diff --git a/clip_large_pretrain_4x256_IF_lr1e-4/20230606_050006/20230606_050006.log b/clip_large_pretrain_4x256_IF_lr1e-4/20230606_050006/20230606_050006.log new file mode 100644 index 0000000000000000000000000000000000000000..19975ae56951655b6a0df9e7a3cf961ec5aa9b63 --- /dev/null +++ b/clip_large_pretrain_4x256_IF_lr1e-4/20230606_050006/20230606_050006.log @@ -0,0 +1,1742 @@ +2023/06/06 05:00:11 - mmengine - INFO - +------------------------------------------------------------ +System environment: + sys.platform: linux + Python: 3.10.9 (main, Mar 8 2023, 10:47:38) [GCC 11.2.0] + CUDA available: True + numpy_random_seed: 1053694385 + GPU 0,1,2,3: NVIDIA A100-SXM4-80GB + CUDA_HOME: /mnt/petrelfs/share/cuda-11.6 + NVCC: Cuda compilation tools, release 11.6, V11.6.124 + GCC: gcc (GCC) 7.5.0 + PyTorch: 1.13.1 + PyTorch compiling details: PyTorch built with: + - GCC 9.3 + - C++ Version: 201402 + - Intel(R) oneAPI Math Kernel Library Version 2021.4-Product Build 20210904 for Intel(R) 64 architecture applications + - Intel(R) MKL-DNN v2.6.0 (Git Hash 52b5f107dd9cf10910aaa19cb47f3abf9b349815) + - OpenMP 201511 (a.k.a. OpenMP 4.5) + - LAPACK is enabled (usually provided by MKL) + - NNPACK is enabled + - CPU capability usage: AVX2 + - CUDA Runtime 11.6 + - NVCC architecture flags: -gencode;arch=compute_37,code=sm_37;-gencode;arch=compute_50,code=sm_50;-gencode;arch=compute_60,code=sm_60;-gencode;arch=compute_61,code=sm_61;-gencode;arch=compute_70,code=sm_70;-gencode;arch=compute_75,code=sm_75;-gencode;arch=compute_80,code=sm_80;-gencode;arch=compute_86,code=sm_86;-gencode;arch=compute_37,code=compute_37 + - CuDNN 8.3.2 (built against CUDA 11.5) + - Magma 2.6.1 + - Build settings: BLAS_INFO=mkl, BUILD_TYPE=Release, CUDA_VERSION=11.6, CUDNN_VERSION=8.3.2, CXX_COMPILER=/opt/rh/devtoolset-9/root/usr/bin/c++, CXX_FLAGS= -fabi-version=11 -Wno-deprecated -fvisibility-inlines-hidden -DUSE_PTHREADPOOL -fopenmp -DNDEBUG -DUSE_KINETO -DUSE_FBGEMM -DUSE_QNNPACK -DUSE_PYTORCH_QNNPACK -DUSE_XNNPACK -DSYMBOLICATE_MOBILE_DEBUG_HANDLE -DEDGE_PROFILER_USE_KINETO -O2 -fPIC -Wno-narrowing -Wall -Wextra -Werror=return-type -Werror=non-virtual-dtor -Wno-missing-field-initializers -Wno-type-limits -Wno-array-bounds -Wno-unknown-pragmas -Wunused-local-typedefs -Wno-unused-parameter -Wno-unused-function -Wno-unused-result -Wno-strict-overflow -Wno-strict-aliasing -Wno-error=deprecated-declarations -Wno-stringop-overflow -Wno-psabi -Wno-error=pedantic -Wno-error=redundant-decls -Wno-error=old-style-cast -fdiagnostics-color=always -faligned-new -Wno-unused-but-set-variable -Wno-maybe-uninitialized -fno-math-errno -fno-trapping-math -Werror=format -Werror=cast-function-type -Wno-stringop-overflow, LAPACK_INFO=mkl, PERF_WITH_AVX=1, PERF_WITH_AVX2=1, PERF_WITH_AVX512=1, TORCH_VERSION=1.13.1, USE_CUDA=ON, USE_CUDNN=ON, USE_EXCEPTION_PTR=1, USE_GFLAGS=OFF, USE_GLOG=OFF, USE_MKL=ON, USE_MKLDNN=ON, USE_MPI=OFF, USE_NCCL=ON, USE_NNPACK=ON, USE_OPENMP=ON, USE_ROCM=OFF, + + TorchVision: 0.14.1 + OpenCV: 4.7.0 + MMEngine: 0.7.3 + +Runtime environment: + cudnn_benchmark: True + mp_cfg: {'mp_start_method': 'fork', 'opencv_num_threads': 0} + dist_cfg: {'backend': 'nccl'} + seed: None + deterministic: False + Distributed launcher: slurm + Distributed training: True + GPU number: 4 +------------------------------------------------------------ + +2023/06/06 05:00:14 - mmengine - INFO - Config: +optim_wrapper = dict( + optimizer=dict( + type='AdamW', lr=0.0001, weight_decay=0.3, _scope_='mmpretrain'), + paramwise_cfg=dict( + custom_keys=dict({ + '.cls_token': dict(decay_mult=0.0), + '.pos_embed': dict(decay_mult=0.0) + })), + type='AmpOptimWrapper', + dtype='bfloat16', + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=False, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=4096) +model = dict( + type='ImageClassifier', + backbone=dict( + frozen_stages=24, + type='VisionTransformer', + arch='l', + img_size=224, + patch_size=14, + drop_rate=0.1, + pre_norm=True, + final_norm=False, + init_cfg=dict( + type='Pretrained', + checkpoint='ckpt/openclip-ViT-L-14.pth', + prefix='backbone')), + neck=dict( + type='CLIPProjection', + in_channels=1024, + out_channels=768, + init_cfg=dict( + type='Pretrained', + checkpoint='ckpt/openclip-ViT-L-14.pth', + prefix='backbone')), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=768, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + init_cfg=None), + init_cfg=dict( + type='TruncNormal', layer=['Conv2d', 'Linear'], std=0.02, bias=0.0), + train_cfg=None) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=128, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/IF80w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/if-dpmsolver++-50-20w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/cc1m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=128, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/if-dpmsolver++-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=128, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/if-dpmsolver++-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/clip_large_pretrain_4x256_IF_lr1e-4' + +2023/06/06 05:00:26 - mmengine - INFO - Hooks will be executed in the following order: +before_run: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_load_checkpoint: +(ABOVE_NORMAL) EMAHook + -------------------- +before_train: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_train_epoch: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(NORMAL ) DistSamplerSeedHook + -------------------- +before_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook + -------------------- +after_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +after_train_epoch: +(NORMAL ) IterTimerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_val_epoch: +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook + -------------------- +before_val_iter: +(NORMAL ) IterTimerHook + -------------------- +after_val_iter: +(NORMAL ) IterTimerHook +(NORMAL ) VisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_val_epoch: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_save_checkpoint: +(ABOVE_NORMAL) EMAHook + -------------------- +after_train: +(VERY_LOW ) CheckpointHook + -------------------- +before_test_epoch: +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook + -------------------- +before_test_iter: +(NORMAL ) IterTimerHook + -------------------- +after_test_iter: +(NORMAL ) IterTimerHook +(NORMAL ) VisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_test_epoch: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_run: +(BELOW_NORMAL) LoggerHook + -------------------- +2023/06/06 05:00:42 - mmengine - INFO - load backbone in model from: ckpt/openclip-ViT-L-14.pth +2023/06/06 05:00:44 - mmengine - WARNING - The model and loaded state dict do not match exactly + +unexpected key in source state_dict: ln1.weight, ln1.bias + +2023/06/06 05:00:44 - mmengine - INFO - load backbone in model from: ckpt/openclip-ViT-L-14.pth +2023/06/06 05:00:46 - mmengine - WARNING - The model and loaded state dict do not match exactly + +unexpected key in source state_dict: cls_token, pos_embed, patch_embed.projection.weight, pre_norm.weight, pre_norm.bias, layers.0.ln1.weight, layers.0.ln1.bias, layers.0.attn.qkv.weight, layers.0.attn.qkv.bias, layers.0.attn.proj.weight, layers.0.attn.proj.bias, layers.0.ln2.weight, layers.0.ln2.bias, layers.0.ffn.layers.0.0.weight, layers.0.ffn.layers.0.0.bias, layers.0.ffn.layers.1.weight, layers.0.ffn.layers.1.bias, layers.1.ln1.weight, layers.1.ln1.bias, layers.1.attn.qkv.weight, layers.1.attn.qkv.bias, layers.1.attn.proj.weight, layers.1.attn.proj.bias, layers.1.ln2.weight, layers.1.ln2.bias, layers.1.ffn.layers.0.0.weight, layers.1.ffn.layers.0.0.bias, layers.1.ffn.layers.1.weight, layers.1.ffn.layers.1.bias, layers.2.ln1.weight, layers.2.ln1.bias, layers.2.attn.qkv.weight, layers.2.attn.qkv.bias, layers.2.attn.proj.weight, layers.2.attn.proj.bias, layers.2.ln2.weight, layers.2.ln2.bias, layers.2.ffn.layers.0.0.weight, layers.2.ffn.layers.0.0.bias, layers.2.ffn.layers.1.weight, layers.2.ffn.layers.1.bias, layers.3.ln1.weight, layers.3.ln1.bias, layers.3.attn.qkv.weight, layers.3.attn.qkv.bias, layers.3.attn.proj.weight, layers.3.attn.proj.bias, layers.3.ln2.weight, layers.3.ln2.bias, layers.3.ffn.layers.0.0.weight, layers.3.ffn.layers.0.0.bias, layers.3.ffn.layers.1.weight, layers.3.ffn.layers.1.bias, layers.4.ln1.weight, layers.4.ln1.bias, layers.4.attn.qkv.weight, layers.4.attn.qkv.bias, layers.4.attn.proj.weight, layers.4.attn.proj.bias, layers.4.ln2.weight, layers.4.ln2.bias, layers.4.ffn.layers.0.0.weight, layers.4.ffn.layers.0.0.bias, layers.4.ffn.layers.1.weight, layers.4.ffn.layers.1.bias, layers.5.ln1.weight, layers.5.ln1.bias, layers.5.attn.qkv.weight, layers.5.attn.qkv.bias, layers.5.attn.proj.weight, layers.5.attn.proj.bias, layers.5.ln2.weight, layers.5.ln2.bias, layers.5.ffn.layers.0.0.weight, layers.5.ffn.layers.0.0.bias, layers.5.ffn.layers.1.weight, layers.5.ffn.layers.1.bias, layers.6.ln1.weight, layers.6.ln1.bias, layers.6.attn.qkv.weight, layers.6.attn.qkv.bias, layers.6.attn.proj.weight, layers.6.attn.proj.bias, layers.6.ln2.weight, layers.6.ln2.bias, layers.6.ffn.layers.0.0.weight, layers.6.ffn.layers.0.0.bias, layers.6.ffn.layers.1.weight, layers.6.ffn.layers.1.bias, layers.7.ln1.weight, layers.7.ln1.bias, layers.7.attn.qkv.weight, layers.7.attn.qkv.bias, layers.7.attn.proj.weight, layers.7.attn.proj.bias, layers.7.ln2.weight, layers.7.ln2.bias, layers.7.ffn.layers.0.0.weight, layers.7.ffn.layers.0.0.bias, layers.7.ffn.layers.1.weight, layers.7.ffn.layers.1.bias, layers.8.ln1.weight, layers.8.ln1.bias, layers.8.attn.qkv.weight, layers.8.attn.qkv.bias, layers.8.attn.proj.weight, layers.8.attn.proj.bias, layers.8.ln2.weight, layers.8.ln2.bias, layers.8.ffn.layers.0.0.weight, layers.8.ffn.layers.0.0.bias, layers.8.ffn.layers.1.weight, layers.8.ffn.layers.1.bias, layers.9.ln1.weight, layers.9.ln1.bias, layers.9.attn.qkv.weight, layers.9.attn.qkv.bias, layers.9.attn.proj.weight, layers.9.attn.proj.bias, layers.9.ln2.weight, layers.9.ln2.bias, layers.9.ffn.layers.0.0.weight, layers.9.ffn.layers.0.0.bias, layers.9.ffn.layers.1.weight, layers.9.ffn.layers.1.bias, layers.10.ln1.weight, layers.10.ln1.bias, layers.10.attn.qkv.weight, layers.10.attn.qkv.bias, layers.10.attn.proj.weight, layers.10.attn.proj.bias, layers.10.ln2.weight, layers.10.ln2.bias, layers.10.ffn.layers.0.0.weight, layers.10.ffn.layers.0.0.bias, layers.10.ffn.layers.1.weight, layers.10.ffn.layers.1.bias, layers.11.ln1.weight, layers.11.ln1.bias, layers.11.attn.qkv.weight, layers.11.attn.qkv.bias, layers.11.attn.proj.weight, layers.11.attn.proj.bias, layers.11.ln2.weight, layers.11.ln2.bias, layers.11.ffn.layers.0.0.weight, layers.11.ffn.layers.0.0.bias, layers.11.ffn.layers.1.weight, layers.11.ffn.layers.1.bias, layers.12.ln1.weight, layers.12.ln1.bias, layers.12.attn.qkv.weight, layers.12.attn.qkv.bias, layers.12.attn.proj.weight, layers.12.attn.proj.bias, layers.12.ln2.weight, layers.12.ln2.bias, layers.12.ffn.layers.0.0.weight, layers.12.ffn.layers.0.0.bias, layers.12.ffn.layers.1.weight, layers.12.ffn.layers.1.bias, layers.13.ln1.weight, layers.13.ln1.bias, layers.13.attn.qkv.weight, layers.13.attn.qkv.bias, layers.13.attn.proj.weight, layers.13.attn.proj.bias, layers.13.ln2.weight, layers.13.ln2.bias, layers.13.ffn.layers.0.0.weight, layers.13.ffn.layers.0.0.bias, layers.13.ffn.layers.1.weight, layers.13.ffn.layers.1.bias, layers.14.ln1.weight, layers.14.ln1.bias, layers.14.attn.qkv.weight, layers.14.attn.qkv.bias, layers.14.attn.proj.weight, layers.14.attn.proj.bias, layers.14.ln2.weight, layers.14.ln2.bias, layers.14.ffn.layers.0.0.weight, layers.14.ffn.layers.0.0.bias, layers.14.ffn.layers.1.weight, layers.14.ffn.layers.1.bias, layers.15.ln1.weight, layers.15.ln1.bias, layers.15.attn.qkv.weight, layers.15.attn.qkv.bias, layers.15.attn.proj.weight, layers.15.attn.proj.bias, layers.15.ln2.weight, layers.15.ln2.bias, layers.15.ffn.layers.0.0.weight, layers.15.ffn.layers.0.0.bias, layers.15.ffn.layers.1.weight, layers.15.ffn.layers.1.bias, layers.16.ln1.weight, layers.16.ln1.bias, layers.16.attn.qkv.weight, layers.16.attn.qkv.bias, layers.16.attn.proj.weight, layers.16.attn.proj.bias, layers.16.ln2.weight, layers.16.ln2.bias, layers.16.ffn.layers.0.0.weight, layers.16.ffn.layers.0.0.bias, layers.16.ffn.layers.1.weight, layers.16.ffn.layers.1.bias, layers.17.ln1.weight, layers.17.ln1.bias, layers.17.attn.qkv.weight, layers.17.attn.qkv.bias, layers.17.attn.proj.weight, layers.17.attn.proj.bias, layers.17.ln2.weight, layers.17.ln2.bias, layers.17.ffn.layers.0.0.weight, layers.17.ffn.layers.0.0.bias, layers.17.ffn.layers.1.weight, layers.17.ffn.layers.1.bias, layers.18.ln1.weight, layers.18.ln1.bias, layers.18.attn.qkv.weight, layers.18.attn.qkv.bias, layers.18.attn.proj.weight, layers.18.attn.proj.bias, layers.18.ln2.weight, layers.18.ln2.bias, layers.18.ffn.layers.0.0.weight, layers.18.ffn.layers.0.0.bias, layers.18.ffn.layers.1.weight, layers.18.ffn.layers.1.bias, layers.19.ln1.weight, layers.19.ln1.bias, layers.19.attn.qkv.weight, layers.19.attn.qkv.bias, layers.19.attn.proj.weight, layers.19.attn.proj.bias, layers.19.ln2.weight, layers.19.ln2.bias, layers.19.ffn.layers.0.0.weight, layers.19.ffn.layers.0.0.bias, layers.19.ffn.layers.1.weight, layers.19.ffn.layers.1.bias, layers.20.ln1.weight, layers.20.ln1.bias, layers.20.attn.qkv.weight, layers.20.attn.qkv.bias, layers.20.attn.proj.weight, layers.20.attn.proj.bias, layers.20.ln2.weight, layers.20.ln2.bias, layers.20.ffn.layers.0.0.weight, layers.20.ffn.layers.0.0.bias, layers.20.ffn.layers.1.weight, layers.20.ffn.layers.1.bias, layers.21.ln1.weight, layers.21.ln1.bias, layers.21.attn.qkv.weight, layers.21.attn.qkv.bias, layers.21.attn.proj.weight, layers.21.attn.proj.bias, layers.21.ln2.weight, layers.21.ln2.bias, layers.21.ffn.layers.0.0.weight, layers.21.ffn.layers.0.0.bias, layers.21.ffn.layers.1.weight, layers.21.ffn.layers.1.bias, layers.22.ln1.weight, layers.22.ln1.bias, layers.22.attn.qkv.weight, layers.22.attn.qkv.bias, layers.22.attn.proj.weight, layers.22.attn.proj.bias, layers.22.ln2.weight, layers.22.ln2.bias, layers.22.ffn.layers.0.0.weight, layers.22.ffn.layers.0.0.bias, layers.22.ffn.layers.1.weight, layers.22.ffn.layers.1.bias, layers.23.ln1.weight, layers.23.ln1.bias, layers.23.attn.qkv.weight, layers.23.attn.qkv.bias, layers.23.attn.proj.weight, layers.23.attn.proj.bias, layers.23.ln2.weight, layers.23.ln2.bias, layers.23.ffn.layers.0.0.weight, layers.23.ffn.layers.0.0.bias, layers.23.ffn.layers.1.weight, layers.23.ffn.layers.1.bias, ln1.weight, ln1.bias + +missing keys in source state_dict: proj + +Name of parameter - Initialization information + +backbone.cls_token - torch.Size([1, 1, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.pos_embed - torch.Size([1, 257, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.patch_embed.projection.weight - torch.Size([1024, 3, 14, 14]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.0.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.0.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.0.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.0.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.0.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.0.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.0.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.0.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.0.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.0.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.0.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.0.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.1.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.1.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.1.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.1.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.1.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.1.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.1.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.1.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.1.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.1.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.1.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.1.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.2.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.2.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.2.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.2.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.2.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.2.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.2.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.2.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.2.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.2.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.2.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.2.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.3.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.3.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.3.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.3.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.3.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.3.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.3.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.3.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.3.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.3.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.3.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.3.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.4.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.4.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.4.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.4.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.4.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.4.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.4.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.4.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.4.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.4.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.4.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.4.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.5.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.5.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.5.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.5.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.5.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.5.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.5.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.5.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.5.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.5.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.5.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.5.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.6.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.6.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.6.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.6.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.6.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.6.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.6.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.6.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.6.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.6.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.6.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.6.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.7.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.7.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.7.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.7.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.7.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.7.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.7.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.7.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.7.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.7.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.7.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.7.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.8.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.8.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.8.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.8.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.8.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.8.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.8.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.8.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.8.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.8.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.8.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.8.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.9.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.9.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.9.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.9.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.9.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.9.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.9.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.9.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.9.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.9.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.9.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.9.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.10.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.10.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.10.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.10.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.10.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.10.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.10.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.10.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.10.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.10.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.10.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.10.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.11.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.11.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.11.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.11.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.11.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.11.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.11.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.11.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.11.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.11.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.11.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.11.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.12.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.12.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.12.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.12.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.12.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.12.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.12.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.12.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.12.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.12.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.12.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.12.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.13.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.13.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.13.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.13.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.13.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.13.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.13.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.13.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.13.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.13.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.13.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.13.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.14.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.14.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.14.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.14.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.14.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.14.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.14.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.14.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.14.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.14.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.14.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.14.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.15.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.15.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.15.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.15.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.15.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.15.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.15.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.15.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.15.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.15.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.15.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.15.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.16.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.16.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.16.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.16.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.16.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.16.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.16.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.16.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.16.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.16.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.16.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.16.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.17.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.17.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.17.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.17.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.17.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.17.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.17.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.17.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.17.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.17.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.17.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.17.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.18.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.18.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.18.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.18.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.18.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.18.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.18.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.18.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.18.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.18.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.18.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.18.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.19.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.19.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.19.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.19.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.19.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.19.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.19.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.19.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.19.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.19.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.19.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.19.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.20.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.20.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.20.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.20.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.20.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.20.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.20.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.20.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.20.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.20.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.20.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.20.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.21.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.21.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.21.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.21.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.21.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.21.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.21.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.21.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.21.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.21.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.21.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.21.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.22.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.22.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.22.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.22.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.22.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.22.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.22.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.22.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.22.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.22.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.22.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.22.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.23.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.23.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.23.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.23.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.23.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.23.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.23.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.23.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.23.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.23.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.23.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.23.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.pre_norm.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.pre_norm.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +neck.proj - torch.Size([1024, 768]): +The value is the same before and after calling `init_weights` of ImageClassifier + +head.fc.weight - torch.Size([2, 768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +head.fc.bias - torch.Size([2]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 +2023/06/06 05:00:46 - mmengine - WARNING - "FileClient" will be deprecated in future. Please use io functions in https://mmengine.readthedocs.io/en/latest/api/fileio.html#file-io +2023/06/06 05:00:46 - mmengine - WARNING - "HardDiskBackend" is the alias of "LocalBackend" and the former will be deprecated in future. +2023/06/06 05:00:46 - mmengine - INFO - Checkpoints will be saved to /mnt/petrelfs/luzeyu/workspace/fakebench/mmpretrain/workdir/clip_large_pretrain_4x256_IF_lr1e-4. +2023/06/06 05:01:55 - mmengine - INFO - Epoch(train) [1][ 100/3937] lr: 9.9999e-05 eta: 7:29:53 time: 0.6355 data_time: 0.0017 memory: 44139 loss: 0.5122 +2023/06/06 05:02:58 - mmengine - INFO - Epoch(train) [1][ 200/3937] lr: 9.9994e-05 eta: 7:11:52 time: 0.6366 data_time: 0.0015 memory: 44139 loss: 0.4937 +2023/06/06 05:04:02 - mmengine - INFO - Epoch(train) [1][ 300/3937] lr: 9.9987e-05 eta: 7:05:16 time: 0.6359 data_time: 0.0015 memory: 44139 loss: 0.4727 +2023/06/06 05:05:05 - mmengine - INFO - Epoch(train) [1][ 400/3937] lr: 9.9977e-05 eta: 7:01:25 time: 0.6360 data_time: 0.0017 memory: 44139 loss: 0.4517 +2023/06/06 05:06:09 - mmengine - INFO - Epoch(train) [1][ 500/3937] lr: 9.9964e-05 eta: 6:58:43 time: 0.6357 data_time: 0.0015 memory: 44139 loss: 0.4451 +2023/06/06 05:07:13 - mmengine - INFO - Epoch(train) [1][ 600/3937] lr: 9.9949e-05 eta: 6:56:35 time: 0.6361 data_time: 0.0015 memory: 44139 loss: 0.4288 +2023/06/06 05:08:16 - mmengine - INFO - Epoch(train) [1][ 700/3937] lr: 9.9930e-05 eta: 6:54:44 time: 0.6368 data_time: 0.0014 memory: 44139 loss: 0.4537 +2023/06/06 05:09:20 - mmengine - INFO - Epoch(train) [1][ 800/3937] lr: 9.9909e-05 eta: 6:53:06 time: 0.6362 data_time: 0.0015 memory: 44139 loss: 0.4423 +2023/06/06 05:10:24 - mmengine - INFO - Epoch(train) [1][ 900/3937] lr: 9.9884e-05 eta: 6:51:36 time: 0.6361 data_time: 0.0017 memory: 44139 loss: 0.4625 +2023/06/06 05:11:27 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_IF_lr1e-4_20230606_050006 +2023/06/06 05:11:27 - mmengine - INFO - Epoch(train) [1][1000/3937] lr: 9.9857e-05 eta: 6:50:13 time: 0.6368 data_time: 0.0016 memory: 44139 loss: 0.4689 +2023/06/06 05:12:31 - mmengine - INFO - Epoch(train) [1][1100/3937] lr: 9.9827e-05 eta: 6:49:09 time: 0.6370 data_time: 0.0016 memory: 44139 loss: 0.4332 +2023/06/06 05:13:35 - mmengine - INFO - Epoch(train) [1][1200/3937] lr: 9.9794e-05 eta: 6:47:51 time: 0.6362 data_time: 0.0014 memory: 44139 loss: 0.4505 +2023/06/06 05:14:40 - mmengine - INFO - Epoch(train) [1][1300/3937] lr: 9.9758e-05 eta: 6:47:02 time: 0.6367 data_time: 0.0015 memory: 44139 loss: 0.4287 +2023/06/06 05:15:43 - mmengine - INFO - Epoch(train) [1][1400/3937] lr: 9.9720e-05 eta: 6:45:46 time: 0.6370 data_time: 0.0014 memory: 44139 loss: 0.4348 +2023/06/06 05:16:47 - mmengine - INFO - Epoch(train) [1][1500/3937] lr: 9.9678e-05 eta: 6:44:31 time: 0.6361 data_time: 0.0014 memory: 44139 loss: 0.4104 +2023/06/06 05:17:51 - mmengine - INFO - Epoch(train) [1][1600/3937] lr: 9.9634e-05 eta: 6:43:18 time: 0.6372 data_time: 0.0015 memory: 44139 loss: 0.4315 +2023/06/06 05:18:55 - mmengine - INFO - Epoch(train) [1][1700/3937] lr: 9.9587e-05 eta: 6:42:10 time: 0.6372 data_time: 0.0015 memory: 44139 loss: 0.3877 +2023/06/06 05:19:59 - mmengine - INFO - Epoch(train) [1][1800/3937] lr: 9.9537e-05 eta: 6:41:00 time: 0.6368 data_time: 0.0015 memory: 44139 loss: 0.4208 +2023/06/06 05:21:02 - mmengine - INFO - Epoch(train) [1][1900/3937] lr: 9.9484e-05 eta: 6:39:48 time: 0.6369 data_time: 0.0016 memory: 44139 loss: 0.3864 +2023/06/06 05:22:06 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_IF_lr1e-4_20230606_050006 +2023/06/06 05:22:06 - mmengine - INFO - Epoch(train) [1][2000/3937] lr: 9.9429e-05 eta: 6:38:41 time: 0.6362 data_time: 0.0018 memory: 44139 loss: 0.4355 +2023/06/06 05:23:10 - mmengine - INFO - Epoch(train) [1][2100/3937] lr: 9.9370e-05 eta: 6:37:31 time: 0.6362 data_time: 0.0015 memory: 44139 loss: 0.4060 +2023/06/06 05:24:13 - mmengine - INFO - Epoch(train) [1][2200/3937] lr: 9.9309e-05 eta: 6:36:21 time: 0.6362 data_time: 0.0015 memory: 44139 loss: 0.4165 +2023/06/06 05:25:17 - mmengine - INFO - Epoch(train) [1][2300/3937] lr: 9.9245e-05 eta: 6:35:13 time: 0.6361 data_time: 0.0017 memory: 44139 loss: 0.3778 +2023/06/06 05:26:22 - mmengine - INFO - Epoch(train) [1][2400/3937] lr: 9.9178e-05 eta: 6:34:23 time: 0.6380 data_time: 0.0021 memory: 44139 loss: 0.3887 +2023/06/06 05:27:26 - mmengine - INFO - Epoch(train) [1][2500/3937] lr: 9.9108e-05 eta: 6:33:13 time: 0.6361 data_time: 0.0015 memory: 44139 loss: 0.3986 +2023/06/06 05:28:29 - mmengine - INFO - Epoch(train) [1][2600/3937] lr: 9.9036e-05 eta: 6:32:05 time: 0.6398 data_time: 0.0017 memory: 44139 loss: 0.3802 +2023/06/06 05:29:33 - mmengine - INFO - Epoch(train) [1][2700/3937] lr: 9.8960e-05 eta: 6:30:56 time: 0.6364 data_time: 0.0015 memory: 44139 loss: 0.4019 +2023/06/06 05:30:37 - mmengine - INFO - Epoch(train) [1][2800/3937] lr: 9.8882e-05 eta: 6:29:48 time: 0.6359 data_time: 0.0014 memory: 44139 loss: 0.3907 +2023/06/06 05:31:40 - mmengine - INFO - Epoch(train) [1][2900/3937] lr: 9.8801e-05 eta: 6:28:39 time: 0.6360 data_time: 0.0015 memory: 44139 loss: 0.3792 +2023/06/06 05:32:44 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_IF_lr1e-4_20230606_050006 +2023/06/06 05:32:44 - mmengine - INFO - Epoch(train) [1][3000/3937] lr: 9.8718e-05 eta: 6:27:32 time: 0.6364 data_time: 0.0014 memory: 44139 loss: 0.3677 +2023/06/06 05:33:48 - mmengine - INFO - Epoch(train) [1][3100/3937] lr: 9.8631e-05 eta: 6:26:25 time: 0.6362 data_time: 0.0017 memory: 44139 loss: 0.4087 +2023/06/06 05:36:14 - mmengine - INFO - Epoch(train) [1][3200/3937] lr: 9.8542e-05 eta: 6:40:52 time: 0.6348 data_time: 0.0015 memory: 44139 loss: 0.4124 +2023/06/06 05:37:17 - mmengine - INFO - Epoch(train) [1][3300/3937] lr: 9.8450e-05 eta: 6:39:14 time: 0.6361 data_time: 0.0015 memory: 44139 loss: 0.3805 +2023/06/06 05:38:21 - mmengine - INFO - Epoch(train) [1][3400/3937] lr: 9.8355e-05 eta: 6:37:39 time: 0.6366 data_time: 0.0015 memory: 44139 loss: 0.3824 +2023/06/06 05:39:25 - mmengine - INFO - Epoch(train) [1][3500/3937] lr: 9.8257e-05 eta: 6:36:11 time: 0.6371 data_time: 0.0014 memory: 44139 loss: 0.3761 +2023/06/06 05:40:29 - mmengine - INFO - Epoch(train) [1][3600/3937] lr: 9.8157e-05 eta: 6:34:39 time: 0.6378 data_time: 0.0015 memory: 44139 loss: 0.3861 +2023/06/06 05:41:33 - mmengine - INFO - Epoch(train) [1][3700/3937] lr: 9.8054e-05 eta: 6:33:09 time: 0.6381 data_time: 0.0016 memory: 44139 loss: 0.3839 +2023/06/06 05:42:36 - mmengine - INFO - Epoch(train) [1][3800/3937] lr: 9.7948e-05 eta: 6:31:40 time: 0.6371 data_time: 0.0015 memory: 44139 loss: 0.3808 +2023/06/06 05:43:40 - mmengine - INFO - Epoch(train) [1][3900/3937] lr: 9.7840e-05 eta: 6:30:14 time: 0.6362 data_time: 0.0014 memory: 44139 loss: 0.3776 +2023/06/06 05:44:04 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_IF_lr1e-4_20230606_050006 +2023/06/06 05:44:04 - mmengine - INFO - Saving checkpoint at 1 epochs +2023/06/06 05:45:38 - mmengine - INFO - Epoch(val) [1][57/57] accuracy/top1: 80.4809 single-label/precision_classwise: [89.67041015625, 73.00264739990234] single-label/recall_classwise: [72.99413299560547, 89.6744155883789] single-label/f1-score_classwise: [80.47744750976562, 80.48423767089844] data_time: 0.0251 time: 1.3206 +2023/06/06 05:46:21 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_IF_lr1e-4_20230606_050006 +2023/06/06 05:46:45 - mmengine - INFO - Epoch(train) [2][ 100/3937] lr: 9.7686e-05 eta: 6:28:47 time: 0.6394 data_time: 0.0016 memory: 44139 loss: 0.3625 +2023/06/06 05:47:49 - mmengine - INFO - Epoch(train) [2][ 200/3937] lr: 9.7571e-05 eta: 6:27:21 time: 0.6369 data_time: 0.0013 memory: 44138 loss: 0.3595 +2023/06/06 05:48:52 - mmengine - INFO - Epoch(train) [2][ 300/3937] lr: 9.7454e-05 eta: 6:25:56 time: 0.6368 data_time: 0.0014 memory: 44138 loss: 0.3616 +2023/06/06 05:49:56 - mmengine - INFO - Epoch(train) [2][ 400/3937] lr: 9.7333e-05 eta: 6:24:33 time: 0.6390 data_time: 0.0014 memory: 44138 loss: 0.3791 +2023/06/06 05:51:00 - mmengine - INFO - Epoch(train) [2][ 500/3937] lr: 9.7210e-05 eta: 6:23:11 time: 0.6482 data_time: 0.0015 memory: 44138 loss: 0.3935 +2023/06/06 05:52:04 - mmengine - INFO - Epoch(train) [2][ 600/3937] lr: 9.7084e-05 eta: 6:21:49 time: 0.6364 data_time: 0.0015 memory: 44138 loss: 0.3580 +2023/06/06 05:53:07 - mmengine - INFO - Epoch(train) [2][ 700/3937] lr: 9.6956e-05 eta: 6:20:27 time: 0.6371 data_time: 0.0016 memory: 44138 loss: 0.3991 +2023/06/06 05:54:11 - mmengine - INFO - Epoch(train) [2][ 800/3937] lr: 9.6825e-05 eta: 6:19:07 time: 0.6358 data_time: 0.0014 memory: 44138 loss: 0.3925 +2023/06/06 05:55:21 - mmengine - INFO - Epoch(train) [2][ 900/3937] lr: 9.6691e-05 eta: 6:18:30 time: 0.6383 data_time: 0.0015 memory: 44138 loss: 0.3729 +2023/06/06 05:56:24 - mmengine - INFO - Epoch(train) [2][1000/3937] lr: 9.6554e-05 eta: 6:17:10 time: 0.6391 data_time: 0.0016 memory: 44138 loss: 0.3283 +2023/06/06 05:57:05 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_IF_lr1e-4_20230606_050006 +2023/06/06 05:57:28 - mmengine - INFO - Epoch(train) [2][1100/3937] lr: 9.6415e-05 eta: 6:15:51 time: 0.6373 data_time: 0.0015 memory: 44138 loss: 0.3831 +2023/06/06 05:58:32 - mmengine - INFO - Epoch(train) [2][1200/3937] lr: 9.6273e-05 eta: 6:14:32 time: 0.6368 data_time: 0.0014 memory: 44138 loss: 0.3891 +2023/06/06 05:59:36 - mmengine - INFO - Epoch(train) [2][1300/3937] lr: 9.6129e-05 eta: 6:13:13 time: 0.6361 data_time: 0.0014 memory: 44138 loss: 0.3717 +2023/06/06 06:00:39 - mmengine - INFO - Epoch(train) [2][1400/3937] lr: 9.5982e-05 eta: 6:11:56 time: 0.6363 data_time: 0.0014 memory: 44138 loss: 0.3645 +2023/06/06 06:01:43 - mmengine - INFO - Epoch(train) [2][1500/3937] lr: 9.5832e-05 eta: 6:10:39 time: 0.6365 data_time: 0.0015 memory: 44138 loss: 0.3516 +2023/06/06 06:02:47 - mmengine - INFO - Epoch(train) [2][1600/3937] lr: 9.5680e-05 eta: 6:09:23 time: 0.6365 data_time: 0.0015 memory: 44138 loss: 0.3908 +2023/06/06 06:03:51 - mmengine - INFO - Epoch(train) [2][1700/3937] lr: 9.5525e-05 eta: 6:08:07 time: 0.6374 data_time: 0.0017 memory: 44138 loss: 0.3329 +2023/06/06 06:04:54 - mmengine - INFO - Epoch(train) [2][1800/3937] lr: 9.5368e-05 eta: 6:06:51 time: 0.6382 data_time: 0.0016 memory: 44138 loss: 0.3408 +2023/06/06 06:05:58 - mmengine - INFO - Epoch(train) [2][1900/3937] lr: 9.5208e-05 eta: 6:05:36 time: 0.6384 data_time: 0.0018 memory: 44138 loss: 0.3576 +2023/06/06 06:07:02 - mmengine - INFO - Epoch(train) [2][2000/3937] lr: 9.5045e-05 eta: 6:04:20 time: 0.6363 data_time: 0.0017 memory: 44138 loss: 0.3598 +2023/06/06 06:07:42 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_IF_lr1e-4_20230606_050006 +2023/06/06 06:08:05 - mmengine - INFO - Epoch(train) [2][2100/3937] lr: 9.4880e-05 eta: 6:03:06 time: 0.6373 data_time: 0.0017 memory: 44138 loss: 0.3374 +2023/06/06 06:09:09 - mmengine - INFO - Epoch(train) [2][2200/3937] lr: 9.4713e-05 eta: 6:01:51 time: 0.6369 data_time: 0.0015 memory: 44138 loss: 0.3715 +2023/06/06 06:10:13 - mmengine - INFO - Epoch(train) [2][2300/3937] lr: 9.4543e-05 eta: 6:00:37 time: 0.6368 data_time: 0.0015 memory: 44138 loss: 0.3563 +2023/06/06 06:11:17 - mmengine - INFO - Epoch(train) [2][2400/3937] lr: 9.4370e-05 eta: 5:59:24 time: 0.6364 data_time: 0.0015 memory: 44138 loss: 0.3385 +2023/06/06 06:12:20 - mmengine - INFO - Epoch(train) [2][2500/3937] lr: 9.4195e-05 eta: 5:58:11 time: 0.6366 data_time: 0.0017 memory: 44138 loss: 0.3685 +2023/06/06 06:13:24 - mmengine - INFO - Epoch(train) [2][2600/3937] lr: 9.4017e-05 eta: 5:56:57 time: 0.6367 data_time: 0.0014 memory: 44138 loss: 0.3705 +2023/06/06 06:14:28 - mmengine - INFO - Epoch(train) [2][2700/3937] lr: 9.3837e-05 eta: 5:55:45 time: 0.6373 data_time: 0.0025 memory: 44138 loss: 0.3419 +2023/06/06 06:15:32 - mmengine - INFO - Epoch(train) [2][2800/3937] lr: 9.3654e-05 eta: 5:54:34 time: 0.6380 data_time: 0.0016 memory: 44138 loss: 0.3687 +2023/06/06 06:16:35 - mmengine - INFO - Epoch(train) [2][2900/3937] lr: 9.3469e-05 eta: 5:53:22 time: 0.6378 data_time: 0.0015 memory: 44138 loss: 0.3347 +2023/06/06 06:17:39 - mmengine - INFO - Epoch(train) [2][3000/3937] lr: 9.3282e-05 eta: 5:52:10 time: 0.6371 data_time: 0.0016 memory: 44138 loss: 0.3643 +2023/06/06 06:18:19 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_IF_lr1e-4_20230606_050006 +2023/06/06 06:18:43 - mmengine - INFO - Epoch(train) [2][3100/3937] lr: 9.3092e-05 eta: 5:50:58 time: 0.6365 data_time: 0.0016 memory: 44138 loss: 0.3295 +2023/06/06 06:19:47 - mmengine - INFO - Epoch(train) [2][3200/3937] lr: 9.2899e-05 eta: 5:49:46 time: 0.6371 data_time: 0.0017 memory: 44138 loss: 0.3336 +2023/06/06 06:20:50 - mmengine - INFO - Epoch(train) [2][3300/3937] lr: 9.2705e-05 eta: 5:48:35 time: 0.6368 data_time: 0.0017 memory: 44138 loss: 0.3379 +2023/06/06 06:21:54 - mmengine - INFO - Epoch(train) [2][3400/3937] lr: 9.2507e-05 eta: 5:47:24 time: 0.6369 data_time: 0.0014 memory: 44138 loss: 0.3514 +2023/06/06 06:22:58 - mmengine - INFO - Epoch(train) [2][3500/3937] lr: 9.2308e-05 eta: 5:46:13 time: 0.6377 data_time: 0.0015 memory: 44138 loss: 0.3443 +2023/06/06 06:24:01 - mmengine - INFO - Epoch(train) [2][3600/3937] lr: 9.2106e-05 eta: 5:45:02 time: 0.6369 data_time: 0.0015 memory: 44138 loss: 0.3348 +2023/06/06 06:25:05 - mmengine - INFO - Epoch(train) [2][3700/3937] lr: 9.1902e-05 eta: 5:43:51 time: 0.6370 data_time: 0.0015 memory: 44138 loss: 0.3649 +2023/06/06 06:26:09 - mmengine - INFO - Epoch(train) [2][3800/3937] lr: 9.1695e-05 eta: 5:42:41 time: 0.6366 data_time: 0.0015 memory: 44138 loss: 0.3500 +2023/06/06 06:27:13 - mmengine - INFO - Epoch(train) [2][3900/3937] lr: 9.1486e-05 eta: 5:41:31 time: 0.6375 data_time: 0.0016 memory: 44138 loss: 0.3576 +2023/06/06 06:27:36 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_IF_lr1e-4_20230606_050006 +2023/06/06 06:27:36 - mmengine - INFO - Saving checkpoint at 2 epochs +2023/06/06 06:29:08 - mmengine - INFO - Epoch(val) [2][57/57] accuracy/top1: 78.8316 single-label/precision_classwise: [96.29875183105469, 68.72116088867188] single-label/recall_classwise: [64.0552978515625, 96.97674560546875] single-label/f1-score_classwise: [76.9353256225586, 80.43981170654297] data_time: 0.0165 time: 1.2614 +2023/06/06 06:30:15 - mmengine - INFO - Epoch(train) [3][ 100/3937] lr: 9.1196e-05 eta: 5:40:07 time: 0.6431 data_time: 0.0015 memory: 44138 loss: 0.3411 +2023/06/06 06:30:31 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_IF_lr1e-4_20230606_050006 +2023/06/06 06:31:18 - mmengine - INFO - Epoch(train) [3][ 200/3937] lr: 9.0981e-05 eta: 5:38:57 time: 0.6373 data_time: 0.0014 memory: 44138 loss: 0.3388 +2023/06/06 06:32:22 - mmengine - INFO - Epoch(train) [3][ 300/3937] lr: 9.0764e-05 eta: 5:37:47 time: 0.6391 data_time: 0.0016 memory: 44138 loss: 0.3596 +2023/06/06 06:33:26 - mmengine - INFO - Epoch(train) [3][ 400/3937] lr: 9.0545e-05 eta: 5:36:38 time: 0.6370 data_time: 0.0016 memory: 44138 loss: 0.3803 +2023/06/06 06:34:30 - mmengine - INFO - Epoch(train) [3][ 500/3937] lr: 9.0324e-05 eta: 5:35:28 time: 0.6369 data_time: 0.0015 memory: 44138 loss: 0.3338 +2023/06/06 06:35:33 - mmengine - INFO - Epoch(train) [3][ 600/3937] lr: 9.0100e-05 eta: 5:34:19 time: 0.6364 data_time: 0.0015 memory: 44138 loss: 0.3524 +2023/06/06 06:36:37 - mmengine - INFO - Epoch(train) [3][ 700/3937] lr: 8.9875e-05 eta: 5:33:10 time: 0.6372 data_time: 0.0018 memory: 44138 loss: 0.3310 +2023/06/06 06:37:41 - mmengine - INFO - Epoch(train) [3][ 800/3937] lr: 8.9647e-05 eta: 5:32:01 time: 0.6366 data_time: 0.0015 memory: 44138 loss: 0.3386 +2023/06/06 06:38:45 - mmengine - INFO - Epoch(train) [3][ 900/3937] lr: 8.9416e-05 eta: 5:30:52 time: 0.6368 data_time: 0.0017 memory: 44138 loss: 0.3177 +2023/06/06 06:39:48 - mmengine - INFO - Epoch(train) [3][1000/3937] lr: 8.9184e-05 eta: 5:29:43 time: 0.6374 data_time: 0.0015 memory: 44138 loss: 0.3489 +2023/06/06 06:40:52 - mmengine - INFO - Epoch(train) [3][1100/3937] lr: 8.8949e-05 eta: 5:28:35 time: 0.6476 data_time: 0.0017 memory: 44138 loss: 0.3389 +2023/06/06 06:41:09 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_IF_lr1e-4_20230606_050006 +2023/06/06 06:41:56 - mmengine - INFO - Epoch(train) [3][1200/3937] lr: 8.8712e-05 eta: 5:27:26 time: 0.6365 data_time: 0.0016 memory: 44138 loss: 0.3414 +2023/06/06 06:43:00 - mmengine - INFO - Epoch(train) [3][1300/3937] lr: 8.8474e-05 eta: 5:26:18 time: 0.6401 data_time: 0.0015 memory: 44138 loss: 0.3254 +2023/06/06 06:44:04 - mmengine - INFO - Epoch(train) [3][1400/3937] lr: 8.8232e-05 eta: 5:25:09 time: 0.6372 data_time: 0.0014 memory: 44138 loss: 0.3486 +2023/06/06 06:45:07 - mmengine - INFO - Epoch(train) [3][1500/3937] lr: 8.7989e-05 eta: 5:24:01 time: 0.6373 data_time: 0.0014 memory: 44138 loss: 0.3306 +2023/06/06 06:46:11 - mmengine - INFO - Epoch(train) [3][1600/3937] lr: 8.7744e-05 eta: 5:22:53 time: 0.6365 data_time: 0.0015 memory: 44138 loss: 0.3267 +2023/06/06 06:47:15 - mmengine - INFO - Epoch(train) [3][1700/3937] lr: 8.7497e-05 eta: 5:21:45 time: 0.6383 data_time: 0.0017 memory: 44138 loss: 0.3426 +2023/06/06 06:48:18 - mmengine - INFO - Epoch(train) [3][1800/3937] lr: 8.7247e-05 eta: 5:20:36 time: 0.6366 data_time: 0.0015 memory: 44138 loss: 0.3574 +2023/06/06 06:49:22 - mmengine - INFO - Epoch(train) [3][1900/3937] lr: 8.6996e-05 eta: 5:19:28 time: 0.6387 data_time: 0.0019 memory: 44138 loss: 0.3522 +2023/06/06 06:50:26 - mmengine - INFO - Epoch(train) [3][2000/3937] lr: 8.6742e-05 eta: 5:18:21 time: 0.6389 data_time: 0.0016 memory: 44138 loss: 0.3137 +2023/06/06 06:51:30 - mmengine - INFO - Epoch(train) [3][2100/3937] lr: 8.6487e-05 eta: 5:17:13 time: 0.6372 data_time: 0.0017 memory: 44138 loss: 0.3321 +2023/06/06 06:51:46 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_IF_lr1e-4_20230606_050006 +2023/06/06 06:52:34 - mmengine - INFO - Epoch(train) [3][2200/3937] lr: 8.6229e-05 eta: 5:16:06 time: 0.6372 data_time: 0.0015 memory: 44138 loss: 0.3593 +2023/06/06 06:53:37 - mmengine - INFO - Epoch(train) [3][2300/3937] lr: 8.5970e-05 eta: 5:14:58 time: 0.6370 data_time: 0.0017 memory: 44138 loss: 0.3654 +2023/06/06 06:54:41 - mmengine - INFO - Epoch(train) [3][2400/3937] lr: 8.5708e-05 eta: 5:13:50 time: 0.6361 data_time: 0.0016 memory: 44138 loss: 0.3468 +2023/06/06 06:55:45 - mmengine - INFO - Epoch(train) [3][2500/3937] lr: 8.5445e-05 eta: 5:12:43 time: 0.6364 data_time: 0.0015 memory: 44138 loss: 0.3359 +2023/06/06 06:56:48 - mmengine - INFO - Epoch(train) [3][2600/3937] lr: 8.5179e-05 eta: 5:11:35 time: 0.6367 data_time: 0.0015 memory: 44138 loss: 0.3342 +2023/06/06 06:57:52 - mmengine - INFO - Epoch(train) [3][2700/3937] lr: 8.4912e-05 eta: 5:10:28 time: 0.6385 data_time: 0.0015 memory: 44138 loss: 0.3113 +2023/06/06 06:58:56 - mmengine - INFO - Epoch(train) [3][2800/3937] lr: 8.4643e-05 eta: 5:09:21 time: 0.6391 data_time: 0.0017 memory: 44138 loss: 0.3383 +2023/06/06 07:00:00 - mmengine - INFO - Epoch(train) [3][2900/3937] lr: 8.4372e-05 eta: 5:08:14 time: 0.6382 data_time: 0.0014 memory: 44138 loss: 0.3224 +2023/06/06 07:01:04 - mmengine - INFO - Epoch(train) [3][3000/3937] lr: 8.4099e-05 eta: 5:07:07 time: 0.6373 data_time: 0.0016 memory: 44138 loss: 0.3374 +2023/06/06 07:02:07 - mmengine - INFO - Epoch(train) [3][3100/3937] lr: 8.3824e-05 eta: 5:06:00 time: 0.6373 data_time: 0.0017 memory: 44138 loss: 0.3260 +2023/06/06 07:02:24 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_IF_lr1e-4_20230606_050006 +2023/06/06 07:03:11 - mmengine - INFO - Epoch(train) [3][3200/3937] lr: 8.3547e-05 eta: 5:04:53 time: 0.6373 data_time: 0.0015 memory: 44138 loss: 0.3232 +2023/06/06 07:04:15 - mmengine - INFO - Epoch(train) [3][3300/3937] lr: 8.3269e-05 eta: 5:03:46 time: 0.6367 data_time: 0.0017 memory: 44138 loss: 0.3508 +2023/06/06 07:05:19 - mmengine - INFO - Epoch(train) [3][3400/3937] lr: 8.2988e-05 eta: 5:02:39 time: 0.6365 data_time: 0.0015 memory: 44138 loss: 0.3453 +2023/06/06 07:06:22 - mmengine - INFO - Epoch(train) [3][3500/3937] lr: 8.2706e-05 eta: 5:01:32 time: 0.6367 data_time: 0.0016 memory: 44138 loss: 0.3596 +2023/06/06 07:07:26 - mmengine - INFO - Epoch(train) [3][3600/3937] lr: 8.2423e-05 eta: 5:00:26 time: 0.6418 data_time: 0.0022 memory: 44138 loss: 0.3164 +2023/06/06 07:08:30 - mmengine - INFO - Epoch(train) [3][3700/3937] lr: 8.2137e-05 eta: 4:59:20 time: 0.6377 data_time: 0.0017 memory: 44138 loss: 0.3380 +2023/06/06 07:09:34 - mmengine - INFO - Epoch(train) [3][3800/3937] lr: 8.1850e-05 eta: 4:58:13 time: 0.6373 data_time: 0.0016 memory: 44138 loss: 0.3182 +2023/06/06 07:10:38 - mmengine - INFO - Epoch(train) [3][3900/3937] lr: 8.1561e-05 eta: 4:57:06 time: 0.6369 data_time: 0.0016 memory: 44138 loss: 0.3442 +2023/06/06 07:11:01 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_IF_lr1e-4_20230606_050006 +2023/06/06 07:11:01 - mmengine - INFO - Saving checkpoint at 3 epochs +2023/06/06 07:12:32 - mmengine - INFO - Epoch(val) [3][57/57] accuracy/top1: 82.1649 single-label/precision_classwise: [97.1237564086914, 72.37509155273438] single-label/recall_classwise: [69.70519256591797, 97.46511840820312] single-label/f1-score_classwise: [81.16133880615234, 83.06686401367188] data_time: 0.0157 time: 1.2586 +2023/06/06 07:13:39 - mmengine - INFO - Epoch(train) [4][ 100/3937] lr: 8.1162e-05 eta: 4:55:43 time: 0.6382 data_time: 0.0031 memory: 44138 loss: 0.3380 +2023/06/06 07:14:36 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_IF_lr1e-4_20230606_050006 +2023/06/06 07:14:43 - mmengine - INFO - Epoch(train) [4][ 200/3937] lr: 8.0869e-05 eta: 4:54:36 time: 0.6364 data_time: 0.0016 memory: 44138 loss: 0.3376 +2023/06/06 07:15:47 - mmengine - INFO - Epoch(train) [4][ 300/3937] lr: 8.0574e-05 eta: 4:53:29 time: 0.6370 data_time: 0.0015 memory: 44138 loss: 0.3617 +2023/06/06 07:16:51 - mmengine - INFO - Epoch(train) [4][ 400/3937] lr: 8.0278e-05 eta: 4:52:23 time: 0.6372 data_time: 0.0016 memory: 44138 loss: 0.3121 +2023/06/06 07:17:54 - mmengine - INFO - Epoch(train) [4][ 500/3937] lr: 7.9980e-05 eta: 4:51:16 time: 0.6371 data_time: 0.0018 memory: 44138 loss: 0.3400 +2023/06/06 07:18:58 - mmengine - INFO - Epoch(train) [4][ 600/3937] lr: 7.9681e-05 eta: 4:50:10 time: 0.6365 data_time: 0.0015 memory: 44138 loss: 0.3378 +2023/06/06 07:20:02 - mmengine - INFO - Epoch(train) [4][ 700/3937] lr: 7.9380e-05 eta: 4:49:04 time: 0.6368 data_time: 0.0016 memory: 44138 loss: 0.3383 +2023/06/06 07:21:06 - mmengine - INFO - Epoch(train) [4][ 800/3937] lr: 7.9077e-05 eta: 4:47:57 time: 0.6374 data_time: 0.0022 memory: 44138 loss: 0.3340 +2023/06/06 07:22:09 - mmengine - INFO - Epoch(train) [4][ 900/3937] lr: 7.8773e-05 eta: 4:46:51 time: 0.6368 data_time: 0.0015 memory: 44138 loss: 0.3364 +2023/06/06 07:23:13 - mmengine - INFO - Epoch(train) [4][1000/3937] lr: 7.8467e-05 eta: 4:45:45 time: 0.6369 data_time: 0.0014 memory: 44138 loss: 0.3190 +2023/06/06 07:24:17 - mmengine - INFO - Epoch(train) [4][1100/3937] lr: 7.8160e-05 eta: 4:44:38 time: 0.6368 data_time: 0.0018 memory: 44138 loss: 0.3354 +2023/06/06 07:25:14 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_IF_lr1e-4_20230606_050006 +2023/06/06 07:25:21 - mmengine - INFO - Epoch(train) [4][1200/3937] lr: 7.7852e-05 eta: 4:43:32 time: 0.6379 data_time: 0.0015 memory: 44138 loss: 0.3120 +2023/06/06 07:26:24 - mmengine - INFO - Epoch(train) [4][1300/3937] lr: 7.7541e-05 eta: 4:42:26 time: 0.6379 data_time: 0.0016 memory: 44138 loss: 0.3525 +2023/06/06 07:27:28 - mmengine - INFO - Epoch(train) [4][1400/3937] lr: 7.7230e-05 eta: 4:41:20 time: 0.6378 data_time: 0.0014 memory: 44138 loss: 0.3322 +2023/06/06 07:28:32 - mmengine - INFO - Epoch(train) [4][1500/3937] lr: 7.6917e-05 eta: 4:40:14 time: 0.6376 data_time: 0.0016 memory: 44138 loss: 0.3502 +2023/06/06 07:29:36 - mmengine - INFO - Epoch(train) [4][1600/3937] lr: 7.6603e-05 eta: 4:39:08 time: 0.6373 data_time: 0.0016 memory: 44138 loss: 0.3586 +2023/06/06 07:30:40 - mmengine - INFO - Epoch(train) [4][1700/3937] lr: 7.6287e-05 eta: 4:38:02 time: 0.6369 data_time: 0.0016 memory: 44138 loss: 0.3289 +2023/06/06 07:31:43 - mmengine - INFO - Epoch(train) [4][1800/3937] lr: 7.5970e-05 eta: 4:36:56 time: 0.6379 data_time: 0.0018 memory: 44138 loss: 0.3384 +2023/06/06 07:32:47 - mmengine - INFO - Epoch(train) [4][1900/3937] lr: 7.5652e-05 eta: 4:35:51 time: 0.6375 data_time: 0.0015 memory: 44138 loss: 0.3200 +2023/06/06 07:33:51 - mmengine - INFO - Epoch(train) [4][2000/3937] lr: 7.5332e-05 eta: 4:34:45 time: 0.6369 data_time: 0.0017 memory: 44138 loss: 0.3320 +2023/06/06 07:34:55 - mmengine - INFO - Epoch(train) [4][2100/3937] lr: 7.5011e-05 eta: 4:33:39 time: 0.6373 data_time: 0.0020 memory: 44138 loss: 0.3438 +2023/06/06 07:35:51 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_IF_lr1e-4_20230606_050006 +2023/06/06 07:35:58 - mmengine - INFO - Epoch(train) [4][2200/3937] lr: 7.4689e-05 eta: 4:32:33 time: 0.6370 data_time: 0.0016 memory: 44138 loss: 0.3016 +2023/06/06 07:37:02 - mmengine - INFO - Epoch(train) [4][2300/3937] lr: 7.4365e-05 eta: 4:31:27 time: 0.6376 data_time: 0.0023 memory: 44138 loss: 0.3412 +2023/06/06 07:38:06 - mmengine - INFO - Epoch(train) [4][2400/3937] lr: 7.4040e-05 eta: 4:30:21 time: 0.6368 data_time: 0.0016 memory: 44138 loss: 0.3275 +2023/06/06 07:39:10 - mmengine - INFO - Epoch(train) [4][2500/3937] lr: 7.3714e-05 eta: 4:29:16 time: 0.6370 data_time: 0.0014 memory: 44138 loss: 0.3261 +2023/06/06 07:40:13 - mmengine - INFO - Epoch(train) [4][2600/3937] lr: 7.3387e-05 eta: 4:28:10 time: 0.6382 data_time: 0.0014 memory: 44138 loss: 0.3429 +2023/06/06 07:41:17 - mmengine - INFO - Epoch(train) [4][2700/3937] lr: 7.3059e-05 eta: 4:27:04 time: 0.6380 data_time: 0.0018 memory: 44138 loss: 0.3348 +2023/06/06 07:42:21 - mmengine - INFO - Epoch(train) [4][2800/3937] lr: 7.2730e-05 eta: 4:25:58 time: 0.6372 data_time: 0.0018 memory: 44138 loss: 0.3406 +2023/06/06 07:43:25 - mmengine - INFO - Epoch(train) [4][2900/3937] lr: 7.2399e-05 eta: 4:24:53 time: 0.6370 data_time: 0.0016 memory: 44138 loss: 0.3422 +2023/06/06 07:44:28 - mmengine - INFO - Epoch(train) [4][3000/3937] lr: 7.2067e-05 eta: 4:23:47 time: 0.6371 data_time: 0.0015 memory: 44138 loss: 0.3132 +2023/06/06 07:45:32 - mmengine - INFO - Epoch(train) [4][3100/3937] lr: 7.1734e-05 eta: 4:22:42 time: 0.6370 data_time: 0.0015 memory: 44138 loss: 0.3458 +2023/06/06 07:46:29 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_IF_lr1e-4_20230606_050006 +2023/06/06 07:46:36 - mmengine - INFO - Epoch(train) [4][3200/3937] lr: 7.1401e-05 eta: 4:21:36 time: 0.6366 data_time: 0.0015 memory: 44138 loss: 0.3669 +2023/06/06 07:47:40 - mmengine - INFO - Epoch(train) [4][3300/3937] lr: 7.1066e-05 eta: 4:20:31 time: 0.6377 data_time: 0.0016 memory: 44138 loss: 0.3488 +2023/06/06 07:48:44 - mmengine - INFO - Epoch(train) [4][3400/3937] lr: 7.0730e-05 eta: 4:19:25 time: 0.6377 data_time: 0.0016 memory: 44138 loss: 0.3177 +2023/06/06 07:49:47 - mmengine - INFO - Epoch(train) [4][3500/3937] lr: 7.0393e-05 eta: 4:18:20 time: 0.6365 data_time: 0.0015 memory: 44138 loss: 0.3466 +2023/06/06 07:50:51 - mmengine - INFO - Epoch(train) [4][3600/3937] lr: 7.0055e-05 eta: 4:17:14 time: 0.6370 data_time: 0.0015 memory: 44138 loss: 0.3368 +2023/06/06 07:51:55 - mmengine - INFO - Epoch(train) [4][3700/3937] lr: 6.9716e-05 eta: 4:16:09 time: 0.6362 data_time: 0.0015 memory: 44138 loss: 0.3190 +2023/06/06 07:52:59 - mmengine - INFO - Epoch(train) [4][3800/3937] lr: 6.9376e-05 eta: 4:15:03 time: 0.6367 data_time: 0.0015 memory: 44138 loss: 0.3387 +2023/06/06 07:54:02 - mmengine - INFO - Epoch(train) [4][3900/3937] lr: 6.9035e-05 eta: 4:13:58 time: 0.6374 data_time: 0.0018 memory: 44138 loss: 0.3340 +2023/06/06 07:54:26 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_IF_lr1e-4_20230606_050006 +2023/06/06 07:54:26 - mmengine - INFO - Saving checkpoint at 4 epochs +2023/06/06 07:55:57 - mmengine - INFO - Epoch(val) [4][57/57] accuracy/top1: 86.2566 single-label/precision_classwise: [97.55258178710938, 77.5588150024414] single-label/recall_classwise: [76.99640655517578, 97.6279067993164] single-label/f1-score_classwise: [86.06407165527344, 86.44381713867188] data_time: 0.0159 time: 1.2590 +2023/06/06 07:57:04 - mmengine - INFO - Epoch(train) [5][ 100/3937] lr: 6.8567e-05 eta: 4:12:33 time: 0.6377 data_time: 0.0016 memory: 44138 loss: 0.3264 +2023/06/06 07:58:08 - mmengine - INFO - Epoch(train) [5][ 200/3937] lr: 6.8224e-05 eta: 4:11:28 time: 0.6375 data_time: 0.0015 memory: 44138 loss: 0.3349 +2023/06/06 07:58:41 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_IF_lr1e-4_20230606_050006 +2023/06/06 07:59:12 - mmengine - INFO - Epoch(train) [5][ 300/3937] lr: 6.7881e-05 eta: 4:10:23 time: 0.6370 data_time: 0.0016 memory: 44138 loss: 0.3428 +2023/06/06 08:00:15 - mmengine - INFO - Epoch(train) [5][ 400/3937] lr: 6.7536e-05 eta: 4:09:17 time: 0.6381 data_time: 0.0026 memory: 44138 loss: 0.3338 +2023/06/06 08:01:19 - mmengine - INFO - Epoch(train) [5][ 500/3937] lr: 6.7191e-05 eta: 4:08:12 time: 0.6389 data_time: 0.0021 memory: 44138 loss: 0.3222 +2023/06/06 08:02:23 - mmengine - INFO - Epoch(train) [5][ 600/3937] lr: 6.6845e-05 eta: 4:07:07 time: 0.6428 data_time: 0.0021 memory: 44138 loss: 0.3249 +2023/06/06 08:03:28 - mmengine - INFO - Epoch(train) [5][ 700/3937] lr: 6.6498e-05 eta: 4:06:02 time: 0.6384 data_time: 0.0018 memory: 44138 loss: 0.3563 +2023/06/06 08:04:32 - mmengine - INFO - Epoch(train) [5][ 800/3937] lr: 6.6151e-05 eta: 4:04:58 time: 0.6386 data_time: 0.0026 memory: 44138 loss: 0.3415 +2023/06/06 08:05:35 - mmengine - INFO - Epoch(train) [5][ 900/3937] lr: 6.5802e-05 eta: 4:03:52 time: 0.6362 data_time: 0.0020 memory: 44138 loss: 0.3410 +2023/06/06 08:06:39 - mmengine - INFO - Epoch(train) [5][1000/3937] lr: 6.5454e-05 eta: 4:02:47 time: 0.6389 data_time: 0.0020 memory: 44138 loss: 0.3617 +2023/06/06 08:07:43 - mmengine - INFO - Epoch(train) [5][1100/3937] lr: 6.5104e-05 eta: 4:01:42 time: 0.6379 data_time: 0.0020 memory: 44138 loss: 0.3619 +2023/06/06 08:08:47 - mmengine - INFO - Epoch(train) [5][1200/3937] lr: 6.4754e-05 eta: 4:00:37 time: 0.6412 data_time: 0.0019 memory: 44138 loss: 0.3187 +2023/06/06 08:09:20 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_IF_lr1e-4_20230606_050006 +2023/06/06 08:09:51 - mmengine - INFO - Epoch(train) [5][1300/3937] lr: 6.4403e-05 eta: 3:59:32 time: 0.6368 data_time: 0.0019 memory: 44138 loss: 0.3272 +2023/06/06 08:10:55 - mmengine - INFO - Epoch(train) [5][1400/3937] lr: 6.4051e-05 eta: 3:58:27 time: 0.6371 data_time: 0.0020 memory: 44138 loss: 0.3496 +2023/06/06 08:11:58 - mmengine - INFO - Epoch(train) [5][1500/3937] lr: 6.3699e-05 eta: 3:57:22 time: 0.6372 data_time: 0.0024 memory: 44138 loss: 0.3175 +2023/06/06 08:13:02 - mmengine - INFO - Epoch(train) [5][1600/3937] lr: 6.3347e-05 eta: 3:56:16 time: 0.6371 data_time: 0.0023 memory: 44138 loss: 0.3220 +2023/06/06 08:14:06 - mmengine - INFO - Epoch(train) [5][1700/3937] lr: 6.2994e-05 eta: 3:55:11 time: 0.6377 data_time: 0.0022 memory: 44138 loss: 0.3163 +2023/06/06 08:15:10 - mmengine - INFO - Epoch(train) [5][1800/3937] lr: 6.2640e-05 eta: 3:54:06 time: 0.6382 data_time: 0.0017 memory: 44138 loss: 0.3425 +2023/06/06 08:16:14 - mmengine - INFO - Epoch(train) [5][1900/3937] lr: 6.2286e-05 eta: 3:53:02 time: 0.6394 data_time: 0.0018 memory: 44138 loss: 0.3069 +2023/06/06 08:17:18 - mmengine - INFO - Epoch(train) [5][2000/3937] lr: 6.1931e-05 eta: 3:51:57 time: 0.6384 data_time: 0.0027 memory: 44138 loss: 0.3705 +2023/06/06 08:18:22 - mmengine - INFO - Epoch(train) [5][2100/3937] lr: 6.1576e-05 eta: 3:50:52 time: 0.6364 data_time: 0.0020 memory: 44138 loss: 0.3499 +2023/06/06 08:19:26 - mmengine - INFO - Epoch(train) [5][2200/3937] lr: 6.1221e-05 eta: 3:49:47 time: 0.6382 data_time: 0.0024 memory: 44138 loss: 0.3240 +2023/06/06 08:19:59 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_IF_lr1e-4_20230606_050006 +2023/06/06 08:20:29 - mmengine - INFO - Epoch(train) [5][2300/3937] lr: 6.0865e-05 eta: 3:48:42 time: 0.6373 data_time: 0.0018 memory: 44138 loss: 0.3075 +2023/06/06 08:21:33 - mmengine - INFO - Epoch(train) [5][2400/3937] lr: 6.0509e-05 eta: 3:47:37 time: 0.6380 data_time: 0.0018 memory: 44138 loss: 0.3451 +2023/06/06 08:22:37 - mmengine - INFO - Epoch(train) [5][2500/3937] lr: 6.0152e-05 eta: 3:46:32 time: 0.6389 data_time: 0.0018 memory: 44138 loss: 0.3387 +2023/06/06 08:23:41 - mmengine - INFO - Epoch(train) [5][2600/3937] lr: 5.9795e-05 eta: 3:45:27 time: 0.6365 data_time: 0.0016 memory: 44138 loss: 0.3282 +2023/06/06 08:24:45 - mmengine - INFO - Epoch(train) [5][2700/3937] lr: 5.9438e-05 eta: 3:44:22 time: 0.6377 data_time: 0.0017 memory: 44138 loss: 0.3411 +2023/06/06 08:25:49 - mmengine - INFO - Epoch(train) [5][2800/3937] lr: 5.9081e-05 eta: 3:43:17 time: 0.6385 data_time: 0.0027 memory: 44138 loss: 0.3036 +2023/06/06 08:26:52 - mmengine - INFO - Epoch(train) [5][2900/3937] lr: 5.8723e-05 eta: 3:42:12 time: 0.6391 data_time: 0.0015 memory: 44138 loss: 0.3333 +2023/06/06 08:27:56 - mmengine - INFO - Epoch(train) [5][3000/3937] lr: 5.8365e-05 eta: 3:41:07 time: 0.6406 data_time: 0.0018 memory: 44138 loss: 0.3187 +2023/06/06 08:29:00 - mmengine - INFO - Epoch(train) [5][3100/3937] lr: 5.8007e-05 eta: 3:40:03 time: 0.6411 data_time: 0.0016 memory: 44138 loss: 0.2957 +2023/06/06 08:30:05 - mmengine - INFO - Epoch(train) [5][3200/3937] lr: 5.7649e-05 eta: 3:38:58 time: 0.6422 data_time: 0.0015 memory: 44138 loss: 0.3227 +2023/06/06 08:30:38 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_IF_lr1e-4_20230606_050006 +2023/06/06 08:31:09 - mmengine - INFO - Epoch(train) [5][3300/3937] lr: 5.7290e-05 eta: 3:37:54 time: 0.6410 data_time: 0.0016 memory: 44138 loss: 0.3376 +2023/06/06 08:32:13 - mmengine - INFO - Epoch(train) [5][3400/3937] lr: 5.6931e-05 eta: 3:36:49 time: 0.6420 data_time: 0.0016 memory: 44138 loss: 0.3269 +2023/06/06 08:33:17 - mmengine - INFO - Epoch(train) [5][3500/3937] lr: 5.6572e-05 eta: 3:35:45 time: 0.6469 data_time: 0.0018 memory: 44138 loss: 0.3279 +2023/06/06 08:34:21 - mmengine - INFO - Epoch(train) [5][3600/3937] lr: 5.6214e-05 eta: 3:34:40 time: 0.6398 data_time: 0.0018 memory: 44138 loss: 0.3242 +2023/06/06 08:35:25 - mmengine - INFO - Epoch(train) [5][3700/3937] lr: 5.5855e-05 eta: 3:33:35 time: 0.6397 data_time: 0.0018 memory: 44138 loss: 0.3336 +2023/06/06 08:36:29 - mmengine - INFO - Epoch(train) [5][3800/3937] lr: 5.5496e-05 eta: 3:32:31 time: 0.6414 data_time: 0.0017 memory: 44138 loss: 0.3647 +2023/06/06 08:37:33 - mmengine - INFO - Epoch(train) [5][3900/3937] lr: 5.5136e-05 eta: 3:31:26 time: 0.6419 data_time: 0.0015 memory: 44138 loss: 0.3502 +2023/06/06 08:37:57 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_IF_lr1e-4_20230606_050006 +2023/06/06 08:37:57 - mmengine - INFO - Saving checkpoint at 5 epochs +2023/06/06 08:39:29 - mmengine - INFO - Epoch(val) [5][57/57] accuracy/top1: 90.2439 single-label/precision_classwise: [97.54212188720703, 83.5861587524414] single-label/recall_classwise: [84.42648315429688, 97.3875961303711] single-label/f1-score_classwise: [90.51165008544922, 89.96060943603516] data_time: 0.0146 time: 1.2585 +2023/06/06 08:40:35 - mmengine - INFO - Epoch(train) [6][ 100/3937] lr: 5.4645e-05 eta: 3:30:00 time: 0.6386 data_time: 0.0016 memory: 44138 loss: 0.3622 +2023/06/06 08:41:39 - mmengine - INFO - Epoch(train) [6][ 200/3937] lr: 5.4285e-05 eta: 3:28:55 time: 0.6384 data_time: 0.0018 memory: 44138 loss: 0.2929 +2023/06/06 08:42:43 - mmengine - INFO - Epoch(train) [6][ 300/3937] lr: 5.3926e-05 eta: 3:27:51 time: 0.6405 data_time: 0.0017 memory: 44138 loss: 0.3145 +2023/06/06 08:42:53 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_IF_lr1e-4_20230606_050006 +2023/06/06 08:43:47 - mmengine - INFO - Epoch(train) [6][ 400/3937] lr: 5.3567e-05 eta: 3:26:46 time: 0.6401 data_time: 0.0029 memory: 44138 loss: 0.3452 +2023/06/06 08:44:51 - mmengine - INFO - Epoch(train) [6][ 500/3937] lr: 5.3209e-05 eta: 3:25:41 time: 0.6493 data_time: 0.0016 memory: 44138 loss: 0.3426 +2023/06/06 08:45:55 - mmengine - INFO - Epoch(train) [6][ 600/3937] lr: 5.2850e-05 eta: 3:24:36 time: 0.6393 data_time: 0.0015 memory: 44138 loss: 0.3071 +2023/06/06 08:46:59 - mmengine - INFO - Epoch(train) [6][ 700/3937] lr: 5.2491e-05 eta: 3:23:31 time: 0.6379 data_time: 0.0017 memory: 44138 loss: 0.3507 +2023/06/06 08:48:03 - mmengine - INFO - Epoch(train) [6][ 800/3937] lr: 5.2133e-05 eta: 3:22:27 time: 0.6376 data_time: 0.0024 memory: 44138 loss: 0.3170 +2023/06/06 08:49:06 - mmengine - INFO - Epoch(train) [6][ 900/3937] lr: 5.1775e-05 eta: 3:21:22 time: 0.6387 data_time: 0.0017 memory: 44138 loss: 0.3608 +2023/06/06 08:50:10 - mmengine - INFO - Epoch(train) [6][1000/3937] lr: 5.1417e-05 eta: 3:20:17 time: 0.6372 data_time: 0.0016 memory: 44138 loss: 0.3422 +2023/06/06 08:51:14 - mmengine - INFO - Epoch(train) [6][1100/3937] lr: 5.1059e-05 eta: 3:19:13 time: 0.6403 data_time: 0.0017 memory: 44138 loss: 0.3411 +2023/06/06 08:52:18 - mmengine - INFO - Epoch(train) [6][1200/3937] lr: 5.0701e-05 eta: 3:18:08 time: 0.6392 data_time: 0.0017 memory: 44138 loss: 0.3372 +2023/06/06 08:53:22 - mmengine - INFO - Epoch(train) [6][1300/3937] lr: 5.0344e-05 eta: 3:17:03 time: 0.6388 data_time: 0.0019 memory: 44138 loss: 0.3402 +2023/06/06 08:53:32 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_IF_lr1e-4_20230606_050006 +2023/06/06 08:54:26 - mmengine - INFO - Epoch(train) [6][1400/3937] lr: 4.9987e-05 eta: 3:15:59 time: 0.6395 data_time: 0.0020 memory: 44138 loss: 0.3344 +2023/06/06 08:55:30 - mmengine - INFO - Epoch(train) [6][1500/3937] lr: 4.9630e-05 eta: 3:14:54 time: 0.6405 data_time: 0.0020 memory: 44138 loss: 0.3596 +2023/06/06 08:56:34 - mmengine - INFO - Epoch(train) [6][1600/3937] lr: 4.9274e-05 eta: 3:13:49 time: 0.6406 data_time: 0.0017 memory: 44138 loss: 0.3008 +2023/06/06 08:57:38 - mmengine - INFO - Epoch(train) [6][1700/3937] lr: 4.8918e-05 eta: 3:12:45 time: 0.6387 data_time: 0.0018 memory: 44138 loss: 0.3156 +2023/06/06 08:58:41 - mmengine - INFO - Epoch(train) [6][1800/3937] lr: 4.8562e-05 eta: 3:11:40 time: 0.6380 data_time: 0.0017 memory: 44138 loss: 0.3158 +2023/06/06 08:59:45 - mmengine - INFO - Epoch(train) [6][1900/3937] lr: 4.8207e-05 eta: 3:10:35 time: 0.6385 data_time: 0.0019 memory: 44138 loss: 0.3264 +2023/06/06 09:00:49 - mmengine - INFO - Epoch(train) [6][2000/3937] lr: 4.7852e-05 eta: 3:09:31 time: 0.6381 data_time: 0.0019 memory: 44138 loss: 0.3240 +2023/06/06 09:01:53 - mmengine - INFO - Epoch(train) [6][2100/3937] lr: 4.7498e-05 eta: 3:08:26 time: 0.6385 data_time: 0.0016 memory: 44138 loss: 0.3161 +2023/06/06 09:02:57 - mmengine - INFO - Epoch(train) [6][2200/3937] lr: 4.7144e-05 eta: 3:07:21 time: 0.6386 data_time: 0.0018 memory: 44138 loss: 0.3347 +2023/06/06 09:04:01 - mmengine - INFO - Epoch(train) [6][2300/3937] lr: 4.6791e-05 eta: 3:06:17 time: 0.6387 data_time: 0.0017 memory: 44138 loss: 0.3308 +2023/06/06 09:04:11 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_IF_lr1e-4_20230606_050006 +2023/06/06 09:05:05 - mmengine - INFO - Epoch(train) [6][2400/3937] lr: 4.6438e-05 eta: 3:05:12 time: 0.6387 data_time: 0.0017 memory: 44138 loss: 0.3387 +2023/06/06 09:06:09 - mmengine - INFO - Epoch(train) [6][2500/3937] lr: 4.6086e-05 eta: 3:04:08 time: 0.6381 data_time: 0.0017 memory: 44138 loss: 0.3311 +2023/06/06 09:07:13 - mmengine - INFO - Epoch(train) [6][2600/3937] lr: 4.5734e-05 eta: 3:03:03 time: 0.6406 data_time: 0.0016 memory: 44138 loss: 0.3468 +2023/06/06 09:08:17 - mmengine - INFO - Epoch(train) [6][2700/3937] lr: 4.5383e-05 eta: 3:01:59 time: 0.6401 data_time: 0.0016 memory: 44138 loss: 0.3607 +2023/06/06 09:09:21 - mmengine - INFO - Epoch(train) [6][2800/3937] lr: 4.5033e-05 eta: 3:00:54 time: 0.6379 data_time: 0.0018 memory: 44138 loss: 0.3262 +2023/06/06 09:10:25 - mmengine - INFO - Epoch(train) [6][2900/3937] lr: 4.4683e-05 eta: 2:59:50 time: 0.6442 data_time: 0.0021 memory: 44138 loss: 0.3510 +2023/06/06 09:11:29 - mmengine - INFO - Epoch(train) [6][3000/3937] lr: 4.4334e-05 eta: 2:58:45 time: 0.6387 data_time: 0.0018 memory: 44138 loss: 0.3286 +2023/06/06 09:12:33 - mmengine - INFO - Epoch(train) [6][3100/3937] lr: 4.3985e-05 eta: 2:57:40 time: 0.6387 data_time: 0.0016 memory: 44138 loss: 0.3489 +2023/06/06 09:13:36 - mmengine - INFO - Epoch(train) [6][3200/3937] lr: 4.3637e-05 eta: 2:56:36 time: 0.6383 data_time: 0.0022 memory: 44138 loss: 0.3292 +2023/06/06 09:14:40 - mmengine - INFO - Epoch(train) [6][3300/3937] lr: 4.3290e-05 eta: 2:55:31 time: 0.6419 data_time: 0.0017 memory: 44138 loss: 0.3328 +2023/06/06 09:14:50 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_IF_lr1e-4_20230606_050006 +2023/06/06 09:15:44 - mmengine - INFO - Epoch(train) [6][3400/3937] lr: 4.2944e-05 eta: 2:54:27 time: 0.6391 data_time: 0.0016 memory: 44138 loss: 0.3286 +2023/06/06 09:16:48 - mmengine - INFO - Epoch(train) [6][3500/3937] lr: 4.2598e-05 eta: 2:53:22 time: 0.6381 data_time: 0.0016 memory: 44138 loss: 0.3221 +2023/06/06 09:17:52 - mmengine - INFO - Epoch(train) [6][3600/3937] lr: 4.2253e-05 eta: 2:52:18 time: 0.6395 data_time: 0.0031 memory: 44138 loss: 0.3422 +2023/06/06 09:18:56 - mmengine - INFO - Epoch(train) [6][3700/3937] lr: 4.1909e-05 eta: 2:51:13 time: 0.6373 data_time: 0.0016 memory: 44138 loss: 0.3500 +2023/06/06 09:20:00 - mmengine - INFO - Epoch(train) [6][3800/3937] lr: 4.1566e-05 eta: 2:50:09 time: 0.6380 data_time: 0.0015 memory: 44138 loss: 0.3381 +2023/06/06 09:21:04 - mmengine - INFO - Epoch(train) [6][3900/3937] lr: 4.1224e-05 eta: 2:49:04 time: 0.6394 data_time: 0.0015 memory: 44138 loss: 0.3400 +2023/06/06 09:21:27 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_IF_lr1e-4_20230606_050006 +2023/06/06 09:21:27 - mmengine - INFO - Saving checkpoint at 6 epochs +2023/06/06 09:22:59 - mmengine - INFO - Epoch(val) [6][57/57] accuracy/top1: 92.4568 single-label/precision_classwise: [97.46580505371094, 87.4250259399414] single-label/recall_classwise: [88.6181411743164, 97.17054748535156] single-label/f1-score_classwise: [92.83162689208984, 92.04053497314453] data_time: 0.0161 time: 1.2590 +2023/06/06 09:24:06 - mmengine - INFO - Epoch(train) [7][ 100/3937] lr: 4.0757e-05 eta: 2:47:38 time: 0.6380 data_time: 0.0016 memory: 44138 loss: 0.3351 +2023/06/06 09:25:10 - mmengine - INFO - Epoch(train) [7][ 200/3937] lr: 4.0416e-05 eta: 2:46:33 time: 0.6393 data_time: 0.0019 memory: 44138 loss: 0.3270 +2023/06/06 09:26:13 - mmengine - INFO - Epoch(train) [7][ 300/3937] lr: 4.0077e-05 eta: 2:45:29 time: 0.6402 data_time: 0.0016 memory: 44138 loss: 0.3321 +2023/06/06 09:27:03 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_IF_lr1e-4_20230606_050006 +2023/06/06 09:27:17 - mmengine - INFO - Epoch(train) [7][ 400/3937] lr: 3.9739e-05 eta: 2:44:24 time: 0.6381 data_time: 0.0017 memory: 44138 loss: 0.3327 +2023/06/06 09:28:21 - mmengine - INFO - Epoch(train) [7][ 500/3937] lr: 3.9402e-05 eta: 2:43:20 time: 0.6379 data_time: 0.0015 memory: 44138 loss: 0.3382 +2023/06/06 09:29:25 - mmengine - INFO - Epoch(train) [7][ 600/3937] lr: 3.9065e-05 eta: 2:42:15 time: 0.6395 data_time: 0.0019 memory: 44138 loss: 0.3125 +2023/06/06 09:30:29 - mmengine - INFO - Epoch(train) [7][ 700/3937] lr: 3.8730e-05 eta: 2:41:11 time: 0.6399 data_time: 0.0016 memory: 44138 loss: 0.3509 +2023/06/06 09:31:33 - mmengine - INFO - Epoch(train) [7][ 800/3937] lr: 3.8396e-05 eta: 2:40:06 time: 0.6391 data_time: 0.0016 memory: 44138 loss: 0.3387 +2023/06/06 09:32:37 - mmengine - INFO - Epoch(train) [7][ 900/3937] lr: 3.8062e-05 eta: 2:39:02 time: 0.6412 data_time: 0.0016 memory: 44138 loss: 0.3242 +2023/06/06 09:33:41 - mmengine - INFO - Epoch(train) [7][1000/3937] lr: 3.7730e-05 eta: 2:37:57 time: 0.6378 data_time: 0.0017 memory: 44138 loss: 0.3211 +2023/06/06 09:34:45 - mmengine - INFO - Epoch(train) [7][1100/3937] lr: 3.7399e-05 eta: 2:36:53 time: 0.6385 data_time: 0.0016 memory: 44138 loss: 0.3298 +2023/06/06 09:35:49 - mmengine - INFO - Epoch(train) [7][1200/3937] lr: 3.7069e-05 eta: 2:35:48 time: 0.6386 data_time: 0.0017 memory: 44138 loss: 0.3449 +2023/06/06 09:36:53 - mmengine - INFO - Epoch(train) [7][1300/3937] lr: 3.6741e-05 eta: 2:34:44 time: 0.6386 data_time: 0.0016 memory: 44138 loss: 0.3130 +2023/06/06 09:37:43 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_IF_lr1e-4_20230606_050006 +2023/06/06 09:37:57 - mmengine - INFO - Epoch(train) [7][1400/3937] lr: 3.6413e-05 eta: 2:33:40 time: 0.6414 data_time: 0.0021 memory: 44138 loss: 0.3471 +2023/06/06 09:39:01 - mmengine - INFO - Epoch(train) [7][1500/3937] lr: 3.6087e-05 eta: 2:32:35 time: 0.6397 data_time: 0.0021 memory: 44138 loss: 0.3027 +2023/06/06 09:40:05 - mmengine - INFO - Epoch(train) [7][1600/3937] lr: 3.5761e-05 eta: 2:31:31 time: 0.6403 data_time: 0.0015 memory: 44138 loss: 0.3238 +2023/06/06 09:41:09 - mmengine - INFO - Epoch(train) [7][1700/3937] lr: 3.5437e-05 eta: 2:30:26 time: 0.6394 data_time: 0.0018 memory: 44138 loss: 0.3504 +2023/06/06 09:42:13 - mmengine - INFO - Epoch(train) [7][1800/3937] lr: 3.5115e-05 eta: 2:29:22 time: 0.6392 data_time: 0.0015 memory: 44138 loss: 0.3467 +2023/06/06 09:43:17 - mmengine - INFO - Epoch(train) [7][1900/3937] lr: 3.4793e-05 eta: 2:28:18 time: 0.6376 data_time: 0.0018 memory: 44138 loss: 0.3153 +2023/06/06 09:44:21 - mmengine - INFO - Epoch(train) [7][2000/3937] lr: 3.4473e-05 eta: 2:27:13 time: 0.6395 data_time: 0.0016 memory: 44138 loss: 0.3325 +2023/06/06 09:45:25 - mmengine - INFO - Epoch(train) [7][2100/3937] lr: 3.4154e-05 eta: 2:26:09 time: 0.6396 data_time: 0.0020 memory: 44138 loss: 0.3500 +2023/06/06 09:46:29 - mmengine - INFO - Epoch(train) [7][2200/3937] lr: 3.3836e-05 eta: 2:25:04 time: 0.6414 data_time: 0.0018 memory: 44138 loss: 0.3323 +2023/06/06 09:47:33 - mmengine - INFO - Epoch(train) [7][2300/3937] lr: 3.3520e-05 eta: 2:24:00 time: 0.6374 data_time: 0.0016 memory: 44138 loss: 0.3465 +2023/06/06 09:48:23 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_IF_lr1e-4_20230606_050006 +2023/06/06 09:48:37 - mmengine - INFO - Epoch(train) [7][2400/3937] lr: 3.3205e-05 eta: 2:22:56 time: 0.6386 data_time: 0.0015 memory: 44138 loss: 0.3382 +2023/06/06 09:49:41 - mmengine - INFO - Epoch(train) [7][2500/3937] lr: 3.2892e-05 eta: 2:21:51 time: 0.6405 data_time: 0.0018 memory: 44138 loss: 0.3324 +2023/06/06 09:50:45 - mmengine - INFO - Epoch(train) [7][2600/3937] lr: 3.2580e-05 eta: 2:20:47 time: 0.6402 data_time: 0.0017 memory: 44138 loss: 0.3484 +2023/06/06 09:51:49 - mmengine - INFO - Epoch(train) [7][2700/3937] lr: 3.2269e-05 eta: 2:19:43 time: 0.6393 data_time: 0.0014 memory: 44138 loss: 0.3480 +2023/06/06 09:52:53 - mmengine - INFO - Epoch(train) [7][2800/3937] lr: 3.1960e-05 eta: 2:18:38 time: 0.6389 data_time: 0.0020 memory: 44138 loss: 0.3267 +2023/06/06 09:53:57 - mmengine - INFO - Epoch(train) [7][2900/3937] lr: 3.1652e-05 eta: 2:17:34 time: 0.6416 data_time: 0.0020 memory: 44138 loss: 0.3208 +2023/06/06 09:55:01 - mmengine - INFO - Epoch(train) [7][3000/3937] lr: 3.1346e-05 eta: 2:16:30 time: 0.6430 data_time: 0.0022 memory: 44138 loss: 0.3511 +2023/06/06 09:56:05 - mmengine - INFO - Epoch(train) [7][3100/3937] lr: 3.1041e-05 eta: 2:15:25 time: 0.6418 data_time: 0.0019 memory: 44138 loss: 0.3995 +2023/06/06 09:57:09 - mmengine - INFO - Epoch(train) [7][3200/3937] lr: 3.0738e-05 eta: 2:14:21 time: 0.6405 data_time: 0.0017 memory: 44138 loss: 0.3602 +2023/06/06 09:58:13 - mmengine - INFO - Epoch(train) [7][3300/3937] lr: 3.0437e-05 eta: 2:13:17 time: 0.6395 data_time: 0.0016 memory: 44138 loss: 0.3430 +2023/06/06 09:59:03 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_IF_lr1e-4_20230606_050006 +2023/06/06 09:59:18 - mmengine - INFO - Epoch(train) [7][3400/3937] lr: 3.0136e-05 eta: 2:12:12 time: 0.6418 data_time: 0.0016 memory: 44138 loss: 0.3421 +2023/06/06 10:00:22 - mmengine - INFO - Epoch(train) [7][3500/3937] lr: 2.9838e-05 eta: 2:11:08 time: 0.6427 data_time: 0.0015 memory: 44138 loss: 0.3353 +2023/06/06 10:01:26 - mmengine - INFO - Epoch(train) [7][3600/3937] lr: 2.9541e-05 eta: 2:10:04 time: 0.6429 data_time: 0.0015 memory: 44138 loss: 0.3391 +2023/06/06 10:02:30 - mmengine - INFO - Epoch(train) [7][3700/3937] lr: 2.9246e-05 eta: 2:08:59 time: 0.6408 data_time: 0.0017 memory: 44138 loss: 0.3578 +2023/06/06 10:03:34 - mmengine - INFO - Epoch(train) [7][3800/3937] lr: 2.8952e-05 eta: 2:07:55 time: 0.6400 data_time: 0.0016 memory: 44138 loss: 0.3137 +2023/06/06 10:04:38 - mmengine - INFO - Epoch(train) [7][3900/3937] lr: 2.8660e-05 eta: 2:06:51 time: 0.6452 data_time: 0.0016 memory: 44138 loss: 0.3143 +2023/06/06 10:05:02 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_IF_lr1e-4_20230606_050006 +2023/06/06 10:05:02 - mmengine - INFO - Saving checkpoint at 7 epochs +2023/06/06 10:06:33 - mmengine - INFO - Epoch(val) [7][57/57] accuracy/top1: 93.6293 single-label/precision_classwise: [97.43363952636719, 89.60853576660156] single-label/recall_classwise: [90.83391571044922, 97.06201171875] single-label/f1-score_classwise: [94.01810455322266, 93.18647003173828] data_time: 0.0161 time: 1.2589 +2023/06/06 10:07:40 - mmengine - INFO - Epoch(train) [8][ 100/3937] lr: 2.8263e-05 eta: 2:05:24 time: 0.6414 data_time: 0.0018 memory: 44138 loss: 0.3279 +2023/06/06 10:08:44 - mmengine - INFO - Epoch(train) [8][ 200/3937] lr: 2.7975e-05 eta: 2:04:19 time: 0.6421 data_time: 0.0024 memory: 44138 loss: 0.3484 +2023/06/06 10:09:48 - mmengine - INFO - Epoch(train) [8][ 300/3937] lr: 2.7689e-05 eta: 2:03:15 time: 0.6495 data_time: 0.0018 memory: 44138 loss: 0.3356 +2023/06/06 10:10:52 - mmengine - INFO - Epoch(train) [8][ 400/3937] lr: 2.7404e-05 eta: 2:02:11 time: 0.6396 data_time: 0.0016 memory: 44138 loss: 0.3365 +2023/06/06 10:11:18 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_IF_lr1e-4_20230606_050006 +2023/06/06 10:11:56 - mmengine - INFO - Epoch(train) [8][ 500/3937] lr: 2.7121e-05 eta: 2:01:07 time: 0.6401 data_time: 0.0017 memory: 44138 loss: 0.3527 +2023/06/06 10:13:00 - mmengine - INFO - Epoch(train) [8][ 600/3937] lr: 2.6840e-05 eta: 2:00:02 time: 0.6407 data_time: 0.0015 memory: 44138 loss: 0.3274 +2023/06/06 10:14:04 - mmengine - INFO - Epoch(train) [8][ 700/3937] lr: 2.6561e-05 eta: 1:58:58 time: 0.6378 data_time: 0.0016 memory: 44138 loss: 0.3499 +2023/06/06 10:15:08 - mmengine - INFO - Epoch(train) [8][ 800/3937] lr: 2.6284e-05 eta: 1:57:53 time: 0.6381 data_time: 0.0015 memory: 44138 loss: 0.3353 +2023/06/06 10:16:12 - mmengine - INFO - Epoch(train) [8][ 900/3937] lr: 2.6008e-05 eta: 1:56:49 time: 0.6378 data_time: 0.0030 memory: 44138 loss: 0.3310 +2023/06/06 10:17:16 - mmengine - INFO - Epoch(train) [8][1000/3937] lr: 2.5735e-05 eta: 1:55:45 time: 0.6385 data_time: 0.0026 memory: 44138 loss: 0.3441 +2023/06/06 10:18:19 - mmengine - INFO - Epoch(train) [8][1100/3937] lr: 2.5463e-05 eta: 1:54:40 time: 0.6370 data_time: 0.0016 memory: 44138 loss: 0.3304 +2023/06/06 10:19:23 - mmengine - INFO - Epoch(train) [8][1200/3937] lr: 2.5193e-05 eta: 1:53:36 time: 0.6368 data_time: 0.0017 memory: 44138 loss: 0.3501 +2023/06/06 10:20:27 - mmengine - INFO - Epoch(train) [8][1300/3937] lr: 2.4925e-05 eta: 1:52:31 time: 0.6371 data_time: 0.0017 memory: 44138 loss: 0.3493 +2023/06/06 10:21:31 - mmengine - INFO - Epoch(train) [8][1400/3937] lr: 2.4659e-05 eta: 1:51:27 time: 0.6362 data_time: 0.0017 memory: 44138 loss: 0.3444 +2023/06/06 10:21:57 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_IF_lr1e-4_20230606_050006 +2023/06/06 10:22:34 - mmengine - INFO - Epoch(train) [8][1500/3937] lr: 2.4394e-05 eta: 1:50:23 time: 0.6368 data_time: 0.0018 memory: 44138 loss: 0.3415 +2023/06/06 10:23:38 - mmengine - INFO - Epoch(train) [8][1600/3937] lr: 2.4132e-05 eta: 1:49:18 time: 0.6387 data_time: 0.0030 memory: 44138 loss: 0.3317 +2023/06/06 10:24:42 - mmengine - INFO - Epoch(train) [8][1700/3937] lr: 2.3872e-05 eta: 1:48:14 time: 0.6383 data_time: 0.0018 memory: 44138 loss: 0.3175 +2023/06/06 10:25:46 - mmengine - INFO - Epoch(train) [8][1800/3937] lr: 2.3613e-05 eta: 1:47:09 time: 0.6373 data_time: 0.0016 memory: 44138 loss: 0.3351 +2023/06/06 10:26:50 - mmengine - INFO - Epoch(train) [8][1900/3937] lr: 2.3357e-05 eta: 1:46:05 time: 0.6380 data_time: 0.0016 memory: 44138 loss: 0.3334 +2023/06/06 10:27:54 - mmengine - INFO - Epoch(train) [8][2000/3937] lr: 2.3103e-05 eta: 1:45:01 time: 0.6373 data_time: 0.0021 memory: 44138 loss: 0.3357 +2023/06/06 10:28:57 - mmengine - INFO - Epoch(train) [8][2100/3937] lr: 2.2851e-05 eta: 1:43:56 time: 0.6381 data_time: 0.0026 memory: 44138 loss: 0.3277 +2023/06/06 10:30:01 - mmengine - INFO - Epoch(train) [8][2200/3937] lr: 2.2600e-05 eta: 1:42:52 time: 0.6372 data_time: 0.0017 memory: 44138 loss: 0.3512 +2023/06/06 10:31:05 - mmengine - INFO - Epoch(train) [8][2300/3937] lr: 2.2352e-05 eta: 1:41:48 time: 0.6360 data_time: 0.0021 memory: 44138 loss: 0.3321 +2023/06/06 10:32:09 - mmengine - INFO - Epoch(train) [8][2400/3937] lr: 2.2106e-05 eta: 1:40:43 time: 0.6363 data_time: 0.0018 memory: 44138 loss: 0.3401 +2023/06/06 10:32:35 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_IF_lr1e-4_20230606_050006 +2023/06/06 10:33:12 - mmengine - INFO - Epoch(train) [8][2500/3937] lr: 2.1862e-05 eta: 1:39:39 time: 0.6382 data_time: 0.0018 memory: 44138 loss: 0.3393 +2023/06/06 10:34:16 - mmengine - INFO - Epoch(train) [8][2600/3937] lr: 2.1620e-05 eta: 1:38:34 time: 0.6372 data_time: 0.0018 memory: 44138 loss: 0.3494 +2023/06/06 10:35:20 - mmengine - INFO - Epoch(train) [8][2700/3937] lr: 2.1380e-05 eta: 1:37:30 time: 0.6456 data_time: 0.0018 memory: 44138 loss: 0.3472 +2023/06/06 10:36:24 - mmengine - INFO - Epoch(train) [8][2800/3937] lr: 2.1143e-05 eta: 1:36:26 time: 0.6369 data_time: 0.0017 memory: 44138 loss: 0.3159 +2023/06/06 10:37:27 - mmengine - INFO - Epoch(train) [8][2900/3937] lr: 2.0907e-05 eta: 1:35:21 time: 0.6377 data_time: 0.0016 memory: 44138 loss: 0.3737 +2023/06/06 10:38:31 - mmengine - INFO - Epoch(train) [8][3000/3937] lr: 2.0674e-05 eta: 1:34:17 time: 0.6369 data_time: 0.0016 memory: 44138 loss: 0.3234 +2023/06/06 10:39:35 - mmengine - INFO - Epoch(train) [8][3100/3937] lr: 2.0443e-05 eta: 1:33:13 time: 0.6370 data_time: 0.0016 memory: 44138 loss: 0.3398 +2023/06/06 10:40:39 - mmengine - INFO - Epoch(train) [8][3200/3937] lr: 2.0214e-05 eta: 1:32:08 time: 0.6368 data_time: 0.0021 memory: 44138 loss: 0.3354 +2023/06/06 10:41:42 - mmengine - INFO - Epoch(train) [8][3300/3937] lr: 1.9987e-05 eta: 1:31:04 time: 0.6389 data_time: 0.0026 memory: 44138 loss: 0.3449 +2023/06/06 10:42:46 - mmengine - INFO - Epoch(train) [8][3400/3937] lr: 1.9763e-05 eta: 1:30:00 time: 0.6366 data_time: 0.0019 memory: 44138 loss: 0.3225 +2023/06/06 10:43:12 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_IF_lr1e-4_20230606_050006 +2023/06/06 10:43:50 - mmengine - INFO - Epoch(train) [8][3500/3937] lr: 1.9541e-05 eta: 1:28:55 time: 0.6378 data_time: 0.0015 memory: 44138 loss: 0.3325 +2023/06/06 10:44:54 - mmengine - INFO - Epoch(train) [8][3600/3937] lr: 1.9321e-05 eta: 1:27:51 time: 0.6369 data_time: 0.0017 memory: 44138 loss: 0.3004 +2023/06/06 10:45:57 - mmengine - INFO - Epoch(train) [8][3700/3937] lr: 1.9103e-05 eta: 1:26:47 time: 0.6371 data_time: 0.0015 memory: 44138 loss: 0.3464 +2023/06/06 10:47:01 - mmengine - INFO - Epoch(train) [8][3800/3937] lr: 1.8888e-05 eta: 1:25:42 time: 0.6450 data_time: 0.0020 memory: 44138 loss: 0.3485 +2023/06/06 10:48:05 - mmengine - INFO - Epoch(train) [8][3900/3937] lr: 1.8675e-05 eta: 1:24:38 time: 0.6373 data_time: 0.0021 memory: 44138 loss: 0.3407 +2023/06/06 10:48:29 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_IF_lr1e-4_20230606_050006 +2023/06/06 10:48:29 - mmengine - INFO - Saving checkpoint at 8 epochs +2023/06/06 10:49:59 - mmengine - INFO - Epoch(val) [8][57/57] accuracy/top1: 94.4400 single-label/precision_classwise: [96.95391082763672, 91.63105773925781] single-label/recall_classwise: [92.8287353515625, 96.41860961914062] single-label/f1-score_classwise: [94.84648895263672, 93.9638900756836] data_time: 0.0144 time: 1.2575 +2023/06/06 10:51:06 - mmengine - INFO - Epoch(train) [9][ 100/3937] lr: 1.8386e-05 eta: 1:23:11 time: 0.6380 data_time: 0.0019 memory: 44138 loss: 0.3471 +2023/06/06 10:52:10 - mmengine - INFO - Epoch(train) [9][ 200/3937] lr: 1.8179e-05 eta: 1:22:06 time: 0.6379 data_time: 0.0024 memory: 44138 loss: 0.3632 +2023/06/06 10:53:13 - mmengine - INFO - Epoch(train) [9][ 300/3937] lr: 1.7974e-05 eta: 1:21:02 time: 0.6378 data_time: 0.0017 memory: 44138 loss: 0.3465 +2023/06/06 10:54:17 - mmengine - INFO - Epoch(train) [9][ 400/3937] lr: 1.7771e-05 eta: 1:19:58 time: 0.6380 data_time: 0.0018 memory: 44138 loss: 0.3592 +2023/06/06 10:55:21 - mmengine - INFO - Epoch(train) [9][ 500/3937] lr: 1.7570e-05 eta: 1:18:53 time: 0.6379 data_time: 0.0017 memory: 44138 loss: 0.3359 +2023/06/06 10:55:24 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_IF_lr1e-4_20230606_050006 +2023/06/06 10:56:25 - mmengine - INFO - Epoch(train) [9][ 600/3937] lr: 1.7372e-05 eta: 1:17:49 time: 0.6384 data_time: 0.0018 memory: 44138 loss: 0.3351 +2023/06/06 10:57:29 - mmengine - INFO - Epoch(train) [9][ 700/3937] lr: 1.7176e-05 eta: 1:16:45 time: 0.6380 data_time: 0.0019 memory: 44138 loss: 0.3564 +2023/06/06 10:58:32 - mmengine - INFO - Epoch(train) [9][ 800/3937] lr: 1.6983e-05 eta: 1:15:41 time: 0.6378 data_time: 0.0016 memory: 44138 loss: 0.3528 +2023/06/06 10:59:36 - mmengine - INFO - Epoch(train) [9][ 900/3937] lr: 1.6792e-05 eta: 1:14:36 time: 0.6371 data_time: 0.0020 memory: 44138 loss: 0.3167 +2023/06/06 11:00:40 - mmengine - INFO - Epoch(train) [9][1000/3937] lr: 1.6604e-05 eta: 1:13:32 time: 0.6367 data_time: 0.0025 memory: 44138 loss: 0.3180 +2023/06/06 11:01:44 - mmengine - INFO - Epoch(train) [9][1100/3937] lr: 1.6418e-05 eta: 1:12:28 time: 0.6368 data_time: 0.0016 memory: 44138 loss: 0.3418 +2023/06/06 11:02:47 - mmengine - INFO - Epoch(train) [9][1200/3937] lr: 1.6234e-05 eta: 1:11:23 time: 0.6368 data_time: 0.0017 memory: 44138 loss: 0.3467 +2023/06/06 11:03:51 - mmengine - INFO - Epoch(train) [9][1300/3937] lr: 1.6053e-05 eta: 1:10:19 time: 0.6372 data_time: 0.0016 memory: 44138 loss: 0.3446 +2023/06/06 11:04:55 - mmengine - INFO - Epoch(train) [9][1400/3937] lr: 1.5874e-05 eta: 1:09:15 time: 0.6367 data_time: 0.0015 memory: 44138 loss: 0.3337 +2023/06/06 11:05:58 - mmengine - INFO - Epoch(train) [9][1500/3937] lr: 1.5698e-05 eta: 1:08:11 time: 0.6364 data_time: 0.0016 memory: 44138 loss: 0.3353 +2023/06/06 11:06:01 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_IF_lr1e-4_20230606_050006 +2023/06/06 11:07:02 - mmengine - INFO - Epoch(train) [9][1600/3937] lr: 1.5524e-05 eta: 1:07:06 time: 0.6365 data_time: 0.0017 memory: 44138 loss: 0.3292 +2023/06/06 11:08:06 - mmengine - INFO - Epoch(train) [9][1700/3937] lr: 1.5353e-05 eta: 1:06:02 time: 0.6375 data_time: 0.0019 memory: 44138 loss: 0.3220 +2023/06/06 11:09:10 - mmengine - INFO - Epoch(train) [9][1800/3937] lr: 1.5185e-05 eta: 1:04:58 time: 0.6366 data_time: 0.0019 memory: 44138 loss: 0.3659 +2023/06/06 11:10:13 - mmengine - INFO - Epoch(train) [9][1900/3937] lr: 1.5019e-05 eta: 1:03:54 time: 0.6365 data_time: 0.0017 memory: 44138 loss: 0.3399 +2023/06/06 11:11:17 - mmengine - INFO - Epoch(train) [9][2000/3937] lr: 1.4855e-05 eta: 1:02:49 time: 0.6376 data_time: 0.0018 memory: 44138 loss: 0.3403 +2023/06/06 11:12:21 - mmengine - INFO - Epoch(train) [9][2100/3937] lr: 1.4694e-05 eta: 1:01:45 time: 0.6372 data_time: 0.0019 memory: 44138 loss: 0.3265 +2023/06/06 11:13:25 - mmengine - INFO - Epoch(train) [9][2200/3937] lr: 1.4536e-05 eta: 1:00:41 time: 0.6368 data_time: 0.0017 memory: 44138 loss: 0.3236 +2023/06/06 11:14:29 - mmengine - INFO - Epoch(train) [9][2300/3937] lr: 1.4380e-05 eta: 0:59:37 time: 0.6374 data_time: 0.0018 memory: 44138 loss: 0.3228 +2023/06/06 11:15:32 - mmengine - INFO - Epoch(train) [9][2400/3937] lr: 1.4227e-05 eta: 0:58:32 time: 0.6372 data_time: 0.0018 memory: 44138 loss: 0.3449 +2023/06/06 11:16:36 - mmengine - INFO - Epoch(train) [9][2500/3937] lr: 1.4076e-05 eta: 0:57:28 time: 0.6375 data_time: 0.0014 memory: 44138 loss: 0.3215 +2023/06/06 11:16:39 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_IF_lr1e-4_20230606_050006 +2023/06/06 11:17:40 - mmengine - INFO - Epoch(train) [9][2600/3937] lr: 1.3928e-05 eta: 0:56:24 time: 0.6371 data_time: 0.0015 memory: 44138 loss: 0.3619 +2023/06/06 11:18:44 - mmengine - INFO - Epoch(train) [9][2700/3937] lr: 1.3783e-05 eta: 0:55:20 time: 0.6370 data_time: 0.0016 memory: 44138 loss: 0.3587 +2023/06/06 11:19:47 - mmengine - INFO - Epoch(train) [9][2800/3937] lr: 1.3640e-05 eta: 0:54:15 time: 0.6373 data_time: 0.0020 memory: 44138 loss: 0.3619 +2023/06/06 11:20:51 - mmengine - INFO - Epoch(train) [9][2900/3937] lr: 1.3500e-05 eta: 0:53:11 time: 0.6366 data_time: 0.0015 memory: 44138 loss: 0.3353 +2023/06/06 11:21:55 - mmengine - INFO - Epoch(train) [9][3000/3937] lr: 1.3362e-05 eta: 0:52:07 time: 0.6360 data_time: 0.0017 memory: 44138 loss: 0.3545 +2023/06/06 11:22:58 - mmengine - INFO - Epoch(train) [9][3100/3937] lr: 1.3227e-05 eta: 0:51:03 time: 0.6367 data_time: 0.0017 memory: 44138 loss: 0.3315 +2023/06/06 11:24:02 - mmengine - INFO - Epoch(train) [9][3200/3937] lr: 1.3095e-05 eta: 0:49:58 time: 0.6364 data_time: 0.0015 memory: 44138 loss: 0.3482 +2023/06/06 11:25:06 - mmengine - INFO - Epoch(train) [9][3300/3937] lr: 1.2966e-05 eta: 0:48:54 time: 0.6360 data_time: 0.0016 memory: 44138 loss: 0.3190 +2023/06/06 11:26:09 - mmengine - INFO - Epoch(train) [9][3400/3937] lr: 1.2839e-05 eta: 0:47:50 time: 0.6377 data_time: 0.0015 memory: 44138 loss: 0.3432 +2023/06/06 11:27:13 - mmengine - INFO - Epoch(train) [9][3500/3937] lr: 1.2715e-05 eta: 0:46:46 time: 0.6384 data_time: 0.0015 memory: 44138 loss: 0.3670 +2023/06/06 11:27:16 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_IF_lr1e-4_20230606_050006 +2023/06/06 11:28:17 - mmengine - INFO - Epoch(train) [9][3600/3937] lr: 1.2593e-05 eta: 0:45:42 time: 0.6370 data_time: 0.0023 memory: 44138 loss: 0.3338 +2023/06/06 11:29:21 - mmengine - INFO - Epoch(train) [9][3700/3937] lr: 1.2474e-05 eta: 0:44:37 time: 0.6370 data_time: 0.0016 memory: 44138 loss: 0.3443 +2023/06/06 11:30:24 - mmengine - INFO - Epoch(train) [9][3800/3937] lr: 1.2358e-05 eta: 0:43:33 time: 0.6361 data_time: 0.0015 memory: 44138 loss: 0.3388 +2023/06/06 11:31:28 - mmengine - INFO - Epoch(train) [9][3900/3937] lr: 1.2245e-05 eta: 0:42:29 time: 0.6371 data_time: 0.0015 memory: 44138 loss: 0.3283 +2023/06/06 11:31:52 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_IF_lr1e-4_20230606_050006 +2023/06/06 11:31:52 - mmengine - INFO - Saving checkpoint at 9 epochs +2023/06/06 11:33:23 - mmengine - INFO - Epoch(val) [9][57/57] accuracy/top1: 94.7497 single-label/precision_classwise: [96.73274993896484, 92.4815444946289] single-label/recall_classwise: [93.63676452636719, 96.11627960205078] single-label/f1-score_classwise: [95.15959167480469, 94.2638931274414] data_time: 0.0181 time: 1.2612 +2023/06/06 11:34:29 - mmengine - INFO - Epoch(train) [10][ 100/3937] lr: 1.2094e-05 eta: 0:41:01 time: 0.6379 data_time: 0.0018 memory: 44138 loss: 0.3731 +2023/06/06 11:35:33 - mmengine - INFO - Epoch(train) [10][ 200/3937] lr: 1.1987e-05 eta: 0:39:57 time: 0.6375 data_time: 0.0018 memory: 44138 loss: 0.3633 +2023/06/06 11:36:37 - mmengine - INFO - Epoch(train) [10][ 300/3937] lr: 1.1883e-05 eta: 0:38:53 time: 0.6378 data_time: 0.0018 memory: 44138 loss: 0.3484 +2023/06/06 11:37:41 - mmengine - INFO - Epoch(train) [10][ 400/3937] lr: 1.1781e-05 eta: 0:37:49 time: 0.6419 data_time: 0.0014 memory: 44138 loss: 0.3425 +2023/06/06 11:38:44 - mmengine - INFO - Epoch(train) [10][ 500/3937] lr: 1.1683e-05 eta: 0:36:45 time: 0.6377 data_time: 0.0021 memory: 44138 loss: 0.3399 +2023/06/06 11:39:27 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_IF_lr1e-4_20230606_050006 +2023/06/06 11:39:48 - mmengine - INFO - Epoch(train) [10][ 600/3937] lr: 1.1587e-05 eta: 0:35:40 time: 0.6365 data_time: 0.0016 memory: 44138 loss: 0.3612 +2023/06/06 11:40:52 - mmengine - INFO - Epoch(train) [10][ 700/3937] lr: 1.1494e-05 eta: 0:34:36 time: 0.6369 data_time: 0.0016 memory: 44138 loss: 0.3209 +2023/06/06 11:41:56 - mmengine - INFO - Epoch(train) [10][ 800/3937] lr: 1.1403e-05 eta: 0:33:32 time: 0.6371 data_time: 0.0015 memory: 44138 loss: 0.3721 +2023/06/06 11:42:59 - mmengine - INFO - Epoch(train) [10][ 900/3937] lr: 1.1316e-05 eta: 0:32:28 time: 0.6368 data_time: 0.0016 memory: 44138 loss: 0.3169 +2023/06/06 11:44:03 - mmengine - INFO - Epoch(train) [10][1000/3937] lr: 1.1231e-05 eta: 0:31:24 time: 0.6372 data_time: 0.0015 memory: 44138 loss: 0.3270 +2023/06/06 11:45:07 - mmengine - INFO - Epoch(train) [10][1100/3937] lr: 1.1149e-05 eta: 0:30:19 time: 0.6380 data_time: 0.0015 memory: 44138 loss: 0.3336 +2023/06/06 11:46:11 - mmengine - INFO - Epoch(train) [10][1200/3937] lr: 1.1070e-05 eta: 0:29:15 time: 0.6376 data_time: 0.0016 memory: 44138 loss: 0.3387 +2023/06/06 11:47:15 - mmengine - INFO - Epoch(train) [10][1300/3937] lr: 1.0993e-05 eta: 0:28:11 time: 0.6385 data_time: 0.0018 memory: 44138 loss: 0.3431 +2023/06/06 11:48:18 - mmengine - INFO - Epoch(train) [10][1400/3937] lr: 1.0920e-05 eta: 0:27:07 time: 0.6373 data_time: 0.0019 memory: 44138 loss: 0.3487 +2023/06/06 11:49:22 - mmengine - INFO - Epoch(train) [10][1500/3937] lr: 1.0849e-05 eta: 0:26:03 time: 0.6379 data_time: 0.0016 memory: 44138 loss: 0.3504 +2023/06/06 11:50:05 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_IF_lr1e-4_20230606_050006 +2023/06/06 11:50:26 - mmengine - INFO - Epoch(train) [10][1600/3937] lr: 1.0781e-05 eta: 0:24:59 time: 0.6369 data_time: 0.0016 memory: 44138 loss: 0.3206 +2023/06/06 11:51:30 - mmengine - INFO - Epoch(train) [10][1700/3937] lr: 1.0716e-05 eta: 0:23:54 time: 0.6376 data_time: 0.0015 memory: 44138 loss: 0.3239 +2023/06/06 11:52:33 - mmengine - INFO - Epoch(train) [10][1800/3937] lr: 1.0653e-05 eta: 0:22:50 time: 0.6372 data_time: 0.0015 memory: 44138 loss: 0.3422 +2023/06/06 11:53:37 - mmengine - INFO - Epoch(train) [10][1900/3937] lr: 1.0594e-05 eta: 0:21:46 time: 0.6374 data_time: 0.0017 memory: 44138 loss: 0.3440 +2023/06/06 11:54:41 - mmengine - INFO - Epoch(train) [10][2000/3937] lr: 1.0537e-05 eta: 0:20:42 time: 0.6368 data_time: 0.0023 memory: 44138 loss: 0.3332 +2023/06/06 11:55:45 - mmengine - INFO - Epoch(train) [10][2100/3937] lr: 1.0483e-05 eta: 0:19:38 time: 0.6377 data_time: 0.0016 memory: 44138 loss: 0.3279 +2023/06/06 11:56:48 - mmengine - INFO - Epoch(train) [10][2200/3937] lr: 1.0432e-05 eta: 0:18:34 time: 0.6364 data_time: 0.0015 memory: 44138 loss: 0.3528 +2023/06/06 11:57:52 - mmengine - INFO - Epoch(train) [10][2300/3937] lr: 1.0384e-05 eta: 0:17:29 time: 0.6370 data_time: 0.0017 memory: 44138 loss: 0.3226 +2023/06/06 11:58:56 - mmengine - INFO - Epoch(train) [10][2400/3937] lr: 1.0338e-05 eta: 0:16:25 time: 0.6369 data_time: 0.0017 memory: 44138 loss: 0.3160 +2023/06/06 12:00:00 - mmengine - INFO - Epoch(train) [10][2500/3937] lr: 1.0296e-05 eta: 0:15:21 time: 0.6380 data_time: 0.0016 memory: 44138 loss: 0.3655 +2023/06/06 12:00:42 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_IF_lr1e-4_20230606_050006 +2023/06/06 12:01:03 - mmengine - INFO - Epoch(train) [10][2600/3937] lr: 1.0256e-05 eta: 0:14:17 time: 0.6383 data_time: 0.0017 memory: 44138 loss: 0.3382 +2023/06/06 12:02:07 - mmengine - INFO - Epoch(train) [10][2700/3937] lr: 1.0219e-05 eta: 0:13:13 time: 0.6372 data_time: 0.0013 memory: 44138 loss: 0.3508 +2023/06/06 12:03:11 - mmengine - INFO - Epoch(train) [10][2800/3937] lr: 1.0185e-05 eta: 0:12:09 time: 0.6375 data_time: 0.0014 memory: 44138 loss: 0.3306 +2023/06/06 12:04:15 - mmengine - INFO - Epoch(train) [10][2900/3937] lr: 1.0154e-05 eta: 0:11:05 time: 0.6382 data_time: 0.0016 memory: 44138 loss: 0.3464 +2023/06/06 12:05:19 - mmengine - INFO - Epoch(train) [10][3000/3937] lr: 1.0126e-05 eta: 0:10:00 time: 0.6375 data_time: 0.0016 memory: 44138 loss: 0.3515 +2023/06/06 12:06:22 - mmengine - INFO - Epoch(train) [10][3100/3937] lr: 1.0101e-05 eta: 0:08:56 time: 0.6369 data_time: 0.0018 memory: 44138 loss: 0.3582 +2023/06/06 12:07:26 - mmengine - INFO - Epoch(train) [10][3200/3937] lr: 1.0078e-05 eta: 0:07:52 time: 0.6378 data_time: 0.0019 memory: 44138 loss: 0.3580 +2023/06/06 12:08:30 - mmengine - INFO - Epoch(train) [10][3300/3937] lr: 1.0058e-05 eta: 0:06:48 time: 0.6377 data_time: 0.0016 memory: 44138 loss: 0.3149 +2023/06/06 12:09:34 - mmengine - INFO - Epoch(train) [10][3400/3937] lr: 1.0041e-05 eta: 0:05:44 time: 0.6372 data_time: 0.0022 memory: 44138 loss: 0.3470 +2023/06/06 12:10:37 - mmengine - INFO - Epoch(train) [10][3500/3937] lr: 1.0027e-05 eta: 0:04:40 time: 0.6371 data_time: 0.0019 memory: 44138 loss: 0.3348 +2023/06/06 12:11:27 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_IF_lr1e-4_20230606_050006 +2023/06/06 12:11:52 - mmengine - INFO - Epoch(train) [10][3600/3937] lr: 1.0016e-05 eta: 0:03:36 time: 0.6461 data_time: 0.0021 memory: 44138 loss: 0.3517 +2023/06/06 12:12:57 - mmengine - INFO - Epoch(train) [10][3700/3937] lr: 1.0008e-05 eta: 0:02:32 time: 0.6565 data_time: 0.0018 memory: 44138 loss: 0.3443 +2023/06/06 12:14:03 - mmengine - INFO - Epoch(train) [10][3800/3937] lr: 1.0003e-05 eta: 0:01:27 time: 0.6375 data_time: 0.0017 memory: 44138 loss: 0.3186 +2023/06/06 12:15:07 - mmengine - INFO - Epoch(train) [10][3900/3937] lr: 1.0000e-05 eta: 0:00:23 time: 0.6365 data_time: 0.0016 memory: 44138 loss: 0.3328 +2023/06/06 12:15:30 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_IF_lr1e-4_20230606_050006 +2023/06/06 12:15:30 - mmengine - INFO - Saving checkpoint at 10 epochs +2023/06/06 12:17:01 - mmengine - INFO - Epoch(val) [10][57/57] accuracy/top1: 94.7705 single-label/precision_classwise: [96.6064224243164, 92.6566390991211] single-label/recall_classwise: [93.80720520019531, 95.9534912109375] single-label/f1-score_classwise: [95.18624114990234, 94.2762451171875] data_time: 0.0153 time: 1.2585 diff --git a/clip_large_pretrain_4x256_IF_lr1e-4/20230606_050006/vis_data/20230606_050006.json b/clip_large_pretrain_4x256_IF_lr1e-4/20230606_050006/vis_data/20230606_050006.json new file mode 100644 index 0000000000000000000000000000000000000000..4e8cd5172c7d8dc61dbd21b8f83e0ab59b30df7c --- /dev/null +++ b/clip_large_pretrain_4x256_IF_lr1e-4/20230606_050006/vis_data/20230606_050006.json @@ -0,0 +1,400 @@ +{"lr": 9.999859583040624e-05, "data_time": 0.0016723871231079102, "loss": 0.512164306640625, "time": 0.6354817628860474, "epoch": 1, "memory": 44139, "step": 100} +{"lr": 9.999432653394512e-05, "data_time": 0.001458597183227539, "loss": 0.49367055892944334, "time": 0.6365986347198487, "epoch": 1, "memory": 44139, "step": 200} +{"lr": 9.998719223919393e-05, "data_time": 0.0015254497528076171, "loss": 0.47274911403656006, "time": 0.6358645677566528, "epoch": 1, "memory": 44139, "step": 300} +{"lr": 9.997719340042649e-05, "data_time": 0.0016599655151367187, "loss": 0.45169744491577146, "time": 0.6360260486602783, "epoch": 1, "memory": 44139, "step": 400} +{"lr": 9.996433065431542e-05, "data_time": 0.0014626264572143554, "loss": 0.445109036564827, "time": 0.6356816291809082, "epoch": 1, "memory": 44139, "step": 500} +{"lr": 9.994860481989157e-05, "data_time": 0.0015146493911743163, "loss": 0.42883872985839844, "time": 0.6361032485961914, "epoch": 1, "memory": 44139, "step": 600} +{"lr": 9.993001689849214e-05, "data_time": 0.0014402151107788086, "loss": 0.45372450947761533, "time": 0.6367725849151611, "epoch": 1, "memory": 44139, "step": 700} +{"lr": 9.990856807369658e-05, "data_time": 0.0015191078186035157, "loss": 0.4422683477401733, "time": 0.636231541633606, "epoch": 1, "memory": 44139, "step": 800} +{"lr": 9.98842597112514e-05, "data_time": 0.001710963249206543, "loss": 0.4624675452709198, "time": 0.6360510587692261, "epoch": 1, "memory": 44139, "step": 900} +{"lr": 9.985709335898336e-05, "data_time": 0.0016065359115600586, "loss": 0.4689463973045349, "time": 0.6368230581283569, "epoch": 1, "memory": 44139, "step": 1000} +{"lr": 9.982707074670047e-05, "data_time": 0.0016260623931884765, "loss": 0.43320480585098264, "time": 0.6370084762573243, "epoch": 1, "memory": 44139, "step": 1100} +{"lr": 9.979419378608245e-05, "data_time": 0.0014093875885009765, "loss": 0.4505138874053955, "time": 0.6361888647079468, "epoch": 1, "memory": 44139, "step": 1200} +{"lr": 9.975846457055832e-05, "data_time": 0.0014659643173217773, "loss": 0.4286786615848541, "time": 0.6366539239883423, "epoch": 1, "memory": 44139, "step": 1300} +{"lr": 9.97198853751737e-05, "data_time": 0.001408529281616211, "loss": 0.434794282913208, "time": 0.6369958639144897, "epoch": 1, "memory": 44139, "step": 1400} +{"lr": 9.967845865644554e-05, "data_time": 0.0014293432235717774, "loss": 0.41036694049835204, "time": 0.6361128330230713, "epoch": 1, "memory": 44139, "step": 1500} +{"lr": 9.963418705220597e-05, "data_time": 0.001466655731201172, "loss": 0.4314510613679886, "time": 0.6372191905975342, "epoch": 1, "memory": 44139, "step": 1600} +{"lr": 9.958707338143413e-05, "data_time": 0.0015425443649291991, "loss": 0.38765864074230194, "time": 0.6372380971908569, "epoch": 1, "memory": 44139, "step": 1700} +{"lr": 9.953712064407689e-05, "data_time": 0.0015309810638427734, "loss": 0.4207657933235168, "time": 0.636822509765625, "epoch": 1, "memory": 44139, "step": 1800} +{"lr": 9.948433202085752e-05, "data_time": 0.0015765666961669923, "loss": 0.38641782104969025, "time": 0.6369469642639161, "epoch": 1, "memory": 44139, "step": 1900} +{"lr": 9.942871087307363e-05, "data_time": 0.0017748832702636718, "loss": 0.43550218641757965, "time": 0.6362056493759155, "epoch": 1, "memory": 44139, "step": 2000} +{"lr": 9.937026074238262e-05, "data_time": 0.001464986801147461, "loss": 0.4059758365154266, "time": 0.6361548900604248, "epoch": 1, "memory": 44139, "step": 2100} +{"lr": 9.930898535057657e-05, "data_time": 0.0015139818191528321, "loss": 0.416530105471611, "time": 0.6361959457397461, "epoch": 1, "memory": 44139, "step": 2200} +{"lr": 9.924488859934492e-05, "data_time": 0.0017367839813232423, "loss": 0.3778152257204056, "time": 0.6361271858215332, "epoch": 1, "memory": 44139, "step": 2300} +{"lr": 9.917797457002633e-05, "data_time": 0.002050018310546875, "loss": 0.38874633610248566, "time": 0.6379516124725342, "epoch": 1, "memory": 44139, "step": 2400} +{"lr": 9.910824752334866e-05, "data_time": 0.0015123844146728515, "loss": 0.39862638115882876, "time": 0.6360560178756713, "epoch": 1, "memory": 44139, "step": 2500} +{"lr": 9.903571189915766e-05, "data_time": 0.0017220258712768554, "loss": 0.38024311065673827, "time": 0.6397989511489868, "epoch": 1, "memory": 44139, "step": 2600} +{"lr": 9.896037231613424e-05, "data_time": 0.001457810401916504, "loss": 0.4019357919692993, "time": 0.6363687753677368, "epoch": 1, "memory": 44139, "step": 2700} +{"lr": 9.888223357150049e-05, "data_time": 0.0013912439346313477, "loss": 0.3906904995441437, "time": 0.6358689785003662, "epoch": 1, "memory": 44139, "step": 2800} +{"lr": 9.880130064071408e-05, "data_time": 0.001487565040588379, "loss": 0.3792046785354614, "time": 0.6360376596450805, "epoch": 1, "memory": 44139, "step": 2900} +{"lr": 9.87175786771517e-05, "data_time": 0.001390385627746582, "loss": 0.36773664355278013, "time": 0.6364351511001587, "epoch": 1, "memory": 44139, "step": 3000} +{"lr": 9.863107301178059e-05, "data_time": 0.0017130613327026368, "loss": 0.4086927860975266, "time": 0.636230754852295, "epoch": 1, "memory": 44139, "step": 3100} +{"lr": 9.854178915281925e-05, "data_time": 0.0015178680419921874, "loss": 0.4124324083328247, "time": 0.6347718477249146, "epoch": 1, "memory": 44139, "step": 3200} +{"lr": 9.84497327853868e-05, "data_time": 0.0015394926071166993, "loss": 0.38047824501991273, "time": 0.6361316919326783, "epoch": 1, "memory": 44139, "step": 3300} +{"lr": 9.83549097711407e-05, "data_time": 0.0015110731124877929, "loss": 0.38240416944026945, "time": 0.6365596532821656, "epoch": 1, "memory": 44139, "step": 3400} +{"lr": 9.825732614790392e-05, "data_time": 0.0014199256896972657, "loss": 0.37611470818519593, "time": 0.6370939493179322, "epoch": 1, "memory": 44139, "step": 3500} +{"lr": 9.815698812928018e-05, "data_time": 0.0015276193618774414, "loss": 0.3861346751451492, "time": 0.6377681970596314, "epoch": 1, "memory": 44139, "step": 3600} +{"lr": 9.805390210425831e-05, "data_time": 0.0015616416931152344, "loss": 0.38390754759311674, "time": 0.6380697011947631, "epoch": 1, "memory": 44139, "step": 3700} +{"lr": 9.794807463680567e-05, "data_time": 0.0015166997909545898, "loss": 0.38078379034996035, "time": 0.6371499061584472, "epoch": 1, "memory": 44139, "step": 3800} +{"lr": 9.783951246544974e-05, "data_time": 0.0013837337493896485, "loss": 0.37763476371765137, "time": 0.636236023902893, "epoch": 1, "memory": 44139, "step": 3900} +{"accuracy/top1": 80.48085021972656, "data_time": 0.025140733049626936, "time": 1.320618048048856, "step": 1} +{"lr": 9.768635527139677e-05, "data_time": 0.0016125202178955077, "loss": 0.3624943017959595, "time": 0.6393807649612426, "epoch": 2, "memory": 44139, "step": 4037} +{"lr": 9.75713397738286e-05, "data_time": 0.0013150930404663085, "loss": 0.35947518050670624, "time": 0.6368603467941284, "epoch": 2, "memory": 44138, "step": 4137} +{"lr": 9.745361356082025e-05, "data_time": 0.0014085769653320312, "loss": 0.3616087704896927, "time": 0.6367685794830322, "epoch": 2, "memory": 44138, "step": 4237} +{"lr": 9.733318412854784e-05, "data_time": 0.0013958930969238282, "loss": 0.3790826857089996, "time": 0.6389569044113159, "epoch": 2, "memory": 44138, "step": 4337} +{"lr": 9.721005914531412e-05, "data_time": 0.0014826536178588867, "loss": 0.3935104191303253, "time": 0.6482341766357422, "epoch": 2, "memory": 44138, "step": 4437} +{"lr": 9.708424645106017e-05, "data_time": 0.0014542818069458007, "loss": 0.3580289512872696, "time": 0.6364399433135987, "epoch": 2, "memory": 44138, "step": 4537} +{"lr": 9.695575405686603e-05, "data_time": 0.0015553951263427735, "loss": 0.3991315305233002, "time": 0.6370989322662354, "epoch": 2, "memory": 44138, "step": 4637} +{"lr": 9.682459014444089e-05, "data_time": 0.0014152288436889648, "loss": 0.3924710273742676, "time": 0.6358236074447632, "epoch": 2, "memory": 44138, "step": 4737} +{"lr": 9.669076306560176e-05, "data_time": 0.0014875173568725587, "loss": 0.3728522062301636, "time": 0.638262152671814, "epoch": 2, "memory": 44138, "step": 4837} +{"lr": 9.655428134174183e-05, "data_time": 0.0015720367431640626, "loss": 0.32827370166778563, "time": 0.6390895128250123, "epoch": 2, "memory": 44138, "step": 4937} +{"lr": 9.641515366328807e-05, "data_time": 0.0014989137649536132, "loss": 0.38309581875801085, "time": 0.6372867584228515, "epoch": 2, "memory": 44138, "step": 5037} +{"lr": 9.627338888914757e-05, "data_time": 0.0014197111129760742, "loss": 0.389060977101326, "time": 0.6368171691894531, "epoch": 2, "memory": 44138, "step": 5137} +{"lr": 9.612899604614378e-05, "data_time": 0.0013750791549682617, "loss": 0.37170601189136504, "time": 0.6360954284667969, "epoch": 2, "memory": 44138, "step": 5237} +{"lr": 9.598198432844131e-05, "data_time": 0.0014049768447875976, "loss": 0.36447584331035615, "time": 0.6362692594528199, "epoch": 2, "memory": 44138, "step": 5337} +{"lr": 9.583236309696081e-05, "data_time": 0.001524806022644043, "loss": 0.3516468346118927, "time": 0.6364728450775147, "epoch": 2, "memory": 44138, "step": 5437} +{"lr": 9.568014187878279e-05, "data_time": 0.001523447036743164, "loss": 0.3908290058374405, "time": 0.6365160942077637, "epoch": 2, "memory": 44138, "step": 5537} +{"lr": 9.552533036654103e-05, "data_time": 0.001683664321899414, "loss": 0.33294848799705506, "time": 0.6374480962753296, "epoch": 2, "memory": 44138, "step": 5637} +{"lr": 9.536793841780541e-05, "data_time": 0.0015509366989135743, "loss": 0.3408015429973602, "time": 0.6381877183914184, "epoch": 2, "memory": 44138, "step": 5737} +{"lr": 9.52079760544543e-05, "data_time": 0.0017657995223999024, "loss": 0.3576482445001602, "time": 0.6383677005767823, "epoch": 2, "memory": 44138, "step": 5837} +{"lr": 9.504545346203619e-05, "data_time": 0.0017487049102783204, "loss": 0.3597940683364868, "time": 0.6362926721572876, "epoch": 2, "memory": 44138, "step": 5937} +{"lr": 9.488038098912122e-05, "data_time": 0.0016869544982910157, "loss": 0.33743955194950104, "time": 0.637329888343811, "epoch": 2, "memory": 44138, "step": 6037} +{"lr": 9.471276914664257e-05, "data_time": 0.0015462160110473633, "loss": 0.3715079724788666, "time": 0.6368890762329101, "epoch": 2, "memory": 44138, "step": 6137} +{"lr": 9.454262860722667e-05, "data_time": 0.0015308618545532226, "loss": 0.35632383823394775, "time": 0.636843204498291, "epoch": 2, "memory": 44138, "step": 6237} +{"lr": 9.436997020451398e-05, "data_time": 0.0014986515045166016, "loss": 0.3385142832994461, "time": 0.6363614797592163, "epoch": 2, "memory": 44138, "step": 6337} +{"lr": 9.419480493246903e-05, "data_time": 0.0016518115997314453, "loss": 0.3685177147388458, "time": 0.6366246461868286, "epoch": 2, "memory": 44138, "step": 6437} +{"lr": 9.401714394468036e-05, "data_time": 0.0014364004135131836, "loss": 0.3705434799194336, "time": 0.636694860458374, "epoch": 2, "memory": 44138, "step": 6537} +{"lr": 9.383699855365022e-05, "data_time": 0.002483630180358887, "loss": 0.3419413596391678, "time": 0.6372604608535767, "epoch": 2, "memory": 44138, "step": 6637} +{"lr": 9.365438023007469e-05, "data_time": 0.0015980005264282227, "loss": 0.3687066316604614, "time": 0.6379871606826782, "epoch": 2, "memory": 44138, "step": 6737} +{"lr": 9.346930060211267e-05, "data_time": 0.001495051383972168, "loss": 0.3346828520298004, "time": 0.6378123998641968, "epoch": 2, "memory": 44138, "step": 6837} +{"lr": 9.328177145464599e-05, "data_time": 0.0016396522521972656, "loss": 0.36425608992576597, "time": 0.6370981931686401, "epoch": 2, "memory": 44138, "step": 6937} +{"lr": 9.309180472852871e-05, "data_time": 0.00156095027923584, "loss": 0.32951463758945465, "time": 0.636542558670044, "epoch": 2, "memory": 44138, "step": 7037} +{"lr": 9.289941251982683e-05, "data_time": 0.0017425060272216798, "loss": 0.33361624777317045, "time": 0.637116551399231, "epoch": 2, "memory": 44138, "step": 7137} +{"lr": 9.270460707904827e-05, "data_time": 0.00169064998626709, "loss": 0.33794479072093964, "time": 0.6367788314819336, "epoch": 2, "memory": 44138, "step": 7237} +{"lr": 9.250740081036243e-05, "data_time": 0.0013814210891723634, "loss": 0.35136326849460603, "time": 0.6368788719177246, "epoch": 2, "memory": 44138, "step": 7337} +{"lr": 9.230780627081088e-05, "data_time": 0.001467418670654297, "loss": 0.3443381518125534, "time": 0.6376749753952027, "epoch": 2, "memory": 44138, "step": 7437} +{"lr": 9.210583616950722e-05, "data_time": 0.0014742136001586914, "loss": 0.33476702272892, "time": 0.636881160736084, "epoch": 2, "memory": 44138, "step": 7537} +{"lr": 9.190150336682842e-05, "data_time": 0.0015477418899536132, "loss": 0.3649121254682541, "time": 0.6370478868484497, "epoch": 2, "memory": 44138, "step": 7637} +{"lr": 9.16948208735955e-05, "data_time": 0.001503157615661621, "loss": 0.3500199973583221, "time": 0.6366247653961181, "epoch": 2, "memory": 44138, "step": 7737} +{"lr": 9.148580185024521e-05, "data_time": 0.001553821563720703, "loss": 0.3575607657432556, "time": 0.6374613523483277, "epoch": 2, "memory": 44138, "step": 7837} +{"accuracy/top1": 78.83163452148438, "data_time": 0.016518901134359425, "time": 1.2614407210514462, "step": 2} +{"lr": 9.119567684509738e-05, "data_time": 0.0015429258346557617, "loss": 0.34105669558048246, "time": 0.6430653810501099, "epoch": 3, "memory": 44138, "step": 7974} +{"lr": 9.098117366468421e-05, "data_time": 0.001436758041381836, "loss": 0.33879759907722473, "time": 0.6373371124267578, "epoch": 3, "memory": 44138, "step": 8074} +{"lr": 9.076437939539471e-05, "data_time": 0.001592397689819336, "loss": 0.359617480635643, "time": 0.6390689849853516, "epoch": 3, "memory": 44138, "step": 8174} +{"lr": 9.05453078415294e-05, "data_time": 0.0015715360641479492, "loss": 0.3803259253501892, "time": 0.6369717359542847, "epoch": 3, "memory": 44138, "step": 8274} +{"lr": 9.032397295239427e-05, "data_time": 0.0015054702758789062, "loss": 0.333760067820549, "time": 0.636887502670288, "epoch": 3, "memory": 44138, "step": 8374} +{"lr": 9.010038882141241e-05, "data_time": 0.0015065431594848632, "loss": 0.35237073004245756, "time": 0.6364066362380981, "epoch": 3, "memory": 44138, "step": 8474} +{"lr": 8.987456968522655e-05, "data_time": 0.0018047094345092773, "loss": 0.33100546300411227, "time": 0.6371907472610474, "epoch": 3, "memory": 44138, "step": 8574} +{"lr": 8.964652992279262e-05, "data_time": 0.0015096664428710938, "loss": 0.3385528564453125, "time": 0.6365949392318726, "epoch": 3, "memory": 44138, "step": 8674} +{"lr": 8.941628405446428e-05, "data_time": 0.0016623258590698243, "loss": 0.3177237302064896, "time": 0.6368241786956788, "epoch": 3, "memory": 44138, "step": 8774} +{"lr": 8.918384674106797e-05, "data_time": 0.0014980077743530274, "loss": 0.3489371448755264, "time": 0.6374441623687744, "epoch": 3, "memory": 44138, "step": 8874} +{"lr": 8.894923278296992e-05, "data_time": 0.0016621828079223632, "loss": 0.33893647193908694, "time": 0.6475806951522827, "epoch": 3, "memory": 44138, "step": 8974} +{"lr": 8.871245711913324e-05, "data_time": 0.001572585105895996, "loss": 0.3414228200912476, "time": 0.6364705324172973, "epoch": 3, "memory": 44138, "step": 9074} +{"lr": 8.847353482616703e-05, "data_time": 0.0015491962432861329, "loss": 0.3254220008850098, "time": 0.6401185989379883, "epoch": 3, "memory": 44138, "step": 9174} +{"lr": 8.823248111736638e-05, "data_time": 0.0014464616775512694, "loss": 0.34861191511154177, "time": 0.6371772766113282, "epoch": 3, "memory": 44138, "step": 9274} +{"lr": 8.798931134174327e-05, "data_time": 0.0014491796493530274, "loss": 0.3306316822767258, "time": 0.6372756004333496, "epoch": 3, "memory": 44138, "step": 9374} +{"lr": 8.774404098304961e-05, "data_time": 0.0015141487121582032, "loss": 0.32666544020175936, "time": 0.6364907026290894, "epoch": 3, "memory": 44138, "step": 9474} +{"lr": 8.749668565879136e-05, "data_time": 0.0017288684844970702, "loss": 0.34261976182460785, "time": 0.6382769823074341, "epoch": 3, "memory": 44138, "step": 9574} +{"lr": 8.724726111923386e-05, "data_time": 0.0014741897583007812, "loss": 0.3573506295681, "time": 0.6365964651107788, "epoch": 3, "memory": 44138, "step": 9674} +{"lr": 8.699578324639898e-05, "data_time": 0.0018654346466064453, "loss": 0.35222489535808565, "time": 0.638684868812561, "epoch": 3, "memory": 44138, "step": 9774} +{"lr": 8.67422680530539e-05, "data_time": 0.0015665531158447266, "loss": 0.3136993333697319, "time": 0.6389204502105713, "epoch": 3, "memory": 44138, "step": 9874} +{"lr": 8.648673168169147e-05, "data_time": 0.001727914810180664, "loss": 0.3321169525384903, "time": 0.6371656894683838, "epoch": 3, "memory": 44138, "step": 9974} +{"lr": 8.622919040350244e-05, "data_time": 0.0015274524688720704, "loss": 0.3592831283807755, "time": 0.6371702671051025, "epoch": 3, "memory": 44138, "step": 10074} +{"lr": 8.596966061733922e-05, "data_time": 0.0016551017761230469, "loss": 0.3653866440057755, "time": 0.6370359659194946, "epoch": 3, "memory": 44138, "step": 10174} +{"lr": 8.570815884867185e-05, "data_time": 0.0015550613403320312, "loss": 0.34676103591918944, "time": 0.6360797643661499, "epoch": 3, "memory": 44138, "step": 10274} +{"lr": 8.544470174853563e-05, "data_time": 0.0014838695526123047, "loss": 0.33594187796115876, "time": 0.636370038986206, "epoch": 3, "memory": 44138, "step": 10374} +{"lr": 8.517930609247077e-05, "data_time": 0.0014545917510986328, "loss": 0.3341986835002899, "time": 0.6366753101348877, "epoch": 3, "memory": 44138, "step": 10474} +{"lr": 8.491198877945477e-05, "data_time": 0.0014988899230957032, "loss": 0.3113416790962219, "time": 0.638522744178772, "epoch": 3, "memory": 44138, "step": 10574} +{"lr": 8.464276683082565e-05, "data_time": 0.0016964435577392577, "loss": 0.3383102774620056, "time": 0.6390778303146363, "epoch": 3, "memory": 44138, "step": 10674} +{"lr": 8.437165738919852e-05, "data_time": 0.0013664484024047852, "loss": 0.3223652243614197, "time": 0.6382468461990356, "epoch": 3, "memory": 44138, "step": 10774} +{"lr": 8.409867771737398e-05, "data_time": 0.0015694379806518554, "loss": 0.33741885125637056, "time": 0.6373280048370361, "epoch": 3, "memory": 44138, "step": 10874} +{"lr": 8.382384519723895e-05, "data_time": 0.0016610860824584962, "loss": 0.3260244011878967, "time": 0.6372844696044921, "epoch": 3, "memory": 44138, "step": 10974} +{"lr": 8.35471773286596e-05, "data_time": 0.0015014171600341796, "loss": 0.32323668897151947, "time": 0.6373264789581299, "epoch": 3, "memory": 44138, "step": 11074} +{"lr": 8.326869172836765e-05, "data_time": 0.0017076730728149414, "loss": 0.3508055955171585, "time": 0.6367055177688599, "epoch": 3, "memory": 44138, "step": 11174} +{"lr": 8.298840612883789e-05, "data_time": 0.0014889240264892578, "loss": 0.34529263377189634, "time": 0.6364981412887574, "epoch": 3, "memory": 44138, "step": 11274} +{"lr": 8.270633837715962e-05, "data_time": 0.0015820026397705077, "loss": 0.3595623314380646, "time": 0.6367258548736572, "epoch": 3, "memory": 44138, "step": 11374} +{"lr": 8.242250643390022e-05, "data_time": 0.0022446632385253905, "loss": 0.31640669256448745, "time": 0.6418020248413085, "epoch": 3, "memory": 44138, "step": 11474} +{"lr": 8.2136928371961e-05, "data_time": 0.0017303705215454101, "loss": 0.3379951626062393, "time": 0.6376960277557373, "epoch": 3, "memory": 44138, "step": 11574} +{"lr": 8.184962237542712e-05, "data_time": 0.0016300678253173828, "loss": 0.3181945890188217, "time": 0.6373231410980225, "epoch": 3, "memory": 44138, "step": 11674} +{"lr": 8.156060673840912e-05, "data_time": 0.0016222000122070312, "loss": 0.34422839283943174, "time": 0.6368931770324707, "epoch": 3, "memory": 44138, "step": 11774} +{"accuracy/top1": 82.16485595703125, "data_time": 0.015650802645190007, "time": 1.2585819014187516, "step": 3} +{"lr": 8.116191337909002e-05, "data_time": 0.0031123638153076174, "loss": 0.338031068444252, "time": 0.6381591558456421, "epoch": 4, "memory": 44138, "step": 11911} +{"lr": 8.086891958166228e-05, "data_time": 0.001564168930053711, "loss": 0.3375686138868332, "time": 0.6364040374755859, "epoch": 4, "memory": 44138, "step": 12011} +{"lr": 8.057427858966412e-05, "data_time": 0.0014896154403686523, "loss": 0.3617121517658234, "time": 0.6370182991027832, "epoch": 4, "memory": 44138, "step": 12111} +{"lr": 8.027800916425949e-05, "data_time": 0.0015753746032714845, "loss": 0.31206669807434084, "time": 0.6372193336486817, "epoch": 4, "memory": 44138, "step": 12211} +{"lr": 7.998013017030234e-05, "data_time": 0.0018462657928466797, "loss": 0.34003559350967405, "time": 0.6371035814285279, "epoch": 4, "memory": 44138, "step": 12311} +{"lr": 7.968066057513528e-05, "data_time": 0.0015268802642822265, "loss": 0.337826144695282, "time": 0.6365351915359497, "epoch": 4, "memory": 44138, "step": 12411} +{"lr": 7.937961944738199e-05, "data_time": 0.0015517950057983398, "loss": 0.33829768002033234, "time": 0.6368270874023437, "epoch": 4, "memory": 44138, "step": 12511} +{"lr": 7.907702595573289e-05, "data_time": 0.0022010326385498045, "loss": 0.3339881062507629, "time": 0.6374033689498901, "epoch": 4, "memory": 44138, "step": 12611} +{"lr": 7.877289936772454e-05, "data_time": 0.001515817642211914, "loss": 0.3363713204860687, "time": 0.6368407011032104, "epoch": 4, "memory": 44138, "step": 12711} +{"lr": 7.8467259048513e-05, "data_time": 0.001432323455810547, "loss": 0.31904266476631166, "time": 0.6369088649749756, "epoch": 4, "memory": 44138, "step": 12811} +{"lr": 7.816012445964079e-05, "data_time": 0.0018038272857666016, "loss": 0.33538403213024137, "time": 0.6368149995803833, "epoch": 4, "memory": 44138, "step": 12911} +{"lr": 7.785151515779718e-05, "data_time": 0.0014866828918457032, "loss": 0.3120292916893959, "time": 0.6378989696502686, "epoch": 4, "memory": 44138, "step": 13011} +{"lr": 7.754145079357376e-05, "data_time": 0.0015709638595581056, "loss": 0.35246534645557404, "time": 0.6378729581832886, "epoch": 4, "memory": 44138, "step": 13111} +{"lr": 7.722995111021218e-05, "data_time": 0.0014418363571166992, "loss": 0.33217440247535707, "time": 0.6377979040145874, "epoch": 4, "memory": 44138, "step": 13211} +{"lr": 7.691703594234788e-05, "data_time": 0.0016176462173461913, "loss": 0.35015018582344054, "time": 0.6375664710998535, "epoch": 4, "memory": 44138, "step": 13311} +{"lr": 7.660272521474673e-05, "data_time": 0.0016233444213867188, "loss": 0.3585609346628189, "time": 0.6372680902481079, "epoch": 4, "memory": 44138, "step": 13411} +{"lr": 7.628703894103628e-05, "data_time": 0.001573491096496582, "loss": 0.328920578956604, "time": 0.6369132041931153, "epoch": 4, "memory": 44138, "step": 13511} +{"lr": 7.596999722243168e-05, "data_time": 0.001769709587097168, "loss": 0.3384198397397995, "time": 0.6378817558288574, "epoch": 4, "memory": 44138, "step": 13611} +{"lr": 7.565162024645551e-05, "data_time": 0.0014981985092163085, "loss": 0.3200004667043686, "time": 0.6374590158462524, "epoch": 4, "memory": 44138, "step": 13711} +{"lr": 7.533192828565237e-05, "data_time": 0.001697683334350586, "loss": 0.33200671672821047, "time": 0.6369019508361816, "epoch": 4, "memory": 44138, "step": 13811} +{"lr": 7.501094169629802e-05, "data_time": 0.0020283222198486327, "loss": 0.34375424683094025, "time": 0.6373275279998779, "epoch": 4, "memory": 44138, "step": 13911} +{"lr": 7.468868091710333e-05, "data_time": 0.0015586376190185546, "loss": 0.30156241059303285, "time": 0.6369643688201905, "epoch": 4, "memory": 44138, "step": 14011} +{"lr": 7.436516646791276e-05, "data_time": 0.0023156166076660155, "loss": 0.3411762535572052, "time": 0.6376494884490966, "epoch": 4, "memory": 44138, "step": 14111} +{"lr": 7.404041894839768e-05, "data_time": 0.0016160249710083009, "loss": 0.32748172283172605, "time": 0.636782431602478, "epoch": 4, "memory": 44138, "step": 14211} +{"lr": 7.371445903674494e-05, "data_time": 0.0014303207397460937, "loss": 0.3261335790157318, "time": 0.6369709491729736, "epoch": 4, "memory": 44138, "step": 14311} +{"lr": 7.338730748834014e-05, "data_time": 0.0014453172683715821, "loss": 0.34288241565227506, "time": 0.6382145643234253, "epoch": 4, "memory": 44138, "step": 14411} +{"lr": 7.305898513444536e-05, "data_time": 0.0017593860626220702, "loss": 0.33481956720352174, "time": 0.6380037546157837, "epoch": 4, "memory": 44138, "step": 14511} +{"lr": 7.27295128808741e-05, "data_time": 0.001836252212524414, "loss": 0.3406389534473419, "time": 0.6371669054031373, "epoch": 4, "memory": 44138, "step": 14611} +{"lr": 7.239891170665884e-05, "data_time": 0.0016165971755981445, "loss": 0.34216139018535613, "time": 0.637013602256775, "epoch": 4, "memory": 44138, "step": 14711} +{"lr": 7.206720266271582e-05, "data_time": 0.001488041877746582, "loss": 0.31315720826387405, "time": 0.6370663404464721, "epoch": 4, "memory": 44138, "step": 14811} +{"lr": 7.173440687050449e-05, "data_time": 0.00149078369140625, "loss": 0.3458180397748947, "time": 0.6370178699493408, "epoch": 4, "memory": 44138, "step": 14911} +{"lr": 7.140054552068265e-05, "data_time": 0.0015344858169555665, "loss": 0.3669417053461075, "time": 0.636637544631958, "epoch": 4, "memory": 44138, "step": 15011} +{"lr": 7.106563987175705e-05, "data_time": 0.0015919208526611328, "loss": 0.348757329583168, "time": 0.6377474308013916, "epoch": 4, "memory": 44138, "step": 15111} +{"lr": 7.072971124873013e-05, "data_time": 0.001621556282043457, "loss": 0.3177102476358414, "time": 0.6377481460571289, "epoch": 4, "memory": 44138, "step": 15211} +{"lr": 7.039278104174143e-05, "data_time": 0.0015107393264770508, "loss": 0.34663744270801544, "time": 0.6364511013031006, "epoch": 4, "memory": 44138, "step": 15311} +{"lr": 7.005487070470615e-05, "data_time": 0.0014693021774291992, "loss": 0.33676148056983946, "time": 0.6370150566101074, "epoch": 4, "memory": 44138, "step": 15411} +{"lr": 6.971600175394891e-05, "data_time": 0.0015363693237304688, "loss": 0.31902232468128205, "time": 0.6362375736236572, "epoch": 4, "memory": 44138, "step": 15511} +{"lr": 6.937619576683373e-05, "data_time": 0.001532578468322754, "loss": 0.3386623293161392, "time": 0.6367444276809693, "epoch": 4, "memory": 44138, "step": 15611} +{"lr": 6.903547438039021e-05, "data_time": 0.0018222808837890625, "loss": 0.3339738607406616, "time": 0.6374045372009277, "epoch": 4, "memory": 44138, "step": 15711} +{"accuracy/top1": 86.2565689086914, "data_time": 0.015925839029509444, "time": 1.2589548990644257, "step": 4} +{"lr": 6.856723955077045e-05, "data_time": 0.001645207405090332, "loss": 0.326394647359848, "time": 0.6376607894897461, "epoch": 5, "memory": 44138, "step": 15848} +{"lr": 6.822443541062844e-05, "data_time": 0.001539325714111328, "loss": 0.33485254645347595, "time": 0.6374549627304077, "epoch": 5, "memory": 44138, "step": 15948} +{"lr": 6.788078920910325e-05, "data_time": 0.001550769805908203, "loss": 0.3428355485200882, "time": 0.6369801759719849, "epoch": 5, "memory": 44138, "step": 16048} +{"lr": 6.753632282774874e-05, "data_time": 0.0025886774063110353, "loss": 0.33377662897109983, "time": 0.6380975246429443, "epoch": 5, "memory": 44138, "step": 16148} +{"lr": 6.719105820034336e-05, "data_time": 0.0020990848541259767, "loss": 0.3222132921218872, "time": 0.6389024019241333, "epoch": 5, "memory": 44138, "step": 16248} +{"lr": 6.68450173114937e-05, "data_time": 0.0021158456802368164, "loss": 0.3249441057443619, "time": 0.6427899599075317, "epoch": 5, "memory": 44138, "step": 16348} +{"lr": 6.64982221952347e-05, "data_time": 0.0017545461654663087, "loss": 0.3563050091266632, "time": 0.6383615732192993, "epoch": 5, "memory": 44138, "step": 16448} +{"lr": 6.615069493362599e-05, "data_time": 0.0026414871215820314, "loss": 0.3414961099624634, "time": 0.6386398553848267, "epoch": 5, "memory": 44138, "step": 16548} +{"lr": 6.580245765534682e-05, "data_time": 0.0019840717315673826, "loss": 0.34103713035583494, "time": 0.6361668825149536, "epoch": 5, "memory": 44138, "step": 16648} +{"lr": 6.54535325342862e-05, "data_time": 0.001967191696166992, "loss": 0.3617107540369034, "time": 0.6388908624649048, "epoch": 5, "memory": 44138, "step": 16748} +{"lr": 6.510394178813155e-05, "data_time": 0.002042698860168457, "loss": 0.36190399825572966, "time": 0.637880277633667, "epoch": 5, "memory": 44138, "step": 16848} +{"lr": 6.475370767695331e-05, "data_time": 0.0019478559494018554, "loss": 0.3186565160751343, "time": 0.6412222623825073, "epoch": 5, "memory": 44138, "step": 16948} +{"lr": 6.440285250178845e-05, "data_time": 0.0019121170043945312, "loss": 0.3272313341498375, "time": 0.6368343114852906, "epoch": 5, "memory": 44138, "step": 17048} +{"lr": 6.405139860321955e-05, "data_time": 0.001958751678466797, "loss": 0.3496026277542114, "time": 0.6370957851409912, "epoch": 5, "memory": 44138, "step": 17148} +{"lr": 6.369936835995306e-05, "data_time": 0.0023630142211914064, "loss": 0.317501437664032, "time": 0.6372188806533814, "epoch": 5, "memory": 44138, "step": 17248} +{"lr": 6.334678418739388e-05, "data_time": 0.0023343563079833984, "loss": 0.3219574272632599, "time": 0.6370770692825317, "epoch": 5, "memory": 44138, "step": 17348} +{"lr": 6.299366853621796e-05, "data_time": 0.0021996259689331054, "loss": 0.3162900283932686, "time": 0.637747049331665, "epoch": 5, "memory": 44138, "step": 17448} +{"lr": 6.264004389094315e-05, "data_time": 0.0016735076904296875, "loss": 0.3424944311380386, "time": 0.6381992340087891, "epoch": 5, "memory": 44138, "step": 17548} +{"lr": 6.22859327684974e-05, "data_time": 0.0017868280410766602, "loss": 0.3068771749734879, "time": 0.6393717050552368, "epoch": 5, "memory": 44138, "step": 17648} +{"lr": 6.193135771678478e-05, "data_time": 0.0027231454849243166, "loss": 0.370467546582222, "time": 0.6384228944778443, "epoch": 5, "memory": 44138, "step": 17748} +{"lr": 6.157634131324983e-05, "data_time": 0.0020483255386352537, "loss": 0.3498892217874527, "time": 0.6363596677780151, "epoch": 5, "memory": 44138, "step": 17848} +{"lr": 6.12209061634402e-05, "data_time": 0.002358865737915039, "loss": 0.3239938378334045, "time": 0.6381728887557984, "epoch": 5, "memory": 44138, "step": 17948} +{"lr": 6.086507489956686e-05, "data_time": 0.0018143177032470704, "loss": 0.30747300684452056, "time": 0.6373126983642579, "epoch": 5, "memory": 44138, "step": 18048} +{"lr": 6.0508870179063345e-05, "data_time": 0.001836705207824707, "loss": 0.34509778916835787, "time": 0.6379811525344848, "epoch": 5, "memory": 44138, "step": 18148} +{"lr": 6.015231468314303e-05, "data_time": 0.0017647981643676759, "loss": 0.33873641192913057, "time": 0.6389262914657593, "epoch": 5, "memory": 44138, "step": 18248} +{"lr": 5.979543111535473e-05, "data_time": 0.0015564680099487305, "loss": 0.3281713366508484, "time": 0.6364551782608032, "epoch": 5, "memory": 44138, "step": 18348} +{"lr": 5.943824220013694e-05, "data_time": 0.0016757965087890625, "loss": 0.3410695821046829, "time": 0.637682032585144, "epoch": 5, "memory": 44138, "step": 18448} +{"lr": 5.908077068137128e-05, "data_time": 0.002691364288330078, "loss": 0.3036420464515686, "time": 0.6384639739990234, "epoch": 5, "memory": 44138, "step": 18548} +{"lr": 5.872303932093392e-05, "data_time": 0.0015374422073364258, "loss": 0.33327831923961637, "time": 0.6390835762023925, "epoch": 5, "memory": 44138, "step": 18648} +{"lr": 5.836507089724633e-05, "data_time": 0.0018391847610473634, "loss": 0.31866020858287813, "time": 0.6406067371368408, "epoch": 5, "memory": 44138, "step": 18748} +{"lr": 5.800688820382498e-05, "data_time": 0.0015901565551757813, "loss": 0.29573112577199934, "time": 0.6411308288574219, "epoch": 5, "memory": 44138, "step": 18848} +{"lr": 5.7648514047829945e-05, "data_time": 0.0015150547027587891, "loss": 0.3226841062307358, "time": 0.6422483205795289, "epoch": 5, "memory": 44138, "step": 18948} +{"lr": 5.7289971248612474e-05, "data_time": 0.0016436338424682616, "loss": 0.3375723406672478, "time": 0.6410388231277466, "epoch": 5, "memory": 44138, "step": 19048} +{"lr": 5.693128263626215e-05, "data_time": 0.001591634750366211, "loss": 0.32685727477073667, "time": 0.6419731378555298, "epoch": 5, "memory": 44138, "step": 19148} +{"lr": 5.6572471050153236e-05, "data_time": 0.0018211603164672852, "loss": 0.32787306010723116, "time": 0.6468555450439453, "epoch": 5, "memory": 44138, "step": 19248} +{"lr": 5.621355933749018e-05, "data_time": 0.0018019676208496094, "loss": 0.32421029806137086, "time": 0.639817190170288, "epoch": 5, "memory": 44138, "step": 19348} +{"lr": 5.5854570351853106e-05, "data_time": 0.0018308401107788087, "loss": 0.3336017966270447, "time": 0.639714241027832, "epoch": 5, "memory": 44138, "step": 19448} +{"lr": 5.5495526951742315e-05, "data_time": 0.0016659021377563477, "loss": 0.36472677886486055, "time": 0.6413752317428589, "epoch": 5, "memory": 44138, "step": 19548} +{"lr": 5.513645199912304e-05, "data_time": 0.0015044450759887696, "loss": 0.35022721588611605, "time": 0.6419339895248413, "epoch": 5, "memory": 44138, "step": 19648} +{"accuracy/top1": 90.24390411376953, "data_time": 0.0146208878221183, "time": 1.258496120058257, "step": 5} +{"lr": 5.464450978670876e-05, "data_time": 0.0016424179077148438, "loss": 0.36218151450157166, "time": 0.6386335134506226, "epoch": 6, "memory": 44138, "step": 19785} +{"lr": 5.4285451361545876e-05, "data_time": 0.001752781867980957, "loss": 0.29286844879388807, "time": 0.6384132862091064, "epoch": 6, "memory": 44138, "step": 19885} +{"lr": 5.3926438435022166e-05, "data_time": 0.0016929149627685548, "loss": 0.31446826457977295, "time": 0.6405260562896729, "epoch": 6, "memory": 44138, "step": 19985} +{"lr": 5.356749386716241e-05, "data_time": 0.0029013872146606444, "loss": 0.345249080657959, "time": 0.6400500535964966, "epoch": 6, "memory": 44138, "step": 20085} +{"lr": 5.320864051363873e-05, "data_time": 0.0015892744064331054, "loss": 0.3425854355096817, "time": 0.6493218898773193, "epoch": 6, "memory": 44138, "step": 20185} +{"lr": 5.2849901224315175e-05, "data_time": 0.0014860868453979493, "loss": 0.30708692967891693, "time": 0.6392604351043701, "epoch": 6, "memory": 44138, "step": 20285} +{"lr": 5.249129884179276e-05, "data_time": 0.001697397232055664, "loss": 0.35072503685951234, "time": 0.6379307031631469, "epoch": 6, "memory": 44138, "step": 20385} +{"lr": 5.2132856199954985e-05, "data_time": 0.0024124622344970704, "loss": 0.31700047850608826, "time": 0.6375947713851928, "epoch": 6, "memory": 44138, "step": 20485} +{"lr": 5.177459612251411e-05, "data_time": 0.001714944839477539, "loss": 0.36079367697238923, "time": 0.6386557340621948, "epoch": 6, "memory": 44138, "step": 20585} +{"lr": 5.141654142155744e-05, "data_time": 0.0015901803970336914, "loss": 0.34223597347736356, "time": 0.6371722221374512, "epoch": 6, "memory": 44138, "step": 20685} +{"lr": 5.105871489609512e-05, "data_time": 0.0016604423522949218, "loss": 0.34108743667602537, "time": 0.640283179283142, "epoch": 6, "memory": 44138, "step": 20785} +{"lr": 5.070113933060828e-05, "data_time": 0.0016683101654052734, "loss": 0.3372240215539932, "time": 0.6392496824264526, "epoch": 6, "memory": 44138, "step": 20885} +{"lr": 5.0343837493598264e-05, "data_time": 0.0018593549728393554, "loss": 0.34016806781291964, "time": 0.6387745380401612, "epoch": 6, "memory": 44138, "step": 20985} +{"lr": 4.998683213613675e-05, "data_time": 0.0020160913467407227, "loss": 0.3343943566083908, "time": 0.6394699335098266, "epoch": 6, "memory": 44138, "step": 21085} +{"lr": 4.963014599041739e-05, "data_time": 0.002039480209350586, "loss": 0.3596372425556183, "time": 0.6404732704162598, "epoch": 6, "memory": 44138, "step": 21185} +{"lr": 4.927380176830793e-05, "data_time": 0.0017155647277832032, "loss": 0.3008223414421082, "time": 0.6406383275985718, "epoch": 6, "memory": 44138, "step": 21285} +{"lr": 4.89178221599044e-05, "data_time": 0.0018315315246582031, "loss": 0.31562944650650027, "time": 0.6387464284896851, "epoch": 6, "memory": 44138, "step": 21385} +{"lr": 4.85622298320861e-05, "data_time": 0.0016685247421264649, "loss": 0.31584973335266114, "time": 0.6380267143249512, "epoch": 6, "memory": 44138, "step": 21485} +{"lr": 4.8207047427072355e-05, "data_time": 0.001862955093383789, "loss": 0.3263800323009491, "time": 0.6384902477264405, "epoch": 6, "memory": 44138, "step": 21585} +{"lr": 4.7852297560980876e-05, "data_time": 0.0018761634826660156, "loss": 0.32396080791950227, "time": 0.6380950689315796, "epoch": 6, "memory": 44138, "step": 21685} +{"lr": 4.749800282238749e-05, "data_time": 0.0015840768814086915, "loss": 0.3161410465836525, "time": 0.6385318994522095, "epoch": 6, "memory": 44138, "step": 21785} +{"lr": 4.714418577088805e-05, "data_time": 0.0017816543579101563, "loss": 0.334672412276268, "time": 0.6386142492294311, "epoch": 6, "memory": 44138, "step": 21885} +{"lr": 4.679086893566184e-05, "data_time": 0.001709890365600586, "loss": 0.3308043345808983, "time": 0.6387040853500366, "epoch": 6, "memory": 44138, "step": 21985} +{"lr": 4.643807481403694e-05, "data_time": 0.001747465133666992, "loss": 0.33866152465343474, "time": 0.6387199401855469, "epoch": 6, "memory": 44138, "step": 22085} +{"lr": 4.6085825870057946e-05, "data_time": 0.0017476558685302734, "loss": 0.33113373517990113, "time": 0.6380645990371704, "epoch": 6, "memory": 44138, "step": 22185} +{"lr": 4.573414453305537e-05, "data_time": 0.0015560388565063477, "loss": 0.34676816761493684, "time": 0.6406392335891724, "epoch": 6, "memory": 44138, "step": 22285} +{"lr": 4.5383053196217514e-05, "data_time": 0.0016381025314331054, "loss": 0.36065601706504824, "time": 0.6400545835494995, "epoch": 6, "memory": 44138, "step": 22385} +{"lr": 4.503257421516477e-05, "data_time": 0.0017625093460083008, "loss": 0.32619393765926363, "time": 0.637947678565979, "epoch": 6, "memory": 44138, "step": 22485} +{"lr": 4.468272990652586e-05, "data_time": 0.002097606658935547, "loss": 0.3510457679629326, "time": 0.6441561937332153, "epoch": 6, "memory": 44138, "step": 22585} +{"lr": 4.433354254651697e-05, "data_time": 0.001828622817993164, "loss": 0.32855255007743833, "time": 0.6386828899383545, "epoch": 6, "memory": 44138, "step": 22685} +{"lr": 4.398503436952338e-05, "data_time": 0.001613950729370117, "loss": 0.3488876402378082, "time": 0.6386906147003174, "epoch": 6, "memory": 44138, "step": 22785} +{"lr": 4.363722756668351e-05, "data_time": 0.002217888832092285, "loss": 0.3292103260755539, "time": 0.6382639646530152, "epoch": 6, "memory": 44138, "step": 22885} +{"lr": 4.329014428447611e-05, "data_time": 0.0016797780990600586, "loss": 0.33284675478935244, "time": 0.6418892860412597, "epoch": 6, "memory": 44138, "step": 22985} +{"lr": 4.294380662331e-05, "data_time": 0.0015721797943115234, "loss": 0.32860801815986634, "time": 0.6390622615814209, "epoch": 6, "memory": 44138, "step": 23085} +{"lr": 4.259823663611678e-05, "data_time": 0.0015774011611938477, "loss": 0.32207544445991515, "time": 0.63813316822052, "epoch": 6, "memory": 44138, "step": 23185} +{"lr": 4.22534563269468e-05, "data_time": 0.0030534744262695314, "loss": 0.34216980934143065, "time": 0.6394953727722168, "epoch": 6, "memory": 44138, "step": 23285} +{"lr": 4.1909487649567694e-05, "data_time": 0.0015613794326782226, "loss": 0.350005766749382, "time": 0.6372884273529053, "epoch": 6, "memory": 44138, "step": 23385} +{"lr": 4.156635250606698e-05, "data_time": 0.0014662027359008789, "loss": 0.33806262016296384, "time": 0.6379681348800659, "epoch": 6, "memory": 44138, "step": 23485} +{"lr": 4.122407274545703e-05, "data_time": 0.0015445470809936524, "loss": 0.3400091290473938, "time": 0.6394245147705078, "epoch": 6, "memory": 44138, "step": 23585} +{"accuracy/top1": 92.45677185058594, "data_time": 0.016118320925482387, "time": 1.2589721515260894, "step": 6} +{"lr": 4.075657788156285e-05, "data_time": 0.0016452789306640625, "loss": 0.33512323498725893, "time": 0.6380365371704102, "epoch": 7, "memory": 44138, "step": 23722} +{"lr": 4.0416412311741566e-05, "data_time": 0.0018833637237548827, "loss": 0.3269672095775604, "time": 0.6393464803695679, "epoch": 7, "memory": 44138, "step": 23822} +{"lr": 4.007717534685821e-05, "data_time": 0.0015670061111450195, "loss": 0.33208352476358416, "time": 0.6402326583862304, "epoch": 7, "memory": 44138, "step": 23922} +{"lr": 3.973888858770982e-05, "data_time": 0.0017290353775024415, "loss": 0.33266387283802035, "time": 0.638140082359314, "epoch": 7, "memory": 44138, "step": 24022} +{"lr": 3.9401573574589535e-05, "data_time": 0.0015340805053710937, "loss": 0.33821841776371003, "time": 0.6378901720046997, "epoch": 7, "memory": 44138, "step": 24122} +{"lr": 3.906525178591511e-05, "data_time": 0.001939082145690918, "loss": 0.3124773234128952, "time": 0.6394755363464355, "epoch": 7, "memory": 44138, "step": 24222} +{"lr": 3.87299446368609e-05, "data_time": 0.0015781164169311524, "loss": 0.3509422570466995, "time": 0.6398543357849121, "epoch": 7, "memory": 44138, "step": 24322} +{"lr": 3.839567347799442e-05, "data_time": 0.0016474008560180664, "loss": 0.33867704272270205, "time": 0.6391457557678223, "epoch": 7, "memory": 44138, "step": 24422} +{"lr": 3.806245959391694e-05, "data_time": 0.0016038894653320312, "loss": 0.3241992652416229, "time": 0.6411542177200318, "epoch": 7, "memory": 44138, "step": 24522} +{"lr": 3.773032420190816e-05, "data_time": 0.0017015695571899413, "loss": 0.32114947885274886, "time": 0.6378494739532471, "epoch": 7, "memory": 44138, "step": 24622} +{"lr": 3.739928845057507e-05, "data_time": 0.0015629291534423827, "loss": 0.3297632187604904, "time": 0.6384621143341065, "epoch": 7, "memory": 44138, "step": 24722} +{"lr": 3.706937341850541e-05, "data_time": 0.0016818523406982421, "loss": 0.34492531418800354, "time": 0.6386454343795777, "epoch": 7, "memory": 44138, "step": 24822} +{"lr": 3.674060011292556e-05, "data_time": 0.0015878677368164062, "loss": 0.3129858136177063, "time": 0.6385512113571167, "epoch": 7, "memory": 44138, "step": 24922} +{"lr": 3.641298946836287e-05, "data_time": 0.0021367788314819334, "loss": 0.3471270501613617, "time": 0.6414084672927857, "epoch": 7, "memory": 44138, "step": 25022} +{"lr": 3.6086562345312595e-05, "data_time": 0.0020767688751220704, "loss": 0.3027243345975876, "time": 0.6397251605987548, "epoch": 7, "memory": 44138, "step": 25122} +{"lr": 3.576133952890974e-05, "data_time": 0.0014804840087890626, "loss": 0.3238443449139595, "time": 0.6403166055679321, "epoch": 7, "memory": 44138, "step": 25222} +{"lr": 3.543734172760541e-05, "data_time": 0.0018121719360351563, "loss": 0.3503607839345932, "time": 0.63937406539917, "epoch": 7, "memory": 44138, "step": 25322} +{"lr": 3.511458957184831e-05, "data_time": 0.001491546630859375, "loss": 0.34674532413482667, "time": 0.6392184734344483, "epoch": 7, "memory": 44138, "step": 25422} +{"lr": 3.479310361277118e-05, "data_time": 0.0017754793167114257, "loss": 0.31534984707832336, "time": 0.637584137916565, "epoch": 7, "memory": 44138, "step": 25522} +{"lr": 3.4472904320882065e-05, "data_time": 0.001627635955810547, "loss": 0.33254554271698, "time": 0.6394550800323486, "epoch": 7, "memory": 44138, "step": 25622} +{"lr": 3.4154012084760864e-05, "data_time": 0.0020392894744873046, "loss": 0.3499720633029938, "time": 0.6396182537078857, "epoch": 7, "memory": 44138, "step": 25722} +{"lr": 3.38364472097612e-05, "data_time": 0.001755666732788086, "loss": 0.3322810471057892, "time": 0.6414047002792358, "epoch": 7, "memory": 44138, "step": 25822} +{"lr": 3.3520229916717485e-05, "data_time": 0.0015645503997802734, "loss": 0.34651918709278107, "time": 0.6374246835708618, "epoch": 7, "memory": 44138, "step": 25922} +{"lr": 3.320538034065722e-05, "data_time": 0.0015337228775024413, "loss": 0.3382045477628708, "time": 0.6386137962341308, "epoch": 7, "memory": 44138, "step": 26022} +{"lr": 3.289191852951901e-05, "data_time": 0.0018460512161254882, "loss": 0.3323641985654831, "time": 0.6404607057571411, "epoch": 7, "memory": 44138, "step": 26122} +{"lr": 3.257986444287605e-05, "data_time": 0.0016823530197143555, "loss": 0.3483747899532318, "time": 0.640232253074646, "epoch": 7, "memory": 44138, "step": 26222} +{"lr": 3.226923795066519e-05, "data_time": 0.0013779163360595702, "loss": 0.3480044960975647, "time": 0.6392740726470947, "epoch": 7, "memory": 44138, "step": 26322} +{"lr": 3.196005883192158e-05, "data_time": 0.0019949197769165037, "loss": 0.32673614621162417, "time": 0.6388867855072021, "epoch": 7, "memory": 44138, "step": 26422} +{"lr": 3.165234677351947e-05, "data_time": 0.0019626379013061523, "loss": 0.3208146706223488, "time": 0.6415780782699585, "epoch": 7, "memory": 44138, "step": 26522} +{"lr": 3.134612136891845e-05, "data_time": 0.0021756172180175783, "loss": 0.3511429995298386, "time": 0.6430078506469726, "epoch": 7, "memory": 44138, "step": 26622} +{"lr": 3.104140211691602e-05, "data_time": 0.001885056495666504, "loss": 0.3994672566652298, "time": 0.6417765617370605, "epoch": 7, "memory": 44138, "step": 26722} +{"lr": 3.073820842040592e-05, "data_time": 0.0016906261444091797, "loss": 0.36018048524856566, "time": 0.6405328273773193, "epoch": 7, "memory": 44138, "step": 26822} +{"lr": 3.043655958514262e-05, "data_time": 0.0015549182891845704, "loss": 0.3429575651884079, "time": 0.6394672155380249, "epoch": 7, "memory": 44138, "step": 26922} +{"lr": 3.013647481851217e-05, "data_time": 0.0016427755355834961, "loss": 0.34207204580307005, "time": 0.641802453994751, "epoch": 7, "memory": 44138, "step": 27022} +{"lr": 2.983797322830899e-05, "data_time": 0.001452159881591797, "loss": 0.3353388488292694, "time": 0.6427386045455933, "epoch": 7, "memory": 44138, "step": 27122} +{"lr": 2.9541073821519346e-05, "data_time": 0.0015214204788208008, "loss": 0.33911772817373276, "time": 0.6428911209106445, "epoch": 7, "memory": 44138, "step": 27222} +{"lr": 2.924579550311095e-05, "data_time": 0.0016516923904418945, "loss": 0.3578227132558823, "time": 0.6407561779022217, "epoch": 7, "memory": 44138, "step": 27322} +{"lr": 2.8952157074829418e-05, "data_time": 0.001589202880859375, "loss": 0.3136764168739319, "time": 0.6399877071380615, "epoch": 7, "memory": 44138, "step": 27422} +{"lr": 2.866017723400071e-05, "data_time": 0.0016010522842407227, "loss": 0.31430790275335313, "time": 0.6452139377593994, "epoch": 7, "memory": 44138, "step": 27522} +{"accuracy/top1": 93.62931060791016, "data_time": 0.01609754151311414, "time": 1.2588747410938657, "step": 7} +{"lr": 2.8262891373132927e-05, "data_time": 0.0017935514450073242, "loss": 0.32785196900367736, "time": 0.641397500038147, "epoch": 8, "memory": 44138, "step": 27659} +{"lr": 2.797491643470029e-05, "data_time": 0.0023957490921020508, "loss": 0.3483874022960663, "time": 0.6421234130859375, "epoch": 8, "memory": 44138, "step": 27759} +{"lr": 2.768866230918266e-05, "data_time": 0.001797795295715332, "loss": 0.33559287488460543, "time": 0.649546217918396, "epoch": 8, "memory": 44138, "step": 27859} +{"lr": 2.7404147223713185e-05, "data_time": 0.001596689224243164, "loss": 0.336510494351387, "time": 0.6395530939102173, "epoch": 8, "memory": 44138, "step": 27959} +{"lr": 2.712138929469222e-05, "data_time": 0.0016678571701049805, "loss": 0.35269678831100465, "time": 0.6401216506958007, "epoch": 8, "memory": 44138, "step": 28059} +{"lr": 2.6840406526633896e-05, "data_time": 0.0015362739562988282, "loss": 0.3273759573698044, "time": 0.6406999349594116, "epoch": 8, "memory": 44138, "step": 28159} +{"lr": 2.6561216811019345e-05, "data_time": 0.00162961483001709, "loss": 0.3498887956142426, "time": 0.6378124237060547, "epoch": 8, "memory": 44138, "step": 28259} +{"lr": 2.6283837925157908e-05, "data_time": 0.001520538330078125, "loss": 0.33529615998268125, "time": 0.6380717277526855, "epoch": 8, "memory": 44138, "step": 28359} +{"lr": 2.6008287531054805e-05, "data_time": 0.0029815912246704103, "loss": 0.3309875041246414, "time": 0.6378012418746948, "epoch": 8, "memory": 44138, "step": 28459} +{"lr": 2.5734583174286796e-05, "data_time": 0.002580428123474121, "loss": 0.34405426383018495, "time": 0.6385094881057739, "epoch": 8, "memory": 44138, "step": 28559} +{"lr": 2.546274228288488e-05, "data_time": 0.0016176462173461913, "loss": 0.33039659112691877, "time": 0.636967945098877, "epoch": 8, "memory": 44138, "step": 28659} +{"lr": 2.5192782166224308e-05, "data_time": 0.0016625642776489258, "loss": 0.3501308798789978, "time": 0.6367563486099244, "epoch": 8, "memory": 44138, "step": 28759} +{"lr": 2.492472001392276e-05, "data_time": 0.0017060279846191407, "loss": 0.3493047595024109, "time": 0.637101149559021, "epoch": 8, "memory": 44138, "step": 28859} +{"lr": 2.4658572894745857e-05, "data_time": 0.0016996622085571288, "loss": 0.34435219764709474, "time": 0.6361613512039185, "epoch": 8, "memory": 44138, "step": 28959} +{"lr": 2.439435775551997e-05, "data_time": 0.0018393993377685547, "loss": 0.3414866477251053, "time": 0.636789059638977, "epoch": 8, "memory": 44138, "step": 29059} +{"lr": 2.4132091420053357e-05, "data_time": 0.002955722808837891, "loss": 0.33168233186006546, "time": 0.6387169122695923, "epoch": 8, "memory": 44138, "step": 29159} +{"lr": 2.387179058806494e-05, "data_time": 0.0017844438552856445, "loss": 0.31749085783958436, "time": 0.6382629156112671, "epoch": 8, "memory": 44138, "step": 29259} +{"lr": 2.361347183412073e-05, "data_time": 0.001632523536682129, "loss": 0.33510020971298216, "time": 0.6372994899749755, "epoch": 8, "memory": 44138, "step": 29359} +{"lr": 2.3357151606578728e-05, "data_time": 0.0015772104263305664, "loss": 0.333350333571434, "time": 0.6379689455032349, "epoch": 8, "memory": 44138, "step": 29459} +{"lr": 2.3102846226541355e-05, "data_time": 0.002072000503540039, "loss": 0.3356959939002991, "time": 0.6372670888900757, "epoch": 8, "memory": 44138, "step": 29559} +{"lr": 2.2850571886816365e-05, "data_time": 0.00260622501373291, "loss": 0.3276965469121933, "time": 0.6381346940994262, "epoch": 8, "memory": 44138, "step": 29659} +{"lr": 2.260034465088568e-05, "data_time": 0.0017112016677856446, "loss": 0.3511845350265503, "time": 0.6372352123260498, "epoch": 8, "memory": 44138, "step": 29759} +{"lr": 2.2352180451882643e-05, "data_time": 0.0021218776702880858, "loss": 0.3321383535861969, "time": 0.6360498666763306, "epoch": 8, "memory": 44138, "step": 29859} +{"lr": 2.210609509157741e-05, "data_time": 0.0017811059951782227, "loss": 0.34005952179431914, "time": 0.6362974882125855, "epoch": 8, "memory": 44138, "step": 29959} +{"lr": 2.1862104239370772e-05, "data_time": 0.0017603158950805664, "loss": 0.33925525546073915, "time": 0.6381688117980957, "epoch": 8, "memory": 44138, "step": 30059} +{"lr": 2.1620223431296407e-05, "data_time": 0.001773238182067871, "loss": 0.34935729801654813, "time": 0.6371684312820435, "epoch": 8, "memory": 44138, "step": 30159} +{"lr": 2.1380468069031744e-05, "data_time": 0.0017689943313598632, "loss": 0.3472258061170578, "time": 0.6456415891647339, "epoch": 8, "memory": 44138, "step": 30259} +{"lr": 2.1142853418917185e-05, "data_time": 0.0017083168029785156, "loss": 0.3159309595823288, "time": 0.6369046211242676, "epoch": 8, "memory": 44138, "step": 30359} +{"lr": 2.0907394610983892e-05, "data_time": 0.0015600919723510742, "loss": 0.3736870944499969, "time": 0.6376642704010009, "epoch": 8, "memory": 44138, "step": 30459} +{"lr": 2.0674106637990643e-05, "data_time": 0.0015815496444702148, "loss": 0.32337228059768675, "time": 0.6369141101837158, "epoch": 8, "memory": 44138, "step": 30559} +{"lr": 2.0443004354468975e-05, "data_time": 0.0016054630279541016, "loss": 0.33981326818466184, "time": 0.6369661092758179, "epoch": 8, "memory": 44138, "step": 30659} +{"lr": 2.0214102475777453e-05, "data_time": 0.002117443084716797, "loss": 0.3353619068861008, "time": 0.6368308067321777, "epoch": 8, "memory": 44138, "step": 30759} +{"lr": 1.9987415577164583e-05, "data_time": 0.002645564079284668, "loss": 0.34488191902637483, "time": 0.6388916492462158, "epoch": 8, "memory": 44138, "step": 30859} +{"lr": 1.9762958092840814e-05, "data_time": 0.0018764257431030274, "loss": 0.3224926322698593, "time": 0.6365635395050049, "epoch": 8, "memory": 44138, "step": 30959} +{"lr": 1.9540744315059334e-05, "data_time": 0.001518082618713379, "loss": 0.3325134128332138, "time": 0.6377947330474854, "epoch": 8, "memory": 44138, "step": 31059} +{"lr": 1.9320788393206132e-05, "data_time": 0.001732802391052246, "loss": 0.30041895508766175, "time": 0.6368714809417725, "epoch": 8, "memory": 44138, "step": 31159} +{"lr": 1.9103104332899027e-05, "data_time": 0.0014935731887817383, "loss": 0.3463741153478622, "time": 0.6371479272842407, "epoch": 8, "memory": 44138, "step": 31259} +{"lr": 1.888770599509573e-05, "data_time": 0.0019794464111328124, "loss": 0.34848603010177615, "time": 0.6449602365493774, "epoch": 8, "memory": 44138, "step": 31359} +{"lr": 1.8674607095211516e-05, "data_time": 0.002109527587890625, "loss": 0.3407303631305695, "time": 0.6373261451721192, "epoch": 8, "memory": 44138, "step": 31459} +{"accuracy/top1": 94.44000244140625, "data_time": 0.014360251097843564, "time": 1.2575075009773518, "step": 8} +{"lr": 1.838641934548488e-05, "data_time": 0.001876235008239746, "loss": 0.34708923697471616, "time": 0.6379916429519653, "epoch": 9, "memory": 44138, "step": 31596} +{"lr": 1.817882403185378e-05, "data_time": 0.0023573875427246095, "loss": 0.36315891742706297, "time": 0.6379354953765869, "epoch": 9, "memory": 44138, "step": 31696} +{"lr": 1.797357329395729e-05, "data_time": 0.0017489194869995117, "loss": 0.34651450216770174, "time": 0.6378476858139038, "epoch": 9, "memory": 44138, "step": 31796} +{"lr": 1.7770680201065646e-05, "data_time": 0.0018070459365844727, "loss": 0.3591558516025543, "time": 0.6380157470703125, "epoch": 9, "memory": 44138, "step": 31896} +{"lr": 1.7570157672326885e-05, "data_time": 0.0016509294509887695, "loss": 0.33588482439517975, "time": 0.6379043579101562, "epoch": 9, "memory": 44138, "step": 31996} +{"lr": 1.7372018475944145e-05, "data_time": 0.0018067598342895509, "loss": 0.3351457595825195, "time": 0.6384174823760986, "epoch": 9, "memory": 44138, "step": 32096} +{"lr": 1.717627522836271e-05, "data_time": 0.0019103050231933593, "loss": 0.35640172362327577, "time": 0.6380068540573121, "epoch": 9, "memory": 44138, "step": 32196} +{"lr": 1.6982940393466685e-05, "data_time": 0.0015610694885253907, "loss": 0.3528045743703842, "time": 0.6377854824066163, "epoch": 9, "memory": 44138, "step": 32296} +{"lr": 1.6792026281785218e-05, "data_time": 0.0020467042922973633, "loss": 0.3167236477136612, "time": 0.6371350765228272, "epoch": 9, "memory": 44138, "step": 32396} +{"lr": 1.660354504970888e-05, "data_time": 0.0024530649185180663, "loss": 0.3179640740156174, "time": 0.6366666078567504, "epoch": 9, "memory": 44138, "step": 32496} +{"lr": 1.641750869871534e-05, "data_time": 0.0016213178634643555, "loss": 0.3418331354856491, "time": 0.6368302345275879, "epoch": 9, "memory": 44138, "step": 32596} +{"lr": 1.6233929074605374e-05, "data_time": 0.0017043352127075195, "loss": 0.34667418599128724, "time": 0.636782455444336, "epoch": 9, "memory": 44138, "step": 32696} +{"lr": 1.6052817866748513e-05, "data_time": 0.0015819787979125977, "loss": 0.3445913285017014, "time": 0.6372241020202637, "epoch": 9, "memory": 44138, "step": 32796} +{"lr": 1.587418660733871e-05, "data_time": 0.0015439748764038085, "loss": 0.33370116651058196, "time": 0.636690616607666, "epoch": 9, "memory": 44138, "step": 32896} +{"lr": 1.5698046670660044e-05, "data_time": 0.0015928506851196288, "loss": 0.33526529371738434, "time": 0.6363592624664307, "epoch": 9, "memory": 44138, "step": 32996} +{"lr": 1.5524409272362524e-05, "data_time": 0.001676321029663086, "loss": 0.329180970788002, "time": 0.6364695787429809, "epoch": 9, "memory": 44138, "step": 33096} +{"lr": 1.5353285468747853e-05, "data_time": 0.0018772840499877929, "loss": 0.3220287948846817, "time": 0.6375333786010742, "epoch": 9, "memory": 44138, "step": 33196} +{"lr": 1.5184686156065442e-05, "data_time": 0.0019403457641601562, "loss": 0.36585857570171354, "time": 0.6365541219711304, "epoch": 9, "memory": 44138, "step": 33296} +{"lr": 1.5018622069818636e-05, "data_time": 0.0016704559326171874, "loss": 0.3399368613958359, "time": 0.6365461111068725, "epoch": 9, "memory": 44138, "step": 33396} +{"lr": 1.4855103784081099e-05, "data_time": 0.0017790555953979491, "loss": 0.3403072267770767, "time": 0.6376025199890136, "epoch": 9, "memory": 44138, "step": 33496} +{"lr": 1.4694141710823508e-05, "data_time": 0.001851367950439453, "loss": 0.3264953374862671, "time": 0.6371854066848754, "epoch": 9, "memory": 44138, "step": 33596} +{"lr": 1.4535746099250614e-05, "data_time": 0.0017159461975097656, "loss": 0.323648601770401, "time": 0.6367975950241089, "epoch": 9, "memory": 44138, "step": 33696} +{"lr": 1.4379927035148576e-05, "data_time": 0.001768183708190918, "loss": 0.3227705806493759, "time": 0.6373627185821533, "epoch": 9, "memory": 44138, "step": 33796} +{"lr": 1.42266944402428e-05, "data_time": 0.0017513513565063476, "loss": 0.3449145913124084, "time": 0.637163496017456, "epoch": 9, "memory": 44138, "step": 33896} +{"lr": 1.4076058071566133e-05, "data_time": 0.0014359474182128907, "loss": 0.3214793801307678, "time": 0.6374832630157471, "epoch": 9, "memory": 44138, "step": 33996} +{"lr": 1.3928027520837635e-05, "data_time": 0.001459956169128418, "loss": 0.3619404137134552, "time": 0.6371097326278686, "epoch": 9, "memory": 44138, "step": 34096} +{"lr": 1.3782612213851759e-05, "data_time": 0.001634979248046875, "loss": 0.3587342739105225, "time": 0.6369773626327515, "epoch": 9, "memory": 44138, "step": 34196} +{"lr": 1.3639821409878263e-05, "data_time": 0.001958751678466797, "loss": 0.36187497079372405, "time": 0.6373254537582398, "epoch": 9, "memory": 44138, "step": 34296} +{"lr": 1.3499664201072546e-05, "data_time": 0.0015430927276611328, "loss": 0.3353217601776123, "time": 0.6366066694259643, "epoch": 9, "memory": 44138, "step": 34396} +{"lr": 1.336214951189674e-05, "data_time": 0.0016913890838623046, "loss": 0.35445632338523864, "time": 0.6360156059265136, "epoch": 9, "memory": 44138, "step": 34496} +{"lr": 1.3227286098551456e-05, "data_time": 0.0017403125762939452, "loss": 0.3315434277057648, "time": 0.6366602659225464, "epoch": 9, "memory": 44138, "step": 34596} +{"lr": 1.3095082548418224e-05, "data_time": 0.0015049219131469727, "loss": 0.3482250928878784, "time": 0.636380672454834, "epoch": 9, "memory": 44138, "step": 34696} +{"lr": 1.2965547279512709e-05, "data_time": 0.001609206199645996, "loss": 0.31899039447307587, "time": 0.6360365152359009, "epoch": 9, "memory": 44138, "step": 34796} +{"lr": 1.2838688539948659e-05, "data_time": 0.0014748811721801759, "loss": 0.34317450523376464, "time": 0.637748908996582, "epoch": 9, "memory": 44138, "step": 34896} +{"lr": 1.2714514407412763e-05, "data_time": 0.0015018701553344727, "loss": 0.36702142357826234, "time": 0.6384203910827637, "epoch": 9, "memory": 44138, "step": 34996} +{"lr": 1.2593032788650256e-05, "data_time": 0.002281594276428223, "loss": 0.33379718363285066, "time": 0.6369554996490479, "epoch": 9, "memory": 44138, "step": 35096} +{"lr": 1.2474251418961484e-05, "data_time": 0.0015685319900512694, "loss": 0.3442933946847916, "time": 0.6369629621505737, "epoch": 9, "memory": 44138, "step": 35196} +{"lr": 1.2358177861709387e-05, "data_time": 0.0015196800231933594, "loss": 0.33880884349346163, "time": 0.6360907793045044, "epoch": 9, "memory": 44138, "step": 35296} +{"lr": 1.2244819507837855e-05, "data_time": 0.0014713048934936524, "loss": 0.3282599478960037, "time": 0.6371447563171386, "epoch": 9, "memory": 44138, "step": 35396} +{"accuracy/top1": 94.74966430664062, "data_time": 0.018098025486387056, "time": 1.2612294616370365, "step": 9} +{"lr": 1.2093939694118031e-05, "data_time": 0.0017571210861206054, "loss": 0.3731055498123169, "time": 0.6378751277923584, "epoch": 10, "memory": 44138, "step": 35533} +{"lr": 1.1987044880988498e-05, "data_time": 0.0017752885818481446, "loss": 0.3633056372404099, "time": 0.6374515295028687, "epoch": 10, "memory": 44138, "step": 35633} +{"lr": 1.1882888903004625e-05, "data_time": 0.0017711400985717773, "loss": 0.3484418958425522, "time": 0.6378175497055054, "epoch": 10, "memory": 44138, "step": 35733} +{"lr": 1.178147839226253e-05, "data_time": 0.0014324665069580078, "loss": 0.34249356389045715, "time": 0.6419201612472534, "epoch": 10, "memory": 44138, "step": 35833} +{"lr": 1.1682819806041682e-05, "data_time": 0.0020990848541259767, "loss": 0.33985520601272584, "time": 0.6377373695373535, "epoch": 10, "memory": 44138, "step": 35933} +{"lr": 1.1586919426393653e-05, "data_time": 0.0016060113906860352, "loss": 0.36118035614490507, "time": 0.6364832639694213, "epoch": 10, "memory": 44138, "step": 36033} +{"lr": 1.1493783359742196e-05, "data_time": 0.0015542745590209962, "loss": 0.32094216644763945, "time": 0.6369356155395508, "epoch": 10, "memory": 44138, "step": 36133} +{"lr": 1.1403417536494367e-05, "data_time": 0.0015214920043945313, "loss": 0.37214413583278655, "time": 0.6370633363723754, "epoch": 10, "memory": 44138, "step": 36233} +{"lr": 1.1315827710662908e-05, "data_time": 0.0016398668289184571, "loss": 0.316905614733696, "time": 0.6368438482284546, "epoch": 10, "memory": 44138, "step": 36333} +{"lr": 1.1231019459499938e-05, "data_time": 0.0014683723449707032, "loss": 0.3269952654838562, "time": 0.6371765613555909, "epoch": 10, "memory": 44138, "step": 36433} +{"lr": 1.1148998183141688e-05, "data_time": 0.0015017986297607422, "loss": 0.3335738182067871, "time": 0.6380161046981812, "epoch": 10, "memory": 44138, "step": 36533} +{"lr": 1.1069769104264779e-05, "data_time": 0.001564335823059082, "loss": 0.33867660760879514, "time": 0.6375886678695679, "epoch": 10, "memory": 44138, "step": 36633} +{"lr": 1.0993337267753594e-05, "data_time": 0.0017877817153930664, "loss": 0.3430941581726074, "time": 0.638480830192566, "epoch": 10, "memory": 44138, "step": 36733} +{"lr": 1.0919707540379088e-05, "data_time": 0.0019340276718139648, "loss": 0.348739692568779, "time": 0.6372841358184814, "epoch": 10, "memory": 44138, "step": 36833} +{"lr": 1.0848884610488858e-05, "data_time": 0.0015627384185791016, "loss": 0.3504202350974083, "time": 0.6378693103790283, "epoch": 10, "memory": 44138, "step": 36933} +{"lr": 1.0780872987708645e-05, "data_time": 0.001573324203491211, "loss": 0.3205928891897202, "time": 0.6368873834609985, "epoch": 10, "memory": 44138, "step": 37033} +{"lr": 1.0715677002655166e-05, "data_time": 0.001503324508666992, "loss": 0.3238894432783127, "time": 0.6375768661499024, "epoch": 10, "memory": 44138, "step": 37133} +{"lr": 1.0653300806660372e-05, "data_time": 0.0014631986618041993, "loss": 0.3422483205795288, "time": 0.6371843576431274, "epoch": 10, "memory": 44138, "step": 37233} +{"lr": 1.0593748371507151e-05, "data_time": 0.0017275571823120116, "loss": 0.34399558007717135, "time": 0.637434458732605, "epoch": 10, "memory": 44138, "step": 37333} +{"lr": 1.0537023489176325e-05, "data_time": 0.002298450469970703, "loss": 0.33317987620830536, "time": 0.6367875814437867, "epoch": 10, "memory": 44138, "step": 37433} +{"lr": 1.0483129771605328e-05, "data_time": 0.0015595674514770508, "loss": 0.32792393267154696, "time": 0.6376729011535645, "epoch": 10, "memory": 44138, "step": 37533} +{"lr": 1.0432070650458095e-05, "data_time": 0.001458287239074707, "loss": 0.3527918815612793, "time": 0.6363785982131958, "epoch": 10, "memory": 44138, "step": 37633} +{"lr": 1.0383849376906634e-05, "data_time": 0.0017002344131469727, "loss": 0.3225879371166229, "time": 0.6369772911071777, "epoch": 10, "memory": 44138, "step": 37733} +{"lr": 1.0338469021423977e-05, "data_time": 0.0016661882400512695, "loss": 0.3160365730524063, "time": 0.6368574619293212, "epoch": 10, "memory": 44138, "step": 37833} +{"lr": 1.029593247358868e-05, "data_time": 0.001589059829711914, "loss": 0.36551602482795714, "time": 0.638017725944519, "epoch": 10, "memory": 44138, "step": 37933} +{"lr": 1.0256242441900827e-05, "data_time": 0.0017339468002319336, "loss": 0.3381687432527542, "time": 0.6383226871490478, "epoch": 10, "memory": 44138, "step": 38033} +{"lr": 1.0219401453609533e-05, "data_time": 0.0013235092163085937, "loss": 0.3508059173822403, "time": 0.6371890068054199, "epoch": 10, "memory": 44138, "step": 38133} +{"lr": 1.018541185455209e-05, "data_time": 0.0014414787292480469, "loss": 0.3306189596652985, "time": 0.6374693393707276, "epoch": 10, "memory": 44138, "step": 38233} +{"lr": 1.015427580900455e-05, "data_time": 0.0016166210174560548, "loss": 0.3464100152254105, "time": 0.6381880760192871, "epoch": 10, "memory": 44138, "step": 38333} +{"lr": 1.0125995299543914e-05, "data_time": 0.0015708684921264648, "loss": 0.3515030652284622, "time": 0.6374882936477662, "epoch": 10, "memory": 44138, "step": 38433} +{"lr": 1.0100572126921902e-05, "data_time": 0.0018432855606079102, "loss": 0.35819247663021087, "time": 0.6369351387023926, "epoch": 10, "memory": 44138, "step": 38533} +{"lr": 1.007800790995031e-05, "data_time": 0.001888418197631836, "loss": 0.3579746216535568, "time": 0.6378156900405884, "epoch": 10, "memory": 44138, "step": 38633} +{"lr": 1.0058304085397905e-05, "data_time": 0.0015785694122314453, "loss": 0.3148579865694046, "time": 0.6376957178115845, "epoch": 10, "memory": 44138, "step": 38733} +{"lr": 1.0041461907898941e-05, "data_time": 0.0021996259689331054, "loss": 0.3470492124557495, "time": 0.6371596574783325, "epoch": 10, "memory": 44138, "step": 38833} +{"lr": 1.0027482449873297e-05, "data_time": 0.0019378900527954102, "loss": 0.3348171353340149, "time": 0.6371330976486206, "epoch": 10, "memory": 44138, "step": 38933} +{"lr": 1.0016366601458162e-05, "data_time": 0.002123451232910156, "loss": 0.35173400938510896, "time": 0.6460501909255981, "epoch": 10, "memory": 44138, "step": 39033} +{"lr": 1.0008115070451358e-05, "data_time": 0.0018494367599487306, "loss": 0.3442584753036499, "time": 0.6565184354782104, "epoch": 10, "memory": 44138, "step": 39133} +{"lr": 1.0002728382266284e-05, "data_time": 0.0016611337661743164, "loss": 0.3185924857854843, "time": 0.6374598264694213, "epoch": 10, "memory": 44138, "step": 39233} +{"lr": 1.0000206879898465e-05, "data_time": 0.001641702651977539, "loss": 0.33281290233135224, "time": 0.6364509344100953, "epoch": 10, "memory": 44138, "step": 39333} +{"accuracy/top1": 94.77053833007812, "data_time": 0.01525562385032917, "time": 1.2584579319789493, "step": 10} diff --git a/clip_large_pretrain_4x256_IF_lr1e-4/20230606_050006/vis_data/config.py b/clip_large_pretrain_4x256_IF_lr1e-4/20230606_050006/vis_data/config.py new file mode 100644 index 0000000000000000000000000000000000000000..98cdad2fa2783bf5a6865ddc1352c928f69609b0 --- /dev/null +++ b/clip_large_pretrain_4x256_IF_lr1e-4/20230606_050006/vis_data/config.py @@ -0,0 +1,251 @@ +optim_wrapper = dict( + optimizer=dict( + type='AdamW', lr=0.0001, weight_decay=0.3, _scope_='mmpretrain'), + paramwise_cfg=dict( + custom_keys=dict({ + '.cls_token': dict(decay_mult=0.0), + '.pos_embed': dict(decay_mult=0.0) + })), + type='AmpOptimWrapper', + dtype='bfloat16', + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=False, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=4096) +model = dict( + type='ImageClassifier', + backbone=dict( + frozen_stages=24, + type='VisionTransformer', + arch='l', + img_size=224, + patch_size=14, + drop_rate=0.1, + pre_norm=True, + final_norm=False, + init_cfg=dict( + type='Pretrained', + checkpoint='ckpt/openclip-ViT-L-14.pth', + prefix='backbone')), + neck=dict( + type='CLIPProjection', + in_channels=1024, + out_channels=768, + init_cfg=dict( + type='Pretrained', + checkpoint='ckpt/openclip-ViT-L-14.pth', + prefix='backbone')), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=768, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + init_cfg=None), + init_cfg=dict( + type='TruncNormal', layer=['Conv2d', 'Linear'], std=0.02, bias=0.0), + train_cfg=None) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=128, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/IF80w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/if-dpmsolver++-50-20w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/cc1m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=128, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/if-dpmsolver++-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=128, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/if-dpmsolver++-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/clip_large_pretrain_4x256_IF_lr1e-4' diff --git a/clip_large_pretrain_4x256_IF_lr1e-4/20230606_050006/vis_data/events.out.tfevents.1685998820.SH-IDC1-10-140-24-119.202894.0 b/clip_large_pretrain_4x256_IF_lr1e-4/20230606_050006/vis_data/events.out.tfevents.1685998820.SH-IDC1-10-140-24-119.202894.0 new file mode 100644 index 0000000000000000000000000000000000000000..f1a4a2ebcbe344ed3d7a29c9a360f5d4820a7417 --- /dev/null +++ b/clip_large_pretrain_4x256_IF_lr1e-4/20230606_050006/vis_data/events.out.tfevents.1685998820.SH-IDC1-10-140-24-119.202894.0 @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ec45ff2b774be6100b4402d7111551a677325e6fee12ae7a0bc1a9a0b3ad6244 +size 1930259 diff --git a/clip_large_pretrain_4x256_IF_lr1e-4/20230606_050006/vis_data/scalars.json b/clip_large_pretrain_4x256_IF_lr1e-4/20230606_050006/vis_data/scalars.json new file mode 100644 index 0000000000000000000000000000000000000000..4e8cd5172c7d8dc61dbd21b8f83e0ab59b30df7c --- /dev/null +++ b/clip_large_pretrain_4x256_IF_lr1e-4/20230606_050006/vis_data/scalars.json @@ -0,0 +1,400 @@ +{"lr": 9.999859583040624e-05, "data_time": 0.0016723871231079102, "loss": 0.512164306640625, "time": 0.6354817628860474, "epoch": 1, "memory": 44139, "step": 100} +{"lr": 9.999432653394512e-05, "data_time": 0.001458597183227539, "loss": 0.49367055892944334, "time": 0.6365986347198487, "epoch": 1, "memory": 44139, "step": 200} +{"lr": 9.998719223919393e-05, "data_time": 0.0015254497528076171, "loss": 0.47274911403656006, "time": 0.6358645677566528, "epoch": 1, "memory": 44139, "step": 300} +{"lr": 9.997719340042649e-05, "data_time": 0.0016599655151367187, "loss": 0.45169744491577146, "time": 0.6360260486602783, "epoch": 1, "memory": 44139, "step": 400} +{"lr": 9.996433065431542e-05, "data_time": 0.0014626264572143554, "loss": 0.445109036564827, "time": 0.6356816291809082, "epoch": 1, "memory": 44139, "step": 500} +{"lr": 9.994860481989157e-05, "data_time": 0.0015146493911743163, "loss": 0.42883872985839844, "time": 0.6361032485961914, "epoch": 1, "memory": 44139, "step": 600} +{"lr": 9.993001689849214e-05, "data_time": 0.0014402151107788086, "loss": 0.45372450947761533, "time": 0.6367725849151611, "epoch": 1, "memory": 44139, "step": 700} +{"lr": 9.990856807369658e-05, "data_time": 0.0015191078186035157, "loss": 0.4422683477401733, "time": 0.636231541633606, "epoch": 1, "memory": 44139, "step": 800} +{"lr": 9.98842597112514e-05, "data_time": 0.001710963249206543, "loss": 0.4624675452709198, "time": 0.6360510587692261, "epoch": 1, "memory": 44139, "step": 900} +{"lr": 9.985709335898336e-05, "data_time": 0.0016065359115600586, "loss": 0.4689463973045349, "time": 0.6368230581283569, "epoch": 1, "memory": 44139, "step": 1000} +{"lr": 9.982707074670047e-05, "data_time": 0.0016260623931884765, "loss": 0.43320480585098264, "time": 0.6370084762573243, "epoch": 1, "memory": 44139, "step": 1100} +{"lr": 9.979419378608245e-05, "data_time": 0.0014093875885009765, "loss": 0.4505138874053955, "time": 0.6361888647079468, "epoch": 1, "memory": 44139, "step": 1200} +{"lr": 9.975846457055832e-05, "data_time": 0.0014659643173217773, "loss": 0.4286786615848541, "time": 0.6366539239883423, "epoch": 1, "memory": 44139, "step": 1300} +{"lr": 9.97198853751737e-05, "data_time": 0.001408529281616211, "loss": 0.434794282913208, "time": 0.6369958639144897, "epoch": 1, "memory": 44139, "step": 1400} +{"lr": 9.967845865644554e-05, "data_time": 0.0014293432235717774, "loss": 0.41036694049835204, "time": 0.6361128330230713, "epoch": 1, "memory": 44139, "step": 1500} +{"lr": 9.963418705220597e-05, "data_time": 0.001466655731201172, "loss": 0.4314510613679886, "time": 0.6372191905975342, "epoch": 1, "memory": 44139, "step": 1600} +{"lr": 9.958707338143413e-05, "data_time": 0.0015425443649291991, "loss": 0.38765864074230194, "time": 0.6372380971908569, "epoch": 1, "memory": 44139, "step": 1700} +{"lr": 9.953712064407689e-05, "data_time": 0.0015309810638427734, "loss": 0.4207657933235168, "time": 0.636822509765625, "epoch": 1, "memory": 44139, "step": 1800} +{"lr": 9.948433202085752e-05, "data_time": 0.0015765666961669923, "loss": 0.38641782104969025, "time": 0.6369469642639161, "epoch": 1, "memory": 44139, "step": 1900} +{"lr": 9.942871087307363e-05, "data_time": 0.0017748832702636718, "loss": 0.43550218641757965, "time": 0.6362056493759155, "epoch": 1, "memory": 44139, "step": 2000} +{"lr": 9.937026074238262e-05, "data_time": 0.001464986801147461, "loss": 0.4059758365154266, "time": 0.6361548900604248, "epoch": 1, "memory": 44139, "step": 2100} +{"lr": 9.930898535057657e-05, "data_time": 0.0015139818191528321, "loss": 0.416530105471611, "time": 0.6361959457397461, "epoch": 1, "memory": 44139, "step": 2200} +{"lr": 9.924488859934492e-05, "data_time": 0.0017367839813232423, "loss": 0.3778152257204056, "time": 0.6361271858215332, "epoch": 1, "memory": 44139, "step": 2300} +{"lr": 9.917797457002633e-05, "data_time": 0.002050018310546875, "loss": 0.38874633610248566, "time": 0.6379516124725342, "epoch": 1, "memory": 44139, "step": 2400} +{"lr": 9.910824752334866e-05, "data_time": 0.0015123844146728515, "loss": 0.39862638115882876, "time": 0.6360560178756713, "epoch": 1, "memory": 44139, "step": 2500} +{"lr": 9.903571189915766e-05, "data_time": 0.0017220258712768554, "loss": 0.38024311065673827, "time": 0.6397989511489868, "epoch": 1, "memory": 44139, "step": 2600} +{"lr": 9.896037231613424e-05, "data_time": 0.001457810401916504, "loss": 0.4019357919692993, "time": 0.6363687753677368, "epoch": 1, "memory": 44139, "step": 2700} +{"lr": 9.888223357150049e-05, "data_time": 0.0013912439346313477, "loss": 0.3906904995441437, "time": 0.6358689785003662, "epoch": 1, "memory": 44139, "step": 2800} +{"lr": 9.880130064071408e-05, "data_time": 0.001487565040588379, "loss": 0.3792046785354614, "time": 0.6360376596450805, "epoch": 1, "memory": 44139, "step": 2900} +{"lr": 9.87175786771517e-05, "data_time": 0.001390385627746582, "loss": 0.36773664355278013, "time": 0.6364351511001587, "epoch": 1, "memory": 44139, "step": 3000} +{"lr": 9.863107301178059e-05, "data_time": 0.0017130613327026368, "loss": 0.4086927860975266, "time": 0.636230754852295, "epoch": 1, "memory": 44139, "step": 3100} +{"lr": 9.854178915281925e-05, "data_time": 0.0015178680419921874, "loss": 0.4124324083328247, "time": 0.6347718477249146, "epoch": 1, "memory": 44139, "step": 3200} +{"lr": 9.84497327853868e-05, "data_time": 0.0015394926071166993, "loss": 0.38047824501991273, "time": 0.6361316919326783, "epoch": 1, "memory": 44139, "step": 3300} +{"lr": 9.83549097711407e-05, "data_time": 0.0015110731124877929, "loss": 0.38240416944026945, "time": 0.6365596532821656, "epoch": 1, "memory": 44139, "step": 3400} +{"lr": 9.825732614790392e-05, "data_time": 0.0014199256896972657, "loss": 0.37611470818519593, "time": 0.6370939493179322, "epoch": 1, "memory": 44139, "step": 3500} +{"lr": 9.815698812928018e-05, "data_time": 0.0015276193618774414, "loss": 0.3861346751451492, "time": 0.6377681970596314, "epoch": 1, "memory": 44139, "step": 3600} +{"lr": 9.805390210425831e-05, "data_time": 0.0015616416931152344, "loss": 0.38390754759311674, "time": 0.6380697011947631, "epoch": 1, "memory": 44139, "step": 3700} +{"lr": 9.794807463680567e-05, "data_time": 0.0015166997909545898, "loss": 0.38078379034996035, "time": 0.6371499061584472, "epoch": 1, "memory": 44139, "step": 3800} +{"lr": 9.783951246544974e-05, "data_time": 0.0013837337493896485, "loss": 0.37763476371765137, "time": 0.636236023902893, "epoch": 1, "memory": 44139, "step": 3900} +{"accuracy/top1": 80.48085021972656, "data_time": 0.025140733049626936, "time": 1.320618048048856, "step": 1} +{"lr": 9.768635527139677e-05, "data_time": 0.0016125202178955077, "loss": 0.3624943017959595, "time": 0.6393807649612426, "epoch": 2, "memory": 44139, "step": 4037} +{"lr": 9.75713397738286e-05, "data_time": 0.0013150930404663085, "loss": 0.35947518050670624, "time": 0.6368603467941284, "epoch": 2, "memory": 44138, "step": 4137} +{"lr": 9.745361356082025e-05, "data_time": 0.0014085769653320312, "loss": 0.3616087704896927, "time": 0.6367685794830322, "epoch": 2, "memory": 44138, "step": 4237} +{"lr": 9.733318412854784e-05, "data_time": 0.0013958930969238282, "loss": 0.3790826857089996, "time": 0.6389569044113159, "epoch": 2, "memory": 44138, "step": 4337} +{"lr": 9.721005914531412e-05, "data_time": 0.0014826536178588867, "loss": 0.3935104191303253, "time": 0.6482341766357422, "epoch": 2, "memory": 44138, "step": 4437} +{"lr": 9.708424645106017e-05, "data_time": 0.0014542818069458007, "loss": 0.3580289512872696, "time": 0.6364399433135987, "epoch": 2, "memory": 44138, "step": 4537} +{"lr": 9.695575405686603e-05, "data_time": 0.0015553951263427735, "loss": 0.3991315305233002, "time": 0.6370989322662354, "epoch": 2, "memory": 44138, "step": 4637} +{"lr": 9.682459014444089e-05, "data_time": 0.0014152288436889648, "loss": 0.3924710273742676, "time": 0.6358236074447632, "epoch": 2, "memory": 44138, "step": 4737} +{"lr": 9.669076306560176e-05, "data_time": 0.0014875173568725587, "loss": 0.3728522062301636, "time": 0.638262152671814, "epoch": 2, "memory": 44138, "step": 4837} +{"lr": 9.655428134174183e-05, "data_time": 0.0015720367431640626, "loss": 0.32827370166778563, "time": 0.6390895128250123, "epoch": 2, "memory": 44138, "step": 4937} +{"lr": 9.641515366328807e-05, "data_time": 0.0014989137649536132, "loss": 0.38309581875801085, "time": 0.6372867584228515, "epoch": 2, "memory": 44138, "step": 5037} +{"lr": 9.627338888914757e-05, "data_time": 0.0014197111129760742, "loss": 0.389060977101326, "time": 0.6368171691894531, "epoch": 2, "memory": 44138, "step": 5137} +{"lr": 9.612899604614378e-05, "data_time": 0.0013750791549682617, "loss": 0.37170601189136504, "time": 0.6360954284667969, "epoch": 2, "memory": 44138, "step": 5237} +{"lr": 9.598198432844131e-05, "data_time": 0.0014049768447875976, "loss": 0.36447584331035615, "time": 0.6362692594528199, "epoch": 2, "memory": 44138, "step": 5337} +{"lr": 9.583236309696081e-05, "data_time": 0.001524806022644043, "loss": 0.3516468346118927, "time": 0.6364728450775147, "epoch": 2, "memory": 44138, "step": 5437} +{"lr": 9.568014187878279e-05, "data_time": 0.001523447036743164, "loss": 0.3908290058374405, "time": 0.6365160942077637, "epoch": 2, "memory": 44138, "step": 5537} +{"lr": 9.552533036654103e-05, "data_time": 0.001683664321899414, "loss": 0.33294848799705506, "time": 0.6374480962753296, "epoch": 2, "memory": 44138, "step": 5637} +{"lr": 9.536793841780541e-05, "data_time": 0.0015509366989135743, "loss": 0.3408015429973602, "time": 0.6381877183914184, "epoch": 2, "memory": 44138, "step": 5737} +{"lr": 9.52079760544543e-05, "data_time": 0.0017657995223999024, "loss": 0.3576482445001602, "time": 0.6383677005767823, "epoch": 2, "memory": 44138, "step": 5837} +{"lr": 9.504545346203619e-05, "data_time": 0.0017487049102783204, "loss": 0.3597940683364868, "time": 0.6362926721572876, "epoch": 2, "memory": 44138, "step": 5937} +{"lr": 9.488038098912122e-05, "data_time": 0.0016869544982910157, "loss": 0.33743955194950104, "time": 0.637329888343811, "epoch": 2, "memory": 44138, "step": 6037} +{"lr": 9.471276914664257e-05, "data_time": 0.0015462160110473633, "loss": 0.3715079724788666, "time": 0.6368890762329101, "epoch": 2, "memory": 44138, "step": 6137} +{"lr": 9.454262860722667e-05, "data_time": 0.0015308618545532226, "loss": 0.35632383823394775, "time": 0.636843204498291, "epoch": 2, "memory": 44138, "step": 6237} +{"lr": 9.436997020451398e-05, "data_time": 0.0014986515045166016, "loss": 0.3385142832994461, "time": 0.6363614797592163, "epoch": 2, "memory": 44138, "step": 6337} +{"lr": 9.419480493246903e-05, "data_time": 0.0016518115997314453, "loss": 0.3685177147388458, "time": 0.6366246461868286, "epoch": 2, "memory": 44138, "step": 6437} +{"lr": 9.401714394468036e-05, "data_time": 0.0014364004135131836, "loss": 0.3705434799194336, "time": 0.636694860458374, "epoch": 2, "memory": 44138, "step": 6537} +{"lr": 9.383699855365022e-05, "data_time": 0.002483630180358887, "loss": 0.3419413596391678, "time": 0.6372604608535767, "epoch": 2, "memory": 44138, "step": 6637} +{"lr": 9.365438023007469e-05, "data_time": 0.0015980005264282227, "loss": 0.3687066316604614, "time": 0.6379871606826782, "epoch": 2, "memory": 44138, "step": 6737} +{"lr": 9.346930060211267e-05, "data_time": 0.001495051383972168, "loss": 0.3346828520298004, "time": 0.6378123998641968, "epoch": 2, "memory": 44138, "step": 6837} +{"lr": 9.328177145464599e-05, "data_time": 0.0016396522521972656, "loss": 0.36425608992576597, "time": 0.6370981931686401, "epoch": 2, "memory": 44138, "step": 6937} +{"lr": 9.309180472852871e-05, "data_time": 0.00156095027923584, "loss": 0.32951463758945465, "time": 0.636542558670044, "epoch": 2, "memory": 44138, "step": 7037} +{"lr": 9.289941251982683e-05, "data_time": 0.0017425060272216798, "loss": 0.33361624777317045, "time": 0.637116551399231, "epoch": 2, "memory": 44138, "step": 7137} +{"lr": 9.270460707904827e-05, "data_time": 0.00169064998626709, "loss": 0.33794479072093964, "time": 0.6367788314819336, "epoch": 2, "memory": 44138, "step": 7237} +{"lr": 9.250740081036243e-05, "data_time": 0.0013814210891723634, "loss": 0.35136326849460603, "time": 0.6368788719177246, "epoch": 2, "memory": 44138, "step": 7337} +{"lr": 9.230780627081088e-05, "data_time": 0.001467418670654297, "loss": 0.3443381518125534, "time": 0.6376749753952027, "epoch": 2, "memory": 44138, "step": 7437} +{"lr": 9.210583616950722e-05, "data_time": 0.0014742136001586914, "loss": 0.33476702272892, "time": 0.636881160736084, "epoch": 2, "memory": 44138, "step": 7537} +{"lr": 9.190150336682842e-05, "data_time": 0.0015477418899536132, "loss": 0.3649121254682541, "time": 0.6370478868484497, "epoch": 2, "memory": 44138, "step": 7637} +{"lr": 9.16948208735955e-05, "data_time": 0.001503157615661621, "loss": 0.3500199973583221, "time": 0.6366247653961181, "epoch": 2, "memory": 44138, "step": 7737} +{"lr": 9.148580185024521e-05, "data_time": 0.001553821563720703, "loss": 0.3575607657432556, "time": 0.6374613523483277, "epoch": 2, "memory": 44138, "step": 7837} +{"accuracy/top1": 78.83163452148438, "data_time": 0.016518901134359425, "time": 1.2614407210514462, "step": 2} +{"lr": 9.119567684509738e-05, "data_time": 0.0015429258346557617, "loss": 0.34105669558048246, "time": 0.6430653810501099, "epoch": 3, "memory": 44138, "step": 7974} +{"lr": 9.098117366468421e-05, "data_time": 0.001436758041381836, "loss": 0.33879759907722473, "time": 0.6373371124267578, "epoch": 3, "memory": 44138, "step": 8074} +{"lr": 9.076437939539471e-05, "data_time": 0.001592397689819336, "loss": 0.359617480635643, "time": 0.6390689849853516, "epoch": 3, "memory": 44138, "step": 8174} +{"lr": 9.05453078415294e-05, "data_time": 0.0015715360641479492, "loss": 0.3803259253501892, "time": 0.6369717359542847, "epoch": 3, "memory": 44138, "step": 8274} +{"lr": 9.032397295239427e-05, "data_time": 0.0015054702758789062, "loss": 0.333760067820549, "time": 0.636887502670288, "epoch": 3, "memory": 44138, "step": 8374} +{"lr": 9.010038882141241e-05, "data_time": 0.0015065431594848632, "loss": 0.35237073004245756, "time": 0.6364066362380981, "epoch": 3, "memory": 44138, "step": 8474} +{"lr": 8.987456968522655e-05, "data_time": 0.0018047094345092773, "loss": 0.33100546300411227, "time": 0.6371907472610474, "epoch": 3, "memory": 44138, "step": 8574} +{"lr": 8.964652992279262e-05, "data_time": 0.0015096664428710938, "loss": 0.3385528564453125, "time": 0.6365949392318726, "epoch": 3, "memory": 44138, "step": 8674} +{"lr": 8.941628405446428e-05, "data_time": 0.0016623258590698243, "loss": 0.3177237302064896, "time": 0.6368241786956788, "epoch": 3, "memory": 44138, "step": 8774} +{"lr": 8.918384674106797e-05, "data_time": 0.0014980077743530274, "loss": 0.3489371448755264, "time": 0.6374441623687744, "epoch": 3, "memory": 44138, "step": 8874} +{"lr": 8.894923278296992e-05, "data_time": 0.0016621828079223632, "loss": 0.33893647193908694, "time": 0.6475806951522827, "epoch": 3, "memory": 44138, "step": 8974} +{"lr": 8.871245711913324e-05, "data_time": 0.001572585105895996, "loss": 0.3414228200912476, "time": 0.6364705324172973, "epoch": 3, "memory": 44138, "step": 9074} +{"lr": 8.847353482616703e-05, "data_time": 0.0015491962432861329, "loss": 0.3254220008850098, "time": 0.6401185989379883, "epoch": 3, "memory": 44138, "step": 9174} +{"lr": 8.823248111736638e-05, "data_time": 0.0014464616775512694, "loss": 0.34861191511154177, "time": 0.6371772766113282, "epoch": 3, "memory": 44138, "step": 9274} +{"lr": 8.798931134174327e-05, "data_time": 0.0014491796493530274, "loss": 0.3306316822767258, "time": 0.6372756004333496, "epoch": 3, "memory": 44138, "step": 9374} +{"lr": 8.774404098304961e-05, "data_time": 0.0015141487121582032, "loss": 0.32666544020175936, "time": 0.6364907026290894, "epoch": 3, "memory": 44138, "step": 9474} +{"lr": 8.749668565879136e-05, "data_time": 0.0017288684844970702, "loss": 0.34261976182460785, "time": 0.6382769823074341, "epoch": 3, "memory": 44138, "step": 9574} +{"lr": 8.724726111923386e-05, "data_time": 0.0014741897583007812, "loss": 0.3573506295681, "time": 0.6365964651107788, "epoch": 3, "memory": 44138, "step": 9674} +{"lr": 8.699578324639898e-05, "data_time": 0.0018654346466064453, "loss": 0.35222489535808565, "time": 0.638684868812561, "epoch": 3, "memory": 44138, "step": 9774} +{"lr": 8.67422680530539e-05, "data_time": 0.0015665531158447266, "loss": 0.3136993333697319, "time": 0.6389204502105713, "epoch": 3, "memory": 44138, "step": 9874} +{"lr": 8.648673168169147e-05, "data_time": 0.001727914810180664, "loss": 0.3321169525384903, "time": 0.6371656894683838, "epoch": 3, "memory": 44138, "step": 9974} +{"lr": 8.622919040350244e-05, "data_time": 0.0015274524688720704, "loss": 0.3592831283807755, "time": 0.6371702671051025, "epoch": 3, "memory": 44138, "step": 10074} +{"lr": 8.596966061733922e-05, "data_time": 0.0016551017761230469, "loss": 0.3653866440057755, "time": 0.6370359659194946, "epoch": 3, "memory": 44138, "step": 10174} +{"lr": 8.570815884867185e-05, "data_time": 0.0015550613403320312, "loss": 0.34676103591918944, "time": 0.6360797643661499, "epoch": 3, "memory": 44138, "step": 10274} +{"lr": 8.544470174853563e-05, "data_time": 0.0014838695526123047, "loss": 0.33594187796115876, "time": 0.636370038986206, "epoch": 3, "memory": 44138, "step": 10374} +{"lr": 8.517930609247077e-05, "data_time": 0.0014545917510986328, "loss": 0.3341986835002899, "time": 0.6366753101348877, "epoch": 3, "memory": 44138, "step": 10474} +{"lr": 8.491198877945477e-05, "data_time": 0.0014988899230957032, "loss": 0.3113416790962219, "time": 0.638522744178772, "epoch": 3, "memory": 44138, "step": 10574} +{"lr": 8.464276683082565e-05, "data_time": 0.0016964435577392577, "loss": 0.3383102774620056, "time": 0.6390778303146363, "epoch": 3, "memory": 44138, "step": 10674} +{"lr": 8.437165738919852e-05, "data_time": 0.0013664484024047852, "loss": 0.3223652243614197, "time": 0.6382468461990356, "epoch": 3, "memory": 44138, "step": 10774} +{"lr": 8.409867771737398e-05, "data_time": 0.0015694379806518554, "loss": 0.33741885125637056, "time": 0.6373280048370361, "epoch": 3, "memory": 44138, "step": 10874} +{"lr": 8.382384519723895e-05, "data_time": 0.0016610860824584962, "loss": 0.3260244011878967, "time": 0.6372844696044921, "epoch": 3, "memory": 44138, "step": 10974} +{"lr": 8.35471773286596e-05, "data_time": 0.0015014171600341796, "loss": 0.32323668897151947, "time": 0.6373264789581299, "epoch": 3, "memory": 44138, "step": 11074} +{"lr": 8.326869172836765e-05, "data_time": 0.0017076730728149414, "loss": 0.3508055955171585, "time": 0.6367055177688599, "epoch": 3, "memory": 44138, "step": 11174} +{"lr": 8.298840612883789e-05, "data_time": 0.0014889240264892578, "loss": 0.34529263377189634, "time": 0.6364981412887574, "epoch": 3, "memory": 44138, "step": 11274} +{"lr": 8.270633837715962e-05, "data_time": 0.0015820026397705077, "loss": 0.3595623314380646, "time": 0.6367258548736572, "epoch": 3, "memory": 44138, "step": 11374} +{"lr": 8.242250643390022e-05, "data_time": 0.0022446632385253905, "loss": 0.31640669256448745, "time": 0.6418020248413085, "epoch": 3, "memory": 44138, "step": 11474} +{"lr": 8.2136928371961e-05, "data_time": 0.0017303705215454101, "loss": 0.3379951626062393, "time": 0.6376960277557373, "epoch": 3, "memory": 44138, "step": 11574} +{"lr": 8.184962237542712e-05, "data_time": 0.0016300678253173828, "loss": 0.3181945890188217, "time": 0.6373231410980225, "epoch": 3, "memory": 44138, "step": 11674} +{"lr": 8.156060673840912e-05, "data_time": 0.0016222000122070312, "loss": 0.34422839283943174, "time": 0.6368931770324707, "epoch": 3, "memory": 44138, "step": 11774} +{"accuracy/top1": 82.16485595703125, "data_time": 0.015650802645190007, "time": 1.2585819014187516, "step": 3} +{"lr": 8.116191337909002e-05, "data_time": 0.0031123638153076174, "loss": 0.338031068444252, "time": 0.6381591558456421, "epoch": 4, "memory": 44138, "step": 11911} +{"lr": 8.086891958166228e-05, "data_time": 0.001564168930053711, "loss": 0.3375686138868332, "time": 0.6364040374755859, "epoch": 4, "memory": 44138, "step": 12011} +{"lr": 8.057427858966412e-05, "data_time": 0.0014896154403686523, "loss": 0.3617121517658234, "time": 0.6370182991027832, "epoch": 4, "memory": 44138, "step": 12111} +{"lr": 8.027800916425949e-05, "data_time": 0.0015753746032714845, "loss": 0.31206669807434084, "time": 0.6372193336486817, "epoch": 4, "memory": 44138, "step": 12211} +{"lr": 7.998013017030234e-05, "data_time": 0.0018462657928466797, "loss": 0.34003559350967405, "time": 0.6371035814285279, "epoch": 4, "memory": 44138, "step": 12311} +{"lr": 7.968066057513528e-05, "data_time": 0.0015268802642822265, "loss": 0.337826144695282, "time": 0.6365351915359497, "epoch": 4, "memory": 44138, "step": 12411} +{"lr": 7.937961944738199e-05, "data_time": 0.0015517950057983398, "loss": 0.33829768002033234, "time": 0.6368270874023437, "epoch": 4, "memory": 44138, "step": 12511} +{"lr": 7.907702595573289e-05, "data_time": 0.0022010326385498045, "loss": 0.3339881062507629, "time": 0.6374033689498901, "epoch": 4, "memory": 44138, "step": 12611} +{"lr": 7.877289936772454e-05, "data_time": 0.001515817642211914, "loss": 0.3363713204860687, "time": 0.6368407011032104, "epoch": 4, "memory": 44138, "step": 12711} +{"lr": 7.8467259048513e-05, "data_time": 0.001432323455810547, "loss": 0.31904266476631166, "time": 0.6369088649749756, "epoch": 4, "memory": 44138, "step": 12811} +{"lr": 7.816012445964079e-05, "data_time": 0.0018038272857666016, "loss": 0.33538403213024137, "time": 0.6368149995803833, "epoch": 4, "memory": 44138, "step": 12911} +{"lr": 7.785151515779718e-05, "data_time": 0.0014866828918457032, "loss": 0.3120292916893959, "time": 0.6378989696502686, "epoch": 4, "memory": 44138, "step": 13011} +{"lr": 7.754145079357376e-05, "data_time": 0.0015709638595581056, "loss": 0.35246534645557404, "time": 0.6378729581832886, "epoch": 4, "memory": 44138, "step": 13111} +{"lr": 7.722995111021218e-05, "data_time": 0.0014418363571166992, "loss": 0.33217440247535707, "time": 0.6377979040145874, "epoch": 4, "memory": 44138, "step": 13211} +{"lr": 7.691703594234788e-05, "data_time": 0.0016176462173461913, "loss": 0.35015018582344054, "time": 0.6375664710998535, "epoch": 4, "memory": 44138, "step": 13311} +{"lr": 7.660272521474673e-05, "data_time": 0.0016233444213867188, "loss": 0.3585609346628189, "time": 0.6372680902481079, "epoch": 4, "memory": 44138, "step": 13411} +{"lr": 7.628703894103628e-05, "data_time": 0.001573491096496582, "loss": 0.328920578956604, "time": 0.6369132041931153, "epoch": 4, "memory": 44138, "step": 13511} +{"lr": 7.596999722243168e-05, "data_time": 0.001769709587097168, "loss": 0.3384198397397995, "time": 0.6378817558288574, "epoch": 4, "memory": 44138, "step": 13611} +{"lr": 7.565162024645551e-05, "data_time": 0.0014981985092163085, "loss": 0.3200004667043686, "time": 0.6374590158462524, "epoch": 4, "memory": 44138, "step": 13711} +{"lr": 7.533192828565237e-05, "data_time": 0.001697683334350586, "loss": 0.33200671672821047, "time": 0.6369019508361816, "epoch": 4, "memory": 44138, "step": 13811} +{"lr": 7.501094169629802e-05, "data_time": 0.0020283222198486327, "loss": 0.34375424683094025, "time": 0.6373275279998779, "epoch": 4, "memory": 44138, "step": 13911} +{"lr": 7.468868091710333e-05, "data_time": 0.0015586376190185546, "loss": 0.30156241059303285, "time": 0.6369643688201905, "epoch": 4, "memory": 44138, "step": 14011} +{"lr": 7.436516646791276e-05, "data_time": 0.0023156166076660155, "loss": 0.3411762535572052, "time": 0.6376494884490966, "epoch": 4, "memory": 44138, "step": 14111} +{"lr": 7.404041894839768e-05, "data_time": 0.0016160249710083009, "loss": 0.32748172283172605, "time": 0.636782431602478, "epoch": 4, "memory": 44138, "step": 14211} +{"lr": 7.371445903674494e-05, "data_time": 0.0014303207397460937, "loss": 0.3261335790157318, "time": 0.6369709491729736, "epoch": 4, "memory": 44138, "step": 14311} +{"lr": 7.338730748834014e-05, "data_time": 0.0014453172683715821, "loss": 0.34288241565227506, "time": 0.6382145643234253, "epoch": 4, "memory": 44138, "step": 14411} +{"lr": 7.305898513444536e-05, "data_time": 0.0017593860626220702, "loss": 0.33481956720352174, "time": 0.6380037546157837, "epoch": 4, "memory": 44138, "step": 14511} +{"lr": 7.27295128808741e-05, "data_time": 0.001836252212524414, "loss": 0.3406389534473419, "time": 0.6371669054031373, "epoch": 4, "memory": 44138, "step": 14611} +{"lr": 7.239891170665884e-05, "data_time": 0.0016165971755981445, "loss": 0.34216139018535613, "time": 0.637013602256775, "epoch": 4, "memory": 44138, "step": 14711} +{"lr": 7.206720266271582e-05, "data_time": 0.001488041877746582, "loss": 0.31315720826387405, "time": 0.6370663404464721, "epoch": 4, "memory": 44138, "step": 14811} +{"lr": 7.173440687050449e-05, "data_time": 0.00149078369140625, "loss": 0.3458180397748947, "time": 0.6370178699493408, "epoch": 4, "memory": 44138, "step": 14911} +{"lr": 7.140054552068265e-05, "data_time": 0.0015344858169555665, "loss": 0.3669417053461075, "time": 0.636637544631958, "epoch": 4, "memory": 44138, "step": 15011} +{"lr": 7.106563987175705e-05, "data_time": 0.0015919208526611328, "loss": 0.348757329583168, "time": 0.6377474308013916, "epoch": 4, "memory": 44138, "step": 15111} +{"lr": 7.072971124873013e-05, "data_time": 0.001621556282043457, "loss": 0.3177102476358414, "time": 0.6377481460571289, "epoch": 4, "memory": 44138, "step": 15211} +{"lr": 7.039278104174143e-05, "data_time": 0.0015107393264770508, "loss": 0.34663744270801544, "time": 0.6364511013031006, "epoch": 4, "memory": 44138, "step": 15311} +{"lr": 7.005487070470615e-05, "data_time": 0.0014693021774291992, "loss": 0.33676148056983946, "time": 0.6370150566101074, "epoch": 4, "memory": 44138, "step": 15411} +{"lr": 6.971600175394891e-05, "data_time": 0.0015363693237304688, "loss": 0.31902232468128205, "time": 0.6362375736236572, "epoch": 4, "memory": 44138, "step": 15511} +{"lr": 6.937619576683373e-05, "data_time": 0.001532578468322754, "loss": 0.3386623293161392, "time": 0.6367444276809693, "epoch": 4, "memory": 44138, "step": 15611} +{"lr": 6.903547438039021e-05, "data_time": 0.0018222808837890625, "loss": 0.3339738607406616, "time": 0.6374045372009277, "epoch": 4, "memory": 44138, "step": 15711} +{"accuracy/top1": 86.2565689086914, "data_time": 0.015925839029509444, "time": 1.2589548990644257, "step": 4} +{"lr": 6.856723955077045e-05, "data_time": 0.001645207405090332, "loss": 0.326394647359848, "time": 0.6376607894897461, "epoch": 5, "memory": 44138, "step": 15848} +{"lr": 6.822443541062844e-05, "data_time": 0.001539325714111328, "loss": 0.33485254645347595, "time": 0.6374549627304077, "epoch": 5, "memory": 44138, "step": 15948} +{"lr": 6.788078920910325e-05, "data_time": 0.001550769805908203, "loss": 0.3428355485200882, "time": 0.6369801759719849, "epoch": 5, "memory": 44138, "step": 16048} +{"lr": 6.753632282774874e-05, "data_time": 0.0025886774063110353, "loss": 0.33377662897109983, "time": 0.6380975246429443, "epoch": 5, "memory": 44138, "step": 16148} +{"lr": 6.719105820034336e-05, "data_time": 0.0020990848541259767, "loss": 0.3222132921218872, "time": 0.6389024019241333, "epoch": 5, "memory": 44138, "step": 16248} +{"lr": 6.68450173114937e-05, "data_time": 0.0021158456802368164, "loss": 0.3249441057443619, "time": 0.6427899599075317, "epoch": 5, "memory": 44138, "step": 16348} +{"lr": 6.64982221952347e-05, "data_time": 0.0017545461654663087, "loss": 0.3563050091266632, "time": 0.6383615732192993, "epoch": 5, "memory": 44138, "step": 16448} +{"lr": 6.615069493362599e-05, "data_time": 0.0026414871215820314, "loss": 0.3414961099624634, "time": 0.6386398553848267, "epoch": 5, "memory": 44138, "step": 16548} +{"lr": 6.580245765534682e-05, "data_time": 0.0019840717315673826, "loss": 0.34103713035583494, "time": 0.6361668825149536, "epoch": 5, "memory": 44138, "step": 16648} +{"lr": 6.54535325342862e-05, "data_time": 0.001967191696166992, "loss": 0.3617107540369034, "time": 0.6388908624649048, "epoch": 5, "memory": 44138, "step": 16748} +{"lr": 6.510394178813155e-05, "data_time": 0.002042698860168457, "loss": 0.36190399825572966, "time": 0.637880277633667, "epoch": 5, "memory": 44138, "step": 16848} +{"lr": 6.475370767695331e-05, "data_time": 0.0019478559494018554, "loss": 0.3186565160751343, "time": 0.6412222623825073, "epoch": 5, "memory": 44138, "step": 16948} +{"lr": 6.440285250178845e-05, "data_time": 0.0019121170043945312, "loss": 0.3272313341498375, "time": 0.6368343114852906, "epoch": 5, "memory": 44138, "step": 17048} +{"lr": 6.405139860321955e-05, "data_time": 0.001958751678466797, "loss": 0.3496026277542114, "time": 0.6370957851409912, "epoch": 5, "memory": 44138, "step": 17148} +{"lr": 6.369936835995306e-05, "data_time": 0.0023630142211914064, "loss": 0.317501437664032, "time": 0.6372188806533814, "epoch": 5, "memory": 44138, "step": 17248} +{"lr": 6.334678418739388e-05, "data_time": 0.0023343563079833984, "loss": 0.3219574272632599, "time": 0.6370770692825317, "epoch": 5, "memory": 44138, "step": 17348} +{"lr": 6.299366853621796e-05, "data_time": 0.0021996259689331054, "loss": 0.3162900283932686, "time": 0.637747049331665, "epoch": 5, "memory": 44138, "step": 17448} +{"lr": 6.264004389094315e-05, "data_time": 0.0016735076904296875, "loss": 0.3424944311380386, "time": 0.6381992340087891, "epoch": 5, "memory": 44138, "step": 17548} +{"lr": 6.22859327684974e-05, "data_time": 0.0017868280410766602, "loss": 0.3068771749734879, "time": 0.6393717050552368, "epoch": 5, "memory": 44138, "step": 17648} +{"lr": 6.193135771678478e-05, "data_time": 0.0027231454849243166, "loss": 0.370467546582222, "time": 0.6384228944778443, "epoch": 5, "memory": 44138, "step": 17748} +{"lr": 6.157634131324983e-05, "data_time": 0.0020483255386352537, "loss": 0.3498892217874527, "time": 0.6363596677780151, "epoch": 5, "memory": 44138, "step": 17848} +{"lr": 6.12209061634402e-05, "data_time": 0.002358865737915039, "loss": 0.3239938378334045, "time": 0.6381728887557984, "epoch": 5, "memory": 44138, "step": 17948} +{"lr": 6.086507489956686e-05, "data_time": 0.0018143177032470704, "loss": 0.30747300684452056, "time": 0.6373126983642579, "epoch": 5, "memory": 44138, "step": 18048} +{"lr": 6.0508870179063345e-05, "data_time": 0.001836705207824707, "loss": 0.34509778916835787, "time": 0.6379811525344848, "epoch": 5, "memory": 44138, "step": 18148} +{"lr": 6.015231468314303e-05, "data_time": 0.0017647981643676759, "loss": 0.33873641192913057, "time": 0.6389262914657593, "epoch": 5, "memory": 44138, "step": 18248} +{"lr": 5.979543111535473e-05, "data_time": 0.0015564680099487305, "loss": 0.3281713366508484, "time": 0.6364551782608032, "epoch": 5, "memory": 44138, "step": 18348} +{"lr": 5.943824220013694e-05, "data_time": 0.0016757965087890625, "loss": 0.3410695821046829, "time": 0.637682032585144, "epoch": 5, "memory": 44138, "step": 18448} +{"lr": 5.908077068137128e-05, "data_time": 0.002691364288330078, "loss": 0.3036420464515686, "time": 0.6384639739990234, "epoch": 5, "memory": 44138, "step": 18548} +{"lr": 5.872303932093392e-05, "data_time": 0.0015374422073364258, "loss": 0.33327831923961637, "time": 0.6390835762023925, "epoch": 5, "memory": 44138, "step": 18648} +{"lr": 5.836507089724633e-05, "data_time": 0.0018391847610473634, "loss": 0.31866020858287813, "time": 0.6406067371368408, "epoch": 5, "memory": 44138, "step": 18748} +{"lr": 5.800688820382498e-05, "data_time": 0.0015901565551757813, "loss": 0.29573112577199934, "time": 0.6411308288574219, "epoch": 5, "memory": 44138, "step": 18848} +{"lr": 5.7648514047829945e-05, "data_time": 0.0015150547027587891, "loss": 0.3226841062307358, "time": 0.6422483205795289, "epoch": 5, "memory": 44138, "step": 18948} +{"lr": 5.7289971248612474e-05, "data_time": 0.0016436338424682616, "loss": 0.3375723406672478, "time": 0.6410388231277466, "epoch": 5, "memory": 44138, "step": 19048} +{"lr": 5.693128263626215e-05, "data_time": 0.001591634750366211, "loss": 0.32685727477073667, "time": 0.6419731378555298, "epoch": 5, "memory": 44138, "step": 19148} +{"lr": 5.6572471050153236e-05, "data_time": 0.0018211603164672852, "loss": 0.32787306010723116, "time": 0.6468555450439453, "epoch": 5, "memory": 44138, "step": 19248} +{"lr": 5.621355933749018e-05, "data_time": 0.0018019676208496094, "loss": 0.32421029806137086, "time": 0.639817190170288, "epoch": 5, "memory": 44138, "step": 19348} +{"lr": 5.5854570351853106e-05, "data_time": 0.0018308401107788087, "loss": 0.3336017966270447, "time": 0.639714241027832, "epoch": 5, "memory": 44138, "step": 19448} +{"lr": 5.5495526951742315e-05, "data_time": 0.0016659021377563477, "loss": 0.36472677886486055, "time": 0.6413752317428589, "epoch": 5, "memory": 44138, "step": 19548} +{"lr": 5.513645199912304e-05, "data_time": 0.0015044450759887696, "loss": 0.35022721588611605, "time": 0.6419339895248413, "epoch": 5, "memory": 44138, "step": 19648} +{"accuracy/top1": 90.24390411376953, "data_time": 0.0146208878221183, "time": 1.258496120058257, "step": 5} +{"lr": 5.464450978670876e-05, "data_time": 0.0016424179077148438, "loss": 0.36218151450157166, "time": 0.6386335134506226, "epoch": 6, "memory": 44138, "step": 19785} +{"lr": 5.4285451361545876e-05, "data_time": 0.001752781867980957, "loss": 0.29286844879388807, "time": 0.6384132862091064, "epoch": 6, "memory": 44138, "step": 19885} +{"lr": 5.3926438435022166e-05, "data_time": 0.0016929149627685548, "loss": 0.31446826457977295, "time": 0.6405260562896729, "epoch": 6, "memory": 44138, "step": 19985} +{"lr": 5.356749386716241e-05, "data_time": 0.0029013872146606444, "loss": 0.345249080657959, "time": 0.6400500535964966, "epoch": 6, "memory": 44138, "step": 20085} +{"lr": 5.320864051363873e-05, "data_time": 0.0015892744064331054, "loss": 0.3425854355096817, "time": 0.6493218898773193, "epoch": 6, "memory": 44138, "step": 20185} +{"lr": 5.2849901224315175e-05, "data_time": 0.0014860868453979493, "loss": 0.30708692967891693, "time": 0.6392604351043701, "epoch": 6, "memory": 44138, "step": 20285} +{"lr": 5.249129884179276e-05, "data_time": 0.001697397232055664, "loss": 0.35072503685951234, "time": 0.6379307031631469, "epoch": 6, "memory": 44138, "step": 20385} +{"lr": 5.2132856199954985e-05, "data_time": 0.0024124622344970704, "loss": 0.31700047850608826, "time": 0.6375947713851928, "epoch": 6, "memory": 44138, "step": 20485} +{"lr": 5.177459612251411e-05, "data_time": 0.001714944839477539, "loss": 0.36079367697238923, "time": 0.6386557340621948, "epoch": 6, "memory": 44138, "step": 20585} +{"lr": 5.141654142155744e-05, "data_time": 0.0015901803970336914, "loss": 0.34223597347736356, "time": 0.6371722221374512, "epoch": 6, "memory": 44138, "step": 20685} +{"lr": 5.105871489609512e-05, "data_time": 0.0016604423522949218, "loss": 0.34108743667602537, "time": 0.640283179283142, "epoch": 6, "memory": 44138, "step": 20785} +{"lr": 5.070113933060828e-05, "data_time": 0.0016683101654052734, "loss": 0.3372240215539932, "time": 0.6392496824264526, "epoch": 6, "memory": 44138, "step": 20885} +{"lr": 5.0343837493598264e-05, "data_time": 0.0018593549728393554, "loss": 0.34016806781291964, "time": 0.6387745380401612, "epoch": 6, "memory": 44138, "step": 20985} +{"lr": 4.998683213613675e-05, "data_time": 0.0020160913467407227, "loss": 0.3343943566083908, "time": 0.6394699335098266, "epoch": 6, "memory": 44138, "step": 21085} +{"lr": 4.963014599041739e-05, "data_time": 0.002039480209350586, "loss": 0.3596372425556183, "time": 0.6404732704162598, "epoch": 6, "memory": 44138, "step": 21185} +{"lr": 4.927380176830793e-05, "data_time": 0.0017155647277832032, "loss": 0.3008223414421082, "time": 0.6406383275985718, "epoch": 6, "memory": 44138, "step": 21285} +{"lr": 4.89178221599044e-05, "data_time": 0.0018315315246582031, "loss": 0.31562944650650027, "time": 0.6387464284896851, "epoch": 6, "memory": 44138, "step": 21385} +{"lr": 4.85622298320861e-05, "data_time": 0.0016685247421264649, "loss": 0.31584973335266114, "time": 0.6380267143249512, "epoch": 6, "memory": 44138, "step": 21485} +{"lr": 4.8207047427072355e-05, "data_time": 0.001862955093383789, "loss": 0.3263800323009491, "time": 0.6384902477264405, "epoch": 6, "memory": 44138, "step": 21585} +{"lr": 4.7852297560980876e-05, "data_time": 0.0018761634826660156, "loss": 0.32396080791950227, "time": 0.6380950689315796, "epoch": 6, "memory": 44138, "step": 21685} +{"lr": 4.749800282238749e-05, "data_time": 0.0015840768814086915, "loss": 0.3161410465836525, "time": 0.6385318994522095, "epoch": 6, "memory": 44138, "step": 21785} +{"lr": 4.714418577088805e-05, "data_time": 0.0017816543579101563, "loss": 0.334672412276268, "time": 0.6386142492294311, "epoch": 6, "memory": 44138, "step": 21885} +{"lr": 4.679086893566184e-05, "data_time": 0.001709890365600586, "loss": 0.3308043345808983, "time": 0.6387040853500366, "epoch": 6, "memory": 44138, "step": 21985} +{"lr": 4.643807481403694e-05, "data_time": 0.001747465133666992, "loss": 0.33866152465343474, "time": 0.6387199401855469, "epoch": 6, "memory": 44138, "step": 22085} +{"lr": 4.6085825870057946e-05, "data_time": 0.0017476558685302734, "loss": 0.33113373517990113, "time": 0.6380645990371704, "epoch": 6, "memory": 44138, "step": 22185} +{"lr": 4.573414453305537e-05, "data_time": 0.0015560388565063477, "loss": 0.34676816761493684, "time": 0.6406392335891724, "epoch": 6, "memory": 44138, "step": 22285} +{"lr": 4.5383053196217514e-05, "data_time": 0.0016381025314331054, "loss": 0.36065601706504824, "time": 0.6400545835494995, "epoch": 6, "memory": 44138, "step": 22385} +{"lr": 4.503257421516477e-05, "data_time": 0.0017625093460083008, "loss": 0.32619393765926363, "time": 0.637947678565979, "epoch": 6, "memory": 44138, "step": 22485} +{"lr": 4.468272990652586e-05, "data_time": 0.002097606658935547, "loss": 0.3510457679629326, "time": 0.6441561937332153, "epoch": 6, "memory": 44138, "step": 22585} +{"lr": 4.433354254651697e-05, "data_time": 0.001828622817993164, "loss": 0.32855255007743833, "time": 0.6386828899383545, "epoch": 6, "memory": 44138, "step": 22685} +{"lr": 4.398503436952338e-05, "data_time": 0.001613950729370117, "loss": 0.3488876402378082, "time": 0.6386906147003174, "epoch": 6, "memory": 44138, "step": 22785} +{"lr": 4.363722756668351e-05, "data_time": 0.002217888832092285, "loss": 0.3292103260755539, "time": 0.6382639646530152, "epoch": 6, "memory": 44138, "step": 22885} +{"lr": 4.329014428447611e-05, "data_time": 0.0016797780990600586, "loss": 0.33284675478935244, "time": 0.6418892860412597, "epoch": 6, "memory": 44138, "step": 22985} +{"lr": 4.294380662331e-05, "data_time": 0.0015721797943115234, "loss": 0.32860801815986634, "time": 0.6390622615814209, "epoch": 6, "memory": 44138, "step": 23085} +{"lr": 4.259823663611678e-05, "data_time": 0.0015774011611938477, "loss": 0.32207544445991515, "time": 0.63813316822052, "epoch": 6, "memory": 44138, "step": 23185} +{"lr": 4.22534563269468e-05, "data_time": 0.0030534744262695314, "loss": 0.34216980934143065, "time": 0.6394953727722168, "epoch": 6, "memory": 44138, "step": 23285} +{"lr": 4.1909487649567694e-05, "data_time": 0.0015613794326782226, "loss": 0.350005766749382, "time": 0.6372884273529053, "epoch": 6, "memory": 44138, "step": 23385} +{"lr": 4.156635250606698e-05, "data_time": 0.0014662027359008789, "loss": 0.33806262016296384, "time": 0.6379681348800659, "epoch": 6, "memory": 44138, "step": 23485} +{"lr": 4.122407274545703e-05, "data_time": 0.0015445470809936524, "loss": 0.3400091290473938, "time": 0.6394245147705078, "epoch": 6, "memory": 44138, "step": 23585} +{"accuracy/top1": 92.45677185058594, "data_time": 0.016118320925482387, "time": 1.2589721515260894, "step": 6} +{"lr": 4.075657788156285e-05, "data_time": 0.0016452789306640625, "loss": 0.33512323498725893, "time": 0.6380365371704102, "epoch": 7, "memory": 44138, "step": 23722} +{"lr": 4.0416412311741566e-05, "data_time": 0.0018833637237548827, "loss": 0.3269672095775604, "time": 0.6393464803695679, "epoch": 7, "memory": 44138, "step": 23822} +{"lr": 4.007717534685821e-05, "data_time": 0.0015670061111450195, "loss": 0.33208352476358416, "time": 0.6402326583862304, "epoch": 7, "memory": 44138, "step": 23922} +{"lr": 3.973888858770982e-05, "data_time": 0.0017290353775024415, "loss": 0.33266387283802035, "time": 0.638140082359314, "epoch": 7, "memory": 44138, "step": 24022} +{"lr": 3.9401573574589535e-05, "data_time": 0.0015340805053710937, "loss": 0.33821841776371003, "time": 0.6378901720046997, "epoch": 7, "memory": 44138, "step": 24122} +{"lr": 3.906525178591511e-05, "data_time": 0.001939082145690918, "loss": 0.3124773234128952, "time": 0.6394755363464355, "epoch": 7, "memory": 44138, "step": 24222} +{"lr": 3.87299446368609e-05, "data_time": 0.0015781164169311524, "loss": 0.3509422570466995, "time": 0.6398543357849121, "epoch": 7, "memory": 44138, "step": 24322} +{"lr": 3.839567347799442e-05, "data_time": 0.0016474008560180664, "loss": 0.33867704272270205, "time": 0.6391457557678223, "epoch": 7, "memory": 44138, "step": 24422} +{"lr": 3.806245959391694e-05, "data_time": 0.0016038894653320312, "loss": 0.3241992652416229, "time": 0.6411542177200318, "epoch": 7, "memory": 44138, "step": 24522} +{"lr": 3.773032420190816e-05, "data_time": 0.0017015695571899413, "loss": 0.32114947885274886, "time": 0.6378494739532471, "epoch": 7, "memory": 44138, "step": 24622} +{"lr": 3.739928845057507e-05, "data_time": 0.0015629291534423827, "loss": 0.3297632187604904, "time": 0.6384621143341065, "epoch": 7, "memory": 44138, "step": 24722} +{"lr": 3.706937341850541e-05, "data_time": 0.0016818523406982421, "loss": 0.34492531418800354, "time": 0.6386454343795777, "epoch": 7, "memory": 44138, "step": 24822} +{"lr": 3.674060011292556e-05, "data_time": 0.0015878677368164062, "loss": 0.3129858136177063, "time": 0.6385512113571167, "epoch": 7, "memory": 44138, "step": 24922} +{"lr": 3.641298946836287e-05, "data_time": 0.0021367788314819334, "loss": 0.3471270501613617, "time": 0.6414084672927857, "epoch": 7, "memory": 44138, "step": 25022} +{"lr": 3.6086562345312595e-05, "data_time": 0.0020767688751220704, "loss": 0.3027243345975876, "time": 0.6397251605987548, "epoch": 7, "memory": 44138, "step": 25122} +{"lr": 3.576133952890974e-05, "data_time": 0.0014804840087890626, "loss": 0.3238443449139595, "time": 0.6403166055679321, "epoch": 7, "memory": 44138, "step": 25222} +{"lr": 3.543734172760541e-05, "data_time": 0.0018121719360351563, "loss": 0.3503607839345932, "time": 0.63937406539917, "epoch": 7, "memory": 44138, "step": 25322} +{"lr": 3.511458957184831e-05, "data_time": 0.001491546630859375, "loss": 0.34674532413482667, "time": 0.6392184734344483, "epoch": 7, "memory": 44138, "step": 25422} +{"lr": 3.479310361277118e-05, "data_time": 0.0017754793167114257, "loss": 0.31534984707832336, "time": 0.637584137916565, "epoch": 7, "memory": 44138, "step": 25522} +{"lr": 3.4472904320882065e-05, "data_time": 0.001627635955810547, "loss": 0.33254554271698, "time": 0.6394550800323486, "epoch": 7, "memory": 44138, "step": 25622} +{"lr": 3.4154012084760864e-05, "data_time": 0.0020392894744873046, "loss": 0.3499720633029938, "time": 0.6396182537078857, "epoch": 7, "memory": 44138, "step": 25722} +{"lr": 3.38364472097612e-05, "data_time": 0.001755666732788086, "loss": 0.3322810471057892, "time": 0.6414047002792358, "epoch": 7, "memory": 44138, "step": 25822} +{"lr": 3.3520229916717485e-05, "data_time": 0.0015645503997802734, "loss": 0.34651918709278107, "time": 0.6374246835708618, "epoch": 7, "memory": 44138, "step": 25922} +{"lr": 3.320538034065722e-05, "data_time": 0.0015337228775024413, "loss": 0.3382045477628708, "time": 0.6386137962341308, "epoch": 7, "memory": 44138, "step": 26022} +{"lr": 3.289191852951901e-05, "data_time": 0.0018460512161254882, "loss": 0.3323641985654831, "time": 0.6404607057571411, "epoch": 7, "memory": 44138, "step": 26122} +{"lr": 3.257986444287605e-05, "data_time": 0.0016823530197143555, "loss": 0.3483747899532318, "time": 0.640232253074646, "epoch": 7, "memory": 44138, "step": 26222} +{"lr": 3.226923795066519e-05, "data_time": 0.0013779163360595702, "loss": 0.3480044960975647, "time": 0.6392740726470947, "epoch": 7, "memory": 44138, "step": 26322} +{"lr": 3.196005883192158e-05, "data_time": 0.0019949197769165037, "loss": 0.32673614621162417, "time": 0.6388867855072021, "epoch": 7, "memory": 44138, "step": 26422} +{"lr": 3.165234677351947e-05, "data_time": 0.0019626379013061523, "loss": 0.3208146706223488, "time": 0.6415780782699585, "epoch": 7, "memory": 44138, "step": 26522} +{"lr": 3.134612136891845e-05, "data_time": 0.0021756172180175783, "loss": 0.3511429995298386, "time": 0.6430078506469726, "epoch": 7, "memory": 44138, "step": 26622} +{"lr": 3.104140211691602e-05, "data_time": 0.001885056495666504, "loss": 0.3994672566652298, "time": 0.6417765617370605, "epoch": 7, "memory": 44138, "step": 26722} +{"lr": 3.073820842040592e-05, "data_time": 0.0016906261444091797, "loss": 0.36018048524856566, "time": 0.6405328273773193, "epoch": 7, "memory": 44138, "step": 26822} +{"lr": 3.043655958514262e-05, "data_time": 0.0015549182891845704, "loss": 0.3429575651884079, "time": 0.6394672155380249, "epoch": 7, "memory": 44138, "step": 26922} +{"lr": 3.013647481851217e-05, "data_time": 0.0016427755355834961, "loss": 0.34207204580307005, "time": 0.641802453994751, "epoch": 7, "memory": 44138, "step": 27022} +{"lr": 2.983797322830899e-05, "data_time": 0.001452159881591797, "loss": 0.3353388488292694, "time": 0.6427386045455933, "epoch": 7, "memory": 44138, "step": 27122} +{"lr": 2.9541073821519346e-05, "data_time": 0.0015214204788208008, "loss": 0.33911772817373276, "time": 0.6428911209106445, "epoch": 7, "memory": 44138, "step": 27222} +{"lr": 2.924579550311095e-05, "data_time": 0.0016516923904418945, "loss": 0.3578227132558823, "time": 0.6407561779022217, "epoch": 7, "memory": 44138, "step": 27322} +{"lr": 2.8952157074829418e-05, "data_time": 0.001589202880859375, "loss": 0.3136764168739319, "time": 0.6399877071380615, "epoch": 7, "memory": 44138, "step": 27422} +{"lr": 2.866017723400071e-05, "data_time": 0.0016010522842407227, "loss": 0.31430790275335313, "time": 0.6452139377593994, "epoch": 7, "memory": 44138, "step": 27522} +{"accuracy/top1": 93.62931060791016, "data_time": 0.01609754151311414, "time": 1.2588747410938657, "step": 7} +{"lr": 2.8262891373132927e-05, "data_time": 0.0017935514450073242, "loss": 0.32785196900367736, "time": 0.641397500038147, "epoch": 8, "memory": 44138, "step": 27659} +{"lr": 2.797491643470029e-05, "data_time": 0.0023957490921020508, "loss": 0.3483874022960663, "time": 0.6421234130859375, "epoch": 8, "memory": 44138, "step": 27759} +{"lr": 2.768866230918266e-05, "data_time": 0.001797795295715332, "loss": 0.33559287488460543, "time": 0.649546217918396, "epoch": 8, "memory": 44138, "step": 27859} +{"lr": 2.7404147223713185e-05, "data_time": 0.001596689224243164, "loss": 0.336510494351387, "time": 0.6395530939102173, "epoch": 8, "memory": 44138, "step": 27959} +{"lr": 2.712138929469222e-05, "data_time": 0.0016678571701049805, "loss": 0.35269678831100465, "time": 0.6401216506958007, "epoch": 8, "memory": 44138, "step": 28059} +{"lr": 2.6840406526633896e-05, "data_time": 0.0015362739562988282, "loss": 0.3273759573698044, "time": 0.6406999349594116, "epoch": 8, "memory": 44138, "step": 28159} +{"lr": 2.6561216811019345e-05, "data_time": 0.00162961483001709, "loss": 0.3498887956142426, "time": 0.6378124237060547, "epoch": 8, "memory": 44138, "step": 28259} +{"lr": 2.6283837925157908e-05, "data_time": 0.001520538330078125, "loss": 0.33529615998268125, "time": 0.6380717277526855, "epoch": 8, "memory": 44138, "step": 28359} +{"lr": 2.6008287531054805e-05, "data_time": 0.0029815912246704103, "loss": 0.3309875041246414, "time": 0.6378012418746948, "epoch": 8, "memory": 44138, "step": 28459} +{"lr": 2.5734583174286796e-05, "data_time": 0.002580428123474121, "loss": 0.34405426383018495, "time": 0.6385094881057739, "epoch": 8, "memory": 44138, "step": 28559} +{"lr": 2.546274228288488e-05, "data_time": 0.0016176462173461913, "loss": 0.33039659112691877, "time": 0.636967945098877, "epoch": 8, "memory": 44138, "step": 28659} +{"lr": 2.5192782166224308e-05, "data_time": 0.0016625642776489258, "loss": 0.3501308798789978, "time": 0.6367563486099244, "epoch": 8, "memory": 44138, "step": 28759} +{"lr": 2.492472001392276e-05, "data_time": 0.0017060279846191407, "loss": 0.3493047595024109, "time": 0.637101149559021, "epoch": 8, "memory": 44138, "step": 28859} +{"lr": 2.4658572894745857e-05, "data_time": 0.0016996622085571288, "loss": 0.34435219764709474, "time": 0.6361613512039185, "epoch": 8, "memory": 44138, "step": 28959} +{"lr": 2.439435775551997e-05, "data_time": 0.0018393993377685547, "loss": 0.3414866477251053, "time": 0.636789059638977, "epoch": 8, "memory": 44138, "step": 29059} +{"lr": 2.4132091420053357e-05, "data_time": 0.002955722808837891, "loss": 0.33168233186006546, "time": 0.6387169122695923, "epoch": 8, "memory": 44138, "step": 29159} +{"lr": 2.387179058806494e-05, "data_time": 0.0017844438552856445, "loss": 0.31749085783958436, "time": 0.6382629156112671, "epoch": 8, "memory": 44138, "step": 29259} +{"lr": 2.361347183412073e-05, "data_time": 0.001632523536682129, "loss": 0.33510020971298216, "time": 0.6372994899749755, "epoch": 8, "memory": 44138, "step": 29359} +{"lr": 2.3357151606578728e-05, "data_time": 0.0015772104263305664, "loss": 0.333350333571434, "time": 0.6379689455032349, "epoch": 8, "memory": 44138, "step": 29459} +{"lr": 2.3102846226541355e-05, "data_time": 0.002072000503540039, "loss": 0.3356959939002991, "time": 0.6372670888900757, "epoch": 8, "memory": 44138, "step": 29559} +{"lr": 2.2850571886816365e-05, "data_time": 0.00260622501373291, "loss": 0.3276965469121933, "time": 0.6381346940994262, "epoch": 8, "memory": 44138, "step": 29659} +{"lr": 2.260034465088568e-05, "data_time": 0.0017112016677856446, "loss": 0.3511845350265503, "time": 0.6372352123260498, "epoch": 8, "memory": 44138, "step": 29759} +{"lr": 2.2352180451882643e-05, "data_time": 0.0021218776702880858, "loss": 0.3321383535861969, "time": 0.6360498666763306, "epoch": 8, "memory": 44138, "step": 29859} +{"lr": 2.210609509157741e-05, "data_time": 0.0017811059951782227, "loss": 0.34005952179431914, "time": 0.6362974882125855, "epoch": 8, "memory": 44138, "step": 29959} +{"lr": 2.1862104239370772e-05, "data_time": 0.0017603158950805664, "loss": 0.33925525546073915, "time": 0.6381688117980957, "epoch": 8, "memory": 44138, "step": 30059} +{"lr": 2.1620223431296407e-05, "data_time": 0.001773238182067871, "loss": 0.34935729801654813, "time": 0.6371684312820435, "epoch": 8, "memory": 44138, "step": 30159} +{"lr": 2.1380468069031744e-05, "data_time": 0.0017689943313598632, "loss": 0.3472258061170578, "time": 0.6456415891647339, "epoch": 8, "memory": 44138, "step": 30259} +{"lr": 2.1142853418917185e-05, "data_time": 0.0017083168029785156, "loss": 0.3159309595823288, "time": 0.6369046211242676, "epoch": 8, "memory": 44138, "step": 30359} +{"lr": 2.0907394610983892e-05, "data_time": 0.0015600919723510742, "loss": 0.3736870944499969, "time": 0.6376642704010009, "epoch": 8, "memory": 44138, "step": 30459} +{"lr": 2.0674106637990643e-05, "data_time": 0.0015815496444702148, "loss": 0.32337228059768675, "time": 0.6369141101837158, "epoch": 8, "memory": 44138, "step": 30559} +{"lr": 2.0443004354468975e-05, "data_time": 0.0016054630279541016, "loss": 0.33981326818466184, "time": 0.6369661092758179, "epoch": 8, "memory": 44138, "step": 30659} +{"lr": 2.0214102475777453e-05, "data_time": 0.002117443084716797, "loss": 0.3353619068861008, "time": 0.6368308067321777, "epoch": 8, "memory": 44138, "step": 30759} +{"lr": 1.9987415577164583e-05, "data_time": 0.002645564079284668, "loss": 0.34488191902637483, "time": 0.6388916492462158, "epoch": 8, "memory": 44138, "step": 30859} +{"lr": 1.9762958092840814e-05, "data_time": 0.0018764257431030274, "loss": 0.3224926322698593, "time": 0.6365635395050049, "epoch": 8, "memory": 44138, "step": 30959} +{"lr": 1.9540744315059334e-05, "data_time": 0.001518082618713379, "loss": 0.3325134128332138, "time": 0.6377947330474854, "epoch": 8, "memory": 44138, "step": 31059} +{"lr": 1.9320788393206132e-05, "data_time": 0.001732802391052246, "loss": 0.30041895508766175, "time": 0.6368714809417725, "epoch": 8, "memory": 44138, "step": 31159} +{"lr": 1.9103104332899027e-05, "data_time": 0.0014935731887817383, "loss": 0.3463741153478622, "time": 0.6371479272842407, "epoch": 8, "memory": 44138, "step": 31259} +{"lr": 1.888770599509573e-05, "data_time": 0.0019794464111328124, "loss": 0.34848603010177615, "time": 0.6449602365493774, "epoch": 8, "memory": 44138, "step": 31359} +{"lr": 1.8674607095211516e-05, "data_time": 0.002109527587890625, "loss": 0.3407303631305695, "time": 0.6373261451721192, "epoch": 8, "memory": 44138, "step": 31459} +{"accuracy/top1": 94.44000244140625, "data_time": 0.014360251097843564, "time": 1.2575075009773518, "step": 8} +{"lr": 1.838641934548488e-05, "data_time": 0.001876235008239746, "loss": 0.34708923697471616, "time": 0.6379916429519653, "epoch": 9, "memory": 44138, "step": 31596} +{"lr": 1.817882403185378e-05, "data_time": 0.0023573875427246095, "loss": 0.36315891742706297, "time": 0.6379354953765869, "epoch": 9, "memory": 44138, "step": 31696} +{"lr": 1.797357329395729e-05, "data_time": 0.0017489194869995117, "loss": 0.34651450216770174, "time": 0.6378476858139038, "epoch": 9, "memory": 44138, "step": 31796} +{"lr": 1.7770680201065646e-05, "data_time": 0.0018070459365844727, "loss": 0.3591558516025543, "time": 0.6380157470703125, "epoch": 9, "memory": 44138, "step": 31896} +{"lr": 1.7570157672326885e-05, "data_time": 0.0016509294509887695, "loss": 0.33588482439517975, "time": 0.6379043579101562, "epoch": 9, "memory": 44138, "step": 31996} +{"lr": 1.7372018475944145e-05, "data_time": 0.0018067598342895509, "loss": 0.3351457595825195, "time": 0.6384174823760986, "epoch": 9, "memory": 44138, "step": 32096} +{"lr": 1.717627522836271e-05, "data_time": 0.0019103050231933593, "loss": 0.35640172362327577, "time": 0.6380068540573121, "epoch": 9, "memory": 44138, "step": 32196} +{"lr": 1.6982940393466685e-05, "data_time": 0.0015610694885253907, "loss": 0.3528045743703842, "time": 0.6377854824066163, "epoch": 9, "memory": 44138, "step": 32296} +{"lr": 1.6792026281785218e-05, "data_time": 0.0020467042922973633, "loss": 0.3167236477136612, "time": 0.6371350765228272, "epoch": 9, "memory": 44138, "step": 32396} +{"lr": 1.660354504970888e-05, "data_time": 0.0024530649185180663, "loss": 0.3179640740156174, "time": 0.6366666078567504, "epoch": 9, "memory": 44138, "step": 32496} +{"lr": 1.641750869871534e-05, "data_time": 0.0016213178634643555, "loss": 0.3418331354856491, "time": 0.6368302345275879, "epoch": 9, "memory": 44138, "step": 32596} +{"lr": 1.6233929074605374e-05, "data_time": 0.0017043352127075195, "loss": 0.34667418599128724, "time": 0.636782455444336, "epoch": 9, "memory": 44138, "step": 32696} +{"lr": 1.6052817866748513e-05, "data_time": 0.0015819787979125977, "loss": 0.3445913285017014, "time": 0.6372241020202637, "epoch": 9, "memory": 44138, "step": 32796} +{"lr": 1.587418660733871e-05, "data_time": 0.0015439748764038085, "loss": 0.33370116651058196, "time": 0.636690616607666, "epoch": 9, "memory": 44138, "step": 32896} +{"lr": 1.5698046670660044e-05, "data_time": 0.0015928506851196288, "loss": 0.33526529371738434, "time": 0.6363592624664307, "epoch": 9, "memory": 44138, "step": 32996} +{"lr": 1.5524409272362524e-05, "data_time": 0.001676321029663086, "loss": 0.329180970788002, "time": 0.6364695787429809, "epoch": 9, "memory": 44138, "step": 33096} +{"lr": 1.5353285468747853e-05, "data_time": 0.0018772840499877929, "loss": 0.3220287948846817, "time": 0.6375333786010742, "epoch": 9, "memory": 44138, "step": 33196} +{"lr": 1.5184686156065442e-05, "data_time": 0.0019403457641601562, "loss": 0.36585857570171354, "time": 0.6365541219711304, "epoch": 9, "memory": 44138, "step": 33296} +{"lr": 1.5018622069818636e-05, "data_time": 0.0016704559326171874, "loss": 0.3399368613958359, "time": 0.6365461111068725, "epoch": 9, "memory": 44138, "step": 33396} +{"lr": 1.4855103784081099e-05, "data_time": 0.0017790555953979491, "loss": 0.3403072267770767, "time": 0.6376025199890136, "epoch": 9, "memory": 44138, "step": 33496} +{"lr": 1.4694141710823508e-05, "data_time": 0.001851367950439453, "loss": 0.3264953374862671, "time": 0.6371854066848754, "epoch": 9, "memory": 44138, "step": 33596} +{"lr": 1.4535746099250614e-05, "data_time": 0.0017159461975097656, "loss": 0.323648601770401, "time": 0.6367975950241089, "epoch": 9, "memory": 44138, "step": 33696} +{"lr": 1.4379927035148576e-05, "data_time": 0.001768183708190918, "loss": 0.3227705806493759, "time": 0.6373627185821533, "epoch": 9, "memory": 44138, "step": 33796} +{"lr": 1.42266944402428e-05, "data_time": 0.0017513513565063476, "loss": 0.3449145913124084, "time": 0.637163496017456, "epoch": 9, "memory": 44138, "step": 33896} +{"lr": 1.4076058071566133e-05, "data_time": 0.0014359474182128907, "loss": 0.3214793801307678, "time": 0.6374832630157471, "epoch": 9, "memory": 44138, "step": 33996} +{"lr": 1.3928027520837635e-05, "data_time": 0.001459956169128418, "loss": 0.3619404137134552, "time": 0.6371097326278686, "epoch": 9, "memory": 44138, "step": 34096} +{"lr": 1.3782612213851759e-05, "data_time": 0.001634979248046875, "loss": 0.3587342739105225, "time": 0.6369773626327515, "epoch": 9, "memory": 44138, "step": 34196} +{"lr": 1.3639821409878263e-05, "data_time": 0.001958751678466797, "loss": 0.36187497079372405, "time": 0.6373254537582398, "epoch": 9, "memory": 44138, "step": 34296} +{"lr": 1.3499664201072546e-05, "data_time": 0.0015430927276611328, "loss": 0.3353217601776123, "time": 0.6366066694259643, "epoch": 9, "memory": 44138, "step": 34396} +{"lr": 1.336214951189674e-05, "data_time": 0.0016913890838623046, "loss": 0.35445632338523864, "time": 0.6360156059265136, "epoch": 9, "memory": 44138, "step": 34496} +{"lr": 1.3227286098551456e-05, "data_time": 0.0017403125762939452, "loss": 0.3315434277057648, "time": 0.6366602659225464, "epoch": 9, "memory": 44138, "step": 34596} +{"lr": 1.3095082548418224e-05, "data_time": 0.0015049219131469727, "loss": 0.3482250928878784, "time": 0.636380672454834, "epoch": 9, "memory": 44138, "step": 34696} +{"lr": 1.2965547279512709e-05, "data_time": 0.001609206199645996, "loss": 0.31899039447307587, "time": 0.6360365152359009, "epoch": 9, "memory": 44138, "step": 34796} +{"lr": 1.2838688539948659e-05, "data_time": 0.0014748811721801759, "loss": 0.34317450523376464, "time": 0.637748908996582, "epoch": 9, "memory": 44138, "step": 34896} +{"lr": 1.2714514407412763e-05, "data_time": 0.0015018701553344727, "loss": 0.36702142357826234, "time": 0.6384203910827637, "epoch": 9, "memory": 44138, "step": 34996} +{"lr": 1.2593032788650256e-05, "data_time": 0.002281594276428223, "loss": 0.33379718363285066, "time": 0.6369554996490479, "epoch": 9, "memory": 44138, "step": 35096} +{"lr": 1.2474251418961484e-05, "data_time": 0.0015685319900512694, "loss": 0.3442933946847916, "time": 0.6369629621505737, "epoch": 9, "memory": 44138, "step": 35196} +{"lr": 1.2358177861709387e-05, "data_time": 0.0015196800231933594, "loss": 0.33880884349346163, "time": 0.6360907793045044, "epoch": 9, "memory": 44138, "step": 35296} +{"lr": 1.2244819507837855e-05, "data_time": 0.0014713048934936524, "loss": 0.3282599478960037, "time": 0.6371447563171386, "epoch": 9, "memory": 44138, "step": 35396} +{"accuracy/top1": 94.74966430664062, "data_time": 0.018098025486387056, "time": 1.2612294616370365, "step": 9} +{"lr": 1.2093939694118031e-05, "data_time": 0.0017571210861206054, "loss": 0.3731055498123169, "time": 0.6378751277923584, "epoch": 10, "memory": 44138, "step": 35533} +{"lr": 1.1987044880988498e-05, "data_time": 0.0017752885818481446, "loss": 0.3633056372404099, "time": 0.6374515295028687, "epoch": 10, "memory": 44138, "step": 35633} +{"lr": 1.1882888903004625e-05, "data_time": 0.0017711400985717773, "loss": 0.3484418958425522, "time": 0.6378175497055054, "epoch": 10, "memory": 44138, "step": 35733} +{"lr": 1.178147839226253e-05, "data_time": 0.0014324665069580078, "loss": 0.34249356389045715, "time": 0.6419201612472534, "epoch": 10, "memory": 44138, "step": 35833} +{"lr": 1.1682819806041682e-05, "data_time": 0.0020990848541259767, "loss": 0.33985520601272584, "time": 0.6377373695373535, "epoch": 10, "memory": 44138, "step": 35933} +{"lr": 1.1586919426393653e-05, "data_time": 0.0016060113906860352, "loss": 0.36118035614490507, "time": 0.6364832639694213, "epoch": 10, "memory": 44138, "step": 36033} +{"lr": 1.1493783359742196e-05, "data_time": 0.0015542745590209962, "loss": 0.32094216644763945, "time": 0.6369356155395508, "epoch": 10, "memory": 44138, "step": 36133} +{"lr": 1.1403417536494367e-05, "data_time": 0.0015214920043945313, "loss": 0.37214413583278655, "time": 0.6370633363723754, "epoch": 10, "memory": 44138, "step": 36233} +{"lr": 1.1315827710662908e-05, "data_time": 0.0016398668289184571, "loss": 0.316905614733696, "time": 0.6368438482284546, "epoch": 10, "memory": 44138, "step": 36333} +{"lr": 1.1231019459499938e-05, "data_time": 0.0014683723449707032, "loss": 0.3269952654838562, "time": 0.6371765613555909, "epoch": 10, "memory": 44138, "step": 36433} +{"lr": 1.1148998183141688e-05, "data_time": 0.0015017986297607422, "loss": 0.3335738182067871, "time": 0.6380161046981812, "epoch": 10, "memory": 44138, "step": 36533} +{"lr": 1.1069769104264779e-05, "data_time": 0.001564335823059082, "loss": 0.33867660760879514, "time": 0.6375886678695679, "epoch": 10, "memory": 44138, "step": 36633} +{"lr": 1.0993337267753594e-05, "data_time": 0.0017877817153930664, "loss": 0.3430941581726074, "time": 0.638480830192566, "epoch": 10, "memory": 44138, "step": 36733} +{"lr": 1.0919707540379088e-05, "data_time": 0.0019340276718139648, "loss": 0.348739692568779, "time": 0.6372841358184814, "epoch": 10, "memory": 44138, "step": 36833} +{"lr": 1.0848884610488858e-05, "data_time": 0.0015627384185791016, "loss": 0.3504202350974083, "time": 0.6378693103790283, "epoch": 10, "memory": 44138, "step": 36933} +{"lr": 1.0780872987708645e-05, "data_time": 0.001573324203491211, "loss": 0.3205928891897202, "time": 0.6368873834609985, "epoch": 10, "memory": 44138, "step": 37033} +{"lr": 1.0715677002655166e-05, "data_time": 0.001503324508666992, "loss": 0.3238894432783127, "time": 0.6375768661499024, "epoch": 10, "memory": 44138, "step": 37133} +{"lr": 1.0653300806660372e-05, "data_time": 0.0014631986618041993, "loss": 0.3422483205795288, "time": 0.6371843576431274, "epoch": 10, "memory": 44138, "step": 37233} +{"lr": 1.0593748371507151e-05, "data_time": 0.0017275571823120116, "loss": 0.34399558007717135, "time": 0.637434458732605, "epoch": 10, "memory": 44138, "step": 37333} +{"lr": 1.0537023489176325e-05, "data_time": 0.002298450469970703, "loss": 0.33317987620830536, "time": 0.6367875814437867, "epoch": 10, "memory": 44138, "step": 37433} +{"lr": 1.0483129771605328e-05, "data_time": 0.0015595674514770508, "loss": 0.32792393267154696, "time": 0.6376729011535645, "epoch": 10, "memory": 44138, "step": 37533} +{"lr": 1.0432070650458095e-05, "data_time": 0.001458287239074707, "loss": 0.3527918815612793, "time": 0.6363785982131958, "epoch": 10, "memory": 44138, "step": 37633} +{"lr": 1.0383849376906634e-05, "data_time": 0.0017002344131469727, "loss": 0.3225879371166229, "time": 0.6369772911071777, "epoch": 10, "memory": 44138, "step": 37733} +{"lr": 1.0338469021423977e-05, "data_time": 0.0016661882400512695, "loss": 0.3160365730524063, "time": 0.6368574619293212, "epoch": 10, "memory": 44138, "step": 37833} +{"lr": 1.029593247358868e-05, "data_time": 0.001589059829711914, "loss": 0.36551602482795714, "time": 0.638017725944519, "epoch": 10, "memory": 44138, "step": 37933} +{"lr": 1.0256242441900827e-05, "data_time": 0.0017339468002319336, "loss": 0.3381687432527542, "time": 0.6383226871490478, "epoch": 10, "memory": 44138, "step": 38033} +{"lr": 1.0219401453609533e-05, "data_time": 0.0013235092163085937, "loss": 0.3508059173822403, "time": 0.6371890068054199, "epoch": 10, "memory": 44138, "step": 38133} +{"lr": 1.018541185455209e-05, "data_time": 0.0014414787292480469, "loss": 0.3306189596652985, "time": 0.6374693393707276, "epoch": 10, "memory": 44138, "step": 38233} +{"lr": 1.015427580900455e-05, "data_time": 0.0016166210174560548, "loss": 0.3464100152254105, "time": 0.6381880760192871, "epoch": 10, "memory": 44138, "step": 38333} +{"lr": 1.0125995299543914e-05, "data_time": 0.0015708684921264648, "loss": 0.3515030652284622, "time": 0.6374882936477662, "epoch": 10, "memory": 44138, "step": 38433} +{"lr": 1.0100572126921902e-05, "data_time": 0.0018432855606079102, "loss": 0.35819247663021087, "time": 0.6369351387023926, "epoch": 10, "memory": 44138, "step": 38533} +{"lr": 1.007800790995031e-05, "data_time": 0.001888418197631836, "loss": 0.3579746216535568, "time": 0.6378156900405884, "epoch": 10, "memory": 44138, "step": 38633} +{"lr": 1.0058304085397905e-05, "data_time": 0.0015785694122314453, "loss": 0.3148579865694046, "time": 0.6376957178115845, "epoch": 10, "memory": 44138, "step": 38733} +{"lr": 1.0041461907898941e-05, "data_time": 0.0021996259689331054, "loss": 0.3470492124557495, "time": 0.6371596574783325, "epoch": 10, "memory": 44138, "step": 38833} +{"lr": 1.0027482449873297e-05, "data_time": 0.0019378900527954102, "loss": 0.3348171353340149, "time": 0.6371330976486206, "epoch": 10, "memory": 44138, "step": 38933} +{"lr": 1.0016366601458162e-05, "data_time": 0.002123451232910156, "loss": 0.35173400938510896, "time": 0.6460501909255981, "epoch": 10, "memory": 44138, "step": 39033} +{"lr": 1.0008115070451358e-05, "data_time": 0.0018494367599487306, "loss": 0.3442584753036499, "time": 0.6565184354782104, "epoch": 10, "memory": 44138, "step": 39133} +{"lr": 1.0002728382266284e-05, "data_time": 0.0016611337661743164, "loss": 0.3185924857854843, "time": 0.6374598264694213, "epoch": 10, "memory": 44138, "step": 39233} +{"lr": 1.0000206879898465e-05, "data_time": 0.001641702651977539, "loss": 0.33281290233135224, "time": 0.6364509344100953, "epoch": 10, "memory": 44138, "step": 39333} +{"accuracy/top1": 94.77053833007812, "data_time": 0.01525562385032917, "time": 1.2584579319789493, "step": 10} diff --git a/clip_large_pretrain_4x256_IF_lr1e-4/20230606_050006/vis_data/vis_image/0000000.png_1.png b/clip_large_pretrain_4x256_IF_lr1e-4/20230606_050006/vis_data/vis_image/0000000.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..7b6499770a830bcd2ba3170079648fa203af923f Binary files /dev/null and b/clip_large_pretrain_4x256_IF_lr1e-4/20230606_050006/vis_data/vis_image/0000000.png_1.png differ diff --git a/clip_large_pretrain_4x256_IF_lr1e-4/20230606_050006/vis_data/vis_image/0000000.png_10.png b/clip_large_pretrain_4x256_IF_lr1e-4/20230606_050006/vis_data/vis_image/0000000.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..c9e3852a9e5e4bdf054009e883208f7aa1e15cdf Binary files /dev/null and b/clip_large_pretrain_4x256_IF_lr1e-4/20230606_050006/vis_data/vis_image/0000000.png_10.png differ diff --git a/clip_large_pretrain_4x256_IF_lr1e-4/20230606_050006/vis_data/vis_image/0000000.png_2.png b/clip_large_pretrain_4x256_IF_lr1e-4/20230606_050006/vis_data/vis_image/0000000.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..3890e849b748e86d6935a5e83f4ad42599a50ffe Binary files /dev/null and b/clip_large_pretrain_4x256_IF_lr1e-4/20230606_050006/vis_data/vis_image/0000000.png_2.png differ diff --git a/clip_large_pretrain_4x256_IF_lr1e-4/20230606_050006/vis_data/vis_image/0000000.png_3.png b/clip_large_pretrain_4x256_IF_lr1e-4/20230606_050006/vis_data/vis_image/0000000.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..8cc51c6f4a56fb4462c68304928054cb94c03351 Binary files /dev/null and b/clip_large_pretrain_4x256_IF_lr1e-4/20230606_050006/vis_data/vis_image/0000000.png_3.png differ diff --git a/clip_large_pretrain_4x256_IF_lr1e-4/20230606_050006/vis_data/vis_image/0000000.png_4.png b/clip_large_pretrain_4x256_IF_lr1e-4/20230606_050006/vis_data/vis_image/0000000.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..e130c6fdd95368a69db67cd7fcc7ab160845f055 Binary files /dev/null and b/clip_large_pretrain_4x256_IF_lr1e-4/20230606_050006/vis_data/vis_image/0000000.png_4.png differ diff --git a/clip_large_pretrain_4x256_IF_lr1e-4/20230606_050006/vis_data/vis_image/0000000.png_5.png b/clip_large_pretrain_4x256_IF_lr1e-4/20230606_050006/vis_data/vis_image/0000000.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..61a9990e9ed76a1a7082805b14d639816410ca25 Binary files /dev/null and b/clip_large_pretrain_4x256_IF_lr1e-4/20230606_050006/vis_data/vis_image/0000000.png_5.png differ diff --git a/clip_large_pretrain_4x256_IF_lr1e-4/20230606_050006/vis_data/vis_image/0000000.png_6.png b/clip_large_pretrain_4x256_IF_lr1e-4/20230606_050006/vis_data/vis_image/0000000.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..5ba3f0074d76faf9d1af5a225e9a5c626d4bda18 Binary files /dev/null and b/clip_large_pretrain_4x256_IF_lr1e-4/20230606_050006/vis_data/vis_image/0000000.png_6.png differ diff --git a/clip_large_pretrain_4x256_IF_lr1e-4/20230606_050006/vis_data/vis_image/0000000.png_7.png b/clip_large_pretrain_4x256_IF_lr1e-4/20230606_050006/vis_data/vis_image/0000000.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..efdd9c3546b722d7472ff38eaeb4ab6e16062438 Binary files /dev/null and b/clip_large_pretrain_4x256_IF_lr1e-4/20230606_050006/vis_data/vis_image/0000000.png_7.png differ diff --git a/clip_large_pretrain_4x256_IF_lr1e-4/20230606_050006/vis_data/vis_image/0000000.png_8.png b/clip_large_pretrain_4x256_IF_lr1e-4/20230606_050006/vis_data/vis_image/0000000.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..4b8c41d4f1fff968f0934049fcd8edb18691b07e Binary files /dev/null and b/clip_large_pretrain_4x256_IF_lr1e-4/20230606_050006/vis_data/vis_image/0000000.png_8.png differ diff --git a/clip_large_pretrain_4x256_IF_lr1e-4/20230606_050006/vis_data/vis_image/0000000.png_9.png b/clip_large_pretrain_4x256_IF_lr1e-4/20230606_050006/vis_data/vis_image/0000000.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..43c3c7615065e6cf762bfcc0be8cd2c9186479c8 Binary files /dev/null and b/clip_large_pretrain_4x256_IF_lr1e-4/20230606_050006/vis_data/vis_image/0000000.png_9.png differ diff --git a/clip_large_pretrain_4x256_IF_lr1e-4/20230606_050006/vis_data/vis_image/000005174.jpg_1.png b/clip_large_pretrain_4x256_IF_lr1e-4/20230606_050006/vis_data/vis_image/000005174.jpg_1.png new file mode 100644 index 0000000000000000000000000000000000000000..7a288872e1fb0cee27303c26788c699586fac109 Binary files /dev/null and b/clip_large_pretrain_4x256_IF_lr1e-4/20230606_050006/vis_data/vis_image/000005174.jpg_1.png differ diff --git a/clip_large_pretrain_4x256_IF_lr1e-4/20230606_050006/vis_data/vis_image/000005174.jpg_10.png b/clip_large_pretrain_4x256_IF_lr1e-4/20230606_050006/vis_data/vis_image/000005174.jpg_10.png new file mode 100644 index 0000000000000000000000000000000000000000..a26c0d9d566a8ac21a83cb1b833a367a981e0eda Binary files /dev/null and b/clip_large_pretrain_4x256_IF_lr1e-4/20230606_050006/vis_data/vis_image/000005174.jpg_10.png differ diff --git a/clip_large_pretrain_4x256_IF_lr1e-4/20230606_050006/vis_data/vis_image/000005174.jpg_2.png b/clip_large_pretrain_4x256_IF_lr1e-4/20230606_050006/vis_data/vis_image/000005174.jpg_2.png new file mode 100644 index 0000000000000000000000000000000000000000..a291b1803c56a02d5387b4f4a2ce6cc1b6a61027 Binary files /dev/null and b/clip_large_pretrain_4x256_IF_lr1e-4/20230606_050006/vis_data/vis_image/000005174.jpg_2.png differ diff --git a/clip_large_pretrain_4x256_IF_lr1e-4/20230606_050006/vis_data/vis_image/000005174.jpg_3.png b/clip_large_pretrain_4x256_IF_lr1e-4/20230606_050006/vis_data/vis_image/000005174.jpg_3.png new file mode 100644 index 0000000000000000000000000000000000000000..223097446b1d7f6396334efe3d318c0131a669c7 Binary files /dev/null and b/clip_large_pretrain_4x256_IF_lr1e-4/20230606_050006/vis_data/vis_image/000005174.jpg_3.png differ diff --git a/clip_large_pretrain_4x256_IF_lr1e-4/20230606_050006/vis_data/vis_image/000005174.jpg_4.png b/clip_large_pretrain_4x256_IF_lr1e-4/20230606_050006/vis_data/vis_image/000005174.jpg_4.png new file mode 100644 index 0000000000000000000000000000000000000000..9f811d585f2aee565500a84670b32766afb8dfeb Binary files /dev/null and b/clip_large_pretrain_4x256_IF_lr1e-4/20230606_050006/vis_data/vis_image/000005174.jpg_4.png differ diff --git a/clip_large_pretrain_4x256_IF_lr1e-4/20230606_050006/vis_data/vis_image/000005174.jpg_5.png b/clip_large_pretrain_4x256_IF_lr1e-4/20230606_050006/vis_data/vis_image/000005174.jpg_5.png new file mode 100644 index 0000000000000000000000000000000000000000..7b70792e01de921f7611d946bb048fa474ff71e5 Binary files /dev/null and b/clip_large_pretrain_4x256_IF_lr1e-4/20230606_050006/vis_data/vis_image/000005174.jpg_5.png differ diff --git a/clip_large_pretrain_4x256_IF_lr1e-4/20230606_050006/vis_data/vis_image/000005174.jpg_6.png b/clip_large_pretrain_4x256_IF_lr1e-4/20230606_050006/vis_data/vis_image/000005174.jpg_6.png new file mode 100644 index 0000000000000000000000000000000000000000..c6ce8f858aba7e488404eccc2e3feca817743050 Binary files /dev/null and b/clip_large_pretrain_4x256_IF_lr1e-4/20230606_050006/vis_data/vis_image/000005174.jpg_6.png differ diff --git a/clip_large_pretrain_4x256_IF_lr1e-4/20230606_050006/vis_data/vis_image/000005174.jpg_7.png b/clip_large_pretrain_4x256_IF_lr1e-4/20230606_050006/vis_data/vis_image/000005174.jpg_7.png new file mode 100644 index 0000000000000000000000000000000000000000..90054efde15b4f516a67e13d3ca502d7758658d9 Binary files /dev/null and b/clip_large_pretrain_4x256_IF_lr1e-4/20230606_050006/vis_data/vis_image/000005174.jpg_7.png differ diff --git a/clip_large_pretrain_4x256_IF_lr1e-4/20230606_050006/vis_data/vis_image/000005174.jpg_8.png b/clip_large_pretrain_4x256_IF_lr1e-4/20230606_050006/vis_data/vis_image/000005174.jpg_8.png new file mode 100644 index 0000000000000000000000000000000000000000..1768622b009b190b5501af6397f4ce73ac6fe5c7 Binary files /dev/null and b/clip_large_pretrain_4x256_IF_lr1e-4/20230606_050006/vis_data/vis_image/000005174.jpg_8.png differ diff --git a/clip_large_pretrain_4x256_IF_lr1e-4/20230606_050006/vis_data/vis_image/000005174.jpg_9.png b/clip_large_pretrain_4x256_IF_lr1e-4/20230606_050006/vis_data/vis_image/000005174.jpg_9.png new file mode 100644 index 0000000000000000000000000000000000000000..603e09508277a53d352ddc4928501bcbecc2f815 Binary files /dev/null and b/clip_large_pretrain_4x256_IF_lr1e-4/20230606_050006/vis_data/vis_image/000005174.jpg_9.png differ diff --git a/clip_large_pretrain_4x256_IF_lr1e-4/clip_large_pretrain_4x256_IF_lr1e-4.py b/clip_large_pretrain_4x256_IF_lr1e-4/clip_large_pretrain_4x256_IF_lr1e-4.py new file mode 100644 index 0000000000000000000000000000000000000000..98cdad2fa2783bf5a6865ddc1352c928f69609b0 --- /dev/null +++ b/clip_large_pretrain_4x256_IF_lr1e-4/clip_large_pretrain_4x256_IF_lr1e-4.py @@ -0,0 +1,251 @@ +optim_wrapper = dict( + optimizer=dict( + type='AdamW', lr=0.0001, weight_decay=0.3, _scope_='mmpretrain'), + paramwise_cfg=dict( + custom_keys=dict({ + '.cls_token': dict(decay_mult=0.0), + '.pos_embed': dict(decay_mult=0.0) + })), + type='AmpOptimWrapper', + dtype='bfloat16', + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=False, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=4096) +model = dict( + type='ImageClassifier', + backbone=dict( + frozen_stages=24, + type='VisionTransformer', + arch='l', + img_size=224, + patch_size=14, + drop_rate=0.1, + pre_norm=True, + final_norm=False, + init_cfg=dict( + type='Pretrained', + checkpoint='ckpt/openclip-ViT-L-14.pth', + prefix='backbone')), + neck=dict( + type='CLIPProjection', + in_channels=1024, + out_channels=768, + init_cfg=dict( + type='Pretrained', + checkpoint='ckpt/openclip-ViT-L-14.pth', + prefix='backbone')), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=768, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + init_cfg=None), + init_cfg=dict( + type='TruncNormal', layer=['Conv2d', 'Linear'], std=0.02, bias=0.0), + train_cfg=None) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=128, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/IF80w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/if-dpmsolver++-50-20w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/cc1m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=128, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/if-dpmsolver++-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=128, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/if-dpmsolver++-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/clip_large_pretrain_4x256_IF_lr1e-4' diff --git a/clip_large_pretrain_4x256_IF_lr1e-4/epoch_1.pth b/clip_large_pretrain_4x256_IF_lr1e-4/epoch_1.pth new file mode 100644 index 0000000000000000000000000000000000000000..8800a29ab2b96e223fbae8dd91bc3dffe6b61860 --- /dev/null +++ b/clip_large_pretrain_4x256_IF_lr1e-4/epoch_1.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:18ee6b59587efecc15bbca73392be42fac140326b53692a1f81b4b5cfd763c22 +size 2438619201 diff --git a/clip_large_pretrain_4x256_IF_lr1e-4/epoch_10.pth b/clip_large_pretrain_4x256_IF_lr1e-4/epoch_10.pth new file mode 100644 index 0000000000000000000000000000000000000000..9b36f5b0a33375635c3b5efaf4356ceb2523e71e --- /dev/null +++ b/clip_large_pretrain_4x256_IF_lr1e-4/epoch_10.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:0ea76eb379604078ed76d9bf2ebfd6f0b134790a51ac16d15b1f3dbace7f6bd9 +size 2441209089 diff --git a/clip_large_pretrain_4x256_IF_lr1e-4/epoch_2.pth b/clip_large_pretrain_4x256_IF_lr1e-4/epoch_2.pth new file mode 100644 index 0000000000000000000000000000000000000000..a064054fbe2f27004107c9d6fb1c6c125d12b884 --- /dev/null +++ b/clip_large_pretrain_4x256_IF_lr1e-4/epoch_2.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:d39604da58d0af42f3809c414df86d67a39dc318eb833c191f84ac44589615cf +size 2438905985 diff --git a/clip_large_pretrain_4x256_IF_lr1e-4/epoch_3.pth b/clip_large_pretrain_4x256_IF_lr1e-4/epoch_3.pth new file mode 100644 index 0000000000000000000000000000000000000000..cf16118c8161f57645e5e1ab05b13214d1f6fa29 --- /dev/null +++ b/clip_large_pretrain_4x256_IF_lr1e-4/epoch_3.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:dd978829e56a03d4f8a8407b24d8cf786ec946a70e357715dd6adec032d98932 +size 2439191873 diff --git a/clip_large_pretrain_4x256_IF_lr1e-4/epoch_4.pth b/clip_large_pretrain_4x256_IF_lr1e-4/epoch_4.pth new file mode 100644 index 0000000000000000000000000000000000000000..c344f1def2bbd3deb0535d1f59e3aa85f9818402 --- /dev/null +++ b/clip_large_pretrain_4x256_IF_lr1e-4/epoch_4.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:923802d3e8d2e586e8ad81992fed3edefcc07b51e3ab0bd7906f46dd717d9c4f +size 2439477761 diff --git a/clip_large_pretrain_4x256_IF_lr1e-4/epoch_5.pth b/clip_large_pretrain_4x256_IF_lr1e-4/epoch_5.pth new file mode 100644 index 0000000000000000000000000000000000000000..e2ade37bfa7376c59c09ab360024c5ca173d2db2 --- /dev/null +++ b/clip_large_pretrain_4x256_IF_lr1e-4/epoch_5.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:da4e397f66ffe41234e9dd8c117c15514e9d42cd53f0b2c4147719f954c04f27 +size 2439764801 diff --git a/clip_large_pretrain_4x256_IF_lr1e-4/epoch_6.pth b/clip_large_pretrain_4x256_IF_lr1e-4/epoch_6.pth new file mode 100644 index 0000000000000000000000000000000000000000..cb379eff69d4efc4a951812a12c383942368bb0b --- /dev/null +++ b/clip_large_pretrain_4x256_IF_lr1e-4/epoch_6.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:0cb0393bfc095953b4e657b643a5b9e8a53b205d37538504994f852e221cf353 +size 2440051393 diff --git a/clip_large_pretrain_4x256_IF_lr1e-4/epoch_7.pth b/clip_large_pretrain_4x256_IF_lr1e-4/epoch_7.pth new file mode 100644 index 0000000000000000000000000000000000000000..67a17afdda0da2d99f782e4e3289732bb1bfea0f --- /dev/null +++ b/clip_large_pretrain_4x256_IF_lr1e-4/epoch_7.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:56a81ddcee48c16d448023f78b1a232d962a948009fefb003b27e204f459c018 +size 2440339009 diff --git a/clip_large_pretrain_4x256_IF_lr1e-4/epoch_8.pth b/clip_large_pretrain_4x256_IF_lr1e-4/epoch_8.pth new file mode 100644 index 0000000000000000000000000000000000000000..c045bf09f26184770ed3c729f8f67cc2c57848ed --- /dev/null +++ b/clip_large_pretrain_4x256_IF_lr1e-4/epoch_8.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:d5a64004fb920448221aa1c7e65d168d509c99e67d5ce1a4f5ce378bb5c55a04 +size 2440629441 diff --git a/clip_large_pretrain_4x256_IF_lr1e-4/epoch_9.pth b/clip_large_pretrain_4x256_IF_lr1e-4/epoch_9.pth new file mode 100644 index 0000000000000000000000000000000000000000..80ce9969d62f0d3cc83009c604e97dff2408a6ad --- /dev/null +++ b/clip_large_pretrain_4x256_IF_lr1e-4/epoch_9.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:e50e0016b0c8998be2ad5a000d1416e45b59fcaf82027d39ba3a2d9892765021 +size 2440919425 diff --git a/clip_large_pretrain_4x256_IF_lr1e-4/last_checkpoint b/clip_large_pretrain_4x256_IF_lr1e-4/last_checkpoint new file mode 100644 index 0000000000000000000000000000000000000000..f17e9190e8419caace80012cb295ce2e94ecc865 --- /dev/null +++ b/clip_large_pretrain_4x256_IF_lr1e-4/last_checkpoint @@ -0,0 +1 @@ +/mnt/petrelfs/luzeyu/workspace/fakebench/mmpretrain/workdir/clip_large_pretrain_4x256_IF_lr1e-4/epoch_10.pth \ No newline at end of file diff --git a/clip_large_pretrain_4x256_all2_lr1e-4/20230606_005614/20230606_005614.log b/clip_large_pretrain_4x256_all2_lr1e-4/20230606_005614/20230606_005614.log new file mode 100644 index 0000000000000000000000000000000000000000..dbe809e94e4855503fee62d37014aba08d9f1923 --- /dev/null +++ b/clip_large_pretrain_4x256_all2_lr1e-4/20230606_005614/20230606_005614.log @@ -0,0 +1,1853 @@ +2023/06/06 00:56:18 - mmengine - INFO - +------------------------------------------------------------ +System environment: + sys.platform: linux + Python: 3.10.9 (main, Mar 8 2023, 10:47:38) [GCC 11.2.0] + CUDA available: True + numpy_random_seed: 2026736370 + GPU 0,1,2,3: NVIDIA A100-SXM4-80GB + CUDA_HOME: /mnt/petrelfs/share/cuda-11.6 + NVCC: Cuda compilation tools, release 11.6, V11.6.124 + GCC: gcc (GCC) 7.5.0 + PyTorch: 1.13.1 + PyTorch compiling details: PyTorch built with: + - GCC 9.3 + - C++ Version: 201402 + - Intel(R) oneAPI Math Kernel Library Version 2021.4-Product Build 20210904 for Intel(R) 64 architecture applications + - Intel(R) MKL-DNN v2.6.0 (Git Hash 52b5f107dd9cf10910aaa19cb47f3abf9b349815) + - OpenMP 201511 (a.k.a. OpenMP 4.5) + - LAPACK is enabled (usually provided by MKL) + - NNPACK is enabled + - CPU capability usage: AVX2 + - CUDA Runtime 11.6 + - NVCC architecture flags: -gencode;arch=compute_37,code=sm_37;-gencode;arch=compute_50,code=sm_50;-gencode;arch=compute_60,code=sm_60;-gencode;arch=compute_61,code=sm_61;-gencode;arch=compute_70,code=sm_70;-gencode;arch=compute_75,code=sm_75;-gencode;arch=compute_80,code=sm_80;-gencode;arch=compute_86,code=sm_86;-gencode;arch=compute_37,code=compute_37 + - CuDNN 8.3.2 (built against CUDA 11.5) + - Magma 2.6.1 + - Build settings: BLAS_INFO=mkl, BUILD_TYPE=Release, CUDA_VERSION=11.6, CUDNN_VERSION=8.3.2, CXX_COMPILER=/opt/rh/devtoolset-9/root/usr/bin/c++, CXX_FLAGS= -fabi-version=11 -Wno-deprecated -fvisibility-inlines-hidden -DUSE_PTHREADPOOL -fopenmp -DNDEBUG -DUSE_KINETO -DUSE_FBGEMM -DUSE_QNNPACK -DUSE_PYTORCH_QNNPACK -DUSE_XNNPACK -DSYMBOLICATE_MOBILE_DEBUG_HANDLE -DEDGE_PROFILER_USE_KINETO -O2 -fPIC -Wno-narrowing -Wall -Wextra -Werror=return-type -Werror=non-virtual-dtor -Wno-missing-field-initializers -Wno-type-limits -Wno-array-bounds -Wno-unknown-pragmas -Wunused-local-typedefs -Wno-unused-parameter -Wno-unused-function -Wno-unused-result -Wno-strict-overflow -Wno-strict-aliasing -Wno-error=deprecated-declarations -Wno-stringop-overflow -Wno-psabi -Wno-error=pedantic -Wno-error=redundant-decls -Wno-error=old-style-cast -fdiagnostics-color=always -faligned-new -Wno-unused-but-set-variable -Wno-maybe-uninitialized -fno-math-errno -fno-trapping-math -Werror=format -Werror=cast-function-type -Wno-stringop-overflow, LAPACK_INFO=mkl, PERF_WITH_AVX=1, PERF_WITH_AVX2=1, PERF_WITH_AVX512=1, TORCH_VERSION=1.13.1, USE_CUDA=ON, USE_CUDNN=ON, USE_EXCEPTION_PTR=1, USE_GFLAGS=OFF, USE_GLOG=OFF, USE_MKL=ON, USE_MKLDNN=ON, USE_MPI=OFF, USE_NCCL=ON, USE_NNPACK=ON, USE_OPENMP=ON, USE_ROCM=OFF, + + TorchVision: 0.14.1 + OpenCV: 4.7.0 + MMEngine: 0.7.3 + +Runtime environment: + cudnn_benchmark: True + mp_cfg: {'mp_start_method': 'fork', 'opencv_num_threads': 0} + dist_cfg: {'backend': 'nccl'} + seed: None + deterministic: False + Distributed launcher: slurm + Distributed training: True + GPU number: 4 +------------------------------------------------------------ + +2023/06/06 00:56:22 - mmengine - INFO - Config: +optim_wrapper = dict( + optimizer=dict( + type='AdamW', lr=0.0001, weight_decay=0.3, _scope_='mmpretrain'), + paramwise_cfg=dict( + custom_keys=dict({ + '.cls_token': dict(decay_mult=0.0), + '.pos_embed': dict(decay_mult=0.0) + })), + type='AmpOptimWrapper', + dtype='bfloat16', + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=False, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=4096) +model = dict( + type='ImageClassifier', + backbone=dict( + frozen_stages=24, + type='VisionTransformer', + arch='l', + img_size=224, + patch_size=14, + drop_rate=0.1, + pre_norm=True, + final_norm=False, + init_cfg=dict( + type='Pretrained', + checkpoint='ckpt/openclip-ViT-L-14.pth', + prefix='backbone')), + neck=dict( + type='CLIPProjection', + in_channels=1024, + out_channels=768, + init_cfg=dict( + type='Pretrained', + checkpoint='ckpt/openclip-ViT-L-14.pth', + prefix='backbone')), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=768, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + init_cfg=None), + init_cfg=dict( + type='TruncNormal', layer=['Conv2d', 'Linear'], std=0.02, bias=0.0), + train_cfg=None) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=128, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/all_0.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/all_1.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3fake8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/cc1m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3real8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=128, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV2-1-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/if-dpmsolver++-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=128, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV2-1-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/if-dpmsolver++-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/clip_large_pretrain_4x256_all2_lr1e-4' + +2023/06/06 00:56:34 - mmengine - INFO - Hooks will be executed in the following order: +before_run: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_load_checkpoint: +(ABOVE_NORMAL) EMAHook + -------------------- +before_train: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_train_epoch: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(NORMAL ) DistSamplerSeedHook + -------------------- +before_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook + -------------------- +after_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +after_train_epoch: +(NORMAL ) IterTimerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_val_epoch: +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook + -------------------- +before_val_iter: +(NORMAL ) IterTimerHook + -------------------- +after_val_iter: +(NORMAL ) IterTimerHook +(NORMAL ) VisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_val_epoch: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_save_checkpoint: +(ABOVE_NORMAL) EMAHook + -------------------- +after_train: +(VERY_LOW ) CheckpointHook + -------------------- +before_test_epoch: +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook + -------------------- +before_test_iter: +(NORMAL ) IterTimerHook + -------------------- +after_test_iter: +(NORMAL ) IterTimerHook +(NORMAL ) VisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_test_epoch: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_run: +(BELOW_NORMAL) LoggerHook + -------------------- +2023/06/06 00:56:54 - mmengine - INFO - load backbone in model from: ckpt/openclip-ViT-L-14.pth +2023/06/06 00:56:56 - mmengine - WARNING - The model and loaded state dict do not match exactly + +unexpected key in source state_dict: ln1.weight, ln1.bias + +2023/06/06 00:56:56 - mmengine - INFO - load backbone in model from: ckpt/openclip-ViT-L-14.pth +2023/06/06 00:56:57 - mmengine - WARNING - The model and loaded state dict do not match exactly + +unexpected key in source state_dict: cls_token, pos_embed, patch_embed.projection.weight, pre_norm.weight, pre_norm.bias, layers.0.ln1.weight, layers.0.ln1.bias, layers.0.attn.qkv.weight, layers.0.attn.qkv.bias, layers.0.attn.proj.weight, layers.0.attn.proj.bias, layers.0.ln2.weight, layers.0.ln2.bias, layers.0.ffn.layers.0.0.weight, layers.0.ffn.layers.0.0.bias, layers.0.ffn.layers.1.weight, layers.0.ffn.layers.1.bias, layers.1.ln1.weight, layers.1.ln1.bias, layers.1.attn.qkv.weight, layers.1.attn.qkv.bias, layers.1.attn.proj.weight, layers.1.attn.proj.bias, layers.1.ln2.weight, layers.1.ln2.bias, layers.1.ffn.layers.0.0.weight, layers.1.ffn.layers.0.0.bias, layers.1.ffn.layers.1.weight, layers.1.ffn.layers.1.bias, layers.2.ln1.weight, layers.2.ln1.bias, layers.2.attn.qkv.weight, layers.2.attn.qkv.bias, layers.2.attn.proj.weight, layers.2.attn.proj.bias, layers.2.ln2.weight, layers.2.ln2.bias, layers.2.ffn.layers.0.0.weight, layers.2.ffn.layers.0.0.bias, layers.2.ffn.layers.1.weight, layers.2.ffn.layers.1.bias, layers.3.ln1.weight, layers.3.ln1.bias, layers.3.attn.qkv.weight, layers.3.attn.qkv.bias, layers.3.attn.proj.weight, layers.3.attn.proj.bias, layers.3.ln2.weight, layers.3.ln2.bias, layers.3.ffn.layers.0.0.weight, layers.3.ffn.layers.0.0.bias, layers.3.ffn.layers.1.weight, layers.3.ffn.layers.1.bias, layers.4.ln1.weight, layers.4.ln1.bias, layers.4.attn.qkv.weight, layers.4.attn.qkv.bias, layers.4.attn.proj.weight, layers.4.attn.proj.bias, layers.4.ln2.weight, layers.4.ln2.bias, layers.4.ffn.layers.0.0.weight, layers.4.ffn.layers.0.0.bias, layers.4.ffn.layers.1.weight, layers.4.ffn.layers.1.bias, layers.5.ln1.weight, layers.5.ln1.bias, layers.5.attn.qkv.weight, layers.5.attn.qkv.bias, layers.5.attn.proj.weight, layers.5.attn.proj.bias, layers.5.ln2.weight, layers.5.ln2.bias, layers.5.ffn.layers.0.0.weight, layers.5.ffn.layers.0.0.bias, layers.5.ffn.layers.1.weight, layers.5.ffn.layers.1.bias, layers.6.ln1.weight, layers.6.ln1.bias, layers.6.attn.qkv.weight, layers.6.attn.qkv.bias, layers.6.attn.proj.weight, layers.6.attn.proj.bias, layers.6.ln2.weight, layers.6.ln2.bias, layers.6.ffn.layers.0.0.weight, layers.6.ffn.layers.0.0.bias, layers.6.ffn.layers.1.weight, layers.6.ffn.layers.1.bias, layers.7.ln1.weight, layers.7.ln1.bias, layers.7.attn.qkv.weight, layers.7.attn.qkv.bias, layers.7.attn.proj.weight, layers.7.attn.proj.bias, layers.7.ln2.weight, layers.7.ln2.bias, layers.7.ffn.layers.0.0.weight, layers.7.ffn.layers.0.0.bias, layers.7.ffn.layers.1.weight, layers.7.ffn.layers.1.bias, layers.8.ln1.weight, layers.8.ln1.bias, layers.8.attn.qkv.weight, layers.8.attn.qkv.bias, layers.8.attn.proj.weight, layers.8.attn.proj.bias, layers.8.ln2.weight, layers.8.ln2.bias, layers.8.ffn.layers.0.0.weight, layers.8.ffn.layers.0.0.bias, layers.8.ffn.layers.1.weight, layers.8.ffn.layers.1.bias, layers.9.ln1.weight, layers.9.ln1.bias, layers.9.attn.qkv.weight, layers.9.attn.qkv.bias, layers.9.attn.proj.weight, layers.9.attn.proj.bias, layers.9.ln2.weight, layers.9.ln2.bias, layers.9.ffn.layers.0.0.weight, layers.9.ffn.layers.0.0.bias, layers.9.ffn.layers.1.weight, layers.9.ffn.layers.1.bias, layers.10.ln1.weight, layers.10.ln1.bias, layers.10.attn.qkv.weight, layers.10.attn.qkv.bias, layers.10.attn.proj.weight, layers.10.attn.proj.bias, layers.10.ln2.weight, layers.10.ln2.bias, layers.10.ffn.layers.0.0.weight, layers.10.ffn.layers.0.0.bias, layers.10.ffn.layers.1.weight, layers.10.ffn.layers.1.bias, layers.11.ln1.weight, layers.11.ln1.bias, layers.11.attn.qkv.weight, layers.11.attn.qkv.bias, layers.11.attn.proj.weight, layers.11.attn.proj.bias, layers.11.ln2.weight, layers.11.ln2.bias, layers.11.ffn.layers.0.0.weight, layers.11.ffn.layers.0.0.bias, layers.11.ffn.layers.1.weight, layers.11.ffn.layers.1.bias, layers.12.ln1.weight, layers.12.ln1.bias, layers.12.attn.qkv.weight, layers.12.attn.qkv.bias, layers.12.attn.proj.weight, layers.12.attn.proj.bias, layers.12.ln2.weight, layers.12.ln2.bias, layers.12.ffn.layers.0.0.weight, layers.12.ffn.layers.0.0.bias, layers.12.ffn.layers.1.weight, layers.12.ffn.layers.1.bias, layers.13.ln1.weight, layers.13.ln1.bias, layers.13.attn.qkv.weight, layers.13.attn.qkv.bias, layers.13.attn.proj.weight, layers.13.attn.proj.bias, layers.13.ln2.weight, layers.13.ln2.bias, layers.13.ffn.layers.0.0.weight, layers.13.ffn.layers.0.0.bias, layers.13.ffn.layers.1.weight, layers.13.ffn.layers.1.bias, layers.14.ln1.weight, layers.14.ln1.bias, layers.14.attn.qkv.weight, layers.14.attn.qkv.bias, layers.14.attn.proj.weight, layers.14.attn.proj.bias, layers.14.ln2.weight, layers.14.ln2.bias, layers.14.ffn.layers.0.0.weight, layers.14.ffn.layers.0.0.bias, layers.14.ffn.layers.1.weight, layers.14.ffn.layers.1.bias, layers.15.ln1.weight, layers.15.ln1.bias, layers.15.attn.qkv.weight, layers.15.attn.qkv.bias, layers.15.attn.proj.weight, layers.15.attn.proj.bias, layers.15.ln2.weight, layers.15.ln2.bias, layers.15.ffn.layers.0.0.weight, layers.15.ffn.layers.0.0.bias, layers.15.ffn.layers.1.weight, layers.15.ffn.layers.1.bias, layers.16.ln1.weight, layers.16.ln1.bias, layers.16.attn.qkv.weight, layers.16.attn.qkv.bias, layers.16.attn.proj.weight, layers.16.attn.proj.bias, layers.16.ln2.weight, layers.16.ln2.bias, layers.16.ffn.layers.0.0.weight, layers.16.ffn.layers.0.0.bias, layers.16.ffn.layers.1.weight, layers.16.ffn.layers.1.bias, layers.17.ln1.weight, layers.17.ln1.bias, layers.17.attn.qkv.weight, layers.17.attn.qkv.bias, layers.17.attn.proj.weight, layers.17.attn.proj.bias, layers.17.ln2.weight, layers.17.ln2.bias, layers.17.ffn.layers.0.0.weight, layers.17.ffn.layers.0.0.bias, layers.17.ffn.layers.1.weight, layers.17.ffn.layers.1.bias, layers.18.ln1.weight, layers.18.ln1.bias, layers.18.attn.qkv.weight, layers.18.attn.qkv.bias, layers.18.attn.proj.weight, layers.18.attn.proj.bias, layers.18.ln2.weight, layers.18.ln2.bias, layers.18.ffn.layers.0.0.weight, layers.18.ffn.layers.0.0.bias, layers.18.ffn.layers.1.weight, layers.18.ffn.layers.1.bias, layers.19.ln1.weight, layers.19.ln1.bias, layers.19.attn.qkv.weight, layers.19.attn.qkv.bias, layers.19.attn.proj.weight, layers.19.attn.proj.bias, layers.19.ln2.weight, layers.19.ln2.bias, layers.19.ffn.layers.0.0.weight, layers.19.ffn.layers.0.0.bias, layers.19.ffn.layers.1.weight, layers.19.ffn.layers.1.bias, layers.20.ln1.weight, layers.20.ln1.bias, layers.20.attn.qkv.weight, layers.20.attn.qkv.bias, layers.20.attn.proj.weight, layers.20.attn.proj.bias, layers.20.ln2.weight, layers.20.ln2.bias, layers.20.ffn.layers.0.0.weight, layers.20.ffn.layers.0.0.bias, layers.20.ffn.layers.1.weight, layers.20.ffn.layers.1.bias, layers.21.ln1.weight, layers.21.ln1.bias, layers.21.attn.qkv.weight, layers.21.attn.qkv.bias, layers.21.attn.proj.weight, layers.21.attn.proj.bias, layers.21.ln2.weight, layers.21.ln2.bias, layers.21.ffn.layers.0.0.weight, layers.21.ffn.layers.0.0.bias, layers.21.ffn.layers.1.weight, layers.21.ffn.layers.1.bias, layers.22.ln1.weight, layers.22.ln1.bias, layers.22.attn.qkv.weight, layers.22.attn.qkv.bias, layers.22.attn.proj.weight, layers.22.attn.proj.bias, layers.22.ln2.weight, layers.22.ln2.bias, layers.22.ffn.layers.0.0.weight, layers.22.ffn.layers.0.0.bias, layers.22.ffn.layers.1.weight, layers.22.ffn.layers.1.bias, layers.23.ln1.weight, layers.23.ln1.bias, layers.23.attn.qkv.weight, layers.23.attn.qkv.bias, layers.23.attn.proj.weight, layers.23.attn.proj.bias, layers.23.ln2.weight, layers.23.ln2.bias, layers.23.ffn.layers.0.0.weight, layers.23.ffn.layers.0.0.bias, layers.23.ffn.layers.1.weight, layers.23.ffn.layers.1.bias, ln1.weight, ln1.bias + +missing keys in source state_dict: proj + +Name of parameter - Initialization information + +backbone.cls_token - torch.Size([1, 1, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.pos_embed - torch.Size([1, 257, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.patch_embed.projection.weight - torch.Size([1024, 3, 14, 14]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.0.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.0.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.0.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.0.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.0.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.0.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.0.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.0.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.0.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.0.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.0.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.0.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.1.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.1.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.1.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.1.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.1.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.1.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.1.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.1.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.1.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.1.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.1.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.1.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.2.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.2.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.2.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.2.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.2.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.2.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.2.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.2.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.2.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.2.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.2.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.2.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.3.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.3.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.3.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.3.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.3.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.3.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.3.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.3.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.3.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.3.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.3.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.3.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.4.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.4.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.4.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.4.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.4.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.4.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.4.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.4.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.4.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.4.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.4.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.4.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.5.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.5.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.5.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.5.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.5.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.5.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.5.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.5.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.5.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.5.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.5.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.5.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.6.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.6.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.6.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.6.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.6.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.6.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.6.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.6.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.6.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.6.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.6.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.6.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.7.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.7.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.7.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.7.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.7.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.7.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.7.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.7.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.7.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.7.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.7.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.7.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.8.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.8.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.8.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.8.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.8.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.8.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.8.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.8.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.8.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.8.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.8.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.8.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.9.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.9.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.9.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.9.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.9.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.9.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.9.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.9.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.9.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.9.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.9.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.9.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.10.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.10.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.10.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.10.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.10.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.10.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.10.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.10.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.10.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.10.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.10.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.10.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.11.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.11.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.11.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.11.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.11.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.11.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.11.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.11.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.11.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.11.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.11.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.11.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.12.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.12.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.12.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.12.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.12.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.12.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.12.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.12.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.12.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.12.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.12.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.12.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.13.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.13.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.13.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.13.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.13.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.13.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.13.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.13.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.13.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.13.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.13.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.13.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.14.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.14.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.14.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.14.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.14.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.14.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.14.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.14.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.14.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.14.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.14.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.14.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.15.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.15.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.15.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.15.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.15.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.15.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.15.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.15.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.15.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.15.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.15.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.15.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.16.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.16.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.16.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.16.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.16.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.16.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.16.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.16.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.16.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.16.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.16.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.16.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.17.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.17.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.17.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.17.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.17.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.17.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.17.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.17.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.17.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.17.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.17.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.17.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.18.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.18.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.18.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.18.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.18.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.18.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.18.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.18.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.18.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.18.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.18.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.18.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.19.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.19.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.19.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.19.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.19.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.19.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.19.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.19.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.19.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.19.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.19.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.19.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.20.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.20.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.20.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.20.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.20.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.20.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.20.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.20.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.20.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.20.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.20.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.20.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.21.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.21.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.21.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.21.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.21.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.21.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.21.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.21.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.21.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.21.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.21.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.21.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.22.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.22.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.22.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.22.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.22.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.22.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.22.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.22.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.22.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.22.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.22.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.22.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.23.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.23.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.23.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.23.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.23.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.23.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.23.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.23.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.23.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.23.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.23.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.23.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.pre_norm.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.pre_norm.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +neck.proj - torch.Size([1024, 768]): +The value is the same before and after calling `init_weights` of ImageClassifier + +head.fc.weight - torch.Size([2, 768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +head.fc.bias - torch.Size([2]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 +2023/06/06 00:56:57 - mmengine - WARNING - "FileClient" will be deprecated in future. Please use io functions in https://mmengine.readthedocs.io/en/latest/api/fileio.html#file-io +2023/06/06 00:56:57 - mmengine - WARNING - "HardDiskBackend" is the alias of "LocalBackend" and the former will be deprecated in future. +2023/06/06 00:56:57 - mmengine - INFO - Checkpoints will be saved to /mnt/petrelfs/luzeyu/workspace/fakebench/mmpretrain/workdir/clip_large_pretrain_4x256_all2_lr1e-4. +2023/06/06 00:58:07 - mmengine - INFO - Epoch(train) [1][ 100/4092] lr: 9.9999e-05 eta: 7:55:06 time: 0.6306 data_time: 0.0018 memory: 44139 loss: 0.5742 +2023/06/06 00:59:11 - mmengine - INFO - Epoch(train) [1][ 200/4092] lr: 9.9995e-05 eta: 7:32:09 time: 0.6331 data_time: 0.0018 memory: 44139 loss: 0.5510 +2023/06/06 01:00:14 - mmengine - INFO - Epoch(train) [1][ 300/4092] lr: 9.9988e-05 eta: 7:23:36 time: 0.6333 data_time: 0.0023 memory: 44139 loss: 0.5507 +2023/06/06 01:01:17 - mmengine - INFO - Epoch(train) [1][ 400/4092] lr: 9.9979e-05 eta: 7:18:35 time: 0.6325 data_time: 0.0017 memory: 44139 loss: 0.5323 +2023/06/06 01:02:21 - mmengine - INFO - Epoch(train) [1][ 500/4092] lr: 9.9967e-05 eta: 7:15:20 time: 0.6465 data_time: 0.0016 memory: 44139 loss: 0.5343 +2023/06/06 01:03:24 - mmengine - INFO - Epoch(train) [1][ 600/4092] lr: 9.9952e-05 eta: 7:13:00 time: 0.6330 data_time: 0.0014 memory: 44139 loss: 0.4938 +2023/06/06 01:04:28 - mmengine - INFO - Epoch(train) [1][ 700/4092] lr: 9.9935e-05 eta: 7:11:00 time: 0.6354 data_time: 0.0016 memory: 44139 loss: 0.5040 +2023/06/06 01:05:31 - mmengine - INFO - Epoch(train) [1][ 800/4092] lr: 9.9915e-05 eta: 7:09:08 time: 0.6346 data_time: 0.0016 memory: 44139 loss: 0.5007 +2023/06/06 01:06:34 - mmengine - INFO - Epoch(train) [1][ 900/4092] lr: 9.9893e-05 eta: 7:07:26 time: 0.6326 data_time: 0.0015 memory: 44139 loss: 0.4955 +2023/06/06 01:07:38 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_all2_lr1e-4_20230606_005614 +2023/06/06 01:07:38 - mmengine - INFO - Epoch(train) [1][1000/4092] lr: 9.9868e-05 eta: 7:05:51 time: 0.6321 data_time: 0.0015 memory: 44139 loss: 0.4976 +2023/06/06 01:08:41 - mmengine - INFO - Epoch(train) [1][1100/4092] lr: 9.9840e-05 eta: 7:04:24 time: 0.6333 data_time: 0.0014 memory: 44139 loss: 0.5405 +2023/06/06 01:09:44 - mmengine - INFO - Epoch(train) [1][1200/4092] lr: 9.9809e-05 eta: 7:03:03 time: 0.6363 data_time: 0.0015 memory: 44139 loss: 0.4754 +2023/06/06 01:10:48 - mmengine - INFO - Epoch(train) [1][1300/4092] lr: 9.9776e-05 eta: 7:01:44 time: 0.6345 data_time: 0.0014 memory: 44139 loss: 0.5028 +2023/06/06 01:11:51 - mmengine - INFO - Epoch(train) [1][1400/4092] lr: 9.9741e-05 eta: 7:00:27 time: 0.6315 data_time: 0.0016 memory: 44139 loss: 0.4924 +2023/06/06 01:12:55 - mmengine - INFO - Epoch(train) [1][1500/4092] lr: 9.9702e-05 eta: 6:59:10 time: 0.6319 data_time: 0.0016 memory: 44139 loss: 0.4951 +2023/06/06 01:13:58 - mmengine - INFO - Epoch(train) [1][1600/4092] lr: 9.9661e-05 eta: 6:57:53 time: 0.6314 data_time: 0.0014 memory: 44139 loss: 0.5086 +2023/06/06 01:15:01 - mmengine - INFO - Epoch(train) [1][1700/4092] lr: 9.9618e-05 eta: 6:56:38 time: 0.6313 data_time: 0.0013 memory: 44139 loss: 0.4961 +2023/06/06 01:16:04 - mmengine - INFO - Epoch(train) [1][1800/4092] lr: 9.9571e-05 eta: 6:55:23 time: 0.6315 data_time: 0.0015 memory: 44139 loss: 0.4864 +2023/06/06 01:17:07 - mmengine - INFO - Epoch(train) [1][1900/4092] lr: 9.9523e-05 eta: 6:54:08 time: 0.6309 data_time: 0.0014 memory: 44139 loss: 0.4863 +2023/06/06 01:18:11 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_all2_lr1e-4_20230606_005614 +2023/06/06 01:18:11 - mmengine - INFO - Epoch(train) [1][2000/4092] lr: 9.9471e-05 eta: 6:52:55 time: 0.6314 data_time: 0.0016 memory: 44139 loss: 0.4824 +2023/06/06 01:19:14 - mmengine - INFO - Epoch(train) [1][2100/4092] lr: 9.9417e-05 eta: 6:51:41 time: 0.6311 data_time: 0.0015 memory: 44139 loss: 0.5120 +2023/06/06 01:20:17 - mmengine - INFO - Epoch(train) [1][2200/4092] lr: 9.9360e-05 eta: 6:50:28 time: 0.6317 data_time: 0.0017 memory: 44139 loss: 0.4707 +2023/06/06 01:21:20 - mmengine - INFO - Epoch(train) [1][2300/4092] lr: 9.9301e-05 eta: 6:49:17 time: 0.6315 data_time: 0.0014 memory: 44139 loss: 0.4904 +2023/06/06 01:22:23 - mmengine - INFO - Epoch(train) [1][2400/4092] lr: 9.9239e-05 eta: 6:48:06 time: 0.6319 data_time: 0.0014 memory: 44139 loss: 0.4934 +2023/06/06 01:23:26 - mmengine - INFO - Epoch(train) [1][2500/4092] lr: 9.9174e-05 eta: 6:46:57 time: 0.6312 data_time: 0.0014 memory: 44139 loss: 0.5018 +2023/06/06 01:24:30 - mmengine - INFO - Epoch(train) [1][2600/4092] lr: 9.9107e-05 eta: 6:45:49 time: 0.6314 data_time: 0.0014 memory: 44139 loss: 0.4846 +2023/06/06 01:25:33 - mmengine - INFO - Epoch(train) [1][2700/4092] lr: 9.9037e-05 eta: 6:44:40 time: 0.6321 data_time: 0.0015 memory: 44139 loss: 0.4840 +2023/06/06 01:27:59 - mmengine - INFO - Epoch(train) [1][2800/4092] lr: 9.8965e-05 eta: 7:02:29 time: 0.6289 data_time: 0.0015 memory: 44139 loss: 0.4953 +2023/06/06 01:29:03 - mmengine - INFO - Epoch(train) [1][2900/4092] lr: 9.8890e-05 eta: 7:00:38 time: 0.6311 data_time: 0.0013 memory: 44139 loss: 0.5100 +2023/06/06 01:30:06 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_all2_lr1e-4_20230606_005614 +2023/06/06 01:30:06 - mmengine - INFO - Epoch(train) [1][3000/4092] lr: 9.8812e-05 eta: 6:58:52 time: 0.6315 data_time: 0.0014 memory: 44139 loss: 0.4527 +2023/06/06 01:31:09 - mmengine - INFO - Epoch(train) [1][3100/4092] lr: 9.8732e-05 eta: 6:57:07 time: 0.6318 data_time: 0.0014 memory: 44139 loss: 0.4773 +2023/06/06 01:32:12 - mmengine - INFO - Epoch(train) [1][3200/4092] lr: 9.8650e-05 eta: 6:55:26 time: 0.6315 data_time: 0.0014 memory: 44139 loss: 0.4770 +2023/06/06 01:33:15 - mmengine - INFO - Epoch(train) [1][3300/4092] lr: 9.8564e-05 eta: 6:53:47 time: 0.6312 data_time: 0.0016 memory: 44139 loss: 0.4589 +2023/06/06 01:34:19 - mmengine - INFO - Epoch(train) [1][3400/4092] lr: 9.8476e-05 eta: 6:52:10 time: 0.6317 data_time: 0.0015 memory: 44139 loss: 0.4754 +2023/06/06 01:35:22 - mmengine - INFO - Epoch(train) [1][3500/4092] lr: 9.8386e-05 eta: 6:50:35 time: 0.6314 data_time: 0.0015 memory: 44139 loss: 0.4615 +2023/06/06 01:36:25 - mmengine - INFO - Epoch(train) [1][3600/4092] lr: 9.8293e-05 eta: 6:49:02 time: 0.6320 data_time: 0.0014 memory: 44139 loss: 0.4558 +2023/06/06 01:37:28 - mmengine - INFO - Epoch(train) [1][3700/4092] lr: 9.8198e-05 eta: 6:47:30 time: 0.6316 data_time: 0.0014 memory: 44139 loss: 0.4740 +2023/06/06 01:38:31 - mmengine - INFO - Epoch(train) [1][3800/4092] lr: 9.8099e-05 eta: 6:46:00 time: 0.6321 data_time: 0.0014 memory: 44139 loss: 0.4786 +2023/06/06 01:39:35 - mmengine - INFO - Epoch(train) [1][3900/4092] lr: 9.7999e-05 eta: 6:44:31 time: 0.6321 data_time: 0.0014 memory: 44139 loss: 0.4451 +2023/06/06 01:40:38 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_all2_lr1e-4_20230606_005614 +2023/06/06 01:40:38 - mmengine - INFO - Epoch(train) [1][4000/4092] lr: 9.7896e-05 eta: 6:43:04 time: 0.6311 data_time: 0.0013 memory: 44139 loss: 0.4625 +2023/06/06 01:41:36 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_all2_lr1e-4_20230606_005614 +2023/06/06 01:41:36 - mmengine - INFO - Saving checkpoint at 1 epochs +2023/06/06 01:43:59 - mmengine - INFO - Epoch(val) [1][100/119] eta: 0:00:25 time: 1.2878 data_time: 0.0007 memory: 44139 +2023/06/06 01:44:42 - mmengine - INFO - Epoch(val) [1][119/119] accuracy/top1: 71.7806 single-label/precision_classwise: [93.49911499023438, 41.82777404785156] single-label/recall_classwise: [68.91189575195312, 82.34883880615234] single-label/f1-score_classwise: [79.34439086914062, 55.4769287109375] data_time: 0.0222 time: 1.3161 +2023/06/06 01:45:49 - mmengine - INFO - Epoch(train) [2][ 100/4092] lr: 9.7691e-05 eta: 6:40:48 time: 0.6318 data_time: 0.0019 memory: 44140 loss: 0.4436 +2023/06/06 01:46:53 - mmengine - INFO - Epoch(train) [2][ 200/4092] lr: 9.7580e-05 eta: 6:39:23 time: 0.6318 data_time: 0.0014 memory: 44140 loss: 0.4713 +2023/06/06 01:47:56 - mmengine - INFO - Epoch(train) [2][ 300/4092] lr: 9.7467e-05 eta: 6:38:00 time: 0.6340 data_time: 0.0014 memory: 44140 loss: 0.4740 +2023/06/06 01:48:59 - mmengine - INFO - Epoch(train) [2][ 400/4092] lr: 9.7352e-05 eta: 6:36:38 time: 0.6323 data_time: 0.0014 memory: 44140 loss: 0.4599 +2023/06/06 01:50:03 - mmengine - INFO - Epoch(train) [2][ 500/4092] lr: 9.7234e-05 eta: 6:35:18 time: 0.6346 data_time: 0.0014 memory: 44140 loss: 0.4425 +2023/06/06 01:51:06 - mmengine - INFO - Epoch(train) [2][ 600/4092] lr: 9.7113e-05 eta: 6:33:57 time: 0.6318 data_time: 0.0014 memory: 44140 loss: 0.4755 +2023/06/06 01:52:09 - mmengine - INFO - Epoch(train) [2][ 700/4092] lr: 9.6990e-05 eta: 6:32:36 time: 0.6312 data_time: 0.0014 memory: 44140 loss: 0.4483 +2023/06/06 01:53:12 - mmengine - INFO - Epoch(train) [2][ 800/4092] lr: 9.6865e-05 eta: 6:31:16 time: 0.6319 data_time: 0.0014 memory: 44140 loss: 0.4435 +2023/06/06 01:54:15 - mmengine - INFO - Epoch(train) [2][ 900/4092] lr: 9.6737e-05 eta: 6:29:57 time: 0.6316 data_time: 0.0013 memory: 44140 loss: 0.4590 +2023/06/06 01:54:20 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_all2_lr1e-4_20230606_005614 +2023/06/06 01:55:19 - mmengine - INFO - Epoch(train) [2][1000/4092] lr: 9.6606e-05 eta: 6:28:39 time: 0.6326 data_time: 0.0013 memory: 44140 loss: 0.4541 +2023/06/06 01:56:22 - mmengine - INFO - Epoch(train) [2][1100/4092] lr: 9.6473e-05 eta: 6:27:21 time: 0.6318 data_time: 0.0015 memory: 44140 loss: 0.4619 +2023/06/06 01:57:25 - mmengine - INFO - Epoch(train) [2][1200/4092] lr: 9.6338e-05 eta: 6:26:03 time: 0.6313 data_time: 0.0013 memory: 44140 loss: 0.4453 +2023/06/06 01:58:28 - mmengine - INFO - Epoch(train) [2][1300/4092] lr: 9.6200e-05 eta: 6:24:46 time: 0.6314 data_time: 0.0016 memory: 44140 loss: 0.4572 +2023/06/06 01:59:31 - mmengine - INFO - Epoch(train) [2][1400/4092] lr: 9.6060e-05 eta: 6:23:29 time: 0.6333 data_time: 0.0015 memory: 44140 loss: 0.4921 +2023/06/06 02:00:35 - mmengine - INFO - Epoch(train) [2][1500/4092] lr: 9.5918e-05 eta: 6:22:13 time: 0.6329 data_time: 0.0017 memory: 44140 loss: 0.4525 +2023/06/06 02:01:38 - mmengine - INFO - Epoch(train) [2][1600/4092] lr: 9.5773e-05 eta: 6:20:58 time: 0.6310 data_time: 0.0015 memory: 44140 loss: 0.4643 +2023/06/06 02:02:41 - mmengine - INFO - Epoch(train) [2][1700/4092] lr: 9.5625e-05 eta: 6:19:43 time: 0.6319 data_time: 0.0015 memory: 44140 loss: 0.4222 +2023/06/06 02:03:44 - mmengine - INFO - Epoch(train) [2][1800/4092] lr: 9.5475e-05 eta: 6:18:28 time: 0.6319 data_time: 0.0014 memory: 44140 loss: 0.4527 +2023/06/06 02:04:47 - mmengine - INFO - Epoch(train) [2][1900/4092] lr: 9.5323e-05 eta: 6:17:14 time: 0.6303 data_time: 0.0016 memory: 44140 loss: 0.4463 +2023/06/06 02:04:53 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_all2_lr1e-4_20230606_005614 +2023/06/06 02:05:51 - mmengine - INFO - Epoch(train) [2][2000/4092] lr: 9.5169e-05 eta: 6:16:00 time: 0.6329 data_time: 0.0015 memory: 44140 loss: 0.4297 +2023/06/06 02:06:54 - mmengine - INFO - Epoch(train) [2][2100/4092] lr: 9.5012e-05 eta: 6:14:46 time: 0.6331 data_time: 0.0016 memory: 44140 loss: 0.4281 +2023/06/06 02:07:57 - mmengine - INFO - Epoch(train) [2][2200/4092] lr: 9.4853e-05 eta: 6:13:33 time: 0.6313 data_time: 0.0014 memory: 44140 loss: 0.4476 +2023/06/06 02:09:00 - mmengine - INFO - Epoch(train) [2][2300/4092] lr: 9.4691e-05 eta: 6:12:20 time: 0.6314 data_time: 0.0014 memory: 44140 loss: 0.4375 +2023/06/06 02:10:04 - mmengine - INFO - Epoch(train) [2][2400/4092] lr: 9.4527e-05 eta: 6:11:07 time: 0.6311 data_time: 0.0015 memory: 44140 loss: 0.4416 +2023/06/06 02:11:07 - mmengine - INFO - Epoch(train) [2][2500/4092] lr: 9.4361e-05 eta: 6:09:55 time: 0.6314 data_time: 0.0014 memory: 44140 loss: 0.4352 +2023/06/06 02:12:10 - mmengine - INFO - Epoch(train) [2][2600/4092] lr: 9.4192e-05 eta: 6:08:43 time: 0.6314 data_time: 0.0014 memory: 44140 loss: 0.4278 +2023/06/06 02:13:13 - mmengine - INFO - Epoch(train) [2][2700/4092] lr: 9.4021e-05 eta: 6:07:31 time: 0.6318 data_time: 0.0014 memory: 44140 loss: 0.4350 +2023/06/06 02:14:16 - mmengine - INFO - Epoch(train) [2][2800/4092] lr: 9.3848e-05 eta: 6:06:20 time: 0.6315 data_time: 0.0014 memory: 44140 loss: 0.4279 +2023/06/06 02:15:20 - mmengine - INFO - Epoch(train) [2][2900/4092] lr: 9.3672e-05 eta: 6:05:08 time: 0.6311 data_time: 0.0014 memory: 44140 loss: 0.4438 +2023/06/06 02:15:25 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_all2_lr1e-4_20230606_005614 +2023/06/06 02:16:23 - mmengine - INFO - Epoch(train) [2][3000/4092] lr: 9.3495e-05 eta: 6:03:58 time: 0.6320 data_time: 0.0014 memory: 44140 loss: 0.4462 +2023/06/06 02:17:26 - mmengine - INFO - Epoch(train) [2][3100/4092] lr: 9.3315e-05 eta: 6:02:47 time: 0.6315 data_time: 0.0014 memory: 44140 loss: 0.4394 +2023/06/06 02:18:29 - mmengine - INFO - Epoch(train) [2][3200/4092] lr: 9.3132e-05 eta: 6:01:36 time: 0.6344 data_time: 0.0014 memory: 44140 loss: 0.4422 +2023/06/06 02:19:33 - mmengine - INFO - Epoch(train) [2][3300/4092] lr: 9.2948e-05 eta: 6:00:26 time: 0.6315 data_time: 0.0014 memory: 44140 loss: 0.4563 +2023/06/06 02:20:36 - mmengine - INFO - Epoch(train) [2][3400/4092] lr: 9.2761e-05 eta: 5:59:16 time: 0.6312 data_time: 0.0014 memory: 44140 loss: 0.4429 +2023/06/06 02:21:39 - mmengine - INFO - Epoch(train) [2][3500/4092] lr: 9.2572e-05 eta: 5:58:06 time: 0.6410 data_time: 0.0016 memory: 44140 loss: 0.4485 +2023/06/06 02:22:42 - mmengine - INFO - Epoch(train) [2][3600/4092] lr: 9.2381e-05 eta: 5:56:56 time: 0.6309 data_time: 0.0015 memory: 44140 loss: 0.4140 +2023/06/06 02:23:45 - mmengine - INFO - Epoch(train) [2][3700/4092] lr: 9.2187e-05 eta: 5:55:46 time: 0.6325 data_time: 0.0015 memory: 44140 loss: 0.4222 +2023/06/06 02:24:49 - mmengine - INFO - Epoch(train) [2][3800/4092] lr: 9.1991e-05 eta: 5:54:36 time: 0.6321 data_time: 0.0014 memory: 44140 loss: 0.4607 +2023/06/06 02:25:52 - mmengine - INFO - Epoch(train) [2][3900/4092] lr: 9.1794e-05 eta: 5:53:27 time: 0.6330 data_time: 0.0014 memory: 44140 loss: 0.4672 +2023/06/06 02:25:57 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_all2_lr1e-4_20230606_005614 +2023/06/06 02:26:55 - mmengine - INFO - Epoch(train) [2][4000/4092] lr: 9.1594e-05 eta: 5:52:18 time: 0.6318 data_time: 0.0015 memory: 44140 loss: 0.4245 +2023/06/06 02:27:53 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_all2_lr1e-4_20230606_005614 +2023/06/06 02:27:53 - mmengine - INFO - Saving checkpoint at 2 epochs +2023/06/06 02:30:13 - mmengine - INFO - Epoch(val) [2][100/119] eta: 0:00:24 time: 1.2874 data_time: 0.0007 memory: 44140 +2023/06/06 02:30:56 - mmengine - INFO - Epoch(val) [2][119/119] accuracy/top1: 72.9457 single-label/precision_classwise: [96.08561706542969, 43.52043151855469] single-label/recall_classwise: [68.387939453125, 89.73643493652344] single-label/f1-score_classwise: [79.90460205078125, 58.614139556884766] data_time: 0.0177 time: 1.2847 +2023/06/06 02:32:03 - mmengine - INFO - Epoch(train) [3][ 100/4092] lr: 9.1204e-05 eta: 5:50:17 time: 0.6333 data_time: 0.0015 memory: 44140 loss: 0.4474 +2023/06/06 02:33:06 - mmengine - INFO - Epoch(train) [3][ 200/4092] lr: 9.0997e-05 eta: 5:49:09 time: 0.6336 data_time: 0.0016 memory: 44140 loss: 0.4212 +2023/06/06 02:34:09 - mmengine - INFO - Epoch(train) [3][ 300/4092] lr: 9.0789e-05 eta: 5:48:00 time: 0.6341 data_time: 0.0014 memory: 44140 loss: 0.4284 +2023/06/06 02:35:13 - mmengine - INFO - Epoch(train) [3][ 400/4092] lr: 9.0579e-05 eta: 5:46:52 time: 0.6312 data_time: 0.0014 memory: 44140 loss: 0.4249 +2023/06/06 02:36:16 - mmengine - INFO - Epoch(train) [3][ 500/4092] lr: 9.0366e-05 eta: 5:45:44 time: 0.6338 data_time: 0.0014 memory: 44140 loss: 0.4253 +2023/06/06 02:37:19 - mmengine - INFO - Epoch(train) [3][ 600/4092] lr: 9.0151e-05 eta: 5:44:36 time: 0.6319 data_time: 0.0014 memory: 44140 loss: 0.4409 +2023/06/06 02:38:23 - mmengine - INFO - Epoch(train) [3][ 700/4092] lr: 8.9935e-05 eta: 5:43:28 time: 0.6324 data_time: 0.0014 memory: 44140 loss: 0.4236 +2023/06/06 02:39:26 - mmengine - INFO - Epoch(train) [3][ 800/4092] lr: 8.9716e-05 eta: 5:42:20 time: 0.6361 data_time: 0.0014 memory: 44140 loss: 0.4090 +2023/06/06 02:39:36 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_all2_lr1e-4_20230606_005614 +2023/06/06 02:40:29 - mmengine - INFO - Epoch(train) [3][ 900/4092] lr: 8.9495e-05 eta: 5:41:13 time: 0.6332 data_time: 0.0016 memory: 44140 loss: 0.4282 +2023/06/06 02:41:33 - mmengine - INFO - Epoch(train) [3][1000/4092] lr: 8.9272e-05 eta: 5:40:05 time: 0.6346 data_time: 0.0016 memory: 44140 loss: 0.4246 +2023/06/06 02:42:36 - mmengine - INFO - Epoch(train) [3][1100/4092] lr: 8.9047e-05 eta: 5:38:58 time: 0.6333 data_time: 0.0014 memory: 44140 loss: 0.4437 +2023/06/06 02:43:40 - mmengine - INFO - Epoch(train) [3][1200/4092] lr: 8.8820e-05 eta: 5:37:51 time: 0.6322 data_time: 0.0013 memory: 44140 loss: 0.4244 +2023/06/06 02:44:43 - mmengine - INFO - Epoch(train) [3][1300/4092] lr: 8.8591e-05 eta: 5:36:44 time: 0.6338 data_time: 0.0014 memory: 44140 loss: 0.4091 +2023/06/06 02:45:46 - mmengine - INFO - Epoch(train) [3][1400/4092] lr: 8.8360e-05 eta: 5:35:36 time: 0.6321 data_time: 0.0014 memory: 44140 loss: 0.4393 +2023/06/06 02:46:50 - mmengine - INFO - Epoch(train) [3][1500/4092] lr: 8.8128e-05 eta: 5:34:29 time: 0.6322 data_time: 0.0013 memory: 44140 loss: 0.4247 +2023/06/06 02:47:53 - mmengine - INFO - Epoch(train) [3][1600/4092] lr: 8.7893e-05 eta: 5:33:22 time: 0.6324 data_time: 0.0016 memory: 44140 loss: 0.4355 +2023/06/06 02:48:57 - mmengine - INFO - Epoch(train) [3][1700/4092] lr: 8.7656e-05 eta: 5:32:15 time: 0.6409 data_time: 0.0016 memory: 44140 loss: 0.4343 +2023/06/06 02:50:00 - mmengine - INFO - Epoch(train) [3][1800/4092] lr: 8.7417e-05 eta: 5:31:08 time: 0.6323 data_time: 0.0015 memory: 44140 loss: 0.4275 +2023/06/06 02:50:10 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_all2_lr1e-4_20230606_005614 +2023/06/06 02:51:03 - mmengine - INFO - Epoch(train) [3][1900/4092] lr: 8.7177e-05 eta: 5:30:01 time: 0.6326 data_time: 0.0017 memory: 44140 loss: 0.4183 +2023/06/06 02:52:06 - mmengine - INFO - Epoch(train) [3][2000/4092] lr: 8.6934e-05 eta: 5:28:53 time: 0.6323 data_time: 0.0015 memory: 44140 loss: 0.4195 +2023/06/06 02:53:10 - mmengine - INFO - Epoch(train) [3][2100/4092] lr: 8.6690e-05 eta: 5:27:46 time: 0.6316 data_time: 0.0015 memory: 44140 loss: 0.4253 +2023/06/06 02:54:13 - mmengine - INFO - Epoch(train) [3][2200/4092] lr: 8.6444e-05 eta: 5:26:39 time: 0.6323 data_time: 0.0016 memory: 44140 loss: 0.4027 +2023/06/06 02:55:16 - mmengine - INFO - Epoch(train) [3][2300/4092] lr: 8.6196e-05 eta: 5:25:33 time: 0.6324 data_time: 0.0014 memory: 44140 loss: 0.4506 +2023/06/06 02:56:20 - mmengine - INFO - Epoch(train) [3][2400/4092] lr: 8.5946e-05 eta: 5:24:26 time: 0.6322 data_time: 0.0014 memory: 44140 loss: 0.4363 +2023/06/06 02:57:23 - mmengine - INFO - Epoch(train) [3][2500/4092] lr: 8.5694e-05 eta: 5:23:20 time: 0.6337 data_time: 0.0016 memory: 44140 loss: 0.4133 +2023/06/06 02:58:26 - mmengine - INFO - Epoch(train) [3][2600/4092] lr: 8.5441e-05 eta: 5:22:13 time: 0.6330 data_time: 0.0015 memory: 44140 loss: 0.4413 +2023/06/06 02:59:30 - mmengine - INFO - Epoch(train) [3][2700/4092] lr: 8.5185e-05 eta: 5:21:07 time: 0.6332 data_time: 0.0014 memory: 44140 loss: 0.4279 +2023/06/06 03:00:33 - mmengine - INFO - Epoch(train) [3][2800/4092] lr: 8.4928e-05 eta: 5:20:00 time: 0.6331 data_time: 0.0014 memory: 44140 loss: 0.4370 +2023/06/06 03:00:43 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_all2_lr1e-4_20230606_005614 +2023/06/06 03:01:36 - mmengine - INFO - Epoch(train) [3][2900/4092] lr: 8.4669e-05 eta: 5:18:55 time: 0.6357 data_time: 0.0013 memory: 44140 loss: 0.4369 +2023/06/06 03:02:40 - mmengine - INFO - Epoch(train) [3][3000/4092] lr: 8.4409e-05 eta: 5:17:48 time: 0.6355 data_time: 0.0017 memory: 44140 loss: 0.4133 +2023/06/06 03:03:43 - mmengine - INFO - Epoch(train) [3][3100/4092] lr: 8.4146e-05 eta: 5:16:42 time: 0.6321 data_time: 0.0016 memory: 44140 loss: 0.4270 +2023/06/06 03:04:46 - mmengine - INFO - Epoch(train) [3][3200/4092] lr: 8.3882e-05 eta: 5:15:35 time: 0.6318 data_time: 0.0014 memory: 44140 loss: 0.4333 +2023/06/06 03:05:49 - mmengine - INFO - Epoch(train) [3][3300/4092] lr: 8.3616e-05 eta: 5:14:29 time: 0.6341 data_time: 0.0018 memory: 44140 loss: 0.4250 +2023/06/06 03:06:53 - mmengine - INFO - Epoch(train) [3][3400/4092] lr: 8.3349e-05 eta: 5:13:23 time: 0.6464 data_time: 0.0022 memory: 44140 loss: 0.4533 +2023/06/06 03:07:56 - mmengine - INFO - Epoch(train) [3][3500/4092] lr: 8.3080e-05 eta: 5:12:17 time: 0.6343 data_time: 0.0023 memory: 44140 loss: 0.4503 +2023/06/06 03:09:00 - mmengine - INFO - Epoch(train) [3][3600/4092] lr: 8.2809e-05 eta: 5:11:11 time: 0.6343 data_time: 0.0025 memory: 44140 loss: 0.4467 +2023/06/06 03:10:03 - mmengine - INFO - Epoch(train) [3][3700/4092] lr: 8.2537e-05 eta: 5:10:05 time: 0.6328 data_time: 0.0017 memory: 44140 loss: 0.4151 +2023/06/06 03:11:06 - mmengine - INFO - Epoch(train) [3][3800/4092] lr: 8.2263e-05 eta: 5:08:59 time: 0.6329 data_time: 0.0020 memory: 44140 loss: 0.4179 +2023/06/06 03:11:16 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_all2_lr1e-4_20230606_005614 +2023/06/06 03:12:10 - mmengine - INFO - Epoch(train) [3][3900/4092] lr: 8.1987e-05 eta: 5:07:54 time: 0.6335 data_time: 0.0024 memory: 44140 loss: 0.4292 +2023/06/06 03:13:13 - mmengine - INFO - Epoch(train) [3][4000/4092] lr: 8.1710e-05 eta: 5:06:48 time: 0.6408 data_time: 0.0023 memory: 44140 loss: 0.4231 +2023/06/06 03:14:11 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_all2_lr1e-4_20230606_005614 +2023/06/06 03:14:11 - mmengine - INFO - Saving checkpoint at 3 epochs +2023/06/06 03:16:33 - mmengine - INFO - Epoch(val) [3][100/119] eta: 0:00:24 time: 1.2880 data_time: 0.0010 memory: 44140 +2023/06/06 03:17:16 - mmengine - INFO - Epoch(val) [3][119/119] accuracy/top1: 75.0476 single-label/precision_classwise: [96.72523498535156, 45.76508712768555] single-label/recall_classwise: [70.66683959960938, 91.18605041503906] single-label/f1-score_classwise: [81.6677474975586, 60.943450927734375] data_time: 0.0200 time: 1.2880 +2023/06/06 03:18:23 - mmengine - INFO - Epoch(train) [4][ 100/4092] lr: 8.1173e-05 eta: 5:04:49 time: 0.6323 data_time: 0.0025 memory: 44140 loss: 0.4168 +2023/06/06 03:19:26 - mmengine - INFO - Epoch(train) [4][ 200/4092] lr: 8.0891e-05 eta: 5:03:44 time: 0.6332 data_time: 0.0021 memory: 44140 loss: 0.4294 +2023/06/06 03:20:29 - mmengine - INFO - Epoch(train) [4][ 300/4092] lr: 8.0608e-05 eta: 5:02:38 time: 0.6329 data_time: 0.0024 memory: 44140 loss: 0.4202 +2023/06/06 03:21:32 - mmengine - INFO - Epoch(train) [4][ 400/4092] lr: 8.0323e-05 eta: 5:01:32 time: 0.6330 data_time: 0.0025 memory: 44140 loss: 0.3883 +2023/06/06 03:22:36 - mmengine - INFO - Epoch(train) [4][ 500/4092] lr: 8.0037e-05 eta: 5:00:27 time: 0.6332 data_time: 0.0024 memory: 44140 loss: 0.4146 +2023/06/06 03:23:39 - mmengine - INFO - Epoch(train) [4][ 600/4092] lr: 7.9749e-05 eta: 4:59:21 time: 0.6315 data_time: 0.0022 memory: 44140 loss: 0.4219 +2023/06/06 03:24:43 - mmengine - INFO - Epoch(train) [4][ 700/4092] lr: 7.9459e-05 eta: 4:58:16 time: 0.6348 data_time: 0.0021 memory: 44140 loss: 0.4341 +2023/06/06 03:24:58 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_all2_lr1e-4_20230606_005614 +2023/06/06 03:25:46 - mmengine - INFO - Epoch(train) [4][ 800/4092] lr: 7.9169e-05 eta: 4:57:10 time: 0.6324 data_time: 0.0020 memory: 44140 loss: 0.4551 +2023/06/06 03:26:49 - mmengine - INFO - Epoch(train) [4][ 900/4092] lr: 7.8877e-05 eta: 4:56:04 time: 0.6329 data_time: 0.0018 memory: 44140 loss: 0.4364 +2023/06/06 03:27:53 - mmengine - INFO - Epoch(train) [4][1000/4092] lr: 7.8583e-05 eta: 4:54:59 time: 0.6334 data_time: 0.0019 memory: 44140 loss: 0.4265 +2023/06/06 03:28:56 - mmengine - INFO - Epoch(train) [4][1100/4092] lr: 7.8288e-05 eta: 4:53:53 time: 0.6325 data_time: 0.0022 memory: 44140 loss: 0.4444 +2023/06/06 03:29:59 - mmengine - INFO - Epoch(train) [4][1200/4092] lr: 7.7992e-05 eta: 4:52:48 time: 0.6307 data_time: 0.0021 memory: 44140 loss: 0.4278 +2023/06/06 03:31:03 - mmengine - INFO - Epoch(train) [4][1300/4092] lr: 7.7694e-05 eta: 4:51:43 time: 0.6340 data_time: 0.0019 memory: 44140 loss: 0.4094 +2023/06/06 03:32:06 - mmengine - INFO - Epoch(train) [4][1400/4092] lr: 7.7395e-05 eta: 4:50:38 time: 0.6329 data_time: 0.0024 memory: 44140 loss: 0.4303 +2023/06/06 03:33:10 - mmengine - INFO - Epoch(train) [4][1500/4092] lr: 7.7095e-05 eta: 4:49:32 time: 0.6322 data_time: 0.0020 memory: 44140 loss: 0.4366 +2023/06/06 03:34:13 - mmengine - INFO - Epoch(train) [4][1600/4092] lr: 7.6793e-05 eta: 4:48:27 time: 0.6337 data_time: 0.0027 memory: 44140 loss: 0.4132 +2023/06/06 03:35:16 - mmengine - INFO - Epoch(train) [4][1700/4092] lr: 7.6490e-05 eta: 4:47:22 time: 0.6324 data_time: 0.0019 memory: 44140 loss: 0.4383 +2023/06/06 03:35:31 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_all2_lr1e-4_20230606_005614 +2023/06/06 03:36:20 - mmengine - INFO - Epoch(train) [4][1800/4092] lr: 7.6186e-05 eta: 4:46:17 time: 0.6326 data_time: 0.0023 memory: 44140 loss: 0.4211 +2023/06/06 03:37:23 - mmengine - INFO - Epoch(train) [4][1900/4092] lr: 7.5881e-05 eta: 4:45:12 time: 0.6322 data_time: 0.0024 memory: 44140 loss: 0.4364 +2023/06/06 03:38:27 - mmengine - INFO - Epoch(train) [4][2000/4092] lr: 7.5574e-05 eta: 4:44:07 time: 0.6362 data_time: 0.0024 memory: 44140 loss: 0.4302 +2023/06/06 03:39:30 - mmengine - INFO - Epoch(train) [4][2100/4092] lr: 7.5266e-05 eta: 4:43:01 time: 0.6330 data_time: 0.0019 memory: 44140 loss: 0.4075 +2023/06/06 03:40:33 - mmengine - INFO - Epoch(train) [4][2200/4092] lr: 7.4957e-05 eta: 4:41:56 time: 0.6322 data_time: 0.0024 memory: 44140 loss: 0.4017 +2023/06/06 03:41:36 - mmengine - INFO - Epoch(train) [4][2300/4092] lr: 7.4647e-05 eta: 4:40:51 time: 0.6337 data_time: 0.0032 memory: 44140 loss: 0.4278 +2023/06/06 03:42:40 - mmengine - INFO - Epoch(train) [4][2400/4092] lr: 7.4336e-05 eta: 4:39:46 time: 0.6332 data_time: 0.0020 memory: 44140 loss: 0.4448 +2023/06/06 03:43:43 - mmengine - INFO - Epoch(train) [4][2500/4092] lr: 7.4023e-05 eta: 4:38:40 time: 0.6318 data_time: 0.0019 memory: 44140 loss: 0.4192 +2023/06/06 03:44:46 - mmengine - INFO - Epoch(train) [4][2600/4092] lr: 7.3709e-05 eta: 4:37:35 time: 0.6316 data_time: 0.0019 memory: 44140 loss: 0.4232 +2023/06/06 03:45:50 - mmengine - INFO - Epoch(train) [4][2700/4092] lr: 7.3395e-05 eta: 4:36:30 time: 0.6323 data_time: 0.0023 memory: 44140 loss: 0.4666 +2023/06/06 03:46:05 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_all2_lr1e-4_20230606_005614 +2023/06/06 03:46:53 - mmengine - INFO - Epoch(train) [4][2800/4092] lr: 7.3079e-05 eta: 4:35:25 time: 0.6315 data_time: 0.0021 memory: 44140 loss: 0.4382 +2023/06/06 03:47:56 - mmengine - INFO - Epoch(train) [4][2900/4092] lr: 7.2762e-05 eta: 4:34:21 time: 0.6321 data_time: 0.0024 memory: 44140 loss: 0.3834 +2023/06/06 03:49:00 - mmengine - INFO - Epoch(train) [4][3000/4092] lr: 7.2444e-05 eta: 4:33:16 time: 0.6340 data_time: 0.0031 memory: 44140 loss: 0.4198 +2023/06/06 03:50:03 - mmengine - INFO - Epoch(train) [4][3100/4092] lr: 7.2125e-05 eta: 4:32:11 time: 0.6334 data_time: 0.0026 memory: 44140 loss: 0.4397 +2023/06/06 03:51:07 - mmengine - INFO - Epoch(train) [4][3200/4092] lr: 7.1805e-05 eta: 4:31:06 time: 0.6322 data_time: 0.0020 memory: 44140 loss: 0.4347 +2023/06/06 03:52:10 - mmengine - INFO - Epoch(train) [4][3300/4092] lr: 7.1484e-05 eta: 4:30:01 time: 0.6348 data_time: 0.0026 memory: 44140 loss: 0.4178 +2023/06/06 03:53:13 - mmengine - INFO - Epoch(train) [4][3400/4092] lr: 7.1162e-05 eta: 4:28:56 time: 0.6323 data_time: 0.0019 memory: 44140 loss: 0.4433 +2023/06/06 03:54:17 - mmengine - INFO - Epoch(train) [4][3500/4092] lr: 7.0839e-05 eta: 4:27:51 time: 0.6344 data_time: 0.0024 memory: 44140 loss: 0.4336 +2023/06/06 03:55:20 - mmengine - INFO - Epoch(train) [4][3600/4092] lr: 7.0515e-05 eta: 4:26:46 time: 0.6330 data_time: 0.0022 memory: 44140 loss: 0.4535 +2023/06/06 03:56:23 - mmengine - INFO - Epoch(train) [4][3700/4092] lr: 7.0191e-05 eta: 4:25:42 time: 0.6319 data_time: 0.0023 memory: 44140 loss: 0.4276 +2023/06/06 03:56:39 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_all2_lr1e-4_20230606_005614 +2023/06/06 03:57:27 - mmengine - INFO - Epoch(train) [4][3800/4092] lr: 6.9865e-05 eta: 4:24:37 time: 0.6330 data_time: 0.0022 memory: 44140 loss: 0.4230 +2023/06/06 03:58:30 - mmengine - INFO - Epoch(train) [4][3900/4092] lr: 6.9538e-05 eta: 4:23:32 time: 0.6325 data_time: 0.0019 memory: 44140 loss: 0.4415 +2023/06/06 03:59:33 - mmengine - INFO - Epoch(train) [4][4000/4092] lr: 6.9211e-05 eta: 4:22:27 time: 0.6328 data_time: 0.0024 memory: 44140 loss: 0.4170 +2023/06/06 04:00:31 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_all2_lr1e-4_20230606_005614 +2023/06/06 04:00:31 - mmengine - INFO - Saving checkpoint at 4 epochs +2023/06/06 04:02:53 - mmengine - INFO - Epoch(val) [4][100/119] eta: 0:00:24 time: 1.2892 data_time: 0.0009 memory: 44140 +2023/06/06 04:03:36 - mmengine - INFO - Epoch(val) [4][119/119] accuracy/top1: 77.1925 single-label/precision_classwise: [96.76256561279297, 48.19059371948242] single-label/recall_classwise: [73.45916748046875, 90.94573974609375] single-label/f1-score_classwise: [83.5157470703125, 62.99905776977539] data_time: 0.0182 time: 1.2875 +2023/06/06 04:04:43 - mmengine - INFO - Epoch(train) [5][ 100/4092] lr: 6.8580e-05 eta: 4:20:27 time: 0.6321 data_time: 0.0017 memory: 44140 loss: 0.4497 +2023/06/06 04:05:46 - mmengine - INFO - Epoch(train) [5][ 200/4092] lr: 6.8250e-05 eta: 4:19:22 time: 0.6332 data_time: 0.0017 memory: 44140 loss: 0.4209 +2023/06/06 04:06:49 - mmengine - INFO - Epoch(train) [5][ 300/4092] lr: 6.7920e-05 eta: 4:18:18 time: 0.6322 data_time: 0.0021 memory: 44140 loss: 0.4413 +2023/06/06 04:07:53 - mmengine - INFO - Epoch(train) [5][ 400/4092] lr: 6.7588e-05 eta: 4:17:13 time: 0.6328 data_time: 0.0022 memory: 44140 loss: 0.4384 +2023/06/06 04:08:56 - mmengine - INFO - Epoch(train) [5][ 500/4092] lr: 6.7256e-05 eta: 4:16:08 time: 0.6331 data_time: 0.0025 memory: 44140 loss: 0.4339 +2023/06/06 04:09:59 - mmengine - INFO - Epoch(train) [5][ 600/4092] lr: 6.6924e-05 eta: 4:15:04 time: 0.6321 data_time: 0.0015 memory: 44140 loss: 0.4453 +2023/06/06 04:10:19 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_all2_lr1e-4_20230606_005614 +2023/06/06 04:11:02 - mmengine - INFO - Epoch(train) [5][ 700/4092] lr: 6.6590e-05 eta: 4:13:59 time: 0.6323 data_time: 0.0017 memory: 44140 loss: 0.4455 +2023/06/06 04:12:06 - mmengine - INFO - Epoch(train) [5][ 800/4092] lr: 6.6256e-05 eta: 4:12:54 time: 0.6320 data_time: 0.0015 memory: 44140 loss: 0.4187 +2023/06/06 04:13:09 - mmengine - INFO - Epoch(train) [5][ 900/4092] lr: 6.5921e-05 eta: 4:11:49 time: 0.6345 data_time: 0.0015 memory: 44140 loss: 0.4166 +2023/06/06 04:14:12 - mmengine - INFO - Epoch(train) [5][1000/4092] lr: 6.5586e-05 eta: 4:10:45 time: 0.6343 data_time: 0.0024 memory: 44140 loss: 0.4171 +2023/06/06 04:15:16 - mmengine - INFO - Epoch(train) [5][1100/4092] lr: 6.5250e-05 eta: 4:09:40 time: 0.6327 data_time: 0.0015 memory: 44140 loss: 0.4422 +2023/06/06 04:16:19 - mmengine - INFO - Epoch(train) [5][1200/4092] lr: 6.4913e-05 eta: 4:08:36 time: 0.6323 data_time: 0.0014 memory: 44140 loss: 0.4665 +2023/06/06 04:17:23 - mmengine - INFO - Epoch(train) [5][1300/4092] lr: 6.4576e-05 eta: 4:07:31 time: 0.6342 data_time: 0.0018 memory: 44140 loss: 0.4295 +2023/06/06 04:18:26 - mmengine - INFO - Epoch(train) [5][1400/4092] lr: 6.4238e-05 eta: 4:06:27 time: 0.6336 data_time: 0.0015 memory: 44140 loss: 0.4353 +2023/06/06 04:19:29 - mmengine - INFO - Epoch(train) [5][1500/4092] lr: 6.3899e-05 eta: 4:05:22 time: 0.6321 data_time: 0.0017 memory: 44140 loss: 0.4146 +2023/06/06 04:20:33 - mmengine - INFO - Epoch(train) [5][1600/4092] lr: 6.3560e-05 eta: 4:04:17 time: 0.6318 data_time: 0.0015 memory: 44140 loss: 0.4556 +2023/06/06 04:20:53 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_all2_lr1e-4_20230606_005614 +2023/06/06 04:21:36 - mmengine - INFO - Epoch(train) [5][1700/4092] lr: 6.3221e-05 eta: 4:03:13 time: 0.6337 data_time: 0.0015 memory: 44140 loss: 0.4441 +2023/06/06 04:22:39 - mmengine - INFO - Epoch(train) [5][1800/4092] lr: 6.2881e-05 eta: 4:02:08 time: 0.6334 data_time: 0.0015 memory: 44140 loss: 0.4030 +2023/06/06 04:23:43 - mmengine - INFO - Epoch(train) [5][1900/4092] lr: 6.2541e-05 eta: 4:01:04 time: 0.6325 data_time: 0.0015 memory: 44140 loss: 0.4508 +2023/06/06 04:24:46 - mmengine - INFO - Epoch(train) [5][2000/4092] lr: 6.2200e-05 eta: 4:00:00 time: 0.6371 data_time: 0.0016 memory: 44140 loss: 0.4132 +2023/06/06 04:25:50 - mmengine - INFO - Epoch(train) [5][2100/4092] lr: 6.1859e-05 eta: 3:58:56 time: 0.6336 data_time: 0.0017 memory: 44140 loss: 0.4684 +2023/06/06 04:26:53 - mmengine - INFO - Epoch(train) [5][2200/4092] lr: 6.1517e-05 eta: 3:57:51 time: 0.6357 data_time: 0.0020 memory: 44140 loss: 0.4058 +2023/06/06 04:27:57 - mmengine - INFO - Epoch(train) [5][2300/4092] lr: 6.1175e-05 eta: 3:56:47 time: 0.6328 data_time: 0.0016 memory: 44140 loss: 0.4473 +2023/06/06 04:29:00 - mmengine - INFO - Epoch(train) [5][2400/4092] lr: 6.0833e-05 eta: 3:55:43 time: 0.6336 data_time: 0.0016 memory: 44140 loss: 0.4328 +2023/06/06 04:30:04 - mmengine - INFO - Epoch(train) [5][2500/4092] lr: 6.0490e-05 eta: 3:54:38 time: 0.6354 data_time: 0.0015 memory: 44140 loss: 0.4395 +2023/06/06 04:31:07 - mmengine - INFO - Epoch(train) [5][2600/4092] lr: 6.0147e-05 eta: 3:53:34 time: 0.6356 data_time: 0.0015 memory: 44140 loss: 0.4286 +2023/06/06 04:31:28 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_all2_lr1e-4_20230606_005614 +2023/06/06 04:32:11 - mmengine - INFO - Epoch(train) [5][2700/4092] lr: 5.9803e-05 eta: 3:52:30 time: 0.6328 data_time: 0.0013 memory: 44140 loss: 0.4076 +2023/06/06 04:33:14 - mmengine - INFO - Epoch(train) [5][2800/4092] lr: 5.9460e-05 eta: 3:51:26 time: 0.6349 data_time: 0.0015 memory: 44140 loss: 0.4158 +2023/06/06 04:34:18 - mmengine - INFO - Epoch(train) [5][2900/4092] lr: 5.9116e-05 eta: 3:50:21 time: 0.6347 data_time: 0.0014 memory: 44140 loss: 0.4413 +2023/06/06 04:35:21 - mmengine - INFO - Epoch(train) [5][3000/4092] lr: 5.8772e-05 eta: 3:49:17 time: 0.6336 data_time: 0.0014 memory: 44140 loss: 0.4374 +2023/06/06 04:36:25 - mmengine - INFO - Epoch(train) [5][3100/4092] lr: 5.8427e-05 eta: 3:48:13 time: 0.6353 data_time: 0.0015 memory: 44140 loss: 0.4362 +2023/06/06 04:37:28 - mmengine - INFO - Epoch(train) [5][3200/4092] lr: 5.8083e-05 eta: 3:47:09 time: 0.6328 data_time: 0.0018 memory: 44140 loss: 0.4268 +2023/06/06 04:38:31 - mmengine - INFO - Epoch(train) [5][3300/4092] lr: 5.7738e-05 eta: 3:46:04 time: 0.6329 data_time: 0.0015 memory: 44140 loss: 0.4766 +2023/06/06 04:39:35 - mmengine - INFO - Epoch(train) [5][3400/4092] lr: 5.7393e-05 eta: 3:45:00 time: 0.6352 data_time: 0.0024 memory: 44140 loss: 0.4457 +2023/06/06 04:40:38 - mmengine - INFO - Epoch(train) [5][3500/4092] lr: 5.7048e-05 eta: 3:43:56 time: 0.6324 data_time: 0.0018 memory: 44140 loss: 0.4370 +2023/06/06 04:41:42 - mmengine - INFO - Epoch(train) [5][3600/4092] lr: 5.6703e-05 eta: 3:42:51 time: 0.6327 data_time: 0.0023 memory: 44140 loss: 0.4330 +2023/06/06 04:42:02 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_all2_lr1e-4_20230606_005614 +2023/06/06 04:42:45 - mmengine - INFO - Epoch(train) [5][3700/4092] lr: 5.6358e-05 eta: 3:41:47 time: 0.6361 data_time: 0.0019 memory: 44140 loss: 0.4219 +2023/06/06 04:43:49 - mmengine - INFO - Epoch(train) [5][3800/4092] lr: 5.6012e-05 eta: 3:40:43 time: 0.6338 data_time: 0.0017 memory: 44140 loss: 0.4352 +2023/06/06 04:44:52 - mmengine - INFO - Epoch(train) [5][3900/4092] lr: 5.5667e-05 eta: 3:39:39 time: 0.6346 data_time: 0.0023 memory: 44140 loss: 0.4157 +2023/06/06 04:45:55 - mmengine - INFO - Epoch(train) [5][4000/4092] lr: 5.5321e-05 eta: 3:38:35 time: 0.6326 data_time: 0.0014 memory: 44140 loss: 0.4114 +2023/06/06 04:46:53 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_all2_lr1e-4_20230606_005614 +2023/06/06 04:46:53 - mmengine - INFO - Saving checkpoint at 5 epochs +2023/06/06 04:49:14 - mmengine - INFO - Epoch(val) [5][100/119] eta: 0:00:24 time: 1.2875 data_time: 0.0007 memory: 44140 +2023/06/06 04:49:57 - mmengine - INFO - Epoch(val) [5][119/119] accuracy/top1: 79.0411 single-label/precision_classwise: [96.61665344238281, 50.51228332519531] single-label/recall_classwise: [76.01371765136719, 90.19379425048828] single-label/f1-score_classwise: [85.08573150634766, 64.75760650634766] data_time: 0.0182 time: 1.2853 +2023/06/06 04:51:04 - mmengine - INFO - Epoch(train) [6][ 100/4092] lr: 5.4658e-05 eta: 3:36:35 time: 0.6321 data_time: 0.0014 memory: 44140 loss: 0.3863 +2023/06/06 04:52:08 - mmengine - INFO - Epoch(train) [6][ 200/4092] lr: 5.4313e-05 eta: 3:35:30 time: 0.6337 data_time: 0.0015 memory: 44140 loss: 0.4308 +2023/06/06 04:53:11 - mmengine - INFO - Epoch(train) [6][ 300/4092] lr: 5.3967e-05 eta: 3:34:26 time: 0.6327 data_time: 0.0016 memory: 44140 loss: 0.4431 +2023/06/06 04:54:14 - mmengine - INFO - Epoch(train) [6][ 400/4092] lr: 5.3622e-05 eta: 3:33:22 time: 0.6327 data_time: 0.0014 memory: 44140 loss: 0.4252 +2023/06/06 04:55:18 - mmengine - INFO - Epoch(train) [6][ 500/4092] lr: 5.3276e-05 eta: 3:32:17 time: 0.6334 data_time: 0.0014 memory: 44140 loss: 0.4332 +2023/06/06 04:55:43 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_all2_lr1e-4_20230606_005614 +2023/06/06 04:56:21 - mmengine - INFO - Epoch(train) [6][ 600/4092] lr: 5.2931e-05 eta: 3:31:13 time: 0.6326 data_time: 0.0015 memory: 44140 loss: 0.4298 +2023/06/06 04:57:24 - mmengine - INFO - Epoch(train) [6][ 700/4092] lr: 5.2586e-05 eta: 3:30:09 time: 0.6339 data_time: 0.0015 memory: 44140 loss: 0.4453 +2023/06/06 04:58:28 - mmengine - INFO - Epoch(train) [6][ 800/4092] lr: 5.2241e-05 eta: 3:29:05 time: 0.6310 data_time: 0.0014 memory: 44140 loss: 0.4186 +2023/06/06 04:59:31 - mmengine - INFO - Epoch(train) [6][ 900/4092] lr: 5.1897e-05 eta: 3:28:00 time: 0.6324 data_time: 0.0015 memory: 44140 loss: 0.4301 +2023/06/06 05:00:34 - mmengine - INFO - Epoch(train) [6][1000/4092] lr: 5.1552e-05 eta: 3:26:56 time: 0.6323 data_time: 0.0018 memory: 44140 loss: 0.4261 +2023/06/06 05:01:37 - mmengine - INFO - Epoch(train) [6][1100/4092] lr: 5.1208e-05 eta: 3:25:52 time: 0.6319 data_time: 0.0023 memory: 44140 loss: 0.4062 +2023/06/06 05:02:41 - mmengine - INFO - Epoch(train) [6][1200/4092] lr: 5.0864e-05 eta: 3:24:48 time: 0.6329 data_time: 0.0017 memory: 44140 loss: 0.4500 +2023/06/06 05:03:44 - mmengine - INFO - Epoch(train) [6][1300/4092] lr: 5.0520e-05 eta: 3:23:43 time: 0.6339 data_time: 0.0015 memory: 44140 loss: 0.4262 +2023/06/06 05:04:47 - mmengine - INFO - Epoch(train) [6][1400/4092] lr: 5.0176e-05 eta: 3:22:39 time: 0.6331 data_time: 0.0017 memory: 44140 loss: 0.3823 +2023/06/06 05:05:51 - mmengine - INFO - Epoch(train) [6][1500/4092] lr: 4.9833e-05 eta: 3:21:35 time: 0.6337 data_time: 0.0016 memory: 44140 loss: 0.4539 +2023/06/06 05:06:16 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_all2_lr1e-4_20230606_005614 +2023/06/06 05:06:54 - mmengine - INFO - Epoch(train) [6][1600/4092] lr: 4.9490e-05 eta: 3:20:31 time: 0.6321 data_time: 0.0018 memory: 44140 loss: 0.4357 +2023/06/06 05:07:57 - mmengine - INFO - Epoch(train) [6][1700/4092] lr: 4.9147e-05 eta: 3:19:26 time: 0.6327 data_time: 0.0015 memory: 44140 loss: 0.4344 +2023/06/06 05:09:01 - mmengine - INFO - Epoch(train) [6][1800/4092] lr: 4.8805e-05 eta: 3:18:22 time: 0.6333 data_time: 0.0014 memory: 44140 loss: 0.4336 +2023/06/06 05:10:04 - mmengine - INFO - Epoch(train) [6][1900/4092] lr: 4.8462e-05 eta: 3:17:18 time: 0.6326 data_time: 0.0015 memory: 44140 loss: 0.4356 +2023/06/06 05:11:07 - mmengine - INFO - Epoch(train) [6][2000/4092] lr: 4.8121e-05 eta: 3:16:14 time: 0.6333 data_time: 0.0014 memory: 44140 loss: 0.4078 +2023/06/06 05:12:11 - mmengine - INFO - Epoch(train) [6][2100/4092] lr: 4.7780e-05 eta: 3:15:10 time: 0.6323 data_time: 0.0015 memory: 44140 loss: 0.4258 +2023/06/06 05:13:14 - mmengine - INFO - Epoch(train) [6][2200/4092] lr: 4.7439e-05 eta: 3:14:06 time: 0.6323 data_time: 0.0016 memory: 44140 loss: 0.4239 +2023/06/06 05:14:17 - mmengine - INFO - Epoch(train) [6][2300/4092] lr: 4.7099e-05 eta: 3:13:02 time: 0.6332 data_time: 0.0015 memory: 44140 loss: 0.4572 +2023/06/06 05:15:21 - mmengine - INFO - Epoch(train) [6][2400/4092] lr: 4.6759e-05 eta: 3:11:57 time: 0.6326 data_time: 0.0015 memory: 44140 loss: 0.4410 +2023/06/06 05:16:24 - mmengine - INFO - Epoch(train) [6][2500/4092] lr: 4.6419e-05 eta: 3:10:53 time: 0.6344 data_time: 0.0017 memory: 44140 loss: 0.4452 +2023/06/06 05:16:49 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_all2_lr1e-4_20230606_005614 +2023/06/06 05:17:27 - mmengine - INFO - Epoch(train) [6][2600/4092] lr: 4.6080e-05 eta: 3:09:49 time: 0.6324 data_time: 0.0024 memory: 44140 loss: 0.4552 +2023/06/06 05:18:31 - mmengine - INFO - Epoch(train) [6][2700/4092] lr: 4.5742e-05 eta: 3:08:45 time: 0.6334 data_time: 0.0014 memory: 44140 loss: 0.4126 +2023/06/06 05:19:34 - mmengine - INFO - Epoch(train) [6][2800/4092] lr: 4.5404e-05 eta: 3:07:41 time: 0.6323 data_time: 0.0014 memory: 44140 loss: 0.3934 +2023/06/06 05:20:37 - mmengine - INFO - Epoch(train) [6][2900/4092] lr: 4.5067e-05 eta: 3:06:37 time: 0.6365 data_time: 0.0015 memory: 44140 loss: 0.4579 +2023/06/06 05:21:41 - mmengine - INFO - Epoch(train) [6][3000/4092] lr: 4.4730e-05 eta: 3:05:33 time: 0.6339 data_time: 0.0016 memory: 44140 loss: 0.4238 +2023/06/06 05:22:44 - mmengine - INFO - Epoch(train) [6][3100/4092] lr: 4.4394e-05 eta: 3:04:29 time: 0.6327 data_time: 0.0016 memory: 44140 loss: 0.4227 +2023/06/06 05:23:47 - mmengine - INFO - Epoch(train) [6][3200/4092] lr: 4.4059e-05 eta: 3:03:25 time: 0.6359 data_time: 0.0020 memory: 44140 loss: 0.4336 +2023/06/06 05:24:51 - mmengine - INFO - Epoch(train) [6][3300/4092] lr: 4.3724e-05 eta: 3:02:21 time: 0.6328 data_time: 0.0015 memory: 44140 loss: 0.4338 +2023/06/06 05:25:54 - mmengine - INFO - Epoch(train) [6][3400/4092] lr: 4.3390e-05 eta: 3:01:16 time: 0.6324 data_time: 0.0019 memory: 44140 loss: 0.4539 +2023/06/06 05:26:57 - mmengine - INFO - Epoch(train) [6][3500/4092] lr: 4.3056e-05 eta: 3:00:12 time: 0.6344 data_time: 0.0016 memory: 44140 loss: 0.4467 +2023/06/06 05:27:23 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_all2_lr1e-4_20230606_005614 +2023/06/06 05:28:01 - mmengine - INFO - Epoch(train) [6][3600/4092] lr: 4.2724e-05 eta: 2:59:08 time: 0.6322 data_time: 0.0018 memory: 44140 loss: 0.4293 +2023/06/06 05:29:04 - mmengine - INFO - Epoch(train) [6][3700/4092] lr: 4.2392e-05 eta: 2:58:04 time: 0.6326 data_time: 0.0015 memory: 44140 loss: 0.4475 +2023/06/06 05:30:08 - mmengine - INFO - Epoch(train) [6][3800/4092] lr: 4.2060e-05 eta: 2:57:00 time: 0.6326 data_time: 0.0017 memory: 44140 loss: 0.4637 +2023/06/06 05:31:11 - mmengine - INFO - Epoch(train) [6][3900/4092] lr: 4.1730e-05 eta: 2:55:56 time: 0.6341 data_time: 0.0018 memory: 44140 loss: 0.4521 +2023/06/06 05:32:14 - mmengine - INFO - Epoch(train) [6][4000/4092] lr: 4.1400e-05 eta: 2:54:52 time: 0.6329 data_time: 0.0015 memory: 44140 loss: 0.4264 +2023/06/06 05:33:12 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_all2_lr1e-4_20230606_005614 +2023/06/06 05:33:12 - mmengine - INFO - Saving checkpoint at 6 epochs +2023/06/06 05:35:33 - mmengine - INFO - Epoch(val) [6][100/119] eta: 0:00:24 time: 1.2877 data_time: 0.0007 memory: 44140 +2023/06/06 05:36:16 - mmengine - INFO - Epoch(val) [6][119/119] accuracy/top1: 80.9394 single-label/precision_classwise: [96.39708709716797, 53.19827651977539] single-label/recall_classwise: [78.7071533203125, 89.16278839111328] single-label/f1-score_classwise: [86.6585464477539, 66.6376953125] data_time: 0.0175 time: 1.2862 +2023/06/06 05:37:22 - mmengine - INFO - Epoch(train) [7][ 100/4092] lr: 4.0769e-05 eta: 2:52:51 time: 0.6331 data_time: 0.0016 memory: 44140 loss: 0.4452 +2023/06/06 05:38:26 - mmengine - INFO - Epoch(train) [7][ 200/4092] lr: 4.0442e-05 eta: 2:51:47 time: 0.6327 data_time: 0.0015 memory: 44140 loss: 0.4411 +2023/06/06 05:39:29 - mmengine - INFO - Epoch(train) [7][ 300/4092] lr: 4.0116e-05 eta: 2:50:43 time: 0.6337 data_time: 0.0015 memory: 44140 loss: 0.4431 +2023/06/06 05:40:32 - mmengine - INFO - Epoch(train) [7][ 400/4092] lr: 3.9790e-05 eta: 2:49:39 time: 0.6330 data_time: 0.0014 memory: 44140 loss: 0.4064 +2023/06/06 05:41:03 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_all2_lr1e-4_20230606_005614 +2023/06/06 05:41:36 - mmengine - INFO - Epoch(train) [7][ 500/4092] lr: 3.9465e-05 eta: 2:48:35 time: 0.6327 data_time: 0.0021 memory: 44140 loss: 0.4140 +2023/06/06 05:42:39 - mmengine - INFO - Epoch(train) [7][ 600/4092] lr: 3.9141e-05 eta: 2:47:31 time: 0.6324 data_time: 0.0015 memory: 44140 loss: 0.4600 +2023/06/06 05:43:43 - mmengine - INFO - Epoch(train) [7][ 700/4092] lr: 3.8819e-05 eta: 2:46:27 time: 0.6326 data_time: 0.0014 memory: 44140 loss: 0.4257 +2023/06/06 05:44:46 - mmengine - INFO - Epoch(train) [7][ 800/4092] lr: 3.8497e-05 eta: 2:45:23 time: 0.6328 data_time: 0.0015 memory: 44140 loss: 0.4564 +2023/06/06 05:45:49 - mmengine - INFO - Epoch(train) [7][ 900/4092] lr: 3.8176e-05 eta: 2:44:19 time: 0.6324 data_time: 0.0014 memory: 44140 loss: 0.4589 +2023/06/06 05:46:52 - mmengine - INFO - Epoch(train) [7][1000/4092] lr: 3.7856e-05 eta: 2:43:15 time: 0.6321 data_time: 0.0024 memory: 44140 loss: 0.4382 +2023/06/06 05:47:56 - mmengine - INFO - Epoch(train) [7][1100/4092] lr: 3.7537e-05 eta: 2:42:11 time: 0.6317 data_time: 0.0019 memory: 44140 loss: 0.4153 +2023/06/06 05:48:59 - mmengine - INFO - Epoch(train) [7][1200/4092] lr: 3.7219e-05 eta: 2:41:07 time: 0.6321 data_time: 0.0022 memory: 44140 loss: 0.4278 +2023/06/06 05:50:02 - mmengine - INFO - Epoch(train) [7][1300/4092] lr: 3.6902e-05 eta: 2:40:03 time: 0.6340 data_time: 0.0020 memory: 44140 loss: 0.4140 +2023/06/06 05:51:06 - mmengine - INFO - Epoch(train) [7][1400/4092] lr: 3.6586e-05 eta: 2:39:00 time: 0.6349 data_time: 0.0018 memory: 44140 loss: 0.4359 +2023/06/06 05:51:37 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_all2_lr1e-4_20230606_005614 +2023/06/06 05:52:10 - mmengine - INFO - Epoch(train) [7][1500/4092] lr: 3.6272e-05 eta: 2:37:56 time: 0.6340 data_time: 0.0019 memory: 44140 loss: 0.4346 +2023/06/06 05:53:13 - mmengine - INFO - Epoch(train) [7][1600/4092] lr: 3.5958e-05 eta: 2:36:52 time: 0.6338 data_time: 0.0020 memory: 44140 loss: 0.4228 +2023/06/06 05:54:20 - mmengine - INFO - Epoch(train) [7][1700/4092] lr: 3.5646e-05 eta: 2:35:50 time: 1.0242 data_time: 0.0273 memory: 44140 loss: 0.4392 +2023/06/06 05:55:24 - mmengine - INFO - Epoch(train) [7][1800/4092] lr: 3.5334e-05 eta: 2:34:46 time: 0.6342 data_time: 0.0025 memory: 44140 loss: 0.4330 +2023/06/06 05:56:27 - mmengine - INFO - Epoch(train) [7][1900/4092] lr: 3.5024e-05 eta: 2:33:42 time: 0.6351 data_time: 0.0027 memory: 44140 loss: 0.4355 +2023/06/06 05:57:31 - mmengine - INFO - Epoch(train) [7][2000/4092] lr: 3.4715e-05 eta: 2:32:38 time: 0.6419 data_time: 0.0017 memory: 44140 loss: 0.4206 +2023/06/06 05:58:35 - mmengine - INFO - Epoch(train) [7][2100/4092] lr: 3.4407e-05 eta: 2:31:35 time: 0.6338 data_time: 0.0016 memory: 44140 loss: 0.4379 +2023/06/06 05:59:38 - mmengine - INFO - Epoch(train) [7][2200/4092] lr: 3.4101e-05 eta: 2:30:31 time: 0.6320 data_time: 0.0015 memory: 44140 loss: 0.4500 +2023/06/06 06:00:41 - mmengine - INFO - Epoch(train) [7][2300/4092] lr: 3.3796e-05 eta: 2:29:27 time: 0.6355 data_time: 0.0015 memory: 44140 loss: 0.4528 +2023/06/06 06:01:45 - mmengine - INFO - Epoch(train) [7][2400/4092] lr: 3.3491e-05 eta: 2:28:23 time: 0.6341 data_time: 0.0021 memory: 44140 loss: 0.4518 +2023/06/06 06:02:15 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_all2_lr1e-4_20230606_005614 +2023/06/06 06:02:48 - mmengine - INFO - Epoch(train) [7][2500/4092] lr: 3.3189e-05 eta: 2:27:19 time: 0.6331 data_time: 0.0018 memory: 44140 loss: 0.4160 +2023/06/06 06:03:51 - mmengine - INFO - Epoch(train) [7][2600/4092] lr: 3.2887e-05 eta: 2:26:15 time: 0.6332 data_time: 0.0023 memory: 44140 loss: 0.4469 +2023/06/06 06:04:55 - mmengine - INFO - Epoch(train) [7][2700/4092] lr: 3.2587e-05 eta: 2:25:11 time: 0.6332 data_time: 0.0025 memory: 44140 loss: 0.4390 +2023/06/06 06:05:58 - mmengine - INFO - Epoch(train) [7][2800/4092] lr: 3.2288e-05 eta: 2:24:07 time: 0.6329 data_time: 0.0015 memory: 44140 loss: 0.4331 +2023/06/06 06:07:02 - mmengine - INFO - Epoch(train) [7][2900/4092] lr: 3.1990e-05 eta: 2:23:03 time: 0.6331 data_time: 0.0023 memory: 44140 loss: 0.4151 +2023/06/06 06:08:05 - mmengine - INFO - Epoch(train) [7][3000/4092] lr: 3.1694e-05 eta: 2:21:59 time: 0.6323 data_time: 0.0019 memory: 44140 loss: 0.4372 +2023/06/06 06:09:08 - mmengine - INFO - Epoch(train) [7][3100/4092] lr: 3.1399e-05 eta: 2:20:55 time: 0.6338 data_time: 0.0020 memory: 44140 loss: 0.4525 +2023/06/06 06:10:12 - mmengine - INFO - Epoch(train) [7][3200/4092] lr: 3.1106e-05 eta: 2:19:51 time: 0.6332 data_time: 0.0018 memory: 44140 loss: 0.4206 +2023/06/06 06:11:15 - mmengine - INFO - Epoch(train) [7][3300/4092] lr: 3.0814e-05 eta: 2:18:48 time: 0.6336 data_time: 0.0018 memory: 44140 loss: 0.4574 +2023/06/06 06:12:18 - mmengine - INFO - Epoch(train) [7][3400/4092] lr: 3.0523e-05 eta: 2:17:44 time: 0.6330 data_time: 0.0020 memory: 44140 loss: 0.4343 +2023/06/06 06:12:49 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_all2_lr1e-4_20230606_005614 +2023/06/06 06:13:22 - mmengine - INFO - Epoch(train) [7][3500/4092] lr: 3.0234e-05 eta: 2:16:40 time: 0.6341 data_time: 0.0018 memory: 44140 loss: 0.4280 +2023/06/06 06:14:25 - mmengine - INFO - Epoch(train) [7][3600/4092] lr: 2.9946e-05 eta: 2:15:36 time: 0.6324 data_time: 0.0021 memory: 44140 loss: 0.4194 +2023/06/06 06:15:28 - mmengine - INFO - Epoch(train) [7][3700/4092] lr: 2.9660e-05 eta: 2:14:32 time: 0.6323 data_time: 0.0018 memory: 44140 loss: 0.4519 +2023/06/06 06:16:32 - mmengine - INFO - Epoch(train) [7][3800/4092] lr: 2.9375e-05 eta: 2:13:28 time: 0.6333 data_time: 0.0024 memory: 44140 loss: 0.4330 +2023/06/06 06:17:35 - mmengine - INFO - Epoch(train) [7][3900/4092] lr: 2.9092e-05 eta: 2:12:24 time: 0.6322 data_time: 0.0021 memory: 44140 loss: 0.4570 +2023/06/06 06:18:38 - mmengine - INFO - Epoch(train) [7][4000/4092] lr: 2.8810e-05 eta: 2:11:20 time: 0.6325 data_time: 0.0017 memory: 44140 loss: 0.4274 +2023/06/06 06:19:36 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_all2_lr1e-4_20230606_005614 +2023/06/06 06:19:36 - mmengine - INFO - Saving checkpoint at 7 epochs +2023/06/06 06:21:58 - mmengine - INFO - Epoch(val) [7][100/119] eta: 0:00:24 time: 1.2877 data_time: 0.0006 memory: 44140 +2023/06/06 06:22:40 - mmengine - INFO - Epoch(val) [7][119/119] accuracy/top1: 82.5083 single-label/precision_classwise: [96.11871337890625, 55.72473907470703] single-label/recall_classwise: [81.03234100341797, 87.94573211669922] single-label/f1-score_classwise: [87.93313598632812, 68.22212982177734] data_time: 0.0170 time: 1.2840 +2023/06/06 06:23:47 - mmengine - INFO - Epoch(train) [8][ 100/4092] lr: 2.8274e-05 eta: 2:09:19 time: 0.6333 data_time: 0.0020 memory: 44140 loss: 0.4547 +2023/06/06 06:24:51 - mmengine - INFO - Epoch(train) [8][ 200/4092] lr: 2.7997e-05 eta: 2:08:15 time: 0.6333 data_time: 0.0015 memory: 44140 loss: 0.4534 +2023/06/06 06:25:54 - mmengine - INFO - Epoch(train) [8][ 300/4092] lr: 2.7721e-05 eta: 2:07:11 time: 0.6471 data_time: 0.0019 memory: 44140 loss: 0.4233 +2023/06/06 06:26:29 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_all2_lr1e-4_20230606_005614 +2023/06/06 06:26:57 - mmengine - INFO - Epoch(train) [8][ 400/4092] lr: 2.7447e-05 eta: 2:06:07 time: 0.6344 data_time: 0.0021 memory: 44140 loss: 0.4301 +2023/06/06 06:28:01 - mmengine - INFO - Epoch(train) [8][ 500/4092] lr: 2.7175e-05 eta: 2:05:03 time: 0.6328 data_time: 0.0018 memory: 44140 loss: 0.4388 +2023/06/06 06:29:04 - mmengine - INFO - Epoch(train) [8][ 600/4092] lr: 2.6904e-05 eta: 2:03:59 time: 0.6319 data_time: 0.0017 memory: 44140 loss: 0.4352 +2023/06/06 06:30:07 - mmengine - INFO - Epoch(train) [8][ 700/4092] lr: 2.6635e-05 eta: 2:02:56 time: 0.6606 data_time: 0.0016 memory: 44140 loss: 0.3998 +2023/06/06 06:31:11 - mmengine - INFO - Epoch(train) [8][ 800/4092] lr: 2.6368e-05 eta: 2:01:52 time: 0.6344 data_time: 0.0016 memory: 44140 loss: 0.4316 +2023/06/06 06:32:14 - mmengine - INFO - Epoch(train) [8][ 900/4092] lr: 2.6102e-05 eta: 2:00:48 time: 0.6356 data_time: 0.0014 memory: 44140 loss: 0.4166 +2023/06/06 06:33:17 - mmengine - INFO - Epoch(train) [8][1000/4092] lr: 2.5838e-05 eta: 1:59:44 time: 0.6324 data_time: 0.0014 memory: 44140 loss: 0.4451 +2023/06/06 06:34:21 - mmengine - INFO - Epoch(train) [8][1100/4092] lr: 2.5576e-05 eta: 1:58:40 time: 0.6329 data_time: 0.0014 memory: 44140 loss: 0.4389 +2023/06/06 06:35:24 - mmengine - INFO - Epoch(train) [8][1200/4092] lr: 2.5315e-05 eta: 1:57:36 time: 0.6327 data_time: 0.0015 memory: 44140 loss: 0.4256 +2023/06/06 06:36:27 - mmengine - INFO - Epoch(train) [8][1300/4092] lr: 2.5056e-05 eta: 1:56:32 time: 0.6322 data_time: 0.0014 memory: 44140 loss: 0.4029 +2023/06/06 06:37:02 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_all2_lr1e-4_20230606_005614 +2023/06/06 06:37:30 - mmengine - INFO - Epoch(train) [8][1400/4092] lr: 2.4799e-05 eta: 1:55:28 time: 0.6318 data_time: 0.0014 memory: 44140 loss: 0.4328 +2023/06/06 06:38:34 - mmengine - INFO - Epoch(train) [8][1500/4092] lr: 2.4544e-05 eta: 1:54:25 time: 0.6323 data_time: 0.0015 memory: 44140 loss: 0.4334 +2023/06/06 06:39:37 - mmengine - INFO - Epoch(train) [8][1600/4092] lr: 2.4291e-05 eta: 1:53:21 time: 0.6311 data_time: 0.0023 memory: 44140 loss: 0.4030 +2023/06/06 06:40:40 - mmengine - INFO - Epoch(train) [8][1700/4092] lr: 2.4039e-05 eta: 1:52:17 time: 0.6329 data_time: 0.0014 memory: 44140 loss: 0.4400 +2023/06/06 06:41:43 - mmengine - INFO - Epoch(train) [8][1800/4092] lr: 2.3789e-05 eta: 1:51:13 time: 0.6322 data_time: 0.0015 memory: 44140 loss: 0.4242 +2023/06/06 06:42:47 - mmengine - INFO - Epoch(train) [8][1900/4092] lr: 2.3541e-05 eta: 1:50:09 time: 0.6333 data_time: 0.0015 memory: 44140 loss: 0.4326 +2023/06/06 06:43:50 - mmengine - INFO - Epoch(train) [8][2000/4092] lr: 2.3295e-05 eta: 1:49:05 time: 0.6321 data_time: 0.0016 memory: 44140 loss: 0.4057 +2023/06/06 06:44:53 - mmengine - INFO - Epoch(train) [8][2100/4092] lr: 2.3051e-05 eta: 1:48:01 time: 0.6324 data_time: 0.0015 memory: 44140 loss: 0.4293 +2023/06/06 06:45:57 - mmengine - INFO - Epoch(train) [8][2200/4092] lr: 2.2809e-05 eta: 1:46:58 time: 0.6320 data_time: 0.0014 memory: 44140 loss: 0.4227 +2023/06/06 06:47:00 - mmengine - INFO - Epoch(train) [8][2300/4092] lr: 2.2568e-05 eta: 1:45:54 time: 0.6322 data_time: 0.0017 memory: 44140 loss: 0.4292 +2023/06/06 06:47:35 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_all2_lr1e-4_20230606_005614 +2023/06/06 06:48:03 - mmengine - INFO - Epoch(train) [8][2400/4092] lr: 2.2330e-05 eta: 1:44:50 time: 0.6315 data_time: 0.0021 memory: 44140 loss: 0.4299 +2023/06/06 06:49:06 - mmengine - INFO - Epoch(train) [8][2500/4092] lr: 2.2093e-05 eta: 1:43:46 time: 0.6322 data_time: 0.0019 memory: 44140 loss: 0.4226 +2023/06/06 06:50:10 - mmengine - INFO - Epoch(train) [8][2600/4092] lr: 2.1858e-05 eta: 1:42:42 time: 0.6334 data_time: 0.0015 memory: 44140 loss: 0.4533 +2023/06/06 06:51:13 - mmengine - INFO - Epoch(train) [8][2700/4092] lr: 2.1626e-05 eta: 1:41:38 time: 0.6318 data_time: 0.0016 memory: 44140 loss: 0.4104 +2023/06/06 06:52:16 - mmengine - INFO - Epoch(train) [8][2800/4092] lr: 2.1395e-05 eta: 1:40:35 time: 0.6330 data_time: 0.0016 memory: 44140 loss: 0.4277 +2023/06/06 06:53:20 - mmengine - INFO - Epoch(train) [8][2900/4092] lr: 2.1166e-05 eta: 1:39:31 time: 0.6349 data_time: 0.0033 memory: 44140 loss: 0.4660 +2023/06/06 06:54:23 - mmengine - INFO - Epoch(train) [8][3000/4092] lr: 2.0939e-05 eta: 1:38:27 time: 0.6327 data_time: 0.0015 memory: 44140 loss: 0.4146 +2023/06/06 06:55:26 - mmengine - INFO - Epoch(train) [8][3100/4092] lr: 2.0715e-05 eta: 1:37:23 time: 0.6330 data_time: 0.0015 memory: 44140 loss: 0.4298 +2023/06/06 06:56:30 - mmengine - INFO - Epoch(train) [8][3200/4092] lr: 2.0492e-05 eta: 1:36:20 time: 0.6326 data_time: 0.0015 memory: 44140 loss: 0.4238 +2023/06/06 06:57:33 - mmengine - INFO - Epoch(train) [8][3300/4092] lr: 2.0271e-05 eta: 1:35:16 time: 0.6332 data_time: 0.0016 memory: 44140 loss: 0.4399 +2023/06/06 06:58:09 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_all2_lr1e-4_20230606_005614 +2023/06/06 06:58:36 - mmengine - INFO - Epoch(train) [8][3400/4092] lr: 2.0052e-05 eta: 1:34:12 time: 0.6324 data_time: 0.0018 memory: 44140 loss: 0.4230 +2023/06/06 06:59:40 - mmengine - INFO - Epoch(train) [8][3500/4092] lr: 1.9836e-05 eta: 1:33:08 time: 0.6349 data_time: 0.0018 memory: 44140 loss: 0.4160 +2023/06/06 07:00:43 - mmengine - INFO - Epoch(train) [8][3600/4092] lr: 1.9621e-05 eta: 1:32:04 time: 0.6338 data_time: 0.0017 memory: 44140 loss: 0.4029 +2023/06/06 07:01:47 - mmengine - INFO - Epoch(train) [8][3700/4092] lr: 1.9409e-05 eta: 1:31:01 time: 0.6326 data_time: 0.0015 memory: 44140 loss: 0.4238 +2023/06/06 07:02:50 - mmengine - INFO - Epoch(train) [8][3800/4092] lr: 1.9198e-05 eta: 1:29:57 time: 0.6321 data_time: 0.0022 memory: 44140 loss: 0.4153 +2023/06/06 07:03:53 - mmengine - INFO - Epoch(train) [8][3900/4092] lr: 1.8990e-05 eta: 1:28:53 time: 0.6325 data_time: 0.0015 memory: 44140 loss: 0.4483 +2023/06/06 07:04:57 - mmengine - INFO - Epoch(train) [8][4000/4092] lr: 1.8784e-05 eta: 1:27:49 time: 0.6318 data_time: 0.0022 memory: 44140 loss: 0.4293 +2023/06/06 07:05:54 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_all2_lr1e-4_20230606_005614 +2023/06/06 07:05:54 - mmengine - INFO - Saving checkpoint at 8 epochs +2023/06/06 07:08:15 - mmengine - INFO - Epoch(val) [8][100/119] eta: 0:00:24 time: 1.2877 data_time: 0.0006 memory: 44140 +2023/06/06 07:08:57 - mmengine - INFO - Epoch(val) [8][119/119] accuracy/top1: 83.3822 single-label/precision_classwise: [95.77450561523438, 57.337921142578125] single-label/recall_classwise: [82.51162719726562, 86.58914184570312] single-label/f1-score_classwise: [88.64974975585938, 68.99107360839844] data_time: 0.0176 time: 1.2849 +2023/06/06 07:10:04 - mmengine - INFO - Epoch(train) [9][ 100/4092] lr: 1.8394e-05 eta: 1:25:48 time: 0.6331 data_time: 0.0015 memory: 44140 loss: 0.4192 +2023/06/06 07:11:08 - mmengine - INFO - Epoch(train) [9][ 200/4092] lr: 1.8194e-05 eta: 1:24:44 time: 0.6337 data_time: 0.0018 memory: 44140 loss: 0.4159 +2023/06/06 07:11:48 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_all2_lr1e-4_20230606_005614 +2023/06/06 07:12:11 - mmengine - INFO - Epoch(train) [9][ 300/4092] lr: 1.7997e-05 eta: 1:23:40 time: 0.6323 data_time: 0.0015 memory: 44140 loss: 0.4478 +2023/06/06 07:13:14 - mmengine - INFO - Epoch(train) [9][ 400/4092] lr: 1.7801e-05 eta: 1:22:36 time: 0.6328 data_time: 0.0016 memory: 44140 loss: 0.4213 +2023/06/06 07:14:18 - mmengine - INFO - Epoch(train) [9][ 500/4092] lr: 1.7608e-05 eta: 1:21:33 time: 0.6325 data_time: 0.0022 memory: 44140 loss: 0.4436 +2023/06/06 07:15:21 - mmengine - INFO - Epoch(train) [9][ 600/4092] lr: 1.7417e-05 eta: 1:20:29 time: 0.6316 data_time: 0.0018 memory: 44140 loss: 0.4177 +2023/06/06 07:16:24 - mmengine - INFO - Epoch(train) [9][ 700/4092] lr: 1.7228e-05 eta: 1:19:25 time: 0.6318 data_time: 0.0017 memory: 44140 loss: 0.4439 +2023/06/06 07:17:27 - mmengine - INFO - Epoch(train) [9][ 800/4092] lr: 1.7041e-05 eta: 1:18:21 time: 0.6323 data_time: 0.0016 memory: 44140 loss: 0.4294 +2023/06/06 07:18:31 - mmengine - INFO - Epoch(train) [9][ 900/4092] lr: 1.6857e-05 eta: 1:17:18 time: 0.6332 data_time: 0.0021 memory: 44140 loss: 0.4280 +2023/06/06 07:19:34 - mmengine - INFO - Epoch(train) [9][1000/4092] lr: 1.6675e-05 eta: 1:16:14 time: 0.6328 data_time: 0.0024 memory: 44140 loss: 0.4527 +2023/06/06 07:20:37 - mmengine - INFO - Epoch(train) [9][1100/4092] lr: 1.6495e-05 eta: 1:15:10 time: 0.6425 data_time: 0.0015 memory: 44140 loss: 0.4159 +2023/06/06 07:21:41 - mmengine - INFO - Epoch(train) [9][1200/4092] lr: 1.6317e-05 eta: 1:14:06 time: 0.6325 data_time: 0.0017 memory: 44140 loss: 0.4096 +2023/06/06 07:22:21 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_all2_lr1e-4_20230606_005614 +2023/06/06 07:22:44 - mmengine - INFO - Epoch(train) [9][1300/4092] lr: 1.6142e-05 eta: 1:13:03 time: 0.6315 data_time: 0.0026 memory: 44140 loss: 0.4171 +2023/06/06 07:23:47 - mmengine - INFO - Epoch(train) [9][1400/4092] lr: 1.5969e-05 eta: 1:11:59 time: 0.6314 data_time: 0.0016 memory: 44140 loss: 0.4286 +2023/06/06 07:24:50 - mmengine - INFO - Epoch(train) [9][1500/4092] lr: 1.5798e-05 eta: 1:10:55 time: 0.6323 data_time: 0.0015 memory: 44140 loss: 0.4414 +2023/06/06 07:25:54 - mmengine - INFO - Epoch(train) [9][1600/4092] lr: 1.5629e-05 eta: 1:09:51 time: 0.6321 data_time: 0.0017 memory: 44140 loss: 0.4408 +2023/06/06 07:26:57 - mmengine - INFO - Epoch(train) [9][1700/4092] lr: 1.5463e-05 eta: 1:08:48 time: 0.6321 data_time: 0.0015 memory: 44140 loss: 0.4242 +2023/06/06 07:28:00 - mmengine - INFO - Epoch(train) [9][1800/4092] lr: 1.5299e-05 eta: 1:07:44 time: 0.6316 data_time: 0.0017 memory: 44140 loss: 0.4343 +2023/06/06 07:29:03 - mmengine - INFO - Epoch(train) [9][1900/4092] lr: 1.5138e-05 eta: 1:06:40 time: 0.6328 data_time: 0.0016 memory: 44140 loss: 0.4663 +2023/06/06 07:30:07 - mmengine - INFO - Epoch(train) [9][2000/4092] lr: 1.4979e-05 eta: 1:05:36 time: 0.6314 data_time: 0.0015 memory: 44140 loss: 0.4332 +2023/06/06 07:31:10 - mmengine - INFO - Epoch(train) [9][2100/4092] lr: 1.4822e-05 eta: 1:04:33 time: 0.6326 data_time: 0.0022 memory: 44140 loss: 0.4290 +2023/06/06 07:32:13 - mmengine - INFO - Epoch(train) [9][2200/4092] lr: 1.4668e-05 eta: 1:03:29 time: 0.6312 data_time: 0.0017 memory: 44140 loss: 0.4446 +2023/06/06 07:32:54 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_all2_lr1e-4_20230606_005614 +2023/06/06 07:33:17 - mmengine - INFO - Epoch(train) [9][2300/4092] lr: 1.4515e-05 eta: 1:02:25 time: 0.6336 data_time: 0.0015 memory: 44140 loss: 0.3901 +2023/06/06 07:34:20 - mmengine - INFO - Epoch(train) [9][2400/4092] lr: 1.4366e-05 eta: 1:01:21 time: 0.6338 data_time: 0.0016 memory: 44140 loss: 0.4295 +2023/06/06 07:35:24 - mmengine - INFO - Epoch(train) [9][2500/4092] lr: 1.4219e-05 eta: 1:00:18 time: 0.6336 data_time: 0.0020 memory: 44140 loss: 0.4015 +2023/06/06 07:36:27 - mmengine - INFO - Epoch(train) [9][2600/4092] lr: 1.4074e-05 eta: 0:59:14 time: 0.6346 data_time: 0.0016 memory: 44140 loss: 0.4517 +2023/06/06 07:37:31 - mmengine - INFO - Epoch(train) [9][2700/4092] lr: 1.3931e-05 eta: 0:58:10 time: 0.6338 data_time: 0.0017 memory: 44140 loss: 0.4484 +2023/06/06 07:38:34 - mmengine - INFO - Epoch(train) [9][2800/4092] lr: 1.3791e-05 eta: 0:57:07 time: 0.6338 data_time: 0.0016 memory: 44140 loss: 0.4113 +2023/06/06 07:39:38 - mmengine - INFO - Epoch(train) [9][2900/4092] lr: 1.3654e-05 eta: 0:56:03 time: 0.6355 data_time: 0.0022 memory: 44140 loss: 0.4379 +2023/06/06 07:40:41 - mmengine - INFO - Epoch(train) [9][3000/4092] lr: 1.3519e-05 eta: 0:54:59 time: 0.6333 data_time: 0.0015 memory: 44140 loss: 0.4329 +2023/06/06 07:41:45 - mmengine - INFO - Epoch(train) [9][3100/4092] lr: 1.3386e-05 eta: 0:53:56 time: 0.6344 data_time: 0.0020 memory: 44140 loss: 0.4024 +2023/06/06 07:42:48 - mmengine - INFO - Epoch(train) [9][3200/4092] lr: 1.3256e-05 eta: 0:52:52 time: 0.6335 data_time: 0.0019 memory: 44140 loss: 0.4243 +2023/06/06 07:43:29 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_all2_lr1e-4_20230606_005614 +2023/06/06 07:43:52 - mmengine - INFO - Epoch(train) [9][3300/4092] lr: 1.3128e-05 eta: 0:51:48 time: 0.6335 data_time: 0.0017 memory: 44140 loss: 0.4423 +2023/06/06 07:44:55 - mmengine - INFO - Epoch(train) [9][3400/4092] lr: 1.3003e-05 eta: 0:50:45 time: 0.6337 data_time: 0.0019 memory: 44140 loss: 0.4114 +2023/06/06 07:45:59 - mmengine - INFO - Epoch(train) [9][3500/4092] lr: 1.2880e-05 eta: 0:49:41 time: 0.6368 data_time: 0.0015 memory: 44140 loss: 0.4362 +2023/06/06 07:47:02 - mmengine - INFO - Epoch(train) [9][3600/4092] lr: 1.2759e-05 eta: 0:48:37 time: 0.6355 data_time: 0.0017 memory: 44140 loss: 0.4443 +2023/06/06 07:48:06 - mmengine - INFO - Epoch(train) [9][3700/4092] lr: 1.2641e-05 eta: 0:47:34 time: 0.6332 data_time: 0.0016 memory: 44140 loss: 0.4355 +2023/06/06 07:49:09 - mmengine - INFO - Epoch(train) [9][3800/4092] lr: 1.2526e-05 eta: 0:46:30 time: 0.6356 data_time: 0.0019 memory: 44140 loss: 0.4494 +2023/06/06 07:50:12 - mmengine - INFO - Epoch(train) [9][3900/4092] lr: 1.2413e-05 eta: 0:45:26 time: 0.6345 data_time: 0.0019 memory: 44140 loss: 0.4139 +2023/06/06 07:51:16 - mmengine - INFO - Epoch(train) [9][4000/4092] lr: 1.2303e-05 eta: 0:44:23 time: 0.6347 data_time: 0.0016 memory: 44140 loss: 0.4212 +2023/06/06 07:52:14 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_all2_lr1e-4_20230606_005614 +2023/06/06 07:52:14 - mmengine - INFO - Saving checkpoint at 9 epochs +2023/06/06 07:54:35 - mmengine - INFO - Epoch(val) [9][100/119] eta: 0:00:24 time: 1.2877 data_time: 0.0007 memory: 44140 +2023/06/06 07:55:17 - mmengine - INFO - Epoch(val) [9][119/119] accuracy/top1: 84.1517 single-label/precision_classwise: [95.78436279296875, 58.75566482543945] single-label/recall_classwise: [83.52587127685547, 86.45736694335938] single-label/f1-score_classwise: [89.236083984375, 69.96424102783203] data_time: 0.0184 time: 1.2860 +2023/06/06 07:56:24 - mmengine - INFO - Epoch(train) [10][ 100/4092] lr: 1.2098e-05 eta: 0:42:21 time: 0.6331 data_time: 0.0018 memory: 44140 loss: 0.4379 +2023/06/06 07:57:10 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_all2_lr1e-4_20230606_005614 +2023/06/06 07:57:28 - mmengine - INFO - Epoch(train) [10][ 200/4092] lr: 1.1995e-05 eta: 0:41:17 time: 0.6329 data_time: 0.0016 memory: 44140 loss: 0.4197 +2023/06/06 07:58:31 - mmengine - INFO - Epoch(train) [10][ 300/4092] lr: 1.1895e-05 eta: 0:40:13 time: 0.6335 data_time: 0.0018 memory: 44140 loss: 0.4558 +2023/06/06 07:59:34 - mmengine - INFO - Epoch(train) [10][ 400/4092] lr: 1.1797e-05 eta: 0:39:10 time: 0.6324 data_time: 0.0020 memory: 44140 loss: 0.4206 +2023/06/06 08:00:38 - mmengine - INFO - Epoch(train) [10][ 500/4092] lr: 1.1701e-05 eta: 0:38:06 time: 0.6353 data_time: 0.0016 memory: 44140 loss: 0.4137 +2023/06/06 08:01:41 - mmengine - INFO - Epoch(train) [10][ 600/4092] lr: 1.1608e-05 eta: 0:37:02 time: 0.6342 data_time: 0.0017 memory: 44140 loss: 0.4046 +2023/06/06 08:02:44 - mmengine - INFO - Epoch(train) [10][ 700/4092] lr: 1.1518e-05 eta: 0:35:59 time: 0.6338 data_time: 0.0021 memory: 44140 loss: 0.4356 +2023/06/06 08:03:48 - mmengine - INFO - Epoch(train) [10][ 800/4092] lr: 1.1430e-05 eta: 0:34:55 time: 0.6319 data_time: 0.0022 memory: 44140 loss: 0.4533 +2023/06/06 08:04:51 - mmengine - INFO - Epoch(train) [10][ 900/4092] lr: 1.1345e-05 eta: 0:33:51 time: 0.6314 data_time: 0.0023 memory: 44140 loss: 0.4262 +2023/06/06 08:05:54 - mmengine - INFO - Epoch(train) [10][1000/4092] lr: 1.1263e-05 eta: 0:32:48 time: 0.6451 data_time: 0.0017 memory: 44140 loss: 0.4054 +2023/06/06 08:06:58 - mmengine - INFO - Epoch(train) [10][1100/4092] lr: 1.1183e-05 eta: 0:31:44 time: 0.6340 data_time: 0.0026 memory: 44140 loss: 0.4339 +2023/06/06 08:07:43 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_all2_lr1e-4_20230606_005614 +2023/06/06 08:08:01 - mmengine - INFO - Epoch(train) [10][1200/4092] lr: 1.1105e-05 eta: 0:30:40 time: 0.6330 data_time: 0.0028 memory: 44140 loss: 0.4344 +2023/06/06 08:09:05 - mmengine - INFO - Epoch(train) [10][1300/4092] lr: 1.1031e-05 eta: 0:29:37 time: 0.6334 data_time: 0.0016 memory: 44140 loss: 0.4391 +2023/06/06 08:10:08 - mmengine - INFO - Epoch(train) [10][1400/4092] lr: 1.0958e-05 eta: 0:28:33 time: 0.6321 data_time: 0.0017 memory: 44140 loss: 0.4351 +2023/06/06 08:11:11 - mmengine - INFO - Epoch(train) [10][1500/4092] lr: 1.0889e-05 eta: 0:27:29 time: 0.6337 data_time: 0.0015 memory: 44140 loss: 0.4341 +2023/06/06 08:12:15 - mmengine - INFO - Epoch(train) [10][1600/4092] lr: 1.0822e-05 eta: 0:26:26 time: 0.6337 data_time: 0.0031 memory: 44140 loss: 0.4434 +2023/06/06 08:13:18 - mmengine - INFO - Epoch(train) [10][1700/4092] lr: 1.0757e-05 eta: 0:25:22 time: 0.6330 data_time: 0.0022 memory: 44140 loss: 0.4389 +2023/06/06 08:14:22 - mmengine - INFO - Epoch(train) [10][1800/4092] lr: 1.0696e-05 eta: 0:24:18 time: 0.6371 data_time: 0.0025 memory: 44140 loss: 0.4290 +2023/06/06 08:15:25 - mmengine - INFO - Epoch(train) [10][1900/4092] lr: 1.0636e-05 eta: 0:23:15 time: 0.6350 data_time: 0.0017 memory: 44140 loss: 0.4352 +2023/06/06 08:16:29 - mmengine - INFO - Epoch(train) [10][2000/4092] lr: 1.0580e-05 eta: 0:22:11 time: 0.6348 data_time: 0.0016 memory: 44140 loss: 0.4078 +2023/06/06 08:17:32 - mmengine - INFO - Epoch(train) [10][2100/4092] lr: 1.0526e-05 eta: 0:21:07 time: 0.6331 data_time: 0.0023 memory: 44140 loss: 0.4292 +2023/06/06 08:18:18 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_all2_lr1e-4_20230606_005614 +2023/06/06 08:18:35 - mmengine - INFO - Epoch(train) [10][2200/4092] lr: 1.0474e-05 eta: 0:20:04 time: 0.6333 data_time: 0.0014 memory: 44140 loss: 0.4246 +2023/06/06 08:19:39 - mmengine - INFO - Epoch(train) [10][2300/4092] lr: 1.0426e-05 eta: 0:19:00 time: 0.6339 data_time: 0.0019 memory: 44140 loss: 0.4188 +2023/06/06 08:20:42 - mmengine - INFO - Epoch(train) [10][2400/4092] lr: 1.0380e-05 eta: 0:17:56 time: 0.6328 data_time: 0.0021 memory: 44140 loss: 0.4077 +2023/06/06 08:21:45 - mmengine - INFO - Epoch(train) [10][2500/4092] lr: 1.0336e-05 eta: 0:16:53 time: 0.6336 data_time: 0.0017 memory: 44140 loss: 0.4142 +2023/06/06 08:22:49 - mmengine - INFO - Epoch(train) [10][2600/4092] lr: 1.0295e-05 eta: 0:15:49 time: 0.6315 data_time: 0.0023 memory: 44140 loss: 0.4428 +2023/06/06 08:23:53 - mmengine - INFO - Epoch(train) [10][2700/4092] lr: 1.0257e-05 eta: 0:14:45 time: 0.6380 data_time: 0.0024 memory: 44140 loss: 0.4394 +2023/06/06 08:24:56 - mmengine - INFO - Epoch(train) [10][2800/4092] lr: 1.0222e-05 eta: 0:13:42 time: 0.6329 data_time: 0.0027 memory: 44140 loss: 0.4387 +2023/06/06 08:26:00 - mmengine - INFO - Epoch(train) [10][2900/4092] lr: 1.0189e-05 eta: 0:12:38 time: 0.6328 data_time: 0.0022 memory: 44140 loss: 0.4340 +2023/06/06 08:27:03 - mmengine - INFO - Epoch(train) [10][3000/4092] lr: 1.0158e-05 eta: 0:11:34 time: 0.6348 data_time: 0.0015 memory: 44140 loss: 0.4061 +2023/06/06 08:28:06 - mmengine - INFO - Epoch(train) [10][3100/4092] lr: 1.0131e-05 eta: 0:10:31 time: 0.6341 data_time: 0.0016 memory: 44140 loss: 0.4352 +2023/06/06 08:28:52 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_all2_lr1e-4_20230606_005614 +2023/06/06 08:29:10 - mmengine - INFO - Epoch(train) [10][3200/4092] lr: 1.0106e-05 eta: 0:09:27 time: 0.6323 data_time: 0.0015 memory: 44140 loss: 0.4138 +2023/06/06 08:30:13 - mmengine - INFO - Epoch(train) [10][3300/4092] lr: 1.0083e-05 eta: 0:08:24 time: 0.6353 data_time: 0.0024 memory: 44140 loss: 0.4161 +2023/06/06 08:31:17 - mmengine - INFO - Epoch(train) [10][3400/4092] lr: 1.0064e-05 eta: 0:07:20 time: 0.6348 data_time: 0.0019 memory: 44140 loss: 0.4420 +2023/06/06 08:32:20 - mmengine - INFO - Epoch(train) [10][3500/4092] lr: 1.0047e-05 eta: 0:06:16 time: 0.6341 data_time: 0.0018 memory: 44140 loss: 0.4455 +2023/06/06 08:33:24 - mmengine - INFO - Epoch(train) [10][3600/4092] lr: 1.0032e-05 eta: 0:05:13 time: 0.6340 data_time: 0.0017 memory: 44140 loss: 0.4343 +2023/06/06 08:34:27 - mmengine - INFO - Epoch(train) [10][3700/4092] lr: 1.0020e-05 eta: 0:04:09 time: 0.6342 data_time: 0.0015 memory: 44140 loss: 0.4428 +2023/06/06 08:35:31 - mmengine - INFO - Epoch(train) [10][3800/4092] lr: 1.0011e-05 eta: 0:03:05 time: 0.6335 data_time: 0.0018 memory: 44140 loss: 0.4355 +2023/06/06 08:36:34 - mmengine - INFO - Epoch(train) [10][3900/4092] lr: 1.0005e-05 eta: 0:02:02 time: 0.6345 data_time: 0.0017 memory: 44140 loss: 0.4310 +2023/06/06 08:37:37 - mmengine - INFO - Epoch(train) [10][4000/4092] lr: 1.0001e-05 eta: 0:00:58 time: 0.6344 data_time: 0.0014 memory: 44140 loss: 0.4473 +2023/06/06 08:38:35 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_all2_lr1e-4_20230606_005614 +2023/06/06 08:38:35 - mmengine - INFO - Saving checkpoint at 10 epochs +2023/06/06 08:40:59 - mmengine - INFO - Epoch(val) [10][100/119] eta: 0:00:24 time: 1.2877 data_time: 0.0007 memory: 44140 +2023/06/06 08:41:41 - mmengine - INFO - Epoch(val) [10][119/119] accuracy/top1: 84.3503 single-label/precision_classwise: [95.57939910888672, 59.226314544677734] single-label/recall_classwise: [83.98670196533203, 85.6899185180664] single-label/f1-score_classwise: [89.40884399414062, 70.04181671142578] data_time: 0.0170 time: 1.2857 diff --git a/clip_large_pretrain_4x256_all2_lr1e-4/20230606_005614/vis_data/20230606_005614.json b/clip_large_pretrain_4x256_all2_lr1e-4/20230606_005614/vis_data/20230606_005614.json new file mode 100644 index 0000000000000000000000000000000000000000..812899d42e2b79b1769959aa303291642cfa8b68 --- /dev/null +++ b/clip_large_pretrain_4x256_all2_lr1e-4/20230606_005614/vis_data/20230606_005614.json @@ -0,0 +1,410 @@ +{"lr": 9.999870019168206e-05, "data_time": 0.0018175363540649414, "loss": 0.5742023468017579, "time": 0.6306015491485596, "epoch": 1, "memory": 44139, "step": 100} +{"lr": 9.99947481934738e-05, "data_time": 0.0018314838409423829, "loss": 0.5509862899780273, "time": 0.6331407070159912, "epoch": 1, "memory": 44139, "step": 200} +{"lr": 9.99881441056942e-05, "data_time": 0.002259683609008789, "loss": 0.5506836891174316, "time": 0.6333303213119507, "epoch": 1, "memory": 44139, "step": 300} +{"lr": 9.997888831760252e-05, "data_time": 0.0016779661178588866, "loss": 0.53231360912323, "time": 0.6325403451919556, "epoch": 1, "memory": 44139, "step": 400} +{"lr": 9.996698137475549e-05, "data_time": 0.0015641212463378905, "loss": 0.5343423008918762, "time": 0.6465238809585572, "epoch": 1, "memory": 44139, "step": 500} +{"lr": 9.99524239789746e-05, "data_time": 0.0014079809188842773, "loss": 0.4938073635101318, "time": 0.6329603433609009, "epoch": 1, "memory": 44139, "step": 600} +{"lr": 9.993521698830466e-05, "data_time": 0.001558542251586914, "loss": 0.5039564847946167, "time": 0.6353706121444702, "epoch": 1, "memory": 44139, "step": 700} +{"lr": 9.991536141696371e-05, "data_time": 0.0015775442123413086, "loss": 0.5006685137748719, "time": 0.6345899105072021, "epoch": 1, "memory": 44139, "step": 800} +{"lr": 9.989285843528274e-05, "data_time": 0.001490950584411621, "loss": 0.49552717208862307, "time": 0.6325678586959839, "epoch": 1, "memory": 44139, "step": 900} +{"lr": 9.986770936963713e-05, "data_time": 0.0014845848083496094, "loss": 0.497565770149231, "time": 0.6320618629455567, "epoch": 1, "memory": 44139, "step": 1000} +{"lr": 9.983991570236818e-05, "data_time": 0.001404738426208496, "loss": 0.5405094265937805, "time": 0.6333181858062744, "epoch": 1, "memory": 44139, "step": 1100} +{"lr": 9.980947907169598e-05, "data_time": 0.0014627456665039062, "loss": 0.47539167404174804, "time": 0.6362954616546631, "epoch": 1, "memory": 44139, "step": 1200} +{"lr": 9.977640127162247e-05, "data_time": 0.001424884796142578, "loss": 0.5027863502502441, "time": 0.6345264911651611, "epoch": 1, "memory": 44139, "step": 1300} +{"lr": 9.974068425182603e-05, "data_time": 0.001611638069152832, "loss": 0.4923910140991211, "time": 0.6314883708953858, "epoch": 1, "memory": 44139, "step": 1400} +{"lr": 9.970233011754657e-05, "data_time": 0.001566004753112793, "loss": 0.4950839042663574, "time": 0.6319402694702149, "epoch": 1, "memory": 44139, "step": 1500} +{"lr": 9.966134112946122e-05, "data_time": 0.0013825416564941407, "loss": 0.5086468577384948, "time": 0.631356167793274, "epoch": 1, "memory": 44139, "step": 1600} +{"lr": 9.961771970355126e-05, "data_time": 0.0013284921646118165, "loss": 0.4961442410945892, "time": 0.6313036203384399, "epoch": 1, "memory": 44139, "step": 1700} +{"lr": 9.957146841095961e-05, "data_time": 0.0014630556106567383, "loss": 0.48642589449882506, "time": 0.6314688682556152, "epoch": 1, "memory": 44139, "step": 1800} +{"lr": 9.952258997783933e-05, "data_time": 0.0013856887817382812, "loss": 0.4863204836845398, "time": 0.6308996438980102, "epoch": 1, "memory": 44139, "step": 1900} +{"lr": 9.947108728519288e-05, "data_time": 0.001604771614074707, "loss": 0.48236514925956725, "time": 0.631430983543396, "epoch": 1, "memory": 44139, "step": 2000} +{"lr": 9.941696336870245e-05, "data_time": 0.001503586769104004, "loss": 0.5119886755943298, "time": 0.631052041053772, "epoch": 1, "memory": 44139, "step": 2100} +{"lr": 9.936022141855094e-05, "data_time": 0.00168154239654541, "loss": 0.4707063287496567, "time": 0.6316580533981323, "epoch": 1, "memory": 44139, "step": 2200} +{"lr": 9.93008647792338e-05, "data_time": 0.001444864273071289, "loss": 0.49044070243835447, "time": 0.6314844846725464, "epoch": 1, "memory": 44139, "step": 2300} +{"lr": 9.923889694936212e-05, "data_time": 0.0014310121536254884, "loss": 0.4934364855289459, "time": 0.6319224596023559, "epoch": 1, "memory": 44139, "step": 2400} +{"lr": 9.917432158145639e-05, "data_time": 0.0014138936996459961, "loss": 0.5017586350440979, "time": 0.6312256097793579, "epoch": 1, "memory": 44139, "step": 2500} +{"lr": 9.91071424817309e-05, "data_time": 0.0014225006103515624, "loss": 0.4845857560634613, "time": 0.6314415693283081, "epoch": 1, "memory": 44139, "step": 2600} +{"lr": 9.903736360986979e-05, "data_time": 0.0014866352081298827, "loss": 0.4840073823928833, "time": 0.6320632696151733, "epoch": 1, "memory": 44139, "step": 2700} +{"lr": 9.89649890787935e-05, "data_time": 0.0014931678771972657, "loss": 0.4953353703022003, "time": 0.6288543939590454, "epoch": 1, "memory": 44139, "step": 2800} +{"lr": 9.889002315441642e-05, "data_time": 0.0013036489486694335, "loss": 0.51003999710083, "time": 0.6310796737670898, "epoch": 1, "memory": 44139, "step": 2900} +{"lr": 9.881247025539515e-05, "data_time": 0.0013866901397705078, "loss": 0.45273745656013487, "time": 0.6315259456634521, "epoch": 1, "memory": 44139, "step": 3000} +{"lr": 9.873233495286844e-05, "data_time": 0.0014451265335083008, "loss": 0.47733883559703827, "time": 0.6317728757858276, "epoch": 1, "memory": 44139, "step": 3100} +{"lr": 9.864962197018756e-05, "data_time": 0.0014280557632446289, "loss": 0.4770282328128815, "time": 0.631502914428711, "epoch": 1, "memory": 44139, "step": 3200} +{"lr": 9.856433618263779e-05, "data_time": 0.0016320705413818359, "loss": 0.45885944068431855, "time": 0.6312127113342285, "epoch": 1, "memory": 44139, "step": 3300} +{"lr": 9.847648261715132e-05, "data_time": 0.0015032291412353516, "loss": 0.4754169762134552, "time": 0.6317023038864136, "epoch": 1, "memory": 44139, "step": 3400} +{"lr": 9.838606645201093e-05, "data_time": 0.0014789342880249024, "loss": 0.46152358651161196, "time": 0.6314425945281983, "epoch": 1, "memory": 44139, "step": 3500} +{"lr": 9.829309301654456e-05, "data_time": 0.0014422178268432618, "loss": 0.45579715967178347, "time": 0.6320044040679932, "epoch": 1, "memory": 44139, "step": 3600} +{"lr": 9.819756779081127e-05, "data_time": 0.0013558149337768554, "loss": 0.47395158410072324, "time": 0.6315643548965454, "epoch": 1, "memory": 44139, "step": 3700} +{"lr": 9.809949640527814e-05, "data_time": 0.001401662826538086, "loss": 0.4785751700401306, "time": 0.6320651054382325, "epoch": 1, "memory": 44139, "step": 3800} +{"lr": 9.799888464048879e-05, "data_time": 0.0013675689697265625, "loss": 0.445124951004982, "time": 0.6320790767669677, "epoch": 1, "memory": 44139, "step": 3900} +{"lr": 9.789573842672242e-05, "data_time": 0.0013480424880981446, "loss": 0.4624592661857605, "time": 0.6311359167098999, "epoch": 1, "memory": 44139, "step": 4000} +{"accuracy/top1": 71.7806167602539, "data_time": 0.022229370950650768, "time": 1.3160906679490034, "step": 1} +{"lr": 9.7690615523361e-05, "data_time": 0.0018679141998291016, "loss": 0.44359633326530457, "time": 0.6317744970321655, "epoch": 2, "memory": 44140, "step": 4192} +{"lr": 9.758010405927656e-05, "data_time": 0.0013559103012084962, "loss": 0.47134405970573423, "time": 0.6318446636199951, "epoch": 2, "memory": 44140, "step": 4292} +{"lr": 9.746708283006492e-05, "data_time": 0.001389288902282715, "loss": 0.47403521537780763, "time": 0.6340468406677247, "epoch": 2, "memory": 44140, "step": 4392} +{"lr": 9.735155849744634e-05, "data_time": 0.0013835430145263672, "loss": 0.4599327057600021, "time": 0.6323191881179809, "epoch": 2, "memory": 44140, "step": 4492} +{"lr": 9.723353787067928e-05, "data_time": 0.0014330148696899414, "loss": 0.44250693917274475, "time": 0.6345848321914673, "epoch": 2, "memory": 44140, "step": 4592} +{"lr": 9.711302790615951e-05, "data_time": 0.0013781070709228515, "loss": 0.4755029261112213, "time": 0.631804084777832, "epoch": 2, "memory": 44140, "step": 4692} +{"lr": 9.699003570700986e-05, "data_time": 0.0013843059539794921, "loss": 0.4483084142208099, "time": 0.631246280670166, "epoch": 2, "memory": 44140, "step": 4792} +{"lr": 9.686456852266142e-05, "data_time": 0.001350736618041992, "loss": 0.4435199677944183, "time": 0.6318500995635986, "epoch": 2, "memory": 44140, "step": 4892} +{"lr": 9.673663374842644e-05, "data_time": 0.0013036489486694335, "loss": 0.45903160572052004, "time": 0.6315513134002686, "epoch": 2, "memory": 44140, "step": 4992} +{"lr": 9.660623892506218e-05, "data_time": 0.0013495445251464843, "loss": 0.45409590005874634, "time": 0.6326237201690674, "epoch": 2, "memory": 44140, "step": 5092} +{"lr": 9.647339173832693e-05, "data_time": 0.001483607292175293, "loss": 0.46190900802612306, "time": 0.6318344831466675, "epoch": 2, "memory": 44140, "step": 5192} +{"lr": 9.633810001852653e-05, "data_time": 0.0013091564178466797, "loss": 0.4453065633773804, "time": 0.6312918424606323, "epoch": 2, "memory": 44140, "step": 5292} +{"lr": 9.620037174005274e-05, "data_time": 0.0016185283660888673, "loss": 0.45716494917869566, "time": 0.631353211402893, "epoch": 2, "memory": 44140, "step": 5392} +{"lr": 9.606021502091389e-05, "data_time": 0.0014839649200439453, "loss": 0.49206840991973877, "time": 0.6333415746688843, "epoch": 2, "memory": 44140, "step": 5492} +{"lr": 9.591763812225569e-05, "data_time": 0.001655721664428711, "loss": 0.4525274097919464, "time": 0.6328972339630127, "epoch": 2, "memory": 44140, "step": 5592} +{"lr": 9.577264944787459e-05, "data_time": 0.0015350818634033204, "loss": 0.46427778005599973, "time": 0.6310419559478759, "epoch": 2, "memory": 44140, "step": 5692} +{"lr": 9.562525754372252e-05, "data_time": 0.0014998674392700194, "loss": 0.4221505582332611, "time": 0.6318845987319947, "epoch": 2, "memory": 44140, "step": 5792} +{"lr": 9.547547109740282e-05, "data_time": 0.0013521671295166015, "loss": 0.4527347683906555, "time": 0.631903076171875, "epoch": 2, "memory": 44140, "step": 5892} +{"lr": 9.532329893765858e-05, "data_time": 0.0015967845916748046, "loss": 0.446257284283638, "time": 0.6303370714187622, "epoch": 2, "memory": 44140, "step": 5992} +{"lr": 9.516875003385223e-05, "data_time": 0.0014877080917358398, "loss": 0.42972763180732726, "time": 0.6328753232955933, "epoch": 2, "memory": 44140, "step": 6092} +{"lr": 9.501183349543653e-05, "data_time": 0.0015574216842651368, "loss": 0.4281292766332626, "time": 0.6330648899078369, "epoch": 2, "memory": 44140, "step": 6192} +{"lr": 9.48525585714179e-05, "data_time": 0.0014125347137451173, "loss": 0.44764353334903717, "time": 0.6313239097595215, "epoch": 2, "memory": 44140, "step": 6292} +{"lr": 9.469093464981119e-05, "data_time": 0.001408553123474121, "loss": 0.43747061491012573, "time": 0.6313658714294433, "epoch": 2, "memory": 44140, "step": 6392} +{"lr": 9.452697125708636e-05, "data_time": 0.0015063762664794921, "loss": 0.44163825511932375, "time": 0.6310720682144165, "epoch": 2, "memory": 44140, "step": 6492} +{"lr": 9.436067805760705e-05, "data_time": 0.001448678970336914, "loss": 0.4352395623922348, "time": 0.6314118862152099, "epoch": 2, "memory": 44140, "step": 6592} +{"lr": 9.419206485306072e-05, "data_time": 0.0014289617538452148, "loss": 0.4278436750173569, "time": 0.6314183235168457, "epoch": 2, "memory": 44140, "step": 6692} +{"lr": 9.4021141581881e-05, "data_time": 0.0013917922973632813, "loss": 0.4349859803915024, "time": 0.6318498373031616, "epoch": 2, "memory": 44140, "step": 6792} +{"lr": 9.384791831866193e-05, "data_time": 0.0014430999755859375, "loss": 0.42788690626621245, "time": 0.6315459489822388, "epoch": 2, "memory": 44140, "step": 6892} +{"lr": 9.367240527356428e-05, "data_time": 0.0014315128326416015, "loss": 0.4437686026096344, "time": 0.6311376094818115, "epoch": 2, "memory": 44140, "step": 6992} +{"lr": 9.349461279171362e-05, "data_time": 0.0014157533645629884, "loss": 0.44617689549922945, "time": 0.6320367097854614, "epoch": 2, "memory": 44140, "step": 7092} +{"lr": 9.33145513525904e-05, "data_time": 0.0013987064361572266, "loss": 0.4393774211406708, "time": 0.6314810991287232, "epoch": 2, "memory": 44140, "step": 7192} +{"lr": 9.313223156941242e-05, "data_time": 0.0013824701309204102, "loss": 0.44216046333312986, "time": 0.6343540191650391, "epoch": 2, "memory": 44140, "step": 7292} +{"lr": 9.294766418850942e-05, "data_time": 0.001382756233215332, "loss": 0.4563088804483414, "time": 0.6314673900604248, "epoch": 2, "memory": 44140, "step": 7392} +{"lr": 9.276086008868929e-05, "data_time": 0.00142822265625, "loss": 0.4428665488958359, "time": 0.6312015771865844, "epoch": 2, "memory": 44140, "step": 7492} +{"lr": 9.257183028059726e-05, "data_time": 0.0016424417495727538, "loss": 0.4485490798950195, "time": 0.641039776802063, "epoch": 2, "memory": 44140, "step": 7592} +{"lr": 9.238058590606654e-05, "data_time": 0.0015211105346679688, "loss": 0.41403867602348327, "time": 0.6308777332305908, "epoch": 2, "memory": 44140, "step": 7692} +{"lr": 9.218713823746189e-05, "data_time": 0.0014520406723022461, "loss": 0.4221742510795593, "time": 0.6324946165084839, "epoch": 2, "memory": 44140, "step": 7792} +{"lr": 9.199149867701492e-05, "data_time": 0.0014151573181152345, "loss": 0.46067075431346893, "time": 0.6320663690567017, "epoch": 2, "memory": 44140, "step": 7892} +{"lr": 9.179367875615232e-05, "data_time": 0.0014046192169189452, "loss": 0.46722287535667417, "time": 0.632951807975769, "epoch": 2, "memory": 44140, "step": 7992} +{"lr": 9.159369013481585e-05, "data_time": 0.0014615058898925781, "loss": 0.42448204159736636, "time": 0.6318077325820923, "epoch": 2, "memory": 44140, "step": 8092} +{"accuracy/top1": 72.94573211669922, "data_time": 0.017711853981018065, "time": 1.2847278495629628, "step": 2} +{"lr": 9.120367596460475e-05, "data_time": 0.0014551162719726562, "loss": 0.4473511815071106, "time": 0.6332897424697876, "epoch": 3, "memory": 44140, "step": 8284} +{"lr": 9.099742266672751e-05, "data_time": 0.0016311883926391601, "loss": 0.4212220788002014, "time": 0.6335990190505981, "epoch": 3, "memory": 44140, "step": 8384} +{"lr": 9.078904760146445e-05, "data_time": 0.0013982772827148438, "loss": 0.4284248322248459, "time": 0.634125804901123, "epoch": 3, "memory": 44140, "step": 8484} +{"lr": 9.057856305090082e-05, "data_time": 0.0013855457305908202, "loss": 0.4249072462320328, "time": 0.6312395095825195, "epoch": 3, "memory": 44140, "step": 8584} +{"lr": 9.036598142145979e-05, "data_time": 0.0014184236526489258, "loss": 0.42530823945999147, "time": 0.6337924242019654, "epoch": 3, "memory": 44140, "step": 8684} +{"lr": 9.015131524317079e-05, "data_time": 0.001400160789489746, "loss": 0.4409077078104019, "time": 0.6319032192230225, "epoch": 3, "memory": 44140, "step": 8784} +{"lr": 8.993457716893124e-05, "data_time": 0.001379108428955078, "loss": 0.423635196685791, "time": 0.6323798179626465, "epoch": 3, "memory": 44140, "step": 8884} +{"lr": 8.971577997376067e-05, "data_time": 0.0013748884201049804, "loss": 0.4090465158224106, "time": 0.6361242771148682, "epoch": 3, "memory": 44140, "step": 8984} +{"lr": 8.949493655404777e-05, "data_time": 0.0016320228576660156, "loss": 0.42824044823646545, "time": 0.6332427024841308, "epoch": 3, "memory": 44140, "step": 9084} +{"lr": 8.927205992679013e-05, "data_time": 0.0016265153884887696, "loss": 0.4245839774608612, "time": 0.6346089363098144, "epoch": 3, "memory": 44140, "step": 9184} +{"lr": 8.9047163228827e-05, "data_time": 0.0014111757278442382, "loss": 0.4436920195817947, "time": 0.6333080291748047, "epoch": 3, "memory": 44140, "step": 9284} +{"lr": 8.88202597160652e-05, "data_time": 0.0013289690017700196, "loss": 0.4244053095579147, "time": 0.6321623802185059, "epoch": 3, "memory": 44140, "step": 9384} +{"lr": 8.859136276269775e-05, "data_time": 0.0013959169387817382, "loss": 0.4091134428977966, "time": 0.6338141441345215, "epoch": 3, "memory": 44140, "step": 9484} +{"lr": 8.836048586041507e-05, "data_time": 0.0014384031295776368, "loss": 0.4392902314662933, "time": 0.6321201086044311, "epoch": 3, "memory": 44140, "step": 9584} +{"lr": 8.812764261761037e-05, "data_time": 0.0013388633728027345, "loss": 0.4246595323085785, "time": 0.6321587562561035, "epoch": 3, "memory": 44140, "step": 9684} +{"lr": 8.789284675857743e-05, "data_time": 0.001620030403137207, "loss": 0.4354802370071411, "time": 0.6323515415191651, "epoch": 3, "memory": 44140, "step": 9784} +{"lr": 8.76561121227013e-05, "data_time": 0.0015564680099487305, "loss": 0.43429321944713595, "time": 0.6408982276916504, "epoch": 3, "memory": 44140, "step": 9884} +{"lr": 8.741745266364282e-05, "data_time": 0.0015223503112792968, "loss": 0.4274744212627411, "time": 0.6323494911193848, "epoch": 3, "memory": 44140, "step": 9984} +{"lr": 8.717688244851635e-05, "data_time": 0.0016958475112915038, "loss": 0.4182525217533112, "time": 0.6326273918151856, "epoch": 3, "memory": 44140, "step": 10084} +{"lr": 8.693441565706024e-05, "data_time": 0.0014996767044067384, "loss": 0.41954270005226135, "time": 0.6322771549224854, "epoch": 3, "memory": 44140, "step": 10184} +{"lr": 8.669006658080132e-05, "data_time": 0.0015003442764282226, "loss": 0.4252572238445282, "time": 0.6315906286239624, "epoch": 3, "memory": 44140, "step": 10284} +{"lr": 8.644384962221226e-05, "data_time": 0.001618814468383789, "loss": 0.4026562452316284, "time": 0.6322749614715576, "epoch": 3, "memory": 44140, "step": 10384} +{"lr": 8.619577929386303e-05, "data_time": 0.0014399051666259765, "loss": 0.4505779087543488, "time": 0.6323697805404663, "epoch": 3, "memory": 44140, "step": 10484} +{"lr": 8.594587021756517e-05, "data_time": 0.0014118909835815429, "loss": 0.43633814454078673, "time": 0.6322449445724487, "epoch": 3, "memory": 44140, "step": 10584} +{"lr": 8.569413712351023e-05, "data_time": 0.0015891551971435546, "loss": 0.4132749974727631, "time": 0.6337039709091187, "epoch": 3, "memory": 44140, "step": 10684} +{"lr": 8.544059484940119e-05, "data_time": 0.0014904499053955077, "loss": 0.4413173168897629, "time": 0.6329502582550048, "epoch": 3, "memory": 44140, "step": 10784} +{"lr": 8.518525833957818e-05, "data_time": 0.0013824939727783204, "loss": 0.4279088795185089, "time": 0.63317391872406, "epoch": 3, "memory": 44140, "step": 10884} +{"lr": 8.49281426441377e-05, "data_time": 0.0014138221740722656, "loss": 0.4370129227638245, "time": 0.6330703735351563, "epoch": 3, "memory": 44140, "step": 10984} +{"lr": 8.466926291804501e-05, "data_time": 0.0012787818908691407, "loss": 0.43694463074207307, "time": 0.6357486248016357, "epoch": 3, "memory": 44140, "step": 11084} +{"lr": 8.440863442024147e-05, "data_time": 0.0017200469970703124, "loss": 0.4132837414741516, "time": 0.6355113744735718, "epoch": 3, "memory": 44140, "step": 11184} +{"lr": 8.414627251274488e-05, "data_time": 0.0015549659729003906, "loss": 0.42699059247970583, "time": 0.6320959329605103, "epoch": 3, "memory": 44140, "step": 11284} +{"lr": 8.388219265974407e-05, "data_time": 0.001401805877685547, "loss": 0.43332897424697875, "time": 0.6318453788757324, "epoch": 3, "memory": 44140, "step": 11384} +{"lr": 8.361641042668715e-05, "data_time": 0.001779031753540039, "loss": 0.4249531179666519, "time": 0.6340765237808228, "epoch": 3, "memory": 44140, "step": 11484} +{"lr": 8.33489414793644e-05, "data_time": 0.0021561384201049805, "loss": 0.4533049076795578, "time": 0.6464383602142334, "epoch": 3, "memory": 44140, "step": 11584} +{"lr": 8.30798015829848e-05, "data_time": 0.0023379802703857424, "loss": 0.4502921670675278, "time": 0.6342859506607056, "epoch": 3, "memory": 44140, "step": 11684} +{"lr": 8.280900660124665e-05, "data_time": 0.002496504783630371, "loss": 0.4467496007680893, "time": 0.6343088388442993, "epoch": 3, "memory": 44140, "step": 11784} +{"lr": 8.253657249540274e-05, "data_time": 0.0016772985458374024, "loss": 0.4151320159435272, "time": 0.6328318119049072, "epoch": 3, "memory": 44140, "step": 11884} +{"lr": 8.226251532331926e-05, "data_time": 0.0020305871963500976, "loss": 0.4178690493106842, "time": 0.6329146146774292, "epoch": 3, "memory": 44140, "step": 11984} +{"lr": 8.198685123852965e-05, "data_time": 0.0024167537689208985, "loss": 0.4292076587677002, "time": 0.6334588766098023, "epoch": 3, "memory": 44140, "step": 12084} +{"lr": 8.17095964892824e-05, "data_time": 0.0023377656936645506, "loss": 0.42308293879032133, "time": 0.6408001184463501, "epoch": 3, "memory": 44140, "step": 12184} +{"accuracy/top1": 75.0475845336914, "data_time": 0.020047646760940552, "time": 1.2880093296368917, "step": 3} +{"lr": 8.11728683565904e-05, "data_time": 0.0024878978729248047, "loss": 0.41675683856010437, "time": 0.6322898387908935, "epoch": 4, "memory": 44140, "step": 12376} +{"lr": 8.089106273476776e-05, "data_time": 0.00207364559173584, "loss": 0.4293960154056549, "time": 0.6331905603408814, "epoch": 4, "memory": 44140, "step": 12476} +{"lr": 8.060773103667566e-05, "data_time": 0.0024184226989746095, "loss": 0.42023499608039855, "time": 0.6328914403915405, "epoch": 4, "memory": 44140, "step": 12576} +{"lr": 8.03228899625083e-05, "data_time": 0.002521252632141113, "loss": 0.38831470608711244, "time": 0.6329829216003418, "epoch": 4, "memory": 44140, "step": 12676} +{"lr": 8.003655630142627e-05, "data_time": 0.0024059534072875975, "loss": 0.41458570659160615, "time": 0.6332283258438111, "epoch": 4, "memory": 44140, "step": 12776} +{"lr": 7.974874693056617e-05, "data_time": 0.002190995216369629, "loss": 0.42189861536026, "time": 0.6314792394638061, "epoch": 4, "memory": 44140, "step": 12876} +{"lr": 7.945947881404618e-05, "data_time": 0.002054762840270996, "loss": 0.43414068818092344, "time": 0.6348303318023681, "epoch": 4, "memory": 44140, "step": 12976} +{"lr": 7.916876900196646e-05, "data_time": 0.0019904375076293945, "loss": 0.4551405757665634, "time": 0.6324449777603149, "epoch": 4, "memory": 44140, "step": 13076} +{"lr": 7.887663462940359e-05, "data_time": 0.0018481731414794922, "loss": 0.4363859176635742, "time": 0.6329013347625733, "epoch": 4, "memory": 44140, "step": 13176} +{"lr": 7.858309291540111e-05, "data_time": 0.0019230127334594726, "loss": 0.4265198588371277, "time": 0.6333821535110473, "epoch": 4, "memory": 44140, "step": 13276} +{"lr": 7.828816116195411e-05, "data_time": 0.0022130727767944334, "loss": 0.4443864405155182, "time": 0.6324852228164672, "epoch": 4, "memory": 44140, "step": 13376} +{"lr": 7.799185675298994e-05, "data_time": 0.002134275436401367, "loss": 0.4278457522392273, "time": 0.6307125806808471, "epoch": 4, "memory": 44140, "step": 13476} +{"lr": 7.769419715334297e-05, "data_time": 0.0019055843353271485, "loss": 0.4094464063644409, "time": 0.6339537620544433, "epoch": 4, "memory": 44140, "step": 13576} +{"lr": 7.739519990772583e-05, "data_time": 0.0023879289627075197, "loss": 0.43034936487674713, "time": 0.6329025745391845, "epoch": 4, "memory": 44140, "step": 13676} +{"lr": 7.709488263969475e-05, "data_time": 0.0020322322845458983, "loss": 0.4365843415260315, "time": 0.6321660041809082, "epoch": 4, "memory": 44140, "step": 13776} +{"lr": 7.67932630506109e-05, "data_time": 0.002697300910949707, "loss": 0.4132125288248062, "time": 0.6336849689483642, "epoch": 4, "memory": 44140, "step": 13876} +{"lr": 7.64903589185975e-05, "data_time": 0.0019088029861450196, "loss": 0.4382692277431488, "time": 0.6323899984359741, "epoch": 4, "memory": 44140, "step": 13976} +{"lr": 7.618618809749134e-05, "data_time": 0.0022614240646362306, "loss": 0.42109595239162445, "time": 0.6325977802276611, "epoch": 4, "memory": 44140, "step": 14076} +{"lr": 7.588076851579028e-05, "data_time": 0.0023856401443481446, "loss": 0.43642314672470095, "time": 0.6321729421615601, "epoch": 4, "memory": 44140, "step": 14176} +{"lr": 7.557411817559755e-05, "data_time": 0.002448272705078125, "loss": 0.4302003622055054, "time": 0.6361776828765869, "epoch": 4, "memory": 44140, "step": 14276} +{"lr": 7.526625515155957e-05, "data_time": 0.0019052743911743164, "loss": 0.4075252503156662, "time": 0.6330398082733154, "epoch": 4, "memory": 44140, "step": 14376} +{"lr": 7.495719758980122e-05, "data_time": 0.0024227142333984376, "loss": 0.40171490609645844, "time": 0.6321627378463746, "epoch": 4, "memory": 44140, "step": 14476} +{"lr": 7.464696370685586e-05, "data_time": 0.0031921863555908203, "loss": 0.42775468826293944, "time": 0.6336983203887939, "epoch": 4, "memory": 44140, "step": 14576} +{"lr": 7.433557178859198e-05, "data_time": 0.0019622087478637696, "loss": 0.44475115835666656, "time": 0.6332455396652221, "epoch": 4, "memory": 44140, "step": 14676} +{"lr": 7.402304018913512e-05, "data_time": 0.0018522500991821288, "loss": 0.41916805803775786, "time": 0.6318212509155273, "epoch": 4, "memory": 44140, "step": 14776} +{"lr": 7.37093873297861e-05, "data_time": 0.0019346237182617187, "loss": 0.4231825232505798, "time": 0.6315581321716308, "epoch": 4, "memory": 44140, "step": 14876} +{"lr": 7.339463169793563e-05, "data_time": 0.002269434928894043, "loss": 0.4665799081325531, "time": 0.6322512626647949, "epoch": 4, "memory": 44140, "step": 14976} +{"lr": 7.307879184597369e-05, "data_time": 0.002082633972167969, "loss": 0.43822734951972964, "time": 0.6315053939819336, "epoch": 4, "memory": 44140, "step": 15076} +{"lr": 7.276188639019704e-05, "data_time": 0.0024214267730712892, "loss": 0.38338116407394407, "time": 0.6320795059204102, "epoch": 4, "memory": 44140, "step": 15176} +{"lr": 7.244393400971127e-05, "data_time": 0.003145432472229004, "loss": 0.4197797864675522, "time": 0.6340341806411743, "epoch": 4, "memory": 44140, "step": 15276} +{"lr": 7.212495344533006e-05, "data_time": 0.00262296199798584, "loss": 0.43974902033805846, "time": 0.6334179639816284, "epoch": 4, "memory": 44140, "step": 15376} +{"lr": 7.180496349847055e-05, "data_time": 0.0020112991333007812, "loss": 0.4346621215343475, "time": 0.6322306871414185, "epoch": 4, "memory": 44140, "step": 15476} +{"lr": 7.148398303004499e-05, "data_time": 0.0026148319244384765, "loss": 0.4178128480911255, "time": 0.6347824096679687, "epoch": 4, "memory": 44140, "step": 15576} +{"lr": 7.11620309593493e-05, "data_time": 0.0018913030624389648, "loss": 0.44328099191188813, "time": 0.6322627544403077, "epoch": 4, "memory": 44140, "step": 15676} +{"lr": 7.083912626294795e-05, "data_time": 0.0023617982864379884, "loss": 0.4336129128932953, "time": 0.6343975782394409, "epoch": 4, "memory": 44140, "step": 15776} +{"lr": 7.051528797355494e-05, "data_time": 0.0022262811660766603, "loss": 0.45349820256233214, "time": 0.6330487251281738, "epoch": 4, "memory": 44140, "step": 15876} +{"lr": 7.019053517891214e-05, "data_time": 0.002340221405029297, "loss": 0.42756457924842833, "time": 0.6318708181381225, "epoch": 4, "memory": 44140, "step": 15976} +{"lr": 6.986488702066473e-05, "data_time": 0.0021955490112304686, "loss": 0.4229933112859726, "time": 0.6330044269561768, "epoch": 4, "memory": 44140, "step": 16076} +{"lr": 6.953836269323245e-05, "data_time": 0.0019453763961791992, "loss": 0.441475448012352, "time": 0.6325273275375366, "epoch": 4, "memory": 44140, "step": 16176} +{"lr": 6.921098144267838e-05, "data_time": 0.002437734603881836, "loss": 0.4169649541378021, "time": 0.6327932834625244, "epoch": 4, "memory": 44140, "step": 16276} +{"accuracy/top1": 77.19245910644531, "data_time": 0.01820972760518392, "time": 1.2874916394551594, "step": 4} +{"lr": 6.858007803673889e-05, "data_time": 0.0017416238784790038, "loss": 0.44968058466911315, "time": 0.6320735692977906, "epoch": 5, "memory": 44140, "step": 16468} +{"lr": 6.825030518869046e-05, "data_time": 0.0017281532287597655, "loss": 0.4209025502204895, "time": 0.6332170486450195, "epoch": 5, "memory": 44140, "step": 16568} +{"lr": 6.791975133846686e-05, "data_time": 0.0021192312240600588, "loss": 0.4412855625152588, "time": 0.6322494268417358, "epoch": 5, "memory": 44140, "step": 16668} +{"lr": 6.758843596964013e-05, "data_time": 0.002190709114074707, "loss": 0.4384043008089066, "time": 0.632795763015747, "epoch": 5, "memory": 44140, "step": 16768} +{"lr": 6.725637861066804e-05, "data_time": 0.002461957931518555, "loss": 0.43394399881362916, "time": 0.6330869674682618, "epoch": 5, "memory": 44140, "step": 16868} +{"lr": 6.692359883374264e-05, "data_time": 0.0015211820602416993, "loss": 0.44528878331184385, "time": 0.6321467399597168, "epoch": 5, "memory": 44140, "step": 16968} +{"lr": 6.659011625363718e-05, "data_time": 0.00165865421295166, "loss": 0.4454505920410156, "time": 0.6323389768600464, "epoch": 5, "memory": 44140, "step": 17068} +{"lr": 6.625595052654927e-05, "data_time": 0.0014862775802612304, "loss": 0.41868581175804137, "time": 0.6319795608520508, "epoch": 5, "memory": 44140, "step": 17168} +{"lr": 6.592112134894315e-05, "data_time": 0.0015000104904174805, "loss": 0.41661250591278076, "time": 0.6344560384750366, "epoch": 5, "memory": 44140, "step": 17268} +{"lr": 6.558564845638802e-05, "data_time": 0.0024153947830200194, "loss": 0.41712948083877566, "time": 0.6343169927597045, "epoch": 5, "memory": 44140, "step": 17368} +{"lr": 6.524955162239518e-05, "data_time": 0.0014957666397094726, "loss": 0.4421685069799423, "time": 0.6326616764068603, "epoch": 5, "memory": 44140, "step": 17468} +{"lr": 6.491285065725238e-05, "data_time": 0.0013540029525756837, "loss": 0.46646787524223327, "time": 0.6323060274124146, "epoch": 5, "memory": 44140, "step": 17568} +{"lr": 6.457556540685618e-05, "data_time": 0.0017585277557373047, "loss": 0.4295129120349884, "time": 0.6341775417327881, "epoch": 5, "memory": 44140, "step": 17668} +{"lr": 6.423771575154208e-05, "data_time": 0.0014645576477050782, "loss": 0.4353230714797974, "time": 0.6336427927017212, "epoch": 5, "memory": 44140, "step": 17768} +{"lr": 6.389932160491309e-05, "data_time": 0.0017275810241699219, "loss": 0.41458328664302824, "time": 0.632085132598877, "epoch": 5, "memory": 44140, "step": 17868} +{"lr": 6.356040291266556e-05, "data_time": 0.0014629364013671875, "loss": 0.45556470155715945, "time": 0.6318282127380371, "epoch": 5, "memory": 44140, "step": 17968} +{"lr": 6.322097965141391e-05, "data_time": 0.0014632463455200196, "loss": 0.4440807819366455, "time": 0.6336803913116456, "epoch": 5, "memory": 44140, "step": 18068} +{"lr": 6.288107182751267e-05, "data_time": 0.0014612913131713868, "loss": 0.40304665863513944, "time": 0.6333890676498413, "epoch": 5, "memory": 44140, "step": 18168} +{"lr": 6.254069947587782e-05, "data_time": 0.0014941930770874024, "loss": 0.45084108710289, "time": 0.6324883222579956, "epoch": 5, "memory": 44140, "step": 18268} +{"lr": 6.219988265880537e-05, "data_time": 0.0015841484069824218, "loss": 0.41318759620189666, "time": 0.6371341943740845, "epoch": 5, "memory": 44140, "step": 18368} +{"lr": 6.185864146478941e-05, "data_time": 0.0017355918884277345, "loss": 0.46838268637657166, "time": 0.6336296319961547, "epoch": 5, "memory": 44140, "step": 18468} +{"lr": 6.151699600733752e-05, "data_time": 0.0019960641860961915, "loss": 0.40581959784030913, "time": 0.635711407661438, "epoch": 5, "memory": 44140, "step": 18568} +{"lr": 6.117496642378554e-05, "data_time": 0.001602506637573242, "loss": 0.44728600382804873, "time": 0.6328262090682983, "epoch": 5, "memory": 44140, "step": 18668} +{"lr": 6.0832572874110315e-05, "data_time": 0.0016323089599609374, "loss": 0.43281572461128237, "time": 0.6335597276687622, "epoch": 5, "memory": 44140, "step": 18768} +{"lr": 6.0489835539741924e-05, "data_time": 0.0014596939086914062, "loss": 0.43951316475868224, "time": 0.6353926420211792, "epoch": 5, "memory": 44140, "step": 18868} +{"lr": 6.0146774622374006e-05, "data_time": 0.0015111684799194336, "loss": 0.4285846263170242, "time": 0.6356100082397461, "epoch": 5, "memory": 44140, "step": 18968} +{"lr": 5.980341034277274e-05, "data_time": 0.0012971162796020508, "loss": 0.4076012700796127, "time": 0.6328089952468872, "epoch": 5, "memory": 44140, "step": 19068} +{"lr": 5.945976293958505e-05, "data_time": 0.0015245914459228516, "loss": 0.4158323615789413, "time": 0.6349369525909424, "epoch": 5, "memory": 44140, "step": 19168} +{"lr": 5.911585266814587e-05, "data_time": 0.0013681650161743164, "loss": 0.44133509397506715, "time": 0.6346670150756836, "epoch": 5, "memory": 44140, "step": 19268} +{"lr": 5.877169979928426e-05, "data_time": 0.0014252185821533204, "loss": 0.4374487638473511, "time": 0.6336438179016113, "epoch": 5, "memory": 44140, "step": 19368} +{"lr": 5.842732461812828e-05, "data_time": 0.0015235424041748046, "loss": 0.43618308901786806, "time": 0.6352621078491211, "epoch": 5, "memory": 44140, "step": 19468} +{"lr": 5.8082747422909646e-05, "data_time": 0.0018047332763671876, "loss": 0.42677063345909116, "time": 0.632779335975647, "epoch": 5, "memory": 44140, "step": 19568} +{"lr": 5.7737988523767406e-05, "data_time": 0.001514101028442383, "loss": 0.4765779942274094, "time": 0.6328549146652221, "epoch": 5, "memory": 44140, "step": 19668} +{"lr": 5.739306824155042e-05, "data_time": 0.0024080276489257812, "loss": 0.4456770420074463, "time": 0.635212755203247, "epoch": 5, "memory": 44140, "step": 19768} +{"lr": 5.704800690661993e-05, "data_time": 0.001786661148071289, "loss": 0.43700712323188784, "time": 0.632424783706665, "epoch": 5, "memory": 44140, "step": 19868} +{"lr": 5.6702824857650994e-05, "data_time": 0.0023291587829589845, "loss": 0.43297325968742373, "time": 0.6327181100845337, "epoch": 5, "memory": 44140, "step": 19968} +{"lr": 5.635754244043411e-05, "data_time": 0.0019201278686523438, "loss": 0.4218790054321289, "time": 0.6360872030258179, "epoch": 5, "memory": 44140, "step": 20068} +{"lr": 5.60121800066754e-05, "data_time": 0.0016581296920776367, "loss": 0.4351540684700012, "time": 0.6338484764099122, "epoch": 5, "memory": 44140, "step": 20168} +{"lr": 5.5666757912797516e-05, "data_time": 0.002289009094238281, "loss": 0.41573789715766907, "time": 0.6346448183059692, "epoch": 5, "memory": 44140, "step": 20268} +{"lr": 5.532129651873955e-05, "data_time": 0.0014219045639038085, "loss": 0.411351689696312, "time": 0.6326027631759643, "epoch": 5, "memory": 44140, "step": 20368} +{"accuracy/top1": 79.04109954833984, "data_time": 0.018181178967158, "time": 1.2852958897749582, "step": 5} +{"lr": 5.465797506068693e-05, "data_time": 0.0014163494110107423, "loss": 0.3863424897193909, "time": 0.6321403026580811, "epoch": 6, "memory": 44140, "step": 20560} +{"lr": 5.431251545044144e-05, "data_time": 0.0014596223831176759, "loss": 0.43077988028526304, "time": 0.6337174415588379, "epoch": 6, "memory": 44140, "step": 20660} +{"lr": 5.396709636204968e-05, "data_time": 0.0015720129013061523, "loss": 0.4430558800697327, "time": 0.6326960325241089, "epoch": 6, "memory": 44140, "step": 20760} +{"lr": 5.3621738155273566e-05, "data_time": 0.001405954360961914, "loss": 0.42521955966949465, "time": 0.6327214717864991, "epoch": 6, "memory": 44140, "step": 20860} +{"lr": 5.3276461186286584e-05, "data_time": 0.0013530492782592774, "loss": 0.4332079946994781, "time": 0.6334285020828248, "epoch": 6, "memory": 44140, "step": 20960} +{"lr": 5.293128580647384e-05, "data_time": 0.0015474081039428711, "loss": 0.4298450410366058, "time": 0.6326422929763794, "epoch": 6, "memory": 44140, "step": 21060} +{"lr": 5.258623236123256e-05, "data_time": 0.0015361309051513672, "loss": 0.44527106285095214, "time": 0.6339249849319458, "epoch": 6, "memory": 44140, "step": 21160} +{"lr": 5.224132118877281e-05, "data_time": 0.0014399051666259765, "loss": 0.4186409264802933, "time": 0.6310220003128052, "epoch": 6, "memory": 44140, "step": 21260} +{"lr": 5.189657261891887e-05, "data_time": 0.0015101194381713866, "loss": 0.43008764982223513, "time": 0.6324338912963867, "epoch": 6, "memory": 44140, "step": 21360} +{"lr": 5.1552006971910974e-05, "data_time": 0.0017882108688354493, "loss": 0.42609631419181826, "time": 0.6323167562484742, "epoch": 6, "memory": 44140, "step": 21460} +{"lr": 5.120764455720728e-05, "data_time": 0.002320599555969238, "loss": 0.40623250007629397, "time": 0.6318898439407349, "epoch": 6, "memory": 44140, "step": 21560} +{"lr": 5.0863505672287034e-05, "data_time": 0.0016577482223510743, "loss": 0.4499622344970703, "time": 0.6329441785812377, "epoch": 6, "memory": 44140, "step": 21660} +{"lr": 5.051961060145421e-05, "data_time": 0.0014844894409179687, "loss": 0.426160603761673, "time": 0.6338789939880372, "epoch": 6, "memory": 44140, "step": 21760} +{"lr": 5.017597961464177e-05, "data_time": 0.0017120838165283203, "loss": 0.38233913779258727, "time": 0.6330686330795288, "epoch": 6, "memory": 44140, "step": 21860} +{"lr": 4.983263296621707e-05, "data_time": 0.0015595912933349609, "loss": 0.45386329889297483, "time": 0.6337078809738159, "epoch": 6, "memory": 44140, "step": 21960} +{"lr": 4.948959089378785e-05, "data_time": 0.0017532587051391601, "loss": 0.43574273586273193, "time": 0.632140064239502, "epoch": 6, "memory": 44140, "step": 22060} +{"lr": 4.91468736170094e-05, "data_time": 0.0014657020568847657, "loss": 0.4344453752040863, "time": 0.6327176570892334, "epoch": 6, "memory": 44140, "step": 22160} +{"lr": 4.880450133639314e-05, "data_time": 0.0014443635940551759, "loss": 0.4336450189352036, "time": 0.6333205223083496, "epoch": 6, "memory": 44140, "step": 22260} +{"lr": 4.8462494232115335e-05, "data_time": 0.0014873266220092774, "loss": 0.43560277223587035, "time": 0.6326103210449219, "epoch": 6, "memory": 44140, "step": 22360} +{"lr": 4.812087246282806e-05, "data_time": 0.001444840431213379, "loss": 0.4077687829732895, "time": 0.6332634925842285, "epoch": 6, "memory": 44140, "step": 22460} +{"lr": 4.7779656164470954e-05, "data_time": 0.001477670669555664, "loss": 0.4258314102888107, "time": 0.6323377609252929, "epoch": 6, "memory": 44140, "step": 22560} +{"lr": 4.7438865449084246e-05, "data_time": 0.0015760183334350586, "loss": 0.42388800382614134, "time": 0.6322774887084961, "epoch": 6, "memory": 44140, "step": 22660} +{"lr": 4.709852040362338e-05, "data_time": 0.001534748077392578, "loss": 0.457219535112381, "time": 0.6332423448562622, "epoch": 6, "memory": 44140, "step": 22760} +{"lr": 4.675864108877499e-05, "data_time": 0.001492929458618164, "loss": 0.44097004532814027, "time": 0.6326010227203369, "epoch": 6, "memory": 44140, "step": 22860} +{"lr": 4.641924753777464e-05, "data_time": 0.001659989356994629, "loss": 0.4451592117547989, "time": 0.6344381093978881, "epoch": 6, "memory": 44140, "step": 22960} +{"lr": 4.608035975522578e-05, "data_time": 0.002403736114501953, "loss": 0.4552268981933594, "time": 0.6323726415634155, "epoch": 6, "memory": 44140, "step": 23060} +{"lr": 4.574199771592085e-05, "data_time": 0.0014498233795166016, "loss": 0.4126337796449661, "time": 0.633358645439148, "epoch": 6, "memory": 44140, "step": 23160} +{"lr": 4.540418136366372e-05, "data_time": 0.001405811309814453, "loss": 0.3933737754821777, "time": 0.6323083400726318, "epoch": 6, "memory": 44140, "step": 23260} +{"lr": 4.506693061009437e-05, "data_time": 0.0014887809753417968, "loss": 0.45793297290802004, "time": 0.6364764451980591, "epoch": 6, "memory": 44140, "step": 23360} +{"lr": 4.473026533351511e-05, "data_time": 0.0015930414199829101, "loss": 0.4237523913383484, "time": 0.6339193105697631, "epoch": 6, "memory": 44140, "step": 23460} +{"lr": 4.439420537771888e-05, "data_time": 0.0016127347946166993, "loss": 0.422735595703125, "time": 0.6327435970306396, "epoch": 6, "memory": 44140, "step": 23560} +{"lr": 4.405877055081978e-05, "data_time": 0.002032613754272461, "loss": 0.4336425542831421, "time": 0.6359319448471069, "epoch": 6, "memory": 44140, "step": 23660} +{"lr": 4.3723980624085373e-05, "data_time": 0.0014849662780761718, "loss": 0.4337951481342316, "time": 0.6327850103378296, "epoch": 6, "memory": 44140, "step": 23760} +{"lr": 4.3389855330771394e-05, "data_time": 0.001927661895751953, "loss": 0.45388599038124083, "time": 0.6324433803558349, "epoch": 6, "memory": 44140, "step": 23860} +{"lr": 4.305641436495869e-05, "data_time": 0.0015739679336547851, "loss": 0.44674779772758483, "time": 0.6343928813934326, "epoch": 6, "memory": 44140, "step": 23960} +{"lr": 4.272367738039222e-05, "data_time": 0.0017908096313476562, "loss": 0.4293277680873871, "time": 0.6322276830673218, "epoch": 6, "memory": 44140, "step": 24060} +{"lr": 4.239166398932278e-05, "data_time": 0.0015479326248168945, "loss": 0.4475178837776184, "time": 0.6325698614120483, "epoch": 6, "memory": 44140, "step": 24160} +{"lr": 4.206039376135108e-05, "data_time": 0.0016879796981811523, "loss": 0.4636631578207016, "time": 0.6325940370559693, "epoch": 6, "memory": 44140, "step": 24260} +{"lr": 4.1729886222273985e-05, "data_time": 0.0018097400665283204, "loss": 0.45206343829631807, "time": 0.634130597114563, "epoch": 6, "memory": 44140, "step": 24360} +{"lr": 4.140016085293389e-05, "data_time": 0.001535654067993164, "loss": 0.4263759821653366, "time": 0.6329107999801635, "epoch": 6, "memory": 44140, "step": 24460} +{"accuracy/top1": 80.93937683105469, "data_time": 0.017462424437204995, "time": 1.2861662606398265, "step": 6} +{"lr": 4.0769351868715556e-05, "data_time": 0.0015712738037109374, "loss": 0.4451601356267929, "time": 0.6331160545349122, "epoch": 7, "memory": 44140, "step": 24652} +{"lr": 4.044202148988293e-05, "data_time": 0.0015348672866821289, "loss": 0.441101410984993, "time": 0.6327461957931518, "epoch": 7, "memory": 44140, "step": 24752} +{"lr": 4.0115549190371644e-05, "data_time": 0.0014959812164306641, "loss": 0.4430814743041992, "time": 0.6337475538253784, "epoch": 7, "memory": 44140, "step": 24852} +{"lr": 3.978995421317808e-05, "data_time": 0.0014377832412719727, "loss": 0.4063778817653656, "time": 0.6330408334732056, "epoch": 7, "memory": 44140, "step": 24952} +{"lr": 3.94652557495873e-05, "data_time": 0.0021008729934692385, "loss": 0.4139744132757187, "time": 0.6326706171035766, "epoch": 7, "memory": 44140, "step": 25052} +{"lr": 3.914147293804207e-05, "data_time": 0.0015067577362060547, "loss": 0.45996909141540526, "time": 0.632449722290039, "epoch": 7, "memory": 44140, "step": 25152} +{"lr": 3.881862486301448e-05, "data_time": 0.0014073610305786132, "loss": 0.4257462203502655, "time": 0.6325594186782837, "epoch": 7, "memory": 44140, "step": 25252} +{"lr": 3.849673055388122e-05, "data_time": 0.0015043973922729491, "loss": 0.4563670545816422, "time": 0.6327755451202393, "epoch": 7, "memory": 44140, "step": 25352} +{"lr": 3.817580898380197e-05, "data_time": 0.0013645410537719727, "loss": 0.4589421510696411, "time": 0.6324453830718995, "epoch": 7, "memory": 44140, "step": 25452} +{"lr": 3.7855879068600975e-05, "data_time": 0.0023953914642333984, "loss": 0.43817236125469206, "time": 0.632078742980957, "epoch": 7, "memory": 44140, "step": 25552} +{"lr": 3.753695966565216e-05, "data_time": 0.001933908462524414, "loss": 0.4153300642967224, "time": 0.6317489624023438, "epoch": 7, "memory": 44140, "step": 25652} +{"lr": 3.7219069572767636e-05, "data_time": 0.0021514177322387697, "loss": 0.42778047621250154, "time": 0.6320550441741943, "epoch": 7, "memory": 44140, "step": 25752} +{"lr": 3.6902227527089725e-05, "data_time": 0.002036738395690918, "loss": 0.41400990784168246, "time": 0.634029746055603, "epoch": 7, "memory": 44140, "step": 25852} +{"lr": 3.65864522039865e-05, "data_time": 0.0018081903457641602, "loss": 0.43593273758888246, "time": 0.6349156856536865, "epoch": 7, "memory": 44140, "step": 25952} +{"lr": 3.627176221595109e-05, "data_time": 0.0019333600997924806, "loss": 0.43460625410079956, "time": 0.6339972734451294, "epoch": 7, "memory": 44140, "step": 26052} +{"lr": 3.595817611150461e-05, "data_time": 0.0020090579986572266, "loss": 0.42278642058372495, "time": 0.6337616443634033, "epoch": 7, "memory": 44140, "step": 26152} +{"lr": 3.5645712374102824e-05, "data_time": 0.027292203903198243, "loss": 0.43919208347797395, "time": 1.0241565704345703, "epoch": 7, "memory": 44140, "step": 26252} +{"lr": 3.533438942104663e-05, "data_time": 0.002497363090515137, "loss": 0.43299540877342224, "time": 0.6341647148132324, "epoch": 7, "memory": 44140, "step": 26352} +{"lr": 3.5024225602396704e-05, "data_time": 0.0026648759841918944, "loss": 0.4354636311531067, "time": 0.6350600242614746, "epoch": 7, "memory": 44140, "step": 26452} +{"lr": 3.4715239199891735e-05, "data_time": 0.0017151594161987304, "loss": 0.4206331014633179, "time": 0.6419120073318482, "epoch": 7, "memory": 44140, "step": 26552} +{"lr": 3.44074484258709e-05, "data_time": 0.0015797376632690429, "loss": 0.43789590895175934, "time": 0.6338491678237915, "epoch": 7, "memory": 44140, "step": 26652} +{"lr": 3.4100871422200426e-05, "data_time": 0.0014853239059448241, "loss": 0.4500276446342468, "time": 0.6319789171218873, "epoch": 7, "memory": 44140, "step": 26752} +{"lr": 3.379552625920419e-05, "data_time": 0.0015302419662475586, "loss": 0.45275644659996034, "time": 0.6354724168777466, "epoch": 7, "memory": 44140, "step": 26852} +{"lr": 3.349143093459868e-05, "data_time": 0.0020882129669189454, "loss": 0.45181081891059877, "time": 0.6340699195861816, "epoch": 7, "memory": 44140, "step": 26952} +{"lr": 3.318860337243226e-05, "data_time": 0.001764535903930664, "loss": 0.4159604340791702, "time": 0.6330552816390991, "epoch": 7, "memory": 44140, "step": 27052} +{"lr": 3.288706142202845e-05, "data_time": 0.0022506237030029295, "loss": 0.4469432055950165, "time": 0.6332366228103637, "epoch": 7, "memory": 44140, "step": 27152} +{"lr": 3.25868228569341e-05, "data_time": 0.002504706382751465, "loss": 0.43903608024120333, "time": 0.6332183837890625, "epoch": 7, "memory": 44140, "step": 27252} +{"lr": 3.228790537387148e-05, "data_time": 0.001451563835144043, "loss": 0.43307424783706666, "time": 0.6329429864883422, "epoch": 7, "memory": 44140, "step": 27352} +{"lr": 3.199032659169556e-05, "data_time": 0.002272152900695801, "loss": 0.41505860090255736, "time": 0.6331403970718383, "epoch": 7, "memory": 44140, "step": 27452} +{"lr": 3.169410405035527e-05, "data_time": 0.001854705810546875, "loss": 0.4372330605983734, "time": 0.6322521448135376, "epoch": 7, "memory": 44140, "step": 27552} +{"lr": 3.139925520985965e-05, "data_time": 0.0020429134368896485, "loss": 0.4525086939334869, "time": 0.6338441848754883, "epoch": 7, "memory": 44140, "step": 27652} +{"lr": 3.1105797449248916e-05, "data_time": 0.001802682876586914, "loss": 0.4206294059753418, "time": 0.6332003355026246, "epoch": 7, "memory": 44140, "step": 27752} +{"lr": 3.081374806556986e-05, "data_time": 0.0018229484558105469, "loss": 0.45739985406398775, "time": 0.6335875511169433, "epoch": 7, "memory": 44140, "step": 27852} +{"lr": 3.052312427285649e-05, "data_time": 0.0020348310470581056, "loss": 0.4342558741569519, "time": 0.633012056350708, "epoch": 7, "memory": 44140, "step": 27952} +{"lr": 3.0233943201115213e-05, "data_time": 0.0018118858337402345, "loss": 0.42804893255233767, "time": 0.6341001272201539, "epoch": 7, "memory": 44140, "step": 28052} +{"lr": 2.9946221895315525e-05, "data_time": 0.0020540952682495117, "loss": 0.41936197578907014, "time": 0.6324288129806519, "epoch": 7, "memory": 44140, "step": 28152} +{"lr": 2.9659977314384953e-05, "data_time": 0.0017561674118041991, "loss": 0.45191813111305235, "time": 0.6322898626327514, "epoch": 7, "memory": 44140, "step": 28252} +{"lr": 2.93752263302096e-05, "data_time": 0.0024352788925170897, "loss": 0.432957860827446, "time": 0.6332912921905518, "epoch": 7, "memory": 44140, "step": 28352} +{"lr": 2.9091985726639714e-05, "data_time": 0.0021311283111572266, "loss": 0.45699959993362427, "time": 0.6321824312210083, "epoch": 7, "memory": 44140, "step": 28452} +{"lr": 2.881027219850035e-05, "data_time": 0.0017208576202392579, "loss": 0.4274042367935181, "time": 0.6325274467468261, "epoch": 7, "memory": 44140, "step": 28552} +{"accuracy/top1": 82.50831604003906, "data_time": 0.017026901245117188, "time": 1.2840092658996582, "step": 7} +{"lr": 2.827372366659638e-05, "data_time": 0.001958155632019043, "loss": 0.45466240048408507, "time": 0.633280897140503, "epoch": 8, "memory": 44140, "step": 28744} +{"lr": 2.7996563897301616e-05, "data_time": 0.0014590978622436523, "loss": 0.45339798033237455, "time": 0.6333338260650635, "epoch": 8, "memory": 44140, "step": 28844} +{"lr": 2.7720995770013028e-05, "data_time": 0.0019378900527954102, "loss": 0.4233066976070404, "time": 0.6471317291259766, "epoch": 8, "memory": 44140, "step": 28944} +{"lr": 2.744703552732299e-05, "data_time": 0.00213165283203125, "loss": 0.43006622791290283, "time": 0.6344423055648803, "epoch": 8, "memory": 44140, "step": 29044} +{"lr": 2.717469931705171e-05, "data_time": 0.0018494606018066406, "loss": 0.4387540936470032, "time": 0.6328137636184692, "epoch": 8, "memory": 44140, "step": 29144} +{"lr": 2.690400319129529e-05, "data_time": 0.0017334699630737304, "loss": 0.435209196805954, "time": 0.6318532466888428, "epoch": 8, "memory": 44140, "step": 29244} +{"lr": 2.6634963105479716e-05, "data_time": 0.0015765666961669923, "loss": 0.39980089366436006, "time": 0.6606237649917602, "epoch": 8, "memory": 44140, "step": 29344} +{"lr": 2.6367594917420177e-05, "data_time": 0.001616835594177246, "loss": 0.4316032290458679, "time": 0.6343820810317993, "epoch": 8, "memory": 44140, "step": 29444} +{"lr": 2.610191438638673e-05, "data_time": 0.0014084100723266602, "loss": 0.4166017174720764, "time": 0.6355518579483033, "epoch": 8, "memory": 44140, "step": 29544} +{"lr": 2.5837937172175032e-05, "data_time": 0.001395273208618164, "loss": 0.4450955927371979, "time": 0.6323782444000244, "epoch": 8, "memory": 44140, "step": 29644} +{"lr": 2.557567883418364e-05, "data_time": 0.001414942741394043, "loss": 0.4389266610145569, "time": 0.6329269170761108, "epoch": 8, "memory": 44140, "step": 29744} +{"lr": 2.531515483049656e-05, "data_time": 0.0015266656875610352, "loss": 0.4255665749311447, "time": 0.6326501131057739, "epoch": 8, "memory": 44140, "step": 29844} +{"lr": 2.5056380516972453e-05, "data_time": 0.001394343376159668, "loss": 0.40291866660118103, "time": 0.6322141408920288, "epoch": 8, "memory": 44140, "step": 29944} +{"lr": 2.4799371146339435e-05, "data_time": 0.0014451265335083008, "loss": 0.43276655673980713, "time": 0.6318119525909424, "epoch": 8, "memory": 44140, "step": 30044} +{"lr": 2.4544141867295954e-05, "data_time": 0.001522397994995117, "loss": 0.4334082961082458, "time": 0.6322718381881713, "epoch": 8, "memory": 44140, "step": 30144} +{"lr": 2.429070772361796e-05, "data_time": 0.0022794008255004883, "loss": 0.40298391580581666, "time": 0.6311321496963501, "epoch": 8, "memory": 44140, "step": 30244} +{"lr": 2.4039083653272075e-05, "data_time": 0.0014112472534179687, "loss": 0.4400285094976425, "time": 0.6329334735870361, "epoch": 8, "memory": 44140, "step": 30344} +{"lr": 2.378928448753535e-05, "data_time": 0.0015388727188110352, "loss": 0.4242362916469574, "time": 0.6322114706039429, "epoch": 8, "memory": 44140, "step": 30444} +{"lr": 2.354132495012082e-05, "data_time": 0.001468515396118164, "loss": 0.43261079490184784, "time": 0.6333138227462769, "epoch": 8, "memory": 44140, "step": 30544} +{"lr": 2.329521965630983e-05, "data_time": 0.0015710353851318359, "loss": 0.40571180880069735, "time": 0.6320971965789794, "epoch": 8, "memory": 44140, "step": 30644} +{"lr": 2.3050983112090472e-05, "data_time": 0.0015341281890869141, "loss": 0.4292501866817474, "time": 0.6324087619781494, "epoch": 8, "memory": 44140, "step": 30744} +{"lr": 2.2808629713302708e-05, "data_time": 0.0014192819595336913, "loss": 0.42273353040218353, "time": 0.6319505214691162, "epoch": 8, "memory": 44140, "step": 30844} +{"lr": 2.2568173744789706e-05, "data_time": 0.0016644716262817383, "loss": 0.4291909635066986, "time": 0.632189679145813, "epoch": 8, "memory": 44140, "step": 30944} +{"lr": 2.2329629379555897e-05, "data_time": 0.0020752191543579102, "loss": 0.4299454927444458, "time": 0.6315293312072754, "epoch": 8, "memory": 44140, "step": 31044} +{"lr": 2.2093010677931667e-05, "data_time": 0.0018827438354492188, "loss": 0.4226302891969681, "time": 0.6322283983230591, "epoch": 8, "memory": 44140, "step": 31144} +{"lr": 2.1858331586744434e-05, "data_time": 0.0014776468276977539, "loss": 0.4532845765352249, "time": 0.6333807468414306, "epoch": 8, "memory": 44140, "step": 31244} +{"lr": 2.1625605938496814e-05, "data_time": 0.0016274452209472656, "loss": 0.41041659116744994, "time": 0.6317580938339233, "epoch": 8, "memory": 44140, "step": 31344} +{"lr": 2.1394847450551186e-05, "data_time": 0.001647639274597168, "loss": 0.42769654393196105, "time": 0.6329609870910644, "epoch": 8, "memory": 44140, "step": 31444} +{"lr": 2.1166069724321078e-05, "data_time": 0.0032933235168457033, "loss": 0.46599029898643496, "time": 0.6349485635757446, "epoch": 8, "memory": 44140, "step": 31544} +{"lr": 2.0939286244469643e-05, "data_time": 0.001453709602355957, "loss": 0.4146469414234161, "time": 0.6327294111251831, "epoch": 8, "memory": 44140, "step": 31644} +{"lr": 2.071451037811473e-05, "data_time": 0.001534104347229004, "loss": 0.42976507544517517, "time": 0.6330198764801025, "epoch": 8, "memory": 44140, "step": 31744} +{"lr": 2.0491755374041045e-05, "data_time": 0.0015379428863525391, "loss": 0.42381633520126344, "time": 0.6326041936874389, "epoch": 8, "memory": 44140, "step": 31844} +{"lr": 2.0271034361919193e-05, "data_time": 0.001643538475036621, "loss": 0.4398748457431793, "time": 0.6331932544708252, "epoch": 8, "memory": 44140, "step": 31944} +{"lr": 2.0052360351531814e-05, "data_time": 0.0018382549285888671, "loss": 0.42296539545059203, "time": 0.6324436664581299, "epoch": 8, "memory": 44140, "step": 32044} +{"lr": 1.983574623200682e-05, "data_time": 0.0017765998840332032, "loss": 0.41598489284515383, "time": 0.6348613739013672, "epoch": 8, "memory": 44140, "step": 32144} +{"lr": 1.9621204771057533e-05, "data_time": 0.0017125606536865234, "loss": 0.4029065787792206, "time": 0.6338467597961426, "epoch": 8, "memory": 44140, "step": 32244} +{"lr": 1.940874861423025e-05, "data_time": 0.0015275001525878907, "loss": 0.4237914115190506, "time": 0.6326350212097168, "epoch": 8, "memory": 44140, "step": 32344} +{"lr": 1.9198390284158783e-05, "data_time": 0.0021889925003051756, "loss": 0.4153091847896576, "time": 0.6321191310882568, "epoch": 8, "memory": 44140, "step": 32444} +{"lr": 1.8990142179826573e-05, "data_time": 0.0014643430709838866, "loss": 0.44834463596343993, "time": 0.63248131275177, "epoch": 8, "memory": 44140, "step": 32544} +{"lr": 1.8784016575835568e-05, "data_time": 0.0021977901458740236, "loss": 0.4292978823184967, "time": 0.6317805051803589, "epoch": 8, "memory": 44140, "step": 32644} +{"accuracy/top1": 83.38215637207031, "data_time": 0.017577195167541505, "time": 1.2848993082841238, "step": 8} +{"lr": 1.8394249604901642e-05, "data_time": 0.0015108585357666016, "loss": 0.4192079246044159, "time": 0.6330900430679322, "epoch": 9, "memory": 44140, "step": 32836} +{"lr": 1.81943907744488e-05, "data_time": 0.0017599344253540039, "loss": 0.4158934384584427, "time": 0.6336795330047608, "epoch": 9, "memory": 44140, "step": 32936} +{"lr": 1.7996701347677866e-05, "data_time": 0.0014974117279052735, "loss": 0.44777504801750184, "time": 0.632328987121582, "epoch": 9, "memory": 44140, "step": 33036} +{"lr": 1.780119297683906e-05, "data_time": 0.0015994787216186523, "loss": 0.42130537927150724, "time": 0.6327724933624268, "epoch": 9, "memory": 44140, "step": 33136} +{"lr": 1.7607877185626483e-05, "data_time": 0.0022485971450805662, "loss": 0.443629652261734, "time": 0.6324627161026001, "epoch": 9, "memory": 44140, "step": 33236} +{"lr": 1.7416765368498675e-05, "data_time": 0.00175323486328125, "loss": 0.417745640873909, "time": 0.6315961599349975, "epoch": 9, "memory": 44140, "step": 33336} +{"lr": 1.7227868790007105e-05, "data_time": 0.001706242561340332, "loss": 0.4439123958349228, "time": 0.6318429708480835, "epoch": 9, "memory": 44140, "step": 33436} +{"lr": 1.7041198584132233e-05, "data_time": 0.00156707763671875, "loss": 0.4293932497501373, "time": 0.6323116064071655, "epoch": 9, "memory": 44140, "step": 33536} +{"lr": 1.6856765753627146e-05, "data_time": 0.0020839691162109373, "loss": 0.4280243694782257, "time": 0.6331812143325806, "epoch": 9, "memory": 44140, "step": 33636} +{"lr": 1.6674581169369163e-05, "data_time": 0.0024164915084838867, "loss": 0.4526894450187683, "time": 0.6328007936477661, "epoch": 9, "memory": 44140, "step": 33736} +{"lr": 1.6494655569718982e-05, "data_time": 0.0015497684478759765, "loss": 0.41594940423965454, "time": 0.6424659252166748, "epoch": 9, "memory": 44140, "step": 33836} +{"lr": 1.6316999559887786e-05, "data_time": 0.0017044544219970703, "loss": 0.40958764851093293, "time": 0.6325393199920655, "epoch": 9, "memory": 44140, "step": 33936} +{"lr": 1.6141623611312105e-05, "data_time": 0.0026323080062866213, "loss": 0.4171371668577194, "time": 0.6314513206481933, "epoch": 9, "memory": 44140, "step": 34036} +{"lr": 1.5968538061036705e-05, "data_time": 0.001636838912963867, "loss": 0.428592312335968, "time": 0.6314349889755249, "epoch": 9, "memory": 44140, "step": 34136} +{"lr": 1.5797753111105157e-05, "data_time": 0.0015050888061523438, "loss": 0.4413925111293793, "time": 0.6323025465011597, "epoch": 9, "memory": 44140, "step": 34236} +{"lr": 1.5629278827958574e-05, "data_time": 0.0017032384872436523, "loss": 0.4407835900783539, "time": 0.6320976972579956, "epoch": 9, "memory": 44140, "step": 34336} +{"lr": 1.5463125141842392e-05, "data_time": 0.0015386581420898438, "loss": 0.4242474019527435, "time": 0.6320900201797486, "epoch": 9, "memory": 44140, "step": 34436} +{"lr": 1.5299301846220796e-05, "data_time": 0.0017403841018676757, "loss": 0.4342947691679001, "time": 0.631589412689209, "epoch": 9, "memory": 44140, "step": 34536} +{"lr": 1.5137818597199751e-05, "data_time": 0.0016096830368041992, "loss": 0.46631333529949187, "time": 0.6328023910522461, "epoch": 9, "memory": 44140, "step": 34636} +{"lr": 1.4978684912957622e-05, "data_time": 0.001450657844543457, "loss": 0.4332442104816437, "time": 0.6314464569091797, "epoch": 9, "memory": 44140, "step": 34736} +{"lr": 1.4821910173184321e-05, "data_time": 0.002198624610900879, "loss": 0.4289512515068054, "time": 0.6325736999511719, "epoch": 9, "memory": 44140, "step": 34836} +{"lr": 1.4667503618528337e-05, "data_time": 0.0017051935195922852, "loss": 0.44457480907440183, "time": 0.6312191724777222, "epoch": 9, "memory": 44140, "step": 34936} +{"lr": 1.4515474350052097e-05, "data_time": 0.0014960050582885742, "loss": 0.39010309278964994, "time": 0.6336281061172485, "epoch": 9, "memory": 44140, "step": 35036} +{"lr": 1.4365831328695616e-05, "data_time": 0.001566457748413086, "loss": 0.42951966226100924, "time": 0.6338382005691529, "epoch": 9, "memory": 44140, "step": 35136} +{"lr": 1.4218583374748169e-05, "data_time": 0.001975345611572266, "loss": 0.40147810578346255, "time": 0.633592939376831, "epoch": 9, "memory": 44140, "step": 35236} +{"lr": 1.4073739167328542e-05, "data_time": 0.0015869617462158203, "loss": 0.4517203629016876, "time": 0.6345815420150757, "epoch": 9, "memory": 44140, "step": 35336} +{"lr": 1.3931307243873344e-05, "data_time": 0.001720762252807617, "loss": 0.44844684302806853, "time": 0.6338479042053222, "epoch": 9, "memory": 44140, "step": 35436} +{"lr": 1.3791295999633911e-05, "data_time": 0.0016016960144042969, "loss": 0.4113050580024719, "time": 0.6337888956069946, "epoch": 9, "memory": 44140, "step": 35536} +{"lr": 1.3653713687181394e-05, "data_time": 0.002195572853088379, "loss": 0.4379051148891449, "time": 0.6354958295822144, "epoch": 9, "memory": 44140, "step": 35636} +{"lr": 1.3518568415920337e-05, "data_time": 0.0015153884887695312, "loss": 0.4328551203012466, "time": 0.6332636594772338, "epoch": 9, "memory": 44140, "step": 35736} +{"lr": 1.3385868151610723e-05, "data_time": 0.0019635915756225585, "loss": 0.40239324867725373, "time": 0.6343563318252563, "epoch": 9, "memory": 44140, "step": 35836} +{"lr": 1.3255620715898442e-05, "data_time": 0.0018749475479125977, "loss": 0.4242888927459717, "time": 0.6334626913070679, "epoch": 9, "memory": 44140, "step": 35936} +{"lr": 1.3127833785854261e-05, "data_time": 0.0017363548278808594, "loss": 0.4422772765159607, "time": 0.6335497379302979, "epoch": 9, "memory": 44140, "step": 36036} +{"lr": 1.3002514893521313e-05, "data_time": 0.0018674850463867188, "loss": 0.4114193767309189, "time": 0.6336813688278198, "epoch": 9, "memory": 44140, "step": 36136} +{"lr": 1.287967142547117e-05, "data_time": 0.00149383544921875, "loss": 0.4362452030181885, "time": 0.636810564994812, "epoch": 9, "memory": 44140, "step": 36236} +{"lr": 1.2759310622368435e-05, "data_time": 0.0017003774642944335, "loss": 0.44429692029953005, "time": 0.6355048418045044, "epoch": 9, "memory": 44140, "step": 36336} +{"lr": 1.2641439578543968e-05, "data_time": 0.0015533685684204102, "loss": 0.43545244038105013, "time": 0.6331853151321412, "epoch": 9, "memory": 44140, "step": 36436} +{"lr": 1.252606524157674e-05, "data_time": 0.001909637451171875, "loss": 0.44942472577095033, "time": 0.6355678081512451, "epoch": 9, "memory": 44140, "step": 36536} +{"lr": 1.2413194411884297e-05, "data_time": 0.001904773712158203, "loss": 0.41389899551868437, "time": 0.6345012426376343, "epoch": 9, "memory": 44140, "step": 36636} +{"lr": 1.2302833742321981e-05, "data_time": 0.001581096649169922, "loss": 0.4211801588535309, "time": 0.6346514225006104, "epoch": 9, "memory": 44140, "step": 36736} +{"accuracy/top1": 84.1517333984375, "data_time": 0.018436970313390096, "time": 1.28598349293073, "step": 9} +{"lr": 1.2098001439139718e-05, "data_time": 0.001784038543701172, "loss": 0.4378559350967407, "time": 0.6330617427825928, "epoch": 10, "memory": 44140, "step": 36928} +{"lr": 1.1995007122352056e-05, "data_time": 0.0016316652297973632, "loss": 0.4197396695613861, "time": 0.6329480171203613, "epoch": 10, "memory": 44140, "step": 37028} +{"lr": 1.1894547614574522e-05, "data_time": 0.0018131494522094726, "loss": 0.45584308803081514, "time": 0.6335228681564331, "epoch": 10, "memory": 44140, "step": 37128} +{"lr": 1.1796628837111806e-05, "data_time": 0.0020069122314453126, "loss": 0.4205508291721344, "time": 0.6323853969573975, "epoch": 10, "memory": 44140, "step": 37228} +{"lr": 1.1701256561512343e-05, "data_time": 0.001579594612121582, "loss": 0.4137354552745819, "time": 0.6353223085403442, "epoch": 10, "memory": 44140, "step": 37328} +{"lr": 1.1608436409228136e-05, "data_time": 0.0017122507095336914, "loss": 0.40455450415611266, "time": 0.6342119455337525, "epoch": 10, "memory": 44140, "step": 37428} +{"lr": 1.1518173851283401e-05, "data_time": 0.0020758390426635744, "loss": 0.43558679819107055, "time": 0.6337926149368286, "epoch": 10, "memory": 44140, "step": 37528} +{"lr": 1.1430474207952176e-05, "data_time": 0.0021812915802001953, "loss": 0.4532723903656006, "time": 0.6319443225860596, "epoch": 10, "memory": 44140, "step": 37628} +{"lr": 1.1345342648444593e-05, "data_time": 0.002339744567871094, "loss": 0.4262354731559753, "time": 0.6314018487930297, "epoch": 10, "memory": 44140, "step": 37728} +{"lr": 1.12627841906023e-05, "data_time": 0.0017124414443969727, "loss": 0.40543001890182495, "time": 0.6451123476028442, "epoch": 10, "memory": 44140, "step": 37828} +{"lr": 1.1182803700602659e-05, "data_time": 0.0025897502899169924, "loss": 0.4339084684848785, "time": 0.6339578628540039, "epoch": 10, "memory": 44140, "step": 37928} +{"lr": 1.110540589267192e-05, "data_time": 0.00283510684967041, "loss": 0.43444119691848754, "time": 0.6329931020736694, "epoch": 10, "memory": 44140, "step": 38028} +{"lr": 1.1030595328807396e-05, "data_time": 0.0016083955764770509, "loss": 0.4390761315822601, "time": 0.6333801984786988, "epoch": 10, "memory": 44140, "step": 38128} +{"lr": 1.0958376418508494e-05, "data_time": 0.0016560077667236329, "loss": 0.4350801944732666, "time": 0.6321488142013549, "epoch": 10, "memory": 44140, "step": 38228} +{"lr": 1.088875341851686e-05, "data_time": 0.0015148401260375976, "loss": 0.43409552574157717, "time": 0.6336699008941651, "epoch": 10, "memory": 44140, "step": 38328} +{"lr": 1.0821730432565486e-05, "data_time": 0.0030703067779541014, "loss": 0.44336821138858795, "time": 0.6336770296096802, "epoch": 10, "memory": 44140, "step": 38428} +{"lr": 1.0757311411136797e-05, "data_time": 0.002206110954284668, "loss": 0.4388790428638458, "time": 0.6329855680465698, "epoch": 10, "memory": 44140, "step": 38528} +{"lr": 1.0695500151229833e-05, "data_time": 0.0025473833084106445, "loss": 0.4290278673171997, "time": 0.6370671987533569, "epoch": 10, "memory": 44140, "step": 38628} +{"lr": 1.0636300296136373e-05, "data_time": 0.0017495155334472656, "loss": 0.435150146484375, "time": 0.6349704265594482, "epoch": 10, "memory": 44140, "step": 38728} +{"lr": 1.0579715335226322e-05, "data_time": 0.0016252517700195313, "loss": 0.40780472159385683, "time": 0.6348392009735108, "epoch": 10, "memory": 44140, "step": 38828} +{"lr": 1.0525748603741896e-05, "data_time": 0.0023051977157592775, "loss": 0.4292386621236801, "time": 0.6330715179443359, "epoch": 10, "memory": 44140, "step": 38928} +{"lr": 1.0474403282601119e-05, "data_time": 0.0014242410659790039, "loss": 0.424554306268692, "time": 0.6333153724670411, "epoch": 10, "memory": 44140, "step": 39028} +{"lr": 1.0425682398210336e-05, "data_time": 0.00188751220703125, "loss": 0.4188223987817764, "time": 0.6339438199996948, "epoch": 10, "memory": 44140, "step": 39128} +{"lr": 1.0379588822285784e-05, "data_time": 0.0021077632904052735, "loss": 0.4077302306890488, "time": 0.6327978134155273, "epoch": 10, "memory": 44140, "step": 39228} +{"lr": 1.0336125271684364e-05, "data_time": 0.0017225503921508788, "loss": 0.41424120366573336, "time": 0.6335647106170654, "epoch": 10, "memory": 44140, "step": 39328} +{"lr": 1.0295294308243494e-05, "data_time": 0.0023066520690917967, "loss": 0.44279840886592864, "time": 0.6315334796905517, "epoch": 10, "memory": 44140, "step": 39428} +{"lr": 1.0257098338630086e-05, "data_time": 0.00244288444519043, "loss": 0.4393835723400116, "time": 0.638019061088562, "epoch": 10, "memory": 44140, "step": 39528} +{"lr": 1.0221539614198729e-05, "data_time": 0.002653050422668457, "loss": 0.43867662250995637, "time": 0.6328872680664063, "epoch": 10, "memory": 44140, "step": 39628} +{"lr": 1.0188620230858978e-05, "data_time": 0.002184557914733887, "loss": 0.43402776420116423, "time": 0.6327784299850464, "epoch": 10, "memory": 44140, "step": 39728} +{"lr": 1.01583421289518e-05, "data_time": 0.0015357255935668946, "loss": 0.4060773134231567, "time": 0.6347954988479614, "epoch": 10, "memory": 44140, "step": 39828} +{"lr": 1.0130707093135206e-05, "data_time": 0.0015917539596557618, "loss": 0.4351502925157547, "time": 0.6341427564620972, "epoch": 10, "memory": 44140, "step": 39928} +{"lr": 1.01057167522791e-05, "data_time": 0.0015056371688842774, "loss": 0.4138208985328674, "time": 0.6322539329528809, "epoch": 10, "memory": 44140, "step": 40028} +{"lr": 1.008337257936921e-05, "data_time": 0.002364039421081543, "loss": 0.41611374318599703, "time": 0.6352760791778564, "epoch": 10, "memory": 44140, "step": 40128} +{"lr": 1.0063675891420294e-05, "data_time": 0.001945805549621582, "loss": 0.4419875293970108, "time": 0.6347781658172608, "epoch": 10, "memory": 44140, "step": 40228} +{"lr": 1.0046627849398548e-05, "data_time": 0.0018296480178833009, "loss": 0.4454681098461151, "time": 0.6340731620788574, "epoch": 10, "memory": 44140, "step": 40328} +{"lr": 1.0032229458153106e-05, "data_time": 0.0016800642013549804, "loss": 0.43427015244960787, "time": 0.6339587450027466, "epoch": 10, "memory": 44140, "step": 40428} +{"lr": 1.0020481566356862e-05, "data_time": 0.001519322395324707, "loss": 0.4427505224943161, "time": 0.6341932058334351, "epoch": 10, "memory": 44140, "step": 40528} +{"lr": 1.0011384866456456e-05, "data_time": 0.0018405437469482422, "loss": 0.435544490814209, "time": 0.633528757095337, "epoch": 10, "memory": 44140, "step": 40628} +{"lr": 1.00049398946314e-05, "data_time": 0.001684713363647461, "loss": 0.4310219496488571, "time": 0.6344710111618042, "epoch": 10, "memory": 44140, "step": 40728} +{"lr": 1.0001147030762536e-05, "data_time": 0.0014252185821533204, "loss": 0.4472906470298767, "time": 0.634446668624878, "epoch": 10, "memory": 44140, "step": 40828} +{"accuracy/top1": 84.35033416748047, "data_time": 0.017048033078511556, "time": 1.2856839835643767, "step": 10} diff --git a/clip_large_pretrain_4x256_all2_lr1e-4/20230606_005614/vis_data/config.py b/clip_large_pretrain_4x256_all2_lr1e-4/20230606_005614/vis_data/config.py new file mode 100644 index 0000000000000000000000000000000000000000..c989f09acfe14651d1114b3ff4a90a7112361e54 --- /dev/null +++ b/clip_large_pretrain_4x256_all2_lr1e-4/20230606_005614/vis_data/config.py @@ -0,0 +1,341 @@ +optim_wrapper = dict( + optimizer=dict( + type='AdamW', lr=0.0001, weight_decay=0.3, _scope_='mmpretrain'), + paramwise_cfg=dict( + custom_keys=dict({ + '.cls_token': dict(decay_mult=0.0), + '.pos_embed': dict(decay_mult=0.0) + })), + type='AmpOptimWrapper', + dtype='bfloat16', + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=False, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=4096) +model = dict( + type='ImageClassifier', + backbone=dict( + frozen_stages=24, + type='VisionTransformer', + arch='l', + img_size=224, + patch_size=14, + drop_rate=0.1, + pre_norm=True, + final_norm=False, + init_cfg=dict( + type='Pretrained', + checkpoint='ckpt/openclip-ViT-L-14.pth', + prefix='backbone')), + neck=dict( + type='CLIPProjection', + in_channels=1024, + out_channels=768, + init_cfg=dict( + type='Pretrained', + checkpoint='ckpt/openclip-ViT-L-14.pth', + prefix='backbone')), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=768, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + init_cfg=None), + init_cfg=dict( + type='TruncNormal', layer=['Conv2d', 'Linear'], std=0.02, bias=0.0), + train_cfg=None) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=128, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/all_0.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/all_1.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3fake8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/cc1m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3real8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=128, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV2-1-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/if-dpmsolver++-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=128, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV2-1-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/if-dpmsolver++-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/clip_large_pretrain_4x256_all2_lr1e-4' diff --git a/clip_large_pretrain_4x256_all2_lr1e-4/20230606_005614/vis_data/events.out.tfevents.1685984187.SH-IDC1-10-140-24-114.130857.0 b/clip_large_pretrain_4x256_all2_lr1e-4/20230606_005614/vis_data/events.out.tfevents.1685984187.SH-IDC1-10-140-24-114.130857.0 new file mode 100644 index 0000000000000000000000000000000000000000..24405943a51d73c419517cbc3716b22953fb13f7 --- /dev/null +++ b/clip_large_pretrain_4x256_all2_lr1e-4/20230606_005614/vis_data/events.out.tfevents.1685984187.SH-IDC1-10-140-24-114.130857.0 @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:7611490e0714f9acc8d9118a294eef07c8ca6aff260ef41ffd6f3d0ee56ba018 +size 3137033 diff --git a/clip_large_pretrain_4x256_all2_lr1e-4/20230606_005614/vis_data/scalars.json b/clip_large_pretrain_4x256_all2_lr1e-4/20230606_005614/vis_data/scalars.json new file mode 100644 index 0000000000000000000000000000000000000000..812899d42e2b79b1769959aa303291642cfa8b68 --- /dev/null +++ b/clip_large_pretrain_4x256_all2_lr1e-4/20230606_005614/vis_data/scalars.json @@ -0,0 +1,410 @@ +{"lr": 9.999870019168206e-05, "data_time": 0.0018175363540649414, "loss": 0.5742023468017579, "time": 0.6306015491485596, "epoch": 1, "memory": 44139, "step": 100} +{"lr": 9.99947481934738e-05, "data_time": 0.0018314838409423829, "loss": 0.5509862899780273, "time": 0.6331407070159912, "epoch": 1, "memory": 44139, "step": 200} +{"lr": 9.99881441056942e-05, "data_time": 0.002259683609008789, "loss": 0.5506836891174316, "time": 0.6333303213119507, "epoch": 1, "memory": 44139, "step": 300} +{"lr": 9.997888831760252e-05, "data_time": 0.0016779661178588866, "loss": 0.53231360912323, "time": 0.6325403451919556, "epoch": 1, "memory": 44139, "step": 400} +{"lr": 9.996698137475549e-05, "data_time": 0.0015641212463378905, "loss": 0.5343423008918762, "time": 0.6465238809585572, "epoch": 1, "memory": 44139, "step": 500} +{"lr": 9.99524239789746e-05, "data_time": 0.0014079809188842773, "loss": 0.4938073635101318, "time": 0.6329603433609009, "epoch": 1, "memory": 44139, "step": 600} +{"lr": 9.993521698830466e-05, "data_time": 0.001558542251586914, "loss": 0.5039564847946167, "time": 0.6353706121444702, "epoch": 1, "memory": 44139, "step": 700} +{"lr": 9.991536141696371e-05, "data_time": 0.0015775442123413086, "loss": 0.5006685137748719, "time": 0.6345899105072021, "epoch": 1, "memory": 44139, "step": 800} +{"lr": 9.989285843528274e-05, "data_time": 0.001490950584411621, "loss": 0.49552717208862307, "time": 0.6325678586959839, "epoch": 1, "memory": 44139, "step": 900} +{"lr": 9.986770936963713e-05, "data_time": 0.0014845848083496094, "loss": 0.497565770149231, "time": 0.6320618629455567, "epoch": 1, "memory": 44139, "step": 1000} +{"lr": 9.983991570236818e-05, "data_time": 0.001404738426208496, "loss": 0.5405094265937805, "time": 0.6333181858062744, "epoch": 1, "memory": 44139, "step": 1100} +{"lr": 9.980947907169598e-05, "data_time": 0.0014627456665039062, "loss": 0.47539167404174804, "time": 0.6362954616546631, "epoch": 1, "memory": 44139, "step": 1200} +{"lr": 9.977640127162247e-05, "data_time": 0.001424884796142578, "loss": 0.5027863502502441, "time": 0.6345264911651611, "epoch": 1, "memory": 44139, "step": 1300} +{"lr": 9.974068425182603e-05, "data_time": 0.001611638069152832, "loss": 0.4923910140991211, "time": 0.6314883708953858, "epoch": 1, "memory": 44139, "step": 1400} +{"lr": 9.970233011754657e-05, "data_time": 0.001566004753112793, "loss": 0.4950839042663574, "time": 0.6319402694702149, "epoch": 1, "memory": 44139, "step": 1500} +{"lr": 9.966134112946122e-05, "data_time": 0.0013825416564941407, "loss": 0.5086468577384948, "time": 0.631356167793274, "epoch": 1, "memory": 44139, "step": 1600} +{"lr": 9.961771970355126e-05, "data_time": 0.0013284921646118165, "loss": 0.4961442410945892, "time": 0.6313036203384399, "epoch": 1, "memory": 44139, "step": 1700} +{"lr": 9.957146841095961e-05, "data_time": 0.0014630556106567383, "loss": 0.48642589449882506, "time": 0.6314688682556152, "epoch": 1, "memory": 44139, "step": 1800} +{"lr": 9.952258997783933e-05, "data_time": 0.0013856887817382812, "loss": 0.4863204836845398, "time": 0.6308996438980102, "epoch": 1, "memory": 44139, "step": 1900} +{"lr": 9.947108728519288e-05, "data_time": 0.001604771614074707, "loss": 0.48236514925956725, "time": 0.631430983543396, "epoch": 1, "memory": 44139, "step": 2000} +{"lr": 9.941696336870245e-05, "data_time": 0.001503586769104004, "loss": 0.5119886755943298, "time": 0.631052041053772, "epoch": 1, "memory": 44139, "step": 2100} +{"lr": 9.936022141855094e-05, "data_time": 0.00168154239654541, "loss": 0.4707063287496567, "time": 0.6316580533981323, "epoch": 1, "memory": 44139, "step": 2200} +{"lr": 9.93008647792338e-05, "data_time": 0.001444864273071289, "loss": 0.49044070243835447, "time": 0.6314844846725464, "epoch": 1, "memory": 44139, "step": 2300} +{"lr": 9.923889694936212e-05, "data_time": 0.0014310121536254884, "loss": 0.4934364855289459, "time": 0.6319224596023559, "epoch": 1, "memory": 44139, "step": 2400} +{"lr": 9.917432158145639e-05, "data_time": 0.0014138936996459961, "loss": 0.5017586350440979, "time": 0.6312256097793579, "epoch": 1, "memory": 44139, "step": 2500} +{"lr": 9.91071424817309e-05, "data_time": 0.0014225006103515624, "loss": 0.4845857560634613, "time": 0.6314415693283081, "epoch": 1, "memory": 44139, "step": 2600} +{"lr": 9.903736360986979e-05, "data_time": 0.0014866352081298827, "loss": 0.4840073823928833, "time": 0.6320632696151733, "epoch": 1, "memory": 44139, "step": 2700} +{"lr": 9.89649890787935e-05, "data_time": 0.0014931678771972657, "loss": 0.4953353703022003, "time": 0.6288543939590454, "epoch": 1, "memory": 44139, "step": 2800} +{"lr": 9.889002315441642e-05, "data_time": 0.0013036489486694335, "loss": 0.51003999710083, "time": 0.6310796737670898, "epoch": 1, "memory": 44139, "step": 2900} +{"lr": 9.881247025539515e-05, "data_time": 0.0013866901397705078, "loss": 0.45273745656013487, "time": 0.6315259456634521, "epoch": 1, "memory": 44139, "step": 3000} +{"lr": 9.873233495286844e-05, "data_time": 0.0014451265335083008, "loss": 0.47733883559703827, "time": 0.6317728757858276, "epoch": 1, "memory": 44139, "step": 3100} +{"lr": 9.864962197018756e-05, "data_time": 0.0014280557632446289, "loss": 0.4770282328128815, "time": 0.631502914428711, "epoch": 1, "memory": 44139, "step": 3200} +{"lr": 9.856433618263779e-05, "data_time": 0.0016320705413818359, "loss": 0.45885944068431855, "time": 0.6312127113342285, "epoch": 1, "memory": 44139, "step": 3300} +{"lr": 9.847648261715132e-05, "data_time": 0.0015032291412353516, "loss": 0.4754169762134552, "time": 0.6317023038864136, "epoch": 1, "memory": 44139, "step": 3400} +{"lr": 9.838606645201093e-05, "data_time": 0.0014789342880249024, "loss": 0.46152358651161196, "time": 0.6314425945281983, "epoch": 1, "memory": 44139, "step": 3500} +{"lr": 9.829309301654456e-05, "data_time": 0.0014422178268432618, "loss": 0.45579715967178347, "time": 0.6320044040679932, "epoch": 1, "memory": 44139, "step": 3600} +{"lr": 9.819756779081127e-05, "data_time": 0.0013558149337768554, "loss": 0.47395158410072324, "time": 0.6315643548965454, "epoch": 1, "memory": 44139, "step": 3700} +{"lr": 9.809949640527814e-05, "data_time": 0.001401662826538086, "loss": 0.4785751700401306, "time": 0.6320651054382325, "epoch": 1, "memory": 44139, "step": 3800} +{"lr": 9.799888464048879e-05, "data_time": 0.0013675689697265625, "loss": 0.445124951004982, "time": 0.6320790767669677, "epoch": 1, "memory": 44139, "step": 3900} +{"lr": 9.789573842672242e-05, "data_time": 0.0013480424880981446, "loss": 0.4624592661857605, "time": 0.6311359167098999, "epoch": 1, "memory": 44139, "step": 4000} +{"accuracy/top1": 71.7806167602539, "data_time": 0.022229370950650768, "time": 1.3160906679490034, "step": 1} +{"lr": 9.7690615523361e-05, "data_time": 0.0018679141998291016, "loss": 0.44359633326530457, "time": 0.6317744970321655, "epoch": 2, "memory": 44140, "step": 4192} +{"lr": 9.758010405927656e-05, "data_time": 0.0013559103012084962, "loss": 0.47134405970573423, "time": 0.6318446636199951, "epoch": 2, "memory": 44140, "step": 4292} +{"lr": 9.746708283006492e-05, "data_time": 0.001389288902282715, "loss": 0.47403521537780763, "time": 0.6340468406677247, "epoch": 2, "memory": 44140, "step": 4392} +{"lr": 9.735155849744634e-05, "data_time": 0.0013835430145263672, "loss": 0.4599327057600021, "time": 0.6323191881179809, "epoch": 2, "memory": 44140, "step": 4492} +{"lr": 9.723353787067928e-05, "data_time": 0.0014330148696899414, "loss": 0.44250693917274475, "time": 0.6345848321914673, "epoch": 2, "memory": 44140, "step": 4592} +{"lr": 9.711302790615951e-05, "data_time": 0.0013781070709228515, "loss": 0.4755029261112213, "time": 0.631804084777832, "epoch": 2, "memory": 44140, "step": 4692} +{"lr": 9.699003570700986e-05, "data_time": 0.0013843059539794921, "loss": 0.4483084142208099, "time": 0.631246280670166, "epoch": 2, "memory": 44140, "step": 4792} +{"lr": 9.686456852266142e-05, "data_time": 0.001350736618041992, "loss": 0.4435199677944183, "time": 0.6318500995635986, "epoch": 2, "memory": 44140, "step": 4892} +{"lr": 9.673663374842644e-05, "data_time": 0.0013036489486694335, "loss": 0.45903160572052004, "time": 0.6315513134002686, "epoch": 2, "memory": 44140, "step": 4992} +{"lr": 9.660623892506218e-05, "data_time": 0.0013495445251464843, "loss": 0.45409590005874634, "time": 0.6326237201690674, "epoch": 2, "memory": 44140, "step": 5092} +{"lr": 9.647339173832693e-05, "data_time": 0.001483607292175293, "loss": 0.46190900802612306, "time": 0.6318344831466675, "epoch": 2, "memory": 44140, "step": 5192} +{"lr": 9.633810001852653e-05, "data_time": 0.0013091564178466797, "loss": 0.4453065633773804, "time": 0.6312918424606323, "epoch": 2, "memory": 44140, "step": 5292} +{"lr": 9.620037174005274e-05, "data_time": 0.0016185283660888673, "loss": 0.45716494917869566, "time": 0.631353211402893, "epoch": 2, "memory": 44140, "step": 5392} +{"lr": 9.606021502091389e-05, "data_time": 0.0014839649200439453, "loss": 0.49206840991973877, "time": 0.6333415746688843, "epoch": 2, "memory": 44140, "step": 5492} +{"lr": 9.591763812225569e-05, "data_time": 0.001655721664428711, "loss": 0.4525274097919464, "time": 0.6328972339630127, "epoch": 2, "memory": 44140, "step": 5592} +{"lr": 9.577264944787459e-05, "data_time": 0.0015350818634033204, "loss": 0.46427778005599973, "time": 0.6310419559478759, "epoch": 2, "memory": 44140, "step": 5692} +{"lr": 9.562525754372252e-05, "data_time": 0.0014998674392700194, "loss": 0.4221505582332611, "time": 0.6318845987319947, "epoch": 2, "memory": 44140, "step": 5792} +{"lr": 9.547547109740282e-05, "data_time": 0.0013521671295166015, "loss": 0.4527347683906555, "time": 0.631903076171875, "epoch": 2, "memory": 44140, "step": 5892} +{"lr": 9.532329893765858e-05, "data_time": 0.0015967845916748046, "loss": 0.446257284283638, "time": 0.6303370714187622, "epoch": 2, "memory": 44140, "step": 5992} +{"lr": 9.516875003385223e-05, "data_time": 0.0014877080917358398, "loss": 0.42972763180732726, "time": 0.6328753232955933, "epoch": 2, "memory": 44140, "step": 6092} +{"lr": 9.501183349543653e-05, "data_time": 0.0015574216842651368, "loss": 0.4281292766332626, "time": 0.6330648899078369, "epoch": 2, "memory": 44140, "step": 6192} +{"lr": 9.48525585714179e-05, "data_time": 0.0014125347137451173, "loss": 0.44764353334903717, "time": 0.6313239097595215, "epoch": 2, "memory": 44140, "step": 6292} +{"lr": 9.469093464981119e-05, "data_time": 0.001408553123474121, "loss": 0.43747061491012573, "time": 0.6313658714294433, "epoch": 2, "memory": 44140, "step": 6392} +{"lr": 9.452697125708636e-05, "data_time": 0.0015063762664794921, "loss": 0.44163825511932375, "time": 0.6310720682144165, "epoch": 2, "memory": 44140, "step": 6492} +{"lr": 9.436067805760705e-05, "data_time": 0.001448678970336914, "loss": 0.4352395623922348, "time": 0.6314118862152099, "epoch": 2, "memory": 44140, "step": 6592} +{"lr": 9.419206485306072e-05, "data_time": 0.0014289617538452148, "loss": 0.4278436750173569, "time": 0.6314183235168457, "epoch": 2, "memory": 44140, "step": 6692} +{"lr": 9.4021141581881e-05, "data_time": 0.0013917922973632813, "loss": 0.4349859803915024, "time": 0.6318498373031616, "epoch": 2, "memory": 44140, "step": 6792} +{"lr": 9.384791831866193e-05, "data_time": 0.0014430999755859375, "loss": 0.42788690626621245, "time": 0.6315459489822388, "epoch": 2, "memory": 44140, "step": 6892} +{"lr": 9.367240527356428e-05, "data_time": 0.0014315128326416015, "loss": 0.4437686026096344, "time": 0.6311376094818115, "epoch": 2, "memory": 44140, "step": 6992} +{"lr": 9.349461279171362e-05, "data_time": 0.0014157533645629884, "loss": 0.44617689549922945, "time": 0.6320367097854614, "epoch": 2, "memory": 44140, "step": 7092} +{"lr": 9.33145513525904e-05, "data_time": 0.0013987064361572266, "loss": 0.4393774211406708, "time": 0.6314810991287232, "epoch": 2, "memory": 44140, "step": 7192} +{"lr": 9.313223156941242e-05, "data_time": 0.0013824701309204102, "loss": 0.44216046333312986, "time": 0.6343540191650391, "epoch": 2, "memory": 44140, "step": 7292} +{"lr": 9.294766418850942e-05, "data_time": 0.001382756233215332, "loss": 0.4563088804483414, "time": 0.6314673900604248, "epoch": 2, "memory": 44140, "step": 7392} +{"lr": 9.276086008868929e-05, "data_time": 0.00142822265625, "loss": 0.4428665488958359, "time": 0.6312015771865844, "epoch": 2, "memory": 44140, "step": 7492} +{"lr": 9.257183028059726e-05, "data_time": 0.0016424417495727538, "loss": 0.4485490798950195, "time": 0.641039776802063, "epoch": 2, "memory": 44140, "step": 7592} +{"lr": 9.238058590606654e-05, "data_time": 0.0015211105346679688, "loss": 0.41403867602348327, "time": 0.6308777332305908, "epoch": 2, "memory": 44140, "step": 7692} +{"lr": 9.218713823746189e-05, "data_time": 0.0014520406723022461, "loss": 0.4221742510795593, "time": 0.6324946165084839, "epoch": 2, "memory": 44140, "step": 7792} +{"lr": 9.199149867701492e-05, "data_time": 0.0014151573181152345, "loss": 0.46067075431346893, "time": 0.6320663690567017, "epoch": 2, "memory": 44140, "step": 7892} +{"lr": 9.179367875615232e-05, "data_time": 0.0014046192169189452, "loss": 0.46722287535667417, "time": 0.632951807975769, "epoch": 2, "memory": 44140, "step": 7992} +{"lr": 9.159369013481585e-05, "data_time": 0.0014615058898925781, "loss": 0.42448204159736636, "time": 0.6318077325820923, "epoch": 2, "memory": 44140, "step": 8092} +{"accuracy/top1": 72.94573211669922, "data_time": 0.017711853981018065, "time": 1.2847278495629628, "step": 2} +{"lr": 9.120367596460475e-05, "data_time": 0.0014551162719726562, "loss": 0.4473511815071106, "time": 0.6332897424697876, "epoch": 3, "memory": 44140, "step": 8284} +{"lr": 9.099742266672751e-05, "data_time": 0.0016311883926391601, "loss": 0.4212220788002014, "time": 0.6335990190505981, "epoch": 3, "memory": 44140, "step": 8384} +{"lr": 9.078904760146445e-05, "data_time": 0.0013982772827148438, "loss": 0.4284248322248459, "time": 0.634125804901123, "epoch": 3, "memory": 44140, "step": 8484} +{"lr": 9.057856305090082e-05, "data_time": 0.0013855457305908202, "loss": 0.4249072462320328, "time": 0.6312395095825195, "epoch": 3, "memory": 44140, "step": 8584} +{"lr": 9.036598142145979e-05, "data_time": 0.0014184236526489258, "loss": 0.42530823945999147, "time": 0.6337924242019654, "epoch": 3, "memory": 44140, "step": 8684} +{"lr": 9.015131524317079e-05, "data_time": 0.001400160789489746, "loss": 0.4409077078104019, "time": 0.6319032192230225, "epoch": 3, "memory": 44140, "step": 8784} +{"lr": 8.993457716893124e-05, "data_time": 0.001379108428955078, "loss": 0.423635196685791, "time": 0.6323798179626465, "epoch": 3, "memory": 44140, "step": 8884} +{"lr": 8.971577997376067e-05, "data_time": 0.0013748884201049804, "loss": 0.4090465158224106, "time": 0.6361242771148682, "epoch": 3, "memory": 44140, "step": 8984} +{"lr": 8.949493655404777e-05, "data_time": 0.0016320228576660156, "loss": 0.42824044823646545, "time": 0.6332427024841308, "epoch": 3, "memory": 44140, "step": 9084} +{"lr": 8.927205992679013e-05, "data_time": 0.0016265153884887696, "loss": 0.4245839774608612, "time": 0.6346089363098144, "epoch": 3, "memory": 44140, "step": 9184} +{"lr": 8.9047163228827e-05, "data_time": 0.0014111757278442382, "loss": 0.4436920195817947, "time": 0.6333080291748047, "epoch": 3, "memory": 44140, "step": 9284} +{"lr": 8.88202597160652e-05, "data_time": 0.0013289690017700196, "loss": 0.4244053095579147, "time": 0.6321623802185059, "epoch": 3, "memory": 44140, "step": 9384} +{"lr": 8.859136276269775e-05, "data_time": 0.0013959169387817382, "loss": 0.4091134428977966, "time": 0.6338141441345215, "epoch": 3, "memory": 44140, "step": 9484} +{"lr": 8.836048586041507e-05, "data_time": 0.0014384031295776368, "loss": 0.4392902314662933, "time": 0.6321201086044311, "epoch": 3, "memory": 44140, "step": 9584} +{"lr": 8.812764261761037e-05, "data_time": 0.0013388633728027345, "loss": 0.4246595323085785, "time": 0.6321587562561035, "epoch": 3, "memory": 44140, "step": 9684} +{"lr": 8.789284675857743e-05, "data_time": 0.001620030403137207, "loss": 0.4354802370071411, "time": 0.6323515415191651, "epoch": 3, "memory": 44140, "step": 9784} +{"lr": 8.76561121227013e-05, "data_time": 0.0015564680099487305, "loss": 0.43429321944713595, "time": 0.6408982276916504, "epoch": 3, "memory": 44140, "step": 9884} +{"lr": 8.741745266364282e-05, "data_time": 0.0015223503112792968, "loss": 0.4274744212627411, "time": 0.6323494911193848, "epoch": 3, "memory": 44140, "step": 9984} +{"lr": 8.717688244851635e-05, "data_time": 0.0016958475112915038, "loss": 0.4182525217533112, "time": 0.6326273918151856, "epoch": 3, "memory": 44140, "step": 10084} +{"lr": 8.693441565706024e-05, "data_time": 0.0014996767044067384, "loss": 0.41954270005226135, "time": 0.6322771549224854, "epoch": 3, "memory": 44140, "step": 10184} +{"lr": 8.669006658080132e-05, "data_time": 0.0015003442764282226, "loss": 0.4252572238445282, "time": 0.6315906286239624, "epoch": 3, "memory": 44140, "step": 10284} +{"lr": 8.644384962221226e-05, "data_time": 0.001618814468383789, "loss": 0.4026562452316284, "time": 0.6322749614715576, "epoch": 3, "memory": 44140, "step": 10384} +{"lr": 8.619577929386303e-05, "data_time": 0.0014399051666259765, "loss": 0.4505779087543488, "time": 0.6323697805404663, "epoch": 3, "memory": 44140, "step": 10484} +{"lr": 8.594587021756517e-05, "data_time": 0.0014118909835815429, "loss": 0.43633814454078673, "time": 0.6322449445724487, "epoch": 3, "memory": 44140, "step": 10584} +{"lr": 8.569413712351023e-05, "data_time": 0.0015891551971435546, "loss": 0.4132749974727631, "time": 0.6337039709091187, "epoch": 3, "memory": 44140, "step": 10684} +{"lr": 8.544059484940119e-05, "data_time": 0.0014904499053955077, "loss": 0.4413173168897629, "time": 0.6329502582550048, "epoch": 3, "memory": 44140, "step": 10784} +{"lr": 8.518525833957818e-05, "data_time": 0.0013824939727783204, "loss": 0.4279088795185089, "time": 0.63317391872406, "epoch": 3, "memory": 44140, "step": 10884} +{"lr": 8.49281426441377e-05, "data_time": 0.0014138221740722656, "loss": 0.4370129227638245, "time": 0.6330703735351563, "epoch": 3, "memory": 44140, "step": 10984} +{"lr": 8.466926291804501e-05, "data_time": 0.0012787818908691407, "loss": 0.43694463074207307, "time": 0.6357486248016357, "epoch": 3, "memory": 44140, "step": 11084} +{"lr": 8.440863442024147e-05, "data_time": 0.0017200469970703124, "loss": 0.4132837414741516, "time": 0.6355113744735718, "epoch": 3, "memory": 44140, "step": 11184} +{"lr": 8.414627251274488e-05, "data_time": 0.0015549659729003906, "loss": 0.42699059247970583, "time": 0.6320959329605103, "epoch": 3, "memory": 44140, "step": 11284} +{"lr": 8.388219265974407e-05, "data_time": 0.001401805877685547, "loss": 0.43332897424697875, "time": 0.6318453788757324, "epoch": 3, "memory": 44140, "step": 11384} +{"lr": 8.361641042668715e-05, "data_time": 0.001779031753540039, "loss": 0.4249531179666519, "time": 0.6340765237808228, "epoch": 3, "memory": 44140, "step": 11484} +{"lr": 8.33489414793644e-05, "data_time": 0.0021561384201049805, "loss": 0.4533049076795578, "time": 0.6464383602142334, "epoch": 3, "memory": 44140, "step": 11584} +{"lr": 8.30798015829848e-05, "data_time": 0.0023379802703857424, "loss": 0.4502921670675278, "time": 0.6342859506607056, "epoch": 3, "memory": 44140, "step": 11684} +{"lr": 8.280900660124665e-05, "data_time": 0.002496504783630371, "loss": 0.4467496007680893, "time": 0.6343088388442993, "epoch": 3, "memory": 44140, "step": 11784} +{"lr": 8.253657249540274e-05, "data_time": 0.0016772985458374024, "loss": 0.4151320159435272, "time": 0.6328318119049072, "epoch": 3, "memory": 44140, "step": 11884} +{"lr": 8.226251532331926e-05, "data_time": 0.0020305871963500976, "loss": 0.4178690493106842, "time": 0.6329146146774292, "epoch": 3, "memory": 44140, "step": 11984} +{"lr": 8.198685123852965e-05, "data_time": 0.0024167537689208985, "loss": 0.4292076587677002, "time": 0.6334588766098023, "epoch": 3, "memory": 44140, "step": 12084} +{"lr": 8.17095964892824e-05, "data_time": 0.0023377656936645506, "loss": 0.42308293879032133, "time": 0.6408001184463501, "epoch": 3, "memory": 44140, "step": 12184} +{"accuracy/top1": 75.0475845336914, "data_time": 0.020047646760940552, "time": 1.2880093296368917, "step": 3} +{"lr": 8.11728683565904e-05, "data_time": 0.0024878978729248047, "loss": 0.41675683856010437, "time": 0.6322898387908935, "epoch": 4, "memory": 44140, "step": 12376} +{"lr": 8.089106273476776e-05, "data_time": 0.00207364559173584, "loss": 0.4293960154056549, "time": 0.6331905603408814, "epoch": 4, "memory": 44140, "step": 12476} +{"lr": 8.060773103667566e-05, "data_time": 0.0024184226989746095, "loss": 0.42023499608039855, "time": 0.6328914403915405, "epoch": 4, "memory": 44140, "step": 12576} +{"lr": 8.03228899625083e-05, "data_time": 0.002521252632141113, "loss": 0.38831470608711244, "time": 0.6329829216003418, "epoch": 4, "memory": 44140, "step": 12676} +{"lr": 8.003655630142627e-05, "data_time": 0.0024059534072875975, "loss": 0.41458570659160615, "time": 0.6332283258438111, "epoch": 4, "memory": 44140, "step": 12776} +{"lr": 7.974874693056617e-05, "data_time": 0.002190995216369629, "loss": 0.42189861536026, "time": 0.6314792394638061, "epoch": 4, "memory": 44140, "step": 12876} +{"lr": 7.945947881404618e-05, "data_time": 0.002054762840270996, "loss": 0.43414068818092344, "time": 0.6348303318023681, "epoch": 4, "memory": 44140, "step": 12976} +{"lr": 7.916876900196646e-05, "data_time": 0.0019904375076293945, "loss": 0.4551405757665634, "time": 0.6324449777603149, "epoch": 4, "memory": 44140, "step": 13076} +{"lr": 7.887663462940359e-05, "data_time": 0.0018481731414794922, "loss": 0.4363859176635742, "time": 0.6329013347625733, "epoch": 4, "memory": 44140, "step": 13176} +{"lr": 7.858309291540111e-05, "data_time": 0.0019230127334594726, "loss": 0.4265198588371277, "time": 0.6333821535110473, "epoch": 4, "memory": 44140, "step": 13276} +{"lr": 7.828816116195411e-05, "data_time": 0.0022130727767944334, "loss": 0.4443864405155182, "time": 0.6324852228164672, "epoch": 4, "memory": 44140, "step": 13376} +{"lr": 7.799185675298994e-05, "data_time": 0.002134275436401367, "loss": 0.4278457522392273, "time": 0.6307125806808471, "epoch": 4, "memory": 44140, "step": 13476} +{"lr": 7.769419715334297e-05, "data_time": 0.0019055843353271485, "loss": 0.4094464063644409, "time": 0.6339537620544433, "epoch": 4, "memory": 44140, "step": 13576} +{"lr": 7.739519990772583e-05, "data_time": 0.0023879289627075197, "loss": 0.43034936487674713, "time": 0.6329025745391845, "epoch": 4, "memory": 44140, "step": 13676} +{"lr": 7.709488263969475e-05, "data_time": 0.0020322322845458983, "loss": 0.4365843415260315, "time": 0.6321660041809082, "epoch": 4, "memory": 44140, "step": 13776} +{"lr": 7.67932630506109e-05, "data_time": 0.002697300910949707, "loss": 0.4132125288248062, "time": 0.6336849689483642, "epoch": 4, "memory": 44140, "step": 13876} +{"lr": 7.64903589185975e-05, "data_time": 0.0019088029861450196, "loss": 0.4382692277431488, "time": 0.6323899984359741, "epoch": 4, "memory": 44140, "step": 13976} +{"lr": 7.618618809749134e-05, "data_time": 0.0022614240646362306, "loss": 0.42109595239162445, "time": 0.6325977802276611, "epoch": 4, "memory": 44140, "step": 14076} +{"lr": 7.588076851579028e-05, "data_time": 0.0023856401443481446, "loss": 0.43642314672470095, "time": 0.6321729421615601, "epoch": 4, "memory": 44140, "step": 14176} +{"lr": 7.557411817559755e-05, "data_time": 0.002448272705078125, "loss": 0.4302003622055054, "time": 0.6361776828765869, "epoch": 4, "memory": 44140, "step": 14276} +{"lr": 7.526625515155957e-05, "data_time": 0.0019052743911743164, "loss": 0.4075252503156662, "time": 0.6330398082733154, "epoch": 4, "memory": 44140, "step": 14376} +{"lr": 7.495719758980122e-05, "data_time": 0.0024227142333984376, "loss": 0.40171490609645844, "time": 0.6321627378463746, "epoch": 4, "memory": 44140, "step": 14476} +{"lr": 7.464696370685586e-05, "data_time": 0.0031921863555908203, "loss": 0.42775468826293944, "time": 0.6336983203887939, "epoch": 4, "memory": 44140, "step": 14576} +{"lr": 7.433557178859198e-05, "data_time": 0.0019622087478637696, "loss": 0.44475115835666656, "time": 0.6332455396652221, "epoch": 4, "memory": 44140, "step": 14676} +{"lr": 7.402304018913512e-05, "data_time": 0.0018522500991821288, "loss": 0.41916805803775786, "time": 0.6318212509155273, "epoch": 4, "memory": 44140, "step": 14776} +{"lr": 7.37093873297861e-05, "data_time": 0.0019346237182617187, "loss": 0.4231825232505798, "time": 0.6315581321716308, "epoch": 4, "memory": 44140, "step": 14876} +{"lr": 7.339463169793563e-05, "data_time": 0.002269434928894043, "loss": 0.4665799081325531, "time": 0.6322512626647949, "epoch": 4, "memory": 44140, "step": 14976} +{"lr": 7.307879184597369e-05, "data_time": 0.002082633972167969, "loss": 0.43822734951972964, "time": 0.6315053939819336, "epoch": 4, "memory": 44140, "step": 15076} +{"lr": 7.276188639019704e-05, "data_time": 0.0024214267730712892, "loss": 0.38338116407394407, "time": 0.6320795059204102, "epoch": 4, "memory": 44140, "step": 15176} +{"lr": 7.244393400971127e-05, "data_time": 0.003145432472229004, "loss": 0.4197797864675522, "time": 0.6340341806411743, "epoch": 4, "memory": 44140, "step": 15276} +{"lr": 7.212495344533006e-05, "data_time": 0.00262296199798584, "loss": 0.43974902033805846, "time": 0.6334179639816284, "epoch": 4, "memory": 44140, "step": 15376} +{"lr": 7.180496349847055e-05, "data_time": 0.0020112991333007812, "loss": 0.4346621215343475, "time": 0.6322306871414185, "epoch": 4, "memory": 44140, "step": 15476} +{"lr": 7.148398303004499e-05, "data_time": 0.0026148319244384765, "loss": 0.4178128480911255, "time": 0.6347824096679687, "epoch": 4, "memory": 44140, "step": 15576} +{"lr": 7.11620309593493e-05, "data_time": 0.0018913030624389648, "loss": 0.44328099191188813, "time": 0.6322627544403077, "epoch": 4, "memory": 44140, "step": 15676} +{"lr": 7.083912626294795e-05, "data_time": 0.0023617982864379884, "loss": 0.4336129128932953, "time": 0.6343975782394409, "epoch": 4, "memory": 44140, "step": 15776} +{"lr": 7.051528797355494e-05, "data_time": 0.0022262811660766603, "loss": 0.45349820256233214, "time": 0.6330487251281738, "epoch": 4, "memory": 44140, "step": 15876} +{"lr": 7.019053517891214e-05, "data_time": 0.002340221405029297, "loss": 0.42756457924842833, "time": 0.6318708181381225, "epoch": 4, "memory": 44140, "step": 15976} +{"lr": 6.986488702066473e-05, "data_time": 0.0021955490112304686, "loss": 0.4229933112859726, "time": 0.6330044269561768, "epoch": 4, "memory": 44140, "step": 16076} +{"lr": 6.953836269323245e-05, "data_time": 0.0019453763961791992, "loss": 0.441475448012352, "time": 0.6325273275375366, "epoch": 4, "memory": 44140, "step": 16176} +{"lr": 6.921098144267838e-05, "data_time": 0.002437734603881836, "loss": 0.4169649541378021, "time": 0.6327932834625244, "epoch": 4, "memory": 44140, "step": 16276} +{"accuracy/top1": 77.19245910644531, "data_time": 0.01820972760518392, "time": 1.2874916394551594, "step": 4} +{"lr": 6.858007803673889e-05, "data_time": 0.0017416238784790038, "loss": 0.44968058466911315, "time": 0.6320735692977906, "epoch": 5, "memory": 44140, "step": 16468} +{"lr": 6.825030518869046e-05, "data_time": 0.0017281532287597655, "loss": 0.4209025502204895, "time": 0.6332170486450195, "epoch": 5, "memory": 44140, "step": 16568} +{"lr": 6.791975133846686e-05, "data_time": 0.0021192312240600588, "loss": 0.4412855625152588, "time": 0.6322494268417358, "epoch": 5, "memory": 44140, "step": 16668} +{"lr": 6.758843596964013e-05, "data_time": 0.002190709114074707, "loss": 0.4384043008089066, "time": 0.632795763015747, "epoch": 5, "memory": 44140, "step": 16768} +{"lr": 6.725637861066804e-05, "data_time": 0.002461957931518555, "loss": 0.43394399881362916, "time": 0.6330869674682618, "epoch": 5, "memory": 44140, "step": 16868} +{"lr": 6.692359883374264e-05, "data_time": 0.0015211820602416993, "loss": 0.44528878331184385, "time": 0.6321467399597168, "epoch": 5, "memory": 44140, "step": 16968} +{"lr": 6.659011625363718e-05, "data_time": 0.00165865421295166, "loss": 0.4454505920410156, "time": 0.6323389768600464, "epoch": 5, "memory": 44140, "step": 17068} +{"lr": 6.625595052654927e-05, "data_time": 0.0014862775802612304, "loss": 0.41868581175804137, "time": 0.6319795608520508, "epoch": 5, "memory": 44140, "step": 17168} +{"lr": 6.592112134894315e-05, "data_time": 0.0015000104904174805, "loss": 0.41661250591278076, "time": 0.6344560384750366, "epoch": 5, "memory": 44140, "step": 17268} +{"lr": 6.558564845638802e-05, "data_time": 0.0024153947830200194, "loss": 0.41712948083877566, "time": 0.6343169927597045, "epoch": 5, "memory": 44140, "step": 17368} +{"lr": 6.524955162239518e-05, "data_time": 0.0014957666397094726, "loss": 0.4421685069799423, "time": 0.6326616764068603, "epoch": 5, "memory": 44140, "step": 17468} +{"lr": 6.491285065725238e-05, "data_time": 0.0013540029525756837, "loss": 0.46646787524223327, "time": 0.6323060274124146, "epoch": 5, "memory": 44140, "step": 17568} +{"lr": 6.457556540685618e-05, "data_time": 0.0017585277557373047, "loss": 0.4295129120349884, "time": 0.6341775417327881, "epoch": 5, "memory": 44140, "step": 17668} +{"lr": 6.423771575154208e-05, "data_time": 0.0014645576477050782, "loss": 0.4353230714797974, "time": 0.6336427927017212, "epoch": 5, "memory": 44140, "step": 17768} +{"lr": 6.389932160491309e-05, "data_time": 0.0017275810241699219, "loss": 0.41458328664302824, "time": 0.632085132598877, "epoch": 5, "memory": 44140, "step": 17868} +{"lr": 6.356040291266556e-05, "data_time": 0.0014629364013671875, "loss": 0.45556470155715945, "time": 0.6318282127380371, "epoch": 5, "memory": 44140, "step": 17968} +{"lr": 6.322097965141391e-05, "data_time": 0.0014632463455200196, "loss": 0.4440807819366455, "time": 0.6336803913116456, "epoch": 5, "memory": 44140, "step": 18068} +{"lr": 6.288107182751267e-05, "data_time": 0.0014612913131713868, "loss": 0.40304665863513944, "time": 0.6333890676498413, "epoch": 5, "memory": 44140, "step": 18168} +{"lr": 6.254069947587782e-05, "data_time": 0.0014941930770874024, "loss": 0.45084108710289, "time": 0.6324883222579956, "epoch": 5, "memory": 44140, "step": 18268} +{"lr": 6.219988265880537e-05, "data_time": 0.0015841484069824218, "loss": 0.41318759620189666, "time": 0.6371341943740845, "epoch": 5, "memory": 44140, "step": 18368} +{"lr": 6.185864146478941e-05, "data_time": 0.0017355918884277345, "loss": 0.46838268637657166, "time": 0.6336296319961547, "epoch": 5, "memory": 44140, "step": 18468} +{"lr": 6.151699600733752e-05, "data_time": 0.0019960641860961915, "loss": 0.40581959784030913, "time": 0.635711407661438, "epoch": 5, "memory": 44140, "step": 18568} +{"lr": 6.117496642378554e-05, "data_time": 0.001602506637573242, "loss": 0.44728600382804873, "time": 0.6328262090682983, "epoch": 5, "memory": 44140, "step": 18668} +{"lr": 6.0832572874110315e-05, "data_time": 0.0016323089599609374, "loss": 0.43281572461128237, "time": 0.6335597276687622, "epoch": 5, "memory": 44140, "step": 18768} +{"lr": 6.0489835539741924e-05, "data_time": 0.0014596939086914062, "loss": 0.43951316475868224, "time": 0.6353926420211792, "epoch": 5, "memory": 44140, "step": 18868} +{"lr": 6.0146774622374006e-05, "data_time": 0.0015111684799194336, "loss": 0.4285846263170242, "time": 0.6356100082397461, "epoch": 5, "memory": 44140, "step": 18968} +{"lr": 5.980341034277274e-05, "data_time": 0.0012971162796020508, "loss": 0.4076012700796127, "time": 0.6328089952468872, "epoch": 5, "memory": 44140, "step": 19068} +{"lr": 5.945976293958505e-05, "data_time": 0.0015245914459228516, "loss": 0.4158323615789413, "time": 0.6349369525909424, "epoch": 5, "memory": 44140, "step": 19168} +{"lr": 5.911585266814587e-05, "data_time": 0.0013681650161743164, "loss": 0.44133509397506715, "time": 0.6346670150756836, "epoch": 5, "memory": 44140, "step": 19268} +{"lr": 5.877169979928426e-05, "data_time": 0.0014252185821533204, "loss": 0.4374487638473511, "time": 0.6336438179016113, "epoch": 5, "memory": 44140, "step": 19368} +{"lr": 5.842732461812828e-05, "data_time": 0.0015235424041748046, "loss": 0.43618308901786806, "time": 0.6352621078491211, "epoch": 5, "memory": 44140, "step": 19468} +{"lr": 5.8082747422909646e-05, "data_time": 0.0018047332763671876, "loss": 0.42677063345909116, "time": 0.632779335975647, "epoch": 5, "memory": 44140, "step": 19568} +{"lr": 5.7737988523767406e-05, "data_time": 0.001514101028442383, "loss": 0.4765779942274094, "time": 0.6328549146652221, "epoch": 5, "memory": 44140, "step": 19668} +{"lr": 5.739306824155042e-05, "data_time": 0.0024080276489257812, "loss": 0.4456770420074463, "time": 0.635212755203247, "epoch": 5, "memory": 44140, "step": 19768} +{"lr": 5.704800690661993e-05, "data_time": 0.001786661148071289, "loss": 0.43700712323188784, "time": 0.632424783706665, "epoch": 5, "memory": 44140, "step": 19868} +{"lr": 5.6702824857650994e-05, "data_time": 0.0023291587829589845, "loss": 0.43297325968742373, "time": 0.6327181100845337, "epoch": 5, "memory": 44140, "step": 19968} +{"lr": 5.635754244043411e-05, "data_time": 0.0019201278686523438, "loss": 0.4218790054321289, "time": 0.6360872030258179, "epoch": 5, "memory": 44140, "step": 20068} +{"lr": 5.60121800066754e-05, "data_time": 0.0016581296920776367, "loss": 0.4351540684700012, "time": 0.6338484764099122, "epoch": 5, "memory": 44140, "step": 20168} +{"lr": 5.5666757912797516e-05, "data_time": 0.002289009094238281, "loss": 0.41573789715766907, "time": 0.6346448183059692, "epoch": 5, "memory": 44140, "step": 20268} +{"lr": 5.532129651873955e-05, "data_time": 0.0014219045639038085, "loss": 0.411351689696312, "time": 0.6326027631759643, "epoch": 5, "memory": 44140, "step": 20368} +{"accuracy/top1": 79.04109954833984, "data_time": 0.018181178967158, "time": 1.2852958897749582, "step": 5} +{"lr": 5.465797506068693e-05, "data_time": 0.0014163494110107423, "loss": 0.3863424897193909, "time": 0.6321403026580811, "epoch": 6, "memory": 44140, "step": 20560} +{"lr": 5.431251545044144e-05, "data_time": 0.0014596223831176759, "loss": 0.43077988028526304, "time": 0.6337174415588379, "epoch": 6, "memory": 44140, "step": 20660} +{"lr": 5.396709636204968e-05, "data_time": 0.0015720129013061523, "loss": 0.4430558800697327, "time": 0.6326960325241089, "epoch": 6, "memory": 44140, "step": 20760} +{"lr": 5.3621738155273566e-05, "data_time": 0.001405954360961914, "loss": 0.42521955966949465, "time": 0.6327214717864991, "epoch": 6, "memory": 44140, "step": 20860} +{"lr": 5.3276461186286584e-05, "data_time": 0.0013530492782592774, "loss": 0.4332079946994781, "time": 0.6334285020828248, "epoch": 6, "memory": 44140, "step": 20960} +{"lr": 5.293128580647384e-05, "data_time": 0.0015474081039428711, "loss": 0.4298450410366058, "time": 0.6326422929763794, "epoch": 6, "memory": 44140, "step": 21060} +{"lr": 5.258623236123256e-05, "data_time": 0.0015361309051513672, "loss": 0.44527106285095214, "time": 0.6339249849319458, "epoch": 6, "memory": 44140, "step": 21160} +{"lr": 5.224132118877281e-05, "data_time": 0.0014399051666259765, "loss": 0.4186409264802933, "time": 0.6310220003128052, "epoch": 6, "memory": 44140, "step": 21260} +{"lr": 5.189657261891887e-05, "data_time": 0.0015101194381713866, "loss": 0.43008764982223513, "time": 0.6324338912963867, "epoch": 6, "memory": 44140, "step": 21360} +{"lr": 5.1552006971910974e-05, "data_time": 0.0017882108688354493, "loss": 0.42609631419181826, "time": 0.6323167562484742, "epoch": 6, "memory": 44140, "step": 21460} +{"lr": 5.120764455720728e-05, "data_time": 0.002320599555969238, "loss": 0.40623250007629397, "time": 0.6318898439407349, "epoch": 6, "memory": 44140, "step": 21560} +{"lr": 5.0863505672287034e-05, "data_time": 0.0016577482223510743, "loss": 0.4499622344970703, "time": 0.6329441785812377, "epoch": 6, "memory": 44140, "step": 21660} +{"lr": 5.051961060145421e-05, "data_time": 0.0014844894409179687, "loss": 0.426160603761673, "time": 0.6338789939880372, "epoch": 6, "memory": 44140, "step": 21760} +{"lr": 5.017597961464177e-05, "data_time": 0.0017120838165283203, "loss": 0.38233913779258727, "time": 0.6330686330795288, "epoch": 6, "memory": 44140, "step": 21860} +{"lr": 4.983263296621707e-05, "data_time": 0.0015595912933349609, "loss": 0.45386329889297483, "time": 0.6337078809738159, "epoch": 6, "memory": 44140, "step": 21960} +{"lr": 4.948959089378785e-05, "data_time": 0.0017532587051391601, "loss": 0.43574273586273193, "time": 0.632140064239502, "epoch": 6, "memory": 44140, "step": 22060} +{"lr": 4.91468736170094e-05, "data_time": 0.0014657020568847657, "loss": 0.4344453752040863, "time": 0.6327176570892334, "epoch": 6, "memory": 44140, "step": 22160} +{"lr": 4.880450133639314e-05, "data_time": 0.0014443635940551759, "loss": 0.4336450189352036, "time": 0.6333205223083496, "epoch": 6, "memory": 44140, "step": 22260} +{"lr": 4.8462494232115335e-05, "data_time": 0.0014873266220092774, "loss": 0.43560277223587035, "time": 0.6326103210449219, "epoch": 6, "memory": 44140, "step": 22360} +{"lr": 4.812087246282806e-05, "data_time": 0.001444840431213379, "loss": 0.4077687829732895, "time": 0.6332634925842285, "epoch": 6, "memory": 44140, "step": 22460} +{"lr": 4.7779656164470954e-05, "data_time": 0.001477670669555664, "loss": 0.4258314102888107, "time": 0.6323377609252929, "epoch": 6, "memory": 44140, "step": 22560} +{"lr": 4.7438865449084246e-05, "data_time": 0.0015760183334350586, "loss": 0.42388800382614134, "time": 0.6322774887084961, "epoch": 6, "memory": 44140, "step": 22660} +{"lr": 4.709852040362338e-05, "data_time": 0.001534748077392578, "loss": 0.457219535112381, "time": 0.6332423448562622, "epoch": 6, "memory": 44140, "step": 22760} +{"lr": 4.675864108877499e-05, "data_time": 0.001492929458618164, "loss": 0.44097004532814027, "time": 0.6326010227203369, "epoch": 6, "memory": 44140, "step": 22860} +{"lr": 4.641924753777464e-05, "data_time": 0.001659989356994629, "loss": 0.4451592117547989, "time": 0.6344381093978881, "epoch": 6, "memory": 44140, "step": 22960} +{"lr": 4.608035975522578e-05, "data_time": 0.002403736114501953, "loss": 0.4552268981933594, "time": 0.6323726415634155, "epoch": 6, "memory": 44140, "step": 23060} +{"lr": 4.574199771592085e-05, "data_time": 0.0014498233795166016, "loss": 0.4126337796449661, "time": 0.633358645439148, "epoch": 6, "memory": 44140, "step": 23160} +{"lr": 4.540418136366372e-05, "data_time": 0.001405811309814453, "loss": 0.3933737754821777, "time": 0.6323083400726318, "epoch": 6, "memory": 44140, "step": 23260} +{"lr": 4.506693061009437e-05, "data_time": 0.0014887809753417968, "loss": 0.45793297290802004, "time": 0.6364764451980591, "epoch": 6, "memory": 44140, "step": 23360} +{"lr": 4.473026533351511e-05, "data_time": 0.0015930414199829101, "loss": 0.4237523913383484, "time": 0.6339193105697631, "epoch": 6, "memory": 44140, "step": 23460} +{"lr": 4.439420537771888e-05, "data_time": 0.0016127347946166993, "loss": 0.422735595703125, "time": 0.6327435970306396, "epoch": 6, "memory": 44140, "step": 23560} +{"lr": 4.405877055081978e-05, "data_time": 0.002032613754272461, "loss": 0.4336425542831421, "time": 0.6359319448471069, "epoch": 6, "memory": 44140, "step": 23660} +{"lr": 4.3723980624085373e-05, "data_time": 0.0014849662780761718, "loss": 0.4337951481342316, "time": 0.6327850103378296, "epoch": 6, "memory": 44140, "step": 23760} +{"lr": 4.3389855330771394e-05, "data_time": 0.001927661895751953, "loss": 0.45388599038124083, "time": 0.6324433803558349, "epoch": 6, "memory": 44140, "step": 23860} +{"lr": 4.305641436495869e-05, "data_time": 0.0015739679336547851, "loss": 0.44674779772758483, "time": 0.6343928813934326, "epoch": 6, "memory": 44140, "step": 23960} +{"lr": 4.272367738039222e-05, "data_time": 0.0017908096313476562, "loss": 0.4293277680873871, "time": 0.6322276830673218, "epoch": 6, "memory": 44140, "step": 24060} +{"lr": 4.239166398932278e-05, "data_time": 0.0015479326248168945, "loss": 0.4475178837776184, "time": 0.6325698614120483, "epoch": 6, "memory": 44140, "step": 24160} +{"lr": 4.206039376135108e-05, "data_time": 0.0016879796981811523, "loss": 0.4636631578207016, "time": 0.6325940370559693, "epoch": 6, "memory": 44140, "step": 24260} +{"lr": 4.1729886222273985e-05, "data_time": 0.0018097400665283204, "loss": 0.45206343829631807, "time": 0.634130597114563, "epoch": 6, "memory": 44140, "step": 24360} +{"lr": 4.140016085293389e-05, "data_time": 0.001535654067993164, "loss": 0.4263759821653366, "time": 0.6329107999801635, "epoch": 6, "memory": 44140, "step": 24460} +{"accuracy/top1": 80.93937683105469, "data_time": 0.017462424437204995, "time": 1.2861662606398265, "step": 6} +{"lr": 4.0769351868715556e-05, "data_time": 0.0015712738037109374, "loss": 0.4451601356267929, "time": 0.6331160545349122, "epoch": 7, "memory": 44140, "step": 24652} +{"lr": 4.044202148988293e-05, "data_time": 0.0015348672866821289, "loss": 0.441101410984993, "time": 0.6327461957931518, "epoch": 7, "memory": 44140, "step": 24752} +{"lr": 4.0115549190371644e-05, "data_time": 0.0014959812164306641, "loss": 0.4430814743041992, "time": 0.6337475538253784, "epoch": 7, "memory": 44140, "step": 24852} +{"lr": 3.978995421317808e-05, "data_time": 0.0014377832412719727, "loss": 0.4063778817653656, "time": 0.6330408334732056, "epoch": 7, "memory": 44140, "step": 24952} +{"lr": 3.94652557495873e-05, "data_time": 0.0021008729934692385, "loss": 0.4139744132757187, "time": 0.6326706171035766, "epoch": 7, "memory": 44140, "step": 25052} +{"lr": 3.914147293804207e-05, "data_time": 0.0015067577362060547, "loss": 0.45996909141540526, "time": 0.632449722290039, "epoch": 7, "memory": 44140, "step": 25152} +{"lr": 3.881862486301448e-05, "data_time": 0.0014073610305786132, "loss": 0.4257462203502655, "time": 0.6325594186782837, "epoch": 7, "memory": 44140, "step": 25252} +{"lr": 3.849673055388122e-05, "data_time": 0.0015043973922729491, "loss": 0.4563670545816422, "time": 0.6327755451202393, "epoch": 7, "memory": 44140, "step": 25352} +{"lr": 3.817580898380197e-05, "data_time": 0.0013645410537719727, "loss": 0.4589421510696411, "time": 0.6324453830718995, "epoch": 7, "memory": 44140, "step": 25452} +{"lr": 3.7855879068600975e-05, "data_time": 0.0023953914642333984, "loss": 0.43817236125469206, "time": 0.632078742980957, "epoch": 7, "memory": 44140, "step": 25552} +{"lr": 3.753695966565216e-05, "data_time": 0.001933908462524414, "loss": 0.4153300642967224, "time": 0.6317489624023438, "epoch": 7, "memory": 44140, "step": 25652} +{"lr": 3.7219069572767636e-05, "data_time": 0.0021514177322387697, "loss": 0.42778047621250154, "time": 0.6320550441741943, "epoch": 7, "memory": 44140, "step": 25752} +{"lr": 3.6902227527089725e-05, "data_time": 0.002036738395690918, "loss": 0.41400990784168246, "time": 0.634029746055603, "epoch": 7, "memory": 44140, "step": 25852} +{"lr": 3.65864522039865e-05, "data_time": 0.0018081903457641602, "loss": 0.43593273758888246, "time": 0.6349156856536865, "epoch": 7, "memory": 44140, "step": 25952} +{"lr": 3.627176221595109e-05, "data_time": 0.0019333600997924806, "loss": 0.43460625410079956, "time": 0.6339972734451294, "epoch": 7, "memory": 44140, "step": 26052} +{"lr": 3.595817611150461e-05, "data_time": 0.0020090579986572266, "loss": 0.42278642058372495, "time": 0.6337616443634033, "epoch": 7, "memory": 44140, "step": 26152} +{"lr": 3.5645712374102824e-05, "data_time": 0.027292203903198243, "loss": 0.43919208347797395, "time": 1.0241565704345703, "epoch": 7, "memory": 44140, "step": 26252} +{"lr": 3.533438942104663e-05, "data_time": 0.002497363090515137, "loss": 0.43299540877342224, "time": 0.6341647148132324, "epoch": 7, "memory": 44140, "step": 26352} +{"lr": 3.5024225602396704e-05, "data_time": 0.0026648759841918944, "loss": 0.4354636311531067, "time": 0.6350600242614746, "epoch": 7, "memory": 44140, "step": 26452} +{"lr": 3.4715239199891735e-05, "data_time": 0.0017151594161987304, "loss": 0.4206331014633179, "time": 0.6419120073318482, "epoch": 7, "memory": 44140, "step": 26552} +{"lr": 3.44074484258709e-05, "data_time": 0.0015797376632690429, "loss": 0.43789590895175934, "time": 0.6338491678237915, "epoch": 7, "memory": 44140, "step": 26652} +{"lr": 3.4100871422200426e-05, "data_time": 0.0014853239059448241, "loss": 0.4500276446342468, "time": 0.6319789171218873, "epoch": 7, "memory": 44140, "step": 26752} +{"lr": 3.379552625920419e-05, "data_time": 0.0015302419662475586, "loss": 0.45275644659996034, "time": 0.6354724168777466, "epoch": 7, "memory": 44140, "step": 26852} +{"lr": 3.349143093459868e-05, "data_time": 0.0020882129669189454, "loss": 0.45181081891059877, "time": 0.6340699195861816, "epoch": 7, "memory": 44140, "step": 26952} +{"lr": 3.318860337243226e-05, "data_time": 0.001764535903930664, "loss": 0.4159604340791702, "time": 0.6330552816390991, "epoch": 7, "memory": 44140, "step": 27052} +{"lr": 3.288706142202845e-05, "data_time": 0.0022506237030029295, "loss": 0.4469432055950165, "time": 0.6332366228103637, "epoch": 7, "memory": 44140, "step": 27152} +{"lr": 3.25868228569341e-05, "data_time": 0.002504706382751465, "loss": 0.43903608024120333, "time": 0.6332183837890625, "epoch": 7, "memory": 44140, "step": 27252} +{"lr": 3.228790537387148e-05, "data_time": 0.001451563835144043, "loss": 0.43307424783706666, "time": 0.6329429864883422, "epoch": 7, "memory": 44140, "step": 27352} +{"lr": 3.199032659169556e-05, "data_time": 0.002272152900695801, "loss": 0.41505860090255736, "time": 0.6331403970718383, "epoch": 7, "memory": 44140, "step": 27452} +{"lr": 3.169410405035527e-05, "data_time": 0.001854705810546875, "loss": 0.4372330605983734, "time": 0.6322521448135376, "epoch": 7, "memory": 44140, "step": 27552} +{"lr": 3.139925520985965e-05, "data_time": 0.0020429134368896485, "loss": 0.4525086939334869, "time": 0.6338441848754883, "epoch": 7, "memory": 44140, "step": 27652} +{"lr": 3.1105797449248916e-05, "data_time": 0.001802682876586914, "loss": 0.4206294059753418, "time": 0.6332003355026246, "epoch": 7, "memory": 44140, "step": 27752} +{"lr": 3.081374806556986e-05, "data_time": 0.0018229484558105469, "loss": 0.45739985406398775, "time": 0.6335875511169433, "epoch": 7, "memory": 44140, "step": 27852} +{"lr": 3.052312427285649e-05, "data_time": 0.0020348310470581056, "loss": 0.4342558741569519, "time": 0.633012056350708, "epoch": 7, "memory": 44140, "step": 27952} +{"lr": 3.0233943201115213e-05, "data_time": 0.0018118858337402345, "loss": 0.42804893255233767, "time": 0.6341001272201539, "epoch": 7, "memory": 44140, "step": 28052} +{"lr": 2.9946221895315525e-05, "data_time": 0.0020540952682495117, "loss": 0.41936197578907014, "time": 0.6324288129806519, "epoch": 7, "memory": 44140, "step": 28152} +{"lr": 2.9659977314384953e-05, "data_time": 0.0017561674118041991, "loss": 0.45191813111305235, "time": 0.6322898626327514, "epoch": 7, "memory": 44140, "step": 28252} +{"lr": 2.93752263302096e-05, "data_time": 0.0024352788925170897, "loss": 0.432957860827446, "time": 0.6332912921905518, "epoch": 7, "memory": 44140, "step": 28352} +{"lr": 2.9091985726639714e-05, "data_time": 0.0021311283111572266, "loss": 0.45699959993362427, "time": 0.6321824312210083, "epoch": 7, "memory": 44140, "step": 28452} +{"lr": 2.881027219850035e-05, "data_time": 0.0017208576202392579, "loss": 0.4274042367935181, "time": 0.6325274467468261, "epoch": 7, "memory": 44140, "step": 28552} +{"accuracy/top1": 82.50831604003906, "data_time": 0.017026901245117188, "time": 1.2840092658996582, "step": 7} +{"lr": 2.827372366659638e-05, "data_time": 0.001958155632019043, "loss": 0.45466240048408507, "time": 0.633280897140503, "epoch": 8, "memory": 44140, "step": 28744} +{"lr": 2.7996563897301616e-05, "data_time": 0.0014590978622436523, "loss": 0.45339798033237455, "time": 0.6333338260650635, "epoch": 8, "memory": 44140, "step": 28844} +{"lr": 2.7720995770013028e-05, "data_time": 0.0019378900527954102, "loss": 0.4233066976070404, "time": 0.6471317291259766, "epoch": 8, "memory": 44140, "step": 28944} +{"lr": 2.744703552732299e-05, "data_time": 0.00213165283203125, "loss": 0.43006622791290283, "time": 0.6344423055648803, "epoch": 8, "memory": 44140, "step": 29044} +{"lr": 2.717469931705171e-05, "data_time": 0.0018494606018066406, "loss": 0.4387540936470032, "time": 0.6328137636184692, "epoch": 8, "memory": 44140, "step": 29144} +{"lr": 2.690400319129529e-05, "data_time": 0.0017334699630737304, "loss": 0.435209196805954, "time": 0.6318532466888428, "epoch": 8, "memory": 44140, "step": 29244} +{"lr": 2.6634963105479716e-05, "data_time": 0.0015765666961669923, "loss": 0.39980089366436006, "time": 0.6606237649917602, "epoch": 8, "memory": 44140, "step": 29344} +{"lr": 2.6367594917420177e-05, "data_time": 0.001616835594177246, "loss": 0.4316032290458679, "time": 0.6343820810317993, "epoch": 8, "memory": 44140, "step": 29444} +{"lr": 2.610191438638673e-05, "data_time": 0.0014084100723266602, "loss": 0.4166017174720764, "time": 0.6355518579483033, "epoch": 8, "memory": 44140, "step": 29544} +{"lr": 2.5837937172175032e-05, "data_time": 0.001395273208618164, "loss": 0.4450955927371979, "time": 0.6323782444000244, "epoch": 8, "memory": 44140, "step": 29644} +{"lr": 2.557567883418364e-05, "data_time": 0.001414942741394043, "loss": 0.4389266610145569, "time": 0.6329269170761108, "epoch": 8, "memory": 44140, "step": 29744} +{"lr": 2.531515483049656e-05, "data_time": 0.0015266656875610352, "loss": 0.4255665749311447, "time": 0.6326501131057739, "epoch": 8, "memory": 44140, "step": 29844} +{"lr": 2.5056380516972453e-05, "data_time": 0.001394343376159668, "loss": 0.40291866660118103, "time": 0.6322141408920288, "epoch": 8, "memory": 44140, "step": 29944} +{"lr": 2.4799371146339435e-05, "data_time": 0.0014451265335083008, "loss": 0.43276655673980713, "time": 0.6318119525909424, "epoch": 8, "memory": 44140, "step": 30044} +{"lr": 2.4544141867295954e-05, "data_time": 0.001522397994995117, "loss": 0.4334082961082458, "time": 0.6322718381881713, "epoch": 8, "memory": 44140, "step": 30144} +{"lr": 2.429070772361796e-05, "data_time": 0.0022794008255004883, "loss": 0.40298391580581666, "time": 0.6311321496963501, "epoch": 8, "memory": 44140, "step": 30244} +{"lr": 2.4039083653272075e-05, "data_time": 0.0014112472534179687, "loss": 0.4400285094976425, "time": 0.6329334735870361, "epoch": 8, "memory": 44140, "step": 30344} +{"lr": 2.378928448753535e-05, "data_time": 0.0015388727188110352, "loss": 0.4242362916469574, "time": 0.6322114706039429, "epoch": 8, "memory": 44140, "step": 30444} +{"lr": 2.354132495012082e-05, "data_time": 0.001468515396118164, "loss": 0.43261079490184784, "time": 0.6333138227462769, "epoch": 8, "memory": 44140, "step": 30544} +{"lr": 2.329521965630983e-05, "data_time": 0.0015710353851318359, "loss": 0.40571180880069735, "time": 0.6320971965789794, "epoch": 8, "memory": 44140, "step": 30644} +{"lr": 2.3050983112090472e-05, "data_time": 0.0015341281890869141, "loss": 0.4292501866817474, "time": 0.6324087619781494, "epoch": 8, "memory": 44140, "step": 30744} +{"lr": 2.2808629713302708e-05, "data_time": 0.0014192819595336913, "loss": 0.42273353040218353, "time": 0.6319505214691162, "epoch": 8, "memory": 44140, "step": 30844} +{"lr": 2.2568173744789706e-05, "data_time": 0.0016644716262817383, "loss": 0.4291909635066986, "time": 0.632189679145813, "epoch": 8, "memory": 44140, "step": 30944} +{"lr": 2.2329629379555897e-05, "data_time": 0.0020752191543579102, "loss": 0.4299454927444458, "time": 0.6315293312072754, "epoch": 8, "memory": 44140, "step": 31044} +{"lr": 2.2093010677931667e-05, "data_time": 0.0018827438354492188, "loss": 0.4226302891969681, "time": 0.6322283983230591, "epoch": 8, "memory": 44140, "step": 31144} +{"lr": 2.1858331586744434e-05, "data_time": 0.0014776468276977539, "loss": 0.4532845765352249, "time": 0.6333807468414306, "epoch": 8, "memory": 44140, "step": 31244} +{"lr": 2.1625605938496814e-05, "data_time": 0.0016274452209472656, "loss": 0.41041659116744994, "time": 0.6317580938339233, "epoch": 8, "memory": 44140, "step": 31344} +{"lr": 2.1394847450551186e-05, "data_time": 0.001647639274597168, "loss": 0.42769654393196105, "time": 0.6329609870910644, "epoch": 8, "memory": 44140, "step": 31444} +{"lr": 2.1166069724321078e-05, "data_time": 0.0032933235168457033, "loss": 0.46599029898643496, "time": 0.6349485635757446, "epoch": 8, "memory": 44140, "step": 31544} +{"lr": 2.0939286244469643e-05, "data_time": 0.001453709602355957, "loss": 0.4146469414234161, "time": 0.6327294111251831, "epoch": 8, "memory": 44140, "step": 31644} +{"lr": 2.071451037811473e-05, "data_time": 0.001534104347229004, "loss": 0.42976507544517517, "time": 0.6330198764801025, "epoch": 8, "memory": 44140, "step": 31744} +{"lr": 2.0491755374041045e-05, "data_time": 0.0015379428863525391, "loss": 0.42381633520126344, "time": 0.6326041936874389, "epoch": 8, "memory": 44140, "step": 31844} +{"lr": 2.0271034361919193e-05, "data_time": 0.001643538475036621, "loss": 0.4398748457431793, "time": 0.6331932544708252, "epoch": 8, "memory": 44140, "step": 31944} +{"lr": 2.0052360351531814e-05, "data_time": 0.0018382549285888671, "loss": 0.42296539545059203, "time": 0.6324436664581299, "epoch": 8, "memory": 44140, "step": 32044} +{"lr": 1.983574623200682e-05, "data_time": 0.0017765998840332032, "loss": 0.41598489284515383, "time": 0.6348613739013672, "epoch": 8, "memory": 44140, "step": 32144} +{"lr": 1.9621204771057533e-05, "data_time": 0.0017125606536865234, "loss": 0.4029065787792206, "time": 0.6338467597961426, "epoch": 8, "memory": 44140, "step": 32244} +{"lr": 1.940874861423025e-05, "data_time": 0.0015275001525878907, "loss": 0.4237914115190506, "time": 0.6326350212097168, "epoch": 8, "memory": 44140, "step": 32344} +{"lr": 1.9198390284158783e-05, "data_time": 0.0021889925003051756, "loss": 0.4153091847896576, "time": 0.6321191310882568, "epoch": 8, "memory": 44140, "step": 32444} +{"lr": 1.8990142179826573e-05, "data_time": 0.0014643430709838866, "loss": 0.44834463596343993, "time": 0.63248131275177, "epoch": 8, "memory": 44140, "step": 32544} +{"lr": 1.8784016575835568e-05, "data_time": 0.0021977901458740236, "loss": 0.4292978823184967, "time": 0.6317805051803589, "epoch": 8, "memory": 44140, "step": 32644} +{"accuracy/top1": 83.38215637207031, "data_time": 0.017577195167541505, "time": 1.2848993082841238, "step": 8} +{"lr": 1.8394249604901642e-05, "data_time": 0.0015108585357666016, "loss": 0.4192079246044159, "time": 0.6330900430679322, "epoch": 9, "memory": 44140, "step": 32836} +{"lr": 1.81943907744488e-05, "data_time": 0.0017599344253540039, "loss": 0.4158934384584427, "time": 0.6336795330047608, "epoch": 9, "memory": 44140, "step": 32936} +{"lr": 1.7996701347677866e-05, "data_time": 0.0014974117279052735, "loss": 0.44777504801750184, "time": 0.632328987121582, "epoch": 9, "memory": 44140, "step": 33036} +{"lr": 1.780119297683906e-05, "data_time": 0.0015994787216186523, "loss": 0.42130537927150724, "time": 0.6327724933624268, "epoch": 9, "memory": 44140, "step": 33136} +{"lr": 1.7607877185626483e-05, "data_time": 0.0022485971450805662, "loss": 0.443629652261734, "time": 0.6324627161026001, "epoch": 9, "memory": 44140, "step": 33236} +{"lr": 1.7416765368498675e-05, "data_time": 0.00175323486328125, "loss": 0.417745640873909, "time": 0.6315961599349975, "epoch": 9, "memory": 44140, "step": 33336} +{"lr": 1.7227868790007105e-05, "data_time": 0.001706242561340332, "loss": 0.4439123958349228, "time": 0.6318429708480835, "epoch": 9, "memory": 44140, "step": 33436} +{"lr": 1.7041198584132233e-05, "data_time": 0.00156707763671875, "loss": 0.4293932497501373, "time": 0.6323116064071655, "epoch": 9, "memory": 44140, "step": 33536} +{"lr": 1.6856765753627146e-05, "data_time": 0.0020839691162109373, "loss": 0.4280243694782257, "time": 0.6331812143325806, "epoch": 9, "memory": 44140, "step": 33636} +{"lr": 1.6674581169369163e-05, "data_time": 0.0024164915084838867, "loss": 0.4526894450187683, "time": 0.6328007936477661, "epoch": 9, "memory": 44140, "step": 33736} +{"lr": 1.6494655569718982e-05, "data_time": 0.0015497684478759765, "loss": 0.41594940423965454, "time": 0.6424659252166748, "epoch": 9, "memory": 44140, "step": 33836} +{"lr": 1.6316999559887786e-05, "data_time": 0.0017044544219970703, "loss": 0.40958764851093293, "time": 0.6325393199920655, "epoch": 9, "memory": 44140, "step": 33936} +{"lr": 1.6141623611312105e-05, "data_time": 0.0026323080062866213, "loss": 0.4171371668577194, "time": 0.6314513206481933, "epoch": 9, "memory": 44140, "step": 34036} +{"lr": 1.5968538061036705e-05, "data_time": 0.001636838912963867, "loss": 0.428592312335968, "time": 0.6314349889755249, "epoch": 9, "memory": 44140, "step": 34136} +{"lr": 1.5797753111105157e-05, "data_time": 0.0015050888061523438, "loss": 0.4413925111293793, "time": 0.6323025465011597, "epoch": 9, "memory": 44140, "step": 34236} +{"lr": 1.5629278827958574e-05, "data_time": 0.0017032384872436523, "loss": 0.4407835900783539, "time": 0.6320976972579956, "epoch": 9, "memory": 44140, "step": 34336} +{"lr": 1.5463125141842392e-05, "data_time": 0.0015386581420898438, "loss": 0.4242474019527435, "time": 0.6320900201797486, "epoch": 9, "memory": 44140, "step": 34436} +{"lr": 1.5299301846220796e-05, "data_time": 0.0017403841018676757, "loss": 0.4342947691679001, "time": 0.631589412689209, "epoch": 9, "memory": 44140, "step": 34536} +{"lr": 1.5137818597199751e-05, "data_time": 0.0016096830368041992, "loss": 0.46631333529949187, "time": 0.6328023910522461, "epoch": 9, "memory": 44140, "step": 34636} +{"lr": 1.4978684912957622e-05, "data_time": 0.001450657844543457, "loss": 0.4332442104816437, "time": 0.6314464569091797, "epoch": 9, "memory": 44140, "step": 34736} +{"lr": 1.4821910173184321e-05, "data_time": 0.002198624610900879, "loss": 0.4289512515068054, "time": 0.6325736999511719, "epoch": 9, "memory": 44140, "step": 34836} +{"lr": 1.4667503618528337e-05, "data_time": 0.0017051935195922852, "loss": 0.44457480907440183, "time": 0.6312191724777222, "epoch": 9, "memory": 44140, "step": 34936} +{"lr": 1.4515474350052097e-05, "data_time": 0.0014960050582885742, "loss": 0.39010309278964994, "time": 0.6336281061172485, "epoch": 9, "memory": 44140, "step": 35036} +{"lr": 1.4365831328695616e-05, "data_time": 0.001566457748413086, "loss": 0.42951966226100924, "time": 0.6338382005691529, "epoch": 9, "memory": 44140, "step": 35136} +{"lr": 1.4218583374748169e-05, "data_time": 0.001975345611572266, "loss": 0.40147810578346255, "time": 0.633592939376831, "epoch": 9, "memory": 44140, "step": 35236} +{"lr": 1.4073739167328542e-05, "data_time": 0.0015869617462158203, "loss": 0.4517203629016876, "time": 0.6345815420150757, "epoch": 9, "memory": 44140, "step": 35336} +{"lr": 1.3931307243873344e-05, "data_time": 0.001720762252807617, "loss": 0.44844684302806853, "time": 0.6338479042053222, "epoch": 9, "memory": 44140, "step": 35436} +{"lr": 1.3791295999633911e-05, "data_time": 0.0016016960144042969, "loss": 0.4113050580024719, "time": 0.6337888956069946, "epoch": 9, "memory": 44140, "step": 35536} +{"lr": 1.3653713687181394e-05, "data_time": 0.002195572853088379, "loss": 0.4379051148891449, "time": 0.6354958295822144, "epoch": 9, "memory": 44140, "step": 35636} +{"lr": 1.3518568415920337e-05, "data_time": 0.0015153884887695312, "loss": 0.4328551203012466, "time": 0.6332636594772338, "epoch": 9, "memory": 44140, "step": 35736} +{"lr": 1.3385868151610723e-05, "data_time": 0.0019635915756225585, "loss": 0.40239324867725373, "time": 0.6343563318252563, "epoch": 9, "memory": 44140, "step": 35836} +{"lr": 1.3255620715898442e-05, "data_time": 0.0018749475479125977, "loss": 0.4242888927459717, "time": 0.6334626913070679, "epoch": 9, "memory": 44140, "step": 35936} +{"lr": 1.3127833785854261e-05, "data_time": 0.0017363548278808594, "loss": 0.4422772765159607, "time": 0.6335497379302979, "epoch": 9, "memory": 44140, "step": 36036} +{"lr": 1.3002514893521313e-05, "data_time": 0.0018674850463867188, "loss": 0.4114193767309189, "time": 0.6336813688278198, "epoch": 9, "memory": 44140, "step": 36136} +{"lr": 1.287967142547117e-05, "data_time": 0.00149383544921875, "loss": 0.4362452030181885, "time": 0.636810564994812, "epoch": 9, "memory": 44140, "step": 36236} +{"lr": 1.2759310622368435e-05, "data_time": 0.0017003774642944335, "loss": 0.44429692029953005, "time": 0.6355048418045044, "epoch": 9, "memory": 44140, "step": 36336} +{"lr": 1.2641439578543968e-05, "data_time": 0.0015533685684204102, "loss": 0.43545244038105013, "time": 0.6331853151321412, "epoch": 9, "memory": 44140, "step": 36436} +{"lr": 1.252606524157674e-05, "data_time": 0.001909637451171875, "loss": 0.44942472577095033, "time": 0.6355678081512451, "epoch": 9, "memory": 44140, "step": 36536} +{"lr": 1.2413194411884297e-05, "data_time": 0.001904773712158203, "loss": 0.41389899551868437, "time": 0.6345012426376343, "epoch": 9, "memory": 44140, "step": 36636} +{"lr": 1.2302833742321981e-05, "data_time": 0.001581096649169922, "loss": 0.4211801588535309, "time": 0.6346514225006104, "epoch": 9, "memory": 44140, "step": 36736} +{"accuracy/top1": 84.1517333984375, "data_time": 0.018436970313390096, "time": 1.28598349293073, "step": 9} +{"lr": 1.2098001439139718e-05, "data_time": 0.001784038543701172, "loss": 0.4378559350967407, "time": 0.6330617427825928, "epoch": 10, "memory": 44140, "step": 36928} +{"lr": 1.1995007122352056e-05, "data_time": 0.0016316652297973632, "loss": 0.4197396695613861, "time": 0.6329480171203613, "epoch": 10, "memory": 44140, "step": 37028} +{"lr": 1.1894547614574522e-05, "data_time": 0.0018131494522094726, "loss": 0.45584308803081514, "time": 0.6335228681564331, "epoch": 10, "memory": 44140, "step": 37128} +{"lr": 1.1796628837111806e-05, "data_time": 0.0020069122314453126, "loss": 0.4205508291721344, "time": 0.6323853969573975, "epoch": 10, "memory": 44140, "step": 37228} +{"lr": 1.1701256561512343e-05, "data_time": 0.001579594612121582, "loss": 0.4137354552745819, "time": 0.6353223085403442, "epoch": 10, "memory": 44140, "step": 37328} +{"lr": 1.1608436409228136e-05, "data_time": 0.0017122507095336914, "loss": 0.40455450415611266, "time": 0.6342119455337525, "epoch": 10, "memory": 44140, "step": 37428} +{"lr": 1.1518173851283401e-05, "data_time": 0.0020758390426635744, "loss": 0.43558679819107055, "time": 0.6337926149368286, "epoch": 10, "memory": 44140, "step": 37528} +{"lr": 1.1430474207952176e-05, "data_time": 0.0021812915802001953, "loss": 0.4532723903656006, "time": 0.6319443225860596, "epoch": 10, "memory": 44140, "step": 37628} +{"lr": 1.1345342648444593e-05, "data_time": 0.002339744567871094, "loss": 0.4262354731559753, "time": 0.6314018487930297, "epoch": 10, "memory": 44140, "step": 37728} +{"lr": 1.12627841906023e-05, "data_time": 0.0017124414443969727, "loss": 0.40543001890182495, "time": 0.6451123476028442, "epoch": 10, "memory": 44140, "step": 37828} +{"lr": 1.1182803700602659e-05, "data_time": 0.0025897502899169924, "loss": 0.4339084684848785, "time": 0.6339578628540039, "epoch": 10, "memory": 44140, "step": 37928} +{"lr": 1.110540589267192e-05, "data_time": 0.00283510684967041, "loss": 0.43444119691848754, "time": 0.6329931020736694, "epoch": 10, "memory": 44140, "step": 38028} +{"lr": 1.1030595328807396e-05, "data_time": 0.0016083955764770509, "loss": 0.4390761315822601, "time": 0.6333801984786988, "epoch": 10, "memory": 44140, "step": 38128} +{"lr": 1.0958376418508494e-05, "data_time": 0.0016560077667236329, "loss": 0.4350801944732666, "time": 0.6321488142013549, "epoch": 10, "memory": 44140, "step": 38228} +{"lr": 1.088875341851686e-05, "data_time": 0.0015148401260375976, "loss": 0.43409552574157717, "time": 0.6336699008941651, "epoch": 10, "memory": 44140, "step": 38328} +{"lr": 1.0821730432565486e-05, "data_time": 0.0030703067779541014, "loss": 0.44336821138858795, "time": 0.6336770296096802, "epoch": 10, "memory": 44140, "step": 38428} +{"lr": 1.0757311411136797e-05, "data_time": 0.002206110954284668, "loss": 0.4388790428638458, "time": 0.6329855680465698, "epoch": 10, "memory": 44140, "step": 38528} +{"lr": 1.0695500151229833e-05, "data_time": 0.0025473833084106445, "loss": 0.4290278673171997, "time": 0.6370671987533569, "epoch": 10, "memory": 44140, "step": 38628} +{"lr": 1.0636300296136373e-05, "data_time": 0.0017495155334472656, "loss": 0.435150146484375, "time": 0.6349704265594482, "epoch": 10, "memory": 44140, "step": 38728} +{"lr": 1.0579715335226322e-05, "data_time": 0.0016252517700195313, "loss": 0.40780472159385683, "time": 0.6348392009735108, "epoch": 10, "memory": 44140, "step": 38828} +{"lr": 1.0525748603741896e-05, "data_time": 0.0023051977157592775, "loss": 0.4292386621236801, "time": 0.6330715179443359, "epoch": 10, "memory": 44140, "step": 38928} +{"lr": 1.0474403282601119e-05, "data_time": 0.0014242410659790039, "loss": 0.424554306268692, "time": 0.6333153724670411, "epoch": 10, "memory": 44140, "step": 39028} +{"lr": 1.0425682398210336e-05, "data_time": 0.00188751220703125, "loss": 0.4188223987817764, "time": 0.6339438199996948, "epoch": 10, "memory": 44140, "step": 39128} +{"lr": 1.0379588822285784e-05, "data_time": 0.0021077632904052735, "loss": 0.4077302306890488, "time": 0.6327978134155273, "epoch": 10, "memory": 44140, "step": 39228} +{"lr": 1.0336125271684364e-05, "data_time": 0.0017225503921508788, "loss": 0.41424120366573336, "time": 0.6335647106170654, "epoch": 10, "memory": 44140, "step": 39328} +{"lr": 1.0295294308243494e-05, "data_time": 0.0023066520690917967, "loss": 0.44279840886592864, "time": 0.6315334796905517, "epoch": 10, "memory": 44140, "step": 39428} +{"lr": 1.0257098338630086e-05, "data_time": 0.00244288444519043, "loss": 0.4393835723400116, "time": 0.638019061088562, "epoch": 10, "memory": 44140, "step": 39528} +{"lr": 1.0221539614198729e-05, "data_time": 0.002653050422668457, "loss": 0.43867662250995637, "time": 0.6328872680664063, "epoch": 10, "memory": 44140, "step": 39628} +{"lr": 1.0188620230858978e-05, "data_time": 0.002184557914733887, "loss": 0.43402776420116423, "time": 0.6327784299850464, "epoch": 10, "memory": 44140, "step": 39728} +{"lr": 1.01583421289518e-05, "data_time": 0.0015357255935668946, "loss": 0.4060773134231567, "time": 0.6347954988479614, "epoch": 10, "memory": 44140, "step": 39828} +{"lr": 1.0130707093135206e-05, "data_time": 0.0015917539596557618, "loss": 0.4351502925157547, "time": 0.6341427564620972, "epoch": 10, "memory": 44140, "step": 39928} +{"lr": 1.01057167522791e-05, "data_time": 0.0015056371688842774, "loss": 0.4138208985328674, "time": 0.6322539329528809, "epoch": 10, "memory": 44140, "step": 40028} +{"lr": 1.008337257936921e-05, "data_time": 0.002364039421081543, "loss": 0.41611374318599703, "time": 0.6352760791778564, "epoch": 10, "memory": 44140, "step": 40128} +{"lr": 1.0063675891420294e-05, "data_time": 0.001945805549621582, "loss": 0.4419875293970108, "time": 0.6347781658172608, "epoch": 10, "memory": 44140, "step": 40228} +{"lr": 1.0046627849398548e-05, "data_time": 0.0018296480178833009, "loss": 0.4454681098461151, "time": 0.6340731620788574, "epoch": 10, "memory": 44140, "step": 40328} +{"lr": 1.0032229458153106e-05, "data_time": 0.0016800642013549804, "loss": 0.43427015244960787, "time": 0.6339587450027466, "epoch": 10, "memory": 44140, "step": 40428} +{"lr": 1.0020481566356862e-05, "data_time": 0.001519322395324707, "loss": 0.4427505224943161, "time": 0.6341932058334351, "epoch": 10, "memory": 44140, "step": 40528} +{"lr": 1.0011384866456456e-05, "data_time": 0.0018405437469482422, "loss": 0.435544490814209, "time": 0.633528757095337, "epoch": 10, "memory": 44140, "step": 40628} +{"lr": 1.00049398946314e-05, "data_time": 0.001684713363647461, "loss": 0.4310219496488571, "time": 0.6344710111618042, "epoch": 10, "memory": 44140, "step": 40728} +{"lr": 1.0001147030762536e-05, "data_time": 0.0014252185821533204, "loss": 0.4472906470298767, "time": 0.634446668624878, "epoch": 10, "memory": 44140, "step": 40828} +{"accuracy/top1": 84.35033416748047, "data_time": 0.017048033078511556, "time": 1.2856839835643767, "step": 10} diff --git a/clip_large_pretrain_4x256_all2_lr1e-4/20230606_005614/vis_data/vis_image/0000000.png_1.png b/clip_large_pretrain_4x256_all2_lr1e-4/20230606_005614/vis_data/vis_image/0000000.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..32b1ea23b5e1eace49b630945ff68199a6951ee9 Binary files /dev/null and b/clip_large_pretrain_4x256_all2_lr1e-4/20230606_005614/vis_data/vis_image/0000000.png_1.png differ diff --git a/clip_large_pretrain_4x256_all2_lr1e-4/20230606_005614/vis_data/vis_image/0000000.png_10.png b/clip_large_pretrain_4x256_all2_lr1e-4/20230606_005614/vis_data/vis_image/0000000.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..d413e549fb546504b32d48cff2a814a7565d2b59 Binary files /dev/null and b/clip_large_pretrain_4x256_all2_lr1e-4/20230606_005614/vis_data/vis_image/0000000.png_10.png differ diff --git a/clip_large_pretrain_4x256_all2_lr1e-4/20230606_005614/vis_data/vis_image/0000000.png_2.png b/clip_large_pretrain_4x256_all2_lr1e-4/20230606_005614/vis_data/vis_image/0000000.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..e7e893218e46c628bf58c8e5091e1ef64b5ee1fc Binary files /dev/null and b/clip_large_pretrain_4x256_all2_lr1e-4/20230606_005614/vis_data/vis_image/0000000.png_2.png differ diff --git a/clip_large_pretrain_4x256_all2_lr1e-4/20230606_005614/vis_data/vis_image/0000000.png_3.png b/clip_large_pretrain_4x256_all2_lr1e-4/20230606_005614/vis_data/vis_image/0000000.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..c268e6e387a2d55784d265426ad9f0e3fb514f47 Binary files /dev/null and b/clip_large_pretrain_4x256_all2_lr1e-4/20230606_005614/vis_data/vis_image/0000000.png_3.png differ diff --git a/clip_large_pretrain_4x256_all2_lr1e-4/20230606_005614/vis_data/vis_image/0000000.png_4.png b/clip_large_pretrain_4x256_all2_lr1e-4/20230606_005614/vis_data/vis_image/0000000.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..fc4cba570ffebc1c3b68843dba9c0daa568b2df8 Binary files /dev/null and b/clip_large_pretrain_4x256_all2_lr1e-4/20230606_005614/vis_data/vis_image/0000000.png_4.png differ diff --git a/clip_large_pretrain_4x256_all2_lr1e-4/20230606_005614/vis_data/vis_image/0000000.png_5.png b/clip_large_pretrain_4x256_all2_lr1e-4/20230606_005614/vis_data/vis_image/0000000.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..fae35240ca2bee6a88287fca0b0c5b02a56fc9b2 Binary files /dev/null and b/clip_large_pretrain_4x256_all2_lr1e-4/20230606_005614/vis_data/vis_image/0000000.png_5.png differ diff --git a/clip_large_pretrain_4x256_all2_lr1e-4/20230606_005614/vis_data/vis_image/0000000.png_6.png b/clip_large_pretrain_4x256_all2_lr1e-4/20230606_005614/vis_data/vis_image/0000000.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..8f2b4708b7bf5687c8f2f136e63a84f44a70533a Binary files /dev/null and b/clip_large_pretrain_4x256_all2_lr1e-4/20230606_005614/vis_data/vis_image/0000000.png_6.png differ diff --git a/clip_large_pretrain_4x256_all2_lr1e-4/20230606_005614/vis_data/vis_image/0000000.png_7.png b/clip_large_pretrain_4x256_all2_lr1e-4/20230606_005614/vis_data/vis_image/0000000.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..34637dc1ecfd7ff7e0e98c53d93be34f6b04e3df Binary files /dev/null and b/clip_large_pretrain_4x256_all2_lr1e-4/20230606_005614/vis_data/vis_image/0000000.png_7.png differ diff --git a/clip_large_pretrain_4x256_all2_lr1e-4/20230606_005614/vis_data/vis_image/0000000.png_8.png b/clip_large_pretrain_4x256_all2_lr1e-4/20230606_005614/vis_data/vis_image/0000000.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..2a6bed02e7e2546a27ed6d3792893e1ebfdaa597 Binary files /dev/null and b/clip_large_pretrain_4x256_all2_lr1e-4/20230606_005614/vis_data/vis_image/0000000.png_8.png differ diff --git a/clip_large_pretrain_4x256_all2_lr1e-4/20230606_005614/vis_data/vis_image/0000000.png_9.png b/clip_large_pretrain_4x256_all2_lr1e-4/20230606_005614/vis_data/vis_image/0000000.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..aa776e51360882fdb52a7358637dd37191768f2a Binary files /dev/null and b/clip_large_pretrain_4x256_all2_lr1e-4/20230606_005614/vis_data/vis_image/0000000.png_9.png differ diff --git a/clip_large_pretrain_4x256_all2_lr1e-4/20230606_005614/vis_data/vis_image/000015337.jpg_1.png b/clip_large_pretrain_4x256_all2_lr1e-4/20230606_005614/vis_data/vis_image/000015337.jpg_1.png new file mode 100644 index 0000000000000000000000000000000000000000..57a9efe097807d59c0da6502c53ba4722fe7809e Binary files /dev/null and b/clip_large_pretrain_4x256_all2_lr1e-4/20230606_005614/vis_data/vis_image/000015337.jpg_1.png differ diff --git a/clip_large_pretrain_4x256_all2_lr1e-4/20230606_005614/vis_data/vis_image/000015337.jpg_10.png b/clip_large_pretrain_4x256_all2_lr1e-4/20230606_005614/vis_data/vis_image/000015337.jpg_10.png new file mode 100644 index 0000000000000000000000000000000000000000..a948857a4c9496313dcce915631a69b379ea885d Binary files /dev/null and b/clip_large_pretrain_4x256_all2_lr1e-4/20230606_005614/vis_data/vis_image/000015337.jpg_10.png differ diff --git a/clip_large_pretrain_4x256_all2_lr1e-4/20230606_005614/vis_data/vis_image/000015337.jpg_2.png b/clip_large_pretrain_4x256_all2_lr1e-4/20230606_005614/vis_data/vis_image/000015337.jpg_2.png new file mode 100644 index 0000000000000000000000000000000000000000..23c5d777aefcaa6f3f3e515bb7f3b8947f68cc51 Binary files /dev/null and b/clip_large_pretrain_4x256_all2_lr1e-4/20230606_005614/vis_data/vis_image/000015337.jpg_2.png differ diff --git a/clip_large_pretrain_4x256_all2_lr1e-4/20230606_005614/vis_data/vis_image/000015337.jpg_3.png b/clip_large_pretrain_4x256_all2_lr1e-4/20230606_005614/vis_data/vis_image/000015337.jpg_3.png new file mode 100644 index 0000000000000000000000000000000000000000..1539907617ee296ee6f626f1fe826467eb7b3d93 Binary files /dev/null and b/clip_large_pretrain_4x256_all2_lr1e-4/20230606_005614/vis_data/vis_image/000015337.jpg_3.png differ diff --git a/clip_large_pretrain_4x256_all2_lr1e-4/20230606_005614/vis_data/vis_image/000015337.jpg_4.png b/clip_large_pretrain_4x256_all2_lr1e-4/20230606_005614/vis_data/vis_image/000015337.jpg_4.png new file mode 100644 index 0000000000000000000000000000000000000000..bbb02d68f3c58bf8b50f9b2a52d15047eff2abc8 Binary files /dev/null and b/clip_large_pretrain_4x256_all2_lr1e-4/20230606_005614/vis_data/vis_image/000015337.jpg_4.png differ diff --git a/clip_large_pretrain_4x256_all2_lr1e-4/20230606_005614/vis_data/vis_image/000015337.jpg_5.png b/clip_large_pretrain_4x256_all2_lr1e-4/20230606_005614/vis_data/vis_image/000015337.jpg_5.png new file mode 100644 index 0000000000000000000000000000000000000000..2d7e689c789092109949774db12e664705fe32f8 Binary files /dev/null and b/clip_large_pretrain_4x256_all2_lr1e-4/20230606_005614/vis_data/vis_image/000015337.jpg_5.png differ diff --git a/clip_large_pretrain_4x256_all2_lr1e-4/20230606_005614/vis_data/vis_image/000015337.jpg_6.png b/clip_large_pretrain_4x256_all2_lr1e-4/20230606_005614/vis_data/vis_image/000015337.jpg_6.png new file mode 100644 index 0000000000000000000000000000000000000000..f5f3d7b0408346d2021db4e3799442499a2793b4 Binary files /dev/null and b/clip_large_pretrain_4x256_all2_lr1e-4/20230606_005614/vis_data/vis_image/000015337.jpg_6.png differ diff --git a/clip_large_pretrain_4x256_all2_lr1e-4/20230606_005614/vis_data/vis_image/000015337.jpg_7.png b/clip_large_pretrain_4x256_all2_lr1e-4/20230606_005614/vis_data/vis_image/000015337.jpg_7.png new file mode 100644 index 0000000000000000000000000000000000000000..a4685dde2a7b809930a1a2bcc9e1f2eb0f60ec8a Binary files /dev/null and b/clip_large_pretrain_4x256_all2_lr1e-4/20230606_005614/vis_data/vis_image/000015337.jpg_7.png differ diff --git a/clip_large_pretrain_4x256_all2_lr1e-4/20230606_005614/vis_data/vis_image/000015337.jpg_8.png b/clip_large_pretrain_4x256_all2_lr1e-4/20230606_005614/vis_data/vis_image/000015337.jpg_8.png new file mode 100644 index 0000000000000000000000000000000000000000..c3b86afe13bb9812e0bf79ca55c2fc7e7171d9ee Binary files /dev/null and b/clip_large_pretrain_4x256_all2_lr1e-4/20230606_005614/vis_data/vis_image/000015337.jpg_8.png differ diff --git a/clip_large_pretrain_4x256_all2_lr1e-4/20230606_005614/vis_data/vis_image/000015337.jpg_9.png b/clip_large_pretrain_4x256_all2_lr1e-4/20230606_005614/vis_data/vis_image/000015337.jpg_9.png new file mode 100644 index 0000000000000000000000000000000000000000..ce29f299193b67db0dd4b0363bb78414a5ef9012 Binary files /dev/null and b/clip_large_pretrain_4x256_all2_lr1e-4/20230606_005614/vis_data/vis_image/000015337.jpg_9.png differ diff --git a/clip_large_pretrain_4x256_all2_lr1e-4/20230606_005614/vis_data/vis_image/0004159.png_1.png b/clip_large_pretrain_4x256_all2_lr1e-4/20230606_005614/vis_data/vis_image/0004159.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..02975b881369c17cee099d755d066a295a0cd215 Binary files /dev/null and b/clip_large_pretrain_4x256_all2_lr1e-4/20230606_005614/vis_data/vis_image/0004159.png_1.png differ diff --git a/clip_large_pretrain_4x256_all2_lr1e-4/20230606_005614/vis_data/vis_image/0004159.png_10.png b/clip_large_pretrain_4x256_all2_lr1e-4/20230606_005614/vis_data/vis_image/0004159.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..9623e8a0448f87ecdd5899bfc7980a71932e1c0e Binary files /dev/null and b/clip_large_pretrain_4x256_all2_lr1e-4/20230606_005614/vis_data/vis_image/0004159.png_10.png differ diff --git a/clip_large_pretrain_4x256_all2_lr1e-4/20230606_005614/vis_data/vis_image/0004159.png_2.png b/clip_large_pretrain_4x256_all2_lr1e-4/20230606_005614/vis_data/vis_image/0004159.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..a93c285142e923584c386c475e73b28f4bf7337f Binary files /dev/null and b/clip_large_pretrain_4x256_all2_lr1e-4/20230606_005614/vis_data/vis_image/0004159.png_2.png differ diff --git a/clip_large_pretrain_4x256_all2_lr1e-4/20230606_005614/vis_data/vis_image/0004159.png_3.png b/clip_large_pretrain_4x256_all2_lr1e-4/20230606_005614/vis_data/vis_image/0004159.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..fcc4e2e054c86872fe940016ede97cb8d18c5efc Binary files /dev/null and b/clip_large_pretrain_4x256_all2_lr1e-4/20230606_005614/vis_data/vis_image/0004159.png_3.png differ diff --git a/clip_large_pretrain_4x256_all2_lr1e-4/20230606_005614/vis_data/vis_image/0004159.png_4.png b/clip_large_pretrain_4x256_all2_lr1e-4/20230606_005614/vis_data/vis_image/0004159.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..4ea296bb8f609d2a099bdc365a688317c4b5360a Binary files /dev/null and b/clip_large_pretrain_4x256_all2_lr1e-4/20230606_005614/vis_data/vis_image/0004159.png_4.png differ diff --git a/clip_large_pretrain_4x256_all2_lr1e-4/20230606_005614/vis_data/vis_image/0004159.png_5.png b/clip_large_pretrain_4x256_all2_lr1e-4/20230606_005614/vis_data/vis_image/0004159.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..86c0b20d8ad96d7cffdcc08f9f1e7dd04ea1df83 Binary files /dev/null and b/clip_large_pretrain_4x256_all2_lr1e-4/20230606_005614/vis_data/vis_image/0004159.png_5.png differ diff --git a/clip_large_pretrain_4x256_all2_lr1e-4/20230606_005614/vis_data/vis_image/0004159.png_6.png b/clip_large_pretrain_4x256_all2_lr1e-4/20230606_005614/vis_data/vis_image/0004159.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..9f5b1338868aae669e1599af6c7eccb0d77d0790 Binary files /dev/null and b/clip_large_pretrain_4x256_all2_lr1e-4/20230606_005614/vis_data/vis_image/0004159.png_6.png differ diff --git a/clip_large_pretrain_4x256_all2_lr1e-4/20230606_005614/vis_data/vis_image/0004159.png_7.png b/clip_large_pretrain_4x256_all2_lr1e-4/20230606_005614/vis_data/vis_image/0004159.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..9bec368571cd6b599b193a1e377d6da400838c64 Binary files /dev/null and b/clip_large_pretrain_4x256_all2_lr1e-4/20230606_005614/vis_data/vis_image/0004159.png_7.png differ diff --git a/clip_large_pretrain_4x256_all2_lr1e-4/20230606_005614/vis_data/vis_image/0004159.png_8.png b/clip_large_pretrain_4x256_all2_lr1e-4/20230606_005614/vis_data/vis_image/0004159.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..d6cb0046fad9869361d1eeb144307d20fa6a4a13 Binary files /dev/null and b/clip_large_pretrain_4x256_all2_lr1e-4/20230606_005614/vis_data/vis_image/0004159.png_8.png differ diff --git a/clip_large_pretrain_4x256_all2_lr1e-4/20230606_005614/vis_data/vis_image/0004159.png_9.png b/clip_large_pretrain_4x256_all2_lr1e-4/20230606_005614/vis_data/vis_image/0004159.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..fa9f680e25d57cbc7be51b24595dadb780dd84e7 Binary files /dev/null and b/clip_large_pretrain_4x256_all2_lr1e-4/20230606_005614/vis_data/vis_image/0004159.png_9.png differ diff --git a/clip_large_pretrain_4x256_all2_lr1e-4/20230606_005614/vis_data/vis_image/0008318.png_1.png b/clip_large_pretrain_4x256_all2_lr1e-4/20230606_005614/vis_data/vis_image/0008318.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..c2e6f752738d71134fd716151d8894eb71540957 Binary files /dev/null and b/clip_large_pretrain_4x256_all2_lr1e-4/20230606_005614/vis_data/vis_image/0008318.png_1.png differ diff --git a/clip_large_pretrain_4x256_all2_lr1e-4/20230606_005614/vis_data/vis_image/0008318.png_10.png b/clip_large_pretrain_4x256_all2_lr1e-4/20230606_005614/vis_data/vis_image/0008318.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..bb49e4721c483bf7baa3d4b53dff4657e1534a2d Binary files /dev/null and b/clip_large_pretrain_4x256_all2_lr1e-4/20230606_005614/vis_data/vis_image/0008318.png_10.png differ diff --git a/clip_large_pretrain_4x256_all2_lr1e-4/20230606_005614/vis_data/vis_image/0008318.png_2.png b/clip_large_pretrain_4x256_all2_lr1e-4/20230606_005614/vis_data/vis_image/0008318.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..80454680ded9bb3706a1a6401a23995dfe7433e4 Binary files /dev/null and b/clip_large_pretrain_4x256_all2_lr1e-4/20230606_005614/vis_data/vis_image/0008318.png_2.png differ diff --git a/clip_large_pretrain_4x256_all2_lr1e-4/20230606_005614/vis_data/vis_image/0008318.png_3.png b/clip_large_pretrain_4x256_all2_lr1e-4/20230606_005614/vis_data/vis_image/0008318.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..e26d8cd5b956d1d4c552a3b14b1d7d3da4a9f72a Binary files /dev/null and b/clip_large_pretrain_4x256_all2_lr1e-4/20230606_005614/vis_data/vis_image/0008318.png_3.png differ diff --git a/clip_large_pretrain_4x256_all2_lr1e-4/20230606_005614/vis_data/vis_image/0008318.png_4.png b/clip_large_pretrain_4x256_all2_lr1e-4/20230606_005614/vis_data/vis_image/0008318.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..43074a35df3f6204c2f12746650273952d0ea9f5 Binary files /dev/null and b/clip_large_pretrain_4x256_all2_lr1e-4/20230606_005614/vis_data/vis_image/0008318.png_4.png differ diff --git a/clip_large_pretrain_4x256_all2_lr1e-4/20230606_005614/vis_data/vis_image/0008318.png_5.png b/clip_large_pretrain_4x256_all2_lr1e-4/20230606_005614/vis_data/vis_image/0008318.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..091e6faa8ea643ed76943e3955cc3c9780fb19ed Binary files /dev/null and b/clip_large_pretrain_4x256_all2_lr1e-4/20230606_005614/vis_data/vis_image/0008318.png_5.png differ diff --git a/clip_large_pretrain_4x256_all2_lr1e-4/20230606_005614/vis_data/vis_image/0008318.png_6.png b/clip_large_pretrain_4x256_all2_lr1e-4/20230606_005614/vis_data/vis_image/0008318.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..b81f3282d526a6c17020fa3c63f3dbb0601031a4 Binary files /dev/null and b/clip_large_pretrain_4x256_all2_lr1e-4/20230606_005614/vis_data/vis_image/0008318.png_6.png differ diff --git a/clip_large_pretrain_4x256_all2_lr1e-4/20230606_005614/vis_data/vis_image/0008318.png_7.png b/clip_large_pretrain_4x256_all2_lr1e-4/20230606_005614/vis_data/vis_image/0008318.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..29ffba7471ab8da642c060f05cc2d67d2e3a8f7e Binary files /dev/null and b/clip_large_pretrain_4x256_all2_lr1e-4/20230606_005614/vis_data/vis_image/0008318.png_7.png differ diff --git a/clip_large_pretrain_4x256_all2_lr1e-4/20230606_005614/vis_data/vis_image/0008318.png_8.png b/clip_large_pretrain_4x256_all2_lr1e-4/20230606_005614/vis_data/vis_image/0008318.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..a5cda8f816d65f8873d120b0775b8ee7370114da Binary files /dev/null and b/clip_large_pretrain_4x256_all2_lr1e-4/20230606_005614/vis_data/vis_image/0008318.png_8.png differ diff --git a/clip_large_pretrain_4x256_all2_lr1e-4/20230606_005614/vis_data/vis_image/0008318.png_9.png b/clip_large_pretrain_4x256_all2_lr1e-4/20230606_005614/vis_data/vis_image/0008318.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..baa28fe8abb94785e8822aa06b8bde64f19294a2 Binary files /dev/null and b/clip_large_pretrain_4x256_all2_lr1e-4/20230606_005614/vis_data/vis_image/0008318.png_9.png differ diff --git a/clip_large_pretrain_4x256_all2_lr1e-4/clip_large_pretrain_4x256_all2_lr1e-4.py b/clip_large_pretrain_4x256_all2_lr1e-4/clip_large_pretrain_4x256_all2_lr1e-4.py new file mode 100644 index 0000000000000000000000000000000000000000..c989f09acfe14651d1114b3ff4a90a7112361e54 --- /dev/null +++ b/clip_large_pretrain_4x256_all2_lr1e-4/clip_large_pretrain_4x256_all2_lr1e-4.py @@ -0,0 +1,341 @@ +optim_wrapper = dict( + optimizer=dict( + type='AdamW', lr=0.0001, weight_decay=0.3, _scope_='mmpretrain'), + paramwise_cfg=dict( + custom_keys=dict({ + '.cls_token': dict(decay_mult=0.0), + '.pos_embed': dict(decay_mult=0.0) + })), + type='AmpOptimWrapper', + dtype='bfloat16', + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=False, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=4096) +model = dict( + type='ImageClassifier', + backbone=dict( + frozen_stages=24, + type='VisionTransformer', + arch='l', + img_size=224, + patch_size=14, + drop_rate=0.1, + pre_norm=True, + final_norm=False, + init_cfg=dict( + type='Pretrained', + checkpoint='ckpt/openclip-ViT-L-14.pth', + prefix='backbone')), + neck=dict( + type='CLIPProjection', + in_channels=1024, + out_channels=768, + init_cfg=dict( + type='Pretrained', + checkpoint='ckpt/openclip-ViT-L-14.pth', + prefix='backbone')), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=768, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + init_cfg=None), + init_cfg=dict( + type='TruncNormal', layer=['Conv2d', 'Linear'], std=0.02, bias=0.0), + train_cfg=None) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=128, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/all_0.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/all_1.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3fake8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/cc1m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3real8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=128, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV2-1-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/if-dpmsolver++-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=128, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV2-1-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/if-dpmsolver++-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/clip_large_pretrain_4x256_all2_lr1e-4' diff --git a/clip_large_pretrain_4x256_all2_lr1e-4/epoch_1.pth b/clip_large_pretrain_4x256_all2_lr1e-4/epoch_1.pth new file mode 100644 index 0000000000000000000000000000000000000000..80b0cae1a89089b13976f9ed4ebb62f54d44e1b4 --- /dev/null +++ b/clip_large_pretrain_4x256_all2_lr1e-4/epoch_1.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:7c473e20d130e4dc0cfaa95813bc46bd2a45a65b9ca86764295d687abc410254 +size 2438637825 diff --git a/clip_large_pretrain_4x256_all2_lr1e-4/epoch_10.pth b/clip_large_pretrain_4x256_all2_lr1e-4/epoch_10.pth new file mode 100644 index 0000000000000000000000000000000000000000..039e9f93e233313e38199a08755edbef05b4743d --- /dev/null +++ b/clip_large_pretrain_4x256_all2_lr1e-4/epoch_10.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:d6357cdda1297cb51f656c93e1b1948a2ad692a0d4162721495bf6746d89e8b2 +size 2441342081 diff --git a/clip_large_pretrain_4x256_all2_lr1e-4/epoch_2.pth b/clip_large_pretrain_4x256_all2_lr1e-4/epoch_2.pth new file mode 100644 index 0000000000000000000000000000000000000000..302f2c66fc72f3f895701b52b0c716b956d724e4 --- /dev/null +++ b/clip_large_pretrain_4x256_all2_lr1e-4/epoch_2.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:9b767628a2b9bbcc6809d1def37ed216e095437e4da2c83eeb4450daccc72006 +size 2438937025 diff --git a/clip_large_pretrain_4x256_all2_lr1e-4/epoch_3.pth b/clip_large_pretrain_4x256_all2_lr1e-4/epoch_3.pth new file mode 100644 index 0000000000000000000000000000000000000000..0ed26a3e07cbaf07f7f62bbcb8bcdd7620d29755 --- /dev/null +++ b/clip_large_pretrain_4x256_all2_lr1e-4/epoch_3.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:5c5616e5cd17292180b2833dff973199638d926efd57f2e7cba862056fe43529 +size 2439236033 diff --git a/clip_large_pretrain_4x256_all2_lr1e-4/epoch_4.pth b/clip_large_pretrain_4x256_all2_lr1e-4/epoch_4.pth new file mode 100644 index 0000000000000000000000000000000000000000..0a5689753e0052b0393f44d13fbe367ec63b96c2 --- /dev/null +++ b/clip_large_pretrain_4x256_all2_lr1e-4/epoch_4.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:0be4b5436806ceba8e348fe98b6e02f1c5ec44f84108db6ef81513cf196f501b +size 2439535681 diff --git a/clip_large_pretrain_4x256_all2_lr1e-4/epoch_5.pth b/clip_large_pretrain_4x256_all2_lr1e-4/epoch_5.pth new file mode 100644 index 0000000000000000000000000000000000000000..0a6e8fff36085c6cfeeef884acb2430a2de265fa --- /dev/null +++ b/clip_large_pretrain_4x256_all2_lr1e-4/epoch_5.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:0109d51598a6a7c3e6c0444e73bf917c6b8264cf903a73c767392df8d3f7b248 +size 2439834689 diff --git a/clip_large_pretrain_4x256_all2_lr1e-4/epoch_6.pth b/clip_large_pretrain_4x256_all2_lr1e-4/epoch_6.pth new file mode 100644 index 0000000000000000000000000000000000000000..bd8ab858f7a23cf24f9628f1ebd44dbe7c16135e --- /dev/null +++ b/clip_large_pretrain_4x256_all2_lr1e-4/epoch_6.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:76633f68ddf4f27af82006665fabff45a9b132f1c7fd4cb8c9a6adb79e298413 +size 2440133505 diff --git a/clip_large_pretrain_4x256_all2_lr1e-4/epoch_7.pth b/clip_large_pretrain_4x256_all2_lr1e-4/epoch_7.pth new file mode 100644 index 0000000000000000000000000000000000000000..e53285398dcd484ac15651831398adac79a57a9f --- /dev/null +++ b/clip_large_pretrain_4x256_all2_lr1e-4/epoch_7.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:14a7793c2155ef0b156f90fc0be67ec0e517a027cbd37797225d1fcf6e55ad57 +size 2440433153 diff --git a/clip_large_pretrain_4x256_all2_lr1e-4/epoch_8.pth b/clip_large_pretrain_4x256_all2_lr1e-4/epoch_8.pth new file mode 100644 index 0000000000000000000000000000000000000000..211e627fcd76f467fa7ab077946b2ab3b1b7972b --- /dev/null +++ b/clip_large_pretrain_4x256_all2_lr1e-4/epoch_8.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:3cfc4b447b56e97e3b5678555b62a404e74bbc95d54d5241e0a09b89b3412145 +size 2440735937 diff --git a/clip_large_pretrain_4x256_all2_lr1e-4/epoch_9.pth b/clip_large_pretrain_4x256_all2_lr1e-4/epoch_9.pth new file mode 100644 index 0000000000000000000000000000000000000000..f3dc0fabc7b1fd7dff470b998badd8b76cd095f7 --- /dev/null +++ b/clip_large_pretrain_4x256_all2_lr1e-4/epoch_9.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:1538a287e6c75d72469c436802e69a36f568ca8dc47fea468af5a2892651f8e9 +size 2441038977 diff --git a/clip_large_pretrain_4x256_all2_lr1e-4/last_checkpoint b/clip_large_pretrain_4x256_all2_lr1e-4/last_checkpoint new file mode 100644 index 0000000000000000000000000000000000000000..63b374154e95935405b00ea1d67856c756e0ec56 --- /dev/null +++ b/clip_large_pretrain_4x256_all2_lr1e-4/last_checkpoint @@ -0,0 +1 @@ +/mnt/petrelfs/luzeyu/workspace/fakebench/mmpretrain/workdir/clip_large_pretrain_4x256_all2_lr1e-4/epoch_10.pth \ No newline at end of file diff --git a/clip_large_pretrain_4x256_all_lr1e-4/20230605_033556/20230605_033556.log b/clip_large_pretrain_4x256_all_lr1e-4/20230605_033556/20230605_033556.log new file mode 100644 index 0000000000000000000000000000000000000000..d09f13fe3eea1c97cf8323a841bc02c55051fc8e --- /dev/null +++ b/clip_large_pretrain_4x256_all_lr1e-4/20230605_033556/20230605_033556.log @@ -0,0 +1,1879 @@ +2023/06/05 03:36:01 - mmengine - INFO - +------------------------------------------------------------ +System environment: + sys.platform: linux + Python: 3.10.9 (main, Mar 8 2023, 10:47:38) [GCC 11.2.0] + CUDA available: True + numpy_random_seed: 1808395070 + GPU 0,1,2,3: NVIDIA A100-SXM4-80GB + CUDA_HOME: /mnt/petrelfs/share/cuda-11.6 + NVCC: Cuda compilation tools, release 11.6, V11.6.124 + GCC: gcc (GCC) 7.5.0 + PyTorch: 1.13.1 + PyTorch compiling details: PyTorch built with: + - GCC 9.3 + - C++ Version: 201402 + - Intel(R) oneAPI Math Kernel Library Version 2021.4-Product Build 20210904 for Intel(R) 64 architecture applications + - Intel(R) MKL-DNN v2.6.0 (Git Hash 52b5f107dd9cf10910aaa19cb47f3abf9b349815) + - OpenMP 201511 (a.k.a. OpenMP 4.5) + - LAPACK is enabled (usually provided by MKL) + - NNPACK is enabled + - CPU capability usage: AVX2 + - CUDA Runtime 11.6 + - NVCC architecture flags: -gencode;arch=compute_37,code=sm_37;-gencode;arch=compute_50,code=sm_50;-gencode;arch=compute_60,code=sm_60;-gencode;arch=compute_61,code=sm_61;-gencode;arch=compute_70,code=sm_70;-gencode;arch=compute_75,code=sm_75;-gencode;arch=compute_80,code=sm_80;-gencode;arch=compute_86,code=sm_86;-gencode;arch=compute_37,code=compute_37 + - CuDNN 8.3.2 (built against CUDA 11.5) + - Magma 2.6.1 + - Build settings: BLAS_INFO=mkl, BUILD_TYPE=Release, CUDA_VERSION=11.6, CUDNN_VERSION=8.3.2, CXX_COMPILER=/opt/rh/devtoolset-9/root/usr/bin/c++, CXX_FLAGS= -fabi-version=11 -Wno-deprecated -fvisibility-inlines-hidden -DUSE_PTHREADPOOL -fopenmp -DNDEBUG -DUSE_KINETO -DUSE_FBGEMM -DUSE_QNNPACK -DUSE_PYTORCH_QNNPACK -DUSE_XNNPACK -DSYMBOLICATE_MOBILE_DEBUG_HANDLE -DEDGE_PROFILER_USE_KINETO -O2 -fPIC -Wno-narrowing -Wall -Wextra -Werror=return-type -Werror=non-virtual-dtor -Wno-missing-field-initializers -Wno-type-limits -Wno-array-bounds -Wno-unknown-pragmas -Wunused-local-typedefs -Wno-unused-parameter -Wno-unused-function -Wno-unused-result -Wno-strict-overflow -Wno-strict-aliasing -Wno-error=deprecated-declarations -Wno-stringop-overflow -Wno-psabi -Wno-error=pedantic -Wno-error=redundant-decls -Wno-error=old-style-cast -fdiagnostics-color=always -faligned-new -Wno-unused-but-set-variable -Wno-maybe-uninitialized -fno-math-errno -fno-trapping-math -Werror=format -Werror=cast-function-type -Wno-stringop-overflow, LAPACK_INFO=mkl, PERF_WITH_AVX=1, PERF_WITH_AVX2=1, PERF_WITH_AVX512=1, TORCH_VERSION=1.13.1, USE_CUDA=ON, USE_CUDNN=ON, USE_EXCEPTION_PTR=1, USE_GFLAGS=OFF, USE_GLOG=OFF, USE_MKL=ON, USE_MKLDNN=ON, USE_MPI=OFF, USE_NCCL=ON, USE_NNPACK=ON, USE_OPENMP=ON, USE_ROCM=OFF, + + TorchVision: 0.14.1 + OpenCV: 4.7.0 + MMEngine: 0.7.3 + +Runtime environment: + cudnn_benchmark: True + mp_cfg: {'mp_start_method': 'fork', 'opencv_num_threads': 0} + dist_cfg: {'backend': 'nccl'} + seed: None + deterministic: False + Distributed launcher: slurm + Distributed training: True + GPU number: 4 +------------------------------------------------------------ + +2023/06/05 03:36:05 - mmengine - INFO - Config: +optim_wrapper = dict( + optimizer=dict( + type='AdamW', lr=0.0001, weight_decay=0.3, _scope_='mmpretrain'), + paramwise_cfg=dict( + custom_keys=dict({ + '.cls_token': dict(decay_mult=0.0), + '.pos_embed': dict(decay_mult=0.0) + })), + type='AmpOptimWrapper', + dtype='bfloat16', + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=False, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=4096) +model = dict( + type='ImageClassifier', + backbone=dict( + frozen_stages=24, + type='VisionTransformer', + arch='l', + img_size=224, + patch_size=14, + drop_rate=0.1, + pre_norm=True, + final_norm=False, + init_cfg=dict( + type='Pretrained', + checkpoint='ckpt/openclip-ViT-L-14.pth', + prefix='backbone')), + neck=dict( + type='CLIPProjection', + in_channels=1024, + out_channels=768, + init_cfg=dict( + type='Pretrained', + checkpoint='ckpt/openclip-ViT-L-14.pth', + prefix='backbone')), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=768, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + init_cfg=None), + init_cfg=dict( + type='TruncNormal', layer=['Conv2d', 'Linear'], std=0.02, bias=0.0), + train_cfg=None) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=128, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/all1.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/all2.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/all3.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3fake8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/cc1m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3real8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=128, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV2-1-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/if-dpmsolver++-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=128, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV2-1-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/if-dpmsolver++-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/clip_large_pretrain_4x256_all_lr1e-4' + +2023/06/05 03:36:18 - mmengine - INFO - Hooks will be executed in the following order: +before_run: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_load_checkpoint: +(ABOVE_NORMAL) EMAHook + -------------------- +before_train: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_train_epoch: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(NORMAL ) DistSamplerSeedHook + -------------------- +before_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook + -------------------- +after_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +after_train_epoch: +(NORMAL ) IterTimerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_val_epoch: +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook + -------------------- +before_val_iter: +(NORMAL ) IterTimerHook + -------------------- +after_val_iter: +(NORMAL ) IterTimerHook +(NORMAL ) VisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_val_epoch: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_save_checkpoint: +(ABOVE_NORMAL) EMAHook + -------------------- +after_train: +(VERY_LOW ) CheckpointHook + -------------------- +before_test_epoch: +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook + -------------------- +before_test_iter: +(NORMAL ) IterTimerHook + -------------------- +after_test_iter: +(NORMAL ) IterTimerHook +(NORMAL ) VisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_test_epoch: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_run: +(BELOW_NORMAL) LoggerHook + -------------------- +2023/06/05 03:36:39 - mmengine - INFO - load backbone in model from: ckpt/openclip-ViT-L-14.pth +2023/06/05 03:36:41 - mmengine - WARNING - The model and loaded state dict do not match exactly + +unexpected key in source state_dict: ln1.weight, ln1.bias + +2023/06/05 03:36:41 - mmengine - INFO - load backbone in model from: ckpt/openclip-ViT-L-14.pth +2023/06/05 03:36:43 - mmengine - WARNING - The model and loaded state dict do not match exactly + +unexpected key in source state_dict: cls_token, pos_embed, patch_embed.projection.weight, pre_norm.weight, pre_norm.bias, layers.0.ln1.weight, layers.0.ln1.bias, layers.0.attn.qkv.weight, layers.0.attn.qkv.bias, layers.0.attn.proj.weight, layers.0.attn.proj.bias, layers.0.ln2.weight, layers.0.ln2.bias, layers.0.ffn.layers.0.0.weight, layers.0.ffn.layers.0.0.bias, layers.0.ffn.layers.1.weight, layers.0.ffn.layers.1.bias, layers.1.ln1.weight, layers.1.ln1.bias, layers.1.attn.qkv.weight, layers.1.attn.qkv.bias, layers.1.attn.proj.weight, layers.1.attn.proj.bias, layers.1.ln2.weight, layers.1.ln2.bias, layers.1.ffn.layers.0.0.weight, layers.1.ffn.layers.0.0.bias, layers.1.ffn.layers.1.weight, layers.1.ffn.layers.1.bias, layers.2.ln1.weight, layers.2.ln1.bias, layers.2.attn.qkv.weight, layers.2.attn.qkv.bias, layers.2.attn.proj.weight, layers.2.attn.proj.bias, layers.2.ln2.weight, layers.2.ln2.bias, layers.2.ffn.layers.0.0.weight, layers.2.ffn.layers.0.0.bias, layers.2.ffn.layers.1.weight, layers.2.ffn.layers.1.bias, layers.3.ln1.weight, layers.3.ln1.bias, layers.3.attn.qkv.weight, layers.3.attn.qkv.bias, layers.3.attn.proj.weight, layers.3.attn.proj.bias, layers.3.ln2.weight, layers.3.ln2.bias, layers.3.ffn.layers.0.0.weight, layers.3.ffn.layers.0.0.bias, layers.3.ffn.layers.1.weight, layers.3.ffn.layers.1.bias, layers.4.ln1.weight, layers.4.ln1.bias, layers.4.attn.qkv.weight, layers.4.attn.qkv.bias, layers.4.attn.proj.weight, layers.4.attn.proj.bias, layers.4.ln2.weight, layers.4.ln2.bias, layers.4.ffn.layers.0.0.weight, layers.4.ffn.layers.0.0.bias, layers.4.ffn.layers.1.weight, layers.4.ffn.layers.1.bias, layers.5.ln1.weight, layers.5.ln1.bias, layers.5.attn.qkv.weight, layers.5.attn.qkv.bias, layers.5.attn.proj.weight, layers.5.attn.proj.bias, layers.5.ln2.weight, layers.5.ln2.bias, layers.5.ffn.layers.0.0.weight, layers.5.ffn.layers.0.0.bias, layers.5.ffn.layers.1.weight, layers.5.ffn.layers.1.bias, layers.6.ln1.weight, layers.6.ln1.bias, layers.6.attn.qkv.weight, layers.6.attn.qkv.bias, layers.6.attn.proj.weight, layers.6.attn.proj.bias, layers.6.ln2.weight, layers.6.ln2.bias, layers.6.ffn.layers.0.0.weight, layers.6.ffn.layers.0.0.bias, layers.6.ffn.layers.1.weight, layers.6.ffn.layers.1.bias, layers.7.ln1.weight, layers.7.ln1.bias, layers.7.attn.qkv.weight, layers.7.attn.qkv.bias, layers.7.attn.proj.weight, layers.7.attn.proj.bias, layers.7.ln2.weight, layers.7.ln2.bias, layers.7.ffn.layers.0.0.weight, layers.7.ffn.layers.0.0.bias, layers.7.ffn.layers.1.weight, layers.7.ffn.layers.1.bias, layers.8.ln1.weight, layers.8.ln1.bias, layers.8.attn.qkv.weight, layers.8.attn.qkv.bias, layers.8.attn.proj.weight, layers.8.attn.proj.bias, layers.8.ln2.weight, layers.8.ln2.bias, layers.8.ffn.layers.0.0.weight, layers.8.ffn.layers.0.0.bias, layers.8.ffn.layers.1.weight, layers.8.ffn.layers.1.bias, layers.9.ln1.weight, layers.9.ln1.bias, layers.9.attn.qkv.weight, layers.9.attn.qkv.bias, layers.9.attn.proj.weight, layers.9.attn.proj.bias, layers.9.ln2.weight, layers.9.ln2.bias, layers.9.ffn.layers.0.0.weight, layers.9.ffn.layers.0.0.bias, layers.9.ffn.layers.1.weight, layers.9.ffn.layers.1.bias, layers.10.ln1.weight, layers.10.ln1.bias, layers.10.attn.qkv.weight, layers.10.attn.qkv.bias, layers.10.attn.proj.weight, layers.10.attn.proj.bias, layers.10.ln2.weight, layers.10.ln2.bias, layers.10.ffn.layers.0.0.weight, layers.10.ffn.layers.0.0.bias, layers.10.ffn.layers.1.weight, layers.10.ffn.layers.1.bias, layers.11.ln1.weight, layers.11.ln1.bias, layers.11.attn.qkv.weight, layers.11.attn.qkv.bias, layers.11.attn.proj.weight, layers.11.attn.proj.bias, layers.11.ln2.weight, layers.11.ln2.bias, layers.11.ffn.layers.0.0.weight, layers.11.ffn.layers.0.0.bias, layers.11.ffn.layers.1.weight, layers.11.ffn.layers.1.bias, layers.12.ln1.weight, layers.12.ln1.bias, layers.12.attn.qkv.weight, layers.12.attn.qkv.bias, layers.12.attn.proj.weight, layers.12.attn.proj.bias, layers.12.ln2.weight, layers.12.ln2.bias, layers.12.ffn.layers.0.0.weight, layers.12.ffn.layers.0.0.bias, layers.12.ffn.layers.1.weight, layers.12.ffn.layers.1.bias, layers.13.ln1.weight, layers.13.ln1.bias, layers.13.attn.qkv.weight, layers.13.attn.qkv.bias, layers.13.attn.proj.weight, layers.13.attn.proj.bias, layers.13.ln2.weight, layers.13.ln2.bias, layers.13.ffn.layers.0.0.weight, layers.13.ffn.layers.0.0.bias, layers.13.ffn.layers.1.weight, layers.13.ffn.layers.1.bias, layers.14.ln1.weight, layers.14.ln1.bias, layers.14.attn.qkv.weight, layers.14.attn.qkv.bias, layers.14.attn.proj.weight, layers.14.attn.proj.bias, layers.14.ln2.weight, layers.14.ln2.bias, layers.14.ffn.layers.0.0.weight, layers.14.ffn.layers.0.0.bias, layers.14.ffn.layers.1.weight, layers.14.ffn.layers.1.bias, layers.15.ln1.weight, layers.15.ln1.bias, layers.15.attn.qkv.weight, layers.15.attn.qkv.bias, layers.15.attn.proj.weight, layers.15.attn.proj.bias, layers.15.ln2.weight, layers.15.ln2.bias, layers.15.ffn.layers.0.0.weight, layers.15.ffn.layers.0.0.bias, layers.15.ffn.layers.1.weight, layers.15.ffn.layers.1.bias, layers.16.ln1.weight, layers.16.ln1.bias, layers.16.attn.qkv.weight, layers.16.attn.qkv.bias, layers.16.attn.proj.weight, layers.16.attn.proj.bias, layers.16.ln2.weight, layers.16.ln2.bias, layers.16.ffn.layers.0.0.weight, layers.16.ffn.layers.0.0.bias, layers.16.ffn.layers.1.weight, layers.16.ffn.layers.1.bias, layers.17.ln1.weight, layers.17.ln1.bias, layers.17.attn.qkv.weight, layers.17.attn.qkv.bias, layers.17.attn.proj.weight, layers.17.attn.proj.bias, layers.17.ln2.weight, layers.17.ln2.bias, layers.17.ffn.layers.0.0.weight, layers.17.ffn.layers.0.0.bias, layers.17.ffn.layers.1.weight, layers.17.ffn.layers.1.bias, layers.18.ln1.weight, layers.18.ln1.bias, layers.18.attn.qkv.weight, layers.18.attn.qkv.bias, layers.18.attn.proj.weight, layers.18.attn.proj.bias, layers.18.ln2.weight, layers.18.ln2.bias, layers.18.ffn.layers.0.0.weight, layers.18.ffn.layers.0.0.bias, layers.18.ffn.layers.1.weight, layers.18.ffn.layers.1.bias, layers.19.ln1.weight, layers.19.ln1.bias, layers.19.attn.qkv.weight, layers.19.attn.qkv.bias, layers.19.attn.proj.weight, layers.19.attn.proj.bias, layers.19.ln2.weight, layers.19.ln2.bias, layers.19.ffn.layers.0.0.weight, layers.19.ffn.layers.0.0.bias, layers.19.ffn.layers.1.weight, layers.19.ffn.layers.1.bias, layers.20.ln1.weight, layers.20.ln1.bias, layers.20.attn.qkv.weight, layers.20.attn.qkv.bias, layers.20.attn.proj.weight, layers.20.attn.proj.bias, layers.20.ln2.weight, layers.20.ln2.bias, layers.20.ffn.layers.0.0.weight, layers.20.ffn.layers.0.0.bias, layers.20.ffn.layers.1.weight, layers.20.ffn.layers.1.bias, layers.21.ln1.weight, layers.21.ln1.bias, layers.21.attn.qkv.weight, layers.21.attn.qkv.bias, layers.21.attn.proj.weight, layers.21.attn.proj.bias, layers.21.ln2.weight, layers.21.ln2.bias, layers.21.ffn.layers.0.0.weight, layers.21.ffn.layers.0.0.bias, layers.21.ffn.layers.1.weight, layers.21.ffn.layers.1.bias, layers.22.ln1.weight, layers.22.ln1.bias, layers.22.attn.qkv.weight, layers.22.attn.qkv.bias, layers.22.attn.proj.weight, layers.22.attn.proj.bias, layers.22.ln2.weight, layers.22.ln2.bias, layers.22.ffn.layers.0.0.weight, layers.22.ffn.layers.0.0.bias, layers.22.ffn.layers.1.weight, layers.22.ffn.layers.1.bias, layers.23.ln1.weight, layers.23.ln1.bias, layers.23.attn.qkv.weight, layers.23.attn.qkv.bias, layers.23.attn.proj.weight, layers.23.attn.proj.bias, layers.23.ln2.weight, layers.23.ln2.bias, layers.23.ffn.layers.0.0.weight, layers.23.ffn.layers.0.0.bias, layers.23.ffn.layers.1.weight, layers.23.ffn.layers.1.bias, ln1.weight, ln1.bias + +missing keys in source state_dict: proj + +Name of parameter - Initialization information + +backbone.cls_token - torch.Size([1, 1, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.pos_embed - torch.Size([1, 257, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.patch_embed.projection.weight - torch.Size([1024, 3, 14, 14]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.0.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.0.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.0.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.0.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.0.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.0.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.0.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.0.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.0.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.0.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.0.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.0.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.1.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.1.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.1.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.1.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.1.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.1.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.1.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.1.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.1.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.1.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.1.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.1.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.2.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.2.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.2.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.2.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.2.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.2.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.2.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.2.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.2.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.2.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.2.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.2.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.3.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.3.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.3.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.3.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.3.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.3.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.3.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.3.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.3.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.3.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.3.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.3.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.4.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.4.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.4.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.4.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.4.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.4.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.4.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.4.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.4.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.4.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.4.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.4.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.5.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.5.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.5.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.5.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.5.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.5.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.5.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.5.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.5.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.5.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.5.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.5.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.6.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.6.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.6.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.6.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.6.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.6.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.6.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.6.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.6.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.6.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.6.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.6.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.7.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.7.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.7.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.7.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.7.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.7.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.7.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.7.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.7.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.7.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.7.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.7.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.8.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.8.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.8.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.8.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.8.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.8.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.8.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.8.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.8.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.8.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.8.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.8.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.9.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.9.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.9.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.9.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.9.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.9.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.9.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.9.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.9.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.9.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.9.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.9.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.10.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.10.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.10.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.10.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.10.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.10.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.10.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.10.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.10.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.10.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.10.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.10.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.11.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.11.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.11.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.11.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.11.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.11.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.11.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.11.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.11.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.11.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.11.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.11.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.12.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.12.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.12.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.12.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.12.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.12.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.12.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.12.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.12.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.12.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.12.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.12.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.13.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.13.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.13.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.13.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.13.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.13.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.13.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.13.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.13.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.13.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.13.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.13.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.14.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.14.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.14.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.14.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.14.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.14.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.14.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.14.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.14.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.14.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.14.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.14.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.15.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.15.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.15.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.15.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.15.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.15.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.15.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.15.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.15.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.15.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.15.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.15.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.16.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.16.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.16.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.16.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.16.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.16.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.16.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.16.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.16.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.16.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.16.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.16.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.17.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.17.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.17.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.17.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.17.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.17.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.17.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.17.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.17.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.17.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.17.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.17.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.18.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.18.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.18.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.18.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.18.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.18.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.18.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.18.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.18.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.18.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.18.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.18.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.19.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.19.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.19.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.19.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.19.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.19.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.19.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.19.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.19.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.19.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.19.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.19.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.20.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.20.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.20.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.20.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.20.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.20.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.20.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.20.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.20.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.20.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.20.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.20.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.21.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.21.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.21.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.21.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.21.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.21.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.21.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.21.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.21.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.21.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.21.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.21.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.22.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.22.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.22.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.22.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.22.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.22.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.22.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.22.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.22.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.22.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.22.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.22.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.23.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.23.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.23.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.23.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.23.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.23.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.23.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.23.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.23.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.23.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.23.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.23.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.pre_norm.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.pre_norm.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +neck.proj - torch.Size([1024, 768]): +The value is the same before and after calling `init_weights` of ImageClassifier + +head.fc.weight - torch.Size([2, 768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +head.fc.bias - torch.Size([2]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 +2023/06/05 03:36:43 - mmengine - WARNING - "FileClient" will be deprecated in future. Please use io functions in https://mmengine.readthedocs.io/en/latest/api/fileio.html#file-io +2023/06/05 03:36:43 - mmengine - WARNING - "HardDiskBackend" is the alias of "LocalBackend" and the former will be deprecated in future. +2023/06/05 03:36:43 - mmengine - INFO - Checkpoints will be saved to /mnt/petrelfs/luzeyu/workspace/fakebench/mmpretrain/workdir/clip_large_pretrain_4x256_all_lr1e-4. +2023/06/05 03:37:53 - mmengine - INFO - Epoch(train) [1][ 100/4111] lr: 9.9999e-05 eta: 8:01:28 time: 0.6297 data_time: 0.0025 memory: 44138 loss: 0.5927 +2023/06/05 03:38:56 - mmengine - INFO - Epoch(train) [1][ 200/4111] lr: 9.9995e-05 eta: 7:34:55 time: 0.6299 data_time: 0.0024 memory: 44138 loss: 0.5558 +2023/06/05 03:39:59 - mmengine - INFO - Epoch(train) [1][ 300/4111] lr: 9.9988e-05 eta: 7:25:26 time: 0.6311 data_time: 0.0024 memory: 44138 loss: 0.5518 +2023/06/05 03:41:02 - mmengine - INFO - Epoch(train) [1][ 400/4111] lr: 9.9979e-05 eta: 7:20:19 time: 0.6294 data_time: 0.0021 memory: 44138 loss: 0.5357 +2023/06/05 03:42:06 - mmengine - INFO - Epoch(train) [1][ 500/4111] lr: 9.9967e-05 eta: 7:17:06 time: 0.6362 data_time: 0.0015 memory: 44138 loss: 0.5294 +2023/06/05 03:43:09 - mmengine - INFO - Epoch(train) [1][ 600/4111] lr: 9.9953e-05 eta: 7:14:21 time: 0.6290 data_time: 0.0016 memory: 44138 loss: 0.5346 +2023/06/05 03:44:12 - mmengine - INFO - Epoch(train) [1][ 700/4111] lr: 9.9936e-05 eta: 7:12:05 time: 0.6303 data_time: 0.0018 memory: 44138 loss: 0.5463 +2023/06/05 03:45:15 - mmengine - INFO - Epoch(train) [1][ 800/4111] lr: 9.9916e-05 eta: 7:10:03 time: 0.6300 data_time: 0.0015 memory: 44138 loss: 0.5194 +2023/06/05 03:46:18 - mmengine - INFO - Epoch(train) [1][ 900/4111] lr: 9.9894e-05 eta: 7:08:14 time: 0.6303 data_time: 0.0016 memory: 44138 loss: 0.5282 +2023/06/05 03:47:21 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_all_lr1e-4_20230605_033556 +2023/06/05 03:47:21 - mmengine - INFO - Epoch(train) [1][1000/4111] lr: 9.9869e-05 eta: 7:06:35 time: 0.6305 data_time: 0.0017 memory: 44138 loss: 0.5352 +2023/06/05 03:48:24 - mmengine - INFO - Epoch(train) [1][1100/4111] lr: 9.9841e-05 eta: 7:05:06 time: 0.6299 data_time: 0.0021 memory: 44138 loss: 0.4952 +2023/06/05 03:49:27 - mmengine - INFO - Epoch(train) [1][1200/4111] lr: 9.9811e-05 eta: 7:03:41 time: 0.6306 data_time: 0.0021 memory: 44138 loss: 0.5200 +2023/06/05 03:50:30 - mmengine - INFO - Epoch(train) [1][1300/4111] lr: 9.9778e-05 eta: 7:02:18 time: 0.6302 data_time: 0.0016 memory: 44138 loss: 0.5058 +2023/06/05 03:51:33 - mmengine - INFO - Epoch(train) [1][1400/4111] lr: 9.9743e-05 eta: 7:00:59 time: 0.6308 data_time: 0.0017 memory: 44138 loss: 0.5007 +2023/06/05 03:52:36 - mmengine - INFO - Epoch(train) [1][1500/4111] lr: 9.9705e-05 eta: 6:59:42 time: 0.6307 data_time: 0.0015 memory: 44138 loss: 0.5251 +2023/06/05 03:53:40 - mmengine - INFO - Epoch(train) [1][1600/4111] lr: 9.9664e-05 eta: 6:58:27 time: 0.6309 data_time: 0.0017 memory: 44138 loss: 0.5359 +2023/06/05 03:54:43 - mmengine - INFO - Epoch(train) [1][1700/4111] lr: 9.9621e-05 eta: 6:57:15 time: 0.6306 data_time: 0.0017 memory: 44138 loss: 0.5241 +2023/06/05 03:55:46 - mmengine - INFO - Epoch(train) [1][1800/4111] lr: 9.9575e-05 eta: 6:56:06 time: 0.6323 data_time: 0.0023 memory: 44138 loss: 0.5213 +2023/06/05 03:56:49 - mmengine - INFO - Epoch(train) [1][1900/4111] lr: 9.9527e-05 eta: 6:54:56 time: 0.6316 data_time: 0.0018 memory: 44138 loss: 0.5202 +2023/06/05 03:57:52 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_all_lr1e-4_20230605_033556 +2023/06/05 03:57:52 - mmengine - INFO - Epoch(train) [1][2000/4111] lr: 9.9476e-05 eta: 6:53:46 time: 0.6295 data_time: 0.0017 memory: 44138 loss: 0.5050 +2023/06/05 03:58:56 - mmengine - INFO - Epoch(train) [1][2100/4111] lr: 9.9422e-05 eta: 6:52:38 time: 0.6308 data_time: 0.0016 memory: 44138 loss: 0.5165 +2023/06/05 03:59:59 - mmengine - INFO - Epoch(train) [1][2200/4111] lr: 9.9366e-05 eta: 6:51:32 time: 0.6295 data_time: 0.0018 memory: 44138 loss: 0.5222 +2023/06/05 04:01:02 - mmengine - INFO - Epoch(train) [1][2300/4111] lr: 9.9307e-05 eta: 6:50:22 time: 0.6309 data_time: 0.0017 memory: 44138 loss: 0.5056 +2023/06/05 04:02:05 - mmengine - INFO - Epoch(train) [1][2400/4111] lr: 9.9246e-05 eta: 6:49:14 time: 0.6310 data_time: 0.0019 memory: 44138 loss: 0.4989 +2023/06/05 04:03:08 - mmengine - INFO - Epoch(train) [1][2500/4111] lr: 9.9182e-05 eta: 6:48:06 time: 0.6320 data_time: 0.0015 memory: 44138 loss: 0.4959 +2023/06/05 04:04:11 - mmengine - INFO - Epoch(train) [1][2600/4111] lr: 9.9115e-05 eta: 6:46:58 time: 0.6303 data_time: 0.0019 memory: 44138 loss: 0.4856 +2023/06/05 04:05:14 - mmengine - INFO - Epoch(train) [1][2700/4111] lr: 9.9046e-05 eta: 6:45:49 time: 0.6311 data_time: 0.0016 memory: 44138 loss: 0.5095 +2023/06/05 04:06:18 - mmengine - INFO - Epoch(train) [1][2800/4111] lr: 9.8974e-05 eta: 6:44:42 time: 0.6291 data_time: 0.0017 memory: 44138 loss: 0.5114 +2023/06/05 04:07:21 - mmengine - INFO - Epoch(train) [1][2900/4111] lr: 9.8900e-05 eta: 6:43:33 time: 0.6291 data_time: 0.0018 memory: 44138 loss: 0.4861 +2023/06/05 04:08:24 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_all_lr1e-4_20230605_033556 +2023/06/05 04:08:24 - mmengine - INFO - Epoch(train) [1][3000/4111] lr: 9.8823e-05 eta: 6:42:25 time: 0.6333 data_time: 0.0014 memory: 44138 loss: 0.5100 +2023/06/05 04:09:27 - mmengine - INFO - Epoch(train) [1][3100/4111] lr: 9.8744e-05 eta: 6:41:18 time: 0.6303 data_time: 0.0015 memory: 44138 loss: 0.4875 +2023/06/05 04:10:30 - mmengine - INFO - Epoch(train) [1][3200/4111] lr: 9.8662e-05 eta: 6:40:12 time: 0.6320 data_time: 0.0019 memory: 44138 loss: 0.5040 +2023/06/05 04:11:33 - mmengine - INFO - Epoch(train) [1][3300/4111] lr: 9.8578e-05 eta: 6:39:05 time: 0.6299 data_time: 0.0016 memory: 44138 loss: 0.4849 +2023/06/05 04:12:36 - mmengine - INFO - Epoch(train) [1][3400/4111] lr: 9.8490e-05 eta: 6:37:58 time: 0.6313 data_time: 0.0015 memory: 44138 loss: 0.5121 +2023/06/05 04:13:39 - mmengine - INFO - Epoch(train) [1][3500/4111] lr: 9.8401e-05 eta: 6:36:53 time: 0.6296 data_time: 0.0016 memory: 44138 loss: 0.4873 +2023/06/05 04:14:42 - mmengine - INFO - Epoch(train) [1][3600/4111] lr: 9.8309e-05 eta: 6:35:47 time: 0.6311 data_time: 0.0018 memory: 44138 loss: 0.4967 +2023/06/05 04:15:45 - mmengine - INFO - Epoch(train) [1][3700/4111] lr: 9.8214e-05 eta: 6:34:41 time: 0.6310 data_time: 0.0017 memory: 44138 loss: 0.5099 +2023/06/05 04:16:48 - mmengine - INFO - Epoch(train) [1][3800/4111] lr: 9.8117e-05 eta: 6:33:36 time: 0.6324 data_time: 0.0018 memory: 44138 loss: 0.4761 +2023/06/05 04:17:51 - mmengine - INFO - Epoch(train) [1][3900/4111] lr: 9.8017e-05 eta: 6:32:31 time: 0.6321 data_time: 0.0017 memory: 44138 loss: 0.4972 +2023/06/05 04:18:54 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_all_lr1e-4_20230605_033556 +2023/06/05 04:18:54 - mmengine - INFO - Epoch(train) [1][4000/4111] lr: 9.7915e-05 eta: 6:31:26 time: 0.6314 data_time: 0.0014 memory: 44138 loss: 0.5002 +2023/06/05 04:19:58 - mmengine - INFO - Epoch(train) [1][4100/4111] lr: 9.7810e-05 eta: 6:30:22 time: 0.6343 data_time: 0.0049 memory: 44138 loss: 0.4714 +2023/06/05 04:20:04 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_all_lr1e-4_20230605_033556 +2023/06/05 04:20:04 - mmengine - INFO - Saving checkpoint at 1 epochs +2023/06/05 04:22:29 - mmengine - INFO - Epoch(val) [1][100/119] eta: 0:00:25 time: 1.2830 data_time: 0.0009 memory: 44138 +2023/06/05 04:23:14 - mmengine - INFO - Epoch(val) [1][119/119] accuracy/top1: 71.2460 single-label/precision_classwise: [96.17422485351562, 41.946285247802734] single-label/recall_classwise: [66.06906127929688, 90.31782531738281] single-label/f1-score_classwise: [78.32855224609375, 57.28684997558594] data_time: 0.0260 time: 1.3183 +2023/06/05 04:24:21 - mmengine - INFO - Epoch(train) [2][ 100/4111] lr: 9.7691e-05 eta: 6:29:39 time: 0.6337 data_time: 0.0024 memory: 44140 loss: 0.4788 +2023/06/05 04:25:24 - mmengine - INFO - Epoch(train) [2][ 200/4111] lr: 9.7581e-05 eta: 6:28:36 time: 0.6336 data_time: 0.0016 memory: 44140 loss: 0.4926 +2023/06/05 04:27:57 - mmengine - INFO - Epoch(train) [2][ 300/4111] lr: 9.7469e-05 eta: 6:39:55 time: 0.6323 data_time: 0.0022 memory: 44140 loss: 0.4985 +2023/06/05 04:29:00 - mmengine - INFO - Epoch(train) [2][ 400/4111] lr: 9.7354e-05 eta: 6:38:33 time: 0.6318 data_time: 0.0019 memory: 44140 loss: 0.4467 +2023/06/05 04:30:04 - mmengine - INFO - Epoch(train) [2][ 500/4111] lr: 9.7236e-05 eta: 6:37:12 time: 0.6334 data_time: 0.0015 memory: 44140 loss: 0.4997 +2023/06/05 04:31:07 - mmengine - INFO - Epoch(train) [2][ 600/4111] lr: 9.7116e-05 eta: 6:35:51 time: 0.6320 data_time: 0.0015 memory: 44140 loss: 0.4525 +2023/06/05 04:32:10 - mmengine - INFO - Epoch(train) [2][ 700/4111] lr: 9.6994e-05 eta: 6:34:30 time: 0.6309 data_time: 0.0018 memory: 44140 loss: 0.4845 +2023/06/05 04:33:13 - mmengine - INFO - Epoch(train) [2][ 800/4111] lr: 9.6869e-05 eta: 6:33:11 time: 0.6335 data_time: 0.0018 memory: 44140 loss: 0.4613 +2023/06/05 04:34:10 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_all_lr1e-4_20230605_033556 +2023/06/05 04:34:17 - mmengine - INFO - Epoch(train) [2][ 900/4111] lr: 9.6742e-05 eta: 6:31:52 time: 0.6328 data_time: 0.0015 memory: 44140 loss: 0.4840 +2023/06/05 04:35:20 - mmengine - INFO - Epoch(train) [2][1000/4111] lr: 9.6612e-05 eta: 6:30:34 time: 0.6317 data_time: 0.0021 memory: 44140 loss: 0.4813 +2023/06/05 04:36:23 - mmengine - INFO - Epoch(train) [2][1100/4111] lr: 9.6480e-05 eta: 6:29:15 time: 0.6318 data_time: 0.0015 memory: 44140 loss: 0.4751 +2023/06/05 04:37:26 - mmengine - INFO - Epoch(train) [2][1200/4111] lr: 9.6346e-05 eta: 6:27:58 time: 0.6307 data_time: 0.0014 memory: 44140 loss: 0.4710 +2023/06/05 04:38:30 - mmengine - INFO - Epoch(train) [2][1300/4111] lr: 9.6209e-05 eta: 6:26:41 time: 0.6319 data_time: 0.0014 memory: 44140 loss: 0.4979 +2023/06/05 04:39:33 - mmengine - INFO - Epoch(train) [2][1400/4111] lr: 9.6069e-05 eta: 6:25:26 time: 0.6350 data_time: 0.0019 memory: 44140 loss: 0.4742 +2023/06/05 04:40:36 - mmengine - INFO - Epoch(train) [2][1500/4111] lr: 9.5928e-05 eta: 6:24:10 time: 0.6317 data_time: 0.0018 memory: 44140 loss: 0.4734 +2023/06/05 04:41:39 - mmengine - INFO - Epoch(train) [2][1600/4111] lr: 9.5783e-05 eta: 6:22:54 time: 0.6309 data_time: 0.0021 memory: 44140 loss: 0.4630 +2023/06/05 04:42:43 - mmengine - INFO - Epoch(train) [2][1700/4111] lr: 9.5637e-05 eta: 6:21:40 time: 0.6313 data_time: 0.0017 memory: 44140 loss: 0.4517 +2023/06/05 04:43:46 - mmengine - INFO - Epoch(train) [2][1800/4111] lr: 9.5488e-05 eta: 6:20:25 time: 0.6332 data_time: 0.0018 memory: 44140 loss: 0.4670 +2023/06/05 04:44:42 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_all_lr1e-4_20230605_033556 +2023/06/05 04:44:49 - mmengine - INFO - Epoch(train) [2][1900/4111] lr: 9.5337e-05 eta: 6:19:12 time: 0.6328 data_time: 0.0016 memory: 44140 loss: 0.4922 +2023/06/05 04:45:53 - mmengine - INFO - Epoch(train) [2][2000/4111] lr: 9.5183e-05 eta: 6:17:58 time: 0.6313 data_time: 0.0016 memory: 44140 loss: 0.4860 +2023/06/05 04:46:56 - mmengine - INFO - Epoch(train) [2][2100/4111] lr: 9.5027e-05 eta: 6:16:44 time: 0.6315 data_time: 0.0015 memory: 44140 loss: 0.4803 +2023/06/05 04:47:59 - mmengine - INFO - Epoch(train) [2][2200/4111] lr: 9.4869e-05 eta: 6:15:31 time: 0.6318 data_time: 0.0016 memory: 44140 loss: 0.4786 +2023/06/05 04:49:02 - mmengine - INFO - Epoch(train) [2][2300/4111] lr: 9.4708e-05 eta: 6:14:18 time: 0.6338 data_time: 0.0020 memory: 44140 loss: 0.4707 +2023/06/05 04:50:06 - mmengine - INFO - Epoch(train) [2][2400/4111] lr: 9.4545e-05 eta: 6:13:06 time: 0.6318 data_time: 0.0017 memory: 44140 loss: 0.4744 +2023/06/05 04:51:09 - mmengine - INFO - Epoch(train) [2][2500/4111] lr: 9.4380e-05 eta: 6:11:54 time: 0.6303 data_time: 0.0015 memory: 44140 loss: 0.4474 +2023/06/05 04:52:12 - mmengine - INFO - Epoch(train) [2][2600/4111] lr: 9.4212e-05 eta: 6:10:42 time: 0.6303 data_time: 0.0017 memory: 44140 loss: 0.4806 +2023/06/05 04:53:16 - mmengine - INFO - Epoch(train) [2][2700/4111] lr: 9.4043e-05 eta: 6:09:30 time: 0.6357 data_time: 0.0017 memory: 44140 loss: 0.4818 +2023/06/05 04:54:19 - mmengine - INFO - Epoch(train) [2][2800/4111] lr: 9.3870e-05 eta: 6:08:19 time: 0.6315 data_time: 0.0015 memory: 44140 loss: 0.4747 +2023/06/05 04:55:15 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_all_lr1e-4_20230605_033556 +2023/06/05 04:55:22 - mmengine - INFO - Epoch(train) [2][2900/4111] lr: 9.3696e-05 eta: 6:07:08 time: 0.6310 data_time: 0.0015 memory: 44140 loss: 0.4685 +2023/06/05 04:56:25 - mmengine - INFO - Epoch(train) [2][3000/4111] lr: 9.3519e-05 eta: 6:05:57 time: 0.6315 data_time: 0.0025 memory: 44140 loss: 0.5209 +2023/06/05 04:57:29 - mmengine - INFO - Epoch(train) [2][3100/4111] lr: 9.3340e-05 eta: 6:04:46 time: 0.6345 data_time: 0.0015 memory: 44140 loss: 0.4742 +2023/06/05 04:58:32 - mmengine - INFO - Epoch(train) [2][3200/4111] lr: 9.3159e-05 eta: 6:03:36 time: 0.6340 data_time: 0.0019 memory: 44140 loss: 0.4483 +2023/06/05 04:59:35 - mmengine - INFO - Epoch(train) [2][3300/4111] lr: 9.2976e-05 eta: 6:02:25 time: 0.6328 data_time: 0.0016 memory: 44140 loss: 0.4724 +2023/06/05 05:00:39 - mmengine - INFO - Epoch(train) [2][3400/4111] lr: 9.2790e-05 eta: 6:01:16 time: 0.6366 data_time: 0.0017 memory: 44140 loss: 0.4809 +2023/06/05 05:01:42 - mmengine - INFO - Epoch(train) [2][3500/4111] lr: 9.2603e-05 eta: 6:00:06 time: 0.6328 data_time: 0.0020 memory: 44140 loss: 0.4781 +2023/06/05 05:02:45 - mmengine - INFO - Epoch(train) [2][3600/4111] lr: 9.2413e-05 eta: 5:58:57 time: 0.6336 data_time: 0.0027 memory: 44140 loss: 0.4530 +2023/06/05 05:03:49 - mmengine - INFO - Epoch(train) [2][3700/4111] lr: 9.2220e-05 eta: 5:57:47 time: 0.6314 data_time: 0.0018 memory: 44140 loss: 0.4671 +2023/06/05 05:04:52 - mmengine - INFO - Epoch(train) [2][3800/4111] lr: 9.2026e-05 eta: 5:56:38 time: 0.6327 data_time: 0.0020 memory: 44140 loss: 0.4647 +2023/06/05 05:05:48 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_all_lr1e-4_20230605_033556 +2023/06/05 05:05:55 - mmengine - INFO - Epoch(train) [2][3900/4111] lr: 9.1829e-05 eta: 5:55:28 time: 0.6310 data_time: 0.0018 memory: 44140 loss: 0.4601 +2023/06/05 05:06:58 - mmengine - INFO - Epoch(train) [2][4000/4111] lr: 9.1631e-05 eta: 5:54:19 time: 0.6321 data_time: 0.0019 memory: 44140 loss: 0.4363 +2023/06/05 05:08:02 - mmengine - INFO - Epoch(train) [2][4100/4111] lr: 9.1430e-05 eta: 5:53:10 time: 0.6357 data_time: 0.0032 memory: 44140 loss: 0.4708 +2023/06/05 05:08:09 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_all_lr1e-4_20230605_033556 +2023/06/05 05:08:09 - mmengine - INFO - Saving checkpoint at 2 epochs +2023/06/05 05:10:30 - mmengine - INFO - Epoch(val) [2][100/119] eta: 0:00:24 time: 1.2824 data_time: 0.0006 memory: 44140 +2023/06/05 05:11:15 - mmengine - INFO - Epoch(val) [2][119/119] accuracy/top1: 83.3722 single-label/precision_classwise: [95.01081085205078, 57.59059143066406] single-label/recall_classwise: [83.22917175292969, 83.89922332763672] single-label/f1-score_classwise: [88.73060607910156, 68.29899597167969] data_time: 0.0194 time: 1.2812 +2023/06/05 05:12:22 - mmengine - INFO - Epoch(train) [3][ 100/4111] lr: 9.1205e-05 eta: 5:52:07 time: 0.6308 data_time: 0.0016 memory: 44140 loss: 0.4665 +2023/06/05 05:13:25 - mmengine - INFO - Epoch(train) [3][ 200/4111] lr: 9.0999e-05 eta: 5:50:57 time: 0.6314 data_time: 0.0017 memory: 44140 loss: 0.4543 +2023/06/05 05:14:28 - mmengine - INFO - Epoch(train) [3][ 300/4111] lr: 9.0792e-05 eta: 5:49:48 time: 0.6316 data_time: 0.0016 memory: 44140 loss: 0.4859 +2023/06/05 05:15:36 - mmengine - INFO - Epoch(train) [3][ 400/4111] lr: 9.0582e-05 eta: 5:48:59 time: 0.6309 data_time: 0.0019 memory: 44140 loss: 0.4938 +2023/06/05 05:16:40 - mmengine - INFO - Epoch(train) [3][ 500/4111] lr: 9.0371e-05 eta: 5:47:50 time: 0.6376 data_time: 0.0029 memory: 44140 loss: 0.4654 +2023/06/05 05:17:43 - mmengine - INFO - Epoch(train) [3][ 600/4111] lr: 9.0157e-05 eta: 5:46:41 time: 0.6311 data_time: 0.0015 memory: 44140 loss: 0.4408 +2023/06/05 05:18:46 - mmengine - INFO - Epoch(train) [3][ 700/4111] lr: 8.9942e-05 eta: 5:45:33 time: 0.6321 data_time: 0.0022 memory: 44140 loss: 0.4499 +2023/06/05 05:19:35 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_all_lr1e-4_20230605_033556 +2023/06/05 05:19:49 - mmengine - INFO - Epoch(train) [3][ 800/4111] lr: 8.9724e-05 eta: 5:44:24 time: 0.6318 data_time: 0.0016 memory: 44140 loss: 0.4765 +2023/06/05 05:20:53 - mmengine - INFO - Epoch(train) [3][ 900/4111] lr: 8.9504e-05 eta: 5:43:16 time: 0.6368 data_time: 0.0019 memory: 44140 loss: 0.4790 +2023/06/05 05:21:56 - mmengine - INFO - Epoch(train) [3][1000/4111] lr: 8.9282e-05 eta: 5:42:08 time: 0.6322 data_time: 0.0017 memory: 44140 loss: 0.4671 +2023/06/05 05:22:59 - mmengine - INFO - Epoch(train) [3][1100/4111] lr: 8.9059e-05 eta: 5:40:59 time: 0.6313 data_time: 0.0019 memory: 44140 loss: 0.4840 +2023/06/05 05:24:02 - mmengine - INFO - Epoch(train) [3][1200/4111] lr: 8.8833e-05 eta: 5:39:51 time: 0.6308 data_time: 0.0016 memory: 44140 loss: 0.4634 +2023/06/05 05:25:05 - mmengine - INFO - Epoch(train) [3][1300/4111] lr: 8.8605e-05 eta: 5:38:43 time: 0.6346 data_time: 0.0016 memory: 44140 loss: 0.4593 +2023/06/05 05:26:09 - mmengine - INFO - Epoch(train) [3][1400/4111] lr: 8.8375e-05 eta: 5:37:35 time: 0.6331 data_time: 0.0015 memory: 44140 loss: 0.4885 +2023/06/05 05:27:12 - mmengine - INFO - Epoch(train) [3][1500/4111] lr: 8.8144e-05 eta: 5:36:27 time: 0.6330 data_time: 0.0016 memory: 44140 loss: 0.4521 +2023/06/05 05:28:15 - mmengine - INFO - Epoch(train) [3][1600/4111] lr: 8.7910e-05 eta: 5:35:20 time: 0.6331 data_time: 0.0016 memory: 44140 loss: 0.4514 +2023/06/05 05:29:19 - mmengine - INFO - Epoch(train) [3][1700/4111] lr: 8.7675e-05 eta: 5:34:12 time: 0.6307 data_time: 0.0015 memory: 44140 loss: 0.4571 +2023/06/05 05:30:08 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_all_lr1e-4_20230605_033556 +2023/06/05 05:30:22 - mmengine - INFO - Epoch(train) [3][1800/4111] lr: 8.7437e-05 eta: 5:33:04 time: 0.6313 data_time: 0.0016 memory: 44140 loss: 0.4745 +2023/06/05 05:31:25 - mmengine - INFO - Epoch(train) [3][1900/4111] lr: 8.7198e-05 eta: 5:31:57 time: 0.6320 data_time: 0.0016 memory: 44140 loss: 0.4391 +2023/06/05 05:32:28 - mmengine - INFO - Epoch(train) [3][2000/4111] lr: 8.6957e-05 eta: 5:30:49 time: 0.6314 data_time: 0.0015 memory: 44140 loss: 0.4638 +2023/06/05 05:33:32 - mmengine - INFO - Epoch(train) [3][2100/4111] lr: 8.6714e-05 eta: 5:29:42 time: 0.6314 data_time: 0.0025 memory: 44140 loss: 0.4337 +2023/06/05 05:34:35 - mmengine - INFO - Epoch(train) [3][2200/4111] lr: 8.6469e-05 eta: 5:28:35 time: 0.6314 data_time: 0.0016 memory: 44140 loss: 0.4856 +2023/06/05 05:35:38 - mmengine - INFO - Epoch(train) [3][2300/4111] lr: 8.6222e-05 eta: 5:27:28 time: 0.6315 data_time: 0.0016 memory: 44140 loss: 0.4622 +2023/06/05 05:36:41 - mmengine - INFO - Epoch(train) [3][2400/4111] lr: 8.5974e-05 eta: 5:26:21 time: 0.6316 data_time: 0.0016 memory: 44140 loss: 0.4460 +2023/06/05 05:37:44 - mmengine - INFO - Epoch(train) [3][2500/4111] lr: 8.5723e-05 eta: 5:25:14 time: 0.6339 data_time: 0.0017 memory: 44140 loss: 0.4644 +2023/06/05 05:38:48 - mmengine - INFO - Epoch(train) [3][2600/4111] lr: 8.5471e-05 eta: 5:24:06 time: 0.6322 data_time: 0.0020 memory: 44140 loss: 0.4641 +2023/06/05 05:39:51 - mmengine - INFO - Epoch(train) [3][2700/4111] lr: 8.5217e-05 eta: 5:22:59 time: 0.6318 data_time: 0.0023 memory: 44140 loss: 0.4477 +2023/06/05 05:40:40 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_all_lr1e-4_20230605_033556 +2023/06/05 05:40:54 - mmengine - INFO - Epoch(train) [3][2800/4111] lr: 8.4962e-05 eta: 5:21:52 time: 0.6307 data_time: 0.0015 memory: 44140 loss: 0.4819 +2023/06/05 05:41:57 - mmengine - INFO - Epoch(train) [3][2900/4111] lr: 8.4704e-05 eta: 5:20:45 time: 0.6301 data_time: 0.0018 memory: 44140 loss: 0.4432 +2023/06/05 05:43:00 - mmengine - INFO - Epoch(train) [3][3000/4111] lr: 8.4445e-05 eta: 5:19:38 time: 0.6304 data_time: 0.0015 memory: 44140 loss: 0.4517 +2023/06/05 05:44:03 - mmengine - INFO - Epoch(train) [3][3100/4111] lr: 8.4184e-05 eta: 5:18:32 time: 0.6331 data_time: 0.0018 memory: 44140 loss: 0.4714 +2023/06/05 05:45:07 - mmengine - INFO - Epoch(train) [3][3200/4111] lr: 8.3921e-05 eta: 5:17:25 time: 0.6334 data_time: 0.0015 memory: 44140 loss: 0.4962 +2023/06/05 05:46:10 - mmengine - INFO - Epoch(train) [3][3300/4111] lr: 8.3657e-05 eta: 5:16:18 time: 0.6299 data_time: 0.0017 memory: 44140 loss: 0.4454 +2023/06/05 05:47:13 - mmengine - INFO - Epoch(train) [3][3400/4111] lr: 8.3391e-05 eta: 5:15:11 time: 0.6316 data_time: 0.0017 memory: 44140 loss: 0.4743 +2023/06/05 05:48:16 - mmengine - INFO - Epoch(train) [3][3500/4111] lr: 8.3123e-05 eta: 5:14:05 time: 0.6316 data_time: 0.0017 memory: 44140 loss: 0.4630 +2023/06/05 05:49:19 - mmengine - INFO - Epoch(train) [3][3600/4111] lr: 8.2854e-05 eta: 5:12:59 time: 0.6338 data_time: 0.0024 memory: 44140 loss: 0.4666 +2023/06/05 05:50:23 - mmengine - INFO - Epoch(train) [3][3700/4111] lr: 8.2583e-05 eta: 5:11:53 time: 0.6342 data_time: 0.0015 memory: 44140 loss: 0.4419 +2023/06/05 05:51:12 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_all_lr1e-4_20230605_033556 +2023/06/05 05:51:26 - mmengine - INFO - Epoch(train) [3][3800/4111] lr: 8.2311e-05 eta: 5:10:47 time: 0.6319 data_time: 0.0021 memory: 44140 loss: 0.4865 +2023/06/05 05:52:29 - mmengine - INFO - Epoch(train) [3][3900/4111] lr: 8.2037e-05 eta: 5:09:41 time: 0.6315 data_time: 0.0016 memory: 44140 loss: 0.4489 +2023/06/05 05:53:33 - mmengine - INFO - Epoch(train) [3][4000/4111] lr: 8.1761e-05 eta: 5:08:35 time: 0.6322 data_time: 0.0017 memory: 44140 loss: 0.4664 +2023/06/05 05:54:36 - mmengine - INFO - Epoch(train) [3][4100/4111] lr: 8.1484e-05 eta: 5:07:29 time: 0.6356 data_time: 0.0034 memory: 44140 loss: 0.4785 +2023/06/05 05:54:43 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_all_lr1e-4_20230605_033556 +2023/06/05 05:54:43 - mmengine - INFO - Saving checkpoint at 3 epochs +2023/06/05 05:57:05 - mmengine - INFO - Epoch(val) [3][100/119] eta: 0:00:24 time: 1.3053 data_time: 0.0016 memory: 44140 +2023/06/05 05:57:49 - mmengine - INFO - Epoch(val) [3][119/119] accuracy/top1: 86.2403 single-label/precision_classwise: [94.41134643554688, 64.08477020263672] single-label/recall_classwise: [87.69648742675781, 80.87596893310547] single-label/f1-score_classwise: [90.93011474609375, 71.50788879394531] data_time: 0.0199 time: 1.2850 +2023/06/05 05:58:56 - mmengine - INFO - Epoch(train) [4][ 100/4111] lr: 8.1174e-05 eta: 5:06:23 time: 0.6307 data_time: 0.0019 memory: 44140 loss: 0.4695 +2023/06/05 05:59:59 - mmengine - INFO - Epoch(train) [4][ 200/4111] lr: 8.0894e-05 eta: 5:05:18 time: 0.6325 data_time: 0.0021 memory: 44140 loss: 0.4693 +2023/06/05 06:01:02 - mmengine - INFO - Epoch(train) [4][ 300/4111] lr: 8.0612e-05 eta: 5:04:11 time: 0.6320 data_time: 0.0018 memory: 44140 loss: 0.4634 +2023/06/05 06:02:05 - mmengine - INFO - Epoch(train) [4][ 400/4111] lr: 8.0328e-05 eta: 5:03:05 time: 0.6311 data_time: 0.0015 memory: 44140 loss: 0.4640 +2023/06/05 06:03:09 - mmengine - INFO - Epoch(train) [4][ 500/4111] lr: 8.0043e-05 eta: 5:01:59 time: 0.6314 data_time: 0.0019 memory: 44140 loss: 0.4621 +2023/06/05 06:04:12 - mmengine - INFO - Epoch(train) [4][ 600/4111] lr: 7.9757e-05 eta: 5:00:53 time: 0.6321 data_time: 0.0015 memory: 44140 loss: 0.4618 +2023/06/05 06:04:54 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_all_lr1e-4_20230605_033556 +2023/06/05 06:05:15 - mmengine - INFO - Epoch(train) [4][ 700/4111] lr: 7.9469e-05 eta: 4:59:48 time: 0.6372 data_time: 0.0017 memory: 44140 loss: 0.4666 +2023/06/05 06:06:18 - mmengine - INFO - Epoch(train) [4][ 800/4111] lr: 7.9180e-05 eta: 4:58:42 time: 0.6319 data_time: 0.0018 memory: 44140 loss: 0.4522 +2023/06/05 06:07:22 - mmengine - INFO - Epoch(train) [4][ 900/4111] lr: 7.8889e-05 eta: 4:57:36 time: 0.6310 data_time: 0.0016 memory: 44140 loss: 0.5025 +2023/06/05 06:08:25 - mmengine - INFO - Epoch(train) [4][1000/4111] lr: 7.8597e-05 eta: 4:56:30 time: 0.6312 data_time: 0.0017 memory: 44140 loss: 0.4567 +2023/06/05 06:09:28 - mmengine - INFO - Epoch(train) [4][1100/4111] lr: 7.8303e-05 eta: 4:55:24 time: 0.6358 data_time: 0.0017 memory: 44140 loss: 0.4541 +2023/06/05 06:10:31 - mmengine - INFO - Epoch(train) [4][1200/4111] lr: 7.8008e-05 eta: 4:54:18 time: 0.6329 data_time: 0.0016 memory: 44140 loss: 0.4546 +2023/06/05 06:11:34 - mmengine - INFO - Epoch(train) [4][1300/4111] lr: 7.7712e-05 eta: 4:53:13 time: 0.6312 data_time: 0.0018 memory: 44140 loss: 0.4589 +2023/06/05 06:12:37 - mmengine - INFO - Epoch(train) [4][1400/4111] lr: 7.7415e-05 eta: 4:52:07 time: 0.6301 data_time: 0.0015 memory: 44140 loss: 0.4681 +2023/06/05 06:13:41 - mmengine - INFO - Epoch(train) [4][1500/4111] lr: 7.7116e-05 eta: 4:51:01 time: 0.6308 data_time: 0.0015 memory: 44140 loss: 0.4672 +2023/06/05 06:14:44 - mmengine - INFO - Epoch(train) [4][1600/4111] lr: 7.6816e-05 eta: 4:49:55 time: 0.6310 data_time: 0.0015 memory: 44140 loss: 0.4781 +2023/06/05 06:15:26 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_all_lr1e-4_20230605_033556 +2023/06/05 06:15:47 - mmengine - INFO - Epoch(train) [4][1700/4111] lr: 7.6514e-05 eta: 4:48:50 time: 0.6314 data_time: 0.0015 memory: 44140 loss: 0.4668 +2023/06/05 06:16:50 - mmengine - INFO - Epoch(train) [4][1800/4111] lr: 7.6212e-05 eta: 4:47:44 time: 0.6320 data_time: 0.0016 memory: 44140 loss: 0.4456 +2023/06/05 06:17:54 - mmengine - INFO - Epoch(train) [4][1900/4111] lr: 7.5908e-05 eta: 4:46:39 time: 0.6321 data_time: 0.0016 memory: 44140 loss: 0.4570 +2023/06/05 06:18:57 - mmengine - INFO - Epoch(train) [4][2000/4111] lr: 7.5603e-05 eta: 4:45:34 time: 0.6325 data_time: 0.0017 memory: 44140 loss: 0.4478 +2023/06/05 06:20:00 - mmengine - INFO - Epoch(train) [4][2100/4111] lr: 7.5296e-05 eta: 4:44:29 time: 0.6317 data_time: 0.0017 memory: 44140 loss: 0.4909 +2023/06/05 06:21:03 - mmengine - INFO - Epoch(train) [4][2200/4111] lr: 7.4989e-05 eta: 4:43:23 time: 0.6344 data_time: 0.0017 memory: 44140 loss: 0.4687 +2023/06/05 06:22:07 - mmengine - INFO - Epoch(train) [4][2300/4111] lr: 7.4680e-05 eta: 4:42:18 time: 0.6328 data_time: 0.0016 memory: 44140 loss: 0.4528 +2023/06/05 06:23:10 - mmengine - INFO - Epoch(train) [4][2400/4111] lr: 7.4370e-05 eta: 4:41:13 time: 0.6316 data_time: 0.0020 memory: 44140 loss: 0.4712 +2023/06/05 06:24:13 - mmengine - INFO - Epoch(train) [4][2500/4111] lr: 7.4059e-05 eta: 4:40:07 time: 0.6318 data_time: 0.0015 memory: 44140 loss: 0.4663 +2023/06/05 06:25:16 - mmengine - INFO - Epoch(train) [4][2600/4111] lr: 7.3747e-05 eta: 4:39:02 time: 0.6321 data_time: 0.0020 memory: 44140 loss: 0.4693 +2023/06/05 06:25:59 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_all_lr1e-4_20230605_033556 +2023/06/05 06:26:19 - mmengine - INFO - Epoch(train) [4][2700/4111] lr: 7.3434e-05 eta: 4:37:57 time: 0.6336 data_time: 0.0015 memory: 44140 loss: 0.4635 +2023/06/05 06:27:23 - mmengine - INFO - Epoch(train) [4][2800/4111] lr: 7.3120e-05 eta: 4:36:51 time: 0.6328 data_time: 0.0017 memory: 44140 loss: 0.4653 +2023/06/05 06:28:26 - mmengine - INFO - Epoch(train) [4][2900/4111] lr: 7.2804e-05 eta: 4:35:46 time: 0.6314 data_time: 0.0017 memory: 44140 loss: 0.4721 +2023/06/05 06:29:29 - mmengine - INFO - Epoch(train) [4][3000/4111] lr: 7.2488e-05 eta: 4:34:41 time: 0.6308 data_time: 0.0015 memory: 44140 loss: 0.4541 +2023/06/05 06:30:32 - mmengine - INFO - Epoch(train) [4][3100/4111] lr: 7.2171e-05 eta: 4:33:36 time: 0.6315 data_time: 0.0018 memory: 44140 loss: 0.4760 +2023/06/05 06:31:35 - mmengine - INFO - Epoch(train) [4][3200/4111] lr: 7.1852e-05 eta: 4:32:31 time: 0.6315 data_time: 0.0015 memory: 44140 loss: 0.4790 +2023/06/05 06:32:39 - mmengine - INFO - Epoch(train) [4][3300/4111] lr: 7.1533e-05 eta: 4:31:26 time: 0.6327 data_time: 0.0015 memory: 44140 loss: 0.4583 +2023/06/05 06:33:42 - mmengine - INFO - Epoch(train) [4][3400/4111] lr: 7.1213e-05 eta: 4:30:20 time: 0.6313 data_time: 0.0015 memory: 44140 loss: 0.4747 +2023/06/05 06:34:45 - mmengine - INFO - Epoch(train) [4][3500/4111] lr: 7.0891e-05 eta: 4:29:15 time: 0.6313 data_time: 0.0023 memory: 44140 loss: 0.4505 +2023/06/05 06:35:48 - mmengine - INFO - Epoch(train) [4][3600/4111] lr: 7.0569e-05 eta: 4:28:10 time: 0.6306 data_time: 0.0015 memory: 44140 loss: 0.4631 +2023/06/05 06:36:31 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_all_lr1e-4_20230605_033556 +2023/06/05 06:36:51 - mmengine - INFO - Epoch(train) [4][3700/4111] lr: 7.0246e-05 eta: 4:27:05 time: 0.6329 data_time: 0.0022 memory: 44140 loss: 0.4630 +2023/06/05 06:37:55 - mmengine - INFO - Epoch(train) [4][3800/4111] lr: 6.9922e-05 eta: 4:26:00 time: 0.6336 data_time: 0.0024 memory: 44140 loss: 0.4596 +2023/06/05 06:38:58 - mmengine - INFO - Epoch(train) [4][3900/4111] lr: 6.9597e-05 eta: 4:24:55 time: 0.6322 data_time: 0.0019 memory: 44140 loss: 0.4482 +2023/06/05 06:40:01 - mmengine - INFO - Epoch(train) [4][4000/4111] lr: 6.9272e-05 eta: 4:23:51 time: 0.6304 data_time: 0.0019 memory: 44140 loss: 0.4800 +2023/06/05 06:41:05 - mmengine - INFO - Epoch(train) [4][4100/4111] lr: 6.8945e-05 eta: 4:22:46 time: 0.6325 data_time: 0.0030 memory: 44140 loss: 0.4720 +2023/06/05 06:41:11 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_all_lr1e-4_20230605_033556 +2023/06/05 06:41:11 - mmengine - INFO - Saving checkpoint at 4 epochs +2023/06/05 06:43:33 - mmengine - INFO - Epoch(val) [4][100/119] eta: 0:00:24 time: 1.2827 data_time: 0.0008 memory: 44140 +2023/06/05 06:44:17 - mmengine - INFO - Epoch(val) [4][119/119] accuracy/top1: 87.5246 single-label/precision_classwise: [94.1501235961914, 67.70805358886719] single-label/recall_classwise: [89.71234893798828, 79.46511840820312] single-label/f1-score_classwise: [91.87767791748047, 73.11697387695312] data_time: 0.0185 time: 1.2800 +2023/06/05 06:45:24 - mmengine - INFO - Epoch(train) [5][ 100/4111] lr: 6.8582e-05 eta: 4:21:39 time: 0.6323 data_time: 0.0017 memory: 44140 loss: 0.4378 +2023/06/05 06:46:27 - mmengine - INFO - Epoch(train) [5][ 200/4111] lr: 6.8253e-05 eta: 4:20:34 time: 0.6316 data_time: 0.0021 memory: 44140 loss: 0.4512 +2023/06/05 06:47:30 - mmengine - INFO - Epoch(train) [5][ 300/4111] lr: 6.7924e-05 eta: 4:19:29 time: 0.6308 data_time: 0.0016 memory: 44140 loss: 0.4462 +2023/06/05 06:48:34 - mmengine - INFO - Epoch(train) [5][ 400/4111] lr: 6.7595e-05 eta: 4:18:24 time: 0.6321 data_time: 0.0020 memory: 44140 loss: 0.4669 +2023/06/05 06:49:37 - mmengine - INFO - Epoch(train) [5][ 500/4111] lr: 6.7264e-05 eta: 4:17:19 time: 0.6316 data_time: 0.0018 memory: 44140 loss: 0.4677 +2023/06/05 06:50:12 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_all_lr1e-4_20230605_033556 +2023/06/05 06:50:40 - mmengine - INFO - Epoch(train) [5][ 600/4111] lr: 6.6933e-05 eta: 4:16:14 time: 0.6323 data_time: 0.0021 memory: 44140 loss: 0.4624 +2023/06/05 06:51:43 - mmengine - INFO - Epoch(train) [5][ 700/4111] lr: 6.6601e-05 eta: 4:15:09 time: 0.6343 data_time: 0.0015 memory: 44140 loss: 0.4679 +2023/06/05 06:52:47 - mmengine - INFO - Epoch(train) [5][ 800/4111] lr: 6.6268e-05 eta: 4:14:05 time: 0.6380 data_time: 0.0026 memory: 44140 loss: 0.4265 +2023/06/05 06:53:50 - mmengine - INFO - Epoch(train) [5][ 900/4111] lr: 6.5935e-05 eta: 4:13:00 time: 0.6318 data_time: 0.0015 memory: 44140 loss: 0.4788 +2023/06/05 06:54:54 - mmengine - INFO - Epoch(train) [5][1000/4111] lr: 6.5601e-05 eta: 4:11:56 time: 0.6331 data_time: 0.0019 memory: 44140 loss: 0.4344 +2023/06/05 06:55:57 - mmengine - INFO - Epoch(train) [5][1100/4111] lr: 6.5267e-05 eta: 4:10:51 time: 0.6322 data_time: 0.0015 memory: 44140 loss: 0.4513 +2023/06/05 06:57:00 - mmengine - INFO - Epoch(train) [5][1200/4111] lr: 6.4932e-05 eta: 4:09:46 time: 0.6336 data_time: 0.0018 memory: 44140 loss: 0.4581 +2023/06/05 06:58:08 - mmengine - INFO - Epoch(train) [5][1300/4111] lr: 6.4596e-05 eta: 4:08:48 time: 0.6305 data_time: 0.0016 memory: 44140 loss: 0.4757 +2023/06/05 06:59:31 - mmengine - INFO - Epoch(train) [5][1400/4111] lr: 6.4260e-05 eta: 4:08:08 time: 0.6313 data_time: 0.0021 memory: 44140 loss: 0.4651 +2023/06/05 07:00:34 - mmengine - INFO - Epoch(train) [5][1500/4111] lr: 6.3923e-05 eta: 4:07:04 time: 0.6321 data_time: 0.0016 memory: 44140 loss: 0.4710 +2023/06/05 07:01:10 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_all_lr1e-4_20230605_033556 +2023/06/05 07:01:38 - mmengine - INFO - Epoch(train) [5][1600/4111] lr: 6.3585e-05 eta: 4:05:59 time: 0.6340 data_time: 0.0026 memory: 44140 loss: 0.4611 +2023/06/05 07:02:41 - mmengine - INFO - Epoch(train) [5][1700/4111] lr: 6.3248e-05 eta: 4:04:54 time: 0.6309 data_time: 0.0016 memory: 44140 loss: 0.4631 +2023/06/05 07:03:44 - mmengine - INFO - Epoch(train) [5][1800/4111] lr: 6.2909e-05 eta: 4:03:49 time: 0.6311 data_time: 0.0020 memory: 44140 loss: 0.4634 +2023/06/05 07:04:48 - mmengine - INFO - Epoch(train) [5][1900/4111] lr: 6.2571e-05 eta: 4:02:44 time: 0.6392 data_time: 0.0015 memory: 44140 loss: 0.4621 +2023/06/05 07:05:51 - mmengine - INFO - Epoch(train) [5][2000/4111] lr: 6.2231e-05 eta: 4:01:40 time: 0.6317 data_time: 0.0020 memory: 44140 loss: 0.4732 +2023/06/05 07:06:54 - mmengine - INFO - Epoch(train) [5][2100/4111] lr: 6.1892e-05 eta: 4:00:35 time: 0.6308 data_time: 0.0016 memory: 44140 loss: 0.4688 +2023/06/05 07:07:57 - mmengine - INFO - Epoch(train) [5][2200/4111] lr: 6.1552e-05 eta: 3:59:30 time: 0.6324 data_time: 0.0022 memory: 44140 loss: 0.4631 +2023/06/05 07:09:01 - mmengine - INFO - Epoch(train) [5][2300/4111] lr: 6.1211e-05 eta: 3:58:25 time: 0.6324 data_time: 0.0016 memory: 44140 loss: 0.4457 +2023/06/05 07:10:04 - mmengine - INFO - Epoch(train) [5][2400/4111] lr: 6.0871e-05 eta: 3:57:20 time: 0.6335 data_time: 0.0019 memory: 44140 loss: 0.4653 +2023/06/05 07:11:07 - mmengine - INFO - Epoch(train) [5][2500/4111] lr: 6.0529e-05 eta: 3:56:16 time: 0.6317 data_time: 0.0015 memory: 44140 loss: 0.4579 +2023/06/05 07:11:43 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_all_lr1e-4_20230605_033556 +2023/06/05 07:12:11 - mmengine - INFO - Epoch(train) [5][2600/4111] lr: 6.0188e-05 eta: 3:55:11 time: 0.6443 data_time: 0.0017 memory: 44140 loss: 0.4558 +2023/06/05 07:13:14 - mmengine - INFO - Epoch(train) [5][2700/4111] lr: 5.9846e-05 eta: 3:54:06 time: 0.6317 data_time: 0.0018 memory: 44140 loss: 0.4572 +2023/06/05 07:14:17 - mmengine - INFO - Epoch(train) [5][2800/4111] lr: 5.9504e-05 eta: 3:53:02 time: 0.6310 data_time: 0.0016 memory: 44140 loss: 0.4678 +2023/06/05 07:15:21 - mmengine - INFO - Epoch(train) [5][2900/4111] lr: 5.9162e-05 eta: 3:51:57 time: 0.6312 data_time: 0.0016 memory: 44140 loss: 0.4622 +2023/06/05 07:16:24 - mmengine - INFO - Epoch(train) [5][3000/4111] lr: 5.8819e-05 eta: 3:50:52 time: 0.6315 data_time: 0.0015 memory: 44140 loss: 0.4836 +2023/06/05 07:17:27 - mmengine - INFO - Epoch(train) [5][3100/4111] lr: 5.8477e-05 eta: 3:49:48 time: 0.6316 data_time: 0.0018 memory: 44140 loss: 0.4748 +2023/06/05 07:18:30 - mmengine - INFO - Epoch(train) [5][3200/4111] lr: 5.8134e-05 eta: 3:48:43 time: 0.6309 data_time: 0.0016 memory: 44140 loss: 0.4699 +2023/06/05 07:19:33 - mmengine - INFO - Epoch(train) [5][3300/4111] lr: 5.7791e-05 eta: 3:47:38 time: 0.6298 data_time: 0.0016 memory: 44140 loss: 0.4809 +2023/06/05 07:20:37 - mmengine - INFO - Epoch(train) [5][3400/4111] lr: 5.7447e-05 eta: 3:46:33 time: 0.6310 data_time: 0.0015 memory: 44140 loss: 0.4683 +2023/06/05 07:21:40 - mmengine - INFO - Epoch(train) [5][3500/4111] lr: 5.7104e-05 eta: 3:45:29 time: 0.6319 data_time: 0.0015 memory: 44140 loss: 0.4481 +2023/06/05 07:22:15 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_all_lr1e-4_20230605_033556 +2023/06/05 07:22:43 - mmengine - INFO - Epoch(train) [5][3600/4111] lr: 5.6760e-05 eta: 3:44:24 time: 0.6312 data_time: 0.0020 memory: 44140 loss: 0.4793 +2023/06/05 07:23:46 - mmengine - INFO - Epoch(train) [5][3700/4111] lr: 5.6417e-05 eta: 3:43:19 time: 0.6306 data_time: 0.0016 memory: 44140 loss: 0.4709 +2023/06/05 07:24:49 - mmengine - INFO - Epoch(train) [5][3800/4111] lr: 5.6073e-05 eta: 3:42:15 time: 0.6302 data_time: 0.0016 memory: 44140 loss: 0.4650 +2023/06/05 07:25:53 - mmengine - INFO - Epoch(train) [5][3900/4111] lr: 5.5729e-05 eta: 3:41:10 time: 0.6335 data_time: 0.0017 memory: 44140 loss: 0.4517 +2023/06/05 07:26:56 - mmengine - INFO - Epoch(train) [5][4000/4111] lr: 5.5385e-05 eta: 3:40:05 time: 0.6305 data_time: 0.0015 memory: 44140 loss: 0.4578 +2023/06/05 07:27:59 - mmengine - INFO - Epoch(train) [5][4100/4111] lr: 5.5041e-05 eta: 3:39:01 time: 0.6329 data_time: 0.0031 memory: 44140 loss: 0.4562 +2023/06/05 07:28:06 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_all_lr1e-4_20230605_033556 +2023/06/05 07:28:06 - mmengine - INFO - Saving checkpoint at 5 epochs +2023/06/05 07:30:28 - mmengine - INFO - Epoch(val) [5][100/119] eta: 0:00:24 time: 1.2827 data_time: 0.0006 memory: 44140 +2023/06/05 07:31:11 - mmengine - INFO - Epoch(val) [5][119/119] accuracy/top1: 88.2909 single-label/precision_classwise: [93.96712493896484, 70.19063568115234] single-label/recall_classwise: [90.95174407958984, 78.48837280273438] single-label/f1-score_classwise: [92.43484497070312, 74.10795593261719] data_time: 0.0194 time: 1.2807 +2023/06/05 07:32:18 - mmengine - INFO - Epoch(train) [6][ 100/4111] lr: 5.4660e-05 eta: 3:37:53 time: 0.6332 data_time: 0.0018 memory: 44140 loss: 0.4751 +2023/06/05 07:33:21 - mmengine - INFO - Epoch(train) [6][ 200/4111] lr: 5.4316e-05 eta: 3:36:48 time: 0.6334 data_time: 0.0017 memory: 44140 loss: 0.4887 +2023/06/05 07:34:24 - mmengine - INFO - Epoch(train) [6][ 300/4111] lr: 5.3972e-05 eta: 3:35:44 time: 0.6323 data_time: 0.0015 memory: 44140 loss: 0.4472 +2023/06/05 07:35:27 - mmengine - INFO - Epoch(train) [6][ 400/4111] lr: 5.3628e-05 eta: 3:34:39 time: 0.6310 data_time: 0.0017 memory: 44140 loss: 0.4418 +2023/06/05 07:35:56 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_all_lr1e-4_20230605_033556 +2023/06/05 07:36:31 - mmengine - INFO - Epoch(train) [6][ 500/4111] lr: 5.3284e-05 eta: 3:33:34 time: 0.6308 data_time: 0.0017 memory: 44140 loss: 0.4762 +2023/06/05 07:37:34 - mmengine - INFO - Epoch(train) [6][ 600/4111] lr: 5.2941e-05 eta: 3:32:30 time: 0.6302 data_time: 0.0018 memory: 44140 loss: 0.4623 +2023/06/05 07:38:37 - mmengine - INFO - Epoch(train) [6][ 700/4111] lr: 5.2597e-05 eta: 3:31:25 time: 0.6306 data_time: 0.0018 memory: 44140 loss: 0.4882 +2023/06/05 07:39:40 - mmengine - INFO - Epoch(train) [6][ 800/4111] lr: 5.2254e-05 eta: 3:30:21 time: 0.6331 data_time: 0.0016 memory: 44140 loss: 0.4594 +2023/06/05 07:40:44 - mmengine - INFO - Epoch(train) [6][ 900/4111] lr: 5.1911e-05 eta: 3:29:16 time: 0.6330 data_time: 0.0017 memory: 44140 loss: 0.4718 +2023/06/05 07:41:47 - mmengine - INFO - Epoch(train) [6][1000/4111] lr: 5.1568e-05 eta: 3:28:12 time: 0.6316 data_time: 0.0018 memory: 44140 loss: 0.4768 +2023/06/05 07:42:50 - mmengine - INFO - Epoch(train) [6][1100/4111] lr: 5.1225e-05 eta: 3:27:07 time: 0.6315 data_time: 0.0015 memory: 44140 loss: 0.4818 +2023/06/05 07:43:53 - mmengine - INFO - Epoch(train) [6][1200/4111] lr: 5.0883e-05 eta: 3:26:03 time: 0.6333 data_time: 0.0016 memory: 44140 loss: 0.4633 +2023/06/05 07:44:57 - mmengine - INFO - Epoch(train) [6][1300/4111] lr: 5.0540e-05 eta: 3:24:59 time: 0.6307 data_time: 0.0016 memory: 44140 loss: 0.4573 +2023/06/05 07:46:00 - mmengine - INFO - Epoch(train) [6][1400/4111] lr: 5.0198e-05 eta: 3:23:54 time: 0.6345 data_time: 0.0015 memory: 44140 loss: 0.4627 +2023/06/05 07:46:28 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_all_lr1e-4_20230605_033556 +2023/06/05 07:47:03 - mmengine - INFO - Epoch(train) [6][1500/4111] lr: 4.9856e-05 eta: 3:22:50 time: 0.6313 data_time: 0.0018 memory: 44140 loss: 0.4584 +2023/06/05 07:48:06 - mmengine - INFO - Epoch(train) [6][1600/4111] lr: 4.9515e-05 eta: 3:21:45 time: 0.6315 data_time: 0.0016 memory: 44140 loss: 0.4805 +2023/06/05 07:49:10 - mmengine - INFO - Epoch(train) [6][1700/4111] lr: 4.9174e-05 eta: 3:20:41 time: 0.6355 data_time: 0.0015 memory: 44140 loss: 0.4729 +2023/06/05 07:50:13 - mmengine - INFO - Epoch(train) [6][1800/4111] lr: 4.8833e-05 eta: 3:19:36 time: 0.6312 data_time: 0.0015 memory: 44140 loss: 0.4489 +2023/06/05 07:51:16 - mmengine - INFO - Epoch(train) [6][1900/4111] lr: 4.8492e-05 eta: 3:18:32 time: 0.6304 data_time: 0.0016 memory: 44140 loss: 0.4645 +2023/06/05 07:52:19 - mmengine - INFO - Epoch(train) [6][2000/4111] lr: 4.8152e-05 eta: 3:17:28 time: 0.6325 data_time: 0.0015 memory: 44140 loss: 0.4664 +2023/06/05 07:53:23 - mmengine - INFO - Epoch(train) [6][2100/4111] lr: 4.7813e-05 eta: 3:16:24 time: 0.6334 data_time: 0.0017 memory: 44140 loss: 0.4645 +2023/06/05 07:54:26 - mmengine - INFO - Epoch(train) [6][2200/4111] lr: 4.7473e-05 eta: 3:15:19 time: 0.6342 data_time: 0.0015 memory: 44140 loss: 0.4704 +2023/06/05 07:55:30 - mmengine - INFO - Epoch(train) [6][2300/4111] lr: 4.7135e-05 eta: 3:14:15 time: 0.6463 data_time: 0.0018 memory: 44140 loss: 0.4463 +2023/06/05 07:56:33 - mmengine - INFO - Epoch(train) [6][2400/4111] lr: 4.6796e-05 eta: 3:13:11 time: 0.6326 data_time: 0.0016 memory: 44140 loss: 0.4846 +2023/06/05 07:57:02 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_all_lr1e-4_20230605_033556 +2023/06/05 07:57:36 - mmengine - INFO - Epoch(train) [6][2500/4111] lr: 4.6458e-05 eta: 3:12:07 time: 0.6323 data_time: 0.0017 memory: 44140 loss: 0.4898 +2023/06/05 07:58:40 - mmengine - INFO - Epoch(train) [6][2600/4111] lr: 4.6121e-05 eta: 3:11:02 time: 0.6312 data_time: 0.0015 memory: 44140 loss: 0.4826 +2023/06/05 07:59:43 - mmengine - INFO - Epoch(train) [6][2700/4111] lr: 4.5784e-05 eta: 3:09:58 time: 0.6318 data_time: 0.0016 memory: 44140 loss: 0.4759 +2023/06/05 08:00:46 - mmengine - INFO - Epoch(train) [6][2800/4111] lr: 4.5448e-05 eta: 3:08:54 time: 0.6339 data_time: 0.0014 memory: 44140 loss: 0.4654 +2023/06/05 08:01:49 - mmengine - INFO - Epoch(train) [6][2900/4111] lr: 4.5112e-05 eta: 3:07:50 time: 0.6313 data_time: 0.0017 memory: 44140 loss: 0.4575 +2023/06/05 08:02:53 - mmengine - INFO - Epoch(train) [6][3000/4111] lr: 4.4777e-05 eta: 3:06:45 time: 0.6351 data_time: 0.0025 memory: 44140 loss: 0.4551 +2023/06/05 08:03:56 - mmengine - INFO - Epoch(train) [6][3100/4111] lr: 4.4442e-05 eta: 3:05:41 time: 0.6322 data_time: 0.0022 memory: 44140 loss: 0.4826 +2023/06/05 08:04:59 - mmengine - INFO - Epoch(train) [6][3200/4111] lr: 4.4108e-05 eta: 3:04:37 time: 0.6314 data_time: 0.0020 memory: 44140 loss: 0.4163 +2023/06/05 08:06:03 - mmengine - INFO - Epoch(train) [6][3300/4111] lr: 4.3775e-05 eta: 3:03:33 time: 0.6315 data_time: 0.0017 memory: 44140 loss: 0.4384 +2023/06/05 08:07:06 - mmengine - INFO - Epoch(train) [6][3400/4111] lr: 4.3442e-05 eta: 3:02:29 time: 0.6403 data_time: 0.0016 memory: 44140 loss: 0.4627 +2023/06/05 08:07:35 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_all_lr1e-4_20230605_033556 +2023/06/05 08:08:09 - mmengine - INFO - Epoch(train) [6][3500/4111] lr: 4.3110e-05 eta: 3:01:24 time: 0.6343 data_time: 0.0016 memory: 44140 loss: 0.4683 +2023/06/05 08:09:13 - mmengine - INFO - Epoch(train) [6][3600/4111] lr: 4.2779e-05 eta: 3:00:20 time: 0.6314 data_time: 0.0023 memory: 44140 loss: 0.4529 +2023/06/05 08:10:16 - mmengine - INFO - Epoch(train) [6][3700/4111] lr: 4.2448e-05 eta: 2:59:16 time: 0.6338 data_time: 0.0014 memory: 44140 loss: 0.4547 +2023/06/05 08:11:19 - mmengine - INFO - Epoch(train) [6][3800/4111] lr: 4.2119e-05 eta: 2:58:12 time: 0.6323 data_time: 0.0018 memory: 44140 loss: 0.4563 +2023/06/05 08:12:23 - mmengine - INFO - Epoch(train) [6][3900/4111] lr: 4.1789e-05 eta: 2:57:08 time: 0.6339 data_time: 0.0018 memory: 44140 loss: 0.4523 +2023/06/05 08:13:26 - mmengine - INFO - Epoch(train) [6][4000/4111] lr: 4.1461e-05 eta: 2:56:04 time: 0.6348 data_time: 0.0016 memory: 44140 loss: 0.4423 +2023/06/05 08:14:29 - mmengine - INFO - Epoch(train) [6][4100/4111] lr: 4.1133e-05 eta: 2:54:59 time: 0.6370 data_time: 0.0030 memory: 44140 loss: 0.4573 +2023/06/05 08:14:36 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_all_lr1e-4_20230605_033556 +2023/06/05 08:14:36 - mmengine - INFO - Saving checkpoint at 6 epochs +2023/06/05 08:16:59 - mmengine - INFO - Epoch(val) [6][100/119] eta: 0:00:24 time: 1.2822 data_time: 0.0008 memory: 44140 +2023/06/05 08:17:42 - mmengine - INFO - Epoch(val) [6][119/119] accuracy/top1: 88.7824 single-label/precision_classwise: [93.91098022460938, 71.82064819335938] single-label/recall_classwise: [91.68192291259766, 78.10077667236328] single-label/f1-score_classwise: [92.78307342529297, 74.82918548583984] data_time: 0.0224 time: 1.2851 +2023/06/05 08:18:49 - mmengine - INFO - Epoch(train) [7][ 100/4111] lr: 4.0771e-05 eta: 2:53:50 time: 0.6327 data_time: 0.0018 memory: 44140 loss: 0.4571 +2023/06/05 08:19:52 - mmengine - INFO - Epoch(train) [7][ 200/4111] lr: 4.0445e-05 eta: 2:52:46 time: 0.6322 data_time: 0.0018 memory: 44140 loss: 0.4359 +2023/06/05 08:20:55 - mmengine - INFO - Epoch(train) [7][ 300/4111] lr: 4.0120e-05 eta: 2:51:42 time: 0.6320 data_time: 0.0015 memory: 44140 loss: 0.4664 +2023/06/05 08:21:17 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_all_lr1e-4_20230605_033556 +2023/06/05 08:21:59 - mmengine - INFO - Epoch(train) [7][ 400/4111] lr: 3.9796e-05 eta: 2:50:38 time: 0.6334 data_time: 0.0017 memory: 44140 loss: 0.4723 +2023/06/05 08:23:02 - mmengine - INFO - Epoch(train) [7][ 500/4111] lr: 3.9473e-05 eta: 2:49:34 time: 0.6347 data_time: 0.0017 memory: 44140 loss: 0.4698 +2023/06/05 08:24:05 - mmengine - INFO - Epoch(train) [7][ 600/4111] lr: 3.9150e-05 eta: 2:48:30 time: 0.6324 data_time: 0.0015 memory: 44140 loss: 0.4644 +2023/06/05 08:25:09 - mmengine - INFO - Epoch(train) [7][ 700/4111] lr: 3.8829e-05 eta: 2:47:26 time: 0.6323 data_time: 0.0020 memory: 44140 loss: 0.4459 +2023/06/05 08:26:12 - mmengine - INFO - Epoch(train) [7][ 800/4111] lr: 3.8509e-05 eta: 2:46:21 time: 0.6324 data_time: 0.0020 memory: 44140 loss: 0.4611 +2023/06/05 08:27:15 - mmengine - INFO - Epoch(train) [7][ 900/4111] lr: 3.8189e-05 eta: 2:45:17 time: 0.6340 data_time: 0.0017 memory: 44140 loss: 0.4709 +2023/06/05 08:28:19 - mmengine - INFO - Epoch(train) [7][1000/4111] lr: 3.7871e-05 eta: 2:44:13 time: 0.6319 data_time: 0.0015 memory: 44140 loss: 0.4548 +2023/06/05 08:29:22 - mmengine - INFO - Epoch(train) [7][1100/4111] lr: 3.7553e-05 eta: 2:43:09 time: 0.6330 data_time: 0.0024 memory: 44140 loss: 0.4800 +2023/06/05 08:30:25 - mmengine - INFO - Epoch(train) [7][1200/4111] lr: 3.7237e-05 eta: 2:42:05 time: 0.6328 data_time: 0.0019 memory: 44140 loss: 0.4594 +2023/06/05 08:31:29 - mmengine - INFO - Epoch(train) [7][1300/4111] lr: 3.6921e-05 eta: 2:41:01 time: 0.6355 data_time: 0.0022 memory: 44140 loss: 0.4604 +2023/06/05 08:31:50 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_all_lr1e-4_20230605_033556 +2023/06/05 08:32:32 - mmengine - INFO - Epoch(train) [7][1400/4111] lr: 3.6607e-05 eta: 2:39:57 time: 0.6340 data_time: 0.0019 memory: 44140 loss: 0.4572 +2023/06/05 08:33:36 - mmengine - INFO - Epoch(train) [7][1500/4111] lr: 3.6294e-05 eta: 2:38:53 time: 0.6331 data_time: 0.0021 memory: 44140 loss: 0.4525 +2023/06/05 08:34:39 - mmengine - INFO - Epoch(train) [7][1600/4111] lr: 3.5981e-05 eta: 2:37:49 time: 0.6327 data_time: 0.0017 memory: 44140 loss: 0.4527 +2023/06/05 08:35:42 - mmengine - INFO - Epoch(train) [7][1700/4111] lr: 3.5670e-05 eta: 2:36:45 time: 0.6316 data_time: 0.0017 memory: 44140 loss: 0.5037 +2023/06/05 08:36:46 - mmengine - INFO - Epoch(train) [7][1800/4111] lr: 3.5360e-05 eta: 2:35:41 time: 0.6320 data_time: 0.0017 memory: 44140 loss: 0.4386 +2023/06/05 08:37:49 - mmengine - INFO - Epoch(train) [7][1900/4111] lr: 3.5051e-05 eta: 2:34:37 time: 0.6332 data_time: 0.0016 memory: 44140 loss: 0.4808 +2023/06/05 08:38:52 - mmengine - INFO - Epoch(train) [7][2000/4111] lr: 3.4744e-05 eta: 2:33:33 time: 0.6322 data_time: 0.0015 memory: 44140 loss: 0.4748 +2023/06/05 08:39:56 - mmengine - INFO - Epoch(train) [7][2100/4111] lr: 3.4437e-05 eta: 2:32:29 time: 0.6352 data_time: 0.0017 memory: 44140 loss: 0.4580 +2023/06/05 08:40:59 - mmengine - INFO - Epoch(train) [7][2200/4111] lr: 3.4132e-05 eta: 2:31:25 time: 0.6322 data_time: 0.0017 memory: 44140 loss: 0.4890 +2023/06/05 08:42:02 - mmengine - INFO - Epoch(train) [7][2300/4111] lr: 3.3828e-05 eta: 2:30:21 time: 0.6335 data_time: 0.0016 memory: 44140 loss: 0.4507 +2023/06/05 08:42:24 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_all_lr1e-4_20230605_033556 +2023/06/05 08:43:06 - mmengine - INFO - Epoch(train) [7][2400/4111] lr: 3.3525e-05 eta: 2:29:17 time: 0.6329 data_time: 0.0022 memory: 44140 loss: 0.4726 +2023/06/05 08:44:09 - mmengine - INFO - Epoch(train) [7][2500/4111] lr: 3.3224e-05 eta: 2:28:13 time: 0.6353 data_time: 0.0019 memory: 44140 loss: 0.4637 +2023/06/05 08:45:12 - mmengine - INFO - Epoch(train) [7][2600/4111] lr: 3.2923e-05 eta: 2:27:09 time: 0.6331 data_time: 0.0017 memory: 44140 loss: 0.4721 +2023/06/05 08:46:16 - mmengine - INFO - Epoch(train) [7][2700/4111] lr: 3.2624e-05 eta: 2:26:05 time: 0.6318 data_time: 0.0016 memory: 44140 loss: 0.4529 +2023/06/05 08:47:19 - mmengine - INFO - Epoch(train) [7][2800/4111] lr: 3.2326e-05 eta: 2:25:01 time: 0.6314 data_time: 0.0017 memory: 44140 loss: 0.4894 +2023/06/05 08:48:22 - mmengine - INFO - Epoch(train) [7][2900/4111] lr: 3.2030e-05 eta: 2:23:57 time: 0.6315 data_time: 0.0015 memory: 44140 loss: 0.4672 +2023/06/05 08:49:26 - mmengine - INFO - Epoch(train) [7][3000/4111] lr: 3.1735e-05 eta: 2:22:53 time: 0.6330 data_time: 0.0019 memory: 44140 loss: 0.4842 +2023/06/05 08:50:29 - mmengine - INFO - Epoch(train) [7][3100/4111] lr: 3.1441e-05 eta: 2:21:49 time: 0.6333 data_time: 0.0018 memory: 44140 loss: 0.4702 +2023/06/05 08:51:32 - mmengine - INFO - Epoch(train) [7][3200/4111] lr: 3.1149e-05 eta: 2:20:45 time: 0.6362 data_time: 0.0015 memory: 44140 loss: 0.4660 +2023/06/05 08:52:35 - mmengine - INFO - Epoch(train) [7][3300/4111] lr: 3.0858e-05 eta: 2:19:41 time: 0.6332 data_time: 0.0020 memory: 44140 loss: 0.4447 +2023/06/05 08:52:57 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_all_lr1e-4_20230605_033556 +2023/06/05 08:53:39 - mmengine - INFO - Epoch(train) [7][3400/4111] lr: 3.0569e-05 eta: 2:18:37 time: 0.6325 data_time: 0.0017 memory: 44140 loss: 0.4772 +2023/06/05 08:54:42 - mmengine - INFO - Epoch(train) [7][3500/4111] lr: 3.0281e-05 eta: 2:17:33 time: 0.6322 data_time: 0.0018 memory: 44140 loss: 0.4586 +2023/06/05 08:55:45 - mmengine - INFO - Epoch(train) [7][3600/4111] lr: 2.9994e-05 eta: 2:16:29 time: 0.6342 data_time: 0.0017 memory: 44140 loss: 0.4595 +2023/06/05 08:56:49 - mmengine - INFO - Epoch(train) [7][3700/4111] lr: 2.9709e-05 eta: 2:15:25 time: 0.6326 data_time: 0.0014 memory: 44140 loss: 0.5164 +2023/06/05 08:57:52 - mmengine - INFO - Epoch(train) [7][3800/4111] lr: 2.9425e-05 eta: 2:14:21 time: 0.6327 data_time: 0.0019 memory: 44140 loss: 0.4485 +2023/06/05 08:58:56 - mmengine - INFO - Epoch(train) [7][3900/4111] lr: 2.9143e-05 eta: 2:13:17 time: 0.6354 data_time: 0.0014 memory: 44140 loss: 0.4667 +2023/06/05 08:59:59 - mmengine - INFO - Epoch(train) [7][4000/4111] lr: 2.8862e-05 eta: 2:12:13 time: 0.6330 data_time: 0.0017 memory: 44140 loss: 0.4680 +2023/06/05 09:01:03 - mmengine - INFO - Epoch(train) [7][4100/4111] lr: 2.8583e-05 eta: 2:11:09 time: 0.6388 data_time: 0.0030 memory: 44140 loss: 0.4719 +2023/06/05 09:01:10 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_all_lr1e-4_20230605_033556 +2023/06/05 09:01:10 - mmengine - INFO - Saving checkpoint at 7 epochs +2023/06/05 09:03:32 - mmengine - INFO - Epoch(val) [7][100/119] eta: 0:00:24 time: 1.2835 data_time: 0.0010 memory: 44140 +2023/06/05 09:04:17 - mmengine - INFO - Epoch(val) [7][119/119] accuracy/top1: 88.9446 single-label/precision_classwise: [93.77130126953125, 72.58860778808594] single-label/recall_classwise: [92.05857849121094, 77.47286224365234] single-label/f1-score_classwise: [92.90705108642578, 74.95124816894531] data_time: 0.0198 time: 1.2823 +2023/06/05 09:05:23 - mmengine - INFO - Epoch(train) [8][ 100/4111] lr: 2.8275e-05 eta: 2:10:00 time: 0.6312 data_time: 0.0020 memory: 44140 loss: 0.4787 +2023/06/05 09:06:27 - mmengine - INFO - Epoch(train) [8][ 200/4111] lr: 2.7999e-05 eta: 2:08:56 time: 0.6324 data_time: 0.0015 memory: 44140 loss: 0.4444 +2023/06/05 09:06:41 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_all_lr1e-4_20230605_033556 +2023/06/05 09:07:30 - mmengine - INFO - Epoch(train) [8][ 300/4111] lr: 2.7725e-05 eta: 2:07:52 time: 0.6328 data_time: 0.0018 memory: 44140 loss: 0.4804 +2023/06/05 09:08:33 - mmengine - INFO - Epoch(train) [8][ 400/4111] lr: 2.7452e-05 eta: 2:06:48 time: 0.6335 data_time: 0.0016 memory: 44140 loss: 0.4573 +2023/06/05 09:09:37 - mmengine - INFO - Epoch(train) [8][ 500/4111] lr: 2.7181e-05 eta: 2:05:44 time: 0.6325 data_time: 0.0016 memory: 44140 loss: 0.4263 +2023/06/05 09:10:40 - mmengine - INFO - Epoch(train) [8][ 600/4111] lr: 2.6911e-05 eta: 2:04:40 time: 0.6335 data_time: 0.0016 memory: 44140 loss: 0.4639 +2023/06/05 09:11:43 - mmengine - INFO - Epoch(train) [8][ 700/4111] lr: 2.6644e-05 eta: 2:03:36 time: 0.6320 data_time: 0.0015 memory: 44140 loss: 0.4891 +2023/06/05 09:12:47 - mmengine - INFO - Epoch(train) [8][ 800/4111] lr: 2.6377e-05 eta: 2:02:32 time: 0.6318 data_time: 0.0017 memory: 44140 loss: 0.4580 +2023/06/05 09:13:50 - mmengine - INFO - Epoch(train) [8][ 900/4111] lr: 2.6113e-05 eta: 2:01:28 time: 0.6310 data_time: 0.0017 memory: 44140 loss: 0.4669 +2023/06/05 09:14:53 - mmengine - INFO - Epoch(train) [8][1000/4111] lr: 2.5850e-05 eta: 2:00:24 time: 0.6320 data_time: 0.0017 memory: 44140 loss: 0.4509 +2023/06/05 09:15:56 - mmengine - INFO - Epoch(train) [8][1100/4111] lr: 2.5589e-05 eta: 1:59:20 time: 0.6311 data_time: 0.0016 memory: 44140 loss: 0.4817 +2023/06/05 09:17:00 - mmengine - INFO - Epoch(train) [8][1200/4111] lr: 2.5330e-05 eta: 1:58:16 time: 0.6341 data_time: 0.0019 memory: 44140 loss: 0.4340 +2023/06/05 09:17:14 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_all_lr1e-4_20230605_033556 +2023/06/05 09:18:03 - mmengine - INFO - Epoch(train) [8][1300/4111] lr: 2.5072e-05 eta: 1:57:12 time: 0.6306 data_time: 0.0015 memory: 44140 loss: 0.4692 +2023/06/05 09:19:06 - mmengine - INFO - Epoch(train) [8][1400/4111] lr: 2.4816e-05 eta: 1:56:08 time: 0.6330 data_time: 0.0016 memory: 44140 loss: 0.4543 +2023/06/05 09:20:09 - mmengine - INFO - Epoch(train) [8][1500/4111] lr: 2.4562e-05 eta: 1:55:05 time: 0.6324 data_time: 0.0016 memory: 44140 loss: 0.4653 +2023/06/05 09:21:13 - mmengine - INFO - Epoch(train) [8][1600/4111] lr: 2.4309e-05 eta: 1:54:01 time: 0.6328 data_time: 0.0018 memory: 44140 loss: 0.4977 +2023/06/05 09:22:16 - mmengine - INFO - Epoch(train) [8][1700/4111] lr: 2.4059e-05 eta: 1:52:57 time: 0.6337 data_time: 0.0025 memory: 44140 loss: 0.4812 +2023/06/05 09:23:20 - mmengine - INFO - Epoch(train) [8][1800/4111] lr: 2.3810e-05 eta: 1:51:53 time: 0.6321 data_time: 0.0015 memory: 44140 loss: 0.4598 +2023/06/05 09:24:23 - mmengine - INFO - Epoch(train) [8][1900/4111] lr: 2.3563e-05 eta: 1:50:49 time: 0.6400 data_time: 0.0023 memory: 44140 loss: 0.4702 +2023/06/05 09:25:26 - mmengine - INFO - Epoch(train) [8][2000/4111] lr: 2.3318e-05 eta: 1:49:45 time: 0.6316 data_time: 0.0016 memory: 44140 loss: 0.4607 +2023/06/05 09:26:29 - mmengine - INFO - Epoch(train) [8][2100/4111] lr: 2.3075e-05 eta: 1:48:41 time: 0.6321 data_time: 0.0016 memory: 44140 loss: 0.4970 +2023/06/05 09:27:33 - mmengine - INFO - Epoch(train) [8][2200/4111] lr: 2.2833e-05 eta: 1:47:37 time: 0.6326 data_time: 0.0016 memory: 44140 loss: 0.4677 +2023/06/05 09:27:47 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_all_lr1e-4_20230605_033556 +2023/06/05 09:28:36 - mmengine - INFO - Epoch(train) [8][2300/4111] lr: 2.2594e-05 eta: 1:46:33 time: 0.6351 data_time: 0.0016 memory: 44140 loss: 0.4667 +2023/06/05 09:29:39 - mmengine - INFO - Epoch(train) [8][2400/4111] lr: 2.2356e-05 eta: 1:45:30 time: 0.6318 data_time: 0.0017 memory: 44140 loss: 0.4802 +2023/06/05 09:30:42 - mmengine - INFO - Epoch(train) [8][2500/4111] lr: 2.2120e-05 eta: 1:44:26 time: 0.6317 data_time: 0.0019 memory: 44140 loss: 0.4583 +2023/06/05 09:31:45 - mmengine - INFO - Epoch(train) [8][2600/4111] lr: 2.1886e-05 eta: 1:43:22 time: 0.6315 data_time: 0.0017 memory: 44140 loss: 0.4647 +2023/06/05 09:32:49 - mmengine - INFO - Epoch(train) [8][2700/4111] lr: 2.1655e-05 eta: 1:42:18 time: 0.6331 data_time: 0.0018 memory: 44140 loss: 0.4615 +2023/06/05 09:33:52 - mmengine - INFO - Epoch(train) [8][2800/4111] lr: 2.1425e-05 eta: 1:41:14 time: 0.6334 data_time: 0.0016 memory: 44140 loss: 0.4511 +2023/06/05 09:34:56 - mmengine - INFO - Epoch(train) [8][2900/4111] lr: 2.1197e-05 eta: 1:40:10 time: 0.6347 data_time: 0.0016 memory: 44140 loss: 0.4822 +2023/06/05 09:35:59 - mmengine - INFO - Epoch(train) [8][3000/4111] lr: 2.0971e-05 eta: 1:39:06 time: 0.6328 data_time: 0.0016 memory: 44140 loss: 0.4560 +2023/06/05 09:37:02 - mmengine - INFO - Epoch(train) [8][3100/4111] lr: 2.0747e-05 eta: 1:38:03 time: 0.6321 data_time: 0.0016 memory: 44140 loss: 0.4886 +2023/06/05 09:38:06 - mmengine - INFO - Epoch(train) [8][3200/4111] lr: 2.0525e-05 eta: 1:36:59 time: 0.6333 data_time: 0.0014 memory: 44140 loss: 0.4754 +2023/06/05 09:38:20 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_all_lr1e-4_20230605_033556 +2023/06/05 09:39:09 - mmengine - INFO - Epoch(train) [8][3300/4111] lr: 2.0305e-05 eta: 1:35:55 time: 0.6328 data_time: 0.0017 memory: 44140 loss: 0.4577 +2023/06/05 09:40:12 - mmengine - INFO - Epoch(train) [8][3400/4111] lr: 2.0087e-05 eta: 1:34:51 time: 0.6324 data_time: 0.0018 memory: 44140 loss: 0.4704 +2023/06/05 09:41:16 - mmengine - INFO - Epoch(train) [8][3500/4111] lr: 1.9871e-05 eta: 1:33:47 time: 0.6318 data_time: 0.0021 memory: 44140 loss: 0.4640 +2023/06/05 09:42:19 - mmengine - INFO - Epoch(train) [8][3600/4111] lr: 1.9657e-05 eta: 1:32:43 time: 0.6317 data_time: 0.0017 memory: 44140 loss: 0.4610 +2023/06/05 09:43:22 - mmengine - INFO - Epoch(train) [8][3700/4111] lr: 1.9445e-05 eta: 1:31:40 time: 0.6316 data_time: 0.0018 memory: 44140 loss: 0.4569 +2023/06/05 09:44:25 - mmengine - INFO - Epoch(train) [8][3800/4111] lr: 1.9235e-05 eta: 1:30:36 time: 0.6332 data_time: 0.0016 memory: 44140 loss: 0.4758 +2023/06/05 09:45:29 - mmengine - INFO - Epoch(train) [8][3900/4111] lr: 1.9028e-05 eta: 1:29:32 time: 0.6356 data_time: 0.0025 memory: 44140 loss: 0.4882 +2023/06/05 09:46:32 - mmengine - INFO - Epoch(train) [8][4000/4111] lr: 1.8822e-05 eta: 1:28:28 time: 0.6315 data_time: 0.0019 memory: 44140 loss: 0.4461 +2023/06/05 09:47:35 - mmengine - INFO - Epoch(train) [8][4100/4111] lr: 1.8619e-05 eta: 1:27:24 time: 0.6334 data_time: 0.0032 memory: 44140 loss: 0.4858 +2023/06/05 09:47:42 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_all_lr1e-4_20230605_033556 +2023/06/05 09:47:42 - mmengine - INFO - Saving checkpoint at 8 epochs +2023/06/05 09:50:04 - mmengine - INFO - Epoch(val) [8][100/119] eta: 0:00:24 time: 1.2819 data_time: 0.0007 memory: 44140 +2023/06/05 09:50:49 - mmengine - INFO - Epoch(val) [8][119/119] accuracy/top1: 89.2193 single-label/precision_classwise: [93.67770385742188, 73.6904525756836] single-label/recall_classwise: [92.5383529663086, 76.99224853515625] single-label/f1-score_classwise: [93.10453796386719, 75.30517578125] data_time: 0.0196 time: 1.2817 +2023/06/05 09:51:56 - mmengine - INFO - Epoch(train) [9][ 100/4111] lr: 1.8395e-05 eta: 1:26:14 time: 0.6302 data_time: 0.0017 memory: 44140 loss: 0.4400 +2023/06/05 09:52:04 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_all_lr1e-4_20230605_033556 +2023/06/05 09:53:00 - mmengine - INFO - Epoch(train) [9][ 200/4111] lr: 1.8196e-05 eta: 1:25:11 time: 0.6339 data_time: 0.0022 memory: 44140 loss: 0.4535 +2023/06/05 09:54:03 - mmengine - INFO - Epoch(train) [9][ 300/4111] lr: 1.7999e-05 eta: 1:24:07 time: 0.6330 data_time: 0.0023 memory: 44140 loss: 0.4614 +2023/06/05 09:55:06 - mmengine - INFO - Epoch(train) [9][ 400/4111] lr: 1.7805e-05 eta: 1:23:03 time: 0.6372 data_time: 0.0031 memory: 44140 loss: 0.4876 +2023/06/05 09:56:10 - mmengine - INFO - Epoch(train) [9][ 500/4111] lr: 1.7612e-05 eta: 1:21:59 time: 0.6351 data_time: 0.0034 memory: 44140 loss: 0.5050 +2023/06/05 09:57:13 - mmengine - INFO - Epoch(train) [9][ 600/4111] lr: 1.7422e-05 eta: 1:20:56 time: 0.6342 data_time: 0.0018 memory: 44140 loss: 0.4839 +2023/06/05 09:58:17 - mmengine - INFO - Epoch(train) [9][ 700/4111] lr: 1.7234e-05 eta: 1:19:52 time: 0.6332 data_time: 0.0016 memory: 44140 loss: 0.4709 +2023/06/05 09:59:20 - mmengine - INFO - Epoch(train) [9][ 800/4111] lr: 1.7048e-05 eta: 1:18:48 time: 0.6323 data_time: 0.0018 memory: 44140 loss: 0.4840 +2023/06/05 10:00:23 - mmengine - INFO - Epoch(train) [9][ 900/4111] lr: 1.6864e-05 eta: 1:17:44 time: 0.6330 data_time: 0.0016 memory: 44140 loss: 0.4804 +2023/06/05 10:01:27 - mmengine - INFO - Epoch(train) [9][1000/4111] lr: 1.6683e-05 eta: 1:16:40 time: 0.6342 data_time: 0.0021 memory: 44140 loss: 0.4645 +2023/06/05 10:02:30 - mmengine - INFO - Epoch(train) [9][1100/4111] lr: 1.6504e-05 eta: 1:15:37 time: 0.6346 data_time: 0.0019 memory: 44140 loss: 0.4759 +2023/06/05 10:02:38 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_all_lr1e-4_20230605_033556 +2023/06/05 10:03:33 - mmengine - INFO - Epoch(train) [9][1200/4111] lr: 1.6327e-05 eta: 1:14:33 time: 0.6338 data_time: 0.0016 memory: 44140 loss: 0.4650 +2023/06/05 10:04:37 - mmengine - INFO - Epoch(train) [9][1300/4111] lr: 1.6152e-05 eta: 1:13:29 time: 0.6335 data_time: 0.0017 memory: 44140 loss: 0.4618 +2023/06/05 10:05:40 - mmengine - INFO - Epoch(train) [9][1400/4111] lr: 1.5980e-05 eta: 1:12:25 time: 0.6311 data_time: 0.0015 memory: 44140 loss: 0.4678 +2023/06/05 10:06:43 - mmengine - INFO - Epoch(train) [9][1500/4111] lr: 1.5810e-05 eta: 1:11:21 time: 0.6365 data_time: 0.0017 memory: 44140 loss: 0.4477 +2023/06/05 10:07:47 - mmengine - INFO - Epoch(train) [9][1600/4111] lr: 1.5642e-05 eta: 1:10:18 time: 0.6341 data_time: 0.0017 memory: 44140 loss: 0.4612 +2023/06/05 10:08:50 - mmengine - INFO - Epoch(train) [9][1700/4111] lr: 1.5476e-05 eta: 1:09:14 time: 0.6324 data_time: 0.0016 memory: 44140 loss: 0.4755 +2023/06/05 10:09:53 - mmengine - INFO - Epoch(train) [9][1800/4111] lr: 1.5313e-05 eta: 1:08:10 time: 0.6312 data_time: 0.0017 memory: 44140 loss: 0.4819 +2023/06/05 10:10:56 - mmengine - INFO - Epoch(train) [9][1900/4111] lr: 1.5152e-05 eta: 1:07:06 time: 0.6299 data_time: 0.0015 memory: 44140 loss: 0.4649 +2023/06/05 10:12:00 - mmengine - INFO - Epoch(train) [9][2000/4111] lr: 1.4993e-05 eta: 1:06:03 time: 0.6310 data_time: 0.0018 memory: 44140 loss: 0.4356 +2023/06/05 10:13:03 - mmengine - INFO - Epoch(train) [9][2100/4111] lr: 1.4837e-05 eta: 1:04:59 time: 0.6351 data_time: 0.0019 memory: 44140 loss: 0.4493 +2023/06/05 10:13:11 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_all_lr1e-4_20230605_033556 +2023/06/05 10:14:06 - mmengine - INFO - Epoch(train) [9][2200/4111] lr: 1.4683e-05 eta: 1:03:55 time: 0.6319 data_time: 0.0014 memory: 44140 loss: 0.4475 +2023/06/05 10:15:10 - mmengine - INFO - Epoch(train) [9][2300/4111] lr: 1.4532e-05 eta: 1:02:51 time: 0.6327 data_time: 0.0016 memory: 44140 loss: 0.4624 +2023/06/05 10:16:13 - mmengine - INFO - Epoch(train) [9][2400/4111] lr: 1.4382e-05 eta: 1:01:47 time: 0.6319 data_time: 0.0021 memory: 44140 loss: 0.4956 +2023/06/05 10:17:16 - mmengine - INFO - Epoch(train) [9][2500/4111] lr: 1.4235e-05 eta: 1:00:44 time: 0.6315 data_time: 0.0015 memory: 44140 loss: 0.4498 +2023/06/05 10:18:19 - mmengine - INFO - Epoch(train) [9][2600/4111] lr: 1.4091e-05 eta: 0:59:40 time: 0.6308 data_time: 0.0016 memory: 44140 loss: 0.4753 +2023/06/05 10:19:23 - mmengine - INFO - Epoch(train) [9][2700/4111] lr: 1.3949e-05 eta: 0:58:36 time: 0.6312 data_time: 0.0016 memory: 44140 loss: 0.4585 +2023/06/05 10:20:26 - mmengine - INFO - Epoch(train) [9][2800/4111] lr: 1.3809e-05 eta: 0:57:32 time: 0.6309 data_time: 0.0017 memory: 44140 loss: 0.4477 +2023/06/05 10:21:29 - mmengine - INFO - Epoch(train) [9][2900/4111] lr: 1.3672e-05 eta: 0:56:29 time: 0.6311 data_time: 0.0017 memory: 44140 loss: 0.4764 +2023/06/05 10:22:32 - mmengine - INFO - Epoch(train) [9][3000/4111] lr: 1.3537e-05 eta: 0:55:25 time: 0.6305 data_time: 0.0017 memory: 44140 loss: 0.4687 +2023/06/05 10:23:35 - mmengine - INFO - Epoch(train) [9][3100/4111] lr: 1.3405e-05 eta: 0:54:21 time: 0.6316 data_time: 0.0017 memory: 44140 loss: 0.4763 +2023/06/05 10:23:43 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_all_lr1e-4_20230605_033556 +2023/06/05 10:24:38 - mmengine - INFO - Epoch(train) [9][3200/4111] lr: 1.3275e-05 eta: 0:53:17 time: 0.6304 data_time: 0.0016 memory: 44140 loss: 0.4672 +2023/06/05 10:25:42 - mmengine - INFO - Epoch(train) [9][3300/4111] lr: 1.3147e-05 eta: 0:52:14 time: 0.6328 data_time: 0.0016 memory: 44140 loss: 0.4551 +2023/06/05 10:26:45 - mmengine - INFO - Epoch(train) [9][3400/4111] lr: 1.3022e-05 eta: 0:51:10 time: 0.6323 data_time: 0.0017 memory: 44140 loss: 0.4441 +2023/06/05 10:27:48 - mmengine - INFO - Epoch(train) [9][3500/4111] lr: 1.2899e-05 eta: 0:50:06 time: 0.6308 data_time: 0.0016 memory: 44140 loss: 0.4863 +2023/06/05 10:28:52 - mmengine - INFO - Epoch(train) [9][3600/4111] lr: 1.2779e-05 eta: 0:49:03 time: 0.6334 data_time: 0.0018 memory: 44140 loss: 0.4475 +2023/06/05 10:29:55 - mmengine - INFO - Epoch(train) [9][3700/4111] lr: 1.2661e-05 eta: 0:47:59 time: 0.6309 data_time: 0.0018 memory: 44140 loss: 0.4629 +2023/06/05 10:30:58 - mmengine - INFO - Epoch(train) [9][3800/4111] lr: 1.2546e-05 eta: 0:46:55 time: 0.6338 data_time: 0.0016 memory: 44140 loss: 0.4707 +2023/06/05 10:32:02 - mmengine - INFO - Epoch(train) [9][3900/4111] lr: 1.2433e-05 eta: 0:45:51 time: 0.6305 data_time: 0.0016 memory: 44140 loss: 0.4491 +2023/06/05 10:33:05 - mmengine - INFO - Epoch(train) [9][4000/4111] lr: 1.2323e-05 eta: 0:44:48 time: 0.6350 data_time: 0.0023 memory: 44140 loss: 0.4553 +2023/06/05 10:34:08 - mmengine - INFO - Epoch(train) [9][4100/4111] lr: 1.2215e-05 eta: 0:43:44 time: 0.6342 data_time: 0.0031 memory: 44140 loss: 0.4479 +2023/06/05 10:34:15 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_all_lr1e-4_20230605_033556 +2023/06/05 10:34:15 - mmengine - INFO - Saving checkpoint at 9 epochs +2023/06/05 10:36:37 - mmengine - INFO - Epoch(val) [9][100/119] eta: 0:00:24 time: 1.2825 data_time: 0.0008 memory: 44140 +2023/06/05 10:37:20 - mmengine - INFO - Epoch(val) [9][119/119] accuracy/top1: 89.4196 single-label/precision_classwise: [93.60687255859375, 74.53064727783203] single-label/recall_classwise: [92.89186096191406, 76.6279067993164] single-label/f1-score_classwise: [93.24799346923828, 75.56472778320312] data_time: 0.0195 time: 1.2815 +2023/06/05 10:37:24 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_all_lr1e-4_20230605_033556 +2023/06/05 10:38:27 - mmengine - INFO - Epoch(train) [10][ 100/4111] lr: 1.2098e-05 eta: 0:42:34 time: 0.6327 data_time: 0.0014 memory: 44140 loss: 0.4737 +2023/06/05 10:39:31 - mmengine - INFO - Epoch(train) [10][ 200/4111] lr: 1.1996e-05 eta: 0:41:30 time: 0.6335 data_time: 0.0016 memory: 44140 loss: 0.4663 +2023/06/05 10:40:34 - mmengine - INFO - Epoch(train) [10][ 300/4111] lr: 1.1896e-05 eta: 0:40:26 time: 0.6322 data_time: 0.0015 memory: 44140 loss: 0.4652 +2023/06/05 10:41:37 - mmengine - INFO - Epoch(train) [10][ 400/4111] lr: 1.1798e-05 eta: 0:39:23 time: 0.6337 data_time: 0.0015 memory: 44140 loss: 0.4798 +2023/06/05 10:44:05 - mmengine - INFO - Epoch(train) [10][ 500/4111] lr: 1.1703e-05 eta: 0:38:27 time: 0.6315 data_time: 0.0017 memory: 44140 loss: 0.4662 +2023/06/05 10:45:09 - mmengine - INFO - Epoch(train) [10][ 600/4111] lr: 1.1611e-05 eta: 0:37:23 time: 0.6319 data_time: 0.0016 memory: 44140 loss: 0.4556 +2023/06/05 10:46:12 - mmengine - INFO - Epoch(train) [10][ 700/4111] lr: 1.1521e-05 eta: 0:36:19 time: 0.6322 data_time: 0.0015 memory: 44140 loss: 0.4515 +2023/06/05 10:47:15 - mmengine - INFO - Epoch(train) [10][ 800/4111] lr: 1.1434e-05 eta: 0:35:15 time: 0.6329 data_time: 0.0017 memory: 44140 loss: 0.4552 +2023/06/05 10:48:18 - mmengine - INFO - Epoch(train) [10][ 900/4111] lr: 1.1349e-05 eta: 0:34:11 time: 0.6328 data_time: 0.0016 memory: 44140 loss: 0.4752 +2023/06/05 10:49:22 - mmengine - INFO - Epoch(train) [10][1000/4111] lr: 1.1267e-05 eta: 0:33:07 time: 0.6330 data_time: 0.0015 memory: 44140 loss: 0.4833 +2023/06/05 10:49:23 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_all_lr1e-4_20230605_033556 +2023/06/05 10:50:25 - mmengine - INFO - Epoch(train) [10][1100/4111] lr: 1.1187e-05 eta: 0:32:03 time: 0.6346 data_time: 0.0016 memory: 44140 loss: 0.4649 +2023/06/05 10:51:29 - mmengine - INFO - Epoch(train) [10][1200/4111] lr: 1.1110e-05 eta: 0:30:59 time: 0.6337 data_time: 0.0017 memory: 44140 loss: 0.5148 +2023/06/05 10:52:32 - mmengine - INFO - Epoch(train) [10][1300/4111] lr: 1.1035e-05 eta: 0:29:55 time: 0.6327 data_time: 0.0018 memory: 44140 loss: 0.4880 +2023/06/05 10:53:35 - mmengine - INFO - Epoch(train) [10][1400/4111] lr: 1.0963e-05 eta: 0:28:52 time: 0.6331 data_time: 0.0017 memory: 44140 loss: 0.4748 +2023/06/05 10:54:39 - mmengine - INFO - Epoch(train) [10][1500/4111] lr: 1.0893e-05 eta: 0:27:48 time: 0.6358 data_time: 0.0016 memory: 44140 loss: 0.4528 +2023/06/05 10:55:42 - mmengine - INFO - Epoch(train) [10][1600/4111] lr: 1.0827e-05 eta: 0:26:44 time: 0.6352 data_time: 0.0017 memory: 44140 loss: 0.4526 +2023/06/05 10:56:46 - mmengine - INFO - Epoch(train) [10][1700/4111] lr: 1.0762e-05 eta: 0:25:40 time: 0.6326 data_time: 0.0019 memory: 44140 loss: 0.4581 +2023/06/05 10:57:49 - mmengine - INFO - Epoch(train) [10][1800/4111] lr: 1.0701e-05 eta: 0:24:36 time: 0.6351 data_time: 0.0022 memory: 44140 loss: 0.4672 +2023/06/05 10:58:52 - mmengine - INFO - Epoch(train) [10][1900/4111] lr: 1.0641e-05 eta: 0:23:32 time: 0.6344 data_time: 0.0019 memory: 44140 loss: 0.4546 +2023/06/05 10:59:56 - mmengine - INFO - Epoch(train) [10][2000/4111] lr: 1.0585e-05 eta: 0:22:28 time: 0.6364 data_time: 0.0022 memory: 44140 loss: 0.4611 +2023/06/05 10:59:57 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_all_lr1e-4_20230605_033556 +2023/06/05 11:00:59 - mmengine - INFO - Epoch(train) [10][2100/4111] lr: 1.0531e-05 eta: 0:21:24 time: 0.6382 data_time: 0.0017 memory: 44140 loss: 0.4460 +2023/06/05 11:02:03 - mmengine - INFO - Epoch(train) [10][2200/4111] lr: 1.0480e-05 eta: 0:20:20 time: 0.6344 data_time: 0.0015 memory: 44140 loss: 0.4586 +2023/06/05 11:03:06 - mmengine - INFO - Epoch(train) [10][2300/4111] lr: 1.0431e-05 eta: 0:19:16 time: 0.6340 data_time: 0.0016 memory: 44140 loss: 0.4806 +2023/06/05 11:04:10 - mmengine - INFO - Epoch(train) [10][2400/4111] lr: 1.0385e-05 eta: 0:18:12 time: 0.6326 data_time: 0.0015 memory: 44140 loss: 0.4657 +2023/06/05 11:05:13 - mmengine - INFO - Epoch(train) [10][2500/4111] lr: 1.0341e-05 eta: 0:17:09 time: 0.6344 data_time: 0.0015 memory: 44140 loss: 0.4758 +2023/06/05 11:06:17 - mmengine - INFO - Epoch(train) [10][2600/4111] lr: 1.0300e-05 eta: 0:16:05 time: 0.6344 data_time: 0.0015 memory: 44140 loss: 0.4728 +2023/06/05 11:07:20 - mmengine - INFO - Epoch(train) [10][2700/4111] lr: 1.0262e-05 eta: 0:15:01 time: 0.6333 data_time: 0.0022 memory: 44140 loss: 0.4543 +2023/06/05 11:08:24 - mmengine - INFO - Epoch(train) [10][2800/4111] lr: 1.0226e-05 eta: 0:13:57 time: 0.6335 data_time: 0.0024 memory: 44140 loss: 0.4590 +2023/06/05 11:09:27 - mmengine - INFO - Epoch(train) [10][2900/4111] lr: 1.0193e-05 eta: 0:12:53 time: 0.6348 data_time: 0.0019 memory: 44140 loss: 0.4657 +2023/06/05 11:10:30 - mmengine - INFO - Epoch(train) [10][3000/4111] lr: 1.0162e-05 eta: 0:11:49 time: 0.6329 data_time: 0.0017 memory: 44140 loss: 0.4662 +2023/06/05 11:10:31 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_all_lr1e-4_20230605_033556 +2023/06/05 11:11:34 - mmengine - INFO - Epoch(train) [10][3100/4111] lr: 1.0135e-05 eta: 0:10:45 time: 0.6337 data_time: 0.0016 memory: 44140 loss: 0.4346 +2023/06/05 11:12:37 - mmengine - INFO - Epoch(train) [10][3200/4111] lr: 1.0109e-05 eta: 0:09:41 time: 0.6331 data_time: 0.0017 memory: 44140 loss: 0.4634 +2023/06/05 11:13:41 - mmengine - INFO - Epoch(train) [10][3300/4111] lr: 1.0087e-05 eta: 0:08:37 time: 0.6338 data_time: 0.0015 memory: 44140 loss: 0.4903 +2023/06/05 11:14:44 - mmengine - INFO - Epoch(train) [10][3400/4111] lr: 1.0067e-05 eta: 0:07:34 time: 0.6360 data_time: 0.0022 memory: 44140 loss: 0.4803 +2023/06/05 11:15:48 - mmengine - INFO - Epoch(train) [10][3500/4111] lr: 1.0049e-05 eta: 0:06:30 time: 0.6344 data_time: 0.0016 memory: 44140 loss: 0.4385 +2023/06/05 11:16:51 - mmengine - INFO - Epoch(train) [10][3600/4111] lr: 1.0034e-05 eta: 0:05:26 time: 0.6349 data_time: 0.0019 memory: 44140 loss: 0.4665 +2023/06/05 11:17:55 - mmengine - INFO - Epoch(train) [10][3700/4111] lr: 1.0022e-05 eta: 0:04:22 time: 0.6346 data_time: 0.0016 memory: 44140 loss: 0.4572 +2023/06/05 11:18:58 - mmengine - INFO - Epoch(train) [10][3800/4111] lr: 1.0013e-05 eta: 0:03:18 time: 0.6333 data_time: 0.0016 memory: 44140 loss: 0.4705 +2023/06/05 11:20:02 - mmengine - INFO - Epoch(train) [10][3900/4111] lr: 1.0006e-05 eta: 0:02:14 time: 0.6342 data_time: 0.0017 memory: 44140 loss: 0.4690 +2023/06/05 11:21:05 - mmengine - INFO - Epoch(train) [10][4000/4111] lr: 1.0002e-05 eta: 0:01:10 time: 0.6397 data_time: 0.0016 memory: 44140 loss: 0.4479 +2023/06/05 11:21:06 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_all_lr1e-4_20230605_033556 +2023/06/05 11:22:09 - mmengine - INFO - Epoch(train) [10][4100/4111] lr: 1.0000e-05 eta: 0:00:07 time: 0.6368 data_time: 0.0032 memory: 44140 loss: 0.4710 +2023/06/05 11:22:15 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_all_lr1e-4_20230605_033556 +2023/06/05 11:22:15 - mmengine - INFO - Saving checkpoint at 10 epochs +2023/06/05 11:24:36 - mmengine - INFO - Epoch(val) [10][100/119] eta: 0:00:24 time: 1.2809 data_time: 0.0008 memory: 44140 +2023/06/05 11:25:19 - mmengine - INFO - Epoch(val) [10][119/119] accuracy/top1: 89.4808 single-label/precision_classwise: [93.50679779052734, 74.95423889160156] single-label/recall_classwise: [93.08966064453125, 76.18605041503906] single-label/f1-score_classwise: [93.2977523803711, 75.56512451171875] data_time: 0.0187 time: 1.2786 diff --git a/clip_large_pretrain_4x256_all_lr1e-4/20230605_033556/vis_data/20230605_033556.json b/clip_large_pretrain_4x256_all_lr1e-4/20230605_033556/vis_data/20230605_033556.json new file mode 100644 index 0000000000000000000000000000000000000000..be6f0f3b722592091fea0ffe54a1a68189ee20f4 --- /dev/null +++ b/clip_large_pretrain_4x256_all_lr1e-4/20230605_033556/vis_data/20230605_033556.json @@ -0,0 +1,420 @@ +{"lr": 9.999871217862936e-05, "data_time": 0.0025150299072265623, "loss": 0.5926630020141601, "time": 0.6296663045883178, "epoch": 1, "memory": 44138, "step": 100} +{"lr": 9.999479662539605e-05, "data_time": 0.0023998022079467773, "loss": 0.5558342933654785, "time": 0.629915189743042, "epoch": 1, "memory": 44138, "step": 200} +{"lr": 9.99882534375654e-05, "data_time": 0.002423381805419922, "loss": 0.5518157005310058, "time": 0.6311023950576782, "epoch": 1, "memory": 44138, "step": 300} +{"lr": 9.997908299725075e-05, "data_time": 0.002051043510437012, "loss": 0.5356977939605713, "time": 0.6294378757476806, "epoch": 1, "memory": 44138, "step": 400} +{"lr": 9.996728583999318e-05, "data_time": 0.00154569149017334, "loss": 0.5294235706329345, "time": 0.6362393140792847, "epoch": 1, "memory": 44138, "step": 500} +{"lr": 9.995286265473058e-05, "data_time": 0.0016171932220458984, "loss": 0.5345909118652343, "time": 0.6289902448654174, "epoch": 1, "memory": 44138, "step": 600} +{"lr": 9.993581428375725e-05, "data_time": 0.001757645606994629, "loss": 0.546266770362854, "time": 0.6302518606185913, "epoch": 1, "memory": 44138, "step": 700} +{"lr": 9.991614172267466e-05, "data_time": 0.001504993438720703, "loss": 0.5193750143051148, "time": 0.6299715280532837, "epoch": 1, "memory": 44138, "step": 800} +{"lr": 9.989384612033355e-05, "data_time": 0.001619243621826172, "loss": 0.5281996011734009, "time": 0.630258321762085, "epoch": 1, "memory": 44138, "step": 900} +{"lr": 9.986892877876658e-05, "data_time": 0.001715373992919922, "loss": 0.5351518869400025, "time": 0.6305138349533081, "epoch": 1, "memory": 44138, "step": 1000} +{"lr": 9.984139115311244e-05, "data_time": 0.002127528190612793, "loss": 0.4951824188232422, "time": 0.6299308776855469, "epoch": 1, "memory": 44138, "step": 1100} +{"lr": 9.981123485153081e-05, "data_time": 0.0020805835723876954, "loss": 0.5200381517410279, "time": 0.6306166172027587, "epoch": 1, "memory": 44138, "step": 1200} +{"lr": 9.977846163510851e-05, "data_time": 0.0016455411911010741, "loss": 0.5057559251785279, "time": 0.6302486896514893, "epoch": 1, "memory": 44138, "step": 1300} +{"lr": 9.974307341775653e-05, "data_time": 0.0017376422882080078, "loss": 0.5007327795028687, "time": 0.6307777881622314, "epoch": 1, "memory": 44138, "step": 1400} +{"lr": 9.970507226609833e-05, "data_time": 0.0014547348022460938, "loss": 0.5251185774803162, "time": 0.6306758403778077, "epoch": 1, "memory": 44138, "step": 1500} +{"lr": 9.966446039934934e-05, "data_time": 0.0017030715942382812, "loss": 0.5359284937381744, "time": 0.6308729887008667, "epoch": 1, "memory": 44138, "step": 1600} +{"lr": 9.962124018918683e-05, "data_time": 0.0017491579055786133, "loss": 0.524097228050232, "time": 0.6306417942047119, "epoch": 1, "memory": 44138, "step": 1700} +{"lr": 9.957541415961212e-05, "data_time": 0.002335810661315918, "loss": 0.5213017463684082, "time": 0.6322964191436767, "epoch": 1, "memory": 44138, "step": 1800} +{"lr": 9.952698498680267e-05, "data_time": 0.0018424272537231445, "loss": 0.5202134847640991, "time": 0.6315749406814575, "epoch": 1, "memory": 44138, "step": 1900} +{"lr": 9.947595549895583e-05, "data_time": 0.0017011404037475587, "loss": 0.5050365686416626, "time": 0.6295128107070923, "epoch": 1, "memory": 44138, "step": 2000} +{"lr": 9.942232867612417e-05, "data_time": 0.0015781402587890625, "loss": 0.5164885520935059, "time": 0.6307832717895507, "epoch": 1, "memory": 44138, "step": 2100} +{"lr": 9.93661076500406e-05, "data_time": 0.0018105506896972656, "loss": 0.5222069263458252, "time": 0.6295430421829223, "epoch": 1, "memory": 44138, "step": 2200} +{"lr": 9.930729570393622e-05, "data_time": 0.0017452716827392578, "loss": 0.5055970311164856, "time": 0.6308633804321289, "epoch": 1, "memory": 44138, "step": 2300} +{"lr": 9.924589627234824e-05, "data_time": 0.0019497632980346679, "loss": 0.49894664287567136, "time": 0.631049394607544, "epoch": 1, "memory": 44138, "step": 2400} +{"lr": 9.918191294091956e-05, "data_time": 0.0015043020248413086, "loss": 0.4959339380264282, "time": 0.6319576978683472, "epoch": 1, "memory": 44138, "step": 2500} +{"lr": 9.91153494461892e-05, "data_time": 0.0019474506378173828, "loss": 0.4856060862541199, "time": 0.6302881717681885, "epoch": 1, "memory": 44138, "step": 2600} +{"lr": 9.904620967537422e-05, "data_time": 0.0016234874725341796, "loss": 0.5094678044319153, "time": 0.6311283111572266, "epoch": 1, "memory": 44138, "step": 2700} +{"lr": 9.897449766614277e-05, "data_time": 0.0016669034957885742, "loss": 0.5114233791828156, "time": 0.6291088342666626, "epoch": 1, "memory": 44138, "step": 2800} +{"lr": 9.890021760637804e-05, "data_time": 0.0017839908599853516, "loss": 0.48612632751464846, "time": 0.6291269302368164, "epoch": 1, "memory": 44138, "step": 2900} +{"lr": 9.882337383393409e-05, "data_time": 0.0013880491256713866, "loss": 0.510037076473236, "time": 0.633272933959961, "epoch": 1, "memory": 44138, "step": 3000} +{"lr": 9.874397083638193e-05, "data_time": 0.001481318473815918, "loss": 0.48753076791763306, "time": 0.6303290605545044, "epoch": 1, "memory": 44138, "step": 3100} +{"lr": 9.866201325074828e-05, "data_time": 0.0018770217895507813, "loss": 0.5040024042129516, "time": 0.6320106744766235, "epoch": 1, "memory": 44138, "step": 3200} +{"lr": 9.857750586324393e-05, "data_time": 0.0016488313674926757, "loss": 0.4848734140396118, "time": 0.6298616647720336, "epoch": 1, "memory": 44138, "step": 3300} +{"lr": 9.84904536089848e-05, "data_time": 0.0014603137969970703, "loss": 0.5120719373226166, "time": 0.6313069105148316, "epoch": 1, "memory": 44138, "step": 3400} +{"lr": 9.840086157170351e-05, "data_time": 0.001648855209350586, "loss": 0.48729785680770876, "time": 0.6295938968658448, "epoch": 1, "memory": 44138, "step": 3500} +{"lr": 9.830873498345255e-05, "data_time": 0.0017504692077636719, "loss": 0.4966985821723938, "time": 0.6311012744903565, "epoch": 1, "memory": 44138, "step": 3600} +{"lr": 9.821407922429864e-05, "data_time": 0.001745462417602539, "loss": 0.5099364876747131, "time": 0.6309836387634278, "epoch": 1, "memory": 44138, "step": 3700} +{"lr": 9.811689982200877e-05, "data_time": 0.001806783676147461, "loss": 0.4760929882526398, "time": 0.632443618774414, "epoch": 1, "memory": 44138, "step": 3800} +{"lr": 9.801720245172702e-05, "data_time": 0.0017486810684204102, "loss": 0.4971849679946899, "time": 0.6320624351501465, "epoch": 1, "memory": 44138, "step": 3900} +{"lr": 9.791499293564358e-05, "data_time": 0.0014429569244384765, "loss": 0.5001693487167358, "time": 0.6314241647720337, "epoch": 1, "memory": 44138, "step": 4000} +{"lr": 9.781027724265439e-05, "data_time": 0.0048599720001220705, "loss": 0.47135170698165896, "time": 0.6343180179595947, "epoch": 1, "memory": 44138, "step": 4100} +{"accuracy/top1": 71.24604797363281, "data_time": 0.025956197947013277, "time": 1.3182958634961552, "step": 1} +{"lr": 9.769111539369434e-05, "data_time": 0.00241999626159668, "loss": 0.4788080334663391, "time": 0.6337323188781738, "epoch": 2, "memory": 44140, "step": 4211} +{"lr": 9.758113190911623e-05, "data_time": 0.0015765190124511718, "loss": 0.49256587624549864, "time": 0.6336324691772461, "epoch": 2, "memory": 44140, "step": 4311} +{"lr": 9.746866174465984e-05, "data_time": 0.0021776914596557616, "loss": 0.4984575003385544, "time": 0.63229341506958, "epoch": 2, "memory": 44140, "step": 4411} +{"lr": 9.735371146842894e-05, "data_time": 0.0018930912017822265, "loss": 0.4467486679553986, "time": 0.6317708253860473, "epoch": 2, "memory": 44140, "step": 4511} +{"lr": 9.723628779336255e-05, "data_time": 0.0014719247817993164, "loss": 0.49971413016319277, "time": 0.6333870410919189, "epoch": 2, "memory": 44140, "step": 4611} +{"lr": 9.711639757684265e-05, "data_time": 0.0014638185501098632, "loss": 0.45254039764404297, "time": 0.6319808959960938, "epoch": 2, "memory": 44140, "step": 4711} +{"lr": 9.699404782029395e-05, "data_time": 0.0017662525177001953, "loss": 0.4845359563827515, "time": 0.6309187889099122, "epoch": 2, "memory": 44140, "step": 4811} +{"lr": 9.686924566877498e-05, "data_time": 0.0018222808837890625, "loss": 0.4613297522068024, "time": 0.6334790229797364, "epoch": 2, "memory": 44140, "step": 4911} +{"lr": 9.674199841056078e-05, "data_time": 0.0015243291854858398, "loss": 0.48396073579788207, "time": 0.6328363418579102, "epoch": 2, "memory": 44140, "step": 5011} +{"lr": 9.661231347671705e-05, "data_time": 0.0021074771881103515, "loss": 0.4813476920127869, "time": 0.6317301511764526, "epoch": 2, "memory": 44140, "step": 5111} +{"lr": 9.64801984406668e-05, "data_time": 0.0014989137649536132, "loss": 0.47513251900672915, "time": 0.6317736387252808, "epoch": 2, "memory": 44140, "step": 5211} +{"lr": 9.634566101774737e-05, "data_time": 0.0014298200607299806, "loss": 0.47101333141326907, "time": 0.6307046175003052, "epoch": 2, "memory": 44140, "step": 5311} +{"lr": 9.62087090647604e-05, "data_time": 0.001444387435913086, "loss": 0.4979289352893829, "time": 0.6319366455078125, "epoch": 2, "memory": 44140, "step": 5411} +{"lr": 9.606935057951247e-05, "data_time": 0.001916360855102539, "loss": 0.4742088675498962, "time": 0.6350284576416015, "epoch": 2, "memory": 44140, "step": 5511} +{"lr": 9.592759370034884e-05, "data_time": 0.0018108129501342774, "loss": 0.4734244287014008, "time": 0.6316951751708985, "epoch": 2, "memory": 44140, "step": 5611} +{"lr": 9.578344670567721e-05, "data_time": 0.0020900964736938477, "loss": 0.46299545764923095, "time": 0.6309096097946167, "epoch": 2, "memory": 44140, "step": 5711} +{"lr": 9.563691801348495e-05, "data_time": 0.0016538619995117188, "loss": 0.4517266273498535, "time": 0.6313366651535034, "epoch": 2, "memory": 44140, "step": 5811} +{"lr": 9.548801618084738e-05, "data_time": 0.0017515897750854492, "loss": 0.4669777274131775, "time": 0.6331912040710449, "epoch": 2, "memory": 44140, "step": 5911} +{"lr": 9.533674990342786e-05, "data_time": 0.001575803756713867, "loss": 0.49224611520767214, "time": 0.6328452348709106, "epoch": 2, "memory": 44140, "step": 6011} +{"lr": 9.518312801497029e-05, "data_time": 0.0015989303588867187, "loss": 0.4860232055187225, "time": 0.6313059568405152, "epoch": 2, "memory": 44140, "step": 6111} +{"lr": 9.502715948678276e-05, "data_time": 0.0015105009078979492, "loss": 0.4802593559026718, "time": 0.6314626455307006, "epoch": 2, "memory": 44140, "step": 6211} +{"lr": 9.486885342721427e-05, "data_time": 0.0015629768371582032, "loss": 0.47859914898872374, "time": 0.631781530380249, "epoch": 2, "memory": 44140, "step": 6311} +{"lr": 9.470821908112202e-05, "data_time": 0.0019846439361572267, "loss": 0.4706738114356995, "time": 0.633806848526001, "epoch": 2, "memory": 44140, "step": 6411} +{"lr": 9.454526582933214e-05, "data_time": 0.0017047643661499024, "loss": 0.47435392141342164, "time": 0.6317571401596069, "epoch": 2, "memory": 44140, "step": 6511} +{"lr": 9.438000318809176e-05, "data_time": 0.0014998912811279297, "loss": 0.44744108319282533, "time": 0.6303221464157105, "epoch": 2, "memory": 44140, "step": 6611} +{"lr": 9.42124408085132e-05, "data_time": 0.0017312526702880859, "loss": 0.4806430101394653, "time": 0.6302761793136596, "epoch": 2, "memory": 44140, "step": 6711} +{"lr": 9.404258847601036e-05, "data_time": 0.0016814947128295898, "loss": 0.4817513346672058, "time": 0.6357363224029541, "epoch": 2, "memory": 44140, "step": 6811} +{"lr": 9.387045610972717e-05, "data_time": 0.001527714729309082, "loss": 0.4747046947479248, "time": 0.6315042495727539, "epoch": 2, "memory": 44140, "step": 6911} +{"lr": 9.369605376195838e-05, "data_time": 0.001538991928100586, "loss": 0.4685009777545929, "time": 0.6310405254364013, "epoch": 2, "memory": 44140, "step": 7011} +{"lr": 9.351939161756269e-05, "data_time": 0.0025190114974975586, "loss": 0.5208580732345581, "time": 0.6315462589263916, "epoch": 2, "memory": 44140, "step": 7111} +{"lr": 9.334047999336764e-05, "data_time": 0.0015179872512817382, "loss": 0.47420313358306887, "time": 0.6344522953033447, "epoch": 2, "memory": 44140, "step": 7211} +{"lr": 9.315932933756752e-05, "data_time": 0.001871800422668457, "loss": 0.4483091592788696, "time": 0.6340370893478393, "epoch": 2, "memory": 44140, "step": 7311} +{"lr": 9.297595022911287e-05, "data_time": 0.001640176773071289, "loss": 0.4723667621612549, "time": 0.6328441619873046, "epoch": 2, "memory": 44140, "step": 7411} +{"lr": 9.279035337709295e-05, "data_time": 0.0017405033111572265, "loss": 0.4809338450431824, "time": 0.6365816354751587, "epoch": 2, "memory": 44140, "step": 7511} +{"lr": 9.260254962011016e-05, "data_time": 0.002028656005859375, "loss": 0.4780664682388306, "time": 0.6328401327133178, "epoch": 2, "memory": 44140, "step": 7611} +{"lr": 9.241254992564702e-05, "data_time": 0.0027015209197998047, "loss": 0.4529854774475098, "time": 0.6335997104644775, "epoch": 2, "memory": 44140, "step": 7711} +{"lr": 9.22203653894259e-05, "data_time": 0.0018162488937377929, "loss": 0.4671182632446289, "time": 0.6313928604125977, "epoch": 2, "memory": 44140, "step": 7811} +{"lr": 9.202600723476098e-05, "data_time": 0.001990199089050293, "loss": 0.4647050857543945, "time": 0.632718563079834, "epoch": 2, "memory": 44140, "step": 7911} +{"lr": 9.182948681190268e-05, "data_time": 0.0018062114715576172, "loss": 0.4600569248199463, "time": 0.6309822559356689, "epoch": 2, "memory": 44140, "step": 8011} +{"lr": 9.163081559737497e-05, "data_time": 0.0019437789916992188, "loss": 0.4363451361656189, "time": 0.632062840461731, "epoch": 2, "memory": 44140, "step": 8111} +{"lr": 9.143000519330517e-05, "data_time": 0.0031851530075073242, "loss": 0.4707728177309036, "time": 0.635717511177063, "epoch": 2, "memory": 44140, "step": 8211} +{"accuracy/top1": 83.37223052978516, "data_time": 0.019402621189753215, "time": 1.2811894913514454, "step": 2} +{"lr": 9.120461478753376e-05, "data_time": 0.0016452312469482423, "loss": 0.46649948954582215, "time": 0.6308094978332519, "epoch": 3, "memory": 44140, "step": 8322} +{"lr": 9.099932932138134e-05, "data_time": 0.001672673225402832, "loss": 0.45427868366241453, "time": 0.6313965082168579, "epoch": 3, "memory": 44140, "step": 8422} +{"lr": 9.079194154362181e-05, "data_time": 0.0016325950622558595, "loss": 0.4859224081039429, "time": 0.6315812349319458, "epoch": 3, "memory": 44140, "step": 8522} +{"lr": 9.058246356541787e-05, "data_time": 0.0018900394439697265, "loss": 0.49377862811088563, "time": 0.6308974981307983, "epoch": 3, "memory": 44140, "step": 8622} +{"lr": 9.037090761999699e-05, "data_time": 0.002896904945373535, "loss": 0.4653532266616821, "time": 0.637605881690979, "epoch": 3, "memory": 44140, "step": 8722} +{"lr": 9.015728606193701e-05, "data_time": 0.0014890909194946289, "loss": 0.44080936908721924, "time": 0.6310546398162842, "epoch": 3, "memory": 44140, "step": 8822} +{"lr": 8.994161136644488e-05, "data_time": 0.0022495746612548827, "loss": 0.44987755119800565, "time": 0.6320571422576904, "epoch": 3, "memory": 44140, "step": 8922} +{"lr": 8.972389612862792e-05, "data_time": 0.0016071796417236328, "loss": 0.47653878927230836, "time": 0.6318201541900634, "epoch": 3, "memory": 44140, "step": 9022} +{"lr": 8.950415306275833e-05, "data_time": 0.001922607421875, "loss": 0.47898975014686584, "time": 0.6368085384368897, "epoch": 3, "memory": 44140, "step": 9122} +{"lr": 8.928239500153059e-05, "data_time": 0.0017330169677734375, "loss": 0.46713787913322447, "time": 0.632226037979126, "epoch": 3, "memory": 44140, "step": 9222} +{"lr": 8.905863489531232e-05, "data_time": 0.0019263505935668945, "loss": 0.4839877486228943, "time": 0.6313059568405152, "epoch": 3, "memory": 44140, "step": 9322} +{"lr": 8.883288581138775e-05, "data_time": 0.0015526056289672852, "loss": 0.4634073913097382, "time": 0.6307883262634277, "epoch": 3, "memory": 44140, "step": 9422} +{"lr": 8.86051609331947e-05, "data_time": 0.0015683889389038086, "loss": 0.4593377411365509, "time": 0.6345818996429443, "epoch": 3, "memory": 44140, "step": 9522} +{"lr": 8.837547355955467e-05, "data_time": 0.001513195037841797, "loss": 0.4885055422782898, "time": 0.6330514192581177, "epoch": 3, "memory": 44140, "step": 9622} +{"lr": 8.814383710389614e-05, "data_time": 0.0015676021575927734, "loss": 0.45213906168937684, "time": 0.6329608917236328, "epoch": 3, "memory": 44140, "step": 9722} +{"lr": 8.791026509347146e-05, "data_time": 0.0016211986541748047, "loss": 0.4513511657714844, "time": 0.6330824851989746, "epoch": 3, "memory": 44140, "step": 9822} +{"lr": 8.767477116856659e-05, "data_time": 0.0014519214630126954, "loss": 0.4570546567440033, "time": 0.6307456970214844, "epoch": 3, "memory": 44140, "step": 9922} +{"lr": 8.743736908170466e-05, "data_time": 0.0016186952590942383, "loss": 0.4745335578918457, "time": 0.6312720060348511, "epoch": 3, "memory": 44140, "step": 10022} +{"lr": 8.719807269684305e-05, "data_time": 0.001607060432434082, "loss": 0.4391482412815094, "time": 0.6320138692855835, "epoch": 3, "memory": 44140, "step": 10122} +{"lr": 8.695689598856356e-05, "data_time": 0.0014997005462646484, "loss": 0.463805741071701, "time": 0.6313628196716309, "epoch": 3, "memory": 44140, "step": 10222} +{"lr": 8.67138530412559e-05, "data_time": 0.0024975061416625975, "loss": 0.4336999773979187, "time": 0.6314276933670044, "epoch": 3, "memory": 44140, "step": 10322} +{"lr": 8.646895804829604e-05, "data_time": 0.0016403913497924805, "loss": 0.4856491804122925, "time": 0.6313858985900879, "epoch": 3, "memory": 44140, "step": 10422} +{"lr": 8.622222531121664e-05, "data_time": 0.00155792236328125, "loss": 0.46217219829559325, "time": 0.6314802169799805, "epoch": 3, "memory": 44140, "step": 10522} +{"lr": 8.597366923887211e-05, "data_time": 0.0015955448150634765, "loss": 0.446012145280838, "time": 0.6316116571426391, "epoch": 3, "memory": 44140, "step": 10622} +{"lr": 8.57233043465972e-05, "data_time": 0.0016615152359008788, "loss": 0.46443553566932677, "time": 0.6339049577713013, "epoch": 3, "memory": 44140, "step": 10722} +{"lr": 8.547114525535937e-05, "data_time": 0.0020400047302246093, "loss": 0.4640778601169586, "time": 0.6322259426116943, "epoch": 3, "memory": 44140, "step": 10822} +{"lr": 8.521720669090474e-05, "data_time": 0.002305126190185547, "loss": 0.4476841986179352, "time": 0.6317931175231933, "epoch": 3, "memory": 44140, "step": 10922} +{"lr": 8.496150348289821e-05, "data_time": 0.0015257835388183595, "loss": 0.4819492340087891, "time": 0.6307430028915405, "epoch": 3, "memory": 44140, "step": 11022} +{"lr": 8.470405056405736e-05, "data_time": 0.0018187999725341798, "loss": 0.44320458769798277, "time": 0.6301448822021485, "epoch": 3, "memory": 44140, "step": 11122} +{"lr": 8.44448629692807e-05, "data_time": 0.001539158821105957, "loss": 0.45170451104640963, "time": 0.6304357528686524, "epoch": 3, "memory": 44140, "step": 11222} +{"lr": 8.418395583476941e-05, "data_time": 0.001814723014831543, "loss": 0.47137278914451597, "time": 0.6330671787261963, "epoch": 3, "memory": 44140, "step": 11322} +{"lr": 8.392134439714336e-05, "data_time": 0.0015060901641845703, "loss": 0.4962335705757141, "time": 0.633410906791687, "epoch": 3, "memory": 44140, "step": 11422} +{"lr": 8.365704399255133e-05, "data_time": 0.0016580343246459961, "loss": 0.44537240266799927, "time": 0.6299348592758178, "epoch": 3, "memory": 44140, "step": 11522} +{"lr": 8.339107005577562e-05, "data_time": 0.0017243623733520508, "loss": 0.4743430554866791, "time": 0.6315998554229736, "epoch": 3, "memory": 44140, "step": 11622} +{"lr": 8.31234381193305e-05, "data_time": 0.0016975641250610352, "loss": 0.46297845244407654, "time": 0.6316205024719238, "epoch": 3, "memory": 44140, "step": 11722} +{"lr": 8.285416381255504e-05, "data_time": 0.0023648977279663087, "loss": 0.4666441261768341, "time": 0.6337792873382568, "epoch": 3, "memory": 44140, "step": 11822} +{"lr": 8.258326286070047e-05, "data_time": 0.0014805555343627929, "loss": 0.4419327437877655, "time": 0.6341650247573852, "epoch": 3, "memory": 44140, "step": 11922} +{"lr": 8.231075108401202e-05, "data_time": 0.002084040641784668, "loss": 0.48647671937942505, "time": 0.6319159269332886, "epoch": 3, "memory": 44140, "step": 12022} +{"lr": 8.203664439680493e-05, "data_time": 0.0015587806701660156, "loss": 0.4488724946975708, "time": 0.6315492868423462, "epoch": 3, "memory": 44140, "step": 12122} +{"lr": 8.176095880653475e-05, "data_time": 0.0016756534576416015, "loss": 0.4664489895105362, "time": 0.6321599960327149, "epoch": 3, "memory": 44140, "step": 12222} +{"lr": 8.148371041286308e-05, "data_time": 0.003419804573059082, "loss": 0.4784762978553772, "time": 0.6355720520019531, "epoch": 3, "memory": 44140, "step": 12322} +{"accuracy/top1": 86.24034118652344, "data_time": 0.019893074035644533, "time": 1.2850286444028218, "step": 3} +{"lr": 8.117415423358443e-05, "data_time": 0.0019217491149902343, "loss": 0.4694810748100281, "time": 0.630733847618103, "epoch": 4, "memory": 44140, "step": 12433} +{"lr": 8.089366155759266e-05, "data_time": 0.0021483182907104494, "loss": 0.4692914724349976, "time": 0.6325457334518433, "epoch": 4, "memory": 44140, "step": 12533} +{"lr": 8.061165672717273e-05, "data_time": 0.001792001724243164, "loss": 0.4633800595998764, "time": 0.631955361366272, "epoch": 4, "memory": 44140, "step": 12633} +{"lr": 8.032815621102116e-05, "data_time": 0.0015488147735595703, "loss": 0.4640433967113495, "time": 0.631095290184021, "epoch": 4, "memory": 44140, "step": 12733} +{"lr": 8.004317656518042e-05, "data_time": 0.0019303321838378905, "loss": 0.4620550960302353, "time": 0.6313628911972046, "epoch": 4, "memory": 44140, "step": 12833} +{"lr": 7.975673443207223e-05, "data_time": 0.0015459299087524415, "loss": 0.4617542505264282, "time": 0.6321180105209351, "epoch": 4, "memory": 44140, "step": 12933} +{"lr": 7.946884653952537e-05, "data_time": 0.0017204999923706055, "loss": 0.4665647357702255, "time": 0.6372065544128418, "epoch": 4, "memory": 44140, "step": 13033} +{"lr": 7.91795296997992e-05, "data_time": 0.0017672538757324218, "loss": 0.45223588943481446, "time": 0.6319003343582154, "epoch": 4, "memory": 44140, "step": 13133} +{"lr": 7.888880080860151e-05, "data_time": 0.0016176462173461913, "loss": 0.502515995502472, "time": 0.631034541130066, "epoch": 4, "memory": 44140, "step": 13233} +{"lr": 7.859667684410205e-05, "data_time": 0.0017225503921508788, "loss": 0.45673911571502684, "time": 0.6312153577804566, "epoch": 4, "memory": 44140, "step": 13333} +{"lr": 7.830317486594082e-05, "data_time": 0.0016968727111816406, "loss": 0.4540987253189087, "time": 0.6358464002609253, "epoch": 4, "memory": 44140, "step": 13433} +{"lr": 7.800831201423207e-05, "data_time": 0.001589202880859375, "loss": 0.45457584857940675, "time": 0.6328500032424926, "epoch": 4, "memory": 44140, "step": 13533} +{"lr": 7.771210550856309e-05, "data_time": 0.0017958402633666993, "loss": 0.45891138911247253, "time": 0.6311897039413452, "epoch": 4, "memory": 44140, "step": 13633} +{"lr": 7.741457264698884e-05, "data_time": 0.0014514923095703125, "loss": 0.4681304693222046, "time": 0.6301425933837891, "epoch": 4, "memory": 44140, "step": 13733} +{"lr": 7.711573080502147e-05, "data_time": 0.0015256881713867187, "loss": 0.46723638772964476, "time": 0.6308440685272216, "epoch": 4, "memory": 44140, "step": 13833} +{"lr": 7.681559743461609e-05, "data_time": 0.0015051603317260743, "loss": 0.47811571061611174, "time": 0.6310416460037231, "epoch": 4, "memory": 44140, "step": 13933} +{"lr": 7.651419006315099e-05, "data_time": 0.0015030145645141602, "loss": 0.4668451607227325, "time": 0.631350326538086, "epoch": 4, "memory": 44140, "step": 14033} +{"lr": 7.621152629240453e-05, "data_time": 0.0016044378280639648, "loss": 0.4456111818552017, "time": 0.6320450782775879, "epoch": 4, "memory": 44140, "step": 14133} +{"lr": 7.590762379752719e-05, "data_time": 0.0016407966613769531, "loss": 0.4569500327110291, "time": 0.6321018695831299, "epoch": 4, "memory": 44140, "step": 14233} +{"lr": 7.560250032600911e-05, "data_time": 0.0016803503036499023, "loss": 0.4478110492229462, "time": 0.632465410232544, "epoch": 4, "memory": 44140, "step": 14333} +{"lr": 7.52961736966437e-05, "data_time": 0.00166778564453125, "loss": 0.4908573359251022, "time": 0.6317228078842163, "epoch": 4, "memory": 44140, "step": 14433} +{"lr": 7.498866179848732e-05, "data_time": 0.0017252683639526368, "loss": 0.4687069535255432, "time": 0.6344333410263061, "epoch": 4, "memory": 44140, "step": 14533} +{"lr": 7.467998258981415e-05, "data_time": 0.001628875732421875, "loss": 0.4527924507856369, "time": 0.632758903503418, "epoch": 4, "memory": 44140, "step": 14633} +{"lr": 7.437015409706804e-05, "data_time": 0.001986527442932129, "loss": 0.47117196321487426, "time": 0.6316200256347656, "epoch": 4, "memory": 44140, "step": 14733} +{"lr": 7.405919441380929e-05, "data_time": 0.00152587890625, "loss": 0.466316694021225, "time": 0.631846261024475, "epoch": 4, "memory": 44140, "step": 14833} +{"lr": 7.374712169965821e-05, "data_time": 0.002025461196899414, "loss": 0.46932312846183777, "time": 0.6320650339126587, "epoch": 4, "memory": 44140, "step": 14933} +{"lr": 7.343395417923461e-05, "data_time": 0.0015321016311645509, "loss": 0.4634749710559845, "time": 0.6335531711578369, "epoch": 4, "memory": 44140, "step": 15033} +{"lr": 7.311971014109342e-05, "data_time": 0.0016558647155761718, "loss": 0.4653380632400513, "time": 0.6328489065170289, "epoch": 4, "memory": 44140, "step": 15133} +{"lr": 7.280440793665682e-05, "data_time": 0.0016974925994873047, "loss": 0.4721198469400406, "time": 0.6314444780349732, "epoch": 4, "memory": 44140, "step": 15233} +{"lr": 7.248806597914234e-05, "data_time": 0.0015108108520507813, "loss": 0.45408596396446227, "time": 0.6307537078857421, "epoch": 4, "memory": 44140, "step": 15333} +{"lr": 7.217070274248765e-05, "data_time": 0.0018209695816040039, "loss": 0.47602528631687163, "time": 0.6314978122711181, "epoch": 4, "memory": 44140, "step": 15433} +{"lr": 7.185233676027203e-05, "data_time": 0.0015051126480102538, "loss": 0.4789700746536255, "time": 0.6314777374267578, "epoch": 4, "memory": 44140, "step": 15533} +{"lr": 7.153298662463338e-05, "data_time": 0.0015069007873535155, "loss": 0.45825580358505247, "time": 0.6326617002487183, "epoch": 4, "memory": 44140, "step": 15633} +{"lr": 7.121267098518291e-05, "data_time": 0.0014716386795043945, "loss": 0.4747273147106171, "time": 0.631275749206543, "epoch": 4, "memory": 44140, "step": 15733} +{"lr": 7.089140854791618e-05, "data_time": 0.0022616147994995116, "loss": 0.45051977038383484, "time": 0.6312662601470947, "epoch": 4, "memory": 44140, "step": 15833} +{"lr": 7.056921807412025e-05, "data_time": 0.0015029668807983398, "loss": 0.4630744159221649, "time": 0.6306298971176147, "epoch": 4, "memory": 44140, "step": 15933} +{"lr": 7.024611837927802e-05, "data_time": 0.0022447586059570314, "loss": 0.46296154260635375, "time": 0.6328754901885987, "epoch": 4, "memory": 44140, "step": 16033} +{"lr": 6.992212833196998e-05, "data_time": 0.0024309158325195312, "loss": 0.45962140560150144, "time": 0.6336203813552856, "epoch": 4, "memory": 44140, "step": 16133} +{"lr": 6.959726685277174e-05, "data_time": 0.0018636226654052735, "loss": 0.44816306233406067, "time": 0.632193922996521, "epoch": 4, "memory": 44140, "step": 16233} +{"lr": 6.927155291314949e-05, "data_time": 0.0018933773040771484, "loss": 0.4800023674964905, "time": 0.6304071426391602, "epoch": 4, "memory": 44140, "step": 16333} +{"lr": 6.894500553435204e-05, "data_time": 0.003003048896789551, "loss": 0.47201485335826876, "time": 0.6325242996215821, "epoch": 4, "memory": 44140, "step": 16433} +{"accuracy/top1": 87.52462005615234, "data_time": 0.018533003330230714, "time": 1.2799886405467986, "step": 4} +{"lr": 6.858158509719903e-05, "data_time": 0.0017267227172851562, "loss": 0.4378457725048065, "time": 0.6323457717895508, "epoch": 5, "memory": 44140, "step": 16544} +{"lr": 6.825334178880171e-05, "data_time": 0.002093982696533203, "loss": 0.45123682022094724, "time": 0.6316104412078858, "epoch": 5, "memory": 44140, "step": 16644} +{"lr": 6.792432450336645e-05, "data_time": 0.0016284465789794921, "loss": 0.4461707055568695, "time": 0.6308274984359741, "epoch": 5, "memory": 44140, "step": 16744} +{"lr": 6.759455245505286e-05, "data_time": 0.0019920825958251952, "loss": 0.46688292622566224, "time": 0.6320695161819458, "epoch": 5, "memory": 44140, "step": 16844} +{"lr": 6.726404490209753e-05, "data_time": 0.0018121719360351563, "loss": 0.46765890121459963, "time": 0.6316281318664551, "epoch": 5, "memory": 44140, "step": 16944} +{"lr": 6.693282114568972e-05, "data_time": 0.0021088361740112306, "loss": 0.4624310553073883, "time": 0.6323110103607178, "epoch": 5, "memory": 44140, "step": 17044} +{"lr": 6.660090052884388e-05, "data_time": 0.0015198707580566407, "loss": 0.4679421305656433, "time": 0.6343421936035156, "epoch": 5, "memory": 44140, "step": 17144} +{"lr": 6.626830243527016e-05, "data_time": 0.002634119987487793, "loss": 0.42645345330238343, "time": 0.6380135536193847, "epoch": 5, "memory": 44140, "step": 17244} +{"lr": 6.593504628824236e-05, "data_time": 0.0015419483184814452, "loss": 0.4788371235132217, "time": 0.6317576169967651, "epoch": 5, "memory": 44140, "step": 17344} +{"lr": 6.560115154946387e-05, "data_time": 0.0018663644790649415, "loss": 0.4343857169151306, "time": 0.6330663442611695, "epoch": 5, "memory": 44140, "step": 17444} +{"lr": 6.526663771793081e-05, "data_time": 0.0015479564666748048, "loss": 0.45125500559806825, "time": 0.6322322130203247, "epoch": 5, "memory": 44140, "step": 17544} +{"lr": 6.493152432879363e-05, "data_time": 0.0018260955810546875, "loss": 0.458109176158905, "time": 0.6336064100265503, "epoch": 5, "memory": 44140, "step": 17644} +{"lr": 6.459583095221596e-05, "data_time": 0.001648092269897461, "loss": 0.4756574332714081, "time": 0.630453372001648, "epoch": 5, "memory": 44140, "step": 17744} +{"lr": 6.425957719223197e-05, "data_time": 0.0021019935607910155, "loss": 0.46511170268058777, "time": 0.6312604427337647, "epoch": 5, "memory": 44140, "step": 17844} +{"lr": 6.392278268560153e-05, "data_time": 0.001582026481628418, "loss": 0.4709571301937103, "time": 0.6320904970169068, "epoch": 5, "memory": 44140, "step": 17944} +{"lr": 6.358546710066316e-05, "data_time": 0.0025523900985717773, "loss": 0.46106258034706116, "time": 0.6339879035949707, "epoch": 5, "memory": 44140, "step": 18044} +{"lr": 6.324765013618573e-05, "data_time": 0.0016246795654296874, "loss": 0.463119113445282, "time": 0.6309337854385376, "epoch": 5, "memory": 44140, "step": 18144} +{"lr": 6.290935152021818e-05, "data_time": 0.001953792572021484, "loss": 0.46339941024780273, "time": 0.6311191082000732, "epoch": 5, "memory": 44140, "step": 18244} +{"lr": 6.2570591008937e-05, "data_time": 0.001540398597717285, "loss": 0.4621423244476318, "time": 0.6392133951187133, "epoch": 5, "memory": 44140, "step": 18344} +{"lr": 6.223138838549294e-05, "data_time": 0.001991677284240723, "loss": 0.473195880651474, "time": 0.6316898107528687, "epoch": 5, "memory": 44140, "step": 18444} +{"lr": 6.18917634588553e-05, "data_time": 0.0016360759735107421, "loss": 0.46876311004161836, "time": 0.630797266960144, "epoch": 5, "memory": 44140, "step": 18544} +{"lr": 6.155173606265555e-05, "data_time": 0.0021683931350708007, "loss": 0.4630735218524933, "time": 0.6323578357696533, "epoch": 5, "memory": 44140, "step": 18644} +{"lr": 6.121132605402843e-05, "data_time": 0.0015950679779052734, "loss": 0.4456640243530273, "time": 0.632435941696167, "epoch": 5, "memory": 44140, "step": 18744} +{"lr": 6.087055331245287e-05, "data_time": 0.0019070625305175782, "loss": 0.4652566611766815, "time": 0.6335298776626587, "epoch": 5, "memory": 44140, "step": 18844} +{"lr": 6.05294377385911e-05, "data_time": 0.0014652967453002929, "loss": 0.4578544795513153, "time": 0.6316671133041382, "epoch": 5, "memory": 44140, "step": 18944} +{"lr": 6.0187999253126355e-05, "data_time": 0.0016525745391845702, "loss": 0.4557887017726898, "time": 0.6442740678787231, "epoch": 5, "memory": 44140, "step": 19044} +{"lr": 5.9846257795599195e-05, "data_time": 0.0017616033554077148, "loss": 0.4571655809879303, "time": 0.631689977645874, "epoch": 5, "memory": 44140, "step": 19144} +{"lr": 5.950423332324355e-05, "data_time": 0.0015979766845703124, "loss": 0.46781941652297976, "time": 0.6310057640075684, "epoch": 5, "memory": 44140, "step": 19244} +{"lr": 5.9161945809820855e-05, "data_time": 0.001575636863708496, "loss": 0.4621806740760803, "time": 0.6312469244003296, "epoch": 5, "memory": 44140, "step": 19344} +{"lr": 5.8819415244453975e-05, "data_time": 0.0014534711837768555, "loss": 0.48360012471675873, "time": 0.6315086364746094, "epoch": 5, "memory": 44140, "step": 19444} +{"lr": 5.8476661630459405e-05, "data_time": 0.001828908920288086, "loss": 0.4748052477836609, "time": 0.6316088199615478, "epoch": 5, "memory": 44140, "step": 19544} +{"lr": 5.81337049841796e-05, "data_time": 0.0015892505645751954, "loss": 0.4698820114135742, "time": 0.6308746814727784, "epoch": 5, "memory": 44140, "step": 19644} +{"lr": 5.779056533381373e-05, "data_time": 0.0015501976013183594, "loss": 0.4809012174606323, "time": 0.6297628164291382, "epoch": 5, "memory": 44140, "step": 19744} +{"lr": 5.744726271824817e-05, "data_time": 0.0015085697174072265, "loss": 0.46830612421035767, "time": 0.6310423612594604, "epoch": 5, "memory": 44140, "step": 19844} +{"lr": 5.710381718588622e-05, "data_time": 0.0015188694000244141, "loss": 0.44808252453804015, "time": 0.6319166421890259, "epoch": 5, "memory": 44140, "step": 19944} +{"lr": 5.676024879347741e-05, "data_time": 0.0020087003707885743, "loss": 0.47926547527313235, "time": 0.631207799911499, "epoch": 5, "memory": 44140, "step": 20044} +{"lr": 5.6416577604945966e-05, "data_time": 0.0016213417053222655, "loss": 0.47091892957687376, "time": 0.6306166887283325, "epoch": 5, "memory": 44140, "step": 20144} +{"lr": 5.6072823690219385e-05, "data_time": 0.0016350746154785156, "loss": 0.46497154235839844, "time": 0.6301866054534913, "epoch": 5, "memory": 44140, "step": 20244} +{"lr": 5.572900712405624e-05, "data_time": 0.0017303228378295898, "loss": 0.45170841813087464, "time": 0.6334760904312133, "epoch": 5, "memory": 44140, "step": 20344} +{"lr": 5.538514798487397e-05, "data_time": 0.0015020132064819335, "loss": 0.45783367156982424, "time": 0.6305166244506836, "epoch": 5, "memory": 44140, "step": 20444} +{"lr": 5.5041266353576e-05, "data_time": 0.0031407833099365234, "loss": 0.45622223913669585, "time": 0.6328708410263062, "epoch": 5, "memory": 44140, "step": 20544} +{"accuracy/top1": 88.29088592529297, "data_time": 0.019408297538757325, "time": 1.280727324883143, "step": 5} +{"lr": 5.4659555783035384e-05, "data_time": 0.0017912149429321288, "loss": 0.47510509490966796, "time": 0.6332332134246826, "epoch": 6, "memory": 44140, "step": 20655} +{"lr": 5.431569258426336e-05, "data_time": 0.0016559600830078126, "loss": 0.4886723875999451, "time": 0.6334224939346313, "epoch": 6, "memory": 44140, "step": 20755} +{"lr": 5.397186934810864e-05, "data_time": 0.0015335559844970703, "loss": 0.4472442388534546, "time": 0.6323338747024536, "epoch": 6, "memory": 44140, "step": 20855} +{"lr": 5.362810615337815e-05, "data_time": 0.0016561508178710937, "loss": 0.4417957067489624, "time": 0.6310176372528076, "epoch": 6, "memory": 44140, "step": 20955} +{"lr": 5.328442307537233e-05, "data_time": 0.0016846418380737304, "loss": 0.4762031018733978, "time": 0.6307538032531739, "epoch": 6, "memory": 44140, "step": 21055} +{"lr": 5.294084018471303e-05, "data_time": 0.0017998933792114258, "loss": 0.46232172250747683, "time": 0.630191969871521, "epoch": 6, "memory": 44140, "step": 21155} +{"lr": 5.259737754617132e-05, "data_time": 0.0018306016921997071, "loss": 0.48820924758911133, "time": 0.6305662631988526, "epoch": 6, "memory": 44140, "step": 21255} +{"lr": 5.2254055217495525e-05, "data_time": 0.0015924930572509765, "loss": 0.45944496989250183, "time": 0.6331133604049682, "epoch": 6, "memory": 44140, "step": 21355} +{"lr": 5.1910893248240283e-05, "data_time": 0.0017383813858032227, "loss": 0.4717793345451355, "time": 0.6329855918884277, "epoch": 6, "memory": 44140, "step": 21455} +{"lr": 5.156791167859532e-05, "data_time": 0.0017553329467773437, "loss": 0.47676384449005127, "time": 0.6315877914428711, "epoch": 6, "memory": 44140, "step": 21555} +{"lr": 5.122513053821533e-05, "data_time": 0.0015356063842773438, "loss": 0.4817984700202942, "time": 0.6315043449401856, "epoch": 6, "memory": 44140, "step": 21655} +{"lr": 5.0882569845050173e-05, "data_time": 0.0015803813934326173, "loss": 0.46333622336387636, "time": 0.6333118438720703, "epoch": 6, "memory": 44140, "step": 21755} +{"lr": 5.054024960417592e-05, "data_time": 0.0015550613403320312, "loss": 0.4572906881570816, "time": 0.6306675672531128, "epoch": 6, "memory": 44140, "step": 21855} +{"lr": 5.0198189806626494e-05, "data_time": 0.0014666318893432617, "loss": 0.46266697645187377, "time": 0.6345197439193726, "epoch": 6, "memory": 44140, "step": 21955} +{"lr": 4.985641042822647e-05, "data_time": 0.0018036365509033203, "loss": 0.45842320919036866, "time": 0.631346607208252, "epoch": 6, "memory": 44140, "step": 22055} +{"lr": 4.951493142842412e-05, "data_time": 0.0015744924545288085, "loss": 0.4805125206708908, "time": 0.6314605951309205, "epoch": 6, "memory": 44140, "step": 22155} +{"lr": 4.917377274912603e-05, "data_time": 0.001524043083190918, "loss": 0.47288902997970583, "time": 0.6355272769927979, "epoch": 6, "memory": 44140, "step": 22255} +{"lr": 4.88329543135327e-05, "data_time": 0.0014655351638793944, "loss": 0.4488997280597687, "time": 0.6312304258346557, "epoch": 6, "memory": 44140, "step": 22355} +{"lr": 4.849249602497467e-05, "data_time": 0.0015840530395507812, "loss": 0.46450841426849365, "time": 0.6303846597671509, "epoch": 6, "memory": 44140, "step": 22455} +{"lr": 4.8152417765750456e-05, "data_time": 0.001480412483215332, "loss": 0.46635721921920775, "time": 0.632472038269043, "epoch": 6, "memory": 44140, "step": 22555} +{"lr": 4.7812739395965355e-05, "data_time": 0.0017144441604614257, "loss": 0.4644902467727661, "time": 0.6333511114120484, "epoch": 6, "memory": 44140, "step": 22655} +{"lr": 4.74734807523717e-05, "data_time": 0.0014958143234252929, "loss": 0.47039506435394285, "time": 0.6341737985610962, "epoch": 6, "memory": 44140, "step": 22755} +{"lr": 4.7134661647210416e-05, "data_time": 0.0018420934677124024, "loss": 0.4463201522827148, "time": 0.6463409900665283, "epoch": 6, "memory": 44140, "step": 22855} +{"lr": 4.679630186705388e-05, "data_time": 0.0016364336013793945, "loss": 0.4845757484436035, "time": 0.6325761079788208, "epoch": 6, "memory": 44140, "step": 22955} +{"lr": 4.6458421171650644e-05, "data_time": 0.0016704320907592774, "loss": 0.4898410618305206, "time": 0.6323488473892211, "epoch": 6, "memory": 44140, "step": 23055} +{"lr": 4.6121039292771295e-05, "data_time": 0.0015418291091918944, "loss": 0.4825673043727875, "time": 0.631195855140686, "epoch": 6, "memory": 44140, "step": 23155} +{"lr": 4.5784175933056197e-05, "data_time": 0.0015616893768310546, "loss": 0.4758934319019318, "time": 0.6317502737045289, "epoch": 6, "memory": 44140, "step": 23255} +{"lr": 4.544785076486494e-05, "data_time": 0.0014406442642211914, "loss": 0.46538543701171875, "time": 0.6338924407958985, "epoch": 6, "memory": 44140, "step": 23355} +{"lr": 4.5112083429127475e-05, "data_time": 0.0016784191131591797, "loss": 0.45752432346343996, "time": 0.6312843322753906, "epoch": 6, "memory": 44140, "step": 23455} +{"lr": 4.477689353419706e-05, "data_time": 0.002481961250305176, "loss": 0.4551388740539551, "time": 0.6350880146026612, "epoch": 6, "memory": 44140, "step": 23555} +{"lr": 4.4442300654705266e-05, "data_time": 0.002202892303466797, "loss": 0.48264811038970945, "time": 0.6322339057922364, "epoch": 6, "memory": 44140, "step": 23655} +{"lr": 4.410832433041866e-05, "data_time": 0.001952815055847168, "loss": 0.416254723072052, "time": 0.6314322233200074, "epoch": 6, "memory": 44140, "step": 23755} +{"lr": 4.377498406509792e-05, "data_time": 0.0017408132553100586, "loss": 0.43838741779327395, "time": 0.6314563035964966, "epoch": 6, "memory": 44140, "step": 23855} +{"lr": 4.344229932535875e-05, "data_time": 0.001574873924255371, "loss": 0.46265836954116824, "time": 0.6403370141983032, "epoch": 6, "memory": 44140, "step": 23955} +{"lr": 4.3110289539535115e-05, "data_time": 0.0015943050384521484, "loss": 0.4682974457740784, "time": 0.6343074083328247, "epoch": 6, "memory": 44140, "step": 24055} +{"lr": 4.2778974096544436e-05, "data_time": 0.002298569679260254, "loss": 0.4528957664966583, "time": 0.6313576698303223, "epoch": 6, "memory": 44140, "step": 24155} +{"lr": 4.2448372344755544e-05, "data_time": 0.0014259815216064453, "loss": 0.45468024015426634, "time": 0.6337838172912598, "epoch": 6, "memory": 44140, "step": 24255} +{"lr": 4.211850359085871e-05, "data_time": 0.0017948389053344727, "loss": 0.4562927335500717, "time": 0.6322719097137451, "epoch": 6, "memory": 44140, "step": 24355} +{"lr": 4.17893870987381e-05, "data_time": 0.0017693281173706056, "loss": 0.45234352350234985, "time": 0.6338937282562256, "epoch": 6, "memory": 44140, "step": 24455} +{"lr": 4.146104208834679e-05, "data_time": 0.0016025304794311523, "loss": 0.44231041669845583, "time": 0.634783124923706, "epoch": 6, "memory": 44140, "step": 24555} +{"lr": 4.113348773458442e-05, "data_time": 0.0030402421951293947, "loss": 0.4572556972503662, "time": 0.6370494604110718, "epoch": 6, "memory": 44140, "step": 24655} +{"accuracy/top1": 88.78241729736328, "data_time": 0.022437427441279092, "time": 1.2850611090660096, "step": 6} +{"lr": 4.0770851520835164e-05, "data_time": 0.0017620325088500977, "loss": 0.4570577025413513, "time": 0.6326758623123169, "epoch": 7, "memory": 44140, "step": 24766} +{"lr": 4.044502815742039e-05, "data_time": 0.0017591476440429687, "loss": 0.4358859181404114, "time": 0.632233715057373, "epoch": 7, "memory": 44140, "step": 24866} +{"lr": 4.012005478445896e-05, "data_time": 0.0015244007110595704, "loss": 0.46642338633537295, "time": 0.6319602489471435, "epoch": 7, "memory": 44140, "step": 24966} +{"lr": 3.9795950379951394e-05, "data_time": 0.0016994953155517578, "loss": 0.47227356433868406, "time": 0.6334277153015136, "epoch": 7, "memory": 44140, "step": 25066} +{"lr": 3.947273387115169e-05, "data_time": 0.001701688766479492, "loss": 0.46978639960289004, "time": 0.6347219705581665, "epoch": 7, "memory": 44140, "step": 25166} +{"lr": 3.9150424133462044e-05, "data_time": 0.0015353679656982422, "loss": 0.4644059896469116, "time": 0.6324105978012085, "epoch": 7, "memory": 44140, "step": 25266} +{"lr": 3.882903998933037e-05, "data_time": 0.002018308639526367, "loss": 0.44588848054409025, "time": 0.6323031663894654, "epoch": 7, "memory": 44140, "step": 25366} +{"lr": 3.8508600207151264e-05, "data_time": 0.0019781827926635743, "loss": 0.461130291223526, "time": 0.6324244976043701, "epoch": 7, "memory": 44140, "step": 25466} +{"lr": 3.818912350016978e-05, "data_time": 0.0016542673110961914, "loss": 0.4709395945072174, "time": 0.6339887857437134, "epoch": 7, "memory": 44140, "step": 25566} +{"lr": 3.787062852538883e-05, "data_time": 0.001529383659362793, "loss": 0.4548145353794098, "time": 0.6318777799606323, "epoch": 7, "memory": 44140, "step": 25666} +{"lr": 3.7553133882479346e-05, "data_time": 0.0023572683334350587, "loss": 0.47995511889457704, "time": 0.6329619407653808, "epoch": 7, "memory": 44140, "step": 25766} +{"lr": 3.723665811269437e-05, "data_time": 0.0018700361251831055, "loss": 0.4594398647546768, "time": 0.632764744758606, "epoch": 7, "memory": 44140, "step": 25866} +{"lr": 3.6921219697786176e-05, "data_time": 0.002150869369506836, "loss": 0.4604384511709213, "time": 0.6355153799057007, "epoch": 7, "memory": 44140, "step": 25966} +{"lr": 3.660683705892677e-05, "data_time": 0.0019325733184814454, "loss": 0.45717291831970214, "time": 0.6339921951293945, "epoch": 7, "memory": 44140, "step": 26066} +{"lr": 3.6293528555632435e-05, "data_time": 0.0020636796951293947, "loss": 0.45245722532272337, "time": 0.633061146736145, "epoch": 7, "memory": 44140, "step": 26166} +{"lr": 3.598131248469134e-05, "data_time": 0.0017244338989257813, "loss": 0.45272687673568723, "time": 0.632715368270874, "epoch": 7, "memory": 44140, "step": 26266} +{"lr": 3.567020707909507e-05, "data_time": 0.0016669511795043945, "loss": 0.5036671459674835, "time": 0.6316426038742066, "epoch": 7, "memory": 44140, "step": 26366} +{"lr": 3.5360230506973935e-05, "data_time": 0.001744699478149414, "loss": 0.4386345148086548, "time": 0.6319615125656128, "epoch": 7, "memory": 44140, "step": 26466} +{"lr": 3.5051400870535877e-05, "data_time": 0.0016463279724121093, "loss": 0.48083417415618895, "time": 0.6331605434417724, "epoch": 7, "memory": 44140, "step": 26566} +{"lr": 3.47437362050094e-05, "data_time": 0.0015377998352050781, "loss": 0.47480157017707825, "time": 0.6321899652481079, "epoch": 7, "memory": 44140, "step": 26666} +{"lr": 3.443725447759023e-05, "data_time": 0.0016852855682373048, "loss": 0.45800946950912474, "time": 0.6351692914962769, "epoch": 7, "memory": 44140, "step": 26766} +{"lr": 3.413197358639224e-05, "data_time": 0.0016905546188354492, "loss": 0.4890470147132874, "time": 0.6322234153747559, "epoch": 7, "memory": 44140, "step": 26866} +{"lr": 3.382791135940194e-05, "data_time": 0.0016348838806152343, "loss": 0.4506791472434998, "time": 0.6334895610809326, "epoch": 7, "memory": 44140, "step": 26966} +{"lr": 3.35250855534377e-05, "data_time": 0.00218658447265625, "loss": 0.47264742851257324, "time": 0.6329361438751221, "epoch": 7, "memory": 44140, "step": 27066} +{"lr": 3.3223513853112476e-05, "data_time": 0.0018648862838745116, "loss": 0.46374369859695436, "time": 0.6352895259857178, "epoch": 7, "memory": 44140, "step": 27166} +{"lr": 3.292321386980115e-05, "data_time": 0.0016809701919555664, "loss": 0.4720534086227417, "time": 0.6330942869186401, "epoch": 7, "memory": 44140, "step": 27266} +{"lr": 3.262420314061217e-05, "data_time": 0.001649951934814453, "loss": 0.4529055833816528, "time": 0.6318412065505982, "epoch": 7, "memory": 44140, "step": 27366} +{"lr": 3.2326499127363195e-05, "data_time": 0.001679682731628418, "loss": 0.489370596408844, "time": 0.6314186334609986, "epoch": 7, "memory": 44140, "step": 27466} +{"lr": 3.2030119215561613e-05, "data_time": 0.0014698028564453125, "loss": 0.46720961928367616, "time": 0.6314525365829468, "epoch": 7, "memory": 44140, "step": 27566} +{"lr": 3.173508071338896e-05, "data_time": 0.0018981695175170898, "loss": 0.48424118757247925, "time": 0.6329678535461426, "epoch": 7, "memory": 44140, "step": 27666} +{"lr": 3.144140085069037e-05, "data_time": 0.0017977476119995118, "loss": 0.4702145099639893, "time": 0.6332960605621338, "epoch": 7, "memory": 44140, "step": 27766} +{"lr": 3.114909677796814e-05, "data_time": 0.0015241622924804688, "loss": 0.46597014665603637, "time": 0.636191463470459, "epoch": 7, "memory": 44140, "step": 27866} +{"lr": 3.085818556538059e-05, "data_time": 0.0020096540451049806, "loss": 0.444723516702652, "time": 0.633235239982605, "epoch": 7, "memory": 44140, "step": 27966} +{"lr": 3.056868420174455e-05, "data_time": 0.0017461776733398438, "loss": 0.4772075653076172, "time": 0.6324796676635742, "epoch": 7, "memory": 44140, "step": 28066} +{"lr": 3.028060959354392e-05, "data_time": 0.0017737388610839845, "loss": 0.45855223536491396, "time": 0.6321711540222168, "epoch": 7, "memory": 44140, "step": 28166} +{"lr": 2.9993978563941876e-05, "data_time": 0.0016729116439819336, "loss": 0.45951600670814513, "time": 0.6341857671737671, "epoch": 7, "memory": 44140, "step": 28266} +{"lr": 2.9708807851798587e-05, "data_time": 0.001439213752746582, "loss": 0.5164131402969361, "time": 0.63258798122406, "epoch": 7, "memory": 44140, "step": 28366} +{"lr": 2.9425114110693717e-05, "data_time": 0.001916360855102539, "loss": 0.448526531457901, "time": 0.632703447341919, "epoch": 7, "memory": 44140, "step": 28466} +{"lr": 2.9142913907953844e-05, "data_time": 0.001385807991027832, "loss": 0.4666679859161377, "time": 0.6354238748550415, "epoch": 7, "memory": 44140, "step": 28566} +{"lr": 2.8862223723684912e-05, "data_time": 0.0016641616821289062, "loss": 0.4680193066596985, "time": 0.6330220699310303, "epoch": 7, "memory": 44140, "step": 28666} +{"lr": 2.8583059949809854e-05, "data_time": 0.002976202964782715, "loss": 0.4719451904296875, "time": 0.6387584209442139, "epoch": 7, "memory": 44140, "step": 28766} +{"accuracy/top1": 88.94461059570312, "data_time": 0.019816283384958902, "time": 1.2822944482167562, "step": 7} +{"lr": 2.827499545208893e-05, "data_time": 0.0020222902297973634, "loss": 0.4786592602729797, "time": 0.6311656951904296, "epoch": 8, "memory": 44140, "step": 28877} +{"lr": 2.7999105784987406e-05, "data_time": 0.0015462875366210938, "loss": 0.4444415599107742, "time": 0.6323636054992676, "epoch": 8, "memory": 44140, "step": 28977} +{"lr": 2.772479293318711e-05, "data_time": 0.001770353317260742, "loss": 0.4803554952144623, "time": 0.6327940464019776, "epoch": 8, "memory": 44140, "step": 29077} +{"lr": 2.7452072916183603e-05, "data_time": 0.0015571832656860352, "loss": 0.4572782635688782, "time": 0.6335251569747925, "epoch": 8, "memory": 44140, "step": 29177} +{"lr": 2.7180961660452883e-05, "data_time": 0.0015721321105957031, "loss": 0.42625134587287905, "time": 0.6325462818145752, "epoch": 8, "memory": 44140, "step": 29277} +{"lr": 2.6911474998521622e-05, "data_time": 0.0015975236892700195, "loss": 0.46394469738006594, "time": 0.6334802627563476, "epoch": 8, "memory": 44140, "step": 29377} +{"lr": 2.664362866804239e-05, "data_time": 0.0015463590621948241, "loss": 0.48913257718086245, "time": 0.6320464134216308, "epoch": 8, "memory": 44140, "step": 29477} +{"lr": 2.637743831087459e-05, "data_time": 0.0017019033432006836, "loss": 0.45803717374801634, "time": 0.6317750215530396, "epoch": 8, "memory": 44140, "step": 29577} +{"lr": 2.6112919472171114e-05, "data_time": 0.0017027854919433594, "loss": 0.4668755173683167, "time": 0.6310395240783692, "epoch": 8, "memory": 44140, "step": 29677} +{"lr": 2.5850087599470434e-05, "data_time": 0.0017415523529052735, "loss": 0.4508780658245087, "time": 0.6319621562957763, "epoch": 8, "memory": 44140, "step": 29777} +{"lr": 2.5588958041794503e-05, "data_time": 0.0015781164169311524, "loss": 0.48173831701278685, "time": 0.6311343431472778, "epoch": 8, "memory": 44140, "step": 29877} +{"lr": 2.5329546048752453e-05, "data_time": 0.001935434341430664, "loss": 0.43395198583602906, "time": 0.6341276645660401, "epoch": 8, "memory": 44140, "step": 29977} +{"lr": 2.5071866769649806e-05, "data_time": 0.0015177488327026366, "loss": 0.46916183829307556, "time": 0.6306435585021972, "epoch": 8, "memory": 44140, "step": 30077} +{"lr": 2.4815935252604222e-05, "data_time": 0.0015572309494018555, "loss": 0.45433741211891177, "time": 0.6330373048782348, "epoch": 8, "memory": 44140, "step": 30177} +{"lr": 2.4561766443666384e-05, "data_time": 0.0016353845596313477, "loss": 0.46527657508850095, "time": 0.6324336051940918, "epoch": 8, "memory": 44140, "step": 30277} +{"lr": 2.4309375185947164e-05, "data_time": 0.0017819166183471679, "loss": 0.4976585924625397, "time": 0.6327979564666748, "epoch": 8, "memory": 44140, "step": 30377} +{"lr": 2.40587762187509e-05, "data_time": 0.002453899383544922, "loss": 0.48123987913131716, "time": 0.6336788415908814, "epoch": 8, "memory": 44140, "step": 30477} +{"lr": 2.3809984176714634e-05, "data_time": 0.0015416860580444336, "loss": 0.45979909896850585, "time": 0.6320846080780029, "epoch": 8, "memory": 44140, "step": 30577} +{"lr": 2.3563013588953406e-05, "data_time": 0.002267742156982422, "loss": 0.47019734382629397, "time": 0.6399927854537963, "epoch": 8, "memory": 44140, "step": 30677} +{"lr": 2.331787887821182e-05, "data_time": 0.0016018152236938477, "loss": 0.46065608263015745, "time": 0.6316389322280884, "epoch": 8, "memory": 44140, "step": 30777} +{"lr": 2.307459436002179e-05, "data_time": 0.0015846729278564454, "loss": 0.4970451831817627, "time": 0.6321336984634399, "epoch": 8, "memory": 44140, "step": 30877} +{"lr": 2.2833174241866527e-05, "data_time": 0.0015520811080932616, "loss": 0.4677172422409058, "time": 0.6325764894485474, "epoch": 8, "memory": 44140, "step": 30977} +{"lr": 2.259363262235086e-05, "data_time": 0.0015968561172485351, "loss": 0.4666645675897598, "time": 0.63511803150177, "epoch": 8, "memory": 44140, "step": 31077} +{"lr": 2.2355983490377735e-05, "data_time": 0.0017240524291992187, "loss": 0.48023677468299864, "time": 0.631782841682434, "epoch": 8, "memory": 44140, "step": 31177} +{"lr": 2.2120240724331546e-05, "data_time": 0.001915311813354492, "loss": 0.45828971862792967, "time": 0.6317096710205078, "epoch": 8, "memory": 44140, "step": 31277} +{"lr": 2.1886418091267445e-05, "data_time": 0.0017271995544433593, "loss": 0.4646733283996582, "time": 0.6314742088317871, "epoch": 8, "memory": 44140, "step": 31377} +{"lr": 2.1654529246107496e-05, "data_time": 0.0018069982528686524, "loss": 0.4614529192447662, "time": 0.6331280946731568, "epoch": 8, "memory": 44140, "step": 31477} +{"lr": 2.1424587730843184e-05, "data_time": 0.0016412019729614257, "loss": 0.45111280083656313, "time": 0.6334101676940918, "epoch": 8, "memory": 44140, "step": 31577} +{"lr": 2.1196606973744593e-05, "data_time": 0.001568770408630371, "loss": 0.48221046328544614, "time": 0.6346508979797363, "epoch": 8, "memory": 44140, "step": 31677} +{"lr": 2.0970600288576148e-05, "data_time": 0.0015542030334472657, "loss": 0.4559887021780014, "time": 0.6328295707702637, "epoch": 8, "memory": 44140, "step": 31777} +{"lr": 2.0746580873819274e-05, "data_time": 0.0016139030456542968, "loss": 0.4885957300662994, "time": 0.6321042537689209, "epoch": 8, "memory": 44140, "step": 31877} +{"lr": 2.0524561811901426e-05, "data_time": 0.0014272212982177734, "loss": 0.475443696975708, "time": 0.6333238363265992, "epoch": 8, "memory": 44140, "step": 31977} +{"lr": 2.0304556068432268e-05, "data_time": 0.001703333854675293, "loss": 0.45770376920700073, "time": 0.6328312158584595, "epoch": 8, "memory": 44140, "step": 32077} +{"lr": 2.0086576491446333e-05, "data_time": 0.001783895492553711, "loss": 0.4704034090042114, "time": 0.6323757171630859, "epoch": 8, "memory": 44140, "step": 32177} +{"lr": 1.987063581065291e-05, "data_time": 0.0021164417266845703, "loss": 0.4640349090099335, "time": 0.6318027973175049, "epoch": 8, "memory": 44140, "step": 32277} +{"lr": 1.965674663669245e-05, "data_time": 0.0016535520553588867, "loss": 0.46104816198348997, "time": 0.63171865940094, "epoch": 8, "memory": 44140, "step": 32377} +{"lr": 1.9444921460400247e-05, "data_time": 0.0017505884170532227, "loss": 0.4569315552711487, "time": 0.6315577745437622, "epoch": 8, "memory": 44140, "step": 32477} +{"lr": 1.923517265207697e-05, "data_time": 0.0016154766082763672, "loss": 0.4757853806018829, "time": 0.6331660270690918, "epoch": 8, "memory": 44140, "step": 32577} +{"lr": 1.9027512460766204e-05, "data_time": 0.0025146484375, "loss": 0.4882297396659851, "time": 0.6356138229370117, "epoch": 8, "memory": 44140, "step": 32677} +{"lr": 1.8821953013539136e-05, "data_time": 0.001882028579711914, "loss": 0.4461084544658661, "time": 0.6315230607986451, "epoch": 8, "memory": 44140, "step": 32777} +{"lr": 1.8618506314786433e-05, "data_time": 0.003214120864868164, "loss": 0.4858038812875748, "time": 0.6333858251571656, "epoch": 8, "memory": 44140, "step": 32877} +{"accuracy/top1": 89.2193374633789, "data_time": 0.01960983673731486, "time": 1.2817478557427724, "step": 8} +{"lr": 1.8395169032542028e-05, "data_time": 0.001709580421447754, "loss": 0.44001683592796326, "time": 0.6302437543869018, "epoch": 9, "memory": 44140, "step": 32988} +{"lr": 1.819621906460573e-05, "data_time": 0.002226090431213379, "loss": 0.45350052416324615, "time": 0.6339422464370728, "epoch": 9, "memory": 44140, "step": 33088} +{"lr": 1.799941838715077e-05, "data_time": 0.002341437339782715, "loss": 0.4614435076713562, "time": 0.6329886674880981, "epoch": 9, "memory": 44140, "step": 33188} +{"lr": 1.7804778493067625e-05, "data_time": 0.00311582088470459, "loss": 0.4876255303621292, "time": 0.6371673583984375, "epoch": 9, "memory": 44140, "step": 33288} +{"lr": 1.761231074905996e-05, "data_time": 0.0033722639083862303, "loss": 0.5050480723381042, "time": 0.6350805044174195, "epoch": 9, "memory": 44140, "step": 33388} +{"lr": 1.742202639498083e-05, "data_time": 0.0017731189727783203, "loss": 0.4838721454143524, "time": 0.6342012643814087, "epoch": 9, "memory": 44140, "step": 33488} +{"lr": 1.7233936543176347e-05, "data_time": 0.0016247987747192382, "loss": 0.47088128328323364, "time": 0.6331941604614257, "epoch": 9, "memory": 44140, "step": 33588} +{"lr": 1.704805217783665e-05, "data_time": 0.0018124341964721679, "loss": 0.48395604491233823, "time": 0.6322825670242309, "epoch": 9, "memory": 44140, "step": 33688} +{"lr": 1.68643841543545e-05, "data_time": 0.0016111373901367188, "loss": 0.4804447889328003, "time": 0.6329972982406616, "epoch": 9, "memory": 44140, "step": 33788} +{"lr": 1.6682943198691327e-05, "data_time": 0.002145981788635254, "loss": 0.46445180773735045, "time": 0.634234094619751, "epoch": 9, "memory": 44140, "step": 33888} +{"lr": 1.6503739906750822e-05, "data_time": 0.0018756866455078125, "loss": 0.4758709967136383, "time": 0.6345801591873169, "epoch": 9, "memory": 44140, "step": 33988} +{"lr": 1.6326784743760187e-05, "data_time": 0.0015656232833862304, "loss": 0.4649648368358612, "time": 0.633787989616394, "epoch": 9, "memory": 44140, "step": 34088} +{"lr": 1.6152088043659013e-05, "data_time": 0.0017421722412109375, "loss": 0.4618089526891708, "time": 0.6334868907928467, "epoch": 9, "memory": 44140, "step": 34188} +{"lr": 1.59796600084957e-05, "data_time": 0.001541447639465332, "loss": 0.46781880855560304, "time": 0.6310805797576904, "epoch": 9, "memory": 44140, "step": 34288} +{"lr": 1.5809510707831687e-05, "data_time": 0.001674985885620117, "loss": 0.4476747512817383, "time": 0.6365415573120117, "epoch": 9, "memory": 44140, "step": 34388} +{"lr": 1.5641650078153473e-05, "data_time": 0.0016707897186279297, "loss": 0.46122092604637144, "time": 0.6341193914413452, "epoch": 9, "memory": 44140, "step": 34488} +{"lr": 1.547608792229235e-05, "data_time": 0.0015656232833862304, "loss": 0.4754705190658569, "time": 0.6324231624603271, "epoch": 9, "memory": 44140, "step": 34588} +{"lr": 1.5312833908851794e-05, "data_time": 0.0017354011535644532, "loss": 0.48189482688903806, "time": 0.6311715364456176, "epoch": 9, "memory": 44140, "step": 34688} +{"lr": 1.5151897571642958e-05, "data_time": 0.001479506492614746, "loss": 0.46493968963623045, "time": 0.6299102306365967, "epoch": 9, "memory": 44140, "step": 34788} +{"lr": 1.4993288309127877e-05, "data_time": 0.001821756362915039, "loss": 0.4356079876422882, "time": 0.630953335762024, "epoch": 9, "memory": 44140, "step": 34888} +{"lr": 1.4837015383870593e-05, "data_time": 0.001851367950439453, "loss": 0.4493475317955017, "time": 0.6351421356201172, "epoch": 9, "memory": 44140, "step": 34988} +{"lr": 1.4683087921996264e-05, "data_time": 0.0014225006103515624, "loss": 0.4474563956260681, "time": 0.6318857669830322, "epoch": 9, "memory": 44140, "step": 35088} +{"lr": 1.453151491265819e-05, "data_time": 0.001576542854309082, "loss": 0.462446403503418, "time": 0.6326565742492676, "epoch": 9, "memory": 44140, "step": 35188} +{"lr": 1.4382305207512847e-05, "data_time": 0.002078413963317871, "loss": 0.4956405937671661, "time": 0.631881856918335, "epoch": 9, "memory": 44140, "step": 35288} +{"lr": 1.423546752020298e-05, "data_time": 0.0015061855316162109, "loss": 0.4497751295566559, "time": 0.6314989328384399, "epoch": 9, "memory": 44140, "step": 35388} +{"lr": 1.4091010425848762e-05, "data_time": 0.0015552520751953125, "loss": 0.47529215812683107, "time": 0.630776333808899, "epoch": 9, "memory": 44140, "step": 35488} +{"lr": 1.3948942360546966e-05, "data_time": 0.0016089916229248048, "loss": 0.4584742307662964, "time": 0.6311586618423461, "epoch": 9, "memory": 44140, "step": 35588} +{"lr": 1.3809271620878346e-05, "data_time": 0.0016643047332763673, "loss": 0.44769375324249266, "time": 0.6308555126190185, "epoch": 9, "memory": 44140, "step": 35688} +{"lr": 1.3672006363423091e-05, "data_time": 0.0017024517059326173, "loss": 0.476364403963089, "time": 0.6311171293258667, "epoch": 9, "memory": 44140, "step": 35788} +{"lr": 1.3537154604284566e-05, "data_time": 0.0016996622085571288, "loss": 0.4687406361103058, "time": 0.6304677248001098, "epoch": 9, "memory": 44140, "step": 35888} +{"lr": 1.3404724218621087e-05, "data_time": 0.001668858528137207, "loss": 0.4763421356678009, "time": 0.6316363573074341, "epoch": 9, "memory": 44140, "step": 35988} +{"lr": 1.3274722940186121e-05, "data_time": 0.0016004323959350586, "loss": 0.4671560525894165, "time": 0.6304128646850586, "epoch": 9, "memory": 44140, "step": 36088} +{"lr": 1.3147158360876541e-05, "data_time": 0.001594233512878418, "loss": 0.45514750480651855, "time": 0.6327695369720459, "epoch": 9, "memory": 44140, "step": 36188} +{"lr": 1.3022037930289352e-05, "data_time": 0.0017365694046020507, "loss": 0.44409971237182616, "time": 0.6322862863540649, "epoch": 9, "memory": 44140, "step": 36288} +{"lr": 1.2899368955286627e-05, "data_time": 0.0015902042388916016, "loss": 0.4862985908985138, "time": 0.6307955503463745, "epoch": 9, "memory": 44140, "step": 36388} +{"lr": 1.2779158599568751e-05, "data_time": 0.0017532587051391601, "loss": 0.44753718972206114, "time": 0.6334474325180054, "epoch": 9, "memory": 44140, "step": 36488} +{"lr": 1.2661413883256134e-05, "data_time": 0.001761150360107422, "loss": 0.46288397908210754, "time": 0.6308752059936523, "epoch": 9, "memory": 44140, "step": 36588} +{"lr": 1.2546141682479203e-05, "data_time": 0.0016102313995361329, "loss": 0.47072811126708985, "time": 0.6337591171264648, "epoch": 9, "memory": 44140, "step": 36688} +{"lr": 1.2433348728976846e-05, "data_time": 0.0015789031982421874, "loss": 0.44911747574806216, "time": 0.6305293321609498, "epoch": 9, "memory": 44140, "step": 36788} +{"lr": 1.2323041609703344e-05, "data_time": 0.0022758722305297853, "loss": 0.45526446104049684, "time": 0.6349687099456787, "epoch": 9, "memory": 44140, "step": 36888} +{"lr": 1.2215226766443614e-05, "data_time": 0.0030605554580688476, "loss": 0.44789217710494994, "time": 0.6341535806655884, "epoch": 9, "memory": 44140, "step": 36988} +{"accuracy/top1": 89.41959381103516, "data_time": 0.01953543027242025, "time": 1.281510070959727, "step": 9} +{"lr": 1.209847850894437e-05, "data_time": 0.001443958282470703, "loss": 0.4737401843070984, "time": 0.6327221870422364, "epoch": 10, "memory": 44140, "step": 37099} +{"lr": 1.1995942849201055e-05, "data_time": 0.0015955448150634765, "loss": 0.4663243889808655, "time": 0.6334791421890259, "epoch": 10, "memory": 44140, "step": 37199} +{"lr": 1.189591856759183e-05, "data_time": 0.001537179946899414, "loss": 0.46519347429275515, "time": 0.6322403669357299, "epoch": 10, "memory": 44140, "step": 37299} +{"lr": 1.1798411505398016e-05, "data_time": 0.0014942169189453125, "loss": 0.479831725358963, "time": 0.6337087154388428, "epoch": 10, "memory": 44140, "step": 37399} +{"lr": 1.1703427356898678e-05, "data_time": 0.0017148494720458985, "loss": 0.4661745309829712, "time": 0.6314796447753906, "epoch": 10, "memory": 44140, "step": 37499} +{"lr": 1.1610971669038223e-05, "data_time": 0.001622915267944336, "loss": 0.4556196093559265, "time": 0.6319317579269409, "epoch": 10, "memory": 44140, "step": 37599} +{"lr": 1.1521049841102396e-05, "data_time": 0.0015362024307250977, "loss": 0.45149661004543307, "time": 0.632202935218811, "epoch": 10, "memory": 44140, "step": 37699} +{"lr": 1.1433667124402997e-05, "data_time": 0.0016666173934936524, "loss": 0.4552126169204712, "time": 0.6328606843948364, "epoch": 10, "memory": 44140, "step": 37799} +{"lr": 1.1348828621971181e-05, "data_time": 0.0016186237335205078, "loss": 0.4752184867858887, "time": 0.6327661752700806, "epoch": 10, "memory": 44140, "step": 37899} +{"lr": 1.1266539288259486e-05, "data_time": 0.0014984846115112306, "loss": 0.48326406478881834, "time": 0.6329694747924804, "epoch": 10, "memory": 44140, "step": 37999} +{"lr": 1.1186803928852477e-05, "data_time": 0.0016207218170166016, "loss": 0.4649399757385254, "time": 0.6346088171005249, "epoch": 10, "memory": 44140, "step": 38099} +{"lr": 1.1109627200186125e-05, "data_time": 0.0017227411270141601, "loss": 0.5148457169532776, "time": 0.6336653470993042, "epoch": 10, "memory": 44140, "step": 38199} +{"lr": 1.1035013609275835e-05, "data_time": 0.00175931453704834, "loss": 0.4879583239555359, "time": 0.6326696634292602, "epoch": 10, "memory": 44140, "step": 38299} +{"lr": 1.0962967513453298e-05, "data_time": 0.001738119125366211, "loss": 0.4747924208641052, "time": 0.6331334590911866, "epoch": 10, "memory": 44140, "step": 38399} +{"lr": 1.0893493120111999e-05, "data_time": 0.0015506267547607422, "loss": 0.45279773473739626, "time": 0.6357613563537597, "epoch": 10, "memory": 44140, "step": 38499} +{"lr": 1.0826594486461509e-05, "data_time": 0.0016738653182983398, "loss": 0.452606725692749, "time": 0.6351911067962647, "epoch": 10, "memory": 44140, "step": 38599} +{"lr": 1.0762275519290557e-05, "data_time": 0.0018566370010375977, "loss": 0.45806434750556946, "time": 0.6325887203216553, "epoch": 10, "memory": 44140, "step": 38699} +{"lr": 1.0700539974738862e-05, "data_time": 0.002175045013427734, "loss": 0.46723602414131166, "time": 0.6351238250732422, "epoch": 10, "memory": 44140, "step": 38799} +{"lr": 1.064139145807783e-05, "data_time": 0.0019116640090942384, "loss": 0.45460622310638427, "time": 0.6343905925750732, "epoch": 10, "memory": 44140, "step": 38899} +{"lr": 1.0584833423499947e-05, "data_time": 0.002245640754699707, "loss": 0.4611483097076416, "time": 0.6363897562026978, "epoch": 10, "memory": 44140, "step": 38999} +{"lr": 1.0530869173917099e-05, "data_time": 0.001709294319152832, "loss": 0.4460208684206009, "time": 0.6381627798080445, "epoch": 10, "memory": 44140, "step": 39099} +{"lr": 1.0479501860767654e-05, "data_time": 0.0015112638473510742, "loss": 0.45862733125686644, "time": 0.634428858757019, "epoch": 10, "memory": 44140, "step": 39199} +{"lr": 1.043073448383247e-05, "data_time": 0.0015918254852294923, "loss": 0.480582058429718, "time": 0.633992338180542, "epoch": 10, "memory": 44140, "step": 39299} +{"lr": 1.038456989105968e-05, "data_time": 0.0014903783798217774, "loss": 0.46571108102798464, "time": 0.6325500965118408, "epoch": 10, "memory": 44140, "step": 39399} +{"lr": 1.034101077839838e-05, "data_time": 0.0015347957611083984, "loss": 0.47578980922698977, "time": 0.634426212310791, "epoch": 10, "memory": 44140, "step": 39499} +{"lr": 1.0300059689641194e-05, "data_time": 0.0014916419982910155, "loss": 0.47278679609298707, "time": 0.6343709230422974, "epoch": 10, "memory": 44140, "step": 39599} +{"lr": 1.0261719016275683e-05, "data_time": 0.002152252197265625, "loss": 0.45428922176361086, "time": 0.633298134803772, "epoch": 10, "memory": 44140, "step": 39699} +{"lr": 1.0225990997344757e-05, "data_time": 0.0024214267730712892, "loss": 0.45897480845451355, "time": 0.6334645748138428, "epoch": 10, "memory": 44140, "step": 39799} +{"lr": 1.019287771931587e-05, "data_time": 0.0018765926361083984, "loss": 0.4657438904047012, "time": 0.6348436355590821, "epoch": 10, "memory": 44140, "step": 39899} +{"lr": 1.0162381115959187e-05, "data_time": 0.001739835739135742, "loss": 0.46620587110519407, "time": 0.6328980445861816, "epoch": 10, "memory": 44140, "step": 39999} +{"lr": 1.0134502968234636e-05, "data_time": 0.0016477108001708984, "loss": 0.43456063866615297, "time": 0.6337140560150146, "epoch": 10, "memory": 44140, "step": 40099} +{"lr": 1.0109244904187918e-05, "data_time": 0.001727747917175293, "loss": 0.46338191628456116, "time": 0.6331120491027832, "epoch": 10, "memory": 44140, "step": 40199} +{"lr": 1.0086608398855446e-05, "data_time": 0.0015302419662475586, "loss": 0.4902698159217834, "time": 0.6338290452957154, "epoch": 10, "memory": 44140, "step": 40299} +{"lr": 1.0066594774178176e-05, "data_time": 0.0022404670715332033, "loss": 0.4802827000617981, "time": 0.6359631538391113, "epoch": 10, "memory": 44140, "step": 40399} +{"lr": 1.0049205198924413e-05, "data_time": 0.001586771011352539, "loss": 0.4385004460811615, "time": 0.6343741178512573, "epoch": 10, "memory": 44140, "step": 40499} +{"lr": 1.0034440688621577e-05, "data_time": 0.0019191980361938476, "loss": 0.46651833057403563, "time": 0.6348704099655151, "epoch": 10, "memory": 44140, "step": 40599} +{"lr": 1.0022302105496896e-05, "data_time": 0.0016030311584472657, "loss": 0.45719313621520996, "time": 0.6345669269561768, "epoch": 10, "memory": 44140, "step": 40699} +{"lr": 1.0012790158427017e-05, "data_time": 0.0015558719635009766, "loss": 0.47053351402282717, "time": 0.6333394289016724, "epoch": 10, "memory": 44140, "step": 40799} +{"lr": 1.000590540289662e-05, "data_time": 0.0016728639602661133, "loss": 0.46899487972259524, "time": 0.6342001914978027, "epoch": 10, "memory": 44140, "step": 40899} +{"lr": 1.0001648240966046e-05, "data_time": 0.0016131639480590821, "loss": 0.4478962004184723, "time": 0.639711880683899, "epoch": 10, "memory": 44140, "step": 40999} +{"lr": 1.0000018921247718e-05, "data_time": 0.0031923294067382813, "loss": 0.4709731459617615, "time": 0.6367666721343994, "epoch": 10, "memory": 44140, "step": 41099} +{"accuracy/top1": 89.48082733154297, "data_time": 0.018739457925160727, "time": 1.2785786350568136, "step": 10} diff --git a/clip_large_pretrain_4x256_all_lr1e-4/20230605_033556/vis_data/config.py b/clip_large_pretrain_4x256_all_lr1e-4/20230605_033556/vis_data/config.py new file mode 100644 index 0000000000000000000000000000000000000000..c8cac25c647871dc6370a72142a368b59b7fbed4 --- /dev/null +++ b/clip_large_pretrain_4x256_all_lr1e-4/20230605_033556/vis_data/config.py @@ -0,0 +1,356 @@ +optim_wrapper = dict( + optimizer=dict( + type='AdamW', lr=0.0001, weight_decay=0.3, _scope_='mmpretrain'), + paramwise_cfg=dict( + custom_keys=dict({ + '.cls_token': dict(decay_mult=0.0), + '.pos_embed': dict(decay_mult=0.0) + })), + type='AmpOptimWrapper', + dtype='bfloat16', + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=False, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=4096) +model = dict( + type='ImageClassifier', + backbone=dict( + frozen_stages=24, + type='VisionTransformer', + arch='l', + img_size=224, + patch_size=14, + drop_rate=0.1, + pre_norm=True, + final_norm=False, + init_cfg=dict( + type='Pretrained', + checkpoint='ckpt/openclip-ViT-L-14.pth', + prefix='backbone')), + neck=dict( + type='CLIPProjection', + in_channels=1024, + out_channels=768, + init_cfg=dict( + type='Pretrained', + checkpoint='ckpt/openclip-ViT-L-14.pth', + prefix='backbone')), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=768, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + init_cfg=None), + init_cfg=dict( + type='TruncNormal', layer=['Conv2d', 'Linear'], std=0.02, bias=0.0), + train_cfg=None) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=128, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/all1.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/all2.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/all3.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3fake8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/cc1m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3real8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=128, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV2-1-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/if-dpmsolver++-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=128, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV2-1-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/if-dpmsolver++-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/clip_large_pretrain_4x256_all_lr1e-4' diff --git a/clip_large_pretrain_4x256_all_lr1e-4/20230605_033556/vis_data/events.out.tfevents.1685907371.SH-IDC1-10-140-24-100.184616.0 b/clip_large_pretrain_4x256_all_lr1e-4/20230605_033556/vis_data/events.out.tfevents.1685907371.SH-IDC1-10-140-24-100.184616.0 new file mode 100644 index 0000000000000000000000000000000000000000..598176d585bbae8bcc3f5a59c20d5b4bf550d316 --- /dev/null +++ b/clip_large_pretrain_4x256_all_lr1e-4/20230605_033556/vis_data/events.out.tfevents.1685907371.SH-IDC1-10-140-24-100.184616.0 @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:97a19a6ff3ea2ce259d6e6de308cae482fadb4c3ef0867d03d1da57ba2e99d2a +size 3102083 diff --git a/clip_large_pretrain_4x256_all_lr1e-4/20230605_033556/vis_data/scalars.json b/clip_large_pretrain_4x256_all_lr1e-4/20230605_033556/vis_data/scalars.json new file mode 100644 index 0000000000000000000000000000000000000000..be6f0f3b722592091fea0ffe54a1a68189ee20f4 --- /dev/null +++ b/clip_large_pretrain_4x256_all_lr1e-4/20230605_033556/vis_data/scalars.json @@ -0,0 +1,420 @@ +{"lr": 9.999871217862936e-05, "data_time": 0.0025150299072265623, "loss": 0.5926630020141601, "time": 0.6296663045883178, "epoch": 1, "memory": 44138, "step": 100} +{"lr": 9.999479662539605e-05, "data_time": 0.0023998022079467773, "loss": 0.5558342933654785, "time": 0.629915189743042, "epoch": 1, "memory": 44138, "step": 200} +{"lr": 9.99882534375654e-05, "data_time": 0.002423381805419922, "loss": 0.5518157005310058, "time": 0.6311023950576782, "epoch": 1, "memory": 44138, "step": 300} +{"lr": 9.997908299725075e-05, "data_time": 0.002051043510437012, "loss": 0.5356977939605713, "time": 0.6294378757476806, "epoch": 1, "memory": 44138, "step": 400} +{"lr": 9.996728583999318e-05, "data_time": 0.00154569149017334, "loss": 0.5294235706329345, "time": 0.6362393140792847, "epoch": 1, "memory": 44138, "step": 500} +{"lr": 9.995286265473058e-05, "data_time": 0.0016171932220458984, "loss": 0.5345909118652343, "time": 0.6289902448654174, "epoch": 1, "memory": 44138, "step": 600} +{"lr": 9.993581428375725e-05, "data_time": 0.001757645606994629, "loss": 0.546266770362854, "time": 0.6302518606185913, "epoch": 1, "memory": 44138, "step": 700} +{"lr": 9.991614172267466e-05, "data_time": 0.001504993438720703, "loss": 0.5193750143051148, "time": 0.6299715280532837, "epoch": 1, "memory": 44138, "step": 800} +{"lr": 9.989384612033355e-05, "data_time": 0.001619243621826172, "loss": 0.5281996011734009, "time": 0.630258321762085, "epoch": 1, "memory": 44138, "step": 900} +{"lr": 9.986892877876658e-05, "data_time": 0.001715373992919922, "loss": 0.5351518869400025, "time": 0.6305138349533081, "epoch": 1, "memory": 44138, "step": 1000} +{"lr": 9.984139115311244e-05, "data_time": 0.002127528190612793, "loss": 0.4951824188232422, "time": 0.6299308776855469, "epoch": 1, "memory": 44138, "step": 1100} +{"lr": 9.981123485153081e-05, "data_time": 0.0020805835723876954, "loss": 0.5200381517410279, "time": 0.6306166172027587, "epoch": 1, "memory": 44138, "step": 1200} +{"lr": 9.977846163510851e-05, "data_time": 0.0016455411911010741, "loss": 0.5057559251785279, "time": 0.6302486896514893, "epoch": 1, "memory": 44138, "step": 1300} +{"lr": 9.974307341775653e-05, "data_time": 0.0017376422882080078, "loss": 0.5007327795028687, "time": 0.6307777881622314, "epoch": 1, "memory": 44138, "step": 1400} +{"lr": 9.970507226609833e-05, "data_time": 0.0014547348022460938, "loss": 0.5251185774803162, "time": 0.6306758403778077, "epoch": 1, "memory": 44138, "step": 1500} +{"lr": 9.966446039934934e-05, "data_time": 0.0017030715942382812, "loss": 0.5359284937381744, "time": 0.6308729887008667, "epoch": 1, "memory": 44138, "step": 1600} +{"lr": 9.962124018918683e-05, "data_time": 0.0017491579055786133, "loss": 0.524097228050232, "time": 0.6306417942047119, "epoch": 1, "memory": 44138, "step": 1700} +{"lr": 9.957541415961212e-05, "data_time": 0.002335810661315918, "loss": 0.5213017463684082, "time": 0.6322964191436767, "epoch": 1, "memory": 44138, "step": 1800} +{"lr": 9.952698498680267e-05, "data_time": 0.0018424272537231445, "loss": 0.5202134847640991, "time": 0.6315749406814575, "epoch": 1, "memory": 44138, "step": 1900} +{"lr": 9.947595549895583e-05, "data_time": 0.0017011404037475587, "loss": 0.5050365686416626, "time": 0.6295128107070923, "epoch": 1, "memory": 44138, "step": 2000} +{"lr": 9.942232867612417e-05, "data_time": 0.0015781402587890625, "loss": 0.5164885520935059, "time": 0.6307832717895507, "epoch": 1, "memory": 44138, "step": 2100} +{"lr": 9.93661076500406e-05, "data_time": 0.0018105506896972656, "loss": 0.5222069263458252, "time": 0.6295430421829223, "epoch": 1, "memory": 44138, "step": 2200} +{"lr": 9.930729570393622e-05, "data_time": 0.0017452716827392578, "loss": 0.5055970311164856, "time": 0.6308633804321289, "epoch": 1, "memory": 44138, "step": 2300} +{"lr": 9.924589627234824e-05, "data_time": 0.0019497632980346679, "loss": 0.49894664287567136, "time": 0.631049394607544, "epoch": 1, "memory": 44138, "step": 2400} +{"lr": 9.918191294091956e-05, "data_time": 0.0015043020248413086, "loss": 0.4959339380264282, "time": 0.6319576978683472, "epoch": 1, "memory": 44138, "step": 2500} +{"lr": 9.91153494461892e-05, "data_time": 0.0019474506378173828, "loss": 0.4856060862541199, "time": 0.6302881717681885, "epoch": 1, "memory": 44138, "step": 2600} +{"lr": 9.904620967537422e-05, "data_time": 0.0016234874725341796, "loss": 0.5094678044319153, "time": 0.6311283111572266, "epoch": 1, "memory": 44138, "step": 2700} +{"lr": 9.897449766614277e-05, "data_time": 0.0016669034957885742, "loss": 0.5114233791828156, "time": 0.6291088342666626, "epoch": 1, "memory": 44138, "step": 2800} +{"lr": 9.890021760637804e-05, "data_time": 0.0017839908599853516, "loss": 0.48612632751464846, "time": 0.6291269302368164, "epoch": 1, "memory": 44138, "step": 2900} +{"lr": 9.882337383393409e-05, "data_time": 0.0013880491256713866, "loss": 0.510037076473236, "time": 0.633272933959961, "epoch": 1, "memory": 44138, "step": 3000} +{"lr": 9.874397083638193e-05, "data_time": 0.001481318473815918, "loss": 0.48753076791763306, "time": 0.6303290605545044, "epoch": 1, "memory": 44138, "step": 3100} +{"lr": 9.866201325074828e-05, "data_time": 0.0018770217895507813, "loss": 0.5040024042129516, "time": 0.6320106744766235, "epoch": 1, "memory": 44138, "step": 3200} +{"lr": 9.857750586324393e-05, "data_time": 0.0016488313674926757, "loss": 0.4848734140396118, "time": 0.6298616647720336, "epoch": 1, "memory": 44138, "step": 3300} +{"lr": 9.84904536089848e-05, "data_time": 0.0014603137969970703, "loss": 0.5120719373226166, "time": 0.6313069105148316, "epoch": 1, "memory": 44138, "step": 3400} +{"lr": 9.840086157170351e-05, "data_time": 0.001648855209350586, "loss": 0.48729785680770876, "time": 0.6295938968658448, "epoch": 1, "memory": 44138, "step": 3500} +{"lr": 9.830873498345255e-05, "data_time": 0.0017504692077636719, "loss": 0.4966985821723938, "time": 0.6311012744903565, "epoch": 1, "memory": 44138, "step": 3600} +{"lr": 9.821407922429864e-05, "data_time": 0.001745462417602539, "loss": 0.5099364876747131, "time": 0.6309836387634278, "epoch": 1, "memory": 44138, "step": 3700} +{"lr": 9.811689982200877e-05, "data_time": 0.001806783676147461, "loss": 0.4760929882526398, "time": 0.632443618774414, "epoch": 1, "memory": 44138, "step": 3800} +{"lr": 9.801720245172702e-05, "data_time": 0.0017486810684204102, "loss": 0.4971849679946899, "time": 0.6320624351501465, "epoch": 1, "memory": 44138, "step": 3900} +{"lr": 9.791499293564358e-05, "data_time": 0.0014429569244384765, "loss": 0.5001693487167358, "time": 0.6314241647720337, "epoch": 1, "memory": 44138, "step": 4000} +{"lr": 9.781027724265439e-05, "data_time": 0.0048599720001220705, "loss": 0.47135170698165896, "time": 0.6343180179595947, "epoch": 1, "memory": 44138, "step": 4100} +{"accuracy/top1": 71.24604797363281, "data_time": 0.025956197947013277, "time": 1.3182958634961552, "step": 1} +{"lr": 9.769111539369434e-05, "data_time": 0.00241999626159668, "loss": 0.4788080334663391, "time": 0.6337323188781738, "epoch": 2, "memory": 44140, "step": 4211} +{"lr": 9.758113190911623e-05, "data_time": 0.0015765190124511718, "loss": 0.49256587624549864, "time": 0.6336324691772461, "epoch": 2, "memory": 44140, "step": 4311} +{"lr": 9.746866174465984e-05, "data_time": 0.0021776914596557616, "loss": 0.4984575003385544, "time": 0.63229341506958, "epoch": 2, "memory": 44140, "step": 4411} +{"lr": 9.735371146842894e-05, "data_time": 0.0018930912017822265, "loss": 0.4467486679553986, "time": 0.6317708253860473, "epoch": 2, "memory": 44140, "step": 4511} +{"lr": 9.723628779336255e-05, "data_time": 0.0014719247817993164, "loss": 0.49971413016319277, "time": 0.6333870410919189, "epoch": 2, "memory": 44140, "step": 4611} +{"lr": 9.711639757684265e-05, "data_time": 0.0014638185501098632, "loss": 0.45254039764404297, "time": 0.6319808959960938, "epoch": 2, "memory": 44140, "step": 4711} +{"lr": 9.699404782029395e-05, "data_time": 0.0017662525177001953, "loss": 0.4845359563827515, "time": 0.6309187889099122, "epoch": 2, "memory": 44140, "step": 4811} +{"lr": 9.686924566877498e-05, "data_time": 0.0018222808837890625, "loss": 0.4613297522068024, "time": 0.6334790229797364, "epoch": 2, "memory": 44140, "step": 4911} +{"lr": 9.674199841056078e-05, "data_time": 0.0015243291854858398, "loss": 0.48396073579788207, "time": 0.6328363418579102, "epoch": 2, "memory": 44140, "step": 5011} +{"lr": 9.661231347671705e-05, "data_time": 0.0021074771881103515, "loss": 0.4813476920127869, "time": 0.6317301511764526, "epoch": 2, "memory": 44140, "step": 5111} +{"lr": 9.64801984406668e-05, "data_time": 0.0014989137649536132, "loss": 0.47513251900672915, "time": 0.6317736387252808, "epoch": 2, "memory": 44140, "step": 5211} +{"lr": 9.634566101774737e-05, "data_time": 0.0014298200607299806, "loss": 0.47101333141326907, "time": 0.6307046175003052, "epoch": 2, "memory": 44140, "step": 5311} +{"lr": 9.62087090647604e-05, "data_time": 0.001444387435913086, "loss": 0.4979289352893829, "time": 0.6319366455078125, "epoch": 2, "memory": 44140, "step": 5411} +{"lr": 9.606935057951247e-05, "data_time": 0.001916360855102539, "loss": 0.4742088675498962, "time": 0.6350284576416015, "epoch": 2, "memory": 44140, "step": 5511} +{"lr": 9.592759370034884e-05, "data_time": 0.0018108129501342774, "loss": 0.4734244287014008, "time": 0.6316951751708985, "epoch": 2, "memory": 44140, "step": 5611} +{"lr": 9.578344670567721e-05, "data_time": 0.0020900964736938477, "loss": 0.46299545764923095, "time": 0.6309096097946167, "epoch": 2, "memory": 44140, "step": 5711} +{"lr": 9.563691801348495e-05, "data_time": 0.0016538619995117188, "loss": 0.4517266273498535, "time": 0.6313366651535034, "epoch": 2, "memory": 44140, "step": 5811} +{"lr": 9.548801618084738e-05, "data_time": 0.0017515897750854492, "loss": 0.4669777274131775, "time": 0.6331912040710449, "epoch": 2, "memory": 44140, "step": 5911} +{"lr": 9.533674990342786e-05, "data_time": 0.001575803756713867, "loss": 0.49224611520767214, "time": 0.6328452348709106, "epoch": 2, "memory": 44140, "step": 6011} +{"lr": 9.518312801497029e-05, "data_time": 0.0015989303588867187, "loss": 0.4860232055187225, "time": 0.6313059568405152, "epoch": 2, "memory": 44140, "step": 6111} +{"lr": 9.502715948678276e-05, "data_time": 0.0015105009078979492, "loss": 0.4802593559026718, "time": 0.6314626455307006, "epoch": 2, "memory": 44140, "step": 6211} +{"lr": 9.486885342721427e-05, "data_time": 0.0015629768371582032, "loss": 0.47859914898872374, "time": 0.631781530380249, "epoch": 2, "memory": 44140, "step": 6311} +{"lr": 9.470821908112202e-05, "data_time": 0.0019846439361572267, "loss": 0.4706738114356995, "time": 0.633806848526001, "epoch": 2, "memory": 44140, "step": 6411} +{"lr": 9.454526582933214e-05, "data_time": 0.0017047643661499024, "loss": 0.47435392141342164, "time": 0.6317571401596069, "epoch": 2, "memory": 44140, "step": 6511} +{"lr": 9.438000318809176e-05, "data_time": 0.0014998912811279297, "loss": 0.44744108319282533, "time": 0.6303221464157105, "epoch": 2, "memory": 44140, "step": 6611} +{"lr": 9.42124408085132e-05, "data_time": 0.0017312526702880859, "loss": 0.4806430101394653, "time": 0.6302761793136596, "epoch": 2, "memory": 44140, "step": 6711} +{"lr": 9.404258847601036e-05, "data_time": 0.0016814947128295898, "loss": 0.4817513346672058, "time": 0.6357363224029541, "epoch": 2, "memory": 44140, "step": 6811} +{"lr": 9.387045610972717e-05, "data_time": 0.001527714729309082, "loss": 0.4747046947479248, "time": 0.6315042495727539, "epoch": 2, "memory": 44140, "step": 6911} +{"lr": 9.369605376195838e-05, "data_time": 0.001538991928100586, "loss": 0.4685009777545929, "time": 0.6310405254364013, "epoch": 2, "memory": 44140, "step": 7011} +{"lr": 9.351939161756269e-05, "data_time": 0.0025190114974975586, "loss": 0.5208580732345581, "time": 0.6315462589263916, "epoch": 2, "memory": 44140, "step": 7111} +{"lr": 9.334047999336764e-05, "data_time": 0.0015179872512817382, "loss": 0.47420313358306887, "time": 0.6344522953033447, "epoch": 2, "memory": 44140, "step": 7211} +{"lr": 9.315932933756752e-05, "data_time": 0.001871800422668457, "loss": 0.4483091592788696, "time": 0.6340370893478393, "epoch": 2, "memory": 44140, "step": 7311} +{"lr": 9.297595022911287e-05, "data_time": 0.001640176773071289, "loss": 0.4723667621612549, "time": 0.6328441619873046, "epoch": 2, "memory": 44140, "step": 7411} +{"lr": 9.279035337709295e-05, "data_time": 0.0017405033111572265, "loss": 0.4809338450431824, "time": 0.6365816354751587, "epoch": 2, "memory": 44140, "step": 7511} +{"lr": 9.260254962011016e-05, "data_time": 0.002028656005859375, "loss": 0.4780664682388306, "time": 0.6328401327133178, "epoch": 2, "memory": 44140, "step": 7611} +{"lr": 9.241254992564702e-05, "data_time": 0.0027015209197998047, "loss": 0.4529854774475098, "time": 0.6335997104644775, "epoch": 2, "memory": 44140, "step": 7711} +{"lr": 9.22203653894259e-05, "data_time": 0.0018162488937377929, "loss": 0.4671182632446289, "time": 0.6313928604125977, "epoch": 2, "memory": 44140, "step": 7811} +{"lr": 9.202600723476098e-05, "data_time": 0.001990199089050293, "loss": 0.4647050857543945, "time": 0.632718563079834, "epoch": 2, "memory": 44140, "step": 7911} +{"lr": 9.182948681190268e-05, "data_time": 0.0018062114715576172, "loss": 0.4600569248199463, "time": 0.6309822559356689, "epoch": 2, "memory": 44140, "step": 8011} +{"lr": 9.163081559737497e-05, "data_time": 0.0019437789916992188, "loss": 0.4363451361656189, "time": 0.632062840461731, "epoch": 2, "memory": 44140, "step": 8111} +{"lr": 9.143000519330517e-05, "data_time": 0.0031851530075073242, "loss": 0.4707728177309036, "time": 0.635717511177063, "epoch": 2, "memory": 44140, "step": 8211} +{"accuracy/top1": 83.37223052978516, "data_time": 0.019402621189753215, "time": 1.2811894913514454, "step": 2} +{"lr": 9.120461478753376e-05, "data_time": 0.0016452312469482423, "loss": 0.46649948954582215, "time": 0.6308094978332519, "epoch": 3, "memory": 44140, "step": 8322} +{"lr": 9.099932932138134e-05, "data_time": 0.001672673225402832, "loss": 0.45427868366241453, "time": 0.6313965082168579, "epoch": 3, "memory": 44140, "step": 8422} +{"lr": 9.079194154362181e-05, "data_time": 0.0016325950622558595, "loss": 0.4859224081039429, "time": 0.6315812349319458, "epoch": 3, "memory": 44140, "step": 8522} +{"lr": 9.058246356541787e-05, "data_time": 0.0018900394439697265, "loss": 0.49377862811088563, "time": 0.6308974981307983, "epoch": 3, "memory": 44140, "step": 8622} +{"lr": 9.037090761999699e-05, "data_time": 0.002896904945373535, "loss": 0.4653532266616821, "time": 0.637605881690979, "epoch": 3, "memory": 44140, "step": 8722} +{"lr": 9.015728606193701e-05, "data_time": 0.0014890909194946289, "loss": 0.44080936908721924, "time": 0.6310546398162842, "epoch": 3, "memory": 44140, "step": 8822} +{"lr": 8.994161136644488e-05, "data_time": 0.0022495746612548827, "loss": 0.44987755119800565, "time": 0.6320571422576904, "epoch": 3, "memory": 44140, "step": 8922} +{"lr": 8.972389612862792e-05, "data_time": 0.0016071796417236328, "loss": 0.47653878927230836, "time": 0.6318201541900634, "epoch": 3, "memory": 44140, "step": 9022} +{"lr": 8.950415306275833e-05, "data_time": 0.001922607421875, "loss": 0.47898975014686584, "time": 0.6368085384368897, "epoch": 3, "memory": 44140, "step": 9122} +{"lr": 8.928239500153059e-05, "data_time": 0.0017330169677734375, "loss": 0.46713787913322447, "time": 0.632226037979126, "epoch": 3, "memory": 44140, "step": 9222} +{"lr": 8.905863489531232e-05, "data_time": 0.0019263505935668945, "loss": 0.4839877486228943, "time": 0.6313059568405152, "epoch": 3, "memory": 44140, "step": 9322} +{"lr": 8.883288581138775e-05, "data_time": 0.0015526056289672852, "loss": 0.4634073913097382, "time": 0.6307883262634277, "epoch": 3, "memory": 44140, "step": 9422} +{"lr": 8.86051609331947e-05, "data_time": 0.0015683889389038086, "loss": 0.4593377411365509, "time": 0.6345818996429443, "epoch": 3, "memory": 44140, "step": 9522} +{"lr": 8.837547355955467e-05, "data_time": 0.001513195037841797, "loss": 0.4885055422782898, "time": 0.6330514192581177, "epoch": 3, "memory": 44140, "step": 9622} +{"lr": 8.814383710389614e-05, "data_time": 0.0015676021575927734, "loss": 0.45213906168937684, "time": 0.6329608917236328, "epoch": 3, "memory": 44140, "step": 9722} +{"lr": 8.791026509347146e-05, "data_time": 0.0016211986541748047, "loss": 0.4513511657714844, "time": 0.6330824851989746, "epoch": 3, "memory": 44140, "step": 9822} +{"lr": 8.767477116856659e-05, "data_time": 0.0014519214630126954, "loss": 0.4570546567440033, "time": 0.6307456970214844, "epoch": 3, "memory": 44140, "step": 9922} +{"lr": 8.743736908170466e-05, "data_time": 0.0016186952590942383, "loss": 0.4745335578918457, "time": 0.6312720060348511, "epoch": 3, "memory": 44140, "step": 10022} +{"lr": 8.719807269684305e-05, "data_time": 0.001607060432434082, "loss": 0.4391482412815094, "time": 0.6320138692855835, "epoch": 3, "memory": 44140, "step": 10122} +{"lr": 8.695689598856356e-05, "data_time": 0.0014997005462646484, "loss": 0.463805741071701, "time": 0.6313628196716309, "epoch": 3, "memory": 44140, "step": 10222} +{"lr": 8.67138530412559e-05, "data_time": 0.0024975061416625975, "loss": 0.4336999773979187, "time": 0.6314276933670044, "epoch": 3, "memory": 44140, "step": 10322} +{"lr": 8.646895804829604e-05, "data_time": 0.0016403913497924805, "loss": 0.4856491804122925, "time": 0.6313858985900879, "epoch": 3, "memory": 44140, "step": 10422} +{"lr": 8.622222531121664e-05, "data_time": 0.00155792236328125, "loss": 0.46217219829559325, "time": 0.6314802169799805, "epoch": 3, "memory": 44140, "step": 10522} +{"lr": 8.597366923887211e-05, "data_time": 0.0015955448150634765, "loss": 0.446012145280838, "time": 0.6316116571426391, "epoch": 3, "memory": 44140, "step": 10622} +{"lr": 8.57233043465972e-05, "data_time": 0.0016615152359008788, "loss": 0.46443553566932677, "time": 0.6339049577713013, "epoch": 3, "memory": 44140, "step": 10722} +{"lr": 8.547114525535937e-05, "data_time": 0.0020400047302246093, "loss": 0.4640778601169586, "time": 0.6322259426116943, "epoch": 3, "memory": 44140, "step": 10822} +{"lr": 8.521720669090474e-05, "data_time": 0.002305126190185547, "loss": 0.4476841986179352, "time": 0.6317931175231933, "epoch": 3, "memory": 44140, "step": 10922} +{"lr": 8.496150348289821e-05, "data_time": 0.0015257835388183595, "loss": 0.4819492340087891, "time": 0.6307430028915405, "epoch": 3, "memory": 44140, "step": 11022} +{"lr": 8.470405056405736e-05, "data_time": 0.0018187999725341798, "loss": 0.44320458769798277, "time": 0.6301448822021485, "epoch": 3, "memory": 44140, "step": 11122} +{"lr": 8.44448629692807e-05, "data_time": 0.001539158821105957, "loss": 0.45170451104640963, "time": 0.6304357528686524, "epoch": 3, "memory": 44140, "step": 11222} +{"lr": 8.418395583476941e-05, "data_time": 0.001814723014831543, "loss": 0.47137278914451597, "time": 0.6330671787261963, "epoch": 3, "memory": 44140, "step": 11322} +{"lr": 8.392134439714336e-05, "data_time": 0.0015060901641845703, "loss": 0.4962335705757141, "time": 0.633410906791687, "epoch": 3, "memory": 44140, "step": 11422} +{"lr": 8.365704399255133e-05, "data_time": 0.0016580343246459961, "loss": 0.44537240266799927, "time": 0.6299348592758178, "epoch": 3, "memory": 44140, "step": 11522} +{"lr": 8.339107005577562e-05, "data_time": 0.0017243623733520508, "loss": 0.4743430554866791, "time": 0.6315998554229736, "epoch": 3, "memory": 44140, "step": 11622} +{"lr": 8.31234381193305e-05, "data_time": 0.0016975641250610352, "loss": 0.46297845244407654, "time": 0.6316205024719238, "epoch": 3, "memory": 44140, "step": 11722} +{"lr": 8.285416381255504e-05, "data_time": 0.0023648977279663087, "loss": 0.4666441261768341, "time": 0.6337792873382568, "epoch": 3, "memory": 44140, "step": 11822} +{"lr": 8.258326286070047e-05, "data_time": 0.0014805555343627929, "loss": 0.4419327437877655, "time": 0.6341650247573852, "epoch": 3, "memory": 44140, "step": 11922} +{"lr": 8.231075108401202e-05, "data_time": 0.002084040641784668, "loss": 0.48647671937942505, "time": 0.6319159269332886, "epoch": 3, "memory": 44140, "step": 12022} +{"lr": 8.203664439680493e-05, "data_time": 0.0015587806701660156, "loss": 0.4488724946975708, "time": 0.6315492868423462, "epoch": 3, "memory": 44140, "step": 12122} +{"lr": 8.176095880653475e-05, "data_time": 0.0016756534576416015, "loss": 0.4664489895105362, "time": 0.6321599960327149, "epoch": 3, "memory": 44140, "step": 12222} +{"lr": 8.148371041286308e-05, "data_time": 0.003419804573059082, "loss": 0.4784762978553772, "time": 0.6355720520019531, "epoch": 3, "memory": 44140, "step": 12322} +{"accuracy/top1": 86.24034118652344, "data_time": 0.019893074035644533, "time": 1.2850286444028218, "step": 3} +{"lr": 8.117415423358443e-05, "data_time": 0.0019217491149902343, "loss": 0.4694810748100281, "time": 0.630733847618103, "epoch": 4, "memory": 44140, "step": 12433} +{"lr": 8.089366155759266e-05, "data_time": 0.0021483182907104494, "loss": 0.4692914724349976, "time": 0.6325457334518433, "epoch": 4, "memory": 44140, "step": 12533} +{"lr": 8.061165672717273e-05, "data_time": 0.001792001724243164, "loss": 0.4633800595998764, "time": 0.631955361366272, "epoch": 4, "memory": 44140, "step": 12633} +{"lr": 8.032815621102116e-05, "data_time": 0.0015488147735595703, "loss": 0.4640433967113495, "time": 0.631095290184021, "epoch": 4, "memory": 44140, "step": 12733} +{"lr": 8.004317656518042e-05, "data_time": 0.0019303321838378905, "loss": 0.4620550960302353, "time": 0.6313628911972046, "epoch": 4, "memory": 44140, "step": 12833} +{"lr": 7.975673443207223e-05, "data_time": 0.0015459299087524415, "loss": 0.4617542505264282, "time": 0.6321180105209351, "epoch": 4, "memory": 44140, "step": 12933} +{"lr": 7.946884653952537e-05, "data_time": 0.0017204999923706055, "loss": 0.4665647357702255, "time": 0.6372065544128418, "epoch": 4, "memory": 44140, "step": 13033} +{"lr": 7.91795296997992e-05, "data_time": 0.0017672538757324218, "loss": 0.45223588943481446, "time": 0.6319003343582154, "epoch": 4, "memory": 44140, "step": 13133} +{"lr": 7.888880080860151e-05, "data_time": 0.0016176462173461913, "loss": 0.502515995502472, "time": 0.631034541130066, "epoch": 4, "memory": 44140, "step": 13233} +{"lr": 7.859667684410205e-05, "data_time": 0.0017225503921508788, "loss": 0.45673911571502684, "time": 0.6312153577804566, "epoch": 4, "memory": 44140, "step": 13333} +{"lr": 7.830317486594082e-05, "data_time": 0.0016968727111816406, "loss": 0.4540987253189087, "time": 0.6358464002609253, "epoch": 4, "memory": 44140, "step": 13433} +{"lr": 7.800831201423207e-05, "data_time": 0.001589202880859375, "loss": 0.45457584857940675, "time": 0.6328500032424926, "epoch": 4, "memory": 44140, "step": 13533} +{"lr": 7.771210550856309e-05, "data_time": 0.0017958402633666993, "loss": 0.45891138911247253, "time": 0.6311897039413452, "epoch": 4, "memory": 44140, "step": 13633} +{"lr": 7.741457264698884e-05, "data_time": 0.0014514923095703125, "loss": 0.4681304693222046, "time": 0.6301425933837891, "epoch": 4, "memory": 44140, "step": 13733} +{"lr": 7.711573080502147e-05, "data_time": 0.0015256881713867187, "loss": 0.46723638772964476, "time": 0.6308440685272216, "epoch": 4, "memory": 44140, "step": 13833} +{"lr": 7.681559743461609e-05, "data_time": 0.0015051603317260743, "loss": 0.47811571061611174, "time": 0.6310416460037231, "epoch": 4, "memory": 44140, "step": 13933} +{"lr": 7.651419006315099e-05, "data_time": 0.0015030145645141602, "loss": 0.4668451607227325, "time": 0.631350326538086, "epoch": 4, "memory": 44140, "step": 14033} +{"lr": 7.621152629240453e-05, "data_time": 0.0016044378280639648, "loss": 0.4456111818552017, "time": 0.6320450782775879, "epoch": 4, "memory": 44140, "step": 14133} +{"lr": 7.590762379752719e-05, "data_time": 0.0016407966613769531, "loss": 0.4569500327110291, "time": 0.6321018695831299, "epoch": 4, "memory": 44140, "step": 14233} +{"lr": 7.560250032600911e-05, "data_time": 0.0016803503036499023, "loss": 0.4478110492229462, "time": 0.632465410232544, "epoch": 4, "memory": 44140, "step": 14333} +{"lr": 7.52961736966437e-05, "data_time": 0.00166778564453125, "loss": 0.4908573359251022, "time": 0.6317228078842163, "epoch": 4, "memory": 44140, "step": 14433} +{"lr": 7.498866179848732e-05, "data_time": 0.0017252683639526368, "loss": 0.4687069535255432, "time": 0.6344333410263061, "epoch": 4, "memory": 44140, "step": 14533} +{"lr": 7.467998258981415e-05, "data_time": 0.001628875732421875, "loss": 0.4527924507856369, "time": 0.632758903503418, "epoch": 4, "memory": 44140, "step": 14633} +{"lr": 7.437015409706804e-05, "data_time": 0.001986527442932129, "loss": 0.47117196321487426, "time": 0.6316200256347656, "epoch": 4, "memory": 44140, "step": 14733} +{"lr": 7.405919441380929e-05, "data_time": 0.00152587890625, "loss": 0.466316694021225, "time": 0.631846261024475, "epoch": 4, "memory": 44140, "step": 14833} +{"lr": 7.374712169965821e-05, "data_time": 0.002025461196899414, "loss": 0.46932312846183777, "time": 0.6320650339126587, "epoch": 4, "memory": 44140, "step": 14933} +{"lr": 7.343395417923461e-05, "data_time": 0.0015321016311645509, "loss": 0.4634749710559845, "time": 0.6335531711578369, "epoch": 4, "memory": 44140, "step": 15033} +{"lr": 7.311971014109342e-05, "data_time": 0.0016558647155761718, "loss": 0.4653380632400513, "time": 0.6328489065170289, "epoch": 4, "memory": 44140, "step": 15133} +{"lr": 7.280440793665682e-05, "data_time": 0.0016974925994873047, "loss": 0.4721198469400406, "time": 0.6314444780349732, "epoch": 4, "memory": 44140, "step": 15233} +{"lr": 7.248806597914234e-05, "data_time": 0.0015108108520507813, "loss": 0.45408596396446227, "time": 0.6307537078857421, "epoch": 4, "memory": 44140, "step": 15333} +{"lr": 7.217070274248765e-05, "data_time": 0.0018209695816040039, "loss": 0.47602528631687163, "time": 0.6314978122711181, "epoch": 4, "memory": 44140, "step": 15433} +{"lr": 7.185233676027203e-05, "data_time": 0.0015051126480102538, "loss": 0.4789700746536255, "time": 0.6314777374267578, "epoch": 4, "memory": 44140, "step": 15533} +{"lr": 7.153298662463338e-05, "data_time": 0.0015069007873535155, "loss": 0.45825580358505247, "time": 0.6326617002487183, "epoch": 4, "memory": 44140, "step": 15633} +{"lr": 7.121267098518291e-05, "data_time": 0.0014716386795043945, "loss": 0.4747273147106171, "time": 0.631275749206543, "epoch": 4, "memory": 44140, "step": 15733} +{"lr": 7.089140854791618e-05, "data_time": 0.0022616147994995116, "loss": 0.45051977038383484, "time": 0.6312662601470947, "epoch": 4, "memory": 44140, "step": 15833} +{"lr": 7.056921807412025e-05, "data_time": 0.0015029668807983398, "loss": 0.4630744159221649, "time": 0.6306298971176147, "epoch": 4, "memory": 44140, "step": 15933} +{"lr": 7.024611837927802e-05, "data_time": 0.0022447586059570314, "loss": 0.46296154260635375, "time": 0.6328754901885987, "epoch": 4, "memory": 44140, "step": 16033} +{"lr": 6.992212833196998e-05, "data_time": 0.0024309158325195312, "loss": 0.45962140560150144, "time": 0.6336203813552856, "epoch": 4, "memory": 44140, "step": 16133} +{"lr": 6.959726685277174e-05, "data_time": 0.0018636226654052735, "loss": 0.44816306233406067, "time": 0.632193922996521, "epoch": 4, "memory": 44140, "step": 16233} +{"lr": 6.927155291314949e-05, "data_time": 0.0018933773040771484, "loss": 0.4800023674964905, "time": 0.6304071426391602, "epoch": 4, "memory": 44140, "step": 16333} +{"lr": 6.894500553435204e-05, "data_time": 0.003003048896789551, "loss": 0.47201485335826876, "time": 0.6325242996215821, "epoch": 4, "memory": 44140, "step": 16433} +{"accuracy/top1": 87.52462005615234, "data_time": 0.018533003330230714, "time": 1.2799886405467986, "step": 4} +{"lr": 6.858158509719903e-05, "data_time": 0.0017267227172851562, "loss": 0.4378457725048065, "time": 0.6323457717895508, "epoch": 5, "memory": 44140, "step": 16544} +{"lr": 6.825334178880171e-05, "data_time": 0.002093982696533203, "loss": 0.45123682022094724, "time": 0.6316104412078858, "epoch": 5, "memory": 44140, "step": 16644} +{"lr": 6.792432450336645e-05, "data_time": 0.0016284465789794921, "loss": 0.4461707055568695, "time": 0.6308274984359741, "epoch": 5, "memory": 44140, "step": 16744} +{"lr": 6.759455245505286e-05, "data_time": 0.0019920825958251952, "loss": 0.46688292622566224, "time": 0.6320695161819458, "epoch": 5, "memory": 44140, "step": 16844} +{"lr": 6.726404490209753e-05, "data_time": 0.0018121719360351563, "loss": 0.46765890121459963, "time": 0.6316281318664551, "epoch": 5, "memory": 44140, "step": 16944} +{"lr": 6.693282114568972e-05, "data_time": 0.0021088361740112306, "loss": 0.4624310553073883, "time": 0.6323110103607178, "epoch": 5, "memory": 44140, "step": 17044} +{"lr": 6.660090052884388e-05, "data_time": 0.0015198707580566407, "loss": 0.4679421305656433, "time": 0.6343421936035156, "epoch": 5, "memory": 44140, "step": 17144} +{"lr": 6.626830243527016e-05, "data_time": 0.002634119987487793, "loss": 0.42645345330238343, "time": 0.6380135536193847, "epoch": 5, "memory": 44140, "step": 17244} +{"lr": 6.593504628824236e-05, "data_time": 0.0015419483184814452, "loss": 0.4788371235132217, "time": 0.6317576169967651, "epoch": 5, "memory": 44140, "step": 17344} +{"lr": 6.560115154946387e-05, "data_time": 0.0018663644790649415, "loss": 0.4343857169151306, "time": 0.6330663442611695, "epoch": 5, "memory": 44140, "step": 17444} +{"lr": 6.526663771793081e-05, "data_time": 0.0015479564666748048, "loss": 0.45125500559806825, "time": 0.6322322130203247, "epoch": 5, "memory": 44140, "step": 17544} +{"lr": 6.493152432879363e-05, "data_time": 0.0018260955810546875, "loss": 0.458109176158905, "time": 0.6336064100265503, "epoch": 5, "memory": 44140, "step": 17644} +{"lr": 6.459583095221596e-05, "data_time": 0.001648092269897461, "loss": 0.4756574332714081, "time": 0.630453372001648, "epoch": 5, "memory": 44140, "step": 17744} +{"lr": 6.425957719223197e-05, "data_time": 0.0021019935607910155, "loss": 0.46511170268058777, "time": 0.6312604427337647, "epoch": 5, "memory": 44140, "step": 17844} +{"lr": 6.392278268560153e-05, "data_time": 0.001582026481628418, "loss": 0.4709571301937103, "time": 0.6320904970169068, "epoch": 5, "memory": 44140, "step": 17944} +{"lr": 6.358546710066316e-05, "data_time": 0.0025523900985717773, "loss": 0.46106258034706116, "time": 0.6339879035949707, "epoch": 5, "memory": 44140, "step": 18044} +{"lr": 6.324765013618573e-05, "data_time": 0.0016246795654296874, "loss": 0.463119113445282, "time": 0.6309337854385376, "epoch": 5, "memory": 44140, "step": 18144} +{"lr": 6.290935152021818e-05, "data_time": 0.001953792572021484, "loss": 0.46339941024780273, "time": 0.6311191082000732, "epoch": 5, "memory": 44140, "step": 18244} +{"lr": 6.2570591008937e-05, "data_time": 0.001540398597717285, "loss": 0.4621423244476318, "time": 0.6392133951187133, "epoch": 5, "memory": 44140, "step": 18344} +{"lr": 6.223138838549294e-05, "data_time": 0.001991677284240723, "loss": 0.473195880651474, "time": 0.6316898107528687, "epoch": 5, "memory": 44140, "step": 18444} +{"lr": 6.18917634588553e-05, "data_time": 0.0016360759735107421, "loss": 0.46876311004161836, "time": 0.630797266960144, "epoch": 5, "memory": 44140, "step": 18544} +{"lr": 6.155173606265555e-05, "data_time": 0.0021683931350708007, "loss": 0.4630735218524933, "time": 0.6323578357696533, "epoch": 5, "memory": 44140, "step": 18644} +{"lr": 6.121132605402843e-05, "data_time": 0.0015950679779052734, "loss": 0.4456640243530273, "time": 0.632435941696167, "epoch": 5, "memory": 44140, "step": 18744} +{"lr": 6.087055331245287e-05, "data_time": 0.0019070625305175782, "loss": 0.4652566611766815, "time": 0.6335298776626587, "epoch": 5, "memory": 44140, "step": 18844} +{"lr": 6.05294377385911e-05, "data_time": 0.0014652967453002929, "loss": 0.4578544795513153, "time": 0.6316671133041382, "epoch": 5, "memory": 44140, "step": 18944} +{"lr": 6.0187999253126355e-05, "data_time": 0.0016525745391845702, "loss": 0.4557887017726898, "time": 0.6442740678787231, "epoch": 5, "memory": 44140, "step": 19044} +{"lr": 5.9846257795599195e-05, "data_time": 0.0017616033554077148, "loss": 0.4571655809879303, "time": 0.631689977645874, "epoch": 5, "memory": 44140, "step": 19144} +{"lr": 5.950423332324355e-05, "data_time": 0.0015979766845703124, "loss": 0.46781941652297976, "time": 0.6310057640075684, "epoch": 5, "memory": 44140, "step": 19244} +{"lr": 5.9161945809820855e-05, "data_time": 0.001575636863708496, "loss": 0.4621806740760803, "time": 0.6312469244003296, "epoch": 5, "memory": 44140, "step": 19344} +{"lr": 5.8819415244453975e-05, "data_time": 0.0014534711837768555, "loss": 0.48360012471675873, "time": 0.6315086364746094, "epoch": 5, "memory": 44140, "step": 19444} +{"lr": 5.8476661630459405e-05, "data_time": 0.001828908920288086, "loss": 0.4748052477836609, "time": 0.6316088199615478, "epoch": 5, "memory": 44140, "step": 19544} +{"lr": 5.81337049841796e-05, "data_time": 0.0015892505645751954, "loss": 0.4698820114135742, "time": 0.6308746814727784, "epoch": 5, "memory": 44140, "step": 19644} +{"lr": 5.779056533381373e-05, "data_time": 0.0015501976013183594, "loss": 0.4809012174606323, "time": 0.6297628164291382, "epoch": 5, "memory": 44140, "step": 19744} +{"lr": 5.744726271824817e-05, "data_time": 0.0015085697174072265, "loss": 0.46830612421035767, "time": 0.6310423612594604, "epoch": 5, "memory": 44140, "step": 19844} +{"lr": 5.710381718588622e-05, "data_time": 0.0015188694000244141, "loss": 0.44808252453804015, "time": 0.6319166421890259, "epoch": 5, "memory": 44140, "step": 19944} +{"lr": 5.676024879347741e-05, "data_time": 0.0020087003707885743, "loss": 0.47926547527313235, "time": 0.631207799911499, "epoch": 5, "memory": 44140, "step": 20044} +{"lr": 5.6416577604945966e-05, "data_time": 0.0016213417053222655, "loss": 0.47091892957687376, "time": 0.6306166887283325, "epoch": 5, "memory": 44140, "step": 20144} +{"lr": 5.6072823690219385e-05, "data_time": 0.0016350746154785156, "loss": 0.46497154235839844, "time": 0.6301866054534913, "epoch": 5, "memory": 44140, "step": 20244} +{"lr": 5.572900712405624e-05, "data_time": 0.0017303228378295898, "loss": 0.45170841813087464, "time": 0.6334760904312133, "epoch": 5, "memory": 44140, "step": 20344} +{"lr": 5.538514798487397e-05, "data_time": 0.0015020132064819335, "loss": 0.45783367156982424, "time": 0.6305166244506836, "epoch": 5, "memory": 44140, "step": 20444} +{"lr": 5.5041266353576e-05, "data_time": 0.0031407833099365234, "loss": 0.45622223913669585, "time": 0.6328708410263062, "epoch": 5, "memory": 44140, "step": 20544} +{"accuracy/top1": 88.29088592529297, "data_time": 0.019408297538757325, "time": 1.280727324883143, "step": 5} +{"lr": 5.4659555783035384e-05, "data_time": 0.0017912149429321288, "loss": 0.47510509490966796, "time": 0.6332332134246826, "epoch": 6, "memory": 44140, "step": 20655} +{"lr": 5.431569258426336e-05, "data_time": 0.0016559600830078126, "loss": 0.4886723875999451, "time": 0.6334224939346313, "epoch": 6, "memory": 44140, "step": 20755} +{"lr": 5.397186934810864e-05, "data_time": 0.0015335559844970703, "loss": 0.4472442388534546, "time": 0.6323338747024536, "epoch": 6, "memory": 44140, "step": 20855} +{"lr": 5.362810615337815e-05, "data_time": 0.0016561508178710937, "loss": 0.4417957067489624, "time": 0.6310176372528076, "epoch": 6, "memory": 44140, "step": 20955} +{"lr": 5.328442307537233e-05, "data_time": 0.0016846418380737304, "loss": 0.4762031018733978, "time": 0.6307538032531739, "epoch": 6, "memory": 44140, "step": 21055} +{"lr": 5.294084018471303e-05, "data_time": 0.0017998933792114258, "loss": 0.46232172250747683, "time": 0.630191969871521, "epoch": 6, "memory": 44140, "step": 21155} +{"lr": 5.259737754617132e-05, "data_time": 0.0018306016921997071, "loss": 0.48820924758911133, "time": 0.6305662631988526, "epoch": 6, "memory": 44140, "step": 21255} +{"lr": 5.2254055217495525e-05, "data_time": 0.0015924930572509765, "loss": 0.45944496989250183, "time": 0.6331133604049682, "epoch": 6, "memory": 44140, "step": 21355} +{"lr": 5.1910893248240283e-05, "data_time": 0.0017383813858032227, "loss": 0.4717793345451355, "time": 0.6329855918884277, "epoch": 6, "memory": 44140, "step": 21455} +{"lr": 5.156791167859532e-05, "data_time": 0.0017553329467773437, "loss": 0.47676384449005127, "time": 0.6315877914428711, "epoch": 6, "memory": 44140, "step": 21555} +{"lr": 5.122513053821533e-05, "data_time": 0.0015356063842773438, "loss": 0.4817984700202942, "time": 0.6315043449401856, "epoch": 6, "memory": 44140, "step": 21655} +{"lr": 5.0882569845050173e-05, "data_time": 0.0015803813934326173, "loss": 0.46333622336387636, "time": 0.6333118438720703, "epoch": 6, "memory": 44140, "step": 21755} +{"lr": 5.054024960417592e-05, "data_time": 0.0015550613403320312, "loss": 0.4572906881570816, "time": 0.6306675672531128, "epoch": 6, "memory": 44140, "step": 21855} +{"lr": 5.0198189806626494e-05, "data_time": 0.0014666318893432617, "loss": 0.46266697645187377, "time": 0.6345197439193726, "epoch": 6, "memory": 44140, "step": 21955} +{"lr": 4.985641042822647e-05, "data_time": 0.0018036365509033203, "loss": 0.45842320919036866, "time": 0.631346607208252, "epoch": 6, "memory": 44140, "step": 22055} +{"lr": 4.951493142842412e-05, "data_time": 0.0015744924545288085, "loss": 0.4805125206708908, "time": 0.6314605951309205, "epoch": 6, "memory": 44140, "step": 22155} +{"lr": 4.917377274912603e-05, "data_time": 0.001524043083190918, "loss": 0.47288902997970583, "time": 0.6355272769927979, "epoch": 6, "memory": 44140, "step": 22255} +{"lr": 4.88329543135327e-05, "data_time": 0.0014655351638793944, "loss": 0.4488997280597687, "time": 0.6312304258346557, "epoch": 6, "memory": 44140, "step": 22355} +{"lr": 4.849249602497467e-05, "data_time": 0.0015840530395507812, "loss": 0.46450841426849365, "time": 0.6303846597671509, "epoch": 6, "memory": 44140, "step": 22455} +{"lr": 4.8152417765750456e-05, "data_time": 0.001480412483215332, "loss": 0.46635721921920775, "time": 0.632472038269043, "epoch": 6, "memory": 44140, "step": 22555} +{"lr": 4.7812739395965355e-05, "data_time": 0.0017144441604614257, "loss": 0.4644902467727661, "time": 0.6333511114120484, "epoch": 6, "memory": 44140, "step": 22655} +{"lr": 4.74734807523717e-05, "data_time": 0.0014958143234252929, "loss": 0.47039506435394285, "time": 0.6341737985610962, "epoch": 6, "memory": 44140, "step": 22755} +{"lr": 4.7134661647210416e-05, "data_time": 0.0018420934677124024, "loss": 0.4463201522827148, "time": 0.6463409900665283, "epoch": 6, "memory": 44140, "step": 22855} +{"lr": 4.679630186705388e-05, "data_time": 0.0016364336013793945, "loss": 0.4845757484436035, "time": 0.6325761079788208, "epoch": 6, "memory": 44140, "step": 22955} +{"lr": 4.6458421171650644e-05, "data_time": 0.0016704320907592774, "loss": 0.4898410618305206, "time": 0.6323488473892211, "epoch": 6, "memory": 44140, "step": 23055} +{"lr": 4.6121039292771295e-05, "data_time": 0.0015418291091918944, "loss": 0.4825673043727875, "time": 0.631195855140686, "epoch": 6, "memory": 44140, "step": 23155} +{"lr": 4.5784175933056197e-05, "data_time": 0.0015616893768310546, "loss": 0.4758934319019318, "time": 0.6317502737045289, "epoch": 6, "memory": 44140, "step": 23255} +{"lr": 4.544785076486494e-05, "data_time": 0.0014406442642211914, "loss": 0.46538543701171875, "time": 0.6338924407958985, "epoch": 6, "memory": 44140, "step": 23355} +{"lr": 4.5112083429127475e-05, "data_time": 0.0016784191131591797, "loss": 0.45752432346343996, "time": 0.6312843322753906, "epoch": 6, "memory": 44140, "step": 23455} +{"lr": 4.477689353419706e-05, "data_time": 0.002481961250305176, "loss": 0.4551388740539551, "time": 0.6350880146026612, "epoch": 6, "memory": 44140, "step": 23555} +{"lr": 4.4442300654705266e-05, "data_time": 0.002202892303466797, "loss": 0.48264811038970945, "time": 0.6322339057922364, "epoch": 6, "memory": 44140, "step": 23655} +{"lr": 4.410832433041866e-05, "data_time": 0.001952815055847168, "loss": 0.416254723072052, "time": 0.6314322233200074, "epoch": 6, "memory": 44140, "step": 23755} +{"lr": 4.377498406509792e-05, "data_time": 0.0017408132553100586, "loss": 0.43838741779327395, "time": 0.6314563035964966, "epoch": 6, "memory": 44140, "step": 23855} +{"lr": 4.344229932535875e-05, "data_time": 0.001574873924255371, "loss": 0.46265836954116824, "time": 0.6403370141983032, "epoch": 6, "memory": 44140, "step": 23955} +{"lr": 4.3110289539535115e-05, "data_time": 0.0015943050384521484, "loss": 0.4682974457740784, "time": 0.6343074083328247, "epoch": 6, "memory": 44140, "step": 24055} +{"lr": 4.2778974096544436e-05, "data_time": 0.002298569679260254, "loss": 0.4528957664966583, "time": 0.6313576698303223, "epoch": 6, "memory": 44140, "step": 24155} +{"lr": 4.2448372344755544e-05, "data_time": 0.0014259815216064453, "loss": 0.45468024015426634, "time": 0.6337838172912598, "epoch": 6, "memory": 44140, "step": 24255} +{"lr": 4.211850359085871e-05, "data_time": 0.0017948389053344727, "loss": 0.4562927335500717, "time": 0.6322719097137451, "epoch": 6, "memory": 44140, "step": 24355} +{"lr": 4.17893870987381e-05, "data_time": 0.0017693281173706056, "loss": 0.45234352350234985, "time": 0.6338937282562256, "epoch": 6, "memory": 44140, "step": 24455} +{"lr": 4.146104208834679e-05, "data_time": 0.0016025304794311523, "loss": 0.44231041669845583, "time": 0.634783124923706, "epoch": 6, "memory": 44140, "step": 24555} +{"lr": 4.113348773458442e-05, "data_time": 0.0030402421951293947, "loss": 0.4572556972503662, "time": 0.6370494604110718, "epoch": 6, "memory": 44140, "step": 24655} +{"accuracy/top1": 88.78241729736328, "data_time": 0.022437427441279092, "time": 1.2850611090660096, "step": 6} +{"lr": 4.0770851520835164e-05, "data_time": 0.0017620325088500977, "loss": 0.4570577025413513, "time": 0.6326758623123169, "epoch": 7, "memory": 44140, "step": 24766} +{"lr": 4.044502815742039e-05, "data_time": 0.0017591476440429687, "loss": 0.4358859181404114, "time": 0.632233715057373, "epoch": 7, "memory": 44140, "step": 24866} +{"lr": 4.012005478445896e-05, "data_time": 0.0015244007110595704, "loss": 0.46642338633537295, "time": 0.6319602489471435, "epoch": 7, "memory": 44140, "step": 24966} +{"lr": 3.9795950379951394e-05, "data_time": 0.0016994953155517578, "loss": 0.47227356433868406, "time": 0.6334277153015136, "epoch": 7, "memory": 44140, "step": 25066} +{"lr": 3.947273387115169e-05, "data_time": 0.001701688766479492, "loss": 0.46978639960289004, "time": 0.6347219705581665, "epoch": 7, "memory": 44140, "step": 25166} +{"lr": 3.9150424133462044e-05, "data_time": 0.0015353679656982422, "loss": 0.4644059896469116, "time": 0.6324105978012085, "epoch": 7, "memory": 44140, "step": 25266} +{"lr": 3.882903998933037e-05, "data_time": 0.002018308639526367, "loss": 0.44588848054409025, "time": 0.6323031663894654, "epoch": 7, "memory": 44140, "step": 25366} +{"lr": 3.8508600207151264e-05, "data_time": 0.0019781827926635743, "loss": 0.461130291223526, "time": 0.6324244976043701, "epoch": 7, "memory": 44140, "step": 25466} +{"lr": 3.818912350016978e-05, "data_time": 0.0016542673110961914, "loss": 0.4709395945072174, "time": 0.6339887857437134, "epoch": 7, "memory": 44140, "step": 25566} +{"lr": 3.787062852538883e-05, "data_time": 0.001529383659362793, "loss": 0.4548145353794098, "time": 0.6318777799606323, "epoch": 7, "memory": 44140, "step": 25666} +{"lr": 3.7553133882479346e-05, "data_time": 0.0023572683334350587, "loss": 0.47995511889457704, "time": 0.6329619407653808, "epoch": 7, "memory": 44140, "step": 25766} +{"lr": 3.723665811269437e-05, "data_time": 0.0018700361251831055, "loss": 0.4594398647546768, "time": 0.632764744758606, "epoch": 7, "memory": 44140, "step": 25866} +{"lr": 3.6921219697786176e-05, "data_time": 0.002150869369506836, "loss": 0.4604384511709213, "time": 0.6355153799057007, "epoch": 7, "memory": 44140, "step": 25966} +{"lr": 3.660683705892677e-05, "data_time": 0.0019325733184814454, "loss": 0.45717291831970214, "time": 0.6339921951293945, "epoch": 7, "memory": 44140, "step": 26066} +{"lr": 3.6293528555632435e-05, "data_time": 0.0020636796951293947, "loss": 0.45245722532272337, "time": 0.633061146736145, "epoch": 7, "memory": 44140, "step": 26166} +{"lr": 3.598131248469134e-05, "data_time": 0.0017244338989257813, "loss": 0.45272687673568723, "time": 0.632715368270874, "epoch": 7, "memory": 44140, "step": 26266} +{"lr": 3.567020707909507e-05, "data_time": 0.0016669511795043945, "loss": 0.5036671459674835, "time": 0.6316426038742066, "epoch": 7, "memory": 44140, "step": 26366} +{"lr": 3.5360230506973935e-05, "data_time": 0.001744699478149414, "loss": 0.4386345148086548, "time": 0.6319615125656128, "epoch": 7, "memory": 44140, "step": 26466} +{"lr": 3.5051400870535877e-05, "data_time": 0.0016463279724121093, "loss": 0.48083417415618895, "time": 0.6331605434417724, "epoch": 7, "memory": 44140, "step": 26566} +{"lr": 3.47437362050094e-05, "data_time": 0.0015377998352050781, "loss": 0.47480157017707825, "time": 0.6321899652481079, "epoch": 7, "memory": 44140, "step": 26666} +{"lr": 3.443725447759023e-05, "data_time": 0.0016852855682373048, "loss": 0.45800946950912474, "time": 0.6351692914962769, "epoch": 7, "memory": 44140, "step": 26766} +{"lr": 3.413197358639224e-05, "data_time": 0.0016905546188354492, "loss": 0.4890470147132874, "time": 0.6322234153747559, "epoch": 7, "memory": 44140, "step": 26866} +{"lr": 3.382791135940194e-05, "data_time": 0.0016348838806152343, "loss": 0.4506791472434998, "time": 0.6334895610809326, "epoch": 7, "memory": 44140, "step": 26966} +{"lr": 3.35250855534377e-05, "data_time": 0.00218658447265625, "loss": 0.47264742851257324, "time": 0.6329361438751221, "epoch": 7, "memory": 44140, "step": 27066} +{"lr": 3.3223513853112476e-05, "data_time": 0.0018648862838745116, "loss": 0.46374369859695436, "time": 0.6352895259857178, "epoch": 7, "memory": 44140, "step": 27166} +{"lr": 3.292321386980115e-05, "data_time": 0.0016809701919555664, "loss": 0.4720534086227417, "time": 0.6330942869186401, "epoch": 7, "memory": 44140, "step": 27266} +{"lr": 3.262420314061217e-05, "data_time": 0.001649951934814453, "loss": 0.4529055833816528, "time": 0.6318412065505982, "epoch": 7, "memory": 44140, "step": 27366} +{"lr": 3.2326499127363195e-05, "data_time": 0.001679682731628418, "loss": 0.489370596408844, "time": 0.6314186334609986, "epoch": 7, "memory": 44140, "step": 27466} +{"lr": 3.2030119215561613e-05, "data_time": 0.0014698028564453125, "loss": 0.46720961928367616, "time": 0.6314525365829468, "epoch": 7, "memory": 44140, "step": 27566} +{"lr": 3.173508071338896e-05, "data_time": 0.0018981695175170898, "loss": 0.48424118757247925, "time": 0.6329678535461426, "epoch": 7, "memory": 44140, "step": 27666} +{"lr": 3.144140085069037e-05, "data_time": 0.0017977476119995118, "loss": 0.4702145099639893, "time": 0.6332960605621338, "epoch": 7, "memory": 44140, "step": 27766} +{"lr": 3.114909677796814e-05, "data_time": 0.0015241622924804688, "loss": 0.46597014665603637, "time": 0.636191463470459, "epoch": 7, "memory": 44140, "step": 27866} +{"lr": 3.085818556538059e-05, "data_time": 0.0020096540451049806, "loss": 0.444723516702652, "time": 0.633235239982605, "epoch": 7, "memory": 44140, "step": 27966} +{"lr": 3.056868420174455e-05, "data_time": 0.0017461776733398438, "loss": 0.4772075653076172, "time": 0.6324796676635742, "epoch": 7, "memory": 44140, "step": 28066} +{"lr": 3.028060959354392e-05, "data_time": 0.0017737388610839845, "loss": 0.45855223536491396, "time": 0.6321711540222168, "epoch": 7, "memory": 44140, "step": 28166} +{"lr": 2.9993978563941876e-05, "data_time": 0.0016729116439819336, "loss": 0.45951600670814513, "time": 0.6341857671737671, "epoch": 7, "memory": 44140, "step": 28266} +{"lr": 2.9708807851798587e-05, "data_time": 0.001439213752746582, "loss": 0.5164131402969361, "time": 0.63258798122406, "epoch": 7, "memory": 44140, "step": 28366} +{"lr": 2.9425114110693717e-05, "data_time": 0.001916360855102539, "loss": 0.448526531457901, "time": 0.632703447341919, "epoch": 7, "memory": 44140, "step": 28466} +{"lr": 2.9142913907953844e-05, "data_time": 0.001385807991027832, "loss": 0.4666679859161377, "time": 0.6354238748550415, "epoch": 7, "memory": 44140, "step": 28566} +{"lr": 2.8862223723684912e-05, "data_time": 0.0016641616821289062, "loss": 0.4680193066596985, "time": 0.6330220699310303, "epoch": 7, "memory": 44140, "step": 28666} +{"lr": 2.8583059949809854e-05, "data_time": 0.002976202964782715, "loss": 0.4719451904296875, "time": 0.6387584209442139, "epoch": 7, "memory": 44140, "step": 28766} +{"accuracy/top1": 88.94461059570312, "data_time": 0.019816283384958902, "time": 1.2822944482167562, "step": 7} +{"lr": 2.827499545208893e-05, "data_time": 0.0020222902297973634, "loss": 0.4786592602729797, "time": 0.6311656951904296, "epoch": 8, "memory": 44140, "step": 28877} +{"lr": 2.7999105784987406e-05, "data_time": 0.0015462875366210938, "loss": 0.4444415599107742, "time": 0.6323636054992676, "epoch": 8, "memory": 44140, "step": 28977} +{"lr": 2.772479293318711e-05, "data_time": 0.001770353317260742, "loss": 0.4803554952144623, "time": 0.6327940464019776, "epoch": 8, "memory": 44140, "step": 29077} +{"lr": 2.7452072916183603e-05, "data_time": 0.0015571832656860352, "loss": 0.4572782635688782, "time": 0.6335251569747925, "epoch": 8, "memory": 44140, "step": 29177} +{"lr": 2.7180961660452883e-05, "data_time": 0.0015721321105957031, "loss": 0.42625134587287905, "time": 0.6325462818145752, "epoch": 8, "memory": 44140, "step": 29277} +{"lr": 2.6911474998521622e-05, "data_time": 0.0015975236892700195, "loss": 0.46394469738006594, "time": 0.6334802627563476, "epoch": 8, "memory": 44140, "step": 29377} +{"lr": 2.664362866804239e-05, "data_time": 0.0015463590621948241, "loss": 0.48913257718086245, "time": 0.6320464134216308, "epoch": 8, "memory": 44140, "step": 29477} +{"lr": 2.637743831087459e-05, "data_time": 0.0017019033432006836, "loss": 0.45803717374801634, "time": 0.6317750215530396, "epoch": 8, "memory": 44140, "step": 29577} +{"lr": 2.6112919472171114e-05, "data_time": 0.0017027854919433594, "loss": 0.4668755173683167, "time": 0.6310395240783692, "epoch": 8, "memory": 44140, "step": 29677} +{"lr": 2.5850087599470434e-05, "data_time": 0.0017415523529052735, "loss": 0.4508780658245087, "time": 0.6319621562957763, "epoch": 8, "memory": 44140, "step": 29777} +{"lr": 2.5588958041794503e-05, "data_time": 0.0015781164169311524, "loss": 0.48173831701278685, "time": 0.6311343431472778, "epoch": 8, "memory": 44140, "step": 29877} +{"lr": 2.5329546048752453e-05, "data_time": 0.001935434341430664, "loss": 0.43395198583602906, "time": 0.6341276645660401, "epoch": 8, "memory": 44140, "step": 29977} +{"lr": 2.5071866769649806e-05, "data_time": 0.0015177488327026366, "loss": 0.46916183829307556, "time": 0.6306435585021972, "epoch": 8, "memory": 44140, "step": 30077} +{"lr": 2.4815935252604222e-05, "data_time": 0.0015572309494018555, "loss": 0.45433741211891177, "time": 0.6330373048782348, "epoch": 8, "memory": 44140, "step": 30177} +{"lr": 2.4561766443666384e-05, "data_time": 0.0016353845596313477, "loss": 0.46527657508850095, "time": 0.6324336051940918, "epoch": 8, "memory": 44140, "step": 30277} +{"lr": 2.4309375185947164e-05, "data_time": 0.0017819166183471679, "loss": 0.4976585924625397, "time": 0.6327979564666748, "epoch": 8, "memory": 44140, "step": 30377} +{"lr": 2.40587762187509e-05, "data_time": 0.002453899383544922, "loss": 0.48123987913131716, "time": 0.6336788415908814, "epoch": 8, "memory": 44140, "step": 30477} +{"lr": 2.3809984176714634e-05, "data_time": 0.0015416860580444336, "loss": 0.45979909896850585, "time": 0.6320846080780029, "epoch": 8, "memory": 44140, "step": 30577} +{"lr": 2.3563013588953406e-05, "data_time": 0.002267742156982422, "loss": 0.47019734382629397, "time": 0.6399927854537963, "epoch": 8, "memory": 44140, "step": 30677} +{"lr": 2.331787887821182e-05, "data_time": 0.0016018152236938477, "loss": 0.46065608263015745, "time": 0.6316389322280884, "epoch": 8, "memory": 44140, "step": 30777} +{"lr": 2.307459436002179e-05, "data_time": 0.0015846729278564454, "loss": 0.4970451831817627, "time": 0.6321336984634399, "epoch": 8, "memory": 44140, "step": 30877} +{"lr": 2.2833174241866527e-05, "data_time": 0.0015520811080932616, "loss": 0.4677172422409058, "time": 0.6325764894485474, "epoch": 8, "memory": 44140, "step": 30977} +{"lr": 2.259363262235086e-05, "data_time": 0.0015968561172485351, "loss": 0.4666645675897598, "time": 0.63511803150177, "epoch": 8, "memory": 44140, "step": 31077} +{"lr": 2.2355983490377735e-05, "data_time": 0.0017240524291992187, "loss": 0.48023677468299864, "time": 0.631782841682434, "epoch": 8, "memory": 44140, "step": 31177} +{"lr": 2.2120240724331546e-05, "data_time": 0.001915311813354492, "loss": 0.45828971862792967, "time": 0.6317096710205078, "epoch": 8, "memory": 44140, "step": 31277} +{"lr": 2.1886418091267445e-05, "data_time": 0.0017271995544433593, "loss": 0.4646733283996582, "time": 0.6314742088317871, "epoch": 8, "memory": 44140, "step": 31377} +{"lr": 2.1654529246107496e-05, "data_time": 0.0018069982528686524, "loss": 0.4614529192447662, "time": 0.6331280946731568, "epoch": 8, "memory": 44140, "step": 31477} +{"lr": 2.1424587730843184e-05, "data_time": 0.0016412019729614257, "loss": 0.45111280083656313, "time": 0.6334101676940918, "epoch": 8, "memory": 44140, "step": 31577} +{"lr": 2.1196606973744593e-05, "data_time": 0.001568770408630371, "loss": 0.48221046328544614, "time": 0.6346508979797363, "epoch": 8, "memory": 44140, "step": 31677} +{"lr": 2.0970600288576148e-05, "data_time": 0.0015542030334472657, "loss": 0.4559887021780014, "time": 0.6328295707702637, "epoch": 8, "memory": 44140, "step": 31777} +{"lr": 2.0746580873819274e-05, "data_time": 0.0016139030456542968, "loss": 0.4885957300662994, "time": 0.6321042537689209, "epoch": 8, "memory": 44140, "step": 31877} +{"lr": 2.0524561811901426e-05, "data_time": 0.0014272212982177734, "loss": 0.475443696975708, "time": 0.6333238363265992, "epoch": 8, "memory": 44140, "step": 31977} +{"lr": 2.0304556068432268e-05, "data_time": 0.001703333854675293, "loss": 0.45770376920700073, "time": 0.6328312158584595, "epoch": 8, "memory": 44140, "step": 32077} +{"lr": 2.0086576491446333e-05, "data_time": 0.001783895492553711, "loss": 0.4704034090042114, "time": 0.6323757171630859, "epoch": 8, "memory": 44140, "step": 32177} +{"lr": 1.987063581065291e-05, "data_time": 0.0021164417266845703, "loss": 0.4640349090099335, "time": 0.6318027973175049, "epoch": 8, "memory": 44140, "step": 32277} +{"lr": 1.965674663669245e-05, "data_time": 0.0016535520553588867, "loss": 0.46104816198348997, "time": 0.63171865940094, "epoch": 8, "memory": 44140, "step": 32377} +{"lr": 1.9444921460400247e-05, "data_time": 0.0017505884170532227, "loss": 0.4569315552711487, "time": 0.6315577745437622, "epoch": 8, "memory": 44140, "step": 32477} +{"lr": 1.923517265207697e-05, "data_time": 0.0016154766082763672, "loss": 0.4757853806018829, "time": 0.6331660270690918, "epoch": 8, "memory": 44140, "step": 32577} +{"lr": 1.9027512460766204e-05, "data_time": 0.0025146484375, "loss": 0.4882297396659851, "time": 0.6356138229370117, "epoch": 8, "memory": 44140, "step": 32677} +{"lr": 1.8821953013539136e-05, "data_time": 0.001882028579711914, "loss": 0.4461084544658661, "time": 0.6315230607986451, "epoch": 8, "memory": 44140, "step": 32777} +{"lr": 1.8618506314786433e-05, "data_time": 0.003214120864868164, "loss": 0.4858038812875748, "time": 0.6333858251571656, "epoch": 8, "memory": 44140, "step": 32877} +{"accuracy/top1": 89.2193374633789, "data_time": 0.01960983673731486, "time": 1.2817478557427724, "step": 8} +{"lr": 1.8395169032542028e-05, "data_time": 0.001709580421447754, "loss": 0.44001683592796326, "time": 0.6302437543869018, "epoch": 9, "memory": 44140, "step": 32988} +{"lr": 1.819621906460573e-05, "data_time": 0.002226090431213379, "loss": 0.45350052416324615, "time": 0.6339422464370728, "epoch": 9, "memory": 44140, "step": 33088} +{"lr": 1.799941838715077e-05, "data_time": 0.002341437339782715, "loss": 0.4614435076713562, "time": 0.6329886674880981, "epoch": 9, "memory": 44140, "step": 33188} +{"lr": 1.7804778493067625e-05, "data_time": 0.00311582088470459, "loss": 0.4876255303621292, "time": 0.6371673583984375, "epoch": 9, "memory": 44140, "step": 33288} +{"lr": 1.761231074905996e-05, "data_time": 0.0033722639083862303, "loss": 0.5050480723381042, "time": 0.6350805044174195, "epoch": 9, "memory": 44140, "step": 33388} +{"lr": 1.742202639498083e-05, "data_time": 0.0017731189727783203, "loss": 0.4838721454143524, "time": 0.6342012643814087, "epoch": 9, "memory": 44140, "step": 33488} +{"lr": 1.7233936543176347e-05, "data_time": 0.0016247987747192382, "loss": 0.47088128328323364, "time": 0.6331941604614257, "epoch": 9, "memory": 44140, "step": 33588} +{"lr": 1.704805217783665e-05, "data_time": 0.0018124341964721679, "loss": 0.48395604491233823, "time": 0.6322825670242309, "epoch": 9, "memory": 44140, "step": 33688} +{"lr": 1.68643841543545e-05, "data_time": 0.0016111373901367188, "loss": 0.4804447889328003, "time": 0.6329972982406616, "epoch": 9, "memory": 44140, "step": 33788} +{"lr": 1.6682943198691327e-05, "data_time": 0.002145981788635254, "loss": 0.46445180773735045, "time": 0.634234094619751, "epoch": 9, "memory": 44140, "step": 33888} +{"lr": 1.6503739906750822e-05, "data_time": 0.0018756866455078125, "loss": 0.4758709967136383, "time": 0.6345801591873169, "epoch": 9, "memory": 44140, "step": 33988} +{"lr": 1.6326784743760187e-05, "data_time": 0.0015656232833862304, "loss": 0.4649648368358612, "time": 0.633787989616394, "epoch": 9, "memory": 44140, "step": 34088} +{"lr": 1.6152088043659013e-05, "data_time": 0.0017421722412109375, "loss": 0.4618089526891708, "time": 0.6334868907928467, "epoch": 9, "memory": 44140, "step": 34188} +{"lr": 1.59796600084957e-05, "data_time": 0.001541447639465332, "loss": 0.46781880855560304, "time": 0.6310805797576904, "epoch": 9, "memory": 44140, "step": 34288} +{"lr": 1.5809510707831687e-05, "data_time": 0.001674985885620117, "loss": 0.4476747512817383, "time": 0.6365415573120117, "epoch": 9, "memory": 44140, "step": 34388} +{"lr": 1.5641650078153473e-05, "data_time": 0.0016707897186279297, "loss": 0.46122092604637144, "time": 0.6341193914413452, "epoch": 9, "memory": 44140, "step": 34488} +{"lr": 1.547608792229235e-05, "data_time": 0.0015656232833862304, "loss": 0.4754705190658569, "time": 0.6324231624603271, "epoch": 9, "memory": 44140, "step": 34588} +{"lr": 1.5312833908851794e-05, "data_time": 0.0017354011535644532, "loss": 0.48189482688903806, "time": 0.6311715364456176, "epoch": 9, "memory": 44140, "step": 34688} +{"lr": 1.5151897571642958e-05, "data_time": 0.001479506492614746, "loss": 0.46493968963623045, "time": 0.6299102306365967, "epoch": 9, "memory": 44140, "step": 34788} +{"lr": 1.4993288309127877e-05, "data_time": 0.001821756362915039, "loss": 0.4356079876422882, "time": 0.630953335762024, "epoch": 9, "memory": 44140, "step": 34888} +{"lr": 1.4837015383870593e-05, "data_time": 0.001851367950439453, "loss": 0.4493475317955017, "time": 0.6351421356201172, "epoch": 9, "memory": 44140, "step": 34988} +{"lr": 1.4683087921996264e-05, "data_time": 0.0014225006103515624, "loss": 0.4474563956260681, "time": 0.6318857669830322, "epoch": 9, "memory": 44140, "step": 35088} +{"lr": 1.453151491265819e-05, "data_time": 0.001576542854309082, "loss": 0.462446403503418, "time": 0.6326565742492676, "epoch": 9, "memory": 44140, "step": 35188} +{"lr": 1.4382305207512847e-05, "data_time": 0.002078413963317871, "loss": 0.4956405937671661, "time": 0.631881856918335, "epoch": 9, "memory": 44140, "step": 35288} +{"lr": 1.423546752020298e-05, "data_time": 0.0015061855316162109, "loss": 0.4497751295566559, "time": 0.6314989328384399, "epoch": 9, "memory": 44140, "step": 35388} +{"lr": 1.4091010425848762e-05, "data_time": 0.0015552520751953125, "loss": 0.47529215812683107, "time": 0.630776333808899, "epoch": 9, "memory": 44140, "step": 35488} +{"lr": 1.3948942360546966e-05, "data_time": 0.0016089916229248048, "loss": 0.4584742307662964, "time": 0.6311586618423461, "epoch": 9, "memory": 44140, "step": 35588} +{"lr": 1.3809271620878346e-05, "data_time": 0.0016643047332763673, "loss": 0.44769375324249266, "time": 0.6308555126190185, "epoch": 9, "memory": 44140, "step": 35688} +{"lr": 1.3672006363423091e-05, "data_time": 0.0017024517059326173, "loss": 0.476364403963089, "time": 0.6311171293258667, "epoch": 9, "memory": 44140, "step": 35788} +{"lr": 1.3537154604284566e-05, "data_time": 0.0016996622085571288, "loss": 0.4687406361103058, "time": 0.6304677248001098, "epoch": 9, "memory": 44140, "step": 35888} +{"lr": 1.3404724218621087e-05, "data_time": 0.001668858528137207, "loss": 0.4763421356678009, "time": 0.6316363573074341, "epoch": 9, "memory": 44140, "step": 35988} +{"lr": 1.3274722940186121e-05, "data_time": 0.0016004323959350586, "loss": 0.4671560525894165, "time": 0.6304128646850586, "epoch": 9, "memory": 44140, "step": 36088} +{"lr": 1.3147158360876541e-05, "data_time": 0.001594233512878418, "loss": 0.45514750480651855, "time": 0.6327695369720459, "epoch": 9, "memory": 44140, "step": 36188} +{"lr": 1.3022037930289352e-05, "data_time": 0.0017365694046020507, "loss": 0.44409971237182616, "time": 0.6322862863540649, "epoch": 9, "memory": 44140, "step": 36288} +{"lr": 1.2899368955286627e-05, "data_time": 0.0015902042388916016, "loss": 0.4862985908985138, "time": 0.6307955503463745, "epoch": 9, "memory": 44140, "step": 36388} +{"lr": 1.2779158599568751e-05, "data_time": 0.0017532587051391601, "loss": 0.44753718972206114, "time": 0.6334474325180054, "epoch": 9, "memory": 44140, "step": 36488} +{"lr": 1.2661413883256134e-05, "data_time": 0.001761150360107422, "loss": 0.46288397908210754, "time": 0.6308752059936523, "epoch": 9, "memory": 44140, "step": 36588} +{"lr": 1.2546141682479203e-05, "data_time": 0.0016102313995361329, "loss": 0.47072811126708985, "time": 0.6337591171264648, "epoch": 9, "memory": 44140, "step": 36688} +{"lr": 1.2433348728976846e-05, "data_time": 0.0015789031982421874, "loss": 0.44911747574806216, "time": 0.6305293321609498, "epoch": 9, "memory": 44140, "step": 36788} +{"lr": 1.2323041609703344e-05, "data_time": 0.0022758722305297853, "loss": 0.45526446104049684, "time": 0.6349687099456787, "epoch": 9, "memory": 44140, "step": 36888} +{"lr": 1.2215226766443614e-05, "data_time": 0.0030605554580688476, "loss": 0.44789217710494994, "time": 0.6341535806655884, "epoch": 9, "memory": 44140, "step": 36988} +{"accuracy/top1": 89.41959381103516, "data_time": 0.01953543027242025, "time": 1.281510070959727, "step": 9} +{"lr": 1.209847850894437e-05, "data_time": 0.001443958282470703, "loss": 0.4737401843070984, "time": 0.6327221870422364, "epoch": 10, "memory": 44140, "step": 37099} +{"lr": 1.1995942849201055e-05, "data_time": 0.0015955448150634765, "loss": 0.4663243889808655, "time": 0.6334791421890259, "epoch": 10, "memory": 44140, "step": 37199} +{"lr": 1.189591856759183e-05, "data_time": 0.001537179946899414, "loss": 0.46519347429275515, "time": 0.6322403669357299, "epoch": 10, "memory": 44140, "step": 37299} +{"lr": 1.1798411505398016e-05, "data_time": 0.0014942169189453125, "loss": 0.479831725358963, "time": 0.6337087154388428, "epoch": 10, "memory": 44140, "step": 37399} +{"lr": 1.1703427356898678e-05, "data_time": 0.0017148494720458985, "loss": 0.4661745309829712, "time": 0.6314796447753906, "epoch": 10, "memory": 44140, "step": 37499} +{"lr": 1.1610971669038223e-05, "data_time": 0.001622915267944336, "loss": 0.4556196093559265, "time": 0.6319317579269409, "epoch": 10, "memory": 44140, "step": 37599} +{"lr": 1.1521049841102396e-05, "data_time": 0.0015362024307250977, "loss": 0.45149661004543307, "time": 0.632202935218811, "epoch": 10, "memory": 44140, "step": 37699} +{"lr": 1.1433667124402997e-05, "data_time": 0.0016666173934936524, "loss": 0.4552126169204712, "time": 0.6328606843948364, "epoch": 10, "memory": 44140, "step": 37799} +{"lr": 1.1348828621971181e-05, "data_time": 0.0016186237335205078, "loss": 0.4752184867858887, "time": 0.6327661752700806, "epoch": 10, "memory": 44140, "step": 37899} +{"lr": 1.1266539288259486e-05, "data_time": 0.0014984846115112306, "loss": 0.48326406478881834, "time": 0.6329694747924804, "epoch": 10, "memory": 44140, "step": 37999} +{"lr": 1.1186803928852477e-05, "data_time": 0.0016207218170166016, "loss": 0.4649399757385254, "time": 0.6346088171005249, "epoch": 10, "memory": 44140, "step": 38099} +{"lr": 1.1109627200186125e-05, "data_time": 0.0017227411270141601, "loss": 0.5148457169532776, "time": 0.6336653470993042, "epoch": 10, "memory": 44140, "step": 38199} +{"lr": 1.1035013609275835e-05, "data_time": 0.00175931453704834, "loss": 0.4879583239555359, "time": 0.6326696634292602, "epoch": 10, "memory": 44140, "step": 38299} +{"lr": 1.0962967513453298e-05, "data_time": 0.001738119125366211, "loss": 0.4747924208641052, "time": 0.6331334590911866, "epoch": 10, "memory": 44140, "step": 38399} +{"lr": 1.0893493120111999e-05, "data_time": 0.0015506267547607422, "loss": 0.45279773473739626, "time": 0.6357613563537597, "epoch": 10, "memory": 44140, "step": 38499} +{"lr": 1.0826594486461509e-05, "data_time": 0.0016738653182983398, "loss": 0.452606725692749, "time": 0.6351911067962647, "epoch": 10, "memory": 44140, "step": 38599} +{"lr": 1.0762275519290557e-05, "data_time": 0.0018566370010375977, "loss": 0.45806434750556946, "time": 0.6325887203216553, "epoch": 10, "memory": 44140, "step": 38699} +{"lr": 1.0700539974738862e-05, "data_time": 0.002175045013427734, "loss": 0.46723602414131166, "time": 0.6351238250732422, "epoch": 10, "memory": 44140, "step": 38799} +{"lr": 1.064139145807783e-05, "data_time": 0.0019116640090942384, "loss": 0.45460622310638427, "time": 0.6343905925750732, "epoch": 10, "memory": 44140, "step": 38899} +{"lr": 1.0584833423499947e-05, "data_time": 0.002245640754699707, "loss": 0.4611483097076416, "time": 0.6363897562026978, "epoch": 10, "memory": 44140, "step": 38999} +{"lr": 1.0530869173917099e-05, "data_time": 0.001709294319152832, "loss": 0.4460208684206009, "time": 0.6381627798080445, "epoch": 10, "memory": 44140, "step": 39099} +{"lr": 1.0479501860767654e-05, "data_time": 0.0015112638473510742, "loss": 0.45862733125686644, "time": 0.634428858757019, "epoch": 10, "memory": 44140, "step": 39199} +{"lr": 1.043073448383247e-05, "data_time": 0.0015918254852294923, "loss": 0.480582058429718, "time": 0.633992338180542, "epoch": 10, "memory": 44140, "step": 39299} +{"lr": 1.038456989105968e-05, "data_time": 0.0014903783798217774, "loss": 0.46571108102798464, "time": 0.6325500965118408, "epoch": 10, "memory": 44140, "step": 39399} +{"lr": 1.034101077839838e-05, "data_time": 0.0015347957611083984, "loss": 0.47578980922698977, "time": 0.634426212310791, "epoch": 10, "memory": 44140, "step": 39499} +{"lr": 1.0300059689641194e-05, "data_time": 0.0014916419982910155, "loss": 0.47278679609298707, "time": 0.6343709230422974, "epoch": 10, "memory": 44140, "step": 39599} +{"lr": 1.0261719016275683e-05, "data_time": 0.002152252197265625, "loss": 0.45428922176361086, "time": 0.633298134803772, "epoch": 10, "memory": 44140, "step": 39699} +{"lr": 1.0225990997344757e-05, "data_time": 0.0024214267730712892, "loss": 0.45897480845451355, "time": 0.6334645748138428, "epoch": 10, "memory": 44140, "step": 39799} +{"lr": 1.019287771931587e-05, "data_time": 0.0018765926361083984, "loss": 0.4657438904047012, "time": 0.6348436355590821, "epoch": 10, "memory": 44140, "step": 39899} +{"lr": 1.0162381115959187e-05, "data_time": 0.001739835739135742, "loss": 0.46620587110519407, "time": 0.6328980445861816, "epoch": 10, "memory": 44140, "step": 39999} +{"lr": 1.0134502968234636e-05, "data_time": 0.0016477108001708984, "loss": 0.43456063866615297, "time": 0.6337140560150146, "epoch": 10, "memory": 44140, "step": 40099} +{"lr": 1.0109244904187918e-05, "data_time": 0.001727747917175293, "loss": 0.46338191628456116, "time": 0.6331120491027832, "epoch": 10, "memory": 44140, "step": 40199} +{"lr": 1.0086608398855446e-05, "data_time": 0.0015302419662475586, "loss": 0.4902698159217834, "time": 0.6338290452957154, "epoch": 10, "memory": 44140, "step": 40299} +{"lr": 1.0066594774178176e-05, "data_time": 0.0022404670715332033, "loss": 0.4802827000617981, "time": 0.6359631538391113, "epoch": 10, "memory": 44140, "step": 40399} +{"lr": 1.0049205198924413e-05, "data_time": 0.001586771011352539, "loss": 0.4385004460811615, "time": 0.6343741178512573, "epoch": 10, "memory": 44140, "step": 40499} +{"lr": 1.0034440688621577e-05, "data_time": 0.0019191980361938476, "loss": 0.46651833057403563, "time": 0.6348704099655151, "epoch": 10, "memory": 44140, "step": 40599} +{"lr": 1.0022302105496896e-05, "data_time": 0.0016030311584472657, "loss": 0.45719313621520996, "time": 0.6345669269561768, "epoch": 10, "memory": 44140, "step": 40699} +{"lr": 1.0012790158427017e-05, "data_time": 0.0015558719635009766, "loss": 0.47053351402282717, "time": 0.6333394289016724, "epoch": 10, "memory": 44140, "step": 40799} +{"lr": 1.000590540289662e-05, "data_time": 0.0016728639602661133, "loss": 0.46899487972259524, "time": 0.6342001914978027, "epoch": 10, "memory": 44140, "step": 40899} +{"lr": 1.0001648240966046e-05, "data_time": 0.0016131639480590821, "loss": 0.4478962004184723, "time": 0.639711880683899, "epoch": 10, "memory": 44140, "step": 40999} +{"lr": 1.0000018921247718e-05, "data_time": 0.0031923294067382813, "loss": 0.4709731459617615, "time": 0.6367666721343994, "epoch": 10, "memory": 44140, "step": 41099} +{"accuracy/top1": 89.48082733154297, "data_time": 0.018739457925160727, "time": 1.2785786350568136, "step": 10} diff --git a/clip_large_pretrain_4x256_all_lr1e-4/20230605_033556/vis_data/vis_image/0000000.png_1.png b/clip_large_pretrain_4x256_all_lr1e-4/20230605_033556/vis_data/vis_image/0000000.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..10e800755a042b2fc20ef31a5403399777b5a9f0 Binary files /dev/null and b/clip_large_pretrain_4x256_all_lr1e-4/20230605_033556/vis_data/vis_image/0000000.png_1.png differ diff --git a/clip_large_pretrain_4x256_all_lr1e-4/20230605_033556/vis_data/vis_image/0000000.png_10.png b/clip_large_pretrain_4x256_all_lr1e-4/20230605_033556/vis_data/vis_image/0000000.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..c5918390ef2df42c54b02340868b9574c29f5c7d Binary files /dev/null and b/clip_large_pretrain_4x256_all_lr1e-4/20230605_033556/vis_data/vis_image/0000000.png_10.png differ diff --git a/clip_large_pretrain_4x256_all_lr1e-4/20230605_033556/vis_data/vis_image/0000000.png_2.png b/clip_large_pretrain_4x256_all_lr1e-4/20230605_033556/vis_data/vis_image/0000000.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..def7c3730ae526b804f79346a2f5beb4a78bdda4 Binary files /dev/null and b/clip_large_pretrain_4x256_all_lr1e-4/20230605_033556/vis_data/vis_image/0000000.png_2.png differ diff --git a/clip_large_pretrain_4x256_all_lr1e-4/20230605_033556/vis_data/vis_image/0000000.png_3.png b/clip_large_pretrain_4x256_all_lr1e-4/20230605_033556/vis_data/vis_image/0000000.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..c9f8147ddc15292a3b3916bb9b420bb9969af2dc Binary files /dev/null and b/clip_large_pretrain_4x256_all_lr1e-4/20230605_033556/vis_data/vis_image/0000000.png_3.png differ diff --git a/clip_large_pretrain_4x256_all_lr1e-4/20230605_033556/vis_data/vis_image/0000000.png_4.png b/clip_large_pretrain_4x256_all_lr1e-4/20230605_033556/vis_data/vis_image/0000000.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..93633baf77b6aea38be2386406d76cc97fc3b786 Binary files /dev/null and b/clip_large_pretrain_4x256_all_lr1e-4/20230605_033556/vis_data/vis_image/0000000.png_4.png differ diff --git a/clip_large_pretrain_4x256_all_lr1e-4/20230605_033556/vis_data/vis_image/0000000.png_5.png b/clip_large_pretrain_4x256_all_lr1e-4/20230605_033556/vis_data/vis_image/0000000.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..ea59011fdd1abd25744aa4db7b3c11b4db3baa95 Binary files /dev/null and b/clip_large_pretrain_4x256_all_lr1e-4/20230605_033556/vis_data/vis_image/0000000.png_5.png differ diff --git a/clip_large_pretrain_4x256_all_lr1e-4/20230605_033556/vis_data/vis_image/0000000.png_6.png b/clip_large_pretrain_4x256_all_lr1e-4/20230605_033556/vis_data/vis_image/0000000.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..62fa9d3e78b6bb90d0bd23b2f503d86668fee89b Binary files /dev/null and b/clip_large_pretrain_4x256_all_lr1e-4/20230605_033556/vis_data/vis_image/0000000.png_6.png differ diff --git a/clip_large_pretrain_4x256_all_lr1e-4/20230605_033556/vis_data/vis_image/0000000.png_7.png b/clip_large_pretrain_4x256_all_lr1e-4/20230605_033556/vis_data/vis_image/0000000.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..8aefa917c6eb6ee4534d995270708b8671fc37c9 Binary files /dev/null and b/clip_large_pretrain_4x256_all_lr1e-4/20230605_033556/vis_data/vis_image/0000000.png_7.png differ diff --git a/clip_large_pretrain_4x256_all_lr1e-4/20230605_033556/vis_data/vis_image/0000000.png_8.png b/clip_large_pretrain_4x256_all_lr1e-4/20230605_033556/vis_data/vis_image/0000000.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..407171e98c2e6e05219a8dea58386fe3b4b5be70 Binary files /dev/null and b/clip_large_pretrain_4x256_all_lr1e-4/20230605_033556/vis_data/vis_image/0000000.png_8.png differ diff --git a/clip_large_pretrain_4x256_all_lr1e-4/20230605_033556/vis_data/vis_image/0000000.png_9.png b/clip_large_pretrain_4x256_all_lr1e-4/20230605_033556/vis_data/vis_image/0000000.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..0dc60b7aee99bba4cd98cc177d95f9134eb07d1c Binary files /dev/null and b/clip_large_pretrain_4x256_all_lr1e-4/20230605_033556/vis_data/vis_image/0000000.png_9.png differ diff --git a/clip_large_pretrain_4x256_all_lr1e-4/20230605_033556/vis_data/vis_image/000015337.jpg_1.png b/clip_large_pretrain_4x256_all_lr1e-4/20230605_033556/vis_data/vis_image/000015337.jpg_1.png new file mode 100644 index 0000000000000000000000000000000000000000..60ad796f94b2c3396062a96eb34c8532b925b0da Binary files /dev/null and b/clip_large_pretrain_4x256_all_lr1e-4/20230605_033556/vis_data/vis_image/000015337.jpg_1.png differ diff --git a/clip_large_pretrain_4x256_all_lr1e-4/20230605_033556/vis_data/vis_image/000015337.jpg_10.png b/clip_large_pretrain_4x256_all_lr1e-4/20230605_033556/vis_data/vis_image/000015337.jpg_10.png new file mode 100644 index 0000000000000000000000000000000000000000..8a3371073f8a95eda036a6325932fe7f50c4405f Binary files /dev/null and b/clip_large_pretrain_4x256_all_lr1e-4/20230605_033556/vis_data/vis_image/000015337.jpg_10.png differ diff --git a/clip_large_pretrain_4x256_all_lr1e-4/20230605_033556/vis_data/vis_image/000015337.jpg_2.png b/clip_large_pretrain_4x256_all_lr1e-4/20230605_033556/vis_data/vis_image/000015337.jpg_2.png new file mode 100644 index 0000000000000000000000000000000000000000..9751782f272ac5af27dc419759165c5fb72206f6 Binary files /dev/null and b/clip_large_pretrain_4x256_all_lr1e-4/20230605_033556/vis_data/vis_image/000015337.jpg_2.png differ diff --git a/clip_large_pretrain_4x256_all_lr1e-4/20230605_033556/vis_data/vis_image/000015337.jpg_3.png b/clip_large_pretrain_4x256_all_lr1e-4/20230605_033556/vis_data/vis_image/000015337.jpg_3.png new file mode 100644 index 0000000000000000000000000000000000000000..a2abbe100490baebb4a6c735fdd2c79e3d984f3c Binary files /dev/null and b/clip_large_pretrain_4x256_all_lr1e-4/20230605_033556/vis_data/vis_image/000015337.jpg_3.png differ diff --git a/clip_large_pretrain_4x256_all_lr1e-4/20230605_033556/vis_data/vis_image/000015337.jpg_4.png b/clip_large_pretrain_4x256_all_lr1e-4/20230605_033556/vis_data/vis_image/000015337.jpg_4.png new file mode 100644 index 0000000000000000000000000000000000000000..7f1ac5c600525d6d491b0a760071a74ade76e6d3 Binary files /dev/null and b/clip_large_pretrain_4x256_all_lr1e-4/20230605_033556/vis_data/vis_image/000015337.jpg_4.png differ diff --git a/clip_large_pretrain_4x256_all_lr1e-4/20230605_033556/vis_data/vis_image/000015337.jpg_5.png b/clip_large_pretrain_4x256_all_lr1e-4/20230605_033556/vis_data/vis_image/000015337.jpg_5.png new file mode 100644 index 0000000000000000000000000000000000000000..39b27bf4749286bc8ee0d696ce2c8bf7454aa455 Binary files /dev/null and b/clip_large_pretrain_4x256_all_lr1e-4/20230605_033556/vis_data/vis_image/000015337.jpg_5.png differ diff --git a/clip_large_pretrain_4x256_all_lr1e-4/20230605_033556/vis_data/vis_image/000015337.jpg_6.png b/clip_large_pretrain_4x256_all_lr1e-4/20230605_033556/vis_data/vis_image/000015337.jpg_6.png new file mode 100644 index 0000000000000000000000000000000000000000..ef020777f45c8dc58d8a163f26de7b7ef4cd475d Binary files /dev/null and b/clip_large_pretrain_4x256_all_lr1e-4/20230605_033556/vis_data/vis_image/000015337.jpg_6.png differ diff --git a/clip_large_pretrain_4x256_all_lr1e-4/20230605_033556/vis_data/vis_image/000015337.jpg_7.png b/clip_large_pretrain_4x256_all_lr1e-4/20230605_033556/vis_data/vis_image/000015337.jpg_7.png new file mode 100644 index 0000000000000000000000000000000000000000..af573be35f2137b72e50ce204608f3f1db9642fa Binary files /dev/null and b/clip_large_pretrain_4x256_all_lr1e-4/20230605_033556/vis_data/vis_image/000015337.jpg_7.png differ diff --git a/clip_large_pretrain_4x256_all_lr1e-4/20230605_033556/vis_data/vis_image/000015337.jpg_8.png b/clip_large_pretrain_4x256_all_lr1e-4/20230605_033556/vis_data/vis_image/000015337.jpg_8.png new file mode 100644 index 0000000000000000000000000000000000000000..3a10b955a3ef9ed75163bbf3f5e59f453edf1c9d Binary files /dev/null and b/clip_large_pretrain_4x256_all_lr1e-4/20230605_033556/vis_data/vis_image/000015337.jpg_8.png differ diff --git a/clip_large_pretrain_4x256_all_lr1e-4/20230605_033556/vis_data/vis_image/000015337.jpg_9.png b/clip_large_pretrain_4x256_all_lr1e-4/20230605_033556/vis_data/vis_image/000015337.jpg_9.png new file mode 100644 index 0000000000000000000000000000000000000000..8131a1f2a30d2b5153f0a47778f9519e361149e5 Binary files /dev/null and b/clip_large_pretrain_4x256_all_lr1e-4/20230605_033556/vis_data/vis_image/000015337.jpg_9.png differ diff --git a/clip_large_pretrain_4x256_all_lr1e-4/20230605_033556/vis_data/vis_image/0004159.png_1.png b/clip_large_pretrain_4x256_all_lr1e-4/20230605_033556/vis_data/vis_image/0004159.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..b47cc6b61e7b064746415794b3cb413c96ca6296 Binary files /dev/null and b/clip_large_pretrain_4x256_all_lr1e-4/20230605_033556/vis_data/vis_image/0004159.png_1.png differ diff --git a/clip_large_pretrain_4x256_all_lr1e-4/20230605_033556/vis_data/vis_image/0004159.png_10.png b/clip_large_pretrain_4x256_all_lr1e-4/20230605_033556/vis_data/vis_image/0004159.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..ad33b083c9ccf182b190c29f7e6925836be1f704 Binary files /dev/null and b/clip_large_pretrain_4x256_all_lr1e-4/20230605_033556/vis_data/vis_image/0004159.png_10.png differ diff --git a/clip_large_pretrain_4x256_all_lr1e-4/20230605_033556/vis_data/vis_image/0004159.png_2.png b/clip_large_pretrain_4x256_all_lr1e-4/20230605_033556/vis_data/vis_image/0004159.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..3a5a498e91fcc1b2d7976df5bd77496b4af5bb87 Binary files /dev/null and b/clip_large_pretrain_4x256_all_lr1e-4/20230605_033556/vis_data/vis_image/0004159.png_2.png differ diff --git a/clip_large_pretrain_4x256_all_lr1e-4/20230605_033556/vis_data/vis_image/0004159.png_3.png b/clip_large_pretrain_4x256_all_lr1e-4/20230605_033556/vis_data/vis_image/0004159.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..059a9f15c22de8de69e39a040ee79aa6bda9ee0b Binary files /dev/null and b/clip_large_pretrain_4x256_all_lr1e-4/20230605_033556/vis_data/vis_image/0004159.png_3.png differ diff --git a/clip_large_pretrain_4x256_all_lr1e-4/20230605_033556/vis_data/vis_image/0004159.png_4.png b/clip_large_pretrain_4x256_all_lr1e-4/20230605_033556/vis_data/vis_image/0004159.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..656ac5edaf7f05ce2cb992b9a5f32d851f86a1b4 Binary files /dev/null and b/clip_large_pretrain_4x256_all_lr1e-4/20230605_033556/vis_data/vis_image/0004159.png_4.png differ diff --git a/clip_large_pretrain_4x256_all_lr1e-4/20230605_033556/vis_data/vis_image/0004159.png_5.png b/clip_large_pretrain_4x256_all_lr1e-4/20230605_033556/vis_data/vis_image/0004159.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..31ef26de30b639fb458bcc99108a4b66b93b4ff4 Binary files /dev/null and b/clip_large_pretrain_4x256_all_lr1e-4/20230605_033556/vis_data/vis_image/0004159.png_5.png differ diff --git a/clip_large_pretrain_4x256_all_lr1e-4/20230605_033556/vis_data/vis_image/0004159.png_6.png b/clip_large_pretrain_4x256_all_lr1e-4/20230605_033556/vis_data/vis_image/0004159.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..f9e651fed068ea6ca00b602c0862ac140bd63300 Binary files /dev/null and b/clip_large_pretrain_4x256_all_lr1e-4/20230605_033556/vis_data/vis_image/0004159.png_6.png differ diff --git a/clip_large_pretrain_4x256_all_lr1e-4/20230605_033556/vis_data/vis_image/0004159.png_7.png b/clip_large_pretrain_4x256_all_lr1e-4/20230605_033556/vis_data/vis_image/0004159.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..8b17049cf2b17e11b7f1d899b7069bc6c8541739 Binary files /dev/null and b/clip_large_pretrain_4x256_all_lr1e-4/20230605_033556/vis_data/vis_image/0004159.png_7.png differ diff --git a/clip_large_pretrain_4x256_all_lr1e-4/20230605_033556/vis_data/vis_image/0004159.png_8.png b/clip_large_pretrain_4x256_all_lr1e-4/20230605_033556/vis_data/vis_image/0004159.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..c630407e30e2e8da449e3d521f8cca9c95543b6c Binary files /dev/null and b/clip_large_pretrain_4x256_all_lr1e-4/20230605_033556/vis_data/vis_image/0004159.png_8.png differ diff --git a/clip_large_pretrain_4x256_all_lr1e-4/20230605_033556/vis_data/vis_image/0004159.png_9.png b/clip_large_pretrain_4x256_all_lr1e-4/20230605_033556/vis_data/vis_image/0004159.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..9ba72f0f344d7fa6d1978b48e3fb4c0a35af5224 Binary files /dev/null and b/clip_large_pretrain_4x256_all_lr1e-4/20230605_033556/vis_data/vis_image/0004159.png_9.png differ diff --git a/clip_large_pretrain_4x256_all_lr1e-4/20230605_033556/vis_data/vis_image/0008318.png_1.png b/clip_large_pretrain_4x256_all_lr1e-4/20230605_033556/vis_data/vis_image/0008318.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..44fbe029ec876c9717f9707b35c6d409962a4c7d Binary files /dev/null and b/clip_large_pretrain_4x256_all_lr1e-4/20230605_033556/vis_data/vis_image/0008318.png_1.png differ diff --git a/clip_large_pretrain_4x256_all_lr1e-4/20230605_033556/vis_data/vis_image/0008318.png_10.png b/clip_large_pretrain_4x256_all_lr1e-4/20230605_033556/vis_data/vis_image/0008318.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..e3a924043143756f22e3d6116999654870b6ceaf Binary files /dev/null and b/clip_large_pretrain_4x256_all_lr1e-4/20230605_033556/vis_data/vis_image/0008318.png_10.png differ diff --git a/clip_large_pretrain_4x256_all_lr1e-4/20230605_033556/vis_data/vis_image/0008318.png_2.png b/clip_large_pretrain_4x256_all_lr1e-4/20230605_033556/vis_data/vis_image/0008318.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..4db7854b0a14dce5f8d43524d78ac528cab5e42f Binary files /dev/null and b/clip_large_pretrain_4x256_all_lr1e-4/20230605_033556/vis_data/vis_image/0008318.png_2.png differ diff --git a/clip_large_pretrain_4x256_all_lr1e-4/20230605_033556/vis_data/vis_image/0008318.png_3.png b/clip_large_pretrain_4x256_all_lr1e-4/20230605_033556/vis_data/vis_image/0008318.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..f080b18baba3902ee240a4e87a065cd675b9e14c Binary files /dev/null and b/clip_large_pretrain_4x256_all_lr1e-4/20230605_033556/vis_data/vis_image/0008318.png_3.png differ diff --git a/clip_large_pretrain_4x256_all_lr1e-4/20230605_033556/vis_data/vis_image/0008318.png_4.png b/clip_large_pretrain_4x256_all_lr1e-4/20230605_033556/vis_data/vis_image/0008318.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..a80816adeff01ba13567d38179f35c71843ee868 Binary files /dev/null and b/clip_large_pretrain_4x256_all_lr1e-4/20230605_033556/vis_data/vis_image/0008318.png_4.png differ diff --git a/clip_large_pretrain_4x256_all_lr1e-4/20230605_033556/vis_data/vis_image/0008318.png_5.png b/clip_large_pretrain_4x256_all_lr1e-4/20230605_033556/vis_data/vis_image/0008318.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..cc3bdfdd51b9ea8506280587d26d3cb64ae01705 Binary files /dev/null and b/clip_large_pretrain_4x256_all_lr1e-4/20230605_033556/vis_data/vis_image/0008318.png_5.png differ diff --git a/clip_large_pretrain_4x256_all_lr1e-4/20230605_033556/vis_data/vis_image/0008318.png_6.png b/clip_large_pretrain_4x256_all_lr1e-4/20230605_033556/vis_data/vis_image/0008318.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..e04b1d8a40960fa85bca375e55e169221c28020e Binary files /dev/null and b/clip_large_pretrain_4x256_all_lr1e-4/20230605_033556/vis_data/vis_image/0008318.png_6.png differ diff --git a/clip_large_pretrain_4x256_all_lr1e-4/20230605_033556/vis_data/vis_image/0008318.png_7.png b/clip_large_pretrain_4x256_all_lr1e-4/20230605_033556/vis_data/vis_image/0008318.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..2f55df950dc0402c34f01ad5fd97302e81c3e2c8 Binary files /dev/null and b/clip_large_pretrain_4x256_all_lr1e-4/20230605_033556/vis_data/vis_image/0008318.png_7.png differ diff --git a/clip_large_pretrain_4x256_all_lr1e-4/20230605_033556/vis_data/vis_image/0008318.png_8.png b/clip_large_pretrain_4x256_all_lr1e-4/20230605_033556/vis_data/vis_image/0008318.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..2c55a7d85455215c2fe82a077fccfe26dfe441f3 Binary files /dev/null and b/clip_large_pretrain_4x256_all_lr1e-4/20230605_033556/vis_data/vis_image/0008318.png_8.png differ diff --git a/clip_large_pretrain_4x256_all_lr1e-4/20230605_033556/vis_data/vis_image/0008318.png_9.png b/clip_large_pretrain_4x256_all_lr1e-4/20230605_033556/vis_data/vis_image/0008318.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..85bf127c773295cfb5d31c79ede8efa0199716c3 Binary files /dev/null and b/clip_large_pretrain_4x256_all_lr1e-4/20230605_033556/vis_data/vis_image/0008318.png_9.png differ diff --git a/clip_large_pretrain_4x256_all_lr1e-4/clip_large_pretrain_4x256_all_lr1e-4.py b/clip_large_pretrain_4x256_all_lr1e-4/clip_large_pretrain_4x256_all_lr1e-4.py new file mode 100644 index 0000000000000000000000000000000000000000..c8cac25c647871dc6370a72142a368b59b7fbed4 --- /dev/null +++ b/clip_large_pretrain_4x256_all_lr1e-4/clip_large_pretrain_4x256_all_lr1e-4.py @@ -0,0 +1,356 @@ +optim_wrapper = dict( + optimizer=dict( + type='AdamW', lr=0.0001, weight_decay=0.3, _scope_='mmpretrain'), + paramwise_cfg=dict( + custom_keys=dict({ + '.cls_token': dict(decay_mult=0.0), + '.pos_embed': dict(decay_mult=0.0) + })), + type='AmpOptimWrapper', + dtype='bfloat16', + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=False, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=4096) +model = dict( + type='ImageClassifier', + backbone=dict( + frozen_stages=24, + type='VisionTransformer', + arch='l', + img_size=224, + patch_size=14, + drop_rate=0.1, + pre_norm=True, + final_norm=False, + init_cfg=dict( + type='Pretrained', + checkpoint='ckpt/openclip-ViT-L-14.pth', + prefix='backbone')), + neck=dict( + type='CLIPProjection', + in_channels=1024, + out_channels=768, + init_cfg=dict( + type='Pretrained', + checkpoint='ckpt/openclip-ViT-L-14.pth', + prefix='backbone')), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=768, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + init_cfg=None), + init_cfg=dict( + type='TruncNormal', layer=['Conv2d', 'Linear'], std=0.02, bias=0.0), + train_cfg=None) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=128, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/all1.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/all2.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/all3.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3fake8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/cc1m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3real8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=128, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV2-1-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/if-dpmsolver++-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=128, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV2-1-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/if-dpmsolver++-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/clip_large_pretrain_4x256_all_lr1e-4' diff --git a/clip_large_pretrain_4x256_all_lr1e-4/epoch_1.pth b/clip_large_pretrain_4x256_all_lr1e-4/epoch_1.pth new file mode 100644 index 0000000000000000000000000000000000000000..8b3138ef1ba7fb5a2597046488f5fe17847b1bdc --- /dev/null +++ b/clip_large_pretrain_4x256_all_lr1e-4/epoch_1.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:fb1acd7febeb2f7a6432c946cd73c97a4a4a1b4cbfb82c6abf387a3115d920ac +size 2438640577 diff --git a/clip_large_pretrain_4x256_all_lr1e-4/epoch_10.pth b/clip_large_pretrain_4x256_all_lr1e-4/epoch_10.pth new file mode 100644 index 0000000000000000000000000000000000000000..e546f2d1cff6971c2c04220e6fd0bcae39ed5987 --- /dev/null +++ b/clip_large_pretrain_4x256_all_lr1e-4/epoch_10.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:51202b43fb037ab5c677d8e095605b394c491506da7b1b5c14cce2542f810f9e +size 2441350337 diff --git a/clip_large_pretrain_4x256_all_lr1e-4/epoch_2.pth b/clip_large_pretrain_4x256_all_lr1e-4/epoch_2.pth new file mode 100644 index 0000000000000000000000000000000000000000..18680fe906fa02165ebabd9548bc8ee3a3205577 --- /dev/null +++ b/clip_large_pretrain_4x256_all_lr1e-4/epoch_2.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:d48e57bbd7e4099b587b9fe47cfeca1fd68e8889fc2aa3bbf39b2803668785bf +size 2438940993 diff --git a/clip_large_pretrain_4x256_all_lr1e-4/epoch_3.pth b/clip_large_pretrain_4x256_all_lr1e-4/epoch_3.pth new file mode 100644 index 0000000000000000000000000000000000000000..01add39e36af916e576c0e268c225c047472366d --- /dev/null +++ b/clip_large_pretrain_4x256_all_lr1e-4/epoch_3.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:631c2a4c13573787ac33c19bdd029321370152725604ffbe2c6789ac2aa03dcc +size 2439240897 diff --git a/clip_large_pretrain_4x256_all_lr1e-4/epoch_4.pth b/clip_large_pretrain_4x256_all_lr1e-4/epoch_4.pth new file mode 100644 index 0000000000000000000000000000000000000000..4afbeb41e346cc83e56cb9981d73056d077633d0 --- /dev/null +++ b/clip_large_pretrain_4x256_all_lr1e-4/epoch_4.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:7c8d979dbe2456820fba801565d44c148a9c34404b9bf826b218e562008f8eae +size 2439540353 diff --git a/clip_large_pretrain_4x256_all_lr1e-4/epoch_5.pth b/clip_large_pretrain_4x256_all_lr1e-4/epoch_5.pth new file mode 100644 index 0000000000000000000000000000000000000000..51517af12cc217822f82b34ee95c3e3288429d7e --- /dev/null +++ b/clip_large_pretrain_4x256_all_lr1e-4/epoch_5.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:2e2811154ca54d4d16a8bbc46d5e883a3a63487c374eaf66f75566b3747f20a3 +size 2439839873 diff --git a/clip_large_pretrain_4x256_all_lr1e-4/epoch_6.pth b/clip_large_pretrain_4x256_all_lr1e-4/epoch_6.pth new file mode 100644 index 0000000000000000000000000000000000000000..551f1f89d50c09ae2bb42d4b47981cfbbb0fde16 --- /dev/null +++ b/clip_large_pretrain_4x256_all_lr1e-4/epoch_6.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:1113d12c286ae3351b2eaf75445291df59cdad2d57f8f8c527595275e54a00a7 +size 2440139457 diff --git a/clip_large_pretrain_4x256_all_lr1e-4/epoch_7.pth b/clip_large_pretrain_4x256_all_lr1e-4/epoch_7.pth new file mode 100644 index 0000000000000000000000000000000000000000..3525238e19fbb60ce709efe4338998dc6f1421a5 --- /dev/null +++ b/clip_large_pretrain_4x256_all_lr1e-4/epoch_7.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:6aed3e131717eb28495fa28f1bfeca6a02805291b164823091df28fc19eec4db +size 2440439809 diff --git a/clip_large_pretrain_4x256_all_lr1e-4/epoch_8.pth b/clip_large_pretrain_4x256_all_lr1e-4/epoch_8.pth new file mode 100644 index 0000000000000000000000000000000000000000..724f9ee3fbb47110c1faaec64fb2ddd06f6c2268 --- /dev/null +++ b/clip_large_pretrain_4x256_all_lr1e-4/epoch_8.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:01933a675742911bc16819d94454502876464cfc3bba3a2b3824cad05854e6a1 +size 2440743489 diff --git a/clip_large_pretrain_4x256_all_lr1e-4/epoch_9.pth b/clip_large_pretrain_4x256_all_lr1e-4/epoch_9.pth new file mode 100644 index 0000000000000000000000000000000000000000..9865f7d2411838ca01571b82353a250f4d04d72e --- /dev/null +++ b/clip_large_pretrain_4x256_all_lr1e-4/epoch_9.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:35fd160c2ab328e2989192b5bea5666604afb07566f922f1ef110fbe1c6796fa +size 2441047169 diff --git a/clip_large_pretrain_4x256_all_lr1e-4/last_checkpoint b/clip_large_pretrain_4x256_all_lr1e-4/last_checkpoint new file mode 100644 index 0000000000000000000000000000000000000000..6ca3e647a66335765a35863a142d22fd630e0e21 --- /dev/null +++ b/clip_large_pretrain_4x256_all_lr1e-4/last_checkpoint @@ -0,0 +1 @@ +/mnt/petrelfs/luzeyu/workspace/fakebench/mmpretrain/workdir/clip_large_pretrain_4x256_all_lr1e-4/epoch_10.pth \ No newline at end of file diff --git a/clip_large_pretrain_4x256_sdv1_lr1e-4_wopostnorm_wneck/20230604_165929/20230604_165929.log b/clip_large_pretrain_4x256_sdv1_lr1e-4_wopostnorm_wneck/20230604_165929/20230604_165929.log new file mode 100644 index 0000000000000000000000000000000000000000..c2445aa28936c8393a025a1644d3fefae1efc018 --- /dev/null +++ b/clip_large_pretrain_4x256_sdv1_lr1e-4_wopostnorm_wneck/20230604_165929/20230604_165929.log @@ -0,0 +1,1324 @@ +2023/06/04 16:59:33 - mmengine - INFO - +------------------------------------------------------------ +System environment: + sys.platform: linux + Python: 3.10.9 (main, Mar 8 2023, 10:47:38) [GCC 11.2.0] + CUDA available: True + numpy_random_seed: 449511118 + GPU 0,1,2,3: NVIDIA A100-SXM4-80GB + CUDA_HOME: /mnt/petrelfs/share/cuda-11.6 + NVCC: Cuda compilation tools, release 11.6, V11.6.124 + GCC: gcc (GCC) 7.5.0 + PyTorch: 1.13.1 + PyTorch compiling details: PyTorch built with: + - GCC 9.3 + - C++ Version: 201402 + - Intel(R) oneAPI Math Kernel Library Version 2021.4-Product Build 20210904 for Intel(R) 64 architecture applications + - Intel(R) MKL-DNN v2.6.0 (Git Hash 52b5f107dd9cf10910aaa19cb47f3abf9b349815) + - OpenMP 201511 (a.k.a. OpenMP 4.5) + - LAPACK is enabled (usually provided by MKL) + - NNPACK is enabled + - CPU capability usage: AVX2 + - CUDA Runtime 11.6 + - NVCC architecture flags: -gencode;arch=compute_37,code=sm_37;-gencode;arch=compute_50,code=sm_50;-gencode;arch=compute_60,code=sm_60;-gencode;arch=compute_61,code=sm_61;-gencode;arch=compute_70,code=sm_70;-gencode;arch=compute_75,code=sm_75;-gencode;arch=compute_80,code=sm_80;-gencode;arch=compute_86,code=sm_86;-gencode;arch=compute_37,code=compute_37 + - CuDNN 8.3.2 (built against CUDA 11.5) + - Magma 2.6.1 + - Build settings: BLAS_INFO=mkl, BUILD_TYPE=Release, CUDA_VERSION=11.6, CUDNN_VERSION=8.3.2, CXX_COMPILER=/opt/rh/devtoolset-9/root/usr/bin/c++, CXX_FLAGS= -fabi-version=11 -Wno-deprecated -fvisibility-inlines-hidden -DUSE_PTHREADPOOL -fopenmp -DNDEBUG -DUSE_KINETO -DUSE_FBGEMM -DUSE_QNNPACK -DUSE_PYTORCH_QNNPACK -DUSE_XNNPACK -DSYMBOLICATE_MOBILE_DEBUG_HANDLE -DEDGE_PROFILER_USE_KINETO -O2 -fPIC -Wno-narrowing -Wall -Wextra -Werror=return-type -Werror=non-virtual-dtor -Wno-missing-field-initializers -Wno-type-limits -Wno-array-bounds -Wno-unknown-pragmas -Wunused-local-typedefs -Wno-unused-parameter -Wno-unused-function -Wno-unused-result -Wno-strict-overflow -Wno-strict-aliasing -Wno-error=deprecated-declarations -Wno-stringop-overflow -Wno-psabi -Wno-error=pedantic -Wno-error=redundant-decls -Wno-error=old-style-cast -fdiagnostics-color=always -faligned-new -Wno-unused-but-set-variable -Wno-maybe-uninitialized -fno-math-errno -fno-trapping-math -Werror=format -Werror=cast-function-type -Wno-stringop-overflow, LAPACK_INFO=mkl, PERF_WITH_AVX=1, PERF_WITH_AVX2=1, PERF_WITH_AVX512=1, TORCH_VERSION=1.13.1, USE_CUDA=ON, USE_CUDNN=ON, USE_EXCEPTION_PTR=1, USE_GFLAGS=OFF, USE_GLOG=OFF, USE_MKL=ON, USE_MKLDNN=ON, USE_MPI=OFF, USE_NCCL=ON, USE_NNPACK=ON, USE_OPENMP=ON, USE_ROCM=OFF, + + TorchVision: 0.14.1 + OpenCV: 4.7.0 + MMEngine: 0.7.3 + +Runtime environment: + cudnn_benchmark: True + mp_cfg: {'mp_start_method': 'fork', 'opencv_num_threads': 0} + dist_cfg: {'backend': 'nccl'} + seed: None + deterministic: False + Distributed launcher: slurm + Distributed training: True + GPU number: 4 +------------------------------------------------------------ + +2023/06/04 16:59:38 - mmengine - INFO - Config: +optim_wrapper = dict( + optimizer=dict( + type='AdamW', lr=0.0001, weight_decay=0.3, _scope_='mmpretrain'), + paramwise_cfg=dict( + custom_keys=dict({ + '.cls_token': dict(decay_mult=0.0), + '.pos_embed': dict(decay_mult=0.0) + })), + type='AmpOptimWrapper', + dtype='bfloat16', + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=False, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=4096) +model = dict( + type='ImageClassifier', + backbone=dict( + frozen_stages=24, + type='VisionTransformer', + arch='l', + img_size=224, + patch_size=14, + drop_rate=0.1, + pre_norm=True, + final_norm=False, + init_cfg=dict( + type='Pretrained', + checkpoint='ckpt/openclip-ViT-L-14.pth', + prefix='backbone')), + neck=dict( + type='CLIPProjection', + in_channels=1024, + out_channels=768, + init_cfg=dict( + type='Pretrained', + checkpoint='ckpt/openclip-ViT-L-14.pth', + prefix='backbone')), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=768, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + init_cfg=None), + init_cfg=dict( + type='TruncNormal', layer=['Conv2d', 'Linear'], std=0.02, bias=0.0), + train_cfg=None) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=128, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stablediffusionV1-5R2-dpmsolver-25-1m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/cc1m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=128, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=128, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/clip_large_pretrain_4x256_sdv1_lr1e-4_wopostnorm_wneck' + +2023/06/04 16:59:51 - mmengine - INFO - Hooks will be executed in the following order: +before_run: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_load_checkpoint: +(ABOVE_NORMAL) EMAHook + -------------------- +before_train: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_train_epoch: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(NORMAL ) DistSamplerSeedHook + -------------------- +before_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook + -------------------- +after_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +after_train_epoch: +(NORMAL ) IterTimerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_val_epoch: +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook + -------------------- +before_val_iter: +(NORMAL ) IterTimerHook + -------------------- +after_val_iter: +(NORMAL ) IterTimerHook +(NORMAL ) VisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_val_epoch: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_save_checkpoint: +(ABOVE_NORMAL) EMAHook + -------------------- +after_train: +(VERY_LOW ) CheckpointHook + -------------------- +before_test_epoch: +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook + -------------------- +before_test_iter: +(NORMAL ) IterTimerHook + -------------------- +after_test_iter: +(NORMAL ) IterTimerHook +(NORMAL ) VisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_test_epoch: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_run: +(BELOW_NORMAL) LoggerHook + -------------------- +2023/06/04 17:00:10 - mmengine - INFO - load backbone in model from: ckpt/openclip-ViT-L-14.pth +2023/06/04 17:00:11 - mmengine - WARNING - The model and loaded state dict do not match exactly + +unexpected key in source state_dict: ln1.weight, ln1.bias + +2023/06/04 17:00:11 - mmengine - INFO - load backbone in model from: ckpt/openclip-ViT-L-14.pth +2023/06/04 17:00:13 - mmengine - WARNING - The model and loaded state dict do not match exactly + +unexpected key in source state_dict: cls_token, pos_embed, patch_embed.projection.weight, pre_norm.weight, pre_norm.bias, layers.0.ln1.weight, layers.0.ln1.bias, layers.0.attn.qkv.weight, layers.0.attn.qkv.bias, layers.0.attn.proj.weight, layers.0.attn.proj.bias, layers.0.ln2.weight, layers.0.ln2.bias, layers.0.ffn.layers.0.0.weight, layers.0.ffn.layers.0.0.bias, layers.0.ffn.layers.1.weight, layers.0.ffn.layers.1.bias, layers.1.ln1.weight, layers.1.ln1.bias, layers.1.attn.qkv.weight, layers.1.attn.qkv.bias, layers.1.attn.proj.weight, layers.1.attn.proj.bias, layers.1.ln2.weight, layers.1.ln2.bias, layers.1.ffn.layers.0.0.weight, layers.1.ffn.layers.0.0.bias, layers.1.ffn.layers.1.weight, layers.1.ffn.layers.1.bias, layers.2.ln1.weight, layers.2.ln1.bias, layers.2.attn.qkv.weight, layers.2.attn.qkv.bias, layers.2.attn.proj.weight, layers.2.attn.proj.bias, layers.2.ln2.weight, layers.2.ln2.bias, layers.2.ffn.layers.0.0.weight, layers.2.ffn.layers.0.0.bias, layers.2.ffn.layers.1.weight, layers.2.ffn.layers.1.bias, layers.3.ln1.weight, layers.3.ln1.bias, layers.3.attn.qkv.weight, layers.3.attn.qkv.bias, layers.3.attn.proj.weight, layers.3.attn.proj.bias, layers.3.ln2.weight, layers.3.ln2.bias, layers.3.ffn.layers.0.0.weight, layers.3.ffn.layers.0.0.bias, layers.3.ffn.layers.1.weight, layers.3.ffn.layers.1.bias, layers.4.ln1.weight, layers.4.ln1.bias, layers.4.attn.qkv.weight, layers.4.attn.qkv.bias, layers.4.attn.proj.weight, layers.4.attn.proj.bias, layers.4.ln2.weight, layers.4.ln2.bias, layers.4.ffn.layers.0.0.weight, layers.4.ffn.layers.0.0.bias, layers.4.ffn.layers.1.weight, layers.4.ffn.layers.1.bias, layers.5.ln1.weight, layers.5.ln1.bias, layers.5.attn.qkv.weight, layers.5.attn.qkv.bias, layers.5.attn.proj.weight, layers.5.attn.proj.bias, layers.5.ln2.weight, layers.5.ln2.bias, layers.5.ffn.layers.0.0.weight, layers.5.ffn.layers.0.0.bias, layers.5.ffn.layers.1.weight, layers.5.ffn.layers.1.bias, layers.6.ln1.weight, layers.6.ln1.bias, layers.6.attn.qkv.weight, layers.6.attn.qkv.bias, layers.6.attn.proj.weight, layers.6.attn.proj.bias, layers.6.ln2.weight, layers.6.ln2.bias, layers.6.ffn.layers.0.0.weight, layers.6.ffn.layers.0.0.bias, layers.6.ffn.layers.1.weight, layers.6.ffn.layers.1.bias, layers.7.ln1.weight, layers.7.ln1.bias, layers.7.attn.qkv.weight, layers.7.attn.qkv.bias, layers.7.attn.proj.weight, layers.7.attn.proj.bias, layers.7.ln2.weight, layers.7.ln2.bias, layers.7.ffn.layers.0.0.weight, layers.7.ffn.layers.0.0.bias, layers.7.ffn.layers.1.weight, layers.7.ffn.layers.1.bias, layers.8.ln1.weight, layers.8.ln1.bias, layers.8.attn.qkv.weight, layers.8.attn.qkv.bias, layers.8.attn.proj.weight, layers.8.attn.proj.bias, layers.8.ln2.weight, layers.8.ln2.bias, layers.8.ffn.layers.0.0.weight, layers.8.ffn.layers.0.0.bias, layers.8.ffn.layers.1.weight, layers.8.ffn.layers.1.bias, layers.9.ln1.weight, layers.9.ln1.bias, layers.9.attn.qkv.weight, layers.9.attn.qkv.bias, layers.9.attn.proj.weight, layers.9.attn.proj.bias, layers.9.ln2.weight, layers.9.ln2.bias, layers.9.ffn.layers.0.0.weight, layers.9.ffn.layers.0.0.bias, layers.9.ffn.layers.1.weight, layers.9.ffn.layers.1.bias, layers.10.ln1.weight, layers.10.ln1.bias, layers.10.attn.qkv.weight, layers.10.attn.qkv.bias, layers.10.attn.proj.weight, layers.10.attn.proj.bias, layers.10.ln2.weight, layers.10.ln2.bias, layers.10.ffn.layers.0.0.weight, layers.10.ffn.layers.0.0.bias, layers.10.ffn.layers.1.weight, layers.10.ffn.layers.1.bias, layers.11.ln1.weight, layers.11.ln1.bias, layers.11.attn.qkv.weight, layers.11.attn.qkv.bias, layers.11.attn.proj.weight, layers.11.attn.proj.bias, layers.11.ln2.weight, layers.11.ln2.bias, layers.11.ffn.layers.0.0.weight, layers.11.ffn.layers.0.0.bias, layers.11.ffn.layers.1.weight, layers.11.ffn.layers.1.bias, layers.12.ln1.weight, layers.12.ln1.bias, layers.12.attn.qkv.weight, layers.12.attn.qkv.bias, layers.12.attn.proj.weight, layers.12.attn.proj.bias, layers.12.ln2.weight, layers.12.ln2.bias, layers.12.ffn.layers.0.0.weight, layers.12.ffn.layers.0.0.bias, layers.12.ffn.layers.1.weight, layers.12.ffn.layers.1.bias, layers.13.ln1.weight, layers.13.ln1.bias, layers.13.attn.qkv.weight, layers.13.attn.qkv.bias, layers.13.attn.proj.weight, layers.13.attn.proj.bias, layers.13.ln2.weight, layers.13.ln2.bias, layers.13.ffn.layers.0.0.weight, layers.13.ffn.layers.0.0.bias, layers.13.ffn.layers.1.weight, layers.13.ffn.layers.1.bias, layers.14.ln1.weight, layers.14.ln1.bias, layers.14.attn.qkv.weight, layers.14.attn.qkv.bias, layers.14.attn.proj.weight, layers.14.attn.proj.bias, layers.14.ln2.weight, layers.14.ln2.bias, layers.14.ffn.layers.0.0.weight, layers.14.ffn.layers.0.0.bias, layers.14.ffn.layers.1.weight, layers.14.ffn.layers.1.bias, layers.15.ln1.weight, layers.15.ln1.bias, layers.15.attn.qkv.weight, layers.15.attn.qkv.bias, layers.15.attn.proj.weight, layers.15.attn.proj.bias, layers.15.ln2.weight, layers.15.ln2.bias, layers.15.ffn.layers.0.0.weight, layers.15.ffn.layers.0.0.bias, layers.15.ffn.layers.1.weight, layers.15.ffn.layers.1.bias, layers.16.ln1.weight, layers.16.ln1.bias, layers.16.attn.qkv.weight, layers.16.attn.qkv.bias, layers.16.attn.proj.weight, layers.16.attn.proj.bias, layers.16.ln2.weight, layers.16.ln2.bias, layers.16.ffn.layers.0.0.weight, layers.16.ffn.layers.0.0.bias, layers.16.ffn.layers.1.weight, layers.16.ffn.layers.1.bias, layers.17.ln1.weight, layers.17.ln1.bias, layers.17.attn.qkv.weight, layers.17.attn.qkv.bias, layers.17.attn.proj.weight, layers.17.attn.proj.bias, layers.17.ln2.weight, layers.17.ln2.bias, layers.17.ffn.layers.0.0.weight, layers.17.ffn.layers.0.0.bias, layers.17.ffn.layers.1.weight, layers.17.ffn.layers.1.bias, layers.18.ln1.weight, layers.18.ln1.bias, layers.18.attn.qkv.weight, layers.18.attn.qkv.bias, layers.18.attn.proj.weight, layers.18.attn.proj.bias, layers.18.ln2.weight, layers.18.ln2.bias, layers.18.ffn.layers.0.0.weight, layers.18.ffn.layers.0.0.bias, layers.18.ffn.layers.1.weight, layers.18.ffn.layers.1.bias, layers.19.ln1.weight, layers.19.ln1.bias, layers.19.attn.qkv.weight, layers.19.attn.qkv.bias, layers.19.attn.proj.weight, layers.19.attn.proj.bias, layers.19.ln2.weight, layers.19.ln2.bias, layers.19.ffn.layers.0.0.weight, layers.19.ffn.layers.0.0.bias, layers.19.ffn.layers.1.weight, layers.19.ffn.layers.1.bias, layers.20.ln1.weight, layers.20.ln1.bias, layers.20.attn.qkv.weight, layers.20.attn.qkv.bias, layers.20.attn.proj.weight, layers.20.attn.proj.bias, layers.20.ln2.weight, layers.20.ln2.bias, layers.20.ffn.layers.0.0.weight, layers.20.ffn.layers.0.0.bias, layers.20.ffn.layers.1.weight, layers.20.ffn.layers.1.bias, layers.21.ln1.weight, layers.21.ln1.bias, layers.21.attn.qkv.weight, layers.21.attn.qkv.bias, layers.21.attn.proj.weight, layers.21.attn.proj.bias, layers.21.ln2.weight, layers.21.ln2.bias, layers.21.ffn.layers.0.0.weight, layers.21.ffn.layers.0.0.bias, layers.21.ffn.layers.1.weight, layers.21.ffn.layers.1.bias, layers.22.ln1.weight, layers.22.ln1.bias, layers.22.attn.qkv.weight, layers.22.attn.qkv.bias, layers.22.attn.proj.weight, layers.22.attn.proj.bias, layers.22.ln2.weight, layers.22.ln2.bias, layers.22.ffn.layers.0.0.weight, layers.22.ffn.layers.0.0.bias, layers.22.ffn.layers.1.weight, layers.22.ffn.layers.1.bias, layers.23.ln1.weight, layers.23.ln1.bias, layers.23.attn.qkv.weight, layers.23.attn.qkv.bias, layers.23.attn.proj.weight, layers.23.attn.proj.bias, layers.23.ln2.weight, layers.23.ln2.bias, layers.23.ffn.layers.0.0.weight, layers.23.ffn.layers.0.0.bias, layers.23.ffn.layers.1.weight, layers.23.ffn.layers.1.bias, ln1.weight, ln1.bias + +missing keys in source state_dict: proj + +Name of parameter - Initialization information + +backbone.cls_token - torch.Size([1, 1, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.pos_embed - torch.Size([1, 257, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.patch_embed.projection.weight - torch.Size([1024, 3, 14, 14]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.0.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.0.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.0.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.0.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.0.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.0.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.0.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.0.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.0.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.0.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.0.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.0.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.1.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.1.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.1.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.1.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.1.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.1.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.1.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.1.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.1.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.1.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.1.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.1.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.2.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.2.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.2.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.2.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.2.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.2.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.2.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.2.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.2.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.2.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.2.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.2.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.3.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.3.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.3.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.3.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.3.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.3.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.3.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.3.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.3.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.3.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.3.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.3.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.4.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.4.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.4.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.4.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.4.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.4.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.4.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.4.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.4.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.4.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.4.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.4.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.5.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.5.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.5.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.5.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.5.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.5.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.5.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.5.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.5.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.5.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.5.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.5.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.6.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.6.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.6.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.6.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.6.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.6.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.6.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.6.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.6.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.6.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.6.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.6.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.7.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.7.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.7.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.7.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.7.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.7.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.7.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.7.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.7.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.7.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.7.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.7.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.8.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.8.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.8.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.8.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.8.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.8.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.8.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.8.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.8.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.8.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.8.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.8.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.9.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.9.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.9.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.9.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.9.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.9.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.9.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.9.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.9.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.9.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.9.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.9.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.10.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.10.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.10.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.10.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.10.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.10.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.10.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.10.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.10.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.10.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.10.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.10.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.11.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.11.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.11.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.11.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.11.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.11.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.11.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.11.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.11.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.11.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.11.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.11.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.12.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.12.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.12.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.12.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.12.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.12.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.12.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.12.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.12.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.12.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.12.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.12.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.13.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.13.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.13.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.13.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.13.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.13.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.13.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.13.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.13.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.13.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.13.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.13.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.14.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.14.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.14.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.14.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.14.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.14.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.14.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.14.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.14.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.14.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.14.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.14.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.15.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.15.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.15.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.15.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.15.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.15.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.15.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.15.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.15.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.15.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.15.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.15.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.16.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.16.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.16.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.16.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.16.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.16.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.16.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.16.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.16.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.16.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.16.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.16.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.17.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.17.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.17.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.17.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.17.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.17.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.17.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.17.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.17.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.17.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.17.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.17.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.18.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.18.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.18.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.18.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.18.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.18.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.18.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.18.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.18.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.18.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.18.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.18.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.19.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.19.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.19.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.19.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.19.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.19.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.19.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.19.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.19.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.19.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.19.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.19.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.20.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.20.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.20.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.20.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.20.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.20.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.20.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.20.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.20.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.20.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.20.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.20.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.21.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.21.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.21.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.21.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.21.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.21.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.21.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.21.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.21.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.21.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.21.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.21.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.22.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.22.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.22.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.22.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.22.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.22.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.22.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.22.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.22.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.22.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.22.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.22.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.23.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.23.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.23.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.23.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.23.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.23.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.23.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.23.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.23.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.23.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.23.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.23.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.pre_norm.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.pre_norm.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +neck.proj - torch.Size([1024, 768]): +The value is the same before and after calling `init_weights` of ImageClassifier + +head.fc.weight - torch.Size([2, 768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +head.fc.bias - torch.Size([2]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 +2023/06/04 17:00:13 - mmengine - WARNING - "FileClient" will be deprecated in future. Please use io functions in https://mmengine.readthedocs.io/en/latest/api/fileio.html#file-io +2023/06/04 17:00:13 - mmengine - WARNING - "HardDiskBackend" is the alias of "LocalBackend" and the former will be deprecated in future. +2023/06/04 17:00:13 - mmengine - INFO - Checkpoints will be saved to /mnt/petrelfs/luzeyu/workspace/fakebench/mmpretrain/workdir/clip_large_pretrain_4x256_sdv1_lr1e-4_wopostnorm_wneck. +2023/06/04 17:01:24 - mmengine - INFO - Epoch(train) [1][ 100/3907] lr: 9.9999e-05 eta: 7:45:19 time: 0.6274 data_time: 0.0020 memory: 44139 loss: 0.5037 +2023/06/04 17:02:27 - mmengine - INFO - Epoch(train) [1][ 200/3907] lr: 9.9994e-05 eta: 7:15:41 time: 0.6289 data_time: 0.0017 memory: 44139 loss: 0.4562 +2023/06/04 17:03:30 - mmengine - INFO - Epoch(train) [1][ 300/3907] lr: 9.9987e-05 eta: 7:05:18 time: 0.6298 data_time: 0.0015 memory: 44139 loss: 0.4655 +2023/06/04 17:04:33 - mmengine - INFO - Epoch(train) [1][ 400/3907] lr: 9.9977e-05 eta: 7:00:01 time: 0.6311 data_time: 0.0014 memory: 44139 loss: 0.4383 +2023/06/04 17:05:36 - mmengine - INFO - Epoch(train) [1][ 500/3907] lr: 9.9964e-05 eta: 6:56:08 time: 0.6297 data_time: 0.0015 memory: 44139 loss: 0.4066 +2023/06/04 17:06:39 - mmengine - INFO - Epoch(train) [1][ 600/3907] lr: 9.9948e-05 eta: 6:53:10 time: 0.6302 data_time: 0.0015 memory: 44139 loss: 0.4381 +2023/06/04 17:07:42 - mmengine - INFO - Epoch(train) [1][ 700/3907] lr: 9.9929e-05 eta: 6:50:50 time: 0.6333 data_time: 0.0016 memory: 44139 loss: 0.4434 +2023/06/04 17:08:45 - mmengine - INFO - Epoch(train) [1][ 800/3907] lr: 9.9907e-05 eta: 6:48:49 time: 0.6296 data_time: 0.0019 memory: 44139 loss: 0.4539 +2023/06/04 17:09:48 - mmengine - INFO - Epoch(train) [1][ 900/3907] lr: 9.9882e-05 eta: 6:46:58 time: 0.6299 data_time: 0.0017 memory: 44139 loss: 0.4135 +2023/06/04 17:10:51 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv1_lr1e-4_20230604_165929 +2023/06/04 17:10:51 - mmengine - INFO - Epoch(train) [1][1000/3907] lr: 9.9855e-05 eta: 6:45:16 time: 0.6296 data_time: 0.0017 memory: 44139 loss: 0.3775 +2023/06/04 17:11:54 - mmengine - INFO - Epoch(train) [1][1100/3907] lr: 9.9824e-05 eta: 6:43:42 time: 0.6292 data_time: 0.0016 memory: 44139 loss: 0.3822 +2023/06/04 17:12:57 - mmengine - INFO - Epoch(train) [1][1200/3907] lr: 9.9791e-05 eta: 6:42:12 time: 0.6307 data_time: 0.0015 memory: 44139 loss: 0.4045 +2023/06/04 17:14:00 - mmengine - INFO - Epoch(train) [1][1300/3907] lr: 9.9755e-05 eta: 6:40:46 time: 0.6284 data_time: 0.0014 memory: 44139 loss: 0.4228 +2023/06/04 17:15:03 - mmengine - INFO - Epoch(train) [1][1400/3907] lr: 9.9716e-05 eta: 6:39:22 time: 0.6284 data_time: 0.0017 memory: 44139 loss: 0.3893 +2023/06/04 17:16:06 - mmengine - INFO - Epoch(train) [1][1500/3907] lr: 9.9674e-05 eta: 6:38:02 time: 0.6301 data_time: 0.0016 memory: 44139 loss: 0.3855 +2023/06/04 17:17:09 - mmengine - INFO - Epoch(train) [1][1600/3907] lr: 9.9629e-05 eta: 6:36:47 time: 0.6283 data_time: 0.0015 memory: 44139 loss: 0.3852 +2023/06/04 17:18:12 - mmengine - INFO - Epoch(train) [1][1700/3907] lr: 9.9581e-05 eta: 6:35:29 time: 0.6285 data_time: 0.0016 memory: 44139 loss: 0.3946 +2023/06/04 17:19:15 - mmengine - INFO - Epoch(train) [1][1800/3907] lr: 9.9530e-05 eta: 6:34:13 time: 0.6283 data_time: 0.0015 memory: 44139 loss: 0.3993 +2023/06/04 17:20:18 - mmengine - INFO - Epoch(train) [1][1900/3907] lr: 9.9476e-05 eta: 6:32:57 time: 0.6278 data_time: 0.0015 memory: 44139 loss: 0.3996 +2023/06/04 17:21:21 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv1_lr1e-4_20230604_165929 +2023/06/04 17:21:21 - mmengine - INFO - Epoch(train) [1][2000/3907] lr: 9.9420e-05 eta: 6:31:42 time: 0.6274 data_time: 0.0018 memory: 44139 loss: 0.3753 +2023/06/04 17:22:24 - mmengine - INFO - Epoch(train) [1][2100/3907] lr: 9.9361e-05 eta: 6:30:29 time: 0.6294 data_time: 0.0015 memory: 44139 loss: 0.3721 +2023/06/04 17:23:26 - mmengine - INFO - Epoch(train) [1][2200/3907] lr: 9.9298e-05 eta: 6:29:17 time: 0.6286 data_time: 0.0015 memory: 44139 loss: 0.3699 +2023/06/04 17:24:29 - mmengine - INFO - Epoch(train) [1][2300/3907] lr: 9.9233e-05 eta: 6:28:05 time: 0.6286 data_time: 0.0015 memory: 44139 loss: 0.3799 +2023/06/04 17:25:32 - mmengine - INFO - Epoch(train) [1][2400/3907] lr: 9.9165e-05 eta: 6:26:54 time: 0.6274 data_time: 0.0015 memory: 44139 loss: 0.3607 +2023/06/04 17:26:35 - mmengine - INFO - Epoch(train) [1][2500/3907] lr: 9.9095e-05 eta: 6:25:45 time: 0.6284 data_time: 0.0016 memory: 44139 loss: 0.3979 +2023/06/04 17:27:38 - mmengine - INFO - Epoch(train) [1][2600/3907] lr: 9.9021e-05 eta: 6:24:36 time: 0.6287 data_time: 0.0014 memory: 44139 loss: 0.3836 +2023/06/04 17:28:41 - mmengine - INFO - Epoch(train) [1][2700/3907] lr: 9.8944e-05 eta: 6:23:27 time: 0.6277 data_time: 0.0016 memory: 44139 loss: 0.3638 +2023/06/04 17:29:44 - mmengine - INFO - Epoch(train) [1][2800/3907] lr: 9.8865e-05 eta: 6:22:18 time: 0.6289 data_time: 0.0017 memory: 44139 loss: 0.3390 +2023/06/04 17:30:46 - mmengine - INFO - Epoch(train) [1][2900/3907] lr: 9.8783e-05 eta: 6:21:10 time: 0.6275 data_time: 0.0015 memory: 44139 loss: 0.3645 +2023/06/04 17:31:49 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv1_lr1e-4_20230604_165929 +2023/06/04 17:31:49 - mmengine - INFO - Epoch(train) [1][3000/3907] lr: 9.8698e-05 eta: 6:20:01 time: 0.6274 data_time: 0.0014 memory: 44139 loss: 0.3369 +2023/06/04 17:32:52 - mmengine - INFO - Epoch(train) [1][3100/3907] lr: 9.8610e-05 eta: 6:18:53 time: 0.6272 data_time: 0.0015 memory: 44139 loss: 0.3641 +2023/06/04 17:33:55 - mmengine - INFO - Epoch(train) [1][3200/3907] lr: 9.8519e-05 eta: 6:17:45 time: 0.6285 data_time: 0.0020 memory: 44139 loss: 0.3568 +2023/06/04 17:34:58 - mmengine - INFO - Epoch(train) [1][3300/3907] lr: 9.8426e-05 eta: 6:16:38 time: 0.6293 data_time: 0.0015 memory: 44139 loss: 0.3258 +2023/06/04 17:36:00 - mmengine - INFO - Epoch(train) [1][3400/3907] lr: 9.8330e-05 eta: 6:15:31 time: 0.6287 data_time: 0.0015 memory: 44139 loss: 0.3230 +2023/06/04 17:37:03 - mmengine - INFO - Epoch(train) [1][3500/3907] lr: 9.8231e-05 eta: 6:14:25 time: 0.6282 data_time: 0.0016 memory: 44139 loss: 0.3450 +2023/06/04 17:38:06 - mmengine - INFO - Epoch(train) [1][3600/3907] lr: 9.8129e-05 eta: 6:13:19 time: 0.6286 data_time: 0.0017 memory: 44139 loss: 0.3551 +2023/06/04 17:39:09 - mmengine - INFO - Epoch(train) [1][3700/3907] lr: 9.8024e-05 eta: 6:12:13 time: 0.6288 data_time: 0.0017 memory: 44139 loss: 0.3679 +2023/06/04 17:40:12 - mmengine - INFO - Epoch(train) [1][3800/3907] lr: 9.7917e-05 eta: 6:11:07 time: 0.6290 data_time: 0.0015 memory: 44139 loss: 0.3537 +2023/06/04 17:41:15 - mmengine - INFO - Epoch(train) [1][3900/3907] lr: 9.7806e-05 eta: 6:10:02 time: 0.6274 data_time: 0.0015 memory: 44139 loss: 0.3821 +2023/06/04 17:41:19 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv1_lr1e-4_20230604_165929 +2023/06/04 17:41:19 - mmengine - INFO - Saving checkpoint at 1 epochs +2023/06/04 17:42:56 - mmengine - INFO - Epoch(val) [1][57/57] accuracy/top1: 81.3020 single-label/precision_classwise: [79.19664764404297, 84.79105377197266] single-label/recall_classwise: [89.61555480957031, 71.09302520751953] single-label/f1-score_classwise: [84.08457946777344, 77.34019470214844] data_time: 0.0441 time: 1.3513 +2023/06/04 17:44:03 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv1_lr1e-4_20230604_165929 +2023/06/04 17:44:08 - mmengine - INFO - Epoch(train) [2][ 100/3907] lr: 9.7685e-05 eta: 6:10:04 time: 0.6321 data_time: 0.0019 memory: 44139 loss: 0.3356 +2023/06/04 17:45:11 - mmengine - INFO - Epoch(train) [2][ 200/3907] lr: 9.7570e-05 eta: 6:08:59 time: 0.6314 data_time: 0.0018 memory: 44138 loss: 0.3296 +2023/06/04 17:46:14 - mmengine - INFO - Epoch(train) [2][ 300/3907] lr: 9.7451e-05 eta: 6:07:55 time: 0.6315 data_time: 0.0017 memory: 44138 loss: 0.3412 +2023/06/04 17:47:17 - mmengine - INFO - Epoch(train) [2][ 400/3907] lr: 9.7329e-05 eta: 6:06:50 time: 0.6293 data_time: 0.0018 memory: 44138 loss: 0.3488 +2023/06/04 17:48:21 - mmengine - INFO - Epoch(train) [2][ 500/3907] lr: 9.7205e-05 eta: 6:05:50 time: 0.6305 data_time: 0.0016 memory: 44138 loss: 0.3149 +2023/06/04 17:49:24 - mmengine - INFO - Epoch(train) [2][ 600/3907] lr: 9.7078e-05 eta: 6:04:44 time: 0.6295 data_time: 0.0014 memory: 44138 loss: 0.3519 +2023/06/04 17:50:27 - mmengine - INFO - Epoch(train) [2][ 700/3907] lr: 9.6949e-05 eta: 6:03:38 time: 0.6305 data_time: 0.0015 memory: 44138 loss: 0.3351 +2023/06/04 17:51:30 - mmengine - INFO - Epoch(train) [2][ 800/3907] lr: 9.6816e-05 eta: 6:02:33 time: 0.6291 data_time: 0.0014 memory: 44138 loss: 0.3389 +2023/06/04 17:52:33 - mmengine - INFO - Epoch(train) [2][ 900/3907] lr: 9.6681e-05 eta: 6:01:27 time: 0.6307 data_time: 0.0016 memory: 44138 loss: 0.3430 +2023/06/04 17:53:36 - mmengine - INFO - Epoch(train) [2][1000/3907] lr: 9.6544e-05 eta: 6:00:23 time: 0.6299 data_time: 0.0016 memory: 44138 loss: 0.3348 diff --git a/clip_large_pretrain_4x256_sdv1_lr1e-4_wopostnorm_wneck/20230604_165929/vis_data/20230604_165929.json b/clip_large_pretrain_4x256_sdv1_lr1e-4_wopostnorm_wneck/20230604_165929/vis_data/20230604_165929.json new file mode 100644 index 0000000000000000000000000000000000000000..4cbcd42356c6323b232bd79f8e7055e107a46561 --- /dev/null +++ b/clip_large_pretrain_4x256_sdv1_lr1e-4_wopostnorm_wneck/20230604_165929/vis_data/20230604_165929.json @@ -0,0 +1,50 @@ +{"lr": 9.999857418382631e-05, "data_time": 0.0020378589630126952, "loss": 0.5037091255187989, "time": 0.6273816823959351, "epoch": 1, "memory": 44139, "step": 100} +{"lr": 9.999423907359506e-05, "data_time": 0.001653289794921875, "loss": 0.4562071800231934, "time": 0.6289210319519043, "epoch": 1, "memory": 44139, "step": 200} +{"lr": 9.998699480412073e-05, "data_time": 0.0015316486358642578, "loss": 0.46545201539993286, "time": 0.6298252820968628, "epoch": 1, "memory": 44139, "step": 300} +{"lr": 9.997684184379085e-05, "data_time": 0.0014403104782104493, "loss": 0.43828535079956055, "time": 0.6311163187026978, "epoch": 1, "memory": 44139, "step": 400} +{"lr": 9.996378084905784e-05, "data_time": 0.001513051986694336, "loss": 0.40659589171409605, "time": 0.6297130823135376, "epoch": 1, "memory": 44139, "step": 500} +{"lr": 9.994781266439667e-05, "data_time": 0.001509380340576172, "loss": 0.4381151467561722, "time": 0.6301935195922852, "epoch": 1, "memory": 44139, "step": 600} +{"lr": 9.992893832225061e-05, "data_time": 0.0015932321548461914, "loss": 0.44335839748382566, "time": 0.6333444833755493, "epoch": 1, "memory": 44139, "step": 700} +{"lr": 9.990715904296405e-05, "data_time": 0.0018578767776489258, "loss": 0.45387116372585296, "time": 0.6295591831207276, "epoch": 1, "memory": 44139, "step": 800} +{"lr": 9.988247623470371e-05, "data_time": 0.0016877174377441407, "loss": 0.4134774267673492, "time": 0.6299310684204101, "epoch": 1, "memory": 44139, "step": 900} +{"lr": 9.985489149336749e-05, "data_time": 0.001697373390197754, "loss": 0.37754403352737426, "time": 0.6295555830001831, "epoch": 1, "memory": 44139, "step": 1000} +{"lr": 9.982440660248187e-05, "data_time": 0.0016101598739624023, "loss": 0.38221445083618166, "time": 0.6291843175888061, "epoch": 1, "memory": 44139, "step": 1100} +{"lr": 9.979102353308574e-05, "data_time": 0.0015021562576293945, "loss": 0.4044693887233734, "time": 0.6307098388671875, "epoch": 1, "memory": 44139, "step": 1200} +{"lr": 9.975474444360358e-05, "data_time": 0.0014406204223632812, "loss": 0.42279147505760195, "time": 0.6283769130706787, "epoch": 1, "memory": 44139, "step": 1300} +{"lr": 9.97155716797056e-05, "data_time": 0.0016569137573242188, "loss": 0.3893288761377335, "time": 0.6284325361251831, "epoch": 1, "memory": 44139, "step": 1400} +{"lr": 9.967350777415632e-05, "data_time": 0.0015644073486328126, "loss": 0.38546980917453766, "time": 0.6300941467285156, "epoch": 1, "memory": 44139, "step": 1500} +{"lr": 9.962855544665059e-05, "data_time": 0.0015160799026489257, "loss": 0.38515550792217257, "time": 0.6283077478408814, "epoch": 1, "memory": 44139, "step": 1600} +{"lr": 9.958071760363773e-05, "data_time": 0.001571059226989746, "loss": 0.3945955365896225, "time": 0.6284991264343261, "epoch": 1, "memory": 44139, "step": 1700} +{"lr": 9.952999733813382e-05, "data_time": 0.0015401124954223632, "loss": 0.39930541515350343, "time": 0.6282625913619995, "epoch": 1, "memory": 44139, "step": 1800} +{"lr": 9.947639792952148e-05, "data_time": 0.0014751195907592774, "loss": 0.3996291786432266, "time": 0.6278280735015869, "epoch": 1, "memory": 44139, "step": 1900} +{"lr": 9.941992284333813e-05, "data_time": 0.0017621040344238282, "loss": 0.3752891838550568, "time": 0.6273766994476319, "epoch": 1, "memory": 44139, "step": 2000} +{"lr": 9.93605757310516e-05, "data_time": 0.001543879508972168, "loss": 0.3720944285392761, "time": 0.6294212341308594, "epoch": 1, "memory": 44139, "step": 2100} +{"lr": 9.929836042982421e-05, "data_time": 0.001537013053894043, "loss": 0.36986070275306704, "time": 0.6286076545715332, "epoch": 1, "memory": 44139, "step": 2200} +{"lr": 9.923328096226467e-05, "data_time": 0.0015148401260375976, "loss": 0.37991597354412077, "time": 0.628557276725769, "epoch": 1, "memory": 44139, "step": 2300} +{"lr": 9.916534153616819e-05, "data_time": 0.0015467166900634765, "loss": 0.36074067652225494, "time": 0.6274428367614746, "epoch": 1, "memory": 44139, "step": 2400} +{"lr": 9.909454654424387e-05, "data_time": 0.001568436622619629, "loss": 0.39794217646121977, "time": 0.6283650636672974, "epoch": 1, "memory": 44139, "step": 2500} +{"lr": 9.902090056383109e-05, "data_time": 0.0014369726181030274, "loss": 0.3836180180311203, "time": 0.6287459135055542, "epoch": 1, "memory": 44139, "step": 2600} +{"lr": 9.894440835660365e-05, "data_time": 0.001600170135498047, "loss": 0.36383976936340334, "time": 0.6277165651321411, "epoch": 1, "memory": 44139, "step": 2700} +{"lr": 9.886507486826162e-05, "data_time": 0.0016936779022216797, "loss": 0.3389661222696304, "time": 0.6289095878601074, "epoch": 1, "memory": 44139, "step": 2800} +{"lr": 9.878290522821154e-05, "data_time": 0.0014937162399291993, "loss": 0.36450161337852477, "time": 0.627508282661438, "epoch": 1, "memory": 44139, "step": 2900} +{"lr": 9.86979047492352e-05, "data_time": 0.0013707637786865234, "loss": 0.33687353432178496, "time": 0.6274271965026855, "epoch": 1, "memory": 44139, "step": 3000} +{"lr": 9.861007892714578e-05, "data_time": 0.0015142440795898437, "loss": 0.36411143839359283, "time": 0.6272186994552612, "epoch": 1, "memory": 44139, "step": 3100} +{"lr": 9.851943344043239e-05, "data_time": 0.001987195014953613, "loss": 0.3568147510290146, "time": 0.6285130739212036, "epoch": 1, "memory": 44139, "step": 3200} +{"lr": 9.842597414989319e-05, "data_time": 0.0015331268310546874, "loss": 0.3257623165845871, "time": 0.6293205738067627, "epoch": 1, "memory": 44139, "step": 3300} +{"lr": 9.832970709825637e-05, "data_time": 0.0015496492385864257, "loss": 0.3229869931936264, "time": 0.6286638021469116, "epoch": 1, "memory": 44139, "step": 3400} +{"lr": 9.823063850978968e-05, "data_time": 0.001575303077697754, "loss": 0.3449849545955658, "time": 0.6282416582107544, "epoch": 1, "memory": 44139, "step": 3500} +{"lr": 9.81287747898975e-05, "data_time": 0.0016646862030029296, "loss": 0.3551407277584076, "time": 0.6286078691482544, "epoch": 1, "memory": 44139, "step": 3600} +{"lr": 9.802412252470715e-05, "data_time": 0.0016801118850708007, "loss": 0.3679080069065094, "time": 0.6287972688674927, "epoch": 1, "memory": 44139, "step": 3700} +{"lr": 9.791668848064258e-05, "data_time": 0.001544189453125, "loss": 0.3536813914775848, "time": 0.6289635181427002, "epoch": 1, "memory": 44139, "step": 3800} +{"lr": 9.780647960398751e-05, "data_time": 0.0014657020568847657, "loss": 0.38214276283979415, "time": 0.6274443864822388, "epoch": 1, "memory": 44139, "step": 3900} +{"accuracy/top1": 81.30197143554688, "data_time": 0.0441409663150185, "time": 1.351335931242558, "step": 1} +{"lr": 9.7685491196646e-05, "data_time": 0.0018814563751220702, "loss": 0.33556934297084806, "time": 0.6321202993392945, "epoch": 2, "memory": 44139, "step": 4007} +{"lr": 9.756956126085676e-05, "data_time": 0.001814866065979004, "loss": 0.3295762687921524, "time": 0.6314014196395874, "epoch": 2, "memory": 44138, "step": 4107} +{"lr": 9.745087893642482e-05, "data_time": 0.001662445068359375, "loss": 0.3412328392267227, "time": 0.6315424919128418, "epoch": 2, "memory": 44138, "step": 4207} +{"lr": 9.73294518969056e-05, "data_time": 0.0018392086029052734, "loss": 0.34878638088703157, "time": 0.6292943239212037, "epoch": 2, "memory": 44138, "step": 4307} +{"lr": 9.72052879933172e-05, "data_time": 0.0016124963760375977, "loss": 0.3149163812398911, "time": 0.6304901361465454, "epoch": 2, "memory": 44138, "step": 4407} +{"lr": 9.707839525363358e-05, "data_time": 0.0014074325561523437, "loss": 0.3519111305475235, "time": 0.6295089006423951, "epoch": 2, "memory": 44138, "step": 4507} +{"lr": 9.69487818822646e-05, "data_time": 0.0015254735946655274, "loss": 0.3351467907428741, "time": 0.6305028438568115, "epoch": 2, "memory": 44138, "step": 4607} +{"lr": 9.681645625952616e-05, "data_time": 0.0013847112655639648, "loss": 0.33887229561805726, "time": 0.6290833473205566, "epoch": 2, "memory": 44138, "step": 4707} +{"lr": 9.668142694109826e-05, "data_time": 0.0015581130981445312, "loss": 0.3430200695991516, "time": 0.6307183027267456, "epoch": 2, "memory": 44138, "step": 4807} +{"lr": 9.654370265747162e-05, "data_time": 0.0015548229217529296, "loss": 0.3347715586423874, "time": 0.6298733949661255, "epoch": 2, "memory": 44138, "step": 4907} diff --git a/clip_large_pretrain_4x256_sdv1_lr1e-4_wopostnorm_wneck/20230604_165929/vis_data/config.py b/clip_large_pretrain_4x256_sdv1_lr1e-4_wopostnorm_wneck/20230604_165929/vis_data/config.py new file mode 100644 index 0000000000000000000000000000000000000000..af1f90081c95d88c89d5cc97dca60c3ea695bcf9 --- /dev/null +++ b/clip_large_pretrain_4x256_sdv1_lr1e-4_wopostnorm_wneck/20230604_165929/vis_data/config.py @@ -0,0 +1,236 @@ +optim_wrapper = dict( + optimizer=dict( + type='AdamW', lr=0.0001, weight_decay=0.3, _scope_='mmpretrain'), + paramwise_cfg=dict( + custom_keys=dict({ + '.cls_token': dict(decay_mult=0.0), + '.pos_embed': dict(decay_mult=0.0) + })), + type='AmpOptimWrapper', + dtype='bfloat16', + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=False, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=4096) +model = dict( + type='ImageClassifier', + backbone=dict( + frozen_stages=24, + type='VisionTransformer', + arch='l', + img_size=224, + patch_size=14, + drop_rate=0.1, + pre_norm=True, + final_norm=False, + init_cfg=dict( + type='Pretrained', + checkpoint='ckpt/openclip-ViT-L-14.pth', + prefix='backbone')), + neck=dict( + type='CLIPProjection', + in_channels=1024, + out_channels=768, + init_cfg=dict( + type='Pretrained', + checkpoint='ckpt/openclip-ViT-L-14.pth', + prefix='backbone')), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=768, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + init_cfg=None), + init_cfg=dict( + type='TruncNormal', layer=['Conv2d', 'Linear'], std=0.02, bias=0.0), + train_cfg=None) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=128, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stablediffusionV1-5R2-dpmsolver-25-1m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/cc1m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=128, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=128, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/clip_large_pretrain_4x256_sdv1_lr1e-4_wopostnorm_wneck' diff --git a/clip_large_pretrain_4x256_sdv1_lr1e-4_wopostnorm_wneck/20230604_165929/vis_data/events.out.tfevents.1685869184.SH-IDC1-10-140-24-125.153228.0 b/clip_large_pretrain_4x256_sdv1_lr1e-4_wopostnorm_wneck/20230604_165929/vis_data/events.out.tfevents.1685869184.SH-IDC1-10-140-24-125.153228.0 new file mode 100644 index 0000000000000000000000000000000000000000..8af2f294ca3c7894110dbb6fef45bc88e8fae5a3 --- /dev/null +++ b/clip_large_pretrain_4x256_sdv1_lr1e-4_wopostnorm_wneck/20230604_165929/vis_data/events.out.tfevents.1685869184.SH-IDC1-10-140-24-125.153228.0 @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:b9f8b3aa0790664d441194d5d8a03763db6826c11c20955e72dd0942acdd236c +size 166893 diff --git a/clip_large_pretrain_4x256_sdv1_lr1e-4_wopostnorm_wneck/20230604_165929/vis_data/scalars.json b/clip_large_pretrain_4x256_sdv1_lr1e-4_wopostnorm_wneck/20230604_165929/vis_data/scalars.json new file mode 100644 index 0000000000000000000000000000000000000000..4cbcd42356c6323b232bd79f8e7055e107a46561 --- /dev/null +++ b/clip_large_pretrain_4x256_sdv1_lr1e-4_wopostnorm_wneck/20230604_165929/vis_data/scalars.json @@ -0,0 +1,50 @@ +{"lr": 9.999857418382631e-05, "data_time": 0.0020378589630126952, "loss": 0.5037091255187989, "time": 0.6273816823959351, "epoch": 1, "memory": 44139, "step": 100} +{"lr": 9.999423907359506e-05, "data_time": 0.001653289794921875, "loss": 0.4562071800231934, "time": 0.6289210319519043, "epoch": 1, "memory": 44139, "step": 200} +{"lr": 9.998699480412073e-05, "data_time": 0.0015316486358642578, "loss": 0.46545201539993286, "time": 0.6298252820968628, "epoch": 1, "memory": 44139, "step": 300} +{"lr": 9.997684184379085e-05, "data_time": 0.0014403104782104493, "loss": 0.43828535079956055, "time": 0.6311163187026978, "epoch": 1, "memory": 44139, "step": 400} +{"lr": 9.996378084905784e-05, "data_time": 0.001513051986694336, "loss": 0.40659589171409605, "time": 0.6297130823135376, "epoch": 1, "memory": 44139, "step": 500} +{"lr": 9.994781266439667e-05, "data_time": 0.001509380340576172, "loss": 0.4381151467561722, "time": 0.6301935195922852, "epoch": 1, "memory": 44139, "step": 600} +{"lr": 9.992893832225061e-05, "data_time": 0.0015932321548461914, "loss": 0.44335839748382566, "time": 0.6333444833755493, "epoch": 1, "memory": 44139, "step": 700} +{"lr": 9.990715904296405e-05, "data_time": 0.0018578767776489258, "loss": 0.45387116372585296, "time": 0.6295591831207276, "epoch": 1, "memory": 44139, "step": 800} +{"lr": 9.988247623470371e-05, "data_time": 0.0016877174377441407, "loss": 0.4134774267673492, "time": 0.6299310684204101, "epoch": 1, "memory": 44139, "step": 900} +{"lr": 9.985489149336749e-05, "data_time": 0.001697373390197754, "loss": 0.37754403352737426, "time": 0.6295555830001831, "epoch": 1, "memory": 44139, "step": 1000} +{"lr": 9.982440660248187e-05, "data_time": 0.0016101598739624023, "loss": 0.38221445083618166, "time": 0.6291843175888061, "epoch": 1, "memory": 44139, "step": 1100} +{"lr": 9.979102353308574e-05, "data_time": 0.0015021562576293945, "loss": 0.4044693887233734, "time": 0.6307098388671875, "epoch": 1, "memory": 44139, "step": 1200} +{"lr": 9.975474444360358e-05, "data_time": 0.0014406204223632812, "loss": 0.42279147505760195, "time": 0.6283769130706787, "epoch": 1, "memory": 44139, "step": 1300} +{"lr": 9.97155716797056e-05, "data_time": 0.0016569137573242188, "loss": 0.3893288761377335, "time": 0.6284325361251831, "epoch": 1, "memory": 44139, "step": 1400} +{"lr": 9.967350777415632e-05, "data_time": 0.0015644073486328126, "loss": 0.38546980917453766, "time": 0.6300941467285156, "epoch": 1, "memory": 44139, "step": 1500} +{"lr": 9.962855544665059e-05, "data_time": 0.0015160799026489257, "loss": 0.38515550792217257, "time": 0.6283077478408814, "epoch": 1, "memory": 44139, "step": 1600} +{"lr": 9.958071760363773e-05, "data_time": 0.001571059226989746, "loss": 0.3945955365896225, "time": 0.6284991264343261, "epoch": 1, "memory": 44139, "step": 1700} +{"lr": 9.952999733813382e-05, "data_time": 0.0015401124954223632, "loss": 0.39930541515350343, "time": 0.6282625913619995, "epoch": 1, "memory": 44139, "step": 1800} +{"lr": 9.947639792952148e-05, "data_time": 0.0014751195907592774, "loss": 0.3996291786432266, "time": 0.6278280735015869, "epoch": 1, "memory": 44139, "step": 1900} +{"lr": 9.941992284333813e-05, "data_time": 0.0017621040344238282, "loss": 0.3752891838550568, "time": 0.6273766994476319, "epoch": 1, "memory": 44139, "step": 2000} +{"lr": 9.93605757310516e-05, "data_time": 0.001543879508972168, "loss": 0.3720944285392761, "time": 0.6294212341308594, "epoch": 1, "memory": 44139, "step": 2100} +{"lr": 9.929836042982421e-05, "data_time": 0.001537013053894043, "loss": 0.36986070275306704, "time": 0.6286076545715332, "epoch": 1, "memory": 44139, "step": 2200} +{"lr": 9.923328096226467e-05, "data_time": 0.0015148401260375976, "loss": 0.37991597354412077, "time": 0.628557276725769, "epoch": 1, "memory": 44139, "step": 2300} +{"lr": 9.916534153616819e-05, "data_time": 0.0015467166900634765, "loss": 0.36074067652225494, "time": 0.6274428367614746, "epoch": 1, "memory": 44139, "step": 2400} +{"lr": 9.909454654424387e-05, "data_time": 0.001568436622619629, "loss": 0.39794217646121977, "time": 0.6283650636672974, "epoch": 1, "memory": 44139, "step": 2500} +{"lr": 9.902090056383109e-05, "data_time": 0.0014369726181030274, "loss": 0.3836180180311203, "time": 0.6287459135055542, "epoch": 1, "memory": 44139, "step": 2600} +{"lr": 9.894440835660365e-05, "data_time": 0.001600170135498047, "loss": 0.36383976936340334, "time": 0.6277165651321411, "epoch": 1, "memory": 44139, "step": 2700} +{"lr": 9.886507486826162e-05, "data_time": 0.0016936779022216797, "loss": 0.3389661222696304, "time": 0.6289095878601074, "epoch": 1, "memory": 44139, "step": 2800} +{"lr": 9.878290522821154e-05, "data_time": 0.0014937162399291993, "loss": 0.36450161337852477, "time": 0.627508282661438, "epoch": 1, "memory": 44139, "step": 2900} +{"lr": 9.86979047492352e-05, "data_time": 0.0013707637786865234, "loss": 0.33687353432178496, "time": 0.6274271965026855, "epoch": 1, "memory": 44139, "step": 3000} +{"lr": 9.861007892714578e-05, "data_time": 0.0015142440795898437, "loss": 0.36411143839359283, "time": 0.6272186994552612, "epoch": 1, "memory": 44139, "step": 3100} +{"lr": 9.851943344043239e-05, "data_time": 0.001987195014953613, "loss": 0.3568147510290146, "time": 0.6285130739212036, "epoch": 1, "memory": 44139, "step": 3200} +{"lr": 9.842597414989319e-05, "data_time": 0.0015331268310546874, "loss": 0.3257623165845871, "time": 0.6293205738067627, "epoch": 1, "memory": 44139, "step": 3300} +{"lr": 9.832970709825637e-05, "data_time": 0.0015496492385864257, "loss": 0.3229869931936264, "time": 0.6286638021469116, "epoch": 1, "memory": 44139, "step": 3400} +{"lr": 9.823063850978968e-05, "data_time": 0.001575303077697754, "loss": 0.3449849545955658, "time": 0.6282416582107544, "epoch": 1, "memory": 44139, "step": 3500} +{"lr": 9.81287747898975e-05, "data_time": 0.0016646862030029296, "loss": 0.3551407277584076, "time": 0.6286078691482544, "epoch": 1, "memory": 44139, "step": 3600} +{"lr": 9.802412252470715e-05, "data_time": 0.0016801118850708007, "loss": 0.3679080069065094, "time": 0.6287972688674927, "epoch": 1, "memory": 44139, "step": 3700} +{"lr": 9.791668848064258e-05, "data_time": 0.001544189453125, "loss": 0.3536813914775848, "time": 0.6289635181427002, "epoch": 1, "memory": 44139, "step": 3800} +{"lr": 9.780647960398751e-05, "data_time": 0.0014657020568847657, "loss": 0.38214276283979415, "time": 0.6274443864822388, "epoch": 1, "memory": 44139, "step": 3900} +{"accuracy/top1": 81.30197143554688, "data_time": 0.0441409663150185, "time": 1.351335931242558, "step": 1} +{"lr": 9.7685491196646e-05, "data_time": 0.0018814563751220702, "loss": 0.33556934297084806, "time": 0.6321202993392945, "epoch": 2, "memory": 44139, "step": 4007} +{"lr": 9.756956126085676e-05, "data_time": 0.001814866065979004, "loss": 0.3295762687921524, "time": 0.6314014196395874, "epoch": 2, "memory": 44138, "step": 4107} +{"lr": 9.745087893642482e-05, "data_time": 0.001662445068359375, "loss": 0.3412328392267227, "time": 0.6315424919128418, "epoch": 2, "memory": 44138, "step": 4207} +{"lr": 9.73294518969056e-05, "data_time": 0.0018392086029052734, "loss": 0.34878638088703157, "time": 0.6292943239212037, "epoch": 2, "memory": 44138, "step": 4307} +{"lr": 9.72052879933172e-05, "data_time": 0.0016124963760375977, "loss": 0.3149163812398911, "time": 0.6304901361465454, "epoch": 2, "memory": 44138, "step": 4407} +{"lr": 9.707839525363358e-05, "data_time": 0.0014074325561523437, "loss": 0.3519111305475235, "time": 0.6295089006423951, "epoch": 2, "memory": 44138, "step": 4507} +{"lr": 9.69487818822646e-05, "data_time": 0.0015254735946655274, "loss": 0.3351467907428741, "time": 0.6305028438568115, "epoch": 2, "memory": 44138, "step": 4607} +{"lr": 9.681645625952616e-05, "data_time": 0.0013847112655639648, "loss": 0.33887229561805726, "time": 0.6290833473205566, "epoch": 2, "memory": 44138, "step": 4707} +{"lr": 9.668142694109826e-05, "data_time": 0.0015581130981445312, "loss": 0.3430200695991516, "time": 0.6307183027267456, "epoch": 2, "memory": 44138, "step": 4807} +{"lr": 9.654370265747162e-05, "data_time": 0.0015548229217529296, "loss": 0.3347715586423874, "time": 0.6298733949661255, "epoch": 2, "memory": 44138, "step": 4907} diff --git a/clip_large_pretrain_4x256_sdv1_lr1e-4_wopostnorm_wneck/20230604_165929/vis_data/vis_image/0000000.png_1.png b/clip_large_pretrain_4x256_sdv1_lr1e-4_wopostnorm_wneck/20230604_165929/vis_data/vis_image/0000000.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..1367617503810e4cc7479ff1283174857704dd4b Binary files /dev/null and b/clip_large_pretrain_4x256_sdv1_lr1e-4_wopostnorm_wneck/20230604_165929/vis_data/vis_image/0000000.png_1.png differ diff --git a/clip_large_pretrain_4x256_sdv1_lr1e-4_wopostnorm_wneck/20230604_165929/vis_data/vis_image/000005174.jpg_1.png b/clip_large_pretrain_4x256_sdv1_lr1e-4_wopostnorm_wneck/20230604_165929/vis_data/vis_image/000005174.jpg_1.png new file mode 100644 index 0000000000000000000000000000000000000000..11db778bbc0dd73cbb7ef4fed0de37726b722d2c Binary files /dev/null and b/clip_large_pretrain_4x256_sdv1_lr1e-4_wopostnorm_wneck/20230604_165929/vis_data/vis_image/000005174.jpg_1.png differ diff --git a/clip_large_pretrain_4x256_sdv1_lr1e-4_wopostnorm_wneck/20230604_175638/20230604_175638.log b/clip_large_pretrain_4x256_sdv1_lr1e-4_wopostnorm_wneck/20230604_175638/20230604_175638.log new file mode 100644 index 0000000000000000000000000000000000000000..b28a5f75b4910ab9cb1d0868b9772744ab8a59e7 --- /dev/null +++ b/clip_large_pretrain_4x256_sdv1_lr1e-4_wopostnorm_wneck/20230604_175638/20230604_175638.log @@ -0,0 +1,1727 @@ +2023/06/04 17:56:43 - mmengine - INFO - +------------------------------------------------------------ +System environment: + sys.platform: linux + Python: 3.10.9 (main, Mar 8 2023, 10:47:38) [GCC 11.2.0] + CUDA available: True + numpy_random_seed: 1114330080 + GPU 0,1,2,3: NVIDIA A100-SXM4-80GB + CUDA_HOME: /mnt/petrelfs/share/cuda-11.6 + NVCC: Cuda compilation tools, release 11.6, V11.6.124 + GCC: gcc (GCC) 7.5.0 + PyTorch: 1.13.1 + PyTorch compiling details: PyTorch built with: + - GCC 9.3 + - C++ Version: 201402 + - Intel(R) oneAPI Math Kernel Library Version 2021.4-Product Build 20210904 for Intel(R) 64 architecture applications + - Intel(R) MKL-DNN v2.6.0 (Git Hash 52b5f107dd9cf10910aaa19cb47f3abf9b349815) + - OpenMP 201511 (a.k.a. OpenMP 4.5) + - LAPACK is enabled (usually provided by MKL) + - NNPACK is enabled + - CPU capability usage: AVX2 + - CUDA Runtime 11.6 + - NVCC architecture flags: -gencode;arch=compute_37,code=sm_37;-gencode;arch=compute_50,code=sm_50;-gencode;arch=compute_60,code=sm_60;-gencode;arch=compute_61,code=sm_61;-gencode;arch=compute_70,code=sm_70;-gencode;arch=compute_75,code=sm_75;-gencode;arch=compute_80,code=sm_80;-gencode;arch=compute_86,code=sm_86;-gencode;arch=compute_37,code=compute_37 + - CuDNN 8.3.2 (built against CUDA 11.5) + - Magma 2.6.1 + - Build settings: BLAS_INFO=mkl, BUILD_TYPE=Release, CUDA_VERSION=11.6, CUDNN_VERSION=8.3.2, CXX_COMPILER=/opt/rh/devtoolset-9/root/usr/bin/c++, CXX_FLAGS= -fabi-version=11 -Wno-deprecated -fvisibility-inlines-hidden -DUSE_PTHREADPOOL -fopenmp -DNDEBUG -DUSE_KINETO -DUSE_FBGEMM -DUSE_QNNPACK -DUSE_PYTORCH_QNNPACK -DUSE_XNNPACK -DSYMBOLICATE_MOBILE_DEBUG_HANDLE -DEDGE_PROFILER_USE_KINETO -O2 -fPIC -Wno-narrowing -Wall -Wextra -Werror=return-type -Werror=non-virtual-dtor -Wno-missing-field-initializers -Wno-type-limits -Wno-array-bounds -Wno-unknown-pragmas -Wunused-local-typedefs -Wno-unused-parameter -Wno-unused-function -Wno-unused-result -Wno-strict-overflow -Wno-strict-aliasing -Wno-error=deprecated-declarations -Wno-stringop-overflow -Wno-psabi -Wno-error=pedantic -Wno-error=redundant-decls -Wno-error=old-style-cast -fdiagnostics-color=always -faligned-new -Wno-unused-but-set-variable -Wno-maybe-uninitialized -fno-math-errno -fno-trapping-math -Werror=format -Werror=cast-function-type -Wno-stringop-overflow, LAPACK_INFO=mkl, PERF_WITH_AVX=1, PERF_WITH_AVX2=1, PERF_WITH_AVX512=1, TORCH_VERSION=1.13.1, USE_CUDA=ON, USE_CUDNN=ON, USE_EXCEPTION_PTR=1, USE_GFLAGS=OFF, USE_GLOG=OFF, USE_MKL=ON, USE_MKLDNN=ON, USE_MPI=OFF, USE_NCCL=ON, USE_NNPACK=ON, USE_OPENMP=ON, USE_ROCM=OFF, + + TorchVision: 0.14.1 + OpenCV: 4.7.0 + MMEngine: 0.7.3 + +Runtime environment: + cudnn_benchmark: True + mp_cfg: {'mp_start_method': 'fork', 'opencv_num_threads': 0} + dist_cfg: {'backend': 'nccl'} + seed: None + deterministic: False + Distributed launcher: slurm + Distributed training: True + GPU number: 4 +------------------------------------------------------------ + +2023/06/04 17:56:47 - mmengine - INFO - Config: +optim_wrapper = dict( + optimizer=dict( + type='AdamW', lr=0.0001, weight_decay=0.3, _scope_='mmpretrain'), + paramwise_cfg=dict( + custom_keys=dict({ + '.cls_token': dict(decay_mult=0.0), + '.pos_embed': dict(decay_mult=0.0) + })), + type='AmpOptimWrapper', + dtype='bfloat16', + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=False, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=4096) +model = dict( + type='ImageClassifier', + backbone=dict( + frozen_stages=24, + type='VisionTransformer', + arch='l', + img_size=224, + patch_size=14, + drop_rate=0.1, + pre_norm=True, + final_norm=False, + init_cfg=dict( + type='Pretrained', + checkpoint='ckpt/openclip-ViT-L-14.pth', + prefix='backbone')), + neck=dict( + type='CLIPProjection', + in_channels=1024, + out_channels=768, + init_cfg=dict( + type='Pretrained', + checkpoint='ckpt/openclip-ViT-L-14.pth', + prefix='backbone')), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=768, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + init_cfg=None), + init_cfg=dict( + type='TruncNormal', layer=['Conv2d', 'Linear'], std=0.02, bias=0.0), + train_cfg=None) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=128, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stablediffusionV1-5R2-dpmsolver-25-1m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/cc1m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=128, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=128, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/clip_large_pretrain_4x256_sdv1_lr1e-4_wopostnorm_wneck' + +2023/06/04 17:57:01 - mmengine - INFO - Hooks will be executed in the following order: +before_run: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_load_checkpoint: +(ABOVE_NORMAL) EMAHook + -------------------- +before_train: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_train_epoch: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(NORMAL ) DistSamplerSeedHook + -------------------- +before_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook + -------------------- +after_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +after_train_epoch: +(NORMAL ) IterTimerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_val_epoch: +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook + -------------------- +before_val_iter: +(NORMAL ) IterTimerHook + -------------------- +after_val_iter: +(NORMAL ) IterTimerHook +(NORMAL ) VisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_val_epoch: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_save_checkpoint: +(ABOVE_NORMAL) EMAHook + -------------------- +after_train: +(VERY_LOW ) CheckpointHook + -------------------- +before_test_epoch: +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook + -------------------- +before_test_iter: +(NORMAL ) IterTimerHook + -------------------- +after_test_iter: +(NORMAL ) IterTimerHook +(NORMAL ) VisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_test_epoch: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_run: +(BELOW_NORMAL) LoggerHook + -------------------- +2023/06/04 17:57:21 - mmengine - INFO - load backbone in model from: ckpt/openclip-ViT-L-14.pth +2023/06/04 17:57:22 - mmengine - WARNING - The model and loaded state dict do not match exactly + +unexpected key in source state_dict: ln1.weight, ln1.bias + +2023/06/04 17:57:22 - mmengine - INFO - load backbone in model from: ckpt/openclip-ViT-L-14.pth +2023/06/04 17:57:24 - mmengine - WARNING - The model and loaded state dict do not match exactly + +unexpected key in source state_dict: cls_token, pos_embed, patch_embed.projection.weight, pre_norm.weight, pre_norm.bias, layers.0.ln1.weight, layers.0.ln1.bias, layers.0.attn.qkv.weight, layers.0.attn.qkv.bias, layers.0.attn.proj.weight, layers.0.attn.proj.bias, layers.0.ln2.weight, layers.0.ln2.bias, layers.0.ffn.layers.0.0.weight, layers.0.ffn.layers.0.0.bias, layers.0.ffn.layers.1.weight, layers.0.ffn.layers.1.bias, layers.1.ln1.weight, layers.1.ln1.bias, layers.1.attn.qkv.weight, layers.1.attn.qkv.bias, layers.1.attn.proj.weight, layers.1.attn.proj.bias, layers.1.ln2.weight, layers.1.ln2.bias, layers.1.ffn.layers.0.0.weight, layers.1.ffn.layers.0.0.bias, layers.1.ffn.layers.1.weight, layers.1.ffn.layers.1.bias, layers.2.ln1.weight, layers.2.ln1.bias, layers.2.attn.qkv.weight, layers.2.attn.qkv.bias, layers.2.attn.proj.weight, layers.2.attn.proj.bias, layers.2.ln2.weight, layers.2.ln2.bias, layers.2.ffn.layers.0.0.weight, layers.2.ffn.layers.0.0.bias, layers.2.ffn.layers.1.weight, layers.2.ffn.layers.1.bias, layers.3.ln1.weight, layers.3.ln1.bias, layers.3.attn.qkv.weight, layers.3.attn.qkv.bias, layers.3.attn.proj.weight, layers.3.attn.proj.bias, layers.3.ln2.weight, layers.3.ln2.bias, layers.3.ffn.layers.0.0.weight, layers.3.ffn.layers.0.0.bias, layers.3.ffn.layers.1.weight, layers.3.ffn.layers.1.bias, layers.4.ln1.weight, layers.4.ln1.bias, layers.4.attn.qkv.weight, layers.4.attn.qkv.bias, layers.4.attn.proj.weight, layers.4.attn.proj.bias, layers.4.ln2.weight, layers.4.ln2.bias, layers.4.ffn.layers.0.0.weight, layers.4.ffn.layers.0.0.bias, layers.4.ffn.layers.1.weight, layers.4.ffn.layers.1.bias, layers.5.ln1.weight, layers.5.ln1.bias, layers.5.attn.qkv.weight, layers.5.attn.qkv.bias, layers.5.attn.proj.weight, layers.5.attn.proj.bias, layers.5.ln2.weight, layers.5.ln2.bias, layers.5.ffn.layers.0.0.weight, layers.5.ffn.layers.0.0.bias, layers.5.ffn.layers.1.weight, layers.5.ffn.layers.1.bias, layers.6.ln1.weight, layers.6.ln1.bias, layers.6.attn.qkv.weight, layers.6.attn.qkv.bias, layers.6.attn.proj.weight, layers.6.attn.proj.bias, layers.6.ln2.weight, layers.6.ln2.bias, layers.6.ffn.layers.0.0.weight, layers.6.ffn.layers.0.0.bias, layers.6.ffn.layers.1.weight, layers.6.ffn.layers.1.bias, layers.7.ln1.weight, layers.7.ln1.bias, layers.7.attn.qkv.weight, layers.7.attn.qkv.bias, layers.7.attn.proj.weight, layers.7.attn.proj.bias, layers.7.ln2.weight, layers.7.ln2.bias, layers.7.ffn.layers.0.0.weight, layers.7.ffn.layers.0.0.bias, layers.7.ffn.layers.1.weight, layers.7.ffn.layers.1.bias, layers.8.ln1.weight, layers.8.ln1.bias, layers.8.attn.qkv.weight, layers.8.attn.qkv.bias, layers.8.attn.proj.weight, layers.8.attn.proj.bias, layers.8.ln2.weight, layers.8.ln2.bias, layers.8.ffn.layers.0.0.weight, layers.8.ffn.layers.0.0.bias, layers.8.ffn.layers.1.weight, layers.8.ffn.layers.1.bias, layers.9.ln1.weight, layers.9.ln1.bias, layers.9.attn.qkv.weight, layers.9.attn.qkv.bias, layers.9.attn.proj.weight, layers.9.attn.proj.bias, layers.9.ln2.weight, layers.9.ln2.bias, layers.9.ffn.layers.0.0.weight, layers.9.ffn.layers.0.0.bias, layers.9.ffn.layers.1.weight, layers.9.ffn.layers.1.bias, layers.10.ln1.weight, layers.10.ln1.bias, layers.10.attn.qkv.weight, layers.10.attn.qkv.bias, layers.10.attn.proj.weight, layers.10.attn.proj.bias, layers.10.ln2.weight, layers.10.ln2.bias, layers.10.ffn.layers.0.0.weight, layers.10.ffn.layers.0.0.bias, layers.10.ffn.layers.1.weight, layers.10.ffn.layers.1.bias, layers.11.ln1.weight, layers.11.ln1.bias, layers.11.attn.qkv.weight, layers.11.attn.qkv.bias, layers.11.attn.proj.weight, layers.11.attn.proj.bias, layers.11.ln2.weight, layers.11.ln2.bias, layers.11.ffn.layers.0.0.weight, layers.11.ffn.layers.0.0.bias, layers.11.ffn.layers.1.weight, layers.11.ffn.layers.1.bias, layers.12.ln1.weight, layers.12.ln1.bias, layers.12.attn.qkv.weight, layers.12.attn.qkv.bias, layers.12.attn.proj.weight, layers.12.attn.proj.bias, layers.12.ln2.weight, layers.12.ln2.bias, layers.12.ffn.layers.0.0.weight, layers.12.ffn.layers.0.0.bias, layers.12.ffn.layers.1.weight, layers.12.ffn.layers.1.bias, layers.13.ln1.weight, layers.13.ln1.bias, layers.13.attn.qkv.weight, layers.13.attn.qkv.bias, layers.13.attn.proj.weight, layers.13.attn.proj.bias, layers.13.ln2.weight, layers.13.ln2.bias, layers.13.ffn.layers.0.0.weight, layers.13.ffn.layers.0.0.bias, layers.13.ffn.layers.1.weight, layers.13.ffn.layers.1.bias, layers.14.ln1.weight, layers.14.ln1.bias, layers.14.attn.qkv.weight, layers.14.attn.qkv.bias, layers.14.attn.proj.weight, layers.14.attn.proj.bias, layers.14.ln2.weight, layers.14.ln2.bias, layers.14.ffn.layers.0.0.weight, layers.14.ffn.layers.0.0.bias, layers.14.ffn.layers.1.weight, layers.14.ffn.layers.1.bias, layers.15.ln1.weight, layers.15.ln1.bias, layers.15.attn.qkv.weight, layers.15.attn.qkv.bias, layers.15.attn.proj.weight, layers.15.attn.proj.bias, layers.15.ln2.weight, layers.15.ln2.bias, layers.15.ffn.layers.0.0.weight, layers.15.ffn.layers.0.0.bias, layers.15.ffn.layers.1.weight, layers.15.ffn.layers.1.bias, layers.16.ln1.weight, layers.16.ln1.bias, layers.16.attn.qkv.weight, layers.16.attn.qkv.bias, layers.16.attn.proj.weight, layers.16.attn.proj.bias, layers.16.ln2.weight, layers.16.ln2.bias, layers.16.ffn.layers.0.0.weight, layers.16.ffn.layers.0.0.bias, layers.16.ffn.layers.1.weight, layers.16.ffn.layers.1.bias, layers.17.ln1.weight, layers.17.ln1.bias, layers.17.attn.qkv.weight, layers.17.attn.qkv.bias, layers.17.attn.proj.weight, layers.17.attn.proj.bias, layers.17.ln2.weight, layers.17.ln2.bias, layers.17.ffn.layers.0.0.weight, layers.17.ffn.layers.0.0.bias, layers.17.ffn.layers.1.weight, layers.17.ffn.layers.1.bias, layers.18.ln1.weight, layers.18.ln1.bias, layers.18.attn.qkv.weight, layers.18.attn.qkv.bias, layers.18.attn.proj.weight, layers.18.attn.proj.bias, layers.18.ln2.weight, layers.18.ln2.bias, layers.18.ffn.layers.0.0.weight, layers.18.ffn.layers.0.0.bias, layers.18.ffn.layers.1.weight, layers.18.ffn.layers.1.bias, layers.19.ln1.weight, layers.19.ln1.bias, layers.19.attn.qkv.weight, layers.19.attn.qkv.bias, layers.19.attn.proj.weight, layers.19.attn.proj.bias, layers.19.ln2.weight, layers.19.ln2.bias, layers.19.ffn.layers.0.0.weight, layers.19.ffn.layers.0.0.bias, layers.19.ffn.layers.1.weight, layers.19.ffn.layers.1.bias, layers.20.ln1.weight, layers.20.ln1.bias, layers.20.attn.qkv.weight, layers.20.attn.qkv.bias, layers.20.attn.proj.weight, layers.20.attn.proj.bias, layers.20.ln2.weight, layers.20.ln2.bias, layers.20.ffn.layers.0.0.weight, layers.20.ffn.layers.0.0.bias, layers.20.ffn.layers.1.weight, layers.20.ffn.layers.1.bias, layers.21.ln1.weight, layers.21.ln1.bias, layers.21.attn.qkv.weight, layers.21.attn.qkv.bias, layers.21.attn.proj.weight, layers.21.attn.proj.bias, layers.21.ln2.weight, layers.21.ln2.bias, layers.21.ffn.layers.0.0.weight, layers.21.ffn.layers.0.0.bias, layers.21.ffn.layers.1.weight, layers.21.ffn.layers.1.bias, layers.22.ln1.weight, layers.22.ln1.bias, layers.22.attn.qkv.weight, layers.22.attn.qkv.bias, layers.22.attn.proj.weight, layers.22.attn.proj.bias, layers.22.ln2.weight, layers.22.ln2.bias, layers.22.ffn.layers.0.0.weight, layers.22.ffn.layers.0.0.bias, layers.22.ffn.layers.1.weight, layers.22.ffn.layers.1.bias, layers.23.ln1.weight, layers.23.ln1.bias, layers.23.attn.qkv.weight, layers.23.attn.qkv.bias, layers.23.attn.proj.weight, layers.23.attn.proj.bias, layers.23.ln2.weight, layers.23.ln2.bias, layers.23.ffn.layers.0.0.weight, layers.23.ffn.layers.0.0.bias, layers.23.ffn.layers.1.weight, layers.23.ffn.layers.1.bias, ln1.weight, ln1.bias + +missing keys in source state_dict: proj + +Name of parameter - Initialization information + +backbone.cls_token - torch.Size([1, 1, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.pos_embed - torch.Size([1, 257, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.patch_embed.projection.weight - torch.Size([1024, 3, 14, 14]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.0.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.0.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.0.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.0.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.0.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.0.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.0.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.0.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.0.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.0.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.0.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.0.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.1.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.1.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.1.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.1.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.1.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.1.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.1.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.1.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.1.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.1.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.1.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.1.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.2.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.2.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.2.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.2.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.2.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.2.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.2.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.2.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.2.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.2.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.2.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.2.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.3.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.3.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.3.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.3.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.3.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.3.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.3.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.3.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.3.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.3.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.3.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.3.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.4.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.4.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.4.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.4.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.4.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.4.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.4.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.4.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.4.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.4.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.4.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.4.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.5.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.5.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.5.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.5.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.5.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.5.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.5.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.5.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.5.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.5.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.5.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.5.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.6.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.6.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.6.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.6.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.6.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.6.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.6.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.6.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.6.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.6.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.6.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.6.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.7.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.7.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.7.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.7.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.7.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.7.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.7.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.7.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.7.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.7.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.7.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.7.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.8.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.8.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.8.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.8.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.8.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.8.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.8.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.8.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.8.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.8.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.8.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.8.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.9.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.9.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.9.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.9.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.9.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.9.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.9.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.9.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.9.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.9.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.9.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.9.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.10.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.10.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.10.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.10.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.10.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.10.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.10.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.10.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.10.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.10.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.10.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.10.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.11.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.11.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.11.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.11.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.11.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.11.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.11.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.11.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.11.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.11.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.11.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.11.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.12.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.12.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.12.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.12.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.12.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.12.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.12.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.12.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.12.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.12.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.12.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.12.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.13.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.13.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.13.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.13.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.13.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.13.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.13.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.13.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.13.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.13.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.13.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.13.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.14.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.14.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.14.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.14.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.14.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.14.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.14.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.14.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.14.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.14.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.14.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.14.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.15.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.15.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.15.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.15.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.15.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.15.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.15.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.15.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.15.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.15.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.15.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.15.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.16.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.16.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.16.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.16.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.16.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.16.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.16.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.16.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.16.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.16.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.16.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.16.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.17.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.17.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.17.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.17.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.17.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.17.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.17.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.17.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.17.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.17.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.17.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.17.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.18.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.18.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.18.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.18.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.18.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.18.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.18.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.18.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.18.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.18.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.18.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.18.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.19.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.19.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.19.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.19.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.19.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.19.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.19.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.19.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.19.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.19.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.19.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.19.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.20.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.20.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.20.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.20.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.20.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.20.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.20.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.20.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.20.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.20.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.20.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.20.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.21.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.21.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.21.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.21.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.21.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.21.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.21.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.21.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.21.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.21.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.21.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.21.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.22.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.22.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.22.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.22.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.22.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.22.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.22.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.22.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.22.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.22.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.22.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.22.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.23.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.23.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.23.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.23.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.23.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.23.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.23.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.23.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.23.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.23.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.23.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.23.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.pre_norm.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.pre_norm.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +neck.proj - torch.Size([1024, 768]): +The value is the same before and after calling `init_weights` of ImageClassifier + +head.fc.weight - torch.Size([2, 768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +head.fc.bias - torch.Size([2]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 +2023/06/04 17:57:24 - mmengine - WARNING - "FileClient" will be deprecated in future. Please use io functions in https://mmengine.readthedocs.io/en/latest/api/fileio.html#file-io +2023/06/04 17:57:24 - mmengine - WARNING - "HardDiskBackend" is the alias of "LocalBackend" and the former will be deprecated in future. +2023/06/04 17:57:24 - mmengine - INFO - Checkpoints will be saved to /mnt/petrelfs/luzeyu/workspace/fakebench/mmpretrain/workdir/clip_large_pretrain_4x256_sdv1_lr1e-4_wopostnorm_wneck. +2023/06/04 17:58:35 - mmengine - INFO - Epoch(train) [1][ 100/3907] lr: 9.9999e-05 eta: 7:41:43 time: 0.6304 data_time: 0.0027 memory: 44139 loss: 0.5272 +2023/06/04 17:59:38 - mmengine - INFO - Epoch(train) [1][ 200/3907] lr: 9.9994e-05 eta: 7:14:52 time: 0.6317 data_time: 0.0020 memory: 44139 loss: 0.4701 +2023/06/04 18:00:41 - mmengine - INFO - Epoch(train) [1][ 300/3907] lr: 9.9987e-05 eta: 7:05:24 time: 0.6329 data_time: 0.0019 memory: 44139 loss: 0.4475 +2023/06/04 18:01:45 - mmengine - INFO - Epoch(train) [1][ 400/3907] lr: 9.9977e-05 eta: 7:00:23 time: 0.6322 data_time: 0.0018 memory: 44139 loss: 0.4509 +2023/06/04 18:02:48 - mmengine - INFO - Epoch(train) [1][ 500/3907] lr: 9.9964e-05 eta: 6:56:44 time: 0.6320 data_time: 0.0018 memory: 44139 loss: 0.4191 +2023/06/04 18:03:51 - mmengine - INFO - Epoch(train) [1][ 600/3907] lr: 9.9948e-05 eta: 6:54:01 time: 0.6332 data_time: 0.0020 memory: 44139 loss: 0.4148 +2023/06/04 18:04:55 - mmengine - INFO - Epoch(train) [1][ 700/3907] lr: 9.9929e-05 eta: 6:51:50 time: 0.6329 data_time: 0.0019 memory: 44139 loss: 0.3993 +2023/06/04 18:05:58 - mmengine - INFO - Epoch(train) [1][ 800/3907] lr: 9.9907e-05 eta: 6:49:50 time: 0.6325 data_time: 0.0021 memory: 44139 loss: 0.4161 +2023/06/04 18:07:01 - mmengine - INFO - Epoch(train) [1][ 900/3907] lr: 9.9882e-05 eta: 6:48:04 time: 0.6325 data_time: 0.0018 memory: 44139 loss: 0.3960 +2023/06/04 18:08:04 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv1_lr1e-4_20230604_175638 +2023/06/04 18:08:04 - mmengine - INFO - Epoch(train) [1][1000/3907] lr: 9.9855e-05 eta: 6:46:25 time: 0.6310 data_time: 0.0021 memory: 44139 loss: 0.4181 +2023/06/04 18:09:08 - mmengine - INFO - Epoch(train) [1][1100/3907] lr: 9.9824e-05 eta: 6:44:54 time: 0.6321 data_time: 0.0016 memory: 44139 loss: 0.4181 +2023/06/04 18:10:11 - mmengine - INFO - Epoch(train) [1][1200/3907] lr: 9.9791e-05 eta: 6:43:31 time: 0.6330 data_time: 0.0017 memory: 44139 loss: 0.4143 +2023/06/04 18:11:14 - mmengine - INFO - Epoch(train) [1][1300/3907] lr: 9.9755e-05 eta: 6:42:08 time: 0.6327 data_time: 0.0018 memory: 44139 loss: 0.4011 +2023/06/04 18:12:18 - mmengine - INFO - Epoch(train) [1][1400/3907] lr: 9.9716e-05 eta: 6:40:47 time: 0.6327 data_time: 0.0018 memory: 44139 loss: 0.4026 +2023/06/04 18:13:21 - mmengine - INFO - Epoch(train) [1][1500/3907] lr: 9.9674e-05 eta: 6:39:29 time: 0.6327 data_time: 0.0016 memory: 44139 loss: 0.3908 +2023/06/04 18:14:24 - mmengine - INFO - Epoch(train) [1][1600/3907] lr: 9.9629e-05 eta: 6:38:12 time: 0.6325 data_time: 0.0017 memory: 44139 loss: 0.3846 +2023/06/04 18:15:27 - mmengine - INFO - Epoch(train) [1][1700/3907] lr: 9.9581e-05 eta: 6:36:58 time: 0.6327 data_time: 0.0017 memory: 44139 loss: 0.4027 +2023/06/04 18:16:31 - mmengine - INFO - Epoch(train) [1][1800/3907] lr: 9.9530e-05 eta: 6:35:46 time: 0.6332 data_time: 0.0018 memory: 44139 loss: 0.3739 +2023/06/04 18:17:34 - mmengine - INFO - Epoch(train) [1][1900/3907] lr: 9.9476e-05 eta: 6:34:34 time: 0.6328 data_time: 0.0018 memory: 44139 loss: 0.3638 +2023/06/04 18:18:37 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv1_lr1e-4_20230604_175638 +2023/06/04 18:18:37 - mmengine - INFO - Epoch(train) [1][2000/3907] lr: 9.9420e-05 eta: 6:33:23 time: 0.6327 data_time: 0.0018 memory: 44139 loss: 0.3912 +2023/06/04 18:19:41 - mmengine - INFO - Epoch(train) [1][2100/3907] lr: 9.9361e-05 eta: 6:32:14 time: 0.6328 data_time: 0.0017 memory: 44139 loss: 0.4008 +2023/06/04 18:20:44 - mmengine - INFO - Epoch(train) [1][2200/3907] lr: 9.9298e-05 eta: 6:31:04 time: 0.6327 data_time: 0.0016 memory: 44139 loss: 0.3407 +2023/06/04 18:21:47 - mmengine - INFO - Epoch(train) [1][2300/3907] lr: 9.9233e-05 eta: 6:29:54 time: 0.6326 data_time: 0.0016 memory: 44139 loss: 0.3782 +2023/06/04 18:22:51 - mmengine - INFO - Epoch(train) [1][2400/3907] lr: 9.9165e-05 eta: 6:28:45 time: 0.6332 data_time: 0.0016 memory: 44139 loss: 0.3638 +2023/06/04 18:23:54 - mmengine - INFO - Epoch(train) [1][2500/3907] lr: 9.9095e-05 eta: 6:27:38 time: 0.6329 data_time: 0.0022 memory: 44139 loss: 0.3505 +2023/06/04 18:24:57 - mmengine - INFO - Epoch(train) [1][2600/3907] lr: 9.9021e-05 eta: 6:26:30 time: 0.6325 data_time: 0.0019 memory: 44139 loss: 0.3692 +2023/06/04 18:26:01 - mmengine - INFO - Epoch(train) [1][2700/3907] lr: 9.8944e-05 eta: 6:25:22 time: 0.6329 data_time: 0.0016 memory: 44139 loss: 0.3849 +2023/06/04 18:27:04 - mmengine - INFO - Epoch(train) [1][2800/3907] lr: 9.8865e-05 eta: 6:24:16 time: 0.6332 data_time: 0.0019 memory: 44139 loss: 0.3473 +2023/06/04 18:28:07 - mmengine - INFO - Epoch(train) [1][2900/3907] lr: 9.8783e-05 eta: 6:23:11 time: 0.6331 data_time: 0.0019 memory: 44139 loss: 0.3706 +2023/06/04 18:29:11 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv1_lr1e-4_20230604_175638 +2023/06/04 18:29:11 - mmengine - INFO - Epoch(train) [1][3000/3907] lr: 9.8698e-05 eta: 6:22:04 time: 0.6325 data_time: 0.0018 memory: 44139 loss: 0.3459 +2023/06/04 18:30:14 - mmengine - INFO - Epoch(train) [1][3100/3907] lr: 9.8610e-05 eta: 6:20:57 time: 0.6328 data_time: 0.0017 memory: 44139 loss: 0.3623 +2023/06/04 18:31:17 - mmengine - INFO - Epoch(train) [1][3200/3907] lr: 9.8519e-05 eta: 6:19:50 time: 0.6327 data_time: 0.0023 memory: 44139 loss: 0.3564 +2023/06/04 18:32:20 - mmengine - INFO - Epoch(train) [1][3300/3907] lr: 9.8426e-05 eta: 6:18:43 time: 0.6363 data_time: 0.0022 memory: 44139 loss: 0.3421 +2023/06/04 18:33:24 - mmengine - INFO - Epoch(train) [1][3400/3907] lr: 9.8330e-05 eta: 6:17:36 time: 0.6324 data_time: 0.0023 memory: 44139 loss: 0.3659 +2023/06/04 18:34:27 - mmengine - INFO - Epoch(train) [1][3500/3907] lr: 9.8231e-05 eta: 6:16:29 time: 0.6318 data_time: 0.0016 memory: 44139 loss: 0.3505 +2023/06/04 18:35:30 - mmengine - INFO - Epoch(train) [1][3600/3907] lr: 9.8129e-05 eta: 6:15:22 time: 0.6315 data_time: 0.0019 memory: 44139 loss: 0.3642 +2023/06/04 18:36:33 - mmengine - INFO - Epoch(train) [1][3700/3907] lr: 9.8024e-05 eta: 6:14:15 time: 0.6308 data_time: 0.0019 memory: 44139 loss: 0.3258 +2023/06/04 18:37:36 - mmengine - INFO - Epoch(train) [1][3800/3907] lr: 9.7917e-05 eta: 6:13:09 time: 0.6322 data_time: 0.0017 memory: 44139 loss: 0.3360 +2023/06/04 18:38:39 - mmengine - INFO - Epoch(train) [1][3900/3907] lr: 9.7806e-05 eta: 6:12:03 time: 0.6308 data_time: 0.0012 memory: 44139 loss: 0.3511 +2023/06/04 18:38:43 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv1_lr1e-4_20230604_175638 +2023/06/04 18:38:43 - mmengine - INFO - Saving checkpoint at 1 epochs +2023/06/04 18:40:20 - mmengine - INFO - Epoch(val) [1][57/57] accuracy/top1: 82.2310 single-label/precision_classwise: [79.2352066040039, 87.52769470214844] single-label/recall_classwise: [91.82501220703125, 70.44961547851562] single-label/f1-score_classwise: [85.06681823730469, 78.06554412841797] data_time: 0.0439 time: 1.3382 +2023/06/04 18:41:23 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv1_lr1e-4_20230604_175638 +2023/06/04 18:41:27 - mmengine - INFO - Epoch(train) [2][ 100/3907] lr: 9.7685e-05 eta: 6:11:27 time: 0.6322 data_time: 0.0018 memory: 44139 loss: 0.3512 +2023/06/04 18:42:30 - mmengine - INFO - Epoch(train) [2][ 200/3907] lr: 9.7570e-05 eta: 6:10:20 time: 0.6322 data_time: 0.0016 memory: 44138 loss: 0.3367 +2023/06/04 18:43:34 - mmengine - INFO - Epoch(train) [2][ 300/3907] lr: 9.7451e-05 eta: 6:09:14 time: 0.6321 data_time: 0.0016 memory: 44138 loss: 0.3411 +2023/06/04 18:44:37 - mmengine - INFO - Epoch(train) [2][ 400/3907] lr: 9.7329e-05 eta: 6:08:09 time: 0.6331 data_time: 0.0019 memory: 44138 loss: 0.3924 +2023/06/04 18:45:40 - mmengine - INFO - Epoch(train) [2][ 500/3907] lr: 9.7205e-05 eta: 6:07:04 time: 0.6322 data_time: 0.0014 memory: 44138 loss: 0.3141 +2023/06/04 18:46:44 - mmengine - INFO - Epoch(train) [2][ 600/3907] lr: 9.7078e-05 eta: 6:05:58 time: 0.6328 data_time: 0.0016 memory: 44138 loss: 0.3529 +2023/06/04 18:47:47 - mmengine - INFO - Epoch(train) [2][ 700/3907] lr: 9.6949e-05 eta: 6:04:54 time: 0.6326 data_time: 0.0015 memory: 44138 loss: 0.3299 +2023/06/04 18:48:50 - mmengine - INFO - Epoch(train) [2][ 800/3907] lr: 9.6816e-05 eta: 6:03:48 time: 0.6329 data_time: 0.0016 memory: 44138 loss: 0.3419 +2023/06/04 18:49:54 - mmengine - INFO - Epoch(train) [2][ 900/3907] lr: 9.6681e-05 eta: 6:02:43 time: 0.6331 data_time: 0.0015 memory: 44138 loss: 0.3642 +2023/06/04 18:50:57 - mmengine - INFO - Epoch(train) [2][1000/3907] lr: 9.6544e-05 eta: 6:01:38 time: 0.6327 data_time: 0.0019 memory: 44138 loss: 0.3296 +2023/06/04 18:51:56 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv1_lr1e-4_20230604_175638 +2023/06/04 18:52:00 - mmengine - INFO - Epoch(train) [2][1100/3907] lr: 9.6403e-05 eta: 6:00:33 time: 0.6320 data_time: 0.0019 memory: 44138 loss: 0.3581 +2023/06/04 18:53:03 - mmengine - INFO - Epoch(train) [2][1200/3907] lr: 9.6260e-05 eta: 5:59:28 time: 0.6329 data_time: 0.0017 memory: 44138 loss: 0.3302 +2023/06/04 18:54:07 - mmengine - INFO - Epoch(train) [2][1300/3907] lr: 9.6114e-05 eta: 5:58:24 time: 0.6422 data_time: 0.0019 memory: 44138 loss: 0.3023 +2023/06/04 18:55:10 - mmengine - INFO - Epoch(train) [2][1400/3907] lr: 9.5966e-05 eta: 5:57:19 time: 0.6328 data_time: 0.0017 memory: 44138 loss: 0.3157 +2023/06/04 18:56:13 - mmengine - INFO - Epoch(train) [2][1500/3907] lr: 9.5815e-05 eta: 5:56:14 time: 0.6319 data_time: 0.0016 memory: 44138 loss: 0.3265 +2023/06/04 18:57:17 - mmengine - INFO - Epoch(train) [2][1600/3907] lr: 9.5661e-05 eta: 5:55:09 time: 0.6324 data_time: 0.0016 memory: 44138 loss: 0.3227 +2023/06/04 18:58:20 - mmengine - INFO - Epoch(train) [2][1700/3907] lr: 9.5505e-05 eta: 5:54:04 time: 0.6321 data_time: 0.0018 memory: 44138 loss: 0.3386 +2023/06/04 18:59:23 - mmengine - INFO - Epoch(train) [2][1800/3907] lr: 9.5346e-05 eta: 5:52:59 time: 0.6321 data_time: 0.0014 memory: 44138 loss: 0.3330 +2023/06/04 19:00:27 - mmengine - INFO - Epoch(train) [2][1900/3907] lr: 9.5184e-05 eta: 5:51:55 time: 0.6323 data_time: 0.0019 memory: 44138 loss: 0.2948 +2023/06/04 19:01:30 - mmengine - INFO - Epoch(train) [2][2000/3907] lr: 9.5020e-05 eta: 5:50:50 time: 0.6327 data_time: 0.0017 memory: 44138 loss: 0.3407 +2023/06/04 19:02:29 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv1_lr1e-4_20230604_175638 +2023/06/04 19:02:33 - mmengine - INFO - Epoch(train) [2][2100/3907] lr: 9.4854e-05 eta: 5:49:46 time: 0.6327 data_time: 0.0019 memory: 44138 loss: 0.3081 +2023/06/04 19:03:36 - mmengine - INFO - Epoch(train) [2][2200/3907] lr: 9.4684e-05 eta: 5:48:41 time: 0.6335 data_time: 0.0015 memory: 44138 loss: 0.3409 +2023/06/04 19:04:40 - mmengine - INFO - Epoch(train) [2][2300/3907] lr: 9.4512e-05 eta: 5:47:38 time: 0.6462 data_time: 0.0165 memory: 44138 loss: 0.3355 +2023/06/04 19:05:43 - mmengine - INFO - Epoch(train) [2][2400/3907] lr: 9.4338e-05 eta: 5:46:34 time: 0.6456 data_time: 0.0015 memory: 44138 loss: 0.3299 +2023/06/04 19:06:47 - mmengine - INFO - Epoch(train) [2][2500/3907] lr: 9.4161e-05 eta: 5:45:29 time: 0.6327 data_time: 0.0016 memory: 44138 loss: 0.3338 +2023/06/04 19:07:50 - mmengine - INFO - Epoch(train) [2][2600/3907] lr: 9.3981e-05 eta: 5:44:26 time: 0.6320 data_time: 0.0018 memory: 44138 loss: 0.3528 +2023/06/04 19:08:53 - mmengine - INFO - Epoch(train) [2][2700/3907] lr: 9.3799e-05 eta: 5:43:21 time: 0.6313 data_time: 0.0017 memory: 44138 loss: 0.3359 +2023/06/04 19:09:56 - mmengine - INFO - Epoch(train) [2][2800/3907] lr: 9.3615e-05 eta: 5:42:16 time: 0.6313 data_time: 0.0016 memory: 44138 loss: 0.3257 +2023/06/04 19:10:59 - mmengine - INFO - Epoch(train) [2][2900/3907] lr: 9.3428e-05 eta: 5:41:11 time: 0.6311 data_time: 0.0018 memory: 44138 loss: 0.3561 +2023/06/04 19:12:03 - mmengine - INFO - Epoch(train) [2][3000/3907] lr: 9.3238e-05 eta: 5:40:06 time: 0.6313 data_time: 0.0015 memory: 44138 loss: 0.3406 +2023/06/04 19:13:01 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv1_lr1e-4_20230604_175638 +2023/06/04 19:13:06 - mmengine - INFO - Epoch(train) [2][3100/3907] lr: 9.3046e-05 eta: 5:39:01 time: 0.6314 data_time: 0.0020 memory: 44138 loss: 0.3190 +2023/06/04 19:14:09 - mmengine - INFO - Epoch(train) [2][3200/3907] lr: 9.2852e-05 eta: 5:37:57 time: 0.6324 data_time: 0.0018 memory: 44138 loss: 0.3082 +2023/06/04 19:15:12 - mmengine - INFO - Epoch(train) [2][3300/3907] lr: 9.2655e-05 eta: 5:36:52 time: 0.6321 data_time: 0.0021 memory: 44138 loss: 0.3210 +2023/06/04 19:16:15 - mmengine - INFO - Epoch(train) [2][3400/3907] lr: 9.2456e-05 eta: 5:35:48 time: 0.6371 data_time: 0.0014 memory: 44138 loss: 0.3304 +2023/06/04 19:17:19 - mmengine - INFO - Epoch(train) [2][3500/3907] lr: 9.2254e-05 eta: 5:34:43 time: 0.6318 data_time: 0.0014 memory: 44138 loss: 0.3070 +2023/06/04 19:18:22 - mmengine - INFO - Epoch(train) [2][3600/3907] lr: 9.2050e-05 eta: 5:33:39 time: 0.6315 data_time: 0.0017 memory: 44138 loss: 0.3144 +2023/06/04 19:19:25 - mmengine - INFO - Epoch(train) [2][3700/3907] lr: 9.1843e-05 eta: 5:32:35 time: 0.6315 data_time: 0.0021 memory: 44138 loss: 0.3164 +2023/06/04 19:20:28 - mmengine - INFO - Epoch(train) [2][3800/3907] lr: 9.1634e-05 eta: 5:31:30 time: 0.6327 data_time: 0.0019 memory: 44138 loss: 0.3164 +2023/06/04 19:21:32 - mmengine - INFO - Epoch(train) [2][3900/3907] lr: 9.1423e-05 eta: 5:30:26 time: 0.6311 data_time: 0.0019 memory: 44138 loss: 0.3063 +2023/06/04 19:21:35 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv1_lr1e-4_20230604_175638 +2023/06/04 19:21:35 - mmengine - INFO - Saving checkpoint at 2 epochs +2023/06/04 19:23:09 - mmengine - INFO - Epoch(val) [2][57/57] accuracy/top1: 86.5454 single-label/precision_classwise: [85.03628540039062, 88.75826263427734] single-label/recall_classwise: [91.7303237915039, 80.17829895019531] single-label/f1-score_classwise: [88.25655364990234, 84.25040435791016] data_time: 0.0348 time: 1.2795 +2023/06/04 19:24:16 - mmengine - INFO - Epoch(train) [3][ 100/3907] lr: 9.1194e-05 eta: 5:29:33 time: 0.6318 data_time: 0.0015 memory: 44138 loss: 0.3150 +2023/06/04 19:25:11 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv1_lr1e-4_20230604_175638 +2023/06/04 19:25:20 - mmengine - INFO - Epoch(train) [3][ 200/3907] lr: 9.0978e-05 eta: 5:28:28 time: 0.6319 data_time: 0.0016 memory: 44138 loss: 0.3322 +2023/06/04 19:26:23 - mmengine - INFO - Epoch(train) [3][ 300/3907] lr: 9.0759e-05 eta: 5:27:24 time: 0.6328 data_time: 0.0021 memory: 44138 loss: 0.3030 +2023/06/04 19:27:26 - mmengine - INFO - Epoch(train) [3][ 400/3907] lr: 9.0539e-05 eta: 5:26:20 time: 0.6329 data_time: 0.0020 memory: 44138 loss: 0.3186 +2023/06/04 19:28:29 - mmengine - INFO - Epoch(train) [3][ 500/3907] lr: 9.0315e-05 eta: 5:25:16 time: 0.6322 data_time: 0.0018 memory: 44138 loss: 0.3062 +2023/06/04 19:29:33 - mmengine - INFO - Epoch(train) [3][ 600/3907] lr: 9.0090e-05 eta: 5:24:11 time: 0.6326 data_time: 0.0016 memory: 44138 loss: 0.2857 +2023/06/04 19:30:36 - mmengine - INFO - Epoch(train) [3][ 700/3907] lr: 8.9862e-05 eta: 5:23:08 time: 0.6335 data_time: 0.0017 memory: 44138 loss: 0.3148 +2023/06/04 19:31:39 - mmengine - INFO - Epoch(train) [3][ 800/3907] lr: 8.9632e-05 eta: 5:22:04 time: 0.6320 data_time: 0.0016 memory: 44138 loss: 0.3290 +2023/06/04 19:32:42 - mmengine - INFO - Epoch(train) [3][ 900/3907] lr: 8.9400e-05 eta: 5:21:00 time: 0.6313 data_time: 0.0015 memory: 44138 loss: 0.3094 +2023/06/04 19:33:46 - mmengine - INFO - Epoch(train) [3][1000/3907] lr: 8.9166e-05 eta: 5:19:55 time: 0.6315 data_time: 0.0016 memory: 44138 loss: 0.3259 +2023/06/04 19:34:49 - mmengine - INFO - Epoch(train) [3][1100/3907] lr: 8.8929e-05 eta: 5:18:51 time: 0.6311 data_time: 0.0014 memory: 44138 loss: 0.3249 +2023/06/04 19:35:43 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv1_lr1e-4_20230604_175638 +2023/06/04 19:35:52 - mmengine - INFO - Epoch(train) [3][1200/3907] lr: 8.8691e-05 eta: 5:17:47 time: 0.6314 data_time: 0.0016 memory: 44138 loss: 0.3466 +2023/06/04 19:36:55 - mmengine - INFO - Epoch(train) [3][1300/3907] lr: 8.8450e-05 eta: 5:16:43 time: 0.6315 data_time: 0.0017 memory: 44138 loss: 0.3328 +2023/06/04 19:37:59 - mmengine - INFO - Epoch(train) [3][1400/3907] lr: 8.8206e-05 eta: 5:15:39 time: 0.6430 data_time: 0.0017 memory: 44138 loss: 0.3047 +2023/06/04 19:39:02 - mmengine - INFO - Epoch(train) [3][1500/3907] lr: 8.7961e-05 eta: 5:14:35 time: 0.6326 data_time: 0.0017 memory: 44138 loss: 0.3104 +2023/06/04 19:40:05 - mmengine - INFO - Epoch(train) [3][1600/3907] lr: 8.7714e-05 eta: 5:13:31 time: 0.6313 data_time: 0.0019 memory: 44138 loss: 0.3151 +2023/06/04 19:41:08 - mmengine - INFO - Epoch(train) [3][1700/3907] lr: 8.7464e-05 eta: 5:12:27 time: 0.6331 data_time: 0.0016 memory: 44138 loss: 0.3046 +2023/06/04 19:42:12 - mmengine - INFO - Epoch(train) [3][1800/3907] lr: 8.7213e-05 eta: 5:11:23 time: 0.6329 data_time: 0.0019 memory: 44138 loss: 0.2984 +2023/06/04 19:43:15 - mmengine - INFO - Epoch(train) [3][1900/3907] lr: 8.6959e-05 eta: 5:10:19 time: 0.6328 data_time: 0.0016 memory: 44138 loss: 0.3286 +2023/06/04 19:44:18 - mmengine - INFO - Epoch(train) [3][2000/3907] lr: 8.6703e-05 eta: 5:09:15 time: 0.6325 data_time: 0.0016 memory: 44138 loss: 0.2875 +2023/06/04 19:45:21 - mmengine - INFO - Epoch(train) [3][2100/3907] lr: 8.6445e-05 eta: 5:08:11 time: 0.6331 data_time: 0.0017 memory: 44138 loss: 0.3135 +2023/06/04 19:46:16 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv1_lr1e-4_20230604_175638 +2023/06/04 19:46:25 - mmengine - INFO - Epoch(train) [3][2200/3907] lr: 8.6186e-05 eta: 5:07:07 time: 0.6326 data_time: 0.0020 memory: 44138 loss: 0.3007 +2023/06/04 19:47:28 - mmengine - INFO - Epoch(train) [3][2300/3907] lr: 8.5924e-05 eta: 5:06:04 time: 0.6326 data_time: 0.0019 memory: 44138 loss: 0.2988 +2023/06/04 19:48:31 - mmengine - INFO - Epoch(train) [3][2400/3907] lr: 8.5660e-05 eta: 5:05:00 time: 0.6326 data_time: 0.0015 memory: 44138 loss: 0.3598 +2023/06/04 19:49:35 - mmengine - INFO - Epoch(train) [3][2500/3907] lr: 8.5394e-05 eta: 5:03:56 time: 0.6326 data_time: 0.0016 memory: 44138 loss: 0.3210 +2023/06/04 19:50:38 - mmengine - INFO - Epoch(train) [3][2600/3907] lr: 8.5126e-05 eta: 5:02:52 time: 0.6336 data_time: 0.0025 memory: 44138 loss: 0.3072 +2023/06/04 19:51:41 - mmengine - INFO - Epoch(train) [3][2700/3907] lr: 8.4856e-05 eta: 5:01:49 time: 0.6324 data_time: 0.0017 memory: 44138 loss: 0.3150 +2023/06/04 19:52:45 - mmengine - INFO - Epoch(train) [3][2800/3907] lr: 8.4585e-05 eta: 5:00:45 time: 0.6329 data_time: 0.0016 memory: 44138 loss: 0.3121 +2023/06/04 19:53:48 - mmengine - INFO - Epoch(train) [3][2900/3907] lr: 8.4311e-05 eta: 4:59:42 time: 0.6328 data_time: 0.0017 memory: 44138 loss: 0.3552 +2023/06/04 19:54:51 - mmengine - INFO - Epoch(train) [3][3000/3907] lr: 8.4036e-05 eta: 4:58:38 time: 0.6336 data_time: 0.0019 memory: 44138 loss: 0.3118 +2023/06/04 19:55:55 - mmengine - INFO - Epoch(train) [3][3100/3907] lr: 8.3758e-05 eta: 4:57:34 time: 0.6317 data_time: 0.0023 memory: 44138 loss: 0.2968 +2023/06/04 19:56:49 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv1_lr1e-4_20230604_175638 +2023/06/04 19:56:58 - mmengine - INFO - Epoch(train) [3][3200/3907] lr: 8.3479e-05 eta: 4:56:30 time: 0.6321 data_time: 0.0016 memory: 44138 loss: 0.3433 +2023/06/04 19:58:01 - mmengine - INFO - Epoch(train) [3][3300/3907] lr: 8.3198e-05 eta: 4:55:26 time: 0.6322 data_time: 0.0015 memory: 44138 loss: 0.3054 +2023/06/04 19:59:04 - mmengine - INFO - Epoch(train) [3][3400/3907] lr: 8.2915e-05 eta: 4:54:23 time: 0.6319 data_time: 0.0023 memory: 44138 loss: 0.3383 +2023/06/04 20:00:08 - mmengine - INFO - Epoch(train) [3][3500/3907] lr: 8.2630e-05 eta: 4:53:19 time: 0.6319 data_time: 0.0020 memory: 44138 loss: 0.2996 +2023/06/04 20:01:11 - mmengine - INFO - Epoch(train) [3][3600/3907] lr: 8.2344e-05 eta: 4:52:15 time: 0.6318 data_time: 0.0016 memory: 44138 loss: 0.3139 +2023/06/04 20:02:14 - mmengine - INFO - Epoch(train) [3][3700/3907] lr: 8.2056e-05 eta: 4:51:11 time: 0.6320 data_time: 0.0016 memory: 44138 loss: 0.2961 +2023/06/04 20:03:17 - mmengine - INFO - Epoch(train) [3][3800/3907] lr: 8.1765e-05 eta: 4:50:07 time: 0.6313 data_time: 0.0015 memory: 44138 loss: 0.3452 +2023/06/04 20:04:21 - mmengine - INFO - Epoch(train) [3][3900/3907] lr: 8.1474e-05 eta: 4:49:04 time: 0.6313 data_time: 0.0013 memory: 44138 loss: 0.3304 +2023/06/04 20:04:25 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv1_lr1e-4_20230604_175638 +2023/06/04 20:04:25 - mmengine - INFO - Saving checkpoint at 3 epochs +2023/06/04 20:05:59 - mmengine - INFO - Epoch(val) [3][57/57] accuracy/top1: 88.3720 single-label/precision_classwise: [86.45936584472656, 91.19828033447266] single-label/recall_classwise: [93.55470275878906, 82.00775146484375] single-label/f1-score_classwise: [89.86720275878906, 86.35919189453125] data_time: 0.0322 time: 1.2769 +2023/06/04 20:07:06 - mmengine - INFO - Epoch(train) [4][ 100/3907] lr: 8.1160e-05 eta: 4:48:04 time: 0.6332 data_time: 0.0015 memory: 44138 loss: 0.2963 +2023/06/04 20:08:10 - mmengine - INFO - Epoch(train) [4][ 200/3907] lr: 8.0864e-05 eta: 4:47:01 time: 0.6327 data_time: 0.0014 memory: 44138 loss: 0.3296 +2023/06/04 20:09:00 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv1_lr1e-4_20230604_175638 +2023/06/04 20:09:13 - mmengine - INFO - Epoch(train) [4][ 300/3907] lr: 8.0567e-05 eta: 4:45:57 time: 0.6334 data_time: 0.0016 memory: 44138 loss: 0.3167 +2023/06/04 20:10:16 - mmengine - INFO - Epoch(train) [4][ 400/3907] lr: 8.0269e-05 eta: 4:44:53 time: 0.6330 data_time: 0.0015 memory: 44138 loss: 0.3112 +2023/06/04 20:11:20 - mmengine - INFO - Epoch(train) [4][ 500/3907] lr: 7.9969e-05 eta: 4:43:50 time: 0.6329 data_time: 0.0015 memory: 44138 loss: 0.3052 +2023/06/04 20:12:23 - mmengine - INFO - Epoch(train) [4][ 600/3907] lr: 7.9667e-05 eta: 4:42:46 time: 0.6325 data_time: 0.0014 memory: 44138 loss: 0.3271 +2023/06/04 20:13:26 - mmengine - INFO - Epoch(train) [4][ 700/3907] lr: 7.9363e-05 eta: 4:41:42 time: 0.6329 data_time: 0.0021 memory: 44138 loss: 0.3198 +2023/06/04 20:14:29 - mmengine - INFO - Epoch(train) [4][ 800/3907] lr: 7.9058e-05 eta: 4:40:39 time: 0.6333 data_time: 0.0015 memory: 44138 loss: 0.3220 +2023/06/04 20:15:33 - mmengine - INFO - Epoch(train) [4][ 900/3907] lr: 7.8752e-05 eta: 4:39:35 time: 0.6323 data_time: 0.0013 memory: 44138 loss: 0.3329 +2023/06/04 20:16:36 - mmengine - INFO - Epoch(train) [4][1000/3907] lr: 7.8444e-05 eta: 4:38:31 time: 0.6328 data_time: 0.0018 memory: 44138 loss: 0.3283 +2023/06/04 20:17:39 - mmengine - INFO - Epoch(train) [4][1100/3907] lr: 7.8134e-05 eta: 4:37:27 time: 0.6352 data_time: 0.0017 memory: 44138 loss: 0.2963 +2023/06/04 20:18:43 - mmengine - INFO - Epoch(train) [4][1200/3907] lr: 7.7823e-05 eta: 4:36:24 time: 0.6332 data_time: 0.0018 memory: 44138 loss: 0.2815 +2023/06/04 20:19:33 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv1_lr1e-4_20230604_175638 +2023/06/04 20:19:46 - mmengine - INFO - Epoch(train) [4][1300/3907] lr: 7.7510e-05 eta: 4:35:20 time: 0.6329 data_time: 0.0015 memory: 44138 loss: 0.2982 +2023/06/04 20:20:49 - mmengine - INFO - Epoch(train) [4][1400/3907] lr: 7.7196e-05 eta: 4:34:17 time: 0.6334 data_time: 0.0021 memory: 44138 loss: 0.3179 +2023/06/04 20:21:53 - mmengine - INFO - Epoch(train) [4][1500/3907] lr: 7.6881e-05 eta: 4:33:13 time: 0.6330 data_time: 0.0018 memory: 44138 loss: 0.3146 +2023/06/04 20:22:56 - mmengine - INFO - Epoch(train) [4][1600/3907] lr: 7.6564e-05 eta: 4:32:10 time: 0.6333 data_time: 0.0018 memory: 44138 loss: 0.3103 +2023/06/04 20:24:00 - mmengine - INFO - Epoch(train) [4][1700/3907] lr: 7.6246e-05 eta: 4:31:06 time: 0.6336 data_time: 0.0021 memory: 44138 loss: 0.3179 +2023/06/04 20:25:03 - mmengine - INFO - Epoch(train) [4][1800/3907] lr: 7.5926e-05 eta: 4:30:03 time: 0.6347 data_time: 0.0016 memory: 44138 loss: 0.3075 +2023/06/04 20:26:06 - mmengine - INFO - Epoch(train) [4][1900/3907] lr: 7.5605e-05 eta: 4:28:59 time: 0.6345 data_time: 0.0015 memory: 44138 loss: 0.3100 +2023/06/04 20:27:10 - mmengine - INFO - Epoch(train) [4][2000/3907] lr: 7.5283e-05 eta: 4:27:56 time: 0.6332 data_time: 0.0018 memory: 44138 loss: 0.3207 +2023/06/04 20:28:13 - mmengine - INFO - Epoch(train) [4][2100/3907] lr: 7.4959e-05 eta: 4:26:52 time: 0.6333 data_time: 0.0022 memory: 44138 loss: 0.3097 +2023/06/04 20:29:17 - mmengine - INFO - Epoch(train) [4][2200/3907] lr: 7.4634e-05 eta: 4:25:49 time: 0.6338 data_time: 0.0021 memory: 44138 loss: 0.3118 +2023/06/04 20:30:07 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv1_lr1e-4_20230604_175638 +2023/06/04 20:30:20 - mmengine - INFO - Epoch(train) [4][2300/3907] lr: 7.4308e-05 eta: 4:24:46 time: 0.6338 data_time: 0.0016 memory: 44138 loss: 0.2965 +2023/06/04 20:31:23 - mmengine - INFO - Epoch(train) [4][2400/3907] lr: 7.3980e-05 eta: 4:23:42 time: 0.6330 data_time: 0.0016 memory: 44138 loss: 0.3057 +2023/06/04 20:32:27 - mmengine - INFO - Epoch(train) [4][2500/3907] lr: 7.3652e-05 eta: 4:22:39 time: 0.6332 data_time: 0.0017 memory: 44138 loss: 0.3190 +2023/06/04 20:33:30 - mmengine - INFO - Epoch(train) [4][2600/3907] lr: 7.3322e-05 eta: 4:21:35 time: 0.6333 data_time: 0.0019 memory: 44138 loss: 0.3105 +2023/06/04 20:34:34 - mmengine - INFO - Epoch(train) [4][2700/3907] lr: 7.2991e-05 eta: 4:20:32 time: 0.6330 data_time: 0.0018 memory: 44138 loss: 0.3334 +2023/06/04 20:35:37 - mmengine - INFO - Epoch(train) [4][2800/3907] lr: 7.2659e-05 eta: 4:19:28 time: 0.6333 data_time: 0.0012 memory: 44138 loss: 0.3280 +2023/06/04 20:36:40 - mmengine - INFO - Epoch(train) [4][2900/3907] lr: 7.2325e-05 eta: 4:18:25 time: 0.6333 data_time: 0.0016 memory: 44138 loss: 0.3360 +2023/06/04 20:37:44 - mmengine - INFO - Epoch(train) [4][3000/3907] lr: 7.1991e-05 eta: 4:17:21 time: 0.6339 data_time: 0.0020 memory: 44138 loss: 0.3111 +2023/06/04 20:38:47 - mmengine - INFO - Epoch(train) [4][3100/3907] lr: 7.1655e-05 eta: 4:16:18 time: 0.6329 data_time: 0.0018 memory: 44138 loss: 0.3148 +2023/06/04 20:39:50 - mmengine - INFO - Epoch(train) [4][3200/3907] lr: 7.1318e-05 eta: 4:15:14 time: 0.6325 data_time: 0.0021 memory: 44138 loss: 0.3174 +2023/06/04 20:40:40 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv1_lr1e-4_20230604_175638 +2023/06/04 20:40:54 - mmengine - INFO - Epoch(train) [4][3300/3907] lr: 7.0981e-05 eta: 4:14:10 time: 0.6315 data_time: 0.0018 memory: 44138 loss: 0.2819 +2023/06/04 20:41:57 - mmengine - INFO - Epoch(train) [4][3400/3907] lr: 7.0642e-05 eta: 4:13:07 time: 0.6328 data_time: 0.0018 memory: 44138 loss: 0.3287 +2023/06/04 20:43:00 - mmengine - INFO - Epoch(train) [4][3500/3907] lr: 7.0302e-05 eta: 4:12:03 time: 0.6321 data_time: 0.0019 memory: 44138 loss: 0.3135 +2023/06/04 20:44:04 - mmengine - INFO - Epoch(train) [4][3600/3907] lr: 6.9961e-05 eta: 4:11:00 time: 0.6325 data_time: 0.0019 memory: 44138 loss: 0.2898 +2023/06/04 20:45:07 - mmengine - INFO - Epoch(train) [4][3700/3907] lr: 6.9620e-05 eta: 4:09:56 time: 0.6332 data_time: 0.0017 memory: 44138 loss: 0.3007 +2023/06/04 20:46:10 - mmengine - INFO - Epoch(train) [4][3800/3907] lr: 6.9277e-05 eta: 4:08:52 time: 0.6334 data_time: 0.0017 memory: 44138 loss: 0.3088 +2023/06/04 20:47:14 - mmengine - INFO - Epoch(train) [4][3900/3907] lr: 6.8933e-05 eta: 4:07:49 time: 0.6310 data_time: 0.0014 memory: 44138 loss: 0.2911 +2023/06/04 20:47:18 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv1_lr1e-4_20230604_175638 +2023/06/04 20:47:18 - mmengine - INFO - Saving checkpoint at 4 epochs +2023/06/04 20:48:51 - mmengine - INFO - Epoch(val) [4][57/57] accuracy/top1: 89.6002 single-label/precision_classwise: [86.67808532714844, 94.16273498535156] single-label/recall_classwise: [95.86515808105469, 81.90697479248047] single-label/f1-score_classwise: [91.04043579101562, 87.60831451416016] data_time: 0.0335 time: 1.2778 +2023/06/04 20:49:59 - mmengine - INFO - Epoch(train) [5][ 100/3907] lr: 6.8565e-05 eta: 4:06:47 time: 0.6340 data_time: 0.0019 memory: 44138 loss: 0.2845 +2023/06/04 20:51:02 - mmengine - INFO - Epoch(train) [5][ 200/3907] lr: 6.8219e-05 eta: 4:05:43 time: 0.6341 data_time: 0.0017 memory: 44138 loss: 0.3264 +2023/06/04 20:52:06 - mmengine - INFO - Epoch(train) [5][ 300/3907] lr: 6.7873e-05 eta: 4:04:40 time: 0.6319 data_time: 0.0016 memory: 44138 loss: 0.3066 +2023/06/04 20:52:51 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv1_lr1e-4_20230604_175638 +2023/06/04 20:53:09 - mmengine - INFO - Epoch(train) [5][ 400/3907] lr: 6.7526e-05 eta: 4:03:36 time: 0.6329 data_time: 0.0017 memory: 44138 loss: 0.2892 +2023/06/04 20:54:12 - mmengine - INFO - Epoch(train) [5][ 500/3907] lr: 6.7178e-05 eta: 4:02:33 time: 0.6326 data_time: 0.0017 memory: 44138 loss: 0.3078 +2023/06/04 20:55:16 - mmengine - INFO - Epoch(train) [5][ 600/3907] lr: 6.6829e-05 eta: 4:01:29 time: 0.6330 data_time: 0.0017 memory: 44138 loss: 0.3136 +2023/06/04 20:56:19 - mmengine - INFO - Epoch(train) [5][ 700/3907] lr: 6.6480e-05 eta: 4:00:26 time: 0.6410 data_time: 0.0018 memory: 44138 loss: 0.2925 +2023/06/04 20:57:22 - mmengine - INFO - Epoch(train) [5][ 800/3907] lr: 6.6129e-05 eta: 3:59:22 time: 0.6332 data_time: 0.0017 memory: 44138 loss: 0.2985 +2023/06/04 20:58:26 - mmengine - INFO - Epoch(train) [5][ 900/3907] lr: 6.5778e-05 eta: 3:58:18 time: 0.6331 data_time: 0.0017 memory: 44138 loss: 0.3070 +2023/06/04 20:59:29 - mmengine - INFO - Epoch(train) [5][1000/3907] lr: 6.5427e-05 eta: 3:57:15 time: 0.6332 data_time: 0.0016 memory: 44138 loss: 0.3102 +2023/06/04 21:00:33 - mmengine - INFO - Epoch(train) [5][1100/3907] lr: 6.5074e-05 eta: 3:56:11 time: 0.6340 data_time: 0.0018 memory: 44138 loss: 0.3002 +2023/06/04 21:01:36 - mmengine - INFO - Epoch(train) [5][1200/3907] lr: 6.4721e-05 eta: 3:55:08 time: 0.6343 data_time: 0.0015 memory: 44138 loss: 0.3075 +2023/06/04 21:02:39 - mmengine - INFO - Epoch(train) [5][1300/3907] lr: 6.4368e-05 eta: 3:54:04 time: 0.6330 data_time: 0.0015 memory: 44138 loss: 0.3246 +2023/06/04 21:03:25 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv1_lr1e-4_20230604_175638 +2023/06/04 21:03:43 - mmengine - INFO - Epoch(train) [5][1400/3907] lr: 6.4014e-05 eta: 3:53:01 time: 0.6320 data_time: 0.0016 memory: 44138 loss: 0.3091 +2023/06/04 21:04:46 - mmengine - INFO - Epoch(train) [5][1500/3907] lr: 6.3659e-05 eta: 3:51:57 time: 0.6321 data_time: 0.0016 memory: 44138 loss: 0.2932 +2023/06/04 21:05:49 - mmengine - INFO - Epoch(train) [5][1600/3907] lr: 6.3303e-05 eta: 3:50:54 time: 0.6327 data_time: 0.0018 memory: 44138 loss: 0.3367 +2023/06/04 21:06:52 - mmengine - INFO - Epoch(train) [5][1700/3907] lr: 6.2948e-05 eta: 3:49:50 time: 0.6335 data_time: 0.0021 memory: 44138 loss: 0.2894 +2023/06/04 21:07:56 - mmengine - INFO - Epoch(train) [5][1800/3907] lr: 6.2591e-05 eta: 3:48:46 time: 0.6327 data_time: 0.0016 memory: 44138 loss: 0.3157 +2023/06/04 21:08:59 - mmengine - INFO - Epoch(train) [5][1900/3907] lr: 6.2234e-05 eta: 3:47:43 time: 0.6338 data_time: 0.0021 memory: 44138 loss: 0.2980 +2023/06/04 21:10:02 - mmengine - INFO - Epoch(train) [5][2000/3907] lr: 6.1877e-05 eta: 3:46:39 time: 0.6328 data_time: 0.0016 memory: 44138 loss: 0.2880 +2023/06/04 21:11:06 - mmengine - INFO - Epoch(train) [5][2100/3907] lr: 6.1519e-05 eta: 3:45:36 time: 0.6327 data_time: 0.0019 memory: 44138 loss: 0.3008 +2023/06/04 21:12:09 - mmengine - INFO - Epoch(train) [5][2200/3907] lr: 6.1161e-05 eta: 3:44:32 time: 0.6323 data_time: 0.0018 memory: 44138 loss: 0.2898 +2023/06/04 21:13:12 - mmengine - INFO - Epoch(train) [5][2300/3907] lr: 6.0802e-05 eta: 3:43:29 time: 0.6336 data_time: 0.0015 memory: 44138 loss: 0.3067 +2023/06/04 21:13:58 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv1_lr1e-4_20230604_175638 +2023/06/04 21:14:16 - mmengine - INFO - Epoch(train) [5][2400/3907] lr: 6.0443e-05 eta: 3:42:25 time: 0.6319 data_time: 0.0016 memory: 44138 loss: 0.2871 +2023/06/04 21:15:19 - mmengine - INFO - Epoch(train) [5][2500/3907] lr: 6.0084e-05 eta: 3:41:21 time: 0.6328 data_time: 0.0020 memory: 44138 loss: 0.2968 +2023/06/04 21:16:22 - mmengine - INFO - Epoch(train) [5][2600/3907] lr: 5.9724e-05 eta: 3:40:18 time: 0.6322 data_time: 0.0017 memory: 44138 loss: 0.2993 +2023/06/04 21:17:25 - mmengine - INFO - Epoch(train) [5][2700/3907] lr: 5.9364e-05 eta: 3:39:14 time: 0.6319 data_time: 0.0016 memory: 44138 loss: 0.3093 +2023/06/04 21:18:29 - mmengine - INFO - Epoch(train) [5][2800/3907] lr: 5.9004e-05 eta: 3:38:11 time: 0.6327 data_time: 0.0017 memory: 44138 loss: 0.3270 +2023/06/04 21:19:32 - mmengine - INFO - Epoch(train) [5][2900/3907] lr: 5.8643e-05 eta: 3:37:07 time: 0.6319 data_time: 0.0016 memory: 44138 loss: 0.3123 +2023/06/04 21:20:35 - mmengine - INFO - Epoch(train) [5][3000/3907] lr: 5.8283e-05 eta: 3:36:03 time: 0.6325 data_time: 0.0020 memory: 44138 loss: 0.3098 +2023/06/04 21:21:38 - mmengine - INFO - Epoch(train) [5][3100/3907] lr: 5.7922e-05 eta: 3:35:00 time: 0.6325 data_time: 0.0020 memory: 44138 loss: 0.3118 +2023/06/04 21:22:42 - mmengine - INFO - Epoch(train) [5][3200/3907] lr: 5.7560e-05 eta: 3:33:56 time: 0.6332 data_time: 0.0016 memory: 44138 loss: 0.3263 +2023/06/04 21:23:45 - mmengine - INFO - Epoch(train) [5][3300/3907] lr: 5.7199e-05 eta: 3:32:53 time: 0.6332 data_time: 0.0016 memory: 44138 loss: 0.3053 +2023/06/04 21:24:31 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv1_lr1e-4_20230604_175638 +2023/06/04 21:24:49 - mmengine - INFO - Epoch(train) [5][3400/3907] lr: 5.6838e-05 eta: 3:31:49 time: 0.6325 data_time: 0.0016 memory: 44138 loss: 0.3196 +2023/06/04 21:25:52 - mmengine - INFO - Epoch(train) [5][3500/3907] lr: 5.6476e-05 eta: 3:30:46 time: 0.6329 data_time: 0.0019 memory: 44138 loss: 0.2998 +2023/06/04 21:26:55 - mmengine - INFO - Epoch(train) [5][3600/3907] lr: 5.6114e-05 eta: 3:29:42 time: 0.6360 data_time: 0.0030 memory: 44138 loss: 0.2985 +2023/06/04 21:27:59 - mmengine - INFO - Epoch(train) [5][3700/3907] lr: 5.5753e-05 eta: 3:28:39 time: 0.6392 data_time: 0.0024 memory: 44138 loss: 0.3244 +2023/06/04 21:29:03 - mmengine - INFO - Epoch(train) [5][3800/3907] lr: 5.5391e-05 eta: 3:27:36 time: 0.6362 data_time: 0.0019 memory: 44138 loss: 0.2790 +2023/06/04 21:30:06 - mmengine - INFO - Epoch(train) [5][3900/3907] lr: 5.5029e-05 eta: 3:26:33 time: 0.6336 data_time: 0.0015 memory: 44138 loss: 0.3025 +2023/06/04 21:30:10 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv1_lr1e-4_20230604_175638 +2023/06/04 21:30:10 - mmengine - INFO - Saving checkpoint at 5 epochs +2023/06/04 21:31:46 - mmengine - INFO - Epoch(val) [5][57/57] accuracy/top1: 89.9551 single-label/precision_classwise: [86.16819763183594, 96.21526336669922] single-label/recall_classwise: [97.41178131103516, 80.79844665527344] single-label/f1-score_classwise: [91.44567108154297, 87.83550262451172] data_time: 0.0375 time: 1.2829 +2023/06/04 21:32:54 - mmengine - INFO - Epoch(train) [6][ 100/3907] lr: 5.4642e-05 eta: 3:25:29 time: 0.6334 data_time: 0.0019 memory: 44138 loss: 0.2939 +2023/06/04 21:33:57 - mmengine - INFO - Epoch(train) [6][ 200/3907] lr: 5.4280e-05 eta: 3:24:26 time: 0.6334 data_time: 0.0016 memory: 44138 loss: 0.3412 +2023/06/04 21:35:01 - mmengine - INFO - Epoch(train) [6][ 300/3907] lr: 5.3918e-05 eta: 3:23:22 time: 0.6372 data_time: 0.0027 memory: 44138 loss: 0.2942 +2023/06/04 21:36:04 - mmengine - INFO - Epoch(train) [6][ 400/3907] lr: 5.3556e-05 eta: 3:22:19 time: 0.6330 data_time: 0.0017 memory: 44138 loss: 0.3073 +2023/06/04 21:36:45 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv1_lr1e-4_20230604_175638 +2023/06/04 21:37:08 - mmengine - INFO - Epoch(train) [6][ 500/3907] lr: 5.3195e-05 eta: 3:21:15 time: 0.6343 data_time: 0.0020 memory: 44138 loss: 0.3229 +2023/06/04 21:38:11 - mmengine - INFO - Epoch(train) [6][ 600/3907] lr: 5.2833e-05 eta: 3:20:12 time: 0.6345 data_time: 0.0023 memory: 44138 loss: 0.3329 +2023/06/04 21:39:14 - mmengine - INFO - Epoch(train) [6][ 700/3907] lr: 5.2472e-05 eta: 3:19:08 time: 0.6340 data_time: 0.0020 memory: 44138 loss: 0.2772 +2023/06/04 21:40:18 - mmengine - INFO - Epoch(train) [6][ 800/3907] lr: 5.2111e-05 eta: 3:18:05 time: 0.6333 data_time: 0.0029 memory: 44138 loss: 0.3145 +2023/06/04 21:41:21 - mmengine - INFO - Epoch(train) [6][ 900/3907] lr: 5.1750e-05 eta: 3:17:01 time: 0.6319 data_time: 0.0016 memory: 44138 loss: 0.3119 +2023/06/04 21:42:25 - mmengine - INFO - Epoch(train) [6][1000/3907] lr: 5.1389e-05 eta: 3:15:58 time: 0.6348 data_time: 0.0027 memory: 44138 loss: 0.2984 +2023/06/04 21:43:28 - mmengine - INFO - Epoch(train) [6][1100/3907] lr: 5.1029e-05 eta: 3:14:55 time: 0.6356 data_time: 0.0022 memory: 44138 loss: 0.2980 +2023/06/04 21:44:32 - mmengine - INFO - Epoch(train) [6][1200/3907] lr: 5.0668e-05 eta: 3:13:51 time: 0.6335 data_time: 0.0018 memory: 44138 loss: 0.3006 +2023/06/04 21:45:35 - mmengine - INFO - Epoch(train) [6][1300/3907] lr: 5.0308e-05 eta: 3:12:48 time: 0.6333 data_time: 0.0018 memory: 44138 loss: 0.3097 +2023/06/04 21:46:38 - mmengine - INFO - Epoch(train) [6][1400/3907] lr: 4.9949e-05 eta: 3:11:44 time: 0.6342 data_time: 0.0025 memory: 44138 loss: 0.3224 +2023/06/04 21:47:20 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv1_lr1e-4_20230604_175638 +2023/06/04 21:47:42 - mmengine - INFO - Epoch(train) [6][1500/3907] lr: 4.9589e-05 eta: 3:10:41 time: 0.6354 data_time: 0.0016 memory: 44138 loss: 0.3052 +2023/06/04 21:48:45 - mmengine - INFO - Epoch(train) [6][1600/3907] lr: 4.9230e-05 eta: 3:09:37 time: 0.6345 data_time: 0.0021 memory: 44138 loss: 0.2933 +2023/06/04 21:49:49 - mmengine - INFO - Epoch(train) [6][1700/3907] lr: 4.8871e-05 eta: 3:08:34 time: 0.6330 data_time: 0.0022 memory: 44138 loss: 0.2998 +2023/06/04 21:50:52 - mmengine - INFO - Epoch(train) [6][1800/3907] lr: 4.8513e-05 eta: 3:07:30 time: 0.6326 data_time: 0.0019 memory: 44138 loss: 0.2937 +2023/06/04 21:51:55 - mmengine - INFO - Epoch(train) [6][1900/3907] lr: 4.8155e-05 eta: 3:06:27 time: 0.6338 data_time: 0.0016 memory: 44138 loss: 0.3262 +2023/06/04 21:52:59 - mmengine - INFO - Epoch(train) [6][2000/3907] lr: 4.7798e-05 eta: 3:05:23 time: 0.6367 data_time: 0.0025 memory: 44138 loss: 0.2895 +2023/06/04 21:54:02 - mmengine - INFO - Epoch(train) [6][2100/3907] lr: 4.7441e-05 eta: 3:04:20 time: 0.6337 data_time: 0.0024 memory: 44138 loss: 0.3103 +2023/06/04 21:55:06 - mmengine - INFO - Epoch(train) [6][2200/3907] lr: 4.7084e-05 eta: 3:03:16 time: 0.6383 data_time: 0.0048 memory: 44138 loss: 0.2789 +2023/06/04 21:56:09 - mmengine - INFO - Epoch(train) [6][2300/3907] lr: 4.6729e-05 eta: 3:02:13 time: 0.6328 data_time: 0.0016 memory: 44138 loss: 0.3134 +2023/06/04 21:57:12 - mmengine - INFO - Epoch(train) [6][2400/3907] lr: 4.6373e-05 eta: 3:01:09 time: 0.6368 data_time: 0.0028 memory: 44138 loss: 0.2875 +2023/06/04 21:57:54 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv1_lr1e-4_20230604_175638 +2023/06/04 21:58:16 - mmengine - INFO - Epoch(train) [6][2500/3907] lr: 4.6018e-05 eta: 3:00:06 time: 0.6371 data_time: 0.0043 memory: 44138 loss: 0.3104 +2023/06/04 21:59:19 - mmengine - INFO - Epoch(train) [6][2600/3907] lr: 4.5664e-05 eta: 2:59:03 time: 0.6331 data_time: 0.0017 memory: 44138 loss: 0.3060 +2023/06/04 22:00:23 - mmengine - INFO - Epoch(train) [6][2700/3907] lr: 4.5310e-05 eta: 2:57:59 time: 0.6324 data_time: 0.0019 memory: 44138 loss: 0.3336 +2023/06/04 22:01:26 - mmengine - INFO - Epoch(train) [6][2800/3907] lr: 4.4957e-05 eta: 2:56:56 time: 0.6362 data_time: 0.0032 memory: 44138 loss: 0.3457 +2023/06/04 22:02:30 - mmengine - INFO - Epoch(train) [6][2900/3907] lr: 4.4605e-05 eta: 2:55:52 time: 0.6330 data_time: 0.0022 memory: 44138 loss: 0.3135 +2023/06/04 22:03:33 - mmengine - INFO - Epoch(train) [6][3000/3907] lr: 4.4253e-05 eta: 2:54:49 time: 0.6337 data_time: 0.0017 memory: 44138 loss: 0.3151 +2023/06/04 22:04:37 - mmengine - INFO - Epoch(train) [6][3100/3907] lr: 4.3902e-05 eta: 2:53:46 time: 0.6340 data_time: 0.0017 memory: 44138 loss: 0.3200 +2023/06/04 22:05:40 - mmengine - INFO - Epoch(train) [6][3200/3907] lr: 4.3552e-05 eta: 2:52:42 time: 0.6322 data_time: 0.0017 memory: 44138 loss: 0.3051 +2023/06/04 22:06:44 - mmengine - INFO - Epoch(train) [6][3300/3907] lr: 4.3202e-05 eta: 2:51:39 time: 0.6345 data_time: 0.0025 memory: 44138 loss: 0.3207 +2023/06/04 22:07:47 - mmengine - INFO - Epoch(train) [6][3400/3907] lr: 4.2854e-05 eta: 2:50:35 time: 0.6325 data_time: 0.0026 memory: 44138 loss: 0.2974 +2023/06/04 22:08:28 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv1_lr1e-4_20230604_175638 +2023/06/04 22:08:51 - mmengine - INFO - Epoch(train) [6][3500/3907] lr: 4.2506e-05 eta: 2:49:32 time: 0.6344 data_time: 0.0019 memory: 44138 loss: 0.2820 +2023/06/04 22:09:54 - mmengine - INFO - Epoch(train) [6][3600/3907] lr: 4.2158e-05 eta: 2:48:28 time: 0.6473 data_time: 0.0020 memory: 44138 loss: 0.3054 +2023/06/04 22:10:58 - mmengine - INFO - Epoch(train) [6][3700/3907] lr: 4.1812e-05 eta: 2:47:25 time: 0.6328 data_time: 0.0017 memory: 44138 loss: 0.3048 +2023/06/04 22:12:01 - mmengine - INFO - Epoch(train) [6][3800/3907] lr: 4.1466e-05 eta: 2:46:21 time: 0.6334 data_time: 0.0017 memory: 44138 loss: 0.2996 +2023/06/04 22:13:04 - mmengine - INFO - Epoch(train) [6][3900/3907] lr: 4.1122e-05 eta: 2:45:18 time: 0.6319 data_time: 0.0014 memory: 44138 loss: 0.3015 +2023/06/04 22:13:08 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv1_lr1e-4_20230604_175638 +2023/06/04 22:13:08 - mmengine - INFO - Saving checkpoint at 6 epochs +2023/06/04 22:14:44 - mmengine - INFO - Epoch(val) [6][57/57] accuracy/top1: 90.2474 single-label/precision_classwise: [86.16844177246094, 97.1073989868164] single-label/recall_classwise: [98.04305267333984, 80.6744155883789] single-label/f1-score_classwise: [91.72301483154297, 88.13142395019531] data_time: 0.0391 time: 1.2822 +2023/06/04 22:15:52 - mmengine - INFO - Epoch(train) [7][ 100/3907] lr: 4.0754e-05 eta: 2:44:13 time: 0.6333 data_time: 0.0024 memory: 44138 loss: 0.3281 +2023/06/04 22:16:55 - mmengine - INFO - Epoch(train) [7][ 200/3907] lr: 4.0411e-05 eta: 2:43:09 time: 0.6317 data_time: 0.0016 memory: 44138 loss: 0.3343 +2023/06/04 22:17:59 - mmengine - INFO - Epoch(train) [7][ 300/3907] lr: 4.0069e-05 eta: 2:42:06 time: 0.6351 data_time: 0.0022 memory: 44138 loss: 0.3084 +2023/06/04 22:19:02 - mmengine - INFO - Epoch(train) [7][ 400/3907] lr: 3.9729e-05 eta: 2:41:02 time: 0.6350 data_time: 0.0030 memory: 44138 loss: 0.3170 +2023/06/04 22:20:06 - mmengine - INFO - Epoch(train) [7][ 500/3907] lr: 3.9389e-05 eta: 2:39:59 time: 0.6345 data_time: 0.0021 memory: 44138 loss: 0.2958 +2023/06/04 22:20:42 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv1_lr1e-4_20230604_175638 +2023/06/04 22:21:09 - mmengine - INFO - Epoch(train) [7][ 600/3907] lr: 3.9050e-05 eta: 2:38:55 time: 0.6339 data_time: 0.0024 memory: 44138 loss: 0.3237 +2023/06/04 22:22:12 - mmengine - INFO - Epoch(train) [7][ 700/3907] lr: 3.8712e-05 eta: 2:37:52 time: 0.6341 data_time: 0.0018 memory: 44138 loss: 0.2734 +2023/06/04 22:23:16 - mmengine - INFO - Epoch(train) [7][ 800/3907] lr: 3.8375e-05 eta: 2:36:49 time: 0.6356 data_time: 0.0019 memory: 44138 loss: 0.3238 +2023/06/04 22:24:19 - mmengine - INFO - Epoch(train) [7][ 900/3907] lr: 3.8039e-05 eta: 2:35:45 time: 0.6321 data_time: 0.0015 memory: 44138 loss: 0.2950 +2023/06/04 22:25:23 - mmengine - INFO - Epoch(train) [7][1000/3907] lr: 3.7705e-05 eta: 2:34:41 time: 0.6325 data_time: 0.0017 memory: 44138 loss: 0.3118 +2023/06/04 22:26:26 - mmengine - INFO - Epoch(train) [7][1100/3907] lr: 3.7371e-05 eta: 2:33:38 time: 0.6333 data_time: 0.0018 memory: 44138 loss: 0.3153 +2023/06/04 22:27:29 - mmengine - INFO - Epoch(train) [7][1200/3907] lr: 3.7039e-05 eta: 2:32:35 time: 0.6328 data_time: 0.0019 memory: 44138 loss: 0.3241 +2023/06/04 22:28:33 - mmengine - INFO - Epoch(train) [7][1300/3907] lr: 3.6708e-05 eta: 2:31:31 time: 0.6323 data_time: 0.0022 memory: 44138 loss: 0.3212 +2023/06/04 22:29:36 - mmengine - INFO - Epoch(train) [7][1400/3907] lr: 3.6378e-05 eta: 2:30:28 time: 0.6450 data_time: 0.0017 memory: 44138 loss: 0.3086 +2023/06/04 22:30:40 - mmengine - INFO - Epoch(train) [7][1500/3907] lr: 3.6049e-05 eta: 2:29:24 time: 0.6321 data_time: 0.0019 memory: 44138 loss: 0.3103 +2023/06/04 22:31:17 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv1_lr1e-4_20230604_175638 +2023/06/04 22:31:43 - mmengine - INFO - Epoch(train) [7][1600/3907] lr: 3.5721e-05 eta: 2:28:21 time: 0.6346 data_time: 0.0025 memory: 44138 loss: 0.2907 +2023/06/04 22:32:47 - mmengine - INFO - Epoch(train) [7][1700/3907] lr: 3.5395e-05 eta: 2:27:17 time: 0.6322 data_time: 0.0022 memory: 44138 loss: 0.3239 +2023/06/04 22:33:50 - mmengine - INFO - Epoch(train) [7][1800/3907] lr: 3.5070e-05 eta: 2:26:14 time: 0.6340 data_time: 0.0022 memory: 44138 loss: 0.3201 +2023/06/04 22:34:54 - mmengine - INFO - Epoch(train) [7][1900/3907] lr: 3.4746e-05 eta: 2:25:10 time: 0.6345 data_time: 0.0023 memory: 44138 loss: 0.2931 +2023/06/04 22:35:57 - mmengine - INFO - Epoch(train) [7][2000/3907] lr: 3.4424e-05 eta: 2:24:07 time: 0.6320 data_time: 0.0015 memory: 44138 loss: 0.3008 +2023/06/04 22:37:00 - mmengine - INFO - Epoch(train) [7][2100/3907] lr: 3.4103e-05 eta: 2:23:03 time: 0.6340 data_time: 0.0029 memory: 44138 loss: 0.2855 +2023/06/04 22:38:04 - mmengine - INFO - Epoch(train) [7][2200/3907] lr: 3.3783e-05 eta: 2:22:00 time: 0.6375 data_time: 0.0033 memory: 44138 loss: 0.3113 +2023/06/04 22:39:07 - mmengine - INFO - Epoch(train) [7][2300/3907] lr: 3.3465e-05 eta: 2:20:56 time: 0.6335 data_time: 0.0022 memory: 44138 loss: 0.2925 +2023/06/04 22:40:11 - mmengine - INFO - Epoch(train) [7][2400/3907] lr: 3.3148e-05 eta: 2:19:53 time: 0.6351 data_time: 0.0018 memory: 44138 loss: 0.3041 +2023/06/04 22:41:14 - mmengine - INFO - Epoch(train) [7][2500/3907] lr: 3.2832e-05 eta: 2:18:50 time: 0.6324 data_time: 0.0015 memory: 44138 loss: 0.3046 +2023/06/04 22:41:51 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv1_lr1e-4_20230604_175638 +2023/06/04 22:42:18 - mmengine - INFO - Epoch(train) [7][2600/3907] lr: 3.2518e-05 eta: 2:17:46 time: 0.6425 data_time: 0.0022 memory: 44138 loss: 0.3483 +2023/06/04 22:43:21 - mmengine - INFO - Epoch(train) [7][2700/3907] lr: 3.2205e-05 eta: 2:16:43 time: 0.6366 data_time: 0.0026 memory: 44138 loss: 0.3097 +2023/06/04 22:44:25 - mmengine - INFO - Epoch(train) [7][2800/3907] lr: 3.1894e-05 eta: 2:15:39 time: 0.6443 data_time: 0.0017 memory: 44138 loss: 0.3339 +2023/06/04 22:45:28 - mmengine - INFO - Epoch(train) [7][2900/3907] lr: 3.1584e-05 eta: 2:14:36 time: 0.6328 data_time: 0.0017 memory: 44138 loss: 0.2889 +2023/06/04 22:46:31 - mmengine - INFO - Epoch(train) [7][3000/3907] lr: 3.1276e-05 eta: 2:13:32 time: 0.6321 data_time: 0.0016 memory: 44138 loss: 0.3262 +2023/06/04 22:47:35 - mmengine - INFO - Epoch(train) [7][3100/3907] lr: 3.0969e-05 eta: 2:12:29 time: 0.6327 data_time: 0.0020 memory: 44138 loss: 0.3024 +2023/06/04 22:48:38 - mmengine - INFO - Epoch(train) [7][3200/3907] lr: 3.0664e-05 eta: 2:11:25 time: 0.6321 data_time: 0.0018 memory: 44138 loss: 0.3252 +2023/06/04 22:49:42 - mmengine - INFO - Epoch(train) [7][3300/3907] lr: 3.0360e-05 eta: 2:10:22 time: 0.6322 data_time: 0.0016 memory: 44138 loss: 0.2930 +2023/06/04 22:50:45 - mmengine - INFO - Epoch(train) [7][3400/3907] lr: 3.0058e-05 eta: 2:09:18 time: 0.6380 data_time: 0.0022 memory: 44138 loss: 0.3042 +2023/06/04 22:51:48 - mmengine - INFO - Epoch(train) [7][3500/3907] lr: 2.9758e-05 eta: 2:08:15 time: 0.6333 data_time: 0.0029 memory: 44138 loss: 0.3156 +2023/06/04 22:52:25 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv1_lr1e-4_20230604_175638 +2023/06/04 22:52:52 - mmengine - INFO - Epoch(train) [7][3600/3907] lr: 2.9459e-05 eta: 2:07:11 time: 0.6332 data_time: 0.0018 memory: 44138 loss: 0.2985 +2023/06/04 22:53:55 - mmengine - INFO - Epoch(train) [7][3700/3907] lr: 2.9162e-05 eta: 2:06:08 time: 0.6411 data_time: 0.0046 memory: 44138 loss: 0.3147 +2023/06/04 22:54:59 - mmengine - INFO - Epoch(train) [7][3800/3907] lr: 2.8867e-05 eta: 2:05:04 time: 0.6333 data_time: 0.0033 memory: 44138 loss: 0.3242 +2023/06/04 22:56:02 - mmengine - INFO - Epoch(train) [7][3900/3907] lr: 2.8573e-05 eta: 2:04:01 time: 0.6328 data_time: 0.0015 memory: 44138 loss: 0.3165 +2023/06/04 22:56:06 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv1_lr1e-4_20230604_175638 +2023/06/04 22:56:06 - mmengine - INFO - Saving checkpoint at 7 epochs +2023/06/04 22:57:42 - mmengine - INFO - Epoch(val) [7][57/57] accuracy/top1: 89.6733 single-label/precision_classwise: [85.07056427001953, 97.80516052246094] single-label/recall_classwise: [98.56069946289062, 78.75968933105469] single-label/f1-score_classwise: [91.32012176513672, 87.25524139404297] data_time: 0.0415 time: 1.2863 +2023/06/04 22:58:50 - mmengine - INFO - Epoch(train) [8][ 100/3907] lr: 2.8261e-05 eta: 2:02:55 time: 0.6316 data_time: 0.0017 memory: 44138 loss: 0.2999 +2023/06/04 22:59:54 - mmengine - INFO - Epoch(train) [8][ 200/3907] lr: 2.7971e-05 eta: 2:01:51 time: 0.6331 data_time: 0.0019 memory: 44138 loss: 0.3440 +2023/06/04 23:00:57 - mmengine - INFO - Epoch(train) [8][ 300/3907] lr: 2.7682e-05 eta: 2:00:48 time: 0.6333 data_time: 0.0016 memory: 44138 loss: 0.3049 +2023/06/04 23:02:00 - mmengine - INFO - Epoch(train) [8][ 400/3907] lr: 2.7395e-05 eta: 1:59:44 time: 0.6331 data_time: 0.0020 memory: 44138 loss: 0.3211 +2023/06/04 23:03:04 - mmengine - INFO - Epoch(train) [8][ 500/3907] lr: 2.7111e-05 eta: 1:58:41 time: 0.6324 data_time: 0.0017 memory: 44138 loss: 0.3021 +2023/06/04 23:04:07 - mmengine - INFO - Epoch(train) [8][ 600/3907] lr: 2.6828e-05 eta: 1:57:37 time: 0.6333 data_time: 0.0021 memory: 44138 loss: 0.3054 +2023/06/04 23:04:40 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv1_lr1e-4_20230604_175638 +2023/06/04 23:05:11 - mmengine - INFO - Epoch(train) [8][ 700/3907] lr: 2.6546e-05 eta: 1:56:34 time: 0.6351 data_time: 0.0017 memory: 44138 loss: 0.3280 +2023/06/04 23:06:14 - mmengine - INFO - Epoch(train) [8][ 800/3907] lr: 2.6267e-05 eta: 1:55:31 time: 0.6350 data_time: 0.0016 memory: 44138 loss: 0.3179 +2023/06/04 23:07:18 - mmengine - INFO - Epoch(train) [8][ 900/3907] lr: 2.5989e-05 eta: 1:54:27 time: 0.6343 data_time: 0.0020 memory: 44138 loss: 0.3035 +2023/06/04 23:08:21 - mmengine - INFO - Epoch(train) [8][1000/3907] lr: 2.5714e-05 eta: 1:53:24 time: 0.6349 data_time: 0.0029 memory: 44138 loss: 0.3111 +2023/06/04 23:09:25 - mmengine - INFO - Epoch(train) [8][1100/3907] lr: 2.5440e-05 eta: 1:52:20 time: 0.6355 data_time: 0.0035 memory: 44138 loss: 0.3269 +2023/06/04 23:10:28 - mmengine - INFO - Epoch(train) [8][1200/3907] lr: 2.5168e-05 eta: 1:51:17 time: 0.6330 data_time: 0.0017 memory: 44138 loss: 0.2808 +2023/06/04 23:11:32 - mmengine - INFO - Epoch(train) [8][1300/3907] lr: 2.4898e-05 eta: 1:50:13 time: 0.6333 data_time: 0.0026 memory: 44138 loss: 0.3033 +2023/06/04 23:12:35 - mmengine - INFO - Epoch(train) [8][1400/3907] lr: 2.4630e-05 eta: 1:49:10 time: 0.6323 data_time: 0.0015 memory: 44138 loss: 0.2709 +2023/06/04 23:13:39 - mmengine - INFO - Epoch(train) [8][1500/3907] lr: 2.4364e-05 eta: 1:48:06 time: 0.6395 data_time: 0.0030 memory: 44138 loss: 0.3023 +2023/06/04 23:14:42 - mmengine - INFO - Epoch(train) [8][1600/3907] lr: 2.4100e-05 eta: 1:47:03 time: 0.6360 data_time: 0.0020 memory: 44138 loss: 0.2782 +2023/06/04 23:15:15 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv1_lr1e-4_20230604_175638 +2023/06/04 23:15:46 - mmengine - INFO - Epoch(train) [8][1700/3907] lr: 2.3838e-05 eta: 1:45:59 time: 0.6334 data_time: 0.0019 memory: 44138 loss: 0.3275 +2023/06/04 23:16:49 - mmengine - INFO - Epoch(train) [8][1800/3907] lr: 2.3578e-05 eta: 1:44:56 time: 0.6330 data_time: 0.0017 memory: 44138 loss: 0.3072 +2023/06/04 23:17:53 - mmengine - INFO - Epoch(train) [8][1900/3907] lr: 2.3320e-05 eta: 1:43:52 time: 0.6342 data_time: 0.0018 memory: 44138 loss: 0.3036 +2023/06/04 23:18:56 - mmengine - INFO - Epoch(train) [8][2000/3907] lr: 2.3064e-05 eta: 1:42:49 time: 0.6338 data_time: 0.0021 memory: 44138 loss: 0.2911 +2023/06/04 23:19:59 - mmengine - INFO - Epoch(train) [8][2100/3907] lr: 2.2810e-05 eta: 1:41:46 time: 0.6339 data_time: 0.0019 memory: 44138 loss: 0.2986 +2023/06/04 23:21:03 - mmengine - INFO - Epoch(train) [8][2200/3907] lr: 2.2558e-05 eta: 1:40:42 time: 0.6327 data_time: 0.0015 memory: 44138 loss: 0.2997 +2023/06/04 23:22:06 - mmengine - INFO - Epoch(train) [8][2300/3907] lr: 2.2309e-05 eta: 1:39:39 time: 0.6375 data_time: 0.0018 memory: 44138 loss: 0.3540 +2023/06/04 23:23:10 - mmengine - INFO - Epoch(train) [8][2400/3907] lr: 2.2061e-05 eta: 1:38:35 time: 0.6361 data_time: 0.0025 memory: 44138 loss: 0.3093 +2023/06/04 23:24:13 - mmengine - INFO - Epoch(train) [8][2500/3907] lr: 2.1816e-05 eta: 1:37:32 time: 0.6323 data_time: 0.0018 memory: 44138 loss: 0.2852 +2023/06/04 23:25:17 - mmengine - INFO - Epoch(train) [8][2600/3907] lr: 2.1572e-05 eta: 1:36:28 time: 0.6345 data_time: 0.0020 memory: 44138 loss: 0.3307 +2023/06/04 23:25:49 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv1_lr1e-4_20230604_175638 +2023/06/04 23:26:20 - mmengine - INFO - Epoch(train) [8][2700/3907] lr: 2.1331e-05 eta: 1:35:25 time: 0.6406 data_time: 0.0019 memory: 44138 loss: 0.3068 +2023/06/04 23:27:24 - mmengine - INFO - Epoch(train) [8][2800/3907] lr: 2.1092e-05 eta: 1:34:21 time: 0.6329 data_time: 0.0019 memory: 44138 loss: 0.2755 +2023/06/04 23:28:27 - mmengine - INFO - Epoch(train) [8][2900/3907] lr: 2.0855e-05 eta: 1:33:18 time: 0.6328 data_time: 0.0020 memory: 44138 loss: 0.2961 +2023/06/04 23:29:31 - mmengine - INFO - Epoch(train) [8][3000/3907] lr: 2.0621e-05 eta: 1:32:14 time: 0.6379 data_time: 0.0053 memory: 44138 loss: 0.3122 +2023/06/04 23:30:34 - mmengine - INFO - Epoch(train) [8][3100/3907] lr: 2.0388e-05 eta: 1:31:11 time: 0.6348 data_time: 0.0018 memory: 44138 loss: 0.3067 +2023/06/04 23:31:38 - mmengine - INFO - Epoch(train) [8][3200/3907] lr: 2.0158e-05 eta: 1:30:07 time: 0.6333 data_time: 0.0019 memory: 44138 loss: 0.3280 +2023/06/04 23:32:41 - mmengine - INFO - Epoch(train) [8][3300/3907] lr: 1.9930e-05 eta: 1:29:04 time: 0.6320 data_time: 0.0016 memory: 44138 loss: 0.2993 +2023/06/04 23:33:44 - mmengine - INFO - Epoch(train) [8][3400/3907] lr: 1.9705e-05 eta: 1:28:00 time: 0.6325 data_time: 0.0016 memory: 44138 loss: 0.3224 +2023/06/04 23:34:48 - mmengine - INFO - Epoch(train) [8][3500/3907] lr: 1.9481e-05 eta: 1:26:57 time: 0.6330 data_time: 0.0020 memory: 44138 loss: 0.3192 +2023/06/04 23:35:51 - mmengine - INFO - Epoch(train) [8][3600/3907] lr: 1.9260e-05 eta: 1:25:53 time: 0.6341 data_time: 0.0022 memory: 44138 loss: 0.2851 +2023/06/04 23:36:24 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv1_lr1e-4_20230604_175638 +2023/06/04 23:36:55 - mmengine - INFO - Epoch(train) [8][3700/3907] lr: 1.9042e-05 eta: 1:24:50 time: 0.6330 data_time: 0.0023 memory: 44138 loss: 0.3149 +2023/06/04 23:37:58 - mmengine - INFO - Epoch(train) [8][3800/3907] lr: 1.8825e-05 eta: 1:23:46 time: 0.6360 data_time: 0.0024 memory: 44138 loss: 0.3388 +2023/06/04 23:39:01 - mmengine - INFO - Epoch(train) [8][3900/3907] lr: 1.8611e-05 eta: 1:22:43 time: 0.6328 data_time: 0.0013 memory: 44138 loss: 0.2871 +2023/06/04 23:39:05 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv1_lr1e-4_20230604_175638 +2023/06/04 23:39:05 - mmengine - INFO - Saving checkpoint at 8 epochs +2023/06/04 23:40:41 - mmengine - INFO - Epoch(val) [8][57/57] accuracy/top1: 89.2592 single-label/precision_classwise: [84.43675994873047, 97.99471282958984] single-label/recall_classwise: [98.70588684082031, 77.6589126586914] single-label/f1-score_classwise: [91.01545715332031, 86.649658203125] data_time: 0.0380 time: 1.2828 +2023/06/04 23:41:49 - mmengine - INFO - Epoch(train) [9][ 100/3907] lr: 1.8385e-05 eta: 1:21:36 time: 0.6345 data_time: 0.0028 memory: 44138 loss: 0.3252 +2023/06/04 23:42:52 - mmengine - INFO - Epoch(train) [9][ 200/3907] lr: 1.8176e-05 eta: 1:20:33 time: 0.6323 data_time: 0.0018 memory: 44138 loss: 0.3209 +2023/06/04 23:43:56 - mmengine - INFO - Epoch(train) [9][ 300/3907] lr: 1.7969e-05 eta: 1:19:29 time: 0.6351 data_time: 0.0019 memory: 44138 loss: 0.3222 +2023/06/04 23:44:59 - mmengine - INFO - Epoch(train) [9][ 400/3907] lr: 1.7765e-05 eta: 1:18:26 time: 0.6327 data_time: 0.0019 memory: 44138 loss: 0.2998 +2023/06/04 23:46:03 - mmengine - INFO - Epoch(train) [9][ 500/3907] lr: 1.7563e-05 eta: 1:17:22 time: 0.6332 data_time: 0.0019 memory: 44138 loss: 0.2856 +2023/06/04 23:47:06 - mmengine - INFO - Epoch(train) [9][ 600/3907] lr: 1.7363e-05 eta: 1:16:19 time: 0.6364 data_time: 0.0023 memory: 44138 loss: 0.3231 +2023/06/04 23:48:10 - mmengine - INFO - Epoch(train) [9][ 700/3907] lr: 1.7166e-05 eta: 1:15:15 time: 0.6363 data_time: 0.0019 memory: 44138 loss: 0.2942 +2023/06/04 23:48:38 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv1_lr1e-4_20230604_175638 +2023/06/04 23:49:13 - mmengine - INFO - Epoch(train) [9][ 800/3907] lr: 1.6971e-05 eta: 1:14:12 time: 0.6340 data_time: 0.0022 memory: 44138 loss: 0.3112 +2023/06/04 23:50:17 - mmengine - INFO - Epoch(train) [9][ 900/3907] lr: 1.6779e-05 eta: 1:13:08 time: 0.6361 data_time: 0.0018 memory: 44138 loss: 0.2856 +2023/06/04 23:51:20 - mmengine - INFO - Epoch(train) [9][1000/3907] lr: 1.6589e-05 eta: 1:12:05 time: 0.6337 data_time: 0.0017 memory: 44138 loss: 0.3068 +2023/06/04 23:52:24 - mmengine - INFO - Epoch(train) [9][1100/3907] lr: 1.6402e-05 eta: 1:11:01 time: 0.6351 data_time: 0.0019 memory: 44138 loss: 0.3107 +2023/06/04 23:53:27 - mmengine - INFO - Epoch(train) [9][1200/3907] lr: 1.6217e-05 eta: 1:09:58 time: 0.6328 data_time: 0.0019 memory: 44138 loss: 0.2819 +2023/06/04 23:54:30 - mmengine - INFO - Epoch(train) [9][1300/3907] lr: 1.6035e-05 eta: 1:08:54 time: 0.6330 data_time: 0.0016 memory: 44138 loss: 0.2919 +2023/06/04 23:55:34 - mmengine - INFO - Epoch(train) [9][1400/3907] lr: 1.5855e-05 eta: 1:07:51 time: 0.6385 data_time: 0.0043 memory: 44138 loss: 0.2940 +2023/06/04 23:56:37 - mmengine - INFO - Epoch(train) [9][1500/3907] lr: 1.5678e-05 eta: 1:06:47 time: 0.6336 data_time: 0.0018 memory: 44138 loss: 0.3216 +2023/06/04 23:57:41 - mmengine - INFO - Epoch(train) [9][1600/3907] lr: 1.5503e-05 eta: 1:05:44 time: 0.6341 data_time: 0.0016 memory: 44138 loss: 0.3360 +2023/06/04 23:58:44 - mmengine - INFO - Epoch(train) [9][1700/3907] lr: 1.5331e-05 eta: 1:04:40 time: 0.6370 data_time: 0.0035 memory: 44138 loss: 0.3019 +2023/06/04 23:59:12 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv1_lr1e-4_20230604_175638 +2023/06/04 23:59:48 - mmengine - INFO - Epoch(train) [9][1800/3907] lr: 1.5162e-05 eta: 1:03:37 time: 0.6329 data_time: 0.0021 memory: 44138 loss: 0.3038 +2023/06/05 00:00:51 - mmengine - INFO - Epoch(train) [9][1900/3907] lr: 1.4995e-05 eta: 1:02:33 time: 0.6366 data_time: 0.0024 memory: 44138 loss: 0.3400 +2023/06/05 00:01:55 - mmengine - INFO - Epoch(train) [9][2000/3907] lr: 1.4830e-05 eta: 1:01:30 time: 0.6355 data_time: 0.0028 memory: 44138 loss: 0.3225 +2023/06/05 00:02:58 - mmengine - INFO - Epoch(train) [9][2100/3907] lr: 1.4668e-05 eta: 1:00:27 time: 0.6335 data_time: 0.0017 memory: 44138 loss: 0.3017 +2023/06/05 00:04:02 - mmengine - INFO - Epoch(train) [9][2200/3907] lr: 1.4509e-05 eta: 0:59:23 time: 0.6333 data_time: 0.0017 memory: 44138 loss: 0.3399 +2023/06/05 00:05:05 - mmengine - INFO - Epoch(train) [9][2300/3907] lr: 1.4353e-05 eta: 0:58:20 time: 0.6344 data_time: 0.0021 memory: 44138 loss: 0.3109 +2023/06/05 00:06:08 - mmengine - INFO - Epoch(train) [9][2400/3907] lr: 1.4199e-05 eta: 0:57:16 time: 0.6326 data_time: 0.0018 memory: 44138 loss: 0.3100 +2023/06/05 00:07:12 - mmengine - INFO - Epoch(train) [9][2500/3907] lr: 1.4047e-05 eta: 0:56:13 time: 0.6346 data_time: 0.0023 memory: 44138 loss: 0.2934 +2023/06/05 00:08:15 - mmengine - INFO - Epoch(train) [9][2600/3907] lr: 1.3899e-05 eta: 0:55:09 time: 0.6446 data_time: 0.0019 memory: 44138 loss: 0.3188 +2023/06/05 00:09:19 - mmengine - INFO - Epoch(train) [9][2700/3907] lr: 1.3753e-05 eta: 0:54:06 time: 0.6368 data_time: 0.0032 memory: 44138 loss: 0.2890 +2023/06/05 00:09:47 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv1_lr1e-4_20230604_175638 +2023/06/05 00:10:22 - mmengine - INFO - Epoch(train) [9][2800/3907] lr: 1.3609e-05 eta: 0:53:02 time: 0.6367 data_time: 0.0019 memory: 44138 loss: 0.3111 +2023/06/05 00:11:26 - mmengine - INFO - Epoch(train) [9][2900/3907] lr: 1.3469e-05 eta: 0:51:59 time: 0.6327 data_time: 0.0017 memory: 44138 loss: 0.3019 +2023/06/05 00:12:29 - mmengine - INFO - Epoch(train) [9][3000/3907] lr: 1.3331e-05 eta: 0:50:55 time: 0.6369 data_time: 0.0024 memory: 44138 loss: 0.3261 +2023/06/05 00:13:33 - mmengine - INFO - Epoch(train) [9][3100/3907] lr: 1.3196e-05 eta: 0:49:52 time: 0.6336 data_time: 0.0019 memory: 44138 loss: 0.2914 +2023/06/05 00:14:36 - mmengine - INFO - Epoch(train) [9][3200/3907] lr: 1.3063e-05 eta: 0:48:48 time: 0.6353 data_time: 0.0039 memory: 44138 loss: 0.3216 +2023/06/05 00:15:39 - mmengine - INFO - Epoch(train) [9][3300/3907] lr: 1.2933e-05 eta: 0:47:45 time: 0.6340 data_time: 0.0017 memory: 44138 loss: 0.3003 +2023/06/05 00:16:43 - mmengine - INFO - Epoch(train) [9][3400/3907] lr: 1.2806e-05 eta: 0:46:41 time: 0.6329 data_time: 0.0018 memory: 44138 loss: 0.3098 +2023/06/05 00:17:46 - mmengine - INFO - Epoch(train) [9][3500/3907] lr: 1.2682e-05 eta: 0:45:38 time: 0.6388 data_time: 0.0016 memory: 44138 loss: 0.3278 +2023/06/05 00:18:50 - mmengine - INFO - Epoch(train) [9][3600/3907] lr: 1.2560e-05 eta: 0:44:34 time: 0.6349 data_time: 0.0035 memory: 44138 loss: 0.2706 +2023/06/05 00:19:53 - mmengine - INFO - Epoch(train) [9][3700/3907] lr: 1.2441e-05 eta: 0:43:31 time: 0.6350 data_time: 0.0022 memory: 44138 loss: 0.3136 +2023/06/05 00:20:21 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv1_lr1e-4_20230604_175638 +2023/06/05 00:20:57 - mmengine - INFO - Epoch(train) [9][3800/3907] lr: 1.2325e-05 eta: 0:42:27 time: 0.6354 data_time: 0.0020 memory: 44138 loss: 0.3295 +2023/06/05 00:22:00 - mmengine - INFO - Epoch(train) [9][3900/3907] lr: 1.2211e-05 eta: 0:41:24 time: 0.6328 data_time: 0.0014 memory: 44138 loss: 0.3240 +2023/06/05 00:22:04 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv1_lr1e-4_20230604_175638 +2023/06/05 00:22:04 - mmengine - INFO - Saving checkpoint at 9 epochs +2023/06/05 00:23:42 - mmengine - INFO - Epoch(val) [9][57/57] accuracy/top1: 88.6712 single-label/precision_classwise: [83.56895446777344, 98.23929595947266] single-label/recall_classwise: [98.88895416259766, 76.12403106689453] single-label/f1-score_classwise: [90.58578491210938, 85.7791748046875] data_time: 0.0370 time: 1.2817 +2023/06/05 00:24:50 - mmengine - INFO - Epoch(train) [10][ 100/3907] lr: 1.2093e-05 eta: 0:40:16 time: 0.6393 data_time: 0.0023 memory: 44138 loss: 0.3085 +2023/06/05 00:25:54 - mmengine - INFO - Epoch(train) [10][ 200/3907] lr: 1.1985e-05 eta: 0:39:13 time: 0.6375 data_time: 0.0022 memory: 44138 loss: 0.2885 +2023/06/05 00:26:58 - mmengine - INFO - Epoch(train) [10][ 300/3907] lr: 1.1881e-05 eta: 0:38:09 time: 0.6339 data_time: 0.0016 memory: 44138 loss: 0.2858 +2023/06/05 00:28:01 - mmengine - INFO - Epoch(train) [10][ 400/3907] lr: 1.1778e-05 eta: 0:37:06 time: 0.6359 data_time: 0.0018 memory: 44138 loss: 0.3285 +2023/06/05 00:29:05 - mmengine - INFO - Epoch(train) [10][ 500/3907] lr: 1.1679e-05 eta: 0:36:02 time: 0.6330 data_time: 0.0016 memory: 44138 loss: 0.3235 +2023/06/05 00:30:08 - mmengine - INFO - Epoch(train) [10][ 600/3907] lr: 1.1583e-05 eta: 0:34:59 time: 0.6355 data_time: 0.0025 memory: 44138 loss: 0.2979 +2023/06/05 00:31:12 - mmengine - INFO - Epoch(train) [10][ 700/3907] lr: 1.1489e-05 eta: 0:33:56 time: 0.6367 data_time: 0.0020 memory: 44138 loss: 0.2975 +2023/06/05 00:32:15 - mmengine - INFO - Epoch(train) [10][ 800/3907] lr: 1.1398e-05 eta: 0:32:52 time: 0.6340 data_time: 0.0019 memory: 44138 loss: 0.3005 +2023/06/05 00:32:39 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv1_lr1e-4_20230604_175638 +2023/06/05 00:33:19 - mmengine - INFO - Epoch(train) [10][ 900/3907] lr: 1.1310e-05 eta: 0:31:49 time: 0.6358 data_time: 0.0022 memory: 44138 loss: 0.3068 +2023/06/05 00:34:22 - mmengine - INFO - Epoch(train) [10][1000/3907] lr: 1.1225e-05 eta: 0:30:45 time: 0.6365 data_time: 0.0018 memory: 44138 loss: 0.2954 +2023/06/05 00:35:26 - mmengine - INFO - Epoch(train) [10][1100/3907] lr: 1.1142e-05 eta: 0:29:42 time: 0.6357 data_time: 0.0044 memory: 44138 loss: 0.3115 +2023/06/05 00:36:29 - mmengine - INFO - Epoch(train) [10][1200/3907] lr: 1.1063e-05 eta: 0:28:38 time: 0.6364 data_time: 0.0024 memory: 44138 loss: 0.3064 +2023/06/05 00:37:33 - mmengine - INFO - Epoch(train) [10][1300/3907] lr: 1.0986e-05 eta: 0:27:35 time: 0.6336 data_time: 0.0021 memory: 44138 loss: 0.3352 +2023/06/05 00:38:36 - mmengine - INFO - Epoch(train) [10][1400/3907] lr: 1.0912e-05 eta: 0:26:31 time: 0.6342 data_time: 0.0021 memory: 44138 loss: 0.3142 +2023/06/05 00:39:40 - mmengine - INFO - Epoch(train) [10][1500/3907] lr: 1.0841e-05 eta: 0:25:28 time: 0.6330 data_time: 0.0016 memory: 44138 loss: 0.3255 +2023/06/05 00:40:44 - mmengine - INFO - Epoch(train) [10][1600/3907] lr: 1.0773e-05 eta: 0:24:24 time: 0.6346 data_time: 0.0018 memory: 44138 loss: 0.3244 +2023/06/05 00:41:47 - mmengine - INFO - Epoch(train) [10][1700/3907] lr: 1.0707e-05 eta: 0:23:21 time: 0.6334 data_time: 0.0018 memory: 44138 loss: 0.3018 +2023/06/05 00:42:51 - mmengine - INFO - Epoch(train) [10][1800/3907] lr: 1.0645e-05 eta: 0:22:17 time: 0.6362 data_time: 0.0018 memory: 44138 loss: 0.3243 +2023/06/05 00:43:14 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv1_lr1e-4_20230604_175638 +2023/06/05 00:43:54 - mmengine - INFO - Epoch(train) [10][1900/3907] lr: 1.0585e-05 eta: 0:21:14 time: 0.6328 data_time: 0.0017 memory: 44138 loss: 0.3089 +2023/06/05 00:44:58 - mmengine - INFO - Epoch(train) [10][2000/3907] lr: 1.0529e-05 eta: 0:20:10 time: 0.6365 data_time: 0.0024 memory: 44138 loss: 0.3017 +2023/06/05 00:46:01 - mmengine - INFO - Epoch(train) [10][2100/3907] lr: 1.0475e-05 eta: 0:19:07 time: 0.6336 data_time: 0.0018 memory: 44138 loss: 0.3050 +2023/06/05 00:47:05 - mmengine - INFO - Epoch(train) [10][2200/3907] lr: 1.0424e-05 eta: 0:18:03 time: 0.6358 data_time: 0.0031 memory: 44138 loss: 0.2934 +2023/06/05 00:48:09 - mmengine - INFO - Epoch(train) [10][2300/3907] lr: 1.0376e-05 eta: 0:17:00 time: 0.6386 data_time: 0.0043 memory: 44138 loss: 0.3326 +2023/06/05 00:49:13 - mmengine - INFO - Epoch(train) [10][2400/3907] lr: 1.0330e-05 eta: 0:15:56 time: 0.6385 data_time: 0.0029 memory: 44138 loss: 0.3196 +2023/06/05 00:50:17 - mmengine - INFO - Epoch(train) [10][2500/3907] lr: 1.0288e-05 eta: 0:14:53 time: 0.6413 data_time: 0.0053 memory: 44138 loss: 0.3192 +2023/06/05 00:51:21 - mmengine - INFO - Epoch(train) [10][2600/3907] lr: 1.0249e-05 eta: 0:13:49 time: 0.6393 data_time: 0.0031 memory: 44138 loss: 0.3143 +2023/06/05 00:52:25 - mmengine - INFO - Epoch(train) [10][2700/3907] lr: 1.0212e-05 eta: 0:12:46 time: 0.6412 data_time: 0.0040 memory: 44138 loss: 0.2617 +2023/06/05 00:53:30 - mmengine - INFO - Epoch(train) [10][2800/3907] lr: 1.0178e-05 eta: 0:11:42 time: 0.6819 data_time: 0.0046 memory: 44138 loss: 0.2991 +2023/06/05 00:53:55 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv1_lr1e-4_20230604_175638 +2023/06/05 00:54:35 - mmengine - INFO - Epoch(train) [10][2900/3907] lr: 1.0148e-05 eta: 0:10:39 time: 0.6426 data_time: 0.0078 memory: 44138 loss: 0.3164 +2023/06/05 00:55:39 - mmengine - INFO - Epoch(train) [10][3000/3907] lr: 1.0120e-05 eta: 0:09:36 time: 0.6328 data_time: 0.0013 memory: 44138 loss: 0.3315 +2023/06/05 00:56:43 - mmengine - INFO - Epoch(train) [10][3100/3907] lr: 1.0095e-05 eta: 0:08:32 time: 0.6323 data_time: 0.0016 memory: 44138 loss: 0.3088 +2023/06/05 00:57:47 - mmengine - INFO - Epoch(train) [10][3200/3907] lr: 1.0073e-05 eta: 0:07:29 time: 0.6596 data_time: 0.0071 memory: 44138 loss: 0.2986 +2023/06/05 00:58:51 - mmengine - INFO - Epoch(train) [10][3300/3907] lr: 1.0054e-05 eta: 0:06:25 time: 0.6320 data_time: 0.0015 memory: 44138 loss: 0.3012 +2023/06/05 00:59:56 - mmengine - INFO - Epoch(train) [10][3400/3907] lr: 1.0038e-05 eta: 0:05:22 time: 0.6363 data_time: 0.0038 memory: 44138 loss: 0.3454 +2023/06/05 01:01:00 - mmengine - INFO - Epoch(train) [10][3500/3907] lr: 1.0024e-05 eta: 0:04:18 time: 0.6622 data_time: 0.0143 memory: 44138 loss: 0.2821 +2023/06/05 01:02:04 - mmengine - INFO - Epoch(train) [10][3600/3907] lr: 1.0014e-05 eta: 0:03:14 time: 0.6328 data_time: 0.0015 memory: 44138 loss: 0.2955 +2023/06/05 01:03:09 - mmengine - INFO - Epoch(train) [10][3700/3907] lr: 1.0006e-05 eta: 0:02:11 time: 0.6371 data_time: 0.0023 memory: 44138 loss: 0.2789 +2023/06/05 01:04:13 - mmengine - INFO - Epoch(train) [10][3800/3907] lr: 1.0002e-05 eta: 0:01:07 time: 0.6573 data_time: 0.0084 memory: 44138 loss: 0.3349 +2023/06/05 01:04:37 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv1_lr1e-4_20230604_175638 +2023/06/05 01:05:18 - mmengine - INFO - Epoch(train) [10][3900/3907] lr: 1.0000e-05 eta: 0:00:04 time: 0.6481 data_time: 0.0033 memory: 44138 loss: 0.3095 +2023/06/05 01:05:22 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv1_lr1e-4_20230604_175638 +2023/06/05 01:05:22 - mmengine - INFO - Saving checkpoint at 10 epochs +2023/06/05 01:07:01 - mmengine - INFO - Epoch(val) [10][57/57] accuracy/top1: 88.5982 single-label/precision_classwise: [83.33332824707031, 98.62557220458984] single-label/recall_classwise: [99.14146423339844, 75.65116119384766] single-label/f1-score_classwise: [90.55265045166016, 85.62403869628906] data_time: 0.1109 time: 1.3538 diff --git a/clip_large_pretrain_4x256_sdv1_lr1e-4_wopostnorm_wneck/20230604_175638/vis_data/20230604_175638.json b/clip_large_pretrain_4x256_sdv1_lr1e-4_wopostnorm_wneck/20230604_175638/vis_data/20230604_175638.json new file mode 100644 index 0000000000000000000000000000000000000000..fde6101e2d597abe1882751764205750c8b3daa9 --- /dev/null +++ b/clip_large_pretrain_4x256_sdv1_lr1e-4_wopostnorm_wneck/20230604_175638/vis_data/20230604_175638.json @@ -0,0 +1,400 @@ +{"lr": 9.999857418382631e-05, "data_time": 0.0026583194732666014, "loss": 0.5272099018096924, "time": 0.6304151773452759, "epoch": 1, "memory": 44139, "step": 100} +{"lr": 9.999423907359506e-05, "data_time": 0.0019651174545288084, "loss": 0.4700825929641724, "time": 0.6316981077194214, "epoch": 1, "memory": 44139, "step": 200} +{"lr": 9.998699480412073e-05, "data_time": 0.001915311813354492, "loss": 0.44749481678009034, "time": 0.6328604698181153, "epoch": 1, "memory": 44139, "step": 300} +{"lr": 9.997684184379085e-05, "data_time": 0.0018089532852172852, "loss": 0.45088441371917726, "time": 0.6322338819503784, "epoch": 1, "memory": 44139, "step": 400} +{"lr": 9.996378084905784e-05, "data_time": 0.0017721176147460938, "loss": 0.41908371448516846, "time": 0.6319724798202515, "epoch": 1, "memory": 44139, "step": 500} +{"lr": 9.994781266439667e-05, "data_time": 0.0020272254943847655, "loss": 0.4147952705621719, "time": 0.6332255363464355, "epoch": 1, "memory": 44139, "step": 600} +{"lr": 9.992893832225061e-05, "data_time": 0.001917409896850586, "loss": 0.3993018567562103, "time": 0.6328696489334107, "epoch": 1, "memory": 44139, "step": 700} +{"lr": 9.990715904296405e-05, "data_time": 0.002123117446899414, "loss": 0.41613166928291323, "time": 0.6325438499450684, "epoch": 1, "memory": 44139, "step": 800} +{"lr": 9.988247623470371e-05, "data_time": 0.0018011569976806641, "loss": 0.395975923538208, "time": 0.6324893712997437, "epoch": 1, "memory": 44139, "step": 900} +{"lr": 9.985489149336749e-05, "data_time": 0.002089095115661621, "loss": 0.4181334525346756, "time": 0.6309988260269165, "epoch": 1, "memory": 44139, "step": 1000} +{"lr": 9.982440660248187e-05, "data_time": 0.0016204833984375, "loss": 0.4181372046470642, "time": 0.6321151494979859, "epoch": 1, "memory": 44139, "step": 1100} +{"lr": 9.979102353308574e-05, "data_time": 0.0017192602157592774, "loss": 0.41433699131011964, "time": 0.632967758178711, "epoch": 1, "memory": 44139, "step": 1200} +{"lr": 9.975474444360358e-05, "data_time": 0.0017733097076416016, "loss": 0.40109002590179443, "time": 0.6327246904373169, "epoch": 1, "memory": 44139, "step": 1300} +{"lr": 9.97155716797056e-05, "data_time": 0.0017506837844848632, "loss": 0.4026054620742798, "time": 0.6326537609100342, "epoch": 1, "memory": 44139, "step": 1400} +{"lr": 9.967350777415632e-05, "data_time": 0.0016286849975585937, "loss": 0.3907728523015976, "time": 0.632654333114624, "epoch": 1, "memory": 44139, "step": 1500} +{"lr": 9.962855544665059e-05, "data_time": 0.0017454147338867188, "loss": 0.3846167206764221, "time": 0.6324703931808472, "epoch": 1, "memory": 44139, "step": 1600} +{"lr": 9.958071760363773e-05, "data_time": 0.0016985654830932618, "loss": 0.40266107618808744, "time": 0.632651424407959, "epoch": 1, "memory": 44139, "step": 1700} +{"lr": 9.952999733813382e-05, "data_time": 0.0018072843551635743, "loss": 0.373909655213356, "time": 0.6331640005111694, "epoch": 1, "memory": 44139, "step": 1800} +{"lr": 9.947639792952148e-05, "data_time": 0.0017575740814208985, "loss": 0.363760969042778, "time": 0.6328064918518066, "epoch": 1, "memory": 44139, "step": 1900} +{"lr": 9.941992284333813e-05, "data_time": 0.0018268108367919921, "loss": 0.39119345545768736, "time": 0.6327016830444336, "epoch": 1, "memory": 44139, "step": 2000} +{"lr": 9.93605757310516e-05, "data_time": 0.001676321029663086, "loss": 0.40079807937145234, "time": 0.6328168153762818, "epoch": 1, "memory": 44139, "step": 2100} +{"lr": 9.929836042982421e-05, "data_time": 0.0016045808792114259, "loss": 0.3407004803419113, "time": 0.6326619863510132, "epoch": 1, "memory": 44139, "step": 2200} +{"lr": 9.923328096226467e-05, "data_time": 0.0016084671020507812, "loss": 0.3782178729772568, "time": 0.6325747966766357, "epoch": 1, "memory": 44139, "step": 2300} +{"lr": 9.916534153616819e-05, "data_time": 0.0016426801681518554, "loss": 0.36378566920757294, "time": 0.6332096576690673, "epoch": 1, "memory": 44139, "step": 2400} +{"lr": 9.909454654424387e-05, "data_time": 0.0021802425384521485, "loss": 0.3504793167114258, "time": 0.6329232931137085, "epoch": 1, "memory": 44139, "step": 2500} +{"lr": 9.902090056383109e-05, "data_time": 0.0019495248794555663, "loss": 0.3691946417093277, "time": 0.6324845552444458, "epoch": 1, "memory": 44139, "step": 2600} +{"lr": 9.894440835660365e-05, "data_time": 0.0015739679336547851, "loss": 0.38485833406448366, "time": 0.6329338550567627, "epoch": 1, "memory": 44139, "step": 2700} +{"lr": 9.886507486826162e-05, "data_time": 0.0018564701080322266, "loss": 0.3473310351371765, "time": 0.63320894241333, "epoch": 1, "memory": 44139, "step": 2800} +{"lr": 9.878290522821154e-05, "data_time": 0.0019229650497436523, "loss": 0.3705559939146042, "time": 0.6331067085266113, "epoch": 1, "memory": 44139, "step": 2900} +{"lr": 9.86979047492352e-05, "data_time": 0.0017976760864257812, "loss": 0.34593615531921384, "time": 0.6325294494628906, "epoch": 1, "memory": 44139, "step": 3000} +{"lr": 9.861007892714578e-05, "data_time": 0.0017056226730346679, "loss": 0.3622700244188309, "time": 0.6328414678573608, "epoch": 1, "memory": 44139, "step": 3100} +{"lr": 9.851943344043239e-05, "data_time": 0.002273368835449219, "loss": 0.3563511461019516, "time": 0.6327307224273682, "epoch": 1, "memory": 44139, "step": 3200} +{"lr": 9.842597414989319e-05, "data_time": 0.0021600246429443358, "loss": 0.3420926809310913, "time": 0.6363249301910401, "epoch": 1, "memory": 44139, "step": 3300} +{"lr": 9.832970709825637e-05, "data_time": 0.0023040771484375, "loss": 0.3658769130706787, "time": 0.6323584318161011, "epoch": 1, "memory": 44139, "step": 3400} +{"lr": 9.823063850978968e-05, "data_time": 0.0016047954559326172, "loss": 0.35051974952220916, "time": 0.6317821979522705, "epoch": 1, "memory": 44139, "step": 3500} +{"lr": 9.81287747898975e-05, "data_time": 0.0018771171569824218, "loss": 0.36420016884803774, "time": 0.6314892530441284, "epoch": 1, "memory": 44139, "step": 3600} +{"lr": 9.802412252470715e-05, "data_time": 0.0019242286682128907, "loss": 0.32584536969661715, "time": 0.6307554960250854, "epoch": 1, "memory": 44139, "step": 3700} +{"lr": 9.791668848064258e-05, "data_time": 0.0017160415649414063, "loss": 0.3360456913709641, "time": 0.6322226762771607, "epoch": 1, "memory": 44139, "step": 3800} +{"lr": 9.780647960398751e-05, "data_time": 0.0012096881866455079, "loss": 0.3510953962802887, "time": 0.630826735496521, "epoch": 1, "memory": 44139, "step": 3900} +{"accuracy/top1": 82.23096466064453, "data_time": 0.04385408602262798, "time": 1.3381989044055604, "step": 1} +{"lr": 9.7685491196646e-05, "data_time": 0.0017798423767089843, "loss": 0.35121058523654936, "time": 0.6321818828582764, "epoch": 2, "memory": 44139, "step": 4007} +{"lr": 9.756956126085676e-05, "data_time": 0.0015968561172485351, "loss": 0.3366678595542908, "time": 0.6321999311447144, "epoch": 2, "memory": 44138, "step": 4107} +{"lr": 9.745087893642482e-05, "data_time": 0.0016131639480590821, "loss": 0.34111056923866273, "time": 0.6321089267730713, "epoch": 2, "memory": 44138, "step": 4207} +{"lr": 9.73294518969056e-05, "data_time": 0.0018948793411254882, "loss": 0.39238911867141724, "time": 0.6330914735794068, "epoch": 2, "memory": 44138, "step": 4307} +{"lr": 9.72052879933172e-05, "data_time": 0.0014326333999633788, "loss": 0.3141148179769516, "time": 0.6321524620056153, "epoch": 2, "memory": 44138, "step": 4407} +{"lr": 9.707839525363358e-05, "data_time": 0.0015880346298217773, "loss": 0.35293120741844175, "time": 0.6328054189682006, "epoch": 2, "memory": 44138, "step": 4507} +{"lr": 9.69487818822646e-05, "data_time": 0.0014887094497680665, "loss": 0.3299088656902313, "time": 0.6326074361801147, "epoch": 2, "memory": 44138, "step": 4607} +{"lr": 9.681645625952616e-05, "data_time": 0.0015972614288330077, "loss": 0.34193129539489747, "time": 0.6329273462295533, "epoch": 2, "memory": 44138, "step": 4707} +{"lr": 9.668142694109826e-05, "data_time": 0.0014832019805908203, "loss": 0.3641922563314438, "time": 0.6330604791641236, "epoch": 2, "memory": 44138, "step": 4807} +{"lr": 9.654370265747162e-05, "data_time": 0.0018642425537109375, "loss": 0.32959853410720824, "time": 0.63274245262146, "epoch": 2, "memory": 44138, "step": 4907} +{"lr": 9.640329231338345e-05, "data_time": 0.0019052267074584962, "loss": 0.35813969671726226, "time": 0.6320257902145385, "epoch": 2, "memory": 44138, "step": 5007} +{"lr": 9.626020498724147e-05, "data_time": 0.0017485618591308594, "loss": 0.3302235662937164, "time": 0.632898998260498, "epoch": 2, "memory": 44138, "step": 5107} +{"lr": 9.611444993053717e-05, "data_time": 0.0018631458282470704, "loss": 0.3023091822862625, "time": 0.6421539783477783, "epoch": 2, "memory": 44138, "step": 5207} +{"lr": 9.596603656724746e-05, "data_time": 0.0016747236251831055, "loss": 0.31570457220077514, "time": 0.6327734231948853, "epoch": 2, "memory": 44138, "step": 5307} +{"lr": 9.58149744932254e-05, "data_time": 0.0016315460205078125, "loss": 0.3265344202518463, "time": 0.6319121599197388, "epoch": 2, "memory": 44138, "step": 5407} +{"lr": 9.566127347557972e-05, "data_time": 0.001626873016357422, "loss": 0.32270238995552064, "time": 0.6323603630065918, "epoch": 2, "memory": 44138, "step": 5507} +{"lr": 9.550494345204367e-05, "data_time": 0.0017716407775878907, "loss": 0.3385892495512962, "time": 0.632108497619629, "epoch": 2, "memory": 44138, "step": 5607} +{"lr": 9.534599453033183e-05, "data_time": 0.0013827800750732422, "loss": 0.33297447562217714, "time": 0.6321195363998413, "epoch": 2, "memory": 44138, "step": 5707} +{"lr": 9.518443698748712e-05, "data_time": 0.0019120216369628907, "loss": 0.2947722613811493, "time": 0.6323132514953613, "epoch": 2, "memory": 44138, "step": 5807} +{"lr": 9.50202812692162e-05, "data_time": 0.001697397232055664, "loss": 0.340665927529335, "time": 0.6326845645904541, "epoch": 2, "memory": 44138, "step": 5907} +{"lr": 9.485353798921386e-05, "data_time": 0.0018517732620239257, "loss": 0.3081132501363754, "time": 0.6327033519744873, "epoch": 2, "memory": 44138, "step": 6007} +{"lr": 9.468421792847712e-05, "data_time": 0.001509237289428711, "loss": 0.34085512459278106, "time": 0.6335102796554566, "epoch": 2, "memory": 44138, "step": 6107} +{"lr": 9.451233203460794e-05, "data_time": 0.016501879692077635, "loss": 0.33548237979412077, "time": 0.6462246656417847, "epoch": 2, "memory": 44138, "step": 6207} +{"lr": 9.433789142110532e-05, "data_time": 0.0015334367752075195, "loss": 0.3299319908022881, "time": 0.6456239700317383, "epoch": 2, "memory": 44138, "step": 6307} +{"lr": 9.416090736664709e-05, "data_time": 0.0016468048095703125, "loss": 0.33380793035030365, "time": 0.6327488899230957, "epoch": 2, "memory": 44138, "step": 6407} +{"lr": 9.398139131436032e-05, "data_time": 0.0018025875091552735, "loss": 0.35275731682777406, "time": 0.6320118188858033, "epoch": 2, "memory": 44138, "step": 6507} +{"lr": 9.379935487108167e-05, "data_time": 0.0017237186431884766, "loss": 0.3359238713979721, "time": 0.6313405752182006, "epoch": 2, "memory": 44138, "step": 6607} +{"lr": 9.361480980660697e-05, "data_time": 0.0015587091445922851, "loss": 0.32569957673549654, "time": 0.6312980651855469, "epoch": 2, "memory": 44138, "step": 6707} +{"lr": 9.342776805292993e-05, "data_time": 0.0017858028411865234, "loss": 0.35612158477306366, "time": 0.6311082124710083, "epoch": 2, "memory": 44138, "step": 6807} +{"lr": 9.323824170347083e-05, "data_time": 0.0015132665634155274, "loss": 0.340649750828743, "time": 0.6312653541564941, "epoch": 2, "memory": 44138, "step": 6907} +{"lr": 9.304624301229471e-05, "data_time": 0.0019707679748535156, "loss": 0.31895554065704346, "time": 0.6313549995422363, "epoch": 2, "memory": 44138, "step": 7007} +{"lr": 9.285178439331897e-05, "data_time": 0.0018143415451049804, "loss": 0.3082291066646576, "time": 0.6324036836624145, "epoch": 2, "memory": 44138, "step": 7107} +{"lr": 9.265487841951087e-05, "data_time": 0.0020533084869384767, "loss": 0.32104428559541703, "time": 0.6320973634719849, "epoch": 2, "memory": 44138, "step": 7207} +{"lr": 9.245553782207421e-05, "data_time": 0.0014142751693725585, "loss": 0.3304094195365906, "time": 0.6371475219726562, "epoch": 2, "memory": 44138, "step": 7307} +{"lr": 9.225377548962671e-05, "data_time": 0.0014187097549438477, "loss": 0.3069656670093536, "time": 0.6318143367767334, "epoch": 2, "memory": 44138, "step": 7407} +{"lr": 9.204960446736601e-05, "data_time": 0.001721811294555664, "loss": 0.31440397799015046, "time": 0.6314618825912476, "epoch": 2, "memory": 44138, "step": 7507} +{"lr": 9.184303795622715e-05, "data_time": 0.002058076858520508, "loss": 0.3164314031600952, "time": 0.6315144062042236, "epoch": 2, "memory": 44138, "step": 7607} +{"lr": 9.163408931202823e-05, "data_time": 0.0019001722335815429, "loss": 0.316406786441803, "time": 0.6327008724212646, "epoch": 2, "memory": 44138, "step": 7707} +{"lr": 9.142277204460729e-05, "data_time": 0.001949620246887207, "loss": 0.30633518695831297, "time": 0.6310815572738647, "epoch": 2, "memory": 44138, "step": 7807} +{"accuracy/top1": 86.54535675048828, "data_time": 0.03482668975303913, "time": 1.2794994937962498, "step": 2} +{"lr": 9.119405492383288e-05, "data_time": 0.0014729261398315429, "loss": 0.3150415584445, "time": 0.631822919845581, "epoch": 3, "memory": 44138, "step": 7914} +{"lr": 9.097787819233205e-05, "data_time": 0.0016135931015014648, "loss": 0.3322458863258362, "time": 0.6319055080413818, "epoch": 3, "memory": 44138, "step": 8014} +{"lr": 9.075937526576627e-05, "data_time": 0.002053642272949219, "loss": 0.3029593124985695, "time": 0.6327821731567382, "epoch": 3, "memory": 44138, "step": 8114} +{"lr": 9.053856027171755e-05, "data_time": 0.0020038366317749025, "loss": 0.3186375737190247, "time": 0.6329057216644287, "epoch": 3, "memory": 44138, "step": 8214} +{"lr": 9.031544748725785e-05, "data_time": 0.0017593622207641602, "loss": 0.3061673820018768, "time": 0.6321552276611329, "epoch": 3, "memory": 44138, "step": 8314} +{"lr": 9.009005133802542e-05, "data_time": 0.0015958547592163086, "loss": 0.28570996820926664, "time": 0.6326016187667847, "epoch": 3, "memory": 44138, "step": 8414} +{"lr": 8.986238639729242e-05, "data_time": 0.0017288923263549805, "loss": 0.3147992193698883, "time": 0.6335033178329468, "epoch": 3, "memory": 44138, "step": 8514} +{"lr": 8.963246738502254e-05, "data_time": 0.001550769805908203, "loss": 0.3290190815925598, "time": 0.6319665431976318, "epoch": 3, "memory": 44138, "step": 8614} +{"lr": 8.940030916691928e-05, "data_time": 0.0015265941619873047, "loss": 0.30940591990947724, "time": 0.6312623262405396, "epoch": 3, "memory": 44138, "step": 8714} +{"lr": 8.91659267534649e-05, "data_time": 0.0015799045562744141, "loss": 0.32590238749980927, "time": 0.6314814805984497, "epoch": 3, "memory": 44138, "step": 8814} +{"lr": 8.892933529894978e-05, "data_time": 0.0014423370361328126, "loss": 0.324916610121727, "time": 0.6310937404632568, "epoch": 3, "memory": 44138, "step": 8914} +{"lr": 8.869055010049283e-05, "data_time": 0.00158236026763916, "loss": 0.3465823620557785, "time": 0.6313714742660522, "epoch": 3, "memory": 44138, "step": 9014} +{"lr": 8.844958659705188e-05, "data_time": 0.001692485809326172, "loss": 0.33284681141376493, "time": 0.6314581394195556, "epoch": 3, "memory": 44138, "step": 9114} +{"lr": 8.82064603684259e-05, "data_time": 0.0017244815826416016, "loss": 0.30471644550561905, "time": 0.6429675579071045, "epoch": 3, "memory": 44138, "step": 9214} +{"lr": 8.796118713424776e-05, "data_time": 0.0017464876174926758, "loss": 0.3103995889425278, "time": 0.632632565498352, "epoch": 3, "memory": 44138, "step": 9314} +{"lr": 8.77137827529676e-05, "data_time": 0.0019337654113769532, "loss": 0.3151204764842987, "time": 0.631326961517334, "epoch": 3, "memory": 44138, "step": 9414} +{"lr": 8.746426322082745e-05, "data_time": 0.0015658140182495117, "loss": 0.3046370282769203, "time": 0.6331208944320679, "epoch": 3, "memory": 44138, "step": 9514} +{"lr": 8.721264467082732e-05, "data_time": 0.0018798351287841798, "loss": 0.29835492074489595, "time": 0.6328758716583252, "epoch": 3, "memory": 44138, "step": 9614} +{"lr": 8.695894337168168e-05, "data_time": 0.0016434669494628906, "loss": 0.3285602807998657, "time": 0.6328323125839234, "epoch": 3, "memory": 44138, "step": 9714} +{"lr": 8.67031757267677e-05, "data_time": 0.0016220808029174805, "loss": 0.28751807659864426, "time": 0.6325338363647461, "epoch": 3, "memory": 44138, "step": 9814} +{"lr": 8.64453582730651e-05, "data_time": 0.0017338991165161133, "loss": 0.3135116085410118, "time": 0.6331475973129272, "epoch": 3, "memory": 44138, "step": 9914} +{"lr": 8.61855076800861e-05, "data_time": 0.001980495452880859, "loss": 0.3006910443305969, "time": 0.6326284408569336, "epoch": 3, "memory": 44138, "step": 10014} +{"lr": 8.592364074879843e-05, "data_time": 0.0018828630447387696, "loss": 0.29880787134170533, "time": 0.6326159238815308, "epoch": 3, "memory": 44138, "step": 10114} +{"lr": 8.565977441053854e-05, "data_time": 0.001508331298828125, "loss": 0.3597894847393036, "time": 0.6325548648834228, "epoch": 3, "memory": 44138, "step": 10214} +{"lr": 8.539392572591724e-05, "data_time": 0.0016089677810668945, "loss": 0.32104514837265014, "time": 0.6325655221939087, "epoch": 3, "memory": 44138, "step": 10314} +{"lr": 8.512611188371633e-05, "data_time": 0.0025423288345336912, "loss": 0.30717981457710264, "time": 0.6336353302001954, "epoch": 3, "memory": 44138, "step": 10414} +{"lr": 8.485635019977725e-05, "data_time": 0.0017320871353149413, "loss": 0.31498719900846484, "time": 0.6323895692825318, "epoch": 3, "memory": 44138, "step": 10514} +{"lr": 8.45846581158818e-05, "data_time": 0.0016342639923095704, "loss": 0.312097804248333, "time": 0.6329158306121826, "epoch": 3, "memory": 44138, "step": 10614} +{"lr": 8.431105319862416e-05, "data_time": 0.0016703605651855469, "loss": 0.35519146025180814, "time": 0.6327781677246094, "epoch": 3, "memory": 44138, "step": 10714} +{"lr": 8.403555313827516e-05, "data_time": 0.001896238327026367, "loss": 0.3117992579936981, "time": 0.6336266040802002, "epoch": 3, "memory": 44138, "step": 10814} +{"lr": 8.375817574763844e-05, "data_time": 0.002271437644958496, "loss": 0.29683640897274016, "time": 0.6317068338394165, "epoch": 3, "memory": 44138, "step": 10914} +{"lr": 8.347893896089871e-05, "data_time": 0.0015553951263427735, "loss": 0.34331596791744234, "time": 0.6320744514465332, "epoch": 3, "memory": 44138, "step": 11014} +{"lr": 8.31978608324625e-05, "data_time": 0.0014697074890136718, "loss": 0.3054161101579666, "time": 0.632155179977417, "epoch": 3, "memory": 44138, "step": 11114} +{"lr": 8.291495953579052e-05, "data_time": 0.0022779464721679687, "loss": 0.3382767826318741, "time": 0.6319034099578857, "epoch": 3, "memory": 44138, "step": 11214} +{"lr": 8.263025336222267e-05, "data_time": 0.001981091499328613, "loss": 0.2996159911155701, "time": 0.631929588317871, "epoch": 3, "memory": 44138, "step": 11314} +{"lr": 8.234376071979551e-05, "data_time": 0.0016178607940673829, "loss": 0.3139187753200531, "time": 0.6318180322647095, "epoch": 3, "memory": 44138, "step": 11414} +{"lr": 8.205550013205187e-05, "data_time": 0.0015609025955200194, "loss": 0.29609236866235733, "time": 0.6319526433944702, "epoch": 3, "memory": 44138, "step": 11514} +{"lr": 8.17654902368435e-05, "data_time": 0.0014995813369750976, "loss": 0.3452329754829407, "time": 0.6313196420669556, "epoch": 3, "memory": 44138, "step": 11614} +{"lr": 8.147374978512582e-05, "data_time": 0.0012678861618041991, "loss": 0.33042644560337064, "time": 0.6313238859176635, "epoch": 3, "memory": 44138, "step": 11714} +{"accuracy/top1": 88.37201690673828, "data_time": 0.03216570410235175, "time": 1.2769481519172932, "step": 3} +{"lr": 8.115969237626578e-05, "data_time": 0.0015007972717285157, "loss": 0.29631754010915756, "time": 0.6331702947616578, "epoch": 4, "memory": 44138, "step": 11821} +{"lr": 8.08644297337245e-05, "data_time": 0.0013971567153930665, "loss": 0.329595622420311, "time": 0.6327028512954712, "epoch": 4, "memory": 44138, "step": 11921} +{"lr": 8.056749479387651e-05, "data_time": 0.0015867948532104492, "loss": 0.3167020112276077, "time": 0.6333897590637207, "epoch": 4, "memory": 44138, "step": 12021} +{"lr": 8.026890675542457e-05, "data_time": 0.0014649629592895508, "loss": 0.31118630319833757, "time": 0.632989764213562, "epoch": 4, "memory": 44138, "step": 12121} +{"lr": 7.996868492395457e-05, "data_time": 0.0014965295791625976, "loss": 0.3052448958158493, "time": 0.6328909397125244, "epoch": 4, "memory": 44138, "step": 12221} +{"lr": 7.966684871068762e-05, "data_time": 0.0013724565505981445, "loss": 0.3271132081747055, "time": 0.6324964046478272, "epoch": 4, "memory": 44138, "step": 12321} +{"lr": 7.936341763122449e-05, "data_time": 0.0020560979843139648, "loss": 0.3197891443967819, "time": 0.6329471826553345, "epoch": 4, "memory": 44138, "step": 12421} +{"lr": 7.905841130428405e-05, "data_time": 0.0015125751495361328, "loss": 0.32204986810684205, "time": 0.6332590103149414, "epoch": 4, "memory": 44138, "step": 12521} +{"lr": 7.875184945043487e-05, "data_time": 0.0012707233428955079, "loss": 0.3328577935695648, "time": 0.6322899103164673, "epoch": 4, "memory": 44138, "step": 12621} +{"lr": 7.844375189082012e-05, "data_time": 0.0017799615859985351, "loss": 0.32825671434402465, "time": 0.6328401565551758, "epoch": 4, "memory": 44138, "step": 12721} +{"lr": 7.813413854587572e-05, "data_time": 0.0017122507095336914, "loss": 0.29628912061452867, "time": 0.635224175453186, "epoch": 4, "memory": 44138, "step": 12821} +{"lr": 7.782302943404291e-05, "data_time": 0.0017526865005493164, "loss": 0.28152785897254945, "time": 0.6331571340560913, "epoch": 4, "memory": 44138, "step": 12921} +{"lr": 7.751044467047337e-05, "data_time": 0.0015371322631835937, "loss": 0.2982440978288651, "time": 0.6328576803207397, "epoch": 4, "memory": 44138, "step": 13021} +{"lr": 7.719640446572907e-05, "data_time": 0.002108931541442871, "loss": 0.31789767146110537, "time": 0.6334184408187866, "epoch": 4, "memory": 44138, "step": 13121} +{"lr": 7.688092912447536e-05, "data_time": 0.0018030166625976562, "loss": 0.3146006315946579, "time": 0.6329826354980469, "epoch": 4, "memory": 44138, "step": 13221} +{"lr": 7.656403904416794e-05, "data_time": 0.001789402961730957, "loss": 0.31028204709291457, "time": 0.6333381652832031, "epoch": 4, "memory": 44138, "step": 13321} +{"lr": 7.624575471373434e-05, "data_time": 0.0021080255508422853, "loss": 0.31789949238300325, "time": 0.6335967540740967, "epoch": 4, "memory": 44138, "step": 13421} +{"lr": 7.592609671224926e-05, "data_time": 0.0016337394714355468, "loss": 0.3074886202812195, "time": 0.6347370624542237, "epoch": 4, "memory": 44138, "step": 13521} +{"lr": 7.560508570760375e-05, "data_time": 0.0015054464340209961, "loss": 0.30995132625102995, "time": 0.6344563722610473, "epoch": 4, "memory": 44138, "step": 13621} +{"lr": 7.528274245516895e-05, "data_time": 0.0017914056777954101, "loss": 0.32072039842605593, "time": 0.633221435546875, "epoch": 4, "memory": 44138, "step": 13721} +{"lr": 7.495908779645405e-05, "data_time": 0.0021927595138549806, "loss": 0.30968077480793, "time": 0.6332769155502319, "epoch": 4, "memory": 44138, "step": 13821} +{"lr": 7.463414265775906e-05, "data_time": 0.002105975151062012, "loss": 0.31177731305360795, "time": 0.6337838411331177, "epoch": 4, "memory": 44138, "step": 13921} +{"lr": 7.430792804882147e-05, "data_time": 0.0015778541564941406, "loss": 0.2964512422680855, "time": 0.6337619304656983, "epoch": 4, "memory": 44138, "step": 14021} +{"lr": 7.398046506145803e-05, "data_time": 0.0016014814376831056, "loss": 0.30565634071826936, "time": 0.6330125570297241, "epoch": 4, "memory": 44138, "step": 14121} +{"lr": 7.365177486820114e-05, "data_time": 0.0016689777374267578, "loss": 0.3190244257450104, "time": 0.6331549167633057, "epoch": 4, "memory": 44138, "step": 14221} +{"lr": 7.332187872092939e-05, "data_time": 0.0019321203231811523, "loss": 0.3105497986078262, "time": 0.6332573890686035, "epoch": 4, "memory": 44138, "step": 14321} +{"lr": 7.299079794949401e-05, "data_time": 0.0018114805221557616, "loss": 0.3333969205617905, "time": 0.6330009937286377, "epoch": 4, "memory": 44138, "step": 14421} +{"lr": 7.265855396033955e-05, "data_time": 0.0012221574783325196, "loss": 0.3279608815908432, "time": 0.6332975864410401, "epoch": 4, "memory": 44138, "step": 14521} +{"lr": 7.232516823512013e-05, "data_time": 0.0015579938888549804, "loss": 0.3360292732715607, "time": 0.6332832098007202, "epoch": 4, "memory": 44138, "step": 14621} +{"lr": 7.199066232930995e-05, "data_time": 0.0019704103469848633, "loss": 0.3111056387424469, "time": 0.6339174985885621, "epoch": 4, "memory": 44138, "step": 14721} +{"lr": 7.165505787080997e-05, "data_time": 0.0017514705657958984, "loss": 0.31476400792598724, "time": 0.6328725337982177, "epoch": 4, "memory": 44138, "step": 14821} +{"lr": 7.13183765585495e-05, "data_time": 0.002126479148864746, "loss": 0.31744611263275146, "time": 0.6325232028961182, "epoch": 4, "memory": 44138, "step": 14921} +{"lr": 7.098064016108317e-05, "data_time": 0.0018365859985351562, "loss": 0.2819396436214447, "time": 0.6314771175384521, "epoch": 4, "memory": 44138, "step": 15021} +{"lr": 7.064187051518379e-05, "data_time": 0.0018094778060913086, "loss": 0.3287270963191986, "time": 0.6328166723251343, "epoch": 4, "memory": 44138, "step": 15121} +{"lr": 7.030208952442949e-05, "data_time": 0.0019420385360717773, "loss": 0.3135231643915176, "time": 0.6320729494094849, "epoch": 4, "memory": 44138, "step": 15221} +{"lr": 6.996131915778851e-05, "data_time": 0.0018641948699951172, "loss": 0.28980848640203477, "time": 0.6325138807296753, "epoch": 4, "memory": 44138, "step": 15321} +{"lr": 6.961958144819853e-05, "data_time": 0.001683354377746582, "loss": 0.30070672035217283, "time": 0.6332222700119019, "epoch": 4, "memory": 44138, "step": 15421} +{"lr": 6.927689849114193e-05, "data_time": 0.0016742944717407227, "loss": 0.3087885230779648, "time": 0.633359694480896, "epoch": 4, "memory": 44138, "step": 15521} +{"lr": 6.893329244321722e-05, "data_time": 0.0014219999313354493, "loss": 0.29107152819633486, "time": 0.6310092449188233, "epoch": 4, "memory": 44138, "step": 15621} +{"accuracy/top1": 89.60022735595703, "data_time": 0.033523913087516, "time": 1.2778359528245598, "step": 4} +{"lr": 6.856463687361792e-05, "data_time": 0.001884150505065918, "loss": 0.2845174640417099, "time": 0.6339736461639405, "epoch": 5, "memory": 44138, "step": 15728} +{"lr": 6.821919068470432e-05, "data_time": 0.00172879695892334, "loss": 0.32641951739788055, "time": 0.6340963840484619, "epoch": 5, "memory": 44138, "step": 15828} +{"lr": 6.787288979235407e-05, "data_time": 0.0016388177871704101, "loss": 0.30662190318107607, "time": 0.6318677425384521, "epoch": 5, "memory": 44138, "step": 15928} +{"lr": 6.752575658708793e-05, "data_time": 0.0016812562942504882, "loss": 0.28915023654699323, "time": 0.6328672409057617, "epoch": 5, "memory": 44138, "step": 16028} +{"lr": 6.71778135132408e-05, "data_time": 0.0017188549041748046, "loss": 0.3077644228935242, "time": 0.6326217889785767, "epoch": 5, "memory": 44138, "step": 16128} +{"lr": 6.682908306751074e-05, "data_time": 0.0016791105270385742, "loss": 0.31356306076049806, "time": 0.6329569816589355, "epoch": 5, "memory": 44138, "step": 16228} +{"lr": 6.64795877975043e-05, "data_time": 0.00176849365234375, "loss": 0.2925348594784737, "time": 0.6409899950027466, "epoch": 5, "memory": 44138, "step": 16328} +{"lr": 6.61293503002787e-05, "data_time": 0.0016947746276855468, "loss": 0.29854719489812853, "time": 0.633159875869751, "epoch": 5, "memory": 44138, "step": 16428} +{"lr": 6.577839322088084e-05, "data_time": 0.001694035530090332, "loss": 0.3069661006331444, "time": 0.6331403255462646, "epoch": 5, "memory": 44138, "step": 16528} +{"lr": 6.542673925088298e-05, "data_time": 0.0016147613525390625, "loss": 0.310231477022171, "time": 0.6331608057022095, "epoch": 5, "memory": 44138, "step": 16628} +{"lr": 6.507441112691584e-05, "data_time": 0.0018262624740600585, "loss": 0.3002298802137375, "time": 0.6339569330215454, "epoch": 5, "memory": 44138, "step": 16728} +{"lr": 6.47214316291983e-05, "data_time": 0.0015388727188110352, "loss": 0.30747975409030914, "time": 0.6342548847198486, "epoch": 5, "memory": 44138, "step": 16828} +{"lr": 6.436782358006481e-05, "data_time": 0.001520848274230957, "loss": 0.3245810091495514, "time": 0.6329957962036132, "epoch": 5, "memory": 44138, "step": 16928} +{"lr": 6.40136098424896e-05, "data_time": 0.0016062259674072266, "loss": 0.30905329883098603, "time": 0.6319875001907349, "epoch": 5, "memory": 44138, "step": 17028} +{"lr": 6.365881331860828e-05, "data_time": 0.0016116619110107422, "loss": 0.2932494878768921, "time": 0.6321008443832398, "epoch": 5, "memory": 44138, "step": 17128} +{"lr": 6.330345694823739e-05, "data_time": 0.0018414974212646485, "loss": 0.3366967052221298, "time": 0.6327253103256225, "epoch": 5, "memory": 44138, "step": 17228} +{"lr": 6.29475637073908e-05, "data_time": 0.002088785171508789, "loss": 0.2893759086728096, "time": 0.6335022449493408, "epoch": 5, "memory": 44138, "step": 17328} +{"lr": 6.259115660679482e-05, "data_time": 0.0016201496124267577, "loss": 0.31572250127792356, "time": 0.632670259475708, "epoch": 5, "memory": 44138, "step": 17428} +{"lr": 6.223425869039963e-05, "data_time": 0.0020895004272460938, "loss": 0.29802556335926056, "time": 0.633821177482605, "epoch": 5, "memory": 44138, "step": 17528} +{"lr": 6.187689303389006e-05, "data_time": 0.0015899658203125, "loss": 0.28801448047161105, "time": 0.6328324317932129, "epoch": 5, "memory": 44138, "step": 17628} +{"lr": 6.151908274319292e-05, "data_time": 0.0019217491149902343, "loss": 0.30075431764125826, "time": 0.6327396869659424, "epoch": 5, "memory": 44138, "step": 17728} +{"lr": 6.116085095298373e-05, "data_time": 0.0018248558044433594, "loss": 0.28984824419021604, "time": 0.6323301553726196, "epoch": 5, "memory": 44138, "step": 17828} +{"lr": 6.080222082519007e-05, "data_time": 0.0015449285507202148, "loss": 0.30669480562210083, "time": 0.6335581541061401, "epoch": 5, "memory": 44138, "step": 17928} +{"lr": 6.0443215547495147e-05, "data_time": 0.0016098737716674805, "loss": 0.2870576292276382, "time": 0.6318952560424804, "epoch": 5, "memory": 44138, "step": 18028} +{"lr": 6.0083858331837714e-05, "data_time": 0.0019602060317993166, "loss": 0.2967754065990448, "time": 0.6327675819396973, "epoch": 5, "memory": 44138, "step": 18128} +{"lr": 5.97241724129116e-05, "data_time": 0.0016579627990722656, "loss": 0.29930940419435503, "time": 0.6322240591049194, "epoch": 5, "memory": 44138, "step": 18228} +{"lr": 5.936418104666328e-05, "data_time": 0.0015895843505859375, "loss": 0.30931147634983064, "time": 0.6318528652191162, "epoch": 5, "memory": 44138, "step": 18328} +{"lr": 5.9003907508788295e-05, "data_time": 0.001702737808227539, "loss": 0.3269601136445999, "time": 0.632736325263977, "epoch": 5, "memory": 44138, "step": 18428} +{"lr": 5.864337509322633e-05, "data_time": 0.0016469240188598632, "loss": 0.3123019427061081, "time": 0.6319441318511962, "epoch": 5, "memory": 44138, "step": 18528} +{"lr": 5.8282607110655194e-05, "data_time": 0.0019517898559570312, "loss": 0.3098306179046631, "time": 0.6324620962142944, "epoch": 5, "memory": 44138, "step": 18628} +{"lr": 5.7921626886983564e-05, "data_time": 0.002015233039855957, "loss": 0.3117759957909584, "time": 0.632539701461792, "epoch": 5, "memory": 44138, "step": 18728} +{"lr": 5.756045776184287e-05, "data_time": 0.0015504121780395507, "loss": 0.3262644737958908, "time": 0.6331888914108277, "epoch": 5, "memory": 44138, "step": 18828} +{"lr": 5.719912308707808e-05, "data_time": 0.0015547513961791993, "loss": 0.30534543097019196, "time": 0.6331577062606811, "epoch": 5, "memory": 44138, "step": 18928} +{"lr": 5.683764622523808e-05, "data_time": 0.0015584230422973633, "loss": 0.31962490975856783, "time": 0.632534408569336, "epoch": 5, "memory": 44138, "step": 19028} +{"lr": 5.647605054806502e-05, "data_time": 0.0019353628158569336, "loss": 0.29977654814720156, "time": 0.6329171180725097, "epoch": 5, "memory": 44138, "step": 19128} +{"lr": 5.6114359434983206e-05, "data_time": 0.0030315876007080077, "loss": 0.29852366000413894, "time": 0.6360118865966797, "epoch": 5, "memory": 44138, "step": 19228} +{"lr": 5.575259627158745e-05, "data_time": 0.002434039115905762, "loss": 0.32444908618927004, "time": 0.6391826391220092, "epoch": 5, "memory": 44138, "step": 19328} +{"lr": 5.539078444813113e-05, "data_time": 0.0018876075744628906, "loss": 0.27897868752479554, "time": 0.6362127542495728, "epoch": 5, "memory": 44138, "step": 19428} +{"lr": 5.5028947358013735e-05, "data_time": 0.0014713525772094727, "loss": 0.30247158110141753, "time": 0.6336456060409545, "epoch": 5, "memory": 44138, "step": 19528} +{"accuracy/top1": 89.95511627197266, "data_time": 0.03747370325285813, "time": 1.2829449217894981, "step": 5} +{"lr": 5.464178020340108e-05, "data_time": 0.0018750667572021485, "loss": 0.293927937746048, "time": 0.6334022521972656, "epoch": 6, "memory": 44138, "step": 19635} +{"lr": 5.427996514795046e-05, "data_time": 0.0016377687454223633, "loss": 0.34120202660560606, "time": 0.6334142208099365, "epoch": 6, "memory": 44138, "step": 19735} +{"lr": 5.3918196647260834e-05, "data_time": 0.0027139902114868162, "loss": 0.2941610261797905, "time": 0.6372280836105346, "epoch": 6, "memory": 44138, "step": 19835} +{"lr": 5.355649809193057e-05, "data_time": 0.0017122507095336914, "loss": 0.3073193609714508, "time": 0.6329587936401367, "epoch": 6, "memory": 44138, "step": 19935} +{"lr": 5.3194892868035765e-05, "data_time": 0.002005934715270996, "loss": 0.32290321588516235, "time": 0.6343416213989258, "epoch": 6, "memory": 44138, "step": 20035} +{"lr": 5.283340435561795e-05, "data_time": 0.0023032665252685548, "loss": 0.3329052716493607, "time": 0.6344602108001709, "epoch": 6, "memory": 44138, "step": 20135} +{"lr": 5.2472055927172495e-05, "data_time": 0.001950240135192871, "loss": 0.2771913021802902, "time": 0.634003472328186, "epoch": 6, "memory": 44138, "step": 20235} +{"lr": 5.2110870946137605e-05, "data_time": 0.0028927087783813476, "loss": 0.31453161835670473, "time": 0.6333108425140381, "epoch": 6, "memory": 44138, "step": 20335} +{"lr": 5.174987276538342e-05, "data_time": 0.0016325950622558595, "loss": 0.3118656262755394, "time": 0.6319082021713257, "epoch": 6, "memory": 44138, "step": 20435} +{"lr": 5.1389084725702425e-05, "data_time": 0.0027080774307250977, "loss": 0.29839026927948, "time": 0.6348451375961304, "epoch": 6, "memory": 44138, "step": 20535} +{"lr": 5.1028530154300045e-05, "data_time": 0.0021979570388793944, "loss": 0.2979643076658249, "time": 0.6355926752090454, "epoch": 6, "memory": 44138, "step": 20635} +{"lr": 5.066823236328663e-05, "data_time": 0.0018099069595336914, "loss": 0.3005822032690048, "time": 0.6334505081176758, "epoch": 6, "memory": 44138, "step": 20735} +{"lr": 5.030821464817006e-05, "data_time": 0.0017623424530029297, "loss": 0.30971167981624603, "time": 0.6332575559616089, "epoch": 6, "memory": 44138, "step": 20835} +{"lr": 4.99485002863494e-05, "data_time": 0.002543354034423828, "loss": 0.3223777174949646, "time": 0.6341951847076416, "epoch": 6, "memory": 44138, "step": 20935} +{"lr": 4.95891125356102e-05, "data_time": 0.0016045570373535156, "loss": 0.305197237432003, "time": 0.6353859663009643, "epoch": 6, "memory": 44138, "step": 21035} +{"lr": 4.923007463262053e-05, "data_time": 0.0020807266235351564, "loss": 0.2932621449232101, "time": 0.6345109701156616, "epoch": 6, "memory": 44138, "step": 21135} +{"lr": 4.887140979142852e-05, "data_time": 0.0022014379501342773, "loss": 0.2998196542263031, "time": 0.6330170869827271, "epoch": 6, "memory": 44138, "step": 21235} +{"lr": 4.851314120196161e-05, "data_time": 0.0019437313079833985, "loss": 0.2937133401632309, "time": 0.6325894355773926, "epoch": 6, "memory": 44138, "step": 21335} +{"lr": 4.815529202852693e-05, "data_time": 0.001574397087097168, "loss": 0.3261815220117569, "time": 0.6337750673294067, "epoch": 6, "memory": 44138, "step": 21435} +{"lr": 4.779788540831406e-05, "data_time": 0.0024583101272583007, "loss": 0.28951150476932525, "time": 0.6366761922836304, "epoch": 6, "memory": 44138, "step": 21535} +{"lr": 4.744094444989843e-05, "data_time": 0.002363729476928711, "loss": 0.31034710109233854, "time": 0.6336637735366821, "epoch": 6, "memory": 44138, "step": 21635} +{"lr": 4.708449223174764e-05, "data_time": 0.004822397232055664, "loss": 0.2788639008998871, "time": 0.6383045196533204, "epoch": 6, "memory": 44138, "step": 21735} +{"lr": 4.672855180072916e-05, "data_time": 0.001599574089050293, "loss": 0.31344456523656844, "time": 0.6327584028244019, "epoch": 6, "memory": 44138, "step": 21835} +{"lr": 4.6373146170620256e-05, "data_time": 0.002849841117858887, "loss": 0.28745528757572175, "time": 0.6367928266525269, "epoch": 6, "memory": 44138, "step": 21935} +{"lr": 4.6018298320619915e-05, "data_time": 0.004253745079040527, "loss": 0.3103777587413788, "time": 0.6370688676834106, "epoch": 6, "memory": 44138, "step": 22035} +{"lr": 4.566403119386318e-05, "data_time": 0.0017075777053833009, "loss": 0.3060238495469093, "time": 0.6330570936203003, "epoch": 6, "memory": 44138, "step": 22135} +{"lr": 4.531036769593769e-05, "data_time": 0.0018994808197021484, "loss": 0.3336349785327911, "time": 0.6324459552764893, "epoch": 6, "memory": 44138, "step": 22235} +{"lr": 4.4957330693402685e-05, "data_time": 0.0031962871551513674, "loss": 0.34569916427135466, "time": 0.636173152923584, "epoch": 6, "memory": 44138, "step": 22335} +{"lr": 4.460494301231059e-05, "data_time": 0.002243232727050781, "loss": 0.313540580868721, "time": 0.6330457925796509, "epoch": 6, "memory": 44138, "step": 22435} +{"lr": 4.425322743673109e-05, "data_time": 0.0016682863235473633, "loss": 0.31509087830781934, "time": 0.6337332963943482, "epoch": 6, "memory": 44138, "step": 22535} +{"lr": 4.390220670727812e-05, "data_time": 0.0017115354537963867, "loss": 0.3199805974960327, "time": 0.6340263605117797, "epoch": 6, "memory": 44138, "step": 22635} +{"lr": 4.3551903519639325e-05, "data_time": 0.0016652584075927735, "loss": 0.30514743030071256, "time": 0.6322040319442749, "epoch": 6, "memory": 44138, "step": 22735} +{"lr": 4.3202340523108894e-05, "data_time": 0.0025270462036132814, "loss": 0.3206787109375, "time": 0.6345431566238403, "epoch": 6, "memory": 44138, "step": 22835} +{"lr": 4.285354031912302e-05, "data_time": 0.0025682687759399415, "loss": 0.2973847329616547, "time": 0.632455563545227, "epoch": 6, "memory": 44138, "step": 22935} +{"lr": 4.2505525459798544e-05, "data_time": 0.0018535852432250977, "loss": 0.28204504400491714, "time": 0.6344072103500367, "epoch": 6, "memory": 44138, "step": 23035} +{"lr": 4.2158318446474836e-05, "data_time": 0.0020078420639038086, "loss": 0.30542626082897184, "time": 0.6473349571228028, "epoch": 6, "memory": 44138, "step": 23135} +{"lr": 4.181194172825905e-05, "data_time": 0.001706695556640625, "loss": 0.3048213630914688, "time": 0.6328002691268921, "epoch": 6, "memory": 44138, "step": 23235} +{"lr": 4.146641770057448e-05, "data_time": 0.001653003692626953, "loss": 0.2996079549193382, "time": 0.6333706140518188, "epoch": 6, "memory": 44138, "step": 23335} +{"lr": 4.112176870371267e-05, "data_time": 0.0014334440231323243, "loss": 0.30148945450782777, "time": 0.6319171190261841, "epoch": 6, "memory": 44138, "step": 23435} +{"accuracy/top1": 90.24738311767578, "data_time": 0.03907531294329413, "time": 1.2821516785128364, "step": 6} +{"lr": 4.0753988582732786e-05, "data_time": 0.0024433135986328125, "loss": 0.32810259610414505, "time": 0.6333051443099975, "epoch": 7, "memory": 44138, "step": 23542} +{"lr": 4.0411221639286815e-05, "data_time": 0.0015944480895996095, "loss": 0.33431318402290344, "time": 0.6316713571548462, "epoch": 7, "memory": 44138, "step": 23642} +{"lr": 4.006939795169518e-05, "data_time": 0.0022157669067382813, "loss": 0.30837578773498536, "time": 0.6351241827011108, "epoch": 7, "memory": 44138, "step": 23742} +{"lr": 3.972853962099913e-05, "data_time": 0.0030332088470458986, "loss": 0.316980317234993, "time": 0.6349954843521118, "epoch": 7, "memory": 44138, "step": 23842} +{"lr": 3.93886686858237e-05, "data_time": 0.0020666122436523438, "loss": 0.29575823396444323, "time": 0.6344683408737183, "epoch": 7, "memory": 44138, "step": 23942} +{"lr": 3.904980712095259e-05, "data_time": 0.002402949333190918, "loss": 0.3236956983804703, "time": 0.6339280843734741, "epoch": 7, "memory": 44138, "step": 24042} +{"lr": 3.871197683590748e-05, "data_time": 0.0018492698669433593, "loss": 0.2733908072113991, "time": 0.6340746402740478, "epoch": 7, "memory": 44138, "step": 24142} +{"lr": 3.837519967353125e-05, "data_time": 0.0019272327423095702, "loss": 0.323754620552063, "time": 0.6356455564498902, "epoch": 7, "memory": 44138, "step": 24242} +{"lr": 3.8039497408575855e-05, "data_time": 0.001482105255126953, "loss": 0.2949719801545143, "time": 0.6321333408355713, "epoch": 7, "memory": 44138, "step": 24342} +{"lr": 3.770489174629432e-05, "data_time": 0.0016910314559936523, "loss": 0.3118149697780609, "time": 0.6324535369873047, "epoch": 7, "memory": 44138, "step": 24442} +{"lr": 3.737140432103754e-05, "data_time": 0.0018266201019287109, "loss": 0.31530108451843264, "time": 0.6332569360733032, "epoch": 7, "memory": 44138, "step": 24542} +{"lr": 3.703905669485545e-05, "data_time": 0.001881718635559082, "loss": 0.32408997416496277, "time": 0.632784104347229, "epoch": 7, "memory": 44138, "step": 24642} +{"lr": 3.6707870356102706e-05, "data_time": 0.0022163152694702148, "loss": 0.3212019979953766, "time": 0.6322555065155029, "epoch": 7, "memory": 44138, "step": 24742} +{"lr": 3.6377866718049484e-05, "data_time": 0.001720571517944336, "loss": 0.30861849188804624, "time": 0.6450377225875854, "epoch": 7, "memory": 44138, "step": 24842} +{"lr": 3.6049067117496965e-05, "data_time": 0.001851654052734375, "loss": 0.3103115975856781, "time": 0.6321022748947144, "epoch": 7, "memory": 44138, "step": 24942} +{"lr": 3.5721492813397755e-05, "data_time": 0.0025243759155273438, "loss": 0.29069556295871735, "time": 0.6345580577850342, "epoch": 7, "memory": 44138, "step": 25042} +{"lr": 3.539516498548151e-05, "data_time": 0.0021893978118896484, "loss": 0.3239056050777435, "time": 0.6321740627288819, "epoch": 7, "memory": 44138, "step": 25142} +{"lr": 3.507010473288518e-05, "data_time": 0.0022469520568847655, "loss": 0.32012275606393814, "time": 0.6339623212814331, "epoch": 7, "memory": 44138, "step": 25242} +{"lr": 3.474633307278909e-05, "data_time": 0.002277231216430664, "loss": 0.2930764466524124, "time": 0.634464693069458, "epoch": 7, "memory": 44138, "step": 25342} +{"lr": 3.4423870939058177e-05, "data_time": 0.0014679431915283203, "loss": 0.30075865536928176, "time": 0.6319833517074585, "epoch": 7, "memory": 44138, "step": 25442} +{"lr": 3.410273918088805e-05, "data_time": 0.002899646759033203, "loss": 0.2855075106024742, "time": 0.6339532613754273, "epoch": 7, "memory": 44138, "step": 25542} +{"lr": 3.37829585614574e-05, "data_time": 0.0033405780792236327, "loss": 0.31129820346832277, "time": 0.6374561548233032, "epoch": 7, "memory": 44138, "step": 25642} +{"lr": 3.3464549756585216e-05, "data_time": 0.002168416976928711, "loss": 0.29246336668729783, "time": 0.6334645986557007, "epoch": 7, "memory": 44138, "step": 25742} +{"lr": 3.314753335339416e-05, "data_time": 0.001820683479309082, "loss": 0.3041336938738823, "time": 0.6351238489151001, "epoch": 7, "memory": 44138, "step": 25842} +{"lr": 3.2831929848979394e-05, "data_time": 0.0015471696853637696, "loss": 0.3046130836009979, "time": 0.6323554515838623, "epoch": 7, "memory": 44138, "step": 25942} +{"lr": 3.251775964908332e-05, "data_time": 0.0022123098373413087, "loss": 0.3483428418636322, "time": 0.6425493478775024, "epoch": 7, "memory": 44138, "step": 26042} +{"lr": 3.22050430667762e-05, "data_time": 0.0025757551193237305, "loss": 0.30965872704982755, "time": 0.6365681886672974, "epoch": 7, "memory": 44138, "step": 26142} +{"lr": 3.189380032114285e-05, "data_time": 0.0017028331756591796, "loss": 0.3338813304901123, "time": 0.6443053960800171, "epoch": 7, "memory": 44138, "step": 26242} +{"lr": 3.1584051535975324e-05, "data_time": 0.001659083366394043, "loss": 0.2889284402132034, "time": 0.632770323753357, "epoch": 7, "memory": 44138, "step": 26342} +{"lr": 3.127581673847175e-05, "data_time": 0.0015621185302734375, "loss": 0.32617907524108886, "time": 0.6320864439010621, "epoch": 7, "memory": 44138, "step": 26442} +{"lr": 3.0969115857941555e-05, "data_time": 0.001957201957702637, "loss": 0.3024232774972916, "time": 0.6326895952224731, "epoch": 7, "memory": 44138, "step": 26542} +{"lr": 3.0663968724516654e-05, "data_time": 0.001799631118774414, "loss": 0.3251842945814133, "time": 0.6321155786514282, "epoch": 7, "memory": 44138, "step": 26642} +{"lr": 3.0360395067869668e-05, "data_time": 0.0016460657119750977, "loss": 0.2930247038602829, "time": 0.6321796178817749, "epoch": 7, "memory": 44138, "step": 26742} +{"lr": 3.0058414515938063e-05, "data_time": 0.0021967649459838866, "loss": 0.30422474443912506, "time": 0.6380234718322754, "epoch": 7, "memory": 44138, "step": 26842} +{"lr": 2.975804659365501e-05, "data_time": 0.002940821647644043, "loss": 0.315599724650383, "time": 0.6333201408386231, "epoch": 7, "memory": 44138, "step": 26942} +{"lr": 2.945931072168714e-05, "data_time": 0.0017727375030517577, "loss": 0.29845210313797, "time": 0.6331922769546509, "epoch": 7, "memory": 44138, "step": 27042} +{"lr": 2.9162226215178855e-05, "data_time": 0.004553484916687012, "loss": 0.31467942893505096, "time": 0.6411008596420288, "epoch": 7, "memory": 44138, "step": 27142} +{"lr": 2.8866812282503322e-05, "data_time": 0.0033054351806640625, "loss": 0.32417497634887693, "time": 0.6332964181900025, "epoch": 7, "memory": 44138, "step": 27242} +{"lr": 2.8573088024020838e-05, "data_time": 0.0015387535095214844, "loss": 0.31645538806915285, "time": 0.6327661275863647, "epoch": 7, "memory": 44138, "step": 27342} +{"accuracy/top1": 89.67329406738281, "data_time": 0.04145223929964263, "time": 1.2863057029658351, "step": 7} +{"lr": 2.8260695817390092e-05, "data_time": 0.001715683937072754, "loss": 0.29987170696258547, "time": 0.6315726280212403, "epoch": 8, "memory": 44138, "step": 27449} +{"lr": 2.7970529402570967e-05, "data_time": 0.0019087076187133789, "loss": 0.3439913898706436, "time": 0.6331199645996094, "epoch": 8, "memory": 44138, "step": 27549} +{"lr": 2.7682110612244214e-05, "data_time": 0.0015655517578125, "loss": 0.3048872321844101, "time": 0.6332799434661865, "epoch": 8, "memory": 44138, "step": 27649} +{"lr": 2.7395458094490283e-05, "data_time": 0.002020597457885742, "loss": 0.3210868567228317, "time": 0.6331084489822387, "epoch": 8, "memory": 44138, "step": 27749} +{"lr": 2.7110590383189015e-05, "data_time": 0.0016994714736938477, "loss": 0.30213554948568344, "time": 0.6323690414428711, "epoch": 8, "memory": 44138, "step": 27849} +{"lr": 2.6827525896821397e-05, "data_time": 0.0020955562591552734, "loss": 0.3053766623139381, "time": 0.633253026008606, "epoch": 8, "memory": 44138, "step": 27949} +{"lr": 2.6546282937278552e-05, "data_time": 0.0016840696334838867, "loss": 0.3279667258262634, "time": 0.6351157426834106, "epoch": 8, "memory": 44138, "step": 28049} +{"lr": 2.6266879688678606e-05, "data_time": 0.0016275405883789062, "loss": 0.31794826686382294, "time": 0.6349692583084107, "epoch": 8, "memory": 44138, "step": 28149} +{"lr": 2.5989334216190738e-05, "data_time": 0.0020150184631347657, "loss": 0.3034849941730499, "time": 0.634264874458313, "epoch": 8, "memory": 44138, "step": 28249} +{"lr": 2.571366446486732e-05, "data_time": 0.0028668880462646485, "loss": 0.31111868768930434, "time": 0.6349168300628663, "epoch": 8, "memory": 44138, "step": 28349} +{"lr": 2.5439888258483665e-05, "data_time": 0.003535032272338867, "loss": 0.3268710315227509, "time": 0.6354778051376343, "epoch": 8, "memory": 44138, "step": 28449} +{"lr": 2.5168023298385384e-05, "data_time": 0.0016741752624511719, "loss": 0.28083030581474305, "time": 0.6329811573028564, "epoch": 8, "memory": 44138, "step": 28549} +{"lr": 2.489808716234412e-05, "data_time": 0.0025573492050170897, "loss": 0.30325927436351774, "time": 0.6332638740539551, "epoch": 8, "memory": 44138, "step": 28649} +{"lr": 2.463009730342111e-05, "data_time": 0.001484251022338867, "loss": 0.27094477862119676, "time": 0.632333779335022, "epoch": 8, "memory": 44138, "step": 28749} +{"lr": 2.436407104883849e-05, "data_time": 0.0029922246932983397, "loss": 0.3023088127374649, "time": 0.6395461320877075, "epoch": 8, "memory": 44138, "step": 28849} +{"lr": 2.4100025598859092e-05, "data_time": 0.0019676923751831056, "loss": 0.2781511783599854, "time": 0.636028528213501, "epoch": 8, "memory": 44138, "step": 28949} +{"lr": 2.3837978025674284e-05, "data_time": 0.0018792152404785156, "loss": 0.3274871453642845, "time": 0.6333925008773804, "epoch": 8, "memory": 44138, "step": 29049} +{"lr": 2.357794527230027e-05, "data_time": 0.001748943328857422, "loss": 0.3072037994861603, "time": 0.6330334424972535, "epoch": 8, "memory": 44138, "step": 29149} +{"lr": 2.33199441514825e-05, "data_time": 0.0018338680267333985, "loss": 0.30363257229328156, "time": 0.634219479560852, "epoch": 8, "memory": 44138, "step": 29249} +{"lr": 2.306399134460859e-05, "data_time": 0.0021389007568359377, "loss": 0.2911437630653381, "time": 0.6338193416595459, "epoch": 8, "memory": 44138, "step": 29349} +{"lr": 2.281010340062995e-05, "data_time": 0.0018537044525146484, "loss": 0.2986374318599701, "time": 0.6339231014251709, "epoch": 8, "memory": 44138, "step": 29449} +{"lr": 2.2558296734991603e-05, "data_time": 0.0014921188354492187, "loss": 0.2997046634554863, "time": 0.6326585531234741, "epoch": 8, "memory": 44138, "step": 29549} +{"lr": 2.2308587628570915e-05, "data_time": 0.001764965057373047, "loss": 0.3540403783321381, "time": 0.6374994516372681, "epoch": 8, "memory": 44138, "step": 29649} +{"lr": 2.2060992226624955e-05, "data_time": 0.002456521987915039, "loss": 0.30933377593755723, "time": 0.6361114978790283, "epoch": 8, "memory": 44138, "step": 29749} +{"lr": 2.181552653774648e-05, "data_time": 0.0018098592758178712, "loss": 0.2852427214384079, "time": 0.6322941064834595, "epoch": 8, "memory": 44138, "step": 29849} +{"lr": 2.1572206432829035e-05, "data_time": 0.0020434379577636717, "loss": 0.33065275847911835, "time": 0.6344693183898926, "epoch": 8, "memory": 44138, "step": 29949} +{"lr": 2.1331047644040754e-05, "data_time": 0.0018612146377563477, "loss": 0.3068340063095093, "time": 0.6406064033508301, "epoch": 8, "memory": 44138, "step": 30049} +{"lr": 2.109206576380705e-05, "data_time": 0.0019402027130126954, "loss": 0.2755319714546204, "time": 0.6329137802124023, "epoch": 8, "memory": 44138, "step": 30149} +{"lr": 2.0855276243802687e-05, "data_time": 0.001954793930053711, "loss": 0.2961136922240257, "time": 0.6327996015548706, "epoch": 8, "memory": 44138, "step": 30249} +{"lr": 2.0620694393952487e-05, "data_time": 0.005257797241210937, "loss": 0.31215171962976457, "time": 0.6379446744918823, "epoch": 8, "memory": 44138, "step": 30349} +{"lr": 2.038833538144169e-05, "data_time": 0.0018012523651123047, "loss": 0.3066970407962799, "time": 0.6348145484924317, "epoch": 8, "memory": 44138, "step": 30449} +{"lr": 2.0158214229735154e-05, "data_time": 0.0019438982009887696, "loss": 0.32797242105007174, "time": 0.6333035707473755, "epoch": 8, "memory": 44138, "step": 30549} +{"lr": 1.9930345817605977e-05, "data_time": 0.0016388416290283204, "loss": 0.29934488534927367, "time": 0.631968903541565, "epoch": 8, "memory": 44138, "step": 30649} +{"lr": 1.970474487817358e-05, "data_time": 0.0015523672103881837, "loss": 0.3224214121699333, "time": 0.632461428642273, "epoch": 8, "memory": 44138, "step": 30749} +{"lr": 1.9481425997951092e-05, "data_time": 0.0019828081130981445, "loss": 0.3191940948367119, "time": 0.6330483436584473, "epoch": 8, "memory": 44138, "step": 30849} +{"lr": 1.9260403615902187e-05, "data_time": 0.0022492170333862304, "loss": 0.2850849464535713, "time": 0.634054708480835, "epoch": 8, "memory": 44138, "step": 30949} +{"lr": 1.9041692022507647e-05, "data_time": 0.0022660493850708008, "loss": 0.3149334788322449, "time": 0.6329516887664794, "epoch": 8, "memory": 44138, "step": 31049} +{"lr": 1.8825305358841125e-05, "data_time": 0.002373361587524414, "loss": 0.33882727921009065, "time": 0.6360174179077148, "epoch": 8, "memory": 44138, "step": 31149} +{"lr": 1.861125761565518e-05, "data_time": 0.0013271331787109374, "loss": 0.28714269697666167, "time": 0.6328182458877564, "epoch": 8, "memory": 44138, "step": 31249} +{"accuracy/top1": 89.25924682617188, "data_time": 0.03800478063780686, "time": 1.2828440871731988, "step": 8} +{"lr": 1.8384832449122864e-05, "data_time": 0.002848911285400391, "loss": 0.3251998543739319, "time": 0.634513258934021, "epoch": 9, "memory": 44138, "step": 31356} +{"lr": 1.817567007332114e-05, "data_time": 0.0018283367156982423, "loss": 0.320851793885231, "time": 0.6322673559188843, "epoch": 9, "memory": 44138, "step": 31456} +{"lr": 1.796888862098825e-05, "data_time": 0.0019241809844970704, "loss": 0.32219129502773286, "time": 0.6350726842880249, "epoch": 9, "memory": 44138, "step": 31556} +{"lr": 1.7764501461839678e-05, "data_time": 0.0019450902938842774, "loss": 0.2998438149690628, "time": 0.6326807737350464, "epoch": 9, "memory": 44138, "step": 31656} +{"lr": 1.7562521810784746e-05, "data_time": 0.0019226312637329102, "loss": 0.2855839133262634, "time": 0.6332160472869873, "epoch": 9, "memory": 44138, "step": 31756} +{"lr": 1.7362962727072444e-05, "data_time": 0.0023490428924560548, "loss": 0.3230756729841232, "time": 0.6363595724105835, "epoch": 9, "memory": 44138, "step": 31856} +{"lr": 1.716583711344682e-05, "data_time": 0.0018647193908691406, "loss": 0.2942240074276924, "time": 0.6362556457519531, "epoch": 9, "memory": 44138, "step": 31956} +{"lr": 1.6971157715312954e-05, "data_time": 0.002173614501953125, "loss": 0.31122791171073916, "time": 0.6339544534683228, "epoch": 9, "memory": 44138, "step": 32056} +{"lr": 1.677893711991274e-05, "data_time": 0.0017704725265502929, "loss": 0.28555778712034224, "time": 0.6361353158950805, "epoch": 9, "memory": 44138, "step": 32156} +{"lr": 1.6589187755511025e-05, "data_time": 0.0017292261123657226, "loss": 0.3068491667509079, "time": 0.6336912155151367, "epoch": 9, "memory": 44138, "step": 32256} +{"lr": 1.640192189059216e-05, "data_time": 0.001858353614807129, "loss": 0.3107178032398224, "time": 0.6351148128509522, "epoch": 9, "memory": 44138, "step": 32356} +{"lr": 1.6217151633066716e-05, "data_time": 0.001909184455871582, "loss": 0.28186499923467634, "time": 0.632826566696167, "epoch": 9, "memory": 44138, "step": 32456} +{"lr": 1.6034888929488564e-05, "data_time": 0.0016216278076171876, "loss": 0.2919092759490013, "time": 0.6329756736755371, "epoch": 9, "memory": 44138, "step": 32556} +{"lr": 1.585514556428258e-05, "data_time": 0.004290103912353516, "loss": 0.2939515233039856, "time": 0.6385218620300293, "epoch": 9, "memory": 44138, "step": 32656} +{"lr": 1.567793315898261e-05, "data_time": 0.001831388473510742, "loss": 0.32163310050964355, "time": 0.6336223840713501, "epoch": 9, "memory": 44138, "step": 32756} +{"lr": 1.5503263171480048e-05, "data_time": 0.0016253471374511718, "loss": 0.33599635064601896, "time": 0.6340978145599365, "epoch": 9, "memory": 44138, "step": 32856} +{"lr": 1.5331146895283165e-05, "data_time": 0.0035316944122314453, "loss": 0.3018835335969925, "time": 0.6369681835174561, "epoch": 9, "memory": 44138, "step": 32956} +{"lr": 1.5161595458786728e-05, "data_time": 0.002083706855773926, "loss": 0.3038186222314835, "time": 0.6329025030136108, "epoch": 9, "memory": 44138, "step": 33056} +{"lr": 1.4994619824552563e-05, "data_time": 0.002429080009460449, "loss": 0.3399715185165405, "time": 0.6365882635116578, "epoch": 9, "memory": 44138, "step": 33156} +{"lr": 1.4830230788600802e-05, "data_time": 0.0028122663497924805, "loss": 0.32248779237270353, "time": 0.6355271339416504, "epoch": 9, "memory": 44138, "step": 33256} +{"lr": 1.4668438979711779e-05, "data_time": 0.001717996597290039, "loss": 0.301661479473114, "time": 0.6334915876388549, "epoch": 9, "memory": 44138, "step": 33356} +{"lr": 1.4509254858738844e-05, "data_time": 0.001712799072265625, "loss": 0.3399364411830902, "time": 0.6332926034927369, "epoch": 9, "memory": 44138, "step": 33456} +{"lr": 1.4352688717931966e-05, "data_time": 0.0021294593811035157, "loss": 0.31088860630989074, "time": 0.6344173431396485, "epoch": 9, "memory": 44138, "step": 33556} +{"lr": 1.4198750680272348e-05, "data_time": 0.0017971038818359376, "loss": 0.30998590886592864, "time": 0.6326429843902588, "epoch": 9, "memory": 44138, "step": 33656} +{"lr": 1.4047450698817857e-05, "data_time": 0.002286982536315918, "loss": 0.29340130686759947, "time": 0.6346447944641114, "epoch": 9, "memory": 44138, "step": 33756} +{"lr": 1.389879855605952e-05, "data_time": 0.0018810272216796876, "loss": 0.3187680631875992, "time": 0.6445939779281616, "epoch": 9, "memory": 44138, "step": 33856} +{"lr": 1.3752803863288976e-05, "data_time": 0.003167533874511719, "loss": 0.2890175938606262, "time": 0.636824107170105, "epoch": 9, "memory": 44138, "step": 33956} +{"lr": 1.3609476059977129e-05, "data_time": 0.0018992662429809571, "loss": 0.3111235052347183, "time": 0.6367466688156128, "epoch": 9, "memory": 44138, "step": 34056} +{"lr": 1.3468824413163754e-05, "data_time": 0.0017231225967407227, "loss": 0.3018523871898651, "time": 0.63269944190979, "epoch": 9, "memory": 44138, "step": 34156} +{"lr": 1.3330858016858395e-05, "data_time": 0.002386045455932617, "loss": 0.3260850816965103, "time": 0.6369416952133179, "epoch": 9, "memory": 44138, "step": 34256} +{"lr": 1.3195585791452309e-05, "data_time": 0.0018776178359985352, "loss": 0.29141670614480975, "time": 0.6336064100265503, "epoch": 9, "memory": 44138, "step": 34356} +{"lr": 1.3063016483141729e-05, "data_time": 0.003920650482177735, "loss": 0.3216202884912491, "time": 0.6353137969970704, "epoch": 9, "memory": 44138, "step": 34456} +{"lr": 1.2933158663362388e-05, "data_time": 0.0017136812210083007, "loss": 0.3003173589706421, "time": 0.6339617252349854, "epoch": 9, "memory": 44138, "step": 34556} +{"lr": 1.2806020728235311e-05, "data_time": 0.0017922878265380859, "loss": 0.3098007678985596, "time": 0.6328829288482666, "epoch": 9, "memory": 44138, "step": 34656} +{"lr": 1.2681610898023963e-05, "data_time": 0.0016156911849975586, "loss": 0.32777359485626223, "time": 0.6387759685516358, "epoch": 9, "memory": 44138, "step": 34756} +{"lr": 1.2559937216602731e-05, "data_time": 0.0035127878189086916, "loss": 0.27056471854448316, "time": 0.6349343299865723, "epoch": 9, "memory": 44138, "step": 34856} +{"lr": 1.2441007550936847e-05, "data_time": 0.0021913766860961912, "loss": 0.3135535687208176, "time": 0.6349970817565918, "epoch": 9, "memory": 44138, "step": 34956} +{"lr": 1.232482959057377e-05, "data_time": 0.0020348072052001954, "loss": 0.32948253154754636, "time": 0.6353827476501465, "epoch": 9, "memory": 44138, "step": 35056} +{"lr": 1.2211410847145969e-05, "data_time": 0.0014485836029052735, "loss": 0.32403912097215654, "time": 0.6327764511108398, "epoch": 9, "memory": 44138, "step": 35156} +{"accuracy/top1": 88.6712417602539, "data_time": 0.03698922847879344, "time": 1.2817480481904129, "step": 9} +{"lr": 1.209311679360931e-05, "data_time": 0.00230712890625, "loss": 0.3084552973508835, "time": 0.6392616510391236, "epoch": 10, "memory": 44138, "step": 35263} +{"lr": 1.1985432727488666e-05, "data_time": 0.0022099971771240234, "loss": 0.28848199993371965, "time": 0.6374534845352173, "epoch": 10, "memory": 44138, "step": 35363} +{"lr": 1.1880529822435289e-05, "data_time": 0.001630401611328125, "loss": 0.28577466011047364, "time": 0.6338502883911132, "epoch": 10, "memory": 44138, "step": 35463} +{"lr": 1.1778414861078726e-05, "data_time": 0.0018352985382080079, "loss": 0.32852563709020616, "time": 0.6359284162521363, "epoch": 10, "memory": 44138, "step": 35563} +{"lr": 1.1679094445790538e-05, "data_time": 0.001617097854614258, "loss": 0.3234549969434738, "time": 0.6330214977264405, "epoch": 10, "memory": 44138, "step": 35663} +{"lr": 1.1582574998257377e-05, "data_time": 0.00251467227935791, "loss": 0.2979410693049431, "time": 0.6355472564697265, "epoch": 10, "memory": 44138, "step": 35763} +{"lr": 1.1488862759065762e-05, "data_time": 0.002043938636779785, "loss": 0.297506745159626, "time": 0.636680793762207, "epoch": 10, "memory": 44138, "step": 35863} +{"lr": 1.1397963787298665e-05, "data_time": 0.0018564701080322266, "loss": 0.30051955580711365, "time": 0.6340206146240235, "epoch": 10, "memory": 44138, "step": 35963} +{"lr": 1.1309883960143664e-05, "data_time": 0.0022110939025878906, "loss": 0.3067521661520004, "time": 0.6357958555221558, "epoch": 10, "memory": 44138, "step": 36063} +{"lr": 1.122462897251302e-05, "data_time": 0.0017559289932250976, "loss": 0.2953896582126617, "time": 0.6364803314208984, "epoch": 10, "memory": 44138, "step": 36163} +{"lr": 1.114220433667542e-05, "data_time": 0.004400992393493652, "loss": 0.31145541816949845, "time": 0.6357387781143189, "epoch": 10, "memory": 44138, "step": 36263} +{"lr": 1.10626153818996e-05, "data_time": 0.0024420976638793944, "loss": 0.30642210841178896, "time": 0.6363560438156128, "epoch": 10, "memory": 44138, "step": 36363} +{"lr": 1.0985867254109761e-05, "data_time": 0.0020621061325073243, "loss": 0.33522696793079376, "time": 0.6336010217666626, "epoch": 10, "memory": 44138, "step": 36463} +{"lr": 1.0911964915552823e-05, "data_time": 0.0021435022354125977, "loss": 0.3141847595572472, "time": 0.6341703414916993, "epoch": 10, "memory": 44138, "step": 36563} +{"lr": 1.0840913144477643e-05, "data_time": 0.0015696287155151367, "loss": 0.32551797330379484, "time": 0.632967185974121, "epoch": 10, "memory": 44138, "step": 36663} +{"lr": 1.0772716534826064e-05, "data_time": 0.0017685174942016601, "loss": 0.3243787676095963, "time": 0.6345966577529907, "epoch": 10, "memory": 44138, "step": 36763} +{"lr": 1.0707379495935828e-05, "data_time": 0.001757049560546875, "loss": 0.30183427035808563, "time": 0.6334445476531982, "epoch": 10, "memory": 44138, "step": 36863} +{"lr": 1.0644906252255541e-05, "data_time": 0.001818084716796875, "loss": 0.32428970336914065, "time": 0.6361659526824951, "epoch": 10, "memory": 44138, "step": 36963} +{"lr": 1.0585300843071527e-05, "data_time": 0.0016587018966674806, "loss": 0.30885269343852995, "time": 0.6327872753143311, "epoch": 10, "memory": 44138, "step": 37063} +{"lr": 1.0528567122246654e-05, "data_time": 0.002350735664367676, "loss": 0.3016982674598694, "time": 0.6364793300628662, "epoch": 10, "memory": 44138, "step": 37163} +{"lr": 1.0474708757971141e-05, "data_time": 0.0017575979232788085, "loss": 0.3050421953201294, "time": 0.6336272716522217, "epoch": 10, "memory": 44138, "step": 37263} +{"lr": 1.042372923252543e-05, "data_time": 0.003053712844848633, "loss": 0.29336825013160706, "time": 0.6358482360839843, "epoch": 10, "memory": 44138, "step": 37363} +{"lr": 1.0375631842054998e-05, "data_time": 0.004346632957458496, "loss": 0.3325827866792679, "time": 0.638576078414917, "epoch": 10, "memory": 44138, "step": 37463} +{"lr": 1.0330419696357254e-05, "data_time": 0.002854728698730469, "loss": 0.31957771331071855, "time": 0.6384903430938721, "epoch": 10, "memory": 44138, "step": 37563} +{"lr": 1.028809571868047e-05, "data_time": 0.005306839942932129, "loss": 0.3192389339208603, "time": 0.6413367748260498, "epoch": 10, "memory": 44138, "step": 37663} +{"lr": 1.0248662645534766e-05, "data_time": 0.003085160255432129, "loss": 0.3142657458782196, "time": 0.6393048286437988, "epoch": 10, "memory": 44138, "step": 37763} +{"lr": 1.0212123026515194e-05, "data_time": 0.0040253400802612305, "loss": 0.2617375895380974, "time": 0.6411966562271119, "epoch": 10, "memory": 44138, "step": 37863} +{"lr": 1.0178479224136881e-05, "data_time": 0.004627513885498047, "loss": 0.2990847289562225, "time": 0.6819361686706543, "epoch": 10, "memory": 44138, "step": 37963} +{"lr": 1.0147733413682289e-05, "data_time": 0.007761240005493164, "loss": 0.3163591355085373, "time": 0.6426003932952881, "epoch": 10, "memory": 44138, "step": 38063} +{"lr": 1.0119887583060527e-05, "data_time": 0.0013313770294189453, "loss": 0.3314773261547089, "time": 0.632814359664917, "epoch": 10, "memory": 44138, "step": 38163} +{"lr": 1.0094943532678871e-05, "data_time": 0.0016098737716674805, "loss": 0.30876474678516386, "time": 0.632322359085083, "epoch": 10, "memory": 44138, "step": 38263} +{"lr": 1.007290287532637e-05, "data_time": 0.007057762145996094, "loss": 0.29862377792596817, "time": 0.6595552682876586, "epoch": 10, "memory": 44138, "step": 38363} +{"lr": 1.0053767036069495e-05, "data_time": 0.001546788215637207, "loss": 0.30123724937438967, "time": 0.6319895267486573, "epoch": 10, "memory": 44138, "step": 38463} +{"lr": 1.003753725216005e-05, "data_time": 0.0037961959838867187, "loss": 0.34542007744312286, "time": 0.6363131999969482, "epoch": 10, "memory": 44138, "step": 38563} +{"lr": 1.0024214572955152e-05, "data_time": 0.01427314281463623, "loss": 0.2820836007595062, "time": 0.6622368097305298, "epoch": 10, "memory": 44138, "step": 38663} +{"lr": 1.0013799859849456e-05, "data_time": 0.001521921157836914, "loss": 0.2955233693122864, "time": 0.6327824831008911, "epoch": 10, "memory": 44138, "step": 38763} +{"lr": 1.0006293786219338e-05, "data_time": 0.002329707145690918, "loss": 0.2788534551858902, "time": 0.6370566606521606, "epoch": 10, "memory": 44138, "step": 38863} +{"lr": 1.0001696837379472e-05, "data_time": 0.008408093452453613, "loss": 0.3348928228020668, "time": 0.6573300123214721, "epoch": 10, "memory": 44138, "step": 38963} +{"lr": 1.0000009310551361e-05, "data_time": 0.003323864936828613, "loss": 0.30951560884714124, "time": 0.6480633735656738, "epoch": 10, "memory": 44138, "step": 39063} +{"accuracy/top1": 88.59817504882812, "data_time": 0.11093969180666167, "time": 1.353807778194033, "step": 10} diff --git a/clip_large_pretrain_4x256_sdv1_lr1e-4_wopostnorm_wneck/20230604_175638/vis_data/config.py b/clip_large_pretrain_4x256_sdv1_lr1e-4_wopostnorm_wneck/20230604_175638/vis_data/config.py new file mode 100644 index 0000000000000000000000000000000000000000..af1f90081c95d88c89d5cc97dca60c3ea695bcf9 --- /dev/null +++ b/clip_large_pretrain_4x256_sdv1_lr1e-4_wopostnorm_wneck/20230604_175638/vis_data/config.py @@ -0,0 +1,236 @@ +optim_wrapper = dict( + optimizer=dict( + type='AdamW', lr=0.0001, weight_decay=0.3, _scope_='mmpretrain'), + paramwise_cfg=dict( + custom_keys=dict({ + '.cls_token': dict(decay_mult=0.0), + '.pos_embed': dict(decay_mult=0.0) + })), + type='AmpOptimWrapper', + dtype='bfloat16', + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=False, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=4096) +model = dict( + type='ImageClassifier', + backbone=dict( + frozen_stages=24, + type='VisionTransformer', + arch='l', + img_size=224, + patch_size=14, + drop_rate=0.1, + pre_norm=True, + final_norm=False, + init_cfg=dict( + type='Pretrained', + checkpoint='ckpt/openclip-ViT-L-14.pth', + prefix='backbone')), + neck=dict( + type='CLIPProjection', + in_channels=1024, + out_channels=768, + init_cfg=dict( + type='Pretrained', + checkpoint='ckpt/openclip-ViT-L-14.pth', + prefix='backbone')), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=768, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + init_cfg=None), + init_cfg=dict( + type='TruncNormal', layer=['Conv2d', 'Linear'], std=0.02, bias=0.0), + train_cfg=None) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=128, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stablediffusionV1-5R2-dpmsolver-25-1m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/cc1m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=128, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=128, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/clip_large_pretrain_4x256_sdv1_lr1e-4_wopostnorm_wneck' diff --git a/clip_large_pretrain_4x256_sdv1_lr1e-4_wopostnorm_wneck/20230604_175638/vis_data/events.out.tfevents.1685872614.SH-IDC1-10-140-24-15.164206.0 b/clip_large_pretrain_4x256_sdv1_lr1e-4_wopostnorm_wneck/20230604_175638/vis_data/events.out.tfevents.1685872614.SH-IDC1-10-140-24-15.164206.0 new file mode 100644 index 0000000000000000000000000000000000000000..a6ce0b97b6130384a882bf3a166416348cbf733d --- /dev/null +++ b/clip_large_pretrain_4x256_sdv1_lr1e-4_wopostnorm_wneck/20230604_175638/vis_data/events.out.tfevents.1685872614.SH-IDC1-10-140-24-15.164206.0 @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:7d1f4f1a7baaa7bd8e13c46868dcab4bfe438198b59e54f218b9cb3f8378ce7c +size 1615529 diff --git a/clip_large_pretrain_4x256_sdv1_lr1e-4_wopostnorm_wneck/20230604_175638/vis_data/scalars.json b/clip_large_pretrain_4x256_sdv1_lr1e-4_wopostnorm_wneck/20230604_175638/vis_data/scalars.json new file mode 100644 index 0000000000000000000000000000000000000000..fde6101e2d597abe1882751764205750c8b3daa9 --- /dev/null +++ b/clip_large_pretrain_4x256_sdv1_lr1e-4_wopostnorm_wneck/20230604_175638/vis_data/scalars.json @@ -0,0 +1,400 @@ +{"lr": 9.999857418382631e-05, "data_time": 0.0026583194732666014, "loss": 0.5272099018096924, "time": 0.6304151773452759, "epoch": 1, "memory": 44139, "step": 100} +{"lr": 9.999423907359506e-05, "data_time": 0.0019651174545288084, "loss": 0.4700825929641724, "time": 0.6316981077194214, "epoch": 1, "memory": 44139, "step": 200} +{"lr": 9.998699480412073e-05, "data_time": 0.001915311813354492, "loss": 0.44749481678009034, "time": 0.6328604698181153, "epoch": 1, "memory": 44139, "step": 300} +{"lr": 9.997684184379085e-05, "data_time": 0.0018089532852172852, "loss": 0.45088441371917726, "time": 0.6322338819503784, "epoch": 1, "memory": 44139, "step": 400} +{"lr": 9.996378084905784e-05, "data_time": 0.0017721176147460938, "loss": 0.41908371448516846, "time": 0.6319724798202515, "epoch": 1, "memory": 44139, "step": 500} +{"lr": 9.994781266439667e-05, "data_time": 0.0020272254943847655, "loss": 0.4147952705621719, "time": 0.6332255363464355, "epoch": 1, "memory": 44139, "step": 600} +{"lr": 9.992893832225061e-05, "data_time": 0.001917409896850586, "loss": 0.3993018567562103, "time": 0.6328696489334107, "epoch": 1, "memory": 44139, "step": 700} +{"lr": 9.990715904296405e-05, "data_time": 0.002123117446899414, "loss": 0.41613166928291323, "time": 0.6325438499450684, "epoch": 1, "memory": 44139, "step": 800} +{"lr": 9.988247623470371e-05, "data_time": 0.0018011569976806641, "loss": 0.395975923538208, "time": 0.6324893712997437, "epoch": 1, "memory": 44139, "step": 900} +{"lr": 9.985489149336749e-05, "data_time": 0.002089095115661621, "loss": 0.4181334525346756, "time": 0.6309988260269165, "epoch": 1, "memory": 44139, "step": 1000} +{"lr": 9.982440660248187e-05, "data_time": 0.0016204833984375, "loss": 0.4181372046470642, "time": 0.6321151494979859, "epoch": 1, "memory": 44139, "step": 1100} +{"lr": 9.979102353308574e-05, "data_time": 0.0017192602157592774, "loss": 0.41433699131011964, "time": 0.632967758178711, "epoch": 1, "memory": 44139, "step": 1200} +{"lr": 9.975474444360358e-05, "data_time": 0.0017733097076416016, "loss": 0.40109002590179443, "time": 0.6327246904373169, "epoch": 1, "memory": 44139, "step": 1300} +{"lr": 9.97155716797056e-05, "data_time": 0.0017506837844848632, "loss": 0.4026054620742798, "time": 0.6326537609100342, "epoch": 1, "memory": 44139, "step": 1400} +{"lr": 9.967350777415632e-05, "data_time": 0.0016286849975585937, "loss": 0.3907728523015976, "time": 0.632654333114624, "epoch": 1, "memory": 44139, "step": 1500} +{"lr": 9.962855544665059e-05, "data_time": 0.0017454147338867188, "loss": 0.3846167206764221, "time": 0.6324703931808472, "epoch": 1, "memory": 44139, "step": 1600} +{"lr": 9.958071760363773e-05, "data_time": 0.0016985654830932618, "loss": 0.40266107618808744, "time": 0.632651424407959, "epoch": 1, "memory": 44139, "step": 1700} +{"lr": 9.952999733813382e-05, "data_time": 0.0018072843551635743, "loss": 0.373909655213356, "time": 0.6331640005111694, "epoch": 1, "memory": 44139, "step": 1800} +{"lr": 9.947639792952148e-05, "data_time": 0.0017575740814208985, "loss": 0.363760969042778, "time": 0.6328064918518066, "epoch": 1, "memory": 44139, "step": 1900} +{"lr": 9.941992284333813e-05, "data_time": 0.0018268108367919921, "loss": 0.39119345545768736, "time": 0.6327016830444336, "epoch": 1, "memory": 44139, "step": 2000} +{"lr": 9.93605757310516e-05, "data_time": 0.001676321029663086, "loss": 0.40079807937145234, "time": 0.6328168153762818, "epoch": 1, "memory": 44139, "step": 2100} +{"lr": 9.929836042982421e-05, "data_time": 0.0016045808792114259, "loss": 0.3407004803419113, "time": 0.6326619863510132, "epoch": 1, "memory": 44139, "step": 2200} +{"lr": 9.923328096226467e-05, "data_time": 0.0016084671020507812, "loss": 0.3782178729772568, "time": 0.6325747966766357, "epoch": 1, "memory": 44139, "step": 2300} +{"lr": 9.916534153616819e-05, "data_time": 0.0016426801681518554, "loss": 0.36378566920757294, "time": 0.6332096576690673, "epoch": 1, "memory": 44139, "step": 2400} +{"lr": 9.909454654424387e-05, "data_time": 0.0021802425384521485, "loss": 0.3504793167114258, "time": 0.6329232931137085, "epoch": 1, "memory": 44139, "step": 2500} +{"lr": 9.902090056383109e-05, "data_time": 0.0019495248794555663, "loss": 0.3691946417093277, "time": 0.6324845552444458, "epoch": 1, "memory": 44139, "step": 2600} +{"lr": 9.894440835660365e-05, "data_time": 0.0015739679336547851, "loss": 0.38485833406448366, "time": 0.6329338550567627, "epoch": 1, "memory": 44139, "step": 2700} +{"lr": 9.886507486826162e-05, "data_time": 0.0018564701080322266, "loss": 0.3473310351371765, "time": 0.63320894241333, "epoch": 1, "memory": 44139, "step": 2800} +{"lr": 9.878290522821154e-05, "data_time": 0.0019229650497436523, "loss": 0.3705559939146042, "time": 0.6331067085266113, "epoch": 1, "memory": 44139, "step": 2900} +{"lr": 9.86979047492352e-05, "data_time": 0.0017976760864257812, "loss": 0.34593615531921384, "time": 0.6325294494628906, "epoch": 1, "memory": 44139, "step": 3000} +{"lr": 9.861007892714578e-05, "data_time": 0.0017056226730346679, "loss": 0.3622700244188309, "time": 0.6328414678573608, "epoch": 1, "memory": 44139, "step": 3100} +{"lr": 9.851943344043239e-05, "data_time": 0.002273368835449219, "loss": 0.3563511461019516, "time": 0.6327307224273682, "epoch": 1, "memory": 44139, "step": 3200} +{"lr": 9.842597414989319e-05, "data_time": 0.0021600246429443358, "loss": 0.3420926809310913, "time": 0.6363249301910401, "epoch": 1, "memory": 44139, "step": 3300} +{"lr": 9.832970709825637e-05, "data_time": 0.0023040771484375, "loss": 0.3658769130706787, "time": 0.6323584318161011, "epoch": 1, "memory": 44139, "step": 3400} +{"lr": 9.823063850978968e-05, "data_time": 0.0016047954559326172, "loss": 0.35051974952220916, "time": 0.6317821979522705, "epoch": 1, "memory": 44139, "step": 3500} +{"lr": 9.81287747898975e-05, "data_time": 0.0018771171569824218, "loss": 0.36420016884803774, "time": 0.6314892530441284, "epoch": 1, "memory": 44139, "step": 3600} +{"lr": 9.802412252470715e-05, "data_time": 0.0019242286682128907, "loss": 0.32584536969661715, "time": 0.6307554960250854, "epoch": 1, "memory": 44139, "step": 3700} +{"lr": 9.791668848064258e-05, "data_time": 0.0017160415649414063, "loss": 0.3360456913709641, "time": 0.6322226762771607, "epoch": 1, "memory": 44139, "step": 3800} +{"lr": 9.780647960398751e-05, "data_time": 0.0012096881866455079, "loss": 0.3510953962802887, "time": 0.630826735496521, "epoch": 1, "memory": 44139, "step": 3900} +{"accuracy/top1": 82.23096466064453, "data_time": 0.04385408602262798, "time": 1.3381989044055604, "step": 1} +{"lr": 9.7685491196646e-05, "data_time": 0.0017798423767089843, "loss": 0.35121058523654936, "time": 0.6321818828582764, "epoch": 2, "memory": 44139, "step": 4007} +{"lr": 9.756956126085676e-05, "data_time": 0.0015968561172485351, "loss": 0.3366678595542908, "time": 0.6321999311447144, "epoch": 2, "memory": 44138, "step": 4107} +{"lr": 9.745087893642482e-05, "data_time": 0.0016131639480590821, "loss": 0.34111056923866273, "time": 0.6321089267730713, "epoch": 2, "memory": 44138, "step": 4207} +{"lr": 9.73294518969056e-05, "data_time": 0.0018948793411254882, "loss": 0.39238911867141724, "time": 0.6330914735794068, "epoch": 2, "memory": 44138, "step": 4307} +{"lr": 9.72052879933172e-05, "data_time": 0.0014326333999633788, "loss": 0.3141148179769516, "time": 0.6321524620056153, "epoch": 2, "memory": 44138, "step": 4407} +{"lr": 9.707839525363358e-05, "data_time": 0.0015880346298217773, "loss": 0.35293120741844175, "time": 0.6328054189682006, "epoch": 2, "memory": 44138, "step": 4507} +{"lr": 9.69487818822646e-05, "data_time": 0.0014887094497680665, "loss": 0.3299088656902313, "time": 0.6326074361801147, "epoch": 2, "memory": 44138, "step": 4607} +{"lr": 9.681645625952616e-05, "data_time": 0.0015972614288330077, "loss": 0.34193129539489747, "time": 0.6329273462295533, "epoch": 2, "memory": 44138, "step": 4707} +{"lr": 9.668142694109826e-05, "data_time": 0.0014832019805908203, "loss": 0.3641922563314438, "time": 0.6330604791641236, "epoch": 2, "memory": 44138, "step": 4807} +{"lr": 9.654370265747162e-05, "data_time": 0.0018642425537109375, "loss": 0.32959853410720824, "time": 0.63274245262146, "epoch": 2, "memory": 44138, "step": 4907} +{"lr": 9.640329231338345e-05, "data_time": 0.0019052267074584962, "loss": 0.35813969671726226, "time": 0.6320257902145385, "epoch": 2, "memory": 44138, "step": 5007} +{"lr": 9.626020498724147e-05, "data_time": 0.0017485618591308594, "loss": 0.3302235662937164, "time": 0.632898998260498, "epoch": 2, "memory": 44138, "step": 5107} +{"lr": 9.611444993053717e-05, "data_time": 0.0018631458282470704, "loss": 0.3023091822862625, "time": 0.6421539783477783, "epoch": 2, "memory": 44138, "step": 5207} +{"lr": 9.596603656724746e-05, "data_time": 0.0016747236251831055, "loss": 0.31570457220077514, "time": 0.6327734231948853, "epoch": 2, "memory": 44138, "step": 5307} +{"lr": 9.58149744932254e-05, "data_time": 0.0016315460205078125, "loss": 0.3265344202518463, "time": 0.6319121599197388, "epoch": 2, "memory": 44138, "step": 5407} +{"lr": 9.566127347557972e-05, "data_time": 0.001626873016357422, "loss": 0.32270238995552064, "time": 0.6323603630065918, "epoch": 2, "memory": 44138, "step": 5507} +{"lr": 9.550494345204367e-05, "data_time": 0.0017716407775878907, "loss": 0.3385892495512962, "time": 0.632108497619629, "epoch": 2, "memory": 44138, "step": 5607} +{"lr": 9.534599453033183e-05, "data_time": 0.0013827800750732422, "loss": 0.33297447562217714, "time": 0.6321195363998413, "epoch": 2, "memory": 44138, "step": 5707} +{"lr": 9.518443698748712e-05, "data_time": 0.0019120216369628907, "loss": 0.2947722613811493, "time": 0.6323132514953613, "epoch": 2, "memory": 44138, "step": 5807} +{"lr": 9.50202812692162e-05, "data_time": 0.001697397232055664, "loss": 0.340665927529335, "time": 0.6326845645904541, "epoch": 2, "memory": 44138, "step": 5907} +{"lr": 9.485353798921386e-05, "data_time": 0.0018517732620239257, "loss": 0.3081132501363754, "time": 0.6327033519744873, "epoch": 2, "memory": 44138, "step": 6007} +{"lr": 9.468421792847712e-05, "data_time": 0.001509237289428711, "loss": 0.34085512459278106, "time": 0.6335102796554566, "epoch": 2, "memory": 44138, "step": 6107} +{"lr": 9.451233203460794e-05, "data_time": 0.016501879692077635, "loss": 0.33548237979412077, "time": 0.6462246656417847, "epoch": 2, "memory": 44138, "step": 6207} +{"lr": 9.433789142110532e-05, "data_time": 0.0015334367752075195, "loss": 0.3299319908022881, "time": 0.6456239700317383, "epoch": 2, "memory": 44138, "step": 6307} +{"lr": 9.416090736664709e-05, "data_time": 0.0016468048095703125, "loss": 0.33380793035030365, "time": 0.6327488899230957, "epoch": 2, "memory": 44138, "step": 6407} +{"lr": 9.398139131436032e-05, "data_time": 0.0018025875091552735, "loss": 0.35275731682777406, "time": 0.6320118188858033, "epoch": 2, "memory": 44138, "step": 6507} +{"lr": 9.379935487108167e-05, "data_time": 0.0017237186431884766, "loss": 0.3359238713979721, "time": 0.6313405752182006, "epoch": 2, "memory": 44138, "step": 6607} +{"lr": 9.361480980660697e-05, "data_time": 0.0015587091445922851, "loss": 0.32569957673549654, "time": 0.6312980651855469, "epoch": 2, "memory": 44138, "step": 6707} +{"lr": 9.342776805292993e-05, "data_time": 0.0017858028411865234, "loss": 0.35612158477306366, "time": 0.6311082124710083, "epoch": 2, "memory": 44138, "step": 6807} +{"lr": 9.323824170347083e-05, "data_time": 0.0015132665634155274, "loss": 0.340649750828743, "time": 0.6312653541564941, "epoch": 2, "memory": 44138, "step": 6907} +{"lr": 9.304624301229471e-05, "data_time": 0.0019707679748535156, "loss": 0.31895554065704346, "time": 0.6313549995422363, "epoch": 2, "memory": 44138, "step": 7007} +{"lr": 9.285178439331897e-05, "data_time": 0.0018143415451049804, "loss": 0.3082291066646576, "time": 0.6324036836624145, "epoch": 2, "memory": 44138, "step": 7107} +{"lr": 9.265487841951087e-05, "data_time": 0.0020533084869384767, "loss": 0.32104428559541703, "time": 0.6320973634719849, "epoch": 2, "memory": 44138, "step": 7207} +{"lr": 9.245553782207421e-05, "data_time": 0.0014142751693725585, "loss": 0.3304094195365906, "time": 0.6371475219726562, "epoch": 2, "memory": 44138, "step": 7307} +{"lr": 9.225377548962671e-05, "data_time": 0.0014187097549438477, "loss": 0.3069656670093536, "time": 0.6318143367767334, "epoch": 2, "memory": 44138, "step": 7407} +{"lr": 9.204960446736601e-05, "data_time": 0.001721811294555664, "loss": 0.31440397799015046, "time": 0.6314618825912476, "epoch": 2, "memory": 44138, "step": 7507} +{"lr": 9.184303795622715e-05, "data_time": 0.002058076858520508, "loss": 0.3164314031600952, "time": 0.6315144062042236, "epoch": 2, "memory": 44138, "step": 7607} +{"lr": 9.163408931202823e-05, "data_time": 0.0019001722335815429, "loss": 0.316406786441803, "time": 0.6327008724212646, "epoch": 2, "memory": 44138, "step": 7707} +{"lr": 9.142277204460729e-05, "data_time": 0.001949620246887207, "loss": 0.30633518695831297, "time": 0.6310815572738647, "epoch": 2, "memory": 44138, "step": 7807} +{"accuracy/top1": 86.54535675048828, "data_time": 0.03482668975303913, "time": 1.2794994937962498, "step": 2} +{"lr": 9.119405492383288e-05, "data_time": 0.0014729261398315429, "loss": 0.3150415584445, "time": 0.631822919845581, "epoch": 3, "memory": 44138, "step": 7914} +{"lr": 9.097787819233205e-05, "data_time": 0.0016135931015014648, "loss": 0.3322458863258362, "time": 0.6319055080413818, "epoch": 3, "memory": 44138, "step": 8014} +{"lr": 9.075937526576627e-05, "data_time": 0.002053642272949219, "loss": 0.3029593124985695, "time": 0.6327821731567382, "epoch": 3, "memory": 44138, "step": 8114} +{"lr": 9.053856027171755e-05, "data_time": 0.0020038366317749025, "loss": 0.3186375737190247, "time": 0.6329057216644287, "epoch": 3, "memory": 44138, "step": 8214} +{"lr": 9.031544748725785e-05, "data_time": 0.0017593622207641602, "loss": 0.3061673820018768, "time": 0.6321552276611329, "epoch": 3, "memory": 44138, "step": 8314} +{"lr": 9.009005133802542e-05, "data_time": 0.0015958547592163086, "loss": 0.28570996820926664, "time": 0.6326016187667847, "epoch": 3, "memory": 44138, "step": 8414} +{"lr": 8.986238639729242e-05, "data_time": 0.0017288923263549805, "loss": 0.3147992193698883, "time": 0.6335033178329468, "epoch": 3, "memory": 44138, "step": 8514} +{"lr": 8.963246738502254e-05, "data_time": 0.001550769805908203, "loss": 0.3290190815925598, "time": 0.6319665431976318, "epoch": 3, "memory": 44138, "step": 8614} +{"lr": 8.940030916691928e-05, "data_time": 0.0015265941619873047, "loss": 0.30940591990947724, "time": 0.6312623262405396, "epoch": 3, "memory": 44138, "step": 8714} +{"lr": 8.91659267534649e-05, "data_time": 0.0015799045562744141, "loss": 0.32590238749980927, "time": 0.6314814805984497, "epoch": 3, "memory": 44138, "step": 8814} +{"lr": 8.892933529894978e-05, "data_time": 0.0014423370361328126, "loss": 0.324916610121727, "time": 0.6310937404632568, "epoch": 3, "memory": 44138, "step": 8914} +{"lr": 8.869055010049283e-05, "data_time": 0.00158236026763916, "loss": 0.3465823620557785, "time": 0.6313714742660522, "epoch": 3, "memory": 44138, "step": 9014} +{"lr": 8.844958659705188e-05, "data_time": 0.001692485809326172, "loss": 0.33284681141376493, "time": 0.6314581394195556, "epoch": 3, "memory": 44138, "step": 9114} +{"lr": 8.82064603684259e-05, "data_time": 0.0017244815826416016, "loss": 0.30471644550561905, "time": 0.6429675579071045, "epoch": 3, "memory": 44138, "step": 9214} +{"lr": 8.796118713424776e-05, "data_time": 0.0017464876174926758, "loss": 0.3103995889425278, "time": 0.632632565498352, "epoch": 3, "memory": 44138, "step": 9314} +{"lr": 8.77137827529676e-05, "data_time": 0.0019337654113769532, "loss": 0.3151204764842987, "time": 0.631326961517334, "epoch": 3, "memory": 44138, "step": 9414} +{"lr": 8.746426322082745e-05, "data_time": 0.0015658140182495117, "loss": 0.3046370282769203, "time": 0.6331208944320679, "epoch": 3, "memory": 44138, "step": 9514} +{"lr": 8.721264467082732e-05, "data_time": 0.0018798351287841798, "loss": 0.29835492074489595, "time": 0.6328758716583252, "epoch": 3, "memory": 44138, "step": 9614} +{"lr": 8.695894337168168e-05, "data_time": 0.0016434669494628906, "loss": 0.3285602807998657, "time": 0.6328323125839234, "epoch": 3, "memory": 44138, "step": 9714} +{"lr": 8.67031757267677e-05, "data_time": 0.0016220808029174805, "loss": 0.28751807659864426, "time": 0.6325338363647461, "epoch": 3, "memory": 44138, "step": 9814} +{"lr": 8.64453582730651e-05, "data_time": 0.0017338991165161133, "loss": 0.3135116085410118, "time": 0.6331475973129272, "epoch": 3, "memory": 44138, "step": 9914} +{"lr": 8.61855076800861e-05, "data_time": 0.001980495452880859, "loss": 0.3006910443305969, "time": 0.6326284408569336, "epoch": 3, "memory": 44138, "step": 10014} +{"lr": 8.592364074879843e-05, "data_time": 0.0018828630447387696, "loss": 0.29880787134170533, "time": 0.6326159238815308, "epoch": 3, "memory": 44138, "step": 10114} +{"lr": 8.565977441053854e-05, "data_time": 0.001508331298828125, "loss": 0.3597894847393036, "time": 0.6325548648834228, "epoch": 3, "memory": 44138, "step": 10214} +{"lr": 8.539392572591724e-05, "data_time": 0.0016089677810668945, "loss": 0.32104514837265014, "time": 0.6325655221939087, "epoch": 3, "memory": 44138, "step": 10314} +{"lr": 8.512611188371633e-05, "data_time": 0.0025423288345336912, "loss": 0.30717981457710264, "time": 0.6336353302001954, "epoch": 3, "memory": 44138, "step": 10414} +{"lr": 8.485635019977725e-05, "data_time": 0.0017320871353149413, "loss": 0.31498719900846484, "time": 0.6323895692825318, "epoch": 3, "memory": 44138, "step": 10514} +{"lr": 8.45846581158818e-05, "data_time": 0.0016342639923095704, "loss": 0.312097804248333, "time": 0.6329158306121826, "epoch": 3, "memory": 44138, "step": 10614} +{"lr": 8.431105319862416e-05, "data_time": 0.0016703605651855469, "loss": 0.35519146025180814, "time": 0.6327781677246094, "epoch": 3, "memory": 44138, "step": 10714} +{"lr": 8.403555313827516e-05, "data_time": 0.001896238327026367, "loss": 0.3117992579936981, "time": 0.6336266040802002, "epoch": 3, "memory": 44138, "step": 10814} +{"lr": 8.375817574763844e-05, "data_time": 0.002271437644958496, "loss": 0.29683640897274016, "time": 0.6317068338394165, "epoch": 3, "memory": 44138, "step": 10914} +{"lr": 8.347893896089871e-05, "data_time": 0.0015553951263427735, "loss": 0.34331596791744234, "time": 0.6320744514465332, "epoch": 3, "memory": 44138, "step": 11014} +{"lr": 8.31978608324625e-05, "data_time": 0.0014697074890136718, "loss": 0.3054161101579666, "time": 0.632155179977417, "epoch": 3, "memory": 44138, "step": 11114} +{"lr": 8.291495953579052e-05, "data_time": 0.0022779464721679687, "loss": 0.3382767826318741, "time": 0.6319034099578857, "epoch": 3, "memory": 44138, "step": 11214} +{"lr": 8.263025336222267e-05, "data_time": 0.001981091499328613, "loss": 0.2996159911155701, "time": 0.631929588317871, "epoch": 3, "memory": 44138, "step": 11314} +{"lr": 8.234376071979551e-05, "data_time": 0.0016178607940673829, "loss": 0.3139187753200531, "time": 0.6318180322647095, "epoch": 3, "memory": 44138, "step": 11414} +{"lr": 8.205550013205187e-05, "data_time": 0.0015609025955200194, "loss": 0.29609236866235733, "time": 0.6319526433944702, "epoch": 3, "memory": 44138, "step": 11514} +{"lr": 8.17654902368435e-05, "data_time": 0.0014995813369750976, "loss": 0.3452329754829407, "time": 0.6313196420669556, "epoch": 3, "memory": 44138, "step": 11614} +{"lr": 8.147374978512582e-05, "data_time": 0.0012678861618041991, "loss": 0.33042644560337064, "time": 0.6313238859176635, "epoch": 3, "memory": 44138, "step": 11714} +{"accuracy/top1": 88.37201690673828, "data_time": 0.03216570410235175, "time": 1.2769481519172932, "step": 3} +{"lr": 8.115969237626578e-05, "data_time": 0.0015007972717285157, "loss": 0.29631754010915756, "time": 0.6331702947616578, "epoch": 4, "memory": 44138, "step": 11821} +{"lr": 8.08644297337245e-05, "data_time": 0.0013971567153930665, "loss": 0.329595622420311, "time": 0.6327028512954712, "epoch": 4, "memory": 44138, "step": 11921} +{"lr": 8.056749479387651e-05, "data_time": 0.0015867948532104492, "loss": 0.3167020112276077, "time": 0.6333897590637207, "epoch": 4, "memory": 44138, "step": 12021} +{"lr": 8.026890675542457e-05, "data_time": 0.0014649629592895508, "loss": 0.31118630319833757, "time": 0.632989764213562, "epoch": 4, "memory": 44138, "step": 12121} +{"lr": 7.996868492395457e-05, "data_time": 0.0014965295791625976, "loss": 0.3052448958158493, "time": 0.6328909397125244, "epoch": 4, "memory": 44138, "step": 12221} +{"lr": 7.966684871068762e-05, "data_time": 0.0013724565505981445, "loss": 0.3271132081747055, "time": 0.6324964046478272, "epoch": 4, "memory": 44138, "step": 12321} +{"lr": 7.936341763122449e-05, "data_time": 0.0020560979843139648, "loss": 0.3197891443967819, "time": 0.6329471826553345, "epoch": 4, "memory": 44138, "step": 12421} +{"lr": 7.905841130428405e-05, "data_time": 0.0015125751495361328, "loss": 0.32204986810684205, "time": 0.6332590103149414, "epoch": 4, "memory": 44138, "step": 12521} +{"lr": 7.875184945043487e-05, "data_time": 0.0012707233428955079, "loss": 0.3328577935695648, "time": 0.6322899103164673, "epoch": 4, "memory": 44138, "step": 12621} +{"lr": 7.844375189082012e-05, "data_time": 0.0017799615859985351, "loss": 0.32825671434402465, "time": 0.6328401565551758, "epoch": 4, "memory": 44138, "step": 12721} +{"lr": 7.813413854587572e-05, "data_time": 0.0017122507095336914, "loss": 0.29628912061452867, "time": 0.635224175453186, "epoch": 4, "memory": 44138, "step": 12821} +{"lr": 7.782302943404291e-05, "data_time": 0.0017526865005493164, "loss": 0.28152785897254945, "time": 0.6331571340560913, "epoch": 4, "memory": 44138, "step": 12921} +{"lr": 7.751044467047337e-05, "data_time": 0.0015371322631835937, "loss": 0.2982440978288651, "time": 0.6328576803207397, "epoch": 4, "memory": 44138, "step": 13021} +{"lr": 7.719640446572907e-05, "data_time": 0.002108931541442871, "loss": 0.31789767146110537, "time": 0.6334184408187866, "epoch": 4, "memory": 44138, "step": 13121} +{"lr": 7.688092912447536e-05, "data_time": 0.0018030166625976562, "loss": 0.3146006315946579, "time": 0.6329826354980469, "epoch": 4, "memory": 44138, "step": 13221} +{"lr": 7.656403904416794e-05, "data_time": 0.001789402961730957, "loss": 0.31028204709291457, "time": 0.6333381652832031, "epoch": 4, "memory": 44138, "step": 13321} +{"lr": 7.624575471373434e-05, "data_time": 0.0021080255508422853, "loss": 0.31789949238300325, "time": 0.6335967540740967, "epoch": 4, "memory": 44138, "step": 13421} +{"lr": 7.592609671224926e-05, "data_time": 0.0016337394714355468, "loss": 0.3074886202812195, "time": 0.6347370624542237, "epoch": 4, "memory": 44138, "step": 13521} +{"lr": 7.560508570760375e-05, "data_time": 0.0015054464340209961, "loss": 0.30995132625102995, "time": 0.6344563722610473, "epoch": 4, "memory": 44138, "step": 13621} +{"lr": 7.528274245516895e-05, "data_time": 0.0017914056777954101, "loss": 0.32072039842605593, "time": 0.633221435546875, "epoch": 4, "memory": 44138, "step": 13721} +{"lr": 7.495908779645405e-05, "data_time": 0.0021927595138549806, "loss": 0.30968077480793, "time": 0.6332769155502319, "epoch": 4, "memory": 44138, "step": 13821} +{"lr": 7.463414265775906e-05, "data_time": 0.002105975151062012, "loss": 0.31177731305360795, "time": 0.6337838411331177, "epoch": 4, "memory": 44138, "step": 13921} +{"lr": 7.430792804882147e-05, "data_time": 0.0015778541564941406, "loss": 0.2964512422680855, "time": 0.6337619304656983, "epoch": 4, "memory": 44138, "step": 14021} +{"lr": 7.398046506145803e-05, "data_time": 0.0016014814376831056, "loss": 0.30565634071826936, "time": 0.6330125570297241, "epoch": 4, "memory": 44138, "step": 14121} +{"lr": 7.365177486820114e-05, "data_time": 0.0016689777374267578, "loss": 0.3190244257450104, "time": 0.6331549167633057, "epoch": 4, "memory": 44138, "step": 14221} +{"lr": 7.332187872092939e-05, "data_time": 0.0019321203231811523, "loss": 0.3105497986078262, "time": 0.6332573890686035, "epoch": 4, "memory": 44138, "step": 14321} +{"lr": 7.299079794949401e-05, "data_time": 0.0018114805221557616, "loss": 0.3333969205617905, "time": 0.6330009937286377, "epoch": 4, "memory": 44138, "step": 14421} +{"lr": 7.265855396033955e-05, "data_time": 0.0012221574783325196, "loss": 0.3279608815908432, "time": 0.6332975864410401, "epoch": 4, "memory": 44138, "step": 14521} +{"lr": 7.232516823512013e-05, "data_time": 0.0015579938888549804, "loss": 0.3360292732715607, "time": 0.6332832098007202, "epoch": 4, "memory": 44138, "step": 14621} +{"lr": 7.199066232930995e-05, "data_time": 0.0019704103469848633, "loss": 0.3111056387424469, "time": 0.6339174985885621, "epoch": 4, "memory": 44138, "step": 14721} +{"lr": 7.165505787080997e-05, "data_time": 0.0017514705657958984, "loss": 0.31476400792598724, "time": 0.6328725337982177, "epoch": 4, "memory": 44138, "step": 14821} +{"lr": 7.13183765585495e-05, "data_time": 0.002126479148864746, "loss": 0.31744611263275146, "time": 0.6325232028961182, "epoch": 4, "memory": 44138, "step": 14921} +{"lr": 7.098064016108317e-05, "data_time": 0.0018365859985351562, "loss": 0.2819396436214447, "time": 0.6314771175384521, "epoch": 4, "memory": 44138, "step": 15021} +{"lr": 7.064187051518379e-05, "data_time": 0.0018094778060913086, "loss": 0.3287270963191986, "time": 0.6328166723251343, "epoch": 4, "memory": 44138, "step": 15121} +{"lr": 7.030208952442949e-05, "data_time": 0.0019420385360717773, "loss": 0.3135231643915176, "time": 0.6320729494094849, "epoch": 4, "memory": 44138, "step": 15221} +{"lr": 6.996131915778851e-05, "data_time": 0.0018641948699951172, "loss": 0.28980848640203477, "time": 0.6325138807296753, "epoch": 4, "memory": 44138, "step": 15321} +{"lr": 6.961958144819853e-05, "data_time": 0.001683354377746582, "loss": 0.30070672035217283, "time": 0.6332222700119019, "epoch": 4, "memory": 44138, "step": 15421} +{"lr": 6.927689849114193e-05, "data_time": 0.0016742944717407227, "loss": 0.3087885230779648, "time": 0.633359694480896, "epoch": 4, "memory": 44138, "step": 15521} +{"lr": 6.893329244321722e-05, "data_time": 0.0014219999313354493, "loss": 0.29107152819633486, "time": 0.6310092449188233, "epoch": 4, "memory": 44138, "step": 15621} +{"accuracy/top1": 89.60022735595703, "data_time": 0.033523913087516, "time": 1.2778359528245598, "step": 4} +{"lr": 6.856463687361792e-05, "data_time": 0.001884150505065918, "loss": 0.2845174640417099, "time": 0.6339736461639405, "epoch": 5, "memory": 44138, "step": 15728} +{"lr": 6.821919068470432e-05, "data_time": 0.00172879695892334, "loss": 0.32641951739788055, "time": 0.6340963840484619, "epoch": 5, "memory": 44138, "step": 15828} +{"lr": 6.787288979235407e-05, "data_time": 0.0016388177871704101, "loss": 0.30662190318107607, "time": 0.6318677425384521, "epoch": 5, "memory": 44138, "step": 15928} +{"lr": 6.752575658708793e-05, "data_time": 0.0016812562942504882, "loss": 0.28915023654699323, "time": 0.6328672409057617, "epoch": 5, "memory": 44138, "step": 16028} +{"lr": 6.71778135132408e-05, "data_time": 0.0017188549041748046, "loss": 0.3077644228935242, "time": 0.6326217889785767, "epoch": 5, "memory": 44138, "step": 16128} +{"lr": 6.682908306751074e-05, "data_time": 0.0016791105270385742, "loss": 0.31356306076049806, "time": 0.6329569816589355, "epoch": 5, "memory": 44138, "step": 16228} +{"lr": 6.64795877975043e-05, "data_time": 0.00176849365234375, "loss": 0.2925348594784737, "time": 0.6409899950027466, "epoch": 5, "memory": 44138, "step": 16328} +{"lr": 6.61293503002787e-05, "data_time": 0.0016947746276855468, "loss": 0.29854719489812853, "time": 0.633159875869751, "epoch": 5, "memory": 44138, "step": 16428} +{"lr": 6.577839322088084e-05, "data_time": 0.001694035530090332, "loss": 0.3069661006331444, "time": 0.6331403255462646, "epoch": 5, "memory": 44138, "step": 16528} +{"lr": 6.542673925088298e-05, "data_time": 0.0016147613525390625, "loss": 0.310231477022171, "time": 0.6331608057022095, "epoch": 5, "memory": 44138, "step": 16628} +{"lr": 6.507441112691584e-05, "data_time": 0.0018262624740600585, "loss": 0.3002298802137375, "time": 0.6339569330215454, "epoch": 5, "memory": 44138, "step": 16728} +{"lr": 6.47214316291983e-05, "data_time": 0.0015388727188110352, "loss": 0.30747975409030914, "time": 0.6342548847198486, "epoch": 5, "memory": 44138, "step": 16828} +{"lr": 6.436782358006481e-05, "data_time": 0.001520848274230957, "loss": 0.3245810091495514, "time": 0.6329957962036132, "epoch": 5, "memory": 44138, "step": 16928} +{"lr": 6.40136098424896e-05, "data_time": 0.0016062259674072266, "loss": 0.30905329883098603, "time": 0.6319875001907349, "epoch": 5, "memory": 44138, "step": 17028} +{"lr": 6.365881331860828e-05, "data_time": 0.0016116619110107422, "loss": 0.2932494878768921, "time": 0.6321008443832398, "epoch": 5, "memory": 44138, "step": 17128} +{"lr": 6.330345694823739e-05, "data_time": 0.0018414974212646485, "loss": 0.3366967052221298, "time": 0.6327253103256225, "epoch": 5, "memory": 44138, "step": 17228} +{"lr": 6.29475637073908e-05, "data_time": 0.002088785171508789, "loss": 0.2893759086728096, "time": 0.6335022449493408, "epoch": 5, "memory": 44138, "step": 17328} +{"lr": 6.259115660679482e-05, "data_time": 0.0016201496124267577, "loss": 0.31572250127792356, "time": 0.632670259475708, "epoch": 5, "memory": 44138, "step": 17428} +{"lr": 6.223425869039963e-05, "data_time": 0.0020895004272460938, "loss": 0.29802556335926056, "time": 0.633821177482605, "epoch": 5, "memory": 44138, "step": 17528} +{"lr": 6.187689303389006e-05, "data_time": 0.0015899658203125, "loss": 0.28801448047161105, "time": 0.6328324317932129, "epoch": 5, "memory": 44138, "step": 17628} +{"lr": 6.151908274319292e-05, "data_time": 0.0019217491149902343, "loss": 0.30075431764125826, "time": 0.6327396869659424, "epoch": 5, "memory": 44138, "step": 17728} +{"lr": 6.116085095298373e-05, "data_time": 0.0018248558044433594, "loss": 0.28984824419021604, "time": 0.6323301553726196, "epoch": 5, "memory": 44138, "step": 17828} +{"lr": 6.080222082519007e-05, "data_time": 0.0015449285507202148, "loss": 0.30669480562210083, "time": 0.6335581541061401, "epoch": 5, "memory": 44138, "step": 17928} +{"lr": 6.0443215547495147e-05, "data_time": 0.0016098737716674805, "loss": 0.2870576292276382, "time": 0.6318952560424804, "epoch": 5, "memory": 44138, "step": 18028} +{"lr": 6.0083858331837714e-05, "data_time": 0.0019602060317993166, "loss": 0.2967754065990448, "time": 0.6327675819396973, "epoch": 5, "memory": 44138, "step": 18128} +{"lr": 5.97241724129116e-05, "data_time": 0.0016579627990722656, "loss": 0.29930940419435503, "time": 0.6322240591049194, "epoch": 5, "memory": 44138, "step": 18228} +{"lr": 5.936418104666328e-05, "data_time": 0.0015895843505859375, "loss": 0.30931147634983064, "time": 0.6318528652191162, "epoch": 5, "memory": 44138, "step": 18328} +{"lr": 5.9003907508788295e-05, "data_time": 0.001702737808227539, "loss": 0.3269601136445999, "time": 0.632736325263977, "epoch": 5, "memory": 44138, "step": 18428} +{"lr": 5.864337509322633e-05, "data_time": 0.0016469240188598632, "loss": 0.3123019427061081, "time": 0.6319441318511962, "epoch": 5, "memory": 44138, "step": 18528} +{"lr": 5.8282607110655194e-05, "data_time": 0.0019517898559570312, "loss": 0.3098306179046631, "time": 0.6324620962142944, "epoch": 5, "memory": 44138, "step": 18628} +{"lr": 5.7921626886983564e-05, "data_time": 0.002015233039855957, "loss": 0.3117759957909584, "time": 0.632539701461792, "epoch": 5, "memory": 44138, "step": 18728} +{"lr": 5.756045776184287e-05, "data_time": 0.0015504121780395507, "loss": 0.3262644737958908, "time": 0.6331888914108277, "epoch": 5, "memory": 44138, "step": 18828} +{"lr": 5.719912308707808e-05, "data_time": 0.0015547513961791993, "loss": 0.30534543097019196, "time": 0.6331577062606811, "epoch": 5, "memory": 44138, "step": 18928} +{"lr": 5.683764622523808e-05, "data_time": 0.0015584230422973633, "loss": 0.31962490975856783, "time": 0.632534408569336, "epoch": 5, "memory": 44138, "step": 19028} +{"lr": 5.647605054806502e-05, "data_time": 0.0019353628158569336, "loss": 0.29977654814720156, "time": 0.6329171180725097, "epoch": 5, "memory": 44138, "step": 19128} +{"lr": 5.6114359434983206e-05, "data_time": 0.0030315876007080077, "loss": 0.29852366000413894, "time": 0.6360118865966797, "epoch": 5, "memory": 44138, "step": 19228} +{"lr": 5.575259627158745e-05, "data_time": 0.002434039115905762, "loss": 0.32444908618927004, "time": 0.6391826391220092, "epoch": 5, "memory": 44138, "step": 19328} +{"lr": 5.539078444813113e-05, "data_time": 0.0018876075744628906, "loss": 0.27897868752479554, "time": 0.6362127542495728, "epoch": 5, "memory": 44138, "step": 19428} +{"lr": 5.5028947358013735e-05, "data_time": 0.0014713525772094727, "loss": 0.30247158110141753, "time": 0.6336456060409545, "epoch": 5, "memory": 44138, "step": 19528} +{"accuracy/top1": 89.95511627197266, "data_time": 0.03747370325285813, "time": 1.2829449217894981, "step": 5} +{"lr": 5.464178020340108e-05, "data_time": 0.0018750667572021485, "loss": 0.293927937746048, "time": 0.6334022521972656, "epoch": 6, "memory": 44138, "step": 19635} +{"lr": 5.427996514795046e-05, "data_time": 0.0016377687454223633, "loss": 0.34120202660560606, "time": 0.6334142208099365, "epoch": 6, "memory": 44138, "step": 19735} +{"lr": 5.3918196647260834e-05, "data_time": 0.0027139902114868162, "loss": 0.2941610261797905, "time": 0.6372280836105346, "epoch": 6, "memory": 44138, "step": 19835} +{"lr": 5.355649809193057e-05, "data_time": 0.0017122507095336914, "loss": 0.3073193609714508, "time": 0.6329587936401367, "epoch": 6, "memory": 44138, "step": 19935} +{"lr": 5.3194892868035765e-05, "data_time": 0.002005934715270996, "loss": 0.32290321588516235, "time": 0.6343416213989258, "epoch": 6, "memory": 44138, "step": 20035} +{"lr": 5.283340435561795e-05, "data_time": 0.0023032665252685548, "loss": 0.3329052716493607, "time": 0.6344602108001709, "epoch": 6, "memory": 44138, "step": 20135} +{"lr": 5.2472055927172495e-05, "data_time": 0.001950240135192871, "loss": 0.2771913021802902, "time": 0.634003472328186, "epoch": 6, "memory": 44138, "step": 20235} +{"lr": 5.2110870946137605e-05, "data_time": 0.0028927087783813476, "loss": 0.31453161835670473, "time": 0.6333108425140381, "epoch": 6, "memory": 44138, "step": 20335} +{"lr": 5.174987276538342e-05, "data_time": 0.0016325950622558595, "loss": 0.3118656262755394, "time": 0.6319082021713257, "epoch": 6, "memory": 44138, "step": 20435} +{"lr": 5.1389084725702425e-05, "data_time": 0.0027080774307250977, "loss": 0.29839026927948, "time": 0.6348451375961304, "epoch": 6, "memory": 44138, "step": 20535} +{"lr": 5.1028530154300045e-05, "data_time": 0.0021979570388793944, "loss": 0.2979643076658249, "time": 0.6355926752090454, "epoch": 6, "memory": 44138, "step": 20635} +{"lr": 5.066823236328663e-05, "data_time": 0.0018099069595336914, "loss": 0.3005822032690048, "time": 0.6334505081176758, "epoch": 6, "memory": 44138, "step": 20735} +{"lr": 5.030821464817006e-05, "data_time": 0.0017623424530029297, "loss": 0.30971167981624603, "time": 0.6332575559616089, "epoch": 6, "memory": 44138, "step": 20835} +{"lr": 4.99485002863494e-05, "data_time": 0.002543354034423828, "loss": 0.3223777174949646, "time": 0.6341951847076416, "epoch": 6, "memory": 44138, "step": 20935} +{"lr": 4.95891125356102e-05, "data_time": 0.0016045570373535156, "loss": 0.305197237432003, "time": 0.6353859663009643, "epoch": 6, "memory": 44138, "step": 21035} +{"lr": 4.923007463262053e-05, "data_time": 0.0020807266235351564, "loss": 0.2932621449232101, "time": 0.6345109701156616, "epoch": 6, "memory": 44138, "step": 21135} +{"lr": 4.887140979142852e-05, "data_time": 0.0022014379501342773, "loss": 0.2998196542263031, "time": 0.6330170869827271, "epoch": 6, "memory": 44138, "step": 21235} +{"lr": 4.851314120196161e-05, "data_time": 0.0019437313079833985, "loss": 0.2937133401632309, "time": 0.6325894355773926, "epoch": 6, "memory": 44138, "step": 21335} +{"lr": 4.815529202852693e-05, "data_time": 0.001574397087097168, "loss": 0.3261815220117569, "time": 0.6337750673294067, "epoch": 6, "memory": 44138, "step": 21435} +{"lr": 4.779788540831406e-05, "data_time": 0.0024583101272583007, "loss": 0.28951150476932525, "time": 0.6366761922836304, "epoch": 6, "memory": 44138, "step": 21535} +{"lr": 4.744094444989843e-05, "data_time": 0.002363729476928711, "loss": 0.31034710109233854, "time": 0.6336637735366821, "epoch": 6, "memory": 44138, "step": 21635} +{"lr": 4.708449223174764e-05, "data_time": 0.004822397232055664, "loss": 0.2788639008998871, "time": 0.6383045196533204, "epoch": 6, "memory": 44138, "step": 21735} +{"lr": 4.672855180072916e-05, "data_time": 0.001599574089050293, "loss": 0.31344456523656844, "time": 0.6327584028244019, "epoch": 6, "memory": 44138, "step": 21835} +{"lr": 4.6373146170620256e-05, "data_time": 0.002849841117858887, "loss": 0.28745528757572175, "time": 0.6367928266525269, "epoch": 6, "memory": 44138, "step": 21935} +{"lr": 4.6018298320619915e-05, "data_time": 0.004253745079040527, "loss": 0.3103777587413788, "time": 0.6370688676834106, "epoch": 6, "memory": 44138, "step": 22035} +{"lr": 4.566403119386318e-05, "data_time": 0.0017075777053833009, "loss": 0.3060238495469093, "time": 0.6330570936203003, "epoch": 6, "memory": 44138, "step": 22135} +{"lr": 4.531036769593769e-05, "data_time": 0.0018994808197021484, "loss": 0.3336349785327911, "time": 0.6324459552764893, "epoch": 6, "memory": 44138, "step": 22235} +{"lr": 4.4957330693402685e-05, "data_time": 0.0031962871551513674, "loss": 0.34569916427135466, "time": 0.636173152923584, "epoch": 6, "memory": 44138, "step": 22335} +{"lr": 4.460494301231059e-05, "data_time": 0.002243232727050781, "loss": 0.313540580868721, "time": 0.6330457925796509, "epoch": 6, "memory": 44138, "step": 22435} +{"lr": 4.425322743673109e-05, "data_time": 0.0016682863235473633, "loss": 0.31509087830781934, "time": 0.6337332963943482, "epoch": 6, "memory": 44138, "step": 22535} +{"lr": 4.390220670727812e-05, "data_time": 0.0017115354537963867, "loss": 0.3199805974960327, "time": 0.6340263605117797, "epoch": 6, "memory": 44138, "step": 22635} +{"lr": 4.3551903519639325e-05, "data_time": 0.0016652584075927735, "loss": 0.30514743030071256, "time": 0.6322040319442749, "epoch": 6, "memory": 44138, "step": 22735} +{"lr": 4.3202340523108894e-05, "data_time": 0.0025270462036132814, "loss": 0.3206787109375, "time": 0.6345431566238403, "epoch": 6, "memory": 44138, "step": 22835} +{"lr": 4.285354031912302e-05, "data_time": 0.0025682687759399415, "loss": 0.2973847329616547, "time": 0.632455563545227, "epoch": 6, "memory": 44138, "step": 22935} +{"lr": 4.2505525459798544e-05, "data_time": 0.0018535852432250977, "loss": 0.28204504400491714, "time": 0.6344072103500367, "epoch": 6, "memory": 44138, "step": 23035} +{"lr": 4.2158318446474836e-05, "data_time": 0.0020078420639038086, "loss": 0.30542626082897184, "time": 0.6473349571228028, "epoch": 6, "memory": 44138, "step": 23135} +{"lr": 4.181194172825905e-05, "data_time": 0.001706695556640625, "loss": 0.3048213630914688, "time": 0.6328002691268921, "epoch": 6, "memory": 44138, "step": 23235} +{"lr": 4.146641770057448e-05, "data_time": 0.001653003692626953, "loss": 0.2996079549193382, "time": 0.6333706140518188, "epoch": 6, "memory": 44138, "step": 23335} +{"lr": 4.112176870371267e-05, "data_time": 0.0014334440231323243, "loss": 0.30148945450782777, "time": 0.6319171190261841, "epoch": 6, "memory": 44138, "step": 23435} +{"accuracy/top1": 90.24738311767578, "data_time": 0.03907531294329413, "time": 1.2821516785128364, "step": 6} +{"lr": 4.0753988582732786e-05, "data_time": 0.0024433135986328125, "loss": 0.32810259610414505, "time": 0.6333051443099975, "epoch": 7, "memory": 44138, "step": 23542} +{"lr": 4.0411221639286815e-05, "data_time": 0.0015944480895996095, "loss": 0.33431318402290344, "time": 0.6316713571548462, "epoch": 7, "memory": 44138, "step": 23642} +{"lr": 4.006939795169518e-05, "data_time": 0.0022157669067382813, "loss": 0.30837578773498536, "time": 0.6351241827011108, "epoch": 7, "memory": 44138, "step": 23742} +{"lr": 3.972853962099913e-05, "data_time": 0.0030332088470458986, "loss": 0.316980317234993, "time": 0.6349954843521118, "epoch": 7, "memory": 44138, "step": 23842} +{"lr": 3.93886686858237e-05, "data_time": 0.0020666122436523438, "loss": 0.29575823396444323, "time": 0.6344683408737183, "epoch": 7, "memory": 44138, "step": 23942} +{"lr": 3.904980712095259e-05, "data_time": 0.002402949333190918, "loss": 0.3236956983804703, "time": 0.6339280843734741, "epoch": 7, "memory": 44138, "step": 24042} +{"lr": 3.871197683590748e-05, "data_time": 0.0018492698669433593, "loss": 0.2733908072113991, "time": 0.6340746402740478, "epoch": 7, "memory": 44138, "step": 24142} +{"lr": 3.837519967353125e-05, "data_time": 0.0019272327423095702, "loss": 0.323754620552063, "time": 0.6356455564498902, "epoch": 7, "memory": 44138, "step": 24242} +{"lr": 3.8039497408575855e-05, "data_time": 0.001482105255126953, "loss": 0.2949719801545143, "time": 0.6321333408355713, "epoch": 7, "memory": 44138, "step": 24342} +{"lr": 3.770489174629432e-05, "data_time": 0.0016910314559936523, "loss": 0.3118149697780609, "time": 0.6324535369873047, "epoch": 7, "memory": 44138, "step": 24442} +{"lr": 3.737140432103754e-05, "data_time": 0.0018266201019287109, "loss": 0.31530108451843264, "time": 0.6332569360733032, "epoch": 7, "memory": 44138, "step": 24542} +{"lr": 3.703905669485545e-05, "data_time": 0.001881718635559082, "loss": 0.32408997416496277, "time": 0.632784104347229, "epoch": 7, "memory": 44138, "step": 24642} +{"lr": 3.6707870356102706e-05, "data_time": 0.0022163152694702148, "loss": 0.3212019979953766, "time": 0.6322555065155029, "epoch": 7, "memory": 44138, "step": 24742} +{"lr": 3.6377866718049484e-05, "data_time": 0.001720571517944336, "loss": 0.30861849188804624, "time": 0.6450377225875854, "epoch": 7, "memory": 44138, "step": 24842} +{"lr": 3.6049067117496965e-05, "data_time": 0.001851654052734375, "loss": 0.3103115975856781, "time": 0.6321022748947144, "epoch": 7, "memory": 44138, "step": 24942} +{"lr": 3.5721492813397755e-05, "data_time": 0.0025243759155273438, "loss": 0.29069556295871735, "time": 0.6345580577850342, "epoch": 7, "memory": 44138, "step": 25042} +{"lr": 3.539516498548151e-05, "data_time": 0.0021893978118896484, "loss": 0.3239056050777435, "time": 0.6321740627288819, "epoch": 7, "memory": 44138, "step": 25142} +{"lr": 3.507010473288518e-05, "data_time": 0.0022469520568847655, "loss": 0.32012275606393814, "time": 0.6339623212814331, "epoch": 7, "memory": 44138, "step": 25242} +{"lr": 3.474633307278909e-05, "data_time": 0.002277231216430664, "loss": 0.2930764466524124, "time": 0.634464693069458, "epoch": 7, "memory": 44138, "step": 25342} +{"lr": 3.4423870939058177e-05, "data_time": 0.0014679431915283203, "loss": 0.30075865536928176, "time": 0.6319833517074585, "epoch": 7, "memory": 44138, "step": 25442} +{"lr": 3.410273918088805e-05, "data_time": 0.002899646759033203, "loss": 0.2855075106024742, "time": 0.6339532613754273, "epoch": 7, "memory": 44138, "step": 25542} +{"lr": 3.37829585614574e-05, "data_time": 0.0033405780792236327, "loss": 0.31129820346832277, "time": 0.6374561548233032, "epoch": 7, "memory": 44138, "step": 25642} +{"lr": 3.3464549756585216e-05, "data_time": 0.002168416976928711, "loss": 0.29246336668729783, "time": 0.6334645986557007, "epoch": 7, "memory": 44138, "step": 25742} +{"lr": 3.314753335339416e-05, "data_time": 0.001820683479309082, "loss": 0.3041336938738823, "time": 0.6351238489151001, "epoch": 7, "memory": 44138, "step": 25842} +{"lr": 3.2831929848979394e-05, "data_time": 0.0015471696853637696, "loss": 0.3046130836009979, "time": 0.6323554515838623, "epoch": 7, "memory": 44138, "step": 25942} +{"lr": 3.251775964908332e-05, "data_time": 0.0022123098373413087, "loss": 0.3483428418636322, "time": 0.6425493478775024, "epoch": 7, "memory": 44138, "step": 26042} +{"lr": 3.22050430667762e-05, "data_time": 0.0025757551193237305, "loss": 0.30965872704982755, "time": 0.6365681886672974, "epoch": 7, "memory": 44138, "step": 26142} +{"lr": 3.189380032114285e-05, "data_time": 0.0017028331756591796, "loss": 0.3338813304901123, "time": 0.6443053960800171, "epoch": 7, "memory": 44138, "step": 26242} +{"lr": 3.1584051535975324e-05, "data_time": 0.001659083366394043, "loss": 0.2889284402132034, "time": 0.632770323753357, "epoch": 7, "memory": 44138, "step": 26342} +{"lr": 3.127581673847175e-05, "data_time": 0.0015621185302734375, "loss": 0.32617907524108886, "time": 0.6320864439010621, "epoch": 7, "memory": 44138, "step": 26442} +{"lr": 3.0969115857941555e-05, "data_time": 0.001957201957702637, "loss": 0.3024232774972916, "time": 0.6326895952224731, "epoch": 7, "memory": 44138, "step": 26542} +{"lr": 3.0663968724516654e-05, "data_time": 0.001799631118774414, "loss": 0.3251842945814133, "time": 0.6321155786514282, "epoch": 7, "memory": 44138, "step": 26642} +{"lr": 3.0360395067869668e-05, "data_time": 0.0016460657119750977, "loss": 0.2930247038602829, "time": 0.6321796178817749, "epoch": 7, "memory": 44138, "step": 26742} +{"lr": 3.0058414515938063e-05, "data_time": 0.0021967649459838866, "loss": 0.30422474443912506, "time": 0.6380234718322754, "epoch": 7, "memory": 44138, "step": 26842} +{"lr": 2.975804659365501e-05, "data_time": 0.002940821647644043, "loss": 0.315599724650383, "time": 0.6333201408386231, "epoch": 7, "memory": 44138, "step": 26942} +{"lr": 2.945931072168714e-05, "data_time": 0.0017727375030517577, "loss": 0.29845210313797, "time": 0.6331922769546509, "epoch": 7, "memory": 44138, "step": 27042} +{"lr": 2.9162226215178855e-05, "data_time": 0.004553484916687012, "loss": 0.31467942893505096, "time": 0.6411008596420288, "epoch": 7, "memory": 44138, "step": 27142} +{"lr": 2.8866812282503322e-05, "data_time": 0.0033054351806640625, "loss": 0.32417497634887693, "time": 0.6332964181900025, "epoch": 7, "memory": 44138, "step": 27242} +{"lr": 2.8573088024020838e-05, "data_time": 0.0015387535095214844, "loss": 0.31645538806915285, "time": 0.6327661275863647, "epoch": 7, "memory": 44138, "step": 27342} +{"accuracy/top1": 89.67329406738281, "data_time": 0.04145223929964263, "time": 1.2863057029658351, "step": 7} +{"lr": 2.8260695817390092e-05, "data_time": 0.001715683937072754, "loss": 0.29987170696258547, "time": 0.6315726280212403, "epoch": 8, "memory": 44138, "step": 27449} +{"lr": 2.7970529402570967e-05, "data_time": 0.0019087076187133789, "loss": 0.3439913898706436, "time": 0.6331199645996094, "epoch": 8, "memory": 44138, "step": 27549} +{"lr": 2.7682110612244214e-05, "data_time": 0.0015655517578125, "loss": 0.3048872321844101, "time": 0.6332799434661865, "epoch": 8, "memory": 44138, "step": 27649} +{"lr": 2.7395458094490283e-05, "data_time": 0.002020597457885742, "loss": 0.3210868567228317, "time": 0.6331084489822387, "epoch": 8, "memory": 44138, "step": 27749} +{"lr": 2.7110590383189015e-05, "data_time": 0.0016994714736938477, "loss": 0.30213554948568344, "time": 0.6323690414428711, "epoch": 8, "memory": 44138, "step": 27849} +{"lr": 2.6827525896821397e-05, "data_time": 0.0020955562591552734, "loss": 0.3053766623139381, "time": 0.633253026008606, "epoch": 8, "memory": 44138, "step": 27949} +{"lr": 2.6546282937278552e-05, "data_time": 0.0016840696334838867, "loss": 0.3279667258262634, "time": 0.6351157426834106, "epoch": 8, "memory": 44138, "step": 28049} +{"lr": 2.6266879688678606e-05, "data_time": 0.0016275405883789062, "loss": 0.31794826686382294, "time": 0.6349692583084107, "epoch": 8, "memory": 44138, "step": 28149} +{"lr": 2.5989334216190738e-05, "data_time": 0.0020150184631347657, "loss": 0.3034849941730499, "time": 0.634264874458313, "epoch": 8, "memory": 44138, "step": 28249} +{"lr": 2.571366446486732e-05, "data_time": 0.0028668880462646485, "loss": 0.31111868768930434, "time": 0.6349168300628663, "epoch": 8, "memory": 44138, "step": 28349} +{"lr": 2.5439888258483665e-05, "data_time": 0.003535032272338867, "loss": 0.3268710315227509, "time": 0.6354778051376343, "epoch": 8, "memory": 44138, "step": 28449} +{"lr": 2.5168023298385384e-05, "data_time": 0.0016741752624511719, "loss": 0.28083030581474305, "time": 0.6329811573028564, "epoch": 8, "memory": 44138, "step": 28549} +{"lr": 2.489808716234412e-05, "data_time": 0.0025573492050170897, "loss": 0.30325927436351774, "time": 0.6332638740539551, "epoch": 8, "memory": 44138, "step": 28649} +{"lr": 2.463009730342111e-05, "data_time": 0.001484251022338867, "loss": 0.27094477862119676, "time": 0.632333779335022, "epoch": 8, "memory": 44138, "step": 28749} +{"lr": 2.436407104883849e-05, "data_time": 0.0029922246932983397, "loss": 0.3023088127374649, "time": 0.6395461320877075, "epoch": 8, "memory": 44138, "step": 28849} +{"lr": 2.4100025598859092e-05, "data_time": 0.0019676923751831056, "loss": 0.2781511783599854, "time": 0.636028528213501, "epoch": 8, "memory": 44138, "step": 28949} +{"lr": 2.3837978025674284e-05, "data_time": 0.0018792152404785156, "loss": 0.3274871453642845, "time": 0.6333925008773804, "epoch": 8, "memory": 44138, "step": 29049} +{"lr": 2.357794527230027e-05, "data_time": 0.001748943328857422, "loss": 0.3072037994861603, "time": 0.6330334424972535, "epoch": 8, "memory": 44138, "step": 29149} +{"lr": 2.33199441514825e-05, "data_time": 0.0018338680267333985, "loss": 0.30363257229328156, "time": 0.634219479560852, "epoch": 8, "memory": 44138, "step": 29249} +{"lr": 2.306399134460859e-05, "data_time": 0.0021389007568359377, "loss": 0.2911437630653381, "time": 0.6338193416595459, "epoch": 8, "memory": 44138, "step": 29349} +{"lr": 2.281010340062995e-05, "data_time": 0.0018537044525146484, "loss": 0.2986374318599701, "time": 0.6339231014251709, "epoch": 8, "memory": 44138, "step": 29449} +{"lr": 2.2558296734991603e-05, "data_time": 0.0014921188354492187, "loss": 0.2997046634554863, "time": 0.6326585531234741, "epoch": 8, "memory": 44138, "step": 29549} +{"lr": 2.2308587628570915e-05, "data_time": 0.001764965057373047, "loss": 0.3540403783321381, "time": 0.6374994516372681, "epoch": 8, "memory": 44138, "step": 29649} +{"lr": 2.2060992226624955e-05, "data_time": 0.002456521987915039, "loss": 0.30933377593755723, "time": 0.6361114978790283, "epoch": 8, "memory": 44138, "step": 29749} +{"lr": 2.181552653774648e-05, "data_time": 0.0018098592758178712, "loss": 0.2852427214384079, "time": 0.6322941064834595, "epoch": 8, "memory": 44138, "step": 29849} +{"lr": 2.1572206432829035e-05, "data_time": 0.0020434379577636717, "loss": 0.33065275847911835, "time": 0.6344693183898926, "epoch": 8, "memory": 44138, "step": 29949} +{"lr": 2.1331047644040754e-05, "data_time": 0.0018612146377563477, "loss": 0.3068340063095093, "time": 0.6406064033508301, "epoch": 8, "memory": 44138, "step": 30049} +{"lr": 2.109206576380705e-05, "data_time": 0.0019402027130126954, "loss": 0.2755319714546204, "time": 0.6329137802124023, "epoch": 8, "memory": 44138, "step": 30149} +{"lr": 2.0855276243802687e-05, "data_time": 0.001954793930053711, "loss": 0.2961136922240257, "time": 0.6327996015548706, "epoch": 8, "memory": 44138, "step": 30249} +{"lr": 2.0620694393952487e-05, "data_time": 0.005257797241210937, "loss": 0.31215171962976457, "time": 0.6379446744918823, "epoch": 8, "memory": 44138, "step": 30349} +{"lr": 2.038833538144169e-05, "data_time": 0.0018012523651123047, "loss": 0.3066970407962799, "time": 0.6348145484924317, "epoch": 8, "memory": 44138, "step": 30449} +{"lr": 2.0158214229735154e-05, "data_time": 0.0019438982009887696, "loss": 0.32797242105007174, "time": 0.6333035707473755, "epoch": 8, "memory": 44138, "step": 30549} +{"lr": 1.9930345817605977e-05, "data_time": 0.0016388416290283204, "loss": 0.29934488534927367, "time": 0.631968903541565, "epoch": 8, "memory": 44138, "step": 30649} +{"lr": 1.970474487817358e-05, "data_time": 0.0015523672103881837, "loss": 0.3224214121699333, "time": 0.632461428642273, "epoch": 8, "memory": 44138, "step": 30749} +{"lr": 1.9481425997951092e-05, "data_time": 0.0019828081130981445, "loss": 0.3191940948367119, "time": 0.6330483436584473, "epoch": 8, "memory": 44138, "step": 30849} +{"lr": 1.9260403615902187e-05, "data_time": 0.0022492170333862304, "loss": 0.2850849464535713, "time": 0.634054708480835, "epoch": 8, "memory": 44138, "step": 30949} +{"lr": 1.9041692022507647e-05, "data_time": 0.0022660493850708008, "loss": 0.3149334788322449, "time": 0.6329516887664794, "epoch": 8, "memory": 44138, "step": 31049} +{"lr": 1.8825305358841125e-05, "data_time": 0.002373361587524414, "loss": 0.33882727921009065, "time": 0.6360174179077148, "epoch": 8, "memory": 44138, "step": 31149} +{"lr": 1.861125761565518e-05, "data_time": 0.0013271331787109374, "loss": 0.28714269697666167, "time": 0.6328182458877564, "epoch": 8, "memory": 44138, "step": 31249} +{"accuracy/top1": 89.25924682617188, "data_time": 0.03800478063780686, "time": 1.2828440871731988, "step": 8} +{"lr": 1.8384832449122864e-05, "data_time": 0.002848911285400391, "loss": 0.3251998543739319, "time": 0.634513258934021, "epoch": 9, "memory": 44138, "step": 31356} +{"lr": 1.817567007332114e-05, "data_time": 0.0018283367156982423, "loss": 0.320851793885231, "time": 0.6322673559188843, "epoch": 9, "memory": 44138, "step": 31456} +{"lr": 1.796888862098825e-05, "data_time": 0.0019241809844970704, "loss": 0.32219129502773286, "time": 0.6350726842880249, "epoch": 9, "memory": 44138, "step": 31556} +{"lr": 1.7764501461839678e-05, "data_time": 0.0019450902938842774, "loss": 0.2998438149690628, "time": 0.6326807737350464, "epoch": 9, "memory": 44138, "step": 31656} +{"lr": 1.7562521810784746e-05, "data_time": 0.0019226312637329102, "loss": 0.2855839133262634, "time": 0.6332160472869873, "epoch": 9, "memory": 44138, "step": 31756} +{"lr": 1.7362962727072444e-05, "data_time": 0.0023490428924560548, "loss": 0.3230756729841232, "time": 0.6363595724105835, "epoch": 9, "memory": 44138, "step": 31856} +{"lr": 1.716583711344682e-05, "data_time": 0.0018647193908691406, "loss": 0.2942240074276924, "time": 0.6362556457519531, "epoch": 9, "memory": 44138, "step": 31956} +{"lr": 1.6971157715312954e-05, "data_time": 0.002173614501953125, "loss": 0.31122791171073916, "time": 0.6339544534683228, "epoch": 9, "memory": 44138, "step": 32056} +{"lr": 1.677893711991274e-05, "data_time": 0.0017704725265502929, "loss": 0.28555778712034224, "time": 0.6361353158950805, "epoch": 9, "memory": 44138, "step": 32156} +{"lr": 1.6589187755511025e-05, "data_time": 0.0017292261123657226, "loss": 0.3068491667509079, "time": 0.6336912155151367, "epoch": 9, "memory": 44138, "step": 32256} +{"lr": 1.640192189059216e-05, "data_time": 0.001858353614807129, "loss": 0.3107178032398224, "time": 0.6351148128509522, "epoch": 9, "memory": 44138, "step": 32356} +{"lr": 1.6217151633066716e-05, "data_time": 0.001909184455871582, "loss": 0.28186499923467634, "time": 0.632826566696167, "epoch": 9, "memory": 44138, "step": 32456} +{"lr": 1.6034888929488564e-05, "data_time": 0.0016216278076171876, "loss": 0.2919092759490013, "time": 0.6329756736755371, "epoch": 9, "memory": 44138, "step": 32556} +{"lr": 1.585514556428258e-05, "data_time": 0.004290103912353516, "loss": 0.2939515233039856, "time": 0.6385218620300293, "epoch": 9, "memory": 44138, "step": 32656} +{"lr": 1.567793315898261e-05, "data_time": 0.001831388473510742, "loss": 0.32163310050964355, "time": 0.6336223840713501, "epoch": 9, "memory": 44138, "step": 32756} +{"lr": 1.5503263171480048e-05, "data_time": 0.0016253471374511718, "loss": 0.33599635064601896, "time": 0.6340978145599365, "epoch": 9, "memory": 44138, "step": 32856} +{"lr": 1.5331146895283165e-05, "data_time": 0.0035316944122314453, "loss": 0.3018835335969925, "time": 0.6369681835174561, "epoch": 9, "memory": 44138, "step": 32956} +{"lr": 1.5161595458786728e-05, "data_time": 0.002083706855773926, "loss": 0.3038186222314835, "time": 0.6329025030136108, "epoch": 9, "memory": 44138, "step": 33056} +{"lr": 1.4994619824552563e-05, "data_time": 0.002429080009460449, "loss": 0.3399715185165405, "time": 0.6365882635116578, "epoch": 9, "memory": 44138, "step": 33156} +{"lr": 1.4830230788600802e-05, "data_time": 0.0028122663497924805, "loss": 0.32248779237270353, "time": 0.6355271339416504, "epoch": 9, "memory": 44138, "step": 33256} +{"lr": 1.4668438979711779e-05, "data_time": 0.001717996597290039, "loss": 0.301661479473114, "time": 0.6334915876388549, "epoch": 9, "memory": 44138, "step": 33356} +{"lr": 1.4509254858738844e-05, "data_time": 0.001712799072265625, "loss": 0.3399364411830902, "time": 0.6332926034927369, "epoch": 9, "memory": 44138, "step": 33456} +{"lr": 1.4352688717931966e-05, "data_time": 0.0021294593811035157, "loss": 0.31088860630989074, "time": 0.6344173431396485, "epoch": 9, "memory": 44138, "step": 33556} +{"lr": 1.4198750680272348e-05, "data_time": 0.0017971038818359376, "loss": 0.30998590886592864, "time": 0.6326429843902588, "epoch": 9, "memory": 44138, "step": 33656} +{"lr": 1.4047450698817857e-05, "data_time": 0.002286982536315918, "loss": 0.29340130686759947, "time": 0.6346447944641114, "epoch": 9, "memory": 44138, "step": 33756} +{"lr": 1.389879855605952e-05, "data_time": 0.0018810272216796876, "loss": 0.3187680631875992, "time": 0.6445939779281616, "epoch": 9, "memory": 44138, "step": 33856} +{"lr": 1.3752803863288976e-05, "data_time": 0.003167533874511719, "loss": 0.2890175938606262, "time": 0.636824107170105, "epoch": 9, "memory": 44138, "step": 33956} +{"lr": 1.3609476059977129e-05, "data_time": 0.0018992662429809571, "loss": 0.3111235052347183, "time": 0.6367466688156128, "epoch": 9, "memory": 44138, "step": 34056} +{"lr": 1.3468824413163754e-05, "data_time": 0.0017231225967407227, "loss": 0.3018523871898651, "time": 0.63269944190979, "epoch": 9, "memory": 44138, "step": 34156} +{"lr": 1.3330858016858395e-05, "data_time": 0.002386045455932617, "loss": 0.3260850816965103, "time": 0.6369416952133179, "epoch": 9, "memory": 44138, "step": 34256} +{"lr": 1.3195585791452309e-05, "data_time": 0.0018776178359985352, "loss": 0.29141670614480975, "time": 0.6336064100265503, "epoch": 9, "memory": 44138, "step": 34356} +{"lr": 1.3063016483141729e-05, "data_time": 0.003920650482177735, "loss": 0.3216202884912491, "time": 0.6353137969970704, "epoch": 9, "memory": 44138, "step": 34456} +{"lr": 1.2933158663362388e-05, "data_time": 0.0017136812210083007, "loss": 0.3003173589706421, "time": 0.6339617252349854, "epoch": 9, "memory": 44138, "step": 34556} +{"lr": 1.2806020728235311e-05, "data_time": 0.0017922878265380859, "loss": 0.3098007678985596, "time": 0.6328829288482666, "epoch": 9, "memory": 44138, "step": 34656} +{"lr": 1.2681610898023963e-05, "data_time": 0.0016156911849975586, "loss": 0.32777359485626223, "time": 0.6387759685516358, "epoch": 9, "memory": 44138, "step": 34756} +{"lr": 1.2559937216602731e-05, "data_time": 0.0035127878189086916, "loss": 0.27056471854448316, "time": 0.6349343299865723, "epoch": 9, "memory": 44138, "step": 34856} +{"lr": 1.2441007550936847e-05, "data_time": 0.0021913766860961912, "loss": 0.3135535687208176, "time": 0.6349970817565918, "epoch": 9, "memory": 44138, "step": 34956} +{"lr": 1.232482959057377e-05, "data_time": 0.0020348072052001954, "loss": 0.32948253154754636, "time": 0.6353827476501465, "epoch": 9, "memory": 44138, "step": 35056} +{"lr": 1.2211410847145969e-05, "data_time": 0.0014485836029052735, "loss": 0.32403912097215654, "time": 0.6327764511108398, "epoch": 9, "memory": 44138, "step": 35156} +{"accuracy/top1": 88.6712417602539, "data_time": 0.03698922847879344, "time": 1.2817480481904129, "step": 9} +{"lr": 1.209311679360931e-05, "data_time": 0.00230712890625, "loss": 0.3084552973508835, "time": 0.6392616510391236, "epoch": 10, "memory": 44138, "step": 35263} +{"lr": 1.1985432727488666e-05, "data_time": 0.0022099971771240234, "loss": 0.28848199993371965, "time": 0.6374534845352173, "epoch": 10, "memory": 44138, "step": 35363} +{"lr": 1.1880529822435289e-05, "data_time": 0.001630401611328125, "loss": 0.28577466011047364, "time": 0.6338502883911132, "epoch": 10, "memory": 44138, "step": 35463} +{"lr": 1.1778414861078726e-05, "data_time": 0.0018352985382080079, "loss": 0.32852563709020616, "time": 0.6359284162521363, "epoch": 10, "memory": 44138, "step": 35563} +{"lr": 1.1679094445790538e-05, "data_time": 0.001617097854614258, "loss": 0.3234549969434738, "time": 0.6330214977264405, "epoch": 10, "memory": 44138, "step": 35663} +{"lr": 1.1582574998257377e-05, "data_time": 0.00251467227935791, "loss": 0.2979410693049431, "time": 0.6355472564697265, "epoch": 10, "memory": 44138, "step": 35763} +{"lr": 1.1488862759065762e-05, "data_time": 0.002043938636779785, "loss": 0.297506745159626, "time": 0.636680793762207, "epoch": 10, "memory": 44138, "step": 35863} +{"lr": 1.1397963787298665e-05, "data_time": 0.0018564701080322266, "loss": 0.30051955580711365, "time": 0.6340206146240235, "epoch": 10, "memory": 44138, "step": 35963} +{"lr": 1.1309883960143664e-05, "data_time": 0.0022110939025878906, "loss": 0.3067521661520004, "time": 0.6357958555221558, "epoch": 10, "memory": 44138, "step": 36063} +{"lr": 1.122462897251302e-05, "data_time": 0.0017559289932250976, "loss": 0.2953896582126617, "time": 0.6364803314208984, "epoch": 10, "memory": 44138, "step": 36163} +{"lr": 1.114220433667542e-05, "data_time": 0.004400992393493652, "loss": 0.31145541816949845, "time": 0.6357387781143189, "epoch": 10, "memory": 44138, "step": 36263} +{"lr": 1.10626153818996e-05, "data_time": 0.0024420976638793944, "loss": 0.30642210841178896, "time": 0.6363560438156128, "epoch": 10, "memory": 44138, "step": 36363} +{"lr": 1.0985867254109761e-05, "data_time": 0.0020621061325073243, "loss": 0.33522696793079376, "time": 0.6336010217666626, "epoch": 10, "memory": 44138, "step": 36463} +{"lr": 1.0911964915552823e-05, "data_time": 0.0021435022354125977, "loss": 0.3141847595572472, "time": 0.6341703414916993, "epoch": 10, "memory": 44138, "step": 36563} +{"lr": 1.0840913144477643e-05, "data_time": 0.0015696287155151367, "loss": 0.32551797330379484, "time": 0.632967185974121, "epoch": 10, "memory": 44138, "step": 36663} +{"lr": 1.0772716534826064e-05, "data_time": 0.0017685174942016601, "loss": 0.3243787676095963, "time": 0.6345966577529907, "epoch": 10, "memory": 44138, "step": 36763} +{"lr": 1.0707379495935828e-05, "data_time": 0.001757049560546875, "loss": 0.30183427035808563, "time": 0.6334445476531982, "epoch": 10, "memory": 44138, "step": 36863} +{"lr": 1.0644906252255541e-05, "data_time": 0.001818084716796875, "loss": 0.32428970336914065, "time": 0.6361659526824951, "epoch": 10, "memory": 44138, "step": 36963} +{"lr": 1.0585300843071527e-05, "data_time": 0.0016587018966674806, "loss": 0.30885269343852995, "time": 0.6327872753143311, "epoch": 10, "memory": 44138, "step": 37063} +{"lr": 1.0528567122246654e-05, "data_time": 0.002350735664367676, "loss": 0.3016982674598694, "time": 0.6364793300628662, "epoch": 10, "memory": 44138, "step": 37163} +{"lr": 1.0474708757971141e-05, "data_time": 0.0017575979232788085, "loss": 0.3050421953201294, "time": 0.6336272716522217, "epoch": 10, "memory": 44138, "step": 37263} +{"lr": 1.042372923252543e-05, "data_time": 0.003053712844848633, "loss": 0.29336825013160706, "time": 0.6358482360839843, "epoch": 10, "memory": 44138, "step": 37363} +{"lr": 1.0375631842054998e-05, "data_time": 0.004346632957458496, "loss": 0.3325827866792679, "time": 0.638576078414917, "epoch": 10, "memory": 44138, "step": 37463} +{"lr": 1.0330419696357254e-05, "data_time": 0.002854728698730469, "loss": 0.31957771331071855, "time": 0.6384903430938721, "epoch": 10, "memory": 44138, "step": 37563} +{"lr": 1.028809571868047e-05, "data_time": 0.005306839942932129, "loss": 0.3192389339208603, "time": 0.6413367748260498, "epoch": 10, "memory": 44138, "step": 37663} +{"lr": 1.0248662645534766e-05, "data_time": 0.003085160255432129, "loss": 0.3142657458782196, "time": 0.6393048286437988, "epoch": 10, "memory": 44138, "step": 37763} +{"lr": 1.0212123026515194e-05, "data_time": 0.0040253400802612305, "loss": 0.2617375895380974, "time": 0.6411966562271119, "epoch": 10, "memory": 44138, "step": 37863} +{"lr": 1.0178479224136881e-05, "data_time": 0.004627513885498047, "loss": 0.2990847289562225, "time": 0.6819361686706543, "epoch": 10, "memory": 44138, "step": 37963} +{"lr": 1.0147733413682289e-05, "data_time": 0.007761240005493164, "loss": 0.3163591355085373, "time": 0.6426003932952881, "epoch": 10, "memory": 44138, "step": 38063} +{"lr": 1.0119887583060527e-05, "data_time": 0.0013313770294189453, "loss": 0.3314773261547089, "time": 0.632814359664917, "epoch": 10, "memory": 44138, "step": 38163} +{"lr": 1.0094943532678871e-05, "data_time": 0.0016098737716674805, "loss": 0.30876474678516386, "time": 0.632322359085083, "epoch": 10, "memory": 44138, "step": 38263} +{"lr": 1.007290287532637e-05, "data_time": 0.007057762145996094, "loss": 0.29862377792596817, "time": 0.6595552682876586, "epoch": 10, "memory": 44138, "step": 38363} +{"lr": 1.0053767036069495e-05, "data_time": 0.001546788215637207, "loss": 0.30123724937438967, "time": 0.6319895267486573, "epoch": 10, "memory": 44138, "step": 38463} +{"lr": 1.003753725216005e-05, "data_time": 0.0037961959838867187, "loss": 0.34542007744312286, "time": 0.6363131999969482, "epoch": 10, "memory": 44138, "step": 38563} +{"lr": 1.0024214572955152e-05, "data_time": 0.01427314281463623, "loss": 0.2820836007595062, "time": 0.6622368097305298, "epoch": 10, "memory": 44138, "step": 38663} +{"lr": 1.0013799859849456e-05, "data_time": 0.001521921157836914, "loss": 0.2955233693122864, "time": 0.6327824831008911, "epoch": 10, "memory": 44138, "step": 38763} +{"lr": 1.0006293786219338e-05, "data_time": 0.002329707145690918, "loss": 0.2788534551858902, "time": 0.6370566606521606, "epoch": 10, "memory": 44138, "step": 38863} +{"lr": 1.0001696837379472e-05, "data_time": 0.008408093452453613, "loss": 0.3348928228020668, "time": 0.6573300123214721, "epoch": 10, "memory": 44138, "step": 38963} +{"lr": 1.0000009310551361e-05, "data_time": 0.003323864936828613, "loss": 0.30951560884714124, "time": 0.6480633735656738, "epoch": 10, "memory": 44138, "step": 39063} +{"accuracy/top1": 88.59817504882812, "data_time": 0.11093969180666167, "time": 1.353807778194033, "step": 10} diff --git a/clip_large_pretrain_4x256_sdv1_lr1e-4_wopostnorm_wneck/20230604_175638/vis_data/vis_image/0000000.png_1.png b/clip_large_pretrain_4x256_sdv1_lr1e-4_wopostnorm_wneck/20230604_175638/vis_data/vis_image/0000000.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..2dd780ddc2c2404c2b0a108e54f8c34058a505cb Binary files /dev/null and b/clip_large_pretrain_4x256_sdv1_lr1e-4_wopostnorm_wneck/20230604_175638/vis_data/vis_image/0000000.png_1.png differ diff --git a/clip_large_pretrain_4x256_sdv1_lr1e-4_wopostnorm_wneck/20230604_175638/vis_data/vis_image/0000000.png_10.png b/clip_large_pretrain_4x256_sdv1_lr1e-4_wopostnorm_wneck/20230604_175638/vis_data/vis_image/0000000.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..840666cf30dc8a2e2a6b9a6ca237cb049584f488 Binary files /dev/null and b/clip_large_pretrain_4x256_sdv1_lr1e-4_wopostnorm_wneck/20230604_175638/vis_data/vis_image/0000000.png_10.png differ diff --git a/clip_large_pretrain_4x256_sdv1_lr1e-4_wopostnorm_wneck/20230604_175638/vis_data/vis_image/0000000.png_2.png b/clip_large_pretrain_4x256_sdv1_lr1e-4_wopostnorm_wneck/20230604_175638/vis_data/vis_image/0000000.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..928b5b7cb6d76eb53aa032cf7401842bb0da8dbe Binary files /dev/null and b/clip_large_pretrain_4x256_sdv1_lr1e-4_wopostnorm_wneck/20230604_175638/vis_data/vis_image/0000000.png_2.png differ diff --git a/clip_large_pretrain_4x256_sdv1_lr1e-4_wopostnorm_wneck/20230604_175638/vis_data/vis_image/0000000.png_3.png b/clip_large_pretrain_4x256_sdv1_lr1e-4_wopostnorm_wneck/20230604_175638/vis_data/vis_image/0000000.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..cf86789aad5bab60c058f589efcea83488d85081 Binary files /dev/null and b/clip_large_pretrain_4x256_sdv1_lr1e-4_wopostnorm_wneck/20230604_175638/vis_data/vis_image/0000000.png_3.png differ diff --git a/clip_large_pretrain_4x256_sdv1_lr1e-4_wopostnorm_wneck/20230604_175638/vis_data/vis_image/0000000.png_4.png b/clip_large_pretrain_4x256_sdv1_lr1e-4_wopostnorm_wneck/20230604_175638/vis_data/vis_image/0000000.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..3e8bbcbb5e3142be9108648f6314a461a88ef9e1 Binary files /dev/null and b/clip_large_pretrain_4x256_sdv1_lr1e-4_wopostnorm_wneck/20230604_175638/vis_data/vis_image/0000000.png_4.png differ diff --git a/clip_large_pretrain_4x256_sdv1_lr1e-4_wopostnorm_wneck/20230604_175638/vis_data/vis_image/0000000.png_5.png b/clip_large_pretrain_4x256_sdv1_lr1e-4_wopostnorm_wneck/20230604_175638/vis_data/vis_image/0000000.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..de61ec9b7c5b9445b5e05533bb9854c5082c0b7c Binary files /dev/null and b/clip_large_pretrain_4x256_sdv1_lr1e-4_wopostnorm_wneck/20230604_175638/vis_data/vis_image/0000000.png_5.png differ diff --git a/clip_large_pretrain_4x256_sdv1_lr1e-4_wopostnorm_wneck/20230604_175638/vis_data/vis_image/0000000.png_6.png b/clip_large_pretrain_4x256_sdv1_lr1e-4_wopostnorm_wneck/20230604_175638/vis_data/vis_image/0000000.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..9486c4cb2557267658ed250ef3c9500887b02ea3 Binary files /dev/null and b/clip_large_pretrain_4x256_sdv1_lr1e-4_wopostnorm_wneck/20230604_175638/vis_data/vis_image/0000000.png_6.png differ diff --git a/clip_large_pretrain_4x256_sdv1_lr1e-4_wopostnorm_wneck/20230604_175638/vis_data/vis_image/0000000.png_7.png b/clip_large_pretrain_4x256_sdv1_lr1e-4_wopostnorm_wneck/20230604_175638/vis_data/vis_image/0000000.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..f3fc202694b75014ac3d4de242971d3821248911 Binary files /dev/null and b/clip_large_pretrain_4x256_sdv1_lr1e-4_wopostnorm_wneck/20230604_175638/vis_data/vis_image/0000000.png_7.png differ diff --git a/clip_large_pretrain_4x256_sdv1_lr1e-4_wopostnorm_wneck/20230604_175638/vis_data/vis_image/0000000.png_8.png b/clip_large_pretrain_4x256_sdv1_lr1e-4_wopostnorm_wneck/20230604_175638/vis_data/vis_image/0000000.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..886b633fa7ffca72553f26717c83fa51cc5254b4 Binary files /dev/null and b/clip_large_pretrain_4x256_sdv1_lr1e-4_wopostnorm_wneck/20230604_175638/vis_data/vis_image/0000000.png_8.png differ diff --git a/clip_large_pretrain_4x256_sdv1_lr1e-4_wopostnorm_wneck/20230604_175638/vis_data/vis_image/0000000.png_9.png b/clip_large_pretrain_4x256_sdv1_lr1e-4_wopostnorm_wneck/20230604_175638/vis_data/vis_image/0000000.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..086a24e0a38c9b3f66a9d7dc56e8df2b615e0c98 Binary files /dev/null and b/clip_large_pretrain_4x256_sdv1_lr1e-4_wopostnorm_wneck/20230604_175638/vis_data/vis_image/0000000.png_9.png differ diff --git a/clip_large_pretrain_4x256_sdv1_lr1e-4_wopostnorm_wneck/20230604_175638/vis_data/vis_image/000005174.jpg_1.png b/clip_large_pretrain_4x256_sdv1_lr1e-4_wopostnorm_wneck/20230604_175638/vis_data/vis_image/000005174.jpg_1.png new file mode 100644 index 0000000000000000000000000000000000000000..d3f64be212d3526446f9b711ab283aa195ed317f Binary files /dev/null and b/clip_large_pretrain_4x256_sdv1_lr1e-4_wopostnorm_wneck/20230604_175638/vis_data/vis_image/000005174.jpg_1.png differ diff --git a/clip_large_pretrain_4x256_sdv1_lr1e-4_wopostnorm_wneck/20230604_175638/vis_data/vis_image/000005174.jpg_10.png b/clip_large_pretrain_4x256_sdv1_lr1e-4_wopostnorm_wneck/20230604_175638/vis_data/vis_image/000005174.jpg_10.png new file mode 100644 index 0000000000000000000000000000000000000000..e9231995e2e67853a3e15e06c2792acfc358c2fa Binary files /dev/null and b/clip_large_pretrain_4x256_sdv1_lr1e-4_wopostnorm_wneck/20230604_175638/vis_data/vis_image/000005174.jpg_10.png differ diff --git a/clip_large_pretrain_4x256_sdv1_lr1e-4_wopostnorm_wneck/20230604_175638/vis_data/vis_image/000005174.jpg_2.png b/clip_large_pretrain_4x256_sdv1_lr1e-4_wopostnorm_wneck/20230604_175638/vis_data/vis_image/000005174.jpg_2.png new file mode 100644 index 0000000000000000000000000000000000000000..ea105f21daf07f3674f14787b88a672d0af44a82 Binary files /dev/null and b/clip_large_pretrain_4x256_sdv1_lr1e-4_wopostnorm_wneck/20230604_175638/vis_data/vis_image/000005174.jpg_2.png differ diff --git a/clip_large_pretrain_4x256_sdv1_lr1e-4_wopostnorm_wneck/20230604_175638/vis_data/vis_image/000005174.jpg_3.png b/clip_large_pretrain_4x256_sdv1_lr1e-4_wopostnorm_wneck/20230604_175638/vis_data/vis_image/000005174.jpg_3.png new file mode 100644 index 0000000000000000000000000000000000000000..b2503c2031c5d9a1f16fb3b7e5a9ca277aa3988b Binary files /dev/null and b/clip_large_pretrain_4x256_sdv1_lr1e-4_wopostnorm_wneck/20230604_175638/vis_data/vis_image/000005174.jpg_3.png differ diff --git a/clip_large_pretrain_4x256_sdv1_lr1e-4_wopostnorm_wneck/20230604_175638/vis_data/vis_image/000005174.jpg_4.png b/clip_large_pretrain_4x256_sdv1_lr1e-4_wopostnorm_wneck/20230604_175638/vis_data/vis_image/000005174.jpg_4.png new file mode 100644 index 0000000000000000000000000000000000000000..ced18cff92b2659ef68d77e3e7daccc2e3409133 Binary files /dev/null and b/clip_large_pretrain_4x256_sdv1_lr1e-4_wopostnorm_wneck/20230604_175638/vis_data/vis_image/000005174.jpg_4.png differ diff --git a/clip_large_pretrain_4x256_sdv1_lr1e-4_wopostnorm_wneck/20230604_175638/vis_data/vis_image/000005174.jpg_5.png b/clip_large_pretrain_4x256_sdv1_lr1e-4_wopostnorm_wneck/20230604_175638/vis_data/vis_image/000005174.jpg_5.png new file mode 100644 index 0000000000000000000000000000000000000000..6e75951c22f24e099f8a4a70a2b6f6ceaa6730c4 Binary files /dev/null and b/clip_large_pretrain_4x256_sdv1_lr1e-4_wopostnorm_wneck/20230604_175638/vis_data/vis_image/000005174.jpg_5.png differ diff --git a/clip_large_pretrain_4x256_sdv1_lr1e-4_wopostnorm_wneck/20230604_175638/vis_data/vis_image/000005174.jpg_6.png b/clip_large_pretrain_4x256_sdv1_lr1e-4_wopostnorm_wneck/20230604_175638/vis_data/vis_image/000005174.jpg_6.png new file mode 100644 index 0000000000000000000000000000000000000000..93d94a2901004f1af6c1b003a2fddf768ed7a51c Binary files /dev/null and b/clip_large_pretrain_4x256_sdv1_lr1e-4_wopostnorm_wneck/20230604_175638/vis_data/vis_image/000005174.jpg_6.png differ diff --git a/clip_large_pretrain_4x256_sdv1_lr1e-4_wopostnorm_wneck/20230604_175638/vis_data/vis_image/000005174.jpg_7.png b/clip_large_pretrain_4x256_sdv1_lr1e-4_wopostnorm_wneck/20230604_175638/vis_data/vis_image/000005174.jpg_7.png new file mode 100644 index 0000000000000000000000000000000000000000..081960917bb74ac00173cb3789d72ff235af7d6a Binary files /dev/null and b/clip_large_pretrain_4x256_sdv1_lr1e-4_wopostnorm_wneck/20230604_175638/vis_data/vis_image/000005174.jpg_7.png differ diff --git a/clip_large_pretrain_4x256_sdv1_lr1e-4_wopostnorm_wneck/20230604_175638/vis_data/vis_image/000005174.jpg_8.png b/clip_large_pretrain_4x256_sdv1_lr1e-4_wopostnorm_wneck/20230604_175638/vis_data/vis_image/000005174.jpg_8.png new file mode 100644 index 0000000000000000000000000000000000000000..0094c8e855982c2093010f681f30db22e2dc3135 Binary files /dev/null and b/clip_large_pretrain_4x256_sdv1_lr1e-4_wopostnorm_wneck/20230604_175638/vis_data/vis_image/000005174.jpg_8.png differ diff --git a/clip_large_pretrain_4x256_sdv1_lr1e-4_wopostnorm_wneck/20230604_175638/vis_data/vis_image/000005174.jpg_9.png b/clip_large_pretrain_4x256_sdv1_lr1e-4_wopostnorm_wneck/20230604_175638/vis_data/vis_image/000005174.jpg_9.png new file mode 100644 index 0000000000000000000000000000000000000000..dc7314f3f6570dbd15951585d941383737ba3f76 Binary files /dev/null and b/clip_large_pretrain_4x256_sdv1_lr1e-4_wopostnorm_wneck/20230604_175638/vis_data/vis_image/000005174.jpg_9.png differ diff --git a/clip_large_pretrain_4x256_sdv1_lr1e-4_wopostnorm_wneck/clip_large_pretrain_4x256_sdv1_lr1e-4.py b/clip_large_pretrain_4x256_sdv1_lr1e-4_wopostnorm_wneck/clip_large_pretrain_4x256_sdv1_lr1e-4.py new file mode 100644 index 0000000000000000000000000000000000000000..af1f90081c95d88c89d5cc97dca60c3ea695bcf9 --- /dev/null +++ b/clip_large_pretrain_4x256_sdv1_lr1e-4_wopostnorm_wneck/clip_large_pretrain_4x256_sdv1_lr1e-4.py @@ -0,0 +1,236 @@ +optim_wrapper = dict( + optimizer=dict( + type='AdamW', lr=0.0001, weight_decay=0.3, _scope_='mmpretrain'), + paramwise_cfg=dict( + custom_keys=dict({ + '.cls_token': dict(decay_mult=0.0), + '.pos_embed': dict(decay_mult=0.0) + })), + type='AmpOptimWrapper', + dtype='bfloat16', + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=False, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=4096) +model = dict( + type='ImageClassifier', + backbone=dict( + frozen_stages=24, + type='VisionTransformer', + arch='l', + img_size=224, + patch_size=14, + drop_rate=0.1, + pre_norm=True, + final_norm=False, + init_cfg=dict( + type='Pretrained', + checkpoint='ckpt/openclip-ViT-L-14.pth', + prefix='backbone')), + neck=dict( + type='CLIPProjection', + in_channels=1024, + out_channels=768, + init_cfg=dict( + type='Pretrained', + checkpoint='ckpt/openclip-ViT-L-14.pth', + prefix='backbone')), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=768, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + init_cfg=None), + init_cfg=dict( + type='TruncNormal', layer=['Conv2d', 'Linear'], std=0.02, bias=0.0), + train_cfg=None) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=128, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stablediffusionV1-5R2-dpmsolver-25-1m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/cc1m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=128, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=128, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/clip_large_pretrain_4x256_sdv1_lr1e-4_wopostnorm_wneck' diff --git a/clip_large_pretrain_4x256_sdv1_lr1e-4_wopostnorm_wneck/epoch_1.pth b/clip_large_pretrain_4x256_sdv1_lr1e-4_wopostnorm_wneck/epoch_1.pth new file mode 100644 index 0000000000000000000000000000000000000000..8706f8060a54da774005ae1a2dedb1a367ca0854 --- /dev/null +++ b/clip_large_pretrain_4x256_sdv1_lr1e-4_wopostnorm_wneck/epoch_1.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:7a3914b67049fc94427aedc76fabc999401cecf007d6f6a2872f1be8c38c40b5 +size 2438616641 diff --git a/clip_large_pretrain_4x256_sdv1_lr1e-4_wopostnorm_wneck/epoch_10.pth b/clip_large_pretrain_4x256_sdv1_lr1e-4_wopostnorm_wneck/epoch_10.pth new file mode 100644 index 0000000000000000000000000000000000000000..447c61cb7a9cbd44f74261372f57ef41eb4937e3 --- /dev/null +++ b/clip_large_pretrain_4x256_sdv1_lr1e-4_wopostnorm_wneck/epoch_10.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:cf7d5963276f1cbc4e27dec05894d4af60b6a1c71cb81f8086e7b6b4105f60d0 +size 2441185857 diff --git a/clip_large_pretrain_4x256_sdv1_lr1e-4_wopostnorm_wneck/epoch_2.pth b/clip_large_pretrain_4x256_sdv1_lr1e-4_wopostnorm_wneck/epoch_2.pth new file mode 100644 index 0000000000000000000000000000000000000000..3895e453edc192230caa8cea438347b5ccb2e514 --- /dev/null +++ b/clip_large_pretrain_4x256_sdv1_lr1e-4_wopostnorm_wneck/epoch_2.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:9138df19619b7d0df0aff441375a8d373c3f898d6cad2fd7de446c2881af6cbf +size 2438901121 diff --git a/clip_large_pretrain_4x256_sdv1_lr1e-4_wopostnorm_wneck/epoch_3.pth b/clip_large_pretrain_4x256_sdv1_lr1e-4_wopostnorm_wneck/epoch_3.pth new file mode 100644 index 0000000000000000000000000000000000000000..a1e266db4134cf59cee8289663b384bd12c4e745 --- /dev/null +++ b/clip_large_pretrain_4x256_sdv1_lr1e-4_wopostnorm_wneck/epoch_3.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:1d69cfd385367252ccf3aca240172ac0c55f2fd02238c36d9ae3ffb4983a16c1 +size 2439185153 diff --git a/clip_large_pretrain_4x256_sdv1_lr1e-4_wopostnorm_wneck/epoch_4.pth b/clip_large_pretrain_4x256_sdv1_lr1e-4_wopostnorm_wneck/epoch_4.pth new file mode 100644 index 0000000000000000000000000000000000000000..93316b4a3f7b2f0ceee2af95c6adee24f5572b85 --- /dev/null +++ b/clip_large_pretrain_4x256_sdv1_lr1e-4_wopostnorm_wneck/epoch_4.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:1e64a0173cf55addb1900c370123592b6ecc620540c39ab7ab6c29a17ff15bd1 +size 2439468609 diff --git a/clip_large_pretrain_4x256_sdv1_lr1e-4_wopostnorm_wneck/epoch_5.pth b/clip_large_pretrain_4x256_sdv1_lr1e-4_wopostnorm_wneck/epoch_5.pth new file mode 100644 index 0000000000000000000000000000000000000000..f96670c5ad4cdb262e6bd8f4ce7a7978fb9c809b --- /dev/null +++ b/clip_large_pretrain_4x256_sdv1_lr1e-4_wopostnorm_wneck/epoch_5.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:47d573230e5cd2718deb310bbc58b376c0107b8b34e1e16b6e1415ff799123f9 +size 2439752193 diff --git a/clip_large_pretrain_4x256_sdv1_lr1e-4_wopostnorm_wneck/epoch_6.pth b/clip_large_pretrain_4x256_sdv1_lr1e-4_wopostnorm_wneck/epoch_6.pth new file mode 100644 index 0000000000000000000000000000000000000000..e7f5544a0db04d2a1e6a67255023e529a7f1fa74 --- /dev/null +++ b/clip_large_pretrain_4x256_sdv1_lr1e-4_wopostnorm_wneck/epoch_6.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:cae81cbc0b9949fb438e1e25dafd40b511b1c7afff9b8843304a9914de10c928 +size 2440036097 diff --git a/clip_large_pretrain_4x256_sdv1_lr1e-4_wopostnorm_wneck/epoch_7.pth b/clip_large_pretrain_4x256_sdv1_lr1e-4_wopostnorm_wneck/epoch_7.pth new file mode 100644 index 0000000000000000000000000000000000000000..0db8d48e2a85a7e598e0257a4992c4425f7df871 --- /dev/null +++ b/clip_large_pretrain_4x256_sdv1_lr1e-4_wopostnorm_wneck/epoch_7.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:aaf38e65425ec1533a29a856ec8a05f5152a70ffd81b40fcdf735fad5fd30168 +size 2440320897 diff --git a/clip_large_pretrain_4x256_sdv1_lr1e-4_wopostnorm_wneck/epoch_8.pth b/clip_large_pretrain_4x256_sdv1_lr1e-4_wopostnorm_wneck/epoch_8.pth new file mode 100644 index 0000000000000000000000000000000000000000..d47dd08bbb20c4b0763088090752d3db1d63c95c --- /dev/null +++ b/clip_large_pretrain_4x256_sdv1_lr1e-4_wopostnorm_wneck/epoch_8.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:20160dc7dbf4ade5b856adf200a9420bca58a0a207a37393d6f192694c4e48c6 +size 2440609089 diff --git a/clip_large_pretrain_4x256_sdv1_lr1e-4_wopostnorm_wneck/epoch_9.pth b/clip_large_pretrain_4x256_sdv1_lr1e-4_wopostnorm_wneck/epoch_9.pth new file mode 100644 index 0000000000000000000000000000000000000000..48822140fed883b1432af982f3f08011a6682cde --- /dev/null +++ b/clip_large_pretrain_4x256_sdv1_lr1e-4_wopostnorm_wneck/epoch_9.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ea5dc96592a421277c58b1f5e12a844c8729b1e62918e85b5b27170b5c07f3fa +size 2440897281 diff --git a/clip_large_pretrain_4x256_sdv1_lr1e-4_wopostnorm_wneck/last_checkpoint b/clip_large_pretrain_4x256_sdv1_lr1e-4_wopostnorm_wneck/last_checkpoint new file mode 100644 index 0000000000000000000000000000000000000000..d3185d0cbf8da329aa04e03ec497f57669a8d397 --- /dev/null +++ b/clip_large_pretrain_4x256_sdv1_lr1e-4_wopostnorm_wneck/last_checkpoint @@ -0,0 +1 @@ +/mnt/petrelfs/luzeyu/workspace/fakebench/mmpretrain/workdir/clip_large_pretrain_4x256_sdv1_lr1e-4_wopostnorm_wneck/epoch_10.pth \ No newline at end of file diff --git a/clip_large_pretrain_4x256_sdv2_lr1e-4/20230604_015520/20230604_015520.log b/clip_large_pretrain_4x256_sdv2_lr1e-4/20230604_015520/20230604_015520.log new file mode 100644 index 0000000000000000000000000000000000000000..f49268ff615d943e7882a58a16776dbb41e604c2 --- /dev/null +++ b/clip_large_pretrain_4x256_sdv2_lr1e-4/20230604_015520/20230604_015520.log @@ -0,0 +1,1710 @@ +2023/06/04 01:55:25 - mmengine - INFO - +------------------------------------------------------------ +System environment: + sys.platform: linux + Python: 3.10.9 (main, Mar 8 2023, 10:47:38) [GCC 11.2.0] + CUDA available: True + numpy_random_seed: 1786672076 + GPU 0,1,2,3: NVIDIA A100-SXM4-80GB + CUDA_HOME: /mnt/petrelfs/share/cuda-11.6 + NVCC: Cuda compilation tools, release 11.6, V11.6.124 + GCC: gcc (GCC) 7.5.0 + PyTorch: 1.13.1 + PyTorch compiling details: PyTorch built with: + - GCC 9.3 + - C++ Version: 201402 + - Intel(R) oneAPI Math Kernel Library Version 2021.4-Product Build 20210904 for Intel(R) 64 architecture applications + - Intel(R) MKL-DNN v2.6.0 (Git Hash 52b5f107dd9cf10910aaa19cb47f3abf9b349815) + - OpenMP 201511 (a.k.a. OpenMP 4.5) + - LAPACK is enabled (usually provided by MKL) + - NNPACK is enabled + - CPU capability usage: AVX2 + - CUDA Runtime 11.6 + - NVCC architecture flags: -gencode;arch=compute_37,code=sm_37;-gencode;arch=compute_50,code=sm_50;-gencode;arch=compute_60,code=sm_60;-gencode;arch=compute_61,code=sm_61;-gencode;arch=compute_70,code=sm_70;-gencode;arch=compute_75,code=sm_75;-gencode;arch=compute_80,code=sm_80;-gencode;arch=compute_86,code=sm_86;-gencode;arch=compute_37,code=compute_37 + - CuDNN 8.3.2 (built against CUDA 11.5) + - Magma 2.6.1 + - Build settings: BLAS_INFO=mkl, BUILD_TYPE=Release, CUDA_VERSION=11.6, CUDNN_VERSION=8.3.2, CXX_COMPILER=/opt/rh/devtoolset-9/root/usr/bin/c++, CXX_FLAGS= -fabi-version=11 -Wno-deprecated -fvisibility-inlines-hidden -DUSE_PTHREADPOOL -fopenmp -DNDEBUG -DUSE_KINETO -DUSE_FBGEMM -DUSE_QNNPACK -DUSE_PYTORCH_QNNPACK -DUSE_XNNPACK -DSYMBOLICATE_MOBILE_DEBUG_HANDLE -DEDGE_PROFILER_USE_KINETO -O2 -fPIC -Wno-narrowing -Wall -Wextra -Werror=return-type -Werror=non-virtual-dtor -Wno-missing-field-initializers -Wno-type-limits -Wno-array-bounds -Wno-unknown-pragmas -Wunused-local-typedefs -Wno-unused-parameter -Wno-unused-function -Wno-unused-result -Wno-strict-overflow -Wno-strict-aliasing -Wno-error=deprecated-declarations -Wno-stringop-overflow -Wno-psabi -Wno-error=pedantic -Wno-error=redundant-decls -Wno-error=old-style-cast -fdiagnostics-color=always -faligned-new -Wno-unused-but-set-variable -Wno-maybe-uninitialized -fno-math-errno -fno-trapping-math -Werror=format -Werror=cast-function-type -Wno-stringop-overflow, LAPACK_INFO=mkl, PERF_WITH_AVX=1, PERF_WITH_AVX2=1, PERF_WITH_AVX512=1, TORCH_VERSION=1.13.1, USE_CUDA=ON, USE_CUDNN=ON, USE_EXCEPTION_PTR=1, USE_GFLAGS=OFF, USE_GLOG=OFF, USE_MKL=ON, USE_MKLDNN=ON, USE_MPI=OFF, USE_NCCL=ON, USE_NNPACK=ON, USE_OPENMP=ON, USE_ROCM=OFF, + + TorchVision: 0.14.1 + OpenCV: 4.7.0 + MMEngine: 0.7.3 + +Runtime environment: + cudnn_benchmark: True + mp_cfg: {'mp_start_method': 'fork', 'opencv_num_threads': 0} + dist_cfg: {'backend': 'nccl'} + seed: None + deterministic: False + Distributed launcher: slurm + Distributed training: True + GPU number: 4 +------------------------------------------------------------ + +2023/06/04 01:55:29 - mmengine - INFO - Config: +optim_wrapper = dict( + optimizer=dict( + type='AdamW', lr=0.0001, weight_decay=0.3, _scope_='mmpretrain'), + paramwise_cfg=dict( + custom_keys=dict({ + '.cls_token': dict(decay_mult=0.0), + '.pos_embed': dict(decay_mult=0.0) + })), + type='AmpOptimWrapper', + dtype='bfloat16', + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=False, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=4096) +model = dict( + type='ImageClassifier', + backbone=dict( + frozen_stages=24, + type='VisionTransformer', + arch='l', + img_size=224, + patch_size=14, + drop_rate=0.1, + pre_norm=True, + init_cfg=dict( + type='Pretrained', + checkpoint='ckpt/openclip-ViT-L-14.pth', + prefix='backbone')), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=1024, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + init_cfg=None), + init_cfg=dict( + type='TruncNormal', layer=['Conv2d', 'Linear'], std=0.02, bias=0.0), + train_cfg=None) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=128, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stablediffusionV2-1-dpmsolver-25-1m.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/cc1m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=128, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV2-1-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=128, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV2-1-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/clip_large_pretrain_4x256_sdv2_lr1e-4' + +2023/06/04 01:55:42 - mmengine - INFO - Hooks will be executed in the following order: +before_run: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_load_checkpoint: +(ABOVE_NORMAL) EMAHook + -------------------- +before_train: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_train_epoch: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(NORMAL ) DistSamplerSeedHook + -------------------- +before_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook + -------------------- +after_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +after_train_epoch: +(NORMAL ) IterTimerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_val_epoch: +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook + -------------------- +before_val_iter: +(NORMAL ) IterTimerHook + -------------------- +after_val_iter: +(NORMAL ) IterTimerHook +(NORMAL ) VisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_val_epoch: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_save_checkpoint: +(ABOVE_NORMAL) EMAHook + -------------------- +after_train: +(VERY_LOW ) CheckpointHook + -------------------- +before_test_epoch: +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook + -------------------- +before_test_iter: +(NORMAL ) IterTimerHook + -------------------- +after_test_iter: +(NORMAL ) IterTimerHook +(NORMAL ) VisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_test_epoch: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_run: +(BELOW_NORMAL) LoggerHook + -------------------- +2023/06/04 01:56:02 - mmengine - INFO - load backbone in model from: ckpt/openclip-ViT-L-14.pth +Name of parameter - Initialization information + +backbone.cls_token - torch.Size([1, 1, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.pos_embed - torch.Size([1, 257, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.patch_embed.projection.weight - torch.Size([1024, 3, 14, 14]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.0.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.0.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.0.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.0.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.0.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.0.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.0.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.0.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.0.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.0.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.0.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.0.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.1.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.1.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.1.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.1.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.1.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.1.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.1.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.1.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.1.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.1.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.1.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.1.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.2.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.2.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.2.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.2.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.2.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.2.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.2.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.2.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.2.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.2.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.2.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.2.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.3.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.3.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.3.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.3.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.3.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.3.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.3.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.3.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.3.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.3.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.3.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.3.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.4.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.4.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.4.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.4.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.4.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.4.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.4.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.4.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.4.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.4.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.4.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.4.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.5.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.5.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.5.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.5.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.5.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.5.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.5.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.5.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.5.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.5.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.5.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.5.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.6.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.6.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.6.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.6.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.6.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.6.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.6.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.6.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.6.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.6.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.6.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.6.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.7.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.7.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.7.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.7.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.7.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.7.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.7.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.7.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.7.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.7.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.7.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.7.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.8.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.8.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.8.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.8.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.8.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.8.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.8.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.8.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.8.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.8.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.8.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.8.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.9.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.9.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.9.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.9.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.9.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.9.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.9.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.9.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.9.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.9.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.9.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.9.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.10.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.10.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.10.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.10.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.10.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.10.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.10.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.10.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.10.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.10.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.10.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.10.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.11.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.11.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.11.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.11.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.11.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.11.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.11.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.11.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.11.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.11.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.11.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.11.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.12.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.12.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.12.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.12.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.12.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.12.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.12.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.12.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.12.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.12.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.12.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.12.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.13.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.13.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.13.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.13.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.13.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.13.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.13.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.13.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.13.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.13.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.13.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.13.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.14.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.14.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.14.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.14.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.14.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.14.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.14.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.14.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.14.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.14.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.14.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.14.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.15.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.15.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.15.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.15.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.15.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.15.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.15.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.15.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.15.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.15.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.15.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.15.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.16.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.16.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.16.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.16.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.16.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.16.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.16.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.16.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.16.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.16.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.16.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.16.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.17.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.17.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.17.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.17.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.17.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.17.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.17.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.17.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.17.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.17.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.17.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.17.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.18.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.18.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.18.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.18.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.18.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.18.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.18.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.18.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.18.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.18.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.18.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.18.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.19.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.19.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.19.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.19.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.19.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.19.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.19.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.19.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.19.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.19.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.19.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.19.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.20.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.20.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.20.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.20.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.20.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.20.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.20.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.20.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.20.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.20.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.20.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.20.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.21.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.21.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.21.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.21.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.21.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.21.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.21.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.21.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.21.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.21.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.21.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.21.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.22.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.22.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.22.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.22.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.22.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.22.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.22.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.22.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.22.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.22.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.22.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.22.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.23.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.23.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.23.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.23.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.23.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.23.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.23.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.23.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.23.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.23.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.23.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.23.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.pre_norm.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.pre_norm.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +head.fc.weight - torch.Size([2, 1024]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +head.fc.bias - torch.Size([2]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 +2023/06/04 01:56:04 - mmengine - WARNING - "FileClient" will be deprecated in future. Please use io functions in https://mmengine.readthedocs.io/en/latest/api/fileio.html#file-io +2023/06/04 01:56:04 - mmengine - WARNING - "HardDiskBackend" is the alias of "LocalBackend" and the former will be deprecated in future. +2023/06/04 01:56:04 - mmengine - INFO - Checkpoints will be saved to /mnt/petrelfs/luzeyu/workspace/fakebench/mmpretrain/workdir/clip_large_pretrain_4x256_sdv2_lr1e-4. +2023/06/04 01:57:13 - mmengine - INFO - Epoch(train) [1][ 100/3907] lr: 9.9999e-05 eta: 7:28:30 time: 0.6303 data_time: 0.0016 memory: 44120 loss: 0.6511 +2023/06/04 01:58:16 - mmengine - INFO - Epoch(train) [1][ 200/3907] lr: 9.9994e-05 eta: 7:07:38 time: 0.6300 data_time: 0.0014 memory: 44120 loss: 0.5780 +2023/06/04 01:59:19 - mmengine - INFO - Epoch(train) [1][ 300/3907] lr: 9.9987e-05 eta: 7:00:24 time: 0.6316 data_time: 0.0015 memory: 44120 loss: 0.5506 +2023/06/04 02:00:23 - mmengine - INFO - Epoch(train) [1][ 400/3907] lr: 9.9977e-05 eta: 6:56:19 time: 0.6333 data_time: 0.0023 memory: 44120 loss: 0.5344 +2023/06/04 02:01:26 - mmengine - INFO - Epoch(train) [1][ 500/3907] lr: 9.9964e-05 eta: 6:53:23 time: 0.6302 data_time: 0.0016 memory: 44120 loss: 0.5074 +2023/06/04 02:02:29 - mmengine - INFO - Epoch(train) [1][ 600/3907] lr: 9.9948e-05 eta: 6:50:56 time: 0.6301 data_time: 0.0016 memory: 44120 loss: 0.4990 +2023/06/04 02:03:32 - mmengine - INFO - Epoch(train) [1][ 700/3907] lr: 9.9929e-05 eta: 6:48:55 time: 0.6292 data_time: 0.0017 memory: 44120 loss: 0.5029 +2023/06/04 02:04:35 - mmengine - INFO - Epoch(train) [1][ 800/3907] lr: 9.9907e-05 eta: 6:47:23 time: 0.6300 data_time: 0.0017 memory: 44120 loss: 0.4679 +2023/06/04 02:05:38 - mmengine - INFO - Epoch(train) [1][ 900/3907] lr: 9.9882e-05 eta: 6:45:44 time: 0.6295 data_time: 0.0016 memory: 44120 loss: 0.4867 +2023/06/04 02:06:41 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230604_015520 +2023/06/04 02:06:41 - mmengine - INFO - Epoch(train) [1][1000/3907] lr: 9.9855e-05 eta: 6:44:12 time: 0.6297 data_time: 0.0015 memory: 44120 loss: 0.4594 +2023/06/04 02:07:44 - mmengine - INFO - Epoch(train) [1][1100/3907] lr: 9.9824e-05 eta: 6:42:46 time: 0.6294 data_time: 0.0014 memory: 44120 loss: 0.4504 +2023/06/04 02:08:47 - mmengine - INFO - Epoch(train) [1][1200/3907] lr: 9.9791e-05 eta: 6:41:22 time: 0.6302 data_time: 0.0015 memory: 44120 loss: 0.4513 +2023/06/04 02:09:51 - mmengine - INFO - Epoch(train) [1][1300/3907] lr: 9.9755e-05 eta: 6:40:08 time: 0.6316 data_time: 0.0016 memory: 44120 loss: 0.4558 +2023/06/04 02:10:54 - mmengine - INFO - Epoch(train) [1][1400/3907] lr: 9.9716e-05 eta: 6:38:50 time: 0.6292 data_time: 0.0015 memory: 44120 loss: 0.4728 +2023/06/04 02:11:57 - mmengine - INFO - Epoch(train) [1][1500/3907] lr: 9.9674e-05 eta: 6:37:34 time: 0.6303 data_time: 0.0015 memory: 44120 loss: 0.4441 +2023/06/04 02:13:00 - mmengine - INFO - Epoch(train) [1][1600/3907] lr: 9.9629e-05 eta: 6:36:20 time: 0.6305 data_time: 0.0016 memory: 44120 loss: 0.4235 +2023/06/04 02:14:03 - mmengine - INFO - Epoch(train) [1][1700/3907] lr: 9.9581e-05 eta: 6:35:08 time: 0.6309 data_time: 0.0019 memory: 44120 loss: 0.4789 +2023/06/04 02:15:06 - mmengine - INFO - Epoch(train) [1][1800/3907] lr: 9.9530e-05 eta: 6:33:56 time: 0.6308 data_time: 0.0021 memory: 44120 loss: 0.4395 +2023/06/04 02:16:09 - mmengine - INFO - Epoch(train) [1][1900/3907] lr: 9.9476e-05 eta: 6:32:48 time: 0.6321 data_time: 0.0018 memory: 44120 loss: 0.4206 +2023/06/04 02:17:12 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230604_015520 +2023/06/04 02:17:12 - mmengine - INFO - Epoch(train) [1][2000/3907] lr: 9.9420e-05 eta: 6:31:39 time: 0.6313 data_time: 0.0016 memory: 44120 loss: 0.4306 +2023/06/04 02:18:15 - mmengine - INFO - Epoch(train) [1][2100/3907] lr: 9.9361e-05 eta: 6:30:33 time: 0.6312 data_time: 0.0021 memory: 44120 loss: 0.4273 +2023/06/04 02:19:18 - mmengine - INFO - Epoch(train) [1][2200/3907] lr: 9.9298e-05 eta: 6:29:25 time: 0.6324 data_time: 0.0017 memory: 44120 loss: 0.4137 +2023/06/04 02:20:22 - mmengine - INFO - Epoch(train) [1][2300/3907] lr: 9.9233e-05 eta: 6:28:19 time: 0.6327 data_time: 0.0017 memory: 44120 loss: 0.4193 +2023/06/04 02:21:25 - mmengine - INFO - Epoch(train) [1][2400/3907] lr: 9.9165e-05 eta: 6:27:12 time: 0.6312 data_time: 0.0016 memory: 44120 loss: 0.4384 +2023/06/04 02:22:28 - mmengine - INFO - Epoch(train) [1][2500/3907] lr: 9.9095e-05 eta: 6:26:06 time: 0.6320 data_time: 0.0014 memory: 44120 loss: 0.4090 +2023/06/04 02:23:31 - mmengine - INFO - Epoch(train) [1][2600/3907] lr: 9.9021e-05 eta: 6:24:59 time: 0.6303 data_time: 0.0017 memory: 44120 loss: 0.4292 +2023/06/04 02:24:34 - mmengine - INFO - Epoch(train) [1][2700/3907] lr: 9.8944e-05 eta: 6:23:52 time: 0.6296 data_time: 0.0016 memory: 44120 loss: 0.4141 +2023/06/04 02:25:37 - mmengine - INFO - Epoch(train) [1][2800/3907] lr: 9.8865e-05 eta: 6:22:47 time: 0.6308 data_time: 0.0016 memory: 44120 loss: 0.3893 +2023/06/04 02:26:40 - mmengine - INFO - Epoch(train) [1][2900/3907] lr: 9.8783e-05 eta: 6:21:40 time: 0.6292 data_time: 0.0014 memory: 44120 loss: 0.3932 +2023/06/04 02:27:44 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230604_015520 +2023/06/04 02:27:44 - mmengine - INFO - Epoch(train) [1][3000/3907] lr: 9.8698e-05 eta: 6:20:35 time: 0.6313 data_time: 0.0021 memory: 44120 loss: 0.3949 +2023/06/04 02:28:47 - mmengine - INFO - Epoch(train) [1][3100/3907] lr: 9.8610e-05 eta: 6:19:29 time: 0.6303 data_time: 0.0016 memory: 44120 loss: 0.3614 +2023/06/04 02:29:50 - mmengine - INFO - Epoch(train) [1][3200/3907] lr: 9.8519e-05 eta: 6:18:24 time: 0.6316 data_time: 0.0016 memory: 44120 loss: 0.4107 +2023/06/04 02:30:53 - mmengine - INFO - Epoch(train) [1][3300/3907] lr: 9.8426e-05 eta: 6:17:20 time: 0.6324 data_time: 0.0016 memory: 44120 loss: 0.3919 +2023/06/04 02:31:56 - mmengine - INFO - Epoch(train) [1][3400/3907] lr: 9.8330e-05 eta: 6:16:16 time: 0.6319 data_time: 0.0022 memory: 44120 loss: 0.3944 +2023/06/04 02:32:59 - mmengine - INFO - Epoch(train) [1][3500/3907] lr: 9.8231e-05 eta: 6:15:12 time: 0.6310 data_time: 0.0018 memory: 44120 loss: 0.4241 +2023/06/04 02:34:03 - mmengine - INFO - Epoch(train) [1][3600/3907] lr: 9.8129e-05 eta: 6:14:08 time: 0.6322 data_time: 0.0017 memory: 44120 loss: 0.4126 +2023/06/04 02:35:06 - mmengine - INFO - Epoch(train) [1][3700/3907] lr: 9.8024e-05 eta: 6:13:05 time: 0.6301 data_time: 0.0018 memory: 44120 loss: 0.3861 +2023/06/04 02:36:09 - mmengine - INFO - Epoch(train) [1][3800/3907] lr: 9.7917e-05 eta: 6:11:59 time: 0.6288 data_time: 0.0016 memory: 44120 loss: 0.4106 +2023/06/04 02:37:12 - mmengine - INFO - Epoch(train) [1][3900/3907] lr: 9.7806e-05 eta: 6:10:53 time: 0.6306 data_time: 0.0017 memory: 44120 loss: 0.3786 +2023/06/04 02:37:16 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230604_015520 +2023/06/04 02:37:16 - mmengine - INFO - Saving checkpoint at 1 epochs +2023/06/04 02:38:53 - mmengine - INFO - Epoch(val) [1][57/57] accuracy/top1: 60.2032 single-label/precision_classwise: [58.161865234375, 91.34615325927734] single-label/recall_classwise: [99.0341567993164, 12.519380569458008] single-label/f1-score_classwise: [73.2844467163086, 22.020727157592773] data_time: 0.0458 time: 1.3493 +2023/06/04 02:39:55 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230604_015520 +2023/06/04 02:39:59 - mmengine - INFO - Epoch(train) [2][ 100/3907] lr: 9.7685e-05 eta: 6:10:07 time: 0.6312 data_time: 0.0024 memory: 44120 loss: 0.4061 +2023/06/04 02:41:02 - mmengine - INFO - Epoch(train) [2][ 200/3907] lr: 9.7570e-05 eta: 6:09:02 time: 0.6301 data_time: 0.0015 memory: 44120 loss: 0.4036 +2023/06/04 02:42:05 - mmengine - INFO - Epoch(train) [2][ 300/3907] lr: 9.7451e-05 eta: 6:07:56 time: 0.6305 data_time: 0.0014 memory: 44120 loss: 0.3966 +2023/06/04 02:43:08 - mmengine - INFO - Epoch(train) [2][ 400/3907] lr: 9.7329e-05 eta: 6:06:50 time: 0.6293 data_time: 0.0015 memory: 44120 loss: 0.4431 +2023/06/04 02:44:11 - mmengine - INFO - Epoch(train) [2][ 500/3907] lr: 9.7205e-05 eta: 6:05:45 time: 0.6296 data_time: 0.0018 memory: 44120 loss: 0.3746 +2023/06/04 02:45:14 - mmengine - INFO - Epoch(train) [2][ 600/3907] lr: 9.7078e-05 eta: 6:04:39 time: 0.6294 data_time: 0.0014 memory: 44120 loss: 0.4080 +2023/06/04 02:46:17 - mmengine - INFO - Epoch(train) [2][ 700/3907] lr: 9.6949e-05 eta: 6:03:33 time: 0.6291 data_time: 0.0014 memory: 44120 loss: 0.3937 +2023/06/04 02:47:20 - mmengine - INFO - Epoch(train) [2][ 800/3907] lr: 9.6816e-05 eta: 6:02:28 time: 0.6295 data_time: 0.0015 memory: 44120 loss: 0.3627 +2023/06/04 02:48:23 - mmengine - INFO - Epoch(train) [2][ 900/3907] lr: 9.6681e-05 eta: 6:01:23 time: 0.6384 data_time: 0.0015 memory: 44120 loss: 0.3746 +2023/06/04 02:49:26 - mmengine - INFO - Epoch(train) [2][1000/3907] lr: 9.6544e-05 eta: 6:00:18 time: 0.6301 data_time: 0.0016 memory: 44120 loss: 0.3812 +2023/06/04 02:50:25 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230604_015520 +2023/06/04 02:50:29 - mmengine - INFO - Epoch(train) [2][1100/3907] lr: 9.6403e-05 eta: 5:59:13 time: 0.6311 data_time: 0.0020 memory: 44120 loss: 0.4129 +2023/06/04 02:51:32 - mmengine - INFO - Epoch(train) [2][1200/3907] lr: 9.6260e-05 eta: 5:58:09 time: 0.6310 data_time: 0.0017 memory: 44120 loss: 0.4010 +2023/06/04 02:52:35 - mmengine - INFO - Epoch(train) [2][1300/3907] lr: 9.6114e-05 eta: 5:57:04 time: 0.6297 data_time: 0.0016 memory: 44120 loss: 0.3777 +2023/06/04 02:53:39 - mmengine - INFO - Epoch(train) [2][1400/3907] lr: 9.5966e-05 eta: 5:56:00 time: 0.6302 data_time: 0.0015 memory: 44120 loss: 0.3880 +2023/06/04 02:54:42 - mmengine - INFO - Epoch(train) [2][1500/3907] lr: 9.5815e-05 eta: 5:54:55 time: 0.6317 data_time: 0.0014 memory: 44120 loss: 0.3734 +2023/06/04 02:55:45 - mmengine - INFO - Epoch(train) [2][1600/3907] lr: 9.5661e-05 eta: 5:53:52 time: 0.6300 data_time: 0.0016 memory: 44120 loss: 0.3854 +2023/06/04 02:56:48 - mmengine - INFO - Epoch(train) [2][1700/3907] lr: 9.5505e-05 eta: 5:52:47 time: 0.6295 data_time: 0.0015 memory: 44120 loss: 0.3805 +2023/06/04 02:57:51 - mmengine - INFO - Epoch(train) [2][1800/3907] lr: 9.5346e-05 eta: 5:51:42 time: 0.6297 data_time: 0.0016 memory: 44120 loss: 0.3834 +2023/06/04 02:58:54 - mmengine - INFO - Epoch(train) [2][1900/3907] lr: 9.5184e-05 eta: 5:50:38 time: 0.6304 data_time: 0.0016 memory: 44120 loss: 0.3680 +2023/06/04 02:59:57 - mmengine - INFO - Epoch(train) [2][2000/3907] lr: 9.5020e-05 eta: 5:49:34 time: 0.6302 data_time: 0.0013 memory: 44120 loss: 0.3973 +2023/06/04 03:00:56 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230604_015520 +2023/06/04 03:01:00 - mmengine - INFO - Epoch(train) [2][2100/3907] lr: 9.4854e-05 eta: 5:48:29 time: 0.6311 data_time: 0.0015 memory: 44120 loss: 0.3713 +2023/06/04 03:02:03 - mmengine - INFO - Epoch(train) [2][2200/3907] lr: 9.4684e-05 eta: 5:47:25 time: 0.6302 data_time: 0.0014 memory: 44120 loss: 0.3753 +2023/06/04 03:03:06 - mmengine - INFO - Epoch(train) [2][2300/3907] lr: 9.4512e-05 eta: 5:46:21 time: 0.6300 data_time: 0.0014 memory: 44120 loss: 0.3761 +2023/06/04 03:04:09 - mmengine - INFO - Epoch(train) [2][2400/3907] lr: 9.4338e-05 eta: 5:45:16 time: 0.6299 data_time: 0.0018 memory: 44120 loss: 0.3866 +2023/06/04 03:05:12 - mmengine - INFO - Epoch(train) [2][2500/3907] lr: 9.4161e-05 eta: 5:44:12 time: 0.6304 data_time: 0.0017 memory: 44120 loss: 0.3936 +2023/06/04 03:06:15 - mmengine - INFO - Epoch(train) [2][2600/3907] lr: 9.3981e-05 eta: 5:43:08 time: 0.6299 data_time: 0.0016 memory: 44120 loss: 0.4170 +2023/06/04 03:07:18 - mmengine - INFO - Epoch(train) [2][2700/3907] lr: 9.3799e-05 eta: 5:42:04 time: 0.6305 data_time: 0.0016 memory: 44120 loss: 0.3835 +2023/06/04 03:08:21 - mmengine - INFO - Epoch(train) [2][2800/3907] lr: 9.3615e-05 eta: 5:40:59 time: 0.6302 data_time: 0.0015 memory: 44120 loss: 0.3864 +2023/06/04 03:09:24 - mmengine - INFO - Epoch(train) [2][2900/3907] lr: 9.3428e-05 eta: 5:39:56 time: 0.6305 data_time: 0.0016 memory: 44120 loss: 0.3615 +2023/06/04 03:10:28 - mmengine - INFO - Epoch(train) [2][3000/3907] lr: 9.3238e-05 eta: 5:38:52 time: 0.6301 data_time: 0.0014 memory: 44120 loss: 0.3628 +2023/06/04 03:11:26 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230604_015520 +2023/06/04 03:11:31 - mmengine - INFO - Epoch(train) [2][3100/3907] lr: 9.3046e-05 eta: 5:37:49 time: 0.6336 data_time: 0.0017 memory: 44120 loss: 0.3922 +2023/06/04 03:12:34 - mmengine - INFO - Epoch(train) [2][3200/3907] lr: 9.2852e-05 eta: 5:36:45 time: 0.6307 data_time: 0.0017 memory: 44120 loss: 0.3859 +2023/06/04 03:13:37 - mmengine - INFO - Epoch(train) [2][3300/3907] lr: 9.2655e-05 eta: 5:35:41 time: 0.6305 data_time: 0.0015 memory: 44120 loss: 0.3957 +2023/06/04 03:14:40 - mmengine - INFO - Epoch(train) [2][3400/3907] lr: 9.2456e-05 eta: 5:34:37 time: 0.6318 data_time: 0.0016 memory: 44120 loss: 0.3874 +2023/06/04 03:15:43 - mmengine - INFO - Epoch(train) [2][3500/3907] lr: 9.2254e-05 eta: 5:33:33 time: 0.6298 data_time: 0.0016 memory: 44120 loss: 0.3795 +2023/06/04 03:16:46 - mmengine - INFO - Epoch(train) [2][3600/3907] lr: 9.2050e-05 eta: 5:32:30 time: 0.6309 data_time: 0.0015 memory: 44120 loss: 0.3857 +2023/06/04 03:17:49 - mmengine - INFO - Epoch(train) [2][3700/3907] lr: 9.1843e-05 eta: 5:31:27 time: 0.6300 data_time: 0.0015 memory: 44120 loss: 0.3903 +2023/06/04 03:18:52 - mmengine - INFO - Epoch(train) [2][3800/3907] lr: 9.1634e-05 eta: 5:30:23 time: 0.6334 data_time: 0.0015 memory: 44120 loss: 0.3988 +2023/06/04 03:19:56 - mmengine - INFO - Epoch(train) [2][3900/3907] lr: 9.1423e-05 eta: 5:29:19 time: 0.6290 data_time: 0.0014 memory: 44120 loss: 0.3804 +2023/06/04 03:19:59 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230604_015520 +2023/06/04 03:19:59 - mmengine - INFO - Saving checkpoint at 2 epochs +2023/06/04 03:21:33 - mmengine - INFO - Epoch(val) [2][57/57] accuracy/top1: 66.7653 single-label/precision_classwise: [62.49751281738281, 96.75978088378906] single-label/recall_classwise: [99.2677230834961, 26.852712631225586] single-label/f1-score_classwise: [76.70357513427734, 42.03883743286133] data_time: 0.0366 time: 1.2851 +2023/06/04 03:22:39 - mmengine - INFO - Epoch(train) [3][ 100/3907] lr: 9.1194e-05 eta: 5:28:21 time: 0.6291 data_time: 0.0013 memory: 44120 loss: 0.3677 +2023/06/04 03:23:33 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230604_015520 +2023/06/04 03:23:42 - mmengine - INFO - Epoch(train) [3][ 200/3907] lr: 9.0978e-05 eta: 5:27:17 time: 0.6294 data_time: 0.0016 memory: 44120 loss: 0.3831 +2023/06/04 03:24:45 - mmengine - INFO - Epoch(train) [3][ 300/3907] lr: 9.0759e-05 eta: 5:26:13 time: 0.6312 data_time: 0.0020 memory: 44120 loss: 0.3798 +2023/06/04 03:25:48 - mmengine - INFO - Epoch(train) [3][ 400/3907] lr: 9.0539e-05 eta: 5:25:09 time: 0.6300 data_time: 0.0019 memory: 44120 loss: 0.3775 +2023/06/04 03:26:51 - mmengine - INFO - Epoch(train) [3][ 500/3907] lr: 9.0315e-05 eta: 5:24:05 time: 0.6303 data_time: 0.0014 memory: 44120 loss: 0.3702 +2023/06/04 03:27:54 - mmengine - INFO - Epoch(train) [3][ 600/3907] lr: 9.0090e-05 eta: 5:23:01 time: 0.6302 data_time: 0.0017 memory: 44120 loss: 0.3814 +2023/06/04 03:28:57 - mmengine - INFO - Epoch(train) [3][ 700/3907] lr: 8.9862e-05 eta: 5:21:58 time: 0.6302 data_time: 0.0017 memory: 44120 loss: 0.3968 +2023/06/04 03:30:00 - mmengine - INFO - Epoch(train) [3][ 800/3907] lr: 8.9632e-05 eta: 5:20:54 time: 0.6328 data_time: 0.0019 memory: 44120 loss: 0.4027 +2023/06/04 03:31:04 - mmengine - INFO - Epoch(train) [3][ 900/3907] lr: 8.9400e-05 eta: 5:19:51 time: 0.6312 data_time: 0.0017 memory: 44120 loss: 0.3704 +2023/06/04 03:32:07 - mmengine - INFO - Epoch(train) [3][1000/3907] lr: 8.9166e-05 eta: 5:18:47 time: 0.6318 data_time: 0.0019 memory: 44120 loss: 0.4036 +2023/06/04 03:33:10 - mmengine - INFO - Epoch(train) [3][1100/3907] lr: 8.8929e-05 eta: 5:17:45 time: 0.6323 data_time: 0.0024 memory: 44120 loss: 0.3695 +2023/06/04 03:34:05 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230604_015520 +2023/06/04 03:34:13 - mmengine - INFO - Epoch(train) [3][1200/3907] lr: 8.8691e-05 eta: 5:16:42 time: 0.6331 data_time: 0.0022 memory: 44120 loss: 0.3578 +2023/06/04 03:35:17 - mmengine - INFO - Epoch(train) [3][1300/3907] lr: 8.8450e-05 eta: 5:15:38 time: 0.6310 data_time: 0.0018 memory: 44120 loss: 0.3782 +2023/06/04 03:36:20 - mmengine - INFO - Epoch(train) [3][1400/3907] lr: 8.8206e-05 eta: 5:14:35 time: 0.6302 data_time: 0.0017 memory: 44120 loss: 0.3636 +2023/06/04 03:37:23 - mmengine - INFO - Epoch(train) [3][1500/3907] lr: 8.7961e-05 eta: 5:13:32 time: 0.6302 data_time: 0.0015 memory: 44120 loss: 0.3884 +2023/06/04 03:38:26 - mmengine - INFO - Epoch(train) [3][1600/3907] lr: 8.7714e-05 eta: 5:12:28 time: 0.6319 data_time: 0.0018 memory: 44120 loss: 0.3720 +2023/06/04 03:39:29 - mmengine - INFO - Epoch(train) [3][1700/3907] lr: 8.7464e-05 eta: 5:11:25 time: 0.6308 data_time: 0.0015 memory: 44120 loss: 0.3773 +2023/06/04 03:40:32 - mmengine - INFO - Epoch(train) [3][1800/3907] lr: 8.7213e-05 eta: 5:10:21 time: 0.6303 data_time: 0.0016 memory: 44120 loss: 0.3679 +2023/06/04 03:41:36 - mmengine - INFO - Epoch(train) [3][1900/3907] lr: 8.6959e-05 eta: 5:09:18 time: 0.6314 data_time: 0.0024 memory: 44120 loss: 0.3706 +2023/06/04 03:42:39 - mmengine - INFO - Epoch(train) [3][2000/3907] lr: 8.6703e-05 eta: 5:08:14 time: 0.6301 data_time: 0.0017 memory: 44120 loss: 0.3868 +2023/06/04 03:43:42 - mmengine - INFO - Epoch(train) [3][2100/3907] lr: 8.6445e-05 eta: 5:07:11 time: 0.6304 data_time: 0.0016 memory: 44120 loss: 0.3731 +2023/06/04 03:44:36 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230604_015520 +2023/06/04 03:44:45 - mmengine - INFO - Epoch(train) [3][2200/3907] lr: 8.6186e-05 eta: 5:06:07 time: 0.6303 data_time: 0.0017 memory: 44120 loss: 0.3727 +2023/06/04 03:45:48 - mmengine - INFO - Epoch(train) [3][2300/3907] lr: 8.5924e-05 eta: 5:05:03 time: 0.6306 data_time: 0.0015 memory: 44120 loss: 0.3646 +2023/06/04 03:46:51 - mmengine - INFO - Epoch(train) [3][2400/3907] lr: 8.5660e-05 eta: 5:04:00 time: 0.6310 data_time: 0.0015 memory: 44120 loss: 0.3594 +2023/06/04 03:47:54 - mmengine - INFO - Epoch(train) [3][2500/3907] lr: 8.5394e-05 eta: 5:02:57 time: 0.6309 data_time: 0.0018 memory: 44120 loss: 0.3431 +2023/06/04 03:48:57 - mmengine - INFO - Epoch(train) [3][2600/3907] lr: 8.5126e-05 eta: 5:01:53 time: 0.6306 data_time: 0.0018 memory: 44120 loss: 0.3686 +2023/06/04 03:50:01 - mmengine - INFO - Epoch(train) [3][2700/3907] lr: 8.4856e-05 eta: 5:00:50 time: 0.6308 data_time: 0.0016 memory: 44120 loss: 0.3896 +2023/06/04 03:51:04 - mmengine - INFO - Epoch(train) [3][2800/3907] lr: 8.4585e-05 eta: 4:59:46 time: 0.6305 data_time: 0.0017 memory: 44120 loss: 0.3783 +2023/06/04 03:52:07 - mmengine - INFO - Epoch(train) [3][2900/3907] lr: 8.4311e-05 eta: 4:58:43 time: 0.6306 data_time: 0.0017 memory: 44120 loss: 0.3652 +2023/06/04 03:53:10 - mmengine - INFO - Epoch(train) [3][3000/3907] lr: 8.4036e-05 eta: 4:57:40 time: 0.6325 data_time: 0.0025 memory: 44120 loss: 0.3763 +2023/06/04 03:54:14 - mmengine - INFO - Epoch(train) [3][3100/3907] lr: 8.3758e-05 eta: 4:56:37 time: 0.6315 data_time: 0.0019 memory: 44120 loss: 0.3439 +2023/06/04 03:55:08 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230604_015520 +2023/06/04 03:55:17 - mmengine - INFO - Epoch(train) [3][3200/3907] lr: 8.3479e-05 eta: 4:55:34 time: 0.6343 data_time: 0.0018 memory: 44120 loss: 0.3857 +2023/06/04 03:56:20 - mmengine - INFO - Epoch(train) [3][3300/3907] lr: 8.3198e-05 eta: 4:54:31 time: 0.6319 data_time: 0.0018 memory: 44120 loss: 0.3935 +2023/06/04 03:57:23 - mmengine - INFO - Epoch(train) [3][3400/3907] lr: 8.2915e-05 eta: 4:53:27 time: 0.6364 data_time: 0.0029 memory: 44120 loss: 0.3598 +2023/06/04 03:58:26 - mmengine - INFO - Epoch(train) [3][3500/3907] lr: 8.2630e-05 eta: 4:52:24 time: 0.6318 data_time: 0.0018 memory: 44120 loss: 0.3610 +2023/06/04 03:59:30 - mmengine - INFO - Epoch(train) [3][3600/3907] lr: 8.2344e-05 eta: 4:51:21 time: 0.6315 data_time: 0.0016 memory: 44120 loss: 0.3520 +2023/06/04 04:00:33 - mmengine - INFO - Epoch(train) [3][3700/3907] lr: 8.2056e-05 eta: 4:50:18 time: 0.6312 data_time: 0.0016 memory: 44120 loss: 0.4015 +2023/06/04 04:01:36 - mmengine - INFO - Epoch(train) [3][3800/3907] lr: 8.1765e-05 eta: 4:49:14 time: 0.6312 data_time: 0.0017 memory: 44120 loss: 0.3792 +2023/06/04 04:02:39 - mmengine - INFO - Epoch(train) [3][3900/3907] lr: 8.1474e-05 eta: 4:48:11 time: 0.6305 data_time: 0.0013 memory: 44120 loss: 0.3917 +2023/06/04 04:02:43 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230604_015520 +2023/06/04 04:02:43 - mmengine - INFO - Saving checkpoint at 3 epochs +2023/06/04 04:04:18 - mmengine - INFO - Epoch(val) [3][57/57] accuracy/top1: 70.7561 single-label/precision_classwise: [65.50458526611328, 97.20646667480469] single-label/recall_classwise: [99.16040802001953, 35.87596893310547] single-label/f1-score_classwise: [78.8930435180664, 52.40926742553711] data_time: 0.0345 time: 1.2825 +2023/06/04 04:05:25 - mmengine - INFO - Epoch(train) [4][ 100/3907] lr: 8.1160e-05 eta: 4:47:09 time: 0.6294 data_time: 0.0014 memory: 44120 loss: 0.3568 +2023/06/04 04:06:28 - mmengine - INFO - Epoch(train) [4][ 200/3907] lr: 8.0864e-05 eta: 4:46:05 time: 0.6308 data_time: 0.0014 memory: 44120 loss: 0.3732 +2023/06/04 04:07:17 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230604_015520 +2023/06/04 04:07:31 - mmengine - INFO - Epoch(train) [4][ 300/3907] lr: 8.0567e-05 eta: 4:45:02 time: 0.6319 data_time: 0.0015 memory: 44120 loss: 0.3886 +2023/06/04 04:08:34 - mmengine - INFO - Epoch(train) [4][ 400/3907] lr: 8.0269e-05 eta: 4:43:58 time: 0.6312 data_time: 0.0015 memory: 44120 loss: 0.3713 +2023/06/04 04:09:37 - mmengine - INFO - Epoch(train) [4][ 500/3907] lr: 7.9969e-05 eta: 4:42:55 time: 0.6296 data_time: 0.0015 memory: 44120 loss: 0.3773 +2023/06/04 04:10:40 - mmengine - INFO - Epoch(train) [4][ 600/3907] lr: 7.9667e-05 eta: 4:41:51 time: 0.6305 data_time: 0.0014 memory: 44120 loss: 0.3723 +2023/06/04 04:11:43 - mmengine - INFO - Epoch(train) [4][ 700/3907] lr: 7.9363e-05 eta: 4:40:48 time: 0.6314 data_time: 0.0015 memory: 44120 loss: 0.3403 +2023/06/04 04:12:46 - mmengine - INFO - Epoch(train) [4][ 800/3907] lr: 7.9058e-05 eta: 4:39:44 time: 0.6303 data_time: 0.0015 memory: 44120 loss: 0.3501 +2023/06/04 04:13:49 - mmengine - INFO - Epoch(train) [4][ 900/3907] lr: 7.8752e-05 eta: 4:38:41 time: 0.6402 data_time: 0.0018 memory: 44120 loss: 0.3872 +2023/06/04 04:14:52 - mmengine - INFO - Epoch(train) [4][1000/3907] lr: 7.8444e-05 eta: 4:37:37 time: 0.6307 data_time: 0.0019 memory: 44120 loss: 0.3647 +2023/06/04 04:15:55 - mmengine - INFO - Epoch(train) [4][1100/3907] lr: 7.8134e-05 eta: 4:36:34 time: 0.6318 data_time: 0.0023 memory: 44120 loss: 0.3562 +2023/06/04 04:16:59 - mmengine - INFO - Epoch(train) [4][1200/3907] lr: 7.7823e-05 eta: 4:35:31 time: 0.6318 data_time: 0.0021 memory: 44120 loss: 0.3545 +2023/06/04 04:17:49 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230604_015520 +2023/06/04 04:18:02 - mmengine - INFO - Epoch(train) [4][1300/3907] lr: 7.7510e-05 eta: 4:34:28 time: 0.6354 data_time: 0.0024 memory: 44120 loss: 0.3912 +2023/06/04 04:19:05 - mmengine - INFO - Epoch(train) [4][1400/3907] lr: 7.7196e-05 eta: 4:33:25 time: 0.6315 data_time: 0.0022 memory: 44120 loss: 0.3548 +2023/06/04 04:20:09 - mmengine - INFO - Epoch(train) [4][1500/3907] lr: 7.6881e-05 eta: 4:32:22 time: 0.6356 data_time: 0.0025 memory: 44120 loss: 0.3395 +2023/06/04 04:21:12 - mmengine - INFO - Epoch(train) [4][1600/3907] lr: 7.6564e-05 eta: 4:31:19 time: 0.6345 data_time: 0.0024 memory: 44120 loss: 0.3590 +2023/06/04 04:22:15 - mmengine - INFO - Epoch(train) [4][1700/3907] lr: 7.6246e-05 eta: 4:30:16 time: 0.6330 data_time: 0.0026 memory: 44120 loss: 0.3762 +2023/06/04 04:23:19 - mmengine - INFO - Epoch(train) [4][1800/3907] lr: 7.5926e-05 eta: 4:29:13 time: 0.6348 data_time: 0.0026 memory: 44120 loss: 0.3627 +2023/06/04 04:24:22 - mmengine - INFO - Epoch(train) [4][1900/3907] lr: 7.5605e-05 eta: 4:28:10 time: 0.6332 data_time: 0.0026 memory: 44120 loss: 0.3569 +2023/06/04 04:25:26 - mmengine - INFO - Epoch(train) [4][2000/3907] lr: 7.5283e-05 eta: 4:27:07 time: 0.6345 data_time: 0.0024 memory: 44120 loss: 0.3352 +2023/06/04 04:26:29 - mmengine - INFO - Epoch(train) [4][2100/3907] lr: 7.4959e-05 eta: 4:26:04 time: 0.6311 data_time: 0.0021 memory: 44120 loss: 0.3555 +2023/06/04 04:27:32 - mmengine - INFO - Epoch(train) [4][2200/3907] lr: 7.4634e-05 eta: 4:25:01 time: 0.6349 data_time: 0.0021 memory: 44120 loss: 0.3551 +2023/06/04 04:28:22 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230604_015520 +2023/06/04 04:28:36 - mmengine - INFO - Epoch(train) [4][2300/3907] lr: 7.4308e-05 eta: 4:23:58 time: 0.6317 data_time: 0.0022 memory: 44120 loss: 0.3711 +2023/06/04 04:29:39 - mmengine - INFO - Epoch(train) [4][2400/3907] lr: 7.3980e-05 eta: 4:22:54 time: 0.6320 data_time: 0.0022 memory: 44120 loss: 0.3771 +2023/06/04 04:30:42 - mmengine - INFO - Epoch(train) [4][2500/3907] lr: 7.3652e-05 eta: 4:21:51 time: 0.6326 data_time: 0.0020 memory: 44120 loss: 0.3532 +2023/06/04 04:31:45 - mmengine - INFO - Epoch(train) [4][2600/3907] lr: 7.3322e-05 eta: 4:20:48 time: 0.6314 data_time: 0.0019 memory: 44120 loss: 0.4026 +2023/06/04 04:32:49 - mmengine - INFO - Epoch(train) [4][2700/3907] lr: 7.2991e-05 eta: 4:19:46 time: 0.6362 data_time: 0.0023 memory: 44120 loss: 0.3727 +2023/06/04 04:33:52 - mmengine - INFO - Epoch(train) [4][2800/3907] lr: 7.2659e-05 eta: 4:18:43 time: 0.6323 data_time: 0.0021 memory: 44120 loss: 0.3621 +2023/06/04 04:34:56 - mmengine - INFO - Epoch(train) [4][2900/3907] lr: 7.2325e-05 eta: 4:17:40 time: 0.6318 data_time: 0.0018 memory: 44120 loss: 0.3518 +2023/06/04 04:35:59 - mmengine - INFO - Epoch(train) [4][3000/3907] lr: 7.1991e-05 eta: 4:16:36 time: 0.6316 data_time: 0.0019 memory: 44120 loss: 0.3624 +2023/06/04 04:37:02 - mmengine - INFO - Epoch(train) [4][3100/3907] lr: 7.1655e-05 eta: 4:15:33 time: 0.6343 data_time: 0.0021 memory: 44120 loss: 0.3663 +2023/06/04 04:38:06 - mmengine - INFO - Epoch(train) [4][3200/3907] lr: 7.1318e-05 eta: 4:14:30 time: 0.6327 data_time: 0.0022 memory: 44120 loss: 0.3525 +2023/06/04 04:38:55 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230604_015520 +2023/06/04 04:39:09 - mmengine - INFO - Epoch(train) [4][3300/3907] lr: 7.0981e-05 eta: 4:13:27 time: 0.6304 data_time: 0.0018 memory: 44120 loss: 0.3712 +2023/06/04 04:40:12 - mmengine - INFO - Epoch(train) [4][3400/3907] lr: 7.0642e-05 eta: 4:12:23 time: 0.6301 data_time: 0.0016 memory: 44120 loss: 0.3629 +2023/06/04 04:41:15 - mmengine - INFO - Epoch(train) [4][3500/3907] lr: 7.0302e-05 eta: 4:11:20 time: 0.6292 data_time: 0.0014 memory: 44120 loss: 0.4025 +2023/06/04 04:42:18 - mmengine - INFO - Epoch(train) [4][3600/3907] lr: 6.9961e-05 eta: 4:10:17 time: 0.6315 data_time: 0.0015 memory: 44120 loss: 0.3519 +2023/06/04 04:43:21 - mmengine - INFO - Epoch(train) [4][3700/3907] lr: 6.9620e-05 eta: 4:09:13 time: 0.6317 data_time: 0.0015 memory: 44120 loss: 0.3676 +2023/06/04 04:44:25 - mmengine - INFO - Epoch(train) [4][3800/3907] lr: 6.9277e-05 eta: 4:08:10 time: 0.6306 data_time: 0.0016 memory: 44120 loss: 0.3885 +2023/06/04 04:45:28 - mmengine - INFO - Epoch(train) [4][3900/3907] lr: 6.8933e-05 eta: 4:07:07 time: 0.6299 data_time: 0.0012 memory: 44120 loss: 0.3632 +2023/06/04 04:45:32 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230604_015520 +2023/06/04 04:45:32 - mmengine - INFO - Saving checkpoint at 4 epochs +2023/06/04 04:47:05 - mmengine - INFO - Epoch(val) [4][57/57] accuracy/top1: 72.9237 single-label/precision_classwise: [67.2444076538086, 97.6181640625] single-label/recall_classwise: [99.19197082519531, 40.66666793823242] single-label/f1-score_classwise: [80.15201568603516, 57.4149055480957] data_time: 0.0339 time: 1.2814 +2023/06/04 04:48:12 - mmengine - INFO - Epoch(train) [5][ 100/3907] lr: 6.8565e-05 eta: 4:06:03 time: 0.6298 data_time: 0.0015 memory: 44120 loss: 0.3604 +2023/06/04 04:49:15 - mmengine - INFO - Epoch(train) [5][ 200/3907] lr: 6.8219e-05 eta: 4:05:00 time: 0.6306 data_time: 0.0015 memory: 44120 loss: 0.3726 +2023/06/04 04:50:18 - mmengine - INFO - Epoch(train) [5][ 300/3907] lr: 6.7873e-05 eta: 4:03:56 time: 0.6314 data_time: 0.0022 memory: 44120 loss: 0.3698 +2023/06/04 04:51:03 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230604_015520 +2023/06/04 04:51:21 - mmengine - INFO - Epoch(train) [5][ 400/3907] lr: 6.7526e-05 eta: 4:02:53 time: 0.6309 data_time: 0.0016 memory: 44120 loss: 0.3606 +2023/06/04 04:52:24 - mmengine - INFO - Epoch(train) [5][ 500/3907] lr: 6.7178e-05 eta: 4:01:49 time: 0.6309 data_time: 0.0014 memory: 44120 loss: 0.3740 +2023/06/04 04:53:27 - mmengine - INFO - Epoch(train) [5][ 600/3907] lr: 6.6829e-05 eta: 4:00:46 time: 0.6325 data_time: 0.0016 memory: 44120 loss: 0.3751 +2023/06/04 04:54:31 - mmengine - INFO - Epoch(train) [5][ 700/3907] lr: 6.6480e-05 eta: 3:59:43 time: 0.6315 data_time: 0.0014 memory: 44120 loss: 0.3717 +2023/06/04 04:55:34 - mmengine - INFO - Epoch(train) [5][ 800/3907] lr: 6.6129e-05 eta: 3:58:40 time: 0.6328 data_time: 0.0015 memory: 44120 loss: 0.3558 +2023/06/04 04:56:37 - mmengine - INFO - Epoch(train) [5][ 900/3907] lr: 6.5778e-05 eta: 3:57:37 time: 0.6312 data_time: 0.0018 memory: 44120 loss: 0.3636 +2023/06/04 04:57:41 - mmengine - INFO - Epoch(train) [5][1000/3907] lr: 6.5427e-05 eta: 3:56:34 time: 0.6371 data_time: 0.0020 memory: 44120 loss: 0.3688 +2023/06/04 04:58:44 - mmengine - INFO - Epoch(train) [5][1100/3907] lr: 6.5074e-05 eta: 3:55:30 time: 0.6312 data_time: 0.0016 memory: 44120 loss: 0.3715 +2023/06/04 04:59:47 - mmengine - INFO - Epoch(train) [5][1200/3907] lr: 6.4721e-05 eta: 3:54:27 time: 0.6323 data_time: 0.0017 memory: 44120 loss: 0.3440 +2023/06/04 05:00:50 - mmengine - INFO - Epoch(train) [5][1300/3907] lr: 6.4368e-05 eta: 3:53:24 time: 0.6310 data_time: 0.0016 memory: 44120 loss: 0.3613 +2023/06/04 05:01:36 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230604_015520 +2023/06/04 05:01:53 - mmengine - INFO - Epoch(train) [5][1400/3907] lr: 6.4014e-05 eta: 3:52:20 time: 0.6305 data_time: 0.0019 memory: 44120 loss: 0.3369 +2023/06/04 05:02:57 - mmengine - INFO - Epoch(train) [5][1500/3907] lr: 6.3659e-05 eta: 3:51:17 time: 0.6301 data_time: 0.0016 memory: 44120 loss: 0.3567 +2023/06/04 05:04:00 - mmengine - INFO - Epoch(train) [5][1600/3907] lr: 6.3303e-05 eta: 3:50:13 time: 0.6305 data_time: 0.0017 memory: 44120 loss: 0.3875 +2023/06/04 05:05:03 - mmengine - INFO - Epoch(train) [5][1700/3907] lr: 6.2948e-05 eta: 3:49:10 time: 0.6324 data_time: 0.0016 memory: 44120 loss: 0.3812 +2023/06/04 05:06:06 - mmengine - INFO - Epoch(train) [5][1800/3907] lr: 6.2591e-05 eta: 3:48:07 time: 0.6315 data_time: 0.0016 memory: 44120 loss: 0.3553 +2023/06/04 05:07:09 - mmengine - INFO - Epoch(train) [5][1900/3907] lr: 6.2234e-05 eta: 3:47:03 time: 0.6326 data_time: 0.0022 memory: 44120 loss: 0.3960 +2023/06/04 05:08:12 - mmengine - INFO - Epoch(train) [5][2000/3907] lr: 6.1877e-05 eta: 3:46:00 time: 0.6313 data_time: 0.0015 memory: 44120 loss: 0.3601 +2023/06/04 05:09:16 - mmengine - INFO - Epoch(train) [5][2100/3907] lr: 6.1519e-05 eta: 3:44:57 time: 0.6310 data_time: 0.0016 memory: 44120 loss: 0.3430 +2023/06/04 05:10:19 - mmengine - INFO - Epoch(train) [5][2200/3907] lr: 6.1161e-05 eta: 3:43:53 time: 0.6306 data_time: 0.0018 memory: 44120 loss: 0.3600 +2023/06/04 05:11:22 - mmengine - INFO - Epoch(train) [5][2300/3907] lr: 6.0802e-05 eta: 3:42:50 time: 0.6309 data_time: 0.0017 memory: 44120 loss: 0.3507 +2023/06/04 05:12:07 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230604_015520 +2023/06/04 05:12:25 - mmengine - INFO - Epoch(train) [5][2400/3907] lr: 6.0443e-05 eta: 3:41:47 time: 0.6305 data_time: 0.0017 memory: 44120 loss: 0.3818 +2023/06/04 05:13:28 - mmengine - INFO - Epoch(train) [5][2500/3907] lr: 6.0084e-05 eta: 3:40:43 time: 0.6308 data_time: 0.0016 memory: 44120 loss: 0.3692 +2023/06/04 05:14:31 - mmengine - INFO - Epoch(train) [5][2600/3907] lr: 5.9724e-05 eta: 3:39:40 time: 0.6305 data_time: 0.0015 memory: 44120 loss: 0.3737 +2023/06/04 05:15:35 - mmengine - INFO - Epoch(train) [5][2700/3907] lr: 5.9364e-05 eta: 3:38:37 time: 0.6305 data_time: 0.0014 memory: 44120 loss: 0.3773 +2023/06/04 05:16:38 - mmengine - INFO - Epoch(train) [5][2800/3907] lr: 5.9004e-05 eta: 3:37:33 time: 0.6312 data_time: 0.0017 memory: 44120 loss: 0.3454 +2023/06/04 05:17:41 - mmengine - INFO - Epoch(train) [5][2900/3907] lr: 5.8643e-05 eta: 3:36:30 time: 0.6328 data_time: 0.0020 memory: 44120 loss: 0.3517 +2023/06/04 05:18:44 - mmengine - INFO - Epoch(train) [5][3000/3907] lr: 5.8283e-05 eta: 3:35:27 time: 0.6313 data_time: 0.0017 memory: 44120 loss: 0.3615 +2023/06/04 05:19:48 - mmengine - INFO - Epoch(train) [5][3100/3907] lr: 5.7922e-05 eta: 3:34:24 time: 0.6343 data_time: 0.0029 memory: 44120 loss: 0.3724 +2023/06/04 05:20:51 - mmengine - INFO - Epoch(train) [5][3200/3907] lr: 5.7560e-05 eta: 3:33:21 time: 0.6325 data_time: 0.0019 memory: 44120 loss: 0.3680 +2023/06/04 05:21:54 - mmengine - INFO - Epoch(train) [5][3300/3907] lr: 5.7199e-05 eta: 3:32:18 time: 0.6394 data_time: 0.0016 memory: 44120 loss: 0.3743 +2023/06/04 05:22:40 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230604_015520 +2023/06/04 05:22:58 - mmengine - INFO - Epoch(train) [5][3400/3907] lr: 5.6838e-05 eta: 3:31:14 time: 0.6325 data_time: 0.0020 memory: 44120 loss: 0.3637 +2023/06/04 05:24:01 - mmengine - INFO - Epoch(train) [5][3500/3907] lr: 5.6476e-05 eta: 3:30:11 time: 0.6316 data_time: 0.0017 memory: 44120 loss: 0.3521 +2023/06/04 05:25:04 - mmengine - INFO - Epoch(train) [5][3600/3907] lr: 5.6114e-05 eta: 3:29:08 time: 0.6313 data_time: 0.0016 memory: 44120 loss: 0.3706 +2023/06/04 05:26:07 - mmengine - INFO - Epoch(train) [5][3700/3907] lr: 5.5753e-05 eta: 3:28:05 time: 0.6312 data_time: 0.0016 memory: 44120 loss: 0.3684 +2023/06/04 05:27:11 - mmengine - INFO - Epoch(train) [5][3800/3907] lr: 5.5391e-05 eta: 3:27:01 time: 0.6324 data_time: 0.0017 memory: 44120 loss: 0.3916 +2023/06/04 05:28:14 - mmengine - INFO - Epoch(train) [5][3900/3907] lr: 5.5029e-05 eta: 3:25:58 time: 0.6315 data_time: 0.0014 memory: 44120 loss: 0.3690 +2023/06/04 05:28:18 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230604_015520 +2023/06/04 05:28:18 - mmengine - INFO - Saving checkpoint at 5 epochs +2023/06/04 05:29:52 - mmengine - INFO - Epoch(val) [5][57/57] accuracy/top1: 74.1867 single-label/precision_classwise: [68.33740234375, 97.43811798095703] single-label/recall_classwise: [99.06571197509766, 43.635658264160156] single-label/f1-score_classwise: [80.88134002685547, 60.27734375] data_time: 0.0338 time: 1.2835 +2023/06/04 05:30:59 - mmengine - INFO - Epoch(train) [6][ 100/3907] lr: 5.4642e-05 eta: 3:24:53 time: 0.6312 data_time: 0.0014 memory: 44120 loss: 0.3427 +2023/06/04 05:32:02 - mmengine - INFO - Epoch(train) [6][ 200/3907] lr: 5.4280e-05 eta: 3:23:50 time: 0.6307 data_time: 0.0014 memory: 44120 loss: 0.3660 +2023/06/04 05:33:05 - mmengine - INFO - Epoch(train) [6][ 300/3907] lr: 5.3918e-05 eta: 3:22:47 time: 0.6346 data_time: 0.0016 memory: 44120 loss: 0.3612 +2023/06/04 05:34:08 - mmengine - INFO - Epoch(train) [6][ 400/3907] lr: 5.3556e-05 eta: 3:21:43 time: 0.6309 data_time: 0.0014 memory: 44120 loss: 0.3365 +2023/06/04 05:34:49 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230604_015520 +2023/06/04 05:35:11 - mmengine - INFO - Epoch(train) [6][ 500/3907] lr: 5.3195e-05 eta: 3:20:40 time: 0.6317 data_time: 0.0014 memory: 44120 loss: 0.3765 +2023/06/04 05:36:15 - mmengine - INFO - Epoch(train) [6][ 600/3907] lr: 5.2833e-05 eta: 3:19:37 time: 0.6314 data_time: 0.0016 memory: 44120 loss: 0.3631 +2023/06/04 05:37:18 - mmengine - INFO - Epoch(train) [6][ 700/3907] lr: 5.2472e-05 eta: 3:18:34 time: 0.6311 data_time: 0.0017 memory: 44120 loss: 0.3929 +2023/06/04 05:38:21 - mmengine - INFO - Epoch(train) [6][ 800/3907] lr: 5.2111e-05 eta: 3:17:30 time: 0.6329 data_time: 0.0016 memory: 44120 loss: 0.3521 +2023/06/04 05:39:24 - mmengine - INFO - Epoch(train) [6][ 900/3907] lr: 5.1750e-05 eta: 3:16:27 time: 0.6312 data_time: 0.0018 memory: 44120 loss: 0.3688 +2023/06/04 05:40:28 - mmengine - INFO - Epoch(train) [6][1000/3907] lr: 5.1389e-05 eta: 3:15:24 time: 0.6309 data_time: 0.0018 memory: 44120 loss: 0.3505 +2023/06/04 05:41:31 - mmengine - INFO - Epoch(train) [6][1100/3907] lr: 5.1029e-05 eta: 3:14:20 time: 0.6307 data_time: 0.0017 memory: 44120 loss: 0.3884 +2023/06/04 05:42:34 - mmengine - INFO - Epoch(train) [6][1200/3907] lr: 5.0668e-05 eta: 3:13:17 time: 0.6327 data_time: 0.0018 memory: 44120 loss: 0.3525 +2023/06/04 05:43:37 - mmengine - INFO - Epoch(train) [6][1300/3907] lr: 5.0308e-05 eta: 3:12:14 time: 0.6351 data_time: 0.0016 memory: 44120 loss: 0.3769 +2023/06/04 05:44:41 - mmengine - INFO - Epoch(train) [6][1400/3907] lr: 4.9949e-05 eta: 3:11:11 time: 0.6432 data_time: 0.0016 memory: 44120 loss: 0.3827 +2023/06/04 05:45:22 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230604_015520 +2023/06/04 05:45:44 - mmengine - INFO - Epoch(train) [6][1500/3907] lr: 4.9589e-05 eta: 3:10:08 time: 0.6320 data_time: 0.0017 memory: 44120 loss: 0.3490 +2023/06/04 05:46:47 - mmengine - INFO - Epoch(train) [6][1600/3907] lr: 4.9230e-05 eta: 3:09:04 time: 0.6321 data_time: 0.0017 memory: 44120 loss: 0.3568 +2023/06/04 05:47:51 - mmengine - INFO - Epoch(train) [6][1700/3907] lr: 4.8871e-05 eta: 3:08:01 time: 0.6317 data_time: 0.0018 memory: 44120 loss: 0.3654 +2023/06/04 05:48:54 - mmengine - INFO - Epoch(train) [6][1800/3907] lr: 4.8513e-05 eta: 3:06:58 time: 0.6306 data_time: 0.0017 memory: 44120 loss: 0.3845 +2023/06/04 05:49:57 - mmengine - INFO - Epoch(train) [6][1900/3907] lr: 4.8155e-05 eta: 3:05:54 time: 0.6317 data_time: 0.0015 memory: 44120 loss: 0.3605 +2023/06/04 05:51:00 - mmengine - INFO - Epoch(train) [6][2000/3907] lr: 4.7798e-05 eta: 3:04:51 time: 0.6315 data_time: 0.0015 memory: 44120 loss: 0.3793 +2023/06/04 05:52:03 - mmengine - INFO - Epoch(train) [6][2100/3907] lr: 4.7441e-05 eta: 3:03:48 time: 0.6309 data_time: 0.0015 memory: 44120 loss: 0.3866 +2023/06/04 05:53:06 - mmengine - INFO - Epoch(train) [6][2200/3907] lr: 4.7084e-05 eta: 3:02:44 time: 0.6312 data_time: 0.0015 memory: 44120 loss: 0.3695 +2023/06/04 05:54:09 - mmengine - INFO - Epoch(train) [6][2300/3907] lr: 4.6729e-05 eta: 3:01:41 time: 0.6308 data_time: 0.0016 memory: 44120 loss: 0.3582 +2023/06/04 05:55:13 - mmengine - INFO - Epoch(train) [6][2400/3907] lr: 4.6373e-05 eta: 3:00:38 time: 0.6326 data_time: 0.0019 memory: 44120 loss: 0.3662 +2023/06/04 05:55:54 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230604_015520 +2023/06/04 05:56:16 - mmengine - INFO - Epoch(train) [6][2500/3907] lr: 4.6018e-05 eta: 2:59:34 time: 0.6340 data_time: 0.0015 memory: 44120 loss: 0.3825 +2023/06/04 05:57:19 - mmengine - INFO - Epoch(train) [6][2600/3907] lr: 4.5664e-05 eta: 2:58:31 time: 0.6347 data_time: 0.0019 memory: 44120 loss: 0.3929 +2023/06/04 05:58:22 - mmengine - INFO - Epoch(train) [6][2700/3907] lr: 4.5310e-05 eta: 2:57:28 time: 0.6304 data_time: 0.0017 memory: 44120 loss: 0.3461 +2023/06/04 05:59:26 - mmengine - INFO - Epoch(train) [6][2800/3907] lr: 4.4957e-05 eta: 2:56:25 time: 0.6325 data_time: 0.0017 memory: 44120 loss: 0.3413 +2023/06/04 06:00:29 - mmengine - INFO - Epoch(train) [6][2900/3907] lr: 4.4605e-05 eta: 2:55:21 time: 0.6322 data_time: 0.0015 memory: 44120 loss: 0.3636 +2023/06/04 06:01:32 - mmengine - INFO - Epoch(train) [6][3000/3907] lr: 4.4253e-05 eta: 2:54:18 time: 0.6305 data_time: 0.0017 memory: 44120 loss: 0.3834 +2023/06/04 06:02:35 - mmengine - INFO - Epoch(train) [6][3100/3907] lr: 4.3902e-05 eta: 2:53:15 time: 0.6302 data_time: 0.0016 memory: 44120 loss: 0.3630 +2023/06/04 06:03:38 - mmengine - INFO - Epoch(train) [6][3200/3907] lr: 4.3552e-05 eta: 2:52:11 time: 0.6307 data_time: 0.0015 memory: 44120 loss: 0.3936 +2023/06/04 06:04:41 - mmengine - INFO - Epoch(train) [6][3300/3907] lr: 4.3202e-05 eta: 2:51:08 time: 0.6331 data_time: 0.0015 memory: 44120 loss: 0.3545 +2023/06/04 06:05:44 - mmengine - INFO - Epoch(train) [6][3400/3907] lr: 4.2854e-05 eta: 2:50:05 time: 0.6330 data_time: 0.0015 memory: 44120 loss: 0.3753 +2023/06/04 06:06:26 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230604_015520 +2023/06/04 06:06:48 - mmengine - INFO - Epoch(train) [6][3500/3907] lr: 4.2506e-05 eta: 2:49:01 time: 0.6303 data_time: 0.0017 memory: 44120 loss: 0.3990 +2023/06/04 06:07:51 - mmengine - INFO - Epoch(train) [6][3600/3907] lr: 4.2158e-05 eta: 2:47:58 time: 0.6320 data_time: 0.0019 memory: 44120 loss: 0.3641 +2023/06/04 06:08:54 - mmengine - INFO - Epoch(train) [6][3700/3907] lr: 4.1812e-05 eta: 2:46:55 time: 0.6305 data_time: 0.0018 memory: 44120 loss: 0.3594 +2023/06/04 06:09:57 - mmengine - INFO - Epoch(train) [6][3800/3907] lr: 4.1466e-05 eta: 2:45:51 time: 0.6327 data_time: 0.0014 memory: 44120 loss: 0.3692 +2023/06/04 06:11:00 - mmengine - INFO - Epoch(train) [6][3900/3907] lr: 4.1122e-05 eta: 2:44:48 time: 0.6310 data_time: 0.0015 memory: 44120 loss: 0.3825 +2023/06/04 06:11:04 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230604_015520 +2023/06/04 06:11:04 - mmengine - INFO - Saving checkpoint at 6 epochs +2023/06/04 06:12:38 - mmengine - INFO - Epoch(val) [6][57/57] accuracy/top1: 74.7608 single-label/precision_classwise: [68.8204116821289, 97.6214599609375] single-label/recall_classwise: [99.10990905761719, 44.86046600341797] single-label/f1-score_classwise: [81.23350524902344, 61.47227478027344] data_time: 0.0361 time: 1.2847 +2023/06/04 06:13:44 - mmengine - INFO - Epoch(train) [7][ 100/3907] lr: 4.0754e-05 eta: 2:43:42 time: 0.6305 data_time: 0.0017 memory: 44120 loss: 0.3907 +2023/06/04 06:14:48 - mmengine - INFO - Epoch(train) [7][ 200/3907] lr: 4.0411e-05 eta: 2:42:39 time: 0.6326 data_time: 0.0020 memory: 44120 loss: 0.3554 +2023/06/04 06:15:51 - mmengine - INFO - Epoch(train) [7][ 300/3907] lr: 4.0069e-05 eta: 2:41:36 time: 0.6350 data_time: 0.0030 memory: 44120 loss: 0.3805 +2023/06/04 06:16:54 - mmengine - INFO - Epoch(train) [7][ 400/3907] lr: 3.9729e-05 eta: 2:40:32 time: 0.6355 data_time: 0.0032 memory: 44120 loss: 0.3926 +2023/06/04 06:17:58 - mmengine - INFO - Epoch(train) [7][ 500/3907] lr: 3.9389e-05 eta: 2:39:29 time: 0.6364 data_time: 0.0028 memory: 44120 loss: 0.3541 +2023/06/04 06:18:34 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230604_015520 +2023/06/04 06:19:01 - mmengine - INFO - Epoch(train) [7][ 600/3907] lr: 3.9050e-05 eta: 2:38:26 time: 0.6305 data_time: 0.0019 memory: 44120 loss: 0.3664 +2023/06/04 06:20:04 - mmengine - INFO - Epoch(train) [7][ 700/3907] lr: 3.8712e-05 eta: 2:37:23 time: 0.6317 data_time: 0.0014 memory: 44120 loss: 0.3534 +2023/06/04 06:21:07 - mmengine - INFO - Epoch(train) [7][ 800/3907] lr: 3.8375e-05 eta: 2:36:19 time: 0.6323 data_time: 0.0018 memory: 44120 loss: 0.3657 +2023/06/04 06:22:11 - mmengine - INFO - Epoch(train) [7][ 900/3907] lr: 3.8039e-05 eta: 2:35:16 time: 0.6302 data_time: 0.0019 memory: 44120 loss: 0.3790 +2023/06/04 06:23:14 - mmengine - INFO - Epoch(train) [7][1000/3907] lr: 3.7705e-05 eta: 2:34:13 time: 0.6311 data_time: 0.0015 memory: 44120 loss: 0.3830 +2023/06/04 06:24:17 - mmengine - INFO - Epoch(train) [7][1100/3907] lr: 3.7371e-05 eta: 2:33:10 time: 0.6312 data_time: 0.0018 memory: 44120 loss: 0.3756 +2023/06/04 06:25:20 - mmengine - INFO - Epoch(train) [7][1200/3907] lr: 3.7039e-05 eta: 2:32:06 time: 0.6304 data_time: 0.0016 memory: 44120 loss: 0.3716 +2023/06/04 06:26:23 - mmengine - INFO - Epoch(train) [7][1300/3907] lr: 3.6708e-05 eta: 2:31:03 time: 0.6307 data_time: 0.0014 memory: 44120 loss: 0.3579 +2023/06/04 06:27:26 - mmengine - INFO - Epoch(train) [7][1400/3907] lr: 3.6378e-05 eta: 2:30:00 time: 0.6336 data_time: 0.0019 memory: 44120 loss: 0.4021 +2023/06/04 06:28:30 - mmengine - INFO - Epoch(train) [7][1500/3907] lr: 3.6049e-05 eta: 2:28:56 time: 0.6413 data_time: 0.0021 memory: 44120 loss: 0.3833 +2023/06/04 06:29:07 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230604_015520 +2023/06/04 06:29:33 - mmengine - INFO - Epoch(train) [7][1600/3907] lr: 3.5721e-05 eta: 2:27:53 time: 0.6323 data_time: 0.0022 memory: 44120 loss: 0.3706 +2023/06/04 06:30:36 - mmengine - INFO - Epoch(train) [7][1700/3907] lr: 3.5395e-05 eta: 2:26:50 time: 0.6330 data_time: 0.0016 memory: 44120 loss: 0.3531 +2023/06/04 06:31:40 - mmengine - INFO - Epoch(train) [7][1800/3907] lr: 3.5070e-05 eta: 2:25:47 time: 0.6315 data_time: 0.0016 memory: 44120 loss: 0.3761 +2023/06/04 06:32:43 - mmengine - INFO - Epoch(train) [7][1900/3907] lr: 3.4746e-05 eta: 2:24:43 time: 0.6312 data_time: 0.0017 memory: 44120 loss: 0.3810 +2023/06/04 06:33:46 - mmengine - INFO - Epoch(train) [7][2000/3907] lr: 3.4424e-05 eta: 2:23:40 time: 0.6312 data_time: 0.0017 memory: 44120 loss: 0.4143 +2023/06/04 06:34:49 - mmengine - INFO - Epoch(train) [7][2100/3907] lr: 3.4103e-05 eta: 2:22:37 time: 0.6315 data_time: 0.0017 memory: 44120 loss: 0.3733 +2023/06/04 06:35:52 - mmengine - INFO - Epoch(train) [7][2200/3907] lr: 3.3783e-05 eta: 2:21:33 time: 0.6352 data_time: 0.0023 memory: 44120 loss: 0.3778 +2023/06/04 06:36:56 - mmengine - INFO - Epoch(train) [7][2300/3907] lr: 3.3465e-05 eta: 2:20:30 time: 0.6337 data_time: 0.0019 memory: 44120 loss: 0.3983 +2023/06/04 06:37:59 - mmengine - INFO - Epoch(train) [7][2400/3907] lr: 3.3148e-05 eta: 2:19:27 time: 0.6315 data_time: 0.0019 memory: 44120 loss: 0.3413 +2023/06/04 06:39:02 - mmengine - INFO - Epoch(train) [7][2500/3907] lr: 3.2832e-05 eta: 2:18:24 time: 0.6326 data_time: 0.0016 memory: 44120 loss: 0.3760 +2023/06/04 06:39:39 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230604_015520 +2023/06/04 06:40:06 - mmengine - INFO - Epoch(train) [7][2600/3907] lr: 3.2518e-05 eta: 2:17:21 time: 0.6311 data_time: 0.0017 memory: 44120 loss: 0.3815 +2023/06/04 06:41:09 - mmengine - INFO - Epoch(train) [7][2700/3907] lr: 3.2205e-05 eta: 2:16:17 time: 0.6312 data_time: 0.0017 memory: 44120 loss: 0.3696 +2023/06/04 06:42:12 - mmengine - INFO - Epoch(train) [7][2800/3907] lr: 3.1894e-05 eta: 2:15:14 time: 0.6324 data_time: 0.0018 memory: 44120 loss: 0.3680 +2023/06/04 06:43:15 - mmengine - INFO - Epoch(train) [7][2900/3907] lr: 3.1584e-05 eta: 2:14:11 time: 0.6307 data_time: 0.0015 memory: 44120 loss: 0.3825 +2023/06/04 06:44:19 - mmengine - INFO - Epoch(train) [7][3000/3907] lr: 3.1276e-05 eta: 2:13:07 time: 0.6309 data_time: 0.0016 memory: 44120 loss: 0.3610 +2023/06/04 06:45:22 - mmengine - INFO - Epoch(train) [7][3100/3907] lr: 3.0969e-05 eta: 2:12:04 time: 0.6306 data_time: 0.0016 memory: 44120 loss: 0.3675 +2023/06/04 06:46:25 - mmengine - INFO - Epoch(train) [7][3200/3907] lr: 3.0664e-05 eta: 2:11:01 time: 0.6312 data_time: 0.0015 memory: 44120 loss: 0.3705 +2023/06/04 06:47:28 - mmengine - INFO - Epoch(train) [7][3300/3907] lr: 3.0360e-05 eta: 2:09:58 time: 0.6323 data_time: 0.0020 memory: 44120 loss: 0.3740 +2023/06/04 06:48:31 - mmengine - INFO - Epoch(train) [7][3400/3907] lr: 3.0058e-05 eta: 2:08:54 time: 0.6343 data_time: 0.0016 memory: 44120 loss: 0.3671 +2023/06/04 06:49:35 - mmengine - INFO - Epoch(train) [7][3500/3907] lr: 2.9758e-05 eta: 2:07:51 time: 0.6317 data_time: 0.0019 memory: 44120 loss: 0.3717 +2023/06/04 06:50:11 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230604_015520 +2023/06/04 06:50:38 - mmengine - INFO - Epoch(train) [7][3600/3907] lr: 2.9459e-05 eta: 2:06:48 time: 0.6339 data_time: 0.0016 memory: 44120 loss: 0.3721 +2023/06/04 06:51:41 - mmengine - INFO - Epoch(train) [7][3700/3907] lr: 2.9162e-05 eta: 2:05:45 time: 0.6303 data_time: 0.0016 memory: 44120 loss: 0.3967 +2023/06/04 06:52:44 - mmengine - INFO - Epoch(train) [7][3800/3907] lr: 2.8867e-05 eta: 2:04:41 time: 0.6320 data_time: 0.0018 memory: 44120 loss: 0.3844 +2023/06/04 06:53:47 - mmengine - INFO - Epoch(train) [7][3900/3907] lr: 2.8573e-05 eta: 2:03:38 time: 0.6294 data_time: 0.0015 memory: 44120 loss: 0.3524 +2023/06/04 06:53:51 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230604_015520 +2023/06/04 06:53:51 - mmengine - INFO - Saving checkpoint at 7 epochs +2023/06/04 06:55:25 - mmengine - INFO - Epoch(val) [7][57/57] accuracy/top1: 75.0357 single-label/precision_classwise: [69.0578842163086, 97.66860961914062] single-label/recall_classwise: [99.11621856689453, 45.465118408203125] single-label/f1-score_classwise: [81.40083312988281, 62.047080993652344] data_time: 0.0318 time: 1.2814 +2023/06/04 06:56:31 - mmengine - INFO - Epoch(train) [8][ 100/3907] lr: 2.8261e-05 eta: 2:02:31 time: 0.6305 data_time: 0.0018 memory: 44120 loss: 0.3888 +2023/06/04 06:57:35 - mmengine - INFO - Epoch(train) [8][ 200/3907] lr: 2.7971e-05 eta: 2:01:28 time: 0.6301 data_time: 0.0014 memory: 44120 loss: 0.3569 +2023/06/04 06:58:38 - mmengine - INFO - Epoch(train) [8][ 300/3907] lr: 2.7682e-05 eta: 2:00:25 time: 0.6309 data_time: 0.0015 memory: 44120 loss: 0.3666 +2023/06/04 06:59:41 - mmengine - INFO - Epoch(train) [8][ 400/3907] lr: 2.7395e-05 eta: 1:59:21 time: 0.6302 data_time: 0.0019 memory: 44120 loss: 0.3775 +2023/06/04 07:00:44 - mmengine - INFO - Epoch(train) [8][ 500/3907] lr: 2.7111e-05 eta: 1:58:18 time: 0.6320 data_time: 0.0017 memory: 44120 loss: 0.3902 +2023/06/04 07:01:47 - mmengine - INFO - Epoch(train) [8][ 600/3907] lr: 2.6828e-05 eta: 1:57:15 time: 0.6325 data_time: 0.0020 memory: 44120 loss: 0.3546 +2023/06/04 07:02:19 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230604_015520 +2023/06/04 07:02:50 - mmengine - INFO - Epoch(train) [8][ 700/3907] lr: 2.6546e-05 eta: 1:56:11 time: 0.6310 data_time: 0.0015 memory: 44120 loss: 0.3699 +2023/06/04 07:03:54 - mmengine - INFO - Epoch(train) [8][ 800/3907] lr: 2.6267e-05 eta: 1:55:08 time: 0.6330 data_time: 0.0023 memory: 44120 loss: 0.3367 +2023/06/04 07:04:57 - mmengine - INFO - Epoch(train) [8][ 900/3907] lr: 2.5989e-05 eta: 1:54:05 time: 0.6312 data_time: 0.0017 memory: 44120 loss: 0.3577 +2023/06/04 07:06:00 - mmengine - INFO - Epoch(train) [8][1000/3907] lr: 2.5714e-05 eta: 1:53:02 time: 0.6319 data_time: 0.0017 memory: 44120 loss: 0.3707 +2023/06/04 07:07:03 - mmengine - INFO - Epoch(train) [8][1100/3907] lr: 2.5440e-05 eta: 1:51:58 time: 0.6332 data_time: 0.0019 memory: 44120 loss: 0.3652 +2023/06/04 07:08:07 - mmengine - INFO - Epoch(train) [8][1200/3907] lr: 2.5168e-05 eta: 1:50:55 time: 0.6309 data_time: 0.0016 memory: 44120 loss: 0.3963 +2023/06/04 07:09:10 - mmengine - INFO - Epoch(train) [8][1300/3907] lr: 2.4898e-05 eta: 1:49:52 time: 0.6313 data_time: 0.0017 memory: 44120 loss: 0.3455 +2023/06/04 07:10:13 - mmengine - INFO - Epoch(train) [8][1400/3907] lr: 2.4630e-05 eta: 1:48:48 time: 0.6302 data_time: 0.0015 memory: 44120 loss: 0.3566 +2023/06/04 07:11:16 - mmengine - INFO - Epoch(train) [8][1500/3907] lr: 2.4364e-05 eta: 1:47:45 time: 0.6316 data_time: 0.0017 memory: 44120 loss: 0.3558 +2023/06/04 07:12:19 - mmengine - INFO - Epoch(train) [8][1600/3907] lr: 2.4100e-05 eta: 1:46:42 time: 0.6453 data_time: 0.0017 memory: 44120 loss: 0.4059 +2023/06/04 07:12:52 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230604_015520 +2023/06/04 07:13:23 - mmengine - INFO - Epoch(train) [8][1700/3907] lr: 2.3838e-05 eta: 1:45:39 time: 0.6318 data_time: 0.0020 memory: 44120 loss: 0.3632 +2023/06/04 07:14:26 - mmengine - INFO - Epoch(train) [8][1800/3907] lr: 2.3578e-05 eta: 1:44:35 time: 0.6327 data_time: 0.0017 memory: 44120 loss: 0.3727 +2023/06/04 07:15:29 - mmengine - INFO - Epoch(train) [8][1900/3907] lr: 2.3320e-05 eta: 1:43:32 time: 0.6314 data_time: 0.0016 memory: 44120 loss: 0.3891 +2023/06/04 07:16:32 - mmengine - INFO - Epoch(train) [8][2000/3907] lr: 2.3064e-05 eta: 1:42:29 time: 0.6343 data_time: 0.0015 memory: 44120 loss: 0.3707 +2023/06/04 07:17:35 - mmengine - INFO - Epoch(train) [8][2100/3907] lr: 2.2810e-05 eta: 1:41:26 time: 0.6312 data_time: 0.0015 memory: 44120 loss: 0.3851 +2023/06/04 07:18:38 - mmengine - INFO - Epoch(train) [8][2200/3907] lr: 2.2558e-05 eta: 1:40:22 time: 0.6336 data_time: 0.0016 memory: 44120 loss: 0.3780 +2023/06/04 07:19:42 - mmengine - INFO - Epoch(train) [8][2300/3907] lr: 2.2309e-05 eta: 1:39:19 time: 0.6311 data_time: 0.0020 memory: 44120 loss: 0.3452 +2023/06/04 07:20:45 - mmengine - INFO - Epoch(train) [8][2400/3907] lr: 2.2061e-05 eta: 1:38:16 time: 0.6307 data_time: 0.0016 memory: 44120 loss: 0.3977 +2023/06/04 07:21:48 - mmengine - INFO - Epoch(train) [8][2500/3907] lr: 2.1816e-05 eta: 1:37:12 time: 0.6324 data_time: 0.0015 memory: 44120 loss: 0.3869 +2023/06/04 07:22:51 - mmengine - INFO - Epoch(train) [8][2600/3907] lr: 2.1572e-05 eta: 1:36:09 time: 0.6331 data_time: 0.0016 memory: 44120 loss: 0.3647 +2023/06/04 07:23:23 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230604_015520 +2023/06/04 07:23:54 - mmengine - INFO - Epoch(train) [8][2700/3907] lr: 2.1331e-05 eta: 1:35:06 time: 0.6307 data_time: 0.0014 memory: 44120 loss: 0.3893 +2023/06/04 07:24:58 - mmengine - INFO - Epoch(train) [8][2800/3907] lr: 2.1092e-05 eta: 1:34:03 time: 0.6316 data_time: 0.0016 memory: 44120 loss: 0.3776 +2023/06/04 07:26:01 - mmengine - INFO - Epoch(train) [8][2900/3907] lr: 2.0855e-05 eta: 1:32:59 time: 0.6313 data_time: 0.0014 memory: 44120 loss: 0.3686 +2023/06/04 07:27:04 - mmengine - INFO - Epoch(train) [8][3000/3907] lr: 2.0621e-05 eta: 1:31:56 time: 0.6353 data_time: 0.0014 memory: 44120 loss: 0.3694 +2023/06/04 07:28:08 - mmengine - INFO - Epoch(train) [8][3100/3907] lr: 2.0388e-05 eta: 1:30:53 time: 0.6312 data_time: 0.0014 memory: 44120 loss: 0.4033 +2023/06/04 07:29:11 - mmengine - INFO - Epoch(train) [8][3200/3907] lr: 2.0158e-05 eta: 1:29:50 time: 0.6298 data_time: 0.0016 memory: 44120 loss: 0.3626 +2023/06/04 07:30:14 - mmengine - INFO - Epoch(train) [8][3300/3907] lr: 1.9930e-05 eta: 1:28:46 time: 0.6325 data_time: 0.0015 memory: 44120 loss: 0.3532 +2023/06/04 07:31:17 - mmengine - INFO - Epoch(train) [8][3400/3907] lr: 1.9705e-05 eta: 1:27:43 time: 0.6316 data_time: 0.0016 memory: 44120 loss: 0.3608 +2023/06/04 07:32:20 - mmengine - INFO - Epoch(train) [8][3500/3907] lr: 1.9481e-05 eta: 1:26:40 time: 0.6312 data_time: 0.0014 memory: 44120 loss: 0.3749 +2023/06/04 07:33:23 - mmengine - INFO - Epoch(train) [8][3600/3907] lr: 1.9260e-05 eta: 1:25:36 time: 0.6321 data_time: 0.0019 memory: 44120 loss: 0.3705 +2023/06/04 07:33:56 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230604_015520 +2023/06/04 07:34:27 - mmengine - INFO - Epoch(train) [8][3700/3907] lr: 1.9042e-05 eta: 1:24:33 time: 0.6339 data_time: 0.0018 memory: 44120 loss: 0.3659 +2023/06/04 07:35:30 - mmengine - INFO - Epoch(train) [8][3800/3907] lr: 1.8825e-05 eta: 1:23:30 time: 0.6306 data_time: 0.0015 memory: 44120 loss: 0.3681 +2023/06/04 07:36:33 - mmengine - INFO - Epoch(train) [8][3900/3907] lr: 1.8611e-05 eta: 1:22:27 time: 0.6309 data_time: 0.0013 memory: 44120 loss: 0.3498 +2023/06/04 07:36:37 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230604_015520 +2023/06/04 07:36:37 - mmengine - INFO - Saving checkpoint at 8 epochs +2023/06/04 07:38:11 - mmengine - INFO - Epoch(val) [8][57/57] accuracy/top1: 75.3175 single-label/precision_classwise: [69.3047866821289, 97.69963836669922] single-label/recall_classwise: [99.11621856689453, 46.093021392822266] single-label/f1-score_classwise: [81.57211303710938, 62.635623931884766] data_time: 0.0382 time: 1.2858 +2023/06/04 07:39:18 - mmengine - INFO - Epoch(train) [9][ 100/3907] lr: 1.8385e-05 eta: 1:21:20 time: 0.6300 data_time: 0.0015 memory: 44120 loss: 0.3670 +2023/06/04 07:40:21 - mmengine - INFO - Epoch(train) [9][ 200/3907] lr: 1.8176e-05 eta: 1:20:16 time: 0.6300 data_time: 0.0016 memory: 44120 loss: 0.3762 +2023/06/04 07:41:24 - mmengine - INFO - Epoch(train) [9][ 300/3907] lr: 1.7969e-05 eta: 1:19:13 time: 0.6309 data_time: 0.0017 memory: 44120 loss: 0.3762 +2023/06/04 07:42:27 - mmengine - INFO - Epoch(train) [9][ 400/3907] lr: 1.7765e-05 eta: 1:18:10 time: 0.6331 data_time: 0.0024 memory: 44120 loss: 0.3647 +2023/06/04 07:43:31 - mmengine - INFO - Epoch(train) [9][ 500/3907] lr: 1.7563e-05 eta: 1:17:07 time: 0.6323 data_time: 0.0018 memory: 44120 loss: 0.3522 +2023/06/04 07:44:34 - mmengine - INFO - Epoch(train) [9][ 600/3907] lr: 1.7363e-05 eta: 1:16:03 time: 0.6366 data_time: 0.0020 memory: 44120 loss: 0.3591 +2023/06/04 07:45:37 - mmengine - INFO - Epoch(train) [9][ 700/3907] lr: 1.7166e-05 eta: 1:15:00 time: 0.6315 data_time: 0.0016 memory: 44120 loss: 0.3492 +2023/06/04 07:46:05 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230604_015520 +2023/06/04 07:46:40 - mmengine - INFO - Epoch(train) [9][ 800/3907] lr: 1.6971e-05 eta: 1:13:57 time: 0.6314 data_time: 0.0019 memory: 44120 loss: 0.3773 +2023/06/04 07:47:44 - mmengine - INFO - Epoch(train) [9][ 900/3907] lr: 1.6779e-05 eta: 1:12:53 time: 0.6309 data_time: 0.0015 memory: 44120 loss: 0.3667 +2023/06/04 07:48:47 - mmengine - INFO - Epoch(train) [9][1000/3907] lr: 1.6589e-05 eta: 1:11:50 time: 0.6315 data_time: 0.0017 memory: 44120 loss: 0.3766 +2023/06/04 07:49:50 - mmengine - INFO - Epoch(train) [9][1100/3907] lr: 1.6402e-05 eta: 1:10:47 time: 0.6338 data_time: 0.0017 memory: 44120 loss: 0.3595 +2023/06/04 07:50:53 - mmengine - INFO - Epoch(train) [9][1200/3907] lr: 1.6217e-05 eta: 1:09:44 time: 0.6327 data_time: 0.0016 memory: 44120 loss: 0.3981 +2023/06/04 07:51:57 - mmengine - INFO - Epoch(train) [9][1300/3907] lr: 1.6035e-05 eta: 1:08:40 time: 0.6316 data_time: 0.0019 memory: 44120 loss: 0.3540 +2023/06/04 07:53:00 - mmengine - INFO - Epoch(train) [9][1400/3907] lr: 1.5855e-05 eta: 1:07:37 time: 0.6316 data_time: 0.0015 memory: 44120 loss: 0.3799 +2023/06/04 07:54:03 - mmengine - INFO - Epoch(train) [9][1500/3907] lr: 1.5678e-05 eta: 1:06:34 time: 0.6313 data_time: 0.0015 memory: 44120 loss: 0.3942 +2023/06/04 07:55:06 - mmengine - INFO - Epoch(train) [9][1600/3907] lr: 1.5503e-05 eta: 1:05:31 time: 0.6313 data_time: 0.0018 memory: 44120 loss: 0.3615 +2023/06/04 07:56:09 - mmengine - INFO - Epoch(train) [9][1700/3907] lr: 1.5331e-05 eta: 1:04:27 time: 0.6343 data_time: 0.0014 memory: 44120 loss: 0.3896 +2023/06/04 07:56:37 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230604_015520 +2023/06/04 07:57:13 - mmengine - INFO - Epoch(train) [9][1800/3907] lr: 1.5162e-05 eta: 1:03:24 time: 0.6307 data_time: 0.0016 memory: 44120 loss: 0.3814 +2023/06/04 07:58:16 - mmengine - INFO - Epoch(train) [9][1900/3907] lr: 1.4995e-05 eta: 1:02:21 time: 0.6312 data_time: 0.0015 memory: 44120 loss: 0.3929 +2023/06/04 07:59:19 - mmengine - INFO - Epoch(train) [9][2000/3907] lr: 1.4830e-05 eta: 1:01:17 time: 0.6309 data_time: 0.0018 memory: 44120 loss: 0.3716 +2023/06/04 08:00:22 - mmengine - INFO - Epoch(train) [9][2100/3907] lr: 1.4668e-05 eta: 1:00:14 time: 0.6308 data_time: 0.0017 memory: 44120 loss: 0.3421 +2023/06/04 08:01:25 - mmengine - INFO - Epoch(train) [9][2200/3907] lr: 1.4509e-05 eta: 0:59:11 time: 0.6317 data_time: 0.0016 memory: 44120 loss: 0.3499 +2023/06/04 08:02:29 - mmengine - INFO - Epoch(train) [9][2300/3907] lr: 1.4353e-05 eta: 0:58:08 time: 0.6311 data_time: 0.0017 memory: 44120 loss: 0.3561 +2023/06/04 08:03:32 - mmengine - INFO - Epoch(train) [9][2400/3907] lr: 1.4199e-05 eta: 0:57:04 time: 0.6342 data_time: 0.0017 memory: 44120 loss: 0.4037 +2023/06/04 08:04:35 - mmengine - INFO - Epoch(train) [9][2500/3907] lr: 1.4047e-05 eta: 0:56:01 time: 0.6315 data_time: 0.0017 memory: 44120 loss: 0.3733 +2023/06/04 08:05:38 - mmengine - INFO - Epoch(train) [9][2600/3907] lr: 1.3899e-05 eta: 0:54:58 time: 0.6326 data_time: 0.0019 memory: 44120 loss: 0.3626 +2023/06/04 08:06:41 - mmengine - INFO - Epoch(train) [9][2700/3907] lr: 1.3753e-05 eta: 0:53:55 time: 0.6308 data_time: 0.0015 memory: 44120 loss: 0.3699 +2023/06/04 08:07:09 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230604_015520 +2023/06/04 08:07:44 - mmengine - INFO - Epoch(train) [9][2800/3907] lr: 1.3609e-05 eta: 0:52:51 time: 0.6303 data_time: 0.0018 memory: 44120 loss: 0.3656 +2023/06/04 08:08:47 - mmengine - INFO - Epoch(train) [9][2900/3907] lr: 1.3469e-05 eta: 0:51:48 time: 0.6306 data_time: 0.0017 memory: 44120 loss: 0.3944 +2023/06/04 08:09:51 - mmengine - INFO - Epoch(train) [9][3000/3907] lr: 1.3331e-05 eta: 0:50:45 time: 0.6309 data_time: 0.0016 memory: 44120 loss: 0.3797 +2023/06/04 08:10:54 - mmengine - INFO - Epoch(train) [9][3100/3907] lr: 1.3196e-05 eta: 0:49:41 time: 0.6310 data_time: 0.0016 memory: 44120 loss: 0.3287 +2023/06/04 08:11:57 - mmengine - INFO - Epoch(train) [9][3200/3907] lr: 1.3063e-05 eta: 0:48:38 time: 0.6328 data_time: 0.0018 memory: 44120 loss: 0.3583 +2023/06/04 08:13:00 - mmengine - INFO - Epoch(train) [9][3300/3907] lr: 1.2933e-05 eta: 0:47:35 time: 0.6322 data_time: 0.0017 memory: 44120 loss: 0.3624 +2023/06/04 08:14:03 - mmengine - INFO - Epoch(train) [9][3400/3907] lr: 1.2806e-05 eta: 0:46:32 time: 0.6311 data_time: 0.0017 memory: 44120 loss: 0.3890 +2023/06/04 08:15:07 - mmengine - INFO - Epoch(train) [9][3500/3907] lr: 1.2682e-05 eta: 0:45:28 time: 0.6311 data_time: 0.0016 memory: 44120 loss: 0.3814 +2023/06/04 08:16:10 - mmengine - INFO - Epoch(train) [9][3600/3907] lr: 1.2560e-05 eta: 0:44:25 time: 0.6319 data_time: 0.0017 memory: 44120 loss: 0.3663 +2023/06/04 08:17:13 - mmengine - INFO - Epoch(train) [9][3700/3907] lr: 1.2441e-05 eta: 0:43:22 time: 0.6319 data_time: 0.0023 memory: 44120 loss: 0.3615 +2023/06/04 08:17:41 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230604_015520 +2023/06/04 08:18:16 - mmengine - INFO - Epoch(train) [9][3800/3907] lr: 1.2325e-05 eta: 0:42:19 time: 0.6342 data_time: 0.0016 memory: 44120 loss: 0.3735 +2023/06/04 08:19:19 - mmengine - INFO - Epoch(train) [9][3900/3907] lr: 1.2211e-05 eta: 0:41:15 time: 0.6308 data_time: 0.0012 memory: 44120 loss: 0.4132 +2023/06/04 08:19:23 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230604_015520 +2023/06/04 08:19:23 - mmengine - INFO - Saving checkpoint at 9 epochs +2023/06/04 08:20:57 - mmengine - INFO - Epoch(val) [9][57/57] accuracy/top1: 75.6376 single-label/precision_classwise: [69.57523345947266, 97.84231567382812] single-label/recall_classwise: [99.16040802001953, 46.75193786621094] single-label/f1-score_classwise: [81.774169921875, 63.271087646484375] data_time: 0.0317 time: 1.2791 +2023/06/04 08:22:04 - mmengine - INFO - Epoch(train) [10][ 100/3907] lr: 1.2093e-05 eta: 0:40:08 time: 0.6304 data_time: 0.0017 memory: 44120 loss: 0.3572 +2023/06/04 08:23:07 - mmengine - INFO - Epoch(train) [10][ 200/3907] lr: 1.1985e-05 eta: 0:39:05 time: 0.6307 data_time: 0.0016 memory: 44120 loss: 0.3765 +2023/06/04 08:24:10 - mmengine - INFO - Epoch(train) [10][ 300/3907] lr: 1.1881e-05 eta: 0:38:01 time: 0.6311 data_time: 0.0017 memory: 44120 loss: 0.3687 +2023/06/04 08:25:14 - mmengine - INFO - Epoch(train) [10][ 400/3907] lr: 1.1778e-05 eta: 0:36:58 time: 0.6301 data_time: 0.0016 memory: 44120 loss: 0.3558 +2023/06/04 08:26:17 - mmengine - INFO - Epoch(train) [10][ 500/3907] lr: 1.1679e-05 eta: 0:35:55 time: 0.6314 data_time: 0.0015 memory: 44120 loss: 0.3804 +2023/06/04 08:27:20 - mmengine - INFO - Epoch(train) [10][ 600/3907] lr: 1.1583e-05 eta: 0:34:52 time: 0.6309 data_time: 0.0017 memory: 44120 loss: 0.3518 +2023/06/04 08:28:23 - mmengine - INFO - Epoch(train) [10][ 700/3907] lr: 1.1489e-05 eta: 0:33:48 time: 0.6328 data_time: 0.0021 memory: 44120 loss: 0.3624 +2023/06/04 08:29:26 - mmengine - INFO - Epoch(train) [10][ 800/3907] lr: 1.1398e-05 eta: 0:32:45 time: 0.6307 data_time: 0.0016 memory: 44120 loss: 0.3868 +2023/06/04 08:29:49 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230604_015520 +2023/06/04 08:30:29 - mmengine - INFO - Epoch(train) [10][ 900/3907] lr: 1.1310e-05 eta: 0:31:42 time: 0.6328 data_time: 0.0017 memory: 44120 loss: 0.3741 +2023/06/04 08:31:32 - mmengine - INFO - Epoch(train) [10][1000/3907] lr: 1.1225e-05 eta: 0:30:39 time: 0.6305 data_time: 0.0016 memory: 44120 loss: 0.3466 +2023/06/04 08:32:36 - mmengine - INFO - Epoch(train) [10][1100/3907] lr: 1.1142e-05 eta: 0:29:35 time: 0.6306 data_time: 0.0018 memory: 44120 loss: 0.3633 +2023/06/04 08:33:39 - mmengine - INFO - Epoch(train) [10][1200/3907] lr: 1.1063e-05 eta: 0:28:32 time: 0.6305 data_time: 0.0016 memory: 44120 loss: 0.3635 +2023/06/04 08:34:42 - mmengine - INFO - Epoch(train) [10][1300/3907] lr: 1.0986e-05 eta: 0:27:29 time: 0.6311 data_time: 0.0017 memory: 44120 loss: 0.3559 +2023/06/04 08:35:45 - mmengine - INFO - Epoch(train) [10][1400/3907] lr: 1.0912e-05 eta: 0:26:25 time: 0.6342 data_time: 0.0016 memory: 44120 loss: 0.3782 +2023/06/04 08:36:48 - mmengine - INFO - Epoch(train) [10][1500/3907] lr: 1.0841e-05 eta: 0:25:22 time: 0.6316 data_time: 0.0022 memory: 44120 loss: 0.3685 +2023/06/04 08:37:52 - mmengine - INFO - Epoch(train) [10][1600/3907] lr: 1.0773e-05 eta: 0:24:19 time: 0.6310 data_time: 0.0016 memory: 44120 loss: 0.3689 +2023/06/04 08:38:55 - mmengine - INFO - Epoch(train) [10][1700/3907] lr: 1.0707e-05 eta: 0:23:16 time: 0.6304 data_time: 0.0015 memory: 44120 loss: 0.3840 +2023/06/04 08:39:58 - mmengine - INFO - Epoch(train) [10][1800/3907] lr: 1.0645e-05 eta: 0:22:12 time: 0.6308 data_time: 0.0016 memory: 44120 loss: 0.3538 +2023/06/04 08:40:21 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230604_015520 +2023/06/04 08:41:01 - mmengine - INFO - Epoch(train) [10][1900/3907] lr: 1.0585e-05 eta: 0:21:09 time: 0.6312 data_time: 0.0016 memory: 44120 loss: 0.3882 +2023/06/04 08:42:04 - mmengine - INFO - Epoch(train) [10][2000/3907] lr: 1.0529e-05 eta: 0:20:06 time: 0.6356 data_time: 0.0018 memory: 44120 loss: 0.3650 +2023/06/04 08:43:07 - mmengine - INFO - Epoch(train) [10][2100/3907] lr: 1.0475e-05 eta: 0:19:03 time: 0.6305 data_time: 0.0017 memory: 44120 loss: 0.3672 +2023/06/04 08:44:11 - mmengine - INFO - Epoch(train) [10][2200/3907] lr: 1.0424e-05 eta: 0:17:59 time: 0.6327 data_time: 0.0015 memory: 44120 loss: 0.3779 +2023/06/04 08:45:14 - mmengine - INFO - Epoch(train) [10][2300/3907] lr: 1.0376e-05 eta: 0:16:56 time: 0.6307 data_time: 0.0016 memory: 44120 loss: 0.3571 +2023/06/04 08:46:17 - mmengine - INFO - Epoch(train) [10][2400/3907] lr: 1.0330e-05 eta: 0:15:53 time: 0.6325 data_time: 0.0018 memory: 44120 loss: 0.3923 +2023/06/04 08:47:20 - mmengine - INFO - Epoch(train) [10][2500/3907] lr: 1.0288e-05 eta: 0:14:50 time: 0.6336 data_time: 0.0018 memory: 44120 loss: 0.3453 +2023/06/04 08:48:23 - mmengine - INFO - Epoch(train) [10][2600/3907] lr: 1.0249e-05 eta: 0:13:46 time: 0.6309 data_time: 0.0016 memory: 44120 loss: 0.3701 +2023/06/04 08:49:27 - mmengine - INFO - Epoch(train) [10][2700/3907] lr: 1.0212e-05 eta: 0:12:43 time: 0.6314 data_time: 0.0018 memory: 44120 loss: 0.3581 +2023/06/04 08:50:30 - mmengine - INFO - Epoch(train) [10][2800/3907] lr: 1.0178e-05 eta: 0:11:40 time: 0.6306 data_time: 0.0016 memory: 44120 loss: 0.3813 +2023/06/04 08:50:53 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230604_015520 +2023/06/04 08:51:33 - mmengine - INFO - Epoch(train) [10][2900/3907] lr: 1.0148e-05 eta: 0:10:37 time: 0.6367 data_time: 0.0017 memory: 44120 loss: 0.3749 +2023/06/04 08:52:36 - mmengine - INFO - Epoch(train) [10][3000/3907] lr: 1.0120e-05 eta: 0:09:33 time: 0.6316 data_time: 0.0017 memory: 44120 loss: 0.3817 +2023/06/04 08:53:39 - mmengine - INFO - Epoch(train) [10][3100/3907] lr: 1.0095e-05 eta: 0:08:30 time: 0.6315 data_time: 0.0017 memory: 44120 loss: 0.3391 +2023/06/04 08:54:43 - mmengine - INFO - Epoch(train) [10][3200/3907] lr: 1.0073e-05 eta: 0:07:27 time: 0.6409 data_time: 0.0017 memory: 44120 loss: 0.3735 +2023/06/04 08:55:46 - mmengine - INFO - Epoch(train) [10][3300/3907] lr: 1.0054e-05 eta: 0:06:23 time: 0.6356 data_time: 0.0024 memory: 44120 loss: 0.3694 +2023/06/04 08:56:49 - mmengine - INFO - Epoch(train) [10][3400/3907] lr: 1.0038e-05 eta: 0:05:20 time: 0.6434 data_time: 0.0017 memory: 44120 loss: 0.3701 +2023/06/04 08:57:53 - mmengine - INFO - Epoch(train) [10][3500/3907] lr: 1.0024e-05 eta: 0:04:17 time: 0.6300 data_time: 0.0016 memory: 44120 loss: 0.3660 +2023/06/04 08:58:56 - mmengine - INFO - Epoch(train) [10][3600/3907] lr: 1.0014e-05 eta: 0:03:14 time: 0.6342 data_time: 0.0029 memory: 44120 loss: 0.3813 +2023/06/04 08:59:59 - mmengine - INFO - Epoch(train) [10][3700/3907] lr: 1.0006e-05 eta: 0:02:10 time: 0.6326 data_time: 0.0022 memory: 44120 loss: 0.3606 +2023/06/04 09:01:02 - mmengine - INFO - Epoch(train) [10][3800/3907] lr: 1.0002e-05 eta: 0:01:07 time: 0.6331 data_time: 0.0016 memory: 44120 loss: 0.3612 +2023/06/04 09:01:26 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230604_015520 +2023/06/04 09:02:05 - mmengine - INFO - Epoch(train) [10][3900/3907] lr: 1.0000e-05 eta: 0:00:04 time: 0.6311 data_time: 0.0012 memory: 44120 loss: 0.3667 +2023/06/04 09:02:09 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230604_015520 +2023/06/04 09:02:09 - mmengine - INFO - Saving checkpoint at 10 epochs +2023/06/04 09:03:43 - mmengine - INFO - Epoch(val) [10][57/57] accuracy/top1: 75.4254 single-label/precision_classwise: [69.43756103515625, 97.37055969238281] single-label/recall_classwise: [98.97734069824219, 46.503875732421875] single-label/f1-score_classwise: [81.61682891845703, 62.945281982421875] data_time: 0.0323 time: 1.2791 diff --git a/clip_large_pretrain_4x256_sdv2_lr1e-4/20230604_015520/vis_data/20230604_015520.json b/clip_large_pretrain_4x256_sdv2_lr1e-4/20230604_015520/vis_data/20230604_015520.json new file mode 100644 index 0000000000000000000000000000000000000000..5593ec92a6acc2947d636d4ce7feebe09bef86c9 --- /dev/null +++ b/clip_large_pretrain_4x256_sdv2_lr1e-4/20230604_015520/vis_data/20230604_015520.json @@ -0,0 +1,400 @@ +{"lr": 9.999857418382631e-05, "data_time": 0.0015535831451416015, "loss": 0.6511398315429687, "time": 0.6302948713302612, "epoch": 1, "memory": 44120, "step": 100} +{"lr": 9.999423907359506e-05, "data_time": 0.0014174461364746093, "loss": 0.5779680252075196, "time": 0.6299609184265137, "epoch": 1, "memory": 44120, "step": 200} +{"lr": 9.998699480412073e-05, "data_time": 0.0014919996261596679, "loss": 0.5505671501159668, "time": 0.6316408395767212, "epoch": 1, "memory": 44120, "step": 300} +{"lr": 9.997684184379085e-05, "data_time": 0.002289700508117676, "loss": 0.5343621253967286, "time": 0.6332749843597412, "epoch": 1, "memory": 44120, "step": 400} +{"lr": 9.996378084905784e-05, "data_time": 0.0016236305236816406, "loss": 0.5073817253112793, "time": 0.6302190542221069, "epoch": 1, "memory": 44120, "step": 500} +{"lr": 9.994781266439667e-05, "data_time": 0.001563096046447754, "loss": 0.4989572048187256, "time": 0.6301367044448852, "epoch": 1, "memory": 44120, "step": 600} +{"lr": 9.992893832225061e-05, "data_time": 0.0016741275787353516, "loss": 0.5028618812561035, "time": 0.6291846990585327, "epoch": 1, "memory": 44120, "step": 700} +{"lr": 9.990715904296405e-05, "data_time": 0.0017473220825195313, "loss": 0.46791136264801025, "time": 0.6299782514572143, "epoch": 1, "memory": 44120, "step": 800} +{"lr": 9.988247623470371e-05, "data_time": 0.0016218662261962891, "loss": 0.4867336988449097, "time": 0.6294743776321411, "epoch": 1, "memory": 44120, "step": 900} +{"lr": 9.985489149336749e-05, "data_time": 0.001532769203186035, "loss": 0.4593863010406494, "time": 0.6297062397003174, "epoch": 1, "memory": 44120, "step": 1000} +{"lr": 9.982440660248187e-05, "data_time": 0.0014293432235717774, "loss": 0.4503697633743286, "time": 0.6294331789016724, "epoch": 1, "memory": 44120, "step": 1100} +{"lr": 9.979102353308574e-05, "data_time": 0.0015051126480102538, "loss": 0.4512750506401062, "time": 0.6301638841629028, "epoch": 1, "memory": 44120, "step": 1200} +{"lr": 9.975474444360358e-05, "data_time": 0.0015797138214111329, "loss": 0.45578144788742064, "time": 0.6316172122955322, "epoch": 1, "memory": 44120, "step": 1300} +{"lr": 9.97155716797056e-05, "data_time": 0.001532125473022461, "loss": 0.4727639019489288, "time": 0.6292188167572021, "epoch": 1, "memory": 44120, "step": 1400} +{"lr": 9.967350777415632e-05, "data_time": 0.0014520645141601562, "loss": 0.44409904479980467, "time": 0.6302995443344116, "epoch": 1, "memory": 44120, "step": 1500} +{"lr": 9.962855544665059e-05, "data_time": 0.0015540361404418946, "loss": 0.4235361576080322, "time": 0.6305093050003052, "epoch": 1, "memory": 44120, "step": 1600} +{"lr": 9.958071760363773e-05, "data_time": 0.0019036293029785155, "loss": 0.4788900136947632, "time": 0.6309087753295899, "epoch": 1, "memory": 44120, "step": 1700} +{"lr": 9.952999733813382e-05, "data_time": 0.0021437406539916992, "loss": 0.4394698917865753, "time": 0.6307754278182983, "epoch": 1, "memory": 44120, "step": 1800} +{"lr": 9.947639792952148e-05, "data_time": 0.0018276691436767579, "loss": 0.4205831170082092, "time": 0.6321241617202759, "epoch": 1, "memory": 44120, "step": 1900} +{"lr": 9.941992284333813e-05, "data_time": 0.001586604118347168, "loss": 0.4306331992149353, "time": 0.6313234567642212, "epoch": 1, "memory": 44120, "step": 2000} +{"lr": 9.93605757310516e-05, "data_time": 0.002110934257507324, "loss": 0.42725415527820587, "time": 0.6311674833297729, "epoch": 1, "memory": 44120, "step": 2100} +{"lr": 9.929836042982421e-05, "data_time": 0.0016530275344848633, "loss": 0.413659405708313, "time": 0.6323507070541382, "epoch": 1, "memory": 44120, "step": 2200} +{"lr": 9.923328096226467e-05, "data_time": 0.0017000436782836914, "loss": 0.41933178901672363, "time": 0.6326930522918701, "epoch": 1, "memory": 44120, "step": 2300} +{"lr": 9.916534153616819e-05, "data_time": 0.0015790700912475587, "loss": 0.4383719086647034, "time": 0.6312283754348755, "epoch": 1, "memory": 44120, "step": 2400} +{"lr": 9.909454654424387e-05, "data_time": 0.0014163970947265625, "loss": 0.40896245241165163, "time": 0.6320044279098511, "epoch": 1, "memory": 44120, "step": 2500} +{"lr": 9.902090056383109e-05, "data_time": 0.0016506671905517577, "loss": 0.4292390912771225, "time": 0.6302929401397706, "epoch": 1, "memory": 44120, "step": 2600} +{"lr": 9.894440835660365e-05, "data_time": 0.0015710830688476563, "loss": 0.41410685181617735, "time": 0.6296397686004639, "epoch": 1, "memory": 44120, "step": 2700} +{"lr": 9.886507486826162e-05, "data_time": 0.0016206979751586914, "loss": 0.3892812252044678, "time": 0.6307731866836548, "epoch": 1, "memory": 44120, "step": 2800} +{"lr": 9.878290522821154e-05, "data_time": 0.0014225006103515624, "loss": 0.39317906498908994, "time": 0.6291845798492431, "epoch": 1, "memory": 44120, "step": 2900} +{"lr": 9.86979047492352e-05, "data_time": 0.0021434307098388674, "loss": 0.3948826640844345, "time": 0.6313328742980957, "epoch": 1, "memory": 44120, "step": 3000} +{"lr": 9.861007892714578e-05, "data_time": 0.0016360759735107421, "loss": 0.36135985851287844, "time": 0.6303182125091553, "epoch": 1, "memory": 44120, "step": 3100} +{"lr": 9.851943344043239e-05, "data_time": 0.001565408706665039, "loss": 0.41074050664901735, "time": 0.6315772056579589, "epoch": 1, "memory": 44120, "step": 3200} +{"lr": 9.842597414989319e-05, "data_time": 0.0015559673309326171, "loss": 0.39188137650489807, "time": 0.6324409484863281, "epoch": 1, "memory": 44120, "step": 3300} +{"lr": 9.832970709825637e-05, "data_time": 0.0021782636642456053, "loss": 0.39438756108283995, "time": 0.6318658828735352, "epoch": 1, "memory": 44120, "step": 3400} +{"lr": 9.823063850978968e-05, "data_time": 0.0017688512802124024, "loss": 0.4241188198328018, "time": 0.6310411214828491, "epoch": 1, "memory": 44120, "step": 3500} +{"lr": 9.81287747898975e-05, "data_time": 0.0016953229904174804, "loss": 0.4126111209392548, "time": 0.6322051525115967, "epoch": 1, "memory": 44120, "step": 3600} +{"lr": 9.802412252470715e-05, "data_time": 0.0017744541168212891, "loss": 0.38610868752002714, "time": 0.6300936222076416, "epoch": 1, "memory": 44120, "step": 3700} +{"lr": 9.791668848064258e-05, "data_time": 0.001561450958251953, "loss": 0.4105738490819931, "time": 0.6288309812545776, "epoch": 1, "memory": 44120, "step": 3800} +{"lr": 9.780647960398751e-05, "data_time": 0.0016667604446411132, "loss": 0.37856527268886564, "time": 0.6306086301803588, "epoch": 1, "memory": 44120, "step": 3900} +{"accuracy/top1": 60.20319366455078, "data_time": 0.04584982520655582, "time": 1.3493186967414723, "step": 1} +{"lr": 9.7685491196646e-05, "data_time": 0.0023908138275146483, "loss": 0.4061006963253021, "time": 0.6311803579330444, "epoch": 2, "memory": 44120, "step": 4007} +{"lr": 9.756956126085676e-05, "data_time": 0.0014954090118408203, "loss": 0.403551721572876, "time": 0.6301365613937377, "epoch": 2, "memory": 44120, "step": 4107} +{"lr": 9.745087893642482e-05, "data_time": 0.0013546466827392579, "loss": 0.3965855985879898, "time": 0.6305136442184448, "epoch": 2, "memory": 44120, "step": 4207} +{"lr": 9.73294518969056e-05, "data_time": 0.0014646053314208984, "loss": 0.4431161880493164, "time": 0.6293402433395385, "epoch": 2, "memory": 44120, "step": 4307} +{"lr": 9.72052879933172e-05, "data_time": 0.001753544807434082, "loss": 0.37462318539619444, "time": 0.6295682430267334, "epoch": 2, "memory": 44120, "step": 4407} +{"lr": 9.707839525363358e-05, "data_time": 0.0014199256896972657, "loss": 0.4080145835876465, "time": 0.6294006824493408, "epoch": 2, "memory": 44120, "step": 4507} +{"lr": 9.69487818822646e-05, "data_time": 0.001447153091430664, "loss": 0.39367065727710726, "time": 0.6290924787521363, "epoch": 2, "memory": 44120, "step": 4607} +{"lr": 9.681645625952616e-05, "data_time": 0.001467132568359375, "loss": 0.36269129514694215, "time": 0.6295343637466431, "epoch": 2, "memory": 44120, "step": 4707} +{"lr": 9.668142694109826e-05, "data_time": 0.001475954055786133, "loss": 0.374642014503479, "time": 0.6383844614028931, "epoch": 2, "memory": 44120, "step": 4807} +{"lr": 9.654370265747162e-05, "data_time": 0.001619863510131836, "loss": 0.38116596043109896, "time": 0.630130934715271, "epoch": 2, "memory": 44120, "step": 4907} +{"lr": 9.640329231338345e-05, "data_time": 0.002016925811767578, "loss": 0.4128776788711548, "time": 0.6310742855072021, "epoch": 2, "memory": 44120, "step": 5007} +{"lr": 9.626020498724147e-05, "data_time": 0.001699209213256836, "loss": 0.40102483332157135, "time": 0.6309804677963257, "epoch": 2, "memory": 44120, "step": 5107} +{"lr": 9.611444993053717e-05, "data_time": 0.0015905141830444337, "loss": 0.37765371799468994, "time": 0.6297220945358276, "epoch": 2, "memory": 44120, "step": 5207} +{"lr": 9.596603656724746e-05, "data_time": 0.0015085220336914062, "loss": 0.38804788887500763, "time": 0.6301882982254028, "epoch": 2, "memory": 44120, "step": 5307} +{"lr": 9.58149744932254e-05, "data_time": 0.0013970613479614257, "loss": 0.3733738332986832, "time": 0.6317455768585205, "epoch": 2, "memory": 44120, "step": 5407} +{"lr": 9.566127347557972e-05, "data_time": 0.0015625476837158204, "loss": 0.3854053646326065, "time": 0.6299889326095581, "epoch": 2, "memory": 44120, "step": 5507} +{"lr": 9.550494345204367e-05, "data_time": 0.0015119791030883789, "loss": 0.38054428398609164, "time": 0.6295350313186645, "epoch": 2, "memory": 44120, "step": 5607} +{"lr": 9.534599453033183e-05, "data_time": 0.0015683650970458984, "loss": 0.38338888585567477, "time": 0.6296643257141114, "epoch": 2, "memory": 44120, "step": 5707} +{"lr": 9.518443698748712e-05, "data_time": 0.001562023162841797, "loss": 0.36804080903530123, "time": 0.6304186344146728, "epoch": 2, "memory": 44120, "step": 5807} +{"lr": 9.50202812692162e-05, "data_time": 0.001341700553894043, "loss": 0.3973271459341049, "time": 0.6301617860794068, "epoch": 2, "memory": 44120, "step": 5907} +{"lr": 9.485353798921386e-05, "data_time": 0.0014605522155761719, "loss": 0.37133278846740725, "time": 0.6311134576797486, "epoch": 2, "memory": 44120, "step": 6007} +{"lr": 9.468421792847712e-05, "data_time": 0.0013740062713623047, "loss": 0.37527437806129454, "time": 0.630155062675476, "epoch": 2, "memory": 44120, "step": 6107} +{"lr": 9.451233203460794e-05, "data_time": 0.0013575315475463866, "loss": 0.3760660618543625, "time": 0.6299977779388428, "epoch": 2, "memory": 44120, "step": 6207} +{"lr": 9.433789142110532e-05, "data_time": 0.0017714977264404296, "loss": 0.3865980327129364, "time": 0.6299216508865356, "epoch": 2, "memory": 44120, "step": 6307} +{"lr": 9.416090736664709e-05, "data_time": 0.0016645431518554688, "loss": 0.3935562491416931, "time": 0.6303906202316284, "epoch": 2, "memory": 44120, "step": 6407} +{"lr": 9.398139131436032e-05, "data_time": 0.0015691995620727539, "loss": 0.4170177221298218, "time": 0.6298570394515991, "epoch": 2, "memory": 44120, "step": 6507} +{"lr": 9.379935487108167e-05, "data_time": 0.001615595817565918, "loss": 0.38354268074035647, "time": 0.630503249168396, "epoch": 2, "memory": 44120, "step": 6607} +{"lr": 9.361480980660697e-05, "data_time": 0.0015149831771850586, "loss": 0.3863678127527237, "time": 0.6301994800567627, "epoch": 2, "memory": 44120, "step": 6707} +{"lr": 9.342776805292993e-05, "data_time": 0.001593160629272461, "loss": 0.36150650680065155, "time": 0.6305224657058716, "epoch": 2, "memory": 44120, "step": 6807} +{"lr": 9.323824170347083e-05, "data_time": 0.0014492273330688477, "loss": 0.3628203600645065, "time": 0.6301002740859986, "epoch": 2, "memory": 44120, "step": 6907} +{"lr": 9.304624301229471e-05, "data_time": 0.0016742706298828124, "loss": 0.3921812415122986, "time": 0.6335572242736817, "epoch": 2, "memory": 44120, "step": 7007} +{"lr": 9.285178439331897e-05, "data_time": 0.0017034769058227538, "loss": 0.38591637313365934, "time": 0.630746603012085, "epoch": 2, "memory": 44120, "step": 7107} +{"lr": 9.265487841951087e-05, "data_time": 0.0014682531356811524, "loss": 0.39570842385292054, "time": 0.6305039882659912, "epoch": 2, "memory": 44120, "step": 7207} +{"lr": 9.245553782207421e-05, "data_time": 0.0016492128372192383, "loss": 0.387408047914505, "time": 0.6317680358886719, "epoch": 2, "memory": 44120, "step": 7307} +{"lr": 9.225377548962671e-05, "data_time": 0.0016398191452026366, "loss": 0.3795217454433441, "time": 0.6297847986221313, "epoch": 2, "memory": 44120, "step": 7407} +{"lr": 9.204960446736601e-05, "data_time": 0.001511359214782715, "loss": 0.38568209707736967, "time": 0.6308626890182495, "epoch": 2, "memory": 44120, "step": 7507} +{"lr": 9.184303795622715e-05, "data_time": 0.0014635086059570312, "loss": 0.39033634662628175, "time": 0.6300491571426392, "epoch": 2, "memory": 44120, "step": 7607} +{"lr": 9.163408931202823e-05, "data_time": 0.001522397994995117, "loss": 0.39881399273872375, "time": 0.6334480285644531, "epoch": 2, "memory": 44120, "step": 7707} +{"lr": 9.142277204460729e-05, "data_time": 0.0013886690139770508, "loss": 0.3804485470056534, "time": 0.6289843797683716, "epoch": 2, "memory": 44120, "step": 7807} +{"accuracy/top1": 66.76525115966797, "data_time": 0.03664940801160089, "time": 1.2851452293067143, "step": 2} +{"lr": 9.119405492383288e-05, "data_time": 0.0013090848922729492, "loss": 0.36767544150352477, "time": 0.6290694952011109, "epoch": 3, "memory": 44120, "step": 7914} +{"lr": 9.097787819233205e-05, "data_time": 0.001606321334838867, "loss": 0.38307862877845766, "time": 0.6293684482574463, "epoch": 3, "memory": 44120, "step": 8014} +{"lr": 9.075937526576627e-05, "data_time": 0.0019600391387939453, "loss": 0.37978521585464475, "time": 0.6312113046646118, "epoch": 3, "memory": 44120, "step": 8114} +{"lr": 9.053856027171755e-05, "data_time": 0.001861906051635742, "loss": 0.37750745117664336, "time": 0.6299858570098877, "epoch": 3, "memory": 44120, "step": 8214} +{"lr": 9.031544748725785e-05, "data_time": 0.001420307159423828, "loss": 0.37020819187164306, "time": 0.6302985429763794, "epoch": 3, "memory": 44120, "step": 8314} +{"lr": 9.009005133802542e-05, "data_time": 0.0016502618789672851, "loss": 0.38140026926994325, "time": 0.6302116394042969, "epoch": 3, "memory": 44120, "step": 8414} +{"lr": 8.986238639729242e-05, "data_time": 0.001695728302001953, "loss": 0.39678852558135985, "time": 0.6301708221435547, "epoch": 3, "memory": 44120, "step": 8514} +{"lr": 8.963246738502254e-05, "data_time": 0.001857590675354004, "loss": 0.4026592016220093, "time": 0.6328332662582398, "epoch": 3, "memory": 44120, "step": 8614} +{"lr": 8.940030916691928e-05, "data_time": 0.0017349720001220703, "loss": 0.3703895002603531, "time": 0.6312051057815552, "epoch": 3, "memory": 44120, "step": 8714} +{"lr": 8.91659267534649e-05, "data_time": 0.0019045829772949218, "loss": 0.4036286801099777, "time": 0.6318345308303833, "epoch": 3, "memory": 44120, "step": 8814} +{"lr": 8.892933529894978e-05, "data_time": 0.002414250373840332, "loss": 0.36947562992572786, "time": 0.6323338270187377, "epoch": 3, "memory": 44120, "step": 8914} +{"lr": 8.869055010049283e-05, "data_time": 0.0022028446197509765, "loss": 0.35781918466091156, "time": 0.6331470251083374, "epoch": 3, "memory": 44120, "step": 9014} +{"lr": 8.844958659705188e-05, "data_time": 0.0017703771591186523, "loss": 0.3781828165054321, "time": 0.630960202217102, "epoch": 3, "memory": 44120, "step": 9114} +{"lr": 8.82064603684259e-05, "data_time": 0.0017389774322509766, "loss": 0.3636253088712692, "time": 0.6301772832870484, "epoch": 3, "memory": 44120, "step": 9214} +{"lr": 8.796118713424776e-05, "data_time": 0.0014866113662719727, "loss": 0.3884113222360611, "time": 0.6302128791809082, "epoch": 3, "memory": 44120, "step": 9314} +{"lr": 8.77137827529676e-05, "data_time": 0.001845383644104004, "loss": 0.3719728320837021, "time": 0.6318903923034668, "epoch": 3, "memory": 44120, "step": 9414} +{"lr": 8.746426322082745e-05, "data_time": 0.0015121936798095704, "loss": 0.37726588547229767, "time": 0.6308449745178223, "epoch": 3, "memory": 44120, "step": 9514} +{"lr": 8.721264467082732e-05, "data_time": 0.0016206979751586914, "loss": 0.36793930530548097, "time": 0.6302549123764039, "epoch": 3, "memory": 44120, "step": 9614} +{"lr": 8.695894337168168e-05, "data_time": 0.002411556243896484, "loss": 0.3705622971057892, "time": 0.6314141273498535, "epoch": 3, "memory": 44120, "step": 9714} +{"lr": 8.67031757267677e-05, "data_time": 0.0017479181289672852, "loss": 0.3867658466100693, "time": 0.6301435470581055, "epoch": 3, "memory": 44120, "step": 9814} +{"lr": 8.64453582730651e-05, "data_time": 0.0016212224960327148, "loss": 0.373067370057106, "time": 0.6304151535034179, "epoch": 3, "memory": 44120, "step": 9914} +{"lr": 8.61855076800861e-05, "data_time": 0.0017477989196777344, "loss": 0.3726970463991165, "time": 0.6302961349487305, "epoch": 3, "memory": 44120, "step": 10014} +{"lr": 8.592364074879843e-05, "data_time": 0.001483011245727539, "loss": 0.3646294206380844, "time": 0.6306333065032959, "epoch": 3, "memory": 44120, "step": 10114} +{"lr": 8.565977441053854e-05, "data_time": 0.0015407085418701171, "loss": 0.3594058692455292, "time": 0.6309885025024414, "epoch": 3, "memory": 44120, "step": 10214} +{"lr": 8.539392572591724e-05, "data_time": 0.0018303394317626953, "loss": 0.34311483502388, "time": 0.6308549880981446, "epoch": 3, "memory": 44120, "step": 10314} +{"lr": 8.512611188371633e-05, "data_time": 0.0018095254898071289, "loss": 0.36856233775615693, "time": 0.6306205511093139, "epoch": 3, "memory": 44120, "step": 10414} +{"lr": 8.485635019977725e-05, "data_time": 0.0015825748443603516, "loss": 0.3896019458770752, "time": 0.630800437927246, "epoch": 3, "memory": 44120, "step": 10514} +{"lr": 8.45846581158818e-05, "data_time": 0.0016936540603637695, "loss": 0.37825354039669035, "time": 0.630472207069397, "epoch": 3, "memory": 44120, "step": 10614} +{"lr": 8.431105319862416e-05, "data_time": 0.0017005681991577148, "loss": 0.3651675909757614, "time": 0.6305620670318604, "epoch": 3, "memory": 44120, "step": 10714} +{"lr": 8.403555313827516e-05, "data_time": 0.002501082420349121, "loss": 0.3763491839170456, "time": 0.6325437784194946, "epoch": 3, "memory": 44120, "step": 10814} +{"lr": 8.375817574763844e-05, "data_time": 0.0019482135772705077, "loss": 0.3438781201839447, "time": 0.6315265893936157, "epoch": 3, "memory": 44120, "step": 10914} +{"lr": 8.347893896089871e-05, "data_time": 0.0018458366394042969, "loss": 0.3857393771409988, "time": 0.6343234062194825, "epoch": 3, "memory": 44120, "step": 11014} +{"lr": 8.31978608324625e-05, "data_time": 0.0017656326293945313, "loss": 0.3934557676315308, "time": 0.6319288015365601, "epoch": 3, "memory": 44120, "step": 11114} +{"lr": 8.291495953579052e-05, "data_time": 0.0029499292373657226, "loss": 0.3598293989896774, "time": 0.6363593578338623, "epoch": 3, "memory": 44120, "step": 11214} +{"lr": 8.263025336222267e-05, "data_time": 0.0017736434936523437, "loss": 0.36102803945541384, "time": 0.6317667007446289, "epoch": 3, "memory": 44120, "step": 11314} +{"lr": 8.234376071979551e-05, "data_time": 0.0015689373016357423, "loss": 0.3519564002752304, "time": 0.6314729452133179, "epoch": 3, "memory": 44120, "step": 11414} +{"lr": 8.205550013205187e-05, "data_time": 0.001592254638671875, "loss": 0.4015227615833282, "time": 0.63122398853302, "epoch": 3, "memory": 44120, "step": 11514} +{"lr": 8.17654902368435e-05, "data_time": 0.0017137765884399415, "loss": 0.3792177468538284, "time": 0.6312206506729126, "epoch": 3, "memory": 44120, "step": 11614} +{"lr": 8.147374978512582e-05, "data_time": 0.0012741327285766602, "loss": 0.39173182249069216, "time": 0.630467700958252, "epoch": 3, "memory": 44120, "step": 11714} +{"accuracy/top1": 70.75606536865234, "data_time": 0.03451025897058947, "time": 1.2825194761670868, "step": 3} +{"lr": 8.115969237626578e-05, "data_time": 0.00140688419342041, "loss": 0.35677762031555177, "time": 0.6293562650680542, "epoch": 4, "memory": 44120, "step": 11821} +{"lr": 8.08644297337245e-05, "data_time": 0.0014370918273925782, "loss": 0.3732084989547729, "time": 0.6307722568511963, "epoch": 4, "memory": 44120, "step": 11921} +{"lr": 8.056749479387651e-05, "data_time": 0.0015021562576293945, "loss": 0.3886274993419647, "time": 0.6319432258605957, "epoch": 4, "memory": 44120, "step": 12021} +{"lr": 8.026890675542457e-05, "data_time": 0.001486063003540039, "loss": 0.37129161059856414, "time": 0.6311835527420044, "epoch": 4, "memory": 44120, "step": 12121} +{"lr": 7.996868492395457e-05, "data_time": 0.0015145301818847655, "loss": 0.3772895783185959, "time": 0.629550838470459, "epoch": 4, "memory": 44120, "step": 12221} +{"lr": 7.966684871068762e-05, "data_time": 0.001385664939880371, "loss": 0.3722930520772934, "time": 0.6305479049682617, "epoch": 4, "memory": 44120, "step": 12321} +{"lr": 7.936341763122449e-05, "data_time": 0.0015053272247314454, "loss": 0.34028970301151273, "time": 0.6314324617385865, "epoch": 4, "memory": 44120, "step": 12421} +{"lr": 7.905841130428405e-05, "data_time": 0.0014670610427856446, "loss": 0.3500517576932907, "time": 0.6303344964981079, "epoch": 4, "memory": 44120, "step": 12521} +{"lr": 7.875184945043487e-05, "data_time": 0.0018103837966918946, "loss": 0.38720719516277313, "time": 0.6401841402053833, "epoch": 4, "memory": 44120, "step": 12621} +{"lr": 7.844375189082012e-05, "data_time": 0.0018702507019042968, "loss": 0.3647387117147446, "time": 0.6307103872299195, "epoch": 4, "memory": 44120, "step": 12721} +{"lr": 7.813413854587572e-05, "data_time": 0.0023189544677734374, "loss": 0.35620381236076354, "time": 0.6318166494369507, "epoch": 4, "memory": 44120, "step": 12821} +{"lr": 7.782302943404291e-05, "data_time": 0.002088785171508789, "loss": 0.3544735312461853, "time": 0.6318197965621948, "epoch": 4, "memory": 44120, "step": 12921} +{"lr": 7.751044467047337e-05, "data_time": 0.002436661720275879, "loss": 0.39118861556053164, "time": 0.6354005336761475, "epoch": 4, "memory": 44120, "step": 13021} +{"lr": 7.719640446572907e-05, "data_time": 0.002176833152770996, "loss": 0.354831263422966, "time": 0.6314675092697144, "epoch": 4, "memory": 44120, "step": 13121} +{"lr": 7.688092912447536e-05, "data_time": 0.0024971961975097656, "loss": 0.33954288959503176, "time": 0.6356311082839966, "epoch": 4, "memory": 44120, "step": 13221} +{"lr": 7.656403904416794e-05, "data_time": 0.0023842573165893553, "loss": 0.358999764919281, "time": 0.6345445871353149, "epoch": 4, "memory": 44120, "step": 13321} +{"lr": 7.624575471373434e-05, "data_time": 0.0026422739028930664, "loss": 0.3761797368526459, "time": 0.6329836130142212, "epoch": 4, "memory": 44120, "step": 13421} +{"lr": 7.592609671224926e-05, "data_time": 0.002575516700744629, "loss": 0.36268415451049807, "time": 0.6348317861557007, "epoch": 4, "memory": 44120, "step": 13521} +{"lr": 7.560508570760375e-05, "data_time": 0.002576899528503418, "loss": 0.3569221556186676, "time": 0.6331686496734619, "epoch": 4, "memory": 44120, "step": 13621} +{"lr": 7.528274245516895e-05, "data_time": 0.0024100780487060548, "loss": 0.3351865142583847, "time": 0.6345189094543457, "epoch": 4, "memory": 44120, "step": 13721} +{"lr": 7.495908779645405e-05, "data_time": 0.0021024465560913084, "loss": 0.3555357575416565, "time": 0.6310706853866577, "epoch": 4, "memory": 44120, "step": 13821} +{"lr": 7.463414265775906e-05, "data_time": 0.0021494150161743162, "loss": 0.3550543010234833, "time": 0.6348968267440795, "epoch": 4, "memory": 44120, "step": 13921} +{"lr": 7.430792804882147e-05, "data_time": 0.0022434711456298826, "loss": 0.37112358808517454, "time": 0.6317383289337158, "epoch": 4, "memory": 44120, "step": 14021} +{"lr": 7.398046506145803e-05, "data_time": 0.0022365570068359373, "loss": 0.3770568609237671, "time": 0.6320046424865723, "epoch": 4, "memory": 44120, "step": 14121} +{"lr": 7.365177486820114e-05, "data_time": 0.0020328521728515624, "loss": 0.3532114714384079, "time": 0.6325595617294312, "epoch": 4, "memory": 44120, "step": 14221} +{"lr": 7.332187872092939e-05, "data_time": 0.001931619644165039, "loss": 0.4026179492473602, "time": 0.6314249515533448, "epoch": 4, "memory": 44120, "step": 14321} +{"lr": 7.299079794949401e-05, "data_time": 0.002328300476074219, "loss": 0.3726774573326111, "time": 0.6362221717834473, "epoch": 4, "memory": 44120, "step": 14421} +{"lr": 7.265855396033955e-05, "data_time": 0.0020601511001586913, "loss": 0.3620989680290222, "time": 0.6323452949523926, "epoch": 4, "memory": 44120, "step": 14521} +{"lr": 7.232516823512013e-05, "data_time": 0.0018093109130859375, "loss": 0.3517916172742844, "time": 0.6318056344985962, "epoch": 4, "memory": 44120, "step": 14621} +{"lr": 7.199066232930995e-05, "data_time": 0.0018860101699829102, "loss": 0.3624063342809677, "time": 0.6316128730773926, "epoch": 4, "memory": 44120, "step": 14721} +{"lr": 7.165505787080997e-05, "data_time": 0.002146315574645996, "loss": 0.3662576645612717, "time": 0.6343168020248413, "epoch": 4, "memory": 44120, "step": 14821} +{"lr": 7.13183765585495e-05, "data_time": 0.002161383628845215, "loss": 0.3525313138961792, "time": 0.6326789855957031, "epoch": 4, "memory": 44120, "step": 14921} +{"lr": 7.098064016108317e-05, "data_time": 0.0018190383911132813, "loss": 0.37121322751045227, "time": 0.6303757429122925, "epoch": 4, "memory": 44120, "step": 15021} +{"lr": 7.064187051518379e-05, "data_time": 0.0016277551651000977, "loss": 0.36289057433605193, "time": 0.6301361322402954, "epoch": 4, "memory": 44120, "step": 15121} +{"lr": 7.030208952442949e-05, "data_time": 0.0014097213745117188, "loss": 0.40250545740127563, "time": 0.6292129755020142, "epoch": 4, "memory": 44120, "step": 15221} +{"lr": 6.996131915778851e-05, "data_time": 0.0014852523803710938, "loss": 0.35194175839424136, "time": 0.6314897298812866, "epoch": 4, "memory": 44120, "step": 15321} +{"lr": 6.961958144819853e-05, "data_time": 0.001454782485961914, "loss": 0.367633318901062, "time": 0.6316639423370362, "epoch": 4, "memory": 44120, "step": 15421} +{"lr": 6.927689849114193e-05, "data_time": 0.0015785932540893556, "loss": 0.38853678703308103, "time": 0.6305542230606079, "epoch": 4, "memory": 44120, "step": 15521} +{"lr": 6.893329244321722e-05, "data_time": 0.0012257814407348633, "loss": 0.36316122114658356, "time": 0.6299135684967041, "epoch": 4, "memory": 44120, "step": 15621} +{"accuracy/top1": 72.92369842529297, "data_time": 0.033927378983333194, "time": 1.2814092964961612, "step": 4} +{"lr": 6.856463687361792e-05, "data_time": 0.0014965295791625976, "loss": 0.36035783737897875, "time": 0.6297960758209229, "epoch": 5, "memory": 44120, "step": 15728} +{"lr": 6.821919068470432e-05, "data_time": 0.0014540910720825194, "loss": 0.37258667349815366, "time": 0.630597186088562, "epoch": 5, "memory": 44120, "step": 15828} +{"lr": 6.787288979235407e-05, "data_time": 0.0021564245223999025, "loss": 0.3697516113519669, "time": 0.6314488410949707, "epoch": 5, "memory": 44120, "step": 15928} +{"lr": 6.752575658708793e-05, "data_time": 0.0016209602355957032, "loss": 0.3605677604675293, "time": 0.6309216022491455, "epoch": 5, "memory": 44120, "step": 16028} +{"lr": 6.71778135132408e-05, "data_time": 0.0013944864273071288, "loss": 0.37399953305721284, "time": 0.6309323072433471, "epoch": 5, "memory": 44120, "step": 16128} +{"lr": 6.682908306751074e-05, "data_time": 0.0015766382217407226, "loss": 0.37510315477848055, "time": 0.6325395107269287, "epoch": 5, "memory": 44120, "step": 16228} +{"lr": 6.64795877975043e-05, "data_time": 0.001371455192565918, "loss": 0.37172494232654574, "time": 0.6315000295639038, "epoch": 5, "memory": 44120, "step": 16328} +{"lr": 6.61293503002787e-05, "data_time": 0.0014748811721801759, "loss": 0.35580343902111056, "time": 0.6327751398086547, "epoch": 5, "memory": 44120, "step": 16428} +{"lr": 6.577839322088084e-05, "data_time": 0.0018471479415893555, "loss": 0.363631945848465, "time": 0.6312255620956421, "epoch": 5, "memory": 44120, "step": 16528} +{"lr": 6.542673925088298e-05, "data_time": 0.001959562301635742, "loss": 0.3688123047351837, "time": 0.6371283531188965, "epoch": 5, "memory": 44120, "step": 16628} +{"lr": 6.507441112691584e-05, "data_time": 0.0015973329544067382, "loss": 0.3715033084154129, "time": 0.6311631917953491, "epoch": 5, "memory": 44120, "step": 16728} +{"lr": 6.47214316291983e-05, "data_time": 0.0017039060592651367, "loss": 0.34398736655712125, "time": 0.6322572946548461, "epoch": 5, "memory": 44120, "step": 16828} +{"lr": 6.436782358006481e-05, "data_time": 0.0015607118606567384, "loss": 0.3612892985343933, "time": 0.6309884309768676, "epoch": 5, "memory": 44120, "step": 16928} +{"lr": 6.40136098424896e-05, "data_time": 0.001902627944946289, "loss": 0.33692377507686616, "time": 0.6304582118988037, "epoch": 5, "memory": 44120, "step": 17028} +{"lr": 6.365881331860828e-05, "data_time": 0.0015949487686157226, "loss": 0.35666256546974184, "time": 0.6300513744354248, "epoch": 5, "memory": 44120, "step": 17128} +{"lr": 6.330345694823739e-05, "data_time": 0.001678800582885742, "loss": 0.38750146329402924, "time": 0.6304916381835938, "epoch": 5, "memory": 44120, "step": 17228} +{"lr": 6.29475637073908e-05, "data_time": 0.0015712738037109374, "loss": 0.3811786353588104, "time": 0.6323616743087769, "epoch": 5, "memory": 44120, "step": 17328} +{"lr": 6.259115660679482e-05, "data_time": 0.0015985727310180663, "loss": 0.3553478717803955, "time": 0.6314690351486206, "epoch": 5, "memory": 44120, "step": 17428} +{"lr": 6.223425869039963e-05, "data_time": 0.0021724462509155273, "loss": 0.39600721895694735, "time": 0.6326065778732299, "epoch": 5, "memory": 44120, "step": 17528} +{"lr": 6.187689303389006e-05, "data_time": 0.001455831527709961, "loss": 0.36005370914936063, "time": 0.6312793254852295, "epoch": 5, "memory": 44120, "step": 17628} +{"lr": 6.151908274319292e-05, "data_time": 0.0015760183334350586, "loss": 0.34296722114086153, "time": 0.6310264587402343, "epoch": 5, "memory": 44120, "step": 17728} +{"lr": 6.116085095298373e-05, "data_time": 0.001760578155517578, "loss": 0.36004377603530885, "time": 0.6306248188018799, "epoch": 5, "memory": 44120, "step": 17828} +{"lr": 6.080222082519007e-05, "data_time": 0.0017028093338012696, "loss": 0.35072051435709, "time": 0.630894660949707, "epoch": 5, "memory": 44120, "step": 17928} +{"lr": 6.0443215547495147e-05, "data_time": 0.0016824960708618163, "loss": 0.38175966739654543, "time": 0.6305303573608398, "epoch": 5, "memory": 44120, "step": 18028} +{"lr": 6.0083858331837714e-05, "data_time": 0.0015622854232788086, "loss": 0.3692367136478424, "time": 0.6308236360549927, "epoch": 5, "memory": 44120, "step": 18128} +{"lr": 5.97241724129116e-05, "data_time": 0.001533842086791992, "loss": 0.3736819863319397, "time": 0.6304581880569458, "epoch": 5, "memory": 44120, "step": 18228} +{"lr": 5.936418104666328e-05, "data_time": 0.0014395475387573241, "loss": 0.37733722627162936, "time": 0.6304697036743164, "epoch": 5, "memory": 44120, "step": 18328} +{"lr": 5.9003907508788295e-05, "data_time": 0.0017235279083251953, "loss": 0.3453709274530411, "time": 0.6312472105026246, "epoch": 5, "memory": 44120, "step": 18428} +{"lr": 5.864337509322633e-05, "data_time": 0.0019505977630615234, "loss": 0.3516500622034073, "time": 0.6328014850616455, "epoch": 5, "memory": 44120, "step": 18528} +{"lr": 5.8282607110655194e-05, "data_time": 0.0017161607742309571, "loss": 0.3615013867616653, "time": 0.631293249130249, "epoch": 5, "memory": 44120, "step": 18628} +{"lr": 5.7921626886983564e-05, "data_time": 0.0028637170791625975, "loss": 0.37235530018806456, "time": 0.6343429327011109, "epoch": 5, "memory": 44120, "step": 18728} +{"lr": 5.756045776184287e-05, "data_time": 0.0018888473510742187, "loss": 0.3680415689945221, "time": 0.6325015544891357, "epoch": 5, "memory": 44120, "step": 18828} +{"lr": 5.719912308707808e-05, "data_time": 0.0016191720962524414, "loss": 0.3743208646774292, "time": 0.6393663644790649, "epoch": 5, "memory": 44120, "step": 18928} +{"lr": 5.683764622523808e-05, "data_time": 0.001971936225891113, "loss": 0.36365104019641875, "time": 0.6324900150299072, "epoch": 5, "memory": 44120, "step": 19028} +{"lr": 5.647605054806502e-05, "data_time": 0.0016748666763305663, "loss": 0.3520662605762482, "time": 0.6315549850463867, "epoch": 5, "memory": 44120, "step": 19128} +{"lr": 5.6114359434983206e-05, "data_time": 0.0016469955444335938, "loss": 0.3705656111240387, "time": 0.6313248634338379, "epoch": 5, "memory": 44120, "step": 19228} +{"lr": 5.575259627158745e-05, "data_time": 0.0015885353088378907, "loss": 0.3683649569749832, "time": 0.631161093711853, "epoch": 5, "memory": 44120, "step": 19328} +{"lr": 5.539078444813113e-05, "data_time": 0.0016903400421142579, "loss": 0.39164201021194456, "time": 0.6324179649353028, "epoch": 5, "memory": 44120, "step": 19428} +{"lr": 5.5028947358013735e-05, "data_time": 0.001396942138671875, "loss": 0.3690459430217743, "time": 0.6315234661102295, "epoch": 5, "memory": 44120, "step": 19528} +{"accuracy/top1": 74.18670654296875, "data_time": 0.033750776586861445, "time": 1.2835038777055412, "step": 5} +{"lr": 5.464178020340108e-05, "data_time": 0.0014005661010742187, "loss": 0.3426802784204483, "time": 0.6311736106872559, "epoch": 6, "memory": 44120, "step": 19635} +{"lr": 5.427996514795046e-05, "data_time": 0.00141599178314209, "loss": 0.36596367657184603, "time": 0.6307308197021484, "epoch": 6, "memory": 44120, "step": 19735} +{"lr": 5.3918196647260834e-05, "data_time": 0.0016403675079345703, "loss": 0.3611610263586044, "time": 0.6345808744430542, "epoch": 6, "memory": 44120, "step": 19835} +{"lr": 5.355649809193057e-05, "data_time": 0.001447582244873047, "loss": 0.3365130931138992, "time": 0.6309124231338501, "epoch": 6, "memory": 44120, "step": 19935} +{"lr": 5.3194892868035765e-05, "data_time": 0.0014417409896850587, "loss": 0.37650632560253144, "time": 0.6317238807678223, "epoch": 6, "memory": 44120, "step": 20035} +{"lr": 5.283340435561795e-05, "data_time": 0.0016066789627075194, "loss": 0.3631355732679367, "time": 0.6313675880432129, "epoch": 6, "memory": 44120, "step": 20135} +{"lr": 5.2472055927172495e-05, "data_time": 0.0017436981201171876, "loss": 0.3929067015647888, "time": 0.631146502494812, "epoch": 6, "memory": 44120, "step": 20235} +{"lr": 5.2110870946137605e-05, "data_time": 0.0015924215316772462, "loss": 0.3521174430847168, "time": 0.6329382658004761, "epoch": 6, "memory": 44120, "step": 20335} +{"lr": 5.174987276538342e-05, "data_time": 0.0017699241638183595, "loss": 0.36878088116645813, "time": 0.6312318325042725, "epoch": 6, "memory": 44120, "step": 20435} +{"lr": 5.1389084725702425e-05, "data_time": 0.0018259286880493164, "loss": 0.3505380988121033, "time": 0.6309262752532959, "epoch": 6, "memory": 44120, "step": 20535} +{"lr": 5.1028530154300045e-05, "data_time": 0.00170440673828125, "loss": 0.3884235739707947, "time": 0.6307322978973389, "epoch": 6, "memory": 44120, "step": 20635} +{"lr": 5.066823236328663e-05, "data_time": 0.0017857313156127929, "loss": 0.35251970291137696, "time": 0.6326623439788819, "epoch": 6, "memory": 44120, "step": 20735} +{"lr": 5.030821464817006e-05, "data_time": 0.0016338586807250976, "loss": 0.376922196149826, "time": 0.63508620262146, "epoch": 6, "memory": 44120, "step": 20835} +{"lr": 4.99485002863494e-05, "data_time": 0.0015831708908081055, "loss": 0.3826912134885788, "time": 0.6432285785675049, "epoch": 6, "memory": 44120, "step": 20935} +{"lr": 4.95891125356102e-05, "data_time": 0.0017174482345581055, "loss": 0.34901047348976133, "time": 0.6320256233215332, "epoch": 6, "memory": 44120, "step": 21035} +{"lr": 4.923007463262053e-05, "data_time": 0.0016624212265014648, "loss": 0.3568054050207138, "time": 0.6320607900619507, "epoch": 6, "memory": 44120, "step": 21135} +{"lr": 4.887140979142852e-05, "data_time": 0.0018236637115478516, "loss": 0.36538914740085604, "time": 0.6317174673080445, "epoch": 6, "memory": 44120, "step": 21235} +{"lr": 4.851314120196161e-05, "data_time": 0.0016582965850830077, "loss": 0.3845347762107849, "time": 0.6306475400924683, "epoch": 6, "memory": 44120, "step": 21335} +{"lr": 4.815529202852693e-05, "data_time": 0.0015470743179321288, "loss": 0.36047182977199554, "time": 0.6317041158676148, "epoch": 6, "memory": 44120, "step": 21435} +{"lr": 4.779788540831406e-05, "data_time": 0.001522231101989746, "loss": 0.3793373882770538, "time": 0.6315422296524048, "epoch": 6, "memory": 44120, "step": 21535} +{"lr": 4.744094444989843e-05, "data_time": 0.0014984607696533203, "loss": 0.38657639026641843, "time": 0.6309186458587647, "epoch": 6, "memory": 44120, "step": 21635} +{"lr": 4.708449223174764e-05, "data_time": 0.001526641845703125, "loss": 0.36951213479042055, "time": 0.6311927080154419, "epoch": 6, "memory": 44120, "step": 21735} +{"lr": 4.672855180072916e-05, "data_time": 0.0015697002410888672, "loss": 0.3582223176956177, "time": 0.6308010578155517, "epoch": 6, "memory": 44120, "step": 21835} +{"lr": 4.6373146170620256e-05, "data_time": 0.0018631219863891602, "loss": 0.3661968469619751, "time": 0.6326027870178222, "epoch": 6, "memory": 44120, "step": 21935} +{"lr": 4.6018298320619915e-05, "data_time": 0.001543712615966797, "loss": 0.3824609011411667, "time": 0.6340478420257568, "epoch": 6, "memory": 44120, "step": 22035} +{"lr": 4.566403119386318e-05, "data_time": 0.0019037723541259766, "loss": 0.39292274713516234, "time": 0.6347447872161865, "epoch": 6, "memory": 44120, "step": 22135} +{"lr": 4.531036769593769e-05, "data_time": 0.0017054080963134766, "loss": 0.3460966289043427, "time": 0.6304093837738037, "epoch": 6, "memory": 44120, "step": 22235} +{"lr": 4.4957330693402685e-05, "data_time": 0.001736593246459961, "loss": 0.3412874609231949, "time": 0.6325237035751343, "epoch": 6, "memory": 44120, "step": 22335} +{"lr": 4.460494301231059e-05, "data_time": 0.0014856100082397462, "loss": 0.3635822504758835, "time": 0.6321806669235229, "epoch": 6, "memory": 44120, "step": 22435} +{"lr": 4.425322743673109e-05, "data_time": 0.0016674757003784179, "loss": 0.38342941701412203, "time": 0.630539870262146, "epoch": 6, "memory": 44120, "step": 22535} +{"lr": 4.390220670727812e-05, "data_time": 0.0016361236572265624, "loss": 0.36300550401210785, "time": 0.6302071332931518, "epoch": 6, "memory": 44120, "step": 22635} +{"lr": 4.3551903519639325e-05, "data_time": 0.0014899253845214843, "loss": 0.3936154991388321, "time": 0.6306586027145386, "epoch": 6, "memory": 44120, "step": 22735} +{"lr": 4.3202340523108894e-05, "data_time": 0.0014934301376342773, "loss": 0.35447017550468446, "time": 0.6331157207489013, "epoch": 6, "memory": 44120, "step": 22835} +{"lr": 4.285354031912302e-05, "data_time": 0.0015354156494140625, "loss": 0.375302791595459, "time": 0.6329831123352051, "epoch": 6, "memory": 44120, "step": 22935} +{"lr": 4.2505525459798544e-05, "data_time": 0.0016644954681396484, "loss": 0.3989559143781662, "time": 0.6303085088729858, "epoch": 6, "memory": 44120, "step": 23035} +{"lr": 4.2158318446474836e-05, "data_time": 0.0019134521484375, "loss": 0.3641472548246384, "time": 0.6320447444915771, "epoch": 6, "memory": 44120, "step": 23135} +{"lr": 4.181194172825905e-05, "data_time": 0.0017750024795532226, "loss": 0.35941425561904905, "time": 0.6305461406707764, "epoch": 6, "memory": 44120, "step": 23235} +{"lr": 4.146641770057448e-05, "data_time": 0.001417231559753418, "loss": 0.3691575676202774, "time": 0.6326620817184448, "epoch": 6, "memory": 44120, "step": 23335} +{"lr": 4.112176870371267e-05, "data_time": 0.001457047462463379, "loss": 0.3825172454118729, "time": 0.6309915781021118, "epoch": 6, "memory": 44120, "step": 23435} +{"accuracy/top1": 74.76079559326172, "data_time": 0.03613086404471562, "time": 1.2846617986415994, "step": 6} +{"lr": 4.0753988582732786e-05, "data_time": 0.0017429351806640624, "loss": 0.3907084405422211, "time": 0.6305223941802979, "epoch": 7, "memory": 44120, "step": 23542} +{"lr": 4.0411221639286815e-05, "data_time": 0.0019685506820678713, "loss": 0.3553955316543579, "time": 0.6326249837875366, "epoch": 7, "memory": 44120, "step": 23642} +{"lr": 4.006939795169518e-05, "data_time": 0.002959156036376953, "loss": 0.38052708506584165, "time": 0.6349667549133301, "epoch": 7, "memory": 44120, "step": 23742} +{"lr": 3.972853962099913e-05, "data_time": 0.003167867660522461, "loss": 0.39260380566120145, "time": 0.6354622840881348, "epoch": 7, "memory": 44120, "step": 23842} +{"lr": 3.93886686858237e-05, "data_time": 0.0027597427368164064, "loss": 0.3540808379650116, "time": 0.6363731861114502, "epoch": 7, "memory": 44120, "step": 23942} +{"lr": 3.904980712095259e-05, "data_time": 0.0018626689910888673, "loss": 0.366429203748703, "time": 0.6304872035980225, "epoch": 7, "memory": 44120, "step": 24042} +{"lr": 3.871197683590748e-05, "data_time": 0.0014447450637817382, "loss": 0.35341583490371703, "time": 0.6317300081253052, "epoch": 7, "memory": 44120, "step": 24142} +{"lr": 3.837519967353125e-05, "data_time": 0.0017804622650146485, "loss": 0.3656893610954285, "time": 0.6323260545730591, "epoch": 7, "memory": 44120, "step": 24242} +{"lr": 3.8039497408575855e-05, "data_time": 0.0018790006637573243, "loss": 0.3789824068546295, "time": 0.6302371263504029, "epoch": 7, "memory": 44120, "step": 24342} +{"lr": 3.770489174629432e-05, "data_time": 0.0015023231506347656, "loss": 0.3829865217208862, "time": 0.6311488628387452, "epoch": 7, "memory": 44120, "step": 24442} +{"lr": 3.737140432103754e-05, "data_time": 0.001765608787536621, "loss": 0.3756262481212616, "time": 0.6312336444854736, "epoch": 7, "memory": 44120, "step": 24542} +{"lr": 3.703905669485545e-05, "data_time": 0.00162050724029541, "loss": 0.3716110110282898, "time": 0.6304387331008912, "epoch": 7, "memory": 44120, "step": 24642} +{"lr": 3.6707870356102706e-05, "data_time": 0.001439046859741211, "loss": 0.35791911482810973, "time": 0.630659818649292, "epoch": 7, "memory": 44120, "step": 24742} +{"lr": 3.6377866718049484e-05, "data_time": 0.0018537044525146484, "loss": 0.4020693838596344, "time": 0.6335522890090942, "epoch": 7, "memory": 44120, "step": 24842} +{"lr": 3.6049067117496965e-05, "data_time": 0.0021165132522583006, "loss": 0.38328524231910704, "time": 0.6413475513458252, "epoch": 7, "memory": 44120, "step": 24942} +{"lr": 3.5721492813397755e-05, "data_time": 0.0021758317947387696, "loss": 0.3705703943967819, "time": 0.6323210716247558, "epoch": 7, "memory": 44120, "step": 25042} +{"lr": 3.539516498548151e-05, "data_time": 0.0016237020492553711, "loss": 0.353135484457016, "time": 0.6329949378967286, "epoch": 7, "memory": 44120, "step": 25142} +{"lr": 3.507010473288518e-05, "data_time": 0.001582765579223633, "loss": 0.3761199861764908, "time": 0.6315385103225708, "epoch": 7, "memory": 44120, "step": 25242} +{"lr": 3.474633307278909e-05, "data_time": 0.0016865968704223634, "loss": 0.38100827038288115, "time": 0.6311712265014648, "epoch": 7, "memory": 44120, "step": 25342} +{"lr": 3.4423870939058177e-05, "data_time": 0.001741790771484375, "loss": 0.4142740875482559, "time": 0.631229019165039, "epoch": 7, "memory": 44120, "step": 25442} +{"lr": 3.410273918088805e-05, "data_time": 0.001669621467590332, "loss": 0.3732988625764847, "time": 0.6314670324325562, "epoch": 7, "memory": 44120, "step": 25542} +{"lr": 3.37829585614574e-05, "data_time": 0.0022763490676879885, "loss": 0.3778089553117752, "time": 0.6351943254470825, "epoch": 7, "memory": 44120, "step": 25642} +{"lr": 3.3464549756585216e-05, "data_time": 0.0018778562545776368, "loss": 0.3983409285545349, "time": 0.6337371349334717, "epoch": 7, "memory": 44120, "step": 25742} +{"lr": 3.314753335339416e-05, "data_time": 0.0019174337387084961, "loss": 0.3413211554288864, "time": 0.6314706802368164, "epoch": 7, "memory": 44120, "step": 25842} +{"lr": 3.2831929848979394e-05, "data_time": 0.001600360870361328, "loss": 0.3760252594947815, "time": 0.6325624704360961, "epoch": 7, "memory": 44120, "step": 25942} +{"lr": 3.251775964908332e-05, "data_time": 0.0016645431518554688, "loss": 0.3815203607082367, "time": 0.6311161756515503, "epoch": 7, "memory": 44120, "step": 26042} +{"lr": 3.22050430667762e-05, "data_time": 0.0017194509506225585, "loss": 0.3696031779050827, "time": 0.6311853408813477, "epoch": 7, "memory": 44120, "step": 26142} +{"lr": 3.189380032114285e-05, "data_time": 0.0017520666122436523, "loss": 0.3679883062839508, "time": 0.6324081182479858, "epoch": 7, "memory": 44120, "step": 26242} +{"lr": 3.1584051535975324e-05, "data_time": 0.0015215873718261719, "loss": 0.3825375556945801, "time": 0.6307426929473877, "epoch": 7, "memory": 44120, "step": 26342} +{"lr": 3.127581673847175e-05, "data_time": 0.001592254638671875, "loss": 0.36102774143218996, "time": 0.6308587789535522, "epoch": 7, "memory": 44120, "step": 26442} +{"lr": 3.0969115857941555e-05, "data_time": 0.0016212940216064453, "loss": 0.367472892999649, "time": 0.6305755615234375, "epoch": 7, "memory": 44120, "step": 26542} +{"lr": 3.0663968724516654e-05, "data_time": 0.001456308364868164, "loss": 0.3704697281122208, "time": 0.6312161445617676, "epoch": 7, "memory": 44120, "step": 26642} +{"lr": 3.0360395067869668e-05, "data_time": 0.0019632577896118164, "loss": 0.3739717960357666, "time": 0.6323014259338379, "epoch": 7, "memory": 44120, "step": 26742} +{"lr": 3.0058414515938063e-05, "data_time": 0.0016052961349487306, "loss": 0.3671251147985458, "time": 0.6343223094940186, "epoch": 7, "memory": 44120, "step": 26842} +{"lr": 2.975804659365501e-05, "data_time": 0.0019020795822143554, "loss": 0.37165725231170654, "time": 0.6316561698913574, "epoch": 7, "memory": 44120, "step": 26942} +{"lr": 2.945931072168714e-05, "data_time": 0.001604461669921875, "loss": 0.3721301585435867, "time": 0.6338523864746094, "epoch": 7, "memory": 44120, "step": 27042} +{"lr": 2.9162226215178855e-05, "data_time": 0.0016022920608520508, "loss": 0.3967426270246506, "time": 0.6302759647369385, "epoch": 7, "memory": 44120, "step": 27142} +{"lr": 2.8866812282503322e-05, "data_time": 0.0017762184143066406, "loss": 0.3843903779983521, "time": 0.632037878036499, "epoch": 7, "memory": 44120, "step": 27242} +{"lr": 2.8573088024020838e-05, "data_time": 0.0015083789825439454, "loss": 0.35235882699489596, "time": 0.6293889284133911, "epoch": 7, "memory": 44120, "step": 27342} +{"accuracy/top1": 75.0356674194336, "data_time": 0.0317530632019043, "time": 1.2813783226342037, "step": 7} +{"lr": 2.8260695817390092e-05, "data_time": 0.0018131256103515625, "loss": 0.3888059794902802, "time": 0.6305115222930908, "epoch": 8, "memory": 44120, "step": 27449} +{"lr": 2.7970529402570967e-05, "data_time": 0.0014161348342895507, "loss": 0.3569154739379883, "time": 0.6301471710205078, "epoch": 8, "memory": 44120, "step": 27549} +{"lr": 2.7682110612244214e-05, "data_time": 0.0015307426452636718, "loss": 0.3666468858718872, "time": 0.6308928728103638, "epoch": 8, "memory": 44120, "step": 27649} +{"lr": 2.7395458094490283e-05, "data_time": 0.001904463768005371, "loss": 0.3775034248828888, "time": 0.6302277326583863, "epoch": 8, "memory": 44120, "step": 27749} +{"lr": 2.7110590383189015e-05, "data_time": 0.0016535282135009765, "loss": 0.39016111493110656, "time": 0.6319559812545776, "epoch": 8, "memory": 44120, "step": 27849} +{"lr": 2.6827525896821397e-05, "data_time": 0.0020310163497924806, "loss": 0.35462755560874937, "time": 0.6325171232223511, "epoch": 8, "memory": 44120, "step": 27949} +{"lr": 2.6546282937278552e-05, "data_time": 0.001477217674255371, "loss": 0.36985248029232026, "time": 0.6309507846832275, "epoch": 8, "memory": 44120, "step": 28049} +{"lr": 2.6266879688678606e-05, "data_time": 0.002318263053894043, "loss": 0.33672429621219635, "time": 0.633042287826538, "epoch": 8, "memory": 44120, "step": 28149} +{"lr": 2.5989334216190738e-05, "data_time": 0.001660633087158203, "loss": 0.3577429085969925, "time": 0.6312180280685424, "epoch": 8, "memory": 44120, "step": 28249} +{"lr": 2.571366446486732e-05, "data_time": 0.001709890365600586, "loss": 0.3706617772579193, "time": 0.6319145679473877, "epoch": 8, "memory": 44120, "step": 28349} +{"lr": 2.5439888258483665e-05, "data_time": 0.001877593994140625, "loss": 0.3651845157146454, "time": 0.6331820726394654, "epoch": 8, "memory": 44120, "step": 28449} +{"lr": 2.5168023298385384e-05, "data_time": 0.001558208465576172, "loss": 0.3963282972574234, "time": 0.6308655977249146, "epoch": 8, "memory": 44120, "step": 28549} +{"lr": 2.489808716234412e-05, "data_time": 0.0017090320587158203, "loss": 0.3455282747745514, "time": 0.6313030242919921, "epoch": 8, "memory": 44120, "step": 28649} +{"lr": 2.463009730342111e-05, "data_time": 0.00146026611328125, "loss": 0.35659017860889436, "time": 0.6301676034927368, "epoch": 8, "memory": 44120, "step": 28749} +{"lr": 2.436407104883849e-05, "data_time": 0.0017405271530151368, "loss": 0.3557533144950867, "time": 0.6315641403198242, "epoch": 8, "memory": 44120, "step": 28849} +{"lr": 2.4100025598859092e-05, "data_time": 0.001744532585144043, "loss": 0.4059371381998062, "time": 0.6453279256820679, "epoch": 8, "memory": 44120, "step": 28949} +{"lr": 2.3837978025674284e-05, "data_time": 0.002048492431640625, "loss": 0.3632131278514862, "time": 0.6317559003829956, "epoch": 8, "memory": 44120, "step": 29049} +{"lr": 2.357794527230027e-05, "data_time": 0.0016649484634399414, "loss": 0.3726804792881012, "time": 0.6326836824417115, "epoch": 8, "memory": 44120, "step": 29149} +{"lr": 2.33199441514825e-05, "data_time": 0.0016494512557983399, "loss": 0.38908624947071074, "time": 0.6314280271530152, "epoch": 8, "memory": 44120, "step": 29249} +{"lr": 2.306399134460859e-05, "data_time": 0.001507735252380371, "loss": 0.3706782728433609, "time": 0.6342707395553588, "epoch": 8, "memory": 44120, "step": 29349} +{"lr": 2.281010340062995e-05, "data_time": 0.0015233278274536133, "loss": 0.3851109206676483, "time": 0.631217622756958, "epoch": 8, "memory": 44120, "step": 29449} +{"lr": 2.2558296734991603e-05, "data_time": 0.0015987157821655273, "loss": 0.37804694175720216, "time": 0.6335761308670044, "epoch": 8, "memory": 44120, "step": 29549} +{"lr": 2.2308587628570915e-05, "data_time": 0.0020048856735229493, "loss": 0.3452103555202484, "time": 0.6310968637466431, "epoch": 8, "memory": 44120, "step": 29649} +{"lr": 2.2060992226624955e-05, "data_time": 0.0015798091888427734, "loss": 0.39766121804714205, "time": 0.630724573135376, "epoch": 8, "memory": 44120, "step": 29749} +{"lr": 2.181552653774648e-05, "data_time": 0.0015205144882202148, "loss": 0.3869244307279587, "time": 0.6324070930480957, "epoch": 8, "memory": 44120, "step": 29849} +{"lr": 2.1572206432829035e-05, "data_time": 0.0015801668167114257, "loss": 0.36472782492637634, "time": 0.6330604553222656, "epoch": 8, "memory": 44120, "step": 29949} +{"lr": 2.1331047644040754e-05, "data_time": 0.0014296770095825195, "loss": 0.3893174022436142, "time": 0.6306674242019653, "epoch": 8, "memory": 44120, "step": 30049} +{"lr": 2.109206576380705e-05, "data_time": 0.001616525650024414, "loss": 0.37757771015167235, "time": 0.6316470146179199, "epoch": 8, "memory": 44120, "step": 30149} +{"lr": 2.0855276243802687e-05, "data_time": 0.001432943344116211, "loss": 0.36857550144195556, "time": 0.6312754392623902, "epoch": 8, "memory": 44120, "step": 30249} +{"lr": 2.0620694393952487e-05, "data_time": 0.001424098014831543, "loss": 0.369435515999794, "time": 0.6353442668914795, "epoch": 8, "memory": 44120, "step": 30349} +{"lr": 2.038833538144169e-05, "data_time": 0.0013814926147460937, "loss": 0.40329310297966003, "time": 0.631174373626709, "epoch": 8, "memory": 44120, "step": 30449} +{"lr": 2.0158214229735154e-05, "data_time": 0.001552748680114746, "loss": 0.36263504028320315, "time": 0.6297932863235474, "epoch": 8, "memory": 44120, "step": 30549} +{"lr": 1.9930345817605977e-05, "data_time": 0.0015177011489868164, "loss": 0.3531788408756256, "time": 0.6324968576431275, "epoch": 8, "memory": 44120, "step": 30649} +{"lr": 1.970474487817358e-05, "data_time": 0.001619577407836914, "loss": 0.3608382433652878, "time": 0.6316331624984741, "epoch": 8, "memory": 44120, "step": 30749} +{"lr": 1.9481425997951092e-05, "data_time": 0.0014397621154785157, "loss": 0.37489425837993623, "time": 0.6312134981155395, "epoch": 8, "memory": 44120, "step": 30849} +{"lr": 1.9260403615902187e-05, "data_time": 0.0019301176071166992, "loss": 0.3705161720514297, "time": 0.6321103572845459, "epoch": 8, "memory": 44120, "step": 30949} +{"lr": 1.9041692022507647e-05, "data_time": 0.0018164873123168944, "loss": 0.36591207385063174, "time": 0.6339050769805908, "epoch": 8, "memory": 44120, "step": 31049} +{"lr": 1.8825305358841125e-05, "data_time": 0.001504969596862793, "loss": 0.3680920213460922, "time": 0.6305970668792724, "epoch": 8, "memory": 44120, "step": 31149} +{"lr": 1.861125761565518e-05, "data_time": 0.001303553581237793, "loss": 0.349801230430603, "time": 0.6308927059173584, "epoch": 8, "memory": 44120, "step": 31249} +{"accuracy/top1": 75.31748962402344, "data_time": 0.03823984491414037, "time": 1.2858252689756196, "step": 8} +{"lr": 1.8384832449122864e-05, "data_time": 0.001545119285583496, "loss": 0.36703485548496245, "time": 0.6299569606781006, "epoch": 9, "memory": 44120, "step": 31356} +{"lr": 1.817567007332114e-05, "data_time": 0.0015564203262329102, "loss": 0.3762494921684265, "time": 0.630024528503418, "epoch": 9, "memory": 44120, "step": 31456} +{"lr": 1.796888862098825e-05, "data_time": 0.0017134189605712891, "loss": 0.3761904388666153, "time": 0.6309181690216065, "epoch": 9, "memory": 44120, "step": 31556} +{"lr": 1.7764501461839678e-05, "data_time": 0.002408552169799805, "loss": 0.3647339463233948, "time": 0.63305504322052, "epoch": 9, "memory": 44120, "step": 31656} +{"lr": 1.7562521810784746e-05, "data_time": 0.0017658233642578124, "loss": 0.35223880410194397, "time": 0.6322914361953735, "epoch": 9, "memory": 44120, "step": 31756} +{"lr": 1.7362962727072444e-05, "data_time": 0.0019680261611938477, "loss": 0.35907942354679107, "time": 0.6366469383239746, "epoch": 9, "memory": 44120, "step": 31856} +{"lr": 1.716583711344682e-05, "data_time": 0.001625204086303711, "loss": 0.34919698536396027, "time": 0.6315120220184326, "epoch": 9, "memory": 44120, "step": 31956} +{"lr": 1.6971157715312954e-05, "data_time": 0.0019192457199096679, "loss": 0.3772891044616699, "time": 0.6313671827316284, "epoch": 9, "memory": 44120, "step": 32056} +{"lr": 1.677893711991274e-05, "data_time": 0.0015111446380615234, "loss": 0.3667439311742783, "time": 0.6309309244155884, "epoch": 9, "memory": 44120, "step": 32156} +{"lr": 1.6589187755511025e-05, "data_time": 0.001652359962463379, "loss": 0.376569926738739, "time": 0.6314516305923462, "epoch": 9, "memory": 44120, "step": 32256} +{"lr": 1.640192189059216e-05, "data_time": 0.0016833066940307618, "loss": 0.35952570140361784, "time": 0.6338250637054443, "epoch": 9, "memory": 44120, "step": 32356} +{"lr": 1.6217151633066716e-05, "data_time": 0.00162813663482666, "loss": 0.39810044765472413, "time": 0.6326664924621582, "epoch": 9, "memory": 44120, "step": 32456} +{"lr": 1.6034888929488564e-05, "data_time": 0.0018696784973144531, "loss": 0.35402290225028993, "time": 0.631584620475769, "epoch": 9, "memory": 44120, "step": 32556} +{"lr": 1.585514556428258e-05, "data_time": 0.001542377471923828, "loss": 0.3798711508512497, "time": 0.631550407409668, "epoch": 9, "memory": 44120, "step": 32656} +{"lr": 1.567793315898261e-05, "data_time": 0.0014659404754638673, "loss": 0.3942319631576538, "time": 0.6313107728958129, "epoch": 9, "memory": 44120, "step": 32756} +{"lr": 1.5503263171480048e-05, "data_time": 0.001752614974975586, "loss": 0.361508384346962, "time": 0.6313312292098999, "epoch": 9, "memory": 44120, "step": 32856} +{"lr": 1.5331146895283165e-05, "data_time": 0.001412200927734375, "loss": 0.38959822356700896, "time": 0.6342740535736084, "epoch": 9, "memory": 44120, "step": 32956} +{"lr": 1.5161595458786728e-05, "data_time": 0.0015943527221679687, "loss": 0.38137592375278473, "time": 0.6306544542312622, "epoch": 9, "memory": 44120, "step": 33056} +{"lr": 1.4994619824552563e-05, "data_time": 0.0015293359756469727, "loss": 0.39286218881607055, "time": 0.6311569929122924, "epoch": 9, "memory": 44120, "step": 33156} +{"lr": 1.4830230788600802e-05, "data_time": 0.0017734050750732421, "loss": 0.37159693241119385, "time": 0.6308529615402222, "epoch": 9, "memory": 44120, "step": 33256} +{"lr": 1.4668438979711779e-05, "data_time": 0.0017318487167358398, "loss": 0.3421124368906021, "time": 0.6308297634124755, "epoch": 9, "memory": 44120, "step": 33356} +{"lr": 1.4509254858738844e-05, "data_time": 0.0015690088272094726, "loss": 0.34994647204875945, "time": 0.6317381381988525, "epoch": 9, "memory": 44120, "step": 33456} +{"lr": 1.4352688717931966e-05, "data_time": 0.0016712427139282226, "loss": 0.3561157017946243, "time": 0.6310806512832642, "epoch": 9, "memory": 44120, "step": 33556} +{"lr": 1.4198750680272348e-05, "data_time": 0.001669001579284668, "loss": 0.4037459522485733, "time": 0.6342191457748413, "epoch": 9, "memory": 44120, "step": 33656} +{"lr": 1.4047450698817857e-05, "data_time": 0.0016637802124023437, "loss": 0.37333865761756896, "time": 0.6314516067504883, "epoch": 9, "memory": 44120, "step": 33756} +{"lr": 1.389879855605952e-05, "data_time": 0.0018546342849731444, "loss": 0.3626085966825485, "time": 0.6326472043991089, "epoch": 9, "memory": 44120, "step": 33856} +{"lr": 1.3752803863288976e-05, "data_time": 0.0015349149703979491, "loss": 0.3698987692594528, "time": 0.6308177471160888, "epoch": 9, "memory": 44120, "step": 33956} +{"lr": 1.3609476059977129e-05, "data_time": 0.0017613649368286132, "loss": 0.36564660370349883, "time": 0.6302955389022827, "epoch": 9, "memory": 44120, "step": 34056} +{"lr": 1.3468824413163754e-05, "data_time": 0.0016710281372070313, "loss": 0.3944234699010849, "time": 0.6306342840194702, "epoch": 9, "memory": 44120, "step": 34156} +{"lr": 1.3330858016858395e-05, "data_time": 0.0016077041625976562, "loss": 0.37970112562179564, "time": 0.6308547019958496, "epoch": 9, "memory": 44120, "step": 34256} +{"lr": 1.3195585791452309e-05, "data_time": 0.0015581846237182617, "loss": 0.32868294715881347, "time": 0.6309695243835449, "epoch": 9, "memory": 44120, "step": 34356} +{"lr": 1.3063016483141729e-05, "data_time": 0.0017665863037109376, "loss": 0.35830931067466737, "time": 0.6328487873077393, "epoch": 9, "memory": 44120, "step": 34456} +{"lr": 1.2933158663362388e-05, "data_time": 0.0016797780990600586, "loss": 0.3624483227729797, "time": 0.6322095394134521, "epoch": 9, "memory": 44120, "step": 34556} +{"lr": 1.2806020728235311e-05, "data_time": 0.001705312728881836, "loss": 0.3889764338731766, "time": 0.6311343431472778, "epoch": 9, "memory": 44120, "step": 34656} +{"lr": 1.2681610898023963e-05, "data_time": 0.0015658140182495117, "loss": 0.3814141720533371, "time": 0.6311052799224853, "epoch": 9, "memory": 44120, "step": 34756} +{"lr": 1.2559937216602731e-05, "data_time": 0.0016845226287841796, "loss": 0.36632585525512695, "time": 0.6318511247634888, "epoch": 9, "memory": 44120, "step": 34856} +{"lr": 1.2441007550936847e-05, "data_time": 0.002275562286376953, "loss": 0.3614902436733246, "time": 0.631909704208374, "epoch": 9, "memory": 44120, "step": 34956} +{"lr": 1.232482959057377e-05, "data_time": 0.0016462087631225586, "loss": 0.3735176056623459, "time": 0.6342405319213867, "epoch": 9, "memory": 44120, "step": 35056} +{"lr": 1.2211410847145969e-05, "data_time": 0.0012430667877197266, "loss": 0.4131546676158905, "time": 0.630808162689209, "epoch": 9, "memory": 44120, "step": 35156} +{"accuracy/top1": 75.6375961303711, "data_time": 0.03174326337617019, "time": 1.279104837055864, "step": 9} +{"lr": 1.209311679360931e-05, "data_time": 0.0017336606979370117, "loss": 0.35722833275794985, "time": 0.6304180383682251, "epoch": 10, "memory": 44120, "step": 35263} +{"lr": 1.1985432727488666e-05, "data_time": 0.0016126632690429688, "loss": 0.3765010148286819, "time": 0.6306754350662231, "epoch": 10, "memory": 44120, "step": 35363} +{"lr": 1.1880529822435289e-05, "data_time": 0.0016975641250610352, "loss": 0.36871753036975863, "time": 0.6311088800430298, "epoch": 10, "memory": 44120, "step": 35463} +{"lr": 1.1778414861078726e-05, "data_time": 0.0015508174896240235, "loss": 0.3558087289333344, "time": 0.6301191568374633, "epoch": 10, "memory": 44120, "step": 35563} +{"lr": 1.1679094445790538e-05, "data_time": 0.001487874984741211, "loss": 0.38038043677806854, "time": 0.6314319849014283, "epoch": 10, "memory": 44120, "step": 35663} +{"lr": 1.1582574998257377e-05, "data_time": 0.0016562461853027344, "loss": 0.35182705521583557, "time": 0.6309397697448731, "epoch": 10, "memory": 44120, "step": 35763} +{"lr": 1.1488862759065762e-05, "data_time": 0.00214688777923584, "loss": 0.3624135583639145, "time": 0.6328227281570434, "epoch": 10, "memory": 44120, "step": 35863} +{"lr": 1.1397963787298665e-05, "data_time": 0.0015983104705810547, "loss": 0.3868138015270233, "time": 0.6307113409042359, "epoch": 10, "memory": 44120, "step": 35963} +{"lr": 1.1309883960143664e-05, "data_time": 0.001660299301147461, "loss": 0.3741405069828033, "time": 0.63277268409729, "epoch": 10, "memory": 44120, "step": 36063} +{"lr": 1.122462897251302e-05, "data_time": 0.0015828371047973632, "loss": 0.34660682678222654, "time": 0.630481481552124, "epoch": 10, "memory": 44120, "step": 36163} +{"lr": 1.114220433667542e-05, "data_time": 0.0017837762832641601, "loss": 0.36329991519451144, "time": 0.6305551052093505, "epoch": 10, "memory": 44120, "step": 36263} +{"lr": 1.10626153818996e-05, "data_time": 0.0016129016876220703, "loss": 0.3634851574897766, "time": 0.6304593086242676, "epoch": 10, "memory": 44120, "step": 36363} +{"lr": 1.0985867254109761e-05, "data_time": 0.0016503572463989259, "loss": 0.35591005384922025, "time": 0.6310620307922363, "epoch": 10, "memory": 44120, "step": 36463} +{"lr": 1.0911964915552823e-05, "data_time": 0.0015620708465576172, "loss": 0.3781615048646927, "time": 0.6342061042785645, "epoch": 10, "memory": 44120, "step": 36563} +{"lr": 1.0840913144477643e-05, "data_time": 0.0022008419036865234, "loss": 0.36848844289779664, "time": 0.6316298246383667, "epoch": 10, "memory": 44120, "step": 36663} +{"lr": 1.0772716534826064e-05, "data_time": 0.0016360998153686524, "loss": 0.36894028782844546, "time": 0.6309942960739136, "epoch": 10, "memory": 44120, "step": 36763} +{"lr": 1.0707379495935828e-05, "data_time": 0.0014685630798339845, "loss": 0.3840280413627625, "time": 0.6303502321243286, "epoch": 10, "memory": 44120, "step": 36863} +{"lr": 1.0644906252255541e-05, "data_time": 0.0015579462051391602, "loss": 0.35381980836391447, "time": 0.6308066368103027, "epoch": 10, "memory": 44120, "step": 36963} +{"lr": 1.0585300843071527e-05, "data_time": 0.0015779972076416016, "loss": 0.3881597608327866, "time": 0.6311758279800415, "epoch": 10, "memory": 44120, "step": 37063} +{"lr": 1.0528567122246654e-05, "data_time": 0.0017542123794555664, "loss": 0.36496404111385344, "time": 0.6356167078018189, "epoch": 10, "memory": 44120, "step": 37163} +{"lr": 1.0474708757971141e-05, "data_time": 0.0016975641250610352, "loss": 0.36717942655086516, "time": 0.6304542541503906, "epoch": 10, "memory": 44120, "step": 37263} +{"lr": 1.042372923252543e-05, "data_time": 0.0015410661697387695, "loss": 0.3778563469648361, "time": 0.6327170372009278, "epoch": 10, "memory": 44120, "step": 37363} +{"lr": 1.0375631842054998e-05, "data_time": 0.0016430139541625977, "loss": 0.357115176320076, "time": 0.6307471752166748, "epoch": 10, "memory": 44120, "step": 37463} +{"lr": 1.0330419696357254e-05, "data_time": 0.0017735719680786132, "loss": 0.39231652915477755, "time": 0.6324982166290283, "epoch": 10, "memory": 44120, "step": 37563} +{"lr": 1.028809571868047e-05, "data_time": 0.0017905950546264649, "loss": 0.3452566027641296, "time": 0.633632755279541, "epoch": 10, "memory": 44120, "step": 37663} +{"lr": 1.0248662645534766e-05, "data_time": 0.0016438722610473632, "loss": 0.37007475793361666, "time": 0.6309433221817017, "epoch": 10, "memory": 44120, "step": 37763} +{"lr": 1.0212123026515194e-05, "data_time": 0.0018476724624633789, "loss": 0.35813745856285095, "time": 0.6313892364501953, "epoch": 10, "memory": 44120, "step": 37863} +{"lr": 1.0178479224136881e-05, "data_time": 0.0015517950057983398, "loss": 0.3813371449708939, "time": 0.6305741310119629, "epoch": 10, "memory": 44120, "step": 37963} +{"lr": 1.0147733413682289e-05, "data_time": 0.001695394515991211, "loss": 0.37492952346801756, "time": 0.6366972923278809, "epoch": 10, "memory": 44120, "step": 38063} +{"lr": 1.0119887583060527e-05, "data_time": 0.001740264892578125, "loss": 0.38165668547153475, "time": 0.631599235534668, "epoch": 10, "memory": 44120, "step": 38163} +{"lr": 1.0094943532678871e-05, "data_time": 0.0016669750213623047, "loss": 0.3390590101480484, "time": 0.6314989566802979, "epoch": 10, "memory": 44120, "step": 38263} +{"lr": 1.007290287532637e-05, "data_time": 0.0016634464263916016, "loss": 0.3734966367483139, "time": 0.6409272432327271, "epoch": 10, "memory": 44120, "step": 38363} +{"lr": 1.0053767036069495e-05, "data_time": 0.002427411079406738, "loss": 0.36938982605934145, "time": 0.6355569362640381, "epoch": 10, "memory": 44120, "step": 38463} +{"lr": 1.003753725216005e-05, "data_time": 0.0017248868942260742, "loss": 0.3700664281845093, "time": 0.6433971405029297, "epoch": 10, "memory": 44120, "step": 38563} +{"lr": 1.0024214572955152e-05, "data_time": 0.0016395092010498048, "loss": 0.3659755736589432, "time": 0.6300317764282226, "epoch": 10, "memory": 44120, "step": 38663} +{"lr": 1.0013799859849456e-05, "data_time": 0.00285491943359375, "loss": 0.381336572766304, "time": 0.6342385768890381, "epoch": 10, "memory": 44120, "step": 38763} +{"lr": 1.0006293786219338e-05, "data_time": 0.0022005081176757813, "loss": 0.3606461644172668, "time": 0.632568097114563, "epoch": 10, "memory": 44120, "step": 38863} +{"lr": 1.0001696837379472e-05, "data_time": 0.001591324806213379, "loss": 0.36118145287036896, "time": 0.6330610275268554, "epoch": 10, "memory": 44120, "step": 38963} +{"lr": 1.0000009310551361e-05, "data_time": 0.0012398004531860352, "loss": 0.36670683324337006, "time": 0.6310634851455689, "epoch": 10, "memory": 44120, "step": 39063} +{"accuracy/top1": 75.42535400390625, "data_time": 0.032334233152455295, "time": 1.2790674094496102, "step": 10} diff --git a/clip_large_pretrain_4x256_sdv2_lr1e-4/20230604_015520/vis_data/config.py b/clip_large_pretrain_4x256_sdv2_lr1e-4/20230604_015520/vis_data/config.py new file mode 100644 index 0000000000000000000000000000000000000000..1170b849c141e564d7e6619aa3d4313cfde1d220 --- /dev/null +++ b/clip_large_pretrain_4x256_sdv2_lr1e-4/20230604_015520/vis_data/config.py @@ -0,0 +1,227 @@ +optim_wrapper = dict( + optimizer=dict( + type='AdamW', lr=0.0001, weight_decay=0.3, _scope_='mmpretrain'), + paramwise_cfg=dict( + custom_keys=dict({ + '.cls_token': dict(decay_mult=0.0), + '.pos_embed': dict(decay_mult=0.0) + })), + type='AmpOptimWrapper', + dtype='bfloat16', + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=False, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=4096) +model = dict( + type='ImageClassifier', + backbone=dict( + frozen_stages=24, + type='VisionTransformer', + arch='l', + img_size=224, + patch_size=14, + drop_rate=0.1, + pre_norm=True, + init_cfg=dict( + type='Pretrained', + checkpoint='ckpt/openclip-ViT-L-14.pth', + prefix='backbone')), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=1024, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + init_cfg=None), + init_cfg=dict( + type='TruncNormal', layer=['Conv2d', 'Linear'], std=0.02, bias=0.0), + train_cfg=None) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=128, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stablediffusionV2-1-dpmsolver-25-1m.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/cc1m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=128, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV2-1-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=128, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV2-1-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/clip_large_pretrain_4x256_sdv2_lr1e-4' diff --git a/clip_large_pretrain_4x256_sdv2_lr1e-4/20230604_015520/vis_data/events.out.tfevents.1685814935.SH-IDC1-10-140-24-19.102915.0 b/clip_large_pretrain_4x256_sdv2_lr1e-4/20230604_015520/vis_data/events.out.tfevents.1685814935.SH-IDC1-10-140-24-19.102915.0 new file mode 100644 index 0000000000000000000000000000000000000000..2cb6fbbb49a73302fa8d5a0137b84ac26fe3e13e --- /dev/null +++ b/clip_large_pretrain_4x256_sdv2_lr1e-4/20230604_015520/vis_data/events.out.tfevents.1685814935.SH-IDC1-10-140-24-19.102915.0 @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:3eed09ebce4a72b42b96e33f102a74e26aa4db4614c4003448f694e67b04b076 +size 1714168 diff --git a/clip_large_pretrain_4x256_sdv2_lr1e-4/20230604_015520/vis_data/scalars.json b/clip_large_pretrain_4x256_sdv2_lr1e-4/20230604_015520/vis_data/scalars.json new file mode 100644 index 0000000000000000000000000000000000000000..5593ec92a6acc2947d636d4ce7feebe09bef86c9 --- /dev/null +++ b/clip_large_pretrain_4x256_sdv2_lr1e-4/20230604_015520/vis_data/scalars.json @@ -0,0 +1,400 @@ +{"lr": 9.999857418382631e-05, "data_time": 0.0015535831451416015, "loss": 0.6511398315429687, "time": 0.6302948713302612, "epoch": 1, "memory": 44120, "step": 100} +{"lr": 9.999423907359506e-05, "data_time": 0.0014174461364746093, "loss": 0.5779680252075196, "time": 0.6299609184265137, "epoch": 1, "memory": 44120, "step": 200} +{"lr": 9.998699480412073e-05, "data_time": 0.0014919996261596679, "loss": 0.5505671501159668, "time": 0.6316408395767212, "epoch": 1, "memory": 44120, "step": 300} +{"lr": 9.997684184379085e-05, "data_time": 0.002289700508117676, "loss": 0.5343621253967286, "time": 0.6332749843597412, "epoch": 1, "memory": 44120, "step": 400} +{"lr": 9.996378084905784e-05, "data_time": 0.0016236305236816406, "loss": 0.5073817253112793, "time": 0.6302190542221069, "epoch": 1, "memory": 44120, "step": 500} +{"lr": 9.994781266439667e-05, "data_time": 0.001563096046447754, "loss": 0.4989572048187256, "time": 0.6301367044448852, "epoch": 1, "memory": 44120, "step": 600} +{"lr": 9.992893832225061e-05, "data_time": 0.0016741275787353516, "loss": 0.5028618812561035, "time": 0.6291846990585327, "epoch": 1, "memory": 44120, "step": 700} +{"lr": 9.990715904296405e-05, "data_time": 0.0017473220825195313, "loss": 0.46791136264801025, "time": 0.6299782514572143, "epoch": 1, "memory": 44120, "step": 800} +{"lr": 9.988247623470371e-05, "data_time": 0.0016218662261962891, "loss": 0.4867336988449097, "time": 0.6294743776321411, "epoch": 1, "memory": 44120, "step": 900} +{"lr": 9.985489149336749e-05, "data_time": 0.001532769203186035, "loss": 0.4593863010406494, "time": 0.6297062397003174, "epoch": 1, "memory": 44120, "step": 1000} +{"lr": 9.982440660248187e-05, "data_time": 0.0014293432235717774, "loss": 0.4503697633743286, "time": 0.6294331789016724, "epoch": 1, "memory": 44120, "step": 1100} +{"lr": 9.979102353308574e-05, "data_time": 0.0015051126480102538, "loss": 0.4512750506401062, "time": 0.6301638841629028, "epoch": 1, "memory": 44120, "step": 1200} +{"lr": 9.975474444360358e-05, "data_time": 0.0015797138214111329, "loss": 0.45578144788742064, "time": 0.6316172122955322, "epoch": 1, "memory": 44120, "step": 1300} +{"lr": 9.97155716797056e-05, "data_time": 0.001532125473022461, "loss": 0.4727639019489288, "time": 0.6292188167572021, "epoch": 1, "memory": 44120, "step": 1400} +{"lr": 9.967350777415632e-05, "data_time": 0.0014520645141601562, "loss": 0.44409904479980467, "time": 0.6302995443344116, "epoch": 1, "memory": 44120, "step": 1500} +{"lr": 9.962855544665059e-05, "data_time": 0.0015540361404418946, "loss": 0.4235361576080322, "time": 0.6305093050003052, "epoch": 1, "memory": 44120, "step": 1600} +{"lr": 9.958071760363773e-05, "data_time": 0.0019036293029785155, "loss": 0.4788900136947632, "time": 0.6309087753295899, "epoch": 1, "memory": 44120, "step": 1700} +{"lr": 9.952999733813382e-05, "data_time": 0.0021437406539916992, "loss": 0.4394698917865753, "time": 0.6307754278182983, "epoch": 1, "memory": 44120, "step": 1800} +{"lr": 9.947639792952148e-05, "data_time": 0.0018276691436767579, "loss": 0.4205831170082092, "time": 0.6321241617202759, "epoch": 1, "memory": 44120, "step": 1900} +{"lr": 9.941992284333813e-05, "data_time": 0.001586604118347168, "loss": 0.4306331992149353, "time": 0.6313234567642212, "epoch": 1, "memory": 44120, "step": 2000} +{"lr": 9.93605757310516e-05, "data_time": 0.002110934257507324, "loss": 0.42725415527820587, "time": 0.6311674833297729, "epoch": 1, "memory": 44120, "step": 2100} +{"lr": 9.929836042982421e-05, "data_time": 0.0016530275344848633, "loss": 0.413659405708313, "time": 0.6323507070541382, "epoch": 1, "memory": 44120, "step": 2200} +{"lr": 9.923328096226467e-05, "data_time": 0.0017000436782836914, "loss": 0.41933178901672363, "time": 0.6326930522918701, "epoch": 1, "memory": 44120, "step": 2300} +{"lr": 9.916534153616819e-05, "data_time": 0.0015790700912475587, "loss": 0.4383719086647034, "time": 0.6312283754348755, "epoch": 1, "memory": 44120, "step": 2400} +{"lr": 9.909454654424387e-05, "data_time": 0.0014163970947265625, "loss": 0.40896245241165163, "time": 0.6320044279098511, "epoch": 1, "memory": 44120, "step": 2500} +{"lr": 9.902090056383109e-05, "data_time": 0.0016506671905517577, "loss": 0.4292390912771225, "time": 0.6302929401397706, "epoch": 1, "memory": 44120, "step": 2600} +{"lr": 9.894440835660365e-05, "data_time": 0.0015710830688476563, "loss": 0.41410685181617735, "time": 0.6296397686004639, "epoch": 1, "memory": 44120, "step": 2700} +{"lr": 9.886507486826162e-05, "data_time": 0.0016206979751586914, "loss": 0.3892812252044678, "time": 0.6307731866836548, "epoch": 1, "memory": 44120, "step": 2800} +{"lr": 9.878290522821154e-05, "data_time": 0.0014225006103515624, "loss": 0.39317906498908994, "time": 0.6291845798492431, "epoch": 1, "memory": 44120, "step": 2900} +{"lr": 9.86979047492352e-05, "data_time": 0.0021434307098388674, "loss": 0.3948826640844345, "time": 0.6313328742980957, "epoch": 1, "memory": 44120, "step": 3000} +{"lr": 9.861007892714578e-05, "data_time": 0.0016360759735107421, "loss": 0.36135985851287844, "time": 0.6303182125091553, "epoch": 1, "memory": 44120, "step": 3100} +{"lr": 9.851943344043239e-05, "data_time": 0.001565408706665039, "loss": 0.41074050664901735, "time": 0.6315772056579589, "epoch": 1, "memory": 44120, "step": 3200} +{"lr": 9.842597414989319e-05, "data_time": 0.0015559673309326171, "loss": 0.39188137650489807, "time": 0.6324409484863281, "epoch": 1, "memory": 44120, "step": 3300} +{"lr": 9.832970709825637e-05, "data_time": 0.0021782636642456053, "loss": 0.39438756108283995, "time": 0.6318658828735352, "epoch": 1, "memory": 44120, "step": 3400} +{"lr": 9.823063850978968e-05, "data_time": 0.0017688512802124024, "loss": 0.4241188198328018, "time": 0.6310411214828491, "epoch": 1, "memory": 44120, "step": 3500} +{"lr": 9.81287747898975e-05, "data_time": 0.0016953229904174804, "loss": 0.4126111209392548, "time": 0.6322051525115967, "epoch": 1, "memory": 44120, "step": 3600} +{"lr": 9.802412252470715e-05, "data_time": 0.0017744541168212891, "loss": 0.38610868752002714, "time": 0.6300936222076416, "epoch": 1, "memory": 44120, "step": 3700} +{"lr": 9.791668848064258e-05, "data_time": 0.001561450958251953, "loss": 0.4105738490819931, "time": 0.6288309812545776, "epoch": 1, "memory": 44120, "step": 3800} +{"lr": 9.780647960398751e-05, "data_time": 0.0016667604446411132, "loss": 0.37856527268886564, "time": 0.6306086301803588, "epoch": 1, "memory": 44120, "step": 3900} +{"accuracy/top1": 60.20319366455078, "data_time": 0.04584982520655582, "time": 1.3493186967414723, "step": 1} +{"lr": 9.7685491196646e-05, "data_time": 0.0023908138275146483, "loss": 0.4061006963253021, "time": 0.6311803579330444, "epoch": 2, "memory": 44120, "step": 4007} +{"lr": 9.756956126085676e-05, "data_time": 0.0014954090118408203, "loss": 0.403551721572876, "time": 0.6301365613937377, "epoch": 2, "memory": 44120, "step": 4107} +{"lr": 9.745087893642482e-05, "data_time": 0.0013546466827392579, "loss": 0.3965855985879898, "time": 0.6305136442184448, "epoch": 2, "memory": 44120, "step": 4207} +{"lr": 9.73294518969056e-05, "data_time": 0.0014646053314208984, "loss": 0.4431161880493164, "time": 0.6293402433395385, "epoch": 2, "memory": 44120, "step": 4307} +{"lr": 9.72052879933172e-05, "data_time": 0.001753544807434082, "loss": 0.37462318539619444, "time": 0.6295682430267334, "epoch": 2, "memory": 44120, "step": 4407} +{"lr": 9.707839525363358e-05, "data_time": 0.0014199256896972657, "loss": 0.4080145835876465, "time": 0.6294006824493408, "epoch": 2, "memory": 44120, "step": 4507} +{"lr": 9.69487818822646e-05, "data_time": 0.001447153091430664, "loss": 0.39367065727710726, "time": 0.6290924787521363, "epoch": 2, "memory": 44120, "step": 4607} +{"lr": 9.681645625952616e-05, "data_time": 0.001467132568359375, "loss": 0.36269129514694215, "time": 0.6295343637466431, "epoch": 2, "memory": 44120, "step": 4707} +{"lr": 9.668142694109826e-05, "data_time": 0.001475954055786133, "loss": 0.374642014503479, "time": 0.6383844614028931, "epoch": 2, "memory": 44120, "step": 4807} +{"lr": 9.654370265747162e-05, "data_time": 0.001619863510131836, "loss": 0.38116596043109896, "time": 0.630130934715271, "epoch": 2, "memory": 44120, "step": 4907} +{"lr": 9.640329231338345e-05, "data_time": 0.002016925811767578, "loss": 0.4128776788711548, "time": 0.6310742855072021, "epoch": 2, "memory": 44120, "step": 5007} +{"lr": 9.626020498724147e-05, "data_time": 0.001699209213256836, "loss": 0.40102483332157135, "time": 0.6309804677963257, "epoch": 2, "memory": 44120, "step": 5107} +{"lr": 9.611444993053717e-05, "data_time": 0.0015905141830444337, "loss": 0.37765371799468994, "time": 0.6297220945358276, "epoch": 2, "memory": 44120, "step": 5207} +{"lr": 9.596603656724746e-05, "data_time": 0.0015085220336914062, "loss": 0.38804788887500763, "time": 0.6301882982254028, "epoch": 2, "memory": 44120, "step": 5307} +{"lr": 9.58149744932254e-05, "data_time": 0.0013970613479614257, "loss": 0.3733738332986832, "time": 0.6317455768585205, "epoch": 2, "memory": 44120, "step": 5407} +{"lr": 9.566127347557972e-05, "data_time": 0.0015625476837158204, "loss": 0.3854053646326065, "time": 0.6299889326095581, "epoch": 2, "memory": 44120, "step": 5507} +{"lr": 9.550494345204367e-05, "data_time": 0.0015119791030883789, "loss": 0.38054428398609164, "time": 0.6295350313186645, "epoch": 2, "memory": 44120, "step": 5607} +{"lr": 9.534599453033183e-05, "data_time": 0.0015683650970458984, "loss": 0.38338888585567477, "time": 0.6296643257141114, "epoch": 2, "memory": 44120, "step": 5707} +{"lr": 9.518443698748712e-05, "data_time": 0.001562023162841797, "loss": 0.36804080903530123, "time": 0.6304186344146728, "epoch": 2, "memory": 44120, "step": 5807} +{"lr": 9.50202812692162e-05, "data_time": 0.001341700553894043, "loss": 0.3973271459341049, "time": 0.6301617860794068, "epoch": 2, "memory": 44120, "step": 5907} +{"lr": 9.485353798921386e-05, "data_time": 0.0014605522155761719, "loss": 0.37133278846740725, "time": 0.6311134576797486, "epoch": 2, "memory": 44120, "step": 6007} +{"lr": 9.468421792847712e-05, "data_time": 0.0013740062713623047, "loss": 0.37527437806129454, "time": 0.630155062675476, "epoch": 2, "memory": 44120, "step": 6107} +{"lr": 9.451233203460794e-05, "data_time": 0.0013575315475463866, "loss": 0.3760660618543625, "time": 0.6299977779388428, "epoch": 2, "memory": 44120, "step": 6207} +{"lr": 9.433789142110532e-05, "data_time": 0.0017714977264404296, "loss": 0.3865980327129364, "time": 0.6299216508865356, "epoch": 2, "memory": 44120, "step": 6307} +{"lr": 9.416090736664709e-05, "data_time": 0.0016645431518554688, "loss": 0.3935562491416931, "time": 0.6303906202316284, "epoch": 2, "memory": 44120, "step": 6407} +{"lr": 9.398139131436032e-05, "data_time": 0.0015691995620727539, "loss": 0.4170177221298218, "time": 0.6298570394515991, "epoch": 2, "memory": 44120, "step": 6507} +{"lr": 9.379935487108167e-05, "data_time": 0.001615595817565918, "loss": 0.38354268074035647, "time": 0.630503249168396, "epoch": 2, "memory": 44120, "step": 6607} +{"lr": 9.361480980660697e-05, "data_time": 0.0015149831771850586, "loss": 0.3863678127527237, "time": 0.6301994800567627, "epoch": 2, "memory": 44120, "step": 6707} +{"lr": 9.342776805292993e-05, "data_time": 0.001593160629272461, "loss": 0.36150650680065155, "time": 0.6305224657058716, "epoch": 2, "memory": 44120, "step": 6807} +{"lr": 9.323824170347083e-05, "data_time": 0.0014492273330688477, "loss": 0.3628203600645065, "time": 0.6301002740859986, "epoch": 2, "memory": 44120, "step": 6907} +{"lr": 9.304624301229471e-05, "data_time": 0.0016742706298828124, "loss": 0.3921812415122986, "time": 0.6335572242736817, "epoch": 2, "memory": 44120, "step": 7007} +{"lr": 9.285178439331897e-05, "data_time": 0.0017034769058227538, "loss": 0.38591637313365934, "time": 0.630746603012085, "epoch": 2, "memory": 44120, "step": 7107} +{"lr": 9.265487841951087e-05, "data_time": 0.0014682531356811524, "loss": 0.39570842385292054, "time": 0.6305039882659912, "epoch": 2, "memory": 44120, "step": 7207} +{"lr": 9.245553782207421e-05, "data_time": 0.0016492128372192383, "loss": 0.387408047914505, "time": 0.6317680358886719, "epoch": 2, "memory": 44120, "step": 7307} +{"lr": 9.225377548962671e-05, "data_time": 0.0016398191452026366, "loss": 0.3795217454433441, "time": 0.6297847986221313, "epoch": 2, "memory": 44120, "step": 7407} +{"lr": 9.204960446736601e-05, "data_time": 0.001511359214782715, "loss": 0.38568209707736967, "time": 0.6308626890182495, "epoch": 2, "memory": 44120, "step": 7507} +{"lr": 9.184303795622715e-05, "data_time": 0.0014635086059570312, "loss": 0.39033634662628175, "time": 0.6300491571426392, "epoch": 2, "memory": 44120, "step": 7607} +{"lr": 9.163408931202823e-05, "data_time": 0.001522397994995117, "loss": 0.39881399273872375, "time": 0.6334480285644531, "epoch": 2, "memory": 44120, "step": 7707} +{"lr": 9.142277204460729e-05, "data_time": 0.0013886690139770508, "loss": 0.3804485470056534, "time": 0.6289843797683716, "epoch": 2, "memory": 44120, "step": 7807} +{"accuracy/top1": 66.76525115966797, "data_time": 0.03664940801160089, "time": 1.2851452293067143, "step": 2} +{"lr": 9.119405492383288e-05, "data_time": 0.0013090848922729492, "loss": 0.36767544150352477, "time": 0.6290694952011109, "epoch": 3, "memory": 44120, "step": 7914} +{"lr": 9.097787819233205e-05, "data_time": 0.001606321334838867, "loss": 0.38307862877845766, "time": 0.6293684482574463, "epoch": 3, "memory": 44120, "step": 8014} +{"lr": 9.075937526576627e-05, "data_time": 0.0019600391387939453, "loss": 0.37978521585464475, "time": 0.6312113046646118, "epoch": 3, "memory": 44120, "step": 8114} +{"lr": 9.053856027171755e-05, "data_time": 0.001861906051635742, "loss": 0.37750745117664336, "time": 0.6299858570098877, "epoch": 3, "memory": 44120, "step": 8214} +{"lr": 9.031544748725785e-05, "data_time": 0.001420307159423828, "loss": 0.37020819187164306, "time": 0.6302985429763794, "epoch": 3, "memory": 44120, "step": 8314} +{"lr": 9.009005133802542e-05, "data_time": 0.0016502618789672851, "loss": 0.38140026926994325, "time": 0.6302116394042969, "epoch": 3, "memory": 44120, "step": 8414} +{"lr": 8.986238639729242e-05, "data_time": 0.001695728302001953, "loss": 0.39678852558135985, "time": 0.6301708221435547, "epoch": 3, "memory": 44120, "step": 8514} +{"lr": 8.963246738502254e-05, "data_time": 0.001857590675354004, "loss": 0.4026592016220093, "time": 0.6328332662582398, "epoch": 3, "memory": 44120, "step": 8614} +{"lr": 8.940030916691928e-05, "data_time": 0.0017349720001220703, "loss": 0.3703895002603531, "time": 0.6312051057815552, "epoch": 3, "memory": 44120, "step": 8714} +{"lr": 8.91659267534649e-05, "data_time": 0.0019045829772949218, "loss": 0.4036286801099777, "time": 0.6318345308303833, "epoch": 3, "memory": 44120, "step": 8814} +{"lr": 8.892933529894978e-05, "data_time": 0.002414250373840332, "loss": 0.36947562992572786, "time": 0.6323338270187377, "epoch": 3, "memory": 44120, "step": 8914} +{"lr": 8.869055010049283e-05, "data_time": 0.0022028446197509765, "loss": 0.35781918466091156, "time": 0.6331470251083374, "epoch": 3, "memory": 44120, "step": 9014} +{"lr": 8.844958659705188e-05, "data_time": 0.0017703771591186523, "loss": 0.3781828165054321, "time": 0.630960202217102, "epoch": 3, "memory": 44120, "step": 9114} +{"lr": 8.82064603684259e-05, "data_time": 0.0017389774322509766, "loss": 0.3636253088712692, "time": 0.6301772832870484, "epoch": 3, "memory": 44120, "step": 9214} +{"lr": 8.796118713424776e-05, "data_time": 0.0014866113662719727, "loss": 0.3884113222360611, "time": 0.6302128791809082, "epoch": 3, "memory": 44120, "step": 9314} +{"lr": 8.77137827529676e-05, "data_time": 0.001845383644104004, "loss": 0.3719728320837021, "time": 0.6318903923034668, "epoch": 3, "memory": 44120, "step": 9414} +{"lr": 8.746426322082745e-05, "data_time": 0.0015121936798095704, "loss": 0.37726588547229767, "time": 0.6308449745178223, "epoch": 3, "memory": 44120, "step": 9514} +{"lr": 8.721264467082732e-05, "data_time": 0.0016206979751586914, "loss": 0.36793930530548097, "time": 0.6302549123764039, "epoch": 3, "memory": 44120, "step": 9614} +{"lr": 8.695894337168168e-05, "data_time": 0.002411556243896484, "loss": 0.3705622971057892, "time": 0.6314141273498535, "epoch": 3, "memory": 44120, "step": 9714} +{"lr": 8.67031757267677e-05, "data_time": 0.0017479181289672852, "loss": 0.3867658466100693, "time": 0.6301435470581055, "epoch": 3, "memory": 44120, "step": 9814} +{"lr": 8.64453582730651e-05, "data_time": 0.0016212224960327148, "loss": 0.373067370057106, "time": 0.6304151535034179, "epoch": 3, "memory": 44120, "step": 9914} +{"lr": 8.61855076800861e-05, "data_time": 0.0017477989196777344, "loss": 0.3726970463991165, "time": 0.6302961349487305, "epoch": 3, "memory": 44120, "step": 10014} +{"lr": 8.592364074879843e-05, "data_time": 0.001483011245727539, "loss": 0.3646294206380844, "time": 0.6306333065032959, "epoch": 3, "memory": 44120, "step": 10114} +{"lr": 8.565977441053854e-05, "data_time": 0.0015407085418701171, "loss": 0.3594058692455292, "time": 0.6309885025024414, "epoch": 3, "memory": 44120, "step": 10214} +{"lr": 8.539392572591724e-05, "data_time": 0.0018303394317626953, "loss": 0.34311483502388, "time": 0.6308549880981446, "epoch": 3, "memory": 44120, "step": 10314} +{"lr": 8.512611188371633e-05, "data_time": 0.0018095254898071289, "loss": 0.36856233775615693, "time": 0.6306205511093139, "epoch": 3, "memory": 44120, "step": 10414} +{"lr": 8.485635019977725e-05, "data_time": 0.0015825748443603516, "loss": 0.3896019458770752, "time": 0.630800437927246, "epoch": 3, "memory": 44120, "step": 10514} +{"lr": 8.45846581158818e-05, "data_time": 0.0016936540603637695, "loss": 0.37825354039669035, "time": 0.630472207069397, "epoch": 3, "memory": 44120, "step": 10614} +{"lr": 8.431105319862416e-05, "data_time": 0.0017005681991577148, "loss": 0.3651675909757614, "time": 0.6305620670318604, "epoch": 3, "memory": 44120, "step": 10714} +{"lr": 8.403555313827516e-05, "data_time": 0.002501082420349121, "loss": 0.3763491839170456, "time": 0.6325437784194946, "epoch": 3, "memory": 44120, "step": 10814} +{"lr": 8.375817574763844e-05, "data_time": 0.0019482135772705077, "loss": 0.3438781201839447, "time": 0.6315265893936157, "epoch": 3, "memory": 44120, "step": 10914} +{"lr": 8.347893896089871e-05, "data_time": 0.0018458366394042969, "loss": 0.3857393771409988, "time": 0.6343234062194825, "epoch": 3, "memory": 44120, "step": 11014} +{"lr": 8.31978608324625e-05, "data_time": 0.0017656326293945313, "loss": 0.3934557676315308, "time": 0.6319288015365601, "epoch": 3, "memory": 44120, "step": 11114} +{"lr": 8.291495953579052e-05, "data_time": 0.0029499292373657226, "loss": 0.3598293989896774, "time": 0.6363593578338623, "epoch": 3, "memory": 44120, "step": 11214} +{"lr": 8.263025336222267e-05, "data_time": 0.0017736434936523437, "loss": 0.36102803945541384, "time": 0.6317667007446289, "epoch": 3, "memory": 44120, "step": 11314} +{"lr": 8.234376071979551e-05, "data_time": 0.0015689373016357423, "loss": 0.3519564002752304, "time": 0.6314729452133179, "epoch": 3, "memory": 44120, "step": 11414} +{"lr": 8.205550013205187e-05, "data_time": 0.001592254638671875, "loss": 0.4015227615833282, "time": 0.63122398853302, "epoch": 3, "memory": 44120, "step": 11514} +{"lr": 8.17654902368435e-05, "data_time": 0.0017137765884399415, "loss": 0.3792177468538284, "time": 0.6312206506729126, "epoch": 3, "memory": 44120, "step": 11614} +{"lr": 8.147374978512582e-05, "data_time": 0.0012741327285766602, "loss": 0.39173182249069216, "time": 0.630467700958252, "epoch": 3, "memory": 44120, "step": 11714} +{"accuracy/top1": 70.75606536865234, "data_time": 0.03451025897058947, "time": 1.2825194761670868, "step": 3} +{"lr": 8.115969237626578e-05, "data_time": 0.00140688419342041, "loss": 0.35677762031555177, "time": 0.6293562650680542, "epoch": 4, "memory": 44120, "step": 11821} +{"lr": 8.08644297337245e-05, "data_time": 0.0014370918273925782, "loss": 0.3732084989547729, "time": 0.6307722568511963, "epoch": 4, "memory": 44120, "step": 11921} +{"lr": 8.056749479387651e-05, "data_time": 0.0015021562576293945, "loss": 0.3886274993419647, "time": 0.6319432258605957, "epoch": 4, "memory": 44120, "step": 12021} +{"lr": 8.026890675542457e-05, "data_time": 0.001486063003540039, "loss": 0.37129161059856414, "time": 0.6311835527420044, "epoch": 4, "memory": 44120, "step": 12121} +{"lr": 7.996868492395457e-05, "data_time": 0.0015145301818847655, "loss": 0.3772895783185959, "time": 0.629550838470459, "epoch": 4, "memory": 44120, "step": 12221} +{"lr": 7.966684871068762e-05, "data_time": 0.001385664939880371, "loss": 0.3722930520772934, "time": 0.6305479049682617, "epoch": 4, "memory": 44120, "step": 12321} +{"lr": 7.936341763122449e-05, "data_time": 0.0015053272247314454, "loss": 0.34028970301151273, "time": 0.6314324617385865, "epoch": 4, "memory": 44120, "step": 12421} +{"lr": 7.905841130428405e-05, "data_time": 0.0014670610427856446, "loss": 0.3500517576932907, "time": 0.6303344964981079, "epoch": 4, "memory": 44120, "step": 12521} +{"lr": 7.875184945043487e-05, "data_time": 0.0018103837966918946, "loss": 0.38720719516277313, "time": 0.6401841402053833, "epoch": 4, "memory": 44120, "step": 12621} +{"lr": 7.844375189082012e-05, "data_time": 0.0018702507019042968, "loss": 0.3647387117147446, "time": 0.6307103872299195, "epoch": 4, "memory": 44120, "step": 12721} +{"lr": 7.813413854587572e-05, "data_time": 0.0023189544677734374, "loss": 0.35620381236076354, "time": 0.6318166494369507, "epoch": 4, "memory": 44120, "step": 12821} +{"lr": 7.782302943404291e-05, "data_time": 0.002088785171508789, "loss": 0.3544735312461853, "time": 0.6318197965621948, "epoch": 4, "memory": 44120, "step": 12921} +{"lr": 7.751044467047337e-05, "data_time": 0.002436661720275879, "loss": 0.39118861556053164, "time": 0.6354005336761475, "epoch": 4, "memory": 44120, "step": 13021} +{"lr": 7.719640446572907e-05, "data_time": 0.002176833152770996, "loss": 0.354831263422966, "time": 0.6314675092697144, "epoch": 4, "memory": 44120, "step": 13121} +{"lr": 7.688092912447536e-05, "data_time": 0.0024971961975097656, "loss": 0.33954288959503176, "time": 0.6356311082839966, "epoch": 4, "memory": 44120, "step": 13221} +{"lr": 7.656403904416794e-05, "data_time": 0.0023842573165893553, "loss": 0.358999764919281, "time": 0.6345445871353149, "epoch": 4, "memory": 44120, "step": 13321} +{"lr": 7.624575471373434e-05, "data_time": 0.0026422739028930664, "loss": 0.3761797368526459, "time": 0.6329836130142212, "epoch": 4, "memory": 44120, "step": 13421} +{"lr": 7.592609671224926e-05, "data_time": 0.002575516700744629, "loss": 0.36268415451049807, "time": 0.6348317861557007, "epoch": 4, "memory": 44120, "step": 13521} +{"lr": 7.560508570760375e-05, "data_time": 0.002576899528503418, "loss": 0.3569221556186676, "time": 0.6331686496734619, "epoch": 4, "memory": 44120, "step": 13621} +{"lr": 7.528274245516895e-05, "data_time": 0.0024100780487060548, "loss": 0.3351865142583847, "time": 0.6345189094543457, "epoch": 4, "memory": 44120, "step": 13721} +{"lr": 7.495908779645405e-05, "data_time": 0.0021024465560913084, "loss": 0.3555357575416565, "time": 0.6310706853866577, "epoch": 4, "memory": 44120, "step": 13821} +{"lr": 7.463414265775906e-05, "data_time": 0.0021494150161743162, "loss": 0.3550543010234833, "time": 0.6348968267440795, "epoch": 4, "memory": 44120, "step": 13921} +{"lr": 7.430792804882147e-05, "data_time": 0.0022434711456298826, "loss": 0.37112358808517454, "time": 0.6317383289337158, "epoch": 4, "memory": 44120, "step": 14021} +{"lr": 7.398046506145803e-05, "data_time": 0.0022365570068359373, "loss": 0.3770568609237671, "time": 0.6320046424865723, "epoch": 4, "memory": 44120, "step": 14121} +{"lr": 7.365177486820114e-05, "data_time": 0.0020328521728515624, "loss": 0.3532114714384079, "time": 0.6325595617294312, "epoch": 4, "memory": 44120, "step": 14221} +{"lr": 7.332187872092939e-05, "data_time": 0.001931619644165039, "loss": 0.4026179492473602, "time": 0.6314249515533448, "epoch": 4, "memory": 44120, "step": 14321} +{"lr": 7.299079794949401e-05, "data_time": 0.002328300476074219, "loss": 0.3726774573326111, "time": 0.6362221717834473, "epoch": 4, "memory": 44120, "step": 14421} +{"lr": 7.265855396033955e-05, "data_time": 0.0020601511001586913, "loss": 0.3620989680290222, "time": 0.6323452949523926, "epoch": 4, "memory": 44120, "step": 14521} +{"lr": 7.232516823512013e-05, "data_time": 0.0018093109130859375, "loss": 0.3517916172742844, "time": 0.6318056344985962, "epoch": 4, "memory": 44120, "step": 14621} +{"lr": 7.199066232930995e-05, "data_time": 0.0018860101699829102, "loss": 0.3624063342809677, "time": 0.6316128730773926, "epoch": 4, "memory": 44120, "step": 14721} +{"lr": 7.165505787080997e-05, "data_time": 0.002146315574645996, "loss": 0.3662576645612717, "time": 0.6343168020248413, "epoch": 4, "memory": 44120, "step": 14821} +{"lr": 7.13183765585495e-05, "data_time": 0.002161383628845215, "loss": 0.3525313138961792, "time": 0.6326789855957031, "epoch": 4, "memory": 44120, "step": 14921} +{"lr": 7.098064016108317e-05, "data_time": 0.0018190383911132813, "loss": 0.37121322751045227, "time": 0.6303757429122925, "epoch": 4, "memory": 44120, "step": 15021} +{"lr": 7.064187051518379e-05, "data_time": 0.0016277551651000977, "loss": 0.36289057433605193, "time": 0.6301361322402954, "epoch": 4, "memory": 44120, "step": 15121} +{"lr": 7.030208952442949e-05, "data_time": 0.0014097213745117188, "loss": 0.40250545740127563, "time": 0.6292129755020142, "epoch": 4, "memory": 44120, "step": 15221} +{"lr": 6.996131915778851e-05, "data_time": 0.0014852523803710938, "loss": 0.35194175839424136, "time": 0.6314897298812866, "epoch": 4, "memory": 44120, "step": 15321} +{"lr": 6.961958144819853e-05, "data_time": 0.001454782485961914, "loss": 0.367633318901062, "time": 0.6316639423370362, "epoch": 4, "memory": 44120, "step": 15421} +{"lr": 6.927689849114193e-05, "data_time": 0.0015785932540893556, "loss": 0.38853678703308103, "time": 0.6305542230606079, "epoch": 4, "memory": 44120, "step": 15521} +{"lr": 6.893329244321722e-05, "data_time": 0.0012257814407348633, "loss": 0.36316122114658356, "time": 0.6299135684967041, "epoch": 4, "memory": 44120, "step": 15621} +{"accuracy/top1": 72.92369842529297, "data_time": 0.033927378983333194, "time": 1.2814092964961612, "step": 4} +{"lr": 6.856463687361792e-05, "data_time": 0.0014965295791625976, "loss": 0.36035783737897875, "time": 0.6297960758209229, "epoch": 5, "memory": 44120, "step": 15728} +{"lr": 6.821919068470432e-05, "data_time": 0.0014540910720825194, "loss": 0.37258667349815366, "time": 0.630597186088562, "epoch": 5, "memory": 44120, "step": 15828} +{"lr": 6.787288979235407e-05, "data_time": 0.0021564245223999025, "loss": 0.3697516113519669, "time": 0.6314488410949707, "epoch": 5, "memory": 44120, "step": 15928} +{"lr": 6.752575658708793e-05, "data_time": 0.0016209602355957032, "loss": 0.3605677604675293, "time": 0.6309216022491455, "epoch": 5, "memory": 44120, "step": 16028} +{"lr": 6.71778135132408e-05, "data_time": 0.0013944864273071288, "loss": 0.37399953305721284, "time": 0.6309323072433471, "epoch": 5, "memory": 44120, "step": 16128} +{"lr": 6.682908306751074e-05, "data_time": 0.0015766382217407226, "loss": 0.37510315477848055, "time": 0.6325395107269287, "epoch": 5, "memory": 44120, "step": 16228} +{"lr": 6.64795877975043e-05, "data_time": 0.001371455192565918, "loss": 0.37172494232654574, "time": 0.6315000295639038, "epoch": 5, "memory": 44120, "step": 16328} +{"lr": 6.61293503002787e-05, "data_time": 0.0014748811721801759, "loss": 0.35580343902111056, "time": 0.6327751398086547, "epoch": 5, "memory": 44120, "step": 16428} +{"lr": 6.577839322088084e-05, "data_time": 0.0018471479415893555, "loss": 0.363631945848465, "time": 0.6312255620956421, "epoch": 5, "memory": 44120, "step": 16528} +{"lr": 6.542673925088298e-05, "data_time": 0.001959562301635742, "loss": 0.3688123047351837, "time": 0.6371283531188965, "epoch": 5, "memory": 44120, "step": 16628} +{"lr": 6.507441112691584e-05, "data_time": 0.0015973329544067382, "loss": 0.3715033084154129, "time": 0.6311631917953491, "epoch": 5, "memory": 44120, "step": 16728} +{"lr": 6.47214316291983e-05, "data_time": 0.0017039060592651367, "loss": 0.34398736655712125, "time": 0.6322572946548461, "epoch": 5, "memory": 44120, "step": 16828} +{"lr": 6.436782358006481e-05, "data_time": 0.0015607118606567384, "loss": 0.3612892985343933, "time": 0.6309884309768676, "epoch": 5, "memory": 44120, "step": 16928} +{"lr": 6.40136098424896e-05, "data_time": 0.001902627944946289, "loss": 0.33692377507686616, "time": 0.6304582118988037, "epoch": 5, "memory": 44120, "step": 17028} +{"lr": 6.365881331860828e-05, "data_time": 0.0015949487686157226, "loss": 0.35666256546974184, "time": 0.6300513744354248, "epoch": 5, "memory": 44120, "step": 17128} +{"lr": 6.330345694823739e-05, "data_time": 0.001678800582885742, "loss": 0.38750146329402924, "time": 0.6304916381835938, "epoch": 5, "memory": 44120, "step": 17228} +{"lr": 6.29475637073908e-05, "data_time": 0.0015712738037109374, "loss": 0.3811786353588104, "time": 0.6323616743087769, "epoch": 5, "memory": 44120, "step": 17328} +{"lr": 6.259115660679482e-05, "data_time": 0.0015985727310180663, "loss": 0.3553478717803955, "time": 0.6314690351486206, "epoch": 5, "memory": 44120, "step": 17428} +{"lr": 6.223425869039963e-05, "data_time": 0.0021724462509155273, "loss": 0.39600721895694735, "time": 0.6326065778732299, "epoch": 5, "memory": 44120, "step": 17528} +{"lr": 6.187689303389006e-05, "data_time": 0.001455831527709961, "loss": 0.36005370914936063, "time": 0.6312793254852295, "epoch": 5, "memory": 44120, "step": 17628} +{"lr": 6.151908274319292e-05, "data_time": 0.0015760183334350586, "loss": 0.34296722114086153, "time": 0.6310264587402343, "epoch": 5, "memory": 44120, "step": 17728} +{"lr": 6.116085095298373e-05, "data_time": 0.001760578155517578, "loss": 0.36004377603530885, "time": 0.6306248188018799, "epoch": 5, "memory": 44120, "step": 17828} +{"lr": 6.080222082519007e-05, "data_time": 0.0017028093338012696, "loss": 0.35072051435709, "time": 0.630894660949707, "epoch": 5, "memory": 44120, "step": 17928} +{"lr": 6.0443215547495147e-05, "data_time": 0.0016824960708618163, "loss": 0.38175966739654543, "time": 0.6305303573608398, "epoch": 5, "memory": 44120, "step": 18028} +{"lr": 6.0083858331837714e-05, "data_time": 0.0015622854232788086, "loss": 0.3692367136478424, "time": 0.6308236360549927, "epoch": 5, "memory": 44120, "step": 18128} +{"lr": 5.97241724129116e-05, "data_time": 0.001533842086791992, "loss": 0.3736819863319397, "time": 0.6304581880569458, "epoch": 5, "memory": 44120, "step": 18228} +{"lr": 5.936418104666328e-05, "data_time": 0.0014395475387573241, "loss": 0.37733722627162936, "time": 0.6304697036743164, "epoch": 5, "memory": 44120, "step": 18328} +{"lr": 5.9003907508788295e-05, "data_time": 0.0017235279083251953, "loss": 0.3453709274530411, "time": 0.6312472105026246, "epoch": 5, "memory": 44120, "step": 18428} +{"lr": 5.864337509322633e-05, "data_time": 0.0019505977630615234, "loss": 0.3516500622034073, "time": 0.6328014850616455, "epoch": 5, "memory": 44120, "step": 18528} +{"lr": 5.8282607110655194e-05, "data_time": 0.0017161607742309571, "loss": 0.3615013867616653, "time": 0.631293249130249, "epoch": 5, "memory": 44120, "step": 18628} +{"lr": 5.7921626886983564e-05, "data_time": 0.0028637170791625975, "loss": 0.37235530018806456, "time": 0.6343429327011109, "epoch": 5, "memory": 44120, "step": 18728} +{"lr": 5.756045776184287e-05, "data_time": 0.0018888473510742187, "loss": 0.3680415689945221, "time": 0.6325015544891357, "epoch": 5, "memory": 44120, "step": 18828} +{"lr": 5.719912308707808e-05, "data_time": 0.0016191720962524414, "loss": 0.3743208646774292, "time": 0.6393663644790649, "epoch": 5, "memory": 44120, "step": 18928} +{"lr": 5.683764622523808e-05, "data_time": 0.001971936225891113, "loss": 0.36365104019641875, "time": 0.6324900150299072, "epoch": 5, "memory": 44120, "step": 19028} +{"lr": 5.647605054806502e-05, "data_time": 0.0016748666763305663, "loss": 0.3520662605762482, "time": 0.6315549850463867, "epoch": 5, "memory": 44120, "step": 19128} +{"lr": 5.6114359434983206e-05, "data_time": 0.0016469955444335938, "loss": 0.3705656111240387, "time": 0.6313248634338379, "epoch": 5, "memory": 44120, "step": 19228} +{"lr": 5.575259627158745e-05, "data_time": 0.0015885353088378907, "loss": 0.3683649569749832, "time": 0.631161093711853, "epoch": 5, "memory": 44120, "step": 19328} +{"lr": 5.539078444813113e-05, "data_time": 0.0016903400421142579, "loss": 0.39164201021194456, "time": 0.6324179649353028, "epoch": 5, "memory": 44120, "step": 19428} +{"lr": 5.5028947358013735e-05, "data_time": 0.001396942138671875, "loss": 0.3690459430217743, "time": 0.6315234661102295, "epoch": 5, "memory": 44120, "step": 19528} +{"accuracy/top1": 74.18670654296875, "data_time": 0.033750776586861445, "time": 1.2835038777055412, "step": 5} +{"lr": 5.464178020340108e-05, "data_time": 0.0014005661010742187, "loss": 0.3426802784204483, "time": 0.6311736106872559, "epoch": 6, "memory": 44120, "step": 19635} +{"lr": 5.427996514795046e-05, "data_time": 0.00141599178314209, "loss": 0.36596367657184603, "time": 0.6307308197021484, "epoch": 6, "memory": 44120, "step": 19735} +{"lr": 5.3918196647260834e-05, "data_time": 0.0016403675079345703, "loss": 0.3611610263586044, "time": 0.6345808744430542, "epoch": 6, "memory": 44120, "step": 19835} +{"lr": 5.355649809193057e-05, "data_time": 0.001447582244873047, "loss": 0.3365130931138992, "time": 0.6309124231338501, "epoch": 6, "memory": 44120, "step": 19935} +{"lr": 5.3194892868035765e-05, "data_time": 0.0014417409896850587, "loss": 0.37650632560253144, "time": 0.6317238807678223, "epoch": 6, "memory": 44120, "step": 20035} +{"lr": 5.283340435561795e-05, "data_time": 0.0016066789627075194, "loss": 0.3631355732679367, "time": 0.6313675880432129, "epoch": 6, "memory": 44120, "step": 20135} +{"lr": 5.2472055927172495e-05, "data_time": 0.0017436981201171876, "loss": 0.3929067015647888, "time": 0.631146502494812, "epoch": 6, "memory": 44120, "step": 20235} +{"lr": 5.2110870946137605e-05, "data_time": 0.0015924215316772462, "loss": 0.3521174430847168, "time": 0.6329382658004761, "epoch": 6, "memory": 44120, "step": 20335} +{"lr": 5.174987276538342e-05, "data_time": 0.0017699241638183595, "loss": 0.36878088116645813, "time": 0.6312318325042725, "epoch": 6, "memory": 44120, "step": 20435} +{"lr": 5.1389084725702425e-05, "data_time": 0.0018259286880493164, "loss": 0.3505380988121033, "time": 0.6309262752532959, "epoch": 6, "memory": 44120, "step": 20535} +{"lr": 5.1028530154300045e-05, "data_time": 0.00170440673828125, "loss": 0.3884235739707947, "time": 0.6307322978973389, "epoch": 6, "memory": 44120, "step": 20635} +{"lr": 5.066823236328663e-05, "data_time": 0.0017857313156127929, "loss": 0.35251970291137696, "time": 0.6326623439788819, "epoch": 6, "memory": 44120, "step": 20735} +{"lr": 5.030821464817006e-05, "data_time": 0.0016338586807250976, "loss": 0.376922196149826, "time": 0.63508620262146, "epoch": 6, "memory": 44120, "step": 20835} +{"lr": 4.99485002863494e-05, "data_time": 0.0015831708908081055, "loss": 0.3826912134885788, "time": 0.6432285785675049, "epoch": 6, "memory": 44120, "step": 20935} +{"lr": 4.95891125356102e-05, "data_time": 0.0017174482345581055, "loss": 0.34901047348976133, "time": 0.6320256233215332, "epoch": 6, "memory": 44120, "step": 21035} +{"lr": 4.923007463262053e-05, "data_time": 0.0016624212265014648, "loss": 0.3568054050207138, "time": 0.6320607900619507, "epoch": 6, "memory": 44120, "step": 21135} +{"lr": 4.887140979142852e-05, "data_time": 0.0018236637115478516, "loss": 0.36538914740085604, "time": 0.6317174673080445, "epoch": 6, "memory": 44120, "step": 21235} +{"lr": 4.851314120196161e-05, "data_time": 0.0016582965850830077, "loss": 0.3845347762107849, "time": 0.6306475400924683, "epoch": 6, "memory": 44120, "step": 21335} +{"lr": 4.815529202852693e-05, "data_time": 0.0015470743179321288, "loss": 0.36047182977199554, "time": 0.6317041158676148, "epoch": 6, "memory": 44120, "step": 21435} +{"lr": 4.779788540831406e-05, "data_time": 0.001522231101989746, "loss": 0.3793373882770538, "time": 0.6315422296524048, "epoch": 6, "memory": 44120, "step": 21535} +{"lr": 4.744094444989843e-05, "data_time": 0.0014984607696533203, "loss": 0.38657639026641843, "time": 0.6309186458587647, "epoch": 6, "memory": 44120, "step": 21635} +{"lr": 4.708449223174764e-05, "data_time": 0.001526641845703125, "loss": 0.36951213479042055, "time": 0.6311927080154419, "epoch": 6, "memory": 44120, "step": 21735} +{"lr": 4.672855180072916e-05, "data_time": 0.0015697002410888672, "loss": 0.3582223176956177, "time": 0.6308010578155517, "epoch": 6, "memory": 44120, "step": 21835} +{"lr": 4.6373146170620256e-05, "data_time": 0.0018631219863891602, "loss": 0.3661968469619751, "time": 0.6326027870178222, "epoch": 6, "memory": 44120, "step": 21935} +{"lr": 4.6018298320619915e-05, "data_time": 0.001543712615966797, "loss": 0.3824609011411667, "time": 0.6340478420257568, "epoch": 6, "memory": 44120, "step": 22035} +{"lr": 4.566403119386318e-05, "data_time": 0.0019037723541259766, "loss": 0.39292274713516234, "time": 0.6347447872161865, "epoch": 6, "memory": 44120, "step": 22135} +{"lr": 4.531036769593769e-05, "data_time": 0.0017054080963134766, "loss": 0.3460966289043427, "time": 0.6304093837738037, "epoch": 6, "memory": 44120, "step": 22235} +{"lr": 4.4957330693402685e-05, "data_time": 0.001736593246459961, "loss": 0.3412874609231949, "time": 0.6325237035751343, "epoch": 6, "memory": 44120, "step": 22335} +{"lr": 4.460494301231059e-05, "data_time": 0.0014856100082397462, "loss": 0.3635822504758835, "time": 0.6321806669235229, "epoch": 6, "memory": 44120, "step": 22435} +{"lr": 4.425322743673109e-05, "data_time": 0.0016674757003784179, "loss": 0.38342941701412203, "time": 0.630539870262146, "epoch": 6, "memory": 44120, "step": 22535} +{"lr": 4.390220670727812e-05, "data_time": 0.0016361236572265624, "loss": 0.36300550401210785, "time": 0.6302071332931518, "epoch": 6, "memory": 44120, "step": 22635} +{"lr": 4.3551903519639325e-05, "data_time": 0.0014899253845214843, "loss": 0.3936154991388321, "time": 0.6306586027145386, "epoch": 6, "memory": 44120, "step": 22735} +{"lr": 4.3202340523108894e-05, "data_time": 0.0014934301376342773, "loss": 0.35447017550468446, "time": 0.6331157207489013, "epoch": 6, "memory": 44120, "step": 22835} +{"lr": 4.285354031912302e-05, "data_time": 0.0015354156494140625, "loss": 0.375302791595459, "time": 0.6329831123352051, "epoch": 6, "memory": 44120, "step": 22935} +{"lr": 4.2505525459798544e-05, "data_time": 0.0016644954681396484, "loss": 0.3989559143781662, "time": 0.6303085088729858, "epoch": 6, "memory": 44120, "step": 23035} +{"lr": 4.2158318446474836e-05, "data_time": 0.0019134521484375, "loss": 0.3641472548246384, "time": 0.6320447444915771, "epoch": 6, "memory": 44120, "step": 23135} +{"lr": 4.181194172825905e-05, "data_time": 0.0017750024795532226, "loss": 0.35941425561904905, "time": 0.6305461406707764, "epoch": 6, "memory": 44120, "step": 23235} +{"lr": 4.146641770057448e-05, "data_time": 0.001417231559753418, "loss": 0.3691575676202774, "time": 0.6326620817184448, "epoch": 6, "memory": 44120, "step": 23335} +{"lr": 4.112176870371267e-05, "data_time": 0.001457047462463379, "loss": 0.3825172454118729, "time": 0.6309915781021118, "epoch": 6, "memory": 44120, "step": 23435} +{"accuracy/top1": 74.76079559326172, "data_time": 0.03613086404471562, "time": 1.2846617986415994, "step": 6} +{"lr": 4.0753988582732786e-05, "data_time": 0.0017429351806640624, "loss": 0.3907084405422211, "time": 0.6305223941802979, "epoch": 7, "memory": 44120, "step": 23542} +{"lr": 4.0411221639286815e-05, "data_time": 0.0019685506820678713, "loss": 0.3553955316543579, "time": 0.6326249837875366, "epoch": 7, "memory": 44120, "step": 23642} +{"lr": 4.006939795169518e-05, "data_time": 0.002959156036376953, "loss": 0.38052708506584165, "time": 0.6349667549133301, "epoch": 7, "memory": 44120, "step": 23742} +{"lr": 3.972853962099913e-05, "data_time": 0.003167867660522461, "loss": 0.39260380566120145, "time": 0.6354622840881348, "epoch": 7, "memory": 44120, "step": 23842} +{"lr": 3.93886686858237e-05, "data_time": 0.0027597427368164064, "loss": 0.3540808379650116, "time": 0.6363731861114502, "epoch": 7, "memory": 44120, "step": 23942} +{"lr": 3.904980712095259e-05, "data_time": 0.0018626689910888673, "loss": 0.366429203748703, "time": 0.6304872035980225, "epoch": 7, "memory": 44120, "step": 24042} +{"lr": 3.871197683590748e-05, "data_time": 0.0014447450637817382, "loss": 0.35341583490371703, "time": 0.6317300081253052, "epoch": 7, "memory": 44120, "step": 24142} +{"lr": 3.837519967353125e-05, "data_time": 0.0017804622650146485, "loss": 0.3656893610954285, "time": 0.6323260545730591, "epoch": 7, "memory": 44120, "step": 24242} +{"lr": 3.8039497408575855e-05, "data_time": 0.0018790006637573243, "loss": 0.3789824068546295, "time": 0.6302371263504029, "epoch": 7, "memory": 44120, "step": 24342} +{"lr": 3.770489174629432e-05, "data_time": 0.0015023231506347656, "loss": 0.3829865217208862, "time": 0.6311488628387452, "epoch": 7, "memory": 44120, "step": 24442} +{"lr": 3.737140432103754e-05, "data_time": 0.001765608787536621, "loss": 0.3756262481212616, "time": 0.6312336444854736, "epoch": 7, "memory": 44120, "step": 24542} +{"lr": 3.703905669485545e-05, "data_time": 0.00162050724029541, "loss": 0.3716110110282898, "time": 0.6304387331008912, "epoch": 7, "memory": 44120, "step": 24642} +{"lr": 3.6707870356102706e-05, "data_time": 0.001439046859741211, "loss": 0.35791911482810973, "time": 0.630659818649292, "epoch": 7, "memory": 44120, "step": 24742} +{"lr": 3.6377866718049484e-05, "data_time": 0.0018537044525146484, "loss": 0.4020693838596344, "time": 0.6335522890090942, "epoch": 7, "memory": 44120, "step": 24842} +{"lr": 3.6049067117496965e-05, "data_time": 0.0021165132522583006, "loss": 0.38328524231910704, "time": 0.6413475513458252, "epoch": 7, "memory": 44120, "step": 24942} +{"lr": 3.5721492813397755e-05, "data_time": 0.0021758317947387696, "loss": 0.3705703943967819, "time": 0.6323210716247558, "epoch": 7, "memory": 44120, "step": 25042} +{"lr": 3.539516498548151e-05, "data_time": 0.0016237020492553711, "loss": 0.353135484457016, "time": 0.6329949378967286, "epoch": 7, "memory": 44120, "step": 25142} +{"lr": 3.507010473288518e-05, "data_time": 0.001582765579223633, "loss": 0.3761199861764908, "time": 0.6315385103225708, "epoch": 7, "memory": 44120, "step": 25242} +{"lr": 3.474633307278909e-05, "data_time": 0.0016865968704223634, "loss": 0.38100827038288115, "time": 0.6311712265014648, "epoch": 7, "memory": 44120, "step": 25342} +{"lr": 3.4423870939058177e-05, "data_time": 0.001741790771484375, "loss": 0.4142740875482559, "time": 0.631229019165039, "epoch": 7, "memory": 44120, "step": 25442} +{"lr": 3.410273918088805e-05, "data_time": 0.001669621467590332, "loss": 0.3732988625764847, "time": 0.6314670324325562, "epoch": 7, "memory": 44120, "step": 25542} +{"lr": 3.37829585614574e-05, "data_time": 0.0022763490676879885, "loss": 0.3778089553117752, "time": 0.6351943254470825, "epoch": 7, "memory": 44120, "step": 25642} +{"lr": 3.3464549756585216e-05, "data_time": 0.0018778562545776368, "loss": 0.3983409285545349, "time": 0.6337371349334717, "epoch": 7, "memory": 44120, "step": 25742} +{"lr": 3.314753335339416e-05, "data_time": 0.0019174337387084961, "loss": 0.3413211554288864, "time": 0.6314706802368164, "epoch": 7, "memory": 44120, "step": 25842} +{"lr": 3.2831929848979394e-05, "data_time": 0.001600360870361328, "loss": 0.3760252594947815, "time": 0.6325624704360961, "epoch": 7, "memory": 44120, "step": 25942} +{"lr": 3.251775964908332e-05, "data_time": 0.0016645431518554688, "loss": 0.3815203607082367, "time": 0.6311161756515503, "epoch": 7, "memory": 44120, "step": 26042} +{"lr": 3.22050430667762e-05, "data_time": 0.0017194509506225585, "loss": 0.3696031779050827, "time": 0.6311853408813477, "epoch": 7, "memory": 44120, "step": 26142} +{"lr": 3.189380032114285e-05, "data_time": 0.0017520666122436523, "loss": 0.3679883062839508, "time": 0.6324081182479858, "epoch": 7, "memory": 44120, "step": 26242} +{"lr": 3.1584051535975324e-05, "data_time": 0.0015215873718261719, "loss": 0.3825375556945801, "time": 0.6307426929473877, "epoch": 7, "memory": 44120, "step": 26342} +{"lr": 3.127581673847175e-05, "data_time": 0.001592254638671875, "loss": 0.36102774143218996, "time": 0.6308587789535522, "epoch": 7, "memory": 44120, "step": 26442} +{"lr": 3.0969115857941555e-05, "data_time": 0.0016212940216064453, "loss": 0.367472892999649, "time": 0.6305755615234375, "epoch": 7, "memory": 44120, "step": 26542} +{"lr": 3.0663968724516654e-05, "data_time": 0.001456308364868164, "loss": 0.3704697281122208, "time": 0.6312161445617676, "epoch": 7, "memory": 44120, "step": 26642} +{"lr": 3.0360395067869668e-05, "data_time": 0.0019632577896118164, "loss": 0.3739717960357666, "time": 0.6323014259338379, "epoch": 7, "memory": 44120, "step": 26742} +{"lr": 3.0058414515938063e-05, "data_time": 0.0016052961349487306, "loss": 0.3671251147985458, "time": 0.6343223094940186, "epoch": 7, "memory": 44120, "step": 26842} +{"lr": 2.975804659365501e-05, "data_time": 0.0019020795822143554, "loss": 0.37165725231170654, "time": 0.6316561698913574, "epoch": 7, "memory": 44120, "step": 26942} +{"lr": 2.945931072168714e-05, "data_time": 0.001604461669921875, "loss": 0.3721301585435867, "time": 0.6338523864746094, "epoch": 7, "memory": 44120, "step": 27042} +{"lr": 2.9162226215178855e-05, "data_time": 0.0016022920608520508, "loss": 0.3967426270246506, "time": 0.6302759647369385, "epoch": 7, "memory": 44120, "step": 27142} +{"lr": 2.8866812282503322e-05, "data_time": 0.0017762184143066406, "loss": 0.3843903779983521, "time": 0.632037878036499, "epoch": 7, "memory": 44120, "step": 27242} +{"lr": 2.8573088024020838e-05, "data_time": 0.0015083789825439454, "loss": 0.35235882699489596, "time": 0.6293889284133911, "epoch": 7, "memory": 44120, "step": 27342} +{"accuracy/top1": 75.0356674194336, "data_time": 0.0317530632019043, "time": 1.2813783226342037, "step": 7} +{"lr": 2.8260695817390092e-05, "data_time": 0.0018131256103515625, "loss": 0.3888059794902802, "time": 0.6305115222930908, "epoch": 8, "memory": 44120, "step": 27449} +{"lr": 2.7970529402570967e-05, "data_time": 0.0014161348342895507, "loss": 0.3569154739379883, "time": 0.6301471710205078, "epoch": 8, "memory": 44120, "step": 27549} +{"lr": 2.7682110612244214e-05, "data_time": 0.0015307426452636718, "loss": 0.3666468858718872, "time": 0.6308928728103638, "epoch": 8, "memory": 44120, "step": 27649} +{"lr": 2.7395458094490283e-05, "data_time": 0.001904463768005371, "loss": 0.3775034248828888, "time": 0.6302277326583863, "epoch": 8, "memory": 44120, "step": 27749} +{"lr": 2.7110590383189015e-05, "data_time": 0.0016535282135009765, "loss": 0.39016111493110656, "time": 0.6319559812545776, "epoch": 8, "memory": 44120, "step": 27849} +{"lr": 2.6827525896821397e-05, "data_time": 0.0020310163497924806, "loss": 0.35462755560874937, "time": 0.6325171232223511, "epoch": 8, "memory": 44120, "step": 27949} +{"lr": 2.6546282937278552e-05, "data_time": 0.001477217674255371, "loss": 0.36985248029232026, "time": 0.6309507846832275, "epoch": 8, "memory": 44120, "step": 28049} +{"lr": 2.6266879688678606e-05, "data_time": 0.002318263053894043, "loss": 0.33672429621219635, "time": 0.633042287826538, "epoch": 8, "memory": 44120, "step": 28149} +{"lr": 2.5989334216190738e-05, "data_time": 0.001660633087158203, "loss": 0.3577429085969925, "time": 0.6312180280685424, "epoch": 8, "memory": 44120, "step": 28249} +{"lr": 2.571366446486732e-05, "data_time": 0.001709890365600586, "loss": 0.3706617772579193, "time": 0.6319145679473877, "epoch": 8, "memory": 44120, "step": 28349} +{"lr": 2.5439888258483665e-05, "data_time": 0.001877593994140625, "loss": 0.3651845157146454, "time": 0.6331820726394654, "epoch": 8, "memory": 44120, "step": 28449} +{"lr": 2.5168023298385384e-05, "data_time": 0.001558208465576172, "loss": 0.3963282972574234, "time": 0.6308655977249146, "epoch": 8, "memory": 44120, "step": 28549} +{"lr": 2.489808716234412e-05, "data_time": 0.0017090320587158203, "loss": 0.3455282747745514, "time": 0.6313030242919921, "epoch": 8, "memory": 44120, "step": 28649} +{"lr": 2.463009730342111e-05, "data_time": 0.00146026611328125, "loss": 0.35659017860889436, "time": 0.6301676034927368, "epoch": 8, "memory": 44120, "step": 28749} +{"lr": 2.436407104883849e-05, "data_time": 0.0017405271530151368, "loss": 0.3557533144950867, "time": 0.6315641403198242, "epoch": 8, "memory": 44120, "step": 28849} +{"lr": 2.4100025598859092e-05, "data_time": 0.001744532585144043, "loss": 0.4059371381998062, "time": 0.6453279256820679, "epoch": 8, "memory": 44120, "step": 28949} +{"lr": 2.3837978025674284e-05, "data_time": 0.002048492431640625, "loss": 0.3632131278514862, "time": 0.6317559003829956, "epoch": 8, "memory": 44120, "step": 29049} +{"lr": 2.357794527230027e-05, "data_time": 0.0016649484634399414, "loss": 0.3726804792881012, "time": 0.6326836824417115, "epoch": 8, "memory": 44120, "step": 29149} +{"lr": 2.33199441514825e-05, "data_time": 0.0016494512557983399, "loss": 0.38908624947071074, "time": 0.6314280271530152, "epoch": 8, "memory": 44120, "step": 29249} +{"lr": 2.306399134460859e-05, "data_time": 0.001507735252380371, "loss": 0.3706782728433609, "time": 0.6342707395553588, "epoch": 8, "memory": 44120, "step": 29349} +{"lr": 2.281010340062995e-05, "data_time": 0.0015233278274536133, "loss": 0.3851109206676483, "time": 0.631217622756958, "epoch": 8, "memory": 44120, "step": 29449} +{"lr": 2.2558296734991603e-05, "data_time": 0.0015987157821655273, "loss": 0.37804694175720216, "time": 0.6335761308670044, "epoch": 8, "memory": 44120, "step": 29549} +{"lr": 2.2308587628570915e-05, "data_time": 0.0020048856735229493, "loss": 0.3452103555202484, "time": 0.6310968637466431, "epoch": 8, "memory": 44120, "step": 29649} +{"lr": 2.2060992226624955e-05, "data_time": 0.0015798091888427734, "loss": 0.39766121804714205, "time": 0.630724573135376, "epoch": 8, "memory": 44120, "step": 29749} +{"lr": 2.181552653774648e-05, "data_time": 0.0015205144882202148, "loss": 0.3869244307279587, "time": 0.6324070930480957, "epoch": 8, "memory": 44120, "step": 29849} +{"lr": 2.1572206432829035e-05, "data_time": 0.0015801668167114257, "loss": 0.36472782492637634, "time": 0.6330604553222656, "epoch": 8, "memory": 44120, "step": 29949} +{"lr": 2.1331047644040754e-05, "data_time": 0.0014296770095825195, "loss": 0.3893174022436142, "time": 0.6306674242019653, "epoch": 8, "memory": 44120, "step": 30049} +{"lr": 2.109206576380705e-05, "data_time": 0.001616525650024414, "loss": 0.37757771015167235, "time": 0.6316470146179199, "epoch": 8, "memory": 44120, "step": 30149} +{"lr": 2.0855276243802687e-05, "data_time": 0.001432943344116211, "loss": 0.36857550144195556, "time": 0.6312754392623902, "epoch": 8, "memory": 44120, "step": 30249} +{"lr": 2.0620694393952487e-05, "data_time": 0.001424098014831543, "loss": 0.369435515999794, "time": 0.6353442668914795, "epoch": 8, "memory": 44120, "step": 30349} +{"lr": 2.038833538144169e-05, "data_time": 0.0013814926147460937, "loss": 0.40329310297966003, "time": 0.631174373626709, "epoch": 8, "memory": 44120, "step": 30449} +{"lr": 2.0158214229735154e-05, "data_time": 0.001552748680114746, "loss": 0.36263504028320315, "time": 0.6297932863235474, "epoch": 8, "memory": 44120, "step": 30549} +{"lr": 1.9930345817605977e-05, "data_time": 0.0015177011489868164, "loss": 0.3531788408756256, "time": 0.6324968576431275, "epoch": 8, "memory": 44120, "step": 30649} +{"lr": 1.970474487817358e-05, "data_time": 0.001619577407836914, "loss": 0.3608382433652878, "time": 0.6316331624984741, "epoch": 8, "memory": 44120, "step": 30749} +{"lr": 1.9481425997951092e-05, "data_time": 0.0014397621154785157, "loss": 0.37489425837993623, "time": 0.6312134981155395, "epoch": 8, "memory": 44120, "step": 30849} +{"lr": 1.9260403615902187e-05, "data_time": 0.0019301176071166992, "loss": 0.3705161720514297, "time": 0.6321103572845459, "epoch": 8, "memory": 44120, "step": 30949} +{"lr": 1.9041692022507647e-05, "data_time": 0.0018164873123168944, "loss": 0.36591207385063174, "time": 0.6339050769805908, "epoch": 8, "memory": 44120, "step": 31049} +{"lr": 1.8825305358841125e-05, "data_time": 0.001504969596862793, "loss": 0.3680920213460922, "time": 0.6305970668792724, "epoch": 8, "memory": 44120, "step": 31149} +{"lr": 1.861125761565518e-05, "data_time": 0.001303553581237793, "loss": 0.349801230430603, "time": 0.6308927059173584, "epoch": 8, "memory": 44120, "step": 31249} +{"accuracy/top1": 75.31748962402344, "data_time": 0.03823984491414037, "time": 1.2858252689756196, "step": 8} +{"lr": 1.8384832449122864e-05, "data_time": 0.001545119285583496, "loss": 0.36703485548496245, "time": 0.6299569606781006, "epoch": 9, "memory": 44120, "step": 31356} +{"lr": 1.817567007332114e-05, "data_time": 0.0015564203262329102, "loss": 0.3762494921684265, "time": 0.630024528503418, "epoch": 9, "memory": 44120, "step": 31456} +{"lr": 1.796888862098825e-05, "data_time": 0.0017134189605712891, "loss": 0.3761904388666153, "time": 0.6309181690216065, "epoch": 9, "memory": 44120, "step": 31556} +{"lr": 1.7764501461839678e-05, "data_time": 0.002408552169799805, "loss": 0.3647339463233948, "time": 0.63305504322052, "epoch": 9, "memory": 44120, "step": 31656} +{"lr": 1.7562521810784746e-05, "data_time": 0.0017658233642578124, "loss": 0.35223880410194397, "time": 0.6322914361953735, "epoch": 9, "memory": 44120, "step": 31756} +{"lr": 1.7362962727072444e-05, "data_time": 0.0019680261611938477, "loss": 0.35907942354679107, "time": 0.6366469383239746, "epoch": 9, "memory": 44120, "step": 31856} +{"lr": 1.716583711344682e-05, "data_time": 0.001625204086303711, "loss": 0.34919698536396027, "time": 0.6315120220184326, "epoch": 9, "memory": 44120, "step": 31956} +{"lr": 1.6971157715312954e-05, "data_time": 0.0019192457199096679, "loss": 0.3772891044616699, "time": 0.6313671827316284, "epoch": 9, "memory": 44120, "step": 32056} +{"lr": 1.677893711991274e-05, "data_time": 0.0015111446380615234, "loss": 0.3667439311742783, "time": 0.6309309244155884, "epoch": 9, "memory": 44120, "step": 32156} +{"lr": 1.6589187755511025e-05, "data_time": 0.001652359962463379, "loss": 0.376569926738739, "time": 0.6314516305923462, "epoch": 9, "memory": 44120, "step": 32256} +{"lr": 1.640192189059216e-05, "data_time": 0.0016833066940307618, "loss": 0.35952570140361784, "time": 0.6338250637054443, "epoch": 9, "memory": 44120, "step": 32356} +{"lr": 1.6217151633066716e-05, "data_time": 0.00162813663482666, "loss": 0.39810044765472413, "time": 0.6326664924621582, "epoch": 9, "memory": 44120, "step": 32456} +{"lr": 1.6034888929488564e-05, "data_time": 0.0018696784973144531, "loss": 0.35402290225028993, "time": 0.631584620475769, "epoch": 9, "memory": 44120, "step": 32556} +{"lr": 1.585514556428258e-05, "data_time": 0.001542377471923828, "loss": 0.3798711508512497, "time": 0.631550407409668, "epoch": 9, "memory": 44120, "step": 32656} +{"lr": 1.567793315898261e-05, "data_time": 0.0014659404754638673, "loss": 0.3942319631576538, "time": 0.6313107728958129, "epoch": 9, "memory": 44120, "step": 32756} +{"lr": 1.5503263171480048e-05, "data_time": 0.001752614974975586, "loss": 0.361508384346962, "time": 0.6313312292098999, "epoch": 9, "memory": 44120, "step": 32856} +{"lr": 1.5331146895283165e-05, "data_time": 0.001412200927734375, "loss": 0.38959822356700896, "time": 0.6342740535736084, "epoch": 9, "memory": 44120, "step": 32956} +{"lr": 1.5161595458786728e-05, "data_time": 0.0015943527221679687, "loss": 0.38137592375278473, "time": 0.6306544542312622, "epoch": 9, "memory": 44120, "step": 33056} +{"lr": 1.4994619824552563e-05, "data_time": 0.0015293359756469727, "loss": 0.39286218881607055, "time": 0.6311569929122924, "epoch": 9, "memory": 44120, "step": 33156} +{"lr": 1.4830230788600802e-05, "data_time": 0.0017734050750732421, "loss": 0.37159693241119385, "time": 0.6308529615402222, "epoch": 9, "memory": 44120, "step": 33256} +{"lr": 1.4668438979711779e-05, "data_time": 0.0017318487167358398, "loss": 0.3421124368906021, "time": 0.6308297634124755, "epoch": 9, "memory": 44120, "step": 33356} +{"lr": 1.4509254858738844e-05, "data_time": 0.0015690088272094726, "loss": 0.34994647204875945, "time": 0.6317381381988525, "epoch": 9, "memory": 44120, "step": 33456} +{"lr": 1.4352688717931966e-05, "data_time": 0.0016712427139282226, "loss": 0.3561157017946243, "time": 0.6310806512832642, "epoch": 9, "memory": 44120, "step": 33556} +{"lr": 1.4198750680272348e-05, "data_time": 0.001669001579284668, "loss": 0.4037459522485733, "time": 0.6342191457748413, "epoch": 9, "memory": 44120, "step": 33656} +{"lr": 1.4047450698817857e-05, "data_time": 0.0016637802124023437, "loss": 0.37333865761756896, "time": 0.6314516067504883, "epoch": 9, "memory": 44120, "step": 33756} +{"lr": 1.389879855605952e-05, "data_time": 0.0018546342849731444, "loss": 0.3626085966825485, "time": 0.6326472043991089, "epoch": 9, "memory": 44120, "step": 33856} +{"lr": 1.3752803863288976e-05, "data_time": 0.0015349149703979491, "loss": 0.3698987692594528, "time": 0.6308177471160888, "epoch": 9, "memory": 44120, "step": 33956} +{"lr": 1.3609476059977129e-05, "data_time": 0.0017613649368286132, "loss": 0.36564660370349883, "time": 0.6302955389022827, "epoch": 9, "memory": 44120, "step": 34056} +{"lr": 1.3468824413163754e-05, "data_time": 0.0016710281372070313, "loss": 0.3944234699010849, "time": 0.6306342840194702, "epoch": 9, "memory": 44120, "step": 34156} +{"lr": 1.3330858016858395e-05, "data_time": 0.0016077041625976562, "loss": 0.37970112562179564, "time": 0.6308547019958496, "epoch": 9, "memory": 44120, "step": 34256} +{"lr": 1.3195585791452309e-05, "data_time": 0.0015581846237182617, "loss": 0.32868294715881347, "time": 0.6309695243835449, "epoch": 9, "memory": 44120, "step": 34356} +{"lr": 1.3063016483141729e-05, "data_time": 0.0017665863037109376, "loss": 0.35830931067466737, "time": 0.6328487873077393, "epoch": 9, "memory": 44120, "step": 34456} +{"lr": 1.2933158663362388e-05, "data_time": 0.0016797780990600586, "loss": 0.3624483227729797, "time": 0.6322095394134521, "epoch": 9, "memory": 44120, "step": 34556} +{"lr": 1.2806020728235311e-05, "data_time": 0.001705312728881836, "loss": 0.3889764338731766, "time": 0.6311343431472778, "epoch": 9, "memory": 44120, "step": 34656} +{"lr": 1.2681610898023963e-05, "data_time": 0.0015658140182495117, "loss": 0.3814141720533371, "time": 0.6311052799224853, "epoch": 9, "memory": 44120, "step": 34756} +{"lr": 1.2559937216602731e-05, "data_time": 0.0016845226287841796, "loss": 0.36632585525512695, "time": 0.6318511247634888, "epoch": 9, "memory": 44120, "step": 34856} +{"lr": 1.2441007550936847e-05, "data_time": 0.002275562286376953, "loss": 0.3614902436733246, "time": 0.631909704208374, "epoch": 9, "memory": 44120, "step": 34956} +{"lr": 1.232482959057377e-05, "data_time": 0.0016462087631225586, "loss": 0.3735176056623459, "time": 0.6342405319213867, "epoch": 9, "memory": 44120, "step": 35056} +{"lr": 1.2211410847145969e-05, "data_time": 0.0012430667877197266, "loss": 0.4131546676158905, "time": 0.630808162689209, "epoch": 9, "memory": 44120, "step": 35156} +{"accuracy/top1": 75.6375961303711, "data_time": 0.03174326337617019, "time": 1.279104837055864, "step": 9} +{"lr": 1.209311679360931e-05, "data_time": 0.0017336606979370117, "loss": 0.35722833275794985, "time": 0.6304180383682251, "epoch": 10, "memory": 44120, "step": 35263} +{"lr": 1.1985432727488666e-05, "data_time": 0.0016126632690429688, "loss": 0.3765010148286819, "time": 0.6306754350662231, "epoch": 10, "memory": 44120, "step": 35363} +{"lr": 1.1880529822435289e-05, "data_time": 0.0016975641250610352, "loss": 0.36871753036975863, "time": 0.6311088800430298, "epoch": 10, "memory": 44120, "step": 35463} +{"lr": 1.1778414861078726e-05, "data_time": 0.0015508174896240235, "loss": 0.3558087289333344, "time": 0.6301191568374633, "epoch": 10, "memory": 44120, "step": 35563} +{"lr": 1.1679094445790538e-05, "data_time": 0.001487874984741211, "loss": 0.38038043677806854, "time": 0.6314319849014283, "epoch": 10, "memory": 44120, "step": 35663} +{"lr": 1.1582574998257377e-05, "data_time": 0.0016562461853027344, "loss": 0.35182705521583557, "time": 0.6309397697448731, "epoch": 10, "memory": 44120, "step": 35763} +{"lr": 1.1488862759065762e-05, "data_time": 0.00214688777923584, "loss": 0.3624135583639145, "time": 0.6328227281570434, "epoch": 10, "memory": 44120, "step": 35863} +{"lr": 1.1397963787298665e-05, "data_time": 0.0015983104705810547, "loss": 0.3868138015270233, "time": 0.6307113409042359, "epoch": 10, "memory": 44120, "step": 35963} +{"lr": 1.1309883960143664e-05, "data_time": 0.001660299301147461, "loss": 0.3741405069828033, "time": 0.63277268409729, "epoch": 10, "memory": 44120, "step": 36063} +{"lr": 1.122462897251302e-05, "data_time": 0.0015828371047973632, "loss": 0.34660682678222654, "time": 0.630481481552124, "epoch": 10, "memory": 44120, "step": 36163} +{"lr": 1.114220433667542e-05, "data_time": 0.0017837762832641601, "loss": 0.36329991519451144, "time": 0.6305551052093505, "epoch": 10, "memory": 44120, "step": 36263} +{"lr": 1.10626153818996e-05, "data_time": 0.0016129016876220703, "loss": 0.3634851574897766, "time": 0.6304593086242676, "epoch": 10, "memory": 44120, "step": 36363} +{"lr": 1.0985867254109761e-05, "data_time": 0.0016503572463989259, "loss": 0.35591005384922025, "time": 0.6310620307922363, "epoch": 10, "memory": 44120, "step": 36463} +{"lr": 1.0911964915552823e-05, "data_time": 0.0015620708465576172, "loss": 0.3781615048646927, "time": 0.6342061042785645, "epoch": 10, "memory": 44120, "step": 36563} +{"lr": 1.0840913144477643e-05, "data_time": 0.0022008419036865234, "loss": 0.36848844289779664, "time": 0.6316298246383667, "epoch": 10, "memory": 44120, "step": 36663} +{"lr": 1.0772716534826064e-05, "data_time": 0.0016360998153686524, "loss": 0.36894028782844546, "time": 0.6309942960739136, "epoch": 10, "memory": 44120, "step": 36763} +{"lr": 1.0707379495935828e-05, "data_time": 0.0014685630798339845, "loss": 0.3840280413627625, "time": 0.6303502321243286, "epoch": 10, "memory": 44120, "step": 36863} +{"lr": 1.0644906252255541e-05, "data_time": 0.0015579462051391602, "loss": 0.35381980836391447, "time": 0.6308066368103027, "epoch": 10, "memory": 44120, "step": 36963} +{"lr": 1.0585300843071527e-05, "data_time": 0.0015779972076416016, "loss": 0.3881597608327866, "time": 0.6311758279800415, "epoch": 10, "memory": 44120, "step": 37063} +{"lr": 1.0528567122246654e-05, "data_time": 0.0017542123794555664, "loss": 0.36496404111385344, "time": 0.6356167078018189, "epoch": 10, "memory": 44120, "step": 37163} +{"lr": 1.0474708757971141e-05, "data_time": 0.0016975641250610352, "loss": 0.36717942655086516, "time": 0.6304542541503906, "epoch": 10, "memory": 44120, "step": 37263} +{"lr": 1.042372923252543e-05, "data_time": 0.0015410661697387695, "loss": 0.3778563469648361, "time": 0.6327170372009278, "epoch": 10, "memory": 44120, "step": 37363} +{"lr": 1.0375631842054998e-05, "data_time": 0.0016430139541625977, "loss": 0.357115176320076, "time": 0.6307471752166748, "epoch": 10, "memory": 44120, "step": 37463} +{"lr": 1.0330419696357254e-05, "data_time": 0.0017735719680786132, "loss": 0.39231652915477755, "time": 0.6324982166290283, "epoch": 10, "memory": 44120, "step": 37563} +{"lr": 1.028809571868047e-05, "data_time": 0.0017905950546264649, "loss": 0.3452566027641296, "time": 0.633632755279541, "epoch": 10, "memory": 44120, "step": 37663} +{"lr": 1.0248662645534766e-05, "data_time": 0.0016438722610473632, "loss": 0.37007475793361666, "time": 0.6309433221817017, "epoch": 10, "memory": 44120, "step": 37763} +{"lr": 1.0212123026515194e-05, "data_time": 0.0018476724624633789, "loss": 0.35813745856285095, "time": 0.6313892364501953, "epoch": 10, "memory": 44120, "step": 37863} +{"lr": 1.0178479224136881e-05, "data_time": 0.0015517950057983398, "loss": 0.3813371449708939, "time": 0.6305741310119629, "epoch": 10, "memory": 44120, "step": 37963} +{"lr": 1.0147733413682289e-05, "data_time": 0.001695394515991211, "loss": 0.37492952346801756, "time": 0.6366972923278809, "epoch": 10, "memory": 44120, "step": 38063} +{"lr": 1.0119887583060527e-05, "data_time": 0.001740264892578125, "loss": 0.38165668547153475, "time": 0.631599235534668, "epoch": 10, "memory": 44120, "step": 38163} +{"lr": 1.0094943532678871e-05, "data_time": 0.0016669750213623047, "loss": 0.3390590101480484, "time": 0.6314989566802979, "epoch": 10, "memory": 44120, "step": 38263} +{"lr": 1.007290287532637e-05, "data_time": 0.0016634464263916016, "loss": 0.3734966367483139, "time": 0.6409272432327271, "epoch": 10, "memory": 44120, "step": 38363} +{"lr": 1.0053767036069495e-05, "data_time": 0.002427411079406738, "loss": 0.36938982605934145, "time": 0.6355569362640381, "epoch": 10, "memory": 44120, "step": 38463} +{"lr": 1.003753725216005e-05, "data_time": 0.0017248868942260742, "loss": 0.3700664281845093, "time": 0.6433971405029297, "epoch": 10, "memory": 44120, "step": 38563} +{"lr": 1.0024214572955152e-05, "data_time": 0.0016395092010498048, "loss": 0.3659755736589432, "time": 0.6300317764282226, "epoch": 10, "memory": 44120, "step": 38663} +{"lr": 1.0013799859849456e-05, "data_time": 0.00285491943359375, "loss": 0.381336572766304, "time": 0.6342385768890381, "epoch": 10, "memory": 44120, "step": 38763} +{"lr": 1.0006293786219338e-05, "data_time": 0.0022005081176757813, "loss": 0.3606461644172668, "time": 0.632568097114563, "epoch": 10, "memory": 44120, "step": 38863} +{"lr": 1.0001696837379472e-05, "data_time": 0.001591324806213379, "loss": 0.36118145287036896, "time": 0.6330610275268554, "epoch": 10, "memory": 44120, "step": 38963} +{"lr": 1.0000009310551361e-05, "data_time": 0.0012398004531860352, "loss": 0.36670683324337006, "time": 0.6310634851455689, "epoch": 10, "memory": 44120, "step": 39063} +{"accuracy/top1": 75.42535400390625, "data_time": 0.032334233152455295, "time": 1.2790674094496102, "step": 10} diff --git a/clip_large_pretrain_4x256_sdv2_lr1e-4/20230604_015520/vis_data/vis_image/0000000.png_1.png b/clip_large_pretrain_4x256_sdv2_lr1e-4/20230604_015520/vis_data/vis_image/0000000.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..e4e73f93521a238e2276f1e5aa6f4694a45a78fc Binary files /dev/null and b/clip_large_pretrain_4x256_sdv2_lr1e-4/20230604_015520/vis_data/vis_image/0000000.png_1.png differ diff --git a/clip_large_pretrain_4x256_sdv2_lr1e-4/20230604_015520/vis_data/vis_image/0000000.png_10.png b/clip_large_pretrain_4x256_sdv2_lr1e-4/20230604_015520/vis_data/vis_image/0000000.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..7df2c2fa42ae3a88eb98e520da280734ee92c5a6 Binary files /dev/null and b/clip_large_pretrain_4x256_sdv2_lr1e-4/20230604_015520/vis_data/vis_image/0000000.png_10.png differ diff --git a/clip_large_pretrain_4x256_sdv2_lr1e-4/20230604_015520/vis_data/vis_image/0000000.png_2.png b/clip_large_pretrain_4x256_sdv2_lr1e-4/20230604_015520/vis_data/vis_image/0000000.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..c1bb8c4f2ef6bd985083bef3792f716221ec433d Binary files /dev/null and b/clip_large_pretrain_4x256_sdv2_lr1e-4/20230604_015520/vis_data/vis_image/0000000.png_2.png differ diff --git a/clip_large_pretrain_4x256_sdv2_lr1e-4/20230604_015520/vis_data/vis_image/0000000.png_3.png b/clip_large_pretrain_4x256_sdv2_lr1e-4/20230604_015520/vis_data/vis_image/0000000.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..f9bcac1b14ce754b285ad870f80a75c13dd2ccca Binary files /dev/null and b/clip_large_pretrain_4x256_sdv2_lr1e-4/20230604_015520/vis_data/vis_image/0000000.png_3.png differ diff --git a/clip_large_pretrain_4x256_sdv2_lr1e-4/20230604_015520/vis_data/vis_image/0000000.png_4.png b/clip_large_pretrain_4x256_sdv2_lr1e-4/20230604_015520/vis_data/vis_image/0000000.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..4bb1714998be5257a319d550cae7470b2ae12057 Binary files /dev/null and b/clip_large_pretrain_4x256_sdv2_lr1e-4/20230604_015520/vis_data/vis_image/0000000.png_4.png differ diff --git a/clip_large_pretrain_4x256_sdv2_lr1e-4/20230604_015520/vis_data/vis_image/0000000.png_5.png b/clip_large_pretrain_4x256_sdv2_lr1e-4/20230604_015520/vis_data/vis_image/0000000.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..8292e25cc1579c711f42923dc878bcb1320da738 Binary files /dev/null and b/clip_large_pretrain_4x256_sdv2_lr1e-4/20230604_015520/vis_data/vis_image/0000000.png_5.png differ diff --git a/clip_large_pretrain_4x256_sdv2_lr1e-4/20230604_015520/vis_data/vis_image/0000000.png_6.png b/clip_large_pretrain_4x256_sdv2_lr1e-4/20230604_015520/vis_data/vis_image/0000000.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..f9dc982bf1acdd5ae55c22c9cc1f7cdc3a5f6ca7 Binary files /dev/null and b/clip_large_pretrain_4x256_sdv2_lr1e-4/20230604_015520/vis_data/vis_image/0000000.png_6.png differ diff --git a/clip_large_pretrain_4x256_sdv2_lr1e-4/20230604_015520/vis_data/vis_image/0000000.png_7.png b/clip_large_pretrain_4x256_sdv2_lr1e-4/20230604_015520/vis_data/vis_image/0000000.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..cdaec193e2c30dae40cd30c0748898039ea8753d Binary files /dev/null and b/clip_large_pretrain_4x256_sdv2_lr1e-4/20230604_015520/vis_data/vis_image/0000000.png_7.png differ diff --git a/clip_large_pretrain_4x256_sdv2_lr1e-4/20230604_015520/vis_data/vis_image/0000000.png_8.png b/clip_large_pretrain_4x256_sdv2_lr1e-4/20230604_015520/vis_data/vis_image/0000000.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..b6ede1281656fef6433becbef588c56880a520a3 Binary files /dev/null and b/clip_large_pretrain_4x256_sdv2_lr1e-4/20230604_015520/vis_data/vis_image/0000000.png_8.png differ diff --git a/clip_large_pretrain_4x256_sdv2_lr1e-4/20230604_015520/vis_data/vis_image/0000000.png_9.png b/clip_large_pretrain_4x256_sdv2_lr1e-4/20230604_015520/vis_data/vis_image/0000000.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..db07f85209ccb7fecc29b0fa75f7833fad766868 Binary files /dev/null and b/clip_large_pretrain_4x256_sdv2_lr1e-4/20230604_015520/vis_data/vis_image/0000000.png_9.png differ diff --git a/clip_large_pretrain_4x256_sdv2_lr1e-4/20230604_015520/vis_data/vis_image/000005174.jpg_1.png b/clip_large_pretrain_4x256_sdv2_lr1e-4/20230604_015520/vis_data/vis_image/000005174.jpg_1.png new file mode 100644 index 0000000000000000000000000000000000000000..eb7ced20379c8470f151c4039195bb29a01a363d Binary files /dev/null and b/clip_large_pretrain_4x256_sdv2_lr1e-4/20230604_015520/vis_data/vis_image/000005174.jpg_1.png differ diff --git a/clip_large_pretrain_4x256_sdv2_lr1e-4/20230604_015520/vis_data/vis_image/000005174.jpg_10.png b/clip_large_pretrain_4x256_sdv2_lr1e-4/20230604_015520/vis_data/vis_image/000005174.jpg_10.png new file mode 100644 index 0000000000000000000000000000000000000000..8df7a28c972942b8dddf4908c7cb9d5085904498 Binary files /dev/null and b/clip_large_pretrain_4x256_sdv2_lr1e-4/20230604_015520/vis_data/vis_image/000005174.jpg_10.png differ diff --git a/clip_large_pretrain_4x256_sdv2_lr1e-4/20230604_015520/vis_data/vis_image/000005174.jpg_2.png b/clip_large_pretrain_4x256_sdv2_lr1e-4/20230604_015520/vis_data/vis_image/000005174.jpg_2.png new file mode 100644 index 0000000000000000000000000000000000000000..b6bfcf5892c6ce01ea12c4d3eb30fad4e82062aa Binary files /dev/null and b/clip_large_pretrain_4x256_sdv2_lr1e-4/20230604_015520/vis_data/vis_image/000005174.jpg_2.png differ diff --git a/clip_large_pretrain_4x256_sdv2_lr1e-4/20230604_015520/vis_data/vis_image/000005174.jpg_3.png b/clip_large_pretrain_4x256_sdv2_lr1e-4/20230604_015520/vis_data/vis_image/000005174.jpg_3.png new file mode 100644 index 0000000000000000000000000000000000000000..e020d62d8cac662a5257c5601c48d6cf5b8558db Binary files /dev/null and b/clip_large_pretrain_4x256_sdv2_lr1e-4/20230604_015520/vis_data/vis_image/000005174.jpg_3.png differ diff --git a/clip_large_pretrain_4x256_sdv2_lr1e-4/20230604_015520/vis_data/vis_image/000005174.jpg_4.png b/clip_large_pretrain_4x256_sdv2_lr1e-4/20230604_015520/vis_data/vis_image/000005174.jpg_4.png new file mode 100644 index 0000000000000000000000000000000000000000..ae527d05752b69c395534697b84dc7594f895dec Binary files /dev/null and b/clip_large_pretrain_4x256_sdv2_lr1e-4/20230604_015520/vis_data/vis_image/000005174.jpg_4.png differ diff --git a/clip_large_pretrain_4x256_sdv2_lr1e-4/20230604_015520/vis_data/vis_image/000005174.jpg_5.png b/clip_large_pretrain_4x256_sdv2_lr1e-4/20230604_015520/vis_data/vis_image/000005174.jpg_5.png new file mode 100644 index 0000000000000000000000000000000000000000..c6b3061510d0fbb909d847268a292b6f60e7d79e Binary files /dev/null and b/clip_large_pretrain_4x256_sdv2_lr1e-4/20230604_015520/vis_data/vis_image/000005174.jpg_5.png differ diff --git a/clip_large_pretrain_4x256_sdv2_lr1e-4/20230604_015520/vis_data/vis_image/000005174.jpg_6.png b/clip_large_pretrain_4x256_sdv2_lr1e-4/20230604_015520/vis_data/vis_image/000005174.jpg_6.png new file mode 100644 index 0000000000000000000000000000000000000000..5917ea866ef8e8605043fc03934514b910c28cee Binary files /dev/null and b/clip_large_pretrain_4x256_sdv2_lr1e-4/20230604_015520/vis_data/vis_image/000005174.jpg_6.png differ diff --git a/clip_large_pretrain_4x256_sdv2_lr1e-4/20230604_015520/vis_data/vis_image/000005174.jpg_7.png b/clip_large_pretrain_4x256_sdv2_lr1e-4/20230604_015520/vis_data/vis_image/000005174.jpg_7.png new file mode 100644 index 0000000000000000000000000000000000000000..2a8c7aaba92c9199c381991fb2ec75a47094034b Binary files /dev/null and b/clip_large_pretrain_4x256_sdv2_lr1e-4/20230604_015520/vis_data/vis_image/000005174.jpg_7.png differ diff --git a/clip_large_pretrain_4x256_sdv2_lr1e-4/20230604_015520/vis_data/vis_image/000005174.jpg_8.png b/clip_large_pretrain_4x256_sdv2_lr1e-4/20230604_015520/vis_data/vis_image/000005174.jpg_8.png new file mode 100644 index 0000000000000000000000000000000000000000..0fc89e20d3cfb48777c8bb1cb1408c694f731e6f Binary files /dev/null and b/clip_large_pretrain_4x256_sdv2_lr1e-4/20230604_015520/vis_data/vis_image/000005174.jpg_8.png differ diff --git a/clip_large_pretrain_4x256_sdv2_lr1e-4/20230604_015520/vis_data/vis_image/000005174.jpg_9.png b/clip_large_pretrain_4x256_sdv2_lr1e-4/20230604_015520/vis_data/vis_image/000005174.jpg_9.png new file mode 100644 index 0000000000000000000000000000000000000000..494638aff94f46a0d1e235e2bb44b451279056ab Binary files /dev/null and b/clip_large_pretrain_4x256_sdv2_lr1e-4/20230604_015520/vis_data/vis_image/000005174.jpg_9.png differ diff --git a/clip_large_pretrain_4x256_sdv2_lr1e-4/clip_large_pretrain_4x256_sdv2_lr1e-4.py b/clip_large_pretrain_4x256_sdv2_lr1e-4/clip_large_pretrain_4x256_sdv2_lr1e-4.py new file mode 100644 index 0000000000000000000000000000000000000000..1170b849c141e564d7e6619aa3d4313cfde1d220 --- /dev/null +++ b/clip_large_pretrain_4x256_sdv2_lr1e-4/clip_large_pretrain_4x256_sdv2_lr1e-4.py @@ -0,0 +1,227 @@ +optim_wrapper = dict( + optimizer=dict( + type='AdamW', lr=0.0001, weight_decay=0.3, _scope_='mmpretrain'), + paramwise_cfg=dict( + custom_keys=dict({ + '.cls_token': dict(decay_mult=0.0), + '.pos_embed': dict(decay_mult=0.0) + })), + type='AmpOptimWrapper', + dtype='bfloat16', + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=False, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=4096) +model = dict( + type='ImageClassifier', + backbone=dict( + frozen_stages=24, + type='VisionTransformer', + arch='l', + img_size=224, + patch_size=14, + drop_rate=0.1, + pre_norm=True, + init_cfg=dict( + type='Pretrained', + checkpoint='ckpt/openclip-ViT-L-14.pth', + prefix='backbone')), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=1024, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + init_cfg=None), + init_cfg=dict( + type='TruncNormal', layer=['Conv2d', 'Linear'], std=0.02, bias=0.0), + train_cfg=None) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=128, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stablediffusionV2-1-dpmsolver-25-1m.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/cc1m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=128, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV2-1-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=128, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV2-1-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/clip_large_pretrain_4x256_sdv2_lr1e-4' diff --git a/clip_large_pretrain_4x256_sdv2_lr1e-4/epoch_1.pth b/clip_large_pretrain_4x256_sdv2_lr1e-4/epoch_1.pth new file mode 100644 index 0000000000000000000000000000000000000000..d2d59b3a7e308be47746b1557b91453ac4aa220e --- /dev/null +++ b/clip_large_pretrain_4x256_sdv2_lr1e-4/epoch_1.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ac94fc91d5b2efcf8d2e180641adce8a041d94434747695a07f2913836da18ea +size 2426056323 diff --git a/clip_large_pretrain_4x256_sdv2_lr1e-4/epoch_10.pth b/clip_large_pretrain_4x256_sdv2_lr1e-4/epoch_10.pth new file mode 100644 index 0000000000000000000000000000000000000000..fa3f04897133d76042dda0097ef276f9d25b70df --- /dev/null +++ b/clip_large_pretrain_4x256_sdv2_lr1e-4/epoch_10.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:70a776d3808196a9e37ae4006125362986f2bd19f9166a19b8b165e61c31c802 +size 2428623491 diff --git a/clip_large_pretrain_4x256_sdv2_lr1e-4/epoch_2.pth b/clip_large_pretrain_4x256_sdv2_lr1e-4/epoch_2.pth new file mode 100644 index 0000000000000000000000000000000000000000..0c21dbbdb5390809d93fd4524a0cce531c94e13a --- /dev/null +++ b/clip_large_pretrain_4x256_sdv2_lr1e-4/epoch_2.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:324b2d44271f8f572406fee44b39900a6bc4388e4cd0ab342fb50fcb522ec756 +size 2426340611 diff --git a/clip_large_pretrain_4x256_sdv2_lr1e-4/epoch_3.pth b/clip_large_pretrain_4x256_sdv2_lr1e-4/epoch_3.pth new file mode 100644 index 0000000000000000000000000000000000000000..d3b779d64e2f3275040c3cceceb5ba9a79fdc027 --- /dev/null +++ b/clip_large_pretrain_4x256_sdv2_lr1e-4/epoch_3.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:41166f48612f7306a9a9c8019adf10d2d5b13fadec7cf1c6822efbf782b49b0e +size 2426624899 diff --git a/clip_large_pretrain_4x256_sdv2_lr1e-4/epoch_4.pth b/clip_large_pretrain_4x256_sdv2_lr1e-4/epoch_4.pth new file mode 100644 index 0000000000000000000000000000000000000000..c3dde03ffe262ed5ed40930b056df2bb45cf9c61 --- /dev/null +++ b/clip_large_pretrain_4x256_sdv2_lr1e-4/epoch_4.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:583b9a27782618d1588cad49ba731f4d136c8bc7c8d5e0dd45a83b13c54b3613 +size 2426908867 diff --git a/clip_large_pretrain_4x256_sdv2_lr1e-4/epoch_5.pth b/clip_large_pretrain_4x256_sdv2_lr1e-4/epoch_5.pth new file mode 100644 index 0000000000000000000000000000000000000000..61ceacf97ccadc48ede09e2baa5f32f291005d30 --- /dev/null +++ b/clip_large_pretrain_4x256_sdv2_lr1e-4/epoch_5.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:7d00fc75e129565dc1f8f7e34cfb8609b90445f17c569467c760491778ccab4a +size 2427192579 diff --git a/clip_large_pretrain_4x256_sdv2_lr1e-4/epoch_6.pth b/clip_large_pretrain_4x256_sdv2_lr1e-4/epoch_6.pth new file mode 100644 index 0000000000000000000000000000000000000000..1cbd70f06f6227627b9749327628f29ba484fc4d --- /dev/null +++ b/clip_large_pretrain_4x256_sdv2_lr1e-4/epoch_6.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:7d53b66b93a8b60b5f61608a81c8e113bc7d46c1e4b87a6b88cd16e3f664aa22 +size 2427476291 diff --git a/clip_large_pretrain_4x256_sdv2_lr1e-4/epoch_7.pth b/clip_large_pretrain_4x256_sdv2_lr1e-4/epoch_7.pth new file mode 100644 index 0000000000000000000000000000000000000000..7fd7db0029b4feac71a821a2a5f212713f5ca392 --- /dev/null +++ b/clip_large_pretrain_4x256_sdv2_lr1e-4/epoch_7.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:7f55b9a43252ab06ba2e8ccb01da806f5a957207de7b87c388cd68dd814277fe +size 2427760835 diff --git a/clip_large_pretrain_4x256_sdv2_lr1e-4/epoch_8.pth b/clip_large_pretrain_4x256_sdv2_lr1e-4/epoch_8.pth new file mode 100644 index 0000000000000000000000000000000000000000..c95e83836b54a122680f0dfc7c5bfdd37c31bb53 --- /dev/null +++ b/clip_large_pretrain_4x256_sdv2_lr1e-4/epoch_8.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:e0874f7a9865a772a68a140dd2be85db23b306523b4b98977abc282eb13d5ec4 +size 2428048451 diff --git a/clip_large_pretrain_4x256_sdv2_lr1e-4/epoch_9.pth b/clip_large_pretrain_4x256_sdv2_lr1e-4/epoch_9.pth new file mode 100644 index 0000000000000000000000000000000000000000..1147bbbaaf8876aba08f75855285e0d171656698 --- /dev/null +++ b/clip_large_pretrain_4x256_sdv2_lr1e-4/epoch_9.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:59a3429cea1c85a6ca8e622da22001ff47464089d2fec18bd340d9db5c6a0ad4 +size 2428336003 diff --git a/clip_large_pretrain_4x256_sdv2_lr1e-4/last_checkpoint b/clip_large_pretrain_4x256_sdv2_lr1e-4/last_checkpoint new file mode 100644 index 0000000000000000000000000000000000000000..9b3970a87e48f0cacee265b1412197334c53f42f --- /dev/null +++ b/clip_large_pretrain_4x256_sdv2_lr1e-4/last_checkpoint @@ -0,0 +1 @@ +/mnt/petrelfs/luzeyu/workspace/fakebench/mmpretrain/workdir/clip_large_pretrain_4x256_sdv2_lr1e-4/epoch_10.pth \ No newline at end of file diff --git a/clip_large_pretrain_4x256_sdv2_lr1e-4_rub_/20230603_200211/20230603_200211.log b/clip_large_pretrain_4x256_sdv2_lr1e-4_rub_/20230603_200211/20230603_200211.log new file mode 100644 index 0000000000000000000000000000000000000000..49c58215ab142d4ca00690051b5131dc8de9686a --- /dev/null +++ b/clip_large_pretrain_4x256_sdv2_lr1e-4_rub_/20230603_200211/20230603_200211.log @@ -0,0 +1,1620 @@ +2023/06/03 20:02:15 - mmengine - INFO - +------------------------------------------------------------ +System environment: + sys.platform: linux + Python: 3.10.9 (main, Mar 8 2023, 10:47:38) [GCC 11.2.0] + CUDA available: True + numpy_random_seed: 786267796 + GPU 0,1,2,3: NVIDIA A100-SXM4-80GB + CUDA_HOME: /mnt/petrelfs/share/cuda-11.6 + NVCC: Cuda compilation tools, release 11.6, V11.6.124 + GCC: gcc (GCC) 7.5.0 + PyTorch: 1.13.1 + PyTorch compiling details: PyTorch built with: + - GCC 9.3 + - C++ Version: 201402 + - Intel(R) oneAPI Math Kernel Library Version 2021.4-Product Build 20210904 for Intel(R) 64 architecture applications + - Intel(R) MKL-DNN v2.6.0 (Git Hash 52b5f107dd9cf10910aaa19cb47f3abf9b349815) + - OpenMP 201511 (a.k.a. OpenMP 4.5) + - LAPACK is enabled (usually provided by MKL) + - NNPACK is enabled + - CPU capability usage: AVX2 + - CUDA Runtime 11.6 + - NVCC architecture flags: -gencode;arch=compute_37,code=sm_37;-gencode;arch=compute_50,code=sm_50;-gencode;arch=compute_60,code=sm_60;-gencode;arch=compute_61,code=sm_61;-gencode;arch=compute_70,code=sm_70;-gencode;arch=compute_75,code=sm_75;-gencode;arch=compute_80,code=sm_80;-gencode;arch=compute_86,code=sm_86;-gencode;arch=compute_37,code=compute_37 + - CuDNN 8.3.2 (built against CUDA 11.5) + - Magma 2.6.1 + - Build settings: BLAS_INFO=mkl, BUILD_TYPE=Release, CUDA_VERSION=11.6, CUDNN_VERSION=8.3.2, CXX_COMPILER=/opt/rh/devtoolset-9/root/usr/bin/c++, CXX_FLAGS= -fabi-version=11 -Wno-deprecated -fvisibility-inlines-hidden -DUSE_PTHREADPOOL -fopenmp -DNDEBUG -DUSE_KINETO -DUSE_FBGEMM -DUSE_QNNPACK -DUSE_PYTORCH_QNNPACK -DUSE_XNNPACK -DSYMBOLICATE_MOBILE_DEBUG_HANDLE -DEDGE_PROFILER_USE_KINETO -O2 -fPIC -Wno-narrowing -Wall -Wextra -Werror=return-type -Werror=non-virtual-dtor -Wno-missing-field-initializers -Wno-type-limits -Wno-array-bounds -Wno-unknown-pragmas -Wunused-local-typedefs -Wno-unused-parameter -Wno-unused-function -Wno-unused-result -Wno-strict-overflow -Wno-strict-aliasing -Wno-error=deprecated-declarations -Wno-stringop-overflow -Wno-psabi -Wno-error=pedantic -Wno-error=redundant-decls -Wno-error=old-style-cast -fdiagnostics-color=always -faligned-new -Wno-unused-but-set-variable -Wno-maybe-uninitialized -fno-math-errno -fno-trapping-math -Werror=format -Werror=cast-function-type -Wno-stringop-overflow, LAPACK_INFO=mkl, PERF_WITH_AVX=1, PERF_WITH_AVX2=1, PERF_WITH_AVX512=1, TORCH_VERSION=1.13.1, USE_CUDA=ON, USE_CUDNN=ON, USE_EXCEPTION_PTR=1, USE_GFLAGS=OFF, USE_GLOG=OFF, USE_MKL=ON, USE_MKLDNN=ON, USE_MPI=OFF, USE_NCCL=ON, USE_NNPACK=ON, USE_OPENMP=ON, USE_ROCM=OFF, + + TorchVision: 0.14.1 + OpenCV: 4.7.0 + MMEngine: 0.7.3 + +Runtime environment: + cudnn_benchmark: True + mp_cfg: {'mp_start_method': 'fork', 'opencv_num_threads': 0} + dist_cfg: {'backend': 'nccl'} + seed: None + deterministic: False + Distributed launcher: slurm + Distributed training: True + GPU number: 4 +------------------------------------------------------------ + +2023/06/03 20:02:19 - mmengine - INFO - Config: +optim_wrapper = dict( + optimizer=dict( + type='AdamW', lr=0.0001, weight_decay=0.3, _scope_='mmpretrain'), + paramwise_cfg=dict( + custom_keys=dict({ + '.cls_token': dict(decay_mult=0.0), + '.pos_embed': dict(decay_mult=0.0) + })), + type='AmpOptimWrapper', + dtype='bfloat16', + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=False, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=4096) +model = dict( + type='ImageClassifier', + backbone=dict( + frozen_stages=24, + type='VisionTransformer', + arch='l', + img_size=224, + patch_size=14, + drop_rate=0.1, + pre_norm=True, + init_cfg=dict( + type='Pretrained', + checkpoint='ckpt/openclip-ViT-L-14.pth', + prefix='backbone')), + neck=dict(type='CLIPProjection', in_channels=1024, out_channels=768), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=768, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + init_cfg=None), + init_cfg=dict( + type='TruncNormal', layer=['Conv2d', 'Linear'], std=0.02, bias=0.0), + train_cfg=None) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=128, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stablediffusionV2-1-dpmsolver-25-1m.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/cc1m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=128, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV2-1-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=128, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV2-1-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/clip_large_pretrain_4x256_sdv2_lr1e-4' + +2023/06/03 20:02:33 - mmengine - INFO - Hooks will be executed in the following order: +before_run: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_load_checkpoint: +(ABOVE_NORMAL) EMAHook + -------------------- +before_train: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_train_epoch: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(NORMAL ) DistSamplerSeedHook + -------------------- +before_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook + -------------------- +after_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +after_train_epoch: +(NORMAL ) IterTimerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_val_epoch: +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook + -------------------- +before_val_iter: +(NORMAL ) IterTimerHook + -------------------- +after_val_iter: +(NORMAL ) IterTimerHook +(NORMAL ) VisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_val_epoch: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_save_checkpoint: +(ABOVE_NORMAL) EMAHook + -------------------- +after_train: +(VERY_LOW ) CheckpointHook + -------------------- +before_test_epoch: +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook + -------------------- +before_test_iter: +(NORMAL ) IterTimerHook + -------------------- +after_test_iter: +(NORMAL ) IterTimerHook +(NORMAL ) VisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_test_epoch: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_run: +(BELOW_NORMAL) LoggerHook + -------------------- +2023/06/03 20:02:53 - mmengine - INFO - load backbone in model from: ckpt/openclip-ViT-L-14.pth +2023/06/03 20:02:54 - mmengine - WARNING - The model and loaded state dict do not match exactly + +unexpected key in source state_dict: ln2.weight, ln2.bias + +missing keys in source state_dict: ln1.weight, ln1.bias + +Name of parameter - Initialization information + +backbone.cls_token - torch.Size([1, 1, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.pos_embed - torch.Size([1, 257, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.patch_embed.projection.weight - torch.Size([1024, 3, 14, 14]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.0.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.0.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.0.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.0.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.0.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.0.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.0.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.0.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.0.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.0.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.0.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.0.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.1.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.1.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.1.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.1.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.1.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.1.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.1.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.1.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.1.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.1.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.1.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.1.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.2.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.2.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.2.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.2.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.2.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.2.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.2.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.2.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.2.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.2.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.2.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.2.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.3.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.3.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.3.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.3.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.3.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.3.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.3.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.3.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.3.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.3.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.3.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.3.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.4.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.4.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.4.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.4.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.4.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.4.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.4.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.4.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.4.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.4.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.4.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.4.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.5.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.5.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.5.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.5.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.5.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.5.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.5.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.5.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.5.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.5.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.5.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.5.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.6.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.6.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.6.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.6.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.6.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.6.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.6.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.6.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.6.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.6.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.6.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.6.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.7.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.7.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.7.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.7.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.7.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.7.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.7.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.7.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.7.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.7.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.7.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.7.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.8.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.8.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.8.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.8.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.8.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.8.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.8.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.8.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.8.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.8.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.8.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.8.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.9.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.9.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.9.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.9.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.9.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.9.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.9.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.9.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.9.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.9.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.9.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.9.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.10.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.10.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.10.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.10.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.10.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.10.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.10.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.10.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.10.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.10.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.10.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.10.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.11.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.11.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.11.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.11.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.11.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.11.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.11.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.11.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.11.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.11.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.11.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.11.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.12.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.12.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.12.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.12.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.12.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.12.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.12.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.12.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.12.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.12.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.12.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.12.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.13.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.13.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.13.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.13.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.13.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.13.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.13.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.13.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.13.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.13.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.13.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.13.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.14.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.14.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.14.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.14.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.14.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.14.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.14.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.14.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.14.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.14.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.14.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.14.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.15.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.15.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.15.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.15.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.15.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.15.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.15.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.15.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.15.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.15.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.15.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.15.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.16.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.16.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.16.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.16.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.16.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.16.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.16.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.16.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.16.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.16.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.16.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.16.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.17.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.17.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.17.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.17.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.17.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.17.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.17.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.17.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.17.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.17.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.17.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.17.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.18.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.18.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.18.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.18.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.18.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.18.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.18.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.18.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.18.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.18.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.18.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.18.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.19.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.19.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.19.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.19.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.19.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.19.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.19.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.19.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.19.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.19.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.19.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.19.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.20.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.20.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.20.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.20.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.20.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.20.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.20.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.20.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.20.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.20.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.20.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.20.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.21.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.21.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.21.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.21.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.21.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.21.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.21.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.21.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.21.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.21.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.21.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.21.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.22.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.22.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.22.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.22.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.22.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.22.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.22.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.22.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.22.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.22.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.22.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.22.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.23.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.23.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.23.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.23.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.23.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.23.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.23.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.23.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.23.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.23.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.23.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.23.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.pre_norm.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.pre_norm.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.ln1.weight - torch.Size([1024]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.ln1.bias - torch.Size([1024]): +The value is the same before and after calling `init_weights` of ImageClassifier + +neck.proj - torch.Size([1024, 768]): +The value is the same before and after calling `init_weights` of ImageClassifier + +head.fc.weight - torch.Size([2, 768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +head.fc.bias - torch.Size([2]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 +2023/06/03 20:02:54 - mmengine - WARNING - "FileClient" will be deprecated in future. Please use io functions in https://mmengine.readthedocs.io/en/latest/api/fileio.html#file-io +2023/06/03 20:02:54 - mmengine - WARNING - "HardDiskBackend" is the alias of "LocalBackend" and the former will be deprecated in future. +2023/06/03 20:02:54 - mmengine - INFO - Checkpoints will be saved to /mnt/petrelfs/luzeyu/workspace/fakebench/mmpretrain/workdir/clip_large_pretrain_4x256_sdv2_lr1e-4. +2023/06/03 20:04:04 - mmengine - INFO - Epoch(train) [1][ 100/3907] lr: 9.9999e-05 eta: 7:30:59 time: 0.6304 data_time: 0.0021 memory: 44139 loss: 0.5255 +2023/06/03 20:05:07 - mmengine - INFO - Epoch(train) [1][ 200/3907] lr: 9.9994e-05 eta: 7:09:36 time: 0.6309 data_time: 0.0016 memory: 44139 loss: 0.4959 +2023/06/03 20:06:10 - mmengine - INFO - Epoch(train) [1][ 300/3907] lr: 9.9987e-05 eta: 7:02:04 time: 0.6312 data_time: 0.0017 memory: 44139 loss: 0.4742 +2023/06/03 20:07:14 - mmengine - INFO - Epoch(train) [1][ 400/3907] lr: 9.9977e-05 eta: 6:57:25 time: 0.6305 data_time: 0.0017 memory: 44139 loss: 0.4523 +2023/06/03 20:08:17 - mmengine - INFO - Epoch(train) [1][ 500/3907] lr: 9.9964e-05 eta: 6:54:11 time: 0.6324 data_time: 0.0017 memory: 44139 loss: 0.4552 +2023/06/03 20:09:20 - mmengine - INFO - Epoch(train) [1][ 600/3907] lr: 9.9948e-05 eta: 6:51:54 time: 0.6456 data_time: 0.0020 memory: 44139 loss: 0.4473 +2023/06/03 20:10:23 - mmengine - INFO - Epoch(train) [1][ 700/3907] lr: 9.9929e-05 eta: 6:49:45 time: 0.6310 data_time: 0.0016 memory: 44139 loss: 0.4602 +2023/06/03 20:11:26 - mmengine - INFO - Epoch(train) [1][ 800/3907] lr: 9.9907e-05 eta: 6:47:52 time: 0.6302 data_time: 0.0025 memory: 44139 loss: 0.4576 +2023/06/03 20:12:29 - mmengine - INFO - Epoch(train) [1][ 900/3907] lr: 9.9882e-05 eta: 6:46:11 time: 0.6299 data_time: 0.0016 memory: 44139 loss: 0.4488 +2023/06/03 20:13:32 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230603_200211 +2023/06/03 20:13:32 - mmengine - INFO - Epoch(train) [1][1000/3907] lr: 9.9855e-05 eta: 6:44:38 time: 0.6338 data_time: 0.0018 memory: 44139 loss: 0.4588 +2023/06/03 20:14:35 - mmengine - INFO - Epoch(train) [1][1100/3907] lr: 9.9824e-05 eta: 6:43:09 time: 0.6302 data_time: 0.0015 memory: 44139 loss: 0.4398 +2023/06/03 20:15:38 - mmengine - INFO - Epoch(train) [1][1200/3907] lr: 9.9791e-05 eta: 6:41:43 time: 0.6293 data_time: 0.0018 memory: 44139 loss: 0.4259 +2023/06/03 20:16:41 - mmengine - INFO - Epoch(train) [1][1300/3907] lr: 9.9755e-05 eta: 6:40:21 time: 0.6309 data_time: 0.0016 memory: 44139 loss: 0.4359 +2023/06/03 20:17:44 - mmengine - INFO - Epoch(train) [1][1400/3907] lr: 9.9716e-05 eta: 6:39:03 time: 0.6302 data_time: 0.0015 memory: 44139 loss: 0.4460 +2023/06/03 20:18:47 - mmengine - INFO - Epoch(train) [1][1500/3907] lr: 9.9674e-05 eta: 6:37:45 time: 0.6295 data_time: 0.0020 memory: 44139 loss: 0.4315 +2023/06/03 20:19:50 - mmengine - INFO - Epoch(train) [1][1600/3907] lr: 9.9629e-05 eta: 6:36:29 time: 0.6294 data_time: 0.0018 memory: 44139 loss: 0.4311 +2023/06/03 20:20:53 - mmengine - INFO - Epoch(train) [1][1700/3907] lr: 9.9581e-05 eta: 6:35:14 time: 0.6304 data_time: 0.0021 memory: 44139 loss: 0.4196 +2023/06/03 20:21:56 - mmengine - INFO - Epoch(train) [1][1800/3907] lr: 9.9530e-05 eta: 6:34:03 time: 0.6298 data_time: 0.0016 memory: 44139 loss: 0.4246 +2023/06/03 20:22:59 - mmengine - INFO - Epoch(train) [1][1900/3907] lr: 9.9476e-05 eta: 6:32:51 time: 0.6293 data_time: 0.0016 memory: 44139 loss: 0.4055 +2023/06/03 20:24:02 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230603_200211 +2023/06/03 20:24:02 - mmengine - INFO - Epoch(train) [1][2000/3907] lr: 9.9420e-05 eta: 6:31:39 time: 0.6292 data_time: 0.0018 memory: 44139 loss: 0.4334 +2023/06/03 20:25:05 - mmengine - INFO - Epoch(train) [1][2100/3907] lr: 9.9361e-05 eta: 6:30:30 time: 0.6301 data_time: 0.0017 memory: 44139 loss: 0.4337 +2023/06/03 20:26:09 - mmengine - INFO - Epoch(train) [1][2200/3907] lr: 9.9298e-05 eta: 6:29:22 time: 0.6305 data_time: 0.0016 memory: 44139 loss: 0.4069 +2023/06/03 20:27:12 - mmengine - INFO - Epoch(train) [1][2300/3907] lr: 9.9233e-05 eta: 6:28:14 time: 0.6303 data_time: 0.0021 memory: 44139 loss: 0.4135 +2023/06/03 20:28:15 - mmengine - INFO - Epoch(train) [1][2400/3907] lr: 9.9165e-05 eta: 6:27:06 time: 0.6308 data_time: 0.0018 memory: 44139 loss: 0.4138 +2023/06/03 20:29:18 - mmengine - INFO - Epoch(train) [1][2500/3907] lr: 9.9095e-05 eta: 6:25:59 time: 0.6300 data_time: 0.0017 memory: 44139 loss: 0.4117 +2023/06/03 20:30:21 - mmengine - INFO - Epoch(train) [1][2600/3907] lr: 9.9021e-05 eta: 6:24:52 time: 0.6318 data_time: 0.0016 memory: 44139 loss: 0.4248 +2023/06/03 20:31:24 - mmengine - INFO - Epoch(train) [1][2700/3907] lr: 9.8944e-05 eta: 6:23:44 time: 0.6298 data_time: 0.0018 memory: 44139 loss: 0.4310 +2023/06/03 20:32:27 - mmengine - INFO - Epoch(train) [1][2800/3907] lr: 9.8865e-05 eta: 6:22:38 time: 0.6302 data_time: 0.0015 memory: 44139 loss: 0.4125 +2023/06/03 20:33:30 - mmengine - INFO - Epoch(train) [1][2900/3907] lr: 9.8783e-05 eta: 6:21:31 time: 0.6295 data_time: 0.0017 memory: 44139 loss: 0.3845 +2023/06/03 20:34:33 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230603_200211 +2023/06/03 20:34:33 - mmengine - INFO - Epoch(train) [1][3000/3907] lr: 9.8698e-05 eta: 6:20:25 time: 0.6312 data_time: 0.0022 memory: 44139 loss: 0.4417 +2023/06/03 20:35:36 - mmengine - INFO - Epoch(train) [1][3100/3907] lr: 9.8610e-05 eta: 6:19:19 time: 0.6298 data_time: 0.0016 memory: 44139 loss: 0.4046 +2023/06/03 20:36:39 - mmengine - INFO - Epoch(train) [1][3200/3907] lr: 9.8519e-05 eta: 6:18:13 time: 0.6302 data_time: 0.0015 memory: 44139 loss: 0.4071 +2023/06/03 20:37:42 - mmengine - INFO - Epoch(train) [1][3300/3907] lr: 9.8426e-05 eta: 6:17:08 time: 0.6304 data_time: 0.0017 memory: 44139 loss: 0.4124 +2023/06/03 20:38:45 - mmengine - INFO - Epoch(train) [1][3400/3907] lr: 9.8330e-05 eta: 6:16:02 time: 0.6294 data_time: 0.0018 memory: 44139 loss: 0.4004 +2023/06/03 20:39:48 - mmengine - INFO - Epoch(train) [1][3500/3907] lr: 9.8231e-05 eta: 6:14:56 time: 0.6309 data_time: 0.0021 memory: 44139 loss: 0.4186 +2023/06/03 20:40:51 - mmengine - INFO - Epoch(train) [1][3600/3907] lr: 9.8129e-05 eta: 6:13:50 time: 0.6299 data_time: 0.0017 memory: 44139 loss: 0.3905 +2023/06/03 20:41:54 - mmengine - INFO - Epoch(train) [1][3700/3907] lr: 9.8024e-05 eta: 6:12:45 time: 0.6293 data_time: 0.0018 memory: 44139 loss: 0.4245 +2023/06/03 20:42:57 - mmengine - INFO - Epoch(train) [1][3800/3907] lr: 9.7917e-05 eta: 6:11:40 time: 0.6294 data_time: 0.0017 memory: 44139 loss: 0.4360 +2023/06/03 20:44:00 - mmengine - INFO - Epoch(train) [1][3900/3907] lr: 9.7806e-05 eta: 6:10:35 time: 0.6291 data_time: 0.0017 memory: 44139 loss: 0.4074 +2023/06/03 20:44:04 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230603_200211 +2023/06/03 20:44:04 - mmengine - INFO - Saving checkpoint at 1 epochs +2023/06/03 20:45:41 - mmengine - INFO - Epoch(val) [1][57/57] accuracy/top1: 78.6577 single-label/precision_classwise: [75.0167465209961, 86.2205581665039] single-label/recall_classwise: [91.8755111694336, 62.42635726928711] single-label/f1-score_classwise: [82.5946273803711, 72.4190673828125] data_time: 0.0428 time: 1.3484 +2023/06/03 20:46:43 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230603_200211 +2023/06/03 20:46:48 - mmengine - INFO - Epoch(train) [2][ 100/3907] lr: 9.7685e-05 eta: 6:09:49 time: 0.6323 data_time: 0.0019 memory: 44139 loss: 0.3870 +2023/06/03 20:47:51 - mmengine - INFO - Epoch(train) [2][ 200/3907] lr: 9.7570e-05 eta: 6:08:45 time: 0.6310 data_time: 0.0019 memory: 44138 loss: 0.3851 +2023/06/03 20:48:54 - mmengine - INFO - Epoch(train) [2][ 300/3907] lr: 9.7451e-05 eta: 6:07:42 time: 0.6309 data_time: 0.0017 memory: 44138 loss: 0.3775 +2023/06/03 20:49:57 - mmengine - INFO - Epoch(train) [2][ 400/3907] lr: 9.7329e-05 eta: 6:06:38 time: 0.6309 data_time: 0.0014 memory: 44138 loss: 0.4018 +2023/06/03 20:51:00 - mmengine - INFO - Epoch(train) [2][ 500/3907] lr: 9.7205e-05 eta: 6:05:33 time: 0.6304 data_time: 0.0017 memory: 44138 loss: 0.4087 +2023/06/03 20:52:04 - mmengine - INFO - Epoch(train) [2][ 600/3907] lr: 9.7078e-05 eta: 6:04:29 time: 0.6331 data_time: 0.0027 memory: 44138 loss: 0.3818 +2023/06/03 20:53:07 - mmengine - INFO - Epoch(train) [2][ 700/3907] lr: 9.6949e-05 eta: 6:03:27 time: 0.6305 data_time: 0.0021 memory: 44138 loss: 0.4072 +2023/06/03 20:54:10 - mmengine - INFO - Epoch(train) [2][ 800/3907] lr: 9.6816e-05 eta: 6:02:22 time: 0.6303 data_time: 0.0017 memory: 44138 loss: 0.3874 +2023/06/03 20:55:13 - mmengine - INFO - Epoch(train) [2][ 900/3907] lr: 9.6681e-05 eta: 6:01:17 time: 0.6310 data_time: 0.0017 memory: 44138 loss: 0.3736 +2023/06/03 20:56:16 - mmengine - INFO - Epoch(train) [2][1000/3907] lr: 9.6544e-05 eta: 6:00:14 time: 0.6312 data_time: 0.0017 memory: 44138 loss: 0.3740 +2023/06/03 20:57:15 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230603_200211 +2023/06/03 20:57:19 - mmengine - INFO - Epoch(train) [2][1100/3907] lr: 9.6403e-05 eta: 5:59:09 time: 0.6307 data_time: 0.0019 memory: 44138 loss: 0.3881 +2023/06/03 20:58:23 - mmengine - INFO - Epoch(train) [2][1200/3907] lr: 9.6260e-05 eta: 5:58:05 time: 0.6308 data_time: 0.0020 memory: 44138 loss: 0.3967 +2023/06/03 20:59:26 - mmengine - INFO - Epoch(train) [2][1300/3907] lr: 9.6114e-05 eta: 5:57:01 time: 0.6299 data_time: 0.0021 memory: 44138 loss: 0.4004 +2023/06/03 21:00:29 - mmengine - INFO - Epoch(train) [2][1400/3907] lr: 9.5966e-05 eta: 5:55:56 time: 0.6303 data_time: 0.0018 memory: 44138 loss: 0.4006 +2023/06/03 21:01:32 - mmengine - INFO - Epoch(train) [2][1500/3907] lr: 9.5815e-05 eta: 5:54:52 time: 0.6306 data_time: 0.0016 memory: 44138 loss: 0.3905 +2023/06/03 21:02:35 - mmengine - INFO - Epoch(train) [2][1600/3907] lr: 9.5661e-05 eta: 5:53:48 time: 0.6308 data_time: 0.0015 memory: 44138 loss: 0.3764 +2023/06/03 21:03:38 - mmengine - INFO - Epoch(train) [2][1700/3907] lr: 9.5505e-05 eta: 5:52:44 time: 0.6313 data_time: 0.0022 memory: 44138 loss: 0.3868 +2023/06/03 21:04:41 - mmengine - INFO - Epoch(train) [2][1800/3907] lr: 9.5346e-05 eta: 5:51:40 time: 0.6302 data_time: 0.0016 memory: 44138 loss: 0.3620 +2023/06/03 21:05:44 - mmengine - INFO - Epoch(train) [2][1900/3907] lr: 9.5184e-05 eta: 5:50:37 time: 0.6325 data_time: 0.0018 memory: 44138 loss: 0.3677 +2023/06/03 21:06:48 - mmengine - INFO - Epoch(train) [2][2000/3907] lr: 9.5020e-05 eta: 5:49:33 time: 0.6313 data_time: 0.0028 memory: 44138 loss: 0.3564 +2023/06/03 21:07:46 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230603_200211 +2023/06/03 21:07:51 - mmengine - INFO - Epoch(train) [2][2100/3907] lr: 9.4854e-05 eta: 5:48:29 time: 0.6307 data_time: 0.0019 memory: 44138 loss: 0.3847 +2023/06/03 21:08:54 - mmengine - INFO - Epoch(train) [2][2200/3907] lr: 9.4684e-05 eta: 5:47:25 time: 0.6314 data_time: 0.0025 memory: 44138 loss: 0.3984 +2023/06/03 21:09:57 - mmengine - INFO - Epoch(train) [2][2300/3907] lr: 9.4512e-05 eta: 5:46:21 time: 0.6316 data_time: 0.0020 memory: 44138 loss: 0.3816 +2023/06/03 21:11:00 - mmengine - INFO - Epoch(train) [2][2400/3907] lr: 9.4338e-05 eta: 5:45:17 time: 0.6314 data_time: 0.0016 memory: 44138 loss: 0.3959 +2023/06/03 21:12:11 - mmengine - INFO - Epoch(train) [2][2500/3907] lr: 9.4161e-05 eta: 5:44:51 time: 0.6291 data_time: 0.0024 memory: 44138 loss: 0.3751 +2023/06/03 21:13:14 - mmengine - INFO - Epoch(train) [2][2600/3907] lr: 9.3981e-05 eta: 5:43:49 time: 0.6312 data_time: 0.0018 memory: 44138 loss: 0.3631 +2023/06/03 21:14:17 - mmengine - INFO - Epoch(train) [2][2700/3907] lr: 9.3799e-05 eta: 5:42:45 time: 0.6317 data_time: 0.0018 memory: 44138 loss: 0.3698 +2023/06/03 21:15:21 - mmengine - INFO - Epoch(train) [2][2800/3907] lr: 9.3615e-05 eta: 5:41:40 time: 0.6312 data_time: 0.0021 memory: 44138 loss: 0.3787 +2023/06/03 21:16:24 - mmengine - INFO - Epoch(train) [2][2900/3907] lr: 9.3428e-05 eta: 5:40:36 time: 0.6308 data_time: 0.0022 memory: 44138 loss: 0.3891 +2023/06/03 21:17:27 - mmengine - INFO - Epoch(train) [2][3000/3907] lr: 9.3238e-05 eta: 5:39:33 time: 0.6332 data_time: 0.0018 memory: 44138 loss: 0.3696 +2023/06/03 21:18:26 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230603_200211 +2023/06/03 21:18:30 - mmengine - INFO - Epoch(train) [2][3100/3907] lr: 9.3046e-05 eta: 5:38:28 time: 0.6316 data_time: 0.0020 memory: 44138 loss: 0.3819 +2023/06/03 21:19:33 - mmengine - INFO - Epoch(train) [2][3200/3907] lr: 9.2852e-05 eta: 5:37:24 time: 0.6299 data_time: 0.0016 memory: 44138 loss: 0.3803 +2023/06/03 21:20:36 - mmengine - INFO - Epoch(train) [2][3300/3907] lr: 9.2655e-05 eta: 5:36:19 time: 0.6308 data_time: 0.0017 memory: 44138 loss: 0.3724 +2023/06/03 21:21:40 - mmengine - INFO - Epoch(train) [2][3400/3907] lr: 9.2456e-05 eta: 5:35:15 time: 0.6313 data_time: 0.0020 memory: 44138 loss: 0.3619 +2023/06/03 21:22:43 - mmengine - INFO - Epoch(train) [2][3500/3907] lr: 9.2254e-05 eta: 5:34:11 time: 0.6308 data_time: 0.0016 memory: 44138 loss: 0.3674 +2023/06/03 21:23:46 - mmengine - INFO - Epoch(train) [2][3600/3907] lr: 9.2050e-05 eta: 5:33:07 time: 0.6305 data_time: 0.0017 memory: 44138 loss: 0.3908 +2023/06/03 21:24:49 - mmengine - INFO - Epoch(train) [2][3700/3907] lr: 9.1843e-05 eta: 5:32:02 time: 0.6309 data_time: 0.0019 memory: 44138 loss: 0.4113 +2023/06/03 21:25:52 - mmengine - INFO - Epoch(train) [2][3800/3907] lr: 9.1634e-05 eta: 5:30:58 time: 0.6308 data_time: 0.0023 memory: 44138 loss: 0.3974 +2023/06/03 21:26:55 - mmengine - INFO - Epoch(train) [2][3900/3907] lr: 9.1423e-05 eta: 5:29:54 time: 0.6297 data_time: 0.0014 memory: 44138 loss: 0.3795 +2023/06/03 21:26:59 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230603_200211 +2023/06/03 21:26:59 - mmengine - INFO - Saving checkpoint at 2 epochs +2023/06/03 21:28:34 - mmengine - INFO - Epoch(val) [2][57/57] accuracy/top1: 78.7203 single-label/precision_classwise: [72.86836242675781, 95.35972595214844] single-label/recall_classwise: [97.80947875976562, 55.27907180786133] single-label/f1-score_classwise: [83.5166015625, 69.98724365234375] data_time: 0.0316 time: 1.2824 +2023/06/03 21:29:39 - mmengine - INFO - Epoch(train) [3][ 100/3907] lr: 9.1194e-05 eta: 5:28:55 time: 0.6310 data_time: 0.0019 memory: 44138 loss: 0.3834 +2023/06/03 21:30:34 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230603_200211 +2023/06/03 21:30:43 - mmengine - INFO - Epoch(train) [3][ 200/3907] lr: 9.0978e-05 eta: 5:27:51 time: 0.6320 data_time: 0.0016 memory: 44138 loss: 0.3921 +2023/06/03 21:31:46 - mmengine - INFO - Epoch(train) [3][ 300/3907] lr: 9.0759e-05 eta: 5:26:47 time: 0.6319 data_time: 0.0021 memory: 44138 loss: 0.3819 +2023/06/03 21:32:49 - mmengine - INFO - Epoch(train) [3][ 400/3907] lr: 9.0539e-05 eta: 5:25:43 time: 0.6310 data_time: 0.0017 memory: 44138 loss: 0.3891 +2023/06/03 21:33:52 - mmengine - INFO - Epoch(train) [3][ 500/3907] lr: 9.0315e-05 eta: 5:24:39 time: 0.6318 data_time: 0.0017 memory: 44138 loss: 0.3636 +2023/06/03 21:34:55 - mmengine - INFO - Epoch(train) [3][ 600/3907] lr: 9.0090e-05 eta: 5:23:35 time: 0.6314 data_time: 0.0014 memory: 44138 loss: 0.3987 +2023/06/03 21:35:59 - mmengine - INFO - Epoch(train) [3][ 700/3907] lr: 8.9862e-05 eta: 5:22:31 time: 0.6311 data_time: 0.0020 memory: 44138 loss: 0.3918 +2023/06/03 21:37:02 - mmengine - INFO - Epoch(train) [3][ 800/3907] lr: 8.9632e-05 eta: 5:21:27 time: 0.6311 data_time: 0.0020 memory: 44138 loss: 0.3671 +2023/06/03 21:38:05 - mmengine - INFO - Epoch(train) [3][ 900/3907] lr: 8.9400e-05 eta: 5:20:23 time: 0.6313 data_time: 0.0020 memory: 44138 loss: 0.3680 +2023/06/03 21:39:08 - mmengine - INFO - Epoch(train) [3][1000/3907] lr: 8.9166e-05 eta: 5:19:19 time: 0.6315 data_time: 0.0020 memory: 44138 loss: 0.3669 +2023/06/03 21:40:11 - mmengine - INFO - Epoch(train) [3][1100/3907] lr: 8.8929e-05 eta: 5:18:16 time: 0.6318 data_time: 0.0017 memory: 44138 loss: 0.3892 +2023/06/03 21:41:06 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230603_200211 +2023/06/03 21:41:14 - mmengine - INFO - Epoch(train) [3][1200/3907] lr: 8.8691e-05 eta: 5:17:12 time: 0.6307 data_time: 0.0016 memory: 44138 loss: 0.4037 +2023/06/03 21:42:18 - mmengine - INFO - Epoch(train) [3][1300/3907] lr: 8.8450e-05 eta: 5:16:08 time: 0.6320 data_time: 0.0015 memory: 44138 loss: 0.3574 +2023/06/03 21:43:21 - mmengine - INFO - Epoch(train) [3][1400/3907] lr: 8.8206e-05 eta: 5:15:04 time: 0.6316 data_time: 0.0020 memory: 44138 loss: 0.3578 +2023/06/03 21:44:24 - mmengine - INFO - Epoch(train) [3][1500/3907] lr: 8.7961e-05 eta: 5:14:01 time: 0.6325 data_time: 0.0020 memory: 44138 loss: 0.3682 +2023/06/03 21:45:27 - mmengine - INFO - Epoch(train) [3][1600/3907] lr: 8.7714e-05 eta: 5:12:57 time: 0.6309 data_time: 0.0018 memory: 44138 loss: 0.3960 +2023/06/03 21:46:30 - mmengine - INFO - Epoch(train) [3][1700/3907] lr: 8.7464e-05 eta: 5:11:53 time: 0.6310 data_time: 0.0014 memory: 44138 loss: 0.3769 +2023/06/03 21:47:34 - mmengine - INFO - Epoch(train) [3][1800/3907] lr: 8.7213e-05 eta: 5:10:49 time: 0.6320 data_time: 0.0022 memory: 44138 loss: 0.3582 +2023/06/03 21:48:37 - mmengine - INFO - Epoch(train) [3][1900/3907] lr: 8.6959e-05 eta: 5:09:46 time: 0.6312 data_time: 0.0020 memory: 44138 loss: 0.3443 +2023/06/03 21:49:40 - mmengine - INFO - Epoch(train) [3][2000/3907] lr: 8.6703e-05 eta: 5:08:42 time: 0.6312 data_time: 0.0016 memory: 44138 loss: 0.3799 +2023/06/03 21:50:43 - mmengine - INFO - Epoch(train) [3][2100/3907] lr: 8.6445e-05 eta: 5:07:38 time: 0.6316 data_time: 0.0017 memory: 44138 loss: 0.3865 +2023/06/03 21:51:37 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230603_200211 +2023/06/03 21:51:46 - mmengine - INFO - Epoch(train) [3][2200/3907] lr: 8.6186e-05 eta: 5:06:34 time: 0.6310 data_time: 0.0015 memory: 44138 loss: 0.3751 +2023/06/03 21:52:49 - mmengine - INFO - Epoch(train) [3][2300/3907] lr: 8.5924e-05 eta: 5:05:30 time: 0.6305 data_time: 0.0017 memory: 44138 loss: 0.3594 +2023/06/03 21:53:52 - mmengine - INFO - Epoch(train) [3][2400/3907] lr: 8.5660e-05 eta: 5:04:26 time: 0.6307 data_time: 0.0020 memory: 44138 loss: 0.4063 +2023/06/03 21:54:56 - mmengine - INFO - Epoch(train) [3][2500/3907] lr: 8.5394e-05 eta: 5:03:23 time: 0.6323 data_time: 0.0018 memory: 44138 loss: 0.3433 +2023/06/03 21:55:59 - mmengine - INFO - Epoch(train) [3][2600/3907] lr: 8.5126e-05 eta: 5:02:19 time: 0.6303 data_time: 0.0021 memory: 44138 loss: 0.3801 +2023/06/03 21:57:02 - mmengine - INFO - Epoch(train) [3][2700/3907] lr: 8.4856e-05 eta: 5:01:15 time: 0.6303 data_time: 0.0020 memory: 44138 loss: 0.4017 +2023/06/03 21:58:05 - mmengine - INFO - Epoch(train) [3][2800/3907] lr: 8.4585e-05 eta: 5:00:11 time: 0.6311 data_time: 0.0021 memory: 44138 loss: 0.3734 +2023/06/03 21:59:08 - mmengine - INFO - Epoch(train) [3][2900/3907] lr: 8.4311e-05 eta: 4:59:07 time: 0.6303 data_time: 0.0020 memory: 44138 loss: 0.3902 +2023/06/03 22:00:11 - mmengine - INFO - Epoch(train) [3][3000/3907] lr: 8.4036e-05 eta: 4:58:04 time: 0.6309 data_time: 0.0016 memory: 44138 loss: 0.3782 +2023/06/03 22:01:15 - mmengine - INFO - Epoch(train) [3][3100/3907] lr: 8.3758e-05 eta: 4:57:01 time: 0.6306 data_time: 0.0018 memory: 44138 loss: 0.3628 +2023/06/03 22:02:09 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230603_200211 +2023/06/03 22:02:18 - mmengine - INFO - Epoch(train) [3][3200/3907] lr: 8.3479e-05 eta: 4:55:57 time: 0.6307 data_time: 0.0023 memory: 44138 loss: 0.3788 +2023/06/03 22:03:21 - mmengine - INFO - Epoch(train) [3][3300/3907] lr: 8.3198e-05 eta: 4:54:53 time: 0.6313 data_time: 0.0020 memory: 44138 loss: 0.3737 +2023/06/03 22:04:24 - mmengine - INFO - Epoch(train) [3][3400/3907] lr: 8.2915e-05 eta: 4:53:49 time: 0.6309 data_time: 0.0018 memory: 44138 loss: 0.3662 +2023/06/03 22:05:27 - mmengine - INFO - Epoch(train) [3][3500/3907] lr: 8.2630e-05 eta: 4:52:46 time: 0.6311 data_time: 0.0015 memory: 44138 loss: 0.3600 +2023/06/03 22:06:30 - mmengine - INFO - Epoch(train) [3][3600/3907] lr: 8.2344e-05 eta: 4:51:42 time: 0.6313 data_time: 0.0018 memory: 44138 loss: 0.3801 +2023/06/03 22:07:33 - mmengine - INFO - Epoch(train) [3][3700/3907] lr: 8.2056e-05 eta: 4:50:38 time: 0.6312 data_time: 0.0025 memory: 44138 loss: 0.3770 +2023/06/03 22:08:36 - mmengine - INFO - Epoch(train) [3][3800/3907] lr: 8.1765e-05 eta: 4:49:35 time: 0.6310 data_time: 0.0018 memory: 44138 loss: 0.3787 +2023/06/03 22:09:40 - mmengine - INFO - Epoch(train) [3][3900/3907] lr: 8.1474e-05 eta: 4:48:31 time: 0.6307 data_time: 0.0016 memory: 44138 loss: 0.3632 +2023/06/03 22:09:43 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230603_200211 +2023/06/03 22:09:43 - mmengine - INFO - Saving checkpoint at 3 epochs +2023/06/03 22:11:18 - mmengine - INFO - Epoch(val) [3][57/57] accuracy/top1: 77.3738 single-label/precision_classwise: [71.36255645751953, 96.45606231689453] single-label/recall_classwise: [98.4596939086914, 51.480621337890625] single-label/f1-score_classwise: [82.7492904663086, 67.13166809082031] data_time: 0.0312 time: 1.2824 +2023/06/03 22:12:25 - mmengine - INFO - Epoch(train) [4][ 100/3907] lr: 8.1160e-05 eta: 4:47:29 time: 0.6302 data_time: 0.0019 memory: 44138 loss: 0.3498 +2023/06/03 22:13:28 - mmengine - INFO - Epoch(train) [4][ 200/3907] lr: 8.0864e-05 eta: 4:46:25 time: 0.6305 data_time: 0.0019 memory: 44138 loss: 0.3540 +2023/06/03 22:14:18 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230603_200211 +2023/06/03 22:14:31 - mmengine - INFO - Epoch(train) [4][ 300/3907] lr: 8.0567e-05 eta: 4:45:22 time: 0.6312 data_time: 0.0018 memory: 44138 loss: 0.3563 +2023/06/03 22:15:34 - mmengine - INFO - Epoch(train) [4][ 400/3907] lr: 8.0269e-05 eta: 4:44:18 time: 0.6315 data_time: 0.0020 memory: 44138 loss: 0.3557 +2023/06/03 22:16:37 - mmengine - INFO - Epoch(train) [4][ 500/3907] lr: 7.9969e-05 eta: 4:43:14 time: 0.6313 data_time: 0.0016 memory: 44138 loss: 0.3599 +2023/06/03 22:17:40 - mmengine - INFO - Epoch(train) [4][ 600/3907] lr: 7.9667e-05 eta: 4:42:11 time: 0.6317 data_time: 0.0021 memory: 44138 loss: 0.3725 +2023/06/03 22:18:44 - mmengine - INFO - Epoch(train) [4][ 700/3907] lr: 7.9363e-05 eta: 4:41:07 time: 0.6317 data_time: 0.0016 memory: 44138 loss: 0.3689 +2023/06/03 22:19:47 - mmengine - INFO - Epoch(train) [4][ 800/3907] lr: 7.9058e-05 eta: 4:40:03 time: 0.6310 data_time: 0.0019 memory: 44138 loss: 0.3983 +2023/06/03 22:20:50 - mmengine - INFO - Epoch(train) [4][ 900/3907] lr: 7.8752e-05 eta: 4:39:00 time: 0.6312 data_time: 0.0019 memory: 44138 loss: 0.3493 +2023/06/03 22:21:53 - mmengine - INFO - Epoch(train) [4][1000/3907] lr: 7.8444e-05 eta: 4:37:56 time: 0.6311 data_time: 0.0019 memory: 44138 loss: 0.3827 +2023/06/03 22:22:56 - mmengine - INFO - Epoch(train) [4][1100/3907] lr: 7.8134e-05 eta: 4:36:52 time: 0.6318 data_time: 0.0018 memory: 44138 loss: 0.3820 +2023/06/03 22:23:59 - mmengine - INFO - Epoch(train) [4][1200/3907] lr: 7.7823e-05 eta: 4:35:49 time: 0.6321 data_time: 0.0017 memory: 44138 loss: 0.3740 +2023/06/03 22:24:49 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230603_200211 +2023/06/03 22:25:03 - mmengine - INFO - Epoch(train) [4][1300/3907] lr: 7.7510e-05 eta: 4:34:46 time: 0.6310 data_time: 0.0017 memory: 44138 loss: 0.3948 +2023/06/03 22:26:06 - mmengine - INFO - Epoch(train) [4][1400/3907] lr: 7.7196e-05 eta: 4:33:42 time: 0.6303 data_time: 0.0018 memory: 44138 loss: 0.3653 +2023/06/03 22:27:09 - mmengine - INFO - Epoch(train) [4][1500/3907] lr: 7.6881e-05 eta: 4:32:38 time: 0.6365 data_time: 0.0024 memory: 44138 loss: 0.3718 +2023/06/03 22:28:12 - mmengine - INFO - Epoch(train) [4][1600/3907] lr: 7.6564e-05 eta: 4:31:35 time: 0.6322 data_time: 0.0022 memory: 44138 loss: 0.3774 +2023/06/03 22:29:15 - mmengine - INFO - Epoch(train) [4][1700/3907] lr: 7.6246e-05 eta: 4:30:32 time: 0.6309 data_time: 0.0023 memory: 44138 loss: 0.3754 +2023/06/03 22:30:19 - mmengine - INFO - Epoch(train) [4][1800/3907] lr: 7.5926e-05 eta: 4:29:29 time: 0.6322 data_time: 0.0019 memory: 44138 loss: 0.3498 +2023/06/03 22:31:22 - mmengine - INFO - Epoch(train) [4][1900/3907] lr: 7.5605e-05 eta: 4:28:25 time: 0.6312 data_time: 0.0016 memory: 44138 loss: 0.3788 +2023/06/03 22:32:25 - mmengine - INFO - Epoch(train) [4][2000/3907] lr: 7.5283e-05 eta: 4:27:22 time: 0.6312 data_time: 0.0020 memory: 44138 loss: 0.3811 +2023/06/03 22:33:28 - mmengine - INFO - Epoch(train) [4][2100/3907] lr: 7.4959e-05 eta: 4:26:18 time: 0.6302 data_time: 0.0018 memory: 44138 loss: 0.3675 +2023/06/03 22:34:32 - mmengine - INFO - Epoch(train) [4][2200/3907] lr: 7.4634e-05 eta: 4:25:15 time: 0.6401 data_time: 0.0020 memory: 44138 loss: 0.3515 +2023/06/03 22:35:21 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230603_200211 +2023/06/03 22:35:35 - mmengine - INFO - Epoch(train) [4][2300/3907] lr: 7.4308e-05 eta: 4:24:11 time: 0.6307 data_time: 0.0017 memory: 44138 loss: 0.3616 +2023/06/03 22:36:38 - mmengine - INFO - Epoch(train) [4][2400/3907] lr: 7.3980e-05 eta: 4:23:08 time: 0.6305 data_time: 0.0015 memory: 44138 loss: 0.3717 +2023/06/03 22:37:41 - mmengine - INFO - Epoch(train) [4][2500/3907] lr: 7.3652e-05 eta: 4:22:05 time: 0.6309 data_time: 0.0015 memory: 44138 loss: 0.4072 +2023/06/03 22:38:44 - mmengine - INFO - Epoch(train) [4][2600/3907] lr: 7.3322e-05 eta: 4:21:01 time: 0.6309 data_time: 0.0018 memory: 44138 loss: 0.3763 +2023/06/03 22:39:47 - mmengine - INFO - Epoch(train) [4][2700/3907] lr: 7.2991e-05 eta: 4:19:57 time: 0.6306 data_time: 0.0018 memory: 44138 loss: 0.3638 +2023/06/03 22:40:51 - mmengine - INFO - Epoch(train) [4][2800/3907] lr: 7.2659e-05 eta: 4:18:54 time: 0.6311 data_time: 0.0019 memory: 44138 loss: 0.3613 +2023/06/03 22:41:54 - mmengine - INFO - Epoch(train) [4][2900/3907] lr: 7.2325e-05 eta: 4:17:51 time: 0.6326 data_time: 0.0016 memory: 44138 loss: 0.3730 +2023/06/03 22:42:57 - mmengine - INFO - Epoch(train) [4][3000/3907] lr: 7.1991e-05 eta: 4:16:47 time: 0.6305 data_time: 0.0018 memory: 44138 loss: 0.3716 +2023/06/03 22:44:00 - mmengine - INFO - Epoch(train) [4][3100/3907] lr: 7.1655e-05 eta: 4:15:44 time: 0.6315 data_time: 0.0017 memory: 44138 loss: 0.3901 +2023/06/03 22:45:03 - mmengine - INFO - Epoch(train) [4][3200/3907] lr: 7.1318e-05 eta: 4:14:40 time: 0.6310 data_time: 0.0024 memory: 44138 loss: 0.3803 +2023/06/03 22:45:53 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230603_200211 +2023/06/03 22:46:07 - mmengine - INFO - Epoch(train) [4][3300/3907] lr: 7.0981e-05 eta: 4:13:37 time: 0.6337 data_time: 0.0015 memory: 44138 loss: 0.3721 +2023/06/03 22:47:10 - mmengine - INFO - Epoch(train) [4][3400/3907] lr: 7.0642e-05 eta: 4:12:33 time: 0.6312 data_time: 0.0015 memory: 44138 loss: 0.3566 +2023/06/03 22:48:13 - mmengine - INFO - Epoch(train) [4][3500/3907] lr: 7.0302e-05 eta: 4:11:30 time: 0.6539 data_time: 0.0017 memory: 44138 loss: 0.3734 +2023/06/03 22:49:16 - mmengine - INFO - Epoch(train) [4][3600/3907] lr: 6.9961e-05 eta: 4:10:27 time: 0.6307 data_time: 0.0020 memory: 44138 loss: 0.3725 +2023/06/03 22:50:19 - mmengine - INFO - Epoch(train) [4][3700/3907] lr: 6.9620e-05 eta: 4:09:23 time: 0.6310 data_time: 0.0015 memory: 44138 loss: 0.3826 +2023/06/03 22:51:22 - mmengine - INFO - Epoch(train) [4][3800/3907] lr: 6.9277e-05 eta: 4:08:20 time: 0.6316 data_time: 0.0025 memory: 44138 loss: 0.3834 +2023/06/03 22:52:26 - mmengine - INFO - Epoch(train) [4][3900/3907] lr: 6.8933e-05 eta: 4:07:16 time: 0.6303 data_time: 0.0014 memory: 44138 loss: 0.3944 +2023/06/03 22:52:30 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230603_200211 +2023/06/03 22:52:30 - mmengine - INFO - Saving checkpoint at 4 epochs +2023/06/03 22:54:04 - mmengine - INFO - Epoch(val) [4][57/57] accuracy/top1: 76.4274 single-label/precision_classwise: [70.36570739746094, 97.23893737792969] single-label/recall_classwise: [98.8700180053711, 48.86821746826172] single-label/f1-score_classwise: [82.21737670898438, 65.04669189453125] data_time: 0.0320 time: 1.2823 +2023/06/03 22:55:10 - mmengine - INFO - Epoch(train) [5][ 100/3907] lr: 6.8565e-05 eta: 4:06:12 time: 0.6303 data_time: 0.0016 memory: 44138 loss: 0.3705 +2023/06/03 22:56:14 - mmengine - INFO - Epoch(train) [5][ 200/3907] lr: 6.8219e-05 eta: 4:05:09 time: 0.6314 data_time: 0.0018 memory: 44138 loss: 0.3432 +2023/06/03 22:57:17 - mmengine - INFO - Epoch(train) [5][ 300/3907] lr: 6.7873e-05 eta: 4:04:05 time: 0.6310 data_time: 0.0016 memory: 44138 loss: 0.3947 +2023/06/03 22:58:02 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230603_200211 +2023/06/03 22:58:20 - mmengine - INFO - Epoch(train) [5][ 400/3907] lr: 6.7526e-05 eta: 4:03:02 time: 0.6306 data_time: 0.0019 memory: 44138 loss: 0.3779 +2023/06/03 22:59:23 - mmengine - INFO - Epoch(train) [5][ 500/3907] lr: 6.7178e-05 eta: 4:01:58 time: 0.6303 data_time: 0.0018 memory: 44138 loss: 0.3887 +2023/06/03 23:00:26 - mmengine - INFO - Epoch(train) [5][ 600/3907] lr: 6.6829e-05 eta: 4:00:55 time: 0.6308 data_time: 0.0025 memory: 44138 loss: 0.3567 +2023/06/03 23:01:29 - mmengine - INFO - Epoch(train) [5][ 700/3907] lr: 6.6480e-05 eta: 3:59:51 time: 0.6314 data_time: 0.0021 memory: 44138 loss: 0.3664 +2023/06/03 23:02:33 - mmengine - INFO - Epoch(train) [5][ 800/3907] lr: 6.6129e-05 eta: 3:58:48 time: 0.6322 data_time: 0.0019 memory: 44138 loss: 0.3559 +2023/06/03 23:03:36 - mmengine - INFO - Epoch(train) [5][ 900/3907] lr: 6.5778e-05 eta: 3:57:45 time: 0.6327 data_time: 0.0014 memory: 44138 loss: 0.3608 +2023/06/03 23:04:39 - mmengine - INFO - Epoch(train) [5][1000/3907] lr: 6.5427e-05 eta: 3:56:42 time: 0.6317 data_time: 0.0020 memory: 44138 loss: 0.3771 +2023/06/03 23:05:42 - mmengine - INFO - Epoch(train) [5][1100/3907] lr: 6.5074e-05 eta: 3:55:38 time: 0.6321 data_time: 0.0014 memory: 44138 loss: 0.3802 +2023/06/03 23:06:45 - mmengine - INFO - Epoch(train) [5][1200/3907] lr: 6.4721e-05 eta: 3:54:35 time: 0.6317 data_time: 0.0019 memory: 44138 loss: 0.3690 +2023/06/03 23:07:49 - mmengine - INFO - Epoch(train) [5][1300/3907] lr: 6.4368e-05 eta: 3:53:31 time: 0.6384 data_time: 0.0019 memory: 44138 loss: 0.3448 +2023/06/03 23:08:34 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230603_200211 +2023/06/03 23:08:52 - mmengine - INFO - Epoch(train) [5][1400/3907] lr: 6.4014e-05 eta: 3:52:28 time: 0.6333 data_time: 0.0017 memory: 44138 loss: 0.3552 +2023/06/03 23:09:55 - mmengine - INFO - Epoch(train) [5][1500/3907] lr: 6.3659e-05 eta: 3:51:25 time: 0.6307 data_time: 0.0018 memory: 44138 loss: 0.3883 +2023/06/03 23:10:58 - mmengine - INFO - Epoch(train) [5][1600/3907] lr: 6.3303e-05 eta: 3:50:21 time: 0.6309 data_time: 0.0022 memory: 44138 loss: 0.3621 +2023/06/03 23:12:02 - mmengine - INFO - Epoch(train) [5][1700/3907] lr: 6.2948e-05 eta: 3:49:18 time: 0.6304 data_time: 0.0018 memory: 44138 loss: 0.3709 +2023/06/03 23:13:05 - mmengine - INFO - Epoch(train) [5][1800/3907] lr: 6.2591e-05 eta: 3:48:15 time: 0.6404 data_time: 0.0017 memory: 44138 loss: 0.3885 +2023/06/03 23:14:08 - mmengine - INFO - Epoch(train) [5][1900/3907] lr: 6.2234e-05 eta: 3:47:11 time: 0.6304 data_time: 0.0022 memory: 44138 loss: 0.3782 +2023/06/03 23:15:11 - mmengine - INFO - Epoch(train) [5][2000/3907] lr: 6.1877e-05 eta: 3:46:08 time: 0.6312 data_time: 0.0023 memory: 44138 loss: 0.3913 +2023/06/03 23:16:14 - mmengine - INFO - Epoch(train) [5][2100/3907] lr: 6.1519e-05 eta: 3:45:04 time: 0.6314 data_time: 0.0025 memory: 44138 loss: 0.3545 +2023/06/03 23:17:17 - mmengine - INFO - Epoch(train) [5][2200/3907] lr: 6.1161e-05 eta: 3:44:01 time: 0.6316 data_time: 0.0017 memory: 44138 loss: 0.3647 +2023/06/03 23:18:21 - mmengine - INFO - Epoch(train) [5][2300/3907] lr: 6.0802e-05 eta: 3:42:57 time: 0.6305 data_time: 0.0020 memory: 44138 loss: 0.3600 +2023/06/03 23:19:06 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230603_200211 +2023/06/03 23:19:24 - mmengine - INFO - Epoch(train) [5][2400/3907] lr: 6.0443e-05 eta: 3:41:54 time: 0.6312 data_time: 0.0020 memory: 44138 loss: 0.3566 +2023/06/03 23:20:27 - mmengine - INFO - Epoch(train) [5][2500/3907] lr: 6.0084e-05 eta: 3:40:51 time: 0.6313 data_time: 0.0015 memory: 44138 loss: 0.3821 +2023/06/03 23:21:30 - mmengine - INFO - Epoch(train) [5][2600/3907] lr: 5.9724e-05 eta: 3:39:47 time: 0.6456 data_time: 0.0017 memory: 44138 loss: 0.3700 +2023/06/03 23:22:34 - mmengine - INFO - Epoch(train) [5][2700/3907] lr: 5.9364e-05 eta: 3:38:44 time: 0.6310 data_time: 0.0019 memory: 44138 loss: 0.3628 +2023/06/03 23:23:37 - mmengine - INFO - Epoch(train) [5][2800/3907] lr: 5.9004e-05 eta: 3:37:41 time: 0.6333 data_time: 0.0015 memory: 44138 loss: 0.3742 +2023/06/03 23:24:40 - mmengine - INFO - Epoch(train) [5][2900/3907] lr: 5.8643e-05 eta: 3:36:38 time: 0.6298 data_time: 0.0017 memory: 44138 loss: 0.3746 +2023/06/03 23:25:43 - mmengine - INFO - Epoch(train) [5][3000/3907] lr: 5.8283e-05 eta: 3:35:34 time: 0.6302 data_time: 0.0016 memory: 44138 loss: 0.3801 +2023/06/03 23:26:46 - mmengine - INFO - Epoch(train) [5][3100/3907] lr: 5.7922e-05 eta: 3:34:31 time: 0.6301 data_time: 0.0014 memory: 44138 loss: 0.3742 +2023/06/03 23:27:49 - mmengine - INFO - Epoch(train) [5][3200/3907] lr: 5.7560e-05 eta: 3:33:27 time: 0.6307 data_time: 0.0019 memory: 44138 loss: 0.3764 +2023/06/03 23:28:53 - mmengine - INFO - Epoch(train) [5][3300/3907] lr: 5.7199e-05 eta: 3:32:24 time: 0.6307 data_time: 0.0024 memory: 44138 loss: 0.3833 +2023/06/03 23:29:38 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230603_200211 +2023/06/03 23:29:56 - mmengine - INFO - Epoch(train) [5][3400/3907] lr: 5.6838e-05 eta: 3:31:20 time: 0.6308 data_time: 0.0016 memory: 44138 loss: 0.3594 +2023/06/03 23:30:59 - mmengine - INFO - Epoch(train) [5][3500/3907] lr: 5.6476e-05 eta: 3:30:17 time: 0.6303 data_time: 0.0018 memory: 44138 loss: 0.3770 +2023/06/03 23:32:02 - mmengine - INFO - Epoch(train) [5][3600/3907] lr: 5.6114e-05 eta: 3:29:13 time: 0.6305 data_time: 0.0016 memory: 44138 loss: 0.3793 +2023/06/03 23:33:05 - mmengine - INFO - Epoch(train) [5][3700/3907] lr: 5.5753e-05 eta: 3:28:10 time: 0.6307 data_time: 0.0027 memory: 44138 loss: 0.3612 +2023/06/03 23:34:08 - mmengine - INFO - Epoch(train) [5][3800/3907] lr: 5.5391e-05 eta: 3:27:06 time: 0.6311 data_time: 0.0015 memory: 44138 loss: 0.3734 +2023/06/03 23:35:11 - mmengine - INFO - Epoch(train) [5][3900/3907] lr: 5.5029e-05 eta: 3:26:03 time: 0.6302 data_time: 0.0014 memory: 44138 loss: 0.3950 +2023/06/03 23:35:15 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230603_200211 +2023/06/03 23:35:15 - mmengine - INFO - Saving checkpoint at 5 epochs +2023/06/03 23:36:48 - mmengine - INFO - Epoch(val) [5][57/57] accuracy/top1: 75.8220 single-label/precision_classwise: [69.77406311035156, 97.55474090576172] single-label/recall_classwise: [99.0341567993164, 47.31782913208008] single-label/f1-score_classwise: [81.86823272705078, 63.726051330566406] data_time: 0.0314 time: 1.2812 +2023/06/03 23:37:55 - mmengine - INFO - Epoch(train) [6][ 100/3907] lr: 5.4642e-05 eta: 3:24:58 time: 0.6306 data_time: 0.0018 memory: 44138 loss: 0.3492 +2023/06/03 23:38:58 - mmengine - INFO - Epoch(train) [6][ 200/3907] lr: 5.4280e-05 eta: 3:23:55 time: 0.6308 data_time: 0.0015 memory: 44138 loss: 0.3644 +2023/06/03 23:40:01 - mmengine - INFO - Epoch(train) [6][ 300/3907] lr: 5.3918e-05 eta: 3:22:51 time: 0.6313 data_time: 0.0017 memory: 44138 loss: 0.3770 +2023/06/03 23:41:04 - mmengine - INFO - Epoch(train) [6][ 400/3907] lr: 5.3556e-05 eta: 3:21:48 time: 0.6310 data_time: 0.0015 memory: 44138 loss: 0.3632 +2023/06/03 23:41:45 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230603_200211 +2023/06/03 23:42:08 - mmengine - INFO - Epoch(train) [6][ 500/3907] lr: 5.3195e-05 eta: 3:20:44 time: 0.6306 data_time: 0.0016 memory: 44138 loss: 0.3822 +2023/06/03 23:43:11 - mmengine - INFO - Epoch(train) [6][ 600/3907] lr: 5.2833e-05 eta: 3:19:41 time: 0.6411 data_time: 0.0017 memory: 44138 loss: 0.3516 +2023/06/03 23:44:14 - mmengine - INFO - Epoch(train) [6][ 700/3907] lr: 5.2472e-05 eta: 3:18:38 time: 0.6313 data_time: 0.0015 memory: 44138 loss: 0.3755 +2023/06/03 23:45:17 - mmengine - INFO - Epoch(train) [6][ 800/3907] lr: 5.2111e-05 eta: 3:17:34 time: 0.6310 data_time: 0.0018 memory: 44138 loss: 0.3542 +2023/06/03 23:46:20 - mmengine - INFO - Epoch(train) [6][ 900/3907] lr: 5.1750e-05 eta: 3:16:31 time: 0.6308 data_time: 0.0016 memory: 44138 loss: 0.3853 +2023/06/03 23:47:23 - mmengine - INFO - Epoch(train) [6][1000/3907] lr: 5.1389e-05 eta: 3:15:27 time: 0.6308 data_time: 0.0020 memory: 44138 loss: 0.3786 +2023/06/03 23:48:26 - mmengine - INFO - Epoch(train) [6][1100/3907] lr: 5.1029e-05 eta: 3:14:24 time: 0.6312 data_time: 0.0018 memory: 44138 loss: 0.3822 +2023/06/03 23:49:30 - mmengine - INFO - Epoch(train) [6][1200/3907] lr: 5.0668e-05 eta: 3:13:21 time: 0.6316 data_time: 0.0017 memory: 44138 loss: 0.3927 +2023/06/03 23:50:33 - mmengine - INFO - Epoch(train) [6][1300/3907] lr: 5.0308e-05 eta: 3:12:17 time: 0.6301 data_time: 0.0017 memory: 44138 loss: 0.3739 +2023/06/03 23:51:36 - mmengine - INFO - Epoch(train) [6][1400/3907] lr: 4.9949e-05 eta: 3:11:14 time: 0.6306 data_time: 0.0017 memory: 44138 loss: 0.3831 +2023/06/03 23:52:17 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230603_200211 +2023/06/03 23:52:39 - mmengine - INFO - Epoch(train) [6][1500/3907] lr: 4.9589e-05 eta: 3:10:10 time: 0.6307 data_time: 0.0017 memory: 44138 loss: 0.3832 +2023/06/03 23:53:42 - mmengine - INFO - Epoch(train) [6][1600/3907] lr: 4.9230e-05 eta: 3:09:07 time: 0.6311 data_time: 0.0021 memory: 44138 loss: 0.3894 +2023/06/03 23:54:45 - mmengine - INFO - Epoch(train) [6][1700/3907] lr: 4.8871e-05 eta: 3:08:04 time: 0.6309 data_time: 0.0014 memory: 44138 loss: 0.3913 +2023/06/03 23:55:48 - mmengine - INFO - Epoch(train) [6][1800/3907] lr: 4.8513e-05 eta: 3:07:00 time: 0.6303 data_time: 0.0019 memory: 44138 loss: 0.3497 +2023/06/03 23:56:51 - mmengine - INFO - Epoch(train) [6][1900/3907] lr: 4.8155e-05 eta: 3:05:57 time: 0.6316 data_time: 0.0026 memory: 44138 loss: 0.3953 +2023/06/03 23:57:55 - mmengine - INFO - Epoch(train) [6][2000/3907] lr: 4.7798e-05 eta: 3:04:53 time: 0.6308 data_time: 0.0018 memory: 44138 loss: 0.3586 +2023/06/03 23:58:58 - mmengine - INFO - Epoch(train) [6][2100/3907] lr: 4.7441e-05 eta: 3:03:50 time: 0.6308 data_time: 0.0014 memory: 44138 loss: 0.3702 +2023/06/04 00:00:01 - mmengine - INFO - Epoch(train) [6][2200/3907] lr: 4.7084e-05 eta: 3:02:47 time: 0.6305 data_time: 0.0016 memory: 44138 loss: 0.3841 +2023/06/04 00:01:04 - mmengine - INFO - Epoch(train) [6][2300/3907] lr: 4.6729e-05 eta: 3:01:43 time: 0.6305 data_time: 0.0016 memory: 44138 loss: 0.3673 +2023/06/04 00:02:07 - mmengine - INFO - Epoch(train) [6][2400/3907] lr: 4.6373e-05 eta: 3:00:40 time: 0.6301 data_time: 0.0019 memory: 44138 loss: 0.3601 +2023/06/04 00:02:48 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230603_200211 +2023/06/04 00:03:10 - mmengine - INFO - Epoch(train) [6][2500/3907] lr: 4.6018e-05 eta: 2:59:36 time: 0.6310 data_time: 0.0015 memory: 44138 loss: 0.3689 +2023/06/04 00:04:13 - mmengine - INFO - Epoch(train) [6][2600/3907] lr: 4.5664e-05 eta: 2:58:33 time: 0.6312 data_time: 0.0015 memory: 44138 loss: 0.3938 +2023/06/04 00:05:16 - mmengine - INFO - Epoch(train) [6][2700/3907] lr: 4.5310e-05 eta: 2:57:30 time: 0.6315 data_time: 0.0021 memory: 44138 loss: 0.3538 +2023/06/04 00:06:19 - mmengine - INFO - Epoch(train) [6][2800/3907] lr: 4.4957e-05 eta: 2:56:26 time: 0.6310 data_time: 0.0023 memory: 44138 loss: 0.3428 +2023/06/04 00:07:22 - mmengine - INFO - Epoch(train) [6][2900/3907] lr: 4.4605e-05 eta: 2:55:23 time: 0.6308 data_time: 0.0017 memory: 44138 loss: 0.3808 +2023/06/04 00:08:25 - mmengine - INFO - Epoch(train) [6][3000/3907] lr: 4.4253e-05 eta: 2:54:19 time: 0.6310 data_time: 0.0016 memory: 44138 loss: 0.3596 +2023/06/04 00:09:29 - mmengine - INFO - Epoch(train) [6][3100/3907] lr: 4.3902e-05 eta: 2:53:16 time: 0.6305 data_time: 0.0016 memory: 44138 loss: 0.3625 +2023/06/04 00:10:32 - mmengine - INFO - Epoch(train) [6][3200/3907] lr: 4.3552e-05 eta: 2:52:13 time: 0.6304 data_time: 0.0019 memory: 44138 loss: 0.3401 +2023/06/04 00:11:35 - mmengine - INFO - Epoch(train) [6][3300/3907] lr: 4.3202e-05 eta: 2:51:09 time: 0.6302 data_time: 0.0017 memory: 44138 loss: 0.3596 +2023/06/04 00:12:38 - mmengine - INFO - Epoch(train) [6][3400/3907] lr: 4.2854e-05 eta: 2:50:06 time: 0.6298 data_time: 0.0017 memory: 44138 loss: 0.3468 +2023/06/04 00:13:19 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230603_200211 +2023/06/04 00:13:41 - mmengine - INFO - Epoch(train) [6][3500/3907] lr: 4.2506e-05 eta: 2:49:02 time: 0.6312 data_time: 0.0016 memory: 44138 loss: 0.3625 +2023/06/04 00:14:44 - mmengine - INFO - Epoch(train) [6][3600/3907] lr: 4.2158e-05 eta: 2:47:59 time: 0.6313 data_time: 0.0015 memory: 44138 loss: 0.3483 +2023/06/04 00:15:47 - mmengine - INFO - Epoch(train) [6][3700/3907] lr: 4.1812e-05 eta: 2:46:56 time: 0.6320 data_time: 0.0019 memory: 44138 loss: 0.3672 +2023/06/04 00:16:50 - mmengine - INFO - Epoch(train) [6][3800/3907] lr: 4.1466e-05 eta: 2:45:53 time: 0.6314 data_time: 0.0018 memory: 44138 loss: 0.3711 +2023/06/04 00:17:54 - mmengine - INFO - Epoch(train) [6][3900/3907] lr: 4.1122e-05 eta: 2:44:49 time: 0.6316 data_time: 0.0014 memory: 44138 loss: 0.3523 +2023/06/04 00:17:58 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230603_200211 +2023/06/04 00:17:58 - mmengine - INFO - Saving checkpoint at 6 epochs +2023/06/04 00:19:32 - mmengine - INFO - Epoch(val) [6][57/57] accuracy/top1: 75.6759 single-label/precision_classwise: [69.64570617675781, 97.52291870117188] single-label/recall_classwise: [99.02783966064453, 47.0] single-label/f1-score_classwise: [81.77765655517578, 63.43045425415039] data_time: 0.0320 time: 1.2841 +2023/06/04 00:20:38 - mmengine - INFO - Epoch(train) [7][ 100/3907] lr: 4.0754e-05 eta: 2:43:43 time: 0.6312 data_time: 0.0019 memory: 44138 loss: 0.3583 +2023/06/04 00:21:42 - mmengine - INFO - Epoch(train) [7][ 200/3907] lr: 4.0411e-05 eta: 2:42:40 time: 0.6309 data_time: 0.0018 memory: 44138 loss: 0.3571 +2023/06/04 00:22:45 - mmengine - INFO - Epoch(train) [7][ 300/3907] lr: 4.0069e-05 eta: 2:41:37 time: 0.6302 data_time: 0.0020 memory: 44138 loss: 0.3721 +2023/06/04 00:23:48 - mmengine - INFO - Epoch(train) [7][ 400/3907] lr: 3.9729e-05 eta: 2:40:33 time: 0.6311 data_time: 0.0015 memory: 44138 loss: 0.3728 +2023/06/04 00:24:51 - mmengine - INFO - Epoch(train) [7][ 500/3907] lr: 3.9389e-05 eta: 2:39:30 time: 0.6308 data_time: 0.0015 memory: 44138 loss: 0.3507 +2023/06/04 00:25:28 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230603_200211 +2023/06/04 00:25:54 - mmengine - INFO - Epoch(train) [7][ 600/3907] lr: 3.9050e-05 eta: 2:38:26 time: 0.6303 data_time: 0.0015 memory: 44138 loss: 0.3819 +2023/06/04 00:26:57 - mmengine - INFO - Epoch(train) [7][ 700/3907] lr: 3.8712e-05 eta: 2:37:23 time: 0.6301 data_time: 0.0019 memory: 44138 loss: 0.4173 +2023/06/04 00:28:00 - mmengine - INFO - Epoch(train) [7][ 800/3907] lr: 3.8375e-05 eta: 2:36:20 time: 0.6307 data_time: 0.0019 memory: 44138 loss: 0.3878 +2023/06/04 00:29:03 - mmengine - INFO - Epoch(train) [7][ 900/3907] lr: 3.8039e-05 eta: 2:35:16 time: 0.6309 data_time: 0.0015 memory: 44138 loss: 0.3851 +2023/06/04 00:30:07 - mmengine - INFO - Epoch(train) [7][1000/3907] lr: 3.7705e-05 eta: 2:34:13 time: 0.6306 data_time: 0.0017 memory: 44138 loss: 0.3569 +2023/06/04 00:31:10 - mmengine - INFO - Epoch(train) [7][1100/3907] lr: 3.7371e-05 eta: 2:33:10 time: 0.6302 data_time: 0.0019 memory: 44138 loss: 0.4022 +2023/06/04 00:32:13 - mmengine - INFO - Epoch(train) [7][1200/3907] lr: 3.7039e-05 eta: 2:32:06 time: 0.6310 data_time: 0.0016 memory: 44138 loss: 0.3660 +2023/06/04 00:33:16 - mmengine - INFO - Epoch(train) [7][1300/3907] lr: 3.6708e-05 eta: 2:31:03 time: 0.6310 data_time: 0.0015 memory: 44138 loss: 0.3924 +2023/06/04 00:34:19 - mmengine - INFO - Epoch(train) [7][1400/3907] lr: 3.6378e-05 eta: 2:30:00 time: 0.6326 data_time: 0.0024 memory: 44138 loss: 0.3813 +2023/06/04 00:35:22 - mmengine - INFO - Epoch(train) [7][1500/3907] lr: 3.6049e-05 eta: 2:28:56 time: 0.6303 data_time: 0.0020 memory: 44138 loss: 0.3725 +2023/06/04 00:35:59 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230603_200211 +2023/06/04 00:36:25 - mmengine - INFO - Epoch(train) [7][1600/3907] lr: 3.5721e-05 eta: 2:27:53 time: 0.6305 data_time: 0.0016 memory: 44138 loss: 0.3616 +2023/06/04 00:37:28 - mmengine - INFO - Epoch(train) [7][1700/3907] lr: 3.5395e-05 eta: 2:26:50 time: 0.6316 data_time: 0.0017 memory: 44138 loss: 0.3746 +2023/06/04 00:38:32 - mmengine - INFO - Epoch(train) [7][1800/3907] lr: 3.5070e-05 eta: 2:25:46 time: 0.6313 data_time: 0.0014 memory: 44138 loss: 0.3749 +2023/06/04 00:39:35 - mmengine - INFO - Epoch(train) [7][1900/3907] lr: 3.4746e-05 eta: 2:24:43 time: 0.6316 data_time: 0.0016 memory: 44138 loss: 0.3537 +2023/06/04 00:40:38 - mmengine - INFO - Epoch(train) [7][2000/3907] lr: 3.4424e-05 eta: 2:23:40 time: 0.6309 data_time: 0.0016 memory: 44138 loss: 0.3501 +2023/06/04 00:41:41 - mmengine - INFO - Epoch(train) [7][2100/3907] lr: 3.4103e-05 eta: 2:22:36 time: 0.6306 data_time: 0.0017 memory: 44138 loss: 0.3691 +2023/06/04 00:42:44 - mmengine - INFO - Epoch(train) [7][2200/3907] lr: 3.3783e-05 eta: 2:21:33 time: 0.6300 data_time: 0.0018 memory: 44138 loss: 0.3769 +2023/06/04 00:43:47 - mmengine - INFO - Epoch(train) [7][2300/3907] lr: 3.3465e-05 eta: 2:20:30 time: 0.6319 data_time: 0.0015 memory: 44138 loss: 0.3290 +2023/06/04 00:44:51 - mmengine - INFO - Epoch(train) [7][2400/3907] lr: 3.3148e-05 eta: 2:19:27 time: 0.6318 data_time: 0.0016 memory: 44138 loss: 0.3829 +2023/06/04 00:45:54 - mmengine - INFO - Epoch(train) [7][2500/3907] lr: 3.2832e-05 eta: 2:18:23 time: 0.6301 data_time: 0.0021 memory: 44138 loss: 0.3834 +2023/06/04 00:46:30 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230603_200211 +2023/06/04 00:46:57 - mmengine - INFO - Epoch(train) [7][2600/3907] lr: 3.2518e-05 eta: 2:17:20 time: 0.6310 data_time: 0.0015 memory: 44138 loss: 0.3593 +2023/06/04 00:48:00 - mmengine - INFO - Epoch(train) [7][2700/3907] lr: 3.2205e-05 eta: 2:16:17 time: 0.6305 data_time: 0.0015 memory: 44138 loss: 0.3614 +2023/06/04 00:49:03 - mmengine - INFO - Epoch(train) [7][2800/3907] lr: 3.1894e-05 eta: 2:15:13 time: 0.6300 data_time: 0.0016 memory: 44138 loss: 0.3845 +2023/06/04 00:50:06 - mmengine - INFO - Epoch(train) [7][2900/3907] lr: 3.1584e-05 eta: 2:14:10 time: 0.6305 data_time: 0.0015 memory: 44138 loss: 0.3566 +2023/06/04 00:51:09 - mmengine - INFO - Epoch(train) [7][3000/3907] lr: 3.1276e-05 eta: 2:13:07 time: 0.6311 data_time: 0.0014 memory: 44138 loss: 0.3782 +2023/06/04 00:52:12 - mmengine - INFO - Epoch(train) [7][3100/3907] lr: 3.0969e-05 eta: 2:12:03 time: 0.6298 data_time: 0.0016 memory: 44138 loss: 0.3849 +2023/06/04 00:53:15 - mmengine - INFO - Epoch(train) [7][3200/3907] lr: 3.0664e-05 eta: 2:11:00 time: 0.6297 data_time: 0.0016 memory: 44138 loss: 0.3654 +2023/06/04 00:54:18 - mmengine - INFO - Epoch(train) [7][3300/3907] lr: 3.0360e-05 eta: 2:09:57 time: 0.6292 data_time: 0.0016 memory: 44138 loss: 0.3949 +2023/06/04 00:55:21 - mmengine - INFO - Epoch(train) [7][3400/3907] lr: 3.0058e-05 eta: 2:08:53 time: 0.6380 data_time: 0.0022 memory: 44138 loss: 0.3886 +2023/06/04 00:56:25 - mmengine - INFO - Epoch(train) [7][3500/3907] lr: 2.9758e-05 eta: 2:07:50 time: 0.6299 data_time: 0.0019 memory: 44138 loss: 0.3547 +2023/06/04 00:57:01 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230603_200211 +2023/06/04 00:57:28 - mmengine - INFO - Epoch(train) [7][3600/3907] lr: 2.9459e-05 eta: 2:06:47 time: 0.6301 data_time: 0.0019 memory: 44138 loss: 0.3702 +2023/06/04 00:58:31 - mmengine - INFO - Epoch(train) [7][3700/3907] lr: 2.9162e-05 eta: 2:05:43 time: 0.6303 data_time: 0.0019 memory: 44138 loss: 0.3572 +2023/06/04 00:59:34 - mmengine - INFO - Epoch(train) [7][3800/3907] lr: 2.8867e-05 eta: 2:04:40 time: 0.6306 data_time: 0.0015 memory: 44138 loss: 0.3858 +2023/06/04 01:00:37 - mmengine - INFO - Epoch(train) [7][3900/3907] lr: 2.8573e-05 eta: 2:03:37 time: 0.6290 data_time: 0.0012 memory: 44138 loss: 0.3748 +2023/06/04 01:00:41 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230603_200211 +2023/06/04 01:00:41 - mmengine - INFO - Saving checkpoint at 7 epochs +2023/06/04 01:02:15 - mmengine - INFO - Epoch(val) [7][57/57] accuracy/top1: 75.4393 single-label/precision_classwise: [69.41889190673828, 97.6505126953125] single-label/recall_classwise: [99.0909652709961, 46.395347595214844] single-label/f1-score_classwise: [81.64252471923828, 62.90398025512695] data_time: 0.0322 time: 1.2809 +2023/06/04 01:03:22 - mmengine - INFO - Epoch(train) [8][ 100/3907] lr: 2.8261e-05 eta: 2:02:30 time: 0.6314 data_time: 0.0015 memory: 44138 loss: 0.3699 +2023/06/04 01:04:25 - mmengine - INFO - Epoch(train) [8][ 200/3907] lr: 2.7971e-05 eta: 2:01:27 time: 0.6314 data_time: 0.0015 memory: 44138 loss: 0.3720 +2023/06/04 01:05:28 - mmengine - INFO - Epoch(train) [8][ 300/3907] lr: 2.7682e-05 eta: 2:00:24 time: 0.6308 data_time: 0.0016 memory: 44138 loss: 0.3988 +2023/06/04 01:06:31 - mmengine - INFO - Epoch(train) [8][ 400/3907] lr: 2.7395e-05 eta: 1:59:20 time: 0.6319 data_time: 0.0021 memory: 44138 loss: 0.3655 +2023/06/04 01:07:34 - mmengine - INFO - Epoch(train) [8][ 500/3907] lr: 2.7111e-05 eta: 1:58:17 time: 0.6304 data_time: 0.0016 memory: 44138 loss: 0.3854 +2023/06/04 01:08:37 - mmengine - INFO - Epoch(train) [8][ 600/3907] lr: 2.6828e-05 eta: 1:57:14 time: 0.6311 data_time: 0.0020 memory: 44138 loss: 0.3751 +2023/06/04 01:09:10 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230603_200211 +2023/06/04 01:09:41 - mmengine - INFO - Epoch(train) [8][ 700/3907] lr: 2.6546e-05 eta: 1:56:11 time: 0.6316 data_time: 0.0022 memory: 44138 loss: 0.3753 +2023/06/04 01:10:44 - mmengine - INFO - Epoch(train) [8][ 800/3907] lr: 2.6267e-05 eta: 1:55:07 time: 0.6311 data_time: 0.0015 memory: 44138 loss: 0.3733 +2023/06/04 01:11:47 - mmengine - INFO - Epoch(train) [8][ 900/3907] lr: 2.5989e-05 eta: 1:54:04 time: 0.6313 data_time: 0.0016 memory: 44138 loss: 0.3868 +2023/06/04 01:12:50 - mmengine - INFO - Epoch(train) [8][1000/3907] lr: 2.5714e-05 eta: 1:53:01 time: 0.6319 data_time: 0.0019 memory: 44138 loss: 0.3725 +2023/06/04 01:13:54 - mmengine - INFO - Epoch(train) [8][1100/3907] lr: 2.5440e-05 eta: 1:51:57 time: 0.6319 data_time: 0.0016 memory: 44138 loss: 0.3547 +2023/06/04 01:14:57 - mmengine - INFO - Epoch(train) [8][1200/3907] lr: 2.5168e-05 eta: 1:50:54 time: 0.6326 data_time: 0.0015 memory: 44138 loss: 0.3674 +2023/06/04 01:16:00 - mmengine - INFO - Epoch(train) [8][1300/3907] lr: 2.4898e-05 eta: 1:49:51 time: 0.6323 data_time: 0.0017 memory: 44138 loss: 0.3771 +2023/06/04 01:17:03 - mmengine - INFO - Epoch(train) [8][1400/3907] lr: 2.4630e-05 eta: 1:48:48 time: 0.6319 data_time: 0.0015 memory: 44138 loss: 0.3820 +2023/06/04 01:18:06 - mmengine - INFO - Epoch(train) [8][1500/3907] lr: 2.4364e-05 eta: 1:47:44 time: 0.6316 data_time: 0.0017 memory: 44138 loss: 0.3760 +2023/06/04 01:19:09 - mmengine - INFO - Epoch(train) [8][1600/3907] lr: 2.4100e-05 eta: 1:46:41 time: 0.6311 data_time: 0.0016 memory: 44138 loss: 0.3573 +2023/06/04 01:19:42 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230603_200211 +2023/06/04 01:20:12 - mmengine - INFO - Epoch(train) [8][1700/3907] lr: 2.3838e-05 eta: 1:45:38 time: 0.6313 data_time: 0.0015 memory: 44138 loss: 0.3562 +2023/06/04 01:21:16 - mmengine - INFO - Epoch(train) [8][1800/3907] lr: 2.3578e-05 eta: 1:44:34 time: 0.6316 data_time: 0.0022 memory: 44138 loss: 0.3465 +2023/06/04 01:22:19 - mmengine - INFO - Epoch(train) [8][1900/3907] lr: 2.3320e-05 eta: 1:43:31 time: 0.6304 data_time: 0.0013 memory: 44138 loss: 0.3562 +2023/06/04 01:23:22 - mmengine - INFO - Epoch(train) [8][2000/3907] lr: 2.3064e-05 eta: 1:42:28 time: 0.6332 data_time: 0.0019 memory: 44138 loss: 0.3802 +2023/06/04 01:24:25 - mmengine - INFO - Epoch(train) [8][2100/3907] lr: 2.2810e-05 eta: 1:41:25 time: 0.6321 data_time: 0.0021 memory: 44138 loss: 0.3789 +2023/06/04 01:25:28 - mmengine - INFO - Epoch(train) [8][2200/3907] lr: 2.2558e-05 eta: 1:40:21 time: 0.6311 data_time: 0.0019 memory: 44138 loss: 0.3890 +2023/06/04 01:26:32 - mmengine - INFO - Epoch(train) [8][2300/3907] lr: 2.2309e-05 eta: 1:39:18 time: 0.6322 data_time: 0.0026 memory: 44138 loss: 0.3646 +2023/06/04 01:27:35 - mmengine - INFO - Epoch(train) [8][2400/3907] lr: 2.2061e-05 eta: 1:38:15 time: 0.6306 data_time: 0.0015 memory: 44138 loss: 0.3819 +2023/06/04 01:28:38 - mmengine - INFO - Epoch(train) [8][2500/3907] lr: 2.1816e-05 eta: 1:37:12 time: 0.6312 data_time: 0.0019 memory: 44138 loss: 0.3884 +2023/06/04 01:29:41 - mmengine - INFO - Epoch(train) [8][2600/3907] lr: 2.1572e-05 eta: 1:36:08 time: 0.6303 data_time: 0.0017 memory: 44138 loss: 0.4010 +2023/06/04 01:30:13 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230603_200211 +2023/06/04 01:30:44 - mmengine - INFO - Epoch(train) [8][2700/3907] lr: 2.1331e-05 eta: 1:35:05 time: 0.6305 data_time: 0.0016 memory: 44138 loss: 0.3727 +2023/06/04 01:31:48 - mmengine - INFO - Epoch(train) [8][2800/3907] lr: 2.1092e-05 eta: 1:34:02 time: 0.6309 data_time: 0.0020 memory: 44138 loss: 0.3734 +2023/06/04 01:32:51 - mmengine - INFO - Epoch(train) [8][2900/3907] lr: 2.0855e-05 eta: 1:32:58 time: 0.6321 data_time: 0.0016 memory: 44138 loss: 0.3565 +2023/06/04 01:33:54 - mmengine - INFO - Epoch(train) [8][3000/3907] lr: 2.0621e-05 eta: 1:31:55 time: 0.6318 data_time: 0.0027 memory: 44138 loss: 0.3870 +2023/06/04 01:34:57 - mmengine - INFO - Epoch(train) [8][3100/3907] lr: 2.0388e-05 eta: 1:30:52 time: 0.6310 data_time: 0.0017 memory: 44138 loss: 0.3577 +2023/06/04 01:36:00 - mmengine - INFO - Epoch(train) [8][3200/3907] lr: 2.0158e-05 eta: 1:29:49 time: 0.6307 data_time: 0.0015 memory: 44138 loss: 0.3523 +2023/06/04 01:37:03 - mmengine - INFO - Epoch(train) [8][3300/3907] lr: 1.9930e-05 eta: 1:28:45 time: 0.6305 data_time: 0.0015 memory: 44138 loss: 0.3910 +2023/06/04 01:38:06 - mmengine - INFO - Epoch(train) [8][3400/3907] lr: 1.9705e-05 eta: 1:27:42 time: 0.6305 data_time: 0.0018 memory: 44138 loss: 0.3844 +2023/06/04 01:39:10 - mmengine - INFO - Epoch(train) [8][3500/3907] lr: 1.9481e-05 eta: 1:26:39 time: 0.6309 data_time: 0.0021 memory: 44138 loss: 0.3644 diff --git a/clip_large_pretrain_4x256_sdv2_lr1e-4_rub_/20230603_200211/vis_data/20230603_200211.json b/clip_large_pretrain_4x256_sdv2_lr1e-4_rub_/20230603_200211/vis_data/20230603_200211.json new file mode 100644 index 0000000000000000000000000000000000000000..d9f552c019633a71994d9bb74d2c5664c92d7532 --- /dev/null +++ b/clip_large_pretrain_4x256_sdv2_lr1e-4_rub_/20230603_200211/vis_data/20230603_200211.json @@ -0,0 +1,315 @@ +{"lr": 9.999857418382631e-05, "data_time": 0.0021013259887695313, "loss": 0.5254676818847657, "time": 0.6303915500640869, "epoch": 1, "memory": 44139, "step": 100} +{"lr": 9.999423907359506e-05, "data_time": 0.001608133316040039, "loss": 0.49592080116271975, "time": 0.6308860301971435, "epoch": 1, "memory": 44139, "step": 200} +{"lr": 9.998699480412073e-05, "data_time": 0.0016800165176391602, "loss": 0.47416927218437194, "time": 0.631201982498169, "epoch": 1, "memory": 44139, "step": 300} +{"lr": 9.997684184379085e-05, "data_time": 0.001680898666381836, "loss": 0.4523438036441803, "time": 0.6304768800735474, "epoch": 1, "memory": 44139, "step": 400} +{"lr": 9.996378084905784e-05, "data_time": 0.0017073631286621093, "loss": 0.4551931142807007, "time": 0.6323943138122559, "epoch": 1, "memory": 44139, "step": 500} +{"lr": 9.994781266439667e-05, "data_time": 0.0019591093063354493, "loss": 0.44729283452033997, "time": 0.6456096410751343, "epoch": 1, "memory": 44139, "step": 600} +{"lr": 9.992893832225061e-05, "data_time": 0.0015645503997802734, "loss": 0.4602384209632874, "time": 0.6309844970703125, "epoch": 1, "memory": 44139, "step": 700} +{"lr": 9.990715904296405e-05, "data_time": 0.002456402778625488, "loss": 0.4575894296169281, "time": 0.6301848411560058, "epoch": 1, "memory": 44139, "step": 800} +{"lr": 9.988247623470371e-05, "data_time": 0.001614689826965332, "loss": 0.4487850069999695, "time": 0.6298971891403198, "epoch": 1, "memory": 44139, "step": 900} +{"lr": 9.985489149336749e-05, "data_time": 0.0017919540405273438, "loss": 0.45876056551933286, "time": 0.6338362693786621, "epoch": 1, "memory": 44139, "step": 1000} +{"lr": 9.982440660248187e-05, "data_time": 0.0015303611755371094, "loss": 0.43976943790912626, "time": 0.6301945209503174, "epoch": 1, "memory": 44139, "step": 1100} +{"lr": 9.979102353308574e-05, "data_time": 0.0018020153045654296, "loss": 0.4259138166904449, "time": 0.6292660236358643, "epoch": 1, "memory": 44139, "step": 1200} +{"lr": 9.975474444360358e-05, "data_time": 0.0016472339630126953, "loss": 0.4358592242002487, "time": 0.6309081792831421, "epoch": 1, "memory": 44139, "step": 1300} +{"lr": 9.97155716797056e-05, "data_time": 0.001531839370727539, "loss": 0.44602505266666415, "time": 0.6301747322082519, "epoch": 1, "memory": 44139, "step": 1400} +{"lr": 9.967350777415632e-05, "data_time": 0.0019945621490478514, "loss": 0.43150107860565184, "time": 0.6295226812362671, "epoch": 1, "memory": 44139, "step": 1500} +{"lr": 9.962855544665059e-05, "data_time": 0.0017927885055541992, "loss": 0.4311340630054474, "time": 0.6293833255767822, "epoch": 1, "memory": 44139, "step": 1600} +{"lr": 9.958071760363773e-05, "data_time": 0.0020589590072631835, "loss": 0.41964859068393706, "time": 0.6303755044937134, "epoch": 1, "memory": 44139, "step": 1700} +{"lr": 9.952999733813382e-05, "data_time": 0.0015864372253417969, "loss": 0.4246463358402252, "time": 0.6298433542251587, "epoch": 1, "memory": 44139, "step": 1800} +{"lr": 9.947639792952148e-05, "data_time": 0.0016413688659667968, "loss": 0.40548085868358613, "time": 0.6293341636657714, "epoch": 1, "memory": 44139, "step": 1900} +{"lr": 9.941992284333813e-05, "data_time": 0.0018304824829101563, "loss": 0.43340319097042085, "time": 0.6292242527008056, "epoch": 1, "memory": 44139, "step": 2000} +{"lr": 9.93605757310516e-05, "data_time": 0.0016559600830078126, "loss": 0.43370539247989653, "time": 0.6301441192626953, "epoch": 1, "memory": 44139, "step": 2100} +{"lr": 9.929836042982421e-05, "data_time": 0.0016306638717651367, "loss": 0.40694511234760283, "time": 0.6304590463638305, "epoch": 1, "memory": 44139, "step": 2200} +{"lr": 9.923328096226467e-05, "data_time": 0.0021354198455810548, "loss": 0.41349645256996154, "time": 0.6303184509277344, "epoch": 1, "memory": 44139, "step": 2300} +{"lr": 9.916534153616819e-05, "data_time": 0.0018451690673828125, "loss": 0.41379840970039367, "time": 0.6308378458023072, "epoch": 1, "memory": 44139, "step": 2400} +{"lr": 9.909454654424387e-05, "data_time": 0.0016844034194946288, "loss": 0.41174737811088563, "time": 0.6300135850906372, "epoch": 1, "memory": 44139, "step": 2500} +{"lr": 9.902090056383109e-05, "data_time": 0.0016065359115600586, "loss": 0.4248443007469177, "time": 0.6317947149276734, "epoch": 1, "memory": 44139, "step": 2600} +{"lr": 9.894440835660365e-05, "data_time": 0.0017657518386840821, "loss": 0.4310387372970581, "time": 0.6297615766525269, "epoch": 1, "memory": 44139, "step": 2700} +{"lr": 9.886507486826162e-05, "data_time": 0.001495027542114258, "loss": 0.41254869401454924, "time": 0.630182147026062, "epoch": 1, "memory": 44139, "step": 2800} +{"lr": 9.878290522821154e-05, "data_time": 0.0017021417617797852, "loss": 0.38445100784301756, "time": 0.6294927358627319, "epoch": 1, "memory": 44139, "step": 2900} +{"lr": 9.86979047492352e-05, "data_time": 0.002216958999633789, "loss": 0.4417062133550644, "time": 0.6311541795730591, "epoch": 1, "memory": 44139, "step": 3000} +{"lr": 9.861007892714578e-05, "data_time": 0.0016395330429077148, "loss": 0.4045764863491058, "time": 0.6297856569290161, "epoch": 1, "memory": 44139, "step": 3100} +{"lr": 9.851943344043239e-05, "data_time": 0.0015326976776123048, "loss": 0.4071483612060547, "time": 0.6301517009735107, "epoch": 1, "memory": 44139, "step": 3200} +{"lr": 9.842597414989319e-05, "data_time": 0.001749730110168457, "loss": 0.41237278282642365, "time": 0.6304476499557495, "epoch": 1, "memory": 44139, "step": 3300} +{"lr": 9.832970709825637e-05, "data_time": 0.001751875877380371, "loss": 0.40038027465343473, "time": 0.6293846130371094, "epoch": 1, "memory": 44139, "step": 3400} +{"lr": 9.823063850978968e-05, "data_time": 0.0021019458770751955, "loss": 0.4185768276453018, "time": 0.6309142112731934, "epoch": 1, "memory": 44139, "step": 3500} +{"lr": 9.81287747898975e-05, "data_time": 0.0016637563705444336, "loss": 0.3904751867055893, "time": 0.6299480676651001, "epoch": 1, "memory": 44139, "step": 3600} +{"lr": 9.802412252470715e-05, "data_time": 0.0018462657928466797, "loss": 0.4244962424039841, "time": 0.6292508125305176, "epoch": 1, "memory": 44139, "step": 3700} +{"lr": 9.791668848064258e-05, "data_time": 0.001714801788330078, "loss": 0.4359990656375885, "time": 0.6293673038482666, "epoch": 1, "memory": 44139, "step": 3800} +{"lr": 9.780647960398751e-05, "data_time": 0.001652693748474121, "loss": 0.4074475705623627, "time": 0.6290539026260376, "epoch": 1, "memory": 44139, "step": 3900} +{"accuracy/top1": 78.65766906738281, "data_time": 0.0428013801574707, "time": 1.3483564602701288, "step": 1} +{"lr": 9.7685491196646e-05, "data_time": 0.001947164535522461, "loss": 0.38701715171337125, "time": 0.6323011636734008, "epoch": 2, "memory": 44139, "step": 4007} +{"lr": 9.756956126085676e-05, "data_time": 0.0019211292266845704, "loss": 0.38512068390846255, "time": 0.6309863805770874, "epoch": 2, "memory": 44138, "step": 4107} +{"lr": 9.745087893642482e-05, "data_time": 0.0017422199249267577, "loss": 0.377513712644577, "time": 0.6308529138565063, "epoch": 2, "memory": 44138, "step": 4207} +{"lr": 9.73294518969056e-05, "data_time": 0.0013633489608764649, "loss": 0.40180844366550444, "time": 0.6309009790420532, "epoch": 2, "memory": 44138, "step": 4307} +{"lr": 9.72052879933172e-05, "data_time": 0.0016982555389404297, "loss": 0.4086743950843811, "time": 0.6303865432739257, "epoch": 2, "memory": 44138, "step": 4407} +{"lr": 9.707839525363358e-05, "data_time": 0.0026702642440795898, "loss": 0.38177225589752195, "time": 0.6331295728683471, "epoch": 2, "memory": 44138, "step": 4507} +{"lr": 9.69487818822646e-05, "data_time": 0.002133893966674805, "loss": 0.40716169476509095, "time": 0.6305238485336304, "epoch": 2, "memory": 44138, "step": 4607} +{"lr": 9.681645625952616e-05, "data_time": 0.0016930580139160156, "loss": 0.3874299794435501, "time": 0.6302786588668823, "epoch": 2, "memory": 44138, "step": 4707} +{"lr": 9.668142694109826e-05, "data_time": 0.0017039060592651367, "loss": 0.3736038327217102, "time": 0.631012749671936, "epoch": 2, "memory": 44138, "step": 4807} +{"lr": 9.654370265747162e-05, "data_time": 0.0016846179962158204, "loss": 0.3739929348230362, "time": 0.6311726570129395, "epoch": 2, "memory": 44138, "step": 4907} +{"lr": 9.640329231338345e-05, "data_time": 0.0018583297729492187, "loss": 0.3880970239639282, "time": 0.6307178497314453, "epoch": 2, "memory": 44138, "step": 5007} +{"lr": 9.626020498724147e-05, "data_time": 0.0020255327224731447, "loss": 0.39673975110054016, "time": 0.6308103799819946, "epoch": 2, "memory": 44138, "step": 5107} +{"lr": 9.611444993053717e-05, "data_time": 0.0020781755447387695, "loss": 0.40036986470222474, "time": 0.6299102783203125, "epoch": 2, "memory": 44138, "step": 5207} +{"lr": 9.596603656724746e-05, "data_time": 0.0018322467803955078, "loss": 0.4005657464265823, "time": 0.6303497076034545, "epoch": 2, "memory": 44138, "step": 5307} +{"lr": 9.58149744932254e-05, "data_time": 0.0016456842422485352, "loss": 0.39048443734645844, "time": 0.6305620908737183, "epoch": 2, "memory": 44138, "step": 5407} +{"lr": 9.566127347557972e-05, "data_time": 0.001516413688659668, "loss": 0.3764130622148514, "time": 0.6307580232620239, "epoch": 2, "memory": 44138, "step": 5507} +{"lr": 9.550494345204367e-05, "data_time": 0.0022286176681518555, "loss": 0.38684943616390227, "time": 0.6312837362289428, "epoch": 2, "memory": 44138, "step": 5607} +{"lr": 9.534599453033183e-05, "data_time": 0.0015774965286254883, "loss": 0.36203564405441285, "time": 0.6302300691604614, "epoch": 2, "memory": 44138, "step": 5707} +{"lr": 9.518443698748712e-05, "data_time": 0.0018428802490234376, "loss": 0.36771643459796904, "time": 0.6324694395065308, "epoch": 2, "memory": 44138, "step": 5807} +{"lr": 9.50202812692162e-05, "data_time": 0.0027725934982299806, "loss": 0.3563918322324753, "time": 0.6313409566879272, "epoch": 2, "memory": 44138, "step": 5907} +{"lr": 9.485353798921386e-05, "data_time": 0.0018866300582885743, "loss": 0.3846891909837723, "time": 0.6306589126586915, "epoch": 2, "memory": 44138, "step": 6007} +{"lr": 9.468421792847712e-05, "data_time": 0.0025237083435058595, "loss": 0.3983618408441544, "time": 0.6313921451568604, "epoch": 2, "memory": 44138, "step": 6107} +{"lr": 9.451233203460794e-05, "data_time": 0.0019637584686279298, "loss": 0.3815899193286896, "time": 0.6315789222717285, "epoch": 2, "memory": 44138, "step": 6207} +{"lr": 9.433789142110532e-05, "data_time": 0.001621842384338379, "loss": 0.3959047496318817, "time": 0.6313664436340332, "epoch": 2, "memory": 44138, "step": 6307} +{"lr": 9.416090736664709e-05, "data_time": 0.0023834705352783203, "loss": 0.37512257099151614, "time": 0.6290892601013184, "epoch": 2, "memory": 44138, "step": 6407} +{"lr": 9.398139131436032e-05, "data_time": 0.0018195152282714845, "loss": 0.3631224125623703, "time": 0.6311782121658325, "epoch": 2, "memory": 44138, "step": 6507} +{"lr": 9.379935487108167e-05, "data_time": 0.0017686843872070312, "loss": 0.369792178273201, "time": 0.6317324876785279, "epoch": 2, "memory": 44138, "step": 6607} +{"lr": 9.361480980660697e-05, "data_time": 0.0020948410034179687, "loss": 0.3787175536155701, "time": 0.6312172174453735, "epoch": 2, "memory": 44138, "step": 6707} +{"lr": 9.342776805292993e-05, "data_time": 0.002226853370666504, "loss": 0.3891234517097473, "time": 0.6308389902114868, "epoch": 2, "memory": 44138, "step": 6807} +{"lr": 9.323824170347083e-05, "data_time": 0.0018370389938354493, "loss": 0.369640052318573, "time": 0.633220887184143, "epoch": 2, "memory": 44138, "step": 6907} +{"lr": 9.304624301229471e-05, "data_time": 0.002008342742919922, "loss": 0.3819348871707916, "time": 0.6316205024719238, "epoch": 2, "memory": 44138, "step": 7007} +{"lr": 9.285178439331897e-05, "data_time": 0.0016373157501220702, "loss": 0.38032921850681306, "time": 0.6298771381378174, "epoch": 2, "memory": 44138, "step": 7107} +{"lr": 9.265487841951087e-05, "data_time": 0.0017477750778198242, "loss": 0.3724153459072113, "time": 0.6307685136795044, "epoch": 2, "memory": 44138, "step": 7207} +{"lr": 9.245553782207421e-05, "data_time": 0.002009415626525879, "loss": 0.36185015439987184, "time": 0.6313466310501099, "epoch": 2, "memory": 44138, "step": 7307} +{"lr": 9.225377548962671e-05, "data_time": 0.0015581607818603515, "loss": 0.3673944413661957, "time": 0.6307828187942505, "epoch": 2, "memory": 44138, "step": 7407} +{"lr": 9.204960446736601e-05, "data_time": 0.0017274379730224609, "loss": 0.3907669007778168, "time": 0.6304670333862304, "epoch": 2, "memory": 44138, "step": 7507} +{"lr": 9.184303795622715e-05, "data_time": 0.0018530845642089843, "loss": 0.4112739682197571, "time": 0.6308581829071045, "epoch": 2, "memory": 44138, "step": 7607} +{"lr": 9.163408931202823e-05, "data_time": 0.002285194396972656, "loss": 0.39737538397312167, "time": 0.6308098316192627, "epoch": 2, "memory": 44138, "step": 7707} +{"lr": 9.142277204460729e-05, "data_time": 0.0014277219772338868, "loss": 0.379548117518425, "time": 0.6297266244888305, "epoch": 2, "memory": 44138, "step": 7807} +{"accuracy/top1": 78.72029876708984, "data_time": 0.03160543277345855, "time": 1.2824266820118344, "step": 2} +{"lr": 9.119405492383288e-05, "data_time": 0.0019155025482177734, "loss": 0.38339483737945557, "time": 0.6309638738632202, "epoch": 3, "memory": 44138, "step": 7914} +{"lr": 9.097787819233205e-05, "data_time": 0.001637101173400879, "loss": 0.39207392930984497, "time": 0.6319971561431885, "epoch": 3, "memory": 44138, "step": 8014} +{"lr": 9.075937526576627e-05, "data_time": 0.0020543336868286133, "loss": 0.3818666934967041, "time": 0.6318645238876343, "epoch": 3, "memory": 44138, "step": 8114} +{"lr": 9.053856027171755e-05, "data_time": 0.001692509651184082, "loss": 0.38909188807010653, "time": 0.630961012840271, "epoch": 3, "memory": 44138, "step": 8214} +{"lr": 9.031544748725785e-05, "data_time": 0.001746964454650879, "loss": 0.36356093883514407, "time": 0.6317759275436401, "epoch": 3, "memory": 44138, "step": 8314} +{"lr": 9.009005133802542e-05, "data_time": 0.0013951778411865235, "loss": 0.3986783742904663, "time": 0.6313511371612549, "epoch": 3, "memory": 44138, "step": 8414} +{"lr": 8.986238639729242e-05, "data_time": 0.0020413875579833986, "loss": 0.3918088436126709, "time": 0.6310792446136475, "epoch": 3, "memory": 44138, "step": 8514} +{"lr": 8.963246738502254e-05, "data_time": 0.0020148038864135744, "loss": 0.36713131368160246, "time": 0.6311418771743774, "epoch": 3, "memory": 44138, "step": 8614} +{"lr": 8.940030916691928e-05, "data_time": 0.0019838571548461913, "loss": 0.36795192062854765, "time": 0.6313153982162476, "epoch": 3, "memory": 44138, "step": 8714} +{"lr": 8.91659267534649e-05, "data_time": 0.0019896984100341796, "loss": 0.3668671786785126, "time": 0.6314969062805176, "epoch": 3, "memory": 44138, "step": 8814} +{"lr": 8.892933529894978e-05, "data_time": 0.0017478227615356444, "loss": 0.3891805768013, "time": 0.6318260908126831, "epoch": 3, "memory": 44138, "step": 8914} +{"lr": 8.869055010049283e-05, "data_time": 0.0015655279159545899, "loss": 0.40365838408470156, "time": 0.6307106256484986, "epoch": 3, "memory": 44138, "step": 9014} +{"lr": 8.844958659705188e-05, "data_time": 0.0014819622039794922, "loss": 0.35741288065910337, "time": 0.6319898366928101, "epoch": 3, "memory": 44138, "step": 9114} +{"lr": 8.82064603684259e-05, "data_time": 0.0020102739334106447, "loss": 0.3577596008777618, "time": 0.6315583705902099, "epoch": 3, "memory": 44138, "step": 9214} +{"lr": 8.796118713424776e-05, "data_time": 0.0019507884979248046, "loss": 0.3682455807924271, "time": 0.6324578762054444, "epoch": 3, "memory": 44138, "step": 9314} +{"lr": 8.77137827529676e-05, "data_time": 0.0018065452575683593, "loss": 0.39598606526851654, "time": 0.6308689117431641, "epoch": 3, "memory": 44138, "step": 9414} +{"lr": 8.746426322082745e-05, "data_time": 0.0014045953750610352, "loss": 0.3769483804702759, "time": 0.6309690237045288, "epoch": 3, "memory": 44138, "step": 9514} +{"lr": 8.721264467082732e-05, "data_time": 0.0022228717803955077, "loss": 0.3582300215959549, "time": 0.6320258617401123, "epoch": 3, "memory": 44138, "step": 9614} +{"lr": 8.695894337168168e-05, "data_time": 0.0019613027572631834, "loss": 0.3443090319633484, "time": 0.6312095642089843, "epoch": 3, "memory": 44138, "step": 9714} +{"lr": 8.67031757267677e-05, "data_time": 0.0016478776931762695, "loss": 0.3798891544342041, "time": 0.6311641454696655, "epoch": 3, "memory": 44138, "step": 9814} +{"lr": 8.64453582730651e-05, "data_time": 0.001678299903869629, "loss": 0.38648802042007446, "time": 0.6316426038742066, "epoch": 3, "memory": 44138, "step": 9914} +{"lr": 8.61855076800861e-05, "data_time": 0.0014892578125, "loss": 0.37505062818527224, "time": 0.6309507846832275, "epoch": 3, "memory": 44138, "step": 10014} +{"lr": 8.592364074879843e-05, "data_time": 0.0017383575439453125, "loss": 0.35937183499336245, "time": 0.630468225479126, "epoch": 3, "memory": 44138, "step": 10114} +{"lr": 8.565977441053854e-05, "data_time": 0.001960349082946777, "loss": 0.4062574774026871, "time": 0.6306670665740967, "epoch": 3, "memory": 44138, "step": 10214} +{"lr": 8.539392572591724e-05, "data_time": 0.001788496971130371, "loss": 0.34328748285770416, "time": 0.6322952032089233, "epoch": 3, "memory": 44138, "step": 10314} +{"lr": 8.512611188371633e-05, "data_time": 0.0020530223846435547, "loss": 0.3801038235425949, "time": 0.6303425788879394, "epoch": 3, "memory": 44138, "step": 10414} +{"lr": 8.485635019977725e-05, "data_time": 0.001952791213989258, "loss": 0.4016746997833252, "time": 0.6302950620651245, "epoch": 3, "memory": 44138, "step": 10514} +{"lr": 8.45846581158818e-05, "data_time": 0.002126765251159668, "loss": 0.37341198325157166, "time": 0.6310628175735473, "epoch": 3, "memory": 44138, "step": 10614} +{"lr": 8.431105319862416e-05, "data_time": 0.0019888877868652344, "loss": 0.3901614427566528, "time": 0.6303201675415039, "epoch": 3, "memory": 44138, "step": 10714} +{"lr": 8.403555313827516e-05, "data_time": 0.0016383886337280273, "loss": 0.37818117439746857, "time": 0.630945348739624, "epoch": 3, "memory": 44138, "step": 10814} +{"lr": 8.375817574763844e-05, "data_time": 0.0017586231231689453, "loss": 0.36275345683097837, "time": 0.6306392908096313, "epoch": 3, "memory": 44138, "step": 10914} +{"lr": 8.347893896089871e-05, "data_time": 0.0022778034210205076, "loss": 0.37881468832492826, "time": 0.6306608438491821, "epoch": 3, "memory": 44138, "step": 11014} +{"lr": 8.31978608324625e-05, "data_time": 0.001977705955505371, "loss": 0.37366268038749695, "time": 0.6312607288360595, "epoch": 3, "memory": 44138, "step": 11114} +{"lr": 8.291495953579052e-05, "data_time": 0.0017975807189941407, "loss": 0.3661737948656082, "time": 0.6308862447738648, "epoch": 3, "memory": 44138, "step": 11214} +{"lr": 8.263025336222267e-05, "data_time": 0.0014571666717529297, "loss": 0.35999304354190825, "time": 0.6310614109039306, "epoch": 3, "memory": 44138, "step": 11314} +{"lr": 8.234376071979551e-05, "data_time": 0.0018079280853271484, "loss": 0.38005903363227844, "time": 0.6313439607620239, "epoch": 3, "memory": 44138, "step": 11414} +{"lr": 8.205550013205187e-05, "data_time": 0.0025040388107299806, "loss": 0.3770012348890305, "time": 0.631182336807251, "epoch": 3, "memory": 44138, "step": 11514} +{"lr": 8.17654902368435e-05, "data_time": 0.0018376588821411132, "loss": 0.37870813608169557, "time": 0.6309926271438598, "epoch": 3, "memory": 44138, "step": 11614} +{"lr": 8.147374978512582e-05, "data_time": 0.0015785694122314453, "loss": 0.36322163343429564, "time": 0.6306546211242676, "epoch": 3, "memory": 44138, "step": 11714} +{"accuracy/top1": 77.37378692626953, "data_time": 0.031213887806596428, "time": 1.2824072385656422, "step": 3} +{"lr": 8.115969237626578e-05, "data_time": 0.001883554458618164, "loss": 0.34975853860378264, "time": 0.6301671504974365, "epoch": 4, "memory": 44138, "step": 11821} +{"lr": 8.08644297337245e-05, "data_time": 0.0018700361251831055, "loss": 0.3540323615074158, "time": 0.6305224418640136, "epoch": 4, "memory": 44138, "step": 11921} +{"lr": 8.056749479387651e-05, "data_time": 0.0018409490585327148, "loss": 0.35627412497997285, "time": 0.6311853170394898, "epoch": 4, "memory": 44138, "step": 12021} +{"lr": 8.026890675542457e-05, "data_time": 0.0019524097442626953, "loss": 0.35569603741168976, "time": 0.6314916849136353, "epoch": 4, "memory": 44138, "step": 12121} +{"lr": 7.996868492395457e-05, "data_time": 0.0015810728073120117, "loss": 0.35987875163555144, "time": 0.6313385248184205, "epoch": 4, "memory": 44138, "step": 12221} +{"lr": 7.966684871068762e-05, "data_time": 0.0020611763000488283, "loss": 0.37253828942775724, "time": 0.6317244291305542, "epoch": 4, "memory": 44138, "step": 12321} +{"lr": 7.936341763122449e-05, "data_time": 0.0016445159912109376, "loss": 0.3688558876514435, "time": 0.6316900968551635, "epoch": 4, "memory": 44138, "step": 12421} +{"lr": 7.905841130428405e-05, "data_time": 0.001904129981994629, "loss": 0.3982546806335449, "time": 0.6310122728347778, "epoch": 4, "memory": 44138, "step": 12521} +{"lr": 7.875184945043487e-05, "data_time": 0.0019101142883300782, "loss": 0.3493445962667465, "time": 0.6312102794647216, "epoch": 4, "memory": 44138, "step": 12621} +{"lr": 7.844375189082012e-05, "data_time": 0.0018593072891235352, "loss": 0.3826897144317627, "time": 0.6311030864715577, "epoch": 4, "memory": 44138, "step": 12721} +{"lr": 7.813413854587572e-05, "data_time": 0.0017517566680908202, "loss": 0.3819606751203537, "time": 0.6317952632904053, "epoch": 4, "memory": 44138, "step": 12821} +{"lr": 7.782302943404291e-05, "data_time": 0.0017009258270263671, "loss": 0.37396695017814635, "time": 0.632053542137146, "epoch": 4, "memory": 44138, "step": 12921} +{"lr": 7.751044467047337e-05, "data_time": 0.001685023307800293, "loss": 0.39482679665088655, "time": 0.6310154438018799, "epoch": 4, "memory": 44138, "step": 13021} +{"lr": 7.719640446572907e-05, "data_time": 0.0017710208892822265, "loss": 0.3653447687625885, "time": 0.6302547931671143, "epoch": 4, "memory": 44138, "step": 13121} +{"lr": 7.688092912447536e-05, "data_time": 0.0024473190307617186, "loss": 0.3717967987060547, "time": 0.6365467786788941, "epoch": 4, "memory": 44138, "step": 13221} +{"lr": 7.656403904416794e-05, "data_time": 0.0021691083908081054, "loss": 0.37738717496395113, "time": 0.6322481155395507, "epoch": 4, "memory": 44138, "step": 13321} +{"lr": 7.624575471373434e-05, "data_time": 0.002275562286376953, "loss": 0.37544898986816405, "time": 0.6308626651763916, "epoch": 4, "memory": 44138, "step": 13421} +{"lr": 7.592609671224926e-05, "data_time": 0.0018621444702148437, "loss": 0.3497780740261078, "time": 0.6322161674499511, "epoch": 4, "memory": 44138, "step": 13521} +{"lr": 7.560508570760375e-05, "data_time": 0.0016445159912109376, "loss": 0.378781920671463, "time": 0.6311922311782837, "epoch": 4, "memory": 44138, "step": 13621} +{"lr": 7.528274245516895e-05, "data_time": 0.0019595146179199217, "loss": 0.38113327622413634, "time": 0.6312441825866699, "epoch": 4, "memory": 44138, "step": 13721} +{"lr": 7.495908779645405e-05, "data_time": 0.001761150360107422, "loss": 0.367531618475914, "time": 0.6302317619323731, "epoch": 4, "memory": 44138, "step": 13821} +{"lr": 7.463414265775906e-05, "data_time": 0.001967191696166992, "loss": 0.35146708190441134, "time": 0.6400554418563843, "epoch": 4, "memory": 44138, "step": 13921} +{"lr": 7.430792804882147e-05, "data_time": 0.0016878366470336914, "loss": 0.3615738064050674, "time": 0.6307236909866333, "epoch": 4, "memory": 44138, "step": 14021} +{"lr": 7.398046506145803e-05, "data_time": 0.0015439510345458985, "loss": 0.3717303067445755, "time": 0.6305463552474976, "epoch": 4, "memory": 44138, "step": 14121} +{"lr": 7.365177486820114e-05, "data_time": 0.0015390872955322265, "loss": 0.40721426606178285, "time": 0.630876612663269, "epoch": 4, "memory": 44138, "step": 14221} +{"lr": 7.332187872092939e-05, "data_time": 0.0018485069274902343, "loss": 0.3763453930616379, "time": 0.6308673620223999, "epoch": 4, "memory": 44138, "step": 14321} +{"lr": 7.299079794949401e-05, "data_time": 0.0017867565155029296, "loss": 0.36380934417247773, "time": 0.6306280612945556, "epoch": 4, "memory": 44138, "step": 14421} +{"lr": 7.265855396033955e-05, "data_time": 0.0018984079360961914, "loss": 0.3613135159015656, "time": 0.6310559034347534, "epoch": 4, "memory": 44138, "step": 14521} +{"lr": 7.232516823512013e-05, "data_time": 0.001582193374633789, "loss": 0.37295305728912354, "time": 0.6325780630111695, "epoch": 4, "memory": 44138, "step": 14621} +{"lr": 7.199066232930995e-05, "data_time": 0.0018008947372436523, "loss": 0.371643990278244, "time": 0.6305253982543946, "epoch": 4, "memory": 44138, "step": 14721} +{"lr": 7.165505787080997e-05, "data_time": 0.0017313241958618164, "loss": 0.3901374012231827, "time": 0.6315348386764527, "epoch": 4, "memory": 44138, "step": 14821} +{"lr": 7.13183765585495e-05, "data_time": 0.0023801326751708984, "loss": 0.3802525758743286, "time": 0.63101806640625, "epoch": 4, "memory": 44138, "step": 14921} +{"lr": 7.098064016108317e-05, "data_time": 0.0014557600021362304, "loss": 0.372085177898407, "time": 0.6336552143096924, "epoch": 4, "memory": 44138, "step": 15021} +{"lr": 7.064187051518379e-05, "data_time": 0.0014697790145874023, "loss": 0.3566492527723312, "time": 0.6312332630157471, "epoch": 4, "memory": 44138, "step": 15121} +{"lr": 7.030208952442949e-05, "data_time": 0.001665806770324707, "loss": 0.37336931228637693, "time": 0.6538846969604493, "epoch": 4, "memory": 44138, "step": 15221} +{"lr": 6.996131915778851e-05, "data_time": 0.0020270586013793946, "loss": 0.3725410372018814, "time": 0.6306520223617553, "epoch": 4, "memory": 44138, "step": 15321} +{"lr": 6.961958144819853e-05, "data_time": 0.001505136489868164, "loss": 0.38264588713645936, "time": 0.6309946298599243, "epoch": 4, "memory": 44138, "step": 15421} +{"lr": 6.927689849114193e-05, "data_time": 0.002476620674133301, "loss": 0.383384981751442, "time": 0.6316012144088745, "epoch": 4, "memory": 44138, "step": 15521} +{"lr": 6.893329244321722e-05, "data_time": 0.001361536979675293, "loss": 0.3943897247314453, "time": 0.6302694797515869, "epoch": 4, "memory": 44138, "step": 15621} +{"accuracy/top1": 76.42740631103516, "data_time": 0.03199244778731774, "time": 1.2822754341980507, "step": 4} +{"lr": 6.856463687361792e-05, "data_time": 0.0015532493591308594, "loss": 0.37046386301517487, "time": 0.6303340911865234, "epoch": 5, "memory": 44138, "step": 15728} +{"lr": 6.821919068470432e-05, "data_time": 0.0017634153366088868, "loss": 0.3431585133075714, "time": 0.6313709020614624, "epoch": 5, "memory": 44138, "step": 15828} +{"lr": 6.787288979235407e-05, "data_time": 0.0015941619873046874, "loss": 0.3947469711303711, "time": 0.6309973001480103, "epoch": 5, "memory": 44138, "step": 15928} +{"lr": 6.752575658708793e-05, "data_time": 0.0019409656524658203, "loss": 0.3779486268758774, "time": 0.6306203842163086, "epoch": 5, "memory": 44138, "step": 16028} +{"lr": 6.71778135132408e-05, "data_time": 0.00177457332611084, "loss": 0.38871109783649443, "time": 0.630304479598999, "epoch": 5, "memory": 44138, "step": 16128} +{"lr": 6.682908306751074e-05, "data_time": 0.0024759531021118163, "loss": 0.35672386884689333, "time": 0.630770468711853, "epoch": 5, "memory": 44138, "step": 16228} +{"lr": 6.64795877975043e-05, "data_time": 0.002137279510498047, "loss": 0.36641431152820586, "time": 0.631433653831482, "epoch": 5, "memory": 44138, "step": 16328} +{"lr": 6.61293503002787e-05, "data_time": 0.0019015073776245117, "loss": 0.35590389668941497, "time": 0.6321771383285523, "epoch": 5, "memory": 44138, "step": 16428} +{"lr": 6.577839322088084e-05, "data_time": 0.0013942718505859375, "loss": 0.3608432561159134, "time": 0.6326557636260987, "epoch": 5, "memory": 44138, "step": 16528} +{"lr": 6.542673925088298e-05, "data_time": 0.001990056037902832, "loss": 0.37712846100330355, "time": 0.631695556640625, "epoch": 5, "memory": 44138, "step": 16628} +{"lr": 6.507441112691584e-05, "data_time": 0.0013965845108032226, "loss": 0.3802467107772827, "time": 0.6321236610412597, "epoch": 5, "memory": 44138, "step": 16728} +{"lr": 6.47214316291983e-05, "data_time": 0.0018872261047363282, "loss": 0.36903893053531645, "time": 0.6317490100860595, "epoch": 5, "memory": 44138, "step": 16828} +{"lr": 6.436782358006481e-05, "data_time": 0.0019455909729003905, "loss": 0.34476652145385744, "time": 0.6384008646011352, "epoch": 5, "memory": 44138, "step": 16928} +{"lr": 6.40136098424896e-05, "data_time": 0.00171051025390625, "loss": 0.3551800101995468, "time": 0.6332657098770141, "epoch": 5, "memory": 44138, "step": 17028} +{"lr": 6.365881331860828e-05, "data_time": 0.0017849206924438477, "loss": 0.3882934868335724, "time": 0.6307220220565796, "epoch": 5, "memory": 44138, "step": 17128} +{"lr": 6.330345694823739e-05, "data_time": 0.0021819353103637697, "loss": 0.36206912696361543, "time": 0.6309105157852173, "epoch": 5, "memory": 44138, "step": 17228} +{"lr": 6.29475637073908e-05, "data_time": 0.0017885923385620116, "loss": 0.37085449546575544, "time": 0.6303831338882446, "epoch": 5, "memory": 44138, "step": 17328} +{"lr": 6.259115660679482e-05, "data_time": 0.0017007827758789063, "loss": 0.3885393291711807, "time": 0.6403779983520508, "epoch": 5, "memory": 44138, "step": 17428} +{"lr": 6.223425869039963e-05, "data_time": 0.002181410789489746, "loss": 0.3782466769218445, "time": 0.6303816556930542, "epoch": 5, "memory": 44138, "step": 17528} +{"lr": 6.187689303389006e-05, "data_time": 0.002347421646118164, "loss": 0.3912949591875076, "time": 0.631197190284729, "epoch": 5, "memory": 44138, "step": 17628} +{"lr": 6.151908274319292e-05, "data_time": 0.0025192022323608397, "loss": 0.35448630303144457, "time": 0.6313546419143676, "epoch": 5, "memory": 44138, "step": 17728} +{"lr": 6.116085095298373e-05, "data_time": 0.001743459701538086, "loss": 0.3646813243627548, "time": 0.6316289901733398, "epoch": 5, "memory": 44138, "step": 17828} +{"lr": 6.080222082519007e-05, "data_time": 0.0020340442657470702, "loss": 0.3599780470132828, "time": 0.6305276155471802, "epoch": 5, "memory": 44138, "step": 17928} +{"lr": 6.0443215547495147e-05, "data_time": 0.0019669294357299804, "loss": 0.3566180944442749, "time": 0.6312304973602295, "epoch": 5, "memory": 44138, "step": 18028} +{"lr": 6.0083858331837714e-05, "data_time": 0.0015351057052612304, "loss": 0.38209750056266784, "time": 0.6313471317291259, "epoch": 5, "memory": 44138, "step": 18128} +{"lr": 5.97241724129116e-05, "data_time": 0.0016611099243164062, "loss": 0.3699777275323868, "time": 0.645628833770752, "epoch": 5, "memory": 44138, "step": 18228} +{"lr": 5.936418104666328e-05, "data_time": 0.0018738031387329102, "loss": 0.3627565443515778, "time": 0.6309934854507446, "epoch": 5, "memory": 44138, "step": 18328} +{"lr": 5.9003907508788295e-05, "data_time": 0.0014870643615722656, "loss": 0.374181854724884, "time": 0.6333391427993774, "epoch": 5, "memory": 44138, "step": 18428} +{"lr": 5.864337509322633e-05, "data_time": 0.0017216205596923828, "loss": 0.37459246814250946, "time": 0.6297661304473877, "epoch": 5, "memory": 44138, "step": 18528} +{"lr": 5.8282607110655194e-05, "data_time": 0.0015652179718017578, "loss": 0.3801063776016235, "time": 0.6301994800567627, "epoch": 5, "memory": 44138, "step": 18628} +{"lr": 5.7921626886983564e-05, "data_time": 0.0013895988464355468, "loss": 0.3741777092218399, "time": 0.6301333427429199, "epoch": 5, "memory": 44138, "step": 18728} +{"lr": 5.756045776184287e-05, "data_time": 0.0019222259521484374, "loss": 0.3763693571090698, "time": 0.6307146549224854, "epoch": 5, "memory": 44138, "step": 18828} +{"lr": 5.719912308707808e-05, "data_time": 0.002425527572631836, "loss": 0.38328993916511533, "time": 0.6307286977767944, "epoch": 5, "memory": 44138, "step": 18928} +{"lr": 5.683764622523808e-05, "data_time": 0.0015813589096069336, "loss": 0.3593562632799149, "time": 0.6307921409606934, "epoch": 5, "memory": 44138, "step": 19028} +{"lr": 5.647605054806502e-05, "data_time": 0.001833629608154297, "loss": 0.37703666687011717, "time": 0.6302602767944336, "epoch": 5, "memory": 44138, "step": 19128} +{"lr": 5.6114359434983206e-05, "data_time": 0.0016495466232299804, "loss": 0.379250505566597, "time": 0.6305471658706665, "epoch": 5, "memory": 44138, "step": 19228} +{"lr": 5.575259627158745e-05, "data_time": 0.0026583194732666014, "loss": 0.3611921459436417, "time": 0.6306925058364868, "epoch": 5, "memory": 44138, "step": 19328} +{"lr": 5.539078444813113e-05, "data_time": 0.0015018224716186524, "loss": 0.373409104347229, "time": 0.6310846328735351, "epoch": 5, "memory": 44138, "step": 19428} +{"lr": 5.5028947358013735e-05, "data_time": 0.0013743162155151368, "loss": 0.39499005377292634, "time": 0.6301813840866088, "epoch": 5, "memory": 44138, "step": 19528} +{"accuracy/top1": 75.8219985961914, "data_time": 0.031351796511946056, "time": 1.2812493381829098, "step": 5} +{"lr": 5.464178020340108e-05, "data_time": 0.0018122673034667968, "loss": 0.34920958876609803, "time": 0.6305671453475952, "epoch": 6, "memory": 44138, "step": 19635} +{"lr": 5.427996514795046e-05, "data_time": 0.001543140411376953, "loss": 0.36444980204105376, "time": 0.6307581424713135, "epoch": 6, "memory": 44138, "step": 19735} +{"lr": 5.3918196647260834e-05, "data_time": 0.0016943454742431641, "loss": 0.3770173192024231, "time": 0.6312621831893921, "epoch": 6, "memory": 44138, "step": 19835} +{"lr": 5.355649809193057e-05, "data_time": 0.0015299558639526368, "loss": 0.3631849020719528, "time": 0.6310431957244873, "epoch": 6, "memory": 44138, "step": 19935} +{"lr": 5.3194892868035765e-05, "data_time": 0.0016496419906616212, "loss": 0.3821670085191727, "time": 0.6306036710739136, "epoch": 6, "memory": 44138, "step": 20035} +{"lr": 5.283340435561795e-05, "data_time": 0.0016690492630004883, "loss": 0.3515541672706604, "time": 0.6410810708999634, "epoch": 6, "memory": 44138, "step": 20135} +{"lr": 5.2472055927172495e-05, "data_time": 0.001516270637512207, "loss": 0.37552843391895296, "time": 0.6313085794448853, "epoch": 6, "memory": 44138, "step": 20235} +{"lr": 5.2110870946137605e-05, "data_time": 0.0017879486083984375, "loss": 0.3542404919862747, "time": 0.6309683084487915, "epoch": 6, "memory": 44138, "step": 20335} +{"lr": 5.174987276538342e-05, "data_time": 0.0015929698944091796, "loss": 0.3852983355522156, "time": 0.6308411359786987, "epoch": 6, "memory": 44138, "step": 20435} +{"lr": 5.1389084725702425e-05, "data_time": 0.002024126052856445, "loss": 0.3785814821720123, "time": 0.6308462142944335, "epoch": 6, "memory": 44138, "step": 20535} +{"lr": 5.1028530154300045e-05, "data_time": 0.0017513036727905273, "loss": 0.3822389543056488, "time": 0.6312130689620972, "epoch": 6, "memory": 44138, "step": 20635} +{"lr": 5.066823236328663e-05, "data_time": 0.00167694091796875, "loss": 0.3927029311656952, "time": 0.631606912612915, "epoch": 6, "memory": 44138, "step": 20735} +{"lr": 5.030821464817006e-05, "data_time": 0.0017026662826538086, "loss": 0.37391828894615176, "time": 0.6300700664520263, "epoch": 6, "memory": 44138, "step": 20835} +{"lr": 4.99485002863494e-05, "data_time": 0.0016702651977539063, "loss": 0.38308541774749755, "time": 0.630565857887268, "epoch": 6, "memory": 44138, "step": 20935} +{"lr": 4.95891125356102e-05, "data_time": 0.001670098304748535, "loss": 0.38320526480674744, "time": 0.6307455062866211, "epoch": 6, "memory": 44138, "step": 21035} +{"lr": 4.923007463262053e-05, "data_time": 0.002085232734680176, "loss": 0.3893862396478653, "time": 0.6310501575469971, "epoch": 6, "memory": 44138, "step": 21135} +{"lr": 4.887140979142852e-05, "data_time": 0.0013954877853393556, "loss": 0.3912886053323746, "time": 0.6308657169342041, "epoch": 6, "memory": 44138, "step": 21235} +{"lr": 4.851314120196161e-05, "data_time": 0.0018996715545654297, "loss": 0.3496919900178909, "time": 0.6302508354187012, "epoch": 6, "memory": 44138, "step": 21335} +{"lr": 4.815529202852693e-05, "data_time": 0.0025632619857788087, "loss": 0.3953156232833862, "time": 0.6316039562225342, "epoch": 6, "memory": 44138, "step": 21435} +{"lr": 4.779788540831406e-05, "data_time": 0.0018422126770019532, "loss": 0.3585584104061127, "time": 0.6307833433151245, "epoch": 6, "memory": 44138, "step": 21535} +{"lr": 4.744094444989843e-05, "data_time": 0.0014423370361328126, "loss": 0.37024637758731843, "time": 0.6308498620986939, "epoch": 6, "memory": 44138, "step": 21635} +{"lr": 4.708449223174764e-05, "data_time": 0.001605224609375, "loss": 0.3841128945350647, "time": 0.6304964542388916, "epoch": 6, "memory": 44138, "step": 21735} +{"lr": 4.672855180072916e-05, "data_time": 0.001566171646118164, "loss": 0.36727446019649507, "time": 0.6305173635482788, "epoch": 6, "memory": 44138, "step": 21835} +{"lr": 4.6373146170620256e-05, "data_time": 0.00189361572265625, "loss": 0.36009582281112673, "time": 0.6301247596740722, "epoch": 6, "memory": 44138, "step": 21935} +{"lr": 4.6018298320619915e-05, "data_time": 0.0015498638153076173, "loss": 0.3688606321811676, "time": 0.630955696105957, "epoch": 6, "memory": 44138, "step": 22035} +{"lr": 4.566403119386318e-05, "data_time": 0.0015400409698486327, "loss": 0.39376863539218904, "time": 0.6312012910842896, "epoch": 6, "memory": 44138, "step": 22135} +{"lr": 4.531036769593769e-05, "data_time": 0.002109670639038086, "loss": 0.35376296639442445, "time": 0.631488585472107, "epoch": 6, "memory": 44138, "step": 22235} +{"lr": 4.4957330693402685e-05, "data_time": 0.002283740043640137, "loss": 0.34284416437149046, "time": 0.6309876441955566, "epoch": 6, "memory": 44138, "step": 22335} +{"lr": 4.460494301231059e-05, "data_time": 0.0016895532608032227, "loss": 0.380840128660202, "time": 0.6308343887329102, "epoch": 6, "memory": 44138, "step": 22435} +{"lr": 4.425322743673109e-05, "data_time": 0.0016279220581054688, "loss": 0.35961690843105315, "time": 0.6309692621231079, "epoch": 6, "memory": 44138, "step": 22535} +{"lr": 4.390220670727812e-05, "data_time": 0.0016263723373413086, "loss": 0.3624699264764786, "time": 0.6305348157882691, "epoch": 6, "memory": 44138, "step": 22635} +{"lr": 4.3551903519639325e-05, "data_time": 0.0019348859786987305, "loss": 0.3400824010372162, "time": 0.630367922782898, "epoch": 6, "memory": 44138, "step": 22735} +{"lr": 4.3202340523108894e-05, "data_time": 0.0016979217529296876, "loss": 0.3596193194389343, "time": 0.6301675081253052, "epoch": 6, "memory": 44138, "step": 22835} +{"lr": 4.285354031912302e-05, "data_time": 0.001671600341796875, "loss": 0.34684542417526243, "time": 0.629757285118103, "epoch": 6, "memory": 44138, "step": 22935} +{"lr": 4.2505525459798544e-05, "data_time": 0.0016098737716674805, "loss": 0.3625393629074097, "time": 0.6311562061309814, "epoch": 6, "memory": 44138, "step": 23035} +{"lr": 4.2158318446474836e-05, "data_time": 0.0014756202697753906, "loss": 0.3483108252286911, "time": 0.6312747240066529, "epoch": 6, "memory": 44138, "step": 23135} +{"lr": 4.181194172825905e-05, "data_time": 0.0018828153610229493, "loss": 0.3671716511249542, "time": 0.6319789409637451, "epoch": 6, "memory": 44138, "step": 23235} +{"lr": 4.146641770057448e-05, "data_time": 0.0018091201782226562, "loss": 0.37112095654010774, "time": 0.6314461708068848, "epoch": 6, "memory": 44138, "step": 23335} +{"lr": 4.112176870371267e-05, "data_time": 0.0014128684997558594, "loss": 0.3523112118244171, "time": 0.6315903425216675, "epoch": 6, "memory": 44138, "step": 23435} +{"accuracy/top1": 75.67586517333984, "data_time": 0.03196217684910215, "time": 1.2841334671809757, "step": 6} +{"lr": 4.0753988582732786e-05, "data_time": 0.001938033103942871, "loss": 0.3583205431699753, "time": 0.6311803340911866, "epoch": 7, "memory": 44138, "step": 23542} +{"lr": 4.0411221639286815e-05, "data_time": 0.0018255949020385743, "loss": 0.35708620846271516, "time": 0.6309070348739624, "epoch": 7, "memory": 44138, "step": 23642} +{"lr": 4.006939795169518e-05, "data_time": 0.002001237869262695, "loss": 0.3720770120620728, "time": 0.6302164554595947, "epoch": 7, "memory": 44138, "step": 23742} +{"lr": 3.972853962099913e-05, "data_time": 0.0015237808227539062, "loss": 0.372759947180748, "time": 0.6311291456222534, "epoch": 7, "memory": 44138, "step": 23842} +{"lr": 3.93886686858237e-05, "data_time": 0.0014948606491088866, "loss": 0.35073335468769073, "time": 0.6307839632034302, "epoch": 7, "memory": 44138, "step": 23942} +{"lr": 3.904980712095259e-05, "data_time": 0.001529860496520996, "loss": 0.3819312959909439, "time": 0.6303291082382202, "epoch": 7, "memory": 44138, "step": 24042} +{"lr": 3.871197683590748e-05, "data_time": 0.0019331693649291993, "loss": 0.4172582268714905, "time": 0.6300901412963867, "epoch": 7, "memory": 44138, "step": 24142} +{"lr": 3.837519967353125e-05, "data_time": 0.0018826723098754883, "loss": 0.3877938807010651, "time": 0.6306633710861206, "epoch": 7, "memory": 44138, "step": 24242} +{"lr": 3.8039497408575855e-05, "data_time": 0.0014815330505371094, "loss": 0.38511968553066256, "time": 0.630943751335144, "epoch": 7, "memory": 44138, "step": 24342} +{"lr": 3.770489174629432e-05, "data_time": 0.0017430305480957032, "loss": 0.35685568749904634, "time": 0.6305532455444336, "epoch": 7, "memory": 44138, "step": 24442} +{"lr": 3.737140432103754e-05, "data_time": 0.0018526792526245117, "loss": 0.40219601094722746, "time": 0.6301730632781982, "epoch": 7, "memory": 44138, "step": 24542} +{"lr": 3.703905669485545e-05, "data_time": 0.0015938758850097656, "loss": 0.36595679819583893, "time": 0.6310053110122681, "epoch": 7, "memory": 44138, "step": 24642} +{"lr": 3.6707870356102706e-05, "data_time": 0.0015116691589355468, "loss": 0.39236302971839904, "time": 0.630971622467041, "epoch": 7, "memory": 44138, "step": 24742} +{"lr": 3.6377866718049484e-05, "data_time": 0.002404189109802246, "loss": 0.3813491195440292, "time": 0.6325625419616699, "epoch": 7, "memory": 44138, "step": 24842} +{"lr": 3.6049067117496965e-05, "data_time": 0.002016139030456543, "loss": 0.3724536061286926, "time": 0.6303442239761352, "epoch": 7, "memory": 44138, "step": 24942} +{"lr": 3.5721492813397755e-05, "data_time": 0.001612710952758789, "loss": 0.3616188108921051, "time": 0.6305454492568969, "epoch": 7, "memory": 44138, "step": 25042} +{"lr": 3.539516498548151e-05, "data_time": 0.0016888618469238282, "loss": 0.3745671182870865, "time": 0.6316226720809937, "epoch": 7, "memory": 44138, "step": 25142} +{"lr": 3.507010473288518e-05, "data_time": 0.0014406442642211914, "loss": 0.37493705153465273, "time": 0.6312808275222779, "epoch": 7, "memory": 44138, "step": 25242} +{"lr": 3.474633307278909e-05, "data_time": 0.0016297578811645507, "loss": 0.35371659994125365, "time": 0.6316396951675415, "epoch": 7, "memory": 44138, "step": 25342} +{"lr": 3.4423870939058177e-05, "data_time": 0.0016299247741699218, "loss": 0.35014588236808775, "time": 0.6309483528137207, "epoch": 7, "memory": 44138, "step": 25442} +{"lr": 3.410273918088805e-05, "data_time": 0.0017478466033935547, "loss": 0.36910676062107084, "time": 0.6305905818939209, "epoch": 7, "memory": 44138, "step": 25542} +{"lr": 3.37829585614574e-05, "data_time": 0.001750349998474121, "loss": 0.376947021484375, "time": 0.6300114631652832, "epoch": 7, "memory": 44138, "step": 25642} +{"lr": 3.3464549756585216e-05, "data_time": 0.0014777421951293946, "loss": 0.32904700338840487, "time": 0.6318575859069824, "epoch": 7, "memory": 44138, "step": 25742} +{"lr": 3.314753335339416e-05, "data_time": 0.0016019105911254882, "loss": 0.3829327464103699, "time": 0.631765604019165, "epoch": 7, "memory": 44138, "step": 25842} +{"lr": 3.2831929848979394e-05, "data_time": 0.002126789093017578, "loss": 0.38341371715068817, "time": 0.6300968170166016, "epoch": 7, "memory": 44138, "step": 25942} +{"lr": 3.251775964908332e-05, "data_time": 0.0015276670455932617, "loss": 0.359295055270195, "time": 0.6310170888900757, "epoch": 7, "memory": 44138, "step": 26042} +{"lr": 3.22050430667762e-05, "data_time": 0.0015214204788208008, "loss": 0.3614420384168625, "time": 0.6304879903793335, "epoch": 7, "memory": 44138, "step": 26142} +{"lr": 3.189380032114285e-05, "data_time": 0.001581263542175293, "loss": 0.38451360166072845, "time": 0.6299646854400635, "epoch": 7, "memory": 44138, "step": 26242} +{"lr": 3.1584051535975324e-05, "data_time": 0.0015254735946655274, "loss": 0.3565573751926422, "time": 0.6304850816726685, "epoch": 7, "memory": 44138, "step": 26342} +{"lr": 3.127581673847175e-05, "data_time": 0.001414346694946289, "loss": 0.3782190531492233, "time": 0.6310838222503662, "epoch": 7, "memory": 44138, "step": 26442} +{"lr": 3.0969115857941555e-05, "data_time": 0.0016266822814941407, "loss": 0.38487481474876406, "time": 0.6298484086990357, "epoch": 7, "memory": 44138, "step": 26542} +{"lr": 3.0663968724516654e-05, "data_time": 0.001608586311340332, "loss": 0.36543085277080534, "time": 0.6297093391418457, "epoch": 7, "memory": 44138, "step": 26642} +{"lr": 3.0360395067869668e-05, "data_time": 0.0016102075576782226, "loss": 0.3948678284883499, "time": 0.629201602935791, "epoch": 7, "memory": 44138, "step": 26742} +{"lr": 3.0058414515938063e-05, "data_time": 0.0021617650985717774, "loss": 0.38861264288425446, "time": 0.6379568815231323, "epoch": 7, "memory": 44138, "step": 26842} +{"lr": 2.975804659365501e-05, "data_time": 0.0018576860427856445, "loss": 0.35465694665908815, "time": 0.6299306392669678, "epoch": 7, "memory": 44138, "step": 26942} +{"lr": 2.945931072168714e-05, "data_time": 0.001920795440673828, "loss": 0.3702373683452606, "time": 0.6300881862640381, "epoch": 7, "memory": 44138, "step": 27042} +{"lr": 2.9162226215178855e-05, "data_time": 0.0019040822982788086, "loss": 0.357158762216568, "time": 0.6302624702453613, "epoch": 7, "memory": 44138, "step": 27142} +{"lr": 2.8866812282503322e-05, "data_time": 0.001497507095336914, "loss": 0.38576229214668273, "time": 0.6305846691131591, "epoch": 7, "memory": 44138, "step": 27242} +{"lr": 2.8573088024020838e-05, "data_time": 0.00119326114654541, "loss": 0.3747771352529526, "time": 0.6289780378341675, "epoch": 7, "memory": 44138, "step": 27342} +{"accuracy/top1": 75.43927001953125, "data_time": 0.03220138056524869, "time": 1.2809336349881928, "step": 7} +{"lr": 2.8260695817390092e-05, "data_time": 0.0014861583709716796, "loss": 0.3698581427335739, "time": 0.6313601732254028, "epoch": 8, "memory": 44138, "step": 27449} +{"lr": 2.7970529402570967e-05, "data_time": 0.0015247583389282227, "loss": 0.3719507813453674, "time": 0.6313715219497681, "epoch": 8, "memory": 44138, "step": 27549} +{"lr": 2.7682110612244214e-05, "data_time": 0.001616692543029785, "loss": 0.39879120886325836, "time": 0.6307795763015747, "epoch": 8, "memory": 44138, "step": 27649} +{"lr": 2.7395458094490283e-05, "data_time": 0.002065730094909668, "loss": 0.36551759839057923, "time": 0.6318770408630371, "epoch": 8, "memory": 44138, "step": 27749} +{"lr": 2.7110590383189015e-05, "data_time": 0.0016478300094604492, "loss": 0.38542689085006715, "time": 0.6304060935974121, "epoch": 8, "memory": 44138, "step": 27849} +{"lr": 2.6827525896821397e-05, "data_time": 0.0019791841506958006, "loss": 0.3750642895698547, "time": 0.6311421155929565, "epoch": 8, "memory": 44138, "step": 27949} +{"lr": 2.6546282937278552e-05, "data_time": 0.0021829128265380858, "loss": 0.37531176805496214, "time": 0.6316208124160767, "epoch": 8, "memory": 44138, "step": 28049} +{"lr": 2.6266879688678606e-05, "data_time": 0.0014534711837768555, "loss": 0.3732608675956726, "time": 0.6311471462249756, "epoch": 8, "memory": 44138, "step": 28149} +{"lr": 2.5989334216190738e-05, "data_time": 0.0015685081481933594, "loss": 0.38682337403297423, "time": 0.6312675476074219, "epoch": 8, "memory": 44138, "step": 28249} +{"lr": 2.571366446486732e-05, "data_time": 0.0018526792526245117, "loss": 0.3724817126989365, "time": 0.6319009780883789, "epoch": 8, "memory": 44138, "step": 28349} +{"lr": 2.5439888258483665e-05, "data_time": 0.0016022443771362305, "loss": 0.35468791127204896, "time": 0.6319311141967774, "epoch": 8, "memory": 44138, "step": 28449} +{"lr": 2.5168023298385384e-05, "data_time": 0.001462268829345703, "loss": 0.3673750728368759, "time": 0.6326402425765991, "epoch": 8, "memory": 44138, "step": 28549} +{"lr": 2.489808716234412e-05, "data_time": 0.0016788244247436523, "loss": 0.37713308334350587, "time": 0.6322790622711182, "epoch": 8, "memory": 44138, "step": 28649} +{"lr": 2.463009730342111e-05, "data_time": 0.0014983654022216798, "loss": 0.3820088759064674, "time": 0.6319272518157959, "epoch": 8, "memory": 44138, "step": 28749} +{"lr": 2.436407104883849e-05, "data_time": 0.0017088651657104492, "loss": 0.3759547084569931, "time": 0.6316488742828369, "epoch": 8, "memory": 44138, "step": 28849} +{"lr": 2.4100025598859092e-05, "data_time": 0.0016340494155883789, "loss": 0.3572659969329834, "time": 0.6310683965682984, "epoch": 8, "memory": 44138, "step": 28949} +{"lr": 2.3837978025674284e-05, "data_time": 0.0015425920486450196, "loss": 0.35619386434555056, "time": 0.631263279914856, "epoch": 8, "memory": 44138, "step": 29049} +{"lr": 2.357794527230027e-05, "data_time": 0.002205491065979004, "loss": 0.3464666038751602, "time": 0.6315906047821045, "epoch": 8, "memory": 44138, "step": 29149} +{"lr": 2.33199441514825e-05, "data_time": 0.0013254165649414062, "loss": 0.356190425157547, "time": 0.6304213047027588, "epoch": 8, "memory": 44138, "step": 29249} +{"lr": 2.306399134460859e-05, "data_time": 0.0018564939498901367, "loss": 0.38022026121616365, "time": 0.6332060098648071, "epoch": 8, "memory": 44138, "step": 29349} +{"lr": 2.281010340062995e-05, "data_time": 0.0021481037139892576, "loss": 0.3789412349462509, "time": 0.632056736946106, "epoch": 8, "memory": 44138, "step": 29449} +{"lr": 2.2558296734991603e-05, "data_time": 0.0018558502197265625, "loss": 0.38903270065784457, "time": 0.6311423540115356, "epoch": 8, "memory": 44138, "step": 29549} +{"lr": 2.2308587628570915e-05, "data_time": 0.002602362632751465, "loss": 0.36461024880409243, "time": 0.6321547746658325, "epoch": 8, "memory": 44138, "step": 29649} +{"lr": 2.2060992226624955e-05, "data_time": 0.0015051603317260743, "loss": 0.381944078207016, "time": 0.6306434154510498, "epoch": 8, "memory": 44138, "step": 29749} +{"lr": 2.181552653774648e-05, "data_time": 0.0018649101257324219, "loss": 0.38841715455055237, "time": 0.631217360496521, "epoch": 8, "memory": 44138, "step": 29849} +{"lr": 2.1572206432829035e-05, "data_time": 0.001702117919921875, "loss": 0.4009664624929428, "time": 0.6303469657897949, "epoch": 8, "memory": 44138, "step": 29949} +{"lr": 2.1331047644040754e-05, "data_time": 0.0016013860702514648, "loss": 0.3727075010538101, "time": 0.630508279800415, "epoch": 8, "memory": 44138, "step": 30049} +{"lr": 2.109206576380705e-05, "data_time": 0.002045083045959473, "loss": 0.37341995239257814, "time": 0.6308868885040283, "epoch": 8, "memory": 44138, "step": 30149} +{"lr": 2.0855276243802687e-05, "data_time": 0.0016182899475097657, "loss": 0.3565258502960205, "time": 0.6321156978607178, "epoch": 8, "memory": 44138, "step": 30249} +{"lr": 2.0620694393952487e-05, "data_time": 0.0026824474334716797, "loss": 0.3869649887084961, "time": 0.6317596673965454, "epoch": 8, "memory": 44138, "step": 30349} +{"lr": 2.038833538144169e-05, "data_time": 0.0017143726348876954, "loss": 0.3577035516500473, "time": 0.6310218811035156, "epoch": 8, "memory": 44138, "step": 30449} +{"lr": 2.0158214229735154e-05, "data_time": 0.0015231609344482423, "loss": 0.35234747231006625, "time": 0.6306694269180297, "epoch": 8, "memory": 44138, "step": 30549} +{"lr": 1.9930345817605977e-05, "data_time": 0.0014863967895507812, "loss": 0.391020405292511, "time": 0.6305213689804077, "epoch": 8, "memory": 44138, "step": 30649} +{"lr": 1.970474487817358e-05, "data_time": 0.0017902851104736328, "loss": 0.38438333570957184, "time": 0.6305299043655396, "epoch": 8, "memory": 44138, "step": 30749} +{"lr": 1.9481425997951092e-05, "data_time": 0.0020719289779663084, "loss": 0.3644045442342758, "time": 0.6309171199798584, "epoch": 8, "memory": 44138, "step": 30849} diff --git a/clip_large_pretrain_4x256_sdv2_lr1e-4_rub_/20230603_200211/vis_data/config.py b/clip_large_pretrain_4x256_sdv2_lr1e-4_rub_/20230603_200211/vis_data/config.py new file mode 100644 index 0000000000000000000000000000000000000000..19e31047e13d469a3faf46963faf1137c9d94750 --- /dev/null +++ b/clip_large_pretrain_4x256_sdv2_lr1e-4_rub_/20230603_200211/vis_data/config.py @@ -0,0 +1,228 @@ +optim_wrapper = dict( + optimizer=dict( + type='AdamW', lr=0.0001, weight_decay=0.3, _scope_='mmpretrain'), + paramwise_cfg=dict( + custom_keys=dict({ + '.cls_token': dict(decay_mult=0.0), + '.pos_embed': dict(decay_mult=0.0) + })), + type='AmpOptimWrapper', + dtype='bfloat16', + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=False, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=4096) +model = dict( + type='ImageClassifier', + backbone=dict( + frozen_stages=24, + type='VisionTransformer', + arch='l', + img_size=224, + patch_size=14, + drop_rate=0.1, + pre_norm=True, + init_cfg=dict( + type='Pretrained', + checkpoint='ckpt/openclip-ViT-L-14.pth', + prefix='backbone')), + neck=dict(type='CLIPProjection', in_channels=1024, out_channels=768), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=768, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + init_cfg=None), + init_cfg=dict( + type='TruncNormal', layer=['Conv2d', 'Linear'], std=0.02, bias=0.0), + train_cfg=None) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=128, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stablediffusionV2-1-dpmsolver-25-1m.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/cc1m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=128, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV2-1-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=128, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV2-1-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/clip_large_pretrain_4x256_sdv2_lr1e-4' diff --git a/clip_large_pretrain_4x256_sdv2_lr1e-4_rub_/20230603_200211/vis_data/events.out.tfevents.1685793746.SH-IDC1-10-140-24-114.117699.0 b/clip_large_pretrain_4x256_sdv2_lr1e-4_rub_/20230603_200211/vis_data/events.out.tfevents.1685793746.SH-IDC1-10-140-24-114.117699.0 new file mode 100644 index 0000000000000000000000000000000000000000..5584037beaa15bdc0c5370d7bd7e8c6a9523100a --- /dev/null +++ b/clip_large_pretrain_4x256_sdv2_lr1e-4_rub_/20230603_200211/vis_data/events.out.tfevents.1685793746.SH-IDC1-10-140-24-114.117699.0 @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:dd498abc17397fd2b5b7f987b99647f85871cb471cc1918aba5feb8747dc22e1 +size 1223053 diff --git a/clip_large_pretrain_4x256_sdv2_lr1e-4_rub_/20230603_200211/vis_data/scalars.json b/clip_large_pretrain_4x256_sdv2_lr1e-4_rub_/20230603_200211/vis_data/scalars.json new file mode 100644 index 0000000000000000000000000000000000000000..d9f552c019633a71994d9bb74d2c5664c92d7532 --- /dev/null +++ b/clip_large_pretrain_4x256_sdv2_lr1e-4_rub_/20230603_200211/vis_data/scalars.json @@ -0,0 +1,315 @@ +{"lr": 9.999857418382631e-05, "data_time": 0.0021013259887695313, "loss": 0.5254676818847657, "time": 0.6303915500640869, "epoch": 1, "memory": 44139, "step": 100} +{"lr": 9.999423907359506e-05, "data_time": 0.001608133316040039, "loss": 0.49592080116271975, "time": 0.6308860301971435, "epoch": 1, "memory": 44139, "step": 200} +{"lr": 9.998699480412073e-05, "data_time": 0.0016800165176391602, "loss": 0.47416927218437194, "time": 0.631201982498169, "epoch": 1, "memory": 44139, "step": 300} +{"lr": 9.997684184379085e-05, "data_time": 0.001680898666381836, "loss": 0.4523438036441803, "time": 0.6304768800735474, "epoch": 1, "memory": 44139, "step": 400} +{"lr": 9.996378084905784e-05, "data_time": 0.0017073631286621093, "loss": 0.4551931142807007, "time": 0.6323943138122559, "epoch": 1, "memory": 44139, "step": 500} +{"lr": 9.994781266439667e-05, "data_time": 0.0019591093063354493, "loss": 0.44729283452033997, "time": 0.6456096410751343, "epoch": 1, "memory": 44139, "step": 600} +{"lr": 9.992893832225061e-05, "data_time": 0.0015645503997802734, "loss": 0.4602384209632874, "time": 0.6309844970703125, "epoch": 1, "memory": 44139, "step": 700} +{"lr": 9.990715904296405e-05, "data_time": 0.002456402778625488, "loss": 0.4575894296169281, "time": 0.6301848411560058, "epoch": 1, "memory": 44139, "step": 800} +{"lr": 9.988247623470371e-05, "data_time": 0.001614689826965332, "loss": 0.4487850069999695, "time": 0.6298971891403198, "epoch": 1, "memory": 44139, "step": 900} +{"lr": 9.985489149336749e-05, "data_time": 0.0017919540405273438, "loss": 0.45876056551933286, "time": 0.6338362693786621, "epoch": 1, "memory": 44139, "step": 1000} +{"lr": 9.982440660248187e-05, "data_time": 0.0015303611755371094, "loss": 0.43976943790912626, "time": 0.6301945209503174, "epoch": 1, "memory": 44139, "step": 1100} +{"lr": 9.979102353308574e-05, "data_time": 0.0018020153045654296, "loss": 0.4259138166904449, "time": 0.6292660236358643, "epoch": 1, "memory": 44139, "step": 1200} +{"lr": 9.975474444360358e-05, "data_time": 0.0016472339630126953, "loss": 0.4358592242002487, "time": 0.6309081792831421, "epoch": 1, "memory": 44139, "step": 1300} +{"lr": 9.97155716797056e-05, "data_time": 0.001531839370727539, "loss": 0.44602505266666415, "time": 0.6301747322082519, "epoch": 1, "memory": 44139, "step": 1400} +{"lr": 9.967350777415632e-05, "data_time": 0.0019945621490478514, "loss": 0.43150107860565184, "time": 0.6295226812362671, "epoch": 1, "memory": 44139, "step": 1500} +{"lr": 9.962855544665059e-05, "data_time": 0.0017927885055541992, "loss": 0.4311340630054474, "time": 0.6293833255767822, "epoch": 1, "memory": 44139, "step": 1600} +{"lr": 9.958071760363773e-05, "data_time": 0.0020589590072631835, "loss": 0.41964859068393706, "time": 0.6303755044937134, "epoch": 1, "memory": 44139, "step": 1700} +{"lr": 9.952999733813382e-05, "data_time": 0.0015864372253417969, "loss": 0.4246463358402252, "time": 0.6298433542251587, "epoch": 1, "memory": 44139, "step": 1800} +{"lr": 9.947639792952148e-05, "data_time": 0.0016413688659667968, "loss": 0.40548085868358613, "time": 0.6293341636657714, "epoch": 1, "memory": 44139, "step": 1900} +{"lr": 9.941992284333813e-05, "data_time": 0.0018304824829101563, "loss": 0.43340319097042085, "time": 0.6292242527008056, "epoch": 1, "memory": 44139, "step": 2000} +{"lr": 9.93605757310516e-05, "data_time": 0.0016559600830078126, "loss": 0.43370539247989653, "time": 0.6301441192626953, "epoch": 1, "memory": 44139, "step": 2100} +{"lr": 9.929836042982421e-05, "data_time": 0.0016306638717651367, "loss": 0.40694511234760283, "time": 0.6304590463638305, "epoch": 1, "memory": 44139, "step": 2200} +{"lr": 9.923328096226467e-05, "data_time": 0.0021354198455810548, "loss": 0.41349645256996154, "time": 0.6303184509277344, "epoch": 1, "memory": 44139, "step": 2300} +{"lr": 9.916534153616819e-05, "data_time": 0.0018451690673828125, "loss": 0.41379840970039367, "time": 0.6308378458023072, "epoch": 1, "memory": 44139, "step": 2400} +{"lr": 9.909454654424387e-05, "data_time": 0.0016844034194946288, "loss": 0.41174737811088563, "time": 0.6300135850906372, "epoch": 1, "memory": 44139, "step": 2500} +{"lr": 9.902090056383109e-05, "data_time": 0.0016065359115600586, "loss": 0.4248443007469177, "time": 0.6317947149276734, "epoch": 1, "memory": 44139, "step": 2600} +{"lr": 9.894440835660365e-05, "data_time": 0.0017657518386840821, "loss": 0.4310387372970581, "time": 0.6297615766525269, "epoch": 1, "memory": 44139, "step": 2700} +{"lr": 9.886507486826162e-05, "data_time": 0.001495027542114258, "loss": 0.41254869401454924, "time": 0.630182147026062, "epoch": 1, "memory": 44139, "step": 2800} +{"lr": 9.878290522821154e-05, "data_time": 0.0017021417617797852, "loss": 0.38445100784301756, "time": 0.6294927358627319, "epoch": 1, "memory": 44139, "step": 2900} +{"lr": 9.86979047492352e-05, "data_time": 0.002216958999633789, "loss": 0.4417062133550644, "time": 0.6311541795730591, "epoch": 1, "memory": 44139, "step": 3000} +{"lr": 9.861007892714578e-05, "data_time": 0.0016395330429077148, "loss": 0.4045764863491058, "time": 0.6297856569290161, "epoch": 1, "memory": 44139, "step": 3100} +{"lr": 9.851943344043239e-05, "data_time": 0.0015326976776123048, "loss": 0.4071483612060547, "time": 0.6301517009735107, "epoch": 1, "memory": 44139, "step": 3200} +{"lr": 9.842597414989319e-05, "data_time": 0.001749730110168457, "loss": 0.41237278282642365, "time": 0.6304476499557495, "epoch": 1, "memory": 44139, "step": 3300} +{"lr": 9.832970709825637e-05, "data_time": 0.001751875877380371, "loss": 0.40038027465343473, "time": 0.6293846130371094, "epoch": 1, "memory": 44139, "step": 3400} +{"lr": 9.823063850978968e-05, "data_time": 0.0021019458770751955, "loss": 0.4185768276453018, "time": 0.6309142112731934, "epoch": 1, "memory": 44139, "step": 3500} +{"lr": 9.81287747898975e-05, "data_time": 0.0016637563705444336, "loss": 0.3904751867055893, "time": 0.6299480676651001, "epoch": 1, "memory": 44139, "step": 3600} +{"lr": 9.802412252470715e-05, "data_time": 0.0018462657928466797, "loss": 0.4244962424039841, "time": 0.6292508125305176, "epoch": 1, "memory": 44139, "step": 3700} +{"lr": 9.791668848064258e-05, "data_time": 0.001714801788330078, "loss": 0.4359990656375885, "time": 0.6293673038482666, "epoch": 1, "memory": 44139, "step": 3800} +{"lr": 9.780647960398751e-05, "data_time": 0.001652693748474121, "loss": 0.4074475705623627, "time": 0.6290539026260376, "epoch": 1, "memory": 44139, "step": 3900} +{"accuracy/top1": 78.65766906738281, "data_time": 0.0428013801574707, "time": 1.3483564602701288, "step": 1} +{"lr": 9.7685491196646e-05, "data_time": 0.001947164535522461, "loss": 0.38701715171337125, "time": 0.6323011636734008, "epoch": 2, "memory": 44139, "step": 4007} +{"lr": 9.756956126085676e-05, "data_time": 0.0019211292266845704, "loss": 0.38512068390846255, "time": 0.6309863805770874, "epoch": 2, "memory": 44138, "step": 4107} +{"lr": 9.745087893642482e-05, "data_time": 0.0017422199249267577, "loss": 0.377513712644577, "time": 0.6308529138565063, "epoch": 2, "memory": 44138, "step": 4207} +{"lr": 9.73294518969056e-05, "data_time": 0.0013633489608764649, "loss": 0.40180844366550444, "time": 0.6309009790420532, "epoch": 2, "memory": 44138, "step": 4307} +{"lr": 9.72052879933172e-05, "data_time": 0.0016982555389404297, "loss": 0.4086743950843811, "time": 0.6303865432739257, "epoch": 2, "memory": 44138, "step": 4407} +{"lr": 9.707839525363358e-05, "data_time": 0.0026702642440795898, "loss": 0.38177225589752195, "time": 0.6331295728683471, "epoch": 2, "memory": 44138, "step": 4507} +{"lr": 9.69487818822646e-05, "data_time": 0.002133893966674805, "loss": 0.40716169476509095, "time": 0.6305238485336304, "epoch": 2, "memory": 44138, "step": 4607} +{"lr": 9.681645625952616e-05, "data_time": 0.0016930580139160156, "loss": 0.3874299794435501, "time": 0.6302786588668823, "epoch": 2, "memory": 44138, "step": 4707} +{"lr": 9.668142694109826e-05, "data_time": 0.0017039060592651367, "loss": 0.3736038327217102, "time": 0.631012749671936, "epoch": 2, "memory": 44138, "step": 4807} +{"lr": 9.654370265747162e-05, "data_time": 0.0016846179962158204, "loss": 0.3739929348230362, "time": 0.6311726570129395, "epoch": 2, "memory": 44138, "step": 4907} +{"lr": 9.640329231338345e-05, "data_time": 0.0018583297729492187, "loss": 0.3880970239639282, "time": 0.6307178497314453, "epoch": 2, "memory": 44138, "step": 5007} +{"lr": 9.626020498724147e-05, "data_time": 0.0020255327224731447, "loss": 0.39673975110054016, "time": 0.6308103799819946, "epoch": 2, "memory": 44138, "step": 5107} +{"lr": 9.611444993053717e-05, "data_time": 0.0020781755447387695, "loss": 0.40036986470222474, "time": 0.6299102783203125, "epoch": 2, "memory": 44138, "step": 5207} +{"lr": 9.596603656724746e-05, "data_time": 0.0018322467803955078, "loss": 0.4005657464265823, "time": 0.6303497076034545, "epoch": 2, "memory": 44138, "step": 5307} +{"lr": 9.58149744932254e-05, "data_time": 0.0016456842422485352, "loss": 0.39048443734645844, "time": 0.6305620908737183, "epoch": 2, "memory": 44138, "step": 5407} +{"lr": 9.566127347557972e-05, "data_time": 0.001516413688659668, "loss": 0.3764130622148514, "time": 0.6307580232620239, "epoch": 2, "memory": 44138, "step": 5507} +{"lr": 9.550494345204367e-05, "data_time": 0.0022286176681518555, "loss": 0.38684943616390227, "time": 0.6312837362289428, "epoch": 2, "memory": 44138, "step": 5607} +{"lr": 9.534599453033183e-05, "data_time": 0.0015774965286254883, "loss": 0.36203564405441285, "time": 0.6302300691604614, "epoch": 2, "memory": 44138, "step": 5707} +{"lr": 9.518443698748712e-05, "data_time": 0.0018428802490234376, "loss": 0.36771643459796904, "time": 0.6324694395065308, "epoch": 2, "memory": 44138, "step": 5807} +{"lr": 9.50202812692162e-05, "data_time": 0.0027725934982299806, "loss": 0.3563918322324753, "time": 0.6313409566879272, "epoch": 2, "memory": 44138, "step": 5907} +{"lr": 9.485353798921386e-05, "data_time": 0.0018866300582885743, "loss": 0.3846891909837723, "time": 0.6306589126586915, "epoch": 2, "memory": 44138, "step": 6007} +{"lr": 9.468421792847712e-05, "data_time": 0.0025237083435058595, "loss": 0.3983618408441544, "time": 0.6313921451568604, "epoch": 2, "memory": 44138, "step": 6107} +{"lr": 9.451233203460794e-05, "data_time": 0.0019637584686279298, "loss": 0.3815899193286896, "time": 0.6315789222717285, "epoch": 2, "memory": 44138, "step": 6207} +{"lr": 9.433789142110532e-05, "data_time": 0.001621842384338379, "loss": 0.3959047496318817, "time": 0.6313664436340332, "epoch": 2, "memory": 44138, "step": 6307} +{"lr": 9.416090736664709e-05, "data_time": 0.0023834705352783203, "loss": 0.37512257099151614, "time": 0.6290892601013184, "epoch": 2, "memory": 44138, "step": 6407} +{"lr": 9.398139131436032e-05, "data_time": 0.0018195152282714845, "loss": 0.3631224125623703, "time": 0.6311782121658325, "epoch": 2, "memory": 44138, "step": 6507} +{"lr": 9.379935487108167e-05, "data_time": 0.0017686843872070312, "loss": 0.369792178273201, "time": 0.6317324876785279, "epoch": 2, "memory": 44138, "step": 6607} +{"lr": 9.361480980660697e-05, "data_time": 0.0020948410034179687, "loss": 0.3787175536155701, "time": 0.6312172174453735, "epoch": 2, "memory": 44138, "step": 6707} +{"lr": 9.342776805292993e-05, "data_time": 0.002226853370666504, "loss": 0.3891234517097473, "time": 0.6308389902114868, "epoch": 2, "memory": 44138, "step": 6807} +{"lr": 9.323824170347083e-05, "data_time": 0.0018370389938354493, "loss": 0.369640052318573, "time": 0.633220887184143, "epoch": 2, "memory": 44138, "step": 6907} +{"lr": 9.304624301229471e-05, "data_time": 0.002008342742919922, "loss": 0.3819348871707916, "time": 0.6316205024719238, "epoch": 2, "memory": 44138, "step": 7007} +{"lr": 9.285178439331897e-05, "data_time": 0.0016373157501220702, "loss": 0.38032921850681306, "time": 0.6298771381378174, "epoch": 2, "memory": 44138, "step": 7107} +{"lr": 9.265487841951087e-05, "data_time": 0.0017477750778198242, "loss": 0.3724153459072113, "time": 0.6307685136795044, "epoch": 2, "memory": 44138, "step": 7207} +{"lr": 9.245553782207421e-05, "data_time": 0.002009415626525879, "loss": 0.36185015439987184, "time": 0.6313466310501099, "epoch": 2, "memory": 44138, "step": 7307} +{"lr": 9.225377548962671e-05, "data_time": 0.0015581607818603515, "loss": 0.3673944413661957, "time": 0.6307828187942505, "epoch": 2, "memory": 44138, "step": 7407} +{"lr": 9.204960446736601e-05, "data_time": 0.0017274379730224609, "loss": 0.3907669007778168, "time": 0.6304670333862304, "epoch": 2, "memory": 44138, "step": 7507} +{"lr": 9.184303795622715e-05, "data_time": 0.0018530845642089843, "loss": 0.4112739682197571, "time": 0.6308581829071045, "epoch": 2, "memory": 44138, "step": 7607} +{"lr": 9.163408931202823e-05, "data_time": 0.002285194396972656, "loss": 0.39737538397312167, "time": 0.6308098316192627, "epoch": 2, "memory": 44138, "step": 7707} +{"lr": 9.142277204460729e-05, "data_time": 0.0014277219772338868, "loss": 0.379548117518425, "time": 0.6297266244888305, "epoch": 2, "memory": 44138, "step": 7807} +{"accuracy/top1": 78.72029876708984, "data_time": 0.03160543277345855, "time": 1.2824266820118344, "step": 2} +{"lr": 9.119405492383288e-05, "data_time": 0.0019155025482177734, "loss": 0.38339483737945557, "time": 0.6309638738632202, "epoch": 3, "memory": 44138, "step": 7914} +{"lr": 9.097787819233205e-05, "data_time": 0.001637101173400879, "loss": 0.39207392930984497, "time": 0.6319971561431885, "epoch": 3, "memory": 44138, "step": 8014} +{"lr": 9.075937526576627e-05, "data_time": 0.0020543336868286133, "loss": 0.3818666934967041, "time": 0.6318645238876343, "epoch": 3, "memory": 44138, "step": 8114} +{"lr": 9.053856027171755e-05, "data_time": 0.001692509651184082, "loss": 0.38909188807010653, "time": 0.630961012840271, "epoch": 3, "memory": 44138, "step": 8214} +{"lr": 9.031544748725785e-05, "data_time": 0.001746964454650879, "loss": 0.36356093883514407, "time": 0.6317759275436401, "epoch": 3, "memory": 44138, "step": 8314} +{"lr": 9.009005133802542e-05, "data_time": 0.0013951778411865235, "loss": 0.3986783742904663, "time": 0.6313511371612549, "epoch": 3, "memory": 44138, "step": 8414} +{"lr": 8.986238639729242e-05, "data_time": 0.0020413875579833986, "loss": 0.3918088436126709, "time": 0.6310792446136475, "epoch": 3, "memory": 44138, "step": 8514} +{"lr": 8.963246738502254e-05, "data_time": 0.0020148038864135744, "loss": 0.36713131368160246, "time": 0.6311418771743774, "epoch": 3, "memory": 44138, "step": 8614} +{"lr": 8.940030916691928e-05, "data_time": 0.0019838571548461913, "loss": 0.36795192062854765, "time": 0.6313153982162476, "epoch": 3, "memory": 44138, "step": 8714} +{"lr": 8.91659267534649e-05, "data_time": 0.0019896984100341796, "loss": 0.3668671786785126, "time": 0.6314969062805176, "epoch": 3, "memory": 44138, "step": 8814} +{"lr": 8.892933529894978e-05, "data_time": 0.0017478227615356444, "loss": 0.3891805768013, "time": 0.6318260908126831, "epoch": 3, "memory": 44138, "step": 8914} +{"lr": 8.869055010049283e-05, "data_time": 0.0015655279159545899, "loss": 0.40365838408470156, "time": 0.6307106256484986, "epoch": 3, "memory": 44138, "step": 9014} +{"lr": 8.844958659705188e-05, "data_time": 0.0014819622039794922, "loss": 0.35741288065910337, "time": 0.6319898366928101, "epoch": 3, "memory": 44138, "step": 9114} +{"lr": 8.82064603684259e-05, "data_time": 0.0020102739334106447, "loss": 0.3577596008777618, "time": 0.6315583705902099, "epoch": 3, "memory": 44138, "step": 9214} +{"lr": 8.796118713424776e-05, "data_time": 0.0019507884979248046, "loss": 0.3682455807924271, "time": 0.6324578762054444, "epoch": 3, "memory": 44138, "step": 9314} +{"lr": 8.77137827529676e-05, "data_time": 0.0018065452575683593, "loss": 0.39598606526851654, "time": 0.6308689117431641, "epoch": 3, "memory": 44138, "step": 9414} +{"lr": 8.746426322082745e-05, "data_time": 0.0014045953750610352, "loss": 0.3769483804702759, "time": 0.6309690237045288, "epoch": 3, "memory": 44138, "step": 9514} +{"lr": 8.721264467082732e-05, "data_time": 0.0022228717803955077, "loss": 0.3582300215959549, "time": 0.6320258617401123, "epoch": 3, "memory": 44138, "step": 9614} +{"lr": 8.695894337168168e-05, "data_time": 0.0019613027572631834, "loss": 0.3443090319633484, "time": 0.6312095642089843, "epoch": 3, "memory": 44138, "step": 9714} +{"lr": 8.67031757267677e-05, "data_time": 0.0016478776931762695, "loss": 0.3798891544342041, "time": 0.6311641454696655, "epoch": 3, "memory": 44138, "step": 9814} +{"lr": 8.64453582730651e-05, "data_time": 0.001678299903869629, "loss": 0.38648802042007446, "time": 0.6316426038742066, "epoch": 3, "memory": 44138, "step": 9914} +{"lr": 8.61855076800861e-05, "data_time": 0.0014892578125, "loss": 0.37505062818527224, "time": 0.6309507846832275, "epoch": 3, "memory": 44138, "step": 10014} +{"lr": 8.592364074879843e-05, "data_time": 0.0017383575439453125, "loss": 0.35937183499336245, "time": 0.630468225479126, "epoch": 3, "memory": 44138, "step": 10114} +{"lr": 8.565977441053854e-05, "data_time": 0.001960349082946777, "loss": 0.4062574774026871, "time": 0.6306670665740967, "epoch": 3, "memory": 44138, "step": 10214} +{"lr": 8.539392572591724e-05, "data_time": 0.001788496971130371, "loss": 0.34328748285770416, "time": 0.6322952032089233, "epoch": 3, "memory": 44138, "step": 10314} +{"lr": 8.512611188371633e-05, "data_time": 0.0020530223846435547, "loss": 0.3801038235425949, "time": 0.6303425788879394, "epoch": 3, "memory": 44138, "step": 10414} +{"lr": 8.485635019977725e-05, "data_time": 0.001952791213989258, "loss": 0.4016746997833252, "time": 0.6302950620651245, "epoch": 3, "memory": 44138, "step": 10514} +{"lr": 8.45846581158818e-05, "data_time": 0.002126765251159668, "loss": 0.37341198325157166, "time": 0.6310628175735473, "epoch": 3, "memory": 44138, "step": 10614} +{"lr": 8.431105319862416e-05, "data_time": 0.0019888877868652344, "loss": 0.3901614427566528, "time": 0.6303201675415039, "epoch": 3, "memory": 44138, "step": 10714} +{"lr": 8.403555313827516e-05, "data_time": 0.0016383886337280273, "loss": 0.37818117439746857, "time": 0.630945348739624, "epoch": 3, "memory": 44138, "step": 10814} +{"lr": 8.375817574763844e-05, "data_time": 0.0017586231231689453, "loss": 0.36275345683097837, "time": 0.6306392908096313, "epoch": 3, "memory": 44138, "step": 10914} +{"lr": 8.347893896089871e-05, "data_time": 0.0022778034210205076, "loss": 0.37881468832492826, "time": 0.6306608438491821, "epoch": 3, "memory": 44138, "step": 11014} +{"lr": 8.31978608324625e-05, "data_time": 0.001977705955505371, "loss": 0.37366268038749695, "time": 0.6312607288360595, "epoch": 3, "memory": 44138, "step": 11114} +{"lr": 8.291495953579052e-05, "data_time": 0.0017975807189941407, "loss": 0.3661737948656082, "time": 0.6308862447738648, "epoch": 3, "memory": 44138, "step": 11214} +{"lr": 8.263025336222267e-05, "data_time": 0.0014571666717529297, "loss": 0.35999304354190825, "time": 0.6310614109039306, "epoch": 3, "memory": 44138, "step": 11314} +{"lr": 8.234376071979551e-05, "data_time": 0.0018079280853271484, "loss": 0.38005903363227844, "time": 0.6313439607620239, "epoch": 3, "memory": 44138, "step": 11414} +{"lr": 8.205550013205187e-05, "data_time": 0.0025040388107299806, "loss": 0.3770012348890305, "time": 0.631182336807251, "epoch": 3, "memory": 44138, "step": 11514} +{"lr": 8.17654902368435e-05, "data_time": 0.0018376588821411132, "loss": 0.37870813608169557, "time": 0.6309926271438598, "epoch": 3, "memory": 44138, "step": 11614} +{"lr": 8.147374978512582e-05, "data_time": 0.0015785694122314453, "loss": 0.36322163343429564, "time": 0.6306546211242676, "epoch": 3, "memory": 44138, "step": 11714} +{"accuracy/top1": 77.37378692626953, "data_time": 0.031213887806596428, "time": 1.2824072385656422, "step": 3} +{"lr": 8.115969237626578e-05, "data_time": 0.001883554458618164, "loss": 0.34975853860378264, "time": 0.6301671504974365, "epoch": 4, "memory": 44138, "step": 11821} +{"lr": 8.08644297337245e-05, "data_time": 0.0018700361251831055, "loss": 0.3540323615074158, "time": 0.6305224418640136, "epoch": 4, "memory": 44138, "step": 11921} +{"lr": 8.056749479387651e-05, "data_time": 0.0018409490585327148, "loss": 0.35627412497997285, "time": 0.6311853170394898, "epoch": 4, "memory": 44138, "step": 12021} +{"lr": 8.026890675542457e-05, "data_time": 0.0019524097442626953, "loss": 0.35569603741168976, "time": 0.6314916849136353, "epoch": 4, "memory": 44138, "step": 12121} +{"lr": 7.996868492395457e-05, "data_time": 0.0015810728073120117, "loss": 0.35987875163555144, "time": 0.6313385248184205, "epoch": 4, "memory": 44138, "step": 12221} +{"lr": 7.966684871068762e-05, "data_time": 0.0020611763000488283, "loss": 0.37253828942775724, "time": 0.6317244291305542, "epoch": 4, "memory": 44138, "step": 12321} +{"lr": 7.936341763122449e-05, "data_time": 0.0016445159912109376, "loss": 0.3688558876514435, "time": 0.6316900968551635, "epoch": 4, "memory": 44138, "step": 12421} +{"lr": 7.905841130428405e-05, "data_time": 0.001904129981994629, "loss": 0.3982546806335449, "time": 0.6310122728347778, "epoch": 4, "memory": 44138, "step": 12521} +{"lr": 7.875184945043487e-05, "data_time": 0.0019101142883300782, "loss": 0.3493445962667465, "time": 0.6312102794647216, "epoch": 4, "memory": 44138, "step": 12621} +{"lr": 7.844375189082012e-05, "data_time": 0.0018593072891235352, "loss": 0.3826897144317627, "time": 0.6311030864715577, "epoch": 4, "memory": 44138, "step": 12721} +{"lr": 7.813413854587572e-05, "data_time": 0.0017517566680908202, "loss": 0.3819606751203537, "time": 0.6317952632904053, "epoch": 4, "memory": 44138, "step": 12821} +{"lr": 7.782302943404291e-05, "data_time": 0.0017009258270263671, "loss": 0.37396695017814635, "time": 0.632053542137146, "epoch": 4, "memory": 44138, "step": 12921} +{"lr": 7.751044467047337e-05, "data_time": 0.001685023307800293, "loss": 0.39482679665088655, "time": 0.6310154438018799, "epoch": 4, "memory": 44138, "step": 13021} +{"lr": 7.719640446572907e-05, "data_time": 0.0017710208892822265, "loss": 0.3653447687625885, "time": 0.6302547931671143, "epoch": 4, "memory": 44138, "step": 13121} +{"lr": 7.688092912447536e-05, "data_time": 0.0024473190307617186, "loss": 0.3717967987060547, "time": 0.6365467786788941, "epoch": 4, "memory": 44138, "step": 13221} +{"lr": 7.656403904416794e-05, "data_time": 0.0021691083908081054, "loss": 0.37738717496395113, "time": 0.6322481155395507, "epoch": 4, "memory": 44138, "step": 13321} +{"lr": 7.624575471373434e-05, "data_time": 0.002275562286376953, "loss": 0.37544898986816405, "time": 0.6308626651763916, "epoch": 4, "memory": 44138, "step": 13421} +{"lr": 7.592609671224926e-05, "data_time": 0.0018621444702148437, "loss": 0.3497780740261078, "time": 0.6322161674499511, "epoch": 4, "memory": 44138, "step": 13521} +{"lr": 7.560508570760375e-05, "data_time": 0.0016445159912109376, "loss": 0.378781920671463, "time": 0.6311922311782837, "epoch": 4, "memory": 44138, "step": 13621} +{"lr": 7.528274245516895e-05, "data_time": 0.0019595146179199217, "loss": 0.38113327622413634, "time": 0.6312441825866699, "epoch": 4, "memory": 44138, "step": 13721} +{"lr": 7.495908779645405e-05, "data_time": 0.001761150360107422, "loss": 0.367531618475914, "time": 0.6302317619323731, "epoch": 4, "memory": 44138, "step": 13821} +{"lr": 7.463414265775906e-05, "data_time": 0.001967191696166992, "loss": 0.35146708190441134, "time": 0.6400554418563843, "epoch": 4, "memory": 44138, "step": 13921} +{"lr": 7.430792804882147e-05, "data_time": 0.0016878366470336914, "loss": 0.3615738064050674, "time": 0.6307236909866333, "epoch": 4, "memory": 44138, "step": 14021} +{"lr": 7.398046506145803e-05, "data_time": 0.0015439510345458985, "loss": 0.3717303067445755, "time": 0.6305463552474976, "epoch": 4, "memory": 44138, "step": 14121} +{"lr": 7.365177486820114e-05, "data_time": 0.0015390872955322265, "loss": 0.40721426606178285, "time": 0.630876612663269, "epoch": 4, "memory": 44138, "step": 14221} +{"lr": 7.332187872092939e-05, "data_time": 0.0018485069274902343, "loss": 0.3763453930616379, "time": 0.6308673620223999, "epoch": 4, "memory": 44138, "step": 14321} +{"lr": 7.299079794949401e-05, "data_time": 0.0017867565155029296, "loss": 0.36380934417247773, "time": 0.6306280612945556, "epoch": 4, "memory": 44138, "step": 14421} +{"lr": 7.265855396033955e-05, "data_time": 0.0018984079360961914, "loss": 0.3613135159015656, "time": 0.6310559034347534, "epoch": 4, "memory": 44138, "step": 14521} +{"lr": 7.232516823512013e-05, "data_time": 0.001582193374633789, "loss": 0.37295305728912354, "time": 0.6325780630111695, "epoch": 4, "memory": 44138, "step": 14621} +{"lr": 7.199066232930995e-05, "data_time": 0.0018008947372436523, "loss": 0.371643990278244, "time": 0.6305253982543946, "epoch": 4, "memory": 44138, "step": 14721} +{"lr": 7.165505787080997e-05, "data_time": 0.0017313241958618164, "loss": 0.3901374012231827, "time": 0.6315348386764527, "epoch": 4, "memory": 44138, "step": 14821} +{"lr": 7.13183765585495e-05, "data_time": 0.0023801326751708984, "loss": 0.3802525758743286, "time": 0.63101806640625, "epoch": 4, "memory": 44138, "step": 14921} +{"lr": 7.098064016108317e-05, "data_time": 0.0014557600021362304, "loss": 0.372085177898407, "time": 0.6336552143096924, "epoch": 4, "memory": 44138, "step": 15021} +{"lr": 7.064187051518379e-05, "data_time": 0.0014697790145874023, "loss": 0.3566492527723312, "time": 0.6312332630157471, "epoch": 4, "memory": 44138, "step": 15121} +{"lr": 7.030208952442949e-05, "data_time": 0.001665806770324707, "loss": 0.37336931228637693, "time": 0.6538846969604493, "epoch": 4, "memory": 44138, "step": 15221} +{"lr": 6.996131915778851e-05, "data_time": 0.0020270586013793946, "loss": 0.3725410372018814, "time": 0.6306520223617553, "epoch": 4, "memory": 44138, "step": 15321} +{"lr": 6.961958144819853e-05, "data_time": 0.001505136489868164, "loss": 0.38264588713645936, "time": 0.6309946298599243, "epoch": 4, "memory": 44138, "step": 15421} +{"lr": 6.927689849114193e-05, "data_time": 0.002476620674133301, "loss": 0.383384981751442, "time": 0.6316012144088745, "epoch": 4, "memory": 44138, "step": 15521} +{"lr": 6.893329244321722e-05, "data_time": 0.001361536979675293, "loss": 0.3943897247314453, "time": 0.6302694797515869, "epoch": 4, "memory": 44138, "step": 15621} +{"accuracy/top1": 76.42740631103516, "data_time": 0.03199244778731774, "time": 1.2822754341980507, "step": 4} +{"lr": 6.856463687361792e-05, "data_time": 0.0015532493591308594, "loss": 0.37046386301517487, "time": 0.6303340911865234, "epoch": 5, "memory": 44138, "step": 15728} +{"lr": 6.821919068470432e-05, "data_time": 0.0017634153366088868, "loss": 0.3431585133075714, "time": 0.6313709020614624, "epoch": 5, "memory": 44138, "step": 15828} +{"lr": 6.787288979235407e-05, "data_time": 0.0015941619873046874, "loss": 0.3947469711303711, "time": 0.6309973001480103, "epoch": 5, "memory": 44138, "step": 15928} +{"lr": 6.752575658708793e-05, "data_time": 0.0019409656524658203, "loss": 0.3779486268758774, "time": 0.6306203842163086, "epoch": 5, "memory": 44138, "step": 16028} +{"lr": 6.71778135132408e-05, "data_time": 0.00177457332611084, "loss": 0.38871109783649443, "time": 0.630304479598999, "epoch": 5, "memory": 44138, "step": 16128} +{"lr": 6.682908306751074e-05, "data_time": 0.0024759531021118163, "loss": 0.35672386884689333, "time": 0.630770468711853, "epoch": 5, "memory": 44138, "step": 16228} +{"lr": 6.64795877975043e-05, "data_time": 0.002137279510498047, "loss": 0.36641431152820586, "time": 0.631433653831482, "epoch": 5, "memory": 44138, "step": 16328} +{"lr": 6.61293503002787e-05, "data_time": 0.0019015073776245117, "loss": 0.35590389668941497, "time": 0.6321771383285523, "epoch": 5, "memory": 44138, "step": 16428} +{"lr": 6.577839322088084e-05, "data_time": 0.0013942718505859375, "loss": 0.3608432561159134, "time": 0.6326557636260987, "epoch": 5, "memory": 44138, "step": 16528} +{"lr": 6.542673925088298e-05, "data_time": 0.001990056037902832, "loss": 0.37712846100330355, "time": 0.631695556640625, "epoch": 5, "memory": 44138, "step": 16628} +{"lr": 6.507441112691584e-05, "data_time": 0.0013965845108032226, "loss": 0.3802467107772827, "time": 0.6321236610412597, "epoch": 5, "memory": 44138, "step": 16728} +{"lr": 6.47214316291983e-05, "data_time": 0.0018872261047363282, "loss": 0.36903893053531645, "time": 0.6317490100860595, "epoch": 5, "memory": 44138, "step": 16828} +{"lr": 6.436782358006481e-05, "data_time": 0.0019455909729003905, "loss": 0.34476652145385744, "time": 0.6384008646011352, "epoch": 5, "memory": 44138, "step": 16928} +{"lr": 6.40136098424896e-05, "data_time": 0.00171051025390625, "loss": 0.3551800101995468, "time": 0.6332657098770141, "epoch": 5, "memory": 44138, "step": 17028} +{"lr": 6.365881331860828e-05, "data_time": 0.0017849206924438477, "loss": 0.3882934868335724, "time": 0.6307220220565796, "epoch": 5, "memory": 44138, "step": 17128} +{"lr": 6.330345694823739e-05, "data_time": 0.0021819353103637697, "loss": 0.36206912696361543, "time": 0.6309105157852173, "epoch": 5, "memory": 44138, "step": 17228} +{"lr": 6.29475637073908e-05, "data_time": 0.0017885923385620116, "loss": 0.37085449546575544, "time": 0.6303831338882446, "epoch": 5, "memory": 44138, "step": 17328} +{"lr": 6.259115660679482e-05, "data_time": 0.0017007827758789063, "loss": 0.3885393291711807, "time": 0.6403779983520508, "epoch": 5, "memory": 44138, "step": 17428} +{"lr": 6.223425869039963e-05, "data_time": 0.002181410789489746, "loss": 0.3782466769218445, "time": 0.6303816556930542, "epoch": 5, "memory": 44138, "step": 17528} +{"lr": 6.187689303389006e-05, "data_time": 0.002347421646118164, "loss": 0.3912949591875076, "time": 0.631197190284729, "epoch": 5, "memory": 44138, "step": 17628} +{"lr": 6.151908274319292e-05, "data_time": 0.0025192022323608397, "loss": 0.35448630303144457, "time": 0.6313546419143676, "epoch": 5, "memory": 44138, "step": 17728} +{"lr": 6.116085095298373e-05, "data_time": 0.001743459701538086, "loss": 0.3646813243627548, "time": 0.6316289901733398, "epoch": 5, "memory": 44138, "step": 17828} +{"lr": 6.080222082519007e-05, "data_time": 0.0020340442657470702, "loss": 0.3599780470132828, "time": 0.6305276155471802, "epoch": 5, "memory": 44138, "step": 17928} +{"lr": 6.0443215547495147e-05, "data_time": 0.0019669294357299804, "loss": 0.3566180944442749, "time": 0.6312304973602295, "epoch": 5, "memory": 44138, "step": 18028} +{"lr": 6.0083858331837714e-05, "data_time": 0.0015351057052612304, "loss": 0.38209750056266784, "time": 0.6313471317291259, "epoch": 5, "memory": 44138, "step": 18128} +{"lr": 5.97241724129116e-05, "data_time": 0.0016611099243164062, "loss": 0.3699777275323868, "time": 0.645628833770752, "epoch": 5, "memory": 44138, "step": 18228} +{"lr": 5.936418104666328e-05, "data_time": 0.0018738031387329102, "loss": 0.3627565443515778, "time": 0.6309934854507446, "epoch": 5, "memory": 44138, "step": 18328} +{"lr": 5.9003907508788295e-05, "data_time": 0.0014870643615722656, "loss": 0.374181854724884, "time": 0.6333391427993774, "epoch": 5, "memory": 44138, "step": 18428} +{"lr": 5.864337509322633e-05, "data_time": 0.0017216205596923828, "loss": 0.37459246814250946, "time": 0.6297661304473877, "epoch": 5, "memory": 44138, "step": 18528} +{"lr": 5.8282607110655194e-05, "data_time": 0.0015652179718017578, "loss": 0.3801063776016235, "time": 0.6301994800567627, "epoch": 5, "memory": 44138, "step": 18628} +{"lr": 5.7921626886983564e-05, "data_time": 0.0013895988464355468, "loss": 0.3741777092218399, "time": 0.6301333427429199, "epoch": 5, "memory": 44138, "step": 18728} +{"lr": 5.756045776184287e-05, "data_time": 0.0019222259521484374, "loss": 0.3763693571090698, "time": 0.6307146549224854, "epoch": 5, "memory": 44138, "step": 18828} +{"lr": 5.719912308707808e-05, "data_time": 0.002425527572631836, "loss": 0.38328993916511533, "time": 0.6307286977767944, "epoch": 5, "memory": 44138, "step": 18928} +{"lr": 5.683764622523808e-05, "data_time": 0.0015813589096069336, "loss": 0.3593562632799149, "time": 0.6307921409606934, "epoch": 5, "memory": 44138, "step": 19028} +{"lr": 5.647605054806502e-05, "data_time": 0.001833629608154297, "loss": 0.37703666687011717, "time": 0.6302602767944336, "epoch": 5, "memory": 44138, "step": 19128} +{"lr": 5.6114359434983206e-05, "data_time": 0.0016495466232299804, "loss": 0.379250505566597, "time": 0.6305471658706665, "epoch": 5, "memory": 44138, "step": 19228} +{"lr": 5.575259627158745e-05, "data_time": 0.0026583194732666014, "loss": 0.3611921459436417, "time": 0.6306925058364868, "epoch": 5, "memory": 44138, "step": 19328} +{"lr": 5.539078444813113e-05, "data_time": 0.0015018224716186524, "loss": 0.373409104347229, "time": 0.6310846328735351, "epoch": 5, "memory": 44138, "step": 19428} +{"lr": 5.5028947358013735e-05, "data_time": 0.0013743162155151368, "loss": 0.39499005377292634, "time": 0.6301813840866088, "epoch": 5, "memory": 44138, "step": 19528} +{"accuracy/top1": 75.8219985961914, "data_time": 0.031351796511946056, "time": 1.2812493381829098, "step": 5} +{"lr": 5.464178020340108e-05, "data_time": 0.0018122673034667968, "loss": 0.34920958876609803, "time": 0.6305671453475952, "epoch": 6, "memory": 44138, "step": 19635} +{"lr": 5.427996514795046e-05, "data_time": 0.001543140411376953, "loss": 0.36444980204105376, "time": 0.6307581424713135, "epoch": 6, "memory": 44138, "step": 19735} +{"lr": 5.3918196647260834e-05, "data_time": 0.0016943454742431641, "loss": 0.3770173192024231, "time": 0.6312621831893921, "epoch": 6, "memory": 44138, "step": 19835} +{"lr": 5.355649809193057e-05, "data_time": 0.0015299558639526368, "loss": 0.3631849020719528, "time": 0.6310431957244873, "epoch": 6, "memory": 44138, "step": 19935} +{"lr": 5.3194892868035765e-05, "data_time": 0.0016496419906616212, "loss": 0.3821670085191727, "time": 0.6306036710739136, "epoch": 6, "memory": 44138, "step": 20035} +{"lr": 5.283340435561795e-05, "data_time": 0.0016690492630004883, "loss": 0.3515541672706604, "time": 0.6410810708999634, "epoch": 6, "memory": 44138, "step": 20135} +{"lr": 5.2472055927172495e-05, "data_time": 0.001516270637512207, "loss": 0.37552843391895296, "time": 0.6313085794448853, "epoch": 6, "memory": 44138, "step": 20235} +{"lr": 5.2110870946137605e-05, "data_time": 0.0017879486083984375, "loss": 0.3542404919862747, "time": 0.6309683084487915, "epoch": 6, "memory": 44138, "step": 20335} +{"lr": 5.174987276538342e-05, "data_time": 0.0015929698944091796, "loss": 0.3852983355522156, "time": 0.6308411359786987, "epoch": 6, "memory": 44138, "step": 20435} +{"lr": 5.1389084725702425e-05, "data_time": 0.002024126052856445, "loss": 0.3785814821720123, "time": 0.6308462142944335, "epoch": 6, "memory": 44138, "step": 20535} +{"lr": 5.1028530154300045e-05, "data_time": 0.0017513036727905273, "loss": 0.3822389543056488, "time": 0.6312130689620972, "epoch": 6, "memory": 44138, "step": 20635} +{"lr": 5.066823236328663e-05, "data_time": 0.00167694091796875, "loss": 0.3927029311656952, "time": 0.631606912612915, "epoch": 6, "memory": 44138, "step": 20735} +{"lr": 5.030821464817006e-05, "data_time": 0.0017026662826538086, "loss": 0.37391828894615176, "time": 0.6300700664520263, "epoch": 6, "memory": 44138, "step": 20835} +{"lr": 4.99485002863494e-05, "data_time": 0.0016702651977539063, "loss": 0.38308541774749755, "time": 0.630565857887268, "epoch": 6, "memory": 44138, "step": 20935} +{"lr": 4.95891125356102e-05, "data_time": 0.001670098304748535, "loss": 0.38320526480674744, "time": 0.6307455062866211, "epoch": 6, "memory": 44138, "step": 21035} +{"lr": 4.923007463262053e-05, "data_time": 0.002085232734680176, "loss": 0.3893862396478653, "time": 0.6310501575469971, "epoch": 6, "memory": 44138, "step": 21135} +{"lr": 4.887140979142852e-05, "data_time": 0.0013954877853393556, "loss": 0.3912886053323746, "time": 0.6308657169342041, "epoch": 6, "memory": 44138, "step": 21235} +{"lr": 4.851314120196161e-05, "data_time": 0.0018996715545654297, "loss": 0.3496919900178909, "time": 0.6302508354187012, "epoch": 6, "memory": 44138, "step": 21335} +{"lr": 4.815529202852693e-05, "data_time": 0.0025632619857788087, "loss": 0.3953156232833862, "time": 0.6316039562225342, "epoch": 6, "memory": 44138, "step": 21435} +{"lr": 4.779788540831406e-05, "data_time": 0.0018422126770019532, "loss": 0.3585584104061127, "time": 0.6307833433151245, "epoch": 6, "memory": 44138, "step": 21535} +{"lr": 4.744094444989843e-05, "data_time": 0.0014423370361328126, "loss": 0.37024637758731843, "time": 0.6308498620986939, "epoch": 6, "memory": 44138, "step": 21635} +{"lr": 4.708449223174764e-05, "data_time": 0.001605224609375, "loss": 0.3841128945350647, "time": 0.6304964542388916, "epoch": 6, "memory": 44138, "step": 21735} +{"lr": 4.672855180072916e-05, "data_time": 0.001566171646118164, "loss": 0.36727446019649507, "time": 0.6305173635482788, "epoch": 6, "memory": 44138, "step": 21835} +{"lr": 4.6373146170620256e-05, "data_time": 0.00189361572265625, "loss": 0.36009582281112673, "time": 0.6301247596740722, "epoch": 6, "memory": 44138, "step": 21935} +{"lr": 4.6018298320619915e-05, "data_time": 0.0015498638153076173, "loss": 0.3688606321811676, "time": 0.630955696105957, "epoch": 6, "memory": 44138, "step": 22035} +{"lr": 4.566403119386318e-05, "data_time": 0.0015400409698486327, "loss": 0.39376863539218904, "time": 0.6312012910842896, "epoch": 6, "memory": 44138, "step": 22135} +{"lr": 4.531036769593769e-05, "data_time": 0.002109670639038086, "loss": 0.35376296639442445, "time": 0.631488585472107, "epoch": 6, "memory": 44138, "step": 22235} +{"lr": 4.4957330693402685e-05, "data_time": 0.002283740043640137, "loss": 0.34284416437149046, "time": 0.6309876441955566, "epoch": 6, "memory": 44138, "step": 22335} +{"lr": 4.460494301231059e-05, "data_time": 0.0016895532608032227, "loss": 0.380840128660202, "time": 0.6308343887329102, "epoch": 6, "memory": 44138, "step": 22435} +{"lr": 4.425322743673109e-05, "data_time": 0.0016279220581054688, "loss": 0.35961690843105315, "time": 0.6309692621231079, "epoch": 6, "memory": 44138, "step": 22535} +{"lr": 4.390220670727812e-05, "data_time": 0.0016263723373413086, "loss": 0.3624699264764786, "time": 0.6305348157882691, "epoch": 6, "memory": 44138, "step": 22635} +{"lr": 4.3551903519639325e-05, "data_time": 0.0019348859786987305, "loss": 0.3400824010372162, "time": 0.630367922782898, "epoch": 6, "memory": 44138, "step": 22735} +{"lr": 4.3202340523108894e-05, "data_time": 0.0016979217529296876, "loss": 0.3596193194389343, "time": 0.6301675081253052, "epoch": 6, "memory": 44138, "step": 22835} +{"lr": 4.285354031912302e-05, "data_time": 0.001671600341796875, "loss": 0.34684542417526243, "time": 0.629757285118103, "epoch": 6, "memory": 44138, "step": 22935} +{"lr": 4.2505525459798544e-05, "data_time": 0.0016098737716674805, "loss": 0.3625393629074097, "time": 0.6311562061309814, "epoch": 6, "memory": 44138, "step": 23035} +{"lr": 4.2158318446474836e-05, "data_time": 0.0014756202697753906, "loss": 0.3483108252286911, "time": 0.6312747240066529, "epoch": 6, "memory": 44138, "step": 23135} +{"lr": 4.181194172825905e-05, "data_time": 0.0018828153610229493, "loss": 0.3671716511249542, "time": 0.6319789409637451, "epoch": 6, "memory": 44138, "step": 23235} +{"lr": 4.146641770057448e-05, "data_time": 0.0018091201782226562, "loss": 0.37112095654010774, "time": 0.6314461708068848, "epoch": 6, "memory": 44138, "step": 23335} +{"lr": 4.112176870371267e-05, "data_time": 0.0014128684997558594, "loss": 0.3523112118244171, "time": 0.6315903425216675, "epoch": 6, "memory": 44138, "step": 23435} +{"accuracy/top1": 75.67586517333984, "data_time": 0.03196217684910215, "time": 1.2841334671809757, "step": 6} +{"lr": 4.0753988582732786e-05, "data_time": 0.001938033103942871, "loss": 0.3583205431699753, "time": 0.6311803340911866, "epoch": 7, "memory": 44138, "step": 23542} +{"lr": 4.0411221639286815e-05, "data_time": 0.0018255949020385743, "loss": 0.35708620846271516, "time": 0.6309070348739624, "epoch": 7, "memory": 44138, "step": 23642} +{"lr": 4.006939795169518e-05, "data_time": 0.002001237869262695, "loss": 0.3720770120620728, "time": 0.6302164554595947, "epoch": 7, "memory": 44138, "step": 23742} +{"lr": 3.972853962099913e-05, "data_time": 0.0015237808227539062, "loss": 0.372759947180748, "time": 0.6311291456222534, "epoch": 7, "memory": 44138, "step": 23842} +{"lr": 3.93886686858237e-05, "data_time": 0.0014948606491088866, "loss": 0.35073335468769073, "time": 0.6307839632034302, "epoch": 7, "memory": 44138, "step": 23942} +{"lr": 3.904980712095259e-05, "data_time": 0.001529860496520996, "loss": 0.3819312959909439, "time": 0.6303291082382202, "epoch": 7, "memory": 44138, "step": 24042} +{"lr": 3.871197683590748e-05, "data_time": 0.0019331693649291993, "loss": 0.4172582268714905, "time": 0.6300901412963867, "epoch": 7, "memory": 44138, "step": 24142} +{"lr": 3.837519967353125e-05, "data_time": 0.0018826723098754883, "loss": 0.3877938807010651, "time": 0.6306633710861206, "epoch": 7, "memory": 44138, "step": 24242} +{"lr": 3.8039497408575855e-05, "data_time": 0.0014815330505371094, "loss": 0.38511968553066256, "time": 0.630943751335144, "epoch": 7, "memory": 44138, "step": 24342} +{"lr": 3.770489174629432e-05, "data_time": 0.0017430305480957032, "loss": 0.35685568749904634, "time": 0.6305532455444336, "epoch": 7, "memory": 44138, "step": 24442} +{"lr": 3.737140432103754e-05, "data_time": 0.0018526792526245117, "loss": 0.40219601094722746, "time": 0.6301730632781982, "epoch": 7, "memory": 44138, "step": 24542} +{"lr": 3.703905669485545e-05, "data_time": 0.0015938758850097656, "loss": 0.36595679819583893, "time": 0.6310053110122681, "epoch": 7, "memory": 44138, "step": 24642} +{"lr": 3.6707870356102706e-05, "data_time": 0.0015116691589355468, "loss": 0.39236302971839904, "time": 0.630971622467041, "epoch": 7, "memory": 44138, "step": 24742} +{"lr": 3.6377866718049484e-05, "data_time": 0.002404189109802246, "loss": 0.3813491195440292, "time": 0.6325625419616699, "epoch": 7, "memory": 44138, "step": 24842} +{"lr": 3.6049067117496965e-05, "data_time": 0.002016139030456543, "loss": 0.3724536061286926, "time": 0.6303442239761352, "epoch": 7, "memory": 44138, "step": 24942} +{"lr": 3.5721492813397755e-05, "data_time": 0.001612710952758789, "loss": 0.3616188108921051, "time": 0.6305454492568969, "epoch": 7, "memory": 44138, "step": 25042} +{"lr": 3.539516498548151e-05, "data_time": 0.0016888618469238282, "loss": 0.3745671182870865, "time": 0.6316226720809937, "epoch": 7, "memory": 44138, "step": 25142} +{"lr": 3.507010473288518e-05, "data_time": 0.0014406442642211914, "loss": 0.37493705153465273, "time": 0.6312808275222779, "epoch": 7, "memory": 44138, "step": 25242} +{"lr": 3.474633307278909e-05, "data_time": 0.0016297578811645507, "loss": 0.35371659994125365, "time": 0.6316396951675415, "epoch": 7, "memory": 44138, "step": 25342} +{"lr": 3.4423870939058177e-05, "data_time": 0.0016299247741699218, "loss": 0.35014588236808775, "time": 0.6309483528137207, "epoch": 7, "memory": 44138, "step": 25442} +{"lr": 3.410273918088805e-05, "data_time": 0.0017478466033935547, "loss": 0.36910676062107084, "time": 0.6305905818939209, "epoch": 7, "memory": 44138, "step": 25542} +{"lr": 3.37829585614574e-05, "data_time": 0.001750349998474121, "loss": 0.376947021484375, "time": 0.6300114631652832, "epoch": 7, "memory": 44138, "step": 25642} +{"lr": 3.3464549756585216e-05, "data_time": 0.0014777421951293946, "loss": 0.32904700338840487, "time": 0.6318575859069824, "epoch": 7, "memory": 44138, "step": 25742} +{"lr": 3.314753335339416e-05, "data_time": 0.0016019105911254882, "loss": 0.3829327464103699, "time": 0.631765604019165, "epoch": 7, "memory": 44138, "step": 25842} +{"lr": 3.2831929848979394e-05, "data_time": 0.002126789093017578, "loss": 0.38341371715068817, "time": 0.6300968170166016, "epoch": 7, "memory": 44138, "step": 25942} +{"lr": 3.251775964908332e-05, "data_time": 0.0015276670455932617, "loss": 0.359295055270195, "time": 0.6310170888900757, "epoch": 7, "memory": 44138, "step": 26042} +{"lr": 3.22050430667762e-05, "data_time": 0.0015214204788208008, "loss": 0.3614420384168625, "time": 0.6304879903793335, "epoch": 7, "memory": 44138, "step": 26142} +{"lr": 3.189380032114285e-05, "data_time": 0.001581263542175293, "loss": 0.38451360166072845, "time": 0.6299646854400635, "epoch": 7, "memory": 44138, "step": 26242} +{"lr": 3.1584051535975324e-05, "data_time": 0.0015254735946655274, "loss": 0.3565573751926422, "time": 0.6304850816726685, "epoch": 7, "memory": 44138, "step": 26342} +{"lr": 3.127581673847175e-05, "data_time": 0.001414346694946289, "loss": 0.3782190531492233, "time": 0.6310838222503662, "epoch": 7, "memory": 44138, "step": 26442} +{"lr": 3.0969115857941555e-05, "data_time": 0.0016266822814941407, "loss": 0.38487481474876406, "time": 0.6298484086990357, "epoch": 7, "memory": 44138, "step": 26542} +{"lr": 3.0663968724516654e-05, "data_time": 0.001608586311340332, "loss": 0.36543085277080534, "time": 0.6297093391418457, "epoch": 7, "memory": 44138, "step": 26642} +{"lr": 3.0360395067869668e-05, "data_time": 0.0016102075576782226, "loss": 0.3948678284883499, "time": 0.629201602935791, "epoch": 7, "memory": 44138, "step": 26742} +{"lr": 3.0058414515938063e-05, "data_time": 0.0021617650985717774, "loss": 0.38861264288425446, "time": 0.6379568815231323, "epoch": 7, "memory": 44138, "step": 26842} +{"lr": 2.975804659365501e-05, "data_time": 0.0018576860427856445, "loss": 0.35465694665908815, "time": 0.6299306392669678, "epoch": 7, "memory": 44138, "step": 26942} +{"lr": 2.945931072168714e-05, "data_time": 0.001920795440673828, "loss": 0.3702373683452606, "time": 0.6300881862640381, "epoch": 7, "memory": 44138, "step": 27042} +{"lr": 2.9162226215178855e-05, "data_time": 0.0019040822982788086, "loss": 0.357158762216568, "time": 0.6302624702453613, "epoch": 7, "memory": 44138, "step": 27142} +{"lr": 2.8866812282503322e-05, "data_time": 0.001497507095336914, "loss": 0.38576229214668273, "time": 0.6305846691131591, "epoch": 7, "memory": 44138, "step": 27242} +{"lr": 2.8573088024020838e-05, "data_time": 0.00119326114654541, "loss": 0.3747771352529526, "time": 0.6289780378341675, "epoch": 7, "memory": 44138, "step": 27342} +{"accuracy/top1": 75.43927001953125, "data_time": 0.03220138056524869, "time": 1.2809336349881928, "step": 7} +{"lr": 2.8260695817390092e-05, "data_time": 0.0014861583709716796, "loss": 0.3698581427335739, "time": 0.6313601732254028, "epoch": 8, "memory": 44138, "step": 27449} +{"lr": 2.7970529402570967e-05, "data_time": 0.0015247583389282227, "loss": 0.3719507813453674, "time": 0.6313715219497681, "epoch": 8, "memory": 44138, "step": 27549} +{"lr": 2.7682110612244214e-05, "data_time": 0.001616692543029785, "loss": 0.39879120886325836, "time": 0.6307795763015747, "epoch": 8, "memory": 44138, "step": 27649} +{"lr": 2.7395458094490283e-05, "data_time": 0.002065730094909668, "loss": 0.36551759839057923, "time": 0.6318770408630371, "epoch": 8, "memory": 44138, "step": 27749} +{"lr": 2.7110590383189015e-05, "data_time": 0.0016478300094604492, "loss": 0.38542689085006715, "time": 0.6304060935974121, "epoch": 8, "memory": 44138, "step": 27849} +{"lr": 2.6827525896821397e-05, "data_time": 0.0019791841506958006, "loss": 0.3750642895698547, "time": 0.6311421155929565, "epoch": 8, "memory": 44138, "step": 27949} +{"lr": 2.6546282937278552e-05, "data_time": 0.0021829128265380858, "loss": 0.37531176805496214, "time": 0.6316208124160767, "epoch": 8, "memory": 44138, "step": 28049} +{"lr": 2.6266879688678606e-05, "data_time": 0.0014534711837768555, "loss": 0.3732608675956726, "time": 0.6311471462249756, "epoch": 8, "memory": 44138, "step": 28149} +{"lr": 2.5989334216190738e-05, "data_time": 0.0015685081481933594, "loss": 0.38682337403297423, "time": 0.6312675476074219, "epoch": 8, "memory": 44138, "step": 28249} +{"lr": 2.571366446486732e-05, "data_time": 0.0018526792526245117, "loss": 0.3724817126989365, "time": 0.6319009780883789, "epoch": 8, "memory": 44138, "step": 28349} +{"lr": 2.5439888258483665e-05, "data_time": 0.0016022443771362305, "loss": 0.35468791127204896, "time": 0.6319311141967774, "epoch": 8, "memory": 44138, "step": 28449} +{"lr": 2.5168023298385384e-05, "data_time": 0.001462268829345703, "loss": 0.3673750728368759, "time": 0.6326402425765991, "epoch": 8, "memory": 44138, "step": 28549} +{"lr": 2.489808716234412e-05, "data_time": 0.0016788244247436523, "loss": 0.37713308334350587, "time": 0.6322790622711182, "epoch": 8, "memory": 44138, "step": 28649} +{"lr": 2.463009730342111e-05, "data_time": 0.0014983654022216798, "loss": 0.3820088759064674, "time": 0.6319272518157959, "epoch": 8, "memory": 44138, "step": 28749} +{"lr": 2.436407104883849e-05, "data_time": 0.0017088651657104492, "loss": 0.3759547084569931, "time": 0.6316488742828369, "epoch": 8, "memory": 44138, "step": 28849} +{"lr": 2.4100025598859092e-05, "data_time": 0.0016340494155883789, "loss": 0.3572659969329834, "time": 0.6310683965682984, "epoch": 8, "memory": 44138, "step": 28949} +{"lr": 2.3837978025674284e-05, "data_time": 0.0015425920486450196, "loss": 0.35619386434555056, "time": 0.631263279914856, "epoch": 8, "memory": 44138, "step": 29049} +{"lr": 2.357794527230027e-05, "data_time": 0.002205491065979004, "loss": 0.3464666038751602, "time": 0.6315906047821045, "epoch": 8, "memory": 44138, "step": 29149} +{"lr": 2.33199441514825e-05, "data_time": 0.0013254165649414062, "loss": 0.356190425157547, "time": 0.6304213047027588, "epoch": 8, "memory": 44138, "step": 29249} +{"lr": 2.306399134460859e-05, "data_time": 0.0018564939498901367, "loss": 0.38022026121616365, "time": 0.6332060098648071, "epoch": 8, "memory": 44138, "step": 29349} +{"lr": 2.281010340062995e-05, "data_time": 0.0021481037139892576, "loss": 0.3789412349462509, "time": 0.632056736946106, "epoch": 8, "memory": 44138, "step": 29449} +{"lr": 2.2558296734991603e-05, "data_time": 0.0018558502197265625, "loss": 0.38903270065784457, "time": 0.6311423540115356, "epoch": 8, "memory": 44138, "step": 29549} +{"lr": 2.2308587628570915e-05, "data_time": 0.002602362632751465, "loss": 0.36461024880409243, "time": 0.6321547746658325, "epoch": 8, "memory": 44138, "step": 29649} +{"lr": 2.2060992226624955e-05, "data_time": 0.0015051603317260743, "loss": 0.381944078207016, "time": 0.6306434154510498, "epoch": 8, "memory": 44138, "step": 29749} +{"lr": 2.181552653774648e-05, "data_time": 0.0018649101257324219, "loss": 0.38841715455055237, "time": 0.631217360496521, "epoch": 8, "memory": 44138, "step": 29849} +{"lr": 2.1572206432829035e-05, "data_time": 0.001702117919921875, "loss": 0.4009664624929428, "time": 0.6303469657897949, "epoch": 8, "memory": 44138, "step": 29949} +{"lr": 2.1331047644040754e-05, "data_time": 0.0016013860702514648, "loss": 0.3727075010538101, "time": 0.630508279800415, "epoch": 8, "memory": 44138, "step": 30049} +{"lr": 2.109206576380705e-05, "data_time": 0.002045083045959473, "loss": 0.37341995239257814, "time": 0.6308868885040283, "epoch": 8, "memory": 44138, "step": 30149} +{"lr": 2.0855276243802687e-05, "data_time": 0.0016182899475097657, "loss": 0.3565258502960205, "time": 0.6321156978607178, "epoch": 8, "memory": 44138, "step": 30249} +{"lr": 2.0620694393952487e-05, "data_time": 0.0026824474334716797, "loss": 0.3869649887084961, "time": 0.6317596673965454, "epoch": 8, "memory": 44138, "step": 30349} +{"lr": 2.038833538144169e-05, "data_time": 0.0017143726348876954, "loss": 0.3577035516500473, "time": 0.6310218811035156, "epoch": 8, "memory": 44138, "step": 30449} +{"lr": 2.0158214229735154e-05, "data_time": 0.0015231609344482423, "loss": 0.35234747231006625, "time": 0.6306694269180297, "epoch": 8, "memory": 44138, "step": 30549} +{"lr": 1.9930345817605977e-05, "data_time": 0.0014863967895507812, "loss": 0.391020405292511, "time": 0.6305213689804077, "epoch": 8, "memory": 44138, "step": 30649} +{"lr": 1.970474487817358e-05, "data_time": 0.0017902851104736328, "loss": 0.38438333570957184, "time": 0.6305299043655396, "epoch": 8, "memory": 44138, "step": 30749} +{"lr": 1.9481425997951092e-05, "data_time": 0.0020719289779663084, "loss": 0.3644045442342758, "time": 0.6309171199798584, "epoch": 8, "memory": 44138, "step": 30849} diff --git a/clip_large_pretrain_4x256_sdv2_lr1e-4_rub_/20230603_200211/vis_data/vis_image/0000000.png_1.png b/clip_large_pretrain_4x256_sdv2_lr1e-4_rub_/20230603_200211/vis_data/vis_image/0000000.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..bd89e1540fdc8b4e655714743642ccbae9b8cafc Binary files /dev/null and b/clip_large_pretrain_4x256_sdv2_lr1e-4_rub_/20230603_200211/vis_data/vis_image/0000000.png_1.png differ diff --git a/clip_large_pretrain_4x256_sdv2_lr1e-4_rub_/20230603_200211/vis_data/vis_image/0000000.png_2.png b/clip_large_pretrain_4x256_sdv2_lr1e-4_rub_/20230603_200211/vis_data/vis_image/0000000.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..9a3a5ef7a24aab16837b9315be42144d77e2ad71 Binary files /dev/null and b/clip_large_pretrain_4x256_sdv2_lr1e-4_rub_/20230603_200211/vis_data/vis_image/0000000.png_2.png differ diff --git a/clip_large_pretrain_4x256_sdv2_lr1e-4_rub_/20230603_200211/vis_data/vis_image/0000000.png_3.png b/clip_large_pretrain_4x256_sdv2_lr1e-4_rub_/20230603_200211/vis_data/vis_image/0000000.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..fe4f5a0fb37aba516c1be8b3a378e1fd7d0c551b Binary files /dev/null and b/clip_large_pretrain_4x256_sdv2_lr1e-4_rub_/20230603_200211/vis_data/vis_image/0000000.png_3.png differ diff --git a/clip_large_pretrain_4x256_sdv2_lr1e-4_rub_/20230603_200211/vis_data/vis_image/0000000.png_4.png b/clip_large_pretrain_4x256_sdv2_lr1e-4_rub_/20230603_200211/vis_data/vis_image/0000000.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..f3672f80359a20044242b397d1bb67a51d1b90d5 Binary files /dev/null and b/clip_large_pretrain_4x256_sdv2_lr1e-4_rub_/20230603_200211/vis_data/vis_image/0000000.png_4.png differ diff --git a/clip_large_pretrain_4x256_sdv2_lr1e-4_rub_/20230603_200211/vis_data/vis_image/0000000.png_5.png b/clip_large_pretrain_4x256_sdv2_lr1e-4_rub_/20230603_200211/vis_data/vis_image/0000000.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..d6bc1cd74ea60b43933b566e4410142676b3756a Binary files /dev/null and b/clip_large_pretrain_4x256_sdv2_lr1e-4_rub_/20230603_200211/vis_data/vis_image/0000000.png_5.png differ diff --git a/clip_large_pretrain_4x256_sdv2_lr1e-4_rub_/20230603_200211/vis_data/vis_image/0000000.png_6.png b/clip_large_pretrain_4x256_sdv2_lr1e-4_rub_/20230603_200211/vis_data/vis_image/0000000.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..fa382c63e596e186cc2e6496256bfd6abe849e85 Binary files /dev/null and b/clip_large_pretrain_4x256_sdv2_lr1e-4_rub_/20230603_200211/vis_data/vis_image/0000000.png_6.png differ diff --git a/clip_large_pretrain_4x256_sdv2_lr1e-4_rub_/20230603_200211/vis_data/vis_image/0000000.png_7.png b/clip_large_pretrain_4x256_sdv2_lr1e-4_rub_/20230603_200211/vis_data/vis_image/0000000.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..b74b80782b17f27a70030bb0d0cf97c63598bc58 Binary files /dev/null and b/clip_large_pretrain_4x256_sdv2_lr1e-4_rub_/20230603_200211/vis_data/vis_image/0000000.png_7.png differ diff --git a/clip_large_pretrain_4x256_sdv2_lr1e-4_rub_/20230603_200211/vis_data/vis_image/000005174.jpg_1.png b/clip_large_pretrain_4x256_sdv2_lr1e-4_rub_/20230603_200211/vis_data/vis_image/000005174.jpg_1.png new file mode 100644 index 0000000000000000000000000000000000000000..d22e6dc1fd05b02eb46613de4603f715f8d5f7fb Binary files /dev/null and b/clip_large_pretrain_4x256_sdv2_lr1e-4_rub_/20230603_200211/vis_data/vis_image/000005174.jpg_1.png differ diff --git a/clip_large_pretrain_4x256_sdv2_lr1e-4_rub_/20230603_200211/vis_data/vis_image/000005174.jpg_2.png b/clip_large_pretrain_4x256_sdv2_lr1e-4_rub_/20230603_200211/vis_data/vis_image/000005174.jpg_2.png new file mode 100644 index 0000000000000000000000000000000000000000..93ce04468f55527b5484b64b7af89ad1aaf43758 Binary files /dev/null and b/clip_large_pretrain_4x256_sdv2_lr1e-4_rub_/20230603_200211/vis_data/vis_image/000005174.jpg_2.png differ diff --git a/clip_large_pretrain_4x256_sdv2_lr1e-4_rub_/20230603_200211/vis_data/vis_image/000005174.jpg_3.png b/clip_large_pretrain_4x256_sdv2_lr1e-4_rub_/20230603_200211/vis_data/vis_image/000005174.jpg_3.png new file mode 100644 index 0000000000000000000000000000000000000000..b7a33b151f76da333da972baf85778a546dc86e1 Binary files /dev/null and b/clip_large_pretrain_4x256_sdv2_lr1e-4_rub_/20230603_200211/vis_data/vis_image/000005174.jpg_3.png differ diff --git a/clip_large_pretrain_4x256_sdv2_lr1e-4_rub_/20230603_200211/vis_data/vis_image/000005174.jpg_4.png b/clip_large_pretrain_4x256_sdv2_lr1e-4_rub_/20230603_200211/vis_data/vis_image/000005174.jpg_4.png new file mode 100644 index 0000000000000000000000000000000000000000..513002685e597b1881d94c797c093b60499b175f Binary files /dev/null and b/clip_large_pretrain_4x256_sdv2_lr1e-4_rub_/20230603_200211/vis_data/vis_image/000005174.jpg_4.png differ diff --git a/clip_large_pretrain_4x256_sdv2_lr1e-4_rub_/20230603_200211/vis_data/vis_image/000005174.jpg_5.png b/clip_large_pretrain_4x256_sdv2_lr1e-4_rub_/20230603_200211/vis_data/vis_image/000005174.jpg_5.png new file mode 100644 index 0000000000000000000000000000000000000000..7c10b8264c3ff6bcdfc27f075e06cf24390af258 Binary files /dev/null and b/clip_large_pretrain_4x256_sdv2_lr1e-4_rub_/20230603_200211/vis_data/vis_image/000005174.jpg_5.png differ diff --git a/clip_large_pretrain_4x256_sdv2_lr1e-4_rub_/20230603_200211/vis_data/vis_image/000005174.jpg_6.png b/clip_large_pretrain_4x256_sdv2_lr1e-4_rub_/20230603_200211/vis_data/vis_image/000005174.jpg_6.png new file mode 100644 index 0000000000000000000000000000000000000000..dd17f646f64ccfb6d2e0f4b0a2cc5a500f2b8dfb Binary files /dev/null and b/clip_large_pretrain_4x256_sdv2_lr1e-4_rub_/20230603_200211/vis_data/vis_image/000005174.jpg_6.png differ diff --git a/clip_large_pretrain_4x256_sdv2_lr1e-4_rub_/20230603_200211/vis_data/vis_image/000005174.jpg_7.png b/clip_large_pretrain_4x256_sdv2_lr1e-4_rub_/20230603_200211/vis_data/vis_image/000005174.jpg_7.png new file mode 100644 index 0000000000000000000000000000000000000000..3466abd6c8c63e73541a7d3d892c04e1ffedcef6 Binary files /dev/null and b/clip_large_pretrain_4x256_sdv2_lr1e-4_rub_/20230603_200211/vis_data/vis_image/000005174.jpg_7.png differ diff --git a/clip_large_pretrain_4x256_sdv2_lr1e-4_rub_/clip_large_pretrain_4x256_sdv2_lr1e-4.py b/clip_large_pretrain_4x256_sdv2_lr1e-4_rub_/clip_large_pretrain_4x256_sdv2_lr1e-4.py new file mode 100644 index 0000000000000000000000000000000000000000..19e31047e13d469a3faf46963faf1137c9d94750 --- /dev/null +++ b/clip_large_pretrain_4x256_sdv2_lr1e-4_rub_/clip_large_pretrain_4x256_sdv2_lr1e-4.py @@ -0,0 +1,228 @@ +optim_wrapper = dict( + optimizer=dict( + type='AdamW', lr=0.0001, weight_decay=0.3, _scope_='mmpretrain'), + paramwise_cfg=dict( + custom_keys=dict({ + '.cls_token': dict(decay_mult=0.0), + '.pos_embed': dict(decay_mult=0.0) + })), + type='AmpOptimWrapper', + dtype='bfloat16', + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=False, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=4096) +model = dict( + type='ImageClassifier', + backbone=dict( + frozen_stages=24, + type='VisionTransformer', + arch='l', + img_size=224, + patch_size=14, + drop_rate=0.1, + pre_norm=True, + init_cfg=dict( + type='Pretrained', + checkpoint='ckpt/openclip-ViT-L-14.pth', + prefix='backbone')), + neck=dict(type='CLIPProjection', in_channels=1024, out_channels=768), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=768, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + init_cfg=None), + init_cfg=dict( + type='TruncNormal', layer=['Conv2d', 'Linear'], std=0.02, bias=0.0), + train_cfg=None) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=128, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stablediffusionV2-1-dpmsolver-25-1m.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/cc1m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=128, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV2-1-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=128, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV2-1-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/clip_large_pretrain_4x256_sdv2_lr1e-4' diff --git a/clip_large_pretrain_4x256_sdv2_lr1e-4_rub_/epoch_1.pth b/clip_large_pretrain_4x256_sdv2_lr1e-4_rub_/epoch_1.pth new file mode 100644 index 0000000000000000000000000000000000000000..d327250ba315a7284facc27c8b9d1e5555abe126 --- /dev/null +++ b/clip_large_pretrain_4x256_sdv2_lr1e-4_rub_/epoch_1.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:436cf2675f675ee1933fef7d05680b3b21d98d93ff8770ba1ff726fd9728652b +size 2438633785 diff --git a/clip_large_pretrain_4x256_sdv2_lr1e-4_rub_/epoch_2.pth b/clip_large_pretrain_4x256_sdv2_lr1e-4_rub_/epoch_2.pth new file mode 100644 index 0000000000000000000000000000000000000000..18e74cd2446ad1794cc95cd947044cc3c13bdeae --- /dev/null +++ b/clip_large_pretrain_4x256_sdv2_lr1e-4_rub_/epoch_2.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:83cb109895770ae786e34c5c9abec34158710fa6efa63aa05e809e3896436cee +size 2438918649 diff --git a/clip_large_pretrain_4x256_sdv2_lr1e-4_rub_/epoch_3.pth b/clip_large_pretrain_4x256_sdv2_lr1e-4_rub_/epoch_3.pth new file mode 100644 index 0000000000000000000000000000000000000000..7cdaba89e0fb84f70d3d2142c99caecbdefbc265 --- /dev/null +++ b/clip_large_pretrain_4x256_sdv2_lr1e-4_rub_/epoch_3.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:4c3e486b2538128834a604c1af9a76459b2248e36f8df5b4006f2d888e2be23c +size 2439203065 diff --git a/clip_large_pretrain_4x256_sdv2_lr1e-4_rub_/epoch_4.pth b/clip_large_pretrain_4x256_sdv2_lr1e-4_rub_/epoch_4.pth new file mode 100644 index 0000000000000000000000000000000000000000..cd5a4ac09b43d53a0b0e04c76a0a94f838f86238 --- /dev/null +++ b/clip_large_pretrain_4x256_sdv2_lr1e-4_rub_/epoch_4.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:0f1d37efbdaee14cada92e03542fe06c7a89e5879a12a3ebdf9ba0adc65a0001 +size 2439486841 diff --git a/clip_large_pretrain_4x256_sdv2_lr1e-4_rub_/epoch_5.pth b/clip_large_pretrain_4x256_sdv2_lr1e-4_rub_/epoch_5.pth new file mode 100644 index 0000000000000000000000000000000000000000..94847374d8419cbd6e9adab81f7cecbc52d949fe --- /dev/null +++ b/clip_large_pretrain_4x256_sdv2_lr1e-4_rub_/epoch_5.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:cf2a5b883f3f12540f4b129c919c8648355b7f603507782c6171fa760a7bafc9 +size 2439770553 diff --git a/clip_large_pretrain_4x256_sdv2_lr1e-4_rub_/epoch_6.pth b/clip_large_pretrain_4x256_sdv2_lr1e-4_rub_/epoch_6.pth new file mode 100644 index 0000000000000000000000000000000000000000..2102f8b19b259aef8df483736ff306c1ec4bc0a7 --- /dev/null +++ b/clip_large_pretrain_4x256_sdv2_lr1e-4_rub_/epoch_6.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:45b506ce72a98e33aacd263a99530b4e68b36904b5217e0621c3cbb7361d152b +size 2440054329 diff --git a/clip_large_pretrain_4x256_sdv2_lr1e-4_rub_/epoch_7.pth b/clip_large_pretrain_4x256_sdv2_lr1e-4_rub_/epoch_7.pth new file mode 100644 index 0000000000000000000000000000000000000000..215bc3e95ab02245a9a9f050d78b6f3d273f2310 --- /dev/null +++ b/clip_large_pretrain_4x256_sdv2_lr1e-4_rub_/epoch_7.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:43e91bb99f60cb07896933ef9aaa89171b2f47d7ac314621eb1d52b3a0f09859 +size 2440338617 diff --git a/clip_large_pretrain_4x256_sdv2_lr1e-4_rub_/last_checkpoint b/clip_large_pretrain_4x256_sdv2_lr1e-4_rub_/last_checkpoint new file mode 100644 index 0000000000000000000000000000000000000000..76847cb1b2b9cb5f30ea20d69ac2a8987deed599 --- /dev/null +++ b/clip_large_pretrain_4x256_sdv2_lr1e-4_rub_/last_checkpoint @@ -0,0 +1 @@ +/mnt/petrelfs/luzeyu/workspace/fakebench/mmpretrain/workdir/clip_large_pretrain_4x256_sdv2_lr1e-4/epoch_7.pth \ No newline at end of file diff --git a/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm/20230604_023242/20230604_023242.log b/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm/20230604_023242/20230604_023242.log new file mode 100644 index 0000000000000000000000000000000000000000..23153b0e5ac58fac9b685cebfddfb72365072230 --- /dev/null +++ b/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm/20230604_023242/20230604_023242.log @@ -0,0 +1,1709 @@ +2023/06/04 02:32:47 - mmengine - INFO - +------------------------------------------------------------ +System environment: + sys.platform: linux + Python: 3.10.9 (main, Mar 8 2023, 10:47:38) [GCC 11.2.0] + CUDA available: True + numpy_random_seed: 1918878949 + GPU 0,1,2,3: NVIDIA A100-SXM4-80GB + CUDA_HOME: /mnt/petrelfs/share/cuda-11.6 + NVCC: Cuda compilation tools, release 11.6, V11.6.124 + GCC: gcc (GCC) 7.5.0 + PyTorch: 1.13.1 + PyTorch compiling details: PyTorch built with: + - GCC 9.3 + - C++ Version: 201402 + - Intel(R) oneAPI Math Kernel Library Version 2021.4-Product Build 20210904 for Intel(R) 64 architecture applications + - Intel(R) MKL-DNN v2.6.0 (Git Hash 52b5f107dd9cf10910aaa19cb47f3abf9b349815) + - OpenMP 201511 (a.k.a. OpenMP 4.5) + - LAPACK is enabled (usually provided by MKL) + - NNPACK is enabled + - CPU capability usage: AVX2 + - CUDA Runtime 11.6 + - NVCC architecture flags: -gencode;arch=compute_37,code=sm_37;-gencode;arch=compute_50,code=sm_50;-gencode;arch=compute_60,code=sm_60;-gencode;arch=compute_61,code=sm_61;-gencode;arch=compute_70,code=sm_70;-gencode;arch=compute_75,code=sm_75;-gencode;arch=compute_80,code=sm_80;-gencode;arch=compute_86,code=sm_86;-gencode;arch=compute_37,code=compute_37 + - CuDNN 8.3.2 (built against CUDA 11.5) + - Magma 2.6.1 + - Build settings: BLAS_INFO=mkl, BUILD_TYPE=Release, CUDA_VERSION=11.6, CUDNN_VERSION=8.3.2, CXX_COMPILER=/opt/rh/devtoolset-9/root/usr/bin/c++, CXX_FLAGS= -fabi-version=11 -Wno-deprecated -fvisibility-inlines-hidden -DUSE_PTHREADPOOL -fopenmp -DNDEBUG -DUSE_KINETO -DUSE_FBGEMM -DUSE_QNNPACK -DUSE_PYTORCH_QNNPACK -DUSE_XNNPACK -DSYMBOLICATE_MOBILE_DEBUG_HANDLE -DEDGE_PROFILER_USE_KINETO -O2 -fPIC -Wno-narrowing -Wall -Wextra -Werror=return-type -Werror=non-virtual-dtor -Wno-missing-field-initializers -Wno-type-limits -Wno-array-bounds -Wno-unknown-pragmas -Wunused-local-typedefs -Wno-unused-parameter -Wno-unused-function -Wno-unused-result -Wno-strict-overflow -Wno-strict-aliasing -Wno-error=deprecated-declarations -Wno-stringop-overflow -Wno-psabi -Wno-error=pedantic -Wno-error=redundant-decls -Wno-error=old-style-cast -fdiagnostics-color=always -faligned-new -Wno-unused-but-set-variable -Wno-maybe-uninitialized -fno-math-errno -fno-trapping-math -Werror=format -Werror=cast-function-type -Wno-stringop-overflow, LAPACK_INFO=mkl, PERF_WITH_AVX=1, PERF_WITH_AVX2=1, PERF_WITH_AVX512=1, TORCH_VERSION=1.13.1, USE_CUDA=ON, USE_CUDNN=ON, USE_EXCEPTION_PTR=1, USE_GFLAGS=OFF, USE_GLOG=OFF, USE_MKL=ON, USE_MKLDNN=ON, USE_MPI=OFF, USE_NCCL=ON, USE_NNPACK=ON, USE_OPENMP=ON, USE_ROCM=OFF, + + TorchVision: 0.14.1 + OpenCV: 4.7.0 + MMEngine: 0.7.3 + +Runtime environment: + cudnn_benchmark: True + mp_cfg: {'mp_start_method': 'fork', 'opencv_num_threads': 0} + dist_cfg: {'backend': 'nccl'} + seed: None + deterministic: False + Distributed launcher: slurm + Distributed training: True + GPU number: 4 +------------------------------------------------------------ + +2023/06/04 02:32:51 - mmengine - INFO - Config: +optim_wrapper = dict( + optimizer=dict( + type='AdamW', lr=0.0001, weight_decay=0.3, _scope_='mmpretrain'), + paramwise_cfg=dict( + custom_keys=dict({ + '.cls_token': dict(decay_mult=0.0), + '.pos_embed': dict(decay_mult=0.0) + })), + type='AmpOptimWrapper', + dtype='bfloat16', + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=False, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=4096) +model = dict( + type='ImageClassifier', + backbone=dict( + frozen_stages=24, + type='VisionTransformer', + arch='l', + img_size=224, + patch_size=14, + drop_rate=0.1, + pre_norm=True, + final_norm=False, + init_cfg=dict( + type='Pretrained', + checkpoint='ckpt/openclip-ViT-L-14.pth', + prefix='backbone')), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=1024, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + init_cfg=None), + init_cfg=dict( + type='TruncNormal', layer=['Conv2d', 'Linear'], std=0.02, bias=0.0), + train_cfg=None) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=128, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stablediffusionV2-1-dpmsolver-25-1m.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/cc1m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=128, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV2-1-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=128, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV2-1-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm' + +2023/06/04 02:33:04 - mmengine - INFO - Hooks will be executed in the following order: +before_run: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_load_checkpoint: +(ABOVE_NORMAL) EMAHook + -------------------- +before_train: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_train_epoch: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(NORMAL ) DistSamplerSeedHook + -------------------- +before_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook + -------------------- +after_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +after_train_epoch: +(NORMAL ) IterTimerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_val_epoch: +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook + -------------------- +before_val_iter: +(NORMAL ) IterTimerHook + -------------------- +after_val_iter: +(NORMAL ) IterTimerHook +(NORMAL ) VisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_val_epoch: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_save_checkpoint: +(ABOVE_NORMAL) EMAHook + -------------------- +after_train: +(VERY_LOW ) CheckpointHook + -------------------- +before_test_epoch: +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook + -------------------- +before_test_iter: +(NORMAL ) IterTimerHook + -------------------- +after_test_iter: +(NORMAL ) IterTimerHook +(NORMAL ) VisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_test_epoch: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_run: +(BELOW_NORMAL) LoggerHook + -------------------- +2023/06/04 02:33:22 - mmengine - INFO - load backbone in model from: ckpt/openclip-ViT-L-14.pth +2023/06/04 02:33:24 - mmengine - WARNING - The model and loaded state dict do not match exactly + +unexpected key in source state_dict: ln1.weight, ln1.bias + +Name of parameter - Initialization information + +backbone.cls_token - torch.Size([1, 1, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.pos_embed - torch.Size([1, 257, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.patch_embed.projection.weight - torch.Size([1024, 3, 14, 14]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.0.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.0.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.0.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.0.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.0.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.0.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.0.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.0.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.0.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.0.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.0.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.0.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.1.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.1.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.1.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.1.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.1.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.1.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.1.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.1.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.1.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.1.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.1.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.1.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.2.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.2.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.2.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.2.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.2.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.2.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.2.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.2.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.2.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.2.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.2.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.2.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.3.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.3.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.3.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.3.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.3.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.3.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.3.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.3.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.3.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.3.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.3.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.3.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.4.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.4.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.4.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.4.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.4.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.4.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.4.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.4.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.4.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.4.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.4.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.4.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.5.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.5.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.5.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.5.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.5.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.5.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.5.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.5.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.5.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.5.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.5.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.5.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.6.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.6.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.6.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.6.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.6.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.6.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.6.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.6.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.6.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.6.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.6.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.6.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.7.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.7.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.7.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.7.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.7.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.7.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.7.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.7.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.7.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.7.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.7.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.7.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.8.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.8.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.8.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.8.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.8.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.8.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.8.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.8.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.8.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.8.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.8.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.8.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.9.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.9.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.9.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.9.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.9.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.9.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.9.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.9.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.9.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.9.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.9.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.9.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.10.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.10.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.10.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.10.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.10.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.10.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.10.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.10.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.10.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.10.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.10.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.10.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.11.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.11.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.11.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.11.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.11.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.11.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.11.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.11.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.11.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.11.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.11.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.11.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.12.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.12.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.12.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.12.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.12.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.12.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.12.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.12.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.12.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.12.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.12.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.12.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.13.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.13.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.13.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.13.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.13.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.13.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.13.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.13.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.13.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.13.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.13.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.13.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.14.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.14.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.14.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.14.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.14.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.14.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.14.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.14.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.14.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.14.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.14.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.14.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.15.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.15.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.15.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.15.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.15.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.15.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.15.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.15.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.15.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.15.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.15.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.15.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.16.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.16.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.16.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.16.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.16.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.16.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.16.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.16.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.16.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.16.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.16.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.16.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.17.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.17.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.17.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.17.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.17.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.17.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.17.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.17.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.17.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.17.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.17.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.17.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.18.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.18.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.18.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.18.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.18.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.18.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.18.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.18.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.18.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.18.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.18.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.18.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.19.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.19.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.19.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.19.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.19.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.19.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.19.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.19.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.19.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.19.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.19.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.19.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.20.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.20.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.20.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.20.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.20.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.20.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.20.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.20.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.20.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.20.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.20.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.20.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.21.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.21.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.21.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.21.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.21.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.21.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.21.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.21.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.21.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.21.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.21.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.21.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.22.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.22.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.22.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.22.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.22.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.22.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.22.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.22.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.22.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.22.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.22.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.22.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.23.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.23.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.23.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.23.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.23.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.23.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.23.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.23.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.23.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.23.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.23.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.23.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.pre_norm.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.pre_norm.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +head.fc.weight - torch.Size([2, 1024]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +head.fc.bias - torch.Size([2]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 +2023/06/04 02:33:24 - mmengine - WARNING - "FileClient" will be deprecated in future. Please use io functions in https://mmengine.readthedocs.io/en/latest/api/fileio.html#file-io +2023/06/04 02:33:24 - mmengine - WARNING - "HardDiskBackend" is the alias of "LocalBackend" and the former will be deprecated in future. +2023/06/04 02:33:24 - mmengine - INFO - Checkpoints will be saved to /mnt/petrelfs/luzeyu/workspace/fakebench/mmpretrain/workdir/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm. +2023/06/04 02:34:33 - mmengine - INFO - Epoch(train) [1][ 100/3907] lr: 9.9999e-05 eta: 7:25:46 time: 0.6276 data_time: 0.0014 memory: 44121 loss: 0.6555 +2023/06/04 02:35:36 - mmengine - INFO - Epoch(train) [1][ 200/3907] lr: 9.9994e-05 eta: 7:06:09 time: 0.6292 data_time: 0.0017 memory: 44121 loss: 0.6053 +2023/06/04 02:36:39 - mmengine - INFO - Epoch(train) [1][ 300/3907] lr: 9.9987e-05 eta: 6:59:03 time: 0.6299 data_time: 0.0015 memory: 44121 loss: 0.6015 +2023/06/04 02:37:42 - mmengine - INFO - Epoch(train) [1][ 400/3907] lr: 9.9977e-05 eta: 6:55:06 time: 0.6311 data_time: 0.0016 memory: 44121 loss: 0.5677 +2023/06/04 02:38:45 - mmengine - INFO - Epoch(train) [1][ 500/3907] lr: 9.9964e-05 eta: 6:52:17 time: 0.6303 data_time: 0.0016 memory: 44121 loss: 0.5292 +2023/06/04 02:39:48 - mmengine - INFO - Epoch(train) [1][ 600/3907] lr: 9.9948e-05 eta: 6:50:05 time: 0.6303 data_time: 0.0014 memory: 44121 loss: 0.4977 +2023/06/04 02:40:51 - mmengine - INFO - Epoch(train) [1][ 700/3907] lr: 9.9929e-05 eta: 6:48:09 time: 0.6306 data_time: 0.0014 memory: 44121 loss: 0.4941 +2023/06/04 02:41:54 - mmengine - INFO - Epoch(train) [1][ 800/3907] lr: 9.9907e-05 eta: 6:46:29 time: 0.6303 data_time: 0.0016 memory: 44121 loss: 0.4833 +2023/06/04 02:42:57 - mmengine - INFO - Epoch(train) [1][ 900/3907] lr: 9.9882e-05 eta: 6:44:56 time: 0.6298 data_time: 0.0015 memory: 44121 loss: 0.5067 +2023/06/04 02:44:00 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230604_023242 +2023/06/04 02:44:00 - mmengine - INFO - Epoch(train) [1][1000/3907] lr: 9.9855e-05 eta: 6:43:33 time: 0.6302 data_time: 0.0015 memory: 44121 loss: 0.4873 +2023/06/04 02:45:03 - mmengine - INFO - Epoch(train) [1][1100/3907] lr: 9.9824e-05 eta: 6:42:12 time: 0.6299 data_time: 0.0015 memory: 44121 loss: 0.4707 +2023/06/04 02:46:06 - mmengine - INFO - Epoch(train) [1][1200/3907] lr: 9.9791e-05 eta: 6:40:54 time: 0.6303 data_time: 0.0014 memory: 44121 loss: 0.4766 +2023/06/04 02:47:10 - mmengine - INFO - Epoch(train) [1][1300/3907] lr: 9.9755e-05 eta: 6:39:38 time: 0.6306 data_time: 0.0014 memory: 44121 loss: 0.4512 +2023/06/04 02:48:13 - mmengine - INFO - Epoch(train) [1][1400/3907] lr: 9.9716e-05 eta: 6:38:23 time: 0.6304 data_time: 0.0014 memory: 44121 loss: 0.4485 +2023/06/04 02:49:16 - mmengine - INFO - Epoch(train) [1][1500/3907] lr: 9.9674e-05 eta: 6:37:11 time: 0.6302 data_time: 0.0013 memory: 44121 loss: 0.4740 +2023/06/04 02:50:19 - mmengine - INFO - Epoch(train) [1][1600/3907] lr: 9.9629e-05 eta: 6:35:58 time: 0.6307 data_time: 0.0013 memory: 44121 loss: 0.4374 +2023/06/04 02:51:22 - mmengine - INFO - Epoch(train) [1][1700/3907] lr: 9.9581e-05 eta: 6:34:46 time: 0.6300 data_time: 0.0014 memory: 44121 loss: 0.4458 +2023/06/04 02:52:25 - mmengine - INFO - Epoch(train) [1][1800/3907] lr: 9.9530e-05 eta: 6:33:35 time: 0.6298 data_time: 0.0014 memory: 44121 loss: 0.4599 +2023/06/04 02:53:28 - mmengine - INFO - Epoch(train) [1][1900/3907] lr: 9.9476e-05 eta: 6:32:27 time: 0.6305 data_time: 0.0015 memory: 44121 loss: 0.4460 +2023/06/04 02:54:31 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230604_023242 +2023/06/04 02:54:31 - mmengine - INFO - Epoch(train) [1][2000/3907] lr: 9.9420e-05 eta: 6:31:17 time: 0.6298 data_time: 0.0017 memory: 44121 loss: 0.4222 +2023/06/04 02:55:34 - mmengine - INFO - Epoch(train) [1][2100/3907] lr: 9.9361e-05 eta: 6:30:08 time: 0.6295 data_time: 0.0017 memory: 44121 loss: 0.4509 +2023/06/04 02:56:37 - mmengine - INFO - Epoch(train) [1][2200/3907] lr: 9.9298e-05 eta: 6:29:00 time: 0.6305 data_time: 0.0015 memory: 44121 loss: 0.4206 +2023/06/04 02:57:40 - mmengine - INFO - Epoch(train) [1][2300/3907] lr: 9.9233e-05 eta: 6:27:53 time: 0.6304 data_time: 0.0015 memory: 44121 loss: 0.4282 +2023/06/04 02:58:43 - mmengine - INFO - Epoch(train) [1][2400/3907] lr: 9.9165e-05 eta: 6:26:45 time: 0.6298 data_time: 0.0014 memory: 44121 loss: 0.4224 +2023/06/04 02:59:46 - mmengine - INFO - Epoch(train) [1][2500/3907] lr: 9.9095e-05 eta: 6:25:38 time: 0.6305 data_time: 0.0014 memory: 44121 loss: 0.4317 +2023/06/04 03:00:49 - mmengine - INFO - Epoch(train) [1][2600/3907] lr: 9.9021e-05 eta: 6:24:32 time: 0.6305 data_time: 0.0013 memory: 44121 loss: 0.4201 +2023/06/04 03:01:52 - mmengine - INFO - Epoch(train) [1][2700/3907] lr: 9.8944e-05 eta: 6:23:25 time: 0.6311 data_time: 0.0015 memory: 44121 loss: 0.3976 +2023/06/04 03:02:55 - mmengine - INFO - Epoch(train) [1][2800/3907] lr: 9.8865e-05 eta: 6:22:19 time: 0.6300 data_time: 0.0014 memory: 44121 loss: 0.3904 +2023/06/04 03:03:58 - mmengine - INFO - Epoch(train) [1][2900/3907] lr: 9.8783e-05 eta: 6:21:14 time: 0.6308 data_time: 0.0015 memory: 44121 loss: 0.4231 +2023/06/04 03:05:01 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230604_023242 +2023/06/04 03:05:01 - mmengine - INFO - Epoch(train) [1][3000/3907] lr: 9.8698e-05 eta: 6:20:09 time: 0.6301 data_time: 0.0015 memory: 44121 loss: 0.4072 +2023/06/04 03:06:04 - mmengine - INFO - Epoch(train) [1][3100/3907] lr: 9.8610e-05 eta: 6:19:04 time: 0.6302 data_time: 0.0015 memory: 44121 loss: 0.4282 +2023/06/04 03:07:08 - mmengine - INFO - Epoch(train) [1][3200/3907] lr: 9.8519e-05 eta: 6:17:59 time: 0.6297 data_time: 0.0015 memory: 44121 loss: 0.4192 +2023/06/04 03:08:11 - mmengine - INFO - Epoch(train) [1][3300/3907] lr: 9.8426e-05 eta: 6:16:54 time: 0.6304 data_time: 0.0014 memory: 44121 loss: 0.4098 +2023/06/04 03:09:14 - mmengine - INFO - Epoch(train) [1][3400/3907] lr: 9.8330e-05 eta: 6:15:49 time: 0.6304 data_time: 0.0014 memory: 44121 loss: 0.3929 +2023/06/04 03:10:17 - mmengine - INFO - Epoch(train) [1][3500/3907] lr: 9.8231e-05 eta: 6:14:44 time: 0.6301 data_time: 0.0016 memory: 44121 loss: 0.4143 +2023/06/04 03:11:20 - mmengine - INFO - Epoch(train) [1][3600/3907] lr: 9.8129e-05 eta: 6:13:40 time: 0.6292 data_time: 0.0018 memory: 44121 loss: 0.4080 +2023/06/04 03:12:23 - mmengine - INFO - Epoch(train) [1][3700/3907] lr: 9.8024e-05 eta: 6:12:35 time: 0.6298 data_time: 0.0015 memory: 44121 loss: 0.3692 +2023/06/04 03:13:26 - mmengine - INFO - Epoch(train) [1][3800/3907] lr: 9.7917e-05 eta: 6:11:30 time: 0.6307 data_time: 0.0014 memory: 44121 loss: 0.3998 +2023/06/04 03:14:29 - mmengine - INFO - Epoch(train) [1][3900/3907] lr: 9.7806e-05 eta: 6:10:26 time: 0.6292 data_time: 0.0012 memory: 44121 loss: 0.3907 +2023/06/04 03:14:33 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230604_023242 +2023/06/04 03:14:33 - mmengine - INFO - Saving checkpoint at 1 epochs +2023/06/04 03:16:09 - mmengine - INFO - Epoch(val) [1][57/57] accuracy/top1: 73.4630 single-label/precision_classwise: [70.59265899658203, 79.97044372558594] single-label/recall_classwise: [88.87696075439453, 54.534881591796875] single-label/f1-score_classwise: [78.68659973144531, 64.84767150878906] data_time: 0.0467 time: 1.3394 +2023/06/04 03:17:12 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230604_023242 +2023/06/04 03:17:16 - mmengine - INFO - Epoch(train) [2][ 100/3907] lr: 9.7685e-05 eta: 6:09:45 time: 0.6315 data_time: 0.0017 memory: 44121 loss: 0.4055 +2023/06/04 03:18:19 - mmengine - INFO - Epoch(train) [2][ 200/3907] lr: 9.7570e-05 eta: 6:08:40 time: 0.6305 data_time: 0.0013 memory: 44120 loss: 0.3878 +2023/06/04 03:19:22 - mmengine - INFO - Epoch(train) [2][ 300/3907] lr: 9.7451e-05 eta: 6:07:35 time: 0.6303 data_time: 0.0016 memory: 44120 loss: 0.3940 +2023/06/04 03:20:25 - mmengine - INFO - Epoch(train) [2][ 400/3907] lr: 9.7329e-05 eta: 6:06:31 time: 0.6314 data_time: 0.0022 memory: 44120 loss: 0.3812 +2023/06/04 03:21:29 - mmengine - INFO - Epoch(train) [2][ 500/3907] lr: 9.7205e-05 eta: 6:05:28 time: 0.6352 data_time: 0.0021 memory: 44120 loss: 0.4011 +2023/06/04 03:22:32 - mmengine - INFO - Epoch(train) [2][ 600/3907] lr: 9.7078e-05 eta: 6:04:25 time: 0.6308 data_time: 0.0016 memory: 44120 loss: 0.3945 +2023/06/04 03:23:35 - mmengine - INFO - Epoch(train) [2][ 700/3907] lr: 9.6949e-05 eta: 6:03:21 time: 0.6310 data_time: 0.0017 memory: 44120 loss: 0.3901 +2023/06/04 03:24:38 - mmengine - INFO - Epoch(train) [2][ 800/3907] lr: 9.6816e-05 eta: 6:02:16 time: 0.6311 data_time: 0.0014 memory: 44120 loss: 0.3786 +2023/06/04 03:25:41 - mmengine - INFO - Epoch(train) [2][ 900/3907] lr: 9.6681e-05 eta: 6:01:12 time: 0.6309 data_time: 0.0016 memory: 44120 loss: 0.3920 +2023/06/04 03:26:45 - mmengine - INFO - Epoch(train) [2][1000/3907] lr: 9.6544e-05 eta: 6:00:09 time: 0.6317 data_time: 0.0014 memory: 44120 loss: 0.3838 +2023/06/04 03:27:43 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230604_023242 +2023/06/04 03:27:48 - mmengine - INFO - Epoch(train) [2][1100/3907] lr: 9.6403e-05 eta: 5:59:05 time: 0.6314 data_time: 0.0016 memory: 44120 loss: 0.3801 +2023/06/04 03:28:51 - mmengine - INFO - Epoch(train) [2][1200/3907] lr: 9.6260e-05 eta: 5:58:01 time: 0.6308 data_time: 0.0014 memory: 44120 loss: 0.3866 +2023/06/04 03:29:54 - mmengine - INFO - Epoch(train) [2][1300/3907] lr: 9.6114e-05 eta: 5:56:57 time: 0.6313 data_time: 0.0014 memory: 44120 loss: 0.3812 +2023/06/04 03:30:57 - mmengine - INFO - Epoch(train) [2][1400/3907] lr: 9.5966e-05 eta: 5:55:54 time: 0.6317 data_time: 0.0015 memory: 44120 loss: 0.3780 +2023/06/04 03:32:01 - mmengine - INFO - Epoch(train) [2][1500/3907] lr: 9.5815e-05 eta: 5:54:51 time: 0.6309 data_time: 0.0014 memory: 44120 loss: 0.3787 +2023/06/04 03:33:04 - mmengine - INFO - Epoch(train) [2][1600/3907] lr: 9.5661e-05 eta: 5:53:47 time: 0.6321 data_time: 0.0014 memory: 44120 loss: 0.3915 +2023/06/04 03:34:07 - mmengine - INFO - Epoch(train) [2][1700/3907] lr: 9.5505e-05 eta: 5:52:43 time: 0.6317 data_time: 0.0015 memory: 44120 loss: 0.3773 +2023/06/04 03:35:10 - mmengine - INFO - Epoch(train) [2][1800/3907] lr: 9.5346e-05 eta: 5:51:39 time: 0.6306 data_time: 0.0015 memory: 44120 loss: 0.3765 +2023/06/04 03:36:13 - mmengine - INFO - Epoch(train) [2][1900/3907] lr: 9.5184e-05 eta: 5:50:36 time: 0.6459 data_time: 0.0014 memory: 44120 loss: 0.3772 +2023/06/04 03:37:16 - mmengine - INFO - Epoch(train) [2][2000/3907] lr: 9.5020e-05 eta: 5:49:32 time: 0.6313 data_time: 0.0016 memory: 44120 loss: 0.3971 +2023/06/04 03:38:15 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230604_023242 +2023/06/04 03:38:20 - mmengine - INFO - Epoch(train) [2][2100/3907] lr: 9.4854e-05 eta: 5:48:29 time: 0.6317 data_time: 0.0018 memory: 44120 loss: 0.3894 +2023/06/04 03:39:23 - mmengine - INFO - Epoch(train) [2][2200/3907] lr: 9.4684e-05 eta: 5:47:25 time: 0.6334 data_time: 0.0014 memory: 44120 loss: 0.3702 +2023/06/04 03:40:26 - mmengine - INFO - Epoch(train) [2][2300/3907] lr: 9.4512e-05 eta: 5:46:21 time: 0.6308 data_time: 0.0014 memory: 44120 loss: 0.3745 +2023/06/04 03:41:29 - mmengine - INFO - Epoch(train) [2][2400/3907] lr: 9.4338e-05 eta: 5:45:18 time: 0.6349 data_time: 0.0015 memory: 44120 loss: 0.3446 +2023/06/04 03:42:32 - mmengine - INFO - Epoch(train) [2][2500/3907] lr: 9.4161e-05 eta: 5:44:15 time: 0.6313 data_time: 0.0015 memory: 44120 loss: 0.3948 +2023/06/04 03:43:36 - mmengine - INFO - Epoch(train) [2][2600/3907] lr: 9.3981e-05 eta: 5:43:12 time: 0.6306 data_time: 0.0014 memory: 44120 loss: 0.3777 +2023/06/04 03:44:39 - mmengine - INFO - Epoch(train) [2][2700/3907] lr: 9.3799e-05 eta: 5:42:08 time: 0.6309 data_time: 0.0015 memory: 44120 loss: 0.3788 +2023/06/04 03:45:42 - mmengine - INFO - Epoch(train) [2][2800/3907] lr: 9.3615e-05 eta: 5:41:05 time: 0.6303 data_time: 0.0014 memory: 44120 loss: 0.3672 +2023/06/04 03:46:45 - mmengine - INFO - Epoch(train) [2][2900/3907] lr: 9.3428e-05 eta: 5:40:01 time: 0.6315 data_time: 0.0014 memory: 44120 loss: 0.3875 +2023/06/04 03:47:48 - mmengine - INFO - Epoch(train) [2][3000/3907] lr: 9.3238e-05 eta: 5:38:58 time: 0.6315 data_time: 0.0014 memory: 44120 loss: 0.3591 +2023/06/04 03:48:47 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230604_023242 +2023/06/04 03:48:52 - mmengine - INFO - Epoch(train) [2][3100/3907] lr: 9.3046e-05 eta: 5:37:54 time: 0.6331 data_time: 0.0016 memory: 44120 loss: 0.3561 +2023/06/04 03:49:55 - mmengine - INFO - Epoch(train) [2][3200/3907] lr: 9.2852e-05 eta: 5:36:51 time: 0.6328 data_time: 0.0017 memory: 44120 loss: 0.3726 +2023/06/04 03:50:58 - mmengine - INFO - Epoch(train) [2][3300/3907] lr: 9.2655e-05 eta: 5:35:47 time: 0.6309 data_time: 0.0016 memory: 44120 loss: 0.3504 +2023/06/04 03:52:01 - mmengine - INFO - Epoch(train) [2][3400/3907] lr: 9.2456e-05 eta: 5:34:44 time: 0.6304 data_time: 0.0016 memory: 44120 loss: 0.3427 +2023/06/04 03:53:04 - mmengine - INFO - Epoch(train) [2][3500/3907] lr: 9.2254e-05 eta: 5:33:40 time: 0.6312 data_time: 0.0014 memory: 44120 loss: 0.3607 +2023/06/04 03:54:07 - mmengine - INFO - Epoch(train) [2][3600/3907] lr: 9.2050e-05 eta: 5:32:36 time: 0.6312 data_time: 0.0015 memory: 44120 loss: 0.3670 +2023/06/04 03:55:11 - mmengine - INFO - Epoch(train) [2][3700/3907] lr: 9.1843e-05 eta: 5:31:33 time: 0.6312 data_time: 0.0014 memory: 44120 loss: 0.3641 +2023/06/04 03:56:14 - mmengine - INFO - Epoch(train) [2][3800/3907] lr: 9.1634e-05 eta: 5:30:29 time: 0.6316 data_time: 0.0014 memory: 44120 loss: 0.3695 +2023/06/04 03:57:17 - mmengine - INFO - Epoch(train) [2][3900/3907] lr: 9.1423e-05 eta: 5:29:26 time: 0.6313 data_time: 0.0012 memory: 44120 loss: 0.3688 +2023/06/04 03:57:21 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230604_023242 +2023/06/04 03:57:21 - mmengine - INFO - Saving checkpoint at 2 epochs +2023/06/04 03:58:55 - mmengine - INFO - Epoch(val) [2][57/57] accuracy/top1: 77.8435 single-label/precision_classwise: [72.1218032836914, 94.5566177368164] single-label/recall_classwise: [97.4812240600586, 53.72868347167969] single-label/f1-score_classwise: [82.9056167602539, 68.52198791503906] data_time: 0.0363 time: 1.2788 +2023/06/04 04:00:02 - mmengine - INFO - Epoch(train) [3][ 100/3907] lr: 9.1194e-05 eta: 5:28:29 time: 0.6318 data_time: 0.0014 memory: 44120 loss: 0.3848 +2023/06/04 04:00:56 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230604_023242 +2023/06/04 04:01:05 - mmengine - INFO - Epoch(train) [3][ 200/3907] lr: 9.0978e-05 eta: 5:27:25 time: 0.6321 data_time: 0.0015 memory: 44120 loss: 0.4015 +2023/06/04 04:02:08 - mmengine - INFO - Epoch(train) [3][ 300/3907] lr: 9.0759e-05 eta: 5:26:22 time: 0.6316 data_time: 0.0015 memory: 44120 loss: 0.3898 +2023/06/04 04:03:12 - mmengine - INFO - Epoch(train) [3][ 400/3907] lr: 9.0539e-05 eta: 5:25:19 time: 0.6336 data_time: 0.0016 memory: 44120 loss: 0.3464 +2023/06/04 04:04:15 - mmengine - INFO - Epoch(train) [3][ 500/3907] lr: 9.0315e-05 eta: 5:24:16 time: 0.6326 data_time: 0.0024 memory: 44120 loss: 0.3698 +2023/06/04 04:05:18 - mmengine - INFO - Epoch(train) [3][ 600/3907] lr: 9.0090e-05 eta: 5:23:14 time: 0.6313 data_time: 0.0015 memory: 44120 loss: 0.3709 +2023/06/04 04:06:22 - mmengine - INFO - Epoch(train) [3][ 700/3907] lr: 8.9862e-05 eta: 5:22:10 time: 0.6314 data_time: 0.0015 memory: 44120 loss: 0.3923 +2023/06/04 04:07:25 - mmengine - INFO - Epoch(train) [3][ 800/3907] lr: 8.9632e-05 eta: 5:21:07 time: 0.6321 data_time: 0.0016 memory: 44120 loss: 0.3743 +2023/06/04 04:08:28 - mmengine - INFO - Epoch(train) [3][ 900/3907] lr: 8.9400e-05 eta: 5:20:05 time: 0.6326 data_time: 0.0015 memory: 44120 loss: 0.3778 +2023/06/04 04:09:32 - mmengine - INFO - Epoch(train) [3][1000/3907] lr: 8.9166e-05 eta: 5:19:01 time: 0.6318 data_time: 0.0014 memory: 44120 loss: 0.3695 +2023/06/04 04:10:35 - mmengine - INFO - Epoch(train) [3][1100/3907] lr: 8.8929e-05 eta: 5:17:58 time: 0.6328 data_time: 0.0015 memory: 44120 loss: 0.3676 +2023/06/04 04:11:29 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230604_023242 +2023/06/04 04:11:38 - mmengine - INFO - Epoch(train) [3][1200/3907] lr: 8.8691e-05 eta: 5:16:55 time: 0.6352 data_time: 0.0014 memory: 44120 loss: 0.3876 +2023/06/04 04:12:42 - mmengine - INFO - Epoch(train) [3][1300/3907] lr: 8.8450e-05 eta: 5:15:52 time: 0.6323 data_time: 0.0015 memory: 44120 loss: 0.3605 +2023/06/04 04:13:45 - mmengine - INFO - Epoch(train) [3][1400/3907] lr: 8.8206e-05 eta: 5:14:49 time: 0.6402 data_time: 0.0016 memory: 44120 loss: 0.3751 +2023/06/04 04:14:48 - mmengine - INFO - Epoch(train) [3][1500/3907] lr: 8.7961e-05 eta: 5:13:46 time: 0.6333 data_time: 0.0016 memory: 44120 loss: 0.3715 +2023/06/04 04:15:52 - mmengine - INFO - Epoch(train) [3][1600/3907] lr: 8.7714e-05 eta: 5:12:43 time: 0.6341 data_time: 0.0015 memory: 44120 loss: 0.3671 +2023/06/04 04:16:55 - mmengine - INFO - Epoch(train) [3][1700/3907] lr: 8.7464e-05 eta: 5:11:40 time: 0.6319 data_time: 0.0016 memory: 44120 loss: 0.3712 +2023/06/04 04:17:58 - mmengine - INFO - Epoch(train) [3][1800/3907] lr: 8.7213e-05 eta: 5:10:37 time: 0.6322 data_time: 0.0015 memory: 44120 loss: 0.3908 +2023/06/04 04:19:02 - mmengine - INFO - Epoch(train) [3][1900/3907] lr: 8.6959e-05 eta: 5:09:33 time: 0.6326 data_time: 0.0016 memory: 44120 loss: 0.3626 +2023/06/04 04:20:05 - mmengine - INFO - Epoch(train) [3][2000/3907] lr: 8.6703e-05 eta: 5:08:30 time: 0.6321 data_time: 0.0017 memory: 44120 loss: 0.3669 +2023/06/04 04:21:08 - mmengine - INFO - Epoch(train) [3][2100/3907] lr: 8.6445e-05 eta: 5:07:26 time: 0.6325 data_time: 0.0014 memory: 44120 loss: 0.3438 +2023/06/04 04:22:02 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230604_023242 +2023/06/04 04:22:11 - mmengine - INFO - Epoch(train) [3][2200/3907] lr: 8.6186e-05 eta: 5:06:23 time: 0.6322 data_time: 0.0014 memory: 44120 loss: 0.3303 +2023/06/04 04:23:14 - mmengine - INFO - Epoch(train) [3][2300/3907] lr: 8.5924e-05 eta: 5:05:20 time: 0.6324 data_time: 0.0015 memory: 44120 loss: 0.3918 +2023/06/04 04:24:18 - mmengine - INFO - Epoch(train) [3][2400/3907] lr: 8.5660e-05 eta: 5:04:17 time: 0.6344 data_time: 0.0014 memory: 44120 loss: 0.3499 +2023/06/04 04:25:21 - mmengine - INFO - Epoch(train) [3][2500/3907] lr: 8.5394e-05 eta: 5:03:13 time: 0.6314 data_time: 0.0014 memory: 44120 loss: 0.3635 +2023/06/04 04:26:24 - mmengine - INFO - Epoch(train) [3][2600/3907] lr: 8.5126e-05 eta: 5:02:10 time: 0.6319 data_time: 0.0014 memory: 44120 loss: 0.3758 +2023/06/04 04:27:28 - mmengine - INFO - Epoch(train) [3][2700/3907] lr: 8.4856e-05 eta: 5:01:07 time: 0.6318 data_time: 0.0014 memory: 44120 loss: 0.3556 +2023/06/04 04:28:31 - mmengine - INFO - Epoch(train) [3][2800/3907] lr: 8.4585e-05 eta: 5:00:04 time: 0.6317 data_time: 0.0014 memory: 44120 loss: 0.3612 +2023/06/04 04:29:34 - mmengine - INFO - Epoch(train) [3][2900/3907] lr: 8.4311e-05 eta: 4:59:00 time: 0.6306 data_time: 0.0014 memory: 44120 loss: 0.3662 +2023/06/04 04:30:37 - mmengine - INFO - Epoch(train) [3][3000/3907] lr: 8.4036e-05 eta: 4:57:57 time: 0.6315 data_time: 0.0015 memory: 44120 loss: 0.3824 +2023/06/04 04:31:41 - mmengine - INFO - Epoch(train) [3][3100/3907] lr: 8.3758e-05 eta: 4:56:54 time: 0.6317 data_time: 0.0014 memory: 44120 loss: 0.3742 +2023/06/04 04:32:35 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230604_023242 +2023/06/04 04:32:44 - mmengine - INFO - Epoch(train) [3][3200/3907] lr: 8.3479e-05 eta: 4:55:50 time: 0.6310 data_time: 0.0014 memory: 44120 loss: 0.3399 +2023/06/04 04:33:47 - mmengine - INFO - Epoch(train) [3][3300/3907] lr: 8.3198e-05 eta: 4:54:47 time: 0.6334 data_time: 0.0015 memory: 44120 loss: 0.3877 +2023/06/04 04:34:50 - mmengine - INFO - Epoch(train) [3][3400/3907] lr: 8.2915e-05 eta: 4:53:44 time: 0.6364 data_time: 0.0013 memory: 44120 loss: 0.3425 +2023/06/04 04:35:54 - mmengine - INFO - Epoch(train) [3][3500/3907] lr: 8.2630e-05 eta: 4:52:41 time: 0.6332 data_time: 0.0015 memory: 44120 loss: 0.3786 +2023/06/04 04:36:57 - mmengine - INFO - Epoch(train) [3][3600/3907] lr: 8.2344e-05 eta: 4:51:37 time: 0.6316 data_time: 0.0014 memory: 44120 loss: 0.3760 +2023/06/04 04:38:00 - mmengine - INFO - Epoch(train) [3][3700/3907] lr: 8.2056e-05 eta: 4:50:34 time: 0.6314 data_time: 0.0015 memory: 44120 loss: 0.3577 +2023/06/04 04:39:04 - mmengine - INFO - Epoch(train) [3][3800/3907] lr: 8.1765e-05 eta: 4:49:31 time: 0.6342 data_time: 0.0015 memory: 44120 loss: 0.3581 +2023/06/04 04:40:07 - mmengine - INFO - Epoch(train) [3][3900/3907] lr: 8.1474e-05 eta: 4:48:28 time: 0.6310 data_time: 0.0012 memory: 44120 loss: 0.3363 +2023/06/04 04:40:11 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230604_023242 +2023/06/04 04:40:11 - mmengine - INFO - Saving checkpoint at 3 epochs +2023/06/04 04:41:44 - mmengine - INFO - Epoch(val) [3][57/57] accuracy/top1: 77.0815 single-label/precision_classwise: [71.14523315429688, 96.01982879638672] single-label/recall_classwise: [98.27661895751953, 51.054264068603516] single-label/f1-score_classwise: [82.53850555419922, 66.66329193115234] data_time: 0.0357 time: 1.2774 +2023/06/04 04:42:51 - mmengine - INFO - Epoch(train) [4][ 100/3907] lr: 8.1160e-05 eta: 4:47:27 time: 0.6328 data_time: 0.0015 memory: 44120 loss: 0.3617 +2023/06/04 04:43:54 - mmengine - INFO - Epoch(train) [4][ 200/3907] lr: 8.0864e-05 eta: 4:46:24 time: 0.6362 data_time: 0.0013 memory: 44120 loss: 0.3635 +2023/06/04 04:44:44 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230604_023242 +2023/06/04 04:44:58 - mmengine - INFO - Epoch(train) [4][ 300/3907] lr: 8.0567e-05 eta: 4:45:21 time: 0.6349 data_time: 0.0014 memory: 44120 loss: 0.3629 +2023/06/04 04:46:01 - mmengine - INFO - Epoch(train) [4][ 400/3907] lr: 8.0269e-05 eta: 4:44:17 time: 0.6332 data_time: 0.0018 memory: 44120 loss: 0.3611 +2023/06/04 04:47:04 - mmengine - INFO - Epoch(train) [4][ 500/3907] lr: 7.9969e-05 eta: 4:43:14 time: 0.6319 data_time: 0.0014 memory: 44120 loss: 0.3571 +2023/06/04 04:48:08 - mmengine - INFO - Epoch(train) [4][ 600/3907] lr: 7.9667e-05 eta: 4:42:11 time: 0.6419 data_time: 0.0015 memory: 44120 loss: 0.3749 +2023/06/04 04:49:11 - mmengine - INFO - Epoch(train) [4][ 700/3907] lr: 7.9363e-05 eta: 4:41:08 time: 0.6329 data_time: 0.0015 memory: 44120 loss: 0.3667 +2023/06/04 04:50:14 - mmengine - INFO - Epoch(train) [4][ 800/3907] lr: 7.9058e-05 eta: 4:40:04 time: 0.6308 data_time: 0.0014 memory: 44120 loss: 0.3752 +2023/06/04 04:51:17 - mmengine - INFO - Epoch(train) [4][ 900/3907] lr: 7.8752e-05 eta: 4:39:01 time: 0.6374 data_time: 0.0015 memory: 44120 loss: 0.3581 +2023/06/04 04:52:21 - mmengine - INFO - Epoch(train) [4][1000/3907] lr: 7.8444e-05 eta: 4:37:58 time: 0.6331 data_time: 0.0013 memory: 44120 loss: 0.3650 +2023/06/04 04:53:24 - mmengine - INFO - Epoch(train) [4][1100/3907] lr: 7.8134e-05 eta: 4:36:55 time: 0.6355 data_time: 0.0015 memory: 44120 loss: 0.3515 +2023/06/04 04:54:28 - mmengine - INFO - Epoch(train) [4][1200/3907] lr: 7.7823e-05 eta: 4:35:52 time: 0.6352 data_time: 0.0015 memory: 44120 loss: 0.3675 +2023/06/04 04:55:18 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230604_023242 +2023/06/04 04:55:31 - mmengine - INFO - Epoch(train) [4][1300/3907] lr: 7.7510e-05 eta: 4:34:48 time: 0.6340 data_time: 0.0016 memory: 44120 loss: 0.3744 +2023/06/04 04:56:34 - mmengine - INFO - Epoch(train) [4][1400/3907] lr: 7.7196e-05 eta: 4:33:45 time: 0.6333 data_time: 0.0015 memory: 44120 loss: 0.3665 +2023/06/04 04:57:38 - mmengine - INFO - Epoch(train) [4][1500/3907] lr: 7.6881e-05 eta: 4:32:43 time: 0.6427 data_time: 0.0016 memory: 44120 loss: 0.3685 +2023/06/04 04:58:41 - mmengine - INFO - Epoch(train) [4][1600/3907] lr: 7.6564e-05 eta: 4:31:40 time: 0.6346 data_time: 0.0015 memory: 44120 loss: 0.3733 +2023/06/04 04:59:45 - mmengine - INFO - Epoch(train) [4][1700/3907] lr: 7.6246e-05 eta: 4:30:36 time: 0.6342 data_time: 0.0015 memory: 44120 loss: 0.3773 +2023/06/04 05:00:48 - mmengine - INFO - Epoch(train) [4][1800/3907] lr: 7.5926e-05 eta: 4:29:33 time: 0.6332 data_time: 0.0015 memory: 44120 loss: 0.3371 +2023/06/04 05:01:51 - mmengine - INFO - Epoch(train) [4][1900/3907] lr: 7.5605e-05 eta: 4:28:30 time: 0.6319 data_time: 0.0014 memory: 44120 loss: 0.3500 +2023/06/04 05:02:55 - mmengine - INFO - Epoch(train) [4][2000/3907] lr: 7.5283e-05 eta: 4:27:26 time: 0.6307 data_time: 0.0016 memory: 44120 loss: 0.3525 +2023/06/04 05:03:58 - mmengine - INFO - Epoch(train) [4][2100/3907] lr: 7.4959e-05 eta: 4:26:23 time: 0.6311 data_time: 0.0015 memory: 44120 loss: 0.3676 +2023/06/04 05:05:01 - mmengine - INFO - Epoch(train) [4][2200/3907] lr: 7.4634e-05 eta: 4:25:19 time: 0.6313 data_time: 0.0015 memory: 44120 loss: 0.3848 +2023/06/04 05:05:51 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230604_023242 +2023/06/04 05:06:04 - mmengine - INFO - Epoch(train) [4][2300/3907] lr: 7.4308e-05 eta: 4:24:16 time: 0.6312 data_time: 0.0016 memory: 44120 loss: 0.3622 +2023/06/04 05:07:07 - mmengine - INFO - Epoch(train) [4][2400/3907] lr: 7.3980e-05 eta: 4:23:12 time: 0.6327 data_time: 0.0016 memory: 44120 loss: 0.3721 +2023/06/04 05:08:11 - mmengine - INFO - Epoch(train) [4][2500/3907] lr: 7.3652e-05 eta: 4:22:09 time: 0.6313 data_time: 0.0017 memory: 44120 loss: 0.3773 +2023/06/04 05:09:14 - mmengine - INFO - Epoch(train) [4][2600/3907] lr: 7.3322e-05 eta: 4:21:05 time: 0.6328 data_time: 0.0014 memory: 44120 loss: 0.3810 +2023/06/04 05:10:17 - mmengine - INFO - Epoch(train) [4][2700/3907] lr: 7.2991e-05 eta: 4:20:02 time: 0.6313 data_time: 0.0015 memory: 44120 loss: 0.3756 +2023/06/04 05:11:20 - mmengine - INFO - Epoch(train) [4][2800/3907] lr: 7.2659e-05 eta: 4:18:59 time: 0.6320 data_time: 0.0014 memory: 44120 loss: 0.3916 +2023/06/04 05:12:24 - mmengine - INFO - Epoch(train) [4][2900/3907] lr: 7.2325e-05 eta: 4:17:55 time: 0.6314 data_time: 0.0016 memory: 44120 loss: 0.3767 +2023/06/04 05:13:27 - mmengine - INFO - Epoch(train) [4][3000/3907] lr: 7.1991e-05 eta: 4:16:52 time: 0.6332 data_time: 0.0014 memory: 44120 loss: 0.3549 +2023/06/04 05:14:30 - mmengine - INFO - Epoch(train) [4][3100/3907] lr: 7.1655e-05 eta: 4:15:48 time: 0.6344 data_time: 0.0014 memory: 44120 loss: 0.3555 +2023/06/04 05:15:33 - mmengine - INFO - Epoch(train) [4][3200/3907] lr: 7.1318e-05 eta: 4:14:45 time: 0.6323 data_time: 0.0017 memory: 44120 loss: 0.3396 +2023/06/04 05:16:23 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230604_023242 +2023/06/04 05:16:37 - mmengine - INFO - Epoch(train) [4][3300/3907] lr: 7.0981e-05 eta: 4:13:42 time: 0.6322 data_time: 0.0016 memory: 44120 loss: 0.3686 +2023/06/04 05:17:40 - mmengine - INFO - Epoch(train) [4][3400/3907] lr: 7.0642e-05 eta: 4:12:39 time: 0.6331 data_time: 0.0013 memory: 44120 loss: 0.3623 +2023/06/04 05:18:44 - mmengine - INFO - Epoch(train) [4][3500/3907] lr: 7.0302e-05 eta: 4:11:36 time: 0.6356 data_time: 0.0014 memory: 44120 loss: 0.3748 +2023/06/04 05:19:47 - mmengine - INFO - Epoch(train) [4][3600/3907] lr: 6.9961e-05 eta: 4:10:32 time: 0.6407 data_time: 0.0015 memory: 44120 loss: 0.3533 +2023/06/04 05:20:50 - mmengine - INFO - Epoch(train) [4][3700/3907] lr: 6.9620e-05 eta: 4:09:29 time: 0.6365 data_time: 0.0023 memory: 44120 loss: 0.3668 +2023/06/04 05:21:54 - mmengine - INFO - Epoch(train) [4][3800/3907] lr: 6.9277e-05 eta: 4:08:26 time: 0.6313 data_time: 0.0015 memory: 44120 loss: 0.3694 +2023/06/04 05:22:57 - mmengine - INFO - Epoch(train) [4][3900/3907] lr: 6.8933e-05 eta: 4:07:23 time: 0.6324 data_time: 0.0012 memory: 44120 loss: 0.3838 +2023/06/04 05:23:01 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230604_023242 +2023/06/04 05:23:01 - mmengine - INFO - Saving checkpoint at 4 epochs +2023/06/04 05:24:34 - mmengine - INFO - Epoch(val) [4][57/57] accuracy/top1: 76.9041 single-label/precision_classwise: [70.8448486328125, 96.96969604492188] single-label/recall_classwise: [98.7248306274414, 50.10852813720703] single-label/f1-score_classwise: [82.49288177490234, 66.07380676269531] data_time: 0.0361 time: 1.2779 +2023/06/04 05:25:42 - mmengine - INFO - Epoch(train) [5][ 100/3907] lr: 6.8565e-05 eta: 4:06:20 time: 0.6328 data_time: 0.0017 memory: 44120 loss: 0.3789 +2023/06/04 05:26:45 - mmengine - INFO - Epoch(train) [5][ 200/3907] lr: 6.8219e-05 eta: 4:05:17 time: 0.6367 data_time: 0.0018 memory: 44120 loss: 0.3688 +2023/06/04 05:27:48 - mmengine - INFO - Epoch(train) [5][ 300/3907] lr: 6.7873e-05 eta: 4:04:13 time: 0.6329 data_time: 0.0018 memory: 44120 loss: 0.3627 +2023/06/04 05:28:34 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230604_023242 +2023/06/04 05:28:52 - mmengine - INFO - Epoch(train) [5][ 400/3907] lr: 6.7526e-05 eta: 4:03:10 time: 0.6354 data_time: 0.0017 memory: 44120 loss: 0.3606 +2023/06/04 05:29:55 - mmengine - INFO - Epoch(train) [5][ 500/3907] lr: 6.7178e-05 eta: 4:02:07 time: 0.6326 data_time: 0.0021 memory: 44120 loss: 0.3683 +2023/06/04 05:30:59 - mmengine - INFO - Epoch(train) [5][ 600/3907] lr: 6.6829e-05 eta: 4:01:04 time: 0.6310 data_time: 0.0017 memory: 44120 loss: 0.3763 +2023/06/04 05:32:02 - mmengine - INFO - Epoch(train) [5][ 700/3907] lr: 6.6480e-05 eta: 4:00:00 time: 0.6318 data_time: 0.0015 memory: 44120 loss: 0.3541 +2023/06/04 05:33:05 - mmengine - INFO - Epoch(train) [5][ 800/3907] lr: 6.6129e-05 eta: 3:58:57 time: 0.6322 data_time: 0.0019 memory: 44120 loss: 0.3661 +2023/06/04 05:34:08 - mmengine - INFO - Epoch(train) [5][ 900/3907] lr: 6.5778e-05 eta: 3:57:53 time: 0.6330 data_time: 0.0014 memory: 44120 loss: 0.3593 +2023/06/04 05:35:12 - mmengine - INFO - Epoch(train) [5][1000/3907] lr: 6.5427e-05 eta: 3:56:50 time: 0.6329 data_time: 0.0015 memory: 44120 loss: 0.3517 +2023/06/04 05:36:15 - mmengine - INFO - Epoch(train) [5][1100/3907] lr: 6.5074e-05 eta: 3:55:47 time: 0.6334 data_time: 0.0019 memory: 44120 loss: 0.3535 +2023/06/04 05:37:19 - mmengine - INFO - Epoch(train) [5][1200/3907] lr: 6.4721e-05 eta: 3:54:44 time: 0.6311 data_time: 0.0014 memory: 44120 loss: 0.3701 +2023/06/04 05:38:22 - mmengine - INFO - Epoch(train) [5][1300/3907] lr: 6.4368e-05 eta: 3:53:40 time: 0.6313 data_time: 0.0016 memory: 44120 loss: 0.3631 +2023/06/04 05:39:07 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230604_023242 +2023/06/04 05:39:25 - mmengine - INFO - Epoch(train) [5][1400/3907] lr: 6.4014e-05 eta: 3:52:37 time: 0.6331 data_time: 0.0015 memory: 44120 loss: 0.3609 +2023/06/04 05:40:28 - mmengine - INFO - Epoch(train) [5][1500/3907] lr: 6.3659e-05 eta: 3:51:34 time: 0.6318 data_time: 0.0015 memory: 44120 loss: 0.3777 +2023/06/04 05:41:31 - mmengine - INFO - Epoch(train) [5][1600/3907] lr: 6.3303e-05 eta: 3:50:30 time: 0.6331 data_time: 0.0015 memory: 44120 loss: 0.3599 +2023/06/04 05:42:35 - mmengine - INFO - Epoch(train) [5][1700/3907] lr: 6.2948e-05 eta: 3:49:27 time: 0.6328 data_time: 0.0015 memory: 44120 loss: 0.3842 +2023/06/04 05:43:38 - mmengine - INFO - Epoch(train) [5][1800/3907] lr: 6.2591e-05 eta: 3:48:23 time: 0.6327 data_time: 0.0017 memory: 44120 loss: 0.3585 +2023/06/04 05:44:41 - mmengine - INFO - Epoch(train) [5][1900/3907] lr: 6.2234e-05 eta: 3:47:20 time: 0.6319 data_time: 0.0014 memory: 44120 loss: 0.3820 +2023/06/04 05:45:45 - mmengine - INFO - Epoch(train) [5][2000/3907] lr: 6.1877e-05 eta: 3:46:17 time: 0.6326 data_time: 0.0015 memory: 44120 loss: 0.3520 +2023/06/04 05:46:48 - mmengine - INFO - Epoch(train) [5][2100/3907] lr: 6.1519e-05 eta: 3:45:13 time: 0.6362 data_time: 0.0015 memory: 44120 loss: 0.3645 +2023/06/04 05:47:52 - mmengine - INFO - Epoch(train) [5][2200/3907] lr: 6.1161e-05 eta: 3:44:10 time: 0.6347 data_time: 0.0016 memory: 44120 loss: 0.3728 +2023/06/04 05:48:55 - mmengine - INFO - Epoch(train) [5][2300/3907] lr: 6.0802e-05 eta: 3:43:07 time: 0.6328 data_time: 0.0017 memory: 44120 loss: 0.3392 +2023/06/04 05:49:41 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230604_023242 +2023/06/04 05:49:59 - mmengine - INFO - Epoch(train) [5][2400/3907] lr: 6.0443e-05 eta: 3:42:04 time: 0.6335 data_time: 0.0016 memory: 44120 loss: 0.3595 +2023/06/04 05:51:02 - mmengine - INFO - Epoch(train) [5][2500/3907] lr: 6.0084e-05 eta: 3:41:01 time: 0.6342 data_time: 0.0016 memory: 44120 loss: 0.3794 +2023/06/04 05:52:05 - mmengine - INFO - Epoch(train) [5][2600/3907] lr: 5.9724e-05 eta: 3:39:57 time: 0.6360 data_time: 0.0016 memory: 44120 loss: 0.3658 +2023/06/04 05:53:09 - mmengine - INFO - Epoch(train) [5][2700/3907] lr: 5.9364e-05 eta: 3:38:54 time: 0.6328 data_time: 0.0015 memory: 44120 loss: 0.3590 +2023/06/04 05:54:12 - mmengine - INFO - Epoch(train) [5][2800/3907] lr: 5.9004e-05 eta: 3:37:51 time: 0.6356 data_time: 0.0016 memory: 44120 loss: 0.3516 +2023/06/04 05:55:16 - mmengine - INFO - Epoch(train) [5][2900/3907] lr: 5.8643e-05 eta: 3:36:48 time: 0.6349 data_time: 0.0014 memory: 44120 loss: 0.3543 +2023/06/04 05:56:19 - mmengine - INFO - Epoch(train) [5][3000/3907] lr: 5.8283e-05 eta: 3:35:45 time: 0.6330 data_time: 0.0017 memory: 44120 loss: 0.3838 +2023/06/04 05:57:22 - mmengine - INFO - Epoch(train) [5][3100/3907] lr: 5.7922e-05 eta: 3:34:41 time: 0.6326 data_time: 0.0016 memory: 44120 loss: 0.3755 +2023/06/04 05:58:26 - mmengine - INFO - Epoch(train) [5][3200/3907] lr: 5.7560e-05 eta: 3:33:38 time: 0.6323 data_time: 0.0015 memory: 44120 loss: 0.3400 +2023/06/04 05:59:29 - mmengine - INFO - Epoch(train) [5][3300/3907] lr: 5.7199e-05 eta: 3:32:34 time: 0.6326 data_time: 0.0016 memory: 44120 loss: 0.3651 +2023/06/04 06:00:15 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230604_023242 +2023/06/04 06:00:32 - mmengine - INFO - Epoch(train) [5][3400/3907] lr: 5.6838e-05 eta: 3:31:31 time: 0.6325 data_time: 0.0015 memory: 44120 loss: 0.3510 +2023/06/04 06:01:36 - mmengine - INFO - Epoch(train) [5][3500/3907] lr: 5.6476e-05 eta: 3:30:28 time: 0.6369 data_time: 0.0015 memory: 44120 loss: 0.3629 +2023/06/04 06:02:39 - mmengine - INFO - Epoch(train) [5][3600/3907] lr: 5.6114e-05 eta: 3:29:25 time: 0.6347 data_time: 0.0015 memory: 44120 loss: 0.3747 +2023/06/04 06:03:43 - mmengine - INFO - Epoch(train) [5][3700/3907] lr: 5.5753e-05 eta: 3:28:22 time: 0.6320 data_time: 0.0015 memory: 44120 loss: 0.3751 +2023/06/04 06:04:46 - mmengine - INFO - Epoch(train) [5][3800/3907] lr: 5.5391e-05 eta: 3:27:18 time: 0.6350 data_time: 0.0015 memory: 44120 loss: 0.3661 +2023/06/04 06:05:50 - mmengine - INFO - Epoch(train) [5][3900/3907] lr: 5.5029e-05 eta: 3:26:15 time: 0.6348 data_time: 0.0014 memory: 44120 loss: 0.3604 +2023/06/04 06:05:54 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230604_023242 +2023/06/04 06:05:54 - mmengine - INFO - Saving checkpoint at 5 epochs +2023/06/04 06:07:28 - mmengine - INFO - Epoch(val) [5][57/57] accuracy/top1: 76.5074 single-label/precision_classwise: [70.40363311767578, 97.52628326416016] single-label/recall_classwise: [98.98995971679688, 48.89922332763672] single-label/f1-score_classwise: [82.28472900390625, 65.13837432861328] data_time: 0.0367 time: 1.2801 +2023/06/04 06:08:34 - mmengine - INFO - Epoch(train) [6][ 100/3907] lr: 5.4642e-05 eta: 3:25:10 time: 0.6327 data_time: 0.0014 memory: 44120 loss: 0.3573 +2023/06/04 06:09:38 - mmengine - INFO - Epoch(train) [6][ 200/3907] lr: 5.4280e-05 eta: 3:24:07 time: 0.6320 data_time: 0.0016 memory: 44120 loss: 0.3743 +2023/06/04 06:10:41 - mmengine - INFO - Epoch(train) [6][ 300/3907] lr: 5.3918e-05 eta: 3:23:04 time: 0.6321 data_time: 0.0014 memory: 44120 loss: 0.3625 +2023/06/04 06:11:45 - mmengine - INFO - Epoch(train) [6][ 400/3907] lr: 5.3556e-05 eta: 3:22:01 time: 0.6332 data_time: 0.0017 memory: 44120 loss: 0.3758 +2023/06/04 06:12:26 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230604_023242 +2023/06/04 06:12:48 - mmengine - INFO - Epoch(train) [6][ 500/3907] lr: 5.3195e-05 eta: 3:20:57 time: 0.6371 data_time: 0.0017 memory: 44120 loss: 0.3593 +2023/06/04 06:13:51 - mmengine - INFO - Epoch(train) [6][ 600/3907] lr: 5.2833e-05 eta: 3:19:54 time: 0.6343 data_time: 0.0015 memory: 44120 loss: 0.3538 +2023/06/04 06:14:55 - mmengine - INFO - Epoch(train) [6][ 700/3907] lr: 5.2472e-05 eta: 3:18:51 time: 0.6345 data_time: 0.0015 memory: 44120 loss: 0.3648 +2023/06/04 06:15:58 - mmengine - INFO - Epoch(train) [6][ 800/3907] lr: 5.2111e-05 eta: 3:17:47 time: 0.6313 data_time: 0.0015 memory: 44120 loss: 0.3619 +2023/06/04 06:17:01 - mmengine - INFO - Epoch(train) [6][ 900/3907] lr: 5.1750e-05 eta: 3:16:44 time: 0.6326 data_time: 0.0014 memory: 44120 loss: 0.3696 +2023/06/04 06:18:05 - mmengine - INFO - Epoch(train) [6][1000/3907] lr: 5.1389e-05 eta: 3:15:40 time: 0.6338 data_time: 0.0015 memory: 44120 loss: 0.3459 +2023/06/04 06:19:08 - mmengine - INFO - Epoch(train) [6][1100/3907] lr: 5.1029e-05 eta: 3:14:37 time: 0.6349 data_time: 0.0017 memory: 44120 loss: 0.3527 +2023/06/04 06:20:12 - mmengine - INFO - Epoch(train) [6][1200/3907] lr: 5.0668e-05 eta: 3:13:34 time: 0.6342 data_time: 0.0018 memory: 44120 loss: 0.3725 +2023/06/04 06:21:15 - mmengine - INFO - Epoch(train) [6][1300/3907] lr: 5.0308e-05 eta: 3:12:31 time: 0.6320 data_time: 0.0014 memory: 44120 loss: 0.3455 +2023/06/04 06:22:18 - mmengine - INFO - Epoch(train) [6][1400/3907] lr: 4.9949e-05 eta: 3:11:27 time: 0.6330 data_time: 0.0014 memory: 44120 loss: 0.3424 +2023/06/04 06:23:00 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230604_023242 +2023/06/04 06:23:22 - mmengine - INFO - Epoch(train) [6][1500/3907] lr: 4.9589e-05 eta: 3:10:24 time: 0.6329 data_time: 0.0016 memory: 44120 loss: 0.3789 +2023/06/04 06:24:25 - mmengine - INFO - Epoch(train) [6][1600/3907] lr: 4.9230e-05 eta: 3:09:21 time: 0.6345 data_time: 0.0016 memory: 44120 loss: 0.3681 +2023/06/04 06:25:29 - mmengine - INFO - Epoch(train) [6][1700/3907] lr: 4.8871e-05 eta: 3:08:17 time: 0.6336 data_time: 0.0016 memory: 44120 loss: 0.3444 +2023/06/04 06:26:32 - mmengine - INFO - Epoch(train) [6][1800/3907] lr: 4.8513e-05 eta: 3:07:14 time: 0.6330 data_time: 0.0016 memory: 44120 loss: 0.3773 +2023/06/04 06:27:35 - mmengine - INFO - Epoch(train) [6][1900/3907] lr: 4.8155e-05 eta: 3:06:11 time: 0.6329 data_time: 0.0014 memory: 44120 loss: 0.3832 +2023/06/04 06:28:39 - mmengine - INFO - Epoch(train) [6][2000/3907] lr: 4.7798e-05 eta: 3:05:07 time: 0.6331 data_time: 0.0017 memory: 44120 loss: 0.3853 +2023/06/04 06:29:42 - mmengine - INFO - Epoch(train) [6][2100/3907] lr: 4.7441e-05 eta: 3:04:04 time: 0.6360 data_time: 0.0017 memory: 44120 loss: 0.3633 +2023/06/04 06:30:46 - mmengine - INFO - Epoch(train) [6][2200/3907] lr: 4.7084e-05 eta: 3:03:01 time: 0.6337 data_time: 0.0015 memory: 44120 loss: 0.3788 +2023/06/04 06:31:49 - mmengine - INFO - Epoch(train) [6][2300/3907] lr: 4.6729e-05 eta: 3:01:57 time: 0.6335 data_time: 0.0015 memory: 44120 loss: 0.3464 +2023/06/04 06:32:52 - mmengine - INFO - Epoch(train) [6][2400/3907] lr: 4.6373e-05 eta: 3:00:54 time: 0.6332 data_time: 0.0015 memory: 44120 loss: 0.3623 +2023/06/04 06:33:34 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230604_023242 +2023/06/04 06:33:56 - mmengine - INFO - Epoch(train) [6][2500/3907] lr: 4.6018e-05 eta: 2:59:51 time: 0.6315 data_time: 0.0013 memory: 44120 loss: 0.3581 +2023/06/04 06:34:59 - mmengine - INFO - Epoch(train) [6][2600/3907] lr: 4.5664e-05 eta: 2:58:47 time: 0.6339 data_time: 0.0014 memory: 44120 loss: 0.3687 +2023/06/04 06:36:02 - mmengine - INFO - Epoch(train) [6][2700/3907] lr: 4.5310e-05 eta: 2:57:44 time: 0.6328 data_time: 0.0015 memory: 44120 loss: 0.3592 +2023/06/04 06:37:06 - mmengine - INFO - Epoch(train) [6][2800/3907] lr: 4.4957e-05 eta: 2:56:41 time: 0.6324 data_time: 0.0015 memory: 44120 loss: 0.3486 +2023/06/04 06:38:09 - mmengine - INFO - Epoch(train) [6][2900/3907] lr: 4.4605e-05 eta: 2:55:37 time: 0.6320 data_time: 0.0015 memory: 44120 loss: 0.3588 +2023/06/04 06:39:12 - mmengine - INFO - Epoch(train) [6][3000/3907] lr: 4.4253e-05 eta: 2:54:34 time: 0.6337 data_time: 0.0015 memory: 44120 loss: 0.3461 +2023/06/04 06:40:16 - mmengine - INFO - Epoch(train) [6][3100/3907] lr: 4.3902e-05 eta: 2:53:31 time: 0.6331 data_time: 0.0014 memory: 44120 loss: 0.3665 +2023/06/04 06:41:19 - mmengine - INFO - Epoch(train) [6][3200/3907] lr: 4.3552e-05 eta: 2:52:27 time: 0.6312 data_time: 0.0016 memory: 44120 loss: 0.3731 +2023/06/04 06:42:22 - mmengine - INFO - Epoch(train) [6][3300/3907] lr: 4.3202e-05 eta: 2:51:24 time: 0.6308 data_time: 0.0014 memory: 44120 loss: 0.3650 +2023/06/04 06:43:25 - mmengine - INFO - Epoch(train) [6][3400/3907] lr: 4.2854e-05 eta: 2:50:20 time: 0.6322 data_time: 0.0016 memory: 44120 loss: 0.3613 +2023/06/04 06:44:07 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230604_023242 +2023/06/04 06:44:29 - mmengine - INFO - Epoch(train) [6][3500/3907] lr: 4.2506e-05 eta: 2:49:17 time: 0.6320 data_time: 0.0014 memory: 44120 loss: 0.3610 +2023/06/04 06:45:32 - mmengine - INFO - Epoch(train) [6][3600/3907] lr: 4.2158e-05 eta: 2:48:14 time: 0.6320 data_time: 0.0015 memory: 44120 loss: 0.3349 +2023/06/04 06:46:35 - mmengine - INFO - Epoch(train) [6][3700/3907] lr: 4.1812e-05 eta: 2:47:10 time: 0.6349 data_time: 0.0016 memory: 44120 loss: 0.3892 +2023/06/04 06:47:38 - mmengine - INFO - Epoch(train) [6][3800/3907] lr: 4.1466e-05 eta: 2:46:07 time: 0.6314 data_time: 0.0015 memory: 44120 loss: 0.3624 +2023/06/04 06:48:42 - mmengine - INFO - Epoch(train) [6][3900/3907] lr: 4.1122e-05 eta: 2:45:03 time: 0.6330 data_time: 0.0013 memory: 44120 loss: 0.3583 +2023/06/04 06:48:46 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230604_023242 +2023/06/04 06:48:46 - mmengine - INFO - Saving checkpoint at 6 epochs +2023/06/04 06:50:20 - mmengine - INFO - Epoch(val) [6][57/57] accuracy/top1: 76.2360 single-label/precision_classwise: [70.17237854003906, 97.3774642944336] single-label/recall_classwise: [98.93946075439453, 48.356590270996094] single-label/f1-score_classwise: [82.10917663574219, 64.62239837646484] data_time: 0.0359 time: 1.2793 +2023/06/04 06:51:27 - mmengine - INFO - Epoch(train) [7][ 100/3907] lr: 4.0754e-05 eta: 2:43:58 time: 0.6335 data_time: 0.0015 memory: 44120 loss: 0.3612 +2023/06/04 06:52:30 - mmengine - INFO - Epoch(train) [7][ 200/3907] lr: 4.0411e-05 eta: 2:42:54 time: 0.6335 data_time: 0.0014 memory: 44120 loss: 0.3565 +2023/06/04 06:53:34 - mmengine - INFO - Epoch(train) [7][ 300/3907] lr: 4.0069e-05 eta: 2:41:51 time: 0.6357 data_time: 0.0015 memory: 44120 loss: 0.3630 +2023/06/04 06:54:37 - mmengine - INFO - Epoch(train) [7][ 400/3907] lr: 3.9729e-05 eta: 2:40:48 time: 0.6355 data_time: 0.0018 memory: 44120 loss: 0.3903 +2023/06/04 06:55:41 - mmengine - INFO - Epoch(train) [7][ 500/3907] lr: 3.9389e-05 eta: 2:39:45 time: 0.6346 data_time: 0.0023 memory: 44120 loss: 0.3425 +2023/06/04 06:56:18 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230604_023242 +2023/06/04 06:56:44 - mmengine - INFO - Epoch(train) [7][ 600/3907] lr: 3.9050e-05 eta: 2:38:41 time: 0.6340 data_time: 0.0013 memory: 44120 loss: 0.3504 +2023/06/04 06:57:48 - mmengine - INFO - Epoch(train) [7][ 700/3907] lr: 3.8712e-05 eta: 2:37:38 time: 0.6357 data_time: 0.0014 memory: 44120 loss: 0.3758 +2023/06/04 06:58:51 - mmengine - INFO - Epoch(train) [7][ 800/3907] lr: 3.8375e-05 eta: 2:36:35 time: 0.6346 data_time: 0.0016 memory: 44120 loss: 0.3494 +2023/06/04 06:59:55 - mmengine - INFO - Epoch(train) [7][ 900/3907] lr: 3.8039e-05 eta: 2:35:31 time: 0.6326 data_time: 0.0015 memory: 44120 loss: 0.3755 +2023/06/04 07:00:58 - mmengine - INFO - Epoch(train) [7][1000/3907] lr: 3.7705e-05 eta: 2:34:28 time: 0.6336 data_time: 0.0015 memory: 44120 loss: 0.3698 +2023/06/04 07:02:01 - mmengine - INFO - Epoch(train) [7][1100/3907] lr: 3.7371e-05 eta: 2:33:25 time: 0.6340 data_time: 0.0015 memory: 44120 loss: 0.3553 +2023/06/04 07:03:05 - mmengine - INFO - Epoch(train) [7][1200/3907] lr: 3.7039e-05 eta: 2:32:21 time: 0.6317 data_time: 0.0016 memory: 44120 loss: 0.3636 +2023/06/04 07:04:08 - mmengine - INFO - Epoch(train) [7][1300/3907] lr: 3.6708e-05 eta: 2:31:18 time: 0.6335 data_time: 0.0017 memory: 44120 loss: 0.3539 +2023/06/04 07:05:12 - mmengine - INFO - Epoch(train) [7][1400/3907] lr: 3.6378e-05 eta: 2:30:15 time: 0.6329 data_time: 0.0015 memory: 44120 loss: 0.3584 +2023/06/04 07:06:15 - mmengine - INFO - Epoch(train) [7][1500/3907] lr: 3.6049e-05 eta: 2:29:11 time: 0.6357 data_time: 0.0016 memory: 44120 loss: 0.3608 +2023/06/04 07:06:52 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230604_023242 +2023/06/04 07:07:18 - mmengine - INFO - Epoch(train) [7][1600/3907] lr: 3.5721e-05 eta: 2:28:08 time: 0.6321 data_time: 0.0016 memory: 44120 loss: 0.3489 +2023/06/04 07:08:22 - mmengine - INFO - Epoch(train) [7][1700/3907] lr: 3.5395e-05 eta: 2:27:05 time: 0.6320 data_time: 0.0014 memory: 44120 loss: 0.3507 +2023/06/04 07:09:25 - mmengine - INFO - Epoch(train) [7][1800/3907] lr: 3.5070e-05 eta: 2:26:01 time: 0.6400 data_time: 0.0015 memory: 44120 loss: 0.3506 +2023/06/04 07:10:29 - mmengine - INFO - Epoch(train) [7][1900/3907] lr: 3.4746e-05 eta: 2:24:58 time: 0.6330 data_time: 0.0015 memory: 44120 loss: 0.3688 +2023/06/04 07:11:32 - mmengine - INFO - Epoch(train) [7][2000/3907] lr: 3.4424e-05 eta: 2:23:55 time: 0.6326 data_time: 0.0015 memory: 44120 loss: 0.3742 +2023/06/04 07:12:35 - mmengine - INFO - Epoch(train) [7][2100/3907] lr: 3.4103e-05 eta: 2:22:51 time: 0.6347 data_time: 0.0014 memory: 44120 loss: 0.3471 +2023/06/04 07:13:39 - mmengine - INFO - Epoch(train) [7][2200/3907] lr: 3.3783e-05 eta: 2:21:48 time: 0.6324 data_time: 0.0014 memory: 44120 loss: 0.3664 +2023/06/04 07:14:42 - mmengine - INFO - Epoch(train) [7][2300/3907] lr: 3.3465e-05 eta: 2:20:45 time: 0.6325 data_time: 0.0014 memory: 44120 loss: 0.3702 +2023/06/04 07:15:46 - mmengine - INFO - Epoch(train) [7][2400/3907] lr: 3.3148e-05 eta: 2:19:41 time: 0.6339 data_time: 0.0014 memory: 44120 loss: 0.3598 +2023/06/04 07:16:49 - mmengine - INFO - Epoch(train) [7][2500/3907] lr: 3.2832e-05 eta: 2:18:38 time: 0.6329 data_time: 0.0015 memory: 44120 loss: 0.3604 +2023/06/04 07:17:26 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230604_023242 +2023/06/04 07:17:53 - mmengine - INFO - Epoch(train) [7][2600/3907] lr: 3.2518e-05 eta: 2:17:35 time: 0.6328 data_time: 0.0015 memory: 44120 loss: 0.3514 +2023/06/04 07:18:56 - mmengine - INFO - Epoch(train) [7][2700/3907] lr: 3.2205e-05 eta: 2:16:31 time: 0.6335 data_time: 0.0015 memory: 44120 loss: 0.3374 +2023/06/04 07:20:00 - mmengine - INFO - Epoch(train) [7][2800/3907] lr: 3.1894e-05 eta: 2:15:28 time: 0.6347 data_time: 0.0015 memory: 44120 loss: 0.3817 +2023/06/04 07:21:03 - mmengine - INFO - Epoch(train) [7][2900/3907] lr: 3.1584e-05 eta: 2:14:25 time: 0.6337 data_time: 0.0014 memory: 44120 loss: 0.3605 +2023/06/04 07:22:06 - mmengine - INFO - Epoch(train) [7][3000/3907] lr: 3.1276e-05 eta: 2:13:21 time: 0.6345 data_time: 0.0015 memory: 44120 loss: 0.3865 +2023/06/04 07:23:10 - mmengine - INFO - Epoch(train) [7][3100/3907] lr: 3.0969e-05 eta: 2:12:18 time: 0.6334 data_time: 0.0018 memory: 44120 loss: 0.3698 +2023/06/04 07:24:13 - mmengine - INFO - Epoch(train) [7][3200/3907] lr: 3.0664e-05 eta: 2:11:15 time: 0.6349 data_time: 0.0015 memory: 44120 loss: 0.3677 +2023/06/04 07:25:17 - mmengine - INFO - Epoch(train) [7][3300/3907] lr: 3.0360e-05 eta: 2:10:11 time: 0.6328 data_time: 0.0015 memory: 44120 loss: 0.3542 +2023/06/04 07:26:20 - mmengine - INFO - Epoch(train) [7][3400/3907] lr: 3.0058e-05 eta: 2:09:08 time: 0.6336 data_time: 0.0014 memory: 44120 loss: 0.3812 +2023/06/04 07:27:24 - mmengine - INFO - Epoch(train) [7][3500/3907] lr: 2.9758e-05 eta: 2:08:05 time: 0.6338 data_time: 0.0014 memory: 44120 loss: 0.3631 +2023/06/04 07:28:00 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230604_023242 +2023/06/04 07:28:27 - mmengine - INFO - Epoch(train) [7][3600/3907] lr: 2.9459e-05 eta: 2:07:01 time: 0.6345 data_time: 0.0014 memory: 44120 loss: 0.3764 +2023/06/04 07:29:30 - mmengine - INFO - Epoch(train) [7][3700/3907] lr: 2.9162e-05 eta: 2:05:58 time: 0.6333 data_time: 0.0015 memory: 44120 loss: 0.3519 +2023/06/04 07:30:34 - mmengine - INFO - Epoch(train) [7][3800/3907] lr: 2.8867e-05 eta: 2:04:55 time: 0.6374 data_time: 0.0015 memory: 44120 loss: 0.3715 +2023/06/04 07:31:37 - mmengine - INFO - Epoch(train) [7][3900/3907] lr: 2.8573e-05 eta: 2:03:51 time: 0.6319 data_time: 0.0014 memory: 44120 loss: 0.3887 +2023/06/04 07:31:41 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230604_023242 +2023/06/04 07:31:41 - mmengine - INFO - Saving checkpoint at 7 epochs +2023/06/04 07:33:15 - mmengine - INFO - Epoch(val) [7][57/57] accuracy/top1: 75.7802 single-label/precision_classwise: [69.73158264160156, 97.59577178955078] single-label/recall_classwise: [99.05309295654297, 47.2015495300293] single-label/f1-score_classwise: [81.84545135498047, 63.629241943359375] data_time: 0.0355 time: 1.2770 +2023/06/04 07:34:21 - mmengine - INFO - Epoch(train) [8][ 100/3907] lr: 2.8261e-05 eta: 2:02:45 time: 0.6340 data_time: 0.0016 memory: 44120 loss: 0.3730 +2023/06/04 07:35:24 - mmengine - INFO - Epoch(train) [8][ 200/3907] lr: 2.7971e-05 eta: 2:01:41 time: 0.6342 data_time: 0.0015 memory: 44120 loss: 0.3552 +2023/06/04 07:36:28 - mmengine - INFO - Epoch(train) [8][ 300/3907] lr: 2.7682e-05 eta: 2:00:38 time: 0.6325 data_time: 0.0015 memory: 44120 loss: 0.3771 +2023/06/04 07:37:31 - mmengine - INFO - Epoch(train) [8][ 400/3907] lr: 2.7395e-05 eta: 1:59:34 time: 0.6348 data_time: 0.0017 memory: 44120 loss: 0.3430 +2023/06/04 07:38:35 - mmengine - INFO - Epoch(train) [8][ 500/3907] lr: 2.7111e-05 eta: 1:58:31 time: 0.6362 data_time: 0.0014 memory: 44120 loss: 0.3939 +2023/06/04 07:39:38 - mmengine - INFO - Epoch(train) [8][ 600/3907] lr: 2.6828e-05 eta: 1:57:28 time: 0.6309 data_time: 0.0016 memory: 44120 loss: 0.3732 +2023/06/04 07:40:10 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230604_023242 +2023/06/04 07:40:42 - mmengine - INFO - Epoch(train) [8][ 700/3907] lr: 2.6546e-05 eta: 1:56:24 time: 0.6349 data_time: 0.0017 memory: 44120 loss: 0.3528 +2023/06/04 07:41:45 - mmengine - INFO - Epoch(train) [8][ 800/3907] lr: 2.6267e-05 eta: 1:55:21 time: 0.6329 data_time: 0.0016 memory: 44120 loss: 0.3531 +2023/06/04 07:42:48 - mmengine - INFO - Epoch(train) [8][ 900/3907] lr: 2.5989e-05 eta: 1:54:18 time: 0.6321 data_time: 0.0016 memory: 44120 loss: 0.3776 +2023/06/04 07:43:52 - mmengine - INFO - Epoch(train) [8][1000/3907] lr: 2.5714e-05 eta: 1:53:14 time: 0.6349 data_time: 0.0015 memory: 44120 loss: 0.3756 +2023/06/04 07:44:55 - mmengine - INFO - Epoch(train) [8][1100/3907] lr: 2.5440e-05 eta: 1:52:11 time: 0.6344 data_time: 0.0015 memory: 44120 loss: 0.3427 +2023/06/04 07:45:59 - mmengine - INFO - Epoch(train) [8][1200/3907] lr: 2.5168e-05 eta: 1:51:07 time: 0.6341 data_time: 0.0014 memory: 44120 loss: 0.3697 +2023/06/04 07:47:02 - mmengine - INFO - Epoch(train) [8][1300/3907] lr: 2.4898e-05 eta: 1:50:04 time: 0.6346 data_time: 0.0014 memory: 44120 loss: 0.3514 +2023/06/04 07:48:05 - mmengine - INFO - Epoch(train) [8][1400/3907] lr: 2.4630e-05 eta: 1:49:01 time: 0.6337 data_time: 0.0016 memory: 44120 loss: 0.3776 +2023/06/04 07:49:09 - mmengine - INFO - Epoch(train) [8][1500/3907] lr: 2.4364e-05 eta: 1:47:57 time: 0.6335 data_time: 0.0015 memory: 44120 loss: 0.3725 +2023/06/04 07:50:12 - mmengine - INFO - Epoch(train) [8][1600/3907] lr: 2.4100e-05 eta: 1:46:54 time: 0.6333 data_time: 0.0015 memory: 44120 loss: 0.3723 +2023/06/04 07:50:45 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230604_023242 +2023/06/04 07:51:16 - mmengine - INFO - Epoch(train) [8][1700/3907] lr: 2.3838e-05 eta: 1:45:51 time: 0.6317 data_time: 0.0014 memory: 44120 loss: 0.3901 +2023/06/04 07:52:19 - mmengine - INFO - Epoch(train) [8][1800/3907] lr: 2.3578e-05 eta: 1:44:47 time: 0.6337 data_time: 0.0014 memory: 44120 loss: 0.3596 +2023/06/04 07:53:22 - mmengine - INFO - Epoch(train) [8][1900/3907] lr: 2.3320e-05 eta: 1:43:44 time: 0.6361 data_time: 0.0014 memory: 44120 loss: 0.3703 +2023/06/04 07:54:26 - mmengine - INFO - Epoch(train) [8][2000/3907] lr: 2.3064e-05 eta: 1:42:41 time: 0.6320 data_time: 0.0015 memory: 44120 loss: 0.3531 +2023/06/04 07:55:29 - mmengine - INFO - Epoch(train) [8][2100/3907] lr: 2.2810e-05 eta: 1:41:37 time: 0.6363 data_time: 0.0016 memory: 44120 loss: 0.3571 +2023/06/04 07:56:33 - mmengine - INFO - Epoch(train) [8][2200/3907] lr: 2.2558e-05 eta: 1:40:34 time: 0.6339 data_time: 0.0014 memory: 44120 loss: 0.4054 +2023/06/04 07:57:36 - mmengine - INFO - Epoch(train) [8][2300/3907] lr: 2.2309e-05 eta: 1:39:30 time: 0.6336 data_time: 0.0017 memory: 44120 loss: 0.3794 +2023/06/04 07:58:40 - mmengine - INFO - Epoch(train) [8][2400/3907] lr: 2.2061e-05 eta: 1:38:27 time: 0.6325 data_time: 0.0014 memory: 44120 loss: 0.3795 +2023/06/04 07:59:43 - mmengine - INFO - Epoch(train) [8][2500/3907] lr: 2.1816e-05 eta: 1:37:24 time: 0.6372 data_time: 0.0014 memory: 44120 loss: 0.3520 +2023/06/04 08:00:47 - mmengine - INFO - Epoch(train) [8][2600/3907] lr: 2.1572e-05 eta: 1:36:20 time: 0.6353 data_time: 0.0019 memory: 44120 loss: 0.3727 +2023/06/04 08:01:19 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230604_023242 +2023/06/04 08:01:50 - mmengine - INFO - Epoch(train) [8][2700/3907] lr: 2.1331e-05 eta: 1:35:17 time: 0.6355 data_time: 0.0017 memory: 44120 loss: 0.3668 +2023/06/04 08:02:54 - mmengine - INFO - Epoch(train) [8][2800/3907] lr: 2.1092e-05 eta: 1:34:14 time: 0.6385 data_time: 0.0016 memory: 44120 loss: 0.3490 +2023/06/04 08:03:57 - mmengine - INFO - Epoch(train) [8][2900/3907] lr: 2.0855e-05 eta: 1:33:10 time: 0.6329 data_time: 0.0017 memory: 44120 loss: 0.3609 +2023/06/04 08:05:01 - mmengine - INFO - Epoch(train) [8][3000/3907] lr: 2.0621e-05 eta: 1:32:07 time: 0.6338 data_time: 0.0014 memory: 44120 loss: 0.3527 +2023/06/04 08:06:04 - mmengine - INFO - Epoch(train) [8][3100/3907] lr: 2.0388e-05 eta: 1:31:04 time: 0.6359 data_time: 0.0015 memory: 44120 loss: 0.3726 +2023/06/04 08:07:08 - mmengine - INFO - Epoch(train) [8][3200/3907] lr: 2.0158e-05 eta: 1:30:00 time: 0.6337 data_time: 0.0016 memory: 44120 loss: 0.3796 +2023/06/04 08:08:11 - mmengine - INFO - Epoch(train) [8][3300/3907] lr: 1.9930e-05 eta: 1:28:57 time: 0.6433 data_time: 0.0016 memory: 44120 loss: 0.3587 +2023/06/04 08:09:15 - mmengine - INFO - Epoch(train) [8][3400/3907] lr: 1.9705e-05 eta: 1:27:54 time: 0.6327 data_time: 0.0015 memory: 44120 loss: 0.3907 +2023/06/04 08:10:18 - mmengine - INFO - Epoch(train) [8][3500/3907] lr: 1.9481e-05 eta: 1:26:50 time: 0.6333 data_time: 0.0015 memory: 44120 loss: 0.3945 +2023/06/04 08:11:21 - mmengine - INFO - Epoch(train) [8][3600/3907] lr: 1.9260e-05 eta: 1:25:47 time: 0.6347 data_time: 0.0016 memory: 44120 loss: 0.3552 +2023/06/04 08:11:54 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230604_023242 +2023/06/04 08:12:25 - mmengine - INFO - Epoch(train) [8][3700/3907] lr: 1.9042e-05 eta: 1:24:43 time: 0.6325 data_time: 0.0015 memory: 44120 loss: 0.3542 +2023/06/04 08:13:28 - mmengine - INFO - Epoch(train) [8][3800/3907] lr: 1.8825e-05 eta: 1:23:40 time: 0.6365 data_time: 0.0018 memory: 44120 loss: 0.3764 +2023/06/04 08:14:32 - mmengine - INFO - Epoch(train) [8][3900/3907] lr: 1.8611e-05 eta: 1:22:37 time: 0.6318 data_time: 0.0013 memory: 44120 loss: 0.3572 +2023/06/04 08:14:36 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230604_023242 +2023/06/04 08:14:36 - mmengine - INFO - Saving checkpoint at 8 epochs +2023/06/04 08:16:10 - mmengine - INFO - Epoch(val) [8][57/57] accuracy/top1: 75.7802 single-label/precision_classwise: [69.68782806396484, 97.98028564453125] single-label/recall_classwise: [99.21090698242188, 47.00775146484375] single-label/f1-score_classwise: [81.86907958984375, 63.533973693847656] data_time: 0.0358 time: 1.2776 +2023/06/04 08:17:16 - mmengine - INFO - Epoch(train) [9][ 100/3907] lr: 1.8385e-05 eta: 1:21:30 time: 0.6344 data_time: 0.0015 memory: 44120 loss: 0.3495 +2023/06/04 08:18:20 - mmengine - INFO - Epoch(train) [9][ 200/3907] lr: 1.8176e-05 eta: 1:20:26 time: 0.6375 data_time: 0.0014 memory: 44120 loss: 0.3595 +2023/06/04 08:19:23 - mmengine - INFO - Epoch(train) [9][ 300/3907] lr: 1.7969e-05 eta: 1:19:23 time: 0.6353 data_time: 0.0028 memory: 44120 loss: 0.3680 +2023/06/04 08:20:27 - mmengine - INFO - Epoch(train) [9][ 400/3907] lr: 1.7765e-05 eta: 1:18:19 time: 0.6322 data_time: 0.0016 memory: 44120 loss: 0.3718 +2023/06/04 08:21:30 - mmengine - INFO - Epoch(train) [9][ 500/3907] lr: 1.7563e-05 eta: 1:17:16 time: 0.6358 data_time: 0.0018 memory: 44120 loss: 0.3680 +2023/06/04 08:22:34 - mmengine - INFO - Epoch(train) [9][ 600/3907] lr: 1.7363e-05 eta: 1:16:13 time: 0.6347 data_time: 0.0015 memory: 44120 loss: 0.3675 +2023/06/04 08:23:37 - mmengine - INFO - Epoch(train) [9][ 700/3907] lr: 1.7166e-05 eta: 1:15:09 time: 0.6332 data_time: 0.0015 memory: 44120 loss: 0.3826 +2023/06/04 08:24:05 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230604_023242 +2023/06/04 08:24:41 - mmengine - INFO - Epoch(train) [9][ 800/3907] lr: 1.6971e-05 eta: 1:14:06 time: 0.6333 data_time: 0.0014 memory: 44120 loss: 0.3639 +2023/06/04 08:25:44 - mmengine - INFO - Epoch(train) [9][ 900/3907] lr: 1.6779e-05 eta: 1:13:03 time: 0.6411 data_time: 0.0015 memory: 44120 loss: 0.3700 +2023/06/04 08:26:47 - mmengine - INFO - Epoch(train) [9][1000/3907] lr: 1.6589e-05 eta: 1:11:59 time: 0.6346 data_time: 0.0014 memory: 44120 loss: 0.3458 +2023/06/04 08:27:51 - mmengine - INFO - Epoch(train) [9][1100/3907] lr: 1.6402e-05 eta: 1:10:56 time: 0.6327 data_time: 0.0016 memory: 44120 loss: 0.3535 +2023/06/04 08:28:54 - mmengine - INFO - Epoch(train) [9][1200/3907] lr: 1.6217e-05 eta: 1:09:52 time: 0.6365 data_time: 0.0017 memory: 44120 loss: 0.3572 +2023/06/04 08:29:58 - mmengine - INFO - Epoch(train) [9][1300/3907] lr: 1.6035e-05 eta: 1:08:49 time: 0.6346 data_time: 0.0020 memory: 44120 loss: 0.3475 +2023/06/04 08:31:01 - mmengine - INFO - Epoch(train) [9][1400/3907] lr: 1.5855e-05 eta: 1:07:46 time: 0.6336 data_time: 0.0018 memory: 44120 loss: 0.3729 +2023/06/04 08:32:05 - mmengine - INFO - Epoch(train) [9][1500/3907] lr: 1.5678e-05 eta: 1:06:42 time: 0.6328 data_time: 0.0015 memory: 44120 loss: 0.3559 +2023/06/04 08:33:08 - mmengine - INFO - Epoch(train) [9][1600/3907] lr: 1.5503e-05 eta: 1:05:39 time: 0.6341 data_time: 0.0016 memory: 44120 loss: 0.3595 +2023/06/04 08:34:11 - mmengine - INFO - Epoch(train) [9][1700/3907] lr: 1.5331e-05 eta: 1:04:35 time: 0.6357 data_time: 0.0016 memory: 44120 loss: 0.3552 +2023/06/04 08:34:39 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230604_023242 +2023/06/04 08:35:15 - mmengine - INFO - Epoch(train) [9][1800/3907] lr: 1.5162e-05 eta: 1:03:32 time: 0.6329 data_time: 0.0018 memory: 44120 loss: 0.3564 +2023/06/04 08:36:18 - mmengine - INFO - Epoch(train) [9][1900/3907] lr: 1.4995e-05 eta: 1:02:29 time: 0.6348 data_time: 0.0016 memory: 44120 loss: 0.3537 +2023/06/04 08:37:22 - mmengine - INFO - Epoch(train) [9][2000/3907] lr: 1.4830e-05 eta: 1:01:25 time: 0.6334 data_time: 0.0016 memory: 44120 loss: 0.3665 +2023/06/04 08:38:25 - mmengine - INFO - Epoch(train) [9][2100/3907] lr: 1.4668e-05 eta: 1:00:22 time: 0.6319 data_time: 0.0015 memory: 44120 loss: 0.3891 +2023/06/04 08:39:29 - mmengine - INFO - Epoch(train) [9][2200/3907] lr: 1.4509e-05 eta: 0:59:19 time: 0.6325 data_time: 0.0015 memory: 44120 loss: 0.3958 +2023/06/04 08:40:32 - mmengine - INFO - Epoch(train) [9][2300/3907] lr: 1.4353e-05 eta: 0:58:15 time: 0.6332 data_time: 0.0017 memory: 44120 loss: 0.3435 +2023/06/04 08:41:36 - mmengine - INFO - Epoch(train) [9][2400/3907] lr: 1.4199e-05 eta: 0:57:12 time: 0.6342 data_time: 0.0017 memory: 44120 loss: 0.3579 +2023/06/04 08:42:39 - mmengine - INFO - Epoch(train) [9][2500/3907] lr: 1.4047e-05 eta: 0:56:08 time: 0.6344 data_time: 0.0016 memory: 44120 loss: 0.3867 +2023/06/04 08:43:43 - mmengine - INFO - Epoch(train) [9][2600/3907] lr: 1.3899e-05 eta: 0:55:05 time: 0.6369 data_time: 0.0014 memory: 44120 loss: 0.3729 +2023/06/04 08:44:46 - mmengine - INFO - Epoch(train) [9][2700/3907] lr: 1.3753e-05 eta: 0:54:02 time: 0.6345 data_time: 0.0016 memory: 44120 loss: 0.3821 +2023/06/04 08:45:14 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230604_023242 +2023/06/04 08:45:49 - mmengine - INFO - Epoch(train) [9][2800/3907] lr: 1.3609e-05 eta: 0:52:58 time: 0.6340 data_time: 0.0016 memory: 44120 loss: 0.3527 +2023/06/04 08:46:53 - mmengine - INFO - Epoch(train) [9][2900/3907] lr: 1.3469e-05 eta: 0:51:55 time: 0.6323 data_time: 0.0014 memory: 44120 loss: 0.3605 +2023/06/04 08:47:56 - mmengine - INFO - Epoch(train) [9][3000/3907] lr: 1.3331e-05 eta: 0:50:51 time: 0.6334 data_time: 0.0014 memory: 44120 loss: 0.3812 +2023/06/04 08:48:59 - mmengine - INFO - Epoch(train) [9][3100/3907] lr: 1.3196e-05 eta: 0:49:48 time: 0.6360 data_time: 0.0014 memory: 44120 loss: 0.3831 +2023/06/04 08:50:03 - mmengine - INFO - Epoch(train) [9][3200/3907] lr: 1.3063e-05 eta: 0:48:45 time: 0.6355 data_time: 0.0014 memory: 44120 loss: 0.3699 +2023/06/04 08:51:06 - mmengine - INFO - Epoch(train) [9][3300/3907] lr: 1.2933e-05 eta: 0:47:41 time: 0.6351 data_time: 0.0014 memory: 44120 loss: 0.3399 +2023/06/04 08:52:10 - mmengine - INFO - Epoch(train) [9][3400/3907] lr: 1.2806e-05 eta: 0:46:38 time: 0.6338 data_time: 0.0016 memory: 44120 loss: 0.3795 +2023/06/04 08:53:13 - mmengine - INFO - Epoch(train) [9][3500/3907] lr: 1.2682e-05 eta: 0:45:34 time: 0.6343 data_time: 0.0016 memory: 44120 loss: 0.3614 +2023/06/04 08:54:17 - mmengine - INFO - Epoch(train) [9][3600/3907] lr: 1.2560e-05 eta: 0:44:31 time: 0.6327 data_time: 0.0015 memory: 44120 loss: 0.3564 +2023/06/04 08:55:20 - mmengine - INFO - Epoch(train) [9][3700/3907] lr: 1.2441e-05 eta: 0:43:28 time: 0.6366 data_time: 0.0014 memory: 44120 loss: 0.3815 +2023/06/04 08:55:48 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230604_023242 +2023/06/04 08:56:24 - mmengine - INFO - Epoch(train) [9][3800/3907] lr: 1.2325e-05 eta: 0:42:24 time: 0.6338 data_time: 0.0015 memory: 44120 loss: 0.3600 +2023/06/04 08:57:27 - mmengine - INFO - Epoch(train) [9][3900/3907] lr: 1.2211e-05 eta: 0:41:21 time: 0.6359 data_time: 0.0013 memory: 44120 loss: 0.3463 +2023/06/04 08:57:31 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230604_023242 +2023/06/04 08:57:31 - mmengine - INFO - Saving checkpoint at 9 epochs +2023/06/04 08:59:05 - mmengine - INFO - Epoch(val) [9][57/57] accuracy/top1: 75.5854 single-label/precision_classwise: [69.53594970703125, 97.77489471435547] single-label/recall_classwise: [99.1351547241211, 46.66666793823242] single-label/f1-score_classwise: [81.73844146728516, 63.178890228271484] data_time: 0.0357 time: 1.2794 +2023/06/04 09:00:11 - mmengine - INFO - Epoch(train) [10][ 100/3907] lr: 1.2093e-05 eta: 0:40:13 time: 0.6345 data_time: 0.0015 memory: 44120 loss: 0.3396 +2023/06/04 09:01:15 - mmengine - INFO - Epoch(train) [10][ 200/3907] lr: 1.1985e-05 eta: 0:39:10 time: 0.6342 data_time: 0.0016 memory: 44120 loss: 0.3449 +2023/06/04 09:02:18 - mmengine - INFO - Epoch(train) [10][ 300/3907] lr: 1.1881e-05 eta: 0:38:06 time: 0.6348 data_time: 0.0016 memory: 44120 loss: 0.3743 +2023/06/04 09:03:21 - mmengine - INFO - Epoch(train) [10][ 400/3907] lr: 1.1778e-05 eta: 0:37:03 time: 0.6322 data_time: 0.0017 memory: 44120 loss: 0.3553 +2023/06/04 09:04:25 - mmengine - INFO - Epoch(train) [10][ 500/3907] lr: 1.1679e-05 eta: 0:36:00 time: 0.6316 data_time: 0.0016 memory: 44120 loss: 0.3930 +2023/06/04 09:05:28 - mmengine - INFO - Epoch(train) [10][ 600/3907] lr: 1.1583e-05 eta: 0:34:56 time: 0.6321 data_time: 0.0017 memory: 44120 loss: 0.3906 +2023/06/04 09:06:31 - mmengine - INFO - Epoch(train) [10][ 700/3907] lr: 1.1489e-05 eta: 0:33:53 time: 0.6356 data_time: 0.0015 memory: 44120 loss: 0.3637 +2023/06/04 09:07:35 - mmengine - INFO - Epoch(train) [10][ 800/3907] lr: 1.1398e-05 eta: 0:32:49 time: 0.6324 data_time: 0.0016 memory: 44120 loss: 0.3630 +2023/06/04 09:07:58 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230604_023242 +2023/06/04 09:08:38 - mmengine - INFO - Epoch(train) [10][ 900/3907] lr: 1.1310e-05 eta: 0:31:46 time: 0.6352 data_time: 0.0014 memory: 44120 loss: 0.3542 +2023/06/04 09:09:42 - mmengine - INFO - Epoch(train) [10][1000/3907] lr: 1.1225e-05 eta: 0:30:43 time: 0.6527 data_time: 0.0016 memory: 44120 loss: 0.3491 +2023/06/04 09:10:45 - mmengine - INFO - Epoch(train) [10][1100/3907] lr: 1.1142e-05 eta: 0:29:39 time: 0.6348 data_time: 0.0015 memory: 44120 loss: 0.3741 +2023/06/04 09:11:49 - mmengine - INFO - Epoch(train) [10][1200/3907] lr: 1.1063e-05 eta: 0:28:36 time: 0.6325 data_time: 0.0015 memory: 44120 loss: 0.3636 +2023/06/04 09:12:52 - mmengine - INFO - Epoch(train) [10][1300/3907] lr: 1.0986e-05 eta: 0:27:32 time: 0.6334 data_time: 0.0018 memory: 44120 loss: 0.3939 +2023/06/04 09:13:55 - mmengine - INFO - Epoch(train) [10][1400/3907] lr: 1.0912e-05 eta: 0:26:29 time: 0.6320 data_time: 0.0019 memory: 44120 loss: 0.3429 +2023/06/04 09:14:59 - mmengine - INFO - Epoch(train) [10][1500/3907] lr: 1.0841e-05 eta: 0:25:26 time: 0.6352 data_time: 0.0015 memory: 44120 loss: 0.3698 +2023/06/04 09:16:02 - mmengine - INFO - Epoch(train) [10][1600/3907] lr: 1.0773e-05 eta: 0:24:22 time: 0.6329 data_time: 0.0015 memory: 44120 loss: 0.3477 +2023/06/04 09:17:06 - mmengine - INFO - Epoch(train) [10][1700/3907] lr: 1.0707e-05 eta: 0:23:19 time: 0.6327 data_time: 0.0016 memory: 44120 loss: 0.3464 +2023/06/04 09:18:09 - mmengine - INFO - Epoch(train) [10][1800/3907] lr: 1.0645e-05 eta: 0:22:15 time: 0.6322 data_time: 0.0017 memory: 44120 loss: 0.3947 +2023/06/04 09:18:33 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230604_023242 +2023/06/04 09:19:13 - mmengine - INFO - Epoch(train) [10][1900/3907] lr: 1.0585e-05 eta: 0:21:12 time: 0.6449 data_time: 0.0016 memory: 44120 loss: 0.3951 +2023/06/04 09:20:16 - mmengine - INFO - Epoch(train) [10][2000/3907] lr: 1.0529e-05 eta: 0:20:09 time: 0.6346 data_time: 0.0015 memory: 44120 loss: 0.3685 +2023/06/04 09:21:20 - mmengine - INFO - Epoch(train) [10][2100/3907] lr: 1.0475e-05 eta: 0:19:05 time: 0.6340 data_time: 0.0015 memory: 44120 loss: 0.3467 +2023/06/04 09:22:23 - mmengine - INFO - Epoch(train) [10][2200/3907] lr: 1.0424e-05 eta: 0:18:02 time: 0.6336 data_time: 0.0015 memory: 44120 loss: 0.3592 +2023/06/04 09:23:27 - mmengine - INFO - Epoch(train) [10][2300/3907] lr: 1.0376e-05 eta: 0:16:58 time: 0.6350 data_time: 0.0017 memory: 44120 loss: 0.3563 +2023/06/04 09:24:30 - mmengine - INFO - Epoch(train) [10][2400/3907] lr: 1.0330e-05 eta: 0:15:55 time: 0.6366 data_time: 0.0018 memory: 44120 loss: 0.3344 +2023/06/04 09:25:34 - mmengine - INFO - Epoch(train) [10][2500/3907] lr: 1.0288e-05 eta: 0:14:52 time: 0.6345 data_time: 0.0015 memory: 44120 loss: 0.3414 +2023/06/04 09:26:37 - mmengine - INFO - Epoch(train) [10][2600/3907] lr: 1.0249e-05 eta: 0:13:48 time: 0.6319 data_time: 0.0017 memory: 44120 loss: 0.3623 +2023/06/04 09:27:41 - mmengine - INFO - Epoch(train) [10][2700/3907] lr: 1.0212e-05 eta: 0:12:45 time: 0.6367 data_time: 0.0019 memory: 44120 loss: 0.3667 +2023/06/04 09:28:44 - mmengine - INFO - Epoch(train) [10][2800/3907] lr: 1.0178e-05 eta: 0:11:41 time: 0.6349 data_time: 0.0015 memory: 44120 loss: 0.3548 +2023/06/04 09:29:08 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230604_023242 +2023/06/04 09:29:48 - mmengine - INFO - Epoch(train) [10][2900/3907] lr: 1.0148e-05 eta: 0:10:38 time: 0.6346 data_time: 0.0016 memory: 44120 loss: 0.3793 +2023/06/04 09:30:51 - mmengine - INFO - Epoch(train) [10][3000/3907] lr: 1.0120e-05 eta: 0:09:35 time: 0.6343 data_time: 0.0014 memory: 44120 loss: 0.3700 +2023/06/04 09:31:55 - mmengine - INFO - Epoch(train) [10][3100/3907] lr: 1.0095e-05 eta: 0:08:31 time: 0.6384 data_time: 0.0014 memory: 44120 loss: 0.3960 +2023/06/04 09:32:59 - mmengine - INFO - Epoch(train) [10][3200/3907] lr: 1.0073e-05 eta: 0:07:28 time: 0.6334 data_time: 0.0021 memory: 44120 loss: 0.3631 +2023/06/04 09:34:02 - mmengine - INFO - Epoch(train) [10][3300/3907] lr: 1.0054e-05 eta: 0:06:24 time: 0.6344 data_time: 0.0017 memory: 44120 loss: 0.3840 +2023/06/04 09:35:05 - mmengine - INFO - Epoch(train) [10][3400/3907] lr: 1.0038e-05 eta: 0:05:21 time: 0.6332 data_time: 0.0014 memory: 44120 loss: 0.3818 +2023/06/04 09:36:09 - mmengine - INFO - Epoch(train) [10][3500/3907] lr: 1.0024e-05 eta: 0:04:18 time: 0.6335 data_time: 0.0017 memory: 44120 loss: 0.3639 +2023/06/04 09:37:12 - mmengine - INFO - Epoch(train) [10][3600/3907] lr: 1.0014e-05 eta: 0:03:14 time: 0.6340 data_time: 0.0015 memory: 44120 loss: 0.3738 +2023/06/04 09:38:16 - mmengine - INFO - Epoch(train) [10][3700/3907] lr: 1.0006e-05 eta: 0:02:11 time: 0.6357 data_time: 0.0019 memory: 44120 loss: 0.3470 +2023/06/04 09:39:19 - mmengine - INFO - Epoch(train) [10][3800/3907] lr: 1.0002e-05 eta: 0:01:07 time: 0.6330 data_time: 0.0014 memory: 44120 loss: 0.3580 +2023/06/04 09:39:43 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230604_023242 +2023/06/04 09:40:23 - mmengine - INFO - Epoch(train) [10][3900/3907] lr: 1.0000e-05 eta: 0:00:04 time: 0.6328 data_time: 0.0013 memory: 44120 loss: 0.3590 +2023/06/04 09:40:27 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230604_023242 +2023/06/04 09:40:27 - mmengine - INFO - Saving checkpoint at 10 epochs +2023/06/04 09:42:01 - mmengine - INFO - Epoch(val) [10][57/57] accuracy/top1: 75.4532 single-label/precision_classwise: [69.40714263916016, 97.87059783935547] single-label/recall_classwise: [99.1793441772461, 46.31782913208008] single-label/f1-score_classwise: [81.66437530517578, 62.87818908691406] data_time: 0.0351 time: 1.2764 diff --git a/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm/20230604_023242/vis_data/20230604_023242.json b/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm/20230604_023242/vis_data/20230604_023242.json new file mode 100644 index 0000000000000000000000000000000000000000..4c318ecb20e8799b32aa51adfb67486042a0e169 --- /dev/null +++ b/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm/20230604_023242/vis_data/20230604_023242.json @@ -0,0 +1,400 @@ +{"lr": 9.999857418382631e-05, "data_time": 0.0013934850692749023, "loss": 0.6554573059082032, "time": 0.6275890350341797, "epoch": 1, "memory": 44121, "step": 100} +{"lr": 9.999423907359506e-05, "data_time": 0.001732015609741211, "loss": 0.6052772521972656, "time": 0.6291704654693604, "epoch": 1, "memory": 44121, "step": 200} +{"lr": 9.998699480412073e-05, "data_time": 0.001477670669555664, "loss": 0.6014793395996094, "time": 0.6299166202545166, "epoch": 1, "memory": 44121, "step": 300} +{"lr": 9.997684184379085e-05, "data_time": 0.0015866518020629882, "loss": 0.5677017211914063, "time": 0.6311460256576538, "epoch": 1, "memory": 44121, "step": 400} +{"lr": 9.996378084905784e-05, "data_time": 0.0016239166259765625, "loss": 0.5291509628295898, "time": 0.6303429365158081, "epoch": 1, "memory": 44121, "step": 500} +{"lr": 9.994781266439667e-05, "data_time": 0.0013891220092773437, "loss": 0.49774875640869143, "time": 0.6303493738174438, "epoch": 1, "memory": 44121, "step": 600} +{"lr": 9.992893832225061e-05, "data_time": 0.001440286636352539, "loss": 0.4941108703613281, "time": 0.6305997133255005, "epoch": 1, "memory": 44121, "step": 700} +{"lr": 9.990715904296405e-05, "data_time": 0.0015773534774780273, "loss": 0.4833106994628906, "time": 0.6303353071212768, "epoch": 1, "memory": 44121, "step": 800} +{"lr": 9.988247623470371e-05, "data_time": 0.001455545425415039, "loss": 0.5067439079284668, "time": 0.6297955513000488, "epoch": 1, "memory": 44121, "step": 900} +{"lr": 9.985489149336749e-05, "data_time": 0.001456451416015625, "loss": 0.48727102279663087, "time": 0.630203366279602, "epoch": 1, "memory": 44121, "step": 1000} +{"lr": 9.982440660248187e-05, "data_time": 0.0014949083328247071, "loss": 0.47072653770446776, "time": 0.6298863410949707, "epoch": 1, "memory": 44121, "step": 1100} +{"lr": 9.979102353308574e-05, "data_time": 0.0013971328735351562, "loss": 0.4766227722167969, "time": 0.6302656412124634, "epoch": 1, "memory": 44121, "step": 1200} +{"lr": 9.975474444360358e-05, "data_time": 0.0014298439025878906, "loss": 0.45115349292755125, "time": 0.6305836915969849, "epoch": 1, "memory": 44121, "step": 1300} +{"lr": 9.97155716797056e-05, "data_time": 0.0014416933059692382, "loss": 0.44846794605255125, "time": 0.6303877830505371, "epoch": 1, "memory": 44121, "step": 1400} +{"lr": 9.967350777415632e-05, "data_time": 0.0013402223587036133, "loss": 0.4739827275276184, "time": 0.6301686763763428, "epoch": 1, "memory": 44121, "step": 1500} +{"lr": 9.962855544665059e-05, "data_time": 0.001266622543334961, "loss": 0.4373971700668335, "time": 0.6306634426116944, "epoch": 1, "memory": 44121, "step": 1600} +{"lr": 9.958071760363773e-05, "data_time": 0.0014128923416137696, "loss": 0.4457846760749817, "time": 0.6300257921218873, "epoch": 1, "memory": 44121, "step": 1700} +{"lr": 9.952999733813382e-05, "data_time": 0.0014392852783203125, "loss": 0.4599105954170227, "time": 0.6298099040985108, "epoch": 1, "memory": 44121, "step": 1800} +{"lr": 9.947639792952148e-05, "data_time": 0.00150909423828125, "loss": 0.4459614634513855, "time": 0.6304698467254639, "epoch": 1, "memory": 44121, "step": 1900} +{"lr": 9.941992284333813e-05, "data_time": 0.0017377376556396485, "loss": 0.42215226888656615, "time": 0.6298106431961059, "epoch": 1, "memory": 44121, "step": 2000} +{"lr": 9.93605757310516e-05, "data_time": 0.0016527652740478515, "loss": 0.45085784792900085, "time": 0.6294882774353028, "epoch": 1, "memory": 44121, "step": 2100} +{"lr": 9.929836042982421e-05, "data_time": 0.0014800786972045898, "loss": 0.4206417202949524, "time": 0.6305043935775757, "epoch": 1, "memory": 44121, "step": 2200} +{"lr": 9.923328096226467e-05, "data_time": 0.001474308967590332, "loss": 0.428151935338974, "time": 0.630351448059082, "epoch": 1, "memory": 44121, "step": 2300} +{"lr": 9.916534153616819e-05, "data_time": 0.0013605356216430664, "loss": 0.4223970860242844, "time": 0.6298309326171875, "epoch": 1, "memory": 44121, "step": 2400} +{"lr": 9.909454654424387e-05, "data_time": 0.0013653755187988282, "loss": 0.43172143697738646, "time": 0.6304721593856811, "epoch": 1, "memory": 44121, "step": 2500} +{"lr": 9.902090056383109e-05, "data_time": 0.0013478517532348633, "loss": 0.42010399103164675, "time": 0.6304569244384766, "epoch": 1, "memory": 44121, "step": 2600} +{"lr": 9.894440835660365e-05, "data_time": 0.0014511585235595704, "loss": 0.39755138754844666, "time": 0.6310604810714722, "epoch": 1, "memory": 44121, "step": 2700} +{"lr": 9.886507486826162e-05, "data_time": 0.0014309167861938476, "loss": 0.3904307305812836, "time": 0.6300219774246216, "epoch": 1, "memory": 44121, "step": 2800} +{"lr": 9.878290522821154e-05, "data_time": 0.001474452018737793, "loss": 0.4230853796005249, "time": 0.630784010887146, "epoch": 1, "memory": 44121, "step": 2900} +{"lr": 9.86979047492352e-05, "data_time": 0.001485919952392578, "loss": 0.40724811553955076, "time": 0.630129361152649, "epoch": 1, "memory": 44121, "step": 3000} +{"lr": 9.861007892714578e-05, "data_time": 0.0015177488327026366, "loss": 0.42819066643714904, "time": 0.6301684856414795, "epoch": 1, "memory": 44121, "step": 3100} +{"lr": 9.851943344043239e-05, "data_time": 0.0015147209167480468, "loss": 0.419231653213501, "time": 0.6296551465988159, "epoch": 1, "memory": 44121, "step": 3200} +{"lr": 9.842597414989319e-05, "data_time": 0.0014043331146240234, "loss": 0.40978620052337644, "time": 0.6303950071334838, "epoch": 1, "memory": 44121, "step": 3300} +{"lr": 9.832970709825637e-05, "data_time": 0.0014115333557128905, "loss": 0.3929266661405563, "time": 0.6304152488708497, "epoch": 1, "memory": 44121, "step": 3400} +{"lr": 9.823063850978968e-05, "data_time": 0.0015639781951904297, "loss": 0.4143178790807724, "time": 0.6300772905349732, "epoch": 1, "memory": 44121, "step": 3500} +{"lr": 9.81287747898975e-05, "data_time": 0.0017814397811889648, "loss": 0.4079999029636383, "time": 0.6292192220687867, "epoch": 1, "memory": 44121, "step": 3600} +{"lr": 9.802412252470715e-05, "data_time": 0.0014991044998168945, "loss": 0.3691905975341797, "time": 0.6297514200210571, "epoch": 1, "memory": 44121, "step": 3700} +{"lr": 9.791668848064258e-05, "data_time": 0.0014405965805053711, "loss": 0.3997865170240402, "time": 0.6306843519210815, "epoch": 1, "memory": 44121, "step": 3800} +{"lr": 9.780647960398751e-05, "data_time": 0.0011514902114868164, "loss": 0.39070974588394164, "time": 0.6291760921478271, "epoch": 1, "memory": 44121, "step": 3900} +{"accuracy/top1": 73.46299743652344, "data_time": 0.046683474590903835, "time": 1.3394464317121004, "step": 1} +{"lr": 9.7685491196646e-05, "data_time": 0.0016675472259521484, "loss": 0.40552919507026675, "time": 0.6314837455749511, "epoch": 2, "memory": 44121, "step": 4007} +{"lr": 9.756956126085676e-05, "data_time": 0.0013395309448242187, "loss": 0.3878222227096558, "time": 0.6305492401123047, "epoch": 2, "memory": 44120, "step": 4107} +{"lr": 9.745087893642482e-05, "data_time": 0.0015935420989990235, "loss": 0.3939862012863159, "time": 0.6303120136260987, "epoch": 2, "memory": 44120, "step": 4207} +{"lr": 9.73294518969056e-05, "data_time": 0.002163076400756836, "loss": 0.381178480386734, "time": 0.6314022541046143, "epoch": 2, "memory": 44120, "step": 4307} +{"lr": 9.72052879933172e-05, "data_time": 0.002112078666687012, "loss": 0.40107953250408174, "time": 0.635162091255188, "epoch": 2, "memory": 44120, "step": 4407} +{"lr": 9.707839525363358e-05, "data_time": 0.0016361236572265624, "loss": 0.39447016417980196, "time": 0.6308466911315918, "epoch": 2, "memory": 44120, "step": 4507} +{"lr": 9.69487818822646e-05, "data_time": 0.0017443418502807618, "loss": 0.39011659622192385, "time": 0.6310233116149903, "epoch": 2, "memory": 44120, "step": 4607} +{"lr": 9.681645625952616e-05, "data_time": 0.0013948917388916016, "loss": 0.37863059639930724, "time": 0.6311232566833496, "epoch": 2, "memory": 44120, "step": 4707} +{"lr": 9.668142694109826e-05, "data_time": 0.0016169309616088866, "loss": 0.3920276552438736, "time": 0.6308682680130004, "epoch": 2, "memory": 44120, "step": 4807} +{"lr": 9.654370265747162e-05, "data_time": 0.001408863067626953, "loss": 0.3837639331817627, "time": 0.6317102670669555, "epoch": 2, "memory": 44120, "step": 4907} +{"lr": 9.640329231338345e-05, "data_time": 0.001631021499633789, "loss": 0.38010249733924867, "time": 0.6313638925552368, "epoch": 2, "memory": 44120, "step": 5007} +{"lr": 9.626020498724147e-05, "data_time": 0.0014079093933105468, "loss": 0.38657009303569795, "time": 0.6307583570480346, "epoch": 2, "memory": 44120, "step": 5107} +{"lr": 9.611444993053717e-05, "data_time": 0.001447010040283203, "loss": 0.3812067717313766, "time": 0.6313483238220214, "epoch": 2, "memory": 44120, "step": 5207} +{"lr": 9.596603656724746e-05, "data_time": 0.00148773193359375, "loss": 0.37799614667892456, "time": 0.6316648721694946, "epoch": 2, "memory": 44120, "step": 5307} +{"lr": 9.58149744932254e-05, "data_time": 0.0013596057891845704, "loss": 0.3787410736083984, "time": 0.6309352874755859, "epoch": 2, "memory": 44120, "step": 5407} +{"lr": 9.566127347557972e-05, "data_time": 0.0014389753341674805, "loss": 0.3914622485637665, "time": 0.6321154356002807, "epoch": 2, "memory": 44120, "step": 5507} +{"lr": 9.550494345204367e-05, "data_time": 0.001526641845703125, "loss": 0.3773065090179443, "time": 0.6316898584365844, "epoch": 2, "memory": 44120, "step": 5607} +{"lr": 9.534599453033183e-05, "data_time": 0.0015011310577392578, "loss": 0.3765223801136017, "time": 0.6306300640106202, "epoch": 2, "memory": 44120, "step": 5707} +{"lr": 9.518443698748712e-05, "data_time": 0.001419973373413086, "loss": 0.377192884683609, "time": 0.6458696842193603, "epoch": 2, "memory": 44120, "step": 5807} +{"lr": 9.50202812692162e-05, "data_time": 0.0016075849533081054, "loss": 0.39714027345180514, "time": 0.6313439846038819, "epoch": 2, "memory": 44120, "step": 5907} +{"lr": 9.485353798921386e-05, "data_time": 0.001783299446105957, "loss": 0.38935945332050326, "time": 0.6316739320755005, "epoch": 2, "memory": 44120, "step": 6007} +{"lr": 9.468421792847712e-05, "data_time": 0.0013911962509155274, "loss": 0.37018243670463563, "time": 0.6333726406097412, "epoch": 2, "memory": 44120, "step": 6107} +{"lr": 9.451233203460794e-05, "data_time": 0.0014320135116577149, "loss": 0.374513179063797, "time": 0.6307969331741333, "epoch": 2, "memory": 44120, "step": 6207} +{"lr": 9.433789142110532e-05, "data_time": 0.0014647960662841797, "loss": 0.34460248947143557, "time": 0.6348870038986206, "epoch": 2, "memory": 44120, "step": 6307} +{"lr": 9.416090736664709e-05, "data_time": 0.0014971494674682617, "loss": 0.3948313921689987, "time": 0.6313304901123047, "epoch": 2, "memory": 44120, "step": 6407} +{"lr": 9.398139131436032e-05, "data_time": 0.0014452457427978516, "loss": 0.37769335210323335, "time": 0.6306396484375, "epoch": 2, "memory": 44120, "step": 6507} +{"lr": 9.379935487108167e-05, "data_time": 0.0014565229415893556, "loss": 0.3787588536739349, "time": 0.6308593034744263, "epoch": 2, "memory": 44120, "step": 6607} +{"lr": 9.361480980660697e-05, "data_time": 0.0013815402984619141, "loss": 0.3671922743320465, "time": 0.6302517652511597, "epoch": 2, "memory": 44120, "step": 6707} +{"lr": 9.342776805292993e-05, "data_time": 0.0014109611511230469, "loss": 0.38747328519821167, "time": 0.6315154552459716, "epoch": 2, "memory": 44120, "step": 6807} +{"lr": 9.323824170347083e-05, "data_time": 0.001415562629699707, "loss": 0.3590896546840668, "time": 0.6315407037734986, "epoch": 2, "memory": 44120, "step": 6907} +{"lr": 9.304624301229471e-05, "data_time": 0.0016486167907714844, "loss": 0.3561199963092804, "time": 0.6330701351165772, "epoch": 2, "memory": 44120, "step": 7007} +{"lr": 9.285178439331897e-05, "data_time": 0.0016980648040771484, "loss": 0.37257237136363985, "time": 0.6327742576599121, "epoch": 2, "memory": 44120, "step": 7107} +{"lr": 9.265487841951087e-05, "data_time": 0.0015543222427368164, "loss": 0.3504183143377304, "time": 0.6309172153472901, "epoch": 2, "memory": 44120, "step": 7207} +{"lr": 9.245553782207421e-05, "data_time": 0.0015695333480834962, "loss": 0.34269999563694, "time": 0.6304365396499634, "epoch": 2, "memory": 44120, "step": 7307} +{"lr": 9.225377548962671e-05, "data_time": 0.0013744831085205078, "loss": 0.36070674657821655, "time": 0.631167221069336, "epoch": 2, "memory": 44120, "step": 7407} +{"lr": 9.204960446736601e-05, "data_time": 0.001514744758605957, "loss": 0.36700805723667146, "time": 0.631222915649414, "epoch": 2, "memory": 44120, "step": 7507} +{"lr": 9.184303795622715e-05, "data_time": 0.0014163494110107423, "loss": 0.3641028881072998, "time": 0.6312353372573852, "epoch": 2, "memory": 44120, "step": 7607} +{"lr": 9.163408931202823e-05, "data_time": 0.001429128646850586, "loss": 0.3695355117321014, "time": 0.631613826751709, "epoch": 2, "memory": 44120, "step": 7707} +{"lr": 9.142277204460729e-05, "data_time": 0.001242852210998535, "loss": 0.3687566339969635, "time": 0.6312731742858887, "epoch": 2, "memory": 44120, "step": 7807} +{"accuracy/top1": 77.84349822998047, "data_time": 0.036329022769270275, "time": 1.278838428957709, "step": 2} +{"lr": 9.119405492383288e-05, "data_time": 0.0014481067657470704, "loss": 0.3847730875015259, "time": 0.6318252086639404, "epoch": 3, "memory": 44120, "step": 7914} +{"lr": 9.097787819233205e-05, "data_time": 0.001453399658203125, "loss": 0.40148198008537295, "time": 0.6320568799972535, "epoch": 3, "memory": 44120, "step": 8014} +{"lr": 9.075937526576627e-05, "data_time": 0.0014786481857299804, "loss": 0.38979090452194215, "time": 0.6316263675689697, "epoch": 3, "memory": 44120, "step": 8114} +{"lr": 9.053856027171755e-05, "data_time": 0.0015566349029541016, "loss": 0.3463983088731766, "time": 0.6335758686065673, "epoch": 3, "memory": 44120, "step": 8214} +{"lr": 9.031544748725785e-05, "data_time": 0.0023714542388916016, "loss": 0.36977761089801786, "time": 0.6326429605484009, "epoch": 3, "memory": 44120, "step": 8314} +{"lr": 9.009005133802542e-05, "data_time": 0.0015105724334716797, "loss": 0.3708623141050339, "time": 0.6312785625457764, "epoch": 3, "memory": 44120, "step": 8414} +{"lr": 8.986238639729242e-05, "data_time": 0.0014684200286865234, "loss": 0.39234643876552583, "time": 0.6314328193664551, "epoch": 3, "memory": 44120, "step": 8514} +{"lr": 8.963246738502254e-05, "data_time": 0.0015577077865600586, "loss": 0.3743011951446533, "time": 0.6321437358856201, "epoch": 3, "memory": 44120, "step": 8614} +{"lr": 8.940030916691928e-05, "data_time": 0.001475071907043457, "loss": 0.3778077602386475, "time": 0.6326121807098388, "epoch": 3, "memory": 44120, "step": 8714} +{"lr": 8.91659267534649e-05, "data_time": 0.0013529062271118164, "loss": 0.3694726645946503, "time": 0.6318154335021973, "epoch": 3, "memory": 44120, "step": 8814} +{"lr": 8.892933529894978e-05, "data_time": 0.0015254497528076171, "loss": 0.36761326491832735, "time": 0.6327833890914917, "epoch": 3, "memory": 44120, "step": 8914} +{"lr": 8.869055010049283e-05, "data_time": 0.001403641700744629, "loss": 0.38764975368976595, "time": 0.6352396965026855, "epoch": 3, "memory": 44120, "step": 9014} +{"lr": 8.844958659705188e-05, "data_time": 0.0014962196350097657, "loss": 0.3604812115430832, "time": 0.632267951965332, "epoch": 3, "memory": 44120, "step": 9114} +{"lr": 8.82064603684259e-05, "data_time": 0.001603245735168457, "loss": 0.37507286071777346, "time": 0.6401543617248535, "epoch": 3, "memory": 44120, "step": 9214} +{"lr": 8.796118713424776e-05, "data_time": 0.0015782594680786132, "loss": 0.37148820161819457, "time": 0.6333001375198364, "epoch": 3, "memory": 44120, "step": 9314} +{"lr": 8.77137827529676e-05, "data_time": 0.0014742612838745117, "loss": 0.3671196222305298, "time": 0.6341067314147949, "epoch": 3, "memory": 44120, "step": 9414} +{"lr": 8.746426322082745e-05, "data_time": 0.0015903234481811524, "loss": 0.3711984634399414, "time": 0.6319302320480347, "epoch": 3, "memory": 44120, "step": 9514} +{"lr": 8.721264467082732e-05, "data_time": 0.0015094518661499024, "loss": 0.3907595783472061, "time": 0.6322148084640503, "epoch": 3, "memory": 44120, "step": 9614} +{"lr": 8.695894337168168e-05, "data_time": 0.001618647575378418, "loss": 0.362550875544548, "time": 0.6325819492340088, "epoch": 3, "memory": 44120, "step": 9714} +{"lr": 8.67031757267677e-05, "data_time": 0.0016858577728271484, "loss": 0.36691901087760925, "time": 0.6320786714553833, "epoch": 3, "memory": 44120, "step": 9814} +{"lr": 8.64453582730651e-05, "data_time": 0.0014262199401855469, "loss": 0.34382962584495547, "time": 0.6324646711349488, "epoch": 3, "memory": 44120, "step": 9914} +{"lr": 8.61855076800861e-05, "data_time": 0.0014376401901245116, "loss": 0.3303049474954605, "time": 0.6322244167327881, "epoch": 3, "memory": 44120, "step": 10014} +{"lr": 8.592364074879843e-05, "data_time": 0.001498246192932129, "loss": 0.3917563557624817, "time": 0.6324445724487304, "epoch": 3, "memory": 44120, "step": 10114} +{"lr": 8.565977441053854e-05, "data_time": 0.0013967514038085937, "loss": 0.3499339759349823, "time": 0.6343879699707031, "epoch": 3, "memory": 44120, "step": 10214} +{"lr": 8.539392572591724e-05, "data_time": 0.0013753652572631836, "loss": 0.3634859323501587, "time": 0.6313601016998291, "epoch": 3, "memory": 44120, "step": 10314} +{"lr": 8.512611188371633e-05, "data_time": 0.001399397850036621, "loss": 0.37576409578323366, "time": 0.6319352149963379, "epoch": 3, "memory": 44120, "step": 10414} +{"lr": 8.485635019977725e-05, "data_time": 0.0013954639434814453, "loss": 0.35563134849071504, "time": 0.6318157196044922, "epoch": 3, "memory": 44120, "step": 10514} +{"lr": 8.45846581158818e-05, "data_time": 0.001403212547302246, "loss": 0.3612090051174164, "time": 0.6316814422607422, "epoch": 3, "memory": 44120, "step": 10614} +{"lr": 8.431105319862416e-05, "data_time": 0.0014333009719848632, "loss": 0.3662464290857315, "time": 0.6306448698043823, "epoch": 3, "memory": 44120, "step": 10714} +{"lr": 8.403555313827516e-05, "data_time": 0.001498270034790039, "loss": 0.38241981863975527, "time": 0.6314601182937623, "epoch": 3, "memory": 44120, "step": 10814} +{"lr": 8.375817574763844e-05, "data_time": 0.0013619184494018555, "loss": 0.37417236268520354, "time": 0.6317417860031128, "epoch": 3, "memory": 44120, "step": 10914} +{"lr": 8.347893896089871e-05, "data_time": 0.0014081716537475586, "loss": 0.33988997936248777, "time": 0.63100106716156, "epoch": 3, "memory": 44120, "step": 11014} +{"lr": 8.31978608324625e-05, "data_time": 0.001481938362121582, "loss": 0.38768957257270814, "time": 0.6333823680877686, "epoch": 3, "memory": 44120, "step": 11114} +{"lr": 8.291495953579052e-05, "data_time": 0.0012996912002563477, "loss": 0.342476087808609, "time": 0.6363723754882813, "epoch": 3, "memory": 44120, "step": 11214} +{"lr": 8.263025336222267e-05, "data_time": 0.001495051383972168, "loss": 0.37864493429660795, "time": 0.6331802368164062, "epoch": 3, "memory": 44120, "step": 11314} +{"lr": 8.234376071979551e-05, "data_time": 0.0014034032821655274, "loss": 0.37599709033966067, "time": 0.6316048860549927, "epoch": 3, "memory": 44120, "step": 11414} +{"lr": 8.205550013205187e-05, "data_time": 0.0015259027481079102, "loss": 0.3576871693134308, "time": 0.6314178228378295, "epoch": 3, "memory": 44120, "step": 11514} +{"lr": 8.17654902368435e-05, "data_time": 0.001470804214477539, "loss": 0.3580753803253174, "time": 0.6342048406600952, "epoch": 3, "memory": 44120, "step": 11614} +{"lr": 8.147374978512582e-05, "data_time": 0.0012356758117675782, "loss": 0.3363444447517395, "time": 0.6309850215911865, "epoch": 3, "memory": 44120, "step": 11714} +{"accuracy/top1": 77.0815200805664, "data_time": 0.03565501344615016, "time": 1.2774384145078987, "step": 3} +{"lr": 8.115969237626578e-05, "data_time": 0.0014601707458496093, "loss": 0.3616588294506073, "time": 0.6327504396438599, "epoch": 4, "memory": 44120, "step": 11821} +{"lr": 8.08644297337245e-05, "data_time": 0.0013072729110717774, "loss": 0.3634976029396057, "time": 0.6361582517623902, "epoch": 4, "memory": 44120, "step": 11921} +{"lr": 8.056749479387651e-05, "data_time": 0.001401519775390625, "loss": 0.36289515495300295, "time": 0.6348573923110962, "epoch": 4, "memory": 44120, "step": 12021} +{"lr": 8.026890675542457e-05, "data_time": 0.0018458604812622071, "loss": 0.3610669583082199, "time": 0.6331987380981445, "epoch": 4, "memory": 44120, "step": 12121} +{"lr": 7.996868492395457e-05, "data_time": 0.0014137506484985351, "loss": 0.35711357891559603, "time": 0.6319075345993042, "epoch": 4, "memory": 44120, "step": 12221} +{"lr": 7.966684871068762e-05, "data_time": 0.0015436172485351562, "loss": 0.37489054501056673, "time": 0.6419233560562134, "epoch": 4, "memory": 44120, "step": 12321} +{"lr": 7.936341763122449e-05, "data_time": 0.0014515399932861328, "loss": 0.366738298535347, "time": 0.6329026937484741, "epoch": 4, "memory": 44120, "step": 12421} +{"lr": 7.905841130428405e-05, "data_time": 0.0014167308807373046, "loss": 0.3751838505268097, "time": 0.6308252334594726, "epoch": 4, "memory": 44120, "step": 12521} +{"lr": 7.875184945043487e-05, "data_time": 0.0014575004577636718, "loss": 0.3581499129533768, "time": 0.6374243974685669, "epoch": 4, "memory": 44120, "step": 12621} +{"lr": 7.844375189082012e-05, "data_time": 0.0013425350189208984, "loss": 0.36496379375457766, "time": 0.6331381559371948, "epoch": 4, "memory": 44120, "step": 12721} +{"lr": 7.813413854587572e-05, "data_time": 0.0014866828918457032, "loss": 0.35153523087501526, "time": 0.6354534387588501, "epoch": 4, "memory": 44120, "step": 12821} +{"lr": 7.782302943404291e-05, "data_time": 0.0014543771743774415, "loss": 0.3674694120883942, "time": 0.6352455615997314, "epoch": 4, "memory": 44120, "step": 12921} +{"lr": 7.751044467047337e-05, "data_time": 0.001568603515625, "loss": 0.3743774712085724, "time": 0.6339632272720337, "epoch": 4, "memory": 44120, "step": 13021} +{"lr": 7.719640446572907e-05, "data_time": 0.0015260696411132813, "loss": 0.3665180325508118, "time": 0.6332921743392944, "epoch": 4, "memory": 44120, "step": 13121} +{"lr": 7.688092912447536e-05, "data_time": 0.0016004562377929688, "loss": 0.3684597373008728, "time": 0.6426959276199341, "epoch": 4, "memory": 44120, "step": 13221} +{"lr": 7.656403904416794e-05, "data_time": 0.00150909423828125, "loss": 0.37329218685626986, "time": 0.6346177339553833, "epoch": 4, "memory": 44120, "step": 13321} +{"lr": 7.624575471373434e-05, "data_time": 0.0014724493026733398, "loss": 0.37726982235908507, "time": 0.6342240333557129, "epoch": 4, "memory": 44120, "step": 13421} +{"lr": 7.592609671224926e-05, "data_time": 0.0014622211456298828, "loss": 0.33714719116687775, "time": 0.6331625699996948, "epoch": 4, "memory": 44120, "step": 13521} +{"lr": 7.560508570760375e-05, "data_time": 0.001430368423461914, "loss": 0.3499594509601593, "time": 0.6319095849990845, "epoch": 4, "memory": 44120, "step": 13621} +{"lr": 7.528274245516895e-05, "data_time": 0.0015727758407592773, "loss": 0.3525034934282303, "time": 0.6307008028030395, "epoch": 4, "memory": 44120, "step": 13721} +{"lr": 7.495908779645405e-05, "data_time": 0.0014667510986328125, "loss": 0.3675632268190384, "time": 0.6310592651367187, "epoch": 4, "memory": 44120, "step": 13821} +{"lr": 7.463414265775906e-05, "data_time": 0.0015061378479003906, "loss": 0.3848166584968567, "time": 0.6312855243682861, "epoch": 4, "memory": 44120, "step": 13921} +{"lr": 7.430792804882147e-05, "data_time": 0.0015873908996582031, "loss": 0.3621916949748993, "time": 0.6312375068664551, "epoch": 4, "memory": 44120, "step": 14021} +{"lr": 7.398046506145803e-05, "data_time": 0.0015889644622802735, "loss": 0.37212550044059756, "time": 0.6326715230941773, "epoch": 4, "memory": 44120, "step": 14121} +{"lr": 7.365177486820114e-05, "data_time": 0.0016717195510864257, "loss": 0.37732054591178893, "time": 0.631341814994812, "epoch": 4, "memory": 44120, "step": 14221} +{"lr": 7.332187872092939e-05, "data_time": 0.0014044284820556641, "loss": 0.3810142785310745, "time": 0.6327552795410156, "epoch": 4, "memory": 44120, "step": 14321} +{"lr": 7.299079794949401e-05, "data_time": 0.001490950584411621, "loss": 0.3755857527256012, "time": 0.6313372135162354, "epoch": 4, "memory": 44120, "step": 14421} +{"lr": 7.265855396033955e-05, "data_time": 0.0014354228973388673, "loss": 0.39156965613365174, "time": 0.6319684267044068, "epoch": 4, "memory": 44120, "step": 14521} +{"lr": 7.232516823512013e-05, "data_time": 0.0015784740447998048, "loss": 0.37674275040626526, "time": 0.6314492702484131, "epoch": 4, "memory": 44120, "step": 14621} +{"lr": 7.199066232930995e-05, "data_time": 0.0014206171035766602, "loss": 0.3549411177635193, "time": 0.6332176208496094, "epoch": 4, "memory": 44120, "step": 14721} +{"lr": 7.165505787080997e-05, "data_time": 0.0014365434646606446, "loss": 0.3555349886417389, "time": 0.6344083309173584, "epoch": 4, "memory": 44120, "step": 14821} +{"lr": 7.13183765585495e-05, "data_time": 0.001679372787475586, "loss": 0.3396206796169281, "time": 0.6323187112808227, "epoch": 4, "memory": 44120, "step": 14921} +{"lr": 7.098064016108317e-05, "data_time": 0.001564335823059082, "loss": 0.3686327874660492, "time": 0.6322042226791382, "epoch": 4, "memory": 44120, "step": 15021} +{"lr": 7.064187051518379e-05, "data_time": 0.0013367652893066407, "loss": 0.3623316496610641, "time": 0.633063554763794, "epoch": 4, "memory": 44120, "step": 15121} +{"lr": 7.030208952442949e-05, "data_time": 0.0014116287231445313, "loss": 0.37482787668704987, "time": 0.6356353282928466, "epoch": 4, "memory": 44120, "step": 15221} +{"lr": 6.996131915778851e-05, "data_time": 0.001506638526916504, "loss": 0.35330727100372317, "time": 0.640656852722168, "epoch": 4, "memory": 44120, "step": 15321} +{"lr": 6.961958144819853e-05, "data_time": 0.002330636978149414, "loss": 0.36677582561969757, "time": 0.6364585876464843, "epoch": 4, "memory": 44120, "step": 15421} +{"lr": 6.927689849114193e-05, "data_time": 0.001501750946044922, "loss": 0.3694306015968323, "time": 0.6313318729400634, "epoch": 4, "memory": 44120, "step": 15521} +{"lr": 6.893329244321722e-05, "data_time": 0.001215219497680664, "loss": 0.3838209182024002, "time": 0.632353687286377, "epoch": 4, "memory": 44120, "step": 15621} +{"accuracy/top1": 76.9040756225586, "data_time": 0.03608762395792994, "time": 1.2779335194620594, "step": 4} +{"lr": 6.856463687361792e-05, "data_time": 0.0017266035079956054, "loss": 0.37894098162651063, "time": 0.632786226272583, "epoch": 5, "memory": 44120, "step": 15728} +{"lr": 6.821919068470432e-05, "data_time": 0.001751112937927246, "loss": 0.368763393163681, "time": 0.6366863489151001, "epoch": 5, "memory": 44120, "step": 15828} +{"lr": 6.787288979235407e-05, "data_time": 0.0018464088439941405, "loss": 0.3627344459295273, "time": 0.6328916549682617, "epoch": 5, "memory": 44120, "step": 15928} +{"lr": 6.752575658708793e-05, "data_time": 0.0017342329025268554, "loss": 0.360625758767128, "time": 0.6354114294052124, "epoch": 5, "memory": 44120, "step": 16028} +{"lr": 6.71778135132408e-05, "data_time": 0.0020811080932617186, "loss": 0.3682739704847336, "time": 0.6326449871063232, "epoch": 5, "memory": 44120, "step": 16128} +{"lr": 6.682908306751074e-05, "data_time": 0.0016627311706542969, "loss": 0.37629181444644927, "time": 0.6309634208679199, "epoch": 5, "memory": 44120, "step": 16228} +{"lr": 6.64795877975043e-05, "data_time": 0.001546168327331543, "loss": 0.35409157872200014, "time": 0.6317749977111816, "epoch": 5, "memory": 44120, "step": 16328} +{"lr": 6.61293503002787e-05, "data_time": 0.0018603801727294922, "loss": 0.3660582721233368, "time": 0.6322298049926758, "epoch": 5, "memory": 44120, "step": 16428} +{"lr": 6.577839322088084e-05, "data_time": 0.0014491796493530274, "loss": 0.359283447265625, "time": 0.6329627513885498, "epoch": 5, "memory": 44120, "step": 16528} +{"lr": 6.542673925088298e-05, "data_time": 0.0015192031860351562, "loss": 0.35167465209960935, "time": 0.6328529834747314, "epoch": 5, "memory": 44120, "step": 16628} +{"lr": 6.507441112691584e-05, "data_time": 0.0018984317779541016, "loss": 0.35347725749015807, "time": 0.6334426164627075, "epoch": 5, "memory": 44120, "step": 16728} +{"lr": 6.47214316291983e-05, "data_time": 0.0013723373413085938, "loss": 0.3700693666934967, "time": 0.631100058555603, "epoch": 5, "memory": 44120, "step": 16828} +{"lr": 6.436782358006481e-05, "data_time": 0.0015852212905883788, "loss": 0.3630936801433563, "time": 0.6312743902206421, "epoch": 5, "memory": 44120, "step": 16928} +{"lr": 6.40136098424896e-05, "data_time": 0.0014849424362182618, "loss": 0.36088619828224183, "time": 0.6330517292022705, "epoch": 5, "memory": 44120, "step": 17028} +{"lr": 6.365881331860828e-05, "data_time": 0.0015062570571899414, "loss": 0.37771726250648496, "time": 0.6318070411682128, "epoch": 5, "memory": 44120, "step": 17128} +{"lr": 6.330345694823739e-05, "data_time": 0.00149993896484375, "loss": 0.3599252313375473, "time": 0.6330976247787475, "epoch": 5, "memory": 44120, "step": 17228} +{"lr": 6.29475637073908e-05, "data_time": 0.0015309810638427734, "loss": 0.38419176936149596, "time": 0.6328319549560547, "epoch": 5, "memory": 44120, "step": 17328} +{"lr": 6.259115660679482e-05, "data_time": 0.0016766786575317383, "loss": 0.3584842085838318, "time": 0.6327142477035522, "epoch": 5, "memory": 44120, "step": 17428} +{"lr": 6.223425869039963e-05, "data_time": 0.0013804674148559571, "loss": 0.3820468544960022, "time": 0.6319448232650757, "epoch": 5, "memory": 44120, "step": 17528} +{"lr": 6.187689303389006e-05, "data_time": 0.0014548301696777344, "loss": 0.3519988000392914, "time": 0.6326465129852294, "epoch": 5, "memory": 44120, "step": 17628} +{"lr": 6.151908274319292e-05, "data_time": 0.0014504432678222657, "loss": 0.36450506150722506, "time": 0.6362475395202637, "epoch": 5, "memory": 44120, "step": 17728} +{"lr": 6.116085095298373e-05, "data_time": 0.0015873193740844726, "loss": 0.3727584004402161, "time": 0.6347034454345704, "epoch": 5, "memory": 44120, "step": 17828} +{"lr": 6.080222082519007e-05, "data_time": 0.0016533374786376954, "loss": 0.33923091292381286, "time": 0.632828950881958, "epoch": 5, "memory": 44120, "step": 17928} +{"lr": 6.0443215547495147e-05, "data_time": 0.001558065414428711, "loss": 0.35954704582691194, "time": 0.6335270404815674, "epoch": 5, "memory": 44120, "step": 18028} +{"lr": 6.0083858331837714e-05, "data_time": 0.001592254638671875, "loss": 0.3793952286243439, "time": 0.6341854333877563, "epoch": 5, "memory": 44120, "step": 18128} +{"lr": 5.97241724129116e-05, "data_time": 0.001589202880859375, "loss": 0.3658044457435608, "time": 0.63597412109375, "epoch": 5, "memory": 44120, "step": 18228} +{"lr": 5.936418104666328e-05, "data_time": 0.001482391357421875, "loss": 0.3590157270431519, "time": 0.6327776670455932, "epoch": 5, "memory": 44120, "step": 18328} +{"lr": 5.9003907508788295e-05, "data_time": 0.0015847921371459962, "loss": 0.3516092628240585, "time": 0.635569429397583, "epoch": 5, "memory": 44120, "step": 18428} +{"lr": 5.864337509322633e-05, "data_time": 0.0013956546783447266, "loss": 0.35428989231586455, "time": 0.6349320888519288, "epoch": 5, "memory": 44120, "step": 18528} +{"lr": 5.8282607110655194e-05, "data_time": 0.0016501426696777343, "loss": 0.3838335871696472, "time": 0.6330335855484008, "epoch": 5, "memory": 44120, "step": 18628} +{"lr": 5.7921626886983564e-05, "data_time": 0.0015996456146240234, "loss": 0.3755250334739685, "time": 0.6326020956039429, "epoch": 5, "memory": 44120, "step": 18728} +{"lr": 5.756045776184287e-05, "data_time": 0.0014918327331542968, "loss": 0.34004478752613065, "time": 0.6322643518447876, "epoch": 5, "memory": 44120, "step": 18828} +{"lr": 5.719912308707808e-05, "data_time": 0.001599264144897461, "loss": 0.3651239424943924, "time": 0.6325971603393554, "epoch": 5, "memory": 44120, "step": 18928} +{"lr": 5.683764622523808e-05, "data_time": 0.0015192031860351562, "loss": 0.35102906823158264, "time": 0.6324996232986451, "epoch": 5, "memory": 44120, "step": 19028} +{"lr": 5.647605054806502e-05, "data_time": 0.0014725923538208008, "loss": 0.3628747522830963, "time": 0.6368881225585937, "epoch": 5, "memory": 44120, "step": 19128} +{"lr": 5.6114359434983206e-05, "data_time": 0.0014778614044189454, "loss": 0.374749031662941, "time": 0.634670615196228, "epoch": 5, "memory": 44120, "step": 19228} +{"lr": 5.575259627158745e-05, "data_time": 0.0015050411224365235, "loss": 0.3750637054443359, "time": 0.6319854497909546, "epoch": 5, "memory": 44120, "step": 19328} +{"lr": 5.539078444813113e-05, "data_time": 0.0015236377716064454, "loss": 0.3660961240530014, "time": 0.6349586963653564, "epoch": 5, "memory": 44120, "step": 19428} +{"lr": 5.5028947358013735e-05, "data_time": 0.0014175653457641601, "loss": 0.3604292690753937, "time": 0.6347747325897217, "epoch": 5, "memory": 44120, "step": 19528} +{"accuracy/top1": 76.50743103027344, "data_time": 0.03666051091818974, "time": 1.2800816831917599, "step": 5} +{"lr": 5.464178020340108e-05, "data_time": 0.001437807083129883, "loss": 0.3572618395090103, "time": 0.632691502571106, "epoch": 6, "memory": 44120, "step": 19635} +{"lr": 5.427996514795046e-05, "data_time": 0.0016089200973510743, "loss": 0.37430056631565095, "time": 0.6319503307342529, "epoch": 6, "memory": 44120, "step": 19735} +{"lr": 5.3918196647260834e-05, "data_time": 0.0014183759689331056, "loss": 0.3624613046646118, "time": 0.6320508241653442, "epoch": 6, "memory": 44120, "step": 19835} +{"lr": 5.355649809193057e-05, "data_time": 0.001658797264099121, "loss": 0.3757551610469818, "time": 0.6331546783447266, "epoch": 6, "memory": 44120, "step": 19935} +{"lr": 5.3194892868035765e-05, "data_time": 0.001749873161315918, "loss": 0.35926517844200134, "time": 0.6371479272842407, "epoch": 6, "memory": 44120, "step": 20035} +{"lr": 5.283340435561795e-05, "data_time": 0.001473379135131836, "loss": 0.3537742137908936, "time": 0.6342557907104492, "epoch": 6, "memory": 44120, "step": 20135} +{"lr": 5.2472055927172495e-05, "data_time": 0.0015133380889892577, "loss": 0.3648320436477661, "time": 0.6345050811767579, "epoch": 6, "memory": 44120, "step": 20235} +{"lr": 5.2110870946137605e-05, "data_time": 0.0014802932739257813, "loss": 0.3618503898382187, "time": 0.6313145160675049, "epoch": 6, "memory": 44120, "step": 20335} +{"lr": 5.174987276538342e-05, "data_time": 0.0014394283294677734, "loss": 0.3696357190608978, "time": 0.632557487487793, "epoch": 6, "memory": 44120, "step": 20435} +{"lr": 5.1389084725702425e-05, "data_time": 0.001480436325073242, "loss": 0.3459394246339798, "time": 0.633821439743042, "epoch": 6, "memory": 44120, "step": 20535} +{"lr": 5.1028530154300045e-05, "data_time": 0.001661992073059082, "loss": 0.35268403589725494, "time": 0.6348913192749024, "epoch": 6, "memory": 44120, "step": 20635} +{"lr": 5.066823236328663e-05, "data_time": 0.0018307924270629882, "loss": 0.37245126664638517, "time": 0.6342251062393188, "epoch": 6, "memory": 44120, "step": 20735} +{"lr": 5.030821464817006e-05, "data_time": 0.0014332294464111327, "loss": 0.3455190181732178, "time": 0.632005262374878, "epoch": 6, "memory": 44120, "step": 20835} +{"lr": 4.99485002863494e-05, "data_time": 0.0014160871505737305, "loss": 0.3423906207084656, "time": 0.633011770248413, "epoch": 6, "memory": 44120, "step": 20935} +{"lr": 4.95891125356102e-05, "data_time": 0.0015711784362792969, "loss": 0.3789345592260361, "time": 0.6328719139099122, "epoch": 6, "memory": 44120, "step": 21035} +{"lr": 4.923007463262053e-05, "data_time": 0.0015725135803222657, "loss": 0.36813676357269287, "time": 0.6345274686813355, "epoch": 6, "memory": 44120, "step": 21135} +{"lr": 4.887140979142852e-05, "data_time": 0.0016031265258789062, "loss": 0.34439248144626616, "time": 0.6335875034332276, "epoch": 6, "memory": 44120, "step": 21235} +{"lr": 4.851314120196161e-05, "data_time": 0.0016109466552734375, "loss": 0.37726256549358367, "time": 0.6330404043197632, "epoch": 6, "memory": 44120, "step": 21335} +{"lr": 4.815529202852693e-05, "data_time": 0.0014396905899047852, "loss": 0.3832100421190262, "time": 0.6328764677047729, "epoch": 6, "memory": 44120, "step": 21435} +{"lr": 4.779788540831406e-05, "data_time": 0.0016587257385253906, "loss": 0.3852983593940735, "time": 0.6330849647521972, "epoch": 6, "memory": 44120, "step": 21535} +{"lr": 4.744094444989843e-05, "data_time": 0.0016937494277954102, "loss": 0.36328046321868895, "time": 0.6359888553619385, "epoch": 6, "memory": 44120, "step": 21635} +{"lr": 4.708449223174764e-05, "data_time": 0.001470804214477539, "loss": 0.37881985306739807, "time": 0.633746862411499, "epoch": 6, "memory": 44120, "step": 21735} +{"lr": 4.672855180072916e-05, "data_time": 0.001506209373474121, "loss": 0.3463868647813797, "time": 0.6335466623306274, "epoch": 6, "memory": 44120, "step": 21835} +{"lr": 4.6373146170620256e-05, "data_time": 0.0015089035034179688, "loss": 0.36230126619338987, "time": 0.6332391977310181, "epoch": 6, "memory": 44120, "step": 21935} +{"lr": 4.6018298320619915e-05, "data_time": 0.0013457298278808593, "loss": 0.3581427901983261, "time": 0.6315054416656494, "epoch": 6, "memory": 44120, "step": 22035} +{"lr": 4.566403119386318e-05, "data_time": 0.0014329671859741211, "loss": 0.36872154772281646, "time": 0.6338589429855347, "epoch": 6, "memory": 44120, "step": 22135} +{"lr": 4.531036769593769e-05, "data_time": 0.0014895200729370117, "loss": 0.3592005521059036, "time": 0.6327705383300781, "epoch": 6, "memory": 44120, "step": 22235} +{"lr": 4.4957330693402685e-05, "data_time": 0.001523733139038086, "loss": 0.34864698350429535, "time": 0.6323814392089844, "epoch": 6, "memory": 44120, "step": 22335} +{"lr": 4.460494301231059e-05, "data_time": 0.001468992233276367, "loss": 0.3588447093963623, "time": 0.6320381879806518, "epoch": 6, "memory": 44120, "step": 22435} +{"lr": 4.425322743673109e-05, "data_time": 0.0015026330947875977, "loss": 0.3461219519376755, "time": 0.6336755514144897, "epoch": 6, "memory": 44120, "step": 22535} +{"lr": 4.390220670727812e-05, "data_time": 0.0014482498168945312, "loss": 0.3664976954460144, "time": 0.6330651044845581, "epoch": 6, "memory": 44120, "step": 22635} +{"lr": 4.3551903519639325e-05, "data_time": 0.0015913963317871094, "loss": 0.37305538058280946, "time": 0.6312130451202392, "epoch": 6, "memory": 44120, "step": 22735} +{"lr": 4.3202340523108894e-05, "data_time": 0.0014281749725341796, "loss": 0.36498062312602997, "time": 0.6307538270950317, "epoch": 6, "memory": 44120, "step": 22835} +{"lr": 4.285354031912302e-05, "data_time": 0.0016157388687133788, "loss": 0.36128232777118685, "time": 0.6321924924850464, "epoch": 6, "memory": 44120, "step": 22935} +{"lr": 4.2505525459798544e-05, "data_time": 0.0014148950576782227, "loss": 0.3610009491443634, "time": 0.6320352554321289, "epoch": 6, "memory": 44120, "step": 23035} +{"lr": 4.2158318446474836e-05, "data_time": 0.0014661073684692383, "loss": 0.3349186420440674, "time": 0.6319817304611206, "epoch": 6, "memory": 44120, "step": 23135} +{"lr": 4.181194172825905e-05, "data_time": 0.0016401052474975587, "loss": 0.3892023950815201, "time": 0.6348764657974243, "epoch": 6, "memory": 44120, "step": 23235} +{"lr": 4.146641770057448e-05, "data_time": 0.0015258073806762695, "loss": 0.3624118328094482, "time": 0.6314297199249268, "epoch": 6, "memory": 44120, "step": 23335} +{"lr": 4.112176870371267e-05, "data_time": 0.0012554168701171876, "loss": 0.35829317569732666, "time": 0.6330458402633667, "epoch": 6, "memory": 44120, "step": 23435} +{"accuracy/top1": 76.23603820800781, "data_time": 0.035895705223083496, "time": 1.2793013720676816, "step": 6} +{"lr": 4.0753988582732786e-05, "data_time": 0.0015159130096435546, "loss": 0.36123780608177186, "time": 0.6334625482559204, "epoch": 7, "memory": 44120, "step": 23542} +{"lr": 4.0411221639286815e-05, "data_time": 0.0013533830642700195, "loss": 0.3564772456884384, "time": 0.6335180282592774, "epoch": 7, "memory": 44120, "step": 23642} +{"lr": 4.006939795169518e-05, "data_time": 0.0015457868576049805, "loss": 0.3629548907279968, "time": 0.6356516361236573, "epoch": 7, "memory": 44120, "step": 23742} +{"lr": 3.972853962099913e-05, "data_time": 0.001765275001525879, "loss": 0.390253359079361, "time": 0.635542917251587, "epoch": 7, "memory": 44120, "step": 23842} +{"lr": 3.93886686858237e-05, "data_time": 0.0022885799407958984, "loss": 0.3424901574850082, "time": 0.6346344709396362, "epoch": 7, "memory": 44120, "step": 23942} +{"lr": 3.904980712095259e-05, "data_time": 0.0013433456420898437, "loss": 0.3503723829984665, "time": 0.6340196847915649, "epoch": 7, "memory": 44120, "step": 24042} +{"lr": 3.871197683590748e-05, "data_time": 0.0013974428176879883, "loss": 0.3757956862449646, "time": 0.6357422828674316, "epoch": 7, "memory": 44120, "step": 24142} +{"lr": 3.837519967353125e-05, "data_time": 0.0015516042709350585, "loss": 0.34941319525241854, "time": 0.6345579147338867, "epoch": 7, "memory": 44120, "step": 24242} +{"lr": 3.8039497408575855e-05, "data_time": 0.001452493667602539, "loss": 0.3754796594381332, "time": 0.6326464414596558, "epoch": 7, "memory": 44120, "step": 24342} +{"lr": 3.770489174629432e-05, "data_time": 0.0014765262603759766, "loss": 0.36980322003364563, "time": 0.6335984945297242, "epoch": 7, "memory": 44120, "step": 24442} +{"lr": 3.737140432103754e-05, "data_time": 0.0015181779861450195, "loss": 0.3552908211946487, "time": 0.634049916267395, "epoch": 7, "memory": 44120, "step": 24542} +{"lr": 3.703905669485545e-05, "data_time": 0.001567983627319336, "loss": 0.36363907158374786, "time": 0.6317353010177612, "epoch": 7, "memory": 44120, "step": 24642} +{"lr": 3.6707870356102706e-05, "data_time": 0.0017004013061523438, "loss": 0.35386523604393005, "time": 0.6334615230560303, "epoch": 7, "memory": 44120, "step": 24742} +{"lr": 3.6377866718049484e-05, "data_time": 0.0014855384826660157, "loss": 0.35835925936698915, "time": 0.6328764438629151, "epoch": 7, "memory": 44120, "step": 24842} +{"lr": 3.6049067117496965e-05, "data_time": 0.0016066074371337891, "loss": 0.3607875853776932, "time": 0.6357440948486328, "epoch": 7, "memory": 44120, "step": 24942} +{"lr": 3.5721492813397755e-05, "data_time": 0.001550602912902832, "loss": 0.3488959699869156, "time": 0.6320597410202027, "epoch": 7, "memory": 44120, "step": 25042} +{"lr": 3.539516498548151e-05, "data_time": 0.0013962984085083008, "loss": 0.3506907194852829, "time": 0.6320287227630615, "epoch": 7, "memory": 44120, "step": 25142} +{"lr": 3.507010473288518e-05, "data_time": 0.0015010356903076172, "loss": 0.35058234333992006, "time": 0.6400053262710571, "epoch": 7, "memory": 44120, "step": 25242} +{"lr": 3.474633307278909e-05, "data_time": 0.0014966964721679688, "loss": 0.36881363689899443, "time": 0.6330070734024048, "epoch": 7, "memory": 44120, "step": 25342} +{"lr": 3.4423870939058177e-05, "data_time": 0.0014649391174316405, "loss": 0.3741642564535141, "time": 0.6326273202896118, "epoch": 7, "memory": 44120, "step": 25442} +{"lr": 3.410273918088805e-05, "data_time": 0.0014224290847778321, "loss": 0.34705342054367067, "time": 0.6346636295318604, "epoch": 7, "memory": 44120, "step": 25542} +{"lr": 3.37829585614574e-05, "data_time": 0.0014030933380126953, "loss": 0.36638697385787966, "time": 0.6324271440505982, "epoch": 7, "memory": 44120, "step": 25642} +{"lr": 3.3464549756585216e-05, "data_time": 0.0014451503753662109, "loss": 0.3701684415340424, "time": 0.6325154781341553, "epoch": 7, "memory": 44120, "step": 25742} +{"lr": 3.314753335339416e-05, "data_time": 0.0013575315475463866, "loss": 0.3597903877496719, "time": 0.6339298725128174, "epoch": 7, "memory": 44120, "step": 25842} +{"lr": 3.2831929848979394e-05, "data_time": 0.0015400886535644532, "loss": 0.3604041874408722, "time": 0.6329383611679077, "epoch": 7, "memory": 44120, "step": 25942} +{"lr": 3.251775964908332e-05, "data_time": 0.0014547109603881836, "loss": 0.35140136778354647, "time": 0.6328030586242676, "epoch": 7, "memory": 44120, "step": 26042} +{"lr": 3.22050430667762e-05, "data_time": 0.0014820098876953125, "loss": 0.3374085336923599, "time": 0.6335373163223267, "epoch": 7, "memory": 44120, "step": 26142} +{"lr": 3.189380032114285e-05, "data_time": 0.0014576911926269531, "loss": 0.38166150748729705, "time": 0.634656023979187, "epoch": 7, "memory": 44120, "step": 26242} +{"lr": 3.1584051535975324e-05, "data_time": 0.0013989448547363282, "loss": 0.36053734123706815, "time": 0.633669090270996, "epoch": 7, "memory": 44120, "step": 26342} +{"lr": 3.127581673847175e-05, "data_time": 0.0015028715133666992, "loss": 0.386484768986702, "time": 0.6345329284667969, "epoch": 7, "memory": 44120, "step": 26442} +{"lr": 3.0969115857941555e-05, "data_time": 0.001757669448852539, "loss": 0.36978948414325713, "time": 0.6333515167236328, "epoch": 7, "memory": 44120, "step": 26542} +{"lr": 3.0663968724516654e-05, "data_time": 0.001509547233581543, "loss": 0.36772051453590393, "time": 0.6349242448806762, "epoch": 7, "memory": 44120, "step": 26642} +{"lr": 3.0360395067869668e-05, "data_time": 0.0015058517456054688, "loss": 0.3541785299777985, "time": 0.6327594518661499, "epoch": 7, "memory": 44120, "step": 26742} +{"lr": 3.0058414515938063e-05, "data_time": 0.0014070749282836914, "loss": 0.3812207102775574, "time": 0.6335640907287597, "epoch": 7, "memory": 44120, "step": 26842} +{"lr": 2.975804659365501e-05, "data_time": 0.001426529884338379, "loss": 0.36305170953273774, "time": 0.6338283777236938, "epoch": 7, "memory": 44120, "step": 26942} +{"lr": 2.945931072168714e-05, "data_time": 0.001401662826538086, "loss": 0.3763676553964615, "time": 0.6344704627990723, "epoch": 7, "memory": 44120, "step": 27042} +{"lr": 2.9162226215178855e-05, "data_time": 0.001462864875793457, "loss": 0.3519167512655258, "time": 0.6332732200622558, "epoch": 7, "memory": 44120, "step": 27142} +{"lr": 2.8866812282503322e-05, "data_time": 0.0015109777450561523, "loss": 0.3715108722448349, "time": 0.6373698472976684, "epoch": 7, "memory": 44120, "step": 27242} +{"lr": 2.8573088024020838e-05, "data_time": 0.0013585090637207031, "loss": 0.38868110477924345, "time": 0.631889820098877, "epoch": 7, "memory": 44120, "step": 27342} +{"accuracy/top1": 75.78024291992188, "data_time": 0.03545153552088244, "time": 1.2770294649847622, "step": 7} +{"lr": 2.8260695817390092e-05, "data_time": 0.0015898466110229493, "loss": 0.37303454279899595, "time": 0.6339924335479736, "epoch": 8, "memory": 44120, "step": 27449} +{"lr": 2.7970529402570967e-05, "data_time": 0.0015257596969604492, "loss": 0.35515395402908323, "time": 0.6342122316360473, "epoch": 8, "memory": 44120, "step": 27549} +{"lr": 2.7682110612244214e-05, "data_time": 0.001466989517211914, "loss": 0.3770622432231903, "time": 0.6325371742248536, "epoch": 8, "memory": 44120, "step": 27649} +{"lr": 2.7395458094490283e-05, "data_time": 0.0017443418502807618, "loss": 0.34302699863910674, "time": 0.6348450660705567, "epoch": 8, "memory": 44120, "step": 27749} +{"lr": 2.7110590383189015e-05, "data_time": 0.0014403343200683593, "loss": 0.393852236866951, "time": 0.6362323045730591, "epoch": 8, "memory": 44120, "step": 27849} +{"lr": 2.6827525896821397e-05, "data_time": 0.001551651954650879, "loss": 0.3731550931930542, "time": 0.6309094905853272, "epoch": 8, "memory": 44120, "step": 27949} +{"lr": 2.6546282937278552e-05, "data_time": 0.0016638517379760742, "loss": 0.3527612864971161, "time": 0.6348623752593994, "epoch": 8, "memory": 44120, "step": 28049} +{"lr": 2.6266879688678606e-05, "data_time": 0.0015616178512573241, "loss": 0.35313935577869415, "time": 0.6329471588134765, "epoch": 8, "memory": 44120, "step": 28149} +{"lr": 2.5989334216190738e-05, "data_time": 0.0015506982803344727, "loss": 0.37757874727249147, "time": 0.632073450088501, "epoch": 8, "memory": 44120, "step": 28249} +{"lr": 2.571366446486732e-05, "data_time": 0.0014993906021118163, "loss": 0.37557385563850404, "time": 0.6349147319793701, "epoch": 8, "memory": 44120, "step": 28349} +{"lr": 2.5439888258483665e-05, "data_time": 0.001457357406616211, "loss": 0.3426795959472656, "time": 0.6343680381774902, "epoch": 8, "memory": 44120, "step": 28449} +{"lr": 2.5168023298385384e-05, "data_time": 0.0014397621154785157, "loss": 0.3696870893239975, "time": 0.6340624332427979, "epoch": 8, "memory": 44120, "step": 28549} +{"lr": 2.489808716234412e-05, "data_time": 0.0014007806777954102, "loss": 0.35144375562667846, "time": 0.6346299171447753, "epoch": 8, "memory": 44120, "step": 28649} +{"lr": 2.463009730342111e-05, "data_time": 0.0015766382217407226, "loss": 0.3776319772005081, "time": 0.6337181568145752, "epoch": 8, "memory": 44120, "step": 28749} +{"lr": 2.436407104883849e-05, "data_time": 0.0014631271362304688, "loss": 0.37245858907699586, "time": 0.6334508895874024, "epoch": 8, "memory": 44120, "step": 28849} +{"lr": 2.4100025598859092e-05, "data_time": 0.0014869451522827148, "loss": 0.37227138578891755, "time": 0.633303451538086, "epoch": 8, "memory": 44120, "step": 28949} +{"lr": 2.3837978025674284e-05, "data_time": 0.0013957262039184571, "loss": 0.3901132047176361, "time": 0.6316985130310059, "epoch": 8, "memory": 44120, "step": 29049} +{"lr": 2.357794527230027e-05, "data_time": 0.0013980388641357422, "loss": 0.3596083104610443, "time": 0.6336965084075927, "epoch": 8, "memory": 44120, "step": 29149} +{"lr": 2.33199441514825e-05, "data_time": 0.0014456748962402345, "loss": 0.37027046382427214, "time": 0.636094617843628, "epoch": 8, "memory": 44120, "step": 29249} +{"lr": 2.306399134460859e-05, "data_time": 0.0014680862426757813, "loss": 0.3530574142932892, "time": 0.6320498704910278, "epoch": 8, "memory": 44120, "step": 29349} +{"lr": 2.281010340062995e-05, "data_time": 0.0015579462051391602, "loss": 0.3571125894784927, "time": 0.6362739562988281, "epoch": 8, "memory": 44120, "step": 29449} +{"lr": 2.2558296734991603e-05, "data_time": 0.001444101333618164, "loss": 0.40539585053920746, "time": 0.6339248895645142, "epoch": 8, "memory": 44120, "step": 29549} +{"lr": 2.2308587628570915e-05, "data_time": 0.0016648292541503907, "loss": 0.37943858206272124, "time": 0.63364098072052, "epoch": 8, "memory": 44120, "step": 29649} +{"lr": 2.2060992226624955e-05, "data_time": 0.0014145851135253906, "loss": 0.37945939898490905, "time": 0.6324527978897094, "epoch": 8, "memory": 44120, "step": 29749} +{"lr": 2.181552653774648e-05, "data_time": 0.0014185428619384766, "loss": 0.35202763974666595, "time": 0.6372021913528443, "epoch": 8, "memory": 44120, "step": 29849} +{"lr": 2.1572206432829035e-05, "data_time": 0.0019323348999023438, "loss": 0.37268022298812864, "time": 0.6352515935897827, "epoch": 8, "memory": 44120, "step": 29949} +{"lr": 2.1331047644040754e-05, "data_time": 0.0016809225082397461, "loss": 0.3668335735797882, "time": 0.6354601621627808, "epoch": 8, "memory": 44120, "step": 30049} +{"lr": 2.109206576380705e-05, "data_time": 0.001648259162902832, "loss": 0.34897190928459165, "time": 0.6384757995605469, "epoch": 8, "memory": 44120, "step": 30149} +{"lr": 2.0855276243802687e-05, "data_time": 0.001694965362548828, "loss": 0.36093118190765383, "time": 0.6328551769256592, "epoch": 8, "memory": 44120, "step": 30249} +{"lr": 2.0620694393952487e-05, "data_time": 0.0014263153076171874, "loss": 0.35270750522613525, "time": 0.6337556123733521, "epoch": 8, "memory": 44120, "step": 30349} +{"lr": 2.038833538144169e-05, "data_time": 0.0014792442321777343, "loss": 0.3726002395153046, "time": 0.6359061241149903, "epoch": 8, "memory": 44120, "step": 30449} +{"lr": 2.0158214229735154e-05, "data_time": 0.00164031982421875, "loss": 0.3795700967311859, "time": 0.6336644411087036, "epoch": 8, "memory": 44120, "step": 30549} +{"lr": 1.9930345817605977e-05, "data_time": 0.0015683650970458984, "loss": 0.35870579779148104, "time": 0.6432867288589478, "epoch": 8, "memory": 44120, "step": 30649} +{"lr": 1.970474487817358e-05, "data_time": 0.0014643192291259766, "loss": 0.3907363682985306, "time": 0.6326995849609375, "epoch": 8, "memory": 44120, "step": 30749} +{"lr": 1.9481425997951092e-05, "data_time": 0.0015496253967285157, "loss": 0.39445521533489225, "time": 0.6333011150360107, "epoch": 8, "memory": 44120, "step": 30849} +{"lr": 1.9260403615902187e-05, "data_time": 0.0015614032745361328, "loss": 0.355153751373291, "time": 0.6346783638000488, "epoch": 8, "memory": 44120, "step": 30949} +{"lr": 1.9041692022507647e-05, "data_time": 0.001454329490661621, "loss": 0.354151850938797, "time": 0.6324937582015991, "epoch": 8, "memory": 44120, "step": 31049} +{"lr": 1.8825305358841125e-05, "data_time": 0.0018217325210571288, "loss": 0.37641663253307345, "time": 0.6365252733230591, "epoch": 8, "memory": 44120, "step": 31149} +{"lr": 1.861125761565518e-05, "data_time": 0.0012953758239746093, "loss": 0.35719003081321715, "time": 0.6318187713623047, "epoch": 8, "memory": 44120, "step": 31249} +{"accuracy/top1": 75.78024291992188, "data_time": 0.03583048952036891, "time": 1.277594253934663, "step": 8} +{"lr": 1.8384832449122864e-05, "data_time": 0.0015422582626342773, "loss": 0.3495414197444916, "time": 0.6344014883041382, "epoch": 9, "memory": 44120, "step": 31356} +{"lr": 1.817567007332114e-05, "data_time": 0.0013895273208618165, "loss": 0.35953572690486907, "time": 0.6375298976898194, "epoch": 9, "memory": 44120, "step": 31456} +{"lr": 1.796888862098825e-05, "data_time": 0.002831554412841797, "loss": 0.36801652014255526, "time": 0.635310435295105, "epoch": 9, "memory": 44120, "step": 31556} +{"lr": 1.7764501461839678e-05, "data_time": 0.001604008674621582, "loss": 0.371783384680748, "time": 0.6321839570999146, "epoch": 9, "memory": 44120, "step": 31656} +{"lr": 1.7562521810784746e-05, "data_time": 0.0017839908599853516, "loss": 0.36798123121261594, "time": 0.6358216762542724, "epoch": 9, "memory": 44120, "step": 31756} +{"lr": 1.7362962727072444e-05, "data_time": 0.0015273809432983399, "loss": 0.36750137209892275, "time": 0.634696364402771, "epoch": 9, "memory": 44120, "step": 31856} +{"lr": 1.716583711344682e-05, "data_time": 0.0014613866806030273, "loss": 0.38259629607200624, "time": 0.6332085132598877, "epoch": 9, "memory": 44120, "step": 31956} +{"lr": 1.6971157715312954e-05, "data_time": 0.001433110237121582, "loss": 0.363897243142128, "time": 0.6333233833312988, "epoch": 9, "memory": 44120, "step": 32056} +{"lr": 1.677893711991274e-05, "data_time": 0.0014777421951293946, "loss": 0.3699881464242935, "time": 0.6410775423049927, "epoch": 9, "memory": 44120, "step": 32156} +{"lr": 1.6589187755511025e-05, "data_time": 0.0014045000076293944, "loss": 0.3458457738161087, "time": 0.6346138715744019, "epoch": 9, "memory": 44120, "step": 32256} +{"lr": 1.640192189059216e-05, "data_time": 0.001602315902709961, "loss": 0.35345048606395724, "time": 0.6327174663543701, "epoch": 9, "memory": 44120, "step": 32356} +{"lr": 1.6217151633066716e-05, "data_time": 0.0016676664352416992, "loss": 0.35717675387859343, "time": 0.6365456581115723, "epoch": 9, "memory": 44120, "step": 32456} +{"lr": 1.6034888929488564e-05, "data_time": 0.001959061622619629, "loss": 0.34748610854148865, "time": 0.634586763381958, "epoch": 9, "memory": 44120, "step": 32556} +{"lr": 1.585514556428258e-05, "data_time": 0.0017581224441528321, "loss": 0.37288335859775545, "time": 0.6336165189743042, "epoch": 9, "memory": 44120, "step": 32656} +{"lr": 1.567793315898261e-05, "data_time": 0.0015058040618896485, "loss": 0.3559491068124771, "time": 0.6327862977981568, "epoch": 9, "memory": 44120, "step": 32756} +{"lr": 1.5503263171480048e-05, "data_time": 0.0016260862350463868, "loss": 0.35945734977722166, "time": 0.6340567588806152, "epoch": 9, "memory": 44120, "step": 32856} +{"lr": 1.5331146895283165e-05, "data_time": 0.0015779733657836914, "loss": 0.35524179339408873, "time": 0.6357394933700562, "epoch": 9, "memory": 44120, "step": 32956} +{"lr": 1.5161595458786728e-05, "data_time": 0.0017821550369262694, "loss": 0.35637607872486116, "time": 0.6328525066375732, "epoch": 9, "memory": 44120, "step": 33056} +{"lr": 1.4994619824552563e-05, "data_time": 0.0016401052474975587, "loss": 0.35372823774814605, "time": 0.6347967624664307, "epoch": 9, "memory": 44120, "step": 33156} +{"lr": 1.4830230788600802e-05, "data_time": 0.001634073257446289, "loss": 0.3664750546216965, "time": 0.6333654403686524, "epoch": 9, "memory": 44120, "step": 33256} +{"lr": 1.4668438979711779e-05, "data_time": 0.0015042781829833984, "loss": 0.38912451863288877, "time": 0.6319226503372193, "epoch": 9, "memory": 44120, "step": 33356} +{"lr": 1.4509254858738844e-05, "data_time": 0.001482391357421875, "loss": 0.3958167463541031, "time": 0.6325299978256226, "epoch": 9, "memory": 44120, "step": 33456} +{"lr": 1.4352688717931966e-05, "data_time": 0.001739835739135742, "loss": 0.3434902191162109, "time": 0.6332091331481934, "epoch": 9, "memory": 44120, "step": 33556} +{"lr": 1.4198750680272348e-05, "data_time": 0.0016980171203613281, "loss": 0.3578742653131485, "time": 0.6342315196990966, "epoch": 9, "memory": 44120, "step": 33656} +{"lr": 1.4047450698817857e-05, "data_time": 0.0016282558441162109, "loss": 0.38665358126163485, "time": 0.6343876600265503, "epoch": 9, "memory": 44120, "step": 33756} +{"lr": 1.389879855605952e-05, "data_time": 0.0013923883438110352, "loss": 0.37287634015083315, "time": 0.6369168996810913, "epoch": 9, "memory": 44120, "step": 33856} +{"lr": 1.3752803863288976e-05, "data_time": 0.0016299009323120118, "loss": 0.38205778896808623, "time": 0.6344584226608276, "epoch": 9, "memory": 44120, "step": 33956} +{"lr": 1.3609476059977129e-05, "data_time": 0.0015775203704833985, "loss": 0.35270113348960874, "time": 0.6340476036071777, "epoch": 9, "memory": 44120, "step": 34056} +{"lr": 1.3468824413163754e-05, "data_time": 0.0013777256011962891, "loss": 0.36054650843143465, "time": 0.6322983026504516, "epoch": 9, "memory": 44120, "step": 34156} +{"lr": 1.3330858016858395e-05, "data_time": 0.001448678970336914, "loss": 0.38119986951351165, "time": 0.6333877801895141, "epoch": 9, "memory": 44120, "step": 34256} +{"lr": 1.3195585791452309e-05, "data_time": 0.0014369487762451172, "loss": 0.38305503129959106, "time": 0.6359586954116822, "epoch": 9, "memory": 44120, "step": 34356} +{"lr": 1.3063016483141729e-05, "data_time": 0.0013981342315673828, "loss": 0.3698997378349304, "time": 0.6355183124542236, "epoch": 9, "memory": 44120, "step": 34456} +{"lr": 1.2933158663362388e-05, "data_time": 0.0014146804809570313, "loss": 0.3398908317089081, "time": 0.6350657701492309, "epoch": 9, "memory": 44120, "step": 34556} +{"lr": 1.2806020728235311e-05, "data_time": 0.0015672922134399414, "loss": 0.37949779331684114, "time": 0.6337861537933349, "epoch": 9, "memory": 44120, "step": 34656} +{"lr": 1.2681610898023963e-05, "data_time": 0.0016001224517822265, "loss": 0.36143686771392824, "time": 0.6342676877975464, "epoch": 9, "memory": 44120, "step": 34756} +{"lr": 1.2559937216602731e-05, "data_time": 0.0014892339706420899, "loss": 0.3563768804073334, "time": 0.6327040672302247, "epoch": 9, "memory": 44120, "step": 34856} +{"lr": 1.2441007550936847e-05, "data_time": 0.001398324966430664, "loss": 0.3814793139696121, "time": 0.6365732669830322, "epoch": 9, "memory": 44120, "step": 34956} +{"lr": 1.232482959057377e-05, "data_time": 0.0014713525772094727, "loss": 0.3600014805793762, "time": 0.6337686538696289, "epoch": 9, "memory": 44120, "step": 35056} +{"lr": 1.2211410847145969e-05, "data_time": 0.0012830972671508789, "loss": 0.34628230333328247, "time": 0.6359486818313599, "epoch": 9, "memory": 44120, "step": 35156} +{"accuracy/top1": 75.58540344238281, "data_time": 0.035728594352459085, "time": 1.2794173380424236, "step": 9} +{"lr": 1.209311679360931e-05, "data_time": 0.001469564437866211, "loss": 0.339600196480751, "time": 0.6344933986663819, "epoch": 10, "memory": 44120, "step": 35263} +{"lr": 1.1985432727488666e-05, "data_time": 0.0016040325164794922, "loss": 0.3448700666427612, "time": 0.6342395782470703, "epoch": 10, "memory": 44120, "step": 35363} +{"lr": 1.1880529822435289e-05, "data_time": 0.0015718460083007813, "loss": 0.37429238557815553, "time": 0.6348059892654419, "epoch": 10, "memory": 44120, "step": 35463} +{"lr": 1.1778414861078726e-05, "data_time": 0.0016747713088989258, "loss": 0.35528212785720825, "time": 0.6322180747985839, "epoch": 10, "memory": 44120, "step": 35563} +{"lr": 1.1679094445790538e-05, "data_time": 0.001582789421081543, "loss": 0.39299889504909513, "time": 0.6315577983856201, "epoch": 10, "memory": 44120, "step": 35663} +{"lr": 1.1582574998257377e-05, "data_time": 0.001683497428894043, "loss": 0.3906119465827942, "time": 0.6320613145828247, "epoch": 10, "memory": 44120, "step": 35763} +{"lr": 1.1488862759065762e-05, "data_time": 0.0015350580215454102, "loss": 0.36365343630313873, "time": 0.6355814218521119, "epoch": 10, "memory": 44120, "step": 35863} +{"lr": 1.1397963787298665e-05, "data_time": 0.001596379280090332, "loss": 0.3629516690969467, "time": 0.6324075937271119, "epoch": 10, "memory": 44120, "step": 35963} +{"lr": 1.1309883960143664e-05, "data_time": 0.0014279842376708984, "loss": 0.3542255997657776, "time": 0.6352359771728515, "epoch": 10, "memory": 44120, "step": 36063} +{"lr": 1.122462897251302e-05, "data_time": 0.0016389846801757812, "loss": 0.3491236180067062, "time": 0.6526760578155517, "epoch": 10, "memory": 44120, "step": 36163} +{"lr": 1.114220433667542e-05, "data_time": 0.0014959096908569336, "loss": 0.3740518867969513, "time": 0.6347596406936645, "epoch": 10, "memory": 44120, "step": 36263} +{"lr": 1.10626153818996e-05, "data_time": 0.0015249252319335938, "loss": 0.3636352241039276, "time": 0.6325485467910766, "epoch": 10, "memory": 44120, "step": 36363} +{"lr": 1.0985867254109761e-05, "data_time": 0.001751875877380371, "loss": 0.3938641458749771, "time": 0.6334201097488403, "epoch": 10, "memory": 44120, "step": 36463} +{"lr": 1.0911964915552823e-05, "data_time": 0.0019232273101806641, "loss": 0.34288425743579865, "time": 0.631958794593811, "epoch": 10, "memory": 44120, "step": 36563} +{"lr": 1.0840913144477643e-05, "data_time": 0.0015038013458251952, "loss": 0.36984099447727203, "time": 0.6352145433425903, "epoch": 10, "memory": 44120, "step": 36663} +{"lr": 1.0772716534826064e-05, "data_time": 0.0015038490295410157, "loss": 0.3477191925048828, "time": 0.6329223394393921, "epoch": 10, "memory": 44120, "step": 36763} +{"lr": 1.0707379495935828e-05, "data_time": 0.0015514373779296875, "loss": 0.34641546905040743, "time": 0.6327091217041015, "epoch": 10, "memory": 44120, "step": 36863} +{"lr": 1.0644906252255541e-05, "data_time": 0.0017029047012329102, "loss": 0.39467841386795044, "time": 0.6321772813796998, "epoch": 10, "memory": 44120, "step": 36963} +{"lr": 1.0585300843071527e-05, "data_time": 0.0015934467315673827, "loss": 0.39513387382030485, "time": 0.6448676824569702, "epoch": 10, "memory": 44120, "step": 37063} +{"lr": 1.0528567122246654e-05, "data_time": 0.0015032768249511718, "loss": 0.3685428321361542, "time": 0.6345503091812134, "epoch": 10, "memory": 44120, "step": 37163} +{"lr": 1.0474708757971141e-05, "data_time": 0.001501607894897461, "loss": 0.3467086672782898, "time": 0.634011173248291, "epoch": 10, "memory": 44120, "step": 37263} +{"lr": 1.042372923252543e-05, "data_time": 0.0015008926391601562, "loss": 0.3592089474201202, "time": 0.6336398124694824, "epoch": 10, "memory": 44120, "step": 37363} +{"lr": 1.0375631842054998e-05, "data_time": 0.001678323745727539, "loss": 0.3563314974308014, "time": 0.634985375404358, "epoch": 10, "memory": 44120, "step": 37463} +{"lr": 1.0330419696357254e-05, "data_time": 0.0017746448516845702, "loss": 0.33439682722091674, "time": 0.6366176128387451, "epoch": 10, "memory": 44120, "step": 37563} +{"lr": 1.028809571868047e-05, "data_time": 0.0015472173690795898, "loss": 0.34144375622272494, "time": 0.6344694852828979, "epoch": 10, "memory": 44120, "step": 37663} +{"lr": 1.0248662645534766e-05, "data_time": 0.0017197608947753906, "loss": 0.3623490482568741, "time": 0.6318809509277343, "epoch": 10, "memory": 44120, "step": 37763} +{"lr": 1.0212123026515194e-05, "data_time": 0.001902318000793457, "loss": 0.36674031615257263, "time": 0.636697244644165, "epoch": 10, "memory": 44120, "step": 37863} +{"lr": 1.0178479224136881e-05, "data_time": 0.001537632942199707, "loss": 0.3548463523387909, "time": 0.6348671197891236, "epoch": 10, "memory": 44120, "step": 37963} +{"lr": 1.0147733413682289e-05, "data_time": 0.001559305191040039, "loss": 0.3793183296918869, "time": 0.6346055507659912, "epoch": 10, "memory": 44120, "step": 38063} +{"lr": 1.0119887583060527e-05, "data_time": 0.0013947725296020509, "loss": 0.36996032297611237, "time": 0.6342608213424683, "epoch": 10, "memory": 44120, "step": 38163} +{"lr": 1.0094943532678871e-05, "data_time": 0.0013764142990112306, "loss": 0.39604129195213317, "time": 0.6383871078491211, "epoch": 10, "memory": 44120, "step": 38263} +{"lr": 1.007290287532637e-05, "data_time": 0.0021173954010009766, "loss": 0.36310293078422545, "time": 0.6334275960922241, "epoch": 10, "memory": 44120, "step": 38363} +{"lr": 1.0053767036069495e-05, "data_time": 0.0017472028732299805, "loss": 0.38397795557975767, "time": 0.6343852758407593, "epoch": 10, "memory": 44120, "step": 38463} +{"lr": 1.003753725216005e-05, "data_time": 0.0014446496963500977, "loss": 0.3818373054265976, "time": 0.633182430267334, "epoch": 10, "memory": 44120, "step": 38563} +{"lr": 1.0024214572955152e-05, "data_time": 0.0016610383987426757, "loss": 0.36392607986927034, "time": 0.6334603786468506, "epoch": 10, "memory": 44120, "step": 38663} +{"lr": 1.0013799859849456e-05, "data_time": 0.0014972209930419922, "loss": 0.37379465997219086, "time": 0.633960509300232, "epoch": 10, "memory": 44120, "step": 38763} +{"lr": 1.0006293786219338e-05, "data_time": 0.0019221782684326171, "loss": 0.34700888097286225, "time": 0.6357094049453735, "epoch": 10, "memory": 44120, "step": 38863} +{"lr": 1.0001696837379472e-05, "data_time": 0.0013655424118041992, "loss": 0.35798477232456205, "time": 0.6330069303512573, "epoch": 10, "memory": 44120, "step": 38963} +{"lr": 1.0000009310551361e-05, "data_time": 0.0013276338577270508, "loss": 0.35898124873638154, "time": 0.6327818870544434, "epoch": 10, "memory": 44120, "step": 39063} +{"accuracy/top1": 75.45318603515625, "data_time": 0.03510504755480536, "time": 1.276444990059425, "step": 10} diff --git a/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm/20230604_023242/vis_data/config.py b/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm/20230604_023242/vis_data/config.py new file mode 100644 index 0000000000000000000000000000000000000000..35ec1e34d29b17c45878985adb3abd9b6709a41c --- /dev/null +++ b/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm/20230604_023242/vis_data/config.py @@ -0,0 +1,228 @@ +optim_wrapper = dict( + optimizer=dict( + type='AdamW', lr=0.0001, weight_decay=0.3, _scope_='mmpretrain'), + paramwise_cfg=dict( + custom_keys=dict({ + '.cls_token': dict(decay_mult=0.0), + '.pos_embed': dict(decay_mult=0.0) + })), + type='AmpOptimWrapper', + dtype='bfloat16', + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=False, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=4096) +model = dict( + type='ImageClassifier', + backbone=dict( + frozen_stages=24, + type='VisionTransformer', + arch='l', + img_size=224, + patch_size=14, + drop_rate=0.1, + pre_norm=True, + final_norm=False, + init_cfg=dict( + type='Pretrained', + checkpoint='ckpt/openclip-ViT-L-14.pth', + prefix='backbone')), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=1024, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + init_cfg=None), + init_cfg=dict( + type='TruncNormal', layer=['Conv2d', 'Linear'], std=0.02, bias=0.0), + train_cfg=None) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=128, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stablediffusionV2-1-dpmsolver-25-1m.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/cc1m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=128, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV2-1-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=128, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV2-1-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm' diff --git a/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm/20230604_023242/vis_data/events.out.tfevents.1685817177.SH-IDC1-10-140-24-19.149327.0 b/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm/20230604_023242/vis_data/events.out.tfevents.1685817177.SH-IDC1-10-140-24-19.149327.0 new file mode 100644 index 0000000000000000000000000000000000000000..0a0c04cbccd83a9bb35703752fe72efb3024b65d --- /dev/null +++ b/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm/20230604_023242/vis_data/events.out.tfevents.1685817177.SH-IDC1-10-140-24-19.149327.0 @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:6ec6b7a7b2d2c0430acce39c9e7e83fa70cebeb253b81f0edc6b8f2021806470 +size 1695630 diff --git a/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm/20230604_023242/vis_data/scalars.json b/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm/20230604_023242/vis_data/scalars.json new file mode 100644 index 0000000000000000000000000000000000000000..4c318ecb20e8799b32aa51adfb67486042a0e169 --- /dev/null +++ b/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm/20230604_023242/vis_data/scalars.json @@ -0,0 +1,400 @@ +{"lr": 9.999857418382631e-05, "data_time": 0.0013934850692749023, "loss": 0.6554573059082032, "time": 0.6275890350341797, "epoch": 1, "memory": 44121, "step": 100} +{"lr": 9.999423907359506e-05, "data_time": 0.001732015609741211, "loss": 0.6052772521972656, "time": 0.6291704654693604, "epoch": 1, "memory": 44121, "step": 200} +{"lr": 9.998699480412073e-05, "data_time": 0.001477670669555664, "loss": 0.6014793395996094, "time": 0.6299166202545166, "epoch": 1, "memory": 44121, "step": 300} +{"lr": 9.997684184379085e-05, "data_time": 0.0015866518020629882, "loss": 0.5677017211914063, "time": 0.6311460256576538, "epoch": 1, "memory": 44121, "step": 400} +{"lr": 9.996378084905784e-05, "data_time": 0.0016239166259765625, "loss": 0.5291509628295898, "time": 0.6303429365158081, "epoch": 1, "memory": 44121, "step": 500} +{"lr": 9.994781266439667e-05, "data_time": 0.0013891220092773437, "loss": 0.49774875640869143, "time": 0.6303493738174438, "epoch": 1, "memory": 44121, "step": 600} +{"lr": 9.992893832225061e-05, "data_time": 0.001440286636352539, "loss": 0.4941108703613281, "time": 0.6305997133255005, "epoch": 1, "memory": 44121, "step": 700} +{"lr": 9.990715904296405e-05, "data_time": 0.0015773534774780273, "loss": 0.4833106994628906, "time": 0.6303353071212768, "epoch": 1, "memory": 44121, "step": 800} +{"lr": 9.988247623470371e-05, "data_time": 0.001455545425415039, "loss": 0.5067439079284668, "time": 0.6297955513000488, "epoch": 1, "memory": 44121, "step": 900} +{"lr": 9.985489149336749e-05, "data_time": 0.001456451416015625, "loss": 0.48727102279663087, "time": 0.630203366279602, "epoch": 1, "memory": 44121, "step": 1000} +{"lr": 9.982440660248187e-05, "data_time": 0.0014949083328247071, "loss": 0.47072653770446776, "time": 0.6298863410949707, "epoch": 1, "memory": 44121, "step": 1100} +{"lr": 9.979102353308574e-05, "data_time": 0.0013971328735351562, "loss": 0.4766227722167969, "time": 0.6302656412124634, "epoch": 1, "memory": 44121, "step": 1200} +{"lr": 9.975474444360358e-05, "data_time": 0.0014298439025878906, "loss": 0.45115349292755125, "time": 0.6305836915969849, "epoch": 1, "memory": 44121, "step": 1300} +{"lr": 9.97155716797056e-05, "data_time": 0.0014416933059692382, "loss": 0.44846794605255125, "time": 0.6303877830505371, "epoch": 1, "memory": 44121, "step": 1400} +{"lr": 9.967350777415632e-05, "data_time": 0.0013402223587036133, "loss": 0.4739827275276184, "time": 0.6301686763763428, "epoch": 1, "memory": 44121, "step": 1500} +{"lr": 9.962855544665059e-05, "data_time": 0.001266622543334961, "loss": 0.4373971700668335, "time": 0.6306634426116944, "epoch": 1, "memory": 44121, "step": 1600} +{"lr": 9.958071760363773e-05, "data_time": 0.0014128923416137696, "loss": 0.4457846760749817, "time": 0.6300257921218873, "epoch": 1, "memory": 44121, "step": 1700} +{"lr": 9.952999733813382e-05, "data_time": 0.0014392852783203125, "loss": 0.4599105954170227, "time": 0.6298099040985108, "epoch": 1, "memory": 44121, "step": 1800} +{"lr": 9.947639792952148e-05, "data_time": 0.00150909423828125, "loss": 0.4459614634513855, "time": 0.6304698467254639, "epoch": 1, "memory": 44121, "step": 1900} +{"lr": 9.941992284333813e-05, "data_time": 0.0017377376556396485, "loss": 0.42215226888656615, "time": 0.6298106431961059, "epoch": 1, "memory": 44121, "step": 2000} +{"lr": 9.93605757310516e-05, "data_time": 0.0016527652740478515, "loss": 0.45085784792900085, "time": 0.6294882774353028, "epoch": 1, "memory": 44121, "step": 2100} +{"lr": 9.929836042982421e-05, "data_time": 0.0014800786972045898, "loss": 0.4206417202949524, "time": 0.6305043935775757, "epoch": 1, "memory": 44121, "step": 2200} +{"lr": 9.923328096226467e-05, "data_time": 0.001474308967590332, "loss": 0.428151935338974, "time": 0.630351448059082, "epoch": 1, "memory": 44121, "step": 2300} +{"lr": 9.916534153616819e-05, "data_time": 0.0013605356216430664, "loss": 0.4223970860242844, "time": 0.6298309326171875, "epoch": 1, "memory": 44121, "step": 2400} +{"lr": 9.909454654424387e-05, "data_time": 0.0013653755187988282, "loss": 0.43172143697738646, "time": 0.6304721593856811, "epoch": 1, "memory": 44121, "step": 2500} +{"lr": 9.902090056383109e-05, "data_time": 0.0013478517532348633, "loss": 0.42010399103164675, "time": 0.6304569244384766, "epoch": 1, "memory": 44121, "step": 2600} +{"lr": 9.894440835660365e-05, "data_time": 0.0014511585235595704, "loss": 0.39755138754844666, "time": 0.6310604810714722, "epoch": 1, "memory": 44121, "step": 2700} +{"lr": 9.886507486826162e-05, "data_time": 0.0014309167861938476, "loss": 0.3904307305812836, "time": 0.6300219774246216, "epoch": 1, "memory": 44121, "step": 2800} +{"lr": 9.878290522821154e-05, "data_time": 0.001474452018737793, "loss": 0.4230853796005249, "time": 0.630784010887146, "epoch": 1, "memory": 44121, "step": 2900} +{"lr": 9.86979047492352e-05, "data_time": 0.001485919952392578, "loss": 0.40724811553955076, "time": 0.630129361152649, "epoch": 1, "memory": 44121, "step": 3000} +{"lr": 9.861007892714578e-05, "data_time": 0.0015177488327026366, "loss": 0.42819066643714904, "time": 0.6301684856414795, "epoch": 1, "memory": 44121, "step": 3100} +{"lr": 9.851943344043239e-05, "data_time": 0.0015147209167480468, "loss": 0.419231653213501, "time": 0.6296551465988159, "epoch": 1, "memory": 44121, "step": 3200} +{"lr": 9.842597414989319e-05, "data_time": 0.0014043331146240234, "loss": 0.40978620052337644, "time": 0.6303950071334838, "epoch": 1, "memory": 44121, "step": 3300} +{"lr": 9.832970709825637e-05, "data_time": 0.0014115333557128905, "loss": 0.3929266661405563, "time": 0.6304152488708497, "epoch": 1, "memory": 44121, "step": 3400} +{"lr": 9.823063850978968e-05, "data_time": 0.0015639781951904297, "loss": 0.4143178790807724, "time": 0.6300772905349732, "epoch": 1, "memory": 44121, "step": 3500} +{"lr": 9.81287747898975e-05, "data_time": 0.0017814397811889648, "loss": 0.4079999029636383, "time": 0.6292192220687867, "epoch": 1, "memory": 44121, "step": 3600} +{"lr": 9.802412252470715e-05, "data_time": 0.0014991044998168945, "loss": 0.3691905975341797, "time": 0.6297514200210571, "epoch": 1, "memory": 44121, "step": 3700} +{"lr": 9.791668848064258e-05, "data_time": 0.0014405965805053711, "loss": 0.3997865170240402, "time": 0.6306843519210815, "epoch": 1, "memory": 44121, "step": 3800} +{"lr": 9.780647960398751e-05, "data_time": 0.0011514902114868164, "loss": 0.39070974588394164, "time": 0.6291760921478271, "epoch": 1, "memory": 44121, "step": 3900} +{"accuracy/top1": 73.46299743652344, "data_time": 0.046683474590903835, "time": 1.3394464317121004, "step": 1} +{"lr": 9.7685491196646e-05, "data_time": 0.0016675472259521484, "loss": 0.40552919507026675, "time": 0.6314837455749511, "epoch": 2, "memory": 44121, "step": 4007} +{"lr": 9.756956126085676e-05, "data_time": 0.0013395309448242187, "loss": 0.3878222227096558, "time": 0.6305492401123047, "epoch": 2, "memory": 44120, "step": 4107} +{"lr": 9.745087893642482e-05, "data_time": 0.0015935420989990235, "loss": 0.3939862012863159, "time": 0.6303120136260987, "epoch": 2, "memory": 44120, "step": 4207} +{"lr": 9.73294518969056e-05, "data_time": 0.002163076400756836, "loss": 0.381178480386734, "time": 0.6314022541046143, "epoch": 2, "memory": 44120, "step": 4307} +{"lr": 9.72052879933172e-05, "data_time": 0.002112078666687012, "loss": 0.40107953250408174, "time": 0.635162091255188, "epoch": 2, "memory": 44120, "step": 4407} +{"lr": 9.707839525363358e-05, "data_time": 0.0016361236572265624, "loss": 0.39447016417980196, "time": 0.6308466911315918, "epoch": 2, "memory": 44120, "step": 4507} +{"lr": 9.69487818822646e-05, "data_time": 0.0017443418502807618, "loss": 0.39011659622192385, "time": 0.6310233116149903, "epoch": 2, "memory": 44120, "step": 4607} +{"lr": 9.681645625952616e-05, "data_time": 0.0013948917388916016, "loss": 0.37863059639930724, "time": 0.6311232566833496, "epoch": 2, "memory": 44120, "step": 4707} +{"lr": 9.668142694109826e-05, "data_time": 0.0016169309616088866, "loss": 0.3920276552438736, "time": 0.6308682680130004, "epoch": 2, "memory": 44120, "step": 4807} +{"lr": 9.654370265747162e-05, "data_time": 0.001408863067626953, "loss": 0.3837639331817627, "time": 0.6317102670669555, "epoch": 2, "memory": 44120, "step": 4907} +{"lr": 9.640329231338345e-05, "data_time": 0.001631021499633789, "loss": 0.38010249733924867, "time": 0.6313638925552368, "epoch": 2, "memory": 44120, "step": 5007} +{"lr": 9.626020498724147e-05, "data_time": 0.0014079093933105468, "loss": 0.38657009303569795, "time": 0.6307583570480346, "epoch": 2, "memory": 44120, "step": 5107} +{"lr": 9.611444993053717e-05, "data_time": 0.001447010040283203, "loss": 0.3812067717313766, "time": 0.6313483238220214, "epoch": 2, "memory": 44120, "step": 5207} +{"lr": 9.596603656724746e-05, "data_time": 0.00148773193359375, "loss": 0.37799614667892456, "time": 0.6316648721694946, "epoch": 2, "memory": 44120, "step": 5307} +{"lr": 9.58149744932254e-05, "data_time": 0.0013596057891845704, "loss": 0.3787410736083984, "time": 0.6309352874755859, "epoch": 2, "memory": 44120, "step": 5407} +{"lr": 9.566127347557972e-05, "data_time": 0.0014389753341674805, "loss": 0.3914622485637665, "time": 0.6321154356002807, "epoch": 2, "memory": 44120, "step": 5507} +{"lr": 9.550494345204367e-05, "data_time": 0.001526641845703125, "loss": 0.3773065090179443, "time": 0.6316898584365844, "epoch": 2, "memory": 44120, "step": 5607} +{"lr": 9.534599453033183e-05, "data_time": 0.0015011310577392578, "loss": 0.3765223801136017, "time": 0.6306300640106202, "epoch": 2, "memory": 44120, "step": 5707} +{"lr": 9.518443698748712e-05, "data_time": 0.001419973373413086, "loss": 0.377192884683609, "time": 0.6458696842193603, "epoch": 2, "memory": 44120, "step": 5807} +{"lr": 9.50202812692162e-05, "data_time": 0.0016075849533081054, "loss": 0.39714027345180514, "time": 0.6313439846038819, "epoch": 2, "memory": 44120, "step": 5907} +{"lr": 9.485353798921386e-05, "data_time": 0.001783299446105957, "loss": 0.38935945332050326, "time": 0.6316739320755005, "epoch": 2, "memory": 44120, "step": 6007} +{"lr": 9.468421792847712e-05, "data_time": 0.0013911962509155274, "loss": 0.37018243670463563, "time": 0.6333726406097412, "epoch": 2, "memory": 44120, "step": 6107} +{"lr": 9.451233203460794e-05, "data_time": 0.0014320135116577149, "loss": 0.374513179063797, "time": 0.6307969331741333, "epoch": 2, "memory": 44120, "step": 6207} +{"lr": 9.433789142110532e-05, "data_time": 0.0014647960662841797, "loss": 0.34460248947143557, "time": 0.6348870038986206, "epoch": 2, "memory": 44120, "step": 6307} +{"lr": 9.416090736664709e-05, "data_time": 0.0014971494674682617, "loss": 0.3948313921689987, "time": 0.6313304901123047, "epoch": 2, "memory": 44120, "step": 6407} +{"lr": 9.398139131436032e-05, "data_time": 0.0014452457427978516, "loss": 0.37769335210323335, "time": 0.6306396484375, "epoch": 2, "memory": 44120, "step": 6507} +{"lr": 9.379935487108167e-05, "data_time": 0.0014565229415893556, "loss": 0.3787588536739349, "time": 0.6308593034744263, "epoch": 2, "memory": 44120, "step": 6607} +{"lr": 9.361480980660697e-05, "data_time": 0.0013815402984619141, "loss": 0.3671922743320465, "time": 0.6302517652511597, "epoch": 2, "memory": 44120, "step": 6707} +{"lr": 9.342776805292993e-05, "data_time": 0.0014109611511230469, "loss": 0.38747328519821167, "time": 0.6315154552459716, "epoch": 2, "memory": 44120, "step": 6807} +{"lr": 9.323824170347083e-05, "data_time": 0.001415562629699707, "loss": 0.3590896546840668, "time": 0.6315407037734986, "epoch": 2, "memory": 44120, "step": 6907} +{"lr": 9.304624301229471e-05, "data_time": 0.0016486167907714844, "loss": 0.3561199963092804, "time": 0.6330701351165772, "epoch": 2, "memory": 44120, "step": 7007} +{"lr": 9.285178439331897e-05, "data_time": 0.0016980648040771484, "loss": 0.37257237136363985, "time": 0.6327742576599121, "epoch": 2, "memory": 44120, "step": 7107} +{"lr": 9.265487841951087e-05, "data_time": 0.0015543222427368164, "loss": 0.3504183143377304, "time": 0.6309172153472901, "epoch": 2, "memory": 44120, "step": 7207} +{"lr": 9.245553782207421e-05, "data_time": 0.0015695333480834962, "loss": 0.34269999563694, "time": 0.6304365396499634, "epoch": 2, "memory": 44120, "step": 7307} +{"lr": 9.225377548962671e-05, "data_time": 0.0013744831085205078, "loss": 0.36070674657821655, "time": 0.631167221069336, "epoch": 2, "memory": 44120, "step": 7407} +{"lr": 9.204960446736601e-05, "data_time": 0.001514744758605957, "loss": 0.36700805723667146, "time": 0.631222915649414, "epoch": 2, "memory": 44120, "step": 7507} +{"lr": 9.184303795622715e-05, "data_time": 0.0014163494110107423, "loss": 0.3641028881072998, "time": 0.6312353372573852, "epoch": 2, "memory": 44120, "step": 7607} +{"lr": 9.163408931202823e-05, "data_time": 0.001429128646850586, "loss": 0.3695355117321014, "time": 0.631613826751709, "epoch": 2, "memory": 44120, "step": 7707} +{"lr": 9.142277204460729e-05, "data_time": 0.001242852210998535, "loss": 0.3687566339969635, "time": 0.6312731742858887, "epoch": 2, "memory": 44120, "step": 7807} +{"accuracy/top1": 77.84349822998047, "data_time": 0.036329022769270275, "time": 1.278838428957709, "step": 2} +{"lr": 9.119405492383288e-05, "data_time": 0.0014481067657470704, "loss": 0.3847730875015259, "time": 0.6318252086639404, "epoch": 3, "memory": 44120, "step": 7914} +{"lr": 9.097787819233205e-05, "data_time": 0.001453399658203125, "loss": 0.40148198008537295, "time": 0.6320568799972535, "epoch": 3, "memory": 44120, "step": 8014} +{"lr": 9.075937526576627e-05, "data_time": 0.0014786481857299804, "loss": 0.38979090452194215, "time": 0.6316263675689697, "epoch": 3, "memory": 44120, "step": 8114} +{"lr": 9.053856027171755e-05, "data_time": 0.0015566349029541016, "loss": 0.3463983088731766, "time": 0.6335758686065673, "epoch": 3, "memory": 44120, "step": 8214} +{"lr": 9.031544748725785e-05, "data_time": 0.0023714542388916016, "loss": 0.36977761089801786, "time": 0.6326429605484009, "epoch": 3, "memory": 44120, "step": 8314} +{"lr": 9.009005133802542e-05, "data_time": 0.0015105724334716797, "loss": 0.3708623141050339, "time": 0.6312785625457764, "epoch": 3, "memory": 44120, "step": 8414} +{"lr": 8.986238639729242e-05, "data_time": 0.0014684200286865234, "loss": 0.39234643876552583, "time": 0.6314328193664551, "epoch": 3, "memory": 44120, "step": 8514} +{"lr": 8.963246738502254e-05, "data_time": 0.0015577077865600586, "loss": 0.3743011951446533, "time": 0.6321437358856201, "epoch": 3, "memory": 44120, "step": 8614} +{"lr": 8.940030916691928e-05, "data_time": 0.001475071907043457, "loss": 0.3778077602386475, "time": 0.6326121807098388, "epoch": 3, "memory": 44120, "step": 8714} +{"lr": 8.91659267534649e-05, "data_time": 0.0013529062271118164, "loss": 0.3694726645946503, "time": 0.6318154335021973, "epoch": 3, "memory": 44120, "step": 8814} +{"lr": 8.892933529894978e-05, "data_time": 0.0015254497528076171, "loss": 0.36761326491832735, "time": 0.6327833890914917, "epoch": 3, "memory": 44120, "step": 8914} +{"lr": 8.869055010049283e-05, "data_time": 0.001403641700744629, "loss": 0.38764975368976595, "time": 0.6352396965026855, "epoch": 3, "memory": 44120, "step": 9014} +{"lr": 8.844958659705188e-05, "data_time": 0.0014962196350097657, "loss": 0.3604812115430832, "time": 0.632267951965332, "epoch": 3, "memory": 44120, "step": 9114} +{"lr": 8.82064603684259e-05, "data_time": 0.001603245735168457, "loss": 0.37507286071777346, "time": 0.6401543617248535, "epoch": 3, "memory": 44120, "step": 9214} +{"lr": 8.796118713424776e-05, "data_time": 0.0015782594680786132, "loss": 0.37148820161819457, "time": 0.6333001375198364, "epoch": 3, "memory": 44120, "step": 9314} +{"lr": 8.77137827529676e-05, "data_time": 0.0014742612838745117, "loss": 0.3671196222305298, "time": 0.6341067314147949, "epoch": 3, "memory": 44120, "step": 9414} +{"lr": 8.746426322082745e-05, "data_time": 0.0015903234481811524, "loss": 0.3711984634399414, "time": 0.6319302320480347, "epoch": 3, "memory": 44120, "step": 9514} +{"lr": 8.721264467082732e-05, "data_time": 0.0015094518661499024, "loss": 0.3907595783472061, "time": 0.6322148084640503, "epoch": 3, "memory": 44120, "step": 9614} +{"lr": 8.695894337168168e-05, "data_time": 0.001618647575378418, "loss": 0.362550875544548, "time": 0.6325819492340088, "epoch": 3, "memory": 44120, "step": 9714} +{"lr": 8.67031757267677e-05, "data_time": 0.0016858577728271484, "loss": 0.36691901087760925, "time": 0.6320786714553833, "epoch": 3, "memory": 44120, "step": 9814} +{"lr": 8.64453582730651e-05, "data_time": 0.0014262199401855469, "loss": 0.34382962584495547, "time": 0.6324646711349488, "epoch": 3, "memory": 44120, "step": 9914} +{"lr": 8.61855076800861e-05, "data_time": 0.0014376401901245116, "loss": 0.3303049474954605, "time": 0.6322244167327881, "epoch": 3, "memory": 44120, "step": 10014} +{"lr": 8.592364074879843e-05, "data_time": 0.001498246192932129, "loss": 0.3917563557624817, "time": 0.6324445724487304, "epoch": 3, "memory": 44120, "step": 10114} +{"lr": 8.565977441053854e-05, "data_time": 0.0013967514038085937, "loss": 0.3499339759349823, "time": 0.6343879699707031, "epoch": 3, "memory": 44120, "step": 10214} +{"lr": 8.539392572591724e-05, "data_time": 0.0013753652572631836, "loss": 0.3634859323501587, "time": 0.6313601016998291, "epoch": 3, "memory": 44120, "step": 10314} +{"lr": 8.512611188371633e-05, "data_time": 0.001399397850036621, "loss": 0.37576409578323366, "time": 0.6319352149963379, "epoch": 3, "memory": 44120, "step": 10414} +{"lr": 8.485635019977725e-05, "data_time": 0.0013954639434814453, "loss": 0.35563134849071504, "time": 0.6318157196044922, "epoch": 3, "memory": 44120, "step": 10514} +{"lr": 8.45846581158818e-05, "data_time": 0.001403212547302246, "loss": 0.3612090051174164, "time": 0.6316814422607422, "epoch": 3, "memory": 44120, "step": 10614} +{"lr": 8.431105319862416e-05, "data_time": 0.0014333009719848632, "loss": 0.3662464290857315, "time": 0.6306448698043823, "epoch": 3, "memory": 44120, "step": 10714} +{"lr": 8.403555313827516e-05, "data_time": 0.001498270034790039, "loss": 0.38241981863975527, "time": 0.6314601182937623, "epoch": 3, "memory": 44120, "step": 10814} +{"lr": 8.375817574763844e-05, "data_time": 0.0013619184494018555, "loss": 0.37417236268520354, "time": 0.6317417860031128, "epoch": 3, "memory": 44120, "step": 10914} +{"lr": 8.347893896089871e-05, "data_time": 0.0014081716537475586, "loss": 0.33988997936248777, "time": 0.63100106716156, "epoch": 3, "memory": 44120, "step": 11014} +{"lr": 8.31978608324625e-05, "data_time": 0.001481938362121582, "loss": 0.38768957257270814, "time": 0.6333823680877686, "epoch": 3, "memory": 44120, "step": 11114} +{"lr": 8.291495953579052e-05, "data_time": 0.0012996912002563477, "loss": 0.342476087808609, "time": 0.6363723754882813, "epoch": 3, "memory": 44120, "step": 11214} +{"lr": 8.263025336222267e-05, "data_time": 0.001495051383972168, "loss": 0.37864493429660795, "time": 0.6331802368164062, "epoch": 3, "memory": 44120, "step": 11314} +{"lr": 8.234376071979551e-05, "data_time": 0.0014034032821655274, "loss": 0.37599709033966067, "time": 0.6316048860549927, "epoch": 3, "memory": 44120, "step": 11414} +{"lr": 8.205550013205187e-05, "data_time": 0.0015259027481079102, "loss": 0.3576871693134308, "time": 0.6314178228378295, "epoch": 3, "memory": 44120, "step": 11514} +{"lr": 8.17654902368435e-05, "data_time": 0.001470804214477539, "loss": 0.3580753803253174, "time": 0.6342048406600952, "epoch": 3, "memory": 44120, "step": 11614} +{"lr": 8.147374978512582e-05, "data_time": 0.0012356758117675782, "loss": 0.3363444447517395, "time": 0.6309850215911865, "epoch": 3, "memory": 44120, "step": 11714} +{"accuracy/top1": 77.0815200805664, "data_time": 0.03565501344615016, "time": 1.2774384145078987, "step": 3} +{"lr": 8.115969237626578e-05, "data_time": 0.0014601707458496093, "loss": 0.3616588294506073, "time": 0.6327504396438599, "epoch": 4, "memory": 44120, "step": 11821} +{"lr": 8.08644297337245e-05, "data_time": 0.0013072729110717774, "loss": 0.3634976029396057, "time": 0.6361582517623902, "epoch": 4, "memory": 44120, "step": 11921} +{"lr": 8.056749479387651e-05, "data_time": 0.001401519775390625, "loss": 0.36289515495300295, "time": 0.6348573923110962, "epoch": 4, "memory": 44120, "step": 12021} +{"lr": 8.026890675542457e-05, "data_time": 0.0018458604812622071, "loss": 0.3610669583082199, "time": 0.6331987380981445, "epoch": 4, "memory": 44120, "step": 12121} +{"lr": 7.996868492395457e-05, "data_time": 0.0014137506484985351, "loss": 0.35711357891559603, "time": 0.6319075345993042, "epoch": 4, "memory": 44120, "step": 12221} +{"lr": 7.966684871068762e-05, "data_time": 0.0015436172485351562, "loss": 0.37489054501056673, "time": 0.6419233560562134, "epoch": 4, "memory": 44120, "step": 12321} +{"lr": 7.936341763122449e-05, "data_time": 0.0014515399932861328, "loss": 0.366738298535347, "time": 0.6329026937484741, "epoch": 4, "memory": 44120, "step": 12421} +{"lr": 7.905841130428405e-05, "data_time": 0.0014167308807373046, "loss": 0.3751838505268097, "time": 0.6308252334594726, "epoch": 4, "memory": 44120, "step": 12521} +{"lr": 7.875184945043487e-05, "data_time": 0.0014575004577636718, "loss": 0.3581499129533768, "time": 0.6374243974685669, "epoch": 4, "memory": 44120, "step": 12621} +{"lr": 7.844375189082012e-05, "data_time": 0.0013425350189208984, "loss": 0.36496379375457766, "time": 0.6331381559371948, "epoch": 4, "memory": 44120, "step": 12721} +{"lr": 7.813413854587572e-05, "data_time": 0.0014866828918457032, "loss": 0.35153523087501526, "time": 0.6354534387588501, "epoch": 4, "memory": 44120, "step": 12821} +{"lr": 7.782302943404291e-05, "data_time": 0.0014543771743774415, "loss": 0.3674694120883942, "time": 0.6352455615997314, "epoch": 4, "memory": 44120, "step": 12921} +{"lr": 7.751044467047337e-05, "data_time": 0.001568603515625, "loss": 0.3743774712085724, "time": 0.6339632272720337, "epoch": 4, "memory": 44120, "step": 13021} +{"lr": 7.719640446572907e-05, "data_time": 0.0015260696411132813, "loss": 0.3665180325508118, "time": 0.6332921743392944, "epoch": 4, "memory": 44120, "step": 13121} +{"lr": 7.688092912447536e-05, "data_time": 0.0016004562377929688, "loss": 0.3684597373008728, "time": 0.6426959276199341, "epoch": 4, "memory": 44120, "step": 13221} +{"lr": 7.656403904416794e-05, "data_time": 0.00150909423828125, "loss": 0.37329218685626986, "time": 0.6346177339553833, "epoch": 4, "memory": 44120, "step": 13321} +{"lr": 7.624575471373434e-05, "data_time": 0.0014724493026733398, "loss": 0.37726982235908507, "time": 0.6342240333557129, "epoch": 4, "memory": 44120, "step": 13421} +{"lr": 7.592609671224926e-05, "data_time": 0.0014622211456298828, "loss": 0.33714719116687775, "time": 0.6331625699996948, "epoch": 4, "memory": 44120, "step": 13521} +{"lr": 7.560508570760375e-05, "data_time": 0.001430368423461914, "loss": 0.3499594509601593, "time": 0.6319095849990845, "epoch": 4, "memory": 44120, "step": 13621} +{"lr": 7.528274245516895e-05, "data_time": 0.0015727758407592773, "loss": 0.3525034934282303, "time": 0.6307008028030395, "epoch": 4, "memory": 44120, "step": 13721} +{"lr": 7.495908779645405e-05, "data_time": 0.0014667510986328125, "loss": 0.3675632268190384, "time": 0.6310592651367187, "epoch": 4, "memory": 44120, "step": 13821} +{"lr": 7.463414265775906e-05, "data_time": 0.0015061378479003906, "loss": 0.3848166584968567, "time": 0.6312855243682861, "epoch": 4, "memory": 44120, "step": 13921} +{"lr": 7.430792804882147e-05, "data_time": 0.0015873908996582031, "loss": 0.3621916949748993, "time": 0.6312375068664551, "epoch": 4, "memory": 44120, "step": 14021} +{"lr": 7.398046506145803e-05, "data_time": 0.0015889644622802735, "loss": 0.37212550044059756, "time": 0.6326715230941773, "epoch": 4, "memory": 44120, "step": 14121} +{"lr": 7.365177486820114e-05, "data_time": 0.0016717195510864257, "loss": 0.37732054591178893, "time": 0.631341814994812, "epoch": 4, "memory": 44120, "step": 14221} +{"lr": 7.332187872092939e-05, "data_time": 0.0014044284820556641, "loss": 0.3810142785310745, "time": 0.6327552795410156, "epoch": 4, "memory": 44120, "step": 14321} +{"lr": 7.299079794949401e-05, "data_time": 0.001490950584411621, "loss": 0.3755857527256012, "time": 0.6313372135162354, "epoch": 4, "memory": 44120, "step": 14421} +{"lr": 7.265855396033955e-05, "data_time": 0.0014354228973388673, "loss": 0.39156965613365174, "time": 0.6319684267044068, "epoch": 4, "memory": 44120, "step": 14521} +{"lr": 7.232516823512013e-05, "data_time": 0.0015784740447998048, "loss": 0.37674275040626526, "time": 0.6314492702484131, "epoch": 4, "memory": 44120, "step": 14621} +{"lr": 7.199066232930995e-05, "data_time": 0.0014206171035766602, "loss": 0.3549411177635193, "time": 0.6332176208496094, "epoch": 4, "memory": 44120, "step": 14721} +{"lr": 7.165505787080997e-05, "data_time": 0.0014365434646606446, "loss": 0.3555349886417389, "time": 0.6344083309173584, "epoch": 4, "memory": 44120, "step": 14821} +{"lr": 7.13183765585495e-05, "data_time": 0.001679372787475586, "loss": 0.3396206796169281, "time": 0.6323187112808227, "epoch": 4, "memory": 44120, "step": 14921} +{"lr": 7.098064016108317e-05, "data_time": 0.001564335823059082, "loss": 0.3686327874660492, "time": 0.6322042226791382, "epoch": 4, "memory": 44120, "step": 15021} +{"lr": 7.064187051518379e-05, "data_time": 0.0013367652893066407, "loss": 0.3623316496610641, "time": 0.633063554763794, "epoch": 4, "memory": 44120, "step": 15121} +{"lr": 7.030208952442949e-05, "data_time": 0.0014116287231445313, "loss": 0.37482787668704987, "time": 0.6356353282928466, "epoch": 4, "memory": 44120, "step": 15221} +{"lr": 6.996131915778851e-05, "data_time": 0.001506638526916504, "loss": 0.35330727100372317, "time": 0.640656852722168, "epoch": 4, "memory": 44120, "step": 15321} +{"lr": 6.961958144819853e-05, "data_time": 0.002330636978149414, "loss": 0.36677582561969757, "time": 0.6364585876464843, "epoch": 4, "memory": 44120, "step": 15421} +{"lr": 6.927689849114193e-05, "data_time": 0.001501750946044922, "loss": 0.3694306015968323, "time": 0.6313318729400634, "epoch": 4, "memory": 44120, "step": 15521} +{"lr": 6.893329244321722e-05, "data_time": 0.001215219497680664, "loss": 0.3838209182024002, "time": 0.632353687286377, "epoch": 4, "memory": 44120, "step": 15621} +{"accuracy/top1": 76.9040756225586, "data_time": 0.03608762395792994, "time": 1.2779335194620594, "step": 4} +{"lr": 6.856463687361792e-05, "data_time": 0.0017266035079956054, "loss": 0.37894098162651063, "time": 0.632786226272583, "epoch": 5, "memory": 44120, "step": 15728} +{"lr": 6.821919068470432e-05, "data_time": 0.001751112937927246, "loss": 0.368763393163681, "time": 0.6366863489151001, "epoch": 5, "memory": 44120, "step": 15828} +{"lr": 6.787288979235407e-05, "data_time": 0.0018464088439941405, "loss": 0.3627344459295273, "time": 0.6328916549682617, "epoch": 5, "memory": 44120, "step": 15928} +{"lr": 6.752575658708793e-05, "data_time": 0.0017342329025268554, "loss": 0.360625758767128, "time": 0.6354114294052124, "epoch": 5, "memory": 44120, "step": 16028} +{"lr": 6.71778135132408e-05, "data_time": 0.0020811080932617186, "loss": 0.3682739704847336, "time": 0.6326449871063232, "epoch": 5, "memory": 44120, "step": 16128} +{"lr": 6.682908306751074e-05, "data_time": 0.0016627311706542969, "loss": 0.37629181444644927, "time": 0.6309634208679199, "epoch": 5, "memory": 44120, "step": 16228} +{"lr": 6.64795877975043e-05, "data_time": 0.001546168327331543, "loss": 0.35409157872200014, "time": 0.6317749977111816, "epoch": 5, "memory": 44120, "step": 16328} +{"lr": 6.61293503002787e-05, "data_time": 0.0018603801727294922, "loss": 0.3660582721233368, "time": 0.6322298049926758, "epoch": 5, "memory": 44120, "step": 16428} +{"lr": 6.577839322088084e-05, "data_time": 0.0014491796493530274, "loss": 0.359283447265625, "time": 0.6329627513885498, "epoch": 5, "memory": 44120, "step": 16528} +{"lr": 6.542673925088298e-05, "data_time": 0.0015192031860351562, "loss": 0.35167465209960935, "time": 0.6328529834747314, "epoch": 5, "memory": 44120, "step": 16628} +{"lr": 6.507441112691584e-05, "data_time": 0.0018984317779541016, "loss": 0.35347725749015807, "time": 0.6334426164627075, "epoch": 5, "memory": 44120, "step": 16728} +{"lr": 6.47214316291983e-05, "data_time": 0.0013723373413085938, "loss": 0.3700693666934967, "time": 0.631100058555603, "epoch": 5, "memory": 44120, "step": 16828} +{"lr": 6.436782358006481e-05, "data_time": 0.0015852212905883788, "loss": 0.3630936801433563, "time": 0.6312743902206421, "epoch": 5, "memory": 44120, "step": 16928} +{"lr": 6.40136098424896e-05, "data_time": 0.0014849424362182618, "loss": 0.36088619828224183, "time": 0.6330517292022705, "epoch": 5, "memory": 44120, "step": 17028} +{"lr": 6.365881331860828e-05, "data_time": 0.0015062570571899414, "loss": 0.37771726250648496, "time": 0.6318070411682128, "epoch": 5, "memory": 44120, "step": 17128} +{"lr": 6.330345694823739e-05, "data_time": 0.00149993896484375, "loss": 0.3599252313375473, "time": 0.6330976247787475, "epoch": 5, "memory": 44120, "step": 17228} +{"lr": 6.29475637073908e-05, "data_time": 0.0015309810638427734, "loss": 0.38419176936149596, "time": 0.6328319549560547, "epoch": 5, "memory": 44120, "step": 17328} +{"lr": 6.259115660679482e-05, "data_time": 0.0016766786575317383, "loss": 0.3584842085838318, "time": 0.6327142477035522, "epoch": 5, "memory": 44120, "step": 17428} +{"lr": 6.223425869039963e-05, "data_time": 0.0013804674148559571, "loss": 0.3820468544960022, "time": 0.6319448232650757, "epoch": 5, "memory": 44120, "step": 17528} +{"lr": 6.187689303389006e-05, "data_time": 0.0014548301696777344, "loss": 0.3519988000392914, "time": 0.6326465129852294, "epoch": 5, "memory": 44120, "step": 17628} +{"lr": 6.151908274319292e-05, "data_time": 0.0014504432678222657, "loss": 0.36450506150722506, "time": 0.6362475395202637, "epoch": 5, "memory": 44120, "step": 17728} +{"lr": 6.116085095298373e-05, "data_time": 0.0015873193740844726, "loss": 0.3727584004402161, "time": 0.6347034454345704, "epoch": 5, "memory": 44120, "step": 17828} +{"lr": 6.080222082519007e-05, "data_time": 0.0016533374786376954, "loss": 0.33923091292381286, "time": 0.632828950881958, "epoch": 5, "memory": 44120, "step": 17928} +{"lr": 6.0443215547495147e-05, "data_time": 0.001558065414428711, "loss": 0.35954704582691194, "time": 0.6335270404815674, "epoch": 5, "memory": 44120, "step": 18028} +{"lr": 6.0083858331837714e-05, "data_time": 0.001592254638671875, "loss": 0.3793952286243439, "time": 0.6341854333877563, "epoch": 5, "memory": 44120, "step": 18128} +{"lr": 5.97241724129116e-05, "data_time": 0.001589202880859375, "loss": 0.3658044457435608, "time": 0.63597412109375, "epoch": 5, "memory": 44120, "step": 18228} +{"lr": 5.936418104666328e-05, "data_time": 0.001482391357421875, "loss": 0.3590157270431519, "time": 0.6327776670455932, "epoch": 5, "memory": 44120, "step": 18328} +{"lr": 5.9003907508788295e-05, "data_time": 0.0015847921371459962, "loss": 0.3516092628240585, "time": 0.635569429397583, "epoch": 5, "memory": 44120, "step": 18428} +{"lr": 5.864337509322633e-05, "data_time": 0.0013956546783447266, "loss": 0.35428989231586455, "time": 0.6349320888519288, "epoch": 5, "memory": 44120, "step": 18528} +{"lr": 5.8282607110655194e-05, "data_time": 0.0016501426696777343, "loss": 0.3838335871696472, "time": 0.6330335855484008, "epoch": 5, "memory": 44120, "step": 18628} +{"lr": 5.7921626886983564e-05, "data_time": 0.0015996456146240234, "loss": 0.3755250334739685, "time": 0.6326020956039429, "epoch": 5, "memory": 44120, "step": 18728} +{"lr": 5.756045776184287e-05, "data_time": 0.0014918327331542968, "loss": 0.34004478752613065, "time": 0.6322643518447876, "epoch": 5, "memory": 44120, "step": 18828} +{"lr": 5.719912308707808e-05, "data_time": 0.001599264144897461, "loss": 0.3651239424943924, "time": 0.6325971603393554, "epoch": 5, "memory": 44120, "step": 18928} +{"lr": 5.683764622523808e-05, "data_time": 0.0015192031860351562, "loss": 0.35102906823158264, "time": 0.6324996232986451, "epoch": 5, "memory": 44120, "step": 19028} +{"lr": 5.647605054806502e-05, "data_time": 0.0014725923538208008, "loss": 0.3628747522830963, "time": 0.6368881225585937, "epoch": 5, "memory": 44120, "step": 19128} +{"lr": 5.6114359434983206e-05, "data_time": 0.0014778614044189454, "loss": 0.374749031662941, "time": 0.634670615196228, "epoch": 5, "memory": 44120, "step": 19228} +{"lr": 5.575259627158745e-05, "data_time": 0.0015050411224365235, "loss": 0.3750637054443359, "time": 0.6319854497909546, "epoch": 5, "memory": 44120, "step": 19328} +{"lr": 5.539078444813113e-05, "data_time": 0.0015236377716064454, "loss": 0.3660961240530014, "time": 0.6349586963653564, "epoch": 5, "memory": 44120, "step": 19428} +{"lr": 5.5028947358013735e-05, "data_time": 0.0014175653457641601, "loss": 0.3604292690753937, "time": 0.6347747325897217, "epoch": 5, "memory": 44120, "step": 19528} +{"accuracy/top1": 76.50743103027344, "data_time": 0.03666051091818974, "time": 1.2800816831917599, "step": 5} +{"lr": 5.464178020340108e-05, "data_time": 0.001437807083129883, "loss": 0.3572618395090103, "time": 0.632691502571106, "epoch": 6, "memory": 44120, "step": 19635} +{"lr": 5.427996514795046e-05, "data_time": 0.0016089200973510743, "loss": 0.37430056631565095, "time": 0.6319503307342529, "epoch": 6, "memory": 44120, "step": 19735} +{"lr": 5.3918196647260834e-05, "data_time": 0.0014183759689331056, "loss": 0.3624613046646118, "time": 0.6320508241653442, "epoch": 6, "memory": 44120, "step": 19835} +{"lr": 5.355649809193057e-05, "data_time": 0.001658797264099121, "loss": 0.3757551610469818, "time": 0.6331546783447266, "epoch": 6, "memory": 44120, "step": 19935} +{"lr": 5.3194892868035765e-05, "data_time": 0.001749873161315918, "loss": 0.35926517844200134, "time": 0.6371479272842407, "epoch": 6, "memory": 44120, "step": 20035} +{"lr": 5.283340435561795e-05, "data_time": 0.001473379135131836, "loss": 0.3537742137908936, "time": 0.6342557907104492, "epoch": 6, "memory": 44120, "step": 20135} +{"lr": 5.2472055927172495e-05, "data_time": 0.0015133380889892577, "loss": 0.3648320436477661, "time": 0.6345050811767579, "epoch": 6, "memory": 44120, "step": 20235} +{"lr": 5.2110870946137605e-05, "data_time": 0.0014802932739257813, "loss": 0.3618503898382187, "time": 0.6313145160675049, "epoch": 6, "memory": 44120, "step": 20335} +{"lr": 5.174987276538342e-05, "data_time": 0.0014394283294677734, "loss": 0.3696357190608978, "time": 0.632557487487793, "epoch": 6, "memory": 44120, "step": 20435} +{"lr": 5.1389084725702425e-05, "data_time": 0.001480436325073242, "loss": 0.3459394246339798, "time": 0.633821439743042, "epoch": 6, "memory": 44120, "step": 20535} +{"lr": 5.1028530154300045e-05, "data_time": 0.001661992073059082, "loss": 0.35268403589725494, "time": 0.6348913192749024, "epoch": 6, "memory": 44120, "step": 20635} +{"lr": 5.066823236328663e-05, "data_time": 0.0018307924270629882, "loss": 0.37245126664638517, "time": 0.6342251062393188, "epoch": 6, "memory": 44120, "step": 20735} +{"lr": 5.030821464817006e-05, "data_time": 0.0014332294464111327, "loss": 0.3455190181732178, "time": 0.632005262374878, "epoch": 6, "memory": 44120, "step": 20835} +{"lr": 4.99485002863494e-05, "data_time": 0.0014160871505737305, "loss": 0.3423906207084656, "time": 0.633011770248413, "epoch": 6, "memory": 44120, "step": 20935} +{"lr": 4.95891125356102e-05, "data_time": 0.0015711784362792969, "loss": 0.3789345592260361, "time": 0.6328719139099122, "epoch": 6, "memory": 44120, "step": 21035} +{"lr": 4.923007463262053e-05, "data_time": 0.0015725135803222657, "loss": 0.36813676357269287, "time": 0.6345274686813355, "epoch": 6, "memory": 44120, "step": 21135} +{"lr": 4.887140979142852e-05, "data_time": 0.0016031265258789062, "loss": 0.34439248144626616, "time": 0.6335875034332276, "epoch": 6, "memory": 44120, "step": 21235} +{"lr": 4.851314120196161e-05, "data_time": 0.0016109466552734375, "loss": 0.37726256549358367, "time": 0.6330404043197632, "epoch": 6, "memory": 44120, "step": 21335} +{"lr": 4.815529202852693e-05, "data_time": 0.0014396905899047852, "loss": 0.3832100421190262, "time": 0.6328764677047729, "epoch": 6, "memory": 44120, "step": 21435} +{"lr": 4.779788540831406e-05, "data_time": 0.0016587257385253906, "loss": 0.3852983593940735, "time": 0.6330849647521972, "epoch": 6, "memory": 44120, "step": 21535} +{"lr": 4.744094444989843e-05, "data_time": 0.0016937494277954102, "loss": 0.36328046321868895, "time": 0.6359888553619385, "epoch": 6, "memory": 44120, "step": 21635} +{"lr": 4.708449223174764e-05, "data_time": 0.001470804214477539, "loss": 0.37881985306739807, "time": 0.633746862411499, "epoch": 6, "memory": 44120, "step": 21735} +{"lr": 4.672855180072916e-05, "data_time": 0.001506209373474121, "loss": 0.3463868647813797, "time": 0.6335466623306274, "epoch": 6, "memory": 44120, "step": 21835} +{"lr": 4.6373146170620256e-05, "data_time": 0.0015089035034179688, "loss": 0.36230126619338987, "time": 0.6332391977310181, "epoch": 6, "memory": 44120, "step": 21935} +{"lr": 4.6018298320619915e-05, "data_time": 0.0013457298278808593, "loss": 0.3581427901983261, "time": 0.6315054416656494, "epoch": 6, "memory": 44120, "step": 22035} +{"lr": 4.566403119386318e-05, "data_time": 0.0014329671859741211, "loss": 0.36872154772281646, "time": 0.6338589429855347, "epoch": 6, "memory": 44120, "step": 22135} +{"lr": 4.531036769593769e-05, "data_time": 0.0014895200729370117, "loss": 0.3592005521059036, "time": 0.6327705383300781, "epoch": 6, "memory": 44120, "step": 22235} +{"lr": 4.4957330693402685e-05, "data_time": 0.001523733139038086, "loss": 0.34864698350429535, "time": 0.6323814392089844, "epoch": 6, "memory": 44120, "step": 22335} +{"lr": 4.460494301231059e-05, "data_time": 0.001468992233276367, "loss": 0.3588447093963623, "time": 0.6320381879806518, "epoch": 6, "memory": 44120, "step": 22435} +{"lr": 4.425322743673109e-05, "data_time": 0.0015026330947875977, "loss": 0.3461219519376755, "time": 0.6336755514144897, "epoch": 6, "memory": 44120, "step": 22535} +{"lr": 4.390220670727812e-05, "data_time": 0.0014482498168945312, "loss": 0.3664976954460144, "time": 0.6330651044845581, "epoch": 6, "memory": 44120, "step": 22635} +{"lr": 4.3551903519639325e-05, "data_time": 0.0015913963317871094, "loss": 0.37305538058280946, "time": 0.6312130451202392, "epoch": 6, "memory": 44120, "step": 22735} +{"lr": 4.3202340523108894e-05, "data_time": 0.0014281749725341796, "loss": 0.36498062312602997, "time": 0.6307538270950317, "epoch": 6, "memory": 44120, "step": 22835} +{"lr": 4.285354031912302e-05, "data_time": 0.0016157388687133788, "loss": 0.36128232777118685, "time": 0.6321924924850464, "epoch": 6, "memory": 44120, "step": 22935} +{"lr": 4.2505525459798544e-05, "data_time": 0.0014148950576782227, "loss": 0.3610009491443634, "time": 0.6320352554321289, "epoch": 6, "memory": 44120, "step": 23035} +{"lr": 4.2158318446474836e-05, "data_time": 0.0014661073684692383, "loss": 0.3349186420440674, "time": 0.6319817304611206, "epoch": 6, "memory": 44120, "step": 23135} +{"lr": 4.181194172825905e-05, "data_time": 0.0016401052474975587, "loss": 0.3892023950815201, "time": 0.6348764657974243, "epoch": 6, "memory": 44120, "step": 23235} +{"lr": 4.146641770057448e-05, "data_time": 0.0015258073806762695, "loss": 0.3624118328094482, "time": 0.6314297199249268, "epoch": 6, "memory": 44120, "step": 23335} +{"lr": 4.112176870371267e-05, "data_time": 0.0012554168701171876, "loss": 0.35829317569732666, "time": 0.6330458402633667, "epoch": 6, "memory": 44120, "step": 23435} +{"accuracy/top1": 76.23603820800781, "data_time": 0.035895705223083496, "time": 1.2793013720676816, "step": 6} +{"lr": 4.0753988582732786e-05, "data_time": 0.0015159130096435546, "loss": 0.36123780608177186, "time": 0.6334625482559204, "epoch": 7, "memory": 44120, "step": 23542} +{"lr": 4.0411221639286815e-05, "data_time": 0.0013533830642700195, "loss": 0.3564772456884384, "time": 0.6335180282592774, "epoch": 7, "memory": 44120, "step": 23642} +{"lr": 4.006939795169518e-05, "data_time": 0.0015457868576049805, "loss": 0.3629548907279968, "time": 0.6356516361236573, "epoch": 7, "memory": 44120, "step": 23742} +{"lr": 3.972853962099913e-05, "data_time": 0.001765275001525879, "loss": 0.390253359079361, "time": 0.635542917251587, "epoch": 7, "memory": 44120, "step": 23842} +{"lr": 3.93886686858237e-05, "data_time": 0.0022885799407958984, "loss": 0.3424901574850082, "time": 0.6346344709396362, "epoch": 7, "memory": 44120, "step": 23942} +{"lr": 3.904980712095259e-05, "data_time": 0.0013433456420898437, "loss": 0.3503723829984665, "time": 0.6340196847915649, "epoch": 7, "memory": 44120, "step": 24042} +{"lr": 3.871197683590748e-05, "data_time": 0.0013974428176879883, "loss": 0.3757956862449646, "time": 0.6357422828674316, "epoch": 7, "memory": 44120, "step": 24142} +{"lr": 3.837519967353125e-05, "data_time": 0.0015516042709350585, "loss": 0.34941319525241854, "time": 0.6345579147338867, "epoch": 7, "memory": 44120, "step": 24242} +{"lr": 3.8039497408575855e-05, "data_time": 0.001452493667602539, "loss": 0.3754796594381332, "time": 0.6326464414596558, "epoch": 7, "memory": 44120, "step": 24342} +{"lr": 3.770489174629432e-05, "data_time": 0.0014765262603759766, "loss": 0.36980322003364563, "time": 0.6335984945297242, "epoch": 7, "memory": 44120, "step": 24442} +{"lr": 3.737140432103754e-05, "data_time": 0.0015181779861450195, "loss": 0.3552908211946487, "time": 0.634049916267395, "epoch": 7, "memory": 44120, "step": 24542} +{"lr": 3.703905669485545e-05, "data_time": 0.001567983627319336, "loss": 0.36363907158374786, "time": 0.6317353010177612, "epoch": 7, "memory": 44120, "step": 24642} +{"lr": 3.6707870356102706e-05, "data_time": 0.0017004013061523438, "loss": 0.35386523604393005, "time": 0.6334615230560303, "epoch": 7, "memory": 44120, "step": 24742} +{"lr": 3.6377866718049484e-05, "data_time": 0.0014855384826660157, "loss": 0.35835925936698915, "time": 0.6328764438629151, "epoch": 7, "memory": 44120, "step": 24842} +{"lr": 3.6049067117496965e-05, "data_time": 0.0016066074371337891, "loss": 0.3607875853776932, "time": 0.6357440948486328, "epoch": 7, "memory": 44120, "step": 24942} +{"lr": 3.5721492813397755e-05, "data_time": 0.001550602912902832, "loss": 0.3488959699869156, "time": 0.6320597410202027, "epoch": 7, "memory": 44120, "step": 25042} +{"lr": 3.539516498548151e-05, "data_time": 0.0013962984085083008, "loss": 0.3506907194852829, "time": 0.6320287227630615, "epoch": 7, "memory": 44120, "step": 25142} +{"lr": 3.507010473288518e-05, "data_time": 0.0015010356903076172, "loss": 0.35058234333992006, "time": 0.6400053262710571, "epoch": 7, "memory": 44120, "step": 25242} +{"lr": 3.474633307278909e-05, "data_time": 0.0014966964721679688, "loss": 0.36881363689899443, "time": 0.6330070734024048, "epoch": 7, "memory": 44120, "step": 25342} +{"lr": 3.4423870939058177e-05, "data_time": 0.0014649391174316405, "loss": 0.3741642564535141, "time": 0.6326273202896118, "epoch": 7, "memory": 44120, "step": 25442} +{"lr": 3.410273918088805e-05, "data_time": 0.0014224290847778321, "loss": 0.34705342054367067, "time": 0.6346636295318604, "epoch": 7, "memory": 44120, "step": 25542} +{"lr": 3.37829585614574e-05, "data_time": 0.0014030933380126953, "loss": 0.36638697385787966, "time": 0.6324271440505982, "epoch": 7, "memory": 44120, "step": 25642} +{"lr": 3.3464549756585216e-05, "data_time": 0.0014451503753662109, "loss": 0.3701684415340424, "time": 0.6325154781341553, "epoch": 7, "memory": 44120, "step": 25742} +{"lr": 3.314753335339416e-05, "data_time": 0.0013575315475463866, "loss": 0.3597903877496719, "time": 0.6339298725128174, "epoch": 7, "memory": 44120, "step": 25842} +{"lr": 3.2831929848979394e-05, "data_time": 0.0015400886535644532, "loss": 0.3604041874408722, "time": 0.6329383611679077, "epoch": 7, "memory": 44120, "step": 25942} +{"lr": 3.251775964908332e-05, "data_time": 0.0014547109603881836, "loss": 0.35140136778354647, "time": 0.6328030586242676, "epoch": 7, "memory": 44120, "step": 26042} +{"lr": 3.22050430667762e-05, "data_time": 0.0014820098876953125, "loss": 0.3374085336923599, "time": 0.6335373163223267, "epoch": 7, "memory": 44120, "step": 26142} +{"lr": 3.189380032114285e-05, "data_time": 0.0014576911926269531, "loss": 0.38166150748729705, "time": 0.634656023979187, "epoch": 7, "memory": 44120, "step": 26242} +{"lr": 3.1584051535975324e-05, "data_time": 0.0013989448547363282, "loss": 0.36053734123706815, "time": 0.633669090270996, "epoch": 7, "memory": 44120, "step": 26342} +{"lr": 3.127581673847175e-05, "data_time": 0.0015028715133666992, "loss": 0.386484768986702, "time": 0.6345329284667969, "epoch": 7, "memory": 44120, "step": 26442} +{"lr": 3.0969115857941555e-05, "data_time": 0.001757669448852539, "loss": 0.36978948414325713, "time": 0.6333515167236328, "epoch": 7, "memory": 44120, "step": 26542} +{"lr": 3.0663968724516654e-05, "data_time": 0.001509547233581543, "loss": 0.36772051453590393, "time": 0.6349242448806762, "epoch": 7, "memory": 44120, "step": 26642} +{"lr": 3.0360395067869668e-05, "data_time": 0.0015058517456054688, "loss": 0.3541785299777985, "time": 0.6327594518661499, "epoch": 7, "memory": 44120, "step": 26742} +{"lr": 3.0058414515938063e-05, "data_time": 0.0014070749282836914, "loss": 0.3812207102775574, "time": 0.6335640907287597, "epoch": 7, "memory": 44120, "step": 26842} +{"lr": 2.975804659365501e-05, "data_time": 0.001426529884338379, "loss": 0.36305170953273774, "time": 0.6338283777236938, "epoch": 7, "memory": 44120, "step": 26942} +{"lr": 2.945931072168714e-05, "data_time": 0.001401662826538086, "loss": 0.3763676553964615, "time": 0.6344704627990723, "epoch": 7, "memory": 44120, "step": 27042} +{"lr": 2.9162226215178855e-05, "data_time": 0.001462864875793457, "loss": 0.3519167512655258, "time": 0.6332732200622558, "epoch": 7, "memory": 44120, "step": 27142} +{"lr": 2.8866812282503322e-05, "data_time": 0.0015109777450561523, "loss": 0.3715108722448349, "time": 0.6373698472976684, "epoch": 7, "memory": 44120, "step": 27242} +{"lr": 2.8573088024020838e-05, "data_time": 0.0013585090637207031, "loss": 0.38868110477924345, "time": 0.631889820098877, "epoch": 7, "memory": 44120, "step": 27342} +{"accuracy/top1": 75.78024291992188, "data_time": 0.03545153552088244, "time": 1.2770294649847622, "step": 7} +{"lr": 2.8260695817390092e-05, "data_time": 0.0015898466110229493, "loss": 0.37303454279899595, "time": 0.6339924335479736, "epoch": 8, "memory": 44120, "step": 27449} +{"lr": 2.7970529402570967e-05, "data_time": 0.0015257596969604492, "loss": 0.35515395402908323, "time": 0.6342122316360473, "epoch": 8, "memory": 44120, "step": 27549} +{"lr": 2.7682110612244214e-05, "data_time": 0.001466989517211914, "loss": 0.3770622432231903, "time": 0.6325371742248536, "epoch": 8, "memory": 44120, "step": 27649} +{"lr": 2.7395458094490283e-05, "data_time": 0.0017443418502807618, "loss": 0.34302699863910674, "time": 0.6348450660705567, "epoch": 8, "memory": 44120, "step": 27749} +{"lr": 2.7110590383189015e-05, "data_time": 0.0014403343200683593, "loss": 0.393852236866951, "time": 0.6362323045730591, "epoch": 8, "memory": 44120, "step": 27849} +{"lr": 2.6827525896821397e-05, "data_time": 0.001551651954650879, "loss": 0.3731550931930542, "time": 0.6309094905853272, "epoch": 8, "memory": 44120, "step": 27949} +{"lr": 2.6546282937278552e-05, "data_time": 0.0016638517379760742, "loss": 0.3527612864971161, "time": 0.6348623752593994, "epoch": 8, "memory": 44120, "step": 28049} +{"lr": 2.6266879688678606e-05, "data_time": 0.0015616178512573241, "loss": 0.35313935577869415, "time": 0.6329471588134765, "epoch": 8, "memory": 44120, "step": 28149} +{"lr": 2.5989334216190738e-05, "data_time": 0.0015506982803344727, "loss": 0.37757874727249147, "time": 0.632073450088501, "epoch": 8, "memory": 44120, "step": 28249} +{"lr": 2.571366446486732e-05, "data_time": 0.0014993906021118163, "loss": 0.37557385563850404, "time": 0.6349147319793701, "epoch": 8, "memory": 44120, "step": 28349} +{"lr": 2.5439888258483665e-05, "data_time": 0.001457357406616211, "loss": 0.3426795959472656, "time": 0.6343680381774902, "epoch": 8, "memory": 44120, "step": 28449} +{"lr": 2.5168023298385384e-05, "data_time": 0.0014397621154785157, "loss": 0.3696870893239975, "time": 0.6340624332427979, "epoch": 8, "memory": 44120, "step": 28549} +{"lr": 2.489808716234412e-05, "data_time": 0.0014007806777954102, "loss": 0.35144375562667846, "time": 0.6346299171447753, "epoch": 8, "memory": 44120, "step": 28649} +{"lr": 2.463009730342111e-05, "data_time": 0.0015766382217407226, "loss": 0.3776319772005081, "time": 0.6337181568145752, "epoch": 8, "memory": 44120, "step": 28749} +{"lr": 2.436407104883849e-05, "data_time": 0.0014631271362304688, "loss": 0.37245858907699586, "time": 0.6334508895874024, "epoch": 8, "memory": 44120, "step": 28849} +{"lr": 2.4100025598859092e-05, "data_time": 0.0014869451522827148, "loss": 0.37227138578891755, "time": 0.633303451538086, "epoch": 8, "memory": 44120, "step": 28949} +{"lr": 2.3837978025674284e-05, "data_time": 0.0013957262039184571, "loss": 0.3901132047176361, "time": 0.6316985130310059, "epoch": 8, "memory": 44120, "step": 29049} +{"lr": 2.357794527230027e-05, "data_time": 0.0013980388641357422, "loss": 0.3596083104610443, "time": 0.6336965084075927, "epoch": 8, "memory": 44120, "step": 29149} +{"lr": 2.33199441514825e-05, "data_time": 0.0014456748962402345, "loss": 0.37027046382427214, "time": 0.636094617843628, "epoch": 8, "memory": 44120, "step": 29249} +{"lr": 2.306399134460859e-05, "data_time": 0.0014680862426757813, "loss": 0.3530574142932892, "time": 0.6320498704910278, "epoch": 8, "memory": 44120, "step": 29349} +{"lr": 2.281010340062995e-05, "data_time": 0.0015579462051391602, "loss": 0.3571125894784927, "time": 0.6362739562988281, "epoch": 8, "memory": 44120, "step": 29449} +{"lr": 2.2558296734991603e-05, "data_time": 0.001444101333618164, "loss": 0.40539585053920746, "time": 0.6339248895645142, "epoch": 8, "memory": 44120, "step": 29549} +{"lr": 2.2308587628570915e-05, "data_time": 0.0016648292541503907, "loss": 0.37943858206272124, "time": 0.63364098072052, "epoch": 8, "memory": 44120, "step": 29649} +{"lr": 2.2060992226624955e-05, "data_time": 0.0014145851135253906, "loss": 0.37945939898490905, "time": 0.6324527978897094, "epoch": 8, "memory": 44120, "step": 29749} +{"lr": 2.181552653774648e-05, "data_time": 0.0014185428619384766, "loss": 0.35202763974666595, "time": 0.6372021913528443, "epoch": 8, "memory": 44120, "step": 29849} +{"lr": 2.1572206432829035e-05, "data_time": 0.0019323348999023438, "loss": 0.37268022298812864, "time": 0.6352515935897827, "epoch": 8, "memory": 44120, "step": 29949} +{"lr": 2.1331047644040754e-05, "data_time": 0.0016809225082397461, "loss": 0.3668335735797882, "time": 0.6354601621627808, "epoch": 8, "memory": 44120, "step": 30049} +{"lr": 2.109206576380705e-05, "data_time": 0.001648259162902832, "loss": 0.34897190928459165, "time": 0.6384757995605469, "epoch": 8, "memory": 44120, "step": 30149} +{"lr": 2.0855276243802687e-05, "data_time": 0.001694965362548828, "loss": 0.36093118190765383, "time": 0.6328551769256592, "epoch": 8, "memory": 44120, "step": 30249} +{"lr": 2.0620694393952487e-05, "data_time": 0.0014263153076171874, "loss": 0.35270750522613525, "time": 0.6337556123733521, "epoch": 8, "memory": 44120, "step": 30349} +{"lr": 2.038833538144169e-05, "data_time": 0.0014792442321777343, "loss": 0.3726002395153046, "time": 0.6359061241149903, "epoch": 8, "memory": 44120, "step": 30449} +{"lr": 2.0158214229735154e-05, "data_time": 0.00164031982421875, "loss": 0.3795700967311859, "time": 0.6336644411087036, "epoch": 8, "memory": 44120, "step": 30549} +{"lr": 1.9930345817605977e-05, "data_time": 0.0015683650970458984, "loss": 0.35870579779148104, "time": 0.6432867288589478, "epoch": 8, "memory": 44120, "step": 30649} +{"lr": 1.970474487817358e-05, "data_time": 0.0014643192291259766, "loss": 0.3907363682985306, "time": 0.6326995849609375, "epoch": 8, "memory": 44120, "step": 30749} +{"lr": 1.9481425997951092e-05, "data_time": 0.0015496253967285157, "loss": 0.39445521533489225, "time": 0.6333011150360107, "epoch": 8, "memory": 44120, "step": 30849} +{"lr": 1.9260403615902187e-05, "data_time": 0.0015614032745361328, "loss": 0.355153751373291, "time": 0.6346783638000488, "epoch": 8, "memory": 44120, "step": 30949} +{"lr": 1.9041692022507647e-05, "data_time": 0.001454329490661621, "loss": 0.354151850938797, "time": 0.6324937582015991, "epoch": 8, "memory": 44120, "step": 31049} +{"lr": 1.8825305358841125e-05, "data_time": 0.0018217325210571288, "loss": 0.37641663253307345, "time": 0.6365252733230591, "epoch": 8, "memory": 44120, "step": 31149} +{"lr": 1.861125761565518e-05, "data_time": 0.0012953758239746093, "loss": 0.35719003081321715, "time": 0.6318187713623047, "epoch": 8, "memory": 44120, "step": 31249} +{"accuracy/top1": 75.78024291992188, "data_time": 0.03583048952036891, "time": 1.277594253934663, "step": 8} +{"lr": 1.8384832449122864e-05, "data_time": 0.0015422582626342773, "loss": 0.3495414197444916, "time": 0.6344014883041382, "epoch": 9, "memory": 44120, "step": 31356} +{"lr": 1.817567007332114e-05, "data_time": 0.0013895273208618165, "loss": 0.35953572690486907, "time": 0.6375298976898194, "epoch": 9, "memory": 44120, "step": 31456} +{"lr": 1.796888862098825e-05, "data_time": 0.002831554412841797, "loss": 0.36801652014255526, "time": 0.635310435295105, "epoch": 9, "memory": 44120, "step": 31556} +{"lr": 1.7764501461839678e-05, "data_time": 0.001604008674621582, "loss": 0.371783384680748, "time": 0.6321839570999146, "epoch": 9, "memory": 44120, "step": 31656} +{"lr": 1.7562521810784746e-05, "data_time": 0.0017839908599853516, "loss": 0.36798123121261594, "time": 0.6358216762542724, "epoch": 9, "memory": 44120, "step": 31756} +{"lr": 1.7362962727072444e-05, "data_time": 0.0015273809432983399, "loss": 0.36750137209892275, "time": 0.634696364402771, "epoch": 9, "memory": 44120, "step": 31856} +{"lr": 1.716583711344682e-05, "data_time": 0.0014613866806030273, "loss": 0.38259629607200624, "time": 0.6332085132598877, "epoch": 9, "memory": 44120, "step": 31956} +{"lr": 1.6971157715312954e-05, "data_time": 0.001433110237121582, "loss": 0.363897243142128, "time": 0.6333233833312988, "epoch": 9, "memory": 44120, "step": 32056} +{"lr": 1.677893711991274e-05, "data_time": 0.0014777421951293946, "loss": 0.3699881464242935, "time": 0.6410775423049927, "epoch": 9, "memory": 44120, "step": 32156} +{"lr": 1.6589187755511025e-05, "data_time": 0.0014045000076293944, "loss": 0.3458457738161087, "time": 0.6346138715744019, "epoch": 9, "memory": 44120, "step": 32256} +{"lr": 1.640192189059216e-05, "data_time": 0.001602315902709961, "loss": 0.35345048606395724, "time": 0.6327174663543701, "epoch": 9, "memory": 44120, "step": 32356} +{"lr": 1.6217151633066716e-05, "data_time": 0.0016676664352416992, "loss": 0.35717675387859343, "time": 0.6365456581115723, "epoch": 9, "memory": 44120, "step": 32456} +{"lr": 1.6034888929488564e-05, "data_time": 0.001959061622619629, "loss": 0.34748610854148865, "time": 0.634586763381958, "epoch": 9, "memory": 44120, "step": 32556} +{"lr": 1.585514556428258e-05, "data_time": 0.0017581224441528321, "loss": 0.37288335859775545, "time": 0.6336165189743042, "epoch": 9, "memory": 44120, "step": 32656} +{"lr": 1.567793315898261e-05, "data_time": 0.0015058040618896485, "loss": 0.3559491068124771, "time": 0.6327862977981568, "epoch": 9, "memory": 44120, "step": 32756} +{"lr": 1.5503263171480048e-05, "data_time": 0.0016260862350463868, "loss": 0.35945734977722166, "time": 0.6340567588806152, "epoch": 9, "memory": 44120, "step": 32856} +{"lr": 1.5331146895283165e-05, "data_time": 0.0015779733657836914, "loss": 0.35524179339408873, "time": 0.6357394933700562, "epoch": 9, "memory": 44120, "step": 32956} +{"lr": 1.5161595458786728e-05, "data_time": 0.0017821550369262694, "loss": 0.35637607872486116, "time": 0.6328525066375732, "epoch": 9, "memory": 44120, "step": 33056} +{"lr": 1.4994619824552563e-05, "data_time": 0.0016401052474975587, "loss": 0.35372823774814605, "time": 0.6347967624664307, "epoch": 9, "memory": 44120, "step": 33156} +{"lr": 1.4830230788600802e-05, "data_time": 0.001634073257446289, "loss": 0.3664750546216965, "time": 0.6333654403686524, "epoch": 9, "memory": 44120, "step": 33256} +{"lr": 1.4668438979711779e-05, "data_time": 0.0015042781829833984, "loss": 0.38912451863288877, "time": 0.6319226503372193, "epoch": 9, "memory": 44120, "step": 33356} +{"lr": 1.4509254858738844e-05, "data_time": 0.001482391357421875, "loss": 0.3958167463541031, "time": 0.6325299978256226, "epoch": 9, "memory": 44120, "step": 33456} +{"lr": 1.4352688717931966e-05, "data_time": 0.001739835739135742, "loss": 0.3434902191162109, "time": 0.6332091331481934, "epoch": 9, "memory": 44120, "step": 33556} +{"lr": 1.4198750680272348e-05, "data_time": 0.0016980171203613281, "loss": 0.3578742653131485, "time": 0.6342315196990966, "epoch": 9, "memory": 44120, "step": 33656} +{"lr": 1.4047450698817857e-05, "data_time": 0.0016282558441162109, "loss": 0.38665358126163485, "time": 0.6343876600265503, "epoch": 9, "memory": 44120, "step": 33756} +{"lr": 1.389879855605952e-05, "data_time": 0.0013923883438110352, "loss": 0.37287634015083315, "time": 0.6369168996810913, "epoch": 9, "memory": 44120, "step": 33856} +{"lr": 1.3752803863288976e-05, "data_time": 0.0016299009323120118, "loss": 0.38205778896808623, "time": 0.6344584226608276, "epoch": 9, "memory": 44120, "step": 33956} +{"lr": 1.3609476059977129e-05, "data_time": 0.0015775203704833985, "loss": 0.35270113348960874, "time": 0.6340476036071777, "epoch": 9, "memory": 44120, "step": 34056} +{"lr": 1.3468824413163754e-05, "data_time": 0.0013777256011962891, "loss": 0.36054650843143465, "time": 0.6322983026504516, "epoch": 9, "memory": 44120, "step": 34156} +{"lr": 1.3330858016858395e-05, "data_time": 0.001448678970336914, "loss": 0.38119986951351165, "time": 0.6333877801895141, "epoch": 9, "memory": 44120, "step": 34256} +{"lr": 1.3195585791452309e-05, "data_time": 0.0014369487762451172, "loss": 0.38305503129959106, "time": 0.6359586954116822, "epoch": 9, "memory": 44120, "step": 34356} +{"lr": 1.3063016483141729e-05, "data_time": 0.0013981342315673828, "loss": 0.3698997378349304, "time": 0.6355183124542236, "epoch": 9, "memory": 44120, "step": 34456} +{"lr": 1.2933158663362388e-05, "data_time": 0.0014146804809570313, "loss": 0.3398908317089081, "time": 0.6350657701492309, "epoch": 9, "memory": 44120, "step": 34556} +{"lr": 1.2806020728235311e-05, "data_time": 0.0015672922134399414, "loss": 0.37949779331684114, "time": 0.6337861537933349, "epoch": 9, "memory": 44120, "step": 34656} +{"lr": 1.2681610898023963e-05, "data_time": 0.0016001224517822265, "loss": 0.36143686771392824, "time": 0.6342676877975464, "epoch": 9, "memory": 44120, "step": 34756} +{"lr": 1.2559937216602731e-05, "data_time": 0.0014892339706420899, "loss": 0.3563768804073334, "time": 0.6327040672302247, "epoch": 9, "memory": 44120, "step": 34856} +{"lr": 1.2441007550936847e-05, "data_time": 0.001398324966430664, "loss": 0.3814793139696121, "time": 0.6365732669830322, "epoch": 9, "memory": 44120, "step": 34956} +{"lr": 1.232482959057377e-05, "data_time": 0.0014713525772094727, "loss": 0.3600014805793762, "time": 0.6337686538696289, "epoch": 9, "memory": 44120, "step": 35056} +{"lr": 1.2211410847145969e-05, "data_time": 0.0012830972671508789, "loss": 0.34628230333328247, "time": 0.6359486818313599, "epoch": 9, "memory": 44120, "step": 35156} +{"accuracy/top1": 75.58540344238281, "data_time": 0.035728594352459085, "time": 1.2794173380424236, "step": 9} +{"lr": 1.209311679360931e-05, "data_time": 0.001469564437866211, "loss": 0.339600196480751, "time": 0.6344933986663819, "epoch": 10, "memory": 44120, "step": 35263} +{"lr": 1.1985432727488666e-05, "data_time": 0.0016040325164794922, "loss": 0.3448700666427612, "time": 0.6342395782470703, "epoch": 10, "memory": 44120, "step": 35363} +{"lr": 1.1880529822435289e-05, "data_time": 0.0015718460083007813, "loss": 0.37429238557815553, "time": 0.6348059892654419, "epoch": 10, "memory": 44120, "step": 35463} +{"lr": 1.1778414861078726e-05, "data_time": 0.0016747713088989258, "loss": 0.35528212785720825, "time": 0.6322180747985839, "epoch": 10, "memory": 44120, "step": 35563} +{"lr": 1.1679094445790538e-05, "data_time": 0.001582789421081543, "loss": 0.39299889504909513, "time": 0.6315577983856201, "epoch": 10, "memory": 44120, "step": 35663} +{"lr": 1.1582574998257377e-05, "data_time": 0.001683497428894043, "loss": 0.3906119465827942, "time": 0.6320613145828247, "epoch": 10, "memory": 44120, "step": 35763} +{"lr": 1.1488862759065762e-05, "data_time": 0.0015350580215454102, "loss": 0.36365343630313873, "time": 0.6355814218521119, "epoch": 10, "memory": 44120, "step": 35863} +{"lr": 1.1397963787298665e-05, "data_time": 0.001596379280090332, "loss": 0.3629516690969467, "time": 0.6324075937271119, "epoch": 10, "memory": 44120, "step": 35963} +{"lr": 1.1309883960143664e-05, "data_time": 0.0014279842376708984, "loss": 0.3542255997657776, "time": 0.6352359771728515, "epoch": 10, "memory": 44120, "step": 36063} +{"lr": 1.122462897251302e-05, "data_time": 0.0016389846801757812, "loss": 0.3491236180067062, "time": 0.6526760578155517, "epoch": 10, "memory": 44120, "step": 36163} +{"lr": 1.114220433667542e-05, "data_time": 0.0014959096908569336, "loss": 0.3740518867969513, "time": 0.6347596406936645, "epoch": 10, "memory": 44120, "step": 36263} +{"lr": 1.10626153818996e-05, "data_time": 0.0015249252319335938, "loss": 0.3636352241039276, "time": 0.6325485467910766, "epoch": 10, "memory": 44120, "step": 36363} +{"lr": 1.0985867254109761e-05, "data_time": 0.001751875877380371, "loss": 0.3938641458749771, "time": 0.6334201097488403, "epoch": 10, "memory": 44120, "step": 36463} +{"lr": 1.0911964915552823e-05, "data_time": 0.0019232273101806641, "loss": 0.34288425743579865, "time": 0.631958794593811, "epoch": 10, "memory": 44120, "step": 36563} +{"lr": 1.0840913144477643e-05, "data_time": 0.0015038013458251952, "loss": 0.36984099447727203, "time": 0.6352145433425903, "epoch": 10, "memory": 44120, "step": 36663} +{"lr": 1.0772716534826064e-05, "data_time": 0.0015038490295410157, "loss": 0.3477191925048828, "time": 0.6329223394393921, "epoch": 10, "memory": 44120, "step": 36763} +{"lr": 1.0707379495935828e-05, "data_time": 0.0015514373779296875, "loss": 0.34641546905040743, "time": 0.6327091217041015, "epoch": 10, "memory": 44120, "step": 36863} +{"lr": 1.0644906252255541e-05, "data_time": 0.0017029047012329102, "loss": 0.39467841386795044, "time": 0.6321772813796998, "epoch": 10, "memory": 44120, "step": 36963} +{"lr": 1.0585300843071527e-05, "data_time": 0.0015934467315673827, "loss": 0.39513387382030485, "time": 0.6448676824569702, "epoch": 10, "memory": 44120, "step": 37063} +{"lr": 1.0528567122246654e-05, "data_time": 0.0015032768249511718, "loss": 0.3685428321361542, "time": 0.6345503091812134, "epoch": 10, "memory": 44120, "step": 37163} +{"lr": 1.0474708757971141e-05, "data_time": 0.001501607894897461, "loss": 0.3467086672782898, "time": 0.634011173248291, "epoch": 10, "memory": 44120, "step": 37263} +{"lr": 1.042372923252543e-05, "data_time": 0.0015008926391601562, "loss": 0.3592089474201202, "time": 0.6336398124694824, "epoch": 10, "memory": 44120, "step": 37363} +{"lr": 1.0375631842054998e-05, "data_time": 0.001678323745727539, "loss": 0.3563314974308014, "time": 0.634985375404358, "epoch": 10, "memory": 44120, "step": 37463} +{"lr": 1.0330419696357254e-05, "data_time": 0.0017746448516845702, "loss": 0.33439682722091674, "time": 0.6366176128387451, "epoch": 10, "memory": 44120, "step": 37563} +{"lr": 1.028809571868047e-05, "data_time": 0.0015472173690795898, "loss": 0.34144375622272494, "time": 0.6344694852828979, "epoch": 10, "memory": 44120, "step": 37663} +{"lr": 1.0248662645534766e-05, "data_time": 0.0017197608947753906, "loss": 0.3623490482568741, "time": 0.6318809509277343, "epoch": 10, "memory": 44120, "step": 37763} +{"lr": 1.0212123026515194e-05, "data_time": 0.001902318000793457, "loss": 0.36674031615257263, "time": 0.636697244644165, "epoch": 10, "memory": 44120, "step": 37863} +{"lr": 1.0178479224136881e-05, "data_time": 0.001537632942199707, "loss": 0.3548463523387909, "time": 0.6348671197891236, "epoch": 10, "memory": 44120, "step": 37963} +{"lr": 1.0147733413682289e-05, "data_time": 0.001559305191040039, "loss": 0.3793183296918869, "time": 0.6346055507659912, "epoch": 10, "memory": 44120, "step": 38063} +{"lr": 1.0119887583060527e-05, "data_time": 0.0013947725296020509, "loss": 0.36996032297611237, "time": 0.6342608213424683, "epoch": 10, "memory": 44120, "step": 38163} +{"lr": 1.0094943532678871e-05, "data_time": 0.0013764142990112306, "loss": 0.39604129195213317, "time": 0.6383871078491211, "epoch": 10, "memory": 44120, "step": 38263} +{"lr": 1.007290287532637e-05, "data_time": 0.0021173954010009766, "loss": 0.36310293078422545, "time": 0.6334275960922241, "epoch": 10, "memory": 44120, "step": 38363} +{"lr": 1.0053767036069495e-05, "data_time": 0.0017472028732299805, "loss": 0.38397795557975767, "time": 0.6343852758407593, "epoch": 10, "memory": 44120, "step": 38463} +{"lr": 1.003753725216005e-05, "data_time": 0.0014446496963500977, "loss": 0.3818373054265976, "time": 0.633182430267334, "epoch": 10, "memory": 44120, "step": 38563} +{"lr": 1.0024214572955152e-05, "data_time": 0.0016610383987426757, "loss": 0.36392607986927034, "time": 0.6334603786468506, "epoch": 10, "memory": 44120, "step": 38663} +{"lr": 1.0013799859849456e-05, "data_time": 0.0014972209930419922, "loss": 0.37379465997219086, "time": 0.633960509300232, "epoch": 10, "memory": 44120, "step": 38763} +{"lr": 1.0006293786219338e-05, "data_time": 0.0019221782684326171, "loss": 0.34700888097286225, "time": 0.6357094049453735, "epoch": 10, "memory": 44120, "step": 38863} +{"lr": 1.0001696837379472e-05, "data_time": 0.0013655424118041992, "loss": 0.35798477232456205, "time": 0.6330069303512573, "epoch": 10, "memory": 44120, "step": 38963} +{"lr": 1.0000009310551361e-05, "data_time": 0.0013276338577270508, "loss": 0.35898124873638154, "time": 0.6327818870544434, "epoch": 10, "memory": 44120, "step": 39063} +{"accuracy/top1": 75.45318603515625, "data_time": 0.03510504755480536, "time": 1.276444990059425, "step": 10} diff --git a/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm/20230604_023242/vis_data/vis_image/0000000.png_1.png b/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm/20230604_023242/vis_data/vis_image/0000000.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..b7fbbd080cb1bcd7d0cc9182e1a3c2415b10d1d4 Binary files /dev/null and b/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm/20230604_023242/vis_data/vis_image/0000000.png_1.png differ diff --git a/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm/20230604_023242/vis_data/vis_image/0000000.png_10.png b/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm/20230604_023242/vis_data/vis_image/0000000.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..ba9f4a3e4278af45ebc4d264e264b14b36834fd1 Binary files /dev/null and b/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm/20230604_023242/vis_data/vis_image/0000000.png_10.png differ diff --git a/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm/20230604_023242/vis_data/vis_image/0000000.png_2.png b/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm/20230604_023242/vis_data/vis_image/0000000.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..44215ed2c5de9d472f6f5e6e7ee64844dc13e493 Binary files /dev/null and b/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm/20230604_023242/vis_data/vis_image/0000000.png_2.png differ diff --git a/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm/20230604_023242/vis_data/vis_image/0000000.png_3.png b/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm/20230604_023242/vis_data/vis_image/0000000.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..4671bacd2fb928627c223953392cd72e39d6acf0 Binary files /dev/null and b/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm/20230604_023242/vis_data/vis_image/0000000.png_3.png differ diff --git a/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm/20230604_023242/vis_data/vis_image/0000000.png_4.png b/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm/20230604_023242/vis_data/vis_image/0000000.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..06689ac2ad736f7b74c5b7fc1faad6ec574d3db6 Binary files /dev/null and b/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm/20230604_023242/vis_data/vis_image/0000000.png_4.png differ diff --git a/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm/20230604_023242/vis_data/vis_image/0000000.png_5.png b/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm/20230604_023242/vis_data/vis_image/0000000.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..5538b238e209b7610f93acc759952c8fd8834675 Binary files /dev/null and b/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm/20230604_023242/vis_data/vis_image/0000000.png_5.png differ diff --git a/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm/20230604_023242/vis_data/vis_image/0000000.png_6.png b/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm/20230604_023242/vis_data/vis_image/0000000.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..d17e0d5df0d4f3f4581058182c5e03f4ba1a0608 Binary files /dev/null and b/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm/20230604_023242/vis_data/vis_image/0000000.png_6.png differ diff --git a/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm/20230604_023242/vis_data/vis_image/0000000.png_7.png b/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm/20230604_023242/vis_data/vis_image/0000000.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..662152664adf873471bee7d74cddd8a85bf16f60 Binary files /dev/null and b/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm/20230604_023242/vis_data/vis_image/0000000.png_7.png differ diff --git a/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm/20230604_023242/vis_data/vis_image/0000000.png_8.png b/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm/20230604_023242/vis_data/vis_image/0000000.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..70ed0afe891b04b242f0f791c3b49d4588bbf809 Binary files /dev/null and b/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm/20230604_023242/vis_data/vis_image/0000000.png_8.png differ diff --git a/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm/20230604_023242/vis_data/vis_image/0000000.png_9.png b/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm/20230604_023242/vis_data/vis_image/0000000.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..649cad36800b7b2fece6564793b3bda8d85856f0 Binary files /dev/null and b/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm/20230604_023242/vis_data/vis_image/0000000.png_9.png differ diff --git a/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm/20230604_023242/vis_data/vis_image/000005174.jpg_1.png b/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm/20230604_023242/vis_data/vis_image/000005174.jpg_1.png new file mode 100644 index 0000000000000000000000000000000000000000..01dc588ac908a9d5bd254f9905a934a9fc70f8c7 Binary files /dev/null and b/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm/20230604_023242/vis_data/vis_image/000005174.jpg_1.png differ diff --git a/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm/20230604_023242/vis_data/vis_image/000005174.jpg_10.png b/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm/20230604_023242/vis_data/vis_image/000005174.jpg_10.png new file mode 100644 index 0000000000000000000000000000000000000000..16be61545381baf300c8e71be6a30f41f22d697d Binary files /dev/null and b/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm/20230604_023242/vis_data/vis_image/000005174.jpg_10.png differ diff --git a/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm/20230604_023242/vis_data/vis_image/000005174.jpg_2.png b/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm/20230604_023242/vis_data/vis_image/000005174.jpg_2.png new file mode 100644 index 0000000000000000000000000000000000000000..25e9696437caefd22c3cfe90cbf99d5cf5fef1ed Binary files /dev/null and b/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm/20230604_023242/vis_data/vis_image/000005174.jpg_2.png differ diff --git a/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm/20230604_023242/vis_data/vis_image/000005174.jpg_3.png b/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm/20230604_023242/vis_data/vis_image/000005174.jpg_3.png new file mode 100644 index 0000000000000000000000000000000000000000..be733a2e65fb37cd7673c67bae6d70b4804e8d9a Binary files /dev/null and b/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm/20230604_023242/vis_data/vis_image/000005174.jpg_3.png differ diff --git a/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm/20230604_023242/vis_data/vis_image/000005174.jpg_4.png b/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm/20230604_023242/vis_data/vis_image/000005174.jpg_4.png new file mode 100644 index 0000000000000000000000000000000000000000..6f24f5e076d0e33b01076b60122d43675e11531a Binary files /dev/null and b/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm/20230604_023242/vis_data/vis_image/000005174.jpg_4.png differ diff --git a/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm/20230604_023242/vis_data/vis_image/000005174.jpg_5.png b/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm/20230604_023242/vis_data/vis_image/000005174.jpg_5.png new file mode 100644 index 0000000000000000000000000000000000000000..16e1f4cda62342ecf03acccf8d0965fb39232a80 Binary files /dev/null and b/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm/20230604_023242/vis_data/vis_image/000005174.jpg_5.png differ diff --git a/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm/20230604_023242/vis_data/vis_image/000005174.jpg_6.png b/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm/20230604_023242/vis_data/vis_image/000005174.jpg_6.png new file mode 100644 index 0000000000000000000000000000000000000000..58a3bbc4fe97dcc011805091cf6cbaeaef1278af Binary files /dev/null and b/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm/20230604_023242/vis_data/vis_image/000005174.jpg_6.png differ diff --git a/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm/20230604_023242/vis_data/vis_image/000005174.jpg_7.png b/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm/20230604_023242/vis_data/vis_image/000005174.jpg_7.png new file mode 100644 index 0000000000000000000000000000000000000000..30b5f751f95f664fcc8017341d351d3b586ede26 Binary files /dev/null and b/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm/20230604_023242/vis_data/vis_image/000005174.jpg_7.png differ diff --git a/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm/20230604_023242/vis_data/vis_image/000005174.jpg_8.png b/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm/20230604_023242/vis_data/vis_image/000005174.jpg_8.png new file mode 100644 index 0000000000000000000000000000000000000000..da1fd3177a348aa37b4bc0e6b7c249be868244b5 Binary files /dev/null and b/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm/20230604_023242/vis_data/vis_image/000005174.jpg_8.png differ diff --git a/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm/20230604_023242/vis_data/vis_image/000005174.jpg_9.png b/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm/20230604_023242/vis_data/vis_image/000005174.jpg_9.png new file mode 100644 index 0000000000000000000000000000000000000000..257a849ccfcf3d04e76d628bf280f1d68186233a Binary files /dev/null and b/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm/20230604_023242/vis_data/vis_image/000005174.jpg_9.png differ diff --git a/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm/clip_large_pretrain_4x256_sdv2_lr1e-4.py b/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm/clip_large_pretrain_4x256_sdv2_lr1e-4.py new file mode 100644 index 0000000000000000000000000000000000000000..35ec1e34d29b17c45878985adb3abd9b6709a41c --- /dev/null +++ b/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm/clip_large_pretrain_4x256_sdv2_lr1e-4.py @@ -0,0 +1,228 @@ +optim_wrapper = dict( + optimizer=dict( + type='AdamW', lr=0.0001, weight_decay=0.3, _scope_='mmpretrain'), + paramwise_cfg=dict( + custom_keys=dict({ + '.cls_token': dict(decay_mult=0.0), + '.pos_embed': dict(decay_mult=0.0) + })), + type='AmpOptimWrapper', + dtype='bfloat16', + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=False, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=4096) +model = dict( + type='ImageClassifier', + backbone=dict( + frozen_stages=24, + type='VisionTransformer', + arch='l', + img_size=224, + patch_size=14, + drop_rate=0.1, + pre_norm=True, + final_norm=False, + init_cfg=dict( + type='Pretrained', + checkpoint='ckpt/openclip-ViT-L-14.pth', + prefix='backbone')), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=1024, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + init_cfg=None), + init_cfg=dict( + type='TruncNormal', layer=['Conv2d', 'Linear'], std=0.02, bias=0.0), + train_cfg=None) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=128, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stablediffusionV2-1-dpmsolver-25-1m.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/cc1m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=128, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV2-1-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=128, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV2-1-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm' diff --git a/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm/epoch_1.pth b/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm/epoch_1.pth new file mode 100644 index 0000000000000000000000000000000000000000..04b1a99a4f725dd0a41a94e7db5bcbab187f1dbd --- /dev/null +++ b/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm/epoch_1.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:f689615c1351a1e849a5cba0277fdba533386bc7d4123a1421d363d57f08ea6b +size 2426037899 diff --git a/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm/epoch_10.pth b/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm/epoch_10.pth new file mode 100644 index 0000000000000000000000000000000000000000..4e6c25d7cc6ab8fbebfdd712c27e97948e0c4cdf --- /dev/null +++ b/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm/epoch_10.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c17cb04d4207be5f5347cbe82a1e28d8c3049434bf72099fe7fe90a98d9f3b05 +size 2428603019 diff --git a/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm/epoch_2.pth b/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm/epoch_2.pth new file mode 100644 index 0000000000000000000000000000000000000000..fba9a9e0243a1e2d0b3e9e69cebb14e8fd9287fa --- /dev/null +++ b/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm/epoch_2.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:571d3d649933ea0fad07de46701919a66aea5ecd48f17a71247c4b523e0c5f74 +size 2426321931 diff --git a/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm/epoch_3.pth b/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm/epoch_3.pth new file mode 100644 index 0000000000000000000000000000000000000000..b6389d6f78fd76c85ee1c00f2b1f212460641928 --- /dev/null +++ b/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm/epoch_3.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:d665650fe17602072218bc93ff8594f86f6e7a5e8c27b3d40afb8ff9cc67bba0 +size 2426605771 diff --git a/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm/epoch_4.pth b/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm/epoch_4.pth new file mode 100644 index 0000000000000000000000000000000000000000..653f3dab964a5a3b5879398ab460f73bf7380fdb --- /dev/null +++ b/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm/epoch_4.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:d16d1ae798117f138e4dca4a75ec79e2e32afbe98555f81f1625a3bd8fa93991 +size 2426889099 diff --git a/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm/epoch_5.pth b/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm/epoch_5.pth new file mode 100644 index 0000000000000000000000000000000000000000..62243c60eb896fa7167e90ba8c825583d77daf5a --- /dev/null +++ b/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm/epoch_5.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:9ade1ee7d5eac5d535a24ed7b65c5ee6e8aa66a8cdc08cc5b18b97d63e747770 +size 2427172747 diff --git a/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm/epoch_6.pth b/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm/epoch_6.pth new file mode 100644 index 0000000000000000000000000000000000000000..2654e7eba5db21270c7687fd58ab1a45bb8aa185 --- /dev/null +++ b/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm/epoch_6.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:df28b322922c1c3595a67e54889f0635fda988fdebe4fab4caf6743fc087835c +size 2427456395 diff --git a/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm/epoch_7.pth b/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm/epoch_7.pth new file mode 100644 index 0000000000000000000000000000000000000000..32291ea51801e0bffae5e17784653bb76fb34402 --- /dev/null +++ b/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm/epoch_7.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:1bb9b8f03e94dd97113a97f37f765882730a5c12b9caca310ca54d9d08a8c9d1 +size 2427740427 diff --git a/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm/epoch_8.pth b/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm/epoch_8.pth new file mode 100644 index 0000000000000000000000000000000000000000..c7facc154efd86deb7e2f17e9176317a68ba4f5d --- /dev/null +++ b/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm/epoch_8.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:bacda0b40cd483bdcc54aca484c4dcf5479c9ccd2c30472532914d919f42faf2 +size 2428028107 diff --git a/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm/epoch_9.pth b/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm/epoch_9.pth new file mode 100644 index 0000000000000000000000000000000000000000..361aa11b6e0749a0cd56058bef60162a21a4a6f2 --- /dev/null +++ b/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm/epoch_9.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:bca9749816b03558f37cb0d4e0bd80b28109194bf486a6bd2b0899709a66e858 +size 2428315595 diff --git a/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm/last_checkpoint b/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm/last_checkpoint new file mode 100644 index 0000000000000000000000000000000000000000..59494c4332afc8250b18ac3b933f0e3940e8d6d4 --- /dev/null +++ b/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm/last_checkpoint @@ -0,0 +1 @@ +/mnt/petrelfs/luzeyu/workspace/fakebench/mmpretrain/workdir/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm/epoch_10.pth \ No newline at end of file diff --git a/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm_wneck/20230604_032014/20230604_032014.log b/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm_wneck/20230604_032014/20230604_032014.log new file mode 100644 index 0000000000000000000000000000000000000000..076fca9776b4f94bbe82389a03c1e618907790fc --- /dev/null +++ b/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm_wneck/20230604_032014/20230604_032014.log @@ -0,0 +1,1727 @@ +2023/06/04 03:20:18 - mmengine - INFO - +------------------------------------------------------------ +System environment: + sys.platform: linux + Python: 3.10.9 (main, Mar 8 2023, 10:47:38) [GCC 11.2.0] + CUDA available: True + numpy_random_seed: 1351706577 + GPU 0,1,2,3: NVIDIA A100-SXM4-80GB + CUDA_HOME: /mnt/petrelfs/share/cuda-11.6 + NVCC: Cuda compilation tools, release 11.6, V11.6.124 + GCC: gcc (GCC) 7.5.0 + PyTorch: 1.13.1 + PyTorch compiling details: PyTorch built with: + - GCC 9.3 + - C++ Version: 201402 + - Intel(R) oneAPI Math Kernel Library Version 2021.4-Product Build 20210904 for Intel(R) 64 architecture applications + - Intel(R) MKL-DNN v2.6.0 (Git Hash 52b5f107dd9cf10910aaa19cb47f3abf9b349815) + - OpenMP 201511 (a.k.a. OpenMP 4.5) + - LAPACK is enabled (usually provided by MKL) + - NNPACK is enabled + - CPU capability usage: AVX2 + - CUDA Runtime 11.6 + - NVCC architecture flags: -gencode;arch=compute_37,code=sm_37;-gencode;arch=compute_50,code=sm_50;-gencode;arch=compute_60,code=sm_60;-gencode;arch=compute_61,code=sm_61;-gencode;arch=compute_70,code=sm_70;-gencode;arch=compute_75,code=sm_75;-gencode;arch=compute_80,code=sm_80;-gencode;arch=compute_86,code=sm_86;-gencode;arch=compute_37,code=compute_37 + - CuDNN 8.3.2 (built against CUDA 11.5) + - Magma 2.6.1 + - Build settings: BLAS_INFO=mkl, BUILD_TYPE=Release, CUDA_VERSION=11.6, CUDNN_VERSION=8.3.2, CXX_COMPILER=/opt/rh/devtoolset-9/root/usr/bin/c++, CXX_FLAGS= -fabi-version=11 -Wno-deprecated -fvisibility-inlines-hidden -DUSE_PTHREADPOOL -fopenmp -DNDEBUG -DUSE_KINETO -DUSE_FBGEMM -DUSE_QNNPACK -DUSE_PYTORCH_QNNPACK -DUSE_XNNPACK -DSYMBOLICATE_MOBILE_DEBUG_HANDLE -DEDGE_PROFILER_USE_KINETO -O2 -fPIC -Wno-narrowing -Wall -Wextra -Werror=return-type -Werror=non-virtual-dtor -Wno-missing-field-initializers -Wno-type-limits -Wno-array-bounds -Wno-unknown-pragmas -Wunused-local-typedefs -Wno-unused-parameter -Wno-unused-function -Wno-unused-result -Wno-strict-overflow -Wno-strict-aliasing -Wno-error=deprecated-declarations -Wno-stringop-overflow -Wno-psabi -Wno-error=pedantic -Wno-error=redundant-decls -Wno-error=old-style-cast -fdiagnostics-color=always -faligned-new -Wno-unused-but-set-variable -Wno-maybe-uninitialized -fno-math-errno -fno-trapping-math -Werror=format -Werror=cast-function-type -Wno-stringop-overflow, LAPACK_INFO=mkl, PERF_WITH_AVX=1, PERF_WITH_AVX2=1, PERF_WITH_AVX512=1, TORCH_VERSION=1.13.1, USE_CUDA=ON, USE_CUDNN=ON, USE_EXCEPTION_PTR=1, USE_GFLAGS=OFF, USE_GLOG=OFF, USE_MKL=ON, USE_MKLDNN=ON, USE_MPI=OFF, USE_NCCL=ON, USE_NNPACK=ON, USE_OPENMP=ON, USE_ROCM=OFF, + + TorchVision: 0.14.1 + OpenCV: 4.7.0 + MMEngine: 0.7.3 + +Runtime environment: + cudnn_benchmark: True + mp_cfg: {'mp_start_method': 'fork', 'opencv_num_threads': 0} + dist_cfg: {'backend': 'nccl'} + seed: None + deterministic: False + Distributed launcher: slurm + Distributed training: True + GPU number: 4 +------------------------------------------------------------ + +2023/06/04 03:20:23 - mmengine - INFO - Config: +optim_wrapper = dict( + optimizer=dict( + type='AdamW', lr=0.0001, weight_decay=0.3, _scope_='mmpretrain'), + paramwise_cfg=dict( + custom_keys=dict({ + '.cls_token': dict(decay_mult=0.0), + '.pos_embed': dict(decay_mult=0.0) + })), + type='AmpOptimWrapper', + dtype='bfloat16', + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=False, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=4096) +model = dict( + type='ImageClassifier', + backbone=dict( + frozen_stages=24, + type='VisionTransformer', + arch='l', + img_size=224, + patch_size=14, + drop_rate=0.1, + pre_norm=True, + final_norm=False, + init_cfg=dict( + type='Pretrained', + checkpoint='ckpt/openclip-ViT-L-14.pth', + prefix='backbone')), + neck=dict( + type='CLIPProjection', + in_channels=1024, + out_channels=768, + init_cfg=dict( + type='Pretrained', + checkpoint='ckpt/openclip-ViT-L-14.pth', + prefix='backbone')), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=768, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + init_cfg=None), + init_cfg=dict( + type='TruncNormal', layer=['Conv2d', 'Linear'], std=0.02, bias=0.0), + train_cfg=None) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=128, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stablediffusionV2-1-dpmsolver-25-1m.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/cc1m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=128, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV2-1-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=128, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV2-1-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm_wneck' + +2023/06/04 03:20:36 - mmengine - INFO - Hooks will be executed in the following order: +before_run: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_load_checkpoint: +(ABOVE_NORMAL) EMAHook + -------------------- +before_train: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_train_epoch: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(NORMAL ) DistSamplerSeedHook + -------------------- +before_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook + -------------------- +after_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +after_train_epoch: +(NORMAL ) IterTimerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_val_epoch: +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook + -------------------- +before_val_iter: +(NORMAL ) IterTimerHook + -------------------- +after_val_iter: +(NORMAL ) IterTimerHook +(NORMAL ) VisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_val_epoch: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_save_checkpoint: +(ABOVE_NORMAL) EMAHook + -------------------- +after_train: +(VERY_LOW ) CheckpointHook + -------------------- +before_test_epoch: +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook + -------------------- +before_test_iter: +(NORMAL ) IterTimerHook + -------------------- +after_test_iter: +(NORMAL ) IterTimerHook +(NORMAL ) VisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_test_epoch: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_run: +(BELOW_NORMAL) LoggerHook + -------------------- +2023/06/04 03:20:57 - mmengine - INFO - load backbone in model from: ckpt/openclip-ViT-L-14.pth +2023/06/04 03:20:58 - mmengine - WARNING - The model and loaded state dict do not match exactly + +unexpected key in source state_dict: ln1.weight, ln1.bias + +2023/06/04 03:20:58 - mmengine - INFO - load backbone in model from: ckpt/openclip-ViT-L-14.pth +2023/06/04 03:21:00 - mmengine - WARNING - The model and loaded state dict do not match exactly + +unexpected key in source state_dict: cls_token, pos_embed, patch_embed.projection.weight, pre_norm.weight, pre_norm.bias, layers.0.ln1.weight, layers.0.ln1.bias, layers.0.attn.qkv.weight, layers.0.attn.qkv.bias, layers.0.attn.proj.weight, layers.0.attn.proj.bias, layers.0.ln2.weight, layers.0.ln2.bias, layers.0.ffn.layers.0.0.weight, layers.0.ffn.layers.0.0.bias, layers.0.ffn.layers.1.weight, layers.0.ffn.layers.1.bias, layers.1.ln1.weight, layers.1.ln1.bias, layers.1.attn.qkv.weight, layers.1.attn.qkv.bias, layers.1.attn.proj.weight, layers.1.attn.proj.bias, layers.1.ln2.weight, layers.1.ln2.bias, layers.1.ffn.layers.0.0.weight, layers.1.ffn.layers.0.0.bias, layers.1.ffn.layers.1.weight, layers.1.ffn.layers.1.bias, layers.2.ln1.weight, layers.2.ln1.bias, layers.2.attn.qkv.weight, layers.2.attn.qkv.bias, layers.2.attn.proj.weight, layers.2.attn.proj.bias, layers.2.ln2.weight, layers.2.ln2.bias, layers.2.ffn.layers.0.0.weight, layers.2.ffn.layers.0.0.bias, layers.2.ffn.layers.1.weight, layers.2.ffn.layers.1.bias, layers.3.ln1.weight, layers.3.ln1.bias, layers.3.attn.qkv.weight, layers.3.attn.qkv.bias, layers.3.attn.proj.weight, layers.3.attn.proj.bias, layers.3.ln2.weight, layers.3.ln2.bias, layers.3.ffn.layers.0.0.weight, layers.3.ffn.layers.0.0.bias, layers.3.ffn.layers.1.weight, layers.3.ffn.layers.1.bias, layers.4.ln1.weight, layers.4.ln1.bias, layers.4.attn.qkv.weight, layers.4.attn.qkv.bias, layers.4.attn.proj.weight, layers.4.attn.proj.bias, layers.4.ln2.weight, layers.4.ln2.bias, layers.4.ffn.layers.0.0.weight, layers.4.ffn.layers.0.0.bias, layers.4.ffn.layers.1.weight, layers.4.ffn.layers.1.bias, layers.5.ln1.weight, layers.5.ln1.bias, layers.5.attn.qkv.weight, layers.5.attn.qkv.bias, layers.5.attn.proj.weight, layers.5.attn.proj.bias, layers.5.ln2.weight, layers.5.ln2.bias, layers.5.ffn.layers.0.0.weight, layers.5.ffn.layers.0.0.bias, layers.5.ffn.layers.1.weight, layers.5.ffn.layers.1.bias, layers.6.ln1.weight, layers.6.ln1.bias, layers.6.attn.qkv.weight, layers.6.attn.qkv.bias, layers.6.attn.proj.weight, layers.6.attn.proj.bias, layers.6.ln2.weight, layers.6.ln2.bias, layers.6.ffn.layers.0.0.weight, layers.6.ffn.layers.0.0.bias, layers.6.ffn.layers.1.weight, layers.6.ffn.layers.1.bias, layers.7.ln1.weight, layers.7.ln1.bias, layers.7.attn.qkv.weight, layers.7.attn.qkv.bias, layers.7.attn.proj.weight, layers.7.attn.proj.bias, layers.7.ln2.weight, layers.7.ln2.bias, layers.7.ffn.layers.0.0.weight, layers.7.ffn.layers.0.0.bias, layers.7.ffn.layers.1.weight, layers.7.ffn.layers.1.bias, layers.8.ln1.weight, layers.8.ln1.bias, layers.8.attn.qkv.weight, layers.8.attn.qkv.bias, layers.8.attn.proj.weight, layers.8.attn.proj.bias, layers.8.ln2.weight, layers.8.ln2.bias, layers.8.ffn.layers.0.0.weight, layers.8.ffn.layers.0.0.bias, layers.8.ffn.layers.1.weight, layers.8.ffn.layers.1.bias, layers.9.ln1.weight, layers.9.ln1.bias, layers.9.attn.qkv.weight, layers.9.attn.qkv.bias, layers.9.attn.proj.weight, layers.9.attn.proj.bias, layers.9.ln2.weight, layers.9.ln2.bias, layers.9.ffn.layers.0.0.weight, layers.9.ffn.layers.0.0.bias, layers.9.ffn.layers.1.weight, layers.9.ffn.layers.1.bias, layers.10.ln1.weight, layers.10.ln1.bias, layers.10.attn.qkv.weight, layers.10.attn.qkv.bias, layers.10.attn.proj.weight, layers.10.attn.proj.bias, layers.10.ln2.weight, layers.10.ln2.bias, layers.10.ffn.layers.0.0.weight, layers.10.ffn.layers.0.0.bias, layers.10.ffn.layers.1.weight, layers.10.ffn.layers.1.bias, layers.11.ln1.weight, layers.11.ln1.bias, layers.11.attn.qkv.weight, layers.11.attn.qkv.bias, layers.11.attn.proj.weight, layers.11.attn.proj.bias, layers.11.ln2.weight, layers.11.ln2.bias, layers.11.ffn.layers.0.0.weight, layers.11.ffn.layers.0.0.bias, layers.11.ffn.layers.1.weight, layers.11.ffn.layers.1.bias, layers.12.ln1.weight, layers.12.ln1.bias, layers.12.attn.qkv.weight, layers.12.attn.qkv.bias, layers.12.attn.proj.weight, layers.12.attn.proj.bias, layers.12.ln2.weight, layers.12.ln2.bias, layers.12.ffn.layers.0.0.weight, layers.12.ffn.layers.0.0.bias, layers.12.ffn.layers.1.weight, layers.12.ffn.layers.1.bias, layers.13.ln1.weight, layers.13.ln1.bias, layers.13.attn.qkv.weight, layers.13.attn.qkv.bias, layers.13.attn.proj.weight, layers.13.attn.proj.bias, layers.13.ln2.weight, layers.13.ln2.bias, layers.13.ffn.layers.0.0.weight, layers.13.ffn.layers.0.0.bias, layers.13.ffn.layers.1.weight, layers.13.ffn.layers.1.bias, layers.14.ln1.weight, layers.14.ln1.bias, layers.14.attn.qkv.weight, layers.14.attn.qkv.bias, layers.14.attn.proj.weight, layers.14.attn.proj.bias, layers.14.ln2.weight, layers.14.ln2.bias, layers.14.ffn.layers.0.0.weight, layers.14.ffn.layers.0.0.bias, layers.14.ffn.layers.1.weight, layers.14.ffn.layers.1.bias, layers.15.ln1.weight, layers.15.ln1.bias, layers.15.attn.qkv.weight, layers.15.attn.qkv.bias, layers.15.attn.proj.weight, layers.15.attn.proj.bias, layers.15.ln2.weight, layers.15.ln2.bias, layers.15.ffn.layers.0.0.weight, layers.15.ffn.layers.0.0.bias, layers.15.ffn.layers.1.weight, layers.15.ffn.layers.1.bias, layers.16.ln1.weight, layers.16.ln1.bias, layers.16.attn.qkv.weight, layers.16.attn.qkv.bias, layers.16.attn.proj.weight, layers.16.attn.proj.bias, layers.16.ln2.weight, layers.16.ln2.bias, layers.16.ffn.layers.0.0.weight, layers.16.ffn.layers.0.0.bias, layers.16.ffn.layers.1.weight, layers.16.ffn.layers.1.bias, layers.17.ln1.weight, layers.17.ln1.bias, layers.17.attn.qkv.weight, layers.17.attn.qkv.bias, layers.17.attn.proj.weight, layers.17.attn.proj.bias, layers.17.ln2.weight, layers.17.ln2.bias, layers.17.ffn.layers.0.0.weight, layers.17.ffn.layers.0.0.bias, layers.17.ffn.layers.1.weight, layers.17.ffn.layers.1.bias, layers.18.ln1.weight, layers.18.ln1.bias, layers.18.attn.qkv.weight, layers.18.attn.qkv.bias, layers.18.attn.proj.weight, layers.18.attn.proj.bias, layers.18.ln2.weight, layers.18.ln2.bias, layers.18.ffn.layers.0.0.weight, layers.18.ffn.layers.0.0.bias, layers.18.ffn.layers.1.weight, layers.18.ffn.layers.1.bias, layers.19.ln1.weight, layers.19.ln1.bias, layers.19.attn.qkv.weight, layers.19.attn.qkv.bias, layers.19.attn.proj.weight, layers.19.attn.proj.bias, layers.19.ln2.weight, layers.19.ln2.bias, layers.19.ffn.layers.0.0.weight, layers.19.ffn.layers.0.0.bias, layers.19.ffn.layers.1.weight, layers.19.ffn.layers.1.bias, layers.20.ln1.weight, layers.20.ln1.bias, layers.20.attn.qkv.weight, layers.20.attn.qkv.bias, layers.20.attn.proj.weight, layers.20.attn.proj.bias, layers.20.ln2.weight, layers.20.ln2.bias, layers.20.ffn.layers.0.0.weight, layers.20.ffn.layers.0.0.bias, layers.20.ffn.layers.1.weight, layers.20.ffn.layers.1.bias, layers.21.ln1.weight, layers.21.ln1.bias, layers.21.attn.qkv.weight, layers.21.attn.qkv.bias, layers.21.attn.proj.weight, layers.21.attn.proj.bias, layers.21.ln2.weight, layers.21.ln2.bias, layers.21.ffn.layers.0.0.weight, layers.21.ffn.layers.0.0.bias, layers.21.ffn.layers.1.weight, layers.21.ffn.layers.1.bias, layers.22.ln1.weight, layers.22.ln1.bias, layers.22.attn.qkv.weight, layers.22.attn.qkv.bias, layers.22.attn.proj.weight, layers.22.attn.proj.bias, layers.22.ln2.weight, layers.22.ln2.bias, layers.22.ffn.layers.0.0.weight, layers.22.ffn.layers.0.0.bias, layers.22.ffn.layers.1.weight, layers.22.ffn.layers.1.bias, layers.23.ln1.weight, layers.23.ln1.bias, layers.23.attn.qkv.weight, layers.23.attn.qkv.bias, layers.23.attn.proj.weight, layers.23.attn.proj.bias, layers.23.ln2.weight, layers.23.ln2.bias, layers.23.ffn.layers.0.0.weight, layers.23.ffn.layers.0.0.bias, layers.23.ffn.layers.1.weight, layers.23.ffn.layers.1.bias, ln1.weight, ln1.bias + +missing keys in source state_dict: proj + +Name of parameter - Initialization information + +backbone.cls_token - torch.Size([1, 1, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.pos_embed - torch.Size([1, 257, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.patch_embed.projection.weight - torch.Size([1024, 3, 14, 14]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.0.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.0.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.0.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.0.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.0.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.0.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.0.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.0.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.0.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.0.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.0.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.0.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.1.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.1.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.1.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.1.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.1.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.1.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.1.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.1.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.1.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.1.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.1.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.1.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.2.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.2.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.2.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.2.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.2.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.2.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.2.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.2.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.2.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.2.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.2.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.2.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.3.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.3.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.3.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.3.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.3.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.3.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.3.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.3.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.3.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.3.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.3.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.3.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.4.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.4.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.4.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.4.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.4.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.4.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.4.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.4.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.4.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.4.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.4.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.4.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.5.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.5.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.5.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.5.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.5.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.5.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.5.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.5.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.5.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.5.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.5.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.5.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.6.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.6.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.6.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.6.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.6.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.6.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.6.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.6.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.6.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.6.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.6.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.6.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.7.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.7.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.7.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.7.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.7.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.7.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.7.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.7.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.7.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.7.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.7.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.7.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.8.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.8.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.8.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.8.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.8.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.8.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.8.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.8.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.8.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.8.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.8.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.8.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.9.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.9.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.9.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.9.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.9.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.9.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.9.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.9.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.9.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.9.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.9.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.9.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.10.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.10.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.10.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.10.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.10.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.10.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.10.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.10.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.10.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.10.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.10.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.10.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.11.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.11.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.11.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.11.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.11.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.11.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.11.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.11.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.11.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.11.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.11.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.11.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.12.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.12.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.12.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.12.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.12.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.12.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.12.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.12.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.12.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.12.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.12.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.12.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.13.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.13.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.13.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.13.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.13.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.13.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.13.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.13.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.13.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.13.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.13.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.13.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.14.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.14.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.14.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.14.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.14.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.14.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.14.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.14.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.14.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.14.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.14.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.14.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.15.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.15.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.15.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.15.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.15.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.15.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.15.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.15.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.15.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.15.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.15.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.15.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.16.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.16.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.16.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.16.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.16.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.16.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.16.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.16.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.16.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.16.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.16.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.16.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.17.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.17.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.17.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.17.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.17.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.17.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.17.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.17.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.17.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.17.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.17.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.17.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.18.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.18.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.18.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.18.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.18.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.18.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.18.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.18.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.18.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.18.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.18.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.18.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.19.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.19.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.19.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.19.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.19.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.19.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.19.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.19.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.19.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.19.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.19.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.19.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.20.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.20.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.20.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.20.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.20.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.20.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.20.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.20.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.20.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.20.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.20.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.20.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.21.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.21.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.21.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.21.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.21.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.21.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.21.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.21.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.21.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.21.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.21.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.21.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.22.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.22.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.22.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.22.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.22.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.22.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.22.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.22.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.22.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.22.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.22.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.22.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.23.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.23.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.23.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.23.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.23.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.23.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.23.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.23.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.23.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.23.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.23.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.23.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.pre_norm.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.pre_norm.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +neck.proj - torch.Size([1024, 768]): +The value is the same before and after calling `init_weights` of ImageClassifier + +head.fc.weight - torch.Size([2, 768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +head.fc.bias - torch.Size([2]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 +2023/06/04 03:21:00 - mmengine - WARNING - "FileClient" will be deprecated in future. Please use io functions in https://mmengine.readthedocs.io/en/latest/api/fileio.html#file-io +2023/06/04 03:21:00 - mmengine - WARNING - "HardDiskBackend" is the alias of "LocalBackend" and the former will be deprecated in future. +2023/06/04 03:21:00 - mmengine - INFO - Checkpoints will be saved to /mnt/petrelfs/luzeyu/workspace/fakebench/mmpretrain/workdir/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm_wneck. +2023/06/04 03:22:09 - mmengine - INFO - Epoch(train) [1][ 100/3907] lr: 9.9999e-05 eta: 7:30:26 time: 0.6352 data_time: 0.0025 memory: 44138 loss: 0.5352 +2023/06/04 03:23:13 - mmengine - INFO - Epoch(train) [1][ 200/3907] lr: 9.9994e-05 eta: 7:10:21 time: 0.6351 data_time: 0.0024 memory: 44138 loss: 0.4755 +2023/06/04 03:24:16 - mmengine - INFO - Epoch(train) [1][ 300/3907] lr: 9.9987e-05 eta: 7:03:01 time: 0.6340 data_time: 0.0019 memory: 44138 loss: 0.4664 +2023/06/04 03:25:20 - mmengine - INFO - Epoch(train) [1][ 400/3907] lr: 9.9977e-05 eta: 6:59:09 time: 0.6367 data_time: 0.0027 memory: 44138 loss: 0.4659 +2023/06/04 03:26:24 - mmengine - INFO - Epoch(train) [1][ 500/3907] lr: 9.9964e-05 eta: 6:56:13 time: 0.6356 data_time: 0.0020 memory: 44138 loss: 0.4894 +2023/06/04 03:27:27 - mmengine - INFO - Epoch(train) [1][ 600/3907] lr: 9.9948e-05 eta: 6:53:54 time: 0.6359 data_time: 0.0017 memory: 44138 loss: 0.4474 +2023/06/04 03:28:31 - mmengine - INFO - Epoch(train) [1][ 700/3907] lr: 9.9929e-05 eta: 6:52:02 time: 0.6359 data_time: 0.0016 memory: 44138 loss: 0.4865 +2023/06/04 03:29:35 - mmengine - INFO - Epoch(train) [1][ 800/3907] lr: 9.9907e-05 eta: 6:50:16 time: 0.6357 data_time: 0.0023 memory: 44138 loss: 0.4405 +2023/06/04 03:30:38 - mmengine - INFO - Epoch(train) [1][ 900/3907] lr: 9.9882e-05 eta: 6:48:43 time: 0.6358 data_time: 0.0016 memory: 44138 loss: 0.4617 +2023/06/04 03:31:42 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230604_032014 +2023/06/04 03:31:42 - mmengine - INFO - Epoch(train) [1][1000/3907] lr: 9.9855e-05 eta: 6:47:16 time: 0.6360 data_time: 0.0020 memory: 44138 loss: 0.4487 +2023/06/04 03:32:46 - mmengine - INFO - Epoch(train) [1][1100/3907] lr: 9.9824e-05 eta: 6:45:55 time: 0.6362 data_time: 0.0021 memory: 44138 loss: 0.4160 +2023/06/04 03:33:49 - mmengine - INFO - Epoch(train) [1][1200/3907] lr: 9.9791e-05 eta: 6:44:33 time: 0.6356 data_time: 0.0018 memory: 44138 loss: 0.4279 +2023/06/04 03:34:53 - mmengine - INFO - Epoch(train) [1][1300/3907] lr: 9.9755e-05 eta: 6:43:12 time: 0.6347 data_time: 0.0016 memory: 44138 loss: 0.4355 +2023/06/04 03:35:56 - mmengine - INFO - Epoch(train) [1][1400/3907] lr: 9.9716e-05 eta: 6:41:54 time: 0.6342 data_time: 0.0017 memory: 44138 loss: 0.4457 +2023/06/04 03:37:00 - mmengine - INFO - Epoch(train) [1][1500/3907] lr: 9.9674e-05 eta: 6:40:39 time: 0.6361 data_time: 0.0017 memory: 44138 loss: 0.4001 +2023/06/04 03:38:03 - mmengine - INFO - Epoch(train) [1][1600/3907] lr: 9.9629e-05 eta: 6:39:25 time: 0.6359 data_time: 0.0015 memory: 44138 loss: 0.4465 +2023/06/04 03:39:07 - mmengine - INFO - Epoch(train) [1][1700/3907] lr: 9.9581e-05 eta: 6:38:15 time: 0.6356 data_time: 0.0014 memory: 44138 loss: 0.4103 +2023/06/04 03:40:11 - mmengine - INFO - Epoch(train) [1][1800/3907] lr: 9.9530e-05 eta: 6:37:03 time: 0.6360 data_time: 0.0017 memory: 44138 loss: 0.4354 +2023/06/04 03:41:14 - mmengine - INFO - Epoch(train) [1][1900/3907] lr: 9.9476e-05 eta: 6:35:53 time: 0.6357 data_time: 0.0017 memory: 44138 loss: 0.4245 +2023/06/04 03:42:18 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230604_032014 +2023/06/04 03:42:18 - mmengine - INFO - Epoch(train) [1][2000/3907] lr: 9.9420e-05 eta: 6:34:42 time: 0.6344 data_time: 0.0020 memory: 44138 loss: 0.4313 +2023/06/04 03:43:21 - mmengine - INFO - Epoch(train) [1][2100/3907] lr: 9.9361e-05 eta: 6:33:31 time: 0.6356 data_time: 0.0017 memory: 44138 loss: 0.4716 +2023/06/04 03:44:25 - mmengine - INFO - Epoch(train) [1][2200/3907] lr: 9.9298e-05 eta: 6:32:24 time: 0.6354 data_time: 0.0021 memory: 44138 loss: 0.4074 +2023/06/04 03:45:29 - mmengine - INFO - Epoch(train) [1][2300/3907] lr: 9.9233e-05 eta: 6:31:16 time: 0.6363 data_time: 0.0018 memory: 44138 loss: 0.3899 +2023/06/04 03:46:32 - mmengine - INFO - Epoch(train) [1][2400/3907] lr: 9.9165e-05 eta: 6:30:09 time: 0.6371 data_time: 0.0024 memory: 44138 loss: 0.3961 +2023/06/04 03:47:36 - mmengine - INFO - Epoch(train) [1][2500/3907] lr: 9.9095e-05 eta: 6:29:05 time: 0.6473 data_time: 0.0020 memory: 44138 loss: 0.3770 +2023/06/04 03:48:40 - mmengine - INFO - Epoch(train) [1][2600/3907] lr: 9.9021e-05 eta: 6:27:58 time: 0.6379 data_time: 0.0020 memory: 44138 loss: 0.4162 +2023/06/04 03:49:43 - mmengine - INFO - Epoch(train) [1][2700/3907] lr: 9.8944e-05 eta: 6:26:53 time: 0.6369 data_time: 0.0017 memory: 44138 loss: 0.4208 +2023/06/04 03:50:47 - mmengine - INFO - Epoch(train) [1][2800/3907] lr: 9.8865e-05 eta: 6:25:47 time: 0.6362 data_time: 0.0018 memory: 44138 loss: 0.4456 +2023/06/04 03:51:51 - mmengine - INFO - Epoch(train) [1][2900/3907] lr: 9.8783e-05 eta: 6:24:41 time: 0.6356 data_time: 0.0017 memory: 44138 loss: 0.4209 +2023/06/04 03:52:54 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230604_032014 +2023/06/04 03:52:54 - mmengine - INFO - Epoch(train) [1][3000/3907] lr: 9.8698e-05 eta: 6:23:34 time: 0.6349 data_time: 0.0016 memory: 44138 loss: 0.4154 +2023/06/04 03:53:58 - mmengine - INFO - Epoch(train) [1][3100/3907] lr: 9.8610e-05 eta: 6:22:27 time: 0.6351 data_time: 0.0027 memory: 44138 loss: 0.4220 +2023/06/04 03:55:01 - mmengine - INFO - Epoch(train) [1][3200/3907] lr: 9.8519e-05 eta: 6:21:21 time: 0.6355 data_time: 0.0021 memory: 44138 loss: 0.4040 +2023/06/04 03:56:05 - mmengine - INFO - Epoch(train) [1][3300/3907] lr: 9.8426e-05 eta: 6:20:16 time: 0.6437 data_time: 0.0020 memory: 44138 loss: 0.3926 +2023/06/04 03:57:09 - mmengine - INFO - Epoch(train) [1][3400/3907] lr: 9.8330e-05 eta: 6:19:10 time: 0.6360 data_time: 0.0016 memory: 44138 loss: 0.3855 +2023/06/04 03:58:12 - mmengine - INFO - Epoch(train) [1][3500/3907] lr: 9.8231e-05 eta: 6:18:05 time: 0.6359 data_time: 0.0017 memory: 44138 loss: 0.4069 +2023/06/04 03:59:16 - mmengine - INFO - Epoch(train) [1][3600/3907] lr: 9.8129e-05 eta: 6:16:59 time: 0.6358 data_time: 0.0017 memory: 44138 loss: 0.3910 +2023/06/04 04:00:20 - mmengine - INFO - Epoch(train) [1][3700/3907] lr: 9.8024e-05 eta: 6:15:55 time: 0.6446 data_time: 0.0016 memory: 44138 loss: 0.4196 +2023/06/04 04:01:23 - mmengine - INFO - Epoch(train) [1][3800/3907] lr: 9.7917e-05 eta: 6:14:50 time: 0.6369 data_time: 0.0016 memory: 44138 loss: 0.3833 +2023/06/04 04:02:27 - mmengine - INFO - Epoch(train) [1][3900/3907] lr: 9.7806e-05 eta: 6:13:44 time: 0.6326 data_time: 0.0013 memory: 44138 loss: 0.3862 +2023/06/04 04:02:31 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230604_032014 +2023/06/04 04:02:31 - mmengine - INFO - Saving checkpoint at 1 epochs +2023/06/04 04:04:13 - mmengine - INFO - Epoch(val) [1][57/57] accuracy/top1: 82.5580 single-label/precision_classwise: [82.6932373046875, 82.37419128417969] single-label/recall_classwise: [86.44656372070312, 77.78294372558594] single-label/f1-score_classwise: [84.52825927734375, 80.01275634765625] data_time: 0.0424 time: 1.3461 +2023/06/04 04:05:15 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230604_032014 +2023/06/04 04:05:19 - mmengine - INFO - Epoch(train) [2][ 100/3907] lr: 9.7685e-05 eta: 6:12:57 time: 0.6351 data_time: 0.0019 memory: 44138 loss: 0.4108 +2023/06/04 04:06:23 - mmengine - INFO - Epoch(train) [2][ 200/3907] lr: 9.7570e-05 eta: 6:11:51 time: 0.6370 data_time: 0.0021 memory: 44138 loss: 0.3643 +2023/06/04 04:07:26 - mmengine - INFO - Epoch(train) [2][ 300/3907] lr: 9.7451e-05 eta: 6:10:46 time: 0.6365 data_time: 0.0017 memory: 44138 loss: 0.4227 +2023/06/04 04:08:30 - mmengine - INFO - Epoch(train) [2][ 400/3907] lr: 9.7329e-05 eta: 6:09:41 time: 0.6369 data_time: 0.0019 memory: 44138 loss: 0.3847 +2023/06/04 04:09:34 - mmengine - INFO - Epoch(train) [2][ 500/3907] lr: 9.7205e-05 eta: 6:08:36 time: 0.6367 data_time: 0.0020 memory: 44138 loss: 0.3927 +2023/06/04 04:10:37 - mmengine - INFO - Epoch(train) [2][ 600/3907] lr: 9.7078e-05 eta: 6:07:31 time: 0.6362 data_time: 0.0018 memory: 44138 loss: 0.4163 +2023/06/04 04:11:41 - mmengine - INFO - Epoch(train) [2][ 700/3907] lr: 9.6949e-05 eta: 6:06:26 time: 0.6365 data_time: 0.0019 memory: 44138 loss: 0.3822 +2023/06/04 04:12:45 - mmengine - INFO - Epoch(train) [2][ 800/3907] lr: 9.6816e-05 eta: 6:05:22 time: 0.6364 data_time: 0.0015 memory: 44138 loss: 0.4025 +2023/06/04 04:13:48 - mmengine - INFO - Epoch(train) [2][ 900/3907] lr: 9.6681e-05 eta: 6:04:17 time: 0.6361 data_time: 0.0013 memory: 44138 loss: 0.3801 +2023/06/04 04:14:52 - mmengine - INFO - Epoch(train) [2][1000/3907] lr: 9.6544e-05 eta: 6:03:12 time: 0.6361 data_time: 0.0014 memory: 44138 loss: 0.3840 +2023/06/04 04:15:51 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230604_032014 +2023/06/04 04:15:56 - mmengine - INFO - Epoch(train) [2][1100/3907] lr: 9.6403e-05 eta: 6:02:07 time: 0.6359 data_time: 0.0018 memory: 44138 loss: 0.3984 +2023/06/04 04:16:59 - mmengine - INFO - Epoch(train) [2][1200/3907] lr: 9.6260e-05 eta: 6:01:02 time: 0.6361 data_time: 0.0017 memory: 44138 loss: 0.3811 +2023/06/04 04:18:03 - mmengine - INFO - Epoch(train) [2][1300/3907] lr: 9.6114e-05 eta: 5:59:59 time: 0.6363 data_time: 0.0014 memory: 44138 loss: 0.3815 +2023/06/04 04:19:07 - mmengine - INFO - Epoch(train) [2][1400/3907] lr: 9.5966e-05 eta: 5:58:54 time: 0.6360 data_time: 0.0016 memory: 44138 loss: 0.3766 +2023/06/04 04:20:10 - mmengine - INFO - Epoch(train) [2][1500/3907] lr: 9.5815e-05 eta: 5:57:49 time: 0.6353 data_time: 0.0018 memory: 44138 loss: 0.3874 +2023/06/04 04:21:14 - mmengine - INFO - Epoch(train) [2][1600/3907] lr: 9.5661e-05 eta: 5:56:44 time: 0.6346 data_time: 0.0023 memory: 44138 loss: 0.4038 +2023/06/04 04:22:18 - mmengine - INFO - Epoch(train) [2][1700/3907] lr: 9.5505e-05 eta: 5:55:39 time: 0.6350 data_time: 0.0017 memory: 44138 loss: 0.3885 +2023/06/04 04:23:21 - mmengine - INFO - Epoch(train) [2][1800/3907] lr: 9.5346e-05 eta: 5:54:34 time: 0.6366 data_time: 0.0021 memory: 44138 loss: 0.4086 +2023/06/04 04:24:25 - mmengine - INFO - Epoch(train) [2][1900/3907] lr: 9.5184e-05 eta: 5:53:30 time: 0.6453 data_time: 0.0014 memory: 44138 loss: 0.3796 +2023/06/04 04:25:29 - mmengine - INFO - Epoch(train) [2][2000/3907] lr: 9.5020e-05 eta: 5:52:26 time: 0.6359 data_time: 0.0014 memory: 44138 loss: 0.3962 +2023/06/04 04:26:28 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230604_032014 +2023/06/04 04:26:32 - mmengine - INFO - Epoch(train) [2][2100/3907] lr: 9.4854e-05 eta: 5:51:22 time: 0.6364 data_time: 0.0026 memory: 44138 loss: 0.3758 +2023/06/04 04:27:36 - mmengine - INFO - Epoch(train) [2][2200/3907] lr: 9.4684e-05 eta: 5:50:18 time: 0.6366 data_time: 0.0020 memory: 44138 loss: 0.4031 +2023/06/04 04:28:40 - mmengine - INFO - Epoch(train) [2][2300/3907] lr: 9.4512e-05 eta: 5:49:14 time: 0.6366 data_time: 0.0016 memory: 44138 loss: 0.3934 +2023/06/04 04:29:43 - mmengine - INFO - Epoch(train) [2][2400/3907] lr: 9.4338e-05 eta: 5:48:09 time: 0.6364 data_time: 0.0017 memory: 44138 loss: 0.3865 +2023/06/04 04:30:47 - mmengine - INFO - Epoch(train) [2][2500/3907] lr: 9.4161e-05 eta: 5:47:05 time: 0.6368 data_time: 0.0014 memory: 44138 loss: 0.4146 +2023/06/04 04:31:51 - mmengine - INFO - Epoch(train) [2][2600/3907] lr: 9.3981e-05 eta: 5:46:01 time: 0.6361 data_time: 0.0018 memory: 44138 loss: 0.3943 +2023/06/04 04:32:54 - mmengine - INFO - Epoch(train) [2][2700/3907] lr: 9.3799e-05 eta: 5:44:57 time: 0.6381 data_time: 0.0015 memory: 44138 loss: 0.4000 +2023/06/04 04:33:58 - mmengine - INFO - Epoch(train) [2][2800/3907] lr: 9.3615e-05 eta: 5:43:53 time: 0.6373 data_time: 0.0015 memory: 44138 loss: 0.3995 +2023/06/04 04:35:02 - mmengine - INFO - Epoch(train) [2][2900/3907] lr: 9.3428e-05 eta: 5:42:49 time: 0.6505 data_time: 0.0016 memory: 44138 loss: 0.3671 +2023/06/04 04:36:06 - mmengine - INFO - Epoch(train) [2][3000/3907] lr: 9.3238e-05 eta: 5:41:45 time: 0.6369 data_time: 0.0017 memory: 44138 loss: 0.3847 +2023/06/04 04:37:05 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230604_032014 +2023/06/04 04:37:09 - mmengine - INFO - Epoch(train) [2][3100/3907] lr: 9.3046e-05 eta: 5:40:42 time: 0.6369 data_time: 0.0020 memory: 44138 loss: 0.3961 +2023/06/04 04:38:13 - mmengine - INFO - Epoch(train) [2][3200/3907] lr: 9.2852e-05 eta: 5:39:38 time: 0.6376 data_time: 0.0014 memory: 44138 loss: 0.3566 +2023/06/04 04:39:17 - mmengine - INFO - Epoch(train) [2][3300/3907] lr: 9.2655e-05 eta: 5:38:34 time: 0.6371 data_time: 0.0016 memory: 44138 loss: 0.3788 +2023/06/04 04:40:21 - mmengine - INFO - Epoch(train) [2][3400/3907] lr: 9.2456e-05 eta: 5:37:30 time: 0.6357 data_time: 0.0017 memory: 44138 loss: 0.3777 +2023/06/04 04:41:24 - mmengine - INFO - Epoch(train) [2][3500/3907] lr: 9.2254e-05 eta: 5:36:25 time: 0.6352 data_time: 0.0017 memory: 44138 loss: 0.3705 +2023/06/04 04:42:28 - mmengine - INFO - Epoch(train) [2][3600/3907] lr: 9.2050e-05 eta: 5:35:21 time: 0.6363 data_time: 0.0020 memory: 44138 loss: 0.4027 +2023/06/04 04:43:31 - mmengine - INFO - Epoch(train) [2][3700/3907] lr: 9.1843e-05 eta: 5:34:16 time: 0.6356 data_time: 0.0016 memory: 44138 loss: 0.3668 +2023/06/04 04:44:35 - mmengine - INFO - Epoch(train) [2][3800/3907] lr: 9.1634e-05 eta: 5:33:12 time: 0.6362 data_time: 0.0013 memory: 44138 loss: 0.3868 +2023/06/04 04:45:39 - mmengine - INFO - Epoch(train) [2][3900/3907] lr: 9.1423e-05 eta: 5:32:09 time: 0.6365 data_time: 0.0016 memory: 44138 loss: 0.4031 +2023/06/04 04:45:43 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230604_032014 +2023/06/04 04:45:43 - mmengine - INFO - Saving checkpoint at 2 epochs +2023/06/04 04:47:17 - mmengine - INFO - Epoch(val) [2][57/57] accuracy/top1: 82.2971 single-label/precision_classwise: [77.39059448242188, 92.8665542602539] single-label/recall_classwise: [95.89672088623047, 65.59689331054688] single-label/f1-score_classwise: [85.65547943115234, 76.88533782958984] data_time: 0.0335 time: 1.2769 +2023/06/04 04:48:24 - mmengine - INFO - Epoch(train) [3][ 100/3907] lr: 9.1194e-05 eta: 5:31:10 time: 0.6362 data_time: 0.0019 memory: 44138 loss: 0.3771 +2023/06/04 04:49:18 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230604_032014 +2023/06/04 04:49:27 - mmengine - INFO - Epoch(train) [3][ 200/3907] lr: 9.0978e-05 eta: 5:30:06 time: 0.6369 data_time: 0.0018 memory: 44138 loss: 0.4005 +2023/06/04 04:50:31 - mmengine - INFO - Epoch(train) [3][ 300/3907] lr: 9.0759e-05 eta: 5:29:02 time: 0.6360 data_time: 0.0019 memory: 44138 loss: 0.3888 +2023/06/04 04:51:35 - mmengine - INFO - Epoch(train) [3][ 400/3907] lr: 9.0539e-05 eta: 5:27:58 time: 0.6370 data_time: 0.0015 memory: 44138 loss: 0.3539 +2023/06/04 04:52:39 - mmengine - INFO - Epoch(train) [3][ 500/3907] lr: 9.0315e-05 eta: 5:26:54 time: 0.6369 data_time: 0.0018 memory: 44138 loss: 0.4057 +2023/06/04 04:53:42 - mmengine - INFO - Epoch(train) [3][ 600/3907] lr: 9.0090e-05 eta: 5:25:50 time: 0.6359 data_time: 0.0016 memory: 44138 loss: 0.3652 +2023/06/04 04:54:46 - mmengine - INFO - Epoch(train) [3][ 700/3907] lr: 8.9862e-05 eta: 5:24:46 time: 0.6368 data_time: 0.0017 memory: 44138 loss: 0.3826 +2023/06/04 04:55:50 - mmengine - INFO - Epoch(train) [3][ 800/3907] lr: 8.9632e-05 eta: 5:23:42 time: 0.6369 data_time: 0.0015 memory: 44138 loss: 0.3654 +2023/06/04 04:56:53 - mmengine - INFO - Epoch(train) [3][ 900/3907] lr: 8.9400e-05 eta: 5:22:39 time: 0.6366 data_time: 0.0016 memory: 44138 loss: 0.3775 +2023/06/04 04:57:57 - mmengine - INFO - Epoch(train) [3][1000/3907] lr: 8.9166e-05 eta: 5:21:34 time: 0.6359 data_time: 0.0014 memory: 44138 loss: 0.3697 +2023/06/04 04:59:01 - mmengine - INFO - Epoch(train) [3][1100/3907] lr: 8.8929e-05 eta: 5:20:30 time: 0.6363 data_time: 0.0019 memory: 44138 loss: 0.3578 +2023/06/04 04:59:55 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230604_032014 +2023/06/04 05:00:04 - mmengine - INFO - Epoch(train) [3][1200/3907] lr: 8.8691e-05 eta: 5:19:25 time: 0.6339 data_time: 0.0018 memory: 44138 loss: 0.3712 +2023/06/04 05:01:08 - mmengine - INFO - Epoch(train) [3][1300/3907] lr: 8.8450e-05 eta: 5:18:21 time: 0.6336 data_time: 0.0015 memory: 44138 loss: 0.3593 +2023/06/04 05:02:11 - mmengine - INFO - Epoch(train) [3][1400/3907] lr: 8.8206e-05 eta: 5:17:16 time: 0.6365 data_time: 0.0025 memory: 44138 loss: 0.3558 +2023/06/04 05:03:15 - mmengine - INFO - Epoch(train) [3][1500/3907] lr: 8.7961e-05 eta: 5:16:13 time: 0.6365 data_time: 0.0018 memory: 44138 loss: 0.3819 +2023/06/04 05:04:19 - mmengine - INFO - Epoch(train) [3][1600/3907] lr: 8.7714e-05 eta: 5:15:09 time: 0.6368 data_time: 0.0020 memory: 44138 loss: 0.4081 +2023/06/04 05:05:22 - mmengine - INFO - Epoch(train) [3][1700/3907] lr: 8.7464e-05 eta: 5:14:05 time: 0.6367 data_time: 0.0015 memory: 44138 loss: 0.3532 +2023/06/04 05:06:26 - mmengine - INFO - Epoch(train) [3][1800/3907] lr: 8.7213e-05 eta: 5:13:01 time: 0.6359 data_time: 0.0018 memory: 44138 loss: 0.3661 +2023/06/04 05:07:30 - mmengine - INFO - Epoch(train) [3][1900/3907] lr: 8.6959e-05 eta: 5:11:56 time: 0.6363 data_time: 0.0014 memory: 44138 loss: 0.3587 +2023/06/04 05:08:33 - mmengine - INFO - Epoch(train) [3][2000/3907] lr: 8.6703e-05 eta: 5:10:52 time: 0.6369 data_time: 0.0020 memory: 44138 loss: 0.3939 +2023/06/04 05:09:37 - mmengine - INFO - Epoch(train) [3][2100/3907] lr: 8.6445e-05 eta: 5:09:48 time: 0.6371 data_time: 0.0015 memory: 44138 loss: 0.3632 +2023/06/04 05:10:32 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230604_032014 +2023/06/04 05:10:41 - mmengine - INFO - Epoch(train) [3][2200/3907] lr: 8.6186e-05 eta: 5:08:44 time: 0.6363 data_time: 0.0017 memory: 44138 loss: 0.3913 +2023/06/04 05:11:45 - mmengine - INFO - Epoch(train) [3][2300/3907] lr: 8.5924e-05 eta: 5:07:41 time: 0.6360 data_time: 0.0015 memory: 44138 loss: 0.3793 +2023/06/04 05:12:48 - mmengine - INFO - Epoch(train) [3][2400/3907] lr: 8.5660e-05 eta: 5:06:37 time: 0.6377 data_time: 0.0015 memory: 44138 loss: 0.3770 +2023/06/04 05:13:52 - mmengine - INFO - Epoch(train) [3][2500/3907] lr: 8.5394e-05 eta: 5:05:33 time: 0.6368 data_time: 0.0016 memory: 44138 loss: 0.3707 +2023/06/04 05:14:56 - mmengine - INFO - Epoch(train) [3][2600/3907] lr: 8.5126e-05 eta: 5:04:29 time: 0.6367 data_time: 0.0018 memory: 44138 loss: 0.3768 +2023/06/04 05:16:00 - mmengine - INFO - Epoch(train) [3][2700/3907] lr: 8.4856e-05 eta: 5:03:25 time: 0.6367 data_time: 0.0016 memory: 44138 loss: 0.3334 +2023/06/04 05:17:03 - mmengine - INFO - Epoch(train) [3][2800/3907] lr: 8.4585e-05 eta: 5:02:21 time: 0.6366 data_time: 0.0015 memory: 44138 loss: 0.3791 +2023/06/04 05:18:07 - mmengine - INFO - Epoch(train) [3][2900/3907] lr: 8.4311e-05 eta: 5:01:17 time: 0.6349 data_time: 0.0016 memory: 44138 loss: 0.3891 +2023/06/04 05:19:10 - mmengine - INFO - Epoch(train) [3][3000/3907] lr: 8.4036e-05 eta: 5:00:13 time: 0.6409 data_time: 0.0019 memory: 44138 loss: 0.3705 +2023/06/04 05:20:14 - mmengine - INFO - Epoch(train) [3][3100/3907] lr: 8.3758e-05 eta: 4:59:10 time: 0.6372 data_time: 0.0023 memory: 44138 loss: 0.3623 +2023/06/04 05:21:09 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230604_032014 +2023/06/04 05:21:18 - mmengine - INFO - Epoch(train) [3][3200/3907] lr: 8.3479e-05 eta: 4:58:06 time: 0.6368 data_time: 0.0016 memory: 44138 loss: 0.3718 +2023/06/04 05:22:22 - mmengine - INFO - Epoch(train) [3][3300/3907] lr: 8.3198e-05 eta: 4:57:02 time: 0.6369 data_time: 0.0013 memory: 44138 loss: 0.3968 +2023/06/04 05:23:26 - mmengine - INFO - Epoch(train) [3][3400/3907] lr: 8.2915e-05 eta: 4:55:58 time: 0.6373 data_time: 0.0020 memory: 44138 loss: 0.3573 +2023/06/04 05:24:29 - mmengine - INFO - Epoch(train) [3][3500/3907] lr: 8.2630e-05 eta: 4:54:55 time: 0.6372 data_time: 0.0018 memory: 44138 loss: 0.3618 +2023/06/04 05:25:33 - mmengine - INFO - Epoch(train) [3][3600/3907] lr: 8.2344e-05 eta: 4:53:51 time: 0.6355 data_time: 0.0016 memory: 44138 loss: 0.3846 +2023/06/04 05:26:37 - mmengine - INFO - Epoch(train) [3][3700/3907] lr: 8.2056e-05 eta: 4:52:47 time: 0.6368 data_time: 0.0018 memory: 44138 loss: 0.3666 +2023/06/04 05:27:40 - mmengine - INFO - Epoch(train) [3][3800/3907] lr: 8.1765e-05 eta: 4:51:43 time: 0.6360 data_time: 0.0018 memory: 44138 loss: 0.3668 +2023/06/04 05:28:44 - mmengine - INFO - Epoch(train) [3][3900/3907] lr: 8.1474e-05 eta: 4:50:39 time: 0.6358 data_time: 0.0017 memory: 44138 loss: 0.3822 +2023/06/04 05:28:48 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230604_032014 +2023/06/04 05:28:48 - mmengine - INFO - Saving checkpoint at 3 epochs +2023/06/04 05:30:23 - mmengine - INFO - Epoch(val) [3][57/57] accuracy/top1: 80.4147 single-label/precision_classwise: [74.61434936523438, 95.46080017089844] single-label/recall_classwise: [97.7084732055664, 59.17829132080078] single-label/f1-score_classwise: [84.6139144897461, 73.0631103515625] data_time: 0.0314 time: 1.2761 +2023/06/04 05:31:30 - mmengine - INFO - Epoch(train) [4][ 100/3907] lr: 8.1160e-05 eta: 4:49:37 time: 0.6379 data_time: 0.0022 memory: 44138 loss: 0.4032 +2023/06/04 05:32:33 - mmengine - INFO - Epoch(train) [4][ 200/3907] lr: 8.0864e-05 eta: 4:48:33 time: 0.6371 data_time: 0.0016 memory: 44138 loss: 0.3555 +2023/06/04 05:33:24 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230604_032014 +2023/06/04 05:33:37 - mmengine - INFO - Epoch(train) [4][ 300/3907] lr: 8.0567e-05 eta: 4:47:29 time: 0.6369 data_time: 0.0015 memory: 44138 loss: 0.3699 +2023/06/04 05:34:41 - mmengine - INFO - Epoch(train) [4][ 400/3907] lr: 8.0269e-05 eta: 4:46:26 time: 0.6353 data_time: 0.0017 memory: 44138 loss: 0.3652 +2023/06/04 05:35:45 - mmengine - INFO - Epoch(train) [4][ 500/3907] lr: 7.9969e-05 eta: 4:45:22 time: 0.6371 data_time: 0.0020 memory: 44138 loss: 0.3850 +2023/06/04 05:36:48 - mmengine - INFO - Epoch(train) [4][ 600/3907] lr: 7.9667e-05 eta: 4:44:18 time: 0.6365 data_time: 0.0021 memory: 44138 loss: 0.3886 +2023/06/04 05:37:52 - mmengine - INFO - Epoch(train) [4][ 700/3907] lr: 7.9363e-05 eta: 4:43:14 time: 0.6373 data_time: 0.0015 memory: 44138 loss: 0.3622 +2023/06/04 05:38:56 - mmengine - INFO - Epoch(train) [4][ 800/3907] lr: 7.9058e-05 eta: 4:42:10 time: 0.6365 data_time: 0.0021 memory: 44138 loss: 0.3656 +2023/06/04 05:40:00 - mmengine - INFO - Epoch(train) [4][ 900/3907] lr: 7.8752e-05 eta: 4:41:06 time: 0.6368 data_time: 0.0022 memory: 44138 loss: 0.3673 +2023/06/04 05:41:03 - mmengine - INFO - Epoch(train) [4][1000/3907] lr: 7.8444e-05 eta: 4:40:02 time: 0.6371 data_time: 0.0020 memory: 44138 loss: 0.3458 +2023/06/04 05:42:07 - mmengine - INFO - Epoch(train) [4][1100/3907] lr: 7.8134e-05 eta: 4:38:59 time: 0.6369 data_time: 0.0015 memory: 44138 loss: 0.3747 +2023/06/04 05:43:11 - mmengine - INFO - Epoch(train) [4][1200/3907] lr: 7.7823e-05 eta: 4:37:55 time: 0.6367 data_time: 0.0023 memory: 44138 loss: 0.3862 +2023/06/04 05:44:01 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230604_032014 +2023/06/04 05:44:15 - mmengine - INFO - Epoch(train) [4][1300/3907] lr: 7.7510e-05 eta: 4:36:51 time: 0.6365 data_time: 0.0013 memory: 44138 loss: 0.3562 +2023/06/04 05:45:18 - mmengine - INFO - Epoch(train) [4][1400/3907] lr: 7.7196e-05 eta: 4:35:47 time: 0.6354 data_time: 0.0017 memory: 44138 loss: 0.3680 +2023/06/04 05:46:22 - mmengine - INFO - Epoch(train) [4][1500/3907] lr: 7.6881e-05 eta: 4:34:43 time: 0.6365 data_time: 0.0017 memory: 44138 loss: 0.3740 +2023/06/04 05:47:26 - mmengine - INFO - Epoch(train) [4][1600/3907] lr: 7.6564e-05 eta: 4:33:39 time: 0.6369 data_time: 0.0017 memory: 44138 loss: 0.3585 +2023/06/04 05:48:29 - mmengine - INFO - Epoch(train) [4][1700/3907] lr: 7.6246e-05 eta: 4:32:35 time: 0.6363 data_time: 0.0019 memory: 44138 loss: 0.3506 +2023/06/04 05:49:33 - mmengine - INFO - Epoch(train) [4][1800/3907] lr: 7.5926e-05 eta: 4:31:31 time: 0.6370 data_time: 0.0015 memory: 44138 loss: 0.3546 +2023/06/04 05:50:37 - mmengine - INFO - Epoch(train) [4][1900/3907] lr: 7.5605e-05 eta: 4:30:27 time: 0.6371 data_time: 0.0015 memory: 44138 loss: 0.3733 +2023/06/04 05:51:40 - mmengine - INFO - Epoch(train) [4][2000/3907] lr: 7.5283e-05 eta: 4:29:24 time: 0.6380 data_time: 0.0016 memory: 44138 loss: 0.3841 +2023/06/04 05:52:44 - mmengine - INFO - Epoch(train) [4][2100/3907] lr: 7.4959e-05 eta: 4:28:20 time: 0.6373 data_time: 0.0018 memory: 44138 loss: 0.3515 +2023/06/04 05:53:48 - mmengine - INFO - Epoch(train) [4][2200/3907] lr: 7.4634e-05 eta: 4:27:16 time: 0.6366 data_time: 0.0016 memory: 44138 loss: 0.3763 +2023/06/04 05:54:38 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230604_032014 +2023/06/04 05:54:52 - mmengine - INFO - Epoch(train) [4][2300/3907] lr: 7.4308e-05 eta: 4:26:12 time: 0.6374 data_time: 0.0022 memory: 44138 loss: 0.3766 +2023/06/04 05:55:55 - mmengine - INFO - Epoch(train) [4][2400/3907] lr: 7.3980e-05 eta: 4:25:08 time: 0.6369 data_time: 0.0018 memory: 44138 loss: 0.4018 +2023/06/04 05:56:59 - mmengine - INFO - Epoch(train) [4][2500/3907] lr: 7.3652e-05 eta: 4:24:05 time: 0.6372 data_time: 0.0016 memory: 44138 loss: 0.3958 +2023/06/04 05:58:03 - mmengine - INFO - Epoch(train) [4][2600/3907] lr: 7.3322e-05 eta: 4:23:01 time: 0.6361 data_time: 0.0016 memory: 44138 loss: 0.3598 +2023/06/04 05:59:06 - mmengine - INFO - Epoch(train) [4][2700/3907] lr: 7.2991e-05 eta: 4:21:57 time: 0.6360 data_time: 0.0016 memory: 44138 loss: 0.3618 +2023/06/04 06:00:10 - mmengine - INFO - Epoch(train) [4][2800/3907] lr: 7.2659e-05 eta: 4:20:53 time: 0.6367 data_time: 0.0017 memory: 44138 loss: 0.3716 +2023/06/04 06:01:14 - mmengine - INFO - Epoch(train) [4][2900/3907] lr: 7.2325e-05 eta: 4:19:49 time: 0.6368 data_time: 0.0016 memory: 44138 loss: 0.3589 +2023/06/04 06:02:18 - mmengine - INFO - Epoch(train) [4][3000/3907] lr: 7.1991e-05 eta: 4:18:45 time: 0.6368 data_time: 0.0013 memory: 44138 loss: 0.3527 +2023/06/04 06:03:21 - mmengine - INFO - Epoch(train) [4][3100/3907] lr: 7.1655e-05 eta: 4:17:41 time: 0.6367 data_time: 0.0023 memory: 44138 loss: 0.3418 +2023/06/04 06:04:25 - mmengine - INFO - Epoch(train) [4][3200/3907] lr: 7.1318e-05 eta: 4:16:38 time: 0.6368 data_time: 0.0015 memory: 44138 loss: 0.3678 +2023/06/04 06:05:16 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230604_032014 +2023/06/04 06:05:29 - mmengine - INFO - Epoch(train) [4][3300/3907] lr: 7.0981e-05 eta: 4:15:34 time: 0.6376 data_time: 0.0017 memory: 44138 loss: 0.3794 +2023/06/04 06:06:33 - mmengine - INFO - Epoch(train) [4][3400/3907] lr: 7.0642e-05 eta: 4:14:30 time: 0.6390 data_time: 0.0026 memory: 44138 loss: 0.3691 +2023/06/04 06:07:36 - mmengine - INFO - Epoch(train) [4][3500/3907] lr: 7.0302e-05 eta: 4:13:26 time: 0.6369 data_time: 0.0018 memory: 44138 loss: 0.3650 +2023/06/04 06:08:40 - mmengine - INFO - Epoch(train) [4][3600/3907] lr: 6.9961e-05 eta: 4:12:23 time: 0.6372 data_time: 0.0016 memory: 44138 loss: 0.3644 +2023/06/04 06:09:44 - mmengine - INFO - Epoch(train) [4][3700/3907] lr: 6.9620e-05 eta: 4:11:19 time: 0.6380 data_time: 0.0017 memory: 44138 loss: 0.3943 +2023/06/04 06:10:48 - mmengine - INFO - Epoch(train) [4][3800/3907] lr: 6.9277e-05 eta: 4:10:16 time: 0.6376 data_time: 0.0017 memory: 44138 loss: 0.3708 +2023/06/04 06:11:52 - mmengine - INFO - Epoch(train) [4][3900/3907] lr: 6.8933e-05 eta: 4:09:12 time: 0.6337 data_time: 0.0013 memory: 44138 loss: 0.3691 +2023/06/04 06:11:56 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230604_032014 +2023/06/04 06:11:56 - mmengine - INFO - Saving checkpoint at 4 epochs +2023/06/04 06:13:31 - mmengine - INFO - Epoch(val) [4][57/57] accuracy/top1: 78.9047 single-label/precision_classwise: [72.86074829101562, 96.47858428955078] single-label/recall_classwise: [98.36500549316406, 55.00775146484375] single-label/f1-score_classwise: [83.71343231201172, 70.066650390625] data_time: 0.0329 time: 1.2754 +2023/06/04 06:14:38 - mmengine - INFO - Epoch(train) [5][ 100/3907] lr: 6.8565e-05 eta: 4:08:07 time: 0.6366 data_time: 0.0016 memory: 44138 loss: 0.3542 +2023/06/04 06:15:41 - mmengine - INFO - Epoch(train) [5][ 200/3907] lr: 6.8219e-05 eta: 4:07:03 time: 0.6366 data_time: 0.0016 memory: 44138 loss: 0.3769 +2023/06/04 06:16:45 - mmengine - INFO - Epoch(train) [5][ 300/3907] lr: 6.7873e-05 eta: 4:05:59 time: 0.6365 data_time: 0.0016 memory: 44138 loss: 0.3566 +2023/06/04 06:17:31 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230604_032014 +2023/06/04 06:17:49 - mmengine - INFO - Epoch(train) [5][ 400/3907] lr: 6.7526e-05 eta: 4:04:55 time: 0.6369 data_time: 0.0016 memory: 44138 loss: 0.3774 +2023/06/04 06:18:52 - mmengine - INFO - Epoch(train) [5][ 500/3907] lr: 6.7178e-05 eta: 4:03:51 time: 0.6354 data_time: 0.0015 memory: 44138 loss: 0.3550 +2023/06/04 06:19:56 - mmengine - INFO - Epoch(train) [5][ 600/3907] lr: 6.6829e-05 eta: 4:02:47 time: 0.6362 data_time: 0.0022 memory: 44138 loss: 0.3727 +2023/06/04 06:21:00 - mmengine - INFO - Epoch(train) [5][ 700/3907] lr: 6.6480e-05 eta: 4:01:44 time: 0.6363 data_time: 0.0016 memory: 44138 loss: 0.3672 +2023/06/04 06:22:03 - mmengine - INFO - Epoch(train) [5][ 800/3907] lr: 6.6129e-05 eta: 4:00:40 time: 0.6365 data_time: 0.0015 memory: 44138 loss: 0.3618 +2023/06/04 06:23:07 - mmengine - INFO - Epoch(train) [5][ 900/3907] lr: 6.5778e-05 eta: 3:59:36 time: 0.6362 data_time: 0.0015 memory: 44138 loss: 0.3405 +2023/06/04 06:24:11 - mmengine - INFO - Epoch(train) [5][1000/3907] lr: 6.5427e-05 eta: 3:58:32 time: 0.6367 data_time: 0.0020 memory: 44138 loss: 0.3624 +2023/06/04 06:25:14 - mmengine - INFO - Epoch(train) [5][1100/3907] lr: 6.5074e-05 eta: 3:57:28 time: 0.6365 data_time: 0.0015 memory: 44138 loss: 0.3517 +2023/06/04 06:26:18 - mmengine - INFO - Epoch(train) [5][1200/3907] lr: 6.4721e-05 eta: 3:56:24 time: 0.6364 data_time: 0.0016 memory: 44138 loss: 0.3515 +2023/06/04 06:27:21 - mmengine - INFO - Epoch(train) [5][1300/3907] lr: 6.4368e-05 eta: 3:55:20 time: 0.6377 data_time: 0.0023 memory: 44138 loss: 0.3611 +2023/06/04 06:28:07 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230604_032014 +2023/06/04 06:28:25 - mmengine - INFO - Epoch(train) [5][1400/3907] lr: 6.4014e-05 eta: 3:54:16 time: 0.6365 data_time: 0.0016 memory: 44138 loss: 0.3713 +2023/06/04 06:29:29 - mmengine - INFO - Epoch(train) [5][1500/3907] lr: 6.3659e-05 eta: 3:53:12 time: 0.6374 data_time: 0.0019 memory: 44138 loss: 0.3826 +2023/06/04 06:30:33 - mmengine - INFO - Epoch(train) [5][1600/3907] lr: 6.3303e-05 eta: 3:52:08 time: 0.6374 data_time: 0.0016 memory: 44138 loss: 0.3695 +2023/06/04 06:31:36 - mmengine - INFO - Epoch(train) [5][1700/3907] lr: 6.2948e-05 eta: 3:51:04 time: 0.6376 data_time: 0.0023 memory: 44138 loss: 0.3665 +2023/06/04 06:32:40 - mmengine - INFO - Epoch(train) [5][1800/3907] lr: 6.2591e-05 eta: 3:50:01 time: 0.6365 data_time: 0.0016 memory: 44138 loss: 0.3638 +2023/06/04 06:33:44 - mmengine - INFO - Epoch(train) [5][1900/3907] lr: 6.2234e-05 eta: 3:48:57 time: 0.6363 data_time: 0.0015 memory: 44138 loss: 0.3768 +2023/06/04 06:34:48 - mmengine - INFO - Epoch(train) [5][2000/3907] lr: 6.1877e-05 eta: 3:47:53 time: 0.6367 data_time: 0.0016 memory: 44138 loss: 0.3679 +2023/06/04 06:35:51 - mmengine - INFO - Epoch(train) [5][2100/3907] lr: 6.1519e-05 eta: 3:46:49 time: 0.6370 data_time: 0.0019 memory: 44138 loss: 0.3669 +2023/06/04 06:36:55 - mmengine - INFO - Epoch(train) [5][2200/3907] lr: 6.1161e-05 eta: 3:45:45 time: 0.6368 data_time: 0.0017 memory: 44138 loss: 0.3675 +2023/06/04 06:37:58 - mmengine - INFO - Epoch(train) [5][2300/3907] lr: 6.0802e-05 eta: 3:44:41 time: 0.6353 data_time: 0.0015 memory: 44138 loss: 0.3405 +2023/06/04 06:38:44 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230604_032014 +2023/06/04 06:39:02 - mmengine - INFO - Epoch(train) [5][2400/3907] lr: 6.0443e-05 eta: 3:43:37 time: 0.6370 data_time: 0.0021 memory: 44138 loss: 0.3934 +2023/06/04 06:40:06 - mmengine - INFO - Epoch(train) [5][2500/3907] lr: 6.0084e-05 eta: 3:42:33 time: 0.6371 data_time: 0.0018 memory: 44138 loss: 0.3674 +2023/06/04 06:41:10 - mmengine - INFO - Epoch(train) [5][2600/3907] lr: 5.9724e-05 eta: 3:41:30 time: 0.6366 data_time: 0.0020 memory: 44138 loss: 0.3526 +2023/06/04 06:42:13 - mmengine - INFO - Epoch(train) [5][2700/3907] lr: 5.9364e-05 eta: 3:40:26 time: 0.6367 data_time: 0.0016 memory: 44138 loss: 0.3733 +2023/06/04 06:43:17 - mmengine - INFO - Epoch(train) [5][2800/3907] lr: 5.9004e-05 eta: 3:39:22 time: 0.6372 data_time: 0.0019 memory: 44138 loss: 0.3880 +2023/06/04 06:44:21 - mmengine - INFO - Epoch(train) [5][2900/3907] lr: 5.8643e-05 eta: 3:38:18 time: 0.6373 data_time: 0.0018 memory: 44138 loss: 0.3906 +2023/06/04 06:45:24 - mmengine - INFO - Epoch(train) [5][3000/3907] lr: 5.8283e-05 eta: 3:37:14 time: 0.6367 data_time: 0.0019 memory: 44138 loss: 0.3969 +2023/06/04 06:46:28 - mmengine - INFO - Epoch(train) [5][3100/3907] lr: 5.7922e-05 eta: 3:36:11 time: 0.6370 data_time: 0.0017 memory: 44138 loss: 0.3747 +2023/06/04 06:47:32 - mmengine - INFO - Epoch(train) [5][3200/3907] lr: 5.7560e-05 eta: 3:35:07 time: 0.6371 data_time: 0.0018 memory: 44138 loss: 0.3617 +2023/06/04 06:48:36 - mmengine - INFO - Epoch(train) [5][3300/3907] lr: 5.7199e-05 eta: 3:34:03 time: 0.6363 data_time: 0.0021 memory: 44138 loss: 0.3736 +2023/06/04 06:49:22 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230604_032014 +2023/06/04 06:49:39 - mmengine - INFO - Epoch(train) [5][3400/3907] lr: 5.6838e-05 eta: 3:32:59 time: 0.6386 data_time: 0.0021 memory: 44138 loss: 0.3510 +2023/06/04 06:50:43 - mmengine - INFO - Epoch(train) [5][3500/3907] lr: 5.6476e-05 eta: 3:31:55 time: 0.6362 data_time: 0.0016 memory: 44138 loss: 0.3507 +2023/06/04 06:51:47 - mmengine - INFO - Epoch(train) [5][3600/3907] lr: 5.6114e-05 eta: 3:30:52 time: 0.6368 data_time: 0.0016 memory: 44138 loss: 0.3837 +2023/06/04 06:52:51 - mmengine - INFO - Epoch(train) [5][3700/3907] lr: 5.5753e-05 eta: 3:29:48 time: 0.6369 data_time: 0.0019 memory: 44138 loss: 0.4033 +2023/06/04 06:53:54 - mmengine - INFO - Epoch(train) [5][3800/3907] lr: 5.5391e-05 eta: 3:28:44 time: 0.6371 data_time: 0.0015 memory: 44138 loss: 0.3668 +2023/06/04 06:54:58 - mmengine - INFO - Epoch(train) [5][3900/3907] lr: 5.5029e-05 eta: 3:27:40 time: 0.6353 data_time: 0.0014 memory: 44138 loss: 0.3448 +2023/06/04 06:55:02 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230604_032014 +2023/06/04 06:55:02 - mmengine - INFO - Saving checkpoint at 5 epochs +2023/06/04 06:56:37 - mmengine - INFO - Epoch(val) [5][57/57] accuracy/top1: 78.0314 single-label/precision_classwise: [71.9728775024414, 96.62985229492188] single-label/recall_classwise: [98.49756622314453, 52.89922332763672] single-label/f1-score_classwise: [83.17163848876953, 68.3698959350586] data_time: 0.0327 time: 1.2760 +2023/06/04 06:57:44 - mmengine - INFO - Epoch(train) [6][ 100/3907] lr: 5.4642e-05 eta: 3:26:35 time: 0.6373 data_time: 0.0025 memory: 44138 loss: 0.3709 +2023/06/04 06:58:47 - mmengine - INFO - Epoch(train) [6][ 200/3907] lr: 5.4280e-05 eta: 3:25:31 time: 0.6355 data_time: 0.0022 memory: 44138 loss: 0.3720 +2023/06/04 06:59:51 - mmengine - INFO - Epoch(train) [6][ 300/3907] lr: 5.3918e-05 eta: 3:24:27 time: 0.6366 data_time: 0.0016 memory: 44138 loss: 0.3533 +2023/06/04 07:00:55 - mmengine - INFO - Epoch(train) [6][ 400/3907] lr: 5.3556e-05 eta: 3:23:23 time: 0.6380 data_time: 0.0015 memory: 44138 loss: 0.3736 +2023/06/04 07:01:36 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230604_032014 +2023/06/04 07:01:59 - mmengine - INFO - Epoch(train) [6][ 500/3907] lr: 5.3195e-05 eta: 3:22:19 time: 0.6368 data_time: 0.0016 memory: 44138 loss: 0.3663 +2023/06/04 07:03:02 - mmengine - INFO - Epoch(train) [6][ 600/3907] lr: 5.2833e-05 eta: 3:21:15 time: 0.6369 data_time: 0.0015 memory: 44138 loss: 0.3887 +2023/06/04 07:04:06 - mmengine - INFO - Epoch(train) [6][ 700/3907] lr: 5.2472e-05 eta: 3:20:11 time: 0.6369 data_time: 0.0015 memory: 44138 loss: 0.3625 +2023/06/04 07:05:10 - mmengine - INFO - Epoch(train) [6][ 800/3907] lr: 5.2111e-05 eta: 3:19:08 time: 0.6373 data_time: 0.0016 memory: 44138 loss: 0.3593 +2023/06/04 07:06:13 - mmengine - INFO - Epoch(train) [6][ 900/3907] lr: 5.1750e-05 eta: 3:18:04 time: 0.6356 data_time: 0.0015 memory: 44138 loss: 0.3751 +2023/06/04 07:07:17 - mmengine - INFO - Epoch(train) [6][1000/3907] lr: 5.1389e-05 eta: 3:17:00 time: 0.6363 data_time: 0.0017 memory: 44138 loss: 0.3484 +2023/06/04 07:08:21 - mmengine - INFO - Epoch(train) [6][1100/3907] lr: 5.1029e-05 eta: 3:15:56 time: 0.6363 data_time: 0.0019 memory: 44138 loss: 0.3503 +2023/06/04 07:09:24 - mmengine - INFO - Epoch(train) [6][1200/3907] lr: 5.0668e-05 eta: 3:14:52 time: 0.6358 data_time: 0.0017 memory: 44138 loss: 0.3673 +2023/06/04 07:10:28 - mmengine - INFO - Epoch(train) [6][1300/3907] lr: 5.0308e-05 eta: 3:13:48 time: 0.6368 data_time: 0.0018 memory: 44138 loss: 0.3620 +2023/06/04 07:11:32 - mmengine - INFO - Epoch(train) [6][1400/3907] lr: 4.9949e-05 eta: 3:12:44 time: 0.6370 data_time: 0.0014 memory: 44138 loss: 0.3625 +2023/06/04 07:12:13 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230604_032014 +2023/06/04 07:12:35 - mmengine - INFO - Epoch(train) [6][1500/3907] lr: 4.9589e-05 eta: 3:11:41 time: 0.6479 data_time: 0.0017 memory: 44138 loss: 0.3599 +2023/06/04 07:13:39 - mmengine - INFO - Epoch(train) [6][1600/3907] lr: 4.9230e-05 eta: 3:10:37 time: 0.6390 data_time: 0.0020 memory: 44138 loss: 0.3876 +2023/06/04 07:14:43 - mmengine - INFO - Epoch(train) [6][1700/3907] lr: 4.8871e-05 eta: 3:09:33 time: 0.6366 data_time: 0.0018 memory: 44138 loss: 0.3581 +2023/06/04 07:15:47 - mmengine - INFO - Epoch(train) [6][1800/3907] lr: 4.8513e-05 eta: 3:08:29 time: 0.6365 data_time: 0.0017 memory: 44138 loss: 0.3812 +2023/06/04 07:16:50 - mmengine - INFO - Epoch(train) [6][1900/3907] lr: 4.8155e-05 eta: 3:07:25 time: 0.6373 data_time: 0.0017 memory: 44138 loss: 0.3990 +2023/06/04 07:17:54 - mmengine - INFO - Epoch(train) [6][2000/3907] lr: 4.7798e-05 eta: 3:06:22 time: 0.6368 data_time: 0.0018 memory: 44138 loss: 0.3996 +2023/06/04 07:18:58 - mmengine - INFO - Epoch(train) [6][2100/3907] lr: 4.7441e-05 eta: 3:05:18 time: 0.6373 data_time: 0.0018 memory: 44138 loss: 0.3687 +2023/06/04 07:20:01 - mmengine - INFO - Epoch(train) [6][2200/3907] lr: 4.7084e-05 eta: 3:04:14 time: 0.6372 data_time: 0.0018 memory: 44138 loss: 0.3547 +2023/06/04 07:21:05 - mmengine - INFO - Epoch(train) [6][2300/3907] lr: 4.6729e-05 eta: 3:03:10 time: 0.6371 data_time: 0.0018 memory: 44138 loss: 0.3685 +2023/06/04 07:22:09 - mmengine - INFO - Epoch(train) [6][2400/3907] lr: 4.6373e-05 eta: 3:02:06 time: 0.6371 data_time: 0.0015 memory: 44138 loss: 0.3577 +2023/06/04 07:22:50 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230604_032014 +2023/06/04 07:23:13 - mmengine - INFO - Epoch(train) [6][2500/3907] lr: 4.6018e-05 eta: 3:01:03 time: 0.6368 data_time: 0.0015 memory: 44138 loss: 0.3813 +2023/06/04 07:24:16 - mmengine - INFO - Epoch(train) [6][2600/3907] lr: 4.5664e-05 eta: 2:59:59 time: 0.6374 data_time: 0.0015 memory: 44138 loss: 0.3584 +2023/06/04 07:25:20 - mmengine - INFO - Epoch(train) [6][2700/3907] lr: 4.5310e-05 eta: 2:58:55 time: 0.6380 data_time: 0.0017 memory: 44138 loss: 0.3853 +2023/06/04 07:26:24 - mmengine - INFO - Epoch(train) [6][2800/3907] lr: 4.4957e-05 eta: 2:57:51 time: 0.6368 data_time: 0.0022 memory: 44138 loss: 0.3722 +2023/06/04 07:27:28 - mmengine - INFO - Epoch(train) [6][2900/3907] lr: 4.4605e-05 eta: 2:56:47 time: 0.6365 data_time: 0.0015 memory: 44138 loss: 0.3711 +2023/06/04 07:28:32 - mmengine - INFO - Epoch(train) [6][3000/3907] lr: 4.4253e-05 eta: 2:55:44 time: 0.6369 data_time: 0.0017 memory: 44138 loss: 0.3819 +2023/06/04 07:29:35 - mmengine - INFO - Epoch(train) [6][3100/3907] lr: 4.3902e-05 eta: 2:54:40 time: 0.6372 data_time: 0.0024 memory: 44138 loss: 0.3397 +2023/06/04 07:30:39 - mmengine - INFO - Epoch(train) [6][3200/3907] lr: 4.3552e-05 eta: 2:53:36 time: 0.6370 data_time: 0.0017 memory: 44138 loss: 0.3634 +2023/06/04 07:31:43 - mmengine - INFO - Epoch(train) [6][3300/3907] lr: 4.3202e-05 eta: 2:52:32 time: 0.6359 data_time: 0.0021 memory: 44138 loss: 0.3723 +2023/06/04 07:32:46 - mmengine - INFO - Epoch(train) [6][3400/3907] lr: 4.2854e-05 eta: 2:51:28 time: 0.6358 data_time: 0.0017 memory: 44138 loss: 0.3690 +2023/06/04 07:33:28 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230604_032014 +2023/06/04 07:33:50 - mmengine - INFO - Epoch(train) [6][3500/3907] lr: 4.2506e-05 eta: 2:50:25 time: 0.6358 data_time: 0.0017 memory: 44138 loss: 0.3821 +2023/06/04 07:34:54 - mmengine - INFO - Epoch(train) [6][3600/3907] lr: 4.2158e-05 eta: 2:49:21 time: 0.6364 data_time: 0.0018 memory: 44138 loss: 0.3863 +2023/06/04 07:35:57 - mmengine - INFO - Epoch(train) [6][3700/3907] lr: 4.1812e-05 eta: 2:48:17 time: 0.6360 data_time: 0.0019 memory: 44138 loss: 0.3671 +2023/06/04 07:37:01 - mmengine - INFO - Epoch(train) [6][3800/3907] lr: 4.1466e-05 eta: 2:47:13 time: 0.6363 data_time: 0.0018 memory: 44138 loss: 0.3668 +2023/06/04 07:38:05 - mmengine - INFO - Epoch(train) [6][3900/3907] lr: 4.1122e-05 eta: 2:46:09 time: 0.6367 data_time: 0.0017 memory: 44138 loss: 0.3598 +2023/06/04 07:38:09 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230604_032014 +2023/06/04 07:38:09 - mmengine - INFO - Saving checkpoint at 6 epochs +2023/06/04 07:39:44 - mmengine - INFO - Epoch(val) [6][57/57] accuracy/top1: 77.8087 single-label/precision_classwise: [71.67727661132812, 97.16517639160156] single-label/recall_classwise: [98.76270294189453, 52.077518463134766] single-label/f1-score_classwise: [83.06785583496094, 67.81063842773438] data_time: 0.0313 time: 1.2764 +2023/06/04 07:40:51 - mmengine - INFO - Epoch(train) [7][ 100/3907] lr: 4.0754e-05 eta: 2:45:03 time: 0.6361 data_time: 0.0015 memory: 44138 loss: 0.3796 +2023/06/04 07:41:54 - mmengine - INFO - Epoch(train) [7][ 200/3907] lr: 4.0411e-05 eta: 2:43:59 time: 0.6369 data_time: 0.0016 memory: 44138 loss: 0.4093 +2023/06/04 07:42:58 - mmengine - INFO - Epoch(train) [7][ 300/3907] lr: 4.0069e-05 eta: 2:42:55 time: 0.6362 data_time: 0.0020 memory: 44138 loss: 0.3521 +2023/06/04 07:44:02 - mmengine - INFO - Epoch(train) [7][ 400/3907] lr: 3.9729e-05 eta: 2:41:51 time: 0.6374 data_time: 0.0014 memory: 44138 loss: 0.3372 +2023/06/04 07:45:05 - mmengine - INFO - Epoch(train) [7][ 500/3907] lr: 3.9389e-05 eta: 2:40:48 time: 0.6368 data_time: 0.0017 memory: 44138 loss: 0.3982 +2023/06/04 07:45:42 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230604_032014 +2023/06/04 07:46:09 - mmengine - INFO - Epoch(train) [7][ 600/3907] lr: 3.9050e-05 eta: 2:39:44 time: 0.6377 data_time: 0.0016 memory: 44138 loss: 0.3759 +2023/06/04 07:47:13 - mmengine - INFO - Epoch(train) [7][ 700/3907] lr: 3.8712e-05 eta: 2:38:40 time: 0.6377 data_time: 0.0020 memory: 44138 loss: 0.3708 +2023/06/04 07:48:17 - mmengine - INFO - Epoch(train) [7][ 800/3907] lr: 3.8375e-05 eta: 2:37:36 time: 0.6366 data_time: 0.0018 memory: 44138 loss: 0.3842 +2023/06/04 07:49:21 - mmengine - INFO - Epoch(train) [7][ 900/3907] lr: 3.8039e-05 eta: 2:36:32 time: 0.6378 data_time: 0.0024 memory: 44138 loss: 0.3891 +2023/06/04 07:50:24 - mmengine - INFO - Epoch(train) [7][1000/3907] lr: 3.7705e-05 eta: 2:35:29 time: 0.6449 data_time: 0.0019 memory: 44138 loss: 0.3842 +2023/06/04 07:51:28 - mmengine - INFO - Epoch(train) [7][1100/3907] lr: 3.7371e-05 eta: 2:34:25 time: 0.6376 data_time: 0.0016 memory: 44138 loss: 0.3763 +2023/06/04 07:52:32 - mmengine - INFO - Epoch(train) [7][1200/3907] lr: 3.7039e-05 eta: 2:33:21 time: 0.6364 data_time: 0.0015 memory: 44138 loss: 0.4037 +2023/06/04 07:53:36 - mmengine - INFO - Epoch(train) [7][1300/3907] lr: 3.6708e-05 eta: 2:32:17 time: 0.6364 data_time: 0.0017 memory: 44138 loss: 0.3758 +2023/06/04 07:54:39 - mmengine - INFO - Epoch(train) [7][1400/3907] lr: 3.6378e-05 eta: 2:31:14 time: 0.6372 data_time: 0.0016 memory: 44138 loss: 0.3691 +2023/06/04 07:55:43 - mmengine - INFO - Epoch(train) [7][1500/3907] lr: 3.6049e-05 eta: 2:30:10 time: 0.6372 data_time: 0.0017 memory: 44138 loss: 0.3494 +2023/06/04 07:56:20 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230604_032014 +2023/06/04 07:56:47 - mmengine - INFO - Epoch(train) [7][1600/3907] lr: 3.5721e-05 eta: 2:29:06 time: 0.6374 data_time: 0.0017 memory: 44138 loss: 0.3785 +2023/06/04 07:57:51 - mmengine - INFO - Epoch(train) [7][1700/3907] lr: 3.5395e-05 eta: 2:28:02 time: 0.6370 data_time: 0.0016 memory: 44138 loss: 0.3621 +2023/06/04 07:58:54 - mmengine - INFO - Epoch(train) [7][1800/3907] lr: 3.5070e-05 eta: 2:26:58 time: 0.6363 data_time: 0.0015 memory: 44138 loss: 0.3825 +2023/06/04 07:59:58 - mmengine - INFO - Epoch(train) [7][1900/3907] lr: 3.4746e-05 eta: 2:25:54 time: 0.6369 data_time: 0.0018 memory: 44138 loss: 0.3769 +2023/06/04 08:01:02 - mmengine - INFO - Epoch(train) [7][2000/3907] lr: 3.4424e-05 eta: 2:24:51 time: 0.6368 data_time: 0.0020 memory: 44138 loss: 0.3705 +2023/06/04 08:02:05 - mmengine - INFO - Epoch(train) [7][2100/3907] lr: 3.4103e-05 eta: 2:23:47 time: 0.6366 data_time: 0.0018 memory: 44138 loss: 0.3651 +2023/06/04 08:03:09 - mmengine - INFO - Epoch(train) [7][2200/3907] lr: 3.3783e-05 eta: 2:22:43 time: 0.6371 data_time: 0.0017 memory: 44138 loss: 0.3871 +2023/06/04 08:04:13 - mmengine - INFO - Epoch(train) [7][2300/3907] lr: 3.3465e-05 eta: 2:21:39 time: 0.6364 data_time: 0.0016 memory: 44138 loss: 0.3867 +2023/06/04 08:05:17 - mmengine - INFO - Epoch(train) [7][2400/3907] lr: 3.3148e-05 eta: 2:20:35 time: 0.6380 data_time: 0.0021 memory: 44138 loss: 0.3789 +2023/06/04 08:06:20 - mmengine - INFO - Epoch(train) [7][2500/3907] lr: 3.2832e-05 eta: 2:19:32 time: 0.6375 data_time: 0.0016 memory: 44138 loss: 0.4013 +2023/06/04 08:06:57 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230604_032014 +2023/06/04 08:07:24 - mmengine - INFO - Epoch(train) [7][2600/3907] lr: 3.2518e-05 eta: 2:18:28 time: 0.6376 data_time: 0.0016 memory: 44138 loss: 0.3687 +2023/06/04 08:08:28 - mmengine - INFO - Epoch(train) [7][2700/3907] lr: 3.2205e-05 eta: 2:17:24 time: 0.6376 data_time: 0.0015 memory: 44138 loss: 0.3866 +2023/06/04 08:09:32 - mmengine - INFO - Epoch(train) [7][2800/3907] lr: 3.1894e-05 eta: 2:16:20 time: 0.6370 data_time: 0.0020 memory: 44138 loss: 0.3730 +2023/06/04 08:10:36 - mmengine - INFO - Epoch(train) [7][2900/3907] lr: 3.1584e-05 eta: 2:15:17 time: 0.6371 data_time: 0.0018 memory: 44138 loss: 0.3776 +2023/06/04 08:11:39 - mmengine - INFO - Epoch(train) [7][3000/3907] lr: 3.1276e-05 eta: 2:14:13 time: 0.6356 data_time: 0.0017 memory: 44138 loss: 0.3637 +2023/06/04 08:12:43 - mmengine - INFO - Epoch(train) [7][3100/3907] lr: 3.0969e-05 eta: 2:13:09 time: 0.6353 data_time: 0.0015 memory: 44138 loss: 0.3562 +2023/06/04 08:13:46 - mmengine - INFO - Epoch(train) [7][3200/3907] lr: 3.0664e-05 eta: 2:12:05 time: 0.6352 data_time: 0.0017 memory: 44138 loss: 0.3679 +2023/06/04 08:14:50 - mmengine - INFO - Epoch(train) [7][3300/3907] lr: 3.0360e-05 eta: 2:11:01 time: 0.6361 data_time: 0.0023 memory: 44138 loss: 0.3840 +2023/06/04 08:15:54 - mmengine - INFO - Epoch(train) [7][3400/3907] lr: 3.0058e-05 eta: 2:09:57 time: 0.6374 data_time: 0.0016 memory: 44138 loss: 0.3893 +2023/06/04 08:16:58 - mmengine - INFO - Epoch(train) [7][3500/3907] lr: 2.9758e-05 eta: 2:08:54 time: 0.6381 data_time: 0.0019 memory: 44138 loss: 0.3752 +2023/06/04 08:17:35 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230604_032014 +2023/06/04 08:18:01 - mmengine - INFO - Epoch(train) [7][3600/3907] lr: 2.9459e-05 eta: 2:07:50 time: 0.6370 data_time: 0.0015 memory: 44138 loss: 0.3807 +2023/06/04 08:19:05 - mmengine - INFO - Epoch(train) [7][3700/3907] lr: 2.9162e-05 eta: 2:06:46 time: 0.6376 data_time: 0.0018 memory: 44138 loss: 0.3943 +2023/06/04 08:20:09 - mmengine - INFO - Epoch(train) [7][3800/3907] lr: 2.8867e-05 eta: 2:05:42 time: 0.6378 data_time: 0.0015 memory: 44138 loss: 0.3626 +2023/06/04 08:21:13 - mmengine - INFO - Epoch(train) [7][3900/3907] lr: 2.8573e-05 eta: 2:04:39 time: 0.6366 data_time: 0.0014 memory: 44138 loss: 0.3706 +2023/06/04 08:21:17 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230604_032014 +2023/06/04 08:21:17 - mmengine - INFO - Saving checkpoint at 7 epochs +2023/06/04 08:22:51 - mmengine - INFO - Epoch(val) [7][57/57] accuracy/top1: 77.2242 single-label/precision_classwise: [71.14134216308594, 97.01095581054688] single-label/recall_classwise: [98.7248306274414, 50.82170486450195] single-label/f1-score_classwise: [82.69352722167969, 66.7005844116211] data_time: 0.0312 time: 1.2760 +2023/06/04 08:23:58 - mmengine - INFO - Epoch(train) [8][ 100/3907] lr: 2.8261e-05 eta: 2:03:32 time: 0.6358 data_time: 0.0018 memory: 44138 loss: 0.3593 +2023/06/04 08:25:02 - mmengine - INFO - Epoch(train) [8][ 200/3907] lr: 2.7971e-05 eta: 2:02:28 time: 0.6351 data_time: 0.0021 memory: 44138 loss: 0.3875 +2023/06/04 08:26:05 - mmengine - INFO - Epoch(train) [8][ 300/3907] lr: 2.7682e-05 eta: 2:01:24 time: 0.6351 data_time: 0.0015 memory: 44138 loss: 0.3756 +2023/06/04 08:27:09 - mmengine - INFO - Epoch(train) [8][ 400/3907] lr: 2.7395e-05 eta: 2:00:20 time: 0.6351 data_time: 0.0019 memory: 44138 loss: 0.3728 +2023/06/04 08:28:12 - mmengine - INFO - Epoch(train) [8][ 500/3907] lr: 2.7111e-05 eta: 1:59:16 time: 0.6347 data_time: 0.0018 memory: 44138 loss: 0.3722 +2023/06/04 08:29:16 - mmengine - INFO - Epoch(train) [8][ 600/3907] lr: 2.6828e-05 eta: 1:58:12 time: 0.6376 data_time: 0.0016 memory: 44138 loss: 0.4010 +2023/06/04 08:29:49 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230604_032014 +2023/06/04 08:30:20 - mmengine - INFO - Epoch(train) [8][ 700/3907] lr: 2.6546e-05 eta: 1:57:08 time: 0.6377 data_time: 0.0026 memory: 44138 loss: 0.3739 +2023/06/04 08:31:23 - mmengine - INFO - Epoch(train) [8][ 800/3907] lr: 2.6267e-05 eta: 1:56:05 time: 0.6351 data_time: 0.0015 memory: 44138 loss: 0.3792 +2023/06/04 08:32:27 - mmengine - INFO - Epoch(train) [8][ 900/3907] lr: 2.5989e-05 eta: 1:55:01 time: 0.6379 data_time: 0.0019 memory: 44138 loss: 0.3811 +2023/06/04 08:33:31 - mmengine - INFO - Epoch(train) [8][1000/3907] lr: 2.5714e-05 eta: 1:53:57 time: 0.6375 data_time: 0.0016 memory: 44138 loss: 0.3933 +2023/06/04 08:34:34 - mmengine - INFO - Epoch(train) [8][1100/3907] lr: 2.5440e-05 eta: 1:52:53 time: 0.6357 data_time: 0.0015 memory: 44138 loss: 0.3568 +2023/06/04 08:35:38 - mmengine - INFO - Epoch(train) [8][1200/3907] lr: 2.5168e-05 eta: 1:51:49 time: 0.6345 data_time: 0.0018 memory: 44138 loss: 0.3460 +2023/06/04 08:36:42 - mmengine - INFO - Epoch(train) [8][1300/3907] lr: 2.4898e-05 eta: 1:50:46 time: 0.6379 data_time: 0.0016 memory: 44138 loss: 0.3555 +2023/06/04 08:37:45 - mmengine - INFO - Epoch(train) [8][1400/3907] lr: 2.4630e-05 eta: 1:49:42 time: 0.6361 data_time: 0.0020 memory: 44138 loss: 0.3855 +2023/06/04 08:38:49 - mmengine - INFO - Epoch(train) [8][1500/3907] lr: 2.4364e-05 eta: 1:48:38 time: 0.6361 data_time: 0.0015 memory: 44138 loss: 0.3764 +2023/06/04 08:39:53 - mmengine - INFO - Epoch(train) [8][1600/3907] lr: 2.4100e-05 eta: 1:47:34 time: 0.6357 data_time: 0.0017 memory: 44138 loss: 0.3731 +2023/06/04 08:40:25 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230604_032014 +2023/06/04 08:40:57 - mmengine - INFO - Epoch(train) [8][1700/3907] lr: 2.3838e-05 eta: 1:46:30 time: 0.6364 data_time: 0.0015 memory: 44138 loss: 0.3520 +2023/06/04 08:42:00 - mmengine - INFO - Epoch(train) [8][1800/3907] lr: 2.3578e-05 eta: 1:45:27 time: 0.6371 data_time: 0.0023 memory: 44138 loss: 0.3523 +2023/06/04 08:43:04 - mmengine - INFO - Epoch(train) [8][1900/3907] lr: 2.3320e-05 eta: 1:44:23 time: 0.6352 data_time: 0.0018 memory: 44138 loss: 0.3900 +2023/06/04 08:44:08 - mmengine - INFO - Epoch(train) [8][2000/3907] lr: 2.3064e-05 eta: 1:43:19 time: 0.6375 data_time: 0.0017 memory: 44138 loss: 0.3788 +2023/06/04 08:45:11 - mmengine - INFO - Epoch(train) [8][2100/3907] lr: 2.2810e-05 eta: 1:42:15 time: 0.6356 data_time: 0.0016 memory: 44138 loss: 0.3825 +2023/06/04 08:46:15 - mmengine - INFO - Epoch(train) [8][2200/3907] lr: 2.2558e-05 eta: 1:41:11 time: 0.6354 data_time: 0.0020 memory: 44138 loss: 0.3868 +2023/06/04 08:47:19 - mmengine - INFO - Epoch(train) [8][2300/3907] lr: 2.2309e-05 eta: 1:40:07 time: 0.6486 data_time: 0.0021 memory: 44138 loss: 0.3400 +2023/06/04 08:48:22 - mmengine - INFO - Epoch(train) [8][2400/3907] lr: 2.2061e-05 eta: 1:39:04 time: 0.6359 data_time: 0.0017 memory: 44138 loss: 0.3570 +2023/06/04 08:49:26 - mmengine - INFO - Epoch(train) [8][2500/3907] lr: 2.1816e-05 eta: 1:38:00 time: 0.6351 data_time: 0.0017 memory: 44138 loss: 0.3637 +2023/06/04 08:50:29 - mmengine - INFO - Epoch(train) [8][2600/3907] lr: 2.1572e-05 eta: 1:36:56 time: 0.6363 data_time: 0.0019 memory: 44138 loss: 0.3287 +2023/06/04 08:51:02 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230604_032014 +2023/06/04 08:51:33 - mmengine - INFO - Epoch(train) [8][2700/3907] lr: 2.1331e-05 eta: 1:35:52 time: 0.6356 data_time: 0.0020 memory: 44138 loss: 0.3922 +2023/06/04 08:52:37 - mmengine - INFO - Epoch(train) [8][2800/3907] lr: 2.1092e-05 eta: 1:34:48 time: 0.6363 data_time: 0.0021 memory: 44138 loss: 0.3737 +2023/06/04 08:53:40 - mmengine - INFO - Epoch(train) [8][2900/3907] lr: 2.0855e-05 eta: 1:33:45 time: 0.6366 data_time: 0.0017 memory: 44138 loss: 0.3949 +2023/06/04 08:54:44 - mmengine - INFO - Epoch(train) [8][3000/3907] lr: 2.0621e-05 eta: 1:32:41 time: 0.6352 data_time: 0.0017 memory: 44138 loss: 0.3693 +2023/06/04 08:55:47 - mmengine - INFO - Epoch(train) [8][3100/3907] lr: 2.0388e-05 eta: 1:31:37 time: 0.6349 data_time: 0.0020 memory: 44138 loss: 0.3421 +2023/06/04 08:56:51 - mmengine - INFO - Epoch(train) [8][3200/3907] lr: 2.0158e-05 eta: 1:30:33 time: 0.6350 data_time: 0.0021 memory: 44138 loss: 0.3605 +2023/06/04 08:57:54 - mmengine - INFO - Epoch(train) [8][3300/3907] lr: 1.9930e-05 eta: 1:29:29 time: 0.6378 data_time: 0.0022 memory: 44138 loss: 0.3508 +2023/06/04 08:58:58 - mmengine - INFO - Epoch(train) [8][3400/3907] lr: 1.9705e-05 eta: 1:28:25 time: 0.6348 data_time: 0.0019 memory: 44138 loss: 0.3833 +2023/06/04 09:00:02 - mmengine - INFO - Epoch(train) [8][3500/3907] lr: 1.9481e-05 eta: 1:27:22 time: 0.6366 data_time: 0.0017 memory: 44138 loss: 0.3771 +2023/06/04 09:01:05 - mmengine - INFO - Epoch(train) [8][3600/3907] lr: 1.9260e-05 eta: 1:26:18 time: 0.6374 data_time: 0.0019 memory: 44138 loss: 0.3743 +2023/06/04 09:01:38 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230604_032014 +2023/06/04 09:02:09 - mmengine - INFO - Epoch(train) [8][3700/3907] lr: 1.9042e-05 eta: 1:25:14 time: 0.6367 data_time: 0.0018 memory: 44138 loss: 0.3877 +2023/06/04 09:03:13 - mmengine - INFO - Epoch(train) [8][3800/3907] lr: 1.8825e-05 eta: 1:24:10 time: 0.6356 data_time: 0.0017 memory: 44138 loss: 0.3610 +2023/06/04 09:04:16 - mmengine - INFO - Epoch(train) [8][3900/3907] lr: 1.8611e-05 eta: 1:23:06 time: 0.6342 data_time: 0.0018 memory: 44138 loss: 0.3503 +2023/06/04 09:04:20 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230604_032014 +2023/06/04 09:04:20 - mmengine - INFO - Saving checkpoint at 8 epochs +2023/06/04 09:05:56 - mmengine - INFO - Epoch(val) [8][57/57] accuracy/top1: 77.1163 single-label/precision_classwise: [71.00489807128906, 97.26416778564453] single-label/recall_classwise: [98.84477233886719, 50.434104919433594] single-label/f1-score_classwise: [82.64323425292969, 66.42503356933594] data_time: 0.0320 time: 1.2753 +2023/06/04 09:07:03 - mmengine - INFO - Epoch(train) [9][ 100/3907] lr: 1.8385e-05 eta: 1:21:59 time: 0.6363 data_time: 0.0018 memory: 44138 loss: 0.3778 +2023/06/04 09:08:07 - mmengine - INFO - Epoch(train) [9][ 200/3907] lr: 1.8176e-05 eta: 1:20:55 time: 0.6455 data_time: 0.0021 memory: 44138 loss: 0.3605 +2023/06/04 09:09:10 - mmengine - INFO - Epoch(train) [9][ 300/3907] lr: 1.7969e-05 eta: 1:19:51 time: 0.6360 data_time: 0.0021 memory: 44138 loss: 0.3793 +2023/06/04 09:10:14 - mmengine - INFO - Epoch(train) [9][ 400/3907] lr: 1.7765e-05 eta: 1:18:47 time: 0.6366 data_time: 0.0018 memory: 44138 loss: 0.3728 +2023/06/04 09:11:18 - mmengine - INFO - Epoch(train) [9][ 500/3907] lr: 1.7563e-05 eta: 1:17:44 time: 0.6365 data_time: 0.0022 memory: 44138 loss: 0.3745 +2023/06/04 09:12:22 - mmengine - INFO - Epoch(train) [9][ 600/3907] lr: 1.7363e-05 eta: 1:16:40 time: 0.6359 data_time: 0.0024 memory: 44138 loss: 0.3474 +2023/06/04 09:13:25 - mmengine - INFO - Epoch(train) [9][ 700/3907] lr: 1.7166e-05 eta: 1:15:36 time: 0.6379 data_time: 0.0025 memory: 44138 loss: 0.3924 +2023/06/04 09:13:53 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230604_032014 +2023/06/04 09:14:29 - mmengine - INFO - Epoch(train) [9][ 800/3907] lr: 1.6971e-05 eta: 1:14:32 time: 0.6388 data_time: 0.0018 memory: 44138 loss: 0.3502 +2023/06/04 09:15:33 - mmengine - INFO - Epoch(train) [9][ 900/3907] lr: 1.6779e-05 eta: 1:13:29 time: 0.6374 data_time: 0.0017 memory: 44138 loss: 0.3612 +2023/06/04 09:16:37 - mmengine - INFO - Epoch(train) [9][1000/3907] lr: 1.6589e-05 eta: 1:12:25 time: 0.6507 data_time: 0.0021 memory: 44138 loss: 0.3506 +2023/06/04 09:17:41 - mmengine - INFO - Epoch(train) [9][1100/3907] lr: 1.6402e-05 eta: 1:11:21 time: 0.6359 data_time: 0.0019 memory: 44138 loss: 0.3661 +2023/06/04 09:18:44 - mmengine - INFO - Epoch(train) [9][1200/3907] lr: 1.6217e-05 eta: 1:10:17 time: 0.6373 data_time: 0.0021 memory: 44138 loss: 0.3614 +2023/06/04 09:19:48 - mmengine - INFO - Epoch(train) [9][1300/3907] lr: 1.6035e-05 eta: 1:09:14 time: 0.6380 data_time: 0.0018 memory: 44138 loss: 0.3695 +2023/06/04 09:20:52 - mmengine - INFO - Epoch(train) [9][1400/3907] lr: 1.5855e-05 eta: 1:08:10 time: 0.6360 data_time: 0.0019 memory: 44138 loss: 0.3652 +2023/06/04 09:21:56 - mmengine - INFO - Epoch(train) [9][1500/3907] lr: 1.5678e-05 eta: 1:07:06 time: 0.6369 data_time: 0.0022 memory: 44138 loss: 0.3885 +2023/06/04 09:23:00 - mmengine - INFO - Epoch(train) [9][1600/3907] lr: 1.5503e-05 eta: 1:06:02 time: 0.6393 data_time: 0.0020 memory: 44138 loss: 0.3810 +2023/06/04 09:24:04 - mmengine - INFO - Epoch(train) [9][1700/3907] lr: 1.5331e-05 eta: 1:04:59 time: 0.6375 data_time: 0.0020 memory: 44138 loss: 0.3712 +2023/06/04 09:24:32 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230604_032014 +2023/06/04 09:25:07 - mmengine - INFO - Epoch(train) [9][1800/3907] lr: 1.5162e-05 eta: 1:03:55 time: 0.6382 data_time: 0.0017 memory: 44138 loss: 0.3558 +2023/06/04 09:26:11 - mmengine - INFO - Epoch(train) [9][1900/3907] lr: 1.4995e-05 eta: 1:02:51 time: 0.6370 data_time: 0.0020 memory: 44138 loss: 0.3715 +2023/06/04 09:27:15 - mmengine - INFO - Epoch(train) [9][2000/3907] lr: 1.4830e-05 eta: 1:01:47 time: 0.6374 data_time: 0.0019 memory: 44138 loss: 0.3914 +2023/06/04 09:28:19 - mmengine - INFO - Epoch(train) [9][2100/3907] lr: 1.4668e-05 eta: 1:00:43 time: 0.6399 data_time: 0.0019 memory: 44138 loss: 0.3708 +2023/06/04 09:29:23 - mmengine - INFO - Epoch(train) [9][2200/3907] lr: 1.4509e-05 eta: 0:59:40 time: 0.6388 data_time: 0.0022 memory: 44138 loss: 0.3788 +2023/06/04 09:30:26 - mmengine - INFO - Epoch(train) [9][2300/3907] lr: 1.4353e-05 eta: 0:58:36 time: 0.6382 data_time: 0.0024 memory: 44138 loss: 0.3610 +2023/06/04 09:31:30 - mmengine - INFO - Epoch(train) [9][2400/3907] lr: 1.4199e-05 eta: 0:57:32 time: 0.6385 data_time: 0.0022 memory: 44138 loss: 0.3617 +2023/06/04 09:32:34 - mmengine - INFO - Epoch(train) [9][2500/3907] lr: 1.4047e-05 eta: 0:56:28 time: 0.6382 data_time: 0.0018 memory: 44138 loss: 0.3511 +2023/06/04 09:33:38 - mmengine - INFO - Epoch(train) [9][2600/3907] lr: 1.3899e-05 eta: 0:55:25 time: 0.6376 data_time: 0.0016 memory: 44138 loss: 0.3801 +2023/06/04 09:34:42 - mmengine - INFO - Epoch(train) [9][2700/3907] lr: 1.3753e-05 eta: 0:54:21 time: 0.6357 data_time: 0.0015 memory: 44138 loss: 0.3761 +2023/06/04 09:35:10 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230604_032014 +2023/06/04 09:35:45 - mmengine - INFO - Epoch(train) [9][2800/3907] lr: 1.3609e-05 eta: 0:53:17 time: 0.6384 data_time: 0.0025 memory: 44138 loss: 0.3872 +2023/06/04 09:36:49 - mmengine - INFO - Epoch(train) [9][2900/3907] lr: 1.3469e-05 eta: 0:52:13 time: 0.6384 data_time: 0.0022 memory: 44138 loss: 0.3785 +2023/06/04 09:37:53 - mmengine - INFO - Epoch(train) [9][3000/3907] lr: 1.3331e-05 eta: 0:51:10 time: 0.6372 data_time: 0.0026 memory: 44138 loss: 0.3989 +2023/06/04 09:38:57 - mmengine - INFO - Epoch(train) [9][3100/3907] lr: 1.3196e-05 eta: 0:50:06 time: 0.6372 data_time: 0.0019 memory: 44138 loss: 0.3729 +2023/06/04 09:40:01 - mmengine - INFO - Epoch(train) [9][3200/3907] lr: 1.3063e-05 eta: 0:49:02 time: 0.6358 data_time: 0.0015 memory: 44138 loss: 0.3519 +2023/06/04 09:41:04 - mmengine - INFO - Epoch(train) [9][3300/3907] lr: 1.2933e-05 eta: 0:47:58 time: 0.6372 data_time: 0.0021 memory: 44138 loss: 0.3785 +2023/06/04 09:42:08 - mmengine - INFO - Epoch(train) [9][3400/3907] lr: 1.2806e-05 eta: 0:46:54 time: 0.6371 data_time: 0.0019 memory: 44138 loss: 0.3695 +2023/06/04 09:43:12 - mmengine - INFO - Epoch(train) [9][3500/3907] lr: 1.2682e-05 eta: 0:45:51 time: 0.6463 data_time: 0.0019 memory: 44138 loss: 0.3665 +2023/06/04 09:44:16 - mmengine - INFO - Epoch(train) [9][3600/3907] lr: 1.2560e-05 eta: 0:44:47 time: 0.6378 data_time: 0.0019 memory: 44138 loss: 0.3604 +2023/06/04 09:45:20 - mmengine - INFO - Epoch(train) [9][3700/3907] lr: 1.2441e-05 eta: 0:43:43 time: 0.6407 data_time: 0.0019 memory: 44138 loss: 0.3926 +2023/06/04 09:45:48 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230604_032014 +2023/06/04 09:46:24 - mmengine - INFO - Epoch(train) [9][3800/3907] lr: 1.2325e-05 eta: 0:42:39 time: 0.6381 data_time: 0.0022 memory: 44138 loss: 0.4025 +2023/06/04 09:47:28 - mmengine - INFO - Epoch(train) [9][3900/3907] lr: 1.2211e-05 eta: 0:41:36 time: 0.6378 data_time: 0.0025 memory: 44138 loss: 0.3599 +2023/06/04 09:47:32 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230604_032014 +2023/06/04 09:47:32 - mmengine - INFO - Saving checkpoint at 9 epochs +2023/06/04 09:49:06 - mmengine - INFO - Epoch(val) [9][57/57] accuracy/top1: 76.7545 single-label/precision_classwise: [70.62877655029297, 97.55313873291016] single-label/recall_classwise: [98.98995971679688, 49.449615478515625] single-label/f1-score_classwise: [82.43828582763672, 65.6309585571289] data_time: 0.0316 time: 1.2749 +2023/06/04 09:50:13 - mmengine - INFO - Epoch(train) [10][ 100/3907] lr: 1.2093e-05 eta: 0:40:28 time: 0.6390 data_time: 0.0028 memory: 44138 loss: 0.3892 +2023/06/04 09:51:17 - mmengine - INFO - Epoch(train) [10][ 200/3907] lr: 1.1985e-05 eta: 0:39:24 time: 0.6366 data_time: 0.0026 memory: 44138 loss: 0.3626 +2023/06/04 09:52:21 - mmengine - INFO - Epoch(train) [10][ 300/3907] lr: 1.1881e-05 eta: 0:38:20 time: 0.6519 data_time: 0.0027 memory: 44138 loss: 0.3831 +2023/06/04 09:53:25 - mmengine - INFO - Epoch(train) [10][ 400/3907] lr: 1.1778e-05 eta: 0:37:16 time: 0.6400 data_time: 0.0029 memory: 44138 loss: 0.3530 +2023/06/04 09:54:29 - mmengine - INFO - Epoch(train) [10][ 500/3907] lr: 1.1679e-05 eta: 0:36:13 time: 0.6372 data_time: 0.0025 memory: 44138 loss: 0.3705 +2023/06/04 09:55:33 - mmengine - INFO - Epoch(train) [10][ 600/3907] lr: 1.1583e-05 eta: 0:35:09 time: 0.6491 data_time: 0.0025 memory: 44138 loss: 0.3651 +2023/06/04 09:56:36 - mmengine - INFO - Epoch(train) [10][ 700/3907] lr: 1.1489e-05 eta: 0:34:05 time: 0.6365 data_time: 0.0023 memory: 44138 loss: 0.3741 +2023/06/04 09:57:40 - mmengine - INFO - Epoch(train) [10][ 800/3907] lr: 1.1398e-05 eta: 0:33:01 time: 0.6382 data_time: 0.0026 memory: 44138 loss: 0.3625 +2023/06/04 09:58:04 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230604_032014 +2023/06/04 09:58:44 - mmengine - INFO - Epoch(train) [10][ 900/3907] lr: 1.1310e-05 eta: 0:31:57 time: 0.6461 data_time: 0.0024 memory: 44138 loss: 0.3426 +2023/06/04 09:59:48 - mmengine - INFO - Epoch(train) [10][1000/3907] lr: 1.1225e-05 eta: 0:30:54 time: 0.6371 data_time: 0.0031 memory: 44138 loss: 0.3682 +2023/06/04 10:00:52 - mmengine - INFO - Epoch(train) [10][1100/3907] lr: 1.1142e-05 eta: 0:29:50 time: 0.6410 data_time: 0.0031 memory: 44138 loss: 0.3780 +2023/06/04 10:01:56 - mmengine - INFO - Epoch(train) [10][1200/3907] lr: 1.1063e-05 eta: 0:28:46 time: 0.6378 data_time: 0.0032 memory: 44138 loss: 0.3740 +2023/06/04 10:03:00 - mmengine - INFO - Epoch(train) [10][1300/3907] lr: 1.0986e-05 eta: 0:27:42 time: 0.6363 data_time: 0.0019 memory: 44138 loss: 0.3587 +2023/06/04 10:04:04 - mmengine - INFO - Epoch(train) [10][1400/3907] lr: 1.0912e-05 eta: 0:26:39 time: 0.6368 data_time: 0.0023 memory: 44138 loss: 0.3492 +2023/06/04 10:05:07 - mmengine - INFO - Epoch(train) [10][1500/3907] lr: 1.0841e-05 eta: 0:25:35 time: 0.6375 data_time: 0.0018 memory: 44138 loss: 0.3662 +2023/06/04 10:06:11 - mmengine - INFO - Epoch(train) [10][1600/3907] lr: 1.0773e-05 eta: 0:24:31 time: 0.6377 data_time: 0.0025 memory: 44138 loss: 0.3560 +2023/06/04 10:07:15 - mmengine - INFO - Epoch(train) [10][1700/3907] lr: 1.0707e-05 eta: 0:23:27 time: 0.6393 data_time: 0.0031 memory: 44138 loss: 0.3666 +2023/06/04 10:08:19 - mmengine - INFO - Epoch(train) [10][1800/3907] lr: 1.0645e-05 eta: 0:22:23 time: 0.6371 data_time: 0.0028 memory: 44138 loss: 0.3827 +2023/06/04 10:08:42 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230604_032014 +2023/06/04 10:09:22 - mmengine - INFO - Epoch(train) [10][1900/3907] lr: 1.0585e-05 eta: 0:21:20 time: 0.6478 data_time: 0.0024 memory: 44138 loss: 0.3794 +2023/06/04 10:10:26 - mmengine - INFO - Epoch(train) [10][2000/3907] lr: 1.0529e-05 eta: 0:20:16 time: 0.6362 data_time: 0.0020 memory: 44138 loss: 0.3628 +2023/06/04 10:11:30 - mmengine - INFO - Epoch(train) [10][2100/3907] lr: 1.0475e-05 eta: 0:19:12 time: 0.6382 data_time: 0.0026 memory: 44138 loss: 0.3968 +2023/06/04 10:12:33 - mmengine - INFO - Epoch(train) [10][2200/3907] lr: 1.0424e-05 eta: 0:18:08 time: 0.6359 data_time: 0.0016 memory: 44138 loss: 0.3656 +2023/06/04 10:13:37 - mmengine - INFO - Epoch(train) [10][2300/3907] lr: 1.0376e-05 eta: 0:17:04 time: 0.6368 data_time: 0.0022 memory: 44138 loss: 0.3627 +2023/06/04 10:14:41 - mmengine - INFO - Epoch(train) [10][2400/3907] lr: 1.0330e-05 eta: 0:16:01 time: 0.6370 data_time: 0.0025 memory: 44138 loss: 0.3799 +2023/06/04 10:15:45 - mmengine - INFO - Epoch(train) [10][2500/3907] lr: 1.0288e-05 eta: 0:14:57 time: 0.6371 data_time: 0.0029 memory: 44138 loss: 0.3676 +2023/06/04 10:16:48 - mmengine - INFO - Epoch(train) [10][2600/3907] lr: 1.0249e-05 eta: 0:13:53 time: 0.6373 data_time: 0.0025 memory: 44138 loss: 0.3494 +2023/06/04 10:17:52 - mmengine - INFO - Epoch(train) [10][2700/3907] lr: 1.0212e-05 eta: 0:12:49 time: 0.6373 data_time: 0.0026 memory: 44138 loss: 0.3732 +2023/06/04 10:18:56 - mmengine - INFO - Epoch(train) [10][2800/3907] lr: 1.0178e-05 eta: 0:11:46 time: 0.6363 data_time: 0.0024 memory: 44138 loss: 0.3445 +2023/06/04 10:19:20 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230604_032014 +2023/06/04 10:20:00 - mmengine - INFO - Epoch(train) [10][2900/3907] lr: 1.0148e-05 eta: 0:10:42 time: 0.6370 data_time: 0.0020 memory: 44138 loss: 0.3546 +2023/06/04 10:21:03 - mmengine - INFO - Epoch(train) [10][3000/3907] lr: 1.0120e-05 eta: 0:09:38 time: 0.6364 data_time: 0.0017 memory: 44138 loss: 0.3736 +2023/06/04 10:22:07 - mmengine - INFO - Epoch(train) [10][3100/3907] lr: 1.0095e-05 eta: 0:08:34 time: 0.6355 data_time: 0.0020 memory: 44138 loss: 0.3643 +2023/06/04 10:23:11 - mmengine - INFO - Epoch(train) [10][3200/3907] lr: 1.0073e-05 eta: 0:07:30 time: 0.6373 data_time: 0.0020 memory: 44138 loss: 0.3745 +2023/06/04 10:24:15 - mmengine - INFO - Epoch(train) [10][3300/3907] lr: 1.0054e-05 eta: 0:06:27 time: 0.6387 data_time: 0.0020 memory: 44138 loss: 0.3922 +2023/06/04 10:25:18 - mmengine - INFO - Epoch(train) [10][3400/3907] lr: 1.0038e-05 eta: 0:05:23 time: 0.6365 data_time: 0.0024 memory: 44138 loss: 0.3649 +2023/06/04 10:26:22 - mmengine - INFO - Epoch(train) [10][3500/3907] lr: 1.0024e-05 eta: 0:04:19 time: 0.6374 data_time: 0.0019 memory: 44138 loss: 0.3524 +2023/06/04 10:27:26 - mmengine - INFO - Epoch(train) [10][3600/3907] lr: 1.0014e-05 eta: 0:03:15 time: 0.6370 data_time: 0.0019 memory: 44138 loss: 0.4032 +2023/06/04 10:28:30 - mmengine - INFO - Epoch(train) [10][3700/3907] lr: 1.0006e-05 eta: 0:02:12 time: 0.6363 data_time: 0.0026 memory: 44138 loss: 0.3864 +2023/06/04 10:29:34 - mmengine - INFO - Epoch(train) [10][3800/3907] lr: 1.0002e-05 eta: 0:01:08 time: 0.6381 data_time: 0.0027 memory: 44138 loss: 0.3724 +2023/06/04 10:29:57 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230604_032014 +2023/06/04 10:30:37 - mmengine - INFO - Epoch(train) [10][3900/3907] lr: 1.0000e-05 eta: 0:00:04 time: 0.6352 data_time: 0.0021 memory: 44138 loss: 0.3813 +2023/06/04 10:30:41 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr1e-4_20230604_032014 +2023/06/04 10:30:41 - mmengine - INFO - Saving checkpoint at 10 epochs +2023/06/04 10:32:15 - mmengine - INFO - Epoch(val) [10][57/57] accuracy/top1: 76.3578 single-label/precision_classwise: [70.28433227539062, 97.37699890136719] single-label/recall_classwise: [98.93315124511719, 48.635658264160156] single-label/f1-score_classwise: [82.18359375, 64.87100982666016] data_time: 0.0321 time: 1.2754 diff --git a/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm_wneck/20230604_032014/vis_data/20230604_032014.json b/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm_wneck/20230604_032014/vis_data/20230604_032014.json new file mode 100644 index 0000000000000000000000000000000000000000..69a77665db0c4734b7fad4999206e586e8e9bb77 --- /dev/null +++ b/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm_wneck/20230604_032014/vis_data/20230604_032014.json @@ -0,0 +1,400 @@ +{"lr": 9.999857418382631e-05, "data_time": 0.002526402473449707, "loss": 0.5351661205291748, "time": 0.6351698398590088, "epoch": 1, "memory": 44138, "step": 100} +{"lr": 9.999423907359506e-05, "data_time": 0.002406454086303711, "loss": 0.4754618167877197, "time": 0.6351021766662598, "epoch": 1, "memory": 44138, "step": 200} +{"lr": 9.998699480412073e-05, "data_time": 0.001868295669555664, "loss": 0.46636803150177003, "time": 0.6340153217315674, "epoch": 1, "memory": 44138, "step": 300} +{"lr": 9.997684184379085e-05, "data_time": 0.002654623985290527, "loss": 0.4659047245979309, "time": 0.6366667032241822, "epoch": 1, "memory": 44138, "step": 400} +{"lr": 9.996378084905784e-05, "data_time": 0.001985454559326172, "loss": 0.4894475668668747, "time": 0.6355506658554078, "epoch": 1, "memory": 44138, "step": 500} +{"lr": 9.994781266439667e-05, "data_time": 0.0017483234405517578, "loss": 0.4473745822906494, "time": 0.6359376192092896, "epoch": 1, "memory": 44138, "step": 600} +{"lr": 9.992893832225061e-05, "data_time": 0.0015566825866699218, "loss": 0.4864815235137939, "time": 0.6358940839767456, "epoch": 1, "memory": 44138, "step": 700} +{"lr": 9.990715904296405e-05, "data_time": 0.0022615671157836916, "loss": 0.4405377238988876, "time": 0.6357339859008789, "epoch": 1, "memory": 44138, "step": 800} +{"lr": 9.988247623470371e-05, "data_time": 0.0015571832656860352, "loss": 0.46169241070747374, "time": 0.6358372211456299, "epoch": 1, "memory": 44138, "step": 900} +{"lr": 9.985489149336749e-05, "data_time": 0.002009463310241699, "loss": 0.44867429733276365, "time": 0.6360357284545899, "epoch": 1, "memory": 44138, "step": 1000} +{"lr": 9.982440660248187e-05, "data_time": 0.0020551443099975585, "loss": 0.4159615635871887, "time": 0.6361851215362548, "epoch": 1, "memory": 44138, "step": 1100} +{"lr": 9.979102353308574e-05, "data_time": 0.001760244369506836, "loss": 0.4278793901205063, "time": 0.6355719804763794, "epoch": 1, "memory": 44138, "step": 1200} +{"lr": 9.975474444360358e-05, "data_time": 0.001583242416381836, "loss": 0.4354752540588379, "time": 0.6347107648849487, "epoch": 1, "memory": 44138, "step": 1300} +{"lr": 9.97155716797056e-05, "data_time": 0.0016881227493286133, "loss": 0.4457062125205994, "time": 0.6341993570327759, "epoch": 1, "memory": 44138, "step": 1400} +{"lr": 9.967350777415632e-05, "data_time": 0.0017208337783813476, "loss": 0.4000787138938904, "time": 0.6360926628112793, "epoch": 1, "memory": 44138, "step": 1500} +{"lr": 9.962855544665059e-05, "data_time": 0.0015161514282226562, "loss": 0.4464715600013733, "time": 0.6359011888504028, "epoch": 1, "memory": 44138, "step": 1600} +{"lr": 9.958071760363773e-05, "data_time": 0.0013989925384521485, "loss": 0.41033613979816436, "time": 0.6355968713760376, "epoch": 1, "memory": 44138, "step": 1700} +{"lr": 9.952999733813382e-05, "data_time": 0.0017291545867919923, "loss": 0.43542540073394775, "time": 0.6360478878021241, "epoch": 1, "memory": 44138, "step": 1800} +{"lr": 9.947639792952148e-05, "data_time": 0.0017313241958618164, "loss": 0.42447437047958375, "time": 0.6356987237930298, "epoch": 1, "memory": 44138, "step": 1900} +{"lr": 9.941992284333813e-05, "data_time": 0.002037668228149414, "loss": 0.4312774628400803, "time": 0.6343778133392334, "epoch": 1, "memory": 44138, "step": 2000} +{"lr": 9.93605757310516e-05, "data_time": 0.0017091035842895508, "loss": 0.47163391709327696, "time": 0.6356060743331909, "epoch": 1, "memory": 44138, "step": 2100} +{"lr": 9.929836042982421e-05, "data_time": 0.002074432373046875, "loss": 0.4073728650808334, "time": 0.6354227781295776, "epoch": 1, "memory": 44138, "step": 2200} +{"lr": 9.923328096226467e-05, "data_time": 0.0017566919326782227, "loss": 0.3898966759443283, "time": 0.6362570762634278, "epoch": 1, "memory": 44138, "step": 2300} +{"lr": 9.916534153616819e-05, "data_time": 0.0023668289184570314, "loss": 0.39612277746200564, "time": 0.6370923280715942, "epoch": 1, "memory": 44138, "step": 2400} +{"lr": 9.909454654424387e-05, "data_time": 0.002035212516784668, "loss": 0.37696029245853424, "time": 0.6472670793533325, "epoch": 1, "memory": 44138, "step": 2500} +{"lr": 9.902090056383109e-05, "data_time": 0.0020287275314331056, "loss": 0.41623030602931976, "time": 0.6379285335540772, "epoch": 1, "memory": 44138, "step": 2600} +{"lr": 9.894440835660365e-05, "data_time": 0.0016750574111938476, "loss": 0.4208056092262268, "time": 0.6368530750274658, "epoch": 1, "memory": 44138, "step": 2700} +{"lr": 9.886507486826162e-05, "data_time": 0.0017534017562866211, "loss": 0.4455771863460541, "time": 0.6362451314926147, "epoch": 1, "memory": 44138, "step": 2800} +{"lr": 9.878290522821154e-05, "data_time": 0.0016971349716186524, "loss": 0.4209083318710327, "time": 0.6355997323989868, "epoch": 1, "memory": 44138, "step": 2900} +{"lr": 9.86979047492352e-05, "data_time": 0.00159454345703125, "loss": 0.41540210843086245, "time": 0.6349119901657104, "epoch": 1, "memory": 44138, "step": 3000} +{"lr": 9.861007892714578e-05, "data_time": 0.0026523828506469726, "loss": 0.4219654887914658, "time": 0.6351418972015381, "epoch": 1, "memory": 44138, "step": 3100} +{"lr": 9.851943344043239e-05, "data_time": 0.002142000198364258, "loss": 0.40397669672966, "time": 0.6354796171188355, "epoch": 1, "memory": 44138, "step": 3200} +{"lr": 9.842597414989319e-05, "data_time": 0.0019694805145263673, "loss": 0.3926267474889755, "time": 0.6436527490615844, "epoch": 1, "memory": 44138, "step": 3300} +{"lr": 9.832970709825637e-05, "data_time": 0.0015901088714599609, "loss": 0.38551358580589296, "time": 0.6360019445419312, "epoch": 1, "memory": 44138, "step": 3400} +{"lr": 9.823063850978968e-05, "data_time": 0.0016624927520751953, "loss": 0.40690678358078003, "time": 0.6358567953109742, "epoch": 1, "memory": 44138, "step": 3500} +{"lr": 9.81287747898975e-05, "data_time": 0.001653742790222168, "loss": 0.39099705815315244, "time": 0.6357744693756103, "epoch": 1, "memory": 44138, "step": 3600} +{"lr": 9.802412252470715e-05, "data_time": 0.0016423702239990235, "loss": 0.41955919861793517, "time": 0.6446315288543701, "epoch": 1, "memory": 44138, "step": 3700} +{"lr": 9.791668848064258e-05, "data_time": 0.0016124248504638672, "loss": 0.3832931399345398, "time": 0.6369137763977051, "epoch": 1, "memory": 44138, "step": 3800} +{"lr": 9.780647960398751e-05, "data_time": 0.0012752056121826173, "loss": 0.3861609011888504, "time": 0.6326295614242554, "epoch": 1, "memory": 44138, "step": 3900} +{"accuracy/top1": 82.55802154541016, "data_time": 0.04241977658188134, "time": 1.3461164340638279, "step": 1} +{"lr": 9.7685491196646e-05, "data_time": 0.0019457101821899413, "loss": 0.410807603597641, "time": 0.6350873231887817, "epoch": 2, "memory": 44138, "step": 4007} +{"lr": 9.756956126085676e-05, "data_time": 0.0020830631256103516, "loss": 0.3643091797828674, "time": 0.6369590520858764, "epoch": 2, "memory": 44138, "step": 4107} +{"lr": 9.745087893642482e-05, "data_time": 0.0017199039459228516, "loss": 0.4226975947618484, "time": 0.6364662170410156, "epoch": 2, "memory": 44138, "step": 4207} +{"lr": 9.73294518969056e-05, "data_time": 0.001866459846496582, "loss": 0.3846674203872681, "time": 0.6369388818740844, "epoch": 2, "memory": 44138, "step": 4307} +{"lr": 9.72052879933172e-05, "data_time": 0.0020048141479492186, "loss": 0.39270468056201935, "time": 0.6366772651672363, "epoch": 2, "memory": 44138, "step": 4407} +{"lr": 9.707839525363358e-05, "data_time": 0.0017948627471923828, "loss": 0.4163460284471512, "time": 0.636162543296814, "epoch": 2, "memory": 44138, "step": 4507} +{"lr": 9.69487818822646e-05, "data_time": 0.0018607378005981445, "loss": 0.38223142027854917, "time": 0.636462950706482, "epoch": 2, "memory": 44138, "step": 4607} +{"lr": 9.681645625952616e-05, "data_time": 0.0014985322952270508, "loss": 0.4024761557579041, "time": 0.6363603353500367, "epoch": 2, "memory": 44138, "step": 4707} +{"lr": 9.668142694109826e-05, "data_time": 0.0013245582580566407, "loss": 0.3801256835460663, "time": 0.6361339092254639, "epoch": 2, "memory": 44138, "step": 4807} +{"lr": 9.654370265747162e-05, "data_time": 0.0013548612594604492, "loss": 0.384027886390686, "time": 0.6361263751983642, "epoch": 2, "memory": 44138, "step": 4907} +{"lr": 9.640329231338345e-05, "data_time": 0.0018148422241210938, "loss": 0.39843118488788604, "time": 0.6359064817428589, "epoch": 2, "memory": 44138, "step": 5007} +{"lr": 9.626020498724147e-05, "data_time": 0.0017451047897338867, "loss": 0.381126669049263, "time": 0.6361303567886353, "epoch": 2, "memory": 44138, "step": 5107} +{"lr": 9.611444993053717e-05, "data_time": 0.0014071464538574219, "loss": 0.3815277427434921, "time": 0.6362509727478027, "epoch": 2, "memory": 44138, "step": 5207} +{"lr": 9.596603656724746e-05, "data_time": 0.0015529394149780273, "loss": 0.3765850216150284, "time": 0.636037039756775, "epoch": 2, "memory": 44138, "step": 5307} +{"lr": 9.58149744932254e-05, "data_time": 0.0018204450607299805, "loss": 0.3873521864414215, "time": 0.6353329658508301, "epoch": 2, "memory": 44138, "step": 5407} +{"lr": 9.566127347557972e-05, "data_time": 0.0022878408432006835, "loss": 0.4038151651620865, "time": 0.6346449375152587, "epoch": 2, "memory": 44138, "step": 5507} +{"lr": 9.550494345204367e-05, "data_time": 0.001725625991821289, "loss": 0.3884665071964264, "time": 0.6349648714065552, "epoch": 2, "memory": 44138, "step": 5607} +{"lr": 9.534599453033183e-05, "data_time": 0.0020513534545898438, "loss": 0.40863217115402223, "time": 0.6366077184677124, "epoch": 2, "memory": 44138, "step": 5707} +{"lr": 9.518443698748712e-05, "data_time": 0.001436305046081543, "loss": 0.37959822118282316, "time": 0.6453171491622924, "epoch": 2, "memory": 44138, "step": 5807} +{"lr": 9.50202812692162e-05, "data_time": 0.0013851642608642579, "loss": 0.3962046951055527, "time": 0.6358612060546875, "epoch": 2, "memory": 44138, "step": 5907} +{"lr": 9.485353798921386e-05, "data_time": 0.0025690078735351564, "loss": 0.3757527589797974, "time": 0.6363880395889282, "epoch": 2, "memory": 44138, "step": 6007} +{"lr": 9.468421792847712e-05, "data_time": 0.0020028829574584963, "loss": 0.40307600796222687, "time": 0.6365566492080689, "epoch": 2, "memory": 44138, "step": 6107} +{"lr": 9.451233203460794e-05, "data_time": 0.001594209671020508, "loss": 0.39338279366493223, "time": 0.6365896940231324, "epoch": 2, "memory": 44138, "step": 6207} +{"lr": 9.433789142110532e-05, "data_time": 0.0016883611679077148, "loss": 0.3865435302257538, "time": 0.6363863706588745, "epoch": 2, "memory": 44138, "step": 6307} +{"lr": 9.416090736664709e-05, "data_time": 0.0013787508010864257, "loss": 0.4145511955022812, "time": 0.6368163585662842, "epoch": 2, "memory": 44138, "step": 6407} +{"lr": 9.398139131436032e-05, "data_time": 0.0017983198165893554, "loss": 0.39429461061954496, "time": 0.6360846519470215, "epoch": 2, "memory": 44138, "step": 6507} +{"lr": 9.379935487108167e-05, "data_time": 0.00151369571685791, "loss": 0.39995901584625243, "time": 0.6380569219589234, "epoch": 2, "memory": 44138, "step": 6607} +{"lr": 9.361480980660697e-05, "data_time": 0.0015455961227416992, "loss": 0.39948015511035917, "time": 0.6372979164123536, "epoch": 2, "memory": 44138, "step": 6707} +{"lr": 9.342776805292993e-05, "data_time": 0.001640486717224121, "loss": 0.36709106862545016, "time": 0.6505163431167602, "epoch": 2, "memory": 44138, "step": 6807} +{"lr": 9.323824170347083e-05, "data_time": 0.0016528844833374023, "loss": 0.38471029996871947, "time": 0.6369179964065552, "epoch": 2, "memory": 44138, "step": 6907} +{"lr": 9.304624301229471e-05, "data_time": 0.0020319461822509766, "loss": 0.39614465534687043, "time": 0.6368730783462524, "epoch": 2, "memory": 44138, "step": 7007} +{"lr": 9.285178439331897e-05, "data_time": 0.0014183759689331056, "loss": 0.35661196112632754, "time": 0.6375965356826783, "epoch": 2, "memory": 44138, "step": 7107} +{"lr": 9.265487841951087e-05, "data_time": 0.0016214609146118163, "loss": 0.37881762683391573, "time": 0.6370727777481079, "epoch": 2, "memory": 44138, "step": 7207} +{"lr": 9.245553782207421e-05, "data_time": 0.0016767740249633788, "loss": 0.3776803106069565, "time": 0.6357251405715942, "epoch": 2, "memory": 44138, "step": 7307} +{"lr": 9.225377548962671e-05, "data_time": 0.0016555070877075195, "loss": 0.37052683234214784, "time": 0.6352493762969971, "epoch": 2, "memory": 44138, "step": 7407} +{"lr": 9.204960446736601e-05, "data_time": 0.0019618749618530275, "loss": 0.40267767906188967, "time": 0.6363161087036133, "epoch": 2, "memory": 44138, "step": 7507} +{"lr": 9.184303795622715e-05, "data_time": 0.001558542251586914, "loss": 0.366828715801239, "time": 0.6356423377990723, "epoch": 2, "memory": 44138, "step": 7607} +{"lr": 9.163408931202823e-05, "data_time": 0.001344776153564453, "loss": 0.38677774369716644, "time": 0.6362397909164429, "epoch": 2, "memory": 44138, "step": 7707} +{"lr": 9.142277204460729e-05, "data_time": 0.0015615463256835938, "loss": 0.40312814712524414, "time": 0.6365349292755127, "epoch": 2, "memory": 44138, "step": 7807} +{"accuracy/top1": 82.29706573486328, "data_time": 0.0335041777840976, "time": 1.2768701479352753, "step": 2} +{"lr": 9.119405492383288e-05, "data_time": 0.0019070625305175782, "loss": 0.3771364837884903, "time": 0.6362051010131836, "epoch": 3, "memory": 44138, "step": 7914} +{"lr": 9.097787819233205e-05, "data_time": 0.0018374204635620117, "loss": 0.4004779279232025, "time": 0.6369052171707154, "epoch": 3, "memory": 44138, "step": 8014} +{"lr": 9.075937526576627e-05, "data_time": 0.0018862247467041015, "loss": 0.3887979418039322, "time": 0.6360315799713134, "epoch": 3, "memory": 44138, "step": 8114} +{"lr": 9.053856027171755e-05, "data_time": 0.0015074491500854492, "loss": 0.35386045277118683, "time": 0.6370097875595093, "epoch": 3, "memory": 44138, "step": 8214} +{"lr": 9.031544748725785e-05, "data_time": 0.0018077135086059571, "loss": 0.405665984749794, "time": 0.6369300842285156, "epoch": 3, "memory": 44138, "step": 8314} +{"lr": 9.009005133802542e-05, "data_time": 0.0015701055526733398, "loss": 0.3652287185192108, "time": 0.6358760118484497, "epoch": 3, "memory": 44138, "step": 8414} +{"lr": 8.986238639729242e-05, "data_time": 0.0016776561737060548, "loss": 0.38257692754268646, "time": 0.6368263959884644, "epoch": 3, "memory": 44138, "step": 8514} +{"lr": 8.963246738502254e-05, "data_time": 0.0014507532119750976, "loss": 0.3653694659471512, "time": 0.6368743658065796, "epoch": 3, "memory": 44138, "step": 8614} +{"lr": 8.940030916691928e-05, "data_time": 0.0016489982604980468, "loss": 0.37748109400272367, "time": 0.6365738630294799, "epoch": 3, "memory": 44138, "step": 8714} +{"lr": 8.91659267534649e-05, "data_time": 0.001427769660949707, "loss": 0.36973799765110016, "time": 0.6358951330184937, "epoch": 3, "memory": 44138, "step": 8814} +{"lr": 8.892933529894978e-05, "data_time": 0.0019328594207763672, "loss": 0.3578315645456314, "time": 0.6362555027008057, "epoch": 3, "memory": 44138, "step": 8914} +{"lr": 8.869055010049283e-05, "data_time": 0.0017607450485229493, "loss": 0.37117459774017336, "time": 0.6339348793029785, "epoch": 3, "memory": 44138, "step": 9014} +{"lr": 8.844958659705188e-05, "data_time": 0.001490616798400879, "loss": 0.3593450516462326, "time": 0.633604621887207, "epoch": 3, "memory": 44138, "step": 9114} +{"lr": 8.82064603684259e-05, "data_time": 0.002527451515197754, "loss": 0.35577426850795746, "time": 0.6364502429962158, "epoch": 3, "memory": 44138, "step": 9214} +{"lr": 8.796118713424776e-05, "data_time": 0.0017650842666625977, "loss": 0.3818786978721619, "time": 0.6364527463912963, "epoch": 3, "memory": 44138, "step": 9314} +{"lr": 8.77137827529676e-05, "data_time": 0.002043271064758301, "loss": 0.4080860525369644, "time": 0.6367826223373413, "epoch": 3, "memory": 44138, "step": 9414} +{"lr": 8.746426322082745e-05, "data_time": 0.0014680862426757813, "loss": 0.35324149429798124, "time": 0.6367222785949707, "epoch": 3, "memory": 44138, "step": 9514} +{"lr": 8.721264467082732e-05, "data_time": 0.0017992019653320312, "loss": 0.3661399602890015, "time": 0.6359201908111572, "epoch": 3, "memory": 44138, "step": 9614} +{"lr": 8.695894337168168e-05, "data_time": 0.0014480113983154296, "loss": 0.35871289670467377, "time": 0.6363116264343261, "epoch": 3, "memory": 44138, "step": 9714} +{"lr": 8.67031757267677e-05, "data_time": 0.001970672607421875, "loss": 0.3939231425523758, "time": 0.6369412183761597, "epoch": 3, "memory": 44138, "step": 9814} +{"lr": 8.64453582730651e-05, "data_time": 0.0014687061309814453, "loss": 0.3631616085767746, "time": 0.6370649099349975, "epoch": 3, "memory": 44138, "step": 9914} +{"lr": 8.61855076800861e-05, "data_time": 0.001672077178955078, "loss": 0.3913224399089813, "time": 0.636255145072937, "epoch": 3, "memory": 44138, "step": 10014} +{"lr": 8.592364074879843e-05, "data_time": 0.001474308967590332, "loss": 0.379348886013031, "time": 0.6359834432601928, "epoch": 3, "memory": 44138, "step": 10114} +{"lr": 8.565977441053854e-05, "data_time": 0.001491856575012207, "loss": 0.3770377814769745, "time": 0.6376715183258057, "epoch": 3, "memory": 44138, "step": 10214} +{"lr": 8.539392572591724e-05, "data_time": 0.0015501976013183594, "loss": 0.37069645822048186, "time": 0.6368112325668335, "epoch": 3, "memory": 44138, "step": 10314} +{"lr": 8.512611188371633e-05, "data_time": 0.0017702817916870118, "loss": 0.37677340507507323, "time": 0.6366872787475586, "epoch": 3, "memory": 44138, "step": 10414} +{"lr": 8.485635019977725e-05, "data_time": 0.0015655279159545899, "loss": 0.33343938291072844, "time": 0.6366911172866822, "epoch": 3, "memory": 44138, "step": 10514} +{"lr": 8.45846581158818e-05, "data_time": 0.00154876708984375, "loss": 0.37909236550331116, "time": 0.6365845680236817, "epoch": 3, "memory": 44138, "step": 10614} +{"lr": 8.431105319862416e-05, "data_time": 0.0016137123107910155, "loss": 0.3890629827976227, "time": 0.6348929643630982, "epoch": 3, "memory": 44138, "step": 10714} +{"lr": 8.403555313827516e-05, "data_time": 0.0019269943237304687, "loss": 0.370540389418602, "time": 0.6409435033798218, "epoch": 3, "memory": 44138, "step": 10814} +{"lr": 8.375817574763844e-05, "data_time": 0.002280569076538086, "loss": 0.3622734367847443, "time": 0.63719801902771, "epoch": 3, "memory": 44138, "step": 10914} +{"lr": 8.347893896089871e-05, "data_time": 0.0016435146331787109, "loss": 0.3717945545911789, "time": 0.6367708683013916, "epoch": 3, "memory": 44138, "step": 11014} +{"lr": 8.31978608324625e-05, "data_time": 0.0013298988342285156, "loss": 0.39679092466831206, "time": 0.6368518590927124, "epoch": 3, "memory": 44138, "step": 11114} +{"lr": 8.291495953579052e-05, "data_time": 0.0019569873809814455, "loss": 0.3572738438844681, "time": 0.6373294115066528, "epoch": 3, "memory": 44138, "step": 11214} +{"lr": 8.263025336222267e-05, "data_time": 0.0018393278121948242, "loss": 0.3617665946483612, "time": 0.6371901750564575, "epoch": 3, "memory": 44138, "step": 11314} +{"lr": 8.234376071979551e-05, "data_time": 0.0015526056289672852, "loss": 0.384552127122879, "time": 0.635496187210083, "epoch": 3, "memory": 44138, "step": 11414} +{"lr": 8.205550013205187e-05, "data_time": 0.0018289804458618164, "loss": 0.3666399359703064, "time": 0.6368457794189453, "epoch": 3, "memory": 44138, "step": 11514} +{"lr": 8.17654902368435e-05, "data_time": 0.0017534017562866211, "loss": 0.3668083220720291, "time": 0.6360332727432251, "epoch": 3, "memory": 44138, "step": 11614} +{"lr": 8.147374978512582e-05, "data_time": 0.001693129539489746, "loss": 0.38223120868206023, "time": 0.6358201026916503, "epoch": 3, "memory": 44138, "step": 11714} +{"accuracy/top1": 80.41474151611328, "data_time": 0.031422812363197064, "time": 1.2761071797074943, "step": 3} +{"lr": 8.115969237626578e-05, "data_time": 0.0022393226623535155, "loss": 0.4031961888074875, "time": 0.6378740072250366, "epoch": 4, "memory": 44138, "step": 11821} +{"lr": 8.08644297337245e-05, "data_time": 0.001559877395629883, "loss": 0.3554630070924759, "time": 0.637148666381836, "epoch": 4, "memory": 44138, "step": 11921} +{"lr": 8.056749479387651e-05, "data_time": 0.0014817476272583007, "loss": 0.3698801577091217, "time": 0.6369105100631713, "epoch": 4, "memory": 44138, "step": 12021} +{"lr": 8.026890675542457e-05, "data_time": 0.0017343282699584961, "loss": 0.3652118921279907, "time": 0.6353127241134644, "epoch": 4, "memory": 44138, "step": 12121} +{"lr": 7.996868492395457e-05, "data_time": 0.0019557952880859377, "loss": 0.3849740356206894, "time": 0.6371478319168091, "epoch": 4, "memory": 44138, "step": 12221} +{"lr": 7.966684871068762e-05, "data_time": 0.002064371109008789, "loss": 0.38858999609947203, "time": 0.6365043163299561, "epoch": 4, "memory": 44138, "step": 12321} +{"lr": 7.936341763122449e-05, "data_time": 0.0015283584594726562, "loss": 0.36215423941612246, "time": 0.6373162031173706, "epoch": 4, "memory": 44138, "step": 12421} +{"lr": 7.905841130428405e-05, "data_time": 0.002115774154663086, "loss": 0.36562091410160064, "time": 0.6364990234375, "epoch": 4, "memory": 44138, "step": 12521} +{"lr": 7.875184945043487e-05, "data_time": 0.0022097349166870115, "loss": 0.3673384487628937, "time": 0.6367980718612671, "epoch": 4, "memory": 44138, "step": 12621} +{"lr": 7.844375189082012e-05, "data_time": 0.0020020008087158203, "loss": 0.3457824647426605, "time": 0.6370929718017578, "epoch": 4, "memory": 44138, "step": 12721} +{"lr": 7.813413854587572e-05, "data_time": 0.001546025276184082, "loss": 0.37467997074127196, "time": 0.6368714332580566, "epoch": 4, "memory": 44138, "step": 12821} +{"lr": 7.782302943404291e-05, "data_time": 0.0023020505905151367, "loss": 0.3861950010061264, "time": 0.6367208242416382, "epoch": 4, "memory": 44138, "step": 12921} +{"lr": 7.751044467047337e-05, "data_time": 0.0013361215591430665, "loss": 0.35624179244041443, "time": 0.636503028869629, "epoch": 4, "memory": 44138, "step": 13021} +{"lr": 7.719640446572907e-05, "data_time": 0.0017018795013427734, "loss": 0.3679825782775879, "time": 0.6353821277618408, "epoch": 4, "memory": 44138, "step": 13121} +{"lr": 7.688092912447536e-05, "data_time": 0.0016974449157714845, "loss": 0.3739969491958618, "time": 0.6364741563796997, "epoch": 4, "memory": 44138, "step": 13221} +{"lr": 7.656403904416794e-05, "data_time": 0.0017170906066894531, "loss": 0.3584659039974213, "time": 0.6369251251220703, "epoch": 4, "memory": 44138, "step": 13321} +{"lr": 7.624575471373434e-05, "data_time": 0.0019341468811035155, "loss": 0.35059358179569244, "time": 0.6363126516342164, "epoch": 4, "memory": 44138, "step": 13421} +{"lr": 7.592609671224926e-05, "data_time": 0.0014975547790527343, "loss": 0.35459600687026976, "time": 0.6369526147842407, "epoch": 4, "memory": 44138, "step": 13521} +{"lr": 7.560508570760375e-05, "data_time": 0.0014723777770996095, "loss": 0.37329457998275756, "time": 0.6371474027633667, "epoch": 4, "memory": 44138, "step": 13621} +{"lr": 7.528274245516895e-05, "data_time": 0.0015514373779296875, "loss": 0.3840697079896927, "time": 0.637981629371643, "epoch": 4, "memory": 44138, "step": 13721} +{"lr": 7.495908779645405e-05, "data_time": 0.0018193721771240234, "loss": 0.35149831771850587, "time": 0.6373069286346436, "epoch": 4, "memory": 44138, "step": 13821} +{"lr": 7.463414265775906e-05, "data_time": 0.0016061067581176758, "loss": 0.3762999713420868, "time": 0.6366349697113037, "epoch": 4, "memory": 44138, "step": 13921} +{"lr": 7.430792804882147e-05, "data_time": 0.002154850959777832, "loss": 0.37663392424583436, "time": 0.6373780965805054, "epoch": 4, "memory": 44138, "step": 14021} +{"lr": 7.398046506145803e-05, "data_time": 0.0017863035202026368, "loss": 0.40180477499961853, "time": 0.6369316577911377, "epoch": 4, "memory": 44138, "step": 14121} +{"lr": 7.365177486820114e-05, "data_time": 0.0015575885772705078, "loss": 0.39575632512569425, "time": 0.6372013330459595, "epoch": 4, "memory": 44138, "step": 14221} +{"lr": 7.332187872092939e-05, "data_time": 0.0016432523727416993, "loss": 0.35980776846408846, "time": 0.6360937118530273, "epoch": 4, "memory": 44138, "step": 14321} +{"lr": 7.299079794949401e-05, "data_time": 0.0016222715377807618, "loss": 0.3617538034915924, "time": 0.6359572410583496, "epoch": 4, "memory": 44138, "step": 14421} +{"lr": 7.265855396033955e-05, "data_time": 0.0017271757125854493, "loss": 0.3715987801551819, "time": 0.6366678237915039, "epoch": 4, "memory": 44138, "step": 14521} +{"lr": 7.232516823512013e-05, "data_time": 0.0015793561935424805, "loss": 0.35890172123909, "time": 0.636816430091858, "epoch": 4, "memory": 44138, "step": 14621} +{"lr": 7.199066232930995e-05, "data_time": 0.0012960195541381835, "loss": 0.35270917415618896, "time": 0.6368032455444336, "epoch": 4, "memory": 44138, "step": 14721} +{"lr": 7.165505787080997e-05, "data_time": 0.002298140525817871, "loss": 0.3418466538190842, "time": 0.6367158412933349, "epoch": 4, "memory": 44138, "step": 14821} +{"lr": 7.13183765585495e-05, "data_time": 0.0015070915222167968, "loss": 0.3678338259458542, "time": 0.6368214845657348, "epoch": 4, "memory": 44138, "step": 14921} +{"lr": 7.098064016108317e-05, "data_time": 0.0017307519912719727, "loss": 0.37935556173324586, "time": 0.6376452684402466, "epoch": 4, "memory": 44138, "step": 15021} +{"lr": 7.064187051518379e-05, "data_time": 0.002585959434509277, "loss": 0.36913132220506667, "time": 0.6390216827392579, "epoch": 4, "memory": 44138, "step": 15121} +{"lr": 7.030208952442949e-05, "data_time": 0.0018384456634521484, "loss": 0.3649781197309494, "time": 0.636927318572998, "epoch": 4, "memory": 44138, "step": 15221} +{"lr": 6.996131915778851e-05, "data_time": 0.0016496658325195312, "loss": 0.3643661648035049, "time": 0.6371755838394165, "epoch": 4, "memory": 44138, "step": 15321} +{"lr": 6.961958144819853e-05, "data_time": 0.0016845464706420898, "loss": 0.3943395346403122, "time": 0.6380192518234253, "epoch": 4, "memory": 44138, "step": 15421} +{"lr": 6.927689849114193e-05, "data_time": 0.0017381429672241211, "loss": 0.37081224024295806, "time": 0.6376275777816772, "epoch": 4, "memory": 44138, "step": 15521} +{"lr": 6.893329244321722e-05, "data_time": 0.0013135433197021484, "loss": 0.3691263049840927, "time": 0.6336902618408203, "epoch": 4, "memory": 44138, "step": 15621} +{"accuracy/top1": 78.90470123291016, "data_time": 0.0329343450480494, "time": 1.2753868966267026, "step": 4} +{"lr": 6.856463687361792e-05, "data_time": 0.001588153839111328, "loss": 0.35422569513320923, "time": 0.6365503549575806, "epoch": 5, "memory": 44138, "step": 15728} +{"lr": 6.821919068470432e-05, "data_time": 0.0015651941299438476, "loss": 0.37694051861763, "time": 0.6365997314453125, "epoch": 5, "memory": 44138, "step": 15828} +{"lr": 6.787288979235407e-05, "data_time": 0.001617741584777832, "loss": 0.3566369891166687, "time": 0.636507534980774, "epoch": 5, "memory": 44138, "step": 15928} +{"lr": 6.752575658708793e-05, "data_time": 0.001557302474975586, "loss": 0.3773763120174408, "time": 0.636857271194458, "epoch": 5, "memory": 44138, "step": 16028} +{"lr": 6.71778135132408e-05, "data_time": 0.0014654874801635742, "loss": 0.355039456486702, "time": 0.6353830337524414, "epoch": 5, "memory": 44138, "step": 16128} +{"lr": 6.682908306751074e-05, "data_time": 0.0022002696990966798, "loss": 0.3726886361837387, "time": 0.6362100601196289, "epoch": 5, "memory": 44138, "step": 16228} +{"lr": 6.64795877975043e-05, "data_time": 0.001607990264892578, "loss": 0.3671862095594406, "time": 0.6362984657287598, "epoch": 5, "memory": 44138, "step": 16328} +{"lr": 6.61293503002787e-05, "data_time": 0.0015233755111694336, "loss": 0.36181028485298156, "time": 0.6364665985107422, "epoch": 5, "memory": 44138, "step": 16428} +{"lr": 6.577839322088084e-05, "data_time": 0.001546788215637207, "loss": 0.3405226916074753, "time": 0.6361546277999878, "epoch": 5, "memory": 44138, "step": 16528} +{"lr": 6.542673925088298e-05, "data_time": 0.0020236968994140625, "loss": 0.3623759776353836, "time": 0.6366914749145508, "epoch": 5, "memory": 44138, "step": 16628} +{"lr": 6.507441112691584e-05, "data_time": 0.001490473747253418, "loss": 0.3516934782266617, "time": 0.6364949226379395, "epoch": 5, "memory": 44138, "step": 16728} +{"lr": 6.47214316291983e-05, "data_time": 0.0015769243240356446, "loss": 0.35145682394504546, "time": 0.6363529205322266, "epoch": 5, "memory": 44138, "step": 16828} +{"lr": 6.436782358006481e-05, "data_time": 0.0022809505462646484, "loss": 0.3611027032136917, "time": 0.637720251083374, "epoch": 5, "memory": 44138, "step": 16928} +{"lr": 6.40136098424896e-05, "data_time": 0.001624298095703125, "loss": 0.37126933336257933, "time": 0.6365201473236084, "epoch": 5, "memory": 44138, "step": 17028} +{"lr": 6.365881331860828e-05, "data_time": 0.0018805503845214845, "loss": 0.3826365262269974, "time": 0.6374353647232056, "epoch": 5, "memory": 44138, "step": 17128} +{"lr": 6.330345694823739e-05, "data_time": 0.0016014814376831056, "loss": 0.36949982643127444, "time": 0.6373626708984375, "epoch": 5, "memory": 44138, "step": 17228} +{"lr": 6.29475637073908e-05, "data_time": 0.0023261308670043945, "loss": 0.36648235023021697, "time": 0.6376047611236573, "epoch": 5, "memory": 44138, "step": 17328} +{"lr": 6.259115660679482e-05, "data_time": 0.0015662908554077148, "loss": 0.36379197239875793, "time": 0.6365245103836059, "epoch": 5, "memory": 44138, "step": 17428} +{"lr": 6.223425869039963e-05, "data_time": 0.0014850616455078125, "loss": 0.37681972086429594, "time": 0.6362675666809082, "epoch": 5, "memory": 44138, "step": 17528} +{"lr": 6.187689303389006e-05, "data_time": 0.0016463279724121093, "loss": 0.36791214942932127, "time": 0.6367289543151855, "epoch": 5, "memory": 44138, "step": 17628} +{"lr": 6.151908274319292e-05, "data_time": 0.001893019676208496, "loss": 0.3668570637702942, "time": 0.6370482683181763, "epoch": 5, "memory": 44138, "step": 17728} +{"lr": 6.116085095298373e-05, "data_time": 0.0017044544219970703, "loss": 0.3675206959247589, "time": 0.6368045806884766, "epoch": 5, "memory": 44138, "step": 17828} +{"lr": 6.080222082519007e-05, "data_time": 0.001479792594909668, "loss": 0.34049825072288514, "time": 0.635310697555542, "epoch": 5, "memory": 44138, "step": 17928} +{"lr": 6.0443215547495147e-05, "data_time": 0.0021247386932373045, "loss": 0.39336840212345126, "time": 0.6370204210281372, "epoch": 5, "memory": 44138, "step": 18028} +{"lr": 6.0083858331837714e-05, "data_time": 0.0018485307693481446, "loss": 0.36739606559276583, "time": 0.6371299743652343, "epoch": 5, "memory": 44138, "step": 18128} +{"lr": 5.97241724129116e-05, "data_time": 0.00196075439453125, "loss": 0.35261605978012084, "time": 0.636580514907837, "epoch": 5, "memory": 44138, "step": 18228} +{"lr": 5.936418104666328e-05, "data_time": 0.0015727043151855468, "loss": 0.3733451128005981, "time": 0.6366521120071411, "epoch": 5, "memory": 44138, "step": 18328} +{"lr": 5.9003907508788295e-05, "data_time": 0.0018732309341430663, "loss": 0.38800789415836334, "time": 0.6371723890304566, "epoch": 5, "memory": 44138, "step": 18428} +{"lr": 5.864337509322633e-05, "data_time": 0.0017942428588867188, "loss": 0.3905669659376144, "time": 0.6373066425323486, "epoch": 5, "memory": 44138, "step": 18528} +{"lr": 5.8282607110655194e-05, "data_time": 0.0019335269927978516, "loss": 0.39688348174095156, "time": 0.6366997241973877, "epoch": 5, "memory": 44138, "step": 18628} +{"lr": 5.7921626886983564e-05, "data_time": 0.001709747314453125, "loss": 0.3747338384389877, "time": 0.6370071411132813, "epoch": 5, "memory": 44138, "step": 18728} +{"lr": 5.756045776184287e-05, "data_time": 0.0018059015274047852, "loss": 0.36166087687015536, "time": 0.6371424436569214, "epoch": 5, "memory": 44138, "step": 18828} +{"lr": 5.719912308707808e-05, "data_time": 0.0020849943161010743, "loss": 0.37360704839229586, "time": 0.6362600088119507, "epoch": 5, "memory": 44138, "step": 18928} +{"lr": 5.683764622523808e-05, "data_time": 0.0020946502685546876, "loss": 0.3509641170501709, "time": 0.6386455297470093, "epoch": 5, "memory": 44138, "step": 19028} +{"lr": 5.647605054806502e-05, "data_time": 0.0016429424285888672, "loss": 0.3507239788770676, "time": 0.6361999273300171, "epoch": 5, "memory": 44138, "step": 19128} +{"lr": 5.6114359434983206e-05, "data_time": 0.0015897750854492188, "loss": 0.383701291680336, "time": 0.6368448257446289, "epoch": 5, "memory": 44138, "step": 19228} +{"lr": 5.575259627158745e-05, "data_time": 0.0018629074096679688, "loss": 0.40328297317028045, "time": 0.636943769454956, "epoch": 5, "memory": 44138, "step": 19328} +{"lr": 5.539078444813113e-05, "data_time": 0.0015216350555419921, "loss": 0.3667666703462601, "time": 0.6371160507202148, "epoch": 5, "memory": 44138, "step": 19428} +{"lr": 5.5028947358013735e-05, "data_time": 0.001392531394958496, "loss": 0.34482610523700713, "time": 0.6352652549743653, "epoch": 5, "memory": 44138, "step": 19528} +{"accuracy/top1": 78.03138732910156, "data_time": 0.03266500604563746, "time": 1.2760491083408225, "step": 5} +{"lr": 5.464178020340108e-05, "data_time": 0.0024876117706298826, "loss": 0.37087360620498655, "time": 0.6373377561569213, "epoch": 6, "memory": 44138, "step": 19635} +{"lr": 5.427996514795046e-05, "data_time": 0.0021753549575805665, "loss": 0.37204011976718904, "time": 0.635506558418274, "epoch": 6, "memory": 44138, "step": 19735} +{"lr": 5.3918196647260834e-05, "data_time": 0.001603412628173828, "loss": 0.35330307185649873, "time": 0.6365888118743896, "epoch": 6, "memory": 44138, "step": 19835} +{"lr": 5.355649809193057e-05, "data_time": 0.001467275619506836, "loss": 0.37361868321895597, "time": 0.6379712581634521, "epoch": 6, "memory": 44138, "step": 19935} +{"lr": 5.3194892868035765e-05, "data_time": 0.001579117774963379, "loss": 0.36632903218269347, "time": 0.636756420135498, "epoch": 6, "memory": 44138, "step": 20035} +{"lr": 5.283340435561795e-05, "data_time": 0.0015182256698608398, "loss": 0.3887178868055344, "time": 0.6369266510009766, "epoch": 6, "memory": 44138, "step": 20135} +{"lr": 5.2472055927172495e-05, "data_time": 0.0014935970306396485, "loss": 0.3624693930149078, "time": 0.6368660926818848, "epoch": 6, "memory": 44138, "step": 20235} +{"lr": 5.2110870946137605e-05, "data_time": 0.0016383647918701172, "loss": 0.3593368470668793, "time": 0.6372593641281128, "epoch": 6, "memory": 44138, "step": 20335} +{"lr": 5.174987276538342e-05, "data_time": 0.0015296459197998047, "loss": 0.3751054733991623, "time": 0.6355548143386841, "epoch": 6, "memory": 44138, "step": 20435} +{"lr": 5.1389084725702425e-05, "data_time": 0.0016545772552490235, "loss": 0.3484298437833786, "time": 0.6362725734710694, "epoch": 6, "memory": 44138, "step": 20535} +{"lr": 5.1028530154300045e-05, "data_time": 0.001936793327331543, "loss": 0.35034675896167755, "time": 0.6363436222076416, "epoch": 6, "memory": 44138, "step": 20635} +{"lr": 5.066823236328663e-05, "data_time": 0.001694488525390625, "loss": 0.36730977296829226, "time": 0.6357667922973633, "epoch": 6, "memory": 44138, "step": 20735} +{"lr": 5.030821464817006e-05, "data_time": 0.0018495798110961914, "loss": 0.3620191186666489, "time": 0.6368024826049805, "epoch": 6, "memory": 44138, "step": 20835} +{"lr": 4.99485002863494e-05, "data_time": 0.0014292001724243164, "loss": 0.3624554634094238, "time": 0.6369625568389893, "epoch": 6, "memory": 44138, "step": 20935} +{"lr": 4.95891125356102e-05, "data_time": 0.001660919189453125, "loss": 0.3599185734987259, "time": 0.6478580236434937, "epoch": 6, "memory": 44138, "step": 21035} +{"lr": 4.923007463262053e-05, "data_time": 0.0020416259765625, "loss": 0.3875850856304169, "time": 0.639004135131836, "epoch": 6, "memory": 44138, "step": 21135} +{"lr": 4.887140979142852e-05, "data_time": 0.0017733812332153321, "loss": 0.35813171565532687, "time": 0.6365517377853394, "epoch": 6, "memory": 44138, "step": 21235} +{"lr": 4.851314120196161e-05, "data_time": 0.0017263650894165038, "loss": 0.3811752527952194, "time": 0.6365441083908081, "epoch": 6, "memory": 44138, "step": 21335} +{"lr": 4.815529202852693e-05, "data_time": 0.0017476558685302734, "loss": 0.3990250527858734, "time": 0.637282919883728, "epoch": 6, "memory": 44138, "step": 21435} +{"lr": 4.779788540831406e-05, "data_time": 0.001805734634399414, "loss": 0.39963223338127135, "time": 0.6367613554000855, "epoch": 6, "memory": 44138, "step": 21535} +{"lr": 4.744094444989843e-05, "data_time": 0.001782989501953125, "loss": 0.3687055140733719, "time": 0.637285566329956, "epoch": 6, "memory": 44138, "step": 21635} +{"lr": 4.708449223174764e-05, "data_time": 0.0017920732498168945, "loss": 0.35472665429115297, "time": 0.6371837377548217, "epoch": 6, "memory": 44138, "step": 21735} +{"lr": 4.672855180072916e-05, "data_time": 0.0017628908157348634, "loss": 0.3685429602861404, "time": 0.6370863676071167, "epoch": 6, "memory": 44138, "step": 21835} +{"lr": 4.6373146170620256e-05, "data_time": 0.0015166044235229493, "loss": 0.35768717229366304, "time": 0.6370740175247193, "epoch": 6, "memory": 44138, "step": 21935} +{"lr": 4.6018298320619915e-05, "data_time": 0.0015221357345581055, "loss": 0.38128423392772676, "time": 0.6368333578109742, "epoch": 6, "memory": 44138, "step": 22035} +{"lr": 4.566403119386318e-05, "data_time": 0.0014801740646362305, "loss": 0.3583536595106125, "time": 0.6373617887496948, "epoch": 6, "memory": 44138, "step": 22135} +{"lr": 4.531036769593769e-05, "data_time": 0.0016695737838745117, "loss": 0.3852864295244217, "time": 0.6379703283309937, "epoch": 6, "memory": 44138, "step": 22235} +{"lr": 4.4957330693402685e-05, "data_time": 0.0021849155426025392, "loss": 0.3721946179866791, "time": 0.636847972869873, "epoch": 6, "memory": 44138, "step": 22335} +{"lr": 4.460494301231059e-05, "data_time": 0.0015393733978271485, "loss": 0.3711391031742096, "time": 0.636501407623291, "epoch": 6, "memory": 44138, "step": 22435} +{"lr": 4.425322743673109e-05, "data_time": 0.0016908645629882812, "loss": 0.381898033618927, "time": 0.6368586540222168, "epoch": 6, "memory": 44138, "step": 22535} +{"lr": 4.390220670727812e-05, "data_time": 0.002420854568481445, "loss": 0.33969086706638335, "time": 0.637241530418396, "epoch": 6, "memory": 44138, "step": 22635} +{"lr": 4.3551903519639325e-05, "data_time": 0.0017018795013427734, "loss": 0.3634351372718811, "time": 0.6370144367218018, "epoch": 6, "memory": 44138, "step": 22735} +{"lr": 4.3202340523108894e-05, "data_time": 0.002078127861022949, "loss": 0.3723094642162323, "time": 0.6359060764312744, "epoch": 6, "memory": 44138, "step": 22835} +{"lr": 4.285354031912302e-05, "data_time": 0.0016663312911987306, "loss": 0.36899976432323456, "time": 0.6357746839523315, "epoch": 6, "memory": 44138, "step": 22935} +{"lr": 4.2505525459798544e-05, "data_time": 0.0017359733581542968, "loss": 0.38210780918598175, "time": 0.635829758644104, "epoch": 6, "memory": 44138, "step": 23035} +{"lr": 4.2158318446474836e-05, "data_time": 0.0018004417419433595, "loss": 0.3862610787153244, "time": 0.6364174604415893, "epoch": 6, "memory": 44138, "step": 23135} +{"lr": 4.181194172825905e-05, "data_time": 0.0018860816955566407, "loss": 0.36710751354694365, "time": 0.6360246896743774, "epoch": 6, "memory": 44138, "step": 23235} +{"lr": 4.146641770057448e-05, "data_time": 0.0017897605895996094, "loss": 0.36676945686340334, "time": 0.6363061666488647, "epoch": 6, "memory": 44138, "step": 23335} +{"lr": 4.112176870371267e-05, "data_time": 0.0017312526702880859, "loss": 0.3598135381937027, "time": 0.6366748332977294, "epoch": 6, "memory": 44138, "step": 23435} +{"accuracy/top1": 77.80870819091797, "data_time": 0.031339752263036265, "time": 1.2763516368537113, "step": 6} +{"lr": 4.0753988582732786e-05, "data_time": 0.00150299072265625, "loss": 0.37958675920963286, "time": 0.6361352205276489, "epoch": 7, "memory": 44138, "step": 23542} +{"lr": 4.0411221639286815e-05, "data_time": 0.0015937328338623048, "loss": 0.409321540594101, "time": 0.6369438171386719, "epoch": 7, "memory": 44138, "step": 23642} +{"lr": 4.006939795169518e-05, "data_time": 0.001965022087097168, "loss": 0.3521025270223618, "time": 0.6362314462661743, "epoch": 7, "memory": 44138, "step": 23742} +{"lr": 3.972853962099913e-05, "data_time": 0.001357722282409668, "loss": 0.33720476031303404, "time": 0.6374253034591675, "epoch": 7, "memory": 44138, "step": 23842} +{"lr": 3.93886686858237e-05, "data_time": 0.0016610383987426757, "loss": 0.39819527566432955, "time": 0.6368105173110962, "epoch": 7, "memory": 44138, "step": 23942} +{"lr": 3.904980712095259e-05, "data_time": 0.0015917778015136718, "loss": 0.3759404569864273, "time": 0.6376616477966308, "epoch": 7, "memory": 44138, "step": 24042} +{"lr": 3.871197683590748e-05, "data_time": 0.001981186866760254, "loss": 0.37075887620449066, "time": 0.6376721620559692, "epoch": 7, "memory": 44138, "step": 24142} +{"lr": 3.837519967353125e-05, "data_time": 0.0017904281616210938, "loss": 0.38417728841304777, "time": 0.6366328001022339, "epoch": 7, "memory": 44138, "step": 24242} +{"lr": 3.8039497408575855e-05, "data_time": 0.0023898839950561523, "loss": 0.3890520006418228, "time": 0.6377934455871582, "epoch": 7, "memory": 44138, "step": 24342} +{"lr": 3.770489174629432e-05, "data_time": 0.0019491195678710937, "loss": 0.38422541320323944, "time": 0.6449474096298218, "epoch": 7, "memory": 44138, "step": 24442} +{"lr": 3.737140432103754e-05, "data_time": 0.0015803337097167968, "loss": 0.3762570321559906, "time": 0.6376269817352295, "epoch": 7, "memory": 44138, "step": 24542} +{"lr": 3.703905669485545e-05, "data_time": 0.0015498161315917968, "loss": 0.4037273466587067, "time": 0.636371660232544, "epoch": 7, "memory": 44138, "step": 24642} +{"lr": 3.6707870356102706e-05, "data_time": 0.0017011404037475587, "loss": 0.3758342534303665, "time": 0.6364459276199341, "epoch": 7, "memory": 44138, "step": 24742} +{"lr": 3.6377866718049484e-05, "data_time": 0.001623845100402832, "loss": 0.3691035211086273, "time": 0.6371749401092529, "epoch": 7, "memory": 44138, "step": 24842} +{"lr": 3.6049067117496965e-05, "data_time": 0.001714324951171875, "loss": 0.34939275979995726, "time": 0.6372305393218994, "epoch": 7, "memory": 44138, "step": 24942} +{"lr": 3.5721492813397755e-05, "data_time": 0.00166628360748291, "loss": 0.378491273522377, "time": 0.6374207973480225, "epoch": 7, "memory": 44138, "step": 25042} +{"lr": 3.539516498548151e-05, "data_time": 0.0015749931335449219, "loss": 0.36208135783672335, "time": 0.6369609832763672, "epoch": 7, "memory": 44138, "step": 25142} +{"lr": 3.507010473288518e-05, "data_time": 0.0014975309371948243, "loss": 0.3824771374464035, "time": 0.6363403081893921, "epoch": 7, "memory": 44138, "step": 25242} +{"lr": 3.474633307278909e-05, "data_time": 0.0018010139465332031, "loss": 0.3768818318843842, "time": 0.6368935346603394, "epoch": 7, "memory": 44138, "step": 25342} +{"lr": 3.4423870939058177e-05, "data_time": 0.001985311508178711, "loss": 0.37051327228546144, "time": 0.6367572546005249, "epoch": 7, "memory": 44138, "step": 25442} +{"lr": 3.410273918088805e-05, "data_time": 0.001759958267211914, "loss": 0.36509051620960237, "time": 0.6366346120834351, "epoch": 7, "memory": 44138, "step": 25542} +{"lr": 3.37829585614574e-05, "data_time": 0.0016704082489013671, "loss": 0.3870923310518265, "time": 0.6371009349822998, "epoch": 7, "memory": 44138, "step": 25642} +{"lr": 3.3464549756585216e-05, "data_time": 0.0016191959381103515, "loss": 0.3867280274629593, "time": 0.6364024162292481, "epoch": 7, "memory": 44138, "step": 25742} +{"lr": 3.314753335339416e-05, "data_time": 0.0020956993103027344, "loss": 0.3788897156715393, "time": 0.6379543781280518, "epoch": 7, "memory": 44138, "step": 25842} +{"lr": 3.2831929848979394e-05, "data_time": 0.0016489505767822265, "loss": 0.40125176012516023, "time": 0.6375192880630494, "epoch": 7, "memory": 44138, "step": 25942} +{"lr": 3.251775964908332e-05, "data_time": 0.0016282081604003906, "loss": 0.36865064799785613, "time": 0.6376344442367554, "epoch": 7, "memory": 44138, "step": 26042} +{"lr": 3.22050430667762e-05, "data_time": 0.0014970779418945312, "loss": 0.38661822080612185, "time": 0.6375813245773315, "epoch": 7, "memory": 44138, "step": 26142} +{"lr": 3.189380032114285e-05, "data_time": 0.001978421211242676, "loss": 0.3730123072862625, "time": 0.6370380878448486, "epoch": 7, "memory": 44138, "step": 26242} +{"lr": 3.1584051535975324e-05, "data_time": 0.001821446418762207, "loss": 0.37764213383197787, "time": 0.6370634317398072, "epoch": 7, "memory": 44138, "step": 26342} +{"lr": 3.127581673847175e-05, "data_time": 0.001682138442993164, "loss": 0.3637180894613266, "time": 0.635551929473877, "epoch": 7, "memory": 44138, "step": 26442} +{"lr": 3.0969115857941555e-05, "data_time": 0.0014571666717529297, "loss": 0.3561870127916336, "time": 0.635262131690979, "epoch": 7, "memory": 44138, "step": 26542} +{"lr": 3.0663968724516654e-05, "data_time": 0.0016916036605834962, "loss": 0.36786005795001986, "time": 0.6352062940597534, "epoch": 7, "memory": 44138, "step": 26642} +{"lr": 3.0360395067869668e-05, "data_time": 0.0023400306701660155, "loss": 0.38404776751995084, "time": 0.6361199378967285, "epoch": 7, "memory": 44138, "step": 26742} +{"lr": 3.0058414515938063e-05, "data_time": 0.0015816926956176759, "loss": 0.38932461738586427, "time": 0.6373611450195312, "epoch": 7, "memory": 44138, "step": 26842} +{"lr": 2.975804659365501e-05, "data_time": 0.001865553855895996, "loss": 0.3752181679010391, "time": 0.6381101131439209, "epoch": 7, "memory": 44138, "step": 26942} +{"lr": 2.945931072168714e-05, "data_time": 0.0015078544616699218, "loss": 0.3806620329618454, "time": 0.6370229959487915, "epoch": 7, "memory": 44138, "step": 27042} +{"lr": 2.9162226215178855e-05, "data_time": 0.001818370819091797, "loss": 0.39434253573417666, "time": 0.6375951528549194, "epoch": 7, "memory": 44138, "step": 27142} +{"lr": 2.8866812282503322e-05, "data_time": 0.0014791250228881835, "loss": 0.36256890296936034, "time": 0.6377808809280395, "epoch": 7, "memory": 44138, "step": 27242} +{"lr": 2.8573088024020838e-05, "data_time": 0.0013997554779052734, "loss": 0.37059479355812075, "time": 0.6366333723068237, "epoch": 7, "memory": 44138, "step": 27342} +{"accuracy/top1": 77.22417449951172, "data_time": 0.031212153105900205, "time": 1.2760090129128818, "step": 7} +{"lr": 2.8260695817390092e-05, "data_time": 0.0017768621444702148, "loss": 0.35927407443523407, "time": 0.6357892274856567, "epoch": 8, "memory": 44138, "step": 27449} +{"lr": 2.7970529402570967e-05, "data_time": 0.002074456214904785, "loss": 0.3875000327825546, "time": 0.635071086883545, "epoch": 8, "memory": 44138, "step": 27549} +{"lr": 2.7682110612244214e-05, "data_time": 0.0014928340911865234, "loss": 0.3755984216928482, "time": 0.6351062059402466, "epoch": 8, "memory": 44138, "step": 27649} +{"lr": 2.7395458094490283e-05, "data_time": 0.001944732666015625, "loss": 0.3727742224931717, "time": 0.6350639343261719, "epoch": 8, "memory": 44138, "step": 27749} +{"lr": 2.7110590383189015e-05, "data_time": 0.0018382310867309571, "loss": 0.37217639684677123, "time": 0.6346689462661743, "epoch": 8, "memory": 44138, "step": 27849} +{"lr": 2.6827525896821397e-05, "data_time": 0.0015794992446899413, "loss": 0.4010009288787842, "time": 0.6375937223434448, "epoch": 8, "memory": 44138, "step": 27949} +{"lr": 2.6546282937278552e-05, "data_time": 0.0026099443435668944, "loss": 0.3738888442516327, "time": 0.6376948118209839, "epoch": 8, "memory": 44138, "step": 28049} +{"lr": 2.6266879688678606e-05, "data_time": 0.0015447378158569337, "loss": 0.37924714386463165, "time": 0.6351224899291992, "epoch": 8, "memory": 44138, "step": 28149} +{"lr": 2.5989334216190738e-05, "data_time": 0.0018911123275756835, "loss": 0.381095290184021, "time": 0.6379172563552856, "epoch": 8, "memory": 44138, "step": 28249} +{"lr": 2.571366446486732e-05, "data_time": 0.0016260862350463868, "loss": 0.39325780272483823, "time": 0.6374745607376099, "epoch": 8, "memory": 44138, "step": 28349} +{"lr": 2.5439888258483665e-05, "data_time": 0.0015037775039672852, "loss": 0.35682632923126223, "time": 0.6357267379760743, "epoch": 8, "memory": 44138, "step": 28449} +{"lr": 2.5168023298385384e-05, "data_time": 0.0017645597457885743, "loss": 0.345992836356163, "time": 0.6345254182815552, "epoch": 8, "memory": 44138, "step": 28549} +{"lr": 2.489808716234412e-05, "data_time": 0.0015947818756103516, "loss": 0.355545736849308, "time": 0.6378972291946411, "epoch": 8, "memory": 44138, "step": 28649} +{"lr": 2.463009730342111e-05, "data_time": 0.002026057243347168, "loss": 0.38551085591316225, "time": 0.6361251831054687, "epoch": 8, "memory": 44138, "step": 28749} +{"lr": 2.436407104883849e-05, "data_time": 0.0015492200851440429, "loss": 0.37640159726142886, "time": 0.6361212730407715, "epoch": 8, "memory": 44138, "step": 28849} +{"lr": 2.4100025598859092e-05, "data_time": 0.0017190933227539062, "loss": 0.3730620265007019, "time": 0.6357465982437134, "epoch": 8, "memory": 44138, "step": 28949} +{"lr": 2.3837978025674284e-05, "data_time": 0.0015191078186035157, "loss": 0.35198096334934237, "time": 0.6363602638244629, "epoch": 8, "memory": 44138, "step": 29049} +{"lr": 2.357794527230027e-05, "data_time": 0.0023094654083251954, "loss": 0.3522915422916412, "time": 0.6370986461639404, "epoch": 8, "memory": 44138, "step": 29149} +{"lr": 2.33199441514825e-05, "data_time": 0.0018253326416015625, "loss": 0.39003380835056306, "time": 0.6351585149765014, "epoch": 8, "memory": 44138, "step": 29249} +{"lr": 2.306399134460859e-05, "data_time": 0.0017494678497314454, "loss": 0.37878172397613524, "time": 0.6375424861907959, "epoch": 8, "memory": 44138, "step": 29349} +{"lr": 2.281010340062995e-05, "data_time": 0.0016418933868408204, "loss": 0.38247411251068114, "time": 0.635607099533081, "epoch": 8, "memory": 44138, "step": 29449} +{"lr": 2.2558296734991603e-05, "data_time": 0.001974749565124512, "loss": 0.38675096333026887, "time": 0.6354138612747192, "epoch": 8, "memory": 44138, "step": 29549} +{"lr": 2.2308587628570915e-05, "data_time": 0.0021421432495117186, "loss": 0.34000577330589293, "time": 0.6485737085342407, "epoch": 8, "memory": 44138, "step": 29649} +{"lr": 2.2060992226624955e-05, "data_time": 0.0017456769943237304, "loss": 0.35701105892658236, "time": 0.6358763694763183, "epoch": 8, "memory": 44138, "step": 29749} +{"lr": 2.181552653774648e-05, "data_time": 0.0016990184783935546, "loss": 0.36373075246810915, "time": 0.6351004123687745, "epoch": 8, "memory": 44138, "step": 29849} +{"lr": 2.1572206432829035e-05, "data_time": 0.0018560647964477538, "loss": 0.3286889672279358, "time": 0.6363370656967163, "epoch": 8, "memory": 44138, "step": 29949} +{"lr": 2.1331047644040754e-05, "data_time": 0.002002406120300293, "loss": 0.39221576750278475, "time": 0.6356395483016968, "epoch": 8, "memory": 44138, "step": 30049} +{"lr": 2.109206576380705e-05, "data_time": 0.002071785926818848, "loss": 0.3737366795539856, "time": 0.6362970352172852, "epoch": 8, "memory": 44138, "step": 30149} +{"lr": 2.0855276243802687e-05, "data_time": 0.0017470836639404297, "loss": 0.3948597311973572, "time": 0.6365963220596313, "epoch": 8, "memory": 44138, "step": 30249} +{"lr": 2.0620694393952487e-05, "data_time": 0.0016836404800415039, "loss": 0.3693047225475311, "time": 0.6352022171020508, "epoch": 8, "memory": 44138, "step": 30349} +{"lr": 2.038833538144169e-05, "data_time": 0.001968693733215332, "loss": 0.34211274087429044, "time": 0.6348795413970947, "epoch": 8, "memory": 44138, "step": 30449} +{"lr": 2.0158214229735154e-05, "data_time": 0.002071523666381836, "loss": 0.3605386734008789, "time": 0.634950852394104, "epoch": 8, "memory": 44138, "step": 30549} +{"lr": 1.9930345817605977e-05, "data_time": 0.002180933952331543, "loss": 0.35083739161491395, "time": 0.6377945661544799, "epoch": 8, "memory": 44138, "step": 30649} +{"lr": 1.970474487817358e-05, "data_time": 0.0018900394439697265, "loss": 0.3833119809627533, "time": 0.6348079681396485, "epoch": 8, "memory": 44138, "step": 30749} +{"lr": 1.9481425997951092e-05, "data_time": 0.001707005500793457, "loss": 0.37710258066654206, "time": 0.6366045713424683, "epoch": 8, "memory": 44138, "step": 30849} +{"lr": 1.9260403615902187e-05, "data_time": 0.001898503303527832, "loss": 0.37433295249938964, "time": 0.6373776912689209, "epoch": 8, "memory": 44138, "step": 30949} +{"lr": 1.9041692022507647e-05, "data_time": 0.0017877340316772461, "loss": 0.3876935660839081, "time": 0.6366706609725952, "epoch": 8, "memory": 44138, "step": 31049} +{"lr": 1.8825305358841125e-05, "data_time": 0.0017325162887573242, "loss": 0.3609554827213287, "time": 0.6356470584869385, "epoch": 8, "memory": 44138, "step": 31149} +{"lr": 1.861125761565518e-05, "data_time": 0.0017855644226074218, "loss": 0.3503482758998871, "time": 0.6341592311859131, "epoch": 8, "memory": 44138, "step": 31249} +{"accuracy/top1": 77.11631774902344, "data_time": 0.031988135699568127, "time": 1.275291237337836, "step": 8} +{"lr": 1.8384832449122864e-05, "data_time": 0.0017569541931152343, "loss": 0.37776540517807006, "time": 0.6363374710083007, "epoch": 9, "memory": 44138, "step": 31356} +{"lr": 1.817567007332114e-05, "data_time": 0.0021444320678710937, "loss": 0.36052057445049285, "time": 0.6455218315124511, "epoch": 9, "memory": 44138, "step": 31456} +{"lr": 1.796888862098825e-05, "data_time": 0.0021024465560913084, "loss": 0.37929793298244474, "time": 0.6360148906707763, "epoch": 9, "memory": 44138, "step": 31556} +{"lr": 1.7764501461839678e-05, "data_time": 0.0017917871475219727, "loss": 0.37280177474021914, "time": 0.636647629737854, "epoch": 9, "memory": 44138, "step": 31656} +{"lr": 1.7562521810784746e-05, "data_time": 0.002162933349609375, "loss": 0.37447566986083985, "time": 0.636457896232605, "epoch": 9, "memory": 44138, "step": 31756} +{"lr": 1.7362962727072444e-05, "data_time": 0.0023943901062011717, "loss": 0.3473565518856049, "time": 0.6358568429946899, "epoch": 9, "memory": 44138, "step": 31856} +{"lr": 1.716583711344682e-05, "data_time": 0.002496647834777832, "loss": 0.3924066424369812, "time": 0.6379173278808594, "epoch": 9, "memory": 44138, "step": 31956} +{"lr": 1.6971157715312954e-05, "data_time": 0.001843571662902832, "loss": 0.35017883479595185, "time": 0.638804841041565, "epoch": 9, "memory": 44138, "step": 32056} +{"lr": 1.677893711991274e-05, "data_time": 0.0016889333724975585, "loss": 0.36119251549243925, "time": 0.6374097347259522, "epoch": 9, "memory": 44138, "step": 32156} +{"lr": 1.6589187755511025e-05, "data_time": 0.0020768404006958007, "loss": 0.35062299072742464, "time": 0.6507104396820068, "epoch": 9, "memory": 44138, "step": 32256} +{"lr": 1.640192189059216e-05, "data_time": 0.0019334077835083008, "loss": 0.3661290407180786, "time": 0.6358798742294312, "epoch": 9, "memory": 44138, "step": 32356} +{"lr": 1.6217151633066716e-05, "data_time": 0.0020746231079101563, "loss": 0.36143653392791747, "time": 0.6372776985168457, "epoch": 9, "memory": 44138, "step": 32456} +{"lr": 1.6034888929488564e-05, "data_time": 0.0018400430679321289, "loss": 0.36948657035827637, "time": 0.6380186319351197, "epoch": 9, "memory": 44138, "step": 32556} +{"lr": 1.585514556428258e-05, "data_time": 0.0018936634063720704, "loss": 0.36524722576141355, "time": 0.6360497236251831, "epoch": 9, "memory": 44138, "step": 32656} +{"lr": 1.567793315898261e-05, "data_time": 0.0021505832672119142, "loss": 0.38852599561214446, "time": 0.6369391202926635, "epoch": 9, "memory": 44138, "step": 32756} +{"lr": 1.5503263171480048e-05, "data_time": 0.001978635787963867, "loss": 0.38102494478225707, "time": 0.6393379449844361, "epoch": 9, "memory": 44138, "step": 32856} +{"lr": 1.5331146895283165e-05, "data_time": 0.0019562959671020506, "loss": 0.37116784453392027, "time": 0.6374693393707276, "epoch": 9, "memory": 44138, "step": 32956} +{"lr": 1.5161595458786728e-05, "data_time": 0.0017194032669067382, "loss": 0.3558268159627914, "time": 0.6382407188415528, "epoch": 9, "memory": 44138, "step": 33056} +{"lr": 1.4994619824552563e-05, "data_time": 0.0019527196884155274, "loss": 0.37153142094612124, "time": 0.63702871799469, "epoch": 9, "memory": 44138, "step": 33156} +{"lr": 1.4830230788600802e-05, "data_time": 0.0019491910934448242, "loss": 0.3914036899805069, "time": 0.6374222755432128, "epoch": 9, "memory": 44138, "step": 33256} +{"lr": 1.4668438979711779e-05, "data_time": 0.001941847801208496, "loss": 0.370823061466217, "time": 0.6398892641067505, "epoch": 9, "memory": 44138, "step": 33356} +{"lr": 1.4509254858738844e-05, "data_time": 0.0022425174713134764, "loss": 0.3787620007991791, "time": 0.6387940168380737, "epoch": 9, "memory": 44138, "step": 33456} +{"lr": 1.4352688717931966e-05, "data_time": 0.0024170875549316406, "loss": 0.3609553128480911, "time": 0.638194227218628, "epoch": 9, "memory": 44138, "step": 33556} +{"lr": 1.4198750680272348e-05, "data_time": 0.0021574735641479493, "loss": 0.3617054492235184, "time": 0.6384587526321411, "epoch": 9, "memory": 44138, "step": 33656} +{"lr": 1.4047450698817857e-05, "data_time": 0.0017816543579101563, "loss": 0.3511413484811783, "time": 0.6382307767868042, "epoch": 9, "memory": 44138, "step": 33756} +{"lr": 1.389879855605952e-05, "data_time": 0.0016302108764648438, "loss": 0.3800592184066772, "time": 0.6376355648040771, "epoch": 9, "memory": 44138, "step": 33856} +{"lr": 1.3752803863288976e-05, "data_time": 0.001538991928100586, "loss": 0.3761305302381516, "time": 0.6357406377792358, "epoch": 9, "memory": 44138, "step": 33956} +{"lr": 1.3609476059977129e-05, "data_time": 0.002507805824279785, "loss": 0.3871601551771164, "time": 0.6384088277816773, "epoch": 9, "memory": 44138, "step": 34056} +{"lr": 1.3468824413163754e-05, "data_time": 0.0021561145782470702, "loss": 0.3784525364637375, "time": 0.6383571624755859, "epoch": 9, "memory": 44138, "step": 34156} +{"lr": 1.3330858016858395e-05, "data_time": 0.0026237010955810548, "loss": 0.398921412229538, "time": 0.6372433662414551, "epoch": 9, "memory": 44138, "step": 34256} +{"lr": 1.3195585791452309e-05, "data_time": 0.0019191741943359376, "loss": 0.37292216420173646, "time": 0.6371522665023803, "epoch": 9, "memory": 44138, "step": 34356} +{"lr": 1.3063016483141729e-05, "data_time": 0.001530003547668457, "loss": 0.3519412994384766, "time": 0.6357794046401978, "epoch": 9, "memory": 44138, "step": 34456} +{"lr": 1.2933158663362388e-05, "data_time": 0.002068305015563965, "loss": 0.3784942656755447, "time": 0.6371706247329711, "epoch": 9, "memory": 44138, "step": 34556} +{"lr": 1.2806020728235311e-05, "data_time": 0.001889181137084961, "loss": 0.3695018321275711, "time": 0.6370637893676758, "epoch": 9, "memory": 44138, "step": 34656} +{"lr": 1.2681610898023963e-05, "data_time": 0.0019256353378295898, "loss": 0.3664830893278122, "time": 0.6462860822677612, "epoch": 9, "memory": 44138, "step": 34756} +{"lr": 1.2559937216602731e-05, "data_time": 0.0019288778305053711, "loss": 0.36039121448993683, "time": 0.6378018140792847, "epoch": 9, "memory": 44138, "step": 34856} +{"lr": 1.2441007550936847e-05, "data_time": 0.0018878459930419921, "loss": 0.39259982407093047, "time": 0.6406694889068604, "epoch": 9, "memory": 44138, "step": 34956} +{"lr": 1.232482959057377e-05, "data_time": 0.0021547555923461916, "loss": 0.40247650146484376, "time": 0.6380646228790283, "epoch": 9, "memory": 44138, "step": 35056} +{"lr": 1.2211410847145969e-05, "data_time": 0.002462339401245117, "loss": 0.3598687320947647, "time": 0.6378409147262574, "epoch": 9, "memory": 44138, "step": 35156} +{"accuracy/top1": 76.75446319580078, "data_time": 0.031586264741831814, "time": 1.2749265555677742, "step": 9} +{"lr": 1.209311679360931e-05, "data_time": 0.002844357490539551, "loss": 0.3891842246055603, "time": 0.6390421867370606, "epoch": 10, "memory": 44138, "step": 35263} +{"lr": 1.1985432727488666e-05, "data_time": 0.0025748729705810545, "loss": 0.36259826123714445, "time": 0.6365508079528809, "epoch": 10, "memory": 44138, "step": 35363} +{"lr": 1.1880529822435289e-05, "data_time": 0.0026526689529418946, "loss": 0.38314634561538696, "time": 0.651921558380127, "epoch": 10, "memory": 44138, "step": 35463} +{"lr": 1.1778414861078726e-05, "data_time": 0.0029461860656738283, "loss": 0.35298205018043516, "time": 0.6399567365646363, "epoch": 10, "memory": 44138, "step": 35563} +{"lr": 1.1679094445790538e-05, "data_time": 0.0024763107299804687, "loss": 0.3704716444015503, "time": 0.6371896028518677, "epoch": 10, "memory": 44138, "step": 35663} +{"lr": 1.1582574998257377e-05, "data_time": 0.002500033378601074, "loss": 0.36506556868553164, "time": 0.6490757465362549, "epoch": 10, "memory": 44138, "step": 35763} +{"lr": 1.1488862759065762e-05, "data_time": 0.0023409128189086914, "loss": 0.37412916123867035, "time": 0.6364859342575073, "epoch": 10, "memory": 44138, "step": 35863} +{"lr": 1.1397963787298665e-05, "data_time": 0.0025920629501342773, "loss": 0.3624534010887146, "time": 0.6381866693496704, "epoch": 10, "memory": 44138, "step": 35963} +{"lr": 1.1309883960143664e-05, "data_time": 0.0023764610290527345, "loss": 0.34261963963508607, "time": 0.6460938453674316, "epoch": 10, "memory": 44138, "step": 36063} +{"lr": 1.122462897251302e-05, "data_time": 0.003063464164733887, "loss": 0.36822369396686555, "time": 0.6370961666107178, "epoch": 10, "memory": 44138, "step": 36163} +{"lr": 1.114220433667542e-05, "data_time": 0.0031174182891845702, "loss": 0.3779571563005447, "time": 0.6410081386566162, "epoch": 10, "memory": 44138, "step": 36263} +{"lr": 1.10626153818996e-05, "data_time": 0.00319979190826416, "loss": 0.37404989302158353, "time": 0.637820553779602, "epoch": 10, "memory": 44138, "step": 36363} +{"lr": 1.0985867254109761e-05, "data_time": 0.001939535140991211, "loss": 0.3587206810712814, "time": 0.636319613456726, "epoch": 10, "memory": 44138, "step": 36463} +{"lr": 1.0911964915552823e-05, "data_time": 0.002281594276428223, "loss": 0.3492222785949707, "time": 0.6368070125579834, "epoch": 10, "memory": 44138, "step": 36563} +{"lr": 1.0840913144477643e-05, "data_time": 0.0018481731414794922, "loss": 0.36618653535842893, "time": 0.6374979972839355, "epoch": 10, "memory": 44138, "step": 36663} +{"lr": 1.0772716534826064e-05, "data_time": 0.0025023698806762697, "loss": 0.3560271501541138, "time": 0.6376560449600219, "epoch": 10, "memory": 44138, "step": 36763} +{"lr": 1.0707379495935828e-05, "data_time": 0.0031203031539916992, "loss": 0.3666188925504684, "time": 0.6392525911331177, "epoch": 10, "memory": 44138, "step": 36863} +{"lr": 1.0644906252255541e-05, "data_time": 0.0027537107467651366, "loss": 0.38266485631465913, "time": 0.6370707988739014, "epoch": 10, "memory": 44138, "step": 36963} +{"lr": 1.0585300843071527e-05, "data_time": 0.002393484115600586, "loss": 0.3794155478477478, "time": 0.6477534532546997, "epoch": 10, "memory": 44138, "step": 37063} +{"lr": 1.0528567122246654e-05, "data_time": 0.002044534683227539, "loss": 0.3628378063440323, "time": 0.6361784696578979, "epoch": 10, "memory": 44138, "step": 37163} +{"lr": 1.0474708757971141e-05, "data_time": 0.0025911808013916017, "loss": 0.3967668205499649, "time": 0.6381731033325195, "epoch": 10, "memory": 44138, "step": 37263} +{"lr": 1.042372923252543e-05, "data_time": 0.0016278743743896485, "loss": 0.36560956537723543, "time": 0.6358978509902954, "epoch": 10, "memory": 44138, "step": 37363} +{"lr": 1.0375631842054998e-05, "data_time": 0.0021711111068725584, "loss": 0.36271997094154357, "time": 0.6368345260620117, "epoch": 10, "memory": 44138, "step": 37463} +{"lr": 1.0330419696357254e-05, "data_time": 0.002533555030822754, "loss": 0.3798720002174377, "time": 0.6369867086410522, "epoch": 10, "memory": 44138, "step": 37563} +{"lr": 1.028809571868047e-05, "data_time": 0.002904343605041504, "loss": 0.36764702796936033, "time": 0.6370655059814453, "epoch": 10, "memory": 44138, "step": 37663} +{"lr": 1.0248662645534766e-05, "data_time": 0.002500581741333008, "loss": 0.34937004148960116, "time": 0.6373331785202027, "epoch": 10, "memory": 44138, "step": 37763} +{"lr": 1.0212123026515194e-05, "data_time": 0.002648138999938965, "loss": 0.3732252478599548, "time": 0.6372593879699707, "epoch": 10, "memory": 44138, "step": 37863} +{"lr": 1.0178479224136881e-05, "data_time": 0.002384328842163086, "loss": 0.3445000916719437, "time": 0.6363013982772827, "epoch": 10, "memory": 44138, "step": 37963} +{"lr": 1.0147733413682289e-05, "data_time": 0.002006673812866211, "loss": 0.35463112890720366, "time": 0.6370181798934936, "epoch": 10, "memory": 44138, "step": 38063} +{"lr": 1.0119887583060527e-05, "data_time": 0.0016669988632202148, "loss": 0.37355350852012636, "time": 0.6363584041595459, "epoch": 10, "memory": 44138, "step": 38163} +{"lr": 1.0094943532678871e-05, "data_time": 0.002047157287597656, "loss": 0.36430065631866454, "time": 0.6355287075042725, "epoch": 10, "memory": 44138, "step": 38263} +{"lr": 1.007290287532637e-05, "data_time": 0.002046823501586914, "loss": 0.3745395392179489, "time": 0.6372630834579468, "epoch": 10, "memory": 44138, "step": 38363} +{"lr": 1.0053767036069495e-05, "data_time": 0.00197749137878418, "loss": 0.3921888291835785, "time": 0.6387291669845581, "epoch": 10, "memory": 44138, "step": 38463} +{"lr": 1.003753725216005e-05, "data_time": 0.002440333366394043, "loss": 0.36490239202976227, "time": 0.6365254640579223, "epoch": 10, "memory": 44138, "step": 38563} +{"lr": 1.0024214572955152e-05, "data_time": 0.0018944740295410156, "loss": 0.35240968465805056, "time": 0.6374494314193726, "epoch": 10, "memory": 44138, "step": 38663} +{"lr": 1.0013799859849456e-05, "data_time": 0.0018710613250732422, "loss": 0.40322225391864774, "time": 0.6369835138320923, "epoch": 10, "memory": 44138, "step": 38763} +{"lr": 1.0006293786219338e-05, "data_time": 0.0025926828384399414, "loss": 0.386372908949852, "time": 0.6362984657287598, "epoch": 10, "memory": 44138, "step": 38863} +{"lr": 1.0001696837379472e-05, "data_time": 0.0026990652084350588, "loss": 0.3724012911319733, "time": 0.6380775451660157, "epoch": 10, "memory": 44138, "step": 38963} +{"lr": 1.0000009310551361e-05, "data_time": 0.0021074295043945314, "loss": 0.38128813207149503, "time": 0.6352448463439941, "epoch": 10, "memory": 44138, "step": 39063} +{"accuracy/top1": 76.35781860351562, "data_time": 0.03211379051208496, "time": 1.2754385717983903, "step": 10} diff --git a/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm_wneck/20230604_032014/vis_data/config.py b/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm_wneck/20230604_032014/vis_data/config.py new file mode 100644 index 0000000000000000000000000000000000000000..455960112013988f06579e1bb256e4040095dfad --- /dev/null +++ b/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm_wneck/20230604_032014/vis_data/config.py @@ -0,0 +1,236 @@ +optim_wrapper = dict( + optimizer=dict( + type='AdamW', lr=0.0001, weight_decay=0.3, _scope_='mmpretrain'), + paramwise_cfg=dict( + custom_keys=dict({ + '.cls_token': dict(decay_mult=0.0), + '.pos_embed': dict(decay_mult=0.0) + })), + type='AmpOptimWrapper', + dtype='bfloat16', + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=False, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=4096) +model = dict( + type='ImageClassifier', + backbone=dict( + frozen_stages=24, + type='VisionTransformer', + arch='l', + img_size=224, + patch_size=14, + drop_rate=0.1, + pre_norm=True, + final_norm=False, + init_cfg=dict( + type='Pretrained', + checkpoint='ckpt/openclip-ViT-L-14.pth', + prefix='backbone')), + neck=dict( + type='CLIPProjection', + in_channels=1024, + out_channels=768, + init_cfg=dict( + type='Pretrained', + checkpoint='ckpt/openclip-ViT-L-14.pth', + prefix='backbone')), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=768, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + init_cfg=None), + init_cfg=dict( + type='TruncNormal', layer=['Conv2d', 'Linear'], std=0.02, bias=0.0), + train_cfg=None) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=128, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stablediffusionV2-1-dpmsolver-25-1m.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/cc1m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=128, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV2-1-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=128, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV2-1-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm_wneck' diff --git a/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm_wneck/20230604_032014/vis_data/events.out.tfevents.1685820029.SH-IDC1-10-140-24-84.232226.0 b/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm_wneck/20230604_032014/vis_data/events.out.tfevents.1685820029.SH-IDC1-10-140-24-84.232226.0 new file mode 100644 index 0000000000000000000000000000000000000000..fdc4edb09bb76ee576d566314d1126b465021c76 --- /dev/null +++ b/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm_wneck/20230604_032014/vis_data/events.out.tfevents.1685820029.SH-IDC1-10-140-24-84.232226.0 @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:6e6baff391ba7e51692162ed96d4a730fb768ad88895d288a781bfc34df7d370 +size 1706705 diff --git a/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm_wneck/20230604_032014/vis_data/scalars.json b/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm_wneck/20230604_032014/vis_data/scalars.json new file mode 100644 index 0000000000000000000000000000000000000000..69a77665db0c4734b7fad4999206e586e8e9bb77 --- /dev/null +++ b/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm_wneck/20230604_032014/vis_data/scalars.json @@ -0,0 +1,400 @@ +{"lr": 9.999857418382631e-05, "data_time": 0.002526402473449707, "loss": 0.5351661205291748, "time": 0.6351698398590088, "epoch": 1, "memory": 44138, "step": 100} +{"lr": 9.999423907359506e-05, "data_time": 0.002406454086303711, "loss": 0.4754618167877197, "time": 0.6351021766662598, "epoch": 1, "memory": 44138, "step": 200} +{"lr": 9.998699480412073e-05, "data_time": 0.001868295669555664, "loss": 0.46636803150177003, "time": 0.6340153217315674, "epoch": 1, "memory": 44138, "step": 300} +{"lr": 9.997684184379085e-05, "data_time": 0.002654623985290527, "loss": 0.4659047245979309, "time": 0.6366667032241822, "epoch": 1, "memory": 44138, "step": 400} +{"lr": 9.996378084905784e-05, "data_time": 0.001985454559326172, "loss": 0.4894475668668747, "time": 0.6355506658554078, "epoch": 1, "memory": 44138, "step": 500} +{"lr": 9.994781266439667e-05, "data_time": 0.0017483234405517578, "loss": 0.4473745822906494, "time": 0.6359376192092896, "epoch": 1, "memory": 44138, "step": 600} +{"lr": 9.992893832225061e-05, "data_time": 0.0015566825866699218, "loss": 0.4864815235137939, "time": 0.6358940839767456, "epoch": 1, "memory": 44138, "step": 700} +{"lr": 9.990715904296405e-05, "data_time": 0.0022615671157836916, "loss": 0.4405377238988876, "time": 0.6357339859008789, "epoch": 1, "memory": 44138, "step": 800} +{"lr": 9.988247623470371e-05, "data_time": 0.0015571832656860352, "loss": 0.46169241070747374, "time": 0.6358372211456299, "epoch": 1, "memory": 44138, "step": 900} +{"lr": 9.985489149336749e-05, "data_time": 0.002009463310241699, "loss": 0.44867429733276365, "time": 0.6360357284545899, "epoch": 1, "memory": 44138, "step": 1000} +{"lr": 9.982440660248187e-05, "data_time": 0.0020551443099975585, "loss": 0.4159615635871887, "time": 0.6361851215362548, "epoch": 1, "memory": 44138, "step": 1100} +{"lr": 9.979102353308574e-05, "data_time": 0.001760244369506836, "loss": 0.4278793901205063, "time": 0.6355719804763794, "epoch": 1, "memory": 44138, "step": 1200} +{"lr": 9.975474444360358e-05, "data_time": 0.001583242416381836, "loss": 0.4354752540588379, "time": 0.6347107648849487, "epoch": 1, "memory": 44138, "step": 1300} +{"lr": 9.97155716797056e-05, "data_time": 0.0016881227493286133, "loss": 0.4457062125205994, "time": 0.6341993570327759, "epoch": 1, "memory": 44138, "step": 1400} +{"lr": 9.967350777415632e-05, "data_time": 0.0017208337783813476, "loss": 0.4000787138938904, "time": 0.6360926628112793, "epoch": 1, "memory": 44138, "step": 1500} +{"lr": 9.962855544665059e-05, "data_time": 0.0015161514282226562, "loss": 0.4464715600013733, "time": 0.6359011888504028, "epoch": 1, "memory": 44138, "step": 1600} +{"lr": 9.958071760363773e-05, "data_time": 0.0013989925384521485, "loss": 0.41033613979816436, "time": 0.6355968713760376, "epoch": 1, "memory": 44138, "step": 1700} +{"lr": 9.952999733813382e-05, "data_time": 0.0017291545867919923, "loss": 0.43542540073394775, "time": 0.6360478878021241, "epoch": 1, "memory": 44138, "step": 1800} +{"lr": 9.947639792952148e-05, "data_time": 0.0017313241958618164, "loss": 0.42447437047958375, "time": 0.6356987237930298, "epoch": 1, "memory": 44138, "step": 1900} +{"lr": 9.941992284333813e-05, "data_time": 0.002037668228149414, "loss": 0.4312774628400803, "time": 0.6343778133392334, "epoch": 1, "memory": 44138, "step": 2000} +{"lr": 9.93605757310516e-05, "data_time": 0.0017091035842895508, "loss": 0.47163391709327696, "time": 0.6356060743331909, "epoch": 1, "memory": 44138, "step": 2100} +{"lr": 9.929836042982421e-05, "data_time": 0.002074432373046875, "loss": 0.4073728650808334, "time": 0.6354227781295776, "epoch": 1, "memory": 44138, "step": 2200} +{"lr": 9.923328096226467e-05, "data_time": 0.0017566919326782227, "loss": 0.3898966759443283, "time": 0.6362570762634278, "epoch": 1, "memory": 44138, "step": 2300} +{"lr": 9.916534153616819e-05, "data_time": 0.0023668289184570314, "loss": 0.39612277746200564, "time": 0.6370923280715942, "epoch": 1, "memory": 44138, "step": 2400} +{"lr": 9.909454654424387e-05, "data_time": 0.002035212516784668, "loss": 0.37696029245853424, "time": 0.6472670793533325, "epoch": 1, "memory": 44138, "step": 2500} +{"lr": 9.902090056383109e-05, "data_time": 0.0020287275314331056, "loss": 0.41623030602931976, "time": 0.6379285335540772, "epoch": 1, "memory": 44138, "step": 2600} +{"lr": 9.894440835660365e-05, "data_time": 0.0016750574111938476, "loss": 0.4208056092262268, "time": 0.6368530750274658, "epoch": 1, "memory": 44138, "step": 2700} +{"lr": 9.886507486826162e-05, "data_time": 0.0017534017562866211, "loss": 0.4455771863460541, "time": 0.6362451314926147, "epoch": 1, "memory": 44138, "step": 2800} +{"lr": 9.878290522821154e-05, "data_time": 0.0016971349716186524, "loss": 0.4209083318710327, "time": 0.6355997323989868, "epoch": 1, "memory": 44138, "step": 2900} +{"lr": 9.86979047492352e-05, "data_time": 0.00159454345703125, "loss": 0.41540210843086245, "time": 0.6349119901657104, "epoch": 1, "memory": 44138, "step": 3000} +{"lr": 9.861007892714578e-05, "data_time": 0.0026523828506469726, "loss": 0.4219654887914658, "time": 0.6351418972015381, "epoch": 1, "memory": 44138, "step": 3100} +{"lr": 9.851943344043239e-05, "data_time": 0.002142000198364258, "loss": 0.40397669672966, "time": 0.6354796171188355, "epoch": 1, "memory": 44138, "step": 3200} +{"lr": 9.842597414989319e-05, "data_time": 0.0019694805145263673, "loss": 0.3926267474889755, "time": 0.6436527490615844, "epoch": 1, "memory": 44138, "step": 3300} +{"lr": 9.832970709825637e-05, "data_time": 0.0015901088714599609, "loss": 0.38551358580589296, "time": 0.6360019445419312, "epoch": 1, "memory": 44138, "step": 3400} +{"lr": 9.823063850978968e-05, "data_time": 0.0016624927520751953, "loss": 0.40690678358078003, "time": 0.6358567953109742, "epoch": 1, "memory": 44138, "step": 3500} +{"lr": 9.81287747898975e-05, "data_time": 0.001653742790222168, "loss": 0.39099705815315244, "time": 0.6357744693756103, "epoch": 1, "memory": 44138, "step": 3600} +{"lr": 9.802412252470715e-05, "data_time": 0.0016423702239990235, "loss": 0.41955919861793517, "time": 0.6446315288543701, "epoch": 1, "memory": 44138, "step": 3700} +{"lr": 9.791668848064258e-05, "data_time": 0.0016124248504638672, "loss": 0.3832931399345398, "time": 0.6369137763977051, "epoch": 1, "memory": 44138, "step": 3800} +{"lr": 9.780647960398751e-05, "data_time": 0.0012752056121826173, "loss": 0.3861609011888504, "time": 0.6326295614242554, "epoch": 1, "memory": 44138, "step": 3900} +{"accuracy/top1": 82.55802154541016, "data_time": 0.04241977658188134, "time": 1.3461164340638279, "step": 1} +{"lr": 9.7685491196646e-05, "data_time": 0.0019457101821899413, "loss": 0.410807603597641, "time": 0.6350873231887817, "epoch": 2, "memory": 44138, "step": 4007} +{"lr": 9.756956126085676e-05, "data_time": 0.0020830631256103516, "loss": 0.3643091797828674, "time": 0.6369590520858764, "epoch": 2, "memory": 44138, "step": 4107} +{"lr": 9.745087893642482e-05, "data_time": 0.0017199039459228516, "loss": 0.4226975947618484, "time": 0.6364662170410156, "epoch": 2, "memory": 44138, "step": 4207} +{"lr": 9.73294518969056e-05, "data_time": 0.001866459846496582, "loss": 0.3846674203872681, "time": 0.6369388818740844, "epoch": 2, "memory": 44138, "step": 4307} +{"lr": 9.72052879933172e-05, "data_time": 0.0020048141479492186, "loss": 0.39270468056201935, "time": 0.6366772651672363, "epoch": 2, "memory": 44138, "step": 4407} +{"lr": 9.707839525363358e-05, "data_time": 0.0017948627471923828, "loss": 0.4163460284471512, "time": 0.636162543296814, "epoch": 2, "memory": 44138, "step": 4507} +{"lr": 9.69487818822646e-05, "data_time": 0.0018607378005981445, "loss": 0.38223142027854917, "time": 0.636462950706482, "epoch": 2, "memory": 44138, "step": 4607} +{"lr": 9.681645625952616e-05, "data_time": 0.0014985322952270508, "loss": 0.4024761557579041, "time": 0.6363603353500367, "epoch": 2, "memory": 44138, "step": 4707} +{"lr": 9.668142694109826e-05, "data_time": 0.0013245582580566407, "loss": 0.3801256835460663, "time": 0.6361339092254639, "epoch": 2, "memory": 44138, "step": 4807} +{"lr": 9.654370265747162e-05, "data_time": 0.0013548612594604492, "loss": 0.384027886390686, "time": 0.6361263751983642, "epoch": 2, "memory": 44138, "step": 4907} +{"lr": 9.640329231338345e-05, "data_time": 0.0018148422241210938, "loss": 0.39843118488788604, "time": 0.6359064817428589, "epoch": 2, "memory": 44138, "step": 5007} +{"lr": 9.626020498724147e-05, "data_time": 0.0017451047897338867, "loss": 0.381126669049263, "time": 0.6361303567886353, "epoch": 2, "memory": 44138, "step": 5107} +{"lr": 9.611444993053717e-05, "data_time": 0.0014071464538574219, "loss": 0.3815277427434921, "time": 0.6362509727478027, "epoch": 2, "memory": 44138, "step": 5207} +{"lr": 9.596603656724746e-05, "data_time": 0.0015529394149780273, "loss": 0.3765850216150284, "time": 0.636037039756775, "epoch": 2, "memory": 44138, "step": 5307} +{"lr": 9.58149744932254e-05, "data_time": 0.0018204450607299805, "loss": 0.3873521864414215, "time": 0.6353329658508301, "epoch": 2, "memory": 44138, "step": 5407} +{"lr": 9.566127347557972e-05, "data_time": 0.0022878408432006835, "loss": 0.4038151651620865, "time": 0.6346449375152587, "epoch": 2, "memory": 44138, "step": 5507} +{"lr": 9.550494345204367e-05, "data_time": 0.001725625991821289, "loss": 0.3884665071964264, "time": 0.6349648714065552, "epoch": 2, "memory": 44138, "step": 5607} +{"lr": 9.534599453033183e-05, "data_time": 0.0020513534545898438, "loss": 0.40863217115402223, "time": 0.6366077184677124, "epoch": 2, "memory": 44138, "step": 5707} +{"lr": 9.518443698748712e-05, "data_time": 0.001436305046081543, "loss": 0.37959822118282316, "time": 0.6453171491622924, "epoch": 2, "memory": 44138, "step": 5807} +{"lr": 9.50202812692162e-05, "data_time": 0.0013851642608642579, "loss": 0.3962046951055527, "time": 0.6358612060546875, "epoch": 2, "memory": 44138, "step": 5907} +{"lr": 9.485353798921386e-05, "data_time": 0.0025690078735351564, "loss": 0.3757527589797974, "time": 0.6363880395889282, "epoch": 2, "memory": 44138, "step": 6007} +{"lr": 9.468421792847712e-05, "data_time": 0.0020028829574584963, "loss": 0.40307600796222687, "time": 0.6365566492080689, "epoch": 2, "memory": 44138, "step": 6107} +{"lr": 9.451233203460794e-05, "data_time": 0.001594209671020508, "loss": 0.39338279366493223, "time": 0.6365896940231324, "epoch": 2, "memory": 44138, "step": 6207} +{"lr": 9.433789142110532e-05, "data_time": 0.0016883611679077148, "loss": 0.3865435302257538, "time": 0.6363863706588745, "epoch": 2, "memory": 44138, "step": 6307} +{"lr": 9.416090736664709e-05, "data_time": 0.0013787508010864257, "loss": 0.4145511955022812, "time": 0.6368163585662842, "epoch": 2, "memory": 44138, "step": 6407} +{"lr": 9.398139131436032e-05, "data_time": 0.0017983198165893554, "loss": 0.39429461061954496, "time": 0.6360846519470215, "epoch": 2, "memory": 44138, "step": 6507} +{"lr": 9.379935487108167e-05, "data_time": 0.00151369571685791, "loss": 0.39995901584625243, "time": 0.6380569219589234, "epoch": 2, "memory": 44138, "step": 6607} +{"lr": 9.361480980660697e-05, "data_time": 0.0015455961227416992, "loss": 0.39948015511035917, "time": 0.6372979164123536, "epoch": 2, "memory": 44138, "step": 6707} +{"lr": 9.342776805292993e-05, "data_time": 0.001640486717224121, "loss": 0.36709106862545016, "time": 0.6505163431167602, "epoch": 2, "memory": 44138, "step": 6807} +{"lr": 9.323824170347083e-05, "data_time": 0.0016528844833374023, "loss": 0.38471029996871947, "time": 0.6369179964065552, "epoch": 2, "memory": 44138, "step": 6907} +{"lr": 9.304624301229471e-05, "data_time": 0.0020319461822509766, "loss": 0.39614465534687043, "time": 0.6368730783462524, "epoch": 2, "memory": 44138, "step": 7007} +{"lr": 9.285178439331897e-05, "data_time": 0.0014183759689331056, "loss": 0.35661196112632754, "time": 0.6375965356826783, "epoch": 2, "memory": 44138, "step": 7107} +{"lr": 9.265487841951087e-05, "data_time": 0.0016214609146118163, "loss": 0.37881762683391573, "time": 0.6370727777481079, "epoch": 2, "memory": 44138, "step": 7207} +{"lr": 9.245553782207421e-05, "data_time": 0.0016767740249633788, "loss": 0.3776803106069565, "time": 0.6357251405715942, "epoch": 2, "memory": 44138, "step": 7307} +{"lr": 9.225377548962671e-05, "data_time": 0.0016555070877075195, "loss": 0.37052683234214784, "time": 0.6352493762969971, "epoch": 2, "memory": 44138, "step": 7407} +{"lr": 9.204960446736601e-05, "data_time": 0.0019618749618530275, "loss": 0.40267767906188967, "time": 0.6363161087036133, "epoch": 2, "memory": 44138, "step": 7507} +{"lr": 9.184303795622715e-05, "data_time": 0.001558542251586914, "loss": 0.366828715801239, "time": 0.6356423377990723, "epoch": 2, "memory": 44138, "step": 7607} +{"lr": 9.163408931202823e-05, "data_time": 0.001344776153564453, "loss": 0.38677774369716644, "time": 0.6362397909164429, "epoch": 2, "memory": 44138, "step": 7707} +{"lr": 9.142277204460729e-05, "data_time": 0.0015615463256835938, "loss": 0.40312814712524414, "time": 0.6365349292755127, "epoch": 2, "memory": 44138, "step": 7807} +{"accuracy/top1": 82.29706573486328, "data_time": 0.0335041777840976, "time": 1.2768701479352753, "step": 2} +{"lr": 9.119405492383288e-05, "data_time": 0.0019070625305175782, "loss": 0.3771364837884903, "time": 0.6362051010131836, "epoch": 3, "memory": 44138, "step": 7914} +{"lr": 9.097787819233205e-05, "data_time": 0.0018374204635620117, "loss": 0.4004779279232025, "time": 0.6369052171707154, "epoch": 3, "memory": 44138, "step": 8014} +{"lr": 9.075937526576627e-05, "data_time": 0.0018862247467041015, "loss": 0.3887979418039322, "time": 0.6360315799713134, "epoch": 3, "memory": 44138, "step": 8114} +{"lr": 9.053856027171755e-05, "data_time": 0.0015074491500854492, "loss": 0.35386045277118683, "time": 0.6370097875595093, "epoch": 3, "memory": 44138, "step": 8214} +{"lr": 9.031544748725785e-05, "data_time": 0.0018077135086059571, "loss": 0.405665984749794, "time": 0.6369300842285156, "epoch": 3, "memory": 44138, "step": 8314} +{"lr": 9.009005133802542e-05, "data_time": 0.0015701055526733398, "loss": 0.3652287185192108, "time": 0.6358760118484497, "epoch": 3, "memory": 44138, "step": 8414} +{"lr": 8.986238639729242e-05, "data_time": 0.0016776561737060548, "loss": 0.38257692754268646, "time": 0.6368263959884644, "epoch": 3, "memory": 44138, "step": 8514} +{"lr": 8.963246738502254e-05, "data_time": 0.0014507532119750976, "loss": 0.3653694659471512, "time": 0.6368743658065796, "epoch": 3, "memory": 44138, "step": 8614} +{"lr": 8.940030916691928e-05, "data_time": 0.0016489982604980468, "loss": 0.37748109400272367, "time": 0.6365738630294799, "epoch": 3, "memory": 44138, "step": 8714} +{"lr": 8.91659267534649e-05, "data_time": 0.001427769660949707, "loss": 0.36973799765110016, "time": 0.6358951330184937, "epoch": 3, "memory": 44138, "step": 8814} +{"lr": 8.892933529894978e-05, "data_time": 0.0019328594207763672, "loss": 0.3578315645456314, "time": 0.6362555027008057, "epoch": 3, "memory": 44138, "step": 8914} +{"lr": 8.869055010049283e-05, "data_time": 0.0017607450485229493, "loss": 0.37117459774017336, "time": 0.6339348793029785, "epoch": 3, "memory": 44138, "step": 9014} +{"lr": 8.844958659705188e-05, "data_time": 0.001490616798400879, "loss": 0.3593450516462326, "time": 0.633604621887207, "epoch": 3, "memory": 44138, "step": 9114} +{"lr": 8.82064603684259e-05, "data_time": 0.002527451515197754, "loss": 0.35577426850795746, "time": 0.6364502429962158, "epoch": 3, "memory": 44138, "step": 9214} +{"lr": 8.796118713424776e-05, "data_time": 0.0017650842666625977, "loss": 0.3818786978721619, "time": 0.6364527463912963, "epoch": 3, "memory": 44138, "step": 9314} +{"lr": 8.77137827529676e-05, "data_time": 0.002043271064758301, "loss": 0.4080860525369644, "time": 0.6367826223373413, "epoch": 3, "memory": 44138, "step": 9414} +{"lr": 8.746426322082745e-05, "data_time": 0.0014680862426757813, "loss": 0.35324149429798124, "time": 0.6367222785949707, "epoch": 3, "memory": 44138, "step": 9514} +{"lr": 8.721264467082732e-05, "data_time": 0.0017992019653320312, "loss": 0.3661399602890015, "time": 0.6359201908111572, "epoch": 3, "memory": 44138, "step": 9614} +{"lr": 8.695894337168168e-05, "data_time": 0.0014480113983154296, "loss": 0.35871289670467377, "time": 0.6363116264343261, "epoch": 3, "memory": 44138, "step": 9714} +{"lr": 8.67031757267677e-05, "data_time": 0.001970672607421875, "loss": 0.3939231425523758, "time": 0.6369412183761597, "epoch": 3, "memory": 44138, "step": 9814} +{"lr": 8.64453582730651e-05, "data_time": 0.0014687061309814453, "loss": 0.3631616085767746, "time": 0.6370649099349975, "epoch": 3, "memory": 44138, "step": 9914} +{"lr": 8.61855076800861e-05, "data_time": 0.001672077178955078, "loss": 0.3913224399089813, "time": 0.636255145072937, "epoch": 3, "memory": 44138, "step": 10014} +{"lr": 8.592364074879843e-05, "data_time": 0.001474308967590332, "loss": 0.379348886013031, "time": 0.6359834432601928, "epoch": 3, "memory": 44138, "step": 10114} +{"lr": 8.565977441053854e-05, "data_time": 0.001491856575012207, "loss": 0.3770377814769745, "time": 0.6376715183258057, "epoch": 3, "memory": 44138, "step": 10214} +{"lr": 8.539392572591724e-05, "data_time": 0.0015501976013183594, "loss": 0.37069645822048186, "time": 0.6368112325668335, "epoch": 3, "memory": 44138, "step": 10314} +{"lr": 8.512611188371633e-05, "data_time": 0.0017702817916870118, "loss": 0.37677340507507323, "time": 0.6366872787475586, "epoch": 3, "memory": 44138, "step": 10414} +{"lr": 8.485635019977725e-05, "data_time": 0.0015655279159545899, "loss": 0.33343938291072844, "time": 0.6366911172866822, "epoch": 3, "memory": 44138, "step": 10514} +{"lr": 8.45846581158818e-05, "data_time": 0.00154876708984375, "loss": 0.37909236550331116, "time": 0.6365845680236817, "epoch": 3, "memory": 44138, "step": 10614} +{"lr": 8.431105319862416e-05, "data_time": 0.0016137123107910155, "loss": 0.3890629827976227, "time": 0.6348929643630982, "epoch": 3, "memory": 44138, "step": 10714} +{"lr": 8.403555313827516e-05, "data_time": 0.0019269943237304687, "loss": 0.370540389418602, "time": 0.6409435033798218, "epoch": 3, "memory": 44138, "step": 10814} +{"lr": 8.375817574763844e-05, "data_time": 0.002280569076538086, "loss": 0.3622734367847443, "time": 0.63719801902771, "epoch": 3, "memory": 44138, "step": 10914} +{"lr": 8.347893896089871e-05, "data_time": 0.0016435146331787109, "loss": 0.3717945545911789, "time": 0.6367708683013916, "epoch": 3, "memory": 44138, "step": 11014} +{"lr": 8.31978608324625e-05, "data_time": 0.0013298988342285156, "loss": 0.39679092466831206, "time": 0.6368518590927124, "epoch": 3, "memory": 44138, "step": 11114} +{"lr": 8.291495953579052e-05, "data_time": 0.0019569873809814455, "loss": 0.3572738438844681, "time": 0.6373294115066528, "epoch": 3, "memory": 44138, "step": 11214} +{"lr": 8.263025336222267e-05, "data_time": 0.0018393278121948242, "loss": 0.3617665946483612, "time": 0.6371901750564575, "epoch": 3, "memory": 44138, "step": 11314} +{"lr": 8.234376071979551e-05, "data_time": 0.0015526056289672852, "loss": 0.384552127122879, "time": 0.635496187210083, "epoch": 3, "memory": 44138, "step": 11414} +{"lr": 8.205550013205187e-05, "data_time": 0.0018289804458618164, "loss": 0.3666399359703064, "time": 0.6368457794189453, "epoch": 3, "memory": 44138, "step": 11514} +{"lr": 8.17654902368435e-05, "data_time": 0.0017534017562866211, "loss": 0.3668083220720291, "time": 0.6360332727432251, "epoch": 3, "memory": 44138, "step": 11614} +{"lr": 8.147374978512582e-05, "data_time": 0.001693129539489746, "loss": 0.38223120868206023, "time": 0.6358201026916503, "epoch": 3, "memory": 44138, "step": 11714} +{"accuracy/top1": 80.41474151611328, "data_time": 0.031422812363197064, "time": 1.2761071797074943, "step": 3} +{"lr": 8.115969237626578e-05, "data_time": 0.0022393226623535155, "loss": 0.4031961888074875, "time": 0.6378740072250366, "epoch": 4, "memory": 44138, "step": 11821} +{"lr": 8.08644297337245e-05, "data_time": 0.001559877395629883, "loss": 0.3554630070924759, "time": 0.637148666381836, "epoch": 4, "memory": 44138, "step": 11921} +{"lr": 8.056749479387651e-05, "data_time": 0.0014817476272583007, "loss": 0.3698801577091217, "time": 0.6369105100631713, "epoch": 4, "memory": 44138, "step": 12021} +{"lr": 8.026890675542457e-05, "data_time": 0.0017343282699584961, "loss": 0.3652118921279907, "time": 0.6353127241134644, "epoch": 4, "memory": 44138, "step": 12121} +{"lr": 7.996868492395457e-05, "data_time": 0.0019557952880859377, "loss": 0.3849740356206894, "time": 0.6371478319168091, "epoch": 4, "memory": 44138, "step": 12221} +{"lr": 7.966684871068762e-05, "data_time": 0.002064371109008789, "loss": 0.38858999609947203, "time": 0.6365043163299561, "epoch": 4, "memory": 44138, "step": 12321} +{"lr": 7.936341763122449e-05, "data_time": 0.0015283584594726562, "loss": 0.36215423941612246, "time": 0.6373162031173706, "epoch": 4, "memory": 44138, "step": 12421} +{"lr": 7.905841130428405e-05, "data_time": 0.002115774154663086, "loss": 0.36562091410160064, "time": 0.6364990234375, "epoch": 4, "memory": 44138, "step": 12521} +{"lr": 7.875184945043487e-05, "data_time": 0.0022097349166870115, "loss": 0.3673384487628937, "time": 0.6367980718612671, "epoch": 4, "memory": 44138, "step": 12621} +{"lr": 7.844375189082012e-05, "data_time": 0.0020020008087158203, "loss": 0.3457824647426605, "time": 0.6370929718017578, "epoch": 4, "memory": 44138, "step": 12721} +{"lr": 7.813413854587572e-05, "data_time": 0.001546025276184082, "loss": 0.37467997074127196, "time": 0.6368714332580566, "epoch": 4, "memory": 44138, "step": 12821} +{"lr": 7.782302943404291e-05, "data_time": 0.0023020505905151367, "loss": 0.3861950010061264, "time": 0.6367208242416382, "epoch": 4, "memory": 44138, "step": 12921} +{"lr": 7.751044467047337e-05, "data_time": 0.0013361215591430665, "loss": 0.35624179244041443, "time": 0.636503028869629, "epoch": 4, "memory": 44138, "step": 13021} +{"lr": 7.719640446572907e-05, "data_time": 0.0017018795013427734, "loss": 0.3679825782775879, "time": 0.6353821277618408, "epoch": 4, "memory": 44138, "step": 13121} +{"lr": 7.688092912447536e-05, "data_time": 0.0016974449157714845, "loss": 0.3739969491958618, "time": 0.6364741563796997, "epoch": 4, "memory": 44138, "step": 13221} +{"lr": 7.656403904416794e-05, "data_time": 0.0017170906066894531, "loss": 0.3584659039974213, "time": 0.6369251251220703, "epoch": 4, "memory": 44138, "step": 13321} +{"lr": 7.624575471373434e-05, "data_time": 0.0019341468811035155, "loss": 0.35059358179569244, "time": 0.6363126516342164, "epoch": 4, "memory": 44138, "step": 13421} +{"lr": 7.592609671224926e-05, "data_time": 0.0014975547790527343, "loss": 0.35459600687026976, "time": 0.6369526147842407, "epoch": 4, "memory": 44138, "step": 13521} +{"lr": 7.560508570760375e-05, "data_time": 0.0014723777770996095, "loss": 0.37329457998275756, "time": 0.6371474027633667, "epoch": 4, "memory": 44138, "step": 13621} +{"lr": 7.528274245516895e-05, "data_time": 0.0015514373779296875, "loss": 0.3840697079896927, "time": 0.637981629371643, "epoch": 4, "memory": 44138, "step": 13721} +{"lr": 7.495908779645405e-05, "data_time": 0.0018193721771240234, "loss": 0.35149831771850587, "time": 0.6373069286346436, "epoch": 4, "memory": 44138, "step": 13821} +{"lr": 7.463414265775906e-05, "data_time": 0.0016061067581176758, "loss": 0.3762999713420868, "time": 0.6366349697113037, "epoch": 4, "memory": 44138, "step": 13921} +{"lr": 7.430792804882147e-05, "data_time": 0.002154850959777832, "loss": 0.37663392424583436, "time": 0.6373780965805054, "epoch": 4, "memory": 44138, "step": 14021} +{"lr": 7.398046506145803e-05, "data_time": 0.0017863035202026368, "loss": 0.40180477499961853, "time": 0.6369316577911377, "epoch": 4, "memory": 44138, "step": 14121} +{"lr": 7.365177486820114e-05, "data_time": 0.0015575885772705078, "loss": 0.39575632512569425, "time": 0.6372013330459595, "epoch": 4, "memory": 44138, "step": 14221} +{"lr": 7.332187872092939e-05, "data_time": 0.0016432523727416993, "loss": 0.35980776846408846, "time": 0.6360937118530273, "epoch": 4, "memory": 44138, "step": 14321} +{"lr": 7.299079794949401e-05, "data_time": 0.0016222715377807618, "loss": 0.3617538034915924, "time": 0.6359572410583496, "epoch": 4, "memory": 44138, "step": 14421} +{"lr": 7.265855396033955e-05, "data_time": 0.0017271757125854493, "loss": 0.3715987801551819, "time": 0.6366678237915039, "epoch": 4, "memory": 44138, "step": 14521} +{"lr": 7.232516823512013e-05, "data_time": 0.0015793561935424805, "loss": 0.35890172123909, "time": 0.636816430091858, "epoch": 4, "memory": 44138, "step": 14621} +{"lr": 7.199066232930995e-05, "data_time": 0.0012960195541381835, "loss": 0.35270917415618896, "time": 0.6368032455444336, "epoch": 4, "memory": 44138, "step": 14721} +{"lr": 7.165505787080997e-05, "data_time": 0.002298140525817871, "loss": 0.3418466538190842, "time": 0.6367158412933349, "epoch": 4, "memory": 44138, "step": 14821} +{"lr": 7.13183765585495e-05, "data_time": 0.0015070915222167968, "loss": 0.3678338259458542, "time": 0.6368214845657348, "epoch": 4, "memory": 44138, "step": 14921} +{"lr": 7.098064016108317e-05, "data_time": 0.0017307519912719727, "loss": 0.37935556173324586, "time": 0.6376452684402466, "epoch": 4, "memory": 44138, "step": 15021} +{"lr": 7.064187051518379e-05, "data_time": 0.002585959434509277, "loss": 0.36913132220506667, "time": 0.6390216827392579, "epoch": 4, "memory": 44138, "step": 15121} +{"lr": 7.030208952442949e-05, "data_time": 0.0018384456634521484, "loss": 0.3649781197309494, "time": 0.636927318572998, "epoch": 4, "memory": 44138, "step": 15221} +{"lr": 6.996131915778851e-05, "data_time": 0.0016496658325195312, "loss": 0.3643661648035049, "time": 0.6371755838394165, "epoch": 4, "memory": 44138, "step": 15321} +{"lr": 6.961958144819853e-05, "data_time": 0.0016845464706420898, "loss": 0.3943395346403122, "time": 0.6380192518234253, "epoch": 4, "memory": 44138, "step": 15421} +{"lr": 6.927689849114193e-05, "data_time": 0.0017381429672241211, "loss": 0.37081224024295806, "time": 0.6376275777816772, "epoch": 4, "memory": 44138, "step": 15521} +{"lr": 6.893329244321722e-05, "data_time": 0.0013135433197021484, "loss": 0.3691263049840927, "time": 0.6336902618408203, "epoch": 4, "memory": 44138, "step": 15621} +{"accuracy/top1": 78.90470123291016, "data_time": 0.0329343450480494, "time": 1.2753868966267026, "step": 4} +{"lr": 6.856463687361792e-05, "data_time": 0.001588153839111328, "loss": 0.35422569513320923, "time": 0.6365503549575806, "epoch": 5, "memory": 44138, "step": 15728} +{"lr": 6.821919068470432e-05, "data_time": 0.0015651941299438476, "loss": 0.37694051861763, "time": 0.6365997314453125, "epoch": 5, "memory": 44138, "step": 15828} +{"lr": 6.787288979235407e-05, "data_time": 0.001617741584777832, "loss": 0.3566369891166687, "time": 0.636507534980774, "epoch": 5, "memory": 44138, "step": 15928} +{"lr": 6.752575658708793e-05, "data_time": 0.001557302474975586, "loss": 0.3773763120174408, "time": 0.636857271194458, "epoch": 5, "memory": 44138, "step": 16028} +{"lr": 6.71778135132408e-05, "data_time": 0.0014654874801635742, "loss": 0.355039456486702, "time": 0.6353830337524414, "epoch": 5, "memory": 44138, "step": 16128} +{"lr": 6.682908306751074e-05, "data_time": 0.0022002696990966798, "loss": 0.3726886361837387, "time": 0.6362100601196289, "epoch": 5, "memory": 44138, "step": 16228} +{"lr": 6.64795877975043e-05, "data_time": 0.001607990264892578, "loss": 0.3671862095594406, "time": 0.6362984657287598, "epoch": 5, "memory": 44138, "step": 16328} +{"lr": 6.61293503002787e-05, "data_time": 0.0015233755111694336, "loss": 0.36181028485298156, "time": 0.6364665985107422, "epoch": 5, "memory": 44138, "step": 16428} +{"lr": 6.577839322088084e-05, "data_time": 0.001546788215637207, "loss": 0.3405226916074753, "time": 0.6361546277999878, "epoch": 5, "memory": 44138, "step": 16528} +{"lr": 6.542673925088298e-05, "data_time": 0.0020236968994140625, "loss": 0.3623759776353836, "time": 0.6366914749145508, "epoch": 5, "memory": 44138, "step": 16628} +{"lr": 6.507441112691584e-05, "data_time": 0.001490473747253418, "loss": 0.3516934782266617, "time": 0.6364949226379395, "epoch": 5, "memory": 44138, "step": 16728} +{"lr": 6.47214316291983e-05, "data_time": 0.0015769243240356446, "loss": 0.35145682394504546, "time": 0.6363529205322266, "epoch": 5, "memory": 44138, "step": 16828} +{"lr": 6.436782358006481e-05, "data_time": 0.0022809505462646484, "loss": 0.3611027032136917, "time": 0.637720251083374, "epoch": 5, "memory": 44138, "step": 16928} +{"lr": 6.40136098424896e-05, "data_time": 0.001624298095703125, "loss": 0.37126933336257933, "time": 0.6365201473236084, "epoch": 5, "memory": 44138, "step": 17028} +{"lr": 6.365881331860828e-05, "data_time": 0.0018805503845214845, "loss": 0.3826365262269974, "time": 0.6374353647232056, "epoch": 5, "memory": 44138, "step": 17128} +{"lr": 6.330345694823739e-05, "data_time": 0.0016014814376831056, "loss": 0.36949982643127444, "time": 0.6373626708984375, "epoch": 5, "memory": 44138, "step": 17228} +{"lr": 6.29475637073908e-05, "data_time": 0.0023261308670043945, "loss": 0.36648235023021697, "time": 0.6376047611236573, "epoch": 5, "memory": 44138, "step": 17328} +{"lr": 6.259115660679482e-05, "data_time": 0.0015662908554077148, "loss": 0.36379197239875793, "time": 0.6365245103836059, "epoch": 5, "memory": 44138, "step": 17428} +{"lr": 6.223425869039963e-05, "data_time": 0.0014850616455078125, "loss": 0.37681972086429594, "time": 0.6362675666809082, "epoch": 5, "memory": 44138, "step": 17528} +{"lr": 6.187689303389006e-05, "data_time": 0.0016463279724121093, "loss": 0.36791214942932127, "time": 0.6367289543151855, "epoch": 5, "memory": 44138, "step": 17628} +{"lr": 6.151908274319292e-05, "data_time": 0.001893019676208496, "loss": 0.3668570637702942, "time": 0.6370482683181763, "epoch": 5, "memory": 44138, "step": 17728} +{"lr": 6.116085095298373e-05, "data_time": 0.0017044544219970703, "loss": 0.3675206959247589, "time": 0.6368045806884766, "epoch": 5, "memory": 44138, "step": 17828} +{"lr": 6.080222082519007e-05, "data_time": 0.001479792594909668, "loss": 0.34049825072288514, "time": 0.635310697555542, "epoch": 5, "memory": 44138, "step": 17928} +{"lr": 6.0443215547495147e-05, "data_time": 0.0021247386932373045, "loss": 0.39336840212345126, "time": 0.6370204210281372, "epoch": 5, "memory": 44138, "step": 18028} +{"lr": 6.0083858331837714e-05, "data_time": 0.0018485307693481446, "loss": 0.36739606559276583, "time": 0.6371299743652343, "epoch": 5, "memory": 44138, "step": 18128} +{"lr": 5.97241724129116e-05, "data_time": 0.00196075439453125, "loss": 0.35261605978012084, "time": 0.636580514907837, "epoch": 5, "memory": 44138, "step": 18228} +{"lr": 5.936418104666328e-05, "data_time": 0.0015727043151855468, "loss": 0.3733451128005981, "time": 0.6366521120071411, "epoch": 5, "memory": 44138, "step": 18328} +{"lr": 5.9003907508788295e-05, "data_time": 0.0018732309341430663, "loss": 0.38800789415836334, "time": 0.6371723890304566, "epoch": 5, "memory": 44138, "step": 18428} +{"lr": 5.864337509322633e-05, "data_time": 0.0017942428588867188, "loss": 0.3905669659376144, "time": 0.6373066425323486, "epoch": 5, "memory": 44138, "step": 18528} +{"lr": 5.8282607110655194e-05, "data_time": 0.0019335269927978516, "loss": 0.39688348174095156, "time": 0.6366997241973877, "epoch": 5, "memory": 44138, "step": 18628} +{"lr": 5.7921626886983564e-05, "data_time": 0.001709747314453125, "loss": 0.3747338384389877, "time": 0.6370071411132813, "epoch": 5, "memory": 44138, "step": 18728} +{"lr": 5.756045776184287e-05, "data_time": 0.0018059015274047852, "loss": 0.36166087687015536, "time": 0.6371424436569214, "epoch": 5, "memory": 44138, "step": 18828} +{"lr": 5.719912308707808e-05, "data_time": 0.0020849943161010743, "loss": 0.37360704839229586, "time": 0.6362600088119507, "epoch": 5, "memory": 44138, "step": 18928} +{"lr": 5.683764622523808e-05, "data_time": 0.0020946502685546876, "loss": 0.3509641170501709, "time": 0.6386455297470093, "epoch": 5, "memory": 44138, "step": 19028} +{"lr": 5.647605054806502e-05, "data_time": 0.0016429424285888672, "loss": 0.3507239788770676, "time": 0.6361999273300171, "epoch": 5, "memory": 44138, "step": 19128} +{"lr": 5.6114359434983206e-05, "data_time": 0.0015897750854492188, "loss": 0.383701291680336, "time": 0.6368448257446289, "epoch": 5, "memory": 44138, "step": 19228} +{"lr": 5.575259627158745e-05, "data_time": 0.0018629074096679688, "loss": 0.40328297317028045, "time": 0.636943769454956, "epoch": 5, "memory": 44138, "step": 19328} +{"lr": 5.539078444813113e-05, "data_time": 0.0015216350555419921, "loss": 0.3667666703462601, "time": 0.6371160507202148, "epoch": 5, "memory": 44138, "step": 19428} +{"lr": 5.5028947358013735e-05, "data_time": 0.001392531394958496, "loss": 0.34482610523700713, "time": 0.6352652549743653, "epoch": 5, "memory": 44138, "step": 19528} +{"accuracy/top1": 78.03138732910156, "data_time": 0.03266500604563746, "time": 1.2760491083408225, "step": 5} +{"lr": 5.464178020340108e-05, "data_time": 0.0024876117706298826, "loss": 0.37087360620498655, "time": 0.6373377561569213, "epoch": 6, "memory": 44138, "step": 19635} +{"lr": 5.427996514795046e-05, "data_time": 0.0021753549575805665, "loss": 0.37204011976718904, "time": 0.635506558418274, "epoch": 6, "memory": 44138, "step": 19735} +{"lr": 5.3918196647260834e-05, "data_time": 0.001603412628173828, "loss": 0.35330307185649873, "time": 0.6365888118743896, "epoch": 6, "memory": 44138, "step": 19835} +{"lr": 5.355649809193057e-05, "data_time": 0.001467275619506836, "loss": 0.37361868321895597, "time": 0.6379712581634521, "epoch": 6, "memory": 44138, "step": 19935} +{"lr": 5.3194892868035765e-05, "data_time": 0.001579117774963379, "loss": 0.36632903218269347, "time": 0.636756420135498, "epoch": 6, "memory": 44138, "step": 20035} +{"lr": 5.283340435561795e-05, "data_time": 0.0015182256698608398, "loss": 0.3887178868055344, "time": 0.6369266510009766, "epoch": 6, "memory": 44138, "step": 20135} +{"lr": 5.2472055927172495e-05, "data_time": 0.0014935970306396485, "loss": 0.3624693930149078, "time": 0.6368660926818848, "epoch": 6, "memory": 44138, "step": 20235} +{"lr": 5.2110870946137605e-05, "data_time": 0.0016383647918701172, "loss": 0.3593368470668793, "time": 0.6372593641281128, "epoch": 6, "memory": 44138, "step": 20335} +{"lr": 5.174987276538342e-05, "data_time": 0.0015296459197998047, "loss": 0.3751054733991623, "time": 0.6355548143386841, "epoch": 6, "memory": 44138, "step": 20435} +{"lr": 5.1389084725702425e-05, "data_time": 0.0016545772552490235, "loss": 0.3484298437833786, "time": 0.6362725734710694, "epoch": 6, "memory": 44138, "step": 20535} +{"lr": 5.1028530154300045e-05, "data_time": 0.001936793327331543, "loss": 0.35034675896167755, "time": 0.6363436222076416, "epoch": 6, "memory": 44138, "step": 20635} +{"lr": 5.066823236328663e-05, "data_time": 0.001694488525390625, "loss": 0.36730977296829226, "time": 0.6357667922973633, "epoch": 6, "memory": 44138, "step": 20735} +{"lr": 5.030821464817006e-05, "data_time": 0.0018495798110961914, "loss": 0.3620191186666489, "time": 0.6368024826049805, "epoch": 6, "memory": 44138, "step": 20835} +{"lr": 4.99485002863494e-05, "data_time": 0.0014292001724243164, "loss": 0.3624554634094238, "time": 0.6369625568389893, "epoch": 6, "memory": 44138, "step": 20935} +{"lr": 4.95891125356102e-05, "data_time": 0.001660919189453125, "loss": 0.3599185734987259, "time": 0.6478580236434937, "epoch": 6, "memory": 44138, "step": 21035} +{"lr": 4.923007463262053e-05, "data_time": 0.0020416259765625, "loss": 0.3875850856304169, "time": 0.639004135131836, "epoch": 6, "memory": 44138, "step": 21135} +{"lr": 4.887140979142852e-05, "data_time": 0.0017733812332153321, "loss": 0.35813171565532687, "time": 0.6365517377853394, "epoch": 6, "memory": 44138, "step": 21235} +{"lr": 4.851314120196161e-05, "data_time": 0.0017263650894165038, "loss": 0.3811752527952194, "time": 0.6365441083908081, "epoch": 6, "memory": 44138, "step": 21335} +{"lr": 4.815529202852693e-05, "data_time": 0.0017476558685302734, "loss": 0.3990250527858734, "time": 0.637282919883728, "epoch": 6, "memory": 44138, "step": 21435} +{"lr": 4.779788540831406e-05, "data_time": 0.001805734634399414, "loss": 0.39963223338127135, "time": 0.6367613554000855, "epoch": 6, "memory": 44138, "step": 21535} +{"lr": 4.744094444989843e-05, "data_time": 0.001782989501953125, "loss": 0.3687055140733719, "time": 0.637285566329956, "epoch": 6, "memory": 44138, "step": 21635} +{"lr": 4.708449223174764e-05, "data_time": 0.0017920732498168945, "loss": 0.35472665429115297, "time": 0.6371837377548217, "epoch": 6, "memory": 44138, "step": 21735} +{"lr": 4.672855180072916e-05, "data_time": 0.0017628908157348634, "loss": 0.3685429602861404, "time": 0.6370863676071167, "epoch": 6, "memory": 44138, "step": 21835} +{"lr": 4.6373146170620256e-05, "data_time": 0.0015166044235229493, "loss": 0.35768717229366304, "time": 0.6370740175247193, "epoch": 6, "memory": 44138, "step": 21935} +{"lr": 4.6018298320619915e-05, "data_time": 0.0015221357345581055, "loss": 0.38128423392772676, "time": 0.6368333578109742, "epoch": 6, "memory": 44138, "step": 22035} +{"lr": 4.566403119386318e-05, "data_time": 0.0014801740646362305, "loss": 0.3583536595106125, "time": 0.6373617887496948, "epoch": 6, "memory": 44138, "step": 22135} +{"lr": 4.531036769593769e-05, "data_time": 0.0016695737838745117, "loss": 0.3852864295244217, "time": 0.6379703283309937, "epoch": 6, "memory": 44138, "step": 22235} +{"lr": 4.4957330693402685e-05, "data_time": 0.0021849155426025392, "loss": 0.3721946179866791, "time": 0.636847972869873, "epoch": 6, "memory": 44138, "step": 22335} +{"lr": 4.460494301231059e-05, "data_time": 0.0015393733978271485, "loss": 0.3711391031742096, "time": 0.636501407623291, "epoch": 6, "memory": 44138, "step": 22435} +{"lr": 4.425322743673109e-05, "data_time": 0.0016908645629882812, "loss": 0.381898033618927, "time": 0.6368586540222168, "epoch": 6, "memory": 44138, "step": 22535} +{"lr": 4.390220670727812e-05, "data_time": 0.002420854568481445, "loss": 0.33969086706638335, "time": 0.637241530418396, "epoch": 6, "memory": 44138, "step": 22635} +{"lr": 4.3551903519639325e-05, "data_time": 0.0017018795013427734, "loss": 0.3634351372718811, "time": 0.6370144367218018, "epoch": 6, "memory": 44138, "step": 22735} +{"lr": 4.3202340523108894e-05, "data_time": 0.002078127861022949, "loss": 0.3723094642162323, "time": 0.6359060764312744, "epoch": 6, "memory": 44138, "step": 22835} +{"lr": 4.285354031912302e-05, "data_time": 0.0016663312911987306, "loss": 0.36899976432323456, "time": 0.6357746839523315, "epoch": 6, "memory": 44138, "step": 22935} +{"lr": 4.2505525459798544e-05, "data_time": 0.0017359733581542968, "loss": 0.38210780918598175, "time": 0.635829758644104, "epoch": 6, "memory": 44138, "step": 23035} +{"lr": 4.2158318446474836e-05, "data_time": 0.0018004417419433595, "loss": 0.3862610787153244, "time": 0.6364174604415893, "epoch": 6, "memory": 44138, "step": 23135} +{"lr": 4.181194172825905e-05, "data_time": 0.0018860816955566407, "loss": 0.36710751354694365, "time": 0.6360246896743774, "epoch": 6, "memory": 44138, "step": 23235} +{"lr": 4.146641770057448e-05, "data_time": 0.0017897605895996094, "loss": 0.36676945686340334, "time": 0.6363061666488647, "epoch": 6, "memory": 44138, "step": 23335} +{"lr": 4.112176870371267e-05, "data_time": 0.0017312526702880859, "loss": 0.3598135381937027, "time": 0.6366748332977294, "epoch": 6, "memory": 44138, "step": 23435} +{"accuracy/top1": 77.80870819091797, "data_time": 0.031339752263036265, "time": 1.2763516368537113, "step": 6} +{"lr": 4.0753988582732786e-05, "data_time": 0.00150299072265625, "loss": 0.37958675920963286, "time": 0.6361352205276489, "epoch": 7, "memory": 44138, "step": 23542} +{"lr": 4.0411221639286815e-05, "data_time": 0.0015937328338623048, "loss": 0.409321540594101, "time": 0.6369438171386719, "epoch": 7, "memory": 44138, "step": 23642} +{"lr": 4.006939795169518e-05, "data_time": 0.001965022087097168, "loss": 0.3521025270223618, "time": 0.6362314462661743, "epoch": 7, "memory": 44138, "step": 23742} +{"lr": 3.972853962099913e-05, "data_time": 0.001357722282409668, "loss": 0.33720476031303404, "time": 0.6374253034591675, "epoch": 7, "memory": 44138, "step": 23842} +{"lr": 3.93886686858237e-05, "data_time": 0.0016610383987426757, "loss": 0.39819527566432955, "time": 0.6368105173110962, "epoch": 7, "memory": 44138, "step": 23942} +{"lr": 3.904980712095259e-05, "data_time": 0.0015917778015136718, "loss": 0.3759404569864273, "time": 0.6376616477966308, "epoch": 7, "memory": 44138, "step": 24042} +{"lr": 3.871197683590748e-05, "data_time": 0.001981186866760254, "loss": 0.37075887620449066, "time": 0.6376721620559692, "epoch": 7, "memory": 44138, "step": 24142} +{"lr": 3.837519967353125e-05, "data_time": 0.0017904281616210938, "loss": 0.38417728841304777, "time": 0.6366328001022339, "epoch": 7, "memory": 44138, "step": 24242} +{"lr": 3.8039497408575855e-05, "data_time": 0.0023898839950561523, "loss": 0.3890520006418228, "time": 0.6377934455871582, "epoch": 7, "memory": 44138, "step": 24342} +{"lr": 3.770489174629432e-05, "data_time": 0.0019491195678710937, "loss": 0.38422541320323944, "time": 0.6449474096298218, "epoch": 7, "memory": 44138, "step": 24442} +{"lr": 3.737140432103754e-05, "data_time": 0.0015803337097167968, "loss": 0.3762570321559906, "time": 0.6376269817352295, "epoch": 7, "memory": 44138, "step": 24542} +{"lr": 3.703905669485545e-05, "data_time": 0.0015498161315917968, "loss": 0.4037273466587067, "time": 0.636371660232544, "epoch": 7, "memory": 44138, "step": 24642} +{"lr": 3.6707870356102706e-05, "data_time": 0.0017011404037475587, "loss": 0.3758342534303665, "time": 0.6364459276199341, "epoch": 7, "memory": 44138, "step": 24742} +{"lr": 3.6377866718049484e-05, "data_time": 0.001623845100402832, "loss": 0.3691035211086273, "time": 0.6371749401092529, "epoch": 7, "memory": 44138, "step": 24842} +{"lr": 3.6049067117496965e-05, "data_time": 0.001714324951171875, "loss": 0.34939275979995726, "time": 0.6372305393218994, "epoch": 7, "memory": 44138, "step": 24942} +{"lr": 3.5721492813397755e-05, "data_time": 0.00166628360748291, "loss": 0.378491273522377, "time": 0.6374207973480225, "epoch": 7, "memory": 44138, "step": 25042} +{"lr": 3.539516498548151e-05, "data_time": 0.0015749931335449219, "loss": 0.36208135783672335, "time": 0.6369609832763672, "epoch": 7, "memory": 44138, "step": 25142} +{"lr": 3.507010473288518e-05, "data_time": 0.0014975309371948243, "loss": 0.3824771374464035, "time": 0.6363403081893921, "epoch": 7, "memory": 44138, "step": 25242} +{"lr": 3.474633307278909e-05, "data_time": 0.0018010139465332031, "loss": 0.3768818318843842, "time": 0.6368935346603394, "epoch": 7, "memory": 44138, "step": 25342} +{"lr": 3.4423870939058177e-05, "data_time": 0.001985311508178711, "loss": 0.37051327228546144, "time": 0.6367572546005249, "epoch": 7, "memory": 44138, "step": 25442} +{"lr": 3.410273918088805e-05, "data_time": 0.001759958267211914, "loss": 0.36509051620960237, "time": 0.6366346120834351, "epoch": 7, "memory": 44138, "step": 25542} +{"lr": 3.37829585614574e-05, "data_time": 0.0016704082489013671, "loss": 0.3870923310518265, "time": 0.6371009349822998, "epoch": 7, "memory": 44138, "step": 25642} +{"lr": 3.3464549756585216e-05, "data_time": 0.0016191959381103515, "loss": 0.3867280274629593, "time": 0.6364024162292481, "epoch": 7, "memory": 44138, "step": 25742} +{"lr": 3.314753335339416e-05, "data_time": 0.0020956993103027344, "loss": 0.3788897156715393, "time": 0.6379543781280518, "epoch": 7, "memory": 44138, "step": 25842} +{"lr": 3.2831929848979394e-05, "data_time": 0.0016489505767822265, "loss": 0.40125176012516023, "time": 0.6375192880630494, "epoch": 7, "memory": 44138, "step": 25942} +{"lr": 3.251775964908332e-05, "data_time": 0.0016282081604003906, "loss": 0.36865064799785613, "time": 0.6376344442367554, "epoch": 7, "memory": 44138, "step": 26042} +{"lr": 3.22050430667762e-05, "data_time": 0.0014970779418945312, "loss": 0.38661822080612185, "time": 0.6375813245773315, "epoch": 7, "memory": 44138, "step": 26142} +{"lr": 3.189380032114285e-05, "data_time": 0.001978421211242676, "loss": 0.3730123072862625, "time": 0.6370380878448486, "epoch": 7, "memory": 44138, "step": 26242} +{"lr": 3.1584051535975324e-05, "data_time": 0.001821446418762207, "loss": 0.37764213383197787, "time": 0.6370634317398072, "epoch": 7, "memory": 44138, "step": 26342} +{"lr": 3.127581673847175e-05, "data_time": 0.001682138442993164, "loss": 0.3637180894613266, "time": 0.635551929473877, "epoch": 7, "memory": 44138, "step": 26442} +{"lr": 3.0969115857941555e-05, "data_time": 0.0014571666717529297, "loss": 0.3561870127916336, "time": 0.635262131690979, "epoch": 7, "memory": 44138, "step": 26542} +{"lr": 3.0663968724516654e-05, "data_time": 0.0016916036605834962, "loss": 0.36786005795001986, "time": 0.6352062940597534, "epoch": 7, "memory": 44138, "step": 26642} +{"lr": 3.0360395067869668e-05, "data_time": 0.0023400306701660155, "loss": 0.38404776751995084, "time": 0.6361199378967285, "epoch": 7, "memory": 44138, "step": 26742} +{"lr": 3.0058414515938063e-05, "data_time": 0.0015816926956176759, "loss": 0.38932461738586427, "time": 0.6373611450195312, "epoch": 7, "memory": 44138, "step": 26842} +{"lr": 2.975804659365501e-05, "data_time": 0.001865553855895996, "loss": 0.3752181679010391, "time": 0.6381101131439209, "epoch": 7, "memory": 44138, "step": 26942} +{"lr": 2.945931072168714e-05, "data_time": 0.0015078544616699218, "loss": 0.3806620329618454, "time": 0.6370229959487915, "epoch": 7, "memory": 44138, "step": 27042} +{"lr": 2.9162226215178855e-05, "data_time": 0.001818370819091797, "loss": 0.39434253573417666, "time": 0.6375951528549194, "epoch": 7, "memory": 44138, "step": 27142} +{"lr": 2.8866812282503322e-05, "data_time": 0.0014791250228881835, "loss": 0.36256890296936034, "time": 0.6377808809280395, "epoch": 7, "memory": 44138, "step": 27242} +{"lr": 2.8573088024020838e-05, "data_time": 0.0013997554779052734, "loss": 0.37059479355812075, "time": 0.6366333723068237, "epoch": 7, "memory": 44138, "step": 27342} +{"accuracy/top1": 77.22417449951172, "data_time": 0.031212153105900205, "time": 1.2760090129128818, "step": 7} +{"lr": 2.8260695817390092e-05, "data_time": 0.0017768621444702148, "loss": 0.35927407443523407, "time": 0.6357892274856567, "epoch": 8, "memory": 44138, "step": 27449} +{"lr": 2.7970529402570967e-05, "data_time": 0.002074456214904785, "loss": 0.3875000327825546, "time": 0.635071086883545, "epoch": 8, "memory": 44138, "step": 27549} +{"lr": 2.7682110612244214e-05, "data_time": 0.0014928340911865234, "loss": 0.3755984216928482, "time": 0.6351062059402466, "epoch": 8, "memory": 44138, "step": 27649} +{"lr": 2.7395458094490283e-05, "data_time": 0.001944732666015625, "loss": 0.3727742224931717, "time": 0.6350639343261719, "epoch": 8, "memory": 44138, "step": 27749} +{"lr": 2.7110590383189015e-05, "data_time": 0.0018382310867309571, "loss": 0.37217639684677123, "time": 0.6346689462661743, "epoch": 8, "memory": 44138, "step": 27849} +{"lr": 2.6827525896821397e-05, "data_time": 0.0015794992446899413, "loss": 0.4010009288787842, "time": 0.6375937223434448, "epoch": 8, "memory": 44138, "step": 27949} +{"lr": 2.6546282937278552e-05, "data_time": 0.0026099443435668944, "loss": 0.3738888442516327, "time": 0.6376948118209839, "epoch": 8, "memory": 44138, "step": 28049} +{"lr": 2.6266879688678606e-05, "data_time": 0.0015447378158569337, "loss": 0.37924714386463165, "time": 0.6351224899291992, "epoch": 8, "memory": 44138, "step": 28149} +{"lr": 2.5989334216190738e-05, "data_time": 0.0018911123275756835, "loss": 0.381095290184021, "time": 0.6379172563552856, "epoch": 8, "memory": 44138, "step": 28249} +{"lr": 2.571366446486732e-05, "data_time": 0.0016260862350463868, "loss": 0.39325780272483823, "time": 0.6374745607376099, "epoch": 8, "memory": 44138, "step": 28349} +{"lr": 2.5439888258483665e-05, "data_time": 0.0015037775039672852, "loss": 0.35682632923126223, "time": 0.6357267379760743, "epoch": 8, "memory": 44138, "step": 28449} +{"lr": 2.5168023298385384e-05, "data_time": 0.0017645597457885743, "loss": 0.345992836356163, "time": 0.6345254182815552, "epoch": 8, "memory": 44138, "step": 28549} +{"lr": 2.489808716234412e-05, "data_time": 0.0015947818756103516, "loss": 0.355545736849308, "time": 0.6378972291946411, "epoch": 8, "memory": 44138, "step": 28649} +{"lr": 2.463009730342111e-05, "data_time": 0.002026057243347168, "loss": 0.38551085591316225, "time": 0.6361251831054687, "epoch": 8, "memory": 44138, "step": 28749} +{"lr": 2.436407104883849e-05, "data_time": 0.0015492200851440429, "loss": 0.37640159726142886, "time": 0.6361212730407715, "epoch": 8, "memory": 44138, "step": 28849} +{"lr": 2.4100025598859092e-05, "data_time": 0.0017190933227539062, "loss": 0.3730620265007019, "time": 0.6357465982437134, "epoch": 8, "memory": 44138, "step": 28949} +{"lr": 2.3837978025674284e-05, "data_time": 0.0015191078186035157, "loss": 0.35198096334934237, "time": 0.6363602638244629, "epoch": 8, "memory": 44138, "step": 29049} +{"lr": 2.357794527230027e-05, "data_time": 0.0023094654083251954, "loss": 0.3522915422916412, "time": 0.6370986461639404, "epoch": 8, "memory": 44138, "step": 29149} +{"lr": 2.33199441514825e-05, "data_time": 0.0018253326416015625, "loss": 0.39003380835056306, "time": 0.6351585149765014, "epoch": 8, "memory": 44138, "step": 29249} +{"lr": 2.306399134460859e-05, "data_time": 0.0017494678497314454, "loss": 0.37878172397613524, "time": 0.6375424861907959, "epoch": 8, "memory": 44138, "step": 29349} +{"lr": 2.281010340062995e-05, "data_time": 0.0016418933868408204, "loss": 0.38247411251068114, "time": 0.635607099533081, "epoch": 8, "memory": 44138, "step": 29449} +{"lr": 2.2558296734991603e-05, "data_time": 0.001974749565124512, "loss": 0.38675096333026887, "time": 0.6354138612747192, "epoch": 8, "memory": 44138, "step": 29549} +{"lr": 2.2308587628570915e-05, "data_time": 0.0021421432495117186, "loss": 0.34000577330589293, "time": 0.6485737085342407, "epoch": 8, "memory": 44138, "step": 29649} +{"lr": 2.2060992226624955e-05, "data_time": 0.0017456769943237304, "loss": 0.35701105892658236, "time": 0.6358763694763183, "epoch": 8, "memory": 44138, "step": 29749} +{"lr": 2.181552653774648e-05, "data_time": 0.0016990184783935546, "loss": 0.36373075246810915, "time": 0.6351004123687745, "epoch": 8, "memory": 44138, "step": 29849} +{"lr": 2.1572206432829035e-05, "data_time": 0.0018560647964477538, "loss": 0.3286889672279358, "time": 0.6363370656967163, "epoch": 8, "memory": 44138, "step": 29949} +{"lr": 2.1331047644040754e-05, "data_time": 0.002002406120300293, "loss": 0.39221576750278475, "time": 0.6356395483016968, "epoch": 8, "memory": 44138, "step": 30049} +{"lr": 2.109206576380705e-05, "data_time": 0.002071785926818848, "loss": 0.3737366795539856, "time": 0.6362970352172852, "epoch": 8, "memory": 44138, "step": 30149} +{"lr": 2.0855276243802687e-05, "data_time": 0.0017470836639404297, "loss": 0.3948597311973572, "time": 0.6365963220596313, "epoch": 8, "memory": 44138, "step": 30249} +{"lr": 2.0620694393952487e-05, "data_time": 0.0016836404800415039, "loss": 0.3693047225475311, "time": 0.6352022171020508, "epoch": 8, "memory": 44138, "step": 30349} +{"lr": 2.038833538144169e-05, "data_time": 0.001968693733215332, "loss": 0.34211274087429044, "time": 0.6348795413970947, "epoch": 8, "memory": 44138, "step": 30449} +{"lr": 2.0158214229735154e-05, "data_time": 0.002071523666381836, "loss": 0.3605386734008789, "time": 0.634950852394104, "epoch": 8, "memory": 44138, "step": 30549} +{"lr": 1.9930345817605977e-05, "data_time": 0.002180933952331543, "loss": 0.35083739161491395, "time": 0.6377945661544799, "epoch": 8, "memory": 44138, "step": 30649} +{"lr": 1.970474487817358e-05, "data_time": 0.0018900394439697265, "loss": 0.3833119809627533, "time": 0.6348079681396485, "epoch": 8, "memory": 44138, "step": 30749} +{"lr": 1.9481425997951092e-05, "data_time": 0.001707005500793457, "loss": 0.37710258066654206, "time": 0.6366045713424683, "epoch": 8, "memory": 44138, "step": 30849} +{"lr": 1.9260403615902187e-05, "data_time": 0.001898503303527832, "loss": 0.37433295249938964, "time": 0.6373776912689209, "epoch": 8, "memory": 44138, "step": 30949} +{"lr": 1.9041692022507647e-05, "data_time": 0.0017877340316772461, "loss": 0.3876935660839081, "time": 0.6366706609725952, "epoch": 8, "memory": 44138, "step": 31049} +{"lr": 1.8825305358841125e-05, "data_time": 0.0017325162887573242, "loss": 0.3609554827213287, "time": 0.6356470584869385, "epoch": 8, "memory": 44138, "step": 31149} +{"lr": 1.861125761565518e-05, "data_time": 0.0017855644226074218, "loss": 0.3503482758998871, "time": 0.6341592311859131, "epoch": 8, "memory": 44138, "step": 31249} +{"accuracy/top1": 77.11631774902344, "data_time": 0.031988135699568127, "time": 1.275291237337836, "step": 8} +{"lr": 1.8384832449122864e-05, "data_time": 0.0017569541931152343, "loss": 0.37776540517807006, "time": 0.6363374710083007, "epoch": 9, "memory": 44138, "step": 31356} +{"lr": 1.817567007332114e-05, "data_time": 0.0021444320678710937, "loss": 0.36052057445049285, "time": 0.6455218315124511, "epoch": 9, "memory": 44138, "step": 31456} +{"lr": 1.796888862098825e-05, "data_time": 0.0021024465560913084, "loss": 0.37929793298244474, "time": 0.6360148906707763, "epoch": 9, "memory": 44138, "step": 31556} +{"lr": 1.7764501461839678e-05, "data_time": 0.0017917871475219727, "loss": 0.37280177474021914, "time": 0.636647629737854, "epoch": 9, "memory": 44138, "step": 31656} +{"lr": 1.7562521810784746e-05, "data_time": 0.002162933349609375, "loss": 0.37447566986083985, "time": 0.636457896232605, "epoch": 9, "memory": 44138, "step": 31756} +{"lr": 1.7362962727072444e-05, "data_time": 0.0023943901062011717, "loss": 0.3473565518856049, "time": 0.6358568429946899, "epoch": 9, "memory": 44138, "step": 31856} +{"lr": 1.716583711344682e-05, "data_time": 0.002496647834777832, "loss": 0.3924066424369812, "time": 0.6379173278808594, "epoch": 9, "memory": 44138, "step": 31956} +{"lr": 1.6971157715312954e-05, "data_time": 0.001843571662902832, "loss": 0.35017883479595185, "time": 0.638804841041565, "epoch": 9, "memory": 44138, "step": 32056} +{"lr": 1.677893711991274e-05, "data_time": 0.0016889333724975585, "loss": 0.36119251549243925, "time": 0.6374097347259522, "epoch": 9, "memory": 44138, "step": 32156} +{"lr": 1.6589187755511025e-05, "data_time": 0.0020768404006958007, "loss": 0.35062299072742464, "time": 0.6507104396820068, "epoch": 9, "memory": 44138, "step": 32256} +{"lr": 1.640192189059216e-05, "data_time": 0.0019334077835083008, "loss": 0.3661290407180786, "time": 0.6358798742294312, "epoch": 9, "memory": 44138, "step": 32356} +{"lr": 1.6217151633066716e-05, "data_time": 0.0020746231079101563, "loss": 0.36143653392791747, "time": 0.6372776985168457, "epoch": 9, "memory": 44138, "step": 32456} +{"lr": 1.6034888929488564e-05, "data_time": 0.0018400430679321289, "loss": 0.36948657035827637, "time": 0.6380186319351197, "epoch": 9, "memory": 44138, "step": 32556} +{"lr": 1.585514556428258e-05, "data_time": 0.0018936634063720704, "loss": 0.36524722576141355, "time": 0.6360497236251831, "epoch": 9, "memory": 44138, "step": 32656} +{"lr": 1.567793315898261e-05, "data_time": 0.0021505832672119142, "loss": 0.38852599561214446, "time": 0.6369391202926635, "epoch": 9, "memory": 44138, "step": 32756} +{"lr": 1.5503263171480048e-05, "data_time": 0.001978635787963867, "loss": 0.38102494478225707, "time": 0.6393379449844361, "epoch": 9, "memory": 44138, "step": 32856} +{"lr": 1.5331146895283165e-05, "data_time": 0.0019562959671020506, "loss": 0.37116784453392027, "time": 0.6374693393707276, "epoch": 9, "memory": 44138, "step": 32956} +{"lr": 1.5161595458786728e-05, "data_time": 0.0017194032669067382, "loss": 0.3558268159627914, "time": 0.6382407188415528, "epoch": 9, "memory": 44138, "step": 33056} +{"lr": 1.4994619824552563e-05, "data_time": 0.0019527196884155274, "loss": 0.37153142094612124, "time": 0.63702871799469, "epoch": 9, "memory": 44138, "step": 33156} +{"lr": 1.4830230788600802e-05, "data_time": 0.0019491910934448242, "loss": 0.3914036899805069, "time": 0.6374222755432128, "epoch": 9, "memory": 44138, "step": 33256} +{"lr": 1.4668438979711779e-05, "data_time": 0.001941847801208496, "loss": 0.370823061466217, "time": 0.6398892641067505, "epoch": 9, "memory": 44138, "step": 33356} +{"lr": 1.4509254858738844e-05, "data_time": 0.0022425174713134764, "loss": 0.3787620007991791, "time": 0.6387940168380737, "epoch": 9, "memory": 44138, "step": 33456} +{"lr": 1.4352688717931966e-05, "data_time": 0.0024170875549316406, "loss": 0.3609553128480911, "time": 0.638194227218628, "epoch": 9, "memory": 44138, "step": 33556} +{"lr": 1.4198750680272348e-05, "data_time": 0.0021574735641479493, "loss": 0.3617054492235184, "time": 0.6384587526321411, "epoch": 9, "memory": 44138, "step": 33656} +{"lr": 1.4047450698817857e-05, "data_time": 0.0017816543579101563, "loss": 0.3511413484811783, "time": 0.6382307767868042, "epoch": 9, "memory": 44138, "step": 33756} +{"lr": 1.389879855605952e-05, "data_time": 0.0016302108764648438, "loss": 0.3800592184066772, "time": 0.6376355648040771, "epoch": 9, "memory": 44138, "step": 33856} +{"lr": 1.3752803863288976e-05, "data_time": 0.001538991928100586, "loss": 0.3761305302381516, "time": 0.6357406377792358, "epoch": 9, "memory": 44138, "step": 33956} +{"lr": 1.3609476059977129e-05, "data_time": 0.002507805824279785, "loss": 0.3871601551771164, "time": 0.6384088277816773, "epoch": 9, "memory": 44138, "step": 34056} +{"lr": 1.3468824413163754e-05, "data_time": 0.0021561145782470702, "loss": 0.3784525364637375, "time": 0.6383571624755859, "epoch": 9, "memory": 44138, "step": 34156} +{"lr": 1.3330858016858395e-05, "data_time": 0.0026237010955810548, "loss": 0.398921412229538, "time": 0.6372433662414551, "epoch": 9, "memory": 44138, "step": 34256} +{"lr": 1.3195585791452309e-05, "data_time": 0.0019191741943359376, "loss": 0.37292216420173646, "time": 0.6371522665023803, "epoch": 9, "memory": 44138, "step": 34356} +{"lr": 1.3063016483141729e-05, "data_time": 0.001530003547668457, "loss": 0.3519412994384766, "time": 0.6357794046401978, "epoch": 9, "memory": 44138, "step": 34456} +{"lr": 1.2933158663362388e-05, "data_time": 0.002068305015563965, "loss": 0.3784942656755447, "time": 0.6371706247329711, "epoch": 9, "memory": 44138, "step": 34556} +{"lr": 1.2806020728235311e-05, "data_time": 0.001889181137084961, "loss": 0.3695018321275711, "time": 0.6370637893676758, "epoch": 9, "memory": 44138, "step": 34656} +{"lr": 1.2681610898023963e-05, "data_time": 0.0019256353378295898, "loss": 0.3664830893278122, "time": 0.6462860822677612, "epoch": 9, "memory": 44138, "step": 34756} +{"lr": 1.2559937216602731e-05, "data_time": 0.0019288778305053711, "loss": 0.36039121448993683, "time": 0.6378018140792847, "epoch": 9, "memory": 44138, "step": 34856} +{"lr": 1.2441007550936847e-05, "data_time": 0.0018878459930419921, "loss": 0.39259982407093047, "time": 0.6406694889068604, "epoch": 9, "memory": 44138, "step": 34956} +{"lr": 1.232482959057377e-05, "data_time": 0.0021547555923461916, "loss": 0.40247650146484376, "time": 0.6380646228790283, "epoch": 9, "memory": 44138, "step": 35056} +{"lr": 1.2211410847145969e-05, "data_time": 0.002462339401245117, "loss": 0.3598687320947647, "time": 0.6378409147262574, "epoch": 9, "memory": 44138, "step": 35156} +{"accuracy/top1": 76.75446319580078, "data_time": 0.031586264741831814, "time": 1.2749265555677742, "step": 9} +{"lr": 1.209311679360931e-05, "data_time": 0.002844357490539551, "loss": 0.3891842246055603, "time": 0.6390421867370606, "epoch": 10, "memory": 44138, "step": 35263} +{"lr": 1.1985432727488666e-05, "data_time": 0.0025748729705810545, "loss": 0.36259826123714445, "time": 0.6365508079528809, "epoch": 10, "memory": 44138, "step": 35363} +{"lr": 1.1880529822435289e-05, "data_time": 0.0026526689529418946, "loss": 0.38314634561538696, "time": 0.651921558380127, "epoch": 10, "memory": 44138, "step": 35463} +{"lr": 1.1778414861078726e-05, "data_time": 0.0029461860656738283, "loss": 0.35298205018043516, "time": 0.6399567365646363, "epoch": 10, "memory": 44138, "step": 35563} +{"lr": 1.1679094445790538e-05, "data_time": 0.0024763107299804687, "loss": 0.3704716444015503, "time": 0.6371896028518677, "epoch": 10, "memory": 44138, "step": 35663} +{"lr": 1.1582574998257377e-05, "data_time": 0.002500033378601074, "loss": 0.36506556868553164, "time": 0.6490757465362549, "epoch": 10, "memory": 44138, "step": 35763} +{"lr": 1.1488862759065762e-05, "data_time": 0.0023409128189086914, "loss": 0.37412916123867035, "time": 0.6364859342575073, "epoch": 10, "memory": 44138, "step": 35863} +{"lr": 1.1397963787298665e-05, "data_time": 0.0025920629501342773, "loss": 0.3624534010887146, "time": 0.6381866693496704, "epoch": 10, "memory": 44138, "step": 35963} +{"lr": 1.1309883960143664e-05, "data_time": 0.0023764610290527345, "loss": 0.34261963963508607, "time": 0.6460938453674316, "epoch": 10, "memory": 44138, "step": 36063} +{"lr": 1.122462897251302e-05, "data_time": 0.003063464164733887, "loss": 0.36822369396686555, "time": 0.6370961666107178, "epoch": 10, "memory": 44138, "step": 36163} +{"lr": 1.114220433667542e-05, "data_time": 0.0031174182891845702, "loss": 0.3779571563005447, "time": 0.6410081386566162, "epoch": 10, "memory": 44138, "step": 36263} +{"lr": 1.10626153818996e-05, "data_time": 0.00319979190826416, "loss": 0.37404989302158353, "time": 0.637820553779602, "epoch": 10, "memory": 44138, "step": 36363} +{"lr": 1.0985867254109761e-05, "data_time": 0.001939535140991211, "loss": 0.3587206810712814, "time": 0.636319613456726, "epoch": 10, "memory": 44138, "step": 36463} +{"lr": 1.0911964915552823e-05, "data_time": 0.002281594276428223, "loss": 0.3492222785949707, "time": 0.6368070125579834, "epoch": 10, "memory": 44138, "step": 36563} +{"lr": 1.0840913144477643e-05, "data_time": 0.0018481731414794922, "loss": 0.36618653535842893, "time": 0.6374979972839355, "epoch": 10, "memory": 44138, "step": 36663} +{"lr": 1.0772716534826064e-05, "data_time": 0.0025023698806762697, "loss": 0.3560271501541138, "time": 0.6376560449600219, "epoch": 10, "memory": 44138, "step": 36763} +{"lr": 1.0707379495935828e-05, "data_time": 0.0031203031539916992, "loss": 0.3666188925504684, "time": 0.6392525911331177, "epoch": 10, "memory": 44138, "step": 36863} +{"lr": 1.0644906252255541e-05, "data_time": 0.0027537107467651366, "loss": 0.38266485631465913, "time": 0.6370707988739014, "epoch": 10, "memory": 44138, "step": 36963} +{"lr": 1.0585300843071527e-05, "data_time": 0.002393484115600586, "loss": 0.3794155478477478, "time": 0.6477534532546997, "epoch": 10, "memory": 44138, "step": 37063} +{"lr": 1.0528567122246654e-05, "data_time": 0.002044534683227539, "loss": 0.3628378063440323, "time": 0.6361784696578979, "epoch": 10, "memory": 44138, "step": 37163} +{"lr": 1.0474708757971141e-05, "data_time": 0.0025911808013916017, "loss": 0.3967668205499649, "time": 0.6381731033325195, "epoch": 10, "memory": 44138, "step": 37263} +{"lr": 1.042372923252543e-05, "data_time": 0.0016278743743896485, "loss": 0.36560956537723543, "time": 0.6358978509902954, "epoch": 10, "memory": 44138, "step": 37363} +{"lr": 1.0375631842054998e-05, "data_time": 0.0021711111068725584, "loss": 0.36271997094154357, "time": 0.6368345260620117, "epoch": 10, "memory": 44138, "step": 37463} +{"lr": 1.0330419696357254e-05, "data_time": 0.002533555030822754, "loss": 0.3798720002174377, "time": 0.6369867086410522, "epoch": 10, "memory": 44138, "step": 37563} +{"lr": 1.028809571868047e-05, "data_time": 0.002904343605041504, "loss": 0.36764702796936033, "time": 0.6370655059814453, "epoch": 10, "memory": 44138, "step": 37663} +{"lr": 1.0248662645534766e-05, "data_time": 0.002500581741333008, "loss": 0.34937004148960116, "time": 0.6373331785202027, "epoch": 10, "memory": 44138, "step": 37763} +{"lr": 1.0212123026515194e-05, "data_time": 0.002648138999938965, "loss": 0.3732252478599548, "time": 0.6372593879699707, "epoch": 10, "memory": 44138, "step": 37863} +{"lr": 1.0178479224136881e-05, "data_time": 0.002384328842163086, "loss": 0.3445000916719437, "time": 0.6363013982772827, "epoch": 10, "memory": 44138, "step": 37963} +{"lr": 1.0147733413682289e-05, "data_time": 0.002006673812866211, "loss": 0.35463112890720366, "time": 0.6370181798934936, "epoch": 10, "memory": 44138, "step": 38063} +{"lr": 1.0119887583060527e-05, "data_time": 0.0016669988632202148, "loss": 0.37355350852012636, "time": 0.6363584041595459, "epoch": 10, "memory": 44138, "step": 38163} +{"lr": 1.0094943532678871e-05, "data_time": 0.002047157287597656, "loss": 0.36430065631866454, "time": 0.6355287075042725, "epoch": 10, "memory": 44138, "step": 38263} +{"lr": 1.007290287532637e-05, "data_time": 0.002046823501586914, "loss": 0.3745395392179489, "time": 0.6372630834579468, "epoch": 10, "memory": 44138, "step": 38363} +{"lr": 1.0053767036069495e-05, "data_time": 0.00197749137878418, "loss": 0.3921888291835785, "time": 0.6387291669845581, "epoch": 10, "memory": 44138, "step": 38463} +{"lr": 1.003753725216005e-05, "data_time": 0.002440333366394043, "loss": 0.36490239202976227, "time": 0.6365254640579223, "epoch": 10, "memory": 44138, "step": 38563} +{"lr": 1.0024214572955152e-05, "data_time": 0.0018944740295410156, "loss": 0.35240968465805056, "time": 0.6374494314193726, "epoch": 10, "memory": 44138, "step": 38663} +{"lr": 1.0013799859849456e-05, "data_time": 0.0018710613250732422, "loss": 0.40322225391864774, "time": 0.6369835138320923, "epoch": 10, "memory": 44138, "step": 38763} +{"lr": 1.0006293786219338e-05, "data_time": 0.0025926828384399414, "loss": 0.386372908949852, "time": 0.6362984657287598, "epoch": 10, "memory": 44138, "step": 38863} +{"lr": 1.0001696837379472e-05, "data_time": 0.0026990652084350588, "loss": 0.3724012911319733, "time": 0.6380775451660157, "epoch": 10, "memory": 44138, "step": 38963} +{"lr": 1.0000009310551361e-05, "data_time": 0.0021074295043945314, "loss": 0.38128813207149503, "time": 0.6352448463439941, "epoch": 10, "memory": 44138, "step": 39063} +{"accuracy/top1": 76.35781860351562, "data_time": 0.03211379051208496, "time": 1.2754385717983903, "step": 10} diff --git a/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm_wneck/20230604_032014/vis_data/vis_image/0000000.png_1.png b/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm_wneck/20230604_032014/vis_data/vis_image/0000000.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..6db2a0b554ae66910f94a67805e51762f397f8f4 Binary files /dev/null and b/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm_wneck/20230604_032014/vis_data/vis_image/0000000.png_1.png differ diff --git a/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm_wneck/20230604_032014/vis_data/vis_image/0000000.png_10.png b/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm_wneck/20230604_032014/vis_data/vis_image/0000000.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..605fbed0e05e2c4fe9356fa19df0c7dbf5a83358 Binary files /dev/null and b/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm_wneck/20230604_032014/vis_data/vis_image/0000000.png_10.png differ diff --git a/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm_wneck/20230604_032014/vis_data/vis_image/0000000.png_2.png b/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm_wneck/20230604_032014/vis_data/vis_image/0000000.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..37737da46d9556fce860df34f1a98439a7e18f5e Binary files /dev/null and b/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm_wneck/20230604_032014/vis_data/vis_image/0000000.png_2.png differ diff --git a/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm_wneck/20230604_032014/vis_data/vis_image/0000000.png_3.png b/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm_wneck/20230604_032014/vis_data/vis_image/0000000.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..c713beb8e2d41d249edf5e647db2462810a9bf9f Binary files /dev/null and b/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm_wneck/20230604_032014/vis_data/vis_image/0000000.png_3.png differ diff --git a/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm_wneck/20230604_032014/vis_data/vis_image/0000000.png_4.png b/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm_wneck/20230604_032014/vis_data/vis_image/0000000.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..ba69ae50defcdab1ee1c8bc83351950663a05702 Binary files /dev/null and b/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm_wneck/20230604_032014/vis_data/vis_image/0000000.png_4.png differ diff --git a/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm_wneck/20230604_032014/vis_data/vis_image/0000000.png_5.png b/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm_wneck/20230604_032014/vis_data/vis_image/0000000.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..37d3c33dd670957a7a0a4f097b8c7370b932b38f Binary files /dev/null and b/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm_wneck/20230604_032014/vis_data/vis_image/0000000.png_5.png differ diff --git a/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm_wneck/20230604_032014/vis_data/vis_image/0000000.png_6.png b/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm_wneck/20230604_032014/vis_data/vis_image/0000000.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..f34e304859f00e9f32fc9533bcc47d7f749e1a27 Binary files /dev/null and b/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm_wneck/20230604_032014/vis_data/vis_image/0000000.png_6.png differ diff --git a/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm_wneck/20230604_032014/vis_data/vis_image/0000000.png_7.png b/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm_wneck/20230604_032014/vis_data/vis_image/0000000.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..0f688e6e95bd93a282dc9f72a8faa044875c69b2 Binary files /dev/null and b/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm_wneck/20230604_032014/vis_data/vis_image/0000000.png_7.png differ diff --git a/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm_wneck/20230604_032014/vis_data/vis_image/0000000.png_8.png b/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm_wneck/20230604_032014/vis_data/vis_image/0000000.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..c1f20c6ed34a9630ce2721db23cd9a633825e543 Binary files /dev/null and b/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm_wneck/20230604_032014/vis_data/vis_image/0000000.png_8.png differ diff --git a/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm_wneck/20230604_032014/vis_data/vis_image/0000000.png_9.png b/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm_wneck/20230604_032014/vis_data/vis_image/0000000.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..45c34b8263a33c5b301c56bc577befed7daf5554 Binary files /dev/null and b/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm_wneck/20230604_032014/vis_data/vis_image/0000000.png_9.png differ diff --git a/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm_wneck/20230604_032014/vis_data/vis_image/000005174.jpg_1.png b/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm_wneck/20230604_032014/vis_data/vis_image/000005174.jpg_1.png new file mode 100644 index 0000000000000000000000000000000000000000..4fc6e4ac816ac8a2aa92bc3bfb96fa58cfc58f1b Binary files /dev/null and b/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm_wneck/20230604_032014/vis_data/vis_image/000005174.jpg_1.png differ diff --git a/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm_wneck/20230604_032014/vis_data/vis_image/000005174.jpg_10.png b/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm_wneck/20230604_032014/vis_data/vis_image/000005174.jpg_10.png new file mode 100644 index 0000000000000000000000000000000000000000..86d9d7cfe76b2e293457e91b2a369aae74a67dad Binary files /dev/null and b/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm_wneck/20230604_032014/vis_data/vis_image/000005174.jpg_10.png differ diff --git a/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm_wneck/20230604_032014/vis_data/vis_image/000005174.jpg_2.png b/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm_wneck/20230604_032014/vis_data/vis_image/000005174.jpg_2.png new file mode 100644 index 0000000000000000000000000000000000000000..28f99b9f6a2b6a62c00d075b490e08cb77e62f56 Binary files /dev/null and b/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm_wneck/20230604_032014/vis_data/vis_image/000005174.jpg_2.png differ diff --git a/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm_wneck/20230604_032014/vis_data/vis_image/000005174.jpg_3.png b/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm_wneck/20230604_032014/vis_data/vis_image/000005174.jpg_3.png new file mode 100644 index 0000000000000000000000000000000000000000..1ad9109671e18764b4602596751e60159af22678 Binary files /dev/null and b/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm_wneck/20230604_032014/vis_data/vis_image/000005174.jpg_3.png differ diff --git a/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm_wneck/20230604_032014/vis_data/vis_image/000005174.jpg_4.png b/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm_wneck/20230604_032014/vis_data/vis_image/000005174.jpg_4.png new file mode 100644 index 0000000000000000000000000000000000000000..57712ae0a03bf3c69c528eb7578b8cdb15b5550b Binary files /dev/null and b/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm_wneck/20230604_032014/vis_data/vis_image/000005174.jpg_4.png differ diff --git a/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm_wneck/20230604_032014/vis_data/vis_image/000005174.jpg_5.png b/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm_wneck/20230604_032014/vis_data/vis_image/000005174.jpg_5.png new file mode 100644 index 0000000000000000000000000000000000000000..b5909299ce9486283f48c6df0ae0aede3cd8c454 Binary files /dev/null and b/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm_wneck/20230604_032014/vis_data/vis_image/000005174.jpg_5.png differ diff --git a/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm_wneck/20230604_032014/vis_data/vis_image/000005174.jpg_6.png b/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm_wneck/20230604_032014/vis_data/vis_image/000005174.jpg_6.png new file mode 100644 index 0000000000000000000000000000000000000000..3245badb64ac32f303fece84d785cbee0e3e23d3 Binary files /dev/null and b/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm_wneck/20230604_032014/vis_data/vis_image/000005174.jpg_6.png differ diff --git a/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm_wneck/20230604_032014/vis_data/vis_image/000005174.jpg_7.png b/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm_wneck/20230604_032014/vis_data/vis_image/000005174.jpg_7.png new file mode 100644 index 0000000000000000000000000000000000000000..b18d54f5b086d22892962f195fd2f1f8c5041192 Binary files /dev/null and b/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm_wneck/20230604_032014/vis_data/vis_image/000005174.jpg_7.png differ diff --git a/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm_wneck/20230604_032014/vis_data/vis_image/000005174.jpg_8.png b/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm_wneck/20230604_032014/vis_data/vis_image/000005174.jpg_8.png new file mode 100644 index 0000000000000000000000000000000000000000..24b0daeb91717a14caa2b100d2c8b8844b8c7569 Binary files /dev/null and b/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm_wneck/20230604_032014/vis_data/vis_image/000005174.jpg_8.png differ diff --git a/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm_wneck/20230604_032014/vis_data/vis_image/000005174.jpg_9.png b/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm_wneck/20230604_032014/vis_data/vis_image/000005174.jpg_9.png new file mode 100644 index 0000000000000000000000000000000000000000..8406b542a9ab97f3fd70dbaf96221ef21f91e9a6 Binary files /dev/null and b/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm_wneck/20230604_032014/vis_data/vis_image/000005174.jpg_9.png differ diff --git a/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm_wneck/clip_large_pretrain_4x256_sdv2_lr1e-4.py b/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm_wneck/clip_large_pretrain_4x256_sdv2_lr1e-4.py new file mode 100644 index 0000000000000000000000000000000000000000..455960112013988f06579e1bb256e4040095dfad --- /dev/null +++ b/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm_wneck/clip_large_pretrain_4x256_sdv2_lr1e-4.py @@ -0,0 +1,236 @@ +optim_wrapper = dict( + optimizer=dict( + type='AdamW', lr=0.0001, weight_decay=0.3, _scope_='mmpretrain'), + paramwise_cfg=dict( + custom_keys=dict({ + '.cls_token': dict(decay_mult=0.0), + '.pos_embed': dict(decay_mult=0.0) + })), + type='AmpOptimWrapper', + dtype='bfloat16', + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=False, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=4096) +model = dict( + type='ImageClassifier', + backbone=dict( + frozen_stages=24, + type='VisionTransformer', + arch='l', + img_size=224, + patch_size=14, + drop_rate=0.1, + pre_norm=True, + final_norm=False, + init_cfg=dict( + type='Pretrained', + checkpoint='ckpt/openclip-ViT-L-14.pth', + prefix='backbone')), + neck=dict( + type='CLIPProjection', + in_channels=1024, + out_channels=768, + init_cfg=dict( + type='Pretrained', + checkpoint='ckpt/openclip-ViT-L-14.pth', + prefix='backbone')), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=768, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + init_cfg=None), + init_cfg=dict( + type='TruncNormal', layer=['Conv2d', 'Linear'], std=0.02, bias=0.0), + train_cfg=None) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=128, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stablediffusionV2-1-dpmsolver-25-1m.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/cc1m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=128, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV2-1-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=128, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV2-1-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm_wneck' diff --git a/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm_wneck/epoch_1.pth b/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm_wneck/epoch_1.pth new file mode 100644 index 0000000000000000000000000000000000000000..927c7575207368b423227a9b4874a2e1545d35b8 --- /dev/null +++ b/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm_wneck/epoch_1.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:a2c0f5cdf5b68907f31f544eb6371a755428539b7d0c017f0e9577337c63da5d +size 2438616321 diff --git a/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm_wneck/epoch_10.pth b/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm_wneck/epoch_10.pth new file mode 100644 index 0000000000000000000000000000000000000000..a8b5b0a76755c350dbe569ef34b8d5df22fa01df --- /dev/null +++ b/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm_wneck/epoch_10.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:0dab73ac622d1209cea3f1bffc3c29d7be07f2615826279cbe9cbdc85031a4c4 +size 2441185921 diff --git a/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm_wneck/epoch_2.pth b/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm_wneck/epoch_2.pth new file mode 100644 index 0000000000000000000000000000000000000000..20779df55b9f97574b56f88cf9040dfee3c693c7 --- /dev/null +++ b/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm_wneck/epoch_2.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:8e9cb8aa917ce2d9c9b144c0c2dbd137135cca2df74743a2f531a1da3b584d63 +size 2438900865 diff --git a/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm_wneck/epoch_3.pth b/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm_wneck/epoch_3.pth new file mode 100644 index 0000000000000000000000000000000000000000..e17dcd0be0154b7739ab40b7b441aeefa2482780 --- /dev/null +++ b/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm_wneck/epoch_3.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:664c2bb3205076aa9c748dd59152da8bd33e74557fc3ec0a375abba2dcf0f340 +size 2439184961 diff --git a/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm_wneck/epoch_4.pth b/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm_wneck/epoch_4.pth new file mode 100644 index 0000000000000000000000000000000000000000..5e2bc5f9e1f4709dc4de7f732eb02cb3f4f87de6 --- /dev/null +++ b/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm_wneck/epoch_4.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:78fa44a2f12c351c367dec57034024b436a3b946586863dfa0cb86bd6e7ad5f0 +size 2439468417 diff --git a/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm_wneck/epoch_5.pth b/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm_wneck/epoch_5.pth new file mode 100644 index 0000000000000000000000000000000000000000..5164014c0b62fa00431de804afe17394c5a2c08e --- /dev/null +++ b/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm_wneck/epoch_5.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:63cfa540a39450e1e3d3f4baec3f7310dff5599e7fcddd06911b20a21253bb3b +size 2439752193 diff --git a/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm_wneck/epoch_6.pth b/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm_wneck/epoch_6.pth new file mode 100644 index 0000000000000000000000000000000000000000..27712a61705a41f99da14bb091899038648abbb7 --- /dev/null +++ b/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm_wneck/epoch_6.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:550af4192e089fc4221de660e7ecd68b09117ffd9ab7a4726142a8c4c4158047 +size 2440036033 diff --git a/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm_wneck/epoch_7.pth b/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm_wneck/epoch_7.pth new file mode 100644 index 0000000000000000000000000000000000000000..9f325001b3360b598182234048e0eb7cdab7400f --- /dev/null +++ b/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm_wneck/epoch_7.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:1abfb3f550d589206b9c3e51aa92261f03c20763667bf014a311df93a45284d6 +size 2440320449 diff --git a/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm_wneck/epoch_8.pth b/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm_wneck/epoch_8.pth new file mode 100644 index 0000000000000000000000000000000000000000..26112a93307d245b87a32901bd2619b0d7c804f0 --- /dev/null +++ b/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm_wneck/epoch_8.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:9d073c6b90bfdaa198b19f47f546c7b0d57a20e36db00a63717643c41c90469d +size 2440608321 diff --git a/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm_wneck/epoch_9.pth b/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm_wneck/epoch_9.pth new file mode 100644 index 0000000000000000000000000000000000000000..c27cf936618effdb2f55703c56764e20c66ef842 --- /dev/null +++ b/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm_wneck/epoch_9.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:453e82f4a68857293b5338c6bbe7b3a22d453179a41f460992ff39dbb807a847 +size 2440896897 diff --git a/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm_wneck/last_checkpoint b/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm_wneck/last_checkpoint new file mode 100644 index 0000000000000000000000000000000000000000..ff1ea69216ce0b6fc0d721db94e4e316cb27f85f --- /dev/null +++ b/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm_wneck/last_checkpoint @@ -0,0 +1 @@ +/mnt/petrelfs/luzeyu/workspace/fakebench/mmpretrain/workdir/clip_large_pretrain_4x256_sdv2_lr1e-4_wopostnorm_wneck/epoch_10.pth \ No newline at end of file diff --git a/clip_large_pretrain_4x256_sdv2_lr3e-4/20230606_050007/20230606_050007.log b/clip_large_pretrain_4x256_sdv2_lr3e-4/20230606_050007/20230606_050007.log new file mode 100644 index 0000000000000000000000000000000000000000..bda967f17e8e5fc4356254d756a287a031a198c4 --- /dev/null +++ b/clip_large_pretrain_4x256_sdv2_lr3e-4/20230606_050007/20230606_050007.log @@ -0,0 +1,280 @@ +2023/06/06 05:00:15 - mmengine - INFO - +------------------------------------------------------------ +System environment: + sys.platform: linux + Python: 3.10.9 (main, Mar 8 2023, 10:47:38) [GCC 11.2.0] + CUDA available: True + numpy_random_seed: 424435667 + GPU 0,1,2,3: NVIDIA A100-SXM4-80GB + CUDA_HOME: /mnt/petrelfs/share/cuda-11.6 + NVCC: Cuda compilation tools, release 11.6, V11.6.124 + GCC: gcc (GCC) 7.5.0 + PyTorch: 1.13.1 + PyTorch compiling details: PyTorch built with: + - GCC 9.3 + - C++ Version: 201402 + - Intel(R) oneAPI Math Kernel Library Version 2021.4-Product Build 20210904 for Intel(R) 64 architecture applications + - Intel(R) MKL-DNN v2.6.0 (Git Hash 52b5f107dd9cf10910aaa19cb47f3abf9b349815) + - OpenMP 201511 (a.k.a. OpenMP 4.5) + - LAPACK is enabled (usually provided by MKL) + - NNPACK is enabled + - CPU capability usage: AVX2 + - CUDA Runtime 11.6 + - NVCC architecture flags: -gencode;arch=compute_37,code=sm_37;-gencode;arch=compute_50,code=sm_50;-gencode;arch=compute_60,code=sm_60;-gencode;arch=compute_61,code=sm_61;-gencode;arch=compute_70,code=sm_70;-gencode;arch=compute_75,code=sm_75;-gencode;arch=compute_80,code=sm_80;-gencode;arch=compute_86,code=sm_86;-gencode;arch=compute_37,code=compute_37 + - CuDNN 8.3.2 (built against CUDA 11.5) + - Magma 2.6.1 + - Build settings: BLAS_INFO=mkl, BUILD_TYPE=Release, CUDA_VERSION=11.6, CUDNN_VERSION=8.3.2, CXX_COMPILER=/opt/rh/devtoolset-9/root/usr/bin/c++, CXX_FLAGS= -fabi-version=11 -Wno-deprecated -fvisibility-inlines-hidden -DUSE_PTHREADPOOL -fopenmp -DNDEBUG -DUSE_KINETO -DUSE_FBGEMM -DUSE_QNNPACK -DUSE_PYTORCH_QNNPACK -DUSE_XNNPACK -DSYMBOLICATE_MOBILE_DEBUG_HANDLE -DEDGE_PROFILER_USE_KINETO -O2 -fPIC -Wno-narrowing -Wall -Wextra -Werror=return-type -Werror=non-virtual-dtor -Wno-missing-field-initializers -Wno-type-limits -Wno-array-bounds -Wno-unknown-pragmas -Wunused-local-typedefs -Wno-unused-parameter -Wno-unused-function -Wno-unused-result -Wno-strict-overflow -Wno-strict-aliasing -Wno-error=deprecated-declarations -Wno-stringop-overflow -Wno-psabi -Wno-error=pedantic -Wno-error=redundant-decls -Wno-error=old-style-cast -fdiagnostics-color=always -faligned-new -Wno-unused-but-set-variable -Wno-maybe-uninitialized -fno-math-errno -fno-trapping-math -Werror=format -Werror=cast-function-type -Wno-stringop-overflow, LAPACK_INFO=mkl, PERF_WITH_AVX=1, PERF_WITH_AVX2=1, PERF_WITH_AVX512=1, TORCH_VERSION=1.13.1, USE_CUDA=ON, USE_CUDNN=ON, USE_EXCEPTION_PTR=1, USE_GFLAGS=OFF, USE_GLOG=OFF, USE_MKL=ON, USE_MKLDNN=ON, USE_MPI=OFF, USE_NCCL=ON, USE_NNPACK=ON, USE_OPENMP=ON, USE_ROCM=OFF, + + TorchVision: 0.14.1 + OpenCV: 4.7.0 + MMEngine: 0.7.3 + +Runtime environment: + cudnn_benchmark: True + mp_cfg: {'mp_start_method': 'fork', 'opencv_num_threads': 0} + dist_cfg: {'backend': 'nccl'} + seed: None + deterministic: False + Distributed launcher: slurm + Distributed training: True + GPU number: 4 +------------------------------------------------------------ + +2023/06/06 05:00:19 - mmengine - INFO - Config: +optim_wrapper = dict( + optimizer=dict( + type='AdamW', lr=0.0003, weight_decay=0.3, _scope_='mmpretrain'), + paramwise_cfg=dict( + custom_keys=dict({ + '.cls_token': dict(decay_mult=0.0), + '.pos_embed': dict(decay_mult=0.0) + })), + type='AmpOptimWrapper', + dtype='bfloat16', + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=False, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=4096) +model = dict( + type='ImageClassifier', + backbone=dict( + frozen_stages=24, + type='VisionTransformer', + arch='l', + img_size=224, + patch_size=14, + drop_rate=0.1, + pre_norm=True, + final_norm=False, + init_cfg=dict( + type='Pretrained', + checkpoint='ckpt/openclip-ViT-L-14.pth', + prefix='backbone')), + neck=dict( + type='CLIPProjection', + in_channels=1024, + out_channels=768, + init_cfg=dict( + type='Pretrained', + checkpoint='ckpt/openclip-ViT-L-14.pth', + prefix='backbone')), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=768, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + init_cfg=None), + init_cfg=dict( + type='TruncNormal', layer=['Conv2d', 'Linear'], std=0.02, bias=0.0), + train_cfg=None) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=128, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stablediffusionV2-1-dpmsolver-25-1m.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/cc1m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=128, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV2-1-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=128, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV2-1-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/clip_large_pretrain_4x256_sdv2_lr3e-4' + diff --git a/clip_large_pretrain_4x256_sdv2_lr3e-4/20230606_050007/vis_data/config.py b/clip_large_pretrain_4x256_sdv2_lr3e-4/20230606_050007/vis_data/config.py new file mode 100644 index 0000000000000000000000000000000000000000..2fe3d2db91303588ef319725b2e1e94dca9ecb05 --- /dev/null +++ b/clip_large_pretrain_4x256_sdv2_lr3e-4/20230606_050007/vis_data/config.py @@ -0,0 +1,236 @@ +optim_wrapper = dict( + optimizer=dict( + type='AdamW', lr=0.0003, weight_decay=0.3, _scope_='mmpretrain'), + paramwise_cfg=dict( + custom_keys=dict({ + '.cls_token': dict(decay_mult=0.0), + '.pos_embed': dict(decay_mult=0.0) + })), + type='AmpOptimWrapper', + dtype='bfloat16', + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=False, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=4096) +model = dict( + type='ImageClassifier', + backbone=dict( + frozen_stages=24, + type='VisionTransformer', + arch='l', + img_size=224, + patch_size=14, + drop_rate=0.1, + pre_norm=True, + final_norm=False, + init_cfg=dict( + type='Pretrained', + checkpoint='ckpt/openclip-ViT-L-14.pth', + prefix='backbone')), + neck=dict( + type='CLIPProjection', + in_channels=1024, + out_channels=768, + init_cfg=dict( + type='Pretrained', + checkpoint='ckpt/openclip-ViT-L-14.pth', + prefix='backbone')), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=768, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + init_cfg=None), + init_cfg=dict( + type='TruncNormal', layer=['Conv2d', 'Linear'], std=0.02, bias=0.0), + train_cfg=None) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=128, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stablediffusionV2-1-dpmsolver-25-1m.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/cc1m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=128, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV2-1-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=128, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV2-1-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/clip_large_pretrain_4x256_sdv2_lr3e-4' diff --git a/clip_large_pretrain_4x256_sdv2_lr3e-4/20230606_052113/20230606_052113.log b/clip_large_pretrain_4x256_sdv2_lr3e-4/20230606_052113/20230606_052113.log new file mode 100644 index 0000000000000000000000000000000000000000..cf8385303e4ddcd3e5ded6322f3048418fe6fafa --- /dev/null +++ b/clip_large_pretrain_4x256_sdv2_lr3e-4/20230606_052113/20230606_052113.log @@ -0,0 +1,1727 @@ +2023/06/06 05:21:18 - mmengine - INFO - +------------------------------------------------------------ +System environment: + sys.platform: linux + Python: 3.10.9 (main, Mar 8 2023, 10:47:38) [GCC 11.2.0] + CUDA available: True + numpy_random_seed: 1258785886 + GPU 0,1,2,3: NVIDIA A100-SXM4-80GB + CUDA_HOME: /mnt/petrelfs/share/cuda-11.6 + NVCC: Cuda compilation tools, release 11.6, V11.6.124 + GCC: gcc (GCC) 7.5.0 + PyTorch: 1.13.1 + PyTorch compiling details: PyTorch built with: + - GCC 9.3 + - C++ Version: 201402 + - Intel(R) oneAPI Math Kernel Library Version 2021.4-Product Build 20210904 for Intel(R) 64 architecture applications + - Intel(R) MKL-DNN v2.6.0 (Git Hash 52b5f107dd9cf10910aaa19cb47f3abf9b349815) + - OpenMP 201511 (a.k.a. OpenMP 4.5) + - LAPACK is enabled (usually provided by MKL) + - NNPACK is enabled + - CPU capability usage: AVX2 + - CUDA Runtime 11.6 + - NVCC architecture flags: -gencode;arch=compute_37,code=sm_37;-gencode;arch=compute_50,code=sm_50;-gencode;arch=compute_60,code=sm_60;-gencode;arch=compute_61,code=sm_61;-gencode;arch=compute_70,code=sm_70;-gencode;arch=compute_75,code=sm_75;-gencode;arch=compute_80,code=sm_80;-gencode;arch=compute_86,code=sm_86;-gencode;arch=compute_37,code=compute_37 + - CuDNN 8.3.2 (built against CUDA 11.5) + - Magma 2.6.1 + - Build settings: BLAS_INFO=mkl, BUILD_TYPE=Release, CUDA_VERSION=11.6, CUDNN_VERSION=8.3.2, CXX_COMPILER=/opt/rh/devtoolset-9/root/usr/bin/c++, CXX_FLAGS= -fabi-version=11 -Wno-deprecated -fvisibility-inlines-hidden -DUSE_PTHREADPOOL -fopenmp -DNDEBUG -DUSE_KINETO -DUSE_FBGEMM -DUSE_QNNPACK -DUSE_PYTORCH_QNNPACK -DUSE_XNNPACK -DSYMBOLICATE_MOBILE_DEBUG_HANDLE -DEDGE_PROFILER_USE_KINETO -O2 -fPIC -Wno-narrowing -Wall -Wextra -Werror=return-type -Werror=non-virtual-dtor -Wno-missing-field-initializers -Wno-type-limits -Wno-array-bounds -Wno-unknown-pragmas -Wunused-local-typedefs -Wno-unused-parameter -Wno-unused-function -Wno-unused-result -Wno-strict-overflow -Wno-strict-aliasing -Wno-error=deprecated-declarations -Wno-stringop-overflow -Wno-psabi -Wno-error=pedantic -Wno-error=redundant-decls -Wno-error=old-style-cast -fdiagnostics-color=always -faligned-new -Wno-unused-but-set-variable -Wno-maybe-uninitialized -fno-math-errno -fno-trapping-math -Werror=format -Werror=cast-function-type -Wno-stringop-overflow, LAPACK_INFO=mkl, PERF_WITH_AVX=1, PERF_WITH_AVX2=1, PERF_WITH_AVX512=1, TORCH_VERSION=1.13.1, USE_CUDA=ON, USE_CUDNN=ON, USE_EXCEPTION_PTR=1, USE_GFLAGS=OFF, USE_GLOG=OFF, USE_MKL=ON, USE_MKLDNN=ON, USE_MPI=OFF, USE_NCCL=ON, USE_NNPACK=ON, USE_OPENMP=ON, USE_ROCM=OFF, + + TorchVision: 0.14.1 + OpenCV: 4.7.0 + MMEngine: 0.7.3 + +Runtime environment: + cudnn_benchmark: True + mp_cfg: {'mp_start_method': 'fork', 'opencv_num_threads': 0} + dist_cfg: {'backend': 'nccl'} + seed: None + deterministic: False + Distributed launcher: slurm + Distributed training: True + GPU number: 4 +------------------------------------------------------------ + +2023/06/06 05:21:22 - mmengine - INFO - Config: +optim_wrapper = dict( + optimizer=dict( + type='AdamW', lr=0.0003, weight_decay=0.3, _scope_='mmpretrain'), + paramwise_cfg=dict( + custom_keys=dict({ + '.cls_token': dict(decay_mult=0.0), + '.pos_embed': dict(decay_mult=0.0) + })), + type='AmpOptimWrapper', + dtype='bfloat16', + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=False, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=4096) +model = dict( + type='ImageClassifier', + backbone=dict( + frozen_stages=24, + type='VisionTransformer', + arch='l', + img_size=224, + patch_size=14, + drop_rate=0.1, + pre_norm=True, + final_norm=False, + init_cfg=dict( + type='Pretrained', + checkpoint='ckpt/openclip-ViT-L-14.pth', + prefix='backbone')), + neck=dict( + type='CLIPProjection', + in_channels=1024, + out_channels=768, + init_cfg=dict( + type='Pretrained', + checkpoint='ckpt/openclip-ViT-L-14.pth', + prefix='backbone')), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=768, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + init_cfg=None), + init_cfg=dict( + type='TruncNormal', layer=['Conv2d', 'Linear'], std=0.02, bias=0.0), + train_cfg=None) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=128, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stablediffusionV2-1-dpmsolver-25-1m.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/cc1m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=128, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV2-1-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=128, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV2-1-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/clip_large_pretrain_4x256_sdv2_lr3e-4' + +2023/06/06 05:21:36 - mmengine - INFO - Hooks will be executed in the following order: +before_run: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_load_checkpoint: +(ABOVE_NORMAL) EMAHook + -------------------- +before_train: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_train_epoch: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(NORMAL ) DistSamplerSeedHook + -------------------- +before_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook + -------------------- +after_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +after_train_epoch: +(NORMAL ) IterTimerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_val_epoch: +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook + -------------------- +before_val_iter: +(NORMAL ) IterTimerHook + -------------------- +after_val_iter: +(NORMAL ) IterTimerHook +(NORMAL ) VisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_val_epoch: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_save_checkpoint: +(ABOVE_NORMAL) EMAHook + -------------------- +after_train: +(VERY_LOW ) CheckpointHook + -------------------- +before_test_epoch: +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook + -------------------- +before_test_iter: +(NORMAL ) IterTimerHook + -------------------- +after_test_iter: +(NORMAL ) IterTimerHook +(NORMAL ) VisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_test_epoch: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_run: +(BELOW_NORMAL) LoggerHook + -------------------- +2023/06/06 05:21:55 - mmengine - INFO - load backbone in model from: ckpt/openclip-ViT-L-14.pth +2023/06/06 05:21:59 - mmengine - WARNING - The model and loaded state dict do not match exactly + +unexpected key in source state_dict: ln1.weight, ln1.bias + +2023/06/06 05:22:00 - mmengine - INFO - load backbone in model from: ckpt/openclip-ViT-L-14.pth +2023/06/06 05:22:01 - mmengine - WARNING - The model and loaded state dict do not match exactly + +unexpected key in source state_dict: cls_token, pos_embed, patch_embed.projection.weight, pre_norm.weight, pre_norm.bias, layers.0.ln1.weight, layers.0.ln1.bias, layers.0.attn.qkv.weight, layers.0.attn.qkv.bias, layers.0.attn.proj.weight, layers.0.attn.proj.bias, layers.0.ln2.weight, layers.0.ln2.bias, layers.0.ffn.layers.0.0.weight, layers.0.ffn.layers.0.0.bias, layers.0.ffn.layers.1.weight, layers.0.ffn.layers.1.bias, layers.1.ln1.weight, layers.1.ln1.bias, layers.1.attn.qkv.weight, layers.1.attn.qkv.bias, layers.1.attn.proj.weight, layers.1.attn.proj.bias, layers.1.ln2.weight, layers.1.ln2.bias, layers.1.ffn.layers.0.0.weight, layers.1.ffn.layers.0.0.bias, layers.1.ffn.layers.1.weight, layers.1.ffn.layers.1.bias, layers.2.ln1.weight, layers.2.ln1.bias, layers.2.attn.qkv.weight, layers.2.attn.qkv.bias, layers.2.attn.proj.weight, layers.2.attn.proj.bias, layers.2.ln2.weight, layers.2.ln2.bias, layers.2.ffn.layers.0.0.weight, layers.2.ffn.layers.0.0.bias, layers.2.ffn.layers.1.weight, layers.2.ffn.layers.1.bias, layers.3.ln1.weight, layers.3.ln1.bias, layers.3.attn.qkv.weight, layers.3.attn.qkv.bias, layers.3.attn.proj.weight, layers.3.attn.proj.bias, layers.3.ln2.weight, layers.3.ln2.bias, layers.3.ffn.layers.0.0.weight, layers.3.ffn.layers.0.0.bias, layers.3.ffn.layers.1.weight, layers.3.ffn.layers.1.bias, layers.4.ln1.weight, layers.4.ln1.bias, layers.4.attn.qkv.weight, layers.4.attn.qkv.bias, layers.4.attn.proj.weight, layers.4.attn.proj.bias, layers.4.ln2.weight, layers.4.ln2.bias, layers.4.ffn.layers.0.0.weight, layers.4.ffn.layers.0.0.bias, layers.4.ffn.layers.1.weight, layers.4.ffn.layers.1.bias, layers.5.ln1.weight, layers.5.ln1.bias, layers.5.attn.qkv.weight, layers.5.attn.qkv.bias, layers.5.attn.proj.weight, layers.5.attn.proj.bias, layers.5.ln2.weight, layers.5.ln2.bias, layers.5.ffn.layers.0.0.weight, layers.5.ffn.layers.0.0.bias, layers.5.ffn.layers.1.weight, layers.5.ffn.layers.1.bias, layers.6.ln1.weight, layers.6.ln1.bias, layers.6.attn.qkv.weight, layers.6.attn.qkv.bias, layers.6.attn.proj.weight, layers.6.attn.proj.bias, layers.6.ln2.weight, layers.6.ln2.bias, layers.6.ffn.layers.0.0.weight, layers.6.ffn.layers.0.0.bias, layers.6.ffn.layers.1.weight, layers.6.ffn.layers.1.bias, layers.7.ln1.weight, layers.7.ln1.bias, layers.7.attn.qkv.weight, layers.7.attn.qkv.bias, layers.7.attn.proj.weight, layers.7.attn.proj.bias, layers.7.ln2.weight, layers.7.ln2.bias, layers.7.ffn.layers.0.0.weight, layers.7.ffn.layers.0.0.bias, layers.7.ffn.layers.1.weight, layers.7.ffn.layers.1.bias, layers.8.ln1.weight, layers.8.ln1.bias, layers.8.attn.qkv.weight, layers.8.attn.qkv.bias, layers.8.attn.proj.weight, layers.8.attn.proj.bias, layers.8.ln2.weight, layers.8.ln2.bias, layers.8.ffn.layers.0.0.weight, layers.8.ffn.layers.0.0.bias, layers.8.ffn.layers.1.weight, layers.8.ffn.layers.1.bias, layers.9.ln1.weight, layers.9.ln1.bias, layers.9.attn.qkv.weight, layers.9.attn.qkv.bias, layers.9.attn.proj.weight, layers.9.attn.proj.bias, layers.9.ln2.weight, layers.9.ln2.bias, layers.9.ffn.layers.0.0.weight, layers.9.ffn.layers.0.0.bias, layers.9.ffn.layers.1.weight, layers.9.ffn.layers.1.bias, layers.10.ln1.weight, layers.10.ln1.bias, layers.10.attn.qkv.weight, layers.10.attn.qkv.bias, layers.10.attn.proj.weight, layers.10.attn.proj.bias, layers.10.ln2.weight, layers.10.ln2.bias, layers.10.ffn.layers.0.0.weight, layers.10.ffn.layers.0.0.bias, layers.10.ffn.layers.1.weight, layers.10.ffn.layers.1.bias, layers.11.ln1.weight, layers.11.ln1.bias, layers.11.attn.qkv.weight, layers.11.attn.qkv.bias, layers.11.attn.proj.weight, layers.11.attn.proj.bias, layers.11.ln2.weight, layers.11.ln2.bias, layers.11.ffn.layers.0.0.weight, layers.11.ffn.layers.0.0.bias, layers.11.ffn.layers.1.weight, layers.11.ffn.layers.1.bias, layers.12.ln1.weight, layers.12.ln1.bias, layers.12.attn.qkv.weight, layers.12.attn.qkv.bias, layers.12.attn.proj.weight, layers.12.attn.proj.bias, layers.12.ln2.weight, layers.12.ln2.bias, layers.12.ffn.layers.0.0.weight, layers.12.ffn.layers.0.0.bias, layers.12.ffn.layers.1.weight, layers.12.ffn.layers.1.bias, layers.13.ln1.weight, layers.13.ln1.bias, layers.13.attn.qkv.weight, layers.13.attn.qkv.bias, layers.13.attn.proj.weight, layers.13.attn.proj.bias, layers.13.ln2.weight, layers.13.ln2.bias, layers.13.ffn.layers.0.0.weight, layers.13.ffn.layers.0.0.bias, layers.13.ffn.layers.1.weight, layers.13.ffn.layers.1.bias, layers.14.ln1.weight, layers.14.ln1.bias, layers.14.attn.qkv.weight, layers.14.attn.qkv.bias, layers.14.attn.proj.weight, layers.14.attn.proj.bias, layers.14.ln2.weight, layers.14.ln2.bias, layers.14.ffn.layers.0.0.weight, layers.14.ffn.layers.0.0.bias, layers.14.ffn.layers.1.weight, layers.14.ffn.layers.1.bias, layers.15.ln1.weight, layers.15.ln1.bias, layers.15.attn.qkv.weight, layers.15.attn.qkv.bias, layers.15.attn.proj.weight, layers.15.attn.proj.bias, layers.15.ln2.weight, layers.15.ln2.bias, layers.15.ffn.layers.0.0.weight, layers.15.ffn.layers.0.0.bias, layers.15.ffn.layers.1.weight, layers.15.ffn.layers.1.bias, layers.16.ln1.weight, layers.16.ln1.bias, layers.16.attn.qkv.weight, layers.16.attn.qkv.bias, layers.16.attn.proj.weight, layers.16.attn.proj.bias, layers.16.ln2.weight, layers.16.ln2.bias, layers.16.ffn.layers.0.0.weight, layers.16.ffn.layers.0.0.bias, layers.16.ffn.layers.1.weight, layers.16.ffn.layers.1.bias, layers.17.ln1.weight, layers.17.ln1.bias, layers.17.attn.qkv.weight, layers.17.attn.qkv.bias, layers.17.attn.proj.weight, layers.17.attn.proj.bias, layers.17.ln2.weight, layers.17.ln2.bias, layers.17.ffn.layers.0.0.weight, layers.17.ffn.layers.0.0.bias, layers.17.ffn.layers.1.weight, layers.17.ffn.layers.1.bias, layers.18.ln1.weight, layers.18.ln1.bias, layers.18.attn.qkv.weight, layers.18.attn.qkv.bias, layers.18.attn.proj.weight, layers.18.attn.proj.bias, layers.18.ln2.weight, layers.18.ln2.bias, layers.18.ffn.layers.0.0.weight, layers.18.ffn.layers.0.0.bias, layers.18.ffn.layers.1.weight, layers.18.ffn.layers.1.bias, layers.19.ln1.weight, layers.19.ln1.bias, layers.19.attn.qkv.weight, layers.19.attn.qkv.bias, layers.19.attn.proj.weight, layers.19.attn.proj.bias, layers.19.ln2.weight, layers.19.ln2.bias, layers.19.ffn.layers.0.0.weight, layers.19.ffn.layers.0.0.bias, layers.19.ffn.layers.1.weight, layers.19.ffn.layers.1.bias, layers.20.ln1.weight, layers.20.ln1.bias, layers.20.attn.qkv.weight, layers.20.attn.qkv.bias, layers.20.attn.proj.weight, layers.20.attn.proj.bias, layers.20.ln2.weight, layers.20.ln2.bias, layers.20.ffn.layers.0.0.weight, layers.20.ffn.layers.0.0.bias, layers.20.ffn.layers.1.weight, layers.20.ffn.layers.1.bias, layers.21.ln1.weight, layers.21.ln1.bias, layers.21.attn.qkv.weight, layers.21.attn.qkv.bias, layers.21.attn.proj.weight, layers.21.attn.proj.bias, layers.21.ln2.weight, layers.21.ln2.bias, layers.21.ffn.layers.0.0.weight, layers.21.ffn.layers.0.0.bias, layers.21.ffn.layers.1.weight, layers.21.ffn.layers.1.bias, layers.22.ln1.weight, layers.22.ln1.bias, layers.22.attn.qkv.weight, layers.22.attn.qkv.bias, layers.22.attn.proj.weight, layers.22.attn.proj.bias, layers.22.ln2.weight, layers.22.ln2.bias, layers.22.ffn.layers.0.0.weight, layers.22.ffn.layers.0.0.bias, layers.22.ffn.layers.1.weight, layers.22.ffn.layers.1.bias, layers.23.ln1.weight, layers.23.ln1.bias, layers.23.attn.qkv.weight, layers.23.attn.qkv.bias, layers.23.attn.proj.weight, layers.23.attn.proj.bias, layers.23.ln2.weight, layers.23.ln2.bias, layers.23.ffn.layers.0.0.weight, layers.23.ffn.layers.0.0.bias, layers.23.ffn.layers.1.weight, layers.23.ffn.layers.1.bias, ln1.weight, ln1.bias + +missing keys in source state_dict: proj + +Name of parameter - Initialization information + +backbone.cls_token - torch.Size([1, 1, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.pos_embed - torch.Size([1, 257, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.patch_embed.projection.weight - torch.Size([1024, 3, 14, 14]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.0.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.0.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.0.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.0.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.0.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.0.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.0.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.0.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.0.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.0.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.0.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.0.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.1.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.1.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.1.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.1.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.1.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.1.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.1.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.1.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.1.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.1.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.1.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.1.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.2.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.2.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.2.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.2.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.2.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.2.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.2.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.2.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.2.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.2.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.2.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.2.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.3.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.3.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.3.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.3.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.3.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.3.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.3.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.3.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.3.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.3.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.3.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.3.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.4.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.4.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.4.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.4.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.4.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.4.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.4.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.4.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.4.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.4.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.4.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.4.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.5.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.5.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.5.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.5.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.5.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.5.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.5.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.5.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.5.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.5.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.5.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.5.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.6.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.6.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.6.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.6.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.6.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.6.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.6.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.6.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.6.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.6.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.6.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.6.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.7.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.7.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.7.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.7.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.7.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.7.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.7.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.7.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.7.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.7.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.7.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.7.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.8.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.8.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.8.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.8.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.8.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.8.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.8.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.8.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.8.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.8.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.8.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.8.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.9.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.9.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.9.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.9.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.9.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.9.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.9.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.9.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.9.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.9.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.9.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.9.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.10.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.10.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.10.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.10.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.10.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.10.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.10.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.10.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.10.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.10.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.10.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.10.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.11.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.11.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.11.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.11.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.11.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.11.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.11.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.11.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.11.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.11.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.11.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.11.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.12.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.12.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.12.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.12.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.12.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.12.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.12.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.12.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.12.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.12.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.12.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.12.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.13.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.13.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.13.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.13.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.13.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.13.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.13.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.13.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.13.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.13.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.13.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.13.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.14.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.14.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.14.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.14.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.14.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.14.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.14.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.14.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.14.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.14.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.14.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.14.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.15.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.15.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.15.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.15.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.15.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.15.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.15.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.15.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.15.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.15.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.15.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.15.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.16.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.16.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.16.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.16.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.16.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.16.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.16.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.16.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.16.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.16.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.16.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.16.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.17.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.17.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.17.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.17.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.17.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.17.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.17.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.17.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.17.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.17.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.17.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.17.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.18.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.18.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.18.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.18.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.18.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.18.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.18.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.18.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.18.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.18.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.18.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.18.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.19.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.19.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.19.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.19.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.19.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.19.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.19.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.19.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.19.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.19.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.19.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.19.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.20.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.20.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.20.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.20.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.20.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.20.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.20.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.20.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.20.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.20.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.20.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.20.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.21.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.21.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.21.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.21.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.21.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.21.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.21.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.21.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.21.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.21.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.21.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.21.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.22.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.22.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.22.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.22.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.22.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.22.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.22.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.22.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.22.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.22.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.22.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.22.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.23.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.23.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.23.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.23.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.23.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.23.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.23.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.23.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.23.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.23.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.23.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.23.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.pre_norm.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.pre_norm.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +neck.proj - torch.Size([1024, 768]): +The value is the same before and after calling `init_weights` of ImageClassifier + +head.fc.weight - torch.Size([2, 768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +head.fc.bias - torch.Size([2]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 +2023/06/06 05:22:01 - mmengine - WARNING - "FileClient" will be deprecated in future. Please use io functions in https://mmengine.readthedocs.io/en/latest/api/fileio.html#file-io +2023/06/06 05:22:01 - mmengine - WARNING - "HardDiskBackend" is the alias of "LocalBackend" and the former will be deprecated in future. +2023/06/06 05:22:01 - mmengine - INFO - Checkpoints will be saved to /mnt/petrelfs/luzeyu/workspace/fakebench/mmpretrain/workdir/clip_large_pretrain_4x256_sdv2_lr3e-4. +2023/06/06 05:23:13 - mmengine - INFO - Epoch(train) [1][ 100/3907] lr: 3.0000e-04 eta: 7:43:35 time: 0.6310 data_time: 0.0017 memory: 44138 loss: 0.4902 +2023/06/06 05:24:16 - mmengine - INFO - Epoch(train) [1][ 200/3907] lr: 2.9998e-04 eta: 7:15:47 time: 0.6329 data_time: 0.0016 memory: 44138 loss: 0.4914 +2023/06/06 05:25:19 - mmengine - INFO - Epoch(train) [1][ 300/3907] lr: 2.9996e-04 eta: 7:06:28 time: 0.6324 data_time: 0.0015 memory: 44138 loss: 0.4494 +2023/06/06 05:26:23 - mmengine - INFO - Epoch(train) [1][ 400/3907] lr: 2.9993e-04 eta: 7:00:59 time: 0.6327 data_time: 0.0017 memory: 44138 loss: 0.4426 +2023/06/06 05:27:26 - mmengine - INFO - Epoch(train) [1][ 500/3907] lr: 2.9988e-04 eta: 6:57:17 time: 0.6334 data_time: 0.0015 memory: 44138 loss: 0.4519 +2023/06/06 05:28:29 - mmengine - INFO - Epoch(train) [1][ 600/3907] lr: 2.9983e-04 eta: 6:54:31 time: 0.6340 data_time: 0.0015 memory: 44138 loss: 0.4173 +2023/06/06 05:29:33 - mmengine - INFO - Epoch(train) [1][ 700/3907] lr: 2.9977e-04 eta: 6:52:16 time: 0.6324 data_time: 0.0016 memory: 44138 loss: 0.4038 +2023/06/06 05:30:36 - mmengine - INFO - Epoch(train) [1][ 800/3907] lr: 2.9970e-04 eta: 6:50:15 time: 0.6337 data_time: 0.0015 memory: 44138 loss: 0.4298 +2023/06/06 05:31:39 - mmengine - INFO - Epoch(train) [1][ 900/3907] lr: 2.9962e-04 eta: 6:48:27 time: 0.6332 data_time: 0.0015 memory: 44138 loss: 0.4194 +2023/06/06 05:32:43 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr3e-4_20230606_052113 +2023/06/06 05:32:43 - mmengine - INFO - Epoch(train) [1][1000/3907] lr: 2.9953e-04 eta: 6:46:49 time: 0.6328 data_time: 0.0015 memory: 44138 loss: 0.4238 +2023/06/06 05:33:46 - mmengine - INFO - Epoch(train) [1][1100/3907] lr: 2.9943e-04 eta: 6:45:18 time: 0.6326 data_time: 0.0015 memory: 44138 loss: 0.4083 +2023/06/06 05:36:10 - mmengine - INFO - Epoch(train) [1][1200/3907] lr: 2.9933e-04 eta: 7:26:02 time: 0.6305 data_time: 0.0014 memory: 44138 loss: 0.4097 +2023/06/06 05:37:13 - mmengine - INFO - Epoch(train) [1][1300/3907] lr: 2.9921e-04 eta: 7:21:17 time: 0.6327 data_time: 0.0014 memory: 44138 loss: 0.3748 +2023/06/06 05:38:16 - mmengine - INFO - Epoch(train) [1][1400/3907] lr: 2.9908e-04 eta: 7:17:07 time: 0.6336 data_time: 0.0015 memory: 44138 loss: 0.3998 +2023/06/06 05:39:20 - mmengine - INFO - Epoch(train) [1][1500/3907] lr: 2.9895e-04 eta: 7:13:20 time: 0.6327 data_time: 0.0016 memory: 44138 loss: 0.3972 +2023/06/06 05:40:23 - mmengine - INFO - Epoch(train) [1][1600/3907] lr: 2.9880e-04 eta: 7:09:52 time: 0.6335 data_time: 0.0016 memory: 44138 loss: 0.4123 +2023/06/06 05:41:26 - mmengine - INFO - Epoch(train) [1][1700/3907] lr: 2.9865e-04 eta: 7:06:42 time: 0.6330 data_time: 0.0016 memory: 44138 loss: 0.4265 +2023/06/06 05:42:30 - mmengine - INFO - Epoch(train) [1][1800/3907] lr: 2.9849e-04 eta: 7:03:51 time: 0.6333 data_time: 0.0017 memory: 44138 loss: 0.3921 +2023/06/06 05:43:33 - mmengine - INFO - Epoch(train) [1][1900/3907] lr: 2.9831e-04 eta: 7:01:10 time: 0.6330 data_time: 0.0014 memory: 44138 loss: 0.3964 +2023/06/06 05:44:37 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr3e-4_20230606_052113 +2023/06/06 05:44:37 - mmengine - INFO - Epoch(train) [1][2000/3907] lr: 2.9813e-04 eta: 6:58:36 time: 0.6349 data_time: 0.0014 memory: 44138 loss: 0.3734 +2023/06/06 05:45:40 - mmengine - INFO - Epoch(train) [1][2100/3907] lr: 2.9794e-04 eta: 6:56:10 time: 0.6333 data_time: 0.0014 memory: 44138 loss: 0.3939 +2023/06/06 05:46:43 - mmengine - INFO - Epoch(train) [1][2200/3907] lr: 2.9774e-04 eta: 6:53:53 time: 0.6339 data_time: 0.0015 memory: 44138 loss: 0.3556 +2023/06/06 05:47:47 - mmengine - INFO - Epoch(train) [1][2300/3907] lr: 2.9753e-04 eta: 6:51:40 time: 0.6328 data_time: 0.0015 memory: 44138 loss: 0.3780 +2023/06/06 05:48:50 - mmengine - INFO - Epoch(train) [1][2400/3907] lr: 2.9731e-04 eta: 6:49:34 time: 0.6330 data_time: 0.0014 memory: 44138 loss: 0.3937 +2023/06/06 05:49:53 - mmengine - INFO - Epoch(train) [1][2500/3907] lr: 2.9708e-04 eta: 6:47:33 time: 0.6332 data_time: 0.0014 memory: 44138 loss: 0.3939 +2023/06/06 05:50:57 - mmengine - INFO - Epoch(train) [1][2600/3907] lr: 2.9685e-04 eta: 6:45:36 time: 0.6337 data_time: 0.0014 memory: 44138 loss: 0.3750 +2023/06/06 05:52:00 - mmengine - INFO - Epoch(train) [1][2700/3907] lr: 2.9660e-04 eta: 6:43:44 time: 0.6326 data_time: 0.0015 memory: 44138 loss: 0.3766 +2023/06/06 05:53:03 - mmengine - INFO - Epoch(train) [1][2800/3907] lr: 2.9634e-04 eta: 6:41:55 time: 0.6333 data_time: 0.0015 memory: 44138 loss: 0.3962 +2023/06/06 05:54:07 - mmengine - INFO - Epoch(train) [1][2900/3907] lr: 2.9608e-04 eta: 6:40:09 time: 0.6331 data_time: 0.0015 memory: 44138 loss: 0.3758 +2023/06/06 05:55:13 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr3e-4_20230606_052113 +2023/06/06 05:55:13 - mmengine - INFO - Epoch(train) [1][3000/3907] lr: 2.9580e-04 eta: 6:39:08 time: 0.6327 data_time: 0.0016 memory: 44138 loss: 0.3716 +2023/06/06 05:56:17 - mmengine - INFO - Epoch(train) [1][3100/3907] lr: 2.9552e-04 eta: 6:37:25 time: 0.6324 data_time: 0.0015 memory: 44138 loss: 0.3768 +2023/06/06 05:57:20 - mmengine - INFO - Epoch(train) [1][3200/3907] lr: 2.9523e-04 eta: 6:35:45 time: 0.6326 data_time: 0.0016 memory: 44138 loss: 0.3876 +2023/06/06 05:58:23 - mmengine - INFO - Epoch(train) [1][3300/3907] lr: 2.9493e-04 eta: 6:34:07 time: 0.6327 data_time: 0.0015 memory: 44138 loss: 0.3849 +2023/06/06 05:59:26 - mmengine - INFO - Epoch(train) [1][3400/3907] lr: 2.9462e-04 eta: 6:32:31 time: 0.6326 data_time: 0.0015 memory: 44138 loss: 0.3955 +2023/06/06 06:00:30 - mmengine - INFO - Epoch(train) [1][3500/3907] lr: 2.9430e-04 eta: 6:30:57 time: 0.6329 data_time: 0.0015 memory: 44138 loss: 0.3836 +2023/06/06 06:01:33 - mmengine - INFO - Epoch(train) [1][3600/3907] lr: 2.9397e-04 eta: 6:29:26 time: 0.6334 data_time: 0.0016 memory: 44138 loss: 0.3765 +2023/06/06 06:02:36 - mmengine - INFO - Epoch(train) [1][3700/3907] lr: 2.9363e-04 eta: 6:27:55 time: 0.6333 data_time: 0.0016 memory: 44138 loss: 0.3621 +2023/06/06 06:03:40 - mmengine - INFO - Epoch(train) [1][3800/3907] lr: 2.9329e-04 eta: 6:26:26 time: 0.6321 data_time: 0.0016 memory: 44138 loss: 0.3587 +2023/06/06 06:04:43 - mmengine - INFO - Epoch(train) [1][3900/3907] lr: 2.9293e-04 eta: 6:24:58 time: 0.6322 data_time: 0.0012 memory: 44138 loss: 0.3765 +2023/06/06 06:04:47 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr3e-4_20230606_052113 +2023/06/06 06:04:47 - mmengine - INFO - Saving checkpoint at 1 epochs +2023/06/06 06:06:34 - mmengine - INFO - Epoch(val) [1][57/57] accuracy/top1: 65.9059 single-label/precision_classwise: [61.83221435546875, 98.3172378540039] single-label/recall_classwise: [99.65911865234375, 24.45736312866211] single-label/f1-score_classwise: [76.31547546386719, 39.170650482177734] data_time: 0.0454 time: 1.3536 +2023/06/06 06:07:36 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr3e-4_20230606_052113 +2023/06/06 06:07:41 - mmengine - INFO - Epoch(train) [2][ 100/3907] lr: 2.9254e-04 eta: 6:23:50 time: 0.6337 data_time: 0.0017 memory: 44138 loss: 0.3807 +2023/06/06 06:08:44 - mmengine - INFO - Epoch(train) [2][ 200/3907] lr: 2.9217e-04 eta: 6:22:25 time: 0.6347 data_time: 0.0015 memory: 44137 loss: 0.3625 +2023/06/06 06:09:48 - mmengine - INFO - Epoch(train) [2][ 300/3907] lr: 2.9179e-04 eta: 6:21:01 time: 0.6334 data_time: 0.0015 memory: 44137 loss: 0.3828 +2023/06/06 06:10:51 - mmengine - INFO - Epoch(train) [2][ 400/3907] lr: 2.9139e-04 eta: 6:19:38 time: 0.6332 data_time: 0.0015 memory: 44137 loss: 0.3831 +2023/06/06 06:11:54 - mmengine - INFO - Epoch(train) [2][ 500/3907] lr: 2.9099e-04 eta: 6:18:16 time: 0.6360 data_time: 0.0014 memory: 44137 loss: 0.3692 +2023/06/06 06:12:58 - mmengine - INFO - Epoch(train) [2][ 600/3907] lr: 2.9059e-04 eta: 6:16:55 time: 0.6345 data_time: 0.0014 memory: 44137 loss: 0.3525 +2023/06/06 06:14:01 - mmengine - INFO - Epoch(train) [2][ 700/3907] lr: 2.9017e-04 eta: 6:15:34 time: 0.6338 data_time: 0.0015 memory: 44137 loss: 0.3589 +2023/06/06 06:15:05 - mmengine - INFO - Epoch(train) [2][ 800/3907] lr: 2.8974e-04 eta: 6:14:13 time: 0.6326 data_time: 0.0014 memory: 44137 loss: 0.3609 +2023/06/06 06:16:08 - mmengine - INFO - Epoch(train) [2][ 900/3907] lr: 2.8931e-04 eta: 6:12:54 time: 0.6342 data_time: 0.0015 memory: 44137 loss: 0.3747 +2023/06/06 06:17:11 - mmengine - INFO - Epoch(train) [2][1000/3907] lr: 2.8886e-04 eta: 6:11:35 time: 0.6345 data_time: 0.0014 memory: 44137 loss: 0.3882 +2023/06/06 06:18:10 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr3e-4_20230606_052113 +2023/06/06 06:18:15 - mmengine - INFO - Epoch(train) [2][1100/3907] lr: 2.8841e-04 eta: 6:10:17 time: 0.6332 data_time: 0.0019 memory: 44137 loss: 0.3529 +2023/06/06 06:19:18 - mmengine - INFO - Epoch(train) [2][1200/3907] lr: 2.8795e-04 eta: 6:08:59 time: 0.6331 data_time: 0.0014 memory: 44137 loss: 0.3874 +2023/06/06 06:20:21 - mmengine - INFO - Epoch(train) [2][1300/3907] lr: 2.8748e-04 eta: 6:07:42 time: 0.6332 data_time: 0.0014 memory: 44137 loss: 0.3844 +2023/06/06 06:21:25 - mmengine - INFO - Epoch(train) [2][1400/3907] lr: 2.8700e-04 eta: 6:06:26 time: 0.6331 data_time: 0.0014 memory: 44137 loss: 0.3736 +2023/06/06 06:22:28 - mmengine - INFO - Epoch(train) [2][1500/3907] lr: 2.8651e-04 eta: 6:05:10 time: 0.6337 data_time: 0.0014 memory: 44137 loss: 0.3806 +2023/06/06 06:23:31 - mmengine - INFO - Epoch(train) [2][1600/3907] lr: 2.8602e-04 eta: 6:03:54 time: 0.6350 data_time: 0.0013 memory: 44137 loss: 0.3710 +2023/06/06 06:24:35 - mmengine - INFO - Epoch(train) [2][1700/3907] lr: 2.8552e-04 eta: 6:02:40 time: 0.6440 data_time: 0.0015 memory: 44137 loss: 0.3771 +2023/06/06 06:25:38 - mmengine - INFO - Epoch(train) [2][1800/3907] lr: 2.8500e-04 eta: 6:01:25 time: 0.6335 data_time: 0.0015 memory: 44137 loss: 0.3843 +2023/06/06 06:26:42 - mmengine - INFO - Epoch(train) [2][1900/3907] lr: 2.8448e-04 eta: 6:00:11 time: 0.6334 data_time: 0.0016 memory: 44137 loss: 0.4046 +2023/06/06 06:27:45 - mmengine - INFO - Epoch(train) [2][2000/3907] lr: 2.8395e-04 eta: 5:58:57 time: 0.6342 data_time: 0.0014 memory: 44137 loss: 0.3640 +2023/06/06 06:28:44 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr3e-4_20230606_052113 +2023/06/06 06:28:48 - mmengine - INFO - Epoch(train) [2][2100/3907] lr: 2.8342e-04 eta: 5:57:43 time: 0.6336 data_time: 0.0015 memory: 44137 loss: 0.3548 +2023/06/06 06:29:52 - mmengine - INFO - Epoch(train) [2][2200/3907] lr: 2.8287e-04 eta: 5:56:30 time: 0.6333 data_time: 0.0014 memory: 44137 loss: 0.3748 +2023/06/06 06:30:55 - mmengine - INFO - Epoch(train) [2][2300/3907] lr: 2.8232e-04 eta: 5:55:17 time: 0.6332 data_time: 0.0014 memory: 44137 loss: 0.3613 +2023/06/06 06:31:59 - mmengine - INFO - Epoch(train) [2][2400/3907] lr: 2.8176e-04 eta: 5:54:05 time: 0.6347 data_time: 0.0015 memory: 44137 loss: 0.3847 +2023/06/06 06:33:02 - mmengine - INFO - Epoch(train) [2][2500/3907] lr: 2.8119e-04 eta: 5:52:53 time: 0.6333 data_time: 0.0014 memory: 44137 loss: 0.3942 +2023/06/06 06:34:06 - mmengine - INFO - Epoch(train) [2][2600/3907] lr: 2.8061e-04 eta: 5:51:41 time: 0.6330 data_time: 0.0014 memory: 44137 loss: 0.3432 +2023/06/06 06:35:09 - mmengine - INFO - Epoch(train) [2][2700/3907] lr: 2.8002e-04 eta: 5:50:30 time: 0.6329 data_time: 0.0013 memory: 44137 loss: 0.4181 +2023/06/06 06:36:12 - mmengine - INFO - Epoch(train) [2][2800/3907] lr: 2.7943e-04 eta: 5:49:18 time: 0.6330 data_time: 0.0014 memory: 44137 loss: 0.3541 +2023/06/06 06:37:16 - mmengine - INFO - Epoch(train) [2][2900/3907] lr: 2.7882e-04 eta: 5:48:06 time: 0.6331 data_time: 0.0015 memory: 44137 loss: 0.3673 +2023/06/06 06:38:19 - mmengine - INFO - Epoch(train) [2][3000/3907] lr: 2.7821e-04 eta: 5:46:55 time: 0.6329 data_time: 0.0015 memory: 44137 loss: 0.3860 +2023/06/06 06:39:18 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr3e-4_20230606_052113 +2023/06/06 06:39:22 - mmengine - INFO - Epoch(train) [2][3100/3907] lr: 2.7759e-04 eta: 5:45:44 time: 0.6331 data_time: 0.0016 memory: 44137 loss: 0.3611 +2023/06/06 06:40:26 - mmengine - INFO - Epoch(train) [2][3200/3907] lr: 2.7697e-04 eta: 5:44:33 time: 0.6328 data_time: 0.0014 memory: 44137 loss: 0.3986 +2023/06/06 06:41:29 - mmengine - INFO - Epoch(train) [2][3300/3907] lr: 2.7633e-04 eta: 5:43:23 time: 0.6332 data_time: 0.0013 memory: 44137 loss: 0.3689 +2023/06/06 06:42:33 - mmengine - INFO - Epoch(train) [2][3400/3907] lr: 2.7569e-04 eta: 5:42:13 time: 0.6328 data_time: 0.0014 memory: 44137 loss: 0.3703 +2023/06/06 06:43:36 - mmengine - INFO - Epoch(train) [2][3500/3907] lr: 2.7504e-04 eta: 5:41:03 time: 0.6328 data_time: 0.0014 memory: 44137 loss: 0.4029 +2023/06/06 06:44:39 - mmengine - INFO - Epoch(train) [2][3600/3907] lr: 2.7438e-04 eta: 5:39:53 time: 0.6334 data_time: 0.0014 memory: 44137 loss: 0.3923 +2023/06/06 06:45:43 - mmengine - INFO - Epoch(train) [2][3700/3907] lr: 2.7372e-04 eta: 5:38:43 time: 0.6332 data_time: 0.0015 memory: 44137 loss: 0.3727 +2023/06/06 06:46:46 - mmengine - INFO - Epoch(train) [2][3800/3907] lr: 2.7304e-04 eta: 5:37:34 time: 0.6326 data_time: 0.0014 memory: 44137 loss: 0.3905 +2023/06/06 06:47:49 - mmengine - INFO - Epoch(train) [2][3900/3907] lr: 2.7236e-04 eta: 5:36:24 time: 0.6329 data_time: 0.0014 memory: 44137 loss: 0.3634 +2023/06/06 06:47:53 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr3e-4_20230606_052113 +2023/06/06 06:47:53 - mmengine - INFO - Saving checkpoint at 2 epochs +2023/06/06 06:49:35 - mmengine - INFO - Epoch(val) [2][57/57] accuracy/top1: 73.0350 single-label/precision_classwise: [67.38280487060547, 97.09217834472656] single-label/recall_classwise: [98.99627685546875, 41.15503692626953] single-label/f1-score_classwise: [80.18611907958984, 57.80705261230469] data_time: 0.0413 time: 1.2851 +2023/06/06 06:50:41 - mmengine - INFO - Epoch(train) [3][ 100/3907] lr: 2.7163e-04 eta: 5:35:21 time: 0.6344 data_time: 0.0015 memory: 44137 loss: 0.3904 +2023/06/06 06:51:36 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr3e-4_20230606_052113 +2023/06/06 06:51:45 - mmengine - INFO - Epoch(train) [3][ 200/3907] lr: 2.7093e-04 eta: 5:34:12 time: 0.6343 data_time: 0.0016 memory: 44137 loss: 0.3502 +2023/06/06 06:52:48 - mmengine - INFO - Epoch(train) [3][ 300/3907] lr: 2.7022e-04 eta: 5:33:03 time: 0.6333 data_time: 0.0015 memory: 44137 loss: 0.3972 +2023/06/06 06:53:52 - mmengine - INFO - Epoch(train) [3][ 400/3907] lr: 2.6951e-04 eta: 5:31:55 time: 0.6353 data_time: 0.0016 memory: 44137 loss: 0.3641 +2023/06/06 06:54:55 - mmengine - INFO - Epoch(train) [3][ 500/3907] lr: 2.6879e-04 eta: 5:30:46 time: 0.6347 data_time: 0.0015 memory: 44137 loss: 0.4046 +2023/06/06 06:55:59 - mmengine - INFO - Epoch(train) [3][ 600/3907] lr: 2.6807e-04 eta: 5:29:38 time: 0.6367 data_time: 0.0014 memory: 44137 loss: 0.3757 +2023/06/06 06:57:02 - mmengine - INFO - Epoch(train) [3][ 700/3907] lr: 2.6733e-04 eta: 5:28:30 time: 0.6332 data_time: 0.0015 memory: 44137 loss: 0.3754 +2023/06/06 06:58:06 - mmengine - INFO - Epoch(train) [3][ 800/3907] lr: 2.6659e-04 eta: 5:27:22 time: 0.6426 data_time: 0.0015 memory: 44137 loss: 0.3896 +2023/06/06 06:59:09 - mmengine - INFO - Epoch(train) [3][ 900/3907] lr: 2.6585e-04 eta: 5:26:14 time: 0.6330 data_time: 0.0014 memory: 44137 loss: 0.3795 +2023/06/06 07:00:12 - mmengine - INFO - Epoch(train) [3][1000/3907] lr: 2.6509e-04 eta: 5:25:05 time: 0.6335 data_time: 0.0014 memory: 44137 loss: 0.3862 +2023/06/06 07:01:16 - mmengine - INFO - Epoch(train) [3][1100/3907] lr: 2.6433e-04 eta: 5:23:57 time: 0.6337 data_time: 0.0014 memory: 44137 loss: 0.4060 +2023/06/06 07:02:10 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr3e-4_20230606_052113 +2023/06/06 07:02:19 - mmengine - INFO - Epoch(train) [3][1200/3907] lr: 2.6356e-04 eta: 5:22:49 time: 0.6342 data_time: 0.0015 memory: 44137 loss: 0.3851 +2023/06/06 07:03:23 - mmengine - INFO - Epoch(train) [3][1300/3907] lr: 2.6278e-04 eta: 5:21:41 time: 0.6333 data_time: 0.0014 memory: 44137 loss: 0.3818 +2023/06/06 07:04:26 - mmengine - INFO - Epoch(train) [3][1400/3907] lr: 2.6200e-04 eta: 5:20:34 time: 0.6335 data_time: 0.0014 memory: 44137 loss: 0.3861 +2023/06/06 07:05:29 - mmengine - INFO - Epoch(train) [3][1500/3907] lr: 2.6121e-04 eta: 5:19:26 time: 0.6337 data_time: 0.0016 memory: 44137 loss: 0.4049 +2023/06/06 07:06:33 - mmengine - INFO - Epoch(train) [3][1600/3907] lr: 2.6041e-04 eta: 5:18:18 time: 0.6340 data_time: 0.0015 memory: 44137 loss: 0.3932 +2023/06/06 07:07:36 - mmengine - INFO - Epoch(train) [3][1700/3907] lr: 2.5961e-04 eta: 5:17:11 time: 0.6334 data_time: 0.0016 memory: 44137 loss: 0.3619 +2023/06/06 07:08:40 - mmengine - INFO - Epoch(train) [3][1800/3907] lr: 2.5880e-04 eta: 5:16:03 time: 0.6326 data_time: 0.0015 memory: 44137 loss: 0.3690 +2023/06/06 07:09:43 - mmengine - INFO - Epoch(train) [3][1900/3907] lr: 2.5798e-04 eta: 5:14:56 time: 0.6339 data_time: 0.0013 memory: 44137 loss: 0.3662 +2023/06/06 07:10:46 - mmengine - INFO - Epoch(train) [3][2000/3907] lr: 2.5715e-04 eta: 5:13:49 time: 0.6333 data_time: 0.0014 memory: 44137 loss: 0.4190 +2023/06/06 07:11:50 - mmengine - INFO - Epoch(train) [3][2100/3907] lr: 2.5632e-04 eta: 5:12:41 time: 0.6336 data_time: 0.0014 memory: 44137 loss: 0.4015 +2023/06/06 07:12:44 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr3e-4_20230606_052113 +2023/06/06 07:12:53 - mmengine - INFO - Epoch(train) [3][2200/3907] lr: 2.5549e-04 eta: 5:11:35 time: 0.6363 data_time: 0.0014 memory: 44137 loss: 0.3521 +2023/06/06 07:13:57 - mmengine - INFO - Epoch(train) [3][2300/3907] lr: 2.5464e-04 eta: 5:10:27 time: 0.6344 data_time: 0.0014 memory: 44137 loss: 0.3942 +2023/06/06 07:15:00 - mmengine - INFO - Epoch(train) [3][2400/3907] lr: 2.5379e-04 eta: 5:09:20 time: 0.6330 data_time: 0.0014 memory: 44137 loss: 0.3686 +2023/06/06 07:16:03 - mmengine - INFO - Epoch(train) [3][2500/3907] lr: 2.5294e-04 eta: 5:08:13 time: 0.6346 data_time: 0.0014 memory: 44137 loss: 0.3932 +2023/06/06 07:17:07 - mmengine - INFO - Epoch(train) [3][2600/3907] lr: 2.5207e-04 eta: 5:07:07 time: 0.6329 data_time: 0.0014 memory: 44137 loss: 0.3807 +2023/06/06 07:18:10 - mmengine - INFO - Epoch(train) [3][2700/3907] lr: 2.5120e-04 eta: 5:06:00 time: 0.6345 data_time: 0.0015 memory: 44137 loss: 0.3918 +2023/06/06 07:19:14 - mmengine - INFO - Epoch(train) [3][2800/3907] lr: 2.5033e-04 eta: 5:04:53 time: 0.6345 data_time: 0.0014 memory: 44137 loss: 0.4117 +2023/06/06 07:20:17 - mmengine - INFO - Epoch(train) [3][2900/3907] lr: 2.4945e-04 eta: 5:03:47 time: 0.6331 data_time: 0.0015 memory: 44137 loss: 0.3975 +2023/06/06 07:21:20 - mmengine - INFO - Epoch(train) [3][3000/3907] lr: 2.4856e-04 eta: 5:02:40 time: 0.6353 data_time: 0.0014 memory: 44137 loss: 0.3954 +2023/06/06 07:22:24 - mmengine - INFO - Epoch(train) [3][3100/3907] lr: 2.4767e-04 eta: 5:01:33 time: 0.6332 data_time: 0.0014 memory: 44137 loss: 0.3588 +2023/06/06 07:23:18 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr3e-4_20230606_052113 +2023/06/06 07:23:27 - mmengine - INFO - Epoch(train) [3][3200/3907] lr: 2.4677e-04 eta: 5:00:27 time: 0.6330 data_time: 0.0014 memory: 44137 loss: 0.4098 +2023/06/06 07:24:30 - mmengine - INFO - Epoch(train) [3][3300/3907] lr: 2.4586e-04 eta: 4:59:20 time: 0.6333 data_time: 0.0014 memory: 44137 loss: 0.4026 +2023/06/06 07:25:34 - mmengine - INFO - Epoch(train) [3][3400/3907] lr: 2.4495e-04 eta: 4:58:14 time: 0.6329 data_time: 0.0016 memory: 44137 loss: 0.3887 +2023/06/06 07:26:37 - mmengine - INFO - Epoch(train) [3][3500/3907] lr: 2.4403e-04 eta: 4:57:07 time: 0.6340 data_time: 0.0015 memory: 44137 loss: 0.3891 +2023/06/06 07:27:41 - mmengine - INFO - Epoch(train) [3][3600/3907] lr: 2.4311e-04 eta: 4:56:01 time: 0.6328 data_time: 0.0015 memory: 44137 loss: 0.4167 +2023/06/06 07:28:44 - mmengine - INFO - Epoch(train) [3][3700/3907] lr: 2.4218e-04 eta: 4:54:55 time: 0.6338 data_time: 0.0015 memory: 44137 loss: 0.3952 +2023/06/06 07:29:48 - mmengine - INFO - Epoch(train) [3][3800/3907] lr: 2.4124e-04 eta: 4:53:49 time: 0.6332 data_time: 0.0016 memory: 44137 loss: 0.3703 +2023/06/06 07:30:51 - mmengine - INFO - Epoch(train) [3][3900/3907] lr: 2.4030e-04 eta: 4:52:43 time: 0.6336 data_time: 0.0011 memory: 44137 loss: 0.4004 +2023/06/06 07:30:55 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr3e-4_20230606_052113 +2023/06/06 07:30:55 - mmengine - INFO - Saving checkpoint at 3 epochs +2023/06/06 07:32:37 - mmengine - INFO - Epoch(val) [3][57/57] accuracy/top1: 73.6474 single-label/precision_classwise: [67.92108917236328, 96.91683959960938] single-label/recall_classwise: [98.89527130126953, 42.643409729003906] single-label/f1-score_classwise: [80.53256225585938, 59.226959228515625] data_time: 0.0359 time: 1.2846 +2023/06/06 07:33:43 - mmengine - INFO - Epoch(train) [4][ 100/3907] lr: 2.3929e-04 eta: 4:51:39 time: 0.6335 data_time: 0.0016 memory: 44137 loss: 0.3860 +2023/06/06 07:34:47 - mmengine - INFO - Epoch(train) [4][ 200/3907] lr: 2.3834e-04 eta: 4:50:33 time: 0.6331 data_time: 0.0015 memory: 44137 loss: 0.3928 +2023/06/06 07:35:37 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr3e-4_20230606_052113 +2023/06/06 07:35:50 - mmengine - INFO - Epoch(train) [4][ 300/3907] lr: 2.3738e-04 eta: 4:49:27 time: 0.6350 data_time: 0.0015 memory: 44137 loss: 0.4065 +2023/06/06 07:36:54 - mmengine - INFO - Epoch(train) [4][ 400/3907] lr: 2.3642e-04 eta: 4:48:21 time: 0.6337 data_time: 0.0015 memory: 44137 loss: 0.3975 +2023/06/06 07:37:57 - mmengine - INFO - Epoch(train) [4][ 500/3907] lr: 2.3545e-04 eta: 4:47:15 time: 0.6337 data_time: 0.0016 memory: 44137 loss: 0.3993 +2023/06/06 07:39:01 - mmengine - INFO - Epoch(train) [4][ 600/3907] lr: 2.3448e-04 eta: 4:46:09 time: 0.6341 data_time: 0.0015 memory: 44137 loss: 0.3807 +2023/06/06 07:40:04 - mmengine - INFO - Epoch(train) [4][ 700/3907] lr: 2.3350e-04 eta: 4:45:03 time: 0.6345 data_time: 0.0014 memory: 44137 loss: 0.3722 +2023/06/06 07:41:07 - mmengine - INFO - Epoch(train) [4][ 800/3907] lr: 2.3252e-04 eta: 4:43:57 time: 0.6342 data_time: 0.0017 memory: 44137 loss: 0.4032 +2023/06/06 07:42:11 - mmengine - INFO - Epoch(train) [4][ 900/3907] lr: 2.3153e-04 eta: 4:42:52 time: 0.6331 data_time: 0.0016 memory: 44137 loss: 0.4019 +2023/06/06 07:43:14 - mmengine - INFO - Epoch(train) [4][1000/3907] lr: 2.3054e-04 eta: 4:41:46 time: 0.6352 data_time: 0.0014 memory: 44137 loss: 0.4120 +2023/06/06 07:44:18 - mmengine - INFO - Epoch(train) [4][1100/3907] lr: 2.2954e-04 eta: 4:40:40 time: 0.6339 data_time: 0.0016 memory: 44137 loss: 0.3979 +2023/06/06 07:45:21 - mmengine - INFO - Epoch(train) [4][1200/3907] lr: 2.2854e-04 eta: 4:39:34 time: 0.6344 data_time: 0.0016 memory: 44137 loss: 0.4025 +2023/06/06 07:46:11 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr3e-4_20230606_052113 +2023/06/06 07:46:24 - mmengine - INFO - Epoch(train) [4][1300/3907] lr: 2.2753e-04 eta: 4:38:29 time: 0.6327 data_time: 0.0017 memory: 44137 loss: 0.3891 +2023/06/06 07:47:28 - mmengine - INFO - Epoch(train) [4][1400/3907] lr: 2.2652e-04 eta: 4:37:23 time: 0.6341 data_time: 0.0015 memory: 44137 loss: 0.3711 +2023/06/06 07:48:31 - mmengine - INFO - Epoch(train) [4][1500/3907] lr: 2.2551e-04 eta: 4:36:17 time: 0.6345 data_time: 0.0015 memory: 44137 loss: 0.3801 +2023/06/06 07:49:35 - mmengine - INFO - Epoch(train) [4][1600/3907] lr: 2.2448e-04 eta: 4:35:12 time: 0.6337 data_time: 0.0014 memory: 44137 loss: 0.3855 +2023/06/06 07:50:38 - mmengine - INFO - Epoch(train) [4][1700/3907] lr: 2.2346e-04 eta: 4:34:07 time: 0.6346 data_time: 0.0014 memory: 44137 loss: 0.3935 +2023/06/06 07:51:42 - mmengine - INFO - Epoch(train) [4][1800/3907] lr: 2.2243e-04 eta: 4:33:01 time: 0.6364 data_time: 0.0014 memory: 44137 loss: 0.3960 +2023/06/06 07:52:45 - mmengine - INFO - Epoch(train) [4][1900/3907] lr: 2.2139e-04 eta: 4:31:56 time: 0.6374 data_time: 0.0015 memory: 44137 loss: 0.4126 +2023/06/06 07:53:49 - mmengine - INFO - Epoch(train) [4][2000/3907] lr: 2.2036e-04 eta: 4:30:51 time: 0.6339 data_time: 0.0014 memory: 44137 loss: 0.4013 +2023/06/06 07:54:52 - mmengine - INFO - Epoch(train) [4][2100/3907] lr: 2.1931e-04 eta: 4:29:45 time: 0.6343 data_time: 0.0014 memory: 44137 loss: 0.4048 +2023/06/06 07:55:56 - mmengine - INFO - Epoch(train) [4][2200/3907] lr: 2.1827e-04 eta: 4:28:40 time: 0.6333 data_time: 0.0014 memory: 44137 loss: 0.4055 +2023/06/06 07:56:46 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr3e-4_20230606_052113 +2023/06/06 07:56:59 - mmengine - INFO - Epoch(train) [4][2300/3907] lr: 2.1721e-04 eta: 4:27:35 time: 0.6355 data_time: 0.0015 memory: 44137 loss: 0.4077 +2023/06/06 07:58:03 - mmengine - INFO - Epoch(train) [4][2400/3907] lr: 2.1616e-04 eta: 4:26:30 time: 0.6345 data_time: 0.0015 memory: 44137 loss: 0.4163 +2023/06/06 07:59:06 - mmengine - INFO - Epoch(train) [4][2500/3907] lr: 2.1510e-04 eta: 4:25:24 time: 0.6341 data_time: 0.0016 memory: 44137 loss: 0.3776 +2023/06/06 08:00:09 - mmengine - INFO - Epoch(train) [4][2600/3907] lr: 2.1404e-04 eta: 4:24:19 time: 0.6339 data_time: 0.0014 memory: 44137 loss: 0.4263 +2023/06/06 08:01:13 - mmengine - INFO - Epoch(train) [4][2700/3907] lr: 2.1297e-04 eta: 4:23:14 time: 0.6336 data_time: 0.0015 memory: 44137 loss: 0.3801 +2023/06/06 08:02:16 - mmengine - INFO - Epoch(train) [4][2800/3907] lr: 2.1190e-04 eta: 4:22:09 time: 0.6334 data_time: 0.0014 memory: 44137 loss: 0.3921 +2023/06/06 08:03:20 - mmengine - INFO - Epoch(train) [4][2900/3907] lr: 2.1083e-04 eta: 4:21:04 time: 0.6341 data_time: 0.0015 memory: 44137 loss: 0.3812 +2023/06/06 08:04:23 - mmengine - INFO - Epoch(train) [4][3000/3907] lr: 2.0975e-04 eta: 4:19:59 time: 0.6346 data_time: 0.0014 memory: 44137 loss: 0.3905 +2023/06/06 08:05:26 - mmengine - INFO - Epoch(train) [4][3100/3907] lr: 2.0867e-04 eta: 4:18:53 time: 0.6332 data_time: 0.0017 memory: 44137 loss: 0.4121 +2023/06/06 08:06:30 - mmengine - INFO - Epoch(train) [4][3200/3907] lr: 2.0758e-04 eta: 4:17:48 time: 0.6328 data_time: 0.0015 memory: 44137 loss: 0.3613 +2023/06/06 08:07:20 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr3e-4_20230606_052113 +2023/06/06 08:07:33 - mmengine - INFO - Epoch(train) [4][3300/3907] lr: 2.0649e-04 eta: 4:16:43 time: 0.6339 data_time: 0.0015 memory: 44137 loss: 0.3770 +2023/06/06 08:08:37 - mmengine - INFO - Epoch(train) [4][3400/3907] lr: 2.0540e-04 eta: 4:15:38 time: 0.6348 data_time: 0.0015 memory: 44137 loss: 0.3904 +2023/06/06 08:09:40 - mmengine - INFO - Epoch(train) [4][3500/3907] lr: 2.0431e-04 eta: 4:14:33 time: 0.6347 data_time: 0.0015 memory: 44137 loss: 0.3929 +2023/06/06 08:10:43 - mmengine - INFO - Epoch(train) [4][3600/3907] lr: 2.0321e-04 eta: 4:13:28 time: 0.6337 data_time: 0.0015 memory: 44137 loss: 0.4146 +2023/06/06 08:11:47 - mmengine - INFO - Epoch(train) [4][3700/3907] lr: 2.0211e-04 eta: 4:12:23 time: 0.6331 data_time: 0.0014 memory: 44137 loss: 0.3817 +2023/06/06 08:12:50 - mmengine - INFO - Epoch(train) [4][3800/3907] lr: 2.0100e-04 eta: 4:11:18 time: 0.6344 data_time: 0.0016 memory: 44137 loss: 0.3898 +2023/06/06 08:13:54 - mmengine - INFO - Epoch(train) [4][3900/3907] lr: 1.9990e-04 eta: 4:10:13 time: 0.6353 data_time: 0.0012 memory: 44137 loss: 0.4089 +2023/06/06 08:13:58 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr3e-4_20230606_052113 +2023/06/06 08:13:58 - mmengine - INFO - Saving checkpoint at 4 epochs +2023/06/06 08:15:43 - mmengine - INFO - Epoch(val) [4][57/57] accuracy/top1: 74.1032 single-label/precision_classwise: [68.3282470703125, 96.79299926757812] single-label/recall_classwise: [98.81951904296875, 43.75193786621094] single-label/f1-score_classwise: [80.79275512695312, 60.26373291015625] data_time: 0.0387 time: 1.2833 +2023/06/06 08:16:49 - mmengine - INFO - Epoch(train) [5][ 100/3907] lr: 1.9871e-04 eta: 4:09:08 time: 0.6335 data_time: 0.0015 memory: 44137 loss: 0.3896 +2023/06/06 08:17:53 - mmengine - INFO - Epoch(train) [5][ 200/3907] lr: 1.9760e-04 eta: 4:08:03 time: 0.6336 data_time: 0.0015 memory: 44137 loss: 0.4006 +2023/06/06 08:18:56 - mmengine - INFO - Epoch(train) [5][ 300/3907] lr: 1.9648e-04 eta: 4:06:58 time: 0.6331 data_time: 0.0019 memory: 44137 loss: 0.3996 +2023/06/06 08:19:42 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr3e-4_20230606_052113 +2023/06/06 08:20:00 - mmengine - INFO - Epoch(train) [5][ 400/3907] lr: 1.9536e-04 eta: 4:05:53 time: 0.6337 data_time: 0.0014 memory: 44137 loss: 0.4185 +2023/06/06 08:21:03 - mmengine - INFO - Epoch(train) [5][ 500/3907] lr: 1.9424e-04 eta: 4:04:48 time: 0.6334 data_time: 0.0014 memory: 44137 loss: 0.3988 +2023/06/06 08:22:07 - mmengine - INFO - Epoch(train) [5][ 600/3907] lr: 1.9312e-04 eta: 4:03:43 time: 0.6339 data_time: 0.0015 memory: 44137 loss: 0.4062 +2023/06/06 08:23:10 - mmengine - INFO - Epoch(train) [5][ 700/3907] lr: 1.9199e-04 eta: 4:02:38 time: 0.6336 data_time: 0.0015 memory: 44137 loss: 0.3885 +2023/06/06 08:24:13 - mmengine - INFO - Epoch(train) [5][ 800/3907] lr: 1.9086e-04 eta: 4:01:33 time: 0.6340 data_time: 0.0016 memory: 44137 loss: 0.4010 +2023/06/06 08:25:17 - mmengine - INFO - Epoch(train) [5][ 900/3907] lr: 1.8973e-04 eta: 4:00:29 time: 0.6329 data_time: 0.0016 memory: 44137 loss: 0.3853 +2023/06/06 08:26:20 - mmengine - INFO - Epoch(train) [5][1000/3907] lr: 1.8860e-04 eta: 3:59:24 time: 0.6334 data_time: 0.0017 memory: 44137 loss: 0.4026 +2023/06/06 08:27:24 - mmengine - INFO - Epoch(train) [5][1100/3907] lr: 1.8746e-04 eta: 3:58:19 time: 0.6331 data_time: 0.0014 memory: 44137 loss: 0.4035 +2023/06/06 08:28:27 - mmengine - INFO - Epoch(train) [5][1200/3907] lr: 1.8632e-04 eta: 3:57:14 time: 0.6331 data_time: 0.0014 memory: 44137 loss: 0.4132 +2023/06/06 08:29:30 - mmengine - INFO - Epoch(train) [5][1300/3907] lr: 1.8519e-04 eta: 3:56:09 time: 0.6340 data_time: 0.0015 memory: 44137 loss: 0.3668 +2023/06/06 08:30:16 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr3e-4_20230606_052113 +2023/06/06 08:30:34 - mmengine - INFO - Epoch(train) [5][1400/3907] lr: 1.8404e-04 eta: 3:55:04 time: 0.6333 data_time: 0.0015 memory: 44137 loss: 0.4048 +2023/06/06 08:31:37 - mmengine - INFO - Epoch(train) [5][1500/3907] lr: 1.8290e-04 eta: 3:54:00 time: 0.6331 data_time: 0.0016 memory: 44137 loss: 0.3937 +2023/06/06 08:32:40 - mmengine - INFO - Epoch(train) [5][1600/3907] lr: 1.8176e-04 eta: 3:52:55 time: 0.6332 data_time: 0.0015 memory: 44137 loss: 0.4019 +2023/06/06 08:33:44 - mmengine - INFO - Epoch(train) [5][1700/3907] lr: 1.8061e-04 eta: 3:51:50 time: 0.6330 data_time: 0.0014 memory: 44137 loss: 0.3819 +2023/06/06 08:34:47 - mmengine - INFO - Epoch(train) [5][1800/3907] lr: 1.7946e-04 eta: 3:50:45 time: 0.6336 data_time: 0.0014 memory: 44137 loss: 0.4042 +2023/06/06 08:35:51 - mmengine - INFO - Epoch(train) [5][1900/3907] lr: 1.7831e-04 eta: 3:49:41 time: 0.6328 data_time: 0.0015 memory: 44137 loss: 0.3941 +2023/06/06 08:36:54 - mmengine - INFO - Epoch(train) [5][2000/3907] lr: 1.7716e-04 eta: 3:48:36 time: 0.6331 data_time: 0.0015 memory: 44137 loss: 0.3920 +2023/06/06 08:37:57 - mmengine - INFO - Epoch(train) [5][2100/3907] lr: 1.7601e-04 eta: 3:47:31 time: 0.6335 data_time: 0.0014 memory: 44137 loss: 0.4196 +2023/06/06 08:39:00 - mmengine - INFO - Epoch(train) [5][2200/3907] lr: 1.7485e-04 eta: 3:46:26 time: 0.6330 data_time: 0.0017 memory: 44137 loss: 0.3883 +2023/06/06 08:40:04 - mmengine - INFO - Epoch(train) [5][2300/3907] lr: 1.7370e-04 eta: 3:45:22 time: 0.6334 data_time: 0.0015 memory: 44137 loss: 0.4159 +2023/06/06 08:40:50 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr3e-4_20230606_052113 +2023/06/06 08:41:07 - mmengine - INFO - Epoch(train) [5][2400/3907] lr: 1.7254e-04 eta: 3:44:17 time: 0.6331 data_time: 0.0013 memory: 44137 loss: 0.3879 +2023/06/06 08:42:11 - mmengine - INFO - Epoch(train) [5][2500/3907] lr: 1.7138e-04 eta: 3:43:12 time: 0.6328 data_time: 0.0014 memory: 44137 loss: 0.4301 +2023/06/06 08:43:14 - mmengine - INFO - Epoch(train) [5][2600/3907] lr: 1.7022e-04 eta: 3:42:08 time: 0.6334 data_time: 0.0016 memory: 44137 loss: 0.4012 +2023/06/06 08:44:17 - mmengine - INFO - Epoch(train) [5][2700/3907] lr: 1.6906e-04 eta: 3:41:03 time: 0.6366 data_time: 0.0015 memory: 44137 loss: 0.3898 +2023/06/06 08:45:21 - mmengine - INFO - Epoch(train) [5][2800/3907] lr: 1.6790e-04 eta: 3:39:59 time: 0.6342 data_time: 0.0014 memory: 44137 loss: 0.4273 +2023/06/06 08:46:24 - mmengine - INFO - Epoch(train) [5][2900/3907] lr: 1.6674e-04 eta: 3:38:54 time: 0.6335 data_time: 0.0015 memory: 44137 loss: 0.3856 +2023/06/06 08:47:28 - mmengine - INFO - Epoch(train) [5][3000/3907] lr: 1.6558e-04 eta: 3:37:50 time: 0.6338 data_time: 0.0015 memory: 44137 loss: 0.3669 +2023/06/06 08:48:31 - mmengine - INFO - Epoch(train) [5][3100/3907] lr: 1.6441e-04 eta: 3:36:45 time: 0.6337 data_time: 0.0014 memory: 44137 loss: 0.4032 +2023/06/06 08:49:35 - mmengine - INFO - Epoch(train) [5][3200/3907] lr: 1.6325e-04 eta: 3:35:41 time: 0.6339 data_time: 0.0015 memory: 44137 loss: 0.4009 +2023/06/06 08:50:38 - mmengine - INFO - Epoch(train) [5][3300/3907] lr: 1.6209e-04 eta: 3:34:36 time: 0.6338 data_time: 0.0015 memory: 44137 loss: 0.3852 +2023/06/06 08:51:24 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr3e-4_20230606_052113 +2023/06/06 08:51:41 - mmengine - INFO - Epoch(train) [5][3400/3907] lr: 1.6092e-04 eta: 3:33:32 time: 0.6344 data_time: 0.0014 memory: 44137 loss: 0.4015 +2023/06/06 08:52:45 - mmengine - INFO - Epoch(train) [5][3500/3907] lr: 1.5976e-04 eta: 3:32:27 time: 0.6346 data_time: 0.0015 memory: 44137 loss: 0.3843 +2023/06/06 08:53:48 - mmengine - INFO - Epoch(train) [5][3600/3907] lr: 1.5859e-04 eta: 3:31:23 time: 0.6335 data_time: 0.0016 memory: 44137 loss: 0.4046 +2023/06/06 08:54:52 - mmengine - INFO - Epoch(train) [5][3700/3907] lr: 1.5743e-04 eta: 3:30:18 time: 0.6338 data_time: 0.0014 memory: 44137 loss: 0.4211 +2023/06/06 08:55:55 - mmengine - INFO - Epoch(train) [5][3800/3907] lr: 1.5626e-04 eta: 3:29:14 time: 0.6340 data_time: 0.0015 memory: 44137 loss: 0.4157 +2023/06/06 08:56:59 - mmengine - INFO - Epoch(train) [5][3900/3907] lr: 1.5509e-04 eta: 3:28:09 time: 0.6333 data_time: 0.0013 memory: 44137 loss: 0.3958 +2023/06/06 08:57:02 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr3e-4_20230606_052113 +2023/06/06 08:57:03 - mmengine - INFO - Saving checkpoint at 5 epochs +2023/06/06 08:58:45 - mmengine - INFO - Epoch(val) [5][57/57] accuracy/top1: 74.8965 single-label/precision_classwise: [69.01339721679688, 96.9291763305664] single-label/recall_classwise: [98.8258285522461, 45.511627197265625] single-label/f1-score_classwise: [81.27189636230469, 61.940181732177734] data_time: 0.0357 time: 1.2823 +2023/06/06 08:59:51 - mmengine - INFO - Epoch(train) [6][ 100/3907] lr: 1.5385e-04 eta: 3:27:03 time: 0.6339 data_time: 0.0016 memory: 44137 loss: 0.4108 +2023/06/06 09:00:55 - mmengine - INFO - Epoch(train) [6][ 200/3907] lr: 1.5268e-04 eta: 3:25:59 time: 0.6333 data_time: 0.0015 memory: 44137 loss: 0.3964 +2023/06/06 09:01:58 - mmengine - INFO - Epoch(train) [6][ 300/3907] lr: 1.5151e-04 eta: 3:24:54 time: 0.6329 data_time: 0.0015 memory: 44137 loss: 0.3975 +2023/06/06 09:03:02 - mmengine - INFO - Epoch(train) [6][ 400/3907] lr: 1.5035e-04 eta: 3:23:50 time: 0.6332 data_time: 0.0014 memory: 44137 loss: 0.3966 +2023/06/06 09:03:43 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr3e-4_20230606_052113 +2023/06/06 09:04:05 - mmengine - INFO - Epoch(train) [6][ 500/3907] lr: 1.4918e-04 eta: 3:22:45 time: 0.6330 data_time: 0.0015 memory: 44137 loss: 0.4015 +2023/06/06 09:05:08 - mmengine - INFO - Epoch(train) [6][ 600/3907] lr: 1.4802e-04 eta: 3:21:41 time: 0.6342 data_time: 0.0015 memory: 44137 loss: 0.4208 +2023/06/06 09:06:12 - mmengine - INFO - Epoch(train) [6][ 700/3907] lr: 1.4685e-04 eta: 3:20:36 time: 0.6332 data_time: 0.0014 memory: 44137 loss: 0.3958 +2023/06/06 09:07:15 - mmengine - INFO - Epoch(train) [6][ 800/3907] lr: 1.4569e-04 eta: 3:19:32 time: 0.6336 data_time: 0.0016 memory: 44137 loss: 0.4158 +2023/06/06 09:08:18 - mmengine - INFO - Epoch(train) [6][ 900/3907] lr: 1.4453e-04 eta: 3:18:27 time: 0.6334 data_time: 0.0016 memory: 44137 loss: 0.4056 +2023/06/06 09:09:22 - mmengine - INFO - Epoch(train) [6][1000/3907] lr: 1.4336e-04 eta: 3:17:23 time: 0.6329 data_time: 0.0014 memory: 44137 loss: 0.4274 +2023/06/06 09:10:25 - mmengine - INFO - Epoch(train) [6][1100/3907] lr: 1.4220e-04 eta: 3:16:19 time: 0.6342 data_time: 0.0015 memory: 44137 loss: 0.4200 +2023/06/06 09:11:29 - mmengine - INFO - Epoch(train) [6][1200/3907] lr: 1.4104e-04 eta: 3:15:15 time: 0.6343 data_time: 0.0015 memory: 44137 loss: 0.4036 +2023/06/06 09:12:32 - mmengine - INFO - Epoch(train) [6][1300/3907] lr: 1.3988e-04 eta: 3:14:10 time: 0.6339 data_time: 0.0015 memory: 44137 loss: 0.3884 +2023/06/06 09:13:35 - mmengine - INFO - Epoch(train) [6][1400/3907] lr: 1.3872e-04 eta: 3:13:06 time: 0.6349 data_time: 0.0016 memory: 44137 loss: 0.4155 +2023/06/06 09:14:17 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr3e-4_20230606_052113 +2023/06/06 09:14:39 - mmengine - INFO - Epoch(train) [6][1500/3907] lr: 1.3756e-04 eta: 3:12:02 time: 0.6335 data_time: 0.0015 memory: 44137 loss: 0.4024 +2023/06/06 09:15:42 - mmengine - INFO - Epoch(train) [6][1600/3907] lr: 1.3641e-04 eta: 3:10:57 time: 0.6332 data_time: 0.0016 memory: 44137 loss: 0.4068 +2023/06/06 09:16:46 - mmengine - INFO - Epoch(train) [6][1700/3907] lr: 1.3525e-04 eta: 3:09:53 time: 0.6350 data_time: 0.0015 memory: 44137 loss: 0.3796 +2023/06/06 09:17:49 - mmengine - INFO - Epoch(train) [6][1800/3907] lr: 1.3410e-04 eta: 3:08:49 time: 0.6359 data_time: 0.0014 memory: 44137 loss: 0.4033 +2023/06/06 09:18:53 - mmengine - INFO - Epoch(train) [6][1900/3907] lr: 1.3294e-04 eta: 3:07:44 time: 0.6332 data_time: 0.0016 memory: 44137 loss: 0.3757 +2023/06/06 09:19:56 - mmengine - INFO - Epoch(train) [6][2000/3907] lr: 1.3179e-04 eta: 3:06:40 time: 0.6364 data_time: 0.0014 memory: 44137 loss: 0.4129 +2023/06/06 09:20:59 - mmengine - INFO - Epoch(train) [6][2100/3907] lr: 1.3064e-04 eta: 3:05:36 time: 0.6335 data_time: 0.0016 memory: 44137 loss: 0.4117 +2023/06/06 09:22:03 - mmengine - INFO - Epoch(train) [6][2200/3907] lr: 1.2949e-04 eta: 3:04:32 time: 0.6339 data_time: 0.0015 memory: 44137 loss: 0.4058 +2023/06/06 09:23:06 - mmengine - INFO - Epoch(train) [6][2300/3907] lr: 1.2835e-04 eta: 3:03:28 time: 0.6340 data_time: 0.0015 memory: 44137 loss: 0.3744 +2023/06/06 09:24:10 - mmengine - INFO - Epoch(train) [6][2400/3907] lr: 1.2720e-04 eta: 3:02:23 time: 0.6359 data_time: 0.0015 memory: 44137 loss: 0.4227 +2023/06/06 09:24:51 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr3e-4_20230606_052113 +2023/06/06 09:25:13 - mmengine - INFO - Epoch(train) [6][2500/3907] lr: 1.2606e-04 eta: 3:01:19 time: 0.6338 data_time: 0.0015 memory: 44137 loss: 0.3869 +2023/06/06 09:26:17 - mmengine - INFO - Epoch(train) [6][2600/3907] lr: 1.2492e-04 eta: 3:00:15 time: 0.6350 data_time: 0.0014 memory: 44137 loss: 0.3986 +2023/06/06 09:27:20 - mmengine - INFO - Epoch(train) [6][2700/3907] lr: 1.2378e-04 eta: 2:59:11 time: 0.6345 data_time: 0.0014 memory: 44137 loss: 0.3789 +2023/06/06 09:28:24 - mmengine - INFO - Epoch(train) [6][2800/3907] lr: 1.2264e-04 eta: 2:58:07 time: 0.6335 data_time: 0.0015 memory: 44137 loss: 0.4107 +2023/06/06 09:29:27 - mmengine - INFO - Epoch(train) [6][2900/3907] lr: 1.2150e-04 eta: 2:57:03 time: 0.6344 data_time: 0.0017 memory: 44137 loss: 0.3939 +2023/06/06 09:30:31 - mmengine - INFO - Epoch(train) [6][3000/3907] lr: 1.2037e-04 eta: 2:55:59 time: 0.6367 data_time: 0.0015 memory: 44137 loss: 0.4375 +2023/06/06 09:31:35 - mmengine - INFO - Epoch(train) [6][3100/3907] lr: 1.1924e-04 eta: 2:54:55 time: 0.6340 data_time: 0.0015 memory: 44137 loss: 0.4152 +2023/06/06 09:32:38 - mmengine - INFO - Epoch(train) [6][3200/3907] lr: 1.1811e-04 eta: 2:53:50 time: 0.6358 data_time: 0.0015 memory: 44137 loss: 0.3900 +2023/06/06 09:33:42 - mmengine - INFO - Epoch(train) [6][3300/3907] lr: 1.1699e-04 eta: 2:52:46 time: 0.6346 data_time: 0.0014 memory: 44137 loss: 0.4125 +2023/06/06 09:34:45 - mmengine - INFO - Epoch(train) [6][3400/3907] lr: 1.1586e-04 eta: 2:51:42 time: 0.6342 data_time: 0.0016 memory: 44137 loss: 0.4143 +2023/06/06 09:35:26 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr3e-4_20230606_052113 +2023/06/06 09:35:49 - mmengine - INFO - Epoch(train) [6][3500/3907] lr: 1.1474e-04 eta: 2:50:38 time: 0.6443 data_time: 0.0015 memory: 44137 loss: 0.4068 +2023/06/06 09:36:52 - mmengine - INFO - Epoch(train) [6][3600/3907] lr: 1.1362e-04 eta: 2:49:34 time: 0.6345 data_time: 0.0015 memory: 44137 loss: 0.4010 +2023/06/06 09:37:56 - mmengine - INFO - Epoch(train) [6][3700/3907] lr: 1.1251e-04 eta: 2:48:30 time: 0.6348 data_time: 0.0014 memory: 44137 loss: 0.4028 +2023/06/06 09:38:59 - mmengine - INFO - Epoch(train) [6][3800/3907] lr: 1.1139e-04 eta: 2:47:26 time: 0.6343 data_time: 0.0015 memory: 44137 loss: 0.4139 +2023/06/06 09:40:03 - mmengine - INFO - Epoch(train) [6][3900/3907] lr: 1.1028e-04 eta: 2:46:22 time: 0.6343 data_time: 0.0013 memory: 44137 loss: 0.3883 +2023/06/06 09:40:07 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr3e-4_20230606_052113 +2023/06/06 09:40:07 - mmengine - INFO - Saving checkpoint at 6 epochs +2023/06/06 09:41:49 - mmengine - INFO - Epoch(val) [6][57/57] accuracy/top1: 74.7051 single-label/precision_classwise: [68.8315658569336, 97.0421142578125] single-label/recall_classwise: [98.88264465332031, 45.015506744384766] single-label/f1-score_classwise: [81.16482543945312, 61.50180435180664] data_time: 0.0362 time: 1.2833 +2023/06/06 09:42:55 - mmengine - INFO - Epoch(train) [7][ 100/3907] lr: 1.0910e-04 eta: 2:45:15 time: 0.6341 data_time: 0.0015 memory: 44137 loss: 0.4047 +2023/06/06 09:43:59 - mmengine - INFO - Epoch(train) [7][ 200/3907] lr: 1.0799e-04 eta: 2:44:11 time: 0.6342 data_time: 0.0014 memory: 44137 loss: 0.4482 +2023/06/06 09:45:02 - mmengine - INFO - Epoch(train) [7][ 300/3907] lr: 1.0689e-04 eta: 2:43:07 time: 0.6344 data_time: 0.0014 memory: 44137 loss: 0.4107 +2023/06/06 09:46:06 - mmengine - INFO - Epoch(train) [7][ 400/3907] lr: 1.0579e-04 eta: 2:42:02 time: 0.6347 data_time: 0.0015 memory: 44137 loss: 0.3960 +2023/06/06 09:47:10 - mmengine - INFO - Epoch(train) [7][ 500/3907] lr: 1.0470e-04 eta: 2:40:59 time: 0.6344 data_time: 0.0016 memory: 44137 loss: 0.3927 +2023/06/06 09:47:46 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr3e-4_20230606_052113 +2023/06/06 09:48:13 - mmengine - INFO - Epoch(train) [7][ 600/3907] lr: 1.0360e-04 eta: 2:39:54 time: 0.6349 data_time: 0.0014 memory: 44137 loss: 0.3906 +2023/06/06 09:49:17 - mmengine - INFO - Epoch(train) [7][ 700/3907] lr: 1.0252e-04 eta: 2:38:50 time: 0.6344 data_time: 0.0013 memory: 44137 loss: 0.3974 +2023/06/06 09:50:20 - mmengine - INFO - Epoch(train) [7][ 800/3907] lr: 1.0143e-04 eta: 2:37:46 time: 0.6349 data_time: 0.0015 memory: 44137 loss: 0.4009 +2023/06/06 09:51:24 - mmengine - INFO - Epoch(train) [7][ 900/3907] lr: 1.0035e-04 eta: 2:36:42 time: 0.6343 data_time: 0.0014 memory: 44137 loss: 0.3946 +2023/06/06 09:52:27 - mmengine - INFO - Epoch(train) [7][1000/3907] lr: 9.9271e-05 eta: 2:35:38 time: 0.6361 data_time: 0.0014 memory: 44137 loss: 0.3930 +2023/06/06 09:53:31 - mmengine - INFO - Epoch(train) [7][1100/3907] lr: 9.8197e-05 eta: 2:34:34 time: 0.6358 data_time: 0.0015 memory: 44137 loss: 0.3951 +2023/06/06 09:54:35 - mmengine - INFO - Epoch(train) [7][1200/3907] lr: 9.7126e-05 eta: 2:33:30 time: 0.6365 data_time: 0.0015 memory: 44137 loss: 0.4105 +2023/06/06 09:55:38 - mmengine - INFO - Epoch(train) [7][1300/3907] lr: 9.6059e-05 eta: 2:32:26 time: 0.6357 data_time: 0.0014 memory: 44137 loss: 0.4261 +2023/06/06 09:56:42 - mmengine - INFO - Epoch(train) [7][1400/3907] lr: 9.4995e-05 eta: 2:31:23 time: 0.6353 data_time: 0.0014 memory: 44137 loss: 0.4296 +2023/06/06 09:57:46 - mmengine - INFO - Epoch(train) [7][1500/3907] lr: 9.3936e-05 eta: 2:30:19 time: 0.6354 data_time: 0.0014 memory: 44137 loss: 0.3760 +2023/06/06 09:58:22 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr3e-4_20230606_052113 +2023/06/06 09:58:49 - mmengine - INFO - Epoch(train) [7][1600/3907] lr: 9.2880e-05 eta: 2:29:15 time: 0.6355 data_time: 0.0017 memory: 44137 loss: 0.4077 +2023/06/06 09:59:53 - mmengine - INFO - Epoch(train) [7][1700/3907] lr: 9.1829e-05 eta: 2:28:11 time: 0.6375 data_time: 0.0015 memory: 44137 loss: 0.3991 +2023/06/06 10:00:56 - mmengine - INFO - Epoch(train) [7][1800/3907] lr: 9.0781e-05 eta: 2:27:07 time: 0.6362 data_time: 0.0015 memory: 44137 loss: 0.3841 +2023/06/06 10:02:00 - mmengine - INFO - Epoch(train) [7][1900/3907] lr: 8.9738e-05 eta: 2:26:03 time: 0.6354 data_time: 0.0016 memory: 44137 loss: 0.3982 +2023/06/06 10:03:03 - mmengine - INFO - Epoch(train) [7][2000/3907] lr: 8.8699e-05 eta: 2:24:59 time: 0.6339 data_time: 0.0015 memory: 44137 loss: 0.4010 +2023/06/06 10:04:07 - mmengine - INFO - Epoch(train) [7][2100/3907] lr: 8.7664e-05 eta: 2:23:55 time: 0.6347 data_time: 0.0014 memory: 44137 loss: 0.4164 +2023/06/06 10:05:10 - mmengine - INFO - Epoch(train) [7][2200/3907] lr: 8.6634e-05 eta: 2:22:51 time: 0.6363 data_time: 0.0015 memory: 44137 loss: 0.4027 +2023/06/06 10:06:14 - mmengine - INFO - Epoch(train) [7][2300/3907] lr: 8.5608e-05 eta: 2:21:47 time: 0.6344 data_time: 0.0015 memory: 44137 loss: 0.4354 +2023/06/06 10:07:17 - mmengine - INFO - Epoch(train) [7][2400/3907] lr: 8.4586e-05 eta: 2:20:43 time: 0.6344 data_time: 0.0015 memory: 44137 loss: 0.3847 +2023/06/06 10:08:21 - mmengine - INFO - Epoch(train) [7][2500/3907] lr: 8.3570e-05 eta: 2:19:39 time: 0.6342 data_time: 0.0015 memory: 44137 loss: 0.3799 +2023/06/06 10:08:58 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr3e-4_20230606_052113 +2023/06/06 10:09:24 - mmengine - INFO - Epoch(train) [7][2600/3907] lr: 8.2557e-05 eta: 2:18:35 time: 0.6345 data_time: 0.0014 memory: 44137 loss: 0.3941 +2023/06/06 10:10:28 - mmengine - INFO - Epoch(train) [7][2700/3907] lr: 8.1550e-05 eta: 2:17:31 time: 0.6345 data_time: 0.0015 memory: 44137 loss: 0.4308 +2023/06/06 10:11:31 - mmengine - INFO - Epoch(train) [7][2800/3907] lr: 8.0547e-05 eta: 2:16:27 time: 0.6340 data_time: 0.0015 memory: 44137 loss: 0.3985 +2023/06/06 10:12:35 - mmengine - INFO - Epoch(train) [7][2900/3907] lr: 7.9549e-05 eta: 2:15:23 time: 0.6340 data_time: 0.0015 memory: 44137 loss: 0.3925 +2023/06/06 10:13:39 - mmengine - INFO - Epoch(train) [7][3000/3907] lr: 7.8555e-05 eta: 2:14:19 time: 0.6360 data_time: 0.0015 memory: 44137 loss: 0.3960 +2023/06/06 10:14:42 - mmengine - INFO - Epoch(train) [7][3100/3907] lr: 7.7567e-05 eta: 2:13:15 time: 0.6359 data_time: 0.0014 memory: 44137 loss: 0.4325 +2023/06/06 10:15:46 - mmengine - INFO - Epoch(train) [7][3200/3907] lr: 7.6584e-05 eta: 2:12:11 time: 0.6353 data_time: 0.0015 memory: 44137 loss: 0.4226 +2023/06/06 10:16:49 - mmengine - INFO - Epoch(train) [7][3300/3907] lr: 7.5606e-05 eta: 2:11:07 time: 0.6345 data_time: 0.0014 memory: 44137 loss: 0.4033 +2023/06/06 10:17:53 - mmengine - INFO - Epoch(train) [7][3400/3907] lr: 7.4633e-05 eta: 2:10:03 time: 0.6340 data_time: 0.0015 memory: 44137 loss: 0.3979 +2023/06/06 10:18:56 - mmengine - INFO - Epoch(train) [7][3500/3907] lr: 7.3665e-05 eta: 2:08:59 time: 0.6358 data_time: 0.0015 memory: 44137 loss: 0.4090 +2023/06/06 10:19:33 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr3e-4_20230606_052113 +2023/06/06 10:20:00 - mmengine - INFO - Epoch(train) [7][3600/3907] lr: 7.2702e-05 eta: 2:07:55 time: 0.6345 data_time: 0.0016 memory: 44137 loss: 0.3982 +2023/06/06 10:21:03 - mmengine - INFO - Epoch(train) [7][3700/3907] lr: 7.1745e-05 eta: 2:06:51 time: 0.6357 data_time: 0.0016 memory: 44137 loss: 0.3984 +2023/06/06 10:22:07 - mmengine - INFO - Epoch(train) [7][3800/3907] lr: 7.0793e-05 eta: 2:05:47 time: 0.6342 data_time: 0.0015 memory: 44137 loss: 0.3969 +2023/06/06 10:23:10 - mmengine - INFO - Epoch(train) [7][3900/3907] lr: 6.9847e-05 eta: 2:04:43 time: 0.6357 data_time: 0.0012 memory: 44137 loss: 0.4011 +2023/06/06 10:23:14 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr3e-4_20230606_052113 +2023/06/06 10:23:14 - mmengine - INFO - Saving checkpoint at 7 epochs +2023/06/06 10:24:57 - mmengine - INFO - Epoch(val) [7][57/57] accuracy/top1: 74.8408 single-label/precision_classwise: [68.9764633178711, 96.81255340576172] single-label/recall_classwise: [98.78164672851562, 45.44186019897461] single-label/f1-score_classwise: [81.23133850097656, 61.85175323486328] data_time: 0.0357 time: 1.2791 +2023/06/06 10:26:04 - mmengine - INFO - Epoch(train) [8][ 100/3907] lr: 6.8840e-05 eta: 2:03:36 time: 0.6349 data_time: 0.0016 memory: 44137 loss: 0.3970 +2023/06/06 10:27:08 - mmengine - INFO - Epoch(train) [8][ 200/3907] lr: 6.7905e-05 eta: 2:02:32 time: 0.6335 data_time: 0.0014 memory: 44137 loss: 0.4097 +2023/06/06 10:28:11 - mmengine - INFO - Epoch(train) [8][ 300/3907] lr: 6.6976e-05 eta: 2:01:28 time: 0.6338 data_time: 0.0017 memory: 44137 loss: 0.4081 +2023/06/06 10:29:14 - mmengine - INFO - Epoch(train) [8][ 400/3907] lr: 6.6052e-05 eta: 2:00:24 time: 0.6338 data_time: 0.0015 memory: 44137 loss: 0.4042 +2023/06/06 10:30:18 - mmengine - INFO - Epoch(train) [8][ 500/3907] lr: 6.5134e-05 eta: 1:59:20 time: 0.6333 data_time: 0.0014 memory: 44137 loss: 0.4124 +2023/06/06 10:31:21 - mmengine - INFO - Epoch(train) [8][ 600/3907] lr: 6.4222e-05 eta: 1:58:16 time: 0.6346 data_time: 0.0015 memory: 44137 loss: 0.4051 +2023/06/06 10:31:54 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr3e-4_20230606_052113 +2023/06/06 10:32:25 - mmengine - INFO - Epoch(train) [8][ 700/3907] lr: 6.3316e-05 eta: 1:57:12 time: 0.6339 data_time: 0.0015 memory: 44137 loss: 0.3989 +2023/06/06 10:33:28 - mmengine - INFO - Epoch(train) [8][ 800/3907] lr: 6.2416e-05 eta: 1:56:08 time: 0.6330 data_time: 0.0016 memory: 44137 loss: 0.3962 +2023/06/06 10:34:31 - mmengine - INFO - Epoch(train) [8][ 900/3907] lr: 6.1521e-05 eta: 1:55:04 time: 0.6331 data_time: 0.0015 memory: 44137 loss: 0.4054 +2023/06/06 10:35:35 - mmengine - INFO - Epoch(train) [8][1000/3907] lr: 6.0633e-05 eta: 1:54:00 time: 0.6334 data_time: 0.0015 memory: 44137 loss: 0.4100 +2023/06/06 10:36:38 - mmengine - INFO - Epoch(train) [8][1100/3907] lr: 5.9751e-05 eta: 1:52:56 time: 0.6335 data_time: 0.0016 memory: 44137 loss: 0.3976 +2023/06/06 10:37:42 - mmengine - INFO - Epoch(train) [8][1200/3907] lr: 5.8875e-05 eta: 1:51:52 time: 0.6329 data_time: 0.0015 memory: 44137 loss: 0.4064 +2023/06/06 10:38:45 - mmengine - INFO - Epoch(train) [8][1300/3907] lr: 5.8005e-05 eta: 1:50:48 time: 0.6331 data_time: 0.0016 memory: 44137 loss: 0.4128 +2023/06/06 10:39:48 - mmengine - INFO - Epoch(train) [8][1400/3907] lr: 5.7141e-05 eta: 1:49:44 time: 0.6349 data_time: 0.0014 memory: 44137 loss: 0.4046 +2023/06/06 10:40:52 - mmengine - INFO - Epoch(train) [8][1500/3907] lr: 5.6284e-05 eta: 1:48:40 time: 0.6336 data_time: 0.0016 memory: 44137 loss: 0.3848 +2023/06/06 10:41:55 - mmengine - INFO - Epoch(train) [8][1600/3907] lr: 5.5433e-05 eta: 1:47:36 time: 0.6359 data_time: 0.0016 memory: 44137 loss: 0.3775 +2023/06/06 10:42:27 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr3e-4_20230606_052113 +2023/06/06 10:42:58 - mmengine - INFO - Epoch(train) [8][1700/3907] lr: 5.4589e-05 eta: 1:46:33 time: 0.6341 data_time: 0.0016 memory: 44137 loss: 0.4161 +2023/06/06 10:44:02 - mmengine - INFO - Epoch(train) [8][1800/3907] lr: 5.3751e-05 eta: 1:45:29 time: 0.6332 data_time: 0.0015 memory: 44137 loss: 0.3929 +2023/06/06 10:45:05 - mmengine - INFO - Epoch(train) [8][1900/3907] lr: 5.2920e-05 eta: 1:44:25 time: 0.6346 data_time: 0.0015 memory: 44137 loss: 0.4086 +2023/06/06 10:46:09 - mmengine - INFO - Epoch(train) [8][2000/3907] lr: 5.2095e-05 eta: 1:43:21 time: 0.6334 data_time: 0.0015 memory: 44137 loss: 0.4202 +2023/06/06 10:47:12 - mmengine - INFO - Epoch(train) [8][2100/3907] lr: 5.1277e-05 eta: 1:42:17 time: 0.6332 data_time: 0.0016 memory: 44137 loss: 0.4079 +2023/06/06 10:48:15 - mmengine - INFO - Epoch(train) [8][2200/3907] lr: 5.0466e-05 eta: 1:41:13 time: 0.6331 data_time: 0.0015 memory: 44137 loss: 0.3835 +2023/06/06 10:49:19 - mmengine - INFO - Epoch(train) [8][2300/3907] lr: 4.9661e-05 eta: 1:40:09 time: 0.6331 data_time: 0.0014 memory: 44137 loss: 0.3971 +2023/06/06 10:50:22 - mmengine - INFO - Epoch(train) [8][2400/3907] lr: 4.8863e-05 eta: 1:39:05 time: 0.6334 data_time: 0.0014 memory: 44137 loss: 0.4183 +2023/06/06 10:51:26 - mmengine - INFO - Epoch(train) [8][2500/3907] lr: 4.8072e-05 eta: 1:38:01 time: 0.6335 data_time: 0.0014 memory: 44137 loss: 0.4206 +2023/06/06 10:52:29 - mmengine - INFO - Epoch(train) [8][2600/3907] lr: 4.7288e-05 eta: 1:36:57 time: 0.6330 data_time: 0.0015 memory: 44137 loss: 0.4409 +2023/06/06 10:53:02 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr3e-4_20230606_052113 +2023/06/06 10:53:33 - mmengine - INFO - Epoch(train) [8][2700/3907] lr: 4.6511e-05 eta: 1:35:53 time: 0.6337 data_time: 0.0015 memory: 44137 loss: 0.4135 +2023/06/06 10:54:36 - mmengine - INFO - Epoch(train) [8][2800/3907] lr: 4.5741e-05 eta: 1:34:50 time: 0.6341 data_time: 0.0015 memory: 44137 loss: 0.3945 +2023/06/06 10:55:39 - mmengine - INFO - Epoch(train) [8][2900/3907] lr: 4.4978e-05 eta: 1:33:46 time: 0.6331 data_time: 0.0016 memory: 44137 loss: 0.4152 +2023/06/06 10:56:43 - mmengine - INFO - Epoch(train) [8][3000/3907] lr: 4.4222e-05 eta: 1:32:42 time: 0.6346 data_time: 0.0016 memory: 44137 loss: 0.4000 +2023/06/06 10:57:46 - mmengine - INFO - Epoch(train) [8][3100/3907] lr: 4.3474e-05 eta: 1:31:38 time: 0.6339 data_time: 0.0015 memory: 44137 loss: 0.4009 +2023/06/06 10:58:50 - mmengine - INFO - Epoch(train) [8][3200/3907] lr: 4.2732e-05 eta: 1:30:34 time: 0.6341 data_time: 0.0015 memory: 44137 loss: 0.4309 +2023/06/06 10:59:53 - mmengine - INFO - Epoch(train) [8][3300/3907] lr: 4.1998e-05 eta: 1:29:30 time: 0.6334 data_time: 0.0015 memory: 44137 loss: 0.4027 +2023/06/06 11:00:56 - mmengine - INFO - Epoch(train) [8][3400/3907] lr: 4.1271e-05 eta: 1:28:26 time: 0.6337 data_time: 0.0015 memory: 44137 loss: 0.4031 +2023/06/06 11:02:00 - mmengine - INFO - Epoch(train) [8][3500/3907] lr: 4.0551e-05 eta: 1:27:22 time: 0.6335 data_time: 0.0016 memory: 44137 loss: 0.4453 +2023/06/06 11:03:03 - mmengine - INFO - Epoch(train) [8][3600/3907] lr: 3.9839e-05 eta: 1:26:18 time: 0.6356 data_time: 0.0015 memory: 44137 loss: 0.3786 +2023/06/06 11:03:36 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr3e-4_20230606_052113 +2023/06/06 11:04:07 - mmengine - INFO - Epoch(train) [8][3700/3907] lr: 3.9134e-05 eta: 1:25:15 time: 0.6338 data_time: 0.0016 memory: 44137 loss: 0.3968 +2023/06/06 11:05:10 - mmengine - INFO - Epoch(train) [8][3800/3907] lr: 3.8437e-05 eta: 1:24:11 time: 0.6350 data_time: 0.0014 memory: 44137 loss: 0.4266 +2023/06/06 11:06:14 - mmengine - INFO - Epoch(train) [8][3900/3907] lr: 3.7747e-05 eta: 1:23:07 time: 0.6341 data_time: 0.0013 memory: 44137 loss: 0.3991 +2023/06/06 11:06:18 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr3e-4_20230606_052113 +2023/06/06 11:06:18 - mmengine - INFO - Saving checkpoint at 8 epochs +2023/06/06 11:08:00 - mmengine - INFO - Epoch(val) [8][57/57] accuracy/top1: 74.3781 single-label/precision_classwise: [68.55816650390625, 96.89935302734375] single-label/recall_classwise: [98.84477233886719, 44.33333206176758] single-label/f1-score_classwise: [80.96173858642578, 60.833953857421875] data_time: 0.0374 time: 1.2846 +2023/06/06 11:09:07 - mmengine - INFO - Epoch(train) [9][ 100/3907] lr: 3.7018e-05 eta: 1:21:59 time: 0.6333 data_time: 0.0017 memory: 44137 loss: 0.4049 +2023/06/06 11:10:10 - mmengine - INFO - Epoch(train) [9][ 200/3907] lr: 3.6344e-05 eta: 1:20:55 time: 0.6341 data_time: 0.0017 memory: 44137 loss: 0.3990 +2023/06/06 11:11:14 - mmengine - INFO - Epoch(train) [9][ 300/3907] lr: 3.5678e-05 eta: 1:19:52 time: 0.6422 data_time: 0.0016 memory: 44137 loss: 0.3981 +2023/06/06 11:12:17 - mmengine - INFO - Epoch(train) [9][ 400/3907] lr: 3.5019e-05 eta: 1:18:48 time: 0.6339 data_time: 0.0015 memory: 44137 loss: 0.3904 +2023/06/06 11:13:21 - mmengine - INFO - Epoch(train) [9][ 500/3907] lr: 3.4368e-05 eta: 1:17:44 time: 0.6337 data_time: 0.0016 memory: 44137 loss: 0.4220 +2023/06/06 11:14:24 - mmengine - INFO - Epoch(train) [9][ 600/3907] lr: 3.3725e-05 eta: 1:16:40 time: 0.6334 data_time: 0.0014 memory: 44137 loss: 0.3847 +2023/06/06 11:15:27 - mmengine - INFO - Epoch(train) [9][ 700/3907] lr: 3.3090e-05 eta: 1:15:36 time: 0.6343 data_time: 0.0014 memory: 44137 loss: 0.3820 +2023/06/06 11:15:55 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr3e-4_20230606_052113 +2023/06/06 11:16:31 - mmengine - INFO - Epoch(train) [9][ 800/3907] lr: 3.2463e-05 eta: 1:14:32 time: 0.6348 data_time: 0.0014 memory: 44137 loss: 0.3988 +2023/06/06 11:17:34 - mmengine - INFO - Epoch(train) [9][ 900/3907] lr: 3.1843e-05 eta: 1:13:28 time: 0.6337 data_time: 0.0014 memory: 44137 loss: 0.4339 +2023/06/06 11:18:38 - mmengine - INFO - Epoch(train) [9][1000/3907] lr: 3.1232e-05 eta: 1:12:25 time: 0.6342 data_time: 0.0017 memory: 44137 loss: 0.4047 +2023/06/06 11:19:41 - mmengine - INFO - Epoch(train) [9][1100/3907] lr: 3.0628e-05 eta: 1:11:21 time: 0.6336 data_time: 0.0016 memory: 44137 loss: 0.3994 +2023/06/06 11:20:45 - mmengine - INFO - Epoch(train) [9][1200/3907] lr: 3.0033e-05 eta: 1:10:17 time: 0.6332 data_time: 0.0014 memory: 44137 loss: 0.3938 +2023/06/06 11:21:48 - mmengine - INFO - Epoch(train) [9][1300/3907] lr: 2.9446e-05 eta: 1:09:13 time: 0.6330 data_time: 0.0015 memory: 44137 loss: 0.4070 +2023/06/06 11:22:51 - mmengine - INFO - Epoch(train) [9][1400/3907] lr: 2.8867e-05 eta: 1:08:09 time: 0.6341 data_time: 0.0015 memory: 44137 loss: 0.4066 +2023/06/06 11:23:55 - mmengine - INFO - Epoch(train) [9][1500/3907] lr: 2.8296e-05 eta: 1:07:05 time: 0.6341 data_time: 0.0015 memory: 44137 loss: 0.4111 +2023/06/06 11:24:58 - mmengine - INFO - Epoch(train) [9][1600/3907] lr: 2.7733e-05 eta: 1:06:02 time: 0.6417 data_time: 0.0017 memory: 44137 loss: 0.4075 +2023/06/06 11:26:02 - mmengine - INFO - Epoch(train) [9][1700/3907] lr: 2.7178e-05 eta: 1:04:58 time: 0.6334 data_time: 0.0016 memory: 44137 loss: 0.4007 +2023/06/06 11:26:30 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr3e-4_20230606_052113 +2023/06/06 11:27:05 - mmengine - INFO - Epoch(train) [9][1800/3907] lr: 2.6632e-05 eta: 1:03:54 time: 0.6338 data_time: 0.0015 memory: 44137 loss: 0.3754 +2023/06/06 11:28:08 - mmengine - INFO - Epoch(train) [9][1900/3907] lr: 2.6094e-05 eta: 1:02:50 time: 0.6333 data_time: 0.0014 memory: 44137 loss: 0.3654 +2023/06/06 11:29:12 - mmengine - INFO - Epoch(train) [9][2000/3907] lr: 2.5564e-05 eta: 1:01:46 time: 0.6350 data_time: 0.0016 memory: 44137 loss: 0.4076 +2023/06/06 11:30:15 - mmengine - INFO - Epoch(train) [9][2100/3907] lr: 2.5043e-05 eta: 1:00:42 time: 0.6345 data_time: 0.0014 memory: 44137 loss: 0.4009 +2023/06/06 11:31:19 - mmengine - INFO - Epoch(train) [9][2200/3907] lr: 2.4530e-05 eta: 0:59:39 time: 0.6335 data_time: 0.0014 memory: 44137 loss: 0.4067 +2023/06/06 11:32:22 - mmengine - INFO - Epoch(train) [9][2300/3907] lr: 2.4025e-05 eta: 0:58:35 time: 0.6336 data_time: 0.0016 memory: 44137 loss: 0.4030 +2023/06/06 11:33:26 - mmengine - INFO - Epoch(train) [9][2400/3907] lr: 2.3529e-05 eta: 0:57:31 time: 0.6336 data_time: 0.0015 memory: 44137 loss: 0.3882 +2023/06/06 11:34:29 - mmengine - INFO - Epoch(train) [9][2500/3907] lr: 2.3042e-05 eta: 0:56:27 time: 0.6340 data_time: 0.0016 memory: 44137 loss: 0.3934 +2023/06/06 11:35:33 - mmengine - INFO - Epoch(train) [9][2600/3907] lr: 2.2563e-05 eta: 0:55:23 time: 0.6359 data_time: 0.0015 memory: 44137 loss: 0.4072 +2023/06/06 11:36:36 - mmengine - INFO - Epoch(train) [9][2700/3907] lr: 2.2092e-05 eta: 0:54:20 time: 0.6342 data_time: 0.0016 memory: 44137 loss: 0.4105 +2023/06/06 11:37:04 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr3e-4_20230606_052113 +2023/06/06 11:37:40 - mmengine - INFO - Epoch(train) [9][2800/3907] lr: 2.1631e-05 eta: 0:53:16 time: 0.6342 data_time: 0.0015 memory: 44137 loss: 0.4228 +2023/06/06 11:38:43 - mmengine - INFO - Epoch(train) [9][2900/3907] lr: 2.1177e-05 eta: 0:52:12 time: 0.6349 data_time: 0.0015 memory: 44137 loss: 0.4199 +2023/06/06 11:39:47 - mmengine - INFO - Epoch(train) [9][3000/3907] lr: 2.0733e-05 eta: 0:51:08 time: 0.6355 data_time: 0.0015 memory: 44137 loss: 0.3760 +2023/06/06 11:40:50 - mmengine - INFO - Epoch(train) [9][3100/3907] lr: 2.0297e-05 eta: 0:50:05 time: 0.6342 data_time: 0.0015 memory: 44137 loss: 0.3991 +2023/06/06 11:41:54 - mmengine - INFO - Epoch(train) [9][3200/3907] lr: 1.9870e-05 eta: 0:49:01 time: 0.6351 data_time: 0.0014 memory: 44137 loss: 0.4381 +2023/06/06 11:42:57 - mmengine - INFO - Epoch(train) [9][3300/3907] lr: 1.9451e-05 eta: 0:47:57 time: 0.6337 data_time: 0.0014 memory: 44137 loss: 0.4118 +2023/06/06 11:44:00 - mmengine - INFO - Epoch(train) [9][3400/3907] lr: 1.9042e-05 eta: 0:46:53 time: 0.6348 data_time: 0.0016 memory: 44137 loss: 0.4093 +2023/06/06 11:45:04 - mmengine - INFO - Epoch(train) [9][3500/3907] lr: 1.8641e-05 eta: 0:45:49 time: 0.6350 data_time: 0.0017 memory: 44137 loss: 0.3998 +2023/06/06 11:46:07 - mmengine - INFO - Epoch(train) [9][3600/3907] lr: 1.8249e-05 eta: 0:44:46 time: 0.6339 data_time: 0.0014 memory: 44137 loss: 0.3997 +2023/06/06 11:47:11 - mmengine - INFO - Epoch(train) [9][3700/3907] lr: 1.7865e-05 eta: 0:43:42 time: 0.6340 data_time: 0.0015 memory: 44137 loss: 0.4115 +2023/06/06 11:47:39 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr3e-4_20230606_052113 +2023/06/06 11:48:14 - mmengine - INFO - Epoch(train) [9][3800/3907] lr: 1.7491e-05 eta: 0:42:38 time: 0.6342 data_time: 0.0016 memory: 44137 loss: 0.4288 +2023/06/06 11:49:18 - mmengine - INFO - Epoch(train) [9][3900/3907] lr: 1.7126e-05 eta: 0:41:34 time: 0.6356 data_time: 0.0013 memory: 44137 loss: 0.4301 +2023/06/06 11:49:22 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr3e-4_20230606_052113 +2023/06/06 11:49:22 - mmengine - INFO - Saving checkpoint at 9 epochs +2023/06/06 11:51:05 - mmengine - INFO - Epoch(val) [9][57/57] accuracy/top1: 74.2876 single-label/precision_classwise: [68.5012435913086, 96.67909240722656] single-label/recall_classwise: [98.76270294189453, 44.23255920410156] single-label/f1-score_classwise: [80.89451599121094, 60.695674896240234] data_time: 0.0376 time: 1.2835 +2023/06/06 11:52:11 - mmengine - INFO - Epoch(train) [10][ 100/3907] lr: 1.6744e-05 eta: 0:40:26 time: 0.6339 data_time: 0.0019 memory: 44137 loss: 0.3890 +2023/06/06 11:53:15 - mmengine - INFO - Epoch(train) [10][ 200/3907] lr: 1.6398e-05 eta: 0:39:23 time: 0.6344 data_time: 0.0015 memory: 44137 loss: 0.4330 +2023/06/06 11:54:18 - mmengine - INFO - Epoch(train) [10][ 300/3907] lr: 1.6059e-05 eta: 0:38:19 time: 0.6342 data_time: 0.0015 memory: 44137 loss: 0.4041 +2023/06/06 11:55:21 - mmengine - INFO - Epoch(train) [10][ 400/3907] lr: 1.5730e-05 eta: 0:37:15 time: 0.6340 data_time: 0.0015 memory: 44137 loss: 0.4127 +2023/06/06 11:56:25 - mmengine - INFO - Epoch(train) [10][ 500/3907] lr: 1.5410e-05 eta: 0:36:11 time: 0.6352 data_time: 0.0015 memory: 44137 loss: 0.4082 +2023/06/06 11:57:28 - mmengine - INFO - Epoch(train) [10][ 600/3907] lr: 1.5099e-05 eta: 0:35:07 time: 0.6344 data_time: 0.0016 memory: 44137 loss: 0.4406 +2023/06/06 11:58:32 - mmengine - INFO - Epoch(train) [10][ 700/3907] lr: 1.4797e-05 eta: 0:34:04 time: 0.6339 data_time: 0.0015 memory: 44137 loss: 0.4013 +2023/06/06 11:59:35 - mmengine - INFO - Epoch(train) [10][ 800/3907] lr: 1.4505e-05 eta: 0:33:00 time: 0.6340 data_time: 0.0015 memory: 44137 loss: 0.4200 +2023/06/06 11:59:59 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr3e-4_20230606_052113 +2023/06/06 12:00:39 - mmengine - INFO - Epoch(train) [10][ 900/3907] lr: 1.4221e-05 eta: 0:31:56 time: 0.6334 data_time: 0.0017 memory: 44137 loss: 0.4198 +2023/06/06 12:01:42 - mmengine - INFO - Epoch(train) [10][1000/3907] lr: 1.3946e-05 eta: 0:30:52 time: 0.6342 data_time: 0.0014 memory: 44137 loss: 0.3851 +2023/06/06 12:02:46 - mmengine - INFO - Epoch(train) [10][1100/3907] lr: 1.3680e-05 eta: 0:29:49 time: 0.6339 data_time: 0.0015 memory: 44137 loss: 0.4090 +2023/06/06 12:03:49 - mmengine - INFO - Epoch(train) [10][1200/3907] lr: 1.3424e-05 eta: 0:28:45 time: 0.6338 data_time: 0.0015 memory: 44137 loss: 0.4391 +2023/06/06 12:04:53 - mmengine - INFO - Epoch(train) [10][1300/3907] lr: 1.3177e-05 eta: 0:27:41 time: 0.6361 data_time: 0.0018 memory: 44137 loss: 0.4089 +2023/06/06 12:05:56 - mmengine - INFO - Epoch(train) [10][1400/3907] lr: 1.2939e-05 eta: 0:26:37 time: 0.6342 data_time: 0.0015 memory: 44137 loss: 0.4454 +2023/06/06 12:07:00 - mmengine - INFO - Epoch(train) [10][1500/3907] lr: 1.2710e-05 eta: 0:25:34 time: 0.6346 data_time: 0.0016 memory: 44137 loss: 0.4100 +2023/06/06 12:08:03 - mmengine - INFO - Epoch(train) [10][1600/3907] lr: 1.2490e-05 eta: 0:24:30 time: 0.6340 data_time: 0.0015 memory: 44137 loss: 0.3992 +2023/06/06 12:09:07 - mmengine - INFO - Epoch(train) [10][1700/3907] lr: 1.2279e-05 eta: 0:23:26 time: 0.6437 data_time: 0.0015 memory: 44137 loss: 0.4182 +2023/06/06 12:10:12 - mmengine - INFO - Epoch(train) [10][1800/3907] lr: 1.2078e-05 eta: 0:22:22 time: 0.6347 data_time: 0.0016 memory: 44137 loss: 0.3996 +2023/06/06 12:10:38 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr3e-4_20230606_052113 +2023/06/06 12:11:21 - mmengine - INFO - Epoch(train) [10][1900/3907] lr: 1.1886e-05 eta: 0:21:19 time: 0.6784 data_time: 0.0018 memory: 44137 loss: 0.4041 +2023/06/06 12:12:35 - mmengine - INFO - Epoch(train) [10][2000/3907] lr: 1.1703e-05 eta: 0:20:16 time: 0.6348 data_time: 0.0014 memory: 44137 loss: 0.3840 +2023/06/06 12:13:46 - mmengine - INFO - Epoch(train) [10][2100/3907] lr: 1.1530e-05 eta: 0:19:12 time: 0.6342 data_time: 0.0015 memory: 44137 loss: 0.3895 +2023/06/06 12:14:50 - mmengine - INFO - Epoch(train) [10][2200/3907] lr: 1.1365e-05 eta: 0:18:09 time: 0.6371 data_time: 0.0014 memory: 44137 loss: 0.3979 +2023/06/06 12:15:54 - mmengine - INFO - Epoch(train) [10][2300/3907] lr: 1.1210e-05 eta: 0:17:05 time: 0.6374 data_time: 0.0015 memory: 44137 loss: 0.4039 +2023/06/06 12:16:57 - mmengine - INFO - Epoch(train) [10][2400/3907] lr: 1.1065e-05 eta: 0:16:01 time: 0.6344 data_time: 0.0016 memory: 44137 loss: 0.3965 +2023/06/06 12:18:01 - mmengine - INFO - Epoch(train) [10][2500/3907] lr: 1.0928e-05 eta: 0:14:57 time: 0.6339 data_time: 0.0019 memory: 44137 loss: 0.4068 +2023/06/06 12:19:04 - mmengine - INFO - Epoch(train) [10][2600/3907] lr: 1.0801e-05 eta: 0:13:53 time: 0.6338 data_time: 0.0015 memory: 44137 loss: 0.4245 +2023/06/06 12:20:08 - mmengine - INFO - Epoch(train) [10][2700/3907] lr: 1.0684e-05 eta: 0:12:50 time: 0.6342 data_time: 0.0016 memory: 44137 loss: 0.4159 +2023/06/06 12:21:11 - mmengine - INFO - Epoch(train) [10][2800/3907] lr: 1.0575e-05 eta: 0:11:46 time: 0.6337 data_time: 0.0014 memory: 44137 loss: 0.4106 +2023/06/06 12:21:34 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr3e-4_20230606_052113 +2023/06/06 12:22:14 - mmengine - INFO - Epoch(train) [10][2900/3907] lr: 1.0476e-05 eta: 0:10:42 time: 0.6342 data_time: 0.0016 memory: 44137 loss: 0.4262 +2023/06/06 12:23:18 - mmengine - INFO - Epoch(train) [10][3000/3907] lr: 1.0386e-05 eta: 0:09:38 time: 0.6350 data_time: 0.0015 memory: 44137 loss: 0.3968 +2023/06/06 12:24:21 - mmengine - INFO - Epoch(train) [10][3100/3907] lr: 1.0306e-05 eta: 0:08:34 time: 0.6333 data_time: 0.0020 memory: 44137 loss: 0.4200 +2023/06/06 12:25:25 - mmengine - INFO - Epoch(train) [10][3200/3907] lr: 1.0235e-05 eta: 0:07:31 time: 0.6336 data_time: 0.0015 memory: 44137 loss: 0.4236 +2023/06/06 12:26:28 - mmengine - INFO - Epoch(train) [10][3300/3907] lr: 1.0173e-05 eta: 0:06:27 time: 0.6347 data_time: 0.0015 memory: 44137 loss: 0.4105 +2023/06/06 12:27:32 - mmengine - INFO - Epoch(train) [10][3400/3907] lr: 1.0121e-05 eta: 0:05:23 time: 0.6334 data_time: 0.0014 memory: 44137 loss: 0.4174 +2023/06/06 12:28:35 - mmengine - INFO - Epoch(train) [10][3500/3907] lr: 1.0078e-05 eta: 0:04:19 time: 0.6333 data_time: 0.0014 memory: 44137 loss: 0.3993 +2023/06/06 12:29:38 - mmengine - INFO - Epoch(train) [10][3600/3907] lr: 1.0044e-05 eta: 0:03:15 time: 0.6335 data_time: 0.0015 memory: 44137 loss: 0.3893 +2023/06/06 12:30:42 - mmengine - INFO - Epoch(train) [10][3700/3907] lr: 1.0020e-05 eta: 0:02:12 time: 0.6342 data_time: 0.0014 memory: 44137 loss: 0.4125 +2023/06/06 12:31:45 - mmengine - INFO - Epoch(train) [10][3800/3907] lr: 1.0005e-05 eta: 0:01:08 time: 0.6334 data_time: 0.0016 memory: 44137 loss: 0.4115 +2023/06/06 12:32:09 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr3e-4_20230606_052113 +2023/06/06 12:32:49 - mmengine - INFO - Epoch(train) [10][3900/3907] lr: 1.0000e-05 eta: 0:00:04 time: 0.6340 data_time: 0.0013 memory: 44137 loss: 0.4028 +2023/06/06 12:32:53 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr3e-4_20230606_052113 +2023/06/06 12:32:53 - mmengine - INFO - Saving checkpoint at 10 epochs +2023/06/06 12:34:35 - mmengine - INFO - Epoch(val) [10][57/57] accuracy/top1: 74.2911 single-label/precision_classwise: [68.50100708007812, 96.71131134033203] single-label/recall_classwise: [98.77532958984375, 44.22480773925781] single-label/f1-score_classwise: [80.8985824584961, 60.69471740722656] data_time: 0.0361 time: 1.2821 diff --git a/clip_large_pretrain_4x256_sdv2_lr3e-4/20230606_052113/vis_data/20230606_052113.json b/clip_large_pretrain_4x256_sdv2_lr3e-4/20230606_052113/vis_data/20230606_052113.json new file mode 100644 index 0000000000000000000000000000000000000000..24d296ee2b82acb398effeea94566fbb57949f7b --- /dev/null +++ b/clip_large_pretrain_4x256_sdv2_lr3e-4/20230606_052113/vis_data/20230606_052113.json @@ -0,0 +1,400 @@ +{"lr": 0.0002999954057034405, "data_time": 0.0016700506210327148, "loss": 0.4902272582054138, "time": 0.6309503555297852, "epoch": 1, "memory": 44138, "step": 100} +{"lr": 0.0002999814370149173, "data_time": 0.0016016960144042969, "loss": 0.49141762852668763, "time": 0.632881760597229, "epoch": 1, "memory": 44138, "step": 200} +{"lr": 0.00029995809436883357, "data_time": 0.0014659404754638673, "loss": 0.4493827402591705, "time": 0.6323542594909668, "epoch": 1, "memory": 44138, "step": 300} +{"lr": 0.00029992537927443715, "data_time": 0.0017125844955444337, "loss": 0.44257314801216124, "time": 0.6326935768127442, "epoch": 1, "memory": 44138, "step": 400} +{"lr": 0.0002998832938469641, "data_time": 0.0015325546264648438, "loss": 0.45193727016448976, "time": 0.6333640098571778, "epoch": 1, "memory": 44138, "step": 500} +{"lr": 0.0002998318408075004, "data_time": 0.0015497446060180665, "loss": 0.41731774210929873, "time": 0.6340202331542969, "epoch": 1, "memory": 44138, "step": 600} +{"lr": 0.0002997710234828077, "data_time": 0.0015933990478515625, "loss": 0.4037692368030548, "time": 0.632411789894104, "epoch": 1, "memory": 44138, "step": 700} +{"lr": 0.00029970084580510675, "data_time": 0.0015311479568481446, "loss": 0.4298222452402115, "time": 0.6337251424789428, "epoch": 1, "memory": 44138, "step": 800} +{"lr": 0.0002996213123118231, "data_time": 0.0015386819839477539, "loss": 0.41936435699462893, "time": 0.6332170009613037, "epoch": 1, "memory": 44138, "step": 900} +{"lr": 0.0002995324281452956, "data_time": 0.0015138149261474609, "loss": 0.4237715512514114, "time": 0.6327500343322754, "epoch": 1, "memory": 44138, "step": 1000} +{"lr": 0.00029943419905244207, "data_time": 0.0015166044235229493, "loss": 0.4082989066839218, "time": 0.6326113700866699, "epoch": 1, "memory": 44138, "step": 1100} +{"lr": 0.000299326631384388, "data_time": 0.0014206647872924804, "loss": 0.4096844166517258, "time": 0.6304686307907105, "epoch": 1, "memory": 44138, "step": 1200} +{"lr": 0.0002992097320960566, "data_time": 0.0014466524124145507, "loss": 0.37484508752822876, "time": 0.6327450037002563, "epoch": 1, "memory": 44138, "step": 1300} +{"lr": 0.00029908350874571896, "data_time": 0.0015225648880004884, "loss": 0.3997973918914795, "time": 0.6336121082305908, "epoch": 1, "memory": 44138, "step": 1400} +{"lr": 0.00029894796949450446, "data_time": 0.0015665054321289062, "loss": 0.39717928469181063, "time": 0.6327163457870484, "epoch": 1, "memory": 44138, "step": 1500} +{"lr": 0.00029880312310587493, "data_time": 0.0015705108642578125, "loss": 0.4122603118419647, "time": 0.6335359811782837, "epoch": 1, "memory": 44138, "step": 1600} +{"lr": 0.00029864897894505576, "data_time": 0.001616358757019043, "loss": 0.4264645129442215, "time": 0.6329888582229615, "epoch": 1, "memory": 44138, "step": 1700} +{"lr": 0.0002984855469784324, "data_time": 0.0016736268997192382, "loss": 0.39207322597503663, "time": 0.6333279371261596, "epoch": 1, "memory": 44138, "step": 1800} +{"lr": 0.0002983128377729035, "data_time": 0.0013751745223999023, "loss": 0.3964429199695587, "time": 0.6329690933227539, "epoch": 1, "memory": 44138, "step": 1900} +{"lr": 0.0002981308624952014, "data_time": 0.0013938665390014649, "loss": 0.37336658239364623, "time": 0.6348889589309692, "epoch": 1, "memory": 44138, "step": 2000} +{"lr": 0.0002979396329111671, "data_time": 0.0014202356338500976, "loss": 0.3938830614089966, "time": 0.6333211183547973, "epoch": 1, "memory": 44138, "step": 2100} +{"lr": 0.00029773916138499, "data_time": 0.001470470428466797, "loss": 0.3555929780006409, "time": 0.6339236736297608, "epoch": 1, "memory": 44138, "step": 2200} +{"lr": 0.00029752946087840954, "data_time": 0.0014882564544677734, "loss": 0.3780499905347824, "time": 0.6328037261962891, "epoch": 1, "memory": 44138, "step": 2300} +{"lr": 0.00029731054494987663, "data_time": 0.0014256000518798827, "loss": 0.39369869530200957, "time": 0.6329768419265747, "epoch": 1, "memory": 44138, "step": 2400} +{"lr": 0.0002970824277536758, "data_time": 0.001444244384765625, "loss": 0.3938509404659271, "time": 0.6332411289215087, "epoch": 1, "memory": 44138, "step": 2500} +{"lr": 0.0002968451240390124, "data_time": 0.0014374971389770508, "loss": 0.37504111528396605, "time": 0.6336678028106689, "epoch": 1, "memory": 44138, "step": 2600} +{"lr": 0.0002965986491490574, "data_time": 0.001483774185180664, "loss": 0.376563623547554, "time": 0.6326481103897095, "epoch": 1, "memory": 44138, "step": 2700} +{"lr": 0.00029634301901995517, "data_time": 0.001518845558166504, "loss": 0.39616626501083374, "time": 0.6332586526870727, "epoch": 1, "memory": 44138, "step": 2800} +{"lr": 0.000296078250179794, "data_time": 0.0015430450439453125, "loss": 0.37575416564941405, "time": 0.6331181526184082, "epoch": 1, "memory": 44138, "step": 2900} +{"lr": 0.0002958043597475368, "data_time": 0.0016086816787719727, "loss": 0.3715750277042389, "time": 0.6327194690704345, "epoch": 1, "memory": 44138, "step": 3000} +{"lr": 0.0002955213654319153, "data_time": 0.0015376806259155273, "loss": 0.3767815738916397, "time": 0.6324351787567138, "epoch": 1, "memory": 44138, "step": 3100} +{"lr": 0.00029522928553028316, "data_time": 0.0016405582427978516, "loss": 0.3875528693199158, "time": 0.6325571060180664, "epoch": 1, "memory": 44138, "step": 3200} +{"lr": 0.0002949281389274349, "data_time": 0.0014760017395019532, "loss": 0.38488066792488096, "time": 0.6327473163604737, "epoch": 1, "memory": 44138, "step": 3300} +{"lr": 0.0002946179450943832, "data_time": 0.0015089988708496093, "loss": 0.39545409083366395, "time": 0.6326494216918945, "epoch": 1, "memory": 44138, "step": 3400} +{"lr": 0.00029429872408710166, "data_time": 0.0014854669570922852, "loss": 0.3835848271846771, "time": 0.6328840970993042, "epoch": 1, "memory": 44138, "step": 3500} +{"lr": 0.0002939704965452269, "data_time": 0.0015824794769287109, "loss": 0.37650732398033143, "time": 0.6334297895431519, "epoch": 1, "memory": 44138, "step": 3600} +{"lr": 0.00029363328369072446, "data_time": 0.0016147375106811523, "loss": 0.36214399933815, "time": 0.6332512617111206, "epoch": 1, "memory": 44138, "step": 3700} +{"lr": 0.00029328710732651664, "data_time": 0.0015808343887329102, "loss": 0.35871745049953463, "time": 0.632096529006958, "epoch": 1, "memory": 44138, "step": 3800} +{"lr": 0.0002929319898350724, "data_time": 0.0012046575546264648, "loss": 0.3764660030603409, "time": 0.6321643352508545, "epoch": 1, "memory": 44138, "step": 3900} +{"accuracy/top1": 65.90585327148438, "data_time": 0.045427987450047544, "time": 1.3536268326274135, "step": 1} +{"lr": 0.0002925421383003051, "data_time": 0.001694941520690918, "loss": 0.3807366907596588, "time": 0.6337023973464966, "epoch": 2, "memory": 44138, "step": 4007} +{"lr": 0.0002921685862849842, "data_time": 0.0014974832534790038, "loss": 0.3624882698059082, "time": 0.6346543073654175, "epoch": 2, "memory": 44137, "step": 4107} +{"lr": 0.0002917861654618148, "data_time": 0.0014608383178710937, "loss": 0.3827891707420349, "time": 0.6333630323410034, "epoch": 2, "memory": 44137, "step": 4207} +{"lr": 0.0002913949005566972, "data_time": 0.0014677047729492188, "loss": 0.3831400215625763, "time": 0.6331991672515869, "epoch": 2, "memory": 44137, "step": 4307} +{"lr": 0.0002909948168673564, "data_time": 0.0014125823974609375, "loss": 0.3692391812801361, "time": 0.6360158443450927, "epoch": 2, "memory": 44137, "step": 4407} +{"lr": 0.00029058594026170915, "data_time": 0.0013590097427368165, "loss": 0.3524982511997223, "time": 0.634480357170105, "epoch": 2, "memory": 44137, "step": 4507} +{"lr": 0.0002901682971761867, "data_time": 0.0015326976776123048, "loss": 0.35894723534584044, "time": 0.6338254690170289, "epoch": 2, "memory": 44137, "step": 4607} +{"lr": 0.00028974191461402947, "data_time": 0.0013803958892822266, "loss": 0.36094541251659396, "time": 0.632565188407898, "epoch": 2, "memory": 44137, "step": 4707} +{"lr": 0.00028930682014353937, "data_time": 0.0014656305313110352, "loss": 0.37468068301677704, "time": 0.634208345413208, "epoch": 2, "memory": 44137, "step": 4807} +{"lr": 0.0002888630418962978, "data_time": 0.001390552520751953, "loss": 0.38818833529949187, "time": 0.6345075607299805, "epoch": 2, "memory": 44137, "step": 4907} +{"lr": 0.00028841060856534716, "data_time": 0.0018620491027832031, "loss": 0.35287671387195585, "time": 0.6331895112991333, "epoch": 2, "memory": 44137, "step": 5007} +{"lr": 0.0002879495494033338, "data_time": 0.0014444351196289062, "loss": 0.38743217587471007, "time": 0.633136248588562, "epoch": 2, "memory": 44137, "step": 5107} +{"lr": 0.00028747989422061985, "data_time": 0.0013684988021850585, "loss": 0.38438221514225007, "time": 0.6332125663757324, "epoch": 2, "memory": 44137, "step": 5207} +{"lr": 0.0002870016733833529, "data_time": 0.0014147520065307616, "loss": 0.37357968688011167, "time": 0.6330580472946167, "epoch": 2, "memory": 44137, "step": 5307} +{"lr": 0.00028651491781150423, "data_time": 0.001379537582397461, "loss": 0.38055368661880495, "time": 0.6336659908294677, "epoch": 2, "memory": 44137, "step": 5407} +{"lr": 0.00028601965897686856, "data_time": 0.0012953758239746093, "loss": 0.37099160850048063, "time": 0.6349786281585693, "epoch": 2, "memory": 44137, "step": 5507} +{"lr": 0.0002855159289010302, "data_time": 0.001461935043334961, "loss": 0.37708382308483124, "time": 0.643999171257019, "epoch": 2, "memory": 44137, "step": 5607} +{"lr": 0.00028500376015329224, "data_time": 0.0014859914779663086, "loss": 0.38429166078567506, "time": 0.633487343788147, "epoch": 2, "memory": 44137, "step": 5707} +{"lr": 0.00028448318584857045, "data_time": 0.0015755414962768555, "loss": 0.4045539706945419, "time": 0.6334191083908081, "epoch": 2, "memory": 44137, "step": 5807} +{"lr": 0.000283954239645253, "data_time": 0.0013811826705932618, "loss": 0.36398411393165586, "time": 0.6341552734375, "epoch": 2, "memory": 44137, "step": 5907} +{"lr": 0.0002834169557430233, "data_time": 0.0014555215835571288, "loss": 0.3547777682542801, "time": 0.6336054325103759, "epoch": 2, "memory": 44137, "step": 6007} +{"lr": 0.00028287136888064903, "data_time": 0.00139312744140625, "loss": 0.37483089566230776, "time": 0.6332645893096924, "epoch": 2, "memory": 44137, "step": 6107} +{"lr": 0.00028231751433373687, "data_time": 0.0014421939849853516, "loss": 0.36126171946525576, "time": 0.6331502199172974, "epoch": 2, "memory": 44137, "step": 6207} +{"lr": 0.00028175542791245077, "data_time": 0.0015299081802368163, "loss": 0.3846882224082947, "time": 0.6347301483154297, "epoch": 2, "memory": 44137, "step": 6307} +{"lr": 0.0002811851459591966, "data_time": 0.0013877391815185548, "loss": 0.3941593259572983, "time": 0.6332796812057495, "epoch": 2, "memory": 44137, "step": 6407} +{"lr": 0.00028060670534627274, "data_time": 0.0014113187789916992, "loss": 0.3431746423244476, "time": 0.6329907894134521, "epoch": 2, "memory": 44137, "step": 6507} +{"lr": 0.00028002014347348626, "data_time": 0.0013391494750976563, "loss": 0.4181104600429535, "time": 0.6329025030136108, "epoch": 2, "memory": 44137, "step": 6607} +{"lr": 0.00027942549826573453, "data_time": 0.0014209985733032227, "loss": 0.35405783355236053, "time": 0.6330006361007691, "epoch": 2, "memory": 44137, "step": 6707} +{"lr": 0.00027882280817055275, "data_time": 0.0015227079391479492, "loss": 0.3673249214887619, "time": 0.6331135272979737, "epoch": 2, "memory": 44137, "step": 6807} +{"lr": 0.000278212112155629, "data_time": 0.0014508724212646484, "loss": 0.3860452800989151, "time": 0.6328620433807373, "epoch": 2, "memory": 44137, "step": 6907} +{"lr": 0.0002775934497062837, "data_time": 0.0015659093856811523, "loss": 0.36105334758758545, "time": 0.6331103563308715, "epoch": 2, "memory": 44137, "step": 7007} +{"lr": 0.00027696686082291747, "data_time": 0.0014065265655517577, "loss": 0.39859779477119445, "time": 0.6327624797821045, "epoch": 2, "memory": 44137, "step": 7107} +{"lr": 0.00027633238601842443, "data_time": 0.0013328313827514649, "loss": 0.3689152032136917, "time": 0.6332187652587891, "epoch": 2, "memory": 44137, "step": 7207} +{"lr": 0.0002756900663155731, "data_time": 0.0013592243194580078, "loss": 0.3702657222747803, "time": 0.6327556848526001, "epoch": 2, "memory": 44137, "step": 7307} +{"lr": 0.0002750399432443529, "data_time": 0.0014490842819213866, "loss": 0.4029386878013611, "time": 0.6327578306198121, "epoch": 2, "memory": 44137, "step": 7407} +{"lr": 0.0002743820588392909, "data_time": 0.0014390945434570312, "loss": 0.3923063039779663, "time": 0.6333898305892944, "epoch": 2, "memory": 44137, "step": 7507} +{"lr": 0.00027371645563673247, "data_time": 0.0015081405639648438, "loss": 0.37271900177001954, "time": 0.6331567287445068, "epoch": 2, "memory": 44137, "step": 7607} +{"lr": 0.0002730431766720916, "data_time": 0.0014412403106689453, "loss": 0.3904691100120544, "time": 0.6326159238815308, "epoch": 2, "memory": 44137, "step": 7707} +{"lr": 0.0002723622654770684, "data_time": 0.0013747930526733399, "loss": 0.3633644700050354, "time": 0.6329151868820191, "epoch": 2, "memory": 44137, "step": 7807} +{"accuracy/top1": 73.03504180908203, "data_time": 0.041268060947286675, "time": 1.2850785748711948, "step": 2} +{"lr": 0.0002716252880879063, "data_time": 0.0014528512954711914, "loss": 0.39044045805931094, "time": 0.6343801021575928, "epoch": 3, "memory": 44137, "step": 7914} +{"lr": 0.000270928718619737, "data_time": 0.0015842676162719726, "loss": 0.350239160656929, "time": 0.6342518091201782, "epoch": 3, "memory": 44137, "step": 8014} +{"lr": 0.0002702246536341361, "data_time": 0.001535344123840332, "loss": 0.3971865147352219, "time": 0.6333378791809082, "epoch": 3, "memory": 44137, "step": 8114} +{"lr": 0.0002695131386533127, "data_time": 0.001560378074645996, "loss": 0.36411244571208956, "time": 0.635346508026123, "epoch": 3, "memory": 44137, "step": 8214} +{"lr": 0.0002687942196811649, "data_time": 0.0014538288116455079, "loss": 0.4046099454164505, "time": 0.63466477394104, "epoch": 3, "memory": 44137, "step": 8314} +{"lr": 0.0002680679432003047, "data_time": 0.0013802051544189453, "loss": 0.37574750781059263, "time": 0.6367144346237182, "epoch": 3, "memory": 44137, "step": 8414} +{"lr": 0.0002673343561690538, "data_time": 0.0014797210693359374, "loss": 0.37542503178119657, "time": 0.633153247833252, "epoch": 3, "memory": 44137, "step": 8514} +{"lr": 0.00026659350601840597, "data_time": 0.0014623641967773438, "loss": 0.38963502645492554, "time": 0.642594575881958, "epoch": 3, "memory": 44137, "step": 8614} +{"lr": 0.0002658454406489622, "data_time": 0.0013822317123413086, "loss": 0.3794644445180893, "time": 0.6330184459686279, "epoch": 3, "memory": 44137, "step": 8714} +{"lr": 0.0002650902084278314, "data_time": 0.0014023303985595704, "loss": 0.3861773073673248, "time": 0.6335261821746826, "epoch": 3, "memory": 44137, "step": 8814} +{"lr": 0.00026432785818550494, "data_time": 0.0014488935470581056, "loss": 0.4059780925512314, "time": 0.6336912870407104, "epoch": 3, "memory": 44137, "step": 8914} +{"lr": 0.00026355843921269914, "data_time": 0.00146939754486084, "loss": 0.3851300150156021, "time": 0.6342427968978882, "epoch": 3, "memory": 44137, "step": 9014} +{"lr": 0.0002627820012571672, "data_time": 0.0013885736465454102, "loss": 0.3818380951881409, "time": 0.6332744121551513, "epoch": 3, "memory": 44137, "step": 9114} +{"lr": 0.00026199859452048346, "data_time": 0.0013898372650146484, "loss": 0.3860693693161011, "time": 0.6335134029388427, "epoch": 3, "memory": 44137, "step": 9214} +{"lr": 0.00026120826965479813, "data_time": 0.0015716552734375, "loss": 0.4048592269420624, "time": 0.6336831331253052, "epoch": 3, "memory": 44137, "step": 9314} +{"lr": 0.0002604110777595625, "data_time": 0.0014738082885742188, "loss": 0.3931962579488754, "time": 0.6340368270874024, "epoch": 3, "memory": 44137, "step": 9414} +{"lr": 0.000259607070378222, "data_time": 0.0016142606735229492, "loss": 0.36188422739505766, "time": 0.6334050178527832, "epoch": 3, "memory": 44137, "step": 9514} +{"lr": 0.00025879629949488836, "data_time": 0.0014542341232299805, "loss": 0.3689572334289551, "time": 0.6326395273208618, "epoch": 3, "memory": 44137, "step": 9614} +{"lr": 0.00025797881753097425, "data_time": 0.0013261795043945312, "loss": 0.3661950796842575, "time": 0.6339035511016846, "epoch": 3, "memory": 44137, "step": 9714} +{"lr": 0.000257154677341807, "data_time": 0.001378941535949707, "loss": 0.4190051108598709, "time": 0.6333179950714112, "epoch": 3, "memory": 44137, "step": 9814} +{"lr": 0.0002563239322132094, "data_time": 0.0014253377914428711, "loss": 0.4014905631542206, "time": 0.6335712909698487, "epoch": 3, "memory": 44137, "step": 9914} +{"lr": 0.00025548663585805504, "data_time": 0.001434659957885742, "loss": 0.352135044336319, "time": 0.6362543106079102, "epoch": 3, "memory": 44137, "step": 10014} +{"lr": 0.0002546428424127946, "data_time": 0.001401805877685547, "loss": 0.3941739499568939, "time": 0.634401512145996, "epoch": 3, "memory": 44137, "step": 10114} +{"lr": 0.00025379260643395755, "data_time": 0.0014047861099243165, "loss": 0.36856828927993773, "time": 0.6330304861068725, "epoch": 3, "memory": 44137, "step": 10214} +{"lr": 0.0002529359828946234, "data_time": 0.0013712644577026367, "loss": 0.39322973489761354, "time": 0.6345672130584716, "epoch": 3, "memory": 44137, "step": 10314} +{"lr": 0.00025207302718086624, "data_time": 0.0014400005340576172, "loss": 0.3807382553815842, "time": 0.6328573226928711, "epoch": 3, "memory": 44137, "step": 10414} +{"lr": 0.00025120379508817534, "data_time": 0.0014844417572021484, "loss": 0.3917868763208389, "time": 0.6344710350036621, "epoch": 3, "memory": 44137, "step": 10514} +{"lr": 0.0002503283428178471, "data_time": 0.0014474868774414062, "loss": 0.41173630952835083, "time": 0.6344508171081543, "epoch": 3, "memory": 44137, "step": 10614} +{"lr": 0.00024944672697335175, "data_time": 0.0015144586563110352, "loss": 0.3974515050649643, "time": 0.6331386089324951, "epoch": 3, "memory": 44137, "step": 10714} +{"lr": 0.00024855900455667294, "data_time": 0.0013801336288452148, "loss": 0.39543994665145876, "time": 0.6353105545043946, "epoch": 3, "memory": 44137, "step": 10814} +{"lr": 0.0002476652329646227, "data_time": 0.0014162063598632812, "loss": 0.3588362991809845, "time": 0.6332003116607666, "epoch": 3, "memory": 44137, "step": 10914} +{"lr": 0.0002467654699851294, "data_time": 0.0013519287109375, "loss": 0.40977767705917356, "time": 0.6330472230911255, "epoch": 3, "memory": 44137, "step": 11014} +{"lr": 0.000245859773793503, "data_time": 0.0014048099517822265, "loss": 0.40263975858688356, "time": 0.6333052396774292, "epoch": 3, "memory": 44137, "step": 11114} +{"lr": 0.0002449482029486725, "data_time": 0.0016291618347167968, "loss": 0.38870387375354765, "time": 0.6329368352890015, "epoch": 3, "memory": 44137, "step": 11214} +{"lr": 0.00024403081638939956, "data_time": 0.0015357494354248046, "loss": 0.38911502361297606, "time": 0.6339688539505005, "epoch": 3, "memory": 44137, "step": 11314} +{"lr": 0.00024310767343046748, "data_time": 0.0015012741088867188, "loss": 0.41671150028705595, "time": 0.6327686548233032, "epoch": 3, "memory": 44137, "step": 11414} +{"lr": 0.00024217883375884914, "data_time": 0.0014972686767578125, "loss": 0.3951862722635269, "time": 0.6337986707687377, "epoch": 3, "memory": 44137, "step": 11514} +{"lr": 0.0002412443574298446, "data_time": 0.0015561342239379882, "loss": 0.37027890980243683, "time": 0.6331808567047119, "epoch": 3, "memory": 44137, "step": 11614} +{"lr": 0.00024030430486319854, "data_time": 0.001108551025390625, "loss": 0.4004129350185394, "time": 0.6335879802703858, "epoch": 3, "memory": 44137, "step": 11714} +{"accuracy/top1": 73.64740753173828, "data_time": 0.03590079833721292, "time": 1.284608479203849, "step": 3} +{"lr": 0.00023929234210131621, "data_time": 0.0015595912933349609, "loss": 0.3860115647315979, "time": 0.6335151195526123, "epoch": 4, "memory": 44137, "step": 11821} +{"lr": 0.0002383409402531276, "data_time": 0.001466655731201172, "loss": 0.3927760601043701, "time": 0.6331015348434448, "epoch": 4, "memory": 44137, "step": 11921} +{"lr": 0.00023738414989139526, "data_time": 0.0014872312545776366, "loss": 0.4064512252807617, "time": 0.6349523067474365, "epoch": 4, "memory": 44137, "step": 12021} +{"lr": 0.0002364220328786055, "data_time": 0.0014856100082397462, "loss": 0.397469300031662, "time": 0.6337270736694336, "epoch": 4, "memory": 44137, "step": 12121} +{"lr": 0.0002354546514216467, "data_time": 0.0015802621841430665, "loss": 0.3992577761411667, "time": 0.6337252855300903, "epoch": 4, "memory": 44137, "step": 12221} +{"lr": 0.00023448206806778656, "data_time": 0.001455974578857422, "loss": 0.3807056397199631, "time": 0.6341383218765259, "epoch": 4, "memory": 44137, "step": 12321} +{"lr": 0.0002335043457006275, "data_time": 0.0014228105545043945, "loss": 0.3721867769956589, "time": 0.6345484018325805, "epoch": 4, "memory": 44137, "step": 12421} +{"lr": 0.00023252154753604175, "data_time": 0.0016866207122802734, "loss": 0.40315210819244385, "time": 0.6342388391494751, "epoch": 4, "memory": 44137, "step": 12521} +{"lr": 0.0002315337371180835, "data_time": 0.0015563488006591797, "loss": 0.40187896490097047, "time": 0.6331060886383056, "epoch": 4, "memory": 44137, "step": 12621} +{"lr": 0.00023054097831488017, "data_time": 0.0014397859573364257, "loss": 0.4120116919279099, "time": 0.6351639270782471, "epoch": 4, "memory": 44137, "step": 12721} +{"lr": 0.00022954333531450405, "data_time": 0.0015584945678710938, "loss": 0.39791989624500274, "time": 0.6338698148727417, "epoch": 4, "memory": 44137, "step": 12821} +{"lr": 0.00022854087262082044, "data_time": 0.0015718460083007813, "loss": 0.4025317281484604, "time": 0.634430193901062, "epoch": 4, "memory": 44137, "step": 12921} +{"lr": 0.00022753365504931858, "data_time": 0.001687312126159668, "loss": 0.3890873521566391, "time": 0.6327084541320801, "epoch": 4, "memory": 44137, "step": 13021} +{"lr": 0.0002265217477229201, "data_time": 0.0014850854873657226, "loss": 0.37108676731586454, "time": 0.6341103792190552, "epoch": 4, "memory": 44137, "step": 13121} +{"lr": 0.0002255052160677689, "data_time": 0.0014713048934936524, "loss": 0.3800925225019455, "time": 0.6344964742660523, "epoch": 4, "memory": 44137, "step": 13221} +{"lr": 0.00022448412580900032, "data_time": 0.0014410734176635743, "loss": 0.38547907769680023, "time": 0.6336946725845337, "epoch": 4, "memory": 44137, "step": 13321} +{"lr": 0.0002234585429664922, "data_time": 0.001412177085876465, "loss": 0.3935152441263199, "time": 0.6346070766448975, "epoch": 4, "memory": 44137, "step": 13421} +{"lr": 0.0002224285338505958, "data_time": 0.001433086395263672, "loss": 0.3959517806768417, "time": 0.6363736391067505, "epoch": 4, "memory": 44137, "step": 13521} +{"lr": 0.00022139416505784887, "data_time": 0.0014551877975463867, "loss": 0.4125809222459793, "time": 0.6373970031738281, "epoch": 4, "memory": 44137, "step": 13621} +{"lr": 0.0002203555034666699, "data_time": 0.0014021635055541993, "loss": 0.40133764743804934, "time": 0.6338953733444214, "epoch": 4, "memory": 44137, "step": 13721} +{"lr": 0.00021931261623303318, "data_time": 0.001403641700744629, "loss": 0.40483948588371277, "time": 0.6342689752578735, "epoch": 4, "memory": 44137, "step": 13821} +{"lr": 0.00021826557078612706, "data_time": 0.0014425992965698242, "loss": 0.40545822083950045, "time": 0.6332755088806152, "epoch": 4, "memory": 44137, "step": 13921} +{"lr": 0.0002172144348239949, "data_time": 0.0014600753784179688, "loss": 0.407717701792717, "time": 0.6354591369628906, "epoch": 4, "memory": 44137, "step": 14021} +{"lr": 0.0002161592763091573, "data_time": 0.0014806032180786134, "loss": 0.4162604361772537, "time": 0.6345121622085571, "epoch": 4, "memory": 44137, "step": 14121} +{"lr": 0.00021510016346421842, "data_time": 0.0015603065490722655, "loss": 0.37755178809165957, "time": 0.6340918064117431, "epoch": 4, "memory": 44137, "step": 14221} +{"lr": 0.0002140371647674538, "data_time": 0.0013860940933227539, "loss": 0.42633184790611267, "time": 0.633940577507019, "epoch": 4, "memory": 44137, "step": 14321} +{"lr": 0.0002129703489483841, "data_time": 0.00148162841796875, "loss": 0.3801177591085434, "time": 0.6336155891418457, "epoch": 4, "memory": 44137, "step": 14421} +{"lr": 0.00021189978498333095, "data_time": 0.0013858795166015625, "loss": 0.3921238601207733, "time": 0.6334233045578003, "epoch": 4, "memory": 44137, "step": 14521} +{"lr": 0.00021082554209095715, "data_time": 0.001500701904296875, "loss": 0.38124058544635775, "time": 0.6341081857681274, "epoch": 4, "memory": 44137, "step": 14621} +{"lr": 0.0002097476897277908, "data_time": 0.0014397621154785157, "loss": 0.3904690682888031, "time": 0.6345602035522461, "epoch": 4, "memory": 44137, "step": 14721} +{"lr": 0.00020866629758373506, "data_time": 0.0017169952392578126, "loss": 0.4120761275291443, "time": 0.6331814527511597, "epoch": 4, "memory": 44137, "step": 14821} +{"lr": 0.0002075814355775621, "data_time": 0.0015081167221069336, "loss": 0.3612970173358917, "time": 0.6327624082565307, "epoch": 4, "memory": 44137, "step": 14921} +{"lr": 0.0002064931738523926, "data_time": 0.001503157615661621, "loss": 0.37701648473739624, "time": 0.6338628053665161, "epoch": 4, "memory": 44137, "step": 15021} +{"lr": 0.00020540158277115988, "data_time": 0.0015111684799194336, "loss": 0.3903593778610229, "time": 0.634803295135498, "epoch": 4, "memory": 44137, "step": 15121} +{"lr": 0.00020430673291206152, "data_time": 0.0015198469161987304, "loss": 0.39292154014110564, "time": 0.634749436378479, "epoch": 4, "memory": 44137, "step": 15221} +{"lr": 0.00020320869506399518, "data_time": 0.0014705657958984375, "loss": 0.4145514190196991, "time": 0.6337226390838623, "epoch": 4, "memory": 44137, "step": 15321} +{"lr": 0.00020210754022198212, "data_time": 0.0014455318450927734, "loss": 0.38167398869991304, "time": 0.6330749750137329, "epoch": 4, "memory": 44137, "step": 15421} +{"lr": 0.00020100333958257622, "data_time": 0.0016491413116455078, "loss": 0.3898289382457733, "time": 0.6344200611114502, "epoch": 4, "memory": 44137, "step": 15521} +{"lr": 0.00019989616453926207, "data_time": 0.001222681999206543, "loss": 0.4088809311389923, "time": 0.6353185415267945, "epoch": 4, "memory": 44137, "step": 15621} +{"accuracy/top1": 74.10319519042969, "data_time": 0.03870173980449808, "time": 1.2833262764174362, "step": 4} +{"lr": 0.00019870827437055215, "data_time": 0.0015145540237426758, "loss": 0.38957429826259615, "time": 0.6335341453552246, "epoch": 5, "memory": 44137, "step": 15728} +{"lr": 0.00019759516998405188, "data_time": 0.0014893293380737304, "loss": 0.4005758613348007, "time": 0.6336127996444703, "epoch": 5, "memory": 44137, "step": 15828} +{"lr": 0.0001964793115531446, "data_time": 0.001859760284423828, "loss": 0.3996329098939896, "time": 0.6330718040466309, "epoch": 5, "memory": 44137, "step": 15928} +{"lr": 0.00019536077122506366, "data_time": 0.0014342784881591797, "loss": 0.41851462721824645, "time": 0.6336797952651978, "epoch": 5, "memory": 44137, "step": 16028} +{"lr": 0.00019423962132044425, "data_time": 0.0013837099075317382, "loss": 0.3988448053598404, "time": 0.6334346771240235, "epoch": 5, "memory": 44137, "step": 16128} +{"lr": 0.00019311593432864616, "data_time": 0.0014795780181884766, "loss": 0.4061956971883774, "time": 0.6338543653488159, "epoch": 5, "memory": 44137, "step": 16228} +{"lr": 0.00019198978290306865, "data_time": 0.0014634370803833009, "loss": 0.3885423570871353, "time": 0.6336427688598633, "epoch": 5, "memory": 44137, "step": 16328} +{"lr": 0.00019086123985645166, "data_time": 0.0016359806060791016, "loss": 0.4009571045637131, "time": 0.6340352773666382, "epoch": 5, "memory": 44137, "step": 16428} +{"lr": 0.00018973037815616829, "data_time": 0.0016285181045532227, "loss": 0.38526704609394075, "time": 0.6328840255737305, "epoch": 5, "memory": 44137, "step": 16528} +{"lr": 0.00018859727091950751, "data_time": 0.0017131567001342773, "loss": 0.40259073078632357, "time": 0.6333629846572876, "epoch": 5, "memory": 44137, "step": 16628} +{"lr": 0.0001874619914089457, "data_time": 0.0014320135116577149, "loss": 0.40346805155277254, "time": 0.6330579996109009, "epoch": 5, "memory": 44137, "step": 16728} +{"lr": 0.0001863246130274106, "data_time": 0.0013761520385742188, "loss": 0.4131861627101898, "time": 0.6331458330154419, "epoch": 5, "memory": 44137, "step": 16828} +{"lr": 0.00018518520931353537, "data_time": 0.0015362739562988282, "loss": 0.366819965839386, "time": 0.6339886665344239, "epoch": 5, "memory": 44137, "step": 16928} +{"lr": 0.00018404385393690312, "data_time": 0.0014780521392822265, "loss": 0.4048200726509094, "time": 0.6333035230636597, "epoch": 5, "memory": 44137, "step": 17028} +{"lr": 0.00018290062069328442, "data_time": 0.0015853643417358398, "loss": 0.39369923770427706, "time": 0.6330546617507935, "epoch": 5, "memory": 44137, "step": 17128} +{"lr": 0.00018175558349986583, "data_time": 0.0014896392822265625, "loss": 0.40191271007061, "time": 0.6332434892654419, "epoch": 5, "memory": 44137, "step": 17228} +{"lr": 0.00018060881639047043, "data_time": 0.0014326810836791993, "loss": 0.38185541331768036, "time": 0.6329642057418823, "epoch": 5, "memory": 44137, "step": 17328} +{"lr": 0.00017946039351077116, "data_time": 0.00143740177154541, "loss": 0.4041925251483917, "time": 0.6335988521575928, "epoch": 5, "memory": 44137, "step": 17428} +{"lr": 0.000178310389113497, "data_time": 0.0015124797821044922, "loss": 0.39412103295326234, "time": 0.6328449487686157, "epoch": 5, "memory": 44137, "step": 17528} +{"lr": 0.00017715887755363153, "data_time": 0.0014708995819091796, "loss": 0.3919717252254486, "time": 0.6330690383911133, "epoch": 5, "memory": 44137, "step": 17628} +{"lr": 0.00017600593328360648, "data_time": 0.0014266729354858398, "loss": 0.4195596158504486, "time": 0.6335479021072388, "epoch": 5, "memory": 44137, "step": 17728} +{"lr": 0.00017485163084848682, "data_time": 0.0016534090042114257, "loss": 0.38832607865333557, "time": 0.6329854011535645, "epoch": 5, "memory": 44137, "step": 17828} +{"lr": 0.00017369604488115154, "data_time": 0.0015085458755493164, "loss": 0.41594268679618834, "time": 0.633428406715393, "epoch": 5, "memory": 44137, "step": 17928} +{"lr": 0.00017253925009746786, "data_time": 0.0013206720352172852, "loss": 0.3879138708114624, "time": 0.6330734968185425, "epoch": 5, "memory": 44137, "step": 18028} +{"lr": 0.0001713813212914606, "data_time": 0.0014251470565795898, "loss": 0.430080246925354, "time": 0.632841682434082, "epoch": 5, "memory": 44137, "step": 18128} +{"lr": 0.00017022233333047657, "data_time": 0.0015694618225097657, "loss": 0.4012255847454071, "time": 0.633358359336853, "epoch": 5, "memory": 44137, "step": 18228} +{"lr": 0.0001690623611503431, "data_time": 0.0015277385711669922, "loss": 0.3898197740316391, "time": 0.6365635633468628, "epoch": 5, "memory": 44137, "step": 18328} +{"lr": 0.00016790147975052384, "data_time": 0.0013677835464477538, "loss": 0.4272844076156616, "time": 0.6342239379882812, "epoch": 5, "memory": 44137, "step": 18428} +{"lr": 0.00016673976418926875, "data_time": 0.0014503002166748047, "loss": 0.38555141985416413, "time": 0.6334893226623535, "epoch": 5, "memory": 44137, "step": 18528} +{"lr": 0.00016557728957876201, "data_time": 0.0015038251876831055, "loss": 0.3669358015060425, "time": 0.6338192701339722, "epoch": 5, "memory": 44137, "step": 18628} +{"lr": 0.0001644141310802647, "data_time": 0.0014124393463134765, "loss": 0.4031758517026901, "time": 0.6337247133255005, "epoch": 5, "memory": 44137, "step": 18728} +{"lr": 0.0001632503638992558, "data_time": 0.0014987230300903321, "loss": 0.4009410500526428, "time": 0.6338726997375488, "epoch": 5, "memory": 44137, "step": 18828} +{"lr": 0.00016208606328056935, "data_time": 0.0014945745468139648, "loss": 0.38522562980651853, "time": 0.6337974071502686, "epoch": 5, "memory": 44137, "step": 18928} +{"lr": 0.00016092130450352928, "data_time": 0.0014091968536376954, "loss": 0.4015228807926178, "time": 0.6343581438064575, "epoch": 5, "memory": 44137, "step": 19028} +{"lr": 0.00015975616287708285, "data_time": 0.001501321792602539, "loss": 0.384323787689209, "time": 0.6346012353897095, "epoch": 5, "memory": 44137, "step": 19128} +{"lr": 0.00015859071373493025, "data_time": 0.001623082160949707, "loss": 0.4046397626399994, "time": 0.6335388422012329, "epoch": 5, "memory": 44137, "step": 19228} +{"lr": 0.00015742503243065513, "data_time": 0.0014320611953735352, "loss": 0.42110501527786254, "time": 0.6338286399841309, "epoch": 5, "memory": 44137, "step": 19328} +{"lr": 0.0001562591943328516, "data_time": 0.001457810401916504, "loss": 0.4157467633485794, "time": 0.6340137958526612, "epoch": 5, "memory": 44137, "step": 19428} +{"lr": 0.0001550932748202511, "data_time": 0.0013442039489746094, "loss": 0.3957761228084564, "time": 0.6333126306533814, "epoch": 5, "memory": 44137, "step": 19528} +{"accuracy/top1": 74.89649200439453, "data_time": 0.035657381189280545, "time": 1.2823072589676956, "step": 5} +{"lr": 0.00015384573621094385, "data_time": 0.001628279685974121, "loss": 0.41077532172203063, "time": 0.6338565111160278, "epoch": 6, "memory": 44137, "step": 19635} +{"lr": 0.00015267988769893654, "data_time": 0.0014618158340454102, "loss": 0.396423476934433, "time": 0.6333179235458374, "epoch": 6, "memory": 44137, "step": 19735} +{"lr": 0.0001515141891967143, "data_time": 0.0014534711837768555, "loss": 0.3975360870361328, "time": 0.6329234600067138, "epoch": 6, "memory": 44137, "step": 19835} +{"lr": 0.00015034871607398357, "data_time": 0.0013907670974731446, "loss": 0.3965549886226654, "time": 0.6331725597381592, "epoch": 6, "memory": 44137, "step": 19935} +{"lr": 0.00014918354368587822, "data_time": 0.0015391111373901367, "loss": 0.4015242636203766, "time": 0.6330177307128906, "epoch": 6, "memory": 44137, "step": 20035} +{"lr": 0.00014801874736808767, "data_time": 0.001451730728149414, "loss": 0.42079220414161683, "time": 0.6342101097106934, "epoch": 6, "memory": 44137, "step": 20135} +{"lr": 0.00014685440243198569, "data_time": 0.001372528076171875, "loss": 0.3958262801170349, "time": 0.6332395315170288, "epoch": 6, "memory": 44137, "step": 20235} +{"lr": 0.00014569058415976212, "data_time": 0.0016178846359252929, "loss": 0.41581864953041076, "time": 0.6336359262466431, "epoch": 6, "memory": 44137, "step": 20335} +{"lr": 0.00014452736779955436, "data_time": 0.0015796661376953126, "loss": 0.4055685460567474, "time": 0.633411455154419, "epoch": 6, "memory": 44137, "step": 20435} +{"lr": 0.0001433648285605824, "data_time": 0.001438474655151367, "loss": 0.42735623717308047, "time": 0.6328923225402832, "epoch": 6, "memory": 44137, "step": 20535} +{"lr": 0.00014220304160828594, "data_time": 0.001488947868347168, "loss": 0.4199580252170563, "time": 0.6342294931411743, "epoch": 6, "memory": 44137, "step": 20635} +{"lr": 0.00014104208205946515, "data_time": 0.001546335220336914, "loss": 0.40364215672016146, "time": 0.6343457937240601, "epoch": 6, "memory": 44137, "step": 20735} +{"lr": 0.00013988202497742296, "data_time": 0.0015279054641723633, "loss": 0.3883852481842041, "time": 0.6339449644088745, "epoch": 6, "memory": 44137, "step": 20835} +{"lr": 0.00013872294536711234, "data_time": 0.0015924692153930664, "loss": 0.41546320617198945, "time": 0.6349325180053711, "epoch": 6, "memory": 44137, "step": 20935} +{"lr": 0.0001375649181702862, "data_time": 0.0015136003494262695, "loss": 0.40236571431159973, "time": 0.6335031747817993, "epoch": 6, "memory": 44137, "step": 21035} +{"lr": 0.0001364080182606529, "data_time": 0.0015846490859985352, "loss": 0.40679511427879333, "time": 0.6332422494888306, "epoch": 6, "memory": 44137, "step": 21135} +{"lr": 0.00013525232043903418, "data_time": 0.0015084981918334962, "loss": 0.3796483725309372, "time": 0.6350220680236817, "epoch": 6, "memory": 44137, "step": 21235} +{"lr": 0.0001340978994285297, "data_time": 0.0013573169708251953, "loss": 0.4032929718494415, "time": 0.6358685970306397, "epoch": 6, "memory": 44137, "step": 21335} +{"lr": 0.00013294482986968493, "data_time": 0.0015769481658935546, "loss": 0.3757345795631409, "time": 0.6332385063171386, "epoch": 6, "memory": 44137, "step": 21435} +{"lr": 0.00013179318631566578, "data_time": 0.0014328956604003906, "loss": 0.41288171112537386, "time": 0.6363572835922241, "epoch": 6, "memory": 44137, "step": 21535} +{"lr": 0.0001306430432274384, "data_time": 0.0015916109085083007, "loss": 0.41170698404312134, "time": 0.6335021495819092, "epoch": 6, "memory": 44137, "step": 21635} +{"lr": 0.00012949447496895343, "data_time": 0.0015311479568481446, "loss": 0.4057985067367554, "time": 0.6339433670043946, "epoch": 6, "memory": 44137, "step": 21735} +{"lr": 0.00012834755580233909, "data_time": 0.001506471633911133, "loss": 0.3743808776140213, "time": 0.6339581251144409, "epoch": 6, "memory": 44137, "step": 21835} +{"lr": 0.00012720235988310005, "data_time": 0.0014614582061767579, "loss": 0.4226959466934204, "time": 0.6358937740325927, "epoch": 6, "memory": 44137, "step": 21935} +{"lr": 0.0001260589612553219, "data_time": 0.0015298128128051758, "loss": 0.3868665754795074, "time": 0.6338008642196655, "epoch": 6, "memory": 44137, "step": 22035} +{"lr": 0.00012491743384688422, "data_time": 0.0013885259628295898, "loss": 0.39861064553260805, "time": 0.6350136041641236, "epoch": 6, "memory": 44137, "step": 22135} +{"lr": 0.0001237778514646806, "data_time": 0.0014439821243286133, "loss": 0.3788717776536942, "time": 0.6345320224761963, "epoch": 6, "memory": 44137, "step": 22235} +{"lr": 0.00012264028778984632, "data_time": 0.0014932394027709962, "loss": 0.4107115536928177, "time": 0.6335056066513062, "epoch": 6, "memory": 44137, "step": 22335} +{"lr": 0.00012150481637299435, "data_time": 0.001719498634338379, "loss": 0.3938616216182709, "time": 0.6344237565994263, "epoch": 6, "memory": 44137, "step": 22435} +{"lr": 0.00012037151062946036, "data_time": 0.001532578468322754, "loss": 0.4374720364809036, "time": 0.6367403745651246, "epoch": 6, "memory": 44137, "step": 22535} +{"lr": 0.00011924044383455631, "data_time": 0.0015212535858154298, "loss": 0.41519361138343813, "time": 0.634022879600525, "epoch": 6, "memory": 44137, "step": 22635} +{"lr": 0.0001181116891188313, "data_time": 0.0014603614807128906, "loss": 0.3899745434522629, "time": 0.6358419179916381, "epoch": 6, "memory": 44137, "step": 22735} +{"lr": 0.0001169853194633444, "data_time": 0.0014418601989746095, "loss": 0.412470543384552, "time": 0.6346251010894776, "epoch": 6, "memory": 44137, "step": 22835} +{"lr": 0.00011586140769494541, "data_time": 0.0015546321868896485, "loss": 0.41434812545776367, "time": 0.6342344522476197, "epoch": 6, "memory": 44137, "step": 22935} +{"lr": 0.00011474002648156654, "data_time": 0.001478719711303711, "loss": 0.4068192094564438, "time": 0.6443174839019775, "epoch": 6, "memory": 44137, "step": 23035} +{"lr": 0.00011362124832752344, "data_time": 0.0015213727951049806, "loss": 0.40099134743213655, "time": 0.6344565629959107, "epoch": 6, "memory": 44137, "step": 23135} +{"lr": 0.00011250514556882821, "data_time": 0.0013526439666748046, "loss": 0.4028135299682617, "time": 0.6347667217254639, "epoch": 6, "memory": 44137, "step": 23235} +{"lr": 0.00011139179036851135, "data_time": 0.0015400171279907227, "loss": 0.413922518491745, "time": 0.6342664957046509, "epoch": 6, "memory": 44137, "step": 23335} +{"lr": 0.00011028125471195667, "data_time": 0.0012638568878173828, "loss": 0.38831880390644075, "time": 0.6342539548873901, "epoch": 6, "memory": 44137, "step": 23435} +{"accuracy/top1": 74.70512390136719, "data_time": 0.03615083365604795, "time": 1.2833006998588299, "step": 6} +{"lr": 0.00010909618543324382, "data_time": 0.0014945507049560548, "loss": 0.4046571433544159, "time": 0.634126091003418, "epoch": 7, "memory": 44137, "step": 23542} +{"lr": 0.00010799171417102892, "data_time": 0.0014247417449951172, "loss": 0.44820262789726256, "time": 0.6341590881347656, "epoch": 7, "memory": 44137, "step": 23642} +{"lr": 0.0001068902822887887, "data_time": 0.001377701759338379, "loss": 0.4107129335403442, "time": 0.6343985080718995, "epoch": 7, "memory": 44137, "step": 23742} +{"lr": 0.00010579196100098985, "data_time": 0.0015125274658203125, "loss": 0.3960062026977539, "time": 0.6347365856170655, "epoch": 7, "memory": 44137, "step": 23842} +{"lr": 0.00010469682132097967, "data_time": 0.0015568494796752929, "loss": 0.3926637411117554, "time": 0.6343899250030518, "epoch": 7, "memory": 44137, "step": 23942} +{"lr": 0.00010360493405639455, "data_time": 0.001444244384765625, "loss": 0.3906110465526581, "time": 0.6349032402038575, "epoch": 7, "memory": 44137, "step": 24042} +{"lr": 0.00010251636980458193, "data_time": 0.0013445615768432617, "loss": 0.39741308987140656, "time": 0.6343841552734375, "epoch": 7, "memory": 44137, "step": 24142} +{"lr": 0.00010143119894803575, "data_time": 0.0014751672744750977, "loss": 0.4008590131998062, "time": 0.6349122524261475, "epoch": 7, "memory": 44137, "step": 24242} +{"lr": 0.00010034949164984562, "data_time": 0.001449894905090332, "loss": 0.3945930600166321, "time": 0.6343448162078857, "epoch": 7, "memory": 44137, "step": 24342} +{"lr": 9.927131784916038e-05, "data_time": 0.0014475107192993164, "loss": 0.3930166274309158, "time": 0.6361006736755371, "epoch": 7, "memory": 44137, "step": 24442} +{"lr": 9.819674725666596e-05, "data_time": 0.0015426874160766602, "loss": 0.3950634658336639, "time": 0.6358383655548095, "epoch": 7, "memory": 44137, "step": 24542} +{"lr": 9.712584935007872e-05, "data_time": 0.0014582157135009765, "loss": 0.41049647629261016, "time": 0.6364786863327027, "epoch": 7, "memory": 44137, "step": 24642} +{"lr": 9.605869336965269e-05, "data_time": 0.0014018535614013672, "loss": 0.4261118292808533, "time": 0.6357121944427491, "epoch": 7, "memory": 44137, "step": 24742} +{"lr": 9.499534831370295e-05, "data_time": 0.001417851448059082, "loss": 0.429586386680603, "time": 0.6352718114852905, "epoch": 7, "memory": 44137, "step": 24842} +{"lr": 9.39358829341445e-05, "data_time": 0.0013863086700439454, "loss": 0.3760415852069855, "time": 0.6354018449783325, "epoch": 7, "memory": 44137, "step": 24942} +{"lr": 9.288036573204672e-05, "data_time": 0.0016774892807006835, "loss": 0.40768120884895326, "time": 0.635544729232788, "epoch": 7, "memory": 44137, "step": 25042} +{"lr": 9.182886495320504e-05, "data_time": 0.0015383481979370118, "loss": 0.39910159409046175, "time": 0.6375270843505859, "epoch": 7, "memory": 44137, "step": 25142} +{"lr": 9.078144858372755e-05, "data_time": 0.0015362977981567382, "loss": 0.38406732082366946, "time": 0.6361876964569092, "epoch": 7, "memory": 44137, "step": 25242} +{"lr": 8.973818434563992e-05, "data_time": 0.001563286781311035, "loss": 0.3982101917266846, "time": 0.6353860378265381, "epoch": 7, "memory": 44137, "step": 25342} +{"lr": 8.86991396925066e-05, "data_time": 0.0014917612075805663, "loss": 0.40099319219589236, "time": 0.6339201927185059, "epoch": 7, "memory": 44137, "step": 25442} +{"lr": 8.766438180506934e-05, "data_time": 0.0014185190200805664, "loss": 0.4164343446493149, "time": 0.6347141981124877, "epoch": 7, "memory": 44137, "step": 25542} +{"lr": 8.66339775869035e-05, "data_time": 0.0015314102172851562, "loss": 0.40273507237434386, "time": 0.6362908124923706, "epoch": 7, "memory": 44137, "step": 25642} +{"lr": 8.560799366009288e-05, "data_time": 0.0015082836151123046, "loss": 0.43540977537631986, "time": 0.6343950033187866, "epoch": 7, "memory": 44137, "step": 25742} +{"lr": 8.458649636092149e-05, "data_time": 0.0014847517013549805, "loss": 0.3846946209669113, "time": 0.6344033241271972, "epoch": 7, "memory": 44137, "step": 25842} +{"lr": 8.356955173558475e-05, "data_time": 0.0014773368835449218, "loss": 0.37987988591194155, "time": 0.6341693878173829, "epoch": 7, "memory": 44137, "step": 25942} +{"lr": 8.255722553591929e-05, "data_time": 0.0014400720596313477, "loss": 0.3940939038991928, "time": 0.6344509601593018, "epoch": 7, "memory": 44137, "step": 26042} +{"lr": 8.154958321515152e-05, "data_time": 0.0015302181243896484, "loss": 0.4308096170425415, "time": 0.6344853639602661, "epoch": 7, "memory": 44137, "step": 26142} +{"lr": 8.054668992366594e-05, "data_time": 0.0015145540237426758, "loss": 0.39854896068573, "time": 0.6340305089950562, "epoch": 7, "memory": 44137, "step": 26242} +{"lr": 7.954861050479249e-05, "data_time": 0.0014557123184204101, "loss": 0.3925297200679779, "time": 0.634020733833313, "epoch": 7, "memory": 44137, "step": 26342} +{"lr": 7.855540949061403e-05, "data_time": 0.0014703035354614257, "loss": 0.39601756930351256, "time": 0.6359855890274048, "epoch": 7, "memory": 44137, "step": 26442} +{"lr": 7.756715109779415e-05, "data_time": 0.0013776302337646484, "loss": 0.43252343833446505, "time": 0.6359425067901612, "epoch": 7, "memory": 44137, "step": 26542} +{"lr": 7.658389922342482e-05, "data_time": 0.001479363441467285, "loss": 0.4226222574710846, "time": 0.6353020429611206, "epoch": 7, "memory": 44137, "step": 26642} +{"lr": 7.560571744089541e-05, "data_time": 0.0014433145523071289, "loss": 0.4033345192670822, "time": 0.6344865798950196, "epoch": 7, "memory": 44137, "step": 26742} +{"lr": 7.463266899578208e-05, "data_time": 0.0014557600021362304, "loss": 0.39786435663700104, "time": 0.6339991092681885, "epoch": 7, "memory": 44137, "step": 26842} +{"lr": 7.366481680175855e-05, "data_time": 0.0014534473419189453, "loss": 0.4089506298303604, "time": 0.6357561826705933, "epoch": 7, "memory": 44137, "step": 26942} +{"lr": 7.270222343652855e-05, "data_time": 0.0015775680541992188, "loss": 0.39817785322666166, "time": 0.6345189571380615, "epoch": 7, "memory": 44137, "step": 27042} +{"lr": 7.174495113777928e-05, "data_time": 0.0016486644744873047, "loss": 0.39844030141830444, "time": 0.6357248067855835, "epoch": 7, "memory": 44137, "step": 27142} +{"lr": 7.079306179915806e-05, "data_time": 0.0015341758728027344, "loss": 0.3968957901000977, "time": 0.6341945409774781, "epoch": 7, "memory": 44137, "step": 27242} +{"lr": 6.984661696627024e-05, "data_time": 0.0012496471405029296, "loss": 0.40110942125320437, "time": 0.6357108592987061, "epoch": 7, "memory": 44137, "step": 27342} +{"accuracy/top1": 74.8408203125, "data_time": 0.0356859749761121, "time": 1.2791170087353936, "step": 7} +{"lr": 6.884001985601584e-05, "data_time": 0.0016091585159301758, "loss": 0.3969531297683716, "time": 0.6348937511444092, "epoch": 8, "memory": 44137, "step": 27449} +{"lr": 6.790503918604319e-05, "data_time": 0.0014080047607421876, "loss": 0.40966375172138214, "time": 0.6334564924240113, "epoch": 8, "memory": 44137, "step": 27549} +{"lr": 6.697568975054603e-05, "data_time": 0.001700115203857422, "loss": 0.40814265608787537, "time": 0.6338228940963745, "epoch": 8, "memory": 44137, "step": 27649} +{"lr": 6.605203163778337e-05, "data_time": 0.001463627815246582, "loss": 0.40422230064868925, "time": 0.6338220834732056, "epoch": 8, "memory": 44137, "step": 27749} +{"lr": 6.513412456803489e-05, "data_time": 0.001416158676147461, "loss": 0.4123924046754837, "time": 0.6332894563674927, "epoch": 8, "memory": 44137, "step": 27849} +{"lr": 6.422202788973933e-05, "data_time": 0.0015094280242919922, "loss": 0.40509442389011385, "time": 0.6346164464950561, "epoch": 8, "memory": 44137, "step": 27949} +{"lr": 6.331580057565697e-05, "data_time": 0.0014674901962280274, "loss": 0.3988806664943695, "time": 0.6339321374893189, "epoch": 8, "memory": 44137, "step": 28049} +{"lr": 6.241550121905724e-05, "data_time": 0.0015756607055664063, "loss": 0.39616633057594297, "time": 0.6330334424972535, "epoch": 8, "memory": 44137, "step": 28149} +{"lr": 6.15211880299298e-05, "data_time": 0.0015095710754394532, "loss": 0.40544065833091736, "time": 0.6330933094024658, "epoch": 8, "memory": 44137, "step": 28249} +{"lr": 6.0632918831221007e-05, "data_time": 0.0014772415161132812, "loss": 0.4100262880325317, "time": 0.6333740234375, "epoch": 8, "memory": 44137, "step": 28349} +{"lr": 5.9750751055095654e-05, "data_time": 0.0015888452529907227, "loss": 0.3976229250431061, "time": 0.6334634780883789, "epoch": 8, "memory": 44137, "step": 28449} +{"lr": 5.887474173922336e-05, "data_time": 0.0014522314071655274, "loss": 0.40636836886405947, "time": 0.6329374074935913, "epoch": 8, "memory": 44137, "step": 28549} +{"lr": 5.80049475230907e-05, "data_time": 0.0016263484954833984, "loss": 0.4128390848636627, "time": 0.6331194162368774, "epoch": 8, "memory": 44137, "step": 28649} +{"lr": 5.7141424644339115e-05, "data_time": 0.0014470577239990234, "loss": 0.4046159446239471, "time": 0.6349116325378418, "epoch": 8, "memory": 44137, "step": 28749} +{"lr": 5.6284228935128753e-05, "data_time": 0.0015562772750854492, "loss": 0.3848160207271576, "time": 0.6335993528366088, "epoch": 8, "memory": 44137, "step": 28849} +{"lr": 5.543341581852875e-05, "data_time": 0.0016240596771240235, "loss": 0.3774960994720459, "time": 0.635904049873352, "epoch": 8, "memory": 44137, "step": 28949} +{"lr": 5.458904030493362e-05, "data_time": 0.001580357551574707, "loss": 0.4161092400550842, "time": 0.6341368436813355, "epoch": 8, "memory": 44137, "step": 29049} +{"lr": 5.3751156988506505e-05, "data_time": 0.0015173912048339843, "loss": 0.3929214417934418, "time": 0.6332409620285034, "epoch": 8, "memory": 44137, "step": 29149} +{"lr": 5.2919820043649535e-05, "data_time": 0.0014882087707519531, "loss": 0.4085879951715469, "time": 0.6345562934875488, "epoch": 8, "memory": 44137, "step": 29249} +{"lr": 5.209508322150057e-05, "data_time": 0.0015078306198120118, "loss": 0.4201653331518173, "time": 0.6334356307983399, "epoch": 8, "memory": 44137, "step": 29349} +{"lr": 5.127699984645856e-05, "data_time": 0.0016011476516723632, "loss": 0.40792628228664396, "time": 0.6332447052001953, "epoch": 8, "memory": 44137, "step": 29449} +{"lr": 5.046562281273526e-05, "data_time": 0.0014960527420043944, "loss": 0.3835161954164505, "time": 0.6331025123596191, "epoch": 8, "memory": 44137, "step": 29549} +{"lr": 4.966100458093549e-05, "data_time": 0.0014417409896850587, "loss": 0.39708831906318665, "time": 0.6330764293670654, "epoch": 8, "memory": 44137, "step": 29649} +{"lr": 4.886319717466545e-05, "data_time": 0.00142214298248291, "loss": 0.4182854235172272, "time": 0.6334197998046875, "epoch": 8, "memory": 44137, "step": 29749} +{"lr": 4.8072252177168467e-05, "data_time": 0.0013968229293823242, "loss": 0.4206300139427185, "time": 0.633488655090332, "epoch": 8, "memory": 44137, "step": 29849} +{"lr": 4.728822072799032e-05, "data_time": 0.0015152692794799805, "loss": 0.4408943563699722, "time": 0.6330051422119141, "epoch": 8, "memory": 44137, "step": 29949} +{"lr": 4.6511153519672786e-05, "data_time": 0.0015388727188110352, "loss": 0.4135405957698822, "time": 0.6336704015731811, "epoch": 8, "memory": 44137, "step": 30049} +{"lr": 4.57411007944756e-05, "data_time": 0.0014841794967651368, "loss": 0.39446223676204684, "time": 0.6341177225112915, "epoch": 8, "memory": 44137, "step": 30149} +{"lr": 4.49781123411285e-05, "data_time": 0.0015529632568359376, "loss": 0.4151736408472061, "time": 0.6330987930297851, "epoch": 8, "memory": 44137, "step": 30249} +{"lr": 4.422223749161159e-05, "data_time": 0.0015906572341918945, "loss": 0.39999428689479827, "time": 0.6345949649810791, "epoch": 8, "memory": 44137, "step": 30349} +{"lr": 4.3473525117965965e-05, "data_time": 0.0015187501907348634, "loss": 0.40094477832317355, "time": 0.6338727712631226, "epoch": 8, "memory": 44137, "step": 30449} +{"lr": 4.2732023629134074e-05, "data_time": 0.0014914751052856445, "loss": 0.43085960745811464, "time": 0.6340967893600464, "epoch": 8, "memory": 44137, "step": 30549} +{"lr": 4.199778096782924e-05, "data_time": 0.0014739990234375, "loss": 0.4026518613100052, "time": 0.6333942413330078, "epoch": 8, "memory": 44137, "step": 30649} +{"lr": 4.127084460743637e-05, "data_time": 0.0014615058898925781, "loss": 0.40305177569389344, "time": 0.6336512088775634, "epoch": 8, "memory": 44137, "step": 30749} +{"lr": 4.0551261548942e-05, "data_time": 0.001582169532775879, "loss": 0.44525461196899413, "time": 0.633476972579956, "epoch": 8, "memory": 44137, "step": 30849} +{"lr": 3.983907831789604e-05, "data_time": 0.001528453826904297, "loss": 0.3785897374153137, "time": 0.6356264591217041, "epoch": 8, "memory": 44137, "step": 30949} +{"lr": 3.913434096140289e-05, "data_time": 0.0015667915344238282, "loss": 0.3967933654785156, "time": 0.6338096618652344, "epoch": 8, "memory": 44137, "step": 31049} +{"lr": 3.8437095045144625e-05, "data_time": 0.0014075040817260742, "loss": 0.42655904293060304, "time": 0.6349545001983643, "epoch": 8, "memory": 44137, "step": 31149} +{"lr": 3.77473856504348e-05, "data_time": 0.001307225227355957, "loss": 0.3991405516862869, "time": 0.6340682744979859, "epoch": 8, "memory": 44137, "step": 31249} +{"accuracy/top1": 74.37806701660156, "data_time": 0.03741505228239914, "time": 1.2846401066615665, "step": 8} +{"lr": 3.701779344716437e-05, "data_time": 0.0016834020614624023, "loss": 0.40488806664943694, "time": 0.633327865600586, "epoch": 9, "memory": 44137, "step": 31356} +{"lr": 3.63438257918037e-05, "data_time": 0.0017286062240600586, "loss": 0.3990287363529205, "time": 0.634128737449646, "epoch": 9, "memory": 44137, "step": 31456} +{"lr": 3.567753000095364e-05, "data_time": 0.001622462272644043, "loss": 0.3980677485466003, "time": 0.642203402519226, "epoch": 9, "memory": 44137, "step": 31556} +{"lr": 3.501894915480852e-05, "data_time": 0.0014738798141479491, "loss": 0.3903617441654205, "time": 0.6338530302047729, "epoch": 9, "memory": 44137, "step": 31656} +{"lr": 3.4368125834743e-05, "data_time": 0.001567530632019043, "loss": 0.422028112411499, "time": 0.6336917877197266, "epoch": 9, "memory": 44137, "step": 31756} +{"lr": 3.372510212055925e-05, "data_time": 0.0014284133911132812, "loss": 0.3847354382276535, "time": 0.6334270238876343, "epoch": 9, "memory": 44137, "step": 31856} +{"lr": 3.308991958776599e-05, "data_time": 0.001447582244873047, "loss": 0.3819938778877258, "time": 0.634311580657959, "epoch": 9, "memory": 44137, "step": 31956} +{"lr": 3.246261930489061e-05, "data_time": 0.0013586997985839844, "loss": 0.39875280261039736, "time": 0.6348443746566772, "epoch": 9, "memory": 44137, "step": 32056} +{"lr": 3.18432418308236e-05, "data_time": 0.0014246463775634765, "loss": 0.4338852077722549, "time": 0.633664870262146, "epoch": 9, "memory": 44137, "step": 32156} +{"lr": 3.123182721219624e-05, "data_time": 0.0016513109207153321, "loss": 0.4046516209840775, "time": 0.6342431783676148, "epoch": 9, "memory": 44137, "step": 32256} +{"lr": 3.0628414980791353e-05, "data_time": 0.0015561342239379882, "loss": 0.39943632781505584, "time": 0.6335716962814331, "epoch": 9, "memory": 44137, "step": 32356} +{"lr": 3.0033044150987488e-05, "data_time": 0.0014463424682617187, "loss": 0.3937799632549286, "time": 0.6331947565078735, "epoch": 9, "memory": 44137, "step": 32456} +{"lr": 2.944575321723602e-05, "data_time": 0.0015320062637329101, "loss": 0.40703162252902986, "time": 0.6329815864562989, "epoch": 9, "memory": 44137, "step": 32556} +{"lr": 2.8866580151572607e-05, "data_time": 0.0014628887176513672, "loss": 0.40660122931003573, "time": 0.6340882301330566, "epoch": 9, "memory": 44137, "step": 32656} +{"lr": 2.829556240116184e-05, "data_time": 0.001535940170288086, "loss": 0.41109093725681306, "time": 0.6341081857681274, "epoch": 9, "memory": 44137, "step": 32756} +{"lr": 2.7732736885876114e-05, "data_time": 0.001671004295349121, "loss": 0.40753025114536284, "time": 0.6416688680648803, "epoch": 9, "memory": 44137, "step": 32856} +{"lr": 2.717813999590862e-05, "data_time": 0.0015972137451171875, "loss": 0.4007302701473236, "time": 0.6333588361740112, "epoch": 9, "memory": 44137, "step": 32956} +{"lr": 2.663180758942035e-05, "data_time": 0.0015102386474609374, "loss": 0.3754425525665283, "time": 0.6337812662124633, "epoch": 9, "memory": 44137, "step": 33056} +{"lr": 2.6093774990221658e-05, "data_time": 0.001367640495300293, "loss": 0.36535353064537046, "time": 0.6333444118499756, "epoch": 9, "memory": 44137, "step": 33156} +{"lr": 2.5564076985488515e-05, "data_time": 0.0015793561935424805, "loss": 0.407567223906517, "time": 0.6349753379821778, "epoch": 9, "memory": 44137, "step": 33256} +{"lr": 2.5042747823512982e-05, "data_time": 0.0013894081115722657, "loss": 0.4009361952543259, "time": 0.634540057182312, "epoch": 9, "memory": 44137, "step": 33356} +{"lr": 2.4529821211489152e-05, "data_time": 0.0014384031295776368, "loss": 0.4067427635192871, "time": 0.633469033241272, "epoch": 9, "memory": 44137, "step": 33456} +{"lr": 2.402533031333377e-05, "data_time": 0.00162813663482666, "loss": 0.4030266135931015, "time": 0.6335771560668946, "epoch": 9, "memory": 44137, "step": 33556} +{"lr": 2.352930774754175e-05, "data_time": 0.0015053749084472656, "loss": 0.38815886378288267, "time": 0.6335853099822998, "epoch": 9, "memory": 44137, "step": 33656} +{"lr": 2.3041785585077355e-05, "data_time": 0.001583099365234375, "loss": 0.3933963686227798, "time": 0.633979344367981, "epoch": 9, "memory": 44137, "step": 33756} +{"lr": 2.256279534730055e-05, "data_time": 0.001464986801147461, "loss": 0.4071730405092239, "time": 0.6359490156173706, "epoch": 9, "memory": 44137, "step": 33856} +{"lr": 2.2092368003928886e-05, "data_time": 0.0015979528427124024, "loss": 0.4104828476905823, "time": 0.6341641664505004, "epoch": 9, "memory": 44137, "step": 33956} +{"lr": 2.1630533971035236e-05, "data_time": 0.0014789342880249024, "loss": 0.42280719578266146, "time": 0.6341844797134399, "epoch": 9, "memory": 44137, "step": 34056} +{"lr": 2.117732310908114e-05, "data_time": 0.001540207862854004, "loss": 0.4198928356170654, "time": 0.6349170684814454, "epoch": 9, "memory": 44137, "step": 34156} +{"lr": 2.0732764720986212e-05, "data_time": 0.0014826297760009766, "loss": 0.37595857977867125, "time": 0.6355086326599121, "epoch": 9, "memory": 44137, "step": 34256} +{"lr": 2.0296887550233324e-05, "data_time": 0.0014709711074829101, "loss": 0.39905501902103424, "time": 0.6342411518096924, "epoch": 9, "memory": 44137, "step": 34356} +{"lr": 1.9869719779010417e-05, "data_time": 0.0014364480972290038, "loss": 0.4381367713212967, "time": 0.6351343154907226, "epoch": 9, "memory": 44137, "step": 34456} +{"lr": 1.9451289026388164e-05, "data_time": 0.0014481782913208009, "loss": 0.4117830693721771, "time": 0.6336531639099121, "epoch": 9, "memory": 44137, "step": 34556} +{"lr": 1.9041622346534353e-05, "data_time": 0.0016322374343872071, "loss": 0.4093270987272263, "time": 0.6347902536392211, "epoch": 9, "memory": 44137, "step": 34656} +{"lr": 1.8640746226964523e-05, "data_time": 0.0016631364822387695, "loss": 0.39976613223552704, "time": 0.6349977016448974, "epoch": 9, "memory": 44137, "step": 34756} +{"lr": 1.82486865868295e-05, "data_time": 0.001429438591003418, "loss": 0.39968317449092866, "time": 0.633928632736206, "epoch": 9, "memory": 44137, "step": 34856} +{"lr": 1.7865468775239498e-05, "data_time": 0.0014545679092407226, "loss": 0.4115006387233734, "time": 0.6340491056442261, "epoch": 9, "memory": 44137, "step": 34956} +{"lr": 1.7491117569625184e-05, "data_time": 0.0015875816345214844, "loss": 0.4287821531295776, "time": 0.6341647624969482, "epoch": 9, "memory": 44137, "step": 35056} +{"lr": 1.7125657174135686e-05, "data_time": 0.0013334035873413086, "loss": 0.43009800612926485, "time": 0.6355794429779053, "epoch": 9, "memory": 44137, "step": 35156} +{"accuracy/top1": 74.28760528564453, "data_time": 0.037636423933094944, "time": 1.2834950767714401, "step": 9} +{"lr": 1.674448744607319e-05, "data_time": 0.0018569469451904298, "loss": 0.3890314280986786, "time": 0.6338708400726318, "epoch": 10, "memory": 44137, "step": 35263} +{"lr": 1.639750545524004e-05, "data_time": 0.0014520883560180664, "loss": 0.4330062627792358, "time": 0.6343757152557373, "epoch": 10, "memory": 44137, "step": 35363} +{"lr": 1.6059484983401464e-05, "data_time": 0.0015063762664794921, "loss": 0.4040870815515518, "time": 0.6342190265655517, "epoch": 10, "memory": 44137, "step": 35463} +{"lr": 1.573044788569705e-05, "data_time": 0.001512908935546875, "loss": 0.4126666158437729, "time": 0.6340214729309082, "epoch": 10, "memory": 44137, "step": 35563} +{"lr": 1.5410415436435178e-05, "data_time": 0.0014666080474853515, "loss": 0.4082415699958801, "time": 0.6351520776748657, "epoch": 10, "memory": 44137, "step": 35663} +{"lr": 1.5099408327717273e-05, "data_time": 0.0015602350234985352, "loss": 0.4406146079301834, "time": 0.6344327211380005, "epoch": 10, "memory": 44137, "step": 35763} +{"lr": 1.479744666809991e-05, "data_time": 0.001470041275024414, "loss": 0.40127586126327514, "time": 0.6339004278182984, "epoch": 10, "memory": 44137, "step": 35863} +{"lr": 1.4504549981294865e-05, "data_time": 0.0015267610549926757, "loss": 0.4200449585914612, "time": 0.6339508295059204, "epoch": 10, "memory": 44137, "step": 35963} +{"lr": 1.4220737204906605e-05, "data_time": 0.001703190803527832, "loss": 0.41975703835487366, "time": 0.6334076881408691, "epoch": 10, "memory": 44137, "step": 36063} +{"lr": 1.3946026689207934e-05, "data_time": 0.0014288663864135743, "loss": 0.3851330608129501, "time": 0.6342244148254395, "epoch": 10, "memory": 44137, "step": 36163} +{"lr": 1.3680436195953514e-05, "data_time": 0.0015428781509399414, "loss": 0.4089700669050217, "time": 0.6338906526565552, "epoch": 10, "memory": 44137, "step": 36263} +{"lr": 1.3423982897231485e-05, "data_time": 0.0014763355255126953, "loss": 0.4391102522611618, "time": 0.6338380575180054, "epoch": 10, "memory": 44137, "step": 36363} +{"lr": 1.3176683374353149e-05, "data_time": 0.0017591476440429687, "loss": 0.4089411228895187, "time": 0.6361191511154175, "epoch": 10, "memory": 44137, "step": 36463} +{"lr": 1.2938553616780832e-05, "data_time": 0.0014814138412475586, "loss": 0.4454068303108215, "time": 0.6342217445373535, "epoch": 10, "memory": 44137, "step": 36563} +{"lr": 1.2709609021094214e-05, "data_time": 0.0015864133834838866, "loss": 0.4099941462278366, "time": 0.6345898389816285, "epoch": 10, "memory": 44137, "step": 36663} +{"lr": 1.2489864389994715e-05, "data_time": 0.0015474081039428711, "loss": 0.39916797280311583, "time": 0.6339625358581543, "epoch": 10, "memory": 44137, "step": 36763} +{"lr": 1.2279333931348431e-05, "data_time": 0.001513504981994629, "loss": 0.4181778907775879, "time": 0.6436614990234375, "epoch": 10, "memory": 44137, "step": 36863} +{"lr": 1.2078031257267545e-05, "data_time": 0.0015755414962768555, "loss": 0.3995991885662079, "time": 0.6347186803817749, "epoch": 10, "memory": 44137, "step": 36963} +{"lr": 1.1885969383230204e-05, "data_time": 0.0017694473266601563, "loss": 0.40407503247261045, "time": 0.6784493446350097, "epoch": 10, "memory": 44137, "step": 37063} +{"lr": 1.1703160727238979e-05, "data_time": 0.0014449834823608398, "loss": 0.38404877483844757, "time": 0.6347826004028321, "epoch": 10, "memory": 44137, "step": 37163} +{"lr": 1.1529617109017937e-05, "data_time": 0.0014947175979614258, "loss": 0.38951668739318845, "time": 0.6342282295227051, "epoch": 10, "memory": 44137, "step": 37263} +{"lr": 1.136534974924845e-05, "data_time": 0.0014008045196533202, "loss": 0.3978728473186493, "time": 0.6371277332305908, "epoch": 10, "memory": 44137, "step": 37363} +{"lr": 1.1210369268843738e-05, "data_time": 0.0014733076095581055, "loss": 0.40388007164001466, "time": 0.637355637550354, "epoch": 10, "memory": 44137, "step": 37463} +{"lr": 1.1064685688262156e-05, "data_time": 0.0015849828720092773, "loss": 0.39653285443782804, "time": 0.6343758821487426, "epoch": 10, "memory": 44137, "step": 37563} +{"lr": 1.0928308426859188e-05, "data_time": 0.0018506050109863281, "loss": 0.4068183064460754, "time": 0.6338836669921875, "epoch": 10, "memory": 44137, "step": 37663} +{"lr": 1.0801246302278603e-05, "data_time": 0.0015090465545654296, "loss": 0.42445961833000184, "time": 0.6338107109069824, "epoch": 10, "memory": 44137, "step": 37763} +{"lr": 1.0683507529882236e-05, "data_time": 0.0015589237213134766, "loss": 0.4159298360347748, "time": 0.6341955423355102, "epoch": 10, "memory": 44137, "step": 37863} +{"lr": 1.057509972221878e-05, "data_time": 0.0014363765716552735, "loss": 0.4106313705444336, "time": 0.6336933612823487, "epoch": 10, "memory": 44137, "step": 37963} +{"lr": 1.047602988853174e-05, "data_time": 0.0015500783920288086, "loss": 0.42624981999397277, "time": 0.6341934680938721, "epoch": 10, "memory": 44137, "step": 38063} +{"lr": 1.0386304434306083e-05, "data_time": 0.0014511346817016602, "loss": 0.3967807799577713, "time": 0.6350108861923218, "epoch": 10, "memory": 44137, "step": 38163} +{"lr": 1.030592916085413e-05, "data_time": 0.0020376920700073244, "loss": 0.42003758549690245, "time": 0.633349871635437, "epoch": 10, "memory": 44137, "step": 38263} +{"lr": 1.0234909264940509e-05, "data_time": 0.0015140056610107421, "loss": 0.4236251085996628, "time": 0.633553409576416, "epoch": 10, "memory": 44137, "step": 38363} +{"lr": 1.0173249338446117e-05, "data_time": 0.0014904499053955077, "loss": 0.4105237126350403, "time": 0.6346558570861817, "epoch": 10, "memory": 44137, "step": 38463} +{"lr": 1.0120953368071232e-05, "data_time": 0.0014421701431274413, "loss": 0.41743309795856476, "time": 0.6333788871765137, "epoch": 10, "memory": 44137, "step": 38563} +{"lr": 1.0078024735077709e-05, "data_time": 0.0014368295669555664, "loss": 0.39933337569236754, "time": 0.6333008289337159, "epoch": 10, "memory": 44137, "step": 38663} +{"lr": 1.004446621507047e-05, "data_time": 0.0014926433563232423, "loss": 0.38928772509098053, "time": 0.6334861278533935, "epoch": 10, "memory": 44137, "step": 38763} +{"lr": 1.0020279977817876e-05, "data_time": 0.0014059782028198243, "loss": 0.4124821960926056, "time": 0.6341943740844727, "epoch": 10, "memory": 44137, "step": 38863} +{"lr": 1.0005467587111622e-05, "data_time": 0.001574540138244629, "loss": 0.41150707602500913, "time": 0.6334386587142944, "epoch": 10, "memory": 44137, "step": 38963} +{"lr": 1.0000030000665491e-05, "data_time": 0.0013365745544433594, "loss": 0.40277799069881437, "time": 0.6339604377746582, "epoch": 10, "memory": 44137, "step": 39063} +{"accuracy/top1": 74.29108428955078, "data_time": 0.03611139593453243, "time": 1.2821046525034412, "step": 10} diff --git a/clip_large_pretrain_4x256_sdv2_lr3e-4/20230606_052113/vis_data/config.py b/clip_large_pretrain_4x256_sdv2_lr3e-4/20230606_052113/vis_data/config.py new file mode 100644 index 0000000000000000000000000000000000000000..2fe3d2db91303588ef319725b2e1e94dca9ecb05 --- /dev/null +++ b/clip_large_pretrain_4x256_sdv2_lr3e-4/20230606_052113/vis_data/config.py @@ -0,0 +1,236 @@ +optim_wrapper = dict( + optimizer=dict( + type='AdamW', lr=0.0003, weight_decay=0.3, _scope_='mmpretrain'), + paramwise_cfg=dict( + custom_keys=dict({ + '.cls_token': dict(decay_mult=0.0), + '.pos_embed': dict(decay_mult=0.0) + })), + type='AmpOptimWrapper', + dtype='bfloat16', + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=False, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=4096) +model = dict( + type='ImageClassifier', + backbone=dict( + frozen_stages=24, + type='VisionTransformer', + arch='l', + img_size=224, + patch_size=14, + drop_rate=0.1, + pre_norm=True, + final_norm=False, + init_cfg=dict( + type='Pretrained', + checkpoint='ckpt/openclip-ViT-L-14.pth', + prefix='backbone')), + neck=dict( + type='CLIPProjection', + in_channels=1024, + out_channels=768, + init_cfg=dict( + type='Pretrained', + checkpoint='ckpt/openclip-ViT-L-14.pth', + prefix='backbone')), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=768, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + init_cfg=None), + init_cfg=dict( + type='TruncNormal', layer=['Conv2d', 'Linear'], std=0.02, bias=0.0), + train_cfg=None) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=128, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stablediffusionV2-1-dpmsolver-25-1m.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/cc1m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=128, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV2-1-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=128, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV2-1-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/clip_large_pretrain_4x256_sdv2_lr3e-4' diff --git a/clip_large_pretrain_4x256_sdv2_lr3e-4/20230606_052113/vis_data/events.out.tfevents.1686000089.SH-IDC1-10-140-24-99.120688.0 b/clip_large_pretrain_4x256_sdv2_lr3e-4/20230606_052113/vis_data/events.out.tfevents.1686000089.SH-IDC1-10-140-24-99.120688.0 new file mode 100644 index 0000000000000000000000000000000000000000..73359875a86f24d194486cf60de132a879d85cf8 --- /dev/null +++ b/clip_large_pretrain_4x256_sdv2_lr3e-4/20230606_052113/vis_data/events.out.tfevents.1686000089.SH-IDC1-10-140-24-99.120688.0 @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:2fbfaf6a8abb26102878c2b8a0a8991bdf28b983b7d3678a3e2518bd9412ec1b +size 1734262 diff --git a/clip_large_pretrain_4x256_sdv2_lr3e-4/20230606_052113/vis_data/scalars.json b/clip_large_pretrain_4x256_sdv2_lr3e-4/20230606_052113/vis_data/scalars.json new file mode 100644 index 0000000000000000000000000000000000000000..24d296ee2b82acb398effeea94566fbb57949f7b --- /dev/null +++ b/clip_large_pretrain_4x256_sdv2_lr3e-4/20230606_052113/vis_data/scalars.json @@ -0,0 +1,400 @@ +{"lr": 0.0002999954057034405, "data_time": 0.0016700506210327148, "loss": 0.4902272582054138, "time": 0.6309503555297852, "epoch": 1, "memory": 44138, "step": 100} +{"lr": 0.0002999814370149173, "data_time": 0.0016016960144042969, "loss": 0.49141762852668763, "time": 0.632881760597229, "epoch": 1, "memory": 44138, "step": 200} +{"lr": 0.00029995809436883357, "data_time": 0.0014659404754638673, "loss": 0.4493827402591705, "time": 0.6323542594909668, "epoch": 1, "memory": 44138, "step": 300} +{"lr": 0.00029992537927443715, "data_time": 0.0017125844955444337, "loss": 0.44257314801216124, "time": 0.6326935768127442, "epoch": 1, "memory": 44138, "step": 400} +{"lr": 0.0002998832938469641, "data_time": 0.0015325546264648438, "loss": 0.45193727016448976, "time": 0.6333640098571778, "epoch": 1, "memory": 44138, "step": 500} +{"lr": 0.0002998318408075004, "data_time": 0.0015497446060180665, "loss": 0.41731774210929873, "time": 0.6340202331542969, "epoch": 1, "memory": 44138, "step": 600} +{"lr": 0.0002997710234828077, "data_time": 0.0015933990478515625, "loss": 0.4037692368030548, "time": 0.632411789894104, "epoch": 1, "memory": 44138, "step": 700} +{"lr": 0.00029970084580510675, "data_time": 0.0015311479568481446, "loss": 0.4298222452402115, "time": 0.6337251424789428, "epoch": 1, "memory": 44138, "step": 800} +{"lr": 0.0002996213123118231, "data_time": 0.0015386819839477539, "loss": 0.41936435699462893, "time": 0.6332170009613037, "epoch": 1, "memory": 44138, "step": 900} +{"lr": 0.0002995324281452956, "data_time": 0.0015138149261474609, "loss": 0.4237715512514114, "time": 0.6327500343322754, "epoch": 1, "memory": 44138, "step": 1000} +{"lr": 0.00029943419905244207, "data_time": 0.0015166044235229493, "loss": 0.4082989066839218, "time": 0.6326113700866699, "epoch": 1, "memory": 44138, "step": 1100} +{"lr": 0.000299326631384388, "data_time": 0.0014206647872924804, "loss": 0.4096844166517258, "time": 0.6304686307907105, "epoch": 1, "memory": 44138, "step": 1200} +{"lr": 0.0002992097320960566, "data_time": 0.0014466524124145507, "loss": 0.37484508752822876, "time": 0.6327450037002563, "epoch": 1, "memory": 44138, "step": 1300} +{"lr": 0.00029908350874571896, "data_time": 0.0015225648880004884, "loss": 0.3997973918914795, "time": 0.6336121082305908, "epoch": 1, "memory": 44138, "step": 1400} +{"lr": 0.00029894796949450446, "data_time": 0.0015665054321289062, "loss": 0.39717928469181063, "time": 0.6327163457870484, "epoch": 1, "memory": 44138, "step": 1500} +{"lr": 0.00029880312310587493, "data_time": 0.0015705108642578125, "loss": 0.4122603118419647, "time": 0.6335359811782837, "epoch": 1, "memory": 44138, "step": 1600} +{"lr": 0.00029864897894505576, "data_time": 0.001616358757019043, "loss": 0.4264645129442215, "time": 0.6329888582229615, "epoch": 1, "memory": 44138, "step": 1700} +{"lr": 0.0002984855469784324, "data_time": 0.0016736268997192382, "loss": 0.39207322597503663, "time": 0.6333279371261596, "epoch": 1, "memory": 44138, "step": 1800} +{"lr": 0.0002983128377729035, "data_time": 0.0013751745223999023, "loss": 0.3964429199695587, "time": 0.6329690933227539, "epoch": 1, "memory": 44138, "step": 1900} +{"lr": 0.0002981308624952014, "data_time": 0.0013938665390014649, "loss": 0.37336658239364623, "time": 0.6348889589309692, "epoch": 1, "memory": 44138, "step": 2000} +{"lr": 0.0002979396329111671, "data_time": 0.0014202356338500976, "loss": 0.3938830614089966, "time": 0.6333211183547973, "epoch": 1, "memory": 44138, "step": 2100} +{"lr": 0.00029773916138499, "data_time": 0.001470470428466797, "loss": 0.3555929780006409, "time": 0.6339236736297608, "epoch": 1, "memory": 44138, "step": 2200} +{"lr": 0.00029752946087840954, "data_time": 0.0014882564544677734, "loss": 0.3780499905347824, "time": 0.6328037261962891, "epoch": 1, "memory": 44138, "step": 2300} +{"lr": 0.00029731054494987663, "data_time": 0.0014256000518798827, "loss": 0.39369869530200957, "time": 0.6329768419265747, "epoch": 1, "memory": 44138, "step": 2400} +{"lr": 0.0002970824277536758, "data_time": 0.001444244384765625, "loss": 0.3938509404659271, "time": 0.6332411289215087, "epoch": 1, "memory": 44138, "step": 2500} +{"lr": 0.0002968451240390124, "data_time": 0.0014374971389770508, "loss": 0.37504111528396605, "time": 0.6336678028106689, "epoch": 1, "memory": 44138, "step": 2600} +{"lr": 0.0002965986491490574, "data_time": 0.001483774185180664, "loss": 0.376563623547554, "time": 0.6326481103897095, "epoch": 1, "memory": 44138, "step": 2700} +{"lr": 0.00029634301901995517, "data_time": 0.001518845558166504, "loss": 0.39616626501083374, "time": 0.6332586526870727, "epoch": 1, "memory": 44138, "step": 2800} +{"lr": 0.000296078250179794, "data_time": 0.0015430450439453125, "loss": 0.37575416564941405, "time": 0.6331181526184082, "epoch": 1, "memory": 44138, "step": 2900} +{"lr": 0.0002958043597475368, "data_time": 0.0016086816787719727, "loss": 0.3715750277042389, "time": 0.6327194690704345, "epoch": 1, "memory": 44138, "step": 3000} +{"lr": 0.0002955213654319153, "data_time": 0.0015376806259155273, "loss": 0.3767815738916397, "time": 0.6324351787567138, "epoch": 1, "memory": 44138, "step": 3100} +{"lr": 0.00029522928553028316, "data_time": 0.0016405582427978516, "loss": 0.3875528693199158, "time": 0.6325571060180664, "epoch": 1, "memory": 44138, "step": 3200} +{"lr": 0.0002949281389274349, "data_time": 0.0014760017395019532, "loss": 0.38488066792488096, "time": 0.6327473163604737, "epoch": 1, "memory": 44138, "step": 3300} +{"lr": 0.0002946179450943832, "data_time": 0.0015089988708496093, "loss": 0.39545409083366395, "time": 0.6326494216918945, "epoch": 1, "memory": 44138, "step": 3400} +{"lr": 0.00029429872408710166, "data_time": 0.0014854669570922852, "loss": 0.3835848271846771, "time": 0.6328840970993042, "epoch": 1, "memory": 44138, "step": 3500} +{"lr": 0.0002939704965452269, "data_time": 0.0015824794769287109, "loss": 0.37650732398033143, "time": 0.6334297895431519, "epoch": 1, "memory": 44138, "step": 3600} +{"lr": 0.00029363328369072446, "data_time": 0.0016147375106811523, "loss": 0.36214399933815, "time": 0.6332512617111206, "epoch": 1, "memory": 44138, "step": 3700} +{"lr": 0.00029328710732651664, "data_time": 0.0015808343887329102, "loss": 0.35871745049953463, "time": 0.632096529006958, "epoch": 1, "memory": 44138, "step": 3800} +{"lr": 0.0002929319898350724, "data_time": 0.0012046575546264648, "loss": 0.3764660030603409, "time": 0.6321643352508545, "epoch": 1, "memory": 44138, "step": 3900} +{"accuracy/top1": 65.90585327148438, "data_time": 0.045427987450047544, "time": 1.3536268326274135, "step": 1} +{"lr": 0.0002925421383003051, "data_time": 0.001694941520690918, "loss": 0.3807366907596588, "time": 0.6337023973464966, "epoch": 2, "memory": 44138, "step": 4007} +{"lr": 0.0002921685862849842, "data_time": 0.0014974832534790038, "loss": 0.3624882698059082, "time": 0.6346543073654175, "epoch": 2, "memory": 44137, "step": 4107} +{"lr": 0.0002917861654618148, "data_time": 0.0014608383178710937, "loss": 0.3827891707420349, "time": 0.6333630323410034, "epoch": 2, "memory": 44137, "step": 4207} +{"lr": 0.0002913949005566972, "data_time": 0.0014677047729492188, "loss": 0.3831400215625763, "time": 0.6331991672515869, "epoch": 2, "memory": 44137, "step": 4307} +{"lr": 0.0002909948168673564, "data_time": 0.0014125823974609375, "loss": 0.3692391812801361, "time": 0.6360158443450927, "epoch": 2, "memory": 44137, "step": 4407} +{"lr": 0.00029058594026170915, "data_time": 0.0013590097427368165, "loss": 0.3524982511997223, "time": 0.634480357170105, "epoch": 2, "memory": 44137, "step": 4507} +{"lr": 0.0002901682971761867, "data_time": 0.0015326976776123048, "loss": 0.35894723534584044, "time": 0.6338254690170289, "epoch": 2, "memory": 44137, "step": 4607} +{"lr": 0.00028974191461402947, "data_time": 0.0013803958892822266, "loss": 0.36094541251659396, "time": 0.632565188407898, "epoch": 2, "memory": 44137, "step": 4707} +{"lr": 0.00028930682014353937, "data_time": 0.0014656305313110352, "loss": 0.37468068301677704, "time": 0.634208345413208, "epoch": 2, "memory": 44137, "step": 4807} +{"lr": 0.0002888630418962978, "data_time": 0.001390552520751953, "loss": 0.38818833529949187, "time": 0.6345075607299805, "epoch": 2, "memory": 44137, "step": 4907} +{"lr": 0.00028841060856534716, "data_time": 0.0018620491027832031, "loss": 0.35287671387195585, "time": 0.6331895112991333, "epoch": 2, "memory": 44137, "step": 5007} +{"lr": 0.0002879495494033338, "data_time": 0.0014444351196289062, "loss": 0.38743217587471007, "time": 0.633136248588562, "epoch": 2, "memory": 44137, "step": 5107} +{"lr": 0.00028747989422061985, "data_time": 0.0013684988021850585, "loss": 0.38438221514225007, "time": 0.6332125663757324, "epoch": 2, "memory": 44137, "step": 5207} +{"lr": 0.0002870016733833529, "data_time": 0.0014147520065307616, "loss": 0.37357968688011167, "time": 0.6330580472946167, "epoch": 2, "memory": 44137, "step": 5307} +{"lr": 0.00028651491781150423, "data_time": 0.001379537582397461, "loss": 0.38055368661880495, "time": 0.6336659908294677, "epoch": 2, "memory": 44137, "step": 5407} +{"lr": 0.00028601965897686856, "data_time": 0.0012953758239746093, "loss": 0.37099160850048063, "time": 0.6349786281585693, "epoch": 2, "memory": 44137, "step": 5507} +{"lr": 0.0002855159289010302, "data_time": 0.001461935043334961, "loss": 0.37708382308483124, "time": 0.643999171257019, "epoch": 2, "memory": 44137, "step": 5607} +{"lr": 0.00028500376015329224, "data_time": 0.0014859914779663086, "loss": 0.38429166078567506, "time": 0.633487343788147, "epoch": 2, "memory": 44137, "step": 5707} +{"lr": 0.00028448318584857045, "data_time": 0.0015755414962768555, "loss": 0.4045539706945419, "time": 0.6334191083908081, "epoch": 2, "memory": 44137, "step": 5807} +{"lr": 0.000283954239645253, "data_time": 0.0013811826705932618, "loss": 0.36398411393165586, "time": 0.6341552734375, "epoch": 2, "memory": 44137, "step": 5907} +{"lr": 0.0002834169557430233, "data_time": 0.0014555215835571288, "loss": 0.3547777682542801, "time": 0.6336054325103759, "epoch": 2, "memory": 44137, "step": 6007} +{"lr": 0.00028287136888064903, "data_time": 0.00139312744140625, "loss": 0.37483089566230776, "time": 0.6332645893096924, "epoch": 2, "memory": 44137, "step": 6107} +{"lr": 0.00028231751433373687, "data_time": 0.0014421939849853516, "loss": 0.36126171946525576, "time": 0.6331502199172974, "epoch": 2, "memory": 44137, "step": 6207} +{"lr": 0.00028175542791245077, "data_time": 0.0015299081802368163, "loss": 0.3846882224082947, "time": 0.6347301483154297, "epoch": 2, "memory": 44137, "step": 6307} +{"lr": 0.0002811851459591966, "data_time": 0.0013877391815185548, "loss": 0.3941593259572983, "time": 0.6332796812057495, "epoch": 2, "memory": 44137, "step": 6407} +{"lr": 0.00028060670534627274, "data_time": 0.0014113187789916992, "loss": 0.3431746423244476, "time": 0.6329907894134521, "epoch": 2, "memory": 44137, "step": 6507} +{"lr": 0.00028002014347348626, "data_time": 0.0013391494750976563, "loss": 0.4181104600429535, "time": 0.6329025030136108, "epoch": 2, "memory": 44137, "step": 6607} +{"lr": 0.00027942549826573453, "data_time": 0.0014209985733032227, "loss": 0.35405783355236053, "time": 0.6330006361007691, "epoch": 2, "memory": 44137, "step": 6707} +{"lr": 0.00027882280817055275, "data_time": 0.0015227079391479492, "loss": 0.3673249214887619, "time": 0.6331135272979737, "epoch": 2, "memory": 44137, "step": 6807} +{"lr": 0.000278212112155629, "data_time": 0.0014508724212646484, "loss": 0.3860452800989151, "time": 0.6328620433807373, "epoch": 2, "memory": 44137, "step": 6907} +{"lr": 0.0002775934497062837, "data_time": 0.0015659093856811523, "loss": 0.36105334758758545, "time": 0.6331103563308715, "epoch": 2, "memory": 44137, "step": 7007} +{"lr": 0.00027696686082291747, "data_time": 0.0014065265655517577, "loss": 0.39859779477119445, "time": 0.6327624797821045, "epoch": 2, "memory": 44137, "step": 7107} +{"lr": 0.00027633238601842443, "data_time": 0.0013328313827514649, "loss": 0.3689152032136917, "time": 0.6332187652587891, "epoch": 2, "memory": 44137, "step": 7207} +{"lr": 0.0002756900663155731, "data_time": 0.0013592243194580078, "loss": 0.3702657222747803, "time": 0.6327556848526001, "epoch": 2, "memory": 44137, "step": 7307} +{"lr": 0.0002750399432443529, "data_time": 0.0014490842819213866, "loss": 0.4029386878013611, "time": 0.6327578306198121, "epoch": 2, "memory": 44137, "step": 7407} +{"lr": 0.0002743820588392909, "data_time": 0.0014390945434570312, "loss": 0.3923063039779663, "time": 0.6333898305892944, "epoch": 2, "memory": 44137, "step": 7507} +{"lr": 0.00027371645563673247, "data_time": 0.0015081405639648438, "loss": 0.37271900177001954, "time": 0.6331567287445068, "epoch": 2, "memory": 44137, "step": 7607} +{"lr": 0.0002730431766720916, "data_time": 0.0014412403106689453, "loss": 0.3904691100120544, "time": 0.6326159238815308, "epoch": 2, "memory": 44137, "step": 7707} +{"lr": 0.0002723622654770684, "data_time": 0.0013747930526733399, "loss": 0.3633644700050354, "time": 0.6329151868820191, "epoch": 2, "memory": 44137, "step": 7807} +{"accuracy/top1": 73.03504180908203, "data_time": 0.041268060947286675, "time": 1.2850785748711948, "step": 2} +{"lr": 0.0002716252880879063, "data_time": 0.0014528512954711914, "loss": 0.39044045805931094, "time": 0.6343801021575928, "epoch": 3, "memory": 44137, "step": 7914} +{"lr": 0.000270928718619737, "data_time": 0.0015842676162719726, "loss": 0.350239160656929, "time": 0.6342518091201782, "epoch": 3, "memory": 44137, "step": 8014} +{"lr": 0.0002702246536341361, "data_time": 0.001535344123840332, "loss": 0.3971865147352219, "time": 0.6333378791809082, "epoch": 3, "memory": 44137, "step": 8114} +{"lr": 0.0002695131386533127, "data_time": 0.001560378074645996, "loss": 0.36411244571208956, "time": 0.635346508026123, "epoch": 3, "memory": 44137, "step": 8214} +{"lr": 0.0002687942196811649, "data_time": 0.0014538288116455079, "loss": 0.4046099454164505, "time": 0.63466477394104, "epoch": 3, "memory": 44137, "step": 8314} +{"lr": 0.0002680679432003047, "data_time": 0.0013802051544189453, "loss": 0.37574750781059263, "time": 0.6367144346237182, "epoch": 3, "memory": 44137, "step": 8414} +{"lr": 0.0002673343561690538, "data_time": 0.0014797210693359374, "loss": 0.37542503178119657, "time": 0.633153247833252, "epoch": 3, "memory": 44137, "step": 8514} +{"lr": 0.00026659350601840597, "data_time": 0.0014623641967773438, "loss": 0.38963502645492554, "time": 0.642594575881958, "epoch": 3, "memory": 44137, "step": 8614} +{"lr": 0.0002658454406489622, "data_time": 0.0013822317123413086, "loss": 0.3794644445180893, "time": 0.6330184459686279, "epoch": 3, "memory": 44137, "step": 8714} +{"lr": 0.0002650902084278314, "data_time": 0.0014023303985595704, "loss": 0.3861773073673248, "time": 0.6335261821746826, "epoch": 3, "memory": 44137, "step": 8814} +{"lr": 0.00026432785818550494, "data_time": 0.0014488935470581056, "loss": 0.4059780925512314, "time": 0.6336912870407104, "epoch": 3, "memory": 44137, "step": 8914} +{"lr": 0.00026355843921269914, "data_time": 0.00146939754486084, "loss": 0.3851300150156021, "time": 0.6342427968978882, "epoch": 3, "memory": 44137, "step": 9014} +{"lr": 0.0002627820012571672, "data_time": 0.0013885736465454102, "loss": 0.3818380951881409, "time": 0.6332744121551513, "epoch": 3, "memory": 44137, "step": 9114} +{"lr": 0.00026199859452048346, "data_time": 0.0013898372650146484, "loss": 0.3860693693161011, "time": 0.6335134029388427, "epoch": 3, "memory": 44137, "step": 9214} +{"lr": 0.00026120826965479813, "data_time": 0.0015716552734375, "loss": 0.4048592269420624, "time": 0.6336831331253052, "epoch": 3, "memory": 44137, "step": 9314} +{"lr": 0.0002604110777595625, "data_time": 0.0014738082885742188, "loss": 0.3931962579488754, "time": 0.6340368270874024, "epoch": 3, "memory": 44137, "step": 9414} +{"lr": 0.000259607070378222, "data_time": 0.0016142606735229492, "loss": 0.36188422739505766, "time": 0.6334050178527832, "epoch": 3, "memory": 44137, "step": 9514} +{"lr": 0.00025879629949488836, "data_time": 0.0014542341232299805, "loss": 0.3689572334289551, "time": 0.6326395273208618, "epoch": 3, "memory": 44137, "step": 9614} +{"lr": 0.00025797881753097425, "data_time": 0.0013261795043945312, "loss": 0.3661950796842575, "time": 0.6339035511016846, "epoch": 3, "memory": 44137, "step": 9714} +{"lr": 0.000257154677341807, "data_time": 0.001378941535949707, "loss": 0.4190051108598709, "time": 0.6333179950714112, "epoch": 3, "memory": 44137, "step": 9814} +{"lr": 0.0002563239322132094, "data_time": 0.0014253377914428711, "loss": 0.4014905631542206, "time": 0.6335712909698487, "epoch": 3, "memory": 44137, "step": 9914} +{"lr": 0.00025548663585805504, "data_time": 0.001434659957885742, "loss": 0.352135044336319, "time": 0.6362543106079102, "epoch": 3, "memory": 44137, "step": 10014} +{"lr": 0.0002546428424127946, "data_time": 0.001401805877685547, "loss": 0.3941739499568939, "time": 0.634401512145996, "epoch": 3, "memory": 44137, "step": 10114} +{"lr": 0.00025379260643395755, "data_time": 0.0014047861099243165, "loss": 0.36856828927993773, "time": 0.6330304861068725, "epoch": 3, "memory": 44137, "step": 10214} +{"lr": 0.0002529359828946234, "data_time": 0.0013712644577026367, "loss": 0.39322973489761354, "time": 0.6345672130584716, "epoch": 3, "memory": 44137, "step": 10314} +{"lr": 0.00025207302718086624, "data_time": 0.0014400005340576172, "loss": 0.3807382553815842, "time": 0.6328573226928711, "epoch": 3, "memory": 44137, "step": 10414} +{"lr": 0.00025120379508817534, "data_time": 0.0014844417572021484, "loss": 0.3917868763208389, "time": 0.6344710350036621, "epoch": 3, "memory": 44137, "step": 10514} +{"lr": 0.0002503283428178471, "data_time": 0.0014474868774414062, "loss": 0.41173630952835083, "time": 0.6344508171081543, "epoch": 3, "memory": 44137, "step": 10614} +{"lr": 0.00024944672697335175, "data_time": 0.0015144586563110352, "loss": 0.3974515050649643, "time": 0.6331386089324951, "epoch": 3, "memory": 44137, "step": 10714} +{"lr": 0.00024855900455667294, "data_time": 0.0013801336288452148, "loss": 0.39543994665145876, "time": 0.6353105545043946, "epoch": 3, "memory": 44137, "step": 10814} +{"lr": 0.0002476652329646227, "data_time": 0.0014162063598632812, "loss": 0.3588362991809845, "time": 0.6332003116607666, "epoch": 3, "memory": 44137, "step": 10914} +{"lr": 0.0002467654699851294, "data_time": 0.0013519287109375, "loss": 0.40977767705917356, "time": 0.6330472230911255, "epoch": 3, "memory": 44137, "step": 11014} +{"lr": 0.000245859773793503, "data_time": 0.0014048099517822265, "loss": 0.40263975858688356, "time": 0.6333052396774292, "epoch": 3, "memory": 44137, "step": 11114} +{"lr": 0.0002449482029486725, "data_time": 0.0016291618347167968, "loss": 0.38870387375354765, "time": 0.6329368352890015, "epoch": 3, "memory": 44137, "step": 11214} +{"lr": 0.00024403081638939956, "data_time": 0.0015357494354248046, "loss": 0.38911502361297606, "time": 0.6339688539505005, "epoch": 3, "memory": 44137, "step": 11314} +{"lr": 0.00024310767343046748, "data_time": 0.0015012741088867188, "loss": 0.41671150028705595, "time": 0.6327686548233032, "epoch": 3, "memory": 44137, "step": 11414} +{"lr": 0.00024217883375884914, "data_time": 0.0014972686767578125, "loss": 0.3951862722635269, "time": 0.6337986707687377, "epoch": 3, "memory": 44137, "step": 11514} +{"lr": 0.0002412443574298446, "data_time": 0.0015561342239379882, "loss": 0.37027890980243683, "time": 0.6331808567047119, "epoch": 3, "memory": 44137, "step": 11614} +{"lr": 0.00024030430486319854, "data_time": 0.001108551025390625, "loss": 0.4004129350185394, "time": 0.6335879802703858, "epoch": 3, "memory": 44137, "step": 11714} +{"accuracy/top1": 73.64740753173828, "data_time": 0.03590079833721292, "time": 1.284608479203849, "step": 3} +{"lr": 0.00023929234210131621, "data_time": 0.0015595912933349609, "loss": 0.3860115647315979, "time": 0.6335151195526123, "epoch": 4, "memory": 44137, "step": 11821} +{"lr": 0.0002383409402531276, "data_time": 0.001466655731201172, "loss": 0.3927760601043701, "time": 0.6331015348434448, "epoch": 4, "memory": 44137, "step": 11921} +{"lr": 0.00023738414989139526, "data_time": 0.0014872312545776366, "loss": 0.4064512252807617, "time": 0.6349523067474365, "epoch": 4, "memory": 44137, "step": 12021} +{"lr": 0.0002364220328786055, "data_time": 0.0014856100082397462, "loss": 0.397469300031662, "time": 0.6337270736694336, "epoch": 4, "memory": 44137, "step": 12121} +{"lr": 0.0002354546514216467, "data_time": 0.0015802621841430665, "loss": 0.3992577761411667, "time": 0.6337252855300903, "epoch": 4, "memory": 44137, "step": 12221} +{"lr": 0.00023448206806778656, "data_time": 0.001455974578857422, "loss": 0.3807056397199631, "time": 0.6341383218765259, "epoch": 4, "memory": 44137, "step": 12321} +{"lr": 0.0002335043457006275, "data_time": 0.0014228105545043945, "loss": 0.3721867769956589, "time": 0.6345484018325805, "epoch": 4, "memory": 44137, "step": 12421} +{"lr": 0.00023252154753604175, "data_time": 0.0016866207122802734, "loss": 0.40315210819244385, "time": 0.6342388391494751, "epoch": 4, "memory": 44137, "step": 12521} +{"lr": 0.0002315337371180835, "data_time": 0.0015563488006591797, "loss": 0.40187896490097047, "time": 0.6331060886383056, "epoch": 4, "memory": 44137, "step": 12621} +{"lr": 0.00023054097831488017, "data_time": 0.0014397859573364257, "loss": 0.4120116919279099, "time": 0.6351639270782471, "epoch": 4, "memory": 44137, "step": 12721} +{"lr": 0.00022954333531450405, "data_time": 0.0015584945678710938, "loss": 0.39791989624500274, "time": 0.6338698148727417, "epoch": 4, "memory": 44137, "step": 12821} +{"lr": 0.00022854087262082044, "data_time": 0.0015718460083007813, "loss": 0.4025317281484604, "time": 0.634430193901062, "epoch": 4, "memory": 44137, "step": 12921} +{"lr": 0.00022753365504931858, "data_time": 0.001687312126159668, "loss": 0.3890873521566391, "time": 0.6327084541320801, "epoch": 4, "memory": 44137, "step": 13021} +{"lr": 0.0002265217477229201, "data_time": 0.0014850854873657226, "loss": 0.37108676731586454, "time": 0.6341103792190552, "epoch": 4, "memory": 44137, "step": 13121} +{"lr": 0.0002255052160677689, "data_time": 0.0014713048934936524, "loss": 0.3800925225019455, "time": 0.6344964742660523, "epoch": 4, "memory": 44137, "step": 13221} +{"lr": 0.00022448412580900032, "data_time": 0.0014410734176635743, "loss": 0.38547907769680023, "time": 0.6336946725845337, "epoch": 4, "memory": 44137, "step": 13321} +{"lr": 0.0002234585429664922, "data_time": 0.001412177085876465, "loss": 0.3935152441263199, "time": 0.6346070766448975, "epoch": 4, "memory": 44137, "step": 13421} +{"lr": 0.0002224285338505958, "data_time": 0.001433086395263672, "loss": 0.3959517806768417, "time": 0.6363736391067505, "epoch": 4, "memory": 44137, "step": 13521} +{"lr": 0.00022139416505784887, "data_time": 0.0014551877975463867, "loss": 0.4125809222459793, "time": 0.6373970031738281, "epoch": 4, "memory": 44137, "step": 13621} +{"lr": 0.0002203555034666699, "data_time": 0.0014021635055541993, "loss": 0.40133764743804934, "time": 0.6338953733444214, "epoch": 4, "memory": 44137, "step": 13721} +{"lr": 0.00021931261623303318, "data_time": 0.001403641700744629, "loss": 0.40483948588371277, "time": 0.6342689752578735, "epoch": 4, "memory": 44137, "step": 13821} +{"lr": 0.00021826557078612706, "data_time": 0.0014425992965698242, "loss": 0.40545822083950045, "time": 0.6332755088806152, "epoch": 4, "memory": 44137, "step": 13921} +{"lr": 0.0002172144348239949, "data_time": 0.0014600753784179688, "loss": 0.407717701792717, "time": 0.6354591369628906, "epoch": 4, "memory": 44137, "step": 14021} +{"lr": 0.0002161592763091573, "data_time": 0.0014806032180786134, "loss": 0.4162604361772537, "time": 0.6345121622085571, "epoch": 4, "memory": 44137, "step": 14121} +{"lr": 0.00021510016346421842, "data_time": 0.0015603065490722655, "loss": 0.37755178809165957, "time": 0.6340918064117431, "epoch": 4, "memory": 44137, "step": 14221} +{"lr": 0.0002140371647674538, "data_time": 0.0013860940933227539, "loss": 0.42633184790611267, "time": 0.633940577507019, "epoch": 4, "memory": 44137, "step": 14321} +{"lr": 0.0002129703489483841, "data_time": 0.00148162841796875, "loss": 0.3801177591085434, "time": 0.6336155891418457, "epoch": 4, "memory": 44137, "step": 14421} +{"lr": 0.00021189978498333095, "data_time": 0.0013858795166015625, "loss": 0.3921238601207733, "time": 0.6334233045578003, "epoch": 4, "memory": 44137, "step": 14521} +{"lr": 0.00021082554209095715, "data_time": 0.001500701904296875, "loss": 0.38124058544635775, "time": 0.6341081857681274, "epoch": 4, "memory": 44137, "step": 14621} +{"lr": 0.0002097476897277908, "data_time": 0.0014397621154785157, "loss": 0.3904690682888031, "time": 0.6345602035522461, "epoch": 4, "memory": 44137, "step": 14721} +{"lr": 0.00020866629758373506, "data_time": 0.0017169952392578126, "loss": 0.4120761275291443, "time": 0.6331814527511597, "epoch": 4, "memory": 44137, "step": 14821} +{"lr": 0.0002075814355775621, "data_time": 0.0015081167221069336, "loss": 0.3612970173358917, "time": 0.6327624082565307, "epoch": 4, "memory": 44137, "step": 14921} +{"lr": 0.0002064931738523926, "data_time": 0.001503157615661621, "loss": 0.37701648473739624, "time": 0.6338628053665161, "epoch": 4, "memory": 44137, "step": 15021} +{"lr": 0.00020540158277115988, "data_time": 0.0015111684799194336, "loss": 0.3903593778610229, "time": 0.634803295135498, "epoch": 4, "memory": 44137, "step": 15121} +{"lr": 0.00020430673291206152, "data_time": 0.0015198469161987304, "loss": 0.39292154014110564, "time": 0.634749436378479, "epoch": 4, "memory": 44137, "step": 15221} +{"lr": 0.00020320869506399518, "data_time": 0.0014705657958984375, "loss": 0.4145514190196991, "time": 0.6337226390838623, "epoch": 4, "memory": 44137, "step": 15321} +{"lr": 0.00020210754022198212, "data_time": 0.0014455318450927734, "loss": 0.38167398869991304, "time": 0.6330749750137329, "epoch": 4, "memory": 44137, "step": 15421} +{"lr": 0.00020100333958257622, "data_time": 0.0016491413116455078, "loss": 0.3898289382457733, "time": 0.6344200611114502, "epoch": 4, "memory": 44137, "step": 15521} +{"lr": 0.00019989616453926207, "data_time": 0.001222681999206543, "loss": 0.4088809311389923, "time": 0.6353185415267945, "epoch": 4, "memory": 44137, "step": 15621} +{"accuracy/top1": 74.10319519042969, "data_time": 0.03870173980449808, "time": 1.2833262764174362, "step": 4} +{"lr": 0.00019870827437055215, "data_time": 0.0015145540237426758, "loss": 0.38957429826259615, "time": 0.6335341453552246, "epoch": 5, "memory": 44137, "step": 15728} +{"lr": 0.00019759516998405188, "data_time": 0.0014893293380737304, "loss": 0.4005758613348007, "time": 0.6336127996444703, "epoch": 5, "memory": 44137, "step": 15828} +{"lr": 0.0001964793115531446, "data_time": 0.001859760284423828, "loss": 0.3996329098939896, "time": 0.6330718040466309, "epoch": 5, "memory": 44137, "step": 15928} +{"lr": 0.00019536077122506366, "data_time": 0.0014342784881591797, "loss": 0.41851462721824645, "time": 0.6336797952651978, "epoch": 5, "memory": 44137, "step": 16028} +{"lr": 0.00019423962132044425, "data_time": 0.0013837099075317382, "loss": 0.3988448053598404, "time": 0.6334346771240235, "epoch": 5, "memory": 44137, "step": 16128} +{"lr": 0.00019311593432864616, "data_time": 0.0014795780181884766, "loss": 0.4061956971883774, "time": 0.6338543653488159, "epoch": 5, "memory": 44137, "step": 16228} +{"lr": 0.00019198978290306865, "data_time": 0.0014634370803833009, "loss": 0.3885423570871353, "time": 0.6336427688598633, "epoch": 5, "memory": 44137, "step": 16328} +{"lr": 0.00019086123985645166, "data_time": 0.0016359806060791016, "loss": 0.4009571045637131, "time": 0.6340352773666382, "epoch": 5, "memory": 44137, "step": 16428} +{"lr": 0.00018973037815616829, "data_time": 0.0016285181045532227, "loss": 0.38526704609394075, "time": 0.6328840255737305, "epoch": 5, "memory": 44137, "step": 16528} +{"lr": 0.00018859727091950751, "data_time": 0.0017131567001342773, "loss": 0.40259073078632357, "time": 0.6333629846572876, "epoch": 5, "memory": 44137, "step": 16628} +{"lr": 0.0001874619914089457, "data_time": 0.0014320135116577149, "loss": 0.40346805155277254, "time": 0.6330579996109009, "epoch": 5, "memory": 44137, "step": 16728} +{"lr": 0.0001863246130274106, "data_time": 0.0013761520385742188, "loss": 0.4131861627101898, "time": 0.6331458330154419, "epoch": 5, "memory": 44137, "step": 16828} +{"lr": 0.00018518520931353537, "data_time": 0.0015362739562988282, "loss": 0.366819965839386, "time": 0.6339886665344239, "epoch": 5, "memory": 44137, "step": 16928} +{"lr": 0.00018404385393690312, "data_time": 0.0014780521392822265, "loss": 0.4048200726509094, "time": 0.6333035230636597, "epoch": 5, "memory": 44137, "step": 17028} +{"lr": 0.00018290062069328442, "data_time": 0.0015853643417358398, "loss": 0.39369923770427706, "time": 0.6330546617507935, "epoch": 5, "memory": 44137, "step": 17128} +{"lr": 0.00018175558349986583, "data_time": 0.0014896392822265625, "loss": 0.40191271007061, "time": 0.6332434892654419, "epoch": 5, "memory": 44137, "step": 17228} +{"lr": 0.00018060881639047043, "data_time": 0.0014326810836791993, "loss": 0.38185541331768036, "time": 0.6329642057418823, "epoch": 5, "memory": 44137, "step": 17328} +{"lr": 0.00017946039351077116, "data_time": 0.00143740177154541, "loss": 0.4041925251483917, "time": 0.6335988521575928, "epoch": 5, "memory": 44137, "step": 17428} +{"lr": 0.000178310389113497, "data_time": 0.0015124797821044922, "loss": 0.39412103295326234, "time": 0.6328449487686157, "epoch": 5, "memory": 44137, "step": 17528} +{"lr": 0.00017715887755363153, "data_time": 0.0014708995819091796, "loss": 0.3919717252254486, "time": 0.6330690383911133, "epoch": 5, "memory": 44137, "step": 17628} +{"lr": 0.00017600593328360648, "data_time": 0.0014266729354858398, "loss": 0.4195596158504486, "time": 0.6335479021072388, "epoch": 5, "memory": 44137, "step": 17728} +{"lr": 0.00017485163084848682, "data_time": 0.0016534090042114257, "loss": 0.38832607865333557, "time": 0.6329854011535645, "epoch": 5, "memory": 44137, "step": 17828} +{"lr": 0.00017369604488115154, "data_time": 0.0015085458755493164, "loss": 0.41594268679618834, "time": 0.633428406715393, "epoch": 5, "memory": 44137, "step": 17928} +{"lr": 0.00017253925009746786, "data_time": 0.0013206720352172852, "loss": 0.3879138708114624, "time": 0.6330734968185425, "epoch": 5, "memory": 44137, "step": 18028} +{"lr": 0.0001713813212914606, "data_time": 0.0014251470565795898, "loss": 0.430080246925354, "time": 0.632841682434082, "epoch": 5, "memory": 44137, "step": 18128} +{"lr": 0.00017022233333047657, "data_time": 0.0015694618225097657, "loss": 0.4012255847454071, "time": 0.633358359336853, "epoch": 5, "memory": 44137, "step": 18228} +{"lr": 0.0001690623611503431, "data_time": 0.0015277385711669922, "loss": 0.3898197740316391, "time": 0.6365635633468628, "epoch": 5, "memory": 44137, "step": 18328} +{"lr": 0.00016790147975052384, "data_time": 0.0013677835464477538, "loss": 0.4272844076156616, "time": 0.6342239379882812, "epoch": 5, "memory": 44137, "step": 18428} +{"lr": 0.00016673976418926875, "data_time": 0.0014503002166748047, "loss": 0.38555141985416413, "time": 0.6334893226623535, "epoch": 5, "memory": 44137, "step": 18528} +{"lr": 0.00016557728957876201, "data_time": 0.0015038251876831055, "loss": 0.3669358015060425, "time": 0.6338192701339722, "epoch": 5, "memory": 44137, "step": 18628} +{"lr": 0.0001644141310802647, "data_time": 0.0014124393463134765, "loss": 0.4031758517026901, "time": 0.6337247133255005, "epoch": 5, "memory": 44137, "step": 18728} +{"lr": 0.0001632503638992558, "data_time": 0.0014987230300903321, "loss": 0.4009410500526428, "time": 0.6338726997375488, "epoch": 5, "memory": 44137, "step": 18828} +{"lr": 0.00016208606328056935, "data_time": 0.0014945745468139648, "loss": 0.38522562980651853, "time": 0.6337974071502686, "epoch": 5, "memory": 44137, "step": 18928} +{"lr": 0.00016092130450352928, "data_time": 0.0014091968536376954, "loss": 0.4015228807926178, "time": 0.6343581438064575, "epoch": 5, "memory": 44137, "step": 19028} +{"lr": 0.00015975616287708285, "data_time": 0.001501321792602539, "loss": 0.384323787689209, "time": 0.6346012353897095, "epoch": 5, "memory": 44137, "step": 19128} +{"lr": 0.00015859071373493025, "data_time": 0.001623082160949707, "loss": 0.4046397626399994, "time": 0.6335388422012329, "epoch": 5, "memory": 44137, "step": 19228} +{"lr": 0.00015742503243065513, "data_time": 0.0014320611953735352, "loss": 0.42110501527786254, "time": 0.6338286399841309, "epoch": 5, "memory": 44137, "step": 19328} +{"lr": 0.0001562591943328516, "data_time": 0.001457810401916504, "loss": 0.4157467633485794, "time": 0.6340137958526612, "epoch": 5, "memory": 44137, "step": 19428} +{"lr": 0.0001550932748202511, "data_time": 0.0013442039489746094, "loss": 0.3957761228084564, "time": 0.6333126306533814, "epoch": 5, "memory": 44137, "step": 19528} +{"accuracy/top1": 74.89649200439453, "data_time": 0.035657381189280545, "time": 1.2823072589676956, "step": 5} +{"lr": 0.00015384573621094385, "data_time": 0.001628279685974121, "loss": 0.41077532172203063, "time": 0.6338565111160278, "epoch": 6, "memory": 44137, "step": 19635} +{"lr": 0.00015267988769893654, "data_time": 0.0014618158340454102, "loss": 0.396423476934433, "time": 0.6333179235458374, "epoch": 6, "memory": 44137, "step": 19735} +{"lr": 0.0001515141891967143, "data_time": 0.0014534711837768555, "loss": 0.3975360870361328, "time": 0.6329234600067138, "epoch": 6, "memory": 44137, "step": 19835} +{"lr": 0.00015034871607398357, "data_time": 0.0013907670974731446, "loss": 0.3965549886226654, "time": 0.6331725597381592, "epoch": 6, "memory": 44137, "step": 19935} +{"lr": 0.00014918354368587822, "data_time": 0.0015391111373901367, "loss": 0.4015242636203766, "time": 0.6330177307128906, "epoch": 6, "memory": 44137, "step": 20035} +{"lr": 0.00014801874736808767, "data_time": 0.001451730728149414, "loss": 0.42079220414161683, "time": 0.6342101097106934, "epoch": 6, "memory": 44137, "step": 20135} +{"lr": 0.00014685440243198569, "data_time": 0.001372528076171875, "loss": 0.3958262801170349, "time": 0.6332395315170288, "epoch": 6, "memory": 44137, "step": 20235} +{"lr": 0.00014569058415976212, "data_time": 0.0016178846359252929, "loss": 0.41581864953041076, "time": 0.6336359262466431, "epoch": 6, "memory": 44137, "step": 20335} +{"lr": 0.00014452736779955436, "data_time": 0.0015796661376953126, "loss": 0.4055685460567474, "time": 0.633411455154419, "epoch": 6, "memory": 44137, "step": 20435} +{"lr": 0.0001433648285605824, "data_time": 0.001438474655151367, "loss": 0.42735623717308047, "time": 0.6328923225402832, "epoch": 6, "memory": 44137, "step": 20535} +{"lr": 0.00014220304160828594, "data_time": 0.001488947868347168, "loss": 0.4199580252170563, "time": 0.6342294931411743, "epoch": 6, "memory": 44137, "step": 20635} +{"lr": 0.00014104208205946515, "data_time": 0.001546335220336914, "loss": 0.40364215672016146, "time": 0.6343457937240601, "epoch": 6, "memory": 44137, "step": 20735} +{"lr": 0.00013988202497742296, "data_time": 0.0015279054641723633, "loss": 0.3883852481842041, "time": 0.6339449644088745, "epoch": 6, "memory": 44137, "step": 20835} +{"lr": 0.00013872294536711234, "data_time": 0.0015924692153930664, "loss": 0.41546320617198945, "time": 0.6349325180053711, "epoch": 6, "memory": 44137, "step": 20935} +{"lr": 0.0001375649181702862, "data_time": 0.0015136003494262695, "loss": 0.40236571431159973, "time": 0.6335031747817993, "epoch": 6, "memory": 44137, "step": 21035} +{"lr": 0.0001364080182606529, "data_time": 0.0015846490859985352, "loss": 0.40679511427879333, "time": 0.6332422494888306, "epoch": 6, "memory": 44137, "step": 21135} +{"lr": 0.00013525232043903418, "data_time": 0.0015084981918334962, "loss": 0.3796483725309372, "time": 0.6350220680236817, "epoch": 6, "memory": 44137, "step": 21235} +{"lr": 0.0001340978994285297, "data_time": 0.0013573169708251953, "loss": 0.4032929718494415, "time": 0.6358685970306397, "epoch": 6, "memory": 44137, "step": 21335} +{"lr": 0.00013294482986968493, "data_time": 0.0015769481658935546, "loss": 0.3757345795631409, "time": 0.6332385063171386, "epoch": 6, "memory": 44137, "step": 21435} +{"lr": 0.00013179318631566578, "data_time": 0.0014328956604003906, "loss": 0.41288171112537386, "time": 0.6363572835922241, "epoch": 6, "memory": 44137, "step": 21535} +{"lr": 0.0001306430432274384, "data_time": 0.0015916109085083007, "loss": 0.41170698404312134, "time": 0.6335021495819092, "epoch": 6, "memory": 44137, "step": 21635} +{"lr": 0.00012949447496895343, "data_time": 0.0015311479568481446, "loss": 0.4057985067367554, "time": 0.6339433670043946, "epoch": 6, "memory": 44137, "step": 21735} +{"lr": 0.00012834755580233909, "data_time": 0.001506471633911133, "loss": 0.3743808776140213, "time": 0.6339581251144409, "epoch": 6, "memory": 44137, "step": 21835} +{"lr": 0.00012720235988310005, "data_time": 0.0014614582061767579, "loss": 0.4226959466934204, "time": 0.6358937740325927, "epoch": 6, "memory": 44137, "step": 21935} +{"lr": 0.0001260589612553219, "data_time": 0.0015298128128051758, "loss": 0.3868665754795074, "time": 0.6338008642196655, "epoch": 6, "memory": 44137, "step": 22035} +{"lr": 0.00012491743384688422, "data_time": 0.0013885259628295898, "loss": 0.39861064553260805, "time": 0.6350136041641236, "epoch": 6, "memory": 44137, "step": 22135} +{"lr": 0.0001237778514646806, "data_time": 0.0014439821243286133, "loss": 0.3788717776536942, "time": 0.6345320224761963, "epoch": 6, "memory": 44137, "step": 22235} +{"lr": 0.00012264028778984632, "data_time": 0.0014932394027709962, "loss": 0.4107115536928177, "time": 0.6335056066513062, "epoch": 6, "memory": 44137, "step": 22335} +{"lr": 0.00012150481637299435, "data_time": 0.001719498634338379, "loss": 0.3938616216182709, "time": 0.6344237565994263, "epoch": 6, "memory": 44137, "step": 22435} +{"lr": 0.00012037151062946036, "data_time": 0.001532578468322754, "loss": 0.4374720364809036, "time": 0.6367403745651246, "epoch": 6, "memory": 44137, "step": 22535} +{"lr": 0.00011924044383455631, "data_time": 0.0015212535858154298, "loss": 0.41519361138343813, "time": 0.634022879600525, "epoch": 6, "memory": 44137, "step": 22635} +{"lr": 0.0001181116891188313, "data_time": 0.0014603614807128906, "loss": 0.3899745434522629, "time": 0.6358419179916381, "epoch": 6, "memory": 44137, "step": 22735} +{"lr": 0.0001169853194633444, "data_time": 0.0014418601989746095, "loss": 0.412470543384552, "time": 0.6346251010894776, "epoch": 6, "memory": 44137, "step": 22835} +{"lr": 0.00011586140769494541, "data_time": 0.0015546321868896485, "loss": 0.41434812545776367, "time": 0.6342344522476197, "epoch": 6, "memory": 44137, "step": 22935} +{"lr": 0.00011474002648156654, "data_time": 0.001478719711303711, "loss": 0.4068192094564438, "time": 0.6443174839019775, "epoch": 6, "memory": 44137, "step": 23035} +{"lr": 0.00011362124832752344, "data_time": 0.0015213727951049806, "loss": 0.40099134743213655, "time": 0.6344565629959107, "epoch": 6, "memory": 44137, "step": 23135} +{"lr": 0.00011250514556882821, "data_time": 0.0013526439666748046, "loss": 0.4028135299682617, "time": 0.6347667217254639, "epoch": 6, "memory": 44137, "step": 23235} +{"lr": 0.00011139179036851135, "data_time": 0.0015400171279907227, "loss": 0.413922518491745, "time": 0.6342664957046509, "epoch": 6, "memory": 44137, "step": 23335} +{"lr": 0.00011028125471195667, "data_time": 0.0012638568878173828, "loss": 0.38831880390644075, "time": 0.6342539548873901, "epoch": 6, "memory": 44137, "step": 23435} +{"accuracy/top1": 74.70512390136719, "data_time": 0.03615083365604795, "time": 1.2833006998588299, "step": 6} +{"lr": 0.00010909618543324382, "data_time": 0.0014945507049560548, "loss": 0.4046571433544159, "time": 0.634126091003418, "epoch": 7, "memory": 44137, "step": 23542} +{"lr": 0.00010799171417102892, "data_time": 0.0014247417449951172, "loss": 0.44820262789726256, "time": 0.6341590881347656, "epoch": 7, "memory": 44137, "step": 23642} +{"lr": 0.0001068902822887887, "data_time": 0.001377701759338379, "loss": 0.4107129335403442, "time": 0.6343985080718995, "epoch": 7, "memory": 44137, "step": 23742} +{"lr": 0.00010579196100098985, "data_time": 0.0015125274658203125, "loss": 0.3960062026977539, "time": 0.6347365856170655, "epoch": 7, "memory": 44137, "step": 23842} +{"lr": 0.00010469682132097967, "data_time": 0.0015568494796752929, "loss": 0.3926637411117554, "time": 0.6343899250030518, "epoch": 7, "memory": 44137, "step": 23942} +{"lr": 0.00010360493405639455, "data_time": 0.001444244384765625, "loss": 0.3906110465526581, "time": 0.6349032402038575, "epoch": 7, "memory": 44137, "step": 24042} +{"lr": 0.00010251636980458193, "data_time": 0.0013445615768432617, "loss": 0.39741308987140656, "time": 0.6343841552734375, "epoch": 7, "memory": 44137, "step": 24142} +{"lr": 0.00010143119894803575, "data_time": 0.0014751672744750977, "loss": 0.4008590131998062, "time": 0.6349122524261475, "epoch": 7, "memory": 44137, "step": 24242} +{"lr": 0.00010034949164984562, "data_time": 0.001449894905090332, "loss": 0.3945930600166321, "time": 0.6343448162078857, "epoch": 7, "memory": 44137, "step": 24342} +{"lr": 9.927131784916038e-05, "data_time": 0.0014475107192993164, "loss": 0.3930166274309158, "time": 0.6361006736755371, "epoch": 7, "memory": 44137, "step": 24442} +{"lr": 9.819674725666596e-05, "data_time": 0.0015426874160766602, "loss": 0.3950634658336639, "time": 0.6358383655548095, "epoch": 7, "memory": 44137, "step": 24542} +{"lr": 9.712584935007872e-05, "data_time": 0.0014582157135009765, "loss": 0.41049647629261016, "time": 0.6364786863327027, "epoch": 7, "memory": 44137, "step": 24642} +{"lr": 9.605869336965269e-05, "data_time": 0.0014018535614013672, "loss": 0.4261118292808533, "time": 0.6357121944427491, "epoch": 7, "memory": 44137, "step": 24742} +{"lr": 9.499534831370295e-05, "data_time": 0.001417851448059082, "loss": 0.429586386680603, "time": 0.6352718114852905, "epoch": 7, "memory": 44137, "step": 24842} +{"lr": 9.39358829341445e-05, "data_time": 0.0013863086700439454, "loss": 0.3760415852069855, "time": 0.6354018449783325, "epoch": 7, "memory": 44137, "step": 24942} +{"lr": 9.288036573204672e-05, "data_time": 0.0016774892807006835, "loss": 0.40768120884895326, "time": 0.635544729232788, "epoch": 7, "memory": 44137, "step": 25042} +{"lr": 9.182886495320504e-05, "data_time": 0.0015383481979370118, "loss": 0.39910159409046175, "time": 0.6375270843505859, "epoch": 7, "memory": 44137, "step": 25142} +{"lr": 9.078144858372755e-05, "data_time": 0.0015362977981567382, "loss": 0.38406732082366946, "time": 0.6361876964569092, "epoch": 7, "memory": 44137, "step": 25242} +{"lr": 8.973818434563992e-05, "data_time": 0.001563286781311035, "loss": 0.3982101917266846, "time": 0.6353860378265381, "epoch": 7, "memory": 44137, "step": 25342} +{"lr": 8.86991396925066e-05, "data_time": 0.0014917612075805663, "loss": 0.40099319219589236, "time": 0.6339201927185059, "epoch": 7, "memory": 44137, "step": 25442} +{"lr": 8.766438180506934e-05, "data_time": 0.0014185190200805664, "loss": 0.4164343446493149, "time": 0.6347141981124877, "epoch": 7, "memory": 44137, "step": 25542} +{"lr": 8.66339775869035e-05, "data_time": 0.0015314102172851562, "loss": 0.40273507237434386, "time": 0.6362908124923706, "epoch": 7, "memory": 44137, "step": 25642} +{"lr": 8.560799366009288e-05, "data_time": 0.0015082836151123046, "loss": 0.43540977537631986, "time": 0.6343950033187866, "epoch": 7, "memory": 44137, "step": 25742} +{"lr": 8.458649636092149e-05, "data_time": 0.0014847517013549805, "loss": 0.3846946209669113, "time": 0.6344033241271972, "epoch": 7, "memory": 44137, "step": 25842} +{"lr": 8.356955173558475e-05, "data_time": 0.0014773368835449218, "loss": 0.37987988591194155, "time": 0.6341693878173829, "epoch": 7, "memory": 44137, "step": 25942} +{"lr": 8.255722553591929e-05, "data_time": 0.0014400720596313477, "loss": 0.3940939038991928, "time": 0.6344509601593018, "epoch": 7, "memory": 44137, "step": 26042} +{"lr": 8.154958321515152e-05, "data_time": 0.0015302181243896484, "loss": 0.4308096170425415, "time": 0.6344853639602661, "epoch": 7, "memory": 44137, "step": 26142} +{"lr": 8.054668992366594e-05, "data_time": 0.0015145540237426758, "loss": 0.39854896068573, "time": 0.6340305089950562, "epoch": 7, "memory": 44137, "step": 26242} +{"lr": 7.954861050479249e-05, "data_time": 0.0014557123184204101, "loss": 0.3925297200679779, "time": 0.634020733833313, "epoch": 7, "memory": 44137, "step": 26342} +{"lr": 7.855540949061403e-05, "data_time": 0.0014703035354614257, "loss": 0.39601756930351256, "time": 0.6359855890274048, "epoch": 7, "memory": 44137, "step": 26442} +{"lr": 7.756715109779415e-05, "data_time": 0.0013776302337646484, "loss": 0.43252343833446505, "time": 0.6359425067901612, "epoch": 7, "memory": 44137, "step": 26542} +{"lr": 7.658389922342482e-05, "data_time": 0.001479363441467285, "loss": 0.4226222574710846, "time": 0.6353020429611206, "epoch": 7, "memory": 44137, "step": 26642} +{"lr": 7.560571744089541e-05, "data_time": 0.0014433145523071289, "loss": 0.4033345192670822, "time": 0.6344865798950196, "epoch": 7, "memory": 44137, "step": 26742} +{"lr": 7.463266899578208e-05, "data_time": 0.0014557600021362304, "loss": 0.39786435663700104, "time": 0.6339991092681885, "epoch": 7, "memory": 44137, "step": 26842} +{"lr": 7.366481680175855e-05, "data_time": 0.0014534473419189453, "loss": 0.4089506298303604, "time": 0.6357561826705933, "epoch": 7, "memory": 44137, "step": 26942} +{"lr": 7.270222343652855e-05, "data_time": 0.0015775680541992188, "loss": 0.39817785322666166, "time": 0.6345189571380615, "epoch": 7, "memory": 44137, "step": 27042} +{"lr": 7.174495113777928e-05, "data_time": 0.0016486644744873047, "loss": 0.39844030141830444, "time": 0.6357248067855835, "epoch": 7, "memory": 44137, "step": 27142} +{"lr": 7.079306179915806e-05, "data_time": 0.0015341758728027344, "loss": 0.3968957901000977, "time": 0.6341945409774781, "epoch": 7, "memory": 44137, "step": 27242} +{"lr": 6.984661696627024e-05, "data_time": 0.0012496471405029296, "loss": 0.40110942125320437, "time": 0.6357108592987061, "epoch": 7, "memory": 44137, "step": 27342} +{"accuracy/top1": 74.8408203125, "data_time": 0.0356859749761121, "time": 1.2791170087353936, "step": 7} +{"lr": 6.884001985601584e-05, "data_time": 0.0016091585159301758, "loss": 0.3969531297683716, "time": 0.6348937511444092, "epoch": 8, "memory": 44137, "step": 27449} +{"lr": 6.790503918604319e-05, "data_time": 0.0014080047607421876, "loss": 0.40966375172138214, "time": 0.6334564924240113, "epoch": 8, "memory": 44137, "step": 27549} +{"lr": 6.697568975054603e-05, "data_time": 0.001700115203857422, "loss": 0.40814265608787537, "time": 0.6338228940963745, "epoch": 8, "memory": 44137, "step": 27649} +{"lr": 6.605203163778337e-05, "data_time": 0.001463627815246582, "loss": 0.40422230064868925, "time": 0.6338220834732056, "epoch": 8, "memory": 44137, "step": 27749} +{"lr": 6.513412456803489e-05, "data_time": 0.001416158676147461, "loss": 0.4123924046754837, "time": 0.6332894563674927, "epoch": 8, "memory": 44137, "step": 27849} +{"lr": 6.422202788973933e-05, "data_time": 0.0015094280242919922, "loss": 0.40509442389011385, "time": 0.6346164464950561, "epoch": 8, "memory": 44137, "step": 27949} +{"lr": 6.331580057565697e-05, "data_time": 0.0014674901962280274, "loss": 0.3988806664943695, "time": 0.6339321374893189, "epoch": 8, "memory": 44137, "step": 28049} +{"lr": 6.241550121905724e-05, "data_time": 0.0015756607055664063, "loss": 0.39616633057594297, "time": 0.6330334424972535, "epoch": 8, "memory": 44137, "step": 28149} +{"lr": 6.15211880299298e-05, "data_time": 0.0015095710754394532, "loss": 0.40544065833091736, "time": 0.6330933094024658, "epoch": 8, "memory": 44137, "step": 28249} +{"lr": 6.0632918831221007e-05, "data_time": 0.0014772415161132812, "loss": 0.4100262880325317, "time": 0.6333740234375, "epoch": 8, "memory": 44137, "step": 28349} +{"lr": 5.9750751055095654e-05, "data_time": 0.0015888452529907227, "loss": 0.3976229250431061, "time": 0.6334634780883789, "epoch": 8, "memory": 44137, "step": 28449} +{"lr": 5.887474173922336e-05, "data_time": 0.0014522314071655274, "loss": 0.40636836886405947, "time": 0.6329374074935913, "epoch": 8, "memory": 44137, "step": 28549} +{"lr": 5.80049475230907e-05, "data_time": 0.0016263484954833984, "loss": 0.4128390848636627, "time": 0.6331194162368774, "epoch": 8, "memory": 44137, "step": 28649} +{"lr": 5.7141424644339115e-05, "data_time": 0.0014470577239990234, "loss": 0.4046159446239471, "time": 0.6349116325378418, "epoch": 8, "memory": 44137, "step": 28749} +{"lr": 5.6284228935128753e-05, "data_time": 0.0015562772750854492, "loss": 0.3848160207271576, "time": 0.6335993528366088, "epoch": 8, "memory": 44137, "step": 28849} +{"lr": 5.543341581852875e-05, "data_time": 0.0016240596771240235, "loss": 0.3774960994720459, "time": 0.635904049873352, "epoch": 8, "memory": 44137, "step": 28949} +{"lr": 5.458904030493362e-05, "data_time": 0.001580357551574707, "loss": 0.4161092400550842, "time": 0.6341368436813355, "epoch": 8, "memory": 44137, "step": 29049} +{"lr": 5.3751156988506505e-05, "data_time": 0.0015173912048339843, "loss": 0.3929214417934418, "time": 0.6332409620285034, "epoch": 8, "memory": 44137, "step": 29149} +{"lr": 5.2919820043649535e-05, "data_time": 0.0014882087707519531, "loss": 0.4085879951715469, "time": 0.6345562934875488, "epoch": 8, "memory": 44137, "step": 29249} +{"lr": 5.209508322150057e-05, "data_time": 0.0015078306198120118, "loss": 0.4201653331518173, "time": 0.6334356307983399, "epoch": 8, "memory": 44137, "step": 29349} +{"lr": 5.127699984645856e-05, "data_time": 0.0016011476516723632, "loss": 0.40792628228664396, "time": 0.6332447052001953, "epoch": 8, "memory": 44137, "step": 29449} +{"lr": 5.046562281273526e-05, "data_time": 0.0014960527420043944, "loss": 0.3835161954164505, "time": 0.6331025123596191, "epoch": 8, "memory": 44137, "step": 29549} +{"lr": 4.966100458093549e-05, "data_time": 0.0014417409896850587, "loss": 0.39708831906318665, "time": 0.6330764293670654, "epoch": 8, "memory": 44137, "step": 29649} +{"lr": 4.886319717466545e-05, "data_time": 0.00142214298248291, "loss": 0.4182854235172272, "time": 0.6334197998046875, "epoch": 8, "memory": 44137, "step": 29749} +{"lr": 4.8072252177168467e-05, "data_time": 0.0013968229293823242, "loss": 0.4206300139427185, "time": 0.633488655090332, "epoch": 8, "memory": 44137, "step": 29849} +{"lr": 4.728822072799032e-05, "data_time": 0.0015152692794799805, "loss": 0.4408943563699722, "time": 0.6330051422119141, "epoch": 8, "memory": 44137, "step": 29949} +{"lr": 4.6511153519672786e-05, "data_time": 0.0015388727188110352, "loss": 0.4135405957698822, "time": 0.6336704015731811, "epoch": 8, "memory": 44137, "step": 30049} +{"lr": 4.57411007944756e-05, "data_time": 0.0014841794967651368, "loss": 0.39446223676204684, "time": 0.6341177225112915, "epoch": 8, "memory": 44137, "step": 30149} +{"lr": 4.49781123411285e-05, "data_time": 0.0015529632568359376, "loss": 0.4151736408472061, "time": 0.6330987930297851, "epoch": 8, "memory": 44137, "step": 30249} +{"lr": 4.422223749161159e-05, "data_time": 0.0015906572341918945, "loss": 0.39999428689479827, "time": 0.6345949649810791, "epoch": 8, "memory": 44137, "step": 30349} +{"lr": 4.3473525117965965e-05, "data_time": 0.0015187501907348634, "loss": 0.40094477832317355, "time": 0.6338727712631226, "epoch": 8, "memory": 44137, "step": 30449} +{"lr": 4.2732023629134074e-05, "data_time": 0.0014914751052856445, "loss": 0.43085960745811464, "time": 0.6340967893600464, "epoch": 8, "memory": 44137, "step": 30549} +{"lr": 4.199778096782924e-05, "data_time": 0.0014739990234375, "loss": 0.4026518613100052, "time": 0.6333942413330078, "epoch": 8, "memory": 44137, "step": 30649} +{"lr": 4.127084460743637e-05, "data_time": 0.0014615058898925781, "loss": 0.40305177569389344, "time": 0.6336512088775634, "epoch": 8, "memory": 44137, "step": 30749} +{"lr": 4.0551261548942e-05, "data_time": 0.001582169532775879, "loss": 0.44525461196899413, "time": 0.633476972579956, "epoch": 8, "memory": 44137, "step": 30849} +{"lr": 3.983907831789604e-05, "data_time": 0.001528453826904297, "loss": 0.3785897374153137, "time": 0.6356264591217041, "epoch": 8, "memory": 44137, "step": 30949} +{"lr": 3.913434096140289e-05, "data_time": 0.0015667915344238282, "loss": 0.3967933654785156, "time": 0.6338096618652344, "epoch": 8, "memory": 44137, "step": 31049} +{"lr": 3.8437095045144625e-05, "data_time": 0.0014075040817260742, "loss": 0.42655904293060304, "time": 0.6349545001983643, "epoch": 8, "memory": 44137, "step": 31149} +{"lr": 3.77473856504348e-05, "data_time": 0.001307225227355957, "loss": 0.3991405516862869, "time": 0.6340682744979859, "epoch": 8, "memory": 44137, "step": 31249} +{"accuracy/top1": 74.37806701660156, "data_time": 0.03741505228239914, "time": 1.2846401066615665, "step": 8} +{"lr": 3.701779344716437e-05, "data_time": 0.0016834020614624023, "loss": 0.40488806664943694, "time": 0.633327865600586, "epoch": 9, "memory": 44137, "step": 31356} +{"lr": 3.63438257918037e-05, "data_time": 0.0017286062240600586, "loss": 0.3990287363529205, "time": 0.634128737449646, "epoch": 9, "memory": 44137, "step": 31456} +{"lr": 3.567753000095364e-05, "data_time": 0.001622462272644043, "loss": 0.3980677485466003, "time": 0.642203402519226, "epoch": 9, "memory": 44137, "step": 31556} +{"lr": 3.501894915480852e-05, "data_time": 0.0014738798141479491, "loss": 0.3903617441654205, "time": 0.6338530302047729, "epoch": 9, "memory": 44137, "step": 31656} +{"lr": 3.4368125834743e-05, "data_time": 0.001567530632019043, "loss": 0.422028112411499, "time": 0.6336917877197266, "epoch": 9, "memory": 44137, "step": 31756} +{"lr": 3.372510212055925e-05, "data_time": 0.0014284133911132812, "loss": 0.3847354382276535, "time": 0.6334270238876343, "epoch": 9, "memory": 44137, "step": 31856} +{"lr": 3.308991958776599e-05, "data_time": 0.001447582244873047, "loss": 0.3819938778877258, "time": 0.634311580657959, "epoch": 9, "memory": 44137, "step": 31956} +{"lr": 3.246261930489061e-05, "data_time": 0.0013586997985839844, "loss": 0.39875280261039736, "time": 0.6348443746566772, "epoch": 9, "memory": 44137, "step": 32056} +{"lr": 3.18432418308236e-05, "data_time": 0.0014246463775634765, "loss": 0.4338852077722549, "time": 0.633664870262146, "epoch": 9, "memory": 44137, "step": 32156} +{"lr": 3.123182721219624e-05, "data_time": 0.0016513109207153321, "loss": 0.4046516209840775, "time": 0.6342431783676148, "epoch": 9, "memory": 44137, "step": 32256} +{"lr": 3.0628414980791353e-05, "data_time": 0.0015561342239379882, "loss": 0.39943632781505584, "time": 0.6335716962814331, "epoch": 9, "memory": 44137, "step": 32356} +{"lr": 3.0033044150987488e-05, "data_time": 0.0014463424682617187, "loss": 0.3937799632549286, "time": 0.6331947565078735, "epoch": 9, "memory": 44137, "step": 32456} +{"lr": 2.944575321723602e-05, "data_time": 0.0015320062637329101, "loss": 0.40703162252902986, "time": 0.6329815864562989, "epoch": 9, "memory": 44137, "step": 32556} +{"lr": 2.8866580151572607e-05, "data_time": 0.0014628887176513672, "loss": 0.40660122931003573, "time": 0.6340882301330566, "epoch": 9, "memory": 44137, "step": 32656} +{"lr": 2.829556240116184e-05, "data_time": 0.001535940170288086, "loss": 0.41109093725681306, "time": 0.6341081857681274, "epoch": 9, "memory": 44137, "step": 32756} +{"lr": 2.7732736885876114e-05, "data_time": 0.001671004295349121, "loss": 0.40753025114536284, "time": 0.6416688680648803, "epoch": 9, "memory": 44137, "step": 32856} +{"lr": 2.717813999590862e-05, "data_time": 0.0015972137451171875, "loss": 0.4007302701473236, "time": 0.6333588361740112, "epoch": 9, "memory": 44137, "step": 32956} +{"lr": 2.663180758942035e-05, "data_time": 0.0015102386474609374, "loss": 0.3754425525665283, "time": 0.6337812662124633, "epoch": 9, "memory": 44137, "step": 33056} +{"lr": 2.6093774990221658e-05, "data_time": 0.001367640495300293, "loss": 0.36535353064537046, "time": 0.6333444118499756, "epoch": 9, "memory": 44137, "step": 33156} +{"lr": 2.5564076985488515e-05, "data_time": 0.0015793561935424805, "loss": 0.407567223906517, "time": 0.6349753379821778, "epoch": 9, "memory": 44137, "step": 33256} +{"lr": 2.5042747823512982e-05, "data_time": 0.0013894081115722657, "loss": 0.4009361952543259, "time": 0.634540057182312, "epoch": 9, "memory": 44137, "step": 33356} +{"lr": 2.4529821211489152e-05, "data_time": 0.0014384031295776368, "loss": 0.4067427635192871, "time": 0.633469033241272, "epoch": 9, "memory": 44137, "step": 33456} +{"lr": 2.402533031333377e-05, "data_time": 0.00162813663482666, "loss": 0.4030266135931015, "time": 0.6335771560668946, "epoch": 9, "memory": 44137, "step": 33556} +{"lr": 2.352930774754175e-05, "data_time": 0.0015053749084472656, "loss": 0.38815886378288267, "time": 0.6335853099822998, "epoch": 9, "memory": 44137, "step": 33656} +{"lr": 2.3041785585077355e-05, "data_time": 0.001583099365234375, "loss": 0.3933963686227798, "time": 0.633979344367981, "epoch": 9, "memory": 44137, "step": 33756} +{"lr": 2.256279534730055e-05, "data_time": 0.001464986801147461, "loss": 0.4071730405092239, "time": 0.6359490156173706, "epoch": 9, "memory": 44137, "step": 33856} +{"lr": 2.2092368003928886e-05, "data_time": 0.0015979528427124024, "loss": 0.4104828476905823, "time": 0.6341641664505004, "epoch": 9, "memory": 44137, "step": 33956} +{"lr": 2.1630533971035236e-05, "data_time": 0.0014789342880249024, "loss": 0.42280719578266146, "time": 0.6341844797134399, "epoch": 9, "memory": 44137, "step": 34056} +{"lr": 2.117732310908114e-05, "data_time": 0.001540207862854004, "loss": 0.4198928356170654, "time": 0.6349170684814454, "epoch": 9, "memory": 44137, "step": 34156} +{"lr": 2.0732764720986212e-05, "data_time": 0.0014826297760009766, "loss": 0.37595857977867125, "time": 0.6355086326599121, "epoch": 9, "memory": 44137, "step": 34256} +{"lr": 2.0296887550233324e-05, "data_time": 0.0014709711074829101, "loss": 0.39905501902103424, "time": 0.6342411518096924, "epoch": 9, "memory": 44137, "step": 34356} +{"lr": 1.9869719779010417e-05, "data_time": 0.0014364480972290038, "loss": 0.4381367713212967, "time": 0.6351343154907226, "epoch": 9, "memory": 44137, "step": 34456} +{"lr": 1.9451289026388164e-05, "data_time": 0.0014481782913208009, "loss": 0.4117830693721771, "time": 0.6336531639099121, "epoch": 9, "memory": 44137, "step": 34556} +{"lr": 1.9041622346534353e-05, "data_time": 0.0016322374343872071, "loss": 0.4093270987272263, "time": 0.6347902536392211, "epoch": 9, "memory": 44137, "step": 34656} +{"lr": 1.8640746226964523e-05, "data_time": 0.0016631364822387695, "loss": 0.39976613223552704, "time": 0.6349977016448974, "epoch": 9, "memory": 44137, "step": 34756} +{"lr": 1.82486865868295e-05, "data_time": 0.001429438591003418, "loss": 0.39968317449092866, "time": 0.633928632736206, "epoch": 9, "memory": 44137, "step": 34856} +{"lr": 1.7865468775239498e-05, "data_time": 0.0014545679092407226, "loss": 0.4115006387233734, "time": 0.6340491056442261, "epoch": 9, "memory": 44137, "step": 34956} +{"lr": 1.7491117569625184e-05, "data_time": 0.0015875816345214844, "loss": 0.4287821531295776, "time": 0.6341647624969482, "epoch": 9, "memory": 44137, "step": 35056} +{"lr": 1.7125657174135686e-05, "data_time": 0.0013334035873413086, "loss": 0.43009800612926485, "time": 0.6355794429779053, "epoch": 9, "memory": 44137, "step": 35156} +{"accuracy/top1": 74.28760528564453, "data_time": 0.037636423933094944, "time": 1.2834950767714401, "step": 9} +{"lr": 1.674448744607319e-05, "data_time": 0.0018569469451904298, "loss": 0.3890314280986786, "time": 0.6338708400726318, "epoch": 10, "memory": 44137, "step": 35263} +{"lr": 1.639750545524004e-05, "data_time": 0.0014520883560180664, "loss": 0.4330062627792358, "time": 0.6343757152557373, "epoch": 10, "memory": 44137, "step": 35363} +{"lr": 1.6059484983401464e-05, "data_time": 0.0015063762664794921, "loss": 0.4040870815515518, "time": 0.6342190265655517, "epoch": 10, "memory": 44137, "step": 35463} +{"lr": 1.573044788569705e-05, "data_time": 0.001512908935546875, "loss": 0.4126666158437729, "time": 0.6340214729309082, "epoch": 10, "memory": 44137, "step": 35563} +{"lr": 1.5410415436435178e-05, "data_time": 0.0014666080474853515, "loss": 0.4082415699958801, "time": 0.6351520776748657, "epoch": 10, "memory": 44137, "step": 35663} +{"lr": 1.5099408327717273e-05, "data_time": 0.0015602350234985352, "loss": 0.4406146079301834, "time": 0.6344327211380005, "epoch": 10, "memory": 44137, "step": 35763} +{"lr": 1.479744666809991e-05, "data_time": 0.001470041275024414, "loss": 0.40127586126327514, "time": 0.6339004278182984, "epoch": 10, "memory": 44137, "step": 35863} +{"lr": 1.4504549981294865e-05, "data_time": 0.0015267610549926757, "loss": 0.4200449585914612, "time": 0.6339508295059204, "epoch": 10, "memory": 44137, "step": 35963} +{"lr": 1.4220737204906605e-05, "data_time": 0.001703190803527832, "loss": 0.41975703835487366, "time": 0.6334076881408691, "epoch": 10, "memory": 44137, "step": 36063} +{"lr": 1.3946026689207934e-05, "data_time": 0.0014288663864135743, "loss": 0.3851330608129501, "time": 0.6342244148254395, "epoch": 10, "memory": 44137, "step": 36163} +{"lr": 1.3680436195953514e-05, "data_time": 0.0015428781509399414, "loss": 0.4089700669050217, "time": 0.6338906526565552, "epoch": 10, "memory": 44137, "step": 36263} +{"lr": 1.3423982897231485e-05, "data_time": 0.0014763355255126953, "loss": 0.4391102522611618, "time": 0.6338380575180054, "epoch": 10, "memory": 44137, "step": 36363} +{"lr": 1.3176683374353149e-05, "data_time": 0.0017591476440429687, "loss": 0.4089411228895187, "time": 0.6361191511154175, "epoch": 10, "memory": 44137, "step": 36463} +{"lr": 1.2938553616780832e-05, "data_time": 0.0014814138412475586, "loss": 0.4454068303108215, "time": 0.6342217445373535, "epoch": 10, "memory": 44137, "step": 36563} +{"lr": 1.2709609021094214e-05, "data_time": 0.0015864133834838866, "loss": 0.4099941462278366, "time": 0.6345898389816285, "epoch": 10, "memory": 44137, "step": 36663} +{"lr": 1.2489864389994715e-05, "data_time": 0.0015474081039428711, "loss": 0.39916797280311583, "time": 0.6339625358581543, "epoch": 10, "memory": 44137, "step": 36763} +{"lr": 1.2279333931348431e-05, "data_time": 0.001513504981994629, "loss": 0.4181778907775879, "time": 0.6436614990234375, "epoch": 10, "memory": 44137, "step": 36863} +{"lr": 1.2078031257267545e-05, "data_time": 0.0015755414962768555, "loss": 0.3995991885662079, "time": 0.6347186803817749, "epoch": 10, "memory": 44137, "step": 36963} +{"lr": 1.1885969383230204e-05, "data_time": 0.0017694473266601563, "loss": 0.40407503247261045, "time": 0.6784493446350097, "epoch": 10, "memory": 44137, "step": 37063} +{"lr": 1.1703160727238979e-05, "data_time": 0.0014449834823608398, "loss": 0.38404877483844757, "time": 0.6347826004028321, "epoch": 10, "memory": 44137, "step": 37163} +{"lr": 1.1529617109017937e-05, "data_time": 0.0014947175979614258, "loss": 0.38951668739318845, "time": 0.6342282295227051, "epoch": 10, "memory": 44137, "step": 37263} +{"lr": 1.136534974924845e-05, "data_time": 0.0014008045196533202, "loss": 0.3978728473186493, "time": 0.6371277332305908, "epoch": 10, "memory": 44137, "step": 37363} +{"lr": 1.1210369268843738e-05, "data_time": 0.0014733076095581055, "loss": 0.40388007164001466, "time": 0.637355637550354, "epoch": 10, "memory": 44137, "step": 37463} +{"lr": 1.1064685688262156e-05, "data_time": 0.0015849828720092773, "loss": 0.39653285443782804, "time": 0.6343758821487426, "epoch": 10, "memory": 44137, "step": 37563} +{"lr": 1.0928308426859188e-05, "data_time": 0.0018506050109863281, "loss": 0.4068183064460754, "time": 0.6338836669921875, "epoch": 10, "memory": 44137, "step": 37663} +{"lr": 1.0801246302278603e-05, "data_time": 0.0015090465545654296, "loss": 0.42445961833000184, "time": 0.6338107109069824, "epoch": 10, "memory": 44137, "step": 37763} +{"lr": 1.0683507529882236e-05, "data_time": 0.0015589237213134766, "loss": 0.4159298360347748, "time": 0.6341955423355102, "epoch": 10, "memory": 44137, "step": 37863} +{"lr": 1.057509972221878e-05, "data_time": 0.0014363765716552735, "loss": 0.4106313705444336, "time": 0.6336933612823487, "epoch": 10, "memory": 44137, "step": 37963} +{"lr": 1.047602988853174e-05, "data_time": 0.0015500783920288086, "loss": 0.42624981999397277, "time": 0.6341934680938721, "epoch": 10, "memory": 44137, "step": 38063} +{"lr": 1.0386304434306083e-05, "data_time": 0.0014511346817016602, "loss": 0.3967807799577713, "time": 0.6350108861923218, "epoch": 10, "memory": 44137, "step": 38163} +{"lr": 1.030592916085413e-05, "data_time": 0.0020376920700073244, "loss": 0.42003758549690245, "time": 0.633349871635437, "epoch": 10, "memory": 44137, "step": 38263} +{"lr": 1.0234909264940509e-05, "data_time": 0.0015140056610107421, "loss": 0.4236251085996628, "time": 0.633553409576416, "epoch": 10, "memory": 44137, "step": 38363} +{"lr": 1.0173249338446117e-05, "data_time": 0.0014904499053955077, "loss": 0.4105237126350403, "time": 0.6346558570861817, "epoch": 10, "memory": 44137, "step": 38463} +{"lr": 1.0120953368071232e-05, "data_time": 0.0014421701431274413, "loss": 0.41743309795856476, "time": 0.6333788871765137, "epoch": 10, "memory": 44137, "step": 38563} +{"lr": 1.0078024735077709e-05, "data_time": 0.0014368295669555664, "loss": 0.39933337569236754, "time": 0.6333008289337159, "epoch": 10, "memory": 44137, "step": 38663} +{"lr": 1.004446621507047e-05, "data_time": 0.0014926433563232423, "loss": 0.38928772509098053, "time": 0.6334861278533935, "epoch": 10, "memory": 44137, "step": 38763} +{"lr": 1.0020279977817876e-05, "data_time": 0.0014059782028198243, "loss": 0.4124821960926056, "time": 0.6341943740844727, "epoch": 10, "memory": 44137, "step": 38863} +{"lr": 1.0005467587111622e-05, "data_time": 0.001574540138244629, "loss": 0.41150707602500913, "time": 0.6334386587142944, "epoch": 10, "memory": 44137, "step": 38963} +{"lr": 1.0000030000665491e-05, "data_time": 0.0013365745544433594, "loss": 0.40277799069881437, "time": 0.6339604377746582, "epoch": 10, "memory": 44137, "step": 39063} +{"accuracy/top1": 74.29108428955078, "data_time": 0.03611139593453243, "time": 1.2821046525034412, "step": 10} diff --git a/clip_large_pretrain_4x256_sdv2_lr3e-4/20230606_052113/vis_data/vis_image/0000000.png_1.png b/clip_large_pretrain_4x256_sdv2_lr3e-4/20230606_052113/vis_data/vis_image/0000000.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..6f3a1e910dca6ccdc91e38540c7e1c84dfd762ae Binary files /dev/null and b/clip_large_pretrain_4x256_sdv2_lr3e-4/20230606_052113/vis_data/vis_image/0000000.png_1.png differ diff --git a/clip_large_pretrain_4x256_sdv2_lr3e-4/20230606_052113/vis_data/vis_image/0000000.png_10.png b/clip_large_pretrain_4x256_sdv2_lr3e-4/20230606_052113/vis_data/vis_image/0000000.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..e1475971452d5b9634dc61100ce54de75e95c24a Binary files /dev/null and b/clip_large_pretrain_4x256_sdv2_lr3e-4/20230606_052113/vis_data/vis_image/0000000.png_10.png differ diff --git a/clip_large_pretrain_4x256_sdv2_lr3e-4/20230606_052113/vis_data/vis_image/0000000.png_2.png b/clip_large_pretrain_4x256_sdv2_lr3e-4/20230606_052113/vis_data/vis_image/0000000.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..2dec5199eaf5d8566bfa22080d9bcfdc9bc5924b Binary files /dev/null and b/clip_large_pretrain_4x256_sdv2_lr3e-4/20230606_052113/vis_data/vis_image/0000000.png_2.png differ diff --git a/clip_large_pretrain_4x256_sdv2_lr3e-4/20230606_052113/vis_data/vis_image/0000000.png_3.png b/clip_large_pretrain_4x256_sdv2_lr3e-4/20230606_052113/vis_data/vis_image/0000000.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..8666a7f2e9e31ba20fcc61699f6a72c2f7140234 Binary files /dev/null and b/clip_large_pretrain_4x256_sdv2_lr3e-4/20230606_052113/vis_data/vis_image/0000000.png_3.png differ diff --git a/clip_large_pretrain_4x256_sdv2_lr3e-4/20230606_052113/vis_data/vis_image/0000000.png_4.png b/clip_large_pretrain_4x256_sdv2_lr3e-4/20230606_052113/vis_data/vis_image/0000000.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..b2e52db585df8e01728761ef6a28e222bea6905b Binary files /dev/null and b/clip_large_pretrain_4x256_sdv2_lr3e-4/20230606_052113/vis_data/vis_image/0000000.png_4.png differ diff --git a/clip_large_pretrain_4x256_sdv2_lr3e-4/20230606_052113/vis_data/vis_image/0000000.png_5.png b/clip_large_pretrain_4x256_sdv2_lr3e-4/20230606_052113/vis_data/vis_image/0000000.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..685bdb9fb9f779c801a2766a18e74a3166072802 Binary files /dev/null and b/clip_large_pretrain_4x256_sdv2_lr3e-4/20230606_052113/vis_data/vis_image/0000000.png_5.png differ diff --git a/clip_large_pretrain_4x256_sdv2_lr3e-4/20230606_052113/vis_data/vis_image/0000000.png_6.png b/clip_large_pretrain_4x256_sdv2_lr3e-4/20230606_052113/vis_data/vis_image/0000000.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..d3597011a4d1a3a4394e62513ae226cd27831f97 Binary files /dev/null and b/clip_large_pretrain_4x256_sdv2_lr3e-4/20230606_052113/vis_data/vis_image/0000000.png_6.png differ diff --git a/clip_large_pretrain_4x256_sdv2_lr3e-4/20230606_052113/vis_data/vis_image/0000000.png_7.png b/clip_large_pretrain_4x256_sdv2_lr3e-4/20230606_052113/vis_data/vis_image/0000000.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..aabd7b936f199fc397080d6982eab1ed4629049d Binary files /dev/null and b/clip_large_pretrain_4x256_sdv2_lr3e-4/20230606_052113/vis_data/vis_image/0000000.png_7.png differ diff --git a/clip_large_pretrain_4x256_sdv2_lr3e-4/20230606_052113/vis_data/vis_image/0000000.png_8.png b/clip_large_pretrain_4x256_sdv2_lr3e-4/20230606_052113/vis_data/vis_image/0000000.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..0034085f4dca30dd5fcc90414b29f7bf8c6288b7 Binary files /dev/null and b/clip_large_pretrain_4x256_sdv2_lr3e-4/20230606_052113/vis_data/vis_image/0000000.png_8.png differ diff --git a/clip_large_pretrain_4x256_sdv2_lr3e-4/20230606_052113/vis_data/vis_image/0000000.png_9.png b/clip_large_pretrain_4x256_sdv2_lr3e-4/20230606_052113/vis_data/vis_image/0000000.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..f2d634d6acfb66660711504c37a872da79b4c731 Binary files /dev/null and b/clip_large_pretrain_4x256_sdv2_lr3e-4/20230606_052113/vis_data/vis_image/0000000.png_9.png differ diff --git a/clip_large_pretrain_4x256_sdv2_lr3e-4/20230606_052113/vis_data/vis_image/000005174.jpg_1.png b/clip_large_pretrain_4x256_sdv2_lr3e-4/20230606_052113/vis_data/vis_image/000005174.jpg_1.png new file mode 100644 index 0000000000000000000000000000000000000000..0e437193b9ae8b65223204955c09977ab16ce234 Binary files /dev/null and b/clip_large_pretrain_4x256_sdv2_lr3e-4/20230606_052113/vis_data/vis_image/000005174.jpg_1.png differ diff --git a/clip_large_pretrain_4x256_sdv2_lr3e-4/20230606_052113/vis_data/vis_image/000005174.jpg_10.png b/clip_large_pretrain_4x256_sdv2_lr3e-4/20230606_052113/vis_data/vis_image/000005174.jpg_10.png new file mode 100644 index 0000000000000000000000000000000000000000..f4915f40582f370cad68cb499ca456b09a462679 Binary files /dev/null and b/clip_large_pretrain_4x256_sdv2_lr3e-4/20230606_052113/vis_data/vis_image/000005174.jpg_10.png differ diff --git a/clip_large_pretrain_4x256_sdv2_lr3e-4/20230606_052113/vis_data/vis_image/000005174.jpg_2.png b/clip_large_pretrain_4x256_sdv2_lr3e-4/20230606_052113/vis_data/vis_image/000005174.jpg_2.png new file mode 100644 index 0000000000000000000000000000000000000000..09fcfad6894422a64edf3622945599948c4edee0 Binary files /dev/null and b/clip_large_pretrain_4x256_sdv2_lr3e-4/20230606_052113/vis_data/vis_image/000005174.jpg_2.png differ diff --git a/clip_large_pretrain_4x256_sdv2_lr3e-4/20230606_052113/vis_data/vis_image/000005174.jpg_3.png b/clip_large_pretrain_4x256_sdv2_lr3e-4/20230606_052113/vis_data/vis_image/000005174.jpg_3.png new file mode 100644 index 0000000000000000000000000000000000000000..d29e999e04b513248a4dfd619c0be617e4ad3afc Binary files /dev/null and b/clip_large_pretrain_4x256_sdv2_lr3e-4/20230606_052113/vis_data/vis_image/000005174.jpg_3.png differ diff --git a/clip_large_pretrain_4x256_sdv2_lr3e-4/20230606_052113/vis_data/vis_image/000005174.jpg_4.png b/clip_large_pretrain_4x256_sdv2_lr3e-4/20230606_052113/vis_data/vis_image/000005174.jpg_4.png new file mode 100644 index 0000000000000000000000000000000000000000..32c35e985d04a921ff590df9ac325f7664441828 Binary files /dev/null and b/clip_large_pretrain_4x256_sdv2_lr3e-4/20230606_052113/vis_data/vis_image/000005174.jpg_4.png differ diff --git a/clip_large_pretrain_4x256_sdv2_lr3e-4/20230606_052113/vis_data/vis_image/000005174.jpg_5.png b/clip_large_pretrain_4x256_sdv2_lr3e-4/20230606_052113/vis_data/vis_image/000005174.jpg_5.png new file mode 100644 index 0000000000000000000000000000000000000000..dde8ae47f907c5a6cf171a707bf4fc8c22594f0b Binary files /dev/null and b/clip_large_pretrain_4x256_sdv2_lr3e-4/20230606_052113/vis_data/vis_image/000005174.jpg_5.png differ diff --git a/clip_large_pretrain_4x256_sdv2_lr3e-4/20230606_052113/vis_data/vis_image/000005174.jpg_6.png b/clip_large_pretrain_4x256_sdv2_lr3e-4/20230606_052113/vis_data/vis_image/000005174.jpg_6.png new file mode 100644 index 0000000000000000000000000000000000000000..ec08f43c412637774b8cfa5ce04483631db88b72 Binary files /dev/null and b/clip_large_pretrain_4x256_sdv2_lr3e-4/20230606_052113/vis_data/vis_image/000005174.jpg_6.png differ diff --git a/clip_large_pretrain_4x256_sdv2_lr3e-4/20230606_052113/vis_data/vis_image/000005174.jpg_7.png b/clip_large_pretrain_4x256_sdv2_lr3e-4/20230606_052113/vis_data/vis_image/000005174.jpg_7.png new file mode 100644 index 0000000000000000000000000000000000000000..dea56bcb9d0d9668c5cb7252349f2cea7d03e13f Binary files /dev/null and b/clip_large_pretrain_4x256_sdv2_lr3e-4/20230606_052113/vis_data/vis_image/000005174.jpg_7.png differ diff --git a/clip_large_pretrain_4x256_sdv2_lr3e-4/20230606_052113/vis_data/vis_image/000005174.jpg_8.png b/clip_large_pretrain_4x256_sdv2_lr3e-4/20230606_052113/vis_data/vis_image/000005174.jpg_8.png new file mode 100644 index 0000000000000000000000000000000000000000..e6f5210bea68aefd186772f6916fb0c25f43b0f6 Binary files /dev/null and b/clip_large_pretrain_4x256_sdv2_lr3e-4/20230606_052113/vis_data/vis_image/000005174.jpg_8.png differ diff --git a/clip_large_pretrain_4x256_sdv2_lr3e-4/20230606_052113/vis_data/vis_image/000005174.jpg_9.png b/clip_large_pretrain_4x256_sdv2_lr3e-4/20230606_052113/vis_data/vis_image/000005174.jpg_9.png new file mode 100644 index 0000000000000000000000000000000000000000..c27c5b841d35b7fcdc77ce3033d4ecdb37f0e2dc Binary files /dev/null and b/clip_large_pretrain_4x256_sdv2_lr3e-4/20230606_052113/vis_data/vis_image/000005174.jpg_9.png differ diff --git a/clip_large_pretrain_4x256_sdv2_lr3e-4/clip_large_pretrain_4x256_sdv2_lr3e-4.py b/clip_large_pretrain_4x256_sdv2_lr3e-4/clip_large_pretrain_4x256_sdv2_lr3e-4.py new file mode 100644 index 0000000000000000000000000000000000000000..2fe3d2db91303588ef319725b2e1e94dca9ecb05 --- /dev/null +++ b/clip_large_pretrain_4x256_sdv2_lr3e-4/clip_large_pretrain_4x256_sdv2_lr3e-4.py @@ -0,0 +1,236 @@ +optim_wrapper = dict( + optimizer=dict( + type='AdamW', lr=0.0003, weight_decay=0.3, _scope_='mmpretrain'), + paramwise_cfg=dict( + custom_keys=dict({ + '.cls_token': dict(decay_mult=0.0), + '.pos_embed': dict(decay_mult=0.0) + })), + type='AmpOptimWrapper', + dtype='bfloat16', + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=False, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=4096) +model = dict( + type='ImageClassifier', + backbone=dict( + frozen_stages=24, + type='VisionTransformer', + arch='l', + img_size=224, + patch_size=14, + drop_rate=0.1, + pre_norm=True, + final_norm=False, + init_cfg=dict( + type='Pretrained', + checkpoint='ckpt/openclip-ViT-L-14.pth', + prefix='backbone')), + neck=dict( + type='CLIPProjection', + in_channels=1024, + out_channels=768, + init_cfg=dict( + type='Pretrained', + checkpoint='ckpt/openclip-ViT-L-14.pth', + prefix='backbone')), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=768, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + init_cfg=None), + init_cfg=dict( + type='TruncNormal', layer=['Conv2d', 'Linear'], std=0.02, bias=0.0), + train_cfg=None) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=128, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stablediffusionV2-1-dpmsolver-25-1m.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/cc1m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=128, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV2-1-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=128, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV2-1-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/clip_large_pretrain_4x256_sdv2_lr3e-4' diff --git a/clip_large_pretrain_4x256_sdv2_lr3e-4/epoch_1.pth b/clip_large_pretrain_4x256_sdv2_lr3e-4/epoch_1.pth new file mode 100644 index 0000000000000000000000000000000000000000..79b973d82cf9990eab11c79837fdecba6a2a9cb4 --- /dev/null +++ b/clip_large_pretrain_4x256_sdv2_lr3e-4/epoch_1.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:4baaa90b788028132a219100b7f60bddc3fb95231ffb23609c9e77ef93d49527 +size 2438616705 diff --git a/clip_large_pretrain_4x256_sdv2_lr3e-4/epoch_10.pth b/clip_large_pretrain_4x256_sdv2_lr3e-4/epoch_10.pth new file mode 100644 index 0000000000000000000000000000000000000000..9720244e210c501aa31638dbe04535d8f402d7c7 --- /dev/null +++ b/clip_large_pretrain_4x256_sdv2_lr3e-4/epoch_10.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:d86a38d5d4761e114f209fe3ece4afc946ce677e9db5d8a17e472ee26df863e3 +size 2441174209 diff --git a/clip_large_pretrain_4x256_sdv2_lr3e-4/epoch_2.pth b/clip_large_pretrain_4x256_sdv2_lr3e-4/epoch_2.pth new file mode 100644 index 0000000000000000000000000000000000000000..29906f7f05281c836d5b33d1ba7e283abc19ad9c --- /dev/null +++ b/clip_large_pretrain_4x256_sdv2_lr3e-4/epoch_2.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:e54a4222fe23a910294b38eb8e15c0a4021495b908964966a7f70dcc4a98b54b +size 2438901249 diff --git a/clip_large_pretrain_4x256_sdv2_lr3e-4/epoch_3.pth b/clip_large_pretrain_4x256_sdv2_lr3e-4/epoch_3.pth new file mode 100644 index 0000000000000000000000000000000000000000..1aa164cee25e88968bb7f28bf703a1fa91a01bf4 --- /dev/null +++ b/clip_large_pretrain_4x256_sdv2_lr3e-4/epoch_3.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:8fff02a0592c0375f69c1626d426c17fe218b31826df9ebd7bf594a3ae7c04b0 +size 2439184641 diff --git a/clip_large_pretrain_4x256_sdv2_lr3e-4/epoch_4.pth b/clip_large_pretrain_4x256_sdv2_lr3e-4/epoch_4.pth new file mode 100644 index 0000000000000000000000000000000000000000..67c9c49f65321203d42978799b9e0f33b9502fac --- /dev/null +++ b/clip_large_pretrain_4x256_sdv2_lr3e-4/epoch_4.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:507f0c6e2b8b03a4de6903c8b9c925d964cfb74b80d042c5285521cb6090ecac +size 2439467969 diff --git a/clip_large_pretrain_4x256_sdv2_lr3e-4/epoch_5.pth b/clip_large_pretrain_4x256_sdv2_lr3e-4/epoch_5.pth new file mode 100644 index 0000000000000000000000000000000000000000..5a799df1cfb8e00476fd10c55d0c5162ea698e26 --- /dev/null +++ b/clip_large_pretrain_4x256_sdv2_lr3e-4/epoch_5.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:5cd9ebb5527eb4483d318bf4e3ff5106d68d04a52bfc98d4a2b5ae4cc3403694 +size 2439751233 diff --git a/clip_large_pretrain_4x256_sdv2_lr3e-4/epoch_6.pth b/clip_large_pretrain_4x256_sdv2_lr3e-4/epoch_6.pth new file mode 100644 index 0000000000000000000000000000000000000000..83ed1815479be05e9bec1fdd6ebc497a078acb4b --- /dev/null +++ b/clip_large_pretrain_4x256_sdv2_lr3e-4/epoch_6.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:dca4b99a3951d562bcfa29947fa592281c430325e741294b76ed534a9c63ad33 +size 2440034305 diff --git a/clip_large_pretrain_4x256_sdv2_lr3e-4/epoch_7.pth b/clip_large_pretrain_4x256_sdv2_lr3e-4/epoch_7.pth new file mode 100644 index 0000000000000000000000000000000000000000..3ad9c8dcdf75ea694ba68df29d4e50d425dcb078 --- /dev/null +++ b/clip_large_pretrain_4x256_sdv2_lr3e-4/epoch_7.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:88037ddaa311c8e8bccd95626cf5a3036f16a91727e8a0eb9d42aa2fda34e922 +size 2440317825 diff --git a/clip_large_pretrain_4x256_sdv2_lr3e-4/epoch_8.pth b/clip_large_pretrain_4x256_sdv2_lr3e-4/epoch_8.pth new file mode 100644 index 0000000000000000000000000000000000000000..61491241dee358500e79eb17f1452e5df7c386c5 --- /dev/null +++ b/clip_large_pretrain_4x256_sdv2_lr3e-4/epoch_8.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:d712faca1f88136931fc69d52e7b8f33c6f0b522f4c3acdd99a5d833b0ab5c82 +size 2440601089 diff --git a/clip_large_pretrain_4x256_sdv2_lr3e-4/epoch_9.pth b/clip_large_pretrain_4x256_sdv2_lr3e-4/epoch_9.pth new file mode 100644 index 0000000000000000000000000000000000000000..ed790be7439a77a07cfb39cf0367158ec72ff503 --- /dev/null +++ b/clip_large_pretrain_4x256_sdv2_lr3e-4/epoch_9.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:07bfe6f9c523c5dd041871bfa30f9206ba13aace54e408be1557b4a00d8568d5 +size 2440887105 diff --git a/clip_large_pretrain_4x256_sdv2_lr3e-4/last_checkpoint b/clip_large_pretrain_4x256_sdv2_lr3e-4/last_checkpoint new file mode 100644 index 0000000000000000000000000000000000000000..7e29eaac7a9c9e647dae46517f5d98f865adb4d5 --- /dev/null +++ b/clip_large_pretrain_4x256_sdv2_lr3e-4/last_checkpoint @@ -0,0 +1 @@ +/mnt/petrelfs/luzeyu/workspace/fakebench/mmpretrain/workdir/clip_large_pretrain_4x256_sdv2_lr3e-4/epoch_10.pth \ No newline at end of file diff --git a/clip_large_pretrain_4x256_sdv2_lr3e-5/20230606_050006/20230606_050006.log b/clip_large_pretrain_4x256_sdv2_lr3e-5/20230606_050006/20230606_050006.log new file mode 100644 index 0000000000000000000000000000000000000000..56efb32a3144d97a3bf6e5ba5ae155a739c4e11f --- /dev/null +++ b/clip_large_pretrain_4x256_sdv2_lr3e-5/20230606_050006/20230606_050006.log @@ -0,0 +1,280 @@ +2023/06/06 05:00:15 - mmengine - INFO - +------------------------------------------------------------ +System environment: + sys.platform: linux + Python: 3.10.9 (main, Mar 8 2023, 10:47:38) [GCC 11.2.0] + CUDA available: True + numpy_random_seed: 943994155 + GPU 0,1,2,3: NVIDIA A100-SXM4-80GB + CUDA_HOME: /mnt/petrelfs/share/cuda-11.6 + NVCC: Cuda compilation tools, release 11.6, V11.6.124 + GCC: gcc (GCC) 7.5.0 + PyTorch: 1.13.1 + PyTorch compiling details: PyTorch built with: + - GCC 9.3 + - C++ Version: 201402 + - Intel(R) oneAPI Math Kernel Library Version 2021.4-Product Build 20210904 for Intel(R) 64 architecture applications + - Intel(R) MKL-DNN v2.6.0 (Git Hash 52b5f107dd9cf10910aaa19cb47f3abf9b349815) + - OpenMP 201511 (a.k.a. OpenMP 4.5) + - LAPACK is enabled (usually provided by MKL) + - NNPACK is enabled + - CPU capability usage: AVX2 + - CUDA Runtime 11.6 + - NVCC architecture flags: -gencode;arch=compute_37,code=sm_37;-gencode;arch=compute_50,code=sm_50;-gencode;arch=compute_60,code=sm_60;-gencode;arch=compute_61,code=sm_61;-gencode;arch=compute_70,code=sm_70;-gencode;arch=compute_75,code=sm_75;-gencode;arch=compute_80,code=sm_80;-gencode;arch=compute_86,code=sm_86;-gencode;arch=compute_37,code=compute_37 + - CuDNN 8.3.2 (built against CUDA 11.5) + - Magma 2.6.1 + - Build settings: BLAS_INFO=mkl, BUILD_TYPE=Release, CUDA_VERSION=11.6, CUDNN_VERSION=8.3.2, CXX_COMPILER=/opt/rh/devtoolset-9/root/usr/bin/c++, CXX_FLAGS= -fabi-version=11 -Wno-deprecated -fvisibility-inlines-hidden -DUSE_PTHREADPOOL -fopenmp -DNDEBUG -DUSE_KINETO -DUSE_FBGEMM -DUSE_QNNPACK -DUSE_PYTORCH_QNNPACK -DUSE_XNNPACK -DSYMBOLICATE_MOBILE_DEBUG_HANDLE -DEDGE_PROFILER_USE_KINETO -O2 -fPIC -Wno-narrowing -Wall -Wextra -Werror=return-type -Werror=non-virtual-dtor -Wno-missing-field-initializers -Wno-type-limits -Wno-array-bounds -Wno-unknown-pragmas -Wunused-local-typedefs -Wno-unused-parameter -Wno-unused-function -Wno-unused-result -Wno-strict-overflow -Wno-strict-aliasing -Wno-error=deprecated-declarations -Wno-stringop-overflow -Wno-psabi -Wno-error=pedantic -Wno-error=redundant-decls -Wno-error=old-style-cast -fdiagnostics-color=always -faligned-new -Wno-unused-but-set-variable -Wno-maybe-uninitialized -fno-math-errno -fno-trapping-math -Werror=format -Werror=cast-function-type -Wno-stringop-overflow, LAPACK_INFO=mkl, PERF_WITH_AVX=1, PERF_WITH_AVX2=1, PERF_WITH_AVX512=1, TORCH_VERSION=1.13.1, USE_CUDA=ON, USE_CUDNN=ON, USE_EXCEPTION_PTR=1, USE_GFLAGS=OFF, USE_GLOG=OFF, USE_MKL=ON, USE_MKLDNN=ON, USE_MPI=OFF, USE_NCCL=ON, USE_NNPACK=ON, USE_OPENMP=ON, USE_ROCM=OFF, + + TorchVision: 0.14.1 + OpenCV: 4.7.0 + MMEngine: 0.7.3 + +Runtime environment: + cudnn_benchmark: True + mp_cfg: {'mp_start_method': 'fork', 'opencv_num_threads': 0} + dist_cfg: {'backend': 'nccl'} + seed: None + deterministic: False + Distributed launcher: slurm + Distributed training: True + GPU number: 4 +------------------------------------------------------------ + +2023/06/06 05:00:19 - mmengine - INFO - Config: +optim_wrapper = dict( + optimizer=dict( + type='AdamW', lr=3e-05, weight_decay=0.3, _scope_='mmpretrain'), + paramwise_cfg=dict( + custom_keys=dict({ + '.cls_token': dict(decay_mult=0.0), + '.pos_embed': dict(decay_mult=0.0) + })), + type='AmpOptimWrapper', + dtype='bfloat16', + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=False, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=4096) +model = dict( + type='ImageClassifier', + backbone=dict( + frozen_stages=24, + type='VisionTransformer', + arch='l', + img_size=224, + patch_size=14, + drop_rate=0.1, + pre_norm=True, + final_norm=False, + init_cfg=dict( + type='Pretrained', + checkpoint='ckpt/openclip-ViT-L-14.pth', + prefix='backbone')), + neck=dict( + type='CLIPProjection', + in_channels=1024, + out_channels=768, + init_cfg=dict( + type='Pretrained', + checkpoint='ckpt/openclip-ViT-L-14.pth', + prefix='backbone')), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=768, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + init_cfg=None), + init_cfg=dict( + type='TruncNormal', layer=['Conv2d', 'Linear'], std=0.02, bias=0.0), + train_cfg=None) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=128, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stablediffusionV2-1-dpmsolver-25-1m.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/cc1m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=128, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV2-1-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=128, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV2-1-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/clip_large_pretrain_4x256_sdv2_lr3e-5' + diff --git a/clip_large_pretrain_4x256_sdv2_lr3e-5/20230606_050006/vis_data/config.py b/clip_large_pretrain_4x256_sdv2_lr3e-5/20230606_050006/vis_data/config.py new file mode 100644 index 0000000000000000000000000000000000000000..e69de29bb2d1d6434b8b29ae775ad8c2e48c5391 diff --git a/clip_large_pretrain_4x256_sdv2_lr3e-5/20230606_052112/20230606_052112.log b/clip_large_pretrain_4x256_sdv2_lr3e-5/20230606_052112/20230606_052112.log new file mode 100644 index 0000000000000000000000000000000000000000..9e3543970fec180d2645d6d7c7632def8a310106 --- /dev/null +++ b/clip_large_pretrain_4x256_sdv2_lr3e-5/20230606_052112/20230606_052112.log @@ -0,0 +1,1727 @@ +2023/06/06 05:21:18 - mmengine - INFO - +------------------------------------------------------------ +System environment: + sys.platform: linux + Python: 3.10.9 (main, Mar 8 2023, 10:47:38) [GCC 11.2.0] + CUDA available: True + numpy_random_seed: 127034622 + GPU 0,1,2,3: NVIDIA A100-SXM4-80GB + CUDA_HOME: /mnt/petrelfs/share/cuda-11.6 + NVCC: Cuda compilation tools, release 11.6, V11.6.124 + GCC: gcc (GCC) 7.5.0 + PyTorch: 1.13.1 + PyTorch compiling details: PyTorch built with: + - GCC 9.3 + - C++ Version: 201402 + - Intel(R) oneAPI Math Kernel Library Version 2021.4-Product Build 20210904 for Intel(R) 64 architecture applications + - Intel(R) MKL-DNN v2.6.0 (Git Hash 52b5f107dd9cf10910aaa19cb47f3abf9b349815) + - OpenMP 201511 (a.k.a. OpenMP 4.5) + - LAPACK is enabled (usually provided by MKL) + - NNPACK is enabled + - CPU capability usage: AVX2 + - CUDA Runtime 11.6 + - NVCC architecture flags: -gencode;arch=compute_37,code=sm_37;-gencode;arch=compute_50,code=sm_50;-gencode;arch=compute_60,code=sm_60;-gencode;arch=compute_61,code=sm_61;-gencode;arch=compute_70,code=sm_70;-gencode;arch=compute_75,code=sm_75;-gencode;arch=compute_80,code=sm_80;-gencode;arch=compute_86,code=sm_86;-gencode;arch=compute_37,code=compute_37 + - CuDNN 8.3.2 (built against CUDA 11.5) + - Magma 2.6.1 + - Build settings: BLAS_INFO=mkl, BUILD_TYPE=Release, CUDA_VERSION=11.6, CUDNN_VERSION=8.3.2, CXX_COMPILER=/opt/rh/devtoolset-9/root/usr/bin/c++, CXX_FLAGS= -fabi-version=11 -Wno-deprecated -fvisibility-inlines-hidden -DUSE_PTHREADPOOL -fopenmp -DNDEBUG -DUSE_KINETO -DUSE_FBGEMM -DUSE_QNNPACK -DUSE_PYTORCH_QNNPACK -DUSE_XNNPACK -DSYMBOLICATE_MOBILE_DEBUG_HANDLE -DEDGE_PROFILER_USE_KINETO -O2 -fPIC -Wno-narrowing -Wall -Wextra -Werror=return-type -Werror=non-virtual-dtor -Wno-missing-field-initializers -Wno-type-limits -Wno-array-bounds -Wno-unknown-pragmas -Wunused-local-typedefs -Wno-unused-parameter -Wno-unused-function -Wno-unused-result -Wno-strict-overflow -Wno-strict-aliasing -Wno-error=deprecated-declarations -Wno-stringop-overflow -Wno-psabi -Wno-error=pedantic -Wno-error=redundant-decls -Wno-error=old-style-cast -fdiagnostics-color=always -faligned-new -Wno-unused-but-set-variable -Wno-maybe-uninitialized -fno-math-errno -fno-trapping-math -Werror=format -Werror=cast-function-type -Wno-stringop-overflow, LAPACK_INFO=mkl, PERF_WITH_AVX=1, PERF_WITH_AVX2=1, PERF_WITH_AVX512=1, TORCH_VERSION=1.13.1, USE_CUDA=ON, USE_CUDNN=ON, USE_EXCEPTION_PTR=1, USE_GFLAGS=OFF, USE_GLOG=OFF, USE_MKL=ON, USE_MKLDNN=ON, USE_MPI=OFF, USE_NCCL=ON, USE_NNPACK=ON, USE_OPENMP=ON, USE_ROCM=OFF, + + TorchVision: 0.14.1 + OpenCV: 4.7.0 + MMEngine: 0.7.3 + +Runtime environment: + cudnn_benchmark: True + mp_cfg: {'mp_start_method': 'fork', 'opencv_num_threads': 0} + dist_cfg: {'backend': 'nccl'} + seed: None + deterministic: False + Distributed launcher: slurm + Distributed training: True + GPU number: 4 +------------------------------------------------------------ + +2023/06/06 05:21:22 - mmengine - INFO - Config: +optim_wrapper = dict( + optimizer=dict( + type='AdamW', lr=3e-05, weight_decay=0.3, _scope_='mmpretrain'), + paramwise_cfg=dict( + custom_keys=dict({ + '.cls_token': dict(decay_mult=0.0), + '.pos_embed': dict(decay_mult=0.0) + })), + type='AmpOptimWrapper', + dtype='bfloat16', + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=False, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=4096) +model = dict( + type='ImageClassifier', + backbone=dict( + frozen_stages=24, + type='VisionTransformer', + arch='l', + img_size=224, + patch_size=14, + drop_rate=0.1, + pre_norm=True, + final_norm=False, + init_cfg=dict( + type='Pretrained', + checkpoint='ckpt/openclip-ViT-L-14.pth', + prefix='backbone')), + neck=dict( + type='CLIPProjection', + in_channels=1024, + out_channels=768, + init_cfg=dict( + type='Pretrained', + checkpoint='ckpt/openclip-ViT-L-14.pth', + prefix='backbone')), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=768, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + init_cfg=None), + init_cfg=dict( + type='TruncNormal', layer=['Conv2d', 'Linear'], std=0.02, bias=0.0), + train_cfg=None) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=128, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stablediffusionV2-1-dpmsolver-25-1m.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/cc1m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=128, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV2-1-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=128, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV2-1-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/clip_large_pretrain_4x256_sdv2_lr3e-5' + +2023/06/06 05:21:36 - mmengine - INFO - Hooks will be executed in the following order: +before_run: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_load_checkpoint: +(ABOVE_NORMAL) EMAHook + -------------------- +before_train: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_train_epoch: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(NORMAL ) DistSamplerSeedHook + -------------------- +before_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook + -------------------- +after_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +after_train_epoch: +(NORMAL ) IterTimerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_val_epoch: +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook + -------------------- +before_val_iter: +(NORMAL ) IterTimerHook + -------------------- +after_val_iter: +(NORMAL ) IterTimerHook +(NORMAL ) VisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_val_epoch: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_save_checkpoint: +(ABOVE_NORMAL) EMAHook + -------------------- +after_train: +(VERY_LOW ) CheckpointHook + -------------------- +before_test_epoch: +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook + -------------------- +before_test_iter: +(NORMAL ) IterTimerHook + -------------------- +after_test_iter: +(NORMAL ) IterTimerHook +(NORMAL ) VisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_test_epoch: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_run: +(BELOW_NORMAL) LoggerHook + -------------------- +2023/06/06 05:21:54 - mmengine - INFO - load backbone in model from: ckpt/openclip-ViT-L-14.pth +2023/06/06 05:21:59 - mmengine - WARNING - The model and loaded state dict do not match exactly + +unexpected key in source state_dict: ln1.weight, ln1.bias + +2023/06/06 05:21:59 - mmengine - INFO - load backbone in model from: ckpt/openclip-ViT-L-14.pth +2023/06/06 05:22:01 - mmengine - WARNING - The model and loaded state dict do not match exactly + +unexpected key in source state_dict: cls_token, pos_embed, patch_embed.projection.weight, pre_norm.weight, pre_norm.bias, layers.0.ln1.weight, layers.0.ln1.bias, layers.0.attn.qkv.weight, layers.0.attn.qkv.bias, layers.0.attn.proj.weight, layers.0.attn.proj.bias, layers.0.ln2.weight, layers.0.ln2.bias, layers.0.ffn.layers.0.0.weight, layers.0.ffn.layers.0.0.bias, layers.0.ffn.layers.1.weight, layers.0.ffn.layers.1.bias, layers.1.ln1.weight, layers.1.ln1.bias, layers.1.attn.qkv.weight, layers.1.attn.qkv.bias, layers.1.attn.proj.weight, layers.1.attn.proj.bias, layers.1.ln2.weight, layers.1.ln2.bias, layers.1.ffn.layers.0.0.weight, layers.1.ffn.layers.0.0.bias, layers.1.ffn.layers.1.weight, layers.1.ffn.layers.1.bias, layers.2.ln1.weight, layers.2.ln1.bias, layers.2.attn.qkv.weight, layers.2.attn.qkv.bias, layers.2.attn.proj.weight, layers.2.attn.proj.bias, layers.2.ln2.weight, layers.2.ln2.bias, layers.2.ffn.layers.0.0.weight, layers.2.ffn.layers.0.0.bias, layers.2.ffn.layers.1.weight, layers.2.ffn.layers.1.bias, layers.3.ln1.weight, layers.3.ln1.bias, layers.3.attn.qkv.weight, layers.3.attn.qkv.bias, layers.3.attn.proj.weight, layers.3.attn.proj.bias, layers.3.ln2.weight, layers.3.ln2.bias, layers.3.ffn.layers.0.0.weight, layers.3.ffn.layers.0.0.bias, layers.3.ffn.layers.1.weight, layers.3.ffn.layers.1.bias, layers.4.ln1.weight, layers.4.ln1.bias, layers.4.attn.qkv.weight, layers.4.attn.qkv.bias, layers.4.attn.proj.weight, layers.4.attn.proj.bias, layers.4.ln2.weight, layers.4.ln2.bias, layers.4.ffn.layers.0.0.weight, layers.4.ffn.layers.0.0.bias, layers.4.ffn.layers.1.weight, layers.4.ffn.layers.1.bias, layers.5.ln1.weight, layers.5.ln1.bias, layers.5.attn.qkv.weight, layers.5.attn.qkv.bias, layers.5.attn.proj.weight, layers.5.attn.proj.bias, layers.5.ln2.weight, layers.5.ln2.bias, layers.5.ffn.layers.0.0.weight, layers.5.ffn.layers.0.0.bias, layers.5.ffn.layers.1.weight, layers.5.ffn.layers.1.bias, layers.6.ln1.weight, layers.6.ln1.bias, layers.6.attn.qkv.weight, layers.6.attn.qkv.bias, layers.6.attn.proj.weight, layers.6.attn.proj.bias, layers.6.ln2.weight, layers.6.ln2.bias, layers.6.ffn.layers.0.0.weight, layers.6.ffn.layers.0.0.bias, layers.6.ffn.layers.1.weight, layers.6.ffn.layers.1.bias, layers.7.ln1.weight, layers.7.ln1.bias, layers.7.attn.qkv.weight, layers.7.attn.qkv.bias, layers.7.attn.proj.weight, layers.7.attn.proj.bias, layers.7.ln2.weight, layers.7.ln2.bias, layers.7.ffn.layers.0.0.weight, layers.7.ffn.layers.0.0.bias, layers.7.ffn.layers.1.weight, layers.7.ffn.layers.1.bias, layers.8.ln1.weight, layers.8.ln1.bias, layers.8.attn.qkv.weight, layers.8.attn.qkv.bias, layers.8.attn.proj.weight, layers.8.attn.proj.bias, layers.8.ln2.weight, layers.8.ln2.bias, layers.8.ffn.layers.0.0.weight, layers.8.ffn.layers.0.0.bias, layers.8.ffn.layers.1.weight, layers.8.ffn.layers.1.bias, layers.9.ln1.weight, layers.9.ln1.bias, layers.9.attn.qkv.weight, layers.9.attn.qkv.bias, layers.9.attn.proj.weight, layers.9.attn.proj.bias, layers.9.ln2.weight, layers.9.ln2.bias, layers.9.ffn.layers.0.0.weight, layers.9.ffn.layers.0.0.bias, layers.9.ffn.layers.1.weight, layers.9.ffn.layers.1.bias, layers.10.ln1.weight, layers.10.ln1.bias, layers.10.attn.qkv.weight, layers.10.attn.qkv.bias, layers.10.attn.proj.weight, layers.10.attn.proj.bias, layers.10.ln2.weight, layers.10.ln2.bias, layers.10.ffn.layers.0.0.weight, layers.10.ffn.layers.0.0.bias, layers.10.ffn.layers.1.weight, layers.10.ffn.layers.1.bias, layers.11.ln1.weight, layers.11.ln1.bias, layers.11.attn.qkv.weight, layers.11.attn.qkv.bias, layers.11.attn.proj.weight, layers.11.attn.proj.bias, layers.11.ln2.weight, layers.11.ln2.bias, layers.11.ffn.layers.0.0.weight, layers.11.ffn.layers.0.0.bias, layers.11.ffn.layers.1.weight, layers.11.ffn.layers.1.bias, layers.12.ln1.weight, layers.12.ln1.bias, layers.12.attn.qkv.weight, layers.12.attn.qkv.bias, layers.12.attn.proj.weight, layers.12.attn.proj.bias, layers.12.ln2.weight, layers.12.ln2.bias, layers.12.ffn.layers.0.0.weight, layers.12.ffn.layers.0.0.bias, layers.12.ffn.layers.1.weight, layers.12.ffn.layers.1.bias, layers.13.ln1.weight, layers.13.ln1.bias, layers.13.attn.qkv.weight, layers.13.attn.qkv.bias, layers.13.attn.proj.weight, layers.13.attn.proj.bias, layers.13.ln2.weight, layers.13.ln2.bias, layers.13.ffn.layers.0.0.weight, layers.13.ffn.layers.0.0.bias, layers.13.ffn.layers.1.weight, layers.13.ffn.layers.1.bias, layers.14.ln1.weight, layers.14.ln1.bias, layers.14.attn.qkv.weight, layers.14.attn.qkv.bias, layers.14.attn.proj.weight, layers.14.attn.proj.bias, layers.14.ln2.weight, layers.14.ln2.bias, layers.14.ffn.layers.0.0.weight, layers.14.ffn.layers.0.0.bias, layers.14.ffn.layers.1.weight, layers.14.ffn.layers.1.bias, layers.15.ln1.weight, layers.15.ln1.bias, layers.15.attn.qkv.weight, layers.15.attn.qkv.bias, layers.15.attn.proj.weight, layers.15.attn.proj.bias, layers.15.ln2.weight, layers.15.ln2.bias, layers.15.ffn.layers.0.0.weight, layers.15.ffn.layers.0.0.bias, layers.15.ffn.layers.1.weight, layers.15.ffn.layers.1.bias, layers.16.ln1.weight, layers.16.ln1.bias, layers.16.attn.qkv.weight, layers.16.attn.qkv.bias, layers.16.attn.proj.weight, layers.16.attn.proj.bias, layers.16.ln2.weight, layers.16.ln2.bias, layers.16.ffn.layers.0.0.weight, layers.16.ffn.layers.0.0.bias, layers.16.ffn.layers.1.weight, layers.16.ffn.layers.1.bias, layers.17.ln1.weight, layers.17.ln1.bias, layers.17.attn.qkv.weight, layers.17.attn.qkv.bias, layers.17.attn.proj.weight, layers.17.attn.proj.bias, layers.17.ln2.weight, layers.17.ln2.bias, layers.17.ffn.layers.0.0.weight, layers.17.ffn.layers.0.0.bias, layers.17.ffn.layers.1.weight, layers.17.ffn.layers.1.bias, layers.18.ln1.weight, layers.18.ln1.bias, layers.18.attn.qkv.weight, layers.18.attn.qkv.bias, layers.18.attn.proj.weight, layers.18.attn.proj.bias, layers.18.ln2.weight, layers.18.ln2.bias, layers.18.ffn.layers.0.0.weight, layers.18.ffn.layers.0.0.bias, layers.18.ffn.layers.1.weight, layers.18.ffn.layers.1.bias, layers.19.ln1.weight, layers.19.ln1.bias, layers.19.attn.qkv.weight, layers.19.attn.qkv.bias, layers.19.attn.proj.weight, layers.19.attn.proj.bias, layers.19.ln2.weight, layers.19.ln2.bias, layers.19.ffn.layers.0.0.weight, layers.19.ffn.layers.0.0.bias, layers.19.ffn.layers.1.weight, layers.19.ffn.layers.1.bias, layers.20.ln1.weight, layers.20.ln1.bias, layers.20.attn.qkv.weight, layers.20.attn.qkv.bias, layers.20.attn.proj.weight, layers.20.attn.proj.bias, layers.20.ln2.weight, layers.20.ln2.bias, layers.20.ffn.layers.0.0.weight, layers.20.ffn.layers.0.0.bias, layers.20.ffn.layers.1.weight, layers.20.ffn.layers.1.bias, layers.21.ln1.weight, layers.21.ln1.bias, layers.21.attn.qkv.weight, layers.21.attn.qkv.bias, layers.21.attn.proj.weight, layers.21.attn.proj.bias, layers.21.ln2.weight, layers.21.ln2.bias, layers.21.ffn.layers.0.0.weight, layers.21.ffn.layers.0.0.bias, layers.21.ffn.layers.1.weight, layers.21.ffn.layers.1.bias, layers.22.ln1.weight, layers.22.ln1.bias, layers.22.attn.qkv.weight, layers.22.attn.qkv.bias, layers.22.attn.proj.weight, layers.22.attn.proj.bias, layers.22.ln2.weight, layers.22.ln2.bias, layers.22.ffn.layers.0.0.weight, layers.22.ffn.layers.0.0.bias, layers.22.ffn.layers.1.weight, layers.22.ffn.layers.1.bias, layers.23.ln1.weight, layers.23.ln1.bias, layers.23.attn.qkv.weight, layers.23.attn.qkv.bias, layers.23.attn.proj.weight, layers.23.attn.proj.bias, layers.23.ln2.weight, layers.23.ln2.bias, layers.23.ffn.layers.0.0.weight, layers.23.ffn.layers.0.0.bias, layers.23.ffn.layers.1.weight, layers.23.ffn.layers.1.bias, ln1.weight, ln1.bias + +missing keys in source state_dict: proj + +Name of parameter - Initialization information + +backbone.cls_token - torch.Size([1, 1, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.pos_embed - torch.Size([1, 257, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.patch_embed.projection.weight - torch.Size([1024, 3, 14, 14]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.0.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.0.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.0.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.0.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.0.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.0.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.0.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.0.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.0.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.0.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.0.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.0.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.1.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.1.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.1.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.1.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.1.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.1.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.1.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.1.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.1.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.1.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.1.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.1.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.2.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.2.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.2.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.2.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.2.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.2.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.2.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.2.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.2.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.2.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.2.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.2.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.3.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.3.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.3.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.3.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.3.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.3.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.3.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.3.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.3.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.3.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.3.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.3.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.4.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.4.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.4.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.4.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.4.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.4.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.4.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.4.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.4.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.4.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.4.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.4.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.5.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.5.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.5.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.5.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.5.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.5.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.5.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.5.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.5.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.5.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.5.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.5.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.6.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.6.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.6.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.6.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.6.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.6.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.6.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.6.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.6.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.6.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.6.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.6.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.7.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.7.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.7.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.7.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.7.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.7.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.7.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.7.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.7.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.7.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.7.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.7.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.8.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.8.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.8.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.8.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.8.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.8.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.8.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.8.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.8.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.8.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.8.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.8.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.9.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.9.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.9.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.9.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.9.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.9.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.9.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.9.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.9.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.9.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.9.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.9.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.10.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.10.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.10.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.10.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.10.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.10.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.10.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.10.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.10.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.10.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.10.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.10.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.11.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.11.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.11.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.11.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.11.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.11.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.11.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.11.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.11.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.11.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.11.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.11.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.12.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.12.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.12.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.12.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.12.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.12.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.12.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.12.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.12.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.12.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.12.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.12.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.13.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.13.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.13.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.13.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.13.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.13.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.13.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.13.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.13.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.13.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.13.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.13.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.14.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.14.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.14.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.14.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.14.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.14.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.14.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.14.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.14.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.14.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.14.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.14.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.15.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.15.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.15.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.15.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.15.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.15.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.15.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.15.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.15.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.15.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.15.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.15.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.16.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.16.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.16.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.16.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.16.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.16.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.16.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.16.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.16.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.16.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.16.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.16.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.17.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.17.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.17.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.17.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.17.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.17.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.17.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.17.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.17.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.17.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.17.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.17.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.18.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.18.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.18.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.18.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.18.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.18.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.18.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.18.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.18.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.18.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.18.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.18.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.19.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.19.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.19.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.19.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.19.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.19.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.19.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.19.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.19.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.19.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.19.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.19.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.20.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.20.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.20.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.20.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.20.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.20.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.20.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.20.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.20.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.20.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.20.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.20.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.21.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.21.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.21.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.21.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.21.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.21.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.21.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.21.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.21.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.21.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.21.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.21.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.22.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.22.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.22.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.22.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.22.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.22.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.22.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.22.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.22.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.22.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.22.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.22.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.23.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.23.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.23.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.23.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.23.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.23.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.23.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.23.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.23.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.23.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.23.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.23.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.pre_norm.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.pre_norm.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +neck.proj - torch.Size([1024, 768]): +The value is the same before and after calling `init_weights` of ImageClassifier + +head.fc.weight - torch.Size([2, 768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +head.fc.bias - torch.Size([2]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 +2023/06/06 05:22:01 - mmengine - WARNING - "FileClient" will be deprecated in future. Please use io functions in https://mmengine.readthedocs.io/en/latest/api/fileio.html#file-io +2023/06/06 05:22:01 - mmengine - WARNING - "HardDiskBackend" is the alias of "LocalBackend" and the former will be deprecated in future. +2023/06/06 05:22:01 - mmengine - INFO - Checkpoints will be saved to /mnt/petrelfs/luzeyu/workspace/fakebench/mmpretrain/workdir/clip_large_pretrain_4x256_sdv2_lr3e-5. +2023/06/06 05:23:13 - mmengine - INFO - Epoch(train) [1][ 100/3907] lr: 3.0000e-05 eta: 7:44:55 time: 0.6327 data_time: 0.0017 memory: 44139 loss: 0.5821 +2023/06/06 05:24:17 - mmengine - INFO - Epoch(train) [1][ 200/3907] lr: 2.9999e-05 eta: 7:17:41 time: 0.6465 data_time: 0.0017 memory: 44139 loss: 0.5244 +2023/06/06 05:25:20 - mmengine - INFO - Epoch(train) [1][ 300/3907] lr: 2.9997e-05 eta: 7:07:48 time: 0.6347 data_time: 0.0015 memory: 44139 loss: 0.5025 +2023/06/06 05:26:24 - mmengine - INFO - Epoch(train) [1][ 400/3907] lr: 2.9995e-05 eta: 7:02:32 time: 0.6352 data_time: 0.0016 memory: 44139 loss: 0.5020 +2023/06/06 05:27:27 - mmengine - INFO - Epoch(train) [1][ 500/3907] lr: 2.9992e-05 eta: 6:58:51 time: 0.6358 data_time: 0.0015 memory: 44139 loss: 0.4893 +2023/06/06 05:28:31 - mmengine - INFO - Epoch(train) [1][ 600/3907] lr: 2.9988e-05 eta: 6:56:11 time: 0.6359 data_time: 0.0015 memory: 44139 loss: 0.4940 +2023/06/06 05:29:35 - mmengine - INFO - Epoch(train) [1][ 700/3907] lr: 2.9984e-05 eta: 6:53:59 time: 0.6356 data_time: 0.0015 memory: 44139 loss: 0.4516 +2023/06/06 05:30:38 - mmengine - INFO - Epoch(train) [1][ 800/3907] lr: 2.9979e-05 eta: 6:51:58 time: 0.6353 data_time: 0.0016 memory: 44139 loss: 0.4653 +2023/06/06 05:31:42 - mmengine - INFO - Epoch(train) [1][ 900/3907] lr: 2.9974e-05 eta: 6:50:15 time: 0.6363 data_time: 0.0016 memory: 44139 loss: 0.4562 +2023/06/06 05:32:46 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr3e-5_20230606_052112 +2023/06/06 05:32:46 - mmengine - INFO - Epoch(train) [1][1000/3907] lr: 2.9968e-05 eta: 6:48:41 time: 0.6352 data_time: 0.0015 memory: 44139 loss: 0.4434 +2023/06/06 05:33:49 - mmengine - INFO - Epoch(train) [1][1100/3907] lr: 2.9961e-05 eta: 6:47:07 time: 0.6362 data_time: 0.0014 memory: 44139 loss: 0.4577 +2023/06/06 05:36:13 - mmengine - INFO - Epoch(train) [1][1200/3907] lr: 2.9954e-05 eta: 7:27:56 time: 0.6349 data_time: 0.0017 memory: 44139 loss: 0.4558 +2023/06/06 05:37:17 - mmengine - INFO - Epoch(train) [1][1300/3907] lr: 2.9945e-05 eta: 7:23:10 time: 0.6344 data_time: 0.0015 memory: 44139 loss: 0.4663 +2023/06/06 05:38:20 - mmengine - INFO - Epoch(train) [1][1400/3907] lr: 2.9937e-05 eta: 7:18:57 time: 0.6346 data_time: 0.0016 memory: 44139 loss: 0.4483 +2023/06/06 05:39:24 - mmengine - INFO - Epoch(train) [1][1500/3907] lr: 2.9927e-05 eta: 7:15:07 time: 0.6364 data_time: 0.0015 memory: 44139 loss: 0.4409 +2023/06/06 05:40:27 - mmengine - INFO - Epoch(train) [1][1600/3907] lr: 2.9917e-05 eta: 7:11:39 time: 0.6369 data_time: 0.0015 memory: 44139 loss: 0.4377 +2023/06/06 05:41:31 - mmengine - INFO - Epoch(train) [1][1700/3907] lr: 2.9907e-05 eta: 7:08:29 time: 0.6341 data_time: 0.0014 memory: 44139 loss: 0.4680 +2023/06/06 05:42:35 - mmengine - INFO - Epoch(train) [1][1800/3907] lr: 2.9896e-05 eta: 7:05:32 time: 0.6356 data_time: 0.0015 memory: 44139 loss: 0.4360 +2023/06/06 05:43:38 - mmengine - INFO - Epoch(train) [1][1900/3907] lr: 2.9884e-05 eta: 7:02:46 time: 0.6354 data_time: 0.0016 memory: 44139 loss: 0.4360 +2023/06/06 05:44:42 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr3e-5_20230606_052112 +2023/06/06 05:44:42 - mmengine - INFO - Epoch(train) [1][2000/3907] lr: 2.9871e-05 eta: 7:00:11 time: 0.6347 data_time: 0.0015 memory: 44139 loss: 0.4763 +2023/06/06 05:45:45 - mmengine - INFO - Epoch(train) [1][2100/3907] lr: 2.9858e-05 eta: 6:57:45 time: 0.6356 data_time: 0.0014 memory: 44139 loss: 0.4316 +2023/06/06 05:46:49 - mmengine - INFO - Epoch(train) [1][2200/3907] lr: 2.9844e-05 eta: 6:55:29 time: 0.6371 data_time: 0.0015 memory: 44139 loss: 0.4738 +2023/06/06 05:47:53 - mmengine - INFO - Epoch(train) [1][2300/3907] lr: 2.9830e-05 eta: 6:53:18 time: 0.6363 data_time: 0.0016 memory: 44139 loss: 0.4565 +2023/06/06 05:48:56 - mmengine - INFO - Epoch(train) [1][2400/3907] lr: 2.9815e-05 eta: 6:51:13 time: 0.6416 data_time: 0.0015 memory: 44139 loss: 0.4229 +2023/06/06 05:50:00 - mmengine - INFO - Epoch(train) [1][2500/3907] lr: 2.9799e-05 eta: 6:49:12 time: 0.6352 data_time: 0.0014 memory: 44139 loss: 0.4524 +2023/06/06 05:51:04 - mmengine - INFO - Epoch(train) [1][2600/3907] lr: 2.9782e-05 eta: 6:47:16 time: 0.6344 data_time: 0.0015 memory: 44139 loss: 0.4300 +2023/06/06 05:52:07 - mmengine - INFO - Epoch(train) [1][2700/3907] lr: 2.9765e-05 eta: 6:45:22 time: 0.6349 data_time: 0.0015 memory: 44139 loss: 0.4235 +2023/06/06 05:53:11 - mmengine - INFO - Epoch(train) [1][2800/3907] lr: 2.9748e-05 eta: 6:43:32 time: 0.6348 data_time: 0.0014 memory: 44139 loss: 0.4433 +2023/06/06 05:54:14 - mmengine - INFO - Epoch(train) [1][2900/3907] lr: 2.9730e-05 eta: 6:41:45 time: 0.6349 data_time: 0.0014 memory: 44139 loss: 0.4557 +2023/06/06 05:55:21 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr3e-5_20230606_052112 +2023/06/06 05:55:21 - mmengine - INFO - Epoch(train) [1][3000/3907] lr: 2.9711e-05 eta: 6:40:41 time: 0.6355 data_time: 0.0015 memory: 44139 loss: 0.4246 +2023/06/06 05:56:25 - mmengine - INFO - Epoch(train) [1][3100/3907] lr: 2.9691e-05 eta: 6:38:59 time: 0.6353 data_time: 0.0015 memory: 44139 loss: 0.4596 +2023/06/06 05:57:28 - mmengine - INFO - Epoch(train) [1][3200/3907] lr: 2.9671e-05 eta: 6:37:19 time: 0.6355 data_time: 0.0016 memory: 44139 loss: 0.4564 +2023/06/06 05:58:32 - mmengine - INFO - Epoch(train) [1][3300/3907] lr: 2.9650e-05 eta: 6:35:42 time: 0.6347 data_time: 0.0015 memory: 44139 loss: 0.4362 +2023/06/06 05:59:35 - mmengine - INFO - Epoch(train) [1][3400/3907] lr: 2.9629e-05 eta: 6:34:06 time: 0.6359 data_time: 0.0017 memory: 44139 loss: 0.4211 +2023/06/06 06:00:39 - mmengine - INFO - Epoch(train) [1][3500/3907] lr: 2.9607e-05 eta: 6:32:32 time: 0.6354 data_time: 0.0015 memory: 44139 loss: 0.4355 +2023/06/06 06:01:43 - mmengine - INFO - Epoch(train) [1][3600/3907] lr: 2.9584e-05 eta: 6:31:00 time: 0.6348 data_time: 0.0016 memory: 44139 loss: 0.4473 +2023/06/06 06:02:46 - mmengine - INFO - Epoch(train) [1][3700/3907] lr: 2.9561e-05 eta: 6:29:29 time: 0.6365 data_time: 0.0014 memory: 44139 loss: 0.4363 +2023/06/06 06:03:50 - mmengine - INFO - Epoch(train) [1][3800/3907] lr: 2.9537e-05 eta: 6:28:00 time: 0.6355 data_time: 0.0016 memory: 44139 loss: 0.4333 +2023/06/06 06:04:53 - mmengine - INFO - Epoch(train) [1][3900/3907] lr: 2.9513e-05 eta: 6:26:32 time: 0.6347 data_time: 0.0012 memory: 44139 loss: 0.4194 +2023/06/06 06:04:57 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr3e-5_20230606_052112 +2023/06/06 06:04:57 - mmengine - INFO - Saving checkpoint at 1 epochs +2023/06/06 06:06:40 - mmengine - INFO - Epoch(val) [1][57/57] accuracy/top1: 80.0738 single-label/precision_classwise: [78.82466888427734, 82.03089904785156] single-label/recall_classwise: [87.29878234863281, 71.20155334472656] single-label/f1-score_classwise: [82.84559631347656, 76.23355865478516] data_time: 0.0479 time: 1.3421 +2023/06/06 06:07:42 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr3e-5_20230606_052112 +2023/06/06 06:07:47 - mmengine - INFO - Epoch(train) [2][ 100/3907] lr: 2.9486e-05 eta: 6:25:22 time: 0.6368 data_time: 0.0016 memory: 44139 loss: 0.4614 +2023/06/06 06:08:50 - mmengine - INFO - Epoch(train) [2][ 200/3907] lr: 2.9460e-05 eta: 6:23:57 time: 0.6372 data_time: 0.0015 memory: 44138 loss: 0.4183 +2023/06/06 06:09:54 - mmengine - INFO - Epoch(train) [2][ 300/3907] lr: 2.9434e-05 eta: 6:22:32 time: 0.6383 data_time: 0.0014 memory: 44138 loss: 0.4502 +2023/06/06 06:10:58 - mmengine - INFO - Epoch(train) [2][ 400/3907] lr: 2.9407e-05 eta: 6:21:09 time: 0.6352 data_time: 0.0014 memory: 44138 loss: 0.4530 +2023/06/06 06:12:01 - mmengine - INFO - Epoch(train) [2][ 500/3907] lr: 2.9379e-05 eta: 6:19:46 time: 0.6358 data_time: 0.0015 memory: 44138 loss: 0.4171 +2023/06/06 06:13:05 - mmengine - INFO - Epoch(train) [2][ 600/3907] lr: 2.9351e-05 eta: 6:18:24 time: 0.6353 data_time: 0.0014 memory: 44138 loss: 0.4244 +2023/06/06 06:14:09 - mmengine - INFO - Epoch(train) [2][ 700/3907] lr: 2.9322e-05 eta: 6:17:02 time: 0.6356 data_time: 0.0015 memory: 44138 loss: 0.4053 +2023/06/06 06:15:12 - mmengine - INFO - Epoch(train) [2][ 800/3907] lr: 2.9293e-05 eta: 6:15:41 time: 0.6364 data_time: 0.0014 memory: 44138 loss: 0.4170 +2023/06/06 06:16:16 - mmengine - INFO - Epoch(train) [2][ 900/3907] lr: 2.9263e-05 eta: 6:14:23 time: 0.6358 data_time: 0.0014 memory: 44138 loss: 0.4180 +2023/06/06 06:17:20 - mmengine - INFO - Epoch(train) [2][1000/3907] lr: 2.9232e-05 eta: 6:13:05 time: 0.6371 data_time: 0.0016 memory: 44138 loss: 0.4058 +2023/06/06 06:18:19 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr3e-5_20230606_052112 +2023/06/06 06:18:23 - mmengine - INFO - Epoch(train) [2][1100/3907] lr: 2.9201e-05 eta: 6:11:47 time: 0.6362 data_time: 0.0017 memory: 44138 loss: 0.4151 +2023/06/06 06:19:27 - mmengine - INFO - Epoch(train) [2][1200/3907] lr: 2.9169e-05 eta: 6:10:30 time: 0.6365 data_time: 0.0015 memory: 44138 loss: 0.4330 +2023/06/06 06:20:31 - mmengine - INFO - Epoch(train) [2][1300/3907] lr: 2.9137e-05 eta: 6:09:14 time: 0.6362 data_time: 0.0014 memory: 44138 loss: 0.4559 +2023/06/06 06:21:35 - mmengine - INFO - Epoch(train) [2][1400/3907] lr: 2.9104e-05 eta: 6:07:58 time: 0.6361 data_time: 0.0015 memory: 44138 loss: 0.4159 +2023/06/06 06:22:38 - mmengine - INFO - Epoch(train) [2][1500/3907] lr: 2.9070e-05 eta: 6:06:41 time: 0.6374 data_time: 0.0014 memory: 44138 loss: 0.4259 +2023/06/06 06:23:42 - mmengine - INFO - Epoch(train) [2][1600/3907] lr: 2.9036e-05 eta: 6:05:25 time: 0.6361 data_time: 0.0015 memory: 44138 loss: 0.4157 +2023/06/06 06:24:46 - mmengine - INFO - Epoch(train) [2][1700/3907] lr: 2.9001e-05 eta: 6:04:10 time: 0.6379 data_time: 0.0015 memory: 44138 loss: 0.4216 +2023/06/06 06:25:49 - mmengine - INFO - Epoch(train) [2][1800/3907] lr: 2.8966e-05 eta: 6:02:55 time: 0.6361 data_time: 0.0014 memory: 44138 loss: 0.4137 +2023/06/06 06:26:53 - mmengine - INFO - Epoch(train) [2][1900/3907] lr: 2.8930e-05 eta: 6:01:40 time: 0.6360 data_time: 0.0014 memory: 44138 loss: 0.4191 +2023/06/06 06:27:56 - mmengine - INFO - Epoch(train) [2][2000/3907] lr: 2.8893e-05 eta: 6:00:26 time: 0.6363 data_time: 0.0016 memory: 44138 loss: 0.4243 +2023/06/06 06:28:56 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr3e-5_20230606_052112 +2023/06/06 06:29:00 - mmengine - INFO - Epoch(train) [2][2100/3907] lr: 2.8856e-05 eta: 5:59:13 time: 0.6360 data_time: 0.0016 memory: 44138 loss: 0.4397 +2023/06/06 06:30:04 - mmengine - INFO - Epoch(train) [2][2200/3907] lr: 2.8819e-05 eta: 5:57:59 time: 0.6349 data_time: 0.0014 memory: 44138 loss: 0.3885 +2023/06/06 06:31:08 - mmengine - INFO - Epoch(train) [2][2300/3907] lr: 2.8781e-05 eta: 5:56:47 time: 0.6370 data_time: 0.0014 memory: 44138 loss: 0.4367 +2023/06/06 06:32:11 - mmengine - INFO - Epoch(train) [2][2400/3907] lr: 2.8742e-05 eta: 5:55:34 time: 0.6358 data_time: 0.0014 memory: 44138 loss: 0.4364 +2023/06/06 06:33:15 - mmengine - INFO - Epoch(train) [2][2500/3907] lr: 2.8702e-05 eta: 5:54:21 time: 0.6351 data_time: 0.0015 memory: 44138 loss: 0.4202 +2023/06/06 06:34:19 - mmengine - INFO - Epoch(train) [2][2600/3907] lr: 2.8663e-05 eta: 5:53:08 time: 0.6363 data_time: 0.0015 memory: 44138 loss: 0.4343 +2023/06/06 06:35:22 - mmengine - INFO - Epoch(train) [2][2700/3907] lr: 2.8622e-05 eta: 5:51:56 time: 0.6358 data_time: 0.0014 memory: 44138 loss: 0.4557 +2023/06/06 06:36:26 - mmengine - INFO - Epoch(train) [2][2800/3907] lr: 2.8581e-05 eta: 5:50:45 time: 0.6372 data_time: 0.0015 memory: 44138 loss: 0.4101 +2023/06/06 06:37:30 - mmengine - INFO - Epoch(train) [2][2900/3907] lr: 2.8540e-05 eta: 5:49:33 time: 0.6367 data_time: 0.0014 memory: 44138 loss: 0.4268 +2023/06/06 06:38:33 - mmengine - INFO - Epoch(train) [2][3000/3907] lr: 2.8497e-05 eta: 5:48:22 time: 0.6367 data_time: 0.0015 memory: 44138 loss: 0.4199 +2023/06/06 06:39:32 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr3e-5_20230606_052112 +2023/06/06 06:39:37 - mmengine - INFO - Epoch(train) [2][3100/3907] lr: 2.8455e-05 eta: 5:47:11 time: 0.6355 data_time: 0.0015 memory: 44138 loss: 0.4295 +2023/06/06 06:40:41 - mmengine - INFO - Epoch(train) [2][3200/3907] lr: 2.8412e-05 eta: 5:46:00 time: 0.6358 data_time: 0.0016 memory: 44138 loss: 0.4226 +2023/06/06 06:41:44 - mmengine - INFO - Epoch(train) [2][3300/3907] lr: 2.8368e-05 eta: 5:44:50 time: 0.6355 data_time: 0.0015 memory: 44138 loss: 0.4122 +2023/06/06 06:42:48 - mmengine - INFO - Epoch(train) [2][3400/3907] lr: 2.8323e-05 eta: 5:43:40 time: 0.6351 data_time: 0.0015 memory: 44138 loss: 0.4387 +2023/06/06 06:43:52 - mmengine - INFO - Epoch(train) [2][3500/3907] lr: 2.8279e-05 eta: 5:42:29 time: 0.6363 data_time: 0.0014 memory: 44138 loss: 0.4387 +2023/06/06 06:44:55 - mmengine - INFO - Epoch(train) [2][3600/3907] lr: 2.8233e-05 eta: 5:41:19 time: 0.6362 data_time: 0.0016 memory: 44138 loss: 0.4182 +2023/06/06 06:45:59 - mmengine - INFO - Epoch(train) [2][3700/3907] lr: 2.8187e-05 eta: 5:40:09 time: 0.6355 data_time: 0.0015 memory: 44138 loss: 0.4316 +2023/06/06 06:47:03 - mmengine - INFO - Epoch(train) [2][3800/3907] lr: 2.8141e-05 eta: 5:38:59 time: 0.6385 data_time: 0.0015 memory: 44138 loss: 0.3920 +2023/06/06 06:48:06 - mmengine - INFO - Epoch(train) [2][3900/3907] lr: 2.8094e-05 eta: 5:37:49 time: 0.6353 data_time: 0.0013 memory: 44138 loss: 0.3885 +2023/06/06 06:48:10 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr3e-5_20230606_052112 +2023/06/06 06:48:10 - mmengine - INFO - Saving checkpoint at 2 epochs +2023/06/06 06:49:52 - mmengine - INFO - Epoch(val) [2][57/57] accuracy/top1: 81.8169 single-label/precision_classwise: [77.02530670166016, 92.15557098388672] single-label/recall_classwise: [95.49270629882812, 65.02325439453125] single-label/f1-score_classwise: [85.27057647705078, 76.24761199951172] data_time: 0.0349 time: 1.2780 +2023/06/06 06:50:59 - mmengine - INFO - Epoch(train) [3][ 100/3907] lr: 2.8043e-05 eta: 5:36:46 time: 0.6348 data_time: 0.0024 memory: 44138 loss: 0.4136 +2023/06/06 06:51:54 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr3e-5_20230606_052112 +2023/06/06 06:52:02 - mmengine - INFO - Epoch(train) [3][ 200/3907] lr: 2.7995e-05 eta: 5:35:36 time: 0.6361 data_time: 0.0018 memory: 44138 loss: 0.3959 +2023/06/06 06:53:06 - mmengine - INFO - Epoch(train) [3][ 300/3907] lr: 2.7947e-05 eta: 5:34:27 time: 0.6361 data_time: 0.0014 memory: 44138 loss: 0.3940 +2023/06/06 06:54:10 - mmengine - INFO - Epoch(train) [3][ 400/3907] lr: 2.7897e-05 eta: 5:33:18 time: 0.6364 data_time: 0.0017 memory: 44138 loss: 0.4299 +2023/06/06 06:55:13 - mmengine - INFO - Epoch(train) [3][ 500/3907] lr: 2.7848e-05 eta: 5:32:09 time: 0.6382 data_time: 0.0015 memory: 44138 loss: 0.4141 +2023/06/06 06:56:17 - mmengine - INFO - Epoch(train) [3][ 600/3907] lr: 2.7798e-05 eta: 5:31:00 time: 0.6369 data_time: 0.0017 memory: 44138 loss: 0.4177 +2023/06/06 06:57:21 - mmengine - INFO - Epoch(train) [3][ 700/3907] lr: 2.7747e-05 eta: 5:29:51 time: 0.6355 data_time: 0.0016 memory: 44138 loss: 0.4205 +2023/06/06 06:58:25 - mmengine - INFO - Epoch(train) [3][ 800/3907] lr: 2.7696e-05 eta: 5:28:43 time: 0.6354 data_time: 0.0017 memory: 44138 loss: 0.4021 +2023/06/06 06:59:28 - mmengine - INFO - Epoch(train) [3][ 900/3907] lr: 2.7645e-05 eta: 5:27:34 time: 0.6379 data_time: 0.0017 memory: 44138 loss: 0.3941 +2023/06/06 07:00:32 - mmengine - INFO - Epoch(train) [3][1000/3907] lr: 2.7592e-05 eta: 5:26:26 time: 0.6357 data_time: 0.0017 memory: 44138 loss: 0.3778 +2023/06/06 07:01:36 - mmengine - INFO - Epoch(train) [3][1100/3907] lr: 2.7540e-05 eta: 5:25:18 time: 0.6357 data_time: 0.0016 memory: 44138 loss: 0.4226 +2023/06/06 07:02:31 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr3e-5_20230606_052112 +2023/06/06 07:02:40 - mmengine - INFO - Epoch(train) [3][1200/3907] lr: 2.7487e-05 eta: 5:24:10 time: 0.6348 data_time: 0.0015 memory: 44138 loss: 0.4193 +2023/06/06 07:03:43 - mmengine - INFO - Epoch(train) [3][1300/3907] lr: 2.7433e-05 eta: 5:23:02 time: 0.6383 data_time: 0.0014 memory: 44138 loss: 0.3944 +2023/06/06 07:04:47 - mmengine - INFO - Epoch(train) [3][1400/3907] lr: 2.7379e-05 eta: 5:21:54 time: 0.6354 data_time: 0.0015 memory: 44138 loss: 0.4156 +2023/06/06 07:05:51 - mmengine - INFO - Epoch(train) [3][1500/3907] lr: 2.7325e-05 eta: 5:20:46 time: 0.6366 data_time: 0.0014 memory: 44138 loss: 0.3944 +2023/06/06 07:06:54 - mmengine - INFO - Epoch(train) [3][1600/3907] lr: 2.7270e-05 eta: 5:19:38 time: 0.6367 data_time: 0.0015 memory: 44138 loss: 0.3986 +2023/06/06 07:07:58 - mmengine - INFO - Epoch(train) [3][1700/3907] lr: 2.7214e-05 eta: 5:18:30 time: 0.6352 data_time: 0.0015 memory: 44138 loss: 0.3906 +2023/06/06 07:09:02 - mmengine - INFO - Epoch(train) [3][1800/3907] lr: 2.7158e-05 eta: 5:17:23 time: 0.6367 data_time: 0.0015 memory: 44138 loss: 0.3704 +2023/06/06 07:10:05 - mmengine - INFO - Epoch(train) [3][1900/3907] lr: 2.7102e-05 eta: 5:16:15 time: 0.6357 data_time: 0.0014 memory: 44138 loss: 0.3927 +2023/06/06 07:11:09 - mmengine - INFO - Epoch(train) [3][2000/3907] lr: 2.7045e-05 eta: 5:15:07 time: 0.6356 data_time: 0.0014 memory: 44138 loss: 0.4295 +2023/06/06 07:12:13 - mmengine - INFO - Epoch(train) [3][2100/3907] lr: 2.6988e-05 eta: 5:14:00 time: 0.6366 data_time: 0.0014 memory: 44138 loss: 0.3927 +2023/06/06 07:13:07 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr3e-5_20230606_052112 +2023/06/06 07:13:16 - mmengine - INFO - Epoch(train) [3][2200/3907] lr: 2.6930e-05 eta: 5:12:52 time: 0.6367 data_time: 0.0015 memory: 44138 loss: 0.3868 +2023/06/06 07:14:20 - mmengine - INFO - Epoch(train) [3][2300/3907] lr: 2.6872e-05 eta: 5:11:45 time: 0.6373 data_time: 0.0014 memory: 44138 loss: 0.4050 +2023/06/06 07:15:23 - mmengine - INFO - Epoch(train) [3][2400/3907] lr: 2.6813e-05 eta: 5:10:38 time: 0.6353 data_time: 0.0015 memory: 44138 loss: 0.4070 +2023/06/06 07:16:27 - mmengine - INFO - Epoch(train) [3][2500/3907] lr: 2.6754e-05 eta: 5:09:31 time: 0.6362 data_time: 0.0015 memory: 44138 loss: 0.4014 +2023/06/06 07:17:31 - mmengine - INFO - Epoch(train) [3][2600/3907] lr: 2.6695e-05 eta: 5:08:23 time: 0.6354 data_time: 0.0015 memory: 44138 loss: 0.4043 +2023/06/06 07:18:35 - mmengine - INFO - Epoch(train) [3][2700/3907] lr: 2.6635e-05 eta: 5:07:16 time: 0.6371 data_time: 0.0015 memory: 44138 loss: 0.3955 +2023/06/06 07:19:38 - mmengine - INFO - Epoch(train) [3][2800/3907] lr: 2.6574e-05 eta: 5:06:09 time: 0.6354 data_time: 0.0016 memory: 44138 loss: 0.3969 +2023/06/06 07:20:42 - mmengine - INFO - Epoch(train) [3][2900/3907] lr: 2.6514e-05 eta: 5:05:03 time: 0.6440 data_time: 0.0015 memory: 44138 loss: 0.3879 +2023/06/06 07:21:45 - mmengine - INFO - Epoch(train) [3][3000/3907] lr: 2.6452e-05 eta: 5:03:55 time: 0.6356 data_time: 0.0015 memory: 44138 loss: 0.4132 +2023/06/06 07:22:49 - mmengine - INFO - Epoch(train) [3][3100/3907] lr: 2.6391e-05 eta: 5:02:49 time: 0.6355 data_time: 0.0015 memory: 44138 loss: 0.3975 +2023/06/06 07:23:44 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr3e-5_20230606_052112 +2023/06/06 07:23:53 - mmengine - INFO - Epoch(train) [3][3200/3907] lr: 2.6329e-05 eta: 5:01:42 time: 0.6355 data_time: 0.0017 memory: 44138 loss: 0.4156 +2023/06/06 07:24:56 - mmengine - INFO - Epoch(train) [3][3300/3907] lr: 2.6266e-05 eta: 5:00:35 time: 0.6360 data_time: 0.0015 memory: 44138 loss: 0.4108 +2023/06/06 07:26:00 - mmengine - INFO - Epoch(train) [3][3400/3907] lr: 2.6203e-05 eta: 4:59:28 time: 0.6377 data_time: 0.0017 memory: 44138 loss: 0.3996 +2023/06/06 07:27:04 - mmengine - INFO - Epoch(train) [3][3500/3907] lr: 2.6140e-05 eta: 4:58:22 time: 0.6386 data_time: 0.0015 memory: 44138 loss: 0.4072 +2023/06/06 07:28:07 - mmengine - INFO - Epoch(train) [3][3600/3907] lr: 2.6076e-05 eta: 4:57:15 time: 0.6379 data_time: 0.0015 memory: 44138 loss: 0.3959 +2023/06/06 07:29:11 - mmengine - INFO - Epoch(train) [3][3700/3907] lr: 2.6012e-05 eta: 4:56:09 time: 0.6354 data_time: 0.0015 memory: 44138 loss: 0.4010 +2023/06/06 07:30:15 - mmengine - INFO - Epoch(train) [3][3800/3907] lr: 2.5948e-05 eta: 4:55:02 time: 0.6385 data_time: 0.0015 memory: 44138 loss: 0.3996 +2023/06/06 07:31:18 - mmengine - INFO - Epoch(train) [3][3900/3907] lr: 2.5883e-05 eta: 4:53:56 time: 0.6359 data_time: 0.0013 memory: 44138 loss: 0.3793 +2023/06/06 07:31:22 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr3e-5_20230606_052112 +2023/06/06 07:31:22 - mmengine - INFO - Saving checkpoint at 3 epochs +2023/06/06 07:33:03 - mmengine - INFO - Epoch(val) [3][57/57] accuracy/top1: 81.4655 single-label/precision_classwise: [76.04796600341797, 94.23998260498047] single-label/recall_classwise: [96.8878173828125, 62.527130126953125] single-label/f1-score_classwise: [85.21221923828125, 75.17591094970703] data_time: 0.0318 time: 1.2766 +2023/06/06 07:34:10 - mmengine - INFO - Epoch(train) [4][ 100/3907] lr: 2.5813e-05 eta: 4:52:51 time: 0.6359 data_time: 0.0015 memory: 44138 loss: 0.4052 +2023/06/06 07:35:13 - mmengine - INFO - Epoch(train) [4][ 200/3907] lr: 2.5748e-05 eta: 4:51:45 time: 0.6373 data_time: 0.0019 memory: 44138 loss: 0.4041 +2023/06/06 07:36:04 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr3e-5_20230606_052112 +2023/06/06 07:36:17 - mmengine - INFO - Epoch(train) [4][ 300/3907] lr: 2.5682e-05 eta: 4:50:39 time: 0.6357 data_time: 0.0014 memory: 44138 loss: 0.3934 +2023/06/06 07:37:21 - mmengine - INFO - Epoch(train) [4][ 400/3907] lr: 2.5615e-05 eta: 4:49:32 time: 0.6355 data_time: 0.0016 memory: 44138 loss: 0.4020 +2023/06/06 07:38:24 - mmengine - INFO - Epoch(train) [4][ 500/3907] lr: 2.5549e-05 eta: 4:48:26 time: 0.6374 data_time: 0.0014 memory: 44138 loss: 0.4026 +2023/06/06 07:39:28 - mmengine - INFO - Epoch(train) [4][ 600/3907] lr: 2.5482e-05 eta: 4:47:20 time: 0.6376 data_time: 0.0014 memory: 44138 loss: 0.3912 +2023/06/06 07:40:32 - mmengine - INFO - Epoch(train) [4][ 700/3907] lr: 2.5414e-05 eta: 4:46:14 time: 0.6359 data_time: 0.0015 memory: 44138 loss: 0.4097 +2023/06/06 07:41:35 - mmengine - INFO - Epoch(train) [4][ 800/3907] lr: 2.5346e-05 eta: 4:45:08 time: 0.6381 data_time: 0.0015 memory: 44138 loss: 0.3840 +2023/06/06 07:42:39 - mmengine - INFO - Epoch(train) [4][ 900/3907] lr: 2.5278e-05 eta: 4:44:02 time: 0.6361 data_time: 0.0015 memory: 44138 loss: 0.4085 +2023/06/06 07:43:43 - mmengine - INFO - Epoch(train) [4][1000/3907] lr: 2.5210e-05 eta: 4:42:56 time: 0.6362 data_time: 0.0015 memory: 44138 loss: 0.3875 +2023/06/06 07:44:46 - mmengine - INFO - Epoch(train) [4][1100/3907] lr: 2.5141e-05 eta: 4:41:50 time: 0.6361 data_time: 0.0014 memory: 44138 loss: 0.4015 +2023/06/06 07:45:50 - mmengine - INFO - Epoch(train) [4][1200/3907] lr: 2.5072e-05 eta: 4:40:44 time: 0.6364 data_time: 0.0014 memory: 44138 loss: 0.3919 +2023/06/06 07:46:41 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr3e-5_20230606_052112 +2023/06/06 07:46:54 - mmengine - INFO - Epoch(train) [4][1300/3907] lr: 2.5002e-05 eta: 4:39:39 time: 0.6352 data_time: 0.0014 memory: 44138 loss: 0.3964 +2023/06/06 07:47:58 - mmengine - INFO - Epoch(train) [4][1400/3907] lr: 2.4933e-05 eta: 4:38:33 time: 0.6373 data_time: 0.0014 memory: 44138 loss: 0.3731 +2023/06/06 07:49:01 - mmengine - INFO - Epoch(train) [4][1500/3907] lr: 2.4862e-05 eta: 4:37:27 time: 0.6367 data_time: 0.0018 memory: 44138 loss: 0.3996 +2023/06/06 07:50:05 - mmengine - INFO - Epoch(train) [4][1600/3907] lr: 2.4792e-05 eta: 4:36:21 time: 0.6357 data_time: 0.0015 memory: 44138 loss: 0.4016 +2023/06/06 07:51:09 - mmengine - INFO - Epoch(train) [4][1700/3907] lr: 2.4721e-05 eta: 4:35:16 time: 0.6371 data_time: 0.0017 memory: 44138 loss: 0.3861 +2023/06/06 07:52:13 - mmengine - INFO - Epoch(train) [4][1800/3907] lr: 2.4650e-05 eta: 4:34:10 time: 0.6360 data_time: 0.0015 memory: 44138 loss: 0.3891 +2023/06/06 07:53:16 - mmengine - INFO - Epoch(train) [4][1900/3907] lr: 2.4579e-05 eta: 4:33:04 time: 0.6397 data_time: 0.0015 memory: 44138 loss: 0.3740 +2023/06/06 07:54:20 - mmengine - INFO - Epoch(train) [4][2000/3907] lr: 2.4507e-05 eta: 4:31:59 time: 0.6453 data_time: 0.0016 memory: 44138 loss: 0.3678 +2023/06/06 07:55:24 - mmengine - INFO - Epoch(train) [4][2100/3907] lr: 2.4435e-05 eta: 4:30:53 time: 0.6363 data_time: 0.0014 memory: 44138 loss: 0.4099 +2023/06/06 07:56:27 - mmengine - INFO - Epoch(train) [4][2200/3907] lr: 2.4363e-05 eta: 4:29:47 time: 0.6355 data_time: 0.0014 memory: 44138 loss: 0.3971 +2023/06/06 07:57:18 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr3e-5_20230606_052112 +2023/06/06 07:57:31 - mmengine - INFO - Epoch(train) [4][2300/3907] lr: 2.4291e-05 eta: 4:28:42 time: 0.6365 data_time: 0.0015 memory: 44138 loss: 0.3685 +2023/06/06 07:58:35 - mmengine - INFO - Epoch(train) [4][2400/3907] lr: 2.4218e-05 eta: 4:27:36 time: 0.6435 data_time: 0.0015 memory: 44138 loss: 0.3989 +2023/06/06 07:59:38 - mmengine - INFO - Epoch(train) [4][2500/3907] lr: 2.4145e-05 eta: 4:26:31 time: 0.6382 data_time: 0.0015 memory: 44138 loss: 0.3914 +2023/06/06 08:00:42 - mmengine - INFO - Epoch(train) [4][2600/3907] lr: 2.4072e-05 eta: 4:25:26 time: 0.6387 data_time: 0.0015 memory: 44138 loss: 0.3884 +2023/06/06 08:01:46 - mmengine - INFO - Epoch(train) [4][2700/3907] lr: 2.3998e-05 eta: 4:24:20 time: 0.6356 data_time: 0.0016 memory: 44138 loss: 0.3905 +2023/06/06 08:02:50 - mmengine - INFO - Epoch(train) [4][2800/3907] lr: 2.3924e-05 eta: 4:23:14 time: 0.6357 data_time: 0.0014 memory: 44138 loss: 0.4043 +2023/06/06 08:03:53 - mmengine - INFO - Epoch(train) [4][2900/3907] lr: 2.3850e-05 eta: 4:22:09 time: 0.6363 data_time: 0.0015 memory: 44138 loss: 0.3877 +2023/06/06 08:04:57 - mmengine - INFO - Epoch(train) [4][3000/3907] lr: 2.3776e-05 eta: 4:21:04 time: 0.6356 data_time: 0.0016 memory: 44138 loss: 0.3915 +2023/06/06 08:06:01 - mmengine - INFO - Epoch(train) [4][3100/3907] lr: 2.3701e-05 eta: 4:19:58 time: 0.6358 data_time: 0.0014 memory: 44138 loss: 0.3847 +2023/06/06 08:07:04 - mmengine - INFO - Epoch(train) [4][3200/3907] lr: 2.3626e-05 eta: 4:18:53 time: 0.6361 data_time: 0.0015 memory: 44138 loss: 0.3819 +2023/06/06 08:07:55 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr3e-5_20230606_052112 +2023/06/06 08:08:08 - mmengine - INFO - Epoch(train) [4][3300/3907] lr: 2.3551e-05 eta: 4:17:47 time: 0.6354 data_time: 0.0015 memory: 44138 loss: 0.3837 +2023/06/06 08:09:12 - mmengine - INFO - Epoch(train) [4][3400/3907] lr: 2.3476e-05 eta: 4:16:42 time: 0.6390 data_time: 0.0015 memory: 44138 loss: 0.4051 +2023/06/06 08:10:16 - mmengine - INFO - Epoch(train) [4][3500/3907] lr: 2.3400e-05 eta: 4:15:37 time: 0.6361 data_time: 0.0015 memory: 44138 loss: 0.3862 +2023/06/06 08:11:19 - mmengine - INFO - Epoch(train) [4][3600/3907] lr: 2.3325e-05 eta: 4:14:32 time: 0.6367 data_time: 0.0015 memory: 44138 loss: 0.3678 +2023/06/06 08:12:23 - mmengine - INFO - Epoch(train) [4][3700/3907] lr: 2.3249e-05 eta: 4:13:27 time: 0.6356 data_time: 0.0015 memory: 44138 loss: 0.3959 +2023/06/06 08:13:27 - mmengine - INFO - Epoch(train) [4][3800/3907] lr: 2.3173e-05 eta: 4:12:21 time: 0.6355 data_time: 0.0014 memory: 44138 loss: 0.3997 +2023/06/06 08:14:30 - mmengine - INFO - Epoch(train) [4][3900/3907] lr: 2.3096e-05 eta: 4:11:16 time: 0.6355 data_time: 0.0013 memory: 44138 loss: 0.3931 +2023/06/06 08:14:34 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr3e-5_20230606_052112 +2023/06/06 08:14:34 - mmengine - INFO - Saving checkpoint at 4 epochs +2023/06/06 08:16:16 - mmengine - INFO - Epoch(val) [4][57/57] accuracy/top1: 80.3138 single-label/precision_classwise: [74.58592987060547, 95.08216857910156] single-label/recall_classwise: [97.5064697265625, 59.2015495300293] single-label/f1-score_classwise: [84.51983642578125, 72.96961212158203] data_time: 0.0321 time: 1.2764 +2023/06/06 08:17:23 - mmengine - INFO - Epoch(train) [5][ 100/3907] lr: 2.3014e-05 eta: 4:10:11 time: 0.6359 data_time: 0.0015 memory: 44138 loss: 0.3916 +2023/06/06 08:18:27 - mmengine - INFO - Epoch(train) [5][ 200/3907] lr: 2.2938e-05 eta: 4:09:05 time: 0.6368 data_time: 0.0016 memory: 44138 loss: 0.3880 +2023/06/06 08:19:31 - mmengine - INFO - Epoch(train) [5][ 300/3907] lr: 2.2861e-05 eta: 4:08:00 time: 0.6365 data_time: 0.0016 memory: 44138 loss: 0.4007 +2023/06/06 08:20:16 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr3e-5_20230606_052112 +2023/06/06 08:20:34 - mmengine - INFO - Epoch(train) [5][ 400/3907] lr: 2.2784e-05 eta: 4:06:55 time: 0.6363 data_time: 0.0016 memory: 44138 loss: 0.4241 +2023/06/06 08:21:38 - mmengine - INFO - Epoch(train) [5][ 500/3907] lr: 2.2706e-05 eta: 4:05:50 time: 0.6356 data_time: 0.0016 memory: 44138 loss: 0.3774 +2023/06/06 08:22:42 - mmengine - INFO - Epoch(train) [5][ 600/3907] lr: 2.2629e-05 eta: 4:04:44 time: 0.6361 data_time: 0.0016 memory: 44138 loss: 0.3699 +2023/06/06 08:23:45 - mmengine - INFO - Epoch(train) [5][ 700/3907] lr: 2.2551e-05 eta: 4:03:39 time: 0.6366 data_time: 0.0016 memory: 44138 loss: 0.4149 +2023/06/06 08:24:49 - mmengine - INFO - Epoch(train) [5][ 800/3907] lr: 2.2473e-05 eta: 4:02:34 time: 0.6381 data_time: 0.0015 memory: 44138 loss: 0.3635 +2023/06/06 08:25:53 - mmengine - INFO - Epoch(train) [5][ 900/3907] lr: 2.2395e-05 eta: 4:01:29 time: 0.6368 data_time: 0.0015 memory: 44138 loss: 0.3736 +2023/06/06 08:26:56 - mmengine - INFO - Epoch(train) [5][1000/3907] lr: 2.2317e-05 eta: 4:00:24 time: 0.6360 data_time: 0.0015 memory: 44138 loss: 0.3605 +2023/06/06 08:28:00 - mmengine - INFO - Epoch(train) [5][1100/3907] lr: 2.2239e-05 eta: 3:59:19 time: 0.6375 data_time: 0.0015 memory: 44138 loss: 0.3709 +2023/06/06 08:29:04 - mmengine - INFO - Epoch(train) [5][1200/3907] lr: 2.2160e-05 eta: 3:58:14 time: 0.6363 data_time: 0.0015 memory: 44138 loss: 0.3899 +2023/06/06 08:30:08 - mmengine - INFO - Epoch(train) [5][1300/3907] lr: 2.2082e-05 eta: 3:57:09 time: 0.6384 data_time: 0.0015 memory: 44138 loss: 0.3937 +2023/06/06 08:30:54 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr3e-5_20230606_052112 +2023/06/06 08:31:12 - mmengine - INFO - Epoch(train) [5][1400/3907] lr: 2.2003e-05 eta: 3:56:04 time: 0.6391 data_time: 0.0014 memory: 44138 loss: 0.3768 +2023/06/06 08:32:15 - mmengine - INFO - Epoch(train) [5][1500/3907] lr: 2.1924e-05 eta: 3:54:59 time: 0.6362 data_time: 0.0015 memory: 44138 loss: 0.4080 +2023/06/06 08:33:19 - mmengine - INFO - Epoch(train) [5][1600/3907] lr: 2.1845e-05 eta: 3:53:55 time: 0.6375 data_time: 0.0016 memory: 44138 loss: 0.3828 +2023/06/06 08:34:23 - mmengine - INFO - Epoch(train) [5][1700/3907] lr: 2.1766e-05 eta: 3:52:50 time: 0.6384 data_time: 0.0014 memory: 44138 loss: 0.3655 +2023/06/06 08:35:27 - mmengine - INFO - Epoch(train) [5][1800/3907] lr: 2.1687e-05 eta: 3:51:45 time: 0.6359 data_time: 0.0017 memory: 44138 loss: 0.3966 +2023/06/06 08:36:30 - mmengine - INFO - Epoch(train) [5][1900/3907] lr: 2.1608e-05 eta: 3:50:40 time: 0.6355 data_time: 0.0015 memory: 44138 loss: 0.3885 +2023/06/06 08:37:34 - mmengine - INFO - Epoch(train) [5][2000/3907] lr: 2.1528e-05 eta: 3:49:35 time: 0.6358 data_time: 0.0014 memory: 44138 loss: 0.3777 +2023/06/06 08:38:38 - mmengine - INFO - Epoch(train) [5][2100/3907] lr: 2.1449e-05 eta: 3:48:30 time: 0.6369 data_time: 0.0016 memory: 44138 loss: 0.3829 +2023/06/06 08:39:41 - mmengine - INFO - Epoch(train) [5][2200/3907] lr: 2.1369e-05 eta: 3:47:25 time: 0.6396 data_time: 0.0016 memory: 44138 loss: 0.3939 +2023/06/06 08:40:45 - mmengine - INFO - Epoch(train) [5][2300/3907] lr: 2.1289e-05 eta: 3:46:20 time: 0.6377 data_time: 0.0014 memory: 44138 loss: 0.3830 +2023/06/06 08:41:31 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr3e-5_20230606_052112 +2023/06/06 08:41:49 - mmengine - INFO - Epoch(train) [5][2400/3907] lr: 2.1210e-05 eta: 3:45:15 time: 0.6373 data_time: 0.0015 memory: 44138 loss: 0.4075 +2023/06/06 08:42:53 - mmengine - INFO - Epoch(train) [5][2500/3907] lr: 2.1130e-05 eta: 3:44:11 time: 0.6363 data_time: 0.0014 memory: 44138 loss: 0.3941 +2023/06/06 08:43:56 - mmengine - INFO - Epoch(train) [5][2600/3907] lr: 2.1050e-05 eta: 3:43:06 time: 0.6369 data_time: 0.0015 memory: 44138 loss: 0.3669 +2023/06/06 08:45:00 - mmengine - INFO - Epoch(train) [5][2700/3907] lr: 2.0970e-05 eta: 3:42:01 time: 0.6359 data_time: 0.0016 memory: 44138 loss: 0.3765 +2023/06/06 08:46:04 - mmengine - INFO - Epoch(train) [5][2800/3907] lr: 2.0890e-05 eta: 3:40:56 time: 0.6384 data_time: 0.0015 memory: 44138 loss: 0.3842 +2023/06/06 08:47:07 - mmengine - INFO - Epoch(train) [5][2900/3907] lr: 2.0810e-05 eta: 3:39:51 time: 0.6357 data_time: 0.0014 memory: 44138 loss: 0.4083 +2023/06/06 08:48:11 - mmengine - INFO - Epoch(train) [5][3000/3907] lr: 2.0729e-05 eta: 3:38:47 time: 0.6359 data_time: 0.0016 memory: 44138 loss: 0.4011 +2023/06/06 08:49:15 - mmengine - INFO - Epoch(train) [5][3100/3907] lr: 2.0649e-05 eta: 3:37:42 time: 0.6411 data_time: 0.0015 memory: 44138 loss: 0.3510 +2023/06/06 08:50:19 - mmengine - INFO - Epoch(train) [5][3200/3907] lr: 2.0569e-05 eta: 3:36:37 time: 0.6372 data_time: 0.0016 memory: 44138 loss: 0.3843 +2023/06/06 08:51:23 - mmengine - INFO - Epoch(train) [5][3300/3907] lr: 2.0489e-05 eta: 3:35:33 time: 0.6371 data_time: 0.0015 memory: 44138 loss: 0.3697 +2023/06/06 08:52:09 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr3e-5_20230606_052112 +2023/06/06 08:52:27 - mmengine - INFO - Epoch(train) [5][3400/3907] lr: 2.0408e-05 eta: 3:34:28 time: 0.6382 data_time: 0.0014 memory: 44138 loss: 0.3669 +2023/06/06 08:53:30 - mmengine - INFO - Epoch(train) [5][3500/3907] lr: 2.0328e-05 eta: 3:33:23 time: 0.6376 data_time: 0.0015 memory: 44138 loss: 0.3740 +2023/06/06 08:54:34 - mmengine - INFO - Epoch(train) [5][3600/3907] lr: 2.0248e-05 eta: 3:32:19 time: 0.6356 data_time: 0.0015 memory: 44138 loss: 0.3798 +2023/06/06 08:55:38 - mmengine - INFO - Epoch(train) [5][3700/3907] lr: 2.0167e-05 eta: 3:31:14 time: 0.6385 data_time: 0.0014 memory: 44138 loss: 0.3770 +2023/06/06 08:56:41 - mmengine - INFO - Epoch(train) [5][3800/3907] lr: 2.0087e-05 eta: 3:30:09 time: 0.6356 data_time: 0.0015 memory: 44138 loss: 0.3950 +2023/06/06 08:57:45 - mmengine - INFO - Epoch(train) [5][3900/3907] lr: 2.0006e-05 eta: 3:29:05 time: 0.6378 data_time: 0.0012 memory: 44138 loss: 0.3776 +2023/06/06 08:57:49 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr3e-5_20230606_052112 +2023/06/06 08:57:49 - mmengine - INFO - Saving checkpoint at 5 epochs +2023/06/06 08:59:30 - mmengine - INFO - Epoch(val) [5][57/57] accuracy/top1: 79.6736 single-label/precision_classwise: [73.83551788330078, 95.44166564941406] single-label/recall_classwise: [97.76528930664062, 57.45736312866211] single-label/f1-score_classwise: [84.13190460205078, 71.7313461303711] data_time: 0.0310 time: 1.2739 +2023/06/06 09:00:37 - mmengine - INFO - Epoch(train) [6][ 100/3907] lr: 1.9920e-05 eta: 3:27:59 time: 0.6389 data_time: 0.0013 memory: 44138 loss: 0.3793 +2023/06/06 09:01:41 - mmengine - INFO - Epoch(train) [6][ 200/3907] lr: 1.9840e-05 eta: 3:26:54 time: 0.6357 data_time: 0.0014 memory: 44138 loss: 0.3913 +2023/06/06 09:02:45 - mmengine - INFO - Epoch(train) [6][ 300/3907] lr: 1.9760e-05 eta: 3:25:49 time: 0.6364 data_time: 0.0015 memory: 44138 loss: 0.4005 +2023/06/06 09:03:48 - mmengine - INFO - Epoch(train) [6][ 400/3907] lr: 1.9679e-05 eta: 3:24:44 time: 0.6368 data_time: 0.0015 memory: 44138 loss: 0.3783 +2023/06/06 09:04:30 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr3e-5_20230606_052112 +2023/06/06 09:04:52 - mmengine - INFO - Epoch(train) [6][ 500/3907] lr: 1.9599e-05 eta: 3:23:40 time: 0.6455 data_time: 0.0015 memory: 44138 loss: 0.3835 +2023/06/06 09:05:56 - mmengine - INFO - Epoch(train) [6][ 600/3907] lr: 1.9519e-05 eta: 3:22:35 time: 0.6390 data_time: 0.0015 memory: 44138 loss: 0.3652 +2023/06/06 09:07:00 - mmengine - INFO - Epoch(train) [6][ 700/3907] lr: 1.9438e-05 eta: 3:21:31 time: 0.6370 data_time: 0.0014 memory: 44138 loss: 0.3933 +2023/06/06 09:08:03 - mmengine - INFO - Epoch(train) [6][ 800/3907] lr: 1.9358e-05 eta: 3:20:26 time: 0.6359 data_time: 0.0014 memory: 44138 loss: 0.3975 +2023/06/06 09:09:07 - mmengine - INFO - Epoch(train) [6][ 900/3907] lr: 1.9278e-05 eta: 3:19:21 time: 0.6433 data_time: 0.0016 memory: 44138 loss: 0.4032 +2023/06/06 09:10:11 - mmengine - INFO - Epoch(train) [6][1000/3907] lr: 1.9198e-05 eta: 3:18:17 time: 0.6373 data_time: 0.0014 memory: 44138 loss: 0.3704 +2023/06/06 09:11:15 - mmengine - INFO - Epoch(train) [6][1100/3907] lr: 1.9117e-05 eta: 3:17:12 time: 0.6361 data_time: 0.0015 memory: 44138 loss: 0.3754 +2023/06/06 09:12:18 - mmengine - INFO - Epoch(train) [6][1200/3907] lr: 1.9037e-05 eta: 3:16:08 time: 0.6362 data_time: 0.0015 memory: 44138 loss: 0.3556 +2023/06/06 09:13:22 - mmengine - INFO - Epoch(train) [6][1300/3907] lr: 1.8957e-05 eta: 3:15:03 time: 0.6368 data_time: 0.0017 memory: 44138 loss: 0.3643 +2023/06/06 09:14:26 - mmengine - INFO - Epoch(train) [6][1400/3907] lr: 1.8877e-05 eta: 3:13:58 time: 0.6382 data_time: 0.0015 memory: 44138 loss: 0.3651 +2023/06/06 09:15:07 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr3e-5_20230606_052112 +2023/06/06 09:15:29 - mmengine - INFO - Epoch(train) [6][1500/3907] lr: 1.8798e-05 eta: 3:12:54 time: 0.6362 data_time: 0.0015 memory: 44138 loss: 0.3655 +2023/06/06 09:16:33 - mmengine - INFO - Epoch(train) [6][1600/3907] lr: 1.8718e-05 eta: 3:11:49 time: 0.6353 data_time: 0.0016 memory: 44138 loss: 0.3602 +2023/06/06 09:17:37 - mmengine - INFO - Epoch(train) [6][1700/3907] lr: 1.8638e-05 eta: 3:10:44 time: 0.6357 data_time: 0.0015 memory: 44138 loss: 0.3669 +2023/06/06 09:18:40 - mmengine - INFO - Epoch(train) [6][1800/3907] lr: 1.8558e-05 eta: 3:09:40 time: 0.6385 data_time: 0.0016 memory: 44138 loss: 0.3765 +2023/06/06 09:19:44 - mmengine - INFO - Epoch(train) [6][1900/3907] lr: 1.8479e-05 eta: 3:08:36 time: 0.6383 data_time: 0.0015 memory: 44138 loss: 0.3942 +2023/06/06 09:20:48 - mmengine - INFO - Epoch(train) [6][2000/3907] lr: 1.8400e-05 eta: 3:07:31 time: 0.6389 data_time: 0.0014 memory: 44138 loss: 0.3796 +2023/06/06 09:21:52 - mmengine - INFO - Epoch(train) [6][2100/3907] lr: 1.8320e-05 eta: 3:06:27 time: 0.6366 data_time: 0.0017 memory: 44138 loss: 0.3982 +2023/06/06 09:22:56 - mmengine - INFO - Epoch(train) [6][2200/3907] lr: 1.8241e-05 eta: 3:05:22 time: 0.6386 data_time: 0.0015 memory: 44138 loss: 0.3645 +2023/06/06 09:23:59 - mmengine - INFO - Epoch(train) [6][2300/3907] lr: 1.8162e-05 eta: 3:04:18 time: 0.6370 data_time: 0.0016 memory: 44138 loss: 0.3877 +2023/06/06 09:25:03 - mmengine - INFO - Epoch(train) [6][2400/3907] lr: 1.8083e-05 eta: 3:03:13 time: 0.6361 data_time: 0.0014 memory: 44138 loss: 0.3965 +2023/06/06 09:25:45 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr3e-5_20230606_052112 +2023/06/06 09:26:07 - mmengine - INFO - Epoch(train) [6][2500/3907] lr: 1.8004e-05 eta: 3:02:09 time: 0.6363 data_time: 0.0015 memory: 44138 loss: 0.3838 +2023/06/06 09:27:10 - mmengine - INFO - Epoch(train) [6][2600/3907] lr: 1.7925e-05 eta: 3:01:04 time: 0.6365 data_time: 0.0016 memory: 44138 loss: 0.4102 +2023/06/06 09:28:14 - mmengine - INFO - Epoch(train) [6][2700/3907] lr: 1.7847e-05 eta: 3:00:00 time: 0.6381 data_time: 0.0015 memory: 44138 loss: 0.3947 +2023/06/06 09:29:18 - mmengine - INFO - Epoch(train) [6][2800/3907] lr: 1.7768e-05 eta: 2:58:55 time: 0.6350 data_time: 0.0015 memory: 44138 loss: 0.3717 +2023/06/06 09:30:22 - mmengine - INFO - Epoch(train) [6][2900/3907] lr: 1.7690e-05 eta: 2:57:51 time: 0.6372 data_time: 0.0014 memory: 44138 loss: 0.3782 +2023/06/06 09:31:25 - mmengine - INFO - Epoch(train) [6][3000/3907] lr: 1.7612e-05 eta: 2:56:46 time: 0.6372 data_time: 0.0015 memory: 44138 loss: 0.3384 +2023/06/06 09:32:29 - mmengine - INFO - Epoch(train) [6][3100/3907] lr: 1.7534e-05 eta: 2:55:42 time: 0.6378 data_time: 0.0015 memory: 44138 loss: 0.3631 +2023/06/06 09:33:33 - mmengine - INFO - Epoch(train) [6][3200/3907] lr: 1.7456e-05 eta: 2:54:37 time: 0.6370 data_time: 0.0017 memory: 44138 loss: 0.3875 +2023/06/06 09:34:37 - mmengine - INFO - Epoch(train) [6][3300/3907] lr: 1.7378e-05 eta: 2:53:33 time: 0.6368 data_time: 0.0015 memory: 44138 loss: 0.3765 +2023/06/06 09:35:40 - mmengine - INFO - Epoch(train) [6][3400/3907] lr: 1.7301e-05 eta: 2:52:29 time: 0.6382 data_time: 0.0014 memory: 44138 loss: 0.3879 +2023/06/06 09:36:22 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr3e-5_20230606_052112 +2023/06/06 09:36:44 - mmengine - INFO - Epoch(train) [6][3500/3907] lr: 1.7223e-05 eta: 2:51:24 time: 0.6376 data_time: 0.0014 memory: 44138 loss: 0.3694 +2023/06/06 09:37:48 - mmengine - INFO - Epoch(train) [6][3600/3907] lr: 1.7146e-05 eta: 2:50:20 time: 0.6368 data_time: 0.0014 memory: 44138 loss: 0.3771 +2023/06/06 09:38:52 - mmengine - INFO - Epoch(train) [6][3700/3907] lr: 1.7069e-05 eta: 2:49:15 time: 0.6374 data_time: 0.0015 memory: 44138 loss: 0.3630 +2023/06/06 09:39:55 - mmengine - INFO - Epoch(train) [6][3800/3907] lr: 1.6993e-05 eta: 2:48:11 time: 0.6366 data_time: 0.0014 memory: 44138 loss: 0.3609 +2023/06/06 09:40:59 - mmengine - INFO - Epoch(train) [6][3900/3907] lr: 1.6916e-05 eta: 2:47:06 time: 0.6344 data_time: 0.0014 memory: 44138 loss: 0.3965 +2023/06/06 09:41:03 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr3e-5_20230606_052112 +2023/06/06 09:41:03 - mmengine - INFO - Saving checkpoint at 6 epochs +2023/06/06 09:42:45 - mmengine - INFO - Epoch(val) [6][57/57] accuracy/top1: 79.6632 single-label/precision_classwise: [73.68944549560547, 96.1533432006836] single-label/recall_classwise: [98.14405822753906, 56.968994140625] single-label/f1-score_classwise: [84.1766128540039, 71.5474853515625] data_time: 0.0322 time: 1.2768 +2023/06/06 09:43:52 - mmengine - INFO - Epoch(train) [7][ 100/3907] lr: 1.6834e-05 eta: 2:45:59 time: 0.6352 data_time: 0.0016 memory: 44138 loss: 0.3702 +2023/06/06 09:44:56 - mmengine - INFO - Epoch(train) [7][ 200/3907] lr: 1.6758e-05 eta: 2:44:55 time: 0.6352 data_time: 0.0015 memory: 44138 loss: 0.3628 +2023/06/06 09:45:59 - mmengine - INFO - Epoch(train) [7][ 300/3907] lr: 1.6682e-05 eta: 2:43:50 time: 0.6370 data_time: 0.0015 memory: 44138 loss: 0.3686 +2023/06/06 09:47:03 - mmengine - INFO - Epoch(train) [7][ 400/3907] lr: 1.6606e-05 eta: 2:42:46 time: 0.6356 data_time: 0.0015 memory: 44138 loss: 0.3560 +2023/06/06 09:48:07 - mmengine - INFO - Epoch(train) [7][ 500/3907] lr: 1.6531e-05 eta: 2:41:41 time: 0.6364 data_time: 0.0015 memory: 44138 loss: 0.3811 +2023/06/06 09:48:44 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr3e-5_20230606_052112 +2023/06/06 09:49:10 - mmengine - INFO - Epoch(train) [7][ 600/3907] lr: 1.6456e-05 eta: 2:40:37 time: 0.6363 data_time: 0.0015 memory: 44138 loss: 0.4050 +2023/06/06 09:50:14 - mmengine - INFO - Epoch(train) [7][ 700/3907] lr: 1.6380e-05 eta: 2:39:32 time: 0.6383 data_time: 0.0015 memory: 44138 loss: 0.3550 +2023/06/06 09:51:18 - mmengine - INFO - Epoch(train) [7][ 800/3907] lr: 1.6306e-05 eta: 2:38:28 time: 0.6361 data_time: 0.0016 memory: 44138 loss: 0.3617 +2023/06/06 09:52:21 - mmengine - INFO - Epoch(train) [7][ 900/3907] lr: 1.6231e-05 eta: 2:37:24 time: 0.6377 data_time: 0.0014 memory: 44138 loss: 0.3531 +2023/06/06 09:53:25 - mmengine - INFO - Epoch(train) [7][1000/3907] lr: 1.6157e-05 eta: 2:36:19 time: 0.6369 data_time: 0.0015 memory: 44138 loss: 0.3748 +2023/06/06 09:54:29 - mmengine - INFO - Epoch(train) [7][1100/3907] lr: 1.6083e-05 eta: 2:35:15 time: 0.6350 data_time: 0.0015 memory: 44138 loss: 0.3918 +2023/06/06 09:55:33 - mmengine - INFO - Epoch(train) [7][1200/3907] lr: 1.6009e-05 eta: 2:34:11 time: 0.6369 data_time: 0.0014 memory: 44138 loss: 0.3528 +2023/06/06 09:56:36 - mmengine - INFO - Epoch(train) [7][1300/3907] lr: 1.5935e-05 eta: 2:33:06 time: 0.6396 data_time: 0.0016 memory: 44138 loss: 0.3765 +2023/06/06 09:57:40 - mmengine - INFO - Epoch(train) [7][1400/3907] lr: 1.5862e-05 eta: 2:32:02 time: 0.6356 data_time: 0.0015 memory: 44138 loss: 0.3666 +2023/06/06 09:58:44 - mmengine - INFO - Epoch(train) [7][1500/3907] lr: 1.5789e-05 eta: 2:30:58 time: 0.6371 data_time: 0.0015 memory: 44138 loss: 0.3824 +2023/06/06 09:59:21 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr3e-5_20230606_052112 +2023/06/06 09:59:48 - mmengine - INFO - Epoch(train) [7][1600/3907] lr: 1.5716e-05 eta: 2:29:53 time: 0.6372 data_time: 0.0015 memory: 44138 loss: 0.3892 +2023/06/06 10:00:51 - mmengine - INFO - Epoch(train) [7][1700/3907] lr: 1.5643e-05 eta: 2:28:49 time: 0.6357 data_time: 0.0015 memory: 44138 loss: 0.3589 +2023/06/06 10:01:55 - mmengine - INFO - Epoch(train) [7][1800/3907] lr: 1.5571e-05 eta: 2:27:45 time: 0.6357 data_time: 0.0014 memory: 44138 loss: 0.3735 +2023/06/06 10:02:59 - mmengine - INFO - Epoch(train) [7][1900/3907] lr: 1.5499e-05 eta: 2:26:40 time: 0.6362 data_time: 0.0015 memory: 44138 loss: 0.3890 +2023/06/06 10:04:02 - mmengine - INFO - Epoch(train) [7][2000/3907] lr: 1.5428e-05 eta: 2:25:36 time: 0.6365 data_time: 0.0015 memory: 44138 loss: 0.3634 +2023/06/06 10:05:06 - mmengine - INFO - Epoch(train) [7][2100/3907] lr: 1.5356e-05 eta: 2:24:32 time: 0.6360 data_time: 0.0016 memory: 44138 loss: 0.3768 +2023/06/06 10:06:10 - mmengine - INFO - Epoch(train) [7][2200/3907] lr: 1.5285e-05 eta: 2:23:27 time: 0.6351 data_time: 0.0015 memory: 44138 loss: 0.3859 +2023/06/06 10:07:13 - mmengine - INFO - Epoch(train) [7][2300/3907] lr: 1.5214e-05 eta: 2:22:23 time: 0.6403 data_time: 0.0014 memory: 44138 loss: 0.3875 +2023/06/06 10:08:17 - mmengine - INFO - Epoch(train) [7][2400/3907] lr: 1.5144e-05 eta: 2:21:19 time: 0.6361 data_time: 0.0014 memory: 44138 loss: 0.3642 +2023/06/06 10:09:21 - mmengine - INFO - Epoch(train) [7][2500/3907] lr: 1.5074e-05 eta: 2:20:14 time: 0.6364 data_time: 0.0020 memory: 44138 loss: 0.3523 +2023/06/06 10:09:58 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr3e-5_20230606_052112 +2023/06/06 10:10:25 - mmengine - INFO - Epoch(train) [7][2600/3907] lr: 1.5004e-05 eta: 2:19:10 time: 0.6363 data_time: 0.0015 memory: 44138 loss: 0.3742 +2023/06/06 10:11:28 - mmengine - INFO - Epoch(train) [7][2700/3907] lr: 1.4934e-05 eta: 2:18:06 time: 0.6378 data_time: 0.0014 memory: 44138 loss: 0.3535 +2023/06/06 10:12:32 - mmengine - INFO - Epoch(train) [7][2800/3907] lr: 1.4865e-05 eta: 2:17:01 time: 0.6352 data_time: 0.0014 memory: 44138 loss: 0.3805 +2023/06/06 10:13:36 - mmengine - INFO - Epoch(train) [7][2900/3907] lr: 1.4796e-05 eta: 2:15:57 time: 0.6355 data_time: 0.0015 memory: 44138 loss: 0.3934 +2023/06/06 10:14:39 - mmengine - INFO - Epoch(train) [7][3000/3907] lr: 1.4728e-05 eta: 2:14:53 time: 0.6362 data_time: 0.0017 memory: 44138 loss: 0.3819 +2023/06/06 10:15:43 - mmengine - INFO - Epoch(train) [7][3100/3907] lr: 1.4660e-05 eta: 2:13:49 time: 0.6362 data_time: 0.0016 memory: 44138 loss: 0.3392 +2023/06/06 10:16:47 - mmengine - INFO - Epoch(train) [7][3200/3907] lr: 1.4592e-05 eta: 2:12:44 time: 0.6364 data_time: 0.0015 memory: 44138 loss: 0.3761 +2023/06/06 10:17:50 - mmengine - INFO - Epoch(train) [7][3300/3907] lr: 1.4525e-05 eta: 2:11:40 time: 0.6365 data_time: 0.0017 memory: 44138 loss: 0.3643 +2023/06/06 10:18:54 - mmengine - INFO - Epoch(train) [7][3400/3907] lr: 1.4457e-05 eta: 2:10:36 time: 0.6364 data_time: 0.0014 memory: 44138 loss: 0.3812 +2023/06/06 10:19:58 - mmengine - INFO - Epoch(train) [7][3500/3907] lr: 1.4391e-05 eta: 2:09:32 time: 0.6356 data_time: 0.0015 memory: 44138 loss: 0.3685 +2023/06/06 10:20:35 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr3e-5_20230606_052112 +2023/06/06 10:21:02 - mmengine - INFO - Epoch(train) [7][3600/3907] lr: 1.4324e-05 eta: 2:08:27 time: 0.6396 data_time: 0.0015 memory: 44138 loss: 0.3588 +2023/06/06 10:22:06 - mmengine - INFO - Epoch(train) [7][3700/3907] lr: 1.4258e-05 eta: 2:07:23 time: 0.6377 data_time: 0.0015 memory: 44138 loss: 0.3918 +2023/06/06 10:23:09 - mmengine - INFO - Epoch(train) [7][3800/3907] lr: 1.4193e-05 eta: 2:06:19 time: 0.6497 data_time: 0.0016 memory: 44138 loss: 0.3637 +2023/06/06 10:24:13 - mmengine - INFO - Epoch(train) [7][3900/3907] lr: 1.4127e-05 eta: 2:05:15 time: 0.6345 data_time: 0.0013 memory: 44138 loss: 0.3829 +2023/06/06 10:24:17 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr3e-5_20230606_052112 +2023/06/06 10:24:17 - mmengine - INFO - Saving checkpoint at 7 epochs +2023/06/06 10:25:55 - mmengine - INFO - Epoch(val) [7][57/57] accuracy/top1: 79.1065 single-label/precision_classwise: [73.08920288085938, 96.33113861083984] single-label/recall_classwise: [98.27661895751953, 55.565895080566406] single-label/f1-score_classwise: [83.8318862915039, 70.47834777832031] data_time: 0.0338 time: 1.2759 +2023/06/06 10:27:02 - mmengine - INFO - Epoch(train) [8][ 100/3907] lr: 1.4058e-05 eta: 2:04:07 time: 0.6369 data_time: 0.0015 memory: 44138 loss: 0.4042 +2023/06/06 10:28:06 - mmengine - INFO - Epoch(train) [8][ 200/3907] lr: 1.3993e-05 eta: 2:03:03 time: 0.6392 data_time: 0.0014 memory: 44138 loss: 0.3983 +2023/06/06 10:29:10 - mmengine - INFO - Epoch(train) [8][ 300/3907] lr: 1.3929e-05 eta: 2:01:59 time: 0.6367 data_time: 0.0015 memory: 44138 loss: 0.3864 +2023/06/06 10:30:13 - mmengine - INFO - Epoch(train) [8][ 400/3907] lr: 1.3866e-05 eta: 2:00:55 time: 0.6354 data_time: 0.0014 memory: 44138 loss: 0.3792 +2023/06/06 10:31:17 - mmengine - INFO - Epoch(train) [8][ 500/3907] lr: 1.3802e-05 eta: 1:59:50 time: 0.6385 data_time: 0.0014 memory: 44138 loss: 0.3679 +2023/06/06 10:32:21 - mmengine - INFO - Epoch(train) [8][ 600/3907] lr: 1.3739e-05 eta: 1:58:46 time: 0.6369 data_time: 0.0014 memory: 44138 loss: 0.3605 +2023/06/06 10:32:54 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr3e-5_20230606_052112 +2023/06/06 10:33:25 - mmengine - INFO - Epoch(train) [8][ 700/3907] lr: 1.3677e-05 eta: 1:57:42 time: 0.6393 data_time: 0.0015 memory: 44138 loss: 0.3706 +2023/06/06 10:34:29 - mmengine - INFO - Epoch(train) [8][ 800/3907] lr: 1.3615e-05 eta: 1:56:38 time: 0.6377 data_time: 0.0014 memory: 44138 loss: 0.3896 +2023/06/06 10:35:32 - mmengine - INFO - Epoch(train) [8][ 900/3907] lr: 1.3553e-05 eta: 1:55:34 time: 0.6371 data_time: 0.0015 memory: 44138 loss: 0.3782 +2023/06/06 10:36:36 - mmengine - INFO - Epoch(train) [8][1000/3907] lr: 1.3492e-05 eta: 1:54:29 time: 0.6376 data_time: 0.0016 memory: 44138 loss: 0.3768 +2023/06/06 10:37:40 - mmengine - INFO - Epoch(train) [8][1100/3907] lr: 1.3431e-05 eta: 1:53:25 time: 0.6351 data_time: 0.0015 memory: 44138 loss: 0.3737 +2023/06/06 10:38:44 - mmengine - INFO - Epoch(train) [8][1200/3907] lr: 1.3371e-05 eta: 1:52:21 time: 0.6368 data_time: 0.0015 memory: 44138 loss: 0.3834 +2023/06/06 10:39:47 - mmengine - INFO - Epoch(train) [8][1300/3907] lr: 1.3311e-05 eta: 1:51:17 time: 0.6374 data_time: 0.0015 memory: 44138 loss: 0.3759 +2023/06/06 10:40:51 - mmengine - INFO - Epoch(train) [8][1400/3907] lr: 1.3251e-05 eta: 1:50:13 time: 0.6370 data_time: 0.0014 memory: 44138 loss: 0.3792 +2023/06/06 10:41:55 - mmengine - INFO - Epoch(train) [8][1500/3907] lr: 1.3192e-05 eta: 1:49:08 time: 0.6371 data_time: 0.0015 memory: 44138 loss: 0.3774 +2023/06/06 10:42:59 - mmengine - INFO - Epoch(train) [8][1600/3907] lr: 1.3133e-05 eta: 1:48:04 time: 0.6372 data_time: 0.0015 memory: 44138 loss: 0.3626 +2023/06/06 10:43:31 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr3e-5_20230606_052112 +2023/06/06 10:44:03 - mmengine - INFO - Epoch(train) [8][1700/3907] lr: 1.3075e-05 eta: 1:47:00 time: 0.6357 data_time: 0.0015 memory: 44138 loss: 0.3762 +2023/06/06 10:45:06 - mmengine - INFO - Epoch(train) [8][1800/3907] lr: 1.3017e-05 eta: 1:45:56 time: 0.6370 data_time: 0.0016 memory: 44138 loss: 0.3778 +2023/06/06 10:46:10 - mmengine - INFO - Epoch(train) [8][1900/3907] lr: 1.2960e-05 eta: 1:44:52 time: 0.6366 data_time: 0.0015 memory: 44138 loss: 0.3494 +2023/06/06 10:47:14 - mmengine - INFO - Epoch(train) [8][2000/3907] lr: 1.2903e-05 eta: 1:43:48 time: 0.6380 data_time: 0.0016 memory: 44138 loss: 0.3399 +2023/06/06 10:48:18 - mmengine - INFO - Epoch(train) [8][2100/3907] lr: 1.2847e-05 eta: 1:42:43 time: 0.6388 data_time: 0.0015 memory: 44138 loss: 0.3769 +2023/06/06 10:49:21 - mmengine - INFO - Epoch(train) [8][2200/3907] lr: 1.2791e-05 eta: 1:41:39 time: 0.6378 data_time: 0.0016 memory: 44138 loss: 0.3620 +2023/06/06 10:50:25 - mmengine - INFO - Epoch(train) [8][2300/3907] lr: 1.2735e-05 eta: 1:40:35 time: 0.6369 data_time: 0.0015 memory: 44138 loss: 0.3535 +2023/06/06 10:51:29 - mmengine - INFO - Epoch(train) [8][2400/3907] lr: 1.2680e-05 eta: 1:39:31 time: 0.6366 data_time: 0.0015 memory: 44138 loss: 0.3676 +2023/06/06 10:52:33 - mmengine - INFO - Epoch(train) [8][2500/3907] lr: 1.2626e-05 eta: 1:38:27 time: 0.6375 data_time: 0.0015 memory: 44138 loss: 0.3606 +2023/06/06 10:53:37 - mmengine - INFO - Epoch(train) [8][2600/3907] lr: 1.2572e-05 eta: 1:37:23 time: 0.6419 data_time: 0.0015 memory: 44138 loss: 0.3713 +2023/06/06 10:54:09 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr3e-5_20230606_052112 +2023/06/06 10:54:41 - mmengine - INFO - Epoch(train) [8][2700/3907] lr: 1.2518e-05 eta: 1:36:19 time: 0.6390 data_time: 0.0015 memory: 44138 loss: 0.3760 +2023/06/06 10:55:44 - mmengine - INFO - Epoch(train) [8][2800/3907] lr: 1.2465e-05 eta: 1:35:14 time: 0.6372 data_time: 0.0015 memory: 44138 loss: 0.3636 +2023/06/06 10:56:48 - mmengine - INFO - Epoch(train) [8][2900/3907] lr: 1.2412e-05 eta: 1:34:10 time: 0.6375 data_time: 0.0017 memory: 44138 loss: 0.3955 +2023/06/06 10:57:52 - mmengine - INFO - Epoch(train) [8][3000/3907] lr: 1.2360e-05 eta: 1:33:06 time: 0.6361 data_time: 0.0015 memory: 44138 loss: 0.3832 +2023/06/06 10:58:55 - mmengine - INFO - Epoch(train) [8][3100/3907] lr: 1.2309e-05 eta: 1:32:02 time: 0.6378 data_time: 0.0018 memory: 44138 loss: 0.3842 +2023/06/06 10:59:59 - mmengine - INFO - Epoch(train) [8][3200/3907] lr: 1.2257e-05 eta: 1:30:58 time: 0.6368 data_time: 0.0016 memory: 44138 loss: 0.3535 +2023/06/06 11:01:03 - mmengine - INFO - Epoch(train) [8][3300/3907] lr: 1.2207e-05 eta: 1:29:54 time: 0.6396 data_time: 0.0016 memory: 44138 loss: 0.3575 +2023/06/06 11:02:07 - mmengine - INFO - Epoch(train) [8][3400/3907] lr: 1.2157e-05 eta: 1:28:50 time: 0.6398 data_time: 0.0016 memory: 44138 loss: 0.3667 +2023/06/06 11:03:11 - mmengine - INFO - Epoch(train) [8][3500/3907] lr: 1.2107e-05 eta: 1:27:45 time: 0.6373 data_time: 0.0015 memory: 44138 loss: 0.3717 +2023/06/06 11:04:15 - mmengine - INFO - Epoch(train) [8][3600/3907] lr: 1.2058e-05 eta: 1:26:41 time: 0.6374 data_time: 0.0016 memory: 44138 loss: 0.3639 +2023/06/06 11:04:47 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr3e-5_20230606_052112 +2023/06/06 11:05:18 - mmengine - INFO - Epoch(train) [8][3700/3907] lr: 1.2009e-05 eta: 1:25:37 time: 0.6376 data_time: 0.0018 memory: 44138 loss: 0.3591 +2023/06/06 11:06:22 - mmengine - INFO - Epoch(train) [8][3800/3907] lr: 1.1961e-05 eta: 1:24:33 time: 0.6374 data_time: 0.0014 memory: 44138 loss: 0.3854 +2023/06/06 11:07:26 - mmengine - INFO - Epoch(train) [8][3900/3907] lr: 1.1914e-05 eta: 1:23:29 time: 0.6368 data_time: 0.0013 memory: 44138 loss: 0.3818 +2023/06/06 11:07:30 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr3e-5_20230606_052112 +2023/06/06 11:07:30 - mmengine - INFO - Saving checkpoint at 8 epochs +2023/06/06 11:09:11 - mmengine - INFO - Epoch(val) [8][57/57] accuracy/top1: 79.1831 single-label/precision_classwise: [73.12781524658203, 96.55404663085938] single-label/recall_classwise: [98.38394165039062, 55.60464859008789] single-label/f1-score_classwise: [83.89632415771484, 70.56913757324219] data_time: 0.0314 time: 1.2761 +2023/06/06 11:10:19 - mmengine - INFO - Epoch(train) [9][ 100/3907] lr: 1.1863e-05 eta: 1:22:21 time: 0.6391 data_time: 0.0017 memory: 44138 loss: 0.3623 +2023/06/06 11:11:22 - mmengine - INFO - Epoch(train) [9][ 200/3907] lr: 1.1817e-05 eta: 1:21:17 time: 0.6372 data_time: 0.0015 memory: 44138 loss: 0.3555 +2023/06/06 11:12:26 - mmengine - INFO - Epoch(train) [9][ 300/3907] lr: 1.1771e-05 eta: 1:20:13 time: 0.6372 data_time: 0.0017 memory: 44138 loss: 0.3709 +2023/06/06 11:13:30 - mmengine - INFO - Epoch(train) [9][ 400/3907] lr: 1.1725e-05 eta: 1:19:09 time: 0.6406 data_time: 0.0014 memory: 44138 loss: 0.3729 +2023/06/06 11:14:34 - mmengine - INFO - Epoch(train) [9][ 500/3907] lr: 1.1681e-05 eta: 1:18:05 time: 0.6392 data_time: 0.0015 memory: 44138 loss: 0.3687 +2023/06/06 11:15:38 - mmengine - INFO - Epoch(train) [9][ 600/3907] lr: 1.1636e-05 eta: 1:17:01 time: 0.6373 data_time: 0.0015 memory: 44138 loss: 0.3786 +2023/06/06 11:16:42 - mmengine - INFO - Epoch(train) [9][ 700/3907] lr: 1.1592e-05 eta: 1:15:56 time: 0.6362 data_time: 0.0014 memory: 44138 loss: 0.3668 +2023/06/06 11:17:10 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr3e-5_20230606_052112 +2023/06/06 11:17:46 - mmengine - INFO - Epoch(train) [9][ 800/3907] lr: 1.1549e-05 eta: 1:14:52 time: 0.6402 data_time: 0.0014 memory: 44138 loss: 0.3583 +2023/06/06 11:18:49 - mmengine - INFO - Epoch(train) [9][ 900/3907] lr: 1.1506e-05 eta: 1:13:48 time: 0.6367 data_time: 0.0014 memory: 44138 loss: 0.3795 +2023/06/06 11:19:53 - mmengine - INFO - Epoch(train) [9][1000/3907] lr: 1.1464e-05 eta: 1:12:44 time: 0.6377 data_time: 0.0015 memory: 44138 loss: 0.3826 +2023/06/06 11:20:57 - mmengine - INFO - Epoch(train) [9][1100/3907] lr: 1.1423e-05 eta: 1:11:40 time: 0.6372 data_time: 0.0014 memory: 44138 loss: 0.3926 +2023/06/06 11:22:01 - mmengine - INFO - Epoch(train) [9][1200/3907] lr: 1.1382e-05 eta: 1:10:36 time: 0.6379 data_time: 0.0015 memory: 44138 loss: 0.3772 +2023/06/06 11:23:05 - mmengine - INFO - Epoch(train) [9][1300/3907] lr: 1.1341e-05 eta: 1:09:32 time: 0.6378 data_time: 0.0018 memory: 44138 loss: 0.3500 +2023/06/06 11:24:09 - mmengine - INFO - Epoch(train) [9][1400/3907] lr: 1.1301e-05 eta: 1:08:28 time: 0.6372 data_time: 0.0015 memory: 44138 loss: 0.3698 +2023/06/06 11:25:13 - mmengine - INFO - Epoch(train) [9][1500/3907] lr: 1.1262e-05 eta: 1:07:24 time: 0.6398 data_time: 0.0015 memory: 44138 loss: 0.3839 +2023/06/06 11:26:17 - mmengine - INFO - Epoch(train) [9][1600/3907] lr: 1.1223e-05 eta: 1:06:20 time: 0.6369 data_time: 0.0015 memory: 44138 loss: 0.3808 +2023/06/06 11:27:20 - mmengine - INFO - Epoch(train) [9][1700/3907] lr: 1.1185e-05 eta: 1:05:15 time: 0.6371 data_time: 0.0014 memory: 44138 loss: 0.3679 +2023/06/06 11:27:49 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr3e-5_20230606_052112 +2023/06/06 11:28:24 - mmengine - INFO - Epoch(train) [9][1800/3907] lr: 1.1147e-05 eta: 1:04:11 time: 0.6390 data_time: 0.0015 memory: 44138 loss: 0.3671 +2023/06/06 11:29:28 - mmengine - INFO - Epoch(train) [9][1900/3907] lr: 1.1110e-05 eta: 1:03:07 time: 0.6382 data_time: 0.0015 memory: 44138 loss: 0.3607 +2023/06/06 11:30:32 - mmengine - INFO - Epoch(train) [9][2000/3907] lr: 1.1073e-05 eta: 1:02:03 time: 0.6375 data_time: 0.0015 memory: 44138 loss: 0.3539 +2023/06/06 11:31:36 - mmengine - INFO - Epoch(train) [9][2100/3907] lr: 1.1037e-05 eta: 1:00:59 time: 0.6386 data_time: 0.0014 memory: 44138 loss: 0.3990 +2023/06/06 11:32:40 - mmengine - INFO - Epoch(train) [9][2200/3907] lr: 1.1002e-05 eta: 0:59:55 time: 0.6386 data_time: 0.0015 memory: 44138 loss: 0.3768 +2023/06/06 11:33:44 - mmengine - INFO - Epoch(train) [9][2300/3907] lr: 1.0967e-05 eta: 0:58:51 time: 0.6383 data_time: 0.0014 memory: 44138 loss: 0.3703 +2023/06/06 11:34:47 - mmengine - INFO - Epoch(train) [9][2400/3907] lr: 1.0933e-05 eta: 0:57:47 time: 0.6401 data_time: 0.0014 memory: 44138 loss: 0.3780 +2023/06/06 11:35:51 - mmengine - INFO - Epoch(train) [9][2500/3907] lr: 1.0899e-05 eta: 0:56:43 time: 0.6357 data_time: 0.0016 memory: 44138 loss: 0.3759 +2023/06/06 11:36:55 - mmengine - INFO - Epoch(train) [9][2600/3907] lr: 1.0866e-05 eta: 0:55:39 time: 0.6368 data_time: 0.0015 memory: 44138 loss: 0.3702 +2023/06/06 11:37:59 - mmengine - INFO - Epoch(train) [9][2700/3907] lr: 1.0834e-05 eta: 0:54:35 time: 0.6379 data_time: 0.0014 memory: 44138 loss: 0.3471 +2023/06/06 11:38:27 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr3e-5_20230606_052112 +2023/06/06 11:39:03 - mmengine - INFO - Epoch(train) [9][2800/3907] lr: 1.0802e-05 eta: 0:53:31 time: 0.6378 data_time: 0.0015 memory: 44138 loss: 0.4077 +2023/06/06 11:40:07 - mmengine - INFO - Epoch(train) [9][2900/3907] lr: 1.0771e-05 eta: 0:52:27 time: 0.6378 data_time: 0.0017 memory: 44138 loss: 0.3521 +2023/06/06 11:41:10 - mmengine - INFO - Epoch(train) [9][3000/3907] lr: 1.0740e-05 eta: 0:51:22 time: 0.6373 data_time: 0.0014 memory: 44138 loss: 0.4087 +2023/06/06 11:42:14 - mmengine - INFO - Epoch(train) [9][3100/3907] lr: 1.0710e-05 eta: 0:50:18 time: 0.6535 data_time: 0.0015 memory: 44138 loss: 0.3581 +2023/06/06 11:43:18 - mmengine - INFO - Epoch(train) [9][3200/3907] lr: 1.0681e-05 eta: 0:49:14 time: 0.6393 data_time: 0.0014 memory: 44138 loss: 0.3922 +2023/06/06 11:44:22 - mmengine - INFO - Epoch(train) [9][3300/3907] lr: 1.0652e-05 eta: 0:48:10 time: 0.6416 data_time: 0.0016 memory: 44138 loss: 0.3732 +2023/06/06 11:45:26 - mmengine - INFO - Epoch(train) [9][3400/3907] lr: 1.0624e-05 eta: 0:47:06 time: 0.6368 data_time: 0.0015 memory: 44138 loss: 0.3640 +2023/06/06 11:46:29 - mmengine - INFO - Epoch(train) [9][3500/3907] lr: 1.0596e-05 eta: 0:46:02 time: 0.6364 data_time: 0.0016 memory: 44138 loss: 0.3692 +2023/06/06 11:47:33 - mmengine - INFO - Epoch(train) [9][3600/3907] lr: 1.0569e-05 eta: 0:44:58 time: 0.6373 data_time: 0.0014 memory: 44138 loss: 0.3812 +2023/06/06 11:48:37 - mmengine - INFO - Epoch(train) [9][3700/3907] lr: 1.0542e-05 eta: 0:43:54 time: 0.6377 data_time: 0.0014 memory: 44138 loss: 0.3598 +2023/06/06 11:49:05 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr3e-5_20230606_052112 +2023/06/06 11:49:41 - mmengine - INFO - Epoch(train) [9][3800/3907] lr: 1.0517e-05 eta: 0:42:50 time: 0.6370 data_time: 0.0015 memory: 44138 loss: 0.3529 +2023/06/06 11:50:44 - mmengine - INFO - Epoch(train) [9][3900/3907] lr: 1.0491e-05 eta: 0:41:46 time: 0.6387 data_time: 0.0014 memory: 44138 loss: 0.3287 +2023/06/06 11:50:48 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr3e-5_20230606_052112 +2023/06/06 11:50:48 - mmengine - INFO - Saving checkpoint at 9 epochs +2023/06/06 11:52:32 - mmengine - INFO - Epoch(val) [9][57/57] accuracy/top1: 79.1622 single-label/precision_classwise: [73.08341217041016, 96.68963623046875] single-label/recall_classwise: [98.45337677001953, 55.472869873046875] single-label/f1-score_classwise: [83.8923110961914, 70.49898529052734] data_time: 0.0324 time: 1.2746 +2023/06/06 11:53:39 - mmengine - INFO - Epoch(train) [10][ 100/3907] lr: 1.0465e-05 eta: 0:40:38 time: 0.6357 data_time: 0.0015 memory: 44138 loss: 0.3764 +2023/06/06 11:54:43 - mmengine - INFO - Epoch(train) [10][ 200/3907] lr: 1.0441e-05 eta: 0:39:34 time: 0.6366 data_time: 0.0015 memory: 44138 loss: 0.3857 +2023/06/06 11:55:47 - mmengine - INFO - Epoch(train) [10][ 300/3907] lr: 1.0418e-05 eta: 0:38:29 time: 0.6388 data_time: 0.0015 memory: 44138 loss: 0.3529 +2023/06/06 11:56:50 - mmengine - INFO - Epoch(train) [10][ 400/3907] lr: 1.0395e-05 eta: 0:37:25 time: 0.6385 data_time: 0.0014 memory: 44138 loss: 0.3618 +2023/06/06 11:57:54 - mmengine - INFO - Epoch(train) [10][ 500/3907] lr: 1.0373e-05 eta: 0:36:21 time: 0.6373 data_time: 0.0015 memory: 44138 loss: 0.3934 +2023/06/06 11:58:58 - mmengine - INFO - Epoch(train) [10][ 600/3907] lr: 1.0352e-05 eta: 0:35:17 time: 0.6389 data_time: 0.0015 memory: 44138 loss: 0.3513 +2023/06/06 12:00:02 - mmengine - INFO - Epoch(train) [10][ 700/3907] lr: 1.0331e-05 eta: 0:34:13 time: 0.6370 data_time: 0.0015 memory: 44138 loss: 0.3731 +2023/06/06 12:01:06 - mmengine - INFO - Epoch(train) [10][ 800/3907] lr: 1.0311e-05 eta: 0:33:09 time: 0.6371 data_time: 0.0015 memory: 44138 loss: 0.3761 +2023/06/06 12:01:29 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr3e-5_20230606_052112 +2023/06/06 12:02:10 - mmengine - INFO - Epoch(train) [10][ 900/3907] lr: 1.0291e-05 eta: 0:32:05 time: 0.6375 data_time: 0.0016 memory: 44138 loss: 0.3635 +2023/06/06 12:03:13 - mmengine - INFO - Epoch(train) [10][1000/3907] lr: 1.0272e-05 eta: 0:31:01 time: 0.6390 data_time: 0.0014 memory: 44138 loss: 0.3746 +2023/06/06 12:04:17 - mmengine - INFO - Epoch(train) [10][1100/3907] lr: 1.0254e-05 eta: 0:29:57 time: 0.6390 data_time: 0.0016 memory: 44138 loss: 0.3888 +2023/06/06 12:05:21 - mmengine - INFO - Epoch(train) [10][1200/3907] lr: 1.0236e-05 eta: 0:28:53 time: 0.6389 data_time: 0.0014 memory: 44138 loss: 0.3702 +2023/06/06 12:06:25 - mmengine - INFO - Epoch(train) [10][1300/3907] lr: 1.0219e-05 eta: 0:27:49 time: 0.6363 data_time: 0.0014 memory: 44138 loss: 0.3894 +2023/06/06 12:07:29 - mmengine - INFO - Epoch(train) [10][1400/3907] lr: 1.0203e-05 eta: 0:26:45 time: 0.6375 data_time: 0.0016 memory: 44138 loss: 0.3747 +2023/06/06 12:08:33 - mmengine - INFO - Epoch(train) [10][1500/3907] lr: 1.0187e-05 eta: 0:25:41 time: 0.6372 data_time: 0.0014 memory: 44138 loss: 0.3911 +2023/06/06 12:09:37 - mmengine - INFO - Epoch(train) [10][1600/3907] lr: 1.0172e-05 eta: 0:24:37 time: 0.6711 data_time: 0.0367 memory: 44138 loss: 0.3456 +2023/06/06 12:10:45 - mmengine - INFO - Epoch(train) [10][1700/3907] lr: 1.0157e-05 eta: 0:23:33 time: 0.6376 data_time: 0.0014 memory: 44138 loss: 0.3497 +2023/06/06 12:12:01 - mmengine - INFO - Epoch(train) [10][1800/3907] lr: 1.0143e-05 eta: 0:22:30 time: 0.6383 data_time: 0.0014 memory: 44138 loss: 0.3668 +2023/06/06 12:12:27 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr3e-5_20230606_052112 +2023/06/06 12:13:08 - mmengine - INFO - Epoch(train) [10][1900/3907] lr: 1.0130e-05 eta: 0:21:26 time: 0.6493 data_time: 0.0014 memory: 44138 loss: 0.3871 +2023/06/06 12:14:13 - mmengine - INFO - Epoch(train) [10][2000/3907] lr: 1.0117e-05 eta: 0:20:22 time: 0.6374 data_time: 0.0014 memory: 44138 loss: 0.3554 +2023/06/06 12:15:17 - mmengine - INFO - Epoch(train) [10][2100/3907] lr: 1.0105e-05 eta: 0:19:18 time: 0.6393 data_time: 0.0014 memory: 44138 loss: 0.3578 +2023/06/06 12:16:21 - mmengine - INFO - Epoch(train) [10][2200/3907] lr: 1.0094e-05 eta: 0:18:13 time: 0.6389 data_time: 0.0016 memory: 44138 loss: 0.3811 +2023/06/06 12:17:25 - mmengine - INFO - Epoch(train) [10][2300/3907] lr: 1.0083e-05 eta: 0:17:09 time: 0.6388 data_time: 0.0015 memory: 44138 loss: 0.3741 +2023/06/06 12:18:29 - mmengine - INFO - Epoch(train) [10][2400/3907] lr: 1.0073e-05 eta: 0:16:05 time: 0.6386 data_time: 0.0016 memory: 44138 loss: 0.3670 +2023/06/06 12:19:33 - mmengine - INFO - Epoch(train) [10][2500/3907] lr: 1.0064e-05 eta: 0:15:01 time: 0.6420 data_time: 0.0015 memory: 44138 loss: 0.3903 +2023/06/06 12:20:37 - mmengine - INFO - Epoch(train) [10][2600/3907] lr: 1.0055e-05 eta: 0:13:57 time: 0.6424 data_time: 0.0017 memory: 44138 loss: 0.3787 +2023/06/06 12:21:41 - mmengine - INFO - Epoch(train) [10][2700/3907] lr: 1.0047e-05 eta: 0:12:53 time: 0.6391 data_time: 0.0015 memory: 44138 loss: 0.3845 +2023/06/06 12:22:45 - mmengine - INFO - Epoch(train) [10][2800/3907] lr: 1.0040e-05 eta: 0:11:49 time: 0.6374 data_time: 0.0015 memory: 44138 loss: 0.3589 +2023/06/06 12:23:09 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr3e-5_20230606_052112 +2023/06/06 12:23:49 - mmengine - INFO - Epoch(train) [10][2900/3907] lr: 1.0033e-05 eta: 0:10:45 time: 0.6393 data_time: 0.0015 memory: 44138 loss: 0.3925 +2023/06/06 12:24:53 - mmengine - INFO - Epoch(train) [10][3000/3907] lr: 1.0027e-05 eta: 0:09:41 time: 0.6386 data_time: 0.0016 memory: 44138 loss: 0.3934 +2023/06/06 12:25:56 - mmengine - INFO - Epoch(train) [10][3100/3907] lr: 1.0021e-05 eta: 0:08:37 time: 0.6383 data_time: 0.0014 memory: 44138 loss: 0.3612 +2023/06/06 12:27:00 - mmengine - INFO - Epoch(train) [10][3200/3907] lr: 1.0016e-05 eta: 0:07:33 time: 0.6386 data_time: 0.0015 memory: 44138 loss: 0.3312 +2023/06/06 12:28:04 - mmengine - INFO - Epoch(train) [10][3300/3907] lr: 1.0012e-05 eta: 0:06:28 time: 0.6370 data_time: 0.0014 memory: 44138 loss: 0.3677 +2023/06/06 12:29:08 - mmengine - INFO - Epoch(train) [10][3400/3907] lr: 1.0008e-05 eta: 0:05:24 time: 0.6383 data_time: 0.0015 memory: 44138 loss: 0.3790 +2023/06/06 12:30:12 - mmengine - INFO - Epoch(train) [10][3500/3907] lr: 1.0005e-05 eta: 0:04:20 time: 0.6404 data_time: 0.0017 memory: 44138 loss: 0.3797 +2023/06/06 12:31:16 - mmengine - INFO - Epoch(train) [10][3600/3907] lr: 1.0003e-05 eta: 0:03:16 time: 0.6484 data_time: 0.0019 memory: 44138 loss: 0.3576 +2023/06/06 12:32:20 - mmengine - INFO - Epoch(train) [10][3700/3907] lr: 1.0001e-05 eta: 0:02:12 time: 0.6407 data_time: 0.0016 memory: 44138 loss: 0.3869 +2023/06/06 12:33:24 - mmengine - INFO - Epoch(train) [10][3800/3907] lr: 1.0000e-05 eta: 0:01:08 time: 0.6391 data_time: 0.0015 memory: 44138 loss: 0.3512 +2023/06/06 12:33:48 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr3e-5_20230606_052112 +2023/06/06 12:34:28 - mmengine - INFO - Epoch(train) [10][3900/3907] lr: 1.0000e-05 eta: 0:00:04 time: 0.6382 data_time: 0.0013 memory: 44138 loss: 0.3589 +2023/06/06 12:34:32 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_sdv2_lr3e-5_20230606_052112 +2023/06/06 12:34:32 - mmengine - INFO - Saving checkpoint at 10 epochs +2023/06/06 12:36:13 - mmengine - INFO - Epoch(val) [10][57/57] accuracy/top1: 78.6646 single-label/precision_classwise: [72.5381851196289, 96.98694610595703] single-label/recall_classwise: [98.63013458251953, 54.14728546142578] single-label/f1-score_classwise: [83.59550476074219, 69.49557495117188] data_time: 0.0317 time: 1.2740 diff --git a/clip_large_pretrain_4x256_sdv2_lr3e-5/20230606_052112/vis_data/20230606_052112.json b/clip_large_pretrain_4x256_sdv2_lr3e-5/20230606_052112/vis_data/20230606_052112.json new file mode 100644 index 0000000000000000000000000000000000000000..519c3486db56bfc3ded9e06a8dc5dd87e2bd9579 --- /dev/null +++ b/clip_large_pretrain_4x256_sdv2_lr3e-5/20230606_052112/vis_data/20230606_052112.json @@ -0,0 +1,400 @@ +{"lr": 2.999968315196157e-05, "data_time": 0.0017328262329101562, "loss": 0.5820598602294922, "time": 0.6326795101165772, "epoch": 1, "memory": 44139, "step": 100} +{"lr": 2.999871979413257e-05, "data_time": 0.0017274379730224609, "loss": 0.5244008064270019, "time": 0.6465194940567016, "epoch": 1, "memory": 44139, "step": 200} +{"lr": 2.999710995647179e-05, "data_time": 0.0015012741088867188, "loss": 0.5025475025177002, "time": 0.6346629619598388, "epoch": 1, "memory": 44139, "step": 300} +{"lr": 2.9994853743065302e-05, "data_time": 0.001559281349182129, "loss": 0.5019857287406921, "time": 0.6351841926574707, "epoch": 1, "memory": 44139, "step": 400} +{"lr": 2.9991951299791448e-05, "data_time": 0.001495051383972168, "loss": 0.4892808198928833, "time": 0.6358315467834472, "epoch": 1, "memory": 44139, "step": 500} +{"lr": 2.9988402814311378e-05, "data_time": 0.0014707803726196288, "loss": 0.4940012454986572, "time": 0.6359200716018677, "epoch": 1, "memory": 44139, "step": 600} +{"lr": 2.998420851605688e-05, "data_time": 0.0014953374862670898, "loss": 0.4516049861907959, "time": 0.6355729103088379, "epoch": 1, "memory": 44139, "step": 700} +{"lr": 2.99793686762156e-05, "data_time": 0.0016302108764648438, "loss": 0.46528034210205077, "time": 0.6352574586868286, "epoch": 1, "memory": 44139, "step": 800} +{"lr": 2.997388360771344e-05, "data_time": 0.0015640497207641602, "loss": 0.45623364448547366, "time": 0.6363352298736572, "epoch": 1, "memory": 44139, "step": 900} +{"lr": 2.9967753665194455e-05, "data_time": 0.0014910221099853516, "loss": 0.44340091943740845, "time": 0.6351892471313476, "epoch": 1, "memory": 44139, "step": 1000} +{"lr": 2.996097924499781e-05, "data_time": 0.0014334678649902343, "loss": 0.45766860246658325, "time": 0.6361861228942871, "epoch": 1, "memory": 44139, "step": 1100} +{"lr": 2.9953560785132187e-05, "data_time": 0.0016579389572143554, "loss": 0.45582705140113833, "time": 0.6349462747573853, "epoch": 1, "memory": 44139, "step": 1200} +{"lr": 2.994549876524743e-05, "data_time": 0.001495194435119629, "loss": 0.466318154335022, "time": 0.634421706199646, "epoch": 1, "memory": 44139, "step": 1300} +{"lr": 2.99367937066036e-05, "data_time": 0.0015588760375976562, "loss": 0.4483098328113556, "time": 0.6346314668655395, "epoch": 1, "memory": 44139, "step": 1400} +{"lr": 2.9927446172037264e-05, "data_time": 0.0014845848083496094, "loss": 0.44086949825286864, "time": 0.6364420890808106, "epoch": 1, "memory": 44139, "step": 1500} +{"lr": 2.9917456765925056e-05, "data_time": 0.0015438556671142577, "loss": 0.4376753866672516, "time": 0.6369487285614014, "epoch": 1, "memory": 44139, "step": 1600} +{"lr": 2.990682613414459e-05, "data_time": 0.001448678970336914, "loss": 0.4680232286453247, "time": 0.6341166734695435, "epoch": 1, "memory": 44139, "step": 1700} +{"lr": 2.989555496403281e-05, "data_time": 0.0014559268951416016, "loss": 0.436040461063385, "time": 0.6355781078338623, "epoch": 1, "memory": 44139, "step": 1800} +{"lr": 2.9883643984341356e-05, "data_time": 0.001564788818359375, "loss": 0.4360331892967224, "time": 0.6353567123413086, "epoch": 1, "memory": 44139, "step": 1900} +{"lr": 2.987109396518962e-05, "data_time": 0.0015111923217773437, "loss": 0.4762925148010254, "time": 0.6346500396728516, "epoch": 1, "memory": 44139, "step": 2000} +{"lr": 2.9857905718015028e-05, "data_time": 0.0014391422271728515, "loss": 0.4315656542778015, "time": 0.6356466770172119, "epoch": 1, "memory": 44139, "step": 2100} +{"lr": 2.984408009552021e-05, "data_time": 0.0015337705612182618, "loss": 0.4738268882036209, "time": 0.6370855808258057, "epoch": 1, "memory": 44139, "step": 2200} +{"lr": 2.9829617991618262e-05, "data_time": 0.0015912532806396484, "loss": 0.4564800500869751, "time": 0.636294960975647, "epoch": 1, "memory": 44139, "step": 2300} +{"lr": 2.981452034137478e-05, "data_time": 0.0015359640121459961, "loss": 0.4229035317897797, "time": 0.6416321277618409, "epoch": 1, "memory": 44139, "step": 2400} +{"lr": 2.979878812094729e-05, "data_time": 0.0013695240020751953, "loss": 0.45235815048217776, "time": 0.6351548194885254, "epoch": 1, "memory": 44139, "step": 2500} +{"lr": 2.978242234752238e-05, "data_time": 0.0015182256698608398, "loss": 0.42996309995651244, "time": 0.6344455242156982, "epoch": 1, "memory": 44139, "step": 2600} +{"lr": 2.9765424079249748e-05, "data_time": 0.001532292366027832, "loss": 0.4235107421875, "time": 0.6349280834197998, "epoch": 1, "memory": 44139, "step": 2700} +{"lr": 2.9747794415173894e-05, "data_time": 0.001432490348815918, "loss": 0.443260133266449, "time": 0.6347655296325684, "epoch": 1, "memory": 44139, "step": 2800} +{"lr": 2.972953449516291e-05, "data_time": 0.001365804672241211, "loss": 0.45572861433029177, "time": 0.6348637580871582, "epoch": 1, "memory": 44139, "step": 2900} +{"lr": 2.9710645499835012e-05, "data_time": 0.0014687538146972655, "loss": 0.4245560824871063, "time": 0.6355278491973877, "epoch": 1, "memory": 44139, "step": 3000} +{"lr": 2.969112865048194e-05, "data_time": 0.0014769077301025391, "loss": 0.4595903933048248, "time": 0.6352854490280151, "epoch": 1, "memory": 44139, "step": 3100} +{"lr": 2.9670985208990228e-05, "data_time": 0.001611328125, "loss": 0.45638911724090575, "time": 0.6354909420013428, "epoch": 1, "memory": 44139, "step": 3200} +{"lr": 2.9650216477759467e-05, "data_time": 0.0015012741088867188, "loss": 0.43623468577861785, "time": 0.6347058534622192, "epoch": 1, "memory": 44139, "step": 3300} +{"lr": 2.9628823799618136e-05, "data_time": 0.0016749143600463868, "loss": 0.4210712373256683, "time": 0.635877537727356, "epoch": 1, "memory": 44139, "step": 3400} +{"lr": 2.960680855773683e-05, "data_time": 0.0014918088912963868, "loss": 0.43551710844039915, "time": 0.6353757381439209, "epoch": 1, "memory": 44139, "step": 3500} +{"lr": 2.958417217553874e-05, "data_time": 0.0016278743743896485, "loss": 0.44734391272068025, "time": 0.6348343610763549, "epoch": 1, "memory": 44139, "step": 3600} +{"lr": 2.956091611660774e-05, "data_time": 0.001421642303466797, "loss": 0.4362662613391876, "time": 0.6364853382110596, "epoch": 1, "memory": 44139, "step": 3700} +{"lr": 2.9537041884593564e-05, "data_time": 0.0015681982040405273, "loss": 0.433257931470871, "time": 0.6355493545532227, "epoch": 1, "memory": 44139, "step": 3800} +{"lr": 2.9512551023114816e-05, "data_time": 0.0012206077575683595, "loss": 0.41935579776763915, "time": 0.6346648693084717, "epoch": 1, "memory": 44139, "step": 3900} +{"accuracy/top1": 80.07376098632812, "data_time": 0.04794620212755705, "time": 1.342139587067721, "step": 1} +{"lr": 2.9485664710372417e-05, "data_time": 0.0015699625015258788, "loss": 0.46139551103115084, "time": 0.6367982387542724, "epoch": 2, "memory": 44139, "step": 4007} +{"lr": 2.9459902502419427e-05, "data_time": 0.0015330553054809571, "loss": 0.41834096908569335, "time": 0.6372418403625488, "epoch": 2, "memory": 44138, "step": 4107} +{"lr": 2.9433528652545812e-05, "data_time": 0.0013655662536621095, "loss": 0.4502277851104736, "time": 0.6382653474807739, "epoch": 2, "memory": 44138, "step": 4207} +{"lr": 2.9406544865986134e-05, "data_time": 0.0013972043991088868, "loss": 0.45304155349731445, "time": 0.6351974248886109, "epoch": 2, "memory": 44138, "step": 4307} +{"lr": 2.9378952887411096e-05, "data_time": 0.0014516353607177735, "loss": 0.4170908659696579, "time": 0.6357728958129882, "epoch": 2, "memory": 44138, "step": 4407} +{"lr": 2.935075450081489e-05, "data_time": 0.0014037132263183595, "loss": 0.42441895306110383, "time": 0.6353183031082154, "epoch": 2, "memory": 44138, "step": 4507} +{"lr": 2.9321951529399665e-05, "data_time": 0.0015067815780639648, "loss": 0.40530370473861693, "time": 0.6356248617172241, "epoch": 2, "memory": 44138, "step": 4607} +{"lr": 2.9292545835457945e-05, "data_time": 0.001408529281616211, "loss": 0.4169813871383667, "time": 0.6364418983459472, "epoch": 2, "memory": 44138, "step": 4707} +{"lr": 2.9262539320251868e-05, "data_time": 0.0013746023178100586, "loss": 0.4180146545171738, "time": 0.6357547760009765, "epoch": 2, "memory": 44138, "step": 4807} +{"lr": 2.923193392389054e-05, "data_time": 0.0016022205352783203, "loss": 0.40583510100841524, "time": 0.6371466875076294, "epoch": 2, "memory": 44138, "step": 4907} +{"lr": 2.9200731625204435e-05, "data_time": 0.0016779899597167969, "loss": 0.41511437594890593, "time": 0.6361659526824951, "epoch": 2, "memory": 44138, "step": 5007} +{"lr": 2.9168934441617457e-05, "data_time": 0.0015182256698608398, "loss": 0.4329532742500305, "time": 0.6365081787109375, "epoch": 2, "memory": 44138, "step": 5107} +{"lr": 2.9136544429016667e-05, "data_time": 0.0014085769653320312, "loss": 0.4559467166662216, "time": 0.6362128734588623, "epoch": 2, "memory": 44138, "step": 5207} +{"lr": 2.9103563681619084e-05, "data_time": 0.0014522075653076172, "loss": 0.41591658890247346, "time": 0.6360562562942504, "epoch": 2, "memory": 44138, "step": 5307} +{"lr": 2.9069994331836577e-05, "data_time": 0.0014405965805053711, "loss": 0.4258942365646362, "time": 0.6373616456985474, "epoch": 2, "memory": 44138, "step": 5407} +{"lr": 2.9035838550137736e-05, "data_time": 0.0015010833740234375, "loss": 0.415695384144783, "time": 0.6360781908035278, "epoch": 2, "memory": 44138, "step": 5507} +{"lr": 2.9001098544907622e-05, "data_time": 0.0015009641647338867, "loss": 0.42156079411506653, "time": 0.6378882646560669, "epoch": 2, "memory": 44138, "step": 5607} +{"lr": 2.8965776562305154e-05, "data_time": 0.0014098405838012696, "loss": 0.4137383341789246, "time": 0.6360904455184937, "epoch": 2, "memory": 44138, "step": 5707} +{"lr": 2.892987488611766e-05, "data_time": 0.0013931989669799805, "loss": 0.4191013127565384, "time": 0.635969614982605, "epoch": 2, "memory": 44138, "step": 5807} +{"lr": 2.88933958376132e-05, "data_time": 0.0015510797500610351, "loss": 0.42429515421390535, "time": 0.6362807750701904, "epoch": 2, "memory": 44138, "step": 5907} +{"lr": 2.8856341775390632e-05, "data_time": 0.001604008674621582, "loss": 0.4397385746240616, "time": 0.6360028743743896, "epoch": 2, "memory": 44138, "step": 6007} +{"lr": 2.8818715095227025e-05, "data_time": 0.0014252901077270509, "loss": 0.3885278284549713, "time": 0.6349270582199097, "epoch": 2, "memory": 44138, "step": 6107} +{"lr": 2.878051822992295e-05, "data_time": 0.0013510465621948242, "loss": 0.4366811364889145, "time": 0.6370023488998413, "epoch": 2, "memory": 44138, "step": 6207} +{"lr": 2.874175364914476e-05, "data_time": 0.0014041900634765626, "loss": 0.4364030212163925, "time": 0.6357920408248902, "epoch": 2, "memory": 44138, "step": 6307} +{"lr": 2.870242385926531e-05, "data_time": 0.0014720439910888671, "loss": 0.4201664745807648, "time": 0.6351218461990357, "epoch": 2, "memory": 44138, "step": 6407} +{"lr": 2.866253140320176e-05, "data_time": 0.0015051126480102538, "loss": 0.4343326359987259, "time": 0.6362585067749024, "epoch": 2, "memory": 44138, "step": 6507} +{"lr": 2.862207886025111e-05, "data_time": 0.0014474868774414062, "loss": 0.4557327330112457, "time": 0.6358387231826782, "epoch": 2, "memory": 44138, "step": 6607} +{"lr": 2.8581068845923533e-05, "data_time": 0.0014994144439697266, "loss": 0.41012815833091737, "time": 0.6372246503829956, "epoch": 2, "memory": 44138, "step": 6707} +{"lr": 2.8539504011773207e-05, "data_time": 0.0014214038848876954, "loss": 0.42677571177482604, "time": 0.6366636514663696, "epoch": 2, "memory": 44138, "step": 6807} +{"lr": 2.849738704522689e-05, "data_time": 0.0014622211456298828, "loss": 0.4198896884918213, "time": 0.6367005825042724, "epoch": 2, "memory": 44138, "step": 6907} +{"lr": 2.845472066941013e-05, "data_time": 0.001542377471923828, "loss": 0.42949950098991396, "time": 0.6354840755462646, "epoch": 2, "memory": 44138, "step": 7007} +{"lr": 2.8411507642971194e-05, "data_time": 0.001567983627319336, "loss": 0.42264110743999483, "time": 0.6358386516571045, "epoch": 2, "memory": 44138, "step": 7107} +{"lr": 2.836775075990287e-05, "data_time": 0.0014949560165405274, "loss": 0.41218724846839905, "time": 0.6354932069778443, "epoch": 2, "memory": 44138, "step": 7207} +{"lr": 2.8323452849361518e-05, "data_time": 0.0015042543411254883, "loss": 0.4387490481138229, "time": 0.6351249694824219, "epoch": 2, "memory": 44138, "step": 7307} +{"lr": 2.827861677548442e-05, "data_time": 0.0014189481735229492, "loss": 0.43870543539524076, "time": 0.6362674236297607, "epoch": 2, "memory": 44138, "step": 7407} +{"lr": 2.8233245437204413e-05, "data_time": 0.0016158103942871093, "loss": 0.4182025045156479, "time": 0.636237359046936, "epoch": 2, "memory": 44138, "step": 7507} +{"lr": 2.8187341768062594e-05, "data_time": 0.0014983892440795898, "loss": 0.43157645463943484, "time": 0.6355327367782593, "epoch": 2, "memory": 44138, "step": 7607} +{"lr": 2.8140908736018534e-05, "data_time": 0.001468658447265625, "loss": 0.3919978946447372, "time": 0.6385063171386719, "epoch": 2, "memory": 44138, "step": 7707} +{"lr": 2.8093949343258448e-05, "data_time": 0.001326274871826172, "loss": 0.3885378032922745, "time": 0.6353389501571656, "epoch": 2, "memory": 44138, "step": 7807} +{"accuracy/top1": 81.8169174194336, "data_time": 0.03494371216872643, "time": 1.2779911715408852, "step": 2} +{"lr": 2.8043123316419856e-05, "data_time": 0.0024315595626831056, "loss": 0.41362959444522857, "time": 0.6347774267196655, "epoch": 3, "memory": 44138, "step": 7914} +{"lr": 2.799508404275313e-05, "data_time": 0.0017800331115722656, "loss": 0.39586409032344816, "time": 0.6360813617706299, "epoch": 3, "memory": 44138, "step": 8014} +{"lr": 2.7946527836849776e-05, "data_time": 0.001385068893432617, "loss": 0.39395827054977417, "time": 0.6360595703125, "epoch": 3, "memory": 44138, "step": 8114} +{"lr": 2.7897457838172418e-05, "data_time": 0.0017013788223266602, "loss": 0.4298724889755249, "time": 0.6363627433776855, "epoch": 3, "memory": 44138, "step": 8214} +{"lr": 2.784787721940371e-05, "data_time": 0.0015061378479003906, "loss": 0.41413705348968505, "time": 0.6381935119628906, "epoch": 3, "memory": 44138, "step": 8314} +{"lr": 2.7797789186241074e-05, "data_time": 0.0017046451568603516, "loss": 0.41770114898681643, "time": 0.6368783235549926, "epoch": 3, "memory": 44138, "step": 8414} +{"lr": 2.774719697718944e-05, "data_time": 0.0015776872634887696, "loss": 0.42050252556800843, "time": 0.6355265855789185, "epoch": 3, "memory": 44138, "step": 8514} +{"lr": 2.7696103863351812e-05, "data_time": 0.0016794443130493165, "loss": 0.40208789706230164, "time": 0.6353798866271972, "epoch": 3, "memory": 44138, "step": 8614} +{"lr": 2.7644513148217896e-05, "data_time": 0.0016884088516235351, "loss": 0.3941322654485703, "time": 0.637868094444275, "epoch": 3, "memory": 44138, "step": 8714} +{"lr": 2.759242816745033e-05, "data_time": 0.0016788959503173829, "loss": 0.3778026938438416, "time": 0.6356910943984986, "epoch": 3, "memory": 44138, "step": 8814} +{"lr": 2.7539852288669356e-05, "data_time": 0.001613783836364746, "loss": 0.4226143151521683, "time": 0.6357138156890869, "epoch": 3, "memory": 44138, "step": 8914} +{"lr": 2.7486788911234612e-05, "data_time": 0.0015007734298706054, "loss": 0.41931558549404147, "time": 0.6348323106765748, "epoch": 3, "memory": 44138, "step": 9014} +{"lr": 2.7433241466025614e-05, "data_time": 0.0013763904571533203, "loss": 0.3943664938211441, "time": 0.6383355617523193, "epoch": 3, "memory": 44138, "step": 9114} +{"lr": 2.7379213415219968e-05, "data_time": 0.0014757394790649414, "loss": 0.4155552923679352, "time": 0.6354357957839966, "epoch": 3, "memory": 44138, "step": 9214} +{"lr": 2.732470825206939e-05, "data_time": 0.001357889175415039, "loss": 0.39439934194087983, "time": 0.6365950584411622, "epoch": 3, "memory": 44138, "step": 9314} +{"lr": 2.7269729500673918e-05, "data_time": 0.001461195945739746, "loss": 0.3985852748155594, "time": 0.6367216348648072, "epoch": 3, "memory": 44138, "step": 9414} +{"lr": 2.7214280715754e-05, "data_time": 0.0014960050582885742, "loss": 0.3905987113714218, "time": 0.6351579666137696, "epoch": 3, "memory": 44138, "step": 9514} +{"lr": 2.7158365482420722e-05, "data_time": 0.0014636516571044922, "loss": 0.37036815881729124, "time": 0.6367456912994385, "epoch": 3, "memory": 44138, "step": 9614} +{"lr": 2.7101987415944003e-05, "data_time": 0.00142214298248291, "loss": 0.39271018505096433, "time": 0.635669493675232, "epoch": 3, "memory": 44138, "step": 9714} +{"lr": 2.7045150161518795e-05, "data_time": 0.0013616561889648437, "loss": 0.42946577072143555, "time": 0.6355520248413086, "epoch": 3, "memory": 44138, "step": 9814} +{"lr": 2.6987857394029402e-05, "data_time": 0.001372051239013672, "loss": 0.39274133145809176, "time": 0.6365905284881592, "epoch": 3, "memory": 44138, "step": 9914} +{"lr": 2.6930112817811945e-05, "data_time": 0.0014690399169921876, "loss": 0.38676917552948, "time": 0.6366588115692139, "epoch": 3, "memory": 44138, "step": 10014} +{"lr": 2.6871920166414833e-05, "data_time": 0.0014314889907836915, "loss": 0.40503534972667693, "time": 0.6373430013656616, "epoch": 3, "memory": 44138, "step": 10114} +{"lr": 2.6813283202357177e-05, "data_time": 0.0015382051467895507, "loss": 0.4070036053657532, "time": 0.6353307485580444, "epoch": 3, "memory": 44138, "step": 10214} +{"lr": 2.675420571688591e-05, "data_time": 0.0015157222747802734, "loss": 0.4014215022325516, "time": 0.6362082958221436, "epoch": 3, "memory": 44138, "step": 10314} +{"lr": 2.6694691529730227e-05, "data_time": 0.0015067100524902345, "loss": 0.4042677044868469, "time": 0.6353816986083984, "epoch": 3, "memory": 44138, "step": 10414} +{"lr": 2.6634744488854987e-05, "data_time": 0.0015247106552124024, "loss": 0.3955463647842407, "time": 0.6370667695999146, "epoch": 3, "memory": 44138, "step": 10514} +{"lr": 2.657436847021168e-05, "data_time": 0.0016013145446777343, "loss": 0.39692915976047516, "time": 0.6353601455688477, "epoch": 3, "memory": 44138, "step": 10614} +{"lr": 2.6513567377487876e-05, "data_time": 0.0015070199966430663, "loss": 0.3878687500953674, "time": 0.6440400123596192, "epoch": 3, "memory": 44138, "step": 10714} +{"lr": 2.645234514185487e-05, "data_time": 0.0015334606170654297, "loss": 0.41319349110126496, "time": 0.6355741500854493, "epoch": 3, "memory": 44138, "step": 10814} +{"lr": 2.6390705721713513e-05, "data_time": 0.0014587163925170899, "loss": 0.39748058319091795, "time": 0.6354933738708496, "epoch": 3, "memory": 44138, "step": 10914} +{"lr": 2.6328653102438134e-05, "data_time": 0.001739335060119629, "loss": 0.41561255156993865, "time": 0.6355241298675537, "epoch": 3, "memory": 44138, "step": 11014} +{"lr": 2.626619129611909e-05, "data_time": 0.001514267921447754, "loss": 0.41075489223003386, "time": 0.6360374689102173, "epoch": 3, "memory": 44138, "step": 11114} +{"lr": 2.6203324341303198e-05, "data_time": 0.001657247543334961, "loss": 0.39961858093738556, "time": 0.6376914262771607, "epoch": 3, "memory": 44138, "step": 11214} +{"lr": 2.6140056302732683e-05, "data_time": 0.0014661550521850586, "loss": 0.40717593431472776, "time": 0.6385722637176514, "epoch": 3, "memory": 44138, "step": 11314} +{"lr": 2.6076391271082312e-05, "data_time": 0.0015054464340209961, "loss": 0.39594255983829496, "time": 0.6378923892974854, "epoch": 3, "memory": 44138, "step": 11414} +{"lr": 2.6012333362694966e-05, "data_time": 0.0014590740203857422, "loss": 0.4010176151990891, "time": 0.6353793621063233, "epoch": 3, "memory": 44138, "step": 11514} +{"lr": 2.5947886719315454e-05, "data_time": 0.0015369176864624024, "loss": 0.39961256980896, "time": 0.6385421276092529, "epoch": 3, "memory": 44138, "step": 11614} +{"lr": 2.58830555078227e-05, "data_time": 0.0012540578842163085, "loss": 0.3792997360229492, "time": 0.6358986854553222, "epoch": 3, "memory": 44138, "step": 11714} +{"accuracy/top1": 81.46550750732422, "data_time": 0.03180969994643639, "time": 1.276645541191101, "step": 3} +{"lr": 2.5813264972520608e-05, "data_time": 0.0014603376388549806, "loss": 0.40517268776893617, "time": 0.6358929395675659, "epoch": 4, "memory": 44138, "step": 11821} +{"lr": 2.574765105195596e-05, "data_time": 0.0018689393997192382, "loss": 0.4040860444307327, "time": 0.6372992992401123, "epoch": 4, "memory": 44138, "step": 11921} +{"lr": 2.5681665509767622e-05, "data_time": 0.0014482498168945312, "loss": 0.39335237741470336, "time": 0.6357096433639526, "epoch": 4, "memory": 44138, "step": 12021} +{"lr": 2.561531261233396e-05, "data_time": 0.0016059637069702148, "loss": 0.4019911468029022, "time": 0.6355145454406739, "epoch": 4, "memory": 44138, "step": 12121} +{"lr": 2.554859664978521e-05, "data_time": 0.0014123201370239257, "loss": 0.4026492089033127, "time": 0.6374299764633179, "epoch": 4, "memory": 44138, "step": 12221} +{"lr": 2.5481521935725978e-05, "data_time": 0.001361989974975586, "loss": 0.3911616176366806, "time": 0.637604808807373, "epoch": 4, "memory": 44138, "step": 12321} +{"lr": 2.541409280695646e-05, "data_time": 0.001536726951599121, "loss": 0.4097395360469818, "time": 0.6359011650085449, "epoch": 4, "memory": 44138, "step": 12421} +{"lr": 2.5346313623192016e-05, "data_time": 0.0014959096908569336, "loss": 0.3840169131755829, "time": 0.6381452322006226, "epoch": 4, "memory": 44138, "step": 12521} +{"lr": 2.5278188766781202e-05, "data_time": 0.0014828205108642577, "loss": 0.40848673284053805, "time": 0.636052131652832, "epoch": 4, "memory": 44138, "step": 12621} +{"lr": 2.520972264242233e-05, "data_time": 0.0015128135681152343, "loss": 0.38745259344577787, "time": 0.6361779928207397, "epoch": 4, "memory": 44138, "step": 12721} +{"lr": 2.5140919676879075e-05, "data_time": 0.001426553726196289, "loss": 0.4014747738838196, "time": 0.6361227035522461, "epoch": 4, "memory": 44138, "step": 12821} +{"lr": 2.5071784318693913e-05, "data_time": 0.001404404640197754, "loss": 0.3918736457824707, "time": 0.6363578557968139, "epoch": 4, "memory": 44138, "step": 12921} +{"lr": 2.5002321037900605e-05, "data_time": 0.0013864278793334962, "loss": 0.396357935667038, "time": 0.6352404117584228, "epoch": 4, "memory": 44138, "step": 13021} +{"lr": 2.4932534325735087e-05, "data_time": 0.0014417648315429687, "loss": 0.3730785220861435, "time": 0.6372972249984741, "epoch": 4, "memory": 44138, "step": 13121} +{"lr": 2.486242869434525e-05, "data_time": 0.0018372535705566406, "loss": 0.39955129027366637, "time": 0.6367447137832641, "epoch": 4, "memory": 44138, "step": 13221} +{"lr": 2.4792008676499066e-05, "data_time": 0.001536393165588379, "loss": 0.40164371728897097, "time": 0.6356911182403564, "epoch": 4, "memory": 44138, "step": 13321} +{"lr": 2.472127882529154e-05, "data_time": 0.001662874221801758, "loss": 0.3860657066106796, "time": 0.6371049642562866, "epoch": 4, "memory": 44138, "step": 13421} +{"lr": 2.4650243713850305e-05, "data_time": 0.0015432596206665038, "loss": 0.38913559913635254, "time": 0.63596670627594, "epoch": 4, "memory": 44138, "step": 13521} +{"lr": 2.4578907935040093e-05, "data_time": 0.0015214204788208008, "loss": 0.37403162717819216, "time": 0.6397278070449829, "epoch": 4, "memory": 44138, "step": 13621} +{"lr": 2.4507276101165596e-05, "data_time": 0.0016132593154907227, "loss": 0.3677979320287704, "time": 0.6453139781951904, "epoch": 4, "memory": 44138, "step": 13721} +{"lr": 2.4435352843673332e-05, "data_time": 0.0014202356338500976, "loss": 0.4099270522594452, "time": 0.6363095760345459, "epoch": 4, "memory": 44138, "step": 13821} +{"lr": 2.4363142812852158e-05, "data_time": 0.0013915538787841798, "loss": 0.39708671867847445, "time": 0.6355254888534546, "epoch": 4, "memory": 44138, "step": 13921} +{"lr": 2.4290650677532606e-05, "data_time": 0.001457071304321289, "loss": 0.3684999763965607, "time": 0.6364887952804565, "epoch": 4, "memory": 44138, "step": 14021} +{"lr": 2.4217881124785113e-05, "data_time": 0.0014864683151245117, "loss": 0.39890529215335846, "time": 0.6435006618499756, "epoch": 4, "memory": 44138, "step": 14121} +{"lr": 2.4144838859616835e-05, "data_time": 0.001450824737548828, "loss": 0.3913849234580994, "time": 0.6382109165191651, "epoch": 4, "memory": 44138, "step": 14221} +{"lr": 2.407152860466747e-05, "data_time": 0.0014965295791625976, "loss": 0.3884093701839447, "time": 0.6387155532836915, "epoch": 4, "memory": 44138, "step": 14321} +{"lr": 2.3997955099903982e-05, "data_time": 0.001557326316833496, "loss": 0.39054149985313413, "time": 0.6356399059295654, "epoch": 4, "memory": 44138, "step": 14421} +{"lr": 2.3924123102314014e-05, "data_time": 0.0013512849807739257, "loss": 0.40433686077594755, "time": 0.6356610059738159, "epoch": 4, "memory": 44138, "step": 14521} +{"lr": 2.385003738559849e-05, "data_time": 0.0014832258224487306, "loss": 0.3877225279808044, "time": 0.6363243579864502, "epoch": 4, "memory": 44138, "step": 14621} +{"lr": 2.3775702739862787e-05, "data_time": 0.0015548229217529296, "loss": 0.3914781719446182, "time": 0.635591459274292, "epoch": 4, "memory": 44138, "step": 14721} +{"lr": 2.370112397130714e-05, "data_time": 0.0014350414276123047, "loss": 0.38474114537239074, "time": 0.6357931137084961, "epoch": 4, "memory": 44138, "step": 14821} +{"lr": 2.3626305901915825e-05, "data_time": 0.0015352249145507812, "loss": 0.38193814754486083, "time": 0.6361060619354248, "epoch": 4, "memory": 44138, "step": 14921} +{"lr": 2.3551253369145433e-05, "data_time": 0.001481914520263672, "loss": 0.3836988091468811, "time": 0.6354096174240113, "epoch": 4, "memory": 44138, "step": 15021} +{"lr": 2.3475971225612054e-05, "data_time": 0.0014744043350219727, "loss": 0.40507255792617797, "time": 0.6389569044113159, "epoch": 4, "memory": 44138, "step": 15121} +{"lr": 2.3400464338777606e-05, "data_time": 0.0015366554260253906, "loss": 0.3861899584531784, "time": 0.6360882759094239, "epoch": 4, "memory": 44138, "step": 15221} +{"lr": 2.332473759063499e-05, "data_time": 0.0014726877212524413, "loss": 0.3677582621574402, "time": 0.636666226387024, "epoch": 4, "memory": 44138, "step": 15321} +{"lr": 2.3248795877392593e-05, "data_time": 0.001471090316772461, "loss": 0.3959373295307159, "time": 0.6355545997619629, "epoch": 4, "memory": 44138, "step": 15421} +{"lr": 2.317264410915764e-05, "data_time": 0.0014323949813842773, "loss": 0.39965182542800903, "time": 0.6355262279510498, "epoch": 4, "memory": 44138, "step": 15521} +{"lr": 2.3096287209618654e-05, "data_time": 0.0012560129165649415, "loss": 0.39307435154914855, "time": 0.6355045795440674, "epoch": 4, "memory": 44138, "step": 15621} +{"accuracy/top1": 80.31383514404297, "data_time": 0.032101787369826745, "time": 1.2763871694433278, "step": 4} +{"lr": 2.3014363749707527e-05, "data_time": 0.0015312433242797852, "loss": 0.3915721088647842, "time": 0.6358813762664794, "epoch": 5, "memory": 44138, "step": 15728} +{"lr": 2.2937597929948786e-05, "data_time": 0.0016015052795410156, "loss": 0.3880497306585312, "time": 0.6367979764938354, "epoch": 5, "memory": 44138, "step": 15828} +{"lr": 2.286064217609303e-05, "data_time": 0.0015599489212036132, "loss": 0.4007106781005859, "time": 0.6364808320999146, "epoch": 5, "memory": 44138, "step": 15928} +{"lr": 2.2783501463811504e-05, "data_time": 0.001565837860107422, "loss": 0.42414140701293945, "time": 0.6363464593887329, "epoch": 5, "memory": 44138, "step": 16028} +{"lr": 2.2706180780734223e-05, "data_time": 0.0015761137008666992, "loss": 0.3773953467607498, "time": 0.635633397102356, "epoch": 5, "memory": 44138, "step": 16128} +{"lr": 2.2628685126127375e-05, "data_time": 0.0015518665313720703, "loss": 0.3699106991291046, "time": 0.636089825630188, "epoch": 5, "memory": 44138, "step": 16228} +{"lr": 2.2551019510570224e-05, "data_time": 0.0016492605209350586, "loss": 0.4149298667907715, "time": 0.6366400957107544, "epoch": 5, "memory": 44138, "step": 16328} +{"lr": 2.247318895563104e-05, "data_time": 0.0014890193939208984, "loss": 0.3634637206792831, "time": 0.6381202697753906, "epoch": 5, "memory": 44138, "step": 16428} +{"lr": 2.2395198493542463e-05, "data_time": 0.0014709234237670898, "loss": 0.37360579073429107, "time": 0.6367754459381103, "epoch": 5, "memory": 44138, "step": 16528} +{"lr": 2.2317053166876117e-05, "data_time": 0.0014590978622436523, "loss": 0.36054129898548126, "time": 0.6359807014465332, "epoch": 5, "memory": 44138, "step": 16628} +{"lr": 2.223875802821662e-05, "data_time": 0.0014612197875976563, "loss": 0.3709414005279541, "time": 0.6374520778656005, "epoch": 5, "memory": 44138, "step": 16728} +{"lr": 2.2160318139834804e-05, "data_time": 0.0015422821044921875, "loss": 0.38987391591072085, "time": 0.6362518787384033, "epoch": 5, "memory": 44138, "step": 16828} +{"lr": 2.2081738573360552e-05, "data_time": 0.0014992237091064453, "loss": 0.3937067031860352, "time": 0.6383609533309936, "epoch": 5, "memory": 44138, "step": 16928} +{"lr": 2.20030244094548e-05, "data_time": 0.0014134645462036133, "loss": 0.37678520679473876, "time": 0.6390758991241455, "epoch": 5, "memory": 44138, "step": 17028} +{"lr": 2.192418073748101e-05, "data_time": 0.0014934539794921875, "loss": 0.40795170664787295, "time": 0.6361664533615112, "epoch": 5, "memory": 44138, "step": 17128} +{"lr": 2.1845212655176195e-05, "data_time": 0.0016251087188720703, "loss": 0.3827826589345932, "time": 0.637525224685669, "epoch": 5, "memory": 44138, "step": 17228} +{"lr": 2.1766125268321264e-05, "data_time": 0.0014215230941772461, "loss": 0.3655057817697525, "time": 0.6384395837783814, "epoch": 5, "memory": 44138, "step": 17328} +{"lr": 2.168692369041089e-05, "data_time": 0.0017093658447265626, "loss": 0.3965724974870682, "time": 0.6359135150909424, "epoch": 5, "memory": 44138, "step": 17428} +{"lr": 2.1607613042322935e-05, "data_time": 0.0014518260955810546, "loss": 0.3884787291288376, "time": 0.6355259895324707, "epoch": 5, "memory": 44138, "step": 17528} +{"lr": 2.1528198451987307e-05, "data_time": 0.001448845863342285, "loss": 0.37770710289478304, "time": 0.6357872724533081, "epoch": 5, "memory": 44138, "step": 17628} +{"lr": 2.1448685054054467e-05, "data_time": 0.001620650291442871, "loss": 0.38291923701763153, "time": 0.6368507146835327, "epoch": 5, "memory": 44138, "step": 17728} +{"lr": 2.1369077989563356e-05, "data_time": 0.0015523672103881837, "loss": 0.39392951428890227, "time": 0.639568567276001, "epoch": 5, "memory": 44138, "step": 17828} +{"lr": 2.128938240560911e-05, "data_time": 0.0014477252960205078, "loss": 0.38296268284320834, "time": 0.6376967668533325, "epoch": 5, "memory": 44138, "step": 17928} +{"lr": 2.120960345501016e-05, "data_time": 0.0014770746231079102, "loss": 0.407487154006958, "time": 0.6372606515884399, "epoch": 5, "memory": 44138, "step": 18028} +{"lr": 2.1129746295975105e-05, "data_time": 0.0014374494552612306, "loss": 0.39411935210227966, "time": 0.6363304376602172, "epoch": 5, "memory": 44138, "step": 18128} +{"lr": 2.1049816091769223e-05, "data_time": 0.0015367746353149414, "loss": 0.3669018536806107, "time": 0.636871600151062, "epoch": 5, "memory": 44138, "step": 18228} +{"lr": 2.096981801038062e-05, "data_time": 0.0016428470611572266, "loss": 0.3765005588531494, "time": 0.6359396696090698, "epoch": 5, "memory": 44138, "step": 18328} +{"lr": 2.0889757224186114e-05, "data_time": 0.001549363136291504, "loss": 0.3841983497142792, "time": 0.638393497467041, "epoch": 5, "memory": 44138, "step": 18428} +{"lr": 2.0809638909616704e-05, "data_time": 0.0014175176620483398, "loss": 0.408274906873703, "time": 0.6357414960861206, "epoch": 5, "memory": 44138, "step": 18528} +{"lr": 2.072946824682303e-05, "data_time": 0.0016066074371337891, "loss": 0.401069301366806, "time": 0.6359381914138794, "epoch": 5, "memory": 44138, "step": 18628} +{"lr": 2.0649250419340373e-05, "data_time": 0.0015076875686645507, "loss": 0.3510246902704239, "time": 0.6410531044006348, "epoch": 5, "memory": 44138, "step": 18728} +{"lr": 2.0568990613753462e-05, "data_time": 0.0015563488006591797, "loss": 0.38433206677436826, "time": 0.6372015953063965, "epoch": 5, "memory": 44138, "step": 18828} +{"lr": 2.0488694019361218e-05, "data_time": 0.001474452018737793, "loss": 0.3696636170148849, "time": 0.6371400833129883, "epoch": 5, "memory": 44138, "step": 18928} +{"lr": 2.0408365827841122e-05, "data_time": 0.001377701759338379, "loss": 0.36692739427089693, "time": 0.6382151365280151, "epoch": 5, "memory": 44138, "step": 19028} +{"lr": 2.0328011232913694e-05, "data_time": 0.00146026611328125, "loss": 0.37398011684417726, "time": 0.6375979900360107, "epoch": 5, "memory": 44138, "step": 19128} +{"lr": 2.024763543000651e-05, "data_time": 0.0015254259109497071, "loss": 0.3797903835773468, "time": 0.6355776309967041, "epoch": 5, "memory": 44138, "step": 19228} +{"lr": 2.0167243615918473e-05, "data_time": 0.0014070034027099609, "loss": 0.37698997259140016, "time": 0.6385165214538574, "epoch": 5, "memory": 44138, "step": 19328} +{"lr": 2.0086840988483663e-05, "data_time": 0.0014674663543701172, "loss": 0.3949741184711456, "time": 0.6355803489685059, "epoch": 5, "memory": 44138, "step": 19428} +{"lr": 2.00064327462353e-05, "data_time": 0.0012489080429077149, "loss": 0.3776142954826355, "time": 0.6377695798873901, "epoch": 5, "memory": 44138, "step": 19528} +{"accuracy/top1": 79.67363739013672, "data_time": 0.030968370108768857, "time": 1.273853819945763, "step": 5} +{"lr": 1.992039560076574e-05, "data_time": 0.0013307809829711914, "loss": 0.37925947904586793, "time": 0.6388778686523438, "epoch": 6, "memory": 44138, "step": 19635} +{"lr": 1.983999225510998e-05, "data_time": 0.0013992071151733398, "loss": 0.3913419783115387, "time": 0.6357207775115967, "epoch": 6, "memory": 44138, "step": 19735} +{"lr": 1.9759599254956633e-05, "data_time": 0.0014631271362304688, "loss": 0.4005438446998596, "time": 0.6363759994506836, "epoch": 6, "memory": 44138, "step": 19835} +{"lr": 1.9679221798216487e-05, "data_time": 0.0014740467071533204, "loss": 0.3782841920852661, "time": 0.636821699142456, "epoch": 6, "memory": 44138, "step": 19935} +{"lr": 1.9598865081795367e-05, "data_time": 0.0015021800994873048, "loss": 0.38352116048336027, "time": 0.645548152923584, "epoch": 6, "memory": 44138, "step": 20035} +{"lr": 1.9518534301258014e-05, "data_time": 0.0015161752700805664, "loss": 0.3651569336652756, "time": 0.6390479564666748, "epoch": 6, "memory": 44138, "step": 20135} +{"lr": 1.9438234650492273e-05, "data_time": 0.0014492034912109374, "loss": 0.39326231777667997, "time": 0.6369673252105713, "epoch": 6, "memory": 44138, "step": 20235} +{"lr": 1.935797132137332e-05, "data_time": 0.001443338394165039, "loss": 0.39745269119739535, "time": 0.6358798027038575, "epoch": 6, "memory": 44138, "step": 20335} +{"lr": 1.927774950342788e-05, "data_time": 0.0015887975692749023, "loss": 0.40316917896270754, "time": 0.6433388233184815, "epoch": 6, "memory": 44138, "step": 20435} +{"lr": 1.9197574383498696e-05, "data_time": 0.0013966560363769531, "loss": 0.3704320967197418, "time": 0.6372830152511597, "epoch": 6, "memory": 44138, "step": 20535} +{"lr": 1.911745114540919e-05, "data_time": 0.0015395402908325196, "loss": 0.3754275530576706, "time": 0.6360685348510742, "epoch": 6, "memory": 44138, "step": 20635} +{"lr": 1.9037384969628367e-05, "data_time": 0.0014511823654174804, "loss": 0.3556096822023392, "time": 0.6362322092056274, "epoch": 6, "memory": 44138, "step": 20735} +{"lr": 1.8957381032935694e-05, "data_time": 0.0016643047332763673, "loss": 0.3643163651227951, "time": 0.636770224571228, "epoch": 6, "memory": 44138, "step": 20835} +{"lr": 1.887744450808657e-05, "data_time": 0.001493072509765625, "loss": 0.3651486158370972, "time": 0.6381950378417969, "epoch": 6, "memory": 44138, "step": 20935} +{"lr": 1.8797580563477768e-05, "data_time": 0.0015011310577392578, "loss": 0.36550105214118955, "time": 0.6361826419830322, "epoch": 6, "memory": 44138, "step": 21035} +{"lr": 1.8717794362813298e-05, "data_time": 0.0016357183456420898, "loss": 0.36020020544528963, "time": 0.6352993965148925, "epoch": 6, "memory": 44138, "step": 21135} +{"lr": 1.863809106477054e-05, "data_time": 0.0015210866928100585, "loss": 0.3669223040342331, "time": 0.6356857061386109, "epoch": 6, "memory": 44138, "step": 21235} +{"lr": 1.8558475822666676e-05, "data_time": 0.0016295671463012694, "loss": 0.3765056669712067, "time": 0.6384711980819702, "epoch": 6, "memory": 44138, "step": 21335} +{"lr": 1.8478953784125537e-05, "data_time": 0.001545262336730957, "loss": 0.39416989386081697, "time": 0.6383140325546265, "epoch": 6, "memory": 44138, "step": 21435} +{"lr": 1.839953009074481e-05, "data_time": 0.0014120817184448242, "loss": 0.3795745223760605, "time": 0.6388851642608643, "epoch": 6, "memory": 44138, "step": 21535} +{"lr": 1.832020987776345e-05, "data_time": 0.0017464876174926758, "loss": 0.3982358336448669, "time": 0.6365885972976685, "epoch": 6, "memory": 44138, "step": 21635} +{"lr": 1.8240998273729876e-05, "data_time": 0.0015010595321655273, "loss": 0.36450940668582915, "time": 0.6385514259338378, "epoch": 6, "memory": 44138, "step": 21735} +{"lr": 1.8161900400170145e-05, "data_time": 0.0015857458114624024, "loss": 0.3877420246601105, "time": 0.6370485305786133, "epoch": 6, "memory": 44138, "step": 21835} +{"lr": 1.8082921371256987e-05, "data_time": 0.0013997554779052734, "loss": 0.39647932052612306, "time": 0.6361183643341064, "epoch": 6, "memory": 44138, "step": 21935} +{"lr": 1.800406629347906e-05, "data_time": 0.0014848709106445312, "loss": 0.38379827737808225, "time": 0.6363045930862427, "epoch": 6, "memory": 44138, "step": 22035} +{"lr": 1.7925340265310823e-05, "data_time": 0.0015642642974853516, "loss": 0.41020860970020295, "time": 0.6364546775817871, "epoch": 6, "memory": 44138, "step": 22135} +{"lr": 1.7846748376882866e-05, "data_time": 0.0015359878540039062, "loss": 0.3946759521961212, "time": 0.6380988359451294, "epoch": 6, "memory": 44138, "step": 22235} +{"lr": 1.7768295709652765e-05, "data_time": 0.001474761962890625, "loss": 0.37165563106536864, "time": 0.6350480556488037, "epoch": 6, "memory": 44138, "step": 22335} +{"lr": 1.768998733607668e-05, "data_time": 0.0013903379440307617, "loss": 0.3782049924135208, "time": 0.6372109889984131, "epoch": 6, "memory": 44138, "step": 22435} +{"lr": 1.7611828319281165e-05, "data_time": 0.0014556169509887696, "loss": 0.33835927248001096, "time": 0.6372325658798218, "epoch": 6, "memory": 44138, "step": 22535} +{"lr": 1.7533823712735976e-05, "data_time": 0.0015301227569580079, "loss": 0.3630715638399124, "time": 0.6378426551818848, "epoch": 6, "memory": 44138, "step": 22635} +{"lr": 1.745597855992727e-05, "data_time": 0.0017188072204589843, "loss": 0.3874973326921463, "time": 0.6369879961013794, "epoch": 6, "memory": 44138, "step": 22735} +{"lr": 1.7378297894031527e-05, "data_time": 0.0014562129974365235, "loss": 0.37654756009578705, "time": 0.6368170261383057, "epoch": 6, "memory": 44138, "step": 22835} +{"lr": 1.730078673759016e-05, "data_time": 0.0014070034027099609, "loss": 0.38792833089828493, "time": 0.6381886959075928, "epoch": 6, "memory": 44138, "step": 22935} +{"lr": 1.7223450102184637e-05, "data_time": 0.001417398452758789, "loss": 0.36938416957855225, "time": 0.6375573873519897, "epoch": 6, "memory": 44138, "step": 23035} +{"lr": 1.7146292988112627e-05, "data_time": 0.0014190912246704102, "loss": 0.37706841230392457, "time": 0.6368034839630127, "epoch": 6, "memory": 44138, "step": 23135} +{"lr": 1.7069320384064597e-05, "data_time": 0.0015091419219970704, "loss": 0.3630496084690094, "time": 0.6373643636703491, "epoch": 6, "memory": 44138, "step": 23235} +{"lr": 1.6992537266801296e-05, "data_time": 0.001432657241821289, "loss": 0.36087783575057986, "time": 0.6366403341293335, "epoch": 6, "memory": 44138, "step": 23335} +{"lr": 1.6915948600831928e-05, "data_time": 0.0014080762863159179, "loss": 0.39652993381023405, "time": 0.6343589067459107, "epoch": 6, "memory": 44138, "step": 23435} +{"accuracy/top1": 79.66320037841797, "data_time": 0.03217871435757341, "time": 1.276825843186214, "step": 6} +{"lr": 1.6834219685058524e-05, "data_time": 0.0015770673751831054, "loss": 0.37019487023353576, "time": 0.6351701974868774, "epoch": 7, "memory": 44138, "step": 23542} +{"lr": 1.675804925318156e-05, "data_time": 0.0014691352844238281, "loss": 0.36275808811187743, "time": 0.6352355480194092, "epoch": 7, "memory": 44138, "step": 23642} +{"lr": 1.6682088433716622e-05, "data_time": 0.00149383544921875, "loss": 0.3685837537050247, "time": 0.637007737159729, "epoch": 7, "memory": 44138, "step": 23742} +{"lr": 1.6606342138006298e-05, "data_time": 0.0014837026596069337, "loss": 0.35596791505813596, "time": 0.6355965852737426, "epoch": 7, "memory": 44138, "step": 23842} +{"lr": 1.653081526352276e-05, "data_time": 0.0015264511108398437, "loss": 0.38109772503376005, "time": 0.6363924741744995, "epoch": 7, "memory": 44138, "step": 23942} +{"lr": 1.6455512693551297e-05, "data_time": 0.0015047311782836914, "loss": 0.4049632966518402, "time": 0.6363247394561767, "epoch": 7, "memory": 44138, "step": 24042} +{"lr": 1.6380439296874487e-05, "data_time": 0.001533174514770508, "loss": 0.35496670603752134, "time": 0.6382855892181396, "epoch": 7, "memory": 44138, "step": 24142} +{"lr": 1.630559992745743e-05, "data_time": 0.0015706062316894532, "loss": 0.36166253983974456, "time": 0.6361469745635986, "epoch": 7, "memory": 44138, "step": 24242} +{"lr": 1.6230999424133908e-05, "data_time": 0.0014258623123168945, "loss": 0.3531336039304733, "time": 0.6377415657043457, "epoch": 7, "memory": 44138, "step": 24342} +{"lr": 1.615664261029346e-05, "data_time": 0.0014956235885620118, "loss": 0.3748207002878189, "time": 0.6369265079498291, "epoch": 7, "memory": 44138, "step": 24442} +{"lr": 1.6082534293569616e-05, "data_time": 0.0015178442001342774, "loss": 0.39182004034519197, "time": 0.635042953491211, "epoch": 7, "memory": 44138, "step": 24542} +{"lr": 1.6008679265529044e-05, "data_time": 0.0014105796813964843, "loss": 0.35284451246261594, "time": 0.6368744850158692, "epoch": 7, "memory": 44138, "step": 24642} +{"lr": 1.593508230136164e-05, "data_time": 0.0015794992446899413, "loss": 0.3764829635620117, "time": 0.639616060256958, "epoch": 7, "memory": 44138, "step": 24742} +{"lr": 1.5861748159571916e-05, "data_time": 0.0015469074249267578, "loss": 0.36659168303012846, "time": 0.6356468677520752, "epoch": 7, "memory": 44138, "step": 24842} +{"lr": 1.5788681581671252e-05, "data_time": 0.0015135765075683593, "loss": 0.3824197918176651, "time": 0.6371435165405274, "epoch": 7, "memory": 44138, "step": 24942} +{"lr": 1.571588729187132e-05, "data_time": 0.0015353918075561523, "loss": 0.38916949927806854, "time": 0.6372353553771972, "epoch": 7, "memory": 44138, "step": 25042} +{"lr": 1.564336999677871e-05, "data_time": 0.0015442848205566406, "loss": 0.358869206905365, "time": 0.6356818675994873, "epoch": 7, "memory": 44138, "step": 25142} +{"lr": 1.5571134385090526e-05, "data_time": 0.001357126235961914, "loss": 0.37347002029418946, "time": 0.6356656789779663, "epoch": 7, "memory": 44138, "step": 25242} +{"lr": 1.54991851272913e-05, "data_time": 0.0014548301696777344, "loss": 0.3890160501003265, "time": 0.6361973285675049, "epoch": 7, "memory": 44138, "step": 25342} +{"lr": 1.5427526875350992e-05, "data_time": 0.0014637947082519532, "loss": 0.3633673697710037, "time": 0.6364731788635254, "epoch": 7, "memory": 44138, "step": 25442} +{"lr": 1.5356164262424192e-05, "data_time": 0.0015940904617309571, "loss": 0.3767683207988739, "time": 0.63603835105896, "epoch": 7, "memory": 44138, "step": 25542} +{"lr": 1.5285101902550614e-05, "data_time": 0.0015026330947875977, "loss": 0.3859305500984192, "time": 0.6351003885269165, "epoch": 7, "memory": 44138, "step": 25642} +{"lr": 1.5214344390356706e-05, "data_time": 0.0014180660247802735, "loss": 0.3874822914600372, "time": 0.6402966260910035, "epoch": 7, "memory": 44138, "step": 25742} +{"lr": 1.5143896300758606e-05, "data_time": 0.0014272689819335937, "loss": 0.36415823400020597, "time": 0.6360689878463746, "epoch": 7, "memory": 44138, "step": 25842} +{"lr": 1.5073762188666339e-05, "data_time": 0.0019725799560546876, "loss": 0.3522688329219818, "time": 0.6363534927368164, "epoch": 7, "memory": 44138, "step": 25942} +{"lr": 1.5003946588689345e-05, "data_time": 0.0014714717864990235, "loss": 0.37421321868896484, "time": 0.6362645626068115, "epoch": 7, "memory": 44138, "step": 26042} +{"lr": 1.4934454014843216e-05, "data_time": 0.001434469223022461, "loss": 0.3535464584827423, "time": 0.6377779722213746, "epoch": 7, "memory": 44138, "step": 26142} +{"lr": 1.4865288960257924e-05, "data_time": 0.0014002799987792968, "loss": 0.3805251866579056, "time": 0.6351898193359375, "epoch": 7, "memory": 44138, "step": 26242} +{"lr": 1.479645589688726e-05, "data_time": 0.0015389680862426757, "loss": 0.3934237599372864, "time": 0.6355166912078858, "epoch": 7, "memory": 44138, "step": 26342} +{"lr": 1.4727959275219693e-05, "data_time": 0.0017250537872314452, "loss": 0.3818521350622177, "time": 0.6361732482910156, "epoch": 7, "memory": 44138, "step": 26442} +{"lr": 1.4659803523990658e-05, "data_time": 0.0016115903854370117, "loss": 0.33923845887184145, "time": 0.6361941576004029, "epoch": 7, "memory": 44138, "step": 26542} +{"lr": 1.4591993049896156e-05, "data_time": 0.0015195608139038086, "loss": 0.37609065473079684, "time": 0.6364243030548096, "epoch": 7, "memory": 44138, "step": 26642} +{"lr": 1.4524532237307841e-05, "data_time": 0.0017479419708251952, "loss": 0.36433753967285154, "time": 0.6365207195281982, "epoch": 7, "memory": 44138, "step": 26742} +{"lr": 1.44574254479896e-05, "data_time": 0.0014411687850952148, "loss": 0.38123677372932435, "time": 0.6364466905593872, "epoch": 7, "memory": 44138, "step": 26842} +{"lr": 1.4390677020815494e-05, "data_time": 0.0014626741409301757, "loss": 0.36849929094314576, "time": 0.6356098651885986, "epoch": 7, "memory": 44138, "step": 26942} +{"lr": 1.4324291271489216e-05, "data_time": 0.001484823226928711, "loss": 0.35879650712013245, "time": 0.639635157585144, "epoch": 7, "memory": 44138, "step": 27042} +{"lr": 1.4258272492265071e-05, "data_time": 0.0015256881713867187, "loss": 0.3917722821235657, "time": 0.637731409072876, "epoch": 7, "memory": 44138, "step": 27142} +{"lr": 1.4192624951670425e-05, "data_time": 0.0016059398651123047, "loss": 0.36367155611515045, "time": 0.6497369050979614, "epoch": 7, "memory": 44138, "step": 27242} +{"lr": 1.4127352894229787e-05, "data_time": 0.001289200782775879, "loss": 0.3828736573457718, "time": 0.6345186471939087, "epoch": 7, "memory": 44138, "step": 27342} +{"accuracy/top1": 79.10650634765625, "data_time": 0.033783953765342975, "time": 1.2759003721434494, "step": 7} +{"lr": 1.4057932403867309e-05, "data_time": 0.0015304088592529297, "loss": 0.4041621804237366, "time": 0.6368994235992431, "epoch": 8, "memory": 44138, "step": 27449} +{"lr": 1.3993450978351854e-05, "data_time": 0.0014450788497924806, "loss": 0.3982856273651123, "time": 0.6391650438308716, "epoch": 8, "memory": 44138, "step": 27549} +{"lr": 1.3929357913834719e-05, "data_time": 0.001488637924194336, "loss": 0.3864494860172272, "time": 0.6367493629455566, "epoch": 8, "memory": 44138, "step": 27649} +{"lr": 1.3865657354333762e-05, "data_time": 0.0014400243759155273, "loss": 0.37924416959285734, "time": 0.6353884220123291, "epoch": 8, "memory": 44138, "step": 27749} +{"lr": 1.3802353418488967e-05, "data_time": 0.0014401674270629883, "loss": 0.3679250717163086, "time": 0.6384751796722412, "epoch": 8, "memory": 44138, "step": 27849} +{"lr": 1.3739450199296088e-05, "data_time": 0.0014028072357177735, "loss": 0.3605092942714691, "time": 0.6369296550750733, "epoch": 8, "memory": 44138, "step": 27949} +{"lr": 1.3676951763842047e-05, "data_time": 0.001517939567565918, "loss": 0.37059405744075774, "time": 0.6392914533615113, "epoch": 8, "memory": 44138, "step": 28049} +{"lr": 1.361486215304198e-05, "data_time": 0.0014487504959106445, "loss": 0.38957723081111906, "time": 0.6377227544784546, "epoch": 8, "memory": 44138, "step": 28149} +{"lr": 1.3553185381377936e-05, "data_time": 0.0014718294143676758, "loss": 0.3781526476144791, "time": 0.6371407032012939, "epoch": 8, "memory": 44138, "step": 28249} +{"lr": 1.3491925436639332e-05, "data_time": 0.0015551328659057617, "loss": 0.3768164038658142, "time": 0.6375884056091309, "epoch": 8, "memory": 44138, "step": 28349} +{"lr": 1.3431086279665111e-05, "data_time": 0.001489877700805664, "loss": 0.37365337610244753, "time": 0.6350898981094361, "epoch": 8, "memory": 44138, "step": 28449} +{"lr": 1.3370671844087644e-05, "data_time": 0.0014866113662719727, "loss": 0.38336474299430845, "time": 0.636806058883667, "epoch": 8, "memory": 44138, "step": 28549} +{"lr": 1.331068603607844e-05, "data_time": 0.001475691795349121, "loss": 0.37587199211120603, "time": 0.6374342918395997, "epoch": 8, "memory": 44138, "step": 28649} +{"lr": 1.3251132734095514e-05, "data_time": 0.0014387845993041992, "loss": 0.3791746735572815, "time": 0.6370241165161132, "epoch": 8, "memory": 44138, "step": 28749} +{"lr": 1.319201578863267e-05, "data_time": 0.0014643907546997071, "loss": 0.3773917078971863, "time": 0.6371464252471923, "epoch": 8, "memory": 44138, "step": 28849} +{"lr": 1.313333902197055e-05, "data_time": 0.0014708518981933593, "loss": 0.36259378492832184, "time": 0.6372398138046265, "epoch": 8, "memory": 44138, "step": 28949} +{"lr": 1.3075106227929447e-05, "data_time": 0.0015167236328125, "loss": 0.37624795734882355, "time": 0.6357067108154297, "epoch": 8, "memory": 44138, "step": 29049} +{"lr": 1.3017321171624083e-05, "data_time": 0.001559138298034668, "loss": 0.3777626246213913, "time": 0.6369559049606324, "epoch": 8, "memory": 44138, "step": 29149} +{"lr": 1.2959987589220086e-05, "data_time": 0.0015105962753295898, "loss": 0.3494486898183823, "time": 0.6365528106689453, "epoch": 8, "memory": 44138, "step": 29249} +{"lr": 1.2903109187692523e-05, "data_time": 0.0015789985656738282, "loss": 0.3398669481277466, "time": 0.6380286931991577, "epoch": 8, "memory": 44138, "step": 29349} +{"lr": 1.2846689644586123e-05, "data_time": 0.0015120506286621094, "loss": 0.376858252286911, "time": 0.6388290643692016, "epoch": 8, "memory": 44138, "step": 29449} +{"lr": 1.2790732607777567e-05, "data_time": 0.0015869379043579102, "loss": 0.36199540793895724, "time": 0.6377752542495727, "epoch": 8, "memory": 44138, "step": 29549} +{"lr": 1.2735241695239598e-05, "data_time": 0.0014661788940429688, "loss": 0.3535286635160446, "time": 0.6368883609771728, "epoch": 8, "memory": 44138, "step": 29649} +{"lr": 1.2680220494807118e-05, "data_time": 0.001450181007385254, "loss": 0.36757898032665254, "time": 0.6365644454956054, "epoch": 8, "memory": 44138, "step": 29749} +{"lr": 1.2625672563945206e-05, "data_time": 0.0014929056167602539, "loss": 0.36055062115192416, "time": 0.6375349521636963, "epoch": 8, "memory": 44138, "step": 29849} +{"lr": 1.2571601429519072e-05, "data_time": 0.0014863252639770509, "loss": 0.37129674255847933, "time": 0.6419340610504151, "epoch": 8, "memory": 44138, "step": 29949} +{"lr": 1.251801058756609e-05, "data_time": 0.001531839370727539, "loss": 0.3759852975606918, "time": 0.6389995098114014, "epoch": 8, "memory": 44138, "step": 30049} +{"lr": 1.2464903503069694e-05, "data_time": 0.0015131711959838867, "loss": 0.363563260436058, "time": 0.6371601581573486, "epoch": 8, "memory": 44138, "step": 30149} +{"lr": 1.2412283609735373e-05, "data_time": 0.0016901731491088868, "loss": 0.3954562932252884, "time": 0.6374919176101684, "epoch": 8, "memory": 44138, "step": 30249} +{"lr": 1.2360154309768638e-05, "data_time": 0.0015392303466796875, "loss": 0.383209890127182, "time": 0.6361282348632813, "epoch": 8, "memory": 44138, "step": 30349} +{"lr": 1.2308518973655091e-05, "data_time": 0.0018133163452148438, "loss": 0.3841525256633759, "time": 0.6378105878829956, "epoch": 8, "memory": 44138, "step": 30449} +{"lr": 1.2257380939942495e-05, "data_time": 0.0015905380249023437, "loss": 0.35346856713294983, "time": 0.6368402242660522, "epoch": 8, "memory": 44138, "step": 30549} +{"lr": 1.2206743515024876e-05, "data_time": 0.001598358154296875, "loss": 0.3574922651052475, "time": 0.6396049499511719, "epoch": 8, "memory": 44138, "step": 30649} +{"lr": 1.2156609972928755e-05, "data_time": 0.0015631437301635743, "loss": 0.3666721910238266, "time": 0.6398011445999146, "epoch": 8, "memory": 44138, "step": 30749} +{"lr": 1.2106983555101508e-05, "data_time": 0.0014890432357788086, "loss": 0.3716928005218506, "time": 0.637281060218811, "epoch": 8, "memory": 44138, "step": 30849} +{"lr": 1.205786747020173e-05, "data_time": 0.0015869140625, "loss": 0.36391824185848237, "time": 0.6374053239822388, "epoch": 8, "memory": 44138, "step": 30949} +{"lr": 1.2009264893891803e-05, "data_time": 0.0017653226852416993, "loss": 0.35911635756492616, "time": 0.6375832319259643, "epoch": 8, "memory": 44138, "step": 31049} +{"lr": 1.1961178968632558e-05, "data_time": 0.001415395736694336, "loss": 0.3853606224060059, "time": 0.6373692512512207, "epoch": 8, "memory": 44138, "step": 31149} +{"lr": 1.19136128034801e-05, "data_time": 0.0012783527374267579, "loss": 0.3817710131406784, "time": 0.6367529392242431, "epoch": 8, "memory": 44138, "step": 31249} +{"accuracy/top1": 79.18305206298828, "data_time": 0.03135753911117028, "time": 1.2761004299953067, "step": 8} +{"lr": 1.1863296099806215e-05, "data_time": 0.0016751527786254884, "loss": 0.3622595489025116, "time": 0.6390992403030396, "epoch": 9, "memory": 44138, "step": 31356} +{"lr": 1.181681557185025e-05, "data_time": 0.001538705825805664, "loss": 0.35549541711807253, "time": 0.6372063398361206, "epoch": 9, "memory": 44138, "step": 31456} +{"lr": 1.1770864137998475e-05, "data_time": 0.0017137765884399415, "loss": 0.37085160315036775, "time": 0.6371675729751587, "epoch": 9, "memory": 44138, "step": 31556} +{"lr": 1.1725444769298765e-05, "data_time": 0.001439523696899414, "loss": 0.3729469805955887, "time": 0.6406498193740845, "epoch": 9, "memory": 44138, "step": 31656} +{"lr": 1.1680560402397639e-05, "data_time": 0.001489710807800293, "loss": 0.3686543405056, "time": 0.6391665458679199, "epoch": 9, "memory": 44138, "step": 31756} +{"lr": 1.1636213939350422e-05, "data_time": 0.0014519453048706054, "loss": 0.3786431461572647, "time": 0.6372768640518188, "epoch": 9, "memory": 44138, "step": 31856} +{"lr": 1.159240824743359e-05, "data_time": 0.0013993024826049806, "loss": 0.36677649319171907, "time": 0.6361638069152832, "epoch": 9, "memory": 44138, "step": 31956} +{"lr": 1.1549146158959366e-05, "data_time": 0.0014185428619384766, "loss": 0.3582907050848007, "time": 0.6401641368865967, "epoch": 9, "memory": 44138, "step": 32056} +{"lr": 1.1506430471092626e-05, "data_time": 0.0014175653457641601, "loss": 0.3795019119977951, "time": 0.636742377281189, "epoch": 9, "memory": 44138, "step": 32156} +{"lr": 1.146426394567e-05, "data_time": 0.001481008529663086, "loss": 0.3826130092144012, "time": 0.6376669645309448, "epoch": 9, "memory": 44138, "step": 32256} +{"lr": 1.142264930902133e-05, "data_time": 0.001440572738647461, "loss": 0.39255124926567075, "time": 0.6372113943099975, "epoch": 9, "memory": 44138, "step": 32356} +{"lr": 1.1381589251793428e-05, "data_time": 0.0014751911163330077, "loss": 0.3772126495838165, "time": 0.6379462003707885, "epoch": 9, "memory": 44138, "step": 32456} +{"lr": 1.1341086428776037e-05, "data_time": 0.0018348217010498047, "loss": 0.34995020031929014, "time": 0.6377904415130615, "epoch": 9, "memory": 44138, "step": 32556} +{"lr": 1.1301143458730242e-05, "data_time": 0.0014760971069335937, "loss": 0.3697828710079193, "time": 0.6371870756149292, "epoch": 9, "memory": 44138, "step": 32656} +{"lr": 1.1261762924219115e-05, "data_time": 0.0015008211135864257, "loss": 0.3838991582393646, "time": 0.6397773504257203, "epoch": 9, "memory": 44138, "step": 32756} +{"lr": 1.1222947371440745e-05, "data_time": 0.0014661073684692383, "loss": 0.38077346980571747, "time": 0.6368762493133545, "epoch": 9, "memory": 44138, "step": 32856} +{"lr": 1.118469931006364e-05, "data_time": 0.0014377355575561524, "loss": 0.36794666945934296, "time": 0.6370911836624146, "epoch": 9, "memory": 44138, "step": 32956} +{"lr": 1.1147021213064413e-05, "data_time": 0.0015420913696289062, "loss": 0.36707726418972014, "time": 0.6390187740325928, "epoch": 9, "memory": 44138, "step": 33056} +{"lr": 1.1109915516567913e-05, "data_time": 0.001514577865600586, "loss": 0.3606869637966156, "time": 0.6381826877593995, "epoch": 9, "memory": 44138, "step": 33156} +{"lr": 1.1073384619689717e-05, "data_time": 0.0014742612838745117, "loss": 0.35385514199733736, "time": 0.6374922275543213, "epoch": 9, "memory": 44138, "step": 33256} +{"lr": 1.1037430884381023e-05, "data_time": 0.0013845443725585937, "loss": 0.39900938272476194, "time": 0.6385854005813598, "epoch": 9, "memory": 44138, "step": 33356} +{"lr": 1.100205663527591e-05, "data_time": 0.0015089035034179688, "loss": 0.37683402597904203, "time": 0.6385986804962158, "epoch": 9, "memory": 44138, "step": 33456} +{"lr": 1.0967264159541028e-05, "data_time": 0.0014016151428222657, "loss": 0.37027874290943147, "time": 0.6383233785629272, "epoch": 9, "memory": 44138, "step": 33556} +{"lr": 1.0933055706727766e-05, "data_time": 0.0014101266860961914, "loss": 0.3779666215181351, "time": 0.6401222229003907, "epoch": 9, "memory": 44138, "step": 33656} +{"lr": 1.089943348862675e-05, "data_time": 0.0015568733215332031, "loss": 0.3759250670671463, "time": 0.6357459306716919, "epoch": 9, "memory": 44138, "step": 33756} +{"lr": 1.0866399679124873e-05, "data_time": 0.0014931678771972657, "loss": 0.37015424072742464, "time": 0.6367595672607422, "epoch": 9, "memory": 44138, "step": 33856} +{"lr": 1.083395641406474e-05, "data_time": 0.0014409065246582032, "loss": 0.34707754850387573, "time": 0.6379493236541748, "epoch": 9, "memory": 44138, "step": 33956} +{"lr": 1.0802105791106526e-05, "data_time": 0.0015029191970825195, "loss": 0.40765505731105806, "time": 0.6377553701400757, "epoch": 9, "memory": 44138, "step": 34056} +{"lr": 1.077084986959243e-05, "data_time": 0.0016844511032104493, "loss": 0.3520644873380661, "time": 0.6377681732177735, "epoch": 9, "memory": 44138, "step": 34156} +{"lr": 1.0740190670413442e-05, "data_time": 0.001387310028076172, "loss": 0.40866038799285886, "time": 0.6372867584228515, "epoch": 9, "memory": 44138, "step": 34256} +{"lr": 1.0710130175878743e-05, "data_time": 0.0014751672744750977, "loss": 0.35814627110958097, "time": 0.6534881591796875, "epoch": 9, "memory": 44138, "step": 34356} +{"lr": 1.0680670329587481e-05, "data_time": 0.0014291048049926759, "loss": 0.39215677678585054, "time": 0.6393326759338379, "epoch": 9, "memory": 44138, "step": 34456} +{"lr": 1.0651813036303164e-05, "data_time": 0.0016228675842285157, "loss": 0.37324203848838805, "time": 0.6415825605392456, "epoch": 9, "memory": 44138, "step": 34556} +{"lr": 1.062356016183046e-05, "data_time": 0.0015181541442871095, "loss": 0.3640313982963562, "time": 0.6368334054946899, "epoch": 9, "memory": 44138, "step": 34656} +{"lr": 1.0595913532894587e-05, "data_time": 0.0015727519989013673, "loss": 0.36915597021579744, "time": 0.6363588094711303, "epoch": 9, "memory": 44138, "step": 34756} +{"lr": 1.0568874937023192e-05, "data_time": 0.0014317989349365234, "loss": 0.3812170743942261, "time": 0.6373331069946289, "epoch": 9, "memory": 44138, "step": 34856} +{"lr": 1.0542446122430764e-05, "data_time": 0.0013901233673095704, "loss": 0.3598202347755432, "time": 0.6376536607742309, "epoch": 9, "memory": 44138, "step": 34956} +{"lr": 1.0516628797905623e-05, "data_time": 0.001470470428466797, "loss": 0.3528856784105301, "time": 0.6370012760162354, "epoch": 9, "memory": 44138, "step": 35056} +{"lr": 1.0491424632699434e-05, "data_time": 0.0013904094696044922, "loss": 0.3287424951791763, "time": 0.6387446880340576, "epoch": 9, "memory": 44138, "step": 35156} +{"accuracy/top1": 79.16217041015625, "data_time": 0.03243000753994646, "time": 1.2745897399968114, "step": 9} +{"lr": 1.0465137065246831e-05, "data_time": 0.0015178918838500977, "loss": 0.37640154659748076, "time": 0.6357495784759521, "epoch": 10, "memory": 44138, "step": 35263} +{"lr": 1.0441207272775563e-05, "data_time": 0.0015198230743408204, "loss": 0.38571797311306, "time": 0.6365582942962646, "epoch": 10, "memory": 44138, "step": 35363} +{"lr": 1.0417895516097024e-05, "data_time": 0.0015020370483398438, "loss": 0.3528754532337189, "time": 0.6387616872787476, "epoch": 10, "memory": 44138, "step": 35463} +{"lr": 1.0395203302462216e-05, "data_time": 0.0014327287673950196, "loss": 0.36181311011314393, "time": 0.6384918212890625, "epoch": 10, "memory": 44138, "step": 35563} +{"lr": 1.0373132099064833e-05, "data_time": 0.0014722585678100587, "loss": 0.39339700937271116, "time": 0.6373436689376831, "epoch": 10, "memory": 44138, "step": 35663} +{"lr": 1.0351683332946347e-05, "data_time": 0.0015189170837402344, "loss": 0.3513374745845795, "time": 0.6388654470443725, "epoch": 10, "memory": 44138, "step": 35763} +{"lr": 1.0330858390903745e-05, "data_time": 0.00150909423828125, "loss": 0.37314617931842803, "time": 0.63696768283844, "epoch": 10, "memory": 44138, "step": 35863} +{"lr": 1.0310658619399925e-05, "data_time": 0.0015008926391601562, "loss": 0.37610665559768675, "time": 0.6370523929595947, "epoch": 10, "memory": 44138, "step": 35963} +{"lr": 1.0291085324476576e-05, "data_time": 0.0015619516372680664, "loss": 0.36353162825107577, "time": 0.6374603271484375, "epoch": 10, "memory": 44138, "step": 36063} +{"lr": 1.0272139771669745e-05, "data_time": 0.0014326333999633788, "loss": 0.37460085153579714, "time": 0.6389540910720826, "epoch": 10, "memory": 44138, "step": 36163} +{"lr": 1.0253823185928048e-05, "data_time": 0.00157470703125, "loss": 0.3888337314128876, "time": 0.6390398740768433, "epoch": 10, "memory": 44138, "step": 36263} +{"lr": 1.0236136751533406e-05, "data_time": 0.0013630151748657226, "loss": 0.3701596975326538, "time": 0.638887357711792, "epoch": 10, "memory": 44138, "step": 36363} +{"lr": 1.0219081612024538e-05, "data_time": 0.0014132976531982422, "loss": 0.38940992653369905, "time": 0.6362844944000244, "epoch": 10, "memory": 44138, "step": 36463} +{"lr": 1.0202658870122986e-05, "data_time": 0.0015787363052368164, "loss": 0.37467766404151914, "time": 0.6374501943588257, "epoch": 10, "memory": 44138, "step": 36563} +{"lr": 1.0186869587661827e-05, "data_time": 0.0014269113540649413, "loss": 0.3911354035139084, "time": 0.6371725797653198, "epoch": 10, "memory": 44138, "step": 36663} +{"lr": 1.0171714785517018e-05, "data_time": 0.036666488647460936, "loss": 0.34564143419265747, "time": 0.6711091518402099, "epoch": 10, "memory": 44138, "step": 36763} +{"lr": 1.0157195443541394e-05, "data_time": 0.0013899803161621094, "loss": 0.3497059136629105, "time": 0.6376233100891113, "epoch": 10, "memory": 44138, "step": 36863} +{"lr": 1.0143312500501317e-05, "data_time": 0.0014259099960327148, "loss": 0.36678563356399535, "time": 0.6382873773574829, "epoch": 10, "memory": 44138, "step": 36963} +{"lr": 1.0130066854015972e-05, "data_time": 0.0014418840408325195, "loss": 0.3871037155389786, "time": 0.6492510318756104, "epoch": 10, "memory": 44138, "step": 37063} +{"lr": 1.0117459360499327e-05, "data_time": 0.0013692140579223632, "loss": 0.35543271601200105, "time": 0.6374026536941528, "epoch": 10, "memory": 44138, "step": 37163} +{"lr": 1.0105490835104763e-05, "data_time": 0.0013768911361694337, "loss": 0.3578047901391983, "time": 0.6393279075622559, "epoch": 10, "memory": 44138, "step": 37263} +{"lr": 1.0094162051672373e-05, "data_time": 0.0015633821487426758, "loss": 0.38107499182224275, "time": 0.6389130115509033, "epoch": 10, "memory": 44138, "step": 37363} +{"lr": 1.0083473742678929e-05, "data_time": 0.0015424489974975586, "loss": 0.37405459880828856, "time": 0.6388011455535889, "epoch": 10, "memory": 44138, "step": 37463} +{"lr": 1.0073426599190536e-05, "data_time": 0.001570272445678711, "loss": 0.36700065732002257, "time": 0.638599944114685, "epoch": 10, "memory": 44138, "step": 37563} +{"lr": 1.0064021270817907e-05, "data_time": 0.0014969825744628907, "loss": 0.390339133143425, "time": 0.6420160055160522, "epoch": 10, "memory": 44138, "step": 37663} +{"lr": 1.0055258365674413e-05, "data_time": 0.0016922950744628906, "loss": 0.37869915962219236, "time": 0.6423586130142211, "epoch": 10, "memory": 44138, "step": 37763} +{"lr": 1.0047138450336735e-05, "data_time": 0.001508331298828125, "loss": 0.38447908461093905, "time": 0.6390578269958496, "epoch": 10, "memory": 44138, "step": 37863} +{"lr": 1.0039662049808213e-05, "data_time": 0.001542806625366211, "loss": 0.35889176428318026, "time": 0.6374271154403687, "epoch": 10, "memory": 44138, "step": 37963} +{"lr": 1.0032829647484969e-05, "data_time": 0.0015082359313964844, "loss": 0.39252043068408965, "time": 0.6393031120300293, "epoch": 10, "memory": 44138, "step": 38063} +{"lr": 1.0026641685124577e-05, "data_time": 0.001631021499633789, "loss": 0.3934075653553009, "time": 0.6386359453201294, "epoch": 10, "memory": 44138, "step": 38163} +{"lr": 1.0021098562817543e-05, "data_time": 0.001405644416809082, "loss": 0.3611509084701538, "time": 0.6383473157882691, "epoch": 10, "memory": 44138, "step": 38263} +{"lr": 1.0016200638961424e-05, "data_time": 0.0015422821044921875, "loss": 0.33123467266559603, "time": 0.6385917663574219, "epoch": 10, "memory": 44138, "step": 38363} +{"lr": 1.001194823023767e-05, "data_time": 0.001431584358215332, "loss": 0.3676534056663513, "time": 0.6369673252105713, "epoch": 10, "memory": 44138, "step": 38463} +{"lr": 1.0008341611591118e-05, "data_time": 0.0014977455139160156, "loss": 0.3790225267410278, "time": 0.6383319616317749, "epoch": 10, "memory": 44138, "step": 38563} +{"lr": 1.0005381016212259e-05, "data_time": 0.0016683340072631836, "loss": 0.3796721279621124, "time": 0.640384316444397, "epoch": 10, "memory": 44138, "step": 38663} +{"lr": 1.0003066635522105e-05, "data_time": 0.001898980140686035, "loss": 0.35755579471588134, "time": 0.6484117984771729, "epoch": 10, "memory": 44138, "step": 38763} +{"lr": 1.0001398619159851e-05, "data_time": 0.001570272445678711, "loss": 0.38691721558570863, "time": 0.6406832695007324, "epoch": 10, "memory": 44138, "step": 38863} +{"lr": 1.0000377074973214e-05, "data_time": 0.0014528751373291016, "loss": 0.35122773945331576, "time": 0.6390617609024047, "epoch": 10, "memory": 44138, "step": 38963} +{"lr": 1.0000002069011414e-05, "data_time": 0.0013076305389404298, "loss": 0.35887025892734525, "time": 0.6382135152816772, "epoch": 10, "memory": 44138, "step": 39063} +{"accuracy/top1": 78.66462707519531, "data_time": 0.031674109656235266, "time": 1.2740361443881332, "step": 10} diff --git a/clip_large_pretrain_4x256_sdv2_lr3e-5/20230606_052112/vis_data/config.py b/clip_large_pretrain_4x256_sdv2_lr3e-5/20230606_052112/vis_data/config.py new file mode 100644 index 0000000000000000000000000000000000000000..8f40f4e1084c3f11ae10e85e06719878b6741bd8 --- /dev/null +++ b/clip_large_pretrain_4x256_sdv2_lr3e-5/20230606_052112/vis_data/config.py @@ -0,0 +1,236 @@ +optim_wrapper = dict( + optimizer=dict( + type='AdamW', lr=3e-05, weight_decay=0.3, _scope_='mmpretrain'), + paramwise_cfg=dict( + custom_keys=dict({ + '.cls_token': dict(decay_mult=0.0), + '.pos_embed': dict(decay_mult=0.0) + })), + type='AmpOptimWrapper', + dtype='bfloat16', + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=False, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=4096) +model = dict( + type='ImageClassifier', + backbone=dict( + frozen_stages=24, + type='VisionTransformer', + arch='l', + img_size=224, + patch_size=14, + drop_rate=0.1, + pre_norm=True, + final_norm=False, + init_cfg=dict( + type='Pretrained', + checkpoint='ckpt/openclip-ViT-L-14.pth', + prefix='backbone')), + neck=dict( + type='CLIPProjection', + in_channels=1024, + out_channels=768, + init_cfg=dict( + type='Pretrained', + checkpoint='ckpt/openclip-ViT-L-14.pth', + prefix='backbone')), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=768, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + init_cfg=None), + init_cfg=dict( + type='TruncNormal', layer=['Conv2d', 'Linear'], std=0.02, bias=0.0), + train_cfg=None) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=128, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stablediffusionV2-1-dpmsolver-25-1m.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/cc1m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=128, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV2-1-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=128, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV2-1-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/clip_large_pretrain_4x256_sdv2_lr3e-5' diff --git a/clip_large_pretrain_4x256_sdv2_lr3e-5/20230606_052112/vis_data/events.out.tfevents.1686000089.SH-IDC1-10-140-24-99.120679.0 b/clip_large_pretrain_4x256_sdv2_lr3e-5/20230606_052112/vis_data/events.out.tfevents.1686000089.SH-IDC1-10-140-24-99.120679.0 new file mode 100644 index 0000000000000000000000000000000000000000..89c6f3a799b3deadc8dda866c9f4584b7b085537 --- /dev/null +++ b/clip_large_pretrain_4x256_sdv2_lr3e-5/20230606_052112/vis_data/events.out.tfevents.1686000089.SH-IDC1-10-140-24-99.120679.0 @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:679c6a08f0f7aa206b2cf8be0917ca22c2019db7388dc34aadf129d4dba44992 +size 1657522 diff --git a/clip_large_pretrain_4x256_sdv2_lr3e-5/20230606_052112/vis_data/scalars.json b/clip_large_pretrain_4x256_sdv2_lr3e-5/20230606_052112/vis_data/scalars.json new file mode 100644 index 0000000000000000000000000000000000000000..519c3486db56bfc3ded9e06a8dc5dd87e2bd9579 --- /dev/null +++ b/clip_large_pretrain_4x256_sdv2_lr3e-5/20230606_052112/vis_data/scalars.json @@ -0,0 +1,400 @@ +{"lr": 2.999968315196157e-05, "data_time": 0.0017328262329101562, "loss": 0.5820598602294922, "time": 0.6326795101165772, "epoch": 1, "memory": 44139, "step": 100} +{"lr": 2.999871979413257e-05, "data_time": 0.0017274379730224609, "loss": 0.5244008064270019, "time": 0.6465194940567016, "epoch": 1, "memory": 44139, "step": 200} +{"lr": 2.999710995647179e-05, "data_time": 0.0015012741088867188, "loss": 0.5025475025177002, "time": 0.6346629619598388, "epoch": 1, "memory": 44139, "step": 300} +{"lr": 2.9994853743065302e-05, "data_time": 0.001559281349182129, "loss": 0.5019857287406921, "time": 0.6351841926574707, "epoch": 1, "memory": 44139, "step": 400} +{"lr": 2.9991951299791448e-05, "data_time": 0.001495051383972168, "loss": 0.4892808198928833, "time": 0.6358315467834472, "epoch": 1, "memory": 44139, "step": 500} +{"lr": 2.9988402814311378e-05, "data_time": 0.0014707803726196288, "loss": 0.4940012454986572, "time": 0.6359200716018677, "epoch": 1, "memory": 44139, "step": 600} +{"lr": 2.998420851605688e-05, "data_time": 0.0014953374862670898, "loss": 0.4516049861907959, "time": 0.6355729103088379, "epoch": 1, "memory": 44139, "step": 700} +{"lr": 2.99793686762156e-05, "data_time": 0.0016302108764648438, "loss": 0.46528034210205077, "time": 0.6352574586868286, "epoch": 1, "memory": 44139, "step": 800} +{"lr": 2.997388360771344e-05, "data_time": 0.0015640497207641602, "loss": 0.45623364448547366, "time": 0.6363352298736572, "epoch": 1, "memory": 44139, "step": 900} +{"lr": 2.9967753665194455e-05, "data_time": 0.0014910221099853516, "loss": 0.44340091943740845, "time": 0.6351892471313476, "epoch": 1, "memory": 44139, "step": 1000} +{"lr": 2.996097924499781e-05, "data_time": 0.0014334678649902343, "loss": 0.45766860246658325, "time": 0.6361861228942871, "epoch": 1, "memory": 44139, "step": 1100} +{"lr": 2.9953560785132187e-05, "data_time": 0.0016579389572143554, "loss": 0.45582705140113833, "time": 0.6349462747573853, "epoch": 1, "memory": 44139, "step": 1200} +{"lr": 2.994549876524743e-05, "data_time": 0.001495194435119629, "loss": 0.466318154335022, "time": 0.634421706199646, "epoch": 1, "memory": 44139, "step": 1300} +{"lr": 2.99367937066036e-05, "data_time": 0.0015588760375976562, "loss": 0.4483098328113556, "time": 0.6346314668655395, "epoch": 1, "memory": 44139, "step": 1400} +{"lr": 2.9927446172037264e-05, "data_time": 0.0014845848083496094, "loss": 0.44086949825286864, "time": 0.6364420890808106, "epoch": 1, "memory": 44139, "step": 1500} +{"lr": 2.9917456765925056e-05, "data_time": 0.0015438556671142577, "loss": 0.4376753866672516, "time": 0.6369487285614014, "epoch": 1, "memory": 44139, "step": 1600} +{"lr": 2.990682613414459e-05, "data_time": 0.001448678970336914, "loss": 0.4680232286453247, "time": 0.6341166734695435, "epoch": 1, "memory": 44139, "step": 1700} +{"lr": 2.989555496403281e-05, "data_time": 0.0014559268951416016, "loss": 0.436040461063385, "time": 0.6355781078338623, "epoch": 1, "memory": 44139, "step": 1800} +{"lr": 2.9883643984341356e-05, "data_time": 0.001564788818359375, "loss": 0.4360331892967224, "time": 0.6353567123413086, "epoch": 1, "memory": 44139, "step": 1900} +{"lr": 2.987109396518962e-05, "data_time": 0.0015111923217773437, "loss": 0.4762925148010254, "time": 0.6346500396728516, "epoch": 1, "memory": 44139, "step": 2000} +{"lr": 2.9857905718015028e-05, "data_time": 0.0014391422271728515, "loss": 0.4315656542778015, "time": 0.6356466770172119, "epoch": 1, "memory": 44139, "step": 2100} +{"lr": 2.984408009552021e-05, "data_time": 0.0015337705612182618, "loss": 0.4738268882036209, "time": 0.6370855808258057, "epoch": 1, "memory": 44139, "step": 2200} +{"lr": 2.9829617991618262e-05, "data_time": 0.0015912532806396484, "loss": 0.4564800500869751, "time": 0.636294960975647, "epoch": 1, "memory": 44139, "step": 2300} +{"lr": 2.981452034137478e-05, "data_time": 0.0015359640121459961, "loss": 0.4229035317897797, "time": 0.6416321277618409, "epoch": 1, "memory": 44139, "step": 2400} +{"lr": 2.979878812094729e-05, "data_time": 0.0013695240020751953, "loss": 0.45235815048217776, "time": 0.6351548194885254, "epoch": 1, "memory": 44139, "step": 2500} +{"lr": 2.978242234752238e-05, "data_time": 0.0015182256698608398, "loss": 0.42996309995651244, "time": 0.6344455242156982, "epoch": 1, "memory": 44139, "step": 2600} +{"lr": 2.9765424079249748e-05, "data_time": 0.001532292366027832, "loss": 0.4235107421875, "time": 0.6349280834197998, "epoch": 1, "memory": 44139, "step": 2700} +{"lr": 2.9747794415173894e-05, "data_time": 0.001432490348815918, "loss": 0.443260133266449, "time": 0.6347655296325684, "epoch": 1, "memory": 44139, "step": 2800} +{"lr": 2.972953449516291e-05, "data_time": 0.001365804672241211, "loss": 0.45572861433029177, "time": 0.6348637580871582, "epoch": 1, "memory": 44139, "step": 2900} +{"lr": 2.9710645499835012e-05, "data_time": 0.0014687538146972655, "loss": 0.4245560824871063, "time": 0.6355278491973877, "epoch": 1, "memory": 44139, "step": 3000} +{"lr": 2.969112865048194e-05, "data_time": 0.0014769077301025391, "loss": 0.4595903933048248, "time": 0.6352854490280151, "epoch": 1, "memory": 44139, "step": 3100} +{"lr": 2.9670985208990228e-05, "data_time": 0.001611328125, "loss": 0.45638911724090575, "time": 0.6354909420013428, "epoch": 1, "memory": 44139, "step": 3200} +{"lr": 2.9650216477759467e-05, "data_time": 0.0015012741088867188, "loss": 0.43623468577861785, "time": 0.6347058534622192, "epoch": 1, "memory": 44139, "step": 3300} +{"lr": 2.9628823799618136e-05, "data_time": 0.0016749143600463868, "loss": 0.4210712373256683, "time": 0.635877537727356, "epoch": 1, "memory": 44139, "step": 3400} +{"lr": 2.960680855773683e-05, "data_time": 0.0014918088912963868, "loss": 0.43551710844039915, "time": 0.6353757381439209, "epoch": 1, "memory": 44139, "step": 3500} +{"lr": 2.958417217553874e-05, "data_time": 0.0016278743743896485, "loss": 0.44734391272068025, "time": 0.6348343610763549, "epoch": 1, "memory": 44139, "step": 3600} +{"lr": 2.956091611660774e-05, "data_time": 0.001421642303466797, "loss": 0.4362662613391876, "time": 0.6364853382110596, "epoch": 1, "memory": 44139, "step": 3700} +{"lr": 2.9537041884593564e-05, "data_time": 0.0015681982040405273, "loss": 0.433257931470871, "time": 0.6355493545532227, "epoch": 1, "memory": 44139, "step": 3800} +{"lr": 2.9512551023114816e-05, "data_time": 0.0012206077575683595, "loss": 0.41935579776763915, "time": 0.6346648693084717, "epoch": 1, "memory": 44139, "step": 3900} +{"accuracy/top1": 80.07376098632812, "data_time": 0.04794620212755705, "time": 1.342139587067721, "step": 1} +{"lr": 2.9485664710372417e-05, "data_time": 0.0015699625015258788, "loss": 0.46139551103115084, "time": 0.6367982387542724, "epoch": 2, "memory": 44139, "step": 4007} +{"lr": 2.9459902502419427e-05, "data_time": 0.0015330553054809571, "loss": 0.41834096908569335, "time": 0.6372418403625488, "epoch": 2, "memory": 44138, "step": 4107} +{"lr": 2.9433528652545812e-05, "data_time": 0.0013655662536621095, "loss": 0.4502277851104736, "time": 0.6382653474807739, "epoch": 2, "memory": 44138, "step": 4207} +{"lr": 2.9406544865986134e-05, "data_time": 0.0013972043991088868, "loss": 0.45304155349731445, "time": 0.6351974248886109, "epoch": 2, "memory": 44138, "step": 4307} +{"lr": 2.9378952887411096e-05, "data_time": 0.0014516353607177735, "loss": 0.4170908659696579, "time": 0.6357728958129882, "epoch": 2, "memory": 44138, "step": 4407} +{"lr": 2.935075450081489e-05, "data_time": 0.0014037132263183595, "loss": 0.42441895306110383, "time": 0.6353183031082154, "epoch": 2, "memory": 44138, "step": 4507} +{"lr": 2.9321951529399665e-05, "data_time": 0.0015067815780639648, "loss": 0.40530370473861693, "time": 0.6356248617172241, "epoch": 2, "memory": 44138, "step": 4607} +{"lr": 2.9292545835457945e-05, "data_time": 0.001408529281616211, "loss": 0.4169813871383667, "time": 0.6364418983459472, "epoch": 2, "memory": 44138, "step": 4707} +{"lr": 2.9262539320251868e-05, "data_time": 0.0013746023178100586, "loss": 0.4180146545171738, "time": 0.6357547760009765, "epoch": 2, "memory": 44138, "step": 4807} +{"lr": 2.923193392389054e-05, "data_time": 0.0016022205352783203, "loss": 0.40583510100841524, "time": 0.6371466875076294, "epoch": 2, "memory": 44138, "step": 4907} +{"lr": 2.9200731625204435e-05, "data_time": 0.0016779899597167969, "loss": 0.41511437594890593, "time": 0.6361659526824951, "epoch": 2, "memory": 44138, "step": 5007} +{"lr": 2.9168934441617457e-05, "data_time": 0.0015182256698608398, "loss": 0.4329532742500305, "time": 0.6365081787109375, "epoch": 2, "memory": 44138, "step": 5107} +{"lr": 2.9136544429016667e-05, "data_time": 0.0014085769653320312, "loss": 0.4559467166662216, "time": 0.6362128734588623, "epoch": 2, "memory": 44138, "step": 5207} +{"lr": 2.9103563681619084e-05, "data_time": 0.0014522075653076172, "loss": 0.41591658890247346, "time": 0.6360562562942504, "epoch": 2, "memory": 44138, "step": 5307} +{"lr": 2.9069994331836577e-05, "data_time": 0.0014405965805053711, "loss": 0.4258942365646362, "time": 0.6373616456985474, "epoch": 2, "memory": 44138, "step": 5407} +{"lr": 2.9035838550137736e-05, "data_time": 0.0015010833740234375, "loss": 0.415695384144783, "time": 0.6360781908035278, "epoch": 2, "memory": 44138, "step": 5507} +{"lr": 2.9001098544907622e-05, "data_time": 0.0015009641647338867, "loss": 0.42156079411506653, "time": 0.6378882646560669, "epoch": 2, "memory": 44138, "step": 5607} +{"lr": 2.8965776562305154e-05, "data_time": 0.0014098405838012696, "loss": 0.4137383341789246, "time": 0.6360904455184937, "epoch": 2, "memory": 44138, "step": 5707} +{"lr": 2.892987488611766e-05, "data_time": 0.0013931989669799805, "loss": 0.4191013127565384, "time": 0.635969614982605, "epoch": 2, "memory": 44138, "step": 5807} +{"lr": 2.88933958376132e-05, "data_time": 0.0015510797500610351, "loss": 0.42429515421390535, "time": 0.6362807750701904, "epoch": 2, "memory": 44138, "step": 5907} +{"lr": 2.8856341775390632e-05, "data_time": 0.001604008674621582, "loss": 0.4397385746240616, "time": 0.6360028743743896, "epoch": 2, "memory": 44138, "step": 6007} +{"lr": 2.8818715095227025e-05, "data_time": 0.0014252901077270509, "loss": 0.3885278284549713, "time": 0.6349270582199097, "epoch": 2, "memory": 44138, "step": 6107} +{"lr": 2.878051822992295e-05, "data_time": 0.0013510465621948242, "loss": 0.4366811364889145, "time": 0.6370023488998413, "epoch": 2, "memory": 44138, "step": 6207} +{"lr": 2.874175364914476e-05, "data_time": 0.0014041900634765626, "loss": 0.4364030212163925, "time": 0.6357920408248902, "epoch": 2, "memory": 44138, "step": 6307} +{"lr": 2.870242385926531e-05, "data_time": 0.0014720439910888671, "loss": 0.4201664745807648, "time": 0.6351218461990357, "epoch": 2, "memory": 44138, "step": 6407} +{"lr": 2.866253140320176e-05, "data_time": 0.0015051126480102538, "loss": 0.4343326359987259, "time": 0.6362585067749024, "epoch": 2, "memory": 44138, "step": 6507} +{"lr": 2.862207886025111e-05, "data_time": 0.0014474868774414062, "loss": 0.4557327330112457, "time": 0.6358387231826782, "epoch": 2, "memory": 44138, "step": 6607} +{"lr": 2.8581068845923533e-05, "data_time": 0.0014994144439697266, "loss": 0.41012815833091737, "time": 0.6372246503829956, "epoch": 2, "memory": 44138, "step": 6707} +{"lr": 2.8539504011773207e-05, "data_time": 0.0014214038848876954, "loss": 0.42677571177482604, "time": 0.6366636514663696, "epoch": 2, "memory": 44138, "step": 6807} +{"lr": 2.849738704522689e-05, "data_time": 0.0014622211456298828, "loss": 0.4198896884918213, "time": 0.6367005825042724, "epoch": 2, "memory": 44138, "step": 6907} +{"lr": 2.845472066941013e-05, "data_time": 0.001542377471923828, "loss": 0.42949950098991396, "time": 0.6354840755462646, "epoch": 2, "memory": 44138, "step": 7007} +{"lr": 2.8411507642971194e-05, "data_time": 0.001567983627319336, "loss": 0.42264110743999483, "time": 0.6358386516571045, "epoch": 2, "memory": 44138, "step": 7107} +{"lr": 2.836775075990287e-05, "data_time": 0.0014949560165405274, "loss": 0.41218724846839905, "time": 0.6354932069778443, "epoch": 2, "memory": 44138, "step": 7207} +{"lr": 2.8323452849361518e-05, "data_time": 0.0015042543411254883, "loss": 0.4387490481138229, "time": 0.6351249694824219, "epoch": 2, "memory": 44138, "step": 7307} +{"lr": 2.827861677548442e-05, "data_time": 0.0014189481735229492, "loss": 0.43870543539524076, "time": 0.6362674236297607, "epoch": 2, "memory": 44138, "step": 7407} +{"lr": 2.8233245437204413e-05, "data_time": 0.0016158103942871093, "loss": 0.4182025045156479, "time": 0.636237359046936, "epoch": 2, "memory": 44138, "step": 7507} +{"lr": 2.8187341768062594e-05, "data_time": 0.0014983892440795898, "loss": 0.43157645463943484, "time": 0.6355327367782593, "epoch": 2, "memory": 44138, "step": 7607} +{"lr": 2.8140908736018534e-05, "data_time": 0.001468658447265625, "loss": 0.3919978946447372, "time": 0.6385063171386719, "epoch": 2, "memory": 44138, "step": 7707} +{"lr": 2.8093949343258448e-05, "data_time": 0.001326274871826172, "loss": 0.3885378032922745, "time": 0.6353389501571656, "epoch": 2, "memory": 44138, "step": 7807} +{"accuracy/top1": 81.8169174194336, "data_time": 0.03494371216872643, "time": 1.2779911715408852, "step": 2} +{"lr": 2.8043123316419856e-05, "data_time": 0.0024315595626831056, "loss": 0.41362959444522857, "time": 0.6347774267196655, "epoch": 3, "memory": 44138, "step": 7914} +{"lr": 2.799508404275313e-05, "data_time": 0.0017800331115722656, "loss": 0.39586409032344816, "time": 0.6360813617706299, "epoch": 3, "memory": 44138, "step": 8014} +{"lr": 2.7946527836849776e-05, "data_time": 0.001385068893432617, "loss": 0.39395827054977417, "time": 0.6360595703125, "epoch": 3, "memory": 44138, "step": 8114} +{"lr": 2.7897457838172418e-05, "data_time": 0.0017013788223266602, "loss": 0.4298724889755249, "time": 0.6363627433776855, "epoch": 3, "memory": 44138, "step": 8214} +{"lr": 2.784787721940371e-05, "data_time": 0.0015061378479003906, "loss": 0.41413705348968505, "time": 0.6381935119628906, "epoch": 3, "memory": 44138, "step": 8314} +{"lr": 2.7797789186241074e-05, "data_time": 0.0017046451568603516, "loss": 0.41770114898681643, "time": 0.6368783235549926, "epoch": 3, "memory": 44138, "step": 8414} +{"lr": 2.774719697718944e-05, "data_time": 0.0015776872634887696, "loss": 0.42050252556800843, "time": 0.6355265855789185, "epoch": 3, "memory": 44138, "step": 8514} +{"lr": 2.7696103863351812e-05, "data_time": 0.0016794443130493165, "loss": 0.40208789706230164, "time": 0.6353798866271972, "epoch": 3, "memory": 44138, "step": 8614} +{"lr": 2.7644513148217896e-05, "data_time": 0.0016884088516235351, "loss": 0.3941322654485703, "time": 0.637868094444275, "epoch": 3, "memory": 44138, "step": 8714} +{"lr": 2.759242816745033e-05, "data_time": 0.0016788959503173829, "loss": 0.3778026938438416, "time": 0.6356910943984986, "epoch": 3, "memory": 44138, "step": 8814} +{"lr": 2.7539852288669356e-05, "data_time": 0.001613783836364746, "loss": 0.4226143151521683, "time": 0.6357138156890869, "epoch": 3, "memory": 44138, "step": 8914} +{"lr": 2.7486788911234612e-05, "data_time": 0.0015007734298706054, "loss": 0.41931558549404147, "time": 0.6348323106765748, "epoch": 3, "memory": 44138, "step": 9014} +{"lr": 2.7433241466025614e-05, "data_time": 0.0013763904571533203, "loss": 0.3943664938211441, "time": 0.6383355617523193, "epoch": 3, "memory": 44138, "step": 9114} +{"lr": 2.7379213415219968e-05, "data_time": 0.0014757394790649414, "loss": 0.4155552923679352, "time": 0.6354357957839966, "epoch": 3, "memory": 44138, "step": 9214} +{"lr": 2.732470825206939e-05, "data_time": 0.001357889175415039, "loss": 0.39439934194087983, "time": 0.6365950584411622, "epoch": 3, "memory": 44138, "step": 9314} +{"lr": 2.7269729500673918e-05, "data_time": 0.001461195945739746, "loss": 0.3985852748155594, "time": 0.6367216348648072, "epoch": 3, "memory": 44138, "step": 9414} +{"lr": 2.7214280715754e-05, "data_time": 0.0014960050582885742, "loss": 0.3905987113714218, "time": 0.6351579666137696, "epoch": 3, "memory": 44138, "step": 9514} +{"lr": 2.7158365482420722e-05, "data_time": 0.0014636516571044922, "loss": 0.37036815881729124, "time": 0.6367456912994385, "epoch": 3, "memory": 44138, "step": 9614} +{"lr": 2.7101987415944003e-05, "data_time": 0.00142214298248291, "loss": 0.39271018505096433, "time": 0.635669493675232, "epoch": 3, "memory": 44138, "step": 9714} +{"lr": 2.7045150161518795e-05, "data_time": 0.0013616561889648437, "loss": 0.42946577072143555, "time": 0.6355520248413086, "epoch": 3, "memory": 44138, "step": 9814} +{"lr": 2.6987857394029402e-05, "data_time": 0.001372051239013672, "loss": 0.39274133145809176, "time": 0.6365905284881592, "epoch": 3, "memory": 44138, "step": 9914} +{"lr": 2.6930112817811945e-05, "data_time": 0.0014690399169921876, "loss": 0.38676917552948, "time": 0.6366588115692139, "epoch": 3, "memory": 44138, "step": 10014} +{"lr": 2.6871920166414833e-05, "data_time": 0.0014314889907836915, "loss": 0.40503534972667693, "time": 0.6373430013656616, "epoch": 3, "memory": 44138, "step": 10114} +{"lr": 2.6813283202357177e-05, "data_time": 0.0015382051467895507, "loss": 0.4070036053657532, "time": 0.6353307485580444, "epoch": 3, "memory": 44138, "step": 10214} +{"lr": 2.675420571688591e-05, "data_time": 0.0015157222747802734, "loss": 0.4014215022325516, "time": 0.6362082958221436, "epoch": 3, "memory": 44138, "step": 10314} +{"lr": 2.6694691529730227e-05, "data_time": 0.0015067100524902345, "loss": 0.4042677044868469, "time": 0.6353816986083984, "epoch": 3, "memory": 44138, "step": 10414} +{"lr": 2.6634744488854987e-05, "data_time": 0.0015247106552124024, "loss": 0.3955463647842407, "time": 0.6370667695999146, "epoch": 3, "memory": 44138, "step": 10514} +{"lr": 2.657436847021168e-05, "data_time": 0.0016013145446777343, "loss": 0.39692915976047516, "time": 0.6353601455688477, "epoch": 3, "memory": 44138, "step": 10614} +{"lr": 2.6513567377487876e-05, "data_time": 0.0015070199966430663, "loss": 0.3878687500953674, "time": 0.6440400123596192, "epoch": 3, "memory": 44138, "step": 10714} +{"lr": 2.645234514185487e-05, "data_time": 0.0015334606170654297, "loss": 0.41319349110126496, "time": 0.6355741500854493, "epoch": 3, "memory": 44138, "step": 10814} +{"lr": 2.6390705721713513e-05, "data_time": 0.0014587163925170899, "loss": 0.39748058319091795, "time": 0.6354933738708496, "epoch": 3, "memory": 44138, "step": 10914} +{"lr": 2.6328653102438134e-05, "data_time": 0.001739335060119629, "loss": 0.41561255156993865, "time": 0.6355241298675537, "epoch": 3, "memory": 44138, "step": 11014} +{"lr": 2.626619129611909e-05, "data_time": 0.001514267921447754, "loss": 0.41075489223003386, "time": 0.6360374689102173, "epoch": 3, "memory": 44138, "step": 11114} +{"lr": 2.6203324341303198e-05, "data_time": 0.001657247543334961, "loss": 0.39961858093738556, "time": 0.6376914262771607, "epoch": 3, "memory": 44138, "step": 11214} +{"lr": 2.6140056302732683e-05, "data_time": 0.0014661550521850586, "loss": 0.40717593431472776, "time": 0.6385722637176514, "epoch": 3, "memory": 44138, "step": 11314} +{"lr": 2.6076391271082312e-05, "data_time": 0.0015054464340209961, "loss": 0.39594255983829496, "time": 0.6378923892974854, "epoch": 3, "memory": 44138, "step": 11414} +{"lr": 2.6012333362694966e-05, "data_time": 0.0014590740203857422, "loss": 0.4010176151990891, "time": 0.6353793621063233, "epoch": 3, "memory": 44138, "step": 11514} +{"lr": 2.5947886719315454e-05, "data_time": 0.0015369176864624024, "loss": 0.39961256980896, "time": 0.6385421276092529, "epoch": 3, "memory": 44138, "step": 11614} +{"lr": 2.58830555078227e-05, "data_time": 0.0012540578842163085, "loss": 0.3792997360229492, "time": 0.6358986854553222, "epoch": 3, "memory": 44138, "step": 11714} +{"accuracy/top1": 81.46550750732422, "data_time": 0.03180969994643639, "time": 1.276645541191101, "step": 3} +{"lr": 2.5813264972520608e-05, "data_time": 0.0014603376388549806, "loss": 0.40517268776893617, "time": 0.6358929395675659, "epoch": 4, "memory": 44138, "step": 11821} +{"lr": 2.574765105195596e-05, "data_time": 0.0018689393997192382, "loss": 0.4040860444307327, "time": 0.6372992992401123, "epoch": 4, "memory": 44138, "step": 11921} +{"lr": 2.5681665509767622e-05, "data_time": 0.0014482498168945312, "loss": 0.39335237741470336, "time": 0.6357096433639526, "epoch": 4, "memory": 44138, "step": 12021} +{"lr": 2.561531261233396e-05, "data_time": 0.0016059637069702148, "loss": 0.4019911468029022, "time": 0.6355145454406739, "epoch": 4, "memory": 44138, "step": 12121} +{"lr": 2.554859664978521e-05, "data_time": 0.0014123201370239257, "loss": 0.4026492089033127, "time": 0.6374299764633179, "epoch": 4, "memory": 44138, "step": 12221} +{"lr": 2.5481521935725978e-05, "data_time": 0.001361989974975586, "loss": 0.3911616176366806, "time": 0.637604808807373, "epoch": 4, "memory": 44138, "step": 12321} +{"lr": 2.541409280695646e-05, "data_time": 0.001536726951599121, "loss": 0.4097395360469818, "time": 0.6359011650085449, "epoch": 4, "memory": 44138, "step": 12421} +{"lr": 2.5346313623192016e-05, "data_time": 0.0014959096908569336, "loss": 0.3840169131755829, "time": 0.6381452322006226, "epoch": 4, "memory": 44138, "step": 12521} +{"lr": 2.5278188766781202e-05, "data_time": 0.0014828205108642577, "loss": 0.40848673284053805, "time": 0.636052131652832, "epoch": 4, "memory": 44138, "step": 12621} +{"lr": 2.520972264242233e-05, "data_time": 0.0015128135681152343, "loss": 0.38745259344577787, "time": 0.6361779928207397, "epoch": 4, "memory": 44138, "step": 12721} +{"lr": 2.5140919676879075e-05, "data_time": 0.001426553726196289, "loss": 0.4014747738838196, "time": 0.6361227035522461, "epoch": 4, "memory": 44138, "step": 12821} +{"lr": 2.5071784318693913e-05, "data_time": 0.001404404640197754, "loss": 0.3918736457824707, "time": 0.6363578557968139, "epoch": 4, "memory": 44138, "step": 12921} +{"lr": 2.5002321037900605e-05, "data_time": 0.0013864278793334962, "loss": 0.396357935667038, "time": 0.6352404117584228, "epoch": 4, "memory": 44138, "step": 13021} +{"lr": 2.4932534325735087e-05, "data_time": 0.0014417648315429687, "loss": 0.3730785220861435, "time": 0.6372972249984741, "epoch": 4, "memory": 44138, "step": 13121} +{"lr": 2.486242869434525e-05, "data_time": 0.0018372535705566406, "loss": 0.39955129027366637, "time": 0.6367447137832641, "epoch": 4, "memory": 44138, "step": 13221} +{"lr": 2.4792008676499066e-05, "data_time": 0.001536393165588379, "loss": 0.40164371728897097, "time": 0.6356911182403564, "epoch": 4, "memory": 44138, "step": 13321} +{"lr": 2.472127882529154e-05, "data_time": 0.001662874221801758, "loss": 0.3860657066106796, "time": 0.6371049642562866, "epoch": 4, "memory": 44138, "step": 13421} +{"lr": 2.4650243713850305e-05, "data_time": 0.0015432596206665038, "loss": 0.38913559913635254, "time": 0.63596670627594, "epoch": 4, "memory": 44138, "step": 13521} +{"lr": 2.4578907935040093e-05, "data_time": 0.0015214204788208008, "loss": 0.37403162717819216, "time": 0.6397278070449829, "epoch": 4, "memory": 44138, "step": 13621} +{"lr": 2.4507276101165596e-05, "data_time": 0.0016132593154907227, "loss": 0.3677979320287704, "time": 0.6453139781951904, "epoch": 4, "memory": 44138, "step": 13721} +{"lr": 2.4435352843673332e-05, "data_time": 0.0014202356338500976, "loss": 0.4099270522594452, "time": 0.6363095760345459, "epoch": 4, "memory": 44138, "step": 13821} +{"lr": 2.4363142812852158e-05, "data_time": 0.0013915538787841798, "loss": 0.39708671867847445, "time": 0.6355254888534546, "epoch": 4, "memory": 44138, "step": 13921} +{"lr": 2.4290650677532606e-05, "data_time": 0.001457071304321289, "loss": 0.3684999763965607, "time": 0.6364887952804565, "epoch": 4, "memory": 44138, "step": 14021} +{"lr": 2.4217881124785113e-05, "data_time": 0.0014864683151245117, "loss": 0.39890529215335846, "time": 0.6435006618499756, "epoch": 4, "memory": 44138, "step": 14121} +{"lr": 2.4144838859616835e-05, "data_time": 0.001450824737548828, "loss": 0.3913849234580994, "time": 0.6382109165191651, "epoch": 4, "memory": 44138, "step": 14221} +{"lr": 2.407152860466747e-05, "data_time": 0.0014965295791625976, "loss": 0.3884093701839447, "time": 0.6387155532836915, "epoch": 4, "memory": 44138, "step": 14321} +{"lr": 2.3997955099903982e-05, "data_time": 0.001557326316833496, "loss": 0.39054149985313413, "time": 0.6356399059295654, "epoch": 4, "memory": 44138, "step": 14421} +{"lr": 2.3924123102314014e-05, "data_time": 0.0013512849807739257, "loss": 0.40433686077594755, "time": 0.6356610059738159, "epoch": 4, "memory": 44138, "step": 14521} +{"lr": 2.385003738559849e-05, "data_time": 0.0014832258224487306, "loss": 0.3877225279808044, "time": 0.6363243579864502, "epoch": 4, "memory": 44138, "step": 14621} +{"lr": 2.3775702739862787e-05, "data_time": 0.0015548229217529296, "loss": 0.3914781719446182, "time": 0.635591459274292, "epoch": 4, "memory": 44138, "step": 14721} +{"lr": 2.370112397130714e-05, "data_time": 0.0014350414276123047, "loss": 0.38474114537239074, "time": 0.6357931137084961, "epoch": 4, "memory": 44138, "step": 14821} +{"lr": 2.3626305901915825e-05, "data_time": 0.0015352249145507812, "loss": 0.38193814754486083, "time": 0.6361060619354248, "epoch": 4, "memory": 44138, "step": 14921} +{"lr": 2.3551253369145433e-05, "data_time": 0.001481914520263672, "loss": 0.3836988091468811, "time": 0.6354096174240113, "epoch": 4, "memory": 44138, "step": 15021} +{"lr": 2.3475971225612054e-05, "data_time": 0.0014744043350219727, "loss": 0.40507255792617797, "time": 0.6389569044113159, "epoch": 4, "memory": 44138, "step": 15121} +{"lr": 2.3400464338777606e-05, "data_time": 0.0015366554260253906, "loss": 0.3861899584531784, "time": 0.6360882759094239, "epoch": 4, "memory": 44138, "step": 15221} +{"lr": 2.332473759063499e-05, "data_time": 0.0014726877212524413, "loss": 0.3677582621574402, "time": 0.636666226387024, "epoch": 4, "memory": 44138, "step": 15321} +{"lr": 2.3248795877392593e-05, "data_time": 0.001471090316772461, "loss": 0.3959373295307159, "time": 0.6355545997619629, "epoch": 4, "memory": 44138, "step": 15421} +{"lr": 2.317264410915764e-05, "data_time": 0.0014323949813842773, "loss": 0.39965182542800903, "time": 0.6355262279510498, "epoch": 4, "memory": 44138, "step": 15521} +{"lr": 2.3096287209618654e-05, "data_time": 0.0012560129165649415, "loss": 0.39307435154914855, "time": 0.6355045795440674, "epoch": 4, "memory": 44138, "step": 15621} +{"accuracy/top1": 80.31383514404297, "data_time": 0.032101787369826745, "time": 1.2763871694433278, "step": 4} +{"lr": 2.3014363749707527e-05, "data_time": 0.0015312433242797852, "loss": 0.3915721088647842, "time": 0.6358813762664794, "epoch": 5, "memory": 44138, "step": 15728} +{"lr": 2.2937597929948786e-05, "data_time": 0.0016015052795410156, "loss": 0.3880497306585312, "time": 0.6367979764938354, "epoch": 5, "memory": 44138, "step": 15828} +{"lr": 2.286064217609303e-05, "data_time": 0.0015599489212036132, "loss": 0.4007106781005859, "time": 0.6364808320999146, "epoch": 5, "memory": 44138, "step": 15928} +{"lr": 2.2783501463811504e-05, "data_time": 0.001565837860107422, "loss": 0.42414140701293945, "time": 0.6363464593887329, "epoch": 5, "memory": 44138, "step": 16028} +{"lr": 2.2706180780734223e-05, "data_time": 0.0015761137008666992, "loss": 0.3773953467607498, "time": 0.635633397102356, "epoch": 5, "memory": 44138, "step": 16128} +{"lr": 2.2628685126127375e-05, "data_time": 0.0015518665313720703, "loss": 0.3699106991291046, "time": 0.636089825630188, "epoch": 5, "memory": 44138, "step": 16228} +{"lr": 2.2551019510570224e-05, "data_time": 0.0016492605209350586, "loss": 0.4149298667907715, "time": 0.6366400957107544, "epoch": 5, "memory": 44138, "step": 16328} +{"lr": 2.247318895563104e-05, "data_time": 0.0014890193939208984, "loss": 0.3634637206792831, "time": 0.6381202697753906, "epoch": 5, "memory": 44138, "step": 16428} +{"lr": 2.2395198493542463e-05, "data_time": 0.0014709234237670898, "loss": 0.37360579073429107, "time": 0.6367754459381103, "epoch": 5, "memory": 44138, "step": 16528} +{"lr": 2.2317053166876117e-05, "data_time": 0.0014590978622436523, "loss": 0.36054129898548126, "time": 0.6359807014465332, "epoch": 5, "memory": 44138, "step": 16628} +{"lr": 2.223875802821662e-05, "data_time": 0.0014612197875976563, "loss": 0.3709414005279541, "time": 0.6374520778656005, "epoch": 5, "memory": 44138, "step": 16728} +{"lr": 2.2160318139834804e-05, "data_time": 0.0015422821044921875, "loss": 0.38987391591072085, "time": 0.6362518787384033, "epoch": 5, "memory": 44138, "step": 16828} +{"lr": 2.2081738573360552e-05, "data_time": 0.0014992237091064453, "loss": 0.3937067031860352, "time": 0.6383609533309936, "epoch": 5, "memory": 44138, "step": 16928} +{"lr": 2.20030244094548e-05, "data_time": 0.0014134645462036133, "loss": 0.37678520679473876, "time": 0.6390758991241455, "epoch": 5, "memory": 44138, "step": 17028} +{"lr": 2.192418073748101e-05, "data_time": 0.0014934539794921875, "loss": 0.40795170664787295, "time": 0.6361664533615112, "epoch": 5, "memory": 44138, "step": 17128} +{"lr": 2.1845212655176195e-05, "data_time": 0.0016251087188720703, "loss": 0.3827826589345932, "time": 0.637525224685669, "epoch": 5, "memory": 44138, "step": 17228} +{"lr": 2.1766125268321264e-05, "data_time": 0.0014215230941772461, "loss": 0.3655057817697525, "time": 0.6384395837783814, "epoch": 5, "memory": 44138, "step": 17328} +{"lr": 2.168692369041089e-05, "data_time": 0.0017093658447265626, "loss": 0.3965724974870682, "time": 0.6359135150909424, "epoch": 5, "memory": 44138, "step": 17428} +{"lr": 2.1607613042322935e-05, "data_time": 0.0014518260955810546, "loss": 0.3884787291288376, "time": 0.6355259895324707, "epoch": 5, "memory": 44138, "step": 17528} +{"lr": 2.1528198451987307e-05, "data_time": 0.001448845863342285, "loss": 0.37770710289478304, "time": 0.6357872724533081, "epoch": 5, "memory": 44138, "step": 17628} +{"lr": 2.1448685054054467e-05, "data_time": 0.001620650291442871, "loss": 0.38291923701763153, "time": 0.6368507146835327, "epoch": 5, "memory": 44138, "step": 17728} +{"lr": 2.1369077989563356e-05, "data_time": 0.0015523672103881837, "loss": 0.39392951428890227, "time": 0.639568567276001, "epoch": 5, "memory": 44138, "step": 17828} +{"lr": 2.128938240560911e-05, "data_time": 0.0014477252960205078, "loss": 0.38296268284320834, "time": 0.6376967668533325, "epoch": 5, "memory": 44138, "step": 17928} +{"lr": 2.120960345501016e-05, "data_time": 0.0014770746231079102, "loss": 0.407487154006958, "time": 0.6372606515884399, "epoch": 5, "memory": 44138, "step": 18028} +{"lr": 2.1129746295975105e-05, "data_time": 0.0014374494552612306, "loss": 0.39411935210227966, "time": 0.6363304376602172, "epoch": 5, "memory": 44138, "step": 18128} +{"lr": 2.1049816091769223e-05, "data_time": 0.0015367746353149414, "loss": 0.3669018536806107, "time": 0.636871600151062, "epoch": 5, "memory": 44138, "step": 18228} +{"lr": 2.096981801038062e-05, "data_time": 0.0016428470611572266, "loss": 0.3765005588531494, "time": 0.6359396696090698, "epoch": 5, "memory": 44138, "step": 18328} +{"lr": 2.0889757224186114e-05, "data_time": 0.001549363136291504, "loss": 0.3841983497142792, "time": 0.638393497467041, "epoch": 5, "memory": 44138, "step": 18428} +{"lr": 2.0809638909616704e-05, "data_time": 0.0014175176620483398, "loss": 0.408274906873703, "time": 0.6357414960861206, "epoch": 5, "memory": 44138, "step": 18528} +{"lr": 2.072946824682303e-05, "data_time": 0.0016066074371337891, "loss": 0.401069301366806, "time": 0.6359381914138794, "epoch": 5, "memory": 44138, "step": 18628} +{"lr": 2.0649250419340373e-05, "data_time": 0.0015076875686645507, "loss": 0.3510246902704239, "time": 0.6410531044006348, "epoch": 5, "memory": 44138, "step": 18728} +{"lr": 2.0568990613753462e-05, "data_time": 0.0015563488006591797, "loss": 0.38433206677436826, "time": 0.6372015953063965, "epoch": 5, "memory": 44138, "step": 18828} +{"lr": 2.0488694019361218e-05, "data_time": 0.001474452018737793, "loss": 0.3696636170148849, "time": 0.6371400833129883, "epoch": 5, "memory": 44138, "step": 18928} +{"lr": 2.0408365827841122e-05, "data_time": 0.001377701759338379, "loss": 0.36692739427089693, "time": 0.6382151365280151, "epoch": 5, "memory": 44138, "step": 19028} +{"lr": 2.0328011232913694e-05, "data_time": 0.00146026611328125, "loss": 0.37398011684417726, "time": 0.6375979900360107, "epoch": 5, "memory": 44138, "step": 19128} +{"lr": 2.024763543000651e-05, "data_time": 0.0015254259109497071, "loss": 0.3797903835773468, "time": 0.6355776309967041, "epoch": 5, "memory": 44138, "step": 19228} +{"lr": 2.0167243615918473e-05, "data_time": 0.0014070034027099609, "loss": 0.37698997259140016, "time": 0.6385165214538574, "epoch": 5, "memory": 44138, "step": 19328} +{"lr": 2.0086840988483663e-05, "data_time": 0.0014674663543701172, "loss": 0.3949741184711456, "time": 0.6355803489685059, "epoch": 5, "memory": 44138, "step": 19428} +{"lr": 2.00064327462353e-05, "data_time": 0.0012489080429077149, "loss": 0.3776142954826355, "time": 0.6377695798873901, "epoch": 5, "memory": 44138, "step": 19528} +{"accuracy/top1": 79.67363739013672, "data_time": 0.030968370108768857, "time": 1.273853819945763, "step": 5} +{"lr": 1.992039560076574e-05, "data_time": 0.0013307809829711914, "loss": 0.37925947904586793, "time": 0.6388778686523438, "epoch": 6, "memory": 44138, "step": 19635} +{"lr": 1.983999225510998e-05, "data_time": 0.0013992071151733398, "loss": 0.3913419783115387, "time": 0.6357207775115967, "epoch": 6, "memory": 44138, "step": 19735} +{"lr": 1.9759599254956633e-05, "data_time": 0.0014631271362304688, "loss": 0.4005438446998596, "time": 0.6363759994506836, "epoch": 6, "memory": 44138, "step": 19835} +{"lr": 1.9679221798216487e-05, "data_time": 0.0014740467071533204, "loss": 0.3782841920852661, "time": 0.636821699142456, "epoch": 6, "memory": 44138, "step": 19935} +{"lr": 1.9598865081795367e-05, "data_time": 0.0015021800994873048, "loss": 0.38352116048336027, "time": 0.645548152923584, "epoch": 6, "memory": 44138, "step": 20035} +{"lr": 1.9518534301258014e-05, "data_time": 0.0015161752700805664, "loss": 0.3651569336652756, "time": 0.6390479564666748, "epoch": 6, "memory": 44138, "step": 20135} +{"lr": 1.9438234650492273e-05, "data_time": 0.0014492034912109374, "loss": 0.39326231777667997, "time": 0.6369673252105713, "epoch": 6, "memory": 44138, "step": 20235} +{"lr": 1.935797132137332e-05, "data_time": 0.001443338394165039, "loss": 0.39745269119739535, "time": 0.6358798027038575, "epoch": 6, "memory": 44138, "step": 20335} +{"lr": 1.927774950342788e-05, "data_time": 0.0015887975692749023, "loss": 0.40316917896270754, "time": 0.6433388233184815, "epoch": 6, "memory": 44138, "step": 20435} +{"lr": 1.9197574383498696e-05, "data_time": 0.0013966560363769531, "loss": 0.3704320967197418, "time": 0.6372830152511597, "epoch": 6, "memory": 44138, "step": 20535} +{"lr": 1.911745114540919e-05, "data_time": 0.0015395402908325196, "loss": 0.3754275530576706, "time": 0.6360685348510742, "epoch": 6, "memory": 44138, "step": 20635} +{"lr": 1.9037384969628367e-05, "data_time": 0.0014511823654174804, "loss": 0.3556096822023392, "time": 0.6362322092056274, "epoch": 6, "memory": 44138, "step": 20735} +{"lr": 1.8957381032935694e-05, "data_time": 0.0016643047332763673, "loss": 0.3643163651227951, "time": 0.636770224571228, "epoch": 6, "memory": 44138, "step": 20835} +{"lr": 1.887744450808657e-05, "data_time": 0.001493072509765625, "loss": 0.3651486158370972, "time": 0.6381950378417969, "epoch": 6, "memory": 44138, "step": 20935} +{"lr": 1.8797580563477768e-05, "data_time": 0.0015011310577392578, "loss": 0.36550105214118955, "time": 0.6361826419830322, "epoch": 6, "memory": 44138, "step": 21035} +{"lr": 1.8717794362813298e-05, "data_time": 0.0016357183456420898, "loss": 0.36020020544528963, "time": 0.6352993965148925, "epoch": 6, "memory": 44138, "step": 21135} +{"lr": 1.863809106477054e-05, "data_time": 0.0015210866928100585, "loss": 0.3669223040342331, "time": 0.6356857061386109, "epoch": 6, "memory": 44138, "step": 21235} +{"lr": 1.8558475822666676e-05, "data_time": 0.0016295671463012694, "loss": 0.3765056669712067, "time": 0.6384711980819702, "epoch": 6, "memory": 44138, "step": 21335} +{"lr": 1.8478953784125537e-05, "data_time": 0.001545262336730957, "loss": 0.39416989386081697, "time": 0.6383140325546265, "epoch": 6, "memory": 44138, "step": 21435} +{"lr": 1.839953009074481e-05, "data_time": 0.0014120817184448242, "loss": 0.3795745223760605, "time": 0.6388851642608643, "epoch": 6, "memory": 44138, "step": 21535} +{"lr": 1.832020987776345e-05, "data_time": 0.0017464876174926758, "loss": 0.3982358336448669, "time": 0.6365885972976685, "epoch": 6, "memory": 44138, "step": 21635} +{"lr": 1.8240998273729876e-05, "data_time": 0.0015010595321655273, "loss": 0.36450940668582915, "time": 0.6385514259338378, "epoch": 6, "memory": 44138, "step": 21735} +{"lr": 1.8161900400170145e-05, "data_time": 0.0015857458114624024, "loss": 0.3877420246601105, "time": 0.6370485305786133, "epoch": 6, "memory": 44138, "step": 21835} +{"lr": 1.8082921371256987e-05, "data_time": 0.0013997554779052734, "loss": 0.39647932052612306, "time": 0.6361183643341064, "epoch": 6, "memory": 44138, "step": 21935} +{"lr": 1.800406629347906e-05, "data_time": 0.0014848709106445312, "loss": 0.38379827737808225, "time": 0.6363045930862427, "epoch": 6, "memory": 44138, "step": 22035} +{"lr": 1.7925340265310823e-05, "data_time": 0.0015642642974853516, "loss": 0.41020860970020295, "time": 0.6364546775817871, "epoch": 6, "memory": 44138, "step": 22135} +{"lr": 1.7846748376882866e-05, "data_time": 0.0015359878540039062, "loss": 0.3946759521961212, "time": 0.6380988359451294, "epoch": 6, "memory": 44138, "step": 22235} +{"lr": 1.7768295709652765e-05, "data_time": 0.001474761962890625, "loss": 0.37165563106536864, "time": 0.6350480556488037, "epoch": 6, "memory": 44138, "step": 22335} +{"lr": 1.768998733607668e-05, "data_time": 0.0013903379440307617, "loss": 0.3782049924135208, "time": 0.6372109889984131, "epoch": 6, "memory": 44138, "step": 22435} +{"lr": 1.7611828319281165e-05, "data_time": 0.0014556169509887696, "loss": 0.33835927248001096, "time": 0.6372325658798218, "epoch": 6, "memory": 44138, "step": 22535} +{"lr": 1.7533823712735976e-05, "data_time": 0.0015301227569580079, "loss": 0.3630715638399124, "time": 0.6378426551818848, "epoch": 6, "memory": 44138, "step": 22635} +{"lr": 1.745597855992727e-05, "data_time": 0.0017188072204589843, "loss": 0.3874973326921463, "time": 0.6369879961013794, "epoch": 6, "memory": 44138, "step": 22735} +{"lr": 1.7378297894031527e-05, "data_time": 0.0014562129974365235, "loss": 0.37654756009578705, "time": 0.6368170261383057, "epoch": 6, "memory": 44138, "step": 22835} +{"lr": 1.730078673759016e-05, "data_time": 0.0014070034027099609, "loss": 0.38792833089828493, "time": 0.6381886959075928, "epoch": 6, "memory": 44138, "step": 22935} +{"lr": 1.7223450102184637e-05, "data_time": 0.001417398452758789, "loss": 0.36938416957855225, "time": 0.6375573873519897, "epoch": 6, "memory": 44138, "step": 23035} +{"lr": 1.7146292988112627e-05, "data_time": 0.0014190912246704102, "loss": 0.37706841230392457, "time": 0.6368034839630127, "epoch": 6, "memory": 44138, "step": 23135} +{"lr": 1.7069320384064597e-05, "data_time": 0.0015091419219970704, "loss": 0.3630496084690094, "time": 0.6373643636703491, "epoch": 6, "memory": 44138, "step": 23235} +{"lr": 1.6992537266801296e-05, "data_time": 0.001432657241821289, "loss": 0.36087783575057986, "time": 0.6366403341293335, "epoch": 6, "memory": 44138, "step": 23335} +{"lr": 1.6915948600831928e-05, "data_time": 0.0014080762863159179, "loss": 0.39652993381023405, "time": 0.6343589067459107, "epoch": 6, "memory": 44138, "step": 23435} +{"accuracy/top1": 79.66320037841797, "data_time": 0.03217871435757341, "time": 1.276825843186214, "step": 6} +{"lr": 1.6834219685058524e-05, "data_time": 0.0015770673751831054, "loss": 0.37019487023353576, "time": 0.6351701974868774, "epoch": 7, "memory": 44138, "step": 23542} +{"lr": 1.675804925318156e-05, "data_time": 0.0014691352844238281, "loss": 0.36275808811187743, "time": 0.6352355480194092, "epoch": 7, "memory": 44138, "step": 23642} +{"lr": 1.6682088433716622e-05, "data_time": 0.00149383544921875, "loss": 0.3685837537050247, "time": 0.637007737159729, "epoch": 7, "memory": 44138, "step": 23742} +{"lr": 1.6606342138006298e-05, "data_time": 0.0014837026596069337, "loss": 0.35596791505813596, "time": 0.6355965852737426, "epoch": 7, "memory": 44138, "step": 23842} +{"lr": 1.653081526352276e-05, "data_time": 0.0015264511108398437, "loss": 0.38109772503376005, "time": 0.6363924741744995, "epoch": 7, "memory": 44138, "step": 23942} +{"lr": 1.6455512693551297e-05, "data_time": 0.0015047311782836914, "loss": 0.4049632966518402, "time": 0.6363247394561767, "epoch": 7, "memory": 44138, "step": 24042} +{"lr": 1.6380439296874487e-05, "data_time": 0.001533174514770508, "loss": 0.35496670603752134, "time": 0.6382855892181396, "epoch": 7, "memory": 44138, "step": 24142} +{"lr": 1.630559992745743e-05, "data_time": 0.0015706062316894532, "loss": 0.36166253983974456, "time": 0.6361469745635986, "epoch": 7, "memory": 44138, "step": 24242} +{"lr": 1.6230999424133908e-05, "data_time": 0.0014258623123168945, "loss": 0.3531336039304733, "time": 0.6377415657043457, "epoch": 7, "memory": 44138, "step": 24342} +{"lr": 1.615664261029346e-05, "data_time": 0.0014956235885620118, "loss": 0.3748207002878189, "time": 0.6369265079498291, "epoch": 7, "memory": 44138, "step": 24442} +{"lr": 1.6082534293569616e-05, "data_time": 0.0015178442001342774, "loss": 0.39182004034519197, "time": 0.635042953491211, "epoch": 7, "memory": 44138, "step": 24542} +{"lr": 1.6008679265529044e-05, "data_time": 0.0014105796813964843, "loss": 0.35284451246261594, "time": 0.6368744850158692, "epoch": 7, "memory": 44138, "step": 24642} +{"lr": 1.593508230136164e-05, "data_time": 0.0015794992446899413, "loss": 0.3764829635620117, "time": 0.639616060256958, "epoch": 7, "memory": 44138, "step": 24742} +{"lr": 1.5861748159571916e-05, "data_time": 0.0015469074249267578, "loss": 0.36659168303012846, "time": 0.6356468677520752, "epoch": 7, "memory": 44138, "step": 24842} +{"lr": 1.5788681581671252e-05, "data_time": 0.0015135765075683593, "loss": 0.3824197918176651, "time": 0.6371435165405274, "epoch": 7, "memory": 44138, "step": 24942} +{"lr": 1.571588729187132e-05, "data_time": 0.0015353918075561523, "loss": 0.38916949927806854, "time": 0.6372353553771972, "epoch": 7, "memory": 44138, "step": 25042} +{"lr": 1.564336999677871e-05, "data_time": 0.0015442848205566406, "loss": 0.358869206905365, "time": 0.6356818675994873, "epoch": 7, "memory": 44138, "step": 25142} +{"lr": 1.5571134385090526e-05, "data_time": 0.001357126235961914, "loss": 0.37347002029418946, "time": 0.6356656789779663, "epoch": 7, "memory": 44138, "step": 25242} +{"lr": 1.54991851272913e-05, "data_time": 0.0014548301696777344, "loss": 0.3890160501003265, "time": 0.6361973285675049, "epoch": 7, "memory": 44138, "step": 25342} +{"lr": 1.5427526875350992e-05, "data_time": 0.0014637947082519532, "loss": 0.3633673697710037, "time": 0.6364731788635254, "epoch": 7, "memory": 44138, "step": 25442} +{"lr": 1.5356164262424192e-05, "data_time": 0.0015940904617309571, "loss": 0.3767683207988739, "time": 0.63603835105896, "epoch": 7, "memory": 44138, "step": 25542} +{"lr": 1.5285101902550614e-05, "data_time": 0.0015026330947875977, "loss": 0.3859305500984192, "time": 0.6351003885269165, "epoch": 7, "memory": 44138, "step": 25642} +{"lr": 1.5214344390356706e-05, "data_time": 0.0014180660247802735, "loss": 0.3874822914600372, "time": 0.6402966260910035, "epoch": 7, "memory": 44138, "step": 25742} +{"lr": 1.5143896300758606e-05, "data_time": 0.0014272689819335937, "loss": 0.36415823400020597, "time": 0.6360689878463746, "epoch": 7, "memory": 44138, "step": 25842} +{"lr": 1.5073762188666339e-05, "data_time": 0.0019725799560546876, "loss": 0.3522688329219818, "time": 0.6363534927368164, "epoch": 7, "memory": 44138, "step": 25942} +{"lr": 1.5003946588689345e-05, "data_time": 0.0014714717864990235, "loss": 0.37421321868896484, "time": 0.6362645626068115, "epoch": 7, "memory": 44138, "step": 26042} +{"lr": 1.4934454014843216e-05, "data_time": 0.001434469223022461, "loss": 0.3535464584827423, "time": 0.6377779722213746, "epoch": 7, "memory": 44138, "step": 26142} +{"lr": 1.4865288960257924e-05, "data_time": 0.0014002799987792968, "loss": 0.3805251866579056, "time": 0.6351898193359375, "epoch": 7, "memory": 44138, "step": 26242} +{"lr": 1.479645589688726e-05, "data_time": 0.0015389680862426757, "loss": 0.3934237599372864, "time": 0.6355166912078858, "epoch": 7, "memory": 44138, "step": 26342} +{"lr": 1.4727959275219693e-05, "data_time": 0.0017250537872314452, "loss": 0.3818521350622177, "time": 0.6361732482910156, "epoch": 7, "memory": 44138, "step": 26442} +{"lr": 1.4659803523990658e-05, "data_time": 0.0016115903854370117, "loss": 0.33923845887184145, "time": 0.6361941576004029, "epoch": 7, "memory": 44138, "step": 26542} +{"lr": 1.4591993049896156e-05, "data_time": 0.0015195608139038086, "loss": 0.37609065473079684, "time": 0.6364243030548096, "epoch": 7, "memory": 44138, "step": 26642} +{"lr": 1.4524532237307841e-05, "data_time": 0.0017479419708251952, "loss": 0.36433753967285154, "time": 0.6365207195281982, "epoch": 7, "memory": 44138, "step": 26742} +{"lr": 1.44574254479896e-05, "data_time": 0.0014411687850952148, "loss": 0.38123677372932435, "time": 0.6364466905593872, "epoch": 7, "memory": 44138, "step": 26842} +{"lr": 1.4390677020815494e-05, "data_time": 0.0014626741409301757, "loss": 0.36849929094314576, "time": 0.6356098651885986, "epoch": 7, "memory": 44138, "step": 26942} +{"lr": 1.4324291271489216e-05, "data_time": 0.001484823226928711, "loss": 0.35879650712013245, "time": 0.639635157585144, "epoch": 7, "memory": 44138, "step": 27042} +{"lr": 1.4258272492265071e-05, "data_time": 0.0015256881713867187, "loss": 0.3917722821235657, "time": 0.637731409072876, "epoch": 7, "memory": 44138, "step": 27142} +{"lr": 1.4192624951670425e-05, "data_time": 0.0016059398651123047, "loss": 0.36367155611515045, "time": 0.6497369050979614, "epoch": 7, "memory": 44138, "step": 27242} +{"lr": 1.4127352894229787e-05, "data_time": 0.001289200782775879, "loss": 0.3828736573457718, "time": 0.6345186471939087, "epoch": 7, "memory": 44138, "step": 27342} +{"accuracy/top1": 79.10650634765625, "data_time": 0.033783953765342975, "time": 1.2759003721434494, "step": 7} +{"lr": 1.4057932403867309e-05, "data_time": 0.0015304088592529297, "loss": 0.4041621804237366, "time": 0.6368994235992431, "epoch": 8, "memory": 44138, "step": 27449} +{"lr": 1.3993450978351854e-05, "data_time": 0.0014450788497924806, "loss": 0.3982856273651123, "time": 0.6391650438308716, "epoch": 8, "memory": 44138, "step": 27549} +{"lr": 1.3929357913834719e-05, "data_time": 0.001488637924194336, "loss": 0.3864494860172272, "time": 0.6367493629455566, "epoch": 8, "memory": 44138, "step": 27649} +{"lr": 1.3865657354333762e-05, "data_time": 0.0014400243759155273, "loss": 0.37924416959285734, "time": 0.6353884220123291, "epoch": 8, "memory": 44138, "step": 27749} +{"lr": 1.3802353418488967e-05, "data_time": 0.0014401674270629883, "loss": 0.3679250717163086, "time": 0.6384751796722412, "epoch": 8, "memory": 44138, "step": 27849} +{"lr": 1.3739450199296088e-05, "data_time": 0.0014028072357177735, "loss": 0.3605092942714691, "time": 0.6369296550750733, "epoch": 8, "memory": 44138, "step": 27949} +{"lr": 1.3676951763842047e-05, "data_time": 0.001517939567565918, "loss": 0.37059405744075774, "time": 0.6392914533615113, "epoch": 8, "memory": 44138, "step": 28049} +{"lr": 1.361486215304198e-05, "data_time": 0.0014487504959106445, "loss": 0.38957723081111906, "time": 0.6377227544784546, "epoch": 8, "memory": 44138, "step": 28149} +{"lr": 1.3553185381377936e-05, "data_time": 0.0014718294143676758, "loss": 0.3781526476144791, "time": 0.6371407032012939, "epoch": 8, "memory": 44138, "step": 28249} +{"lr": 1.3491925436639332e-05, "data_time": 0.0015551328659057617, "loss": 0.3768164038658142, "time": 0.6375884056091309, "epoch": 8, "memory": 44138, "step": 28349} +{"lr": 1.3431086279665111e-05, "data_time": 0.001489877700805664, "loss": 0.37365337610244753, "time": 0.6350898981094361, "epoch": 8, "memory": 44138, "step": 28449} +{"lr": 1.3370671844087644e-05, "data_time": 0.0014866113662719727, "loss": 0.38336474299430845, "time": 0.636806058883667, "epoch": 8, "memory": 44138, "step": 28549} +{"lr": 1.331068603607844e-05, "data_time": 0.001475691795349121, "loss": 0.37587199211120603, "time": 0.6374342918395997, "epoch": 8, "memory": 44138, "step": 28649} +{"lr": 1.3251132734095514e-05, "data_time": 0.0014387845993041992, "loss": 0.3791746735572815, "time": 0.6370241165161132, "epoch": 8, "memory": 44138, "step": 28749} +{"lr": 1.319201578863267e-05, "data_time": 0.0014643907546997071, "loss": 0.3773917078971863, "time": 0.6371464252471923, "epoch": 8, "memory": 44138, "step": 28849} +{"lr": 1.313333902197055e-05, "data_time": 0.0014708518981933593, "loss": 0.36259378492832184, "time": 0.6372398138046265, "epoch": 8, "memory": 44138, "step": 28949} +{"lr": 1.3075106227929447e-05, "data_time": 0.0015167236328125, "loss": 0.37624795734882355, "time": 0.6357067108154297, "epoch": 8, "memory": 44138, "step": 29049} +{"lr": 1.3017321171624083e-05, "data_time": 0.001559138298034668, "loss": 0.3777626246213913, "time": 0.6369559049606324, "epoch": 8, "memory": 44138, "step": 29149} +{"lr": 1.2959987589220086e-05, "data_time": 0.0015105962753295898, "loss": 0.3494486898183823, "time": 0.6365528106689453, "epoch": 8, "memory": 44138, "step": 29249} +{"lr": 1.2903109187692523e-05, "data_time": 0.0015789985656738282, "loss": 0.3398669481277466, "time": 0.6380286931991577, "epoch": 8, "memory": 44138, "step": 29349} +{"lr": 1.2846689644586123e-05, "data_time": 0.0015120506286621094, "loss": 0.376858252286911, "time": 0.6388290643692016, "epoch": 8, "memory": 44138, "step": 29449} +{"lr": 1.2790732607777567e-05, "data_time": 0.0015869379043579102, "loss": 0.36199540793895724, "time": 0.6377752542495727, "epoch": 8, "memory": 44138, "step": 29549} +{"lr": 1.2735241695239598e-05, "data_time": 0.0014661788940429688, "loss": 0.3535286635160446, "time": 0.6368883609771728, "epoch": 8, "memory": 44138, "step": 29649} +{"lr": 1.2680220494807118e-05, "data_time": 0.001450181007385254, "loss": 0.36757898032665254, "time": 0.6365644454956054, "epoch": 8, "memory": 44138, "step": 29749} +{"lr": 1.2625672563945206e-05, "data_time": 0.0014929056167602539, "loss": 0.36055062115192416, "time": 0.6375349521636963, "epoch": 8, "memory": 44138, "step": 29849} +{"lr": 1.2571601429519072e-05, "data_time": 0.0014863252639770509, "loss": 0.37129674255847933, "time": 0.6419340610504151, "epoch": 8, "memory": 44138, "step": 29949} +{"lr": 1.251801058756609e-05, "data_time": 0.001531839370727539, "loss": 0.3759852975606918, "time": 0.6389995098114014, "epoch": 8, "memory": 44138, "step": 30049} +{"lr": 1.2464903503069694e-05, "data_time": 0.0015131711959838867, "loss": 0.363563260436058, "time": 0.6371601581573486, "epoch": 8, "memory": 44138, "step": 30149} +{"lr": 1.2412283609735373e-05, "data_time": 0.0016901731491088868, "loss": 0.3954562932252884, "time": 0.6374919176101684, "epoch": 8, "memory": 44138, "step": 30249} +{"lr": 1.2360154309768638e-05, "data_time": 0.0015392303466796875, "loss": 0.383209890127182, "time": 0.6361282348632813, "epoch": 8, "memory": 44138, "step": 30349} +{"lr": 1.2308518973655091e-05, "data_time": 0.0018133163452148438, "loss": 0.3841525256633759, "time": 0.6378105878829956, "epoch": 8, "memory": 44138, "step": 30449} +{"lr": 1.2257380939942495e-05, "data_time": 0.0015905380249023437, "loss": 0.35346856713294983, "time": 0.6368402242660522, "epoch": 8, "memory": 44138, "step": 30549} +{"lr": 1.2206743515024876e-05, "data_time": 0.001598358154296875, "loss": 0.3574922651052475, "time": 0.6396049499511719, "epoch": 8, "memory": 44138, "step": 30649} +{"lr": 1.2156609972928755e-05, "data_time": 0.0015631437301635743, "loss": 0.3666721910238266, "time": 0.6398011445999146, "epoch": 8, "memory": 44138, "step": 30749} +{"lr": 1.2106983555101508e-05, "data_time": 0.0014890432357788086, "loss": 0.3716928005218506, "time": 0.637281060218811, "epoch": 8, "memory": 44138, "step": 30849} +{"lr": 1.205786747020173e-05, "data_time": 0.0015869140625, "loss": 0.36391824185848237, "time": 0.6374053239822388, "epoch": 8, "memory": 44138, "step": 30949} +{"lr": 1.2009264893891803e-05, "data_time": 0.0017653226852416993, "loss": 0.35911635756492616, "time": 0.6375832319259643, "epoch": 8, "memory": 44138, "step": 31049} +{"lr": 1.1961178968632558e-05, "data_time": 0.001415395736694336, "loss": 0.3853606224060059, "time": 0.6373692512512207, "epoch": 8, "memory": 44138, "step": 31149} +{"lr": 1.19136128034801e-05, "data_time": 0.0012783527374267579, "loss": 0.3817710131406784, "time": 0.6367529392242431, "epoch": 8, "memory": 44138, "step": 31249} +{"accuracy/top1": 79.18305206298828, "data_time": 0.03135753911117028, "time": 1.2761004299953067, "step": 8} +{"lr": 1.1863296099806215e-05, "data_time": 0.0016751527786254884, "loss": 0.3622595489025116, "time": 0.6390992403030396, "epoch": 9, "memory": 44138, "step": 31356} +{"lr": 1.181681557185025e-05, "data_time": 0.001538705825805664, "loss": 0.35549541711807253, "time": 0.6372063398361206, "epoch": 9, "memory": 44138, "step": 31456} +{"lr": 1.1770864137998475e-05, "data_time": 0.0017137765884399415, "loss": 0.37085160315036775, "time": 0.6371675729751587, "epoch": 9, "memory": 44138, "step": 31556} +{"lr": 1.1725444769298765e-05, "data_time": 0.001439523696899414, "loss": 0.3729469805955887, "time": 0.6406498193740845, "epoch": 9, "memory": 44138, "step": 31656} +{"lr": 1.1680560402397639e-05, "data_time": 0.001489710807800293, "loss": 0.3686543405056, "time": 0.6391665458679199, "epoch": 9, "memory": 44138, "step": 31756} +{"lr": 1.1636213939350422e-05, "data_time": 0.0014519453048706054, "loss": 0.3786431461572647, "time": 0.6372768640518188, "epoch": 9, "memory": 44138, "step": 31856} +{"lr": 1.159240824743359e-05, "data_time": 0.0013993024826049806, "loss": 0.36677649319171907, "time": 0.6361638069152832, "epoch": 9, "memory": 44138, "step": 31956} +{"lr": 1.1549146158959366e-05, "data_time": 0.0014185428619384766, "loss": 0.3582907050848007, "time": 0.6401641368865967, "epoch": 9, "memory": 44138, "step": 32056} +{"lr": 1.1506430471092626e-05, "data_time": 0.0014175653457641601, "loss": 0.3795019119977951, "time": 0.636742377281189, "epoch": 9, "memory": 44138, "step": 32156} +{"lr": 1.146426394567e-05, "data_time": 0.001481008529663086, "loss": 0.3826130092144012, "time": 0.6376669645309448, "epoch": 9, "memory": 44138, "step": 32256} +{"lr": 1.142264930902133e-05, "data_time": 0.001440572738647461, "loss": 0.39255124926567075, "time": 0.6372113943099975, "epoch": 9, "memory": 44138, "step": 32356} +{"lr": 1.1381589251793428e-05, "data_time": 0.0014751911163330077, "loss": 0.3772126495838165, "time": 0.6379462003707885, "epoch": 9, "memory": 44138, "step": 32456} +{"lr": 1.1341086428776037e-05, "data_time": 0.0018348217010498047, "loss": 0.34995020031929014, "time": 0.6377904415130615, "epoch": 9, "memory": 44138, "step": 32556} +{"lr": 1.1301143458730242e-05, "data_time": 0.0014760971069335937, "loss": 0.3697828710079193, "time": 0.6371870756149292, "epoch": 9, "memory": 44138, "step": 32656} +{"lr": 1.1261762924219115e-05, "data_time": 0.0015008211135864257, "loss": 0.3838991582393646, "time": 0.6397773504257203, "epoch": 9, "memory": 44138, "step": 32756} +{"lr": 1.1222947371440745e-05, "data_time": 0.0014661073684692383, "loss": 0.38077346980571747, "time": 0.6368762493133545, "epoch": 9, "memory": 44138, "step": 32856} +{"lr": 1.118469931006364e-05, "data_time": 0.0014377355575561524, "loss": 0.36794666945934296, "time": 0.6370911836624146, "epoch": 9, "memory": 44138, "step": 32956} +{"lr": 1.1147021213064413e-05, "data_time": 0.0015420913696289062, "loss": 0.36707726418972014, "time": 0.6390187740325928, "epoch": 9, "memory": 44138, "step": 33056} +{"lr": 1.1109915516567913e-05, "data_time": 0.001514577865600586, "loss": 0.3606869637966156, "time": 0.6381826877593995, "epoch": 9, "memory": 44138, "step": 33156} +{"lr": 1.1073384619689717e-05, "data_time": 0.0014742612838745117, "loss": 0.35385514199733736, "time": 0.6374922275543213, "epoch": 9, "memory": 44138, "step": 33256} +{"lr": 1.1037430884381023e-05, "data_time": 0.0013845443725585937, "loss": 0.39900938272476194, "time": 0.6385854005813598, "epoch": 9, "memory": 44138, "step": 33356} +{"lr": 1.100205663527591e-05, "data_time": 0.0015089035034179688, "loss": 0.37683402597904203, "time": 0.6385986804962158, "epoch": 9, "memory": 44138, "step": 33456} +{"lr": 1.0967264159541028e-05, "data_time": 0.0014016151428222657, "loss": 0.37027874290943147, "time": 0.6383233785629272, "epoch": 9, "memory": 44138, "step": 33556} +{"lr": 1.0933055706727766e-05, "data_time": 0.0014101266860961914, "loss": 0.3779666215181351, "time": 0.6401222229003907, "epoch": 9, "memory": 44138, "step": 33656} +{"lr": 1.089943348862675e-05, "data_time": 0.0015568733215332031, "loss": 0.3759250670671463, "time": 0.6357459306716919, "epoch": 9, "memory": 44138, "step": 33756} +{"lr": 1.0866399679124873e-05, "data_time": 0.0014931678771972657, "loss": 0.37015424072742464, "time": 0.6367595672607422, "epoch": 9, "memory": 44138, "step": 33856} +{"lr": 1.083395641406474e-05, "data_time": 0.0014409065246582032, "loss": 0.34707754850387573, "time": 0.6379493236541748, "epoch": 9, "memory": 44138, "step": 33956} +{"lr": 1.0802105791106526e-05, "data_time": 0.0015029191970825195, "loss": 0.40765505731105806, "time": 0.6377553701400757, "epoch": 9, "memory": 44138, "step": 34056} +{"lr": 1.077084986959243e-05, "data_time": 0.0016844511032104493, "loss": 0.3520644873380661, "time": 0.6377681732177735, "epoch": 9, "memory": 44138, "step": 34156} +{"lr": 1.0740190670413442e-05, "data_time": 0.001387310028076172, "loss": 0.40866038799285886, "time": 0.6372867584228515, "epoch": 9, "memory": 44138, "step": 34256} +{"lr": 1.0710130175878743e-05, "data_time": 0.0014751672744750977, "loss": 0.35814627110958097, "time": 0.6534881591796875, "epoch": 9, "memory": 44138, "step": 34356} +{"lr": 1.0680670329587481e-05, "data_time": 0.0014291048049926759, "loss": 0.39215677678585054, "time": 0.6393326759338379, "epoch": 9, "memory": 44138, "step": 34456} +{"lr": 1.0651813036303164e-05, "data_time": 0.0016228675842285157, "loss": 0.37324203848838805, "time": 0.6415825605392456, "epoch": 9, "memory": 44138, "step": 34556} +{"lr": 1.062356016183046e-05, "data_time": 0.0015181541442871095, "loss": 0.3640313982963562, "time": 0.6368334054946899, "epoch": 9, "memory": 44138, "step": 34656} +{"lr": 1.0595913532894587e-05, "data_time": 0.0015727519989013673, "loss": 0.36915597021579744, "time": 0.6363588094711303, "epoch": 9, "memory": 44138, "step": 34756} +{"lr": 1.0568874937023192e-05, "data_time": 0.0014317989349365234, "loss": 0.3812170743942261, "time": 0.6373331069946289, "epoch": 9, "memory": 44138, "step": 34856} +{"lr": 1.0542446122430764e-05, "data_time": 0.0013901233673095704, "loss": 0.3598202347755432, "time": 0.6376536607742309, "epoch": 9, "memory": 44138, "step": 34956} +{"lr": 1.0516628797905623e-05, "data_time": 0.001470470428466797, "loss": 0.3528856784105301, "time": 0.6370012760162354, "epoch": 9, "memory": 44138, "step": 35056} +{"lr": 1.0491424632699434e-05, "data_time": 0.0013904094696044922, "loss": 0.3287424951791763, "time": 0.6387446880340576, "epoch": 9, "memory": 44138, "step": 35156} +{"accuracy/top1": 79.16217041015625, "data_time": 0.03243000753994646, "time": 1.2745897399968114, "step": 9} +{"lr": 1.0465137065246831e-05, "data_time": 0.0015178918838500977, "loss": 0.37640154659748076, "time": 0.6357495784759521, "epoch": 10, "memory": 44138, "step": 35263} +{"lr": 1.0441207272775563e-05, "data_time": 0.0015198230743408204, "loss": 0.38571797311306, "time": 0.6365582942962646, "epoch": 10, "memory": 44138, "step": 35363} +{"lr": 1.0417895516097024e-05, "data_time": 0.0015020370483398438, "loss": 0.3528754532337189, "time": 0.6387616872787476, "epoch": 10, "memory": 44138, "step": 35463} +{"lr": 1.0395203302462216e-05, "data_time": 0.0014327287673950196, "loss": 0.36181311011314393, "time": 0.6384918212890625, "epoch": 10, "memory": 44138, "step": 35563} +{"lr": 1.0373132099064833e-05, "data_time": 0.0014722585678100587, "loss": 0.39339700937271116, "time": 0.6373436689376831, "epoch": 10, "memory": 44138, "step": 35663} +{"lr": 1.0351683332946347e-05, "data_time": 0.0015189170837402344, "loss": 0.3513374745845795, "time": 0.6388654470443725, "epoch": 10, "memory": 44138, "step": 35763} +{"lr": 1.0330858390903745e-05, "data_time": 0.00150909423828125, "loss": 0.37314617931842803, "time": 0.63696768283844, "epoch": 10, "memory": 44138, "step": 35863} +{"lr": 1.0310658619399925e-05, "data_time": 0.0015008926391601562, "loss": 0.37610665559768675, "time": 0.6370523929595947, "epoch": 10, "memory": 44138, "step": 35963} +{"lr": 1.0291085324476576e-05, "data_time": 0.0015619516372680664, "loss": 0.36353162825107577, "time": 0.6374603271484375, "epoch": 10, "memory": 44138, "step": 36063} +{"lr": 1.0272139771669745e-05, "data_time": 0.0014326333999633788, "loss": 0.37460085153579714, "time": 0.6389540910720826, "epoch": 10, "memory": 44138, "step": 36163} +{"lr": 1.0253823185928048e-05, "data_time": 0.00157470703125, "loss": 0.3888337314128876, "time": 0.6390398740768433, "epoch": 10, "memory": 44138, "step": 36263} +{"lr": 1.0236136751533406e-05, "data_time": 0.0013630151748657226, "loss": 0.3701596975326538, "time": 0.638887357711792, "epoch": 10, "memory": 44138, "step": 36363} +{"lr": 1.0219081612024538e-05, "data_time": 0.0014132976531982422, "loss": 0.38940992653369905, "time": 0.6362844944000244, "epoch": 10, "memory": 44138, "step": 36463} +{"lr": 1.0202658870122986e-05, "data_time": 0.0015787363052368164, "loss": 0.37467766404151914, "time": 0.6374501943588257, "epoch": 10, "memory": 44138, "step": 36563} +{"lr": 1.0186869587661827e-05, "data_time": 0.0014269113540649413, "loss": 0.3911354035139084, "time": 0.6371725797653198, "epoch": 10, "memory": 44138, "step": 36663} +{"lr": 1.0171714785517018e-05, "data_time": 0.036666488647460936, "loss": 0.34564143419265747, "time": 0.6711091518402099, "epoch": 10, "memory": 44138, "step": 36763} +{"lr": 1.0157195443541394e-05, "data_time": 0.0013899803161621094, "loss": 0.3497059136629105, "time": 0.6376233100891113, "epoch": 10, "memory": 44138, "step": 36863} +{"lr": 1.0143312500501317e-05, "data_time": 0.0014259099960327148, "loss": 0.36678563356399535, "time": 0.6382873773574829, "epoch": 10, "memory": 44138, "step": 36963} +{"lr": 1.0130066854015972e-05, "data_time": 0.0014418840408325195, "loss": 0.3871037155389786, "time": 0.6492510318756104, "epoch": 10, "memory": 44138, "step": 37063} +{"lr": 1.0117459360499327e-05, "data_time": 0.0013692140579223632, "loss": 0.35543271601200105, "time": 0.6374026536941528, "epoch": 10, "memory": 44138, "step": 37163} +{"lr": 1.0105490835104763e-05, "data_time": 0.0013768911361694337, "loss": 0.3578047901391983, "time": 0.6393279075622559, "epoch": 10, "memory": 44138, "step": 37263} +{"lr": 1.0094162051672373e-05, "data_time": 0.0015633821487426758, "loss": 0.38107499182224275, "time": 0.6389130115509033, "epoch": 10, "memory": 44138, "step": 37363} +{"lr": 1.0083473742678929e-05, "data_time": 0.0015424489974975586, "loss": 0.37405459880828856, "time": 0.6388011455535889, "epoch": 10, "memory": 44138, "step": 37463} +{"lr": 1.0073426599190536e-05, "data_time": 0.001570272445678711, "loss": 0.36700065732002257, "time": 0.638599944114685, "epoch": 10, "memory": 44138, "step": 37563} +{"lr": 1.0064021270817907e-05, "data_time": 0.0014969825744628907, "loss": 0.390339133143425, "time": 0.6420160055160522, "epoch": 10, "memory": 44138, "step": 37663} +{"lr": 1.0055258365674413e-05, "data_time": 0.0016922950744628906, "loss": 0.37869915962219236, "time": 0.6423586130142211, "epoch": 10, "memory": 44138, "step": 37763} +{"lr": 1.0047138450336735e-05, "data_time": 0.001508331298828125, "loss": 0.38447908461093905, "time": 0.6390578269958496, "epoch": 10, "memory": 44138, "step": 37863} +{"lr": 1.0039662049808213e-05, "data_time": 0.001542806625366211, "loss": 0.35889176428318026, "time": 0.6374271154403687, "epoch": 10, "memory": 44138, "step": 37963} +{"lr": 1.0032829647484969e-05, "data_time": 0.0015082359313964844, "loss": 0.39252043068408965, "time": 0.6393031120300293, "epoch": 10, "memory": 44138, "step": 38063} +{"lr": 1.0026641685124577e-05, "data_time": 0.001631021499633789, "loss": 0.3934075653553009, "time": 0.6386359453201294, "epoch": 10, "memory": 44138, "step": 38163} +{"lr": 1.0021098562817543e-05, "data_time": 0.001405644416809082, "loss": 0.3611509084701538, "time": 0.6383473157882691, "epoch": 10, "memory": 44138, "step": 38263} +{"lr": 1.0016200638961424e-05, "data_time": 0.0015422821044921875, "loss": 0.33123467266559603, "time": 0.6385917663574219, "epoch": 10, "memory": 44138, "step": 38363} +{"lr": 1.001194823023767e-05, "data_time": 0.001431584358215332, "loss": 0.3676534056663513, "time": 0.6369673252105713, "epoch": 10, "memory": 44138, "step": 38463} +{"lr": 1.0008341611591118e-05, "data_time": 0.0014977455139160156, "loss": 0.3790225267410278, "time": 0.6383319616317749, "epoch": 10, "memory": 44138, "step": 38563} +{"lr": 1.0005381016212259e-05, "data_time": 0.0016683340072631836, "loss": 0.3796721279621124, "time": 0.640384316444397, "epoch": 10, "memory": 44138, "step": 38663} +{"lr": 1.0003066635522105e-05, "data_time": 0.001898980140686035, "loss": 0.35755579471588134, "time": 0.6484117984771729, "epoch": 10, "memory": 44138, "step": 38763} +{"lr": 1.0001398619159851e-05, "data_time": 0.001570272445678711, "loss": 0.38691721558570863, "time": 0.6406832695007324, "epoch": 10, "memory": 44138, "step": 38863} +{"lr": 1.0000377074973214e-05, "data_time": 0.0014528751373291016, "loss": 0.35122773945331576, "time": 0.6390617609024047, "epoch": 10, "memory": 44138, "step": 38963} +{"lr": 1.0000002069011414e-05, "data_time": 0.0013076305389404298, "loss": 0.35887025892734525, "time": 0.6382135152816772, "epoch": 10, "memory": 44138, "step": 39063} +{"accuracy/top1": 78.66462707519531, "data_time": 0.031674109656235266, "time": 1.2740361443881332, "step": 10} diff --git a/clip_large_pretrain_4x256_sdv2_lr3e-5/20230606_052112/vis_data/vis_image/0000000.png_1.png b/clip_large_pretrain_4x256_sdv2_lr3e-5/20230606_052112/vis_data/vis_image/0000000.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..cef053057bfc317a9dab0d276f235fedab6562b3 Binary files /dev/null and b/clip_large_pretrain_4x256_sdv2_lr3e-5/20230606_052112/vis_data/vis_image/0000000.png_1.png differ diff --git a/clip_large_pretrain_4x256_sdv2_lr3e-5/20230606_052112/vis_data/vis_image/0000000.png_10.png b/clip_large_pretrain_4x256_sdv2_lr3e-5/20230606_052112/vis_data/vis_image/0000000.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..ffb72e035a8aaca316f89e95c970e953477e318f Binary files /dev/null and b/clip_large_pretrain_4x256_sdv2_lr3e-5/20230606_052112/vis_data/vis_image/0000000.png_10.png differ diff --git a/clip_large_pretrain_4x256_sdv2_lr3e-5/20230606_052112/vis_data/vis_image/0000000.png_2.png b/clip_large_pretrain_4x256_sdv2_lr3e-5/20230606_052112/vis_data/vis_image/0000000.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..92fff22bad4c3e3c0d5e4748445debbe067241fa Binary files /dev/null and b/clip_large_pretrain_4x256_sdv2_lr3e-5/20230606_052112/vis_data/vis_image/0000000.png_2.png differ diff --git a/clip_large_pretrain_4x256_sdv2_lr3e-5/20230606_052112/vis_data/vis_image/0000000.png_3.png b/clip_large_pretrain_4x256_sdv2_lr3e-5/20230606_052112/vis_data/vis_image/0000000.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..887a52ab5a49f062e0e5e0c4860d37f5a3192c8b Binary files /dev/null and b/clip_large_pretrain_4x256_sdv2_lr3e-5/20230606_052112/vis_data/vis_image/0000000.png_3.png differ diff --git a/clip_large_pretrain_4x256_sdv2_lr3e-5/20230606_052112/vis_data/vis_image/0000000.png_4.png b/clip_large_pretrain_4x256_sdv2_lr3e-5/20230606_052112/vis_data/vis_image/0000000.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..6b8e4d064df53f0b1049c274422c3da7dedab129 Binary files /dev/null and b/clip_large_pretrain_4x256_sdv2_lr3e-5/20230606_052112/vis_data/vis_image/0000000.png_4.png differ diff --git a/clip_large_pretrain_4x256_sdv2_lr3e-5/20230606_052112/vis_data/vis_image/0000000.png_5.png b/clip_large_pretrain_4x256_sdv2_lr3e-5/20230606_052112/vis_data/vis_image/0000000.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..1a56de064b0e60a848f5a2cbb924a1d2ad136f75 Binary files /dev/null and b/clip_large_pretrain_4x256_sdv2_lr3e-5/20230606_052112/vis_data/vis_image/0000000.png_5.png differ diff --git a/clip_large_pretrain_4x256_sdv2_lr3e-5/20230606_052112/vis_data/vis_image/0000000.png_6.png b/clip_large_pretrain_4x256_sdv2_lr3e-5/20230606_052112/vis_data/vis_image/0000000.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..1fa7ad72ba281a600ca961cb656cad17c6725bf1 Binary files /dev/null and b/clip_large_pretrain_4x256_sdv2_lr3e-5/20230606_052112/vis_data/vis_image/0000000.png_6.png differ diff --git a/clip_large_pretrain_4x256_sdv2_lr3e-5/20230606_052112/vis_data/vis_image/0000000.png_7.png b/clip_large_pretrain_4x256_sdv2_lr3e-5/20230606_052112/vis_data/vis_image/0000000.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..9512d4d939997338ac8747eaee8815c80852f3b1 Binary files /dev/null and b/clip_large_pretrain_4x256_sdv2_lr3e-5/20230606_052112/vis_data/vis_image/0000000.png_7.png differ diff --git a/clip_large_pretrain_4x256_sdv2_lr3e-5/20230606_052112/vis_data/vis_image/0000000.png_8.png b/clip_large_pretrain_4x256_sdv2_lr3e-5/20230606_052112/vis_data/vis_image/0000000.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..0741c58ac6907580be78affc954d26ce46f60050 Binary files /dev/null and b/clip_large_pretrain_4x256_sdv2_lr3e-5/20230606_052112/vis_data/vis_image/0000000.png_8.png differ diff --git a/clip_large_pretrain_4x256_sdv2_lr3e-5/20230606_052112/vis_data/vis_image/0000000.png_9.png b/clip_large_pretrain_4x256_sdv2_lr3e-5/20230606_052112/vis_data/vis_image/0000000.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..f5b0dac40ea9895a50d77da9a32a996cce23c490 Binary files /dev/null and b/clip_large_pretrain_4x256_sdv2_lr3e-5/20230606_052112/vis_data/vis_image/0000000.png_9.png differ diff --git a/clip_large_pretrain_4x256_sdv2_lr3e-5/20230606_052112/vis_data/vis_image/000005174.jpg_1.png b/clip_large_pretrain_4x256_sdv2_lr3e-5/20230606_052112/vis_data/vis_image/000005174.jpg_1.png new file mode 100644 index 0000000000000000000000000000000000000000..c998c570d226c621df84ab76d2935e33d807756c Binary files /dev/null and b/clip_large_pretrain_4x256_sdv2_lr3e-5/20230606_052112/vis_data/vis_image/000005174.jpg_1.png differ diff --git a/clip_large_pretrain_4x256_sdv2_lr3e-5/20230606_052112/vis_data/vis_image/000005174.jpg_10.png b/clip_large_pretrain_4x256_sdv2_lr3e-5/20230606_052112/vis_data/vis_image/000005174.jpg_10.png new file mode 100644 index 0000000000000000000000000000000000000000..58db6ac9449849afbb10e6a9eca7e07850a5be31 Binary files /dev/null and b/clip_large_pretrain_4x256_sdv2_lr3e-5/20230606_052112/vis_data/vis_image/000005174.jpg_10.png differ diff --git a/clip_large_pretrain_4x256_sdv2_lr3e-5/20230606_052112/vis_data/vis_image/000005174.jpg_2.png b/clip_large_pretrain_4x256_sdv2_lr3e-5/20230606_052112/vis_data/vis_image/000005174.jpg_2.png new file mode 100644 index 0000000000000000000000000000000000000000..75cadc4b582cde916379160752e6db7f5d14f881 Binary files /dev/null and b/clip_large_pretrain_4x256_sdv2_lr3e-5/20230606_052112/vis_data/vis_image/000005174.jpg_2.png differ diff --git a/clip_large_pretrain_4x256_sdv2_lr3e-5/20230606_052112/vis_data/vis_image/000005174.jpg_3.png b/clip_large_pretrain_4x256_sdv2_lr3e-5/20230606_052112/vis_data/vis_image/000005174.jpg_3.png new file mode 100644 index 0000000000000000000000000000000000000000..56d4c017405a72d2cfb5ecd6bc92a8a4c130db91 Binary files /dev/null and b/clip_large_pretrain_4x256_sdv2_lr3e-5/20230606_052112/vis_data/vis_image/000005174.jpg_3.png differ diff --git a/clip_large_pretrain_4x256_sdv2_lr3e-5/20230606_052112/vis_data/vis_image/000005174.jpg_4.png b/clip_large_pretrain_4x256_sdv2_lr3e-5/20230606_052112/vis_data/vis_image/000005174.jpg_4.png new file mode 100644 index 0000000000000000000000000000000000000000..e5459bec6811e41526c326cc901a498bf33b99f7 Binary files /dev/null and b/clip_large_pretrain_4x256_sdv2_lr3e-5/20230606_052112/vis_data/vis_image/000005174.jpg_4.png differ diff --git a/clip_large_pretrain_4x256_sdv2_lr3e-5/20230606_052112/vis_data/vis_image/000005174.jpg_5.png b/clip_large_pretrain_4x256_sdv2_lr3e-5/20230606_052112/vis_data/vis_image/000005174.jpg_5.png new file mode 100644 index 0000000000000000000000000000000000000000..8c357746928031b31f001f35d3257c12ccd1376c Binary files /dev/null and b/clip_large_pretrain_4x256_sdv2_lr3e-5/20230606_052112/vis_data/vis_image/000005174.jpg_5.png differ diff --git a/clip_large_pretrain_4x256_sdv2_lr3e-5/20230606_052112/vis_data/vis_image/000005174.jpg_6.png b/clip_large_pretrain_4x256_sdv2_lr3e-5/20230606_052112/vis_data/vis_image/000005174.jpg_6.png new file mode 100644 index 0000000000000000000000000000000000000000..010524ad92d0bf7f0b140e72ae456bc86cc23a9c Binary files /dev/null and b/clip_large_pretrain_4x256_sdv2_lr3e-5/20230606_052112/vis_data/vis_image/000005174.jpg_6.png differ diff --git a/clip_large_pretrain_4x256_sdv2_lr3e-5/20230606_052112/vis_data/vis_image/000005174.jpg_7.png b/clip_large_pretrain_4x256_sdv2_lr3e-5/20230606_052112/vis_data/vis_image/000005174.jpg_7.png new file mode 100644 index 0000000000000000000000000000000000000000..d4d35623ecd9f5e533e90616a9ce78ec96cb3b29 Binary files /dev/null and b/clip_large_pretrain_4x256_sdv2_lr3e-5/20230606_052112/vis_data/vis_image/000005174.jpg_7.png differ diff --git a/clip_large_pretrain_4x256_sdv2_lr3e-5/20230606_052112/vis_data/vis_image/000005174.jpg_8.png b/clip_large_pretrain_4x256_sdv2_lr3e-5/20230606_052112/vis_data/vis_image/000005174.jpg_8.png new file mode 100644 index 0000000000000000000000000000000000000000..fc9aca01ed0a1bd384b678d727638fc22bce6179 Binary files /dev/null and b/clip_large_pretrain_4x256_sdv2_lr3e-5/20230606_052112/vis_data/vis_image/000005174.jpg_8.png differ diff --git a/clip_large_pretrain_4x256_sdv2_lr3e-5/20230606_052112/vis_data/vis_image/000005174.jpg_9.png b/clip_large_pretrain_4x256_sdv2_lr3e-5/20230606_052112/vis_data/vis_image/000005174.jpg_9.png new file mode 100644 index 0000000000000000000000000000000000000000..b3caebbf09c33cb05658e1ce1c7834a9c818ff39 Binary files /dev/null and b/clip_large_pretrain_4x256_sdv2_lr3e-5/20230606_052112/vis_data/vis_image/000005174.jpg_9.png differ diff --git a/clip_large_pretrain_4x256_sdv2_lr3e-5/clip_large_pretrain_4x256_sdv2_lr3e-5.py b/clip_large_pretrain_4x256_sdv2_lr3e-5/clip_large_pretrain_4x256_sdv2_lr3e-5.py new file mode 100644 index 0000000000000000000000000000000000000000..8f40f4e1084c3f11ae10e85e06719878b6741bd8 --- /dev/null +++ b/clip_large_pretrain_4x256_sdv2_lr3e-5/clip_large_pretrain_4x256_sdv2_lr3e-5.py @@ -0,0 +1,236 @@ +optim_wrapper = dict( + optimizer=dict( + type='AdamW', lr=3e-05, weight_decay=0.3, _scope_='mmpretrain'), + paramwise_cfg=dict( + custom_keys=dict({ + '.cls_token': dict(decay_mult=0.0), + '.pos_embed': dict(decay_mult=0.0) + })), + type='AmpOptimWrapper', + dtype='bfloat16', + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=False, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=4096) +model = dict( + type='ImageClassifier', + backbone=dict( + frozen_stages=24, + type='VisionTransformer', + arch='l', + img_size=224, + patch_size=14, + drop_rate=0.1, + pre_norm=True, + final_norm=False, + init_cfg=dict( + type='Pretrained', + checkpoint='ckpt/openclip-ViT-L-14.pth', + prefix='backbone')), + neck=dict( + type='CLIPProjection', + in_channels=1024, + out_channels=768, + init_cfg=dict( + type='Pretrained', + checkpoint='ckpt/openclip-ViT-L-14.pth', + prefix='backbone')), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=768, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + init_cfg=None), + init_cfg=dict( + type='TruncNormal', layer=['Conv2d', 'Linear'], std=0.02, bias=0.0), + train_cfg=None) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=128, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stablediffusionV2-1-dpmsolver-25-1m.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/cc1m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=128, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV2-1-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=128, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV2-1-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/clip_large_pretrain_4x256_sdv2_lr3e-5' diff --git a/clip_large_pretrain_4x256_sdv2_lr3e-5/epoch_1.pth b/clip_large_pretrain_4x256_sdv2_lr3e-5/epoch_1.pth new file mode 100644 index 0000000000000000000000000000000000000000..8a334bc57663c6937f54af53f48e4fd092ea07a9 --- /dev/null +++ b/clip_large_pretrain_4x256_sdv2_lr3e-5/epoch_1.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:3ec984d0ba3ce173b89ee0c64205777a4d2429de2e1e1b49ce0f460e98ed06e4 +size 2438617665 diff --git a/clip_large_pretrain_4x256_sdv2_lr3e-5/epoch_10.pth b/clip_large_pretrain_4x256_sdv2_lr3e-5/epoch_10.pth new file mode 100644 index 0000000000000000000000000000000000000000..877463c58767290ef3b06ca02aa2714a1d163136 --- /dev/null +++ b/clip_large_pretrain_4x256_sdv2_lr3e-5/epoch_10.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:92fb435b7b7d55e7597299d5863e8c0d639c6bf983c366bfd1b0af13d0d07464 +size 2441207169 diff --git a/clip_large_pretrain_4x256_sdv2_lr3e-5/epoch_2.pth b/clip_large_pretrain_4x256_sdv2_lr3e-5/epoch_2.pth new file mode 100644 index 0000000000000000000000000000000000000000..b36a866acfabcadac9fdcb01218b594e50ab931e --- /dev/null +++ b/clip_large_pretrain_4x256_sdv2_lr3e-5/epoch_2.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:20e93a4715cfad18790a30fcb0b009b2b6441940f63f170e7eec841746f5ec42 +size 2438906689 diff --git a/clip_large_pretrain_4x256_sdv2_lr3e-5/epoch_3.pth b/clip_large_pretrain_4x256_sdv2_lr3e-5/epoch_3.pth new file mode 100644 index 0000000000000000000000000000000000000000..1f2a39c8abde3f05106388fb538352a4bd39a554 --- /dev/null +++ b/clip_large_pretrain_4x256_sdv2_lr3e-5/epoch_3.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:018c03f9fcd9b97fc817d44445d6bd6abba00d58574327e94f82b2c1f66efe7b +size 2439193665 diff --git a/clip_large_pretrain_4x256_sdv2_lr3e-5/epoch_4.pth b/clip_large_pretrain_4x256_sdv2_lr3e-5/epoch_4.pth new file mode 100644 index 0000000000000000000000000000000000000000..ca90a0a55e44741da574437bca4fbd92c1937ec5 --- /dev/null +++ b/clip_large_pretrain_4x256_sdv2_lr3e-5/epoch_4.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c2b0d045deb58920931263e9fa6c70f540f4fc00b01ccb077e210979c512970e +size 2439481281 diff --git a/clip_large_pretrain_4x256_sdv2_lr3e-5/epoch_5.pth b/clip_large_pretrain_4x256_sdv2_lr3e-5/epoch_5.pth new file mode 100644 index 0000000000000000000000000000000000000000..2adc6dfde5603b080fe865e41754a22c2774187b --- /dev/null +++ b/clip_large_pretrain_4x256_sdv2_lr3e-5/epoch_5.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:806e760819ac1247f73fed5865b060802c9635fc8694a889e9226113095de6c8 +size 2439768641 diff --git a/clip_large_pretrain_4x256_sdv2_lr3e-5/epoch_6.pth b/clip_large_pretrain_4x256_sdv2_lr3e-5/epoch_6.pth new file mode 100644 index 0000000000000000000000000000000000000000..354789f5a7691fe37c89861fd102c8ed658a2001 --- /dev/null +++ b/clip_large_pretrain_4x256_sdv2_lr3e-5/epoch_6.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c4221340f18acacb1c2c12ec36552756c40ef65d49f34de045fb94c68c5a7ac1 +size 2440056449 diff --git a/clip_large_pretrain_4x256_sdv2_lr3e-5/epoch_7.pth b/clip_large_pretrain_4x256_sdv2_lr3e-5/epoch_7.pth new file mode 100644 index 0000000000000000000000000000000000000000..f847ddc99665e962d8c71b0187cb04f9957a9fbb --- /dev/null +++ b/clip_large_pretrain_4x256_sdv2_lr3e-5/epoch_7.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ac52045d24d3ad6621ebac3ee796901f99610fdeb7596344e1789188cae75016 +size 2440343937 diff --git a/clip_large_pretrain_4x256_sdv2_lr3e-5/epoch_8.pth b/clip_large_pretrain_4x256_sdv2_lr3e-5/epoch_8.pth new file mode 100644 index 0000000000000000000000000000000000000000..d11ba0c36c88581830c874b342391598134f1d6b --- /dev/null +++ b/clip_large_pretrain_4x256_sdv2_lr3e-5/epoch_8.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ec0928a182110ed1af8522b57f750162ab99a56919d369f1681e14ea0ebd3eb6 +size 2440631553 diff --git a/clip_large_pretrain_4x256_sdv2_lr3e-5/epoch_9.pth b/clip_large_pretrain_4x256_sdv2_lr3e-5/epoch_9.pth new file mode 100644 index 0000000000000000000000000000000000000000..f7795a7c52f4b780be01cdc660c097e3194a5339 --- /dev/null +++ b/clip_large_pretrain_4x256_sdv2_lr3e-5/epoch_9.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:91bfafbb44a0652daddb3996dfeb3185a8bc3d1fc8ed8347a7385d2aea637b14 +size 2440919297 diff --git a/clip_large_pretrain_4x256_sdv2_lr3e-5/last_checkpoint b/clip_large_pretrain_4x256_sdv2_lr3e-5/last_checkpoint new file mode 100644 index 0000000000000000000000000000000000000000..b2dd7abbb3d0b44c1d6cd07121e734d56bb25acf --- /dev/null +++ b/clip_large_pretrain_4x256_sdv2_lr3e-5/last_checkpoint @@ -0,0 +1 @@ +/mnt/petrelfs/luzeyu/workspace/fakebench/mmpretrain/workdir/clip_large_pretrain_4x256_sdv2_lr3e-5/epoch_10.pth \ No newline at end of file diff --git a/clip_large_pretrain_4x256_stylegan3_lr1e-4/20230605_145625/20230605_145625.log b/clip_large_pretrain_4x256_stylegan3_lr1e-4/20230605_145625/20230605_145625.log new file mode 100644 index 0000000000000000000000000000000000000000..ebc8010b6490fe790a66a6f580b1537621b26e1c --- /dev/null +++ b/clip_large_pretrain_4x256_stylegan3_lr1e-4/20230605_145625/20230605_145625.log @@ -0,0 +1,1879 @@ +2023/06/05 14:56:29 - mmengine - INFO - +------------------------------------------------------------ +System environment: + sys.platform: linux + Python: 3.10.9 (main, Mar 8 2023, 10:47:38) [GCC 11.2.0] + CUDA available: True + numpy_random_seed: 242847178 + GPU 0,1,2,3: NVIDIA A100-SXM4-80GB + CUDA_HOME: /mnt/petrelfs/share/cuda-11.6 + NVCC: Cuda compilation tools, release 11.6, V11.6.124 + GCC: gcc (GCC) 7.5.0 + PyTorch: 1.13.1 + PyTorch compiling details: PyTorch built with: + - GCC 9.3 + - C++ Version: 201402 + - Intel(R) oneAPI Math Kernel Library Version 2021.4-Product Build 20210904 for Intel(R) 64 architecture applications + - Intel(R) MKL-DNN v2.6.0 (Git Hash 52b5f107dd9cf10910aaa19cb47f3abf9b349815) + - OpenMP 201511 (a.k.a. OpenMP 4.5) + - LAPACK is enabled (usually provided by MKL) + - NNPACK is enabled + - CPU capability usage: AVX2 + - CUDA Runtime 11.6 + - NVCC architecture flags: -gencode;arch=compute_37,code=sm_37;-gencode;arch=compute_50,code=sm_50;-gencode;arch=compute_60,code=sm_60;-gencode;arch=compute_61,code=sm_61;-gencode;arch=compute_70,code=sm_70;-gencode;arch=compute_75,code=sm_75;-gencode;arch=compute_80,code=sm_80;-gencode;arch=compute_86,code=sm_86;-gencode;arch=compute_37,code=compute_37 + - CuDNN 8.3.2 (built against CUDA 11.5) + - Magma 2.6.1 + - Build settings: BLAS_INFO=mkl, BUILD_TYPE=Release, CUDA_VERSION=11.6, CUDNN_VERSION=8.3.2, CXX_COMPILER=/opt/rh/devtoolset-9/root/usr/bin/c++, CXX_FLAGS= -fabi-version=11 -Wno-deprecated -fvisibility-inlines-hidden -DUSE_PTHREADPOOL -fopenmp -DNDEBUG -DUSE_KINETO -DUSE_FBGEMM -DUSE_QNNPACK -DUSE_PYTORCH_QNNPACK -DUSE_XNNPACK -DSYMBOLICATE_MOBILE_DEBUG_HANDLE -DEDGE_PROFILER_USE_KINETO -O2 -fPIC -Wno-narrowing -Wall -Wextra -Werror=return-type -Werror=non-virtual-dtor -Wno-missing-field-initializers -Wno-type-limits -Wno-array-bounds -Wno-unknown-pragmas -Wunused-local-typedefs -Wno-unused-parameter -Wno-unused-function -Wno-unused-result -Wno-strict-overflow -Wno-strict-aliasing -Wno-error=deprecated-declarations -Wno-stringop-overflow -Wno-psabi -Wno-error=pedantic -Wno-error=redundant-decls -Wno-error=old-style-cast -fdiagnostics-color=always -faligned-new -Wno-unused-but-set-variable -Wno-maybe-uninitialized -fno-math-errno -fno-trapping-math -Werror=format -Werror=cast-function-type -Wno-stringop-overflow, LAPACK_INFO=mkl, PERF_WITH_AVX=1, PERF_WITH_AVX2=1, PERF_WITH_AVX512=1, TORCH_VERSION=1.13.1, USE_CUDA=ON, USE_CUDNN=ON, USE_EXCEPTION_PTR=1, USE_GFLAGS=OFF, USE_GLOG=OFF, USE_MKL=ON, USE_MKLDNN=ON, USE_MPI=OFF, USE_NCCL=ON, USE_NNPACK=ON, USE_OPENMP=ON, USE_ROCM=OFF, + + TorchVision: 0.14.1 + OpenCV: 4.7.0 + MMEngine: 0.7.3 + +Runtime environment: + cudnn_benchmark: True + mp_cfg: {'mp_start_method': 'fork', 'opencv_num_threads': 0} + dist_cfg: {'backend': 'nccl'} + seed: None + deterministic: False + Distributed launcher: slurm + Distributed training: True + GPU number: 4 +------------------------------------------------------------ + +2023/06/05 14:56:33 - mmengine - INFO - Config: +optim_wrapper = dict( + optimizer=dict( + type='AdamW', lr=0.0001, weight_decay=0.3, _scope_='mmpretrain'), + paramwise_cfg=dict( + custom_keys=dict({ + '.cls_token': dict(decay_mult=0.0), + '.pos_embed': dict(decay_mult=0.0) + })), + type='AmpOptimWrapper', + dtype='bfloat16', + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=False, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=4096) +model = dict( + type='ImageClassifier', + backbone=dict( + frozen_stages=24, + type='VisionTransformer', + arch='l', + img_size=224, + patch_size=14, + drop_rate=0.1, + pre_norm=True, + final_norm=False, + init_cfg=dict( + type='Pretrained', + checkpoint='ckpt/openclip-ViT-L-14.pth', + prefix='backbone')), + neck=dict( + type='CLIPProjection', + in_channels=1024, + out_channels=768, + init_cfg=dict( + type='Pretrained', + checkpoint='ckpt/openclip-ViT-L-14.pth', + prefix='backbone')), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=768, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + init_cfg=None), + init_cfg=dict( + type='TruncNormal', layer=['Conv2d', 'Linear'], std=0.02, bias=0.0), + train_cfg=None) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=128, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/all1.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/all2.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/all3.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3fake8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/cc1m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3real8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=128, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV2-1-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/if-dpmsolver++-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=128, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV2-1-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/if-dpmsolver++-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/clip_large_pretrain_4x256_stylegan3_lr1e-4' + +2023/06/05 14:56:47 - mmengine - INFO - Hooks will be executed in the following order: +before_run: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_load_checkpoint: +(ABOVE_NORMAL) EMAHook + -------------------- +before_train: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_train_epoch: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(NORMAL ) DistSamplerSeedHook + -------------------- +before_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook + -------------------- +after_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +after_train_epoch: +(NORMAL ) IterTimerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_val_epoch: +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook + -------------------- +before_val_iter: +(NORMAL ) IterTimerHook + -------------------- +after_val_iter: +(NORMAL ) IterTimerHook +(NORMAL ) VisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_val_epoch: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_save_checkpoint: +(ABOVE_NORMAL) EMAHook + -------------------- +after_train: +(VERY_LOW ) CheckpointHook + -------------------- +before_test_epoch: +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook + -------------------- +before_test_iter: +(NORMAL ) IterTimerHook + -------------------- +after_test_iter: +(NORMAL ) IterTimerHook +(NORMAL ) VisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_test_epoch: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_run: +(BELOW_NORMAL) LoggerHook + -------------------- +2023/06/05 14:57:06 - mmengine - INFO - load backbone in model from: ckpt/openclip-ViT-L-14.pth +2023/06/05 14:57:07 - mmengine - WARNING - The model and loaded state dict do not match exactly + +unexpected key in source state_dict: ln1.weight, ln1.bias + +2023/06/05 14:57:07 - mmengine - INFO - load backbone in model from: ckpt/openclip-ViT-L-14.pth +2023/06/05 14:57:09 - mmengine - WARNING - The model and loaded state dict do not match exactly + +unexpected key in source state_dict: cls_token, pos_embed, patch_embed.projection.weight, pre_norm.weight, pre_norm.bias, layers.0.ln1.weight, layers.0.ln1.bias, layers.0.attn.qkv.weight, layers.0.attn.qkv.bias, layers.0.attn.proj.weight, layers.0.attn.proj.bias, layers.0.ln2.weight, layers.0.ln2.bias, layers.0.ffn.layers.0.0.weight, layers.0.ffn.layers.0.0.bias, layers.0.ffn.layers.1.weight, layers.0.ffn.layers.1.bias, layers.1.ln1.weight, layers.1.ln1.bias, layers.1.attn.qkv.weight, layers.1.attn.qkv.bias, layers.1.attn.proj.weight, layers.1.attn.proj.bias, layers.1.ln2.weight, layers.1.ln2.bias, layers.1.ffn.layers.0.0.weight, layers.1.ffn.layers.0.0.bias, layers.1.ffn.layers.1.weight, layers.1.ffn.layers.1.bias, layers.2.ln1.weight, layers.2.ln1.bias, layers.2.attn.qkv.weight, layers.2.attn.qkv.bias, layers.2.attn.proj.weight, layers.2.attn.proj.bias, layers.2.ln2.weight, layers.2.ln2.bias, layers.2.ffn.layers.0.0.weight, layers.2.ffn.layers.0.0.bias, layers.2.ffn.layers.1.weight, layers.2.ffn.layers.1.bias, layers.3.ln1.weight, layers.3.ln1.bias, layers.3.attn.qkv.weight, layers.3.attn.qkv.bias, layers.3.attn.proj.weight, layers.3.attn.proj.bias, layers.3.ln2.weight, layers.3.ln2.bias, layers.3.ffn.layers.0.0.weight, layers.3.ffn.layers.0.0.bias, layers.3.ffn.layers.1.weight, layers.3.ffn.layers.1.bias, layers.4.ln1.weight, layers.4.ln1.bias, layers.4.attn.qkv.weight, layers.4.attn.qkv.bias, layers.4.attn.proj.weight, layers.4.attn.proj.bias, layers.4.ln2.weight, layers.4.ln2.bias, layers.4.ffn.layers.0.0.weight, layers.4.ffn.layers.0.0.bias, layers.4.ffn.layers.1.weight, layers.4.ffn.layers.1.bias, layers.5.ln1.weight, layers.5.ln1.bias, layers.5.attn.qkv.weight, layers.5.attn.qkv.bias, layers.5.attn.proj.weight, layers.5.attn.proj.bias, layers.5.ln2.weight, layers.5.ln2.bias, layers.5.ffn.layers.0.0.weight, layers.5.ffn.layers.0.0.bias, layers.5.ffn.layers.1.weight, layers.5.ffn.layers.1.bias, layers.6.ln1.weight, layers.6.ln1.bias, layers.6.attn.qkv.weight, layers.6.attn.qkv.bias, layers.6.attn.proj.weight, layers.6.attn.proj.bias, layers.6.ln2.weight, layers.6.ln2.bias, layers.6.ffn.layers.0.0.weight, layers.6.ffn.layers.0.0.bias, layers.6.ffn.layers.1.weight, layers.6.ffn.layers.1.bias, layers.7.ln1.weight, layers.7.ln1.bias, layers.7.attn.qkv.weight, layers.7.attn.qkv.bias, layers.7.attn.proj.weight, layers.7.attn.proj.bias, layers.7.ln2.weight, layers.7.ln2.bias, layers.7.ffn.layers.0.0.weight, layers.7.ffn.layers.0.0.bias, layers.7.ffn.layers.1.weight, layers.7.ffn.layers.1.bias, layers.8.ln1.weight, layers.8.ln1.bias, layers.8.attn.qkv.weight, layers.8.attn.qkv.bias, layers.8.attn.proj.weight, layers.8.attn.proj.bias, layers.8.ln2.weight, layers.8.ln2.bias, layers.8.ffn.layers.0.0.weight, layers.8.ffn.layers.0.0.bias, layers.8.ffn.layers.1.weight, layers.8.ffn.layers.1.bias, layers.9.ln1.weight, layers.9.ln1.bias, layers.9.attn.qkv.weight, layers.9.attn.qkv.bias, layers.9.attn.proj.weight, layers.9.attn.proj.bias, layers.9.ln2.weight, layers.9.ln2.bias, layers.9.ffn.layers.0.0.weight, layers.9.ffn.layers.0.0.bias, layers.9.ffn.layers.1.weight, layers.9.ffn.layers.1.bias, layers.10.ln1.weight, layers.10.ln1.bias, layers.10.attn.qkv.weight, layers.10.attn.qkv.bias, layers.10.attn.proj.weight, layers.10.attn.proj.bias, layers.10.ln2.weight, layers.10.ln2.bias, layers.10.ffn.layers.0.0.weight, layers.10.ffn.layers.0.0.bias, layers.10.ffn.layers.1.weight, layers.10.ffn.layers.1.bias, layers.11.ln1.weight, layers.11.ln1.bias, layers.11.attn.qkv.weight, layers.11.attn.qkv.bias, layers.11.attn.proj.weight, layers.11.attn.proj.bias, layers.11.ln2.weight, layers.11.ln2.bias, layers.11.ffn.layers.0.0.weight, layers.11.ffn.layers.0.0.bias, layers.11.ffn.layers.1.weight, layers.11.ffn.layers.1.bias, layers.12.ln1.weight, layers.12.ln1.bias, layers.12.attn.qkv.weight, layers.12.attn.qkv.bias, layers.12.attn.proj.weight, layers.12.attn.proj.bias, layers.12.ln2.weight, layers.12.ln2.bias, layers.12.ffn.layers.0.0.weight, layers.12.ffn.layers.0.0.bias, layers.12.ffn.layers.1.weight, layers.12.ffn.layers.1.bias, layers.13.ln1.weight, layers.13.ln1.bias, layers.13.attn.qkv.weight, layers.13.attn.qkv.bias, layers.13.attn.proj.weight, layers.13.attn.proj.bias, layers.13.ln2.weight, layers.13.ln2.bias, layers.13.ffn.layers.0.0.weight, layers.13.ffn.layers.0.0.bias, layers.13.ffn.layers.1.weight, layers.13.ffn.layers.1.bias, layers.14.ln1.weight, layers.14.ln1.bias, layers.14.attn.qkv.weight, layers.14.attn.qkv.bias, layers.14.attn.proj.weight, layers.14.attn.proj.bias, layers.14.ln2.weight, layers.14.ln2.bias, layers.14.ffn.layers.0.0.weight, layers.14.ffn.layers.0.0.bias, layers.14.ffn.layers.1.weight, layers.14.ffn.layers.1.bias, layers.15.ln1.weight, layers.15.ln1.bias, layers.15.attn.qkv.weight, layers.15.attn.qkv.bias, layers.15.attn.proj.weight, layers.15.attn.proj.bias, layers.15.ln2.weight, layers.15.ln2.bias, layers.15.ffn.layers.0.0.weight, layers.15.ffn.layers.0.0.bias, layers.15.ffn.layers.1.weight, layers.15.ffn.layers.1.bias, layers.16.ln1.weight, layers.16.ln1.bias, layers.16.attn.qkv.weight, layers.16.attn.qkv.bias, layers.16.attn.proj.weight, layers.16.attn.proj.bias, layers.16.ln2.weight, layers.16.ln2.bias, layers.16.ffn.layers.0.0.weight, layers.16.ffn.layers.0.0.bias, layers.16.ffn.layers.1.weight, layers.16.ffn.layers.1.bias, layers.17.ln1.weight, layers.17.ln1.bias, layers.17.attn.qkv.weight, layers.17.attn.qkv.bias, layers.17.attn.proj.weight, layers.17.attn.proj.bias, layers.17.ln2.weight, layers.17.ln2.bias, layers.17.ffn.layers.0.0.weight, layers.17.ffn.layers.0.0.bias, layers.17.ffn.layers.1.weight, layers.17.ffn.layers.1.bias, layers.18.ln1.weight, layers.18.ln1.bias, layers.18.attn.qkv.weight, layers.18.attn.qkv.bias, layers.18.attn.proj.weight, layers.18.attn.proj.bias, layers.18.ln2.weight, layers.18.ln2.bias, layers.18.ffn.layers.0.0.weight, layers.18.ffn.layers.0.0.bias, layers.18.ffn.layers.1.weight, layers.18.ffn.layers.1.bias, layers.19.ln1.weight, layers.19.ln1.bias, layers.19.attn.qkv.weight, layers.19.attn.qkv.bias, layers.19.attn.proj.weight, layers.19.attn.proj.bias, layers.19.ln2.weight, layers.19.ln2.bias, layers.19.ffn.layers.0.0.weight, layers.19.ffn.layers.0.0.bias, layers.19.ffn.layers.1.weight, layers.19.ffn.layers.1.bias, layers.20.ln1.weight, layers.20.ln1.bias, layers.20.attn.qkv.weight, layers.20.attn.qkv.bias, layers.20.attn.proj.weight, layers.20.attn.proj.bias, layers.20.ln2.weight, layers.20.ln2.bias, layers.20.ffn.layers.0.0.weight, layers.20.ffn.layers.0.0.bias, layers.20.ffn.layers.1.weight, layers.20.ffn.layers.1.bias, layers.21.ln1.weight, layers.21.ln1.bias, layers.21.attn.qkv.weight, layers.21.attn.qkv.bias, layers.21.attn.proj.weight, layers.21.attn.proj.bias, layers.21.ln2.weight, layers.21.ln2.bias, layers.21.ffn.layers.0.0.weight, layers.21.ffn.layers.0.0.bias, layers.21.ffn.layers.1.weight, layers.21.ffn.layers.1.bias, layers.22.ln1.weight, layers.22.ln1.bias, layers.22.attn.qkv.weight, layers.22.attn.qkv.bias, layers.22.attn.proj.weight, layers.22.attn.proj.bias, layers.22.ln2.weight, layers.22.ln2.bias, layers.22.ffn.layers.0.0.weight, layers.22.ffn.layers.0.0.bias, layers.22.ffn.layers.1.weight, layers.22.ffn.layers.1.bias, layers.23.ln1.weight, layers.23.ln1.bias, layers.23.attn.qkv.weight, layers.23.attn.qkv.bias, layers.23.attn.proj.weight, layers.23.attn.proj.bias, layers.23.ln2.weight, layers.23.ln2.bias, layers.23.ffn.layers.0.0.weight, layers.23.ffn.layers.0.0.bias, layers.23.ffn.layers.1.weight, layers.23.ffn.layers.1.bias, ln1.weight, ln1.bias + +missing keys in source state_dict: proj + +Name of parameter - Initialization information + +backbone.cls_token - torch.Size([1, 1, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.pos_embed - torch.Size([1, 257, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.patch_embed.projection.weight - torch.Size([1024, 3, 14, 14]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.0.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.0.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.0.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.0.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.0.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.0.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.0.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.0.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.0.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.0.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.0.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.0.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.1.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.1.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.1.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.1.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.1.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.1.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.1.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.1.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.1.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.1.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.1.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.1.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.2.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.2.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.2.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.2.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.2.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.2.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.2.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.2.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.2.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.2.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.2.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.2.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.3.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.3.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.3.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.3.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.3.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.3.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.3.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.3.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.3.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.3.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.3.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.3.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.4.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.4.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.4.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.4.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.4.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.4.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.4.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.4.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.4.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.4.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.4.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.4.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.5.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.5.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.5.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.5.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.5.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.5.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.5.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.5.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.5.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.5.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.5.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.5.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.6.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.6.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.6.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.6.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.6.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.6.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.6.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.6.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.6.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.6.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.6.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.6.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.7.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.7.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.7.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.7.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.7.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.7.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.7.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.7.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.7.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.7.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.7.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.7.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.8.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.8.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.8.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.8.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.8.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.8.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.8.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.8.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.8.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.8.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.8.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.8.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.9.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.9.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.9.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.9.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.9.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.9.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.9.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.9.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.9.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.9.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.9.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.9.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.10.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.10.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.10.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.10.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.10.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.10.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.10.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.10.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.10.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.10.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.10.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.10.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.11.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.11.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.11.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.11.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.11.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.11.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.11.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.11.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.11.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.11.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.11.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.11.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.12.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.12.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.12.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.12.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.12.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.12.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.12.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.12.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.12.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.12.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.12.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.12.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.13.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.13.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.13.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.13.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.13.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.13.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.13.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.13.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.13.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.13.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.13.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.13.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.14.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.14.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.14.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.14.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.14.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.14.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.14.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.14.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.14.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.14.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.14.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.14.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.15.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.15.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.15.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.15.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.15.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.15.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.15.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.15.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.15.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.15.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.15.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.15.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.16.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.16.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.16.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.16.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.16.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.16.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.16.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.16.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.16.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.16.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.16.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.16.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.17.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.17.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.17.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.17.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.17.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.17.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.17.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.17.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.17.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.17.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.17.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.17.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.18.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.18.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.18.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.18.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.18.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.18.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.18.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.18.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.18.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.18.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.18.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.18.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.19.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.19.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.19.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.19.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.19.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.19.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.19.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.19.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.19.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.19.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.19.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.19.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.20.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.20.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.20.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.20.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.20.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.20.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.20.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.20.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.20.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.20.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.20.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.20.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.21.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.21.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.21.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.21.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.21.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.21.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.21.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.21.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.21.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.21.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.21.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.21.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.22.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.22.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.22.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.22.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.22.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.22.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.22.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.22.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.22.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.22.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.22.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.22.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.23.ln1.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.23.ln1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.23.attn.qkv.weight - torch.Size([3072, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.23.attn.qkv.bias - torch.Size([3072]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.23.attn.proj.weight - torch.Size([1024, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.23.attn.proj.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.23.ln2.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.23.ln2.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.23.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.23.ffn.layers.0.0.bias - torch.Size([4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.23.ffn.layers.1.weight - torch.Size([1024, 4096]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.layers.23.ffn.layers.1.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.pre_norm.weight - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +backbone.pre_norm.bias - torch.Size([1024]): +PretrainedInit: load from ckpt/openclip-ViT-L-14.pth + +neck.proj - torch.Size([1024, 768]): +The value is the same before and after calling `init_weights` of ImageClassifier + +head.fc.weight - torch.Size([2, 768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +head.fc.bias - torch.Size([2]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 +2023/06/05 14:57:09 - mmengine - WARNING - "FileClient" will be deprecated in future. Please use io functions in https://mmengine.readthedocs.io/en/latest/api/fileio.html#file-io +2023/06/05 14:57:09 - mmengine - WARNING - "HardDiskBackend" is the alias of "LocalBackend" and the former will be deprecated in future. +2023/06/05 14:57:09 - mmengine - INFO - Checkpoints will be saved to /mnt/petrelfs/luzeyu/workspace/fakebench/mmpretrain/workdir/clip_large_pretrain_4x256_stylegan3_lr1e-4. +2023/06/05 14:58:18 - mmengine - INFO - Epoch(train) [1][ 100/4111] lr: 9.9999e-05 eta: 7:54:46 time: 0.6316 data_time: 0.0015 memory: 44139 loss: 0.5830 +2023/06/05 14:59:22 - mmengine - INFO - Epoch(train) [1][ 200/4111] lr: 9.9995e-05 eta: 7:32:36 time: 0.6347 data_time: 0.0016 memory: 44139 loss: 0.5629 +2023/06/05 15:00:25 - mmengine - INFO - Epoch(train) [1][ 300/4111] lr: 9.9988e-05 eta: 7:24:34 time: 0.6324 data_time: 0.0016 memory: 44139 loss: 0.5593 +2023/06/05 15:01:28 - mmengine - INFO - Epoch(train) [1][ 400/4111] lr: 9.9979e-05 eta: 7:20:03 time: 0.6338 data_time: 0.0015 memory: 44139 loss: 0.5523 +2023/06/05 15:02:32 - mmengine - INFO - Epoch(train) [1][ 500/4111] lr: 9.9967e-05 eta: 7:16:50 time: 0.6318 data_time: 0.0015 memory: 44139 loss: 0.5318 +2023/06/05 15:03:35 - mmengine - INFO - Epoch(train) [1][ 600/4111] lr: 9.9953e-05 eta: 7:14:20 time: 0.6330 data_time: 0.0017 memory: 44139 loss: 0.5652 +2023/06/05 15:04:38 - mmengine - INFO - Epoch(train) [1][ 700/4111] lr: 9.9936e-05 eta: 7:12:26 time: 0.6331 data_time: 0.0016 memory: 44139 loss: 0.5029 +2023/06/05 15:05:42 - mmengine - INFO - Epoch(train) [1][ 800/4111] lr: 9.9916e-05 eta: 7:10:37 time: 0.6333 data_time: 0.0015 memory: 44139 loss: 0.5037 +2023/06/05 15:06:45 - mmengine - INFO - Epoch(train) [1][ 900/4111] lr: 9.9894e-05 eta: 7:09:00 time: 0.6338 data_time: 0.0015 memory: 44139 loss: 0.5305 +2023/06/05 15:07:48 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_stylegan3_lr1e-4_20230605_145625 +2023/06/05 15:07:48 - mmengine - INFO - Epoch(train) [1][1000/4111] lr: 9.9869e-05 eta: 7:07:32 time: 0.6328 data_time: 0.0014 memory: 44139 loss: 0.5346 +2023/06/05 15:08:52 - mmengine - INFO - Epoch(train) [1][1100/4111] lr: 9.9841e-05 eta: 7:06:04 time: 0.6335 data_time: 0.0015 memory: 44139 loss: 0.5431 +2023/06/05 15:09:55 - mmengine - INFO - Epoch(train) [1][1200/4111] lr: 9.9811e-05 eta: 7:04:38 time: 0.6322 data_time: 0.0016 memory: 44139 loss: 0.5502 +2023/06/05 15:10:58 - mmengine - INFO - Epoch(train) [1][1300/4111] lr: 9.9778e-05 eta: 7:03:19 time: 0.6332 data_time: 0.0014 memory: 44139 loss: 0.4987 +2023/06/05 15:12:02 - mmengine - INFO - Epoch(train) [1][1400/4111] lr: 9.9743e-05 eta: 7:02:01 time: 0.6326 data_time: 0.0016 memory: 44139 loss: 0.5177 +2023/06/05 15:13:05 - mmengine - INFO - Epoch(train) [1][1500/4111] lr: 9.9705e-05 eta: 7:00:46 time: 0.6342 data_time: 0.0015 memory: 44139 loss: 0.4972 +2023/06/05 15:14:08 - mmengine - INFO - Epoch(train) [1][1600/4111] lr: 9.9664e-05 eta: 6:59:31 time: 0.6341 data_time: 0.0014 memory: 44139 loss: 0.5204 +2023/06/05 15:15:12 - mmengine - INFO - Epoch(train) [1][1700/4111] lr: 9.9621e-05 eta: 6:58:20 time: 0.6350 data_time: 0.0014 memory: 44139 loss: 0.5304 +2023/06/05 15:16:15 - mmengine - INFO - Epoch(train) [1][1800/4111] lr: 9.9575e-05 eta: 6:57:10 time: 0.6334 data_time: 0.0014 memory: 44139 loss: 0.4993 +2023/06/05 15:17:18 - mmengine - INFO - Epoch(train) [1][1900/4111] lr: 9.9527e-05 eta: 6:55:58 time: 0.6329 data_time: 0.0014 memory: 44139 loss: 0.5130 +2023/06/05 15:18:22 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_stylegan3_lr1e-4_20230605_145625 +2023/06/05 15:18:22 - mmengine - INFO - Epoch(train) [1][2000/4111] lr: 9.9476e-05 eta: 6:54:48 time: 0.6344 data_time: 0.0014 memory: 44139 loss: 0.4818 +2023/06/05 15:19:25 - mmengine - INFO - Epoch(train) [1][2100/4111] lr: 9.9422e-05 eta: 6:53:39 time: 0.6328 data_time: 0.0014 memory: 44139 loss: 0.4995 +2023/06/05 15:20:28 - mmengine - INFO - Epoch(train) [1][2200/4111] lr: 9.9366e-05 eta: 6:52:30 time: 0.6326 data_time: 0.0014 memory: 44139 loss: 0.5224 +2023/06/05 15:21:32 - mmengine - INFO - Epoch(train) [1][2300/4111] lr: 9.9307e-05 eta: 6:51:23 time: 0.6406 data_time: 0.0015 memory: 44139 loss: 0.5021 +2023/06/05 15:22:35 - mmengine - INFO - Epoch(train) [1][2400/4111] lr: 9.9246e-05 eta: 6:50:15 time: 0.6327 data_time: 0.0019 memory: 44139 loss: 0.5077 +2023/06/05 15:24:00 - mmengine - INFO - Epoch(train) [1][2500/4111] lr: 9.9182e-05 eta: 6:54:38 time: 0.6323 data_time: 0.0014 memory: 44139 loss: 0.5193 +2023/06/05 15:25:23 - mmengine - INFO - Epoch(train) [1][2600/4111] lr: 9.9115e-05 eta: 6:58:12 time: 0.6341 data_time: 0.0015 memory: 44139 loss: 0.4992 +2023/06/05 15:26:27 - mmengine - INFO - Epoch(train) [1][2700/4111] lr: 9.9046e-05 eta: 6:56:43 time: 0.6350 data_time: 0.0015 memory: 44139 loss: 0.4964 +2023/06/05 15:27:30 - mmengine - INFO - Epoch(train) [1][2800/4111] lr: 9.8974e-05 eta: 6:55:15 time: 0.6336 data_time: 0.0017 memory: 44139 loss: 0.5007 +2023/06/05 15:28:33 - mmengine - INFO - Epoch(train) [1][2900/4111] lr: 9.8900e-05 eta: 6:53:49 time: 0.6339 data_time: 0.0015 memory: 44139 loss: 0.5037 +2023/06/05 15:29:37 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_stylegan3_lr1e-4_20230605_145625 +2023/06/05 15:29:37 - mmengine - INFO - Epoch(train) [1][3000/4111] lr: 9.8823e-05 eta: 6:52:23 time: 0.6328 data_time: 0.0014 memory: 44139 loss: 0.4871 +2023/06/05 15:30:40 - mmengine - INFO - Epoch(train) [1][3100/4111] lr: 9.8744e-05 eta: 6:50:59 time: 0.6331 data_time: 0.0016 memory: 44139 loss: 0.5132 +2023/06/05 15:31:44 - mmengine - INFO - Epoch(train) [1][3200/4111] lr: 9.8662e-05 eta: 6:49:38 time: 0.6419 data_time: 0.0015 memory: 44139 loss: 0.4801 +2023/06/05 15:32:47 - mmengine - INFO - Epoch(train) [1][3300/4111] lr: 9.8578e-05 eta: 6:48:18 time: 0.6353 data_time: 0.0015 memory: 44139 loss: 0.4806 +2023/06/05 15:33:51 - mmengine - INFO - Epoch(train) [1][3400/4111] lr: 9.8490e-05 eta: 6:46:58 time: 0.6341 data_time: 0.0014 memory: 44139 loss: 0.4827 +2023/06/05 15:34:54 - mmengine - INFO - Epoch(train) [1][3500/4111] lr: 9.8401e-05 eta: 6:45:38 time: 0.6333 data_time: 0.0016 memory: 44139 loss: 0.5124 +2023/06/05 15:35:57 - mmengine - INFO - Epoch(train) [1][3600/4111] lr: 9.8309e-05 eta: 6:44:19 time: 0.6321 data_time: 0.0020 memory: 44139 loss: 0.4695 +2023/06/05 15:37:01 - mmengine - INFO - Epoch(train) [1][3700/4111] lr: 9.8214e-05 eta: 6:43:00 time: 0.6328 data_time: 0.0015 memory: 44139 loss: 0.4676 +2023/06/05 15:38:04 - mmengine - INFO - Epoch(train) [1][3800/4111] lr: 9.8117e-05 eta: 6:41:43 time: 0.6344 data_time: 0.0015 memory: 44139 loss: 0.4806 +2023/06/05 15:39:07 - mmengine - INFO - Epoch(train) [1][3900/4111] lr: 9.8017e-05 eta: 6:40:26 time: 0.6329 data_time: 0.0018 memory: 44139 loss: 0.4921 +2023/06/05 15:40:10 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_stylegan3_lr1e-4_20230605_145625 +2023/06/05 15:40:10 - mmengine - INFO - Epoch(train) [1][4000/4111] lr: 9.7915e-05 eta: 6:39:09 time: 0.6322 data_time: 0.0015 memory: 44139 loss: 0.5180 +2023/06/05 15:41:14 - mmengine - INFO - Epoch(train) [1][4100/4111] lr: 9.7810e-05 eta: 6:37:53 time: 0.6353 data_time: 0.0029 memory: 44139 loss: 0.5091 +2023/06/05 15:41:21 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_stylegan3_lr1e-4_20230605_145625 +2023/06/05 15:41:21 - mmengine - INFO - Saving checkpoint at 1 epochs +2023/06/05 15:43:45 - mmengine - INFO - Epoch(val) [1][100/119] eta: 0:00:25 time: 1.2823 data_time: 0.0007 memory: 44139 +2023/06/05 15:44:28 - mmengine - INFO - Epoch(val) [1][119/119] accuracy/top1: 79.6717 single-label/precision_classwise: [93.3755111694336, 51.55799865722656] single-label/recall_classwise: [79.81609344482422, 79.13953399658203] single-label/f1-score_classwise: [86.06501007080078, 62.43844985961914] data_time: 0.0231 time: 1.3137 +2023/06/05 15:45:34 - mmengine - INFO - Epoch(train) [2][ 100/4111] lr: 9.7691e-05 eta: 6:36:57 time: 0.6324 data_time: 0.0014 memory: 44140 loss: 0.4650 +2023/06/05 15:46:38 - mmengine - INFO - Epoch(train) [2][ 200/4111] lr: 9.7581e-05 eta: 6:35:42 time: 0.6337 data_time: 0.0015 memory: 44140 loss: 0.4936 +2023/06/05 15:47:41 - mmengine - INFO - Epoch(train) [2][ 300/4111] lr: 9.7469e-05 eta: 6:34:29 time: 0.6390 data_time: 0.0016 memory: 44140 loss: 0.4635 +2023/06/05 15:48:44 - mmengine - INFO - Epoch(train) [2][ 400/4111] lr: 9.7354e-05 eta: 6:33:15 time: 0.6334 data_time: 0.0017 memory: 44140 loss: 0.4712 +2023/06/05 15:49:48 - mmengine - INFO - Epoch(train) [2][ 500/4111] lr: 9.7236e-05 eta: 6:32:02 time: 0.6328 data_time: 0.0014 memory: 44140 loss: 0.4786 +2023/06/05 15:50:51 - mmengine - INFO - Epoch(train) [2][ 600/4111] lr: 9.7116e-05 eta: 6:30:49 time: 0.6326 data_time: 0.0015 memory: 44140 loss: 0.4785 +2023/06/05 15:51:54 - mmengine - INFO - Epoch(train) [2][ 700/4111] lr: 9.6994e-05 eta: 6:29:36 time: 0.6336 data_time: 0.0014 memory: 44140 loss: 0.4658 +2023/06/05 15:52:58 - mmengine - INFO - Epoch(train) [2][ 800/4111] lr: 9.6869e-05 eta: 6:28:25 time: 0.6338 data_time: 0.0013 memory: 44140 loss: 0.4925 +2023/06/05 15:55:22 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_stylegan3_lr1e-4_20230605_145625 +2023/06/05 15:55:29 - mmengine - INFO - Epoch(train) [2][ 900/4111] lr: 9.6742e-05 eta: 6:37:42 time: 0.6337 data_time: 0.0016 memory: 44140 loss: 0.4578 +2023/06/05 15:56:32 - mmengine - INFO - Epoch(train) [2][1000/4111] lr: 9.6612e-05 eta: 6:36:17 time: 0.6327 data_time: 0.0015 memory: 44140 loss: 0.4974 +2023/06/05 15:57:35 - mmengine - INFO - Epoch(train) [2][1100/4111] lr: 9.6480e-05 eta: 6:34:52 time: 0.6334 data_time: 0.0014 memory: 44140 loss: 0.4713 +2023/06/05 15:58:39 - mmengine - INFO - Epoch(train) [2][1200/4111] lr: 9.6346e-05 eta: 6:33:28 time: 0.6328 data_time: 0.0015 memory: 44140 loss: 0.4649 +2023/06/05 15:59:42 - mmengine - INFO - Epoch(train) [2][1300/4111] lr: 9.6209e-05 eta: 6:32:05 time: 0.6339 data_time: 0.0014 memory: 44140 loss: 0.4518 +2023/06/05 16:00:45 - mmengine - INFO - Epoch(train) [2][1400/4111] lr: 9.6069e-05 eta: 6:30:43 time: 0.6334 data_time: 0.0015 memory: 44140 loss: 0.4486 +2023/06/05 16:01:49 - mmengine - INFO - Epoch(train) [2][1500/4111] lr: 9.5928e-05 eta: 6:29:21 time: 0.6343 data_time: 0.0013 memory: 44140 loss: 0.4892 +2023/06/05 16:02:52 - mmengine - INFO - Epoch(train) [2][1600/4111] lr: 9.5783e-05 eta: 6:28:00 time: 0.6324 data_time: 0.0014 memory: 44140 loss: 0.4853 +2023/06/05 16:03:55 - mmengine - INFO - Epoch(train) [2][1700/4111] lr: 9.5637e-05 eta: 6:26:39 time: 0.6330 data_time: 0.0015 memory: 44140 loss: 0.4755 +2023/06/05 16:04:59 - mmengine - INFO - Epoch(train) [2][1800/4111] lr: 9.5488e-05 eta: 6:25:20 time: 0.6336 data_time: 0.0015 memory: 44140 loss: 0.5018 +2023/06/05 16:05:55 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_stylegan3_lr1e-4_20230605_145625 +2023/06/05 16:06:02 - mmengine - INFO - Epoch(train) [2][1900/4111] lr: 9.5337e-05 eta: 6:24:01 time: 0.6343 data_time: 0.0015 memory: 44140 loss: 0.4598 +2023/06/05 16:07:06 - mmengine - INFO - Epoch(train) [2][2000/4111] lr: 9.5183e-05 eta: 6:22:42 time: 0.6327 data_time: 0.0014 memory: 44140 loss: 0.4625 +2023/06/05 16:08:09 - mmengine - INFO - Epoch(train) [2][2100/4111] lr: 9.5027e-05 eta: 6:21:23 time: 0.6341 data_time: 0.0014 memory: 44140 loss: 0.4712 +2023/06/05 16:09:12 - mmengine - INFO - Epoch(train) [2][2200/4111] lr: 9.4869e-05 eta: 6:20:06 time: 0.6332 data_time: 0.0014 memory: 44140 loss: 0.4923 +2023/06/05 16:10:16 - mmengine - INFO - Epoch(train) [2][2300/4111] lr: 9.4708e-05 eta: 6:18:49 time: 0.6341 data_time: 0.0016 memory: 44140 loss: 0.4904 +2023/06/05 16:11:19 - mmengine - INFO - Epoch(train) [2][2400/4111] lr: 9.4545e-05 eta: 6:17:32 time: 0.6332 data_time: 0.0015 memory: 44140 loss: 0.4785 +2023/06/05 16:12:22 - mmengine - INFO - Epoch(train) [2][2500/4111] lr: 9.4380e-05 eta: 6:16:15 time: 0.6329 data_time: 0.0015 memory: 44140 loss: 0.4459 +2023/06/05 16:13:26 - mmengine - INFO - Epoch(train) [2][2600/4111] lr: 9.4212e-05 eta: 6:14:59 time: 0.6406 data_time: 0.0014 memory: 44140 loss: 0.4434 +2023/06/05 16:14:29 - mmengine - INFO - Epoch(train) [2][2700/4111] lr: 9.4043e-05 eta: 6:13:44 time: 0.6336 data_time: 0.0014 memory: 44140 loss: 0.4675 +2023/06/05 16:15:33 - mmengine - INFO - Epoch(train) [2][2800/4111] lr: 9.3870e-05 eta: 6:12:28 time: 0.6351 data_time: 0.0013 memory: 44140 loss: 0.4625 +2023/06/05 16:16:29 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_stylegan3_lr1e-4_20230605_145625 +2023/06/05 16:16:36 - mmengine - INFO - Epoch(train) [2][2900/4111] lr: 9.3696e-05 eta: 6:11:13 time: 0.6327 data_time: 0.0014 memory: 44140 loss: 0.4640 +2023/06/05 16:17:39 - mmengine - INFO - Epoch(train) [2][3000/4111] lr: 9.3519e-05 eta: 6:09:59 time: 0.6348 data_time: 0.0015 memory: 44140 loss: 0.4867 +2023/06/05 16:18:43 - mmengine - INFO - Epoch(train) [2][3100/4111] lr: 9.3340e-05 eta: 6:08:45 time: 0.6329 data_time: 0.0014 memory: 44140 loss: 0.4709 +2023/06/05 16:19:46 - mmengine - INFO - Epoch(train) [2][3200/4111] lr: 9.3159e-05 eta: 6:07:31 time: 0.6385 data_time: 0.0013 memory: 44140 loss: 0.4843 +2023/06/05 16:20:50 - mmengine - INFO - Epoch(train) [2][3300/4111] lr: 9.2976e-05 eta: 6:06:18 time: 0.6329 data_time: 0.0015 memory: 44140 loss: 0.4710 +2023/06/05 16:21:53 - mmengine - INFO - Epoch(train) [2][3400/4111] lr: 9.2790e-05 eta: 6:05:04 time: 0.6322 data_time: 0.0015 memory: 44140 loss: 0.4563 +2023/06/05 16:22:56 - mmengine - INFO - Epoch(train) [2][3500/4111] lr: 9.2603e-05 eta: 6:03:50 time: 0.6325 data_time: 0.0014 memory: 44140 loss: 0.4974 +2023/06/05 16:23:59 - mmengine - INFO - Epoch(train) [2][3600/4111] lr: 9.2413e-05 eta: 6:02:37 time: 0.6410 data_time: 0.0014 memory: 44140 loss: 0.4968 +2023/06/05 16:25:03 - mmengine - INFO - Epoch(train) [2][3700/4111] lr: 9.2220e-05 eta: 6:01:25 time: 0.6350 data_time: 0.0015 memory: 44140 loss: 0.4634 +2023/06/05 16:26:06 - mmengine - INFO - Epoch(train) [2][3800/4111] lr: 9.2026e-05 eta: 6:00:13 time: 0.6337 data_time: 0.0014 memory: 44140 loss: 0.4575 +2023/06/05 16:27:03 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_stylegan3_lr1e-4_20230605_145625 +2023/06/05 16:27:10 - mmengine - INFO - Epoch(train) [2][3900/4111] lr: 9.1829e-05 eta: 5:59:00 time: 0.6323 data_time: 0.0015 memory: 44140 loss: 0.4827 +2023/06/05 16:28:13 - mmengine - INFO - Epoch(train) [2][4000/4111] lr: 9.1631e-05 eta: 5:57:48 time: 0.6330 data_time: 0.0014 memory: 44140 loss: 0.4647 +2023/06/05 16:29:17 - mmengine - INFO - Epoch(train) [2][4100/4111] lr: 9.1430e-05 eta: 5:56:36 time: 0.6347 data_time: 0.0035 memory: 44140 loss: 0.4715 +2023/06/05 16:29:23 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_stylegan3_lr1e-4_20230605_145625 +2023/06/05 16:29:23 - mmengine - INFO - Saving checkpoint at 2 epochs +2023/06/05 16:31:44 - mmengine - INFO - Epoch(val) [2][100/119] eta: 0:00:24 time: 1.2822 data_time: 0.0007 memory: 44140 +2023/06/05 16:32:27 - mmengine - INFO - Epoch(val) [2][119/119] accuracy/top1: 84.4099 single-label/precision_classwise: [94.01409149169922, 60.15287780761719] single-label/recall_classwise: [85.63011932373047, 79.91473388671875] single-label/f1-score_classwise: [89.62646484375, 68.63972473144531] data_time: 0.0179 time: 1.2802 +2023/06/05 16:33:33 - mmengine - INFO - Epoch(train) [3][ 100/4111] lr: 9.1205e-05 eta: 5:55:29 time: 0.6324 data_time: 0.0017 memory: 44140 loss: 0.5030 +2023/06/05 16:34:37 - mmengine - INFO - Epoch(train) [3][ 200/4111] lr: 9.0999e-05 eta: 5:54:18 time: 0.6347 data_time: 0.0015 memory: 44140 loss: 0.4575 +2023/06/05 16:35:40 - mmengine - INFO - Epoch(train) [3][ 300/4111] lr: 9.0792e-05 eta: 5:53:06 time: 0.6327 data_time: 0.0016 memory: 44140 loss: 0.4557 +2023/06/05 16:36:44 - mmengine - INFO - Epoch(train) [3][ 400/4111] lr: 9.0582e-05 eta: 5:51:56 time: 0.6334 data_time: 0.0016 memory: 44140 loss: 0.4725 +2023/06/05 16:37:47 - mmengine - INFO - Epoch(train) [3][ 500/4111] lr: 9.0371e-05 eta: 5:50:45 time: 0.6333 data_time: 0.0015 memory: 44140 loss: 0.4796 +2023/06/05 16:38:50 - mmengine - INFO - Epoch(train) [3][ 600/4111] lr: 9.0157e-05 eta: 5:49:34 time: 0.6340 data_time: 0.0014 memory: 44140 loss: 0.4742 +2023/06/05 16:39:54 - mmengine - INFO - Epoch(train) [3][ 700/4111] lr: 8.9942e-05 eta: 5:48:24 time: 0.6336 data_time: 0.0014 memory: 44140 loss: 0.4322 +2023/06/05 16:40:43 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_stylegan3_lr1e-4_20230605_145625 +2023/06/05 16:40:57 - mmengine - INFO - Epoch(train) [3][ 800/4111] lr: 8.9724e-05 eta: 5:47:14 time: 0.6357 data_time: 0.0016 memory: 44140 loss: 0.4728 +2023/06/05 16:42:01 - mmengine - INFO - Epoch(train) [3][ 900/4111] lr: 8.9504e-05 eta: 5:46:04 time: 0.6339 data_time: 0.0016 memory: 44140 loss: 0.4606 +2023/06/05 16:43:04 - mmengine - INFO - Epoch(train) [3][1000/4111] lr: 8.9282e-05 eta: 5:44:53 time: 0.6340 data_time: 0.0014 memory: 44140 loss: 0.4547 +2023/06/05 16:44:08 - mmengine - INFO - Epoch(train) [3][1100/4111] lr: 8.9059e-05 eta: 5:43:43 time: 0.6331 data_time: 0.0014 memory: 44140 loss: 0.4499 +2023/06/05 16:45:11 - mmengine - INFO - Epoch(train) [3][1200/4111] lr: 8.8833e-05 eta: 5:42:34 time: 0.6324 data_time: 0.0015 memory: 44140 loss: 0.4654 +2023/06/05 16:46:14 - mmengine - INFO - Epoch(train) [3][1300/4111] lr: 8.8605e-05 eta: 5:41:24 time: 0.6329 data_time: 0.0017 memory: 44140 loss: 0.4494 +2023/06/05 16:47:18 - mmengine - INFO - Epoch(train) [3][1400/4111] lr: 8.8375e-05 eta: 5:40:14 time: 0.6353 data_time: 0.0015 memory: 44140 loss: 0.4445 +2023/06/05 16:48:21 - mmengine - INFO - Epoch(train) [3][1500/4111] lr: 8.8144e-05 eta: 5:39:05 time: 0.6360 data_time: 0.0016 memory: 44140 loss: 0.4465 +2023/06/05 16:49:25 - mmengine - INFO - Epoch(train) [3][1600/4111] lr: 8.7910e-05 eta: 5:37:56 time: 0.6347 data_time: 0.0014 memory: 44140 loss: 0.4556 +2023/06/05 16:50:28 - mmengine - INFO - Epoch(train) [3][1700/4111] lr: 8.7675e-05 eta: 5:36:47 time: 0.6333 data_time: 0.0016 memory: 44140 loss: 0.4593 +2023/06/05 16:51:18 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_stylegan3_lr1e-4_20230605_145625 +2023/06/05 16:51:32 - mmengine - INFO - Epoch(train) [3][1800/4111] lr: 8.7437e-05 eta: 5:35:38 time: 0.6338 data_time: 0.0015 memory: 44140 loss: 0.4937 +2023/06/05 16:52:35 - mmengine - INFO - Epoch(train) [3][1900/4111] lr: 8.7198e-05 eta: 5:34:29 time: 0.6339 data_time: 0.0014 memory: 44140 loss: 0.4547 +2023/06/05 16:53:39 - mmengine - INFO - Epoch(train) [3][2000/4111] lr: 8.6957e-05 eta: 5:33:20 time: 0.6335 data_time: 0.0015 memory: 44140 loss: 0.4673 +2023/06/05 16:54:42 - mmengine - INFO - Epoch(train) [3][2100/4111] lr: 8.6714e-05 eta: 5:32:11 time: 0.6340 data_time: 0.0015 memory: 44140 loss: 0.4839 +2023/06/05 16:55:45 - mmengine - INFO - Epoch(train) [3][2200/4111] lr: 8.6469e-05 eta: 5:31:03 time: 0.6354 data_time: 0.0015 memory: 44140 loss: 0.4854 +2023/06/05 16:56:49 - mmengine - INFO - Epoch(train) [3][2300/4111] lr: 8.6222e-05 eta: 5:29:55 time: 0.6336 data_time: 0.0016 memory: 44140 loss: 0.4552 +2023/06/05 16:57:52 - mmengine - INFO - Epoch(train) [3][2400/4111] lr: 8.5974e-05 eta: 5:28:46 time: 0.6341 data_time: 0.0016 memory: 44140 loss: 0.4661 +2023/06/05 16:58:56 - mmengine - INFO - Epoch(train) [3][2500/4111] lr: 8.5723e-05 eta: 5:27:38 time: 0.6334 data_time: 0.0016 memory: 44140 loss: 0.4548 +2023/06/05 16:59:59 - mmengine - INFO - Epoch(train) [3][2600/4111] lr: 8.5471e-05 eta: 5:26:30 time: 0.6353 data_time: 0.0014 memory: 44140 loss: 0.4696 +2023/06/05 17:01:03 - mmengine - INFO - Epoch(train) [3][2700/4111] lr: 8.5217e-05 eta: 5:25:22 time: 0.6340 data_time: 0.0015 memory: 44140 loss: 0.4486 +2023/06/05 17:01:52 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_stylegan3_lr1e-4_20230605_145625 +2023/06/05 17:02:06 - mmengine - INFO - Epoch(train) [3][2800/4111] lr: 8.4962e-05 eta: 5:24:14 time: 0.6334 data_time: 0.0014 memory: 44140 loss: 0.4574 +2023/06/05 17:03:10 - mmengine - INFO - Epoch(train) [3][2900/4111] lr: 8.4704e-05 eta: 5:23:06 time: 0.6347 data_time: 0.0015 memory: 44140 loss: 0.4755 +2023/06/05 17:04:13 - mmengine - INFO - Epoch(train) [3][3000/4111] lr: 8.4445e-05 eta: 5:21:58 time: 0.6337 data_time: 0.0015 memory: 44140 loss: 0.4251 +2023/06/05 17:05:17 - mmengine - INFO - Epoch(train) [3][3100/4111] lr: 8.4184e-05 eta: 5:20:50 time: 0.6348 data_time: 0.0016 memory: 44140 loss: 0.4576 +2023/06/05 17:06:20 - mmengine - INFO - Epoch(train) [3][3200/4111] lr: 8.3921e-05 eta: 5:19:43 time: 0.6338 data_time: 0.0017 memory: 44140 loss: 0.4814 +2023/06/05 17:07:24 - mmengine - INFO - Epoch(train) [3][3300/4111] lr: 8.3657e-05 eta: 5:18:35 time: 0.6358 data_time: 0.0015 memory: 44140 loss: 0.4783 +2023/06/05 17:08:27 - mmengine - INFO - Epoch(train) [3][3400/4111] lr: 8.3391e-05 eta: 5:17:27 time: 0.6339 data_time: 0.0014 memory: 44140 loss: 0.4661 +2023/06/05 17:09:30 - mmengine - INFO - Epoch(train) [3][3500/4111] lr: 8.3123e-05 eta: 5:16:20 time: 0.6336 data_time: 0.0017 memory: 44140 loss: 0.4642 +2023/06/05 17:10:34 - mmengine - INFO - Epoch(train) [3][3600/4111] lr: 8.2854e-05 eta: 5:15:12 time: 0.6348 data_time: 0.0014 memory: 44140 loss: 0.4624 +2023/06/05 17:11:37 - mmengine - INFO - Epoch(train) [3][3700/4111] lr: 8.2583e-05 eta: 5:14:05 time: 0.6375 data_time: 0.0022 memory: 44140 loss: 0.4447 +2023/06/05 17:12:27 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_stylegan3_lr1e-4_20230605_145625 +2023/06/05 17:12:41 - mmengine - INFO - Epoch(train) [3][3800/4111] lr: 8.2311e-05 eta: 5:12:58 time: 0.6340 data_time: 0.0016 memory: 44140 loss: 0.4797 +2023/06/05 17:13:44 - mmengine - INFO - Epoch(train) [3][3900/4111] lr: 8.2037e-05 eta: 5:11:51 time: 0.6324 data_time: 0.0017 memory: 44140 loss: 0.4820 +2023/06/05 17:14:47 - mmengine - INFO - Epoch(train) [3][4000/4111] lr: 8.1761e-05 eta: 5:10:43 time: 0.6330 data_time: 0.0015 memory: 44140 loss: 0.4437 +2023/06/05 17:15:51 - mmengine - INFO - Epoch(train) [3][4100/4111] lr: 8.1484e-05 eta: 5:09:36 time: 0.6359 data_time: 0.0031 memory: 44140 loss: 0.4542 +2023/06/05 17:15:58 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_stylegan3_lr1e-4_20230605_145625 +2023/06/05 17:15:58 - mmengine - INFO - Saving checkpoint at 3 epochs +2023/06/05 17:18:19 - mmengine - INFO - Epoch(val) [3][100/119] eta: 0:00:24 time: 1.2927 data_time: 0.0007 memory: 44140 +2023/06/05 17:19:01 - mmengine - INFO - Epoch(val) [3][119/119] accuracy/top1: 86.5763 single-label/precision_classwise: [93.9571762084961, 65.35624694824219] single-label/recall_classwise: [88.63287353515625, 79.0] single-label/f1-score_classwise: [91.21739196777344, 71.53335571289062] data_time: 0.0178 time: 1.2792 +2023/06/05 17:20:08 - mmengine - INFO - Epoch(train) [4][ 100/4111] lr: 8.1174e-05 eta: 5:08:30 time: 0.6362 data_time: 0.0016 memory: 44140 loss: 0.4561 +2023/06/05 17:21:12 - mmengine - INFO - Epoch(train) [4][ 200/4111] lr: 8.0894e-05 eta: 5:07:23 time: 0.6330 data_time: 0.0014 memory: 44140 loss: 0.4487 +2023/06/05 17:22:15 - mmengine - INFO - Epoch(train) [4][ 300/4111] lr: 8.0612e-05 eta: 5:06:16 time: 0.6329 data_time: 0.0016 memory: 44140 loss: 0.4526 +2023/06/05 17:23:19 - mmengine - INFO - Epoch(train) [4][ 400/4111] lr: 8.0328e-05 eta: 5:05:09 time: 0.6337 data_time: 0.0015 memory: 44140 loss: 0.4790 +2023/06/05 17:24:22 - mmengine - INFO - Epoch(train) [4][ 500/4111] lr: 8.0043e-05 eta: 5:04:02 time: 0.6340 data_time: 0.0016 memory: 44140 loss: 0.4166 +2023/06/05 17:25:25 - mmengine - INFO - Epoch(train) [4][ 600/4111] lr: 7.9757e-05 eta: 5:02:55 time: 0.6360 data_time: 0.0014 memory: 44140 loss: 0.4761 +2023/06/05 17:26:08 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_stylegan3_lr1e-4_20230605_145625 +2023/06/05 17:26:29 - mmengine - INFO - Epoch(train) [4][ 700/4111] lr: 7.9469e-05 eta: 5:01:48 time: 0.6362 data_time: 0.0017 memory: 44140 loss: 0.4425 +2023/06/05 17:27:32 - mmengine - INFO - Epoch(train) [4][ 800/4111] lr: 7.9180e-05 eta: 5:00:42 time: 0.6345 data_time: 0.0018 memory: 44140 loss: 0.4640 +2023/06/05 17:28:36 - mmengine - INFO - Epoch(train) [4][ 900/4111] lr: 7.8889e-05 eta: 4:59:35 time: 0.6354 data_time: 0.0018 memory: 44140 loss: 0.4716 +2023/06/05 17:29:39 - mmengine - INFO - Epoch(train) [4][1000/4111] lr: 7.8597e-05 eta: 4:58:28 time: 0.6351 data_time: 0.0021 memory: 44140 loss: 0.4530 +2023/06/05 17:30:43 - mmengine - INFO - Epoch(train) [4][1100/4111] lr: 7.8303e-05 eta: 4:57:22 time: 0.6351 data_time: 0.0022 memory: 44140 loss: 0.4628 +2023/06/05 17:31:47 - mmengine - INFO - Epoch(train) [4][1200/4111] lr: 7.8008e-05 eta: 4:56:16 time: 0.6344 data_time: 0.0022 memory: 44140 loss: 0.4498 +2023/06/05 17:32:50 - mmengine - INFO - Epoch(train) [4][1300/4111] lr: 7.7712e-05 eta: 4:55:09 time: 0.6327 data_time: 0.0021 memory: 44140 loss: 0.4608 +2023/06/05 17:33:54 - mmengine - INFO - Epoch(train) [4][1400/4111] lr: 7.7415e-05 eta: 4:54:03 time: 0.6350 data_time: 0.0016 memory: 44140 loss: 0.4272 +2023/06/05 17:34:57 - mmengine - INFO - Epoch(train) [4][1500/4111] lr: 7.7116e-05 eta: 4:52:57 time: 0.6337 data_time: 0.0018 memory: 44140 loss: 0.4416 +2023/06/05 17:36:00 - mmengine - INFO - Epoch(train) [4][1600/4111] lr: 7.6816e-05 eta: 4:51:50 time: 0.6331 data_time: 0.0015 memory: 44140 loss: 0.4656 +2023/06/05 17:36:43 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_stylegan3_lr1e-4_20230605_145625 +2023/06/05 17:37:04 - mmengine - INFO - Epoch(train) [4][1700/4111] lr: 7.6514e-05 eta: 4:50:44 time: 0.6336 data_time: 0.0015 memory: 44140 loss: 0.4717 +2023/06/05 17:38:07 - mmengine - INFO - Epoch(train) [4][1800/4111] lr: 7.6212e-05 eta: 4:49:37 time: 0.6328 data_time: 0.0022 memory: 44140 loss: 0.4877 +2023/06/05 17:39:11 - mmengine - INFO - Epoch(train) [4][1900/4111] lr: 7.5908e-05 eta: 4:48:31 time: 0.6333 data_time: 0.0018 memory: 44140 loss: 0.4612 +2023/06/05 17:40:14 - mmengine - INFO - Epoch(train) [4][2000/4111] lr: 7.5603e-05 eta: 4:47:25 time: 0.6343 data_time: 0.0018 memory: 44140 loss: 0.4622 +2023/06/05 17:41:18 - mmengine - INFO - Epoch(train) [4][2100/4111] lr: 7.5296e-05 eta: 4:46:19 time: 0.6360 data_time: 0.0013 memory: 44140 loss: 0.4644 +2023/06/05 17:42:21 - mmengine - INFO - Epoch(train) [4][2200/4111] lr: 7.4989e-05 eta: 4:45:13 time: 0.6339 data_time: 0.0016 memory: 44140 loss: 0.4922 +2023/06/05 17:43:25 - mmengine - INFO - Epoch(train) [4][2300/4111] lr: 7.4680e-05 eta: 4:44:07 time: 0.6380 data_time: 0.0015 memory: 44140 loss: 0.4758 +2023/06/05 17:44:28 - mmengine - INFO - Epoch(train) [4][2400/4111] lr: 7.4370e-05 eta: 4:43:01 time: 0.6327 data_time: 0.0018 memory: 44140 loss: 0.4429 +2023/06/05 17:45:32 - mmengine - INFO - Epoch(train) [4][2500/4111] lr: 7.4059e-05 eta: 4:41:55 time: 0.6339 data_time: 0.0015 memory: 44140 loss: 0.4586 +2023/06/05 17:46:35 - mmengine - INFO - Epoch(train) [4][2600/4111] lr: 7.3747e-05 eta: 4:40:49 time: 0.6329 data_time: 0.0017 memory: 44140 loss: 0.4808 +2023/06/05 17:47:17 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_stylegan3_lr1e-4_20230605_145625 +2023/06/05 17:47:38 - mmengine - INFO - Epoch(train) [4][2700/4111] lr: 7.3434e-05 eta: 4:39:43 time: 0.6341 data_time: 0.0015 memory: 44140 loss: 0.4389 +2023/06/05 17:48:42 - mmengine - INFO - Epoch(train) [4][2800/4111] lr: 7.3120e-05 eta: 4:38:36 time: 0.6334 data_time: 0.0017 memory: 44140 loss: 0.4692 +2023/06/05 17:49:45 - mmengine - INFO - Epoch(train) [4][2900/4111] lr: 7.2804e-05 eta: 4:37:30 time: 0.6360 data_time: 0.0018 memory: 44140 loss: 0.4431 +2023/06/05 17:50:49 - mmengine - INFO - Epoch(train) [4][3000/4111] lr: 7.2488e-05 eta: 4:36:25 time: 0.6328 data_time: 0.0021 memory: 44140 loss: 0.4778 +2023/06/05 17:51:52 - mmengine - INFO - Epoch(train) [4][3100/4111] lr: 7.2171e-05 eta: 4:35:19 time: 0.6350 data_time: 0.0024 memory: 44140 loss: 0.4468 +2023/06/05 17:52:55 - mmengine - INFO - Epoch(train) [4][3200/4111] lr: 7.1852e-05 eta: 4:34:13 time: 0.6345 data_time: 0.0016 memory: 44140 loss: 0.4608 +2023/06/05 17:53:59 - mmengine - INFO - Epoch(train) [4][3300/4111] lr: 7.1533e-05 eta: 4:33:07 time: 0.6354 data_time: 0.0016 memory: 44140 loss: 0.4724 +2023/06/05 17:55:03 - mmengine - INFO - Epoch(train) [4][3400/4111] lr: 7.1213e-05 eta: 4:32:02 time: 0.6337 data_time: 0.0018 memory: 44140 loss: 0.4933 +2023/06/05 17:56:06 - mmengine - INFO - Epoch(train) [4][3500/4111] lr: 7.0891e-05 eta: 4:30:56 time: 0.6343 data_time: 0.0015 memory: 44140 loss: 0.4477 +2023/06/05 17:57:09 - mmengine - INFO - Epoch(train) [4][3600/4111] lr: 7.0569e-05 eta: 4:29:50 time: 0.6341 data_time: 0.0016 memory: 44140 loss: 0.4720 +2023/06/05 17:57:52 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_stylegan3_lr1e-4_20230605_145625 +2023/06/05 17:58:13 - mmengine - INFO - Epoch(train) [4][3700/4111] lr: 7.0246e-05 eta: 4:28:45 time: 0.6353 data_time: 0.0015 memory: 44140 loss: 0.4624 +2023/06/05 17:59:17 - mmengine - INFO - Epoch(train) [4][3800/4111] lr: 6.9922e-05 eta: 4:27:39 time: 0.6347 data_time: 0.0015 memory: 44140 loss: 0.4482 +2023/06/05 18:00:20 - mmengine - INFO - Epoch(train) [4][3900/4111] lr: 6.9597e-05 eta: 4:26:34 time: 0.6340 data_time: 0.0020 memory: 44140 loss: 0.4535 +2023/06/05 18:01:24 - mmengine - INFO - Epoch(train) [4][4000/4111] lr: 6.9272e-05 eta: 4:25:28 time: 0.6350 data_time: 0.0016 memory: 44140 loss: 0.4347 +2023/06/05 18:02:27 - mmengine - INFO - Epoch(train) [4][4100/4111] lr: 6.8945e-05 eta: 4:24:23 time: 0.6419 data_time: 0.0030 memory: 44140 loss: 0.4881 +2023/06/05 18:02:34 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_stylegan3_lr1e-4_20230605_145625 +2023/06/05 18:02:34 - mmengine - INFO - Saving checkpoint at 4 epochs +2023/06/05 18:04:54 - mmengine - INFO - Epoch(val) [4][100/119] eta: 0:00:24 time: 1.2813 data_time: 0.0005 memory: 44140 +2023/06/05 18:05:37 - mmengine - INFO - Epoch(val) [4][119/119] accuracy/top1: 87.7497 single-label/precision_classwise: [94.0125961303711, 68.5192642211914] single-label/recall_classwise: [90.1668701171875, 78.84495544433594] single-label/f1-score_classwise: [92.04957580566406, 73.32035064697266] data_time: 0.0179 time: 1.2803 +2023/06/05 18:06:45 - mmengine - INFO - Epoch(train) [5][ 100/4111] lr: 6.8582e-05 eta: 4:23:16 time: 0.6371 data_time: 0.0016 memory: 44140 loss: 0.4589 +2023/06/05 18:07:48 - mmengine - INFO - Epoch(train) [5][ 200/4111] lr: 6.8253e-05 eta: 4:22:11 time: 0.6339 data_time: 0.0020 memory: 44140 loss: 0.4647 +2023/06/05 18:08:52 - mmengine - INFO - Epoch(train) [5][ 300/4111] lr: 6.7924e-05 eta: 4:21:05 time: 0.6346 data_time: 0.0016 memory: 44140 loss: 0.4443 +2023/06/05 18:09:55 - mmengine - INFO - Epoch(train) [5][ 400/4111] lr: 6.7595e-05 eta: 4:20:00 time: 0.6344 data_time: 0.0016 memory: 44140 loss: 0.4744 +2023/06/05 18:10:59 - mmengine - INFO - Epoch(train) [5][ 500/4111] lr: 6.7264e-05 eta: 4:18:54 time: 0.6344 data_time: 0.0015 memory: 44140 loss: 0.4733 +2023/06/05 18:11:34 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_stylegan3_lr1e-4_20230605_145625 +2023/06/05 18:12:02 - mmengine - INFO - Epoch(train) [5][ 600/4111] lr: 6.6933e-05 eta: 4:17:49 time: 0.6350 data_time: 0.0018 memory: 44140 loss: 0.4643 +2023/06/05 18:13:06 - mmengine - INFO - Epoch(train) [5][ 700/4111] lr: 6.6601e-05 eta: 4:16:44 time: 0.6338 data_time: 0.0024 memory: 44140 loss: 0.4681 +2023/06/05 18:14:09 - mmengine - INFO - Epoch(train) [5][ 800/4111] lr: 6.6268e-05 eta: 4:15:38 time: 0.6355 data_time: 0.0021 memory: 44140 loss: 0.4910 +2023/06/05 18:15:13 - mmengine - INFO - Epoch(train) [5][ 900/4111] lr: 6.5935e-05 eta: 4:14:33 time: 0.6352 data_time: 0.0015 memory: 44140 loss: 0.4762 +2023/06/05 18:16:16 - mmengine - INFO - Epoch(train) [5][1000/4111] lr: 6.5601e-05 eta: 4:13:28 time: 0.6338 data_time: 0.0018 memory: 44140 loss: 0.4894 +2023/06/05 18:17:20 - mmengine - INFO - Epoch(train) [5][1100/4111] lr: 6.5267e-05 eta: 4:12:22 time: 0.6351 data_time: 0.0021 memory: 44140 loss: 0.4618 +2023/06/05 18:18:23 - mmengine - INFO - Epoch(train) [5][1200/4111] lr: 6.4932e-05 eta: 4:11:17 time: 0.6341 data_time: 0.0016 memory: 44140 loss: 0.4718 +2023/06/05 18:19:27 - mmengine - INFO - Epoch(train) [5][1300/4111] lr: 6.4596e-05 eta: 4:10:12 time: 0.6344 data_time: 0.0015 memory: 44140 loss: 0.4258 +2023/06/05 18:20:30 - mmengine - INFO - Epoch(train) [5][1400/4111] lr: 6.4260e-05 eta: 4:09:06 time: 0.6341 data_time: 0.0016 memory: 44140 loss: 0.4689 +2023/06/05 18:21:34 - mmengine - INFO - Epoch(train) [5][1500/4111] lr: 6.3923e-05 eta: 4:08:01 time: 0.6331 data_time: 0.0020 memory: 44140 loss: 0.4547 +2023/06/05 18:22:09 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_stylegan3_lr1e-4_20230605_145625 +2023/06/05 18:22:37 - mmengine - INFO - Epoch(train) [5][1600/4111] lr: 6.3585e-05 eta: 4:06:56 time: 0.6337 data_time: 0.0014 memory: 44140 loss: 0.4536 +2023/06/05 18:23:40 - mmengine - INFO - Epoch(train) [5][1700/4111] lr: 6.3248e-05 eta: 4:05:51 time: 0.6357 data_time: 0.0017 memory: 44140 loss: 0.4633 +2023/06/05 18:24:44 - mmengine - INFO - Epoch(train) [5][1800/4111] lr: 6.2909e-05 eta: 4:04:45 time: 0.6344 data_time: 0.0015 memory: 44140 loss: 0.4437 +2023/06/05 18:25:47 - mmengine - INFO - Epoch(train) [5][1900/4111] lr: 6.2571e-05 eta: 4:03:40 time: 0.6336 data_time: 0.0021 memory: 44140 loss: 0.4458 +2023/06/05 18:26:51 - mmengine - INFO - Epoch(train) [5][2000/4111] lr: 6.2231e-05 eta: 4:02:35 time: 0.6340 data_time: 0.0016 memory: 44140 loss: 0.4612 +2023/06/05 18:27:54 - mmengine - INFO - Epoch(train) [5][2100/4111] lr: 6.1892e-05 eta: 4:01:30 time: 0.6347 data_time: 0.0015 memory: 44140 loss: 0.4664 +2023/06/05 18:28:58 - mmengine - INFO - Epoch(train) [5][2200/4111] lr: 6.1552e-05 eta: 4:00:25 time: 0.6347 data_time: 0.0015 memory: 44140 loss: 0.4526 +2023/06/05 18:30:01 - mmengine - INFO - Epoch(train) [5][2300/4111] lr: 6.1211e-05 eta: 3:59:20 time: 0.6335 data_time: 0.0017 memory: 44140 loss: 0.4700 +2023/06/05 18:31:05 - mmengine - INFO - Epoch(train) [5][2400/4111] lr: 6.0871e-05 eta: 3:58:15 time: 0.6337 data_time: 0.0022 memory: 44140 loss: 0.4398 +2023/06/05 18:32:08 - mmengine - INFO - Epoch(train) [5][2500/4111] lr: 6.0529e-05 eta: 3:57:09 time: 0.6336 data_time: 0.0020 memory: 44140 loss: 0.4636 +2023/06/05 18:32:44 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_stylegan3_lr1e-4_20230605_145625 +2023/06/05 18:33:12 - mmengine - INFO - Epoch(train) [5][2600/4111] lr: 6.0188e-05 eta: 3:56:04 time: 0.6342 data_time: 0.0015 memory: 44140 loss: 0.4714 +2023/06/05 18:34:15 - mmengine - INFO - Epoch(train) [5][2700/4111] lr: 5.9846e-05 eta: 3:54:59 time: 0.6363 data_time: 0.0015 memory: 44140 loss: 0.4585 +2023/06/05 18:35:19 - mmengine - INFO - Epoch(train) [5][2800/4111] lr: 5.9504e-05 eta: 3:53:54 time: 0.6350 data_time: 0.0025 memory: 44140 loss: 0.4662 +2023/06/05 18:36:48 - mmengine - INFO - Epoch(train) [5][2900/4111] lr: 5.9162e-05 eta: 3:53:19 time: 3.2730 data_time: 0.0017 memory: 44140 loss: 0.4833 +2023/06/05 18:37:52 - mmengine - INFO - Epoch(train) [5][3000/4111] lr: 5.8819e-05 eta: 3:52:14 time: 0.6358 data_time: 0.0017 memory: 44140 loss: 0.4421 +2023/06/05 18:38:55 - mmengine - INFO - Epoch(train) [5][3100/4111] lr: 5.8477e-05 eta: 3:51:09 time: 0.6354 data_time: 0.0019 memory: 44140 loss: 0.4587 +2023/06/05 18:39:59 - mmengine - INFO - Epoch(train) [5][3200/4111] lr: 5.8134e-05 eta: 3:50:03 time: 0.6365 data_time: 0.0019 memory: 44140 loss: 0.4553 +2023/06/05 18:41:03 - mmengine - INFO - Epoch(train) [5][3300/4111] lr: 5.7791e-05 eta: 3:48:58 time: 0.6352 data_time: 0.0016 memory: 44140 loss: 0.4569 +2023/06/05 18:42:06 - mmengine - INFO - Epoch(train) [5][3400/4111] lr: 5.7447e-05 eta: 3:47:53 time: 0.6342 data_time: 0.0015 memory: 44140 loss: 0.4753 +2023/06/05 18:43:10 - mmengine - INFO - Epoch(train) [5][3500/4111] lr: 5.7104e-05 eta: 3:46:48 time: 0.6349 data_time: 0.0015 memory: 44140 loss: 0.4783 +2023/06/05 18:43:45 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_stylegan3_lr1e-4_20230605_145625 +2023/06/05 18:44:13 - mmengine - INFO - Epoch(train) [5][3600/4111] lr: 5.6760e-05 eta: 3:45:43 time: 0.6435 data_time: 0.0020 memory: 44140 loss: 0.4497 +2023/06/05 18:45:17 - mmengine - INFO - Epoch(train) [5][3700/4111] lr: 5.6417e-05 eta: 3:44:38 time: 0.6335 data_time: 0.0022 memory: 44140 loss: 0.4790 +2023/06/05 18:46:20 - mmengine - INFO - Epoch(train) [5][3800/4111] lr: 5.6073e-05 eta: 3:43:33 time: 0.6346 data_time: 0.0014 memory: 44140 loss: 0.4539 +2023/06/05 18:47:24 - mmengine - INFO - Epoch(train) [5][3900/4111] lr: 5.5729e-05 eta: 3:42:28 time: 0.6350 data_time: 0.0016 memory: 44140 loss: 0.4731 +2023/06/05 18:48:27 - mmengine - INFO - Epoch(train) [5][4000/4111] lr: 5.5385e-05 eta: 3:41:22 time: 0.6349 data_time: 0.0020 memory: 44140 loss: 0.4750 +2023/06/05 18:49:30 - mmengine - INFO - Epoch(train) [5][4100/4111] lr: 5.5041e-05 eta: 3:40:17 time: 0.6351 data_time: 0.0037 memory: 44140 loss: 0.4441 +2023/06/05 18:49:37 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_stylegan3_lr1e-4_20230605_145625 +2023/06/05 18:49:37 - mmengine - INFO - Saving checkpoint at 5 epochs +2023/06/05 18:51:59 - mmengine - INFO - Epoch(val) [5][100/119] eta: 0:00:24 time: 1.2956 data_time: 0.0008 memory: 44140 +2023/06/05 18:52:41 - mmengine - INFO - Epoch(val) [5][119/119] accuracy/top1: 88.1866 single-label/precision_classwise: [93.75121307373047, 70.189208984375] single-label/recall_classwise: [91.04854583740234, 77.6434097290039] single-label/f1-score_classwise: [92.38011169433594, 73.72837829589844] data_time: 0.0174 time: 1.2803 +2023/06/05 18:53:48 - mmengine - INFO - Epoch(train) [6][ 100/4111] lr: 5.4660e-05 eta: 3:39:09 time: 0.6340 data_time: 0.0015 memory: 44140 loss: 0.4841 +2023/06/05 18:54:52 - mmengine - INFO - Epoch(train) [6][ 200/4111] lr: 5.4316e-05 eta: 3:38:04 time: 0.6352 data_time: 0.0014 memory: 44140 loss: 0.4501 +2023/06/05 18:55:56 - mmengine - INFO - Epoch(train) [6][ 300/4111] lr: 5.3972e-05 eta: 3:36:59 time: 0.6360 data_time: 0.0014 memory: 44140 loss: 0.4396 +2023/06/05 18:56:59 - mmengine - INFO - Epoch(train) [6][ 400/4111] lr: 5.3628e-05 eta: 3:35:54 time: 0.6340 data_time: 0.0015 memory: 44140 loss: 0.4615 +2023/06/05 18:57:28 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_stylegan3_lr1e-4_20230605_145625 +2023/06/05 18:58:06 - mmengine - INFO - Epoch(train) [6][ 500/4111] lr: 5.3284e-05 eta: 3:34:52 time: 0.6345 data_time: 0.0016 memory: 44140 loss: 0.4510 +2023/06/05 18:59:10 - mmengine - INFO - Epoch(train) [6][ 600/4111] lr: 5.2941e-05 eta: 3:33:47 time: 0.6439 data_time: 0.0016 memory: 44140 loss: 0.4729 +2023/06/05 19:00:18 - mmengine - INFO - Epoch(train) [6][ 700/4111] lr: 5.2597e-05 eta: 3:32:46 time: 0.6338 data_time: 0.0015 memory: 44140 loss: 0.4968 +2023/06/05 19:01:21 - mmengine - INFO - Epoch(train) [6][ 800/4111] lr: 5.2254e-05 eta: 3:31:41 time: 0.6347 data_time: 0.0016 memory: 44140 loss: 0.4382 +2023/06/05 19:02:25 - mmengine - INFO - Epoch(train) [6][ 900/4111] lr: 5.1911e-05 eta: 3:30:36 time: 0.6341 data_time: 0.0017 memory: 44140 loss: 0.4446 +2023/06/05 19:03:28 - mmengine - INFO - Epoch(train) [6][1000/4111] lr: 5.1568e-05 eta: 3:29:31 time: 0.6355 data_time: 0.0020 memory: 44140 loss: 0.4478 +2023/06/05 19:04:33 - mmengine - INFO - Epoch(train) [6][1100/4111] lr: 5.1225e-05 eta: 3:28:28 time: 0.6340 data_time: 0.0015 memory: 44140 loss: 0.4768 +2023/06/05 19:05:37 - mmengine - INFO - Epoch(train) [6][1200/4111] lr: 5.0883e-05 eta: 3:27:23 time: 0.6328 data_time: 0.0017 memory: 44140 loss: 0.4739 +2023/06/05 19:06:40 - mmengine - INFO - Epoch(train) [6][1300/4111] lr: 5.0540e-05 eta: 3:26:18 time: 0.6351 data_time: 0.0016 memory: 44140 loss: 0.4730 +2023/06/05 19:07:44 - mmengine - INFO - Epoch(train) [6][1400/4111] lr: 5.0198e-05 eta: 3:25:13 time: 0.6342 data_time: 0.0020 memory: 44140 loss: 0.4475 +2023/06/05 19:08:12 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_stylegan3_lr1e-4_20230605_145625 +2023/06/05 19:08:47 - mmengine - INFO - Epoch(train) [6][1500/4111] lr: 4.9856e-05 eta: 3:24:08 time: 0.6335 data_time: 0.0019 memory: 44140 loss: 0.4802 +2023/06/05 19:09:51 - mmengine - INFO - Epoch(train) [6][1600/4111] lr: 4.9515e-05 eta: 3:23:02 time: 0.6339 data_time: 0.0013 memory: 44140 loss: 0.4867 +2023/06/05 19:10:54 - mmengine - INFO - Epoch(train) [6][1700/4111] lr: 4.9174e-05 eta: 3:21:58 time: 0.6358 data_time: 0.0015 memory: 44140 loss: 0.4754 +2023/06/05 19:11:58 - mmengine - INFO - Epoch(train) [6][1800/4111] lr: 4.8833e-05 eta: 3:20:53 time: 0.6340 data_time: 0.0016 memory: 44140 loss: 0.4750 +2023/06/05 19:13:01 - mmengine - INFO - Epoch(train) [6][1900/4111] lr: 4.8492e-05 eta: 3:19:48 time: 0.6330 data_time: 0.0020 memory: 44140 loss: 0.4520 +2023/06/05 19:14:11 - mmengine - INFO - Epoch(train) [6][2000/4111] lr: 4.8152e-05 eta: 3:18:48 time: 1.2933 data_time: 0.0017 memory: 44140 loss: 0.4769 +2023/06/05 19:15:15 - mmengine - INFO - Epoch(train) [6][2100/4111] lr: 4.7813e-05 eta: 3:17:43 time: 0.6353 data_time: 0.0018 memory: 44140 loss: 0.4721 +2023/06/05 19:16:18 - mmengine - INFO - Epoch(train) [6][2200/4111] lr: 4.7473e-05 eta: 3:16:38 time: 0.6358 data_time: 0.0015 memory: 44140 loss: 0.4756 +2023/06/05 19:17:22 - mmengine - INFO - Epoch(train) [6][2300/4111] lr: 4.7135e-05 eta: 3:15:33 time: 0.6349 data_time: 0.0017 memory: 44140 loss: 0.4574 +2023/06/05 19:18:25 - mmengine - INFO - Epoch(train) [6][2400/4111] lr: 4.6796e-05 eta: 3:14:29 time: 0.6357 data_time: 0.0024 memory: 44140 loss: 0.4630 +2023/06/05 19:18:54 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_stylegan3_lr1e-4_20230605_145625 +2023/06/05 19:19:29 - mmengine - INFO - Epoch(train) [6][2500/4111] lr: 4.6458e-05 eta: 3:13:24 time: 0.6368 data_time: 0.0023 memory: 44140 loss: 0.4364 +2023/06/05 19:20:33 - mmengine - INFO - Epoch(train) [6][2600/4111] lr: 4.6121e-05 eta: 3:12:19 time: 0.6361 data_time: 0.0020 memory: 44140 loss: 0.4442 +2023/06/05 19:21:36 - mmengine - INFO - Epoch(train) [6][2700/4111] lr: 4.5784e-05 eta: 3:11:14 time: 0.6338 data_time: 0.0015 memory: 44140 loss: 0.4735 +2023/06/05 19:22:40 - mmengine - INFO - Epoch(train) [6][2800/4111] lr: 4.5448e-05 eta: 3:10:09 time: 0.6346 data_time: 0.0020 memory: 44140 loss: 0.4579 +2023/06/05 19:23:43 - mmengine - INFO - Epoch(train) [6][2900/4111] lr: 4.5112e-05 eta: 3:09:05 time: 0.6372 data_time: 0.0017 memory: 44140 loss: 0.4638 +2023/06/05 19:24:47 - mmengine - INFO - Epoch(train) [6][3000/4111] lr: 4.4777e-05 eta: 3:08:00 time: 0.6379 data_time: 0.0016 memory: 44140 loss: 0.4767 +2023/06/05 19:25:51 - mmengine - INFO - Epoch(train) [6][3100/4111] lr: 4.4442e-05 eta: 3:06:55 time: 0.6348 data_time: 0.0023 memory: 44140 loss: 0.4359 +2023/06/05 19:26:54 - mmengine - INFO - Epoch(train) [6][3200/4111] lr: 4.4108e-05 eta: 3:05:50 time: 0.6339 data_time: 0.0018 memory: 44140 loss: 0.4762 +2023/06/05 19:27:57 - mmengine - INFO - Epoch(train) [6][3300/4111] lr: 4.3775e-05 eta: 3:04:46 time: 0.6349 data_time: 0.0016 memory: 44140 loss: 0.4504 +2023/06/05 19:29:01 - mmengine - INFO - Epoch(train) [6][3400/4111] lr: 4.3442e-05 eta: 3:03:41 time: 0.6344 data_time: 0.0020 memory: 44140 loss: 0.4763 +2023/06/05 19:29:30 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_stylegan3_lr1e-4_20230605_145625 +2023/06/05 19:30:04 - mmengine - INFO - Epoch(train) [6][3500/4111] lr: 4.3110e-05 eta: 3:02:36 time: 0.6344 data_time: 0.0015 memory: 44140 loss: 0.4579 +2023/06/05 19:31:08 - mmengine - INFO - Epoch(train) [6][3600/4111] lr: 4.2779e-05 eta: 3:01:31 time: 0.6338 data_time: 0.0019 memory: 44140 loss: 0.4684 +2023/06/05 19:32:33 - mmengine - INFO - Epoch(train) [6][3700/4111] lr: 4.2448e-05 eta: 3:00:42 time: 0.6424 data_time: 0.0021 memory: 44140 loss: 0.4568 +2023/06/05 19:33:37 - mmengine - INFO - Epoch(train) [6][3800/4111] lr: 4.2119e-05 eta: 2:59:37 time: 0.6419 data_time: 0.0016 memory: 44140 loss: 0.4580 +2023/06/05 19:34:42 - mmengine - INFO - Epoch(train) [6][3900/4111] lr: 4.1789e-05 eta: 2:58:33 time: 0.6362 data_time: 0.0015 memory: 44140 loss: 0.4642 +2023/06/05 19:35:45 - mmengine - INFO - Epoch(train) [6][4000/4111] lr: 4.1461e-05 eta: 2:57:28 time: 0.6338 data_time: 0.0019 memory: 44140 loss: 0.4529 +2023/06/05 19:36:49 - mmengine - INFO - Epoch(train) [6][4100/4111] lr: 4.1133e-05 eta: 2:56:23 time: 0.6348 data_time: 0.0034 memory: 44140 loss: 0.4552 +2023/06/05 19:36:55 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_stylegan3_lr1e-4_20230605_145625 +2023/06/05 19:36:55 - mmengine - INFO - Saving checkpoint at 6 epochs +2023/06/05 19:39:16 - mmengine - INFO - Epoch(val) [6][100/119] eta: 0:00:24 time: 1.2820 data_time: 0.0006 memory: 44140 +2023/06/05 19:39:59 - mmengine - INFO - Epoch(val) [6][119/119] accuracy/top1: 88.9016 single-label/precision_classwise: [93.75281524658203, 72.47786712646484] single-label/recall_classwise: [92.02070617675781, 77.41085052490234] single-label/f1-score_classwise: [92.8786849975586, 74.8631820678711] data_time: 0.0180 time: 1.2792 +2023/06/05 19:41:06 - mmengine - INFO - Epoch(train) [7][ 100/4111] lr: 4.0771e-05 eta: 2:55:13 time: 0.6349 data_time: 0.0016 memory: 44140 loss: 0.4697 +2023/06/05 19:42:10 - mmengine - INFO - Epoch(train) [7][ 200/4111] lr: 4.0445e-05 eta: 2:54:09 time: 0.6336 data_time: 0.0017 memory: 44140 loss: 0.4612 +2023/06/05 19:43:13 - mmengine - INFO - Epoch(train) [7][ 300/4111] lr: 4.0120e-05 eta: 2:53:04 time: 0.6338 data_time: 0.0020 memory: 44140 loss: 0.4595 +2023/06/05 19:43:35 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_stylegan3_lr1e-4_20230605_145625 +2023/06/05 19:44:17 - mmengine - INFO - Epoch(train) [7][ 400/4111] lr: 3.9796e-05 eta: 2:51:59 time: 0.6349 data_time: 0.0017 memory: 44140 loss: 0.4820 +2023/06/05 19:45:21 - mmengine - INFO - Epoch(train) [7][ 500/4111] lr: 3.9473e-05 eta: 2:50:54 time: 0.6361 data_time: 0.0017 memory: 44140 loss: 0.4521 +2023/06/05 19:46:24 - mmengine - INFO - Epoch(train) [7][ 600/4111] lr: 3.9150e-05 eta: 2:49:50 time: 0.6361 data_time: 0.0016 memory: 44140 loss: 0.4474 +2023/06/05 19:47:28 - mmengine - INFO - Epoch(train) [7][ 700/4111] lr: 3.8829e-05 eta: 2:48:45 time: 0.6384 data_time: 0.0015 memory: 44140 loss: 0.4652 +2023/06/05 19:48:32 - mmengine - INFO - Epoch(train) [7][ 800/4111] lr: 3.8509e-05 eta: 2:47:40 time: 0.6370 data_time: 0.0016 memory: 44140 loss: 0.4558 +2023/06/05 19:49:35 - mmengine - INFO - Epoch(train) [7][ 900/4111] lr: 3.8189e-05 eta: 2:46:36 time: 0.6361 data_time: 0.0016 memory: 44140 loss: 0.4810 +2023/06/05 19:50:39 - mmengine - INFO - Epoch(train) [7][1000/4111] lr: 3.7871e-05 eta: 2:45:31 time: 0.6340 data_time: 0.0016 memory: 44140 loss: 0.4564 +2023/06/05 19:51:42 - mmengine - INFO - Epoch(train) [7][1100/4111] lr: 3.7553e-05 eta: 2:44:26 time: 0.6338 data_time: 0.0017 memory: 44140 loss: 0.4706 +2023/06/05 19:52:46 - mmengine - INFO - Epoch(train) [7][1200/4111] lr: 3.7237e-05 eta: 2:43:21 time: 0.6334 data_time: 0.0018 memory: 44140 loss: 0.4648 +2023/06/05 19:53:49 - mmengine - INFO - Epoch(train) [7][1300/4111] lr: 3.6921e-05 eta: 2:42:16 time: 0.6359 data_time: 0.0018 memory: 44140 loss: 0.4563 +2023/06/05 19:54:11 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_stylegan3_lr1e-4_20230605_145625 +2023/06/05 19:54:53 - mmengine - INFO - Epoch(train) [7][1400/4111] lr: 3.6607e-05 eta: 2:41:12 time: 0.6336 data_time: 0.0015 memory: 44140 loss: 0.4827 +2023/06/05 19:55:57 - mmengine - INFO - Epoch(train) [7][1500/4111] lr: 3.6294e-05 eta: 2:40:07 time: 0.6375 data_time: 0.0018 memory: 44140 loss: 0.4337 +2023/06/05 19:57:00 - mmengine - INFO - Epoch(train) [7][1600/4111] lr: 3.5981e-05 eta: 2:39:02 time: 0.6347 data_time: 0.0015 memory: 44140 loss: 0.4628 +2023/06/05 19:58:04 - mmengine - INFO - Epoch(train) [7][1700/4111] lr: 3.5670e-05 eta: 2:37:58 time: 0.6342 data_time: 0.0015 memory: 44140 loss: 0.4708 +2023/06/05 19:59:07 - mmengine - INFO - Epoch(train) [7][1800/4111] lr: 3.5360e-05 eta: 2:36:53 time: 0.6349 data_time: 0.0016 memory: 44140 loss: 0.4480 +2023/06/05 20:00:11 - mmengine - INFO - Epoch(train) [7][1900/4111] lr: 3.5051e-05 eta: 2:35:48 time: 0.6340 data_time: 0.0015 memory: 44140 loss: 0.4461 +2023/06/05 20:01:15 - mmengine - INFO - Epoch(train) [7][2000/4111] lr: 3.4744e-05 eta: 2:34:44 time: 0.6366 data_time: 0.0023 memory: 44140 loss: 0.4630 +2023/06/05 20:02:18 - mmengine - INFO - Epoch(train) [7][2100/4111] lr: 3.4437e-05 eta: 2:33:39 time: 0.6364 data_time: 0.0016 memory: 44140 loss: 0.4408 +2023/06/05 20:03:22 - mmengine - INFO - Epoch(train) [7][2200/4111] lr: 3.4132e-05 eta: 2:32:34 time: 0.6370 data_time: 0.0015 memory: 44140 loss: 0.4819 +2023/06/05 20:04:25 - mmengine - INFO - Epoch(train) [7][2300/4111] lr: 3.3828e-05 eta: 2:31:30 time: 0.6383 data_time: 0.0017 memory: 44140 loss: 0.4886 +2023/06/05 20:04:47 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_stylegan3_lr1e-4_20230605_145625 +2023/06/05 20:05:29 - mmengine - INFO - Epoch(train) [7][2400/4111] lr: 3.3525e-05 eta: 2:30:25 time: 0.6335 data_time: 0.0015 memory: 44140 loss: 0.4552 +2023/06/05 20:06:32 - mmengine - INFO - Epoch(train) [7][2500/4111] lr: 3.3224e-05 eta: 2:29:20 time: 0.6357 data_time: 0.0015 memory: 44140 loss: 0.4468 +2023/06/05 20:07:36 - mmengine - INFO - Epoch(train) [7][2600/4111] lr: 3.2923e-05 eta: 2:28:16 time: 0.6373 data_time: 0.0015 memory: 44140 loss: 0.4654 +2023/06/05 20:08:40 - mmengine - INFO - Epoch(train) [7][2700/4111] lr: 3.2624e-05 eta: 2:27:11 time: 0.6342 data_time: 0.0024 memory: 44140 loss: 0.4521 +2023/06/05 20:09:43 - mmengine - INFO - Epoch(train) [7][2800/4111] lr: 3.2326e-05 eta: 2:26:07 time: 0.6367 data_time: 0.0016 memory: 44140 loss: 0.4758 +2023/06/05 20:10:47 - mmengine - INFO - Epoch(train) [7][2900/4111] lr: 3.2030e-05 eta: 2:25:02 time: 0.6339 data_time: 0.0014 memory: 44140 loss: 0.4543 +2023/06/05 20:11:50 - mmengine - INFO - Epoch(train) [7][3000/4111] lr: 3.1735e-05 eta: 2:23:57 time: 0.6339 data_time: 0.0016 memory: 44140 loss: 0.4432 +2023/06/05 20:12:54 - mmengine - INFO - Epoch(train) [7][3100/4111] lr: 3.1441e-05 eta: 2:22:53 time: 0.6348 data_time: 0.0016 memory: 44140 loss: 0.4632 +2023/06/05 20:13:57 - mmengine - INFO - Epoch(train) [7][3200/4111] lr: 3.1149e-05 eta: 2:21:48 time: 0.6365 data_time: 0.0016 memory: 44140 loss: 0.4473 +2023/06/05 20:15:01 - mmengine - INFO - Epoch(train) [7][3300/4111] lr: 3.0858e-05 eta: 2:20:44 time: 0.6338 data_time: 0.0015 memory: 44140 loss: 0.4641 +2023/06/05 20:15:22 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_stylegan3_lr1e-4_20230605_145625 +2023/06/05 20:16:04 - mmengine - INFO - Epoch(train) [7][3400/4111] lr: 3.0569e-05 eta: 2:19:39 time: 0.6346 data_time: 0.0017 memory: 44140 loss: 0.4819 +2023/06/05 20:17:08 - mmengine - INFO - Epoch(train) [7][3500/4111] lr: 3.0281e-05 eta: 2:18:34 time: 0.6338 data_time: 0.0016 memory: 44140 loss: 0.4812 +2023/06/05 20:18:11 - mmengine - INFO - Epoch(train) [7][3600/4111] lr: 2.9994e-05 eta: 2:17:30 time: 0.6373 data_time: 0.0014 memory: 44140 loss: 0.4763 +2023/06/05 20:19:15 - mmengine - INFO - Epoch(train) [7][3700/4111] lr: 2.9709e-05 eta: 2:16:25 time: 0.6369 data_time: 0.0017 memory: 44140 loss: 0.4820 +2023/06/05 20:20:19 - mmengine - INFO - Epoch(train) [7][3800/4111] lr: 2.9425e-05 eta: 2:15:21 time: 0.6357 data_time: 0.0016 memory: 44140 loss: 0.4442 +2023/06/05 20:21:22 - mmengine - INFO - Epoch(train) [7][3900/4111] lr: 2.9143e-05 eta: 2:14:16 time: 0.6350 data_time: 0.0019 memory: 44140 loss: 0.4637 +2023/06/05 20:22:26 - mmengine - INFO - Epoch(train) [7][4000/4111] lr: 2.8862e-05 eta: 2:13:12 time: 0.6347 data_time: 0.0016 memory: 44140 loss: 0.4616 +2023/06/05 20:23:29 - mmengine - INFO - Epoch(train) [7][4100/4111] lr: 2.8583e-05 eta: 2:12:07 time: 0.6384 data_time: 0.0029 memory: 44140 loss: 0.4638 +2023/06/05 20:23:36 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_stylegan3_lr1e-4_20230605_145625 +2023/06/05 20:23:36 - mmengine - INFO - Saving checkpoint at 7 epochs +2023/06/05 20:25:57 - mmengine - INFO - Epoch(val) [7][100/119] eta: 0:00:24 time: 1.2816 data_time: 0.0009 memory: 44140 +2023/06/05 20:26:39 - mmengine - INFO - Epoch(val) [7][119/119] accuracy/top1: 89.1482 single-label/precision_classwise: [93.697998046875, 73.40763092041016] single-label/recall_classwise: [92.41841125488281, 77.10077667236328] single-label/f1-score_classwise: [93.05380249023438, 75.20889282226562] data_time: 0.0179 time: 1.2793 +2023/06/05 20:27:50 - mmengine - INFO - Epoch(train) [8][ 100/4111] lr: 2.8275e-05 eta: 2:10:59 time: 0.6343 data_time: 0.0014 memory: 44140 loss: 0.4637 +2023/06/05 20:28:53 - mmengine - INFO - Epoch(train) [8][ 200/4111] lr: 2.7999e-05 eta: 2:09:54 time: 0.6339 data_time: 0.0019 memory: 44140 loss: 0.4610 +2023/06/05 20:29:08 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_stylegan3_lr1e-4_20230605_145625 +2023/06/05 20:29:57 - mmengine - INFO - Epoch(train) [8][ 300/4111] lr: 2.7725e-05 eta: 2:08:49 time: 0.6352 data_time: 0.0015 memory: 44140 loss: 0.4674 +2023/06/05 20:31:00 - mmengine - INFO - Epoch(train) [8][ 400/4111] lr: 2.7452e-05 eta: 2:07:45 time: 0.6331 data_time: 0.0016 memory: 44140 loss: 0.4786 +2023/06/05 20:32:03 - mmengine - INFO - Epoch(train) [8][ 500/4111] lr: 2.7181e-05 eta: 2:06:40 time: 0.6341 data_time: 0.0016 memory: 44140 loss: 0.4577 +2023/06/05 20:33:07 - mmengine - INFO - Epoch(train) [8][ 600/4111] lr: 2.6911e-05 eta: 2:05:36 time: 0.6377 data_time: 0.0014 memory: 44140 loss: 0.5010 +2023/06/05 20:34:10 - mmengine - INFO - Epoch(train) [8][ 700/4111] lr: 2.6644e-05 eta: 2:04:31 time: 0.6344 data_time: 0.0022 memory: 44140 loss: 0.4641 +2023/06/05 20:35:14 - mmengine - INFO - Epoch(train) [8][ 800/4111] lr: 2.6377e-05 eta: 2:03:27 time: 0.6332 data_time: 0.0021 memory: 44140 loss: 0.4771 +2023/06/05 20:36:17 - mmengine - INFO - Epoch(train) [8][ 900/4111] lr: 2.6113e-05 eta: 2:02:22 time: 0.6331 data_time: 0.0025 memory: 44140 loss: 0.4558 +2023/06/05 20:37:24 - mmengine - INFO - Epoch(train) [8][1000/4111] lr: 2.5850e-05 eta: 2:01:19 time: 0.6368 data_time: 0.0016 memory: 44140 loss: 0.4631 +2023/06/05 20:38:27 - mmengine - INFO - Epoch(train) [8][1100/4111] lr: 2.5589e-05 eta: 2:00:14 time: 0.6337 data_time: 0.0019 memory: 44140 loss: 0.4599 +2023/06/05 20:39:31 - mmengine - INFO - Epoch(train) [8][1200/4111] lr: 2.5330e-05 eta: 1:59:10 time: 0.6336 data_time: 0.0022 memory: 44140 loss: 0.4491 +2023/06/05 20:39:45 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_stylegan3_lr1e-4_20230605_145625 +2023/06/05 20:40:34 - mmengine - INFO - Epoch(train) [8][1300/4111] lr: 2.5072e-05 eta: 1:58:05 time: 0.6342 data_time: 0.0016 memory: 44140 loss: 0.4482 +2023/06/05 20:41:38 - mmengine - INFO - Epoch(train) [8][1400/4111] lr: 2.4816e-05 eta: 1:57:01 time: 0.6327 data_time: 0.0015 memory: 44140 loss: 0.4638 +2023/06/05 20:42:41 - mmengine - INFO - Epoch(train) [8][1500/4111] lr: 2.4562e-05 eta: 1:55:56 time: 0.6346 data_time: 0.0016 memory: 44140 loss: 0.4553 +2023/06/05 20:43:44 - mmengine - INFO - Epoch(train) [8][1600/4111] lr: 2.4309e-05 eta: 1:54:52 time: 0.6331 data_time: 0.0017 memory: 44140 loss: 0.4282 +2023/06/05 20:44:48 - mmengine - INFO - Epoch(train) [8][1700/4111] lr: 2.4059e-05 eta: 1:53:47 time: 0.6337 data_time: 0.0019 memory: 44140 loss: 0.4766 +2023/06/05 20:45:51 - mmengine - INFO - Epoch(train) [8][1800/4111] lr: 2.3810e-05 eta: 1:52:43 time: 0.6341 data_time: 0.0016 memory: 44140 loss: 0.4766 +2023/06/05 20:46:55 - mmengine - INFO - Epoch(train) [8][1900/4111] lr: 2.3563e-05 eta: 1:51:38 time: 0.6353 data_time: 0.0017 memory: 44140 loss: 0.4585 +2023/06/05 20:47:58 - mmengine - INFO - Epoch(train) [8][2000/4111] lr: 2.3318e-05 eta: 1:50:34 time: 0.6356 data_time: 0.0015 memory: 44140 loss: 0.4651 +2023/06/05 20:49:02 - mmengine - INFO - Epoch(train) [8][2100/4111] lr: 2.3075e-05 eta: 1:49:29 time: 0.6331 data_time: 0.0028 memory: 44140 loss: 0.4758 +2023/06/05 20:50:05 - mmengine - INFO - Epoch(train) [8][2200/4111] lr: 2.2833e-05 eta: 1:48:25 time: 0.6363 data_time: 0.0024 memory: 44140 loss: 0.4503 +2023/06/05 20:50:20 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_stylegan3_lr1e-4_20230605_145625 +2023/06/05 20:51:09 - mmengine - INFO - Epoch(train) [8][2300/4111] lr: 2.2594e-05 eta: 1:47:21 time: 0.6360 data_time: 0.0022 memory: 44140 loss: 0.4671 +2023/06/05 20:52:21 - mmengine - INFO - Epoch(train) [8][2400/4111] lr: 2.2356e-05 eta: 1:46:19 time: 0.6359 data_time: 0.0016 memory: 44140 loss: 0.4379 +2023/06/05 20:53:25 - mmengine - INFO - Epoch(train) [8][2500/4111] lr: 2.2120e-05 eta: 1:45:15 time: 0.6332 data_time: 0.0019 memory: 44140 loss: 0.4569 +2023/06/05 20:54:28 - mmengine - INFO - Epoch(train) [8][2600/4111] lr: 2.1886e-05 eta: 1:44:10 time: 0.6347 data_time: 0.0018 memory: 44140 loss: 0.4609 +2023/06/05 20:55:31 - mmengine - INFO - Epoch(train) [8][2700/4111] lr: 2.1655e-05 eta: 1:43:06 time: 0.6342 data_time: 0.0015 memory: 44140 loss: 0.4707 +2023/06/05 20:56:35 - mmengine - INFO - Epoch(train) [8][2800/4111] lr: 2.1425e-05 eta: 1:42:01 time: 0.6375 data_time: 0.0015 memory: 44140 loss: 0.4530 +2023/06/05 20:57:38 - mmengine - INFO - Epoch(train) [8][2900/4111] lr: 2.1197e-05 eta: 1:40:57 time: 0.6349 data_time: 0.0017 memory: 44140 loss: 0.4490 +2023/06/05 20:58:42 - mmengine - INFO - Epoch(train) [8][3000/4111] lr: 2.0971e-05 eta: 1:39:52 time: 0.6335 data_time: 0.0019 memory: 44140 loss: 0.4901 +2023/06/05 20:59:45 - mmengine - INFO - Epoch(train) [8][3100/4111] lr: 2.0747e-05 eta: 1:38:48 time: 0.6334 data_time: 0.0014 memory: 44140 loss: 0.4664 +2023/06/05 21:00:49 - mmengine - INFO - Epoch(train) [8][3200/4111] lr: 2.0525e-05 eta: 1:37:43 time: 0.6335 data_time: 0.0017 memory: 44140 loss: 0.4452 +2023/06/05 21:01:03 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_stylegan3_lr1e-4_20230605_145625 +2023/06/05 21:01:52 - mmengine - INFO - Epoch(train) [8][3300/4111] lr: 2.0305e-05 eta: 1:36:39 time: 0.6342 data_time: 0.0016 memory: 44140 loss: 0.4638 +2023/06/05 21:02:56 - mmengine - INFO - Epoch(train) [8][3400/4111] lr: 2.0087e-05 eta: 1:35:35 time: 0.6342 data_time: 0.0015 memory: 44140 loss: 0.4697 +2023/06/05 21:03:59 - mmengine - INFO - Epoch(train) [8][3500/4111] lr: 1.9871e-05 eta: 1:34:30 time: 0.6344 data_time: 0.0017 memory: 44140 loss: 0.4618 +2023/06/05 21:05:03 - mmengine - INFO - Epoch(train) [8][3600/4111] lr: 1.9657e-05 eta: 1:33:26 time: 0.6343 data_time: 0.0016 memory: 44140 loss: 0.4942 +2023/06/05 21:06:06 - mmengine - INFO - Epoch(train) [8][3700/4111] lr: 1.9445e-05 eta: 1:32:21 time: 0.6342 data_time: 0.0017 memory: 44140 loss: 0.4498 +2023/06/05 21:07:09 - mmengine - INFO - Epoch(train) [8][3800/4111] lr: 1.9235e-05 eta: 1:31:17 time: 0.6333 data_time: 0.0023 memory: 44140 loss: 0.4614 +2023/06/05 21:08:13 - mmengine - INFO - Epoch(train) [8][3900/4111] lr: 1.9028e-05 eta: 1:30:13 time: 0.6401 data_time: 0.0022 memory: 44140 loss: 0.4647 +2023/06/05 21:09:16 - mmengine - INFO - Epoch(train) [8][4000/4111] lr: 1.8822e-05 eta: 1:29:08 time: 0.6349 data_time: 0.0016 memory: 44140 loss: 0.4693 +2023/06/05 21:10:20 - mmengine - INFO - Epoch(train) [8][4100/4111] lr: 1.8619e-05 eta: 1:28:04 time: 0.6349 data_time: 0.0031 memory: 44140 loss: 0.4744 +2023/06/05 21:10:27 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_stylegan3_lr1e-4_20230605_145625 +2023/06/05 21:10:27 - mmengine - INFO - Saving checkpoint at 8 epochs +2023/06/05 21:12:48 - mmengine - INFO - Epoch(val) [8][100/119] eta: 0:00:24 time: 1.2812 data_time: 0.0005 memory: 44140 +2023/06/05 21:13:31 - mmengine - INFO - Epoch(val) [8][119/119] accuracy/top1: 89.2044 single-label/precision_classwise: [93.40461730957031, 74.16812133789062] single-label/recall_classwise: [92.8287353515625, 75.85271453857422] single-label/f1-score_classwise: [93.11579132080078, 75.00096130371094] data_time: 0.0173 time: 1.2791 +2023/06/05 21:14:38 - mmengine - INFO - Epoch(train) [9][ 100/4111] lr: 1.8395e-05 eta: 1:26:53 time: 0.6335 data_time: 0.0015 memory: 44140 loss: 0.4579 +2023/06/05 21:14:45 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_stylegan3_lr1e-4_20230605_145625 +2023/06/05 21:15:41 - mmengine - INFO - Epoch(train) [9][ 200/4111] lr: 1.8196e-05 eta: 1:25:49 time: 0.6338 data_time: 0.0019 memory: 44140 loss: 0.4705 +2023/06/05 21:16:45 - mmengine - INFO - Epoch(train) [9][ 300/4111] lr: 1.7999e-05 eta: 1:24:45 time: 0.6357 data_time: 0.0015 memory: 44140 loss: 0.4662 +2023/06/05 21:17:48 - mmengine - INFO - Epoch(train) [9][ 400/4111] lr: 1.7805e-05 eta: 1:23:40 time: 0.6382 data_time: 0.0015 memory: 44140 loss: 0.4730 +2023/06/05 21:18:52 - mmengine - INFO - Epoch(train) [9][ 500/4111] lr: 1.7612e-05 eta: 1:22:36 time: 0.6361 data_time: 0.0015 memory: 44140 loss: 0.4685 +2023/06/05 21:19:56 - mmengine - INFO - Epoch(train) [9][ 600/4111] lr: 1.7422e-05 eta: 1:21:32 time: 0.6350 data_time: 0.0014 memory: 44140 loss: 0.4582 +2023/06/05 21:21:25 - mmengine - INFO - Epoch(train) [9][ 700/4111] lr: 1.7234e-05 eta: 1:20:33 time: 0.6354 data_time: 0.0025 memory: 44140 loss: 0.4637 +2023/06/05 21:22:29 - mmengine - INFO - Epoch(train) [9][ 800/4111] lr: 1.7048e-05 eta: 1:19:29 time: 0.6345 data_time: 0.0015 memory: 44140 loss: 0.4439 +2023/06/05 21:23:33 - mmengine - INFO - Epoch(train) [9][ 900/4111] lr: 1.6864e-05 eta: 1:18:24 time: 0.6378 data_time: 0.0015 memory: 44140 loss: 0.4598 +2023/06/05 21:24:36 - mmengine - INFO - Epoch(train) [9][1000/4111] lr: 1.6683e-05 eta: 1:17:20 time: 0.6352 data_time: 0.0017 memory: 44140 loss: 0.4574 +2023/06/05 21:25:40 - mmengine - INFO - Epoch(train) [9][1100/4111] lr: 1.6504e-05 eta: 1:16:16 time: 0.6343 data_time: 0.0016 memory: 44140 loss: 0.4601 +2023/06/05 21:25:47 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_stylegan3_lr1e-4_20230605_145625 +2023/06/05 21:26:43 - mmengine - INFO - Epoch(train) [9][1200/4111] lr: 1.6327e-05 eta: 1:15:11 time: 0.6351 data_time: 0.0014 memory: 44140 loss: 0.4575 +2023/06/05 21:27:47 - mmengine - INFO - Epoch(train) [9][1300/4111] lr: 1.6152e-05 eta: 1:14:07 time: 0.6353 data_time: 0.0016 memory: 44140 loss: 0.4719 +2023/06/05 21:28:50 - mmengine - INFO - Epoch(train) [9][1400/4111] lr: 1.5980e-05 eta: 1:13:02 time: 0.6334 data_time: 0.0016 memory: 44140 loss: 0.4594 +2023/06/05 21:29:54 - mmengine - INFO - Epoch(train) [9][1500/4111] lr: 1.5810e-05 eta: 1:11:58 time: 0.6336 data_time: 0.0015 memory: 44140 loss: 0.4614 +2023/06/05 21:30:57 - mmengine - INFO - Epoch(train) [9][1600/4111] lr: 1.5642e-05 eta: 1:10:54 time: 0.6345 data_time: 0.0015 memory: 44140 loss: 0.4630 +2023/06/05 21:32:01 - mmengine - INFO - Epoch(train) [9][1700/4111] lr: 1.5476e-05 eta: 1:09:49 time: 0.6345 data_time: 0.0016 memory: 44140 loss: 0.4735 +2023/06/05 21:33:04 - mmengine - INFO - Epoch(train) [9][1800/4111] lr: 1.5313e-05 eta: 1:08:45 time: 0.6344 data_time: 0.0016 memory: 44140 loss: 0.4546 +2023/06/05 21:34:08 - mmengine - INFO - Epoch(train) [9][1900/4111] lr: 1.5152e-05 eta: 1:07:40 time: 0.6345 data_time: 0.0016 memory: 44140 loss: 0.4379 +2023/06/05 21:35:12 - mmengine - INFO - Epoch(train) [9][2000/4111] lr: 1.4993e-05 eta: 1:06:36 time: 0.6344 data_time: 0.0015 memory: 44140 loss: 0.4720 +2023/06/05 21:36:15 - mmengine - INFO - Epoch(train) [9][2100/4111] lr: 1.4837e-05 eta: 1:05:32 time: 0.6357 data_time: 0.0015 memory: 44140 loss: 0.4597 +2023/06/05 21:36:23 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_stylegan3_lr1e-4_20230605_145625 +2023/06/05 21:37:19 - mmengine - INFO - Epoch(train) [9][2200/4111] lr: 1.4683e-05 eta: 1:04:27 time: 0.6355 data_time: 0.0016 memory: 44140 loss: 0.4681 +2023/06/05 21:38:22 - mmengine - INFO - Epoch(train) [9][2300/4111] lr: 1.4532e-05 eta: 1:03:23 time: 0.6344 data_time: 0.0015 memory: 44140 loss: 0.4502 +2023/06/05 21:39:26 - mmengine - INFO - Epoch(train) [9][2400/4111] lr: 1.4382e-05 eta: 1:02:19 time: 0.6335 data_time: 0.0015 memory: 44140 loss: 0.4791 +2023/06/05 21:40:29 - mmengine - INFO - Epoch(train) [9][2500/4111] lr: 1.4235e-05 eta: 1:01:14 time: 0.6337 data_time: 0.0015 memory: 44140 loss: 0.4548 +2023/06/05 21:41:33 - mmengine - INFO - Epoch(train) [9][2600/4111] lr: 1.4091e-05 eta: 1:00:10 time: 0.6353 data_time: 0.0016 memory: 44140 loss: 0.4755 +2023/06/05 21:42:36 - mmengine - INFO - Epoch(train) [9][2700/4111] lr: 1.3949e-05 eta: 0:59:06 time: 0.6348 data_time: 0.0014 memory: 44140 loss: 0.4671 +2023/06/05 21:43:40 - mmengine - INFO - Epoch(train) [9][2800/4111] lr: 1.3809e-05 eta: 0:58:01 time: 0.6343 data_time: 0.0016 memory: 44140 loss: 0.4549 +2023/06/05 21:44:43 - mmengine - INFO - Epoch(train) [9][2900/4111] lr: 1.3672e-05 eta: 0:56:57 time: 0.6360 data_time: 0.0024 memory: 44140 loss: 0.4772 +2023/06/05 21:45:47 - mmengine - INFO - Epoch(train) [9][3000/4111] lr: 1.3537e-05 eta: 0:55:53 time: 0.6336 data_time: 0.0016 memory: 44140 loss: 0.4681 +2023/06/05 21:46:56 - mmengine - INFO - Epoch(train) [9][3100/4111] lr: 1.3405e-05 eta: 0:54:49 time: 0.6359 data_time: 0.0016 memory: 44140 loss: 0.4779 +2023/06/05 21:47:04 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_stylegan3_lr1e-4_20230605_145625 +2023/06/05 21:48:00 - mmengine - INFO - Epoch(train) [9][3200/4111] lr: 1.3275e-05 eta: 0:53:45 time: 0.6339 data_time: 0.0015 memory: 44140 loss: 0.4315 +2023/06/05 21:49:04 - mmengine - INFO - Epoch(train) [9][3300/4111] lr: 1.3147e-05 eta: 0:52:41 time: 0.6344 data_time: 0.0017 memory: 44140 loss: 0.4663 +2023/06/05 21:50:07 - mmengine - INFO - Epoch(train) [9][3400/4111] lr: 1.3022e-05 eta: 0:51:36 time: 0.6352 data_time: 0.0015 memory: 44140 loss: 0.4640 +2023/06/05 21:51:10 - mmengine - INFO - Epoch(train) [9][3500/4111] lr: 1.2899e-05 eta: 0:50:32 time: 0.6343 data_time: 0.0015 memory: 44140 loss: 0.4838 +2023/06/05 21:52:14 - mmengine - INFO - Epoch(train) [9][3600/4111] lr: 1.2779e-05 eta: 0:49:28 time: 0.6342 data_time: 0.0016 memory: 44140 loss: 0.4731 +2023/06/05 21:53:18 - mmengine - INFO - Epoch(train) [9][3700/4111] lr: 1.2661e-05 eta: 0:48:23 time: 0.6362 data_time: 0.0017 memory: 44140 loss: 0.4541 +2023/06/05 21:54:26 - mmengine - INFO - Epoch(train) [9][3800/4111] lr: 1.2546e-05 eta: 0:47:20 time: 0.6346 data_time: 0.0019 memory: 44140 loss: 0.4696 +2023/06/05 21:55:30 - mmengine - INFO - Epoch(train) [9][3900/4111] lr: 1.2433e-05 eta: 0:46:15 time: 0.6349 data_time: 0.0015 memory: 44140 loss: 0.4489 +2023/06/05 21:56:33 - mmengine - INFO - Epoch(train) [9][4000/4111] lr: 1.2323e-05 eta: 0:45:11 time: 0.6346 data_time: 0.0015 memory: 44140 loss: 0.4627 +2023/06/05 21:57:37 - mmengine - INFO - Epoch(train) [9][4100/4111] lr: 1.2215e-05 eta: 0:44:07 time: 0.6346 data_time: 0.0030 memory: 44140 loss: 0.4508 +2023/06/05 21:57:44 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_stylegan3_lr1e-4_20230605_145625 +2023/06/05 21:57:44 - mmengine - INFO - Saving checkpoint at 9 epochs +2023/06/05 22:00:04 - mmengine - INFO - Epoch(val) [9][100/119] eta: 0:00:24 time: 1.2820 data_time: 0.0005 memory: 44140 +2023/06/05 22:00:47 - mmengine - INFO - Epoch(val) [9][119/119] accuracy/top1: 89.2938 single-label/precision_classwise: [93.42317199707031, 74.45061492919922] single-label/recall_classwise: [92.92974090576172, 75.89922332763672] single-label/f1-score_classwise: [93.1758041381836, 75.16793823242188] data_time: 0.0179 time: 1.2799 +2023/06/05 22:00:52 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_stylegan3_lr1e-4_20230605_145625 +2023/06/05 22:01:54 - mmengine - INFO - Epoch(train) [10][ 100/4111] lr: 1.2098e-05 eta: 0:42:56 time: 0.6364 data_time: 0.0015 memory: 44140 loss: 0.4583 +2023/06/05 22:02:58 - mmengine - INFO - Epoch(train) [10][ 200/4111] lr: 1.1996e-05 eta: 0:41:52 time: 0.6366 data_time: 0.0016 memory: 44140 loss: 0.4593 +2023/06/05 22:04:02 - mmengine - INFO - Epoch(train) [10][ 300/4111] lr: 1.1896e-05 eta: 0:40:47 time: 0.6373 data_time: 0.0015 memory: 44140 loss: 0.4432 +2023/06/05 22:05:05 - mmengine - INFO - Epoch(train) [10][ 400/4111] lr: 1.1798e-05 eta: 0:39:43 time: 0.6338 data_time: 0.0019 memory: 44140 loss: 0.4560 +2023/06/05 22:06:09 - mmengine - INFO - Epoch(train) [10][ 500/4111] lr: 1.1703e-05 eta: 0:38:39 time: 0.6351 data_time: 0.0014 memory: 44140 loss: 0.4433 +2023/06/05 22:07:12 - mmengine - INFO - Epoch(train) [10][ 600/4111] lr: 1.1611e-05 eta: 0:37:34 time: 0.6349 data_time: 0.0014 memory: 44140 loss: 0.4576 +2023/06/05 22:08:16 - mmengine - INFO - Epoch(train) [10][ 700/4111] lr: 1.1521e-05 eta: 0:36:30 time: 0.6339 data_time: 0.0016 memory: 44140 loss: 0.4530 +2023/06/05 22:09:19 - mmengine - INFO - Epoch(train) [10][ 800/4111] lr: 1.1434e-05 eta: 0:35:26 time: 0.6333 data_time: 0.0023 memory: 44140 loss: 0.4676 +2023/06/05 22:10:23 - mmengine - INFO - Epoch(train) [10][ 900/4111] lr: 1.1349e-05 eta: 0:34:21 time: 0.6339 data_time: 0.0016 memory: 44140 loss: 0.4379 +2023/06/05 22:11:26 - mmengine - INFO - Epoch(train) [10][1000/4111] lr: 1.1267e-05 eta: 0:33:17 time: 0.6358 data_time: 0.0017 memory: 44140 loss: 0.4574 +2023/06/05 22:11:27 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_stylegan3_lr1e-4_20230605_145625 +2023/06/05 22:12:30 - mmengine - INFO - Epoch(train) [10][1100/4111] lr: 1.1187e-05 eta: 0:32:13 time: 0.6345 data_time: 0.0023 memory: 44140 loss: 0.4866 +2023/06/05 22:13:33 - mmengine - INFO - Epoch(train) [10][1200/4111] lr: 1.1110e-05 eta: 0:31:09 time: 0.6335 data_time: 0.0016 memory: 44140 loss: 0.4816 +2023/06/05 22:14:37 - mmengine - INFO - Epoch(train) [10][1300/4111] lr: 1.1035e-05 eta: 0:30:04 time: 0.6328 data_time: 0.0019 memory: 44140 loss: 0.4572 +2023/06/05 22:15:40 - mmengine - INFO - Epoch(train) [10][1400/4111] lr: 1.0963e-05 eta: 0:29:00 time: 0.6344 data_time: 0.0015 memory: 44140 loss: 0.4768 +2023/06/05 22:16:44 - mmengine - INFO - Epoch(train) [10][1500/4111] lr: 1.0893e-05 eta: 0:27:56 time: 0.6342 data_time: 0.0016 memory: 44140 loss: 0.4904 +2023/06/05 22:17:47 - mmengine - INFO - Epoch(train) [10][1600/4111] lr: 1.0827e-05 eta: 0:26:52 time: 0.6397 data_time: 0.0022 memory: 44140 loss: 0.4628 +2023/06/05 22:18:51 - mmengine - INFO - Epoch(train) [10][1700/4111] lr: 1.0762e-05 eta: 0:25:47 time: 0.6332 data_time: 0.0017 memory: 44140 loss: 0.4267 +2023/06/05 22:19:54 - mmengine - INFO - Epoch(train) [10][1800/4111] lr: 1.0701e-05 eta: 0:24:43 time: 0.6334 data_time: 0.0019 memory: 44140 loss: 0.4881 +2023/06/05 22:20:58 - mmengine - INFO - Epoch(train) [10][1900/4111] lr: 1.0641e-05 eta: 0:23:39 time: 0.6340 data_time: 0.0017 memory: 44140 loss: 0.4750 +2023/06/05 22:22:01 - mmengine - INFO - Epoch(train) [10][2000/4111] lr: 1.0585e-05 eta: 0:22:35 time: 0.6364 data_time: 0.0015 memory: 44140 loss: 0.5053 +2023/06/05 22:22:02 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_stylegan3_lr1e-4_20230605_145625 +2023/06/05 22:23:05 - mmengine - INFO - Epoch(train) [10][2100/4111] lr: 1.0531e-05 eta: 0:21:30 time: 0.6340 data_time: 0.0015 memory: 44140 loss: 0.4437 +2023/06/05 22:24:08 - mmengine - INFO - Epoch(train) [10][2200/4111] lr: 1.0480e-05 eta: 0:20:26 time: 0.6327 data_time: 0.0017 memory: 44140 loss: 0.4714 +2023/06/05 22:25:12 - mmengine - INFO - Epoch(train) [10][2300/4111] lr: 1.0431e-05 eta: 0:19:22 time: 0.6349 data_time: 0.0018 memory: 44140 loss: 0.4750 +2023/06/05 22:26:15 - mmengine - INFO - Epoch(train) [10][2400/4111] lr: 1.0385e-05 eta: 0:18:18 time: 0.6353 data_time: 0.0016 memory: 44140 loss: 0.4722 +2023/06/05 22:27:19 - mmengine - INFO - Epoch(train) [10][2500/4111] lr: 1.0341e-05 eta: 0:17:14 time: 0.6341 data_time: 0.0015 memory: 44140 loss: 0.4292 +2023/06/05 22:28:23 - mmengine - INFO - Epoch(train) [10][2600/4111] lr: 1.0300e-05 eta: 0:16:09 time: 0.6352 data_time: 0.0016 memory: 44140 loss: 0.4675 +2023/06/05 22:29:26 - mmengine - INFO - Epoch(train) [10][2700/4111] lr: 1.0262e-05 eta: 0:15:05 time: 0.6353 data_time: 0.0015 memory: 44140 loss: 0.4756 +2023/06/05 22:30:30 - mmengine - INFO - Epoch(train) [10][2800/4111] lr: 1.0226e-05 eta: 0:14:01 time: 0.6369 data_time: 0.0019 memory: 44140 loss: 0.4694 +2023/06/05 22:31:34 - mmengine - INFO - Epoch(train) [10][2900/4111] lr: 1.0193e-05 eta: 0:12:57 time: 0.6351 data_time: 0.0017 memory: 44140 loss: 0.4684 +2023/06/05 22:32:37 - mmengine - INFO - Epoch(train) [10][3000/4111] lr: 1.0162e-05 eta: 0:11:53 time: 0.6356 data_time: 0.0018 memory: 44140 loss: 0.4788 +2023/06/05 22:32:38 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_stylegan3_lr1e-4_20230605_145625 +2023/06/05 22:33:41 - mmengine - INFO - Epoch(train) [10][3100/4111] lr: 1.0135e-05 eta: 0:10:48 time: 0.6391 data_time: 0.0015 memory: 44140 loss: 0.4751 +2023/06/05 22:34:44 - mmengine - INFO - Epoch(train) [10][3200/4111] lr: 1.0109e-05 eta: 0:09:44 time: 0.6355 data_time: 0.0015 memory: 44140 loss: 0.5071 +2023/06/05 22:35:48 - mmengine - INFO - Epoch(train) [10][3300/4111] lr: 1.0087e-05 eta: 0:08:40 time: 0.6339 data_time: 0.0017 memory: 44140 loss: 0.4607 +2023/06/05 22:36:51 - mmengine - INFO - Epoch(train) [10][3400/4111] lr: 1.0067e-05 eta: 0:07:36 time: 0.6332 data_time: 0.0016 memory: 44140 loss: 0.4886 +2023/06/05 22:37:55 - mmengine - INFO - Epoch(train) [10][3500/4111] lr: 1.0049e-05 eta: 0:06:32 time: 0.6341 data_time: 0.0015 memory: 44140 loss: 0.4588 +2023/06/05 22:38:58 - mmengine - INFO - Epoch(train) [10][3600/4111] lr: 1.0034e-05 eta: 0:05:27 time: 0.6348 data_time: 0.0016 memory: 44140 loss: 0.4638 +2023/06/05 22:40:02 - mmengine - INFO - Epoch(train) [10][3700/4111] lr: 1.0022e-05 eta: 0:04:23 time: 0.6357 data_time: 0.0015 memory: 44140 loss: 0.4812 +2023/06/05 22:41:06 - mmengine - INFO - Epoch(train) [10][3800/4111] lr: 1.0013e-05 eta: 0:03:19 time: 0.6465 data_time: 0.0139 memory: 44140 loss: 0.4870 +2023/06/05 22:42:09 - mmengine - INFO - Epoch(train) [10][3900/4111] lr: 1.0006e-05 eta: 0:02:15 time: 0.6362 data_time: 0.0017 memory: 44140 loss: 0.4826 +2023/06/05 22:43:13 - mmengine - INFO - Epoch(train) [10][4000/4111] lr: 1.0002e-05 eta: 0:01:11 time: 0.6367 data_time: 0.0022 memory: 44140 loss: 0.4957 +2023/06/05 22:43:13 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_stylegan3_lr1e-4_20230605_145625 +2023/06/05 22:44:16 - mmengine - INFO - Epoch(train) [10][4100/4111] lr: 1.0000e-05 eta: 0:00:07 time: 0.6364 data_time: 0.0031 memory: 44140 loss: 0.4520 +2023/06/05 22:44:23 - mmengine - INFO - Exp name: clip_large_pretrain_4x256_stylegan3_lr1e-4_20230605_145625 +2023/06/05 22:44:23 - mmengine - INFO - Saving checkpoint at 10 epochs +2023/06/05 22:46:44 - mmengine - INFO - Epoch(val) [10][100/119] eta: 0:00:24 time: 1.2819 data_time: 0.0008 memory: 44140 +2023/06/05 22:47:26 - mmengine - INFO - Epoch(val) [10][119/119] accuracy/top1: 89.4444 single-label/precision_classwise: [93.37535858154297, 75.09619903564453] single-label/recall_classwise: [93.19066619873047, 75.6434097290039] single-label/f1-score_classwise: [93.28292083740234, 75.36881256103516] data_time: 0.0178 time: 1.2793 diff --git a/clip_large_pretrain_4x256_stylegan3_lr1e-4/20230605_145625/vis_data/20230605_145625.json b/clip_large_pretrain_4x256_stylegan3_lr1e-4/20230605_145625/vis_data/20230605_145625.json new file mode 100644 index 0000000000000000000000000000000000000000..d4c0de8dafed7889fdf44dfd2a12da71f06c1a52 --- /dev/null +++ b/clip_large_pretrain_4x256_stylegan3_lr1e-4/20230605_145625/vis_data/20230605_145625.json @@ -0,0 +1,420 @@ +{"lr": 9.999871217862936e-05, "data_time": 0.0015280723571777343, "loss": 0.5829696655273438, "time": 0.6316387891769409, "epoch": 1, "memory": 44139, "step": 100} +{"lr": 9.999479662539605e-05, "data_time": 0.0016149282455444336, "loss": 0.5629205703735352, "time": 0.6347221851348877, "epoch": 1, "memory": 44139, "step": 200} +{"lr": 9.99882534375654e-05, "data_time": 0.0016250133514404297, "loss": 0.5592653274536132, "time": 0.632369875907898, "epoch": 1, "memory": 44139, "step": 300} +{"lr": 9.997908299725075e-05, "data_time": 0.0015004873275756836, "loss": 0.5522704601287842, "time": 0.6337758302688599, "epoch": 1, "memory": 44139, "step": 400} +{"lr": 9.996728583999318e-05, "data_time": 0.0015186309814453126, "loss": 0.5318122386932373, "time": 0.6317744255065918, "epoch": 1, "memory": 44139, "step": 500} +{"lr": 9.995286265473058e-05, "data_time": 0.0017360925674438476, "loss": 0.5652305126190186, "time": 0.6330352783203125, "epoch": 1, "memory": 44139, "step": 600} +{"lr": 9.993581428375725e-05, "data_time": 0.001550602912902832, "loss": 0.5028891801834107, "time": 0.6331093072891235, "epoch": 1, "memory": 44139, "step": 700} +{"lr": 9.991614172267466e-05, "data_time": 0.001519465446472168, "loss": 0.5037070035934448, "time": 0.6332561016082764, "epoch": 1, "memory": 44139, "step": 800} +{"lr": 9.989384612033355e-05, "data_time": 0.0014886856079101562, "loss": 0.5305287957191467, "time": 0.6337663888931274, "epoch": 1, "memory": 44139, "step": 900} +{"lr": 9.986892877876658e-05, "data_time": 0.0013533830642700195, "loss": 0.5345773220062255, "time": 0.6327988624572753, "epoch": 1, "memory": 44139, "step": 1000} +{"lr": 9.984139115311244e-05, "data_time": 0.0015159368515014649, "loss": 0.5430529594421387, "time": 0.6334908246994019, "epoch": 1, "memory": 44139, "step": 1100} +{"lr": 9.981123485153081e-05, "data_time": 0.0015604257583618163, "loss": 0.550173544883728, "time": 0.6322064161300659, "epoch": 1, "memory": 44139, "step": 1200} +{"lr": 9.977846163510851e-05, "data_time": 0.0014005184173583984, "loss": 0.4987194061279297, "time": 0.6331788063049316, "epoch": 1, "memory": 44139, "step": 1300} +{"lr": 9.974307341775653e-05, "data_time": 0.0016389608383178711, "loss": 0.5176548004150391, "time": 0.6326443910598755, "epoch": 1, "memory": 44139, "step": 1400} +{"lr": 9.970507226609833e-05, "data_time": 0.0014583349227905273, "loss": 0.49719810485839844, "time": 0.6342327356338501, "epoch": 1, "memory": 44139, "step": 1500} +{"lr": 9.966446039934934e-05, "data_time": 0.0014329671859741211, "loss": 0.5203955173492432, "time": 0.6341060400009155, "epoch": 1, "memory": 44139, "step": 1600} +{"lr": 9.962124018918683e-05, "data_time": 0.0013983011245727538, "loss": 0.5304009437561035, "time": 0.6350225448608399, "epoch": 1, "memory": 44139, "step": 1700} +{"lr": 9.957541415961212e-05, "data_time": 0.0013764858245849609, "loss": 0.49931039810180666, "time": 0.6334275484085083, "epoch": 1, "memory": 44139, "step": 1800} +{"lr": 9.952698498680267e-05, "data_time": 0.0013906240463256836, "loss": 0.5129714608192444, "time": 0.6329006910324096, "epoch": 1, "memory": 44139, "step": 1900} +{"lr": 9.947595549895583e-05, "data_time": 0.0014127492904663086, "loss": 0.48182153701782227, "time": 0.63444082736969, "epoch": 1, "memory": 44139, "step": 2000} +{"lr": 9.942232867612417e-05, "data_time": 0.001363086700439453, "loss": 0.4994787693023682, "time": 0.6327764987945557, "epoch": 1, "memory": 44139, "step": 2100} +{"lr": 9.93661076500406e-05, "data_time": 0.0014455795288085937, "loss": 0.5223768472671508, "time": 0.632605504989624, "epoch": 1, "memory": 44139, "step": 2200} +{"lr": 9.930729570393622e-05, "data_time": 0.0014699697494506836, "loss": 0.5021018862724305, "time": 0.6406325101852417, "epoch": 1, "memory": 44139, "step": 2300} +{"lr": 9.924589627234824e-05, "data_time": 0.0019201278686523438, "loss": 0.5076982319355011, "time": 0.6326788187026977, "epoch": 1, "memory": 44139, "step": 2400} +{"lr": 9.918191294091956e-05, "data_time": 0.0014031887054443359, "loss": 0.5193401515483856, "time": 0.6322619199752808, "epoch": 1, "memory": 44139, "step": 2500} +{"lr": 9.91153494461892e-05, "data_time": 0.0015448808670043945, "loss": 0.4991978645324707, "time": 0.6341390609741211, "epoch": 1, "memory": 44139, "step": 2600} +{"lr": 9.904620967537422e-05, "data_time": 0.0015372753143310547, "loss": 0.49636522531509397, "time": 0.6350481271743774, "epoch": 1, "memory": 44139, "step": 2700} +{"lr": 9.897449766614277e-05, "data_time": 0.0016783714294433595, "loss": 0.5006751537322998, "time": 0.6336184740066528, "epoch": 1, "memory": 44139, "step": 2800} +{"lr": 9.890021760637804e-05, "data_time": 0.0014549493789672852, "loss": 0.5036589443683624, "time": 0.6338624715805053, "epoch": 1, "memory": 44139, "step": 2900} +{"lr": 9.882337383393409e-05, "data_time": 0.001423048973083496, "loss": 0.4870572656393051, "time": 0.6328262329101563, "epoch": 1, "memory": 44139, "step": 3000} +{"lr": 9.874397083638193e-05, "data_time": 0.001641249656677246, "loss": 0.5131533741950989, "time": 0.6331220149993897, "epoch": 1, "memory": 44139, "step": 3100} +{"lr": 9.866201325074828e-05, "data_time": 0.0014507770538330078, "loss": 0.48008095026016234, "time": 0.6418931007385253, "epoch": 1, "memory": 44139, "step": 3200} +{"lr": 9.857750586324393e-05, "data_time": 0.001474308967590332, "loss": 0.480618292093277, "time": 0.6352507829666137, "epoch": 1, "memory": 44139, "step": 3300} +{"lr": 9.84904536089848e-05, "data_time": 0.0013846158981323242, "loss": 0.4827093005180359, "time": 0.6340517044067383, "epoch": 1, "memory": 44139, "step": 3400} +{"lr": 9.840086157170351e-05, "data_time": 0.0016017675399780274, "loss": 0.5124086380004883, "time": 0.633268666267395, "epoch": 1, "memory": 44139, "step": 3500} +{"lr": 9.830873498345255e-05, "data_time": 0.0020368099212646484, "loss": 0.469501256942749, "time": 0.6321164608001709, "epoch": 1, "memory": 44139, "step": 3600} +{"lr": 9.821407922429864e-05, "data_time": 0.0015007972717285157, "loss": 0.4675758719444275, "time": 0.6327744722366333, "epoch": 1, "memory": 44139, "step": 3700} +{"lr": 9.811689982200877e-05, "data_time": 0.0014689207077026368, "loss": 0.48063214421272277, "time": 0.6344436407089233, "epoch": 1, "memory": 44139, "step": 3800} +{"lr": 9.801720245172702e-05, "data_time": 0.0017609119415283204, "loss": 0.4920886754989624, "time": 0.6328966617584229, "epoch": 1, "memory": 44139, "step": 3900} +{"lr": 9.791499293564358e-05, "data_time": 0.0015110969543457031, "loss": 0.5180442333221436, "time": 0.6321677446365357, "epoch": 1, "memory": 44139, "step": 4000} +{"lr": 9.781027724265439e-05, "data_time": 0.002924323081970215, "loss": 0.5090823292732238, "time": 0.6352986574172974, "epoch": 1, "memory": 44139, "step": 4100} +{"accuracy/top1": 79.6716537475586, "data_time": 0.023074809242697322, "time": 1.3137203725446172, "step": 1} +{"lr": 9.769111539369434e-05, "data_time": 0.001417851448059082, "loss": 0.464958930015564, "time": 0.6323960781097412, "epoch": 2, "memory": 44140, "step": 4211} +{"lr": 9.758113190911623e-05, "data_time": 0.001508808135986328, "loss": 0.4936454653739929, "time": 0.6337024450302124, "epoch": 2, "memory": 44140, "step": 4311} +{"lr": 9.746866174465984e-05, "data_time": 0.0016089677810668945, "loss": 0.4635092854499817, "time": 0.6390480756759643, "epoch": 2, "memory": 44140, "step": 4411} +{"lr": 9.735371146842894e-05, "data_time": 0.0017129898071289063, "loss": 0.4712280809879303, "time": 0.6333528280258178, "epoch": 2, "memory": 44140, "step": 4511} +{"lr": 9.723628779336255e-05, "data_time": 0.0014141321182250977, "loss": 0.4785924434661865, "time": 0.6327715158462525, "epoch": 2, "memory": 44140, "step": 4611} +{"lr": 9.711639757684265e-05, "data_time": 0.0014632940292358398, "loss": 0.4785075306892395, "time": 0.632625937461853, "epoch": 2, "memory": 44140, "step": 4711} +{"lr": 9.699404782029395e-05, "data_time": 0.0013666629791259765, "loss": 0.4657517999410629, "time": 0.6335940837860108, "epoch": 2, "memory": 44140, "step": 4811} +{"lr": 9.686924566877498e-05, "data_time": 0.001330256462097168, "loss": 0.4924519777297974, "time": 0.6338048458099366, "epoch": 2, "memory": 44140, "step": 4911} +{"lr": 9.674199841056078e-05, "data_time": 0.0016278743743896485, "loss": 0.4578102469444275, "time": 0.6337306022644043, "epoch": 2, "memory": 44140, "step": 5011} +{"lr": 9.661231347671705e-05, "data_time": 0.0014542341232299805, "loss": 0.4974474787712097, "time": 0.6327146291732788, "epoch": 2, "memory": 44140, "step": 5111} +{"lr": 9.64801984406668e-05, "data_time": 0.00137786865234375, "loss": 0.47127229571342466, "time": 0.6333929538726807, "epoch": 2, "memory": 44140, "step": 5211} +{"lr": 9.634566101774737e-05, "data_time": 0.001451420783996582, "loss": 0.4649467468261719, "time": 0.6328048229217529, "epoch": 2, "memory": 44140, "step": 5311} +{"lr": 9.62087090647604e-05, "data_time": 0.001425766944885254, "loss": 0.4517827838659286, "time": 0.6339376926422119, "epoch": 2, "memory": 44140, "step": 5411} +{"lr": 9.606935057951247e-05, "data_time": 0.0014910459518432616, "loss": 0.448582249879837, "time": 0.6333621501922607, "epoch": 2, "memory": 44140, "step": 5511} +{"lr": 9.592759370034884e-05, "data_time": 0.0013057708740234375, "loss": 0.4891594350337982, "time": 0.6343136310577393, "epoch": 2, "memory": 44140, "step": 5611} +{"lr": 9.578344670567721e-05, "data_time": 0.0014177322387695312, "loss": 0.48531308174133303, "time": 0.6324173927307128, "epoch": 2, "memory": 44140, "step": 5711} +{"lr": 9.563691801348495e-05, "data_time": 0.0014531612396240234, "loss": 0.47551594376564027, "time": 0.6330333948135376, "epoch": 2, "memory": 44140, "step": 5811} +{"lr": 9.548801618084738e-05, "data_time": 0.0015380144119262694, "loss": 0.5018477857112884, "time": 0.6336065053939819, "epoch": 2, "memory": 44140, "step": 5911} +{"lr": 9.533674990342786e-05, "data_time": 0.001472616195678711, "loss": 0.45980127453804015, "time": 0.6343237161636353, "epoch": 2, "memory": 44140, "step": 6011} +{"lr": 9.518312801497029e-05, "data_time": 0.0014411449432373048, "loss": 0.462513792514801, "time": 0.6327403783798218, "epoch": 2, "memory": 44140, "step": 6111} +{"lr": 9.502715948678276e-05, "data_time": 0.0014458179473876953, "loss": 0.47115838527679443, "time": 0.6341397285461425, "epoch": 2, "memory": 44140, "step": 6211} +{"lr": 9.486885342721427e-05, "data_time": 0.0013930082321166992, "loss": 0.49234930574893954, "time": 0.6331521511077881, "epoch": 2, "memory": 44140, "step": 6311} +{"lr": 9.470821908112202e-05, "data_time": 0.0015508174896240235, "loss": 0.4903668165206909, "time": 0.6341366529464721, "epoch": 2, "memory": 44140, "step": 6411} +{"lr": 9.454526582933214e-05, "data_time": 0.0015165328979492188, "loss": 0.478467321395874, "time": 0.6331990718841553, "epoch": 2, "memory": 44140, "step": 6511} +{"lr": 9.438000318809176e-05, "data_time": 0.0014827966690063477, "loss": 0.44587414264678954, "time": 0.6328572988510132, "epoch": 2, "memory": 44140, "step": 6611} +{"lr": 9.42124408085132e-05, "data_time": 0.0014327049255371093, "loss": 0.4434362292289734, "time": 0.6405970573425293, "epoch": 2, "memory": 44140, "step": 6711} +{"lr": 9.404258847601036e-05, "data_time": 0.0013992786407470703, "loss": 0.467532616853714, "time": 0.6336398363113404, "epoch": 2, "memory": 44140, "step": 6811} +{"lr": 9.387045610972717e-05, "data_time": 0.0013389110565185547, "loss": 0.4624688357114792, "time": 0.6350803852081299, "epoch": 2, "memory": 44140, "step": 6911} +{"lr": 9.369605376195838e-05, "data_time": 0.0013905048370361328, "loss": 0.46402873992919924, "time": 0.6326971769332885, "epoch": 2, "memory": 44140, "step": 7011} +{"lr": 9.351939161756269e-05, "data_time": 0.001452493667602539, "loss": 0.48670401573181155, "time": 0.6348248720169067, "epoch": 2, "memory": 44140, "step": 7111} +{"lr": 9.334047999336764e-05, "data_time": 0.001385951042175293, "loss": 0.4709049880504608, "time": 0.6329421520233154, "epoch": 2, "memory": 44140, "step": 7211} +{"lr": 9.315932933756752e-05, "data_time": 0.0013277053833007813, "loss": 0.48430325388908385, "time": 0.6384798049926758, "epoch": 2, "memory": 44140, "step": 7311} +{"lr": 9.297595022911287e-05, "data_time": 0.0014803886413574218, "loss": 0.4709764182567596, "time": 0.6329472303390503, "epoch": 2, "memory": 44140, "step": 7411} +{"lr": 9.279035337709295e-05, "data_time": 0.001465606689453125, "loss": 0.4562969863414764, "time": 0.6321838855743408, "epoch": 2, "memory": 44140, "step": 7511} +{"lr": 9.260254962011016e-05, "data_time": 0.0014004230499267579, "loss": 0.4973665475845337, "time": 0.6325476884841919, "epoch": 2, "memory": 44140, "step": 7611} +{"lr": 9.241254992564702e-05, "data_time": 0.0014434814453125, "loss": 0.496775871515274, "time": 0.6410278081893921, "epoch": 2, "memory": 44140, "step": 7711} +{"lr": 9.22203653894259e-05, "data_time": 0.0014738321304321289, "loss": 0.46343018114566803, "time": 0.6349817752838135, "epoch": 2, "memory": 44140, "step": 7811} +{"lr": 9.202600723476098e-05, "data_time": 0.0014462232589721679, "loss": 0.457493257522583, "time": 0.6337049722671508, "epoch": 2, "memory": 44140, "step": 7911} +{"lr": 9.182948681190268e-05, "data_time": 0.0014832973480224609, "loss": 0.48266295790672303, "time": 0.6323026657104492, "epoch": 2, "memory": 44140, "step": 8011} +{"lr": 9.163081559737497e-05, "data_time": 0.0014436960220336914, "loss": 0.46472718715667727, "time": 0.6329636096954345, "epoch": 2, "memory": 44140, "step": 8111} +{"lr": 9.143000519330517e-05, "data_time": 0.0034996747970581056, "loss": 0.47150681614875795, "time": 0.634698486328125, "epoch": 2, "memory": 44140, "step": 8211} +{"accuracy/top1": 84.409912109375, "data_time": 0.01794877052307129, "time": 1.280187298854192, "step": 2} +{"lr": 9.120461478753376e-05, "data_time": 0.0016698837280273438, "loss": 0.5030023932456971, "time": 0.6323754787445068, "epoch": 3, "memory": 44140, "step": 8322} +{"lr": 9.099932932138134e-05, "data_time": 0.0015458345413208007, "loss": 0.45751610994338987, "time": 0.6347354412078857, "epoch": 3, "memory": 44140, "step": 8422} +{"lr": 9.079194154362181e-05, "data_time": 0.001558208465576172, "loss": 0.4556926995515823, "time": 0.6327492475509644, "epoch": 3, "memory": 44140, "step": 8522} +{"lr": 9.058246356541787e-05, "data_time": 0.0015683650970458984, "loss": 0.47245320677757263, "time": 0.6333695650100708, "epoch": 3, "memory": 44140, "step": 8622} +{"lr": 9.037090761999699e-05, "data_time": 0.0014644861221313477, "loss": 0.47963533997535707, "time": 0.6332862138748169, "epoch": 3, "memory": 44140, "step": 8722} +{"lr": 9.015728606193701e-05, "data_time": 0.0013886690139770508, "loss": 0.47418411672115324, "time": 0.6339986562728882, "epoch": 3, "memory": 44140, "step": 8822} +{"lr": 8.994161136644488e-05, "data_time": 0.0014088869094848633, "loss": 0.4321967601776123, "time": 0.6335744142532349, "epoch": 3, "memory": 44140, "step": 8922} +{"lr": 8.972389612862792e-05, "data_time": 0.0015585660934448243, "loss": 0.47279823422431944, "time": 0.6356832265853882, "epoch": 3, "memory": 44140, "step": 9022} +{"lr": 8.950415306275833e-05, "data_time": 0.0015932083129882812, "loss": 0.4606236100196838, "time": 0.6338690996170044, "epoch": 3, "memory": 44140, "step": 9122} +{"lr": 8.928239500153059e-05, "data_time": 0.0014430999755859375, "loss": 0.4547202467918396, "time": 0.6340134382247925, "epoch": 3, "memory": 44140, "step": 9222} +{"lr": 8.905863489531232e-05, "data_time": 0.0014305591583251953, "loss": 0.4499098300933838, "time": 0.6330832958221435, "epoch": 3, "memory": 44140, "step": 9322} +{"lr": 8.883288581138775e-05, "data_time": 0.0015149831771850586, "loss": 0.4653953969478607, "time": 0.6323587656021118, "epoch": 3, "memory": 44140, "step": 9422} +{"lr": 8.86051609331947e-05, "data_time": 0.0016928434371948243, "loss": 0.4494109511375427, "time": 0.6328791379928589, "epoch": 3, "memory": 44140, "step": 9522} +{"lr": 8.837547355955467e-05, "data_time": 0.0014689207077026368, "loss": 0.44448021054267883, "time": 0.6352570533752442, "epoch": 3, "memory": 44140, "step": 9622} +{"lr": 8.814383710389614e-05, "data_time": 0.001623678207397461, "loss": 0.4464535593986511, "time": 0.6360183000564575, "epoch": 3, "memory": 44140, "step": 9722} +{"lr": 8.791026509347146e-05, "data_time": 0.0014097213745117188, "loss": 0.45558726191520693, "time": 0.6346526145935059, "epoch": 3, "memory": 44140, "step": 9822} +{"lr": 8.767477116856659e-05, "data_time": 0.0016045093536376954, "loss": 0.45931233167648317, "time": 0.6333139657974243, "epoch": 3, "memory": 44140, "step": 9922} +{"lr": 8.743736908170466e-05, "data_time": 0.0014598846435546875, "loss": 0.4937210500240326, "time": 0.6338386774063111, "epoch": 3, "memory": 44140, "step": 10022} +{"lr": 8.719807269684305e-05, "data_time": 0.00140688419342041, "loss": 0.4547117233276367, "time": 0.6339017152786255, "epoch": 3, "memory": 44140, "step": 10122} +{"lr": 8.695689598856356e-05, "data_time": 0.001477670669555664, "loss": 0.4673060417175293, "time": 0.6335069179534912, "epoch": 3, "memory": 44140, "step": 10222} +{"lr": 8.67138530412559e-05, "data_time": 0.0014937639236450196, "loss": 0.4838897228240967, "time": 0.633981990814209, "epoch": 3, "memory": 44140, "step": 10322} +{"lr": 8.646895804829604e-05, "data_time": 0.0015043258666992188, "loss": 0.48540433049201964, "time": 0.6353799819946289, "epoch": 3, "memory": 44140, "step": 10422} +{"lr": 8.622222531121664e-05, "data_time": 0.0016132593154907227, "loss": 0.45520622134208677, "time": 0.6336303234100342, "epoch": 3, "memory": 44140, "step": 10522} +{"lr": 8.597366923887211e-05, "data_time": 0.0015883207321166991, "loss": 0.4660855710506439, "time": 0.6341121912002563, "epoch": 3, "memory": 44140, "step": 10622} +{"lr": 8.57233043465972e-05, "data_time": 0.001558232307434082, "loss": 0.4547513484954834, "time": 0.6333629131317139, "epoch": 3, "memory": 44140, "step": 10722} +{"lr": 8.547114525535937e-05, "data_time": 0.001423811912536621, "loss": 0.4696462869644165, "time": 0.635311222076416, "epoch": 3, "memory": 44140, "step": 10822} +{"lr": 8.521720669090474e-05, "data_time": 0.001542949676513672, "loss": 0.44859603345394133, "time": 0.6340259313583374, "epoch": 3, "memory": 44140, "step": 10922} +{"lr": 8.496150348289821e-05, "data_time": 0.0013875961303710938, "loss": 0.4574402332305908, "time": 0.6333667039871216, "epoch": 3, "memory": 44140, "step": 11022} +{"lr": 8.470405056405736e-05, "data_time": 0.0014806509017944336, "loss": 0.47548220455646517, "time": 0.6347490787506104, "epoch": 3, "memory": 44140, "step": 11122} +{"lr": 8.44448629692807e-05, "data_time": 0.0015207767486572266, "loss": 0.42512376308441163, "time": 0.6337342500686646, "epoch": 3, "memory": 44140, "step": 11222} +{"lr": 8.418395583476941e-05, "data_time": 0.0016474485397338866, "loss": 0.45764694511890414, "time": 0.6348349571228027, "epoch": 3, "memory": 44140, "step": 11322} +{"lr": 8.392134439714336e-05, "data_time": 0.0016633272171020508, "loss": 0.48138769865036013, "time": 0.6338225603103638, "epoch": 3, "memory": 44140, "step": 11422} +{"lr": 8.365704399255133e-05, "data_time": 0.0015475988388061524, "loss": 0.47832974791526794, "time": 0.635752534866333, "epoch": 3, "memory": 44140, "step": 11522} +{"lr": 8.339107005577562e-05, "data_time": 0.0013602495193481446, "loss": 0.46606722474098206, "time": 0.6338636159896851, "epoch": 3, "memory": 44140, "step": 11622} +{"lr": 8.31234381193305e-05, "data_time": 0.001706409454345703, "loss": 0.46423605680465696, "time": 0.6335791826248169, "epoch": 3, "memory": 44140, "step": 11722} +{"lr": 8.285416381255504e-05, "data_time": 0.001396942138671875, "loss": 0.4623516917228699, "time": 0.6347856760025025, "epoch": 3, "memory": 44140, "step": 11822} +{"lr": 8.258326286070047e-05, "data_time": 0.002235102653503418, "loss": 0.44467554688453675, "time": 0.6375418424606323, "epoch": 3, "memory": 44140, "step": 11922} +{"lr": 8.231075108401202e-05, "data_time": 0.001590728759765625, "loss": 0.4796634018421173, "time": 0.6340131998062134, "epoch": 3, "memory": 44140, "step": 12022} +{"lr": 8.203664439680493e-05, "data_time": 0.0017249584197998047, "loss": 0.4819711267948151, "time": 0.6324265480041504, "epoch": 3, "memory": 44140, "step": 12122} +{"lr": 8.176095880653475e-05, "data_time": 0.0014985322952270508, "loss": 0.4436904340982437, "time": 0.6329562187194824, "epoch": 3, "memory": 44140, "step": 12222} +{"lr": 8.148371041286308e-05, "data_time": 0.0031274080276489256, "loss": 0.45416841506958006, "time": 0.6359366178512573, "epoch": 3, "memory": 44140, "step": 12322} +{"accuracy/top1": 86.57630920410156, "data_time": 0.017765897512435912, "time": 1.279219208161036, "step": 3} +{"lr": 8.117415423358443e-05, "data_time": 0.0016343116760253907, "loss": 0.45611473321914675, "time": 0.6361605644226074, "epoch": 4, "memory": 44140, "step": 12433} +{"lr": 8.089366155759266e-05, "data_time": 0.001431417465209961, "loss": 0.448657700419426, "time": 0.6329862833023071, "epoch": 4, "memory": 44140, "step": 12533} +{"lr": 8.061165672717273e-05, "data_time": 0.0016322374343872071, "loss": 0.4526179552078247, "time": 0.6328766822814942, "epoch": 4, "memory": 44140, "step": 12633} +{"lr": 8.032815621102116e-05, "data_time": 0.0014635801315307617, "loss": 0.4790175259113312, "time": 0.6336992502212524, "epoch": 4, "memory": 44140, "step": 12733} +{"lr": 8.004317656518042e-05, "data_time": 0.0016112565994262696, "loss": 0.41661471128463745, "time": 0.6339985609054566, "epoch": 4, "memory": 44140, "step": 12833} +{"lr": 7.975673443207223e-05, "data_time": 0.0014017581939697265, "loss": 0.47605031728744507, "time": 0.6359712362289429, "epoch": 4, "memory": 44140, "step": 12933} +{"lr": 7.946884653952537e-05, "data_time": 0.001673579216003418, "loss": 0.44245787262916564, "time": 0.6362274169921875, "epoch": 4, "memory": 44140, "step": 13033} +{"lr": 7.91795296997992e-05, "data_time": 0.0018304109573364258, "loss": 0.46397862434387205, "time": 0.6344977617263794, "epoch": 4, "memory": 44140, "step": 13133} +{"lr": 7.888880080860151e-05, "data_time": 0.0017838716506958009, "loss": 0.47162403762340543, "time": 0.6353833913803101, "epoch": 4, "memory": 44140, "step": 13233} +{"lr": 7.859667684410205e-05, "data_time": 0.0021419286727905273, "loss": 0.4530477821826935, "time": 0.6351371049880982, "epoch": 4, "memory": 44140, "step": 13333} +{"lr": 7.830317486594082e-05, "data_time": 0.002172255516052246, "loss": 0.46278727650642393, "time": 0.6350919961929321, "epoch": 4, "memory": 44140, "step": 13433} +{"lr": 7.800831201423207e-05, "data_time": 0.0021567583084106446, "loss": 0.4498206853866577, "time": 0.6344151735305786, "epoch": 4, "memory": 44140, "step": 13533} +{"lr": 7.771210550856309e-05, "data_time": 0.00211029052734375, "loss": 0.46084251403808596, "time": 0.6326980590820312, "epoch": 4, "memory": 44140, "step": 13633} +{"lr": 7.741457264698884e-05, "data_time": 0.0016243696212768556, "loss": 0.4272093951702118, "time": 0.6350467681884766, "epoch": 4, "memory": 44140, "step": 13733} +{"lr": 7.711573080502147e-05, "data_time": 0.001788187026977539, "loss": 0.441609525680542, "time": 0.6337425231933593, "epoch": 4, "memory": 44140, "step": 13833} +{"lr": 7.681559743461609e-05, "data_time": 0.001470804214477539, "loss": 0.46558114886283875, "time": 0.6330921173095703, "epoch": 4, "memory": 44140, "step": 13933} +{"lr": 7.651419006315099e-05, "data_time": 0.0014998674392700194, "loss": 0.4717443287372589, "time": 0.6336032390594483, "epoch": 4, "memory": 44140, "step": 14033} +{"lr": 7.621152629240453e-05, "data_time": 0.0021607637405395507, "loss": 0.4876570016145706, "time": 0.632820463180542, "epoch": 4, "memory": 44140, "step": 14133} +{"lr": 7.590762379752719e-05, "data_time": 0.0017596721649169923, "loss": 0.4612249732017517, "time": 0.6332582473754883, "epoch": 4, "memory": 44140, "step": 14233} +{"lr": 7.560250032600911e-05, "data_time": 0.0018387317657470703, "loss": 0.46215103268623353, "time": 0.6342531681060791, "epoch": 4, "memory": 44140, "step": 14333} +{"lr": 7.52961736966437e-05, "data_time": 0.0012931108474731445, "loss": 0.4643726646900177, "time": 0.6360110282897949, "epoch": 4, "memory": 44140, "step": 14433} +{"lr": 7.498866179848732e-05, "data_time": 0.0015786170959472656, "loss": 0.4921614289283752, "time": 0.6338835239410401, "epoch": 4, "memory": 44140, "step": 14533} +{"lr": 7.467998258981415e-05, "data_time": 0.0014732837677001952, "loss": 0.4758285641670227, "time": 0.6380360841751098, "epoch": 4, "memory": 44140, "step": 14633} +{"lr": 7.437015409706804e-05, "data_time": 0.0018257617950439454, "loss": 0.44287298917770385, "time": 0.6327397108078003, "epoch": 4, "memory": 44140, "step": 14733} +{"lr": 7.405919441380929e-05, "data_time": 0.0014877557754516602, "loss": 0.4586209297180176, "time": 0.6339401006698608, "epoch": 4, "memory": 44140, "step": 14833} +{"lr": 7.374712169965821e-05, "data_time": 0.0017371654510498046, "loss": 0.48081800937652586, "time": 0.6329304218292237, "epoch": 4, "memory": 44140, "step": 14933} +{"lr": 7.343395417923461e-05, "data_time": 0.0015257596969604492, "loss": 0.4388641655445099, "time": 0.6340760231018067, "epoch": 4, "memory": 44140, "step": 15033} +{"lr": 7.311971014109342e-05, "data_time": 0.00172271728515625, "loss": 0.4692396819591522, "time": 0.6333871126174927, "epoch": 4, "memory": 44140, "step": 15133} +{"lr": 7.280440793665682e-05, "data_time": 0.0017844676971435548, "loss": 0.4430750787258148, "time": 0.6360200881958008, "epoch": 4, "memory": 44140, "step": 15233} +{"lr": 7.248806597914234e-05, "data_time": 0.002052974700927734, "loss": 0.47781543135643006, "time": 0.6328158617019654, "epoch": 4, "memory": 44140, "step": 15333} +{"lr": 7.217070274248765e-05, "data_time": 0.0024039745330810547, "loss": 0.44683332443237306, "time": 0.634963083267212, "epoch": 4, "memory": 44140, "step": 15433} +{"lr": 7.185233676027203e-05, "data_time": 0.001649928092956543, "loss": 0.46081181764602663, "time": 0.6344746589660645, "epoch": 4, "memory": 44140, "step": 15533} +{"lr": 7.153298662463338e-05, "data_time": 0.001600337028503418, "loss": 0.4723763644695282, "time": 0.6354378461837769, "epoch": 4, "memory": 44140, "step": 15633} +{"lr": 7.121267098518291e-05, "data_time": 0.0017853498458862305, "loss": 0.49327149987220764, "time": 0.6336778163909912, "epoch": 4, "memory": 44140, "step": 15733} +{"lr": 7.089140854791618e-05, "data_time": 0.0014856100082397462, "loss": 0.4476708710193634, "time": 0.6343159675598145, "epoch": 4, "memory": 44140, "step": 15833} +{"lr": 7.056921807412025e-05, "data_time": 0.001572871208190918, "loss": 0.4719863414764404, "time": 0.634116792678833, "epoch": 4, "memory": 44140, "step": 15933} +{"lr": 7.024611837927802e-05, "data_time": 0.0014816999435424804, "loss": 0.4623782217502594, "time": 0.6353331089019776, "epoch": 4, "memory": 44140, "step": 16033} +{"lr": 6.992212833196998e-05, "data_time": 0.0015210628509521485, "loss": 0.4481693387031555, "time": 0.6347303628921509, "epoch": 4, "memory": 44140, "step": 16133} +{"lr": 6.959726685277174e-05, "data_time": 0.002022385597229004, "loss": 0.4535117208957672, "time": 0.6339698791503906, "epoch": 4, "memory": 44140, "step": 16233} +{"lr": 6.927155291314949e-05, "data_time": 0.0016282558441162109, "loss": 0.43472830057144163, "time": 0.6349962711334228, "epoch": 4, "memory": 44140, "step": 16333} +{"lr": 6.894500553435204e-05, "data_time": 0.002988910675048828, "loss": 0.48811282515525817, "time": 0.6419486999511719, "epoch": 4, "memory": 44140, "step": 16433} +{"accuracy/top1": 87.74970245361328, "data_time": 0.01793476939201355, "time": 1.2803063730398814, "step": 4} +{"lr": 6.858158509719903e-05, "data_time": 0.0015555858612060548, "loss": 0.4589222550392151, "time": 0.6371300935745239, "epoch": 5, "memory": 44140, "step": 16544} +{"lr": 6.825334178880171e-05, "data_time": 0.0020379543304443358, "loss": 0.46468005776405336, "time": 0.6338776588439942, "epoch": 5, "memory": 44140, "step": 16644} +{"lr": 6.792432450336645e-05, "data_time": 0.0015887022018432617, "loss": 0.44431387782096865, "time": 0.6346333742141723, "epoch": 5, "memory": 44140, "step": 16744} +{"lr": 6.759455245505286e-05, "data_time": 0.0016175270080566405, "loss": 0.47444817423820496, "time": 0.6343838214874268, "epoch": 5, "memory": 44140, "step": 16844} +{"lr": 6.726404490209753e-05, "data_time": 0.001459050178527832, "loss": 0.4732815623283386, "time": 0.6343556404113769, "epoch": 5, "memory": 44140, "step": 16944} +{"lr": 6.693282114568972e-05, "data_time": 0.0018039703369140624, "loss": 0.46433616280555723, "time": 0.6349769353866577, "epoch": 5, "memory": 44140, "step": 17044} +{"lr": 6.660090052884388e-05, "data_time": 0.0024118423461914062, "loss": 0.46810369193553925, "time": 0.6337874412536622, "epoch": 5, "memory": 44140, "step": 17144} +{"lr": 6.626830243527016e-05, "data_time": 0.002102088928222656, "loss": 0.4910249650478363, "time": 0.6354984045028687, "epoch": 5, "memory": 44140, "step": 17244} +{"lr": 6.593504628824236e-05, "data_time": 0.0015336275100708008, "loss": 0.47623333930969236, "time": 0.6352442264556885, "epoch": 5, "memory": 44140, "step": 17344} +{"lr": 6.560115154946387e-05, "data_time": 0.0017632484436035157, "loss": 0.48938114047050474, "time": 0.6337780952453613, "epoch": 5, "memory": 44140, "step": 17444} +{"lr": 6.526663771793081e-05, "data_time": 0.002121281623840332, "loss": 0.461805933713913, "time": 0.6351408720016479, "epoch": 5, "memory": 44140, "step": 17544} +{"lr": 6.493152432879363e-05, "data_time": 0.0015882253646850586, "loss": 0.471802419424057, "time": 0.6341345071792602, "epoch": 5, "memory": 44140, "step": 17644} +{"lr": 6.459583095221596e-05, "data_time": 0.001488327980041504, "loss": 0.4258020281791687, "time": 0.6344020128250122, "epoch": 5, "memory": 44140, "step": 17744} +{"lr": 6.425957719223197e-05, "data_time": 0.0016002893447875977, "loss": 0.4689207494258881, "time": 0.6341097593307495, "epoch": 5, "memory": 44140, "step": 17844} +{"lr": 6.392278268560153e-05, "data_time": 0.0019644021987915037, "loss": 0.45474722385406496, "time": 0.6330618143081665, "epoch": 5, "memory": 44140, "step": 17944} +{"lr": 6.358546710066316e-05, "data_time": 0.001446080207824707, "loss": 0.4536068499088287, "time": 0.6336838960647583, "epoch": 5, "memory": 44140, "step": 18044} +{"lr": 6.324765013618573e-05, "data_time": 0.0016654014587402343, "loss": 0.46334036290645597, "time": 0.635744571685791, "epoch": 5, "memory": 44140, "step": 18144} +{"lr": 6.290935152021818e-05, "data_time": 0.0014859914779663086, "loss": 0.44370869994163514, "time": 0.6343780994415283, "epoch": 5, "memory": 44140, "step": 18244} +{"lr": 6.2570591008937e-05, "data_time": 0.0021237611770629885, "loss": 0.4457757234573364, "time": 0.6336251735687256, "epoch": 5, "memory": 44140, "step": 18344} +{"lr": 6.223138838549294e-05, "data_time": 0.0016193628311157227, "loss": 0.4611522316932678, "time": 0.634003758430481, "epoch": 5, "memory": 44140, "step": 18444} +{"lr": 6.18917634588553e-05, "data_time": 0.0014549732208251954, "loss": 0.466436493396759, "time": 0.6346589803695679, "epoch": 5, "memory": 44140, "step": 18544} +{"lr": 6.155173606265555e-05, "data_time": 0.0015205144882202148, "loss": 0.4525952637195587, "time": 0.6346549987792969, "epoch": 5, "memory": 44140, "step": 18644} +{"lr": 6.121132605402843e-05, "data_time": 0.0017121315002441406, "loss": 0.47000470757484436, "time": 0.6335221529006958, "epoch": 5, "memory": 44140, "step": 18744} +{"lr": 6.087055331245287e-05, "data_time": 0.0021596670150756834, "loss": 0.4398204982280731, "time": 0.6336713790893554, "epoch": 5, "memory": 44140, "step": 18844} +{"lr": 6.05294377385911e-05, "data_time": 0.0020489931106567384, "loss": 0.4636018514633179, "time": 0.6335805654525757, "epoch": 5, "memory": 44140, "step": 18944} +{"lr": 6.0187999253126355e-05, "data_time": 0.0014872074127197266, "loss": 0.47138131856918336, "time": 0.6342474222183228, "epoch": 5, "memory": 44140, "step": 19044} +{"lr": 5.9846257795599195e-05, "data_time": 0.0014727354049682618, "loss": 0.458501136302948, "time": 0.6362945079803467, "epoch": 5, "memory": 44140, "step": 19144} +{"lr": 5.950423332324355e-05, "data_time": 0.0025372028350830076, "loss": 0.4662016898393631, "time": 0.6349953174591064, "epoch": 5, "memory": 44140, "step": 19244} +{"lr": 5.9161945809820855e-05, "data_time": 0.001747465133666992, "loss": 0.4833052158355713, "time": 3.2729888200759887, "epoch": 5, "memory": 44140, "step": 19344} +{"lr": 5.8819415244453975e-05, "data_time": 0.001704835891723633, "loss": 0.4420823872089386, "time": 0.6357810258865356, "epoch": 5, "memory": 44140, "step": 19444} +{"lr": 5.8476661630459405e-05, "data_time": 0.0019277334213256836, "loss": 0.4586881101131439, "time": 0.6353569984436035, "epoch": 5, "memory": 44140, "step": 19544} +{"lr": 5.81337049841796e-05, "data_time": 0.001938033103942871, "loss": 0.4552523583173752, "time": 0.6364953756332398, "epoch": 5, "memory": 44140, "step": 19644} +{"lr": 5.779056533381373e-05, "data_time": 0.0015594244003295898, "loss": 0.4569220900535583, "time": 0.6351578235626221, "epoch": 5, "memory": 44140, "step": 19744} +{"lr": 5.744726271824817e-05, "data_time": 0.0014957666397094726, "loss": 0.47525659799575803, "time": 0.6341636657714844, "epoch": 5, "memory": 44140, "step": 19844} +{"lr": 5.710381718588622e-05, "data_time": 0.0015401840209960938, "loss": 0.47827436923980715, "time": 0.63492271900177, "epoch": 5, "memory": 44140, "step": 19944} +{"lr": 5.676024879347741e-05, "data_time": 0.002040863037109375, "loss": 0.44967913031578066, "time": 0.6435336112976074, "epoch": 5, "memory": 44140, "step": 20044} +{"lr": 5.6416577604945966e-05, "data_time": 0.0022011280059814455, "loss": 0.4790433168411255, "time": 0.6335307359695435, "epoch": 5, "memory": 44140, "step": 20144} +{"lr": 5.6072823690219385e-05, "data_time": 0.00143125057220459, "loss": 0.45392078161239624, "time": 0.6346201419830322, "epoch": 5, "memory": 44140, "step": 20244} +{"lr": 5.572900712405624e-05, "data_time": 0.001572728157043457, "loss": 0.47307010293006896, "time": 0.6350109815597534, "epoch": 5, "memory": 44140, "step": 20344} +{"lr": 5.538514798487397e-05, "data_time": 0.0020049095153808595, "loss": 0.4750144124031067, "time": 0.6349259614944458, "epoch": 5, "memory": 44140, "step": 20444} +{"lr": 5.5041266353576e-05, "data_time": 0.003738856315612793, "loss": 0.4440926045179367, "time": 0.6351094007492065, "epoch": 5, "memory": 44140, "step": 20544} +{"accuracy/top1": 88.1866226196289, "data_time": 0.017412426074345906, "time": 1.2803153574466706, "step": 5} +{"lr": 5.4659555783035384e-05, "data_time": 0.0014635086059570312, "loss": 0.48406948447227477, "time": 0.6340425491333008, "epoch": 6, "memory": 44140, "step": 20655} +{"lr": 5.431569258426336e-05, "data_time": 0.001412653923034668, "loss": 0.4501304030418396, "time": 0.6351540565490723, "epoch": 6, "memory": 44140, "step": 20755} +{"lr": 5.397186934810864e-05, "data_time": 0.0013927698135375976, "loss": 0.43961726427078246, "time": 0.6360284328460694, "epoch": 6, "memory": 44140, "step": 20855} +{"lr": 5.362810615337815e-05, "data_time": 0.0014640331268310548, "loss": 0.46147736310958865, "time": 0.6339771509170532, "epoch": 6, "memory": 44140, "step": 20955} +{"lr": 5.328442307537233e-05, "data_time": 0.0016271114349365235, "loss": 0.4509885013103485, "time": 0.634494423866272, "epoch": 6, "memory": 44140, "step": 21055} +{"lr": 5.294084018471303e-05, "data_time": 0.0015717506408691405, "loss": 0.47292686700820924, "time": 0.6439121723175049, "epoch": 6, "memory": 44140, "step": 21155} +{"lr": 5.259737754617132e-05, "data_time": 0.0015347957611083984, "loss": 0.49679508805274963, "time": 0.6337537050247193, "epoch": 6, "memory": 44140, "step": 21255} +{"lr": 5.2254055217495525e-05, "data_time": 0.001613020896911621, "loss": 0.4382395207881927, "time": 0.6347167015075683, "epoch": 6, "memory": 44140, "step": 21355} +{"lr": 5.1910893248240283e-05, "data_time": 0.001726388931274414, "loss": 0.4445831775665283, "time": 0.6341121435165405, "epoch": 6, "memory": 44140, "step": 21455} +{"lr": 5.156791167859532e-05, "data_time": 0.002004122734069824, "loss": 0.4478444576263428, "time": 0.6355179786682129, "epoch": 6, "memory": 44140, "step": 21555} +{"lr": 5.122513053821533e-05, "data_time": 0.0014799118041992187, "loss": 0.4767975151538849, "time": 0.6340090990066528, "epoch": 6, "memory": 44140, "step": 21655} +{"lr": 5.0882569845050173e-05, "data_time": 0.0016569852828979491, "loss": 0.47388242483139037, "time": 0.6328195333480835, "epoch": 6, "memory": 44140, "step": 21755} +{"lr": 5.054024960417592e-05, "data_time": 0.0015819787979125977, "loss": 0.4729978024959564, "time": 0.6351083278656006, "epoch": 6, "memory": 44140, "step": 21855} +{"lr": 5.0198189806626494e-05, "data_time": 0.0020102262496948242, "loss": 0.44750158190727235, "time": 0.6341897249221802, "epoch": 6, "memory": 44140, "step": 21955} +{"lr": 4.985641042822647e-05, "data_time": 0.0019378900527954102, "loss": 0.48021815419197084, "time": 0.6335058689117432, "epoch": 6, "memory": 44140, "step": 22055} +{"lr": 4.951493142842412e-05, "data_time": 0.0013368844985961914, "loss": 0.4866527020931244, "time": 0.6338645219802856, "epoch": 6, "memory": 44140, "step": 22155} +{"lr": 4.917377274912603e-05, "data_time": 0.0015365362167358398, "loss": 0.47538907527923585, "time": 0.635843300819397, "epoch": 6, "memory": 44140, "step": 22255} +{"lr": 4.88329543135327e-05, "data_time": 0.001574230194091797, "loss": 0.47498087882995604, "time": 0.6340251445770264, "epoch": 6, "memory": 44140, "step": 22355} +{"lr": 4.849249602497467e-05, "data_time": 0.0020350217819213867, "loss": 0.4520275712013245, "time": 0.6329787969589233, "epoch": 6, "memory": 44140, "step": 22455} +{"lr": 4.8152417765750456e-05, "data_time": 0.001703786849975586, "loss": 0.4768667906522751, "time": 1.2932875871658325, "epoch": 6, "memory": 44140, "step": 22555} +{"lr": 4.7812739395965355e-05, "data_time": 0.0017891883850097655, "loss": 0.47207836508750917, "time": 0.6352556705474853, "epoch": 6, "memory": 44140, "step": 22655} +{"lr": 4.74734807523717e-05, "data_time": 0.0015207052230834961, "loss": 0.4756499171257019, "time": 0.6357563257217407, "epoch": 6, "memory": 44140, "step": 22755} +{"lr": 4.7134661647210416e-05, "data_time": 0.001726675033569336, "loss": 0.4574288845062256, "time": 0.6348897457122803, "epoch": 6, "memory": 44140, "step": 22855} +{"lr": 4.679630186705388e-05, "data_time": 0.002379608154296875, "loss": 0.4630141377449036, "time": 0.635697340965271, "epoch": 6, "memory": 44140, "step": 22955} +{"lr": 4.6458421171650644e-05, "data_time": 0.002309274673461914, "loss": 0.4364140689373016, "time": 0.6367731094360352, "epoch": 6, "memory": 44140, "step": 23055} +{"lr": 4.6121039292771295e-05, "data_time": 0.001967716217041016, "loss": 0.4441856205463409, "time": 0.636125898361206, "epoch": 6, "memory": 44140, "step": 23155} +{"lr": 4.5784175933056197e-05, "data_time": 0.0015056848526000977, "loss": 0.4734744668006897, "time": 0.6337905406951905, "epoch": 6, "memory": 44140, "step": 23255} +{"lr": 4.544785076486494e-05, "data_time": 0.0019669294357299804, "loss": 0.457925409078598, "time": 0.6346128463745118, "epoch": 6, "memory": 44140, "step": 23355} +{"lr": 4.5112083429127475e-05, "data_time": 0.0016997337341308593, "loss": 0.46377404034137726, "time": 0.6371508836746216, "epoch": 6, "memory": 44140, "step": 23455} +{"lr": 4.477689353419706e-05, "data_time": 0.0015819311141967774, "loss": 0.4767437160015106, "time": 0.6379316806793213, "epoch": 6, "memory": 44140, "step": 23555} +{"lr": 4.4442300654705266e-05, "data_time": 0.00226438045501709, "loss": 0.4358688294887543, "time": 0.6347510576248169, "epoch": 6, "memory": 44140, "step": 23655} +{"lr": 4.410832433041866e-05, "data_time": 0.0018082618713378905, "loss": 0.4761827826499939, "time": 0.6339036226272583, "epoch": 6, "memory": 44140, "step": 23755} +{"lr": 4.377498406509792e-05, "data_time": 0.0016064167022705079, "loss": 0.45044424533843996, "time": 0.6349463939666748, "epoch": 6, "memory": 44140, "step": 23855} +{"lr": 4.344229932535875e-05, "data_time": 0.001959872245788574, "loss": 0.4763031482696533, "time": 0.6344331264495849, "epoch": 6, "memory": 44140, "step": 23955} +{"lr": 4.3110289539535115e-05, "data_time": 0.0015465736389160157, "loss": 0.45789788365364076, "time": 0.6343870878219604, "epoch": 6, "memory": 44140, "step": 24055} +{"lr": 4.2778974096544436e-05, "data_time": 0.0019354581832885741, "loss": 0.46838202476501467, "time": 0.6338257551193237, "epoch": 6, "memory": 44140, "step": 24155} +{"lr": 4.2448372344755544e-05, "data_time": 0.0021352529525756835, "loss": 0.45679605603218076, "time": 0.6424296617507934, "epoch": 6, "memory": 44140, "step": 24255} +{"lr": 4.211850359085871e-05, "data_time": 0.00158236026763916, "loss": 0.4579767048358917, "time": 0.6419151306152344, "epoch": 6, "memory": 44140, "step": 24355} +{"lr": 4.17893870987381e-05, "data_time": 0.001468825340270996, "loss": 0.4642179012298584, "time": 0.6362257957458496, "epoch": 6, "memory": 44140, "step": 24455} +{"lr": 4.146104208834679e-05, "data_time": 0.0018604755401611327, "loss": 0.4529073446989059, "time": 0.6338111877441406, "epoch": 6, "memory": 44140, "step": 24555} +{"lr": 4.113348773458442e-05, "data_time": 0.003417181968688965, "loss": 0.4551727056503296, "time": 0.6347980737686157, "epoch": 6, "memory": 44140, "step": 24655} +{"accuracy/top1": 88.90158081054688, "data_time": 0.017988532781600952, "time": 1.2791539053122203, "step": 6} +{"lr": 4.0770851520835164e-05, "data_time": 0.0016108989715576173, "loss": 0.4696533143520355, "time": 0.6349081993103027, "epoch": 7, "memory": 44140, "step": 24766} +{"lr": 4.044502815742039e-05, "data_time": 0.00169525146484375, "loss": 0.46119399070739747, "time": 0.6335535764694213, "epoch": 7, "memory": 44140, "step": 24866} +{"lr": 4.012005478445896e-05, "data_time": 0.002030205726623535, "loss": 0.45948365330696106, "time": 0.6337975025177002, "epoch": 7, "memory": 44140, "step": 24966} +{"lr": 3.9795950379951394e-05, "data_time": 0.0016697406768798827, "loss": 0.4820225715637207, "time": 0.6349336862564087, "epoch": 7, "memory": 44140, "step": 25066} +{"lr": 3.947273387115169e-05, "data_time": 0.0016877174377441407, "loss": 0.45211027264595033, "time": 0.6361408233642578, "epoch": 7, "memory": 44140, "step": 25166} +{"lr": 3.9150424133462044e-05, "data_time": 0.0015559911727905274, "loss": 0.4473924934864044, "time": 0.6361247777938843, "epoch": 7, "memory": 44140, "step": 25266} +{"lr": 3.882903998933037e-05, "data_time": 0.0014920949935913086, "loss": 0.4652102291584015, "time": 0.6383752584457397, "epoch": 7, "memory": 44140, "step": 25366} +{"lr": 3.8508600207151264e-05, "data_time": 0.00156095027923584, "loss": 0.4557941198348999, "time": 0.6369585990905762, "epoch": 7, "memory": 44140, "step": 25466} +{"lr": 3.818912350016978e-05, "data_time": 0.0015782594680786132, "loss": 0.4809683561325073, "time": 0.6361147642135621, "epoch": 7, "memory": 44140, "step": 25566} +{"lr": 3.787062852538883e-05, "data_time": 0.0015544891357421875, "loss": 0.45643723011016846, "time": 0.6339664220809936, "epoch": 7, "memory": 44140, "step": 25666} +{"lr": 3.7553133882479346e-05, "data_time": 0.0016622066497802735, "loss": 0.4706435680389404, "time": 0.6337752103805542, "epoch": 7, "memory": 44140, "step": 25766} +{"lr": 3.723665811269437e-05, "data_time": 0.0017843961715698243, "loss": 0.46476571559906005, "time": 0.6333802938461304, "epoch": 7, "memory": 44140, "step": 25866} +{"lr": 3.6921219697786176e-05, "data_time": 0.0018469333648681641, "loss": 0.4563186526298523, "time": 0.6359025955200195, "epoch": 7, "memory": 44140, "step": 25966} +{"lr": 3.660683705892677e-05, "data_time": 0.001450324058532715, "loss": 0.4826504051685333, "time": 0.633551549911499, "epoch": 7, "memory": 44140, "step": 26066} +{"lr": 3.6293528555632435e-05, "data_time": 0.0017829179763793946, "loss": 0.4337110996246338, "time": 0.6374663829803466, "epoch": 7, "memory": 44140, "step": 26166} +{"lr": 3.598131248469134e-05, "data_time": 0.0014847278594970702, "loss": 0.46282885074615476, "time": 0.6346645355224609, "epoch": 7, "memory": 44140, "step": 26266} +{"lr": 3.567020707909507e-05, "data_time": 0.001493549346923828, "loss": 0.47078515887260436, "time": 0.6341926574707031, "epoch": 7, "memory": 44140, "step": 26366} +{"lr": 3.5360230506973935e-05, "data_time": 0.001625227928161621, "loss": 0.4480311840772629, "time": 0.6349181175231934, "epoch": 7, "memory": 44140, "step": 26466} +{"lr": 3.5051400870535877e-05, "data_time": 0.001460099220275879, "loss": 0.4460991144180298, "time": 0.6339737415313721, "epoch": 7, "memory": 44140, "step": 26566} +{"lr": 3.47437362050094e-05, "data_time": 0.0022795438766479493, "loss": 0.46299099624156953, "time": 0.6366446256637573, "epoch": 7, "memory": 44140, "step": 26666} +{"lr": 3.443725447759023e-05, "data_time": 0.0016202211380004882, "loss": 0.4408200979232788, "time": 0.6363564014434815, "epoch": 7, "memory": 44140, "step": 26766} +{"lr": 3.413197358639224e-05, "data_time": 0.001509714126586914, "loss": 0.4818792164325714, "time": 0.6370244503021241, "epoch": 7, "memory": 44140, "step": 26866} +{"lr": 3.382791135940194e-05, "data_time": 0.0017197132110595703, "loss": 0.4885852813720703, "time": 0.6382949590682984, "epoch": 7, "memory": 44140, "step": 26966} +{"lr": 3.35250855534377e-05, "data_time": 0.001477670669555664, "loss": 0.4552358388900757, "time": 0.6335381746292115, "epoch": 7, "memory": 44140, "step": 27066} +{"lr": 3.3223513853112476e-05, "data_time": 0.0014629125595092773, "loss": 0.4468356966972351, "time": 0.6357050657272338, "epoch": 7, "memory": 44140, "step": 27166} +{"lr": 3.292321386980115e-05, "data_time": 0.0015483379364013671, "loss": 0.4653731107711792, "time": 0.6372542142868042, "epoch": 7, "memory": 44140, "step": 27266} +{"lr": 3.262420314061217e-05, "data_time": 0.0023773908615112305, "loss": 0.45212759971618655, "time": 0.6342232704162598, "epoch": 7, "memory": 44140, "step": 27366} +{"lr": 3.2326499127363195e-05, "data_time": 0.0016300201416015625, "loss": 0.47584818601608275, "time": 0.6367329359054565, "epoch": 7, "memory": 44140, "step": 27466} +{"lr": 3.2030119215561613e-05, "data_time": 0.0014003515243530273, "loss": 0.4543128728866577, "time": 0.6339415073394775, "epoch": 7, "memory": 44140, "step": 27566} +{"lr": 3.173508071338896e-05, "data_time": 0.0016157865524291993, "loss": 0.44322863817214964, "time": 0.6338517665863037, "epoch": 7, "memory": 44140, "step": 27666} +{"lr": 3.144140085069037e-05, "data_time": 0.0016385555267333985, "loss": 0.463171112537384, "time": 0.6348424673080444, "epoch": 7, "memory": 44140, "step": 27766} +{"lr": 3.114909677796814e-05, "data_time": 0.0015558481216430664, "loss": 0.44728931188583376, "time": 0.6365219831466675, "epoch": 7, "memory": 44140, "step": 27866} +{"lr": 3.085818556538059e-05, "data_time": 0.0014888286590576173, "loss": 0.46412636041641236, "time": 0.6337711095809937, "epoch": 7, "memory": 44140, "step": 27966} +{"lr": 3.056868420174455e-05, "data_time": 0.0017244338989257813, "loss": 0.48188812732696534, "time": 0.634553861618042, "epoch": 7, "memory": 44140, "step": 28066} +{"lr": 3.028060959354392e-05, "data_time": 0.0015617609024047852, "loss": 0.48123701810836794, "time": 0.6338025331497192, "epoch": 7, "memory": 44140, "step": 28166} +{"lr": 2.9993978563941876e-05, "data_time": 0.001407027244567871, "loss": 0.4762677103281021, "time": 0.637341046333313, "epoch": 7, "memory": 44140, "step": 28266} +{"lr": 2.9708807851798587e-05, "data_time": 0.0017408370971679688, "loss": 0.4819815993309021, "time": 0.636882209777832, "epoch": 7, "memory": 44140, "step": 28366} +{"lr": 2.9425114110693717e-05, "data_time": 0.001581263542175293, "loss": 0.4441631197929382, "time": 0.6356881141662598, "epoch": 7, "memory": 44140, "step": 28466} +{"lr": 2.9142913907953844e-05, "data_time": 0.0019063472747802735, "loss": 0.4636713624000549, "time": 0.6350440740585327, "epoch": 7, "memory": 44140, "step": 28566} +{"lr": 2.8862223723684912e-05, "data_time": 0.0015549421310424804, "loss": 0.46156834065914154, "time": 0.6346679925918579, "epoch": 7, "memory": 44140, "step": 28666} +{"lr": 2.8583059949809854e-05, "data_time": 0.002940654754638672, "loss": 0.4637675046920776, "time": 0.6384005784988404, "epoch": 7, "memory": 44140, "step": 28766} +{"accuracy/top1": 89.14817810058594, "data_time": 0.01785958210627238, "time": 1.2793065150578817, "step": 7} +{"lr": 2.827499545208893e-05, "data_time": 0.0014168500900268554, "loss": 0.46368653178215025, "time": 0.6343302726745605, "epoch": 8, "memory": 44140, "step": 28877} +{"lr": 2.7999105784987406e-05, "data_time": 0.0018784523010253907, "loss": 0.46099224090576174, "time": 0.6338589191436768, "epoch": 8, "memory": 44140, "step": 28977} +{"lr": 2.772479293318711e-05, "data_time": 0.0015350341796875, "loss": 0.4673727661371231, "time": 0.6351502656936645, "epoch": 8, "memory": 44140, "step": 29077} +{"lr": 2.7452072916183603e-05, "data_time": 0.0015818119049072266, "loss": 0.47864570617675783, "time": 0.6330815553665161, "epoch": 8, "memory": 44140, "step": 29177} +{"lr": 2.7180961660452883e-05, "data_time": 0.0015813827514648438, "loss": 0.4577386975288391, "time": 0.6341041326522827, "epoch": 8, "memory": 44140, "step": 29277} +{"lr": 2.6911474998521622e-05, "data_time": 0.0013978004455566407, "loss": 0.5009500443935394, "time": 0.6377395868301392, "epoch": 8, "memory": 44140, "step": 29377} +{"lr": 2.664362866804239e-05, "data_time": 0.002186417579650879, "loss": 0.46405903697013856, "time": 0.6343747854232789, "epoch": 8, "memory": 44140, "step": 29477} +{"lr": 2.637743831087459e-05, "data_time": 0.0020719051361083986, "loss": 0.4770909249782562, "time": 0.6331545829772949, "epoch": 8, "memory": 44140, "step": 29577} +{"lr": 2.6112919472171114e-05, "data_time": 0.002500581741333008, "loss": 0.4557668209075928, "time": 0.6330735683441162, "epoch": 8, "memory": 44140, "step": 29677} +{"lr": 2.5850087599470434e-05, "data_time": 0.0016175031661987305, "loss": 0.4631189674139023, "time": 0.6367992401123047, "epoch": 8, "memory": 44140, "step": 29777} +{"lr": 2.5588958041794503e-05, "data_time": 0.0018662691116333007, "loss": 0.45991411805152893, "time": 0.6337423086166382, "epoch": 8, "memory": 44140, "step": 29877} +{"lr": 2.5329546048752453e-05, "data_time": 0.002153301239013672, "loss": 0.4490981578826904, "time": 0.633574390411377, "epoch": 8, "memory": 44140, "step": 29977} +{"lr": 2.5071866769649806e-05, "data_time": 0.0016371726989746094, "loss": 0.4482205629348755, "time": 0.6342392206192017, "epoch": 8, "memory": 44140, "step": 30077} +{"lr": 2.4815935252604222e-05, "data_time": 0.001477670669555664, "loss": 0.4637909531593323, "time": 0.6326849460601807, "epoch": 8, "memory": 44140, "step": 30177} +{"lr": 2.4561766443666384e-05, "data_time": 0.0015606641769409179, "loss": 0.455281537771225, "time": 0.6346306800842285, "epoch": 8, "memory": 44140, "step": 30277} +{"lr": 2.4309375185947164e-05, "data_time": 0.001670694351196289, "loss": 0.42819930613040924, "time": 0.6330586910247803, "epoch": 8, "memory": 44140, "step": 30377} +{"lr": 2.40587762187509e-05, "data_time": 0.0019073247909545898, "loss": 0.4765880346298218, "time": 0.633723521232605, "epoch": 8, "memory": 44140, "step": 30477} +{"lr": 2.3809984176714634e-05, "data_time": 0.0016132354736328124, "loss": 0.47656781077384947, "time": 0.634052038192749, "epoch": 8, "memory": 44140, "step": 30577} +{"lr": 2.3563013588953406e-05, "data_time": 0.0016823053359985352, "loss": 0.45848439931869506, "time": 0.6352833986282349, "epoch": 8, "memory": 44140, "step": 30677} +{"lr": 2.331787887821182e-05, "data_time": 0.0014740228652954102, "loss": 0.46513075828552247, "time": 0.6356109619140625, "epoch": 8, "memory": 44140, "step": 30777} +{"lr": 2.307459436002179e-05, "data_time": 0.002760791778564453, "loss": 0.47583167552947997, "time": 0.633095121383667, "epoch": 8, "memory": 44140, "step": 30877} +{"lr": 2.2833174241866527e-05, "data_time": 0.002381706237792969, "loss": 0.45027904510498046, "time": 0.6363095998764038, "epoch": 8, "memory": 44140, "step": 30977} +{"lr": 2.259363262235086e-05, "data_time": 0.0022494316101074217, "loss": 0.4671119451522827, "time": 0.6359816789627075, "epoch": 8, "memory": 44140, "step": 31077} +{"lr": 2.2355983490377735e-05, "data_time": 0.0015765666961669923, "loss": 0.43791429698467255, "time": 0.6359403133392334, "epoch": 8, "memory": 44140, "step": 31177} +{"lr": 2.2120240724331546e-05, "data_time": 0.0018905401229858398, "loss": 0.456916469335556, "time": 0.6331723928451538, "epoch": 8, "memory": 44140, "step": 31277} +{"lr": 2.1886418091267445e-05, "data_time": 0.0018166542053222657, "loss": 0.46086585223674775, "time": 0.6346737384796143, "epoch": 8, "memory": 44140, "step": 31377} +{"lr": 2.1654529246107496e-05, "data_time": 0.0015304803848266602, "loss": 0.47067301273345946, "time": 0.6341936826705933, "epoch": 8, "memory": 44140, "step": 31477} +{"lr": 2.1424587730843184e-05, "data_time": 0.0015278100967407227, "loss": 0.4529714584350586, "time": 0.6375127077102661, "epoch": 8, "memory": 44140, "step": 31577} +{"lr": 2.1196606973744593e-05, "data_time": 0.0016810178756713867, "loss": 0.4490090847015381, "time": 0.6349388837814331, "epoch": 8, "memory": 44140, "step": 31677} +{"lr": 2.0970600288576148e-05, "data_time": 0.0018979787826538086, "loss": 0.4901163220405579, "time": 0.633464765548706, "epoch": 8, "memory": 44140, "step": 31777} +{"lr": 2.0746580873819274e-05, "data_time": 0.0013696908950805663, "loss": 0.46635947823524476, "time": 0.6334488153457641, "epoch": 8, "memory": 44140, "step": 31877} +{"lr": 2.0524561811901426e-05, "data_time": 0.0016888856887817382, "loss": 0.44515493512153625, "time": 0.6334747791290283, "epoch": 8, "memory": 44140, "step": 31977} +{"lr": 2.0304556068432268e-05, "data_time": 0.0016237258911132812, "loss": 0.46383466124534606, "time": 0.6341516494750976, "epoch": 8, "memory": 44140, "step": 32077} +{"lr": 2.0086576491446333e-05, "data_time": 0.0014827728271484375, "loss": 0.46970545649528506, "time": 0.6342121124267578, "epoch": 8, "memory": 44140, "step": 32177} +{"lr": 1.987063581065291e-05, "data_time": 0.0016895532608032227, "loss": 0.4618035197257996, "time": 0.6343858242034912, "epoch": 8, "memory": 44140, "step": 32277} +{"lr": 1.965674663669245e-05, "data_time": 0.0016300201416015625, "loss": 0.49416699409484866, "time": 0.6343100309371948, "epoch": 8, "memory": 44140, "step": 32377} +{"lr": 1.9444921460400247e-05, "data_time": 0.0016714572906494141, "loss": 0.4498171627521515, "time": 0.6342238664627076, "epoch": 8, "memory": 44140, "step": 32477} +{"lr": 1.923517265207697e-05, "data_time": 0.0022983312606811523, "loss": 0.4613679885864258, "time": 0.6332913398742676, "epoch": 8, "memory": 44140, "step": 32577} +{"lr": 1.9027512460766204e-05, "data_time": 0.002181529998779297, "loss": 0.46468991935253146, "time": 0.640119194984436, "epoch": 8, "memory": 44140, "step": 32677} +{"lr": 1.8821953013539136e-05, "data_time": 0.001618170738220215, "loss": 0.46930948495864866, "time": 0.6349443912506103, "epoch": 8, "memory": 44140, "step": 32777} +{"lr": 1.8618506314786433e-05, "data_time": 0.003096127510070801, "loss": 0.47440574765205384, "time": 0.6349209785461426, "epoch": 8, "memory": 44140, "step": 32877} +{"accuracy/top1": 89.2044448852539, "data_time": 0.017275114854176838, "time": 1.2790930946667989, "step": 8} +{"lr": 1.8395169032542028e-05, "data_time": 0.0014571428298950195, "loss": 0.45794283151626586, "time": 0.6334787845611572, "epoch": 9, "memory": 44140, "step": 32988} +{"lr": 1.819621906460573e-05, "data_time": 0.0018780231475830078, "loss": 0.47048654556274416, "time": 0.6338154077529907, "epoch": 9, "memory": 44140, "step": 33088} +{"lr": 1.799941838715077e-05, "data_time": 0.001538395881652832, "loss": 0.4661583721637726, "time": 0.6356528043746948, "epoch": 9, "memory": 44140, "step": 33188} +{"lr": 1.7804778493067625e-05, "data_time": 0.0015492200851440429, "loss": 0.47302921414375304, "time": 0.6382202625274658, "epoch": 9, "memory": 44140, "step": 33288} +{"lr": 1.761231074905996e-05, "data_time": 0.0014631032943725586, "loss": 0.46851676106452944, "time": 0.6360684633255005, "epoch": 9, "memory": 44140, "step": 33388} +{"lr": 1.742202639498083e-05, "data_time": 0.001425457000732422, "loss": 0.45815710425376893, "time": 0.6349555015563965, "epoch": 9, "memory": 44140, "step": 33488} +{"lr": 1.7233936543176347e-05, "data_time": 0.002461886405944824, "loss": 0.46372724175453184, "time": 0.6354443073272705, "epoch": 9, "memory": 44140, "step": 33588} +{"lr": 1.704805217783665e-05, "data_time": 0.001511859893798828, "loss": 0.44393640756607056, "time": 0.6344957113265991, "epoch": 9, "memory": 44140, "step": 33688} +{"lr": 1.68643841543545e-05, "data_time": 0.0014844655990600586, "loss": 0.4598296880722046, "time": 0.6378105640411377, "epoch": 9, "memory": 44140, "step": 33788} +{"lr": 1.6682943198691327e-05, "data_time": 0.0017051458358764648, "loss": 0.45735839009284973, "time": 0.6351711273193359, "epoch": 9, "memory": 44140, "step": 33888} +{"lr": 1.6503739906750822e-05, "data_time": 0.0016067981719970702, "loss": 0.46009152233600614, "time": 0.6342929840087891, "epoch": 9, "memory": 44140, "step": 33988} +{"lr": 1.6326784743760187e-05, "data_time": 0.0014171838760375977, "loss": 0.4574501931667328, "time": 0.6351300477981567, "epoch": 9, "memory": 44140, "step": 34088} +{"lr": 1.6152088043659013e-05, "data_time": 0.001588892936706543, "loss": 0.47189466953277587, "time": 0.6352554082870483, "epoch": 9, "memory": 44140, "step": 34188} +{"lr": 1.59796600084957e-05, "data_time": 0.0015816926956176759, "loss": 0.4594050168991089, "time": 0.6334317445755004, "epoch": 9, "memory": 44140, "step": 34288} +{"lr": 1.5809510707831687e-05, "data_time": 0.0015261173248291016, "loss": 0.4613716512918472, "time": 0.6336474418640137, "epoch": 9, "memory": 44140, "step": 34388} +{"lr": 1.5641650078153473e-05, "data_time": 0.0014529705047607421, "loss": 0.4630406856536865, "time": 0.6344540119171143, "epoch": 9, "memory": 44140, "step": 34488} +{"lr": 1.547608792229235e-05, "data_time": 0.0016174793243408203, "loss": 0.4735118567943573, "time": 0.6345244884490967, "epoch": 9, "memory": 44140, "step": 34588} +{"lr": 1.5312833908851794e-05, "data_time": 0.0015510320663452148, "loss": 0.4545616626739502, "time": 0.6343519210815429, "epoch": 9, "memory": 44140, "step": 34688} +{"lr": 1.5151897571642958e-05, "data_time": 0.0015728473663330078, "loss": 0.437881326675415, "time": 0.6345327138900757, "epoch": 9, "memory": 44140, "step": 34788} +{"lr": 1.4993288309127877e-05, "data_time": 0.0014739990234375, "loss": 0.47203379273414614, "time": 0.6343764305114746, "epoch": 9, "memory": 44140, "step": 34888} +{"lr": 1.4837015383870593e-05, "data_time": 0.0014815330505371094, "loss": 0.45972618758678435, "time": 0.635708999633789, "epoch": 9, "memory": 44140, "step": 34988} +{"lr": 1.4683087921996264e-05, "data_time": 0.0016397953033447266, "loss": 0.4681494772434235, "time": 0.635512399673462, "epoch": 9, "memory": 44140, "step": 35088} +{"lr": 1.453151491265819e-05, "data_time": 0.0014656543731689452, "loss": 0.45016089677810667, "time": 0.6344328880310058, "epoch": 9, "memory": 44140, "step": 35188} +{"lr": 1.4382305207512847e-05, "data_time": 0.0014738321304321289, "loss": 0.4790822803974152, "time": 0.6335061311721801, "epoch": 9, "memory": 44140, "step": 35288} +{"lr": 1.423546752020298e-05, "data_time": 0.0015030384063720703, "loss": 0.4548185706138611, "time": 0.6337298393249512, "epoch": 9, "memory": 44140, "step": 35388} +{"lr": 1.4091010425848762e-05, "data_time": 0.0015840530395507812, "loss": 0.4755293786525726, "time": 0.6353351354599, "epoch": 9, "memory": 44140, "step": 35488} +{"lr": 1.3948942360546966e-05, "data_time": 0.001415586471557617, "loss": 0.46705070734024046, "time": 0.6347707986831665, "epoch": 9, "memory": 44140, "step": 35588} +{"lr": 1.3809271620878346e-05, "data_time": 0.001620006561279297, "loss": 0.45487709641456603, "time": 0.6343161821365356, "epoch": 9, "memory": 44140, "step": 35688} +{"lr": 1.3672006363423091e-05, "data_time": 0.0024388313293457033, "loss": 0.47723528146743777, "time": 0.6360401391983033, "epoch": 9, "memory": 44140, "step": 35788} +{"lr": 1.3537154604284566e-05, "data_time": 0.0016469240188598632, "loss": 0.46806180775165557, "time": 0.6335711717605591, "epoch": 9, "memory": 44140, "step": 35888} +{"lr": 1.3404724218621087e-05, "data_time": 0.0015513896942138672, "loss": 0.4778871476650238, "time": 0.6358850955963135, "epoch": 9, "memory": 44140, "step": 35988} +{"lr": 1.3274722940186121e-05, "data_time": 0.0015336990356445313, "loss": 0.43152810335159303, "time": 0.6338732481002808, "epoch": 9, "memory": 44140, "step": 36088} +{"lr": 1.3147158360876541e-05, "data_time": 0.0017436742782592773, "loss": 0.4663323938846588, "time": 0.634447169303894, "epoch": 9, "memory": 44140, "step": 36188} +{"lr": 1.3022037930289352e-05, "data_time": 0.0014832019805908203, "loss": 0.46402403712272644, "time": 0.6352090835571289, "epoch": 9, "memory": 44140, "step": 36288} +{"lr": 1.2899368955286627e-05, "data_time": 0.001495528221130371, "loss": 0.4838336706161499, "time": 0.6343191146850586, "epoch": 9, "memory": 44140, "step": 36388} +{"lr": 1.2779158599568751e-05, "data_time": 0.001556730270385742, "loss": 0.4730865597724915, "time": 0.6341584205627442, "epoch": 9, "memory": 44140, "step": 36488} +{"lr": 1.2661413883256134e-05, "data_time": 0.0017279386520385742, "loss": 0.4540800988674164, "time": 0.636177945137024, "epoch": 9, "memory": 44140, "step": 36588} +{"lr": 1.2546141682479203e-05, "data_time": 0.0018882274627685547, "loss": 0.46962815523147583, "time": 0.634635066986084, "epoch": 9, "memory": 44140, "step": 36688} +{"lr": 1.2433348728976846e-05, "data_time": 0.0015498638153076173, "loss": 0.4489203214645386, "time": 0.6349020719528198, "epoch": 9, "memory": 44140, "step": 36788} +{"lr": 1.2323041609703344e-05, "data_time": 0.0015301465988159179, "loss": 0.4627111107110977, "time": 0.6345993280410767, "epoch": 9, "memory": 44140, "step": 36888} +{"lr": 1.2215226766443614e-05, "data_time": 0.003014397621154785, "loss": 0.4508346140384674, "time": 0.6346036672592164, "epoch": 9, "memory": 44140, "step": 36988} +{"accuracy/top1": 89.29381561279297, "data_time": 0.017911229530970255, "time": 1.279884970188141, "step": 9} +{"lr": 1.209847850894437e-05, "data_time": 0.0014863014221191406, "loss": 0.4583283841609955, "time": 0.6364388465881348, "epoch": 10, "memory": 44140, "step": 37099} +{"lr": 1.1995942849201055e-05, "data_time": 0.0015822887420654298, "loss": 0.4592619836330414, "time": 0.6366263866424561, "epoch": 10, "memory": 44140, "step": 37199} +{"lr": 1.189591856759183e-05, "data_time": 0.00153656005859375, "loss": 0.44315301179885863, "time": 0.6373314380645752, "epoch": 10, "memory": 44140, "step": 37299} +{"lr": 1.1798411505398016e-05, "data_time": 0.0019263982772827148, "loss": 0.45604652166366577, "time": 0.63377845287323, "epoch": 10, "memory": 44140, "step": 37399} +{"lr": 1.1703427356898678e-05, "data_time": 0.0014147520065307616, "loss": 0.4433305740356445, "time": 0.6351155519485474, "epoch": 10, "memory": 44140, "step": 37499} +{"lr": 1.1610971669038223e-05, "data_time": 0.0014008760452270507, "loss": 0.45763781666755676, "time": 0.6348682403564453, "epoch": 10, "memory": 44140, "step": 37599} +{"lr": 1.1521049841102396e-05, "data_time": 0.001594209671020508, "loss": 0.45304534435272215, "time": 0.6338958501815796, "epoch": 10, "memory": 44140, "step": 37699} +{"lr": 1.1433667124402997e-05, "data_time": 0.0023214340209960936, "loss": 0.4675642132759094, "time": 0.633301067352295, "epoch": 10, "memory": 44140, "step": 37799} +{"lr": 1.1348828621971181e-05, "data_time": 0.0015839815139770507, "loss": 0.43786188364028933, "time": 0.6339093446731567, "epoch": 10, "memory": 44140, "step": 37899} +{"lr": 1.1266539288259486e-05, "data_time": 0.0017474889755249023, "loss": 0.45742044448852537, "time": 0.6358333826065063, "epoch": 10, "memory": 44140, "step": 37999} +{"lr": 1.1186803928852477e-05, "data_time": 0.0023247241973876954, "loss": 0.4865990221500397, "time": 0.634549617767334, "epoch": 10, "memory": 44140, "step": 38099} +{"lr": 1.1109627200186125e-05, "data_time": 0.0015635967254638671, "loss": 0.48159350752830504, "time": 0.6335470676422119, "epoch": 10, "memory": 44140, "step": 38199} +{"lr": 1.1035013609275835e-05, "data_time": 0.0018955469131469727, "loss": 0.4572065591812134, "time": 0.6327837467193603, "epoch": 10, "memory": 44140, "step": 38299} +{"lr": 1.0962967513453298e-05, "data_time": 0.001545119285583496, "loss": 0.47677528858184814, "time": 0.6344039440155029, "epoch": 10, "memory": 44140, "step": 38399} +{"lr": 1.0893493120111999e-05, "data_time": 0.0015576601028442383, "loss": 0.49043325185775755, "time": 0.6342111110687256, "epoch": 10, "memory": 44140, "step": 38499} +{"lr": 1.0826594486461509e-05, "data_time": 0.002152705192565918, "loss": 0.4628440201282501, "time": 0.6397070407867431, "epoch": 10, "memory": 44140, "step": 38599} +{"lr": 1.0762275519290557e-05, "data_time": 0.0016562938690185547, "loss": 0.42665415406227114, "time": 0.6332319498062133, "epoch": 10, "memory": 44140, "step": 38699} +{"lr": 1.0700539974738862e-05, "data_time": 0.0019118309020996094, "loss": 0.4881063044071198, "time": 0.6334288120269775, "epoch": 10, "memory": 44140, "step": 38799} +{"lr": 1.064139145807783e-05, "data_time": 0.0016589641571044921, "loss": 0.47498711943626404, "time": 0.63395094871521, "epoch": 10, "memory": 44140, "step": 38899} +{"lr": 1.0584833423499947e-05, "data_time": 0.0015327215194702148, "loss": 0.5052730917930603, "time": 0.6363633632659912, "epoch": 10, "memory": 44140, "step": 38999} +{"lr": 1.0530869173917099e-05, "data_time": 0.001548910140991211, "loss": 0.4437434017658234, "time": 0.6339998006820678, "epoch": 10, "memory": 44140, "step": 39099} +{"lr": 1.0479501860767654e-05, "data_time": 0.001680135726928711, "loss": 0.47136734426021576, "time": 0.6326848983764648, "epoch": 10, "memory": 44140, "step": 39199} +{"lr": 1.043073448383247e-05, "data_time": 0.0018276214599609376, "loss": 0.4749999761581421, "time": 0.6348501920700074, "epoch": 10, "memory": 44140, "step": 39299} +{"lr": 1.038456989105968e-05, "data_time": 0.0016482114791870118, "loss": 0.4722422778606415, "time": 0.6353384256362915, "epoch": 10, "memory": 44140, "step": 39399} +{"lr": 1.034101077839838e-05, "data_time": 0.0015259504318237305, "loss": 0.4292121887207031, "time": 0.634094500541687, "epoch": 10, "memory": 44140, "step": 39499} +{"lr": 1.0300059689641194e-05, "data_time": 0.0015766143798828125, "loss": 0.46753338277339934, "time": 0.6351865053176879, "epoch": 10, "memory": 44140, "step": 39599} +{"lr": 1.0261719016275683e-05, "data_time": 0.0014925479888916015, "loss": 0.47564561367034913, "time": 0.6352946758270264, "epoch": 10, "memory": 44140, "step": 39699} +{"lr": 1.0225990997344757e-05, "data_time": 0.0018812656402587891, "loss": 0.4693823516368866, "time": 0.636930513381958, "epoch": 10, "memory": 44140, "step": 39799} +{"lr": 1.019287771931587e-05, "data_time": 0.001675868034362793, "loss": 0.468352347612381, "time": 0.635081434249878, "epoch": 10, "memory": 44140, "step": 39899} +{"lr": 1.0162381115959187e-05, "data_time": 0.00176546573638916, "loss": 0.47879849672317504, "time": 0.6356399059295654, "epoch": 10, "memory": 44140, "step": 39999} +{"lr": 1.0134502968234636e-05, "data_time": 0.0015333175659179687, "loss": 0.47510187029838563, "time": 0.6390942096710205, "epoch": 10, "memory": 44140, "step": 40099} +{"lr": 1.0109244904187918e-05, "data_time": 0.001466655731201172, "loss": 0.5071095883846283, "time": 0.6355346441268921, "epoch": 10, "memory": 44140, "step": 40199} +{"lr": 1.0086608398855446e-05, "data_time": 0.0016642570495605468, "loss": 0.4606593608856201, "time": 0.6338963031768798, "epoch": 10, "memory": 44140, "step": 40299} +{"lr": 1.0066594774178176e-05, "data_time": 0.0016315937042236327, "loss": 0.4885964959859848, "time": 0.6331888914108277, "epoch": 10, "memory": 44140, "step": 40399} +{"lr": 1.0049205198924413e-05, "data_time": 0.0015462875366210938, "loss": 0.4587945997714996, "time": 0.6341197729110718, "epoch": 10, "memory": 44140, "step": 40499} +{"lr": 1.0034440688621577e-05, "data_time": 0.0015654802322387696, "loss": 0.46378846764564513, "time": 0.6347687482833863, "epoch": 10, "memory": 44140, "step": 40599} +{"lr": 1.0022302105496896e-05, "data_time": 0.0015238523483276367, "loss": 0.481249725818634, "time": 0.6357029676437378, "epoch": 10, "memory": 44140, "step": 40699} +{"lr": 1.0012790158427017e-05, "data_time": 0.013905096054077148, "loss": 0.48698058128356936, "time": 0.6465101718902588, "epoch": 10, "memory": 44140, "step": 40799} +{"lr": 1.000590540289662e-05, "data_time": 0.0017086982727050782, "loss": 0.4825940251350403, "time": 0.6361988306045532, "epoch": 10, "memory": 44140, "step": 40899} +{"lr": 1.0001648240966046e-05, "data_time": 0.002201271057128906, "loss": 0.49574007987976076, "time": 0.6367281436920166, "epoch": 10, "memory": 44140, "step": 40999} +{"lr": 1.0000018921247718e-05, "data_time": 0.0031217336654663086, "loss": 0.4519974410533905, "time": 0.6363741159439087, "epoch": 10, "memory": 44140, "step": 41099} +{"accuracy/top1": 89.44441986083984, "data_time": 0.01780688762664795, "time": 1.279326872030894, "step": 10} diff --git a/clip_large_pretrain_4x256_stylegan3_lr1e-4/20230605_145625/vis_data/config.py b/clip_large_pretrain_4x256_stylegan3_lr1e-4/20230605_145625/vis_data/config.py new file mode 100644 index 0000000000000000000000000000000000000000..fcc09894d5b0708c1c12dca7f367c370989270ef --- /dev/null +++ b/clip_large_pretrain_4x256_stylegan3_lr1e-4/20230605_145625/vis_data/config.py @@ -0,0 +1,356 @@ +optim_wrapper = dict( + optimizer=dict( + type='AdamW', lr=0.0001, weight_decay=0.3, _scope_='mmpretrain'), + paramwise_cfg=dict( + custom_keys=dict({ + '.cls_token': dict(decay_mult=0.0), + '.pos_embed': dict(decay_mult=0.0) + })), + type='AmpOptimWrapper', + dtype='bfloat16', + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=False, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=4096) +model = dict( + type='ImageClassifier', + backbone=dict( + frozen_stages=24, + type='VisionTransformer', + arch='l', + img_size=224, + patch_size=14, + drop_rate=0.1, + pre_norm=True, + final_norm=False, + init_cfg=dict( + type='Pretrained', + checkpoint='ckpt/openclip-ViT-L-14.pth', + prefix='backbone')), + neck=dict( + type='CLIPProjection', + in_channels=1024, + out_channels=768, + init_cfg=dict( + type='Pretrained', + checkpoint='ckpt/openclip-ViT-L-14.pth', + prefix='backbone')), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=768, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + init_cfg=None), + init_cfg=dict( + type='TruncNormal', layer=['Conv2d', 'Linear'], std=0.02, bias=0.0), + train_cfg=None) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=128, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/all1.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/all2.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/all3.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3fake8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/cc1m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3real8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=128, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV2-1-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/if-dpmsolver++-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=128, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV2-1-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/if-dpmsolver++-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/clip_large_pretrain_4x256_stylegan3_lr1e-4' diff --git a/clip_large_pretrain_4x256_stylegan3_lr1e-4/20230605_145625/vis_data/events.out.tfevents.1685948199.SH-IDC1-10-140-24-103.206295.0 b/clip_large_pretrain_4x256_stylegan3_lr1e-4/20230605_145625/vis_data/events.out.tfevents.1685948199.SH-IDC1-10-140-24-103.206295.0 new file mode 100644 index 0000000000000000000000000000000000000000..22b92c06d5edd62fd05421c9d9db3ae7ebe57811 --- /dev/null +++ b/clip_large_pretrain_4x256_stylegan3_lr1e-4/20230605_145625/vis_data/events.out.tfevents.1685948199.SH-IDC1-10-140-24-103.206295.0 @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:79727c97db3625041575f735252511f0e807e9929ebe2a3cf794e1bfc9588a49 +size 2992949 diff --git a/clip_large_pretrain_4x256_stylegan3_lr1e-4/20230605_145625/vis_data/scalars.json b/clip_large_pretrain_4x256_stylegan3_lr1e-4/20230605_145625/vis_data/scalars.json new file mode 100644 index 0000000000000000000000000000000000000000..d4c0de8dafed7889fdf44dfd2a12da71f06c1a52 --- /dev/null +++ b/clip_large_pretrain_4x256_stylegan3_lr1e-4/20230605_145625/vis_data/scalars.json @@ -0,0 +1,420 @@ +{"lr": 9.999871217862936e-05, "data_time": 0.0015280723571777343, "loss": 0.5829696655273438, "time": 0.6316387891769409, "epoch": 1, "memory": 44139, "step": 100} +{"lr": 9.999479662539605e-05, "data_time": 0.0016149282455444336, "loss": 0.5629205703735352, "time": 0.6347221851348877, "epoch": 1, "memory": 44139, "step": 200} +{"lr": 9.99882534375654e-05, "data_time": 0.0016250133514404297, "loss": 0.5592653274536132, "time": 0.632369875907898, "epoch": 1, "memory": 44139, "step": 300} +{"lr": 9.997908299725075e-05, "data_time": 0.0015004873275756836, "loss": 0.5522704601287842, "time": 0.6337758302688599, "epoch": 1, "memory": 44139, "step": 400} +{"lr": 9.996728583999318e-05, "data_time": 0.0015186309814453126, "loss": 0.5318122386932373, "time": 0.6317744255065918, "epoch": 1, "memory": 44139, "step": 500} +{"lr": 9.995286265473058e-05, "data_time": 0.0017360925674438476, "loss": 0.5652305126190186, "time": 0.6330352783203125, "epoch": 1, "memory": 44139, "step": 600} +{"lr": 9.993581428375725e-05, "data_time": 0.001550602912902832, "loss": 0.5028891801834107, "time": 0.6331093072891235, "epoch": 1, "memory": 44139, "step": 700} +{"lr": 9.991614172267466e-05, "data_time": 0.001519465446472168, "loss": 0.5037070035934448, "time": 0.6332561016082764, "epoch": 1, "memory": 44139, "step": 800} +{"lr": 9.989384612033355e-05, "data_time": 0.0014886856079101562, "loss": 0.5305287957191467, "time": 0.6337663888931274, "epoch": 1, "memory": 44139, "step": 900} +{"lr": 9.986892877876658e-05, "data_time": 0.0013533830642700195, "loss": 0.5345773220062255, "time": 0.6327988624572753, "epoch": 1, "memory": 44139, "step": 1000} +{"lr": 9.984139115311244e-05, "data_time": 0.0015159368515014649, "loss": 0.5430529594421387, "time": 0.6334908246994019, "epoch": 1, "memory": 44139, "step": 1100} +{"lr": 9.981123485153081e-05, "data_time": 0.0015604257583618163, "loss": 0.550173544883728, "time": 0.6322064161300659, "epoch": 1, "memory": 44139, "step": 1200} +{"lr": 9.977846163510851e-05, "data_time": 0.0014005184173583984, "loss": 0.4987194061279297, "time": 0.6331788063049316, "epoch": 1, "memory": 44139, "step": 1300} +{"lr": 9.974307341775653e-05, "data_time": 0.0016389608383178711, "loss": 0.5176548004150391, "time": 0.6326443910598755, "epoch": 1, "memory": 44139, "step": 1400} +{"lr": 9.970507226609833e-05, "data_time": 0.0014583349227905273, "loss": 0.49719810485839844, "time": 0.6342327356338501, "epoch": 1, "memory": 44139, "step": 1500} +{"lr": 9.966446039934934e-05, "data_time": 0.0014329671859741211, "loss": 0.5203955173492432, "time": 0.6341060400009155, "epoch": 1, "memory": 44139, "step": 1600} +{"lr": 9.962124018918683e-05, "data_time": 0.0013983011245727538, "loss": 0.5304009437561035, "time": 0.6350225448608399, "epoch": 1, "memory": 44139, "step": 1700} +{"lr": 9.957541415961212e-05, "data_time": 0.0013764858245849609, "loss": 0.49931039810180666, "time": 0.6334275484085083, "epoch": 1, "memory": 44139, "step": 1800} +{"lr": 9.952698498680267e-05, "data_time": 0.0013906240463256836, "loss": 0.5129714608192444, "time": 0.6329006910324096, "epoch": 1, "memory": 44139, "step": 1900} +{"lr": 9.947595549895583e-05, "data_time": 0.0014127492904663086, "loss": 0.48182153701782227, "time": 0.63444082736969, "epoch": 1, "memory": 44139, "step": 2000} +{"lr": 9.942232867612417e-05, "data_time": 0.001363086700439453, "loss": 0.4994787693023682, "time": 0.6327764987945557, "epoch": 1, "memory": 44139, "step": 2100} +{"lr": 9.93661076500406e-05, "data_time": 0.0014455795288085937, "loss": 0.5223768472671508, "time": 0.632605504989624, "epoch": 1, "memory": 44139, "step": 2200} +{"lr": 9.930729570393622e-05, "data_time": 0.0014699697494506836, "loss": 0.5021018862724305, "time": 0.6406325101852417, "epoch": 1, "memory": 44139, "step": 2300} +{"lr": 9.924589627234824e-05, "data_time": 0.0019201278686523438, "loss": 0.5076982319355011, "time": 0.6326788187026977, "epoch": 1, "memory": 44139, "step": 2400} +{"lr": 9.918191294091956e-05, "data_time": 0.0014031887054443359, "loss": 0.5193401515483856, "time": 0.6322619199752808, "epoch": 1, "memory": 44139, "step": 2500} +{"lr": 9.91153494461892e-05, "data_time": 0.0015448808670043945, "loss": 0.4991978645324707, "time": 0.6341390609741211, "epoch": 1, "memory": 44139, "step": 2600} +{"lr": 9.904620967537422e-05, "data_time": 0.0015372753143310547, "loss": 0.49636522531509397, "time": 0.6350481271743774, "epoch": 1, "memory": 44139, "step": 2700} +{"lr": 9.897449766614277e-05, "data_time": 0.0016783714294433595, "loss": 0.5006751537322998, "time": 0.6336184740066528, "epoch": 1, "memory": 44139, "step": 2800} +{"lr": 9.890021760637804e-05, "data_time": 0.0014549493789672852, "loss": 0.5036589443683624, "time": 0.6338624715805053, "epoch": 1, "memory": 44139, "step": 2900} +{"lr": 9.882337383393409e-05, "data_time": 0.001423048973083496, "loss": 0.4870572656393051, "time": 0.6328262329101563, "epoch": 1, "memory": 44139, "step": 3000} +{"lr": 9.874397083638193e-05, "data_time": 0.001641249656677246, "loss": 0.5131533741950989, "time": 0.6331220149993897, "epoch": 1, "memory": 44139, "step": 3100} +{"lr": 9.866201325074828e-05, "data_time": 0.0014507770538330078, "loss": 0.48008095026016234, "time": 0.6418931007385253, "epoch": 1, "memory": 44139, "step": 3200} +{"lr": 9.857750586324393e-05, "data_time": 0.001474308967590332, "loss": 0.480618292093277, "time": 0.6352507829666137, "epoch": 1, "memory": 44139, "step": 3300} +{"lr": 9.84904536089848e-05, "data_time": 0.0013846158981323242, "loss": 0.4827093005180359, "time": 0.6340517044067383, "epoch": 1, "memory": 44139, "step": 3400} +{"lr": 9.840086157170351e-05, "data_time": 0.0016017675399780274, "loss": 0.5124086380004883, "time": 0.633268666267395, "epoch": 1, "memory": 44139, "step": 3500} +{"lr": 9.830873498345255e-05, "data_time": 0.0020368099212646484, "loss": 0.469501256942749, "time": 0.6321164608001709, "epoch": 1, "memory": 44139, "step": 3600} +{"lr": 9.821407922429864e-05, "data_time": 0.0015007972717285157, "loss": 0.4675758719444275, "time": 0.6327744722366333, "epoch": 1, "memory": 44139, "step": 3700} +{"lr": 9.811689982200877e-05, "data_time": 0.0014689207077026368, "loss": 0.48063214421272277, "time": 0.6344436407089233, "epoch": 1, "memory": 44139, "step": 3800} +{"lr": 9.801720245172702e-05, "data_time": 0.0017609119415283204, "loss": 0.4920886754989624, "time": 0.6328966617584229, "epoch": 1, "memory": 44139, "step": 3900} +{"lr": 9.791499293564358e-05, "data_time": 0.0015110969543457031, "loss": 0.5180442333221436, "time": 0.6321677446365357, "epoch": 1, "memory": 44139, "step": 4000} +{"lr": 9.781027724265439e-05, "data_time": 0.002924323081970215, "loss": 0.5090823292732238, "time": 0.6352986574172974, "epoch": 1, "memory": 44139, "step": 4100} +{"accuracy/top1": 79.6716537475586, "data_time": 0.023074809242697322, "time": 1.3137203725446172, "step": 1} +{"lr": 9.769111539369434e-05, "data_time": 0.001417851448059082, "loss": 0.464958930015564, "time": 0.6323960781097412, "epoch": 2, "memory": 44140, "step": 4211} +{"lr": 9.758113190911623e-05, "data_time": 0.001508808135986328, "loss": 0.4936454653739929, "time": 0.6337024450302124, "epoch": 2, "memory": 44140, "step": 4311} +{"lr": 9.746866174465984e-05, "data_time": 0.0016089677810668945, "loss": 0.4635092854499817, "time": 0.6390480756759643, "epoch": 2, "memory": 44140, "step": 4411} +{"lr": 9.735371146842894e-05, "data_time": 0.0017129898071289063, "loss": 0.4712280809879303, "time": 0.6333528280258178, "epoch": 2, "memory": 44140, "step": 4511} +{"lr": 9.723628779336255e-05, "data_time": 0.0014141321182250977, "loss": 0.4785924434661865, "time": 0.6327715158462525, "epoch": 2, "memory": 44140, "step": 4611} +{"lr": 9.711639757684265e-05, "data_time": 0.0014632940292358398, "loss": 0.4785075306892395, "time": 0.632625937461853, "epoch": 2, "memory": 44140, "step": 4711} +{"lr": 9.699404782029395e-05, "data_time": 0.0013666629791259765, "loss": 0.4657517999410629, "time": 0.6335940837860108, "epoch": 2, "memory": 44140, "step": 4811} +{"lr": 9.686924566877498e-05, "data_time": 0.001330256462097168, "loss": 0.4924519777297974, "time": 0.6338048458099366, "epoch": 2, "memory": 44140, "step": 4911} +{"lr": 9.674199841056078e-05, "data_time": 0.0016278743743896485, "loss": 0.4578102469444275, "time": 0.6337306022644043, "epoch": 2, "memory": 44140, "step": 5011} +{"lr": 9.661231347671705e-05, "data_time": 0.0014542341232299805, "loss": 0.4974474787712097, "time": 0.6327146291732788, "epoch": 2, "memory": 44140, "step": 5111} +{"lr": 9.64801984406668e-05, "data_time": 0.00137786865234375, "loss": 0.47127229571342466, "time": 0.6333929538726807, "epoch": 2, "memory": 44140, "step": 5211} +{"lr": 9.634566101774737e-05, "data_time": 0.001451420783996582, "loss": 0.4649467468261719, "time": 0.6328048229217529, "epoch": 2, "memory": 44140, "step": 5311} +{"lr": 9.62087090647604e-05, "data_time": 0.001425766944885254, "loss": 0.4517827838659286, "time": 0.6339376926422119, "epoch": 2, "memory": 44140, "step": 5411} +{"lr": 9.606935057951247e-05, "data_time": 0.0014910459518432616, "loss": 0.448582249879837, "time": 0.6333621501922607, "epoch": 2, "memory": 44140, "step": 5511} +{"lr": 9.592759370034884e-05, "data_time": 0.0013057708740234375, "loss": 0.4891594350337982, "time": 0.6343136310577393, "epoch": 2, "memory": 44140, "step": 5611} +{"lr": 9.578344670567721e-05, "data_time": 0.0014177322387695312, "loss": 0.48531308174133303, "time": 0.6324173927307128, "epoch": 2, "memory": 44140, "step": 5711} +{"lr": 9.563691801348495e-05, "data_time": 0.0014531612396240234, "loss": 0.47551594376564027, "time": 0.6330333948135376, "epoch": 2, "memory": 44140, "step": 5811} +{"lr": 9.548801618084738e-05, "data_time": 0.0015380144119262694, "loss": 0.5018477857112884, "time": 0.6336065053939819, "epoch": 2, "memory": 44140, "step": 5911} +{"lr": 9.533674990342786e-05, "data_time": 0.001472616195678711, "loss": 0.45980127453804015, "time": 0.6343237161636353, "epoch": 2, "memory": 44140, "step": 6011} +{"lr": 9.518312801497029e-05, "data_time": 0.0014411449432373048, "loss": 0.462513792514801, "time": 0.6327403783798218, "epoch": 2, "memory": 44140, "step": 6111} +{"lr": 9.502715948678276e-05, "data_time": 0.0014458179473876953, "loss": 0.47115838527679443, "time": 0.6341397285461425, "epoch": 2, "memory": 44140, "step": 6211} +{"lr": 9.486885342721427e-05, "data_time": 0.0013930082321166992, "loss": 0.49234930574893954, "time": 0.6331521511077881, "epoch": 2, "memory": 44140, "step": 6311} +{"lr": 9.470821908112202e-05, "data_time": 0.0015508174896240235, "loss": 0.4903668165206909, "time": 0.6341366529464721, "epoch": 2, "memory": 44140, "step": 6411} +{"lr": 9.454526582933214e-05, "data_time": 0.0015165328979492188, "loss": 0.478467321395874, "time": 0.6331990718841553, "epoch": 2, "memory": 44140, "step": 6511} +{"lr": 9.438000318809176e-05, "data_time": 0.0014827966690063477, "loss": 0.44587414264678954, "time": 0.6328572988510132, "epoch": 2, "memory": 44140, "step": 6611} +{"lr": 9.42124408085132e-05, "data_time": 0.0014327049255371093, "loss": 0.4434362292289734, "time": 0.6405970573425293, "epoch": 2, "memory": 44140, "step": 6711} +{"lr": 9.404258847601036e-05, "data_time": 0.0013992786407470703, "loss": 0.467532616853714, "time": 0.6336398363113404, "epoch": 2, "memory": 44140, "step": 6811} +{"lr": 9.387045610972717e-05, "data_time": 0.0013389110565185547, "loss": 0.4624688357114792, "time": 0.6350803852081299, "epoch": 2, "memory": 44140, "step": 6911} +{"lr": 9.369605376195838e-05, "data_time": 0.0013905048370361328, "loss": 0.46402873992919924, "time": 0.6326971769332885, "epoch": 2, "memory": 44140, "step": 7011} +{"lr": 9.351939161756269e-05, "data_time": 0.001452493667602539, "loss": 0.48670401573181155, "time": 0.6348248720169067, "epoch": 2, "memory": 44140, "step": 7111} +{"lr": 9.334047999336764e-05, "data_time": 0.001385951042175293, "loss": 0.4709049880504608, "time": 0.6329421520233154, "epoch": 2, "memory": 44140, "step": 7211} +{"lr": 9.315932933756752e-05, "data_time": 0.0013277053833007813, "loss": 0.48430325388908385, "time": 0.6384798049926758, "epoch": 2, "memory": 44140, "step": 7311} +{"lr": 9.297595022911287e-05, "data_time": 0.0014803886413574218, "loss": 0.4709764182567596, "time": 0.6329472303390503, "epoch": 2, "memory": 44140, "step": 7411} +{"lr": 9.279035337709295e-05, "data_time": 0.001465606689453125, "loss": 0.4562969863414764, "time": 0.6321838855743408, "epoch": 2, "memory": 44140, "step": 7511} +{"lr": 9.260254962011016e-05, "data_time": 0.0014004230499267579, "loss": 0.4973665475845337, "time": 0.6325476884841919, "epoch": 2, "memory": 44140, "step": 7611} +{"lr": 9.241254992564702e-05, "data_time": 0.0014434814453125, "loss": 0.496775871515274, "time": 0.6410278081893921, "epoch": 2, "memory": 44140, "step": 7711} +{"lr": 9.22203653894259e-05, "data_time": 0.0014738321304321289, "loss": 0.46343018114566803, "time": 0.6349817752838135, "epoch": 2, "memory": 44140, "step": 7811} +{"lr": 9.202600723476098e-05, "data_time": 0.0014462232589721679, "loss": 0.457493257522583, "time": 0.6337049722671508, "epoch": 2, "memory": 44140, "step": 7911} +{"lr": 9.182948681190268e-05, "data_time": 0.0014832973480224609, "loss": 0.48266295790672303, "time": 0.6323026657104492, "epoch": 2, "memory": 44140, "step": 8011} +{"lr": 9.163081559737497e-05, "data_time": 0.0014436960220336914, "loss": 0.46472718715667727, "time": 0.6329636096954345, "epoch": 2, "memory": 44140, "step": 8111} +{"lr": 9.143000519330517e-05, "data_time": 0.0034996747970581056, "loss": 0.47150681614875795, "time": 0.634698486328125, "epoch": 2, "memory": 44140, "step": 8211} +{"accuracy/top1": 84.409912109375, "data_time": 0.01794877052307129, "time": 1.280187298854192, "step": 2} +{"lr": 9.120461478753376e-05, "data_time": 0.0016698837280273438, "loss": 0.5030023932456971, "time": 0.6323754787445068, "epoch": 3, "memory": 44140, "step": 8322} +{"lr": 9.099932932138134e-05, "data_time": 0.0015458345413208007, "loss": 0.45751610994338987, "time": 0.6347354412078857, "epoch": 3, "memory": 44140, "step": 8422} +{"lr": 9.079194154362181e-05, "data_time": 0.001558208465576172, "loss": 0.4556926995515823, "time": 0.6327492475509644, "epoch": 3, "memory": 44140, "step": 8522} +{"lr": 9.058246356541787e-05, "data_time": 0.0015683650970458984, "loss": 0.47245320677757263, "time": 0.6333695650100708, "epoch": 3, "memory": 44140, "step": 8622} +{"lr": 9.037090761999699e-05, "data_time": 0.0014644861221313477, "loss": 0.47963533997535707, "time": 0.6332862138748169, "epoch": 3, "memory": 44140, "step": 8722} +{"lr": 9.015728606193701e-05, "data_time": 0.0013886690139770508, "loss": 0.47418411672115324, "time": 0.6339986562728882, "epoch": 3, "memory": 44140, "step": 8822} +{"lr": 8.994161136644488e-05, "data_time": 0.0014088869094848633, "loss": 0.4321967601776123, "time": 0.6335744142532349, "epoch": 3, "memory": 44140, "step": 8922} +{"lr": 8.972389612862792e-05, "data_time": 0.0015585660934448243, "loss": 0.47279823422431944, "time": 0.6356832265853882, "epoch": 3, "memory": 44140, "step": 9022} +{"lr": 8.950415306275833e-05, "data_time": 0.0015932083129882812, "loss": 0.4606236100196838, "time": 0.6338690996170044, "epoch": 3, "memory": 44140, "step": 9122} +{"lr": 8.928239500153059e-05, "data_time": 0.0014430999755859375, "loss": 0.4547202467918396, "time": 0.6340134382247925, "epoch": 3, "memory": 44140, "step": 9222} +{"lr": 8.905863489531232e-05, "data_time": 0.0014305591583251953, "loss": 0.4499098300933838, "time": 0.6330832958221435, "epoch": 3, "memory": 44140, "step": 9322} +{"lr": 8.883288581138775e-05, "data_time": 0.0015149831771850586, "loss": 0.4653953969478607, "time": 0.6323587656021118, "epoch": 3, "memory": 44140, "step": 9422} +{"lr": 8.86051609331947e-05, "data_time": 0.0016928434371948243, "loss": 0.4494109511375427, "time": 0.6328791379928589, "epoch": 3, "memory": 44140, "step": 9522} +{"lr": 8.837547355955467e-05, "data_time": 0.0014689207077026368, "loss": 0.44448021054267883, "time": 0.6352570533752442, "epoch": 3, "memory": 44140, "step": 9622} +{"lr": 8.814383710389614e-05, "data_time": 0.001623678207397461, "loss": 0.4464535593986511, "time": 0.6360183000564575, "epoch": 3, "memory": 44140, "step": 9722} +{"lr": 8.791026509347146e-05, "data_time": 0.0014097213745117188, "loss": 0.45558726191520693, "time": 0.6346526145935059, "epoch": 3, "memory": 44140, "step": 9822} +{"lr": 8.767477116856659e-05, "data_time": 0.0016045093536376954, "loss": 0.45931233167648317, "time": 0.6333139657974243, "epoch": 3, "memory": 44140, "step": 9922} +{"lr": 8.743736908170466e-05, "data_time": 0.0014598846435546875, "loss": 0.4937210500240326, "time": 0.6338386774063111, "epoch": 3, "memory": 44140, "step": 10022} +{"lr": 8.719807269684305e-05, "data_time": 0.00140688419342041, "loss": 0.4547117233276367, "time": 0.6339017152786255, "epoch": 3, "memory": 44140, "step": 10122} +{"lr": 8.695689598856356e-05, "data_time": 0.001477670669555664, "loss": 0.4673060417175293, "time": 0.6335069179534912, "epoch": 3, "memory": 44140, "step": 10222} +{"lr": 8.67138530412559e-05, "data_time": 0.0014937639236450196, "loss": 0.4838897228240967, "time": 0.633981990814209, "epoch": 3, "memory": 44140, "step": 10322} +{"lr": 8.646895804829604e-05, "data_time": 0.0015043258666992188, "loss": 0.48540433049201964, "time": 0.6353799819946289, "epoch": 3, "memory": 44140, "step": 10422} +{"lr": 8.622222531121664e-05, "data_time": 0.0016132593154907227, "loss": 0.45520622134208677, "time": 0.6336303234100342, "epoch": 3, "memory": 44140, "step": 10522} +{"lr": 8.597366923887211e-05, "data_time": 0.0015883207321166991, "loss": 0.4660855710506439, "time": 0.6341121912002563, "epoch": 3, "memory": 44140, "step": 10622} +{"lr": 8.57233043465972e-05, "data_time": 0.001558232307434082, "loss": 0.4547513484954834, "time": 0.6333629131317139, "epoch": 3, "memory": 44140, "step": 10722} +{"lr": 8.547114525535937e-05, "data_time": 0.001423811912536621, "loss": 0.4696462869644165, "time": 0.635311222076416, "epoch": 3, "memory": 44140, "step": 10822} +{"lr": 8.521720669090474e-05, "data_time": 0.001542949676513672, "loss": 0.44859603345394133, "time": 0.6340259313583374, "epoch": 3, "memory": 44140, "step": 10922} +{"lr": 8.496150348289821e-05, "data_time": 0.0013875961303710938, "loss": 0.4574402332305908, "time": 0.6333667039871216, "epoch": 3, "memory": 44140, "step": 11022} +{"lr": 8.470405056405736e-05, "data_time": 0.0014806509017944336, "loss": 0.47548220455646517, "time": 0.6347490787506104, "epoch": 3, "memory": 44140, "step": 11122} +{"lr": 8.44448629692807e-05, "data_time": 0.0015207767486572266, "loss": 0.42512376308441163, "time": 0.6337342500686646, "epoch": 3, "memory": 44140, "step": 11222} +{"lr": 8.418395583476941e-05, "data_time": 0.0016474485397338866, "loss": 0.45764694511890414, "time": 0.6348349571228027, "epoch": 3, "memory": 44140, "step": 11322} +{"lr": 8.392134439714336e-05, "data_time": 0.0016633272171020508, "loss": 0.48138769865036013, "time": 0.6338225603103638, "epoch": 3, "memory": 44140, "step": 11422} +{"lr": 8.365704399255133e-05, "data_time": 0.0015475988388061524, "loss": 0.47832974791526794, "time": 0.635752534866333, "epoch": 3, "memory": 44140, "step": 11522} +{"lr": 8.339107005577562e-05, "data_time": 0.0013602495193481446, "loss": 0.46606722474098206, "time": 0.6338636159896851, "epoch": 3, "memory": 44140, "step": 11622} +{"lr": 8.31234381193305e-05, "data_time": 0.001706409454345703, "loss": 0.46423605680465696, "time": 0.6335791826248169, "epoch": 3, "memory": 44140, "step": 11722} +{"lr": 8.285416381255504e-05, "data_time": 0.001396942138671875, "loss": 0.4623516917228699, "time": 0.6347856760025025, "epoch": 3, "memory": 44140, "step": 11822} +{"lr": 8.258326286070047e-05, "data_time": 0.002235102653503418, "loss": 0.44467554688453675, "time": 0.6375418424606323, "epoch": 3, "memory": 44140, "step": 11922} +{"lr": 8.231075108401202e-05, "data_time": 0.001590728759765625, "loss": 0.4796634018421173, "time": 0.6340131998062134, "epoch": 3, "memory": 44140, "step": 12022} +{"lr": 8.203664439680493e-05, "data_time": 0.0017249584197998047, "loss": 0.4819711267948151, "time": 0.6324265480041504, "epoch": 3, "memory": 44140, "step": 12122} +{"lr": 8.176095880653475e-05, "data_time": 0.0014985322952270508, "loss": 0.4436904340982437, "time": 0.6329562187194824, "epoch": 3, "memory": 44140, "step": 12222} +{"lr": 8.148371041286308e-05, "data_time": 0.0031274080276489256, "loss": 0.45416841506958006, "time": 0.6359366178512573, "epoch": 3, "memory": 44140, "step": 12322} +{"accuracy/top1": 86.57630920410156, "data_time": 0.017765897512435912, "time": 1.279219208161036, "step": 3} +{"lr": 8.117415423358443e-05, "data_time": 0.0016343116760253907, "loss": 0.45611473321914675, "time": 0.6361605644226074, "epoch": 4, "memory": 44140, "step": 12433} +{"lr": 8.089366155759266e-05, "data_time": 0.001431417465209961, "loss": 0.448657700419426, "time": 0.6329862833023071, "epoch": 4, "memory": 44140, "step": 12533} +{"lr": 8.061165672717273e-05, "data_time": 0.0016322374343872071, "loss": 0.4526179552078247, "time": 0.6328766822814942, "epoch": 4, "memory": 44140, "step": 12633} +{"lr": 8.032815621102116e-05, "data_time": 0.0014635801315307617, "loss": 0.4790175259113312, "time": 0.6336992502212524, "epoch": 4, "memory": 44140, "step": 12733} +{"lr": 8.004317656518042e-05, "data_time": 0.0016112565994262696, "loss": 0.41661471128463745, "time": 0.6339985609054566, "epoch": 4, "memory": 44140, "step": 12833} +{"lr": 7.975673443207223e-05, "data_time": 0.0014017581939697265, "loss": 0.47605031728744507, "time": 0.6359712362289429, "epoch": 4, "memory": 44140, "step": 12933} +{"lr": 7.946884653952537e-05, "data_time": 0.001673579216003418, "loss": 0.44245787262916564, "time": 0.6362274169921875, "epoch": 4, "memory": 44140, "step": 13033} +{"lr": 7.91795296997992e-05, "data_time": 0.0018304109573364258, "loss": 0.46397862434387205, "time": 0.6344977617263794, "epoch": 4, "memory": 44140, "step": 13133} +{"lr": 7.888880080860151e-05, "data_time": 0.0017838716506958009, "loss": 0.47162403762340543, "time": 0.6353833913803101, "epoch": 4, "memory": 44140, "step": 13233} +{"lr": 7.859667684410205e-05, "data_time": 0.0021419286727905273, "loss": 0.4530477821826935, "time": 0.6351371049880982, "epoch": 4, "memory": 44140, "step": 13333} +{"lr": 7.830317486594082e-05, "data_time": 0.002172255516052246, "loss": 0.46278727650642393, "time": 0.6350919961929321, "epoch": 4, "memory": 44140, "step": 13433} +{"lr": 7.800831201423207e-05, "data_time": 0.0021567583084106446, "loss": 0.4498206853866577, "time": 0.6344151735305786, "epoch": 4, "memory": 44140, "step": 13533} +{"lr": 7.771210550856309e-05, "data_time": 0.00211029052734375, "loss": 0.46084251403808596, "time": 0.6326980590820312, "epoch": 4, "memory": 44140, "step": 13633} +{"lr": 7.741457264698884e-05, "data_time": 0.0016243696212768556, "loss": 0.4272093951702118, "time": 0.6350467681884766, "epoch": 4, "memory": 44140, "step": 13733} +{"lr": 7.711573080502147e-05, "data_time": 0.001788187026977539, "loss": 0.441609525680542, "time": 0.6337425231933593, "epoch": 4, "memory": 44140, "step": 13833} +{"lr": 7.681559743461609e-05, "data_time": 0.001470804214477539, "loss": 0.46558114886283875, "time": 0.6330921173095703, "epoch": 4, "memory": 44140, "step": 13933} +{"lr": 7.651419006315099e-05, "data_time": 0.0014998674392700194, "loss": 0.4717443287372589, "time": 0.6336032390594483, "epoch": 4, "memory": 44140, "step": 14033} +{"lr": 7.621152629240453e-05, "data_time": 0.0021607637405395507, "loss": 0.4876570016145706, "time": 0.632820463180542, "epoch": 4, "memory": 44140, "step": 14133} +{"lr": 7.590762379752719e-05, "data_time": 0.0017596721649169923, "loss": 0.4612249732017517, "time": 0.6332582473754883, "epoch": 4, "memory": 44140, "step": 14233} +{"lr": 7.560250032600911e-05, "data_time": 0.0018387317657470703, "loss": 0.46215103268623353, "time": 0.6342531681060791, "epoch": 4, "memory": 44140, "step": 14333} +{"lr": 7.52961736966437e-05, "data_time": 0.0012931108474731445, "loss": 0.4643726646900177, "time": 0.6360110282897949, "epoch": 4, "memory": 44140, "step": 14433} +{"lr": 7.498866179848732e-05, "data_time": 0.0015786170959472656, "loss": 0.4921614289283752, "time": 0.6338835239410401, "epoch": 4, "memory": 44140, "step": 14533} +{"lr": 7.467998258981415e-05, "data_time": 0.0014732837677001952, "loss": 0.4758285641670227, "time": 0.6380360841751098, "epoch": 4, "memory": 44140, "step": 14633} +{"lr": 7.437015409706804e-05, "data_time": 0.0018257617950439454, "loss": 0.44287298917770385, "time": 0.6327397108078003, "epoch": 4, "memory": 44140, "step": 14733} +{"lr": 7.405919441380929e-05, "data_time": 0.0014877557754516602, "loss": 0.4586209297180176, "time": 0.6339401006698608, "epoch": 4, "memory": 44140, "step": 14833} +{"lr": 7.374712169965821e-05, "data_time": 0.0017371654510498046, "loss": 0.48081800937652586, "time": 0.6329304218292237, "epoch": 4, "memory": 44140, "step": 14933} +{"lr": 7.343395417923461e-05, "data_time": 0.0015257596969604492, "loss": 0.4388641655445099, "time": 0.6340760231018067, "epoch": 4, "memory": 44140, "step": 15033} +{"lr": 7.311971014109342e-05, "data_time": 0.00172271728515625, "loss": 0.4692396819591522, "time": 0.6333871126174927, "epoch": 4, "memory": 44140, "step": 15133} +{"lr": 7.280440793665682e-05, "data_time": 0.0017844676971435548, "loss": 0.4430750787258148, "time": 0.6360200881958008, "epoch": 4, "memory": 44140, "step": 15233} +{"lr": 7.248806597914234e-05, "data_time": 0.002052974700927734, "loss": 0.47781543135643006, "time": 0.6328158617019654, "epoch": 4, "memory": 44140, "step": 15333} +{"lr": 7.217070274248765e-05, "data_time": 0.0024039745330810547, "loss": 0.44683332443237306, "time": 0.634963083267212, "epoch": 4, "memory": 44140, "step": 15433} +{"lr": 7.185233676027203e-05, "data_time": 0.001649928092956543, "loss": 0.46081181764602663, "time": 0.6344746589660645, "epoch": 4, "memory": 44140, "step": 15533} +{"lr": 7.153298662463338e-05, "data_time": 0.001600337028503418, "loss": 0.4723763644695282, "time": 0.6354378461837769, "epoch": 4, "memory": 44140, "step": 15633} +{"lr": 7.121267098518291e-05, "data_time": 0.0017853498458862305, "loss": 0.49327149987220764, "time": 0.6336778163909912, "epoch": 4, "memory": 44140, "step": 15733} +{"lr": 7.089140854791618e-05, "data_time": 0.0014856100082397462, "loss": 0.4476708710193634, "time": 0.6343159675598145, "epoch": 4, "memory": 44140, "step": 15833} +{"lr": 7.056921807412025e-05, "data_time": 0.001572871208190918, "loss": 0.4719863414764404, "time": 0.634116792678833, "epoch": 4, "memory": 44140, "step": 15933} +{"lr": 7.024611837927802e-05, "data_time": 0.0014816999435424804, "loss": 0.4623782217502594, "time": 0.6353331089019776, "epoch": 4, "memory": 44140, "step": 16033} +{"lr": 6.992212833196998e-05, "data_time": 0.0015210628509521485, "loss": 0.4481693387031555, "time": 0.6347303628921509, "epoch": 4, "memory": 44140, "step": 16133} +{"lr": 6.959726685277174e-05, "data_time": 0.002022385597229004, "loss": 0.4535117208957672, "time": 0.6339698791503906, "epoch": 4, "memory": 44140, "step": 16233} +{"lr": 6.927155291314949e-05, "data_time": 0.0016282558441162109, "loss": 0.43472830057144163, "time": 0.6349962711334228, "epoch": 4, "memory": 44140, "step": 16333} +{"lr": 6.894500553435204e-05, "data_time": 0.002988910675048828, "loss": 0.48811282515525817, "time": 0.6419486999511719, "epoch": 4, "memory": 44140, "step": 16433} +{"accuracy/top1": 87.74970245361328, "data_time": 0.01793476939201355, "time": 1.2803063730398814, "step": 4} +{"lr": 6.858158509719903e-05, "data_time": 0.0015555858612060548, "loss": 0.4589222550392151, "time": 0.6371300935745239, "epoch": 5, "memory": 44140, "step": 16544} +{"lr": 6.825334178880171e-05, "data_time": 0.0020379543304443358, "loss": 0.46468005776405336, "time": 0.6338776588439942, "epoch": 5, "memory": 44140, "step": 16644} +{"lr": 6.792432450336645e-05, "data_time": 0.0015887022018432617, "loss": 0.44431387782096865, "time": 0.6346333742141723, "epoch": 5, "memory": 44140, "step": 16744} +{"lr": 6.759455245505286e-05, "data_time": 0.0016175270080566405, "loss": 0.47444817423820496, "time": 0.6343838214874268, "epoch": 5, "memory": 44140, "step": 16844} +{"lr": 6.726404490209753e-05, "data_time": 0.001459050178527832, "loss": 0.4732815623283386, "time": 0.6343556404113769, "epoch": 5, "memory": 44140, "step": 16944} +{"lr": 6.693282114568972e-05, "data_time": 0.0018039703369140624, "loss": 0.46433616280555723, "time": 0.6349769353866577, "epoch": 5, "memory": 44140, "step": 17044} +{"lr": 6.660090052884388e-05, "data_time": 0.0024118423461914062, "loss": 0.46810369193553925, "time": 0.6337874412536622, "epoch": 5, "memory": 44140, "step": 17144} +{"lr": 6.626830243527016e-05, "data_time": 0.002102088928222656, "loss": 0.4910249650478363, "time": 0.6354984045028687, "epoch": 5, "memory": 44140, "step": 17244} +{"lr": 6.593504628824236e-05, "data_time": 0.0015336275100708008, "loss": 0.47623333930969236, "time": 0.6352442264556885, "epoch": 5, "memory": 44140, "step": 17344} +{"lr": 6.560115154946387e-05, "data_time": 0.0017632484436035157, "loss": 0.48938114047050474, "time": 0.6337780952453613, "epoch": 5, "memory": 44140, "step": 17444} +{"lr": 6.526663771793081e-05, "data_time": 0.002121281623840332, "loss": 0.461805933713913, "time": 0.6351408720016479, "epoch": 5, "memory": 44140, "step": 17544} +{"lr": 6.493152432879363e-05, "data_time": 0.0015882253646850586, "loss": 0.471802419424057, "time": 0.6341345071792602, "epoch": 5, "memory": 44140, "step": 17644} +{"lr": 6.459583095221596e-05, "data_time": 0.001488327980041504, "loss": 0.4258020281791687, "time": 0.6344020128250122, "epoch": 5, "memory": 44140, "step": 17744} +{"lr": 6.425957719223197e-05, "data_time": 0.0016002893447875977, "loss": 0.4689207494258881, "time": 0.6341097593307495, "epoch": 5, "memory": 44140, "step": 17844} +{"lr": 6.392278268560153e-05, "data_time": 0.0019644021987915037, "loss": 0.45474722385406496, "time": 0.6330618143081665, "epoch": 5, "memory": 44140, "step": 17944} +{"lr": 6.358546710066316e-05, "data_time": 0.001446080207824707, "loss": 0.4536068499088287, "time": 0.6336838960647583, "epoch": 5, "memory": 44140, "step": 18044} +{"lr": 6.324765013618573e-05, "data_time": 0.0016654014587402343, "loss": 0.46334036290645597, "time": 0.635744571685791, "epoch": 5, "memory": 44140, "step": 18144} +{"lr": 6.290935152021818e-05, "data_time": 0.0014859914779663086, "loss": 0.44370869994163514, "time": 0.6343780994415283, "epoch": 5, "memory": 44140, "step": 18244} +{"lr": 6.2570591008937e-05, "data_time": 0.0021237611770629885, "loss": 0.4457757234573364, "time": 0.6336251735687256, "epoch": 5, "memory": 44140, "step": 18344} +{"lr": 6.223138838549294e-05, "data_time": 0.0016193628311157227, "loss": 0.4611522316932678, "time": 0.634003758430481, "epoch": 5, "memory": 44140, "step": 18444} +{"lr": 6.18917634588553e-05, "data_time": 0.0014549732208251954, "loss": 0.466436493396759, "time": 0.6346589803695679, "epoch": 5, "memory": 44140, "step": 18544} +{"lr": 6.155173606265555e-05, "data_time": 0.0015205144882202148, "loss": 0.4525952637195587, "time": 0.6346549987792969, "epoch": 5, "memory": 44140, "step": 18644} +{"lr": 6.121132605402843e-05, "data_time": 0.0017121315002441406, "loss": 0.47000470757484436, "time": 0.6335221529006958, "epoch": 5, "memory": 44140, "step": 18744} +{"lr": 6.087055331245287e-05, "data_time": 0.0021596670150756834, "loss": 0.4398204982280731, "time": 0.6336713790893554, "epoch": 5, "memory": 44140, "step": 18844} +{"lr": 6.05294377385911e-05, "data_time": 0.0020489931106567384, "loss": 0.4636018514633179, "time": 0.6335805654525757, "epoch": 5, "memory": 44140, "step": 18944} +{"lr": 6.0187999253126355e-05, "data_time": 0.0014872074127197266, "loss": 0.47138131856918336, "time": 0.6342474222183228, "epoch": 5, "memory": 44140, "step": 19044} +{"lr": 5.9846257795599195e-05, "data_time": 0.0014727354049682618, "loss": 0.458501136302948, "time": 0.6362945079803467, "epoch": 5, "memory": 44140, "step": 19144} +{"lr": 5.950423332324355e-05, "data_time": 0.0025372028350830076, "loss": 0.4662016898393631, "time": 0.6349953174591064, "epoch": 5, "memory": 44140, "step": 19244} +{"lr": 5.9161945809820855e-05, "data_time": 0.001747465133666992, "loss": 0.4833052158355713, "time": 3.2729888200759887, "epoch": 5, "memory": 44140, "step": 19344} +{"lr": 5.8819415244453975e-05, "data_time": 0.001704835891723633, "loss": 0.4420823872089386, "time": 0.6357810258865356, "epoch": 5, "memory": 44140, "step": 19444} +{"lr": 5.8476661630459405e-05, "data_time": 0.0019277334213256836, "loss": 0.4586881101131439, "time": 0.6353569984436035, "epoch": 5, "memory": 44140, "step": 19544} +{"lr": 5.81337049841796e-05, "data_time": 0.001938033103942871, "loss": 0.4552523583173752, "time": 0.6364953756332398, "epoch": 5, "memory": 44140, "step": 19644} +{"lr": 5.779056533381373e-05, "data_time": 0.0015594244003295898, "loss": 0.4569220900535583, "time": 0.6351578235626221, "epoch": 5, "memory": 44140, "step": 19744} +{"lr": 5.744726271824817e-05, "data_time": 0.0014957666397094726, "loss": 0.47525659799575803, "time": 0.6341636657714844, "epoch": 5, "memory": 44140, "step": 19844} +{"lr": 5.710381718588622e-05, "data_time": 0.0015401840209960938, "loss": 0.47827436923980715, "time": 0.63492271900177, "epoch": 5, "memory": 44140, "step": 19944} +{"lr": 5.676024879347741e-05, "data_time": 0.002040863037109375, "loss": 0.44967913031578066, "time": 0.6435336112976074, "epoch": 5, "memory": 44140, "step": 20044} +{"lr": 5.6416577604945966e-05, "data_time": 0.0022011280059814455, "loss": 0.4790433168411255, "time": 0.6335307359695435, "epoch": 5, "memory": 44140, "step": 20144} +{"lr": 5.6072823690219385e-05, "data_time": 0.00143125057220459, "loss": 0.45392078161239624, "time": 0.6346201419830322, "epoch": 5, "memory": 44140, "step": 20244} +{"lr": 5.572900712405624e-05, "data_time": 0.001572728157043457, "loss": 0.47307010293006896, "time": 0.6350109815597534, "epoch": 5, "memory": 44140, "step": 20344} +{"lr": 5.538514798487397e-05, "data_time": 0.0020049095153808595, "loss": 0.4750144124031067, "time": 0.6349259614944458, "epoch": 5, "memory": 44140, "step": 20444} +{"lr": 5.5041266353576e-05, "data_time": 0.003738856315612793, "loss": 0.4440926045179367, "time": 0.6351094007492065, "epoch": 5, "memory": 44140, "step": 20544} +{"accuracy/top1": 88.1866226196289, "data_time": 0.017412426074345906, "time": 1.2803153574466706, "step": 5} +{"lr": 5.4659555783035384e-05, "data_time": 0.0014635086059570312, "loss": 0.48406948447227477, "time": 0.6340425491333008, "epoch": 6, "memory": 44140, "step": 20655} +{"lr": 5.431569258426336e-05, "data_time": 0.001412653923034668, "loss": 0.4501304030418396, "time": 0.6351540565490723, "epoch": 6, "memory": 44140, "step": 20755} +{"lr": 5.397186934810864e-05, "data_time": 0.0013927698135375976, "loss": 0.43961726427078246, "time": 0.6360284328460694, "epoch": 6, "memory": 44140, "step": 20855} +{"lr": 5.362810615337815e-05, "data_time": 0.0014640331268310548, "loss": 0.46147736310958865, "time": 0.6339771509170532, "epoch": 6, "memory": 44140, "step": 20955} +{"lr": 5.328442307537233e-05, "data_time": 0.0016271114349365235, "loss": 0.4509885013103485, "time": 0.634494423866272, "epoch": 6, "memory": 44140, "step": 21055} +{"lr": 5.294084018471303e-05, "data_time": 0.0015717506408691405, "loss": 0.47292686700820924, "time": 0.6439121723175049, "epoch": 6, "memory": 44140, "step": 21155} +{"lr": 5.259737754617132e-05, "data_time": 0.0015347957611083984, "loss": 0.49679508805274963, "time": 0.6337537050247193, "epoch": 6, "memory": 44140, "step": 21255} +{"lr": 5.2254055217495525e-05, "data_time": 0.001613020896911621, "loss": 0.4382395207881927, "time": 0.6347167015075683, "epoch": 6, "memory": 44140, "step": 21355} +{"lr": 5.1910893248240283e-05, "data_time": 0.001726388931274414, "loss": 0.4445831775665283, "time": 0.6341121435165405, "epoch": 6, "memory": 44140, "step": 21455} +{"lr": 5.156791167859532e-05, "data_time": 0.002004122734069824, "loss": 0.4478444576263428, "time": 0.6355179786682129, "epoch": 6, "memory": 44140, "step": 21555} +{"lr": 5.122513053821533e-05, "data_time": 0.0014799118041992187, "loss": 0.4767975151538849, "time": 0.6340090990066528, "epoch": 6, "memory": 44140, "step": 21655} +{"lr": 5.0882569845050173e-05, "data_time": 0.0016569852828979491, "loss": 0.47388242483139037, "time": 0.6328195333480835, "epoch": 6, "memory": 44140, "step": 21755} +{"lr": 5.054024960417592e-05, "data_time": 0.0015819787979125977, "loss": 0.4729978024959564, "time": 0.6351083278656006, "epoch": 6, "memory": 44140, "step": 21855} +{"lr": 5.0198189806626494e-05, "data_time": 0.0020102262496948242, "loss": 0.44750158190727235, "time": 0.6341897249221802, "epoch": 6, "memory": 44140, "step": 21955} +{"lr": 4.985641042822647e-05, "data_time": 0.0019378900527954102, "loss": 0.48021815419197084, "time": 0.6335058689117432, "epoch": 6, "memory": 44140, "step": 22055} +{"lr": 4.951493142842412e-05, "data_time": 0.0013368844985961914, "loss": 0.4866527020931244, "time": 0.6338645219802856, "epoch": 6, "memory": 44140, "step": 22155} +{"lr": 4.917377274912603e-05, "data_time": 0.0015365362167358398, "loss": 0.47538907527923585, "time": 0.635843300819397, "epoch": 6, "memory": 44140, "step": 22255} +{"lr": 4.88329543135327e-05, "data_time": 0.001574230194091797, "loss": 0.47498087882995604, "time": 0.6340251445770264, "epoch": 6, "memory": 44140, "step": 22355} +{"lr": 4.849249602497467e-05, "data_time": 0.0020350217819213867, "loss": 0.4520275712013245, "time": 0.6329787969589233, "epoch": 6, "memory": 44140, "step": 22455} +{"lr": 4.8152417765750456e-05, "data_time": 0.001703786849975586, "loss": 0.4768667906522751, "time": 1.2932875871658325, "epoch": 6, "memory": 44140, "step": 22555} +{"lr": 4.7812739395965355e-05, "data_time": 0.0017891883850097655, "loss": 0.47207836508750917, "time": 0.6352556705474853, "epoch": 6, "memory": 44140, "step": 22655} +{"lr": 4.74734807523717e-05, "data_time": 0.0015207052230834961, "loss": 0.4756499171257019, "time": 0.6357563257217407, "epoch": 6, "memory": 44140, "step": 22755} +{"lr": 4.7134661647210416e-05, "data_time": 0.001726675033569336, "loss": 0.4574288845062256, "time": 0.6348897457122803, "epoch": 6, "memory": 44140, "step": 22855} +{"lr": 4.679630186705388e-05, "data_time": 0.002379608154296875, "loss": 0.4630141377449036, "time": 0.635697340965271, "epoch": 6, "memory": 44140, "step": 22955} +{"lr": 4.6458421171650644e-05, "data_time": 0.002309274673461914, "loss": 0.4364140689373016, "time": 0.6367731094360352, "epoch": 6, "memory": 44140, "step": 23055} +{"lr": 4.6121039292771295e-05, "data_time": 0.001967716217041016, "loss": 0.4441856205463409, "time": 0.636125898361206, "epoch": 6, "memory": 44140, "step": 23155} +{"lr": 4.5784175933056197e-05, "data_time": 0.0015056848526000977, "loss": 0.4734744668006897, "time": 0.6337905406951905, "epoch": 6, "memory": 44140, "step": 23255} +{"lr": 4.544785076486494e-05, "data_time": 0.0019669294357299804, "loss": 0.457925409078598, "time": 0.6346128463745118, "epoch": 6, "memory": 44140, "step": 23355} +{"lr": 4.5112083429127475e-05, "data_time": 0.0016997337341308593, "loss": 0.46377404034137726, "time": 0.6371508836746216, "epoch": 6, "memory": 44140, "step": 23455} +{"lr": 4.477689353419706e-05, "data_time": 0.0015819311141967774, "loss": 0.4767437160015106, "time": 0.6379316806793213, "epoch": 6, "memory": 44140, "step": 23555} +{"lr": 4.4442300654705266e-05, "data_time": 0.00226438045501709, "loss": 0.4358688294887543, "time": 0.6347510576248169, "epoch": 6, "memory": 44140, "step": 23655} +{"lr": 4.410832433041866e-05, "data_time": 0.0018082618713378905, "loss": 0.4761827826499939, "time": 0.6339036226272583, "epoch": 6, "memory": 44140, "step": 23755} +{"lr": 4.377498406509792e-05, "data_time": 0.0016064167022705079, "loss": 0.45044424533843996, "time": 0.6349463939666748, "epoch": 6, "memory": 44140, "step": 23855} +{"lr": 4.344229932535875e-05, "data_time": 0.001959872245788574, "loss": 0.4763031482696533, "time": 0.6344331264495849, "epoch": 6, "memory": 44140, "step": 23955} +{"lr": 4.3110289539535115e-05, "data_time": 0.0015465736389160157, "loss": 0.45789788365364076, "time": 0.6343870878219604, "epoch": 6, "memory": 44140, "step": 24055} +{"lr": 4.2778974096544436e-05, "data_time": 0.0019354581832885741, "loss": 0.46838202476501467, "time": 0.6338257551193237, "epoch": 6, "memory": 44140, "step": 24155} +{"lr": 4.2448372344755544e-05, "data_time": 0.0021352529525756835, "loss": 0.45679605603218076, "time": 0.6424296617507934, "epoch": 6, "memory": 44140, "step": 24255} +{"lr": 4.211850359085871e-05, "data_time": 0.00158236026763916, "loss": 0.4579767048358917, "time": 0.6419151306152344, "epoch": 6, "memory": 44140, "step": 24355} +{"lr": 4.17893870987381e-05, "data_time": 0.001468825340270996, "loss": 0.4642179012298584, "time": 0.6362257957458496, "epoch": 6, "memory": 44140, "step": 24455} +{"lr": 4.146104208834679e-05, "data_time": 0.0018604755401611327, "loss": 0.4529073446989059, "time": 0.6338111877441406, "epoch": 6, "memory": 44140, "step": 24555} +{"lr": 4.113348773458442e-05, "data_time": 0.003417181968688965, "loss": 0.4551727056503296, "time": 0.6347980737686157, "epoch": 6, "memory": 44140, "step": 24655} +{"accuracy/top1": 88.90158081054688, "data_time": 0.017988532781600952, "time": 1.2791539053122203, "step": 6} +{"lr": 4.0770851520835164e-05, "data_time": 0.0016108989715576173, "loss": 0.4696533143520355, "time": 0.6349081993103027, "epoch": 7, "memory": 44140, "step": 24766} +{"lr": 4.044502815742039e-05, "data_time": 0.00169525146484375, "loss": 0.46119399070739747, "time": 0.6335535764694213, "epoch": 7, "memory": 44140, "step": 24866} +{"lr": 4.012005478445896e-05, "data_time": 0.002030205726623535, "loss": 0.45948365330696106, "time": 0.6337975025177002, "epoch": 7, "memory": 44140, "step": 24966} +{"lr": 3.9795950379951394e-05, "data_time": 0.0016697406768798827, "loss": 0.4820225715637207, "time": 0.6349336862564087, "epoch": 7, "memory": 44140, "step": 25066} +{"lr": 3.947273387115169e-05, "data_time": 0.0016877174377441407, "loss": 0.45211027264595033, "time": 0.6361408233642578, "epoch": 7, "memory": 44140, "step": 25166} +{"lr": 3.9150424133462044e-05, "data_time": 0.0015559911727905274, "loss": 0.4473924934864044, "time": 0.6361247777938843, "epoch": 7, "memory": 44140, "step": 25266} +{"lr": 3.882903998933037e-05, "data_time": 0.0014920949935913086, "loss": 0.4652102291584015, "time": 0.6383752584457397, "epoch": 7, "memory": 44140, "step": 25366} +{"lr": 3.8508600207151264e-05, "data_time": 0.00156095027923584, "loss": 0.4557941198348999, "time": 0.6369585990905762, "epoch": 7, "memory": 44140, "step": 25466} +{"lr": 3.818912350016978e-05, "data_time": 0.0015782594680786132, "loss": 0.4809683561325073, "time": 0.6361147642135621, "epoch": 7, "memory": 44140, "step": 25566} +{"lr": 3.787062852538883e-05, "data_time": 0.0015544891357421875, "loss": 0.45643723011016846, "time": 0.6339664220809936, "epoch": 7, "memory": 44140, "step": 25666} +{"lr": 3.7553133882479346e-05, "data_time": 0.0016622066497802735, "loss": 0.4706435680389404, "time": 0.6337752103805542, "epoch": 7, "memory": 44140, "step": 25766} +{"lr": 3.723665811269437e-05, "data_time": 0.0017843961715698243, "loss": 0.46476571559906005, "time": 0.6333802938461304, "epoch": 7, "memory": 44140, "step": 25866} +{"lr": 3.6921219697786176e-05, "data_time": 0.0018469333648681641, "loss": 0.4563186526298523, "time": 0.6359025955200195, "epoch": 7, "memory": 44140, "step": 25966} +{"lr": 3.660683705892677e-05, "data_time": 0.001450324058532715, "loss": 0.4826504051685333, "time": 0.633551549911499, "epoch": 7, "memory": 44140, "step": 26066} +{"lr": 3.6293528555632435e-05, "data_time": 0.0017829179763793946, "loss": 0.4337110996246338, "time": 0.6374663829803466, "epoch": 7, "memory": 44140, "step": 26166} +{"lr": 3.598131248469134e-05, "data_time": 0.0014847278594970702, "loss": 0.46282885074615476, "time": 0.6346645355224609, "epoch": 7, "memory": 44140, "step": 26266} +{"lr": 3.567020707909507e-05, "data_time": 0.001493549346923828, "loss": 0.47078515887260436, "time": 0.6341926574707031, "epoch": 7, "memory": 44140, "step": 26366} +{"lr": 3.5360230506973935e-05, "data_time": 0.001625227928161621, "loss": 0.4480311840772629, "time": 0.6349181175231934, "epoch": 7, "memory": 44140, "step": 26466} +{"lr": 3.5051400870535877e-05, "data_time": 0.001460099220275879, "loss": 0.4460991144180298, "time": 0.6339737415313721, "epoch": 7, "memory": 44140, "step": 26566} +{"lr": 3.47437362050094e-05, "data_time": 0.0022795438766479493, "loss": 0.46299099624156953, "time": 0.6366446256637573, "epoch": 7, "memory": 44140, "step": 26666} +{"lr": 3.443725447759023e-05, "data_time": 0.0016202211380004882, "loss": 0.4408200979232788, "time": 0.6363564014434815, "epoch": 7, "memory": 44140, "step": 26766} +{"lr": 3.413197358639224e-05, "data_time": 0.001509714126586914, "loss": 0.4818792164325714, "time": 0.6370244503021241, "epoch": 7, "memory": 44140, "step": 26866} +{"lr": 3.382791135940194e-05, "data_time": 0.0017197132110595703, "loss": 0.4885852813720703, "time": 0.6382949590682984, "epoch": 7, "memory": 44140, "step": 26966} +{"lr": 3.35250855534377e-05, "data_time": 0.001477670669555664, "loss": 0.4552358388900757, "time": 0.6335381746292115, "epoch": 7, "memory": 44140, "step": 27066} +{"lr": 3.3223513853112476e-05, "data_time": 0.0014629125595092773, "loss": 0.4468356966972351, "time": 0.6357050657272338, "epoch": 7, "memory": 44140, "step": 27166} +{"lr": 3.292321386980115e-05, "data_time": 0.0015483379364013671, "loss": 0.4653731107711792, "time": 0.6372542142868042, "epoch": 7, "memory": 44140, "step": 27266} +{"lr": 3.262420314061217e-05, "data_time": 0.0023773908615112305, "loss": 0.45212759971618655, "time": 0.6342232704162598, "epoch": 7, "memory": 44140, "step": 27366} +{"lr": 3.2326499127363195e-05, "data_time": 0.0016300201416015625, "loss": 0.47584818601608275, "time": 0.6367329359054565, "epoch": 7, "memory": 44140, "step": 27466} +{"lr": 3.2030119215561613e-05, "data_time": 0.0014003515243530273, "loss": 0.4543128728866577, "time": 0.6339415073394775, "epoch": 7, "memory": 44140, "step": 27566} +{"lr": 3.173508071338896e-05, "data_time": 0.0016157865524291993, "loss": 0.44322863817214964, "time": 0.6338517665863037, "epoch": 7, "memory": 44140, "step": 27666} +{"lr": 3.144140085069037e-05, "data_time": 0.0016385555267333985, "loss": 0.463171112537384, "time": 0.6348424673080444, "epoch": 7, "memory": 44140, "step": 27766} +{"lr": 3.114909677796814e-05, "data_time": 0.0015558481216430664, "loss": 0.44728931188583376, "time": 0.6365219831466675, "epoch": 7, "memory": 44140, "step": 27866} +{"lr": 3.085818556538059e-05, "data_time": 0.0014888286590576173, "loss": 0.46412636041641236, "time": 0.6337711095809937, "epoch": 7, "memory": 44140, "step": 27966} +{"lr": 3.056868420174455e-05, "data_time": 0.0017244338989257813, "loss": 0.48188812732696534, "time": 0.634553861618042, "epoch": 7, "memory": 44140, "step": 28066} +{"lr": 3.028060959354392e-05, "data_time": 0.0015617609024047852, "loss": 0.48123701810836794, "time": 0.6338025331497192, "epoch": 7, "memory": 44140, "step": 28166} +{"lr": 2.9993978563941876e-05, "data_time": 0.001407027244567871, "loss": 0.4762677103281021, "time": 0.637341046333313, "epoch": 7, "memory": 44140, "step": 28266} +{"lr": 2.9708807851798587e-05, "data_time": 0.0017408370971679688, "loss": 0.4819815993309021, "time": 0.636882209777832, "epoch": 7, "memory": 44140, "step": 28366} +{"lr": 2.9425114110693717e-05, "data_time": 0.001581263542175293, "loss": 0.4441631197929382, "time": 0.6356881141662598, "epoch": 7, "memory": 44140, "step": 28466} +{"lr": 2.9142913907953844e-05, "data_time": 0.0019063472747802735, "loss": 0.4636713624000549, "time": 0.6350440740585327, "epoch": 7, "memory": 44140, "step": 28566} +{"lr": 2.8862223723684912e-05, "data_time": 0.0015549421310424804, "loss": 0.46156834065914154, "time": 0.6346679925918579, "epoch": 7, "memory": 44140, "step": 28666} +{"lr": 2.8583059949809854e-05, "data_time": 0.002940654754638672, "loss": 0.4637675046920776, "time": 0.6384005784988404, "epoch": 7, "memory": 44140, "step": 28766} +{"accuracy/top1": 89.14817810058594, "data_time": 0.01785958210627238, "time": 1.2793065150578817, "step": 7} +{"lr": 2.827499545208893e-05, "data_time": 0.0014168500900268554, "loss": 0.46368653178215025, "time": 0.6343302726745605, "epoch": 8, "memory": 44140, "step": 28877} +{"lr": 2.7999105784987406e-05, "data_time": 0.0018784523010253907, "loss": 0.46099224090576174, "time": 0.6338589191436768, "epoch": 8, "memory": 44140, "step": 28977} +{"lr": 2.772479293318711e-05, "data_time": 0.0015350341796875, "loss": 0.4673727661371231, "time": 0.6351502656936645, "epoch": 8, "memory": 44140, "step": 29077} +{"lr": 2.7452072916183603e-05, "data_time": 0.0015818119049072266, "loss": 0.47864570617675783, "time": 0.6330815553665161, "epoch": 8, "memory": 44140, "step": 29177} +{"lr": 2.7180961660452883e-05, "data_time": 0.0015813827514648438, "loss": 0.4577386975288391, "time": 0.6341041326522827, "epoch": 8, "memory": 44140, "step": 29277} +{"lr": 2.6911474998521622e-05, "data_time": 0.0013978004455566407, "loss": 0.5009500443935394, "time": 0.6377395868301392, "epoch": 8, "memory": 44140, "step": 29377} +{"lr": 2.664362866804239e-05, "data_time": 0.002186417579650879, "loss": 0.46405903697013856, "time": 0.6343747854232789, "epoch": 8, "memory": 44140, "step": 29477} +{"lr": 2.637743831087459e-05, "data_time": 0.0020719051361083986, "loss": 0.4770909249782562, "time": 0.6331545829772949, "epoch": 8, "memory": 44140, "step": 29577} +{"lr": 2.6112919472171114e-05, "data_time": 0.002500581741333008, "loss": 0.4557668209075928, "time": 0.6330735683441162, "epoch": 8, "memory": 44140, "step": 29677} +{"lr": 2.5850087599470434e-05, "data_time": 0.0016175031661987305, "loss": 0.4631189674139023, "time": 0.6367992401123047, "epoch": 8, "memory": 44140, "step": 29777} +{"lr": 2.5588958041794503e-05, "data_time": 0.0018662691116333007, "loss": 0.45991411805152893, "time": 0.6337423086166382, "epoch": 8, "memory": 44140, "step": 29877} +{"lr": 2.5329546048752453e-05, "data_time": 0.002153301239013672, "loss": 0.4490981578826904, "time": 0.633574390411377, "epoch": 8, "memory": 44140, "step": 29977} +{"lr": 2.5071866769649806e-05, "data_time": 0.0016371726989746094, "loss": 0.4482205629348755, "time": 0.6342392206192017, "epoch": 8, "memory": 44140, "step": 30077} +{"lr": 2.4815935252604222e-05, "data_time": 0.001477670669555664, "loss": 0.4637909531593323, "time": 0.6326849460601807, "epoch": 8, "memory": 44140, "step": 30177} +{"lr": 2.4561766443666384e-05, "data_time": 0.0015606641769409179, "loss": 0.455281537771225, "time": 0.6346306800842285, "epoch": 8, "memory": 44140, "step": 30277} +{"lr": 2.4309375185947164e-05, "data_time": 0.001670694351196289, "loss": 0.42819930613040924, "time": 0.6330586910247803, "epoch": 8, "memory": 44140, "step": 30377} +{"lr": 2.40587762187509e-05, "data_time": 0.0019073247909545898, "loss": 0.4765880346298218, "time": 0.633723521232605, "epoch": 8, "memory": 44140, "step": 30477} +{"lr": 2.3809984176714634e-05, "data_time": 0.0016132354736328124, "loss": 0.47656781077384947, "time": 0.634052038192749, "epoch": 8, "memory": 44140, "step": 30577} +{"lr": 2.3563013588953406e-05, "data_time": 0.0016823053359985352, "loss": 0.45848439931869506, "time": 0.6352833986282349, "epoch": 8, "memory": 44140, "step": 30677} +{"lr": 2.331787887821182e-05, "data_time": 0.0014740228652954102, "loss": 0.46513075828552247, "time": 0.6356109619140625, "epoch": 8, "memory": 44140, "step": 30777} +{"lr": 2.307459436002179e-05, "data_time": 0.002760791778564453, "loss": 0.47583167552947997, "time": 0.633095121383667, "epoch": 8, "memory": 44140, "step": 30877} +{"lr": 2.2833174241866527e-05, "data_time": 0.002381706237792969, "loss": 0.45027904510498046, "time": 0.6363095998764038, "epoch": 8, "memory": 44140, "step": 30977} +{"lr": 2.259363262235086e-05, "data_time": 0.0022494316101074217, "loss": 0.4671119451522827, "time": 0.6359816789627075, "epoch": 8, "memory": 44140, "step": 31077} +{"lr": 2.2355983490377735e-05, "data_time": 0.0015765666961669923, "loss": 0.43791429698467255, "time": 0.6359403133392334, "epoch": 8, "memory": 44140, "step": 31177} +{"lr": 2.2120240724331546e-05, "data_time": 0.0018905401229858398, "loss": 0.456916469335556, "time": 0.6331723928451538, "epoch": 8, "memory": 44140, "step": 31277} +{"lr": 2.1886418091267445e-05, "data_time": 0.0018166542053222657, "loss": 0.46086585223674775, "time": 0.6346737384796143, "epoch": 8, "memory": 44140, "step": 31377} +{"lr": 2.1654529246107496e-05, "data_time": 0.0015304803848266602, "loss": 0.47067301273345946, "time": 0.6341936826705933, "epoch": 8, "memory": 44140, "step": 31477} +{"lr": 2.1424587730843184e-05, "data_time": 0.0015278100967407227, "loss": 0.4529714584350586, "time": 0.6375127077102661, "epoch": 8, "memory": 44140, "step": 31577} +{"lr": 2.1196606973744593e-05, "data_time": 0.0016810178756713867, "loss": 0.4490090847015381, "time": 0.6349388837814331, "epoch": 8, "memory": 44140, "step": 31677} +{"lr": 2.0970600288576148e-05, "data_time": 0.0018979787826538086, "loss": 0.4901163220405579, "time": 0.633464765548706, "epoch": 8, "memory": 44140, "step": 31777} +{"lr": 2.0746580873819274e-05, "data_time": 0.0013696908950805663, "loss": 0.46635947823524476, "time": 0.6334488153457641, "epoch": 8, "memory": 44140, "step": 31877} +{"lr": 2.0524561811901426e-05, "data_time": 0.0016888856887817382, "loss": 0.44515493512153625, "time": 0.6334747791290283, "epoch": 8, "memory": 44140, "step": 31977} +{"lr": 2.0304556068432268e-05, "data_time": 0.0016237258911132812, "loss": 0.46383466124534606, "time": 0.6341516494750976, "epoch": 8, "memory": 44140, "step": 32077} +{"lr": 2.0086576491446333e-05, "data_time": 0.0014827728271484375, "loss": 0.46970545649528506, "time": 0.6342121124267578, "epoch": 8, "memory": 44140, "step": 32177} +{"lr": 1.987063581065291e-05, "data_time": 0.0016895532608032227, "loss": 0.4618035197257996, "time": 0.6343858242034912, "epoch": 8, "memory": 44140, "step": 32277} +{"lr": 1.965674663669245e-05, "data_time": 0.0016300201416015625, "loss": 0.49416699409484866, "time": 0.6343100309371948, "epoch": 8, "memory": 44140, "step": 32377} +{"lr": 1.9444921460400247e-05, "data_time": 0.0016714572906494141, "loss": 0.4498171627521515, "time": 0.6342238664627076, "epoch": 8, "memory": 44140, "step": 32477} +{"lr": 1.923517265207697e-05, "data_time": 0.0022983312606811523, "loss": 0.4613679885864258, "time": 0.6332913398742676, "epoch": 8, "memory": 44140, "step": 32577} +{"lr": 1.9027512460766204e-05, "data_time": 0.002181529998779297, "loss": 0.46468991935253146, "time": 0.640119194984436, "epoch": 8, "memory": 44140, "step": 32677} +{"lr": 1.8821953013539136e-05, "data_time": 0.001618170738220215, "loss": 0.46930948495864866, "time": 0.6349443912506103, "epoch": 8, "memory": 44140, "step": 32777} +{"lr": 1.8618506314786433e-05, "data_time": 0.003096127510070801, "loss": 0.47440574765205384, "time": 0.6349209785461426, "epoch": 8, "memory": 44140, "step": 32877} +{"accuracy/top1": 89.2044448852539, "data_time": 0.017275114854176838, "time": 1.2790930946667989, "step": 8} +{"lr": 1.8395169032542028e-05, "data_time": 0.0014571428298950195, "loss": 0.45794283151626586, "time": 0.6334787845611572, "epoch": 9, "memory": 44140, "step": 32988} +{"lr": 1.819621906460573e-05, "data_time": 0.0018780231475830078, "loss": 0.47048654556274416, "time": 0.6338154077529907, "epoch": 9, "memory": 44140, "step": 33088} +{"lr": 1.799941838715077e-05, "data_time": 0.001538395881652832, "loss": 0.4661583721637726, "time": 0.6356528043746948, "epoch": 9, "memory": 44140, "step": 33188} +{"lr": 1.7804778493067625e-05, "data_time": 0.0015492200851440429, "loss": 0.47302921414375304, "time": 0.6382202625274658, "epoch": 9, "memory": 44140, "step": 33288} +{"lr": 1.761231074905996e-05, "data_time": 0.0014631032943725586, "loss": 0.46851676106452944, "time": 0.6360684633255005, "epoch": 9, "memory": 44140, "step": 33388} +{"lr": 1.742202639498083e-05, "data_time": 0.001425457000732422, "loss": 0.45815710425376893, "time": 0.6349555015563965, "epoch": 9, "memory": 44140, "step": 33488} +{"lr": 1.7233936543176347e-05, "data_time": 0.002461886405944824, "loss": 0.46372724175453184, "time": 0.6354443073272705, "epoch": 9, "memory": 44140, "step": 33588} +{"lr": 1.704805217783665e-05, "data_time": 0.001511859893798828, "loss": 0.44393640756607056, "time": 0.6344957113265991, "epoch": 9, "memory": 44140, "step": 33688} +{"lr": 1.68643841543545e-05, "data_time": 0.0014844655990600586, "loss": 0.4598296880722046, "time": 0.6378105640411377, "epoch": 9, "memory": 44140, "step": 33788} +{"lr": 1.6682943198691327e-05, "data_time": 0.0017051458358764648, "loss": 0.45735839009284973, "time": 0.6351711273193359, "epoch": 9, "memory": 44140, "step": 33888} +{"lr": 1.6503739906750822e-05, "data_time": 0.0016067981719970702, "loss": 0.46009152233600614, "time": 0.6342929840087891, "epoch": 9, "memory": 44140, "step": 33988} +{"lr": 1.6326784743760187e-05, "data_time": 0.0014171838760375977, "loss": 0.4574501931667328, "time": 0.6351300477981567, "epoch": 9, "memory": 44140, "step": 34088} +{"lr": 1.6152088043659013e-05, "data_time": 0.001588892936706543, "loss": 0.47189466953277587, "time": 0.6352554082870483, "epoch": 9, "memory": 44140, "step": 34188} +{"lr": 1.59796600084957e-05, "data_time": 0.0015816926956176759, "loss": 0.4594050168991089, "time": 0.6334317445755004, "epoch": 9, "memory": 44140, "step": 34288} +{"lr": 1.5809510707831687e-05, "data_time": 0.0015261173248291016, "loss": 0.4613716512918472, "time": 0.6336474418640137, "epoch": 9, "memory": 44140, "step": 34388} +{"lr": 1.5641650078153473e-05, "data_time": 0.0014529705047607421, "loss": 0.4630406856536865, "time": 0.6344540119171143, "epoch": 9, "memory": 44140, "step": 34488} +{"lr": 1.547608792229235e-05, "data_time": 0.0016174793243408203, "loss": 0.4735118567943573, "time": 0.6345244884490967, "epoch": 9, "memory": 44140, "step": 34588} +{"lr": 1.5312833908851794e-05, "data_time": 0.0015510320663452148, "loss": 0.4545616626739502, "time": 0.6343519210815429, "epoch": 9, "memory": 44140, "step": 34688} +{"lr": 1.5151897571642958e-05, "data_time": 0.0015728473663330078, "loss": 0.437881326675415, "time": 0.6345327138900757, "epoch": 9, "memory": 44140, "step": 34788} +{"lr": 1.4993288309127877e-05, "data_time": 0.0014739990234375, "loss": 0.47203379273414614, "time": 0.6343764305114746, "epoch": 9, "memory": 44140, "step": 34888} +{"lr": 1.4837015383870593e-05, "data_time": 0.0014815330505371094, "loss": 0.45972618758678435, "time": 0.635708999633789, "epoch": 9, "memory": 44140, "step": 34988} +{"lr": 1.4683087921996264e-05, "data_time": 0.0016397953033447266, "loss": 0.4681494772434235, "time": 0.635512399673462, "epoch": 9, "memory": 44140, "step": 35088} +{"lr": 1.453151491265819e-05, "data_time": 0.0014656543731689452, "loss": 0.45016089677810667, "time": 0.6344328880310058, "epoch": 9, "memory": 44140, "step": 35188} +{"lr": 1.4382305207512847e-05, "data_time": 0.0014738321304321289, "loss": 0.4790822803974152, "time": 0.6335061311721801, "epoch": 9, "memory": 44140, "step": 35288} +{"lr": 1.423546752020298e-05, "data_time": 0.0015030384063720703, "loss": 0.4548185706138611, "time": 0.6337298393249512, "epoch": 9, "memory": 44140, "step": 35388} +{"lr": 1.4091010425848762e-05, "data_time": 0.0015840530395507812, "loss": 0.4755293786525726, "time": 0.6353351354599, "epoch": 9, "memory": 44140, "step": 35488} +{"lr": 1.3948942360546966e-05, "data_time": 0.001415586471557617, "loss": 0.46705070734024046, "time": 0.6347707986831665, "epoch": 9, "memory": 44140, "step": 35588} +{"lr": 1.3809271620878346e-05, "data_time": 0.001620006561279297, "loss": 0.45487709641456603, "time": 0.6343161821365356, "epoch": 9, "memory": 44140, "step": 35688} +{"lr": 1.3672006363423091e-05, "data_time": 0.0024388313293457033, "loss": 0.47723528146743777, "time": 0.6360401391983033, "epoch": 9, "memory": 44140, "step": 35788} +{"lr": 1.3537154604284566e-05, "data_time": 0.0016469240188598632, "loss": 0.46806180775165557, "time": 0.6335711717605591, "epoch": 9, "memory": 44140, "step": 35888} +{"lr": 1.3404724218621087e-05, "data_time": 0.0015513896942138672, "loss": 0.4778871476650238, "time": 0.6358850955963135, "epoch": 9, "memory": 44140, "step": 35988} +{"lr": 1.3274722940186121e-05, "data_time": 0.0015336990356445313, "loss": 0.43152810335159303, "time": 0.6338732481002808, "epoch": 9, "memory": 44140, "step": 36088} +{"lr": 1.3147158360876541e-05, "data_time": 0.0017436742782592773, "loss": 0.4663323938846588, "time": 0.634447169303894, "epoch": 9, "memory": 44140, "step": 36188} +{"lr": 1.3022037930289352e-05, "data_time": 0.0014832019805908203, "loss": 0.46402403712272644, "time": 0.6352090835571289, "epoch": 9, "memory": 44140, "step": 36288} +{"lr": 1.2899368955286627e-05, "data_time": 0.001495528221130371, "loss": 0.4838336706161499, "time": 0.6343191146850586, "epoch": 9, "memory": 44140, "step": 36388} +{"lr": 1.2779158599568751e-05, "data_time": 0.001556730270385742, "loss": 0.4730865597724915, "time": 0.6341584205627442, "epoch": 9, "memory": 44140, "step": 36488} +{"lr": 1.2661413883256134e-05, "data_time": 0.0017279386520385742, "loss": 0.4540800988674164, "time": 0.636177945137024, "epoch": 9, "memory": 44140, "step": 36588} +{"lr": 1.2546141682479203e-05, "data_time": 0.0018882274627685547, "loss": 0.46962815523147583, "time": 0.634635066986084, "epoch": 9, "memory": 44140, "step": 36688} +{"lr": 1.2433348728976846e-05, "data_time": 0.0015498638153076173, "loss": 0.4489203214645386, "time": 0.6349020719528198, "epoch": 9, "memory": 44140, "step": 36788} +{"lr": 1.2323041609703344e-05, "data_time": 0.0015301465988159179, "loss": 0.4627111107110977, "time": 0.6345993280410767, "epoch": 9, "memory": 44140, "step": 36888} +{"lr": 1.2215226766443614e-05, "data_time": 0.003014397621154785, "loss": 0.4508346140384674, "time": 0.6346036672592164, "epoch": 9, "memory": 44140, "step": 36988} +{"accuracy/top1": 89.29381561279297, "data_time": 0.017911229530970255, "time": 1.279884970188141, "step": 9} +{"lr": 1.209847850894437e-05, "data_time": 0.0014863014221191406, "loss": 0.4583283841609955, "time": 0.6364388465881348, "epoch": 10, "memory": 44140, "step": 37099} +{"lr": 1.1995942849201055e-05, "data_time": 0.0015822887420654298, "loss": 0.4592619836330414, "time": 0.6366263866424561, "epoch": 10, "memory": 44140, "step": 37199} +{"lr": 1.189591856759183e-05, "data_time": 0.00153656005859375, "loss": 0.44315301179885863, "time": 0.6373314380645752, "epoch": 10, "memory": 44140, "step": 37299} +{"lr": 1.1798411505398016e-05, "data_time": 0.0019263982772827148, "loss": 0.45604652166366577, "time": 0.63377845287323, "epoch": 10, "memory": 44140, "step": 37399} +{"lr": 1.1703427356898678e-05, "data_time": 0.0014147520065307616, "loss": 0.4433305740356445, "time": 0.6351155519485474, "epoch": 10, "memory": 44140, "step": 37499} +{"lr": 1.1610971669038223e-05, "data_time": 0.0014008760452270507, "loss": 0.45763781666755676, "time": 0.6348682403564453, "epoch": 10, "memory": 44140, "step": 37599} +{"lr": 1.1521049841102396e-05, "data_time": 0.001594209671020508, "loss": 0.45304534435272215, "time": 0.6338958501815796, "epoch": 10, "memory": 44140, "step": 37699} +{"lr": 1.1433667124402997e-05, "data_time": 0.0023214340209960936, "loss": 0.4675642132759094, "time": 0.633301067352295, "epoch": 10, "memory": 44140, "step": 37799} +{"lr": 1.1348828621971181e-05, "data_time": 0.0015839815139770507, "loss": 0.43786188364028933, "time": 0.6339093446731567, "epoch": 10, "memory": 44140, "step": 37899} +{"lr": 1.1266539288259486e-05, "data_time": 0.0017474889755249023, "loss": 0.45742044448852537, "time": 0.6358333826065063, "epoch": 10, "memory": 44140, "step": 37999} +{"lr": 1.1186803928852477e-05, "data_time": 0.0023247241973876954, "loss": 0.4865990221500397, "time": 0.634549617767334, "epoch": 10, "memory": 44140, "step": 38099} +{"lr": 1.1109627200186125e-05, "data_time": 0.0015635967254638671, "loss": 0.48159350752830504, "time": 0.6335470676422119, "epoch": 10, "memory": 44140, "step": 38199} +{"lr": 1.1035013609275835e-05, "data_time": 0.0018955469131469727, "loss": 0.4572065591812134, "time": 0.6327837467193603, "epoch": 10, "memory": 44140, "step": 38299} +{"lr": 1.0962967513453298e-05, "data_time": 0.001545119285583496, "loss": 0.47677528858184814, "time": 0.6344039440155029, "epoch": 10, "memory": 44140, "step": 38399} +{"lr": 1.0893493120111999e-05, "data_time": 0.0015576601028442383, "loss": 0.49043325185775755, "time": 0.6342111110687256, "epoch": 10, "memory": 44140, "step": 38499} +{"lr": 1.0826594486461509e-05, "data_time": 0.002152705192565918, "loss": 0.4628440201282501, "time": 0.6397070407867431, "epoch": 10, "memory": 44140, "step": 38599} +{"lr": 1.0762275519290557e-05, "data_time": 0.0016562938690185547, "loss": 0.42665415406227114, "time": 0.6332319498062133, "epoch": 10, "memory": 44140, "step": 38699} +{"lr": 1.0700539974738862e-05, "data_time": 0.0019118309020996094, "loss": 0.4881063044071198, "time": 0.6334288120269775, "epoch": 10, "memory": 44140, "step": 38799} +{"lr": 1.064139145807783e-05, "data_time": 0.0016589641571044921, "loss": 0.47498711943626404, "time": 0.63395094871521, "epoch": 10, "memory": 44140, "step": 38899} +{"lr": 1.0584833423499947e-05, "data_time": 0.0015327215194702148, "loss": 0.5052730917930603, "time": 0.6363633632659912, "epoch": 10, "memory": 44140, "step": 38999} +{"lr": 1.0530869173917099e-05, "data_time": 0.001548910140991211, "loss": 0.4437434017658234, "time": 0.6339998006820678, "epoch": 10, "memory": 44140, "step": 39099} +{"lr": 1.0479501860767654e-05, "data_time": 0.001680135726928711, "loss": 0.47136734426021576, "time": 0.6326848983764648, "epoch": 10, "memory": 44140, "step": 39199} +{"lr": 1.043073448383247e-05, "data_time": 0.0018276214599609376, "loss": 0.4749999761581421, "time": 0.6348501920700074, "epoch": 10, "memory": 44140, "step": 39299} +{"lr": 1.038456989105968e-05, "data_time": 0.0016482114791870118, "loss": 0.4722422778606415, "time": 0.6353384256362915, "epoch": 10, "memory": 44140, "step": 39399} +{"lr": 1.034101077839838e-05, "data_time": 0.0015259504318237305, "loss": 0.4292121887207031, "time": 0.634094500541687, "epoch": 10, "memory": 44140, "step": 39499} +{"lr": 1.0300059689641194e-05, "data_time": 0.0015766143798828125, "loss": 0.46753338277339934, "time": 0.6351865053176879, "epoch": 10, "memory": 44140, "step": 39599} +{"lr": 1.0261719016275683e-05, "data_time": 0.0014925479888916015, "loss": 0.47564561367034913, "time": 0.6352946758270264, "epoch": 10, "memory": 44140, "step": 39699} +{"lr": 1.0225990997344757e-05, "data_time": 0.0018812656402587891, "loss": 0.4693823516368866, "time": 0.636930513381958, "epoch": 10, "memory": 44140, "step": 39799} +{"lr": 1.019287771931587e-05, "data_time": 0.001675868034362793, "loss": 0.468352347612381, "time": 0.635081434249878, "epoch": 10, "memory": 44140, "step": 39899} +{"lr": 1.0162381115959187e-05, "data_time": 0.00176546573638916, "loss": 0.47879849672317504, "time": 0.6356399059295654, "epoch": 10, "memory": 44140, "step": 39999} +{"lr": 1.0134502968234636e-05, "data_time": 0.0015333175659179687, "loss": 0.47510187029838563, "time": 0.6390942096710205, "epoch": 10, "memory": 44140, "step": 40099} +{"lr": 1.0109244904187918e-05, "data_time": 0.001466655731201172, "loss": 0.5071095883846283, "time": 0.6355346441268921, "epoch": 10, "memory": 44140, "step": 40199} +{"lr": 1.0086608398855446e-05, "data_time": 0.0016642570495605468, "loss": 0.4606593608856201, "time": 0.6338963031768798, "epoch": 10, "memory": 44140, "step": 40299} +{"lr": 1.0066594774178176e-05, "data_time": 0.0016315937042236327, "loss": 0.4885964959859848, "time": 0.6331888914108277, "epoch": 10, "memory": 44140, "step": 40399} +{"lr": 1.0049205198924413e-05, "data_time": 0.0015462875366210938, "loss": 0.4587945997714996, "time": 0.6341197729110718, "epoch": 10, "memory": 44140, "step": 40499} +{"lr": 1.0034440688621577e-05, "data_time": 0.0015654802322387696, "loss": 0.46378846764564513, "time": 0.6347687482833863, "epoch": 10, "memory": 44140, "step": 40599} +{"lr": 1.0022302105496896e-05, "data_time": 0.0015238523483276367, "loss": 0.481249725818634, "time": 0.6357029676437378, "epoch": 10, "memory": 44140, "step": 40699} +{"lr": 1.0012790158427017e-05, "data_time": 0.013905096054077148, "loss": 0.48698058128356936, "time": 0.6465101718902588, "epoch": 10, "memory": 44140, "step": 40799} +{"lr": 1.000590540289662e-05, "data_time": 0.0017086982727050782, "loss": 0.4825940251350403, "time": 0.6361988306045532, "epoch": 10, "memory": 44140, "step": 40899} +{"lr": 1.0001648240966046e-05, "data_time": 0.002201271057128906, "loss": 0.49574007987976076, "time": 0.6367281436920166, "epoch": 10, "memory": 44140, "step": 40999} +{"lr": 1.0000018921247718e-05, "data_time": 0.0031217336654663086, "loss": 0.4519974410533905, "time": 0.6363741159439087, "epoch": 10, "memory": 44140, "step": 41099} +{"accuracy/top1": 89.44441986083984, "data_time": 0.01780688762664795, "time": 1.279326872030894, "step": 10} diff --git a/clip_large_pretrain_4x256_stylegan3_lr1e-4/20230605_145625/vis_data/vis_image/0000000.png_1.png b/clip_large_pretrain_4x256_stylegan3_lr1e-4/20230605_145625/vis_data/vis_image/0000000.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..40eca72f0683d38a62211a6f1b759fb2d4cbdffb Binary files /dev/null and b/clip_large_pretrain_4x256_stylegan3_lr1e-4/20230605_145625/vis_data/vis_image/0000000.png_1.png differ diff --git a/clip_large_pretrain_4x256_stylegan3_lr1e-4/20230605_145625/vis_data/vis_image/0000000.png_10.png b/clip_large_pretrain_4x256_stylegan3_lr1e-4/20230605_145625/vis_data/vis_image/0000000.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..587111b3b17dbaf5914fac77488ee051db82086d Binary files /dev/null and b/clip_large_pretrain_4x256_stylegan3_lr1e-4/20230605_145625/vis_data/vis_image/0000000.png_10.png differ diff --git a/clip_large_pretrain_4x256_stylegan3_lr1e-4/20230605_145625/vis_data/vis_image/0000000.png_2.png b/clip_large_pretrain_4x256_stylegan3_lr1e-4/20230605_145625/vis_data/vis_image/0000000.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..35378c5f70897c09e5379e4e869f42e533554590 Binary files /dev/null and b/clip_large_pretrain_4x256_stylegan3_lr1e-4/20230605_145625/vis_data/vis_image/0000000.png_2.png differ diff --git a/clip_large_pretrain_4x256_stylegan3_lr1e-4/20230605_145625/vis_data/vis_image/0000000.png_3.png b/clip_large_pretrain_4x256_stylegan3_lr1e-4/20230605_145625/vis_data/vis_image/0000000.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..347c0236734c99d3685deb5232be4e4533768bf3 Binary files /dev/null and b/clip_large_pretrain_4x256_stylegan3_lr1e-4/20230605_145625/vis_data/vis_image/0000000.png_3.png differ diff --git a/clip_large_pretrain_4x256_stylegan3_lr1e-4/20230605_145625/vis_data/vis_image/0000000.png_4.png b/clip_large_pretrain_4x256_stylegan3_lr1e-4/20230605_145625/vis_data/vis_image/0000000.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..1748829a1286846b7e62b173b52068d9cc1102dd Binary files /dev/null and b/clip_large_pretrain_4x256_stylegan3_lr1e-4/20230605_145625/vis_data/vis_image/0000000.png_4.png differ diff --git a/clip_large_pretrain_4x256_stylegan3_lr1e-4/20230605_145625/vis_data/vis_image/0000000.png_5.png b/clip_large_pretrain_4x256_stylegan3_lr1e-4/20230605_145625/vis_data/vis_image/0000000.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..840841fb5443bf115458a0bbcb1c8853eb2cff15 Binary files /dev/null and b/clip_large_pretrain_4x256_stylegan3_lr1e-4/20230605_145625/vis_data/vis_image/0000000.png_5.png differ diff --git a/clip_large_pretrain_4x256_stylegan3_lr1e-4/20230605_145625/vis_data/vis_image/0000000.png_6.png b/clip_large_pretrain_4x256_stylegan3_lr1e-4/20230605_145625/vis_data/vis_image/0000000.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..6319c2aa94c79234206bc91aaef1a93bbf5c0885 Binary files /dev/null and b/clip_large_pretrain_4x256_stylegan3_lr1e-4/20230605_145625/vis_data/vis_image/0000000.png_6.png differ diff --git a/clip_large_pretrain_4x256_stylegan3_lr1e-4/20230605_145625/vis_data/vis_image/0000000.png_7.png b/clip_large_pretrain_4x256_stylegan3_lr1e-4/20230605_145625/vis_data/vis_image/0000000.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..63760929314986a69352bf7d819781206468a4d6 Binary files /dev/null and b/clip_large_pretrain_4x256_stylegan3_lr1e-4/20230605_145625/vis_data/vis_image/0000000.png_7.png differ diff --git a/clip_large_pretrain_4x256_stylegan3_lr1e-4/20230605_145625/vis_data/vis_image/0000000.png_8.png b/clip_large_pretrain_4x256_stylegan3_lr1e-4/20230605_145625/vis_data/vis_image/0000000.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..c9e86c99d3467d8186963d4aa8445f802bed2d22 Binary files /dev/null and b/clip_large_pretrain_4x256_stylegan3_lr1e-4/20230605_145625/vis_data/vis_image/0000000.png_8.png differ diff --git a/clip_large_pretrain_4x256_stylegan3_lr1e-4/20230605_145625/vis_data/vis_image/0000000.png_9.png b/clip_large_pretrain_4x256_stylegan3_lr1e-4/20230605_145625/vis_data/vis_image/0000000.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..e7519ac68eb8f95e8699af09cfbedc76d5fc56d9 Binary files /dev/null and b/clip_large_pretrain_4x256_stylegan3_lr1e-4/20230605_145625/vis_data/vis_image/0000000.png_9.png differ diff --git a/clip_large_pretrain_4x256_stylegan3_lr1e-4/20230605_145625/vis_data/vis_image/000015337.jpg_1.png b/clip_large_pretrain_4x256_stylegan3_lr1e-4/20230605_145625/vis_data/vis_image/000015337.jpg_1.png new file mode 100644 index 0000000000000000000000000000000000000000..c436ae345299c52775a5fb48004e3df6850c0699 Binary files /dev/null and b/clip_large_pretrain_4x256_stylegan3_lr1e-4/20230605_145625/vis_data/vis_image/000015337.jpg_1.png differ diff --git a/clip_large_pretrain_4x256_stylegan3_lr1e-4/20230605_145625/vis_data/vis_image/000015337.jpg_10.png b/clip_large_pretrain_4x256_stylegan3_lr1e-4/20230605_145625/vis_data/vis_image/000015337.jpg_10.png new file mode 100644 index 0000000000000000000000000000000000000000..9f9fc756fdbf32fccd6cf7ead73beae80744c77f Binary files /dev/null and b/clip_large_pretrain_4x256_stylegan3_lr1e-4/20230605_145625/vis_data/vis_image/000015337.jpg_10.png differ diff --git a/clip_large_pretrain_4x256_stylegan3_lr1e-4/20230605_145625/vis_data/vis_image/000015337.jpg_2.png b/clip_large_pretrain_4x256_stylegan3_lr1e-4/20230605_145625/vis_data/vis_image/000015337.jpg_2.png new file mode 100644 index 0000000000000000000000000000000000000000..1c33f4e944357062a8eb15c43f3024b8b22d5c48 Binary files /dev/null and b/clip_large_pretrain_4x256_stylegan3_lr1e-4/20230605_145625/vis_data/vis_image/000015337.jpg_2.png differ diff --git a/clip_large_pretrain_4x256_stylegan3_lr1e-4/20230605_145625/vis_data/vis_image/000015337.jpg_3.png b/clip_large_pretrain_4x256_stylegan3_lr1e-4/20230605_145625/vis_data/vis_image/000015337.jpg_3.png new file mode 100644 index 0000000000000000000000000000000000000000..749fa52fd9d42cb836f877586b3290a6a5fc441b Binary files /dev/null and b/clip_large_pretrain_4x256_stylegan3_lr1e-4/20230605_145625/vis_data/vis_image/000015337.jpg_3.png differ diff --git a/clip_large_pretrain_4x256_stylegan3_lr1e-4/20230605_145625/vis_data/vis_image/000015337.jpg_4.png b/clip_large_pretrain_4x256_stylegan3_lr1e-4/20230605_145625/vis_data/vis_image/000015337.jpg_4.png new file mode 100644 index 0000000000000000000000000000000000000000..be09a8ca92387d8d9cdc411c3c7821993296e47a Binary files /dev/null and b/clip_large_pretrain_4x256_stylegan3_lr1e-4/20230605_145625/vis_data/vis_image/000015337.jpg_4.png differ diff --git a/clip_large_pretrain_4x256_stylegan3_lr1e-4/20230605_145625/vis_data/vis_image/000015337.jpg_5.png b/clip_large_pretrain_4x256_stylegan3_lr1e-4/20230605_145625/vis_data/vis_image/000015337.jpg_5.png new file mode 100644 index 0000000000000000000000000000000000000000..c0976c598302d188082262c0d33d9622725c85c5 Binary files /dev/null and b/clip_large_pretrain_4x256_stylegan3_lr1e-4/20230605_145625/vis_data/vis_image/000015337.jpg_5.png differ diff --git a/clip_large_pretrain_4x256_stylegan3_lr1e-4/20230605_145625/vis_data/vis_image/000015337.jpg_6.png b/clip_large_pretrain_4x256_stylegan3_lr1e-4/20230605_145625/vis_data/vis_image/000015337.jpg_6.png new file mode 100644 index 0000000000000000000000000000000000000000..bf0f5c800aea7c2a79c551ed1d74aec711198755 Binary files /dev/null and b/clip_large_pretrain_4x256_stylegan3_lr1e-4/20230605_145625/vis_data/vis_image/000015337.jpg_6.png differ diff --git a/clip_large_pretrain_4x256_stylegan3_lr1e-4/20230605_145625/vis_data/vis_image/000015337.jpg_7.png b/clip_large_pretrain_4x256_stylegan3_lr1e-4/20230605_145625/vis_data/vis_image/000015337.jpg_7.png new file mode 100644 index 0000000000000000000000000000000000000000..fe19b2327808533a5307dad441a2038eb9a9a878 Binary files /dev/null and b/clip_large_pretrain_4x256_stylegan3_lr1e-4/20230605_145625/vis_data/vis_image/000015337.jpg_7.png differ diff --git a/clip_large_pretrain_4x256_stylegan3_lr1e-4/20230605_145625/vis_data/vis_image/000015337.jpg_8.png b/clip_large_pretrain_4x256_stylegan3_lr1e-4/20230605_145625/vis_data/vis_image/000015337.jpg_8.png new file mode 100644 index 0000000000000000000000000000000000000000..fb56834eac5992967f243ea04683287a426448a0 Binary files /dev/null and b/clip_large_pretrain_4x256_stylegan3_lr1e-4/20230605_145625/vis_data/vis_image/000015337.jpg_8.png differ diff --git a/clip_large_pretrain_4x256_stylegan3_lr1e-4/20230605_145625/vis_data/vis_image/000015337.jpg_9.png b/clip_large_pretrain_4x256_stylegan3_lr1e-4/20230605_145625/vis_data/vis_image/000015337.jpg_9.png new file mode 100644 index 0000000000000000000000000000000000000000..c3dfd4cba703d831ebeb04b7653261ba718229bf Binary files /dev/null and b/clip_large_pretrain_4x256_stylegan3_lr1e-4/20230605_145625/vis_data/vis_image/000015337.jpg_9.png differ diff --git a/clip_large_pretrain_4x256_stylegan3_lr1e-4/20230605_145625/vis_data/vis_image/0004159.png_1.png b/clip_large_pretrain_4x256_stylegan3_lr1e-4/20230605_145625/vis_data/vis_image/0004159.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..1989f290c52740754b2249527b535d757846ceac Binary files /dev/null and b/clip_large_pretrain_4x256_stylegan3_lr1e-4/20230605_145625/vis_data/vis_image/0004159.png_1.png differ diff --git a/clip_large_pretrain_4x256_stylegan3_lr1e-4/20230605_145625/vis_data/vis_image/0004159.png_10.png b/clip_large_pretrain_4x256_stylegan3_lr1e-4/20230605_145625/vis_data/vis_image/0004159.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..2168b46bb787457fec716ce746703362a55c809e Binary files /dev/null and b/clip_large_pretrain_4x256_stylegan3_lr1e-4/20230605_145625/vis_data/vis_image/0004159.png_10.png differ diff --git a/clip_large_pretrain_4x256_stylegan3_lr1e-4/20230605_145625/vis_data/vis_image/0004159.png_2.png b/clip_large_pretrain_4x256_stylegan3_lr1e-4/20230605_145625/vis_data/vis_image/0004159.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..d0b691de14eee9baffb12bfa440a56cc4b2388c4 Binary files /dev/null and b/clip_large_pretrain_4x256_stylegan3_lr1e-4/20230605_145625/vis_data/vis_image/0004159.png_2.png differ diff --git a/clip_large_pretrain_4x256_stylegan3_lr1e-4/20230605_145625/vis_data/vis_image/0004159.png_3.png b/clip_large_pretrain_4x256_stylegan3_lr1e-4/20230605_145625/vis_data/vis_image/0004159.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..00f2f68363a0c2c90b9a9f4761bca7c3478337ff Binary files /dev/null and b/clip_large_pretrain_4x256_stylegan3_lr1e-4/20230605_145625/vis_data/vis_image/0004159.png_3.png differ diff --git a/clip_large_pretrain_4x256_stylegan3_lr1e-4/20230605_145625/vis_data/vis_image/0004159.png_4.png b/clip_large_pretrain_4x256_stylegan3_lr1e-4/20230605_145625/vis_data/vis_image/0004159.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..21a0d7ba641dce91ad98770c4bd48ba6e5052153 Binary files /dev/null and b/clip_large_pretrain_4x256_stylegan3_lr1e-4/20230605_145625/vis_data/vis_image/0004159.png_4.png differ diff --git a/clip_large_pretrain_4x256_stylegan3_lr1e-4/20230605_145625/vis_data/vis_image/0004159.png_5.png b/clip_large_pretrain_4x256_stylegan3_lr1e-4/20230605_145625/vis_data/vis_image/0004159.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..a5089a5c4a063e52361c0c4513ed96214bd4990b Binary files /dev/null and b/clip_large_pretrain_4x256_stylegan3_lr1e-4/20230605_145625/vis_data/vis_image/0004159.png_5.png differ diff --git a/clip_large_pretrain_4x256_stylegan3_lr1e-4/20230605_145625/vis_data/vis_image/0004159.png_6.png b/clip_large_pretrain_4x256_stylegan3_lr1e-4/20230605_145625/vis_data/vis_image/0004159.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..3f909c62e2dd697176c2ef25018b8bb3235d6e7a Binary files /dev/null and b/clip_large_pretrain_4x256_stylegan3_lr1e-4/20230605_145625/vis_data/vis_image/0004159.png_6.png differ diff --git a/clip_large_pretrain_4x256_stylegan3_lr1e-4/20230605_145625/vis_data/vis_image/0004159.png_7.png b/clip_large_pretrain_4x256_stylegan3_lr1e-4/20230605_145625/vis_data/vis_image/0004159.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..4ba2875639313e81ec1bec76e0aa7e6f4becb459 Binary files /dev/null and b/clip_large_pretrain_4x256_stylegan3_lr1e-4/20230605_145625/vis_data/vis_image/0004159.png_7.png differ diff --git a/clip_large_pretrain_4x256_stylegan3_lr1e-4/20230605_145625/vis_data/vis_image/0004159.png_8.png b/clip_large_pretrain_4x256_stylegan3_lr1e-4/20230605_145625/vis_data/vis_image/0004159.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..963c19ebd2e49b1e2e0cd29f8076abe3288bd95e Binary files /dev/null and b/clip_large_pretrain_4x256_stylegan3_lr1e-4/20230605_145625/vis_data/vis_image/0004159.png_8.png differ diff --git a/clip_large_pretrain_4x256_stylegan3_lr1e-4/20230605_145625/vis_data/vis_image/0004159.png_9.png b/clip_large_pretrain_4x256_stylegan3_lr1e-4/20230605_145625/vis_data/vis_image/0004159.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..bdf69ea57097824f3a44bd6374934f2838616869 Binary files /dev/null and b/clip_large_pretrain_4x256_stylegan3_lr1e-4/20230605_145625/vis_data/vis_image/0004159.png_9.png differ diff --git a/clip_large_pretrain_4x256_stylegan3_lr1e-4/20230605_145625/vis_data/vis_image/0008318.png_1.png b/clip_large_pretrain_4x256_stylegan3_lr1e-4/20230605_145625/vis_data/vis_image/0008318.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..8ffc2629614c9babccb64a40ab91f57fc5feba84 Binary files /dev/null and b/clip_large_pretrain_4x256_stylegan3_lr1e-4/20230605_145625/vis_data/vis_image/0008318.png_1.png differ diff --git a/clip_large_pretrain_4x256_stylegan3_lr1e-4/20230605_145625/vis_data/vis_image/0008318.png_10.png b/clip_large_pretrain_4x256_stylegan3_lr1e-4/20230605_145625/vis_data/vis_image/0008318.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..a5394913342b4b242153e644c8cf9266de1d1d77 Binary files /dev/null and b/clip_large_pretrain_4x256_stylegan3_lr1e-4/20230605_145625/vis_data/vis_image/0008318.png_10.png differ diff --git a/clip_large_pretrain_4x256_stylegan3_lr1e-4/20230605_145625/vis_data/vis_image/0008318.png_2.png b/clip_large_pretrain_4x256_stylegan3_lr1e-4/20230605_145625/vis_data/vis_image/0008318.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..24074903cf4d74320520b7f8e8272aa553c89e7d Binary files /dev/null and b/clip_large_pretrain_4x256_stylegan3_lr1e-4/20230605_145625/vis_data/vis_image/0008318.png_2.png differ diff --git a/clip_large_pretrain_4x256_stylegan3_lr1e-4/20230605_145625/vis_data/vis_image/0008318.png_3.png b/clip_large_pretrain_4x256_stylegan3_lr1e-4/20230605_145625/vis_data/vis_image/0008318.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..f36584f515b5a408fabe55cfe4ad7dd6c08b3563 Binary files /dev/null and b/clip_large_pretrain_4x256_stylegan3_lr1e-4/20230605_145625/vis_data/vis_image/0008318.png_3.png differ diff --git a/clip_large_pretrain_4x256_stylegan3_lr1e-4/20230605_145625/vis_data/vis_image/0008318.png_4.png b/clip_large_pretrain_4x256_stylegan3_lr1e-4/20230605_145625/vis_data/vis_image/0008318.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..1d7a9d3692df39d674819575618fdd938fccd3be Binary files /dev/null and b/clip_large_pretrain_4x256_stylegan3_lr1e-4/20230605_145625/vis_data/vis_image/0008318.png_4.png differ diff --git a/clip_large_pretrain_4x256_stylegan3_lr1e-4/20230605_145625/vis_data/vis_image/0008318.png_5.png b/clip_large_pretrain_4x256_stylegan3_lr1e-4/20230605_145625/vis_data/vis_image/0008318.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..7196453dfa92c17b4a23c60c7edc22f74859024d Binary files /dev/null and b/clip_large_pretrain_4x256_stylegan3_lr1e-4/20230605_145625/vis_data/vis_image/0008318.png_5.png differ diff --git a/clip_large_pretrain_4x256_stylegan3_lr1e-4/20230605_145625/vis_data/vis_image/0008318.png_6.png b/clip_large_pretrain_4x256_stylegan3_lr1e-4/20230605_145625/vis_data/vis_image/0008318.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..e47a3a41613ff18a1f9f619ea6f29c9690991f81 Binary files /dev/null and b/clip_large_pretrain_4x256_stylegan3_lr1e-4/20230605_145625/vis_data/vis_image/0008318.png_6.png differ diff --git a/clip_large_pretrain_4x256_stylegan3_lr1e-4/20230605_145625/vis_data/vis_image/0008318.png_7.png b/clip_large_pretrain_4x256_stylegan3_lr1e-4/20230605_145625/vis_data/vis_image/0008318.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..45607817f57038de88e05d00c4176d0f816f2228 Binary files /dev/null and b/clip_large_pretrain_4x256_stylegan3_lr1e-4/20230605_145625/vis_data/vis_image/0008318.png_7.png differ diff --git a/clip_large_pretrain_4x256_stylegan3_lr1e-4/20230605_145625/vis_data/vis_image/0008318.png_8.png b/clip_large_pretrain_4x256_stylegan3_lr1e-4/20230605_145625/vis_data/vis_image/0008318.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..1a2526dc378f466d544117dfb63ef2ba7617e139 Binary files /dev/null and b/clip_large_pretrain_4x256_stylegan3_lr1e-4/20230605_145625/vis_data/vis_image/0008318.png_8.png differ diff --git a/clip_large_pretrain_4x256_stylegan3_lr1e-4/20230605_145625/vis_data/vis_image/0008318.png_9.png b/clip_large_pretrain_4x256_stylegan3_lr1e-4/20230605_145625/vis_data/vis_image/0008318.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..582ba55c829ba52cd09c2b192d8bf5addfe257fd Binary files /dev/null and b/clip_large_pretrain_4x256_stylegan3_lr1e-4/20230605_145625/vis_data/vis_image/0008318.png_9.png differ diff --git a/clip_large_pretrain_4x256_stylegan3_lr1e-4/clip_large_pretrain_4x256_stylegan3_lr1e-4.py b/clip_large_pretrain_4x256_stylegan3_lr1e-4/clip_large_pretrain_4x256_stylegan3_lr1e-4.py new file mode 100644 index 0000000000000000000000000000000000000000..fcc09894d5b0708c1c12dca7f367c370989270ef --- /dev/null +++ b/clip_large_pretrain_4x256_stylegan3_lr1e-4/clip_large_pretrain_4x256_stylegan3_lr1e-4.py @@ -0,0 +1,356 @@ +optim_wrapper = dict( + optimizer=dict( + type='AdamW', lr=0.0001, weight_decay=0.3, _scope_='mmpretrain'), + paramwise_cfg=dict( + custom_keys=dict({ + '.cls_token': dict(decay_mult=0.0), + '.pos_embed': dict(decay_mult=0.0) + })), + type='AmpOptimWrapper', + dtype='bfloat16', + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=False, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=4096) +model = dict( + type='ImageClassifier', + backbone=dict( + frozen_stages=24, + type='VisionTransformer', + arch='l', + img_size=224, + patch_size=14, + drop_rate=0.1, + pre_norm=True, + final_norm=False, + init_cfg=dict( + type='Pretrained', + checkpoint='ckpt/openclip-ViT-L-14.pth', + prefix='backbone')), + neck=dict( + type='CLIPProjection', + in_channels=1024, + out_channels=768, + init_cfg=dict( + type='Pretrained', + checkpoint='ckpt/openclip-ViT-L-14.pth', + prefix='backbone')), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=768, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + init_cfg=None), + init_cfg=dict( + type='TruncNormal', layer=['Conv2d', 'Linear'], std=0.02, bias=0.0), + train_cfg=None) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=128, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/all1.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/all2.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/all3.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3fake8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/cc1m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3real8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=128, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV2-1-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/if-dpmsolver++-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=128, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV2-1-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/if-dpmsolver++-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/clip_large_pretrain_4x256_stylegan3_lr1e-4' diff --git a/clip_large_pretrain_4x256_stylegan3_lr1e-4/epoch_1.pth b/clip_large_pretrain_4x256_stylegan3_lr1e-4/epoch_1.pth new file mode 100644 index 0000000000000000000000000000000000000000..571930ef69370bf35e4ff60f7a7fa4903479415b --- /dev/null +++ b/clip_large_pretrain_4x256_stylegan3_lr1e-4/epoch_1.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:90bcc843e883b481597efab97ef592d3742b9fe31153690dc0471e8405dbdecc +size 2438640129 diff --git a/clip_large_pretrain_4x256_stylegan3_lr1e-4/epoch_10.pth b/clip_large_pretrain_4x256_stylegan3_lr1e-4/epoch_10.pth new file mode 100644 index 0000000000000000000000000000000000000000..190f99ea98d59f8e44d02dc8db0cb1975b8d71ce --- /dev/null +++ b/clip_large_pretrain_4x256_stylegan3_lr1e-4/epoch_10.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:218745d7e56e26694a3dfe4137a471677cd7d0fdbb7cd97c465859af4fcef2ef +size 2441349569 diff --git a/clip_large_pretrain_4x256_stylegan3_lr1e-4/epoch_2.pth b/clip_large_pretrain_4x256_stylegan3_lr1e-4/epoch_2.pth new file mode 100644 index 0000000000000000000000000000000000000000..dd5728c3fbcedeebe29d97a014e07db5fd270b94 --- /dev/null +++ b/clip_large_pretrain_4x256_stylegan3_lr1e-4/epoch_2.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c487e01ae6ede419b78273d79d2588f2b84f53f76464d60b73407673f829a681 +size 2438940161 diff --git a/clip_large_pretrain_4x256_stylegan3_lr1e-4/epoch_3.pth b/clip_large_pretrain_4x256_stylegan3_lr1e-4/epoch_3.pth new file mode 100644 index 0000000000000000000000000000000000000000..8cf3b1913941bb163563970be30d9e03baa0a88b --- /dev/null +++ b/clip_large_pretrain_4x256_stylegan3_lr1e-4/epoch_3.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:133576f2e28d576d909623fad66d7a8d2af82ea562faed42581e0175912ee1d7 +size 2439239681 diff --git a/clip_large_pretrain_4x256_stylegan3_lr1e-4/epoch_4.pth b/clip_large_pretrain_4x256_stylegan3_lr1e-4/epoch_4.pth new file mode 100644 index 0000000000000000000000000000000000000000..f8e44a502871625dbdfdff92964cc2fdf474da39 --- /dev/null +++ b/clip_large_pretrain_4x256_stylegan3_lr1e-4/epoch_4.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:0ff5374fba6112032a31f6ae697576374bc67ae0fc30dc278b80623dfe507790 +size 2439539329 diff --git a/clip_large_pretrain_4x256_stylegan3_lr1e-4/epoch_5.pth b/clip_large_pretrain_4x256_stylegan3_lr1e-4/epoch_5.pth new file mode 100644 index 0000000000000000000000000000000000000000..dcf76266f3cc0e1cafe16dee3e9e52797381166e --- /dev/null +++ b/clip_large_pretrain_4x256_stylegan3_lr1e-4/epoch_5.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:fe57b417d0e63bb14e6d024ce3861d0d65f8b70457177394d123f0b9f5e16937 +size 2439839105 diff --git a/clip_large_pretrain_4x256_stylegan3_lr1e-4/epoch_6.pth b/clip_large_pretrain_4x256_stylegan3_lr1e-4/epoch_6.pth new file mode 100644 index 0000000000000000000000000000000000000000..b82a1541a734e20aa6a79fdb69074b2446da1f26 --- /dev/null +++ b/clip_large_pretrain_4x256_stylegan3_lr1e-4/epoch_6.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:23996833433722daec4157799f72015099effff95a763daa5a3cedb925753ce1 +size 2440138817 diff --git a/clip_large_pretrain_4x256_stylegan3_lr1e-4/epoch_7.pth b/clip_large_pretrain_4x256_stylegan3_lr1e-4/epoch_7.pth new file mode 100644 index 0000000000000000000000000000000000000000..2de1c4fc48345eca2bf6be4b24622896f9a172e7 --- /dev/null +++ b/clip_large_pretrain_4x256_stylegan3_lr1e-4/epoch_7.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:05a545cffcb3a57e72710fe68b65aa19a21c25cbd1b76ff786b484aa189f0c9a +size 2440439233 diff --git a/clip_large_pretrain_4x256_stylegan3_lr1e-4/epoch_8.pth b/clip_large_pretrain_4x256_stylegan3_lr1e-4/epoch_8.pth new file mode 100644 index 0000000000000000000000000000000000000000..5f8e8f2d8e17fd040d77759172528b585092dea8 --- /dev/null +++ b/clip_large_pretrain_4x256_stylegan3_lr1e-4/epoch_8.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:dc1017bf38ddda18c6a18bf06930464bb873ee2df0918605c67537349cf1ed3d +size 2440742913 diff --git a/clip_large_pretrain_4x256_stylegan3_lr1e-4/epoch_9.pth b/clip_large_pretrain_4x256_stylegan3_lr1e-4/epoch_9.pth new file mode 100644 index 0000000000000000000000000000000000000000..a7088bc4d96945f29f8ad571e5282cc9e87e2505 --- /dev/null +++ b/clip_large_pretrain_4x256_stylegan3_lr1e-4/epoch_9.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:0b5735616e3c776663517fd37fbde58904adafab4511de2094e43413d9684d13 +size 2441046209 diff --git a/clip_large_pretrain_4x256_stylegan3_lr1e-4/last_checkpoint b/clip_large_pretrain_4x256_stylegan3_lr1e-4/last_checkpoint new file mode 100644 index 0000000000000000000000000000000000000000..39ccaacaa8d4e5f5b12d9b2d7ad0b413a28d2ae9 --- /dev/null +++ b/clip_large_pretrain_4x256_stylegan3_lr1e-4/last_checkpoint @@ -0,0 +1 @@ +/mnt/petrelfs/luzeyu/workspace/fakebench/mmpretrain/workdir/clip_large_pretrain_4x256_stylegan3_lr1e-4/epoch_10.pth \ No newline at end of file diff --git a/convnext_base_8xb512_4e-3lr_5m/20230531_172616/20230531_172616.log b/convnext_base_8xb512_4e-3lr_5m/20230531_172616/20230531_172616.log new file mode 100644 index 0000000000000000000000000000000000000000..278963a17a4f554edaba410a0195ae01f5035b95 --- /dev/null +++ b/convnext_base_8xb512_4e-3lr_5m/20230531_172616/20230531_172616.log @@ -0,0 +1,1933 @@ +2023/05/31 17:26:23 - mmengine - INFO - +------------------------------------------------------------ +System environment: + sys.platform: linux + Python: 3.10.9 (main, Mar 8 2023, 10:47:38) [GCC 11.2.0] + CUDA available: True + numpy_random_seed: 1509869627 + GPU 0,1,2,3,4,5,6,7: NVIDIA A100-SXM4-80GB + CUDA_HOME: /mnt/petrelfs/share/cuda-11.6 + NVCC: Cuda compilation tools, release 11.6, V11.6.124 + GCC: gcc (GCC) 4.8.5 20150623 (Red Hat 4.8.5-44) + PyTorch: 1.13.1 + PyTorch compiling details: PyTorch built with: + - GCC 9.3 + - C++ Version: 201402 + - Intel(R) oneAPI Math Kernel Library Version 2021.4-Product Build 20210904 for Intel(R) 64 architecture applications + - Intel(R) MKL-DNN v2.6.0 (Git Hash 52b5f107dd9cf10910aaa19cb47f3abf9b349815) + - OpenMP 201511 (a.k.a. OpenMP 4.5) + - LAPACK is enabled (usually provided by MKL) + - NNPACK is enabled + - CPU capability usage: AVX2 + - CUDA Runtime 11.6 + - NVCC architecture flags: -gencode;arch=compute_37,code=sm_37;-gencode;arch=compute_50,code=sm_50;-gencode;arch=compute_60,code=sm_60;-gencode;arch=compute_61,code=sm_61;-gencode;arch=compute_70,code=sm_70;-gencode;arch=compute_75,code=sm_75;-gencode;arch=compute_80,code=sm_80;-gencode;arch=compute_86,code=sm_86;-gencode;arch=compute_37,code=compute_37 + - CuDNN 8.3.2 (built against CUDA 11.5) + - Magma 2.6.1 + - Build settings: BLAS_INFO=mkl, BUILD_TYPE=Release, CUDA_VERSION=11.6, CUDNN_VERSION=8.3.2, CXX_COMPILER=/opt/rh/devtoolset-9/root/usr/bin/c++, CXX_FLAGS= -fabi-version=11 -Wno-deprecated -fvisibility-inlines-hidden -DUSE_PTHREADPOOL -fopenmp -DNDEBUG -DUSE_KINETO -DUSE_FBGEMM -DUSE_QNNPACK -DUSE_PYTORCH_QNNPACK -DUSE_XNNPACK -DSYMBOLICATE_MOBILE_DEBUG_HANDLE -DEDGE_PROFILER_USE_KINETO -O2 -fPIC -Wno-narrowing -Wall -Wextra -Werror=return-type -Werror=non-virtual-dtor -Wno-missing-field-initializers -Wno-type-limits -Wno-array-bounds -Wno-unknown-pragmas -Wunused-local-typedefs -Wno-unused-parameter -Wno-unused-function -Wno-unused-result -Wno-strict-overflow -Wno-strict-aliasing -Wno-error=deprecated-declarations -Wno-stringop-overflow -Wno-psabi -Wno-error=pedantic -Wno-error=redundant-decls -Wno-error=old-style-cast -fdiagnostics-color=always -faligned-new -Wno-unused-but-set-variable -Wno-maybe-uninitialized -fno-math-errno -fno-trapping-math -Werror=format -Werror=cast-function-type -Wno-stringop-overflow, LAPACK_INFO=mkl, PERF_WITH_AVX=1, PERF_WITH_AVX2=1, PERF_WITH_AVX512=1, TORCH_VERSION=1.13.1, USE_CUDA=ON, USE_CUDNN=ON, USE_EXCEPTION_PTR=1, USE_GFLAGS=OFF, USE_GLOG=OFF, USE_MKL=ON, USE_MKLDNN=ON, USE_MPI=OFF, USE_NCCL=ON, USE_NNPACK=ON, USE_OPENMP=ON, USE_ROCM=OFF, + + TorchVision: 0.14.1 + OpenCV: 4.7.0 + MMEngine: 0.7.3 + +Runtime environment: + cudnn_benchmark: True + mp_cfg: {'mp_start_method': 'fork', 'opencv_num_threads': 0} + dist_cfg: {'backend': 'nccl'} + seed: None + deterministic: False + Distributed launcher: slurm + Distributed training: True + GPU number: 8 +------------------------------------------------------------ + +2023/05/31 17:26:24 - mmengine - INFO - Config: +optim_wrapper = dict( + optimizer=dict( + type='AdamW', + lr=0.004, + weight_decay=0.05, + eps=1e-08, + betas=(0.9, 0.999), + _scope_='mmpretrain'), + paramwise_cfg=dict( + norm_decay_mult=0.0, + bias_decay_mult=0.0, + flat_decay_mult=0.0, + custom_keys=dict({ + '.absolute_pos_embed': dict(decay_mult=0.0), + '.relative_position_bias_table': dict(decay_mult=0.0) + })), + type='AmpOptimWrapper', + dtype='bfloat16', + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=True, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=20, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=4096) +model = dict( + type='ImageClassifier', + backbone=dict(type='ConvNeXt', arch='base', drop_path_rate=0.5), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=1024, + loss=dict( + type='LabelSmoothLoss', label_smooth_val=0.1, mode='original'), + init_cfg=None), + init_cfg=dict( + type='TruncNormal', layer=['Conv2d', 'Linear'], std=0.02, bias=0.0), + train_cfg=None) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=512, + num_workers=10, + dataset=dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stablediffusionV1-5R2-dpmsolver-25-5m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/convnext_base_8xb512_4e-3lr_5m' + +2023/05/31 17:26:28 - mmengine - INFO - Hooks will be executed in the following order: +before_run: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_load_checkpoint: +(ABOVE_NORMAL) EMAHook + -------------------- +before_train: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_train_epoch: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(NORMAL ) DistSamplerSeedHook + -------------------- +before_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook + -------------------- +after_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +after_train_epoch: +(NORMAL ) IterTimerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_val_epoch: +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook + -------------------- +before_val_iter: +(NORMAL ) IterTimerHook + -------------------- +after_val_iter: +(NORMAL ) IterTimerHook +(NORMAL ) VisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_val_epoch: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_save_checkpoint: +(ABOVE_NORMAL) EMAHook + -------------------- +after_train: +(VERY_LOW ) CheckpointHook + -------------------- +before_test_epoch: +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook + -------------------- +before_test_iter: +(NORMAL ) IterTimerHook + -------------------- +after_test_iter: +(NORMAL ) IterTimerHook +(NORMAL ) VisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_test_epoch: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_run: +(BELOW_NORMAL) LoggerHook + -------------------- +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.0.0.bias:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.0.1.weight:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.0.1.bias:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.1.0.weight:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.1.0.bias:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.1.1.bias:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.2.0.weight:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.2.0.bias:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.2.1.bias:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.3.0.weight:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.3.0.bias:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.3.1.bias:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.0.0.gamma:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.0.0.depthwise_conv.bias:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.0.0.norm.weight:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.0.0.norm.bias:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.0.0.pointwise_conv1.bias:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.0.0.pointwise_conv2.bias:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.0.1.gamma:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.0.1.depthwise_conv.bias:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.0.1.norm.weight:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.0.1.norm.bias:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.0.1.pointwise_conv1.bias:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.0.1.pointwise_conv2.bias:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.0.2.gamma:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.0.2.depthwise_conv.bias:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.0.2.norm.weight:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.0.2.norm.bias:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.0.2.pointwise_conv1.bias:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.0.2.pointwise_conv2.bias:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.1.0.gamma:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.1.0.depthwise_conv.bias:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.1.0.norm.weight:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.1.0.norm.bias:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.1.0.pointwise_conv1.bias:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.1.0.pointwise_conv2.bias:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.1.1.gamma:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.1.1.depthwise_conv.bias:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.1.1.norm.weight:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.1.1.norm.bias:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.1.1.pointwise_conv1.bias:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.1.1.pointwise_conv2.bias:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.1.2.gamma:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.1.2.depthwise_conv.bias:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.1.2.norm.weight:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.1.2.norm.bias:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.1.2.pointwise_conv1.bias:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.1.2.pointwise_conv2.bias:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.2.0.gamma:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.2.0.depthwise_conv.bias:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.2.0.norm.weight:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.2.0.norm.bias:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.2.0.pointwise_conv1.bias:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.2.0.pointwise_conv2.bias:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.2.1.gamma:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.2.1.depthwise_conv.bias:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.2.1.norm.weight:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.2.1.norm.bias:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.2.1.pointwise_conv1.bias:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.2.1.pointwise_conv2.bias:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.2.2.gamma:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.2.2.depthwise_conv.bias:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.2.2.norm.weight:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.2.2.norm.bias:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.2.2.pointwise_conv1.bias:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.2.2.pointwise_conv2.bias:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.2.3.gamma:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.2.3.depthwise_conv.bias:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.2.3.norm.weight:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.2.3.norm.bias:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.2.3.pointwise_conv1.bias:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.2.3.pointwise_conv2.bias:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.2.4.gamma:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.2.4.depthwise_conv.bias:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.2.4.norm.weight:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.2.4.norm.bias:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.2.4.pointwise_conv1.bias:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.2.4.pointwise_conv2.bias:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.2.5.gamma:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.2.5.depthwise_conv.bias:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.2.5.norm.weight:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.2.5.norm.bias:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.2.5.pointwise_conv1.bias:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.2.5.pointwise_conv2.bias:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.2.6.gamma:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.2.6.depthwise_conv.bias:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.2.6.norm.weight:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.2.6.norm.bias:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.2.6.pointwise_conv1.bias:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.2.6.pointwise_conv2.bias:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.2.7.gamma:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.2.7.depthwise_conv.bias:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.2.7.norm.weight:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.2.7.norm.bias:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.2.7.pointwise_conv1.bias:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.2.7.pointwise_conv2.bias:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.2.8.gamma:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.2.8.depthwise_conv.bias:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.2.8.norm.weight:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.2.8.norm.bias:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.2.8.pointwise_conv1.bias:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.2.8.pointwise_conv2.bias:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.2.9.gamma:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.2.9.depthwise_conv.bias:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.2.9.norm.weight:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.2.9.norm.bias:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.2.9.pointwise_conv1.bias:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.2.9.pointwise_conv2.bias:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.2.10.gamma:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.2.10.depthwise_conv.bias:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.2.10.norm.weight:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.2.10.norm.bias:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.2.10.pointwise_conv1.bias:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.2.10.pointwise_conv2.bias:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.2.11.gamma:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.2.11.depthwise_conv.bias:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.2.11.norm.weight:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.2.11.norm.bias:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.2.11.pointwise_conv1.bias:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.2.11.pointwise_conv2.bias:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.2.12.gamma:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.2.12.depthwise_conv.bias:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.2.12.norm.weight:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.2.12.norm.bias:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.2.12.pointwise_conv1.bias:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.2.12.pointwise_conv2.bias:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.2.13.gamma:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.2.13.depthwise_conv.bias:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.2.13.norm.weight:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.2.13.norm.bias:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.2.13.pointwise_conv1.bias:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.2.13.pointwise_conv2.bias:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.2.14.gamma:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.2.14.depthwise_conv.bias:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.2.14.norm.weight:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.2.14.norm.bias:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.2.14.pointwise_conv1.bias:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.2.14.pointwise_conv2.bias:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.2.15.gamma:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.2.15.depthwise_conv.bias:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.2.15.norm.weight:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.2.15.norm.bias:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.2.15.pointwise_conv1.bias:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.2.15.pointwise_conv2.bias:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.2.16.gamma:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.2.16.depthwise_conv.bias:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.2.16.norm.weight:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.2.16.norm.bias:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.2.16.pointwise_conv1.bias:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.2.16.pointwise_conv2.bias:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.2.17.gamma:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.2.17.depthwise_conv.bias:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.2.17.norm.weight:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.2.17.norm.bias:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.2.17.pointwise_conv1.bias:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.2.17.pointwise_conv2.bias:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.2.18.gamma:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.2.18.depthwise_conv.bias:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.2.18.norm.weight:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.2.18.norm.bias:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.2.18.pointwise_conv1.bias:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.2.18.pointwise_conv2.bias:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.2.19.gamma:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.2.19.depthwise_conv.bias:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.2.19.norm.weight:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.2.19.norm.bias:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.2.19.pointwise_conv1.bias:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.2.19.pointwise_conv2.bias:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.2.20.gamma:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.2.20.depthwise_conv.bias:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.2.20.norm.weight:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.2.20.norm.bias:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.2.20.pointwise_conv1.bias:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.2.20.pointwise_conv2.bias:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.2.21.gamma:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.2.21.depthwise_conv.bias:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.2.21.norm.weight:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.2.21.norm.bias:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.2.21.pointwise_conv1.bias:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.2.21.pointwise_conv2.bias:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.2.22.gamma:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.2.22.depthwise_conv.bias:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.2.22.norm.weight:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.2.22.norm.bias:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.2.22.pointwise_conv1.bias:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.2.22.pointwise_conv2.bias:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.2.23.gamma:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.2.23.depthwise_conv.bias:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.2.23.norm.weight:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.2.23.norm.bias:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.2.23.pointwise_conv1.bias:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.2.23.pointwise_conv2.bias:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.2.24.gamma:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.2.24.depthwise_conv.bias:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.2.24.norm.weight:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.2.24.norm.bias:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.2.24.pointwise_conv1.bias:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.2.24.pointwise_conv2.bias:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.2.25.gamma:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.2.25.depthwise_conv.bias:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.2.25.norm.weight:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.2.25.norm.bias:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.2.25.pointwise_conv1.bias:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.2.25.pointwise_conv2.bias:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.2.26.gamma:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.2.26.depthwise_conv.bias:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.2.26.norm.weight:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.2.26.norm.bias:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.2.26.pointwise_conv1.bias:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.2.26.pointwise_conv2.bias:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.3.0.gamma:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.3.0.depthwise_conv.bias:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.3.0.norm.weight:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.3.0.norm.bias:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.3.0.pointwise_conv1.bias:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.3.0.pointwise_conv2.bias:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.3.1.gamma:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.3.1.depthwise_conv.bias:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.3.1.norm.weight:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.3.1.norm.bias:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.3.1.pointwise_conv1.bias:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.3.1.pointwise_conv2.bias:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.3.2.gamma:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.3.2.depthwise_conv.bias:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.3.2.norm.weight:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.3.2.norm.bias:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.3.2.pointwise_conv1.bias:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.stages.3.2.pointwise_conv2.bias:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.norm3.weight:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- backbone.norm3.bias:weight_decay=0.0 +2023/05/31 17:26:56 - mmengine - INFO - paramwise_options -- head.fc.bias:weight_decay=0.0 +Name of parameter - Initialization information + +backbone.downsample_layers.0.0.weight - torch.Size([128, 3, 4, 4]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.downsample_layers.0.0.bias - torch.Size([128]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.downsample_layers.0.1.weight - torch.Size([128]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.downsample_layers.0.1.bias - torch.Size([128]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.downsample_layers.1.0.weight - torch.Size([128]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.downsample_layers.1.0.bias - torch.Size([128]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.downsample_layers.1.1.weight - torch.Size([256, 128, 2, 2]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.downsample_layers.1.1.bias - torch.Size([256]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.downsample_layers.2.0.weight - torch.Size([256]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.downsample_layers.2.0.bias - torch.Size([256]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.downsample_layers.2.1.weight - torch.Size([512, 256, 2, 2]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.downsample_layers.2.1.bias - torch.Size([512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.downsample_layers.3.0.weight - torch.Size([512]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.downsample_layers.3.0.bias - torch.Size([512]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.downsample_layers.3.1.weight - torch.Size([1024, 512, 2, 2]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.downsample_layers.3.1.bias - torch.Size([1024]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.0.gamma - torch.Size([128]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.0.0.depthwise_conv.weight - torch.Size([128, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.0.depthwise_conv.bias - torch.Size([128]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.0.norm.weight - torch.Size([128]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.0.0.norm.bias - torch.Size([128]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.0.0.pointwise_conv1.weight - torch.Size([512, 128]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.0.pointwise_conv1.bias - torch.Size([512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.0.pointwise_conv2.weight - torch.Size([128, 512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.0.pointwise_conv2.bias - torch.Size([128]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.1.gamma - torch.Size([128]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.0.1.depthwise_conv.weight - torch.Size([128, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.1.depthwise_conv.bias - torch.Size([128]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.1.norm.weight - torch.Size([128]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.0.1.norm.bias - torch.Size([128]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.0.1.pointwise_conv1.weight - torch.Size([512, 128]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.1.pointwise_conv1.bias - torch.Size([512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.1.pointwise_conv2.weight - torch.Size([128, 512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.1.pointwise_conv2.bias - torch.Size([128]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.2.gamma - torch.Size([128]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.0.2.depthwise_conv.weight - torch.Size([128, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.2.depthwise_conv.bias - torch.Size([128]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.2.norm.weight - torch.Size([128]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.0.2.norm.bias - torch.Size([128]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.0.2.pointwise_conv1.weight - torch.Size([512, 128]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.2.pointwise_conv1.bias - torch.Size([512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.2.pointwise_conv2.weight - torch.Size([128, 512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.2.pointwise_conv2.bias - torch.Size([128]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.0.gamma - torch.Size([256]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.1.0.depthwise_conv.weight - torch.Size([256, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.0.depthwise_conv.bias - torch.Size([256]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.0.norm.weight - torch.Size([256]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.1.0.norm.bias - torch.Size([256]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.1.0.pointwise_conv1.weight - torch.Size([1024, 256]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.0.pointwise_conv1.bias - torch.Size([1024]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.0.pointwise_conv2.weight - torch.Size([256, 1024]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.0.pointwise_conv2.bias - torch.Size([256]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.1.gamma - torch.Size([256]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.1.1.depthwise_conv.weight - torch.Size([256, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.1.depthwise_conv.bias - torch.Size([256]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.1.norm.weight - torch.Size([256]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.1.1.norm.bias - torch.Size([256]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.1.1.pointwise_conv1.weight - torch.Size([1024, 256]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.1.pointwise_conv1.bias - torch.Size([1024]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.1.pointwise_conv2.weight - torch.Size([256, 1024]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.1.pointwise_conv2.bias - torch.Size([256]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.2.gamma - torch.Size([256]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.1.2.depthwise_conv.weight - torch.Size([256, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.2.depthwise_conv.bias - torch.Size([256]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.2.norm.weight - torch.Size([256]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.1.2.norm.bias - torch.Size([256]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.1.2.pointwise_conv1.weight - torch.Size([1024, 256]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.2.pointwise_conv1.bias - torch.Size([1024]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.2.pointwise_conv2.weight - torch.Size([256, 1024]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.2.pointwise_conv2.bias - torch.Size([256]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.0.gamma - torch.Size([512]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.0.depthwise_conv.weight - torch.Size([512, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.0.depthwise_conv.bias - torch.Size([512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.0.norm.weight - torch.Size([512]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.0.norm.bias - torch.Size([512]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.0.pointwise_conv1.weight - torch.Size([2048, 512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.0.pointwise_conv1.bias - torch.Size([2048]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.0.pointwise_conv2.weight - torch.Size([512, 2048]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.0.pointwise_conv2.bias - torch.Size([512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.1.gamma - torch.Size([512]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.1.depthwise_conv.weight - torch.Size([512, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.1.depthwise_conv.bias - torch.Size([512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.1.norm.weight - torch.Size([512]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.1.norm.bias - torch.Size([512]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.1.pointwise_conv1.weight - torch.Size([2048, 512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.1.pointwise_conv1.bias - torch.Size([2048]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.1.pointwise_conv2.weight - torch.Size([512, 2048]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.1.pointwise_conv2.bias - torch.Size([512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.2.gamma - torch.Size([512]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.2.depthwise_conv.weight - torch.Size([512, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.2.depthwise_conv.bias - torch.Size([512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.2.norm.weight - torch.Size([512]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.2.norm.bias - torch.Size([512]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.2.pointwise_conv1.weight - torch.Size([2048, 512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.2.pointwise_conv1.bias - torch.Size([2048]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.2.pointwise_conv2.weight - torch.Size([512, 2048]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.2.pointwise_conv2.bias - torch.Size([512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.3.gamma - torch.Size([512]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.3.depthwise_conv.weight - torch.Size([512, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.3.depthwise_conv.bias - torch.Size([512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.3.norm.weight - torch.Size([512]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.3.norm.bias - torch.Size([512]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.3.pointwise_conv1.weight - torch.Size([2048, 512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.3.pointwise_conv1.bias - torch.Size([2048]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.3.pointwise_conv2.weight - torch.Size([512, 2048]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.3.pointwise_conv2.bias - torch.Size([512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.4.gamma - torch.Size([512]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.4.depthwise_conv.weight - torch.Size([512, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.4.depthwise_conv.bias - torch.Size([512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.4.norm.weight - torch.Size([512]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.4.norm.bias - torch.Size([512]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.4.pointwise_conv1.weight - torch.Size([2048, 512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.4.pointwise_conv1.bias - torch.Size([2048]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.4.pointwise_conv2.weight - torch.Size([512, 2048]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.4.pointwise_conv2.bias - torch.Size([512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.5.gamma - torch.Size([512]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.5.depthwise_conv.weight - torch.Size([512, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.5.depthwise_conv.bias - torch.Size([512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.5.norm.weight - torch.Size([512]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.5.norm.bias - torch.Size([512]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.5.pointwise_conv1.weight - torch.Size([2048, 512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.5.pointwise_conv1.bias - torch.Size([2048]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.5.pointwise_conv2.weight - torch.Size([512, 2048]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.5.pointwise_conv2.bias - torch.Size([512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.6.gamma - torch.Size([512]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.6.depthwise_conv.weight - torch.Size([512, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.6.depthwise_conv.bias - torch.Size([512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.6.norm.weight - torch.Size([512]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.6.norm.bias - torch.Size([512]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.6.pointwise_conv1.weight - torch.Size([2048, 512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.6.pointwise_conv1.bias - torch.Size([2048]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.6.pointwise_conv2.weight - torch.Size([512, 2048]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.6.pointwise_conv2.bias - torch.Size([512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.7.gamma - torch.Size([512]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.7.depthwise_conv.weight - torch.Size([512, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.7.depthwise_conv.bias - torch.Size([512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.7.norm.weight - torch.Size([512]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.7.norm.bias - torch.Size([512]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.7.pointwise_conv1.weight - torch.Size([2048, 512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.7.pointwise_conv1.bias - torch.Size([2048]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.7.pointwise_conv2.weight - torch.Size([512, 2048]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.7.pointwise_conv2.bias - torch.Size([512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.8.gamma - torch.Size([512]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.8.depthwise_conv.weight - torch.Size([512, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.8.depthwise_conv.bias - torch.Size([512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.8.norm.weight - torch.Size([512]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.8.norm.bias - torch.Size([512]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.8.pointwise_conv1.weight - torch.Size([2048, 512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.8.pointwise_conv1.bias - torch.Size([2048]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.8.pointwise_conv2.weight - torch.Size([512, 2048]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.8.pointwise_conv2.bias - torch.Size([512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.9.gamma - torch.Size([512]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.9.depthwise_conv.weight - torch.Size([512, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.9.depthwise_conv.bias - torch.Size([512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.9.norm.weight - torch.Size([512]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.9.norm.bias - torch.Size([512]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.9.pointwise_conv1.weight - torch.Size([2048, 512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.9.pointwise_conv1.bias - torch.Size([2048]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.9.pointwise_conv2.weight - torch.Size([512, 2048]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.9.pointwise_conv2.bias - torch.Size([512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.10.gamma - torch.Size([512]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.10.depthwise_conv.weight - torch.Size([512, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.10.depthwise_conv.bias - torch.Size([512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.10.norm.weight - torch.Size([512]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.10.norm.bias - torch.Size([512]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.10.pointwise_conv1.weight - torch.Size([2048, 512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.10.pointwise_conv1.bias - torch.Size([2048]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.10.pointwise_conv2.weight - torch.Size([512, 2048]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.10.pointwise_conv2.bias - torch.Size([512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.11.gamma - torch.Size([512]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.11.depthwise_conv.weight - torch.Size([512, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.11.depthwise_conv.bias - torch.Size([512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.11.norm.weight - torch.Size([512]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.11.norm.bias - torch.Size([512]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.11.pointwise_conv1.weight - torch.Size([2048, 512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.11.pointwise_conv1.bias - torch.Size([2048]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.11.pointwise_conv2.weight - torch.Size([512, 2048]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.11.pointwise_conv2.bias - torch.Size([512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.12.gamma - torch.Size([512]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.12.depthwise_conv.weight - torch.Size([512, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.12.depthwise_conv.bias - torch.Size([512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.12.norm.weight - torch.Size([512]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.12.norm.bias - torch.Size([512]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.12.pointwise_conv1.weight - torch.Size([2048, 512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.12.pointwise_conv1.bias - torch.Size([2048]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.12.pointwise_conv2.weight - torch.Size([512, 2048]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.12.pointwise_conv2.bias - torch.Size([512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.13.gamma - torch.Size([512]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.13.depthwise_conv.weight - torch.Size([512, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.13.depthwise_conv.bias - torch.Size([512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.13.norm.weight - torch.Size([512]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.13.norm.bias - torch.Size([512]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.13.pointwise_conv1.weight - torch.Size([2048, 512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.13.pointwise_conv1.bias - torch.Size([2048]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.13.pointwise_conv2.weight - torch.Size([512, 2048]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.13.pointwise_conv2.bias - torch.Size([512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.14.gamma - torch.Size([512]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.14.depthwise_conv.weight - torch.Size([512, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.14.depthwise_conv.bias - torch.Size([512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.14.norm.weight - torch.Size([512]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.14.norm.bias - torch.Size([512]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.14.pointwise_conv1.weight - torch.Size([2048, 512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.14.pointwise_conv1.bias - torch.Size([2048]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.14.pointwise_conv2.weight - torch.Size([512, 2048]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.14.pointwise_conv2.bias - torch.Size([512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.15.gamma - torch.Size([512]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.15.depthwise_conv.weight - torch.Size([512, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.15.depthwise_conv.bias - torch.Size([512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.15.norm.weight - torch.Size([512]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.15.norm.bias - torch.Size([512]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.15.pointwise_conv1.weight - torch.Size([2048, 512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.15.pointwise_conv1.bias - torch.Size([2048]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.15.pointwise_conv2.weight - torch.Size([512, 2048]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.15.pointwise_conv2.bias - torch.Size([512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.16.gamma - torch.Size([512]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.16.depthwise_conv.weight - torch.Size([512, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.16.depthwise_conv.bias - torch.Size([512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.16.norm.weight - torch.Size([512]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.16.norm.bias - torch.Size([512]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.16.pointwise_conv1.weight - torch.Size([2048, 512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.16.pointwise_conv1.bias - torch.Size([2048]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.16.pointwise_conv2.weight - torch.Size([512, 2048]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.16.pointwise_conv2.bias - torch.Size([512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.17.gamma - torch.Size([512]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.17.depthwise_conv.weight - torch.Size([512, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.17.depthwise_conv.bias - torch.Size([512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.17.norm.weight - torch.Size([512]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.17.norm.bias - torch.Size([512]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.17.pointwise_conv1.weight - torch.Size([2048, 512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.17.pointwise_conv1.bias - torch.Size([2048]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.17.pointwise_conv2.weight - torch.Size([512, 2048]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.17.pointwise_conv2.bias - torch.Size([512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.18.gamma - torch.Size([512]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.18.depthwise_conv.weight - torch.Size([512, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.18.depthwise_conv.bias - torch.Size([512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.18.norm.weight - torch.Size([512]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.18.norm.bias - torch.Size([512]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.18.pointwise_conv1.weight - torch.Size([2048, 512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.18.pointwise_conv1.bias - torch.Size([2048]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.18.pointwise_conv2.weight - torch.Size([512, 2048]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.18.pointwise_conv2.bias - torch.Size([512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.19.gamma - torch.Size([512]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.19.depthwise_conv.weight - torch.Size([512, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.19.depthwise_conv.bias - torch.Size([512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.19.norm.weight - torch.Size([512]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.19.norm.bias - torch.Size([512]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.19.pointwise_conv1.weight - torch.Size([2048, 512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.19.pointwise_conv1.bias - torch.Size([2048]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.19.pointwise_conv2.weight - torch.Size([512, 2048]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.19.pointwise_conv2.bias - torch.Size([512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.20.gamma - torch.Size([512]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.20.depthwise_conv.weight - torch.Size([512, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.20.depthwise_conv.bias - torch.Size([512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.20.norm.weight - torch.Size([512]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.20.norm.bias - torch.Size([512]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.20.pointwise_conv1.weight - torch.Size([2048, 512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.20.pointwise_conv1.bias - torch.Size([2048]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.20.pointwise_conv2.weight - torch.Size([512, 2048]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.20.pointwise_conv2.bias - torch.Size([512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.21.gamma - torch.Size([512]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.21.depthwise_conv.weight - torch.Size([512, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.21.depthwise_conv.bias - torch.Size([512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.21.norm.weight - torch.Size([512]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.21.norm.bias - torch.Size([512]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.21.pointwise_conv1.weight - torch.Size([2048, 512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.21.pointwise_conv1.bias - torch.Size([2048]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.21.pointwise_conv2.weight - torch.Size([512, 2048]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.21.pointwise_conv2.bias - torch.Size([512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.22.gamma - torch.Size([512]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.22.depthwise_conv.weight - torch.Size([512, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.22.depthwise_conv.bias - torch.Size([512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.22.norm.weight - torch.Size([512]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.22.norm.bias - torch.Size([512]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.22.pointwise_conv1.weight - torch.Size([2048, 512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.22.pointwise_conv1.bias - torch.Size([2048]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.22.pointwise_conv2.weight - torch.Size([512, 2048]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.22.pointwise_conv2.bias - torch.Size([512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.23.gamma - torch.Size([512]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.23.depthwise_conv.weight - torch.Size([512, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.23.depthwise_conv.bias - torch.Size([512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.23.norm.weight - torch.Size([512]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.23.norm.bias - torch.Size([512]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.23.pointwise_conv1.weight - torch.Size([2048, 512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.23.pointwise_conv1.bias - torch.Size([2048]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.23.pointwise_conv2.weight - torch.Size([512, 2048]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.23.pointwise_conv2.bias - torch.Size([512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.24.gamma - torch.Size([512]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.24.depthwise_conv.weight - torch.Size([512, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.24.depthwise_conv.bias - torch.Size([512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.24.norm.weight - torch.Size([512]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.24.norm.bias - torch.Size([512]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.24.pointwise_conv1.weight - torch.Size([2048, 512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.24.pointwise_conv1.bias - torch.Size([2048]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.24.pointwise_conv2.weight - torch.Size([512, 2048]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.24.pointwise_conv2.bias - torch.Size([512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.25.gamma - torch.Size([512]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.25.depthwise_conv.weight - torch.Size([512, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.25.depthwise_conv.bias - torch.Size([512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.25.norm.weight - torch.Size([512]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.25.norm.bias - torch.Size([512]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.25.pointwise_conv1.weight - torch.Size([2048, 512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.25.pointwise_conv1.bias - torch.Size([2048]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.25.pointwise_conv2.weight - torch.Size([512, 2048]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.25.pointwise_conv2.bias - torch.Size([512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.26.gamma - torch.Size([512]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.26.depthwise_conv.weight - torch.Size([512, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.26.depthwise_conv.bias - torch.Size([512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.26.norm.weight - torch.Size([512]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.26.norm.bias - torch.Size([512]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.26.pointwise_conv1.weight - torch.Size([2048, 512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.26.pointwise_conv1.bias - torch.Size([2048]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.26.pointwise_conv2.weight - torch.Size([512, 2048]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.26.pointwise_conv2.bias - torch.Size([512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.0.gamma - torch.Size([1024]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.3.0.depthwise_conv.weight - torch.Size([1024, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.0.depthwise_conv.bias - torch.Size([1024]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.0.norm.weight - torch.Size([1024]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.3.0.norm.bias - torch.Size([1024]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.3.0.pointwise_conv1.weight - torch.Size([4096, 1024]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.0.pointwise_conv1.bias - torch.Size([4096]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.0.pointwise_conv2.weight - torch.Size([1024, 4096]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.0.pointwise_conv2.bias - torch.Size([1024]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.1.gamma - torch.Size([1024]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.3.1.depthwise_conv.weight - torch.Size([1024, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.1.depthwise_conv.bias - torch.Size([1024]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.1.norm.weight - torch.Size([1024]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.3.1.norm.bias - torch.Size([1024]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.3.1.pointwise_conv1.weight - torch.Size([4096, 1024]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.1.pointwise_conv1.bias - torch.Size([4096]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.1.pointwise_conv2.weight - torch.Size([1024, 4096]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.1.pointwise_conv2.bias - torch.Size([1024]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.2.gamma - torch.Size([1024]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.3.2.depthwise_conv.weight - torch.Size([1024, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.2.depthwise_conv.bias - torch.Size([1024]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.2.norm.weight - torch.Size([1024]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.3.2.norm.bias - torch.Size([1024]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.3.2.pointwise_conv1.weight - torch.Size([4096, 1024]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.2.pointwise_conv1.bias - torch.Size([4096]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.2.pointwise_conv2.weight - torch.Size([1024, 4096]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.2.pointwise_conv2.bias - torch.Size([1024]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.norm3.weight - torch.Size([1024]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.norm3.bias - torch.Size([1024]): +The value is the same before and after calling `init_weights` of ImageClassifier + +head.fc.weight - torch.Size([2, 1024]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +head.fc.bias - torch.Size([2]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 +2023/05/31 17:26:57 - mmengine - WARNING - "FileClient" will be deprecated in future. Please use io functions in https://mmengine.readthedocs.io/en/latest/api/fileio.html#file-io +2023/05/31 17:26:57 - mmengine - WARNING - "HardDiskBackend" is the alias of "LocalBackend" and the former will be deprecated in future. +2023/05/31 17:26:57 - mmengine - INFO - Checkpoints will be saved to /mnt/petrelfs/luzeyu/workspace/fakebench/mmpretrain/workdir/convnext_base_8xb512_4e-3lr_5m. +2023/05/31 17:31:03 - mmengine - INFO - Epoch(train) [1][ 100/1440] lr: 4.0000e-03 eta: 19:36:12 time: 2.2626 data_time: 0.2941 memory: 65239 loss: 0.6472 +2023/05/31 17:34:49 - mmengine - INFO - Epoch(train) [1][ 200/1440] lr: 4.0000e-03 eta: 18:44:16 time: 2.2346 data_time: 0.0133 memory: 65239 loss: 0.5990 +2023/05/31 17:38:36 - mmengine - INFO - Epoch(train) [1][ 300/1440] lr: 4.0000e-03 eta: 18:27:06 time: 2.2422 data_time: 0.0018 memory: 65239 loss: 0.5900 +2023/05/31 17:42:18 - mmengine - INFO - Epoch(train) [1][ 400/1440] lr: 4.0000e-03 eta: 18:09:32 time: 2.1565 data_time: 0.0021 memory: 65239 loss: 0.5610 +2023/05/31 17:45:51 - mmengine - INFO - Epoch(train) [1][ 500/1440] lr: 4.0000e-03 eta: 17:49:34 time: 2.0359 data_time: 0.0019 memory: 65239 loss: 0.5474 +2023/05/31 17:49:20 - mmengine - INFO - Epoch(train) [1][ 600/1440] lr: 4.0000e-03 eta: 17:31:51 time: 1.9755 data_time: 0.0018 memory: 65239 loss: 0.5487 +2023/05/31 17:52:47 - mmengine - INFO - Epoch(train) [1][ 700/1440] lr: 4.0000e-03 eta: 17:17:02 time: 2.1713 data_time: 0.0017 memory: 65239 loss: 0.5407 +2023/05/31 17:56:12 - mmengine - INFO - Epoch(train) [1][ 800/1440] lr: 4.0000e-03 eta: 17:03:51 time: 1.9163 data_time: 0.0019 memory: 65239 loss: 0.5277 +2023/05/31 17:59:38 - mmengine - INFO - Epoch(train) [1][ 900/1440] lr: 4.0000e-03 eta: 16:53:17 time: 2.0221 data_time: 0.0021 memory: 65239 loss: 0.5092 +2023/05/31 18:03:01 - mmengine - INFO - Exp name: convnext_base_8xb512_fake5m_20230531_172616 +2023/05/31 18:03:01 - mmengine - INFO - Epoch(train) [1][1000/1440] lr: 4.0000e-03 eta: 16:42:31 time: 2.0095 data_time: 0.0016 memory: 65239 loss: 0.5061 +2023/05/31 18:06:19 - mmengine - INFO - Epoch(train) [1][1100/1440] lr: 4.0000e-03 eta: 16:31:27 time: 2.0268 data_time: 0.0018 memory: 65239 loss: 0.4942 +2023/05/31 18:09:38 - mmengine - INFO - Epoch(train) [1][1200/1440] lr: 4.0000e-03 eta: 16:21:30 time: 2.2057 data_time: 0.0024 memory: 65239 loss: 0.4754 +2023/05/31 18:12:49 - mmengine - INFO - Epoch(train) [1][1300/1440] lr: 4.0000e-03 eta: 16:10:11 time: 1.8691 data_time: 0.0019 memory: 65239 loss: 0.4647 +2023/05/31 18:15:51 - mmengine - INFO - Epoch(train) [1][1400/1440] lr: 4.0000e-03 eta: 15:56:58 time: 1.7549 data_time: 0.0021 memory: 65239 loss: 0.4821 +2023/05/31 18:17:00 - mmengine - INFO - Exp name: convnext_base_8xb512_fake5m_20230531_172616 +2023/05/31 18:17:00 - mmengine - INFO - Saving checkpoint at 1 epochs +2023/05/31 18:17:20 - mmengine - INFO - Epoch(val) [1][8/8] accuracy/top1: 63.9543 single-label/precision_classwise: [100.0, 0.0] single-label/recall_classwise: [63.95429229736328, 0.0] single-label/f1-score_classwise: [78.01478576660156, 0.0] data_time: 0.5046 time: 1.4219 +2023/05/31 18:20:44 - mmengine - INFO - Epoch(train) [2][ 100/1440] lr: 3.9754e-03 eta: 15:45:54 time: 1.7892 data_time: 0.0020 memory: 65242 loss: 0.4259 +2023/05/31 18:23:48 - mmengine - INFO - Epoch(train) [2][ 200/1440] lr: 3.9754e-03 eta: 15:35:47 time: 1.9047 data_time: 0.0017 memory: 65239 loss: 0.4148 +2023/05/31 18:26:54 - mmengine - INFO - Epoch(train) [2][ 300/1440] lr: 3.9754e-03 eta: 15:26:58 time: 1.7980 data_time: 0.0018 memory: 65239 loss: 0.4111 +2023/05/31 18:29:58 - mmengine - INFO - Epoch(train) [2][ 400/1440] lr: 3.9754e-03 eta: 15:18:14 time: 1.7964 data_time: 0.0019 memory: 65239 loss: 0.4050 +2023/05/31 18:32:54 - mmengine - INFO - Epoch(train) [2][ 500/1440] lr: 3.9754e-03 eta: 15:08:20 time: 1.7555 data_time: 0.0018 memory: 65239 loss: 0.3943 +2023/05/31 18:34:41 - mmengine - INFO - Exp name: convnext_base_8xb512_fake5m_20230531_172616 +2023/05/31 18:35:55 - mmengine - INFO - Epoch(train) [2][ 600/1440] lr: 3.9754e-03 eta: 15:00:08 time: 1.8628 data_time: 0.0020 memory: 65239 loss: 0.3766 +2023/05/31 18:38:56 - mmengine - INFO - Epoch(train) [2][ 700/1440] lr: 3.9754e-03 eta: 14:52:27 time: 1.9785 data_time: 0.0024 memory: 65239 loss: 0.3674 +2023/05/31 18:41:56 - mmengine - INFO - Epoch(train) [2][ 800/1440] lr: 3.9754e-03 eta: 14:44:57 time: 1.8699 data_time: 0.0019 memory: 65239 loss: 0.3727 +2023/05/31 18:44:54 - mmengine - INFO - Epoch(train) [2][ 900/1440] lr: 3.9754e-03 eta: 14:37:28 time: 1.7702 data_time: 0.0020 memory: 65239 loss: 0.3354 +2023/05/31 18:47:59 - mmengine - INFO - Epoch(train) [2][1000/1440] lr: 3.9754e-03 eta: 14:31:41 time: 2.0686 data_time: 0.0023 memory: 65239 loss: 0.3583 +2023/05/31 18:51:24 - mmengine - INFO - Epoch(train) [2][1100/1440] lr: 3.9754e-03 eta: 14:29:37 time: 1.9711 data_time: 0.0017 memory: 65239 loss: 0.3402 +2023/05/31 18:54:49 - mmengine - INFO - Epoch(train) [2][1200/1440] lr: 3.9754e-03 eta: 14:27:13 time: 1.9335 data_time: 0.0021 memory: 65239 loss: 0.3163 +2023/05/31 18:58:08 - mmengine - INFO - Epoch(train) [2][1300/1440] lr: 3.9754e-03 eta: 14:24:01 time: 2.1501 data_time: 0.0018 memory: 65239 loss: 0.3279 +2023/05/31 19:01:22 - mmengine - INFO - Epoch(train) [2][1400/1440] lr: 3.9754e-03 eta: 14:19:55 time: 1.7612 data_time: 0.0017 memory: 65239 loss: 0.3360 +2023/05/31 19:02:32 - mmengine - INFO - Exp name: convnext_base_8xb512_fake5m_20230531_172616 +2023/05/31 19:02:32 - mmengine - INFO - Saving checkpoint at 2 epochs +2023/05/31 19:02:50 - mmengine - INFO - Epoch(val) [2][8/8] accuracy/top1: 69.5537 single-label/precision_classwise: [100.0, 0.0] single-label/recall_classwise: [69.5536880493164, 0.0] single-label/f1-score_classwise: [82.04326629638672, 0.0] data_time: 0.3871 time: 1.0008 +2023/05/31 19:06:10 - mmengine - INFO - Epoch(train) [3][ 100/1440] lr: 3.9024e-03 eta: 14:14:01 time: 1.8419 data_time: 0.0020 memory: 65239 loss: 0.2829 +2023/05/31 19:06:50 - mmengine - INFO - Exp name: convnext_base_8xb512_fake5m_20230531_172616 +2023/05/31 19:09:36 - mmengine - INFO - Epoch(train) [3][ 200/1440] lr: 3.9024e-03 eta: 14:11:46 time: 2.0271 data_time: 0.0025 memory: 65239 loss: 0.2768 +2023/05/31 19:12:54 - mmengine - INFO - Epoch(train) [3][ 300/1440] lr: 3.9024e-03 eta: 14:08:23 time: 1.8426 data_time: 0.0019 memory: 65239 loss: 0.5766 +2023/05/31 19:15:58 - mmengine - INFO - Epoch(train) [3][ 400/1440] lr: 3.9024e-03 eta: 14:03:10 time: 1.8540 data_time: 0.0021 memory: 65239 loss: 0.2859 +2023/05/31 19:19:05 - mmengine - INFO - Epoch(train) [3][ 500/1440] lr: 3.9024e-03 eta: 13:58:29 time: 1.8345 data_time: 0.0019 memory: 65239 loss: 0.2681 +2023/05/31 19:22:09 - mmengine - INFO - Epoch(train) [3][ 600/1440] lr: 3.9024e-03 eta: 13:53:31 time: 1.8542 data_time: 0.0017 memory: 65239 loss: 0.2809 +2023/05/31 19:25:18 - mmengine - INFO - Epoch(train) [3][ 700/1440] lr: 3.9024e-03 eta: 13:49:08 time: 1.9766 data_time: 0.0017 memory: 65239 loss: 0.2797 +2023/05/31 19:28:25 - mmengine - INFO - Epoch(train) [3][ 800/1440] lr: 3.9024e-03 eta: 13:44:43 time: 1.8463 data_time: 0.0018 memory: 65239 loss: 0.2430 +2023/05/31 19:31:27 - mmengine - INFO - Epoch(train) [3][ 900/1440] lr: 3.9024e-03 eta: 13:39:46 time: 1.7899 data_time: 0.0023 memory: 65239 loss: 0.5906 +2023/05/31 19:34:27 - mmengine - INFO - Epoch(train) [3][1000/1440] lr: 3.9024e-03 eta: 13:34:44 time: 1.8255 data_time: 0.0022 memory: 65239 loss: 0.5313 +2023/05/31 19:37:25 - mmengine - INFO - Epoch(train) [3][1100/1440] lr: 3.9024e-03 eta: 13:29:33 time: 1.7710 data_time: 0.0018 memory: 65239 loss: 0.4591 +2023/05/31 19:38:00 - mmengine - INFO - Exp name: convnext_base_8xb512_fake5m_20230531_172616 +2023/05/31 19:40:21 - mmengine - INFO - Epoch(train) [3][1200/1440] lr: 3.9024e-03 eta: 13:24:18 time: 1.7722 data_time: 0.0020 memory: 65239 loss: 0.3252 +2023/05/31 19:43:16 - mmengine - INFO - Epoch(train) [3][1300/1440] lr: 3.9024e-03 eta: 13:19:07 time: 1.7606 data_time: 0.0019 memory: 65239 loss: 0.2585 +2023/05/31 19:46:12 - mmengine - INFO - Epoch(train) [3][1400/1440] lr: 3.9024e-03 eta: 13:14:02 time: 1.7476 data_time: 0.0019 memory: 65239 loss: 0.2461 +2023/05/31 19:47:21 - mmengine - INFO - Exp name: convnext_base_8xb512_fake5m_20230531_172616 +2023/05/31 19:47:21 - mmengine - INFO - Saving checkpoint at 3 epochs +2023/05/31 19:47:39 - mmengine - INFO - Epoch(val) [3][8/8] accuracy/top1: 69.6926 single-label/precision_classwise: [100.0, 0.0] single-label/recall_classwise: [69.69257354736328, 0.0] single-label/f1-score_classwise: [82.13980865478516, 0.0] data_time: 0.3738 time: 0.9856 +2023/05/31 19:50:51 - mmengine - INFO - Epoch(train) [4][ 100/1440] lr: 3.7826e-03 eta: 13:08:27 time: 1.7445 data_time: 0.0020 memory: 65239 loss: 0.3203 +2023/05/31 19:53:46 - mmengine - INFO - Epoch(train) [4][ 200/1440] lr: 3.7826e-03 eta: 13:03:32 time: 1.7491 data_time: 0.0024 memory: 65239 loss: 0.3310 +2023/05/31 19:56:42 - mmengine - INFO - Epoch(train) [4][ 300/1440] lr: 3.7826e-03 eta: 12:58:43 time: 1.7604 data_time: 0.0021 memory: 65239 loss: 0.2420 +2023/05/31 19:59:37 - mmengine - INFO - Epoch(train) [4][ 400/1440] lr: 3.7826e-03 eta: 12:54:02 time: 1.7569 data_time: 0.0023 memory: 65239 loss: 0.2289 +2023/05/31 20:02:34 - mmengine - INFO - Epoch(train) [4][ 500/1440] lr: 3.7826e-03 eta: 12:49:28 time: 1.7629 data_time: 0.0021 memory: 65239 loss: 0.2276 +2023/05/31 20:05:29 - mmengine - INFO - Epoch(train) [4][ 600/1440] lr: 3.7826e-03 eta: 12:44:52 time: 1.7658 data_time: 0.0023 memory: 65239 loss: 0.2624 +2023/05/31 20:07:50 - mmengine - INFO - Exp name: convnext_base_8xb512_fake5m_20230531_172616 +2023/05/31 20:08:25 - mmengine - INFO - Epoch(train) [4][ 700/1440] lr: 3.7826e-03 eta: 12:40:22 time: 1.7475 data_time: 0.0020 memory: 65239 loss: 0.2266 +2023/05/31 20:11:21 - mmengine - INFO - Epoch(train) [4][ 800/1440] lr: 3.7826e-03 eta: 12:35:55 time: 1.7470 data_time: 0.0022 memory: 65239 loss: 0.4686 +2023/05/31 20:14:17 - mmengine - INFO - Epoch(train) [4][ 900/1440] lr: 3.7826e-03 eta: 12:31:33 time: 1.7661 data_time: 0.0020 memory: 65239 loss: 0.4011 +2023/05/31 20:17:12 - mmengine - INFO - Epoch(train) [4][1000/1440] lr: 3.7826e-03 eta: 12:27:12 time: 1.7664 data_time: 0.0016 memory: 65239 loss: 0.2450 +2023/05/31 20:20:08 - mmengine - INFO - Epoch(train) [4][1100/1440] lr: 3.7826e-03 eta: 12:22:57 time: 1.7601 data_time: 0.0020 memory: 65239 loss: 0.2323 +2023/05/31 20:23:04 - mmengine - INFO - Epoch(train) [4][1200/1440] lr: 3.7826e-03 eta: 12:18:43 time: 1.7838 data_time: 0.0019 memory: 65239 loss: 0.2376 +2023/05/31 20:25:59 - mmengine - INFO - Epoch(train) [4][1300/1440] lr: 3.7826e-03 eta: 12:14:30 time: 1.7745 data_time: 0.0018 memory: 65239 loss: 0.5021 +2023/05/31 20:28:55 - mmengine - INFO - Epoch(train) [4][1400/1440] lr: 3.7826e-03 eta: 12:10:21 time: 1.7649 data_time: 0.0018 memory: 65239 loss: 0.4459 +2023/05/31 20:30:04 - mmengine - INFO - Exp name: convnext_base_8xb512_fake5m_20230531_172616 +2023/05/31 20:30:04 - mmengine - INFO - Saving checkpoint at 4 epochs +2023/05/31 20:30:21 - mmengine - INFO - Epoch(val) [4][8/8] accuracy/top1: 71.3781 single-label/precision_classwise: [100.0, 0.0] single-label/recall_classwise: [71.37806701660156, 0.0] single-label/f1-score_classwise: [83.2989501953125, 0.0] data_time: 0.3789 time: 0.9783 +2023/05/31 20:33:34 - mmengine - INFO - Epoch(train) [5][ 100/1440] lr: 3.6190e-03 eta: 12:05:42 time: 1.7463 data_time: 0.0018 memory: 65239 loss: 0.3983 +2023/05/31 20:36:30 - mmengine - INFO - Epoch(train) [5][ 200/1440] lr: 3.6190e-03 eta: 12:01:38 time: 1.7427 data_time: 0.0019 memory: 65239 loss: 0.2763 +2023/05/31 20:37:41 - mmengine - INFO - Exp name: convnext_base_8xb512_fake5m_20230531_172616 +2023/05/31 20:39:26 - mmengine - INFO - Epoch(train) [5][ 300/1440] lr: 3.6190e-03 eta: 11:57:35 time: 1.7476 data_time: 0.0019 memory: 65239 loss: 0.2459 +2023/05/31 20:42:22 - mmengine - INFO - Epoch(train) [5][ 400/1440] lr: 3.6190e-03 eta: 11:53:37 time: 1.7563 data_time: 0.0024 memory: 65239 loss: 0.2366 +2023/05/31 20:45:18 - mmengine - INFO - Epoch(train) [5][ 500/1440] lr: 3.6190e-03 eta: 11:49:43 time: 1.7643 data_time: 0.0023 memory: 65239 loss: 0.2190 +2023/05/31 20:48:14 - mmengine - INFO - Epoch(train) [5][ 600/1440] lr: 3.6190e-03 eta: 11:45:48 time: 1.7546 data_time: 0.0028 memory: 65239 loss: 0.2193 +2023/05/31 20:51:10 - mmengine - INFO - Epoch(train) [5][ 700/1440] lr: 3.6190e-03 eta: 11:41:55 time: 1.7667 data_time: 0.0020 memory: 65239 loss: 0.2206 +2023/05/31 20:54:06 - mmengine - INFO - Epoch(train) [5][ 800/1440] lr: 3.6190e-03 eta: 11:38:03 time: 1.7723 data_time: 0.0023 memory: 65239 loss: 0.2114 +2023/05/31 20:57:02 - mmengine - INFO - Epoch(train) [5][ 900/1440] lr: 3.6190e-03 eta: 11:34:13 time: 1.7407 data_time: 0.0022 memory: 65239 loss: 0.2156 +2023/05/31 20:59:58 - mmengine - INFO - Epoch(train) [5][1000/1440] lr: 3.6190e-03 eta: 11:30:25 time: 1.7773 data_time: 0.0026 memory: 65239 loss: 0.6568 +2023/05/31 21:02:53 - mmengine - INFO - Epoch(train) [5][1100/1440] lr: 3.6190e-03 eta: 11:26:38 time: 1.7450 data_time: 0.0023 memory: 65239 loss: 0.5246 +2023/05/31 21:05:49 - mmengine - INFO - Epoch(train) [5][1200/1440] lr: 3.6190e-03 eta: 11:22:54 time: 1.7539 data_time: 0.0031 memory: 65239 loss: 0.4265 +2023/05/31 21:07:00 - mmengine - INFO - Exp name: convnext_base_8xb512_fake5m_20230531_172616 +2023/05/31 21:08:46 - mmengine - INFO - Epoch(train) [5][1300/1440] lr: 3.6190e-03 eta: 11:19:11 time: 1.7702 data_time: 0.0029 memory: 65239 loss: 0.2813 +2023/05/31 21:11:41 - mmengine - INFO - Epoch(train) [5][1400/1440] lr: 3.6190e-03 eta: 11:15:28 time: 1.7621 data_time: 0.0023 memory: 65239 loss: 0.5837 +2023/05/31 21:12:50 - mmengine - INFO - Exp name: convnext_base_8xb512_fake5m_20230531_172616 +2023/05/31 21:12:50 - mmengine - INFO - Saving checkpoint at 5 epochs +2023/05/31 21:13:09 - mmengine - INFO - Epoch(val) [5][8/8] accuracy/top1: 85.3040 single-label/precision_classwise: [100.0, 0.0] single-label/recall_classwise: [85.303955078125, 0.0] single-label/f1-score_classwise: [92.06922149658203, 0.0] data_time: 0.3962 time: 0.9961 +2023/05/31 21:16:23 - mmengine - INFO - Epoch(train) [6][ 100/1440] lr: 3.4157e-03 eta: 11:11:09 time: 1.7607 data_time: 0.0028 memory: 65239 loss: 0.3881 +2023/05/31 21:19:19 - mmengine - INFO - Epoch(train) [6][ 200/1440] lr: 3.4157e-03 eta: 11:07:29 time: 1.7476 data_time: 0.0024 memory: 65239 loss: 0.3754 +2023/05/31 21:22:16 - mmengine - INFO - Epoch(train) [6][ 300/1440] lr: 3.4157e-03 eta: 11:03:52 time: 1.7580 data_time: 0.0029 memory: 65239 loss: 0.2555 +2023/05/31 21:25:11 - mmengine - INFO - Epoch(train) [6][ 400/1440] lr: 3.4157e-03 eta: 11:00:12 time: 1.7511 data_time: 0.0034 memory: 65239 loss: 0.2288 +2023/05/31 21:28:07 - mmengine - INFO - Epoch(train) [6][ 500/1440] lr: 3.4157e-03 eta: 10:56:36 time: 1.7480 data_time: 0.0028 memory: 65239 loss: 0.2238 +2023/05/31 21:31:03 - mmengine - INFO - Epoch(train) [6][ 600/1440] lr: 3.4157e-03 eta: 10:53:00 time: 1.7521 data_time: 0.0022 memory: 65239 loss: 0.2270 +2023/05/31 21:33:59 - mmengine - INFO - Epoch(train) [6][ 700/1440] lr: 3.4157e-03 eta: 10:49:25 time: 1.7525 data_time: 0.0018 memory: 65239 loss: 0.2133 +2023/05/31 21:36:56 - mmengine - INFO - Exp name: convnext_base_8xb512_fake5m_20230531_172616 +2023/05/31 21:36:56 - mmengine - INFO - Epoch(train) [6][ 800/1440] lr: 3.4157e-03 eta: 10:45:53 time: 1.7570 data_time: 0.0025 memory: 65239 loss: 0.2118 +2023/05/31 21:39:52 - mmengine - INFO - Epoch(train) [6][ 900/1440] lr: 3.4157e-03 eta: 10:42:20 time: 1.7542 data_time: 0.0020 memory: 65239 loss: 0.2100 +2023/05/31 21:42:48 - mmengine - INFO - Epoch(train) [6][1000/1440] lr: 3.4157e-03 eta: 10:38:49 time: 1.7548 data_time: 0.0025 memory: 65239 loss: 0.6073 +2023/05/31 21:45:44 - mmengine - INFO - Epoch(train) [6][1100/1440] lr: 3.4157e-03 eta: 10:35:18 time: 1.7570 data_time: 0.0036 memory: 65239 loss: 0.4654 +2023/05/31 21:48:40 - mmengine - INFO - Epoch(train) [6][1200/1440] lr: 3.4157e-03 eta: 10:31:48 time: 1.7672 data_time: 0.0018 memory: 65239 loss: 0.3930 +2023/05/31 21:51:36 - mmengine - INFO - Epoch(train) [6][1300/1440] lr: 3.4157e-03 eta: 10:28:19 time: 1.7647 data_time: 0.0018 memory: 65239 loss: 0.5148 +2023/05/31 21:54:32 - mmengine - INFO - Epoch(train) [6][1400/1440] lr: 3.4157e-03 eta: 10:24:52 time: 1.7740 data_time: 0.0018 memory: 65239 loss: 0.3334 +2023/05/31 21:55:42 - mmengine - INFO - Exp name: convnext_base_8xb512_fake5m_20230531_172616 +2023/05/31 21:55:42 - mmengine - INFO - Saving checkpoint at 6 epochs +2023/05/31 21:55:59 - mmengine - INFO - Epoch(val) [6][8/8] accuracy/top1: 97.9042 single-label/precision_classwise: [100.0, 0.0] single-label/recall_classwise: [97.9041748046875, 0.0] single-label/f1-score_classwise: [98.94099426269531, 0.0] data_time: 0.3861 time: 0.9885 +2023/05/31 21:59:13 - mmengine - INFO - Epoch(train) [7][ 100/1440] lr: 3.1776e-03 eta: 10:20:38 time: 1.7676 data_time: 0.0021 memory: 65239 loss: 0.2411 +2023/05/31 22:02:09 - mmengine - INFO - Epoch(train) [7][ 200/1440] lr: 3.1776e-03 eta: 10:17:11 time: 1.7444 data_time: 0.0021 memory: 65239 loss: 0.2192 +2023/05/31 22:05:05 - mmengine - INFO - Epoch(train) [7][ 300/1440] lr: 3.1776e-03 eta: 10:13:45 time: 1.7490 data_time: 0.0024 memory: 65239 loss: 0.2204 +2023/05/31 22:06:51 - mmengine - INFO - Exp name: convnext_base_8xb512_fake5m_20230531_172616 +2023/05/31 22:08:02 - mmengine - INFO - Epoch(train) [7][ 400/1440] lr: 3.1776e-03 eta: 10:10:20 time: 1.7567 data_time: 0.0019 memory: 65239 loss: 0.2154 +2023/05/31 22:10:58 - mmengine - INFO - Epoch(train) [7][ 500/1440] lr: 3.1776e-03 eta: 10:06:54 time: 1.7602 data_time: 0.0023 memory: 65239 loss: 0.2103 +2023/05/31 22:13:54 - mmengine - INFO - Epoch(train) [7][ 600/1440] lr: 3.1776e-03 eta: 10:03:30 time: 1.7623 data_time: 0.0021 memory: 65239 loss: 0.2122 +2023/05/31 22:16:49 - mmengine - INFO - Epoch(train) [7][ 700/1440] lr: 3.1776e-03 eta: 10:00:05 time: 1.7444 data_time: 0.0020 memory: 65239 loss: 0.2121 +2023/05/31 22:19:45 - mmengine - INFO - Epoch(train) [7][ 800/1440] lr: 3.1776e-03 eta: 9:56:41 time: 1.7660 data_time: 0.0019 memory: 65239 loss: 0.4852 +2023/05/31 22:22:41 - mmengine - INFO - Epoch(train) [7][ 900/1440] lr: 3.1776e-03 eta: 9:53:18 time: 1.7837 data_time: 0.0027 memory: 65239 loss: 0.3856 +2023/05/31 22:25:36 - mmengine - INFO - Epoch(train) [7][1000/1440] lr: 3.1776e-03 eta: 9:49:54 time: 1.7563 data_time: 0.0024 memory: 65239 loss: 0.3505 +2023/05/31 22:28:32 - mmengine - INFO - Epoch(train) [7][1100/1440] lr: 3.1776e-03 eta: 9:46:32 time: 1.7536 data_time: 0.0021 memory: 65239 loss: 0.3022 +2023/05/31 22:31:28 - mmengine - INFO - Epoch(train) [7][1200/1440] lr: 3.1776e-03 eta: 9:43:11 time: 1.7506 data_time: 0.0018 memory: 65239 loss: 0.3539 +2023/05/31 22:34:24 - mmengine - INFO - Epoch(train) [7][1300/1440] lr: 3.1776e-03 eta: 9:39:50 time: 1.7663 data_time: 0.0018 memory: 65239 loss: 0.2612 +2023/05/31 22:36:09 - mmengine - INFO - Exp name: convnext_base_8xb512_fake5m_20230531_172616 +2023/05/31 22:37:20 - mmengine - INFO - Epoch(train) [7][1400/1440] lr: 3.1776e-03 eta: 9:36:29 time: 1.7686 data_time: 0.0019 memory: 65239 loss: 0.2376 +2023/05/31 22:38:29 - mmengine - INFO - Exp name: convnext_base_8xb512_fake5m_20230531_172616 +2023/05/31 22:38:29 - mmengine - INFO - Saving checkpoint at 7 epochs +2023/05/31 22:38:47 - mmengine - INFO - Epoch(val) [7][8/8] accuracy/top1: 99.5581 single-label/precision_classwise: [100.0, 0.0] single-label/recall_classwise: [99.55810546875, 0.0] single-label/f1-score_classwise: [99.778564453125, 0.0] data_time: 0.3981 time: 1.0004 +2023/05/31 22:42:01 - mmengine - INFO - Epoch(train) [8][ 100/1440] lr: 2.9107e-03 eta: 9:32:19 time: 1.7455 data_time: 0.0018 memory: 65239 loss: 0.2252 +2023/05/31 22:44:57 - mmengine - INFO - Epoch(train) [8][ 200/1440] lr: 2.9107e-03 eta: 9:28:59 time: 1.7489 data_time: 0.0029 memory: 65239 loss: 0.5732 +2023/05/31 22:47:53 - mmengine - INFO - Epoch(train) [8][ 300/1440] lr: 2.9107e-03 eta: 9:25:40 time: 1.7536 data_time: 0.0020 memory: 65239 loss: 0.3643 +2023/05/31 22:50:48 - mmengine - INFO - Epoch(train) [8][ 400/1440] lr: 2.9107e-03 eta: 9:22:20 time: 1.7556 data_time: 0.0031 memory: 65239 loss: 0.3816 +2023/05/31 22:53:56 - mmengine - INFO - Epoch(train) [8][ 500/1440] lr: 2.9107e-03 eta: 9:19:22 time: 1.7726 data_time: 0.0021 memory: 65239 loss: 0.3120 +2023/05/31 22:56:52 - mmengine - INFO - Epoch(train) [8][ 600/1440] lr: 2.9107e-03 eta: 9:16:04 time: 1.7544 data_time: 0.0034 memory: 65239 loss: 0.5555 +2023/05/31 22:59:47 - mmengine - INFO - Epoch(train) [8][ 700/1440] lr: 2.9107e-03 eta: 9:12:45 time: 1.7589 data_time: 0.0023 memory: 65239 loss: 0.3435 +2023/05/31 23:02:43 - mmengine - INFO - Epoch(train) [8][ 800/1440] lr: 2.9107e-03 eta: 9:09:27 time: 1.7548 data_time: 0.0023 memory: 65239 loss: 0.3372 +2023/05/31 23:05:39 - mmengine - INFO - Epoch(train) [8][ 900/1440] lr: 2.9107e-03 eta: 9:06:11 time: 1.7703 data_time: 0.0019 memory: 65239 loss: 0.2628 +2023/05/31 23:06:14 - mmengine - INFO - Exp name: convnext_base_8xb512_fake5m_20230531_172616 +2023/05/31 23:08:35 - mmengine - INFO - Epoch(train) [8][1000/1440] lr: 2.9107e-03 eta: 9:02:54 time: 1.7411 data_time: 0.0028 memory: 65239 loss: 0.5019 +2023/05/31 23:11:31 - mmengine - INFO - Epoch(train) [8][1100/1440] lr: 2.9107e-03 eta: 8:59:38 time: 1.7550 data_time: 0.0022 memory: 65239 loss: 0.3080 +2023/05/31 23:14:27 - mmengine - INFO - Epoch(train) [8][1200/1440] lr: 2.9107e-03 eta: 8:56:23 time: 1.7574 data_time: 0.0027 memory: 65239 loss: 0.2971 +2023/05/31 23:17:23 - mmengine - INFO - Epoch(train) [8][1300/1440] lr: 2.9107e-03 eta: 8:53:07 time: 1.7569 data_time: 0.0021 memory: 65239 loss: 0.2355 +2023/05/31 23:20:19 - mmengine - INFO - Epoch(train) [8][1400/1440] lr: 2.9107e-03 eta: 8:49:51 time: 1.7784 data_time: 0.0036 memory: 65239 loss: 0.2233 +2023/05/31 23:21:28 - mmengine - INFO - Exp name: convnext_base_8xb512_fake5m_20230531_172616 +2023/05/31 23:21:28 - mmengine - INFO - Saving checkpoint at 8 epochs +2023/05/31 23:21:46 - mmengine - INFO - Epoch(val) [8][8/8] accuracy/top1: 99.8043 single-label/precision_classwise: [100.0, 0.0] single-label/recall_classwise: [99.80430603027344, 0.0] single-label/f1-score_classwise: [99.90206146240234, 0.0] data_time: 0.3886 time: 1.0026 +2023/05/31 23:24:59 - mmengine - INFO - Epoch(train) [9][ 100/1440] lr: 2.6215e-03 eta: 8:45:42 time: 1.7626 data_time: 0.0022 memory: 65239 loss: 0.2151 +2023/05/31 23:27:55 - mmengine - INFO - Epoch(train) [9][ 200/1440] lr: 2.6215e-03 eta: 8:42:27 time: 1.7824 data_time: 0.0019 memory: 65239 loss: 0.2132 +2023/05/31 23:30:50 - mmengine - INFO - Epoch(train) [9][ 300/1440] lr: 2.6215e-03 eta: 8:39:12 time: 1.7803 data_time: 0.0021 memory: 65239 loss: 0.2547 +2023/05/31 23:33:46 - mmengine - INFO - Epoch(train) [9][ 400/1440] lr: 2.6215e-03 eta: 8:35:57 time: 1.7546 data_time: 0.0022 memory: 65239 loss: 0.3476 +2023/05/31 23:36:06 - mmengine - INFO - Exp name: convnext_base_8xb512_fake5m_20230531_172616 +2023/05/31 23:36:42 - mmengine - INFO - Epoch(train) [9][ 500/1440] lr: 2.6215e-03 eta: 8:32:43 time: 1.7414 data_time: 0.0022 memory: 65239 loss: 0.2268 +2023/05/31 23:39:38 - mmengine - INFO - Epoch(train) [9][ 600/1440] lr: 2.6215e-03 eta: 8:29:30 time: 1.7546 data_time: 0.0018 memory: 65239 loss: 0.2123 +2023/05/31 23:42:34 - mmengine - INFO - Epoch(train) [9][ 700/1440] lr: 2.6215e-03 eta: 8:26:17 time: 1.7565 data_time: 0.0017 memory: 65239 loss: 0.2129 +2023/05/31 23:45:30 - mmengine - INFO - Epoch(train) [9][ 800/1440] lr: 2.6215e-03 eta: 8:23:04 time: 1.7676 data_time: 0.0018 memory: 65239 loss: 0.2079 +2023/05/31 23:48:26 - mmengine - INFO - Epoch(train) [9][ 900/1440] lr: 2.6215e-03 eta: 8:19:51 time: 1.7574 data_time: 0.0019 memory: 65239 loss: 0.2084 +2023/05/31 23:51:22 - mmengine - INFO - Epoch(train) [9][1000/1440] lr: 2.6215e-03 eta: 8:16:39 time: 1.7647 data_time: 0.0021 memory: 65239 loss: 0.2068 +2023/05/31 23:54:18 - mmengine - INFO - Epoch(train) [9][1100/1440] lr: 2.6215e-03 eta: 8:13:26 time: 1.7684 data_time: 0.0019 memory: 65239 loss: 0.2077 +2023/05/31 23:57:14 - mmengine - INFO - Epoch(train) [9][1200/1440] lr: 2.6215e-03 eta: 8:10:15 time: 1.7409 data_time: 0.0019 memory: 65239 loss: 0.2053 +2023/06/01 00:00:10 - mmengine - INFO - Epoch(train) [9][1300/1440] lr: 2.6215e-03 eta: 8:07:03 time: 1.7627 data_time: 0.0017 memory: 65239 loss: 0.2063 +2023/06/01 00:03:06 - mmengine - INFO - Epoch(train) [9][1400/1440] lr: 2.6215e-03 eta: 8:03:52 time: 1.7660 data_time: 0.0020 memory: 65239 loss: 0.2151 +2023/06/01 00:04:15 - mmengine - INFO - Exp name: convnext_base_8xb512_fake5m_20230531_172616 +2023/06/01 00:04:15 - mmengine - INFO - Saving checkpoint at 9 epochs +2023/06/01 00:04:33 - mmengine - INFO - Epoch(val) [9][8/8] accuracy/top1: 99.9053 single-label/precision_classwise: [100.0, 0.0] single-label/recall_classwise: [99.90530395507812, 0.0] single-label/f1-score_classwise: [99.95262908935547, 0.0] data_time: 0.3988 time: 0.9993 +2023/06/01 00:06:00 - mmengine - INFO - Exp name: convnext_base_8xb512_fake5m_20230531_172616 +2023/06/01 00:07:45 - mmengine - INFO - Epoch(train) [10][ 100/1440] lr: 2.3171e-03 eta: 7:59:43 time: 1.7667 data_time: 0.0019 memory: 65239 loss: 0.2084 +2023/06/01 00:10:42 - mmengine - INFO - Epoch(train) [10][ 200/1440] lr: 2.3171e-03 eta: 7:56:32 time: 1.7612 data_time: 0.0021 memory: 65239 loss: 0.2047 +2023/06/01 00:13:37 - mmengine - INFO - Epoch(train) [10][ 300/1440] lr: 2.3171e-03 eta: 7:53:21 time: 1.7642 data_time: 0.0022 memory: 65239 loss: 0.2036 +2023/06/01 00:16:34 - mmengine - INFO - Epoch(train) [10][ 400/1440] lr: 2.3171e-03 eta: 7:50:11 time: 1.7749 data_time: 0.0021 memory: 65239 loss: 0.2058 +2023/06/01 00:19:31 - mmengine - INFO - Epoch(train) [10][ 500/1440] lr: 2.3171e-03 eta: 7:47:02 time: 1.7784 data_time: 0.0018 memory: 65239 loss: 0.2053 +2023/06/01 00:22:27 - mmengine - INFO - Epoch(train) [10][ 600/1440] lr: 2.3171e-03 eta: 7:43:51 time: 1.7422 data_time: 0.0020 memory: 65239 loss: 0.2044 +2023/06/01 00:25:22 - mmengine - INFO - Epoch(train) [10][ 700/1440] lr: 2.3171e-03 eta: 7:40:41 time: 1.7587 data_time: 0.0018 memory: 65239 loss: 0.2042 +2023/06/01 00:28:19 - mmengine - INFO - Epoch(train) [10][ 800/1440] lr: 2.3171e-03 eta: 7:37:31 time: 1.7431 data_time: 0.0019 memory: 65239 loss: 0.2045 +2023/06/01 00:31:15 - mmengine - INFO - Epoch(train) [10][ 900/1440] lr: 2.3171e-03 eta: 7:34:22 time: 1.7516 data_time: 0.0022 memory: 65239 loss: 0.5548 +2023/06/01 00:34:11 - mmengine - INFO - Epoch(train) [10][1000/1440] lr: 2.3171e-03 eta: 7:31:13 time: 1.7507 data_time: 0.0020 memory: 65239 loss: 0.4270 +2023/06/01 00:35:21 - mmengine - INFO - Exp name: convnext_base_8xb512_fake5m_20230531_172616 +2023/06/01 00:37:07 - mmengine - INFO - Epoch(train) [10][1100/1440] lr: 2.3171e-03 eta: 7:28:03 time: 1.7591 data_time: 0.0026 memory: 65239 loss: 0.5489 +2023/06/01 00:40:02 - mmengine - INFO - Epoch(train) [10][1200/1440] lr: 2.3171e-03 eta: 7:24:54 time: 1.7965 data_time: 0.0021 memory: 65239 loss: 0.3515 +2023/06/01 00:42:58 - mmengine - INFO - Epoch(train) [10][1300/1440] lr: 2.3171e-03 eta: 7:21:45 time: 1.7634 data_time: 0.0020 memory: 65239 loss: 0.2223 +2023/06/01 00:45:54 - mmengine - INFO - Epoch(train) [10][1400/1440] lr: 2.3171e-03 eta: 7:18:36 time: 1.7719 data_time: 0.0036 memory: 65239 loss: 0.2146 +2023/06/01 00:47:03 - mmengine - INFO - Exp name: convnext_base_8xb512_fake5m_20230531_172616 +2023/06/01 00:47:03 - mmengine - INFO - Saving checkpoint at 10 epochs +2023/06/01 00:47:22 - mmengine - INFO - Epoch(val) [10][8/8] accuracy/top1: 99.9053 single-label/precision_classwise: [100.0, 0.0] single-label/recall_classwise: [99.90530395507812, 0.0] single-label/f1-score_classwise: [99.95262908935547, 0.0] data_time: 0.3941 time: 1.0054 +2023/06/01 00:50:36 - mmengine - INFO - Epoch(train) [11][ 100/1440] lr: 2.0050e-03 eta: 7:14:29 time: 1.7537 data_time: 0.0022 memory: 65239 loss: 0.2082 +2023/06/01 00:53:31 - mmengine - INFO - Epoch(train) [11][ 200/1440] lr: 2.0050e-03 eta: 7:11:20 time: 1.7521 data_time: 0.0020 memory: 65239 loss: 0.2096 +2023/06/01 00:56:27 - mmengine - INFO - Epoch(train) [11][ 300/1440] lr: 2.0050e-03 eta: 7:08:11 time: 1.7487 data_time: 0.0019 memory: 65239 loss: 0.2138 +2023/06/01 00:59:22 - mmengine - INFO - Epoch(train) [11][ 400/1440] lr: 2.0050e-03 eta: 7:05:03 time: 1.7755 data_time: 0.0018 memory: 65239 loss: 0.2068 +2023/06/01 01:02:18 - mmengine - INFO - Epoch(train) [11][ 500/1440] lr: 2.0050e-03 eta: 7:01:55 time: 1.7551 data_time: 0.0018 memory: 65239 loss: 0.2052 +2023/06/01 01:05:14 - mmengine - INFO - Exp name: convnext_base_8xb512_fake5m_20230531_172616 +2023/06/01 01:05:14 - mmengine - INFO - Epoch(train) [11][ 600/1440] lr: 2.0050e-03 eta: 6:58:47 time: 1.7574 data_time: 0.0018 memory: 65239 loss: 0.2075 +2023/06/01 01:09:03 - mmengine - INFO - Epoch(train) [11][ 700/1440] lr: 2.0050e-03 eta: 6:56:27 time: 1.7552 data_time: 0.0026 memory: 65239 loss: 0.2066 +2023/06/01 01:11:59 - mmengine - INFO - Epoch(train) [11][ 800/1440] lr: 2.0050e-03 eta: 6:53:19 time: 1.7734 data_time: 0.0037 memory: 65239 loss: 0.2023 +2023/06/01 01:14:56 - mmengine - INFO - Epoch(train) [11][ 900/1440] lr: 2.0050e-03 eta: 6:50:12 time: 1.7670 data_time: 0.0028 memory: 65239 loss: 0.2042 +2023/06/01 01:17:52 - mmengine - INFO - Epoch(train) [11][1000/1440] lr: 2.0050e-03 eta: 6:47:04 time: 1.7714 data_time: 0.0031 memory: 65239 loss: 0.2037 +2023/06/01 01:20:47 - mmengine - INFO - Epoch(train) [11][1100/1440] lr: 2.0050e-03 eta: 6:43:56 time: 1.7453 data_time: 0.0028 memory: 65239 loss: 0.2071 +2023/06/01 01:23:43 - mmengine - INFO - Epoch(train) [11][1200/1440] lr: 2.0050e-03 eta: 6:40:48 time: 1.7550 data_time: 0.0022 memory: 65239 loss: 0.2048 +2023/06/01 01:26:38 - mmengine - INFO - Epoch(train) [11][1300/1440] lr: 2.0050e-03 eta: 6:37:40 time: 1.7572 data_time: 0.0019 memory: 65239 loss: 0.2043 +2023/06/01 01:29:34 - mmengine - INFO - Epoch(train) [11][1400/1440] lr: 2.0050e-03 eta: 6:34:33 time: 1.7670 data_time: 0.0040 memory: 65239 loss: 0.2048 +2023/06/01 01:30:44 - mmengine - INFO - Exp name: convnext_base_8xb512_fake5m_20230531_172616 +2023/06/01 01:30:44 - mmengine - INFO - Saving checkpoint at 11 epochs +2023/06/01 01:31:03 - mmengine - INFO - Epoch(val) [11][8/8] accuracy/top1: 99.8927 single-label/precision_classwise: [100.0, 0.0] single-label/recall_classwise: [99.89268493652344, 0.0] single-label/f1-score_classwise: [99.9463119506836, 0.0] data_time: 0.3898 time: 1.0219 +2023/06/01 01:34:15 - mmengine - INFO - Epoch(train) [12][ 100/1440] lr: 1.6929e-03 eta: 6:30:24 time: 1.7769 data_time: 0.0031 memory: 65239 loss: 0.2039 +2023/06/01 01:36:01 - mmengine - INFO - Exp name: convnext_base_8xb512_fake5m_20230531_172616 +2023/06/01 01:37:11 - mmengine - INFO - Epoch(train) [12][ 200/1440] lr: 1.6929e-03 eta: 6:27:16 time: 1.7418 data_time: 0.0026 memory: 65239 loss: 0.2013 +2023/06/01 01:40:07 - mmengine - INFO - Epoch(train) [12][ 300/1440] lr: 1.6929e-03 eta: 6:24:09 time: 1.7805 data_time: 0.0022 memory: 65239 loss: 0.2042 +2023/06/01 01:43:03 - mmengine - INFO - Epoch(train) [12][ 400/1440] lr: 1.6929e-03 eta: 6:21:03 time: 1.7641 data_time: 0.0018 memory: 65239 loss: 0.2034 +2023/06/01 01:45:59 - mmengine - INFO - Epoch(train) [12][ 500/1440] lr: 1.6929e-03 eta: 6:17:56 time: 1.7781 data_time: 0.0019 memory: 65239 loss: 0.2023 +2023/06/01 01:48:55 - mmengine - INFO - Epoch(train) [12][ 600/1440] lr: 1.6929e-03 eta: 6:14:49 time: 1.7608 data_time: 0.0023 memory: 65239 loss: 0.2011 +2023/06/01 01:51:51 - mmengine - INFO - Epoch(train) [12][ 700/1440] lr: 1.6929e-03 eta: 6:11:43 time: 1.7540 data_time: 0.0020 memory: 65239 loss: 0.2036 +2023/06/01 01:54:46 - mmengine - INFO - Epoch(train) [12][ 800/1440] lr: 1.6929e-03 eta: 6:08:37 time: 1.7440 data_time: 0.0018 memory: 65239 loss: 0.3651 +2023/06/01 01:57:42 - mmengine - INFO - Epoch(train) [12][ 900/1440] lr: 1.6929e-03 eta: 6:05:30 time: 1.7422 data_time: 0.0019 memory: 65239 loss: 0.5715 +2023/06/01 02:00:39 - mmengine - INFO - Epoch(train) [12][1000/1440] lr: 1.6929e-03 eta: 6:02:24 time: 1.7717 data_time: 0.0021 memory: 65239 loss: 0.4542 +2023/06/01 02:03:34 - mmengine - INFO - Epoch(train) [12][1100/1440] lr: 1.6929e-03 eta: 5:59:18 time: 1.7605 data_time: 0.0034 memory: 65239 loss: 0.3863 +2023/06/01 02:05:20 - mmengine - INFO - Exp name: convnext_base_8xb512_fake5m_20230531_172616 +2023/06/01 02:06:30 - mmengine - INFO - Epoch(train) [12][1200/1440] lr: 1.6929e-03 eta: 5:56:13 time: 1.7848 data_time: 0.0030 memory: 65239 loss: 0.3558 +2023/06/01 02:09:26 - mmengine - INFO - Epoch(train) [12][1300/1440] lr: 1.6929e-03 eta: 5:53:07 time: 1.7429 data_time: 0.0020 memory: 65239 loss: 0.3454 +2023/06/01 02:12:23 - mmengine - INFO - Epoch(train) [12][1400/1440] lr: 1.6929e-03 eta: 5:50:02 time: 1.7764 data_time: 0.0036 memory: 65239 loss: 0.3287 +2023/06/01 02:13:32 - mmengine - INFO - Exp name: convnext_base_8xb512_fake5m_20230531_172616 +2023/06/01 02:13:32 - mmengine - INFO - Saving checkpoint at 12 epochs +2023/06/01 02:13:50 - mmengine - INFO - Epoch(val) [12][8/8] accuracy/top1: 99.8990 single-label/precision_classwise: [100.0, 0.0] single-label/recall_classwise: [99.89899444580078, 0.0] single-label/f1-score_classwise: [99.94947814941406, 0.0] data_time: 0.3878 time: 1.0037 +2023/06/01 02:17:03 - mmengine - INFO - Epoch(train) [13][ 100/1440] lr: 1.3885e-03 eta: 5:45:52 time: 1.7988 data_time: 0.0020 memory: 65239 loss: 0.3024 +2023/06/01 02:19:59 - mmengine - INFO - Epoch(train) [13][ 200/1440] lr: 1.3885e-03 eta: 5:42:47 time: 1.7548 data_time: 0.0018 memory: 65239 loss: 0.2907 +2023/06/01 02:22:55 - mmengine - INFO - Epoch(train) [13][ 300/1440] lr: 1.3885e-03 eta: 5:39:42 time: 1.7595 data_time: 0.0019 memory: 65239 loss: 0.2900 +2023/06/01 02:25:51 - mmengine - INFO - Epoch(train) [13][ 400/1440] lr: 1.3885e-03 eta: 5:36:36 time: 1.7445 data_time: 0.0019 memory: 65239 loss: 0.2832 +2023/06/01 02:28:47 - mmengine - INFO - Epoch(train) [13][ 500/1440] lr: 1.3885e-03 eta: 5:33:31 time: 1.7530 data_time: 0.0023 memory: 65239 loss: 0.2616 +2023/06/01 02:31:43 - mmengine - INFO - Epoch(train) [13][ 600/1440] lr: 1.3885e-03 eta: 5:30:26 time: 1.7702 data_time: 0.0021 memory: 65239 loss: 0.2644 +2023/06/01 02:34:39 - mmengine - INFO - Epoch(train) [13][ 700/1440] lr: 1.3885e-03 eta: 5:27:22 time: 1.7654 data_time: 0.0018 memory: 65239 loss: 0.2523 +2023/06/01 02:35:14 - mmengine - INFO - Exp name: convnext_base_8xb512_fake5m_20230531_172616 +2023/06/01 02:37:36 - mmengine - INFO - Epoch(train) [13][ 800/1440] lr: 1.3885e-03 eta: 5:24:17 time: 1.8094 data_time: 0.0022 memory: 65239 loss: 0.2498 +2023/06/01 02:40:32 - mmengine - INFO - Epoch(train) [13][ 900/1440] lr: 1.3885e-03 eta: 5:21:12 time: 1.7453 data_time: 0.0022 memory: 65239 loss: 0.2371 +2023/06/01 02:43:28 - mmengine - INFO - Epoch(train) [13][1000/1440] lr: 1.3885e-03 eta: 5:18:08 time: 1.7417 data_time: 0.0020 memory: 65239 loss: 0.2248 +2023/06/01 02:46:24 - mmengine - INFO - Epoch(train) [13][1100/1440] lr: 1.3885e-03 eta: 5:15:03 time: 1.7676 data_time: 0.0025 memory: 65239 loss: 0.2166 +2023/06/01 02:49:20 - mmengine - INFO - Epoch(train) [13][1200/1440] lr: 1.3885e-03 eta: 5:11:59 time: 1.7819 data_time: 0.0023 memory: 65239 loss: 0.2205 +2023/06/01 02:52:16 - mmengine - INFO - Epoch(train) [13][1300/1440] lr: 1.3885e-03 eta: 5:08:55 time: 1.7519 data_time: 0.0021 memory: 65239 loss: 0.2135 +2023/06/01 02:55:13 - mmengine - INFO - Epoch(train) [13][1400/1440] lr: 1.3885e-03 eta: 5:05:51 time: 1.7685 data_time: 0.0022 memory: 65239 loss: 0.2121 +2023/06/01 02:56:22 - mmengine - INFO - Exp name: convnext_base_8xb512_fake5m_20230531_172616 +2023/06/01 02:56:22 - mmengine - INFO - Saving checkpoint at 13 epochs +2023/06/01 02:56:40 - mmengine - INFO - Epoch(val) [13][8/8] accuracy/top1: 99.9179 single-label/precision_classwise: [100.0, 0.0] single-label/recall_classwise: [99.91793823242188, 0.0] single-label/f1-score_classwise: [99.95895385742188, 0.0] data_time: 0.3861 time: 0.9991 +2023/06/01 02:59:53 - mmengine - INFO - Epoch(train) [14][ 100/1440] lr: 1.0993e-03 eta: 5:01:41 time: 1.7865 data_time: 0.0021 memory: 65239 loss: 0.2108 +2023/06/01 03:02:49 - mmengine - INFO - Epoch(train) [14][ 200/1440] lr: 1.0993e-03 eta: 4:58:37 time: 1.7428 data_time: 0.0019 memory: 65239 loss: 0.2129 +2023/06/01 03:05:10 - mmengine - INFO - Exp name: convnext_base_8xb512_fake5m_20230531_172616 +2023/06/01 03:05:45 - mmengine - INFO - Epoch(train) [14][ 300/1440] lr: 1.0993e-03 eta: 4:55:33 time: 1.7599 data_time: 0.0020 memory: 65239 loss: 0.2063 +2023/06/01 03:08:41 - mmengine - INFO - Epoch(train) [14][ 400/1440] lr: 1.0993e-03 eta: 4:52:29 time: 1.7628 data_time: 0.0024 memory: 65239 loss: 0.2080 +2023/06/01 03:11:38 - mmengine - INFO - Epoch(train) [14][ 500/1440] lr: 1.0993e-03 eta: 4:49:25 time: 1.7662 data_time: 0.0022 memory: 65239 loss: 0.2065 +2023/06/01 03:14:34 - mmengine - INFO - Epoch(train) [14][ 600/1440] lr: 1.0993e-03 eta: 4:46:22 time: 1.7968 data_time: 0.0020 memory: 65239 loss: 0.2084 +2023/06/01 03:17:30 - mmengine - INFO - Epoch(train) [14][ 700/1440] lr: 1.0993e-03 eta: 4:43:17 time: 1.7569 data_time: 0.0020 memory: 65239 loss: 0.2056 +2023/06/01 03:20:25 - mmengine - INFO - Epoch(train) [14][ 800/1440] lr: 1.0993e-03 eta: 4:40:14 time: 1.7733 data_time: 0.0021 memory: 65239 loss: 0.2044 +2023/06/01 03:23:22 - mmengine - INFO - Epoch(train) [14][ 900/1440] lr: 1.0993e-03 eta: 4:37:10 time: 1.7678 data_time: 0.0018 memory: 65239 loss: 0.2066 +2023/06/01 03:26:17 - mmengine - INFO - Epoch(train) [14][1000/1440] lr: 1.0993e-03 eta: 4:34:06 time: 1.7441 data_time: 0.0019 memory: 65239 loss: 0.2065 +2023/06/01 03:29:13 - mmengine - INFO - Epoch(train) [14][1100/1440] lr: 1.0993e-03 eta: 4:31:03 time: 1.7735 data_time: 0.0021 memory: 65239 loss: 0.2033 +2023/06/01 03:32:09 - mmengine - INFO - Epoch(train) [14][1200/1440] lr: 1.0993e-03 eta: 4:27:59 time: 1.7670 data_time: 0.0020 memory: 65239 loss: 0.2052 +2023/06/01 03:34:30 - mmengine - INFO - Exp name: convnext_base_8xb512_fake5m_20230531_172616 +2023/06/01 03:35:05 - mmengine - INFO - Epoch(train) [14][1300/1440] lr: 1.0993e-03 eta: 4:24:56 time: 1.7802 data_time: 0.0019 memory: 65239 loss: 0.2046 +2023/06/01 03:38:01 - mmengine - INFO - Epoch(train) [14][1400/1440] lr: 1.0993e-03 eta: 4:21:53 time: 1.7677 data_time: 0.0018 memory: 65239 loss: 0.2039 +2023/06/01 03:39:10 - mmengine - INFO - Exp name: convnext_base_8xb512_fake5m_20230531_172616 +2023/06/01 03:39:10 - mmengine - INFO - Saving checkpoint at 14 epochs +2023/06/01 03:39:27 - mmengine - INFO - Epoch(val) [14][8/8] accuracy/top1: 99.9432 single-label/precision_classwise: [100.0, 0.0] single-label/recall_classwise: [99.94318389892578, 0.0] single-label/f1-score_classwise: [99.97158813476562, 0.0] data_time: 0.3936 time: 0.9923 +2023/06/01 03:42:40 - mmengine - INFO - Epoch(train) [15][ 100/1440] lr: 8.3237e-04 eta: 4:17:43 time: 1.8027 data_time: 0.0021 memory: 65239 loss: 0.2032 +2023/06/01 03:45:36 - mmengine - INFO - Epoch(train) [15][ 200/1440] lr: 8.3237e-04 eta: 4:14:40 time: 1.7563 data_time: 0.0024 memory: 65239 loss: 0.2022 +2023/06/01 03:48:32 - mmengine - INFO - Epoch(train) [15][ 300/1440] lr: 8.3237e-04 eta: 4:11:37 time: 1.7427 data_time: 0.0021 memory: 65239 loss: 0.2044 +2023/06/01 03:51:29 - mmengine - INFO - Epoch(train) [15][ 400/1440] lr: 8.3237e-04 eta: 4:08:34 time: 1.7548 data_time: 0.0019 memory: 65239 loss: 0.1999 +2023/06/01 03:54:25 - mmengine - INFO - Epoch(train) [15][ 500/1440] lr: 8.3237e-04 eta: 4:05:31 time: 1.7528 data_time: 0.0021 memory: 65239 loss: 0.2024 +2023/06/01 03:57:21 - mmengine - INFO - Epoch(train) [15][ 600/1440] lr: 8.3237e-04 eta: 4:02:28 time: 1.7569 data_time: 0.0021 memory: 65239 loss: 0.2020 +2023/06/01 04:00:18 - mmengine - INFO - Epoch(train) [15][ 700/1440] lr: 8.3237e-04 eta: 3:59:26 time: 1.7670 data_time: 0.0018 memory: 65239 loss: 0.2038 +2023/06/01 04:03:13 - mmengine - INFO - Epoch(train) [15][ 800/1440] lr: 8.3237e-04 eta: 3:56:23 time: 1.7570 data_time: 0.0020 memory: 65239 loss: 0.2031 +2023/06/01 04:04:23 - mmengine - INFO - Exp name: convnext_base_8xb512_fake5m_20230531_172616 +2023/06/01 04:06:09 - mmengine - INFO - Epoch(train) [15][ 900/1440] lr: 8.3237e-04 eta: 3:53:20 time: 1.7781 data_time: 0.0018 memory: 65239 loss: 0.2023 +2023/06/01 04:09:06 - mmengine - INFO - Epoch(train) [15][1000/1440] lr: 8.3237e-04 eta: 3:50:17 time: 1.7662 data_time: 0.0024 memory: 65239 loss: 0.2008 +2023/06/01 04:12:02 - mmengine - INFO - Epoch(train) [15][1100/1440] lr: 8.3237e-04 eta: 3:47:15 time: 1.7482 data_time: 0.0020 memory: 65239 loss: 0.2030 +2023/06/01 04:14:59 - mmengine - INFO - Epoch(train) [15][1200/1440] lr: 8.3237e-04 eta: 3:44:13 time: 1.7631 data_time: 0.0023 memory: 65239 loss: 0.2092 +2023/06/01 04:17:55 - mmengine - INFO - Epoch(train) [15][1300/1440] lr: 8.3237e-04 eta: 3:41:10 time: 1.7783 data_time: 0.0030 memory: 65239 loss: 0.2023 +2023/06/01 04:20:51 - mmengine - INFO - Epoch(train) [15][1400/1440] lr: 8.3237e-04 eta: 3:38:08 time: 1.7537 data_time: 0.0024 memory: 65239 loss: 0.2046 +2023/06/01 04:22:00 - mmengine - INFO - Exp name: convnext_base_8xb512_fake5m_20230531_172616 +2023/06/01 04:22:00 - mmengine - INFO - Saving checkpoint at 15 epochs +2023/06/01 04:22:18 - mmengine - INFO - Epoch(val) [15][8/8] accuracy/top1: 99.9495 single-label/precision_classwise: [100.0, 0.0] single-label/recall_classwise: [99.94949340820312, 0.0] single-label/f1-score_classwise: [99.97473907470703, 0.0] data_time: 0.4084 time: 1.0092 +2023/06/01 04:25:31 - mmengine - INFO - Epoch(train) [16][ 100/1440] lr: 5.9432e-04 eta: 3:33:58 time: 1.7593 data_time: 0.0019 memory: 65239 loss: 0.2029 +2023/06/01 04:28:27 - mmengine - INFO - Epoch(train) [16][ 200/1440] lr: 5.9432e-04 eta: 3:30:55 time: 1.7554 data_time: 0.0024 memory: 65239 loss: 0.2003 +2023/06/01 04:31:23 - mmengine - INFO - Epoch(train) [16][ 300/1440] lr: 5.9432e-04 eta: 3:27:53 time: 1.7688 data_time: 0.0019 memory: 65239 loss: 0.2022 +2023/06/01 04:34:19 - mmengine - INFO - Exp name: convnext_base_8xb512_fake5m_20230531_172616 +2023/06/01 04:34:19 - mmengine - INFO - Epoch(train) [16][ 400/1440] lr: 5.9432e-04 eta: 3:24:51 time: 1.7510 data_time: 0.0023 memory: 65239 loss: 0.2029 +2023/06/01 04:37:15 - mmengine - INFO - Epoch(train) [16][ 500/1440] lr: 5.9432e-04 eta: 3:21:49 time: 1.7643 data_time: 0.0028 memory: 65239 loss: 0.2023 +2023/06/01 04:40:11 - mmengine - INFO - Epoch(train) [16][ 600/1440] lr: 5.9432e-04 eta: 3:18:46 time: 1.7592 data_time: 0.0024 memory: 65239 loss: 0.2015 +2023/06/01 04:43:07 - mmengine - INFO - Epoch(train) [16][ 700/1440] lr: 5.9432e-04 eta: 3:15:44 time: 1.7889 data_time: 0.0018 memory: 65239 loss: 0.2009 +2023/06/01 04:46:03 - mmengine - INFO - Epoch(train) [16][ 800/1440] lr: 5.9432e-04 eta: 3:12:42 time: 1.7687 data_time: 0.0020 memory: 65239 loss: 0.2024 +2023/06/01 04:48:59 - mmengine - INFO - Epoch(train) [16][ 900/1440] lr: 5.9432e-04 eta: 3:09:40 time: 1.7791 data_time: 0.0029 memory: 65239 loss: 0.2006 +2023/06/01 04:51:55 - mmengine - INFO - Epoch(train) [16][1000/1440] lr: 5.9432e-04 eta: 3:06:38 time: 1.7573 data_time: 0.0020 memory: 65239 loss: 0.2000 +2023/06/01 04:54:52 - mmengine - INFO - Epoch(train) [16][1100/1440] lr: 5.9432e-04 eta: 3:03:37 time: 1.7696 data_time: 0.0025 memory: 65239 loss: 0.2006 +2023/06/01 04:57:48 - mmengine - INFO - Epoch(train) [16][1200/1440] lr: 5.9432e-04 eta: 3:00:35 time: 1.7552 data_time: 0.0020 memory: 65239 loss: 0.2020 +2023/06/01 05:00:44 - mmengine - INFO - Epoch(train) [16][1300/1440] lr: 5.9432e-04 eta: 2:57:33 time: 1.7679 data_time: 0.0026 memory: 65239 loss: 0.2027 +2023/06/01 05:03:40 - mmengine - INFO - Exp name: convnext_base_8xb512_fake5m_20230531_172616 +2023/06/01 05:03:40 - mmengine - INFO - Epoch(train) [16][1400/1440] lr: 5.9432e-04 eta: 2:54:31 time: 1.7544 data_time: 0.0021 memory: 65239 loss: 0.1994 +2023/06/01 05:04:50 - mmengine - INFO - Exp name: convnext_base_8xb512_fake5m_20230531_172616 +2023/06/01 05:04:50 - mmengine - INFO - Saving checkpoint at 16 epochs +2023/06/01 05:05:08 - mmengine - INFO - Epoch(val) [16][8/8] accuracy/top1: 99.9684 single-label/precision_classwise: [100.0, 0.0] single-label/recall_classwise: [99.96842956542969, 0.0] single-label/f1-score_classwise: [99.98420715332031, 0.0] data_time: 0.3825 time: 1.0150 +2023/06/01 05:08:21 - mmengine - INFO - Epoch(train) [17][ 100/1440] lr: 3.9101e-04 eta: 2:50:21 time: 1.7573 data_time: 0.0020 memory: 65239 loss: 0.2008 +2023/06/01 05:11:17 - mmengine - INFO - Epoch(train) [17][ 200/1440] lr: 3.9101e-04 eta: 2:47:19 time: 1.7546 data_time: 0.0027 memory: 65239 loss: 0.2009 +2023/06/01 05:14:13 - mmengine - INFO - Epoch(train) [17][ 300/1440] lr: 3.9101e-04 eta: 2:44:18 time: 1.7777 data_time: 0.0019 memory: 65239 loss: 0.2001 +2023/06/01 05:17:09 - mmengine - INFO - Epoch(train) [17][ 400/1440] lr: 3.9101e-04 eta: 2:41:16 time: 1.7529 data_time: 0.0019 memory: 65239 loss: 0.2014 +2023/06/01 05:20:05 - mmengine - INFO - Epoch(train) [17][ 500/1440] lr: 3.9101e-04 eta: 2:38:15 time: 1.7707 data_time: 0.0022 memory: 65239 loss: 0.2008 +2023/06/01 05:23:00 - mmengine - INFO - Epoch(train) [17][ 600/1440] lr: 3.9101e-04 eta: 2:35:13 time: 1.7420 data_time: 0.0021 memory: 65239 loss: 0.2005 +2023/06/01 05:25:57 - mmengine - INFO - Epoch(train) [17][ 700/1440] lr: 3.9101e-04 eta: 2:32:12 time: 1.7537 data_time: 0.0020 memory: 65239 loss: 0.2009 +2023/06/01 05:28:53 - mmengine - INFO - Epoch(train) [17][ 800/1440] lr: 3.9101e-04 eta: 2:29:10 time: 1.7783 data_time: 0.0022 memory: 65239 loss: 0.1998 +2023/06/01 05:31:49 - mmengine - INFO - Epoch(train) [17][ 900/1440] lr: 3.9101e-04 eta: 2:26:09 time: 1.7694 data_time: 0.0019 memory: 65239 loss: 0.2008 +2023/06/01 05:33:34 - mmengine - INFO - Exp name: convnext_base_8xb512_fake5m_20230531_172616 +2023/06/01 05:34:44 - mmengine - INFO - Epoch(train) [17][1000/1440] lr: 3.9101e-04 eta: 2:23:07 time: 1.7735 data_time: 0.0020 memory: 65239 loss: 0.2020 +2023/06/01 05:37:40 - mmengine - INFO - Epoch(train) [17][1100/1440] lr: 3.9101e-04 eta: 2:20:06 time: 1.7670 data_time: 0.0027 memory: 65239 loss: 0.2002 +2023/06/01 05:40:35 - mmengine - INFO - Epoch(train) [17][1200/1440] lr: 3.9101e-04 eta: 2:17:05 time: 1.7412 data_time: 0.0024 memory: 65239 loss: 0.1992 +2023/06/01 05:43:31 - mmengine - INFO - Epoch(train) [17][1300/1440] lr: 3.9101e-04 eta: 2:14:04 time: 1.7559 data_time: 0.0020 memory: 65239 loss: 0.1998 +2023/06/01 05:46:27 - mmengine - INFO - Epoch(train) [17][1400/1440] lr: 3.9101e-04 eta: 2:11:02 time: 1.7551 data_time: 0.0021 memory: 65239 loss: 0.1996 +2023/06/01 05:47:36 - mmengine - INFO - Exp name: convnext_base_8xb512_fake5m_20230531_172616 +2023/06/01 05:47:36 - mmengine - INFO - Saving checkpoint at 17 epochs +2023/06/01 05:47:54 - mmengine - INFO - Epoch(val) [17][8/8] accuracy/top1: 99.9811 single-label/precision_classwise: [100.0, 0.0] single-label/recall_classwise: [99.98106384277344, 0.0] single-label/f1-score_classwise: [99.99053192138672, 0.0] data_time: 0.3806 time: 0.9813 +2023/06/01 05:51:07 - mmengine - INFO - Epoch(train) [18][ 100/1440] lr: 2.2744e-04 eta: 2:06:52 time: 1.7524 data_time: 0.0018 memory: 65239 loss: 0.2004 +2023/06/01 05:54:03 - mmengine - INFO - Epoch(train) [18][ 200/1440] lr: 2.2744e-04 eta: 2:03:50 time: 1.7502 data_time: 0.0020 memory: 65239 loss: 0.1993 +2023/06/01 05:56:59 - mmengine - INFO - Epoch(train) [18][ 300/1440] lr: 2.2744e-04 eta: 2:00:49 time: 1.7579 data_time: 0.0019 memory: 65239 loss: 0.1993 +2023/06/01 05:59:54 - mmengine - INFO - Epoch(train) [18][ 400/1440] lr: 2.2744e-04 eta: 1:57:48 time: 1.7430 data_time: 0.0021 memory: 65239 loss: 0.2005 +2023/06/01 06:02:50 - mmengine - INFO - Epoch(train) [18][ 500/1440] lr: 2.2744e-04 eta: 1:54:47 time: 1.7415 data_time: 0.0023 memory: 65239 loss: 0.1993 +2023/06/01 06:03:25 - mmengine - INFO - Exp name: convnext_base_8xb512_fake5m_20230531_172616 +2023/06/01 06:05:47 - mmengine - INFO - Epoch(train) [18][ 600/1440] lr: 2.2744e-04 eta: 1:51:46 time: 1.7595 data_time: 0.0022 memory: 65239 loss: 0.2001 +2023/06/01 06:08:43 - mmengine - INFO - Epoch(train) [18][ 700/1440] lr: 2.2744e-04 eta: 1:48:45 time: 1.7668 data_time: 0.0022 memory: 65239 loss: 0.2001 +2023/06/01 06:11:40 - mmengine - INFO - Epoch(train) [18][ 800/1440] lr: 2.2744e-04 eta: 1:45:45 time: 1.7585 data_time: 0.0021 memory: 65239 loss: 0.1998 +2023/06/01 06:14:36 - mmengine - INFO - Epoch(train) [18][ 900/1440] lr: 2.2744e-04 eta: 1:42:44 time: 1.7640 data_time: 0.0021 memory: 65239 loss: 0.1996 +2023/06/01 06:17:32 - mmengine - INFO - Epoch(train) [18][1000/1440] lr: 2.2744e-04 eta: 1:39:43 time: 1.7666 data_time: 0.0018 memory: 65239 loss: 0.2000 +2023/06/01 06:20:28 - mmengine - INFO - Epoch(train) [18][1100/1440] lr: 2.2744e-04 eta: 1:36:42 time: 1.7692 data_time: 0.0019 memory: 65239 loss: 0.2011 +2023/06/01 06:23:24 - mmengine - INFO - Epoch(train) [18][1200/1440] lr: 2.2744e-04 eta: 1:33:42 time: 1.7574 data_time: 0.0021 memory: 65239 loss: 0.1992 +2023/06/01 06:26:21 - mmengine - INFO - Epoch(train) [18][1300/1440] lr: 2.2744e-04 eta: 1:30:41 time: 1.7659 data_time: 0.0020 memory: 65239 loss: 0.2005 +2023/06/01 06:29:17 - mmengine - INFO - Epoch(train) [18][1400/1440] lr: 2.2744e-04 eta: 1:27:40 time: 1.7594 data_time: 0.0026 memory: 65239 loss: 0.1996 +2023/06/01 06:30:26 - mmengine - INFO - Exp name: convnext_base_8xb512_fake5m_20230531_172616 +2023/06/01 06:30:26 - mmengine - INFO - Saving checkpoint at 18 epochs +2023/06/01 06:30:44 - mmengine - INFO - Epoch(val) [18][8/8] accuracy/top1: 99.9874 single-label/precision_classwise: [100.0, 0.0] single-label/recall_classwise: [99.98737335205078, 0.0] single-label/f1-score_classwise: [99.99369049072266, 0.0] data_time: 0.3822 time: 0.9836 +2023/06/01 06:33:22 - mmengine - INFO - Exp name: convnext_base_8xb512_fake5m_20230531_172616 +2023/06/01 06:33:57 - mmengine - INFO - Epoch(train) [19][ 100/1440] lr: 1.0764e-04 eta: 1:23:29 time: 1.7427 data_time: 0.0024 memory: 65239 loss: 0.1998 +2023/06/01 06:36:53 - mmengine - INFO - Epoch(train) [19][ 200/1440] lr: 1.0764e-04 eta: 1:20:29 time: 1.7590 data_time: 0.0017 memory: 65239 loss: 0.2003 +2023/06/01 06:39:49 - mmengine - INFO - Epoch(train) [19][ 300/1440] lr: 1.0764e-04 eta: 1:17:28 time: 1.7673 data_time: 0.0024 memory: 65239 loss: 0.1999 +2023/06/01 06:42:45 - mmengine - INFO - Epoch(train) [19][ 400/1440] lr: 1.0764e-04 eta: 1:14:27 time: 1.7429 data_time: 0.0021 memory: 65239 loss: 0.1987 +2023/06/01 06:45:40 - mmengine - INFO - Epoch(train) [19][ 500/1440] lr: 1.0764e-04 eta: 1:11:27 time: 1.7448 data_time: 0.0021 memory: 65239 loss: 0.1986 +2023/06/01 06:48:36 - mmengine - INFO - Epoch(train) [19][ 600/1440] lr: 1.0764e-04 eta: 1:08:26 time: 1.7682 data_time: 0.0022 memory: 65239 loss: 0.1990 +2023/06/01 06:51:32 - mmengine - INFO - Epoch(train) [19][ 700/1440] lr: 1.0764e-04 eta: 1:05:26 time: 1.7537 data_time: 0.0022 memory: 65239 loss: 0.1993 +2023/06/01 06:54:28 - mmengine - INFO - Epoch(train) [19][ 800/1440] lr: 1.0764e-04 eta: 1:02:25 time: 1.7775 data_time: 0.0024 memory: 65239 loss: 0.1993 +2023/06/01 06:57:24 - mmengine - INFO - Epoch(train) [19][ 900/1440] lr: 1.0764e-04 eta: 0:59:25 time: 1.7686 data_time: 0.0022 memory: 65239 loss: 0.1995 +2023/06/01 07:00:20 - mmengine - INFO - Epoch(train) [19][1000/1440] lr: 1.0764e-04 eta: 0:56:25 time: 1.7774 data_time: 0.0024 memory: 65239 loss: 0.1993 +2023/06/01 07:02:41 - mmengine - INFO - Exp name: convnext_base_8xb512_fake5m_20230531_172616 +2023/06/01 07:03:16 - mmengine - INFO - Epoch(train) [19][1100/1440] lr: 1.0764e-04 eta: 0:53:24 time: 1.7925 data_time: 0.0022 memory: 65239 loss: 0.2005 +2023/06/01 07:06:12 - mmengine - INFO - Epoch(train) [19][1200/1440] lr: 1.0764e-04 eta: 0:50:24 time: 1.7672 data_time: 0.0020 memory: 65239 loss: 0.1989 +2023/06/01 07:09:08 - mmengine - INFO - Epoch(train) [19][1300/1440] lr: 1.0764e-04 eta: 0:47:24 time: 1.7406 data_time: 0.0020 memory: 65239 loss: 0.1994 +2023/06/01 07:12:04 - mmengine - INFO - Epoch(train) [19][1400/1440] lr: 1.0764e-04 eta: 0:44:24 time: 1.7579 data_time: 0.0021 memory: 65239 loss: 0.1993 +2023/06/01 07:13:13 - mmengine - INFO - Exp name: convnext_base_8xb512_fake5m_20230531_172616 +2023/06/01 07:13:13 - mmengine - INFO - Saving checkpoint at 19 epochs +2023/06/01 07:13:32 - mmengine - INFO - Epoch(val) [19][8/8] accuracy/top1: 99.9874 single-label/precision_classwise: [100.0, 0.0] single-label/recall_classwise: [99.98737335205078, 0.0] single-label/f1-score_classwise: [99.99369049072266, 0.0] data_time: 0.3952 time: 0.9961 +2023/06/01 07:16:45 - mmengine - INFO - Epoch(train) [20][ 100/1440] lr: 3.4562e-05 eta: 0:40:12 time: 1.7411 data_time: 0.0027 memory: 65239 loss: 0.1986 +2023/06/01 07:19:42 - mmengine - INFO - Epoch(train) [20][ 200/1440] lr: 3.4562e-05 eta: 0:37:12 time: 1.7546 data_time: 0.0025 memory: 65239 loss: 0.1993 +2023/06/01 07:22:38 - mmengine - INFO - Epoch(train) [20][ 300/1440] lr: 3.4562e-05 eta: 0:34:12 time: 1.7523 data_time: 0.0018 memory: 65239 loss: 0.2006 +2023/06/01 07:25:34 - mmengine - INFO - Epoch(train) [20][ 400/1440] lr: 3.4562e-05 eta: 0:31:11 time: 1.7582 data_time: 0.0021 memory: 65239 loss: 0.1987 +2023/06/01 07:28:30 - mmengine - INFO - Epoch(train) [20][ 500/1440] lr: 3.4562e-05 eta: 0:28:11 time: 1.7569 data_time: 0.0021 memory: 65239 loss: 0.1993 +2023/06/01 07:31:26 - mmengine - INFO - Epoch(train) [20][ 600/1440] lr: 3.4562e-05 eta: 0:25:11 time: 1.7810 data_time: 0.0020 memory: 65239 loss: 0.1991 +2023/06/01 07:32:36 - mmengine - INFO - Exp name: convnext_base_8xb512_fake5m_20230531_172616 +2023/06/01 07:34:22 - mmengine - INFO - Epoch(train) [20][ 700/1440] lr: 3.4562e-05 eta: 0:22:11 time: 1.7551 data_time: 0.0019 memory: 65239 loss: 0.1994 +2023/06/01 07:37:18 - mmengine - INFO - Epoch(train) [20][ 800/1440] lr: 3.4562e-05 eta: 0:19:11 time: 1.7554 data_time: 0.0020 memory: 65239 loss: 0.1999 +2023/06/01 07:40:14 - mmengine - INFO - Epoch(train) [20][ 900/1440] lr: 3.4562e-05 eta: 0:16:11 time: 1.7436 data_time: 0.0020 memory: 65239 loss: 0.1990 +2023/06/01 07:43:10 - mmengine - INFO - Epoch(train) [20][1000/1440] lr: 3.4562e-05 eta: 0:13:11 time: 1.7679 data_time: 0.0021 memory: 65239 loss: 0.1993 +2023/06/01 07:46:06 - mmengine - INFO - Epoch(train) [20][1100/1440] lr: 3.4562e-05 eta: 0:10:11 time: 1.7443 data_time: 0.0018 memory: 65239 loss: 0.1990 +2023/06/01 07:49:02 - mmengine - INFO - Epoch(train) [20][1200/1440] lr: 3.4562e-05 eta: 0:07:11 time: 1.7789 data_time: 0.0025 memory: 65239 loss: 0.1995 +2023/06/01 07:51:58 - mmengine - INFO - Epoch(train) [20][1300/1440] lr: 3.4562e-05 eta: 0:04:11 time: 1.7606 data_time: 0.0021 memory: 65239 loss: 0.1993 +2023/06/01 07:54:54 - mmengine - INFO - Epoch(train) [20][1400/1440] lr: 3.4562e-05 eta: 0:01:11 time: 1.7534 data_time: 0.0021 memory: 65239 loss: 0.1988 +2023/06/01 07:56:03 - mmengine - INFO - Exp name: convnext_base_8xb512_fake5m_20230531_172616 +2023/06/01 07:56:03 - mmengine - INFO - Saving checkpoint at 20 epochs +2023/06/01 07:56:21 - mmengine - INFO - Epoch(val) [20][8/8] accuracy/top1: 99.9874 single-label/precision_classwise: [100.0, 0.0] single-label/recall_classwise: [99.98737335205078, 0.0] single-label/f1-score_classwise: [99.99369049072266, 0.0] data_time: 0.3839 time: 0.9837 diff --git a/convnext_base_8xb512_4e-3lr_5m/20230531_172616/vis_data/20230531_172616.json b/convnext_base_8xb512_4e-3lr_5m/20230531_172616/vis_data/20230531_172616.json new file mode 100644 index 0000000000000000000000000000000000000000..4b7faa1cda74a291f2bc77753253d42f76dcbcbf --- /dev/null +++ b/convnext_base_8xb512_4e-3lr_5m/20230531_172616/vis_data/20230531_172616.json @@ -0,0 +1,300 @@ +{"lr": 0.004, "data_time": 0.294070839881897, "loss": 0.6471519112586975, "time": 2.2626111030578615, "epoch": 1, "memory": 65239, "step": 100} +{"lr": 0.004, "data_time": 0.013324451446533204, "loss": 0.5990125954151153, "time": 2.234609270095825, "epoch": 1, "memory": 65239, "step": 200} +{"lr": 0.004, "data_time": 0.0017860889434814452, "loss": 0.5899681389331818, "time": 2.242182660102844, "epoch": 1, "memory": 65239, "step": 300} +{"lr": 0.004, "data_time": 0.0020656824111938477, "loss": 0.5609820008277893, "time": 2.1564761877059935, "epoch": 1, "memory": 65239, "step": 400} +{"lr": 0.004, "data_time": 0.0018647193908691406, "loss": 0.5474038362503052, "time": 2.0359162092208862, "epoch": 1, "memory": 65239, "step": 500} +{"lr": 0.004, "data_time": 0.0018175840377807617, "loss": 0.5487338185310364, "time": 1.975503969192505, "epoch": 1, "memory": 65239, "step": 600} +{"lr": 0.004, "data_time": 0.0017319679260253905, "loss": 0.5407381355762482, "time": 2.171300530433655, "epoch": 1, "memory": 65239, "step": 700} +{"lr": 0.004, "data_time": 0.0019083499908447265, "loss": 0.5277312874794007, "time": 1.9163430690765382, "epoch": 1, "memory": 65239, "step": 800} +{"lr": 0.004, "data_time": 0.0021228551864624023, "loss": 0.5092132538557053, "time": 2.0220649003982545, "epoch": 1, "memory": 65239, "step": 900} +{"lr": 0.004, "data_time": 0.0016291379928588868, "loss": 0.5061135679483414, "time": 2.0095176219940187, "epoch": 1, "memory": 65239, "step": 1000} +{"lr": 0.004, "data_time": 0.0017848253250122071, "loss": 0.49420292377471925, "time": 2.0268460750579833, "epoch": 1, "memory": 65239, "step": 1100} +{"lr": 0.004, "data_time": 0.002391219139099121, "loss": 0.4754354238510132, "time": 2.205688810348511, "epoch": 1, "memory": 65239, "step": 1200} +{"lr": 0.004, "data_time": 0.0019346952438354492, "loss": 0.46469809114933014, "time": 1.8690577268600463, "epoch": 1, "memory": 65239, "step": 1300} +{"lr": 0.004, "data_time": 0.002057504653930664, "loss": 0.4820683717727661, "time": 1.7549256801605224, "epoch": 1, "memory": 65239, "step": 1400} +{"accuracy/top1": 63.95429229736328, "data_time": 0.5045796036720276, "time": 1.421905368566513, "step": 1} +{"lr": 0.0039754382394873, "data_time": 0.0020109891891479494, "loss": 0.4259195327758789, "time": 1.78918559551239, "epoch": 2, "memory": 65242, "step": 1540} +{"lr": 0.0039754382394873, "data_time": 0.0016762733459472657, "loss": 0.41477752327919004, "time": 1.9046591520309448, "epoch": 2, "memory": 65239, "step": 1640} +{"lr": 0.0039754382394873, "data_time": 0.0018329143524169922, "loss": 0.41110491156578066, "time": 1.797971487045288, "epoch": 2, "memory": 65239, "step": 1740} +{"lr": 0.0039754382394873, "data_time": 0.0019385099411010742, "loss": 0.40502460300922394, "time": 1.7963937044143676, "epoch": 2, "memory": 65239, "step": 1840} +{"lr": 0.0039754382394873, "data_time": 0.001766204833984375, "loss": 0.3943407952785492, "time": 1.7555464506149292, "epoch": 2, "memory": 65239, "step": 1940} +{"lr": 0.0039754382394873, "data_time": 0.0020206928253173827, "loss": 0.37663411200046537, "time": 1.86280198097229, "epoch": 2, "memory": 65239, "step": 2040} +{"lr": 0.0039754382394873, "data_time": 0.0024490833282470705, "loss": 0.3673877328634262, "time": 1.9785310268402099, "epoch": 2, "memory": 65239, "step": 2140} +{"lr": 0.0039754382394873, "data_time": 0.001880502700805664, "loss": 0.3726718842983246, "time": 1.8698872089385987, "epoch": 2, "memory": 65239, "step": 2240} +{"lr": 0.0039754382394873, "data_time": 0.002016782760620117, "loss": 0.33537166714668276, "time": 1.7702200651168822, "epoch": 2, "memory": 65239, "step": 2340} +{"lr": 0.0039754382394873, "data_time": 0.0023378849029541014, "loss": 0.3582533121109009, "time": 2.0686433792114256, "epoch": 2, "memory": 65239, "step": 2440} +{"lr": 0.0039754382394873, "data_time": 0.001741170883178711, "loss": 0.34021255373954773, "time": 1.971092939376831, "epoch": 2, "memory": 65239, "step": 2540} +{"lr": 0.0039754382394873, "data_time": 0.0021079301834106444, "loss": 0.31628977358341215, "time": 1.933544397354126, "epoch": 2, "memory": 65239, "step": 2640} +{"lr": 0.0039754382394873, "data_time": 0.0017555475234985352, "loss": 0.32789141237735747, "time": 2.150077152252197, "epoch": 2, "memory": 65239, "step": 2740} +{"lr": 0.0039754382394873, "data_time": 0.0016577482223510743, "loss": 0.335993355512619, "time": 1.7611670255661012, "epoch": 2, "memory": 65239, "step": 2840} +{"accuracy/top1": 69.5536880493164, "data_time": 0.38713765144348145, "time": 1.000772926542494, "step": 2} +{"lr": 0.0039023577500088323, "data_time": 0.002031755447387695, "loss": 0.28288253843784333, "time": 1.8418720960617065, "epoch": 3, "memory": 65239, "step": 2980} +{"lr": 0.0039023577500088323, "data_time": 0.0024559736251831056, "loss": 0.27681420743465424, "time": 2.0270755767822264, "epoch": 3, "memory": 65239, "step": 3080} +{"lr": 0.0039023577500088323, "data_time": 0.0018558025360107422, "loss": 0.5766315579414367, "time": 1.8425831556320191, "epoch": 3, "memory": 65239, "step": 3180} +{"lr": 0.0039023577500088323, "data_time": 0.0020722150802612305, "loss": 0.2858594566583633, "time": 1.8540190696716308, "epoch": 3, "memory": 65239, "step": 3280} +{"lr": 0.0039023577500088323, "data_time": 0.0018726587295532227, "loss": 0.26811360716819765, "time": 1.8345449924468995, "epoch": 3, "memory": 65239, "step": 3380} +{"lr": 0.0039023577500088323, "data_time": 0.0017278432846069337, "loss": 0.28085304200649264, "time": 1.854233455657959, "epoch": 3, "memory": 65239, "step": 3480} +{"lr": 0.0039023577500088323, "data_time": 0.001746535301208496, "loss": 0.27968792617321014, "time": 1.9766229152679444, "epoch": 3, "memory": 65239, "step": 3580} +{"lr": 0.0039023577500088323, "data_time": 0.0017764806747436524, "loss": 0.24300129860639572, "time": 1.8463210821151734, "epoch": 3, "memory": 65239, "step": 3680} +{"lr": 0.0039023577500088323, "data_time": 0.0023488283157348635, "loss": 0.5905640900135041, "time": 1.7898751020431518, "epoch": 3, "memory": 65239, "step": 3780} +{"lr": 0.0039023577500088323, "data_time": 0.0021797895431518556, "loss": 0.531296843290329, "time": 1.8254986524581909, "epoch": 3, "memory": 65239, "step": 3880} +{"lr": 0.0039023577500088323, "data_time": 0.0017969846725463868, "loss": 0.4591155737638474, "time": 1.7710155248641968, "epoch": 3, "memory": 65239, "step": 3980} +{"lr": 0.0039023577500088323, "data_time": 0.0019678354263305666, "loss": 0.32523842751979826, "time": 1.7721887350082397, "epoch": 3, "memory": 65239, "step": 4080} +{"lr": 0.0039023577500088323, "data_time": 0.0019093036651611328, "loss": 0.25851487517356875, "time": 1.7606155157089234, "epoch": 3, "memory": 65239, "step": 4180} +{"lr": 0.0039023577500088323, "data_time": 0.0019119739532470702, "loss": 0.24610384106636046, "time": 1.7476157188415526, "epoch": 3, "memory": 65239, "step": 4280} +{"accuracy/top1": 69.69256591796875, "data_time": 0.3738433255089654, "time": 0.9855900075700548, "step": 3} +{"lr": 0.0037825580157557948, "data_time": 0.0020201683044433595, "loss": 0.3203301072120667, "time": 1.744491958618164, "epoch": 4, "memory": 65239, "step": 4420} +{"lr": 0.0037825580157557948, "data_time": 0.002407479286193848, "loss": 0.3310075968503952, "time": 1.7491080999374389, "epoch": 4, "memory": 65239, "step": 4520} +{"lr": 0.0037825580157557948, "data_time": 0.002125263214111328, "loss": 0.24202582985162735, "time": 1.7603554725646973, "epoch": 4, "memory": 65239, "step": 4620} +{"lr": 0.0037825580157557948, "data_time": 0.002288627624511719, "loss": 0.22887523621320724, "time": 1.7569090127944946, "epoch": 4, "memory": 65239, "step": 4720} +{"lr": 0.0037825580157557948, "data_time": 0.0020905494689941405, "loss": 0.2276250422000885, "time": 1.7629335641860961, "epoch": 4, "memory": 65239, "step": 4820} +{"lr": 0.0037825580157557948, "data_time": 0.0022696256637573242, "loss": 0.26235032826662064, "time": 1.7657853841781617, "epoch": 4, "memory": 65239, "step": 4920} +{"lr": 0.0037825580157557948, "data_time": 0.0019611835479736326, "loss": 0.22663307785987855, "time": 1.7475191593170165, "epoch": 4, "memory": 65239, "step": 5020} +{"lr": 0.0037825580157557948, "data_time": 0.0021947622299194336, "loss": 0.46859192550182344, "time": 1.7470179080963135, "epoch": 4, "memory": 65239, "step": 5120} +{"lr": 0.0037825580157557948, "data_time": 0.0019683599472045898, "loss": 0.4010765999555588, "time": 1.7661454916000365, "epoch": 4, "memory": 65239, "step": 5220} +{"lr": 0.0037825580157557948, "data_time": 0.0015542268753051757, "loss": 0.2449846625328064, "time": 1.766442322731018, "epoch": 4, "memory": 65239, "step": 5320} +{"lr": 0.0037825580157557948, "data_time": 0.00200197696685791, "loss": 0.2322591319680214, "time": 1.7601410150527954, "epoch": 4, "memory": 65239, "step": 5420} +{"lr": 0.0037825580157557948, "data_time": 0.0018733739852905273, "loss": 0.23762350678443908, "time": 1.783822774887085, "epoch": 4, "memory": 65239, "step": 5520} +{"lr": 0.0037825580157557948, "data_time": 0.0017696619033813477, "loss": 0.5020914494991302, "time": 1.7745041847229004, "epoch": 4, "memory": 65239, "step": 5620} +{"lr": 0.0037825580157557948, "data_time": 0.001827239990234375, "loss": 0.4458970695734024, "time": 1.7648678064346313, "epoch": 4, "memory": 65239, "step": 5720} +{"accuracy/top1": 71.37806701660156, "data_time": 0.3789208200242784, "time": 0.9783332347869873, "step": 4} +{"lr": 0.0036189889037780212, "data_time": 0.001812291145324707, "loss": 0.3982940405607224, "time": 1.746271300315857, "epoch": 5, "memory": 65239, "step": 5860} +{"lr": 0.0036189889037780212, "data_time": 0.0019017457962036133, "loss": 0.2763418614864349, "time": 1.742712378501892, "epoch": 5, "memory": 65239, "step": 5960} +{"lr": 0.0036189889037780212, "data_time": 0.001874399185180664, "loss": 0.2458963692188263, "time": 1.7476083040237427, "epoch": 5, "memory": 65239, "step": 6060} +{"lr": 0.0036189889037780212, "data_time": 0.002372145652770996, "loss": 0.23664892315864564, "time": 1.7563019752502442, "epoch": 5, "memory": 65239, "step": 6160} +{"lr": 0.0036189889037780212, "data_time": 0.0022600412368774412, "loss": 0.21899010241031647, "time": 1.764320707321167, "epoch": 5, "memory": 65239, "step": 6260} +{"lr": 0.0036189889037780212, "data_time": 0.0027506351470947266, "loss": 0.2192614659667015, "time": 1.75459566116333, "epoch": 5, "memory": 65239, "step": 6360} +{"lr": 0.0036189889037780212, "data_time": 0.00204160213470459, "loss": 0.2205602914094925, "time": 1.7666844367980956, "epoch": 5, "memory": 65239, "step": 6460} +{"lr": 0.0036189889037780212, "data_time": 0.002318286895751953, "loss": 0.21142199784517288, "time": 1.772250247001648, "epoch": 5, "memory": 65239, "step": 6560} +{"lr": 0.0036189889037780212, "data_time": 0.002233147621154785, "loss": 0.21560410261154175, "time": 1.7406726598739624, "epoch": 5, "memory": 65239, "step": 6660} +{"lr": 0.0036189889037780212, "data_time": 0.002552437782287598, "loss": 0.6568452656269074, "time": 1.777324390411377, "epoch": 5, "memory": 65239, "step": 6760} +{"lr": 0.0036189889037780212, "data_time": 0.0023006916046142576, "loss": 0.5245528280735016, "time": 1.7449957847595214, "epoch": 5, "memory": 65239, "step": 6860} +{"lr": 0.0036189889037780212, "data_time": 0.003089642524719238, "loss": 0.42651675045490267, "time": 1.7538987398147583, "epoch": 5, "memory": 65239, "step": 6960} +{"lr": 0.0036189889037780212, "data_time": 0.0029165029525756838, "loss": 0.2812874525785446, "time": 1.7702352046966552, "epoch": 5, "memory": 65239, "step": 7060} +{"lr": 0.0036189889037780212, "data_time": 0.0022844314575195313, "loss": 0.5836776614189148, "time": 1.7620587348937988, "epoch": 5, "memory": 65239, "step": 7160} +{"accuracy/top1": 85.303955078125, "data_time": 0.39623305532667374, "time": 0.9961348639594184, "step": 5} +{"lr": 0.003415678028467163, "data_time": 0.0028115034103393553, "loss": 0.38810711205005644, "time": 1.7607410192489623, "epoch": 6, "memory": 65239, "step": 7300} +{"lr": 0.003415678028467163, "data_time": 0.002412891387939453, "loss": 0.3754238337278366, "time": 1.7475921392440796, "epoch": 6, "memory": 65239, "step": 7400} +{"lr": 0.003415678028467163, "data_time": 0.0028981685638427733, "loss": 0.2555116951465607, "time": 1.757999587059021, "epoch": 6, "memory": 65239, "step": 7500} +{"lr": 0.003415678028467163, "data_time": 0.003374743461608887, "loss": 0.2288053587079048, "time": 1.7510651111602784, "epoch": 6, "memory": 65239, "step": 7600} +{"lr": 0.003415678028467163, "data_time": 0.002768135070800781, "loss": 0.22382745742797852, "time": 1.747952365875244, "epoch": 6, "memory": 65239, "step": 7700} +{"lr": 0.003415678028467163, "data_time": 0.002213597297668457, "loss": 0.2270168751478195, "time": 1.752075982093811, "epoch": 6, "memory": 65239, "step": 7800} +{"lr": 0.003415678028467163, "data_time": 0.0018398761749267578, "loss": 0.21332418769598008, "time": 1.7525475025177002, "epoch": 6, "memory": 65239, "step": 7900} +{"lr": 0.003415678028467163, "data_time": 0.002478933334350586, "loss": 0.21184402257204055, "time": 1.7569594144821168, "epoch": 6, "memory": 65239, "step": 8000} +{"lr": 0.003415678028467163, "data_time": 0.001969575881958008, "loss": 0.21002853214740752, "time": 1.7542494773864745, "epoch": 6, "memory": 65239, "step": 8100} +{"lr": 0.003415678028467163, "data_time": 0.0024922847747802734, "loss": 0.6072520792484284, "time": 1.7548253297805787, "epoch": 6, "memory": 65239, "step": 8200} +{"lr": 0.003415678028467163, "data_time": 0.003552103042602539, "loss": 0.4654354006052017, "time": 1.7570080280303955, "epoch": 6, "memory": 65239, "step": 8300} +{"lr": 0.003415678028467163, "data_time": 0.0018169641494750976, "loss": 0.3930463522672653, "time": 1.7672250270843506, "epoch": 6, "memory": 65239, "step": 8400} +{"lr": 0.003415678028467163, "data_time": 0.0017796754837036133, "loss": 0.5148335725069046, "time": 1.7646772623062135, "epoch": 6, "memory": 65239, "step": 8500} +{"lr": 0.003415678028467163, "data_time": 0.0018166065216064452, "loss": 0.33338751196861266, "time": 1.7740228414535522, "epoch": 6, "memory": 65239, "step": 8600} +{"accuracy/top1": 97.90416717529297, "data_time": 0.3860985967848036, "time": 0.9885270065731473, "step": 6} +{"lr": 0.003177631578323485, "data_time": 0.002128195762634277, "loss": 0.24110675156116484, "time": 1.767635989189148, "epoch": 7, "memory": 65239, "step": 8740} +{"lr": 0.003177631578323485, "data_time": 0.0021002769470214845, "loss": 0.2191752091050148, "time": 1.7444259405136109, "epoch": 7, "memory": 65239, "step": 8840} +{"lr": 0.003177631578323485, "data_time": 0.002385258674621582, "loss": 0.22043028473854065, "time": 1.7490360736846924, "epoch": 7, "memory": 65239, "step": 8940} +{"lr": 0.003177631578323485, "data_time": 0.0018976688385009765, "loss": 0.21541695594787597, "time": 1.7567009687423707, "epoch": 7, "memory": 65239, "step": 9040} +{"lr": 0.003177631578323485, "data_time": 0.0023168087005615234, "loss": 0.2103339597582817, "time": 1.7602417945861817, "epoch": 7, "memory": 65239, "step": 9140} +{"lr": 0.003177631578323485, "data_time": 0.0020807981491088867, "loss": 0.21217475831508636, "time": 1.7623361825942994, "epoch": 7, "memory": 65239, "step": 9240} +{"lr": 0.003177631578323485, "data_time": 0.001966047286987305, "loss": 0.21213506907224655, "time": 1.7444286823272706, "epoch": 7, "memory": 65239, "step": 9340} +{"lr": 0.003177631578323485, "data_time": 0.0018904924392700196, "loss": 0.4851627379655838, "time": 1.7660134553909301, "epoch": 7, "memory": 65239, "step": 9440} +{"lr": 0.003177631578323485, "data_time": 0.002680373191833496, "loss": 0.38564896285533906, "time": 1.7836931943893433, "epoch": 7, "memory": 65239, "step": 9540} +{"lr": 0.003177631578323485, "data_time": 0.002378702163696289, "loss": 0.35051958858966825, "time": 1.7562671899795532, "epoch": 7, "memory": 65239, "step": 9640} +{"lr": 0.003177631578323485, "data_time": 0.0020656824111938477, "loss": 0.3022203117609024, "time": 1.753632402420044, "epoch": 7, "memory": 65239, "step": 9740} +{"lr": 0.003177631578323485, "data_time": 0.0017815589904785155, "loss": 0.3538883596658707, "time": 1.7506144046783447, "epoch": 7, "memory": 65239, "step": 9840} +{"lr": 0.003177631578323485, "data_time": 0.0018436908721923828, "loss": 0.2612239092588425, "time": 1.7662548780441285, "epoch": 7, "memory": 65239, "step": 9940} +{"lr": 0.003177631578323485, "data_time": 0.001869511604309082, "loss": 0.23757199048995972, "time": 1.768561887741089, "epoch": 7, "memory": 65239, "step": 10040} +{"accuracy/top1": 99.55810546875, "data_time": 0.39805518256293404, "time": 1.000428729587131, "step": 7} +{"lr": 0.0029107110469803964, "data_time": 0.0018477678298950196, "loss": 0.22523382157087327, "time": 1.7455479621887207, "epoch": 8, "memory": 65239, "step": 10180} +{"lr": 0.0029107110469803964, "data_time": 0.002920699119567871, "loss": 0.5731547236442566, "time": 1.748858380317688, "epoch": 8, "memory": 65239, "step": 10280} +{"lr": 0.0029107110469803964, "data_time": 0.0020402193069458006, "loss": 0.36425108909606935, "time": 1.7535511016845704, "epoch": 8, "memory": 65239, "step": 10380} +{"lr": 0.0029107110469803964, "data_time": 0.003080892562866211, "loss": 0.38161486983299253, "time": 1.7556456327438354, "epoch": 8, "memory": 65239, "step": 10480} +{"lr": 0.0029107110469803964, "data_time": 0.002080416679382324, "loss": 0.31201172769069674, "time": 1.7726204156875611, "epoch": 8, "memory": 65239, "step": 10580} +{"lr": 0.0029107110469803964, "data_time": 0.0033840656280517576, "loss": 0.555474603176117, "time": 1.7544145107269287, "epoch": 8, "memory": 65239, "step": 10680} +{"lr": 0.0029107110469803964, "data_time": 0.0023167610168457033, "loss": 0.3435333579778671, "time": 1.7588919639587401, "epoch": 8, "memory": 65239, "step": 10780} +{"lr": 0.0029107110469803964, "data_time": 0.002349996566772461, "loss": 0.337172144651413, "time": 1.7547831535339355, "epoch": 8, "memory": 65239, "step": 10880} +{"lr": 0.0029107110469803964, "data_time": 0.0019315242767333984, "loss": 0.2627667888998985, "time": 1.770263934135437, "epoch": 8, "memory": 65239, "step": 10980} +{"lr": 0.0029107110469803964, "data_time": 0.002792072296142578, "loss": 0.5019217729568481, "time": 1.7411481380462646, "epoch": 8, "memory": 65239, "step": 11080} +{"lr": 0.0029107110469803964, "data_time": 0.002167630195617676, "loss": 0.30798786878585815, "time": 1.7550288677215575, "epoch": 8, "memory": 65239, "step": 11180} +{"lr": 0.0029107110469803964, "data_time": 0.0027318477630615236, "loss": 0.2970837473869324, "time": 1.7573724508285522, "epoch": 8, "memory": 65239, "step": 11280} +{"lr": 0.0029107110469803964, "data_time": 0.002062225341796875, "loss": 0.23551699072122573, "time": 1.756939435005188, "epoch": 8, "memory": 65239, "step": 11380} +{"lr": 0.0029107110469803964, "data_time": 0.0036136388778686525, "loss": 0.2233336478471756, "time": 1.7784118890762328, "epoch": 8, "memory": 65239, "step": 11480} +{"accuracy/top1": 99.8042984008789, "data_time": 0.388608005311754, "time": 1.0026252534654405, "step": 8} +{"lr": 0.0026214889037780207, "data_time": 0.0021722793579101564, "loss": 0.21508563458919525, "time": 1.7626146078109741, "epoch": 9, "memory": 65239, "step": 11620} +{"lr": 0.0026214889037780207, "data_time": 0.0019434213638305664, "loss": 0.21320223510265351, "time": 1.782397484779358, "epoch": 9, "memory": 65239, "step": 11720} +{"lr": 0.0026214889037780207, "data_time": 0.002144598960876465, "loss": 0.2547208309173584, "time": 1.7802840709686278, "epoch": 9, "memory": 65239, "step": 11820} +{"lr": 0.0026214889037780207, "data_time": 0.0022049903869628905, "loss": 0.34757152795791624, "time": 1.7546339273452758, "epoch": 9, "memory": 65239, "step": 11920} +{"lr": 0.0026214889037780207, "data_time": 0.0021652460098266603, "loss": 0.2267567276954651, "time": 1.7413899898529053, "epoch": 9, "memory": 65239, "step": 12020} +{"lr": 0.0026214889037780207, "data_time": 0.0018102645874023438, "loss": 0.21234106719493867, "time": 1.7545764207839967, "epoch": 9, "memory": 65239, "step": 12120} +{"lr": 0.0026214889037780207, "data_time": 0.0017374515533447265, "loss": 0.21293012648820878, "time": 1.7564863681793212, "epoch": 9, "memory": 65239, "step": 12220} +{"lr": 0.0026214889037780207, "data_time": 0.0018054485321044923, "loss": 0.2079003170132637, "time": 1.7676341533660889, "epoch": 9, "memory": 65239, "step": 12320} +{"lr": 0.0026214889037780207, "data_time": 0.0019423484802246094, "loss": 0.20840659886598586, "time": 1.7573559522628783, "epoch": 9, "memory": 65239, "step": 12420} +{"lr": 0.0026214889037780207, "data_time": 0.00207064151763916, "loss": 0.20683594048023224, "time": 1.764697504043579, "epoch": 9, "memory": 65239, "step": 12520} +{"lr": 0.0026214889037780207, "data_time": 0.0019136905670166016, "loss": 0.2077335923910141, "time": 1.768392562866211, "epoch": 9, "memory": 65239, "step": 12620} +{"lr": 0.0026214889037780207, "data_time": 0.0019383192062377929, "loss": 0.20528330355882646, "time": 1.7409001111984252, "epoch": 9, "memory": 65239, "step": 12720} +{"lr": 0.0026214889037780207, "data_time": 0.0017258405685424804, "loss": 0.20630260556936264, "time": 1.762709903717041, "epoch": 9, "memory": 65239, "step": 12820} +{"lr": 0.0026214889037780207, "data_time": 0.0019710302352905274, "loss": 0.21508312076330185, "time": 1.7659692525863648, "epoch": 9, "memory": 65239, "step": 12920} +{"accuracy/top1": 99.90530395507812, "data_time": 0.39879780345492893, "time": 0.9993477397494845, "step": 9} +{"lr": 0.002317086757755261, "data_time": 0.0018691539764404297, "loss": 0.20842716991901397, "time": 1.766720747947693, "epoch": 10, "memory": 65239, "step": 13060} +{"lr": 0.002317086757755261, "data_time": 0.002149224281311035, "loss": 0.20473891198635102, "time": 1.7611877202987671, "epoch": 10, "memory": 65239, "step": 13160} +{"lr": 0.002317086757755261, "data_time": 0.0022081851959228514, "loss": 0.2035791978240013, "time": 1.764173984527588, "epoch": 10, "memory": 65239, "step": 13260} +{"lr": 0.002317086757755261, "data_time": 0.002052903175354004, "loss": 0.20580856800079345, "time": 1.7749219417572022, "epoch": 10, "memory": 65239, "step": 13360} +{"lr": 0.002317086757755261, "data_time": 0.00182650089263916, "loss": 0.20526792109012604, "time": 1.7783968448638916, "epoch": 10, "memory": 65239, "step": 13460} +{"lr": 0.002317086757755261, "data_time": 0.0020101070404052734, "loss": 0.20435237288475036, "time": 1.7422078132629395, "epoch": 10, "memory": 65239, "step": 13560} +{"lr": 0.002317086757755261, "data_time": 0.0017679929733276367, "loss": 0.20416932553052902, "time": 1.758651614189148, "epoch": 10, "memory": 65239, "step": 13660} +{"lr": 0.002317086757755261, "data_time": 0.00191497802734375, "loss": 0.20448124706745147, "time": 1.7431493759155274, "epoch": 10, "memory": 65239, "step": 13760} +{"lr": 0.002317086757755261, "data_time": 0.002168416976928711, "loss": 0.5548220098018646, "time": 1.7515681743621827, "epoch": 10, "memory": 65239, "step": 13860} +{"lr": 0.002317086757755261, "data_time": 0.002024054527282715, "loss": 0.42696404457092285, "time": 1.750724482536316, "epoch": 10, "memory": 65239, "step": 13960} +{"lr": 0.002317086757755261, "data_time": 0.00256807804107666, "loss": 0.5489125967025756, "time": 1.7591268301010132, "epoch": 10, "memory": 65239, "step": 14060} +{"lr": 0.002317086757755261, "data_time": 0.002050447463989258, "loss": 0.35148289799690247, "time": 1.7965302467346191, "epoch": 10, "memory": 65239, "step": 14160} +{"lr": 0.002317086757755261, "data_time": 0.0019504070281982423, "loss": 0.22230225503444673, "time": 1.7634393453598023, "epoch": 10, "memory": 65239, "step": 14260} +{"lr": 0.002317086757755261, "data_time": 0.0035539865493774414, "loss": 0.2145804226398468, "time": 1.7718512773513795, "epoch": 10, "memory": 65239, "step": 14360} +{"accuracy/top1": 99.90530395507812, "data_time": 0.3941118187374539, "time": 1.00543393029107, "step": 10} +{"lr": 0.0020050000000000003, "data_time": 0.0021576881408691406, "loss": 0.20816791653633118, "time": 1.7536620140075683, "epoch": 11, "memory": 65239, "step": 14500} +{"lr": 0.0020050000000000003, "data_time": 0.0019799232482910155, "loss": 0.20956154614686967, "time": 1.7521398782730102, "epoch": 11, "memory": 65239, "step": 14600} +{"lr": 0.0020050000000000003, "data_time": 0.0019007682800292968, "loss": 0.21384180039167405, "time": 1.748732900619507, "epoch": 11, "memory": 65239, "step": 14700} +{"lr": 0.0020050000000000003, "data_time": 0.0018473386764526368, "loss": 0.20683404803276062, "time": 1.7754571437835693, "epoch": 11, "memory": 65239, "step": 14800} +{"lr": 0.0020050000000000003, "data_time": 0.0017994403839111327, "loss": 0.20519542843103408, "time": 1.7551137447357177, "epoch": 11, "memory": 65239, "step": 14900} +{"lr": 0.0020050000000000003, "data_time": 0.001813507080078125, "loss": 0.2074568524956703, "time": 1.7574237823486327, "epoch": 11, "memory": 65239, "step": 15000} +{"lr": 0.0020050000000000003, "data_time": 0.0026179075241088865, "loss": 0.20661062747240067, "time": 1.7552197456359864, "epoch": 11, "memory": 65239, "step": 15100} +{"lr": 0.0020050000000000003, "data_time": 0.0037160158157348634, "loss": 0.20226958841085435, "time": 1.7733661651611328, "epoch": 11, "memory": 65239, "step": 15200} +{"lr": 0.0020050000000000003, "data_time": 0.0027885913848876955, "loss": 0.20422090888023375, "time": 1.7669978618621827, "epoch": 11, "memory": 65239, "step": 15300} +{"lr": 0.0020050000000000003, "data_time": 0.0031191825866699217, "loss": 0.203707717359066, "time": 1.7714092969894408, "epoch": 11, "memory": 65239, "step": 15400} +{"lr": 0.0020050000000000003, "data_time": 0.0027695655822753905, "loss": 0.2071462318301201, "time": 1.745333456993103, "epoch": 11, "memory": 65239, "step": 15500} +{"lr": 0.0020050000000000003, "data_time": 0.002219891548156738, "loss": 0.20478703677654267, "time": 1.7550196409225465, "epoch": 11, "memory": 65239, "step": 15600} +{"lr": 0.0020050000000000003, "data_time": 0.0018514871597290039, "loss": 0.20427894294261933, "time": 1.7571877479553222, "epoch": 11, "memory": 65239, "step": 15700} +{"lr": 0.0020050000000000003, "data_time": 0.0040299415588378905, "loss": 0.20476067066192627, "time": 1.7670121431350707, "epoch": 11, "memory": 65239, "step": 15800} +{"accuracy/top1": 99.8926773071289, "data_time": 0.389821105533176, "time": 1.0218744277954102, "step": 11} +{"lr": 0.0016929132422447402, "data_time": 0.0031202077865600587, "loss": 0.20389499366283417, "time": 1.7768501043319702, "epoch": 12, "memory": 65239, "step": 15940} +{"lr": 0.0016929132422447402, "data_time": 0.002581787109375, "loss": 0.20126746445894242, "time": 1.7417608737945556, "epoch": 12, "memory": 65239, "step": 16040} +{"lr": 0.0016929132422447402, "data_time": 0.0021525144577026365, "loss": 0.20422982573509216, "time": 1.7804574966430664, "epoch": 12, "memory": 65239, "step": 16140} +{"lr": 0.0016929132422447402, "data_time": 0.0018012523651123047, "loss": 0.20338739305734635, "time": 1.764135479927063, "epoch": 12, "memory": 65239, "step": 16240} +{"lr": 0.0016929132422447402, "data_time": 0.0018938064575195312, "loss": 0.20231839418411254, "time": 1.7781321048736571, "epoch": 12, "memory": 65239, "step": 16340} +{"lr": 0.0016929132422447402, "data_time": 0.002283930778503418, "loss": 0.20108201503753662, "time": 1.760820174217224, "epoch": 12, "memory": 65239, "step": 16440} +{"lr": 0.0016929132422447402, "data_time": 0.0019663572311401367, "loss": 0.20355645418167115, "time": 1.7540451288223267, "epoch": 12, "memory": 65239, "step": 16540} +{"lr": 0.0016929132422447402, "data_time": 0.001800203323364258, "loss": 0.36508008986711504, "time": 1.7439824342727661, "epoch": 12, "memory": 65239, "step": 16640} +{"lr": 0.0016929132422447402, "data_time": 0.001930832862854004, "loss": 0.5714943587779999, "time": 1.7421794891357423, "epoch": 12, "memory": 65239, "step": 16740} +{"lr": 0.0016929132422447402, "data_time": 0.0020593643188476563, "loss": 0.454214009642601, "time": 1.7717114448547364, "epoch": 12, "memory": 65239, "step": 16840} +{"lr": 0.0016929132422447402, "data_time": 0.0034413576126098634, "loss": 0.3863367438316345, "time": 1.7605345964431762, "epoch": 12, "memory": 65239, "step": 16940} +{"lr": 0.0016929132422447402, "data_time": 0.003021430969238281, "loss": 0.3557781934738159, "time": 1.7847723007202148, "epoch": 12, "memory": 65239, "step": 17040} +{"lr": 0.0016929132422447402, "data_time": 0.002015042304992676, "loss": 0.34542907774448395, "time": 1.7428524494171143, "epoch": 12, "memory": 65239, "step": 17140} +{"lr": 0.0016929132422447402, "data_time": 0.003604888916015625, "loss": 0.32865734696388244, "time": 1.7763700246810914, "epoch": 12, "memory": 65239, "step": 17240} +{"accuracy/top1": 99.89899444580078, "data_time": 0.3877944416469998, "time": 1.0037411318884955, "step": 12} +{"lr": 0.0013885110962219803, "data_time": 0.0019809961318969726, "loss": 0.3023691028356552, "time": 1.798832631111145, "epoch": 13, "memory": 65239, "step": 17380} +{"lr": 0.0013885110962219803, "data_time": 0.0017806291580200195, "loss": 0.29072235226631166, "time": 1.754764461517334, "epoch": 13, "memory": 65239, "step": 17480} +{"lr": 0.0013885110962219803, "data_time": 0.0018604040145874024, "loss": 0.28999619483947753, "time": 1.759545087814331, "epoch": 13, "memory": 65239, "step": 17580} +{"lr": 0.0013885110962219803, "data_time": 0.0019051074981689454, "loss": 0.28324767053127287, "time": 1.7445470333099364, "epoch": 13, "memory": 65239, "step": 17680} +{"lr": 0.0013885110962219803, "data_time": 0.0023175716400146485, "loss": 0.2616253077983856, "time": 1.7530099630355835, "epoch": 13, "memory": 65239, "step": 17780} +{"lr": 0.0013885110962219803, "data_time": 0.002057051658630371, "loss": 0.2643518656492233, "time": 1.7701531887054442, "epoch": 13, "memory": 65239, "step": 17880} +{"lr": 0.0013885110962219803, "data_time": 0.0018391370773315429, "loss": 0.25228428691625593, "time": 1.7654434442520142, "epoch": 13, "memory": 65239, "step": 17980} +{"lr": 0.0013885110962219803, "data_time": 0.0021905899047851562, "loss": 0.24978435337543486, "time": 1.809389328956604, "epoch": 13, "memory": 65239, "step": 18080} +{"lr": 0.0013885110962219803, "data_time": 0.002230405807495117, "loss": 0.23705584406852723, "time": 1.7452591180801391, "epoch": 13, "memory": 65239, "step": 18180} +{"lr": 0.0013885110962219803, "data_time": 0.0019921064376831055, "loss": 0.22479892820119857, "time": 1.7417317390441895, "epoch": 13, "memory": 65239, "step": 18280} +{"lr": 0.0013885110962219803, "data_time": 0.0024779319763183595, "loss": 0.21664273738861084, "time": 1.7676319360733033, "epoch": 13, "memory": 65239, "step": 18380} +{"lr": 0.0013885110962219803, "data_time": 0.002303028106689453, "loss": 0.22049114406108855, "time": 1.781920313835144, "epoch": 13, "memory": 65239, "step": 18480} +{"lr": 0.0013885110962219803, "data_time": 0.0021100759506225584, "loss": 0.21351015269756318, "time": 1.7518767356872558, "epoch": 13, "memory": 65239, "step": 18580} +{"lr": 0.0013885110962219803, "data_time": 0.002191305160522461, "loss": 0.21210558712482452, "time": 1.768466329574585, "epoch": 13, "memory": 65239, "step": 18680} +{"accuracy/top1": 99.91793060302734, "data_time": 0.3861401081085205, "time": 0.9990745650397407, "step": 13} +{"lr": 0.0010992889530196043, "data_time": 0.0020858287811279298, "loss": 0.21081105172634124, "time": 1.7864968538284303, "epoch": 14, "memory": 65239, "step": 18820} +{"lr": 0.0010992889530196043, "data_time": 0.0019132375717163086, "loss": 0.21286335587501526, "time": 1.742831325531006, "epoch": 14, "memory": 65239, "step": 18920} +{"lr": 0.0010992889530196043, "data_time": 0.0020255804061889648, "loss": 0.20626734793186188, "time": 1.7599242448806762, "epoch": 14, "memory": 65239, "step": 19020} +{"lr": 0.0010992889530196043, "data_time": 0.002443075180053711, "loss": 0.2080485165119171, "time": 1.7627905130386352, "epoch": 14, "memory": 65239, "step": 19120} +{"lr": 0.0010992889530196043, "data_time": 0.002242445945739746, "loss": 0.20652599781751632, "time": 1.766191792488098, "epoch": 14, "memory": 65239, "step": 19220} +{"lr": 0.0010992889530196043, "data_time": 0.002003169059753418, "loss": 0.20841969251632692, "time": 1.796801495552063, "epoch": 14, "memory": 65239, "step": 19320} +{"lr": 0.0010992889530196043, "data_time": 0.0019716262817382813, "loss": 0.20560517013072968, "time": 1.7568519115447998, "epoch": 14, "memory": 65239, "step": 19420} +{"lr": 0.0010992889530196043, "data_time": 0.0020728111267089844, "loss": 0.20440366715192795, "time": 1.7732841968536377, "epoch": 14, "memory": 65239, "step": 19520} +{"lr": 0.0010992889530196043, "data_time": 0.0017874002456665038, "loss": 0.20662380307912825, "time": 1.7677780866622925, "epoch": 14, "memory": 65239, "step": 19620} +{"lr": 0.0010992889530196043, "data_time": 0.0019253015518188477, "loss": 0.20646163523197175, "time": 1.74409761428833, "epoch": 14, "memory": 65239, "step": 19720} +{"lr": 0.0010992889530196043, "data_time": 0.0020572185516357423, "loss": 0.20328553169965743, "time": 1.7734734058380126, "epoch": 14, "memory": 65239, "step": 19820} +{"lr": 0.0010992889530196043, "data_time": 0.001986050605773926, "loss": 0.20519484877586364, "time": 1.7670406103134155, "epoch": 14, "memory": 65239, "step": 19920} +{"lr": 0.0010992889530196043, "data_time": 0.00192568302154541, "loss": 0.20460060685873033, "time": 1.7802386999130249, "epoch": 14, "memory": 65239, "step": 20020} +{"lr": 0.0010992889530196043, "data_time": 0.0018279075622558594, "loss": 0.20388774275779725, "time": 1.7676955461502075, "epoch": 14, "memory": 65239, "step": 20120} +{"accuracy/top1": 99.94318389892578, "data_time": 0.39358721839057076, "time": 0.9922700723012289, "step": 14} +{"lr": 0.0008323684216765164, "data_time": 0.0020709991455078124, "loss": 0.2032449021935463, "time": 1.802698302268982, "epoch": 15, "memory": 65239, "step": 20260} +{"lr": 0.0008323684216765164, "data_time": 0.002353334426879883, "loss": 0.20223529934883117, "time": 1.756337785720825, "epoch": 15, "memory": 65239, "step": 20360} +{"lr": 0.0008323684216765164, "data_time": 0.002085280418395996, "loss": 0.2043743059039116, "time": 1.7427149534225463, "epoch": 15, "memory": 65239, "step": 20460} +{"lr": 0.0008323684216765164, "data_time": 0.001897430419921875, "loss": 0.19989629983901977, "time": 1.7547791719436645, "epoch": 15, "memory": 65239, "step": 20560} +{"lr": 0.0008323684216765164, "data_time": 0.002122235298156738, "loss": 0.20236582458019256, "time": 1.7527594089508056, "epoch": 15, "memory": 65239, "step": 20660} +{"lr": 0.0008323684216765164, "data_time": 0.002094531059265137, "loss": 0.20196168571710588, "time": 1.7569403171539306, "epoch": 15, "memory": 65239, "step": 20760} +{"lr": 0.0008323684216765164, "data_time": 0.0017537355422973632, "loss": 0.2038114219903946, "time": 1.7669857263565063, "epoch": 15, "memory": 65239, "step": 20860} +{"lr": 0.0008323684216765164, "data_time": 0.0019979000091552733, "loss": 0.20305096805095674, "time": 1.7569674253463745, "epoch": 15, "memory": 65239, "step": 20960} +{"lr": 0.0008323684216765164, "data_time": 0.001834726333618164, "loss": 0.20225291848182678, "time": 1.7780596017837524, "epoch": 15, "memory": 65239, "step": 21060} +{"lr": 0.0008323684216765164, "data_time": 0.002405691146850586, "loss": 0.20076795816421508, "time": 1.7662408828735352, "epoch": 15, "memory": 65239, "step": 21160} +{"lr": 0.0008323684216765164, "data_time": 0.0019542455673217775, "loss": 0.20304083526134492, "time": 1.7481752395629884, "epoch": 15, "memory": 65239, "step": 21260} +{"lr": 0.0008323684216765164, "data_time": 0.0023180961608886717, "loss": 0.20917652249336244, "time": 1.7630841493606568, "epoch": 15, "memory": 65239, "step": 21360} +{"lr": 0.0008323684216765164, "data_time": 0.0029805183410644533, "loss": 0.20227044969797134, "time": 1.778252363204956, "epoch": 15, "memory": 65239, "step": 21460} +{"lr": 0.0008323684216765164, "data_time": 0.0023592233657836912, "loss": 0.2046443834900856, "time": 1.753699493408203, "epoch": 15, "memory": 65239, "step": 21560} +{"accuracy/top1": 99.94949340820312, "data_time": 0.40842943721347386, "time": 1.0091813405354817, "step": 15} +{"lr": 0.0005943219715328379, "data_time": 0.0019431114196777344, "loss": 0.20289774239063263, "time": 1.759325337409973, "epoch": 16, "memory": 65239, "step": 21700} +{"lr": 0.0005943219715328379, "data_time": 0.0024357080459594727, "loss": 0.2002728119492531, "time": 1.7553544759750366, "epoch": 16, "memory": 65239, "step": 21800} +{"lr": 0.0005943219715328379, "data_time": 0.0019134521484375, "loss": 0.20216303318738937, "time": 1.7687824964523315, "epoch": 16, "memory": 65239, "step": 21900} +{"lr": 0.0005943219715328379, "data_time": 0.002260923385620117, "loss": 0.20292162895202637, "time": 1.7510209798812866, "epoch": 16, "memory": 65239, "step": 22000} +{"lr": 0.0005943219715328379, "data_time": 0.0027961492538452148, "loss": 0.20225557684898376, "time": 1.7643477201461792, "epoch": 16, "memory": 65239, "step": 22100} +{"lr": 0.0005943219715328379, "data_time": 0.0023709774017333985, "loss": 0.20151180624961854, "time": 1.7591939449310303, "epoch": 16, "memory": 65239, "step": 22200} +{"lr": 0.0005943219715328379, "data_time": 0.001816558837890625, "loss": 0.20086881071329116, "time": 1.7888813018798828, "epoch": 16, "memory": 65239, "step": 22300} +{"lr": 0.0005943219715328379, "data_time": 0.002017974853515625, "loss": 0.20243337303400039, "time": 1.7686737537384034, "epoch": 16, "memory": 65239, "step": 22400} +{"lr": 0.0005943219715328379, "data_time": 0.002935194969177246, "loss": 0.20057529658079148, "time": 1.7791025161743164, "epoch": 16, "memory": 65239, "step": 22500} +{"lr": 0.0005943219715328379, "data_time": 0.0020410299301147463, "loss": 0.20000529438257217, "time": 1.757346272468567, "epoch": 16, "memory": 65239, "step": 22600} +{"lr": 0.0005943219715328379, "data_time": 0.002509284019470215, "loss": 0.20058979988098144, "time": 1.7695966005325316, "epoch": 16, "memory": 65239, "step": 22700} +{"lr": 0.0005943219715328379, "data_time": 0.0020275115966796875, "loss": 0.2020442008972168, "time": 1.7551882028579713, "epoch": 16, "memory": 65239, "step": 22800} +{"lr": 0.0005943219715328379, "data_time": 0.0026473522186279295, "loss": 0.20274824202060698, "time": 1.7679340362548828, "epoch": 16, "memory": 65239, "step": 22900} +{"lr": 0.0005943219715328379, "data_time": 0.0020511627197265627, "loss": 0.1994107723236084, "time": 1.7544353008270264, "epoch": 16, "memory": 65239, "step": 23000} +{"accuracy/top1": 99.96842956542969, "data_time": 0.3824879328409831, "time": 1.0150129795074463, "step": 16} +{"lr": 0.0003910110962219801, "data_time": 0.002009677886962891, "loss": 0.20077932626008987, "time": 1.757280707359314, "epoch": 17, "memory": 65239, "step": 23140} +{"lr": 0.0003910110962219801, "data_time": 0.0026701927185058595, "loss": 0.20092237442731858, "time": 1.7545795440673828, "epoch": 17, "memory": 65239, "step": 23240} +{"lr": 0.0003910110962219801, "data_time": 0.001892423629760742, "loss": 0.20009273141622544, "time": 1.7777115821838378, "epoch": 17, "memory": 65239, "step": 23340} +{"lr": 0.0003910110962219801, "data_time": 0.0018848657608032226, "loss": 0.20136475265026094, "time": 1.7528565645217895, "epoch": 17, "memory": 65239, "step": 23440} +{"lr": 0.0003910110962219801, "data_time": 0.00219578742980957, "loss": 0.20083101391792296, "time": 1.7707293510437012, "epoch": 17, "memory": 65239, "step": 23540} +{"lr": 0.0003910110962219801, "data_time": 0.002124381065368652, "loss": 0.2004728764295578, "time": 1.7419605016708375, "epoch": 17, "memory": 65239, "step": 23640} +{"lr": 0.0003910110962219801, "data_time": 0.002007460594177246, "loss": 0.20091038942337036, "time": 1.7536983013153076, "epoch": 17, "memory": 65239, "step": 23740} +{"lr": 0.0003910110962219801, "data_time": 0.0021686792373657227, "loss": 0.1997741401195526, "time": 1.778288984298706, "epoch": 17, "memory": 65239, "step": 23840} +{"lr": 0.0003910110962219801, "data_time": 0.0018552064895629883, "loss": 0.20084185302257537, "time": 1.7693876266479491, "epoch": 17, "memory": 65239, "step": 23940} +{"lr": 0.0003910110962219801, "data_time": 0.001966524124145508, "loss": 0.20196772068738938, "time": 1.7735004901885987, "epoch": 17, "memory": 65239, "step": 24040} +{"lr": 0.0003910110962219801, "data_time": 0.002670621871948242, "loss": 0.20015702694654464, "time": 1.7670262098312377, "epoch": 17, "memory": 65239, "step": 24140} +{"lr": 0.0003910110962219801, "data_time": 0.002411818504333496, "loss": 0.19915879815816878, "time": 1.741189670562744, "epoch": 17, "memory": 65239, "step": 24240} +{"lr": 0.0003910110962219801, "data_time": 0.0019959211349487305, "loss": 0.19983991235494614, "time": 1.7559121608734132, "epoch": 17, "memory": 65239, "step": 24340} +{"lr": 0.0003910110962219801, "data_time": 0.00214080810546875, "loss": 0.19959970712661743, "time": 1.755110502243042, "epoch": 17, "memory": 65239, "step": 24440} +{"accuracy/top1": 99.9810562133789, "data_time": 0.3806263605753581, "time": 0.981286711162991, "step": 17} +{"lr": 0.00022744198424420629, "data_time": 0.001790928840637207, "loss": 0.2004450038075447, "time": 1.7523632764816284, "epoch": 18, "memory": 65239, "step": 24580} +{"lr": 0.00022744198424420629, "data_time": 0.002045464515686035, "loss": 0.19925308525562285, "time": 1.7501772403717042, "epoch": 18, "memory": 65239, "step": 24680} +{"lr": 0.00022744198424420629, "data_time": 0.0019402027130126954, "loss": 0.19926874488592147, "time": 1.7578789472579956, "epoch": 18, "memory": 65239, "step": 24780} +{"lr": 0.00022744198424420629, "data_time": 0.002113199234008789, "loss": 0.20051436126232147, "time": 1.7429935216903687, "epoch": 18, "memory": 65239, "step": 24880} +{"lr": 0.00022744198424420629, "data_time": 0.0023162841796875, "loss": 0.19934873580932616, "time": 1.7414918184280395, "epoch": 18, "memory": 65239, "step": 24980} +{"lr": 0.00022744198424420629, "data_time": 0.002202749252319336, "loss": 0.20008279681205748, "time": 1.759545636177063, "epoch": 18, "memory": 65239, "step": 25080} +{"lr": 0.00022744198424420629, "data_time": 0.0021761178970336912, "loss": 0.20010497868061067, "time": 1.7668198108673097, "epoch": 18, "memory": 65239, "step": 25180} +{"lr": 0.00022744198424420629, "data_time": 0.002141261100769043, "loss": 0.19980052262544631, "time": 1.75847430229187, "epoch": 18, "memory": 65239, "step": 25280} +{"lr": 0.00022744198424420629, "data_time": 0.002082538604736328, "loss": 0.19960130006074905, "time": 1.7639854669570922, "epoch": 18, "memory": 65239, "step": 25380} +{"lr": 0.00022744198424420629, "data_time": 0.0018153905868530274, "loss": 0.20003917217254638, "time": 1.7665620565414428, "epoch": 18, "memory": 65239, "step": 25480} +{"lr": 0.00022744198424420629, "data_time": 0.001948070526123047, "loss": 0.20112757831811906, "time": 1.7692194700241088, "epoch": 18, "memory": 65239, "step": 25580} +{"lr": 0.00022744198424420629, "data_time": 0.002145218849182129, "loss": 0.19917107373476028, "time": 1.7573678493499756, "epoch": 18, "memory": 65239, "step": 25680} +{"lr": 0.00022744198424420629, "data_time": 0.0020323991775512695, "loss": 0.20046317726373672, "time": 1.7658787488937377, "epoch": 18, "memory": 65239, "step": 25780} +{"lr": 0.00022744198424420629, "data_time": 0.0025656461715698243, "loss": 0.19956554025411605, "time": 1.759355902671814, "epoch": 18, "memory": 65239, "step": 25880} +{"accuracy/top1": 99.98737335205078, "data_time": 0.3821522659725613, "time": 0.9835881657070584, "step": 18} +{"lr": 0.00010764224999116871, "data_time": 0.002393293380737305, "loss": 0.19981290847063066, "time": 1.7427427768707275, "epoch": 19, "memory": 65239, "step": 26020} +{"lr": 0.00010764224999116871, "data_time": 0.0017134666442871094, "loss": 0.20034122914075853, "time": 1.758993673324585, "epoch": 19, "memory": 65239, "step": 26120} +{"lr": 0.00010764224999116871, "data_time": 0.0023830890655517577, "loss": 0.19991809427738189, "time": 1.7672643899917602, "epoch": 19, "memory": 65239, "step": 26220} +{"lr": 0.00010764224999116871, "data_time": 0.0021048545837402343, "loss": 0.19866206049919127, "time": 1.74294593334198, "epoch": 19, "memory": 65239, "step": 26320} +{"lr": 0.00010764224999116871, "data_time": 0.0021180152893066407, "loss": 0.19864848852157593, "time": 1.7447846412658692, "epoch": 19, "memory": 65239, "step": 26420} +{"lr": 0.00010764224999116871, "data_time": 0.002152705192565918, "loss": 0.19899600148200988, "time": 1.768186354637146, "epoch": 19, "memory": 65239, "step": 26520} +{"lr": 0.00010764224999116871, "data_time": 0.0022055387496948244, "loss": 0.19933582693338395, "time": 1.753677487373352, "epoch": 19, "memory": 65239, "step": 26620} +{"lr": 0.00010764224999116871, "data_time": 0.002377963066101074, "loss": 0.19929366260766984, "time": 1.7775017261505126, "epoch": 19, "memory": 65239, "step": 26720} +{"lr": 0.00010764224999116871, "data_time": 0.0021529436111450196, "loss": 0.1995449736714363, "time": 1.7686182022094727, "epoch": 19, "memory": 65239, "step": 26820} +{"lr": 0.00010764224999116871, "data_time": 0.0023652315139770508, "loss": 0.19931880980730057, "time": 1.7774329662322998, "epoch": 19, "memory": 65239, "step": 26920} +{"lr": 0.00010764224999116871, "data_time": 0.0022436141967773437, "loss": 0.20051181316375732, "time": 1.7925169229507447, "epoch": 19, "memory": 65239, "step": 27020} +{"lr": 0.00010764224999116871, "data_time": 0.0020327091217041014, "loss": 0.19891057461500167, "time": 1.7671783685684204, "epoch": 19, "memory": 65239, "step": 27120} +{"lr": 0.00010764224999116871, "data_time": 0.002000594139099121, "loss": 0.19943817853927612, "time": 1.7406118154525756, "epoch": 19, "memory": 65239, "step": 27220} +{"lr": 0.00010764224999116871, "data_time": 0.0021384477615356444, "loss": 0.19929901659488677, "time": 1.7579497337341308, "epoch": 19, "memory": 65239, "step": 27320} +{"accuracy/top1": 99.98737335205078, "data_time": 0.3951967822180854, "time": 0.9960782527923584, "step": 19} +{"lr": 3.456176051270038e-05, "data_time": 0.002686166763305664, "loss": 0.19861729741096495, "time": 1.7411158561706543, "epoch": 20, "memory": 65239, "step": 27460} +{"lr": 3.456176051270038e-05, "data_time": 0.002486705780029297, "loss": 0.19931466579437257, "time": 1.7545936346054076, "epoch": 20, "memory": 65239, "step": 27560} +{"lr": 3.456176051270038e-05, "data_time": 0.0018121004104614258, "loss": 0.2005652129650116, "time": 1.7522696256637573, "epoch": 20, "memory": 65239, "step": 27660} +{"lr": 3.456176051270038e-05, "data_time": 0.0021408796310424805, "loss": 0.19871699512004853, "time": 1.758206844329834, "epoch": 20, "memory": 65239, "step": 27760} +{"lr": 3.456176051270038e-05, "data_time": 0.0021448135375976562, "loss": 0.19933580756187438, "time": 1.7568791151046752, "epoch": 20, "memory": 65239, "step": 27860} +{"lr": 3.456176051270038e-05, "data_time": 0.0019590139389038088, "loss": 0.1991402491927147, "time": 1.7809900283813476, "epoch": 20, "memory": 65239, "step": 27960} +{"lr": 3.456176051270038e-05, "data_time": 0.0018506526947021484, "loss": 0.19941504597663878, "time": 1.7550737619400025, "epoch": 20, "memory": 65239, "step": 28060} +{"lr": 3.456176051270038e-05, "data_time": 0.001998114585876465, "loss": 0.19986934661865235, "time": 1.7554445028305055, "epoch": 20, "memory": 65239, "step": 28160} +{"lr": 3.456176051270038e-05, "data_time": 0.0020247936248779298, "loss": 0.1989905744791031, "time": 1.7436093091964722, "epoch": 20, "memory": 65239, "step": 28260} +{"lr": 3.456176051270038e-05, "data_time": 0.0021062135696411134, "loss": 0.1992984741926193, "time": 1.7679325580596923, "epoch": 20, "memory": 65239, "step": 28360} +{"lr": 3.456176051270038e-05, "data_time": 0.0017922401428222656, "loss": 0.19895447343587874, "time": 1.7443328380584717, "epoch": 20, "memory": 65239, "step": 28460} +{"lr": 3.456176051270038e-05, "data_time": 0.0025100231170654295, "loss": 0.19951033294200898, "time": 1.7789090394973754, "epoch": 20, "memory": 65239, "step": 28560} +{"lr": 3.456176051270038e-05, "data_time": 0.0020704030990600585, "loss": 0.19933993071317674, "time": 1.7605851650238038, "epoch": 20, "memory": 65239, "step": 28660} +{"lr": 3.456176051270038e-05, "data_time": 0.0021323919296264648, "loss": 0.19881039708852768, "time": 1.7534290552139282, "epoch": 20, "memory": 65239, "step": 28760} +{"accuracy/top1": 99.98737335205078, "data_time": 0.38388779428270126, "time": 0.9837023682064481, "step": 20} diff --git a/convnext_base_8xb512_4e-3lr_5m/20230531_172616/vis_data/config.py b/convnext_base_8xb512_4e-3lr_5m/20230531_172616/vis_data/config.py new file mode 100644 index 0000000000000000000000000000000000000000..5c21ac599661bd296a59c3b80c82afd4217314f8 --- /dev/null +++ b/convnext_base_8xb512_4e-3lr_5m/20230531_172616/vis_data/config.py @@ -0,0 +1,170 @@ +optim_wrapper = dict( + optimizer=dict( + type='AdamW', + lr=0.004, + weight_decay=0.05, + eps=1e-08, + betas=(0.9, 0.999), + _scope_='mmpretrain'), + paramwise_cfg=dict( + norm_decay_mult=0.0, + bias_decay_mult=0.0, + flat_decay_mult=0.0, + custom_keys=dict({ + '.absolute_pos_embed': dict(decay_mult=0.0), + '.relative_position_bias_table': dict(decay_mult=0.0) + })), + type='AmpOptimWrapper', + dtype='bfloat16', + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=True, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=20, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=4096) +model = dict( + type='ImageClassifier', + backbone=dict(type='ConvNeXt', arch='base', drop_path_rate=0.5), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=1024, + loss=dict( + type='LabelSmoothLoss', label_smooth_val=0.1, mode='original'), + init_cfg=None), + init_cfg=dict( + type='TruncNormal', layer=['Conv2d', 'Linear'], std=0.02, bias=0.0), + train_cfg=None) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=512, + num_workers=10, + dataset=dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stablediffusionV1-5R2-dpmsolver-25-5m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/convnext_base_8xb512_4e-3lr_5m' diff --git a/convnext_base_8xb512_4e-3lr_5m/20230531_172616/vis_data/events.out.tfevents.1685525186.SH-IDC1-10-140-24-115.149105.0 b/convnext_base_8xb512_4e-3lr_5m/20230531_172616/vis_data/events.out.tfevents.1685525186.SH-IDC1-10-140-24-115.149105.0 new file mode 100644 index 0000000000000000000000000000000000000000..bd643a433f769cee9956548161928e6164a6598b --- /dev/null +++ b/convnext_base_8xb512_4e-3lr_5m/20230531_172616/vis_data/events.out.tfevents.1685525186.SH-IDC1-10-140-24-115.149105.0 @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:53ca30b1d247c95d4686709da0dd8774f2ae8572943f1e9f82b2f34afb7a82c1 +size 1500974 diff --git a/convnext_base_8xb512_4e-3lr_5m/20230531_172616/vis_data/scalars.json b/convnext_base_8xb512_4e-3lr_5m/20230531_172616/vis_data/scalars.json new file mode 100644 index 0000000000000000000000000000000000000000..4b7faa1cda74a291f2bc77753253d42f76dcbcbf --- /dev/null +++ b/convnext_base_8xb512_4e-3lr_5m/20230531_172616/vis_data/scalars.json @@ -0,0 +1,300 @@ +{"lr": 0.004, "data_time": 0.294070839881897, "loss": 0.6471519112586975, "time": 2.2626111030578615, "epoch": 1, "memory": 65239, "step": 100} +{"lr": 0.004, "data_time": 0.013324451446533204, "loss": 0.5990125954151153, "time": 2.234609270095825, "epoch": 1, "memory": 65239, "step": 200} +{"lr": 0.004, "data_time": 0.0017860889434814452, "loss": 0.5899681389331818, "time": 2.242182660102844, "epoch": 1, "memory": 65239, "step": 300} +{"lr": 0.004, "data_time": 0.0020656824111938477, "loss": 0.5609820008277893, "time": 2.1564761877059935, "epoch": 1, "memory": 65239, "step": 400} +{"lr": 0.004, "data_time": 0.0018647193908691406, "loss": 0.5474038362503052, "time": 2.0359162092208862, "epoch": 1, "memory": 65239, "step": 500} +{"lr": 0.004, "data_time": 0.0018175840377807617, "loss": 0.5487338185310364, "time": 1.975503969192505, "epoch": 1, "memory": 65239, "step": 600} +{"lr": 0.004, "data_time": 0.0017319679260253905, "loss": 0.5407381355762482, "time": 2.171300530433655, "epoch": 1, "memory": 65239, "step": 700} +{"lr": 0.004, "data_time": 0.0019083499908447265, "loss": 0.5277312874794007, "time": 1.9163430690765382, "epoch": 1, "memory": 65239, "step": 800} +{"lr": 0.004, "data_time": 0.0021228551864624023, "loss": 0.5092132538557053, "time": 2.0220649003982545, "epoch": 1, "memory": 65239, "step": 900} +{"lr": 0.004, "data_time": 0.0016291379928588868, "loss": 0.5061135679483414, "time": 2.0095176219940187, "epoch": 1, "memory": 65239, "step": 1000} +{"lr": 0.004, "data_time": 0.0017848253250122071, "loss": 0.49420292377471925, "time": 2.0268460750579833, "epoch": 1, "memory": 65239, "step": 1100} +{"lr": 0.004, "data_time": 0.002391219139099121, "loss": 0.4754354238510132, "time": 2.205688810348511, "epoch": 1, "memory": 65239, "step": 1200} +{"lr": 0.004, "data_time": 0.0019346952438354492, "loss": 0.46469809114933014, "time": 1.8690577268600463, "epoch": 1, "memory": 65239, "step": 1300} +{"lr": 0.004, "data_time": 0.002057504653930664, "loss": 0.4820683717727661, "time": 1.7549256801605224, "epoch": 1, "memory": 65239, "step": 1400} +{"accuracy/top1": 63.95429229736328, "data_time": 0.5045796036720276, "time": 1.421905368566513, "step": 1} +{"lr": 0.0039754382394873, "data_time": 0.0020109891891479494, "loss": 0.4259195327758789, "time": 1.78918559551239, "epoch": 2, "memory": 65242, "step": 1540} +{"lr": 0.0039754382394873, "data_time": 0.0016762733459472657, "loss": 0.41477752327919004, "time": 1.9046591520309448, "epoch": 2, "memory": 65239, "step": 1640} +{"lr": 0.0039754382394873, "data_time": 0.0018329143524169922, "loss": 0.41110491156578066, "time": 1.797971487045288, "epoch": 2, "memory": 65239, "step": 1740} +{"lr": 0.0039754382394873, "data_time": 0.0019385099411010742, "loss": 0.40502460300922394, "time": 1.7963937044143676, "epoch": 2, "memory": 65239, "step": 1840} +{"lr": 0.0039754382394873, "data_time": 0.001766204833984375, "loss": 0.3943407952785492, "time": 1.7555464506149292, "epoch": 2, "memory": 65239, "step": 1940} +{"lr": 0.0039754382394873, "data_time": 0.0020206928253173827, "loss": 0.37663411200046537, "time": 1.86280198097229, "epoch": 2, "memory": 65239, "step": 2040} +{"lr": 0.0039754382394873, "data_time": 0.0024490833282470705, "loss": 0.3673877328634262, "time": 1.9785310268402099, "epoch": 2, "memory": 65239, "step": 2140} +{"lr": 0.0039754382394873, "data_time": 0.001880502700805664, "loss": 0.3726718842983246, "time": 1.8698872089385987, "epoch": 2, "memory": 65239, "step": 2240} +{"lr": 0.0039754382394873, "data_time": 0.002016782760620117, "loss": 0.33537166714668276, "time": 1.7702200651168822, "epoch": 2, "memory": 65239, "step": 2340} +{"lr": 0.0039754382394873, "data_time": 0.0023378849029541014, "loss": 0.3582533121109009, "time": 2.0686433792114256, "epoch": 2, "memory": 65239, "step": 2440} +{"lr": 0.0039754382394873, "data_time": 0.001741170883178711, "loss": 0.34021255373954773, "time": 1.971092939376831, "epoch": 2, "memory": 65239, "step": 2540} +{"lr": 0.0039754382394873, "data_time": 0.0021079301834106444, "loss": 0.31628977358341215, "time": 1.933544397354126, "epoch": 2, "memory": 65239, "step": 2640} +{"lr": 0.0039754382394873, "data_time": 0.0017555475234985352, "loss": 0.32789141237735747, "time": 2.150077152252197, "epoch": 2, "memory": 65239, "step": 2740} +{"lr": 0.0039754382394873, "data_time": 0.0016577482223510743, "loss": 0.335993355512619, "time": 1.7611670255661012, "epoch": 2, "memory": 65239, "step": 2840} +{"accuracy/top1": 69.5536880493164, "data_time": 0.38713765144348145, "time": 1.000772926542494, "step": 2} +{"lr": 0.0039023577500088323, "data_time": 0.002031755447387695, "loss": 0.28288253843784333, "time": 1.8418720960617065, "epoch": 3, "memory": 65239, "step": 2980} +{"lr": 0.0039023577500088323, "data_time": 0.0024559736251831056, "loss": 0.27681420743465424, "time": 2.0270755767822264, "epoch": 3, "memory": 65239, "step": 3080} +{"lr": 0.0039023577500088323, "data_time": 0.0018558025360107422, "loss": 0.5766315579414367, "time": 1.8425831556320191, "epoch": 3, "memory": 65239, "step": 3180} +{"lr": 0.0039023577500088323, "data_time": 0.0020722150802612305, "loss": 0.2858594566583633, "time": 1.8540190696716308, "epoch": 3, "memory": 65239, "step": 3280} +{"lr": 0.0039023577500088323, "data_time": 0.0018726587295532227, "loss": 0.26811360716819765, "time": 1.8345449924468995, "epoch": 3, "memory": 65239, "step": 3380} +{"lr": 0.0039023577500088323, "data_time": 0.0017278432846069337, "loss": 0.28085304200649264, "time": 1.854233455657959, "epoch": 3, "memory": 65239, "step": 3480} +{"lr": 0.0039023577500088323, "data_time": 0.001746535301208496, "loss": 0.27968792617321014, "time": 1.9766229152679444, "epoch": 3, "memory": 65239, "step": 3580} +{"lr": 0.0039023577500088323, "data_time": 0.0017764806747436524, "loss": 0.24300129860639572, "time": 1.8463210821151734, "epoch": 3, "memory": 65239, "step": 3680} +{"lr": 0.0039023577500088323, "data_time": 0.0023488283157348635, "loss": 0.5905640900135041, "time": 1.7898751020431518, "epoch": 3, "memory": 65239, "step": 3780} +{"lr": 0.0039023577500088323, "data_time": 0.0021797895431518556, "loss": 0.531296843290329, "time": 1.8254986524581909, "epoch": 3, "memory": 65239, "step": 3880} +{"lr": 0.0039023577500088323, "data_time": 0.0017969846725463868, "loss": 0.4591155737638474, "time": 1.7710155248641968, "epoch": 3, "memory": 65239, "step": 3980} +{"lr": 0.0039023577500088323, "data_time": 0.0019678354263305666, "loss": 0.32523842751979826, "time": 1.7721887350082397, "epoch": 3, "memory": 65239, "step": 4080} +{"lr": 0.0039023577500088323, "data_time": 0.0019093036651611328, "loss": 0.25851487517356875, "time": 1.7606155157089234, "epoch": 3, "memory": 65239, "step": 4180} +{"lr": 0.0039023577500088323, "data_time": 0.0019119739532470702, "loss": 0.24610384106636046, "time": 1.7476157188415526, "epoch": 3, "memory": 65239, "step": 4280} +{"accuracy/top1": 69.69256591796875, "data_time": 0.3738433255089654, "time": 0.9855900075700548, "step": 3} +{"lr": 0.0037825580157557948, "data_time": 0.0020201683044433595, "loss": 0.3203301072120667, "time": 1.744491958618164, "epoch": 4, "memory": 65239, "step": 4420} +{"lr": 0.0037825580157557948, "data_time": 0.002407479286193848, "loss": 0.3310075968503952, "time": 1.7491080999374389, "epoch": 4, "memory": 65239, "step": 4520} +{"lr": 0.0037825580157557948, "data_time": 0.002125263214111328, "loss": 0.24202582985162735, "time": 1.7603554725646973, "epoch": 4, "memory": 65239, "step": 4620} +{"lr": 0.0037825580157557948, "data_time": 0.002288627624511719, "loss": 0.22887523621320724, "time": 1.7569090127944946, "epoch": 4, "memory": 65239, "step": 4720} +{"lr": 0.0037825580157557948, "data_time": 0.0020905494689941405, "loss": 0.2276250422000885, "time": 1.7629335641860961, "epoch": 4, "memory": 65239, "step": 4820} +{"lr": 0.0037825580157557948, "data_time": 0.0022696256637573242, "loss": 0.26235032826662064, "time": 1.7657853841781617, "epoch": 4, "memory": 65239, "step": 4920} +{"lr": 0.0037825580157557948, "data_time": 0.0019611835479736326, "loss": 0.22663307785987855, "time": 1.7475191593170165, "epoch": 4, "memory": 65239, "step": 5020} +{"lr": 0.0037825580157557948, "data_time": 0.0021947622299194336, "loss": 0.46859192550182344, "time": 1.7470179080963135, "epoch": 4, "memory": 65239, "step": 5120} +{"lr": 0.0037825580157557948, "data_time": 0.0019683599472045898, "loss": 0.4010765999555588, "time": 1.7661454916000365, "epoch": 4, "memory": 65239, "step": 5220} +{"lr": 0.0037825580157557948, "data_time": 0.0015542268753051757, "loss": 0.2449846625328064, "time": 1.766442322731018, "epoch": 4, "memory": 65239, "step": 5320} +{"lr": 0.0037825580157557948, "data_time": 0.00200197696685791, "loss": 0.2322591319680214, "time": 1.7601410150527954, "epoch": 4, "memory": 65239, "step": 5420} +{"lr": 0.0037825580157557948, "data_time": 0.0018733739852905273, "loss": 0.23762350678443908, "time": 1.783822774887085, "epoch": 4, "memory": 65239, "step": 5520} +{"lr": 0.0037825580157557948, "data_time": 0.0017696619033813477, "loss": 0.5020914494991302, "time": 1.7745041847229004, "epoch": 4, "memory": 65239, "step": 5620} +{"lr": 0.0037825580157557948, "data_time": 0.001827239990234375, "loss": 0.4458970695734024, "time": 1.7648678064346313, "epoch": 4, "memory": 65239, "step": 5720} +{"accuracy/top1": 71.37806701660156, "data_time": 0.3789208200242784, "time": 0.9783332347869873, "step": 4} +{"lr": 0.0036189889037780212, "data_time": 0.001812291145324707, "loss": 0.3982940405607224, "time": 1.746271300315857, "epoch": 5, "memory": 65239, "step": 5860} +{"lr": 0.0036189889037780212, "data_time": 0.0019017457962036133, "loss": 0.2763418614864349, "time": 1.742712378501892, "epoch": 5, "memory": 65239, "step": 5960} +{"lr": 0.0036189889037780212, "data_time": 0.001874399185180664, "loss": 0.2458963692188263, "time": 1.7476083040237427, "epoch": 5, "memory": 65239, "step": 6060} +{"lr": 0.0036189889037780212, "data_time": 0.002372145652770996, "loss": 0.23664892315864564, "time": 1.7563019752502442, "epoch": 5, "memory": 65239, "step": 6160} +{"lr": 0.0036189889037780212, "data_time": 0.0022600412368774412, "loss": 0.21899010241031647, "time": 1.764320707321167, "epoch": 5, "memory": 65239, "step": 6260} +{"lr": 0.0036189889037780212, "data_time": 0.0027506351470947266, "loss": 0.2192614659667015, "time": 1.75459566116333, "epoch": 5, "memory": 65239, "step": 6360} +{"lr": 0.0036189889037780212, "data_time": 0.00204160213470459, "loss": 0.2205602914094925, "time": 1.7666844367980956, "epoch": 5, "memory": 65239, "step": 6460} +{"lr": 0.0036189889037780212, "data_time": 0.002318286895751953, "loss": 0.21142199784517288, "time": 1.772250247001648, "epoch": 5, "memory": 65239, "step": 6560} +{"lr": 0.0036189889037780212, "data_time": 0.002233147621154785, "loss": 0.21560410261154175, "time": 1.7406726598739624, "epoch": 5, "memory": 65239, "step": 6660} +{"lr": 0.0036189889037780212, "data_time": 0.002552437782287598, "loss": 0.6568452656269074, "time": 1.777324390411377, "epoch": 5, "memory": 65239, "step": 6760} +{"lr": 0.0036189889037780212, "data_time": 0.0023006916046142576, "loss": 0.5245528280735016, "time": 1.7449957847595214, "epoch": 5, "memory": 65239, "step": 6860} +{"lr": 0.0036189889037780212, "data_time": 0.003089642524719238, "loss": 0.42651675045490267, "time": 1.7538987398147583, "epoch": 5, "memory": 65239, "step": 6960} +{"lr": 0.0036189889037780212, "data_time": 0.0029165029525756838, "loss": 0.2812874525785446, "time": 1.7702352046966552, "epoch": 5, "memory": 65239, "step": 7060} +{"lr": 0.0036189889037780212, "data_time": 0.0022844314575195313, "loss": 0.5836776614189148, "time": 1.7620587348937988, "epoch": 5, "memory": 65239, "step": 7160} +{"accuracy/top1": 85.303955078125, "data_time": 0.39623305532667374, "time": 0.9961348639594184, "step": 5} +{"lr": 0.003415678028467163, "data_time": 0.0028115034103393553, "loss": 0.38810711205005644, "time": 1.7607410192489623, "epoch": 6, "memory": 65239, "step": 7300} +{"lr": 0.003415678028467163, "data_time": 0.002412891387939453, "loss": 0.3754238337278366, "time": 1.7475921392440796, "epoch": 6, "memory": 65239, "step": 7400} +{"lr": 0.003415678028467163, "data_time": 0.0028981685638427733, "loss": 0.2555116951465607, "time": 1.757999587059021, "epoch": 6, "memory": 65239, "step": 7500} +{"lr": 0.003415678028467163, "data_time": 0.003374743461608887, "loss": 0.2288053587079048, "time": 1.7510651111602784, "epoch": 6, "memory": 65239, "step": 7600} +{"lr": 0.003415678028467163, "data_time": 0.002768135070800781, "loss": 0.22382745742797852, "time": 1.747952365875244, "epoch": 6, "memory": 65239, "step": 7700} +{"lr": 0.003415678028467163, "data_time": 0.002213597297668457, "loss": 0.2270168751478195, "time": 1.752075982093811, "epoch": 6, "memory": 65239, "step": 7800} +{"lr": 0.003415678028467163, "data_time": 0.0018398761749267578, "loss": 0.21332418769598008, "time": 1.7525475025177002, "epoch": 6, "memory": 65239, "step": 7900} +{"lr": 0.003415678028467163, "data_time": 0.002478933334350586, "loss": 0.21184402257204055, "time": 1.7569594144821168, "epoch": 6, "memory": 65239, "step": 8000} +{"lr": 0.003415678028467163, "data_time": 0.001969575881958008, "loss": 0.21002853214740752, "time": 1.7542494773864745, "epoch": 6, "memory": 65239, "step": 8100} +{"lr": 0.003415678028467163, "data_time": 0.0024922847747802734, "loss": 0.6072520792484284, "time": 1.7548253297805787, "epoch": 6, "memory": 65239, "step": 8200} +{"lr": 0.003415678028467163, "data_time": 0.003552103042602539, "loss": 0.4654354006052017, "time": 1.7570080280303955, "epoch": 6, "memory": 65239, "step": 8300} +{"lr": 0.003415678028467163, "data_time": 0.0018169641494750976, "loss": 0.3930463522672653, "time": 1.7672250270843506, "epoch": 6, "memory": 65239, "step": 8400} +{"lr": 0.003415678028467163, "data_time": 0.0017796754837036133, "loss": 0.5148335725069046, "time": 1.7646772623062135, "epoch": 6, "memory": 65239, "step": 8500} +{"lr": 0.003415678028467163, "data_time": 0.0018166065216064452, "loss": 0.33338751196861266, "time": 1.7740228414535522, "epoch": 6, "memory": 65239, "step": 8600} +{"accuracy/top1": 97.90416717529297, "data_time": 0.3860985967848036, "time": 0.9885270065731473, "step": 6} +{"lr": 0.003177631578323485, "data_time": 0.002128195762634277, "loss": 0.24110675156116484, "time": 1.767635989189148, "epoch": 7, "memory": 65239, "step": 8740} +{"lr": 0.003177631578323485, "data_time": 0.0021002769470214845, "loss": 0.2191752091050148, "time": 1.7444259405136109, "epoch": 7, "memory": 65239, "step": 8840} +{"lr": 0.003177631578323485, "data_time": 0.002385258674621582, "loss": 0.22043028473854065, "time": 1.7490360736846924, "epoch": 7, "memory": 65239, "step": 8940} +{"lr": 0.003177631578323485, "data_time": 0.0018976688385009765, "loss": 0.21541695594787597, "time": 1.7567009687423707, "epoch": 7, "memory": 65239, "step": 9040} +{"lr": 0.003177631578323485, "data_time": 0.0023168087005615234, "loss": 0.2103339597582817, "time": 1.7602417945861817, "epoch": 7, "memory": 65239, "step": 9140} +{"lr": 0.003177631578323485, "data_time": 0.0020807981491088867, "loss": 0.21217475831508636, "time": 1.7623361825942994, "epoch": 7, "memory": 65239, "step": 9240} +{"lr": 0.003177631578323485, "data_time": 0.001966047286987305, "loss": 0.21213506907224655, "time": 1.7444286823272706, "epoch": 7, "memory": 65239, "step": 9340} +{"lr": 0.003177631578323485, "data_time": 0.0018904924392700196, "loss": 0.4851627379655838, "time": 1.7660134553909301, "epoch": 7, "memory": 65239, "step": 9440} +{"lr": 0.003177631578323485, "data_time": 0.002680373191833496, "loss": 0.38564896285533906, "time": 1.7836931943893433, "epoch": 7, "memory": 65239, "step": 9540} +{"lr": 0.003177631578323485, "data_time": 0.002378702163696289, "loss": 0.35051958858966825, "time": 1.7562671899795532, "epoch": 7, "memory": 65239, "step": 9640} +{"lr": 0.003177631578323485, "data_time": 0.0020656824111938477, "loss": 0.3022203117609024, "time": 1.753632402420044, "epoch": 7, "memory": 65239, "step": 9740} +{"lr": 0.003177631578323485, "data_time": 0.0017815589904785155, "loss": 0.3538883596658707, "time": 1.7506144046783447, "epoch": 7, "memory": 65239, "step": 9840} +{"lr": 0.003177631578323485, "data_time": 0.0018436908721923828, "loss": 0.2612239092588425, "time": 1.7662548780441285, "epoch": 7, "memory": 65239, "step": 9940} +{"lr": 0.003177631578323485, "data_time": 0.001869511604309082, "loss": 0.23757199048995972, "time": 1.768561887741089, "epoch": 7, "memory": 65239, "step": 10040} +{"accuracy/top1": 99.55810546875, "data_time": 0.39805518256293404, "time": 1.000428729587131, "step": 7} +{"lr": 0.0029107110469803964, "data_time": 0.0018477678298950196, "loss": 0.22523382157087327, "time": 1.7455479621887207, "epoch": 8, "memory": 65239, "step": 10180} +{"lr": 0.0029107110469803964, "data_time": 0.002920699119567871, "loss": 0.5731547236442566, "time": 1.748858380317688, "epoch": 8, "memory": 65239, "step": 10280} +{"lr": 0.0029107110469803964, "data_time": 0.0020402193069458006, "loss": 0.36425108909606935, "time": 1.7535511016845704, "epoch": 8, "memory": 65239, "step": 10380} +{"lr": 0.0029107110469803964, "data_time": 0.003080892562866211, "loss": 0.38161486983299253, "time": 1.7556456327438354, "epoch": 8, "memory": 65239, "step": 10480} +{"lr": 0.0029107110469803964, "data_time": 0.002080416679382324, "loss": 0.31201172769069674, "time": 1.7726204156875611, "epoch": 8, "memory": 65239, "step": 10580} +{"lr": 0.0029107110469803964, "data_time": 0.0033840656280517576, "loss": 0.555474603176117, "time": 1.7544145107269287, "epoch": 8, "memory": 65239, "step": 10680} +{"lr": 0.0029107110469803964, "data_time": 0.0023167610168457033, "loss": 0.3435333579778671, "time": 1.7588919639587401, "epoch": 8, "memory": 65239, "step": 10780} +{"lr": 0.0029107110469803964, "data_time": 0.002349996566772461, "loss": 0.337172144651413, "time": 1.7547831535339355, "epoch": 8, "memory": 65239, "step": 10880} +{"lr": 0.0029107110469803964, "data_time": 0.0019315242767333984, "loss": 0.2627667888998985, "time": 1.770263934135437, "epoch": 8, "memory": 65239, "step": 10980} +{"lr": 0.0029107110469803964, "data_time": 0.002792072296142578, "loss": 0.5019217729568481, "time": 1.7411481380462646, "epoch": 8, "memory": 65239, "step": 11080} +{"lr": 0.0029107110469803964, "data_time": 0.002167630195617676, "loss": 0.30798786878585815, "time": 1.7550288677215575, "epoch": 8, "memory": 65239, "step": 11180} +{"lr": 0.0029107110469803964, "data_time": 0.0027318477630615236, "loss": 0.2970837473869324, "time": 1.7573724508285522, "epoch": 8, "memory": 65239, "step": 11280} +{"lr": 0.0029107110469803964, "data_time": 0.002062225341796875, "loss": 0.23551699072122573, "time": 1.756939435005188, "epoch": 8, "memory": 65239, "step": 11380} +{"lr": 0.0029107110469803964, "data_time": 0.0036136388778686525, "loss": 0.2233336478471756, "time": 1.7784118890762328, "epoch": 8, "memory": 65239, "step": 11480} +{"accuracy/top1": 99.8042984008789, "data_time": 0.388608005311754, "time": 1.0026252534654405, "step": 8} +{"lr": 0.0026214889037780207, "data_time": 0.0021722793579101564, "loss": 0.21508563458919525, "time": 1.7626146078109741, "epoch": 9, "memory": 65239, "step": 11620} +{"lr": 0.0026214889037780207, "data_time": 0.0019434213638305664, "loss": 0.21320223510265351, "time": 1.782397484779358, "epoch": 9, "memory": 65239, "step": 11720} +{"lr": 0.0026214889037780207, "data_time": 0.002144598960876465, "loss": 0.2547208309173584, "time": 1.7802840709686278, "epoch": 9, "memory": 65239, "step": 11820} +{"lr": 0.0026214889037780207, "data_time": 0.0022049903869628905, "loss": 0.34757152795791624, "time": 1.7546339273452758, "epoch": 9, "memory": 65239, "step": 11920} +{"lr": 0.0026214889037780207, "data_time": 0.0021652460098266603, "loss": 0.2267567276954651, "time": 1.7413899898529053, "epoch": 9, "memory": 65239, "step": 12020} +{"lr": 0.0026214889037780207, "data_time": 0.0018102645874023438, "loss": 0.21234106719493867, "time": 1.7545764207839967, "epoch": 9, "memory": 65239, "step": 12120} +{"lr": 0.0026214889037780207, "data_time": 0.0017374515533447265, "loss": 0.21293012648820878, "time": 1.7564863681793212, "epoch": 9, "memory": 65239, "step": 12220} +{"lr": 0.0026214889037780207, "data_time": 0.0018054485321044923, "loss": 0.2079003170132637, "time": 1.7676341533660889, "epoch": 9, "memory": 65239, "step": 12320} +{"lr": 0.0026214889037780207, "data_time": 0.0019423484802246094, "loss": 0.20840659886598586, "time": 1.7573559522628783, "epoch": 9, "memory": 65239, "step": 12420} +{"lr": 0.0026214889037780207, "data_time": 0.00207064151763916, "loss": 0.20683594048023224, "time": 1.764697504043579, "epoch": 9, "memory": 65239, "step": 12520} +{"lr": 0.0026214889037780207, "data_time": 0.0019136905670166016, "loss": 0.2077335923910141, "time": 1.768392562866211, "epoch": 9, "memory": 65239, "step": 12620} +{"lr": 0.0026214889037780207, "data_time": 0.0019383192062377929, "loss": 0.20528330355882646, "time": 1.7409001111984252, "epoch": 9, "memory": 65239, "step": 12720} +{"lr": 0.0026214889037780207, "data_time": 0.0017258405685424804, "loss": 0.20630260556936264, "time": 1.762709903717041, "epoch": 9, "memory": 65239, "step": 12820} +{"lr": 0.0026214889037780207, "data_time": 0.0019710302352905274, "loss": 0.21508312076330185, "time": 1.7659692525863648, "epoch": 9, "memory": 65239, "step": 12920} +{"accuracy/top1": 99.90530395507812, "data_time": 0.39879780345492893, "time": 0.9993477397494845, "step": 9} +{"lr": 0.002317086757755261, "data_time": 0.0018691539764404297, "loss": 0.20842716991901397, "time": 1.766720747947693, "epoch": 10, "memory": 65239, "step": 13060} +{"lr": 0.002317086757755261, "data_time": 0.002149224281311035, "loss": 0.20473891198635102, "time": 1.7611877202987671, "epoch": 10, "memory": 65239, "step": 13160} +{"lr": 0.002317086757755261, "data_time": 0.0022081851959228514, "loss": 0.2035791978240013, "time": 1.764173984527588, "epoch": 10, "memory": 65239, "step": 13260} +{"lr": 0.002317086757755261, "data_time": 0.002052903175354004, "loss": 0.20580856800079345, "time": 1.7749219417572022, "epoch": 10, "memory": 65239, "step": 13360} +{"lr": 0.002317086757755261, "data_time": 0.00182650089263916, "loss": 0.20526792109012604, "time": 1.7783968448638916, "epoch": 10, "memory": 65239, "step": 13460} +{"lr": 0.002317086757755261, "data_time": 0.0020101070404052734, "loss": 0.20435237288475036, "time": 1.7422078132629395, "epoch": 10, "memory": 65239, "step": 13560} +{"lr": 0.002317086757755261, "data_time": 0.0017679929733276367, "loss": 0.20416932553052902, "time": 1.758651614189148, "epoch": 10, "memory": 65239, "step": 13660} +{"lr": 0.002317086757755261, "data_time": 0.00191497802734375, "loss": 0.20448124706745147, "time": 1.7431493759155274, "epoch": 10, "memory": 65239, "step": 13760} +{"lr": 0.002317086757755261, "data_time": 0.002168416976928711, "loss": 0.5548220098018646, "time": 1.7515681743621827, "epoch": 10, "memory": 65239, "step": 13860} +{"lr": 0.002317086757755261, "data_time": 0.002024054527282715, "loss": 0.42696404457092285, "time": 1.750724482536316, "epoch": 10, "memory": 65239, "step": 13960} +{"lr": 0.002317086757755261, "data_time": 0.00256807804107666, "loss": 0.5489125967025756, "time": 1.7591268301010132, "epoch": 10, "memory": 65239, "step": 14060} +{"lr": 0.002317086757755261, "data_time": 0.002050447463989258, "loss": 0.35148289799690247, "time": 1.7965302467346191, "epoch": 10, "memory": 65239, "step": 14160} +{"lr": 0.002317086757755261, "data_time": 0.0019504070281982423, "loss": 0.22230225503444673, "time": 1.7634393453598023, "epoch": 10, "memory": 65239, "step": 14260} +{"lr": 0.002317086757755261, "data_time": 0.0035539865493774414, "loss": 0.2145804226398468, "time": 1.7718512773513795, "epoch": 10, "memory": 65239, "step": 14360} +{"accuracy/top1": 99.90530395507812, "data_time": 0.3941118187374539, "time": 1.00543393029107, "step": 10} +{"lr": 0.0020050000000000003, "data_time": 0.0021576881408691406, "loss": 0.20816791653633118, "time": 1.7536620140075683, "epoch": 11, "memory": 65239, "step": 14500} +{"lr": 0.0020050000000000003, "data_time": 0.0019799232482910155, "loss": 0.20956154614686967, "time": 1.7521398782730102, "epoch": 11, "memory": 65239, "step": 14600} +{"lr": 0.0020050000000000003, "data_time": 0.0019007682800292968, "loss": 0.21384180039167405, "time": 1.748732900619507, "epoch": 11, "memory": 65239, "step": 14700} +{"lr": 0.0020050000000000003, "data_time": 0.0018473386764526368, "loss": 0.20683404803276062, "time": 1.7754571437835693, "epoch": 11, "memory": 65239, "step": 14800} +{"lr": 0.0020050000000000003, "data_time": 0.0017994403839111327, "loss": 0.20519542843103408, "time": 1.7551137447357177, "epoch": 11, "memory": 65239, "step": 14900} +{"lr": 0.0020050000000000003, "data_time": 0.001813507080078125, "loss": 0.2074568524956703, "time": 1.7574237823486327, "epoch": 11, "memory": 65239, "step": 15000} +{"lr": 0.0020050000000000003, "data_time": 0.0026179075241088865, "loss": 0.20661062747240067, "time": 1.7552197456359864, "epoch": 11, "memory": 65239, "step": 15100} +{"lr": 0.0020050000000000003, "data_time": 0.0037160158157348634, "loss": 0.20226958841085435, "time": 1.7733661651611328, "epoch": 11, "memory": 65239, "step": 15200} +{"lr": 0.0020050000000000003, "data_time": 0.0027885913848876955, "loss": 0.20422090888023375, "time": 1.7669978618621827, "epoch": 11, "memory": 65239, "step": 15300} +{"lr": 0.0020050000000000003, "data_time": 0.0031191825866699217, "loss": 0.203707717359066, "time": 1.7714092969894408, "epoch": 11, "memory": 65239, "step": 15400} +{"lr": 0.0020050000000000003, "data_time": 0.0027695655822753905, "loss": 0.2071462318301201, "time": 1.745333456993103, "epoch": 11, "memory": 65239, "step": 15500} +{"lr": 0.0020050000000000003, "data_time": 0.002219891548156738, "loss": 0.20478703677654267, "time": 1.7550196409225465, "epoch": 11, "memory": 65239, "step": 15600} +{"lr": 0.0020050000000000003, "data_time": 0.0018514871597290039, "loss": 0.20427894294261933, "time": 1.7571877479553222, "epoch": 11, "memory": 65239, "step": 15700} +{"lr": 0.0020050000000000003, "data_time": 0.0040299415588378905, "loss": 0.20476067066192627, "time": 1.7670121431350707, "epoch": 11, "memory": 65239, "step": 15800} +{"accuracy/top1": 99.8926773071289, "data_time": 0.389821105533176, "time": 1.0218744277954102, "step": 11} +{"lr": 0.0016929132422447402, "data_time": 0.0031202077865600587, "loss": 0.20389499366283417, "time": 1.7768501043319702, "epoch": 12, "memory": 65239, "step": 15940} +{"lr": 0.0016929132422447402, "data_time": 0.002581787109375, "loss": 0.20126746445894242, "time": 1.7417608737945556, "epoch": 12, "memory": 65239, "step": 16040} +{"lr": 0.0016929132422447402, "data_time": 0.0021525144577026365, "loss": 0.20422982573509216, "time": 1.7804574966430664, "epoch": 12, "memory": 65239, "step": 16140} +{"lr": 0.0016929132422447402, "data_time": 0.0018012523651123047, "loss": 0.20338739305734635, "time": 1.764135479927063, "epoch": 12, "memory": 65239, "step": 16240} +{"lr": 0.0016929132422447402, "data_time": 0.0018938064575195312, "loss": 0.20231839418411254, "time": 1.7781321048736571, "epoch": 12, "memory": 65239, "step": 16340} +{"lr": 0.0016929132422447402, "data_time": 0.002283930778503418, "loss": 0.20108201503753662, "time": 1.760820174217224, "epoch": 12, "memory": 65239, "step": 16440} +{"lr": 0.0016929132422447402, "data_time": 0.0019663572311401367, "loss": 0.20355645418167115, "time": 1.7540451288223267, "epoch": 12, "memory": 65239, "step": 16540} +{"lr": 0.0016929132422447402, "data_time": 0.001800203323364258, "loss": 0.36508008986711504, "time": 1.7439824342727661, "epoch": 12, "memory": 65239, "step": 16640} +{"lr": 0.0016929132422447402, "data_time": 0.001930832862854004, "loss": 0.5714943587779999, "time": 1.7421794891357423, "epoch": 12, "memory": 65239, "step": 16740} +{"lr": 0.0016929132422447402, "data_time": 0.0020593643188476563, "loss": 0.454214009642601, "time": 1.7717114448547364, "epoch": 12, "memory": 65239, "step": 16840} +{"lr": 0.0016929132422447402, "data_time": 0.0034413576126098634, "loss": 0.3863367438316345, "time": 1.7605345964431762, "epoch": 12, "memory": 65239, "step": 16940} +{"lr": 0.0016929132422447402, "data_time": 0.003021430969238281, "loss": 0.3557781934738159, "time": 1.7847723007202148, "epoch": 12, "memory": 65239, "step": 17040} +{"lr": 0.0016929132422447402, "data_time": 0.002015042304992676, "loss": 0.34542907774448395, "time": 1.7428524494171143, "epoch": 12, "memory": 65239, "step": 17140} +{"lr": 0.0016929132422447402, "data_time": 0.003604888916015625, "loss": 0.32865734696388244, "time": 1.7763700246810914, "epoch": 12, "memory": 65239, "step": 17240} +{"accuracy/top1": 99.89899444580078, "data_time": 0.3877944416469998, "time": 1.0037411318884955, "step": 12} +{"lr": 0.0013885110962219803, "data_time": 0.0019809961318969726, "loss": 0.3023691028356552, "time": 1.798832631111145, "epoch": 13, "memory": 65239, "step": 17380} +{"lr": 0.0013885110962219803, "data_time": 0.0017806291580200195, "loss": 0.29072235226631166, "time": 1.754764461517334, "epoch": 13, "memory": 65239, "step": 17480} +{"lr": 0.0013885110962219803, "data_time": 0.0018604040145874024, "loss": 0.28999619483947753, "time": 1.759545087814331, "epoch": 13, "memory": 65239, "step": 17580} +{"lr": 0.0013885110962219803, "data_time": 0.0019051074981689454, "loss": 0.28324767053127287, "time": 1.7445470333099364, "epoch": 13, "memory": 65239, "step": 17680} +{"lr": 0.0013885110962219803, "data_time": 0.0023175716400146485, "loss": 0.2616253077983856, "time": 1.7530099630355835, "epoch": 13, "memory": 65239, "step": 17780} +{"lr": 0.0013885110962219803, "data_time": 0.002057051658630371, "loss": 0.2643518656492233, "time": 1.7701531887054442, "epoch": 13, "memory": 65239, "step": 17880} +{"lr": 0.0013885110962219803, "data_time": 0.0018391370773315429, "loss": 0.25228428691625593, "time": 1.7654434442520142, "epoch": 13, "memory": 65239, "step": 17980} +{"lr": 0.0013885110962219803, "data_time": 0.0021905899047851562, "loss": 0.24978435337543486, "time": 1.809389328956604, "epoch": 13, "memory": 65239, "step": 18080} +{"lr": 0.0013885110962219803, "data_time": 0.002230405807495117, "loss": 0.23705584406852723, "time": 1.7452591180801391, "epoch": 13, "memory": 65239, "step": 18180} +{"lr": 0.0013885110962219803, "data_time": 0.0019921064376831055, "loss": 0.22479892820119857, "time": 1.7417317390441895, "epoch": 13, "memory": 65239, "step": 18280} +{"lr": 0.0013885110962219803, "data_time": 0.0024779319763183595, "loss": 0.21664273738861084, "time": 1.7676319360733033, "epoch": 13, "memory": 65239, "step": 18380} +{"lr": 0.0013885110962219803, "data_time": 0.002303028106689453, "loss": 0.22049114406108855, "time": 1.781920313835144, "epoch": 13, "memory": 65239, "step": 18480} +{"lr": 0.0013885110962219803, "data_time": 0.0021100759506225584, "loss": 0.21351015269756318, "time": 1.7518767356872558, "epoch": 13, "memory": 65239, "step": 18580} +{"lr": 0.0013885110962219803, "data_time": 0.002191305160522461, "loss": 0.21210558712482452, "time": 1.768466329574585, "epoch": 13, "memory": 65239, "step": 18680} +{"accuracy/top1": 99.91793060302734, "data_time": 0.3861401081085205, "time": 0.9990745650397407, "step": 13} +{"lr": 0.0010992889530196043, "data_time": 0.0020858287811279298, "loss": 0.21081105172634124, "time": 1.7864968538284303, "epoch": 14, "memory": 65239, "step": 18820} +{"lr": 0.0010992889530196043, "data_time": 0.0019132375717163086, "loss": 0.21286335587501526, "time": 1.742831325531006, "epoch": 14, "memory": 65239, "step": 18920} +{"lr": 0.0010992889530196043, "data_time": 0.0020255804061889648, "loss": 0.20626734793186188, "time": 1.7599242448806762, "epoch": 14, "memory": 65239, "step": 19020} +{"lr": 0.0010992889530196043, "data_time": 0.002443075180053711, "loss": 0.2080485165119171, "time": 1.7627905130386352, "epoch": 14, "memory": 65239, "step": 19120} +{"lr": 0.0010992889530196043, "data_time": 0.002242445945739746, "loss": 0.20652599781751632, "time": 1.766191792488098, "epoch": 14, "memory": 65239, "step": 19220} +{"lr": 0.0010992889530196043, "data_time": 0.002003169059753418, "loss": 0.20841969251632692, "time": 1.796801495552063, "epoch": 14, "memory": 65239, "step": 19320} +{"lr": 0.0010992889530196043, "data_time": 0.0019716262817382813, "loss": 0.20560517013072968, "time": 1.7568519115447998, "epoch": 14, "memory": 65239, "step": 19420} +{"lr": 0.0010992889530196043, "data_time": 0.0020728111267089844, "loss": 0.20440366715192795, "time": 1.7732841968536377, "epoch": 14, "memory": 65239, "step": 19520} +{"lr": 0.0010992889530196043, "data_time": 0.0017874002456665038, "loss": 0.20662380307912825, "time": 1.7677780866622925, "epoch": 14, "memory": 65239, "step": 19620} +{"lr": 0.0010992889530196043, "data_time": 0.0019253015518188477, "loss": 0.20646163523197175, "time": 1.74409761428833, "epoch": 14, "memory": 65239, "step": 19720} +{"lr": 0.0010992889530196043, "data_time": 0.0020572185516357423, "loss": 0.20328553169965743, "time": 1.7734734058380126, "epoch": 14, "memory": 65239, "step": 19820} +{"lr": 0.0010992889530196043, "data_time": 0.001986050605773926, "loss": 0.20519484877586364, "time": 1.7670406103134155, "epoch": 14, "memory": 65239, "step": 19920} +{"lr": 0.0010992889530196043, "data_time": 0.00192568302154541, "loss": 0.20460060685873033, "time": 1.7802386999130249, "epoch": 14, "memory": 65239, "step": 20020} +{"lr": 0.0010992889530196043, "data_time": 0.0018279075622558594, "loss": 0.20388774275779725, "time": 1.7676955461502075, "epoch": 14, "memory": 65239, "step": 20120} +{"accuracy/top1": 99.94318389892578, "data_time": 0.39358721839057076, "time": 0.9922700723012289, "step": 14} +{"lr": 0.0008323684216765164, "data_time": 0.0020709991455078124, "loss": 0.2032449021935463, "time": 1.802698302268982, "epoch": 15, "memory": 65239, "step": 20260} +{"lr": 0.0008323684216765164, "data_time": 0.002353334426879883, "loss": 0.20223529934883117, "time": 1.756337785720825, "epoch": 15, "memory": 65239, "step": 20360} +{"lr": 0.0008323684216765164, "data_time": 0.002085280418395996, "loss": 0.2043743059039116, "time": 1.7427149534225463, "epoch": 15, "memory": 65239, "step": 20460} +{"lr": 0.0008323684216765164, "data_time": 0.001897430419921875, "loss": 0.19989629983901977, "time": 1.7547791719436645, "epoch": 15, "memory": 65239, "step": 20560} +{"lr": 0.0008323684216765164, "data_time": 0.002122235298156738, "loss": 0.20236582458019256, "time": 1.7527594089508056, "epoch": 15, "memory": 65239, "step": 20660} +{"lr": 0.0008323684216765164, "data_time": 0.002094531059265137, "loss": 0.20196168571710588, "time": 1.7569403171539306, "epoch": 15, "memory": 65239, "step": 20760} +{"lr": 0.0008323684216765164, "data_time": 0.0017537355422973632, "loss": 0.2038114219903946, "time": 1.7669857263565063, "epoch": 15, "memory": 65239, "step": 20860} +{"lr": 0.0008323684216765164, "data_time": 0.0019979000091552733, "loss": 0.20305096805095674, "time": 1.7569674253463745, "epoch": 15, "memory": 65239, "step": 20960} +{"lr": 0.0008323684216765164, "data_time": 0.001834726333618164, "loss": 0.20225291848182678, "time": 1.7780596017837524, "epoch": 15, "memory": 65239, "step": 21060} +{"lr": 0.0008323684216765164, "data_time": 0.002405691146850586, "loss": 0.20076795816421508, "time": 1.7662408828735352, "epoch": 15, "memory": 65239, "step": 21160} +{"lr": 0.0008323684216765164, "data_time": 0.0019542455673217775, "loss": 0.20304083526134492, "time": 1.7481752395629884, "epoch": 15, "memory": 65239, "step": 21260} +{"lr": 0.0008323684216765164, "data_time": 0.0023180961608886717, "loss": 0.20917652249336244, "time": 1.7630841493606568, "epoch": 15, "memory": 65239, "step": 21360} +{"lr": 0.0008323684216765164, "data_time": 0.0029805183410644533, "loss": 0.20227044969797134, "time": 1.778252363204956, "epoch": 15, "memory": 65239, "step": 21460} +{"lr": 0.0008323684216765164, "data_time": 0.0023592233657836912, "loss": 0.2046443834900856, "time": 1.753699493408203, "epoch": 15, "memory": 65239, "step": 21560} +{"accuracy/top1": 99.94949340820312, "data_time": 0.40842943721347386, "time": 1.0091813405354817, "step": 15} +{"lr": 0.0005943219715328379, "data_time": 0.0019431114196777344, "loss": 0.20289774239063263, "time": 1.759325337409973, "epoch": 16, "memory": 65239, "step": 21700} +{"lr": 0.0005943219715328379, "data_time": 0.0024357080459594727, "loss": 0.2002728119492531, "time": 1.7553544759750366, "epoch": 16, "memory": 65239, "step": 21800} +{"lr": 0.0005943219715328379, "data_time": 0.0019134521484375, "loss": 0.20216303318738937, "time": 1.7687824964523315, "epoch": 16, "memory": 65239, "step": 21900} +{"lr": 0.0005943219715328379, "data_time": 0.002260923385620117, "loss": 0.20292162895202637, "time": 1.7510209798812866, "epoch": 16, "memory": 65239, "step": 22000} +{"lr": 0.0005943219715328379, "data_time": 0.0027961492538452148, "loss": 0.20225557684898376, "time": 1.7643477201461792, "epoch": 16, "memory": 65239, "step": 22100} +{"lr": 0.0005943219715328379, "data_time": 0.0023709774017333985, "loss": 0.20151180624961854, "time": 1.7591939449310303, "epoch": 16, "memory": 65239, "step": 22200} +{"lr": 0.0005943219715328379, "data_time": 0.001816558837890625, "loss": 0.20086881071329116, "time": 1.7888813018798828, "epoch": 16, "memory": 65239, "step": 22300} +{"lr": 0.0005943219715328379, "data_time": 0.002017974853515625, "loss": 0.20243337303400039, "time": 1.7686737537384034, "epoch": 16, "memory": 65239, "step": 22400} +{"lr": 0.0005943219715328379, "data_time": 0.002935194969177246, "loss": 0.20057529658079148, "time": 1.7791025161743164, "epoch": 16, "memory": 65239, "step": 22500} +{"lr": 0.0005943219715328379, "data_time": 0.0020410299301147463, "loss": 0.20000529438257217, "time": 1.757346272468567, "epoch": 16, "memory": 65239, "step": 22600} +{"lr": 0.0005943219715328379, "data_time": 0.002509284019470215, "loss": 0.20058979988098144, "time": 1.7695966005325316, "epoch": 16, "memory": 65239, "step": 22700} +{"lr": 0.0005943219715328379, "data_time": 0.0020275115966796875, "loss": 0.2020442008972168, "time": 1.7551882028579713, "epoch": 16, "memory": 65239, "step": 22800} +{"lr": 0.0005943219715328379, "data_time": 0.0026473522186279295, "loss": 0.20274824202060698, "time": 1.7679340362548828, "epoch": 16, "memory": 65239, "step": 22900} +{"lr": 0.0005943219715328379, "data_time": 0.0020511627197265627, "loss": 0.1994107723236084, "time": 1.7544353008270264, "epoch": 16, "memory": 65239, "step": 23000} +{"accuracy/top1": 99.96842956542969, "data_time": 0.3824879328409831, "time": 1.0150129795074463, "step": 16} +{"lr": 0.0003910110962219801, "data_time": 0.002009677886962891, "loss": 0.20077932626008987, "time": 1.757280707359314, "epoch": 17, "memory": 65239, "step": 23140} +{"lr": 0.0003910110962219801, "data_time": 0.0026701927185058595, "loss": 0.20092237442731858, "time": 1.7545795440673828, "epoch": 17, "memory": 65239, "step": 23240} +{"lr": 0.0003910110962219801, "data_time": 0.001892423629760742, "loss": 0.20009273141622544, "time": 1.7777115821838378, "epoch": 17, "memory": 65239, "step": 23340} +{"lr": 0.0003910110962219801, "data_time": 0.0018848657608032226, "loss": 0.20136475265026094, "time": 1.7528565645217895, "epoch": 17, "memory": 65239, "step": 23440} +{"lr": 0.0003910110962219801, "data_time": 0.00219578742980957, "loss": 0.20083101391792296, "time": 1.7707293510437012, "epoch": 17, "memory": 65239, "step": 23540} +{"lr": 0.0003910110962219801, "data_time": 0.002124381065368652, "loss": 0.2004728764295578, "time": 1.7419605016708375, "epoch": 17, "memory": 65239, "step": 23640} +{"lr": 0.0003910110962219801, "data_time": 0.002007460594177246, "loss": 0.20091038942337036, "time": 1.7536983013153076, "epoch": 17, "memory": 65239, "step": 23740} +{"lr": 0.0003910110962219801, "data_time": 0.0021686792373657227, "loss": 0.1997741401195526, "time": 1.778288984298706, "epoch": 17, "memory": 65239, "step": 23840} +{"lr": 0.0003910110962219801, "data_time": 0.0018552064895629883, "loss": 0.20084185302257537, "time": 1.7693876266479491, "epoch": 17, "memory": 65239, "step": 23940} +{"lr": 0.0003910110962219801, "data_time": 0.001966524124145508, "loss": 0.20196772068738938, "time": 1.7735004901885987, "epoch": 17, "memory": 65239, "step": 24040} +{"lr": 0.0003910110962219801, "data_time": 0.002670621871948242, "loss": 0.20015702694654464, "time": 1.7670262098312377, "epoch": 17, "memory": 65239, "step": 24140} +{"lr": 0.0003910110962219801, "data_time": 0.002411818504333496, "loss": 0.19915879815816878, "time": 1.741189670562744, "epoch": 17, "memory": 65239, "step": 24240} +{"lr": 0.0003910110962219801, "data_time": 0.0019959211349487305, "loss": 0.19983991235494614, "time": 1.7559121608734132, "epoch": 17, "memory": 65239, "step": 24340} +{"lr": 0.0003910110962219801, "data_time": 0.00214080810546875, "loss": 0.19959970712661743, "time": 1.755110502243042, "epoch": 17, "memory": 65239, "step": 24440} +{"accuracy/top1": 99.9810562133789, "data_time": 0.3806263605753581, "time": 0.981286711162991, "step": 17} +{"lr": 0.00022744198424420629, "data_time": 0.001790928840637207, "loss": 0.2004450038075447, "time": 1.7523632764816284, "epoch": 18, "memory": 65239, "step": 24580} +{"lr": 0.00022744198424420629, "data_time": 0.002045464515686035, "loss": 0.19925308525562285, "time": 1.7501772403717042, "epoch": 18, "memory": 65239, "step": 24680} +{"lr": 0.00022744198424420629, "data_time": 0.0019402027130126954, "loss": 0.19926874488592147, "time": 1.7578789472579956, "epoch": 18, "memory": 65239, "step": 24780} +{"lr": 0.00022744198424420629, "data_time": 0.002113199234008789, "loss": 0.20051436126232147, "time": 1.7429935216903687, "epoch": 18, "memory": 65239, "step": 24880} +{"lr": 0.00022744198424420629, "data_time": 0.0023162841796875, "loss": 0.19934873580932616, "time": 1.7414918184280395, "epoch": 18, "memory": 65239, "step": 24980} +{"lr": 0.00022744198424420629, "data_time": 0.002202749252319336, "loss": 0.20008279681205748, "time": 1.759545636177063, "epoch": 18, "memory": 65239, "step": 25080} +{"lr": 0.00022744198424420629, "data_time": 0.0021761178970336912, "loss": 0.20010497868061067, "time": 1.7668198108673097, "epoch": 18, "memory": 65239, "step": 25180} +{"lr": 0.00022744198424420629, "data_time": 0.002141261100769043, "loss": 0.19980052262544631, "time": 1.75847430229187, "epoch": 18, "memory": 65239, "step": 25280} +{"lr": 0.00022744198424420629, "data_time": 0.002082538604736328, "loss": 0.19960130006074905, "time": 1.7639854669570922, "epoch": 18, "memory": 65239, "step": 25380} +{"lr": 0.00022744198424420629, "data_time": 0.0018153905868530274, "loss": 0.20003917217254638, "time": 1.7665620565414428, "epoch": 18, "memory": 65239, "step": 25480} +{"lr": 0.00022744198424420629, "data_time": 0.001948070526123047, "loss": 0.20112757831811906, "time": 1.7692194700241088, "epoch": 18, "memory": 65239, "step": 25580} +{"lr": 0.00022744198424420629, "data_time": 0.002145218849182129, "loss": 0.19917107373476028, "time": 1.7573678493499756, "epoch": 18, "memory": 65239, "step": 25680} +{"lr": 0.00022744198424420629, "data_time": 0.0020323991775512695, "loss": 0.20046317726373672, "time": 1.7658787488937377, "epoch": 18, "memory": 65239, "step": 25780} +{"lr": 0.00022744198424420629, "data_time": 0.0025656461715698243, "loss": 0.19956554025411605, "time": 1.759355902671814, "epoch": 18, "memory": 65239, "step": 25880} +{"accuracy/top1": 99.98737335205078, "data_time": 0.3821522659725613, "time": 0.9835881657070584, "step": 18} +{"lr": 0.00010764224999116871, "data_time": 0.002393293380737305, "loss": 0.19981290847063066, "time": 1.7427427768707275, "epoch": 19, "memory": 65239, "step": 26020} +{"lr": 0.00010764224999116871, "data_time": 0.0017134666442871094, "loss": 0.20034122914075853, "time": 1.758993673324585, "epoch": 19, "memory": 65239, "step": 26120} +{"lr": 0.00010764224999116871, "data_time": 0.0023830890655517577, "loss": 0.19991809427738189, "time": 1.7672643899917602, "epoch": 19, "memory": 65239, "step": 26220} +{"lr": 0.00010764224999116871, "data_time": 0.0021048545837402343, "loss": 0.19866206049919127, "time": 1.74294593334198, "epoch": 19, "memory": 65239, "step": 26320} +{"lr": 0.00010764224999116871, "data_time": 0.0021180152893066407, "loss": 0.19864848852157593, "time": 1.7447846412658692, "epoch": 19, "memory": 65239, "step": 26420} +{"lr": 0.00010764224999116871, "data_time": 0.002152705192565918, "loss": 0.19899600148200988, "time": 1.768186354637146, "epoch": 19, "memory": 65239, "step": 26520} +{"lr": 0.00010764224999116871, "data_time": 0.0022055387496948244, "loss": 0.19933582693338395, "time": 1.753677487373352, "epoch": 19, "memory": 65239, "step": 26620} +{"lr": 0.00010764224999116871, "data_time": 0.002377963066101074, "loss": 0.19929366260766984, "time": 1.7775017261505126, "epoch": 19, "memory": 65239, "step": 26720} +{"lr": 0.00010764224999116871, "data_time": 0.0021529436111450196, "loss": 0.1995449736714363, "time": 1.7686182022094727, "epoch": 19, "memory": 65239, "step": 26820} +{"lr": 0.00010764224999116871, "data_time": 0.0023652315139770508, "loss": 0.19931880980730057, "time": 1.7774329662322998, "epoch": 19, "memory": 65239, "step": 26920} +{"lr": 0.00010764224999116871, "data_time": 0.0022436141967773437, "loss": 0.20051181316375732, "time": 1.7925169229507447, "epoch": 19, "memory": 65239, "step": 27020} +{"lr": 0.00010764224999116871, "data_time": 0.0020327091217041014, "loss": 0.19891057461500167, "time": 1.7671783685684204, "epoch": 19, "memory": 65239, "step": 27120} +{"lr": 0.00010764224999116871, "data_time": 0.002000594139099121, "loss": 0.19943817853927612, "time": 1.7406118154525756, "epoch": 19, "memory": 65239, "step": 27220} +{"lr": 0.00010764224999116871, "data_time": 0.0021384477615356444, "loss": 0.19929901659488677, "time": 1.7579497337341308, "epoch": 19, "memory": 65239, "step": 27320} +{"accuracy/top1": 99.98737335205078, "data_time": 0.3951967822180854, "time": 0.9960782527923584, "step": 19} +{"lr": 3.456176051270038e-05, "data_time": 0.002686166763305664, "loss": 0.19861729741096495, "time": 1.7411158561706543, "epoch": 20, "memory": 65239, "step": 27460} +{"lr": 3.456176051270038e-05, "data_time": 0.002486705780029297, "loss": 0.19931466579437257, "time": 1.7545936346054076, "epoch": 20, "memory": 65239, "step": 27560} +{"lr": 3.456176051270038e-05, "data_time": 0.0018121004104614258, "loss": 0.2005652129650116, "time": 1.7522696256637573, "epoch": 20, "memory": 65239, "step": 27660} +{"lr": 3.456176051270038e-05, "data_time": 0.0021408796310424805, "loss": 0.19871699512004853, "time": 1.758206844329834, "epoch": 20, "memory": 65239, "step": 27760} +{"lr": 3.456176051270038e-05, "data_time": 0.0021448135375976562, "loss": 0.19933580756187438, "time": 1.7568791151046752, "epoch": 20, "memory": 65239, "step": 27860} +{"lr": 3.456176051270038e-05, "data_time": 0.0019590139389038088, "loss": 0.1991402491927147, "time": 1.7809900283813476, "epoch": 20, "memory": 65239, "step": 27960} +{"lr": 3.456176051270038e-05, "data_time": 0.0018506526947021484, "loss": 0.19941504597663878, "time": 1.7550737619400025, "epoch": 20, "memory": 65239, "step": 28060} +{"lr": 3.456176051270038e-05, "data_time": 0.001998114585876465, "loss": 0.19986934661865235, "time": 1.7554445028305055, "epoch": 20, "memory": 65239, "step": 28160} +{"lr": 3.456176051270038e-05, "data_time": 0.0020247936248779298, "loss": 0.1989905744791031, "time": 1.7436093091964722, "epoch": 20, "memory": 65239, "step": 28260} +{"lr": 3.456176051270038e-05, "data_time": 0.0021062135696411134, "loss": 0.1992984741926193, "time": 1.7679325580596923, "epoch": 20, "memory": 65239, "step": 28360} +{"lr": 3.456176051270038e-05, "data_time": 0.0017922401428222656, "loss": 0.19895447343587874, "time": 1.7443328380584717, "epoch": 20, "memory": 65239, "step": 28460} +{"lr": 3.456176051270038e-05, "data_time": 0.0025100231170654295, "loss": 0.19951033294200898, "time": 1.7789090394973754, "epoch": 20, "memory": 65239, "step": 28560} +{"lr": 3.456176051270038e-05, "data_time": 0.0020704030990600585, "loss": 0.19933993071317674, "time": 1.7605851650238038, "epoch": 20, "memory": 65239, "step": 28660} +{"lr": 3.456176051270038e-05, "data_time": 0.0021323919296264648, "loss": 0.19881039708852768, "time": 1.7534290552139282, "epoch": 20, "memory": 65239, "step": 28760} +{"accuracy/top1": 99.98737335205078, "data_time": 0.38388779428270126, "time": 0.9837023682064481, "step": 20} diff --git a/convnext_base_8xb512_4e-3lr_5m/20230531_172616/vis_data/vis_image/0000000.png_1.png b/convnext_base_8xb512_4e-3lr_5m/20230531_172616/vis_data/vis_image/0000000.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..6f264824565adf0630fdfc2e331f1ad6f2495172 Binary files /dev/null and b/convnext_base_8xb512_4e-3lr_5m/20230531_172616/vis_data/vis_image/0000000.png_1.png differ diff --git a/convnext_base_8xb512_4e-3lr_5m/20230531_172616/vis_data/vis_image/0000000.png_10.png b/convnext_base_8xb512_4e-3lr_5m/20230531_172616/vis_data/vis_image/0000000.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..735eb30c6d4a289c06416cbe6a29f4c3f5fd2291 Binary files /dev/null and b/convnext_base_8xb512_4e-3lr_5m/20230531_172616/vis_data/vis_image/0000000.png_10.png differ diff --git a/convnext_base_8xb512_4e-3lr_5m/20230531_172616/vis_data/vis_image/0000000.png_11.png b/convnext_base_8xb512_4e-3lr_5m/20230531_172616/vis_data/vis_image/0000000.png_11.png new file mode 100644 index 0000000000000000000000000000000000000000..5dd93c02c1fad9212e595b965d83c2e69c99319d Binary files /dev/null and b/convnext_base_8xb512_4e-3lr_5m/20230531_172616/vis_data/vis_image/0000000.png_11.png differ diff --git a/convnext_base_8xb512_4e-3lr_5m/20230531_172616/vis_data/vis_image/0000000.png_12.png b/convnext_base_8xb512_4e-3lr_5m/20230531_172616/vis_data/vis_image/0000000.png_12.png new file mode 100644 index 0000000000000000000000000000000000000000..5dd93c02c1fad9212e595b965d83c2e69c99319d Binary files /dev/null and b/convnext_base_8xb512_4e-3lr_5m/20230531_172616/vis_data/vis_image/0000000.png_12.png differ diff --git a/convnext_base_8xb512_4e-3lr_5m/20230531_172616/vis_data/vis_image/0000000.png_13.png b/convnext_base_8xb512_4e-3lr_5m/20230531_172616/vis_data/vis_image/0000000.png_13.png new file mode 100644 index 0000000000000000000000000000000000000000..5dd93c02c1fad9212e595b965d83c2e69c99319d Binary files /dev/null and b/convnext_base_8xb512_4e-3lr_5m/20230531_172616/vis_data/vis_image/0000000.png_13.png differ diff --git a/convnext_base_8xb512_4e-3lr_5m/20230531_172616/vis_data/vis_image/0000000.png_14.png b/convnext_base_8xb512_4e-3lr_5m/20230531_172616/vis_data/vis_image/0000000.png_14.png new file mode 100644 index 0000000000000000000000000000000000000000..5dd93c02c1fad9212e595b965d83c2e69c99319d Binary files /dev/null and b/convnext_base_8xb512_4e-3lr_5m/20230531_172616/vis_data/vis_image/0000000.png_14.png differ diff --git a/convnext_base_8xb512_4e-3lr_5m/20230531_172616/vis_data/vis_image/0000000.png_15.png b/convnext_base_8xb512_4e-3lr_5m/20230531_172616/vis_data/vis_image/0000000.png_15.png new file mode 100644 index 0000000000000000000000000000000000000000..5dd93c02c1fad9212e595b965d83c2e69c99319d Binary files /dev/null and b/convnext_base_8xb512_4e-3lr_5m/20230531_172616/vis_data/vis_image/0000000.png_15.png differ diff --git a/convnext_base_8xb512_4e-3lr_5m/20230531_172616/vis_data/vis_image/0000000.png_16.png b/convnext_base_8xb512_4e-3lr_5m/20230531_172616/vis_data/vis_image/0000000.png_16.png new file mode 100644 index 0000000000000000000000000000000000000000..5dd93c02c1fad9212e595b965d83c2e69c99319d Binary files /dev/null and b/convnext_base_8xb512_4e-3lr_5m/20230531_172616/vis_data/vis_image/0000000.png_16.png differ diff --git a/convnext_base_8xb512_4e-3lr_5m/20230531_172616/vis_data/vis_image/0000000.png_17.png b/convnext_base_8xb512_4e-3lr_5m/20230531_172616/vis_data/vis_image/0000000.png_17.png new file mode 100644 index 0000000000000000000000000000000000000000..5dd93c02c1fad9212e595b965d83c2e69c99319d Binary files /dev/null and b/convnext_base_8xb512_4e-3lr_5m/20230531_172616/vis_data/vis_image/0000000.png_17.png differ diff --git a/convnext_base_8xb512_4e-3lr_5m/20230531_172616/vis_data/vis_image/0000000.png_18.png b/convnext_base_8xb512_4e-3lr_5m/20230531_172616/vis_data/vis_image/0000000.png_18.png new file mode 100644 index 0000000000000000000000000000000000000000..5dd93c02c1fad9212e595b965d83c2e69c99319d Binary files /dev/null and b/convnext_base_8xb512_4e-3lr_5m/20230531_172616/vis_data/vis_image/0000000.png_18.png differ diff --git a/convnext_base_8xb512_4e-3lr_5m/20230531_172616/vis_data/vis_image/0000000.png_19.png b/convnext_base_8xb512_4e-3lr_5m/20230531_172616/vis_data/vis_image/0000000.png_19.png new file mode 100644 index 0000000000000000000000000000000000000000..5dd93c02c1fad9212e595b965d83c2e69c99319d Binary files /dev/null and b/convnext_base_8xb512_4e-3lr_5m/20230531_172616/vis_data/vis_image/0000000.png_19.png differ diff --git a/convnext_base_8xb512_4e-3lr_5m/20230531_172616/vis_data/vis_image/0000000.png_2.png b/convnext_base_8xb512_4e-3lr_5m/20230531_172616/vis_data/vis_image/0000000.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..6f264824565adf0630fdfc2e331f1ad6f2495172 Binary files /dev/null and b/convnext_base_8xb512_4e-3lr_5m/20230531_172616/vis_data/vis_image/0000000.png_2.png differ diff --git a/convnext_base_8xb512_4e-3lr_5m/20230531_172616/vis_data/vis_image/0000000.png_20.png b/convnext_base_8xb512_4e-3lr_5m/20230531_172616/vis_data/vis_image/0000000.png_20.png new file mode 100644 index 0000000000000000000000000000000000000000..f86f1c3663d5c8be96cce83a359c576789781e55 Binary files /dev/null and b/convnext_base_8xb512_4e-3lr_5m/20230531_172616/vis_data/vis_image/0000000.png_20.png differ diff --git a/convnext_base_8xb512_4e-3lr_5m/20230531_172616/vis_data/vis_image/0000000.png_3.png b/convnext_base_8xb512_4e-3lr_5m/20230531_172616/vis_data/vis_image/0000000.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..6f264824565adf0630fdfc2e331f1ad6f2495172 Binary files /dev/null and b/convnext_base_8xb512_4e-3lr_5m/20230531_172616/vis_data/vis_image/0000000.png_3.png differ diff --git a/convnext_base_8xb512_4e-3lr_5m/20230531_172616/vis_data/vis_image/0000000.png_4.png b/convnext_base_8xb512_4e-3lr_5m/20230531_172616/vis_data/vis_image/0000000.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..6f264824565adf0630fdfc2e331f1ad6f2495172 Binary files /dev/null and b/convnext_base_8xb512_4e-3lr_5m/20230531_172616/vis_data/vis_image/0000000.png_4.png differ diff --git a/convnext_base_8xb512_4e-3lr_5m/20230531_172616/vis_data/vis_image/0000000.png_5.png b/convnext_base_8xb512_4e-3lr_5m/20230531_172616/vis_data/vis_image/0000000.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..3068bccfa8ed5f3ad7e0110060b2f984e12a99fb Binary files /dev/null and b/convnext_base_8xb512_4e-3lr_5m/20230531_172616/vis_data/vis_image/0000000.png_5.png differ diff --git a/convnext_base_8xb512_4e-3lr_5m/20230531_172616/vis_data/vis_image/0000000.png_6.png b/convnext_base_8xb512_4e-3lr_5m/20230531_172616/vis_data/vis_image/0000000.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..f86f1c3663d5c8be96cce83a359c576789781e55 Binary files /dev/null and b/convnext_base_8xb512_4e-3lr_5m/20230531_172616/vis_data/vis_image/0000000.png_6.png differ diff --git a/convnext_base_8xb512_4e-3lr_5m/20230531_172616/vis_data/vis_image/0000000.png_7.png b/convnext_base_8xb512_4e-3lr_5m/20230531_172616/vis_data/vis_image/0000000.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..735eb30c6d4a289c06416cbe6a29f4c3f5fd2291 Binary files /dev/null and b/convnext_base_8xb512_4e-3lr_5m/20230531_172616/vis_data/vis_image/0000000.png_7.png differ diff --git a/convnext_base_8xb512_4e-3lr_5m/20230531_172616/vis_data/vis_image/0000000.png_8.png b/convnext_base_8xb512_4e-3lr_5m/20230531_172616/vis_data/vis_image/0000000.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..8fdd29eba7045e367ec6aeb6fea13424393515b7 Binary files /dev/null and b/convnext_base_8xb512_4e-3lr_5m/20230531_172616/vis_data/vis_image/0000000.png_8.png differ diff --git a/convnext_base_8xb512_4e-3lr_5m/20230531_172616/vis_data/vis_image/0000000.png_9.png b/convnext_base_8xb512_4e-3lr_5m/20230531_172616/vis_data/vis_image/0000000.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..8fdd29eba7045e367ec6aeb6fea13424393515b7 Binary files /dev/null and b/convnext_base_8xb512_4e-3lr_5m/20230531_172616/vis_data/vis_image/0000000.png_9.png differ diff --git a/convnext_base_8xb512_4e-3lr_5m/convnext_base_8xb512_fake5m.py b/convnext_base_8xb512_4e-3lr_5m/convnext_base_8xb512_fake5m.py new file mode 100644 index 0000000000000000000000000000000000000000..5c21ac599661bd296a59c3b80c82afd4217314f8 --- /dev/null +++ b/convnext_base_8xb512_4e-3lr_5m/convnext_base_8xb512_fake5m.py @@ -0,0 +1,170 @@ +optim_wrapper = dict( + optimizer=dict( + type='AdamW', + lr=0.004, + weight_decay=0.05, + eps=1e-08, + betas=(0.9, 0.999), + _scope_='mmpretrain'), + paramwise_cfg=dict( + norm_decay_mult=0.0, + bias_decay_mult=0.0, + flat_decay_mult=0.0, + custom_keys=dict({ + '.absolute_pos_embed': dict(decay_mult=0.0), + '.relative_position_bias_table': dict(decay_mult=0.0) + })), + type='AmpOptimWrapper', + dtype='bfloat16', + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=True, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=20, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=4096) +model = dict( + type='ImageClassifier', + backbone=dict(type='ConvNeXt', arch='base', drop_path_rate=0.5), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=1024, + loss=dict( + type='LabelSmoothLoss', label_smooth_val=0.1, mode='original'), + init_cfg=None), + init_cfg=dict( + type='TruncNormal', layer=['Conv2d', 'Linear'], std=0.02, bias=0.0), + train_cfg=None) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=512, + num_workers=10, + dataset=dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stablediffusionV1-5R2-dpmsolver-25-5m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/convnext_base_8xb512_4e-3lr_5m' diff --git a/convnext_base_8xb512_4e-3lr_5m/epoch_1.pth b/convnext_base_8xb512_4e-3lr_5m/epoch_1.pth new file mode 100644 index 0000000000000000000000000000000000000000..56b251b2cb60c7e06f2c7ef20b13416584b79634 --- /dev/null +++ b/convnext_base_8xb512_4e-3lr_5m/epoch_1.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:1cc92d20e1667049c081201ae23b3d18f4fc917a5f16049255e37527ba68c661 +size 1401795744 diff --git a/convnext_base_8xb512_4e-3lr_5m/epoch_10.pth b/convnext_base_8xb512_4e-3lr_5m/epoch_10.pth new file mode 100644 index 0000000000000000000000000000000000000000..d391edc5e56cd17a2ffb9ec4f6fd1debee1b6548 --- /dev/null +++ b/convnext_base_8xb512_4e-3lr_5m/epoch_10.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ed2804ddba24f39f8db157b97dee1a8cdd6093bada48d085b7441486a13890d2 +size 1402744096 diff --git a/convnext_base_8xb512_4e-3lr_5m/epoch_11.pth b/convnext_base_8xb512_4e-3lr_5m/epoch_11.pth new file mode 100644 index 0000000000000000000000000000000000000000..d29d8a3f3593c52e9ef89af22e2ddfd6f57d06b4 --- /dev/null +++ b/convnext_base_8xb512_4e-3lr_5m/epoch_11.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c6cab0e4d3163aef867f16d9ea94b73feae72a2c653e83227213056498204142 +size 1402849888 diff --git a/convnext_base_8xb512_4e-3lr_5m/epoch_12.pth b/convnext_base_8xb512_4e-3lr_5m/epoch_12.pth new file mode 100644 index 0000000000000000000000000000000000000000..be08264d7a74c82f5e48741d1fa0145d336f827c --- /dev/null +++ b/convnext_base_8xb512_4e-3lr_5m/epoch_12.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:b4388c4c88da99506c01fa9dfe353a3834effc62f4f3b54b1d38f4502c8a2c3b +size 1402958560 diff --git a/convnext_base_8xb512_4e-3lr_5m/epoch_13.pth b/convnext_base_8xb512_4e-3lr_5m/epoch_13.pth new file mode 100644 index 0000000000000000000000000000000000000000..9f73bc8e5958e3b434d24de31feeee2d6faf0d4d --- /dev/null +++ b/convnext_base_8xb512_4e-3lr_5m/epoch_13.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:eb551641218255d7b90302abe271b93c0e249e9d128ece4986e31f7fe066c305 +size 1403065568 diff --git a/convnext_base_8xb512_4e-3lr_5m/epoch_14.pth b/convnext_base_8xb512_4e-3lr_5m/epoch_14.pth new file mode 100644 index 0000000000000000000000000000000000000000..1bc70b8eb581f9931700a481c330f580225ca221 --- /dev/null +++ b/convnext_base_8xb512_4e-3lr_5m/epoch_14.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:12ad3263d52ccd9f25190f48256804f4bcf8868dc04695439dc733dcb2e2aaab +size 1403171168 diff --git a/convnext_base_8xb512_4e-3lr_5m/epoch_15.pth b/convnext_base_8xb512_4e-3lr_5m/epoch_15.pth new file mode 100644 index 0000000000000000000000000000000000000000..e511a9030a80b47ff1cc0fab0874eeff725f8288 --- /dev/null +++ b/convnext_base_8xb512_4e-3lr_5m/epoch_15.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:b761d93a2164bb33fc63fedfff51ab278870128a5059f05ea69a9501f085668b +size 1403275360 diff --git a/convnext_base_8xb512_4e-3lr_5m/epoch_16.pth b/convnext_base_8xb512_4e-3lr_5m/epoch_16.pth new file mode 100644 index 0000000000000000000000000000000000000000..8005de5b1ef047fd1c4daa2faf58df71ea70c6cd --- /dev/null +++ b/convnext_base_8xb512_4e-3lr_5m/epoch_16.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:f9a55f2a7ea277b8e1eac62dc965107d375aeb8e47d47bbb36cc68bd0914c969 +size 1403381088 diff --git a/convnext_base_8xb512_4e-3lr_5m/epoch_17.pth b/convnext_base_8xb512_4e-3lr_5m/epoch_17.pth new file mode 100644 index 0000000000000000000000000000000000000000..3b75d1783919c888323e94718fd64e7fc168f6a0 --- /dev/null +++ b/convnext_base_8xb512_4e-3lr_5m/epoch_17.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:33c3e0c213471443758516033a268fe33407d57c8604d6837ef0c664b9d557c1 +size 1403489696 diff --git a/convnext_base_8xb512_4e-3lr_5m/epoch_18.pth b/convnext_base_8xb512_4e-3lr_5m/epoch_18.pth new file mode 100644 index 0000000000000000000000000000000000000000..854d42ecd33d2008c3b6c018e169fb508f19494f --- /dev/null +++ b/convnext_base_8xb512_4e-3lr_5m/epoch_18.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:7abd045244232c2e869dc2a77741a349ede68f286ecbca1d9219c87faac836fd +size 1403597856 diff --git a/convnext_base_8xb512_4e-3lr_5m/epoch_19.pth b/convnext_base_8xb512_4e-3lr_5m/epoch_19.pth new file mode 100644 index 0000000000000000000000000000000000000000..5ad9b1d994637c3337493f477021e956ce4e0349 --- /dev/null +++ b/convnext_base_8xb512_4e-3lr_5m/epoch_19.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:2b9bf3ccb971b6efaf3c0426892883d7d84007d36462a6ee7312b5989b8285dc +size 1403701664 diff --git a/convnext_base_8xb512_4e-3lr_5m/epoch_2.pth b/convnext_base_8xb512_4e-3lr_5m/epoch_2.pth new file mode 100644 index 0000000000000000000000000000000000000000..3368c2e09c2316ac453bfbd12306ceef55c6c101 --- /dev/null +++ b/convnext_base_8xb512_4e-3lr_5m/epoch_2.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:709360581cc1b9cfb4029e515c0e8c8b9715bc9ce64aed8810e3f330541905e4 +size 1401902304 diff --git a/convnext_base_8xb512_4e-3lr_5m/epoch_20.pth b/convnext_base_8xb512_4e-3lr_5m/epoch_20.pth new file mode 100644 index 0000000000000000000000000000000000000000..fc28fa2da74303f46d052f19ba63594c20584f41 --- /dev/null +++ b/convnext_base_8xb512_4e-3lr_5m/epoch_20.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:f10e12d22139329ebf94ab44554900eaef4065818d65b6e78e0e94f60af2f753 +size 1403803872 diff --git a/convnext_base_8xb512_4e-3lr_5m/epoch_3.pth b/convnext_base_8xb512_4e-3lr_5m/epoch_3.pth new file mode 100644 index 0000000000000000000000000000000000000000..a18e9b681e283eff856400c4f3cbe05530d96a81 --- /dev/null +++ b/convnext_base_8xb512_4e-3lr_5m/epoch_3.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:59394afe306c19dfd4879efc14a2d4bd76393e32bb2635c0f7333bf83535af5b +size 1402006240 diff --git a/convnext_base_8xb512_4e-3lr_5m/epoch_4.pth b/convnext_base_8xb512_4e-3lr_5m/epoch_4.pth new file mode 100644 index 0000000000000000000000000000000000000000..0cdb6437ef15a1159ecfad07b4d4ba4ba1867035 --- /dev/null +++ b/convnext_base_8xb512_4e-3lr_5m/epoch_4.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:cc2ba742cde61530e9bc8514ced854722bd137fa5dbca7f1c655564a0f47cce5 +size 1402111712 diff --git a/convnext_base_8xb512_4e-3lr_5m/epoch_5.pth b/convnext_base_8xb512_4e-3lr_5m/epoch_5.pth new file mode 100644 index 0000000000000000000000000000000000000000..382175af23d534e07ea1cf8a087cf0f747345bb2 --- /dev/null +++ b/convnext_base_8xb512_4e-3lr_5m/epoch_5.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:94535b73178da559d197a35a7d928d203a14ea6f7279cd0e2e31d619331260a7 +size 1402220384 diff --git a/convnext_base_8xb512_4e-3lr_5m/epoch_6.pth b/convnext_base_8xb512_4e-3lr_5m/epoch_6.pth new file mode 100644 index 0000000000000000000000000000000000000000..1ffcf1219c2b6ef2e630eb959c4581b0e26b5aac --- /dev/null +++ b/convnext_base_8xb512_4e-3lr_5m/epoch_6.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:9832a115902cecff8a605fb00ec0d393abed92fcf2a4e6e20852cb50a56aace5 +size 1402326048 diff --git a/convnext_base_8xb512_4e-3lr_5m/epoch_7.pth b/convnext_base_8xb512_4e-3lr_5m/epoch_7.pth new file mode 100644 index 0000000000000000000000000000000000000000..265a3cfd89ba40e3afecadf84cd8edb64a25116e --- /dev/null +++ b/convnext_base_8xb512_4e-3lr_5m/epoch_7.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:0a8ddb971faa67cfebd6074fd6745b25e847f79ed52a5bc1f52ddc7d26c5377f +size 1402434528 diff --git a/convnext_base_8xb512_4e-3lr_5m/epoch_8.pth b/convnext_base_8xb512_4e-3lr_5m/epoch_8.pth new file mode 100644 index 0000000000000000000000000000000000000000..44cb3fa7b6d3347102dc5bbecce3b02458bf6332 --- /dev/null +++ b/convnext_base_8xb512_4e-3lr_5m/epoch_8.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:825f47d023502039fdd916e8801bc2541d0bbec575e855e4f4bbdda3ed2157dd +size 1402537312 diff --git a/convnext_base_8xb512_4e-3lr_5m/epoch_9.pth b/convnext_base_8xb512_4e-3lr_5m/epoch_9.pth new file mode 100644 index 0000000000000000000000000000000000000000..4d9f5045cb688323a7e1557f8cb15c24e2e1e5fd --- /dev/null +++ b/convnext_base_8xb512_4e-3lr_5m/epoch_9.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:1c5a6358443381607827224790ad350a0401419ad48d2eeb13312f3054518d54 +size 1402639904 diff --git a/convnext_base_8xb512_4e-3lr_5m/last_checkpoint b/convnext_base_8xb512_4e-3lr_5m/last_checkpoint new file mode 100644 index 0000000000000000000000000000000000000000..b04cbf1c90951d6749f0c8b1f4bf8b299cea8f06 --- /dev/null +++ b/convnext_base_8xb512_4e-3lr_5m/last_checkpoint @@ -0,0 +1 @@ +/mnt/petrelfs/luzeyu/workspace/fakebench/mmpretrain/workdir/convnext_base_8xb512_4e-3lr_5m/epoch_20.pth \ No newline at end of file diff --git a/convnext_small_4xb256_4e-3lr_5m/20230531_173252/20230531_173252.log b/convnext_small_4xb256_4e-3lr_5m/20230531_173252/20230531_173252.log new file mode 100644 index 0000000000000000000000000000000000000000..c32c92ed2bf9523b36ee5b09851b118bc529dfb6 --- /dev/null +++ b/convnext_small_4xb256_4e-3lr_5m/20230531_173252/20230531_173252.log @@ -0,0 +1,2609 @@ +2023/05/31 17:32:56 - mmengine - INFO - +------------------------------------------------------------ +System environment: + sys.platform: linux + Python: 3.10.9 (main, Mar 8 2023, 10:47:38) [GCC 11.2.0] + CUDA available: True + numpy_random_seed: 151848001 + GPU 0,1,2,3: NVIDIA A100-SXM4-80GB + CUDA_HOME: /mnt/petrelfs/share/cuda-11.6 + NVCC: Cuda compilation tools, release 11.6, V11.6.124 + GCC: gcc (GCC) 4.8.5 20150623 (Red Hat 4.8.5-44) + PyTorch: 1.13.1 + PyTorch compiling details: PyTorch built with: + - GCC 9.3 + - C++ Version: 201402 + - Intel(R) oneAPI Math Kernel Library Version 2021.4-Product Build 20210904 for Intel(R) 64 architecture applications + - Intel(R) MKL-DNN v2.6.0 (Git Hash 52b5f107dd9cf10910aaa19cb47f3abf9b349815) + - OpenMP 201511 (a.k.a. OpenMP 4.5) + - LAPACK is enabled (usually provided by MKL) + - NNPACK is enabled + - CPU capability usage: AVX2 + - CUDA Runtime 11.6 + - NVCC architecture flags: -gencode;arch=compute_37,code=sm_37;-gencode;arch=compute_50,code=sm_50;-gencode;arch=compute_60,code=sm_60;-gencode;arch=compute_61,code=sm_61;-gencode;arch=compute_70,code=sm_70;-gencode;arch=compute_75,code=sm_75;-gencode;arch=compute_80,code=sm_80;-gencode;arch=compute_86,code=sm_86;-gencode;arch=compute_37,code=compute_37 + - CuDNN 8.3.2 (built against CUDA 11.5) + - Magma 2.6.1 + - Build settings: BLAS_INFO=mkl, BUILD_TYPE=Release, CUDA_VERSION=11.6, CUDNN_VERSION=8.3.2, CXX_COMPILER=/opt/rh/devtoolset-9/root/usr/bin/c++, CXX_FLAGS= -fabi-version=11 -Wno-deprecated -fvisibility-inlines-hidden -DUSE_PTHREADPOOL -fopenmp -DNDEBUG -DUSE_KINETO -DUSE_FBGEMM -DUSE_QNNPACK -DUSE_PYTORCH_QNNPACK -DUSE_XNNPACK -DSYMBOLICATE_MOBILE_DEBUG_HANDLE -DEDGE_PROFILER_USE_KINETO -O2 -fPIC -Wno-narrowing -Wall -Wextra -Werror=return-type -Werror=non-virtual-dtor -Wno-missing-field-initializers -Wno-type-limits -Wno-array-bounds -Wno-unknown-pragmas -Wunused-local-typedefs -Wno-unused-parameter -Wno-unused-function -Wno-unused-result -Wno-strict-overflow -Wno-strict-aliasing -Wno-error=deprecated-declarations -Wno-stringop-overflow -Wno-psabi -Wno-error=pedantic -Wno-error=redundant-decls -Wno-error=old-style-cast -fdiagnostics-color=always -faligned-new -Wno-unused-but-set-variable -Wno-maybe-uninitialized -fno-math-errno -fno-trapping-math -Werror=format -Werror=cast-function-type -Wno-stringop-overflow, LAPACK_INFO=mkl, PERF_WITH_AVX=1, PERF_WITH_AVX2=1, PERF_WITH_AVX512=1, TORCH_VERSION=1.13.1, USE_CUDA=ON, USE_CUDNN=ON, USE_EXCEPTION_PTR=1, USE_GFLAGS=OFF, USE_GLOG=OFF, USE_MKL=ON, USE_MKLDNN=ON, USE_MPI=OFF, USE_NCCL=ON, USE_NNPACK=ON, USE_OPENMP=ON, USE_ROCM=OFF, + + TorchVision: 0.14.1 + OpenCV: 4.7.0 + MMEngine: 0.7.3 + +Runtime environment: + cudnn_benchmark: True + mp_cfg: {'mp_start_method': 'fork', 'opencv_num_threads': 0} + dist_cfg: {'backend': 'nccl'} + seed: None + deterministic: False + Distributed launcher: slurm + Distributed training: True + GPU number: 4 +------------------------------------------------------------ + +2023/05/31 17:32:57 - mmengine - INFO - Config: +optim_wrapper = dict( + optimizer=dict( + type='AdamW', + lr=0.004, + weight_decay=0.05, + eps=1e-08, + betas=(0.9, 0.999), + _scope_='mmpretrain'), + paramwise_cfg=dict( + norm_decay_mult=0.0, + bias_decay_mult=0.0, + flat_decay_mult=0.0, + custom_keys=dict({ + '.absolute_pos_embed': dict(decay_mult=0.0), + '.relative_position_bias_table': dict(decay_mult=0.0) + })), + type='AmpOptimWrapper', + dtype='bfloat16', + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=True, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=20, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=4096) +model = dict( + type='ImageClassifier', + backbone=dict(type='ConvNeXt', arch='small', drop_path_rate=0.4), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=768, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + init_cfg=None), + init_cfg=dict( + type='TruncNormal', layer=['Conv2d', 'Linear'], std=0.02, bias=0.0), + train_cfg=None) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stablediffusionV1-5R2-dpmsolver-25-5m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/convnext_small_4xb256_4e-3lr_5m' + +2023/05/31 17:33:00 - mmengine - INFO - Hooks will be executed in the following order: +before_run: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_load_checkpoint: +(ABOVE_NORMAL) EMAHook + -------------------- +before_train: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_train_epoch: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(NORMAL ) DistSamplerSeedHook + -------------------- +before_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook + -------------------- +after_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +after_train_epoch: +(NORMAL ) IterTimerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_val_epoch: +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook + -------------------- +before_val_iter: +(NORMAL ) IterTimerHook + -------------------- +after_val_iter: +(NORMAL ) IterTimerHook +(NORMAL ) VisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_val_epoch: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_save_checkpoint: +(ABOVE_NORMAL) EMAHook + -------------------- +after_train: +(VERY_LOW ) CheckpointHook + -------------------- +before_test_epoch: +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook + -------------------- +before_test_iter: +(NORMAL ) IterTimerHook + -------------------- +after_test_iter: +(NORMAL ) IterTimerHook +(NORMAL ) VisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_test_epoch: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_run: +(BELOW_NORMAL) LoggerHook + -------------------- +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.0.0.bias:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.0.1.weight:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.0.1.bias:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.1.0.weight:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.1.0.bias:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.1.1.bias:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.2.0.weight:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.2.0.bias:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.2.1.bias:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.3.0.weight:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.3.0.bias:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.3.1.bias:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.0.0.gamma:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.0.0.depthwise_conv.bias:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.0.0.norm.weight:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.0.0.norm.bias:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.0.0.pointwise_conv1.bias:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.0.0.pointwise_conv2.bias:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.0.1.gamma:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.0.1.depthwise_conv.bias:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.0.1.norm.weight:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.0.1.norm.bias:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.0.1.pointwise_conv1.bias:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.0.1.pointwise_conv2.bias:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.0.2.gamma:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.0.2.depthwise_conv.bias:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.0.2.norm.weight:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.0.2.norm.bias:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.0.2.pointwise_conv1.bias:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.0.2.pointwise_conv2.bias:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.1.0.gamma:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.1.0.depthwise_conv.bias:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.1.0.norm.weight:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.1.0.norm.bias:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.1.0.pointwise_conv1.bias:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.1.0.pointwise_conv2.bias:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.1.1.gamma:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.1.1.depthwise_conv.bias:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.1.1.norm.weight:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.1.1.norm.bias:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.1.1.pointwise_conv1.bias:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.1.1.pointwise_conv2.bias:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.1.2.gamma:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.1.2.depthwise_conv.bias:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.1.2.norm.weight:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.1.2.norm.bias:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.1.2.pointwise_conv1.bias:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.1.2.pointwise_conv2.bias:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.2.0.gamma:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.2.0.depthwise_conv.bias:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.2.0.norm.weight:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.2.0.norm.bias:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.2.0.pointwise_conv1.bias:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.2.0.pointwise_conv2.bias:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.2.1.gamma:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.2.1.depthwise_conv.bias:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.2.1.norm.weight:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.2.1.norm.bias:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.2.1.pointwise_conv1.bias:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.2.1.pointwise_conv2.bias:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.2.2.gamma:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.2.2.depthwise_conv.bias:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.2.2.norm.weight:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.2.2.norm.bias:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.2.2.pointwise_conv1.bias:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.2.2.pointwise_conv2.bias:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.2.3.gamma:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.2.3.depthwise_conv.bias:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.2.3.norm.weight:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.2.3.norm.bias:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.2.3.pointwise_conv1.bias:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.2.3.pointwise_conv2.bias:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.2.4.gamma:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.2.4.depthwise_conv.bias:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.2.4.norm.weight:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.2.4.norm.bias:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.2.4.pointwise_conv1.bias:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.2.4.pointwise_conv2.bias:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.2.5.gamma:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.2.5.depthwise_conv.bias:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.2.5.norm.weight:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.2.5.norm.bias:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.2.5.pointwise_conv1.bias:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.2.5.pointwise_conv2.bias:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.2.6.gamma:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.2.6.depthwise_conv.bias:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.2.6.norm.weight:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.2.6.norm.bias:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.2.6.pointwise_conv1.bias:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.2.6.pointwise_conv2.bias:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.2.7.gamma:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.2.7.depthwise_conv.bias:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.2.7.norm.weight:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.2.7.norm.bias:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.2.7.pointwise_conv1.bias:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.2.7.pointwise_conv2.bias:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.2.8.gamma:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.2.8.depthwise_conv.bias:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.2.8.norm.weight:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.2.8.norm.bias:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.2.8.pointwise_conv1.bias:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.2.8.pointwise_conv2.bias:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.2.9.gamma:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.2.9.depthwise_conv.bias:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.2.9.norm.weight:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.2.9.norm.bias:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.2.9.pointwise_conv1.bias:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.2.9.pointwise_conv2.bias:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.2.10.gamma:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.2.10.depthwise_conv.bias:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.2.10.norm.weight:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.2.10.norm.bias:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.2.10.pointwise_conv1.bias:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.2.10.pointwise_conv2.bias:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.2.11.gamma:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.2.11.depthwise_conv.bias:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.2.11.norm.weight:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.2.11.norm.bias:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.2.11.pointwise_conv1.bias:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.2.11.pointwise_conv2.bias:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.2.12.gamma:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.2.12.depthwise_conv.bias:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.2.12.norm.weight:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.2.12.norm.bias:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.2.12.pointwise_conv1.bias:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.2.12.pointwise_conv2.bias:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.2.13.gamma:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.2.13.depthwise_conv.bias:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.2.13.norm.weight:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.2.13.norm.bias:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.2.13.pointwise_conv1.bias:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.2.13.pointwise_conv2.bias:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.2.14.gamma:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.2.14.depthwise_conv.bias:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.2.14.norm.weight:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.2.14.norm.bias:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.2.14.pointwise_conv1.bias:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.2.14.pointwise_conv2.bias:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.2.15.gamma:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.2.15.depthwise_conv.bias:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.2.15.norm.weight:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.2.15.norm.bias:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.2.15.pointwise_conv1.bias:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.2.15.pointwise_conv2.bias:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.2.16.gamma:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.2.16.depthwise_conv.bias:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.2.16.norm.weight:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.2.16.norm.bias:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.2.16.pointwise_conv1.bias:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.2.16.pointwise_conv2.bias:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.2.17.gamma:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.2.17.depthwise_conv.bias:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.2.17.norm.weight:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.2.17.norm.bias:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.2.17.pointwise_conv1.bias:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.2.17.pointwise_conv2.bias:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.2.18.gamma:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.2.18.depthwise_conv.bias:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.2.18.norm.weight:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.2.18.norm.bias:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.2.18.pointwise_conv1.bias:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.2.18.pointwise_conv2.bias:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.2.19.gamma:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.2.19.depthwise_conv.bias:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.2.19.norm.weight:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.2.19.norm.bias:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.2.19.pointwise_conv1.bias:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.2.19.pointwise_conv2.bias:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.2.20.gamma:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.2.20.depthwise_conv.bias:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.2.20.norm.weight:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.2.20.norm.bias:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.2.20.pointwise_conv1.bias:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.2.20.pointwise_conv2.bias:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.2.21.gamma:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.2.21.depthwise_conv.bias:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.2.21.norm.weight:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.2.21.norm.bias:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.2.21.pointwise_conv1.bias:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.2.21.pointwise_conv2.bias:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.2.22.gamma:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.2.22.depthwise_conv.bias:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.2.22.norm.weight:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.2.22.norm.bias:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.2.22.pointwise_conv1.bias:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.2.22.pointwise_conv2.bias:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.2.23.gamma:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.2.23.depthwise_conv.bias:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.2.23.norm.weight:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.2.23.norm.bias:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.2.23.pointwise_conv1.bias:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.2.23.pointwise_conv2.bias:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.2.24.gamma:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.2.24.depthwise_conv.bias:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.2.24.norm.weight:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.2.24.norm.bias:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.2.24.pointwise_conv1.bias:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.2.24.pointwise_conv2.bias:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.2.25.gamma:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.2.25.depthwise_conv.bias:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.2.25.norm.weight:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.2.25.norm.bias:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.2.25.pointwise_conv1.bias:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.2.25.pointwise_conv2.bias:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.2.26.gamma:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.2.26.depthwise_conv.bias:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.2.26.norm.weight:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.2.26.norm.bias:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.2.26.pointwise_conv1.bias:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.2.26.pointwise_conv2.bias:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.3.0.gamma:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.3.0.depthwise_conv.bias:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.3.0.norm.weight:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.3.0.norm.bias:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.3.0.pointwise_conv1.bias:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.3.0.pointwise_conv2.bias:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.3.1.gamma:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.3.1.depthwise_conv.bias:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.3.1.norm.weight:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.3.1.norm.bias:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.3.1.pointwise_conv1.bias:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.3.1.pointwise_conv2.bias:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.3.2.gamma:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.3.2.depthwise_conv.bias:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.3.2.norm.weight:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.3.2.norm.bias:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.3.2.pointwise_conv1.bias:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.stages.3.2.pointwise_conv2.bias:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.norm3.weight:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- backbone.norm3.bias:weight_decay=0.0 +2023/05/31 17:33:27 - mmengine - INFO - paramwise_options -- head.fc.bias:weight_decay=0.0 +Name of parameter - Initialization information + +backbone.downsample_layers.0.0.weight - torch.Size([96, 3, 4, 4]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.downsample_layers.0.0.bias - torch.Size([96]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.downsample_layers.0.1.weight - torch.Size([96]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.downsample_layers.0.1.bias - torch.Size([96]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.downsample_layers.1.0.weight - torch.Size([96]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.downsample_layers.1.0.bias - torch.Size([96]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.downsample_layers.1.1.weight - torch.Size([192, 96, 2, 2]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.downsample_layers.1.1.bias - torch.Size([192]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.downsample_layers.2.0.weight - torch.Size([192]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.downsample_layers.2.0.bias - torch.Size([192]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.downsample_layers.2.1.weight - torch.Size([384, 192, 2, 2]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.downsample_layers.2.1.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.downsample_layers.3.0.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.downsample_layers.3.0.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.downsample_layers.3.1.weight - torch.Size([768, 384, 2, 2]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.downsample_layers.3.1.bias - torch.Size([768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.0.gamma - torch.Size([96]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.0.0.depthwise_conv.weight - torch.Size([96, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.0.depthwise_conv.bias - torch.Size([96]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.0.norm.weight - torch.Size([96]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.0.0.norm.bias - torch.Size([96]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.0.0.pointwise_conv1.weight - torch.Size([384, 96]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.0.pointwise_conv1.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.0.pointwise_conv2.weight - torch.Size([96, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.0.pointwise_conv2.bias - torch.Size([96]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.1.gamma - torch.Size([96]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.0.1.depthwise_conv.weight - torch.Size([96, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.1.depthwise_conv.bias - torch.Size([96]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.1.norm.weight - torch.Size([96]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.0.1.norm.bias - torch.Size([96]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.0.1.pointwise_conv1.weight - torch.Size([384, 96]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.1.pointwise_conv1.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.1.pointwise_conv2.weight - torch.Size([96, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.1.pointwise_conv2.bias - torch.Size([96]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.2.gamma - torch.Size([96]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.0.2.depthwise_conv.weight - torch.Size([96, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.2.depthwise_conv.bias - torch.Size([96]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.2.norm.weight - torch.Size([96]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.0.2.norm.bias - torch.Size([96]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.0.2.pointwise_conv1.weight - torch.Size([384, 96]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.2.pointwise_conv1.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.2.pointwise_conv2.weight - torch.Size([96, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.2.pointwise_conv2.bias - torch.Size([96]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.0.gamma - torch.Size([192]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.1.0.depthwise_conv.weight - torch.Size([192, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.0.depthwise_conv.bias - torch.Size([192]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.0.norm.weight - torch.Size([192]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.1.0.norm.bias - torch.Size([192]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.1.0.pointwise_conv1.weight - torch.Size([768, 192]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.0.pointwise_conv1.bias - torch.Size([768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.0.pointwise_conv2.weight - torch.Size([192, 768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.0.pointwise_conv2.bias - torch.Size([192]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.1.gamma - torch.Size([192]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.1.1.depthwise_conv.weight - torch.Size([192, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.1.depthwise_conv.bias - torch.Size([192]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.1.norm.weight - torch.Size([192]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.1.1.norm.bias - torch.Size([192]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.1.1.pointwise_conv1.weight - torch.Size([768, 192]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.1.pointwise_conv1.bias - torch.Size([768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.1.pointwise_conv2.weight - torch.Size([192, 768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.1.pointwise_conv2.bias - torch.Size([192]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.2.gamma - torch.Size([192]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.1.2.depthwise_conv.weight - torch.Size([192, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.2.depthwise_conv.bias - torch.Size([192]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.2.norm.weight - torch.Size([192]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.1.2.norm.bias - torch.Size([192]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.1.2.pointwise_conv1.weight - torch.Size([768, 192]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.2.pointwise_conv1.bias - torch.Size([768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.2.pointwise_conv2.weight - torch.Size([192, 768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.2.pointwise_conv2.bias - torch.Size([192]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.0.gamma - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.0.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.0.depthwise_conv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.0.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.0.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.0.pointwise_conv1.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.0.pointwise_conv1.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.0.pointwise_conv2.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.0.pointwise_conv2.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.1.gamma - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.1.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.1.depthwise_conv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.1.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.1.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.1.pointwise_conv1.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.1.pointwise_conv1.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.1.pointwise_conv2.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.1.pointwise_conv2.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.2.gamma - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.2.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.2.depthwise_conv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.2.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.2.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.2.pointwise_conv1.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.2.pointwise_conv1.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.2.pointwise_conv2.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.2.pointwise_conv2.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.3.gamma - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.3.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.3.depthwise_conv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.3.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.3.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.3.pointwise_conv1.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.3.pointwise_conv1.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.3.pointwise_conv2.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.3.pointwise_conv2.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.4.gamma - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.4.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.4.depthwise_conv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.4.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.4.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.4.pointwise_conv1.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.4.pointwise_conv1.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.4.pointwise_conv2.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.4.pointwise_conv2.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.5.gamma - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.5.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.5.depthwise_conv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.5.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.5.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.5.pointwise_conv1.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.5.pointwise_conv1.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.5.pointwise_conv2.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.5.pointwise_conv2.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.6.gamma - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.6.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.6.depthwise_conv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.6.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.6.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.6.pointwise_conv1.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.6.pointwise_conv1.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.6.pointwise_conv2.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.6.pointwise_conv2.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.7.gamma - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.7.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.7.depthwise_conv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.7.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.7.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.7.pointwise_conv1.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.7.pointwise_conv1.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.7.pointwise_conv2.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.7.pointwise_conv2.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.8.gamma - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.8.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.8.depthwise_conv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.8.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.8.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.8.pointwise_conv1.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.8.pointwise_conv1.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.8.pointwise_conv2.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.8.pointwise_conv2.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.9.gamma - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.9.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.9.depthwise_conv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.9.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.9.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.9.pointwise_conv1.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.9.pointwise_conv1.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.9.pointwise_conv2.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.9.pointwise_conv2.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.10.gamma - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.10.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.10.depthwise_conv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.10.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.10.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.10.pointwise_conv1.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.10.pointwise_conv1.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.10.pointwise_conv2.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.10.pointwise_conv2.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.11.gamma - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.11.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.11.depthwise_conv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.11.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.11.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.11.pointwise_conv1.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.11.pointwise_conv1.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.11.pointwise_conv2.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.11.pointwise_conv2.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.12.gamma - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.12.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.12.depthwise_conv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.12.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.12.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.12.pointwise_conv1.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.12.pointwise_conv1.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.12.pointwise_conv2.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.12.pointwise_conv2.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.13.gamma - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.13.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.13.depthwise_conv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.13.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.13.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.13.pointwise_conv1.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.13.pointwise_conv1.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.13.pointwise_conv2.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.13.pointwise_conv2.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.14.gamma - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.14.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.14.depthwise_conv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.14.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.14.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.14.pointwise_conv1.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.14.pointwise_conv1.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.14.pointwise_conv2.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.14.pointwise_conv2.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.15.gamma - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.15.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.15.depthwise_conv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.15.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.15.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.15.pointwise_conv1.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.15.pointwise_conv1.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.15.pointwise_conv2.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.15.pointwise_conv2.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.16.gamma - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.16.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.16.depthwise_conv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.16.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.16.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.16.pointwise_conv1.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.16.pointwise_conv1.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.16.pointwise_conv2.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.16.pointwise_conv2.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.17.gamma - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.17.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.17.depthwise_conv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.17.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.17.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.17.pointwise_conv1.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.17.pointwise_conv1.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.17.pointwise_conv2.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.17.pointwise_conv2.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.18.gamma - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.18.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.18.depthwise_conv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.18.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.18.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.18.pointwise_conv1.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.18.pointwise_conv1.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.18.pointwise_conv2.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.18.pointwise_conv2.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.19.gamma - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.19.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.19.depthwise_conv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.19.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.19.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.19.pointwise_conv1.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.19.pointwise_conv1.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.19.pointwise_conv2.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.19.pointwise_conv2.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.20.gamma - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.20.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.20.depthwise_conv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.20.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.20.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.20.pointwise_conv1.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.20.pointwise_conv1.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.20.pointwise_conv2.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.20.pointwise_conv2.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.21.gamma - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.21.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.21.depthwise_conv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.21.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.21.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.21.pointwise_conv1.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.21.pointwise_conv1.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.21.pointwise_conv2.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.21.pointwise_conv2.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.22.gamma - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.22.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.22.depthwise_conv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.22.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.22.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.22.pointwise_conv1.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.22.pointwise_conv1.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.22.pointwise_conv2.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.22.pointwise_conv2.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.23.gamma - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.23.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.23.depthwise_conv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.23.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.23.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.23.pointwise_conv1.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.23.pointwise_conv1.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.23.pointwise_conv2.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.23.pointwise_conv2.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.24.gamma - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.24.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.24.depthwise_conv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.24.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.24.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.24.pointwise_conv1.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.24.pointwise_conv1.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.24.pointwise_conv2.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.24.pointwise_conv2.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.25.gamma - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.25.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.25.depthwise_conv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.25.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.25.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.25.pointwise_conv1.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.25.pointwise_conv1.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.25.pointwise_conv2.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.25.pointwise_conv2.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.26.gamma - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.26.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.26.depthwise_conv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.26.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.26.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.26.pointwise_conv1.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.26.pointwise_conv1.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.26.pointwise_conv2.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.26.pointwise_conv2.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.0.gamma - torch.Size([768]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.3.0.depthwise_conv.weight - torch.Size([768, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.0.depthwise_conv.bias - torch.Size([768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.0.norm.weight - torch.Size([768]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.3.0.norm.bias - torch.Size([768]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.3.0.pointwise_conv1.weight - torch.Size([3072, 768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.0.pointwise_conv1.bias - torch.Size([3072]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.0.pointwise_conv2.weight - torch.Size([768, 3072]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.0.pointwise_conv2.bias - torch.Size([768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.1.gamma - torch.Size([768]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.3.1.depthwise_conv.weight - torch.Size([768, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.1.depthwise_conv.bias - torch.Size([768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.1.norm.weight - torch.Size([768]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.3.1.norm.bias - torch.Size([768]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.3.1.pointwise_conv1.weight - torch.Size([3072, 768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.1.pointwise_conv1.bias - torch.Size([3072]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.1.pointwise_conv2.weight - torch.Size([768, 3072]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.1.pointwise_conv2.bias - torch.Size([768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.2.gamma - torch.Size([768]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.3.2.depthwise_conv.weight - torch.Size([768, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.2.depthwise_conv.bias - torch.Size([768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.2.norm.weight - torch.Size([768]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.3.2.norm.bias - torch.Size([768]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.3.2.pointwise_conv1.weight - torch.Size([3072, 768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.2.pointwise_conv1.bias - torch.Size([3072]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.2.pointwise_conv2.weight - torch.Size([768, 3072]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.2.pointwise_conv2.bias - torch.Size([768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.norm3.weight - torch.Size([768]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.norm3.bias - torch.Size([768]): +The value is the same before and after calling `init_weights` of ImageClassifier + +head.fc.weight - torch.Size([2, 768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +head.fc.bias - torch.Size([2]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 +2023/05/31 17:33:28 - mmengine - WARNING - "FileClient" will be deprecated in future. Please use io functions in https://mmengine.readthedocs.io/en/latest/api/fileio.html#file-io +2023/05/31 17:33:28 - mmengine - WARNING - "HardDiskBackend" is the alias of "LocalBackend" and the former will be deprecated in future. +2023/05/31 17:33:28 - mmengine - INFO - Checkpoints will be saved to /mnt/petrelfs/luzeyu/workspace/fakebench/mmpretrain/workdir/convnext_small_4xb256_4e-3lr_5m. +2023/05/31 17:35:30 - mmengine - INFO - Epoch(train) [1][ 100/5758] lr: 4.0000e-03 eta: 1 day, 15:02:45 time: 1.1037 data_time: 0.2584 memory: 25071 loss: 0.6135 +2023/05/31 17:37:22 - mmengine - INFO - Epoch(train) [1][ 200/5758] lr: 4.0000e-03 eta: 1 day, 13:21:05 time: 1.0568 data_time: 0.0020 memory: 25071 loss: 0.5921 +2023/05/31 17:39:12 - mmengine - INFO - Epoch(train) [1][ 300/5758] lr: 4.0000e-03 eta: 1 day, 12:35:36 time: 1.0625 data_time: 0.0015 memory: 25071 loss: 0.5693 +2023/05/31 17:41:03 - mmengine - INFO - Epoch(train) [1][ 400/5758] lr: 4.0000e-03 eta: 1 day, 12:17:30 time: 1.0132 data_time: 0.0018 memory: 25071 loss: 0.5560 +2023/05/31 17:42:49 - mmengine - INFO - Epoch(train) [1][ 500/5758] lr: 4.0000e-03 eta: 1 day, 11:45:23 time: 1.0576 data_time: 0.0018 memory: 25071 loss: 0.5387 +2023/05/31 17:44:33 - mmengine - INFO - Epoch(train) [1][ 600/5758] lr: 4.0000e-03 eta: 1 day, 11:17:54 time: 1.0222 data_time: 0.0015 memory: 25071 loss: 0.5206 +2023/05/31 17:46:16 - mmengine - INFO - Epoch(train) [1][ 700/5758] lr: 4.0000e-03 eta: 1 day, 10:54:07 time: 1.0578 data_time: 0.0011 memory: 25071 loss: 0.5122 +2023/05/31 17:47:55 - mmengine - INFO - Epoch(train) [1][ 800/5758] lr: 4.0000e-03 eta: 1 day, 10:24:46 time: 0.9828 data_time: 0.0019 memory: 25071 loss: 0.5337 +2023/05/31 17:49:35 - mmengine - INFO - Epoch(train) [1][ 900/5758] lr: 4.0000e-03 eta: 1 day, 10:05:36 time: 1.0143 data_time: 0.0015 memory: 25071 loss: 0.4814 +2023/05/31 17:51:11 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_20230531_173252 +2023/05/31 17:51:11 - mmengine - INFO - Epoch(train) [1][1000/5758] lr: 4.0000e-03 eta: 1 day, 9:43:16 time: 1.0005 data_time: 0.0016 memory: 25071 loss: 0.4887 +2023/05/31 17:52:51 - mmengine - INFO - Epoch(train) [1][1100/5758] lr: 4.0000e-03 eta: 1 day, 9:29:42 time: 0.9755 data_time: 0.0021 memory: 25071 loss: 0.4961 +2023/05/31 17:54:30 - mmengine - INFO - Epoch(train) [1][1200/5758] lr: 4.0000e-03 eta: 1 day, 9:17:48 time: 0.9678 data_time: 0.0019 memory: 25071 loss: 0.4578 +2023/05/31 17:56:11 - mmengine - INFO - Epoch(train) [1][1300/5758] lr: 4.0000e-03 eta: 1 day, 9:09:25 time: 1.0431 data_time: 0.0017 memory: 25071 loss: 0.4498 +2023/05/31 17:57:50 - mmengine - INFO - Epoch(train) [1][1400/5758] lr: 4.0000e-03 eta: 1 day, 9:00:00 time: 0.9596 data_time: 0.0015 memory: 25071 loss: 0.4368 +2023/05/31 17:59:31 - mmengine - INFO - Epoch(train) [1][1500/5758] lr: 4.0000e-03 eta: 1 day, 8:53:57 time: 0.9089 data_time: 0.0018 memory: 25071 loss: 0.5320 +2023/05/31 18:01:11 - mmengine - INFO - Epoch(train) [1][1600/5758] lr: 4.0000e-03 eta: 1 day, 8:47:28 time: 0.9892 data_time: 0.0015 memory: 25071 loss: 0.4219 +2023/05/31 18:02:49 - mmengine - INFO - Epoch(train) [1][1700/5758] lr: 4.0000e-03 eta: 1 day, 8:38:27 time: 0.9990 data_time: 0.0017 memory: 25071 loss: 0.4142 +2023/05/31 18:04:24 - mmengine - INFO - Epoch(train) [1][1800/5758] lr: 4.0000e-03 eta: 1 day, 8:27:46 time: 1.1330 data_time: 0.0014 memory: 25071 loss: 0.3887 +2023/05/31 18:06:00 - mmengine - INFO - Epoch(train) [1][1900/5758] lr: 4.0000e-03 eta: 1 day, 8:19:25 time: 0.9008 data_time: 0.0014 memory: 25071 loss: 0.3969 +2023/05/31 18:07:34 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_20230531_173252 +2023/05/31 18:07:34 - mmengine - INFO - Epoch(train) [1][2000/5758] lr: 4.0000e-03 eta: 1 day, 8:09:46 time: 1.0660 data_time: 0.0014 memory: 25071 loss: 0.3817 +2023/05/31 18:09:11 - mmengine - INFO - Epoch(train) [1][2100/5758] lr: 4.0000e-03 eta: 1 day, 8:02:39 time: 1.0678 data_time: 0.0016 memory: 25071 loss: 0.4004 +2023/05/31 18:10:42 - mmengine - INFO - Epoch(train) [1][2200/5758] lr: 4.0000e-03 eta: 1 day, 7:52:14 time: 0.8623 data_time: 0.0015 memory: 25071 loss: 0.3590 +2023/05/31 18:12:12 - mmengine - INFO - Epoch(train) [1][2300/5758] lr: 4.0000e-03 eta: 1 day, 7:40:20 time: 0.8956 data_time: 0.0015 memory: 25071 loss: 0.2965 +2023/05/31 18:13:41 - mmengine - INFO - Epoch(train) [1][2400/5758] lr: 4.0000e-03 eta: 1 day, 7:29:22 time: 0.9355 data_time: 0.0015 memory: 25071 loss: 0.3110 +2023/05/31 18:15:08 - mmengine - INFO - Epoch(train) [1][2500/5758] lr: 4.0000e-03 eta: 1 day, 7:17:56 time: 0.8769 data_time: 0.0016 memory: 25071 loss: 0.2893 +2023/05/31 18:16:33 - mmengine - INFO - Epoch(train) [1][2600/5758] lr: 4.0000e-03 eta: 1 day, 7:05:08 time: 0.8800 data_time: 0.0020 memory: 25071 loss: 0.2850 +2023/05/31 18:18:04 - mmengine - INFO - Epoch(train) [1][2700/5758] lr: 4.0000e-03 eta: 1 day, 6:57:39 time: 0.9100 data_time: 0.0014 memory: 25071 loss: 0.2896 +2023/05/31 18:19:42 - mmengine - INFO - Epoch(train) [1][2800/5758] lr: 4.0000e-03 eta: 1 day, 6:55:00 time: 0.9779 data_time: 0.0015 memory: 25071 loss: 0.3225 +2023/05/31 18:21:16 - mmengine - INFO - Epoch(train) [1][2900/5758] lr: 4.0000e-03 eta: 1 day, 6:50:30 time: 0.9152 data_time: 0.0013 memory: 25071 loss: 0.2772 +2023/05/31 18:22:52 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_20230531_173252 +2023/05/31 18:22:52 - mmengine - INFO - Epoch(train) [1][3000/5758] lr: 4.0000e-03 eta: 1 day, 6:47:02 time: 0.9313 data_time: 0.0020 memory: 25071 loss: 0.2216 +2023/05/31 18:24:27 - mmengine - INFO - Epoch(train) [1][3100/5758] lr: 4.0000e-03 eta: 1 day, 6:42:54 time: 0.9810 data_time: 0.0017 memory: 25071 loss: 0.2503 +2023/05/31 18:26:03 - mmengine - INFO - Epoch(train) [1][3200/5758] lr: 4.0000e-03 eta: 1 day, 6:39:59 time: 0.9946 data_time: 0.0014 memory: 25071 loss: 0.2145 +2023/05/31 18:27:36 - mmengine - INFO - Epoch(train) [1][3300/5758] lr: 4.0000e-03 eta: 1 day, 6:35:12 time: 0.9269 data_time: 0.0025 memory: 25071 loss: 0.2393 +2023/05/31 18:29:08 - mmengine - INFO - Epoch(train) [1][3400/5758] lr: 4.0000e-03 eta: 1 day, 6:30:04 time: 0.9523 data_time: 0.0017 memory: 25071 loss: 0.2123 +2023/05/31 18:30:38 - mmengine - INFO - Epoch(train) [1][3500/5758] lr: 4.0000e-03 eta: 1 day, 6:23:36 time: 0.9016 data_time: 0.0015 memory: 25071 loss: 0.2155 +2023/05/31 18:32:07 - mmengine - INFO - Epoch(train) [1][3600/5758] lr: 4.0000e-03 eta: 1 day, 6:17:30 time: 1.0201 data_time: 0.0224 memory: 25071 loss: 0.2121 +2023/05/31 18:33:37 - mmengine - INFO - Epoch(train) [1][3700/5758] lr: 4.0000e-03 eta: 1 day, 6:12:02 time: 0.8195 data_time: 0.0015 memory: 25071 loss: 0.1702 +2023/05/31 18:35:09 - mmengine - INFO - Epoch(train) [1][3800/5758] lr: 4.0000e-03 eta: 1 day, 6:07:50 time: 1.0012 data_time: 0.0014 memory: 25071 loss: 0.1883 +2023/05/31 18:36:37 - mmengine - INFO - Epoch(train) [1][3900/5758] lr: 4.0000e-03 eta: 1 day, 6:01:21 time: 0.8362 data_time: 0.0015 memory: 25071 loss: 0.1630 +2023/05/31 18:38:07 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_20230531_173252 +2023/05/31 18:38:07 - mmengine - INFO - Epoch(train) [1][4000/5758] lr: 4.0000e-03 eta: 1 day, 5:56:27 time: 0.8800 data_time: 0.0015 memory: 25071 loss: 0.1774 +2023/05/31 18:39:37 - mmengine - INFO - Epoch(train) [1][4100/5758] lr: 4.0000e-03 eta: 1 day, 5:51:57 time: 0.9191 data_time: 0.0016 memory: 25071 loss: 0.1531 +2023/05/31 18:41:08 - mmengine - INFO - Epoch(train) [1][4200/5758] lr: 4.0000e-03 eta: 1 day, 5:47:53 time: 0.9319 data_time: 0.0019 memory: 25071 loss: 0.1537 +2023/05/31 18:42:39 - mmengine - INFO - Epoch(train) [1][4300/5758] lr: 4.0000e-03 eta: 1 day, 5:43:35 time: 0.8985 data_time: 0.0016 memory: 25071 loss: 0.1297 +2023/05/31 18:44:09 - mmengine - INFO - Epoch(train) [1][4400/5758] lr: 4.0000e-03 eta: 1 day, 5:39:24 time: 0.8515 data_time: 0.0014 memory: 25071 loss: 0.1656 +2023/05/31 18:45:39 - mmengine - INFO - Epoch(train) [1][4500/5758] lr: 4.0000e-03 eta: 1 day, 5:35:04 time: 0.9596 data_time: 0.0013 memory: 25071 loss: 0.1451 +2023/05/31 18:47:10 - mmengine - INFO - Epoch(train) [1][4600/5758] lr: 4.0000e-03 eta: 1 day, 5:31:21 time: 0.9197 data_time: 0.0014 memory: 25071 loss: 0.1515 +2023/05/31 18:48:51 - mmengine - INFO - Epoch(train) [1][4700/5758] lr: 4.0000e-03 eta: 1 day, 5:31:47 time: 0.9126 data_time: 0.0022 memory: 25071 loss: 0.1102 +2023/05/31 18:50:38 - mmengine - INFO - Epoch(train) [1][4800/5758] lr: 4.0000e-03 eta: 1 day, 5:34:02 time: 0.9485 data_time: 0.0018 memory: 25071 loss: 0.1259 +2023/05/31 18:52:23 - mmengine - INFO - Epoch(train) [1][4900/5758] lr: 4.0000e-03 eta: 1 day, 5:35:44 time: 1.0158 data_time: 0.0012 memory: 25071 loss: 0.0999 +2023/05/31 18:54:05 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_20230531_173252 +2023/05/31 18:54:05 - mmengine - INFO - Epoch(train) [1][5000/5758] lr: 4.0000e-03 eta: 1 day, 5:36:10 time: 0.9244 data_time: 0.0014 memory: 25071 loss: 0.1139 +2023/05/31 18:55:47 - mmengine - INFO - Epoch(train) [1][5100/5758] lr: 4.0000e-03 eta: 1 day, 5:36:21 time: 0.9541 data_time: 0.0016 memory: 25071 loss: 0.1296 +2023/05/31 18:57:23 - mmengine - INFO - Epoch(train) [1][5200/5758] lr: 4.0000e-03 eta: 1 day, 5:34:25 time: 0.8874 data_time: 0.0022 memory: 25071 loss: 0.0850 +2023/05/31 18:59:01 - mmengine - INFO - Epoch(train) [1][5300/5758] lr: 4.0000e-03 eta: 1 day, 5:33:19 time: 0.9663 data_time: 0.0015 memory: 25071 loss: 0.0990 +2023/05/31 19:00:38 - mmengine - INFO - Epoch(train) [1][5400/5758] lr: 4.0000e-03 eta: 1 day, 5:31:44 time: 0.9788 data_time: 0.0020 memory: 25071 loss: 0.0649 +2023/05/31 19:02:07 - mmengine - INFO - Epoch(train) [1][5500/5758] lr: 4.0000e-03 eta: 1 day, 5:27:34 time: 0.8545 data_time: 0.0015 memory: 25071 loss: 0.0742 +2023/05/31 19:03:37 - mmengine - INFO - Epoch(train) [1][5600/5758] lr: 4.0000e-03 eta: 1 day, 5:23:51 time: 0.9005 data_time: 0.0014 memory: 25071 loss: 0.0684 +2023/05/31 19:05:11 - mmengine - INFO - Epoch(train) [1][5700/5758] lr: 4.0000e-03 eta: 1 day, 5:21:18 time: 1.0014 data_time: 0.0017 memory: 25071 loss: 0.0647 +2023/05/31 19:06:07 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_20230531_173252 +2023/05/31 19:06:07 - mmengine - INFO - Saving checkpoint at 1 epochs +2023/05/31 19:06:27 - mmengine - INFO - Epoch(val) [1][16/16] accuracy/top1: 71.3339 single-label/precision_classwise: [100.0, 0.0] single-label/recall_classwise: [71.33387756347656, 0.0] single-label/f1-score_classwise: [83.26885223388672, 0.0] data_time: 0.2987 time: 0.8163 +2023/05/31 19:08:22 - mmengine - INFO - Epoch(train) [2][ 100/5758] lr: 3.9754e-03 eta: 1 day, 5:24:16 time: 0.9840 data_time: 0.0016 memory: 25074 loss: 0.4454 +2023/05/31 19:10:12 - mmengine - INFO - Epoch(train) [2][ 200/5758] lr: 3.9754e-03 eta: 1 day, 5:26:30 time: 0.9971 data_time: 0.0021 memory: 25074 loss: 0.4071 +2023/05/31 19:11:01 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_20230531_173252 +2023/05/31 19:11:59 - mmengine - INFO - Epoch(train) [2][ 300/5758] lr: 3.9754e-03 eta: 1 day, 5:27:59 time: 1.0498 data_time: 0.0013 memory: 25074 loss: 0.2401 +2023/05/31 19:13:32 - mmengine - INFO - Epoch(train) [2][ 400/5758] lr: 3.9754e-03 eta: 1 day, 5:25:06 time: 0.9421 data_time: 0.0016 memory: 25074 loss: 0.2418 +2023/05/31 19:15:07 - mmengine - INFO - Epoch(train) [2][ 500/5758] lr: 3.9754e-03 eta: 1 day, 5:22:56 time: 0.8745 data_time: 0.0018 memory: 25074 loss: 0.0975 +2023/05/31 19:16:41 - mmengine - INFO - Epoch(train) [2][ 600/5758] lr: 3.9754e-03 eta: 1 day, 5:20:23 time: 0.9066 data_time: 0.0022 memory: 25074 loss: 0.0629 +2023/05/31 19:18:13 - mmengine - INFO - Epoch(train) [2][ 700/5758] lr: 3.9754e-03 eta: 1 day, 5:17:22 time: 0.9185 data_time: 0.0019 memory: 25074 loss: 0.0615 +2023/05/31 19:19:46 - mmengine - INFO - Epoch(train) [2][ 800/5758] lr: 3.9754e-03 eta: 1 day, 5:14:27 time: 0.8759 data_time: 0.0015 memory: 25074 loss: 0.0705 +2023/05/31 19:21:18 - mmengine - INFO - Epoch(train) [2][ 900/5758] lr: 3.9754e-03 eta: 1 day, 5:11:31 time: 0.9211 data_time: 0.0016 memory: 25074 loss: 0.0725 +2023/05/31 19:22:48 - mmengine - INFO - Epoch(train) [2][1000/5758] lr: 3.9754e-03 eta: 1 day, 5:08:08 time: 1.0462 data_time: 0.0015 memory: 25074 loss: 0.0804 +2023/05/31 19:24:19 - mmengine - INFO - Epoch(train) [2][1100/5758] lr: 3.9754e-03 eta: 1 day, 5:05:02 time: 0.8867 data_time: 0.0019 memory: 25074 loss: 0.0675 +2023/05/31 19:25:52 - mmengine - INFO - Epoch(train) [2][1200/5758] lr: 3.9754e-03 eta: 1 day, 5:02:22 time: 0.8682 data_time: 0.0016 memory: 25074 loss: 0.0524 +2023/05/31 19:26:33 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_20230531_173252 +2023/05/31 19:27:26 - mmengine - INFO - Epoch(train) [2][1300/5758] lr: 3.9754e-03 eta: 1 day, 5:00:14 time: 0.9431 data_time: 0.0019 memory: 25074 loss: 0.0816 +2023/05/31 19:29:01 - mmengine - INFO - Epoch(train) [2][1400/5758] lr: 3.9754e-03 eta: 1 day, 4:58:11 time: 0.9383 data_time: 0.0021 memory: 25074 loss: 0.0454 +2023/05/31 19:30:34 - mmengine - INFO - Epoch(train) [2][1500/5758] lr: 3.9754e-03 eta: 1 day, 4:55:48 time: 0.8265 data_time: 0.0018 memory: 25074 loss: 0.0543 +2023/05/31 19:32:06 - mmengine - INFO - Epoch(train) [2][1600/5758] lr: 3.9754e-03 eta: 1 day, 4:52:52 time: 0.9104 data_time: 0.0023 memory: 25074 loss: 0.0614 +2023/05/31 19:33:35 - mmengine - INFO - Epoch(train) [2][1700/5758] lr: 3.9754e-03 eta: 1 day, 4:49:38 time: 0.8602 data_time: 0.0013 memory: 25074 loss: 0.0497 +2023/05/31 19:35:06 - mmengine - INFO - Epoch(train) [2][1800/5758] lr: 3.9754e-03 eta: 1 day, 4:46:38 time: 0.9751 data_time: 0.0024 memory: 25074 loss: 0.0506 +2023/05/31 19:36:32 - mmengine - INFO - Epoch(train) [2][1900/5758] lr: 3.9754e-03 eta: 1 day, 4:42:36 time: 0.7512 data_time: 0.0018 memory: 25074 loss: 0.5002 +2023/05/31 19:37:56 - mmengine - INFO - Epoch(train) [2][2000/5758] lr: 3.9754e-03 eta: 1 day, 4:38:23 time: 0.8483 data_time: 0.0014 memory: 25074 loss: 0.4878 +2023/05/31 19:39:22 - mmengine - INFO - Epoch(train) [2][2100/5758] lr: 3.9754e-03 eta: 1 day, 4:34:25 time: 0.8354 data_time: 0.0019 memory: 25074 loss: 0.4480 +2023/05/31 19:40:50 - mmengine - INFO - Epoch(train) [2][2200/5758] lr: 3.9754e-03 eta: 1 day, 4:30:57 time: 0.8249 data_time: 0.0019 memory: 25074 loss: 0.2145 +2023/05/31 19:41:26 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_20230531_173252 +2023/05/31 19:42:15 - mmengine - INFO - Epoch(train) [2][2300/5758] lr: 3.9754e-03 eta: 1 day, 4:27:07 time: 0.8128 data_time: 0.0020 memory: 25074 loss: 0.0894 +2023/05/31 19:43:41 - mmengine - INFO - Epoch(train) [2][2400/5758] lr: 3.9754e-03 eta: 1 day, 4:23:20 time: 0.8492 data_time: 0.0019 memory: 25074 loss: 0.0481 +2023/05/31 19:45:05 - mmengine - INFO - Epoch(train) [2][2500/5758] lr: 3.9754e-03 eta: 1 day, 4:19:11 time: 0.8987 data_time: 0.0022 memory: 25074 loss: 0.0582 +2023/05/31 19:46:24 - mmengine - INFO - Epoch(train) [2][2600/5758] lr: 3.9754e-03 eta: 1 day, 4:14:15 time: 0.7931 data_time: 0.0016 memory: 25074 loss: 0.0499 +2023/05/31 19:47:46 - mmengine - INFO - Epoch(train) [2][2700/5758] lr: 3.9754e-03 eta: 1 day, 4:09:45 time: 0.8968 data_time: 0.0023 memory: 25074 loss: 0.0692 +2023/05/31 19:49:07 - mmengine - INFO - Epoch(train) [2][2800/5758] lr: 3.9754e-03 eta: 1 day, 4:05:23 time: 0.8449 data_time: 0.0022 memory: 25074 loss: 0.0493 +2023/05/31 19:50:26 - mmengine - INFO - Epoch(train) [2][2900/5758] lr: 3.9754e-03 eta: 1 day, 4:00:33 time: 0.7910 data_time: 0.0027 memory: 25074 loss: 0.0368 +2023/05/31 19:51:46 - mmengine - INFO - Epoch(train) [2][3000/5758] lr: 3.9754e-03 eta: 1 day, 3:55:57 time: 0.7935 data_time: 0.0017 memory: 25074 loss: 0.0463 +2023/05/31 19:53:07 - mmengine - INFO - Epoch(train) [2][3100/5758] lr: 3.9754e-03 eta: 1 day, 3:51:37 time: 0.8450 data_time: 0.0017 memory: 25074 loss: 0.0452 +2023/05/31 19:54:28 - mmengine - INFO - Epoch(train) [2][3200/5758] lr: 3.9754e-03 eta: 1 day, 3:47:31 time: 0.8160 data_time: 0.0022 memory: 25074 loss: 0.0492 +2023/05/31 19:55:02 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_20230531_173252 +2023/05/31 19:55:51 - mmengine - INFO - Epoch(train) [2][3300/5758] lr: 3.9754e-03 eta: 1 day, 3:43:39 time: 0.8127 data_time: 0.0021 memory: 25074 loss: 0.0413 +2023/05/31 19:57:13 - mmengine - INFO - Epoch(train) [2][3400/5758] lr: 3.9754e-03 eta: 1 day, 3:39:43 time: 0.8368 data_time: 0.0014 memory: 25074 loss: 0.0402 +2023/05/31 19:58:36 - mmengine - INFO - Epoch(train) [2][3500/5758] lr: 3.9754e-03 eta: 1 day, 3:36:04 time: 0.8621 data_time: 0.0015 memory: 25074 loss: 0.0767 +2023/05/31 20:00:00 - mmengine - INFO - Epoch(train) [2][3600/5758] lr: 3.9754e-03 eta: 1 day, 3:32:46 time: 0.7853 data_time: 0.0017 memory: 25074 loss: 0.0275 +2023/05/31 20:01:23 - mmengine - INFO - Epoch(train) [2][3700/5758] lr: 3.9754e-03 eta: 1 day, 3:29:15 time: 0.8450 data_time: 0.0018 memory: 25074 loss: 0.0290 +2023/05/31 20:02:48 - mmengine - INFO - Epoch(train) [2][3800/5758] lr: 3.9754e-03 eta: 1 day, 3:25:58 time: 0.8254 data_time: 0.0016 memory: 25074 loss: 0.0305 +2023/05/31 20:04:11 - mmengine - INFO - Epoch(train) [2][3900/5758] lr: 3.9754e-03 eta: 1 day, 3:22:38 time: 0.9337 data_time: 0.0017 memory: 25074 loss: 0.0382 +2023/05/31 20:05:35 - mmengine - INFO - Epoch(train) [2][4000/5758] lr: 3.9754e-03 eta: 1 day, 3:19:16 time: 0.8491 data_time: 0.0016 memory: 25074 loss: 0.0606 +2023/05/31 20:07:02 - mmengine - INFO - Epoch(train) [2][4100/5758] lr: 3.9754e-03 eta: 1 day, 3:16:39 time: 0.9263 data_time: 0.0018 memory: 25074 loss: 0.0396 +2023/05/31 20:08:32 - mmengine - INFO - Epoch(train) [2][4200/5758] lr: 3.9754e-03 eta: 1 day, 3:14:34 time: 0.9409 data_time: 0.0016 memory: 25074 loss: 0.0302 +2023/05/31 20:09:10 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_20230531_173252 +2023/05/31 20:10:02 - mmengine - INFO - Epoch(train) [2][4300/5758] lr: 3.9754e-03 eta: 1 day, 3:12:21 time: 0.8561 data_time: 0.0016 memory: 25074 loss: 0.0355 +2023/05/31 20:11:30 - mmengine - INFO - Epoch(train) [2][4400/5758] lr: 3.9754e-03 eta: 1 day, 3:09:51 time: 0.8507 data_time: 0.0020 memory: 25074 loss: 0.0249 +2023/05/31 20:12:59 - mmengine - INFO - Epoch(train) [2][4500/5758] lr: 3.9754e-03 eta: 1 day, 3:07:44 time: 0.9130 data_time: 0.0017 memory: 25074 loss: 0.0247 +2023/05/31 20:14:28 - mmengine - INFO - Epoch(train) [2][4600/5758] lr: 3.9754e-03 eta: 1 day, 3:05:28 time: 0.8855 data_time: 0.0013 memory: 25074 loss: 0.4744 +2023/05/31 20:15:55 - mmengine - INFO - Epoch(train) [2][4700/5758] lr: 3.9754e-03 eta: 1 day, 3:02:55 time: 0.8346 data_time: 0.0012 memory: 25074 loss: 0.2807 +2023/05/31 20:17:22 - mmengine - INFO - Epoch(train) [2][4800/5758] lr: 3.9754e-03 eta: 1 day, 3:00:21 time: 0.9134 data_time: 0.0013 memory: 25074 loss: 0.5717 +2023/05/31 20:18:53 - mmengine - INFO - Epoch(train) [2][4900/5758] lr: 3.9754e-03 eta: 1 day, 2:58:26 time: 0.8857 data_time: 0.0015 memory: 25074 loss: 0.5091 +2023/05/31 20:20:24 - mmengine - INFO - Epoch(train) [2][5000/5758] lr: 3.9754e-03 eta: 1 day, 2:56:31 time: 0.9126 data_time: 0.0020 memory: 25074 loss: 0.3839 +2023/05/31 20:21:53 - mmengine - INFO - Epoch(train) [2][5100/5758] lr: 3.9754e-03 eta: 1 day, 2:54:27 time: 0.9338 data_time: 0.0021 memory: 25074 loss: 0.4082 +2023/05/31 20:23:19 - mmengine - INFO - Epoch(train) [2][5200/5758] lr: 3.9754e-03 eta: 1 day, 2:51:43 time: 0.7848 data_time: 0.0013 memory: 25074 loss: 0.3371 +2023/05/31 20:23:56 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_20230531_173252 +2023/05/31 20:24:47 - mmengine - INFO - Epoch(train) [2][5300/5758] lr: 3.9754e-03 eta: 1 day, 2:49:29 time: 0.8530 data_time: 0.0013 memory: 25074 loss: 0.3264 +2023/05/31 20:26:18 - mmengine - INFO - Epoch(train) [2][5400/5758] lr: 3.9754e-03 eta: 1 day, 2:47:37 time: 0.9370 data_time: 0.0013 memory: 25074 loss: 0.1496 +2023/05/31 20:27:48 - mmengine - INFO - Epoch(train) [2][5500/5758] lr: 3.9754e-03 eta: 1 day, 2:45:44 time: 0.8929 data_time: 0.0013 memory: 25074 loss: 0.0755 +2023/05/31 20:29:16 - mmengine - INFO - Epoch(train) [2][5600/5758] lr: 3.9754e-03 eta: 1 day, 2:43:21 time: 0.8584 data_time: 0.0022 memory: 25074 loss: 0.7301 +2023/05/31 20:30:45 - mmengine - INFO - Epoch(train) [2][5700/5758] lr: 3.9754e-03 eta: 1 day, 2:41:22 time: 0.9257 data_time: 0.0014 memory: 25074 loss: 0.6506 +2023/05/31 20:31:36 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_20230531_173252 +2023/05/31 20:31:36 - mmengine - INFO - Saving checkpoint at 2 epochs +2023/05/31 20:31:55 - mmengine - INFO - Epoch(val) [2][16/16] accuracy/top1: 100.0000 single-label/precision_classwise: [100.0, 0.0] single-label/recall_classwise: [100.0, 0.0] single-label/f1-score_classwise: [100.0, 0.0] data_time: 0.2579 time: 0.6472 +2023/05/31 20:33:33 - mmengine - INFO - Epoch(train) [3][ 100/5758] lr: 3.9024e-03 eta: 1 day, 2:39:10 time: 0.8862 data_time: 0.0020 memory: 25074 loss: 0.5309 +2023/05/31 20:35:06 - mmengine - INFO - Epoch(train) [3][ 200/5758] lr: 3.9024e-03 eta: 1 day, 2:37:42 time: 0.9476 data_time: 0.0017 memory: 25074 loss: 0.4704 +2023/05/31 20:36:34 - mmengine - INFO - Epoch(train) [3][ 300/5758] lr: 3.9024e-03 eta: 1 day, 2:35:32 time: 0.8306 data_time: 0.0023 memory: 25074 loss: 0.4299 +2023/05/31 20:38:02 - mmengine - INFO - Epoch(train) [3][ 400/5758] lr: 3.9024e-03 eta: 1 day, 2:33:13 time: 0.9029 data_time: 0.0019 memory: 25074 loss: 0.3691 +2023/05/31 20:39:17 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_20230531_173252 +2023/05/31 20:39:30 - mmengine - INFO - Epoch(train) [3][ 500/5758] lr: 3.9024e-03 eta: 1 day, 2:31:03 time: 0.8394 data_time: 0.0020 memory: 25074 loss: 0.3560 +2023/05/31 20:40:59 - mmengine - INFO - Epoch(train) [3][ 600/5758] lr: 3.9024e-03 eta: 1 day, 2:29:01 time: 0.9999 data_time: 0.0019 memory: 25074 loss: 0.3092 +2023/05/31 20:42:27 - mmengine - INFO - Epoch(train) [3][ 700/5758] lr: 3.9024e-03 eta: 1 day, 2:26:52 time: 0.8323 data_time: 0.0020 memory: 25074 loss: 0.2898 +2023/05/31 20:43:55 - mmengine - INFO - Epoch(train) [3][ 800/5758] lr: 3.9024e-03 eta: 1 day, 2:24:42 time: 0.8981 data_time: 0.0020 memory: 25074 loss: 0.2832 +2023/05/31 20:45:25 - mmengine - INFO - Epoch(train) [3][ 900/5758] lr: 3.9024e-03 eta: 1 day, 2:22:44 time: 0.8585 data_time: 0.0022 memory: 25074 loss: 0.2297 +2023/05/31 20:46:56 - mmengine - INFO - Epoch(train) [3][1000/5758] lr: 3.9024e-03 eta: 1 day, 2:21:01 time: 0.9057 data_time: 0.0031 memory: 25074 loss: 0.2236 +2023/05/31 20:48:24 - mmengine - INFO - Epoch(train) [3][1100/5758] lr: 3.9024e-03 eta: 1 day, 2:18:56 time: 0.9525 data_time: 0.0022 memory: 25074 loss: 0.2103 +2023/05/31 20:49:54 - mmengine - INFO - Epoch(train) [3][1200/5758] lr: 3.9024e-03 eta: 1 day, 2:17:00 time: 0.8641 data_time: 0.0022 memory: 25074 loss: 0.1976 +2023/05/31 20:51:20 - mmengine - INFO - Epoch(train) [3][1300/5758] lr: 3.9024e-03 eta: 1 day, 2:14:39 time: 0.8039 data_time: 0.0020 memory: 25074 loss: 0.2243 +2023/05/31 20:52:47 - mmengine - INFO - Epoch(train) [3][1400/5758] lr: 3.9024e-03 eta: 1 day, 2:12:22 time: 0.8570 data_time: 0.0018 memory: 25074 loss: 0.1269 +2023/05/31 20:54:01 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_20230531_173252 +2023/05/31 20:54:14 - mmengine - INFO - Epoch(train) [3][1500/5758] lr: 3.9024e-03 eta: 1 day, 2:10:14 time: 0.9195 data_time: 0.0023 memory: 25074 loss: 0.1365 +2023/05/31 20:55:44 - mmengine - INFO - Epoch(train) [3][1600/5758] lr: 3.9024e-03 eta: 1 day, 2:08:25 time: 0.9113 data_time: 0.0026 memory: 25074 loss: 0.1514 +2023/05/31 20:57:12 - mmengine - INFO - Epoch(train) [3][1700/5758] lr: 3.9024e-03 eta: 1 day, 2:06:16 time: 0.8294 data_time: 0.0020 memory: 25074 loss: 0.1175 +2023/05/31 20:58:40 - mmengine - INFO - Epoch(train) [3][1800/5758] lr: 3.9024e-03 eta: 1 day, 2:04:11 time: 0.9456 data_time: 0.0019 memory: 25074 loss: 0.0811 +2023/05/31 21:00:05 - mmengine - INFO - Epoch(train) [3][1900/5758] lr: 3.9024e-03 eta: 1 day, 2:01:50 time: 0.8765 data_time: 0.0015 memory: 25074 loss: 0.0680 +2023/05/31 21:01:32 - mmengine - INFO - Epoch(train) [3][2000/5758] lr: 3.9024e-03 eta: 1 day, 1:59:34 time: 0.8908 data_time: 0.0019 memory: 25074 loss: 0.0674 +2023/05/31 21:02:57 - mmengine - INFO - Epoch(train) [3][2100/5758] lr: 3.9024e-03 eta: 1 day, 1:57:08 time: 0.8105 data_time: 0.0014 memory: 25074 loss: 0.0715 +2023/05/31 21:04:23 - mmengine - INFO - Epoch(train) [3][2200/5758] lr: 3.9024e-03 eta: 1 day, 1:54:54 time: 0.8420 data_time: 0.0013 memory: 25074 loss: 0.0617 +2023/05/31 21:05:48 - mmengine - INFO - Epoch(train) [3][2300/5758] lr: 3.9024e-03 eta: 1 day, 1:52:34 time: 0.7945 data_time: 0.0021 memory: 25074 loss: 0.0454 +2023/05/31 21:07:16 - mmengine - INFO - Epoch(train) [3][2400/5758] lr: 3.9024e-03 eta: 1 day, 1:50:33 time: 0.8628 data_time: 0.0018 memory: 25074 loss: 0.0485 +2023/05/31 21:08:31 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_20230531_173252 +2023/05/31 21:08:44 - mmengine - INFO - Epoch(train) [3][2500/5758] lr: 3.9024e-03 eta: 1 day, 1:48:31 time: 0.8765 data_time: 0.0022 memory: 25074 loss: 0.0497 +2023/05/31 21:10:10 - mmengine - INFO - Epoch(train) [3][2600/5758] lr: 3.9024e-03 eta: 1 day, 1:46:18 time: 0.7989 data_time: 0.0016 memory: 25074 loss: 0.0700 +2023/05/31 21:11:35 - mmengine - INFO - Epoch(train) [3][2700/5758] lr: 3.9024e-03 eta: 1 day, 1:43:55 time: 0.8280 data_time: 0.0014 memory: 25074 loss: 0.0505 +2023/05/31 21:13:01 - mmengine - INFO - Epoch(train) [3][2800/5758] lr: 3.9024e-03 eta: 1 day, 1:41:43 time: 0.8680 data_time: 0.0016 memory: 25074 loss: 0.0507 +2023/05/31 21:14:26 - mmengine - INFO - Epoch(train) [3][2900/5758] lr: 3.9024e-03 eta: 1 day, 1:39:27 time: 0.8109 data_time: 0.0024 memory: 25074 loss: 0.0536 +2023/05/31 21:15:51 - mmengine - INFO - Epoch(train) [3][3000/5758] lr: 3.9024e-03 eta: 1 day, 1:37:09 time: 0.8751 data_time: 0.0022 memory: 25074 loss: 0.0404 +2023/05/31 21:17:15 - mmengine - INFO - Epoch(train) [3][3100/5758] lr: 3.9024e-03 eta: 1 day, 1:34:46 time: 0.8424 data_time: 0.0021 memory: 25074 loss: 0.0417 +2023/05/31 21:18:40 - mmengine - INFO - Epoch(train) [3][3200/5758] lr: 3.9024e-03 eta: 1 day, 1:32:27 time: 0.8206 data_time: 0.0014 memory: 25074 loss: 0.1048 +2023/05/31 21:20:05 - mmengine - INFO - Epoch(train) [3][3300/5758] lr: 3.9024e-03 eta: 1 day, 1:30:08 time: 0.8065 data_time: 0.0017 memory: 25074 loss: 0.5778 +2023/05/31 21:21:30 - mmengine - INFO - Epoch(train) [3][3400/5758] lr: 3.9024e-03 eta: 1 day, 1:27:57 time: 0.8386 data_time: 0.0020 memory: 25074 loss: 0.4217 +2023/05/31 21:22:43 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_20230531_173252 +2023/05/31 21:22:56 - mmengine - INFO - Epoch(train) [3][3500/5758] lr: 3.9024e-03 eta: 1 day, 1:25:51 time: 0.8775 data_time: 0.0014 memory: 25074 loss: 0.3421 +2023/05/31 21:24:26 - mmengine - INFO - Epoch(train) [3][3600/5758] lr: 3.9024e-03 eta: 1 day, 1:24:05 time: 0.8238 data_time: 0.0021 memory: 25074 loss: 0.2083 +2023/05/31 21:25:55 - mmengine - INFO - Epoch(train) [3][3700/5758] lr: 3.9024e-03 eta: 1 day, 1:22:23 time: 0.8463 data_time: 0.0019 memory: 25074 loss: 0.0917 +2023/05/31 21:27:25 - mmengine - INFO - Epoch(train) [3][3800/5758] lr: 3.9024e-03 eta: 1 day, 1:20:41 time: 0.8383 data_time: 0.0014 memory: 25074 loss: 0.0916 +2023/05/31 21:28:54 - mmengine - INFO - Epoch(train) [3][3900/5758] lr: 3.9024e-03 eta: 1 day, 1:18:51 time: 0.8325 data_time: 0.0015 memory: 25074 loss: 0.0407 +2023/05/31 21:30:20 - mmengine - INFO - Epoch(train) [3][4000/5758] lr: 3.9024e-03 eta: 1 day, 1:16:45 time: 0.8978 data_time: 0.0016 memory: 25074 loss: 0.0456 +2023/05/31 21:31:46 - mmengine - INFO - Epoch(train) [3][4100/5758] lr: 3.9024e-03 eta: 1 day, 1:14:41 time: 0.9011 data_time: 0.0015 memory: 25074 loss: 0.0519 +2023/05/31 21:33:15 - mmengine - INFO - Epoch(train) [3][4200/5758] lr: 3.9024e-03 eta: 1 day, 1:12:55 time: 0.9069 data_time: 0.0016 memory: 25074 loss: 0.0254 +2023/05/31 21:34:42 - mmengine - INFO - Epoch(train) [3][4300/5758] lr: 3.9024e-03 eta: 1 day, 1:10:58 time: 0.8678 data_time: 0.0015 memory: 25074 loss: 0.0338 +2023/05/31 21:36:10 - mmengine - INFO - Epoch(train) [3][4400/5758] lr: 3.9024e-03 eta: 1 day, 1:09:04 time: 0.8704 data_time: 0.0015 memory: 25074 loss: 0.0263 +2023/05/31 21:37:26 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_20230531_173252 +2023/05/31 21:37:39 - mmengine - INFO - Epoch(train) [3][4500/5758] lr: 3.9024e-03 eta: 1 day, 1:07:17 time: 0.8488 data_time: 0.0015 memory: 25074 loss: 0.0240 +2023/05/31 21:39:05 - mmengine - INFO - Epoch(train) [3][4600/5758] lr: 3.9024e-03 eta: 1 day, 1:05:18 time: 0.8706 data_time: 0.0015 memory: 25074 loss: 0.0922 +2023/05/31 21:40:34 - mmengine - INFO - Epoch(train) [3][4700/5758] lr: 3.9024e-03 eta: 1 day, 1:03:30 time: 0.9396 data_time: 0.0015 memory: 25074 loss: 0.0390 +2023/05/31 21:42:02 - mmengine - INFO - Epoch(train) [3][4800/5758] lr: 3.9024e-03 eta: 1 day, 1:01:44 time: 0.9234 data_time: 0.0013 memory: 25074 loss: 0.0330 +2023/05/31 21:43:33 - mmengine - INFO - Epoch(train) [3][4900/5758] lr: 3.9024e-03 eta: 1 day, 1:00:09 time: 0.9162 data_time: 0.0013 memory: 25074 loss: 0.0281 +2023/05/31 21:44:56 - mmengine - INFO - Epoch(train) [3][5000/5758] lr: 3.9024e-03 eta: 1 day, 0:57:52 time: 0.8040 data_time: 0.0017 memory: 25074 loss: 0.0249 +2023/05/31 21:46:20 - mmengine - INFO - Epoch(train) [3][5100/5758] lr: 3.9024e-03 eta: 1 day, 0:55:34 time: 0.8070 data_time: 0.0013 memory: 25074 loss: 0.0303 +2023/05/31 21:47:47 - mmengine - INFO - Epoch(train) [3][5200/5758] lr: 3.9024e-03 eta: 1 day, 0:53:41 time: 0.9348 data_time: 0.0015 memory: 25074 loss: 0.3830 +2023/05/31 21:49:14 - mmengine - INFO - Epoch(train) [3][5300/5758] lr: 3.9024e-03 eta: 1 day, 0:51:46 time: 0.8739 data_time: 0.0014 memory: 25074 loss: 0.1110 +2023/05/31 21:50:39 - mmengine - INFO - Epoch(train) [3][5400/5758] lr: 3.9024e-03 eta: 1 day, 0:49:42 time: 0.8635 data_time: 0.0015 memory: 25074 loss: 0.1760 +2023/05/31 21:51:55 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_20230531_173252 +2023/05/31 21:52:08 - mmengine - INFO - Epoch(train) [3][5500/5758] lr: 3.9024e-03 eta: 1 day, 0:47:58 time: 0.8102 data_time: 0.0013 memory: 25074 loss: 0.0392 +2023/05/31 21:53:38 - mmengine - INFO - Epoch(train) [3][5600/5758] lr: 3.9024e-03 eta: 1 day, 0:46:23 time: 0.9931 data_time: 0.0015 memory: 25074 loss: 0.0389 +2023/05/31 21:55:06 - mmengine - INFO - Epoch(train) [3][5700/5758] lr: 3.9024e-03 eta: 1 day, 0:44:33 time: 0.9239 data_time: 0.0020 memory: 25074 loss: 0.0255 +2023/05/31 21:55:54 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_20230531_173252 +2023/05/31 21:55:54 - mmengine - INFO - Saving checkpoint at 3 epochs +2023/05/31 21:56:14 - mmengine - INFO - Epoch(val) [3][16/16] accuracy/top1: 100.0000 single-label/precision_classwise: [100.0, 0.0] single-label/recall_classwise: [100.0, 0.0] single-label/f1-score_classwise: [100.0, 0.0] data_time: 0.2431 time: 0.6312 +2023/05/31 21:57:48 - mmengine - INFO - Epoch(train) [4][ 100/5758] lr: 3.7826e-03 eta: 1 day, 0:42:06 time: 0.8111 data_time: 0.0020 memory: 25074 loss: 0.0245 +2023/05/31 21:59:16 - mmengine - INFO - Epoch(train) [4][ 200/5758] lr: 3.7826e-03 eta: 1 day, 0:40:17 time: 0.8649 data_time: 0.0026 memory: 25074 loss: 0.0179 +2023/05/31 22:00:43 - mmengine - INFO - Epoch(train) [4][ 300/5758] lr: 3.7826e-03 eta: 1 day, 0:38:23 time: 0.8393 data_time: 0.0021 memory: 25074 loss: 0.0241 +2023/05/31 22:02:13 - mmengine - INFO - Epoch(train) [4][ 400/5758] lr: 3.7826e-03 eta: 1 day, 0:36:46 time: 0.8186 data_time: 0.0023 memory: 25074 loss: 0.0163 +2023/05/31 22:03:43 - mmengine - INFO - Epoch(train) [4][ 500/5758] lr: 3.7826e-03 eta: 1 day, 0:35:13 time: 0.8512 data_time: 0.0025 memory: 25074 loss: 0.0226 +2023/05/31 22:05:11 - mmengine - INFO - Epoch(train) [4][ 600/5758] lr: 3.7826e-03 eta: 1 day, 0:33:26 time: 0.8193 data_time: 0.0021 memory: 25074 loss: 0.0241 +2023/05/31 22:06:44 - mmengine - INFO - Epoch(train) [4][ 700/5758] lr: 3.7826e-03 eta: 1 day, 0:32:04 time: 0.8932 data_time: 0.0024 memory: 25074 loss: 0.0193 +2023/05/31 22:07:05 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_20230531_173252 +2023/05/31 22:08:08 - mmengine - INFO - Epoch(train) [4][ 800/5758] lr: 3.7826e-03 eta: 1 day, 0:29:56 time: 0.8333 data_time: 0.0021 memory: 25074 loss: 0.0156 +2023/05/31 22:09:29 - mmengine - INFO - Epoch(train) [4][ 900/5758] lr: 3.7826e-03 eta: 1 day, 0:27:36 time: 0.8374 data_time: 0.0022 memory: 25074 loss: 0.3952 +2023/05/31 22:10:53 - mmengine - INFO - Epoch(train) [4][1000/5758] lr: 3.7826e-03 eta: 1 day, 0:25:26 time: 0.8616 data_time: 0.0027 memory: 25074 loss: 0.3264 +2023/05/31 22:12:16 - mmengine - INFO - Epoch(train) [4][1100/5758] lr: 3.7826e-03 eta: 1 day, 0:23:16 time: 0.8392 data_time: 0.0016 memory: 25074 loss: 0.2365 +2023/05/31 22:13:38 - mmengine - INFO - Epoch(train) [4][1200/5758] lr: 3.7826e-03 eta: 1 day, 0:21:00 time: 0.8178 data_time: 0.0021 memory: 25074 loss: 0.3408 +2023/05/31 22:15:05 - mmengine - INFO - Epoch(train) [4][1300/5758] lr: 3.7826e-03 eta: 1 day, 0:19:10 time: 0.8166 data_time: 0.0025 memory: 25074 loss: 0.1928 +2023/05/31 22:16:29 - mmengine - INFO - Epoch(train) [4][1400/5758] lr: 3.7826e-03 eta: 1 day, 0:17:05 time: 0.8161 data_time: 0.0014 memory: 25074 loss: 0.0683 +2023/05/31 22:17:53 - mmengine - INFO - Epoch(train) [4][1500/5758] lr: 3.7826e-03 eta: 1 day, 0:15:02 time: 0.8285 data_time: 0.0015 memory: 25074 loss: 0.0416 +2023/05/31 22:19:14 - mmengine - INFO - Epoch(train) [4][1600/5758] lr: 3.7826e-03 eta: 1 day, 0:12:43 time: 0.8103 data_time: 0.0014 memory: 25074 loss: 0.0294 +2023/05/31 22:20:38 - mmengine - INFO - Epoch(train) [4][1700/5758] lr: 3.7826e-03 eta: 1 day, 0:10:37 time: 0.8247 data_time: 0.0016 memory: 25074 loss: 0.0448 +2023/05/31 22:20:59 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_20230531_173252 +2023/05/31 22:22:00 - mmengine - INFO - Epoch(train) [4][1800/5758] lr: 3.7826e-03 eta: 1 day, 0:08:22 time: 0.8886 data_time: 0.0017 memory: 25074 loss: 0.0440 +2023/05/31 22:23:22 - mmengine - INFO - Epoch(train) [4][1900/5758] lr: 3.7826e-03 eta: 1 day, 0:06:12 time: 0.8481 data_time: 0.0019 memory: 25074 loss: 0.0296 +2023/05/31 22:24:46 - mmengine - INFO - Epoch(train) [4][2000/5758] lr: 3.7826e-03 eta: 1 day, 0:04:06 time: 0.9165 data_time: 0.0022 memory: 25074 loss: 0.0245 +2023/05/31 22:26:06 - mmengine - INFO - Epoch(train) [4][2100/5758] lr: 3.7826e-03 eta: 1 day, 0:01:46 time: 0.7630 data_time: 0.0016 memory: 25074 loss: 0.0210 +2023/05/31 22:27:28 - mmengine - INFO - Epoch(train) [4][2200/5758] lr: 3.7826e-03 eta: 23:59:38 time: 0.8513 data_time: 0.0017 memory: 25074 loss: 0.9873 +2023/05/31 22:28:53 - mmengine - INFO - Epoch(train) [4][2300/5758] lr: 3.7826e-03 eta: 23:57:39 time: 0.8507 data_time: 0.0017 memory: 25074 loss: 0.6863 +2023/05/31 22:30:15 - mmengine - INFO - Epoch(train) [4][2400/5758] lr: 3.7826e-03 eta: 23:55:32 time: 0.8141 data_time: 0.0020 memory: 25074 loss: 0.6228 +2023/05/31 22:31:39 - mmengine - INFO - Epoch(train) [4][2500/5758] lr: 3.7826e-03 eta: 23:53:29 time: 0.8711 data_time: 0.0022 memory: 25074 loss: 0.6230 +2023/05/31 22:32:59 - mmengine - INFO - Epoch(train) [4][2600/5758] lr: 3.7826e-03 eta: 23:51:10 time: 0.7790 data_time: 0.0016 memory: 25074 loss: 0.5823 +2023/05/31 22:34:20 - mmengine - INFO - Epoch(train) [4][2700/5758] lr: 3.7826e-03 eta: 23:48:57 time: 0.8447 data_time: 0.0021 memory: 25074 loss: 0.5363 +2023/05/31 22:34:41 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_20230531_173252 +2023/05/31 22:35:41 - mmengine - INFO - Epoch(train) [4][2800/5758] lr: 3.7826e-03 eta: 23:46:45 time: 0.8090 data_time: 0.0014 memory: 25074 loss: 0.5363 +2023/05/31 22:37:03 - mmengine - INFO - Epoch(train) [4][2900/5758] lr: 3.7826e-03 eta: 23:44:36 time: 0.7996 data_time: 0.0017 memory: 25074 loss: 0.5080 +2023/05/31 22:38:27 - mmengine - INFO - Epoch(train) [4][3000/5758] lr: 3.7826e-03 eta: 23:42:37 time: 0.8065 data_time: 0.0025 memory: 25074 loss: 0.4858 +2023/05/31 22:39:50 - mmengine - INFO - Epoch(train) [4][3100/5758] lr: 3.7826e-03 eta: 23:40:35 time: 0.8865 data_time: 0.0018 memory: 25074 loss: 0.4641 +2023/05/31 22:41:09 - mmengine - INFO - Epoch(train) [4][3200/5758] lr: 3.7826e-03 eta: 23:38:15 time: 0.8055 data_time: 0.0020 memory: 25074 loss: 0.4372 +2023/05/31 22:42:31 - mmengine - INFO - Epoch(train) [4][3300/5758] lr: 3.7826e-03 eta: 23:36:07 time: 0.8218 data_time: 0.0017 memory: 25074 loss: 0.4226 +2023/05/31 22:43:52 - mmengine - INFO - Epoch(train) [4][3400/5758] lr: 3.7826e-03 eta: 23:33:57 time: 0.7935 data_time: 0.0022 memory: 25074 loss: 0.4066 +2023/05/31 22:45:15 - mmengine - INFO - Epoch(train) [4][3500/5758] lr: 3.7826e-03 eta: 23:31:59 time: 0.8749 data_time: 0.0027 memory: 25074 loss: 0.4147 +2023/05/31 22:46:37 - mmengine - INFO - Epoch(train) [4][3600/5758] lr: 3.7826e-03 eta: 23:29:53 time: 0.8124 data_time: 0.0032 memory: 25074 loss: 0.3632 +2023/05/31 22:47:59 - mmengine - INFO - Epoch(train) [4][3700/5758] lr: 3.7826e-03 eta: 23:27:50 time: 0.8755 data_time: 0.0019 memory: 25074 loss: 0.3610 +2023/05/31 22:48:22 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_20230531_173252 +2023/05/31 22:49:26 - mmengine - INFO - Epoch(train) [4][3800/5758] lr: 3.7826e-03 eta: 23:26:06 time: 0.8620 data_time: 0.0017 memory: 25074 loss: 0.3225 +2023/05/31 22:50:51 - mmengine - INFO - Epoch(train) [4][3900/5758] lr: 3.7826e-03 eta: 23:24:15 time: 0.9055 data_time: 0.0018 memory: 25074 loss: 0.3151 +2023/05/31 22:52:14 - mmengine - INFO - Epoch(train) [4][4000/5758] lr: 3.7826e-03 eta: 23:22:17 time: 0.7909 data_time: 0.0019 memory: 25074 loss: 0.3306 +2023/05/31 22:53:35 - mmengine - INFO - Epoch(train) [4][4100/5758] lr: 3.7826e-03 eta: 23:20:08 time: 0.7740 data_time: 0.0014 memory: 25074 loss: 0.2978 +2023/05/31 22:54:54 - mmengine - INFO - Epoch(train) [4][4200/5758] lr: 3.7826e-03 eta: 23:17:54 time: 0.8174 data_time: 0.0018 memory: 25074 loss: 0.2693 +2023/05/31 22:56:14 - mmengine - INFO - Epoch(train) [4][4300/5758] lr: 3.7826e-03 eta: 23:15:42 time: 0.7717 data_time: 0.0016 memory: 25074 loss: 0.2740 +2023/05/31 22:57:34 - mmengine - INFO - Epoch(train) [4][4400/5758] lr: 3.7826e-03 eta: 23:13:34 time: 0.8106 data_time: 0.0017 memory: 25074 loss: 0.2659 +2023/05/31 22:58:55 - mmengine - INFO - Epoch(train) [4][4500/5758] lr: 3.7826e-03 eta: 23:11:26 time: 0.8459 data_time: 0.0019 memory: 25074 loss: 0.2477 +2023/05/31 23:00:17 - mmengine - INFO - Epoch(train) [4][4600/5758] lr: 3.7826e-03 eta: 23:09:28 time: 0.8613 data_time: 0.0021 memory: 25074 loss: 0.2430 +2023/05/31 23:01:38 - mmengine - INFO - Epoch(train) [4][4700/5758] lr: 3.7826e-03 eta: 23:07:21 time: 0.7738 data_time: 0.0015 memory: 25074 loss: 0.2166 +2023/05/31 23:02:00 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_20230531_173252 +2023/05/31 23:02:59 - mmengine - INFO - Epoch(train) [4][4800/5758] lr: 3.7826e-03 eta: 23:05:17 time: 0.8245 data_time: 0.0018 memory: 25074 loss: 0.2208 +2023/05/31 23:04:22 - mmengine - INFO - Epoch(train) [4][4900/5758] lr: 3.7826e-03 eta: 23:03:20 time: 0.8350 data_time: 0.0020 memory: 25074 loss: 0.2657 +2023/05/31 23:05:45 - mmengine - INFO - Epoch(train) [4][5000/5758] lr: 3.7826e-03 eta: 23:01:27 time: 0.9054 data_time: 0.0013 memory: 25074 loss: 0.1990 +2023/05/31 23:07:08 - mmengine - INFO - Epoch(train) [4][5100/5758] lr: 3.7826e-03 eta: 22:59:30 time: 0.8268 data_time: 0.0015 memory: 25074 loss: 0.1768 +2023/05/31 23:08:29 - mmengine - INFO - Epoch(train) [4][5200/5758] lr: 3.7826e-03 eta: 22:57:27 time: 0.8090 data_time: 0.0028 memory: 25074 loss: 0.1815 +2023/05/31 23:09:50 - mmengine - INFO - Epoch(train) [4][5300/5758] lr: 3.7826e-03 eta: 22:55:25 time: 0.7551 data_time: 0.0025 memory: 25074 loss: 0.1644 +2023/05/31 23:11:12 - mmengine - INFO - Epoch(train) [4][5400/5758] lr: 3.7826e-03 eta: 22:53:25 time: 0.9045 data_time: 0.0014 memory: 25074 loss: 0.2193 +2023/05/31 23:12:32 - mmengine - INFO - Epoch(train) [4][5500/5758] lr: 3.7826e-03 eta: 22:51:21 time: 0.8169 data_time: 0.0019 memory: 25074 loss: 0.2136 +2023/05/31 23:13:53 - mmengine - INFO - Epoch(train) [4][5600/5758] lr: 3.7826e-03 eta: 22:49:19 time: 0.9057 data_time: 0.0017 memory: 25074 loss: 0.1699 +2023/05/31 23:15:16 - mmengine - INFO - Epoch(train) [4][5700/5758] lr: 3.7826e-03 eta: 22:47:27 time: 0.8601 data_time: 0.0020 memory: 25074 loss: 0.1526 +2023/05/31 23:15:38 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_20230531_173252 +2023/05/31 23:16:03 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_20230531_173252 +2023/05/31 23:16:03 - mmengine - INFO - Saving checkpoint at 4 epochs +2023/05/31 23:16:22 - mmengine - INFO - Epoch(val) [4][16/16] accuracy/top1: 100.0000 single-label/precision_classwise: [100.0, 0.0] single-label/recall_classwise: [100.0, 0.0] single-label/f1-score_classwise: [100.0, 0.0] data_time: 0.2402 time: 0.6277 +2023/05/31 23:17:51 - mmengine - INFO - Epoch(train) [5][ 100/5758] lr: 3.6190e-03 eta: 22:44:49 time: 0.7362 data_time: 0.0028 memory: 25074 loss: 0.1410 +2023/05/31 23:19:13 - mmengine - INFO - Epoch(train) [5][ 200/5758] lr: 3.6190e-03 eta: 22:42:52 time: 0.7838 data_time: 0.0024 memory: 25074 loss: 0.1311 +2023/05/31 23:20:38 - mmengine - INFO - Epoch(train) [5][ 300/5758] lr: 3.6190e-03 eta: 22:41:08 time: 0.7731 data_time: 0.0022 memory: 25074 loss: 0.1291 +2023/05/31 23:22:02 - mmengine - INFO - Epoch(train) [5][ 400/5758] lr: 3.6190e-03 eta: 22:39:17 time: 0.8351 data_time: 0.0021 memory: 25074 loss: 0.1430 +2023/05/31 23:23:24 - mmengine - INFO - Epoch(train) [5][ 500/5758] lr: 3.6190e-03 eta: 22:37:21 time: 0.8327 data_time: 0.0020 memory: 25074 loss: 0.1098 +2023/05/31 23:24:43 - mmengine - INFO - Epoch(train) [5][ 600/5758] lr: 3.6190e-03 eta: 22:35:15 time: 0.8027 data_time: 0.0021 memory: 25074 loss: 0.1494 +2023/05/31 23:26:05 - mmengine - INFO - Epoch(train) [5][ 700/5758] lr: 3.6190e-03 eta: 22:33:18 time: 0.8583 data_time: 0.0016 memory: 25074 loss: 0.1001 +2023/05/31 23:27:27 - mmengine - INFO - Epoch(train) [5][ 800/5758] lr: 3.6190e-03 eta: 22:31:24 time: 0.8757 data_time: 0.0023 memory: 25074 loss: 0.1192 +2023/05/31 23:28:48 - mmengine - INFO - Epoch(train) [5][ 900/5758] lr: 3.6190e-03 eta: 22:29:27 time: 0.7414 data_time: 0.0020 memory: 25074 loss: 0.0992 +2023/05/31 23:29:43 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_20230531_173252 +2023/05/31 23:30:10 - mmengine - INFO - Epoch(train) [5][1000/5758] lr: 3.6190e-03 eta: 22:27:31 time: 0.8163 data_time: 0.0018 memory: 25074 loss: 0.5075 +2023/05/31 23:31:32 - mmengine - INFO - Epoch(train) [5][1100/5758] lr: 3.6190e-03 eta: 22:25:38 time: 0.7812 data_time: 0.0021 memory: 25074 loss: 0.2001 +2023/05/31 23:32:53 - mmengine - INFO - Epoch(train) [5][1200/5758] lr: 3.6190e-03 eta: 22:23:41 time: 0.9015 data_time: 0.0019 memory: 25074 loss: 0.1282 +2023/05/31 23:34:13 - mmengine - INFO - Epoch(train) [5][1300/5758] lr: 3.6190e-03 eta: 22:21:39 time: 0.7681 data_time: 0.0019 memory: 25074 loss: 0.0742 +2023/05/31 23:35:35 - mmengine - INFO - Epoch(train) [5][1400/5758] lr: 3.6190e-03 eta: 22:19:45 time: 0.8026 data_time: 0.0017 memory: 25074 loss: 0.1416 +2023/05/31 23:36:58 - mmengine - INFO - Epoch(train) [5][1500/5758] lr: 3.6190e-03 eta: 22:17:55 time: 1.0372 data_time: 0.0021 memory: 25074 loss: 0.0798 +2023/05/31 23:38:20 - mmengine - INFO - Epoch(train) [5][1600/5758] lr: 3.6190e-03 eta: 22:16:03 time: 0.8295 data_time: 0.0020 memory: 25074 loss: 0.0977 +2023/05/31 23:39:42 - mmengine - INFO - Epoch(train) [5][1700/5758] lr: 3.6190e-03 eta: 22:14:11 time: 0.8011 data_time: 0.0015 memory: 25074 loss: 0.0776 +2023/05/31 23:41:03 - mmengine - INFO - Epoch(train) [5][1800/5758] lr: 3.6190e-03 eta: 22:12:15 time: 0.7675 data_time: 0.0018 memory: 25074 loss: 0.0825 +2023/05/31 23:42:23 - mmengine - INFO - Epoch(train) [5][1900/5758] lr: 3.6190e-03 eta: 22:10:15 time: 0.8050 data_time: 0.0023 memory: 25074 loss: 0.1585 +2023/05/31 23:43:18 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_20230531_173252 +2023/05/31 23:43:44 - mmengine - INFO - Epoch(train) [5][2000/5758] lr: 3.6190e-03 eta: 22:08:19 time: 0.8166 data_time: 0.0022 memory: 25074 loss: 0.0716 +2023/05/31 23:45:07 - mmengine - INFO - Epoch(train) [5][2100/5758] lr: 3.6190e-03 eta: 22:06:33 time: 0.9307 data_time: 0.0021 memory: 25074 loss: 0.0528 +2023/05/31 23:46:27 - mmengine - INFO - Epoch(train) [5][2200/5758] lr: 3.6190e-03 eta: 22:04:33 time: 0.7789 data_time: 0.0016 memory: 25074 loss: 0.0555 +2023/05/31 23:47:48 - mmengine - INFO - Epoch(train) [5][2300/5758] lr: 3.6190e-03 eta: 22:02:40 time: 0.7973 data_time: 0.0028 memory: 25074 loss: 0.0683 +2023/05/31 23:49:10 - mmengine - INFO - Epoch(train) [5][2400/5758] lr: 3.6190e-03 eta: 22:00:48 time: 0.8169 data_time: 0.0019 memory: 25074 loss: 0.1626 +2023/05/31 23:50:32 - mmengine - INFO - Epoch(train) [5][2500/5758] lr: 3.6190e-03 eta: 21:58:58 time: 0.8373 data_time: 0.0020 memory: 25074 loss: 0.0514 +2023/05/31 23:51:52 - mmengine - INFO - Epoch(train) [5][2600/5758] lr: 3.6190e-03 eta: 21:57:00 time: 0.8783 data_time: 0.0022 memory: 25074 loss: 0.0432 +2023/05/31 23:53:10 - mmengine - INFO - Epoch(train) [5][2700/5758] lr: 3.6190e-03 eta: 21:54:58 time: 0.7888 data_time: 0.0023 memory: 25074 loss: 0.0879 +2023/05/31 23:54:31 - mmengine - INFO - Epoch(train) [5][2800/5758] lr: 3.6190e-03 eta: 21:53:05 time: 0.8403 data_time: 0.0021 memory: 25074 loss: 0.0391 +2023/05/31 23:55:54 - mmengine - INFO - Epoch(train) [5][2900/5758] lr: 3.6190e-03 eta: 21:51:17 time: 0.8360 data_time: 0.0019 memory: 25074 loss: 0.3418 +2023/05/31 23:56:49 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_20230531_173252 +2023/05/31 23:57:14 - mmengine - INFO - Epoch(train) [5][3000/5758] lr: 3.6190e-03 eta: 21:49:22 time: 0.7920 data_time: 0.0015 memory: 25074 loss: 0.0532 +2023/05/31 23:58:36 - mmengine - INFO - Epoch(train) [5][3100/5758] lr: 3.6190e-03 eta: 21:47:33 time: 0.8395 data_time: 0.0018 memory: 25074 loss: 0.0482 +2023/05/31 23:59:58 - mmengine - INFO - Epoch(train) [5][3200/5758] lr: 3.6190e-03 eta: 21:45:43 time: 0.8029 data_time: 0.0022 memory: 25074 loss: 0.5153 +2023/06/01 00:01:18 - mmengine - INFO - Epoch(train) [5][3300/5758] lr: 3.6190e-03 eta: 21:43:49 time: 0.8329 data_time: 0.0022 memory: 25074 loss: 0.2922 +2023/06/01 00:02:39 - mmengine - INFO - Epoch(train) [5][3400/5758] lr: 3.6190e-03 eta: 21:41:55 time: 0.8191 data_time: 0.0021 memory: 25074 loss: 0.1692 +2023/06/01 00:03:59 - mmengine - INFO - Epoch(train) [5][3500/5758] lr: 3.6190e-03 eta: 21:40:02 time: 0.8360 data_time: 0.0020 memory: 25074 loss: 0.0718 +2023/06/01 00:05:20 - mmengine - INFO - Epoch(train) [5][3600/5758] lr: 3.6190e-03 eta: 21:38:09 time: 0.8147 data_time: 0.0019 memory: 25074 loss: 0.0571 +2023/06/01 00:06:40 - mmengine - INFO - Epoch(train) [5][3700/5758] lr: 3.6190e-03 eta: 21:36:16 time: 0.7919 data_time: 0.0022 memory: 25074 loss: 0.0399 +2023/06/01 00:08:01 - mmengine - INFO - Epoch(train) [5][3800/5758] lr: 3.6190e-03 eta: 21:34:26 time: 0.7573 data_time: 0.0017 memory: 25074 loss: 0.0239 +2023/06/01 00:09:22 - mmengine - INFO - Epoch(train) [5][3900/5758] lr: 3.6190e-03 eta: 21:32:33 time: 0.8103 data_time: 0.0018 memory: 25074 loss: 0.0489 +2023/06/01 00:10:16 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_20230531_173252 +2023/06/01 00:10:42 - mmengine - INFO - Epoch(train) [5][4000/5758] lr: 3.6190e-03 eta: 21:30:40 time: 0.8083 data_time: 0.0020 memory: 25074 loss: 0.0225 +2023/06/01 00:12:03 - mmengine - INFO - Epoch(train) [5][4100/5758] lr: 3.6190e-03 eta: 21:28:49 time: 0.8008 data_time: 0.0024 memory: 25074 loss: 0.0307 +2023/06/01 00:13:23 - mmengine - INFO - Epoch(train) [5][4200/5758] lr: 3.6190e-03 eta: 21:26:57 time: 0.7805 data_time: 0.0019 memory: 25074 loss: 0.0404 +2023/06/01 00:14:43 - mmengine - INFO - Epoch(train) [5][4300/5758] lr: 3.6190e-03 eta: 21:25:05 time: 0.8094 data_time: 0.0015 memory: 25074 loss: 0.0219 +2023/06/01 00:16:04 - mmengine - INFO - Epoch(train) [5][4400/5758] lr: 3.6190e-03 eta: 21:23:15 time: 0.8124 data_time: 0.0025 memory: 25074 loss: 0.0218 +2023/06/01 00:17:20 - mmengine - INFO - Epoch(train) [5][4500/5758] lr: 3.6190e-03 eta: 21:21:07 time: 0.7535 data_time: 0.0016 memory: 25074 loss: 0.0240 +2023/06/01 00:18:39 - mmengine - INFO - Epoch(train) [5][4600/5758] lr: 3.6190e-03 eta: 21:19:14 time: 0.7911 data_time: 0.0022 memory: 25074 loss: 0.2776 +2023/06/01 00:20:00 - mmengine - INFO - Epoch(train) [5][4700/5758] lr: 3.6190e-03 eta: 21:17:24 time: 0.7885 data_time: 0.0015 memory: 25074 loss: 0.0282 +2023/06/01 00:21:20 - mmengine - INFO - Epoch(train) [5][4800/5758] lr: 3.6190e-03 eta: 21:15:33 time: 0.7951 data_time: 0.0021 memory: 25074 loss: 0.0301 +2023/06/01 00:22:40 - mmengine - INFO - Epoch(train) [5][4900/5758] lr: 3.6190e-03 eta: 21:13:41 time: 0.7606 data_time: 0.0017 memory: 25074 loss: 0.0214 +2023/06/01 00:23:37 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_20230531_173252 +2023/06/01 00:24:02 - mmengine - INFO - Epoch(train) [5][5000/5758] lr: 3.6190e-03 eta: 21:11:57 time: 0.8411 data_time: 0.0020 memory: 25074 loss: 0.0209 +2023/06/01 00:25:22 - mmengine - INFO - Epoch(train) [5][5100/5758] lr: 3.6190e-03 eta: 21:10:06 time: 0.7920 data_time: 0.0023 memory: 25074 loss: 0.0308 +2023/06/01 00:26:44 - mmengine - INFO - Epoch(train) [5][5200/5758] lr: 3.6190e-03 eta: 21:08:21 time: 0.8144 data_time: 0.0018 memory: 25074 loss: 0.7453 +2023/06/01 00:28:04 - mmengine - INFO - Epoch(train) [5][5300/5758] lr: 3.6190e-03 eta: 21:06:28 time: 0.7838 data_time: 0.0017 memory: 25074 loss: 0.5257 +2023/06/01 00:29:24 - mmengine - INFO - Epoch(train) [5][5400/5758] lr: 3.6190e-03 eta: 21:04:40 time: 0.7523 data_time: 0.0016 memory: 25074 loss: 0.3817 +2023/06/01 00:30:44 - mmengine - INFO - Epoch(train) [5][5500/5758] lr: 3.6190e-03 eta: 21:02:48 time: 0.7719 data_time: 0.0016 memory: 25074 loss: 0.2653 +2023/06/01 00:32:03 - mmengine - INFO - Epoch(train) [5][5600/5758] lr: 3.6190e-03 eta: 21:00:55 time: 0.8409 data_time: 0.0018 memory: 25074 loss: 0.5123 +2023/06/01 00:33:23 - mmengine - INFO - Epoch(train) [5][5700/5758] lr: 3.6190e-03 eta: 20:59:06 time: 0.7827 data_time: 0.0016 memory: 25074 loss: 0.2876 +2023/06/01 00:34:09 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_20230531_173252 +2023/06/01 00:34:09 - mmengine - INFO - Saving checkpoint at 5 epochs +2023/06/01 00:34:28 - mmengine - INFO - Epoch(val) [5][16/16] accuracy/top1: 100.0000 single-label/precision_classwise: [100.0, 0.0] single-label/recall_classwise: [100.0, 0.0] single-label/f1-score_classwise: [100.0, 0.0] data_time: 0.2348 time: 0.6202 +2023/06/01 00:35:55 - mmengine - INFO - Epoch(train) [6][ 100/5758] lr: 3.4157e-03 eta: 20:56:33 time: 0.7910 data_time: 0.0018 memory: 25074 loss: 0.1717 +2023/06/01 00:37:15 - mmengine - INFO - Epoch(train) [6][ 200/5758] lr: 3.4157e-03 eta: 20:54:42 time: 0.8032 data_time: 0.0018 memory: 25074 loss: 0.0633 +2023/06/01 00:37:23 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_20230531_173252 +2023/06/01 00:38:34 - mmengine - INFO - Epoch(train) [6][ 300/5758] lr: 3.4157e-03 eta: 20:52:53 time: 0.7925 data_time: 0.0019 memory: 25074 loss: 0.0350 +2023/06/01 00:39:55 - mmengine - INFO - Epoch(train) [6][ 400/5758] lr: 3.4157e-03 eta: 20:51:06 time: 0.8220 data_time: 0.0024 memory: 25074 loss: 0.0298 +2023/06/01 00:41:16 - mmengine - INFO - Epoch(train) [6][ 500/5758] lr: 3.4157e-03 eta: 20:49:19 time: 0.7986 data_time: 0.0015 memory: 25074 loss: 0.0448 +2023/06/01 00:42:36 - mmengine - INFO - Epoch(train) [6][ 600/5758] lr: 3.4157e-03 eta: 20:47:31 time: 0.7858 data_time: 0.0024 memory: 25074 loss: 0.0284 +2023/06/01 00:43:56 - mmengine - INFO - Epoch(train) [6][ 700/5758] lr: 3.4157e-03 eta: 20:45:44 time: 0.8088 data_time: 0.0019 memory: 25074 loss: 0.0374 +2023/06/01 00:45:17 - mmengine - INFO - Epoch(train) [6][ 800/5758] lr: 3.4157e-03 eta: 20:43:58 time: 0.7831 data_time: 0.0023 memory: 25074 loss: 0.0244 +2023/06/01 00:46:38 - mmengine - INFO - Epoch(train) [6][ 900/5758] lr: 3.4157e-03 eta: 20:42:12 time: 0.8390 data_time: 0.0027 memory: 25074 loss: 0.0272 +2023/06/01 00:47:58 - mmengine - INFO - Epoch(train) [6][1000/5758] lr: 3.4157e-03 eta: 20:40:24 time: 0.7135 data_time: 0.0023 memory: 25074 loss: 0.0431 +2023/06/01 00:49:20 - mmengine - INFO - Epoch(train) [6][1100/5758] lr: 3.4157e-03 eta: 20:38:41 time: 0.8424 data_time: 0.0020 memory: 25074 loss: 0.0196 +2023/06/01 00:50:41 - mmengine - INFO - Epoch(train) [6][1200/5758] lr: 3.4157e-03 eta: 20:36:58 time: 0.7580 data_time: 0.0021 memory: 25074 loss: 0.0258 +2023/06/01 00:50:49 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_20230531_173252 +2023/06/01 00:52:02 - mmengine - INFO - Epoch(train) [6][1300/5758] lr: 3.4157e-03 eta: 20:35:12 time: 0.8078 data_time: 0.0020 memory: 25074 loss: 0.0160 +2023/06/01 00:53:22 - mmengine - INFO - Epoch(train) [6][1400/5758] lr: 3.4157e-03 eta: 20:33:26 time: 0.8338 data_time: 0.0022 memory: 25074 loss: 0.0233 +2023/06/01 00:54:41 - mmengine - INFO - Epoch(train) [6][1500/5758] lr: 3.4157e-03 eta: 20:31:36 time: 0.7886 data_time: 0.0017 memory: 25074 loss: 0.0263 +2023/06/01 00:56:00 - mmengine - INFO - Epoch(train) [6][1600/5758] lr: 3.4157e-03 eta: 20:29:47 time: 0.7699 data_time: 0.0017 memory: 25074 loss: 0.0194 +2023/06/01 00:57:21 - mmengine - INFO - Epoch(train) [6][1700/5758] lr: 3.4157e-03 eta: 20:28:02 time: 0.7650 data_time: 0.0019 memory: 25074 loss: 0.0245 +2023/06/01 00:58:41 - mmengine - INFO - Epoch(train) [6][1800/5758] lr: 3.4157e-03 eta: 20:26:16 time: 0.7781 data_time: 0.0014 memory: 25074 loss: 0.9105 +2023/06/01 01:00:00 - mmengine - INFO - Epoch(train) [6][1900/5758] lr: 3.4157e-03 eta: 20:24:27 time: 0.7518 data_time: 0.0016 memory: 25074 loss: 0.6870 +2023/06/01 01:01:19 - mmengine - INFO - Epoch(train) [6][2000/5758] lr: 3.4157e-03 eta: 20:22:38 time: 0.8090 data_time: 0.0018 memory: 25074 loss: 0.6859 +2023/06/01 01:02:38 - mmengine - INFO - Epoch(train) [6][2100/5758] lr: 3.4157e-03 eta: 20:20:50 time: 0.8212 data_time: 0.0016 memory: 25074 loss: 0.6863 +2023/06/01 01:03:57 - mmengine - INFO - Epoch(train) [6][2200/5758] lr: 3.4157e-03 eta: 20:19:01 time: 0.7603 data_time: 0.0015 memory: 25074 loss: 0.6893 +2023/06/01 01:04:05 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_20230531_173252 +2023/06/01 01:05:18 - mmengine - INFO - Epoch(train) [6][2300/5758] lr: 3.4157e-03 eta: 20:17:19 time: 0.7956 data_time: 0.0017 memory: 25074 loss: 0.6890 +2023/06/01 01:06:38 - mmengine - INFO - Epoch(train) [6][2400/5758] lr: 3.4157e-03 eta: 20:15:33 time: 0.8149 data_time: 0.0014 memory: 25074 loss: 0.6837 +2023/06/01 01:09:00 - mmengine - INFO - Epoch(train) [6][2500/5758] lr: 3.4157e-03 eta: 20:16:33 time: 0.7687 data_time: 0.0615 memory: 25074 loss: 0.6795 +2023/06/01 01:10:12 - mmengine - INFO - Epoch(train) [6][2600/5758] lr: 3.4157e-03 eta: 20:14:27 time: 0.7203 data_time: 0.0013 memory: 25074 loss: 0.6829 +2023/06/01 01:11:30 - mmengine - INFO - Epoch(train) [6][2700/5758] lr: 3.4157e-03 eta: 20:12:36 time: 0.8305 data_time: 0.0014 memory: 25074 loss: 0.6826 +2023/06/01 01:12:46 - mmengine - INFO - Epoch(train) [6][2800/5758] lr: 3.4157e-03 eta: 20:10:39 time: 0.8051 data_time: 0.0018 memory: 25074 loss: 0.6878 +2023/06/01 01:14:01 - mmengine - INFO - Epoch(train) [6][2900/5758] lr: 3.4157e-03 eta: 20:08:40 time: 0.7747 data_time: 0.0014 memory: 25074 loss: 0.6877 +2023/06/01 01:15:17 - mmengine - INFO - Epoch(train) [6][3000/5758] lr: 3.4157e-03 eta: 20:06:45 time: 0.7355 data_time: 0.0015 memory: 25074 loss: 0.6831 +2023/06/01 01:16:33 - mmengine - INFO - Epoch(train) [6][3100/5758] lr: 3.4157e-03 eta: 20:04:50 time: 0.7252 data_time: 0.0018 memory: 25074 loss: 0.6801 +2023/06/01 01:17:50 - mmengine - INFO - Epoch(train) [6][3200/5758] lr: 3.4157e-03 eta: 20:02:58 time: 0.7361 data_time: 0.0016 memory: 25074 loss: 0.6558 +2023/06/01 01:17:57 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_20230531_173252 +2023/06/01 01:19:04 - mmengine - INFO - Epoch(train) [6][3300/5758] lr: 3.4157e-03 eta: 20:00:59 time: 0.7466 data_time: 0.0020 memory: 25074 loss: 0.6396 +2023/06/01 01:20:20 - mmengine - INFO - Epoch(train) [6][3400/5758] lr: 3.4157e-03 eta: 19:59:04 time: 0.7108 data_time: 0.0015 memory: 25074 loss: 0.5973 +2023/06/01 01:21:36 - mmengine - INFO - Epoch(train) [6][3500/5758] lr: 3.4157e-03 eta: 19:57:11 time: 0.7146 data_time: 0.0015 memory: 25074 loss: 0.5695 +2023/06/01 01:22:56 - mmengine - INFO - Epoch(train) [6][3600/5758] lr: 3.4157e-03 eta: 19:55:27 time: 0.7867 data_time: 0.0019 memory: 25074 loss: 0.5324 +2023/06/01 01:24:13 - mmengine - INFO - Epoch(train) [6][3700/5758] lr: 3.4157e-03 eta: 19:53:34 time: 0.7765 data_time: 0.0017 memory: 25074 loss: 0.5393 +2023/06/01 01:25:30 - mmengine - INFO - Epoch(train) [6][3800/5758] lr: 3.4157e-03 eta: 19:51:42 time: 0.8012 data_time: 0.0020 memory: 25074 loss: 0.4775 +2023/06/01 01:26:51 - mmengine - INFO - Epoch(train) [6][3900/5758] lr: 3.4157e-03 eta: 19:50:03 time: 0.7955 data_time: 0.0019 memory: 25074 loss: 0.4625 +2023/06/01 01:28:12 - mmengine - INFO - Epoch(train) [6][4000/5758] lr: 3.4157e-03 eta: 19:48:23 time: 0.8331 data_time: 0.0021 memory: 25074 loss: 0.4338 +2023/06/01 01:29:34 - mmengine - INFO - Epoch(train) [6][4100/5758] lr: 3.4157e-03 eta: 19:46:43 time: 0.8732 data_time: 0.0017 memory: 25074 loss: 0.3927 +2023/06/01 01:30:53 - mmengine - INFO - Epoch(train) [6][4200/5758] lr: 3.4157e-03 eta: 19:44:58 time: 0.8073 data_time: 0.0019 memory: 25074 loss: 0.3973 +2023/06/01 01:31:00 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_20230531_173252 +2023/06/01 01:32:11 - mmengine - INFO - Epoch(train) [6][4300/5758] lr: 3.4157e-03 eta: 19:43:11 time: 0.8198 data_time: 0.0019 memory: 25074 loss: 0.3450 +2023/06/01 01:33:30 - mmengine - INFO - Epoch(train) [6][4400/5758] lr: 3.4157e-03 eta: 19:41:27 time: 0.8256 data_time: 0.0017 memory: 25074 loss: 0.3450 +2023/06/01 01:34:50 - mmengine - INFO - Epoch(train) [6][4500/5758] lr: 3.4157e-03 eta: 19:39:44 time: 0.8408 data_time: 0.0016 memory: 25074 loss: 0.3008 +2023/06/01 01:36:11 - mmengine - INFO - Epoch(train) [6][4600/5758] lr: 3.4157e-03 eta: 19:38:03 time: 0.8218 data_time: 0.0018 memory: 25074 loss: 0.2861 +2023/06/01 01:37:34 - mmengine - INFO - Epoch(train) [6][4700/5758] lr: 3.4157e-03 eta: 19:36:29 time: 0.8694 data_time: 0.0021 memory: 25074 loss: 0.2748 +2023/06/01 01:38:55 - mmengine - INFO - Epoch(train) [6][4800/5758] lr: 3.4157e-03 eta: 19:34:49 time: 0.8727 data_time: 0.0017 memory: 25074 loss: 0.2403 +2023/06/01 01:40:16 - mmengine - INFO - Epoch(train) [6][4900/5758] lr: 3.4157e-03 eta: 19:33:11 time: 0.8610 data_time: 0.0024 memory: 25074 loss: 0.1954 +2023/06/01 01:41:36 - mmengine - INFO - Epoch(train) [6][5000/5758] lr: 3.4157e-03 eta: 19:31:27 time: 0.7790 data_time: 0.0020 memory: 25074 loss: 0.2252 +2023/06/01 01:42:57 - mmengine - INFO - Epoch(train) [6][5100/5758] lr: 3.4157e-03 eta: 19:29:49 time: 0.8055 data_time: 0.0020 memory: 25074 loss: 0.2168 +2023/06/01 01:44:17 - mmengine - INFO - Epoch(train) [6][5200/5758] lr: 3.4157e-03 eta: 19:28:08 time: 0.7842 data_time: 0.0019 memory: 25074 loss: 0.1787 +2023/06/01 01:44:26 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_20230531_173252 +2023/06/01 01:45:39 - mmengine - INFO - Epoch(train) [6][5300/5758] lr: 3.4157e-03 eta: 19:26:31 time: 0.8257 data_time: 0.0021 memory: 25074 loss: 0.3113 +2023/06/01 01:47:01 - mmengine - INFO - Epoch(train) [6][5400/5758] lr: 3.4157e-03 eta: 19:24:53 time: 0.7858 data_time: 0.0019 memory: 25074 loss: 0.1942 +2023/06/01 01:48:23 - mmengine - INFO - Epoch(train) [6][5500/5758] lr: 3.4157e-03 eta: 19:23:17 time: 0.8522 data_time: 0.0022 memory: 25074 loss: 0.1610 +2023/06/01 01:49:44 - mmengine - INFO - Epoch(train) [6][5600/5758] lr: 3.4157e-03 eta: 19:21:39 time: 0.8028 data_time: 0.0022 memory: 25074 loss: 0.1281 +2023/06/01 01:51:07 - mmengine - INFO - Epoch(train) [6][5700/5758] lr: 3.4157e-03 eta: 19:20:03 time: 0.8249 data_time: 0.0020 memory: 25074 loss: 0.0844 +2023/06/01 01:51:53 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_20230531_173252 +2023/06/01 01:51:53 - mmengine - INFO - Saving checkpoint at 6 epochs +2023/06/01 01:52:12 - mmengine - INFO - Epoch(val) [6][16/16] accuracy/top1: 100.0000 single-label/precision_classwise: [100.0, 0.0] single-label/recall_classwise: [100.0, 0.0] single-label/f1-score_classwise: [100.0, 0.0] data_time: 0.2262 time: 0.6205 +2023/06/01 01:53:41 - mmengine - INFO - Epoch(train) [7][ 100/5758] lr: 3.1776e-03 eta: 19:17:43 time: 0.7843 data_time: 0.0017 memory: 25074 loss: 0.0823 +2023/06/01 01:55:02 - mmengine - INFO - Epoch(train) [7][ 200/5758] lr: 3.1776e-03 eta: 19:16:07 time: 0.7684 data_time: 0.0023 memory: 25074 loss: 0.0858 +2023/06/01 01:56:25 - mmengine - INFO - Epoch(train) [7][ 300/5758] lr: 3.1776e-03 eta: 19:14:31 time: 0.8051 data_time: 0.0017 memory: 25074 loss: 0.0816 +2023/06/01 01:57:48 - mmengine - INFO - Epoch(train) [7][ 400/5758] lr: 3.1776e-03 eta: 19:12:58 time: 0.8809 data_time: 0.0019 memory: 25074 loss: 0.1065 +2023/06/01 01:58:30 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_20230531_173252 +2023/06/01 01:59:11 - mmengine - INFO - Epoch(train) [7][ 500/5758] lr: 3.1776e-03 eta: 19:11:24 time: 0.7991 data_time: 0.0022 memory: 25074 loss: 0.0550 +2023/06/01 02:00:32 - mmengine - INFO - Epoch(train) [7][ 600/5758] lr: 3.1776e-03 eta: 19:09:47 time: 0.8405 data_time: 0.0021 memory: 25074 loss: 0.0631 +2023/06/01 02:01:54 - mmengine - INFO - Epoch(train) [7][ 700/5758] lr: 3.1776e-03 eta: 19:08:11 time: 0.8097 data_time: 0.0020 memory: 25074 loss: 0.0530 +2023/06/01 02:03:17 - mmengine - INFO - Epoch(train) [7][ 800/5758] lr: 3.1776e-03 eta: 19:06:37 time: 0.8274 data_time: 0.0023 memory: 25074 loss: 0.0548 +2023/06/01 02:04:40 - mmengine - INFO - Epoch(train) [7][ 900/5758] lr: 3.1776e-03 eta: 19:05:04 time: 0.8286 data_time: 0.0019 memory: 25074 loss: 0.0599 +2023/06/01 02:06:02 - mmengine - INFO - Epoch(train) [7][1000/5758] lr: 3.1776e-03 eta: 19:03:27 time: 0.8614 data_time: 0.0021 memory: 25074 loss: 0.0369 +2023/06/01 02:07:24 - mmengine - INFO - Epoch(train) [7][1100/5758] lr: 3.1776e-03 eta: 19:01:52 time: 0.8106 data_time: 0.0020 memory: 25074 loss: 0.0259 +2023/06/01 02:08:47 - mmengine - INFO - Epoch(train) [7][1200/5758] lr: 3.1776e-03 eta: 19:00:19 time: 0.7781 data_time: 0.0016 memory: 25074 loss: 0.0549 +2023/06/01 02:10:07 - mmengine - INFO - Epoch(train) [7][1300/5758] lr: 3.1776e-03 eta: 18:58:39 time: 0.8170 data_time: 0.0018 memory: 25074 loss: 0.0450 +2023/06/01 02:11:30 - mmengine - INFO - Epoch(train) [7][1400/5758] lr: 3.1776e-03 eta: 18:57:07 time: 0.9130 data_time: 0.0028 memory: 25074 loss: 0.0345 +2023/06/01 02:12:12 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_20230531_173252 +2023/06/01 02:12:52 - mmengine - INFO - Epoch(train) [7][1500/5758] lr: 3.1776e-03 eta: 18:55:31 time: 0.8565 data_time: 0.0017 memory: 25074 loss: 0.5377 +2023/06/01 02:14:13 - mmengine - INFO - Epoch(train) [7][1600/5758] lr: 3.1776e-03 eta: 18:53:54 time: 0.8048 data_time: 0.0023 memory: 25074 loss: 0.3871 +2023/06/01 02:15:34 - mmengine - INFO - Epoch(train) [7][1700/5758] lr: 3.1776e-03 eta: 18:52:17 time: 0.8087 data_time: 0.0019 memory: 25074 loss: 0.2551 +2023/06/01 02:16:55 - mmengine - INFO - Epoch(train) [7][1800/5758] lr: 3.1776e-03 eta: 18:50:40 time: 0.7995 data_time: 0.0016 memory: 25074 loss: 0.1017 +2023/06/01 02:18:16 - mmengine - INFO - Epoch(train) [7][1900/5758] lr: 3.1776e-03 eta: 18:49:03 time: 0.7761 data_time: 0.0018 memory: 25074 loss: 0.0550 +2023/06/01 02:19:38 - mmengine - INFO - Epoch(train) [7][2000/5758] lr: 3.1776e-03 eta: 18:47:28 time: 0.8210 data_time: 0.0016 memory: 25074 loss: 0.0397 +2023/06/01 02:21:01 - mmengine - INFO - Epoch(train) [7][2100/5758] lr: 3.1776e-03 eta: 18:45:54 time: 0.8073 data_time: 0.0020 memory: 25074 loss: 0.0337 +2023/06/01 02:22:22 - mmengine - INFO - Epoch(train) [7][2200/5758] lr: 3.1776e-03 eta: 18:44:17 time: 0.8329 data_time: 0.0015 memory: 25074 loss: 0.0510 +2023/06/01 02:23:42 - mmengine - INFO - Epoch(train) [7][2300/5758] lr: 3.1776e-03 eta: 18:42:40 time: 0.8236 data_time: 0.0018 memory: 25074 loss: 0.0374 +2023/06/01 02:25:03 - mmengine - INFO - Epoch(train) [7][2400/5758] lr: 3.1776e-03 eta: 18:41:03 time: 0.7547 data_time: 0.0018 memory: 25074 loss: 0.0335 +2023/06/01 02:25:45 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_20230531_173252 +2023/06/01 02:26:25 - mmengine - INFO - Epoch(train) [7][2500/5758] lr: 3.1776e-03 eta: 18:39:27 time: 0.8111 data_time: 0.0015 memory: 25074 loss: 0.0330 +2023/06/01 02:27:48 - mmengine - INFO - Epoch(train) [7][2600/5758] lr: 3.1776e-03 eta: 18:37:55 time: 0.7962 data_time: 0.0020 memory: 25074 loss: 0.0280 +2023/06/01 02:29:09 - mmengine - INFO - Epoch(train) [7][2700/5758] lr: 3.1776e-03 eta: 18:36:19 time: 0.8027 data_time: 0.0015 memory: 25074 loss: 0.0194 +2023/06/01 02:30:30 - mmengine - INFO - Epoch(train) [7][2800/5758] lr: 3.1776e-03 eta: 18:34:43 time: 0.7862 data_time: 0.0013 memory: 25074 loss: 0.4901 +2023/06/01 02:31:51 - mmengine - INFO - Epoch(train) [7][2900/5758] lr: 3.1776e-03 eta: 18:33:07 time: 0.8663 data_time: 0.0016 memory: 25074 loss: 0.3818 +2023/06/01 02:33:12 - mmengine - INFO - Epoch(train) [7][3000/5758] lr: 3.1776e-03 eta: 18:31:31 time: 0.8492 data_time: 0.0015 memory: 25074 loss: 0.3002 +2023/06/01 02:34:33 - mmengine - INFO - Epoch(train) [7][3100/5758] lr: 3.1776e-03 eta: 18:29:54 time: 0.7984 data_time: 0.0015 memory: 25074 loss: 0.2209 +2023/06/01 02:35:55 - mmengine - INFO - Epoch(train) [7][3200/5758] lr: 3.1776e-03 eta: 18:28:20 time: 0.8250 data_time: 0.0019 memory: 25074 loss: 0.1824 +2023/06/01 02:37:16 - mmengine - INFO - Epoch(train) [7][3300/5758] lr: 3.1776e-03 eta: 18:26:45 time: 0.7949 data_time: 0.0029 memory: 25074 loss: 0.0875 +2023/06/01 02:38:39 - mmengine - INFO - Epoch(train) [7][3400/5758] lr: 3.1776e-03 eta: 18:25:13 time: 0.8398 data_time: 0.0020 memory: 25074 loss: 0.0338 +2023/06/01 02:39:21 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_20230531_173252 +2023/06/01 02:40:00 - mmengine - INFO - Epoch(train) [7][3500/5758] lr: 3.1776e-03 eta: 18:23:38 time: 0.7995 data_time: 0.0018 memory: 25074 loss: 0.0624 +2023/06/01 02:41:22 - mmengine - INFO - Epoch(train) [7][3600/5758] lr: 3.1776e-03 eta: 18:22:03 time: 0.7869 data_time: 0.0017 memory: 25074 loss: 0.0265 +2023/06/01 02:42:43 - mmengine - INFO - Epoch(train) [7][3700/5758] lr: 3.1776e-03 eta: 18:20:27 time: 0.8373 data_time: 0.0021 memory: 25074 loss: 0.0237 +2023/06/01 02:44:06 - mmengine - INFO - Epoch(train) [7][3800/5758] lr: 3.1776e-03 eta: 18:18:56 time: 0.9143 data_time: 0.0021 memory: 25074 loss: 0.0428 +2023/06/01 02:45:27 - mmengine - INFO - Epoch(train) [7][3900/5758] lr: 3.1776e-03 eta: 18:17:21 time: 0.7826 data_time: 0.0024 memory: 25074 loss: 0.0203 +2023/06/01 02:46:48 - mmengine - INFO - Epoch(train) [7][4000/5758] lr: 3.1776e-03 eta: 18:15:46 time: 0.8423 data_time: 0.0016 memory: 25074 loss: 0.0184 +2023/06/01 02:48:10 - mmengine - INFO - Epoch(train) [7][4100/5758] lr: 3.1776e-03 eta: 18:14:11 time: 0.8426 data_time: 0.0017 memory: 25074 loss: 0.0317 +2023/06/01 02:49:31 - mmengine - INFO - Epoch(train) [7][4200/5758] lr: 3.1776e-03 eta: 18:12:37 time: 0.7893 data_time: 0.0019 memory: 25074 loss: 0.0680 +2023/06/01 02:50:54 - mmengine - INFO - Epoch(train) [7][4300/5758] lr: 3.1776e-03 eta: 18:11:06 time: 0.8710 data_time: 0.0022 memory: 25074 loss: 0.0158 +2023/06/01 02:52:17 - mmengine - INFO - Epoch(train) [7][4400/5758] lr: 3.1776e-03 eta: 18:09:33 time: 0.8603 data_time: 0.0018 memory: 25074 loss: 0.0189 +2023/06/01 02:52:58 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_20230531_173252 +2023/06/01 02:53:39 - mmengine - INFO - Epoch(train) [7][4500/5758] lr: 3.1776e-03 eta: 18:08:01 time: 0.8458 data_time: 0.0016 memory: 25074 loss: 0.0184 +2023/06/01 02:55:02 - mmengine - INFO - Epoch(train) [7][4600/5758] lr: 3.1776e-03 eta: 18:06:30 time: 0.8067 data_time: 0.0017 memory: 25074 loss: 0.0191 +2023/06/01 02:56:24 - mmengine - INFO - Epoch(train) [7][4700/5758] lr: 3.1776e-03 eta: 18:04:57 time: 0.8204 data_time: 0.0021 memory: 25074 loss: 0.0132 +2023/06/01 02:57:46 - mmengine - INFO - Epoch(train) [7][4800/5758] lr: 3.1776e-03 eta: 18:03:24 time: 0.7940 data_time: 0.0016 memory: 25074 loss: 0.0251 +2023/06/01 02:59:09 - mmengine - INFO - Epoch(train) [7][4900/5758] lr: 3.1776e-03 eta: 18:01:53 time: 0.9075 data_time: 0.0016 memory: 25074 loss: 0.0212 +2023/06/01 03:00:32 - mmengine - INFO - Epoch(train) [7][5000/5758] lr: 3.1776e-03 eta: 18:00:22 time: 0.7914 data_time: 0.0014 memory: 25074 loss: 0.6647 +2023/06/01 03:01:54 - mmengine - INFO - Epoch(train) [7][5100/5758] lr: 3.1776e-03 eta: 17:58:48 time: 0.8294 data_time: 0.0015 memory: 25074 loss: 0.6365 +2023/06/01 03:03:15 - mmengine - INFO - Epoch(train) [7][5200/5758] lr: 3.1776e-03 eta: 17:57:13 time: 0.7879 data_time: 0.0016 memory: 25074 loss: 0.6290 +2023/06/01 03:04:37 - mmengine - INFO - Epoch(train) [7][5300/5758] lr: 3.1776e-03 eta: 17:55:41 time: 0.8508 data_time: 0.0018 memory: 25074 loss: 0.6199 +2023/06/01 03:05:58 - mmengine - INFO - Epoch(train) [7][5400/5758] lr: 3.1776e-03 eta: 17:54:07 time: 0.8093 data_time: 0.0017 memory: 25074 loss: 0.6043 +2023/06/01 03:06:40 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_20230531_173252 +2023/06/01 03:07:19 - mmengine - INFO - Epoch(train) [7][5500/5758] lr: 3.1776e-03 eta: 17:52:32 time: 0.7967 data_time: 0.0021 memory: 25074 loss: 0.5974 +2023/06/01 03:08:42 - mmengine - INFO - Epoch(train) [7][5600/5758] lr: 3.1776e-03 eta: 17:51:02 time: 0.7883 data_time: 0.0016 memory: 25074 loss: 0.5756 +2023/06/01 03:10:05 - mmengine - INFO - Epoch(train) [7][5700/5758] lr: 3.1776e-03 eta: 17:49:30 time: 0.8141 data_time: 0.0013 memory: 25074 loss: 0.5639 +2023/06/01 03:10:52 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_20230531_173252 +2023/06/01 03:10:52 - mmengine - INFO - Saving checkpoint at 7 epochs +2023/06/01 03:11:10 - mmengine - INFO - Epoch(val) [7][16/16] accuracy/top1: 100.0000 single-label/precision_classwise: [100.0, 0.0] single-label/recall_classwise: [100.0, 0.0] single-label/f1-score_classwise: [100.0, 0.0] data_time: 0.2255 time: 0.6097 +2023/06/01 03:12:39 - mmengine - INFO - Epoch(train) [8][ 100/5758] lr: 2.9107e-03 eta: 17:47:17 time: 0.7893 data_time: 0.0018 memory: 25074 loss: 0.5362 +2023/06/01 03:14:00 - mmengine - INFO - Epoch(train) [8][ 200/5758] lr: 2.9107e-03 eta: 17:45:43 time: 0.8145 data_time: 0.0021 memory: 25074 loss: 0.5108 +2023/06/01 03:15:22 - mmengine - INFO - Epoch(train) [8][ 300/5758] lr: 2.9107e-03 eta: 17:44:10 time: 0.8064 data_time: 0.0027 memory: 25074 loss: 0.4770 +2023/06/01 03:16:44 - mmengine - INFO - Epoch(train) [8][ 400/5758] lr: 2.9107e-03 eta: 17:42:37 time: 0.7941 data_time: 0.0021 memory: 25074 loss: 0.4469 +2023/06/01 03:18:02 - mmengine - INFO - Epoch(train) [8][ 500/5758] lr: 2.9107e-03 eta: 17:40:58 time: 0.7754 data_time: 0.0014 memory: 25074 loss: 0.4381 +2023/06/01 03:19:22 - mmengine - INFO - Epoch(train) [8][ 600/5758] lr: 2.9107e-03 eta: 17:39:23 time: 0.8069 data_time: 0.0020 memory: 25074 loss: 0.4025 +2023/06/01 03:20:40 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_20230531_173252 +2023/06/01 03:20:44 - mmengine - INFO - Epoch(train) [8][ 700/5758] lr: 2.9107e-03 eta: 17:37:51 time: 0.8387 data_time: 0.0016 memory: 25074 loss: 0.3908 +2023/06/01 03:22:07 - mmengine - INFO - Epoch(train) [8][ 800/5758] lr: 2.9107e-03 eta: 17:36:21 time: 0.7854 data_time: 0.0015 memory: 25074 loss: 0.3861 +2023/06/01 03:23:30 - mmengine - INFO - Epoch(train) [8][ 900/5758] lr: 2.9107e-03 eta: 17:34:50 time: 0.8259 data_time: 0.0020 memory: 25074 loss: 0.3816 +2023/06/01 03:24:50 - mmengine - INFO - Epoch(train) [8][1000/5758] lr: 2.9107e-03 eta: 17:33:15 time: 0.8218 data_time: 0.0014 memory: 25074 loss: 0.3461 +2023/06/01 03:26:11 - mmengine - INFO - Epoch(train) [8][1100/5758] lr: 2.9107e-03 eta: 17:31:41 time: 0.8307 data_time: 0.0017 memory: 25074 loss: 0.3338 +2023/06/01 03:27:34 - mmengine - INFO - Epoch(train) [8][1200/5758] lr: 2.9107e-03 eta: 17:30:11 time: 0.8040 data_time: 0.0013 memory: 25074 loss: 0.3185 +2023/06/01 03:28:57 - mmengine - INFO - Epoch(train) [8][1300/5758] lr: 2.9107e-03 eta: 17:28:41 time: 0.8786 data_time: 0.0014 memory: 25074 loss: 0.3485 +2023/06/01 03:30:20 - mmengine - INFO - Epoch(train) [8][1400/5758] lr: 2.9107e-03 eta: 17:27:10 time: 0.8875 data_time: 0.0014 memory: 25074 loss: 0.2783 +2023/06/01 03:31:42 - mmengine - INFO - Epoch(train) [8][1500/5758] lr: 2.9107e-03 eta: 17:25:39 time: 0.8517 data_time: 0.0013 memory: 25074 loss: 0.3125 +2023/06/01 03:33:04 - mmengine - INFO - Epoch(train) [8][1600/5758] lr: 2.9107e-03 eta: 17:24:06 time: 0.8422 data_time: 0.0014 memory: 25074 loss: 0.2974 +2023/06/01 03:34:21 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_20230531_173252 +2023/06/01 03:34:25 - mmengine - INFO - Epoch(train) [8][1700/5758] lr: 2.9107e-03 eta: 17:22:34 time: 0.7978 data_time: 0.0016 memory: 25074 loss: 0.2832 +2023/06/01 03:35:48 - mmengine - INFO - Epoch(train) [8][1800/5758] lr: 2.9107e-03 eta: 17:21:04 time: 0.8021 data_time: 0.0013 memory: 25074 loss: 0.2620 +2023/06/01 03:37:09 - mmengine - INFO - Epoch(train) [8][1900/5758] lr: 2.9107e-03 eta: 17:19:30 time: 0.7949 data_time: 0.0015 memory: 25074 loss: 0.2613 +2023/06/01 03:38:32 - mmengine - INFO - Epoch(train) [8][2000/5758] lr: 2.9107e-03 eta: 17:18:00 time: 0.8208 data_time: 0.0014 memory: 25074 loss: 0.2649 +2023/06/01 03:39:55 - mmengine - INFO - Epoch(train) [8][2100/5758] lr: 2.9107e-03 eta: 17:16:30 time: 0.8449 data_time: 0.0014 memory: 25074 loss: 0.2695 +2023/06/01 03:41:15 - mmengine - INFO - Epoch(train) [8][2200/5758] lr: 2.9107e-03 eta: 17:14:56 time: 0.8236 data_time: 0.0023 memory: 25074 loss: 0.2874 +2023/06/01 03:42:36 - mmengine - INFO - Epoch(train) [8][2300/5758] lr: 2.9107e-03 eta: 17:13:23 time: 0.8314 data_time: 0.0021 memory: 25074 loss: 0.2447 +2023/06/01 03:43:58 - mmengine - INFO - Epoch(train) [8][2400/5758] lr: 2.9107e-03 eta: 17:11:51 time: 0.8474 data_time: 0.0021 memory: 25074 loss: 0.2398 +2023/06/01 03:45:19 - mmengine - INFO - Epoch(train) [8][2500/5758] lr: 2.9107e-03 eta: 17:10:18 time: 0.7882 data_time: 0.0022 memory: 25074 loss: 0.2349 +2023/06/01 03:46:42 - mmengine - INFO - Epoch(train) [8][2600/5758] lr: 2.9107e-03 eta: 17:08:49 time: 0.8113 data_time: 0.0023 memory: 25074 loss: 0.2343 +2023/06/01 03:47:57 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_20230531_173252 +2023/06/01 03:48:02 - mmengine - INFO - Epoch(train) [8][2700/5758] lr: 2.9107e-03 eta: 17:07:14 time: 0.8289 data_time: 0.0024 memory: 25074 loss: 0.2269 +2023/06/01 03:49:24 - mmengine - INFO - Epoch(train) [8][2800/5758] lr: 2.9107e-03 eta: 17:05:43 time: 0.7623 data_time: 0.0019 memory: 25074 loss: 0.2178 +2023/06/01 03:50:44 - mmengine - INFO - Epoch(train) [8][2900/5758] lr: 2.9107e-03 eta: 17:04:09 time: 0.7880 data_time: 0.0022 memory: 25074 loss: 0.2173 +2023/06/01 03:52:05 - mmengine - INFO - Epoch(train) [8][3000/5758] lr: 2.9107e-03 eta: 17:02:35 time: 0.8095 data_time: 0.0021 memory: 25074 loss: 0.2091 +2023/06/01 03:53:26 - mmengine - INFO - Epoch(train) [8][3100/5758] lr: 2.9107e-03 eta: 17:01:03 time: 0.8101 data_time: 0.0019 memory: 25074 loss: 0.1714 +2023/06/01 03:54:49 - mmengine - INFO - Epoch(train) [8][3200/5758] lr: 2.9107e-03 eta: 16:59:34 time: 0.8593 data_time: 0.0020 memory: 25074 loss: 0.2080 +2023/06/01 03:56:11 - mmengine - INFO - Epoch(train) [8][3300/5758] lr: 2.9107e-03 eta: 16:58:03 time: 0.8132 data_time: 0.0016 memory: 25074 loss: 0.1868 +2023/06/01 03:57:33 - mmengine - INFO - Epoch(train) [8][3400/5758] lr: 2.9107e-03 eta: 16:56:32 time: 0.7774 data_time: 0.0022 memory: 25074 loss: 0.1665 +2023/06/01 03:58:54 - mmengine - INFO - Epoch(train) [8][3500/5758] lr: 2.9107e-03 eta: 16:55:00 time: 0.8707 data_time: 0.0021 memory: 25074 loss: 0.1845 +2023/06/01 04:00:17 - mmengine - INFO - Epoch(train) [8][3600/5758] lr: 2.9107e-03 eta: 16:53:30 time: 0.7724 data_time: 0.0024 memory: 25074 loss: 0.1662 +2023/06/01 04:01:29 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_20230531_173252 +2023/06/01 04:01:34 - mmengine - INFO - Epoch(train) [8][3700/5758] lr: 2.9107e-03 eta: 16:51:51 time: 0.7987 data_time: 0.0017 memory: 25074 loss: 0.1619 +2023/06/01 04:02:56 - mmengine - INFO - Epoch(train) [8][3800/5758] lr: 2.9107e-03 eta: 16:50:21 time: 0.8377 data_time: 0.0014 memory: 25074 loss: 0.1478 +2023/06/01 04:04:19 - mmengine - INFO - Epoch(train) [8][3900/5758] lr: 2.9107e-03 eta: 16:48:52 time: 0.8408 data_time: 0.0015 memory: 25074 loss: 0.1582 +2023/06/01 04:05:40 - mmengine - INFO - Epoch(train) [8][4000/5758] lr: 2.9107e-03 eta: 16:47:20 time: 0.8367 data_time: 0.0016 memory: 25074 loss: 0.1327 +2023/06/01 04:07:04 - mmengine - INFO - Epoch(train) [8][4100/5758] lr: 2.9107e-03 eta: 16:45:52 time: 0.8209 data_time: 0.0017 memory: 25074 loss: 0.1277 +2023/06/01 04:08:25 - mmengine - INFO - Epoch(train) [8][4200/5758] lr: 2.9107e-03 eta: 16:44:20 time: 0.8249 data_time: 0.0017 memory: 25074 loss: 0.1300 +2023/06/01 04:09:48 - mmengine - INFO - Epoch(train) [8][4300/5758] lr: 2.9107e-03 eta: 16:42:51 time: 0.7803 data_time: 0.0018 memory: 25074 loss: 0.1400 +2023/06/01 04:11:10 - mmengine - INFO - Epoch(train) [8][4400/5758] lr: 2.9107e-03 eta: 16:41:21 time: 0.8321 data_time: 0.0019 memory: 25074 loss: 0.1150 +2023/06/01 04:12:31 - mmengine - INFO - Epoch(train) [8][4500/5758] lr: 2.9107e-03 eta: 16:39:48 time: 0.8187 data_time: 0.0019 memory: 25074 loss: 0.1016 +2023/06/01 04:13:53 - mmengine - INFO - Epoch(train) [8][4600/5758] lr: 2.9107e-03 eta: 16:38:18 time: 0.7879 data_time: 0.0021 memory: 25074 loss: 0.0986 +2023/06/01 04:15:11 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_20230531_173252 +2023/06/01 04:15:15 - mmengine - INFO - Epoch(train) [8][4700/5758] lr: 2.9107e-03 eta: 16:36:49 time: 0.7465 data_time: 0.0029 memory: 25074 loss: 0.0792 +2023/06/01 04:16:37 - mmengine - INFO - Epoch(train) [8][4800/5758] lr: 2.9107e-03 eta: 16:35:17 time: 0.7901 data_time: 0.0026 memory: 25074 loss: 0.0931 +2023/06/01 04:17:59 - mmengine - INFO - Epoch(train) [8][4900/5758] lr: 2.9107e-03 eta: 16:33:47 time: 0.8152 data_time: 0.0021 memory: 25074 loss: 0.0984 +2023/06/01 04:19:22 - mmengine - INFO - Epoch(train) [8][5000/5758] lr: 2.9107e-03 eta: 16:32:18 time: 0.8074 data_time: 0.0022 memory: 25074 loss: 0.0778 +2023/06/01 04:20:44 - mmengine - INFO - Epoch(train) [8][5100/5758] lr: 2.9107e-03 eta: 16:30:48 time: 0.8148 data_time: 0.0018 memory: 25074 loss: 0.0694 +2023/06/01 04:22:05 - mmengine - INFO - Epoch(train) [8][5200/5758] lr: 2.9107e-03 eta: 16:29:16 time: 0.8064 data_time: 0.0014 memory: 25074 loss: 0.0627 +2023/06/01 04:23:27 - mmengine - INFO - Epoch(train) [8][5300/5758] lr: 2.9107e-03 eta: 16:27:46 time: 0.7676 data_time: 0.0017 memory: 25074 loss: 0.0830 +2023/06/01 04:24:49 - mmengine - INFO - Epoch(train) [8][5400/5758] lr: 2.9107e-03 eta: 16:26:16 time: 0.8051 data_time: 0.0019 memory: 25074 loss: 0.0650 +2023/06/01 04:26:10 - mmengine - INFO - Epoch(train) [8][5500/5758] lr: 2.9107e-03 eta: 16:24:45 time: 0.8106 data_time: 0.0023 memory: 25074 loss: 0.0770 +2023/06/01 04:27:31 - mmengine - INFO - Epoch(train) [8][5600/5758] lr: 2.9107e-03 eta: 16:23:13 time: 0.8026 data_time: 0.0015 memory: 25074 loss: 0.0426 +2023/06/01 04:28:48 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_20230531_173252 +2023/06/01 04:28:52 - mmengine - INFO - Epoch(train) [8][5700/5758] lr: 2.9107e-03 eta: 16:21:42 time: 0.8076 data_time: 0.0020 memory: 25074 loss: 0.0647 +2023/06/01 04:29:43 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_20230531_173252 +2023/06/01 04:29:43 - mmengine - INFO - Saving checkpoint at 8 epochs +2023/06/01 04:30:03 - mmengine - INFO - Epoch(val) [8][16/16] accuracy/top1: 100.0000 single-label/precision_classwise: [100.0, 0.0] single-label/recall_classwise: [100.0, 0.0] single-label/f1-score_classwise: [100.0, 0.0] data_time: 0.2316 time: 0.6260 +2023/06/01 04:31:32 - mmengine - INFO - Epoch(train) [9][ 100/5758] lr: 2.6215e-03 eta: 16:19:36 time: 0.7938 data_time: 0.0020 memory: 25074 loss: 0.0473 +2023/06/01 04:32:54 - mmengine - INFO - Epoch(train) [9][ 200/5758] lr: 2.6215e-03 eta: 16:18:06 time: 0.8486 data_time: 0.0014 memory: 25074 loss: 0.0379 +2023/06/01 04:34:17 - mmengine - INFO - Epoch(train) [9][ 300/5758] lr: 2.6215e-03 eta: 16:16:37 time: 0.8137 data_time: 0.0022 memory: 25074 loss: 0.0486 +2023/06/01 04:35:38 - mmengine - INFO - Epoch(train) [9][ 400/5758] lr: 2.6215e-03 eta: 16:15:05 time: 0.8413 data_time: 0.0022 memory: 25074 loss: 0.0392 +2023/06/01 04:36:58 - mmengine - INFO - Epoch(train) [9][ 500/5758] lr: 2.6215e-03 eta: 16:13:34 time: 0.8025 data_time: 0.0021 memory: 25074 loss: 0.0501 +2023/06/01 04:38:21 - mmengine - INFO - Epoch(train) [9][ 600/5758] lr: 2.6215e-03 eta: 16:12:04 time: 0.8108 data_time: 0.0019 memory: 25074 loss: 0.0653 +2023/06/01 04:39:42 - mmengine - INFO - Epoch(train) [9][ 700/5758] lr: 2.6215e-03 eta: 16:10:33 time: 0.7812 data_time: 0.0016 memory: 25074 loss: 0.0389 +2023/06/01 04:41:01 - mmengine - INFO - Epoch(train) [9][ 800/5758] lr: 2.6215e-03 eta: 16:08:59 time: 0.8008 data_time: 0.0023 memory: 25074 loss: 0.0421 +2023/06/01 04:42:24 - mmengine - INFO - Epoch(train) [9][ 900/5758] lr: 2.6215e-03 eta: 16:07:31 time: 0.7815 data_time: 0.0017 memory: 25074 loss: 0.0425 +2023/06/01 04:42:54 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_20230531_173252 +2023/06/01 04:43:45 - mmengine - INFO - Epoch(train) [9][1000/5758] lr: 2.6215e-03 eta: 16:06:01 time: 0.8031 data_time: 0.0023 memory: 25074 loss: 0.0496 +2023/06/01 04:45:08 - mmengine - INFO - Epoch(train) [9][1100/5758] lr: 2.6215e-03 eta: 16:04:32 time: 0.8107 data_time: 0.0018 memory: 25074 loss: 0.0354 +2023/06/01 04:46:31 - mmengine - INFO - Epoch(train) [9][1200/5758] lr: 2.6215e-03 eta: 16:03:04 time: 0.8082 data_time: 0.0021 memory: 25074 loss: 0.0424 +2023/06/01 04:47:54 - mmengine - INFO - Epoch(train) [9][1300/5758] lr: 2.6215e-03 eta: 16:01:36 time: 0.8648 data_time: 0.0016 memory: 25074 loss: 0.0406 +2023/06/01 04:49:16 - mmengine - INFO - Epoch(train) [9][1400/5758] lr: 2.6215e-03 eta: 16:00:07 time: 0.8223 data_time: 0.0023 memory: 25074 loss: 0.0584 +2023/06/01 04:50:38 - mmengine - INFO - Epoch(train) [9][1500/5758] lr: 2.6215e-03 eta: 15:58:36 time: 0.7938 data_time: 0.0019 memory: 25074 loss: 0.0369 +2023/06/01 04:51:59 - mmengine - INFO - Epoch(train) [9][1600/5758] lr: 2.6215e-03 eta: 15:57:06 time: 0.7814 data_time: 0.0021 memory: 25074 loss: 0.0269 +2023/06/01 04:53:20 - mmengine - INFO - Epoch(train) [9][1700/5758] lr: 2.6215e-03 eta: 15:55:34 time: 0.7761 data_time: 0.0018 memory: 25074 loss: 0.0410 +2023/06/01 04:54:42 - mmengine - INFO - Epoch(train) [9][1800/5758] lr: 2.6215e-03 eta: 15:54:05 time: 0.8536 data_time: 0.0019 memory: 25074 loss: 0.0302 +2023/06/01 04:56:04 - mmengine - INFO - Epoch(train) [9][1900/5758] lr: 2.6215e-03 eta: 15:52:36 time: 0.8638 data_time: 0.0019 memory: 25074 loss: 0.1361 +2023/06/01 04:56:34 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_20230531_173252 +2023/06/01 04:57:26 - mmengine - INFO - Epoch(train) [9][2000/5758] lr: 2.6215e-03 eta: 15:51:07 time: 0.7930 data_time: 0.0021 memory: 25074 loss: 0.0336 +2023/06/01 04:58:47 - mmengine - INFO - Epoch(train) [9][2100/5758] lr: 2.6215e-03 eta: 15:49:36 time: 0.7864 data_time: 0.0017 memory: 25074 loss: 0.0320 +2023/06/01 05:00:08 - mmengine - INFO - Epoch(train) [9][2200/5758] lr: 2.6215e-03 eta: 15:48:05 time: 0.8411 data_time: 0.0023 memory: 25074 loss: 0.0256 +2023/06/01 05:01:31 - mmengine - INFO - Epoch(train) [9][2300/5758] lr: 2.6215e-03 eta: 15:46:37 time: 0.7890 data_time: 0.0015 memory: 25074 loss: 0.0304 +2023/06/01 05:02:52 - mmengine - INFO - Epoch(train) [9][2400/5758] lr: 2.6215e-03 eta: 15:45:07 time: 0.8084 data_time: 0.0020 memory: 25074 loss: 0.0360 +2023/06/01 05:04:14 - mmengine - INFO - Epoch(train) [9][2500/5758] lr: 2.6215e-03 eta: 15:43:37 time: 0.7925 data_time: 0.0023 memory: 25074 loss: 0.0213 +2023/06/01 05:05:37 - mmengine - INFO - Epoch(train) [9][2600/5758] lr: 2.6215e-03 eta: 15:42:10 time: 0.8228 data_time: 0.0025 memory: 25074 loss: 0.0444 +2023/06/01 05:06:58 - mmengine - INFO - Epoch(train) [9][2700/5758] lr: 2.6215e-03 eta: 15:40:39 time: 0.7708 data_time: 0.0022 memory: 25074 loss: 0.0278 +2023/06/01 05:08:20 - mmengine - INFO - Epoch(train) [9][2800/5758] lr: 2.6215e-03 eta: 15:39:10 time: 0.8667 data_time: 0.0017 memory: 25074 loss: 0.0206 +2023/06/01 05:09:42 - mmengine - INFO - Epoch(train) [9][2900/5758] lr: 2.6215e-03 eta: 15:37:41 time: 0.8383 data_time: 0.0017 memory: 25074 loss: 0.0244 +2023/06/01 05:10:11 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_20230531_173252 +2023/06/01 05:11:03 - mmengine - INFO - Epoch(train) [9][3000/5758] lr: 2.6215e-03 eta: 15:36:10 time: 0.8165 data_time: 0.0015 memory: 25074 loss: 0.0269 +2023/06/01 05:12:26 - mmengine - INFO - Epoch(train) [9][3100/5758] lr: 2.6215e-03 eta: 15:34:44 time: 0.8457 data_time: 0.0017 memory: 25074 loss: 0.0242 +2023/06/01 05:13:49 - mmengine - INFO - Epoch(train) [9][3200/5758] lr: 2.6215e-03 eta: 15:33:15 time: 0.8112 data_time: 0.0014 memory: 25074 loss: 0.0352 +2023/06/01 05:15:10 - mmengine - INFO - Epoch(train) [9][3300/5758] lr: 2.6215e-03 eta: 15:31:46 time: 0.8152 data_time: 0.0019 memory: 25074 loss: 0.2452 +2023/06/01 05:16:31 - mmengine - INFO - Epoch(train) [9][3400/5758] lr: 2.6215e-03 eta: 15:30:16 time: 0.8276 data_time: 0.0014 memory: 25074 loss: 0.0282 +2023/06/01 05:17:54 - mmengine - INFO - Epoch(train) [9][3500/5758] lr: 2.6215e-03 eta: 15:28:47 time: 0.7728 data_time: 0.0015 memory: 25074 loss: 0.0234 +2023/06/01 05:19:16 - mmengine - INFO - Epoch(train) [9][3600/5758] lr: 2.6215e-03 eta: 15:27:19 time: 0.8695 data_time: 0.0017 memory: 25074 loss: 0.0234 +2023/06/01 05:20:39 - mmengine - INFO - Epoch(train) [9][3700/5758] lr: 2.6215e-03 eta: 15:25:51 time: 0.8025 data_time: 0.0014 memory: 25074 loss: 0.0151 +2023/06/01 05:21:57 - mmengine - INFO - Epoch(train) [9][3800/5758] lr: 2.6215e-03 eta: 15:24:17 time: 0.8283 data_time: 0.0017 memory: 25074 loss: 0.0320 +2023/06/01 05:23:17 - mmengine - INFO - Epoch(train) [9][3900/5758] lr: 2.6215e-03 eta: 15:22:46 time: 0.7985 data_time: 0.0023 memory: 25074 loss: 0.0139 +2023/06/01 05:23:47 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_20230531_173252 +2023/06/01 05:24:39 - mmengine - INFO - Epoch(train) [9][4000/5758] lr: 2.6215e-03 eta: 15:21:16 time: 0.7803 data_time: 0.0020 memory: 25074 loss: 0.0388 +2023/06/01 05:25:58 - mmengine - INFO - Epoch(train) [9][4100/5758] lr: 2.6215e-03 eta: 15:19:44 time: 0.7932 data_time: 0.0015 memory: 25074 loss: 0.0264 +2023/06/01 05:27:20 - mmengine - INFO - Epoch(train) [9][4200/5758] lr: 2.6215e-03 eta: 15:18:15 time: 0.8094 data_time: 0.0020 memory: 25074 loss: 0.0261 +2023/06/01 05:28:41 - mmengine - INFO - Epoch(train) [9][4300/5758] lr: 2.6215e-03 eta: 15:16:46 time: 0.8204 data_time: 0.0025 memory: 25074 loss: 0.0220 +2023/06/01 05:30:02 - mmengine - INFO - Epoch(train) [9][4400/5758] lr: 2.6215e-03 eta: 15:15:16 time: 0.7742 data_time: 0.0019 memory: 25074 loss: 0.0302 +2023/06/01 05:31:24 - mmengine - INFO - Epoch(train) [9][4500/5758] lr: 2.6215e-03 eta: 15:13:47 time: 0.8320 data_time: 0.0026 memory: 25074 loss: 0.0197 +2023/06/01 05:32:44 - mmengine - INFO - Epoch(train) [9][4600/5758] lr: 2.6215e-03 eta: 15:12:16 time: 0.8003 data_time: 0.0019 memory: 25074 loss: 0.0249 +2023/06/01 05:34:06 - mmengine - INFO - Epoch(train) [9][4700/5758] lr: 2.6215e-03 eta: 15:10:48 time: 0.8219 data_time: 0.0015 memory: 25074 loss: 0.0292 +2023/06/01 05:35:28 - mmengine - INFO - Epoch(train) [9][4800/5758] lr: 2.6215e-03 eta: 15:09:19 time: 0.7957 data_time: 0.0016 memory: 25074 loss: 0.0230 +2023/06/01 05:36:48 - mmengine - INFO - Epoch(train) [9][4900/5758] lr: 2.6215e-03 eta: 15:07:48 time: 0.8086 data_time: 0.0019 memory: 25074 loss: 0.0224 +2023/06/01 05:37:18 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_20230531_173252 +2023/06/01 05:38:10 - mmengine - INFO - Epoch(train) [9][5000/5758] lr: 2.6215e-03 eta: 15:06:19 time: 0.7607 data_time: 0.0018 memory: 25074 loss: 0.0186 +2023/06/01 05:39:32 - mmengine - INFO - Epoch(train) [9][5100/5758] lr: 2.6215e-03 eta: 15:04:51 time: 0.8408 data_time: 0.0020 memory: 25074 loss: 0.0331 +2023/06/01 05:40:53 - mmengine - INFO - Epoch(train) [9][5200/5758] lr: 2.6215e-03 eta: 15:03:21 time: 0.8428 data_time: 0.0016 memory: 25074 loss: 0.0149 +2023/06/01 05:42:13 - mmengine - INFO - Epoch(train) [9][5300/5758] lr: 2.6215e-03 eta: 15:01:51 time: 0.8012 data_time: 0.0027 memory: 25074 loss: 0.0286 +2023/06/01 05:43:35 - mmengine - INFO - Epoch(train) [9][5400/5758] lr: 2.6215e-03 eta: 15:00:22 time: 0.8069 data_time: 0.0024 memory: 25074 loss: 0.0266 +2023/06/01 05:44:55 - mmengine - INFO - Epoch(train) [9][5500/5758] lr: 2.6215e-03 eta: 14:58:52 time: 0.8241 data_time: 0.0019 memory: 25074 loss: 0.0203 +2023/06/01 05:46:17 - mmengine - INFO - Epoch(train) [9][5600/5758] lr: 2.6215e-03 eta: 14:57:23 time: 0.8194 data_time: 0.0019 memory: 25074 loss: 0.0232 +2023/06/01 05:47:39 - mmengine - INFO - Epoch(train) [9][5700/5758] lr: 2.6215e-03 eta: 14:55:56 time: 0.8480 data_time: 0.0022 memory: 25074 loss: 0.0144 +2023/06/01 05:48:26 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_20230531_173252 +2023/06/01 05:48:26 - mmengine - INFO - Saving checkpoint at 9 epochs +2023/06/01 05:48:45 - mmengine - INFO - Epoch(val) [9][16/16] accuracy/top1: 99.3308 single-label/precision_classwise: [100.0, 0.0] single-label/recall_classwise: [99.33084869384766, 0.0] single-label/f1-score_classwise: [99.664306640625, 0.0] data_time: 0.2375 time: 0.6248 +2023/06/01 05:50:14 - mmengine - INFO - Epoch(train) [10][ 100/5758] lr: 2.3171e-03 eta: 14:53:44 time: 0.8207 data_time: 0.0020 memory: 25074 loss: 0.0175 +2023/06/01 05:51:18 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_20230531_173252 +2023/06/01 05:51:36 - mmengine - INFO - Epoch(train) [10][ 200/5758] lr: 2.3171e-03 eta: 14:52:16 time: 0.7813 data_time: 0.0016 memory: 25074 loss: 0.0177 +2023/06/01 05:52:57 - mmengine - INFO - Epoch(train) [10][ 300/5758] lr: 2.3171e-03 eta: 14:50:46 time: 0.8036 data_time: 0.0025 memory: 25074 loss: 0.0118 +2023/06/01 05:54:20 - mmengine - INFO - Epoch(train) [10][ 400/5758] lr: 2.3171e-03 eta: 14:49:19 time: 0.7797 data_time: 0.0017 memory: 25074 loss: 0.0121 +2023/06/01 05:55:40 - mmengine - INFO - Epoch(train) [10][ 500/5758] lr: 2.3171e-03 eta: 14:47:49 time: 0.8295 data_time: 0.0017 memory: 25074 loss: 0.0131 +2023/06/01 05:57:03 - mmengine - INFO - Epoch(train) [10][ 600/5758] lr: 2.3171e-03 eta: 14:46:22 time: 0.8365 data_time: 0.0016 memory: 25074 loss: 0.0196 +2023/06/01 05:58:25 - mmengine - INFO - Epoch(train) [10][ 700/5758] lr: 2.3171e-03 eta: 14:44:54 time: 0.8344 data_time: 0.0026 memory: 25074 loss: 0.0111 +2023/06/01 05:59:45 - mmengine - INFO - Epoch(train) [10][ 800/5758] lr: 2.3171e-03 eta: 14:43:24 time: 0.7452 data_time: 0.0020 memory: 25074 loss: 0.0180 +2023/06/01 06:01:09 - mmengine - INFO - Epoch(train) [10][ 900/5758] lr: 2.3171e-03 eta: 14:41:58 time: 0.7868 data_time: 0.0018 memory: 25074 loss: 0.0188 +2023/06/01 06:02:29 - mmengine - INFO - Epoch(train) [10][1000/5758] lr: 2.3171e-03 eta: 14:40:28 time: 0.7551 data_time: 0.0026 memory: 25074 loss: 0.0201 +2023/06/01 06:03:50 - mmengine - INFO - Epoch(train) [10][1100/5758] lr: 2.3171e-03 eta: 14:38:58 time: 0.7882 data_time: 0.0021 memory: 25074 loss: 0.0149 +2023/06/01 06:04:54 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_20230531_173252 +2023/06/01 06:05:12 - mmengine - INFO - Epoch(train) [10][1200/5758] lr: 2.3171e-03 eta: 14:37:30 time: 0.8209 data_time: 0.0020 memory: 25074 loss: 0.0153 +2023/06/01 06:06:33 - mmengine - INFO - Epoch(train) [10][1300/5758] lr: 2.3171e-03 eta: 14:36:02 time: 0.7906 data_time: 0.0017 memory: 25074 loss: 0.0131 +2023/06/01 06:07:56 - mmengine - INFO - Epoch(train) [10][1400/5758] lr: 2.3171e-03 eta: 14:34:34 time: 0.7908 data_time: 0.0018 memory: 25074 loss: 0.0138 +2023/06/01 06:09:16 - mmengine - INFO - Epoch(train) [10][1500/5758] lr: 2.3171e-03 eta: 14:33:04 time: 0.7822 data_time: 0.0022 memory: 25074 loss: 0.0182 +2023/06/01 06:10:39 - mmengine - INFO - Epoch(train) [10][1600/5758] lr: 2.3171e-03 eta: 14:31:37 time: 0.8082 data_time: 0.0027 memory: 25074 loss: 0.0141 +2023/06/01 06:12:00 - mmengine - INFO - Epoch(train) [10][1700/5758] lr: 2.3171e-03 eta: 14:30:08 time: 0.8055 data_time: 0.0020 memory: 25074 loss: 0.0160 +2023/06/01 06:13:23 - mmengine - INFO - Epoch(train) [10][1800/5758] lr: 2.3171e-03 eta: 14:28:42 time: 0.8374 data_time: 0.0020 memory: 25074 loss: 0.0092 +2023/06/01 06:14:47 - mmengine - INFO - Epoch(train) [10][1900/5758] lr: 2.3171e-03 eta: 14:27:16 time: 0.8446 data_time: 0.0017 memory: 25074 loss: 0.0152 +2023/06/01 06:16:09 - mmengine - INFO - Epoch(train) [10][2000/5758] lr: 2.3171e-03 eta: 14:25:48 time: 0.8119 data_time: 0.0019 memory: 25074 loss: 0.0116 +2023/06/01 06:17:32 - mmengine - INFO - Epoch(train) [10][2100/5758] lr: 2.3171e-03 eta: 14:24:22 time: 0.8511 data_time: 0.0022 memory: 25074 loss: 0.0124 +2023/06/01 06:18:36 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_20230531_173252 +2023/06/01 06:18:54 - mmengine - INFO - Epoch(train) [10][2200/5758] lr: 2.3171e-03 eta: 14:22:54 time: 0.8317 data_time: 0.0018 memory: 25074 loss: 0.0117 +2023/06/01 06:20:17 - mmengine - INFO - Epoch(train) [10][2300/5758] lr: 2.3171e-03 eta: 14:21:27 time: 0.8639 data_time: 0.0026 memory: 25074 loss: 0.0246 +2023/06/01 06:21:39 - mmengine - INFO - Epoch(train) [10][2400/5758] lr: 2.3171e-03 eta: 14:20:00 time: 0.7997 data_time: 0.0022 memory: 25074 loss: 0.0215 +2023/06/01 06:23:00 - mmengine - INFO - Epoch(train) [10][2500/5758] lr: 2.3171e-03 eta: 14:18:31 time: 0.8154 data_time: 0.0026 memory: 25074 loss: 0.0185 +2023/06/01 06:24:21 - mmengine - INFO - Epoch(train) [10][2600/5758] lr: 2.3171e-03 eta: 14:17:02 time: 0.8356 data_time: 0.0022 memory: 25074 loss: 0.6881 +2023/06/01 06:25:43 - mmengine - INFO - Epoch(train) [10][2700/5758] lr: 2.3171e-03 eta: 14:15:34 time: 0.8412 data_time: 0.0021 memory: 25074 loss: 0.6841 +2023/06/01 06:27:05 - mmengine - INFO - Epoch(train) [10][2800/5758] lr: 2.3171e-03 eta: 14:14:07 time: 0.8091 data_time: 0.0022 memory: 25074 loss: 0.6862 +2023/06/01 06:28:28 - mmengine - INFO - Epoch(train) [10][2900/5758] lr: 2.3171e-03 eta: 14:12:40 time: 0.8718 data_time: 0.0019 memory: 25074 loss: 0.6753 +2023/06/01 06:29:50 - mmengine - INFO - Epoch(train) [10][3000/5758] lr: 2.3171e-03 eta: 14:11:13 time: 0.8286 data_time: 0.0019 memory: 25074 loss: 0.5613 +2023/06/01 06:31:12 - mmengine - INFO - Epoch(train) [10][3100/5758] lr: 2.3171e-03 eta: 14:09:45 time: 0.8195 data_time: 0.0024 memory: 25074 loss: 0.4973 +2023/06/01 06:32:16 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_20230531_173252 +2023/06/01 06:32:33 - mmengine - INFO - Epoch(train) [10][3200/5758] lr: 2.3171e-03 eta: 14:08:17 time: 0.8235 data_time: 0.0015 memory: 25074 loss: 0.4726 +2023/06/01 06:33:55 - mmengine - INFO - Epoch(train) [10][3300/5758] lr: 2.3171e-03 eta: 14:06:49 time: 0.7902 data_time: 0.0016 memory: 25074 loss: 0.4224 +2023/06/01 06:35:18 - mmengine - INFO - Epoch(train) [10][3400/5758] lr: 2.3171e-03 eta: 14:05:22 time: 0.7914 data_time: 0.0023 memory: 25074 loss: 0.4256 +2023/06/01 06:36:41 - mmengine - INFO - Epoch(train) [10][3500/5758] lr: 2.3171e-03 eta: 14:03:56 time: 0.8414 data_time: 0.0018 memory: 25074 loss: 0.3666 +2023/06/01 06:38:03 - mmengine - INFO - Epoch(train) [10][3600/5758] lr: 2.3171e-03 eta: 14:02:29 time: 0.7918 data_time: 0.0016 memory: 25074 loss: 0.3317 +2023/06/01 06:39:27 - mmengine - INFO - Epoch(train) [10][3700/5758] lr: 2.3171e-03 eta: 14:01:03 time: 0.8647 data_time: 0.0022 memory: 25074 loss: 0.2783 +2023/06/01 06:40:49 - mmengine - INFO - Epoch(train) [10][3800/5758] lr: 2.3171e-03 eta: 13:59:36 time: 0.7787 data_time: 0.0023 memory: 25074 loss: 0.2840 +2023/06/01 06:42:12 - mmengine - INFO - Epoch(train) [10][3900/5758] lr: 2.3171e-03 eta: 13:58:09 time: 0.8275 data_time: 0.0023 memory: 25074 loss: 0.2700 +2023/06/01 06:43:34 - mmengine - INFO - Epoch(train) [10][4000/5758] lr: 2.3171e-03 eta: 13:56:42 time: 0.8666 data_time: 0.0025 memory: 25074 loss: 0.2077 +2023/06/01 06:44:56 - mmengine - INFO - Epoch(train) [10][4100/5758] lr: 2.3171e-03 eta: 13:55:15 time: 0.8322 data_time: 0.0022 memory: 25074 loss: 0.1953 +2023/06/01 06:46:00 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_20230531_173252 +2023/06/01 06:46:18 - mmengine - INFO - Epoch(train) [10][4200/5758] lr: 2.3171e-03 eta: 13:53:48 time: 0.8302 data_time: 0.0020 memory: 25074 loss: 0.1650 +2023/06/01 06:47:39 - mmengine - INFO - Epoch(train) [10][4300/5758] lr: 2.3171e-03 eta: 13:52:19 time: 0.7660 data_time: 0.0022 memory: 25074 loss: 0.1104 +2023/06/01 06:49:02 - mmengine - INFO - Epoch(train) [10][4400/5758] lr: 2.3171e-03 eta: 13:50:53 time: 0.9145 data_time: 0.0023 memory: 25074 loss: 0.0837 +2023/06/01 06:50:25 - mmengine - INFO - Epoch(train) [10][4500/5758] lr: 2.3171e-03 eta: 13:49:27 time: 0.8665 data_time: 0.0025 memory: 25074 loss: 0.0865 +2023/06/01 06:51:48 - mmengine - INFO - Epoch(train) [10][4600/5758] lr: 2.3171e-03 eta: 13:48:01 time: 0.9035 data_time: 0.0022 memory: 25074 loss: 0.0692 +2023/06/01 06:53:09 - mmengine - INFO - Epoch(train) [10][4700/5758] lr: 2.3171e-03 eta: 13:46:33 time: 0.8001 data_time: 0.0015 memory: 25074 loss: 0.0423 +2023/06/01 06:54:32 - mmengine - INFO - Epoch(train) [10][4800/5758] lr: 2.3171e-03 eta: 13:45:06 time: 0.8566 data_time: 0.0020 memory: 25074 loss: 0.0463 +2023/06/01 06:55:53 - mmengine - INFO - Epoch(train) [10][4900/5758] lr: 2.3171e-03 eta: 13:43:38 time: 0.8226 data_time: 0.0017 memory: 25074 loss: 0.0312 +2023/06/01 06:57:14 - mmengine - INFO - Epoch(train) [10][5000/5758] lr: 2.3171e-03 eta: 13:42:10 time: 0.8224 data_time: 0.0026 memory: 25074 loss: 0.0410 +2023/06/01 06:58:35 - mmengine - INFO - Epoch(train) [10][5100/5758] lr: 2.3171e-03 eta: 13:40:41 time: 0.7936 data_time: 0.0020 memory: 25074 loss: 0.0424 +2023/06/01 06:59:39 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_20230531_173252 +2023/06/01 06:59:57 - mmengine - INFO - Epoch(train) [10][5200/5758] lr: 2.3171e-03 eta: 13:39:14 time: 0.7964 data_time: 0.0025 memory: 25074 loss: 0.0250 +2023/06/01 07:01:19 - mmengine - INFO - Epoch(train) [10][5300/5758] lr: 2.3171e-03 eta: 13:37:47 time: 0.8108 data_time: 0.0023 memory: 25074 loss: 0.0514 +2023/06/01 07:02:42 - mmengine - INFO - Epoch(train) [10][5400/5758] lr: 2.3171e-03 eta: 13:36:21 time: 0.8978 data_time: 0.0021 memory: 25074 loss: 0.0274 +2023/06/01 07:04:04 - mmengine - INFO - Epoch(train) [10][5500/5758] lr: 2.3171e-03 eta: 13:34:54 time: 0.8460 data_time: 0.0024 memory: 25074 loss: 0.0313 +2023/06/01 07:05:27 - mmengine - INFO - Epoch(train) [10][5600/5758] lr: 2.3171e-03 eta: 13:33:28 time: 0.8046 data_time: 0.0026 memory: 25074 loss: 0.0202 +2023/06/01 07:06:49 - mmengine - INFO - Epoch(train) [10][5700/5758] lr: 2.3171e-03 eta: 13:32:00 time: 0.8253 data_time: 0.0024 memory: 25074 loss: 0.0205 +2023/06/01 07:07:36 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_20230531_173252 +2023/06/01 07:07:36 - mmengine - INFO - Saving checkpoint at 10 epochs +2023/06/01 07:07:54 - mmengine - INFO - Epoch(val) [10][16/16] accuracy/top1: 100.0000 single-label/precision_classwise: [100.0, 0.0] single-label/recall_classwise: [100.0, 0.0] single-label/f1-score_classwise: [100.0, 0.0] data_time: 0.2228 time: 0.6108 +2023/06/01 07:09:25 - mmengine - INFO - Epoch(train) [11][ 100/5758] lr: 2.0050e-03 eta: 13:29:51 time: 0.7896 data_time: 0.0026 memory: 25074 loss: 0.0207 +2023/06/01 07:10:48 - mmengine - INFO - Epoch(train) [11][ 200/5758] lr: 2.0050e-03 eta: 13:28:24 time: 0.8085 data_time: 0.0021 memory: 25074 loss: 0.0184 +2023/06/01 07:12:07 - mmengine - INFO - Epoch(train) [11][ 300/5758] lr: 2.0050e-03 eta: 13:26:55 time: 0.7892 data_time: 0.0019 memory: 25074 loss: 0.0260 +2023/06/01 07:13:29 - mmengine - INFO - Epoch(train) [11][ 400/5758] lr: 2.0050e-03 eta: 13:25:28 time: 0.8114 data_time: 0.0018 memory: 25074 loss: 0.0138 +2023/06/01 07:13:46 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_20230531_173252 +2023/06/01 07:14:51 - mmengine - INFO - Epoch(train) [11][ 500/5758] lr: 2.0050e-03 eta: 13:24:01 time: 0.8287 data_time: 0.0017 memory: 25074 loss: 0.0174 +2023/06/01 07:16:13 - mmengine - INFO - Epoch(train) [11][ 600/5758] lr: 2.0050e-03 eta: 13:22:34 time: 0.7944 data_time: 0.0019 memory: 25074 loss: 0.0243 +2023/06/01 07:17:35 - mmengine - INFO - Epoch(train) [11][ 700/5758] lr: 2.0050e-03 eta: 13:21:07 time: 0.7873 data_time: 0.0029 memory: 25074 loss: 0.0201 +2023/06/01 07:18:57 - mmengine - INFO - Epoch(train) [11][ 800/5758] lr: 2.0050e-03 eta: 13:19:40 time: 0.8720 data_time: 0.0029 memory: 25074 loss: 0.0144 +2023/06/01 07:20:21 - mmengine - INFO - Epoch(train) [11][ 900/5758] lr: 2.0050e-03 eta: 13:18:15 time: 0.8665 data_time: 0.0025 memory: 25074 loss: 0.0190 +2023/06/01 07:21:43 - mmengine - INFO - Epoch(train) [11][1000/5758] lr: 2.0050e-03 eta: 13:16:48 time: 0.7974 data_time: 0.0020 memory: 25074 loss: 0.0185 +2023/06/01 07:23:05 - mmengine - INFO - Epoch(train) [11][1100/5758] lr: 2.0050e-03 eta: 13:15:21 time: 0.7817 data_time: 0.0018 memory: 25074 loss: 0.0580 +2023/06/01 07:24:27 - mmengine - INFO - Epoch(train) [11][1200/5758] lr: 2.0050e-03 eta: 13:13:54 time: 0.8420 data_time: 0.0017 memory: 25074 loss: 0.0130 +2023/06/01 07:25:49 - mmengine - INFO - Epoch(train) [11][1300/5758] lr: 2.0050e-03 eta: 13:12:28 time: 0.7899 data_time: 0.0022 memory: 25074 loss: 0.0250 +2023/06/01 07:27:12 - mmengine - INFO - Epoch(train) [11][1400/5758] lr: 2.0050e-03 eta: 13:11:01 time: 0.8453 data_time: 0.0021 memory: 25074 loss: 0.0197 +2023/06/01 07:27:28 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_20230531_173252 +2023/06/01 07:28:33 - mmengine - INFO - Epoch(train) [11][1500/5758] lr: 2.0050e-03 eta: 13:09:34 time: 0.7987 data_time: 0.0023 memory: 25074 loss: 0.0098 +2023/06/01 07:29:56 - mmengine - INFO - Epoch(train) [11][1600/5758] lr: 2.0050e-03 eta: 13:08:07 time: 0.8288 data_time: 0.0019 memory: 25074 loss: 0.0116 +2023/06/01 07:31:19 - mmengine - INFO - Epoch(train) [11][1700/5758] lr: 2.0050e-03 eta: 13:06:42 time: 0.8637 data_time: 0.0020 memory: 25074 loss: 0.0153 +2023/06/01 07:32:41 - mmengine - INFO - Epoch(train) [11][1800/5758] lr: 2.0050e-03 eta: 13:05:15 time: 0.8098 data_time: 0.0022 memory: 25074 loss: 0.0148 +2023/06/01 07:34:02 - mmengine - INFO - Epoch(train) [11][1900/5758] lr: 2.0050e-03 eta: 13:03:47 time: 0.8413 data_time: 0.0025 memory: 25074 loss: 0.0159 +2023/06/01 07:35:25 - mmengine - INFO - Epoch(train) [11][2000/5758] lr: 2.0050e-03 eta: 13:02:21 time: 0.7999 data_time: 0.0022 memory: 25074 loss: 0.5658 +2023/06/01 07:36:47 - mmengine - INFO - Epoch(train) [11][2100/5758] lr: 2.0050e-03 eta: 13:00:55 time: 0.8692 data_time: 0.0019 memory: 25074 loss: 0.4891 +2023/06/01 07:38:09 - mmengine - INFO - Epoch(train) [11][2200/5758] lr: 2.0050e-03 eta: 12:59:28 time: 0.8257 data_time: 0.0017 memory: 25074 loss: 0.4329 +2023/06/01 07:39:32 - mmengine - INFO - Epoch(train) [11][2300/5758] lr: 2.0050e-03 eta: 12:58:02 time: 0.8492 data_time: 0.0016 memory: 25074 loss: 0.3923 +2023/06/01 07:40:53 - mmengine - INFO - Epoch(train) [11][2400/5758] lr: 2.0050e-03 eta: 12:56:34 time: 0.8267 data_time: 0.0022 memory: 25074 loss: 0.3799 +2023/06/01 07:41:09 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_20230531_173252 +2023/06/01 07:42:14 - mmengine - INFO - Epoch(train) [11][2500/5758] lr: 2.0050e-03 eta: 12:55:07 time: 0.7978 data_time: 0.0022 memory: 25074 loss: 0.3877 +2023/06/01 07:43:37 - mmengine - INFO - Epoch(train) [11][2600/5758] lr: 2.0050e-03 eta: 12:53:41 time: 0.8705 data_time: 0.0024 memory: 25074 loss: 0.3053 +2023/06/01 07:44:59 - mmengine - INFO - Epoch(train) [11][2700/5758] lr: 2.0050e-03 eta: 12:52:15 time: 0.7852 data_time: 0.0021 memory: 25074 loss: 0.2704 +2023/06/01 07:46:22 - mmengine - INFO - Epoch(train) [11][2800/5758] lr: 2.0050e-03 eta: 12:50:49 time: 0.8448 data_time: 0.0024 memory: 25074 loss: 0.2515 +2023/06/01 07:47:42 - mmengine - INFO - Epoch(train) [11][2900/5758] lr: 2.0050e-03 eta: 12:49:20 time: 0.8257 data_time: 0.0015 memory: 25074 loss: 0.1319 +2023/06/01 07:48:59 - mmengine - INFO - Epoch(train) [11][3000/5758] lr: 2.0050e-03 eta: 12:47:49 time: 0.7994 data_time: 0.0015 memory: 25074 loss: 0.0617 +2023/06/01 07:50:18 - mmengine - INFO - Epoch(train) [11][3100/5758] lr: 2.0050e-03 eta: 12:46:20 time: 0.8148 data_time: 0.0015 memory: 25074 loss: 0.0287 +2023/06/01 07:51:40 - mmengine - INFO - Epoch(train) [11][3200/5758] lr: 2.0050e-03 eta: 12:44:53 time: 0.7376 data_time: 0.0014 memory: 25074 loss: 0.0306 +2023/06/01 07:53:02 - mmengine - INFO - Epoch(train) [11][3300/5758] lr: 2.0050e-03 eta: 12:43:26 time: 0.8499 data_time: 0.0015 memory: 25074 loss: 0.0208 +2023/06/01 07:54:23 - mmengine - INFO - Epoch(train) [11][3400/5758] lr: 2.0050e-03 eta: 12:42:00 time: 0.8704 data_time: 0.0022 memory: 25074 loss: 0.0297 +2023/06/01 07:54:39 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_20230531_173252 +2023/06/01 07:55:44 - mmengine - INFO - Epoch(train) [11][3500/5758] lr: 2.0050e-03 eta: 12:40:32 time: 0.8545 data_time: 0.0023 memory: 25074 loss: 0.0147 +2023/06/01 07:57:07 - mmengine - INFO - Epoch(train) [11][3600/5758] lr: 2.0050e-03 eta: 12:39:07 time: 0.8331 data_time: 0.0019 memory: 25074 loss: 0.0234 +2023/06/01 07:58:31 - mmengine - INFO - Epoch(train) [11][3700/5758] lr: 2.0050e-03 eta: 12:37:41 time: 0.8547 data_time: 0.0018 memory: 25074 loss: 0.0217 +2023/06/01 07:59:54 - mmengine - INFO - Epoch(train) [11][3800/5758] lr: 2.0050e-03 eta: 12:36:16 time: 0.7948 data_time: 0.0015 memory: 25074 loss: 0.0193 +2023/06/01 08:01:18 - mmengine - INFO - Epoch(train) [11][3900/5758] lr: 2.0050e-03 eta: 12:34:51 time: 0.8642 data_time: 0.0017 memory: 25074 loss: 0.0179 +2023/06/01 08:02:42 - mmengine - INFO - Epoch(train) [11][4000/5758] lr: 2.0050e-03 eta: 12:33:27 time: 0.8782 data_time: 0.0024 memory: 25074 loss: 0.0182 +2023/06/01 08:04:05 - mmengine - INFO - Epoch(train) [11][4100/5758] lr: 2.0050e-03 eta: 12:32:01 time: 0.8442 data_time: 0.0015 memory: 25074 loss: 0.0158 +2023/06/01 08:05:29 - mmengine - INFO - Epoch(train) [11][4200/5758] lr: 2.0050e-03 eta: 12:30:37 time: 0.8643 data_time: 0.0021 memory: 25074 loss: 0.0352 +2023/06/01 08:06:55 - mmengine - INFO - Epoch(train) [11][4300/5758] lr: 2.0050e-03 eta: 12:29:13 time: 0.9276 data_time: 0.0023 memory: 25074 loss: 0.0172 +2023/06/01 08:08:18 - mmengine - INFO - Epoch(train) [11][4400/5758] lr: 2.0050e-03 eta: 12:27:47 time: 0.8090 data_time: 0.0021 memory: 25074 loss: 0.0138 +2023/06/01 08:08:35 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_20230531_173252 +2023/06/01 08:09:43 - mmengine - INFO - Epoch(train) [11][4500/5758] lr: 2.0050e-03 eta: 12:26:24 time: 0.8844 data_time: 0.0017 memory: 25074 loss: 0.0128 +2023/06/01 08:11:07 - mmengine - INFO - Epoch(train) [11][4600/5758] lr: 2.0050e-03 eta: 12:24:59 time: 0.8389 data_time: 0.0018 memory: 25074 loss: 0.0173 +2023/06/01 08:12:32 - mmengine - INFO - Epoch(train) [11][4700/5758] lr: 2.0050e-03 eta: 12:23:36 time: 0.8496 data_time: 0.0015 memory: 25074 loss: 0.0152 +2023/06/01 08:13:59 - mmengine - INFO - Epoch(train) [11][4800/5758] lr: 2.0050e-03 eta: 12:22:13 time: 0.9049 data_time: 0.0027 memory: 25074 loss: 0.0134 +2023/06/01 08:15:23 - mmengine - INFO - Epoch(train) [11][4900/5758] lr: 2.0050e-03 eta: 12:20:48 time: 0.8099 data_time: 0.0019 memory: 25074 loss: 0.0127 +2023/06/01 08:16:49 - mmengine - INFO - Epoch(train) [11][5000/5758] lr: 2.0050e-03 eta: 12:19:25 time: 0.8768 data_time: 0.0014 memory: 25074 loss: 0.0189 +2023/06/01 08:18:13 - mmengine - INFO - Epoch(train) [11][5100/5758] lr: 2.0050e-03 eta: 12:18:01 time: 0.8187 data_time: 0.0022 memory: 25074 loss: 0.0098 +2023/06/01 08:19:37 - mmengine - INFO - Epoch(train) [11][5200/5758] lr: 2.0050e-03 eta: 12:16:36 time: 0.8418 data_time: 0.0021 memory: 25074 loss: 0.0167 +2023/06/01 08:21:01 - mmengine - INFO - Epoch(train) [11][5300/5758] lr: 2.0050e-03 eta: 12:15:12 time: 0.8278 data_time: 0.0017 memory: 25074 loss: 0.0103 +2023/06/01 08:22:25 - mmengine - INFO - Epoch(train) [11][5400/5758] lr: 2.0050e-03 eta: 12:13:47 time: 0.8843 data_time: 0.0020 memory: 25074 loss: 0.0097 +2023/06/01 08:22:42 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_20230531_173252 +2023/06/01 08:23:49 - mmengine - INFO - Epoch(train) [11][5500/5758] lr: 2.0050e-03 eta: 12:12:22 time: 0.7989 data_time: 0.0020 memory: 25074 loss: 0.0108 +2023/06/01 08:25:13 - mmengine - INFO - Epoch(train) [11][5600/5758] lr: 2.0050e-03 eta: 12:10:57 time: 0.8645 data_time: 0.0017 memory: 25074 loss: 0.0130 +2023/06/01 08:26:37 - mmengine - INFO - Epoch(train) [11][5700/5758] lr: 2.0050e-03 eta: 12:09:32 time: 0.7989 data_time: 0.0015 memory: 25074 loss: 0.0108 +2023/06/01 08:27:24 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_20230531_173252 +2023/06/01 08:27:24 - mmengine - INFO - Saving checkpoint at 11 epochs +2023/06/01 08:27:44 - mmengine - INFO - Epoch(val) [11][16/16] accuracy/top1: 100.0000 single-label/precision_classwise: [100.0, 0.0] single-label/recall_classwise: [100.0, 0.0] single-label/f1-score_classwise: [100.0, 0.0] data_time: 0.2367 time: 0.6233 +2023/06/01 08:29:15 - mmengine - INFO - Epoch(train) [12][ 100/5758] lr: 1.6929e-03 eta: 12:07:24 time: 0.8847 data_time: 0.0017 memory: 25074 loss: 0.0126 +2023/06/01 08:30:41 - mmengine - INFO - Epoch(train) [12][ 200/5758] lr: 1.6929e-03 eta: 12:06:01 time: 0.8630 data_time: 0.0022 memory: 25074 loss: 0.0124 +2023/06/01 08:32:05 - mmengine - INFO - Epoch(train) [12][ 300/5758] lr: 1.6929e-03 eta: 12:04:36 time: 0.8486 data_time: 0.0022 memory: 25074 loss: 0.0069 +2023/06/01 08:33:28 - mmengine - INFO - Epoch(train) [12][ 400/5758] lr: 1.6929e-03 eta: 12:03:11 time: 0.8659 data_time: 0.0014 memory: 25074 loss: 0.0096 +2023/06/01 08:34:52 - mmengine - INFO - Epoch(train) [12][ 500/5758] lr: 1.6929e-03 eta: 12:01:46 time: 0.8526 data_time: 0.0019 memory: 25074 loss: 0.0052 +2023/06/01 08:36:17 - mmengine - INFO - Epoch(train) [12][ 600/5758] lr: 1.6929e-03 eta: 12:00:22 time: 0.8559 data_time: 0.0015 memory: 25074 loss: 0.5257 +2023/06/01 08:37:09 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_20230531_173252 +2023/06/01 08:37:41 - mmengine - INFO - Epoch(train) [12][ 700/5758] lr: 1.6929e-03 eta: 11:58:57 time: 0.8218 data_time: 0.0024 memory: 25074 loss: 0.0174 +2023/06/01 08:39:06 - mmengine - INFO - Epoch(train) [12][ 800/5758] lr: 1.6929e-03 eta: 11:57:34 time: 0.8379 data_time: 0.0019 memory: 25074 loss: 0.0184 +2023/06/01 08:40:31 - mmengine - INFO - Epoch(train) [12][ 900/5758] lr: 1.6929e-03 eta: 11:56:09 time: 0.8501 data_time: 0.0016 memory: 25074 loss: 0.0157 +2023/06/01 08:41:56 - mmengine - INFO - Epoch(train) [12][1000/5758] lr: 1.6929e-03 eta: 11:54:45 time: 0.8576 data_time: 0.0016 memory: 25074 loss: 0.0174 +2023/06/01 08:43:19 - mmengine - INFO - Epoch(train) [12][1100/5758] lr: 1.6929e-03 eta: 11:53:20 time: 0.8440 data_time: 0.0018 memory: 25074 loss: 0.0087 +2023/06/01 08:44:45 - mmengine - INFO - Epoch(train) [12][1200/5758] lr: 1.6929e-03 eta: 11:51:57 time: 0.8774 data_time: 0.0017 memory: 25074 loss: 0.0048 +2023/06/01 08:46:11 - mmengine - INFO - Epoch(train) [12][1300/5758] lr: 1.6929e-03 eta: 11:50:33 time: 0.8194 data_time: 0.0016 memory: 25074 loss: 0.0122 +2023/06/01 08:47:36 - mmengine - INFO - Epoch(train) [12][1400/5758] lr: 1.6929e-03 eta: 11:49:10 time: 0.8386 data_time: 0.0014 memory: 25074 loss: 0.0067 +2023/06/01 08:49:00 - mmengine - INFO - Epoch(train) [12][1500/5758] lr: 1.6929e-03 eta: 11:47:45 time: 0.8878 data_time: 0.0017 memory: 25074 loss: 0.0128 +2023/06/01 08:50:21 - mmengine - INFO - Epoch(train) [12][1600/5758] lr: 1.6929e-03 eta: 11:46:18 time: 0.7722 data_time: 0.0016 memory: 25074 loss: 0.0047 +2023/06/01 08:51:10 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_20230531_173252 +2023/06/01 08:51:42 - mmengine - INFO - Epoch(train) [12][1700/5758] lr: 1.6929e-03 eta: 11:44:51 time: 0.8705 data_time: 0.0017 memory: 25074 loss: 0.0095 +2023/06/01 08:53:08 - mmengine - INFO - Epoch(train) [12][1800/5758] lr: 1.6929e-03 eta: 11:43:28 time: 0.8121 data_time: 0.0017 memory: 25074 loss: 0.0117 +2023/06/01 08:54:34 - mmengine - INFO - Epoch(train) [12][1900/5758] lr: 1.6929e-03 eta: 11:42:04 time: 0.8418 data_time: 0.0023 memory: 25074 loss: 0.0113 +2023/06/01 08:55:59 - mmengine - INFO - Epoch(train) [12][2000/5758] lr: 1.6929e-03 eta: 11:40:41 time: 0.8386 data_time: 0.0016 memory: 25074 loss: 0.0080 +2023/06/01 08:57:24 - mmengine - INFO - Epoch(train) [12][2100/5758] lr: 1.6929e-03 eta: 11:39:17 time: 0.9014 data_time: 0.0015 memory: 25074 loss: 0.0139 +2023/06/01 08:58:48 - mmengine - INFO - Epoch(train) [12][2200/5758] lr: 1.6929e-03 eta: 11:37:52 time: 0.8340 data_time: 0.0015 memory: 25074 loss: 0.0080 +2023/06/01 09:00:13 - mmengine - INFO - Epoch(train) [12][2300/5758] lr: 1.6929e-03 eta: 11:36:29 time: 0.8477 data_time: 0.0015 memory: 25074 loss: 0.0104 +2023/06/01 09:01:37 - mmengine - INFO - Epoch(train) [12][2400/5758] lr: 1.6929e-03 eta: 11:35:04 time: 0.8631 data_time: 0.0014 memory: 25074 loss: 0.0102 +2023/06/01 09:03:02 - mmengine - INFO - Epoch(train) [12][2500/5758] lr: 1.6929e-03 eta: 11:33:40 time: 0.8444 data_time: 0.0019 memory: 25074 loss: 0.0047 +2023/06/01 09:04:26 - mmengine - INFO - Epoch(train) [12][2600/5758] lr: 1.6929e-03 eta: 11:32:15 time: 0.8416 data_time: 0.0015 memory: 25074 loss: 0.0079 +2023/06/01 09:05:19 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_20230531_173252 +2023/06/01 09:05:52 - mmengine - INFO - Epoch(train) [12][2700/5758] lr: 1.6929e-03 eta: 11:30:52 time: 0.8767 data_time: 0.0015 memory: 25074 loss: 0.0047 +2023/06/01 09:07:17 - mmengine - INFO - Epoch(train) [12][2800/5758] lr: 1.6929e-03 eta: 11:29:28 time: 0.7927 data_time: 0.0017 memory: 25074 loss: 0.6126 +2023/06/01 09:08:41 - mmengine - INFO - Epoch(train) [12][2900/5758] lr: 1.6929e-03 eta: 11:28:03 time: 0.7909 data_time: 0.0015 memory: 25074 loss: 0.6206 +2023/06/01 09:10:06 - mmengine - INFO - Epoch(train) [12][3000/5758] lr: 1.6929e-03 eta: 11:26:39 time: 0.8380 data_time: 0.0015 memory: 25074 loss: 0.5530 +2023/06/01 09:11:30 - mmengine - INFO - Epoch(train) [12][3100/5758] lr: 1.6929e-03 eta: 11:25:15 time: 0.8861 data_time: 0.0016 memory: 25074 loss: 0.4458 +2023/06/01 09:13:20 - mmengine - INFO - Epoch(train) [12][3200/5758] lr: 1.6929e-03 eta: 11:24:09 time: 2.0128 data_time: 0.0019 memory: 25074 loss: 0.3750 +2023/06/01 09:14:57 - mmengine - INFO - Epoch(train) [12][3300/5758] lr: 1.6929e-03 eta: 11:22:53 time: 0.8876 data_time: 0.0028 memory: 25074 loss: 0.3053 +2023/06/01 09:16:22 - mmengine - INFO - Epoch(train) [12][3400/5758] lr: 1.6929e-03 eta: 11:21:29 time: 0.8399 data_time: 0.0019 memory: 25074 loss: 0.2901 +2023/06/01 09:17:45 - mmengine - INFO - Epoch(train) [12][3500/5758] lr: 1.6929e-03 eta: 11:20:04 time: 0.8448 data_time: 0.0019 memory: 25074 loss: 0.3042 +2023/06/01 09:19:10 - mmengine - INFO - Epoch(train) [12][3600/5758] lr: 1.6929e-03 eta: 11:18:40 time: 0.8764 data_time: 0.0027 memory: 25074 loss: 0.2410 +2023/06/01 09:20:02 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_20230531_173252 +2023/06/01 09:20:35 - mmengine - INFO - Epoch(train) [12][3700/5758] lr: 1.6929e-03 eta: 11:17:16 time: 0.8524 data_time: 0.0022 memory: 25074 loss: 0.2603 +2023/06/01 09:22:00 - mmengine - INFO - Epoch(train) [12][3800/5758] lr: 1.6929e-03 eta: 11:15:52 time: 0.8215 data_time: 0.0017 memory: 25074 loss: 0.2199 +2023/06/01 09:23:24 - mmengine - INFO - Epoch(train) [12][3900/5758] lr: 1.6929e-03 eta: 11:14:27 time: 0.8601 data_time: 0.0017 memory: 25074 loss: 0.1977 +2023/06/01 09:24:50 - mmengine - INFO - Epoch(train) [12][4000/5758] lr: 1.6929e-03 eta: 11:13:03 time: 0.8335 data_time: 0.0017 memory: 25074 loss: 0.2058 +2023/06/01 09:26:14 - mmengine - INFO - Epoch(train) [12][4100/5758] lr: 1.6929e-03 eta: 11:11:39 time: 0.8400 data_time: 0.0015 memory: 25074 loss: 0.1739 +2023/06/01 09:27:40 - mmengine - INFO - Epoch(train) [12][4200/5758] lr: 1.6929e-03 eta: 11:10:15 time: 0.8414 data_time: 0.0017 memory: 25074 loss: 0.1664 +2023/06/01 09:29:05 - mmengine - INFO - Epoch(train) [12][4300/5758] lr: 1.6929e-03 eta: 11:08:51 time: 0.8928 data_time: 0.0016 memory: 25074 loss: 0.1331 +2023/06/01 09:30:29 - mmengine - INFO - Epoch(train) [12][4400/5758] lr: 1.6929e-03 eta: 11:07:27 time: 0.8252 data_time: 0.0023 memory: 25074 loss: 0.0725 +2023/06/01 09:31:54 - mmengine - INFO - Epoch(train) [12][4500/5758] lr: 1.6929e-03 eta: 11:06:03 time: 0.8944 data_time: 0.0017 memory: 25074 loss: 0.0562 +2023/06/01 09:33:32 - mmengine - INFO - Epoch(train) [12][4600/5758] lr: 1.6929e-03 eta: 11:04:48 time: 0.9269 data_time: 0.0022 memory: 25074 loss: 0.0484 +2023/06/01 09:34:23 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_20230531_173252 +2023/06/01 09:34:56 - mmengine - INFO - Epoch(train) [12][4700/5758] lr: 1.6929e-03 eta: 11:03:23 time: 0.8407 data_time: 0.0018 memory: 25074 loss: 0.0265 +2023/06/01 09:36:21 - mmengine - INFO - Epoch(train) [12][4800/5758] lr: 1.6929e-03 eta: 11:01:59 time: 0.8709 data_time: 0.0020 memory: 25074 loss: 0.0359 +2023/06/01 09:37:47 - mmengine - INFO - Epoch(train) [12][4900/5758] lr: 1.6929e-03 eta: 11:00:35 time: 0.8374 data_time: 0.0022 memory: 25074 loss: 0.0321 +2023/06/01 09:39:10 - mmengine - INFO - Epoch(train) [12][5000/5758] lr: 1.6929e-03 eta: 10:59:10 time: 0.8044 data_time: 0.0027 memory: 25074 loss: 0.0322 +2023/06/01 09:40:36 - mmengine - INFO - Epoch(train) [12][5100/5758] lr: 1.6929e-03 eta: 10:57:46 time: 0.8549 data_time: 0.0020 memory: 25074 loss: 0.0260 +2023/06/01 09:42:00 - mmengine - INFO - Epoch(train) [12][5200/5758] lr: 1.6929e-03 eta: 10:56:21 time: 0.8272 data_time: 0.0019 memory: 25074 loss: 0.0193 +2023/06/01 09:43:24 - mmengine - INFO - Epoch(train) [12][5300/5758] lr: 1.6929e-03 eta: 10:54:57 time: 0.9095 data_time: 0.0021 memory: 25074 loss: 0.0188 +2023/06/01 09:44:49 - mmengine - INFO - Epoch(train) [12][5400/5758] lr: 1.6929e-03 eta: 10:53:33 time: 0.7878 data_time: 0.0022 memory: 25074 loss: 0.0276 +2023/06/01 09:46:15 - mmengine - INFO - Epoch(train) [12][5500/5758] lr: 1.6929e-03 eta: 10:52:09 time: 0.8668 data_time: 0.0023 memory: 25074 loss: 0.0198 +2023/06/01 09:47:39 - mmengine - INFO - Epoch(train) [12][5600/5758] lr: 1.6929e-03 eta: 10:50:45 time: 0.8528 data_time: 0.0029 memory: 25074 loss: 0.0221 +2023/06/01 09:48:30 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_20230531_173252 +2023/06/01 09:49:02 - mmengine - INFO - Epoch(train) [12][5700/5758] lr: 1.6929e-03 eta: 10:49:19 time: 0.8659 data_time: 0.0022 memory: 25074 loss: 0.0213 +2023/06/01 09:49:51 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_20230531_173252 +2023/06/01 09:49:51 - mmengine - INFO - Saving checkpoint at 12 epochs +2023/06/01 09:50:10 - mmengine - INFO - Epoch(val) [12][16/16] accuracy/top1: 100.0000 single-label/precision_classwise: [100.0, 0.0] single-label/recall_classwise: [100.0, 0.0] single-label/f1-score_classwise: [100.0, 0.0] data_time: 0.2396 time: 0.6476 +2023/06/01 09:51:41 - mmengine - INFO - Epoch(train) [13][ 100/5758] lr: 1.3885e-03 eta: 10:47:10 time: 0.8154 data_time: 0.0022 memory: 25074 loss: 0.0151 +2023/06/01 09:53:04 - mmengine - INFO - Epoch(train) [13][ 200/5758] lr: 1.3885e-03 eta: 10:45:44 time: 0.8094 data_time: 0.0023 memory: 25074 loss: 0.0128 +2023/06/01 09:54:28 - mmengine - INFO - Epoch(train) [13][ 300/5758] lr: 1.3885e-03 eta: 10:44:20 time: 0.8432 data_time: 0.0028 memory: 25074 loss: 0.0146 +2023/06/01 09:55:54 - mmengine - INFO - Epoch(train) [13][ 400/5758] lr: 1.3885e-03 eta: 10:42:56 time: 0.8767 data_time: 0.0020 memory: 25074 loss: 0.0190 +2023/06/01 09:57:20 - mmengine - INFO - Epoch(train) [13][ 500/5758] lr: 1.3885e-03 eta: 10:41:32 time: 0.8396 data_time: 0.0025 memory: 25074 loss: 0.0278 +2023/06/01 09:58:45 - mmengine - INFO - Epoch(train) [13][ 600/5758] lr: 1.3885e-03 eta: 10:40:08 time: 0.8231 data_time: 0.0019 memory: 25074 loss: 0.0105 +2023/06/01 10:00:11 - mmengine - INFO - Epoch(train) [13][ 700/5758] lr: 1.3885e-03 eta: 10:38:45 time: 0.8371 data_time: 0.0017 memory: 25074 loss: 0.0100 +2023/06/01 10:01:38 - mmengine - INFO - Epoch(train) [13][ 800/5758] lr: 1.3885e-03 eta: 10:37:22 time: 0.8564 data_time: 0.0022 memory: 25074 loss: 0.0084 +2023/06/01 10:03:03 - mmengine - INFO - Epoch(train) [13][ 900/5758] lr: 1.3885e-03 eta: 10:35:58 time: 0.9230 data_time: 0.0022 memory: 25074 loss: 0.0126 +2023/06/01 10:03:08 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_20230531_173252 +2023/06/01 10:04:28 - mmengine - INFO - Epoch(train) [13][1000/5758] lr: 1.3885e-03 eta: 10:34:34 time: 0.8306 data_time: 0.0021 memory: 25074 loss: 0.0064 +2023/06/01 10:05:52 - mmengine - INFO - Epoch(train) [13][1100/5758] lr: 1.3885e-03 eta: 10:33:09 time: 0.8508 data_time: 0.0021 memory: 25074 loss: 0.0105 +2023/06/01 10:07:18 - mmengine - INFO - Epoch(train) [13][1200/5758] lr: 1.3885e-03 eta: 10:31:45 time: 0.8331 data_time: 0.0020 memory: 25074 loss: 0.0174 +2023/06/01 10:08:45 - mmengine - INFO - Epoch(train) [13][1300/5758] lr: 1.3885e-03 eta: 10:30:22 time: 0.8436 data_time: 0.0019 memory: 25074 loss: 0.0175 +2023/06/01 10:10:10 - mmengine - INFO - Epoch(train) [13][1400/5758] lr: 1.3885e-03 eta: 10:28:58 time: 0.8395 data_time: 0.0019 memory: 25074 loss: 0.0113 +2023/06/01 10:11:35 - mmengine - INFO - Epoch(train) [13][1500/5758] lr: 1.3885e-03 eta: 10:27:34 time: 0.8369 data_time: 0.0022 memory: 25074 loss: 0.0235 +2023/06/01 10:13:01 - mmengine - INFO - Epoch(train) [13][1600/5758] lr: 1.3885e-03 eta: 10:26:10 time: 0.8792 data_time: 0.0026 memory: 25074 loss: 0.0064 +2023/06/01 10:14:26 - mmengine - INFO - Epoch(train) [13][1700/5758] lr: 1.3885e-03 eta: 10:24:46 time: 0.8195 data_time: 0.0022 memory: 25074 loss: 0.0120 +2023/06/01 10:15:53 - mmengine - INFO - Epoch(train) [13][1800/5758] lr: 1.3885e-03 eta: 10:23:23 time: 0.8364 data_time: 0.0020 memory: 25074 loss: 0.0083 +2023/06/01 10:17:18 - mmengine - INFO - Epoch(train) [13][1900/5758] lr: 1.3885e-03 eta: 10:21:59 time: 0.8775 data_time: 0.0026 memory: 25074 loss: 0.0128 +2023/06/01 10:17:22 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_20230531_173252 +2023/06/01 10:18:43 - mmengine - INFO - Epoch(train) [13][2000/5758] lr: 1.3885e-03 eta: 10:20:35 time: 0.8430 data_time: 0.0021 memory: 25074 loss: 0.0170 +2023/06/01 10:20:09 - mmengine - INFO - Epoch(train) [13][2100/5758] lr: 1.3885e-03 eta: 10:19:11 time: 0.8354 data_time: 0.0020 memory: 25074 loss: 0.0083 +2023/06/01 10:21:34 - mmengine - INFO - Epoch(train) [13][2200/5758] lr: 1.3885e-03 eta: 10:17:47 time: 0.8647 data_time: 0.0025 memory: 25074 loss: 0.0133 +2023/06/01 10:23:00 - mmengine - INFO - Epoch(train) [13][2300/5758] lr: 1.3885e-03 eta: 10:16:23 time: 0.8228 data_time: 0.0021 memory: 25074 loss: 0.0046 +2023/06/01 10:24:24 - mmengine - INFO - Epoch(train) [13][2400/5758] lr: 1.3885e-03 eta: 10:14:59 time: 0.8153 data_time: 0.0015 memory: 25074 loss: 0.0054 +2023/06/01 10:25:44 - mmengine - INFO - Epoch(train) [13][2500/5758] lr: 1.3885e-03 eta: 10:13:31 time: 0.8246 data_time: 0.0019 memory: 25074 loss: 0.0088 +2023/06/01 10:27:05 - mmengine - INFO - Epoch(train) [13][2600/5758] lr: 1.3885e-03 eta: 10:12:05 time: 0.7858 data_time: 0.0024 memory: 25074 loss: 0.6324 +2023/06/01 10:28:27 - mmengine - INFO - Epoch(train) [13][2700/5758] lr: 1.3885e-03 eta: 10:10:39 time: 0.8469 data_time: 0.0021 memory: 25074 loss: 0.4092 +2023/06/01 10:29:52 - mmengine - INFO - Epoch(train) [13][2800/5758] lr: 1.3885e-03 eta: 10:09:14 time: 0.8175 data_time: 0.0019 memory: 25074 loss: 0.3070 +2023/06/01 10:31:18 - mmengine - INFO - Epoch(train) [13][2900/5758] lr: 1.3885e-03 eta: 10:07:51 time: 0.8586 data_time: 0.0018 memory: 25074 loss: 0.2704 +2023/06/01 10:31:22 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_20230531_173252 +2023/06/01 10:32:42 - mmengine - INFO - Epoch(train) [13][3000/5758] lr: 1.3885e-03 eta: 10:06:26 time: 0.8356 data_time: 0.0023 memory: 25074 loss: 0.2208 +2023/06/01 10:34:06 - mmengine - INFO - Epoch(train) [13][3100/5758] lr: 1.3885e-03 eta: 10:05:02 time: 0.8541 data_time: 0.0024 memory: 25074 loss: 0.1664 +2023/06/01 10:35:33 - mmengine - INFO - Epoch(train) [13][3200/5758] lr: 1.3885e-03 eta: 10:03:39 time: 0.8588 data_time: 0.0017 memory: 25074 loss: 0.0688 +2023/06/01 10:37:00 - mmengine - INFO - Epoch(train) [13][3300/5758] lr: 1.3885e-03 eta: 10:02:15 time: 0.8716 data_time: 0.0018 memory: 25074 loss: 0.0412 +2023/06/01 10:38:26 - mmengine - INFO - Epoch(train) [13][3400/5758] lr: 1.3885e-03 eta: 10:00:52 time: 0.8019 data_time: 0.0023 memory: 25074 loss: 0.0353 +2023/06/01 10:39:52 - mmengine - INFO - Epoch(train) [13][3500/5758] lr: 1.3885e-03 eta: 9:59:28 time: 0.8152 data_time: 0.0021 memory: 25074 loss: 0.0256 +2023/06/01 10:41:18 - mmengine - INFO - Epoch(train) [13][3600/5758] lr: 1.3885e-03 eta: 9:58:05 time: 0.8970 data_time: 0.0026 memory: 25074 loss: 0.0196 +2023/06/01 10:42:43 - mmengine - INFO - Epoch(train) [13][3700/5758] lr: 1.3885e-03 eta: 9:56:40 time: 0.8235 data_time: 0.0025 memory: 25074 loss: 0.0193 +2023/06/01 10:44:08 - mmengine - INFO - Epoch(train) [13][3800/5758] lr: 1.3885e-03 eta: 9:55:16 time: 0.8693 data_time: 0.0029 memory: 25074 loss: 0.0148 +2023/06/01 10:45:33 - mmengine - INFO - Epoch(train) [13][3900/5758] lr: 1.3885e-03 eta: 9:53:51 time: 0.9129 data_time: 0.0024 memory: 25074 loss: 0.0473 +2023/06/01 10:45:37 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_20230531_173252 +2023/06/01 10:46:59 - mmengine - INFO - Epoch(train) [13][4000/5758] lr: 1.3885e-03 eta: 9:52:28 time: 0.9482 data_time: 0.0023 memory: 25074 loss: 0.2544 +2023/06/01 10:48:26 - mmengine - INFO - Epoch(train) [13][4100/5758] lr: 1.3885e-03 eta: 9:51:05 time: 0.8873 data_time: 0.0025 memory: 25074 loss: 0.0586 +2023/06/01 10:49:53 - mmengine - INFO - Epoch(train) [13][4200/5758] lr: 1.3885e-03 eta: 9:49:42 time: 0.8858 data_time: 0.0029 memory: 25074 loss: 0.0363 +2023/06/01 10:51:18 - mmengine - INFO - Epoch(train) [13][4300/5758] lr: 1.3885e-03 eta: 9:48:18 time: 0.8361 data_time: 0.0020 memory: 25074 loss: 0.0183 +2023/06/01 10:52:44 - mmengine - INFO - Epoch(train) [13][4400/5758] lr: 1.3885e-03 eta: 9:46:54 time: 0.8416 data_time: 0.0018 memory: 25074 loss: 0.0225 +2023/06/01 10:54:12 - mmengine - INFO - Epoch(train) [13][4500/5758] lr: 1.3885e-03 eta: 9:45:31 time: 0.9435 data_time: 0.0022 memory: 25074 loss: 0.0200 +2023/06/01 10:55:39 - mmengine - INFO - Epoch(train) [13][4600/5758] lr: 1.3885e-03 eta: 9:44:08 time: 0.8584 data_time: 0.0022 memory: 25074 loss: 0.0063 +2023/06/01 10:57:06 - mmengine - INFO - Epoch(train) [13][4700/5758] lr: 1.3885e-03 eta: 9:42:45 time: 0.8500 data_time: 0.0020 memory: 25074 loss: 0.0158 +2023/06/01 10:58:32 - mmengine - INFO - Epoch(train) [13][4800/5758] lr: 1.3885e-03 eta: 9:41:21 time: 0.8819 data_time: 0.0025 memory: 25074 loss: 0.0109 +2023/06/01 10:59:57 - mmengine - INFO - Epoch(train) [13][4900/5758] lr: 1.3885e-03 eta: 9:39:57 time: 0.8574 data_time: 0.0019 memory: 25074 loss: 0.0097 +2023/06/01 11:00:02 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_20230531_173252 +2023/06/01 11:01:23 - mmengine - INFO - Epoch(train) [13][5000/5758] lr: 1.3885e-03 eta: 9:38:33 time: 0.8520 data_time: 0.0023 memory: 25074 loss: 0.0216 +2023/06/01 11:02:47 - mmengine - INFO - Epoch(train) [13][5100/5758] lr: 1.3885e-03 eta: 9:37:09 time: 0.9194 data_time: 0.0026 memory: 25074 loss: 0.0122 +2023/06/01 11:04:11 - mmengine - INFO - Epoch(train) [13][5200/5758] lr: 1.3885e-03 eta: 9:35:44 time: 0.8667 data_time: 0.0017 memory: 25074 loss: 0.0090 +2023/06/01 11:05:37 - mmengine - INFO - Epoch(train) [13][5300/5758] lr: 1.3885e-03 eta: 9:34:20 time: 0.8207 data_time: 0.0024 memory: 25074 loss: 0.0094 +2023/06/01 11:07:00 - mmengine - INFO - Epoch(train) [13][5400/5758] lr: 1.3885e-03 eta: 9:32:54 time: 0.8605 data_time: 0.0021 memory: 25074 loss: 0.0091 +2023/06/01 11:08:23 - mmengine - INFO - Epoch(train) [13][5500/5758] lr: 1.3885e-03 eta: 9:31:29 time: 0.8658 data_time: 0.0019 memory: 25074 loss: 0.0064 +2023/06/01 11:09:46 - mmengine - INFO - Epoch(train) [13][5600/5758] lr: 1.3885e-03 eta: 9:30:04 time: 0.8472 data_time: 0.0021 memory: 25074 loss: 0.0172 +2023/06/01 11:11:10 - mmengine - INFO - Epoch(train) [13][5700/5758] lr: 1.3885e-03 eta: 9:28:39 time: 0.8367 data_time: 0.0024 memory: 25074 loss: 0.0111 +2023/06/01 11:11:59 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_20230531_173252 +2023/06/01 11:11:59 - mmengine - INFO - Saving checkpoint at 13 epochs +2023/06/01 11:12:18 - mmengine - INFO - Epoch(val) [13][16/16] accuracy/top1: 99.9369 single-label/precision_classwise: [100.0, 0.0] single-label/recall_classwise: [99.93687438964844, 0.0] single-label/f1-score_classwise: [99.96842956542969, 0.0] data_time: 0.2643 time: 0.6520 +2023/06/01 11:13:50 - mmengine - INFO - Epoch(train) [14][ 100/5758] lr: 1.0993e-03 eta: 9:26:29 time: 0.8829 data_time: 0.0022 memory: 25074 loss: 0.0060 +2023/06/01 11:14:31 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_20230531_173252 +2023/06/01 11:15:17 - mmengine - INFO - Epoch(train) [14][ 200/5758] lr: 1.0993e-03 eta: 9:25:06 time: 0.9074 data_time: 0.0019 memory: 25074 loss: 0.0025 +2023/06/01 11:16:43 - mmengine - INFO - Epoch(train) [14][ 300/5758] lr: 1.0993e-03 eta: 9:23:42 time: 0.8835 data_time: 0.0026 memory: 25074 loss: 0.0047 +2023/06/01 11:18:06 - mmengine - INFO - Epoch(train) [14][ 400/5758] lr: 1.0993e-03 eta: 9:22:17 time: 0.8699 data_time: 0.0019 memory: 25074 loss: 0.0124 +2023/06/01 11:19:30 - mmengine - INFO - Epoch(train) [14][ 500/5758] lr: 1.0993e-03 eta: 9:20:52 time: 0.8357 data_time: 0.0020 memory: 25074 loss: 0.0090 +2023/06/01 11:20:54 - mmengine - INFO - Epoch(train) [14][ 600/5758] lr: 1.0993e-03 eta: 9:19:27 time: 0.8575 data_time: 0.0021 memory: 25074 loss: 0.0077 +2023/06/01 11:22:22 - mmengine - INFO - Epoch(train) [14][ 700/5758] lr: 1.0993e-03 eta: 9:18:04 time: 0.8786 data_time: 0.0017 memory: 25074 loss: 0.0072 +2023/06/01 11:23:49 - mmengine - INFO - Epoch(train) [14][ 800/5758] lr: 1.0993e-03 eta: 9:16:41 time: 0.9102 data_time: 0.0023 memory: 25074 loss: 0.0094 +2023/06/01 11:25:16 - mmengine - INFO - Epoch(train) [14][ 900/5758] lr: 1.0993e-03 eta: 9:15:18 time: 0.9484 data_time: 0.0021 memory: 25074 loss: 0.0063 +2023/06/01 11:26:44 - mmengine - INFO - Epoch(train) [14][1000/5758] lr: 1.0993e-03 eta: 9:13:55 time: 0.8318 data_time: 0.0021 memory: 25074 loss: 0.0025 +2023/06/01 11:28:10 - mmengine - INFO - Epoch(train) [14][1100/5758] lr: 1.0993e-03 eta: 9:12:31 time: 0.8653 data_time: 0.0024 memory: 25074 loss: 0.0106 +2023/06/01 11:28:52 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_20230531_173252 +2023/06/01 11:29:36 - mmengine - INFO - Epoch(train) [14][1200/5758] lr: 1.0993e-03 eta: 9:11:08 time: 0.8250 data_time: 0.0016 memory: 25074 loss: 0.0046 +2023/06/01 11:31:02 - mmengine - INFO - Epoch(train) [14][1300/5758] lr: 1.0993e-03 eta: 9:09:44 time: 0.8512 data_time: 0.0027 memory: 25074 loss: 0.0058 +2023/06/01 11:32:30 - mmengine - INFO - Epoch(train) [14][1400/5758] lr: 1.0993e-03 eta: 9:08:21 time: 0.9002 data_time: 0.0023 memory: 25074 loss: 0.0057 +2023/06/01 11:33:59 - mmengine - INFO - Epoch(train) [14][1500/5758] lr: 1.0993e-03 eta: 9:06:59 time: 0.8559 data_time: 0.0020 memory: 25074 loss: 0.0079 +2023/06/01 11:35:28 - mmengine - INFO - Epoch(train) [14][1600/5758] lr: 1.0993e-03 eta: 9:05:36 time: 0.8499 data_time: 0.0021 memory: 25074 loss: 0.0034 +2023/06/01 11:36:57 - mmengine - INFO - Epoch(train) [14][1700/5758] lr: 1.0993e-03 eta: 9:04:14 time: 0.8800 data_time: 0.0027 memory: 25074 loss: 0.0107 +2023/06/01 11:38:26 - mmengine - INFO - Epoch(train) [14][1800/5758] lr: 1.0993e-03 eta: 9:02:51 time: 0.8587 data_time: 0.0024 memory: 25074 loss: 0.0053 +2023/06/01 11:39:54 - mmengine - INFO - Epoch(train) [14][1900/5758] lr: 1.0993e-03 eta: 9:01:28 time: 0.9227 data_time: 0.0025 memory: 25074 loss: 0.0077 +2023/06/01 11:41:20 - mmengine - INFO - Epoch(train) [14][2000/5758] lr: 1.0993e-03 eta: 9:00:04 time: 0.8529 data_time: 0.0022 memory: 25074 loss: 0.0114 +2023/06/01 11:42:46 - mmengine - INFO - Epoch(train) [14][2100/5758] lr: 1.0993e-03 eta: 8:58:41 time: 0.8198 data_time: 0.0023 memory: 25074 loss: 0.0048 +2023/06/01 11:43:24 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_20230531_173252 +2023/06/01 11:44:08 - mmengine - INFO - Epoch(train) [14][2200/5758] lr: 1.0993e-03 eta: 8:57:15 time: 0.7742 data_time: 0.0015 memory: 25074 loss: 0.0091 +2023/06/01 11:45:26 - mmengine - INFO - Epoch(train) [14][2300/5758] lr: 1.0993e-03 eta: 8:55:47 time: 0.8021 data_time: 0.0015 memory: 25074 loss: 0.0048 +2023/06/01 11:46:49 - mmengine - INFO - Epoch(train) [14][2400/5758] lr: 1.0993e-03 eta: 8:54:22 time: 0.7940 data_time: 0.0020 memory: 25074 loss: 0.0043 +2023/06/01 11:48:12 - mmengine - INFO - Epoch(train) [14][2500/5758] lr: 1.0993e-03 eta: 8:52:56 time: 0.8532 data_time: 0.0019 memory: 25074 loss: 0.0065 +2023/06/01 11:49:36 - mmengine - INFO - Epoch(train) [14][2600/5758] lr: 1.0993e-03 eta: 8:51:31 time: 0.7793 data_time: 0.0014 memory: 25074 loss: 0.0092 +2023/06/01 11:50:58 - mmengine - INFO - Epoch(train) [14][2700/5758] lr: 1.0993e-03 eta: 8:50:06 time: 0.8605 data_time: 0.0014 memory: 25074 loss: 0.0042 +2023/06/01 11:52:23 - mmengine - INFO - Epoch(train) [14][2800/5758] lr: 1.0993e-03 eta: 8:48:41 time: 0.8461 data_time: 0.0014 memory: 25074 loss: 0.0034 +2023/06/01 11:53:48 - mmengine - INFO - Epoch(train) [14][2900/5758] lr: 1.0993e-03 eta: 8:47:17 time: 0.8399 data_time: 0.0024 memory: 25074 loss: 0.0122 +2023/06/01 11:55:10 - mmengine - INFO - Epoch(train) [14][3000/5758] lr: 1.0993e-03 eta: 8:45:51 time: 0.8112 data_time: 0.0017 memory: 25074 loss: 0.0035 +2023/06/01 11:56:33 - mmengine - INFO - Epoch(train) [14][3100/5758] lr: 1.0993e-03 eta: 8:44:26 time: 0.8249 data_time: 0.0018 memory: 25074 loss: 0.0011 +2023/06/01 11:57:12 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_20230531_173252 +2023/06/01 11:57:56 - mmengine - INFO - Epoch(train) [14][3200/5758] lr: 1.0993e-03 eta: 8:43:00 time: 0.8364 data_time: 0.0025 memory: 25074 loss: 0.0101 +2023/06/01 11:59:17 - mmengine - INFO - Epoch(train) [14][3300/5758] lr: 1.0993e-03 eta: 8:41:34 time: 0.8085 data_time: 0.0026 memory: 25074 loss: 0.0073 +2023/06/01 12:00:38 - mmengine - INFO - Epoch(train) [14][3400/5758] lr: 1.0993e-03 eta: 8:40:08 time: 0.8355 data_time: 0.0023 memory: 25074 loss: 0.0097 +2023/06/01 12:01:58 - mmengine - INFO - Epoch(train) [14][3500/5758] lr: 1.0993e-03 eta: 8:38:41 time: 0.8418 data_time: 0.0023 memory: 25074 loss: 0.0069 +2023/06/01 12:03:21 - mmengine - INFO - Epoch(train) [14][3600/5758] lr: 1.0993e-03 eta: 8:37:16 time: 0.7828 data_time: 0.0019 memory: 25074 loss: 0.0057 +2023/06/01 12:04:42 - mmengine - INFO - Epoch(train) [14][3700/5758] lr: 1.0993e-03 eta: 8:35:50 time: 0.8424 data_time: 0.0024 memory: 25074 loss: 0.0085 +2023/06/01 12:06:04 - mmengine - INFO - Epoch(train) [14][3800/5758] lr: 1.0993e-03 eta: 8:34:24 time: 0.8067 data_time: 0.0020 memory: 25074 loss: 0.0078 +2023/06/01 12:07:26 - mmengine - INFO - Epoch(train) [14][3900/5758] lr: 1.0993e-03 eta: 8:32:58 time: 0.8274 data_time: 0.0022 memory: 25074 loss: 0.0046 +2023/06/01 12:08:46 - mmengine - INFO - Epoch(train) [14][4000/5758] lr: 1.0993e-03 eta: 8:31:31 time: 0.7971 data_time: 0.0017 memory: 25074 loss: 0.0110 +2023/06/01 12:10:09 - mmengine - INFO - Epoch(train) [14][4100/5758] lr: 1.0993e-03 eta: 8:30:06 time: 0.8197 data_time: 0.0022 memory: 25074 loss: 0.0107 +2023/06/01 12:10:46 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_20230531_173252 +2023/06/01 12:11:31 - mmengine - INFO - Epoch(train) [14][4200/5758] lr: 1.0993e-03 eta: 8:28:40 time: 0.8274 data_time: 0.0026 memory: 25074 loss: 0.0026 +2023/06/01 12:12:54 - mmengine - INFO - Epoch(train) [14][4300/5758] lr: 1.0993e-03 eta: 8:27:15 time: 0.8829 data_time: 0.0019 memory: 25074 loss: 0.0069 +2023/06/01 12:14:16 - mmengine - INFO - Epoch(train) [14][4400/5758] lr: 1.0993e-03 eta: 8:25:49 time: 0.8273 data_time: 0.0023 memory: 25074 loss: 0.0029 +2023/06/01 12:15:38 - mmengine - INFO - Epoch(train) [14][4500/5758] lr: 1.0993e-03 eta: 8:24:24 time: 0.8197 data_time: 0.0020 memory: 25074 loss: 0.0032 +2023/06/01 12:17:00 - mmengine - INFO - Epoch(train) [14][4600/5758] lr: 1.0993e-03 eta: 8:22:58 time: 0.8215 data_time: 0.0019 memory: 25074 loss: 0.0017 +2023/06/01 12:18:24 - mmengine - INFO - Epoch(train) [14][4700/5758] lr: 1.0993e-03 eta: 8:21:33 time: 0.8593 data_time: 0.0021 memory: 25074 loss: 0.0084 +2023/06/01 12:19:48 - mmengine - INFO - Epoch(train) [14][4800/5758] lr: 1.0993e-03 eta: 8:20:08 time: 0.8848 data_time: 0.0021 memory: 25074 loss: 0.0056 +2023/06/01 12:21:12 - mmengine - INFO - Epoch(train) [14][4900/5758] lr: 1.0993e-03 eta: 8:18:44 time: 0.8677 data_time: 0.0019 memory: 25074 loss: 0.0101 +2023/06/01 12:22:35 - mmengine - INFO - Epoch(train) [14][5000/5758] lr: 1.0993e-03 eta: 8:17:19 time: 0.8044 data_time: 0.0019 memory: 25074 loss: 0.0105 +2023/06/01 12:24:01 - mmengine - INFO - Epoch(train) [14][5100/5758] lr: 1.0993e-03 eta: 8:15:55 time: 0.8106 data_time: 0.0017 memory: 25074 loss: 0.0049 +2023/06/01 12:24:39 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_20230531_173252 +2023/06/01 12:25:22 - mmengine - INFO - Epoch(train) [14][5200/5758] lr: 1.0993e-03 eta: 8:14:29 time: 0.7801 data_time: 0.0020 memory: 25074 loss: 0.0027 +2023/06/01 12:26:45 - mmengine - INFO - Epoch(train) [14][5300/5758] lr: 1.0993e-03 eta: 8:13:03 time: 0.8744 data_time: 0.0016 memory: 25074 loss: 0.0039 +2023/06/01 12:28:06 - mmengine - INFO - Epoch(train) [14][5400/5758] lr: 1.0993e-03 eta: 8:11:38 time: 0.8251 data_time: 0.0015 memory: 25074 loss: 0.0070 +2023/06/01 12:29:28 - mmengine - INFO - Epoch(train) [14][5500/5758] lr: 1.0993e-03 eta: 8:10:12 time: 0.8248 data_time: 0.0020 memory: 25074 loss: 0.0071 +2023/06/01 12:30:50 - mmengine - INFO - Epoch(train) [14][5600/5758] lr: 1.0993e-03 eta: 8:08:46 time: 0.8233 data_time: 0.0022 memory: 25074 loss: 0.0075 +2023/06/01 12:32:11 - mmengine - INFO - Epoch(train) [14][5700/5758] lr: 1.0993e-03 eta: 8:07:20 time: 0.7905 data_time: 0.0017 memory: 25074 loss: 0.0051 +2023/06/01 12:32:59 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_20230531_173252 +2023/06/01 12:32:59 - mmengine - INFO - Saving checkpoint at 14 epochs +2023/06/01 12:33:17 - mmengine - INFO - Epoch(val) [14][16/16] accuracy/top1: 99.9874 single-label/precision_classwise: [100.0, 0.0] single-label/recall_classwise: [99.98737335205078, 0.0] single-label/f1-score_classwise: [99.99369049072266, 0.0] data_time: 0.2449 time: 0.6382 +2023/06/01 12:34:47 - mmengine - INFO - Epoch(train) [15][ 100/5758] lr: 8.3237e-04 eta: 8:05:08 time: 0.8145 data_time: 0.0021 memory: 25074 loss: 0.0082 +2023/06/01 12:36:08 - mmengine - INFO - Epoch(train) [15][ 200/5758] lr: 8.3237e-04 eta: 8:03:42 time: 0.7930 data_time: 0.0017 memory: 25074 loss: 0.0009 +2023/06/01 12:37:30 - mmengine - INFO - Epoch(train) [15][ 300/5758] lr: 8.3237e-04 eta: 8:02:17 time: 0.8182 data_time: 0.0017 memory: 25074 loss: 0.0074 +2023/06/01 12:38:41 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_20230531_173252 +2023/06/01 12:38:50 - mmengine - INFO - Epoch(train) [15][ 400/5758] lr: 8.3237e-04 eta: 8:00:50 time: 0.7837 data_time: 0.0019 memory: 25074 loss: 0.0030 +2023/06/01 12:40:13 - mmengine - INFO - Epoch(train) [15][ 500/5758] lr: 8.3237e-04 eta: 7:59:25 time: 0.8062 data_time: 0.0017 memory: 25074 loss: 0.0063 +2023/06/01 12:41:33 - mmengine - INFO - Epoch(train) [15][ 600/5758] lr: 8.3237e-04 eta: 7:57:59 time: 0.7703 data_time: 0.0016 memory: 25074 loss: 0.0045 +2023/06/01 12:42:54 - mmengine - INFO - Epoch(train) [15][ 700/5758] lr: 8.3237e-04 eta: 7:56:33 time: 0.8155 data_time: 0.0020 memory: 25074 loss: 0.0062 +2023/06/01 12:44:16 - mmengine - INFO - Epoch(train) [15][ 800/5758] lr: 8.3237e-04 eta: 7:55:08 time: 0.8437 data_time: 0.0020 memory: 25074 loss: 0.0076 +2023/06/01 12:45:39 - mmengine - INFO - Epoch(train) [15][ 900/5758] lr: 8.3237e-04 eta: 7:53:42 time: 0.7929 data_time: 0.0017 memory: 25074 loss: 0.0222 +2023/06/01 12:47:00 - mmengine - INFO - Epoch(train) [15][1000/5758] lr: 8.3237e-04 eta: 7:52:16 time: 0.7770 data_time: 0.0016 memory: 25074 loss: 0.0040 +2023/06/01 12:48:22 - mmengine - INFO - Epoch(train) [15][1100/5758] lr: 8.3237e-04 eta: 7:50:51 time: 0.8139 data_time: 0.0025 memory: 25074 loss: 0.0044 +2023/06/01 12:49:45 - mmengine - INFO - Epoch(train) [15][1200/5758] lr: 8.3237e-04 eta: 7:49:26 time: 0.8529 data_time: 0.0018 memory: 25074 loss: 0.0032 +2023/06/01 12:51:06 - mmengine - INFO - Epoch(train) [15][1300/5758] lr: 8.3237e-04 eta: 7:48:00 time: 0.8502 data_time: 0.0014 memory: 25074 loss: 0.0038 +2023/06/01 12:52:19 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_20230531_173252 +2023/06/01 12:52:29 - mmengine - INFO - Epoch(train) [15][1400/5758] lr: 8.3237e-04 eta: 7:46:35 time: 0.7973 data_time: 0.0014 memory: 25074 loss: 0.0046 +2023/06/01 12:53:48 - mmengine - INFO - Epoch(train) [15][1500/5758] lr: 8.3237e-04 eta: 7:45:08 time: 0.7630 data_time: 0.0021 memory: 25074 loss: 0.0030 +2023/06/01 12:55:10 - mmengine - INFO - Epoch(train) [15][1600/5758] lr: 8.3237e-04 eta: 7:43:43 time: 0.8240 data_time: 0.0017 memory: 25074 loss: 0.0034 +2023/06/01 12:56:31 - mmengine - INFO - Epoch(train) [15][1700/5758] lr: 8.3237e-04 eta: 7:42:17 time: 0.8517 data_time: 0.0015 memory: 25074 loss: 0.0033 +2023/06/01 12:57:52 - mmengine - INFO - Epoch(train) [15][1800/5758] lr: 8.3237e-04 eta: 7:40:51 time: 0.8344 data_time: 0.0016 memory: 25074 loss: 0.0069 +2023/06/01 12:59:15 - mmengine - INFO - Epoch(train) [15][1900/5758] lr: 8.3237e-04 eta: 7:39:26 time: 0.8258 data_time: 0.0026 memory: 25074 loss: 0.0046 +2023/06/01 13:00:35 - mmengine - INFO - Epoch(train) [15][2000/5758] lr: 8.3237e-04 eta: 7:38:00 time: 0.7366 data_time: 0.0016 memory: 25074 loss: 0.0028 +2023/06/01 13:01:58 - mmengine - INFO - Epoch(train) [15][2100/5758] lr: 8.3237e-04 eta: 7:36:35 time: 0.9010 data_time: 0.0016 memory: 25074 loss: 0.0041 +2023/06/01 13:03:22 - mmengine - INFO - Epoch(train) [15][2200/5758] lr: 8.3237e-04 eta: 7:35:11 time: 0.8672 data_time: 0.0022 memory: 25074 loss: 0.0032 +2023/06/01 13:04:46 - mmengine - INFO - Epoch(train) [15][2300/5758] lr: 8.3237e-04 eta: 7:33:46 time: 0.8019 data_time: 0.0015 memory: 25074 loss: 0.0036 +2023/06/01 13:05:58 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_20230531_173252 +2023/06/01 13:06:08 - mmengine - INFO - Epoch(train) [15][2400/5758] lr: 8.3237e-04 eta: 7:32:20 time: 0.8077 data_time: 0.0016 memory: 25074 loss: 0.0034 +2023/06/01 13:07:31 - mmengine - INFO - Epoch(train) [15][2500/5758] lr: 8.3237e-04 eta: 7:30:55 time: 0.7925 data_time: 0.0022 memory: 25074 loss: 0.0026 +2023/06/01 13:08:54 - mmengine - INFO - Epoch(train) [15][2600/5758] lr: 8.3237e-04 eta: 7:29:30 time: 0.8363 data_time: 0.0016 memory: 25074 loss: 0.0026 +2023/06/01 13:10:17 - mmengine - INFO - Epoch(train) [15][2700/5758] lr: 8.3237e-04 eta: 7:28:05 time: 0.8434 data_time: 0.0026 memory: 25074 loss: 0.0013 +2023/06/01 13:11:41 - mmengine - INFO - Epoch(train) [15][2800/5758] lr: 8.3237e-04 eta: 7:26:41 time: 0.8503 data_time: 0.0015 memory: 25074 loss: 0.0040 +2023/06/01 13:13:04 - mmengine - INFO - Epoch(train) [15][2900/5758] lr: 8.3237e-04 eta: 7:25:16 time: 0.8352 data_time: 0.0020 memory: 25074 loss: 0.0017 +2023/06/01 13:14:28 - mmengine - INFO - Epoch(train) [15][3000/5758] lr: 8.3237e-04 eta: 7:23:51 time: 0.8708 data_time: 0.0016 memory: 25074 loss: 0.0037 +2023/06/01 13:15:53 - mmengine - INFO - Epoch(train) [15][3100/5758] lr: 8.3237e-04 eta: 7:22:27 time: 0.8685 data_time: 0.0019 memory: 25074 loss: 0.0023 +2023/06/01 13:17:27 - mmengine - INFO - Epoch(train) [15][3200/5758] lr: 8.3237e-04 eta: 7:21:06 time: 0.9061 data_time: 0.0015 memory: 25074 loss: 0.0080 +2023/06/01 13:18:49 - mmengine - INFO - Epoch(train) [15][3300/5758] lr: 8.3237e-04 eta: 7:19:41 time: 0.7958 data_time: 0.0020 memory: 25074 loss: 0.0023 +2023/06/01 13:20:05 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_20230531_173252 +2023/06/01 13:20:15 - mmengine - INFO - Epoch(train) [15][3400/5758] lr: 8.3237e-04 eta: 7:18:17 time: 0.8611 data_time: 0.0022 memory: 25074 loss: 0.0024 +2023/06/01 13:21:39 - mmengine - INFO - Epoch(train) [15][3500/5758] lr: 8.3237e-04 eta: 7:16:52 time: 0.8284 data_time: 0.0022 memory: 25074 loss: 0.0028 +2023/06/01 13:23:02 - mmengine - INFO - Epoch(train) [15][3600/5758] lr: 8.3237e-04 eta: 7:15:28 time: 0.8111 data_time: 0.0021 memory: 25074 loss: 0.0021 +2023/06/01 13:24:27 - mmengine - INFO - Epoch(train) [15][3700/5758] lr: 8.3237e-04 eta: 7:14:03 time: 0.8174 data_time: 0.0022 memory: 25074 loss: 0.0019 +2023/06/01 13:25:50 - mmengine - INFO - Epoch(train) [15][3800/5758] lr: 8.3237e-04 eta: 7:12:39 time: 0.8291 data_time: 0.0021 memory: 25074 loss: 0.0028 +2023/06/01 13:27:17 - mmengine - INFO - Epoch(train) [15][3900/5758] lr: 8.3237e-04 eta: 7:11:15 time: 0.8151 data_time: 0.0024 memory: 25074 loss: 0.0019 +2023/06/01 13:28:44 - mmengine - INFO - Epoch(train) [15][4000/5758] lr: 8.3237e-04 eta: 7:09:51 time: 0.8477 data_time: 0.0020 memory: 25074 loss: 0.0044 +2023/06/01 13:30:11 - mmengine - INFO - Epoch(train) [15][4100/5758] lr: 8.3237e-04 eta: 7:08:28 time: 0.8792 data_time: 0.0021 memory: 25074 loss: 0.0070 +2023/06/01 13:31:33 - mmengine - INFO - Epoch(train) [15][4200/5758] lr: 8.3237e-04 eta: 7:07:03 time: 0.8185 data_time: 0.0025 memory: 25074 loss: 0.0017 +2023/06/01 13:32:57 - mmengine - INFO - Epoch(train) [15][4300/5758] lr: 8.3237e-04 eta: 7:05:38 time: 0.8046 data_time: 0.0025 memory: 25074 loss: 0.0031 +2023/06/01 13:34:11 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_20230531_173252 +2023/06/01 13:34:21 - mmengine - INFO - Epoch(train) [15][4400/5758] lr: 8.3237e-04 eta: 7:04:13 time: 0.8695 data_time: 0.0018 memory: 25074 loss: 0.0020 +2023/06/01 13:35:45 - mmengine - INFO - Epoch(train) [15][4500/5758] lr: 8.3237e-04 eta: 7:02:49 time: 0.7930 data_time: 0.0016 memory: 25074 loss: 0.0016 +2023/06/01 13:37:10 - mmengine - INFO - Epoch(train) [15][4600/5758] lr: 8.3237e-04 eta: 7:01:25 time: 0.9227 data_time: 0.0021 memory: 25074 loss: 0.0031 +2023/06/01 13:38:32 - mmengine - INFO - Epoch(train) [15][4700/5758] lr: 8.3237e-04 eta: 6:59:59 time: 0.8556 data_time: 0.0016 memory: 25074 loss: 0.0043 +2023/06/01 13:39:57 - mmengine - INFO - Epoch(train) [15][4800/5758] lr: 8.3237e-04 eta: 6:58:35 time: 0.8675 data_time: 0.0017 memory: 25074 loss: 0.0056 +2023/06/01 13:41:21 - mmengine - INFO - Epoch(train) [15][4900/5758] lr: 8.3237e-04 eta: 6:57:10 time: 0.8150 data_time: 0.0017 memory: 25074 loss: 0.0049 +2023/06/01 13:42:45 - mmengine - INFO - Epoch(train) [15][5000/5758] lr: 8.3237e-04 eta: 6:55:46 time: 0.9148 data_time: 0.0023 memory: 25074 loss: 0.0025 +2023/06/01 13:44:08 - mmengine - INFO - Epoch(train) [15][5100/5758] lr: 8.3237e-04 eta: 6:54:21 time: 0.8296 data_time: 0.0019 memory: 25074 loss: 0.0025 +2023/06/01 13:45:33 - mmengine - INFO - Epoch(train) [15][5200/5758] lr: 8.3237e-04 eta: 6:52:57 time: 0.8046 data_time: 0.0030 memory: 25074 loss: 0.0059 +2023/06/01 13:46:56 - mmengine - INFO - Epoch(train) [15][5300/5758] lr: 8.3237e-04 eta: 6:51:32 time: 0.8163 data_time: 0.0021 memory: 25074 loss: 0.0048 +2023/06/01 13:48:10 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_20230531_173252 +2023/06/01 13:48:20 - mmengine - INFO - Epoch(train) [15][5400/5758] lr: 8.3237e-04 eta: 6:50:07 time: 0.8333 data_time: 0.0026 memory: 25074 loss: 0.0042 +2023/06/01 13:49:42 - mmengine - INFO - Epoch(train) [15][5500/5758] lr: 8.3237e-04 eta: 6:48:42 time: 0.8160 data_time: 0.0020 memory: 25074 loss: 0.0024 +2023/06/01 13:51:05 - mmengine - INFO - Epoch(train) [15][5600/5758] lr: 8.3237e-04 eta: 6:47:17 time: 0.8412 data_time: 0.0019 memory: 25074 loss: 0.0033 +2023/06/01 13:52:27 - mmengine - INFO - Epoch(train) [15][5700/5758] lr: 8.3237e-04 eta: 6:45:52 time: 0.8080 data_time: 0.0018 memory: 25074 loss: 0.0039 +2023/06/01 13:53:15 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_20230531_173252 +2023/06/01 13:53:15 - mmengine - INFO - Saving checkpoint at 15 epochs +2023/06/01 13:53:34 - mmengine - INFO - Epoch(val) [15][16/16] accuracy/top1: 100.0000 single-label/precision_classwise: [100.0, 0.0] single-label/recall_classwise: [100.0, 0.0] single-label/f1-score_classwise: [100.0, 0.0] data_time: 0.2341 time: 0.6188 +2023/06/01 13:55:04 - mmengine - INFO - Epoch(train) [16][ 100/5758] lr: 5.9432e-04 eta: 6:43:40 time: 0.8635 data_time: 0.0019 memory: 25074 loss: 0.0025 +2023/06/01 13:56:28 - mmengine - INFO - Epoch(train) [16][ 200/5758] lr: 5.9432e-04 eta: 6:42:15 time: 0.8420 data_time: 0.0024 memory: 25074 loss: 0.0034 +2023/06/01 13:57:51 - mmengine - INFO - Epoch(train) [16][ 300/5758] lr: 5.9432e-04 eta: 6:40:51 time: 0.7625 data_time: 0.0021 memory: 25074 loss: 0.0030 +2023/06/01 13:59:15 - mmengine - INFO - Epoch(train) [16][ 400/5758] lr: 5.9432e-04 eta: 6:39:26 time: 0.8603 data_time: 0.0021 memory: 25074 loss: 0.0035 +2023/06/01 14:00:38 - mmengine - INFO - Epoch(train) [16][ 500/5758] lr: 5.9432e-04 eta: 6:38:01 time: 0.7584 data_time: 0.0020 memory: 25074 loss: 0.0015 +2023/06/01 14:02:01 - mmengine - INFO - Epoch(train) [16][ 600/5758] lr: 5.9432e-04 eta: 6:36:36 time: 0.8660 data_time: 0.0018 memory: 25074 loss: 0.0032 +2023/06/01 14:02:26 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_20230531_173252 +2023/06/01 14:03:24 - mmengine - INFO - Epoch(train) [16][ 700/5758] lr: 5.9432e-04 eta: 6:35:11 time: 0.7694 data_time: 0.0018 memory: 25074 loss: 0.0005 +2023/06/01 14:04:47 - mmengine - INFO - Epoch(train) [16][ 800/5758] lr: 5.9432e-04 eta: 6:33:47 time: 0.8643 data_time: 0.0020 memory: 25074 loss: 0.0007 +2023/06/01 14:06:10 - mmengine - INFO - Epoch(train) [16][ 900/5758] lr: 5.9432e-04 eta: 6:32:22 time: 0.8344 data_time: 0.0027 memory: 25074 loss: 0.0025 +2023/06/01 14:07:33 - mmengine - INFO - Epoch(train) [16][1000/5758] lr: 5.9432e-04 eta: 6:30:57 time: 0.8328 data_time: 0.0017 memory: 25074 loss: 0.0021 +2023/06/01 14:08:56 - mmengine - INFO - Epoch(train) [16][1100/5758] lr: 5.9432e-04 eta: 6:29:32 time: 0.8298 data_time: 0.0018 memory: 25074 loss: 0.0033 +2023/06/01 14:10:20 - mmengine - INFO - Epoch(train) [16][1200/5758] lr: 5.9432e-04 eta: 6:28:07 time: 0.8674 data_time: 0.0015 memory: 25074 loss: 0.0030 +2023/06/01 14:11:43 - mmengine - INFO - Epoch(train) [16][1300/5758] lr: 5.9432e-04 eta: 6:26:43 time: 0.8187 data_time: 0.0027 memory: 25074 loss: 0.0004 +2023/06/01 14:13:08 - mmengine - INFO - Epoch(train) [16][1400/5758] lr: 5.9432e-04 eta: 6:25:18 time: 0.9097 data_time: 0.0023 memory: 25074 loss: 0.0027 +2023/06/01 14:14:32 - mmengine - INFO - Epoch(train) [16][1500/5758] lr: 5.9432e-04 eta: 6:23:54 time: 0.8454 data_time: 0.0019 memory: 25074 loss: 0.0023 +2023/06/01 14:15:56 - mmengine - INFO - Epoch(train) [16][1600/5758] lr: 5.9432e-04 eta: 6:22:29 time: 0.8380 data_time: 0.0019 memory: 25074 loss: 0.0039 +2023/06/01 14:16:21 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_20230531_173252 +2023/06/01 14:17:21 - mmengine - INFO - Epoch(train) [16][1700/5758] lr: 5.9432e-04 eta: 6:21:05 time: 0.8268 data_time: 0.0026 memory: 25074 loss: 0.0022 +2023/06/01 14:18:46 - mmengine - INFO - Epoch(train) [16][1800/5758] lr: 5.9432e-04 eta: 6:19:41 time: 0.8713 data_time: 0.0025 memory: 25074 loss: 0.0016 +2023/06/01 14:20:09 - mmengine - INFO - Epoch(train) [16][1900/5758] lr: 5.9432e-04 eta: 6:18:16 time: 0.8646 data_time: 0.0015 memory: 25074 loss: 0.0030 +2023/06/01 14:21:34 - mmengine - INFO - Epoch(train) [16][2000/5758] lr: 5.9432e-04 eta: 6:16:52 time: 0.8700 data_time: 0.0016 memory: 25074 loss: 0.0009 +2023/06/01 14:23:02 - mmengine - INFO - Epoch(train) [16][2100/5758] lr: 5.9432e-04 eta: 6:15:28 time: 0.8489 data_time: 0.0020 memory: 25074 loss: 0.0027 +2023/06/01 14:24:26 - mmengine - INFO - Epoch(train) [16][2200/5758] lr: 5.9432e-04 eta: 6:14:04 time: 0.8313 data_time: 0.0025 memory: 25074 loss: 0.0015 +2023/06/01 14:25:51 - mmengine - INFO - Epoch(train) [16][2300/5758] lr: 5.9432e-04 eta: 6:12:40 time: 0.8532 data_time: 0.0019 memory: 25074 loss: 0.0009 +2023/06/01 14:27:16 - mmengine - INFO - Epoch(train) [16][2400/5758] lr: 5.9432e-04 eta: 6:11:15 time: 0.8587 data_time: 0.0022 memory: 25074 loss: 0.0008 +2023/06/01 14:28:42 - mmengine - INFO - Epoch(train) [16][2500/5758] lr: 5.9432e-04 eta: 6:09:52 time: 0.8913 data_time: 0.0016 memory: 25074 loss: 0.0037 +2023/06/01 14:30:13 - mmengine - INFO - Epoch(train) [16][2600/5758] lr: 5.9432e-04 eta: 6:08:29 time: 0.9534 data_time: 0.0038 memory: 25074 loss: 0.0010 +2023/06/01 14:30:39 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_20230531_173252 +2023/06/01 14:31:41 - mmengine - INFO - Epoch(train) [16][2700/5758] lr: 5.9432e-04 eta: 6:07:06 time: 0.9603 data_time: 0.0021 memory: 25074 loss: 0.0016 +2023/06/01 14:33:10 - mmengine - INFO - Epoch(train) [16][2800/5758] lr: 5.9432e-04 eta: 6:05:42 time: 0.9243 data_time: 0.0022 memory: 25074 loss: 0.0005 +2023/06/01 14:34:41 - mmengine - INFO - Epoch(train) [16][2900/5758] lr: 5.9432e-04 eta: 6:04:20 time: 0.9090 data_time: 0.0023 memory: 25074 loss: 0.0108 +2023/06/01 14:36:10 - mmengine - INFO - Epoch(train) [16][3000/5758] lr: 5.9432e-04 eta: 6:02:57 time: 0.8929 data_time: 0.0015 memory: 25074 loss: 0.0019 +2023/06/01 14:37:36 - mmengine - INFO - Epoch(train) [16][3100/5758] lr: 5.9432e-04 eta: 6:01:33 time: 0.9010 data_time: 0.0016 memory: 25074 loss: 0.0008 +2023/06/01 14:39:01 - mmengine - INFO - Epoch(train) [16][3200/5758] lr: 5.9432e-04 eta: 6:00:08 time: 0.8695 data_time: 0.0015 memory: 25074 loss: 0.0018 +2023/06/01 14:40:26 - mmengine - INFO - Epoch(train) [16][3300/5758] lr: 5.9432e-04 eta: 5:58:44 time: 0.8774 data_time: 0.0016 memory: 25074 loss: 0.0006 +2023/06/01 14:41:50 - mmengine - INFO - Epoch(train) [16][3400/5758] lr: 5.9432e-04 eta: 5:57:20 time: 0.8348 data_time: 0.0016 memory: 25074 loss: 0.0029 +2023/06/01 14:43:15 - mmengine - INFO - Epoch(train) [16][3500/5758] lr: 5.9432e-04 eta: 5:55:55 time: 0.8082 data_time: 0.0023 memory: 25074 loss: 0.0031 +2023/06/01 14:44:38 - mmengine - INFO - Epoch(train) [16][3600/5758] lr: 5.9432e-04 eta: 5:54:31 time: 0.7833 data_time: 0.0016 memory: 25074 loss: 0.0014 +2023/06/01 14:45:02 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_20230531_173252 +2023/06/01 14:46:00 - mmengine - INFO - Epoch(train) [16][3700/5758] lr: 5.9432e-04 eta: 5:53:05 time: 0.8314 data_time: 0.0016 memory: 25074 loss: 0.0019 +2023/06/01 14:47:25 - mmengine - INFO - Epoch(train) [16][3800/5758] lr: 5.9432e-04 eta: 5:51:41 time: 0.8320 data_time: 0.0019 memory: 25074 loss: 0.0008 +2023/06/01 14:48:48 - mmengine - INFO - Epoch(train) [16][3900/5758] lr: 5.9432e-04 eta: 5:50:16 time: 0.8242 data_time: 0.0014 memory: 25074 loss: 0.0035 +2023/06/01 14:50:12 - mmengine - INFO - Epoch(train) [16][4000/5758] lr: 5.9432e-04 eta: 5:48:52 time: 0.8209 data_time: 0.0024 memory: 25074 loss: 0.0022 +2023/06/01 14:51:34 - mmengine - INFO - Epoch(train) [16][4100/5758] lr: 5.9432e-04 eta: 5:47:27 time: 0.7744 data_time: 0.0027 memory: 25074 loss: 0.0025 +2023/06/01 14:52:56 - mmengine - INFO - Epoch(train) [16][4200/5758] lr: 5.9432e-04 eta: 5:46:01 time: 0.7740 data_time: 0.0016 memory: 25074 loss: 0.0040 +2023/06/01 14:54:18 - mmengine - INFO - Epoch(train) [16][4300/5758] lr: 5.9432e-04 eta: 5:44:36 time: 0.8169 data_time: 0.0025 memory: 25074 loss: 0.0027 +2023/06/01 14:55:38 - mmengine - INFO - Epoch(train) [16][4400/5758] lr: 5.9432e-04 eta: 5:43:11 time: 0.8274 data_time: 0.0024 memory: 25074 loss: 0.0066 +2023/06/01 14:56:57 - mmengine - INFO - Epoch(train) [16][4500/5758] lr: 5.9432e-04 eta: 5:41:45 time: 0.8252 data_time: 0.0016 memory: 25074 loss: 0.0027 +2023/06/01 14:58:17 - mmengine - INFO - Epoch(train) [16][4600/5758] lr: 5.9432e-04 eta: 5:40:19 time: 0.7876 data_time: 0.0017 memory: 25074 loss: 0.0054 +2023/06/01 14:58:42 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_20230531_173252 +2023/06/01 14:59:38 - mmengine - INFO - Epoch(train) [16][4700/5758] lr: 5.9432e-04 eta: 5:38:54 time: 0.8089 data_time: 0.0026 memory: 25074 loss: 0.0014 +2023/06/01 15:00:58 - mmengine - INFO - Epoch(train) [16][4800/5758] lr: 5.9432e-04 eta: 5:37:28 time: 0.7445 data_time: 0.0015 memory: 25074 loss: 0.0009 +2023/06/01 15:02:16 - mmengine - INFO - Epoch(train) [16][4900/5758] lr: 5.9432e-04 eta: 5:36:02 time: 0.7698 data_time: 0.0021 memory: 25074 loss: 0.0019 +2023/06/01 15:03:36 - mmengine - INFO - Epoch(train) [16][5000/5758] lr: 5.9432e-04 eta: 5:34:37 time: 0.7804 data_time: 0.0017 memory: 25074 loss: 0.0008 +2023/06/01 15:04:57 - mmengine - INFO - Epoch(train) [16][5100/5758] lr: 5.9432e-04 eta: 5:33:12 time: 0.7461 data_time: 0.0020 memory: 25074 loss: 0.0056 +2023/06/01 15:06:17 - mmengine - INFO - Epoch(train) [16][5200/5758] lr: 5.9432e-04 eta: 5:31:46 time: 0.8317 data_time: 0.0025 memory: 25074 loss: 0.0006 +2023/06/01 15:07:37 - mmengine - INFO - Epoch(train) [16][5300/5758] lr: 5.9432e-04 eta: 5:30:20 time: 0.7502 data_time: 0.0016 memory: 25074 loss: 0.0013 +2023/06/01 15:08:56 - mmengine - INFO - Epoch(train) [16][5400/5758] lr: 5.9432e-04 eta: 5:28:55 time: 0.7772 data_time: 0.0020 memory: 25074 loss: 0.0012 diff --git a/convnext_small_4xb256_4e-3lr_5m/20230531_173252/vis_data/20230531_173252.json b/convnext_small_4xb256_4e-3lr_5m/20230531_173252/vis_data/20230531_173252.json new file mode 100644 index 0000000000000000000000000000000000000000..afc30d82d119faef960e30e57ceab1cc0a51bf14 --- /dev/null +++ b/convnext_small_4xb256_4e-3lr_5m/20230531_173252/vis_data/20230531_173252.json @@ -0,0 +1,924 @@ +{"lr": 0.004, "data_time": 0.25836734771728515, "loss": 0.6135261058807373, "time": 1.1036933660507202, "epoch": 1, "memory": 25071, "step": 100} +{"lr": 0.004, "data_time": 0.001967930793762207, "loss": 0.5921411991119385, "time": 1.0567732810974122, "epoch": 1, "memory": 25071, "step": 200} +{"lr": 0.004, "data_time": 0.0014708280563354493, "loss": 0.5693366527557373, "time": 1.0624513149261474, "epoch": 1, "memory": 25071, "step": 300} +{"lr": 0.004, "data_time": 0.0017904520034790039, "loss": 0.5559554100036621, "time": 1.0132100105285644, "epoch": 1, "memory": 25071, "step": 400} +{"lr": 0.004, "data_time": 0.0018054723739624023, "loss": 0.5386565923690796, "time": 1.057649564743042, "epoch": 1, "memory": 25071, "step": 500} +{"lr": 0.004, "data_time": 0.001491236686706543, "loss": 0.520584225654602, "time": 1.0222002744674683, "epoch": 1, "memory": 25071, "step": 600} +{"lr": 0.004, "data_time": 0.0010788917541503906, "loss": 0.5122485876083374, "time": 1.0578108787536622, "epoch": 1, "memory": 25071, "step": 700} +{"lr": 0.004, "data_time": 0.0018996477127075195, "loss": 0.5337131738662719, "time": 0.9827592134475708, "epoch": 1, "memory": 25071, "step": 800} +{"lr": 0.004, "data_time": 0.0014658689498901368, "loss": 0.48136401176452637, "time": 1.0142579555511475, "epoch": 1, "memory": 25071, "step": 900} +{"lr": 0.004, "data_time": 0.0016474485397338866, "loss": 0.48867201805114746, "time": 1.0005245208740234, "epoch": 1, "memory": 25071, "step": 1000} +{"lr": 0.004, "data_time": 0.0021308183670043944, "loss": 0.4960619449615479, "time": 0.9755245447158813, "epoch": 1, "memory": 25071, "step": 1100} +{"lr": 0.004, "data_time": 0.0018622636795043944, "loss": 0.45781826972961426, "time": 0.9678232431411743, "epoch": 1, "memory": 25071, "step": 1200} +{"lr": 0.004, "data_time": 0.0016861677169799805, "loss": 0.4497794032096863, "time": 1.0431145668029784, "epoch": 1, "memory": 25071, "step": 1300} +{"lr": 0.004, "data_time": 0.0015056848526000977, "loss": 0.4368001639842987, "time": 0.9595765352249146, "epoch": 1, "memory": 25071, "step": 1400} +{"lr": 0.004, "data_time": 0.0017606496810913086, "loss": 0.5319863557815552, "time": 0.9088726043701172, "epoch": 1, "memory": 25071, "step": 1500} +{"lr": 0.004, "data_time": 0.0015338659286499023, "loss": 0.4219388782978058, "time": 0.9892490386962891, "epoch": 1, "memory": 25071, "step": 1600} +{"lr": 0.004, "data_time": 0.0016805887222290038, "loss": 0.41420637667179105, "time": 0.9990060091018677, "epoch": 1, "memory": 25071, "step": 1700} +{"lr": 0.004, "data_time": 0.001367664337158203, "loss": 0.3887391030788422, "time": 1.133042073249817, "epoch": 1, "memory": 25071, "step": 1800} +{"lr": 0.004, "data_time": 0.0014392852783203125, "loss": 0.3968598783016205, "time": 0.9008044719696044, "epoch": 1, "memory": 25071, "step": 1900} +{"lr": 0.004, "data_time": 0.0013582229614257813, "loss": 0.38166859149932864, "time": 1.0659605503082275, "epoch": 1, "memory": 25071, "step": 2000} +{"lr": 0.004, "data_time": 0.0015635013580322266, "loss": 0.40037643909454346, "time": 1.067807149887085, "epoch": 1, "memory": 25071, "step": 2100} +{"lr": 0.004, "data_time": 0.0015240907669067383, "loss": 0.3590026319026947, "time": 0.8623078346252442, "epoch": 1, "memory": 25071, "step": 2200} +{"lr": 0.004, "data_time": 0.0014862537384033204, "loss": 0.2965458869934082, "time": 0.8956292152404786, "epoch": 1, "memory": 25071, "step": 2300} +{"lr": 0.004, "data_time": 0.0015390634536743164, "loss": 0.31096053421497344, "time": 0.9354721546173096, "epoch": 1, "memory": 25071, "step": 2400} +{"lr": 0.004, "data_time": 0.0015926837921142578, "loss": 0.28933600038290025, "time": 0.8769220352172852, "epoch": 1, "memory": 25071, "step": 2500} +{"lr": 0.004, "data_time": 0.001988649368286133, "loss": 0.2849620833992958, "time": 0.8799722909927368, "epoch": 1, "memory": 25071, "step": 2600} +{"lr": 0.004, "data_time": 0.0013595342636108399, "loss": 0.2896030187606812, "time": 0.9100133657455445, "epoch": 1, "memory": 25071, "step": 2700} +{"lr": 0.004, "data_time": 0.0014533758163452148, "loss": 0.32246443927288054, "time": 0.9779120445251465, "epoch": 1, "memory": 25071, "step": 2800} +{"lr": 0.004, "data_time": 0.001288127899169922, "loss": 0.27721202224493025, "time": 0.9152437686920166, "epoch": 1, "memory": 25071, "step": 2900} +{"lr": 0.004, "data_time": 0.0019807338714599608, "loss": 0.22157122939825058, "time": 0.931296968460083, "epoch": 1, "memory": 25071, "step": 3000} +{"lr": 0.004, "data_time": 0.001652097702026367, "loss": 0.2503370255231857, "time": 0.9809789896011353, "epoch": 1, "memory": 25071, "step": 3100} +{"lr": 0.004, "data_time": 0.0013518571853637696, "loss": 0.21450963467359543, "time": 0.9946384191513061, "epoch": 1, "memory": 25071, "step": 3200} +{"lr": 0.004, "data_time": 0.002450108528137207, "loss": 0.239343623816967, "time": 0.9268593072891236, "epoch": 1, "memory": 25071, "step": 3300} +{"lr": 0.004, "data_time": 0.0017489194869995117, "loss": 0.21228042244911194, "time": 0.9522735357284546, "epoch": 1, "memory": 25071, "step": 3400} +{"lr": 0.004, "data_time": 0.0014758110046386719, "loss": 0.21550967246294023, "time": 0.9016278266906739, "epoch": 1, "memory": 25071, "step": 3500} +{"lr": 0.004, "data_time": 0.022446608543395995, "loss": 0.21206613928079604, "time": 1.0201257705688476, "epoch": 1, "memory": 25071, "step": 3600} +{"lr": 0.004, "data_time": 0.0015426874160766602, "loss": 0.1702307105064392, "time": 0.8195062875747681, "epoch": 1, "memory": 25071, "step": 3700} +{"lr": 0.004, "data_time": 0.0014066219329833985, "loss": 0.18829978108406067, "time": 1.001184868812561, "epoch": 1, "memory": 25071, "step": 3800} +{"lr": 0.004, "data_time": 0.0014562368392944335, "loss": 0.16298147067427635, "time": 0.836207914352417, "epoch": 1, "memory": 25071, "step": 3900} +{"lr": 0.004, "data_time": 0.0014770746231079102, "loss": 0.1773746654391289, "time": 0.8800336360931397, "epoch": 1, "memory": 25071, "step": 4000} +{"lr": 0.004, "data_time": 0.001606893539428711, "loss": 0.15313689559698104, "time": 0.9190765142440795, "epoch": 1, "memory": 25071, "step": 4100} +{"lr": 0.004, "data_time": 0.0019302129745483398, "loss": 0.15373844504356385, "time": 0.9319148778915405, "epoch": 1, "memory": 25071, "step": 4200} +{"lr": 0.004, "data_time": 0.0015819787979125977, "loss": 0.12970864176750183, "time": 0.89850013256073, "epoch": 1, "memory": 25071, "step": 4300} +{"lr": 0.004, "data_time": 0.0014078617095947266, "loss": 0.1656157925724983, "time": 0.8514695882797241, "epoch": 1, "memory": 25071, "step": 4400} +{"lr": 0.004, "data_time": 0.0013335227966308593, "loss": 0.1450781174004078, "time": 0.9596346616744995, "epoch": 1, "memory": 25071, "step": 4500} +{"lr": 0.004, "data_time": 0.0013980865478515625, "loss": 0.15148803144693374, "time": 0.9196593523025512, "epoch": 1, "memory": 25071, "step": 4600} +{"lr": 0.004, "data_time": 0.002222132682800293, "loss": 0.11021999567747116, "time": 0.9126356601715088, "epoch": 1, "memory": 25071, "step": 4700} +{"lr": 0.004, "data_time": 0.0018005132675170898, "loss": 0.12593574821949005, "time": 0.9484604358673095, "epoch": 1, "memory": 25071, "step": 4800} +{"lr": 0.004, "data_time": 0.0012134790420532226, "loss": 0.09986786693334579, "time": 1.0157971143722535, "epoch": 1, "memory": 25071, "step": 4900} +{"lr": 0.004, "data_time": 0.0014298677444458008, "loss": 0.11388952732086181, "time": 0.9243617296218872, "epoch": 1, "memory": 25071, "step": 5000} +{"lr": 0.004, "data_time": 0.001612401008605957, "loss": 0.12963890954852103, "time": 0.9540765523910523, "epoch": 1, "memory": 25071, "step": 5100} +{"lr": 0.004, "data_time": 0.0021638393402099608, "loss": 0.0849734254181385, "time": 0.8873657941818237, "epoch": 1, "memory": 25071, "step": 5200} +{"lr": 0.004, "data_time": 0.0015216112136840821, "loss": 0.09903554022312164, "time": 0.966349720954895, "epoch": 1, "memory": 25071, "step": 5300} +{"lr": 0.004, "data_time": 0.0020122289657592773, "loss": 0.06487535983324051, "time": 0.9787596464157104, "epoch": 1, "memory": 25071, "step": 5400} +{"lr": 0.004, "data_time": 0.0015494585037231444, "loss": 0.07417001202702522, "time": 0.8545481443405152, "epoch": 1, "memory": 25071, "step": 5500} +{"lr": 0.004, "data_time": 0.0014139890670776367, "loss": 0.06840806379914284, "time": 0.9005010604858399, "epoch": 1, "memory": 25071, "step": 5600} +{"lr": 0.004, "data_time": 0.0016618967056274414, "loss": 0.06471074298024178, "time": 1.0014020204544067, "epoch": 1, "memory": 25071, "step": 5700} +{"accuracy/top1": 71.33387756347656, "data_time": 0.29866595566272736, "time": 0.8162815719842911, "step": 1} +{"lr": 0.0039754382394873, "data_time": 0.0015654325485229491, "loss": 0.44536194801330564, "time": 0.9839568614959717, "epoch": 2, "memory": 25074, "step": 5858} +{"lr": 0.0039754382394873, "data_time": 0.002077460289001465, "loss": 0.4071248471736908, "time": 0.9970962762832641, "epoch": 2, "memory": 25074, "step": 5958} +{"lr": 0.0039754382394873, "data_time": 0.0013185977935791016, "loss": 0.24007392972707747, "time": 1.0498293399810792, "epoch": 2, "memory": 25074, "step": 6058} +{"lr": 0.0039754382394873, "data_time": 0.0015950202941894531, "loss": 0.24183081835508347, "time": 0.9420676946640014, "epoch": 2, "memory": 25074, "step": 6158} +{"lr": 0.0039754382394873, "data_time": 0.0018357276916503907, "loss": 0.09749895296990871, "time": 0.8744754314422607, "epoch": 2, "memory": 25074, "step": 6258} +{"lr": 0.0039754382394873, "data_time": 0.002217006683349609, "loss": 0.0628897450864315, "time": 0.9066163539886475, "epoch": 2, "memory": 25074, "step": 6358} +{"lr": 0.0039754382394873, "data_time": 0.001868891716003418, "loss": 0.0615274328738451, "time": 0.9184701681137085, "epoch": 2, "memory": 25074, "step": 6458} +{"lr": 0.0039754382394873, "data_time": 0.0014720201492309571, "loss": 0.07046853713691234, "time": 0.8759130716323853, "epoch": 2, "memory": 25074, "step": 6558} +{"lr": 0.0039754382394873, "data_time": 0.0015817642211914062, "loss": 0.07254206612706185, "time": 0.9210872888565064, "epoch": 2, "memory": 25074, "step": 6658} +{"lr": 0.0039754382394873, "data_time": 0.0014837980270385742, "loss": 0.08038320019841194, "time": 1.0461889982223511, "epoch": 2, "memory": 25074, "step": 6758} +{"lr": 0.0039754382394873, "data_time": 0.001940441131591797, "loss": 0.06745437271893025, "time": 0.8867187023162841, "epoch": 2, "memory": 25074, "step": 6858} +{"lr": 0.0039754382394873, "data_time": 0.0015705347061157227, "loss": 0.05236329883337021, "time": 0.8681844234466553, "epoch": 2, "memory": 25074, "step": 6958} +{"lr": 0.0039754382394873, "data_time": 0.0019382953643798829, "loss": 0.08161418288946151, "time": 0.9430710315704346, "epoch": 2, "memory": 25074, "step": 7058} +{"lr": 0.0039754382394873, "data_time": 0.0021197319030761717, "loss": 0.045391821675002574, "time": 0.938313889503479, "epoch": 2, "memory": 25074, "step": 7158} +{"lr": 0.0039754382394873, "data_time": 0.0018213033676147462, "loss": 0.05432411562651396, "time": 0.82647705078125, "epoch": 2, "memory": 25074, "step": 7258} +{"lr": 0.0039754382394873, "data_time": 0.0022699117660522463, "loss": 0.06139286048710346, "time": 0.9103602409362793, "epoch": 2, "memory": 25074, "step": 7358} +{"lr": 0.0039754382394873, "data_time": 0.0012668371200561523, "loss": 0.049718914180994035, "time": 0.860216474533081, "epoch": 2, "memory": 25074, "step": 7458} +{"lr": 0.0039754382394873, "data_time": 0.0023814916610717775, "loss": 0.050588475167751314, "time": 0.9751445293426514, "epoch": 2, "memory": 25074, "step": 7558} +{"lr": 0.0039754382394873, "data_time": 0.0017695188522338866, "loss": 0.5002466946840286, "time": 0.7511817693710328, "epoch": 2, "memory": 25074, "step": 7658} +{"lr": 0.0039754382394873, "data_time": 0.0013837099075317382, "loss": 0.4878327250480652, "time": 0.8483304500579834, "epoch": 2, "memory": 25074, "step": 7758} +{"lr": 0.0039754382394873, "data_time": 0.0019185066223144532, "loss": 0.4479573845863342, "time": 0.8353627443313598, "epoch": 2, "memory": 25074, "step": 7858} +{"lr": 0.0039754382394873, "data_time": 0.0019244909286499023, "loss": 0.21451487988233567, "time": 0.8248765468597412, "epoch": 2, "memory": 25074, "step": 7958} +{"lr": 0.0039754382394873, "data_time": 0.0020031452178955076, "loss": 0.08937919065356255, "time": 0.8127747297286987, "epoch": 2, "memory": 25074, "step": 8058} +{"lr": 0.0039754382394873, "data_time": 0.0019345521926879884, "loss": 0.048111673444509506, "time": 0.8491933107376098, "epoch": 2, "memory": 25074, "step": 8158} +{"lr": 0.0039754382394873, "data_time": 0.002181816101074219, "loss": 0.05818078741431236, "time": 0.8987390756607055, "epoch": 2, "memory": 25074, "step": 8258} +{"lr": 0.0039754382394873, "data_time": 0.0016150474548339844, "loss": 0.04987697713077068, "time": 0.7931256771087647, "epoch": 2, "memory": 25074, "step": 8358} +{"lr": 0.0039754382394873, "data_time": 0.002272367477416992, "loss": 0.06922312416136264, "time": 0.8968161344528198, "epoch": 2, "memory": 25074, "step": 8458} +{"lr": 0.0039754382394873, "data_time": 0.0022190093994140627, "loss": 0.04928675331175327, "time": 0.8449273586273194, "epoch": 2, "memory": 25074, "step": 8558} +{"lr": 0.0039754382394873, "data_time": 0.0026656150817871093, "loss": 0.03678979128599167, "time": 0.7909621715545654, "epoch": 2, "memory": 25074, "step": 8658} +{"lr": 0.0039754382394873, "data_time": 0.0016669988632202148, "loss": 0.04629333112388849, "time": 0.7934857606887817, "epoch": 2, "memory": 25074, "step": 8758} +{"lr": 0.0039754382394873, "data_time": 0.0017243385314941405, "loss": 0.04524355810135603, "time": 0.8450065851211548, "epoch": 2, "memory": 25074, "step": 8858} +{"lr": 0.0039754382394873, "data_time": 0.0021995067596435546, "loss": 0.049249463342130186, "time": 0.8159966468811035, "epoch": 2, "memory": 25074, "step": 8958} +{"lr": 0.0039754382394873, "data_time": 0.0020696163177490235, "loss": 0.041333705000579354, "time": 0.812670636177063, "epoch": 2, "memory": 25074, "step": 9058} +{"lr": 0.0039754382394873, "data_time": 0.001447772979736328, "loss": 0.040220907609909776, "time": 0.8368125915527344, "epoch": 2, "memory": 25074, "step": 9158} +{"lr": 0.0039754382394873, "data_time": 0.001545858383178711, "loss": 0.07665113694965839, "time": 0.8620558261871338, "epoch": 2, "memory": 25074, "step": 9258} +{"lr": 0.0039754382394873, "data_time": 0.001687169075012207, "loss": 0.0275498004630208, "time": 0.7853478908538818, "epoch": 2, "memory": 25074, "step": 9358} +{"lr": 0.0039754382394873, "data_time": 0.0018344879150390624, "loss": 0.028978040814399718, "time": 0.8449751615524292, "epoch": 2, "memory": 25074, "step": 9458} +{"lr": 0.0039754382394873, "data_time": 0.0015711069107055664, "loss": 0.030457904562354088, "time": 0.8253869771957397, "epoch": 2, "memory": 25074, "step": 9558} +{"lr": 0.0039754382394873, "data_time": 0.0017217159271240234, "loss": 0.0381824167445302, "time": 0.9336783409118652, "epoch": 2, "memory": 25074, "step": 9658} +{"lr": 0.0039754382394873, "data_time": 0.0015754461288452148, "loss": 0.06059279851615429, "time": 0.8491053342819214, "epoch": 2, "memory": 25074, "step": 9758} +{"lr": 0.0039754382394873, "data_time": 0.0018334627151489259, "loss": 0.039567502215504646, "time": 0.9262634754180908, "epoch": 2, "memory": 25074, "step": 9858} +{"lr": 0.0039754382394873, "data_time": 0.0015657901763916015, "loss": 0.03016723138280213, "time": 0.9408579587936401, "epoch": 2, "memory": 25074, "step": 9958} +{"lr": 0.0039754382394873, "data_time": 0.0015697717666625977, "loss": 0.03549889884889126, "time": 0.8561189889907836, "epoch": 2, "memory": 25074, "step": 10058} +{"lr": 0.0039754382394873, "data_time": 0.001992487907409668, "loss": 0.024865062534809114, "time": 0.8507202863693237, "epoch": 2, "memory": 25074, "step": 10158} +{"lr": 0.0039754382394873, "data_time": 0.001700282096862793, "loss": 0.024713566433638333, "time": 0.9130228757858276, "epoch": 2, "memory": 25074, "step": 10258} +{"lr": 0.0039754382394873, "data_time": 0.001290130615234375, "loss": 0.47437719702720643, "time": 0.8855489730834961, "epoch": 2, "memory": 25074, "step": 10358} +{"lr": 0.0039754382394873, "data_time": 0.0012105464935302734, "loss": 0.28070435076951983, "time": 0.8346230030059815, "epoch": 2, "memory": 25074, "step": 10458} +{"lr": 0.0039754382394873, "data_time": 0.0013357162475585937, "loss": 0.5716702461242675, "time": 0.9133694171905518, "epoch": 2, "memory": 25074, "step": 10558} +{"lr": 0.0039754382394873, "data_time": 0.0014510393142700196, "loss": 0.5090821623802185, "time": 0.885738730430603, "epoch": 2, "memory": 25074, "step": 10658} +{"lr": 0.0039754382394873, "data_time": 0.002016592025756836, "loss": 0.38386878967285154, "time": 0.912585973739624, "epoch": 2, "memory": 25074, "step": 10758} +{"lr": 0.0039754382394873, "data_time": 0.0020873546600341797, "loss": 0.4081609070301056, "time": 0.9337555170059204, "epoch": 2, "memory": 25074, "step": 10858} +{"lr": 0.0039754382394873, "data_time": 0.0012762784957885743, "loss": 0.33712922036647797, "time": 0.7847714900970459, "epoch": 2, "memory": 25074, "step": 10958} +{"lr": 0.0039754382394873, "data_time": 0.00125885009765625, "loss": 0.326380181312561, "time": 0.8530193090438842, "epoch": 2, "memory": 25074, "step": 11058} +{"lr": 0.0039754382394873, "data_time": 0.001339125633239746, "loss": 0.1495766393840313, "time": 0.9370234251022339, "epoch": 2, "memory": 25074, "step": 11158} +{"lr": 0.0039754382394873, "data_time": 0.0013068199157714843, "loss": 0.07547324076294899, "time": 0.8928805828094483, "epoch": 2, "memory": 25074, "step": 11258} +{"lr": 0.0039754382394873, "data_time": 0.002223658561706543, "loss": 0.7301055625081062, "time": 0.8583959817886353, "epoch": 2, "memory": 25074, "step": 11358} +{"lr": 0.0039754382394873, "data_time": 0.0014155149459838868, "loss": 0.6506040573120118, "time": 0.9257060527801514, "epoch": 2, "memory": 25074, "step": 11458} +{"accuracy/top1": 100.0, "data_time": 0.25789012628443103, "time": 0.6472268385045669, "step": 2} +{"lr": 0.0039023577500088323, "data_time": 0.0020456314086914062, "loss": 0.530897855758667, "time": 0.8861788988113404, "epoch": 3, "memory": 25074, "step": 11616} +{"lr": 0.0039023577500088323, "data_time": 0.0016747236251831055, "loss": 0.47043827176094055, "time": 0.9475600481033325, "epoch": 3, "memory": 25074, "step": 11716} +{"lr": 0.0039023577500088323, "data_time": 0.002315211296081543, "loss": 0.4299037456512451, "time": 0.830585265159607, "epoch": 3, "memory": 25074, "step": 11816} +{"lr": 0.0039023577500088323, "data_time": 0.001876497268676758, "loss": 0.3691077768802643, "time": 0.9028676509857178, "epoch": 3, "memory": 25074, "step": 11916} +{"lr": 0.0039023577500088323, "data_time": 0.0020478010177612306, "loss": 0.35603734850883484, "time": 0.8394214153289795, "epoch": 3, "memory": 25074, "step": 12016} +{"lr": 0.0039023577500088323, "data_time": 0.001947927474975586, "loss": 0.30922357439994813, "time": 0.999903678894043, "epoch": 3, "memory": 25074, "step": 12116} +{"lr": 0.0039023577500088323, "data_time": 0.00202484130859375, "loss": 0.28984666168689727, "time": 0.8322690010070801, "epoch": 3, "memory": 25074, "step": 12216} +{"lr": 0.0039023577500088323, "data_time": 0.0020009756088256838, "loss": 0.2831827044487, "time": 0.8980642795562744, "epoch": 3, "memory": 25074, "step": 12316} +{"lr": 0.0039023577500088323, "data_time": 0.002239632606506348, "loss": 0.22968373447656631, "time": 0.8585394859313965, "epoch": 3, "memory": 25074, "step": 12416} +{"lr": 0.0039023577500088323, "data_time": 0.003071236610412598, "loss": 0.22362501174211502, "time": 0.9056859016418457, "epoch": 3, "memory": 25074, "step": 12516} +{"lr": 0.0039023577500088323, "data_time": 0.0021527767181396484, "loss": 0.2103023812174797, "time": 0.9525051593780518, "epoch": 3, "memory": 25074, "step": 12616} +{"lr": 0.0039023577500088323, "data_time": 0.0021797895431518556, "loss": 0.19760508984327316, "time": 0.8640597343444825, "epoch": 3, "memory": 25074, "step": 12716} +{"lr": 0.0039023577500088323, "data_time": 0.0019652605056762694, "loss": 0.22433767467737198, "time": 0.8039330005645752, "epoch": 3, "memory": 25074, "step": 12816} +{"lr": 0.0039023577500088323, "data_time": 0.0018321752548217773, "loss": 0.12686453387141228, "time": 0.8569660425186157, "epoch": 3, "memory": 25074, "step": 12916} +{"lr": 0.0039023577500088323, "data_time": 0.0022662878036499023, "loss": 0.13652086406946182, "time": 0.9195060968399048, "epoch": 3, "memory": 25074, "step": 13016} +{"lr": 0.0039023577500088323, "data_time": 0.00256044864654541, "loss": 0.15135828405618668, "time": 0.9112863063812255, "epoch": 3, "memory": 25074, "step": 13116} +{"lr": 0.0039023577500088323, "data_time": 0.002023744583129883, "loss": 0.11753128990530967, "time": 0.8293555021286011, "epoch": 3, "memory": 25074, "step": 13216} +{"lr": 0.0039023577500088323, "data_time": 0.0019275903701782226, "loss": 0.0810539573431015, "time": 0.9455954790115356, "epoch": 3, "memory": 25074, "step": 13316} +{"lr": 0.0039023577500088323, "data_time": 0.0014641523361206056, "loss": 0.06795164383947849, "time": 0.8764861822128296, "epoch": 3, "memory": 25074, "step": 13416} +{"lr": 0.0039023577500088323, "data_time": 0.0019464492797851562, "loss": 0.0674210075289011, "time": 0.8908105373382569, "epoch": 3, "memory": 25074, "step": 13516} +{"lr": 0.0039023577500088323, "data_time": 0.0014425039291381836, "loss": 0.07151532731950283, "time": 0.8104778051376342, "epoch": 3, "memory": 25074, "step": 13616} +{"lr": 0.0039023577500088323, "data_time": 0.0012932777404785155, "loss": 0.0617364514619112, "time": 0.8420161008834839, "epoch": 3, "memory": 25074, "step": 13716} +{"lr": 0.0039023577500088323, "data_time": 0.0020517826080322264, "loss": 0.045361613482236864, "time": 0.794519567489624, "epoch": 3, "memory": 25074, "step": 13816} +{"lr": 0.0039023577500088323, "data_time": 0.0017646074295043946, "loss": 0.04847529921680689, "time": 0.8627906799316406, "epoch": 3, "memory": 25074, "step": 13916} +{"lr": 0.0039023577500088323, "data_time": 0.0022499561309814453, "loss": 0.04969416819512844, "time": 0.8765430450439453, "epoch": 3, "memory": 25074, "step": 14016} +{"lr": 0.0039023577500088323, "data_time": 0.001618647575378418, "loss": 0.07003878727555275, "time": 0.7989153385162353, "epoch": 3, "memory": 25074, "step": 14116} +{"lr": 0.0039023577500088323, "data_time": 0.0013932466506958007, "loss": 0.05054851789027452, "time": 0.8280042886734009, "epoch": 3, "memory": 25074, "step": 14216} +{"lr": 0.0039023577500088323, "data_time": 0.0016368627548217773, "loss": 0.050657306984066966, "time": 0.8679801940917968, "epoch": 3, "memory": 25074, "step": 14316} +{"lr": 0.0039023577500088323, "data_time": 0.002403879165649414, "loss": 0.05359412152320146, "time": 0.810921311378479, "epoch": 3, "memory": 25074, "step": 14416} +{"lr": 0.0039023577500088323, "data_time": 0.002183365821838379, "loss": 0.04041057452559471, "time": 0.8750509023666382, "epoch": 3, "memory": 25074, "step": 14516} +{"lr": 0.0039023577500088323, "data_time": 0.002136683464050293, "loss": 0.04170214645564556, "time": 0.8423972845077514, "epoch": 3, "memory": 25074, "step": 14616} +{"lr": 0.0039023577500088323, "data_time": 0.001434493064880371, "loss": 0.10478606373071671, "time": 0.8206373929977417, "epoch": 3, "memory": 25074, "step": 14716} +{"lr": 0.0039023577500088323, "data_time": 0.0016542911529541016, "loss": 0.5778481483459472, "time": 0.8065363883972168, "epoch": 3, "memory": 25074, "step": 14816} +{"lr": 0.0039023577500088323, "data_time": 0.0019530534744262695, "loss": 0.42171688079833985, "time": 0.838587474822998, "epoch": 3, "memory": 25074, "step": 14916} +{"lr": 0.0039023577500088323, "data_time": 0.001415562629699707, "loss": 0.3420998930931091, "time": 0.8774507284164429, "epoch": 3, "memory": 25074, "step": 15016} +{"lr": 0.0039023577500088323, "data_time": 0.0020879268646240234, "loss": 0.2082872837781906, "time": 0.8238179922103882, "epoch": 3, "memory": 25074, "step": 15116} +{"lr": 0.0039023577500088323, "data_time": 0.001863384246826172, "loss": 0.09171477109193801, "time": 0.8462811708450317, "epoch": 3, "memory": 25074, "step": 15216} +{"lr": 0.0039023577500088323, "data_time": 0.001365351676940918, "loss": 0.09157317467033863, "time": 0.8382880210876464, "epoch": 3, "memory": 25074, "step": 15316} +{"lr": 0.0039023577500088323, "data_time": 0.001530599594116211, "loss": 0.040719654969871046, "time": 0.8324625730514527, "epoch": 3, "memory": 25074, "step": 15416} +{"lr": 0.0039023577500088323, "data_time": 0.0016404151916503905, "loss": 0.04555123336613178, "time": 0.8977967262268066, "epoch": 3, "memory": 25074, "step": 15516} +{"lr": 0.0039023577500088323, "data_time": 0.0014584064483642578, "loss": 0.0518822580575943, "time": 0.9010794162750244, "epoch": 3, "memory": 25074, "step": 15616} +{"lr": 0.0039023577500088323, "data_time": 0.0015751123428344727, "loss": 0.025403516925871374, "time": 0.9069263458251953, "epoch": 3, "memory": 25074, "step": 15716} +{"lr": 0.0039023577500088323, "data_time": 0.0014905691146850585, "loss": 0.03381628515198827, "time": 0.8677921533584595, "epoch": 3, "memory": 25074, "step": 15816} +{"lr": 0.0039023577500088323, "data_time": 0.0014578580856323242, "loss": 0.02633177926763892, "time": 0.8704387664794921, "epoch": 3, "memory": 25074, "step": 15916} +{"lr": 0.0039023577500088323, "data_time": 0.0014814615249633789, "loss": 0.02399464203044772, "time": 0.8488096475601197, "epoch": 3, "memory": 25074, "step": 16016} +{"lr": 0.0039023577500088323, "data_time": 0.0014820575714111328, "loss": 0.09220785070210695, "time": 0.8705696821212768, "epoch": 3, "memory": 25074, "step": 16116} +{"lr": 0.0039023577500088323, "data_time": 0.0015499114990234375, "loss": 0.039041274785995485, "time": 0.9396117687225342, "epoch": 3, "memory": 25074, "step": 16216} +{"lr": 0.0039023577500088323, "data_time": 0.0012744665145874023, "loss": 0.033044479042291644, "time": 0.9233942985534668, "epoch": 3, "memory": 25074, "step": 16316} +{"lr": 0.0039023577500088323, "data_time": 0.001346421241760254, "loss": 0.028087299875915052, "time": 0.9162482738494873, "epoch": 3, "memory": 25074, "step": 16416} +{"lr": 0.0039023577500088323, "data_time": 0.0017097949981689452, "loss": 0.02493842667900026, "time": 0.8040107250213623, "epoch": 3, "memory": 25074, "step": 16516} +{"lr": 0.0039023577500088323, "data_time": 0.0013321161270141602, "loss": 0.030253164237365127, "time": 0.8069973707199096, "epoch": 3, "memory": 25074, "step": 16616} +{"lr": 0.0039023577500088323, "data_time": 0.0014808893203735352, "loss": 0.38303807079792024, "time": 0.9347812414169312, "epoch": 3, "memory": 25074, "step": 16716} +{"lr": 0.0039023577500088323, "data_time": 0.0013840675354003906, "loss": 0.11098072454333305, "time": 0.8739282369613648, "epoch": 3, "memory": 25074, "step": 16816} +{"lr": 0.0039023577500088323, "data_time": 0.0015062808990478516, "loss": 0.17602913081645966, "time": 0.8635448694229126, "epoch": 3, "memory": 25074, "step": 16916} +{"lr": 0.0039023577500088323, "data_time": 0.0013004064559936524, "loss": 0.03915222808718681, "time": 0.8101547002792359, "epoch": 3, "memory": 25074, "step": 17016} +{"lr": 0.0039023577500088323, "data_time": 0.001514577865600586, "loss": 0.03889876324683428, "time": 0.9931037902832032, "epoch": 3, "memory": 25074, "step": 17116} +{"lr": 0.0039023577500088323, "data_time": 0.00196683406829834, "loss": 0.025535664753988385, "time": 0.9238677978515625, "epoch": 3, "memory": 25074, "step": 17216} +{"accuracy/top1": 100.0, "data_time": 0.24308482338400447, "time": 0.6311884767868939, "step": 3} +{"lr": 0.0037825580157557948, "data_time": 0.0019972562789916993, "loss": 0.024469496868550777, "time": 0.8110650300979614, "epoch": 4, "memory": 25074, "step": 17374} +{"lr": 0.0037825580157557948, "data_time": 0.0026037216186523436, "loss": 0.017897700518369676, "time": 0.864920711517334, "epoch": 4, "memory": 25074, "step": 17474} +{"lr": 0.0037825580157557948, "data_time": 0.002062368392944336, "loss": 0.024089607782661913, "time": 0.8393401622772216, "epoch": 4, "memory": 25074, "step": 17574} +{"lr": 0.0037825580157557948, "data_time": 0.0023013591766357423, "loss": 0.016335582453757526, "time": 0.8185791730880737, "epoch": 4, "memory": 25074, "step": 17674} +{"lr": 0.0037825580157557948, "data_time": 0.0025336742401123047, "loss": 0.02260943572036922, "time": 0.8512113332748413, "epoch": 4, "memory": 25074, "step": 17774} +{"lr": 0.0037825580157557948, "data_time": 0.002070331573486328, "loss": 0.024111079657450317, "time": 0.8192685365676879, "epoch": 4, "memory": 25074, "step": 17874} +{"lr": 0.0037825580157557948, "data_time": 0.0024126529693603515, "loss": 0.01927987444214523, "time": 0.8932074785232544, "epoch": 4, "memory": 25074, "step": 17974} +{"lr": 0.0037825580157557948, "data_time": 0.0021356821060180666, "loss": 0.015596888307482005, "time": 0.8333463430404663, "epoch": 4, "memory": 25074, "step": 18074} +{"lr": 0.0037825580157557948, "data_time": 0.0021569252014160155, "loss": 0.3951702803373337, "time": 0.837399172782898, "epoch": 4, "memory": 25074, "step": 18174} +{"lr": 0.0037825580157557948, "data_time": 0.0026964664459228514, "loss": 0.3264499008655548, "time": 0.8615692138671875, "epoch": 4, "memory": 25074, "step": 18274} +{"lr": 0.0037825580157557948, "data_time": 0.0015542745590209962, "loss": 0.2364593267440796, "time": 0.8392498970031739, "epoch": 4, "memory": 25074, "step": 18374} +{"lr": 0.0037825580157557948, "data_time": 0.0021144628524780275, "loss": 0.3408315718173981, "time": 0.8177517652511597, "epoch": 4, "memory": 25074, "step": 18474} +{"lr": 0.0037825580157557948, "data_time": 0.0024758577346801758, "loss": 0.19284270852804183, "time": 0.8166410446166992, "epoch": 4, "memory": 25074, "step": 18574} +{"lr": 0.0037825580157557948, "data_time": 0.0014041900634765626, "loss": 0.06831820160150529, "time": 0.816063928604126, "epoch": 4, "memory": 25074, "step": 18674} +{"lr": 0.0037825580157557948, "data_time": 0.0015158891677856446, "loss": 0.04160851426422596, "time": 0.8284622669219971, "epoch": 4, "memory": 25074, "step": 18774} +{"lr": 0.0037825580157557948, "data_time": 0.0013856887817382812, "loss": 0.029360038321465255, "time": 0.8102623224258423, "epoch": 4, "memory": 25074, "step": 18874} +{"lr": 0.0037825580157557948, "data_time": 0.0015660524368286133, "loss": 0.04482399821281433, "time": 0.8247085332870483, "epoch": 4, "memory": 25074, "step": 18974} +{"lr": 0.0037825580157557948, "data_time": 0.0017116308212280274, "loss": 0.043982202745974065, "time": 0.8885748147964477, "epoch": 4, "memory": 25074, "step": 19074} +{"lr": 0.0037825580157557948, "data_time": 0.0018639087677001954, "loss": 0.029614633601158857, "time": 0.8481374025344849, "epoch": 4, "memory": 25074, "step": 19174} +{"lr": 0.0037825580157557948, "data_time": 0.0021905422210693358, "loss": 0.024476656597107647, "time": 0.9164541006088257, "epoch": 4, "memory": 25074, "step": 19274} +{"lr": 0.0037825580157557948, "data_time": 0.0015946388244628905, "loss": 0.020974504435434936, "time": 0.7629599809646607, "epoch": 4, "memory": 25074, "step": 19374} +{"lr": 0.0037825580157557948, "data_time": 0.0016913175582885741, "loss": 0.987266993522644, "time": 0.8512732744216919, "epoch": 4, "memory": 25074, "step": 19474} +{"lr": 0.0037825580157557948, "data_time": 0.0016631841659545898, "loss": 0.686273193359375, "time": 0.8507126569747925, "epoch": 4, "memory": 25074, "step": 19574} +{"lr": 0.0037825580157557948, "data_time": 0.0019798040390014648, "loss": 0.6228355407714844, "time": 0.8141442537307739, "epoch": 4, "memory": 25074, "step": 19674} +{"lr": 0.0037825580157557948, "data_time": 0.0022313594818115234, "loss": 0.6229887008666992, "time": 0.8711118221282959, "epoch": 4, "memory": 25074, "step": 19774} +{"lr": 0.0037825580157557948, "data_time": 0.0015506982803344727, "loss": 0.5822794914245606, "time": 0.7789529323577881, "epoch": 4, "memory": 25074, "step": 19874} +{"lr": 0.0037825580157557948, "data_time": 0.0021116256713867186, "loss": 0.5362835168838501, "time": 0.8447186708450317, "epoch": 4, "memory": 25074, "step": 19974} +{"lr": 0.0037825580157557948, "data_time": 0.00142059326171875, "loss": 0.5363489866256714, "time": 0.8090475797653198, "epoch": 4, "memory": 25074, "step": 20074} +{"lr": 0.0037825580157557948, "data_time": 0.001723003387451172, "loss": 0.5079730987548828, "time": 0.7995928049087524, "epoch": 4, "memory": 25074, "step": 20174} +{"lr": 0.0037825580157557948, "data_time": 0.0025243282318115233, "loss": 0.48581901788711546, "time": 0.8065108060836792, "epoch": 4, "memory": 25074, "step": 20274} +{"lr": 0.0037825580157557948, "data_time": 0.0017581224441528321, "loss": 0.46407254934310915, "time": 0.8865103721618652, "epoch": 4, "memory": 25074, "step": 20374} +{"lr": 0.0037825580157557948, "data_time": 0.002043271064758301, "loss": 0.43721904754638674, "time": 0.8054994583129883, "epoch": 4, "memory": 25074, "step": 20474} +{"lr": 0.0037825580157557948, "data_time": 0.0016558647155761718, "loss": 0.42262349724769593, "time": 0.8218477725982666, "epoch": 4, "memory": 25074, "step": 20574} +{"lr": 0.0037825580157557948, "data_time": 0.002169299125671387, "loss": 0.40657807886600494, "time": 0.7935243129730225, "epoch": 4, "memory": 25074, "step": 20674} +{"lr": 0.0037825580157557948, "data_time": 0.0026780128479003905, "loss": 0.41469231247901917, "time": 0.8749217987060547, "epoch": 4, "memory": 25074, "step": 20774} +{"lr": 0.0037825580157557948, "data_time": 0.0031522274017333984, "loss": 0.36320524513721464, "time": 0.8123588562011719, "epoch": 4, "memory": 25074, "step": 20874} +{"lr": 0.0037825580157557948, "data_time": 0.001869368553161621, "loss": 0.36104739606380465, "time": 0.8755192041397095, "epoch": 4, "memory": 25074, "step": 20974} +{"lr": 0.0037825580157557948, "data_time": 0.0016811370849609375, "loss": 0.32246618568897245, "time": 0.8619577407836914, "epoch": 4, "memory": 25074, "step": 21074} +{"lr": 0.0037825580157557948, "data_time": 0.001760721206665039, "loss": 0.31507766246795654, "time": 0.9055258750915527, "epoch": 4, "memory": 25074, "step": 21174} +{"lr": 0.0037825580157557948, "data_time": 0.0018574953079223632, "loss": 0.33061912953853606, "time": 0.7908989429473877, "epoch": 4, "memory": 25074, "step": 21274} +{"lr": 0.0037825580157557948, "data_time": 0.0014024019241333009, "loss": 0.29776401072740555, "time": 0.7740065813064575, "epoch": 4, "memory": 25074, "step": 21374} +{"lr": 0.0037825580157557948, "data_time": 0.0017714262008666991, "loss": 0.2693121373653412, "time": 0.8174318552017212, "epoch": 4, "memory": 25074, "step": 21474} +{"lr": 0.0037825580157557948, "data_time": 0.0016476869583129882, "loss": 0.27400467842817305, "time": 0.7716795682907105, "epoch": 4, "memory": 25074, "step": 21574} +{"lr": 0.0037825580157557948, "data_time": 0.0016970872879028321, "loss": 0.26591013967990873, "time": 0.8105545759201049, "epoch": 4, "memory": 25074, "step": 21674} +{"lr": 0.0037825580157557948, "data_time": 0.0018676996231079101, "loss": 0.24770168960094452, "time": 0.8458791971206665, "epoch": 4, "memory": 25074, "step": 21774} +{"lr": 0.0037825580157557948, "data_time": 0.002076530456542969, "loss": 0.24303943514823914, "time": 0.861296558380127, "epoch": 4, "memory": 25074, "step": 21874} +{"lr": 0.0037825580157557948, "data_time": 0.0015017986297607422, "loss": 0.21657509952783585, "time": 0.7737709999084472, "epoch": 4, "memory": 25074, "step": 21974} +{"lr": 0.0037825580157557948, "data_time": 0.0017820358276367187, "loss": 0.22080043107271194, "time": 0.8244883775711059, "epoch": 4, "memory": 25074, "step": 22074} +{"lr": 0.0037825580157557948, "data_time": 0.0020010948181152345, "loss": 0.2656919628381729, "time": 0.8349753618240356, "epoch": 4, "memory": 25074, "step": 22174} +{"lr": 0.0037825580157557948, "data_time": 0.0013096332550048828, "loss": 0.19900043457746505, "time": 0.9053715467453003, "epoch": 4, "memory": 25074, "step": 22274} +{"lr": 0.0037825580157557948, "data_time": 0.0014603853225708008, "loss": 0.1767716571688652, "time": 0.8267655372619629, "epoch": 4, "memory": 25074, "step": 22374} +{"lr": 0.0037825580157557948, "data_time": 0.0028059720993041993, "loss": 0.18151189237833024, "time": 0.8090237855911255, "epoch": 4, "memory": 25074, "step": 22474} +{"lr": 0.0037825580157557948, "data_time": 0.0025251626968383787, "loss": 0.16442398279905318, "time": 0.755122184753418, "epoch": 4, "memory": 25074, "step": 22574} +{"lr": 0.0037825580157557948, "data_time": 0.001403474807739258, "loss": 0.219259575009346, "time": 0.9044676065444947, "epoch": 4, "memory": 25074, "step": 22674} +{"lr": 0.0037825580157557948, "data_time": 0.0019288301467895509, "loss": 0.21363774687051773, "time": 0.8168543100357055, "epoch": 4, "memory": 25074, "step": 22774} +{"lr": 0.0037825580157557948, "data_time": 0.0016801834106445312, "loss": 0.16989538371562957, "time": 0.9056608200073242, "epoch": 4, "memory": 25074, "step": 22874} +{"lr": 0.0037825580157557948, "data_time": 0.0020328760147094727, "loss": 0.1526303119957447, "time": 0.8601443052291871, "epoch": 4, "memory": 25074, "step": 22974} +{"accuracy/top1": 100.0, "data_time": 0.2402495917151956, "time": 0.6277029374066521, "step": 4} +{"lr": 0.0036189889037780212, "data_time": 0.002791881561279297, "loss": 0.1409598857164383, "time": 0.7361682176589965, "epoch": 5, "memory": 25074, "step": 23132} +{"lr": 0.0036189889037780212, "data_time": 0.002370285987854004, "loss": 0.13107298240065574, "time": 0.7838279008865356, "epoch": 5, "memory": 25074, "step": 23232} +{"lr": 0.0036189889037780212, "data_time": 0.002247118949890137, "loss": 0.1290941208600998, "time": 0.7731497287750244, "epoch": 5, "memory": 25074, "step": 23332} +{"lr": 0.0036189889037780212, "data_time": 0.0020523309707641602, "loss": 0.14304085224866867, "time": 0.835112190246582, "epoch": 5, "memory": 25074, "step": 23432} +{"lr": 0.0036189889037780212, "data_time": 0.0020323753356933593, "loss": 0.10982143506407738, "time": 0.8326578378677368, "epoch": 5, "memory": 25074, "step": 23532} +{"lr": 0.0036189889037780212, "data_time": 0.0021000146865844727, "loss": 0.14942233264446259, "time": 0.8026610374450683, "epoch": 5, "memory": 25074, "step": 23632} +{"lr": 0.0036189889037780212, "data_time": 0.0015743017196655274, "loss": 0.10007348507642747, "time": 0.8582959651947022, "epoch": 5, "memory": 25074, "step": 23732} +{"lr": 0.0036189889037780212, "data_time": 0.0023128509521484373, "loss": 0.11920400932431222, "time": 0.8757483005523682, "epoch": 5, "memory": 25074, "step": 23832} +{"lr": 0.0036189889037780212, "data_time": 0.0019636154174804688, "loss": 0.09923064410686493, "time": 0.7413851976394653, "epoch": 5, "memory": 25074, "step": 23932} +{"lr": 0.0036189889037780212, "data_time": 0.0018323898315429688, "loss": 0.5074950933456421, "time": 0.8162568092346192, "epoch": 5, "memory": 25074, "step": 24032} +{"lr": 0.0036189889037780212, "data_time": 0.002145814895629883, "loss": 0.2000981852412224, "time": 0.7811573505401611, "epoch": 5, "memory": 25074, "step": 24132} +{"lr": 0.0036189889037780212, "data_time": 0.0018929481506347657, "loss": 0.1282152235507965, "time": 0.9015360593795776, "epoch": 5, "memory": 25074, "step": 24232} +{"lr": 0.0036189889037780212, "data_time": 0.001922917366027832, "loss": 0.07424237728118896, "time": 0.7680579423904419, "epoch": 5, "memory": 25074, "step": 24332} +{"lr": 0.0036189889037780212, "data_time": 0.0016839981079101562, "loss": 0.1415588803589344, "time": 0.8025635004043579, "epoch": 5, "memory": 25074, "step": 24432} +{"lr": 0.0036189889037780212, "data_time": 0.0021357297897338866, "loss": 0.07980295084416866, "time": 1.0372174501419067, "epoch": 5, "memory": 25074, "step": 24532} +{"lr": 0.0036189889037780212, "data_time": 0.001970791816711426, "loss": 0.09773364216089249, "time": 0.8294796705245971, "epoch": 5, "memory": 25074, "step": 24632} +{"lr": 0.0036189889037780212, "data_time": 0.0015277624130249023, "loss": 0.07760869041085243, "time": 0.8010515213012696, "epoch": 5, "memory": 25074, "step": 24732} +{"lr": 0.0036189889037780212, "data_time": 0.0018419981002807616, "loss": 0.08253993056714534, "time": 0.7675093173980713, "epoch": 5, "memory": 25074, "step": 24832} +{"lr": 0.0036189889037780212, "data_time": 0.002334880828857422, "loss": 0.15845463946461677, "time": 0.8050356388092041, "epoch": 5, "memory": 25074, "step": 24932} +{"lr": 0.0036189889037780212, "data_time": 0.0022067785263061523, "loss": 0.0716423649340868, "time": 0.8166382074356079, "epoch": 5, "memory": 25074, "step": 25032} +{"lr": 0.0036189889037780212, "data_time": 0.0020513534545898438, "loss": 0.05277598164975643, "time": 0.9307221889495849, "epoch": 5, "memory": 25074, "step": 25132} +{"lr": 0.0036189889037780212, "data_time": 0.0015638113021850587, "loss": 0.055456786043941976, "time": 0.7789030075073242, "epoch": 5, "memory": 25074, "step": 25232} +{"lr": 0.0036189889037780212, "data_time": 0.002756953239440918, "loss": 0.06829598098993302, "time": 0.7972937822341919, "epoch": 5, "memory": 25074, "step": 25332} +{"lr": 0.0036189889037780212, "data_time": 0.0018646717071533203, "loss": 0.1625942789018154, "time": 0.8168657064437866, "epoch": 5, "memory": 25074, "step": 25432} +{"lr": 0.0036189889037780212, "data_time": 0.0019771814346313476, "loss": 0.05140396654605865, "time": 0.8372904300689697, "epoch": 5, "memory": 25074, "step": 25532} +{"lr": 0.0036189889037780212, "data_time": 0.0021970987319946287, "loss": 0.04322529584169388, "time": 0.8783126831054687, "epoch": 5, "memory": 25074, "step": 25632} +{"lr": 0.0036189889037780212, "data_time": 0.002334904670715332, "loss": 0.08789301030337811, "time": 0.7887928724288941, "epoch": 5, "memory": 25074, "step": 25732} +{"lr": 0.0036189889037780212, "data_time": 0.002146148681640625, "loss": 0.03907903414219618, "time": 0.8402746915817261, "epoch": 5, "memory": 25074, "step": 25832} +{"lr": 0.0036189889037780212, "data_time": 0.001930546760559082, "loss": 0.34184410572052004, "time": 0.8359517574310302, "epoch": 5, "memory": 25074, "step": 25932} +{"lr": 0.0036189889037780212, "data_time": 0.0015136480331420898, "loss": 0.053155054431408645, "time": 0.7919988393783569, "epoch": 5, "memory": 25074, "step": 26032} +{"lr": 0.0036189889037780212, "data_time": 0.0017973899841308594, "loss": 0.04823228809982538, "time": 0.8395230054855347, "epoch": 5, "memory": 25074, "step": 26132} +{"lr": 0.0036189889037780212, "data_time": 0.002233123779296875, "loss": 0.5153215765953064, "time": 0.8028883218765259, "epoch": 5, "memory": 25074, "step": 26232} +{"lr": 0.0036189889037780212, "data_time": 0.002189040184020996, "loss": 0.29221029579639435, "time": 0.8329488515853882, "epoch": 5, "memory": 25074, "step": 26332} +{"lr": 0.0036189889037780212, "data_time": 0.0020819664001464843, "loss": 0.1691926673054695, "time": 0.8191149234771729, "epoch": 5, "memory": 25074, "step": 26432} +{"lr": 0.0036189889037780212, "data_time": 0.002014946937561035, "loss": 0.071760324947536, "time": 0.8359919786453247, "epoch": 5, "memory": 25074, "step": 26532} +{"lr": 0.0036189889037780212, "data_time": 0.001936483383178711, "loss": 0.05709798783063889, "time": 0.8147225618362427, "epoch": 5, "memory": 25074, "step": 26632} +{"lr": 0.0036189889037780212, "data_time": 0.0022314310073852537, "loss": 0.03994304072111845, "time": 0.7918617963790894, "epoch": 5, "memory": 25074, "step": 26732} +{"lr": 0.0036189889037780212, "data_time": 0.0016577720642089843, "loss": 0.023929027561098336, "time": 0.7573018550872803, "epoch": 5, "memory": 25074, "step": 26832} +{"lr": 0.0036189889037780212, "data_time": 0.001753520965576172, "loss": 0.048914790898561475, "time": 0.8102521896362305, "epoch": 5, "memory": 25074, "step": 26932} +{"lr": 0.0036189889037780212, "data_time": 0.0019876241683959963, "loss": 0.022528800275176765, "time": 0.8083161830902099, "epoch": 5, "memory": 25074, "step": 27032} +{"lr": 0.0036189889037780212, "data_time": 0.002398037910461426, "loss": 0.03067467249929905, "time": 0.8007509708404541, "epoch": 5, "memory": 25074, "step": 27132} +{"lr": 0.0036189889037780212, "data_time": 0.0018538713455200195, "loss": 0.04042974412441254, "time": 0.7805203437805176, "epoch": 5, "memory": 25074, "step": 27232} +{"lr": 0.0036189889037780212, "data_time": 0.0015351533889770507, "loss": 0.02192025911062956, "time": 0.8094108581542969, "epoch": 5, "memory": 25074, "step": 27332} +{"lr": 0.0036189889037780212, "data_time": 0.002463555335998535, "loss": 0.021817026380449533, "time": 0.8124203443527221, "epoch": 5, "memory": 25074, "step": 27432} +{"lr": 0.0036189889037780212, "data_time": 0.0015733003616333007, "loss": 0.023982952069491147, "time": 0.7534865856170654, "epoch": 5, "memory": 25074, "step": 27532} +{"lr": 0.0036189889037780212, "data_time": 0.0021764039993286133, "loss": 0.277609246969223, "time": 0.7911419153213501, "epoch": 5, "memory": 25074, "step": 27632} +{"lr": 0.0036189889037780212, "data_time": 0.0014683961868286132, "loss": 0.02815322340466082, "time": 0.7884535789489746, "epoch": 5, "memory": 25074, "step": 27732} +{"lr": 0.0036189889037780212, "data_time": 0.0021384000778198243, "loss": 0.030094720143824815, "time": 0.7951118230819703, "epoch": 5, "memory": 25074, "step": 27832} +{"lr": 0.0036189889037780212, "data_time": 0.0016850471496582032, "loss": 0.021415264159440995, "time": 0.7606137752532959, "epoch": 5, "memory": 25074, "step": 27932} +{"lr": 0.0036189889037780212, "data_time": 0.0019913434982299803, "loss": 0.020857601333409546, "time": 0.8410851955413818, "epoch": 5, "memory": 25074, "step": 28032} +{"lr": 0.0036189889037780212, "data_time": 0.002279233932495117, "loss": 0.030796349234879016, "time": 0.7919528961181641, "epoch": 5, "memory": 25074, "step": 28132} +{"lr": 0.0036189889037780212, "data_time": 0.0018125057220458984, "loss": 0.7452868342399597, "time": 0.8143720149993896, "epoch": 5, "memory": 25074, "step": 28232} +{"lr": 0.0036189889037780212, "data_time": 0.0017468452453613282, "loss": 0.5256998062133789, "time": 0.7838029623031616, "epoch": 5, "memory": 25074, "step": 28332} +{"lr": 0.0036189889037780212, "data_time": 0.0015678882598876953, "loss": 0.3817441314458847, "time": 0.7522627592086792, "epoch": 5, "memory": 25074, "step": 28432} +{"lr": 0.0036189889037780212, "data_time": 0.0015955686569213868, "loss": 0.2653026685118675, "time": 0.7718812465667725, "epoch": 5, "memory": 25074, "step": 28532} +{"lr": 0.0036189889037780212, "data_time": 0.0018198490142822266, "loss": 0.5123111963272095, "time": 0.8409427165985107, "epoch": 5, "memory": 25074, "step": 28632} +{"lr": 0.0036189889037780212, "data_time": 0.0015866756439208984, "loss": 0.2875907853245735, "time": 0.7826916933059692, "epoch": 5, "memory": 25074, "step": 28732} +{"accuracy/top1": 100.0, "data_time": 0.23484166930703557, "time": 0.6201901716344497, "step": 5} +{"lr": 0.003415678028467163, "data_time": 0.0018018245697021484, "loss": 0.17174191698431968, "time": 0.7909611940383912, "epoch": 6, "memory": 25074, "step": 28890} +{"lr": 0.003415678028467163, "data_time": 0.0017652511596679688, "loss": 0.0633200965821743, "time": 0.8031851530075074, "epoch": 6, "memory": 25074, "step": 28990} +{"lr": 0.003415678028467163, "data_time": 0.0019463062286376952, "loss": 0.03496393393725157, "time": 0.7924827337265015, "epoch": 6, "memory": 25074, "step": 29090} +{"lr": 0.003415678028467163, "data_time": 0.0024018049240112304, "loss": 0.02982586110010743, "time": 0.821974778175354, "epoch": 6, "memory": 25074, "step": 29190} +{"lr": 0.003415678028467163, "data_time": 0.0015446901321411132, "loss": 0.04476702082902193, "time": 0.7985591888427734, "epoch": 6, "memory": 25074, "step": 29290} +{"lr": 0.003415678028467163, "data_time": 0.0023563623428344725, "loss": 0.028352122567594052, "time": 0.7858057975769043, "epoch": 6, "memory": 25074, "step": 29390} +{"lr": 0.003415678028467163, "data_time": 0.0019240856170654296, "loss": 0.0374476021155715, "time": 0.8087836265563965, "epoch": 6, "memory": 25074, "step": 29490} +{"lr": 0.003415678028467163, "data_time": 0.0023302078247070313, "loss": 0.024446723610162736, "time": 0.7831089735031128, "epoch": 6, "memory": 25074, "step": 29590} +{"lr": 0.003415678028467163, "data_time": 0.0026931524276733398, "loss": 0.027217399375513197, "time": 0.8390152931213379, "epoch": 6, "memory": 25074, "step": 29690} +{"lr": 0.003415678028467163, "data_time": 0.0023102045059204103, "loss": 0.04310166211798787, "time": 0.7135088682174683, "epoch": 6, "memory": 25074, "step": 29790} +{"lr": 0.003415678028467163, "data_time": 0.002012801170349121, "loss": 0.01963822771795094, "time": 0.8424479722976684, "epoch": 6, "memory": 25074, "step": 29890} +{"lr": 0.003415678028467163, "data_time": 0.0021408796310424805, "loss": 0.025779563747346402, "time": 0.7580039501190186, "epoch": 6, "memory": 25074, "step": 29990} +{"lr": 0.003415678028467163, "data_time": 0.0019848823547363283, "loss": 0.0160231560934335, "time": 0.8077895164489746, "epoch": 6, "memory": 25074, "step": 30090} +{"lr": 0.003415678028467163, "data_time": 0.0022003173828125, "loss": 0.023282242938876152, "time": 0.8337722539901733, "epoch": 6, "memory": 25074, "step": 30190} +{"lr": 0.003415678028467163, "data_time": 0.0017018318176269531, "loss": 0.026293031871318817, "time": 0.7885631799697876, "epoch": 6, "memory": 25074, "step": 30290} +{"lr": 0.003415678028467163, "data_time": 0.001690506935119629, "loss": 0.019426302472129463, "time": 0.7699220418930054, "epoch": 6, "memory": 25074, "step": 30390} +{"lr": 0.003415678028467163, "data_time": 0.001882314682006836, "loss": 0.024549905210733414, "time": 0.7650113105773926, "epoch": 6, "memory": 25074, "step": 30490} +{"lr": 0.003415678028467163, "data_time": 0.0013676881790161133, "loss": 0.9105373382568359, "time": 0.7781325578689575, "epoch": 6, "memory": 25074, "step": 30590} +{"lr": 0.003415678028467163, "data_time": 0.001557326316833496, "loss": 0.6870330810546875, "time": 0.7517716884613037, "epoch": 6, "memory": 25074, "step": 30690} +{"lr": 0.003415678028467163, "data_time": 0.0018422365188598632, "loss": 0.6858840942382812, "time": 0.809003210067749, "epoch": 6, "memory": 25074, "step": 30790} +{"lr": 0.003415678028467163, "data_time": 0.0016306638717651367, "loss": 0.6863113403320312, "time": 0.8211899518966674, "epoch": 6, "memory": 25074, "step": 30890} +{"lr": 0.003415678028467163, "data_time": 0.001467752456665039, "loss": 0.689276123046875, "time": 0.760313606262207, "epoch": 6, "memory": 25074, "step": 30990} +{"lr": 0.003415678028467163, "data_time": 0.0016613483428955078, "loss": 0.6889556884765625, "time": 0.7955705165863037, "epoch": 6, "memory": 25074, "step": 31090} +{"lr": 0.003415678028467163, "data_time": 0.001390528678894043, "loss": 0.6836715698242187, "time": 0.814949369430542, "epoch": 6, "memory": 25074, "step": 31190} +{"lr": 0.003415678028467163, "data_time": 0.06151556968688965, "loss": 0.6795379638671875, "time": 0.7687273979187011, "epoch": 6, "memory": 25074, "step": 31290} +{"lr": 0.003415678028467163, "data_time": 0.0012712955474853515, "loss": 0.6829338073730469, "time": 0.7202790260314942, "epoch": 6, "memory": 25074, "step": 31390} +{"lr": 0.003415678028467163, "data_time": 0.0014473199844360352, "loss": 0.6826400756835938, "time": 0.8305152893066406, "epoch": 6, "memory": 25074, "step": 31490} +{"lr": 0.003415678028467163, "data_time": 0.0018189430236816406, "loss": 0.6878402709960938, "time": 0.8051482677459717, "epoch": 6, "memory": 25074, "step": 31590} +{"lr": 0.003415678028467163, "data_time": 0.0013915538787841798, "loss": 0.6876800537109375, "time": 0.7747416973114014, "epoch": 6, "memory": 25074, "step": 31690} +{"lr": 0.003415678028467163, "data_time": 0.0015104293823242187, "loss": 0.6831192016601563, "time": 0.7355213165283203, "epoch": 6, "memory": 25074, "step": 31790} +{"lr": 0.003415678028467163, "data_time": 0.0018348217010498047, "loss": 0.6800773620605469, "time": 0.7252334833145142, "epoch": 6, "memory": 25074, "step": 31890} +{"lr": 0.003415678028467163, "data_time": 0.0016130924224853516, "loss": 0.6558177947998047, "time": 0.7361180305480957, "epoch": 6, "memory": 25074, "step": 31990} +{"lr": 0.003415678028467163, "data_time": 0.002025461196899414, "loss": 0.6396217346191406, "time": 0.7465869903564453, "epoch": 6, "memory": 25074, "step": 32090} +{"lr": 0.003415678028467163, "data_time": 0.001508045196533203, "loss": 0.5973075866699219, "time": 0.7108067750930787, "epoch": 6, "memory": 25074, "step": 32190} +{"lr": 0.003415678028467163, "data_time": 0.0015272140502929688, "loss": 0.5695059776306153, "time": 0.7145503282546997, "epoch": 6, "memory": 25074, "step": 32290} +{"lr": 0.003415678028467163, "data_time": 0.0019104957580566406, "loss": 0.5324402332305909, "time": 0.7866726398468018, "epoch": 6, "memory": 25074, "step": 32390} +{"lr": 0.003415678028467163, "data_time": 0.0016667366027832032, "loss": 0.5393310070037842, "time": 0.7764627933502197, "epoch": 6, "memory": 25074, "step": 32490} +{"lr": 0.003415678028467163, "data_time": 0.0019632816314697266, "loss": 0.4775394558906555, "time": 0.8011926412582397, "epoch": 6, "memory": 25074, "step": 32590} +{"lr": 0.003415678028467163, "data_time": 0.0019040107727050781, "loss": 0.46249977946281434, "time": 0.7955193758010864, "epoch": 6, "memory": 25074, "step": 32690} +{"lr": 0.003415678028467163, "data_time": 0.0021024227142333986, "loss": 0.43382647037506106, "time": 0.833111572265625, "epoch": 6, "memory": 25074, "step": 32790} +{"lr": 0.003415678028467163, "data_time": 0.001708245277404785, "loss": 0.39271233677864076, "time": 0.8732306241989136, "epoch": 6, "memory": 25074, "step": 32890} +{"lr": 0.003415678028467163, "data_time": 0.0018872737884521485, "loss": 0.39725507497787477, "time": 0.8073127508163452, "epoch": 6, "memory": 25074, "step": 32990} +{"lr": 0.003415678028467163, "data_time": 0.0018921613693237305, "loss": 0.34503606855869295, "time": 0.8197930812835693, "epoch": 6, "memory": 25074, "step": 33090} +{"lr": 0.003415678028467163, "data_time": 0.001664447784423828, "loss": 0.34498479068279264, "time": 0.825571870803833, "epoch": 6, "memory": 25074, "step": 33190} +{"lr": 0.003415678028467163, "data_time": 0.0015811443328857422, "loss": 0.30077325105667113, "time": 0.8408125638961792, "epoch": 6, "memory": 25074, "step": 33290} +{"lr": 0.003415678028467163, "data_time": 0.0018062353134155273, "loss": 0.28608314096927645, "time": 0.8217522144317627, "epoch": 6, "memory": 25074, "step": 33390} +{"lr": 0.003415678028467163, "data_time": 0.0020669937133789063, "loss": 0.27477487176656723, "time": 0.8694106578826905, "epoch": 6, "memory": 25074, "step": 33490} +{"lr": 0.003415678028467163, "data_time": 0.0017047166824340821, "loss": 0.24026929885149, "time": 0.8727396726608276, "epoch": 6, "memory": 25074, "step": 33590} +{"lr": 0.003415678028467163, "data_time": 0.0023725271224975587, "loss": 0.19539487808942796, "time": 0.8610022783279419, "epoch": 6, "memory": 25074, "step": 33690} +{"lr": 0.003415678028467163, "data_time": 0.0020278453826904296, "loss": 0.22524606734514235, "time": 0.7790164947509766, "epoch": 6, "memory": 25074, "step": 33790} +{"lr": 0.003415678028467163, "data_time": 0.001953601837158203, "loss": 0.21684082001447677, "time": 0.8055158138275147, "epoch": 6, "memory": 25074, "step": 33890} +{"lr": 0.003415678028467163, "data_time": 0.001944732666015625, "loss": 0.17871028035879136, "time": 0.7842403173446655, "epoch": 6, "memory": 25074, "step": 33990} +{"lr": 0.003415678028467163, "data_time": 0.002075338363647461, "loss": 0.3113388612866402, "time": 0.8257120609283447, "epoch": 6, "memory": 25074, "step": 34090} +{"lr": 0.003415678028467163, "data_time": 0.0019263029098510742, "loss": 0.19416865408420564, "time": 0.785835599899292, "epoch": 6, "memory": 25074, "step": 34190} +{"lr": 0.003415678028467163, "data_time": 0.0021706342697143553, "loss": 0.16097232401371003, "time": 0.8521947145462037, "epoch": 6, "memory": 25074, "step": 34290} +{"lr": 0.003415678028467163, "data_time": 0.0021689891815185546, "loss": 0.12806736528873444, "time": 0.8027761936187744, "epoch": 6, "memory": 25074, "step": 34390} +{"lr": 0.003415678028467163, "data_time": 0.001958608627319336, "loss": 0.08435219004750252, "time": 0.8249452114105225, "epoch": 6, "memory": 25074, "step": 34490} +{"accuracy/top1": 100.0, "data_time": 0.22624763320474064, "time": 0.6205435079686782, "step": 6} +{"lr": 0.003177631578323485, "data_time": 0.0017069339752197265, "loss": 0.08231379240751266, "time": 0.7843175649642944, "epoch": 7, "memory": 25074, "step": 34648} +{"lr": 0.003177631578323485, "data_time": 0.0022977590560913086, "loss": 0.08582820855081082, "time": 0.7684028387069702, "epoch": 7, "memory": 25074, "step": 34748} +{"lr": 0.003177631578323485, "data_time": 0.001708054542541504, "loss": 0.08164712898433209, "time": 0.8050656318664551, "epoch": 7, "memory": 25074, "step": 34848} +{"lr": 0.003177631578323485, "data_time": 0.0018536806106567382, "loss": 0.10651303324848413, "time": 0.8809154272079468, "epoch": 7, "memory": 25074, "step": 34948} +{"lr": 0.003177631578323485, "data_time": 0.002226734161376953, "loss": 0.055012472718954084, "time": 0.7990552902221679, "epoch": 7, "memory": 25074, "step": 35048} +{"lr": 0.003177631578323485, "data_time": 0.002142786979675293, "loss": 0.06312011554837227, "time": 0.8404693365097046, "epoch": 7, "memory": 25074, "step": 35148} +{"lr": 0.003177631578323485, "data_time": 0.001969790458679199, "loss": 0.053021703660488126, "time": 0.8096812725067138, "epoch": 7, "memory": 25074, "step": 35248} +{"lr": 0.003177631578323485, "data_time": 0.0022991418838500975, "loss": 0.05477383937686682, "time": 0.8274132013320923, "epoch": 7, "memory": 25074, "step": 35348} +{"lr": 0.003177631578323485, "data_time": 0.001857590675354004, "loss": 0.059876644238829616, "time": 0.8286162137985229, "epoch": 7, "memory": 25074, "step": 35448} +{"lr": 0.003177631578323485, "data_time": 0.0020510673522949217, "loss": 0.03685490805655718, "time": 0.8613656282424926, "epoch": 7, "memory": 25074, "step": 35548} +{"lr": 0.003177631578323485, "data_time": 0.0019825220108032225, "loss": 0.02592898178845644, "time": 0.8106374502182007, "epoch": 7, "memory": 25074, "step": 35648} +{"lr": 0.003177631578323485, "data_time": 0.0016251325607299805, "loss": 0.05490502044558525, "time": 0.7780727624893189, "epoch": 7, "memory": 25074, "step": 35748} +{"lr": 0.003177631578323485, "data_time": 0.0018387556076049805, "loss": 0.04498982932418585, "time": 0.8170005559921265, "epoch": 7, "memory": 25074, "step": 35848} +{"lr": 0.003177631578323485, "data_time": 0.0027972936630249025, "loss": 0.034482643753290174, "time": 0.9129539012908936, "epoch": 7, "memory": 25074, "step": 35948} +{"lr": 0.003177631578323485, "data_time": 0.0016614913940429688, "loss": 0.5376731395721436, "time": 0.8564528703689576, "epoch": 7, "memory": 25074, "step": 36048} +{"lr": 0.003177631578323485, "data_time": 0.0023099660873413088, "loss": 0.3870724499225616, "time": 0.8048108577728271, "epoch": 7, "memory": 25074, "step": 36148} +{"lr": 0.003177631578323485, "data_time": 0.0018640518188476562, "loss": 0.25513831079006194, "time": 0.8086620092391967, "epoch": 7, "memory": 25074, "step": 36248} +{"lr": 0.003177631578323485, "data_time": 0.0016347885131835938, "loss": 0.10172961838543415, "time": 0.7994915008544922, "epoch": 7, "memory": 25074, "step": 36348} +{"lr": 0.003177631578323485, "data_time": 0.0017717599868774414, "loss": 0.05500081535428762, "time": 0.7761198282241821, "epoch": 7, "memory": 25074, "step": 36448} +{"lr": 0.003177631578323485, "data_time": 0.0016062021255493163, "loss": 0.03965157438069582, "time": 0.82099609375, "epoch": 7, "memory": 25074, "step": 36548} +{"lr": 0.003177631578323485, "data_time": 0.001989150047302246, "loss": 0.033694837940856814, "time": 0.8073045492172242, "epoch": 7, "memory": 25074, "step": 36648} +{"lr": 0.003177631578323485, "data_time": 0.0015305519104003907, "loss": 0.051020176336169246, "time": 0.8328912734985352, "epoch": 7, "memory": 25074, "step": 36748} +{"lr": 0.003177631578323485, "data_time": 0.0017588138580322266, "loss": 0.03742875270545483, "time": 0.8235810279846192, "epoch": 7, "memory": 25074, "step": 36848} +{"lr": 0.003177631578323485, "data_time": 0.001764059066772461, "loss": 0.03348587434738874, "time": 0.7547449350357056, "epoch": 7, "memory": 25074, "step": 36948} +{"lr": 0.003177631578323485, "data_time": 0.0014856100082397462, "loss": 0.03297441750764847, "time": 0.811091947555542, "epoch": 7, "memory": 25074, "step": 37048} +{"lr": 0.003177631578323485, "data_time": 0.002025747299194336, "loss": 0.027984938118606806, "time": 0.7961791515350342, "epoch": 7, "memory": 25074, "step": 37148} +{"lr": 0.003177631578323485, "data_time": 0.001494598388671875, "loss": 0.019442465668544174, "time": 0.8026607275009155, "epoch": 7, "memory": 25074, "step": 37248} +{"lr": 0.003177631578323485, "data_time": 0.0013259649276733398, "loss": 0.4900744140148163, "time": 0.7862446784973145, "epoch": 7, "memory": 25074, "step": 37348} +{"lr": 0.003177631578323485, "data_time": 0.0015650272369384765, "loss": 0.38184359967708587, "time": 0.8663106203079224, "epoch": 7, "memory": 25074, "step": 37448} +{"lr": 0.003177631578323485, "data_time": 0.0014769077301025391, "loss": 0.3002003714442253, "time": 0.8492149591445923, "epoch": 7, "memory": 25074, "step": 37548} +{"lr": 0.003177631578323485, "data_time": 0.001516580581665039, "loss": 0.22087419629096985, "time": 0.7983831644058228, "epoch": 7, "memory": 25074, "step": 37648} +{"lr": 0.003177631578323485, "data_time": 0.001896071434020996, "loss": 0.1823909968137741, "time": 0.8249658346176147, "epoch": 7, "memory": 25074, "step": 37748} +{"lr": 0.003177631578323485, "data_time": 0.0029141902923583984, "loss": 0.0875114869326353, "time": 0.7948890686035156, "epoch": 7, "memory": 25074, "step": 37848} +{"lr": 0.003177631578323485, "data_time": 0.001998114585876465, "loss": 0.03376759458333254, "time": 0.8398072242736816, "epoch": 7, "memory": 25074, "step": 37948} +{"lr": 0.003177631578323485, "data_time": 0.0017653226852416993, "loss": 0.06241262629628182, "time": 0.7995390892028809, "epoch": 7, "memory": 25074, "step": 38048} +{"lr": 0.003177631578323485, "data_time": 0.0017367124557495118, "loss": 0.026532156532630323, "time": 0.7869459390640259, "epoch": 7, "memory": 25074, "step": 38148} +{"lr": 0.003177631578323485, "data_time": 0.0021256446838378907, "loss": 0.023698237910866737, "time": 0.837284255027771, "epoch": 7, "memory": 25074, "step": 38248} +{"lr": 0.003177631578323485, "data_time": 0.002065277099609375, "loss": 0.04279262721538544, "time": 0.9142627954483032, "epoch": 7, "memory": 25074, "step": 38348} +{"lr": 0.003177631578323485, "data_time": 0.0024343013763427736, "loss": 0.020273453602567316, "time": 0.7826040983200073, "epoch": 7, "memory": 25074, "step": 38448} +{"lr": 0.003177631578323485, "data_time": 0.0016155242919921875, "loss": 0.01841148789972067, "time": 0.8423229694366455, "epoch": 7, "memory": 25074, "step": 38548} +{"lr": 0.003177631578323485, "data_time": 0.0017494440078735351, "loss": 0.031658689863979815, "time": 0.8426234483718872, "epoch": 7, "memory": 25074, "step": 38648} +{"lr": 0.003177631578323485, "data_time": 0.0018595218658447265, "loss": 0.06796988733112812, "time": 0.7892760753631591, "epoch": 7, "memory": 25074, "step": 38748} +{"lr": 0.003177631578323485, "data_time": 0.002170705795288086, "loss": 0.015778914047405123, "time": 0.8710393190383912, "epoch": 7, "memory": 25074, "step": 38848} +{"lr": 0.003177631578323485, "data_time": 0.0017850399017333984, "loss": 0.018908139877021314, "time": 0.8602873325347901, "epoch": 7, "memory": 25074, "step": 38948} +{"lr": 0.003177631578323485, "data_time": 0.0016251087188720703, "loss": 0.01837123604491353, "time": 0.8457715511322021, "epoch": 7, "memory": 25074, "step": 39048} +{"lr": 0.003177631578323485, "data_time": 0.0017221927642822265, "loss": 0.01905709123238921, "time": 0.8067245244979858, "epoch": 7, "memory": 25074, "step": 39148} +{"lr": 0.003177631578323485, "data_time": 0.00210263729095459, "loss": 0.013219493068754673, "time": 0.8204404830932617, "epoch": 7, "memory": 25074, "step": 39248} +{"lr": 0.003177631578323485, "data_time": 0.001557326316833496, "loss": 0.025120821595191956, "time": 0.7940375328063964, "epoch": 7, "memory": 25074, "step": 39348} +{"lr": 0.003177631578323485, "data_time": 0.0016313314437866211, "loss": 0.02124868012033403, "time": 0.9074863195419312, "epoch": 7, "memory": 25074, "step": 39448} +{"lr": 0.003177631578323485, "data_time": 0.0014491319656372071, "loss": 0.6646780014038086, "time": 0.7914240837097168, "epoch": 7, "memory": 25074, "step": 39548} +{"lr": 0.003177631578323485, "data_time": 0.001481318473815918, "loss": 0.6365413665771484, "time": 0.8293594598770142, "epoch": 7, "memory": 25074, "step": 39648} +{"lr": 0.003177631578323485, "data_time": 0.0016228437423706054, "loss": 0.6289838790893555, "time": 0.7879393100738525, "epoch": 7, "memory": 25074, "step": 39748} +{"lr": 0.003177631578323485, "data_time": 0.0018411874771118164, "loss": 0.6199359893798828, "time": 0.8507879257202149, "epoch": 7, "memory": 25074, "step": 39848} +{"lr": 0.003177631578323485, "data_time": 0.0017439842224121094, "loss": 0.6042745590209961, "time": 0.8092621564865112, "epoch": 7, "memory": 25074, "step": 39948} +{"lr": 0.003177631578323485, "data_time": 0.0021016836166381837, "loss": 0.5974456787109375, "time": 0.7967112302780152, "epoch": 7, "memory": 25074, "step": 40048} +{"lr": 0.003177631578323485, "data_time": 0.0015798568725585937, "loss": 0.5755602836608886, "time": 0.7882761001586914, "epoch": 7, "memory": 25074, "step": 40148} +{"lr": 0.003177631578323485, "data_time": 0.0012903928756713868, "loss": 0.5639491558074952, "time": 0.8141366958618164, "epoch": 7, "memory": 25074, "step": 40248} +{"accuracy/top1": 100.0, "data_time": 0.22547212769003475, "time": 0.6097386163823745, "step": 7} +{"lr": 0.0029107110469803964, "data_time": 0.001800227165222168, "loss": 0.5362431287765503, "time": 0.7892891883850097, "epoch": 8, "memory": 25074, "step": 40406} +{"lr": 0.0029107110469803964, "data_time": 0.0021252870559692384, "loss": 0.5108250856399537, "time": 0.8145163297653198, "epoch": 8, "memory": 25074, "step": 40506} +{"lr": 0.0029107110469803964, "data_time": 0.0026922702789306642, "loss": 0.47698734998703, "time": 0.8064170360565186, "epoch": 8, "memory": 25074, "step": 40606} +{"lr": 0.0029107110469803964, "data_time": 0.002124929428100586, "loss": 0.4469404309988022, "time": 0.7941334486007691, "epoch": 8, "memory": 25074, "step": 40706} +{"lr": 0.0029107110469803964, "data_time": 0.0013925552368164063, "loss": 0.43811762928962705, "time": 0.7753614187240601, "epoch": 8, "memory": 25074, "step": 40806} +{"lr": 0.0029107110469803964, "data_time": 0.001997566223144531, "loss": 0.40253242254257204, "time": 0.8069181680679322, "epoch": 8, "memory": 25074, "step": 40906} +{"lr": 0.0029107110469803964, "data_time": 0.0015981435775756837, "loss": 0.39082804024219514, "time": 0.8386501789093017, "epoch": 8, "memory": 25074, "step": 41006} +{"lr": 0.0029107110469803964, "data_time": 0.0014615535736083984, "loss": 0.3861152559518814, "time": 0.785384202003479, "epoch": 8, "memory": 25074, "step": 41106} +{"lr": 0.0029107110469803964, "data_time": 0.001982879638671875, "loss": 0.38163594603538514, "time": 0.8258617162704468, "epoch": 8, "memory": 25074, "step": 41206} +{"lr": 0.0029107110469803964, "data_time": 0.0014397144317626954, "loss": 0.3460691422224045, "time": 0.8217541933059692, "epoch": 8, "memory": 25074, "step": 41306} +{"lr": 0.0029107110469803964, "data_time": 0.001651453971862793, "loss": 0.33383490443229674, "time": 0.8306535720825196, "epoch": 8, "memory": 25074, "step": 41406} +{"lr": 0.0029107110469803964, "data_time": 0.0013466358184814453, "loss": 0.31845238506793977, "time": 0.8040345668792724, "epoch": 8, "memory": 25074, "step": 41506} +{"lr": 0.0029107110469803964, "data_time": 0.0013608455657958985, "loss": 0.34845821261405946, "time": 0.8786239385604858, "epoch": 8, "memory": 25074, "step": 41606} +{"lr": 0.0029107110469803964, "data_time": 0.0013670921325683594, "loss": 0.278292253613472, "time": 0.8874995946884155, "epoch": 8, "memory": 25074, "step": 41706} +{"lr": 0.0029107110469803964, "data_time": 0.0013338565826416016, "loss": 0.31247431337833403, "time": 0.8517024755477905, "epoch": 8, "memory": 25074, "step": 41806} +{"lr": 0.0029107110469803964, "data_time": 0.0014211416244506836, "loss": 0.2974017262458801, "time": 0.8422387838363647, "epoch": 8, "memory": 25074, "step": 41906} +{"lr": 0.0029107110469803964, "data_time": 0.0015972614288330077, "loss": 0.28323196023702624, "time": 0.7977926969528198, "epoch": 8, "memory": 25074, "step": 42006} +{"lr": 0.0029107110469803964, "data_time": 0.0013134002685546875, "loss": 0.2619663327932358, "time": 0.802070426940918, "epoch": 8, "memory": 25074, "step": 42106} +{"lr": 0.0029107110469803964, "data_time": 0.0014792442321777343, "loss": 0.26130121052265165, "time": 0.7949286222457885, "epoch": 8, "memory": 25074, "step": 42206} +{"lr": 0.0029107110469803964, "data_time": 0.001403498649597168, "loss": 0.264859938621521, "time": 0.8208257913589477, "epoch": 8, "memory": 25074, "step": 42306} +{"lr": 0.0029107110469803964, "data_time": 0.001445770263671875, "loss": 0.2695415988564491, "time": 0.844876766204834, "epoch": 8, "memory": 25074, "step": 42406} +{"lr": 0.0029107110469803964, "data_time": 0.002265477180480957, "loss": 0.2874342992901802, "time": 0.8235554456710815, "epoch": 8, "memory": 25074, "step": 42506} +{"lr": 0.0029107110469803964, "data_time": 0.0021061420440673826, "loss": 0.24466877281665803, "time": 0.8314491510391235, "epoch": 8, "memory": 25074, "step": 42606} +{"lr": 0.0029107110469803964, "data_time": 0.0020581960678100588, "loss": 0.23984656482934952, "time": 0.8473775148391723, "epoch": 8, "memory": 25074, "step": 42706} +{"lr": 0.0029107110469803964, "data_time": 0.00217893123626709, "loss": 0.23493181318044662, "time": 0.7882248878479003, "epoch": 8, "memory": 25074, "step": 42806} +{"lr": 0.0029107110469803964, "data_time": 0.0023426532745361326, "loss": 0.23431477695703506, "time": 0.8113497257232666, "epoch": 8, "memory": 25074, "step": 42906} +{"lr": 0.0029107110469803964, "data_time": 0.0024402379989624024, "loss": 0.2268846094608307, "time": 0.8288700580596924, "epoch": 8, "memory": 25074, "step": 43006} +{"lr": 0.0029107110469803964, "data_time": 0.0019439220428466796, "loss": 0.21778939068317413, "time": 0.7623370170593262, "epoch": 8, "memory": 25074, "step": 43106} +{"lr": 0.0029107110469803964, "data_time": 0.0022038936614990233, "loss": 0.2172636419534683, "time": 0.7880473852157592, "epoch": 8, "memory": 25074, "step": 43206} +{"lr": 0.0029107110469803964, "data_time": 0.002065873146057129, "loss": 0.2091320648789406, "time": 0.8095156192779541, "epoch": 8, "memory": 25074, "step": 43306} +{"lr": 0.0029107110469803964, "data_time": 0.0018571853637695313, "loss": 0.17143266126513482, "time": 0.810135793685913, "epoch": 8, "memory": 25074, "step": 43406} +{"lr": 0.0029107110469803964, "data_time": 0.002043342590332031, "loss": 0.20797404944896697, "time": 0.8592607259750367, "epoch": 8, "memory": 25074, "step": 43506} +{"lr": 0.0029107110469803964, "data_time": 0.0015845775604248046, "loss": 0.18679274320602418, "time": 0.813219141960144, "epoch": 8, "memory": 25074, "step": 43606} +{"lr": 0.0029107110469803964, "data_time": 0.00219423770904541, "loss": 0.16646159291267396, "time": 0.7774173498153687, "epoch": 8, "memory": 25074, "step": 43706} +{"lr": 0.0029107110469803964, "data_time": 0.0021094560623168947, "loss": 0.1844821035861969, "time": 0.8706939220428467, "epoch": 8, "memory": 25074, "step": 43806} +{"lr": 0.0029107110469803964, "data_time": 0.002427840232849121, "loss": 0.16617073863744736, "time": 0.7723523139953613, "epoch": 8, "memory": 25074, "step": 43906} +{"lr": 0.0029107110469803964, "data_time": 0.0016525983810424805, "loss": 0.16185368224978447, "time": 0.798746919631958, "epoch": 8, "memory": 25074, "step": 44006} +{"lr": 0.0029107110469803964, "data_time": 0.0013972043991088868, "loss": 0.14784356951713562, "time": 0.8377389192581177, "epoch": 8, "memory": 25074, "step": 44106} +{"lr": 0.0029107110469803964, "data_time": 0.0014557123184204101, "loss": 0.15821471065282822, "time": 0.8408293485641479, "epoch": 8, "memory": 25074, "step": 44206} +{"lr": 0.0029107110469803964, "data_time": 0.0015608787536621094, "loss": 0.13272587954998016, "time": 0.836663794517517, "epoch": 8, "memory": 25074, "step": 44306} +{"lr": 0.0029107110469803964, "data_time": 0.0016949176788330078, "loss": 0.12774054631590842, "time": 0.820939040184021, "epoch": 8, "memory": 25074, "step": 44406} +{"lr": 0.0029107110469803964, "data_time": 0.0017126798629760742, "loss": 0.1299731768667698, "time": 0.8248666763305664, "epoch": 8, "memory": 25074, "step": 44506} +{"lr": 0.0029107110469803964, "data_time": 0.0018166542053222657, "loss": 0.13998845592141151, "time": 0.780283761024475, "epoch": 8, "memory": 25074, "step": 44606} +{"lr": 0.0029107110469803964, "data_time": 0.0018818140029907226, "loss": 0.11496206223964692, "time": 0.832054591178894, "epoch": 8, "memory": 25074, "step": 44706} +{"lr": 0.0029107110469803964, "data_time": 0.001940298080444336, "loss": 0.1016205944120884, "time": 0.8187478303909301, "epoch": 8, "memory": 25074, "step": 44806} +{"lr": 0.0029107110469803964, "data_time": 0.00205533504486084, "loss": 0.09862435236573219, "time": 0.7878509283065795, "epoch": 8, "memory": 25074, "step": 44906} +{"lr": 0.0029107110469803964, "data_time": 0.00294034481048584, "loss": 0.07922623381018638, "time": 0.7465308904647827, "epoch": 8, "memory": 25074, "step": 45006} +{"lr": 0.0029107110469803964, "data_time": 0.002628660202026367, "loss": 0.09312311969697476, "time": 0.7900871753692627, "epoch": 8, "memory": 25074, "step": 45106} +{"lr": 0.0029107110469803964, "data_time": 0.0020933628082275392, "loss": 0.09835705608129501, "time": 0.8151920795440674, "epoch": 8, "memory": 25074, "step": 45206} +{"lr": 0.0029107110469803964, "data_time": 0.0021629810333251955, "loss": 0.07780469469726085, "time": 0.807362413406372, "epoch": 8, "memory": 25074, "step": 45306} +{"lr": 0.0029107110469803964, "data_time": 0.0018307924270629882, "loss": 0.06943958066403866, "time": 0.8148200273513794, "epoch": 8, "memory": 25074, "step": 45406} +{"lr": 0.0029107110469803964, "data_time": 0.0014321565628051757, "loss": 0.06271064318716527, "time": 0.8063824415206909, "epoch": 8, "memory": 25074, "step": 45506} +{"lr": 0.0029107110469803964, "data_time": 0.0016618967056274414, "loss": 0.08301684185862541, "time": 0.7676124334335327, "epoch": 8, "memory": 25074, "step": 45606} +{"lr": 0.0029107110469803964, "data_time": 0.0019184350967407227, "loss": 0.06498735733330249, "time": 0.8050572395324707, "epoch": 8, "memory": 25074, "step": 45706} +{"lr": 0.0029107110469803964, "data_time": 0.002342534065246582, "loss": 0.07699587866663933, "time": 0.8105920076370239, "epoch": 8, "memory": 25074, "step": 45806} +{"lr": 0.0029107110469803964, "data_time": 0.001531982421875, "loss": 0.042582137137651445, "time": 0.8026448488235474, "epoch": 8, "memory": 25074, "step": 45906} +{"lr": 0.0029107110469803964, "data_time": 0.001994681358337402, "loss": 0.064727021753788, "time": 0.8076073884963989, "epoch": 8, "memory": 25074, "step": 46006} +{"accuracy/top1": 100.0, "data_time": 0.23159482899834127, "time": 0.625993434120627, "step": 8} +{"lr": 0.0026214889037780207, "data_time": 0.001961421966552734, "loss": 0.047271404415369034, "time": 0.793843674659729, "epoch": 9, "memory": 25074, "step": 46164} +{"lr": 0.0026214889037780207, "data_time": 0.0014253854751586914, "loss": 0.037891981471329925, "time": 0.8486331462860107, "epoch": 9, "memory": 25074, "step": 46264} +{"lr": 0.0026214889037780207, "data_time": 0.002214503288269043, "loss": 0.04856832753866911, "time": 0.8137295246124268, "epoch": 9, "memory": 25074, "step": 46364} +{"lr": 0.0026214889037780207, "data_time": 0.0021689891815185546, "loss": 0.03917246051132679, "time": 0.8412569761276245, "epoch": 9, "memory": 25074, "step": 46464} +{"lr": 0.0026214889037780207, "data_time": 0.0021068334579467775, "loss": 0.050056328624486925, "time": 0.8025254487991333, "epoch": 9, "memory": 25074, "step": 46564} +{"lr": 0.0026214889037780207, "data_time": 0.0019266605377197266, "loss": 0.06529371645301581, "time": 0.8107960939407348, "epoch": 9, "memory": 25074, "step": 46664} +{"lr": 0.0026214889037780207, "data_time": 0.0016360282897949219, "loss": 0.038940389081835745, "time": 0.7812315940856933, "epoch": 9, "memory": 25074, "step": 46764} +{"lr": 0.0026214889037780207, "data_time": 0.0023120641708374023, "loss": 0.04207463189959526, "time": 0.8007621765136719, "epoch": 9, "memory": 25074, "step": 46864} +{"lr": 0.0026214889037780207, "data_time": 0.0017193794250488282, "loss": 0.04248841069638729, "time": 0.7815124988555908, "epoch": 9, "memory": 25074, "step": 46964} +{"lr": 0.0026214889037780207, "data_time": 0.0023185253143310548, "loss": 0.04956267960369587, "time": 0.8030549049377441, "epoch": 9, "memory": 25074, "step": 47064} +{"lr": 0.0026214889037780207, "data_time": 0.0018356800079345702, "loss": 0.03543620202690363, "time": 0.8107479333877563, "epoch": 9, "memory": 25074, "step": 47164} +{"lr": 0.0026214889037780207, "data_time": 0.002096390724182129, "loss": 0.04235930442810058, "time": 0.808166241645813, "epoch": 9, "memory": 25074, "step": 47264} +{"lr": 0.0026214889037780207, "data_time": 0.0015815019607543946, "loss": 0.040592977032065394, "time": 0.86477210521698, "epoch": 9, "memory": 25074, "step": 47364} +{"lr": 0.0026214889037780207, "data_time": 0.002258706092834473, "loss": 0.058388377726078036, "time": 0.8222603797912598, "epoch": 9, "memory": 25074, "step": 47464} +{"lr": 0.0026214889037780207, "data_time": 0.0018778324127197265, "loss": 0.036899886652827264, "time": 0.7938441276550293, "epoch": 9, "memory": 25074, "step": 47564} +{"lr": 0.0026214889037780207, "data_time": 0.0020696163177490235, "loss": 0.026893947273492813, "time": 0.7813984870910644, "epoch": 9, "memory": 25074, "step": 47664} +{"lr": 0.0026214889037780207, "data_time": 0.0018400430679321289, "loss": 0.0410335611552, "time": 0.7760570526123047, "epoch": 9, "memory": 25074, "step": 47764} +{"lr": 0.0026214889037780207, "data_time": 0.00188138484954834, "loss": 0.030237929755821825, "time": 0.8536426305770874, "epoch": 9, "memory": 25074, "step": 47864} +{"lr": 0.0026214889037780207, "data_time": 0.0019226312637329102, "loss": 0.13606493994593621, "time": 0.8637728929519654, "epoch": 9, "memory": 25074, "step": 47964} +{"lr": 0.0026214889037780207, "data_time": 0.00211482048034668, "loss": 0.03358338922262192, "time": 0.7930330276489258, "epoch": 9, "memory": 25074, "step": 48064} +{"lr": 0.0026214889037780207, "data_time": 0.001715874671936035, "loss": 0.0319957572966814, "time": 0.7864399194717407, "epoch": 9, "memory": 25074, "step": 48164} +{"lr": 0.0026214889037780207, "data_time": 0.0022603750228881838, "loss": 0.025551164289936423, "time": 0.8411364555358887, "epoch": 9, "memory": 25074, "step": 48264} +{"lr": 0.0026214889037780207, "data_time": 0.0015340089797973634, "loss": 0.030427138321101665, "time": 0.7890354871749878, "epoch": 9, "memory": 25074, "step": 48364} +{"lr": 0.0026214889037780207, "data_time": 0.0020235300064086912, "loss": 0.036000782065093516, "time": 0.8084030628204346, "epoch": 9, "memory": 25074, "step": 48464} +{"lr": 0.0026214889037780207, "data_time": 0.0022765636444091798, "loss": 0.021300629805773498, "time": 0.7924983978271485, "epoch": 9, "memory": 25074, "step": 48564} +{"lr": 0.0026214889037780207, "data_time": 0.0025313138961791993, "loss": 0.04439207892864942, "time": 0.8228082180023193, "epoch": 9, "memory": 25074, "step": 48664} +{"lr": 0.0026214889037780207, "data_time": 0.002218770980834961, "loss": 0.02781707257963717, "time": 0.7708287239074707, "epoch": 9, "memory": 25074, "step": 48764} +{"lr": 0.0026214889037780207, "data_time": 0.001650547981262207, "loss": 0.02063184455037117, "time": 0.8667472839355469, "epoch": 9, "memory": 25074, "step": 48864} +{"lr": 0.0026214889037780207, "data_time": 0.0016929388046264648, "loss": 0.024441056232899428, "time": 0.8383315563201904, "epoch": 9, "memory": 25074, "step": 48964} +{"lr": 0.0026214889037780207, "data_time": 0.0014557838439941406, "loss": 0.02691945619881153, "time": 0.8165060758590699, "epoch": 9, "memory": 25074, "step": 49064} +{"lr": 0.0026214889037780207, "data_time": 0.0017290592193603515, "loss": 0.0241798035800457, "time": 0.8456665277481079, "epoch": 9, "memory": 25074, "step": 49164} +{"lr": 0.0026214889037780207, "data_time": 0.0014475584030151367, "loss": 0.03516687611117959, "time": 0.8111721277236938, "epoch": 9, "memory": 25074, "step": 49264} +{"lr": 0.0026214889037780207, "data_time": 0.0018590927124023438, "loss": 0.24516677521169186, "time": 0.8151656150817871, "epoch": 9, "memory": 25074, "step": 49364} +{"lr": 0.0026214889037780207, "data_time": 0.001399683952331543, "loss": 0.028192812763154506, "time": 0.8275673389434814, "epoch": 9, "memory": 25074, "step": 49464} +{"lr": 0.0026214889037780207, "data_time": 0.001490330696105957, "loss": 0.023393982416018845, "time": 0.7727564096450805, "epoch": 9, "memory": 25074, "step": 49564} +{"lr": 0.0026214889037780207, "data_time": 0.0016549110412597656, "loss": 0.02338564246892929, "time": 0.8695456981658936, "epoch": 9, "memory": 25074, "step": 49664} +{"lr": 0.0026214889037780207, "data_time": 0.001441788673400879, "loss": 0.015078999381512404, "time": 0.802464509010315, "epoch": 9, "memory": 25074, "step": 49764} +{"lr": 0.0026214889037780207, "data_time": 0.0016698122024536132, "loss": 0.03197599779814482, "time": 0.8282928705215454, "epoch": 9, "memory": 25074, "step": 49864} +{"lr": 0.0026214889037780207, "data_time": 0.0022622823715209963, "loss": 0.013944790884852409, "time": 0.7984851837158203, "epoch": 9, "memory": 25074, "step": 49964} +{"lr": 0.0026214889037780207, "data_time": 0.0019774436950683594, "loss": 0.03877650052309036, "time": 0.7803465366363526, "epoch": 9, "memory": 25074, "step": 50064} +{"lr": 0.0026214889037780207, "data_time": 0.0015271425247192383, "loss": 0.026444716192781926, "time": 0.7932117938995361, "epoch": 9, "memory": 25074, "step": 50164} +{"lr": 0.0026214889037780207, "data_time": 0.001999688148498535, "loss": 0.02609200058504939, "time": 0.8093926191329956, "epoch": 9, "memory": 25074, "step": 50264} +{"lr": 0.0026214889037780207, "data_time": 0.0024790525436401366, "loss": 0.022045370377600192, "time": 0.8204033613204956, "epoch": 9, "memory": 25074, "step": 50364} +{"lr": 0.0026214889037780207, "data_time": 0.0019163131713867188, "loss": 0.030200400948524476, "time": 0.7742255210876465, "epoch": 9, "memory": 25074, "step": 50464} +{"lr": 0.0026214889037780207, "data_time": 0.002627658843994141, "loss": 0.019695536652579903, "time": 0.832004976272583, "epoch": 9, "memory": 25074, "step": 50564} +{"lr": 0.0026214889037780207, "data_time": 0.0019025802612304688, "loss": 0.024905043840408325, "time": 0.8002906799316406, "epoch": 9, "memory": 25074, "step": 50664} +{"lr": 0.0026214889037780207, "data_time": 0.0015072107315063476, "loss": 0.02924757357686758, "time": 0.8218582391738891, "epoch": 9, "memory": 25074, "step": 50764} +{"lr": 0.0026214889037780207, "data_time": 0.0015533685684204102, "loss": 0.0230212087277323, "time": 0.7957111835479737, "epoch": 9, "memory": 25074, "step": 50864} +{"lr": 0.0026214889037780207, "data_time": 0.0018848896026611329, "loss": 0.022353645507246257, "time": 0.8086437463760376, "epoch": 9, "memory": 25074, "step": 50964} +{"lr": 0.0026214889037780207, "data_time": 0.0018225669860839843, "loss": 0.018633402604609727, "time": 0.7607362508773804, "epoch": 9, "memory": 25074, "step": 51064} +{"lr": 0.0026214889037780207, "data_time": 0.0019716978073120116, "loss": 0.03313821600750089, "time": 0.840787124633789, "epoch": 9, "memory": 25074, "step": 51164} +{"lr": 0.0026214889037780207, "data_time": 0.0016269683837890625, "loss": 0.014874533098191024, "time": 0.8427674531936645, "epoch": 9, "memory": 25074, "step": 51264} +{"lr": 0.0026214889037780207, "data_time": 0.002698802947998047, "loss": 0.028587420424446463, "time": 0.8011973857879638, "epoch": 9, "memory": 25074, "step": 51364} +{"lr": 0.0026214889037780207, "data_time": 0.002422475814819336, "loss": 0.02664147736504674, "time": 0.8069389343261719, "epoch": 9, "memory": 25074, "step": 51464} +{"lr": 0.0026214889037780207, "data_time": 0.0018821954727172852, "loss": 0.020317592099308966, "time": 0.8240701913833618, "epoch": 9, "memory": 25074, "step": 51564} +{"lr": 0.0026214889037780207, "data_time": 0.0019106626510620116, "loss": 0.023197530955076217, "time": 0.8193970680236816, "epoch": 9, "memory": 25074, "step": 51664} +{"lr": 0.0026214889037780207, "data_time": 0.0021815061569213866, "loss": 0.014408213505521416, "time": 0.8479761600494384, "epoch": 9, "memory": 25074, "step": 51764} +{"accuracy/top1": 99.33084869384766, "data_time": 0.23750793232637293, "time": 0.6247797713560217, "step": 9} +{"lr": 0.002317086757755261, "data_time": 0.0020456314086914062, "loss": 0.017532415268942714, "time": 0.8207318782806396, "epoch": 10, "memory": 25074, "step": 51922} +{"lr": 0.002317086757755261, "data_time": 0.001619434356689453, "loss": 0.01765131545253098, "time": 0.7812523126602173, "epoch": 10, "memory": 25074, "step": 52022} +{"lr": 0.002317086757755261, "data_time": 0.0024901628494262695, "loss": 0.011763214785605668, "time": 0.8035857677459717, "epoch": 10, "memory": 25074, "step": 52122} +{"lr": 0.002317086757755261, "data_time": 0.0017036914825439454, "loss": 0.012079137982800603, "time": 0.7796544313430787, "epoch": 10, "memory": 25074, "step": 52222} +{"lr": 0.002317086757755261, "data_time": 0.0016983509063720702, "loss": 0.013122422154992819, "time": 0.8294822454452515, "epoch": 10, "memory": 25074, "step": 52322} +{"lr": 0.002317086757755261, "data_time": 0.001611948013305664, "loss": 0.019555166084319352, "time": 0.8364537000656128, "epoch": 10, "memory": 25074, "step": 52422} +{"lr": 0.002317086757755261, "data_time": 0.0026428699493408203, "loss": 0.011063139233738185, "time": 0.8343900680541992, "epoch": 10, "memory": 25074, "step": 52522} +{"lr": 0.002317086757755261, "data_time": 0.0020388364791870117, "loss": 0.018025414459407328, "time": 0.7451565027236938, "epoch": 10, "memory": 25074, "step": 52622} +{"lr": 0.002317086757755261, "data_time": 0.0018421173095703124, "loss": 0.018842556141316892, "time": 0.7868381261825561, "epoch": 10, "memory": 25074, "step": 52722} +{"lr": 0.002317086757755261, "data_time": 0.0026170492172241213, "loss": 0.020071753207594155, "time": 0.755088472366333, "epoch": 10, "memory": 25074, "step": 52822} +{"lr": 0.002317086757755261, "data_time": 0.0021312713623046877, "loss": 0.014937166590243578, "time": 0.7882421255111695, "epoch": 10, "memory": 25074, "step": 52922} +{"lr": 0.002317086757755261, "data_time": 0.0020072460174560547, "loss": 0.015283338679000735, "time": 0.8208979368209839, "epoch": 10, "memory": 25074, "step": 53022} +{"lr": 0.002317086757755261, "data_time": 0.0017373323440551757, "loss": 0.013104923209175468, "time": 0.7905618190765381, "epoch": 10, "memory": 25074, "step": 53122} +{"lr": 0.002317086757755261, "data_time": 0.0017949342727661133, "loss": 0.013780922582373022, "time": 0.7907957077026367, "epoch": 10, "memory": 25074, "step": 53222} +{"lr": 0.002317086757755261, "data_time": 0.0022233247756958006, "loss": 0.018247692426666618, "time": 0.7822386264801026, "epoch": 10, "memory": 25074, "step": 53322} +{"lr": 0.002317086757755261, "data_time": 0.002651524543762207, "loss": 0.014068604446947575, "time": 0.8082148551940918, "epoch": 10, "memory": 25074, "step": 53422} +{"lr": 0.002317086757755261, "data_time": 0.0020285606384277343, "loss": 0.016047554556280375, "time": 0.8054810047149659, "epoch": 10, "memory": 25074, "step": 53522} +{"lr": 0.002317086757755261, "data_time": 0.0020494461059570312, "loss": 0.009224567469209433, "time": 0.8373507022857666, "epoch": 10, "memory": 25074, "step": 53622} +{"lr": 0.002317086757755261, "data_time": 0.0017382383346557617, "loss": 0.015223241131752729, "time": 0.8445773839950561, "epoch": 10, "memory": 25074, "step": 53722} +{"lr": 0.002317086757755261, "data_time": 0.0018725633621215821, "loss": 0.011556544108316302, "time": 0.8119404315948486, "epoch": 10, "memory": 25074, "step": 53822} +{"lr": 0.002317086757755261, "data_time": 0.0021619319915771483, "loss": 0.012437142711132766, "time": 0.8511190176010132, "epoch": 10, "memory": 25074, "step": 53922} +{"lr": 0.002317086757755261, "data_time": 0.0018029212951660156, "loss": 0.011740866350010037, "time": 0.8316761970520019, "epoch": 10, "memory": 25074, "step": 54022} +{"lr": 0.002317086757755261, "data_time": 0.0025862932205200197, "loss": 0.024602488055825232, "time": 0.8639458656311035, "epoch": 10, "memory": 25074, "step": 54122} +{"lr": 0.002317086757755261, "data_time": 0.0022039890289306642, "loss": 0.02148634297773242, "time": 0.7997440338134766, "epoch": 10, "memory": 25074, "step": 54222} +{"lr": 0.002317086757755261, "data_time": 0.0026303529739379883, "loss": 0.0185382345225662, "time": 0.8154025316238404, "epoch": 10, "memory": 25074, "step": 54322} +{"lr": 0.002317086757755261, "data_time": 0.00218808650970459, "loss": 0.6881393432617188, "time": 0.8355687379837036, "epoch": 10, "memory": 25074, "step": 54422} +{"lr": 0.002317086757755261, "data_time": 0.0020766973495483397, "loss": 0.6841361999511719, "time": 0.8412147521972656, "epoch": 10, "memory": 25074, "step": 54522} +{"lr": 0.002317086757755261, "data_time": 0.0022381305694580077, "loss": 0.686181640625, "time": 0.8091365098953247, "epoch": 10, "memory": 25074, "step": 54622} +{"lr": 0.002317086757755261, "data_time": 0.0019045829772949218, "loss": 0.6752777099609375, "time": 0.8718130111694335, "epoch": 10, "memory": 25074, "step": 54722} +{"lr": 0.002317086757755261, "data_time": 0.0018952131271362306, "loss": 0.5613418817520142, "time": 0.8286389827728271, "epoch": 10, "memory": 25074, "step": 54822} +{"lr": 0.002317086757755261, "data_time": 0.002397894859313965, "loss": 0.49733060896396636, "time": 0.8195470571517944, "epoch": 10, "memory": 25074, "step": 54922} +{"lr": 0.002317086757755261, "data_time": 0.0015130758285522461, "loss": 0.4726357638835907, "time": 0.8234517812728882, "epoch": 10, "memory": 25074, "step": 55022} +{"lr": 0.002317086757755261, "data_time": 0.001624751091003418, "loss": 0.422371917963028, "time": 0.7901685953140258, "epoch": 10, "memory": 25074, "step": 55122} +{"lr": 0.002317086757755261, "data_time": 0.0022609710693359377, "loss": 0.4255555093288422, "time": 0.7914380073547364, "epoch": 10, "memory": 25074, "step": 55222} +{"lr": 0.002317086757755261, "data_time": 0.001772165298461914, "loss": 0.3666388154029846, "time": 0.8414067983627319, "epoch": 10, "memory": 25074, "step": 55322} +{"lr": 0.002317086757755261, "data_time": 0.0015523910522460937, "loss": 0.3317451149225235, "time": 0.7918429613113404, "epoch": 10, "memory": 25074, "step": 55422} +{"lr": 0.002317086757755261, "data_time": 0.002152752876281738, "loss": 0.27832788676023484, "time": 0.8647343873977661, "epoch": 10, "memory": 25074, "step": 55522} +{"lr": 0.002317086757755261, "data_time": 0.002333998680114746, "loss": 0.2840075805783272, "time": 0.7786744832992554, "epoch": 10, "memory": 25074, "step": 55622} +{"lr": 0.002317086757755261, "data_time": 0.0023336410522460938, "loss": 0.2699961230158806, "time": 0.8274954557418823, "epoch": 10, "memory": 25074, "step": 55722} +{"lr": 0.002317086757755261, "data_time": 0.0024749279022216798, "loss": 0.20767785906791686, "time": 0.8666459083557129, "epoch": 10, "memory": 25074, "step": 55822} +{"lr": 0.002317086757755261, "data_time": 0.0022413015365600588, "loss": 0.19525945335626602, "time": 0.8322144031524659, "epoch": 10, "memory": 25074, "step": 55922} +{"lr": 0.002317086757755261, "data_time": 0.0020354032516479493, "loss": 0.16501691937446594, "time": 0.8302124261856079, "epoch": 10, "memory": 25074, "step": 56022} +{"lr": 0.002317086757755261, "data_time": 0.002210378646850586, "loss": 0.11037591993808746, "time": 0.7660328149795532, "epoch": 10, "memory": 25074, "step": 56122} +{"lr": 0.002317086757755261, "data_time": 0.0023299217224121093, "loss": 0.08365992046892642, "time": 0.9144795179367066, "epoch": 10, "memory": 25074, "step": 56222} +{"lr": 0.002317086757755261, "data_time": 0.002490592002868652, "loss": 0.08645744733512402, "time": 0.8665079116821289, "epoch": 10, "memory": 25074, "step": 56322} +{"lr": 0.002317086757755261, "data_time": 0.0022429943084716795, "loss": 0.0692336829379201, "time": 0.9035349369049073, "epoch": 10, "memory": 25074, "step": 56422} +{"lr": 0.002317086757755261, "data_time": 0.0014755725860595703, "loss": 0.04233815483748913, "time": 0.8001418828964233, "epoch": 10, "memory": 25074, "step": 56522} +{"lr": 0.002317086757755261, "data_time": 0.0019971132278442383, "loss": 0.04625342469662428, "time": 0.8565793037414551, "epoch": 10, "memory": 25074, "step": 56622} +{"lr": 0.002317086757755261, "data_time": 0.0017195463180541993, "loss": 0.03124313671141863, "time": 0.8225848913192749, "epoch": 10, "memory": 25074, "step": 56722} +{"lr": 0.002317086757755261, "data_time": 0.0026178836822509767, "loss": 0.04096714742481709, "time": 0.8223660469055176, "epoch": 10, "memory": 25074, "step": 56822} +{"lr": 0.002317086757755261, "data_time": 0.0019570350646972655, "loss": 0.04236628897488117, "time": 0.7936403274536132, "epoch": 10, "memory": 25074, "step": 56922} +{"lr": 0.002317086757755261, "data_time": 0.0025371313095092773, "loss": 0.025017997808754446, "time": 0.7964055061340332, "epoch": 10, "memory": 25074, "step": 57022} +{"lr": 0.002317086757755261, "data_time": 0.002306365966796875, "loss": 0.05144036617130041, "time": 0.8108157157897949, "epoch": 10, "memory": 25074, "step": 57122} +{"lr": 0.002317086757755261, "data_time": 0.002100038528442383, "loss": 0.02744342922233045, "time": 0.8977809429168702, "epoch": 10, "memory": 25074, "step": 57222} +{"lr": 0.002317086757755261, "data_time": 0.0023747682571411133, "loss": 0.03131036190316081, "time": 0.8460140705108643, "epoch": 10, "memory": 25074, "step": 57322} +{"lr": 0.002317086757755261, "data_time": 0.002607393264770508, "loss": 0.02022187914699316, "time": 0.8046281814575196, "epoch": 10, "memory": 25074, "step": 57422} +{"lr": 0.002317086757755261, "data_time": 0.0024229049682617187, "loss": 0.020474710408598183, "time": 0.8252543687820435, "epoch": 10, "memory": 25074, "step": 57522} +{"accuracy/top1": 100.0, "data_time": 0.22276956894818475, "time": 0.6108220044304343, "step": 10} +{"lr": 0.0020050000000000003, "data_time": 0.002590775489807129, "loss": 0.02065365444868803, "time": 0.7896236181259155, "epoch": 11, "memory": 25074, "step": 57680} +{"lr": 0.0020050000000000003, "data_time": 0.002123236656188965, "loss": 0.0183746303897351, "time": 0.8085314273834229, "epoch": 11, "memory": 25074, "step": 57780} +{"lr": 0.0020050000000000003, "data_time": 0.0019198417663574218, "loss": 0.02596295215189457, "time": 0.7891613721847535, "epoch": 11, "memory": 25074, "step": 57880} +{"lr": 0.0020050000000000003, "data_time": 0.0018343925476074219, "loss": 0.013839405495673418, "time": 0.8113912582397461, "epoch": 11, "memory": 25074, "step": 57980} +{"lr": 0.0020050000000000003, "data_time": 0.0016790390014648437, "loss": 0.017425126023590566, "time": 0.8286957740783691, "epoch": 11, "memory": 25074, "step": 58080} +{"lr": 0.0020050000000000003, "data_time": 0.0019089937210083007, "loss": 0.024284788640215993, "time": 0.7943507194519043, "epoch": 11, "memory": 25074, "step": 58180} +{"lr": 0.0020050000000000003, "data_time": 0.0028844833374023437, "loss": 0.020067624375224115, "time": 0.7873490810394287, "epoch": 11, "memory": 25074, "step": 58280} +{"lr": 0.0020050000000000003, "data_time": 0.0029201507568359375, "loss": 0.01442998107522726, "time": 0.8720103740692139, "epoch": 11, "memory": 25074, "step": 58380} +{"lr": 0.0020050000000000003, "data_time": 0.0024564504623413087, "loss": 0.018959759408608078, "time": 0.8664995908737183, "epoch": 11, "memory": 25074, "step": 58480} +{"lr": 0.0020050000000000003, "data_time": 0.0020102500915527345, "loss": 0.018543228320777417, "time": 0.7973664999008179, "epoch": 11, "memory": 25074, "step": 58580} +{"lr": 0.0020050000000000003, "data_time": 0.0018429994583129884, "loss": 0.05799570456147194, "time": 0.7817375898361206, "epoch": 11, "memory": 25074, "step": 58680} +{"lr": 0.0020050000000000003, "data_time": 0.0016614437103271485, "loss": 0.012952214712277055, "time": 0.8419855833053589, "epoch": 11, "memory": 25074, "step": 58780} +{"lr": 0.0020050000000000003, "data_time": 0.002156186103820801, "loss": 0.025047481711953878, "time": 0.7899405241012574, "epoch": 11, "memory": 25074, "step": 58880} +{"lr": 0.0020050000000000003, "data_time": 0.0021374702453613283, "loss": 0.019668174767866732, "time": 0.845258641242981, "epoch": 11, "memory": 25074, "step": 58980} +{"lr": 0.0020050000000000003, "data_time": 0.002274632453918457, "loss": 0.009758525481447577, "time": 0.7986643552780152, "epoch": 11, "memory": 25074, "step": 59080} +{"lr": 0.0020050000000000003, "data_time": 0.0019394159317016602, "loss": 0.011583693930879236, "time": 0.8287673711776733, "epoch": 11, "memory": 25074, "step": 59180} +{"lr": 0.0020050000000000003, "data_time": 0.0020122766494750977, "loss": 0.01529764630831778, "time": 0.8637459754943848, "epoch": 11, "memory": 25074, "step": 59280} +{"lr": 0.0020050000000000003, "data_time": 0.0021901369094848634, "loss": 0.01475684279575944, "time": 0.8098446607589722, "epoch": 11, "memory": 25074, "step": 59380} +{"lr": 0.0020050000000000003, "data_time": 0.002501654624938965, "loss": 0.015876112086698414, "time": 0.8413092613220214, "epoch": 11, "memory": 25074, "step": 59480} +{"lr": 0.0020050000000000003, "data_time": 0.00217742919921875, "loss": 0.5657811641693116, "time": 0.7998675107955933, "epoch": 11, "memory": 25074, "step": 59580} +{"lr": 0.0020050000000000003, "data_time": 0.0019238948822021484, "loss": 0.4891375541687012, "time": 0.8692360639572143, "epoch": 11, "memory": 25074, "step": 59680} +{"lr": 0.0020050000000000003, "data_time": 0.001653289794921875, "loss": 0.43288591504096985, "time": 0.8257172346115113, "epoch": 11, "memory": 25074, "step": 59780} +{"lr": 0.0020050000000000003, "data_time": 0.0016234159469604493, "loss": 0.392276793718338, "time": 0.8491599082946777, "epoch": 11, "memory": 25074, "step": 59880} +{"lr": 0.0020050000000000003, "data_time": 0.002200007438659668, "loss": 0.3798608660697937, "time": 0.8266980171203613, "epoch": 11, "memory": 25074, "step": 59980} +{"lr": 0.0020050000000000003, "data_time": 0.0021578073501586914, "loss": 0.3877251446247101, "time": 0.7977796792984009, "epoch": 11, "memory": 25074, "step": 60080} +{"lr": 0.0020050000000000003, "data_time": 0.0023580074310302736, "loss": 0.30527042746543886, "time": 0.870452618598938, "epoch": 11, "memory": 25074, "step": 60180} +{"lr": 0.0020050000000000003, "data_time": 0.0020999908447265625, "loss": 0.27039543241262437, "time": 0.7851626396179199, "epoch": 11, "memory": 25074, "step": 60280} +{"lr": 0.0020050000000000003, "data_time": 0.002391409873962402, "loss": 0.25152166187763214, "time": 0.8447658777236938, "epoch": 11, "memory": 25074, "step": 60380} +{"lr": 0.0020050000000000003, "data_time": 0.0015321493148803711, "loss": 0.13193194791674615, "time": 0.8256754159927369, "epoch": 11, "memory": 25074, "step": 60480} +{"lr": 0.0020050000000000003, "data_time": 0.0015456199645996094, "loss": 0.06172098498791456, "time": 0.7993576765060425, "epoch": 11, "memory": 25074, "step": 60580} +{"lr": 0.0020050000000000003, "data_time": 0.0015208959579467774, "loss": 0.028707681596279143, "time": 0.814827823638916, "epoch": 11, "memory": 25074, "step": 60680} +{"lr": 0.0020050000000000003, "data_time": 0.0014363527297973633, "loss": 0.030554246716201307, "time": 0.73760826587677, "epoch": 11, "memory": 25074, "step": 60780} +{"lr": 0.0020050000000000003, "data_time": 0.0014954328536987305, "loss": 0.020832169940695165, "time": 0.8498701333999634, "epoch": 11, "memory": 25074, "step": 60880} +{"lr": 0.0020050000000000003, "data_time": 0.002184128761291504, "loss": 0.029661014582961796, "time": 0.8704250812530517, "epoch": 11, "memory": 25074, "step": 60980} +{"lr": 0.0020050000000000003, "data_time": 0.0022843599319458006, "loss": 0.01465114252641797, "time": 0.8545256853103638, "epoch": 11, "memory": 25074, "step": 61080} +{"lr": 0.0020050000000000003, "data_time": 0.0019321918487548828, "loss": 0.023434247262775897, "time": 0.8331084251403809, "epoch": 11, "memory": 25074, "step": 61180} +{"lr": 0.0020050000000000003, "data_time": 0.0018091201782226562, "loss": 0.021718163136392833, "time": 0.8547056674957275, "epoch": 11, "memory": 25074, "step": 61280} +{"lr": 0.0020050000000000003, "data_time": 0.0014871597290039063, "loss": 0.019336781091988088, "time": 0.7947772979736328, "epoch": 11, "memory": 25074, "step": 61380} +{"lr": 0.0020050000000000003, "data_time": 0.001659226417541504, "loss": 0.01791371051222086, "time": 0.864168906211853, "epoch": 11, "memory": 25074, "step": 61480} +{"lr": 0.0020050000000000003, "data_time": 0.0023895025253295897, "loss": 0.018238531658425927, "time": 0.8782007694244385, "epoch": 11, "memory": 25074, "step": 61580} +{"lr": 0.0020050000000000003, "data_time": 0.0014923095703125, "loss": 0.015828208066523075, "time": 0.8442202806472778, "epoch": 11, "memory": 25074, "step": 61680} +{"lr": 0.0020050000000000003, "data_time": 0.0020627498626708983, "loss": 0.03517089206725359, "time": 0.8642852306365967, "epoch": 11, "memory": 25074, "step": 61780} +{"lr": 0.0020050000000000003, "data_time": 0.0022553205490112305, "loss": 0.017217851197347044, "time": 0.9275934219360351, "epoch": 11, "memory": 25074, "step": 61880} +{"lr": 0.0020050000000000003, "data_time": 0.002085423469543457, "loss": 0.013848322676494718, "time": 0.8090350389480591, "epoch": 11, "memory": 25074, "step": 61980} +{"lr": 0.0020050000000000003, "data_time": 0.001747465133666992, "loss": 0.012819849653169513, "time": 0.8843717098236084, "epoch": 11, "memory": 25074, "step": 62080} +{"lr": 0.0020050000000000003, "data_time": 0.0017876863479614259, "loss": 0.01733230291865766, "time": 0.8388864278793335, "epoch": 11, "memory": 25074, "step": 62180} +{"lr": 0.0020050000000000003, "data_time": 0.0014911413192749024, "loss": 0.01523120547644794, "time": 0.8495623588562011, "epoch": 11, "memory": 25074, "step": 62280} +{"lr": 0.0020050000000000003, "data_time": 0.0026740789413452148, "loss": 0.013444093475118279, "time": 0.9048556566238404, "epoch": 11, "memory": 25074, "step": 62380} +{"lr": 0.0020050000000000003, "data_time": 0.0018796205520629882, "loss": 0.012744836742058396, "time": 0.8099363803863525, "epoch": 11, "memory": 25074, "step": 62480} +{"lr": 0.0020050000000000003, "data_time": 0.0014407873153686524, "loss": 0.01894613322801888, "time": 0.876837968826294, "epoch": 11, "memory": 25074, "step": 62580} +{"lr": 0.0020050000000000003, "data_time": 0.002218937873840332, "loss": 0.009766071895137429, "time": 0.8186978340148926, "epoch": 11, "memory": 25074, "step": 62680} +{"lr": 0.0020050000000000003, "data_time": 0.002142596244812012, "loss": 0.016696813562884927, "time": 0.8417745590209961, "epoch": 11, "memory": 25074, "step": 62780} +{"lr": 0.0020050000000000003, "data_time": 0.001732778549194336, "loss": 0.010334298480302095, "time": 0.8277750492095948, "epoch": 11, "memory": 25074, "step": 62880} +{"lr": 0.0020050000000000003, "data_time": 0.001989459991455078, "loss": 0.00968954348936677, "time": 0.8843014717102051, "epoch": 11, "memory": 25074, "step": 62980} +{"lr": 0.0020050000000000003, "data_time": 0.0019947052001953124, "loss": 0.010796508565545083, "time": 0.798948335647583, "epoch": 11, "memory": 25074, "step": 63080} +{"lr": 0.0020050000000000003, "data_time": 0.0016904354095458984, "loss": 0.013044635811820627, "time": 0.8644941329956055, "epoch": 11, "memory": 25074, "step": 63180} +{"lr": 0.0020050000000000003, "data_time": 0.0014651298522949218, "loss": 0.010796272382140159, "time": 0.7989150047302246, "epoch": 11, "memory": 25074, "step": 63280} +{"accuracy/top1": 100.0, "data_time": 0.23673754579880657, "time": 0.6232916888068704, "step": 11} +{"lr": 0.0016929132422447402, "data_time": 0.0016835689544677734, "loss": 0.012621474638581275, "time": 0.8847478151321411, "epoch": 12, "memory": 25074, "step": 63438} +{"lr": 0.0016929132422447402, "data_time": 0.002158713340759277, "loss": 0.012400662759318948, "time": 0.8630192279815674, "epoch": 12, "memory": 25074, "step": 63538} +{"lr": 0.0016929132422447402, "data_time": 0.0022430658340454102, "loss": 0.006929849926382303, "time": 0.8485779285430908, "epoch": 12, "memory": 25074, "step": 63638} +{"lr": 0.0016929132422447402, "data_time": 0.0014281272888183594, "loss": 0.009629213670268654, "time": 0.8658886432647706, "epoch": 12, "memory": 25074, "step": 63738} +{"lr": 0.0016929132422447402, "data_time": 0.0018830299377441406, "loss": 0.005235561076551676, "time": 0.8525608539581299, "epoch": 12, "memory": 25074, "step": 63838} +{"lr": 0.0016929132422447402, "data_time": 0.001458120346069336, "loss": 0.5257198572158813, "time": 0.8559411525726318, "epoch": 12, "memory": 25074, "step": 63938} +{"lr": 0.0016929132422447402, "data_time": 0.0024147987365722655, "loss": 0.01740764332935214, "time": 0.8217525959014893, "epoch": 12, "memory": 25074, "step": 64038} +{"lr": 0.0016929132422447402, "data_time": 0.0018601417541503906, "loss": 0.018427724577486516, "time": 0.8378716230392456, "epoch": 12, "memory": 25074, "step": 64138} +{"lr": 0.0016929132422447402, "data_time": 0.0016304492950439454, "loss": 0.015688306698575617, "time": 0.8500861167907715, "epoch": 12, "memory": 25074, "step": 64238} +{"lr": 0.0016929132422447402, "data_time": 0.0016458988189697265, "loss": 0.01742241098545492, "time": 0.8576305150985718, "epoch": 12, "memory": 25074, "step": 64338} +{"lr": 0.0016929132422447402, "data_time": 0.0018129348754882812, "loss": 0.00865716771222651, "time": 0.8440456390380859, "epoch": 12, "memory": 25074, "step": 64438} +{"lr": 0.0016929132422447402, "data_time": 0.0016907453536987305, "loss": 0.004777575191110372, "time": 0.8773534297943115, "epoch": 12, "memory": 25074, "step": 64538} +{"lr": 0.0016929132422447402, "data_time": 0.0016074895858764649, "loss": 0.012243752693757414, "time": 0.8194139003753662, "epoch": 12, "memory": 25074, "step": 64638} +{"lr": 0.0016929132422447402, "data_time": 0.0013826370239257812, "loss": 0.006700974376872182, "time": 0.8386459112167358, "epoch": 12, "memory": 25074, "step": 64738} +{"lr": 0.0016929132422447402, "data_time": 0.0017354011535644532, "loss": 0.012806080281734467, "time": 0.8877508878707886, "epoch": 12, "memory": 25074, "step": 64838} +{"lr": 0.0016929132422447402, "data_time": 0.0015664339065551759, "loss": 0.004690782725811004, "time": 0.7721720457077026, "epoch": 12, "memory": 25074, "step": 64938} +{"lr": 0.0016929132422447402, "data_time": 0.0017460107803344727, "loss": 0.009485487034544349, "time": 0.8704520225524902, "epoch": 12, "memory": 25074, "step": 65038} +{"lr": 0.0016929132422447402, "data_time": 0.0016896247863769532, "loss": 0.01166194025427103, "time": 0.8121005058288574, "epoch": 12, "memory": 25074, "step": 65138} +{"lr": 0.0016929132422447402, "data_time": 0.002256011962890625, "loss": 0.011341179488226771, "time": 0.8417533159255981, "epoch": 12, "memory": 25074, "step": 65238} +{"lr": 0.0016929132422447402, "data_time": 0.00157623291015625, "loss": 0.007958573894575239, "time": 0.8386371374130249, "epoch": 12, "memory": 25074, "step": 65338} +{"lr": 0.0016929132422447402, "data_time": 0.0014529705047607421, "loss": 0.013911510119214654, "time": 0.9013768196105957, "epoch": 12, "memory": 25074, "step": 65438} +{"lr": 0.0016929132422447402, "data_time": 0.0015138626098632813, "loss": 0.00801685922779143, "time": 0.8339585065841675, "epoch": 12, "memory": 25074, "step": 65538} +{"lr": 0.0016929132422447402, "data_time": 0.0014876127243041992, "loss": 0.010362574364989996, "time": 0.84767165184021, "epoch": 12, "memory": 25074, "step": 65638} +{"lr": 0.0016929132422447402, "data_time": 0.001378011703491211, "loss": 0.01023637312464416, "time": 0.8630808830261231, "epoch": 12, "memory": 25074, "step": 65738} +{"lr": 0.0016929132422447402, "data_time": 0.0019390106201171875, "loss": 0.004722489602863788, "time": 0.8443884134292603, "epoch": 12, "memory": 25074, "step": 65838} +{"lr": 0.0016929132422447402, "data_time": 0.0015073299407958984, "loss": 0.007929569622501732, "time": 0.8416267395019531, "epoch": 12, "memory": 25074, "step": 65938} +{"lr": 0.0016929132422447402, "data_time": 0.001519918441772461, "loss": 0.004670946905389428, "time": 0.8766562461853027, "epoch": 12, "memory": 25074, "step": 66038} +{"lr": 0.0016929132422447402, "data_time": 0.0016965150833129882, "loss": 0.6126410746946931, "time": 0.7927167415618896, "epoch": 12, "memory": 25074, "step": 66138} +{"lr": 0.0016929132422447402, "data_time": 0.0015101432800292969, "loss": 0.6205726623535156, "time": 0.7908918619155884, "epoch": 12, "memory": 25074, "step": 66238} +{"lr": 0.0016929132422447402, "data_time": 0.001480698585510254, "loss": 0.5530441761016845, "time": 0.8379903316497803, "epoch": 12, "memory": 25074, "step": 66338} +{"lr": 0.0016929132422447402, "data_time": 0.0015945911407470703, "loss": 0.44581436514854433, "time": 0.8860613107681274, "epoch": 12, "memory": 25074, "step": 66438} +{"lr": 0.0016929132422447402, "data_time": 0.0018938064575195312, "loss": 0.3750150799751282, "time": 2.01279821395874, "epoch": 12, "memory": 25074, "step": 66538} +{"lr": 0.0016929132422447402, "data_time": 0.0027787208557128905, "loss": 0.3052941858768463, "time": 0.8875884056091309, "epoch": 12, "memory": 25074, "step": 66638} +{"lr": 0.0016929132422447402, "data_time": 0.001857924461364746, "loss": 0.29005714058876036, "time": 0.839936900138855, "epoch": 12, "memory": 25074, "step": 66738} +{"lr": 0.0016929132422447402, "data_time": 0.0019493579864501953, "loss": 0.3041937619447708, "time": 0.8447667121887207, "epoch": 12, "memory": 25074, "step": 66838} +{"lr": 0.0016929132422447402, "data_time": 0.002667999267578125, "loss": 0.24104223847389222, "time": 0.8763674020767211, "epoch": 12, "memory": 25074, "step": 66938} +{"lr": 0.0016929132422447402, "data_time": 0.0022324323654174805, "loss": 0.26029926389455793, "time": 0.8523849487304688, "epoch": 12, "memory": 25074, "step": 67038} +{"lr": 0.0016929132422447402, "data_time": 0.0017183065414428712, "loss": 0.21992358714342117, "time": 0.8215179204940796, "epoch": 12, "memory": 25074, "step": 67138} +{"lr": 0.0016929132422447402, "data_time": 0.0016553401947021484, "loss": 0.19771873950958252, "time": 0.8601011753082275, "epoch": 12, "memory": 25074, "step": 67238} +{"lr": 0.0016929132422447402, "data_time": 0.0017284154891967773, "loss": 0.20580636113882064, "time": 0.8334522485733032, "epoch": 12, "memory": 25074, "step": 67338} +{"lr": 0.0016929132422447402, "data_time": 0.0014740228652954102, "loss": 0.17389717549085618, "time": 0.8400430917739868, "epoch": 12, "memory": 25074, "step": 67438} +{"lr": 0.0016929132422447402, "data_time": 0.0016913890838623046, "loss": 0.1664424017071724, "time": 0.8414290904998779, "epoch": 12, "memory": 25074, "step": 67538} +{"lr": 0.0016929132422447402, "data_time": 0.001644730567932129, "loss": 0.13310553431510924, "time": 0.8928435325622559, "epoch": 12, "memory": 25074, "step": 67638} +{"lr": 0.0016929132422447402, "data_time": 0.0023133277893066405, "loss": 0.07253217957913875, "time": 0.8251712322235107, "epoch": 12, "memory": 25074, "step": 67738} +{"lr": 0.0016929132422447402, "data_time": 0.001665210723876953, "loss": 0.05623281970620155, "time": 0.8944040536880493, "epoch": 12, "memory": 25074, "step": 67838} +{"lr": 0.0016929132422447402, "data_time": 0.002199721336364746, "loss": 0.048439373075962064, "time": 0.9268851518630982, "epoch": 12, "memory": 25074, "step": 67938} +{"lr": 0.0016929132422447402, "data_time": 0.0018225431442260743, "loss": 0.026531974226236342, "time": 0.8406747579574585, "epoch": 12, "memory": 25074, "step": 68038} +{"lr": 0.0016929132422447402, "data_time": 0.0020012617111206054, "loss": 0.0358680609613657, "time": 0.8708724737167358, "epoch": 12, "memory": 25074, "step": 68138} +{"lr": 0.0016929132422447402, "data_time": 0.0021620750427246093, "loss": 0.03205973412841558, "time": 0.8374172449111938, "epoch": 12, "memory": 25074, "step": 68238} +{"lr": 0.0016929132422447402, "data_time": 0.0027259349822998046, "loss": 0.03224460240453482, "time": 0.8043625354766846, "epoch": 12, "memory": 25074, "step": 68338} +{"lr": 0.0016929132422447402, "data_time": 0.0020184040069580076, "loss": 0.026007355935871602, "time": 0.8549070835113526, "epoch": 12, "memory": 25074, "step": 68438} +{"lr": 0.0016929132422447402, "data_time": 0.001947927474975586, "loss": 0.019273914955556394, "time": 0.8272217035293579, "epoch": 12, "memory": 25074, "step": 68538} +{"lr": 0.0016929132422447402, "data_time": 0.002057814598083496, "loss": 0.018786594970151782, "time": 0.9094939231872559, "epoch": 12, "memory": 25074, "step": 68638} +{"lr": 0.0016929132422447402, "data_time": 0.0022021770477294923, "loss": 0.027599555673077703, "time": 0.7877516508102417, "epoch": 12, "memory": 25074, "step": 68738} +{"lr": 0.0016929132422447402, "data_time": 0.002342677116394043, "loss": 0.019824191089719535, "time": 0.8667730093002319, "epoch": 12, "memory": 25074, "step": 68838} +{"lr": 0.0016929132422447402, "data_time": 0.0029083013534545897, "loss": 0.02208682969212532, "time": 0.8528003215789794, "epoch": 12, "memory": 25074, "step": 68938} +{"lr": 0.0016929132422447402, "data_time": 0.0021807193756103516, "loss": 0.021340499771758913, "time": 0.8658803462982178, "epoch": 12, "memory": 25074, "step": 69038} +{"accuracy/top1": 100.0, "data_time": 0.239563689512365, "time": 0.6475922079647288, "step": 12} +{"lr": 0.0013885110962219803, "data_time": 0.0021898031234741213, "loss": 0.015078532509505749, "time": 0.8153757095336914, "epoch": 13, "memory": 25074, "step": 69196} +{"lr": 0.0013885110962219803, "data_time": 0.0022696256637573242, "loss": 0.012806871673092246, "time": 0.8093753099441529, "epoch": 13, "memory": 25074, "step": 69296} +{"lr": 0.0013885110962219803, "data_time": 0.002752876281738281, "loss": 0.01458560605533421, "time": 0.8431886196136474, "epoch": 13, "memory": 25074, "step": 69396} +{"lr": 0.0013885110962219803, "data_time": 0.0020204782485961914, "loss": 0.018982653878629206, "time": 0.8766860961914062, "epoch": 13, "memory": 25074, "step": 69496} +{"lr": 0.0013885110962219803, "data_time": 0.0024851322174072265, "loss": 0.027763380389660596, "time": 0.8396484375, "epoch": 13, "memory": 25074, "step": 69596} +{"lr": 0.0013885110962219803, "data_time": 0.001862788200378418, "loss": 0.01050863009877503, "time": 0.8230806827545166, "epoch": 13, "memory": 25074, "step": 69696} +{"lr": 0.0013885110962219803, "data_time": 0.0017082929611206056, "loss": 0.00995070575736463, "time": 0.837144422531128, "epoch": 13, "memory": 25074, "step": 69796} +{"lr": 0.0013885110962219803, "data_time": 0.0021940231323242187, "loss": 0.008433992881327868, "time": 0.8563901424407959, "epoch": 13, "memory": 25074, "step": 69896} +{"lr": 0.0013885110962219803, "data_time": 0.002205348014831543, "loss": 0.012554180575534701, "time": 0.9230027198791504, "epoch": 13, "memory": 25074, "step": 69996} +{"lr": 0.0013885110962219803, "data_time": 0.002058887481689453, "loss": 0.006387828104197979, "time": 0.8305997371673584, "epoch": 13, "memory": 25074, "step": 70096} +{"lr": 0.0013885110962219803, "data_time": 0.0020680665969848634, "loss": 0.010489308135583996, "time": 0.8507598161697387, "epoch": 13, "memory": 25074, "step": 70196} +{"lr": 0.0013885110962219803, "data_time": 0.0019878149032592773, "loss": 0.017410028772428632, "time": 0.8330593824386596, "epoch": 13, "memory": 25074, "step": 70296} +{"lr": 0.0013885110962219803, "data_time": 0.001918792724609375, "loss": 0.01754007306881249, "time": 0.8435761451721191, "epoch": 13, "memory": 25074, "step": 70396} +{"lr": 0.0013885110962219803, "data_time": 0.0019415855407714845, "loss": 0.011260713404044508, "time": 0.8395402669906616, "epoch": 13, "memory": 25074, "step": 70496} +{"lr": 0.0013885110962219803, "data_time": 0.002218461036682129, "loss": 0.023512054467573762, "time": 0.8369058609008789, "epoch": 13, "memory": 25074, "step": 70596} +{"lr": 0.0013885110962219803, "data_time": 0.002631545066833496, "loss": 0.006385938404127956, "time": 0.8792079210281372, "epoch": 13, "memory": 25074, "step": 70696} +{"lr": 0.0013885110962219803, "data_time": 0.002178788185119629, "loss": 0.01204793006181717, "time": 0.8195004463195801, "epoch": 13, "memory": 25074, "step": 70796} +{"lr": 0.0013885110962219803, "data_time": 0.0019639968872070313, "loss": 0.008308718353509903, "time": 0.8363892316818238, "epoch": 13, "memory": 25074, "step": 70896} +{"lr": 0.0013885110962219803, "data_time": 0.0026275634765625, "loss": 0.012846501916646958, "time": 0.8775410890579224, "epoch": 13, "memory": 25074, "step": 70996} +{"lr": 0.0013885110962219803, "data_time": 0.0021003246307373046, "loss": 0.01695290096104145, "time": 0.8429610252380371, "epoch": 13, "memory": 25074, "step": 71096} +{"lr": 0.0013885110962219803, "data_time": 0.0019771337509155275, "loss": 0.008343653427436948, "time": 0.8354400873184205, "epoch": 13, "memory": 25074, "step": 71196} +{"lr": 0.0013885110962219803, "data_time": 0.0025455713272094726, "loss": 0.013309257617220283, "time": 0.8646746397018432, "epoch": 13, "memory": 25074, "step": 71296} +{"lr": 0.0013885110962219803, "data_time": 0.002090191841125488, "loss": 0.004584446968510747, "time": 0.8228408336639405, "epoch": 13, "memory": 25074, "step": 71396} +{"lr": 0.0013885110962219803, "data_time": 0.0014577150344848634, "loss": 0.005361990164965391, "time": 0.8153012752532959, "epoch": 13, "memory": 25074, "step": 71496} +{"lr": 0.0013885110962219803, "data_time": 0.001925206184387207, "loss": 0.008768650144338608, "time": 0.8246352672576904, "epoch": 13, "memory": 25074, "step": 71596} +{"lr": 0.0013885110962219803, "data_time": 0.002426481246948242, "loss": 0.6324028015136719, "time": 0.7858124494552612, "epoch": 13, "memory": 25074, "step": 71696} +{"lr": 0.0013885110962219803, "data_time": 0.002113151550292969, "loss": 0.40916086733341217, "time": 0.8469485759735107, "epoch": 13, "memory": 25074, "step": 71796} +{"lr": 0.0013885110962219803, "data_time": 0.0018510818481445312, "loss": 0.30697071850299834, "time": 0.8175181150436401, "epoch": 13, "memory": 25074, "step": 71896} +{"lr": 0.0013885110962219803, "data_time": 0.0017873048782348633, "loss": 0.2704250529408455, "time": 0.8586045265197754, "epoch": 13, "memory": 25074, "step": 71996} +{"lr": 0.0013885110962219803, "data_time": 0.0023411035537719725, "loss": 0.22080647200345993, "time": 0.835635781288147, "epoch": 13, "memory": 25074, "step": 72096} +{"lr": 0.0013885110962219803, "data_time": 0.002394556999206543, "loss": 0.16636738702654838, "time": 0.8540791273117065, "epoch": 13, "memory": 25074, "step": 72196} +{"lr": 0.0013885110962219803, "data_time": 0.001729440689086914, "loss": 0.06882301755249501, "time": 0.8588104486465454, "epoch": 13, "memory": 25074, "step": 72296} +{"lr": 0.0013885110962219803, "data_time": 0.00183258056640625, "loss": 0.04123944826424122, "time": 0.8716161727905274, "epoch": 13, "memory": 25074, "step": 72396} +{"lr": 0.0013885110962219803, "data_time": 0.002329111099243164, "loss": 0.03534002546221018, "time": 0.8019241333007813, "epoch": 13, "memory": 25074, "step": 72496} +{"lr": 0.0013885110962219803, "data_time": 0.002138566970825195, "loss": 0.025562022486701606, "time": 0.8151591062545777, "epoch": 13, "memory": 25074, "step": 72596} +{"lr": 0.0013885110962219803, "data_time": 0.0025525808334350584, "loss": 0.01956577431410551, "time": 0.8970122814178467, "epoch": 13, "memory": 25074, "step": 72696} +{"lr": 0.0013885110962219803, "data_time": 0.0024777650833129883, "loss": 0.019316222751513124, "time": 0.8234839677810669, "epoch": 13, "memory": 25074, "step": 72796} +{"lr": 0.0013885110962219803, "data_time": 0.002869844436645508, "loss": 0.014829408191144467, "time": 0.869322943687439, "epoch": 13, "memory": 25074, "step": 72896} +{"lr": 0.0013885110962219803, "data_time": 0.002419590950012207, "loss": 0.04732629181817174, "time": 0.9129178047180175, "epoch": 13, "memory": 25074, "step": 72996} +{"lr": 0.0013885110962219803, "data_time": 0.0023486137390136717, "loss": 0.2543853297829628, "time": 0.9481786251068115, "epoch": 13, "memory": 25074, "step": 73096} +{"lr": 0.0013885110962219803, "data_time": 0.002463889122009277, "loss": 0.05858225151896477, "time": 0.8872850179672241, "epoch": 13, "memory": 25074, "step": 73196} +{"lr": 0.0013885110962219803, "data_time": 0.002854776382446289, "loss": 0.036336281523108485, "time": 0.8857977151870727, "epoch": 13, "memory": 25074, "step": 73296} +{"lr": 0.0013885110962219803, "data_time": 0.0019756317138671874, "loss": 0.018332358822226523, "time": 0.8360615730285644, "epoch": 13, "memory": 25074, "step": 73396} +{"lr": 0.0013885110962219803, "data_time": 0.0018157482147216797, "loss": 0.022497400641441345, "time": 0.8416365623474121, "epoch": 13, "memory": 25074, "step": 73496} +{"lr": 0.0013885110962219803, "data_time": 0.0021916866302490235, "loss": 0.01998066296800971, "time": 0.9434623479843139, "epoch": 13, "memory": 25074, "step": 73596} +{"lr": 0.0013885110962219803, "data_time": 0.002246522903442383, "loss": 0.006339082960039377, "time": 0.8584026336669922, "epoch": 13, "memory": 25074, "step": 73696} +{"lr": 0.0013885110962219803, "data_time": 0.001978421211242676, "loss": 0.01584367179311812, "time": 0.8500343799591065, "epoch": 13, "memory": 25074, "step": 73796} +{"lr": 0.0013885110962219803, "data_time": 0.0025147914886474608, "loss": 0.010897502861917019, "time": 0.8818669557571411, "epoch": 13, "memory": 25074, "step": 73896} +{"lr": 0.0013885110962219803, "data_time": 0.0018570899963378906, "loss": 0.0097136911470443, "time": 0.8574341058731079, "epoch": 13, "memory": 25074, "step": 73996} +{"lr": 0.0013885110962219803, "data_time": 0.002330303192138672, "loss": 0.021634470019489527, "time": 0.8520456790924072, "epoch": 13, "memory": 25074, "step": 74096} +{"lr": 0.0013885110962219803, "data_time": 0.0025841236114501954, "loss": 0.012225399538874627, "time": 0.919404149055481, "epoch": 13, "memory": 25074, "step": 74196} +{"lr": 0.0013885110962219803, "data_time": 0.0017344951629638672, "loss": 0.008988901786506177, "time": 0.8667396545410156, "epoch": 13, "memory": 25074, "step": 74296} +{"lr": 0.0013885110962219803, "data_time": 0.00243377685546875, "loss": 0.009445293573662639, "time": 0.8206976175308227, "epoch": 13, "memory": 25074, "step": 74396} +{"lr": 0.0013885110962219803, "data_time": 0.002050924301147461, "loss": 0.009059466188773513, "time": 0.8605088472366333, "epoch": 13, "memory": 25074, "step": 74496} +{"lr": 0.0013885110962219803, "data_time": 0.001897287368774414, "loss": 0.006352568743750453, "time": 0.8658196210861206, "epoch": 13, "memory": 25074, "step": 74596} +{"lr": 0.0013885110962219803, "data_time": 0.0020585060119628906, "loss": 0.017209505615755915, "time": 0.8472270488739013, "epoch": 13, "memory": 25074, "step": 74696} +{"lr": 0.0013885110962219803, "data_time": 0.0023885488510131834, "loss": 0.011076639639213681, "time": 0.8367435693740845, "epoch": 13, "memory": 25074, "step": 74796} +{"accuracy/top1": 99.9368667602539, "data_time": 0.2643093221327838, "time": 0.6519988705130184, "step": 13} +{"lr": 0.0010992889530196043, "data_time": 0.0021567344665527344, "loss": 0.006026058923453093, "time": 0.8828518390655518, "epoch": 14, "memory": 25074, "step": 74954} +{"lr": 0.0010992889530196043, "data_time": 0.0018913745880126953, "loss": 0.0025050125550478697, "time": 0.9073744297027588, "epoch": 14, "memory": 25074, "step": 75054} +{"lr": 0.0010992889530196043, "data_time": 0.0026072025299072265, "loss": 0.004734926158562303, "time": 0.8835417270660401, "epoch": 14, "memory": 25074, "step": 75154} +{"lr": 0.0010992889530196043, "data_time": 0.0018737316131591797, "loss": 0.012393732834607363, "time": 0.8698746204376221, "epoch": 14, "memory": 25074, "step": 75254} +{"lr": 0.0010992889530196043, "data_time": 0.002016615867614746, "loss": 0.008986308984458447, "time": 0.8357090950012207, "epoch": 14, "memory": 25074, "step": 75354} +{"lr": 0.0010992889530196043, "data_time": 0.0021437406539916992, "loss": 0.007734660571441054, "time": 0.8574966430664063, "epoch": 14, "memory": 25074, "step": 75454} +{"lr": 0.0010992889530196043, "data_time": 0.0016841888427734375, "loss": 0.007169878343120217, "time": 0.8786022186279296, "epoch": 14, "memory": 25074, "step": 75554} +{"lr": 0.0010992889530196043, "data_time": 0.002342367172241211, "loss": 0.009355614241212607, "time": 0.9101663589477539, "epoch": 14, "memory": 25074, "step": 75654} +{"lr": 0.0010992889530196043, "data_time": 0.0021462440490722656, "loss": 0.006325017893686891, "time": 0.9483930587768554, "epoch": 14, "memory": 25074, "step": 75754} +{"lr": 0.0010992889530196043, "data_time": 0.002126932144165039, "loss": 0.0025268110912293196, "time": 0.8317933320999146, "epoch": 14, "memory": 25074, "step": 75854} +{"lr": 0.0010992889530196043, "data_time": 0.002440953254699707, "loss": 0.010621646838262676, "time": 0.8653169870376587, "epoch": 14, "memory": 25074, "step": 75954} +{"lr": 0.0010992889530196043, "data_time": 0.001569676399230957, "loss": 0.004551400756463408, "time": 0.825012469291687, "epoch": 14, "memory": 25074, "step": 76054} +{"lr": 0.0010992889530196043, "data_time": 0.002653026580810547, "loss": 0.0058206333313137295, "time": 0.8512198448181152, "epoch": 14, "memory": 25074, "step": 76154} +{"lr": 0.0010992889530196043, "data_time": 0.0023299455642700195, "loss": 0.005658266041427851, "time": 0.9002356052398681, "epoch": 14, "memory": 25074, "step": 76254} +{"lr": 0.0010992889530196043, "data_time": 0.0019682168960571287, "loss": 0.007896796939894557, "time": 0.8558631896972656, "epoch": 14, "memory": 25074, "step": 76354} +{"lr": 0.0010992889530196043, "data_time": 0.0021454572677612306, "loss": 0.003358024824410677, "time": 0.8499433279037476, "epoch": 14, "memory": 25074, "step": 76454} +{"lr": 0.0010992889530196043, "data_time": 0.0027106046676635743, "loss": 0.010712708998471498, "time": 0.8800162553787232, "epoch": 14, "memory": 25074, "step": 76554} +{"lr": 0.0010992889530196043, "data_time": 0.002442073822021484, "loss": 0.00526658478192985, "time": 0.8586807966232299, "epoch": 14, "memory": 25074, "step": 76654} +{"lr": 0.0010992889530196043, "data_time": 0.0024852752685546875, "loss": 0.007747081574052572, "time": 0.9226675033569336, "epoch": 14, "memory": 25074, "step": 76754} +{"lr": 0.0010992889530196043, "data_time": 0.00218052864074707, "loss": 0.011363293835893273, "time": 0.852879548072815, "epoch": 14, "memory": 25074, "step": 76854} +{"lr": 0.0010992889530196043, "data_time": 0.0023080587387084963, "loss": 0.004773387638852, "time": 0.8198220014572144, "epoch": 14, "memory": 25074, "step": 76954} +{"lr": 0.0010992889530196043, "data_time": 0.001535344123840332, "loss": 0.00914477938786149, "time": 0.7742030382156372, "epoch": 14, "memory": 25074, "step": 77054} +{"lr": 0.0010992889530196043, "data_time": 0.0015209197998046875, "loss": 0.004816683242097497, "time": 0.802119255065918, "epoch": 14, "memory": 25074, "step": 77154} +{"lr": 0.0010992889530196043, "data_time": 0.002005314826965332, "loss": 0.004340975638478994, "time": 0.7940043926239013, "epoch": 14, "memory": 25074, "step": 77254} +{"lr": 0.0010992889530196043, "data_time": 0.0018709182739257812, "loss": 0.006500071287155152, "time": 0.8532385349273681, "epoch": 14, "memory": 25074, "step": 77354} +{"lr": 0.0010992889530196043, "data_time": 0.001375126838684082, "loss": 0.009208159986883402, "time": 0.7792670488357544, "epoch": 14, "memory": 25074, "step": 77454} +{"lr": 0.0010992889530196043, "data_time": 0.001407027244567871, "loss": 0.004191431915387511, "time": 0.8605144023895264, "epoch": 14, "memory": 25074, "step": 77554} +{"lr": 0.0010992889530196043, "data_time": 0.0013914823532104493, "loss": 0.0033865841571241615, "time": 0.8461498737335205, "epoch": 14, "memory": 25074, "step": 77654} +{"lr": 0.0010992889530196043, "data_time": 0.0024385452270507812, "loss": 0.012164186779409647, "time": 0.8398956775665283, "epoch": 14, "memory": 25074, "step": 77754} +{"lr": 0.0010992889530196043, "data_time": 0.0017299413681030273, "loss": 0.003520938800647855, "time": 0.81116943359375, "epoch": 14, "memory": 25074, "step": 77854} +{"lr": 0.0010992889530196043, "data_time": 0.0018192291259765624, "loss": 0.0010594731196761132, "time": 0.8249178886413574, "epoch": 14, "memory": 25074, "step": 77954} +{"lr": 0.0010992889530196043, "data_time": 0.002476000785827637, "loss": 0.010088300658389926, "time": 0.8364016056060791, "epoch": 14, "memory": 25074, "step": 78054} +{"lr": 0.0010992889530196043, "data_time": 0.0025847673416137694, "loss": 0.007270760508254171, "time": 0.8084653377532959, "epoch": 14, "memory": 25074, "step": 78154} +{"lr": 0.0010992889530196043, "data_time": 0.0023340225219726563, "loss": 0.009715610602870583, "time": 0.8354634046554565, "epoch": 14, "memory": 25074, "step": 78254} +{"lr": 0.0010992889530196043, "data_time": 0.0023162364959716797, "loss": 0.006866844790056348, "time": 0.8418134927749634, "epoch": 14, "memory": 25074, "step": 78354} +{"lr": 0.0010992889530196043, "data_time": 0.0018935441970825196, "loss": 0.005691180005669594, "time": 0.7827688932418824, "epoch": 14, "memory": 25074, "step": 78454} +{"lr": 0.0010992889530196043, "data_time": 0.0023929357528686525, "loss": 0.008494042512029409, "time": 0.8424341917037964, "epoch": 14, "memory": 25074, "step": 78554} +{"lr": 0.0010992889530196043, "data_time": 0.002049756050109863, "loss": 0.0077940210234373804, "time": 0.8067271709442139, "epoch": 14, "memory": 25074, "step": 78654} +{"lr": 0.0010992889530196043, "data_time": 0.0021519899368286134, "loss": 0.0046489877626299855, "time": 0.8273500919342041, "epoch": 14, "memory": 25074, "step": 78754} +{"lr": 0.0010992889530196043, "data_time": 0.0017299175262451172, "loss": 0.010990736912935973, "time": 0.7971392154693604, "epoch": 14, "memory": 25074, "step": 78854} +{"lr": 0.0010992889530196043, "data_time": 0.002199697494506836, "loss": 0.010737274214625359, "time": 0.8197074174880982, "epoch": 14, "memory": 25074, "step": 78954} +{"lr": 0.0010992889530196043, "data_time": 0.0025701999664306642, "loss": 0.0026431609876453876, "time": 0.8274394035339355, "epoch": 14, "memory": 25074, "step": 79054} +{"lr": 0.0010992889530196043, "data_time": 0.0018967866897583007, "loss": 0.006859848322346806, "time": 0.8829192399978638, "epoch": 14, "memory": 25074, "step": 79154} +{"lr": 0.0010992889530196043, "data_time": 0.002288532257080078, "loss": 0.0029487856198102238, "time": 0.8272616863250732, "epoch": 14, "memory": 25074, "step": 79254} +{"lr": 0.0010992889530196043, "data_time": 0.0020235300064086912, "loss": 0.0031577238813042642, "time": 0.8196623802185059, "epoch": 14, "memory": 25074, "step": 79354} +{"lr": 0.0010992889530196043, "data_time": 0.0019021034240722656, "loss": 0.0017039770260453223, "time": 0.821526026725769, "epoch": 14, "memory": 25074, "step": 79454} +{"lr": 0.0010992889530196043, "data_time": 0.002126264572143555, "loss": 0.00837366795167327, "time": 0.8593217372894287, "epoch": 14, "memory": 25074, "step": 79554} +{"lr": 0.0010992889530196043, "data_time": 0.002129483222961426, "loss": 0.00560099733993411, "time": 0.8847751379013061, "epoch": 14, "memory": 25074, "step": 79654} +{"lr": 0.0010992889530196043, "data_time": 0.0018660545349121094, "loss": 0.01007423377595842, "time": 0.867729663848877, "epoch": 14, "memory": 25074, "step": 79754} +{"lr": 0.0010992889530196043, "data_time": 0.0018967866897583007, "loss": 0.01046025725081563, "time": 0.804381251335144, "epoch": 14, "memory": 25074, "step": 79854} +{"lr": 0.0010992889530196043, "data_time": 0.0017420530319213867, "loss": 0.004879824677482247, "time": 0.8106240034103394, "epoch": 14, "memory": 25074, "step": 79954} +{"lr": 0.0010992889530196043, "data_time": 0.001981806755065918, "loss": 0.0026554435025900602, "time": 0.7800873756408692, "epoch": 14, "memory": 25074, "step": 80054} +{"lr": 0.0010992889530196043, "data_time": 0.0015960693359375, "loss": 0.0038610049057751896, "time": 0.8743934154510498, "epoch": 14, "memory": 25074, "step": 80154} +{"lr": 0.0010992889530196043, "data_time": 0.0014801263809204102, "loss": 0.006989907007664442, "time": 0.8250518798828125, "epoch": 14, "memory": 25074, "step": 80254} +{"lr": 0.0010992889530196043, "data_time": 0.001961159706115723, "loss": 0.00710472296923399, "time": 0.8247511863708497, "epoch": 14, "memory": 25074, "step": 80354} +{"lr": 0.0010992889530196043, "data_time": 0.002167797088623047, "loss": 0.0075253006536513565, "time": 0.8233062982559204, "epoch": 14, "memory": 25074, "step": 80454} +{"lr": 0.0010992889530196043, "data_time": 0.0016500711441040038, "loss": 0.005123985139653087, "time": 0.7905114412307739, "epoch": 14, "memory": 25074, "step": 80554} +{"accuracy/top1": 99.98737335205078, "data_time": 0.24486652542563045, "time": 0.638165039174697, "step": 14} +{"lr": 0.0008323684216765164, "data_time": 0.002087116241455078, "loss": 0.008190978318452835, "time": 0.8144507646560669, "epoch": 15, "memory": 25074, "step": 80712} +{"lr": 0.0008323684216765164, "data_time": 0.0017007112503051758, "loss": 0.0008685978595167398, "time": 0.7929686069488525, "epoch": 15, "memory": 25074, "step": 80812} +{"lr": 0.0008323684216765164, "data_time": 0.0017495870590209961, "loss": 0.007414591545239091, "time": 0.8182244300842285, "epoch": 15, "memory": 25074, "step": 80912} +{"lr": 0.0008323684216765164, "data_time": 0.001892399787902832, "loss": 0.0029814619570970534, "time": 0.7836994409561158, "epoch": 15, "memory": 25074, "step": 81012} +{"lr": 0.0008323684216765164, "data_time": 0.0016854047775268556, "loss": 0.006268685963004827, "time": 0.8062455654144287, "epoch": 15, "memory": 25074, "step": 81112} +{"lr": 0.0008323684216765164, "data_time": 0.001596975326538086, "loss": 0.004488473618403077, "time": 0.7703092098236084, "epoch": 15, "memory": 25074, "step": 81212} +{"lr": 0.0008323684216765164, "data_time": 0.002002906799316406, "loss": 0.006217769905924797, "time": 0.8155091524124145, "epoch": 15, "memory": 25074, "step": 81312} +{"lr": 0.0008323684216765164, "data_time": 0.002004671096801758, "loss": 0.007581920549273491, "time": 0.8436862230300903, "epoch": 15, "memory": 25074, "step": 81412} +{"lr": 0.0008323684216765164, "data_time": 0.0016895532608032227, "loss": 0.022166385781019926, "time": 0.7929153680801392, "epoch": 15, "memory": 25074, "step": 81512} +{"lr": 0.0008323684216765164, "data_time": 0.0015766620635986328, "loss": 0.003970115818083287, "time": 0.7769886016845703, "epoch": 15, "memory": 25074, "step": 81612} +{"lr": 0.0008323684216765164, "data_time": 0.0025185585021972657, "loss": 0.004429465113207698, "time": 0.8138776063919068, "epoch": 15, "memory": 25074, "step": 81712} +{"lr": 0.0008323684216765164, "data_time": 0.001848292350769043, "loss": 0.0031737711280584334, "time": 0.852851128578186, "epoch": 15, "memory": 25074, "step": 81812} +{"lr": 0.0008323684216765164, "data_time": 0.001421499252319336, "loss": 0.0038157122209668158, "time": 0.8502236366271972, "epoch": 15, "memory": 25074, "step": 81912} +{"lr": 0.0008323684216765164, "data_time": 0.001393437385559082, "loss": 0.004559839516878128, "time": 0.7973427772521973, "epoch": 15, "memory": 25074, "step": 82012} +{"lr": 0.0008323684216765164, "data_time": 0.0021329641342163084, "loss": 0.0030257120728492738, "time": 0.763046407699585, "epoch": 15, "memory": 25074, "step": 82112} +{"lr": 0.0008323684216765164, "data_time": 0.001698756217956543, "loss": 0.0033781186677515505, "time": 0.8239712238311767, "epoch": 15, "memory": 25074, "step": 82212} +{"lr": 0.0008323684216765164, "data_time": 0.0015157938003540039, "loss": 0.0033322657458484174, "time": 0.8517455577850341, "epoch": 15, "memory": 25074, "step": 82312} +{"lr": 0.0008323684216765164, "data_time": 0.001606583595275879, "loss": 0.006936464365571738, "time": 0.8343741178512574, "epoch": 15, "memory": 25074, "step": 82412} +{"lr": 0.0008323684216765164, "data_time": 0.0025830745697021486, "loss": 0.0046401942614465955, "time": 0.8258126497268676, "epoch": 15, "memory": 25074, "step": 82512} +{"lr": 0.0008323684216765164, "data_time": 0.0016058921813964845, "loss": 0.002779401373118162, "time": 0.7366345882415771, "epoch": 15, "memory": 25074, "step": 82612} +{"lr": 0.0008323684216765164, "data_time": 0.001593470573425293, "loss": 0.004100637137889862, "time": 0.901034951210022, "epoch": 15, "memory": 25074, "step": 82712} +{"lr": 0.0008323684216765164, "data_time": 0.0021658897399902343, "loss": 0.0032065569888800383, "time": 0.8671699523925781, "epoch": 15, "memory": 25074, "step": 82812} +{"lr": 0.0008323684216765164, "data_time": 0.0015111684799194336, "loss": 0.0036294298246502877, "time": 0.8019484519958496, "epoch": 15, "memory": 25074, "step": 82912} +{"lr": 0.0008323684216765164, "data_time": 0.0016389846801757812, "loss": 0.0033681821078062057, "time": 0.8077052116394043, "epoch": 15, "memory": 25074, "step": 83012} +{"lr": 0.0008323684216765164, "data_time": 0.0021733522415161135, "loss": 0.002631246857345104, "time": 0.7924538135528565, "epoch": 15, "memory": 25074, "step": 83112} +{"lr": 0.0008323684216765164, "data_time": 0.0016282081604003906, "loss": 0.0025754041504114865, "time": 0.8362874507904052, "epoch": 15, "memory": 25074, "step": 83212} +{"lr": 0.0008323684216765164, "data_time": 0.0026250839233398437, "loss": 0.0013220942579209804, "time": 0.8433773279190063, "epoch": 15, "memory": 25074, "step": 83312} +{"lr": 0.0008323684216765164, "data_time": 0.0014664649963378907, "loss": 0.003958883741870522, "time": 0.8503497123718262, "epoch": 15, "memory": 25074, "step": 83412} +{"lr": 0.0008323684216765164, "data_time": 0.0019595861434936524, "loss": 0.0016685916110873223, "time": 0.835192608833313, "epoch": 15, "memory": 25074, "step": 83512} +{"lr": 0.0008323684216765164, "data_time": 0.0015546560287475586, "loss": 0.003671620553359389, "time": 0.8708398580551148, "epoch": 15, "memory": 25074, "step": 83612} +{"lr": 0.0008323684216765164, "data_time": 0.0019462108612060547, "loss": 0.0023081721272319557, "time": 0.8684589862823486, "epoch": 15, "memory": 25074, "step": 83712} +{"lr": 0.0008323684216765164, "data_time": 0.0015102148056030274, "loss": 0.007967055262997746, "time": 0.9060788869857788, "epoch": 15, "memory": 25074, "step": 83812} +{"lr": 0.0008323684216765164, "data_time": 0.001999187469482422, "loss": 0.002258659154176712, "time": 0.7957812070846557, "epoch": 15, "memory": 25074, "step": 83912} +{"lr": 0.0008323684216765164, "data_time": 0.002222752571105957, "loss": 0.0024430623278021814, "time": 0.8611499547958374, "epoch": 15, "memory": 25074, "step": 84012} +{"lr": 0.0008323684216765164, "data_time": 0.0021514415740966795, "loss": 0.002764499979093671, "time": 0.8284383773803711, "epoch": 15, "memory": 25074, "step": 84112} +{"lr": 0.0008323684216765164, "data_time": 0.002145242691040039, "loss": 0.002132095117121935, "time": 0.8111137866973877, "epoch": 15, "memory": 25074, "step": 84212} +{"lr": 0.0008323684216765164, "data_time": 0.002154636383056641, "loss": 0.0018910645972937345, "time": 0.8173701763153076, "epoch": 15, "memory": 25074, "step": 84312} +{"lr": 0.0008323684216765164, "data_time": 0.0021457672119140625, "loss": 0.0027873084880411627, "time": 0.8290945053100586, "epoch": 15, "memory": 25074, "step": 84412} +{"lr": 0.0008323684216765164, "data_time": 0.002359342575073242, "loss": 0.001872691372409463, "time": 0.8151299476623535, "epoch": 15, "memory": 25074, "step": 84512} +{"lr": 0.0008323684216765164, "data_time": 0.002001833915710449, "loss": 0.004440096439793706, "time": 0.8477199077606201, "epoch": 15, "memory": 25074, "step": 84612} +{"lr": 0.0008323684216765164, "data_time": 0.002122354507446289, "loss": 0.0070302918087691065, "time": 0.8791698455810547, "epoch": 15, "memory": 25074, "step": 84712} +{"lr": 0.0008323684216765164, "data_time": 0.002477121353149414, "loss": 0.0017334240023046733, "time": 0.818484902381897, "epoch": 15, "memory": 25074, "step": 84812} +{"lr": 0.0008323684216765164, "data_time": 0.0024985551834106447, "loss": 0.0031128203961998226, "time": 0.8045818567276001, "epoch": 15, "memory": 25074, "step": 84912} +{"lr": 0.0008323684216765164, "data_time": 0.001815032958984375, "loss": 0.0020079654175788166, "time": 0.8695090293884278, "epoch": 15, "memory": 25074, "step": 85012} +{"lr": 0.0008323684216765164, "data_time": 0.0015797138214111329, "loss": 0.0015763045754283666, "time": 0.7929683685302734, "epoch": 15, "memory": 25074, "step": 85112} +{"lr": 0.0008323684216765164, "data_time": 0.0021280527114868166, "loss": 0.0030572716146707534, "time": 0.9226945877075196, "epoch": 15, "memory": 25074, "step": 85212} +{"lr": 0.0008323684216765164, "data_time": 0.0016113519668579102, "loss": 0.0042770013213157656, "time": 0.8555640459060669, "epoch": 15, "memory": 25074, "step": 85312} +{"lr": 0.0008323684216765164, "data_time": 0.0016779899597167969, "loss": 0.0056419032625854015, "time": 0.8675498962402344, "epoch": 15, "memory": 25074, "step": 85412} +{"lr": 0.0008323684216765164, "data_time": 0.0016929149627685548, "loss": 0.004891542112454772, "time": 0.8150385856628418, "epoch": 15, "memory": 25074, "step": 85512} +{"lr": 0.0008323684216765164, "data_time": 0.0023407697677612304, "loss": 0.0024810972157865764, "time": 0.9147989749908447, "epoch": 15, "memory": 25074, "step": 85612} +{"lr": 0.0008323684216765164, "data_time": 0.0019356966018676757, "loss": 0.0024863758590072393, "time": 0.829637885093689, "epoch": 15, "memory": 25074, "step": 85712} +{"lr": 0.0008323684216765164, "data_time": 0.0029778480529785156, "loss": 0.005940651195123792, "time": 0.8046359777450561, "epoch": 15, "memory": 25074, "step": 85812} +{"lr": 0.0008323684216765164, "data_time": 0.0020565509796142576, "loss": 0.00481735453940928, "time": 0.8162915229797363, "epoch": 15, "memory": 25074, "step": 85912} +{"lr": 0.0008323684216765164, "data_time": 0.002616596221923828, "loss": 0.004171461844816804, "time": 0.8333085536956787, "epoch": 15, "memory": 25074, "step": 86012} +{"lr": 0.0008323684216765164, "data_time": 0.0020272493362426757, "loss": 0.0023523079231381415, "time": 0.816011929512024, "epoch": 15, "memory": 25074, "step": 86112} +{"lr": 0.0008323684216765164, "data_time": 0.0019292354583740235, "loss": 0.0032898181583732367, "time": 0.8411876678466796, "epoch": 15, "memory": 25074, "step": 86212} +{"lr": 0.0008323684216765164, "data_time": 0.0017971992492675781, "loss": 0.0039269427303224805, "time": 0.8080334424972534, "epoch": 15, "memory": 25074, "step": 86312} +{"accuracy/top1": 100.0, "data_time": 0.23414185467888327, "time": 0.6187553966746611, "step": 15} +{"lr": 0.0005943219715328379, "data_time": 0.001903700828552246, "loss": 0.0025037748273462057, "time": 0.8635496377944947, "epoch": 16, "memory": 25074, "step": 86470} +{"lr": 0.0005943219715328379, "data_time": 0.0023732900619506834, "loss": 0.0033647256903350355, "time": 0.8420201539993286, "epoch": 16, "memory": 25074, "step": 86570} +{"lr": 0.0005943219715328379, "data_time": 0.002093648910522461, "loss": 0.002997089829295874, "time": 0.7624979257583618, "epoch": 16, "memory": 25074, "step": 86670} +{"lr": 0.0005943219715328379, "data_time": 0.0020543336868286133, "loss": 0.003493775613605976, "time": 0.860261869430542, "epoch": 16, "memory": 25074, "step": 86770} +{"lr": 0.0005943219715328379, "data_time": 0.002039337158203125, "loss": 0.001480706874281168, "time": 0.7583719968795777, "epoch": 16, "memory": 25074, "step": 86870} +{"lr": 0.0005943219715328379, "data_time": 0.0018027544021606446, "loss": 0.0032122720498591662, "time": 0.8659549236297608, "epoch": 16, "memory": 25074, "step": 86970} +{"lr": 0.0005943219715328379, "data_time": 0.0017798185348510743, "loss": 0.0005405058618634939, "time": 0.7693642616271973, "epoch": 16, "memory": 25074, "step": 87070} +{"lr": 0.0005943219715328379, "data_time": 0.0020461797714233397, "loss": 0.0006689358502626419, "time": 0.8643156051635742, "epoch": 16, "memory": 25074, "step": 87170} +{"lr": 0.0005943219715328379, "data_time": 0.002682352066040039, "loss": 0.002496235119178891, "time": 0.8343918085098266, "epoch": 16, "memory": 25074, "step": 87270} +{"lr": 0.0005943219715328379, "data_time": 0.0016872644424438476, "loss": 0.0020704213995486496, "time": 0.8328126668930054, "epoch": 16, "memory": 25074, "step": 87370} +{"lr": 0.0005943219715328379, "data_time": 0.0017588138580322266, "loss": 0.0033198816701769827, "time": 0.8297866582870483, "epoch": 16, "memory": 25074, "step": 87470} +{"lr": 0.0005943219715328379, "data_time": 0.0015356302261352538, "loss": 0.0030173714738339187, "time": 0.8674288988113403, "epoch": 16, "memory": 25074, "step": 87570} +{"lr": 0.0005943219715328379, "data_time": 0.002687549591064453, "loss": 0.0004096377640962601, "time": 0.8186800479888916, "epoch": 16, "memory": 25074, "step": 87670} +{"lr": 0.0005943219715328379, "data_time": 0.0023172855377197265, "loss": 0.002651927527040243, "time": 0.9097333669662475, "epoch": 16, "memory": 25074, "step": 87770} +{"lr": 0.0005943219715328379, "data_time": 0.001935577392578125, "loss": 0.002328659035265446, "time": 0.8453935384750366, "epoch": 16, "memory": 25074, "step": 87870} +{"lr": 0.0005943219715328379, "data_time": 0.0019171476364135743, "loss": 0.003917717421427369, "time": 0.8380070924758911, "epoch": 16, "memory": 25074, "step": 87970} +{"lr": 0.0005943219715328379, "data_time": 0.0025748252868652345, "loss": 0.0021569257602095603, "time": 0.826769208908081, "epoch": 16, "memory": 25074, "step": 88070} +{"lr": 0.0005943219715328379, "data_time": 0.00251309871673584, "loss": 0.001567025063559413, "time": 0.8712774991989136, "epoch": 16, "memory": 25074, "step": 88170} +{"lr": 0.0005943219715328379, "data_time": 0.0015096664428710938, "loss": 0.003016546554863453, "time": 0.8645804643630981, "epoch": 16, "memory": 25074, "step": 88270} +{"lr": 0.0005943219715328379, "data_time": 0.0015955448150634765, "loss": 0.0008537331130355596, "time": 0.8699898719787598, "epoch": 16, "memory": 25074, "step": 88370} +{"lr": 0.0005943219715328379, "data_time": 0.0020115852355957033, "loss": 0.002749787736684084, "time": 0.8488941669464112, "epoch": 16, "memory": 25074, "step": 88470} +{"lr": 0.0005943219715328379, "data_time": 0.0024833202362060545, "loss": 0.001474676700308919, "time": 0.8312921524047852, "epoch": 16, "memory": 25074, "step": 88570} +{"lr": 0.0005943219715328379, "data_time": 0.0019132137298583985, "loss": 0.0009348067920655012, "time": 0.8532140970230102, "epoch": 16, "memory": 25074, "step": 88670} +{"lr": 0.0005943219715328379, "data_time": 0.002227473258972168, "loss": 0.0007522489875555038, "time": 0.858740258216858, "epoch": 16, "memory": 25074, "step": 88770} +{"lr": 0.0005943219715328379, "data_time": 0.0015662431716918946, "loss": 0.0037273469381034373, "time": 0.8913288831710815, "epoch": 16, "memory": 25074, "step": 88870} +{"lr": 0.0005943219715328379, "data_time": 0.0037634849548339845, "loss": 0.0010302946902811527, "time": 0.9533833742141724, "epoch": 16, "memory": 25074, "step": 88970} +{"lr": 0.0005943219715328379, "data_time": 0.0020781040191650392, "loss": 0.0015971431974321603, "time": 0.9602782726287842, "epoch": 16, "memory": 25074, "step": 89070} +{"lr": 0.0005943219715328379, "data_time": 0.002176856994628906, "loss": 0.0005267304833978415, "time": 0.924289870262146, "epoch": 16, "memory": 25074, "step": 89170} +{"lr": 0.0005943219715328379, "data_time": 0.002349090576171875, "loss": 0.010791574558243155, "time": 0.9090284585952759, "epoch": 16, "memory": 25074, "step": 89270} +{"lr": 0.0005943219715328379, "data_time": 0.0015442848205566406, "loss": 0.001905987225472927, "time": 0.892949104309082, "epoch": 16, "memory": 25074, "step": 89370} +{"lr": 0.0005943219715328379, "data_time": 0.0015676259994506837, "loss": 0.0007513926830142736, "time": 0.9010032892227173, "epoch": 16, "memory": 25074, "step": 89470} +{"lr": 0.0005943219715328379, "data_time": 0.0015018224716186524, "loss": 0.0017784679774194955, "time": 0.8695396423339844, "epoch": 16, "memory": 25074, "step": 89570} +{"lr": 0.0005943219715328379, "data_time": 0.0015874624252319336, "loss": 0.0006417270749807357, "time": 0.877429747581482, "epoch": 16, "memory": 25074, "step": 89670} +{"lr": 0.0005943219715328379, "data_time": 0.0016166210174560548, "loss": 0.00294205192476511, "time": 0.8348247289657593, "epoch": 16, "memory": 25074, "step": 89770} +{"lr": 0.0005943219715328379, "data_time": 0.0022717952728271485, "loss": 0.003139247512444854, "time": 0.808187747001648, "epoch": 16, "memory": 25074, "step": 89870} +{"lr": 0.0005943219715328379, "data_time": 0.0016261100769042968, "loss": 0.0013865420129150151, "time": 0.7832516670227051, "epoch": 16, "memory": 25074, "step": 89970} +{"lr": 0.0005943219715328379, "data_time": 0.0015677690505981445, "loss": 0.0018713842146098614, "time": 0.8313507795333862, "epoch": 16, "memory": 25074, "step": 90070} +{"lr": 0.0005943219715328379, "data_time": 0.0019213438034057617, "loss": 0.0007648433558642865, "time": 0.8319806337356568, "epoch": 16, "memory": 25074, "step": 90170} +{"lr": 0.0005943219715328379, "data_time": 0.0014399290084838867, "loss": 0.003510078974068165, "time": 0.8241634845733643, "epoch": 16, "memory": 25074, "step": 90270} +{"lr": 0.0005943219715328379, "data_time": 0.002419281005859375, "loss": 0.0022074762731790543, "time": 0.8208592176437378, "epoch": 16, "memory": 25074, "step": 90370} +{"lr": 0.0005943219715328379, "data_time": 0.0026560306549072267, "loss": 0.0024664254393428563, "time": 0.7743782520294189, "epoch": 16, "memory": 25074, "step": 90470} +{"lr": 0.0005943219715328379, "data_time": 0.0016248703002929687, "loss": 0.004019773378968239, "time": 0.7739948749542236, "epoch": 16, "memory": 25074, "step": 90570} +{"lr": 0.0005943219715328379, "data_time": 0.002538037300109863, "loss": 0.0027326020412147046, "time": 0.816907525062561, "epoch": 16, "memory": 25074, "step": 90670} +{"lr": 0.0005943219715328379, "data_time": 0.002382326126098633, "loss": 0.006594126066192985, "time": 0.8274320363998413, "epoch": 16, "memory": 25074, "step": 90770} +{"lr": 0.0005943219715328379, "data_time": 0.0016460180282592773, "loss": 0.002716598007827997, "time": 0.8251569509506226, "epoch": 16, "memory": 25074, "step": 90870} +{"lr": 0.0005943219715328379, "data_time": 0.0017352104187011719, "loss": 0.005413290625438094, "time": 0.7875924110412598, "epoch": 16, "memory": 25074, "step": 90970} +{"lr": 0.0005943219715328379, "data_time": 0.0025635957717895508, "loss": 0.001384977111592889, "time": 0.8089169263839722, "epoch": 16, "memory": 25074, "step": 91070} +{"lr": 0.0005943219715328379, "data_time": 0.0015411853790283203, "loss": 0.0009327524341642857, "time": 0.7445062875747681, "epoch": 16, "memory": 25074, "step": 91170} +{"lr": 0.0005943219715328379, "data_time": 0.0021345853805541993, "loss": 0.0019064848311245441, "time": 0.7698233842849731, "epoch": 16, "memory": 25074, "step": 91270} +{"lr": 0.0005943219715328379, "data_time": 0.0017052650451660155, "loss": 0.0007534346543252469, "time": 0.7804375648498535, "epoch": 16, "memory": 25074, "step": 91370} +{"lr": 0.0005943219715328379, "data_time": 0.002021384239196777, "loss": 0.005605431459844113, "time": 0.7461480855941772, "epoch": 16, "memory": 25074, "step": 91470} +{"lr": 0.0005943219715328379, "data_time": 0.0025379419326782226, "loss": 0.000628881249576807, "time": 0.8316758632659912, "epoch": 16, "memory": 25074, "step": 91570} +{"lr": 0.0005943219715328379, "data_time": 0.001603841781616211, "loss": 0.0013007433619350195, "time": 0.7501929759979248, "epoch": 16, "memory": 25074, "step": 91670} +{"lr": 0.0005943219715328379, "data_time": 0.002016425132751465, "loss": 0.0012229129672050477, "time": 0.7771937847137451, "epoch": 16, "memory": 25074, "step": 91770} diff --git a/convnext_small_4xb256_4e-3lr_5m/20230531_173252/vis_data/config.py b/convnext_small_4xb256_4e-3lr_5m/20230531_173252/vis_data/config.py new file mode 100644 index 0000000000000000000000000000000000000000..675945812b0b3cd1ce1dff570c0730faa54378a1 --- /dev/null +++ b/convnext_small_4xb256_4e-3lr_5m/20230531_173252/vis_data/config.py @@ -0,0 +1,169 @@ +optim_wrapper = dict( + optimizer=dict( + type='AdamW', + lr=0.004, + weight_decay=0.05, + eps=1e-08, + betas=(0.9, 0.999), + _scope_='mmpretrain'), + paramwise_cfg=dict( + norm_decay_mult=0.0, + bias_decay_mult=0.0, + flat_decay_mult=0.0, + custom_keys=dict({ + '.absolute_pos_embed': dict(decay_mult=0.0), + '.relative_position_bias_table': dict(decay_mult=0.0) + })), + type='AmpOptimWrapper', + dtype='bfloat16', + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=True, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=20, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=4096) +model = dict( + type='ImageClassifier', + backbone=dict(type='ConvNeXt', arch='small', drop_path_rate=0.4), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=768, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + init_cfg=None), + init_cfg=dict( + type='TruncNormal', layer=['Conv2d', 'Linear'], std=0.02, bias=0.0), + train_cfg=None) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stablediffusionV1-5R2-dpmsolver-25-5m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/convnext_small_4xb256_4e-3lr_5m' diff --git a/convnext_small_4xb256_4e-3lr_5m/20230531_173252/vis_data/events.out.tfevents.1685525579.SH-IDC1-10-140-24-119.223945.0 b/convnext_small_4xb256_4e-3lr_5m/20230531_173252/vis_data/events.out.tfevents.1685525579.SH-IDC1-10-140-24-119.223945.0 new file mode 100644 index 0000000000000000000000000000000000000000..e2f3f721ac9181a65743534036c3901c83c80009 --- /dev/null +++ b/convnext_small_4xb256_4e-3lr_5m/20230531_173252/vis_data/events.out.tfevents.1685525579.SH-IDC1-10-140-24-119.223945.0 @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:d1c29b35b39e0f08e0bfb93c7efdc69d0cb98627924014ee48e92a9d7438e76c +size 1315612 diff --git a/convnext_small_4xb256_4e-3lr_5m/20230531_173252/vis_data/scalars.json b/convnext_small_4xb256_4e-3lr_5m/20230531_173252/vis_data/scalars.json new file mode 100644 index 0000000000000000000000000000000000000000..afc30d82d119faef960e30e57ceab1cc0a51bf14 --- /dev/null +++ b/convnext_small_4xb256_4e-3lr_5m/20230531_173252/vis_data/scalars.json @@ -0,0 +1,924 @@ +{"lr": 0.004, "data_time": 0.25836734771728515, "loss": 0.6135261058807373, "time": 1.1036933660507202, "epoch": 1, "memory": 25071, "step": 100} +{"lr": 0.004, "data_time": 0.001967930793762207, "loss": 0.5921411991119385, "time": 1.0567732810974122, "epoch": 1, "memory": 25071, "step": 200} +{"lr": 0.004, "data_time": 0.0014708280563354493, "loss": 0.5693366527557373, "time": 1.0624513149261474, "epoch": 1, "memory": 25071, "step": 300} +{"lr": 0.004, "data_time": 0.0017904520034790039, "loss": 0.5559554100036621, "time": 1.0132100105285644, "epoch": 1, "memory": 25071, "step": 400} +{"lr": 0.004, "data_time": 0.0018054723739624023, "loss": 0.5386565923690796, "time": 1.057649564743042, "epoch": 1, "memory": 25071, "step": 500} +{"lr": 0.004, "data_time": 0.001491236686706543, "loss": 0.520584225654602, "time": 1.0222002744674683, "epoch": 1, "memory": 25071, "step": 600} +{"lr": 0.004, "data_time": 0.0010788917541503906, "loss": 0.5122485876083374, "time": 1.0578108787536622, "epoch": 1, "memory": 25071, "step": 700} +{"lr": 0.004, "data_time": 0.0018996477127075195, "loss": 0.5337131738662719, "time": 0.9827592134475708, "epoch": 1, "memory": 25071, "step": 800} +{"lr": 0.004, "data_time": 0.0014658689498901368, "loss": 0.48136401176452637, "time": 1.0142579555511475, "epoch": 1, "memory": 25071, "step": 900} +{"lr": 0.004, "data_time": 0.0016474485397338866, "loss": 0.48867201805114746, "time": 1.0005245208740234, "epoch": 1, "memory": 25071, "step": 1000} +{"lr": 0.004, "data_time": 0.0021308183670043944, "loss": 0.4960619449615479, "time": 0.9755245447158813, "epoch": 1, "memory": 25071, "step": 1100} +{"lr": 0.004, "data_time": 0.0018622636795043944, "loss": 0.45781826972961426, "time": 0.9678232431411743, "epoch": 1, "memory": 25071, "step": 1200} +{"lr": 0.004, "data_time": 0.0016861677169799805, "loss": 0.4497794032096863, "time": 1.0431145668029784, "epoch": 1, "memory": 25071, "step": 1300} +{"lr": 0.004, "data_time": 0.0015056848526000977, "loss": 0.4368001639842987, "time": 0.9595765352249146, "epoch": 1, "memory": 25071, "step": 1400} +{"lr": 0.004, "data_time": 0.0017606496810913086, "loss": 0.5319863557815552, "time": 0.9088726043701172, "epoch": 1, "memory": 25071, "step": 1500} +{"lr": 0.004, "data_time": 0.0015338659286499023, "loss": 0.4219388782978058, "time": 0.9892490386962891, "epoch": 1, "memory": 25071, "step": 1600} +{"lr": 0.004, "data_time": 0.0016805887222290038, "loss": 0.41420637667179105, "time": 0.9990060091018677, "epoch": 1, "memory": 25071, "step": 1700} +{"lr": 0.004, "data_time": 0.001367664337158203, "loss": 0.3887391030788422, "time": 1.133042073249817, "epoch": 1, "memory": 25071, "step": 1800} +{"lr": 0.004, "data_time": 0.0014392852783203125, "loss": 0.3968598783016205, "time": 0.9008044719696044, "epoch": 1, "memory": 25071, "step": 1900} +{"lr": 0.004, "data_time": 0.0013582229614257813, "loss": 0.38166859149932864, "time": 1.0659605503082275, "epoch": 1, "memory": 25071, "step": 2000} +{"lr": 0.004, "data_time": 0.0015635013580322266, "loss": 0.40037643909454346, "time": 1.067807149887085, "epoch": 1, "memory": 25071, "step": 2100} +{"lr": 0.004, "data_time": 0.0015240907669067383, "loss": 0.3590026319026947, "time": 0.8623078346252442, "epoch": 1, "memory": 25071, "step": 2200} +{"lr": 0.004, "data_time": 0.0014862537384033204, "loss": 0.2965458869934082, "time": 0.8956292152404786, "epoch": 1, "memory": 25071, "step": 2300} +{"lr": 0.004, "data_time": 0.0015390634536743164, "loss": 0.31096053421497344, "time": 0.9354721546173096, "epoch": 1, "memory": 25071, "step": 2400} +{"lr": 0.004, "data_time": 0.0015926837921142578, "loss": 0.28933600038290025, "time": 0.8769220352172852, "epoch": 1, "memory": 25071, "step": 2500} +{"lr": 0.004, "data_time": 0.001988649368286133, "loss": 0.2849620833992958, "time": 0.8799722909927368, "epoch": 1, "memory": 25071, "step": 2600} +{"lr": 0.004, "data_time": 0.0013595342636108399, "loss": 0.2896030187606812, "time": 0.9100133657455445, "epoch": 1, "memory": 25071, "step": 2700} +{"lr": 0.004, "data_time": 0.0014533758163452148, "loss": 0.32246443927288054, "time": 0.9779120445251465, "epoch": 1, "memory": 25071, "step": 2800} +{"lr": 0.004, "data_time": 0.001288127899169922, "loss": 0.27721202224493025, "time": 0.9152437686920166, "epoch": 1, "memory": 25071, "step": 2900} +{"lr": 0.004, "data_time": 0.0019807338714599608, "loss": 0.22157122939825058, "time": 0.931296968460083, "epoch": 1, "memory": 25071, "step": 3000} +{"lr": 0.004, "data_time": 0.001652097702026367, "loss": 0.2503370255231857, "time": 0.9809789896011353, "epoch": 1, "memory": 25071, "step": 3100} +{"lr": 0.004, "data_time": 0.0013518571853637696, "loss": 0.21450963467359543, "time": 0.9946384191513061, "epoch": 1, "memory": 25071, "step": 3200} +{"lr": 0.004, "data_time": 0.002450108528137207, "loss": 0.239343623816967, "time": 0.9268593072891236, "epoch": 1, "memory": 25071, "step": 3300} +{"lr": 0.004, "data_time": 0.0017489194869995117, "loss": 0.21228042244911194, "time": 0.9522735357284546, "epoch": 1, "memory": 25071, "step": 3400} +{"lr": 0.004, "data_time": 0.0014758110046386719, "loss": 0.21550967246294023, "time": 0.9016278266906739, "epoch": 1, "memory": 25071, "step": 3500} +{"lr": 0.004, "data_time": 0.022446608543395995, "loss": 0.21206613928079604, "time": 1.0201257705688476, "epoch": 1, "memory": 25071, "step": 3600} +{"lr": 0.004, "data_time": 0.0015426874160766602, "loss": 0.1702307105064392, "time": 0.8195062875747681, "epoch": 1, "memory": 25071, "step": 3700} +{"lr": 0.004, "data_time": 0.0014066219329833985, "loss": 0.18829978108406067, "time": 1.001184868812561, "epoch": 1, "memory": 25071, "step": 3800} +{"lr": 0.004, "data_time": 0.0014562368392944335, "loss": 0.16298147067427635, "time": 0.836207914352417, "epoch": 1, "memory": 25071, "step": 3900} +{"lr": 0.004, "data_time": 0.0014770746231079102, "loss": 0.1773746654391289, "time": 0.8800336360931397, "epoch": 1, "memory": 25071, "step": 4000} +{"lr": 0.004, "data_time": 0.001606893539428711, "loss": 0.15313689559698104, "time": 0.9190765142440795, "epoch": 1, "memory": 25071, "step": 4100} +{"lr": 0.004, "data_time": 0.0019302129745483398, "loss": 0.15373844504356385, "time": 0.9319148778915405, "epoch": 1, "memory": 25071, "step": 4200} +{"lr": 0.004, "data_time": 0.0015819787979125977, "loss": 0.12970864176750183, "time": 0.89850013256073, "epoch": 1, "memory": 25071, "step": 4300} +{"lr": 0.004, "data_time": 0.0014078617095947266, "loss": 0.1656157925724983, "time": 0.8514695882797241, "epoch": 1, "memory": 25071, "step": 4400} +{"lr": 0.004, "data_time": 0.0013335227966308593, "loss": 0.1450781174004078, "time": 0.9596346616744995, "epoch": 1, "memory": 25071, "step": 4500} +{"lr": 0.004, "data_time": 0.0013980865478515625, "loss": 0.15148803144693374, "time": 0.9196593523025512, "epoch": 1, "memory": 25071, "step": 4600} +{"lr": 0.004, "data_time": 0.002222132682800293, "loss": 0.11021999567747116, "time": 0.9126356601715088, "epoch": 1, "memory": 25071, "step": 4700} +{"lr": 0.004, "data_time": 0.0018005132675170898, "loss": 0.12593574821949005, "time": 0.9484604358673095, "epoch": 1, "memory": 25071, "step": 4800} +{"lr": 0.004, "data_time": 0.0012134790420532226, "loss": 0.09986786693334579, "time": 1.0157971143722535, "epoch": 1, "memory": 25071, "step": 4900} +{"lr": 0.004, "data_time": 0.0014298677444458008, "loss": 0.11388952732086181, "time": 0.9243617296218872, "epoch": 1, "memory": 25071, "step": 5000} +{"lr": 0.004, "data_time": 0.001612401008605957, "loss": 0.12963890954852103, "time": 0.9540765523910523, "epoch": 1, "memory": 25071, "step": 5100} +{"lr": 0.004, "data_time": 0.0021638393402099608, "loss": 0.0849734254181385, "time": 0.8873657941818237, "epoch": 1, "memory": 25071, "step": 5200} +{"lr": 0.004, "data_time": 0.0015216112136840821, "loss": 0.09903554022312164, "time": 0.966349720954895, "epoch": 1, "memory": 25071, "step": 5300} +{"lr": 0.004, "data_time": 0.0020122289657592773, "loss": 0.06487535983324051, "time": 0.9787596464157104, "epoch": 1, "memory": 25071, "step": 5400} +{"lr": 0.004, "data_time": 0.0015494585037231444, "loss": 0.07417001202702522, "time": 0.8545481443405152, "epoch": 1, "memory": 25071, "step": 5500} +{"lr": 0.004, "data_time": 0.0014139890670776367, "loss": 0.06840806379914284, "time": 0.9005010604858399, "epoch": 1, "memory": 25071, "step": 5600} +{"lr": 0.004, "data_time": 0.0016618967056274414, "loss": 0.06471074298024178, "time": 1.0014020204544067, "epoch": 1, "memory": 25071, "step": 5700} +{"accuracy/top1": 71.33387756347656, "data_time": 0.29866595566272736, "time": 0.8162815719842911, "step": 1} +{"lr": 0.0039754382394873, "data_time": 0.0015654325485229491, "loss": 0.44536194801330564, "time": 0.9839568614959717, "epoch": 2, "memory": 25074, "step": 5858} +{"lr": 0.0039754382394873, "data_time": 0.002077460289001465, "loss": 0.4071248471736908, "time": 0.9970962762832641, "epoch": 2, "memory": 25074, "step": 5958} +{"lr": 0.0039754382394873, "data_time": 0.0013185977935791016, "loss": 0.24007392972707747, "time": 1.0498293399810792, "epoch": 2, "memory": 25074, "step": 6058} +{"lr": 0.0039754382394873, "data_time": 0.0015950202941894531, "loss": 0.24183081835508347, "time": 0.9420676946640014, "epoch": 2, "memory": 25074, "step": 6158} +{"lr": 0.0039754382394873, "data_time": 0.0018357276916503907, "loss": 0.09749895296990871, "time": 0.8744754314422607, "epoch": 2, "memory": 25074, "step": 6258} +{"lr": 0.0039754382394873, "data_time": 0.002217006683349609, "loss": 0.0628897450864315, "time": 0.9066163539886475, "epoch": 2, "memory": 25074, "step": 6358} +{"lr": 0.0039754382394873, "data_time": 0.001868891716003418, "loss": 0.0615274328738451, "time": 0.9184701681137085, "epoch": 2, "memory": 25074, "step": 6458} +{"lr": 0.0039754382394873, "data_time": 0.0014720201492309571, "loss": 0.07046853713691234, "time": 0.8759130716323853, "epoch": 2, "memory": 25074, "step": 6558} +{"lr": 0.0039754382394873, "data_time": 0.0015817642211914062, "loss": 0.07254206612706185, "time": 0.9210872888565064, "epoch": 2, "memory": 25074, "step": 6658} +{"lr": 0.0039754382394873, "data_time": 0.0014837980270385742, "loss": 0.08038320019841194, "time": 1.0461889982223511, "epoch": 2, "memory": 25074, "step": 6758} +{"lr": 0.0039754382394873, "data_time": 0.001940441131591797, "loss": 0.06745437271893025, "time": 0.8867187023162841, "epoch": 2, "memory": 25074, "step": 6858} +{"lr": 0.0039754382394873, "data_time": 0.0015705347061157227, "loss": 0.05236329883337021, "time": 0.8681844234466553, "epoch": 2, "memory": 25074, "step": 6958} +{"lr": 0.0039754382394873, "data_time": 0.0019382953643798829, "loss": 0.08161418288946151, "time": 0.9430710315704346, "epoch": 2, "memory": 25074, "step": 7058} +{"lr": 0.0039754382394873, "data_time": 0.0021197319030761717, "loss": 0.045391821675002574, "time": 0.938313889503479, "epoch": 2, "memory": 25074, "step": 7158} +{"lr": 0.0039754382394873, "data_time": 0.0018213033676147462, "loss": 0.05432411562651396, "time": 0.82647705078125, "epoch": 2, "memory": 25074, "step": 7258} +{"lr": 0.0039754382394873, "data_time": 0.0022699117660522463, "loss": 0.06139286048710346, "time": 0.9103602409362793, "epoch": 2, "memory": 25074, "step": 7358} +{"lr": 0.0039754382394873, "data_time": 0.0012668371200561523, "loss": 0.049718914180994035, "time": 0.860216474533081, "epoch": 2, "memory": 25074, "step": 7458} +{"lr": 0.0039754382394873, "data_time": 0.0023814916610717775, "loss": 0.050588475167751314, "time": 0.9751445293426514, "epoch": 2, "memory": 25074, "step": 7558} +{"lr": 0.0039754382394873, "data_time": 0.0017695188522338866, "loss": 0.5002466946840286, "time": 0.7511817693710328, "epoch": 2, "memory": 25074, "step": 7658} +{"lr": 0.0039754382394873, "data_time": 0.0013837099075317382, "loss": 0.4878327250480652, "time": 0.8483304500579834, "epoch": 2, "memory": 25074, "step": 7758} +{"lr": 0.0039754382394873, "data_time": 0.0019185066223144532, "loss": 0.4479573845863342, "time": 0.8353627443313598, "epoch": 2, "memory": 25074, "step": 7858} +{"lr": 0.0039754382394873, "data_time": 0.0019244909286499023, "loss": 0.21451487988233567, "time": 0.8248765468597412, "epoch": 2, "memory": 25074, "step": 7958} +{"lr": 0.0039754382394873, "data_time": 0.0020031452178955076, "loss": 0.08937919065356255, "time": 0.8127747297286987, "epoch": 2, "memory": 25074, "step": 8058} +{"lr": 0.0039754382394873, "data_time": 0.0019345521926879884, "loss": 0.048111673444509506, "time": 0.8491933107376098, "epoch": 2, "memory": 25074, "step": 8158} +{"lr": 0.0039754382394873, "data_time": 0.002181816101074219, "loss": 0.05818078741431236, "time": 0.8987390756607055, "epoch": 2, "memory": 25074, "step": 8258} +{"lr": 0.0039754382394873, "data_time": 0.0016150474548339844, "loss": 0.04987697713077068, "time": 0.7931256771087647, "epoch": 2, "memory": 25074, "step": 8358} +{"lr": 0.0039754382394873, "data_time": 0.002272367477416992, "loss": 0.06922312416136264, "time": 0.8968161344528198, "epoch": 2, "memory": 25074, "step": 8458} +{"lr": 0.0039754382394873, "data_time": 0.0022190093994140627, "loss": 0.04928675331175327, "time": 0.8449273586273194, "epoch": 2, "memory": 25074, "step": 8558} +{"lr": 0.0039754382394873, "data_time": 0.0026656150817871093, "loss": 0.03678979128599167, "time": 0.7909621715545654, "epoch": 2, "memory": 25074, "step": 8658} +{"lr": 0.0039754382394873, "data_time": 0.0016669988632202148, "loss": 0.04629333112388849, "time": 0.7934857606887817, "epoch": 2, "memory": 25074, "step": 8758} +{"lr": 0.0039754382394873, "data_time": 0.0017243385314941405, "loss": 0.04524355810135603, "time": 0.8450065851211548, "epoch": 2, "memory": 25074, "step": 8858} +{"lr": 0.0039754382394873, "data_time": 0.0021995067596435546, "loss": 0.049249463342130186, "time": 0.8159966468811035, "epoch": 2, "memory": 25074, "step": 8958} +{"lr": 0.0039754382394873, "data_time": 0.0020696163177490235, "loss": 0.041333705000579354, "time": 0.812670636177063, "epoch": 2, "memory": 25074, "step": 9058} +{"lr": 0.0039754382394873, "data_time": 0.001447772979736328, "loss": 0.040220907609909776, "time": 0.8368125915527344, "epoch": 2, "memory": 25074, "step": 9158} +{"lr": 0.0039754382394873, "data_time": 0.001545858383178711, "loss": 0.07665113694965839, "time": 0.8620558261871338, "epoch": 2, "memory": 25074, "step": 9258} +{"lr": 0.0039754382394873, "data_time": 0.001687169075012207, "loss": 0.0275498004630208, "time": 0.7853478908538818, "epoch": 2, "memory": 25074, "step": 9358} +{"lr": 0.0039754382394873, "data_time": 0.0018344879150390624, "loss": 0.028978040814399718, "time": 0.8449751615524292, "epoch": 2, "memory": 25074, "step": 9458} +{"lr": 0.0039754382394873, "data_time": 0.0015711069107055664, "loss": 0.030457904562354088, "time": 0.8253869771957397, "epoch": 2, "memory": 25074, "step": 9558} +{"lr": 0.0039754382394873, "data_time": 0.0017217159271240234, "loss": 0.0381824167445302, "time": 0.9336783409118652, "epoch": 2, "memory": 25074, "step": 9658} +{"lr": 0.0039754382394873, "data_time": 0.0015754461288452148, "loss": 0.06059279851615429, "time": 0.8491053342819214, "epoch": 2, "memory": 25074, "step": 9758} +{"lr": 0.0039754382394873, "data_time": 0.0018334627151489259, "loss": 0.039567502215504646, "time": 0.9262634754180908, "epoch": 2, "memory": 25074, "step": 9858} +{"lr": 0.0039754382394873, "data_time": 0.0015657901763916015, "loss": 0.03016723138280213, "time": 0.9408579587936401, "epoch": 2, "memory": 25074, "step": 9958} +{"lr": 0.0039754382394873, "data_time": 0.0015697717666625977, "loss": 0.03549889884889126, "time": 0.8561189889907836, "epoch": 2, "memory": 25074, "step": 10058} +{"lr": 0.0039754382394873, "data_time": 0.001992487907409668, "loss": 0.024865062534809114, "time": 0.8507202863693237, "epoch": 2, "memory": 25074, "step": 10158} +{"lr": 0.0039754382394873, "data_time": 0.001700282096862793, "loss": 0.024713566433638333, "time": 0.9130228757858276, "epoch": 2, "memory": 25074, "step": 10258} +{"lr": 0.0039754382394873, "data_time": 0.001290130615234375, "loss": 0.47437719702720643, "time": 0.8855489730834961, "epoch": 2, "memory": 25074, "step": 10358} +{"lr": 0.0039754382394873, "data_time": 0.0012105464935302734, "loss": 0.28070435076951983, "time": 0.8346230030059815, "epoch": 2, "memory": 25074, "step": 10458} +{"lr": 0.0039754382394873, "data_time": 0.0013357162475585937, "loss": 0.5716702461242675, "time": 0.9133694171905518, "epoch": 2, "memory": 25074, "step": 10558} +{"lr": 0.0039754382394873, "data_time": 0.0014510393142700196, "loss": 0.5090821623802185, "time": 0.885738730430603, "epoch": 2, "memory": 25074, "step": 10658} +{"lr": 0.0039754382394873, "data_time": 0.002016592025756836, "loss": 0.38386878967285154, "time": 0.912585973739624, "epoch": 2, "memory": 25074, "step": 10758} +{"lr": 0.0039754382394873, "data_time": 0.0020873546600341797, "loss": 0.4081609070301056, "time": 0.9337555170059204, "epoch": 2, "memory": 25074, "step": 10858} +{"lr": 0.0039754382394873, "data_time": 0.0012762784957885743, "loss": 0.33712922036647797, "time": 0.7847714900970459, "epoch": 2, "memory": 25074, "step": 10958} +{"lr": 0.0039754382394873, "data_time": 0.00125885009765625, "loss": 0.326380181312561, "time": 0.8530193090438842, "epoch": 2, "memory": 25074, "step": 11058} +{"lr": 0.0039754382394873, "data_time": 0.001339125633239746, "loss": 0.1495766393840313, "time": 0.9370234251022339, "epoch": 2, "memory": 25074, "step": 11158} +{"lr": 0.0039754382394873, "data_time": 0.0013068199157714843, "loss": 0.07547324076294899, "time": 0.8928805828094483, "epoch": 2, "memory": 25074, "step": 11258} +{"lr": 0.0039754382394873, "data_time": 0.002223658561706543, "loss": 0.7301055625081062, "time": 0.8583959817886353, "epoch": 2, "memory": 25074, "step": 11358} +{"lr": 0.0039754382394873, "data_time": 0.0014155149459838868, "loss": 0.6506040573120118, "time": 0.9257060527801514, "epoch": 2, "memory": 25074, "step": 11458} +{"accuracy/top1": 100.0, "data_time": 0.25789012628443103, "time": 0.6472268385045669, "step": 2} +{"lr": 0.0039023577500088323, "data_time": 0.0020456314086914062, "loss": 0.530897855758667, "time": 0.8861788988113404, "epoch": 3, "memory": 25074, "step": 11616} +{"lr": 0.0039023577500088323, "data_time": 0.0016747236251831055, "loss": 0.47043827176094055, "time": 0.9475600481033325, "epoch": 3, "memory": 25074, "step": 11716} +{"lr": 0.0039023577500088323, "data_time": 0.002315211296081543, "loss": 0.4299037456512451, "time": 0.830585265159607, "epoch": 3, "memory": 25074, "step": 11816} +{"lr": 0.0039023577500088323, "data_time": 0.001876497268676758, "loss": 0.3691077768802643, "time": 0.9028676509857178, "epoch": 3, "memory": 25074, "step": 11916} +{"lr": 0.0039023577500088323, "data_time": 0.0020478010177612306, "loss": 0.35603734850883484, "time": 0.8394214153289795, "epoch": 3, "memory": 25074, "step": 12016} +{"lr": 0.0039023577500088323, "data_time": 0.001947927474975586, "loss": 0.30922357439994813, "time": 0.999903678894043, "epoch": 3, "memory": 25074, "step": 12116} +{"lr": 0.0039023577500088323, "data_time": 0.00202484130859375, "loss": 0.28984666168689727, "time": 0.8322690010070801, "epoch": 3, "memory": 25074, "step": 12216} +{"lr": 0.0039023577500088323, "data_time": 0.0020009756088256838, "loss": 0.2831827044487, "time": 0.8980642795562744, "epoch": 3, "memory": 25074, "step": 12316} +{"lr": 0.0039023577500088323, "data_time": 0.002239632606506348, "loss": 0.22968373447656631, "time": 0.8585394859313965, "epoch": 3, "memory": 25074, "step": 12416} +{"lr": 0.0039023577500088323, "data_time": 0.003071236610412598, "loss": 0.22362501174211502, "time": 0.9056859016418457, "epoch": 3, "memory": 25074, "step": 12516} +{"lr": 0.0039023577500088323, "data_time": 0.0021527767181396484, "loss": 0.2103023812174797, "time": 0.9525051593780518, "epoch": 3, "memory": 25074, "step": 12616} +{"lr": 0.0039023577500088323, "data_time": 0.0021797895431518556, "loss": 0.19760508984327316, "time": 0.8640597343444825, "epoch": 3, "memory": 25074, "step": 12716} +{"lr": 0.0039023577500088323, "data_time": 0.0019652605056762694, "loss": 0.22433767467737198, "time": 0.8039330005645752, "epoch": 3, "memory": 25074, "step": 12816} +{"lr": 0.0039023577500088323, "data_time": 0.0018321752548217773, "loss": 0.12686453387141228, "time": 0.8569660425186157, "epoch": 3, "memory": 25074, "step": 12916} +{"lr": 0.0039023577500088323, "data_time": 0.0022662878036499023, "loss": 0.13652086406946182, "time": 0.9195060968399048, "epoch": 3, "memory": 25074, "step": 13016} +{"lr": 0.0039023577500088323, "data_time": 0.00256044864654541, "loss": 0.15135828405618668, "time": 0.9112863063812255, "epoch": 3, "memory": 25074, "step": 13116} +{"lr": 0.0039023577500088323, "data_time": 0.002023744583129883, "loss": 0.11753128990530967, "time": 0.8293555021286011, "epoch": 3, "memory": 25074, "step": 13216} +{"lr": 0.0039023577500088323, "data_time": 0.0019275903701782226, "loss": 0.0810539573431015, "time": 0.9455954790115356, "epoch": 3, "memory": 25074, "step": 13316} +{"lr": 0.0039023577500088323, "data_time": 0.0014641523361206056, "loss": 0.06795164383947849, "time": 0.8764861822128296, "epoch": 3, "memory": 25074, "step": 13416} +{"lr": 0.0039023577500088323, "data_time": 0.0019464492797851562, "loss": 0.0674210075289011, "time": 0.8908105373382569, "epoch": 3, "memory": 25074, "step": 13516} +{"lr": 0.0039023577500088323, "data_time": 0.0014425039291381836, "loss": 0.07151532731950283, "time": 0.8104778051376342, "epoch": 3, "memory": 25074, "step": 13616} +{"lr": 0.0039023577500088323, "data_time": 0.0012932777404785155, "loss": 0.0617364514619112, "time": 0.8420161008834839, "epoch": 3, "memory": 25074, "step": 13716} +{"lr": 0.0039023577500088323, "data_time": 0.0020517826080322264, "loss": 0.045361613482236864, "time": 0.794519567489624, "epoch": 3, "memory": 25074, "step": 13816} +{"lr": 0.0039023577500088323, "data_time": 0.0017646074295043946, "loss": 0.04847529921680689, "time": 0.8627906799316406, "epoch": 3, "memory": 25074, "step": 13916} +{"lr": 0.0039023577500088323, "data_time": 0.0022499561309814453, "loss": 0.04969416819512844, "time": 0.8765430450439453, "epoch": 3, "memory": 25074, "step": 14016} +{"lr": 0.0039023577500088323, "data_time": 0.001618647575378418, "loss": 0.07003878727555275, "time": 0.7989153385162353, "epoch": 3, "memory": 25074, "step": 14116} +{"lr": 0.0039023577500088323, "data_time": 0.0013932466506958007, "loss": 0.05054851789027452, "time": 0.8280042886734009, "epoch": 3, "memory": 25074, "step": 14216} +{"lr": 0.0039023577500088323, "data_time": 0.0016368627548217773, "loss": 0.050657306984066966, "time": 0.8679801940917968, "epoch": 3, "memory": 25074, "step": 14316} +{"lr": 0.0039023577500088323, "data_time": 0.002403879165649414, "loss": 0.05359412152320146, "time": 0.810921311378479, "epoch": 3, "memory": 25074, "step": 14416} +{"lr": 0.0039023577500088323, "data_time": 0.002183365821838379, "loss": 0.04041057452559471, "time": 0.8750509023666382, "epoch": 3, "memory": 25074, "step": 14516} +{"lr": 0.0039023577500088323, "data_time": 0.002136683464050293, "loss": 0.04170214645564556, "time": 0.8423972845077514, "epoch": 3, "memory": 25074, "step": 14616} +{"lr": 0.0039023577500088323, "data_time": 0.001434493064880371, "loss": 0.10478606373071671, "time": 0.8206373929977417, "epoch": 3, "memory": 25074, "step": 14716} +{"lr": 0.0039023577500088323, "data_time": 0.0016542911529541016, "loss": 0.5778481483459472, "time": 0.8065363883972168, "epoch": 3, "memory": 25074, "step": 14816} +{"lr": 0.0039023577500088323, "data_time": 0.0019530534744262695, "loss": 0.42171688079833985, "time": 0.838587474822998, "epoch": 3, "memory": 25074, "step": 14916} +{"lr": 0.0039023577500088323, "data_time": 0.001415562629699707, "loss": 0.3420998930931091, "time": 0.8774507284164429, "epoch": 3, "memory": 25074, "step": 15016} +{"lr": 0.0039023577500088323, "data_time": 0.0020879268646240234, "loss": 0.2082872837781906, "time": 0.8238179922103882, "epoch": 3, "memory": 25074, "step": 15116} +{"lr": 0.0039023577500088323, "data_time": 0.001863384246826172, "loss": 0.09171477109193801, "time": 0.8462811708450317, "epoch": 3, "memory": 25074, "step": 15216} +{"lr": 0.0039023577500088323, "data_time": 0.001365351676940918, "loss": 0.09157317467033863, "time": 0.8382880210876464, "epoch": 3, "memory": 25074, "step": 15316} +{"lr": 0.0039023577500088323, "data_time": 0.001530599594116211, "loss": 0.040719654969871046, "time": 0.8324625730514527, "epoch": 3, "memory": 25074, "step": 15416} +{"lr": 0.0039023577500088323, "data_time": 0.0016404151916503905, "loss": 0.04555123336613178, "time": 0.8977967262268066, "epoch": 3, "memory": 25074, "step": 15516} +{"lr": 0.0039023577500088323, "data_time": 0.0014584064483642578, "loss": 0.0518822580575943, "time": 0.9010794162750244, "epoch": 3, "memory": 25074, "step": 15616} +{"lr": 0.0039023577500088323, "data_time": 0.0015751123428344727, "loss": 0.025403516925871374, "time": 0.9069263458251953, "epoch": 3, "memory": 25074, "step": 15716} +{"lr": 0.0039023577500088323, "data_time": 0.0014905691146850585, "loss": 0.03381628515198827, "time": 0.8677921533584595, "epoch": 3, "memory": 25074, "step": 15816} +{"lr": 0.0039023577500088323, "data_time": 0.0014578580856323242, "loss": 0.02633177926763892, "time": 0.8704387664794921, "epoch": 3, "memory": 25074, "step": 15916} +{"lr": 0.0039023577500088323, "data_time": 0.0014814615249633789, "loss": 0.02399464203044772, "time": 0.8488096475601197, "epoch": 3, "memory": 25074, "step": 16016} +{"lr": 0.0039023577500088323, "data_time": 0.0014820575714111328, "loss": 0.09220785070210695, "time": 0.8705696821212768, "epoch": 3, "memory": 25074, "step": 16116} +{"lr": 0.0039023577500088323, "data_time": 0.0015499114990234375, "loss": 0.039041274785995485, "time": 0.9396117687225342, "epoch": 3, "memory": 25074, "step": 16216} +{"lr": 0.0039023577500088323, "data_time": 0.0012744665145874023, "loss": 0.033044479042291644, "time": 0.9233942985534668, "epoch": 3, "memory": 25074, "step": 16316} +{"lr": 0.0039023577500088323, "data_time": 0.001346421241760254, "loss": 0.028087299875915052, "time": 0.9162482738494873, "epoch": 3, "memory": 25074, "step": 16416} +{"lr": 0.0039023577500088323, "data_time": 0.0017097949981689452, "loss": 0.02493842667900026, "time": 0.8040107250213623, "epoch": 3, "memory": 25074, "step": 16516} +{"lr": 0.0039023577500088323, "data_time": 0.0013321161270141602, "loss": 0.030253164237365127, "time": 0.8069973707199096, "epoch": 3, "memory": 25074, "step": 16616} +{"lr": 0.0039023577500088323, "data_time": 0.0014808893203735352, "loss": 0.38303807079792024, "time": 0.9347812414169312, "epoch": 3, "memory": 25074, "step": 16716} +{"lr": 0.0039023577500088323, "data_time": 0.0013840675354003906, "loss": 0.11098072454333305, "time": 0.8739282369613648, "epoch": 3, "memory": 25074, "step": 16816} +{"lr": 0.0039023577500088323, "data_time": 0.0015062808990478516, "loss": 0.17602913081645966, "time": 0.8635448694229126, "epoch": 3, "memory": 25074, "step": 16916} +{"lr": 0.0039023577500088323, "data_time": 0.0013004064559936524, "loss": 0.03915222808718681, "time": 0.8101547002792359, "epoch": 3, "memory": 25074, "step": 17016} +{"lr": 0.0039023577500088323, "data_time": 0.001514577865600586, "loss": 0.03889876324683428, "time": 0.9931037902832032, "epoch": 3, "memory": 25074, "step": 17116} +{"lr": 0.0039023577500088323, "data_time": 0.00196683406829834, "loss": 0.025535664753988385, "time": 0.9238677978515625, "epoch": 3, "memory": 25074, "step": 17216} +{"accuracy/top1": 100.0, "data_time": 0.24308482338400447, "time": 0.6311884767868939, "step": 3} +{"lr": 0.0037825580157557948, "data_time": 0.0019972562789916993, "loss": 0.024469496868550777, "time": 0.8110650300979614, "epoch": 4, "memory": 25074, "step": 17374} +{"lr": 0.0037825580157557948, "data_time": 0.0026037216186523436, "loss": 0.017897700518369676, "time": 0.864920711517334, "epoch": 4, "memory": 25074, "step": 17474} +{"lr": 0.0037825580157557948, "data_time": 0.002062368392944336, "loss": 0.024089607782661913, "time": 0.8393401622772216, "epoch": 4, "memory": 25074, "step": 17574} +{"lr": 0.0037825580157557948, "data_time": 0.0023013591766357423, "loss": 0.016335582453757526, "time": 0.8185791730880737, "epoch": 4, "memory": 25074, "step": 17674} +{"lr": 0.0037825580157557948, "data_time": 0.0025336742401123047, "loss": 0.02260943572036922, "time": 0.8512113332748413, "epoch": 4, "memory": 25074, "step": 17774} +{"lr": 0.0037825580157557948, "data_time": 0.002070331573486328, "loss": 0.024111079657450317, "time": 0.8192685365676879, "epoch": 4, "memory": 25074, "step": 17874} +{"lr": 0.0037825580157557948, "data_time": 0.0024126529693603515, "loss": 0.01927987444214523, "time": 0.8932074785232544, "epoch": 4, "memory": 25074, "step": 17974} +{"lr": 0.0037825580157557948, "data_time": 0.0021356821060180666, "loss": 0.015596888307482005, "time": 0.8333463430404663, "epoch": 4, "memory": 25074, "step": 18074} +{"lr": 0.0037825580157557948, "data_time": 0.0021569252014160155, "loss": 0.3951702803373337, "time": 0.837399172782898, "epoch": 4, "memory": 25074, "step": 18174} +{"lr": 0.0037825580157557948, "data_time": 0.0026964664459228514, "loss": 0.3264499008655548, "time": 0.8615692138671875, "epoch": 4, "memory": 25074, "step": 18274} +{"lr": 0.0037825580157557948, "data_time": 0.0015542745590209962, "loss": 0.2364593267440796, "time": 0.8392498970031739, "epoch": 4, "memory": 25074, "step": 18374} +{"lr": 0.0037825580157557948, "data_time": 0.0021144628524780275, "loss": 0.3408315718173981, "time": 0.8177517652511597, "epoch": 4, "memory": 25074, "step": 18474} +{"lr": 0.0037825580157557948, "data_time": 0.0024758577346801758, "loss": 0.19284270852804183, "time": 0.8166410446166992, "epoch": 4, "memory": 25074, "step": 18574} +{"lr": 0.0037825580157557948, "data_time": 0.0014041900634765626, "loss": 0.06831820160150529, "time": 0.816063928604126, "epoch": 4, "memory": 25074, "step": 18674} +{"lr": 0.0037825580157557948, "data_time": 0.0015158891677856446, "loss": 0.04160851426422596, "time": 0.8284622669219971, "epoch": 4, "memory": 25074, "step": 18774} +{"lr": 0.0037825580157557948, "data_time": 0.0013856887817382812, "loss": 0.029360038321465255, "time": 0.8102623224258423, "epoch": 4, "memory": 25074, "step": 18874} +{"lr": 0.0037825580157557948, "data_time": 0.0015660524368286133, "loss": 0.04482399821281433, "time": 0.8247085332870483, "epoch": 4, "memory": 25074, "step": 18974} +{"lr": 0.0037825580157557948, "data_time": 0.0017116308212280274, "loss": 0.043982202745974065, "time": 0.8885748147964477, "epoch": 4, "memory": 25074, "step": 19074} +{"lr": 0.0037825580157557948, "data_time": 0.0018639087677001954, "loss": 0.029614633601158857, "time": 0.8481374025344849, "epoch": 4, "memory": 25074, "step": 19174} +{"lr": 0.0037825580157557948, "data_time": 0.0021905422210693358, "loss": 0.024476656597107647, "time": 0.9164541006088257, "epoch": 4, "memory": 25074, "step": 19274} +{"lr": 0.0037825580157557948, "data_time": 0.0015946388244628905, "loss": 0.020974504435434936, "time": 0.7629599809646607, "epoch": 4, "memory": 25074, "step": 19374} +{"lr": 0.0037825580157557948, "data_time": 0.0016913175582885741, "loss": 0.987266993522644, "time": 0.8512732744216919, "epoch": 4, "memory": 25074, "step": 19474} +{"lr": 0.0037825580157557948, "data_time": 0.0016631841659545898, "loss": 0.686273193359375, "time": 0.8507126569747925, "epoch": 4, "memory": 25074, "step": 19574} +{"lr": 0.0037825580157557948, "data_time": 0.0019798040390014648, "loss": 0.6228355407714844, "time": 0.8141442537307739, "epoch": 4, "memory": 25074, "step": 19674} +{"lr": 0.0037825580157557948, "data_time": 0.0022313594818115234, "loss": 0.6229887008666992, "time": 0.8711118221282959, "epoch": 4, "memory": 25074, "step": 19774} +{"lr": 0.0037825580157557948, "data_time": 0.0015506982803344727, "loss": 0.5822794914245606, "time": 0.7789529323577881, "epoch": 4, "memory": 25074, "step": 19874} +{"lr": 0.0037825580157557948, "data_time": 0.0021116256713867186, "loss": 0.5362835168838501, "time": 0.8447186708450317, "epoch": 4, "memory": 25074, "step": 19974} +{"lr": 0.0037825580157557948, "data_time": 0.00142059326171875, "loss": 0.5363489866256714, "time": 0.8090475797653198, "epoch": 4, "memory": 25074, "step": 20074} +{"lr": 0.0037825580157557948, "data_time": 0.001723003387451172, "loss": 0.5079730987548828, "time": 0.7995928049087524, "epoch": 4, "memory": 25074, "step": 20174} +{"lr": 0.0037825580157557948, "data_time": 0.0025243282318115233, "loss": 0.48581901788711546, "time": 0.8065108060836792, "epoch": 4, "memory": 25074, "step": 20274} +{"lr": 0.0037825580157557948, "data_time": 0.0017581224441528321, "loss": 0.46407254934310915, "time": 0.8865103721618652, "epoch": 4, "memory": 25074, "step": 20374} +{"lr": 0.0037825580157557948, "data_time": 0.002043271064758301, "loss": 0.43721904754638674, "time": 0.8054994583129883, "epoch": 4, "memory": 25074, "step": 20474} +{"lr": 0.0037825580157557948, "data_time": 0.0016558647155761718, "loss": 0.42262349724769593, "time": 0.8218477725982666, "epoch": 4, "memory": 25074, "step": 20574} +{"lr": 0.0037825580157557948, "data_time": 0.002169299125671387, "loss": 0.40657807886600494, "time": 0.7935243129730225, "epoch": 4, "memory": 25074, "step": 20674} +{"lr": 0.0037825580157557948, "data_time": 0.0026780128479003905, "loss": 0.41469231247901917, "time": 0.8749217987060547, "epoch": 4, "memory": 25074, "step": 20774} +{"lr": 0.0037825580157557948, "data_time": 0.0031522274017333984, "loss": 0.36320524513721464, "time": 0.8123588562011719, "epoch": 4, "memory": 25074, "step": 20874} +{"lr": 0.0037825580157557948, "data_time": 0.001869368553161621, "loss": 0.36104739606380465, "time": 0.8755192041397095, "epoch": 4, "memory": 25074, "step": 20974} +{"lr": 0.0037825580157557948, "data_time": 0.0016811370849609375, "loss": 0.32246618568897245, "time": 0.8619577407836914, "epoch": 4, "memory": 25074, "step": 21074} +{"lr": 0.0037825580157557948, "data_time": 0.001760721206665039, "loss": 0.31507766246795654, "time": 0.9055258750915527, "epoch": 4, "memory": 25074, "step": 21174} +{"lr": 0.0037825580157557948, "data_time": 0.0018574953079223632, "loss": 0.33061912953853606, "time": 0.7908989429473877, "epoch": 4, "memory": 25074, "step": 21274} +{"lr": 0.0037825580157557948, "data_time": 0.0014024019241333009, "loss": 0.29776401072740555, "time": 0.7740065813064575, "epoch": 4, "memory": 25074, "step": 21374} +{"lr": 0.0037825580157557948, "data_time": 0.0017714262008666991, "loss": 0.2693121373653412, "time": 0.8174318552017212, "epoch": 4, "memory": 25074, "step": 21474} +{"lr": 0.0037825580157557948, "data_time": 0.0016476869583129882, "loss": 0.27400467842817305, "time": 0.7716795682907105, "epoch": 4, "memory": 25074, "step": 21574} +{"lr": 0.0037825580157557948, "data_time": 0.0016970872879028321, "loss": 0.26591013967990873, "time": 0.8105545759201049, "epoch": 4, "memory": 25074, "step": 21674} +{"lr": 0.0037825580157557948, "data_time": 0.0018676996231079101, "loss": 0.24770168960094452, "time": 0.8458791971206665, "epoch": 4, "memory": 25074, "step": 21774} +{"lr": 0.0037825580157557948, "data_time": 0.002076530456542969, "loss": 0.24303943514823914, "time": 0.861296558380127, "epoch": 4, "memory": 25074, "step": 21874} +{"lr": 0.0037825580157557948, "data_time": 0.0015017986297607422, "loss": 0.21657509952783585, "time": 0.7737709999084472, "epoch": 4, "memory": 25074, "step": 21974} +{"lr": 0.0037825580157557948, "data_time": 0.0017820358276367187, "loss": 0.22080043107271194, "time": 0.8244883775711059, "epoch": 4, "memory": 25074, "step": 22074} +{"lr": 0.0037825580157557948, "data_time": 0.0020010948181152345, "loss": 0.2656919628381729, "time": 0.8349753618240356, "epoch": 4, "memory": 25074, "step": 22174} +{"lr": 0.0037825580157557948, "data_time": 0.0013096332550048828, "loss": 0.19900043457746505, "time": 0.9053715467453003, "epoch": 4, "memory": 25074, "step": 22274} +{"lr": 0.0037825580157557948, "data_time": 0.0014603853225708008, "loss": 0.1767716571688652, "time": 0.8267655372619629, "epoch": 4, "memory": 25074, "step": 22374} +{"lr": 0.0037825580157557948, "data_time": 0.0028059720993041993, "loss": 0.18151189237833024, "time": 0.8090237855911255, "epoch": 4, "memory": 25074, "step": 22474} +{"lr": 0.0037825580157557948, "data_time": 0.0025251626968383787, "loss": 0.16442398279905318, "time": 0.755122184753418, "epoch": 4, "memory": 25074, "step": 22574} +{"lr": 0.0037825580157557948, "data_time": 0.001403474807739258, "loss": 0.219259575009346, "time": 0.9044676065444947, "epoch": 4, "memory": 25074, "step": 22674} +{"lr": 0.0037825580157557948, "data_time": 0.0019288301467895509, "loss": 0.21363774687051773, "time": 0.8168543100357055, "epoch": 4, "memory": 25074, "step": 22774} +{"lr": 0.0037825580157557948, "data_time": 0.0016801834106445312, "loss": 0.16989538371562957, "time": 0.9056608200073242, "epoch": 4, "memory": 25074, "step": 22874} +{"lr": 0.0037825580157557948, "data_time": 0.0020328760147094727, "loss": 0.1526303119957447, "time": 0.8601443052291871, "epoch": 4, "memory": 25074, "step": 22974} +{"accuracy/top1": 100.0, "data_time": 0.2402495917151956, "time": 0.6277029374066521, "step": 4} +{"lr": 0.0036189889037780212, "data_time": 0.002791881561279297, "loss": 0.1409598857164383, "time": 0.7361682176589965, "epoch": 5, "memory": 25074, "step": 23132} +{"lr": 0.0036189889037780212, "data_time": 0.002370285987854004, "loss": 0.13107298240065574, "time": 0.7838279008865356, "epoch": 5, "memory": 25074, "step": 23232} +{"lr": 0.0036189889037780212, "data_time": 0.002247118949890137, "loss": 0.1290941208600998, "time": 0.7731497287750244, "epoch": 5, "memory": 25074, "step": 23332} +{"lr": 0.0036189889037780212, "data_time": 0.0020523309707641602, "loss": 0.14304085224866867, "time": 0.835112190246582, "epoch": 5, "memory": 25074, "step": 23432} +{"lr": 0.0036189889037780212, "data_time": 0.0020323753356933593, "loss": 0.10982143506407738, "time": 0.8326578378677368, "epoch": 5, "memory": 25074, "step": 23532} +{"lr": 0.0036189889037780212, "data_time": 0.0021000146865844727, "loss": 0.14942233264446259, "time": 0.8026610374450683, "epoch": 5, "memory": 25074, "step": 23632} +{"lr": 0.0036189889037780212, "data_time": 0.0015743017196655274, "loss": 0.10007348507642747, "time": 0.8582959651947022, "epoch": 5, "memory": 25074, "step": 23732} +{"lr": 0.0036189889037780212, "data_time": 0.0023128509521484373, "loss": 0.11920400932431222, "time": 0.8757483005523682, "epoch": 5, "memory": 25074, "step": 23832} +{"lr": 0.0036189889037780212, "data_time": 0.0019636154174804688, "loss": 0.09923064410686493, "time": 0.7413851976394653, "epoch": 5, "memory": 25074, "step": 23932} +{"lr": 0.0036189889037780212, "data_time": 0.0018323898315429688, "loss": 0.5074950933456421, "time": 0.8162568092346192, "epoch": 5, "memory": 25074, "step": 24032} +{"lr": 0.0036189889037780212, "data_time": 0.002145814895629883, "loss": 0.2000981852412224, "time": 0.7811573505401611, "epoch": 5, "memory": 25074, "step": 24132} +{"lr": 0.0036189889037780212, "data_time": 0.0018929481506347657, "loss": 0.1282152235507965, "time": 0.9015360593795776, "epoch": 5, "memory": 25074, "step": 24232} +{"lr": 0.0036189889037780212, "data_time": 0.001922917366027832, "loss": 0.07424237728118896, "time": 0.7680579423904419, "epoch": 5, "memory": 25074, "step": 24332} +{"lr": 0.0036189889037780212, "data_time": 0.0016839981079101562, "loss": 0.1415588803589344, "time": 0.8025635004043579, "epoch": 5, "memory": 25074, "step": 24432} +{"lr": 0.0036189889037780212, "data_time": 0.0021357297897338866, "loss": 0.07980295084416866, "time": 1.0372174501419067, "epoch": 5, "memory": 25074, "step": 24532} +{"lr": 0.0036189889037780212, "data_time": 0.001970791816711426, "loss": 0.09773364216089249, "time": 0.8294796705245971, "epoch": 5, "memory": 25074, "step": 24632} +{"lr": 0.0036189889037780212, "data_time": 0.0015277624130249023, "loss": 0.07760869041085243, "time": 0.8010515213012696, "epoch": 5, "memory": 25074, "step": 24732} +{"lr": 0.0036189889037780212, "data_time": 0.0018419981002807616, "loss": 0.08253993056714534, "time": 0.7675093173980713, "epoch": 5, "memory": 25074, "step": 24832} +{"lr": 0.0036189889037780212, "data_time": 0.002334880828857422, "loss": 0.15845463946461677, "time": 0.8050356388092041, "epoch": 5, "memory": 25074, "step": 24932} +{"lr": 0.0036189889037780212, "data_time": 0.0022067785263061523, "loss": 0.0716423649340868, "time": 0.8166382074356079, "epoch": 5, "memory": 25074, "step": 25032} +{"lr": 0.0036189889037780212, "data_time": 0.0020513534545898438, "loss": 0.05277598164975643, "time": 0.9307221889495849, "epoch": 5, "memory": 25074, "step": 25132} +{"lr": 0.0036189889037780212, "data_time": 0.0015638113021850587, "loss": 0.055456786043941976, "time": 0.7789030075073242, "epoch": 5, "memory": 25074, "step": 25232} +{"lr": 0.0036189889037780212, "data_time": 0.002756953239440918, "loss": 0.06829598098993302, "time": 0.7972937822341919, "epoch": 5, "memory": 25074, "step": 25332} +{"lr": 0.0036189889037780212, "data_time": 0.0018646717071533203, "loss": 0.1625942789018154, "time": 0.8168657064437866, "epoch": 5, "memory": 25074, "step": 25432} +{"lr": 0.0036189889037780212, "data_time": 0.0019771814346313476, "loss": 0.05140396654605865, "time": 0.8372904300689697, "epoch": 5, "memory": 25074, "step": 25532} +{"lr": 0.0036189889037780212, "data_time": 0.0021970987319946287, "loss": 0.04322529584169388, "time": 0.8783126831054687, "epoch": 5, "memory": 25074, "step": 25632} +{"lr": 0.0036189889037780212, "data_time": 0.002334904670715332, "loss": 0.08789301030337811, "time": 0.7887928724288941, "epoch": 5, "memory": 25074, "step": 25732} +{"lr": 0.0036189889037780212, "data_time": 0.002146148681640625, "loss": 0.03907903414219618, "time": 0.8402746915817261, "epoch": 5, "memory": 25074, "step": 25832} +{"lr": 0.0036189889037780212, "data_time": 0.001930546760559082, "loss": 0.34184410572052004, "time": 0.8359517574310302, "epoch": 5, "memory": 25074, "step": 25932} +{"lr": 0.0036189889037780212, "data_time": 0.0015136480331420898, "loss": 0.053155054431408645, "time": 0.7919988393783569, "epoch": 5, "memory": 25074, "step": 26032} +{"lr": 0.0036189889037780212, "data_time": 0.0017973899841308594, "loss": 0.04823228809982538, "time": 0.8395230054855347, "epoch": 5, "memory": 25074, "step": 26132} +{"lr": 0.0036189889037780212, "data_time": 0.002233123779296875, "loss": 0.5153215765953064, "time": 0.8028883218765259, "epoch": 5, "memory": 25074, "step": 26232} +{"lr": 0.0036189889037780212, "data_time": 0.002189040184020996, "loss": 0.29221029579639435, "time": 0.8329488515853882, "epoch": 5, "memory": 25074, "step": 26332} +{"lr": 0.0036189889037780212, "data_time": 0.0020819664001464843, "loss": 0.1691926673054695, "time": 0.8191149234771729, "epoch": 5, "memory": 25074, "step": 26432} +{"lr": 0.0036189889037780212, "data_time": 0.002014946937561035, "loss": 0.071760324947536, "time": 0.8359919786453247, "epoch": 5, "memory": 25074, "step": 26532} +{"lr": 0.0036189889037780212, "data_time": 0.001936483383178711, "loss": 0.05709798783063889, "time": 0.8147225618362427, "epoch": 5, "memory": 25074, "step": 26632} +{"lr": 0.0036189889037780212, "data_time": 0.0022314310073852537, "loss": 0.03994304072111845, "time": 0.7918617963790894, "epoch": 5, "memory": 25074, "step": 26732} +{"lr": 0.0036189889037780212, "data_time": 0.0016577720642089843, "loss": 0.023929027561098336, "time": 0.7573018550872803, "epoch": 5, "memory": 25074, "step": 26832} +{"lr": 0.0036189889037780212, "data_time": 0.001753520965576172, "loss": 0.048914790898561475, "time": 0.8102521896362305, "epoch": 5, "memory": 25074, "step": 26932} +{"lr": 0.0036189889037780212, "data_time": 0.0019876241683959963, "loss": 0.022528800275176765, "time": 0.8083161830902099, "epoch": 5, "memory": 25074, "step": 27032} +{"lr": 0.0036189889037780212, "data_time": 0.002398037910461426, "loss": 0.03067467249929905, "time": 0.8007509708404541, "epoch": 5, "memory": 25074, "step": 27132} +{"lr": 0.0036189889037780212, "data_time": 0.0018538713455200195, "loss": 0.04042974412441254, "time": 0.7805203437805176, "epoch": 5, "memory": 25074, "step": 27232} +{"lr": 0.0036189889037780212, "data_time": 0.0015351533889770507, "loss": 0.02192025911062956, "time": 0.8094108581542969, "epoch": 5, "memory": 25074, "step": 27332} +{"lr": 0.0036189889037780212, "data_time": 0.002463555335998535, "loss": 0.021817026380449533, "time": 0.8124203443527221, "epoch": 5, "memory": 25074, "step": 27432} +{"lr": 0.0036189889037780212, "data_time": 0.0015733003616333007, "loss": 0.023982952069491147, "time": 0.7534865856170654, "epoch": 5, "memory": 25074, "step": 27532} +{"lr": 0.0036189889037780212, "data_time": 0.0021764039993286133, "loss": 0.277609246969223, "time": 0.7911419153213501, "epoch": 5, "memory": 25074, "step": 27632} +{"lr": 0.0036189889037780212, "data_time": 0.0014683961868286132, "loss": 0.02815322340466082, "time": 0.7884535789489746, "epoch": 5, "memory": 25074, "step": 27732} +{"lr": 0.0036189889037780212, "data_time": 0.0021384000778198243, "loss": 0.030094720143824815, "time": 0.7951118230819703, "epoch": 5, "memory": 25074, "step": 27832} +{"lr": 0.0036189889037780212, "data_time": 0.0016850471496582032, "loss": 0.021415264159440995, "time": 0.7606137752532959, "epoch": 5, "memory": 25074, "step": 27932} +{"lr": 0.0036189889037780212, "data_time": 0.0019913434982299803, "loss": 0.020857601333409546, "time": 0.8410851955413818, "epoch": 5, "memory": 25074, "step": 28032} +{"lr": 0.0036189889037780212, "data_time": 0.002279233932495117, "loss": 0.030796349234879016, "time": 0.7919528961181641, "epoch": 5, "memory": 25074, "step": 28132} +{"lr": 0.0036189889037780212, "data_time": 0.0018125057220458984, "loss": 0.7452868342399597, "time": 0.8143720149993896, "epoch": 5, "memory": 25074, "step": 28232} +{"lr": 0.0036189889037780212, "data_time": 0.0017468452453613282, "loss": 0.5256998062133789, "time": 0.7838029623031616, "epoch": 5, "memory": 25074, "step": 28332} +{"lr": 0.0036189889037780212, "data_time": 0.0015678882598876953, "loss": 0.3817441314458847, "time": 0.7522627592086792, "epoch": 5, "memory": 25074, "step": 28432} +{"lr": 0.0036189889037780212, "data_time": 0.0015955686569213868, "loss": 0.2653026685118675, "time": 0.7718812465667725, "epoch": 5, "memory": 25074, "step": 28532} +{"lr": 0.0036189889037780212, "data_time": 0.0018198490142822266, "loss": 0.5123111963272095, "time": 0.8409427165985107, "epoch": 5, "memory": 25074, "step": 28632} +{"lr": 0.0036189889037780212, "data_time": 0.0015866756439208984, "loss": 0.2875907853245735, "time": 0.7826916933059692, "epoch": 5, "memory": 25074, "step": 28732} +{"accuracy/top1": 100.0, "data_time": 0.23484166930703557, "time": 0.6201901716344497, "step": 5} +{"lr": 0.003415678028467163, "data_time": 0.0018018245697021484, "loss": 0.17174191698431968, "time": 0.7909611940383912, "epoch": 6, "memory": 25074, "step": 28890} +{"lr": 0.003415678028467163, "data_time": 0.0017652511596679688, "loss": 0.0633200965821743, "time": 0.8031851530075074, "epoch": 6, "memory": 25074, "step": 28990} +{"lr": 0.003415678028467163, "data_time": 0.0019463062286376952, "loss": 0.03496393393725157, "time": 0.7924827337265015, "epoch": 6, "memory": 25074, "step": 29090} +{"lr": 0.003415678028467163, "data_time": 0.0024018049240112304, "loss": 0.02982586110010743, "time": 0.821974778175354, "epoch": 6, "memory": 25074, "step": 29190} +{"lr": 0.003415678028467163, "data_time": 0.0015446901321411132, "loss": 0.04476702082902193, "time": 0.7985591888427734, "epoch": 6, "memory": 25074, "step": 29290} +{"lr": 0.003415678028467163, "data_time": 0.0023563623428344725, "loss": 0.028352122567594052, "time": 0.7858057975769043, "epoch": 6, "memory": 25074, "step": 29390} +{"lr": 0.003415678028467163, "data_time": 0.0019240856170654296, "loss": 0.0374476021155715, "time": 0.8087836265563965, "epoch": 6, "memory": 25074, "step": 29490} +{"lr": 0.003415678028467163, "data_time": 0.0023302078247070313, "loss": 0.024446723610162736, "time": 0.7831089735031128, "epoch": 6, "memory": 25074, "step": 29590} +{"lr": 0.003415678028467163, "data_time": 0.0026931524276733398, "loss": 0.027217399375513197, "time": 0.8390152931213379, "epoch": 6, "memory": 25074, "step": 29690} +{"lr": 0.003415678028467163, "data_time": 0.0023102045059204103, "loss": 0.04310166211798787, "time": 0.7135088682174683, "epoch": 6, "memory": 25074, "step": 29790} +{"lr": 0.003415678028467163, "data_time": 0.002012801170349121, "loss": 0.01963822771795094, "time": 0.8424479722976684, "epoch": 6, "memory": 25074, "step": 29890} +{"lr": 0.003415678028467163, "data_time": 0.0021408796310424805, "loss": 0.025779563747346402, "time": 0.7580039501190186, "epoch": 6, "memory": 25074, "step": 29990} +{"lr": 0.003415678028467163, "data_time": 0.0019848823547363283, "loss": 0.0160231560934335, "time": 0.8077895164489746, "epoch": 6, "memory": 25074, "step": 30090} +{"lr": 0.003415678028467163, "data_time": 0.0022003173828125, "loss": 0.023282242938876152, "time": 0.8337722539901733, "epoch": 6, "memory": 25074, "step": 30190} +{"lr": 0.003415678028467163, "data_time": 0.0017018318176269531, "loss": 0.026293031871318817, "time": 0.7885631799697876, "epoch": 6, "memory": 25074, "step": 30290} +{"lr": 0.003415678028467163, "data_time": 0.001690506935119629, "loss": 0.019426302472129463, "time": 0.7699220418930054, "epoch": 6, "memory": 25074, "step": 30390} +{"lr": 0.003415678028467163, "data_time": 0.001882314682006836, "loss": 0.024549905210733414, "time": 0.7650113105773926, "epoch": 6, "memory": 25074, "step": 30490} +{"lr": 0.003415678028467163, "data_time": 0.0013676881790161133, "loss": 0.9105373382568359, "time": 0.7781325578689575, "epoch": 6, "memory": 25074, "step": 30590} +{"lr": 0.003415678028467163, "data_time": 0.001557326316833496, "loss": 0.6870330810546875, "time": 0.7517716884613037, "epoch": 6, "memory": 25074, "step": 30690} +{"lr": 0.003415678028467163, "data_time": 0.0018422365188598632, "loss": 0.6858840942382812, "time": 0.809003210067749, "epoch": 6, "memory": 25074, "step": 30790} +{"lr": 0.003415678028467163, "data_time": 0.0016306638717651367, "loss": 0.6863113403320312, "time": 0.8211899518966674, "epoch": 6, "memory": 25074, "step": 30890} +{"lr": 0.003415678028467163, "data_time": 0.001467752456665039, "loss": 0.689276123046875, "time": 0.760313606262207, "epoch": 6, "memory": 25074, "step": 30990} +{"lr": 0.003415678028467163, "data_time": 0.0016613483428955078, "loss": 0.6889556884765625, "time": 0.7955705165863037, "epoch": 6, "memory": 25074, "step": 31090} +{"lr": 0.003415678028467163, "data_time": 0.001390528678894043, "loss": 0.6836715698242187, "time": 0.814949369430542, "epoch": 6, "memory": 25074, "step": 31190} +{"lr": 0.003415678028467163, "data_time": 0.06151556968688965, "loss": 0.6795379638671875, "time": 0.7687273979187011, "epoch": 6, "memory": 25074, "step": 31290} +{"lr": 0.003415678028467163, "data_time": 0.0012712955474853515, "loss": 0.6829338073730469, "time": 0.7202790260314942, "epoch": 6, "memory": 25074, "step": 31390} +{"lr": 0.003415678028467163, "data_time": 0.0014473199844360352, "loss": 0.6826400756835938, "time": 0.8305152893066406, "epoch": 6, "memory": 25074, "step": 31490} +{"lr": 0.003415678028467163, "data_time": 0.0018189430236816406, "loss": 0.6878402709960938, "time": 0.8051482677459717, "epoch": 6, "memory": 25074, "step": 31590} +{"lr": 0.003415678028467163, "data_time": 0.0013915538787841798, "loss": 0.6876800537109375, "time": 0.7747416973114014, "epoch": 6, "memory": 25074, "step": 31690} +{"lr": 0.003415678028467163, "data_time": 0.0015104293823242187, "loss": 0.6831192016601563, "time": 0.7355213165283203, "epoch": 6, "memory": 25074, "step": 31790} +{"lr": 0.003415678028467163, "data_time": 0.0018348217010498047, "loss": 0.6800773620605469, "time": 0.7252334833145142, "epoch": 6, "memory": 25074, "step": 31890} +{"lr": 0.003415678028467163, "data_time": 0.0016130924224853516, "loss": 0.6558177947998047, "time": 0.7361180305480957, "epoch": 6, "memory": 25074, "step": 31990} +{"lr": 0.003415678028467163, "data_time": 0.002025461196899414, "loss": 0.6396217346191406, "time": 0.7465869903564453, "epoch": 6, "memory": 25074, "step": 32090} +{"lr": 0.003415678028467163, "data_time": 0.001508045196533203, "loss": 0.5973075866699219, "time": 0.7108067750930787, "epoch": 6, "memory": 25074, "step": 32190} +{"lr": 0.003415678028467163, "data_time": 0.0015272140502929688, "loss": 0.5695059776306153, "time": 0.7145503282546997, "epoch": 6, "memory": 25074, "step": 32290} +{"lr": 0.003415678028467163, "data_time": 0.0019104957580566406, "loss": 0.5324402332305909, "time": 0.7866726398468018, "epoch": 6, "memory": 25074, "step": 32390} +{"lr": 0.003415678028467163, "data_time": 0.0016667366027832032, "loss": 0.5393310070037842, "time": 0.7764627933502197, "epoch": 6, "memory": 25074, "step": 32490} +{"lr": 0.003415678028467163, "data_time": 0.0019632816314697266, "loss": 0.4775394558906555, "time": 0.8011926412582397, "epoch": 6, "memory": 25074, "step": 32590} +{"lr": 0.003415678028467163, "data_time": 0.0019040107727050781, "loss": 0.46249977946281434, "time": 0.7955193758010864, "epoch": 6, "memory": 25074, "step": 32690} +{"lr": 0.003415678028467163, "data_time": 0.0021024227142333986, "loss": 0.43382647037506106, "time": 0.833111572265625, "epoch": 6, "memory": 25074, "step": 32790} +{"lr": 0.003415678028467163, "data_time": 0.001708245277404785, "loss": 0.39271233677864076, "time": 0.8732306241989136, "epoch": 6, "memory": 25074, "step": 32890} +{"lr": 0.003415678028467163, "data_time": 0.0018872737884521485, "loss": 0.39725507497787477, "time": 0.8073127508163452, "epoch": 6, "memory": 25074, "step": 32990} +{"lr": 0.003415678028467163, "data_time": 0.0018921613693237305, "loss": 0.34503606855869295, "time": 0.8197930812835693, "epoch": 6, "memory": 25074, "step": 33090} +{"lr": 0.003415678028467163, "data_time": 0.001664447784423828, "loss": 0.34498479068279264, "time": 0.825571870803833, "epoch": 6, "memory": 25074, "step": 33190} +{"lr": 0.003415678028467163, "data_time": 0.0015811443328857422, "loss": 0.30077325105667113, "time": 0.8408125638961792, "epoch": 6, "memory": 25074, "step": 33290} +{"lr": 0.003415678028467163, "data_time": 0.0018062353134155273, "loss": 0.28608314096927645, "time": 0.8217522144317627, "epoch": 6, "memory": 25074, "step": 33390} +{"lr": 0.003415678028467163, "data_time": 0.0020669937133789063, "loss": 0.27477487176656723, "time": 0.8694106578826905, "epoch": 6, "memory": 25074, "step": 33490} +{"lr": 0.003415678028467163, "data_time": 0.0017047166824340821, "loss": 0.24026929885149, "time": 0.8727396726608276, "epoch": 6, "memory": 25074, "step": 33590} +{"lr": 0.003415678028467163, "data_time": 0.0023725271224975587, "loss": 0.19539487808942796, "time": 0.8610022783279419, "epoch": 6, "memory": 25074, "step": 33690} +{"lr": 0.003415678028467163, "data_time": 0.0020278453826904296, "loss": 0.22524606734514235, "time": 0.7790164947509766, "epoch": 6, "memory": 25074, "step": 33790} +{"lr": 0.003415678028467163, "data_time": 0.001953601837158203, "loss": 0.21684082001447677, "time": 0.8055158138275147, "epoch": 6, "memory": 25074, "step": 33890} +{"lr": 0.003415678028467163, "data_time": 0.001944732666015625, "loss": 0.17871028035879136, "time": 0.7842403173446655, "epoch": 6, "memory": 25074, "step": 33990} +{"lr": 0.003415678028467163, "data_time": 0.002075338363647461, "loss": 0.3113388612866402, "time": 0.8257120609283447, "epoch": 6, "memory": 25074, "step": 34090} +{"lr": 0.003415678028467163, "data_time": 0.0019263029098510742, "loss": 0.19416865408420564, "time": 0.785835599899292, "epoch": 6, "memory": 25074, "step": 34190} +{"lr": 0.003415678028467163, "data_time": 0.0021706342697143553, "loss": 0.16097232401371003, "time": 0.8521947145462037, "epoch": 6, "memory": 25074, "step": 34290} +{"lr": 0.003415678028467163, "data_time": 0.0021689891815185546, "loss": 0.12806736528873444, "time": 0.8027761936187744, "epoch": 6, "memory": 25074, "step": 34390} +{"lr": 0.003415678028467163, "data_time": 0.001958608627319336, "loss": 0.08435219004750252, "time": 0.8249452114105225, "epoch": 6, "memory": 25074, "step": 34490} +{"accuracy/top1": 100.0, "data_time": 0.22624763320474064, "time": 0.6205435079686782, "step": 6} +{"lr": 0.003177631578323485, "data_time": 0.0017069339752197265, "loss": 0.08231379240751266, "time": 0.7843175649642944, "epoch": 7, "memory": 25074, "step": 34648} +{"lr": 0.003177631578323485, "data_time": 0.0022977590560913086, "loss": 0.08582820855081082, "time": 0.7684028387069702, "epoch": 7, "memory": 25074, "step": 34748} +{"lr": 0.003177631578323485, "data_time": 0.001708054542541504, "loss": 0.08164712898433209, "time": 0.8050656318664551, "epoch": 7, "memory": 25074, "step": 34848} +{"lr": 0.003177631578323485, "data_time": 0.0018536806106567382, "loss": 0.10651303324848413, "time": 0.8809154272079468, "epoch": 7, "memory": 25074, "step": 34948} +{"lr": 0.003177631578323485, "data_time": 0.002226734161376953, "loss": 0.055012472718954084, "time": 0.7990552902221679, "epoch": 7, "memory": 25074, "step": 35048} +{"lr": 0.003177631578323485, "data_time": 0.002142786979675293, "loss": 0.06312011554837227, "time": 0.8404693365097046, "epoch": 7, "memory": 25074, "step": 35148} +{"lr": 0.003177631578323485, "data_time": 0.001969790458679199, "loss": 0.053021703660488126, "time": 0.8096812725067138, "epoch": 7, "memory": 25074, "step": 35248} +{"lr": 0.003177631578323485, "data_time": 0.0022991418838500975, "loss": 0.05477383937686682, "time": 0.8274132013320923, "epoch": 7, "memory": 25074, "step": 35348} +{"lr": 0.003177631578323485, "data_time": 0.001857590675354004, "loss": 0.059876644238829616, "time": 0.8286162137985229, "epoch": 7, "memory": 25074, "step": 35448} +{"lr": 0.003177631578323485, "data_time": 0.0020510673522949217, "loss": 0.03685490805655718, "time": 0.8613656282424926, "epoch": 7, "memory": 25074, "step": 35548} +{"lr": 0.003177631578323485, "data_time": 0.0019825220108032225, "loss": 0.02592898178845644, "time": 0.8106374502182007, "epoch": 7, "memory": 25074, "step": 35648} +{"lr": 0.003177631578323485, "data_time": 0.0016251325607299805, "loss": 0.05490502044558525, "time": 0.7780727624893189, "epoch": 7, "memory": 25074, "step": 35748} +{"lr": 0.003177631578323485, "data_time": 0.0018387556076049805, "loss": 0.04498982932418585, "time": 0.8170005559921265, "epoch": 7, "memory": 25074, "step": 35848} +{"lr": 0.003177631578323485, "data_time": 0.0027972936630249025, "loss": 0.034482643753290174, "time": 0.9129539012908936, "epoch": 7, "memory": 25074, "step": 35948} +{"lr": 0.003177631578323485, "data_time": 0.0016614913940429688, "loss": 0.5376731395721436, "time": 0.8564528703689576, "epoch": 7, "memory": 25074, "step": 36048} +{"lr": 0.003177631578323485, "data_time": 0.0023099660873413088, "loss": 0.3870724499225616, "time": 0.8048108577728271, "epoch": 7, "memory": 25074, "step": 36148} +{"lr": 0.003177631578323485, "data_time": 0.0018640518188476562, "loss": 0.25513831079006194, "time": 0.8086620092391967, "epoch": 7, "memory": 25074, "step": 36248} +{"lr": 0.003177631578323485, "data_time": 0.0016347885131835938, "loss": 0.10172961838543415, "time": 0.7994915008544922, "epoch": 7, "memory": 25074, "step": 36348} +{"lr": 0.003177631578323485, "data_time": 0.0017717599868774414, "loss": 0.05500081535428762, "time": 0.7761198282241821, "epoch": 7, "memory": 25074, "step": 36448} +{"lr": 0.003177631578323485, "data_time": 0.0016062021255493163, "loss": 0.03965157438069582, "time": 0.82099609375, "epoch": 7, "memory": 25074, "step": 36548} +{"lr": 0.003177631578323485, "data_time": 0.001989150047302246, "loss": 0.033694837940856814, "time": 0.8073045492172242, "epoch": 7, "memory": 25074, "step": 36648} +{"lr": 0.003177631578323485, "data_time": 0.0015305519104003907, "loss": 0.051020176336169246, "time": 0.8328912734985352, "epoch": 7, "memory": 25074, "step": 36748} +{"lr": 0.003177631578323485, "data_time": 0.0017588138580322266, "loss": 0.03742875270545483, "time": 0.8235810279846192, "epoch": 7, "memory": 25074, "step": 36848} +{"lr": 0.003177631578323485, "data_time": 0.001764059066772461, "loss": 0.03348587434738874, "time": 0.7547449350357056, "epoch": 7, "memory": 25074, "step": 36948} +{"lr": 0.003177631578323485, "data_time": 0.0014856100082397462, "loss": 0.03297441750764847, "time": 0.811091947555542, "epoch": 7, "memory": 25074, "step": 37048} +{"lr": 0.003177631578323485, "data_time": 0.002025747299194336, "loss": 0.027984938118606806, "time": 0.7961791515350342, "epoch": 7, "memory": 25074, "step": 37148} +{"lr": 0.003177631578323485, "data_time": 0.001494598388671875, "loss": 0.019442465668544174, "time": 0.8026607275009155, "epoch": 7, "memory": 25074, "step": 37248} +{"lr": 0.003177631578323485, "data_time": 0.0013259649276733398, "loss": 0.4900744140148163, "time": 0.7862446784973145, "epoch": 7, "memory": 25074, "step": 37348} +{"lr": 0.003177631578323485, "data_time": 0.0015650272369384765, "loss": 0.38184359967708587, "time": 0.8663106203079224, "epoch": 7, "memory": 25074, "step": 37448} +{"lr": 0.003177631578323485, "data_time": 0.0014769077301025391, "loss": 0.3002003714442253, "time": 0.8492149591445923, "epoch": 7, "memory": 25074, "step": 37548} +{"lr": 0.003177631578323485, "data_time": 0.001516580581665039, "loss": 0.22087419629096985, "time": 0.7983831644058228, "epoch": 7, "memory": 25074, "step": 37648} +{"lr": 0.003177631578323485, "data_time": 0.001896071434020996, "loss": 0.1823909968137741, "time": 0.8249658346176147, "epoch": 7, "memory": 25074, "step": 37748} +{"lr": 0.003177631578323485, "data_time": 0.0029141902923583984, "loss": 0.0875114869326353, "time": 0.7948890686035156, "epoch": 7, "memory": 25074, "step": 37848} +{"lr": 0.003177631578323485, "data_time": 0.001998114585876465, "loss": 0.03376759458333254, "time": 0.8398072242736816, "epoch": 7, "memory": 25074, "step": 37948} +{"lr": 0.003177631578323485, "data_time": 0.0017653226852416993, "loss": 0.06241262629628182, "time": 0.7995390892028809, "epoch": 7, "memory": 25074, "step": 38048} +{"lr": 0.003177631578323485, "data_time": 0.0017367124557495118, "loss": 0.026532156532630323, "time": 0.7869459390640259, "epoch": 7, "memory": 25074, "step": 38148} +{"lr": 0.003177631578323485, "data_time": 0.0021256446838378907, "loss": 0.023698237910866737, "time": 0.837284255027771, "epoch": 7, "memory": 25074, "step": 38248} +{"lr": 0.003177631578323485, "data_time": 0.002065277099609375, "loss": 0.04279262721538544, "time": 0.9142627954483032, "epoch": 7, "memory": 25074, "step": 38348} +{"lr": 0.003177631578323485, "data_time": 0.0024343013763427736, "loss": 0.020273453602567316, "time": 0.7826040983200073, "epoch": 7, "memory": 25074, "step": 38448} +{"lr": 0.003177631578323485, "data_time": 0.0016155242919921875, "loss": 0.01841148789972067, "time": 0.8423229694366455, "epoch": 7, "memory": 25074, "step": 38548} +{"lr": 0.003177631578323485, "data_time": 0.0017494440078735351, "loss": 0.031658689863979815, "time": 0.8426234483718872, "epoch": 7, "memory": 25074, "step": 38648} +{"lr": 0.003177631578323485, "data_time": 0.0018595218658447265, "loss": 0.06796988733112812, "time": 0.7892760753631591, "epoch": 7, "memory": 25074, "step": 38748} +{"lr": 0.003177631578323485, "data_time": 0.002170705795288086, "loss": 0.015778914047405123, "time": 0.8710393190383912, "epoch": 7, "memory": 25074, "step": 38848} +{"lr": 0.003177631578323485, "data_time": 0.0017850399017333984, "loss": 0.018908139877021314, "time": 0.8602873325347901, "epoch": 7, "memory": 25074, "step": 38948} +{"lr": 0.003177631578323485, "data_time": 0.0016251087188720703, "loss": 0.01837123604491353, "time": 0.8457715511322021, "epoch": 7, "memory": 25074, "step": 39048} +{"lr": 0.003177631578323485, "data_time": 0.0017221927642822265, "loss": 0.01905709123238921, "time": 0.8067245244979858, "epoch": 7, "memory": 25074, "step": 39148} +{"lr": 0.003177631578323485, "data_time": 0.00210263729095459, "loss": 0.013219493068754673, "time": 0.8204404830932617, "epoch": 7, "memory": 25074, "step": 39248} +{"lr": 0.003177631578323485, "data_time": 0.001557326316833496, "loss": 0.025120821595191956, "time": 0.7940375328063964, "epoch": 7, "memory": 25074, "step": 39348} +{"lr": 0.003177631578323485, "data_time": 0.0016313314437866211, "loss": 0.02124868012033403, "time": 0.9074863195419312, "epoch": 7, "memory": 25074, "step": 39448} +{"lr": 0.003177631578323485, "data_time": 0.0014491319656372071, "loss": 0.6646780014038086, "time": 0.7914240837097168, "epoch": 7, "memory": 25074, "step": 39548} +{"lr": 0.003177631578323485, "data_time": 0.001481318473815918, "loss": 0.6365413665771484, "time": 0.8293594598770142, "epoch": 7, "memory": 25074, "step": 39648} +{"lr": 0.003177631578323485, "data_time": 0.0016228437423706054, "loss": 0.6289838790893555, "time": 0.7879393100738525, "epoch": 7, "memory": 25074, "step": 39748} +{"lr": 0.003177631578323485, "data_time": 0.0018411874771118164, "loss": 0.6199359893798828, "time": 0.8507879257202149, "epoch": 7, "memory": 25074, "step": 39848} +{"lr": 0.003177631578323485, "data_time": 0.0017439842224121094, "loss": 0.6042745590209961, "time": 0.8092621564865112, "epoch": 7, "memory": 25074, "step": 39948} +{"lr": 0.003177631578323485, "data_time": 0.0021016836166381837, "loss": 0.5974456787109375, "time": 0.7967112302780152, "epoch": 7, "memory": 25074, "step": 40048} +{"lr": 0.003177631578323485, "data_time": 0.0015798568725585937, "loss": 0.5755602836608886, "time": 0.7882761001586914, "epoch": 7, "memory": 25074, "step": 40148} +{"lr": 0.003177631578323485, "data_time": 0.0012903928756713868, "loss": 0.5639491558074952, "time": 0.8141366958618164, "epoch": 7, "memory": 25074, "step": 40248} +{"accuracy/top1": 100.0, "data_time": 0.22547212769003475, "time": 0.6097386163823745, "step": 7} +{"lr": 0.0029107110469803964, "data_time": 0.001800227165222168, "loss": 0.5362431287765503, "time": 0.7892891883850097, "epoch": 8, "memory": 25074, "step": 40406} +{"lr": 0.0029107110469803964, "data_time": 0.0021252870559692384, "loss": 0.5108250856399537, "time": 0.8145163297653198, "epoch": 8, "memory": 25074, "step": 40506} +{"lr": 0.0029107110469803964, "data_time": 0.0026922702789306642, "loss": 0.47698734998703, "time": 0.8064170360565186, "epoch": 8, "memory": 25074, "step": 40606} +{"lr": 0.0029107110469803964, "data_time": 0.002124929428100586, "loss": 0.4469404309988022, "time": 0.7941334486007691, "epoch": 8, "memory": 25074, "step": 40706} +{"lr": 0.0029107110469803964, "data_time": 0.0013925552368164063, "loss": 0.43811762928962705, "time": 0.7753614187240601, "epoch": 8, "memory": 25074, "step": 40806} +{"lr": 0.0029107110469803964, "data_time": 0.001997566223144531, "loss": 0.40253242254257204, "time": 0.8069181680679322, "epoch": 8, "memory": 25074, "step": 40906} +{"lr": 0.0029107110469803964, "data_time": 0.0015981435775756837, "loss": 0.39082804024219514, "time": 0.8386501789093017, "epoch": 8, "memory": 25074, "step": 41006} +{"lr": 0.0029107110469803964, "data_time": 0.0014615535736083984, "loss": 0.3861152559518814, "time": 0.785384202003479, "epoch": 8, "memory": 25074, "step": 41106} +{"lr": 0.0029107110469803964, "data_time": 0.001982879638671875, "loss": 0.38163594603538514, "time": 0.8258617162704468, "epoch": 8, "memory": 25074, "step": 41206} +{"lr": 0.0029107110469803964, "data_time": 0.0014397144317626954, "loss": 0.3460691422224045, "time": 0.8217541933059692, "epoch": 8, "memory": 25074, "step": 41306} +{"lr": 0.0029107110469803964, "data_time": 0.001651453971862793, "loss": 0.33383490443229674, "time": 0.8306535720825196, "epoch": 8, "memory": 25074, "step": 41406} +{"lr": 0.0029107110469803964, "data_time": 0.0013466358184814453, "loss": 0.31845238506793977, "time": 0.8040345668792724, "epoch": 8, "memory": 25074, "step": 41506} +{"lr": 0.0029107110469803964, "data_time": 0.0013608455657958985, "loss": 0.34845821261405946, "time": 0.8786239385604858, "epoch": 8, "memory": 25074, "step": 41606} +{"lr": 0.0029107110469803964, "data_time": 0.0013670921325683594, "loss": 0.278292253613472, "time": 0.8874995946884155, "epoch": 8, "memory": 25074, "step": 41706} +{"lr": 0.0029107110469803964, "data_time": 0.0013338565826416016, "loss": 0.31247431337833403, "time": 0.8517024755477905, "epoch": 8, "memory": 25074, "step": 41806} +{"lr": 0.0029107110469803964, "data_time": 0.0014211416244506836, "loss": 0.2974017262458801, "time": 0.8422387838363647, "epoch": 8, "memory": 25074, "step": 41906} +{"lr": 0.0029107110469803964, "data_time": 0.0015972614288330077, "loss": 0.28323196023702624, "time": 0.7977926969528198, "epoch": 8, "memory": 25074, "step": 42006} +{"lr": 0.0029107110469803964, "data_time": 0.0013134002685546875, "loss": 0.2619663327932358, "time": 0.802070426940918, "epoch": 8, "memory": 25074, "step": 42106} +{"lr": 0.0029107110469803964, "data_time": 0.0014792442321777343, "loss": 0.26130121052265165, "time": 0.7949286222457885, "epoch": 8, "memory": 25074, "step": 42206} +{"lr": 0.0029107110469803964, "data_time": 0.001403498649597168, "loss": 0.264859938621521, "time": 0.8208257913589477, "epoch": 8, "memory": 25074, "step": 42306} +{"lr": 0.0029107110469803964, "data_time": 0.001445770263671875, "loss": 0.2695415988564491, "time": 0.844876766204834, "epoch": 8, "memory": 25074, "step": 42406} +{"lr": 0.0029107110469803964, "data_time": 0.002265477180480957, "loss": 0.2874342992901802, "time": 0.8235554456710815, "epoch": 8, "memory": 25074, "step": 42506} +{"lr": 0.0029107110469803964, "data_time": 0.0021061420440673826, "loss": 0.24466877281665803, "time": 0.8314491510391235, "epoch": 8, "memory": 25074, "step": 42606} +{"lr": 0.0029107110469803964, "data_time": 0.0020581960678100588, "loss": 0.23984656482934952, "time": 0.8473775148391723, "epoch": 8, "memory": 25074, "step": 42706} +{"lr": 0.0029107110469803964, "data_time": 0.00217893123626709, "loss": 0.23493181318044662, "time": 0.7882248878479003, "epoch": 8, "memory": 25074, "step": 42806} +{"lr": 0.0029107110469803964, "data_time": 0.0023426532745361326, "loss": 0.23431477695703506, "time": 0.8113497257232666, "epoch": 8, "memory": 25074, "step": 42906} +{"lr": 0.0029107110469803964, "data_time": 0.0024402379989624024, "loss": 0.2268846094608307, "time": 0.8288700580596924, "epoch": 8, "memory": 25074, "step": 43006} +{"lr": 0.0029107110469803964, "data_time": 0.0019439220428466796, "loss": 0.21778939068317413, "time": 0.7623370170593262, "epoch": 8, "memory": 25074, "step": 43106} +{"lr": 0.0029107110469803964, "data_time": 0.0022038936614990233, "loss": 0.2172636419534683, "time": 0.7880473852157592, "epoch": 8, "memory": 25074, "step": 43206} +{"lr": 0.0029107110469803964, "data_time": 0.002065873146057129, "loss": 0.2091320648789406, "time": 0.8095156192779541, "epoch": 8, "memory": 25074, "step": 43306} +{"lr": 0.0029107110469803964, "data_time": 0.0018571853637695313, "loss": 0.17143266126513482, "time": 0.810135793685913, "epoch": 8, "memory": 25074, "step": 43406} +{"lr": 0.0029107110469803964, "data_time": 0.002043342590332031, "loss": 0.20797404944896697, "time": 0.8592607259750367, "epoch": 8, "memory": 25074, "step": 43506} +{"lr": 0.0029107110469803964, "data_time": 0.0015845775604248046, "loss": 0.18679274320602418, "time": 0.813219141960144, "epoch": 8, "memory": 25074, "step": 43606} +{"lr": 0.0029107110469803964, "data_time": 0.00219423770904541, "loss": 0.16646159291267396, "time": 0.7774173498153687, "epoch": 8, "memory": 25074, "step": 43706} +{"lr": 0.0029107110469803964, "data_time": 0.0021094560623168947, "loss": 0.1844821035861969, "time": 0.8706939220428467, "epoch": 8, "memory": 25074, "step": 43806} +{"lr": 0.0029107110469803964, "data_time": 0.002427840232849121, "loss": 0.16617073863744736, "time": 0.7723523139953613, "epoch": 8, "memory": 25074, "step": 43906} +{"lr": 0.0029107110469803964, "data_time": 0.0016525983810424805, "loss": 0.16185368224978447, "time": 0.798746919631958, "epoch": 8, "memory": 25074, "step": 44006} +{"lr": 0.0029107110469803964, "data_time": 0.0013972043991088868, "loss": 0.14784356951713562, "time": 0.8377389192581177, "epoch": 8, "memory": 25074, "step": 44106} +{"lr": 0.0029107110469803964, "data_time": 0.0014557123184204101, "loss": 0.15821471065282822, "time": 0.8408293485641479, "epoch": 8, "memory": 25074, "step": 44206} +{"lr": 0.0029107110469803964, "data_time": 0.0015608787536621094, "loss": 0.13272587954998016, "time": 0.836663794517517, "epoch": 8, "memory": 25074, "step": 44306} +{"lr": 0.0029107110469803964, "data_time": 0.0016949176788330078, "loss": 0.12774054631590842, "time": 0.820939040184021, "epoch": 8, "memory": 25074, "step": 44406} +{"lr": 0.0029107110469803964, "data_time": 0.0017126798629760742, "loss": 0.1299731768667698, "time": 0.8248666763305664, "epoch": 8, "memory": 25074, "step": 44506} +{"lr": 0.0029107110469803964, "data_time": 0.0018166542053222657, "loss": 0.13998845592141151, "time": 0.780283761024475, "epoch": 8, "memory": 25074, "step": 44606} +{"lr": 0.0029107110469803964, "data_time": 0.0018818140029907226, "loss": 0.11496206223964692, "time": 0.832054591178894, "epoch": 8, "memory": 25074, "step": 44706} +{"lr": 0.0029107110469803964, "data_time": 0.001940298080444336, "loss": 0.1016205944120884, "time": 0.8187478303909301, "epoch": 8, "memory": 25074, "step": 44806} +{"lr": 0.0029107110469803964, "data_time": 0.00205533504486084, "loss": 0.09862435236573219, "time": 0.7878509283065795, "epoch": 8, "memory": 25074, "step": 44906} +{"lr": 0.0029107110469803964, "data_time": 0.00294034481048584, "loss": 0.07922623381018638, "time": 0.7465308904647827, "epoch": 8, "memory": 25074, "step": 45006} +{"lr": 0.0029107110469803964, "data_time": 0.002628660202026367, "loss": 0.09312311969697476, "time": 0.7900871753692627, "epoch": 8, "memory": 25074, "step": 45106} +{"lr": 0.0029107110469803964, "data_time": 0.0020933628082275392, "loss": 0.09835705608129501, "time": 0.8151920795440674, "epoch": 8, "memory": 25074, "step": 45206} +{"lr": 0.0029107110469803964, "data_time": 0.0021629810333251955, "loss": 0.07780469469726085, "time": 0.807362413406372, "epoch": 8, "memory": 25074, "step": 45306} +{"lr": 0.0029107110469803964, "data_time": 0.0018307924270629882, "loss": 0.06943958066403866, "time": 0.8148200273513794, "epoch": 8, "memory": 25074, "step": 45406} +{"lr": 0.0029107110469803964, "data_time": 0.0014321565628051757, "loss": 0.06271064318716527, "time": 0.8063824415206909, "epoch": 8, "memory": 25074, "step": 45506} +{"lr": 0.0029107110469803964, "data_time": 0.0016618967056274414, "loss": 0.08301684185862541, "time": 0.7676124334335327, "epoch": 8, "memory": 25074, "step": 45606} +{"lr": 0.0029107110469803964, "data_time": 0.0019184350967407227, "loss": 0.06498735733330249, "time": 0.8050572395324707, "epoch": 8, "memory": 25074, "step": 45706} +{"lr": 0.0029107110469803964, "data_time": 0.002342534065246582, "loss": 0.07699587866663933, "time": 0.8105920076370239, "epoch": 8, "memory": 25074, "step": 45806} +{"lr": 0.0029107110469803964, "data_time": 0.001531982421875, "loss": 0.042582137137651445, "time": 0.8026448488235474, "epoch": 8, "memory": 25074, "step": 45906} +{"lr": 0.0029107110469803964, "data_time": 0.001994681358337402, "loss": 0.064727021753788, "time": 0.8076073884963989, "epoch": 8, "memory": 25074, "step": 46006} +{"accuracy/top1": 100.0, "data_time": 0.23159482899834127, "time": 0.625993434120627, "step": 8} +{"lr": 0.0026214889037780207, "data_time": 0.001961421966552734, "loss": 0.047271404415369034, "time": 0.793843674659729, "epoch": 9, "memory": 25074, "step": 46164} +{"lr": 0.0026214889037780207, "data_time": 0.0014253854751586914, "loss": 0.037891981471329925, "time": 0.8486331462860107, "epoch": 9, "memory": 25074, "step": 46264} +{"lr": 0.0026214889037780207, "data_time": 0.002214503288269043, "loss": 0.04856832753866911, "time": 0.8137295246124268, "epoch": 9, "memory": 25074, "step": 46364} +{"lr": 0.0026214889037780207, "data_time": 0.0021689891815185546, "loss": 0.03917246051132679, "time": 0.8412569761276245, "epoch": 9, "memory": 25074, "step": 46464} +{"lr": 0.0026214889037780207, "data_time": 0.0021068334579467775, "loss": 0.050056328624486925, "time": 0.8025254487991333, "epoch": 9, "memory": 25074, "step": 46564} +{"lr": 0.0026214889037780207, "data_time": 0.0019266605377197266, "loss": 0.06529371645301581, "time": 0.8107960939407348, "epoch": 9, "memory": 25074, "step": 46664} +{"lr": 0.0026214889037780207, "data_time": 0.0016360282897949219, "loss": 0.038940389081835745, "time": 0.7812315940856933, "epoch": 9, "memory": 25074, "step": 46764} +{"lr": 0.0026214889037780207, "data_time": 0.0023120641708374023, "loss": 0.04207463189959526, "time": 0.8007621765136719, "epoch": 9, "memory": 25074, "step": 46864} +{"lr": 0.0026214889037780207, "data_time": 0.0017193794250488282, "loss": 0.04248841069638729, "time": 0.7815124988555908, "epoch": 9, "memory": 25074, "step": 46964} +{"lr": 0.0026214889037780207, "data_time": 0.0023185253143310548, "loss": 0.04956267960369587, "time": 0.8030549049377441, "epoch": 9, "memory": 25074, "step": 47064} +{"lr": 0.0026214889037780207, "data_time": 0.0018356800079345702, "loss": 0.03543620202690363, "time": 0.8107479333877563, "epoch": 9, "memory": 25074, "step": 47164} +{"lr": 0.0026214889037780207, "data_time": 0.002096390724182129, "loss": 0.04235930442810058, "time": 0.808166241645813, "epoch": 9, "memory": 25074, "step": 47264} +{"lr": 0.0026214889037780207, "data_time": 0.0015815019607543946, "loss": 0.040592977032065394, "time": 0.86477210521698, "epoch": 9, "memory": 25074, "step": 47364} +{"lr": 0.0026214889037780207, "data_time": 0.002258706092834473, "loss": 0.058388377726078036, "time": 0.8222603797912598, "epoch": 9, "memory": 25074, "step": 47464} +{"lr": 0.0026214889037780207, "data_time": 0.0018778324127197265, "loss": 0.036899886652827264, "time": 0.7938441276550293, "epoch": 9, "memory": 25074, "step": 47564} +{"lr": 0.0026214889037780207, "data_time": 0.0020696163177490235, "loss": 0.026893947273492813, "time": 0.7813984870910644, "epoch": 9, "memory": 25074, "step": 47664} +{"lr": 0.0026214889037780207, "data_time": 0.0018400430679321289, "loss": 0.0410335611552, "time": 0.7760570526123047, "epoch": 9, "memory": 25074, "step": 47764} +{"lr": 0.0026214889037780207, "data_time": 0.00188138484954834, "loss": 0.030237929755821825, "time": 0.8536426305770874, "epoch": 9, "memory": 25074, "step": 47864} +{"lr": 0.0026214889037780207, "data_time": 0.0019226312637329102, "loss": 0.13606493994593621, "time": 0.8637728929519654, "epoch": 9, "memory": 25074, "step": 47964} +{"lr": 0.0026214889037780207, "data_time": 0.00211482048034668, "loss": 0.03358338922262192, "time": 0.7930330276489258, "epoch": 9, "memory": 25074, "step": 48064} +{"lr": 0.0026214889037780207, "data_time": 0.001715874671936035, "loss": 0.0319957572966814, "time": 0.7864399194717407, "epoch": 9, "memory": 25074, "step": 48164} +{"lr": 0.0026214889037780207, "data_time": 0.0022603750228881838, "loss": 0.025551164289936423, "time": 0.8411364555358887, "epoch": 9, "memory": 25074, "step": 48264} +{"lr": 0.0026214889037780207, "data_time": 0.0015340089797973634, "loss": 0.030427138321101665, "time": 0.7890354871749878, "epoch": 9, "memory": 25074, "step": 48364} +{"lr": 0.0026214889037780207, "data_time": 0.0020235300064086912, "loss": 0.036000782065093516, "time": 0.8084030628204346, "epoch": 9, "memory": 25074, "step": 48464} +{"lr": 0.0026214889037780207, "data_time": 0.0022765636444091798, "loss": 0.021300629805773498, "time": 0.7924983978271485, "epoch": 9, "memory": 25074, "step": 48564} +{"lr": 0.0026214889037780207, "data_time": 0.0025313138961791993, "loss": 0.04439207892864942, "time": 0.8228082180023193, "epoch": 9, "memory": 25074, "step": 48664} +{"lr": 0.0026214889037780207, "data_time": 0.002218770980834961, "loss": 0.02781707257963717, "time": 0.7708287239074707, "epoch": 9, "memory": 25074, "step": 48764} +{"lr": 0.0026214889037780207, "data_time": 0.001650547981262207, "loss": 0.02063184455037117, "time": 0.8667472839355469, "epoch": 9, "memory": 25074, "step": 48864} +{"lr": 0.0026214889037780207, "data_time": 0.0016929388046264648, "loss": 0.024441056232899428, "time": 0.8383315563201904, "epoch": 9, "memory": 25074, "step": 48964} +{"lr": 0.0026214889037780207, "data_time": 0.0014557838439941406, "loss": 0.02691945619881153, "time": 0.8165060758590699, "epoch": 9, "memory": 25074, "step": 49064} +{"lr": 0.0026214889037780207, "data_time": 0.0017290592193603515, "loss": 0.0241798035800457, "time": 0.8456665277481079, "epoch": 9, "memory": 25074, "step": 49164} +{"lr": 0.0026214889037780207, "data_time": 0.0014475584030151367, "loss": 0.03516687611117959, "time": 0.8111721277236938, "epoch": 9, "memory": 25074, "step": 49264} +{"lr": 0.0026214889037780207, "data_time": 0.0018590927124023438, "loss": 0.24516677521169186, "time": 0.8151656150817871, "epoch": 9, "memory": 25074, "step": 49364} +{"lr": 0.0026214889037780207, "data_time": 0.001399683952331543, "loss": 0.028192812763154506, "time": 0.8275673389434814, "epoch": 9, "memory": 25074, "step": 49464} +{"lr": 0.0026214889037780207, "data_time": 0.001490330696105957, "loss": 0.023393982416018845, "time": 0.7727564096450805, "epoch": 9, "memory": 25074, "step": 49564} +{"lr": 0.0026214889037780207, "data_time": 0.0016549110412597656, "loss": 0.02338564246892929, "time": 0.8695456981658936, "epoch": 9, "memory": 25074, "step": 49664} +{"lr": 0.0026214889037780207, "data_time": 0.001441788673400879, "loss": 0.015078999381512404, "time": 0.802464509010315, "epoch": 9, "memory": 25074, "step": 49764} +{"lr": 0.0026214889037780207, "data_time": 0.0016698122024536132, "loss": 0.03197599779814482, "time": 0.8282928705215454, "epoch": 9, "memory": 25074, "step": 49864} +{"lr": 0.0026214889037780207, "data_time": 0.0022622823715209963, "loss": 0.013944790884852409, "time": 0.7984851837158203, "epoch": 9, "memory": 25074, "step": 49964} +{"lr": 0.0026214889037780207, "data_time": 0.0019774436950683594, "loss": 0.03877650052309036, "time": 0.7803465366363526, "epoch": 9, "memory": 25074, "step": 50064} +{"lr": 0.0026214889037780207, "data_time": 0.0015271425247192383, "loss": 0.026444716192781926, "time": 0.7932117938995361, "epoch": 9, "memory": 25074, "step": 50164} +{"lr": 0.0026214889037780207, "data_time": 0.001999688148498535, "loss": 0.02609200058504939, "time": 0.8093926191329956, "epoch": 9, "memory": 25074, "step": 50264} +{"lr": 0.0026214889037780207, "data_time": 0.0024790525436401366, "loss": 0.022045370377600192, "time": 0.8204033613204956, "epoch": 9, "memory": 25074, "step": 50364} +{"lr": 0.0026214889037780207, "data_time": 0.0019163131713867188, "loss": 0.030200400948524476, "time": 0.7742255210876465, "epoch": 9, "memory": 25074, "step": 50464} +{"lr": 0.0026214889037780207, "data_time": 0.002627658843994141, "loss": 0.019695536652579903, "time": 0.832004976272583, "epoch": 9, "memory": 25074, "step": 50564} +{"lr": 0.0026214889037780207, "data_time": 0.0019025802612304688, "loss": 0.024905043840408325, "time": 0.8002906799316406, "epoch": 9, "memory": 25074, "step": 50664} +{"lr": 0.0026214889037780207, "data_time": 0.0015072107315063476, "loss": 0.02924757357686758, "time": 0.8218582391738891, "epoch": 9, "memory": 25074, "step": 50764} +{"lr": 0.0026214889037780207, "data_time": 0.0015533685684204102, "loss": 0.0230212087277323, "time": 0.7957111835479737, "epoch": 9, "memory": 25074, "step": 50864} +{"lr": 0.0026214889037780207, "data_time": 0.0018848896026611329, "loss": 0.022353645507246257, "time": 0.8086437463760376, "epoch": 9, "memory": 25074, "step": 50964} +{"lr": 0.0026214889037780207, "data_time": 0.0018225669860839843, "loss": 0.018633402604609727, "time": 0.7607362508773804, "epoch": 9, "memory": 25074, "step": 51064} +{"lr": 0.0026214889037780207, "data_time": 0.0019716978073120116, "loss": 0.03313821600750089, "time": 0.840787124633789, "epoch": 9, "memory": 25074, "step": 51164} +{"lr": 0.0026214889037780207, "data_time": 0.0016269683837890625, "loss": 0.014874533098191024, "time": 0.8427674531936645, "epoch": 9, "memory": 25074, "step": 51264} +{"lr": 0.0026214889037780207, "data_time": 0.002698802947998047, "loss": 0.028587420424446463, "time": 0.8011973857879638, "epoch": 9, "memory": 25074, "step": 51364} +{"lr": 0.0026214889037780207, "data_time": 0.002422475814819336, "loss": 0.02664147736504674, "time": 0.8069389343261719, "epoch": 9, "memory": 25074, "step": 51464} +{"lr": 0.0026214889037780207, "data_time": 0.0018821954727172852, "loss": 0.020317592099308966, "time": 0.8240701913833618, "epoch": 9, "memory": 25074, "step": 51564} +{"lr": 0.0026214889037780207, "data_time": 0.0019106626510620116, "loss": 0.023197530955076217, "time": 0.8193970680236816, "epoch": 9, "memory": 25074, "step": 51664} +{"lr": 0.0026214889037780207, "data_time": 0.0021815061569213866, "loss": 0.014408213505521416, "time": 0.8479761600494384, "epoch": 9, "memory": 25074, "step": 51764} +{"accuracy/top1": 99.33084869384766, "data_time": 0.23750793232637293, "time": 0.6247797713560217, "step": 9} +{"lr": 0.002317086757755261, "data_time": 0.0020456314086914062, "loss": 0.017532415268942714, "time": 0.8207318782806396, "epoch": 10, "memory": 25074, "step": 51922} +{"lr": 0.002317086757755261, "data_time": 0.001619434356689453, "loss": 0.01765131545253098, "time": 0.7812523126602173, "epoch": 10, "memory": 25074, "step": 52022} +{"lr": 0.002317086757755261, "data_time": 0.0024901628494262695, "loss": 0.011763214785605668, "time": 0.8035857677459717, "epoch": 10, "memory": 25074, "step": 52122} +{"lr": 0.002317086757755261, "data_time": 0.0017036914825439454, "loss": 0.012079137982800603, "time": 0.7796544313430787, "epoch": 10, "memory": 25074, "step": 52222} +{"lr": 0.002317086757755261, "data_time": 0.0016983509063720702, "loss": 0.013122422154992819, "time": 0.8294822454452515, "epoch": 10, "memory": 25074, "step": 52322} +{"lr": 0.002317086757755261, "data_time": 0.001611948013305664, "loss": 0.019555166084319352, "time": 0.8364537000656128, "epoch": 10, "memory": 25074, "step": 52422} +{"lr": 0.002317086757755261, "data_time": 0.0026428699493408203, "loss": 0.011063139233738185, "time": 0.8343900680541992, "epoch": 10, "memory": 25074, "step": 52522} +{"lr": 0.002317086757755261, "data_time": 0.0020388364791870117, "loss": 0.018025414459407328, "time": 0.7451565027236938, "epoch": 10, "memory": 25074, "step": 52622} +{"lr": 0.002317086757755261, "data_time": 0.0018421173095703124, "loss": 0.018842556141316892, "time": 0.7868381261825561, "epoch": 10, "memory": 25074, "step": 52722} +{"lr": 0.002317086757755261, "data_time": 0.0026170492172241213, "loss": 0.020071753207594155, "time": 0.755088472366333, "epoch": 10, "memory": 25074, "step": 52822} +{"lr": 0.002317086757755261, "data_time": 0.0021312713623046877, "loss": 0.014937166590243578, "time": 0.7882421255111695, "epoch": 10, "memory": 25074, "step": 52922} +{"lr": 0.002317086757755261, "data_time": 0.0020072460174560547, "loss": 0.015283338679000735, "time": 0.8208979368209839, "epoch": 10, "memory": 25074, "step": 53022} +{"lr": 0.002317086757755261, "data_time": 0.0017373323440551757, "loss": 0.013104923209175468, "time": 0.7905618190765381, "epoch": 10, "memory": 25074, "step": 53122} +{"lr": 0.002317086757755261, "data_time": 0.0017949342727661133, "loss": 0.013780922582373022, "time": 0.7907957077026367, "epoch": 10, "memory": 25074, "step": 53222} +{"lr": 0.002317086757755261, "data_time": 0.0022233247756958006, "loss": 0.018247692426666618, "time": 0.7822386264801026, "epoch": 10, "memory": 25074, "step": 53322} +{"lr": 0.002317086757755261, "data_time": 0.002651524543762207, "loss": 0.014068604446947575, "time": 0.8082148551940918, "epoch": 10, "memory": 25074, "step": 53422} +{"lr": 0.002317086757755261, "data_time": 0.0020285606384277343, "loss": 0.016047554556280375, "time": 0.8054810047149659, "epoch": 10, "memory": 25074, "step": 53522} +{"lr": 0.002317086757755261, "data_time": 0.0020494461059570312, "loss": 0.009224567469209433, "time": 0.8373507022857666, "epoch": 10, "memory": 25074, "step": 53622} +{"lr": 0.002317086757755261, "data_time": 0.0017382383346557617, "loss": 0.015223241131752729, "time": 0.8445773839950561, "epoch": 10, "memory": 25074, "step": 53722} +{"lr": 0.002317086757755261, "data_time": 0.0018725633621215821, "loss": 0.011556544108316302, "time": 0.8119404315948486, "epoch": 10, "memory": 25074, "step": 53822} +{"lr": 0.002317086757755261, "data_time": 0.0021619319915771483, "loss": 0.012437142711132766, "time": 0.8511190176010132, "epoch": 10, "memory": 25074, "step": 53922} +{"lr": 0.002317086757755261, "data_time": 0.0018029212951660156, "loss": 0.011740866350010037, "time": 0.8316761970520019, "epoch": 10, "memory": 25074, "step": 54022} +{"lr": 0.002317086757755261, "data_time": 0.0025862932205200197, "loss": 0.024602488055825232, "time": 0.8639458656311035, "epoch": 10, "memory": 25074, "step": 54122} +{"lr": 0.002317086757755261, "data_time": 0.0022039890289306642, "loss": 0.02148634297773242, "time": 0.7997440338134766, "epoch": 10, "memory": 25074, "step": 54222} +{"lr": 0.002317086757755261, "data_time": 0.0026303529739379883, "loss": 0.0185382345225662, "time": 0.8154025316238404, "epoch": 10, "memory": 25074, "step": 54322} +{"lr": 0.002317086757755261, "data_time": 0.00218808650970459, "loss": 0.6881393432617188, "time": 0.8355687379837036, "epoch": 10, "memory": 25074, "step": 54422} +{"lr": 0.002317086757755261, "data_time": 0.0020766973495483397, "loss": 0.6841361999511719, "time": 0.8412147521972656, "epoch": 10, "memory": 25074, "step": 54522} +{"lr": 0.002317086757755261, "data_time": 0.0022381305694580077, "loss": 0.686181640625, "time": 0.8091365098953247, "epoch": 10, "memory": 25074, "step": 54622} +{"lr": 0.002317086757755261, "data_time": 0.0019045829772949218, "loss": 0.6752777099609375, "time": 0.8718130111694335, "epoch": 10, "memory": 25074, "step": 54722} +{"lr": 0.002317086757755261, "data_time": 0.0018952131271362306, "loss": 0.5613418817520142, "time": 0.8286389827728271, "epoch": 10, "memory": 25074, "step": 54822} +{"lr": 0.002317086757755261, "data_time": 0.002397894859313965, "loss": 0.49733060896396636, "time": 0.8195470571517944, "epoch": 10, "memory": 25074, "step": 54922} +{"lr": 0.002317086757755261, "data_time": 0.0015130758285522461, "loss": 0.4726357638835907, "time": 0.8234517812728882, "epoch": 10, "memory": 25074, "step": 55022} +{"lr": 0.002317086757755261, "data_time": 0.001624751091003418, "loss": 0.422371917963028, "time": 0.7901685953140258, "epoch": 10, "memory": 25074, "step": 55122} +{"lr": 0.002317086757755261, "data_time": 0.0022609710693359377, "loss": 0.4255555093288422, "time": 0.7914380073547364, "epoch": 10, "memory": 25074, "step": 55222} +{"lr": 0.002317086757755261, "data_time": 0.001772165298461914, "loss": 0.3666388154029846, "time": 0.8414067983627319, "epoch": 10, "memory": 25074, "step": 55322} +{"lr": 0.002317086757755261, "data_time": 0.0015523910522460937, "loss": 0.3317451149225235, "time": 0.7918429613113404, "epoch": 10, "memory": 25074, "step": 55422} +{"lr": 0.002317086757755261, "data_time": 0.002152752876281738, "loss": 0.27832788676023484, "time": 0.8647343873977661, "epoch": 10, "memory": 25074, "step": 55522} +{"lr": 0.002317086757755261, "data_time": 0.002333998680114746, "loss": 0.2840075805783272, "time": 0.7786744832992554, "epoch": 10, "memory": 25074, "step": 55622} +{"lr": 0.002317086757755261, "data_time": 0.0023336410522460938, "loss": 0.2699961230158806, "time": 0.8274954557418823, "epoch": 10, "memory": 25074, "step": 55722} +{"lr": 0.002317086757755261, "data_time": 0.0024749279022216798, "loss": 0.20767785906791686, "time": 0.8666459083557129, "epoch": 10, "memory": 25074, "step": 55822} +{"lr": 0.002317086757755261, "data_time": 0.0022413015365600588, "loss": 0.19525945335626602, "time": 0.8322144031524659, "epoch": 10, "memory": 25074, "step": 55922} +{"lr": 0.002317086757755261, "data_time": 0.0020354032516479493, "loss": 0.16501691937446594, "time": 0.8302124261856079, "epoch": 10, "memory": 25074, "step": 56022} +{"lr": 0.002317086757755261, "data_time": 0.002210378646850586, "loss": 0.11037591993808746, "time": 0.7660328149795532, "epoch": 10, "memory": 25074, "step": 56122} +{"lr": 0.002317086757755261, "data_time": 0.0023299217224121093, "loss": 0.08365992046892642, "time": 0.9144795179367066, "epoch": 10, "memory": 25074, "step": 56222} +{"lr": 0.002317086757755261, "data_time": 0.002490592002868652, "loss": 0.08645744733512402, "time": 0.8665079116821289, "epoch": 10, "memory": 25074, "step": 56322} +{"lr": 0.002317086757755261, "data_time": 0.0022429943084716795, "loss": 0.0692336829379201, "time": 0.9035349369049073, "epoch": 10, "memory": 25074, "step": 56422} +{"lr": 0.002317086757755261, "data_time": 0.0014755725860595703, "loss": 0.04233815483748913, "time": 0.8001418828964233, "epoch": 10, "memory": 25074, "step": 56522} +{"lr": 0.002317086757755261, "data_time": 0.0019971132278442383, "loss": 0.04625342469662428, "time": 0.8565793037414551, "epoch": 10, "memory": 25074, "step": 56622} +{"lr": 0.002317086757755261, "data_time": 0.0017195463180541993, "loss": 0.03124313671141863, "time": 0.8225848913192749, "epoch": 10, "memory": 25074, "step": 56722} +{"lr": 0.002317086757755261, "data_time": 0.0026178836822509767, "loss": 0.04096714742481709, "time": 0.8223660469055176, "epoch": 10, "memory": 25074, "step": 56822} +{"lr": 0.002317086757755261, "data_time": 0.0019570350646972655, "loss": 0.04236628897488117, "time": 0.7936403274536132, "epoch": 10, "memory": 25074, "step": 56922} +{"lr": 0.002317086757755261, "data_time": 0.0025371313095092773, "loss": 0.025017997808754446, "time": 0.7964055061340332, "epoch": 10, "memory": 25074, "step": 57022} +{"lr": 0.002317086757755261, "data_time": 0.002306365966796875, "loss": 0.05144036617130041, "time": 0.8108157157897949, "epoch": 10, "memory": 25074, "step": 57122} +{"lr": 0.002317086757755261, "data_time": 0.002100038528442383, "loss": 0.02744342922233045, "time": 0.8977809429168702, "epoch": 10, "memory": 25074, "step": 57222} +{"lr": 0.002317086757755261, "data_time": 0.0023747682571411133, "loss": 0.03131036190316081, "time": 0.8460140705108643, "epoch": 10, "memory": 25074, "step": 57322} +{"lr": 0.002317086757755261, "data_time": 0.002607393264770508, "loss": 0.02022187914699316, "time": 0.8046281814575196, "epoch": 10, "memory": 25074, "step": 57422} +{"lr": 0.002317086757755261, "data_time": 0.0024229049682617187, "loss": 0.020474710408598183, "time": 0.8252543687820435, "epoch": 10, "memory": 25074, "step": 57522} +{"accuracy/top1": 100.0, "data_time": 0.22276956894818475, "time": 0.6108220044304343, "step": 10} +{"lr": 0.0020050000000000003, "data_time": 0.002590775489807129, "loss": 0.02065365444868803, "time": 0.7896236181259155, "epoch": 11, "memory": 25074, "step": 57680} +{"lr": 0.0020050000000000003, "data_time": 0.002123236656188965, "loss": 0.0183746303897351, "time": 0.8085314273834229, "epoch": 11, "memory": 25074, "step": 57780} +{"lr": 0.0020050000000000003, "data_time": 0.0019198417663574218, "loss": 0.02596295215189457, "time": 0.7891613721847535, "epoch": 11, "memory": 25074, "step": 57880} +{"lr": 0.0020050000000000003, "data_time": 0.0018343925476074219, "loss": 0.013839405495673418, "time": 0.8113912582397461, "epoch": 11, "memory": 25074, "step": 57980} +{"lr": 0.0020050000000000003, "data_time": 0.0016790390014648437, "loss": 0.017425126023590566, "time": 0.8286957740783691, "epoch": 11, "memory": 25074, "step": 58080} +{"lr": 0.0020050000000000003, "data_time": 0.0019089937210083007, "loss": 0.024284788640215993, "time": 0.7943507194519043, "epoch": 11, "memory": 25074, "step": 58180} +{"lr": 0.0020050000000000003, "data_time": 0.0028844833374023437, "loss": 0.020067624375224115, "time": 0.7873490810394287, "epoch": 11, "memory": 25074, "step": 58280} +{"lr": 0.0020050000000000003, "data_time": 0.0029201507568359375, "loss": 0.01442998107522726, "time": 0.8720103740692139, "epoch": 11, "memory": 25074, "step": 58380} +{"lr": 0.0020050000000000003, "data_time": 0.0024564504623413087, "loss": 0.018959759408608078, "time": 0.8664995908737183, "epoch": 11, "memory": 25074, "step": 58480} +{"lr": 0.0020050000000000003, "data_time": 0.0020102500915527345, "loss": 0.018543228320777417, "time": 0.7973664999008179, "epoch": 11, "memory": 25074, "step": 58580} +{"lr": 0.0020050000000000003, "data_time": 0.0018429994583129884, "loss": 0.05799570456147194, "time": 0.7817375898361206, "epoch": 11, "memory": 25074, "step": 58680} +{"lr": 0.0020050000000000003, "data_time": 0.0016614437103271485, "loss": 0.012952214712277055, "time": 0.8419855833053589, "epoch": 11, "memory": 25074, "step": 58780} +{"lr": 0.0020050000000000003, "data_time": 0.002156186103820801, "loss": 0.025047481711953878, "time": 0.7899405241012574, "epoch": 11, "memory": 25074, "step": 58880} +{"lr": 0.0020050000000000003, "data_time": 0.0021374702453613283, "loss": 0.019668174767866732, "time": 0.845258641242981, "epoch": 11, "memory": 25074, "step": 58980} +{"lr": 0.0020050000000000003, "data_time": 0.002274632453918457, "loss": 0.009758525481447577, "time": 0.7986643552780152, "epoch": 11, "memory": 25074, "step": 59080} +{"lr": 0.0020050000000000003, "data_time": 0.0019394159317016602, "loss": 0.011583693930879236, "time": 0.8287673711776733, "epoch": 11, "memory": 25074, "step": 59180} +{"lr": 0.0020050000000000003, "data_time": 0.0020122766494750977, "loss": 0.01529764630831778, "time": 0.8637459754943848, "epoch": 11, "memory": 25074, "step": 59280} +{"lr": 0.0020050000000000003, "data_time": 0.0021901369094848634, "loss": 0.01475684279575944, "time": 0.8098446607589722, "epoch": 11, "memory": 25074, "step": 59380} +{"lr": 0.0020050000000000003, "data_time": 0.002501654624938965, "loss": 0.015876112086698414, "time": 0.8413092613220214, "epoch": 11, "memory": 25074, "step": 59480} +{"lr": 0.0020050000000000003, "data_time": 0.00217742919921875, "loss": 0.5657811641693116, "time": 0.7998675107955933, "epoch": 11, "memory": 25074, "step": 59580} +{"lr": 0.0020050000000000003, "data_time": 0.0019238948822021484, "loss": 0.4891375541687012, "time": 0.8692360639572143, "epoch": 11, "memory": 25074, "step": 59680} +{"lr": 0.0020050000000000003, "data_time": 0.001653289794921875, "loss": 0.43288591504096985, "time": 0.8257172346115113, "epoch": 11, "memory": 25074, "step": 59780} +{"lr": 0.0020050000000000003, "data_time": 0.0016234159469604493, "loss": 0.392276793718338, "time": 0.8491599082946777, "epoch": 11, "memory": 25074, "step": 59880} +{"lr": 0.0020050000000000003, "data_time": 0.002200007438659668, "loss": 0.3798608660697937, "time": 0.8266980171203613, "epoch": 11, "memory": 25074, "step": 59980} +{"lr": 0.0020050000000000003, "data_time": 0.0021578073501586914, "loss": 0.3877251446247101, "time": 0.7977796792984009, "epoch": 11, "memory": 25074, "step": 60080} +{"lr": 0.0020050000000000003, "data_time": 0.0023580074310302736, "loss": 0.30527042746543886, "time": 0.870452618598938, "epoch": 11, "memory": 25074, "step": 60180} +{"lr": 0.0020050000000000003, "data_time": 0.0020999908447265625, "loss": 0.27039543241262437, "time": 0.7851626396179199, "epoch": 11, "memory": 25074, "step": 60280} +{"lr": 0.0020050000000000003, "data_time": 0.002391409873962402, "loss": 0.25152166187763214, "time": 0.8447658777236938, "epoch": 11, "memory": 25074, "step": 60380} +{"lr": 0.0020050000000000003, "data_time": 0.0015321493148803711, "loss": 0.13193194791674615, "time": 0.8256754159927369, "epoch": 11, "memory": 25074, "step": 60480} +{"lr": 0.0020050000000000003, "data_time": 0.0015456199645996094, "loss": 0.06172098498791456, "time": 0.7993576765060425, "epoch": 11, "memory": 25074, "step": 60580} +{"lr": 0.0020050000000000003, "data_time": 0.0015208959579467774, "loss": 0.028707681596279143, "time": 0.814827823638916, "epoch": 11, "memory": 25074, "step": 60680} +{"lr": 0.0020050000000000003, "data_time": 0.0014363527297973633, "loss": 0.030554246716201307, "time": 0.73760826587677, "epoch": 11, "memory": 25074, "step": 60780} +{"lr": 0.0020050000000000003, "data_time": 0.0014954328536987305, "loss": 0.020832169940695165, "time": 0.8498701333999634, "epoch": 11, "memory": 25074, "step": 60880} +{"lr": 0.0020050000000000003, "data_time": 0.002184128761291504, "loss": 0.029661014582961796, "time": 0.8704250812530517, "epoch": 11, "memory": 25074, "step": 60980} +{"lr": 0.0020050000000000003, "data_time": 0.0022843599319458006, "loss": 0.01465114252641797, "time": 0.8545256853103638, "epoch": 11, "memory": 25074, "step": 61080} +{"lr": 0.0020050000000000003, "data_time": 0.0019321918487548828, "loss": 0.023434247262775897, "time": 0.8331084251403809, "epoch": 11, "memory": 25074, "step": 61180} +{"lr": 0.0020050000000000003, "data_time": 0.0018091201782226562, "loss": 0.021718163136392833, "time": 0.8547056674957275, "epoch": 11, "memory": 25074, "step": 61280} +{"lr": 0.0020050000000000003, "data_time": 0.0014871597290039063, "loss": 0.019336781091988088, "time": 0.7947772979736328, "epoch": 11, "memory": 25074, "step": 61380} +{"lr": 0.0020050000000000003, "data_time": 0.001659226417541504, "loss": 0.01791371051222086, "time": 0.864168906211853, "epoch": 11, "memory": 25074, "step": 61480} +{"lr": 0.0020050000000000003, "data_time": 0.0023895025253295897, "loss": 0.018238531658425927, "time": 0.8782007694244385, "epoch": 11, "memory": 25074, "step": 61580} +{"lr": 0.0020050000000000003, "data_time": 0.0014923095703125, "loss": 0.015828208066523075, "time": 0.8442202806472778, "epoch": 11, "memory": 25074, "step": 61680} +{"lr": 0.0020050000000000003, "data_time": 0.0020627498626708983, "loss": 0.03517089206725359, "time": 0.8642852306365967, "epoch": 11, "memory": 25074, "step": 61780} +{"lr": 0.0020050000000000003, "data_time": 0.0022553205490112305, "loss": 0.017217851197347044, "time": 0.9275934219360351, "epoch": 11, "memory": 25074, "step": 61880} +{"lr": 0.0020050000000000003, "data_time": 0.002085423469543457, "loss": 0.013848322676494718, "time": 0.8090350389480591, "epoch": 11, "memory": 25074, "step": 61980} +{"lr": 0.0020050000000000003, "data_time": 0.001747465133666992, "loss": 0.012819849653169513, "time": 0.8843717098236084, "epoch": 11, "memory": 25074, "step": 62080} +{"lr": 0.0020050000000000003, "data_time": 0.0017876863479614259, "loss": 0.01733230291865766, "time": 0.8388864278793335, "epoch": 11, "memory": 25074, "step": 62180} +{"lr": 0.0020050000000000003, "data_time": 0.0014911413192749024, "loss": 0.01523120547644794, "time": 0.8495623588562011, "epoch": 11, "memory": 25074, "step": 62280} +{"lr": 0.0020050000000000003, "data_time": 0.0026740789413452148, "loss": 0.013444093475118279, "time": 0.9048556566238404, "epoch": 11, "memory": 25074, "step": 62380} +{"lr": 0.0020050000000000003, "data_time": 0.0018796205520629882, "loss": 0.012744836742058396, "time": 0.8099363803863525, "epoch": 11, "memory": 25074, "step": 62480} +{"lr": 0.0020050000000000003, "data_time": 0.0014407873153686524, "loss": 0.01894613322801888, "time": 0.876837968826294, "epoch": 11, "memory": 25074, "step": 62580} +{"lr": 0.0020050000000000003, "data_time": 0.002218937873840332, "loss": 0.009766071895137429, "time": 0.8186978340148926, "epoch": 11, "memory": 25074, "step": 62680} +{"lr": 0.0020050000000000003, "data_time": 0.002142596244812012, "loss": 0.016696813562884927, "time": 0.8417745590209961, "epoch": 11, "memory": 25074, "step": 62780} +{"lr": 0.0020050000000000003, "data_time": 0.001732778549194336, "loss": 0.010334298480302095, "time": 0.8277750492095948, "epoch": 11, "memory": 25074, "step": 62880} +{"lr": 0.0020050000000000003, "data_time": 0.001989459991455078, "loss": 0.00968954348936677, "time": 0.8843014717102051, "epoch": 11, "memory": 25074, "step": 62980} +{"lr": 0.0020050000000000003, "data_time": 0.0019947052001953124, "loss": 0.010796508565545083, "time": 0.798948335647583, "epoch": 11, "memory": 25074, "step": 63080} +{"lr": 0.0020050000000000003, "data_time": 0.0016904354095458984, "loss": 0.013044635811820627, "time": 0.8644941329956055, "epoch": 11, "memory": 25074, "step": 63180} +{"lr": 0.0020050000000000003, "data_time": 0.0014651298522949218, "loss": 0.010796272382140159, "time": 0.7989150047302246, "epoch": 11, "memory": 25074, "step": 63280} +{"accuracy/top1": 100.0, "data_time": 0.23673754579880657, "time": 0.6232916888068704, "step": 11} +{"lr": 0.0016929132422447402, "data_time": 0.0016835689544677734, "loss": 0.012621474638581275, "time": 0.8847478151321411, "epoch": 12, "memory": 25074, "step": 63438} +{"lr": 0.0016929132422447402, "data_time": 0.002158713340759277, "loss": 0.012400662759318948, "time": 0.8630192279815674, "epoch": 12, "memory": 25074, "step": 63538} +{"lr": 0.0016929132422447402, "data_time": 0.0022430658340454102, "loss": 0.006929849926382303, "time": 0.8485779285430908, "epoch": 12, "memory": 25074, "step": 63638} +{"lr": 0.0016929132422447402, "data_time": 0.0014281272888183594, "loss": 0.009629213670268654, "time": 0.8658886432647706, "epoch": 12, "memory": 25074, "step": 63738} +{"lr": 0.0016929132422447402, "data_time": 0.0018830299377441406, "loss": 0.005235561076551676, "time": 0.8525608539581299, "epoch": 12, "memory": 25074, "step": 63838} +{"lr": 0.0016929132422447402, "data_time": 0.001458120346069336, "loss": 0.5257198572158813, "time": 0.8559411525726318, "epoch": 12, "memory": 25074, "step": 63938} +{"lr": 0.0016929132422447402, "data_time": 0.0024147987365722655, "loss": 0.01740764332935214, "time": 0.8217525959014893, "epoch": 12, "memory": 25074, "step": 64038} +{"lr": 0.0016929132422447402, "data_time": 0.0018601417541503906, "loss": 0.018427724577486516, "time": 0.8378716230392456, "epoch": 12, "memory": 25074, "step": 64138} +{"lr": 0.0016929132422447402, "data_time": 0.0016304492950439454, "loss": 0.015688306698575617, "time": 0.8500861167907715, "epoch": 12, "memory": 25074, "step": 64238} +{"lr": 0.0016929132422447402, "data_time": 0.0016458988189697265, "loss": 0.01742241098545492, "time": 0.8576305150985718, "epoch": 12, "memory": 25074, "step": 64338} +{"lr": 0.0016929132422447402, "data_time": 0.0018129348754882812, "loss": 0.00865716771222651, "time": 0.8440456390380859, "epoch": 12, "memory": 25074, "step": 64438} +{"lr": 0.0016929132422447402, "data_time": 0.0016907453536987305, "loss": 0.004777575191110372, "time": 0.8773534297943115, "epoch": 12, "memory": 25074, "step": 64538} +{"lr": 0.0016929132422447402, "data_time": 0.0016074895858764649, "loss": 0.012243752693757414, "time": 0.8194139003753662, "epoch": 12, "memory": 25074, "step": 64638} +{"lr": 0.0016929132422447402, "data_time": 0.0013826370239257812, "loss": 0.006700974376872182, "time": 0.8386459112167358, "epoch": 12, "memory": 25074, "step": 64738} +{"lr": 0.0016929132422447402, "data_time": 0.0017354011535644532, "loss": 0.012806080281734467, "time": 0.8877508878707886, "epoch": 12, "memory": 25074, "step": 64838} +{"lr": 0.0016929132422447402, "data_time": 0.0015664339065551759, "loss": 0.004690782725811004, "time": 0.7721720457077026, "epoch": 12, "memory": 25074, "step": 64938} +{"lr": 0.0016929132422447402, "data_time": 0.0017460107803344727, "loss": 0.009485487034544349, "time": 0.8704520225524902, "epoch": 12, "memory": 25074, "step": 65038} +{"lr": 0.0016929132422447402, "data_time": 0.0016896247863769532, "loss": 0.01166194025427103, "time": 0.8121005058288574, "epoch": 12, "memory": 25074, "step": 65138} +{"lr": 0.0016929132422447402, "data_time": 0.002256011962890625, "loss": 0.011341179488226771, "time": 0.8417533159255981, "epoch": 12, "memory": 25074, "step": 65238} +{"lr": 0.0016929132422447402, "data_time": 0.00157623291015625, "loss": 0.007958573894575239, "time": 0.8386371374130249, "epoch": 12, "memory": 25074, "step": 65338} +{"lr": 0.0016929132422447402, "data_time": 0.0014529705047607421, "loss": 0.013911510119214654, "time": 0.9013768196105957, "epoch": 12, "memory": 25074, "step": 65438} +{"lr": 0.0016929132422447402, "data_time": 0.0015138626098632813, "loss": 0.00801685922779143, "time": 0.8339585065841675, "epoch": 12, "memory": 25074, "step": 65538} +{"lr": 0.0016929132422447402, "data_time": 0.0014876127243041992, "loss": 0.010362574364989996, "time": 0.84767165184021, "epoch": 12, "memory": 25074, "step": 65638} +{"lr": 0.0016929132422447402, "data_time": 0.001378011703491211, "loss": 0.01023637312464416, "time": 0.8630808830261231, "epoch": 12, "memory": 25074, "step": 65738} +{"lr": 0.0016929132422447402, "data_time": 0.0019390106201171875, "loss": 0.004722489602863788, "time": 0.8443884134292603, "epoch": 12, "memory": 25074, "step": 65838} +{"lr": 0.0016929132422447402, "data_time": 0.0015073299407958984, "loss": 0.007929569622501732, "time": 0.8416267395019531, "epoch": 12, "memory": 25074, "step": 65938} +{"lr": 0.0016929132422447402, "data_time": 0.001519918441772461, "loss": 0.004670946905389428, "time": 0.8766562461853027, "epoch": 12, "memory": 25074, "step": 66038} +{"lr": 0.0016929132422447402, "data_time": 0.0016965150833129882, "loss": 0.6126410746946931, "time": 0.7927167415618896, "epoch": 12, "memory": 25074, "step": 66138} +{"lr": 0.0016929132422447402, "data_time": 0.0015101432800292969, "loss": 0.6205726623535156, "time": 0.7908918619155884, "epoch": 12, "memory": 25074, "step": 66238} +{"lr": 0.0016929132422447402, "data_time": 0.001480698585510254, "loss": 0.5530441761016845, "time": 0.8379903316497803, "epoch": 12, "memory": 25074, "step": 66338} +{"lr": 0.0016929132422447402, "data_time": 0.0015945911407470703, "loss": 0.44581436514854433, "time": 0.8860613107681274, "epoch": 12, "memory": 25074, "step": 66438} +{"lr": 0.0016929132422447402, "data_time": 0.0018938064575195312, "loss": 0.3750150799751282, "time": 2.01279821395874, "epoch": 12, "memory": 25074, "step": 66538} +{"lr": 0.0016929132422447402, "data_time": 0.0027787208557128905, "loss": 0.3052941858768463, "time": 0.8875884056091309, "epoch": 12, "memory": 25074, "step": 66638} +{"lr": 0.0016929132422447402, "data_time": 0.001857924461364746, "loss": 0.29005714058876036, "time": 0.839936900138855, "epoch": 12, "memory": 25074, "step": 66738} +{"lr": 0.0016929132422447402, "data_time": 0.0019493579864501953, "loss": 0.3041937619447708, "time": 0.8447667121887207, "epoch": 12, "memory": 25074, "step": 66838} +{"lr": 0.0016929132422447402, "data_time": 0.002667999267578125, "loss": 0.24104223847389222, "time": 0.8763674020767211, "epoch": 12, "memory": 25074, "step": 66938} +{"lr": 0.0016929132422447402, "data_time": 0.0022324323654174805, "loss": 0.26029926389455793, "time": 0.8523849487304688, "epoch": 12, "memory": 25074, "step": 67038} +{"lr": 0.0016929132422447402, "data_time": 0.0017183065414428712, "loss": 0.21992358714342117, "time": 0.8215179204940796, "epoch": 12, "memory": 25074, "step": 67138} +{"lr": 0.0016929132422447402, "data_time": 0.0016553401947021484, "loss": 0.19771873950958252, "time": 0.8601011753082275, "epoch": 12, "memory": 25074, "step": 67238} +{"lr": 0.0016929132422447402, "data_time": 0.0017284154891967773, "loss": 0.20580636113882064, "time": 0.8334522485733032, "epoch": 12, "memory": 25074, "step": 67338} +{"lr": 0.0016929132422447402, "data_time": 0.0014740228652954102, "loss": 0.17389717549085618, "time": 0.8400430917739868, "epoch": 12, "memory": 25074, "step": 67438} +{"lr": 0.0016929132422447402, "data_time": 0.0016913890838623046, "loss": 0.1664424017071724, "time": 0.8414290904998779, "epoch": 12, "memory": 25074, "step": 67538} +{"lr": 0.0016929132422447402, "data_time": 0.001644730567932129, "loss": 0.13310553431510924, "time": 0.8928435325622559, "epoch": 12, "memory": 25074, "step": 67638} +{"lr": 0.0016929132422447402, "data_time": 0.0023133277893066405, "loss": 0.07253217957913875, "time": 0.8251712322235107, "epoch": 12, "memory": 25074, "step": 67738} +{"lr": 0.0016929132422447402, "data_time": 0.001665210723876953, "loss": 0.05623281970620155, "time": 0.8944040536880493, "epoch": 12, "memory": 25074, "step": 67838} +{"lr": 0.0016929132422447402, "data_time": 0.002199721336364746, "loss": 0.048439373075962064, "time": 0.9268851518630982, "epoch": 12, "memory": 25074, "step": 67938} +{"lr": 0.0016929132422447402, "data_time": 0.0018225431442260743, "loss": 0.026531974226236342, "time": 0.8406747579574585, "epoch": 12, "memory": 25074, "step": 68038} +{"lr": 0.0016929132422447402, "data_time": 0.0020012617111206054, "loss": 0.0358680609613657, "time": 0.8708724737167358, "epoch": 12, "memory": 25074, "step": 68138} +{"lr": 0.0016929132422447402, "data_time": 0.0021620750427246093, "loss": 0.03205973412841558, "time": 0.8374172449111938, "epoch": 12, "memory": 25074, "step": 68238} +{"lr": 0.0016929132422447402, "data_time": 0.0027259349822998046, "loss": 0.03224460240453482, "time": 0.8043625354766846, "epoch": 12, "memory": 25074, "step": 68338} +{"lr": 0.0016929132422447402, "data_time": 0.0020184040069580076, "loss": 0.026007355935871602, "time": 0.8549070835113526, "epoch": 12, "memory": 25074, "step": 68438} +{"lr": 0.0016929132422447402, "data_time": 0.001947927474975586, "loss": 0.019273914955556394, "time": 0.8272217035293579, "epoch": 12, "memory": 25074, "step": 68538} +{"lr": 0.0016929132422447402, "data_time": 0.002057814598083496, "loss": 0.018786594970151782, "time": 0.9094939231872559, "epoch": 12, "memory": 25074, "step": 68638} +{"lr": 0.0016929132422447402, "data_time": 0.0022021770477294923, "loss": 0.027599555673077703, "time": 0.7877516508102417, "epoch": 12, "memory": 25074, "step": 68738} +{"lr": 0.0016929132422447402, "data_time": 0.002342677116394043, "loss": 0.019824191089719535, "time": 0.8667730093002319, "epoch": 12, "memory": 25074, "step": 68838} +{"lr": 0.0016929132422447402, "data_time": 0.0029083013534545897, "loss": 0.02208682969212532, "time": 0.8528003215789794, "epoch": 12, "memory": 25074, "step": 68938} +{"lr": 0.0016929132422447402, "data_time": 0.0021807193756103516, "loss": 0.021340499771758913, "time": 0.8658803462982178, "epoch": 12, "memory": 25074, "step": 69038} +{"accuracy/top1": 100.0, "data_time": 0.239563689512365, "time": 0.6475922079647288, "step": 12} +{"lr": 0.0013885110962219803, "data_time": 0.0021898031234741213, "loss": 0.015078532509505749, "time": 0.8153757095336914, "epoch": 13, "memory": 25074, "step": 69196} +{"lr": 0.0013885110962219803, "data_time": 0.0022696256637573242, "loss": 0.012806871673092246, "time": 0.8093753099441529, "epoch": 13, "memory": 25074, "step": 69296} +{"lr": 0.0013885110962219803, "data_time": 0.002752876281738281, "loss": 0.01458560605533421, "time": 0.8431886196136474, "epoch": 13, "memory": 25074, "step": 69396} +{"lr": 0.0013885110962219803, "data_time": 0.0020204782485961914, "loss": 0.018982653878629206, "time": 0.8766860961914062, "epoch": 13, "memory": 25074, "step": 69496} +{"lr": 0.0013885110962219803, "data_time": 0.0024851322174072265, "loss": 0.027763380389660596, "time": 0.8396484375, "epoch": 13, "memory": 25074, "step": 69596} +{"lr": 0.0013885110962219803, "data_time": 0.001862788200378418, "loss": 0.01050863009877503, "time": 0.8230806827545166, "epoch": 13, "memory": 25074, "step": 69696} +{"lr": 0.0013885110962219803, "data_time": 0.0017082929611206056, "loss": 0.00995070575736463, "time": 0.837144422531128, "epoch": 13, "memory": 25074, "step": 69796} +{"lr": 0.0013885110962219803, "data_time": 0.0021940231323242187, "loss": 0.008433992881327868, "time": 0.8563901424407959, "epoch": 13, "memory": 25074, "step": 69896} +{"lr": 0.0013885110962219803, "data_time": 0.002205348014831543, "loss": 0.012554180575534701, "time": 0.9230027198791504, "epoch": 13, "memory": 25074, "step": 69996} +{"lr": 0.0013885110962219803, "data_time": 0.002058887481689453, "loss": 0.006387828104197979, "time": 0.8305997371673584, "epoch": 13, "memory": 25074, "step": 70096} +{"lr": 0.0013885110962219803, "data_time": 0.0020680665969848634, "loss": 0.010489308135583996, "time": 0.8507598161697387, "epoch": 13, "memory": 25074, "step": 70196} +{"lr": 0.0013885110962219803, "data_time": 0.0019878149032592773, "loss": 0.017410028772428632, "time": 0.8330593824386596, "epoch": 13, "memory": 25074, "step": 70296} +{"lr": 0.0013885110962219803, "data_time": 0.001918792724609375, "loss": 0.01754007306881249, "time": 0.8435761451721191, "epoch": 13, "memory": 25074, "step": 70396} +{"lr": 0.0013885110962219803, "data_time": 0.0019415855407714845, "loss": 0.011260713404044508, "time": 0.8395402669906616, "epoch": 13, "memory": 25074, "step": 70496} +{"lr": 0.0013885110962219803, "data_time": 0.002218461036682129, "loss": 0.023512054467573762, "time": 0.8369058609008789, "epoch": 13, "memory": 25074, "step": 70596} +{"lr": 0.0013885110962219803, "data_time": 0.002631545066833496, "loss": 0.006385938404127956, "time": 0.8792079210281372, "epoch": 13, "memory": 25074, "step": 70696} +{"lr": 0.0013885110962219803, "data_time": 0.002178788185119629, "loss": 0.01204793006181717, "time": 0.8195004463195801, "epoch": 13, "memory": 25074, "step": 70796} +{"lr": 0.0013885110962219803, "data_time": 0.0019639968872070313, "loss": 0.008308718353509903, "time": 0.8363892316818238, "epoch": 13, "memory": 25074, "step": 70896} +{"lr": 0.0013885110962219803, "data_time": 0.0026275634765625, "loss": 0.012846501916646958, "time": 0.8775410890579224, "epoch": 13, "memory": 25074, "step": 70996} +{"lr": 0.0013885110962219803, "data_time": 0.0021003246307373046, "loss": 0.01695290096104145, "time": 0.8429610252380371, "epoch": 13, "memory": 25074, "step": 71096} +{"lr": 0.0013885110962219803, "data_time": 0.0019771337509155275, "loss": 0.008343653427436948, "time": 0.8354400873184205, "epoch": 13, "memory": 25074, "step": 71196} +{"lr": 0.0013885110962219803, "data_time": 0.0025455713272094726, "loss": 0.013309257617220283, "time": 0.8646746397018432, "epoch": 13, "memory": 25074, "step": 71296} +{"lr": 0.0013885110962219803, "data_time": 0.002090191841125488, "loss": 0.004584446968510747, "time": 0.8228408336639405, "epoch": 13, "memory": 25074, "step": 71396} +{"lr": 0.0013885110962219803, "data_time": 0.0014577150344848634, "loss": 0.005361990164965391, "time": 0.8153012752532959, "epoch": 13, "memory": 25074, "step": 71496} +{"lr": 0.0013885110962219803, "data_time": 0.001925206184387207, "loss": 0.008768650144338608, "time": 0.8246352672576904, "epoch": 13, "memory": 25074, "step": 71596} +{"lr": 0.0013885110962219803, "data_time": 0.002426481246948242, "loss": 0.6324028015136719, "time": 0.7858124494552612, "epoch": 13, "memory": 25074, "step": 71696} +{"lr": 0.0013885110962219803, "data_time": 0.002113151550292969, "loss": 0.40916086733341217, "time": 0.8469485759735107, "epoch": 13, "memory": 25074, "step": 71796} +{"lr": 0.0013885110962219803, "data_time": 0.0018510818481445312, "loss": 0.30697071850299834, "time": 0.8175181150436401, "epoch": 13, "memory": 25074, "step": 71896} +{"lr": 0.0013885110962219803, "data_time": 0.0017873048782348633, "loss": 0.2704250529408455, "time": 0.8586045265197754, "epoch": 13, "memory": 25074, "step": 71996} +{"lr": 0.0013885110962219803, "data_time": 0.0023411035537719725, "loss": 0.22080647200345993, "time": 0.835635781288147, "epoch": 13, "memory": 25074, "step": 72096} +{"lr": 0.0013885110962219803, "data_time": 0.002394556999206543, "loss": 0.16636738702654838, "time": 0.8540791273117065, "epoch": 13, "memory": 25074, "step": 72196} +{"lr": 0.0013885110962219803, "data_time": 0.001729440689086914, "loss": 0.06882301755249501, "time": 0.8588104486465454, "epoch": 13, "memory": 25074, "step": 72296} +{"lr": 0.0013885110962219803, "data_time": 0.00183258056640625, "loss": 0.04123944826424122, "time": 0.8716161727905274, "epoch": 13, "memory": 25074, "step": 72396} +{"lr": 0.0013885110962219803, "data_time": 0.002329111099243164, "loss": 0.03534002546221018, "time": 0.8019241333007813, "epoch": 13, "memory": 25074, "step": 72496} +{"lr": 0.0013885110962219803, "data_time": 0.002138566970825195, "loss": 0.025562022486701606, "time": 0.8151591062545777, "epoch": 13, "memory": 25074, "step": 72596} +{"lr": 0.0013885110962219803, "data_time": 0.0025525808334350584, "loss": 0.01956577431410551, "time": 0.8970122814178467, "epoch": 13, "memory": 25074, "step": 72696} +{"lr": 0.0013885110962219803, "data_time": 0.0024777650833129883, "loss": 0.019316222751513124, "time": 0.8234839677810669, "epoch": 13, "memory": 25074, "step": 72796} +{"lr": 0.0013885110962219803, "data_time": 0.002869844436645508, "loss": 0.014829408191144467, "time": 0.869322943687439, "epoch": 13, "memory": 25074, "step": 72896} +{"lr": 0.0013885110962219803, "data_time": 0.002419590950012207, "loss": 0.04732629181817174, "time": 0.9129178047180175, "epoch": 13, "memory": 25074, "step": 72996} +{"lr": 0.0013885110962219803, "data_time": 0.0023486137390136717, "loss": 0.2543853297829628, "time": 0.9481786251068115, "epoch": 13, "memory": 25074, "step": 73096} +{"lr": 0.0013885110962219803, "data_time": 0.002463889122009277, "loss": 0.05858225151896477, "time": 0.8872850179672241, "epoch": 13, "memory": 25074, "step": 73196} +{"lr": 0.0013885110962219803, "data_time": 0.002854776382446289, "loss": 0.036336281523108485, "time": 0.8857977151870727, "epoch": 13, "memory": 25074, "step": 73296} +{"lr": 0.0013885110962219803, "data_time": 0.0019756317138671874, "loss": 0.018332358822226523, "time": 0.8360615730285644, "epoch": 13, "memory": 25074, "step": 73396} +{"lr": 0.0013885110962219803, "data_time": 0.0018157482147216797, "loss": 0.022497400641441345, "time": 0.8416365623474121, "epoch": 13, "memory": 25074, "step": 73496} +{"lr": 0.0013885110962219803, "data_time": 0.0021916866302490235, "loss": 0.01998066296800971, "time": 0.9434623479843139, "epoch": 13, "memory": 25074, "step": 73596} +{"lr": 0.0013885110962219803, "data_time": 0.002246522903442383, "loss": 0.006339082960039377, "time": 0.8584026336669922, "epoch": 13, "memory": 25074, "step": 73696} +{"lr": 0.0013885110962219803, "data_time": 0.001978421211242676, "loss": 0.01584367179311812, "time": 0.8500343799591065, "epoch": 13, "memory": 25074, "step": 73796} +{"lr": 0.0013885110962219803, "data_time": 0.0025147914886474608, "loss": 0.010897502861917019, "time": 0.8818669557571411, "epoch": 13, "memory": 25074, "step": 73896} +{"lr": 0.0013885110962219803, "data_time": 0.0018570899963378906, "loss": 0.0097136911470443, "time": 0.8574341058731079, "epoch": 13, "memory": 25074, "step": 73996} +{"lr": 0.0013885110962219803, "data_time": 0.002330303192138672, "loss": 0.021634470019489527, "time": 0.8520456790924072, "epoch": 13, "memory": 25074, "step": 74096} +{"lr": 0.0013885110962219803, "data_time": 0.0025841236114501954, "loss": 0.012225399538874627, "time": 0.919404149055481, "epoch": 13, "memory": 25074, "step": 74196} +{"lr": 0.0013885110962219803, "data_time": 0.0017344951629638672, "loss": 0.008988901786506177, "time": 0.8667396545410156, "epoch": 13, "memory": 25074, "step": 74296} +{"lr": 0.0013885110962219803, "data_time": 0.00243377685546875, "loss": 0.009445293573662639, "time": 0.8206976175308227, "epoch": 13, "memory": 25074, "step": 74396} +{"lr": 0.0013885110962219803, "data_time": 0.002050924301147461, "loss": 0.009059466188773513, "time": 0.8605088472366333, "epoch": 13, "memory": 25074, "step": 74496} +{"lr": 0.0013885110962219803, "data_time": 0.001897287368774414, "loss": 0.006352568743750453, "time": 0.8658196210861206, "epoch": 13, "memory": 25074, "step": 74596} +{"lr": 0.0013885110962219803, "data_time": 0.0020585060119628906, "loss": 0.017209505615755915, "time": 0.8472270488739013, "epoch": 13, "memory": 25074, "step": 74696} +{"lr": 0.0013885110962219803, "data_time": 0.0023885488510131834, "loss": 0.011076639639213681, "time": 0.8367435693740845, "epoch": 13, "memory": 25074, "step": 74796} +{"accuracy/top1": 99.9368667602539, "data_time": 0.2643093221327838, "time": 0.6519988705130184, "step": 13} +{"lr": 0.0010992889530196043, "data_time": 0.0021567344665527344, "loss": 0.006026058923453093, "time": 0.8828518390655518, "epoch": 14, "memory": 25074, "step": 74954} +{"lr": 0.0010992889530196043, "data_time": 0.0018913745880126953, "loss": 0.0025050125550478697, "time": 0.9073744297027588, "epoch": 14, "memory": 25074, "step": 75054} +{"lr": 0.0010992889530196043, "data_time": 0.0026072025299072265, "loss": 0.004734926158562303, "time": 0.8835417270660401, "epoch": 14, "memory": 25074, "step": 75154} +{"lr": 0.0010992889530196043, "data_time": 0.0018737316131591797, "loss": 0.012393732834607363, "time": 0.8698746204376221, "epoch": 14, "memory": 25074, "step": 75254} +{"lr": 0.0010992889530196043, "data_time": 0.002016615867614746, "loss": 0.008986308984458447, "time": 0.8357090950012207, "epoch": 14, "memory": 25074, "step": 75354} +{"lr": 0.0010992889530196043, "data_time": 0.0021437406539916992, "loss": 0.007734660571441054, "time": 0.8574966430664063, "epoch": 14, "memory": 25074, "step": 75454} +{"lr": 0.0010992889530196043, "data_time": 0.0016841888427734375, "loss": 0.007169878343120217, "time": 0.8786022186279296, "epoch": 14, "memory": 25074, "step": 75554} +{"lr": 0.0010992889530196043, "data_time": 0.002342367172241211, "loss": 0.009355614241212607, "time": 0.9101663589477539, "epoch": 14, "memory": 25074, "step": 75654} +{"lr": 0.0010992889530196043, "data_time": 0.0021462440490722656, "loss": 0.006325017893686891, "time": 0.9483930587768554, "epoch": 14, "memory": 25074, "step": 75754} +{"lr": 0.0010992889530196043, "data_time": 0.002126932144165039, "loss": 0.0025268110912293196, "time": 0.8317933320999146, "epoch": 14, "memory": 25074, "step": 75854} +{"lr": 0.0010992889530196043, "data_time": 0.002440953254699707, "loss": 0.010621646838262676, "time": 0.8653169870376587, "epoch": 14, "memory": 25074, "step": 75954} +{"lr": 0.0010992889530196043, "data_time": 0.001569676399230957, "loss": 0.004551400756463408, "time": 0.825012469291687, "epoch": 14, "memory": 25074, "step": 76054} +{"lr": 0.0010992889530196043, "data_time": 0.002653026580810547, "loss": 0.0058206333313137295, "time": 0.8512198448181152, "epoch": 14, "memory": 25074, "step": 76154} +{"lr": 0.0010992889530196043, "data_time": 0.0023299455642700195, "loss": 0.005658266041427851, "time": 0.9002356052398681, "epoch": 14, "memory": 25074, "step": 76254} +{"lr": 0.0010992889530196043, "data_time": 0.0019682168960571287, "loss": 0.007896796939894557, "time": 0.8558631896972656, "epoch": 14, "memory": 25074, "step": 76354} +{"lr": 0.0010992889530196043, "data_time": 0.0021454572677612306, "loss": 0.003358024824410677, "time": 0.8499433279037476, "epoch": 14, "memory": 25074, "step": 76454} +{"lr": 0.0010992889530196043, "data_time": 0.0027106046676635743, "loss": 0.010712708998471498, "time": 0.8800162553787232, "epoch": 14, "memory": 25074, "step": 76554} +{"lr": 0.0010992889530196043, "data_time": 0.002442073822021484, "loss": 0.00526658478192985, "time": 0.8586807966232299, "epoch": 14, "memory": 25074, "step": 76654} +{"lr": 0.0010992889530196043, "data_time": 0.0024852752685546875, "loss": 0.007747081574052572, "time": 0.9226675033569336, "epoch": 14, "memory": 25074, "step": 76754} +{"lr": 0.0010992889530196043, "data_time": 0.00218052864074707, "loss": 0.011363293835893273, "time": 0.852879548072815, "epoch": 14, "memory": 25074, "step": 76854} +{"lr": 0.0010992889530196043, "data_time": 0.0023080587387084963, "loss": 0.004773387638852, "time": 0.8198220014572144, "epoch": 14, "memory": 25074, "step": 76954} +{"lr": 0.0010992889530196043, "data_time": 0.001535344123840332, "loss": 0.00914477938786149, "time": 0.7742030382156372, "epoch": 14, "memory": 25074, "step": 77054} +{"lr": 0.0010992889530196043, "data_time": 0.0015209197998046875, "loss": 0.004816683242097497, "time": 0.802119255065918, "epoch": 14, "memory": 25074, "step": 77154} +{"lr": 0.0010992889530196043, "data_time": 0.002005314826965332, "loss": 0.004340975638478994, "time": 0.7940043926239013, "epoch": 14, "memory": 25074, "step": 77254} +{"lr": 0.0010992889530196043, "data_time": 0.0018709182739257812, "loss": 0.006500071287155152, "time": 0.8532385349273681, "epoch": 14, "memory": 25074, "step": 77354} +{"lr": 0.0010992889530196043, "data_time": 0.001375126838684082, "loss": 0.009208159986883402, "time": 0.7792670488357544, "epoch": 14, "memory": 25074, "step": 77454} +{"lr": 0.0010992889530196043, "data_time": 0.001407027244567871, "loss": 0.004191431915387511, "time": 0.8605144023895264, "epoch": 14, "memory": 25074, "step": 77554} +{"lr": 0.0010992889530196043, "data_time": 0.0013914823532104493, "loss": 0.0033865841571241615, "time": 0.8461498737335205, "epoch": 14, "memory": 25074, "step": 77654} +{"lr": 0.0010992889530196043, "data_time": 0.0024385452270507812, "loss": 0.012164186779409647, "time": 0.8398956775665283, "epoch": 14, "memory": 25074, "step": 77754} +{"lr": 0.0010992889530196043, "data_time": 0.0017299413681030273, "loss": 0.003520938800647855, "time": 0.81116943359375, "epoch": 14, "memory": 25074, "step": 77854} +{"lr": 0.0010992889530196043, "data_time": 0.0018192291259765624, "loss": 0.0010594731196761132, "time": 0.8249178886413574, "epoch": 14, "memory": 25074, "step": 77954} +{"lr": 0.0010992889530196043, "data_time": 0.002476000785827637, "loss": 0.010088300658389926, "time": 0.8364016056060791, "epoch": 14, "memory": 25074, "step": 78054} +{"lr": 0.0010992889530196043, "data_time": 0.0025847673416137694, "loss": 0.007270760508254171, "time": 0.8084653377532959, "epoch": 14, "memory": 25074, "step": 78154} +{"lr": 0.0010992889530196043, "data_time": 0.0023340225219726563, "loss": 0.009715610602870583, "time": 0.8354634046554565, "epoch": 14, "memory": 25074, "step": 78254} +{"lr": 0.0010992889530196043, "data_time": 0.0023162364959716797, "loss": 0.006866844790056348, "time": 0.8418134927749634, "epoch": 14, "memory": 25074, "step": 78354} +{"lr": 0.0010992889530196043, "data_time": 0.0018935441970825196, "loss": 0.005691180005669594, "time": 0.7827688932418824, "epoch": 14, "memory": 25074, "step": 78454} +{"lr": 0.0010992889530196043, "data_time": 0.0023929357528686525, "loss": 0.008494042512029409, "time": 0.8424341917037964, "epoch": 14, "memory": 25074, "step": 78554} +{"lr": 0.0010992889530196043, "data_time": 0.002049756050109863, "loss": 0.0077940210234373804, "time": 0.8067271709442139, "epoch": 14, "memory": 25074, "step": 78654} +{"lr": 0.0010992889530196043, "data_time": 0.0021519899368286134, "loss": 0.0046489877626299855, "time": 0.8273500919342041, "epoch": 14, "memory": 25074, "step": 78754} +{"lr": 0.0010992889530196043, "data_time": 0.0017299175262451172, "loss": 0.010990736912935973, "time": 0.7971392154693604, "epoch": 14, "memory": 25074, "step": 78854} +{"lr": 0.0010992889530196043, "data_time": 0.002199697494506836, "loss": 0.010737274214625359, "time": 0.8197074174880982, "epoch": 14, "memory": 25074, "step": 78954} +{"lr": 0.0010992889530196043, "data_time": 0.0025701999664306642, "loss": 0.0026431609876453876, "time": 0.8274394035339355, "epoch": 14, "memory": 25074, "step": 79054} +{"lr": 0.0010992889530196043, "data_time": 0.0018967866897583007, "loss": 0.006859848322346806, "time": 0.8829192399978638, "epoch": 14, "memory": 25074, "step": 79154} +{"lr": 0.0010992889530196043, "data_time": 0.002288532257080078, "loss": 0.0029487856198102238, "time": 0.8272616863250732, "epoch": 14, "memory": 25074, "step": 79254} +{"lr": 0.0010992889530196043, "data_time": 0.0020235300064086912, "loss": 0.0031577238813042642, "time": 0.8196623802185059, "epoch": 14, "memory": 25074, "step": 79354} +{"lr": 0.0010992889530196043, "data_time": 0.0019021034240722656, "loss": 0.0017039770260453223, "time": 0.821526026725769, "epoch": 14, "memory": 25074, "step": 79454} +{"lr": 0.0010992889530196043, "data_time": 0.002126264572143555, "loss": 0.00837366795167327, "time": 0.8593217372894287, "epoch": 14, "memory": 25074, "step": 79554} +{"lr": 0.0010992889530196043, "data_time": 0.002129483222961426, "loss": 0.00560099733993411, "time": 0.8847751379013061, "epoch": 14, "memory": 25074, "step": 79654} +{"lr": 0.0010992889530196043, "data_time": 0.0018660545349121094, "loss": 0.01007423377595842, "time": 0.867729663848877, "epoch": 14, "memory": 25074, "step": 79754} +{"lr": 0.0010992889530196043, "data_time": 0.0018967866897583007, "loss": 0.01046025725081563, "time": 0.804381251335144, "epoch": 14, "memory": 25074, "step": 79854} +{"lr": 0.0010992889530196043, "data_time": 0.0017420530319213867, "loss": 0.004879824677482247, "time": 0.8106240034103394, "epoch": 14, "memory": 25074, "step": 79954} +{"lr": 0.0010992889530196043, "data_time": 0.001981806755065918, "loss": 0.0026554435025900602, "time": 0.7800873756408692, "epoch": 14, "memory": 25074, "step": 80054} +{"lr": 0.0010992889530196043, "data_time": 0.0015960693359375, "loss": 0.0038610049057751896, "time": 0.8743934154510498, "epoch": 14, "memory": 25074, "step": 80154} +{"lr": 0.0010992889530196043, "data_time": 0.0014801263809204102, "loss": 0.006989907007664442, "time": 0.8250518798828125, "epoch": 14, "memory": 25074, "step": 80254} +{"lr": 0.0010992889530196043, "data_time": 0.001961159706115723, "loss": 0.00710472296923399, "time": 0.8247511863708497, "epoch": 14, "memory": 25074, "step": 80354} +{"lr": 0.0010992889530196043, "data_time": 0.002167797088623047, "loss": 0.0075253006536513565, "time": 0.8233062982559204, "epoch": 14, "memory": 25074, "step": 80454} +{"lr": 0.0010992889530196043, "data_time": 0.0016500711441040038, "loss": 0.005123985139653087, "time": 0.7905114412307739, "epoch": 14, "memory": 25074, "step": 80554} +{"accuracy/top1": 99.98737335205078, "data_time": 0.24486652542563045, "time": 0.638165039174697, "step": 14} +{"lr": 0.0008323684216765164, "data_time": 0.002087116241455078, "loss": 0.008190978318452835, "time": 0.8144507646560669, "epoch": 15, "memory": 25074, "step": 80712} +{"lr": 0.0008323684216765164, "data_time": 0.0017007112503051758, "loss": 0.0008685978595167398, "time": 0.7929686069488525, "epoch": 15, "memory": 25074, "step": 80812} +{"lr": 0.0008323684216765164, "data_time": 0.0017495870590209961, "loss": 0.007414591545239091, "time": 0.8182244300842285, "epoch": 15, "memory": 25074, "step": 80912} +{"lr": 0.0008323684216765164, "data_time": 0.001892399787902832, "loss": 0.0029814619570970534, "time": 0.7836994409561158, "epoch": 15, "memory": 25074, "step": 81012} +{"lr": 0.0008323684216765164, "data_time": 0.0016854047775268556, "loss": 0.006268685963004827, "time": 0.8062455654144287, "epoch": 15, "memory": 25074, "step": 81112} +{"lr": 0.0008323684216765164, "data_time": 0.001596975326538086, "loss": 0.004488473618403077, "time": 0.7703092098236084, "epoch": 15, "memory": 25074, "step": 81212} +{"lr": 0.0008323684216765164, "data_time": 0.002002906799316406, "loss": 0.006217769905924797, "time": 0.8155091524124145, "epoch": 15, "memory": 25074, "step": 81312} +{"lr": 0.0008323684216765164, "data_time": 0.002004671096801758, "loss": 0.007581920549273491, "time": 0.8436862230300903, "epoch": 15, "memory": 25074, "step": 81412} +{"lr": 0.0008323684216765164, "data_time": 0.0016895532608032227, "loss": 0.022166385781019926, "time": 0.7929153680801392, "epoch": 15, "memory": 25074, "step": 81512} +{"lr": 0.0008323684216765164, "data_time": 0.0015766620635986328, "loss": 0.003970115818083287, "time": 0.7769886016845703, "epoch": 15, "memory": 25074, "step": 81612} +{"lr": 0.0008323684216765164, "data_time": 0.0025185585021972657, "loss": 0.004429465113207698, "time": 0.8138776063919068, "epoch": 15, "memory": 25074, "step": 81712} +{"lr": 0.0008323684216765164, "data_time": 0.001848292350769043, "loss": 0.0031737711280584334, "time": 0.852851128578186, "epoch": 15, "memory": 25074, "step": 81812} +{"lr": 0.0008323684216765164, "data_time": 0.001421499252319336, "loss": 0.0038157122209668158, "time": 0.8502236366271972, "epoch": 15, "memory": 25074, "step": 81912} +{"lr": 0.0008323684216765164, "data_time": 0.001393437385559082, "loss": 0.004559839516878128, "time": 0.7973427772521973, "epoch": 15, "memory": 25074, "step": 82012} +{"lr": 0.0008323684216765164, "data_time": 0.0021329641342163084, "loss": 0.0030257120728492738, "time": 0.763046407699585, "epoch": 15, "memory": 25074, "step": 82112} +{"lr": 0.0008323684216765164, "data_time": 0.001698756217956543, "loss": 0.0033781186677515505, "time": 0.8239712238311767, "epoch": 15, "memory": 25074, "step": 82212} +{"lr": 0.0008323684216765164, "data_time": 0.0015157938003540039, "loss": 0.0033322657458484174, "time": 0.8517455577850341, "epoch": 15, "memory": 25074, "step": 82312} +{"lr": 0.0008323684216765164, "data_time": 0.001606583595275879, "loss": 0.006936464365571738, "time": 0.8343741178512574, "epoch": 15, "memory": 25074, "step": 82412} +{"lr": 0.0008323684216765164, "data_time": 0.0025830745697021486, "loss": 0.0046401942614465955, "time": 0.8258126497268676, "epoch": 15, "memory": 25074, "step": 82512} +{"lr": 0.0008323684216765164, "data_time": 0.0016058921813964845, "loss": 0.002779401373118162, "time": 0.7366345882415771, "epoch": 15, "memory": 25074, "step": 82612} +{"lr": 0.0008323684216765164, "data_time": 0.001593470573425293, "loss": 0.004100637137889862, "time": 0.901034951210022, "epoch": 15, "memory": 25074, "step": 82712} +{"lr": 0.0008323684216765164, "data_time": 0.0021658897399902343, "loss": 0.0032065569888800383, "time": 0.8671699523925781, "epoch": 15, "memory": 25074, "step": 82812} +{"lr": 0.0008323684216765164, "data_time": 0.0015111684799194336, "loss": 0.0036294298246502877, "time": 0.8019484519958496, "epoch": 15, "memory": 25074, "step": 82912} +{"lr": 0.0008323684216765164, "data_time": 0.0016389846801757812, "loss": 0.0033681821078062057, "time": 0.8077052116394043, "epoch": 15, "memory": 25074, "step": 83012} +{"lr": 0.0008323684216765164, "data_time": 0.0021733522415161135, "loss": 0.002631246857345104, "time": 0.7924538135528565, "epoch": 15, "memory": 25074, "step": 83112} +{"lr": 0.0008323684216765164, "data_time": 0.0016282081604003906, "loss": 0.0025754041504114865, "time": 0.8362874507904052, "epoch": 15, "memory": 25074, "step": 83212} +{"lr": 0.0008323684216765164, "data_time": 0.0026250839233398437, "loss": 0.0013220942579209804, "time": 0.8433773279190063, "epoch": 15, "memory": 25074, "step": 83312} +{"lr": 0.0008323684216765164, "data_time": 0.0014664649963378907, "loss": 0.003958883741870522, "time": 0.8503497123718262, "epoch": 15, "memory": 25074, "step": 83412} +{"lr": 0.0008323684216765164, "data_time": 0.0019595861434936524, "loss": 0.0016685916110873223, "time": 0.835192608833313, "epoch": 15, "memory": 25074, "step": 83512} +{"lr": 0.0008323684216765164, "data_time": 0.0015546560287475586, "loss": 0.003671620553359389, "time": 0.8708398580551148, "epoch": 15, "memory": 25074, "step": 83612} +{"lr": 0.0008323684216765164, "data_time": 0.0019462108612060547, "loss": 0.0023081721272319557, "time": 0.8684589862823486, "epoch": 15, "memory": 25074, "step": 83712} +{"lr": 0.0008323684216765164, "data_time": 0.0015102148056030274, "loss": 0.007967055262997746, "time": 0.9060788869857788, "epoch": 15, "memory": 25074, "step": 83812} +{"lr": 0.0008323684216765164, "data_time": 0.001999187469482422, "loss": 0.002258659154176712, "time": 0.7957812070846557, "epoch": 15, "memory": 25074, "step": 83912} +{"lr": 0.0008323684216765164, "data_time": 0.002222752571105957, "loss": 0.0024430623278021814, "time": 0.8611499547958374, "epoch": 15, "memory": 25074, "step": 84012} +{"lr": 0.0008323684216765164, "data_time": 0.0021514415740966795, "loss": 0.002764499979093671, "time": 0.8284383773803711, "epoch": 15, "memory": 25074, "step": 84112} +{"lr": 0.0008323684216765164, "data_time": 0.002145242691040039, "loss": 0.002132095117121935, "time": 0.8111137866973877, "epoch": 15, "memory": 25074, "step": 84212} +{"lr": 0.0008323684216765164, "data_time": 0.002154636383056641, "loss": 0.0018910645972937345, "time": 0.8173701763153076, "epoch": 15, "memory": 25074, "step": 84312} +{"lr": 0.0008323684216765164, "data_time": 0.0021457672119140625, "loss": 0.0027873084880411627, "time": 0.8290945053100586, "epoch": 15, "memory": 25074, "step": 84412} +{"lr": 0.0008323684216765164, "data_time": 0.002359342575073242, "loss": 0.001872691372409463, "time": 0.8151299476623535, "epoch": 15, "memory": 25074, "step": 84512} +{"lr": 0.0008323684216765164, "data_time": 0.002001833915710449, "loss": 0.004440096439793706, "time": 0.8477199077606201, "epoch": 15, "memory": 25074, "step": 84612} +{"lr": 0.0008323684216765164, "data_time": 0.002122354507446289, "loss": 0.0070302918087691065, "time": 0.8791698455810547, "epoch": 15, "memory": 25074, "step": 84712} +{"lr": 0.0008323684216765164, "data_time": 0.002477121353149414, "loss": 0.0017334240023046733, "time": 0.818484902381897, "epoch": 15, "memory": 25074, "step": 84812} +{"lr": 0.0008323684216765164, "data_time": 0.0024985551834106447, "loss": 0.0031128203961998226, "time": 0.8045818567276001, "epoch": 15, "memory": 25074, "step": 84912} +{"lr": 0.0008323684216765164, "data_time": 0.001815032958984375, "loss": 0.0020079654175788166, "time": 0.8695090293884278, "epoch": 15, "memory": 25074, "step": 85012} +{"lr": 0.0008323684216765164, "data_time": 0.0015797138214111329, "loss": 0.0015763045754283666, "time": 0.7929683685302734, "epoch": 15, "memory": 25074, "step": 85112} +{"lr": 0.0008323684216765164, "data_time": 0.0021280527114868166, "loss": 0.0030572716146707534, "time": 0.9226945877075196, "epoch": 15, "memory": 25074, "step": 85212} +{"lr": 0.0008323684216765164, "data_time": 0.0016113519668579102, "loss": 0.0042770013213157656, "time": 0.8555640459060669, "epoch": 15, "memory": 25074, "step": 85312} +{"lr": 0.0008323684216765164, "data_time": 0.0016779899597167969, "loss": 0.0056419032625854015, "time": 0.8675498962402344, "epoch": 15, "memory": 25074, "step": 85412} +{"lr": 0.0008323684216765164, "data_time": 0.0016929149627685548, "loss": 0.004891542112454772, "time": 0.8150385856628418, "epoch": 15, "memory": 25074, "step": 85512} +{"lr": 0.0008323684216765164, "data_time": 0.0023407697677612304, "loss": 0.0024810972157865764, "time": 0.9147989749908447, "epoch": 15, "memory": 25074, "step": 85612} +{"lr": 0.0008323684216765164, "data_time": 0.0019356966018676757, "loss": 0.0024863758590072393, "time": 0.829637885093689, "epoch": 15, "memory": 25074, "step": 85712} +{"lr": 0.0008323684216765164, "data_time": 0.0029778480529785156, "loss": 0.005940651195123792, "time": 0.8046359777450561, "epoch": 15, "memory": 25074, "step": 85812} +{"lr": 0.0008323684216765164, "data_time": 0.0020565509796142576, "loss": 0.00481735453940928, "time": 0.8162915229797363, "epoch": 15, "memory": 25074, "step": 85912} +{"lr": 0.0008323684216765164, "data_time": 0.002616596221923828, "loss": 0.004171461844816804, "time": 0.8333085536956787, "epoch": 15, "memory": 25074, "step": 86012} +{"lr": 0.0008323684216765164, "data_time": 0.0020272493362426757, "loss": 0.0023523079231381415, "time": 0.816011929512024, "epoch": 15, "memory": 25074, "step": 86112} +{"lr": 0.0008323684216765164, "data_time": 0.0019292354583740235, "loss": 0.0032898181583732367, "time": 0.8411876678466796, "epoch": 15, "memory": 25074, "step": 86212} +{"lr": 0.0008323684216765164, "data_time": 0.0017971992492675781, "loss": 0.0039269427303224805, "time": 0.8080334424972534, "epoch": 15, "memory": 25074, "step": 86312} +{"accuracy/top1": 100.0, "data_time": 0.23414185467888327, "time": 0.6187553966746611, "step": 15} +{"lr": 0.0005943219715328379, "data_time": 0.001903700828552246, "loss": 0.0025037748273462057, "time": 0.8635496377944947, "epoch": 16, "memory": 25074, "step": 86470} +{"lr": 0.0005943219715328379, "data_time": 0.0023732900619506834, "loss": 0.0033647256903350355, "time": 0.8420201539993286, "epoch": 16, "memory": 25074, "step": 86570} +{"lr": 0.0005943219715328379, "data_time": 0.002093648910522461, "loss": 0.002997089829295874, "time": 0.7624979257583618, "epoch": 16, "memory": 25074, "step": 86670} +{"lr": 0.0005943219715328379, "data_time": 0.0020543336868286133, "loss": 0.003493775613605976, "time": 0.860261869430542, "epoch": 16, "memory": 25074, "step": 86770} +{"lr": 0.0005943219715328379, "data_time": 0.002039337158203125, "loss": 0.001480706874281168, "time": 0.7583719968795777, "epoch": 16, "memory": 25074, "step": 86870} +{"lr": 0.0005943219715328379, "data_time": 0.0018027544021606446, "loss": 0.0032122720498591662, "time": 0.8659549236297608, "epoch": 16, "memory": 25074, "step": 86970} +{"lr": 0.0005943219715328379, "data_time": 0.0017798185348510743, "loss": 0.0005405058618634939, "time": 0.7693642616271973, "epoch": 16, "memory": 25074, "step": 87070} +{"lr": 0.0005943219715328379, "data_time": 0.0020461797714233397, "loss": 0.0006689358502626419, "time": 0.8643156051635742, "epoch": 16, "memory": 25074, "step": 87170} +{"lr": 0.0005943219715328379, "data_time": 0.002682352066040039, "loss": 0.002496235119178891, "time": 0.8343918085098266, "epoch": 16, "memory": 25074, "step": 87270} +{"lr": 0.0005943219715328379, "data_time": 0.0016872644424438476, "loss": 0.0020704213995486496, "time": 0.8328126668930054, "epoch": 16, "memory": 25074, "step": 87370} +{"lr": 0.0005943219715328379, "data_time": 0.0017588138580322266, "loss": 0.0033198816701769827, "time": 0.8297866582870483, "epoch": 16, "memory": 25074, "step": 87470} +{"lr": 0.0005943219715328379, "data_time": 0.0015356302261352538, "loss": 0.0030173714738339187, "time": 0.8674288988113403, "epoch": 16, "memory": 25074, "step": 87570} +{"lr": 0.0005943219715328379, "data_time": 0.002687549591064453, "loss": 0.0004096377640962601, "time": 0.8186800479888916, "epoch": 16, "memory": 25074, "step": 87670} +{"lr": 0.0005943219715328379, "data_time": 0.0023172855377197265, "loss": 0.002651927527040243, "time": 0.9097333669662475, "epoch": 16, "memory": 25074, "step": 87770} +{"lr": 0.0005943219715328379, "data_time": 0.001935577392578125, "loss": 0.002328659035265446, "time": 0.8453935384750366, "epoch": 16, "memory": 25074, "step": 87870} +{"lr": 0.0005943219715328379, "data_time": 0.0019171476364135743, "loss": 0.003917717421427369, "time": 0.8380070924758911, "epoch": 16, "memory": 25074, "step": 87970} +{"lr": 0.0005943219715328379, "data_time": 0.0025748252868652345, "loss": 0.0021569257602095603, "time": 0.826769208908081, "epoch": 16, "memory": 25074, "step": 88070} +{"lr": 0.0005943219715328379, "data_time": 0.00251309871673584, "loss": 0.001567025063559413, "time": 0.8712774991989136, "epoch": 16, "memory": 25074, "step": 88170} +{"lr": 0.0005943219715328379, "data_time": 0.0015096664428710938, "loss": 0.003016546554863453, "time": 0.8645804643630981, "epoch": 16, "memory": 25074, "step": 88270} +{"lr": 0.0005943219715328379, "data_time": 0.0015955448150634765, "loss": 0.0008537331130355596, "time": 0.8699898719787598, "epoch": 16, "memory": 25074, "step": 88370} +{"lr": 0.0005943219715328379, "data_time": 0.0020115852355957033, "loss": 0.002749787736684084, "time": 0.8488941669464112, "epoch": 16, "memory": 25074, "step": 88470} +{"lr": 0.0005943219715328379, "data_time": 0.0024833202362060545, "loss": 0.001474676700308919, "time": 0.8312921524047852, "epoch": 16, "memory": 25074, "step": 88570} +{"lr": 0.0005943219715328379, "data_time": 0.0019132137298583985, "loss": 0.0009348067920655012, "time": 0.8532140970230102, "epoch": 16, "memory": 25074, "step": 88670} +{"lr": 0.0005943219715328379, "data_time": 0.002227473258972168, "loss": 0.0007522489875555038, "time": 0.858740258216858, "epoch": 16, "memory": 25074, "step": 88770} +{"lr": 0.0005943219715328379, "data_time": 0.0015662431716918946, "loss": 0.0037273469381034373, "time": 0.8913288831710815, "epoch": 16, "memory": 25074, "step": 88870} +{"lr": 0.0005943219715328379, "data_time": 0.0037634849548339845, "loss": 0.0010302946902811527, "time": 0.9533833742141724, "epoch": 16, "memory": 25074, "step": 88970} +{"lr": 0.0005943219715328379, "data_time": 0.0020781040191650392, "loss": 0.0015971431974321603, "time": 0.9602782726287842, "epoch": 16, "memory": 25074, "step": 89070} +{"lr": 0.0005943219715328379, "data_time": 0.002176856994628906, "loss": 0.0005267304833978415, "time": 0.924289870262146, "epoch": 16, "memory": 25074, "step": 89170} +{"lr": 0.0005943219715328379, "data_time": 0.002349090576171875, "loss": 0.010791574558243155, "time": 0.9090284585952759, "epoch": 16, "memory": 25074, "step": 89270} +{"lr": 0.0005943219715328379, "data_time": 0.0015442848205566406, "loss": 0.001905987225472927, "time": 0.892949104309082, "epoch": 16, "memory": 25074, "step": 89370} +{"lr": 0.0005943219715328379, "data_time": 0.0015676259994506837, "loss": 0.0007513926830142736, "time": 0.9010032892227173, "epoch": 16, "memory": 25074, "step": 89470} +{"lr": 0.0005943219715328379, "data_time": 0.0015018224716186524, "loss": 0.0017784679774194955, "time": 0.8695396423339844, "epoch": 16, "memory": 25074, "step": 89570} +{"lr": 0.0005943219715328379, "data_time": 0.0015874624252319336, "loss": 0.0006417270749807357, "time": 0.877429747581482, "epoch": 16, "memory": 25074, "step": 89670} +{"lr": 0.0005943219715328379, "data_time": 0.0016166210174560548, "loss": 0.00294205192476511, "time": 0.8348247289657593, "epoch": 16, "memory": 25074, "step": 89770} +{"lr": 0.0005943219715328379, "data_time": 0.0022717952728271485, "loss": 0.003139247512444854, "time": 0.808187747001648, "epoch": 16, "memory": 25074, "step": 89870} +{"lr": 0.0005943219715328379, "data_time": 0.0016261100769042968, "loss": 0.0013865420129150151, "time": 0.7832516670227051, "epoch": 16, "memory": 25074, "step": 89970} +{"lr": 0.0005943219715328379, "data_time": 0.0015677690505981445, "loss": 0.0018713842146098614, "time": 0.8313507795333862, "epoch": 16, "memory": 25074, "step": 90070} +{"lr": 0.0005943219715328379, "data_time": 0.0019213438034057617, "loss": 0.0007648433558642865, "time": 0.8319806337356568, "epoch": 16, "memory": 25074, "step": 90170} +{"lr": 0.0005943219715328379, "data_time": 0.0014399290084838867, "loss": 0.003510078974068165, "time": 0.8241634845733643, "epoch": 16, "memory": 25074, "step": 90270} +{"lr": 0.0005943219715328379, "data_time": 0.002419281005859375, "loss": 0.0022074762731790543, "time": 0.8208592176437378, "epoch": 16, "memory": 25074, "step": 90370} +{"lr": 0.0005943219715328379, "data_time": 0.0026560306549072267, "loss": 0.0024664254393428563, "time": 0.7743782520294189, "epoch": 16, "memory": 25074, "step": 90470} +{"lr": 0.0005943219715328379, "data_time": 0.0016248703002929687, "loss": 0.004019773378968239, "time": 0.7739948749542236, "epoch": 16, "memory": 25074, "step": 90570} +{"lr": 0.0005943219715328379, "data_time": 0.002538037300109863, "loss": 0.0027326020412147046, "time": 0.816907525062561, "epoch": 16, "memory": 25074, "step": 90670} +{"lr": 0.0005943219715328379, "data_time": 0.002382326126098633, "loss": 0.006594126066192985, "time": 0.8274320363998413, "epoch": 16, "memory": 25074, "step": 90770} +{"lr": 0.0005943219715328379, "data_time": 0.0016460180282592773, "loss": 0.002716598007827997, "time": 0.8251569509506226, "epoch": 16, "memory": 25074, "step": 90870} +{"lr": 0.0005943219715328379, "data_time": 0.0017352104187011719, "loss": 0.005413290625438094, "time": 0.7875924110412598, "epoch": 16, "memory": 25074, "step": 90970} +{"lr": 0.0005943219715328379, "data_time": 0.0025635957717895508, "loss": 0.001384977111592889, "time": 0.8089169263839722, "epoch": 16, "memory": 25074, "step": 91070} +{"lr": 0.0005943219715328379, "data_time": 0.0015411853790283203, "loss": 0.0009327524341642857, "time": 0.7445062875747681, "epoch": 16, "memory": 25074, "step": 91170} +{"lr": 0.0005943219715328379, "data_time": 0.0021345853805541993, "loss": 0.0019064848311245441, "time": 0.7698233842849731, "epoch": 16, "memory": 25074, "step": 91270} +{"lr": 0.0005943219715328379, "data_time": 0.0017052650451660155, "loss": 0.0007534346543252469, "time": 0.7804375648498535, "epoch": 16, "memory": 25074, "step": 91370} +{"lr": 0.0005943219715328379, "data_time": 0.002021384239196777, "loss": 0.005605431459844113, "time": 0.7461480855941772, "epoch": 16, "memory": 25074, "step": 91470} +{"lr": 0.0005943219715328379, "data_time": 0.0025379419326782226, "loss": 0.000628881249576807, "time": 0.8316758632659912, "epoch": 16, "memory": 25074, "step": 91570} +{"lr": 0.0005943219715328379, "data_time": 0.001603841781616211, "loss": 0.0013007433619350195, "time": 0.7501929759979248, "epoch": 16, "memory": 25074, "step": 91670} +{"lr": 0.0005943219715328379, "data_time": 0.002016425132751465, "loss": 0.0012229129672050477, "time": 0.7771937847137451, "epoch": 16, "memory": 25074, "step": 91770} diff --git a/convnext_small_4xb256_4e-3lr_5m/20230531_173252/vis_data/vis_image/0000000.png_1.png b/convnext_small_4xb256_4e-3lr_5m/20230531_173252/vis_data/vis_image/0000000.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..6f264824565adf0630fdfc2e331f1ad6f2495172 Binary files /dev/null and b/convnext_small_4xb256_4e-3lr_5m/20230531_173252/vis_data/vis_image/0000000.png_1.png differ diff --git a/convnext_small_4xb256_4e-3lr_5m/20230531_173252/vis_data/vis_image/0000000.png_10.png b/convnext_small_4xb256_4e-3lr_5m/20230531_173252/vis_data/vis_image/0000000.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..934769c1f2bb2402694533d16138c8dbc1ce4a97 Binary files /dev/null and b/convnext_small_4xb256_4e-3lr_5m/20230531_173252/vis_data/vis_image/0000000.png_10.png differ diff --git a/convnext_small_4xb256_4e-3lr_5m/20230531_173252/vis_data/vis_image/0000000.png_11.png b/convnext_small_4xb256_4e-3lr_5m/20230531_173252/vis_data/vis_image/0000000.png_11.png new file mode 100644 index 0000000000000000000000000000000000000000..dbd00a21bdda4f44b04908f3673b5c490c33ef18 Binary files /dev/null and b/convnext_small_4xb256_4e-3lr_5m/20230531_173252/vis_data/vis_image/0000000.png_11.png differ diff --git a/convnext_small_4xb256_4e-3lr_5m/20230531_173252/vis_data/vis_image/0000000.png_12.png b/convnext_small_4xb256_4e-3lr_5m/20230531_173252/vis_data/vis_image/0000000.png_12.png new file mode 100644 index 0000000000000000000000000000000000000000..0498c212162131517529b3d9090a1f8df25b93f4 Binary files /dev/null and b/convnext_small_4xb256_4e-3lr_5m/20230531_173252/vis_data/vis_image/0000000.png_12.png differ diff --git a/convnext_small_4xb256_4e-3lr_5m/20230531_173252/vis_data/vis_image/0000000.png_13.png b/convnext_small_4xb256_4e-3lr_5m/20230531_173252/vis_data/vis_image/0000000.png_13.png new file mode 100644 index 0000000000000000000000000000000000000000..78086a4ac43d9c8da460aa6d9b24e940253194a5 Binary files /dev/null and b/convnext_small_4xb256_4e-3lr_5m/20230531_173252/vis_data/vis_image/0000000.png_13.png differ diff --git a/convnext_small_4xb256_4e-3lr_5m/20230531_173252/vis_data/vis_image/0000000.png_14.png b/convnext_small_4xb256_4e-3lr_5m/20230531_173252/vis_data/vis_image/0000000.png_14.png new file mode 100644 index 0000000000000000000000000000000000000000..b2aa1aaaf9f97ce8b19f3bbce6b7024e76a7ba4c Binary files /dev/null and b/convnext_small_4xb256_4e-3lr_5m/20230531_173252/vis_data/vis_image/0000000.png_14.png differ diff --git a/convnext_small_4xb256_4e-3lr_5m/20230531_173252/vis_data/vis_image/0000000.png_15.png b/convnext_small_4xb256_4e-3lr_5m/20230531_173252/vis_data/vis_image/0000000.png_15.png new file mode 100644 index 0000000000000000000000000000000000000000..6f264824565adf0630fdfc2e331f1ad6f2495172 Binary files /dev/null and b/convnext_small_4xb256_4e-3lr_5m/20230531_173252/vis_data/vis_image/0000000.png_15.png differ diff --git a/convnext_small_4xb256_4e-3lr_5m/20230531_173252/vis_data/vis_image/0000000.png_2.png b/convnext_small_4xb256_4e-3lr_5m/20230531_173252/vis_data/vis_image/0000000.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..e1606975577487c3b0be6dd199d6f550e6fa62c5 Binary files /dev/null and b/convnext_small_4xb256_4e-3lr_5m/20230531_173252/vis_data/vis_image/0000000.png_2.png differ diff --git a/convnext_small_4xb256_4e-3lr_5m/20230531_173252/vis_data/vis_image/0000000.png_3.png b/convnext_small_4xb256_4e-3lr_5m/20230531_173252/vis_data/vis_image/0000000.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..3068bccfa8ed5f3ad7e0110060b2f984e12a99fb Binary files /dev/null and b/convnext_small_4xb256_4e-3lr_5m/20230531_173252/vis_data/vis_image/0000000.png_3.png differ diff --git a/convnext_small_4xb256_4e-3lr_5m/20230531_173252/vis_data/vis_image/0000000.png_4.png b/convnext_small_4xb256_4e-3lr_5m/20230531_173252/vis_data/vis_image/0000000.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..735eb30c6d4a289c06416cbe6a29f4c3f5fd2291 Binary files /dev/null and b/convnext_small_4xb256_4e-3lr_5m/20230531_173252/vis_data/vis_image/0000000.png_4.png differ diff --git a/convnext_small_4xb256_4e-3lr_5m/20230531_173252/vis_data/vis_image/0000000.png_5.png b/convnext_small_4xb256_4e-3lr_5m/20230531_173252/vis_data/vis_image/0000000.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..1ad0d513db818562ea7f70c0e772de827724e16f Binary files /dev/null and b/convnext_small_4xb256_4e-3lr_5m/20230531_173252/vis_data/vis_image/0000000.png_5.png differ diff --git a/convnext_small_4xb256_4e-3lr_5m/20230531_173252/vis_data/vis_image/0000000.png_6.png b/convnext_small_4xb256_4e-3lr_5m/20230531_173252/vis_data/vis_image/0000000.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..4eb7a27eaf8becfc9f84383d9afc43ebe1f725b7 Binary files /dev/null and b/convnext_small_4xb256_4e-3lr_5m/20230531_173252/vis_data/vis_image/0000000.png_6.png differ diff --git a/convnext_small_4xb256_4e-3lr_5m/20230531_173252/vis_data/vis_image/0000000.png_7.png b/convnext_small_4xb256_4e-3lr_5m/20230531_173252/vis_data/vis_image/0000000.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..0b012ee8d3450751b93494c45d561703813ef77e Binary files /dev/null and b/convnext_small_4xb256_4e-3lr_5m/20230531_173252/vis_data/vis_image/0000000.png_7.png differ diff --git a/convnext_small_4xb256_4e-3lr_5m/20230531_173252/vis_data/vis_image/0000000.png_8.png b/convnext_small_4xb256_4e-3lr_5m/20230531_173252/vis_data/vis_image/0000000.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..dbd00a21bdda4f44b04908f3673b5c490c33ef18 Binary files /dev/null and b/convnext_small_4xb256_4e-3lr_5m/20230531_173252/vis_data/vis_image/0000000.png_8.png differ diff --git a/convnext_small_4xb256_4e-3lr_5m/20230531_173252/vis_data/vis_image/0000000.png_9.png b/convnext_small_4xb256_4e-3lr_5m/20230531_173252/vis_data/vis_image/0000000.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..359ab8b499d0c0d2b132b04283ae399f01124a8b Binary files /dev/null and b/convnext_small_4xb256_4e-3lr_5m/20230531_173252/vis_data/vis_image/0000000.png_9.png differ diff --git a/convnext_small_4xb256_4e-3lr_5m/convnext_small_4xb256_fake5m.py b/convnext_small_4xb256_4e-3lr_5m/convnext_small_4xb256_fake5m.py new file mode 100644 index 0000000000000000000000000000000000000000..675945812b0b3cd1ce1dff570c0730faa54378a1 --- /dev/null +++ b/convnext_small_4xb256_4e-3lr_5m/convnext_small_4xb256_fake5m.py @@ -0,0 +1,169 @@ +optim_wrapper = dict( + optimizer=dict( + type='AdamW', + lr=0.004, + weight_decay=0.05, + eps=1e-08, + betas=(0.9, 0.999), + _scope_='mmpretrain'), + paramwise_cfg=dict( + norm_decay_mult=0.0, + bias_decay_mult=0.0, + flat_decay_mult=0.0, + custom_keys=dict({ + '.absolute_pos_embed': dict(decay_mult=0.0), + '.relative_position_bias_table': dict(decay_mult=0.0) + })), + type='AmpOptimWrapper', + dtype='bfloat16', + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=True, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=20, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=4096) +model = dict( + type='ImageClassifier', + backbone=dict(type='ConvNeXt', arch='small', drop_path_rate=0.4), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=768, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + init_cfg=None), + init_cfg=dict( + type='TruncNormal', layer=['Conv2d', 'Linear'], std=0.02, bias=0.0), + train_cfg=None) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stablediffusionV1-5R2-dpmsolver-25-5m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/convnext_small_4xb256_4e-3lr_5m' diff --git a/convnext_small_4xb256_4e-3lr_5m/epoch_1.pth b/convnext_small_4xb256_4e-3lr_5m/epoch_1.pth new file mode 100644 index 0000000000000000000000000000000000000000..4d5becab25bfd1192e26ba8e9426abde138bc09e --- /dev/null +++ b/convnext_small_4xb256_4e-3lr_5m/epoch_1.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:3d1cf779fc7c49988df6c2d5e28d33039aead39cc35d3e40043735753793e22c +size 792317408 diff --git a/convnext_small_4xb256_4e-3lr_5m/epoch_10.pth b/convnext_small_4xb256_4e-3lr_5m/epoch_10.pth new file mode 100644 index 0000000000000000000000000000000000000000..b319fe337cb89e1bf337759baf191709109d2d58 --- /dev/null +++ b/convnext_small_4xb256_4e-3lr_5m/epoch_10.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:9e45fa3b6b3e319708026c75f59cad0715d84e448f1a097fb180be9e063d9ab7 +size 796093408 diff --git a/convnext_small_4xb256_4e-3lr_5m/epoch_11.pth b/convnext_small_4xb256_4e-3lr_5m/epoch_11.pth new file mode 100644 index 0000000000000000000000000000000000000000..585d9b083b7691e32962b196dbd4a0d5c0a1db93 --- /dev/null +++ b/convnext_small_4xb256_4e-3lr_5m/epoch_11.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:f7654b0bdc9dfa98fb3d368f655c33be3c5dfa01119d342258504087322dd601 +size 796513568 diff --git a/convnext_small_4xb256_4e-3lr_5m/epoch_12.pth b/convnext_small_4xb256_4e-3lr_5m/epoch_12.pth new file mode 100644 index 0000000000000000000000000000000000000000..649cec26dce73c7588555f77b0f871a94c99c010 --- /dev/null +++ b/convnext_small_4xb256_4e-3lr_5m/epoch_12.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:bf746107885d6b2f83291cae2f217d67577f3597d5c85d8186e6c9f7b6ca3a4a +size 796944224 diff --git a/convnext_small_4xb256_4e-3lr_5m/epoch_13.pth b/convnext_small_4xb256_4e-3lr_5m/epoch_13.pth new file mode 100644 index 0000000000000000000000000000000000000000..668834143cd981ec16f974399babb45e8be55b69 --- /dev/null +++ b/convnext_small_4xb256_4e-3lr_5m/epoch_13.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:9dad2b11c8ada8a71a1454fb9a62a18bb9a5c61deddfe07a7feecd81dc1510c0 +size 797369632 diff --git a/convnext_small_4xb256_4e-3lr_5m/epoch_14.pth b/convnext_small_4xb256_4e-3lr_5m/epoch_14.pth new file mode 100644 index 0000000000000000000000000000000000000000..a321bee1e0746db01e0c10143468a196bc60e615 --- /dev/null +++ b/convnext_small_4xb256_4e-3lr_5m/epoch_14.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:636ea4eb1c51d3f0b2e9a005d2cb629359a8e7e7a7b3ce1f916e8013ce43e29a +size 797786656 diff --git a/convnext_small_4xb256_4e-3lr_5m/epoch_15.pth b/convnext_small_4xb256_4e-3lr_5m/epoch_15.pth new file mode 100644 index 0000000000000000000000000000000000000000..77a87c2f9ea7fb2cc78d7668fb8efe8eee59fa0d --- /dev/null +++ b/convnext_small_4xb256_4e-3lr_5m/epoch_15.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:30dbe1b5ced0786fe4fc8ccff492bece3ab26f4e5bf9ef1ec0c3f9d0ed654b76 +size 798196256 diff --git a/convnext_small_4xb256_4e-3lr_5m/epoch_2.pth b/convnext_small_4xb256_4e-3lr_5m/epoch_2.pth new file mode 100644 index 0000000000000000000000000000000000000000..75acb2393d270c2bd58365a728057f686be7c522 --- /dev/null +++ b/convnext_small_4xb256_4e-3lr_5m/epoch_2.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:2d3a704ece89828cda56cf19852dd8ac8ca6b0800a6b584a3f0085d4c79918a9 +size 792738848 diff --git a/convnext_small_4xb256_4e-3lr_5m/epoch_3.pth b/convnext_small_4xb256_4e-3lr_5m/epoch_3.pth new file mode 100644 index 0000000000000000000000000000000000000000..6514b70193596d4a6f16c0e644a57317d9d222d5 --- /dev/null +++ b/convnext_small_4xb256_4e-3lr_5m/epoch_3.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:a80906f9eaf0673bd9e1b117be32653b68139c3b3b2a4164739c4eb9a3da22e1 +size 793153440 diff --git a/convnext_small_4xb256_4e-3lr_5m/epoch_4.pth b/convnext_small_4xb256_4e-3lr_5m/epoch_4.pth new file mode 100644 index 0000000000000000000000000000000000000000..5654137f2becd45f6b75a4ca2ef5fa79217bc591 --- /dev/null +++ b/convnext_small_4xb256_4e-3lr_5m/epoch_4.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:250509efb25c69919aea5d29f293d620ee964524030c0678c0c3ac1a1475bc7f +size 793574112 diff --git a/convnext_small_4xb256_4e-3lr_5m/epoch_5.pth b/convnext_small_4xb256_4e-3lr_5m/epoch_5.pth new file mode 100644 index 0000000000000000000000000000000000000000..b37da917d9b0fdfc47ec6d69d621f79c87813b54 --- /dev/null +++ b/convnext_small_4xb256_4e-3lr_5m/epoch_5.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:1c2570775babc73580dafb617ad735af03b9fd0cedd9d9a11ad55ba6a8d11341 +size 794006880 diff --git a/convnext_small_4xb256_4e-3lr_5m/epoch_6.pth b/convnext_small_4xb256_4e-3lr_5m/epoch_6.pth new file mode 100644 index 0000000000000000000000000000000000000000..6eb88605b8bfa8efecd5e5272ef7fd086ee24146 --- /dev/null +++ b/convnext_small_4xb256_4e-3lr_5m/epoch_6.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:84439a78540c2c9dad329dfd3ee8fb84e80a0cdd2f93fb3eacf77621b88d86de +size 794427296 diff --git a/convnext_small_4xb256_4e-3lr_5m/epoch_7.pth b/convnext_small_4xb256_4e-3lr_5m/epoch_7.pth new file mode 100644 index 0000000000000000000000000000000000000000..db7b3c4bb63994e2c27909890262cb16032c5799 --- /dev/null +++ b/convnext_small_4xb256_4e-3lr_5m/epoch_7.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:1e50570fecf5bf4cb0d44a1f1f3f6b8f43a5dad8bc6d623a73e11fffafbb5ef5 +size 794859232 diff --git a/convnext_small_4xb256_4e-3lr_5m/epoch_8.pth b/convnext_small_4xb256_4e-3lr_5m/epoch_8.pth new file mode 100644 index 0000000000000000000000000000000000000000..4eb3a7aef2b322234c860b7cd98a356e71fcba11 --- /dev/null +++ b/convnext_small_4xb256_4e-3lr_5m/epoch_8.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:de477f93a513b287d0099f187df0cbe2744e27c4cec21dbcc0a6eb54d0d081a5 +size 795268896 diff --git a/convnext_small_4xb256_4e-3lr_5m/epoch_9.pth b/convnext_small_4xb256_4e-3lr_5m/epoch_9.pth new file mode 100644 index 0000000000000000000000000000000000000000..81b9382b2638f810fd0b72023f2e48ab32e0abdd --- /dev/null +++ b/convnext_small_4xb256_4e-3lr_5m/epoch_9.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c59458a320b4c6cc1543430ea50bd268116ce86f7fe91ae9a5e73726b71edafe +size 795678304 diff --git a/convnext_small_4xb256_4e-3lr_5m/last_checkpoint b/convnext_small_4xb256_4e-3lr_5m/last_checkpoint new file mode 100644 index 0000000000000000000000000000000000000000..9db9a6e186edf58dff29332d7dbcade0a244e376 --- /dev/null +++ b/convnext_small_4xb256_4e-3lr_5m/last_checkpoint @@ -0,0 +1 @@ +/mnt/petrelfs/luzeyu/workspace/fakebench/mmpretrain/workdir/convnext_small_4xb256_4e-3lr_5m/epoch_15.pth \ No newline at end of file diff --git a/convnext_small_4xb256_IF_1m_lr4e-4_aug_1e-1/20230606_050018/20230606_050018.log b/convnext_small_4xb256_IF_1m_lr4e-4_aug_1e-1/20230606_050018/20230606_050018.log new file mode 100644 index 0000000000000000000000000000000000000000..2764555ea3e965450c5836c7570ddae12e44f839 --- /dev/null +++ b/convnext_small_4xb256_IF_1m_lr4e-4_aug_1e-1/20230606_050018/20230606_050018.log @@ -0,0 +1,1857 @@ +2023/06/06 05:00:22 - mmengine - INFO - +------------------------------------------------------------ +System environment: + sys.platform: linux + Python: 3.10.9 (main, Mar 8 2023, 10:47:38) [GCC 11.2.0] + CUDA available: True + numpy_random_seed: 215766092 + GPU 0,1,2,3: NVIDIA A100-SXM4-80GB + CUDA_HOME: /mnt/petrelfs/share/cuda-11.6 + NVCC: Cuda compilation tools, release 11.6, V11.6.124 + GCC: gcc (GCC) 7.5.0 + PyTorch: 1.13.1 + PyTorch compiling details: PyTorch built with: + - GCC 9.3 + - C++ Version: 201402 + - Intel(R) oneAPI Math Kernel Library Version 2021.4-Product Build 20210904 for Intel(R) 64 architecture applications + - Intel(R) MKL-DNN v2.6.0 (Git Hash 52b5f107dd9cf10910aaa19cb47f3abf9b349815) + - OpenMP 201511 (a.k.a. OpenMP 4.5) + - LAPACK is enabled (usually provided by MKL) + - NNPACK is enabled + - CPU capability usage: AVX2 + - CUDA Runtime 11.6 + - NVCC architecture flags: -gencode;arch=compute_37,code=sm_37;-gencode;arch=compute_50,code=sm_50;-gencode;arch=compute_60,code=sm_60;-gencode;arch=compute_61,code=sm_61;-gencode;arch=compute_70,code=sm_70;-gencode;arch=compute_75,code=sm_75;-gencode;arch=compute_80,code=sm_80;-gencode;arch=compute_86,code=sm_86;-gencode;arch=compute_37,code=compute_37 + - CuDNN 8.3.2 (built against CUDA 11.5) + - Magma 2.6.1 + - Build settings: BLAS_INFO=mkl, BUILD_TYPE=Release, CUDA_VERSION=11.6, CUDNN_VERSION=8.3.2, CXX_COMPILER=/opt/rh/devtoolset-9/root/usr/bin/c++, CXX_FLAGS= -fabi-version=11 -Wno-deprecated -fvisibility-inlines-hidden -DUSE_PTHREADPOOL -fopenmp -DNDEBUG -DUSE_KINETO -DUSE_FBGEMM -DUSE_QNNPACK -DUSE_PYTORCH_QNNPACK -DUSE_XNNPACK -DSYMBOLICATE_MOBILE_DEBUG_HANDLE -DEDGE_PROFILER_USE_KINETO -O2 -fPIC -Wno-narrowing -Wall -Wextra -Werror=return-type -Werror=non-virtual-dtor -Wno-missing-field-initializers -Wno-type-limits -Wno-array-bounds -Wno-unknown-pragmas -Wunused-local-typedefs -Wno-unused-parameter -Wno-unused-function -Wno-unused-result -Wno-strict-overflow -Wno-strict-aliasing -Wno-error=deprecated-declarations -Wno-stringop-overflow -Wno-psabi -Wno-error=pedantic -Wno-error=redundant-decls -Wno-error=old-style-cast -fdiagnostics-color=always -faligned-new -Wno-unused-but-set-variable -Wno-maybe-uninitialized -fno-math-errno -fno-trapping-math -Werror=format -Werror=cast-function-type -Wno-stringop-overflow, LAPACK_INFO=mkl, PERF_WITH_AVX=1, PERF_WITH_AVX2=1, PERF_WITH_AVX512=1, TORCH_VERSION=1.13.1, USE_CUDA=ON, USE_CUDNN=ON, USE_EXCEPTION_PTR=1, USE_GFLAGS=OFF, USE_GLOG=OFF, USE_MKL=ON, USE_MKLDNN=ON, USE_MPI=OFF, USE_NCCL=ON, USE_NNPACK=ON, USE_OPENMP=ON, USE_ROCM=OFF, + + TorchVision: 0.14.1 + OpenCV: 4.7.0 + MMEngine: 0.7.3 + +Runtime environment: + cudnn_benchmark: True + mp_cfg: {'mp_start_method': 'fork', 'opencv_num_threads': 0} + dist_cfg: {'backend': 'nccl'} + seed: None + deterministic: False + Distributed launcher: slurm + Distributed training: True + GPU number: 4 +------------------------------------------------------------ + +2023/06/06 05:00:25 - mmengine - INFO - Config: +optim_wrapper = dict( + optimizer=dict( + type='AdamW', + lr=0.0004, + weight_decay=0.05, + eps=1e-08, + betas=(0.9, 0.999), + _scope_='mmpretrain'), + paramwise_cfg=dict( + norm_decay_mult=0.0, + bias_decay_mult=0.0, + flat_decay_mult=0.0, + custom_keys=dict({ + '.absolute_pos_embed': dict(decay_mult=0.0), + '.relative_position_bias_table': dict(decay_mult=0.0) + })), + type='AmpOptimWrapper', + dtype='bfloat16', + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=False, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=4096) +model = dict( + type='ImageClassifier', + backbone=dict( + frozen_stages=2, + type='ConvNeXt', + arch='small', + drop_path_rate=0.4, + init_cfg=dict( + type='Pretrained', + checkpoint= + 'https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth', + prefix='backbone')), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=768, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + init_cfg=None), + init_cfg=dict( + type='TruncNormal', layer=['Conv2d', 'Linear'], std=0.02, bias=0.0), + train_cfg=None) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/IF80w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/if-dpmsolver++-50-20w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/cc1m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/if-dpmsolver++-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/if-dpmsolver++-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/convnext_small_4xb256_IF_1m_lr4e-4_aug_1e-1' + +2023/06/06 05:00:35 - mmengine - INFO - Hooks will be executed in the following order: +before_run: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_load_checkpoint: +(ABOVE_NORMAL) EMAHook + -------------------- +before_train: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_train_epoch: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(NORMAL ) DistSamplerSeedHook + -------------------- +before_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook + -------------------- +after_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +after_train_epoch: +(NORMAL ) IterTimerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_val_epoch: +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook + -------------------- +before_val_iter: +(NORMAL ) IterTimerHook + -------------------- +after_val_iter: +(NORMAL ) IterTimerHook +(NORMAL ) VisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_val_epoch: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_save_checkpoint: +(ABOVE_NORMAL) EMAHook + -------------------- +after_train: +(VERY_LOW ) CheckpointHook + -------------------- +before_test_epoch: +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook + -------------------- +before_test_iter: +(NORMAL ) IterTimerHook + -------------------- +after_test_iter: +(NORMAL ) IterTimerHook +(NORMAL ) VisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_test_epoch: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_run: +(BELOW_NORMAL) LoggerHook + -------------------- +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.2.0.weight:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.2.0.bias:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.2.1.bias:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.3.0.weight:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.3.0.bias:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.3.1.bias:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.0.gamma:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.0.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.0.norm.weight:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.0.norm.bias:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.0.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.0.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.1.gamma:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.1.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.1.norm.weight:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.1.norm.bias:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.1.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.1.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.2.gamma:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.2.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.2.norm.weight:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.2.norm.bias:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.2.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.2.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.3.gamma:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.3.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.3.norm.weight:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.3.norm.bias:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.3.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.3.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.4.gamma:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.4.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.4.norm.weight:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.4.norm.bias:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.4.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.4.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.5.gamma:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.5.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.5.norm.weight:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.5.norm.bias:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.5.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.5.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.6.gamma:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.6.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.6.norm.weight:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.6.norm.bias:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.6.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.6.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.7.gamma:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.7.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.7.norm.weight:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.7.norm.bias:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.7.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.7.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.8.gamma:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.8.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.8.norm.weight:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.8.norm.bias:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.8.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.8.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.9.gamma:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.9.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.9.norm.weight:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.9.norm.bias:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.9.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.9.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.10.gamma:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.10.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.10.norm.weight:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.10.norm.bias:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.10.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.10.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.11.gamma:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.11.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.11.norm.weight:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.11.norm.bias:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.11.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.11.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.12.gamma:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.12.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.12.norm.weight:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.12.norm.bias:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.12.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.12.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.13.gamma:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.13.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.13.norm.weight:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.13.norm.bias:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.13.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.13.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.14.gamma:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.14.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.14.norm.weight:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.14.norm.bias:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.14.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.14.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.15.gamma:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.15.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.15.norm.weight:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.15.norm.bias:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.15.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.15.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.16.gamma:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.16.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.16.norm.weight:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.16.norm.bias:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.16.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.16.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.17.gamma:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.17.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.17.norm.weight:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.17.norm.bias:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.17.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.17.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.18.gamma:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.18.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.18.norm.weight:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.18.norm.bias:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.18.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.18.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.19.gamma:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.19.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.19.norm.weight:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.19.norm.bias:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.19.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.19.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.20.gamma:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.20.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.20.norm.weight:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.20.norm.bias:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.20.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.20.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.21.gamma:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.21.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.21.norm.weight:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.21.norm.bias:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.21.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.21.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.22.gamma:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.22.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.22.norm.weight:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.22.norm.bias:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.22.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.22.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.23.gamma:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.23.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.23.norm.weight:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.23.norm.bias:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.23.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.23.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.24.gamma:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.24.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.24.norm.weight:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.24.norm.bias:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.24.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.24.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.25.gamma:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.25.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.25.norm.weight:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.25.norm.bias:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.25.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.25.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.26.gamma:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.26.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.26.norm.weight:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.26.norm.bias:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.26.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.26.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.3.0.gamma:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.3.0.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.3.0.norm.weight:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.3.0.norm.bias:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.3.0.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.3.0.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.3.1.gamma:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.3.1.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.3.1.norm.weight:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.3.1.norm.bias:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.3.1.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.3.1.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.3.2.gamma:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.3.2.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.3.2.norm.weight:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.3.2.norm.bias:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.3.2.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.stages.3.2.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.norm3.weight:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- backbone.norm3.bias:weight_decay=0.0 +2023/06/06 05:00:48 - mmengine - INFO - paramwise_options -- head.fc.bias:weight_decay=0.0 +2023/06/06 05:00:52 - mmengine - INFO - load backbone in model from: https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth +Name of parameter - Initialization information + +backbone.downsample_layers.0.0.weight - torch.Size([96, 3, 4, 4]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.0.0.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.0.1.weight - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.0.1.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.1.0.weight - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.1.0.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.1.1.weight - torch.Size([192, 96, 2, 2]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.1.1.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.2.0.weight - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.2.0.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.2.1.weight - torch.Size([384, 192, 2, 2]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.2.1.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.3.0.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.3.0.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.3.1.weight - torch.Size([768, 384, 2, 2]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.3.1.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.0.gamma - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.0.depthwise_conv.weight - torch.Size([96, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.0.depthwise_conv.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.0.norm.weight - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.0.norm.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.0.pointwise_conv1.weight - torch.Size([384, 96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.0.pointwise_conv1.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.0.pointwise_conv2.weight - torch.Size([96, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.0.pointwise_conv2.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.1.gamma - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.1.depthwise_conv.weight - torch.Size([96, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.1.depthwise_conv.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.1.norm.weight - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.1.norm.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.1.pointwise_conv1.weight - torch.Size([384, 96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.1.pointwise_conv1.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.1.pointwise_conv2.weight - torch.Size([96, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.1.pointwise_conv2.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.2.gamma - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.2.depthwise_conv.weight - torch.Size([96, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.2.depthwise_conv.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.2.norm.weight - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.2.norm.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.2.pointwise_conv1.weight - torch.Size([384, 96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.2.pointwise_conv1.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.2.pointwise_conv2.weight - torch.Size([96, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.2.pointwise_conv2.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.0.gamma - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.0.depthwise_conv.weight - torch.Size([192, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.0.depthwise_conv.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.0.norm.weight - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.0.norm.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.0.pointwise_conv1.weight - torch.Size([768, 192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.0.pointwise_conv1.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.0.pointwise_conv2.weight - torch.Size([192, 768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.0.pointwise_conv2.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.1.gamma - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.1.depthwise_conv.weight - torch.Size([192, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.1.depthwise_conv.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.1.norm.weight - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.1.norm.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.1.pointwise_conv1.weight - torch.Size([768, 192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.1.pointwise_conv1.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.1.pointwise_conv2.weight - torch.Size([192, 768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.1.pointwise_conv2.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.2.gamma - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.2.depthwise_conv.weight - torch.Size([192, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.2.depthwise_conv.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.2.norm.weight - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.2.norm.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.2.pointwise_conv1.weight - torch.Size([768, 192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.2.pointwise_conv1.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.2.pointwise_conv2.weight - torch.Size([192, 768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.2.pointwise_conv2.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.0.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.0.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.0.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.0.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.0.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.0.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.0.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.0.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.0.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.1.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.1.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.1.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.1.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.1.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.1.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.1.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.1.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.1.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.2.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.2.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.2.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.2.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.2.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.2.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.2.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.2.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.2.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.3.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.3.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.3.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.3.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.3.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.3.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.3.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.3.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.3.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.4.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.4.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.4.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.4.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.4.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.4.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.4.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.4.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.4.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.5.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.5.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.5.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.5.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.5.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.5.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.5.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.5.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.5.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.6.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.6.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.6.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.6.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.6.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.6.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.6.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.6.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.6.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.7.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.7.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.7.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.7.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.7.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.7.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.7.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.7.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.7.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.8.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.8.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.8.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.8.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.8.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.8.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.8.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.8.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.8.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.9.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.9.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.9.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.9.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.9.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.9.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.9.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.9.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.9.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.10.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.10.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.10.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.10.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.10.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.10.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.10.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.10.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.10.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.11.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.11.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.11.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.11.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.11.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.11.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.11.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.11.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.11.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.12.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.12.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.12.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.12.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.12.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.12.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.12.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.12.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.12.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.13.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.13.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.13.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.13.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.13.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.13.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.13.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.13.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.13.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.14.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.14.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.14.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.14.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.14.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.14.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.14.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.14.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.14.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.15.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.15.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.15.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.15.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.15.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.15.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.15.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.15.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.15.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.16.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.16.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.16.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.16.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.16.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.16.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.16.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.16.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.16.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.17.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.17.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.17.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.17.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.17.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.17.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.17.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.17.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.17.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.18.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.18.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.18.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.18.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.18.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.18.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.18.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.18.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.18.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.19.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.19.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.19.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.19.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.19.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.19.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.19.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.19.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.19.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.20.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.20.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.20.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.20.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.20.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.20.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.20.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.20.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.20.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.21.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.21.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.21.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.21.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.21.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.21.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.21.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.21.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.21.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.22.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.22.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.22.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.22.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.22.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.22.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.22.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.22.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.22.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.23.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.23.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.23.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.23.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.23.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.23.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.23.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.23.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.23.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.24.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.24.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.24.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.24.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.24.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.24.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.24.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.24.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.24.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.25.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.25.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.25.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.25.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.25.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.25.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.25.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.25.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.25.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.26.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.26.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.26.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.26.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.26.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.26.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.26.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.26.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.26.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.0.gamma - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.0.depthwise_conv.weight - torch.Size([768, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.0.depthwise_conv.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.0.norm.weight - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.0.norm.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.0.pointwise_conv1.weight - torch.Size([3072, 768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.0.pointwise_conv1.bias - torch.Size([3072]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.0.pointwise_conv2.weight - torch.Size([768, 3072]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.0.pointwise_conv2.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.1.gamma - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.1.depthwise_conv.weight - torch.Size([768, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.1.depthwise_conv.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.1.norm.weight - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.1.norm.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.1.pointwise_conv1.weight - torch.Size([3072, 768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.1.pointwise_conv1.bias - torch.Size([3072]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.1.pointwise_conv2.weight - torch.Size([768, 3072]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.1.pointwise_conv2.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.2.gamma - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.2.depthwise_conv.weight - torch.Size([768, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.2.depthwise_conv.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.2.norm.weight - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.2.norm.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.2.pointwise_conv1.weight - torch.Size([3072, 768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.2.pointwise_conv1.bias - torch.Size([3072]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.2.pointwise_conv2.weight - torch.Size([768, 3072]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.2.pointwise_conv2.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.norm3.weight - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.norm3.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +head.fc.weight - torch.Size([2, 768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +head.fc.bias - torch.Size([2]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 +2023/06/06 05:00:53 - mmengine - WARNING - "FileClient" will be deprecated in future. Please use io functions in https://mmengine.readthedocs.io/en/latest/api/fileio.html#file-io +2023/06/06 05:00:53 - mmengine - WARNING - "HardDiskBackend" is the alias of "LocalBackend" and the former will be deprecated in future. +2023/06/06 05:00:53 - mmengine - INFO - Checkpoints will be saved to /mnt/petrelfs/luzeyu/workspace/fakebench/mmpretrain/workdir/convnext_small_4xb256_IF_1m_lr4e-4_aug_1e-1. +2023/06/06 05:01:53 - mmengine - INFO - Epoch(train) [1][ 100/1969] lr: 3.9998e-04 eta: 3:17:58 time: 0.5191 data_time: 0.0018 memory: 15578 loss: 0.0276 +2023/06/06 05:02:48 - mmengine - INFO - Epoch(train) [1][ 200/1969] lr: 3.9990e-04 eta: 3:07:02 time: 0.5323 data_time: 0.0019 memory: 15578 loss: 0.0272 +2023/06/06 05:03:40 - mmengine - INFO - Epoch(train) [1][ 300/1969] lr: 3.9978e-04 eta: 2:59:52 time: 0.5156 data_time: 0.0018 memory: 15578 loss: 0.0233 +2023/06/06 05:04:32 - mmengine - INFO - Epoch(train) [1][ 400/1969] lr: 3.9960e-04 eta: 2:56:05 time: 0.5271 data_time: 0.0020 memory: 15578 loss: 0.0200 +2023/06/06 05:05:24 - mmengine - INFO - Epoch(train) [1][ 500/1969] lr: 3.9938e-04 eta: 2:53:17 time: 0.5171 data_time: 0.0019 memory: 15578 loss: 0.0201 +2023/06/06 05:06:16 - mmengine - INFO - Epoch(train) [1][ 600/1969] lr: 3.9911e-04 eta: 2:51:14 time: 0.5175 data_time: 0.0021 memory: 15578 loss: 0.0169 +2023/06/06 05:07:07 - mmengine - INFO - Epoch(train) [1][ 700/1969] lr: 3.9879e-04 eta: 2:49:25 time: 0.5222 data_time: 0.0018 memory: 15578 loss: 0.0140 +2023/06/06 05:08:00 - mmengine - INFO - Epoch(train) [1][ 800/1969] lr: 3.9842e-04 eta: 2:47:56 time: 0.5156 data_time: 0.0016 memory: 15578 loss: 0.0139 +2023/06/06 05:08:51 - mmengine - INFO - Epoch(train) [1][ 900/1969] lr: 3.9800e-04 eta: 2:46:27 time: 0.5159 data_time: 0.0017 memory: 15578 loss: 0.0138 +2023/06/06 05:09:43 - mmengine - INFO - Exp name: convnext_small_4xb256_IF_1m_lr4e-4_aug_1e-1_20230606_050018 +2023/06/06 05:09:43 - mmengine - INFO - Epoch(train) [1][1000/1969] lr: 3.9753e-04 eta: 2:45:12 time: 0.5156 data_time: 0.0019 memory: 15578 loss: 0.0190 +2023/06/06 05:10:35 - mmengine - INFO - Epoch(train) [1][1100/1969] lr: 3.9701e-04 eta: 2:43:56 time: 0.5165 data_time: 0.0017 memory: 15578 loss: 0.0113 +2023/06/06 05:11:27 - mmengine - INFO - Epoch(train) [1][1200/1969] lr: 3.9644e-04 eta: 2:42:46 time: 0.5162 data_time: 0.0017 memory: 15578 loss: 0.0147 +2023/06/06 05:12:18 - mmengine - INFO - Epoch(train) [1][1300/1969] lr: 3.9583e-04 eta: 2:41:38 time: 0.5205 data_time: 0.0017 memory: 15578 loss: 0.0091 +2023/06/06 05:13:10 - mmengine - INFO - Epoch(train) [1][1400/1969] lr: 3.9516e-04 eta: 2:40:34 time: 0.5156 data_time: 0.0016 memory: 15578 loss: 0.0160 +2023/06/06 05:14:02 - mmengine - INFO - Epoch(train) [1][1500/1969] lr: 3.9445e-04 eta: 2:39:35 time: 0.5266 data_time: 0.0017 memory: 15578 loss: 0.0193 +2023/06/06 05:14:54 - mmengine - INFO - Epoch(train) [1][1600/1969] lr: 3.9369e-04 eta: 2:38:34 time: 0.5160 data_time: 0.0016 memory: 15578 loss: 0.0182 +2023/06/06 05:15:46 - mmengine - INFO - Epoch(train) [1][1700/1969] lr: 3.9288e-04 eta: 2:37:35 time: 0.5178 data_time: 0.0016 memory: 15578 loss: 0.0126 +2023/06/06 05:16:38 - mmengine - INFO - Epoch(train) [1][1800/1969] lr: 3.9202e-04 eta: 2:36:36 time: 0.5148 data_time: 0.0017 memory: 15578 loss: 0.0199 +2023/06/06 05:17:30 - mmengine - INFO - Epoch(train) [1][1900/1969] lr: 3.9112e-04 eta: 2:35:36 time: 0.5147 data_time: 0.0015 memory: 15578 loss: 0.0129 +2023/06/06 05:18:05 - mmengine - INFO - Exp name: convnext_small_4xb256_IF_1m_lr4e-4_aug_1e-1_20230606_050018 +2023/06/06 05:18:05 - mmengine - INFO - Saving checkpoint at 1 epochs +2023/06/06 05:18:38 - mmengine - INFO - Epoch(val) [1][29/29] accuracy/top1: 96.2527 single-label/precision_classwise: [95.33283996582031, 97.45524597167969] single-label/recall_classwise: [97.99886322021484, 94.10852813720703] single-label/f1-score_classwise: [96.64746856689453, 95.75265502929688] data_time: 0.1395 time: 0.6055 +2023/06/06 05:19:04 - mmengine - INFO - Exp name: convnext_small_4xb256_IF_1m_lr4e-4_aug_1e-1_20230606_050018 +2023/06/06 05:19:41 - mmengine - INFO - Epoch(train) [2][ 100/1969] lr: 3.8948e-04 eta: 2:35:31 time: 0.5443 data_time: 0.0289 memory: 15578 loss: 0.0155 +2023/06/06 05:20:42 - mmengine - INFO - Epoch(train) [2][ 200/1969] lr: 3.8845e-04 eta: 2:35:45 time: 0.5310 data_time: 0.0016 memory: 15575 loss: 0.0102 +2023/06/06 05:21:36 - mmengine - INFO - Epoch(train) [2][ 300/1969] lr: 3.8737e-04 eta: 2:34:54 time: 0.5833 data_time: 0.0356 memory: 15575 loss: 0.0151 +2023/06/06 05:22:30 - mmengine - INFO - Epoch(train) [2][ 400/1969] lr: 3.8625e-04 eta: 2:34:07 time: 0.5234 data_time: 0.0017 memory: 15575 loss: 0.0146 +2023/06/06 05:23:23 - mmengine - INFO - Epoch(train) [2][ 500/1969] lr: 3.8508e-04 eta: 2:33:14 time: 0.5206 data_time: 0.0016 memory: 15575 loss: 0.0144 +2023/06/06 05:24:22 - mmengine - INFO - Epoch(train) [2][ 600/1969] lr: 3.8386e-04 eta: 2:32:57 time: 0.5145 data_time: 0.0016 memory: 15575 loss: 0.0123 +2023/06/06 05:25:27 - mmengine - INFO - Epoch(train) [2][ 700/1969] lr: 3.8260e-04 eta: 2:33:16 time: 1.0784 data_time: 0.2809 memory: 15575 loss: 0.0136 +2023/06/06 05:26:21 - mmengine - INFO - Epoch(train) [2][ 800/1969] lr: 3.8129e-04 eta: 2:32:22 time: 0.6251 data_time: 0.0018 memory: 15575 loss: 0.0075 +2023/06/06 05:27:14 - mmengine - INFO - Epoch(train) [2][ 900/1969] lr: 3.7994e-04 eta: 2:31:23 time: 0.5329 data_time: 0.0015 memory: 15575 loss: 0.0121 +2023/06/06 05:28:08 - mmengine - INFO - Epoch(train) [2][1000/1969] lr: 3.7854e-04 eta: 2:30:29 time: 0.7000 data_time: 0.0015 memory: 15575 loss: 0.0114 +2023/06/06 05:28:25 - mmengine - INFO - Exp name: convnext_small_4xb256_IF_1m_lr4e-4_aug_1e-1_20230606_050018 +2023/06/06 05:29:03 - mmengine - INFO - Epoch(train) [2][1100/1969] lr: 3.7710e-04 eta: 2:29:37 time: 0.5768 data_time: 0.0016 memory: 15575 loss: 0.0136 +2023/06/06 05:29:56 - mmengine - INFO - Epoch(train) [2][1200/1969] lr: 3.7562e-04 eta: 2:28:38 time: 0.5170 data_time: 0.0016 memory: 15575 loss: 0.0149 +2023/06/06 05:30:51 - mmengine - INFO - Epoch(train) [2][1300/1969] lr: 3.7409e-04 eta: 2:27:48 time: 0.5152 data_time: 0.0015 memory: 15575 loss: 0.0066 +2023/06/06 05:31:44 - mmengine - INFO - Epoch(train) [2][1400/1969] lr: 3.7252e-04 eta: 2:26:50 time: 0.5232 data_time: 0.0017 memory: 15575 loss: 0.0142 +2023/06/06 05:32:38 - mmengine - INFO - Epoch(train) [2][1500/1969] lr: 3.7090e-04 eta: 2:25:56 time: 0.5400 data_time: 0.0259 memory: 15575 loss: 0.0122 +2023/06/06 05:33:31 - mmengine - INFO - Epoch(train) [2][1600/1969] lr: 3.6925e-04 eta: 2:25:00 time: 0.5146 data_time: 0.0015 memory: 15575 loss: 0.0069 +2023/06/06 05:35:56 - mmengine - INFO - Epoch(train) [2][1700/1969] lr: 3.6755e-04 eta: 2:30:44 time: 9.8493 data_time: 9.3115 memory: 15575 loss: 0.0122 +2023/06/06 05:36:50 - mmengine - INFO - Epoch(train) [2][1800/1969] lr: 3.6581e-04 eta: 2:29:37 time: 0.5151 data_time: 0.0015 memory: 15575 loss: 0.0142 +2023/06/06 05:37:44 - mmengine - INFO - Epoch(train) [2][1900/1969] lr: 3.6403e-04 eta: 2:28:30 time: 0.5575 data_time: 0.0017 memory: 15575 loss: 0.0105 +2023/06/06 05:38:21 - mmengine - INFO - Exp name: convnext_small_4xb256_IF_1m_lr4e-4_aug_1e-1_20230606_050018 +2023/06/06 05:38:21 - mmengine - INFO - Saving checkpoint at 2 epochs +2023/06/06 05:38:52 - mmengine - INFO - Epoch(val) [2][29/29] accuracy/top1: 98.8031 single-label/precision_classwise: [98.09446716308594, 99.70704650878906] single-label/recall_classwise: [99.76642608642578, 97.62015533447266] single-label/f1-score_classwise: [98.92337799072266, 98.65255737304688] data_time: 0.1627 time: 0.5490 +2023/06/06 05:39:34 - mmengine - INFO - Exp name: convnext_small_4xb256_IF_1m_lr4e-4_aug_1e-1_20230606_050018 +2023/06/06 05:39:54 - mmengine - INFO - Epoch(train) [3][ 100/1969] lr: 3.6093e-04 eta: 2:27:09 time: 0.5472 data_time: 0.0015 memory: 15575 loss: 0.0087 +2023/06/06 05:40:49 - mmengine - INFO - Epoch(train) [3][ 200/1969] lr: 3.5904e-04 eta: 2:26:05 time: 0.5169 data_time: 0.0020 memory: 15575 loss: 0.0096 +2023/06/06 05:41:41 - mmengine - INFO - Epoch(train) [3][ 300/1969] lr: 3.5711e-04 eta: 2:24:52 time: 0.5255 data_time: 0.0014 memory: 15575 loss: 0.0089 +2023/06/06 05:42:34 - mmengine - INFO - Epoch(train) [3][ 400/1969] lr: 3.5515e-04 eta: 2:23:46 time: 0.5239 data_time: 0.0015 memory: 15575 loss: 0.0153 +2023/06/06 05:43:27 - mmengine - INFO - Epoch(train) [3][ 500/1969] lr: 3.5314e-04 eta: 2:22:39 time: 0.5234 data_time: 0.0015 memory: 15575 loss: 0.0166 +2023/06/06 05:44:20 - mmengine - INFO - Epoch(train) [3][ 600/1969] lr: 3.5110e-04 eta: 2:21:32 time: 0.5160 data_time: 0.0014 memory: 15575 loss: 0.0132 +2023/06/06 05:45:12 - mmengine - INFO - Epoch(train) [3][ 700/1969] lr: 3.4902e-04 eta: 2:20:23 time: 0.5216 data_time: 0.0014 memory: 15575 loss: 0.0061 +2023/06/06 05:46:05 - mmengine - INFO - Epoch(train) [3][ 800/1969] lr: 3.4691e-04 eta: 2:19:18 time: 0.5343 data_time: 0.0016 memory: 15575 loss: 0.0122 +2023/06/06 05:46:58 - mmengine - INFO - Epoch(train) [3][ 900/1969] lr: 3.4475e-04 eta: 2:18:14 time: 0.5542 data_time: 0.0017 memory: 15575 loss: 0.0134 +2023/06/06 05:47:52 - mmengine - INFO - Epoch(train) [3][1000/1969] lr: 3.4257e-04 eta: 2:17:13 time: 0.5288 data_time: 0.0016 memory: 15575 loss: 0.0143 +2023/06/06 05:48:26 - mmengine - INFO - Exp name: convnext_small_4xb256_IF_1m_lr4e-4_aug_1e-1_20230606_050018 +2023/06/06 05:48:46 - mmengine - INFO - Epoch(train) [3][1100/1969] lr: 3.4034e-04 eta: 2:16:11 time: 0.5186 data_time: 0.0016 memory: 15575 loss: 0.0137 +2023/06/06 05:49:39 - mmengine - INFO - Epoch(train) [3][1200/1969] lr: 3.3809e-04 eta: 2:15:07 time: 0.5228 data_time: 0.0020 memory: 15575 loss: 0.0126 +2023/06/06 05:50:32 - mmengine - INFO - Epoch(train) [3][1300/1969] lr: 3.3580e-04 eta: 2:14:04 time: 0.5421 data_time: 0.0016 memory: 15575 loss: 0.0049 +2023/06/06 05:51:26 - mmengine - INFO - Epoch(train) [3][1400/1969] lr: 3.3347e-04 eta: 2:13:04 time: 0.5370 data_time: 0.0015 memory: 15575 loss: 0.0121 +2023/06/06 05:52:18 - mmengine - INFO - Epoch(train) [3][1500/1969] lr: 3.3112e-04 eta: 2:11:59 time: 0.5184 data_time: 0.0016 memory: 15575 loss: 0.0083 +2023/06/06 05:53:11 - mmengine - INFO - Epoch(train) [3][1600/1969] lr: 3.2873e-04 eta: 2:10:55 time: 0.5149 data_time: 0.0016 memory: 15575 loss: 0.0123 +2023/06/06 05:54:03 - mmengine - INFO - Epoch(train) [3][1700/1969] lr: 3.2631e-04 eta: 2:09:52 time: 0.5159 data_time: 0.0017 memory: 15575 loss: 0.0152 +2023/06/06 05:55:08 - mmengine - INFO - Epoch(train) [3][1800/1969] lr: 3.2385e-04 eta: 2:09:19 time: 0.5275 data_time: 0.0146 memory: 15575 loss: 0.0131 +2023/06/06 05:56:00 - mmengine - INFO - Epoch(train) [3][1900/1969] lr: 3.2137e-04 eta: 2:08:15 time: 0.5170 data_time: 0.0015 memory: 15575 loss: 0.0084 +2023/06/06 05:56:35 - mmengine - INFO - Exp name: convnext_small_4xb256_IF_1m_lr4e-4_aug_1e-1_20230606_050018 +2023/06/06 05:56:35 - mmengine - INFO - Saving checkpoint at 3 epochs +2023/06/06 05:57:03 - mmengine - INFO - Epoch(val) [3][29/29] accuracy/top1: 99.2485 single-label/precision_classwise: [98.72762298583984, 99.90557098388672] single-label/recall_classwise: [99.92424774169922, 98.4186019897461] single-label/f1-score_classwise: [99.32232666015625, 99.15650939941406] data_time: 0.0524 time: 0.4355 +2023/06/06 05:57:56 - mmengine - INFO - Exp name: convnext_small_4xb256_IF_1m_lr4e-4_aug_1e-1_20230606_050018 +2023/06/06 05:58:00 - mmengine - INFO - Epoch(train) [4][ 100/1969] lr: 3.1711e-04 eta: 2:06:37 time: 0.5155 data_time: 0.0017 memory: 15575 loss: 0.0074 +2023/06/06 05:58:52 - mmengine - INFO - Epoch(train) [4][ 200/1969] lr: 3.1455e-04 eta: 2:05:33 time: 0.5166 data_time: 0.0014 memory: 15575 loss: 0.0124 +2023/06/06 05:59:44 - mmengine - INFO - Epoch(train) [4][ 300/1969] lr: 3.1196e-04 eta: 2:04:30 time: 0.5158 data_time: 0.0016 memory: 15575 loss: 0.0079 +2023/06/06 06:00:35 - mmengine - INFO - Epoch(train) [4][ 400/1969] lr: 3.0935e-04 eta: 2:03:27 time: 0.5182 data_time: 0.0016 memory: 15575 loss: 0.0116 +2023/06/06 06:01:27 - mmengine - INFO - Epoch(train) [4][ 500/1969] lr: 3.0671e-04 eta: 2:02:24 time: 0.5150 data_time: 0.0015 memory: 15575 loss: 0.0080 +2023/06/06 06:02:19 - mmengine - INFO - Epoch(train) [4][ 600/1969] lr: 3.0404e-04 eta: 2:01:22 time: 0.5186 data_time: 0.0014 memory: 15575 loss: 0.0074 +2023/06/06 06:03:11 - mmengine - INFO - Epoch(train) [4][ 700/1969] lr: 3.0135e-04 eta: 2:00:20 time: 0.5151 data_time: 0.0016 memory: 15575 loss: 0.0131 +2023/06/06 06:04:03 - mmengine - INFO - Epoch(train) [4][ 800/1969] lr: 2.9863e-04 eta: 1:59:18 time: 0.5164 data_time: 0.0016 memory: 15575 loss: 0.0086 +2023/06/06 06:04:54 - mmengine - INFO - Epoch(train) [4][ 900/1969] lr: 2.9589e-04 eta: 1:58:17 time: 0.5147 data_time: 0.0016 memory: 15575 loss: 0.0058 +2023/06/06 06:05:46 - mmengine - INFO - Epoch(train) [4][1000/1969] lr: 2.9313e-04 eta: 1:57:16 time: 0.5156 data_time: 0.0015 memory: 15575 loss: 0.0094 +2023/06/06 06:06:35 - mmengine - INFO - Exp name: convnext_small_4xb256_IF_1m_lr4e-4_aug_1e-1_20230606_050018 +2023/06/06 06:06:38 - mmengine - INFO - Epoch(train) [4][1100/1969] lr: 2.9034e-04 eta: 1:56:15 time: 0.5153 data_time: 0.0016 memory: 15575 loss: 0.0115 +2023/06/06 06:07:30 - mmengine - INFO - Epoch(train) [4][1200/1969] lr: 2.8753e-04 eta: 1:55:14 time: 0.5146 data_time: 0.0015 memory: 15575 loss: 0.0070 +2023/06/06 06:08:21 - mmengine - INFO - Epoch(train) [4][1300/1969] lr: 2.8470e-04 eta: 1:54:13 time: 0.5149 data_time: 0.0014 memory: 15575 loss: 0.0113 +2023/06/06 06:09:13 - mmengine - INFO - Epoch(train) [4][1400/1969] lr: 2.8185e-04 eta: 1:53:13 time: 0.5161 data_time: 0.0016 memory: 15575 loss: 0.0062 +2023/06/06 06:10:05 - mmengine - INFO - Epoch(train) [4][1500/1969] lr: 2.7898e-04 eta: 1:52:13 time: 0.5146 data_time: 0.0016 memory: 15575 loss: 0.0064 +2023/06/06 06:10:57 - mmengine - INFO - Epoch(train) [4][1600/1969] lr: 2.7610e-04 eta: 1:51:13 time: 0.5230 data_time: 0.0020 memory: 15575 loss: 0.0088 +2023/06/06 06:11:48 - mmengine - INFO - Epoch(train) [4][1700/1969] lr: 2.7319e-04 eta: 1:50:14 time: 0.5147 data_time: 0.0015 memory: 15575 loss: 0.0099 +2023/06/06 06:12:40 - mmengine - INFO - Epoch(train) [4][1800/1969] lr: 2.7027e-04 eta: 1:49:14 time: 0.5196 data_time: 0.0015 memory: 15575 loss: 0.0146 +2023/06/06 06:13:32 - mmengine - INFO - Epoch(train) [4][1900/1969] lr: 2.6733e-04 eta: 1:48:15 time: 0.5275 data_time: 0.0014 memory: 15575 loss: 0.0130 +2023/06/06 06:14:07 - mmengine - INFO - Exp name: convnext_small_4xb256_IF_1m_lr4e-4_aug_1e-1_20230606_050018 +2023/06/06 06:14:07 - mmengine - INFO - Saving checkpoint at 4 epochs +2023/06/06 06:14:38 - mmengine - INFO - Epoch(val) [4][29/29] accuracy/top1: 99.3772 single-label/precision_classwise: [98.95598602294922, 99.90584564208984] single-label/recall_classwise: [99.92424774169922, 98.70542907714844] single-label/f1-score_classwise: [99.43775939941406, 99.30200958251953] data_time: 0.1419 time: 0.5246 +2023/06/06 06:15:35 - mmengine - INFO - Epoch(train) [5][ 100/1969] lr: 2.6232e-04 eta: 1:46:43 time: 0.5156 data_time: 0.0016 memory: 15575 loss: 0.0123 +2023/06/06 06:15:47 - mmengine - INFO - Exp name: convnext_small_4xb256_IF_1m_lr4e-4_aug_1e-1_20230606_050018 +2023/06/06 06:16:27 - mmengine - INFO - Epoch(train) [5][ 200/1969] lr: 2.5934e-04 eta: 1:45:44 time: 0.5150 data_time: 0.0016 memory: 15575 loss: 0.0134 +2023/06/06 06:17:18 - mmengine - INFO - Epoch(train) [5][ 300/1969] lr: 2.5635e-04 eta: 1:44:45 time: 0.5153 data_time: 0.0016 memory: 15575 loss: 0.0065 +2023/06/06 06:18:10 - mmengine - INFO - Epoch(train) [5][ 400/1969] lr: 2.5334e-04 eta: 1:43:47 time: 0.5196 data_time: 0.0016 memory: 15575 loss: 0.0108 +2023/06/06 06:19:02 - mmengine - INFO - Epoch(train) [5][ 500/1969] lr: 2.5032e-04 eta: 1:42:48 time: 0.5149 data_time: 0.0015 memory: 15575 loss: 0.0088 +2023/06/06 06:19:54 - mmengine - INFO - Epoch(train) [5][ 600/1969] lr: 2.4729e-04 eta: 1:41:50 time: 0.5263 data_time: 0.0015 memory: 15575 loss: 0.0063 +2023/06/06 06:20:46 - mmengine - INFO - Epoch(train) [5][ 700/1969] lr: 2.4424e-04 eta: 1:40:52 time: 0.5154 data_time: 0.0015 memory: 15575 loss: 0.0079 +2023/06/06 06:21:37 - mmengine - INFO - Epoch(train) [5][ 800/1969] lr: 2.4119e-04 eta: 1:39:54 time: 0.5215 data_time: 0.0015 memory: 15575 loss: 0.0048 +2023/06/06 06:22:29 - mmengine - INFO - Epoch(train) [5][ 900/1969] lr: 2.3813e-04 eta: 1:38:57 time: 0.5149 data_time: 0.0015 memory: 15575 loss: 0.0070 +2023/06/06 06:23:21 - mmengine - INFO - Epoch(train) [5][1000/1969] lr: 2.3506e-04 eta: 1:37:59 time: 0.5176 data_time: 0.0015 memory: 15575 loss: 0.0110 +2023/06/06 06:24:13 - mmengine - INFO - Epoch(train) [5][1100/1969] lr: 2.3198e-04 eta: 1:37:01 time: 0.5148 data_time: 0.0015 memory: 15575 loss: 0.0076 +2023/06/06 06:24:25 - mmengine - INFO - Exp name: convnext_small_4xb256_IF_1m_lr4e-4_aug_1e-1_20230606_050018 +2023/06/06 06:25:04 - mmengine - INFO - Epoch(train) [5][1200/1969] lr: 2.2890e-04 eta: 1:36:04 time: 0.5237 data_time: 0.0016 memory: 15575 loss: 0.0087 +2023/06/06 06:25:56 - mmengine - INFO - Epoch(train) [5][1300/1969] lr: 2.2581e-04 eta: 1:35:07 time: 0.5152 data_time: 0.0015 memory: 15575 loss: 0.0083 +2023/06/06 06:26:48 - mmengine - INFO - Epoch(train) [5][1400/1969] lr: 2.2271e-04 eta: 1:34:09 time: 0.5195 data_time: 0.0016 memory: 15575 loss: 0.0099 +2023/06/06 06:27:39 - mmengine - INFO - Epoch(train) [5][1500/1969] lr: 2.1961e-04 eta: 1:33:12 time: 0.5150 data_time: 0.0017 memory: 15575 loss: 0.0069 +2023/06/06 06:28:31 - mmengine - INFO - Epoch(train) [5][1600/1969] lr: 2.1651e-04 eta: 1:32:16 time: 0.5200 data_time: 0.0017 memory: 15575 loss: 0.0072 +2023/06/06 06:29:23 - mmengine - INFO - Epoch(train) [5][1700/1969] lr: 2.1340e-04 eta: 1:31:19 time: 0.5150 data_time: 0.0015 memory: 15575 loss: 0.0115 +2023/06/06 06:30:15 - mmengine - INFO - Epoch(train) [5][1800/1969] lr: 2.1029e-04 eta: 1:30:22 time: 0.5204 data_time: 0.0014 memory: 15575 loss: 0.0060 +2023/06/06 06:31:07 - mmengine - INFO - Epoch(train) [5][1900/1969] lr: 2.0718e-04 eta: 1:29:26 time: 0.5229 data_time: 0.0015 memory: 15575 loss: 0.0060 +2023/06/06 06:31:42 - mmengine - INFO - Exp name: convnext_small_4xb256_IF_1m_lr4e-4_aug_1e-1_20230606_050018 +2023/06/06 06:31:42 - mmengine - INFO - Saving checkpoint at 5 epochs +2023/06/06 06:32:11 - mmengine - INFO - Epoch(val) [5][29/29] accuracy/top1: 99.5616 single-label/precision_classwise: [99.2911376953125, 99.8984375] single-label/recall_classwise: [99.91793823242188, 99.12403106689453] single-label/f1-score_classwise: [99.60355377197266, 99.50973510742188] data_time: 0.0743 time: 0.4616 +2023/06/06 06:33:09 - mmengine - INFO - Epoch(train) [6][ 100/1969] lr: 2.0192e-04 eta: 1:27:56 time: 0.5155 data_time: 0.0017 memory: 15575 loss: 0.0088 +2023/06/06 06:33:37 - mmengine - INFO - Exp name: convnext_small_4xb256_IF_1m_lr4e-4_aug_1e-1_20230606_050018 +2023/06/06 06:34:01 - mmengine - INFO - Epoch(train) [6][ 200/1969] lr: 1.9881e-04 eta: 1:27:00 time: 0.5151 data_time: 0.0017 memory: 15575 loss: 0.0091 +2023/06/06 06:34:53 - mmengine - INFO - Epoch(train) [6][ 300/1969] lr: 1.9570e-04 eta: 1:26:03 time: 0.5162 data_time: 0.0017 memory: 15575 loss: 0.0115 +2023/06/06 06:35:44 - mmengine - INFO - Epoch(train) [6][ 400/1969] lr: 1.9259e-04 eta: 1:25:07 time: 0.5160 data_time: 0.0018 memory: 15575 loss: 0.0093 +2023/06/06 06:36:42 - mmengine - INFO - Epoch(train) [6][ 500/1969] lr: 1.8949e-04 eta: 1:24:16 time: 0.5230 data_time: 0.0016 memory: 15575 loss: 0.0096 +2023/06/06 06:37:33 - mmengine - INFO - Epoch(train) [6][ 600/1969] lr: 1.8639e-04 eta: 1:23:20 time: 0.5279 data_time: 0.0015 memory: 15575 loss: 0.0081 +2023/06/06 06:38:25 - mmengine - INFO - Epoch(train) [6][ 700/1969] lr: 1.8330e-04 eta: 1:22:23 time: 0.5157 data_time: 0.0014 memory: 15575 loss: 0.0115 +2023/06/06 06:39:17 - mmengine - INFO - Epoch(train) [6][ 800/1969] lr: 1.8021e-04 eta: 1:21:28 time: 0.5202 data_time: 0.0019 memory: 15575 loss: 0.0095 +2023/06/06 06:40:09 - mmengine - INFO - Epoch(train) [6][ 900/1969] lr: 1.7713e-04 eta: 1:20:32 time: 0.5150 data_time: 0.0016 memory: 15575 loss: 0.0050 +2023/06/06 06:41:01 - mmengine - INFO - Epoch(train) [6][1000/1969] lr: 1.7405e-04 eta: 1:19:36 time: 0.5304 data_time: 0.0017 memory: 15575 loss: 0.0065 +2023/06/06 06:41:53 - mmengine - INFO - Epoch(train) [6][1100/1969] lr: 1.7098e-04 eta: 1:18:40 time: 0.5151 data_time: 0.0017 memory: 15575 loss: 0.0081 +2023/06/06 06:42:21 - mmengine - INFO - Exp name: convnext_small_4xb256_IF_1m_lr4e-4_aug_1e-1_20230606_050018 +2023/06/06 06:42:45 - mmengine - INFO - Epoch(train) [6][1200/1969] lr: 1.6792e-04 eta: 1:17:45 time: 0.5237 data_time: 0.0019 memory: 15575 loss: 0.0069 +2023/06/06 06:43:36 - mmengine - INFO - Epoch(train) [6][1300/1969] lr: 1.6487e-04 eta: 1:16:49 time: 0.5151 data_time: 0.0017 memory: 15575 loss: 0.0066 +2023/06/06 06:44:28 - mmengine - INFO - Epoch(train) [6][1400/1969] lr: 1.6183e-04 eta: 1:15:53 time: 0.5165 data_time: 0.0018 memory: 15575 loss: 0.0040 +2023/06/06 06:45:20 - mmengine - INFO - Epoch(train) [6][1500/1969] lr: 1.5881e-04 eta: 1:14:58 time: 0.5148 data_time: 0.0017 memory: 15575 loss: 0.0070 +2023/06/06 06:46:12 - mmengine - INFO - Epoch(train) [6][1600/1969] lr: 1.5579e-04 eta: 1:14:03 time: 0.5205 data_time: 0.0018 memory: 15575 loss: 0.0062 +2023/06/06 06:47:04 - mmengine - INFO - Epoch(train) [6][1700/1969] lr: 1.5278e-04 eta: 1:13:07 time: 0.5165 data_time: 0.0015 memory: 15575 loss: 0.0080 +2023/06/06 06:47:55 - mmengine - INFO - Epoch(train) [6][1800/1969] lr: 1.4979e-04 eta: 1:12:12 time: 0.5187 data_time: 0.0017 memory: 15575 loss: 0.0077 +2023/06/06 06:48:47 - mmengine - INFO - Epoch(train) [6][1900/1969] lr: 1.4682e-04 eta: 1:11:17 time: 0.5151 data_time: 0.0017 memory: 15575 loss: 0.0091 +2023/06/06 06:49:23 - mmengine - INFO - Exp name: convnext_small_4xb256_IF_1m_lr4e-4_aug_1e-1_20230606_050018 +2023/06/06 06:49:23 - mmengine - INFO - Saving checkpoint at 6 epochs +2023/06/06 06:49:55 - mmengine - INFO - Epoch(val) [6][29/29] accuracy/top1: 99.6555 single-label/precision_classwise: [99.45337677001953, 99.90643310546875] single-label/recall_classwise: [99.92424774169922, 99.32557678222656] single-label/f1-score_classwise: [99.6882553100586, 99.61515045166016] data_time: 0.1735 time: 0.5614 +2023/06/06 06:50:52 - mmengine - INFO - Epoch(train) [7][ 100/1969] lr: 1.4182e-04 eta: 1:09:47 time: 0.5168 data_time: 0.0017 memory: 15575 loss: 0.0114 +2023/06/06 06:51:37 - mmengine - INFO - Exp name: convnext_small_4xb256_IF_1m_lr4e-4_aug_1e-1_20230606_050018 +2023/06/06 06:51:44 - mmengine - INFO - Epoch(train) [7][ 200/1969] lr: 1.3888e-04 eta: 1:08:52 time: 0.5173 data_time: 0.0017 memory: 15575 loss: 0.0043 +2023/06/06 06:52:36 - mmengine - INFO - Epoch(train) [7][ 300/1969] lr: 1.3597e-04 eta: 1:07:57 time: 0.5162 data_time: 0.0015 memory: 15575 loss: 0.0104 +2023/06/06 06:53:28 - mmengine - INFO - Epoch(train) [7][ 400/1969] lr: 1.3307e-04 eta: 1:07:02 time: 0.5148 data_time: 0.0015 memory: 15575 loss: 0.0077 +2023/06/06 06:54:20 - mmengine - INFO - Epoch(train) [7][ 500/1969] lr: 1.3018e-04 eta: 1:06:07 time: 0.5159 data_time: 0.0014 memory: 15575 loss: 0.0058 +2023/06/06 06:55:12 - mmengine - INFO - Epoch(train) [7][ 600/1969] lr: 1.2732e-04 eta: 1:05:12 time: 0.5161 data_time: 0.0015 memory: 15575 loss: 0.0089 +2023/06/06 06:56:04 - mmengine - INFO - Epoch(train) [7][ 700/1969] lr: 1.2448e-04 eta: 1:04:17 time: 0.5153 data_time: 0.0014 memory: 15575 loss: 0.0055 +2023/06/06 06:56:55 - mmengine - INFO - Epoch(train) [7][ 800/1969] lr: 1.2165e-04 eta: 1:03:22 time: 0.5157 data_time: 0.0015 memory: 15575 loss: 0.0048 +2023/06/06 06:57:47 - mmengine - INFO - Epoch(train) [7][ 900/1969] lr: 1.1885e-04 eta: 1:02:27 time: 0.5161 data_time: 0.0015 memory: 15575 loss: 0.0044 +2023/06/06 06:58:39 - mmengine - INFO - Epoch(train) [7][1000/1969] lr: 1.1607e-04 eta: 1:01:33 time: 0.5158 data_time: 0.0016 memory: 15575 loss: 0.0044 +2023/06/06 06:59:31 - mmengine - INFO - Epoch(train) [7][1100/1969] lr: 1.1331e-04 eta: 1:00:38 time: 0.5146 data_time: 0.0015 memory: 15575 loss: 0.0060 +2023/06/06 07:00:15 - mmengine - INFO - Exp name: convnext_small_4xb256_IF_1m_lr4e-4_aug_1e-1_20230606_050018 +2023/06/06 07:00:23 - mmengine - INFO - Epoch(train) [7][1200/1969] lr: 1.1058e-04 eta: 0:59:43 time: 0.5149 data_time: 0.0016 memory: 15575 loss: 0.0036 +2023/06/06 07:01:15 - mmengine - INFO - Epoch(train) [7][1300/1969] lr: 1.0787e-04 eta: 0:58:49 time: 0.5153 data_time: 0.0016 memory: 15575 loss: 0.0099 +2023/06/06 07:02:06 - mmengine - INFO - Epoch(train) [7][1400/1969] lr: 1.0518e-04 eta: 0:57:54 time: 0.5147 data_time: 0.0016 memory: 15575 loss: 0.0070 +2023/06/06 07:02:58 - mmengine - INFO - Epoch(train) [7][1500/1969] lr: 1.0252e-04 eta: 0:57:00 time: 0.5151 data_time: 0.0017 memory: 15575 loss: 0.0126 +2023/06/06 07:03:50 - mmengine - INFO - Epoch(train) [7][1600/1969] lr: 9.9889e-05 eta: 0:56:05 time: 0.5149 data_time: 0.0016 memory: 15575 loss: 0.0054 +2023/06/06 07:04:42 - mmengine - INFO - Epoch(train) [7][1700/1969] lr: 9.7282e-05 eta: 0:55:11 time: 0.5155 data_time: 0.0015 memory: 15575 loss: 0.0079 +2023/06/06 07:05:33 - mmengine - INFO - Epoch(train) [7][1800/1969] lr: 9.4703e-05 eta: 0:54:16 time: 0.5147 data_time: 0.0015 memory: 15575 loss: 0.0058 +2023/06/06 07:06:25 - mmengine - INFO - Epoch(train) [7][1900/1969] lr: 9.2151e-05 eta: 0:53:22 time: 0.5150 data_time: 0.0015 memory: 15575 loss: 0.0061 +2023/06/06 07:07:01 - mmengine - INFO - Exp name: convnext_small_4xb256_IF_1m_lr4e-4_aug_1e-1_20230606_050018 +2023/06/06 07:07:01 - mmengine - INFO - Saving checkpoint at 7 epochs +2023/06/06 07:07:29 - mmengine - INFO - Epoch(val) [7][29/29] accuracy/top1: 99.7182 single-label/precision_classwise: [99.53482818603516, 99.9454574584961] single-label/recall_classwise: [99.955810546875, 99.42636108398438] single-label/f1-score_classwise: [99.744873046875, 99.68522644042969] data_time: 0.0519 time: 0.4389 +2023/06/06 07:08:26 - mmengine - INFO - Epoch(train) [8][ 100/1969] lr: 8.7904e-05 eta: 0:51:52 time: 0.5150 data_time: 0.0017 memory: 15575 loss: 0.0033 +2023/06/06 07:09:18 - mmengine - INFO - Epoch(train) [8][ 200/1969] lr: 8.5432e-05 eta: 0:50:58 time: 0.5222 data_time: 0.0016 memory: 15575 loss: 0.0034 +2023/06/06 07:09:27 - mmengine - INFO - Exp name: convnext_small_4xb256_IF_1m_lr4e-4_aug_1e-1_20230606_050018 +2023/06/06 07:10:10 - mmengine - INFO - Epoch(train) [8][ 300/1969] lr: 8.2989e-05 eta: 0:50:03 time: 0.5161 data_time: 0.0017 memory: 15575 loss: 0.0033 +2023/06/06 07:11:01 - mmengine - INFO - Epoch(train) [8][ 400/1969] lr: 8.0578e-05 eta: 0:49:09 time: 0.5145 data_time: 0.0016 memory: 15575 loss: 0.0035 +2023/06/06 07:11:53 - mmengine - INFO - Epoch(train) [8][ 500/1969] lr: 7.8198e-05 eta: 0:48:15 time: 0.5149 data_time: 0.0018 memory: 15575 loss: 0.0065 +2023/06/06 07:12:45 - mmengine - INFO - Epoch(train) [8][ 600/1969] lr: 7.5851e-05 eta: 0:47:21 time: 0.5302 data_time: 0.0016 memory: 15575 loss: 0.0046 +2023/06/06 07:13:37 - mmengine - INFO - Epoch(train) [8][ 700/1969] lr: 7.3536e-05 eta: 0:46:27 time: 0.5154 data_time: 0.0016 memory: 15575 loss: 0.0038 +2023/06/06 07:14:29 - mmengine - INFO - Epoch(train) [8][ 800/1969] lr: 7.1255e-05 eta: 0:45:32 time: 0.5161 data_time: 0.0015 memory: 15575 loss: 0.0036 +2023/06/06 07:15:20 - mmengine - INFO - Epoch(train) [8][ 900/1969] lr: 6.9008e-05 eta: 0:44:38 time: 0.5152 data_time: 0.0015 memory: 15575 loss: 0.0069 +2023/06/06 07:16:12 - mmengine - INFO - Epoch(train) [8][1000/1969] lr: 6.6796e-05 eta: 0:43:44 time: 0.5150 data_time: 0.0016 memory: 15575 loss: 0.0063 +2023/06/06 07:17:04 - mmengine - INFO - Epoch(train) [8][1100/1969] lr: 6.4618e-05 eta: 0:42:50 time: 0.5160 data_time: 0.0018 memory: 15575 loss: 0.0052 +2023/06/06 07:17:56 - mmengine - INFO - Epoch(train) [8][1200/1969] lr: 6.2477e-05 eta: 0:41:56 time: 0.5145 data_time: 0.0015 memory: 15575 loss: 0.0036 +2023/06/06 07:18:05 - mmengine - INFO - Exp name: convnext_small_4xb256_IF_1m_lr4e-4_aug_1e-1_20230606_050018 +2023/06/06 07:18:48 - mmengine - INFO - Epoch(train) [8][1300/1969] lr: 6.0372e-05 eta: 0:41:02 time: 0.5158 data_time: 0.0015 memory: 15575 loss: 0.0036 +2023/06/06 07:19:39 - mmengine - INFO - Epoch(train) [8][1400/1969] lr: 5.8303e-05 eta: 0:40:08 time: 0.5167 data_time: 0.0016 memory: 15575 loss: 0.0043 +2023/06/06 07:20:31 - mmengine - INFO - Epoch(train) [8][1500/1969] lr: 5.6272e-05 eta: 0:39:14 time: 0.5148 data_time: 0.0016 memory: 15575 loss: 0.0051 +2023/06/06 07:21:23 - mmengine - INFO - Epoch(train) [8][1600/1969] lr: 5.4279e-05 eta: 0:38:20 time: 0.5186 data_time: 0.0015 memory: 15575 loss: 0.0023 +2023/06/06 07:22:15 - mmengine - INFO - Epoch(train) [8][1700/1969] lr: 5.2324e-05 eta: 0:37:27 time: 0.5155 data_time: 0.0016 memory: 15575 loss: 0.0068 +2023/06/06 07:23:06 - mmengine - INFO - Epoch(train) [8][1800/1969] lr: 5.0408e-05 eta: 0:36:33 time: 0.5153 data_time: 0.0015 memory: 15575 loss: 0.0045 +2023/06/06 07:23:58 - mmengine - INFO - Epoch(train) [8][1900/1969] lr: 4.8532e-05 eta: 0:35:39 time: 0.5148 data_time: 0.0015 memory: 15575 loss: 0.0057 +2023/06/06 07:24:34 - mmengine - INFO - Exp name: convnext_small_4xb256_IF_1m_lr4e-4_aug_1e-1_20230606_050018 +2023/06/06 07:24:34 - mmengine - INFO - Saving checkpoint at 8 epochs +2023/06/06 07:25:05 - mmengine - INFO - Epoch(val) [8][29/29] accuracy/top1: 99.7634 single-label/precision_classwise: [99.62247467041016, 99.93772888183594] single-label/recall_classwise: [99.94949340820312, 99.53488159179688] single-label/f1-score_classwise: [99.78572082519531, 99.73590087890625] data_time: 0.1759 time: 0.5638 +2023/06/06 07:26:04 - mmengine - INFO - Epoch(train) [9][ 100/1969] lr: 4.5451e-05 eta: 0:34:10 time: 0.5240 data_time: 0.0015 memory: 15575 loss: 0.0062 +2023/06/06 07:26:56 - mmengine - INFO - Epoch(train) [9][ 200/1969] lr: 4.3683e-05 eta: 0:33:16 time: 0.5194 data_time: 0.0016 memory: 15575 loss: 0.0047 +2023/06/06 07:27:21 - mmengine - INFO - Exp name: convnext_small_4xb256_IF_1m_lr4e-4_aug_1e-1_20230606_050018 +2023/06/06 07:27:48 - mmengine - INFO - Epoch(train) [9][ 300/1969] lr: 4.1955e-05 eta: 0:32:22 time: 0.5179 data_time: 0.0015 memory: 15575 loss: 0.0094 +2023/06/06 07:28:40 - mmengine - INFO - Epoch(train) [9][ 400/1969] lr: 4.0269e-05 eta: 0:31:28 time: 0.5255 data_time: 0.0015 memory: 15575 loss: 0.0035 +2023/06/06 07:29:32 - mmengine - INFO - Epoch(train) [9][ 500/1969] lr: 3.8626e-05 eta: 0:30:35 time: 0.5147 data_time: 0.0015 memory: 15575 loss: 0.0026 +2023/06/06 07:30:23 - mmengine - INFO - Epoch(train) [9][ 600/1969] lr: 3.7024e-05 eta: 0:29:41 time: 0.5324 data_time: 0.0015 memory: 15575 loss: 0.0060 +2023/06/06 07:31:15 - mmengine - INFO - Epoch(train) [9][ 700/1969] lr: 3.5465e-05 eta: 0:28:47 time: 0.5146 data_time: 0.0015 memory: 15575 loss: 0.0054 +2023/06/06 07:32:07 - mmengine - INFO - Epoch(train) [9][ 800/1969] lr: 3.3950e-05 eta: 0:27:54 time: 0.5209 data_time: 0.0016 memory: 15575 loss: 0.0066 +2023/06/06 07:32:59 - mmengine - INFO - Epoch(train) [9][ 900/1969] lr: 3.2478e-05 eta: 0:27:00 time: 0.5150 data_time: 0.0017 memory: 15575 loss: 0.0030 +2023/06/06 07:33:51 - mmengine - INFO - Epoch(train) [9][1000/1969] lr: 3.1049e-05 eta: 0:26:06 time: 0.5389 data_time: 0.0015 memory: 15575 loss: 0.0032 +2023/06/06 07:34:42 - mmengine - INFO - Epoch(train) [9][1100/1969] lr: 2.9666e-05 eta: 0:25:13 time: 0.5170 data_time: 0.0017 memory: 15575 loss: 0.0047 +2023/06/06 07:35:34 - mmengine - INFO - Epoch(train) [9][1200/1969] lr: 2.8326e-05 eta: 0:24:19 time: 0.5318 data_time: 0.0017 memory: 15575 loss: 0.0032 +2023/06/06 07:35:59 - mmengine - INFO - Exp name: convnext_small_4xb256_IF_1m_lr4e-4_aug_1e-1_20230606_050018 +2023/06/06 07:36:26 - mmengine - INFO - Epoch(train) [9][1300/1969] lr: 2.7032e-05 eta: 0:23:26 time: 0.5228 data_time: 0.0016 memory: 15575 loss: 0.0058 +2023/06/06 07:37:18 - mmengine - INFO - Epoch(train) [9][1400/1969] lr: 2.5783e-05 eta: 0:22:32 time: 0.5210 data_time: 0.0017 memory: 15575 loss: 0.0050 +2023/06/06 07:38:10 - mmengine - INFO - Epoch(train) [9][1500/1969] lr: 2.4580e-05 eta: 0:21:39 time: 0.5191 data_time: 0.0015 memory: 15575 loss: 0.0072 +2023/06/06 07:39:02 - mmengine - INFO - Epoch(train) [9][1600/1969] lr: 2.3422e-05 eta: 0:20:45 time: 0.5155 data_time: 0.0015 memory: 15575 loss: 0.0028 +2023/06/06 07:39:53 - mmengine - INFO - Epoch(train) [9][1700/1969] lr: 2.2311e-05 eta: 0:19:52 time: 0.5212 data_time: 0.0016 memory: 15575 loss: 0.0031 +2023/06/06 07:40:45 - mmengine - INFO - Epoch(train) [9][1800/1969] lr: 2.1246e-05 eta: 0:18:58 time: 0.5150 data_time: 0.0016 memory: 15575 loss: 0.0047 +2023/06/06 07:41:37 - mmengine - INFO - Epoch(train) [9][1900/1969] lr: 2.0229e-05 eta: 0:18:05 time: 0.5147 data_time: 0.0015 memory: 15575 loss: 0.0027 +2023/06/06 07:42:12 - mmengine - INFO - Exp name: convnext_small_4xb256_IF_1m_lr4e-4_aug_1e-1_20230606_050018 +2023/06/06 07:42:12 - mmengine - INFO - Saving checkpoint at 9 epochs +2023/06/06 07:42:43 - mmengine - INFO - Epoch(val) [9][29/29] accuracy/top1: 99.7878 single-label/precision_classwise: [99.66012573242188, 99.94554138183594] single-label/recall_classwise: [99.955810546875, 99.5813980102539] single-label/f1-score_classwise: [99.80774688720703, 99.76313781738281] data_time: 0.1560 time: 0.5419 +2023/06/06 07:43:41 - mmengine - INFO - Epoch(train) [10][ 100/1969] lr: 1.8615e-05 eta: 0:16:35 time: 0.5160 data_time: 0.0016 memory: 15575 loss: 0.0077 +2023/06/06 07:44:33 - mmengine - INFO - Epoch(train) [10][ 200/1969] lr: 1.7725e-05 eta: 0:15:42 time: 0.5250 data_time: 0.0015 memory: 15575 loss: 0.0054 +2023/06/06 07:45:13 - mmengine - INFO - Exp name: convnext_small_4xb256_IF_1m_lr4e-4_aug_1e-1_20230606_050018 +2023/06/06 07:45:24 - mmengine - INFO - Epoch(train) [10][ 300/1969] lr: 1.6881e-05 eta: 0:14:48 time: 0.5152 data_time: 0.0015 memory: 15575 loss: 0.0073 +2023/06/06 07:46:16 - mmengine - INFO - Epoch(train) [10][ 400/1969] lr: 1.6086e-05 eta: 0:13:55 time: 0.5257 data_time: 0.0019 memory: 15575 loss: 0.0048 +2023/06/06 07:47:09 - mmengine - INFO - Epoch(train) [10][ 500/1969] lr: 1.5339e-05 eta: 0:13:02 time: 0.5185 data_time: 0.0023 memory: 15575 loss: 0.0087 +2023/06/06 07:48:00 - mmengine - INFO - Epoch(train) [10][ 600/1969] lr: 1.4640e-05 eta: 0:12:08 time: 0.5172 data_time: 0.0016 memory: 15575 loss: 0.0069 +2023/06/06 07:48:52 - mmengine - INFO - Epoch(train) [10][ 700/1969] lr: 1.3990e-05 eta: 0:11:15 time: 0.5169 data_time: 0.0017 memory: 15575 loss: 0.0027 +2023/06/06 07:49:44 - mmengine - INFO - Epoch(train) [10][ 800/1969] lr: 1.3388e-05 eta: 0:10:22 time: 0.5187 data_time: 0.0015 memory: 15575 loss: 0.0065 +2023/06/06 07:50:36 - mmengine - INFO - Epoch(train) [10][ 900/1969] lr: 1.2835e-05 eta: 0:09:28 time: 0.5152 data_time: 0.0015 memory: 15575 loss: 0.0061 +2023/06/06 07:51:28 - mmengine - INFO - Epoch(train) [10][1000/1969] lr: 1.2331e-05 eta: 0:08:35 time: 0.5146 data_time: 0.0015 memory: 15575 loss: 0.0015 +2023/06/06 07:52:20 - mmengine - INFO - Epoch(train) [10][1100/1969] lr: 1.1876e-05 eta: 0:07:42 time: 0.5165 data_time: 0.0016 memory: 15575 loss: 0.0043 +2023/06/06 07:53:12 - mmengine - INFO - Epoch(train) [10][1200/1969] lr: 1.1470e-05 eta: 0:06:49 time: 0.5141 data_time: 0.0015 memory: 15575 loss: 0.0031 +2023/06/06 07:53:53 - mmengine - INFO - Exp name: convnext_small_4xb256_IF_1m_lr4e-4_aug_1e-1_20230606_050018 +2023/06/06 07:54:03 - mmengine - INFO - Epoch(train) [10][1300/1969] lr: 1.1113e-05 eta: 0:05:55 time: 0.5144 data_time: 0.0015 memory: 15575 loss: 0.0063 +2023/06/06 07:54:55 - mmengine - INFO - Epoch(train) [10][1400/1969] lr: 1.0806e-05 eta: 0:05:02 time: 0.5168 data_time: 0.0015 memory: 15575 loss: 0.0017 +2023/06/06 07:55:47 - mmengine - INFO - Epoch(train) [10][1500/1969] lr: 1.0548e-05 eta: 0:04:09 time: 0.5162 data_time: 0.0018 memory: 15575 loss: 0.0056 +2023/06/06 07:56:39 - mmengine - INFO - Epoch(train) [10][1600/1969] lr: 1.0340e-05 eta: 0:03:16 time: 0.5177 data_time: 0.0021 memory: 15575 loss: 0.0032 +2023/06/06 07:57:31 - mmengine - INFO - Epoch(train) [10][1700/1969] lr: 1.0181e-05 eta: 0:02:22 time: 0.5150 data_time: 0.0018 memory: 15575 loss: 0.0030 +2023/06/06 07:58:23 - mmengine - INFO - Epoch(train) [10][1800/1969] lr: 1.0072e-05 eta: 0:01:29 time: 0.5157 data_time: 0.0016 memory: 15575 loss: 0.0030 +2023/06/06 07:59:14 - mmengine - INFO - Epoch(train) [10][1900/1969] lr: 1.0012e-05 eta: 0:00:36 time: 0.5153 data_time: 0.0016 memory: 15575 loss: 0.0025 +2023/06/06 07:59:50 - mmengine - INFO - Exp name: convnext_small_4xb256_IF_1m_lr4e-4_aug_1e-1_20230606_050018 +2023/06/06 07:59:50 - mmengine - INFO - Saving checkpoint at 10 epochs +2023/06/06 08:00:22 - mmengine - INFO - Epoch(val) [10][29/29] accuracy/top1: 99.8330 single-label/precision_classwise: [99.76053619384766, 99.92230987548828] single-label/recall_classwise: [99.93687438964844, 99.70542907714844] single-label/f1-score_classwise: [99.84861755371094, 99.81375122070312] data_time: 0.1844 time: 0.5707 diff --git a/convnext_small_4xb256_IF_1m_lr4e-4_aug_1e-1/20230606_050018/vis_data/20230606_050018.json b/convnext_small_4xb256_IF_1m_lr4e-4_aug_1e-1/20230606_050018/vis_data/20230606_050018.json new file mode 100644 index 0000000000000000000000000000000000000000..982821765f6724cd912ce3a691dc9a71e530a3b5 --- /dev/null +++ b/convnext_small_4xb256_IF_1m_lr4e-4_aug_1e-1/20230606_050018/vis_data/20230606_050018.json @@ -0,0 +1,200 @@ +{"lr": 0.0003999756737991541, "data_time": 0.0018261909484863282, "loss": 0.027601023856550454, "time": 0.5191150665283203, "epoch": 1, "memory": 15578, "step": 100} +{"lr": 0.00039990171605141573, "data_time": 0.0018618345260620118, "loss": 0.0271825865842402, "time": 0.5322962045669556, "epoch": 1, "memory": 15578, "step": 200} +{"lr": 0.00039977814310179305, "data_time": 0.0018401622772216796, "loss": 0.02326606847345829, "time": 0.515569543838501, "epoch": 1, "memory": 15578, "step": 300} +{"lr": 0.00039960498640766363, "data_time": 0.0019962310791015623, "loss": 0.01996779362671077, "time": 0.5271080255508422, "epoch": 1, "memory": 15578, "step": 400} +{"lr": 0.00039938229004870185, "data_time": 0.001923537254333496, "loss": 0.020116635179147125, "time": 0.517112421989441, "epoch": 1, "memory": 15578, "step": 500} +{"lr": 0.00039911011071566005, "data_time": 0.0020764827728271484, "loss": 0.016853199154138566, "time": 0.5174846649169922, "epoch": 1, "memory": 15578, "step": 600} +{"lr": 0.0003987885176959352, "data_time": 0.0018217325210571288, "loss": 0.013987581757828594, "time": 0.5222087144851685, "epoch": 1, "memory": 15578, "step": 700} +{"lr": 0.0003984175928559308, "data_time": 0.0016354560852050782, "loss": 0.013943490013480187, "time": 0.5155891418457031, "epoch": 1, "memory": 15578, "step": 800} +{"lr": 0.00039799743062021714, "data_time": 0.001679682731628418, "loss": 0.01380026382394135, "time": 0.515896224975586, "epoch": 1, "memory": 15578, "step": 900} +{"lr": 0.0003975281379474946, "data_time": 0.0019191980361938476, "loss": 0.018961887620389462, "time": 0.5156231641769409, "epoch": 1, "memory": 15578, "step": 1000} +{"lr": 0.00039700983430336476, "data_time": 0.0016831398010253907, "loss": 0.01128587475977838, "time": 0.5164831638336181, "epoch": 1, "memory": 15578, "step": 1100} +{"lr": 0.0003964426516299177, "data_time": 0.0017139196395874023, "loss": 0.014654634799808264, "time": 0.5162296772003174, "epoch": 1, "memory": 15578, "step": 1200} +{"lr": 0.0003958267343121465, "data_time": 0.0017230987548828125, "loss": 0.009067555936053396, "time": 0.520532774925232, "epoch": 1, "memory": 15578, "step": 1300} +{"lr": 0.00039516223914119037, "data_time": 0.0016436100006103516, "loss": 0.01604834762401879, "time": 0.5156150817871094, "epoch": 1, "memory": 15578, "step": 1400} +{"lr": 0.0003944493352744212, "data_time": 0.001742100715637207, "loss": 0.019340876024216413, "time": 0.52655189037323, "epoch": 1, "memory": 15578, "step": 1500} +{"lr": 0.0003936882041923822, "data_time": 0.0016018867492675782, "loss": 0.018161868583410977, "time": 0.5159793376922608, "epoch": 1, "memory": 15578, "step": 1600} +{"lr": 0.0003928790396525886, "data_time": 0.0016326189041137695, "loss": 0.012575972964987159, "time": 0.5178181886672973, "epoch": 1, "memory": 15578, "step": 1700} +{"lr": 0.00039202204764020483, "data_time": 0.001667356491088867, "loss": 0.01994553143158555, "time": 0.5147627353668213, "epoch": 1, "memory": 15578, "step": 1800} +{"lr": 0.00039111744631560716, "data_time": 0.0015305757522583007, "loss": 0.012879175506532193, "time": 0.514723825454712, "epoch": 1, "memory": 15578, "step": 1900} +{"accuracy/top1": 96.25273895263672, "data_time": 0.13952156593059672, "time": 0.6054766918050831, "step": 1} +{"lr": 0.0003894811013774212, "data_time": 0.02893216609954834, "loss": 0.015500097675248981, "time": 0.5442700624465943, "epoch": 2, "memory": 15578, "step": 2069} +{"lr": 0.00038844960758563737, "data_time": 0.001591062545776367, "loss": 0.010155213065445423, "time": 0.5309667825698853, "epoch": 2, "memory": 15575, "step": 2169} +{"lr": 0.00038737141390069504, "data_time": 0.0356330156326294, "loss": 0.015067616989836097, "time": 0.5833153009414673, "epoch": 2, "memory": 15575, "step": 2269} +{"lr": 0.0003862467947932271, "data_time": 0.0016663551330566406, "loss": 0.014586024405434727, "time": 0.5234311342239379, "epoch": 2, "memory": 15575, "step": 2369} +{"lr": 0.000385076036552165, "data_time": 0.001562786102294922, "loss": 0.014382174983620644, "time": 0.5206002235412598, "epoch": 2, "memory": 15575, "step": 2469} +{"lr": 0.00038385943721186026, "data_time": 0.0015935182571411132, "loss": 0.012308013578876852, "time": 0.5144658327102661, "epoch": 2, "memory": 15575, "step": 2569} +{"lr": 0.0003825973064762139, "data_time": 0.2808521032333374, "loss": 0.013550773914903402, "time": 1.0783581256866455, "epoch": 2, "memory": 15575, "step": 2669} +{"lr": 0.000381289965639839, "data_time": 0.001841425895690918, "loss": 0.007510989252477884, "time": 0.6251338005065918, "epoch": 2, "memory": 15575, "step": 2769} +{"lr": 0.00037993774750626663, "data_time": 0.0014961481094360352, "loss": 0.012146124336868524, "time": 0.5329279184341431, "epoch": 2, "memory": 15575, "step": 2869} +{"lr": 0.00037854099630322966, "data_time": 0.0015124082565307617, "loss": 0.011426049657166005, "time": 0.7000055313110352, "epoch": 2, "memory": 15575, "step": 2969} +{"lr": 0.00037710006759503174, "data_time": 0.0015774250030517578, "loss": 0.013603793689981103, "time": 0.5768113613128663, "epoch": 2, "memory": 15575, "step": 3069} +{"lr": 0.00037561532819203284, "data_time": 0.0015952587127685547, "loss": 0.014887896692380309, "time": 0.5170141458511353, "epoch": 2, "memory": 15575, "step": 3169} +{"lr": 0.00037408715605727376, "data_time": 0.001523876190185547, "loss": 0.006633514491841197, "time": 0.5152340412139893, "epoch": 2, "memory": 15575, "step": 3269} +{"lr": 0.00037251594021025933, "data_time": 0.0016954421997070312, "loss": 0.014243566524237394, "time": 0.5231961488723755, "epoch": 2, "memory": 15575, "step": 3369} +{"lr": 0.00037090208062792615, "data_time": 0.02590627670288086, "loss": 0.012183473305776715, "time": 0.5400282144546509, "epoch": 2, "memory": 15575, "step": 3469} +{"lr": 0.00036924598814282443, "data_time": 0.001516437530517578, "loss": 0.006949825631454587, "time": 0.5146150112152099, "epoch": 2, "memory": 15575, "step": 3569} +{"lr": 0.00036754808433853343, "data_time": 9.311471533775329, "loss": 0.012170182494446635, "time": 9.849288773536681, "epoch": 2, "memory": 15575, "step": 3669} +{"lr": 0.00036580880144233954, "data_time": 0.0014627695083618164, "loss": 0.014173916587606073, "time": 0.5150933504104614, "epoch": 2, "memory": 15575, "step": 3769} +{"lr": 0.0003640285822152096, "data_time": 0.0016908884048461915, "loss": 0.010514096030965448, "time": 0.5575487852096558, "epoch": 2, "memory": 15575, "step": 3869} +{"accuracy/top1": 98.80310821533203, "data_time": 0.1626681089401245, "time": 0.5489826043446858, "step": 2} +{"lr": 0.0003609282333974528, "data_time": 0.0014587163925170899, "loss": 0.00871057934127748, "time": 0.5471591472625732, "epoch": 3, "memory": 15575, "step": 4038} +{"lr": 0.0003590401733608027, "data_time": 0.0020009279251098633, "loss": 0.009571832278743387, "time": 0.5169068813323975, "epoch": 3, "memory": 15575, "step": 4138} +{"lr": 0.00035711290005063985, "data_time": 0.001433587074279785, "loss": 0.008919509826228023, "time": 0.5254595518112183, "epoch": 3, "memory": 15575, "step": 4238} +{"lr": 0.0003551469040837546, "data_time": 0.0015285491943359375, "loss": 0.015251696854829789, "time": 0.5239350318908691, "epoch": 3, "memory": 15575, "step": 4338} +{"lr": 0.00035314268593437953, "data_time": 0.0015092849731445312, "loss": 0.016632861783728005, "time": 0.5233675003051758, "epoch": 3, "memory": 15575, "step": 4438} +{"lr": 0.00035110075580678697, "data_time": 0.0014284849166870117, "loss": 0.013199200900271536, "time": 0.5159765243530273, "epoch": 3, "memory": 15575, "step": 4538} +{"lr": 0.00034902163350540724, "data_time": 0.0014449596405029298, "loss": 0.006127667799592018, "time": 0.5216275691986084, "epoch": 3, "memory": 15575, "step": 4638} +{"lr": 0.0003469058483025052, "data_time": 0.0015788793563842774, "loss": 0.012180058006197213, "time": 0.5343188047409058, "epoch": 3, "memory": 15575, "step": 4738} +{"lr": 0.0003447539388034461, "data_time": 0.001650857925415039, "loss": 0.01343419598415494, "time": 0.554154348373413, "epoch": 3, "memory": 15575, "step": 4838} +{"lr": 0.0003425664528095857, "data_time": 0.001610136032104492, "loss": 0.014271644130349159, "time": 0.528766679763794, "epoch": 3, "memory": 15575, "step": 4938} +{"lr": 0.00034034394717881845, "data_time": 0.0015502691268920899, "loss": 0.01365456241182983, "time": 0.5186124563217163, "epoch": 3, "memory": 15575, "step": 5038} +{"lr": 0.00033808698768382034, "data_time": 0.0019941091537475585, "loss": 0.012560153007507324, "time": 0.522815203666687, "epoch": 3, "memory": 15575, "step": 5138} +{"lr": 0.00033579614886802443, "data_time": 0.0015717506408691405, "loss": 0.004915389930829406, "time": 0.5420629978179932, "epoch": 3, "memory": 15575, "step": 5238} +{"lr": 0.00033347201389936123, "data_time": 0.0014863252639770509, "loss": 0.01211695084348321, "time": 0.5369882583618164, "epoch": 3, "memory": 15575, "step": 5338} +{"lr": 0.0003311151744218034, "data_time": 0.0015731573104858399, "loss": 0.008279312448576093, "time": 0.5183897495269776, "epoch": 3, "memory": 15575, "step": 5438} +{"lr": 0.0003287262304047557, "data_time": 0.001571822166442871, "loss": 0.012255780724808573, "time": 0.5148802042007447, "epoch": 3, "memory": 15575, "step": 5538} +{"lr": 0.0003263057899903226, "data_time": 0.0017143487930297852, "loss": 0.01518821376375854, "time": 0.5159362554550171, "epoch": 3, "memory": 15575, "step": 5638} +{"lr": 0.00032385446933849547, "data_time": 0.014629149436950683, "loss": 0.013086866587400436, "time": 0.5274892568588256, "epoch": 3, "memory": 15575, "step": 5738} +{"lr": 0.0003213728924703014, "data_time": 0.0015346765518188476, "loss": 0.008438705652952194, "time": 0.5169786930084228, "epoch": 3, "memory": 15575, "step": 5838} +{"accuracy/top1": 99.24845886230469, "data_time": 0.05241813659667969, "time": 0.43549114068349204, "step": 3} +{"lr": 0.00031711202351665616, "data_time": 0.0016689777374267578, "loss": 0.007415917282924056, "time": 0.5154834747314453, "epoch": 4, "memory": 15575, "step": 6007} +{"lr": 0.0003145522137064149, "data_time": 0.0014388084411621094, "loss": 0.012401284277439117, "time": 0.5166043281555176, "epoch": 4, "memory": 15575, "step": 6107} +{"lr": 0.0003119645157107334, "data_time": 0.001614832878112793, "loss": 0.007888593943789601, "time": 0.5158061981201172, "epoch": 4, "memory": 15575, "step": 6207} +{"lr": 0.00030934958826756915, "data_time": 0.0015721321105957031, "loss": 0.011648882133886217, "time": 0.518235182762146, "epoch": 4, "memory": 15575, "step": 6307} +{"lr": 0.0003067080970465515, "data_time": 0.0015164613723754883, "loss": 0.008028108440339566, "time": 0.5149527311325073, "epoch": 4, "memory": 15575, "step": 6407} +{"lr": 0.00030404071447952345, "data_time": 0.001444864273071289, "loss": 0.007447708304971457, "time": 0.5185810565948487, "epoch": 4, "memory": 15575, "step": 6507} +{"lr": 0.0003013481195893653, "data_time": 0.0015978813171386719, "loss": 0.013097852282226086, "time": 0.5151091337203979, "epoch": 4, "memory": 15575, "step": 6607} +{"lr": 0.0002986309978171385, "data_time": 0.0015954971313476562, "loss": 0.008581136912107467, "time": 0.5163673639297486, "epoch": 4, "memory": 15575, "step": 6707} +{"lr": 0.0002958900408475955, "data_time": 0.0016129016876220703, "loss": 0.005765685345977545, "time": 0.5147150039672852, "epoch": 4, "memory": 15575, "step": 6807} +{"lr": 0.0002931259464331015, "data_time": 0.001521730422973633, "loss": 0.009429838322103024, "time": 0.5155909299850464, "epoch": 4, "memory": 15575, "step": 6907} +{"lr": 0.00029033941821601176, "data_time": 0.0015855550765991211, "loss": 0.011490005301311612, "time": 0.5152943849563598, "epoch": 4, "memory": 15575, "step": 7007} +{"lr": 0.0002875311655495464, "data_time": 0.0014770030975341797, "loss": 0.006993796117603779, "time": 0.5145813465118408, "epoch": 4, "memory": 15575, "step": 7107} +{"lr": 0.0002847019033172168, "data_time": 0.001403641700744629, "loss": 0.011287294235080481, "time": 0.5149314403533936, "epoch": 4, "memory": 15575, "step": 7207} +{"lr": 0.0002818523517508384, "data_time": 0.0015740394592285156, "loss": 0.006218388443812728, "time": 0.5160921812057495, "epoch": 4, "memory": 15575, "step": 7307} +{"lr": 0.00027898323624718594, "data_time": 0.0015975475311279298, "loss": 0.006412314670160413, "time": 0.5145673274993896, "epoch": 4, "memory": 15575, "step": 7407} +{"lr": 0.000276095287183333, "data_time": 0.002046346664428711, "loss": 0.008772781072184443, "time": 0.5230429172515869, "epoch": 4, "memory": 15575, "step": 7507} +{"lr": 0.00027318923973072227, "data_time": 0.0014729261398315429, "loss": 0.00985402320511639, "time": 0.5146645307540894, "epoch": 4, "memory": 15575, "step": 7607} +{"lr": 0.00027026583366801727, "data_time": 0.0015037775039672852, "loss": 0.0145890639629215, "time": 0.5196113109588623, "epoch": 4, "memory": 15575, "step": 7707} +{"lr": 0.0002673258131927808, "data_time": 0.0014482736587524414, "loss": 0.012970298808068038, "time": 0.5275277137756348, "epoch": 4, "memory": 15575, "step": 7807} +{"accuracy/top1": 99.377197265625, "data_time": 0.14185065428415936, "time": 0.5245868841807048, "step": 4} +{"lr": 0.0002623215076436499, "data_time": 0.0015915870666503907, "loss": 0.01226809537038207, "time": 0.5155995845794678, "epoch": 5, "memory": 15575, "step": 7976} +{"lr": 0.00025934052007928784, "data_time": 0.00157623291015625, "loss": 0.013417656626552343, "time": 0.5149784326553345, "epoch": 5, "memory": 15575, "step": 8076} +{"lr": 0.0002563456993074567, "data_time": 0.0016045808792114259, "loss": 0.006499800039455295, "time": 0.5153000116348266, "epoch": 5, "memory": 15575, "step": 8176} +{"lr": 0.00025333780770542177, "data_time": 0.0015806436538696289, "loss": 0.010765056544914841, "time": 0.5195995092391967, "epoch": 5, "memory": 15575, "step": 8276} +{"lr": 0.00025031761097782693, "data_time": 0.001491379737854004, "loss": 0.008847670070827008, "time": 0.51485755443573, "epoch": 5, "memory": 15575, "step": 8376} +{"lr": 0.0002472858779617731, "data_time": 0.0015269041061401368, "loss": 0.006283209845423699, "time": 0.5262893915176392, "epoch": 5, "memory": 15575, "step": 8476} +{"lr": 0.0002442433804310988, "data_time": 0.0015306472778320312, "loss": 0.00790382451377809, "time": 0.5153609275817871, "epoch": 5, "memory": 15575, "step": 8576} +{"lr": 0.00024119089289991304, "data_time": 0.0014877557754516602, "loss": 0.0048039590008556845, "time": 0.5214916706085205, "epoch": 5, "memory": 15575, "step": 8676} +{"lr": 0.00023812919242543494, "data_time": 0.001499772071838379, "loss": 0.0070088079664856195, "time": 0.514876914024353, "epoch": 5, "memory": 15575, "step": 8776} +{"lr": 0.00023505905841017684, "data_time": 0.0015369176864624024, "loss": 0.011009765835478903, "time": 0.5175901651382446, "epoch": 5, "memory": 15575, "step": 8876} +{"lr": 0.00023198127240353698, "data_time": 0.0014757394790649414, "loss": 0.007589173363521695, "time": 0.5148205280303955, "epoch": 5, "memory": 15575, "step": 8976} +{"lr": 0.00022889661790284532, "data_time": 0.0015509366989135743, "loss": 0.008709517400711774, "time": 0.5236516475677491, "epoch": 5, "memory": 15575, "step": 9076} +{"lr": 0.000225805880153911, "data_time": 0.001490616798400879, "loss": 0.008318354608491063, "time": 0.5151755809783936, "epoch": 5, "memory": 15575, "step": 9176} +{"lr": 0.00022270984595112745, "data_time": 0.0015845537185668946, "loss": 0.00993576324544847, "time": 0.5195048809051513, "epoch": 5, "memory": 15575, "step": 9276} +{"lr": 0.00021960930343718077, "data_time": 0.001716303825378418, "loss": 0.0068968711886554955, "time": 0.5150446176528931, "epoch": 5, "memory": 15575, "step": 9376} +{"lr": 0.00021650504190241605, "data_time": 0.001743459701538086, "loss": 0.007159140007570386, "time": 0.5200284719467163, "epoch": 5, "memory": 15575, "step": 9476} +{"lr": 0.00021339785158391245, "data_time": 0.0015363693237304688, "loss": 0.011549755837768317, "time": 0.5149904251098633, "epoch": 5, "memory": 15575, "step": 9576} +{"lr": 0.00021028852346431534, "data_time": 0.0014166831970214844, "loss": 0.006046575214713812, "time": 0.5203746557235718, "epoch": 5, "memory": 15575, "step": 9676} +{"lr": 0.00020717784907047976, "data_time": 0.0014749288558959961, "loss": 0.006021002354100346, "time": 0.5229143142700196, "epoch": 5, "memory": 15575, "step": 9776} +{"accuracy/top1": 99.56160736083984, "data_time": 0.07427238623301188, "time": 0.46157800356547035, "step": 5} +{"lr": 0.0002019199632200586, "data_time": 0.001688075065612793, "loss": 0.008768323250114918, "time": 0.5154914140701294, "epoch": 6, "memory": 15575, "step": 9945} +{"lr": 0.00019880959773413697, "data_time": 0.0016509294509887695, "loss": 0.009114690544083714, "time": 0.5151286840438842, "epoch": 6, "memory": 15575, "step": 10045} +{"lr": 0.00019570080810944632, "data_time": 0.0016642332077026368, "loss": 0.011529341479763389, "time": 0.5162417888641357, "epoch": 6, "memory": 15575, "step": 10145} +{"lr": 0.00019259438573575975, "data_time": 0.001756906509399414, "loss": 0.009273130353540182, "time": 0.5160192251205444, "epoch": 6, "memory": 15575, "step": 10245} +{"lr": 0.00018949112140023044, "data_time": 0.001581430435180664, "loss": 0.009626655094325542, "time": 0.523019528388977, "epoch": 6, "memory": 15575, "step": 10345} +{"lr": 0.00018639180508608498, "data_time": 0.0015453100204467773, "loss": 0.008061748836189509, "time": 0.5278511047363281, "epoch": 6, "memory": 15575, "step": 10445} +{"lr": 0.00018329722577152123, "data_time": 0.0014238595962524415, "loss": 0.011502831522375346, "time": 0.5156992197036743, "epoch": 6, "memory": 15575, "step": 10545} +{"lr": 0.0001802081712288607, "data_time": 0.001892828941345215, "loss": 0.009451930969953537, "time": 0.5201520442962646, "epoch": 6, "memory": 15575, "step": 10645} +{"lr": 0.00017712542782401123, "data_time": 0.001572561264038086, "loss": 0.005022558709606528, "time": 0.5149688243865966, "epoch": 6, "memory": 15575, "step": 10745} +{"lr": 0.00017404978031628336, "data_time": 0.0017352104187011719, "loss": 0.00650988407433033, "time": 0.530413818359375, "epoch": 6, "memory": 15575, "step": 10845} +{"lr": 0.00017098201165861962, "data_time": 0.0016693353652954101, "loss": 0.008117228886112571, "time": 0.5151396274566651, "epoch": 6, "memory": 15575, "step": 10945} +{"lr": 0.00016792290279828072, "data_time": 0.0018962860107421876, "loss": 0.006933824857696891, "time": 0.5237004041671753, "epoch": 6, "memory": 15575, "step": 11045} +{"lr": 0.00016487323247804387, "data_time": 0.0017143964767456054, "loss": 0.006620669225230813, "time": 0.5151191234588623, "epoch": 6, "memory": 15575, "step": 11145} +{"lr": 0.00016183377703796263, "data_time": 0.0017733573913574219, "loss": 0.0039889195933938025, "time": 0.5164686441421509, "epoch": 6, "memory": 15575, "step": 11245} +{"lr": 0.00015880531021773676, "data_time": 0.001653432846069336, "loss": 0.006959619652479887, "time": 0.514758563041687, "epoch": 6, "memory": 15575, "step": 11345} +{"lr": 0.00015578860295974566, "data_time": 0.001767420768737793, "loss": 0.006196077354252339, "time": 0.5205242872238159, "epoch": 6, "memory": 15575, "step": 11445} +{"lr": 0.0001527844232127924, "data_time": 0.0015105485916137695, "loss": 0.008009692560881377, "time": 0.5164531230926513, "epoch": 6, "memory": 15575, "step": 11545} +{"lr": 0.000149793535736612, "data_time": 0.001721334457397461, "loss": 0.0076533563435077666, "time": 0.5187115669250488, "epoch": 6, "memory": 15575, "step": 11645} +{"lr": 0.00014681670190718876, "data_time": 0.0017121553421020508, "loss": 0.009085492882877588, "time": 0.5150811433792114, "epoch": 6, "memory": 15575, "step": 11745} +{"accuracy/top1": 99.65554809570312, "data_time": 0.17346168359120687, "time": 0.5613858938217163, "step": 6} +{"lr": 0.00014181991425636192, "data_time": 0.0016644716262817383, "loss": 0.01141593037173152, "time": 0.5167794704437256, "epoch": 7, "memory": 15575, "step": 11914} +{"lr": 0.00013888463509383338, "data_time": 0.0017268896102905274, "loss": 0.004260255442932248, "time": 0.5172703981399536, "epoch": 7, "memory": 15575, "step": 12014} +{"lr": 0.000135966186604959, "data_time": 0.00152740478515625, "loss": 0.010442282911390066, "time": 0.5161826372146606, "epoch": 7, "memory": 15575, "step": 12114} +{"lr": 0.0001330653117252752, "data_time": 0.0015177726745605469, "loss": 0.007746391324326396, "time": 0.51475031375885, "epoch": 7, "memory": 15575, "step": 12214} +{"lr": 0.00013018274891668863, "data_time": 0.0014116287231445313, "loss": 0.005766073195263744, "time": 0.5159190177917481, "epoch": 7, "memory": 15575, "step": 12314} +{"lr": 0.000127319231979489, "data_time": 0.001488971710205078, "loss": 0.008930237777531148, "time": 0.5161234855651855, "epoch": 7, "memory": 15575, "step": 12414} +{"lr": 0.0001244754898655487, "data_time": 0.001428842544555664, "loss": 0.005514901271089912, "time": 0.5152976512908936, "epoch": 7, "memory": 15575, "step": 12514} +{"lr": 0.0001216522464927582, "data_time": 0.0015146970748901368, "loss": 0.004802424600347877, "time": 0.5157246828079224, "epoch": 7, "memory": 15575, "step": 12614} +{"lr": 0.00011885022056074007, "data_time": 0.0014697551727294923, "loss": 0.0044075089506804945, "time": 0.5160865068435669, "epoch": 7, "memory": 15575, "step": 12714} +{"lr": 0.00011607012536789545, "data_time": 0.0016083002090454101, "loss": 0.004432854941114783, "time": 0.5158129930496216, "epoch": 7, "memory": 15575, "step": 12814} +{"lr": 0.00011331266862982092, "data_time": 0.0015475749969482422, "loss": 0.006024764757603407, "time": 0.514604139328003, "epoch": 7, "memory": 15575, "step": 12914} +{"lr": 0.00011057855229914932, "data_time": 0.0015961885452270507, "loss": 0.003552483953535557, "time": 0.5149378538131714, "epoch": 7, "memory": 15575, "step": 13014} +{"lr": 0.0001078684723868573, "data_time": 0.0015608549118041992, "loss": 0.009875560691580177, "time": 0.5153422832489014, "epoch": 7, "memory": 15575, "step": 13114} +{"lr": 0.00010518311878508484, "data_time": 0.0015584945678710938, "loss": 0.006951159657910466, "time": 0.5147438049316406, "epoch": 7, "memory": 15575, "step": 13214} +{"lr": 0.00010252317509151276, "data_time": 0.0016586065292358398, "loss": 0.01264434470795095, "time": 0.5150826692581176, "epoch": 7, "memory": 15575, "step": 13314} +{"lr": 9.98893184353425e-05, "data_time": 0.0016132593154907227, "loss": 0.005360524170100689, "time": 0.5149397850036621, "epoch": 7, "memory": 15575, "step": 13414} +{"lr": 9.728221930492262e-05, "data_time": 0.0014868974685668945, "loss": 0.00791693925857544, "time": 0.5155071496963501, "epoch": 7, "memory": 15575, "step": 13514} +{"lr": 9.470254137706575e-05, "data_time": 0.0014949083328247071, "loss": 0.005792217794805765, "time": 0.5146904945373535, "epoch": 7, "memory": 15575, "step": 13614} +{"lr": 9.215094134809975e-05, "data_time": 0.0014746904373168946, "loss": 0.006089560361579061, "time": 0.5150094985961914, "epoch": 7, "memory": 15575, "step": 13714} +{"accuracy/top1": 99.71817779541016, "data_time": 0.051924705505371094, "time": 0.43892056941986085, "step": 7} +{"lr": 8.790437228679661e-05, "data_time": 0.001678800582885742, "loss": 0.003303104406222701, "time": 0.5149797677993775, "epoch": 8, "memory": 15575, "step": 13883} +{"lr": 8.543150502338196e-05, "data_time": 0.001611018180847168, "loss": 0.00341587932780385, "time": 0.5222368717193604, "epoch": 8, "memory": 15575, "step": 13983} +{"lr": 8.298907574244125e-05, "data_time": 0.0017382144927978516, "loss": 0.003288163850083947, "time": 0.516103720664978, "epoch": 8, "memory": 15575, "step": 14083} +{"lr": 8.057770620156745e-05, "data_time": 0.0016478776931762695, "loss": 0.003450537147000432, "time": 0.5145237684249878, "epoch": 8, "memory": 15575, "step": 14183} +{"lr": 7.819801025162347e-05, "data_time": 0.0017688512802124024, "loss": 0.006510276347398758, "time": 0.514856481552124, "epoch": 8, "memory": 15575, "step": 14283} +{"lr": 7.585059368047708e-05, "data_time": 0.0015503644943237304, "loss": 0.004638825636357069, "time": 0.5301702737808227, "epoch": 8, "memory": 15575, "step": 14383} +{"lr": 7.353605405878812e-05, "data_time": 0.0016034603118896484, "loss": 0.0037703410256654025, "time": 0.5154333353042603, "epoch": 8, "memory": 15575, "step": 14483} +{"lr": 7.125498058788827e-05, "data_time": 0.0014626741409301757, "loss": 0.0036244274117052553, "time": 0.5160910606384277, "epoch": 8, "memory": 15575, "step": 14583} +{"lr": 6.900795394979036e-05, "data_time": 0.0015135526657104493, "loss": 0.006894288817420602, "time": 0.5152117729187011, "epoch": 8, "memory": 15575, "step": 14683} +{"lr": 6.679554615936711e-05, "data_time": 0.0015542030334472657, "loss": 0.006287419935688376, "time": 0.5149774551391602, "epoch": 8, "memory": 15575, "step": 14783} +{"lr": 6.461832041873565e-05, "data_time": 0.0018001556396484374, "loss": 0.005158521374687553, "time": 0.5160490512847901, "epoch": 8, "memory": 15575, "step": 14883} +{"lr": 6.247683097388664e-05, "data_time": 0.001475954055786133, "loss": 0.0035997253842651846, "time": 0.5145102739334106, "epoch": 8, "memory": 15575, "step": 14983} +{"lr": 6.037162297359161e-05, "data_time": 0.0015129327774047851, "loss": 0.0036070691421628, "time": 0.5158472776412963, "epoch": 8, "memory": 15575, "step": 15083} +{"lr": 5.830323233062794e-05, "data_time": 0.0015598058700561524, "loss": 0.004280280461534858, "time": 0.5166856288909912, "epoch": 8, "memory": 15575, "step": 15183} +{"lr": 5.6272185585353725e-05, "data_time": 0.0015989303588867187, "loss": 0.005098620429635048, "time": 0.5148352146148681, "epoch": 8, "memory": 15575, "step": 15283} +{"lr": 5.4278999771668655e-05, "data_time": 0.0014949321746826171, "loss": 0.002327388757839799, "time": 0.518629002571106, "epoch": 8, "memory": 15575, "step": 15383} +{"lr": 5.2324182285395304e-05, "data_time": 0.0016213178634643555, "loss": 0.006777837499976158, "time": 0.5155083894729614, "epoch": 8, "memory": 15575, "step": 15483} +{"lr": 5.0408230755113716e-05, "data_time": 0.0014513731002807617, "loss": 0.004475796734914184, "time": 0.5153226613998413, "epoch": 8, "memory": 15575, "step": 15583} +{"lr": 4.853163291548285e-05, "data_time": 0.0015436172485351562, "loss": 0.005731246387585998, "time": 0.5148451089859009, "epoch": 8, "memory": 15575, "step": 15683} +{"accuracy/top1": 99.7634048461914, "data_time": 0.17594437599182128, "time": 0.5637810707092286, "step": 8} +{"lr": 4.5450966211010106e-05, "data_time": 0.0014884233474731444, "loss": 0.006236531725153327, "time": 0.5239831447601319, "epoch": 9, "memory": 15575, "step": 15852} +{"lr": 4.368257232393176e-05, "data_time": 0.0016297578811645507, "loss": 0.004661645088344812, "time": 0.5194293260574341, "epoch": 9, "memory": 15575, "step": 15952} +{"lr": 4.195524424634855e-05, "data_time": 0.001526045799255371, "loss": 0.009385608416050673, "time": 0.517946434020996, "epoch": 9, "memory": 15575, "step": 16052} +{"lr": 4.0269421695945e-05, "data_time": 0.0015024662017822266, "loss": 0.003542406717315316, "time": 0.5254890441894531, "epoch": 9, "memory": 15575, "step": 16152} +{"lr": 3.862553382454128e-05, "data_time": 0.0015067338943481445, "loss": 0.002612195489928126, "time": 0.5146977186203003, "epoch": 9, "memory": 15575, "step": 16252} +{"lr": 3.702399910884571e-05, "data_time": 0.0015241146087646485, "loss": 0.005953837977722287, "time": 0.532414722442627, "epoch": 9, "memory": 15575, "step": 16352} +{"lr": 3.5465225243925684e-05, "data_time": 0.0015434026718139648, "loss": 0.005354569805786013, "time": 0.5145599126815796, "epoch": 9, "memory": 15575, "step": 16452} +{"lr": 3.394960903942234e-05, "data_time": 0.0016269922256469727, "loss": 0.00658784625120461, "time": 0.5209419250488281, "epoch": 9, "memory": 15575, "step": 16552} +{"lr": 3.247753631853657e-05, "data_time": 0.0016747236251831055, "loss": 0.0029929411597549916, "time": 0.5149687767028809, "epoch": 9, "memory": 15575, "step": 16652} +{"lr": 3.104938181981207e-05, "data_time": 0.0014681339263916016, "loss": 0.00320582278072834, "time": 0.5388503789901733, "epoch": 9, "memory": 15575, "step": 16752} +{"lr": 2.966550910173992e-05, "data_time": 0.0017337560653686523, "loss": 0.004748571198433638, "time": 0.5170167446136474, "epoch": 9, "memory": 15575, "step": 16852} +{"lr": 2.8326270450208946e-05, "data_time": 0.0016788244247436523, "loss": 0.003234737832099199, "time": 0.5318400144577027, "epoch": 9, "memory": 15575, "step": 16952} +{"lr": 2.7032006788826352e-05, "data_time": 0.0016044855117797851, "loss": 0.005758630717173219, "time": 0.5228027820587158, "epoch": 9, "memory": 15575, "step": 17052} +{"lr": 2.5783047592130032e-05, "data_time": 0.0016762018203735352, "loss": 0.005023986147716641, "time": 0.5210402488708497, "epoch": 9, "memory": 15575, "step": 17152} +{"lr": 2.4579710801716043e-05, "data_time": 0.0014997243881225586, "loss": 0.007177395233884454, "time": 0.5191366910934448, "epoch": 9, "memory": 15575, "step": 17252} +{"lr": 2.3422302745301813e-05, "data_time": 0.0015058517456054688, "loss": 0.0028006411157548428, "time": 0.5155412912368774, "epoch": 9, "memory": 15575, "step": 17352} +{"lr": 2.2311118058745624e-05, "data_time": 0.0016066074371337891, "loss": 0.003138246154412627, "time": 0.5212033271789551, "epoch": 9, "memory": 15575, "step": 17452} +{"lr": 2.1246439611042613e-05, "data_time": 0.001579618453979492, "loss": 0.004682178189978003, "time": 0.5149863719940185, "epoch": 9, "memory": 15575, "step": 17552} +{"lr": 2.0228538432315955e-05, "data_time": 0.0015074491500854492, "loss": 0.002692938270047307, "time": 0.5146836996078491, "epoch": 9, "memory": 15575, "step": 17652} +{"accuracy/top1": 99.78775787353516, "data_time": 0.15599579811096193, "time": 0.5418736934661865, "step": 9} +{"lr": 1.861533120597442e-05, "data_time": 0.0016471624374389648, "loss": 0.007706991024315357, "time": 0.5160431623458862, "epoch": 10, "memory": 15575, "step": 17821} +{"lr": 1.7724513410024274e-05, "data_time": 0.0014929056167602539, "loss": 0.005444771517068148, "time": 0.5249947786331177, "epoch": 10, "memory": 15575, "step": 17921} +{"lr": 1.6881369443034795e-05, "data_time": 0.0014614105224609376, "loss": 0.0072738665621727705, "time": 0.5152039527893066, "epoch": 10, "memory": 15575, "step": 18021} +{"lr": 1.6086113940151352e-05, "data_time": 0.0019154310226440429, "loss": 0.004786178935319185, "time": 0.525653076171875, "epoch": 10, "memory": 15575, "step": 18121} +{"lr": 1.5338949345780815e-05, "data_time": 0.002294325828552246, "loss": 0.008724371390417218, "time": 0.5185094594955444, "epoch": 10, "memory": 15575, "step": 18221} +{"lr": 1.4640065862056652e-05, "data_time": 0.0016380786895751954, "loss": 0.006897278735414148, "time": 0.5171728134155273, "epoch": 10, "memory": 15575, "step": 18321} +{"lr": 1.3989641400419702e-05, "data_time": 0.0017490863800048828, "loss": 0.002737410692498088, "time": 0.5169026374816894, "epoch": 10, "memory": 15575, "step": 18421} +{"lr": 1.33878415363284e-05, "data_time": 0.0015343427658081055, "loss": 0.006530545791611076, "time": 0.5186675071716309, "epoch": 10, "memory": 15575, "step": 18521} +{"lr": 1.2834819467108807e-05, "data_time": 0.0015212774276733398, "loss": 0.006148951360955834, "time": 0.5152447700500489, "epoch": 10, "memory": 15575, "step": 18621} +{"lr": 1.2330715972956002e-05, "data_time": 0.001495218276977539, "loss": 0.0015209995210170746, "time": 0.5145811080932617, "epoch": 10, "memory": 15575, "step": 18721} +{"lr": 1.187565938109641e-05, "data_time": 0.0015730619430541991, "loss": 0.004278835840523243, "time": 0.5164597511291504, "epoch": 10, "memory": 15575, "step": 18821} +{"lr": 1.1469765533119946e-05, "data_time": 0.001512455940246582, "loss": 0.003100361954420805, "time": 0.5141446828842163, "epoch": 10, "memory": 15575, "step": 18921} +{"lr": 1.1113137755490966e-05, "data_time": 0.001491880416870117, "loss": 0.0062726660631597046, "time": 0.5143632173538208, "epoch": 10, "memory": 15575, "step": 19021} +{"lr": 1.0805866833244799e-05, "data_time": 0.0015006780624389649, "loss": 0.0017133825924247504, "time": 0.5168334722518921, "epoch": 10, "memory": 15575, "step": 19121} +{"lr": 1.0548030986877194e-05, "data_time": 0.001772165298461914, "loss": 0.005550199327990413, "time": 0.5161619663238526, "epoch": 10, "memory": 15575, "step": 19221} +{"lr": 1.0339695852431895e-05, "data_time": 0.002076864242553711, "loss": 0.0031654906924813986, "time": 0.5176642894744873, "epoch": 10, "memory": 15575, "step": 19321} +{"lr": 1.0180914464792136e-05, "data_time": 0.0017699480056762695, "loss": 0.0029759655240923165, "time": 0.5150105714797973, "epoch": 10, "memory": 15575, "step": 19421} +{"lr": 1.007172724417983e-05, "data_time": 0.0015714168548583984, "loss": 0.00298995072953403, "time": 0.5156848907470704, "epoch": 10, "memory": 15575, "step": 19521} +{"lr": 1.0012161985865822e-05, "data_time": 0.001559138298034668, "loss": 0.0024837352801114322, "time": 0.5152529001235961, "epoch": 10, "memory": 15575, "step": 19621} +{"accuracy/top1": 99.83299255371094, "data_time": 0.18442506790161134, "time": 0.5707194010416666, "step": 10} diff --git a/convnext_small_4xb256_IF_1m_lr4e-4_aug_1e-1/20230606_050018/vis_data/config.py b/convnext_small_4xb256_IF_1m_lr4e-4_aug_1e-1/20230606_050018/vis_data/config.py new file mode 100644 index 0000000000000000000000000000000000000000..6cbb68ea0841a58c5da7a681ef0548c7d7ab66f7 --- /dev/null +++ b/convnext_small_4xb256_IF_1m_lr4e-4_aug_1e-1/20230606_050018/vis_data/config.py @@ -0,0 +1,264 @@ +optim_wrapper = dict( + optimizer=dict( + type='AdamW', + lr=0.0004, + weight_decay=0.05, + eps=1e-08, + betas=(0.9, 0.999), + _scope_='mmpretrain'), + paramwise_cfg=dict( + norm_decay_mult=0.0, + bias_decay_mult=0.0, + flat_decay_mult=0.0, + custom_keys=dict({ + '.absolute_pos_embed': dict(decay_mult=0.0), + '.relative_position_bias_table': dict(decay_mult=0.0) + })), + type='AmpOptimWrapper', + dtype='bfloat16', + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=False, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=4096) +model = dict( + type='ImageClassifier', + backbone=dict( + frozen_stages=2, + type='ConvNeXt', + arch='small', + drop_path_rate=0.4, + init_cfg=dict( + type='Pretrained', + checkpoint= + 'https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth', + prefix='backbone')), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=768, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + init_cfg=None), + init_cfg=dict( + type='TruncNormal', layer=['Conv2d', 'Linear'], std=0.02, bias=0.0), + train_cfg=None) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/IF80w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/if-dpmsolver++-50-20w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/cc1m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/if-dpmsolver++-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/if-dpmsolver++-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/convnext_small_4xb256_IF_1m_lr4e-4_aug_1e-1' diff --git a/convnext_small_4xb256_IF_1m_lr4e-4_aug_1e-1/20230606_050018/vis_data/events.out.tfevents.1685998830.SH-IDC1-10-140-24-119.203264.0 b/convnext_small_4xb256_IF_1m_lr4e-4_aug_1e-1/20230606_050018/vis_data/events.out.tfevents.1685998830.SH-IDC1-10-140-24-119.203264.0 new file mode 100644 index 0000000000000000000000000000000000000000..253e04c59531d82cb1fe33ae41835d36e7b1c9d2 --- /dev/null +++ b/convnext_small_4xb256_IF_1m_lr4e-4_aug_1e-1/20230606_050018/vis_data/events.out.tfevents.1685998830.SH-IDC1-10-140-24-119.203264.0 @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:440634b957a952b0c027cbf9ef03bc831f9031e30b97f760e2ba98f14d651138 +size 1939669 diff --git a/convnext_small_4xb256_IF_1m_lr4e-4_aug_1e-1/20230606_050018/vis_data/scalars.json b/convnext_small_4xb256_IF_1m_lr4e-4_aug_1e-1/20230606_050018/vis_data/scalars.json new file mode 100644 index 0000000000000000000000000000000000000000..982821765f6724cd912ce3a691dc9a71e530a3b5 --- /dev/null +++ b/convnext_small_4xb256_IF_1m_lr4e-4_aug_1e-1/20230606_050018/vis_data/scalars.json @@ -0,0 +1,200 @@ +{"lr": 0.0003999756737991541, "data_time": 0.0018261909484863282, "loss": 0.027601023856550454, "time": 0.5191150665283203, "epoch": 1, "memory": 15578, "step": 100} +{"lr": 0.00039990171605141573, "data_time": 0.0018618345260620118, "loss": 0.0271825865842402, "time": 0.5322962045669556, "epoch": 1, "memory": 15578, "step": 200} +{"lr": 0.00039977814310179305, "data_time": 0.0018401622772216796, "loss": 0.02326606847345829, "time": 0.515569543838501, "epoch": 1, "memory": 15578, "step": 300} +{"lr": 0.00039960498640766363, "data_time": 0.0019962310791015623, "loss": 0.01996779362671077, "time": 0.5271080255508422, "epoch": 1, "memory": 15578, "step": 400} +{"lr": 0.00039938229004870185, "data_time": 0.001923537254333496, "loss": 0.020116635179147125, "time": 0.517112421989441, "epoch": 1, "memory": 15578, "step": 500} +{"lr": 0.00039911011071566005, "data_time": 0.0020764827728271484, "loss": 0.016853199154138566, "time": 0.5174846649169922, "epoch": 1, "memory": 15578, "step": 600} +{"lr": 0.0003987885176959352, "data_time": 0.0018217325210571288, "loss": 0.013987581757828594, "time": 0.5222087144851685, "epoch": 1, "memory": 15578, "step": 700} +{"lr": 0.0003984175928559308, "data_time": 0.0016354560852050782, "loss": 0.013943490013480187, "time": 0.5155891418457031, "epoch": 1, "memory": 15578, "step": 800} +{"lr": 0.00039799743062021714, "data_time": 0.001679682731628418, "loss": 0.01380026382394135, "time": 0.515896224975586, "epoch": 1, "memory": 15578, "step": 900} +{"lr": 0.0003975281379474946, "data_time": 0.0019191980361938476, "loss": 0.018961887620389462, "time": 0.5156231641769409, "epoch": 1, "memory": 15578, "step": 1000} +{"lr": 0.00039700983430336476, "data_time": 0.0016831398010253907, "loss": 0.01128587475977838, "time": 0.5164831638336181, "epoch": 1, "memory": 15578, "step": 1100} +{"lr": 0.0003964426516299177, "data_time": 0.0017139196395874023, "loss": 0.014654634799808264, "time": 0.5162296772003174, "epoch": 1, "memory": 15578, "step": 1200} +{"lr": 0.0003958267343121465, "data_time": 0.0017230987548828125, "loss": 0.009067555936053396, "time": 0.520532774925232, "epoch": 1, "memory": 15578, "step": 1300} +{"lr": 0.00039516223914119037, "data_time": 0.0016436100006103516, "loss": 0.01604834762401879, "time": 0.5156150817871094, "epoch": 1, "memory": 15578, "step": 1400} +{"lr": 0.0003944493352744212, "data_time": 0.001742100715637207, "loss": 0.019340876024216413, "time": 0.52655189037323, "epoch": 1, "memory": 15578, "step": 1500} +{"lr": 0.0003936882041923822, "data_time": 0.0016018867492675782, "loss": 0.018161868583410977, "time": 0.5159793376922608, "epoch": 1, "memory": 15578, "step": 1600} +{"lr": 0.0003928790396525886, "data_time": 0.0016326189041137695, "loss": 0.012575972964987159, "time": 0.5178181886672973, "epoch": 1, "memory": 15578, "step": 1700} +{"lr": 0.00039202204764020483, "data_time": 0.001667356491088867, "loss": 0.01994553143158555, "time": 0.5147627353668213, "epoch": 1, "memory": 15578, "step": 1800} +{"lr": 0.00039111744631560716, "data_time": 0.0015305757522583007, "loss": 0.012879175506532193, "time": 0.514723825454712, "epoch": 1, "memory": 15578, "step": 1900} +{"accuracy/top1": 96.25273895263672, "data_time": 0.13952156593059672, "time": 0.6054766918050831, "step": 1} +{"lr": 0.0003894811013774212, "data_time": 0.02893216609954834, "loss": 0.015500097675248981, "time": 0.5442700624465943, "epoch": 2, "memory": 15578, "step": 2069} +{"lr": 0.00038844960758563737, "data_time": 0.001591062545776367, "loss": 0.010155213065445423, "time": 0.5309667825698853, "epoch": 2, "memory": 15575, "step": 2169} +{"lr": 0.00038737141390069504, "data_time": 0.0356330156326294, "loss": 0.015067616989836097, "time": 0.5833153009414673, "epoch": 2, "memory": 15575, "step": 2269} +{"lr": 0.0003862467947932271, "data_time": 0.0016663551330566406, "loss": 0.014586024405434727, "time": 0.5234311342239379, "epoch": 2, "memory": 15575, "step": 2369} +{"lr": 0.000385076036552165, "data_time": 0.001562786102294922, "loss": 0.014382174983620644, "time": 0.5206002235412598, "epoch": 2, "memory": 15575, "step": 2469} +{"lr": 0.00038385943721186026, "data_time": 0.0015935182571411132, "loss": 0.012308013578876852, "time": 0.5144658327102661, "epoch": 2, "memory": 15575, "step": 2569} +{"lr": 0.0003825973064762139, "data_time": 0.2808521032333374, "loss": 0.013550773914903402, "time": 1.0783581256866455, "epoch": 2, "memory": 15575, "step": 2669} +{"lr": 0.000381289965639839, "data_time": 0.001841425895690918, "loss": 0.007510989252477884, "time": 0.6251338005065918, "epoch": 2, "memory": 15575, "step": 2769} +{"lr": 0.00037993774750626663, "data_time": 0.0014961481094360352, "loss": 0.012146124336868524, "time": 0.5329279184341431, "epoch": 2, "memory": 15575, "step": 2869} +{"lr": 0.00037854099630322966, "data_time": 0.0015124082565307617, "loss": 0.011426049657166005, "time": 0.7000055313110352, "epoch": 2, "memory": 15575, "step": 2969} +{"lr": 0.00037710006759503174, "data_time": 0.0015774250030517578, "loss": 0.013603793689981103, "time": 0.5768113613128663, "epoch": 2, "memory": 15575, "step": 3069} +{"lr": 0.00037561532819203284, "data_time": 0.0015952587127685547, "loss": 0.014887896692380309, "time": 0.5170141458511353, "epoch": 2, "memory": 15575, "step": 3169} +{"lr": 0.00037408715605727376, "data_time": 0.001523876190185547, "loss": 0.006633514491841197, "time": 0.5152340412139893, "epoch": 2, "memory": 15575, "step": 3269} +{"lr": 0.00037251594021025933, "data_time": 0.0016954421997070312, "loss": 0.014243566524237394, "time": 0.5231961488723755, "epoch": 2, "memory": 15575, "step": 3369} +{"lr": 0.00037090208062792615, "data_time": 0.02590627670288086, "loss": 0.012183473305776715, "time": 0.5400282144546509, "epoch": 2, "memory": 15575, "step": 3469} +{"lr": 0.00036924598814282443, "data_time": 0.001516437530517578, "loss": 0.006949825631454587, "time": 0.5146150112152099, "epoch": 2, "memory": 15575, "step": 3569} +{"lr": 0.00036754808433853343, "data_time": 9.311471533775329, "loss": 0.012170182494446635, "time": 9.849288773536681, "epoch": 2, "memory": 15575, "step": 3669} +{"lr": 0.00036580880144233954, "data_time": 0.0014627695083618164, "loss": 0.014173916587606073, "time": 0.5150933504104614, "epoch": 2, "memory": 15575, "step": 3769} +{"lr": 0.0003640285822152096, "data_time": 0.0016908884048461915, "loss": 0.010514096030965448, "time": 0.5575487852096558, "epoch": 2, "memory": 15575, "step": 3869} +{"accuracy/top1": 98.80310821533203, "data_time": 0.1626681089401245, "time": 0.5489826043446858, "step": 2} +{"lr": 0.0003609282333974528, "data_time": 0.0014587163925170899, "loss": 0.00871057934127748, "time": 0.5471591472625732, "epoch": 3, "memory": 15575, "step": 4038} +{"lr": 0.0003590401733608027, "data_time": 0.0020009279251098633, "loss": 0.009571832278743387, "time": 0.5169068813323975, "epoch": 3, "memory": 15575, "step": 4138} +{"lr": 0.00035711290005063985, "data_time": 0.001433587074279785, "loss": 0.008919509826228023, "time": 0.5254595518112183, "epoch": 3, "memory": 15575, "step": 4238} +{"lr": 0.0003551469040837546, "data_time": 0.0015285491943359375, "loss": 0.015251696854829789, "time": 0.5239350318908691, "epoch": 3, "memory": 15575, "step": 4338} +{"lr": 0.00035314268593437953, "data_time": 0.0015092849731445312, "loss": 0.016632861783728005, "time": 0.5233675003051758, "epoch": 3, "memory": 15575, "step": 4438} +{"lr": 0.00035110075580678697, "data_time": 0.0014284849166870117, "loss": 0.013199200900271536, "time": 0.5159765243530273, "epoch": 3, "memory": 15575, "step": 4538} +{"lr": 0.00034902163350540724, "data_time": 0.0014449596405029298, "loss": 0.006127667799592018, "time": 0.5216275691986084, "epoch": 3, "memory": 15575, "step": 4638} +{"lr": 0.0003469058483025052, "data_time": 0.0015788793563842774, "loss": 0.012180058006197213, "time": 0.5343188047409058, "epoch": 3, "memory": 15575, "step": 4738} +{"lr": 0.0003447539388034461, "data_time": 0.001650857925415039, "loss": 0.01343419598415494, "time": 0.554154348373413, "epoch": 3, "memory": 15575, "step": 4838} +{"lr": 0.0003425664528095857, "data_time": 0.001610136032104492, "loss": 0.014271644130349159, "time": 0.528766679763794, "epoch": 3, "memory": 15575, "step": 4938} +{"lr": 0.00034034394717881845, "data_time": 0.0015502691268920899, "loss": 0.01365456241182983, "time": 0.5186124563217163, "epoch": 3, "memory": 15575, "step": 5038} +{"lr": 0.00033808698768382034, "data_time": 0.0019941091537475585, "loss": 0.012560153007507324, "time": 0.522815203666687, "epoch": 3, "memory": 15575, "step": 5138} +{"lr": 0.00033579614886802443, "data_time": 0.0015717506408691405, "loss": 0.004915389930829406, "time": 0.5420629978179932, "epoch": 3, "memory": 15575, "step": 5238} +{"lr": 0.00033347201389936123, "data_time": 0.0014863252639770509, "loss": 0.01211695084348321, "time": 0.5369882583618164, "epoch": 3, "memory": 15575, "step": 5338} +{"lr": 0.0003311151744218034, "data_time": 0.0015731573104858399, "loss": 0.008279312448576093, "time": 0.5183897495269776, "epoch": 3, "memory": 15575, "step": 5438} +{"lr": 0.0003287262304047557, "data_time": 0.001571822166442871, "loss": 0.012255780724808573, "time": 0.5148802042007447, "epoch": 3, "memory": 15575, "step": 5538} +{"lr": 0.0003263057899903226, "data_time": 0.0017143487930297852, "loss": 0.01518821376375854, "time": 0.5159362554550171, "epoch": 3, "memory": 15575, "step": 5638} +{"lr": 0.00032385446933849547, "data_time": 0.014629149436950683, "loss": 0.013086866587400436, "time": 0.5274892568588256, "epoch": 3, "memory": 15575, "step": 5738} +{"lr": 0.0003213728924703014, "data_time": 0.0015346765518188476, "loss": 0.008438705652952194, "time": 0.5169786930084228, "epoch": 3, "memory": 15575, "step": 5838} +{"accuracy/top1": 99.24845886230469, "data_time": 0.05241813659667969, "time": 0.43549114068349204, "step": 3} +{"lr": 0.00031711202351665616, "data_time": 0.0016689777374267578, "loss": 0.007415917282924056, "time": 0.5154834747314453, "epoch": 4, "memory": 15575, "step": 6007} +{"lr": 0.0003145522137064149, "data_time": 0.0014388084411621094, "loss": 0.012401284277439117, "time": 0.5166043281555176, "epoch": 4, "memory": 15575, "step": 6107} +{"lr": 0.0003119645157107334, "data_time": 0.001614832878112793, "loss": 0.007888593943789601, "time": 0.5158061981201172, "epoch": 4, "memory": 15575, "step": 6207} +{"lr": 0.00030934958826756915, "data_time": 0.0015721321105957031, "loss": 0.011648882133886217, "time": 0.518235182762146, "epoch": 4, "memory": 15575, "step": 6307} +{"lr": 0.0003067080970465515, "data_time": 0.0015164613723754883, "loss": 0.008028108440339566, "time": 0.5149527311325073, "epoch": 4, "memory": 15575, "step": 6407} +{"lr": 0.00030404071447952345, "data_time": 0.001444864273071289, "loss": 0.007447708304971457, "time": 0.5185810565948487, "epoch": 4, "memory": 15575, "step": 6507} +{"lr": 0.0003013481195893653, "data_time": 0.0015978813171386719, "loss": 0.013097852282226086, "time": 0.5151091337203979, "epoch": 4, "memory": 15575, "step": 6607} +{"lr": 0.0002986309978171385, "data_time": 0.0015954971313476562, "loss": 0.008581136912107467, "time": 0.5163673639297486, "epoch": 4, "memory": 15575, "step": 6707} +{"lr": 0.0002958900408475955, "data_time": 0.0016129016876220703, "loss": 0.005765685345977545, "time": 0.5147150039672852, "epoch": 4, "memory": 15575, "step": 6807} +{"lr": 0.0002931259464331015, "data_time": 0.001521730422973633, "loss": 0.009429838322103024, "time": 0.5155909299850464, "epoch": 4, "memory": 15575, "step": 6907} +{"lr": 0.00029033941821601176, "data_time": 0.0015855550765991211, "loss": 0.011490005301311612, "time": 0.5152943849563598, "epoch": 4, "memory": 15575, "step": 7007} +{"lr": 0.0002875311655495464, "data_time": 0.0014770030975341797, "loss": 0.006993796117603779, "time": 0.5145813465118408, "epoch": 4, "memory": 15575, "step": 7107} +{"lr": 0.0002847019033172168, "data_time": 0.001403641700744629, "loss": 0.011287294235080481, "time": 0.5149314403533936, "epoch": 4, "memory": 15575, "step": 7207} +{"lr": 0.0002818523517508384, "data_time": 0.0015740394592285156, "loss": 0.006218388443812728, "time": 0.5160921812057495, "epoch": 4, "memory": 15575, "step": 7307} +{"lr": 0.00027898323624718594, "data_time": 0.0015975475311279298, "loss": 0.006412314670160413, "time": 0.5145673274993896, "epoch": 4, "memory": 15575, "step": 7407} +{"lr": 0.000276095287183333, "data_time": 0.002046346664428711, "loss": 0.008772781072184443, "time": 0.5230429172515869, "epoch": 4, "memory": 15575, "step": 7507} +{"lr": 0.00027318923973072227, "data_time": 0.0014729261398315429, "loss": 0.00985402320511639, "time": 0.5146645307540894, "epoch": 4, "memory": 15575, "step": 7607} +{"lr": 0.00027026583366801727, "data_time": 0.0015037775039672852, "loss": 0.0145890639629215, "time": 0.5196113109588623, "epoch": 4, "memory": 15575, "step": 7707} +{"lr": 0.0002673258131927808, "data_time": 0.0014482736587524414, "loss": 0.012970298808068038, "time": 0.5275277137756348, "epoch": 4, "memory": 15575, "step": 7807} +{"accuracy/top1": 99.377197265625, "data_time": 0.14185065428415936, "time": 0.5245868841807048, "step": 4} +{"lr": 0.0002623215076436499, "data_time": 0.0015915870666503907, "loss": 0.01226809537038207, "time": 0.5155995845794678, "epoch": 5, "memory": 15575, "step": 7976} +{"lr": 0.00025934052007928784, "data_time": 0.00157623291015625, "loss": 0.013417656626552343, "time": 0.5149784326553345, "epoch": 5, "memory": 15575, "step": 8076} +{"lr": 0.0002563456993074567, "data_time": 0.0016045808792114259, "loss": 0.006499800039455295, "time": 0.5153000116348266, "epoch": 5, "memory": 15575, "step": 8176} +{"lr": 0.00025333780770542177, "data_time": 0.0015806436538696289, "loss": 0.010765056544914841, "time": 0.5195995092391967, "epoch": 5, "memory": 15575, "step": 8276} +{"lr": 0.00025031761097782693, "data_time": 0.001491379737854004, "loss": 0.008847670070827008, "time": 0.51485755443573, "epoch": 5, "memory": 15575, "step": 8376} +{"lr": 0.0002472858779617731, "data_time": 0.0015269041061401368, "loss": 0.006283209845423699, "time": 0.5262893915176392, "epoch": 5, "memory": 15575, "step": 8476} +{"lr": 0.0002442433804310988, "data_time": 0.0015306472778320312, "loss": 0.00790382451377809, "time": 0.5153609275817871, "epoch": 5, "memory": 15575, "step": 8576} +{"lr": 0.00024119089289991304, "data_time": 0.0014877557754516602, "loss": 0.0048039590008556845, "time": 0.5214916706085205, "epoch": 5, "memory": 15575, "step": 8676} +{"lr": 0.00023812919242543494, "data_time": 0.001499772071838379, "loss": 0.0070088079664856195, "time": 0.514876914024353, "epoch": 5, "memory": 15575, "step": 8776} +{"lr": 0.00023505905841017684, "data_time": 0.0015369176864624024, "loss": 0.011009765835478903, "time": 0.5175901651382446, "epoch": 5, "memory": 15575, "step": 8876} +{"lr": 0.00023198127240353698, "data_time": 0.0014757394790649414, "loss": 0.007589173363521695, "time": 0.5148205280303955, "epoch": 5, "memory": 15575, "step": 8976} +{"lr": 0.00022889661790284532, "data_time": 0.0015509366989135743, "loss": 0.008709517400711774, "time": 0.5236516475677491, "epoch": 5, "memory": 15575, "step": 9076} +{"lr": 0.000225805880153911, "data_time": 0.001490616798400879, "loss": 0.008318354608491063, "time": 0.5151755809783936, "epoch": 5, "memory": 15575, "step": 9176} +{"lr": 0.00022270984595112745, "data_time": 0.0015845537185668946, "loss": 0.00993576324544847, "time": 0.5195048809051513, "epoch": 5, "memory": 15575, "step": 9276} +{"lr": 0.00021960930343718077, "data_time": 0.001716303825378418, "loss": 0.0068968711886554955, "time": 0.5150446176528931, "epoch": 5, "memory": 15575, "step": 9376} +{"lr": 0.00021650504190241605, "data_time": 0.001743459701538086, "loss": 0.007159140007570386, "time": 0.5200284719467163, "epoch": 5, "memory": 15575, "step": 9476} +{"lr": 0.00021339785158391245, "data_time": 0.0015363693237304688, "loss": 0.011549755837768317, "time": 0.5149904251098633, "epoch": 5, "memory": 15575, "step": 9576} +{"lr": 0.00021028852346431534, "data_time": 0.0014166831970214844, "loss": 0.006046575214713812, "time": 0.5203746557235718, "epoch": 5, "memory": 15575, "step": 9676} +{"lr": 0.00020717784907047976, "data_time": 0.0014749288558959961, "loss": 0.006021002354100346, "time": 0.5229143142700196, "epoch": 5, "memory": 15575, "step": 9776} +{"accuracy/top1": 99.56160736083984, "data_time": 0.07427238623301188, "time": 0.46157800356547035, "step": 5} +{"lr": 0.0002019199632200586, "data_time": 0.001688075065612793, "loss": 0.008768323250114918, "time": 0.5154914140701294, "epoch": 6, "memory": 15575, "step": 9945} +{"lr": 0.00019880959773413697, "data_time": 0.0016509294509887695, "loss": 0.009114690544083714, "time": 0.5151286840438842, "epoch": 6, "memory": 15575, "step": 10045} +{"lr": 0.00019570080810944632, "data_time": 0.0016642332077026368, "loss": 0.011529341479763389, "time": 0.5162417888641357, "epoch": 6, "memory": 15575, "step": 10145} +{"lr": 0.00019259438573575975, "data_time": 0.001756906509399414, "loss": 0.009273130353540182, "time": 0.5160192251205444, "epoch": 6, "memory": 15575, "step": 10245} +{"lr": 0.00018949112140023044, "data_time": 0.001581430435180664, "loss": 0.009626655094325542, "time": 0.523019528388977, "epoch": 6, "memory": 15575, "step": 10345} +{"lr": 0.00018639180508608498, "data_time": 0.0015453100204467773, "loss": 0.008061748836189509, "time": 0.5278511047363281, "epoch": 6, "memory": 15575, "step": 10445} +{"lr": 0.00018329722577152123, "data_time": 0.0014238595962524415, "loss": 0.011502831522375346, "time": 0.5156992197036743, "epoch": 6, "memory": 15575, "step": 10545} +{"lr": 0.0001802081712288607, "data_time": 0.001892828941345215, "loss": 0.009451930969953537, "time": 0.5201520442962646, "epoch": 6, "memory": 15575, "step": 10645} +{"lr": 0.00017712542782401123, "data_time": 0.001572561264038086, "loss": 0.005022558709606528, "time": 0.5149688243865966, "epoch": 6, "memory": 15575, "step": 10745} +{"lr": 0.00017404978031628336, "data_time": 0.0017352104187011719, "loss": 0.00650988407433033, "time": 0.530413818359375, "epoch": 6, "memory": 15575, "step": 10845} +{"lr": 0.00017098201165861962, "data_time": 0.0016693353652954101, "loss": 0.008117228886112571, "time": 0.5151396274566651, "epoch": 6, "memory": 15575, "step": 10945} +{"lr": 0.00016792290279828072, "data_time": 0.0018962860107421876, "loss": 0.006933824857696891, "time": 0.5237004041671753, "epoch": 6, "memory": 15575, "step": 11045} +{"lr": 0.00016487323247804387, "data_time": 0.0017143964767456054, "loss": 0.006620669225230813, "time": 0.5151191234588623, "epoch": 6, "memory": 15575, "step": 11145} +{"lr": 0.00016183377703796263, "data_time": 0.0017733573913574219, "loss": 0.0039889195933938025, "time": 0.5164686441421509, "epoch": 6, "memory": 15575, "step": 11245} +{"lr": 0.00015880531021773676, "data_time": 0.001653432846069336, "loss": 0.006959619652479887, "time": 0.514758563041687, "epoch": 6, "memory": 15575, "step": 11345} +{"lr": 0.00015578860295974566, "data_time": 0.001767420768737793, "loss": 0.006196077354252339, "time": 0.5205242872238159, "epoch": 6, "memory": 15575, "step": 11445} +{"lr": 0.0001527844232127924, "data_time": 0.0015105485916137695, "loss": 0.008009692560881377, "time": 0.5164531230926513, "epoch": 6, "memory": 15575, "step": 11545} +{"lr": 0.000149793535736612, "data_time": 0.001721334457397461, "loss": 0.0076533563435077666, "time": 0.5187115669250488, "epoch": 6, "memory": 15575, "step": 11645} +{"lr": 0.00014681670190718876, "data_time": 0.0017121553421020508, "loss": 0.009085492882877588, "time": 0.5150811433792114, "epoch": 6, "memory": 15575, "step": 11745} +{"accuracy/top1": 99.65554809570312, "data_time": 0.17346168359120687, "time": 0.5613858938217163, "step": 6} +{"lr": 0.00014181991425636192, "data_time": 0.0016644716262817383, "loss": 0.01141593037173152, "time": 0.5167794704437256, "epoch": 7, "memory": 15575, "step": 11914} +{"lr": 0.00013888463509383338, "data_time": 0.0017268896102905274, "loss": 0.004260255442932248, "time": 0.5172703981399536, "epoch": 7, "memory": 15575, "step": 12014} +{"lr": 0.000135966186604959, "data_time": 0.00152740478515625, "loss": 0.010442282911390066, "time": 0.5161826372146606, "epoch": 7, "memory": 15575, "step": 12114} +{"lr": 0.0001330653117252752, "data_time": 0.0015177726745605469, "loss": 0.007746391324326396, "time": 0.51475031375885, "epoch": 7, "memory": 15575, "step": 12214} +{"lr": 0.00013018274891668863, "data_time": 0.0014116287231445313, "loss": 0.005766073195263744, "time": 0.5159190177917481, "epoch": 7, "memory": 15575, "step": 12314} +{"lr": 0.000127319231979489, "data_time": 0.001488971710205078, "loss": 0.008930237777531148, "time": 0.5161234855651855, "epoch": 7, "memory": 15575, "step": 12414} +{"lr": 0.0001244754898655487, "data_time": 0.001428842544555664, "loss": 0.005514901271089912, "time": 0.5152976512908936, "epoch": 7, "memory": 15575, "step": 12514} +{"lr": 0.0001216522464927582, "data_time": 0.0015146970748901368, "loss": 0.004802424600347877, "time": 0.5157246828079224, "epoch": 7, "memory": 15575, "step": 12614} +{"lr": 0.00011885022056074007, "data_time": 0.0014697551727294923, "loss": 0.0044075089506804945, "time": 0.5160865068435669, "epoch": 7, "memory": 15575, "step": 12714} +{"lr": 0.00011607012536789545, "data_time": 0.0016083002090454101, "loss": 0.004432854941114783, "time": 0.5158129930496216, "epoch": 7, "memory": 15575, "step": 12814} +{"lr": 0.00011331266862982092, "data_time": 0.0015475749969482422, "loss": 0.006024764757603407, "time": 0.514604139328003, "epoch": 7, "memory": 15575, "step": 12914} +{"lr": 0.00011057855229914932, "data_time": 0.0015961885452270507, "loss": 0.003552483953535557, "time": 0.5149378538131714, "epoch": 7, "memory": 15575, "step": 13014} +{"lr": 0.0001078684723868573, "data_time": 0.0015608549118041992, "loss": 0.009875560691580177, "time": 0.5153422832489014, "epoch": 7, "memory": 15575, "step": 13114} +{"lr": 0.00010518311878508484, "data_time": 0.0015584945678710938, "loss": 0.006951159657910466, "time": 0.5147438049316406, "epoch": 7, "memory": 15575, "step": 13214} +{"lr": 0.00010252317509151276, "data_time": 0.0016586065292358398, "loss": 0.01264434470795095, "time": 0.5150826692581176, "epoch": 7, "memory": 15575, "step": 13314} +{"lr": 9.98893184353425e-05, "data_time": 0.0016132593154907227, "loss": 0.005360524170100689, "time": 0.5149397850036621, "epoch": 7, "memory": 15575, "step": 13414} +{"lr": 9.728221930492262e-05, "data_time": 0.0014868974685668945, "loss": 0.00791693925857544, "time": 0.5155071496963501, "epoch": 7, "memory": 15575, "step": 13514} +{"lr": 9.470254137706575e-05, "data_time": 0.0014949083328247071, "loss": 0.005792217794805765, "time": 0.5146904945373535, "epoch": 7, "memory": 15575, "step": 13614} +{"lr": 9.215094134809975e-05, "data_time": 0.0014746904373168946, "loss": 0.006089560361579061, "time": 0.5150094985961914, "epoch": 7, "memory": 15575, "step": 13714} +{"accuracy/top1": 99.71817779541016, "data_time": 0.051924705505371094, "time": 0.43892056941986085, "step": 7} +{"lr": 8.790437228679661e-05, "data_time": 0.001678800582885742, "loss": 0.003303104406222701, "time": 0.5149797677993775, "epoch": 8, "memory": 15575, "step": 13883} +{"lr": 8.543150502338196e-05, "data_time": 0.001611018180847168, "loss": 0.00341587932780385, "time": 0.5222368717193604, "epoch": 8, "memory": 15575, "step": 13983} +{"lr": 8.298907574244125e-05, "data_time": 0.0017382144927978516, "loss": 0.003288163850083947, "time": 0.516103720664978, "epoch": 8, "memory": 15575, "step": 14083} +{"lr": 8.057770620156745e-05, "data_time": 0.0016478776931762695, "loss": 0.003450537147000432, "time": 0.5145237684249878, "epoch": 8, "memory": 15575, "step": 14183} +{"lr": 7.819801025162347e-05, "data_time": 0.0017688512802124024, "loss": 0.006510276347398758, "time": 0.514856481552124, "epoch": 8, "memory": 15575, "step": 14283} +{"lr": 7.585059368047708e-05, "data_time": 0.0015503644943237304, "loss": 0.004638825636357069, "time": 0.5301702737808227, "epoch": 8, "memory": 15575, "step": 14383} +{"lr": 7.353605405878812e-05, "data_time": 0.0016034603118896484, "loss": 0.0037703410256654025, "time": 0.5154333353042603, "epoch": 8, "memory": 15575, "step": 14483} +{"lr": 7.125498058788827e-05, "data_time": 0.0014626741409301757, "loss": 0.0036244274117052553, "time": 0.5160910606384277, "epoch": 8, "memory": 15575, "step": 14583} +{"lr": 6.900795394979036e-05, "data_time": 0.0015135526657104493, "loss": 0.006894288817420602, "time": 0.5152117729187011, "epoch": 8, "memory": 15575, "step": 14683} +{"lr": 6.679554615936711e-05, "data_time": 0.0015542030334472657, "loss": 0.006287419935688376, "time": 0.5149774551391602, "epoch": 8, "memory": 15575, "step": 14783} +{"lr": 6.461832041873565e-05, "data_time": 0.0018001556396484374, "loss": 0.005158521374687553, "time": 0.5160490512847901, "epoch": 8, "memory": 15575, "step": 14883} +{"lr": 6.247683097388664e-05, "data_time": 0.001475954055786133, "loss": 0.0035997253842651846, "time": 0.5145102739334106, "epoch": 8, "memory": 15575, "step": 14983} +{"lr": 6.037162297359161e-05, "data_time": 0.0015129327774047851, "loss": 0.0036070691421628, "time": 0.5158472776412963, "epoch": 8, "memory": 15575, "step": 15083} +{"lr": 5.830323233062794e-05, "data_time": 0.0015598058700561524, "loss": 0.004280280461534858, "time": 0.5166856288909912, "epoch": 8, "memory": 15575, "step": 15183} +{"lr": 5.6272185585353725e-05, "data_time": 0.0015989303588867187, "loss": 0.005098620429635048, "time": 0.5148352146148681, "epoch": 8, "memory": 15575, "step": 15283} +{"lr": 5.4278999771668655e-05, "data_time": 0.0014949321746826171, "loss": 0.002327388757839799, "time": 0.518629002571106, "epoch": 8, "memory": 15575, "step": 15383} +{"lr": 5.2324182285395304e-05, "data_time": 0.0016213178634643555, "loss": 0.006777837499976158, "time": 0.5155083894729614, "epoch": 8, "memory": 15575, "step": 15483} +{"lr": 5.0408230755113716e-05, "data_time": 0.0014513731002807617, "loss": 0.004475796734914184, "time": 0.5153226613998413, "epoch": 8, "memory": 15575, "step": 15583} +{"lr": 4.853163291548285e-05, "data_time": 0.0015436172485351562, "loss": 0.005731246387585998, "time": 0.5148451089859009, "epoch": 8, "memory": 15575, "step": 15683} +{"accuracy/top1": 99.7634048461914, "data_time": 0.17594437599182128, "time": 0.5637810707092286, "step": 8} +{"lr": 4.5450966211010106e-05, "data_time": 0.0014884233474731444, "loss": 0.006236531725153327, "time": 0.5239831447601319, "epoch": 9, "memory": 15575, "step": 15852} +{"lr": 4.368257232393176e-05, "data_time": 0.0016297578811645507, "loss": 0.004661645088344812, "time": 0.5194293260574341, "epoch": 9, "memory": 15575, "step": 15952} +{"lr": 4.195524424634855e-05, "data_time": 0.001526045799255371, "loss": 0.009385608416050673, "time": 0.517946434020996, "epoch": 9, "memory": 15575, "step": 16052} +{"lr": 4.0269421695945e-05, "data_time": 0.0015024662017822266, "loss": 0.003542406717315316, "time": 0.5254890441894531, "epoch": 9, "memory": 15575, "step": 16152} +{"lr": 3.862553382454128e-05, "data_time": 0.0015067338943481445, "loss": 0.002612195489928126, "time": 0.5146977186203003, "epoch": 9, "memory": 15575, "step": 16252} +{"lr": 3.702399910884571e-05, "data_time": 0.0015241146087646485, "loss": 0.005953837977722287, "time": 0.532414722442627, "epoch": 9, "memory": 15575, "step": 16352} +{"lr": 3.5465225243925684e-05, "data_time": 0.0015434026718139648, "loss": 0.005354569805786013, "time": 0.5145599126815796, "epoch": 9, "memory": 15575, "step": 16452} +{"lr": 3.394960903942234e-05, "data_time": 0.0016269922256469727, "loss": 0.00658784625120461, "time": 0.5209419250488281, "epoch": 9, "memory": 15575, "step": 16552} +{"lr": 3.247753631853657e-05, "data_time": 0.0016747236251831055, "loss": 0.0029929411597549916, "time": 0.5149687767028809, "epoch": 9, "memory": 15575, "step": 16652} +{"lr": 3.104938181981207e-05, "data_time": 0.0014681339263916016, "loss": 0.00320582278072834, "time": 0.5388503789901733, "epoch": 9, "memory": 15575, "step": 16752} +{"lr": 2.966550910173992e-05, "data_time": 0.0017337560653686523, "loss": 0.004748571198433638, "time": 0.5170167446136474, "epoch": 9, "memory": 15575, "step": 16852} +{"lr": 2.8326270450208946e-05, "data_time": 0.0016788244247436523, "loss": 0.003234737832099199, "time": 0.5318400144577027, "epoch": 9, "memory": 15575, "step": 16952} +{"lr": 2.7032006788826352e-05, "data_time": 0.0016044855117797851, "loss": 0.005758630717173219, "time": 0.5228027820587158, "epoch": 9, "memory": 15575, "step": 17052} +{"lr": 2.5783047592130032e-05, "data_time": 0.0016762018203735352, "loss": 0.005023986147716641, "time": 0.5210402488708497, "epoch": 9, "memory": 15575, "step": 17152} +{"lr": 2.4579710801716043e-05, "data_time": 0.0014997243881225586, "loss": 0.007177395233884454, "time": 0.5191366910934448, "epoch": 9, "memory": 15575, "step": 17252} +{"lr": 2.3422302745301813e-05, "data_time": 0.0015058517456054688, "loss": 0.0028006411157548428, "time": 0.5155412912368774, "epoch": 9, "memory": 15575, "step": 17352} +{"lr": 2.2311118058745624e-05, "data_time": 0.0016066074371337891, "loss": 0.003138246154412627, "time": 0.5212033271789551, "epoch": 9, "memory": 15575, "step": 17452} +{"lr": 2.1246439611042613e-05, "data_time": 0.001579618453979492, "loss": 0.004682178189978003, "time": 0.5149863719940185, "epoch": 9, "memory": 15575, "step": 17552} +{"lr": 2.0228538432315955e-05, "data_time": 0.0015074491500854492, "loss": 0.002692938270047307, "time": 0.5146836996078491, "epoch": 9, "memory": 15575, "step": 17652} +{"accuracy/top1": 99.78775787353516, "data_time": 0.15599579811096193, "time": 0.5418736934661865, "step": 9} +{"lr": 1.861533120597442e-05, "data_time": 0.0016471624374389648, "loss": 0.007706991024315357, "time": 0.5160431623458862, "epoch": 10, "memory": 15575, "step": 17821} +{"lr": 1.7724513410024274e-05, "data_time": 0.0014929056167602539, "loss": 0.005444771517068148, "time": 0.5249947786331177, "epoch": 10, "memory": 15575, "step": 17921} +{"lr": 1.6881369443034795e-05, "data_time": 0.0014614105224609376, "loss": 0.0072738665621727705, "time": 0.5152039527893066, "epoch": 10, "memory": 15575, "step": 18021} +{"lr": 1.6086113940151352e-05, "data_time": 0.0019154310226440429, "loss": 0.004786178935319185, "time": 0.525653076171875, "epoch": 10, "memory": 15575, "step": 18121} +{"lr": 1.5338949345780815e-05, "data_time": 0.002294325828552246, "loss": 0.008724371390417218, "time": 0.5185094594955444, "epoch": 10, "memory": 15575, "step": 18221} +{"lr": 1.4640065862056652e-05, "data_time": 0.0016380786895751954, "loss": 0.006897278735414148, "time": 0.5171728134155273, "epoch": 10, "memory": 15575, "step": 18321} +{"lr": 1.3989641400419702e-05, "data_time": 0.0017490863800048828, "loss": 0.002737410692498088, "time": 0.5169026374816894, "epoch": 10, "memory": 15575, "step": 18421} +{"lr": 1.33878415363284e-05, "data_time": 0.0015343427658081055, "loss": 0.006530545791611076, "time": 0.5186675071716309, "epoch": 10, "memory": 15575, "step": 18521} +{"lr": 1.2834819467108807e-05, "data_time": 0.0015212774276733398, "loss": 0.006148951360955834, "time": 0.5152447700500489, "epoch": 10, "memory": 15575, "step": 18621} +{"lr": 1.2330715972956002e-05, "data_time": 0.001495218276977539, "loss": 0.0015209995210170746, "time": 0.5145811080932617, "epoch": 10, "memory": 15575, "step": 18721} +{"lr": 1.187565938109641e-05, "data_time": 0.0015730619430541991, "loss": 0.004278835840523243, "time": 0.5164597511291504, "epoch": 10, "memory": 15575, "step": 18821} +{"lr": 1.1469765533119946e-05, "data_time": 0.001512455940246582, "loss": 0.003100361954420805, "time": 0.5141446828842163, "epoch": 10, "memory": 15575, "step": 18921} +{"lr": 1.1113137755490966e-05, "data_time": 0.001491880416870117, "loss": 0.0062726660631597046, "time": 0.5143632173538208, "epoch": 10, "memory": 15575, "step": 19021} +{"lr": 1.0805866833244799e-05, "data_time": 0.0015006780624389649, "loss": 0.0017133825924247504, "time": 0.5168334722518921, "epoch": 10, "memory": 15575, "step": 19121} +{"lr": 1.0548030986877194e-05, "data_time": 0.001772165298461914, "loss": 0.005550199327990413, "time": 0.5161619663238526, "epoch": 10, "memory": 15575, "step": 19221} +{"lr": 1.0339695852431895e-05, "data_time": 0.002076864242553711, "loss": 0.0031654906924813986, "time": 0.5176642894744873, "epoch": 10, "memory": 15575, "step": 19321} +{"lr": 1.0180914464792136e-05, "data_time": 0.0017699480056762695, "loss": 0.0029759655240923165, "time": 0.5150105714797973, "epoch": 10, "memory": 15575, "step": 19421} +{"lr": 1.007172724417983e-05, "data_time": 0.0015714168548583984, "loss": 0.00298995072953403, "time": 0.5156848907470704, "epoch": 10, "memory": 15575, "step": 19521} +{"lr": 1.0012161985865822e-05, "data_time": 0.001559138298034668, "loss": 0.0024837352801114322, "time": 0.5152529001235961, "epoch": 10, "memory": 15575, "step": 19621} +{"accuracy/top1": 99.83299255371094, "data_time": 0.18442506790161134, "time": 0.5707194010416666, "step": 10} diff --git a/convnext_small_4xb256_IF_1m_lr4e-4_aug_1e-1/20230606_050018/vis_data/vis_image/0000000.png_1.png b/convnext_small_4xb256_IF_1m_lr4e-4_aug_1e-1/20230606_050018/vis_data/vis_image/0000000.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..67696d0d96c5ac20dec419fe36f4fa85197b047b Binary files /dev/null and b/convnext_small_4xb256_IF_1m_lr4e-4_aug_1e-1/20230606_050018/vis_data/vis_image/0000000.png_1.png differ diff --git a/convnext_small_4xb256_IF_1m_lr4e-4_aug_1e-1/20230606_050018/vis_data/vis_image/0000000.png_10.png b/convnext_small_4xb256_IF_1m_lr4e-4_aug_1e-1/20230606_050018/vis_data/vis_image/0000000.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..d707d9be7335b096d8cf68e7ca07fa32bd48992a Binary files /dev/null and b/convnext_small_4xb256_IF_1m_lr4e-4_aug_1e-1/20230606_050018/vis_data/vis_image/0000000.png_10.png differ diff --git a/convnext_small_4xb256_IF_1m_lr4e-4_aug_1e-1/20230606_050018/vis_data/vis_image/0000000.png_2.png b/convnext_small_4xb256_IF_1m_lr4e-4_aug_1e-1/20230606_050018/vis_data/vis_image/0000000.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..7b29449cd69bb0ba614841f3d13e8ad90c60e19c Binary files /dev/null and b/convnext_small_4xb256_IF_1m_lr4e-4_aug_1e-1/20230606_050018/vis_data/vis_image/0000000.png_2.png differ diff --git a/convnext_small_4xb256_IF_1m_lr4e-4_aug_1e-1/20230606_050018/vis_data/vis_image/0000000.png_3.png b/convnext_small_4xb256_IF_1m_lr4e-4_aug_1e-1/20230606_050018/vis_data/vis_image/0000000.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..bd804ff14e827dce06f80c5d1dd06f72a52e201f Binary files /dev/null and b/convnext_small_4xb256_IF_1m_lr4e-4_aug_1e-1/20230606_050018/vis_data/vis_image/0000000.png_3.png differ diff --git a/convnext_small_4xb256_IF_1m_lr4e-4_aug_1e-1/20230606_050018/vis_data/vis_image/0000000.png_4.png b/convnext_small_4xb256_IF_1m_lr4e-4_aug_1e-1/20230606_050018/vis_data/vis_image/0000000.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..9c858d026c0a48cb44b1a30643e68f8f63c3da07 Binary files /dev/null and b/convnext_small_4xb256_IF_1m_lr4e-4_aug_1e-1/20230606_050018/vis_data/vis_image/0000000.png_4.png differ diff --git a/convnext_small_4xb256_IF_1m_lr4e-4_aug_1e-1/20230606_050018/vis_data/vis_image/0000000.png_5.png b/convnext_small_4xb256_IF_1m_lr4e-4_aug_1e-1/20230606_050018/vis_data/vis_image/0000000.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..87979f310aaa7fcada4d1ea90d89844474668534 Binary files /dev/null and b/convnext_small_4xb256_IF_1m_lr4e-4_aug_1e-1/20230606_050018/vis_data/vis_image/0000000.png_5.png differ diff --git a/convnext_small_4xb256_IF_1m_lr4e-4_aug_1e-1/20230606_050018/vis_data/vis_image/0000000.png_6.png b/convnext_small_4xb256_IF_1m_lr4e-4_aug_1e-1/20230606_050018/vis_data/vis_image/0000000.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..60857a19e4676138a0e73e152349ff74801e7859 Binary files /dev/null and b/convnext_small_4xb256_IF_1m_lr4e-4_aug_1e-1/20230606_050018/vis_data/vis_image/0000000.png_6.png differ diff --git a/convnext_small_4xb256_IF_1m_lr4e-4_aug_1e-1/20230606_050018/vis_data/vis_image/0000000.png_7.png b/convnext_small_4xb256_IF_1m_lr4e-4_aug_1e-1/20230606_050018/vis_data/vis_image/0000000.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..484e28db096cb6e82da962f017f540079379f379 Binary files /dev/null and b/convnext_small_4xb256_IF_1m_lr4e-4_aug_1e-1/20230606_050018/vis_data/vis_image/0000000.png_7.png differ diff --git a/convnext_small_4xb256_IF_1m_lr4e-4_aug_1e-1/20230606_050018/vis_data/vis_image/0000000.png_8.png b/convnext_small_4xb256_IF_1m_lr4e-4_aug_1e-1/20230606_050018/vis_data/vis_image/0000000.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..928f5a0004508659a808a386c97697f3d25cbc1b Binary files /dev/null and b/convnext_small_4xb256_IF_1m_lr4e-4_aug_1e-1/20230606_050018/vis_data/vis_image/0000000.png_8.png differ diff --git a/convnext_small_4xb256_IF_1m_lr4e-4_aug_1e-1/20230606_050018/vis_data/vis_image/0000000.png_9.png b/convnext_small_4xb256_IF_1m_lr4e-4_aug_1e-1/20230606_050018/vis_data/vis_image/0000000.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..0d0f147f74344c517a195317286e2c93de35c246 Binary files /dev/null and b/convnext_small_4xb256_IF_1m_lr4e-4_aug_1e-1/20230606_050018/vis_data/vis_image/0000000.png_9.png differ diff --git a/convnext_small_4xb256_IF_1m_lr4e-4_aug_1e-1/20230606_050018/vis_data/vis_image/000005174.jpg_1.png b/convnext_small_4xb256_IF_1m_lr4e-4_aug_1e-1/20230606_050018/vis_data/vis_image/000005174.jpg_1.png new file mode 100644 index 0000000000000000000000000000000000000000..d73b205bcdd08760dfd2cd87acaf3bd944da77ae Binary files /dev/null and b/convnext_small_4xb256_IF_1m_lr4e-4_aug_1e-1/20230606_050018/vis_data/vis_image/000005174.jpg_1.png differ diff --git a/convnext_small_4xb256_IF_1m_lr4e-4_aug_1e-1/20230606_050018/vis_data/vis_image/000005174.jpg_10.png b/convnext_small_4xb256_IF_1m_lr4e-4_aug_1e-1/20230606_050018/vis_data/vis_image/000005174.jpg_10.png new file mode 100644 index 0000000000000000000000000000000000000000..adab01f539f8eff1307dc2559aaddcafaac81c63 Binary files /dev/null and b/convnext_small_4xb256_IF_1m_lr4e-4_aug_1e-1/20230606_050018/vis_data/vis_image/000005174.jpg_10.png differ diff --git a/convnext_small_4xb256_IF_1m_lr4e-4_aug_1e-1/20230606_050018/vis_data/vis_image/000005174.jpg_2.png b/convnext_small_4xb256_IF_1m_lr4e-4_aug_1e-1/20230606_050018/vis_data/vis_image/000005174.jpg_2.png new file mode 100644 index 0000000000000000000000000000000000000000..ca6ee2d3d33b744722f75b9c6a1a26eddbe95906 Binary files /dev/null and b/convnext_small_4xb256_IF_1m_lr4e-4_aug_1e-1/20230606_050018/vis_data/vis_image/000005174.jpg_2.png differ diff --git a/convnext_small_4xb256_IF_1m_lr4e-4_aug_1e-1/20230606_050018/vis_data/vis_image/000005174.jpg_3.png b/convnext_small_4xb256_IF_1m_lr4e-4_aug_1e-1/20230606_050018/vis_data/vis_image/000005174.jpg_3.png new file mode 100644 index 0000000000000000000000000000000000000000..95dc8a573dc0d3888fa04550ed8a607e06923c50 Binary files /dev/null and b/convnext_small_4xb256_IF_1m_lr4e-4_aug_1e-1/20230606_050018/vis_data/vis_image/000005174.jpg_3.png differ diff --git a/convnext_small_4xb256_IF_1m_lr4e-4_aug_1e-1/20230606_050018/vis_data/vis_image/000005174.jpg_4.png b/convnext_small_4xb256_IF_1m_lr4e-4_aug_1e-1/20230606_050018/vis_data/vis_image/000005174.jpg_4.png new file mode 100644 index 0000000000000000000000000000000000000000..17ccd700b0799e644d6fea4b7243c9bb2f47b47f Binary files /dev/null and b/convnext_small_4xb256_IF_1m_lr4e-4_aug_1e-1/20230606_050018/vis_data/vis_image/000005174.jpg_4.png differ diff --git a/convnext_small_4xb256_IF_1m_lr4e-4_aug_1e-1/20230606_050018/vis_data/vis_image/000005174.jpg_5.png b/convnext_small_4xb256_IF_1m_lr4e-4_aug_1e-1/20230606_050018/vis_data/vis_image/000005174.jpg_5.png new file mode 100644 index 0000000000000000000000000000000000000000..16159960eb6d84a45122afc46ccf7b34fc19790b Binary files /dev/null and b/convnext_small_4xb256_IF_1m_lr4e-4_aug_1e-1/20230606_050018/vis_data/vis_image/000005174.jpg_5.png differ diff --git a/convnext_small_4xb256_IF_1m_lr4e-4_aug_1e-1/20230606_050018/vis_data/vis_image/000005174.jpg_6.png b/convnext_small_4xb256_IF_1m_lr4e-4_aug_1e-1/20230606_050018/vis_data/vis_image/000005174.jpg_6.png new file mode 100644 index 0000000000000000000000000000000000000000..400f34e8468e50c0ea16f7eab95ae843e7d8d9d9 Binary files /dev/null and b/convnext_small_4xb256_IF_1m_lr4e-4_aug_1e-1/20230606_050018/vis_data/vis_image/000005174.jpg_6.png differ diff --git a/convnext_small_4xb256_IF_1m_lr4e-4_aug_1e-1/20230606_050018/vis_data/vis_image/000005174.jpg_7.png b/convnext_small_4xb256_IF_1m_lr4e-4_aug_1e-1/20230606_050018/vis_data/vis_image/000005174.jpg_7.png new file mode 100644 index 0000000000000000000000000000000000000000..d83373cf2d183f06d4098ac9d3625595ad29cc60 Binary files /dev/null and b/convnext_small_4xb256_IF_1m_lr4e-4_aug_1e-1/20230606_050018/vis_data/vis_image/000005174.jpg_7.png differ diff --git a/convnext_small_4xb256_IF_1m_lr4e-4_aug_1e-1/20230606_050018/vis_data/vis_image/000005174.jpg_8.png b/convnext_small_4xb256_IF_1m_lr4e-4_aug_1e-1/20230606_050018/vis_data/vis_image/000005174.jpg_8.png new file mode 100644 index 0000000000000000000000000000000000000000..b6dfca6df02353380b69b40d9a41d23ac543d4a9 Binary files /dev/null and b/convnext_small_4xb256_IF_1m_lr4e-4_aug_1e-1/20230606_050018/vis_data/vis_image/000005174.jpg_8.png differ diff --git a/convnext_small_4xb256_IF_1m_lr4e-4_aug_1e-1/20230606_050018/vis_data/vis_image/000005174.jpg_9.png b/convnext_small_4xb256_IF_1m_lr4e-4_aug_1e-1/20230606_050018/vis_data/vis_image/000005174.jpg_9.png new file mode 100644 index 0000000000000000000000000000000000000000..1abc15a756c6a9c9a10d0115605b5812364f63a2 Binary files /dev/null and b/convnext_small_4xb256_IF_1m_lr4e-4_aug_1e-1/20230606_050018/vis_data/vis_image/000005174.jpg_9.png differ diff --git a/convnext_small_4xb256_IF_1m_lr4e-4_aug_1e-1/convnext_small_4xb256_IF_1m_lr4e-4_aug_1e-1.py b/convnext_small_4xb256_IF_1m_lr4e-4_aug_1e-1/convnext_small_4xb256_IF_1m_lr4e-4_aug_1e-1.py new file mode 100644 index 0000000000000000000000000000000000000000..6cbb68ea0841a58c5da7a681ef0548c7d7ab66f7 --- /dev/null +++ b/convnext_small_4xb256_IF_1m_lr4e-4_aug_1e-1/convnext_small_4xb256_IF_1m_lr4e-4_aug_1e-1.py @@ -0,0 +1,264 @@ +optim_wrapper = dict( + optimizer=dict( + type='AdamW', + lr=0.0004, + weight_decay=0.05, + eps=1e-08, + betas=(0.9, 0.999), + _scope_='mmpretrain'), + paramwise_cfg=dict( + norm_decay_mult=0.0, + bias_decay_mult=0.0, + flat_decay_mult=0.0, + custom_keys=dict({ + '.absolute_pos_embed': dict(decay_mult=0.0), + '.relative_position_bias_table': dict(decay_mult=0.0) + })), + type='AmpOptimWrapper', + dtype='bfloat16', + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=False, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=4096) +model = dict( + type='ImageClassifier', + backbone=dict( + frozen_stages=2, + type='ConvNeXt', + arch='small', + drop_path_rate=0.4, + init_cfg=dict( + type='Pretrained', + checkpoint= + 'https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth', + prefix='backbone')), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=768, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + init_cfg=None), + init_cfg=dict( + type='TruncNormal', layer=['Conv2d', 'Linear'], std=0.02, bias=0.0), + train_cfg=None) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/IF80w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/if-dpmsolver++-50-20w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/cc1m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/if-dpmsolver++-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/if-dpmsolver++-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/convnext_small_4xb256_IF_1m_lr4e-4_aug_1e-1' diff --git a/convnext_small_4xb256_IF_1m_lr4e-4_aug_1e-1/epoch_1.pth b/convnext_small_4xb256_IF_1m_lr4e-4_aug_1e-1/epoch_1.pth new file mode 100644 index 0000000000000000000000000000000000000000..53ba6d270166c98e9f46f1d6d4e232643211e5fe --- /dev/null +++ b/convnext_small_4xb256_IF_1m_lr4e-4_aug_1e-1/epoch_1.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:e8b8991ba8332c6afcc6584efc3f2a078ac3ad252715dc166dd95672ee6beead +size 782112474 diff --git a/convnext_small_4xb256_IF_1m_lr4e-4_aug_1e-1/epoch_10.pth b/convnext_small_4xb256_IF_1m_lr4e-4_aug_1e-1/epoch_10.pth new file mode 100644 index 0000000000000000000000000000000000000000..4f2df0f994a6962fd4f862c4bd89bd24ec5a20c0 --- /dev/null +++ b/convnext_small_4xb256_IF_1m_lr4e-4_aug_1e-1/epoch_10.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:762dab12b87081b9a0d6b0218558d94544c3aee0277b6be831820769f71f4ada +size 783395738 diff --git a/convnext_small_4xb256_IF_1m_lr4e-4_aug_1e-1/epoch_2.pth b/convnext_small_4xb256_IF_1m_lr4e-4_aug_1e-1/epoch_2.pth new file mode 100644 index 0000000000000000000000000000000000000000..b7e0941b909443aa64eaad4af687dda39d0982d2 --- /dev/null +++ b/convnext_small_4xb256_IF_1m_lr4e-4_aug_1e-1/epoch_2.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:969cc781edcb8e03865009a641d57e8c7caf6619e28686a2c33e8bf159f673d1 +size 782256410 diff --git a/convnext_small_4xb256_IF_1m_lr4e-4_aug_1e-1/epoch_3.pth b/convnext_small_4xb256_IF_1m_lr4e-4_aug_1e-1/epoch_3.pth new file mode 100644 index 0000000000000000000000000000000000000000..c54986d56c1bfac01cfb341a179945ff79c296d3 --- /dev/null +++ b/convnext_small_4xb256_IF_1m_lr4e-4_aug_1e-1/epoch_3.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:1327d3e395d4abda6cc2a13d234c6c0c64d06bafdb8a2222e3f22ff62ff9b822 +size 782399002 diff --git a/convnext_small_4xb256_IF_1m_lr4e-4_aug_1e-1/epoch_4.pth b/convnext_small_4xb256_IF_1m_lr4e-4_aug_1e-1/epoch_4.pth new file mode 100644 index 0000000000000000000000000000000000000000..4faa8619e983a70e1e4a08c15bd39fb6b7f3b9af --- /dev/null +++ b/convnext_small_4xb256_IF_1m_lr4e-4_aug_1e-1/epoch_4.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:6c52b20bb6b63afebd4b27dd2eeee31a534a0eba7822ab8890f4264130b16f5f +size 782541530 diff --git a/convnext_small_4xb256_IF_1m_lr4e-4_aug_1e-1/epoch_5.pth b/convnext_small_4xb256_IF_1m_lr4e-4_aug_1e-1/epoch_5.pth new file mode 100644 index 0000000000000000000000000000000000000000..d9668c1063488e7b016aa292f57bd8e702564418 --- /dev/null +++ b/convnext_small_4xb256_IF_1m_lr4e-4_aug_1e-1/epoch_5.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:32d391a06b12a99e3daab9b879233335e3d9c79194716d129fcffe1fd87e39e3 +size 782683674 diff --git a/convnext_small_4xb256_IF_1m_lr4e-4_aug_1e-1/epoch_6.pth b/convnext_small_4xb256_IF_1m_lr4e-4_aug_1e-1/epoch_6.pth new file mode 100644 index 0000000000000000000000000000000000000000..215478da169105db1d38b9ce8e7c4a4c88698ab2 --- /dev/null +++ b/convnext_small_4xb256_IF_1m_lr4e-4_aug_1e-1/epoch_6.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:8c89141b35d38ddb7b1ccd17f94e0e1429cc685dbf63c9ed11fce51f6a9e2226 +size 782825818 diff --git a/convnext_small_4xb256_IF_1m_lr4e-4_aug_1e-1/epoch_7.pth b/convnext_small_4xb256_IF_1m_lr4e-4_aug_1e-1/epoch_7.pth new file mode 100644 index 0000000000000000000000000000000000000000..11e313c21a336dcf2976d1990c22ee504224c92d --- /dev/null +++ b/convnext_small_4xb256_IF_1m_lr4e-4_aug_1e-1/epoch_7.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:8c0999d5ad8795b0ffe923261d54e9cebb4a74a7ca42e515d0826c4bcee6d0d2 +size 782967834 diff --git a/convnext_small_4xb256_IF_1m_lr4e-4_aug_1e-1/epoch_8.pth b/convnext_small_4xb256_IF_1m_lr4e-4_aug_1e-1/epoch_8.pth new file mode 100644 index 0000000000000000000000000000000000000000..ff9d3ff3be45a9c876003c07222773bea81c8531 --- /dev/null +++ b/convnext_small_4xb256_IF_1m_lr4e-4_aug_1e-1/epoch_8.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ac23223a0fae08968e3d66a8ee4391a752ce91a5c7285b20530eba79ed64fb7b +size 783109402 diff --git a/convnext_small_4xb256_IF_1m_lr4e-4_aug_1e-1/epoch_9.pth b/convnext_small_4xb256_IF_1m_lr4e-4_aug_1e-1/epoch_9.pth new file mode 100644 index 0000000000000000000000000000000000000000..d4f704cdcb320c86bfb09fe61fdb59155299c473 --- /dev/null +++ b/convnext_small_4xb256_IF_1m_lr4e-4_aug_1e-1/epoch_9.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:0d038e6378aff3272bbd247215b961f43a2e7fdf24e80d88bca327ba6417b831 +size 783251930 diff --git a/convnext_small_4xb256_IF_1m_lr4e-4_aug_1e-1/last_checkpoint b/convnext_small_4xb256_IF_1m_lr4e-4_aug_1e-1/last_checkpoint new file mode 100644 index 0000000000000000000000000000000000000000..820b87c1b18d9eda2916964cd7b0410778b20cb6 --- /dev/null +++ b/convnext_small_4xb256_IF_1m_lr4e-4_aug_1e-1/last_checkpoint @@ -0,0 +1 @@ +/mnt/petrelfs/luzeyu/workspace/fakebench/mmpretrain/workdir/convnext_small_4xb256_IF_1m_lr4e-4_aug_1e-1/epoch_10.pth \ No newline at end of file diff --git a/convnext_small_4xb256_IF_1m_lr4e-4_aug_5e-1/20230606_050039/20230606_050039.log b/convnext_small_4xb256_IF_1m_lr4e-4_aug_5e-1/20230606_050039/20230606_050039.log new file mode 100644 index 0000000000000000000000000000000000000000..05a662f555756638c5f9adff0690397899e68f76 --- /dev/null +++ b/convnext_small_4xb256_IF_1m_lr4e-4_aug_5e-1/20230606_050039/20230606_050039.log @@ -0,0 +1,1618 @@ +2023/06/06 05:00:46 - mmengine - INFO - +------------------------------------------------------------ +System environment: + sys.platform: linux + Python: 3.10.9 (main, Mar 8 2023, 10:47:38) [GCC 11.2.0] + CUDA available: True + numpy_random_seed: 2142801404 + GPU 0,1,2,3: NVIDIA A100-SXM4-80GB + CUDA_HOME: /mnt/petrelfs/share/cuda-11.6 + NVCC: Cuda compilation tools, release 11.6, V11.6.124 + GCC: gcc (GCC) 7.5.0 + PyTorch: 1.13.1 + PyTorch compiling details: PyTorch built with: + - GCC 9.3 + - C++ Version: 201402 + - Intel(R) oneAPI Math Kernel Library Version 2021.4-Product Build 20210904 for Intel(R) 64 architecture applications + - Intel(R) MKL-DNN v2.6.0 (Git Hash 52b5f107dd9cf10910aaa19cb47f3abf9b349815) + - OpenMP 201511 (a.k.a. OpenMP 4.5) + - LAPACK is enabled (usually provided by MKL) + - NNPACK is enabled + - CPU capability usage: AVX2 + - CUDA Runtime 11.6 + - NVCC architecture flags: -gencode;arch=compute_37,code=sm_37;-gencode;arch=compute_50,code=sm_50;-gencode;arch=compute_60,code=sm_60;-gencode;arch=compute_61,code=sm_61;-gencode;arch=compute_70,code=sm_70;-gencode;arch=compute_75,code=sm_75;-gencode;arch=compute_80,code=sm_80;-gencode;arch=compute_86,code=sm_86;-gencode;arch=compute_37,code=compute_37 + - CuDNN 8.3.2 (built against CUDA 11.5) + - Magma 2.6.1 + - Build settings: BLAS_INFO=mkl, BUILD_TYPE=Release, CUDA_VERSION=11.6, CUDNN_VERSION=8.3.2, CXX_COMPILER=/opt/rh/devtoolset-9/root/usr/bin/c++, CXX_FLAGS= -fabi-version=11 -Wno-deprecated -fvisibility-inlines-hidden -DUSE_PTHREADPOOL -fopenmp -DNDEBUG -DUSE_KINETO -DUSE_FBGEMM -DUSE_QNNPACK -DUSE_PYTORCH_QNNPACK -DUSE_XNNPACK -DSYMBOLICATE_MOBILE_DEBUG_HANDLE -DEDGE_PROFILER_USE_KINETO -O2 -fPIC -Wno-narrowing -Wall -Wextra -Werror=return-type -Werror=non-virtual-dtor -Wno-missing-field-initializers -Wno-type-limits -Wno-array-bounds -Wno-unknown-pragmas -Wunused-local-typedefs -Wno-unused-parameter -Wno-unused-function -Wno-unused-result -Wno-strict-overflow -Wno-strict-aliasing -Wno-error=deprecated-declarations -Wno-stringop-overflow -Wno-psabi -Wno-error=pedantic -Wno-error=redundant-decls -Wno-error=old-style-cast -fdiagnostics-color=always -faligned-new -Wno-unused-but-set-variable -Wno-maybe-uninitialized -fno-math-errno -fno-trapping-math -Werror=format -Werror=cast-function-type -Wno-stringop-overflow, LAPACK_INFO=mkl, PERF_WITH_AVX=1, PERF_WITH_AVX2=1, PERF_WITH_AVX512=1, TORCH_VERSION=1.13.1, USE_CUDA=ON, USE_CUDNN=ON, USE_EXCEPTION_PTR=1, USE_GFLAGS=OFF, USE_GLOG=OFF, USE_MKL=ON, USE_MKLDNN=ON, USE_MPI=OFF, USE_NCCL=ON, USE_NNPACK=ON, USE_OPENMP=ON, USE_ROCM=OFF, + + TorchVision: 0.14.1 + OpenCV: 4.7.0 + MMEngine: 0.7.3 + +Runtime environment: + cudnn_benchmark: True + mp_cfg: {'mp_start_method': 'fork', 'opencv_num_threads': 0} + dist_cfg: {'backend': 'nccl'} + seed: None + deterministic: False + Distributed launcher: slurm + Distributed training: True + GPU number: 4 +------------------------------------------------------------ + +2023/06/06 05:00:50 - mmengine - INFO - Config: +optim_wrapper = dict( + optimizer=dict( + type='AdamW', + lr=0.0004, + weight_decay=0.05, + eps=1e-08, + betas=(0.9, 0.999), + _scope_='mmpretrain'), + paramwise_cfg=dict( + norm_decay_mult=0.0, + bias_decay_mult=0.0, + flat_decay_mult=0.0, + custom_keys=dict({ + '.absolute_pos_embed': dict(decay_mult=0.0), + '.relative_position_bias_table': dict(decay_mult=0.0) + })), + type='AmpOptimWrapper', + dtype='bfloat16', + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=False, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=4096) +model = dict( + type='ImageClassifier', + backbone=dict( + frozen_stages=2, + type='ConvNeXt', + arch='small', + drop_path_rate=0.4, + init_cfg=dict( + type='Pretrained', + checkpoint= + 'https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth', + prefix='backbone')), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=768, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + init_cfg=None), + init_cfg=dict( + type='TruncNormal', layer=['Conv2d', 'Linear'], std=0.02, bias=0.0), + train_cfg=None) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/IF80w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/if-dpmsolver++-50-20w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/cc1m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/if-dpmsolver++-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/if-dpmsolver++-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/convnext_small_4xb256_IF_1m_lr4e-4_aug_5e-1' + +2023/06/06 05:01:01 - mmengine - INFO - Hooks will be executed in the following order: +before_run: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_load_checkpoint: +(ABOVE_NORMAL) EMAHook + -------------------- +before_train: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_train_epoch: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(NORMAL ) DistSamplerSeedHook + -------------------- +before_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook + -------------------- +after_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +after_train_epoch: +(NORMAL ) IterTimerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_val_epoch: +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook + -------------------- +before_val_iter: +(NORMAL ) IterTimerHook + -------------------- +after_val_iter: +(NORMAL ) IterTimerHook +(NORMAL ) VisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_val_epoch: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_save_checkpoint: +(ABOVE_NORMAL) EMAHook + -------------------- +after_train: +(VERY_LOW ) CheckpointHook + -------------------- +before_test_epoch: +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook + -------------------- +before_test_iter: +(NORMAL ) IterTimerHook + -------------------- +after_test_iter: +(NORMAL ) IterTimerHook +(NORMAL ) VisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_test_epoch: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_run: +(BELOW_NORMAL) LoggerHook + -------------------- +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.2.0.weight:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.2.0.bias:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.2.1.bias:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.3.0.weight:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.3.0.bias:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.3.1.bias:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.0.gamma:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.0.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.0.norm.weight:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.0.norm.bias:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.0.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.0.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.1.gamma:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.1.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.1.norm.weight:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.1.norm.bias:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.1.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.1.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.2.gamma:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.2.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.2.norm.weight:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.2.norm.bias:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.2.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.2.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.3.gamma:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.3.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.3.norm.weight:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.3.norm.bias:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.3.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.3.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.4.gamma:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.4.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.4.norm.weight:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.4.norm.bias:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.4.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.4.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.5.gamma:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.5.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.5.norm.weight:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.5.norm.bias:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.5.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.5.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.6.gamma:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.6.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.6.norm.weight:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.6.norm.bias:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.6.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.6.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.7.gamma:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.7.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.7.norm.weight:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.7.norm.bias:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.7.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.7.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.8.gamma:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.8.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.8.norm.weight:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.8.norm.bias:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.8.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.8.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.9.gamma:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.9.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.9.norm.weight:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.9.norm.bias:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.9.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.9.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.10.gamma:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.10.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.10.norm.weight:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.10.norm.bias:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.10.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.10.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.11.gamma:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.11.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.11.norm.weight:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.11.norm.bias:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.11.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.11.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.12.gamma:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.12.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.12.norm.weight:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.12.norm.bias:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.12.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.12.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.13.gamma:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.13.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.13.norm.weight:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.13.norm.bias:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.13.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.13.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.14.gamma:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.14.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.14.norm.weight:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.14.norm.bias:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.14.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.14.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.15.gamma:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.15.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.15.norm.weight:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.15.norm.bias:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.15.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.15.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.16.gamma:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.16.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.16.norm.weight:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.16.norm.bias:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.16.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.16.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.17.gamma:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.17.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.17.norm.weight:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.17.norm.bias:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.17.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.17.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.18.gamma:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.18.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.18.norm.weight:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.18.norm.bias:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.18.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.18.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.19.gamma:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.19.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.19.norm.weight:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.19.norm.bias:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.19.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.19.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.20.gamma:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.20.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.20.norm.weight:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.20.norm.bias:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.20.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.20.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.21.gamma:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.21.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.21.norm.weight:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.21.norm.bias:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.21.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.21.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.22.gamma:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.22.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.22.norm.weight:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.22.norm.bias:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.22.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.22.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.23.gamma:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.23.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.23.norm.weight:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.23.norm.bias:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.23.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.23.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.24.gamma:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.24.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.24.norm.weight:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.24.norm.bias:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.24.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.24.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.25.gamma:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.25.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.25.norm.weight:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.25.norm.bias:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.25.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.25.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.26.gamma:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.26.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.26.norm.weight:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.26.norm.bias:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.26.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.26.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.3.0.gamma:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.3.0.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.3.0.norm.weight:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.3.0.norm.bias:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.3.0.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.3.0.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.3.1.gamma:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.3.1.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.3.1.norm.weight:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.3.1.norm.bias:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.3.1.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.3.1.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.3.2.gamma:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.3.2.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.3.2.norm.weight:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.3.2.norm.bias:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.3.2.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.stages.3.2.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.norm3.weight:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- backbone.norm3.bias:weight_decay=0.0 +2023/06/06 05:01:14 - mmengine - INFO - paramwise_options -- head.fc.bias:weight_decay=0.0 +2023/06/06 05:01:18 - mmengine - INFO - load backbone in model from: https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth +Name of parameter - Initialization information + +backbone.downsample_layers.0.0.weight - torch.Size([96, 3, 4, 4]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.0.0.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.0.1.weight - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.0.1.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.1.0.weight - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.1.0.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.1.1.weight - torch.Size([192, 96, 2, 2]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.1.1.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.2.0.weight - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.2.0.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.2.1.weight - torch.Size([384, 192, 2, 2]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.2.1.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.3.0.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.3.0.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.3.1.weight - torch.Size([768, 384, 2, 2]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.3.1.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.0.gamma - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.0.depthwise_conv.weight - torch.Size([96, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.0.depthwise_conv.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.0.norm.weight - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.0.norm.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.0.pointwise_conv1.weight - torch.Size([384, 96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.0.pointwise_conv1.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.0.pointwise_conv2.weight - torch.Size([96, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.0.pointwise_conv2.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.1.gamma - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.1.depthwise_conv.weight - torch.Size([96, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.1.depthwise_conv.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.1.norm.weight - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.1.norm.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.1.pointwise_conv1.weight - torch.Size([384, 96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.1.pointwise_conv1.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.1.pointwise_conv2.weight - torch.Size([96, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.1.pointwise_conv2.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.2.gamma - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.2.depthwise_conv.weight - torch.Size([96, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.2.depthwise_conv.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.2.norm.weight - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.2.norm.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.2.pointwise_conv1.weight - torch.Size([384, 96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.2.pointwise_conv1.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.2.pointwise_conv2.weight - torch.Size([96, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.2.pointwise_conv2.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.0.gamma - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.0.depthwise_conv.weight - torch.Size([192, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.0.depthwise_conv.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.0.norm.weight - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.0.norm.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.0.pointwise_conv1.weight - torch.Size([768, 192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.0.pointwise_conv1.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.0.pointwise_conv2.weight - torch.Size([192, 768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.0.pointwise_conv2.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.1.gamma - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.1.depthwise_conv.weight - torch.Size([192, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.1.depthwise_conv.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.1.norm.weight - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.1.norm.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.1.pointwise_conv1.weight - torch.Size([768, 192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.1.pointwise_conv1.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.1.pointwise_conv2.weight - torch.Size([192, 768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.1.pointwise_conv2.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.2.gamma - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.2.depthwise_conv.weight - torch.Size([192, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.2.depthwise_conv.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.2.norm.weight - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.2.norm.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.2.pointwise_conv1.weight - torch.Size([768, 192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.2.pointwise_conv1.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.2.pointwise_conv2.weight - torch.Size([192, 768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.2.pointwise_conv2.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.0.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.0.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.0.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.0.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.0.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.0.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.0.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.0.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.0.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.1.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.1.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.1.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.1.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.1.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.1.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.1.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.1.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.1.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.2.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.2.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.2.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.2.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.2.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.2.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.2.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.2.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.2.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.3.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.3.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.3.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.3.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.3.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.3.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.3.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.3.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.3.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.4.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.4.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.4.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.4.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.4.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.4.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.4.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.4.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.4.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.5.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.5.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.5.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.5.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.5.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.5.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.5.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.5.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.5.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.6.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.6.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.6.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.6.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.6.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.6.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.6.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.6.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.6.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.7.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.7.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.7.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.7.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.7.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.7.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.7.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.7.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.7.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.8.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.8.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.8.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.8.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.8.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.8.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.8.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.8.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.8.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.9.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.9.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.9.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.9.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.9.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.9.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.9.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.9.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.9.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.10.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.10.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.10.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.10.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.10.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.10.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.10.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.10.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.10.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.11.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.11.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.11.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.11.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.11.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.11.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.11.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.11.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.11.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.12.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.12.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.12.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.12.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.12.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.12.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.12.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.12.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.12.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.13.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.13.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.13.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.13.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.13.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.13.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.13.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.13.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.13.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.14.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.14.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.14.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.14.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.14.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.14.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.14.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.14.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.14.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.15.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.15.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.15.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.15.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.15.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.15.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.15.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.15.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.15.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.16.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.16.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.16.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.16.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.16.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.16.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.16.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.16.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.16.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.17.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.17.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.17.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.17.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.17.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.17.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.17.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.17.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.17.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.18.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.18.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.18.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.18.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.18.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.18.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.18.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.18.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.18.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.19.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.19.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.19.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.19.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.19.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.19.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.19.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.19.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.19.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.20.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.20.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.20.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.20.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.20.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.20.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.20.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.20.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.20.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.21.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.21.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.21.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.21.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.21.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.21.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.21.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.21.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.21.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.22.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.22.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.22.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.22.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.22.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.22.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.22.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.22.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.22.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.23.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.23.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.23.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.23.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.23.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.23.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.23.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.23.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.23.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.24.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.24.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.24.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.24.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.24.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.24.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.24.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.24.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.24.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.25.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.25.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.25.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.25.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.25.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.25.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.25.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.25.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.25.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.26.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.26.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.26.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.26.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.26.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.26.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.26.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.26.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.26.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.0.gamma - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.0.depthwise_conv.weight - torch.Size([768, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.0.depthwise_conv.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.0.norm.weight - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.0.norm.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.0.pointwise_conv1.weight - torch.Size([3072, 768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.0.pointwise_conv1.bias - torch.Size([3072]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.0.pointwise_conv2.weight - torch.Size([768, 3072]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.0.pointwise_conv2.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.1.gamma - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.1.depthwise_conv.weight - torch.Size([768, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.1.depthwise_conv.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.1.norm.weight - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.1.norm.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.1.pointwise_conv1.weight - torch.Size([3072, 768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.1.pointwise_conv1.bias - torch.Size([3072]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.1.pointwise_conv2.weight - torch.Size([768, 3072]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.1.pointwise_conv2.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.2.gamma - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.2.depthwise_conv.weight - torch.Size([768, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.2.depthwise_conv.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.2.norm.weight - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.2.norm.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.2.pointwise_conv1.weight - torch.Size([3072, 768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.2.pointwise_conv1.bias - torch.Size([3072]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.2.pointwise_conv2.weight - torch.Size([768, 3072]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.2.pointwise_conv2.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.norm3.weight - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.norm3.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +head.fc.weight - torch.Size([2, 768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +head.fc.bias - torch.Size([2]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 +2023/06/06 05:01:18 - mmengine - WARNING - "FileClient" will be deprecated in future. Please use io functions in https://mmengine.readthedocs.io/en/latest/api/fileio.html#file-io +2023/06/06 05:01:18 - mmengine - WARNING - "HardDiskBackend" is the alias of "LocalBackend" and the former will be deprecated in future. +2023/06/06 05:01:18 - mmengine - INFO - Checkpoints will be saved to /mnt/petrelfs/luzeyu/workspace/fakebench/mmpretrain/workdir/convnext_small_4xb256_IF_1m_lr4e-4_aug_5e-1. diff --git a/convnext_small_4xb256_IF_1m_lr4e-4_aug_5e-1/20230606_050039/vis_data/config.py b/convnext_small_4xb256_IF_1m_lr4e-4_aug_5e-1/20230606_050039/vis_data/config.py new file mode 100644 index 0000000000000000000000000000000000000000..387bb5b9606b514ed2e86ad5d7d57200ef9f0bd9 --- /dev/null +++ b/convnext_small_4xb256_IF_1m_lr4e-4_aug_5e-1/20230606_050039/vis_data/config.py @@ -0,0 +1,264 @@ +optim_wrapper = dict( + optimizer=dict( + type='AdamW', + lr=0.0004, + weight_decay=0.05, + eps=1e-08, + betas=(0.9, 0.999), + _scope_='mmpretrain'), + paramwise_cfg=dict( + norm_decay_mult=0.0, + bias_decay_mult=0.0, + flat_decay_mult=0.0, + custom_keys=dict({ + '.absolute_pos_embed': dict(decay_mult=0.0), + '.relative_position_bias_table': dict(decay_mult=0.0) + })), + type='AmpOptimWrapper', + dtype='bfloat16', + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=False, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=4096) +model = dict( + type='ImageClassifier', + backbone=dict( + frozen_stages=2, + type='ConvNeXt', + arch='small', + drop_path_rate=0.4, + init_cfg=dict( + type='Pretrained', + checkpoint= + 'https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth', + prefix='backbone')), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=768, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + init_cfg=None), + init_cfg=dict( + type='TruncNormal', layer=['Conv2d', 'Linear'], std=0.02, bias=0.0), + train_cfg=None) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/IF80w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/if-dpmsolver++-50-20w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/cc1m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/if-dpmsolver++-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/if-dpmsolver++-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/convnext_small_4xb256_IF_1m_lr4e-4_aug_5e-1' diff --git a/convnext_small_4xb256_IF_1m_lr4e-4_aug_5e-1/20230606_050039/vis_data/events.out.tfevents.1685998856.SH-IDC1-10-140-24-124.191205.0 b/convnext_small_4xb256_IF_1m_lr4e-4_aug_5e-1/20230606_050039/vis_data/events.out.tfevents.1685998856.SH-IDC1-10-140-24-124.191205.0 new file mode 100644 index 0000000000000000000000000000000000000000..bfa1feef81998b51e27a2936914eaa9be7768e2c --- /dev/null +++ b/convnext_small_4xb256_IF_1m_lr4e-4_aug_5e-1/20230606_050039/vis_data/events.out.tfevents.1685998856.SH-IDC1-10-140-24-124.191205.0 @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:9cd872435b01446d8935604204bdc412b50f5680c1e210053a4f9282f4ac7688 +size 10121 diff --git a/convnext_small_4xb256_IF_1m_lr4e-4_aug_5e-1/20230606_050439/20230606_050439.log b/convnext_small_4xb256_IF_1m_lr4e-4_aug_5e-1/20230606_050439/20230606_050439.log new file mode 100644 index 0000000000000000000000000000000000000000..cc4ba80e1d53d5be2b607351b5bae0fbb2e12703 --- /dev/null +++ b/convnext_small_4xb256_IF_1m_lr4e-4_aug_5e-1/20230606_050439/20230606_050439.log @@ -0,0 +1,1618 @@ +2023/06/06 05:04:45 - mmengine - INFO - +------------------------------------------------------------ +System environment: + sys.platform: linux + Python: 3.10.9 (main, Mar 8 2023, 10:47:38) [GCC 11.2.0] + CUDA available: True + numpy_random_seed: 867830763 + GPU 0,1,2,3: NVIDIA A100-SXM4-80GB + CUDA_HOME: /mnt/petrelfs/share/cuda-11.6 + NVCC: Cuda compilation tools, release 11.6, V11.6.124 + GCC: gcc (GCC) 7.5.0 + PyTorch: 1.13.1 + PyTorch compiling details: PyTorch built with: + - GCC 9.3 + - C++ Version: 201402 + - Intel(R) oneAPI Math Kernel Library Version 2021.4-Product Build 20210904 for Intel(R) 64 architecture applications + - Intel(R) MKL-DNN v2.6.0 (Git Hash 52b5f107dd9cf10910aaa19cb47f3abf9b349815) + - OpenMP 201511 (a.k.a. OpenMP 4.5) + - LAPACK is enabled (usually provided by MKL) + - NNPACK is enabled + - CPU capability usage: AVX2 + - CUDA Runtime 11.6 + - NVCC architecture flags: -gencode;arch=compute_37,code=sm_37;-gencode;arch=compute_50,code=sm_50;-gencode;arch=compute_60,code=sm_60;-gencode;arch=compute_61,code=sm_61;-gencode;arch=compute_70,code=sm_70;-gencode;arch=compute_75,code=sm_75;-gencode;arch=compute_80,code=sm_80;-gencode;arch=compute_86,code=sm_86;-gencode;arch=compute_37,code=compute_37 + - CuDNN 8.3.2 (built against CUDA 11.5) + - Magma 2.6.1 + - Build settings: BLAS_INFO=mkl, BUILD_TYPE=Release, CUDA_VERSION=11.6, CUDNN_VERSION=8.3.2, CXX_COMPILER=/opt/rh/devtoolset-9/root/usr/bin/c++, CXX_FLAGS= -fabi-version=11 -Wno-deprecated -fvisibility-inlines-hidden -DUSE_PTHREADPOOL -fopenmp -DNDEBUG -DUSE_KINETO -DUSE_FBGEMM -DUSE_QNNPACK -DUSE_PYTORCH_QNNPACK -DUSE_XNNPACK -DSYMBOLICATE_MOBILE_DEBUG_HANDLE -DEDGE_PROFILER_USE_KINETO -O2 -fPIC -Wno-narrowing -Wall -Wextra -Werror=return-type -Werror=non-virtual-dtor -Wno-missing-field-initializers -Wno-type-limits -Wno-array-bounds -Wno-unknown-pragmas -Wunused-local-typedefs -Wno-unused-parameter -Wno-unused-function -Wno-unused-result -Wno-strict-overflow -Wno-strict-aliasing -Wno-error=deprecated-declarations -Wno-stringop-overflow -Wno-psabi -Wno-error=pedantic -Wno-error=redundant-decls -Wno-error=old-style-cast -fdiagnostics-color=always -faligned-new -Wno-unused-but-set-variable -Wno-maybe-uninitialized -fno-math-errno -fno-trapping-math -Werror=format -Werror=cast-function-type -Wno-stringop-overflow, LAPACK_INFO=mkl, PERF_WITH_AVX=1, PERF_WITH_AVX2=1, PERF_WITH_AVX512=1, TORCH_VERSION=1.13.1, USE_CUDA=ON, USE_CUDNN=ON, USE_EXCEPTION_PTR=1, USE_GFLAGS=OFF, USE_GLOG=OFF, USE_MKL=ON, USE_MKLDNN=ON, USE_MPI=OFF, USE_NCCL=ON, USE_NNPACK=ON, USE_OPENMP=ON, USE_ROCM=OFF, + + TorchVision: 0.14.1 + OpenCV: 4.7.0 + MMEngine: 0.7.3 + +Runtime environment: + cudnn_benchmark: True + mp_cfg: {'mp_start_method': 'fork', 'opencv_num_threads': 0} + dist_cfg: {'backend': 'nccl'} + seed: None + deterministic: False + Distributed launcher: slurm + Distributed training: True + GPU number: 4 +------------------------------------------------------------ + +2023/06/06 05:04:49 - mmengine - INFO - Config: +optim_wrapper = dict( + optimizer=dict( + type='AdamW', + lr=0.0004, + weight_decay=0.05, + eps=1e-08, + betas=(0.9, 0.999), + _scope_='mmpretrain'), + paramwise_cfg=dict( + norm_decay_mult=0.0, + bias_decay_mult=0.0, + flat_decay_mult=0.0, + custom_keys=dict({ + '.absolute_pos_embed': dict(decay_mult=0.0), + '.relative_position_bias_table': dict(decay_mult=0.0) + })), + type='AmpOptimWrapper', + dtype='bfloat16', + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=False, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=4096) +model = dict( + type='ImageClassifier', + backbone=dict( + frozen_stages=2, + type='ConvNeXt', + arch='small', + drop_path_rate=0.4, + init_cfg=dict( + type='Pretrained', + checkpoint= + 'https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth', + prefix='backbone')), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=768, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + init_cfg=None), + init_cfg=dict( + type='TruncNormal', layer=['Conv2d', 'Linear'], std=0.02, bias=0.0), + train_cfg=None) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/IF80w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/if-dpmsolver++-50-20w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/cc1m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/if-dpmsolver++-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/if-dpmsolver++-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/convnext_small_4xb256_IF_1m_lr4e-4_aug_5e-1' + +2023/06/06 05:04:59 - mmengine - INFO - Hooks will be executed in the following order: +before_run: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_load_checkpoint: +(ABOVE_NORMAL) EMAHook + -------------------- +before_train: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_train_epoch: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(NORMAL ) DistSamplerSeedHook + -------------------- +before_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook + -------------------- +after_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +after_train_epoch: +(NORMAL ) IterTimerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_val_epoch: +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook + -------------------- +before_val_iter: +(NORMAL ) IterTimerHook + -------------------- +after_val_iter: +(NORMAL ) IterTimerHook +(NORMAL ) VisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_val_epoch: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_save_checkpoint: +(ABOVE_NORMAL) EMAHook + -------------------- +after_train: +(VERY_LOW ) CheckpointHook + -------------------- +before_test_epoch: +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook + -------------------- +before_test_iter: +(NORMAL ) IterTimerHook + -------------------- +after_test_iter: +(NORMAL ) IterTimerHook +(NORMAL ) VisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_test_epoch: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_run: +(BELOW_NORMAL) LoggerHook + -------------------- +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.2.0.weight:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.2.0.bias:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.2.1.bias:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.3.0.weight:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.3.0.bias:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.3.1.bias:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.2.0.gamma:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.2.0.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.2.0.norm.weight:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.2.0.norm.bias:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.2.0.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.2.0.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.2.1.gamma:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.2.1.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.2.1.norm.weight:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.2.1.norm.bias:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.2.1.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.2.1.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.2.2.gamma:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.2.2.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.2.2.norm.weight:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.2.2.norm.bias:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.2.2.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.2.2.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.2.3.gamma:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.2.3.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.2.3.norm.weight:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.2.3.norm.bias:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.2.3.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.2.3.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.2.4.gamma:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.2.4.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.2.4.norm.weight:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.2.4.norm.bias:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.2.4.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.2.4.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.2.5.gamma:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.2.5.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.2.5.norm.weight:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.2.5.norm.bias:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.2.5.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.2.5.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.2.6.gamma:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.2.6.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.2.6.norm.weight:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.2.6.norm.bias:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.2.6.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.2.6.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.2.7.gamma:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.2.7.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.2.7.norm.weight:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.2.7.norm.bias:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.2.7.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.2.7.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.2.8.gamma:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.2.8.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.2.8.norm.weight:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.2.8.norm.bias:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.2.8.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.2.8.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.2.9.gamma:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.2.9.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.2.9.norm.weight:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.2.9.norm.bias:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.2.9.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.2.9.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.2.10.gamma:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.2.10.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.2.10.norm.weight:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.2.10.norm.bias:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.2.10.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.2.10.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.2.11.gamma:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.2.11.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.2.11.norm.weight:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.2.11.norm.bias:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.2.11.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.2.11.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.2.12.gamma:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.2.12.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.2.12.norm.weight:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.2.12.norm.bias:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.2.12.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.2.12.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.2.13.gamma:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.2.13.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.2.13.norm.weight:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.2.13.norm.bias:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.2.13.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.2.13.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.2.14.gamma:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.2.14.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.2.14.norm.weight:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.2.14.norm.bias:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.2.14.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.2.14.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.2.15.gamma:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.2.15.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.2.15.norm.weight:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.2.15.norm.bias:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.2.15.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.2.15.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.2.16.gamma:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.2.16.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.2.16.norm.weight:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.2.16.norm.bias:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.2.16.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.2.16.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.2.17.gamma:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.2.17.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.2.17.norm.weight:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.2.17.norm.bias:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.2.17.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.2.17.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.2.18.gamma:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.2.18.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.2.18.norm.weight:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.2.18.norm.bias:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.2.18.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.2.18.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.2.19.gamma:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.2.19.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.2.19.norm.weight:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.2.19.norm.bias:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.2.19.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.2.19.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.2.20.gamma:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.2.20.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.2.20.norm.weight:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.2.20.norm.bias:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.2.20.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.2.20.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.2.21.gamma:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.2.21.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.2.21.norm.weight:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.2.21.norm.bias:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.2.21.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.2.21.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.2.22.gamma:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.2.22.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.2.22.norm.weight:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.2.22.norm.bias:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.2.22.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.2.22.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.2.23.gamma:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.2.23.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.2.23.norm.weight:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.2.23.norm.bias:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.2.23.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.2.23.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.2.24.gamma:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.2.24.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.2.24.norm.weight:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.2.24.norm.bias:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.2.24.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.2.24.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.2.25.gamma:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.2.25.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.2.25.norm.weight:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.2.25.norm.bias:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.2.25.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.2.25.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.2.26.gamma:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.2.26.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.2.26.norm.weight:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.2.26.norm.bias:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.2.26.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.2.26.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.3.0.gamma:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.3.0.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.3.0.norm.weight:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.3.0.norm.bias:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.3.0.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.3.0.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.3.1.gamma:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.3.1.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.3.1.norm.weight:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.3.1.norm.bias:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.3.1.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.3.1.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.3.2.gamma:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.3.2.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.3.2.norm.weight:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.3.2.norm.bias:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.3.2.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.stages.3.2.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.norm3.weight:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- backbone.norm3.bias:weight_decay=0.0 +2023/06/06 05:05:11 - mmengine - INFO - paramwise_options -- head.fc.bias:weight_decay=0.0 +2023/06/06 05:05:15 - mmengine - INFO - load backbone in model from: https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth +Name of parameter - Initialization information + +backbone.downsample_layers.0.0.weight - torch.Size([96, 3, 4, 4]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.0.0.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.0.1.weight - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.0.1.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.1.0.weight - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.1.0.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.1.1.weight - torch.Size([192, 96, 2, 2]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.1.1.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.2.0.weight - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.2.0.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.2.1.weight - torch.Size([384, 192, 2, 2]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.2.1.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.3.0.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.3.0.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.3.1.weight - torch.Size([768, 384, 2, 2]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.3.1.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.0.gamma - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.0.depthwise_conv.weight - torch.Size([96, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.0.depthwise_conv.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.0.norm.weight - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.0.norm.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.0.pointwise_conv1.weight - torch.Size([384, 96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.0.pointwise_conv1.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.0.pointwise_conv2.weight - torch.Size([96, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.0.pointwise_conv2.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.1.gamma - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.1.depthwise_conv.weight - torch.Size([96, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.1.depthwise_conv.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.1.norm.weight - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.1.norm.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.1.pointwise_conv1.weight - torch.Size([384, 96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.1.pointwise_conv1.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.1.pointwise_conv2.weight - torch.Size([96, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.1.pointwise_conv2.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.2.gamma - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.2.depthwise_conv.weight - torch.Size([96, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.2.depthwise_conv.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.2.norm.weight - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.2.norm.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.2.pointwise_conv1.weight - torch.Size([384, 96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.2.pointwise_conv1.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.2.pointwise_conv2.weight - torch.Size([96, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.2.pointwise_conv2.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.0.gamma - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.0.depthwise_conv.weight - torch.Size([192, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.0.depthwise_conv.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.0.norm.weight - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.0.norm.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.0.pointwise_conv1.weight - torch.Size([768, 192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.0.pointwise_conv1.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.0.pointwise_conv2.weight - torch.Size([192, 768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.0.pointwise_conv2.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.1.gamma - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.1.depthwise_conv.weight - torch.Size([192, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.1.depthwise_conv.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.1.norm.weight - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.1.norm.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.1.pointwise_conv1.weight - torch.Size([768, 192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.1.pointwise_conv1.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.1.pointwise_conv2.weight - torch.Size([192, 768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.1.pointwise_conv2.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.2.gamma - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.2.depthwise_conv.weight - torch.Size([192, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.2.depthwise_conv.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.2.norm.weight - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.2.norm.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.2.pointwise_conv1.weight - torch.Size([768, 192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.2.pointwise_conv1.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.2.pointwise_conv2.weight - torch.Size([192, 768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.2.pointwise_conv2.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.0.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.0.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.0.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.0.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.0.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.0.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.0.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.0.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.0.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.1.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.1.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.1.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.1.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.1.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.1.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.1.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.1.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.1.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.2.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.2.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.2.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.2.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.2.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.2.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.2.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.2.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.2.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.3.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.3.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.3.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.3.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.3.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.3.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.3.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.3.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.3.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.4.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.4.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.4.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.4.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.4.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.4.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.4.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.4.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.4.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.5.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.5.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.5.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.5.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.5.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.5.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.5.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.5.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.5.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.6.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.6.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.6.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.6.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.6.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.6.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.6.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.6.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.6.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.7.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.7.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.7.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.7.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.7.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.7.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.7.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.7.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.7.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.8.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.8.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.8.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.8.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.8.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.8.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.8.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.8.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.8.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.9.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.9.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.9.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.9.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.9.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.9.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.9.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.9.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.9.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.10.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.10.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.10.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.10.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.10.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.10.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.10.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.10.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.10.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.11.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.11.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.11.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.11.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.11.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.11.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.11.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.11.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.11.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.12.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.12.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.12.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.12.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.12.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.12.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.12.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.12.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.12.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.13.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.13.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.13.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.13.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.13.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.13.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.13.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.13.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.13.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.14.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.14.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.14.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.14.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.14.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.14.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.14.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.14.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.14.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.15.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.15.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.15.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.15.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.15.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.15.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.15.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.15.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.15.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.16.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.16.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.16.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.16.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.16.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.16.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.16.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.16.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.16.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.17.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.17.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.17.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.17.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.17.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.17.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.17.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.17.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.17.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.18.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.18.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.18.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.18.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.18.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.18.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.18.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.18.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.18.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.19.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.19.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.19.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.19.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.19.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.19.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.19.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.19.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.19.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.20.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.20.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.20.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.20.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.20.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.20.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.20.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.20.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.20.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.21.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.21.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.21.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.21.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.21.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.21.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.21.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.21.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.21.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.22.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.22.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.22.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.22.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.22.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.22.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.22.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.22.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.22.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.23.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.23.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.23.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.23.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.23.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.23.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.23.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.23.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.23.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.24.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.24.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.24.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.24.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.24.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.24.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.24.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.24.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.24.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.25.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.25.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.25.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.25.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.25.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.25.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.25.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.25.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.25.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.26.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.26.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.26.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.26.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.26.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.26.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.26.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.26.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.26.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.0.gamma - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.0.depthwise_conv.weight - torch.Size([768, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.0.depthwise_conv.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.0.norm.weight - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.0.norm.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.0.pointwise_conv1.weight - torch.Size([3072, 768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.0.pointwise_conv1.bias - torch.Size([3072]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.0.pointwise_conv2.weight - torch.Size([768, 3072]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.0.pointwise_conv2.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.1.gamma - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.1.depthwise_conv.weight - torch.Size([768, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.1.depthwise_conv.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.1.norm.weight - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.1.norm.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.1.pointwise_conv1.weight - torch.Size([3072, 768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.1.pointwise_conv1.bias - torch.Size([3072]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.1.pointwise_conv2.weight - torch.Size([768, 3072]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.1.pointwise_conv2.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.2.gamma - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.2.depthwise_conv.weight - torch.Size([768, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.2.depthwise_conv.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.2.norm.weight - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.2.norm.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.2.pointwise_conv1.weight - torch.Size([3072, 768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.2.pointwise_conv1.bias - torch.Size([3072]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.2.pointwise_conv2.weight - torch.Size([768, 3072]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.2.pointwise_conv2.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.norm3.weight - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.norm3.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +head.fc.weight - torch.Size([2, 768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +head.fc.bias - torch.Size([2]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 +2023/06/06 05:05:16 - mmengine - WARNING - "FileClient" will be deprecated in future. Please use io functions in https://mmengine.readthedocs.io/en/latest/api/fileio.html#file-io +2023/06/06 05:05:16 - mmengine - WARNING - "HardDiskBackend" is the alias of "LocalBackend" and the former will be deprecated in future. +2023/06/06 05:05:16 - mmengine - INFO - Checkpoints will be saved to /mnt/petrelfs/luzeyu/workspace/fakebench/mmpretrain/workdir/convnext_small_4xb256_IF_1m_lr4e-4_aug_5e-1. diff --git a/convnext_small_4xb256_IF_1m_lr4e-4_aug_5e-1/20230606_050439/vis_data/config.py b/convnext_small_4xb256_IF_1m_lr4e-4_aug_5e-1/20230606_050439/vis_data/config.py new file mode 100644 index 0000000000000000000000000000000000000000..387bb5b9606b514ed2e86ad5d7d57200ef9f0bd9 --- /dev/null +++ b/convnext_small_4xb256_IF_1m_lr4e-4_aug_5e-1/20230606_050439/vis_data/config.py @@ -0,0 +1,264 @@ +optim_wrapper = dict( + optimizer=dict( + type='AdamW', + lr=0.0004, + weight_decay=0.05, + eps=1e-08, + betas=(0.9, 0.999), + _scope_='mmpretrain'), + paramwise_cfg=dict( + norm_decay_mult=0.0, + bias_decay_mult=0.0, + flat_decay_mult=0.0, + custom_keys=dict({ + '.absolute_pos_embed': dict(decay_mult=0.0), + '.relative_position_bias_table': dict(decay_mult=0.0) + })), + type='AmpOptimWrapper', + dtype='bfloat16', + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=False, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=4096) +model = dict( + type='ImageClassifier', + backbone=dict( + frozen_stages=2, + type='ConvNeXt', + arch='small', + drop_path_rate=0.4, + init_cfg=dict( + type='Pretrained', + checkpoint= + 'https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth', + prefix='backbone')), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=768, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + init_cfg=None), + init_cfg=dict( + type='TruncNormal', layer=['Conv2d', 'Linear'], std=0.02, bias=0.0), + train_cfg=None) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/IF80w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/if-dpmsolver++-50-20w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/cc1m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/if-dpmsolver++-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/if-dpmsolver++-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/convnext_small_4xb256_IF_1m_lr4e-4_aug_5e-1' diff --git a/convnext_small_4xb256_IF_1m_lr4e-4_aug_5e-1/20230606_050439/vis_data/events.out.tfevents.1685999094.SH-IDC1-10-140-24-124.204947.0 b/convnext_small_4xb256_IF_1m_lr4e-4_aug_5e-1/20230606_050439/vis_data/events.out.tfevents.1685999094.SH-IDC1-10-140-24-124.204947.0 new file mode 100644 index 0000000000000000000000000000000000000000..e9e9630a3d9444150667edd69444eceb75278590 --- /dev/null +++ b/convnext_small_4xb256_IF_1m_lr4e-4_aug_5e-1/20230606_050439/vis_data/events.out.tfevents.1685999094.SH-IDC1-10-140-24-124.204947.0 @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:0b9138f43e3e542b0875b612241ad89f76a4de01d9a0401761093bb201ad2c91 +size 10121 diff --git a/convnext_small_4xb256_IF_1m_lr4e-4_aug_5e-1/20230606_051640/20230606_051640.log b/convnext_small_4xb256_IF_1m_lr4e-4_aug_5e-1/20230606_051640/20230606_051640.log new file mode 100644 index 0000000000000000000000000000000000000000..1c0cc3c6247cecebb80483c11106036bcd1aec77 --- /dev/null +++ b/convnext_small_4xb256_IF_1m_lr4e-4_aug_5e-1/20230606_051640/20230606_051640.log @@ -0,0 +1,1857 @@ +2023/06/06 05:16:45 - mmengine - INFO - +------------------------------------------------------------ +System environment: + sys.platform: linux + Python: 3.10.9 (main, Mar 8 2023, 10:47:38) [GCC 11.2.0] + CUDA available: True + numpy_random_seed: 129913838 + GPU 0,1,2,3: NVIDIA A100-SXM4-80GB + CUDA_HOME: /mnt/petrelfs/share/cuda-11.6 + NVCC: Cuda compilation tools, release 11.6, V11.6.124 + GCC: gcc (GCC) 7.5.0 + PyTorch: 1.13.1 + PyTorch compiling details: PyTorch built with: + - GCC 9.3 + - C++ Version: 201402 + - Intel(R) oneAPI Math Kernel Library Version 2021.4-Product Build 20210904 for Intel(R) 64 architecture applications + - Intel(R) MKL-DNN v2.6.0 (Git Hash 52b5f107dd9cf10910aaa19cb47f3abf9b349815) + - OpenMP 201511 (a.k.a. OpenMP 4.5) + - LAPACK is enabled (usually provided by MKL) + - NNPACK is enabled + - CPU capability usage: AVX2 + - CUDA Runtime 11.6 + - NVCC architecture flags: -gencode;arch=compute_37,code=sm_37;-gencode;arch=compute_50,code=sm_50;-gencode;arch=compute_60,code=sm_60;-gencode;arch=compute_61,code=sm_61;-gencode;arch=compute_70,code=sm_70;-gencode;arch=compute_75,code=sm_75;-gencode;arch=compute_80,code=sm_80;-gencode;arch=compute_86,code=sm_86;-gencode;arch=compute_37,code=compute_37 + - CuDNN 8.3.2 (built against CUDA 11.5) + - Magma 2.6.1 + - Build settings: BLAS_INFO=mkl, BUILD_TYPE=Release, CUDA_VERSION=11.6, CUDNN_VERSION=8.3.2, CXX_COMPILER=/opt/rh/devtoolset-9/root/usr/bin/c++, CXX_FLAGS= -fabi-version=11 -Wno-deprecated -fvisibility-inlines-hidden -DUSE_PTHREADPOOL -fopenmp -DNDEBUG -DUSE_KINETO -DUSE_FBGEMM -DUSE_QNNPACK -DUSE_PYTORCH_QNNPACK -DUSE_XNNPACK -DSYMBOLICATE_MOBILE_DEBUG_HANDLE -DEDGE_PROFILER_USE_KINETO -O2 -fPIC -Wno-narrowing -Wall -Wextra -Werror=return-type -Werror=non-virtual-dtor -Wno-missing-field-initializers -Wno-type-limits -Wno-array-bounds -Wno-unknown-pragmas -Wunused-local-typedefs -Wno-unused-parameter -Wno-unused-function -Wno-unused-result -Wno-strict-overflow -Wno-strict-aliasing -Wno-error=deprecated-declarations -Wno-stringop-overflow -Wno-psabi -Wno-error=pedantic -Wno-error=redundant-decls -Wno-error=old-style-cast -fdiagnostics-color=always -faligned-new -Wno-unused-but-set-variable -Wno-maybe-uninitialized -fno-math-errno -fno-trapping-math -Werror=format -Werror=cast-function-type -Wno-stringop-overflow, LAPACK_INFO=mkl, PERF_WITH_AVX=1, PERF_WITH_AVX2=1, PERF_WITH_AVX512=1, TORCH_VERSION=1.13.1, USE_CUDA=ON, USE_CUDNN=ON, USE_EXCEPTION_PTR=1, USE_GFLAGS=OFF, USE_GLOG=OFF, USE_MKL=ON, USE_MKLDNN=ON, USE_MPI=OFF, USE_NCCL=ON, USE_NNPACK=ON, USE_OPENMP=ON, USE_ROCM=OFF, + + TorchVision: 0.14.1 + OpenCV: 4.7.0 + MMEngine: 0.7.3 + +Runtime environment: + cudnn_benchmark: True + mp_cfg: {'mp_start_method': 'fork', 'opencv_num_threads': 0} + dist_cfg: {'backend': 'nccl'} + seed: None + deterministic: False + Distributed launcher: slurm + Distributed training: True + GPU number: 4 +------------------------------------------------------------ + +2023/06/06 05:16:48 - mmengine - INFO - Config: +optim_wrapper = dict( + optimizer=dict( + type='AdamW', + lr=0.0004, + weight_decay=0.05, + eps=1e-08, + betas=(0.9, 0.999), + _scope_='mmpretrain'), + paramwise_cfg=dict( + norm_decay_mult=0.0, + bias_decay_mult=0.0, + flat_decay_mult=0.0, + custom_keys=dict({ + '.absolute_pos_embed': dict(decay_mult=0.0), + '.relative_position_bias_table': dict(decay_mult=0.0) + })), + type='AmpOptimWrapper', + dtype='bfloat16', + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=False, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=4096) +model = dict( + type='ImageClassifier', + backbone=dict( + frozen_stages=2, + type='ConvNeXt', + arch='small', + drop_path_rate=0.4, + init_cfg=dict( + type='Pretrained', + checkpoint= + 'https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth', + prefix='backbone')), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=768, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + init_cfg=None), + init_cfg=dict( + type='TruncNormal', layer=['Conv2d', 'Linear'], std=0.02, bias=0.0), + train_cfg=None) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/IF80w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/if-dpmsolver++-50-20w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/cc1m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/if-dpmsolver++-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/if-dpmsolver++-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/convnext_small_4xb256_IF_1m_lr4e-4_aug_5e-1' + +2023/06/06 05:16:59 - mmengine - INFO - Hooks will be executed in the following order: +before_run: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_load_checkpoint: +(ABOVE_NORMAL) EMAHook + -------------------- +before_train: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_train_epoch: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(NORMAL ) DistSamplerSeedHook + -------------------- +before_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook + -------------------- +after_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +after_train_epoch: +(NORMAL ) IterTimerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_val_epoch: +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook + -------------------- +before_val_iter: +(NORMAL ) IterTimerHook + -------------------- +after_val_iter: +(NORMAL ) IterTimerHook +(NORMAL ) VisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_val_epoch: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_save_checkpoint: +(ABOVE_NORMAL) EMAHook + -------------------- +after_train: +(VERY_LOW ) CheckpointHook + -------------------- +before_test_epoch: +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook + -------------------- +before_test_iter: +(NORMAL ) IterTimerHook + -------------------- +after_test_iter: +(NORMAL ) IterTimerHook +(NORMAL ) VisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_test_epoch: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_run: +(BELOW_NORMAL) LoggerHook + -------------------- +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.2.0.weight:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.2.0.bias:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.2.1.bias:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.3.0.weight:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.3.0.bias:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.3.1.bias:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.2.0.gamma:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.2.0.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.2.0.norm.weight:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.2.0.norm.bias:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.2.0.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.2.0.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.2.1.gamma:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.2.1.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.2.1.norm.weight:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.2.1.norm.bias:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.2.1.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.2.1.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.2.2.gamma:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.2.2.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.2.2.norm.weight:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.2.2.norm.bias:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.2.2.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.2.2.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.2.3.gamma:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.2.3.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.2.3.norm.weight:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.2.3.norm.bias:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.2.3.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.2.3.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.2.4.gamma:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.2.4.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.2.4.norm.weight:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.2.4.norm.bias:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.2.4.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.2.4.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.2.5.gamma:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.2.5.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.2.5.norm.weight:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.2.5.norm.bias:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.2.5.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.2.5.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.2.6.gamma:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.2.6.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.2.6.norm.weight:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.2.6.norm.bias:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.2.6.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.2.6.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.2.7.gamma:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.2.7.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.2.7.norm.weight:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.2.7.norm.bias:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.2.7.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.2.7.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.2.8.gamma:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.2.8.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.2.8.norm.weight:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.2.8.norm.bias:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.2.8.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.2.8.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.2.9.gamma:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.2.9.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.2.9.norm.weight:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.2.9.norm.bias:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.2.9.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.2.9.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.2.10.gamma:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.2.10.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.2.10.norm.weight:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.2.10.norm.bias:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.2.10.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.2.10.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.2.11.gamma:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.2.11.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.2.11.norm.weight:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.2.11.norm.bias:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.2.11.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.2.11.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.2.12.gamma:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.2.12.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.2.12.norm.weight:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.2.12.norm.bias:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.2.12.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.2.12.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.2.13.gamma:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.2.13.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.2.13.norm.weight:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.2.13.norm.bias:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.2.13.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.2.13.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.2.14.gamma:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.2.14.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.2.14.norm.weight:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.2.14.norm.bias:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.2.14.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.2.14.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.2.15.gamma:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.2.15.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.2.15.norm.weight:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.2.15.norm.bias:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.2.15.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.2.15.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.2.16.gamma:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.2.16.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.2.16.norm.weight:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.2.16.norm.bias:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.2.16.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.2.16.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.2.17.gamma:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.2.17.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.2.17.norm.weight:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.2.17.norm.bias:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.2.17.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.2.17.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.2.18.gamma:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.2.18.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.2.18.norm.weight:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.2.18.norm.bias:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.2.18.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.2.18.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.2.19.gamma:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.2.19.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.2.19.norm.weight:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.2.19.norm.bias:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.2.19.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.2.19.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.2.20.gamma:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.2.20.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.2.20.norm.weight:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.2.20.norm.bias:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.2.20.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.2.20.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.2.21.gamma:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.2.21.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.2.21.norm.weight:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.2.21.norm.bias:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.2.21.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.2.21.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.2.22.gamma:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.2.22.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.2.22.norm.weight:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.2.22.norm.bias:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.2.22.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.2.22.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.2.23.gamma:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.2.23.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.2.23.norm.weight:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.2.23.norm.bias:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.2.23.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.2.23.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.2.24.gamma:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.2.24.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.2.24.norm.weight:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.2.24.norm.bias:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.2.24.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.2.24.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.2.25.gamma:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.2.25.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.2.25.norm.weight:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.2.25.norm.bias:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.2.25.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.2.25.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.2.26.gamma:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.2.26.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.2.26.norm.weight:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.2.26.norm.bias:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.2.26.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.2.26.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.3.0.gamma:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.3.0.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.3.0.norm.weight:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.3.0.norm.bias:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.3.0.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.3.0.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.3.1.gamma:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.3.1.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.3.1.norm.weight:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.3.1.norm.bias:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.3.1.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.3.1.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.3.2.gamma:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.3.2.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.3.2.norm.weight:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.3.2.norm.bias:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.3.2.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.stages.3.2.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.norm3.weight:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- backbone.norm3.bias:weight_decay=0.0 +2023/06/06 05:17:12 - mmengine - INFO - paramwise_options -- head.fc.bias:weight_decay=0.0 +2023/06/06 05:17:17 - mmengine - INFO - load backbone in model from: https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth +Name of parameter - Initialization information + +backbone.downsample_layers.0.0.weight - torch.Size([96, 3, 4, 4]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.0.0.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.0.1.weight - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.0.1.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.1.0.weight - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.1.0.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.1.1.weight - torch.Size([192, 96, 2, 2]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.1.1.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.2.0.weight - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.2.0.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.2.1.weight - torch.Size([384, 192, 2, 2]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.2.1.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.3.0.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.3.0.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.3.1.weight - torch.Size([768, 384, 2, 2]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.3.1.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.0.gamma - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.0.depthwise_conv.weight - torch.Size([96, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.0.depthwise_conv.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.0.norm.weight - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.0.norm.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.0.pointwise_conv1.weight - torch.Size([384, 96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.0.pointwise_conv1.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.0.pointwise_conv2.weight - torch.Size([96, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.0.pointwise_conv2.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.1.gamma - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.1.depthwise_conv.weight - torch.Size([96, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.1.depthwise_conv.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.1.norm.weight - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.1.norm.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.1.pointwise_conv1.weight - torch.Size([384, 96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.1.pointwise_conv1.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.1.pointwise_conv2.weight - torch.Size([96, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.1.pointwise_conv2.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.2.gamma - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.2.depthwise_conv.weight - torch.Size([96, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.2.depthwise_conv.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.2.norm.weight - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.2.norm.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.2.pointwise_conv1.weight - torch.Size([384, 96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.2.pointwise_conv1.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.2.pointwise_conv2.weight - torch.Size([96, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.2.pointwise_conv2.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.0.gamma - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.0.depthwise_conv.weight - torch.Size([192, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.0.depthwise_conv.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.0.norm.weight - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.0.norm.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.0.pointwise_conv1.weight - torch.Size([768, 192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.0.pointwise_conv1.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.0.pointwise_conv2.weight - torch.Size([192, 768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.0.pointwise_conv2.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.1.gamma - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.1.depthwise_conv.weight - torch.Size([192, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.1.depthwise_conv.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.1.norm.weight - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.1.norm.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.1.pointwise_conv1.weight - torch.Size([768, 192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.1.pointwise_conv1.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.1.pointwise_conv2.weight - torch.Size([192, 768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.1.pointwise_conv2.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.2.gamma - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.2.depthwise_conv.weight - torch.Size([192, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.2.depthwise_conv.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.2.norm.weight - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.2.norm.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.2.pointwise_conv1.weight - torch.Size([768, 192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.2.pointwise_conv1.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.2.pointwise_conv2.weight - torch.Size([192, 768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.2.pointwise_conv2.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.0.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.0.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.0.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.0.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.0.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.0.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.0.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.0.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.0.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.1.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.1.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.1.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.1.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.1.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.1.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.1.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.1.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.1.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.2.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.2.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.2.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.2.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.2.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.2.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.2.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.2.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.2.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.3.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.3.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.3.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.3.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.3.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.3.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.3.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.3.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.3.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.4.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.4.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.4.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.4.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.4.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.4.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.4.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.4.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.4.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.5.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.5.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.5.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.5.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.5.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.5.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.5.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.5.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.5.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.6.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.6.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.6.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.6.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.6.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.6.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.6.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.6.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.6.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.7.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.7.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.7.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.7.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.7.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.7.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.7.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.7.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.7.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.8.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.8.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.8.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.8.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.8.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.8.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.8.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.8.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.8.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.9.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.9.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.9.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.9.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.9.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.9.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.9.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.9.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.9.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.10.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.10.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.10.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.10.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.10.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.10.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.10.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.10.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.10.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.11.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.11.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.11.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.11.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.11.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.11.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.11.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.11.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.11.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.12.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.12.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.12.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.12.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.12.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.12.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.12.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.12.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.12.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.13.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.13.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.13.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.13.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.13.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.13.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.13.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.13.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.13.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.14.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.14.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.14.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.14.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.14.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.14.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.14.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.14.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.14.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.15.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.15.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.15.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.15.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.15.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.15.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.15.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.15.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.15.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.16.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.16.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.16.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.16.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.16.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.16.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.16.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.16.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.16.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.17.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.17.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.17.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.17.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.17.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.17.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.17.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.17.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.17.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.18.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.18.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.18.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.18.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.18.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.18.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.18.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.18.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.18.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.19.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.19.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.19.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.19.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.19.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.19.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.19.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.19.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.19.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.20.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.20.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.20.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.20.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.20.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.20.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.20.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.20.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.20.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.21.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.21.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.21.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.21.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.21.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.21.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.21.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.21.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.21.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.22.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.22.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.22.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.22.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.22.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.22.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.22.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.22.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.22.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.23.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.23.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.23.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.23.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.23.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.23.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.23.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.23.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.23.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.24.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.24.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.24.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.24.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.24.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.24.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.24.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.24.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.24.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.25.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.25.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.25.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.25.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.25.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.25.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.25.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.25.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.25.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.26.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.26.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.26.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.26.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.26.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.26.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.26.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.26.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.26.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.0.gamma - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.0.depthwise_conv.weight - torch.Size([768, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.0.depthwise_conv.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.0.norm.weight - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.0.norm.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.0.pointwise_conv1.weight - torch.Size([3072, 768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.0.pointwise_conv1.bias - torch.Size([3072]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.0.pointwise_conv2.weight - torch.Size([768, 3072]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.0.pointwise_conv2.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.1.gamma - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.1.depthwise_conv.weight - torch.Size([768, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.1.depthwise_conv.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.1.norm.weight - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.1.norm.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.1.pointwise_conv1.weight - torch.Size([3072, 768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.1.pointwise_conv1.bias - torch.Size([3072]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.1.pointwise_conv2.weight - torch.Size([768, 3072]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.1.pointwise_conv2.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.2.gamma - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.2.depthwise_conv.weight - torch.Size([768, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.2.depthwise_conv.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.2.norm.weight - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.2.norm.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.2.pointwise_conv1.weight - torch.Size([3072, 768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.2.pointwise_conv1.bias - torch.Size([3072]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.2.pointwise_conv2.weight - torch.Size([768, 3072]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.2.pointwise_conv2.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.norm3.weight - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.norm3.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +head.fc.weight - torch.Size([2, 768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +head.fc.bias - torch.Size([2]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 +2023/06/06 05:17:17 - mmengine - WARNING - "FileClient" will be deprecated in future. Please use io functions in https://mmengine.readthedocs.io/en/latest/api/fileio.html#file-io +2023/06/06 05:17:17 - mmengine - WARNING - "HardDiskBackend" is the alias of "LocalBackend" and the former will be deprecated in future. +2023/06/06 05:17:17 - mmengine - INFO - Checkpoints will be saved to /mnt/petrelfs/luzeyu/workspace/fakebench/mmpretrain/workdir/convnext_small_4xb256_IF_1m_lr4e-4_aug_5e-1. +2023/06/06 05:18:29 - mmengine - INFO - Epoch(train) [1][ 100/1969] lr: 3.9998e-04 eta: 3:57:07 time: 0.6124 data_time: 0.0017 memory: 15587 loss: 0.0946 +2023/06/06 05:19:44 - mmengine - INFO - Epoch(train) [1][ 200/1969] lr: 3.9990e-04 eta: 3:58:12 time: 0.6957 data_time: 0.0297 memory: 15587 loss: 0.0585 +2023/06/06 05:20:56 - mmengine - INFO - Epoch(train) [1][ 300/1969] lr: 3.9978e-04 eta: 3:56:20 time: 0.6685 data_time: 0.0017 memory: 15587 loss: 0.0583 +2023/06/06 05:22:01 - mmengine - INFO - Epoch(train) [1][ 400/1969] lr: 3.9960e-04 eta: 3:48:19 time: 0.6553 data_time: 0.0017 memory: 15587 loss: 0.0593 +2023/06/06 05:23:07 - mmengine - INFO - Epoch(train) [1][ 500/1969] lr: 3.9938e-04 eta: 3:44:08 time: 0.6385 data_time: 0.0017 memory: 15587 loss: 0.0539 +2023/06/06 05:24:14 - mmengine - INFO - Epoch(train) [1][ 600/1969] lr: 3.9911e-04 eta: 3:41:17 time: 0.6029 data_time: 0.0016 memory: 15587 loss: 0.0574 +2023/06/06 05:25:28 - mmengine - INFO - Epoch(train) [1][ 700/1969] lr: 3.9879e-04 eta: 3:42:15 time: 1.1324 data_time: 0.0016 memory: 15587 loss: 0.0563 +2023/06/06 05:26:30 - mmengine - INFO - Epoch(train) [1][ 800/1969] lr: 3.9842e-04 eta: 3:37:32 time: 0.6036 data_time: 0.0018 memory: 15587 loss: 0.0601 +2023/06/06 05:27:33 - mmengine - INFO - Epoch(train) [1][ 900/1969] lr: 3.9800e-04 eta: 3:34:21 time: 0.5760 data_time: 0.0015 memory: 15587 loss: 0.0546 +2023/06/06 05:28:38 - mmengine - INFO - Exp name: convnext_small_4xb256_IF_1m_lr4e-4_aug_5e-1_20230606_051640 +2023/06/06 05:28:38 - mmengine - INFO - Epoch(train) [1][1000/1969] lr: 3.9753e-04 eta: 3:32:02 time: 0.5958 data_time: 0.0016 memory: 15587 loss: 0.0415 +2023/06/06 05:29:39 - mmengine - INFO - Epoch(train) [1][1100/1969] lr: 3.9701e-04 eta: 3:29:04 time: 0.6263 data_time: 0.0017 memory: 15587 loss: 0.0481 +2023/06/06 05:30:44 - mmengine - INFO - Epoch(train) [1][1200/1969] lr: 3.9644e-04 eta: 3:27:10 time: 0.6445 data_time: 0.0016 memory: 15587 loss: 0.0421 +2023/06/06 05:31:46 - mmengine - INFO - Epoch(train) [1][1300/1969] lr: 3.9583e-04 eta: 3:24:55 time: 0.6142 data_time: 0.0015 memory: 15587 loss: 0.0477 +2023/06/06 05:32:49 - mmengine - INFO - Epoch(train) [1][1400/1969] lr: 3.9516e-04 eta: 3:23:01 time: 0.5965 data_time: 0.0016 memory: 15587 loss: 0.0333 +2023/06/06 05:33:51 - mmengine - INFO - Epoch(train) [1][1500/1969] lr: 3.9445e-04 eta: 3:20:51 time: 0.6076 data_time: 0.0016 memory: 15587 loss: 0.0398 +2023/06/06 05:36:25 - mmengine - INFO - Epoch(train) [1][1600/1969] lr: 3.9369e-04 eta: 3:36:21 time: 0.6154 data_time: 0.0015 memory: 15587 loss: 0.0466 +2023/06/06 05:37:28 - mmengine - INFO - Epoch(train) [1][1700/1969] lr: 3.9288e-04 eta: 3:33:41 time: 0.6293 data_time: 0.0016 memory: 15587 loss: 0.0350 +2023/06/06 05:38:38 - mmengine - INFO - Epoch(train) [1][1800/1969] lr: 3.9202e-04 eta: 3:32:12 time: 1.3417 data_time: 0.0015 memory: 15587 loss: 0.0342 +2023/06/06 05:39:44 - mmengine - INFO - Epoch(train) [1][1900/1969] lr: 3.9112e-04 eta: 3:30:12 time: 0.6439 data_time: 0.0015 memory: 15587 loss: 0.0412 +2023/06/06 05:40:29 - mmengine - INFO - Exp name: convnext_small_4xb256_IF_1m_lr4e-4_aug_5e-1_20230606_051640 +2023/06/06 05:40:29 - mmengine - INFO - Saving checkpoint at 1 epochs +2023/06/06 05:41:05 - mmengine - INFO - Epoch(val) [1][29/29] accuracy/top1: 80.1260 single-label/precision_classwise: [73.56348419189453, 99.58609008789062] single-label/recall_classwise: [99.81061553955078, 55.9534912109375] single-label/f1-score_classwise: [84.70027160644531, 71.64979553222656] data_time: 0.2369 time: 0.7003 +2023/06/06 05:41:31 - mmengine - INFO - Exp name: convnext_small_4xb256_IF_1m_lr4e-4_aug_5e-1_20230606_051640 +2023/06/06 05:42:15 - mmengine - INFO - Epoch(train) [2][ 100/1969] lr: 3.8948e-04 eta: 3:27:30 time: 0.6675 data_time: 0.0017 memory: 15587 loss: 0.0321 +2023/06/06 05:43:17 - mmengine - INFO - Epoch(train) [2][ 200/1969] lr: 3.8845e-04 eta: 3:25:07 time: 0.6055 data_time: 0.0018 memory: 15585 loss: 0.0265 +2023/06/06 05:44:19 - mmengine - INFO - Epoch(train) [2][ 300/1969] lr: 3.8737e-04 eta: 3:22:54 time: 0.6007 data_time: 0.0016 memory: 15585 loss: 0.0378 +2023/06/06 05:45:19 - mmengine - INFO - Epoch(train) [2][ 400/1969] lr: 3.8625e-04 eta: 3:20:37 time: 0.6381 data_time: 0.0015 memory: 15585 loss: 0.0368 +2023/06/06 05:46:22 - mmengine - INFO - Epoch(train) [2][ 500/1969] lr: 3.8508e-04 eta: 3:18:37 time: 0.5798 data_time: 0.0015 memory: 15585 loss: 0.0336 +2023/06/06 05:47:24 - mmengine - INFO - Epoch(train) [2][ 600/1969] lr: 3.8386e-04 eta: 3:16:39 time: 0.6176 data_time: 0.0020 memory: 15585 loss: 0.0396 +2023/06/06 05:48:27 - mmengine - INFO - Epoch(train) [2][ 700/1969] lr: 3.8260e-04 eta: 3:14:55 time: 0.6705 data_time: 0.0015 memory: 15585 loss: 0.0287 +2023/06/06 05:49:30 - mmengine - INFO - Epoch(train) [2][ 800/1969] lr: 3.8129e-04 eta: 3:13:10 time: 0.6355 data_time: 0.0017 memory: 15585 loss: 0.0337 +2023/06/06 05:50:33 - mmengine - INFO - Epoch(train) [2][ 900/1969] lr: 3.7994e-04 eta: 3:11:30 time: 0.6274 data_time: 0.0016 memory: 15585 loss: 0.0346 +2023/06/06 05:51:36 - mmengine - INFO - Epoch(train) [2][1000/1969] lr: 3.7854e-04 eta: 3:09:55 time: 0.6428 data_time: 0.0016 memory: 15585 loss: 0.0430 +2023/06/06 05:51:56 - mmengine - INFO - Exp name: convnext_small_4xb256_IF_1m_lr4e-4_aug_5e-1_20230606_051640 +2023/06/06 05:52:40 - mmengine - INFO - Epoch(train) [2][1100/1969] lr: 3.7710e-04 eta: 3:08:22 time: 0.6416 data_time: 0.0014 memory: 15585 loss: 0.0325 +2023/06/06 05:53:43 - mmengine - INFO - Epoch(train) [2][1200/1969] lr: 3.7562e-04 eta: 3:06:49 time: 0.6325 data_time: 0.0017 memory: 15585 loss: 0.0321 +2023/06/06 05:54:56 - mmengine - INFO - Epoch(train) [2][1300/1969] lr: 3.7409e-04 eta: 3:06:06 time: 0.6006 data_time: 0.0019 memory: 15585 loss: 0.0295 +2023/06/06 05:55:56 - mmengine - INFO - Epoch(train) [2][1400/1969] lr: 3.7252e-04 eta: 3:04:17 time: 0.6230 data_time: 0.0016 memory: 15585 loss: 0.0266 +2023/06/06 05:56:51 - mmengine - INFO - Epoch(train) [2][1500/1969] lr: 3.7090e-04 eta: 3:02:10 time: 0.5775 data_time: 0.0016 memory: 15585 loss: 0.0236 +2023/06/06 05:57:49 - mmengine - INFO - Epoch(train) [2][1600/1969] lr: 3.6925e-04 eta: 3:00:20 time: 0.5890 data_time: 0.0015 memory: 15585 loss: 0.0292 +2023/06/06 05:58:47 - mmengine - INFO - Epoch(train) [2][1700/1969] lr: 3.6755e-04 eta: 2:58:36 time: 0.5980 data_time: 0.0015 memory: 15585 loss: 0.0280 +2023/06/06 05:59:44 - mmengine - INFO - Epoch(train) [2][1800/1969] lr: 3.6581e-04 eta: 2:56:47 time: 0.6007 data_time: 0.0015 memory: 15585 loss: 0.0328 +2023/06/06 06:00:41 - mmengine - INFO - Epoch(train) [2][1900/1969] lr: 3.6403e-04 eta: 2:55:01 time: 0.5462 data_time: 0.0015 memory: 15585 loss: 0.0343 +2023/06/06 06:01:20 - mmengine - INFO - Exp name: convnext_small_4xb256_IF_1m_lr4e-4_aug_5e-1_20230606_051640 +2023/06/06 06:01:20 - mmengine - INFO - Saving checkpoint at 2 epochs +2023/06/06 06:01:54 - mmengine - INFO - Epoch(val) [2][29/29] accuracy/top1: 91.7887 single-label/precision_classwise: [87.12751007080078, 99.78273010253906] single-label/recall_classwise: [99.85481262207031, 81.88372039794922] single-label/f1-score_classwise: [93.0580062866211, 89.95146179199219] data_time: 0.2112 time: 0.5998 +2023/06/06 06:02:36 - mmengine - INFO - Exp name: convnext_small_4xb256_IF_1m_lr4e-4_aug_5e-1_20230606_051640 +2023/06/06 06:02:57 - mmengine - INFO - Epoch(train) [3][ 100/1969] lr: 3.6093e-04 eta: 2:52:28 time: 0.5542 data_time: 0.0372 memory: 15585 loss: 0.0245 +2023/06/06 06:03:56 - mmengine - INFO - Epoch(train) [3][ 200/1969] lr: 3.5904e-04 eta: 2:50:54 time: 0.6017 data_time: 0.0835 memory: 15585 loss: 0.0372 +2023/06/06 06:04:54 - mmengine - INFO - Epoch(train) [3][ 300/1969] lr: 3.5711e-04 eta: 2:49:20 time: 0.6125 data_time: 0.0586 memory: 15585 loss: 0.0278 +2023/06/06 06:05:53 - mmengine - INFO - Epoch(train) [3][ 400/1969] lr: 3.5515e-04 eta: 2:47:50 time: 0.6046 data_time: 0.0022 memory: 15585 loss: 0.0296 +2023/06/06 06:06:51 - mmengine - INFO - Epoch(train) [3][ 500/1969] lr: 3.5314e-04 eta: 2:46:20 time: 0.5971 data_time: 0.0016 memory: 15585 loss: 0.0318 +2023/06/06 06:07:52 - mmengine - INFO - Epoch(train) [3][ 600/1969] lr: 3.5110e-04 eta: 2:45:00 time: 0.6521 data_time: 0.0017 memory: 15585 loss: 0.0282 +2023/06/06 06:08:53 - mmengine - INFO - Epoch(train) [3][ 700/1969] lr: 3.4902e-04 eta: 2:43:39 time: 0.5890 data_time: 0.0019 memory: 15585 loss: 0.0231 +2023/06/06 06:09:51 - mmengine - INFO - Epoch(train) [3][ 800/1969] lr: 3.4691e-04 eta: 2:42:11 time: 0.5729 data_time: 0.0018 memory: 15585 loss: 0.0262 +2023/06/06 06:10:49 - mmengine - INFO - Epoch(train) [3][ 900/1969] lr: 3.4475e-04 eta: 2:40:45 time: 0.6035 data_time: 0.0016 memory: 15585 loss: 0.0280 +2023/06/06 06:11:48 - mmengine - INFO - Epoch(train) [3][1000/1969] lr: 3.4257e-04 eta: 2:39:23 time: 0.6211 data_time: 0.0019 memory: 15585 loss: 0.0212 +2023/06/06 06:12:25 - mmengine - INFO - Exp name: convnext_small_4xb256_IF_1m_lr4e-4_aug_5e-1_20230606_051640 +2023/06/06 06:12:47 - mmengine - INFO - Epoch(train) [3][1100/1969] lr: 3.4034e-04 eta: 2:37:59 time: 0.5433 data_time: 0.0017 memory: 15585 loss: 0.0261 +2023/06/06 06:13:46 - mmengine - INFO - Epoch(train) [3][1200/1969] lr: 3.3809e-04 eta: 2:36:38 time: 0.5565 data_time: 0.0015 memory: 15585 loss: 0.0309 +2023/06/06 06:14:43 - mmengine - INFO - Epoch(train) [3][1300/1969] lr: 3.3580e-04 eta: 2:35:14 time: 0.6117 data_time: 0.0015 memory: 15585 loss: 0.0419 +2023/06/06 06:15:41 - mmengine - INFO - Epoch(train) [3][1400/1969] lr: 3.3347e-04 eta: 2:33:51 time: 0.5814 data_time: 0.0018 memory: 15585 loss: 0.0235 +2023/06/06 06:16:38 - mmengine - INFO - Epoch(train) [3][1500/1969] lr: 3.3112e-04 eta: 2:32:28 time: 0.5733 data_time: 0.0016 memory: 15585 loss: 0.0306 +2023/06/06 06:17:37 - mmengine - INFO - Epoch(train) [3][1600/1969] lr: 3.2873e-04 eta: 2:31:10 time: 0.6124 data_time: 0.0017 memory: 15585 loss: 0.0240 +2023/06/06 06:18:34 - mmengine - INFO - Epoch(train) [3][1700/1969] lr: 3.2631e-04 eta: 2:29:48 time: 0.5683 data_time: 0.0015 memory: 15585 loss: 0.0239 +2023/06/06 06:19:32 - mmengine - INFO - Epoch(train) [3][1800/1969] lr: 3.2385e-04 eta: 2:28:30 time: 0.6199 data_time: 0.0017 memory: 15585 loss: 0.0306 +2023/06/06 06:20:32 - mmengine - INFO - Epoch(train) [3][1900/1969] lr: 3.2137e-04 eta: 2:27:17 time: 0.6046 data_time: 0.0018 memory: 15585 loss: 0.0205 +2023/06/06 06:21:11 - mmengine - INFO - Exp name: convnext_small_4xb256_IF_1m_lr4e-4_aug_5e-1_20230606_051640 +2023/06/06 06:21:11 - mmengine - INFO - Saving checkpoint at 3 epochs +2023/06/06 06:21:42 - mmengine - INFO - Epoch(val) [3][29/29] accuracy/top1: 95.7308 single-label/precision_classwise: [92.92680358886719, 99.80374145507812] single-label/recall_classwise: [99.85481262207031, 90.66667175292969] single-label/f1-score_classwise: [96.26631927490234, 95.01605224609375] data_time: 0.1200 time: 0.5051 +2023/06/06 06:22:42 - mmengine - INFO - Exp name: convnext_small_4xb256_IF_1m_lr4e-4_aug_5e-1_20230606_051640 +2023/06/06 06:22:45 - mmengine - INFO - Epoch(train) [4][ 100/1969] lr: 3.1711e-04 eta: 2:25:17 time: 0.5588 data_time: 0.0021 memory: 15585 loss: 0.0231 +2023/06/06 06:23:44 - mmengine - INFO - Epoch(train) [4][ 200/1969] lr: 3.1455e-04 eta: 2:24:02 time: 0.6028 data_time: 0.0015 memory: 15585 loss: 0.0258 +2023/06/06 06:24:46 - mmengine - INFO - Epoch(train) [4][ 300/1969] lr: 3.1196e-04 eta: 2:22:54 time: 0.6135 data_time: 0.0016 memory: 15585 loss: 0.0368 +2023/06/06 06:25:45 - mmengine - INFO - Epoch(train) [4][ 400/1969] lr: 3.0935e-04 eta: 2:21:41 time: 0.5793 data_time: 0.0015 memory: 15585 loss: 0.0173 +2023/06/06 06:26:43 - mmengine - INFO - Epoch(train) [4][ 500/1969] lr: 3.0671e-04 eta: 2:20:27 time: 0.5791 data_time: 0.0017 memory: 15585 loss: 0.0241 +2023/06/06 06:27:43 - mmengine - INFO - Epoch(train) [4][ 600/1969] lr: 3.0404e-04 eta: 2:19:16 time: 0.6501 data_time: 0.0016 memory: 15585 loss: 0.0160 +2023/06/06 06:28:39 - mmengine - INFO - Epoch(train) [4][ 700/1969] lr: 3.0135e-04 eta: 2:17:59 time: 0.5738 data_time: 0.0017 memory: 15585 loss: 0.0246 +2023/06/06 06:29:37 - mmengine - INFO - Epoch(train) [4][ 800/1969] lr: 2.9863e-04 eta: 2:16:46 time: 0.6157 data_time: 0.0016 memory: 15585 loss: 0.0288 +2023/06/06 06:30:36 - mmengine - INFO - Epoch(train) [4][ 900/1969] lr: 2.9589e-04 eta: 2:15:33 time: 0.5952 data_time: 0.0016 memory: 15585 loss: 0.0193 +2023/06/06 06:31:34 - mmengine - INFO - Epoch(train) [4][1000/1969] lr: 2.9313e-04 eta: 2:14:22 time: 0.6250 data_time: 0.0017 memory: 15585 loss: 0.0233 +2023/06/06 06:32:27 - mmengine - INFO - Exp name: convnext_small_4xb256_IF_1m_lr4e-4_aug_5e-1_20230606_051640 +2023/06/06 06:32:31 - mmengine - INFO - Epoch(train) [4][1100/1969] lr: 2.9034e-04 eta: 2:13:06 time: 0.5614 data_time: 0.0020 memory: 15585 loss: 0.0292 +2023/06/06 06:33:30 - mmengine - INFO - Epoch(train) [4][1200/1969] lr: 2.8753e-04 eta: 2:11:58 time: 0.5563 data_time: 0.0015 memory: 15585 loss: 0.0245 +2023/06/06 06:34:28 - mmengine - INFO - Epoch(train) [4][1300/1969] lr: 2.8470e-04 eta: 2:10:45 time: 0.5465 data_time: 0.0017 memory: 15585 loss: 0.0220 +2023/06/06 06:35:25 - mmengine - INFO - Epoch(train) [4][1400/1969] lr: 2.8185e-04 eta: 2:09:33 time: 0.5955 data_time: 0.0016 memory: 15585 loss: 0.0204 +2023/06/06 06:36:34 - mmengine - INFO - Epoch(train) [4][1500/1969] lr: 2.7898e-04 eta: 2:08:40 time: 0.5859 data_time: 0.0017 memory: 15585 loss: 0.0220 +2023/06/06 06:37:32 - mmengine - INFO - Epoch(train) [4][1600/1969] lr: 2.7610e-04 eta: 2:07:29 time: 0.6029 data_time: 0.0015 memory: 15585 loss: 0.0233 +2023/06/06 06:38:28 - mmengine - INFO - Epoch(train) [4][1700/1969] lr: 2.7319e-04 eta: 2:06:16 time: 0.6054 data_time: 0.0018 memory: 15585 loss: 0.0286 +2023/06/06 06:39:26 - mmengine - INFO - Epoch(train) [4][1800/1969] lr: 2.7027e-04 eta: 2:05:06 time: 0.5769 data_time: 0.0016 memory: 15585 loss: 0.0270 +2023/06/06 06:40:25 - mmengine - INFO - Epoch(train) [4][1900/1969] lr: 2.6733e-04 eta: 2:03:58 time: 0.5727 data_time: 0.0016 memory: 15585 loss: 0.0288 +2023/06/06 06:41:04 - mmengine - INFO - Exp name: convnext_small_4xb256_IF_1m_lr4e-4_aug_5e-1_20230606_051640 +2023/06/06 06:41:04 - mmengine - INFO - Saving checkpoint at 4 epochs +2023/06/06 06:41:39 - mmengine - INFO - Epoch(val) [4][29/29] accuracy/top1: 97.6201 single-label/precision_classwise: [95.96094512939453, 99.85308837890625] single-label/recall_classwise: [99.88636779785156, 94.83720397949219] single-label/f1-score_classwise: [97.88431549072266, 97.2805404663086] data_time: 0.2257 time: 0.6116 +2023/06/06 06:42:44 - mmengine - INFO - Epoch(train) [5][ 100/1969] lr: 2.6232e-04 eta: 2:02:10 time: 0.5897 data_time: 0.0447 memory: 15585 loss: 0.0287 +2023/06/06 06:42:58 - mmengine - INFO - Exp name: convnext_small_4xb256_IF_1m_lr4e-4_aug_5e-1_20230606_051640 +2023/06/06 06:43:43 - mmengine - INFO - Epoch(train) [5][ 200/1969] lr: 2.5934e-04 eta: 2:01:02 time: 0.5518 data_time: 0.0297 memory: 15585 loss: 0.0290 +2023/06/06 06:44:46 - mmengine - INFO - Epoch(train) [5][ 300/1969] lr: 2.5635e-04 eta: 2:00:00 time: 0.6342 data_time: 0.0015 memory: 15585 loss: 0.0286 +2023/06/06 06:45:44 - mmengine - INFO - Epoch(train) [5][ 400/1969] lr: 2.5334e-04 eta: 1:58:51 time: 0.5839 data_time: 0.0015 memory: 15585 loss: 0.0195 +2023/06/06 06:46:43 - mmengine - INFO - Epoch(train) [5][ 500/1969] lr: 2.5032e-04 eta: 1:57:44 time: 0.5738 data_time: 0.0017 memory: 15585 loss: 0.0239 +2023/06/06 06:47:41 - mmengine - INFO - Epoch(train) [5][ 600/1969] lr: 2.4729e-04 eta: 1:56:36 time: 0.5874 data_time: 0.0015 memory: 15585 loss: 0.0223 +2023/06/06 06:48:40 - mmengine - INFO - Epoch(train) [5][ 700/1969] lr: 2.4424e-04 eta: 1:55:29 time: 0.6105 data_time: 0.0016 memory: 15585 loss: 0.0301 +2023/06/06 06:49:36 - mmengine - INFO - Epoch(train) [5][ 800/1969] lr: 2.4119e-04 eta: 1:54:19 time: 0.5406 data_time: 0.0015 memory: 15585 loss: 0.0180 +2023/06/06 06:50:33 - mmengine - INFO - Epoch(train) [5][ 900/1969] lr: 2.3813e-04 eta: 1:53:10 time: 0.5435 data_time: 0.0016 memory: 15585 loss: 0.0222 +2023/06/06 06:51:31 - mmengine - INFO - Epoch(train) [5][1000/1969] lr: 2.3506e-04 eta: 1:52:02 time: 0.5551 data_time: 0.0016 memory: 15585 loss: 0.0233 +2023/06/06 06:52:30 - mmengine - INFO - Epoch(train) [5][1100/1969] lr: 2.3198e-04 eta: 1:50:57 time: 0.5643 data_time: 0.0016 memory: 15585 loss: 0.0247 +2023/06/06 06:52:45 - mmengine - INFO - Exp name: convnext_small_4xb256_IF_1m_lr4e-4_aug_5e-1_20230606_051640 +2023/06/06 06:53:28 - mmengine - INFO - Epoch(train) [5][1200/1969] lr: 2.2890e-04 eta: 1:49:50 time: 0.6122 data_time: 0.0014 memory: 15585 loss: 0.0208 +2023/06/06 06:54:26 - mmengine - INFO - Epoch(train) [5][1300/1969] lr: 2.2581e-04 eta: 1:48:43 time: 0.5742 data_time: 0.0016 memory: 15585 loss: 0.0174 +2023/06/06 06:55:25 - mmengine - INFO - Epoch(train) [5][1400/1969] lr: 2.2271e-04 eta: 1:47:38 time: 0.5884 data_time: 0.0018 memory: 15585 loss: 0.0208 +2023/06/06 06:56:23 - mmengine - INFO - Epoch(train) [5][1500/1969] lr: 2.1961e-04 eta: 1:46:31 time: 0.5784 data_time: 0.0014 memory: 15585 loss: 0.0137 +2023/06/06 06:57:22 - mmengine - INFO - Epoch(train) [5][1600/1969] lr: 2.1651e-04 eta: 1:45:25 time: 0.5637 data_time: 0.0017 memory: 15585 loss: 0.0255 +2023/06/06 06:58:21 - mmengine - INFO - Epoch(train) [5][1700/1969] lr: 2.1340e-04 eta: 1:44:20 time: 0.5901 data_time: 0.0019 memory: 15585 loss: 0.0204 +2023/06/06 06:59:19 - mmengine - INFO - Epoch(train) [5][1800/1969] lr: 2.1029e-04 eta: 1:43:15 time: 0.5614 data_time: 0.0016 memory: 15585 loss: 0.0191 +2023/06/06 07:00:17 - mmengine - INFO - Epoch(train) [5][1900/1969] lr: 2.0718e-04 eta: 1:42:09 time: 0.5584 data_time: 0.0017 memory: 15585 loss: 0.0213 +2023/06/06 07:00:56 - mmengine - INFO - Exp name: convnext_small_4xb256_IF_1m_lr4e-4_aug_5e-1_20230606_051640 +2023/06/06 07:00:56 - mmengine - INFO - Saving checkpoint at 5 epochs +2023/06/06 07:01:30 - mmengine - INFO - Epoch(val) [5][29/29] accuracy/top1: 98.4308 single-label/precision_classwise: [97.40635681152344, 99.76017761230469] single-label/recall_classwise: [99.81061553955078, 96.73643493652344] single-label/f1-score_classwise: [98.59382629394531, 98.22504425048828] data_time: 0.2134 time: 0.6043 +2023/06/06 07:02:35 - mmengine - INFO - Epoch(train) [6][ 100/1969] lr: 2.0192e-04 eta: 1:40:24 time: 0.6086 data_time: 0.0017 memory: 15585 loss: 0.0149 +2023/06/06 07:03:08 - mmengine - INFO - Exp name: convnext_small_4xb256_IF_1m_lr4e-4_aug_5e-1_20230606_051640 +2023/06/06 07:03:35 - mmengine - INFO - Epoch(train) [6][ 200/1969] lr: 1.9881e-04 eta: 1:39:20 time: 0.5879 data_time: 0.0016 memory: 15585 loss: 0.0175 +2023/06/06 07:04:34 - mmengine - INFO - Epoch(train) [6][ 300/1969] lr: 1.9570e-04 eta: 1:38:16 time: 0.6431 data_time: 0.0016 memory: 15585 loss: 0.0187 +2023/06/06 07:05:32 - mmengine - INFO - Epoch(train) [6][ 400/1969] lr: 1.9259e-04 eta: 1:37:11 time: 0.5860 data_time: 0.0014 memory: 15585 loss: 0.0194 +2023/06/06 07:06:30 - mmengine - INFO - Epoch(train) [6][ 500/1969] lr: 1.8949e-04 eta: 1:36:06 time: 0.5449 data_time: 0.0015 memory: 15585 loss: 0.0274 +2023/06/06 07:07:26 - mmengine - INFO - Epoch(train) [6][ 600/1969] lr: 1.8639e-04 eta: 1:34:59 time: 0.5434 data_time: 0.0018 memory: 15585 loss: 0.0181 +2023/06/06 07:08:25 - mmengine - INFO - Epoch(train) [6][ 700/1969] lr: 1.8330e-04 eta: 1:33:54 time: 0.5646 data_time: 0.0015 memory: 15585 loss: 0.0125 +2023/06/06 07:09:22 - mmengine - INFO - Epoch(train) [6][ 800/1969] lr: 1.8021e-04 eta: 1:32:49 time: 0.5423 data_time: 0.0017 memory: 15585 loss: 0.0149 +2023/06/06 07:10:19 - mmengine - INFO - Epoch(train) [6][ 900/1969] lr: 1.7713e-04 eta: 1:31:44 time: 0.5615 data_time: 0.0016 memory: 15585 loss: 0.0249 +2023/06/06 07:11:17 - mmengine - INFO - Epoch(train) [6][1000/1969] lr: 1.7405e-04 eta: 1:30:39 time: 0.5385 data_time: 0.0018 memory: 15585 loss: 0.0139 +2023/06/06 07:12:14 - mmengine - INFO - Epoch(train) [6][1100/1969] lr: 1.7098e-04 eta: 1:29:34 time: 0.5790 data_time: 0.0017 memory: 15585 loss: 0.0157 +2023/06/06 07:12:46 - mmengine - INFO - Exp name: convnext_small_4xb256_IF_1m_lr4e-4_aug_5e-1_20230606_051640 +2023/06/06 07:13:12 - mmengine - INFO - Epoch(train) [6][1200/1969] lr: 1.6792e-04 eta: 1:28:30 time: 0.5652 data_time: 0.0017 memory: 15585 loss: 0.0195 +2023/06/06 07:14:10 - mmengine - INFO - Epoch(train) [6][1300/1969] lr: 1.6487e-04 eta: 1:27:26 time: 0.5819 data_time: 0.0014 memory: 15585 loss: 0.0226 +2023/06/06 07:15:08 - mmengine - INFO - Epoch(train) [6][1400/1969] lr: 1.6183e-04 eta: 1:26:22 time: 0.5798 data_time: 0.0022 memory: 15585 loss: 0.0189 +2023/06/06 07:16:06 - mmengine - INFO - Epoch(train) [6][1500/1969] lr: 1.5881e-04 eta: 1:25:18 time: 0.5688 data_time: 0.0016 memory: 15585 loss: 0.0159 +2023/06/06 07:17:05 - mmengine - INFO - Epoch(train) [6][1600/1969] lr: 1.5579e-04 eta: 1:24:15 time: 0.5816 data_time: 0.0016 memory: 15585 loss: 0.0205 +2023/06/06 07:18:05 - mmengine - INFO - Epoch(train) [6][1700/1969] lr: 1.5278e-04 eta: 1:23:13 time: 0.6482 data_time: 0.0019 memory: 15585 loss: 0.0168 +2023/06/06 07:19:03 - mmengine - INFO - Epoch(train) [6][1800/1969] lr: 1.4979e-04 eta: 1:22:09 time: 0.6013 data_time: 0.0017 memory: 15585 loss: 0.0273 +2023/06/06 07:20:03 - mmengine - INFO - Epoch(train) [6][1900/1969] lr: 1.4682e-04 eta: 1:21:07 time: 0.5793 data_time: 0.0015 memory: 15585 loss: 0.0200 +2023/06/06 07:20:43 - mmengine - INFO - Exp name: convnext_small_4xb256_IF_1m_lr4e-4_aug_5e-1_20230606_051640 +2023/06/06 07:20:43 - mmengine - INFO - Saving checkpoint at 6 epochs +2023/06/06 07:21:17 - mmengine - INFO - Epoch(val) [6][29/29] accuracy/top1: 98.8379 single-label/precision_classwise: [98.11356353759766, 99.76239776611328] single-label/recall_classwise: [99.81061553955078, 97.6434097290039] single-label/f1-score_classwise: [98.95481872558594, 98.6915283203125] data_time: 0.2203 time: 0.6101 +2023/06/06 07:22:22 - mmengine - INFO - Epoch(train) [7][ 100/1969] lr: 1.4182e-04 eta: 1:19:25 time: 0.5759 data_time: 0.0016 memory: 15585 loss: 0.0138 +2023/06/06 07:23:13 - mmengine - INFO - Exp name: convnext_small_4xb256_IF_1m_lr4e-4_aug_5e-1_20230606_051640 +2023/06/06 07:23:21 - mmengine - INFO - Epoch(train) [7][ 200/1969] lr: 1.3888e-04 eta: 1:18:22 time: 0.5869 data_time: 0.0017 memory: 15585 loss: 0.0167 +2023/06/06 07:24:20 - mmengine - INFO - Epoch(train) [7][ 300/1969] lr: 1.3597e-04 eta: 1:17:19 time: 0.5900 data_time: 0.0015 memory: 15585 loss: 0.0199 +2023/06/06 07:25:18 - mmengine - INFO - Epoch(train) [7][ 400/1969] lr: 1.3307e-04 eta: 1:16:16 time: 0.6056 data_time: 0.0017 memory: 15585 loss: 0.0137 +2023/06/06 07:26:19 - mmengine - INFO - Epoch(train) [7][ 500/1969] lr: 1.3018e-04 eta: 1:15:15 time: 0.5968 data_time: 0.0017 memory: 15585 loss: 0.0157 +2023/06/06 07:27:17 - mmengine - INFO - Epoch(train) [7][ 600/1969] lr: 1.2732e-04 eta: 1:14:12 time: 0.5849 data_time: 0.0015 memory: 15585 loss: 0.0221 +2023/06/06 07:28:17 - mmengine - INFO - Epoch(train) [7][ 700/1969] lr: 1.2448e-04 eta: 1:13:09 time: 0.6183 data_time: 0.0016 memory: 15585 loss: 0.0159 +2023/06/06 07:29:14 - mmengine - INFO - Epoch(train) [7][ 800/1969] lr: 1.2165e-04 eta: 1:12:06 time: 0.5871 data_time: 0.0019 memory: 15585 loss: 0.0143 +2023/06/06 07:30:13 - mmengine - INFO - Epoch(train) [7][ 900/1969] lr: 1.1885e-04 eta: 1:11:04 time: 0.5565 data_time: 0.0015 memory: 15585 loss: 0.0146 +2023/06/06 07:31:14 - mmengine - INFO - Epoch(train) [7][1000/1969] lr: 1.1607e-04 eta: 1:10:02 time: 0.7400 data_time: 0.0017 memory: 15585 loss: 0.0114 +2023/06/06 07:32:13 - mmengine - INFO - Epoch(train) [7][1100/1969] lr: 1.1331e-04 eta: 1:09:00 time: 0.5874 data_time: 0.0022 memory: 15585 loss: 0.0191 +2023/06/06 07:33:06 - mmengine - INFO - Exp name: convnext_small_4xb256_IF_1m_lr4e-4_aug_5e-1_20230606_051640 +2023/06/06 07:33:13 - mmengine - INFO - Epoch(train) [7][1200/1969] lr: 1.1058e-04 eta: 1:07:59 time: 0.5711 data_time: 0.0017 memory: 15585 loss: 0.0185 +2023/06/06 07:34:11 - mmengine - INFO - Epoch(train) [7][1300/1969] lr: 1.0787e-04 eta: 1:06:56 time: 0.6221 data_time: 0.0016 memory: 15585 loss: 0.0207 +2023/06/06 07:35:10 - mmengine - INFO - Epoch(train) [7][1400/1969] lr: 1.0518e-04 eta: 1:05:54 time: 0.5703 data_time: 0.0017 memory: 15585 loss: 0.0137 +2023/06/06 07:36:09 - mmengine - INFO - Epoch(train) [7][1500/1969] lr: 1.0252e-04 eta: 1:04:52 time: 0.5903 data_time: 0.0017 memory: 15585 loss: 0.0160 +2023/06/06 07:37:08 - mmengine - INFO - Epoch(train) [7][1600/1969] lr: 9.9889e-05 eta: 1:03:50 time: 0.5261 data_time: 0.0015 memory: 15585 loss: 0.0096 +2023/06/06 07:38:09 - mmengine - INFO - Epoch(train) [7][1700/1969] lr: 9.7282e-05 eta: 1:02:49 time: 0.5468 data_time: 0.0019 memory: 15585 loss: 0.0186 +2023/06/06 07:39:08 - mmengine - INFO - Epoch(train) [7][1800/1969] lr: 9.4703e-05 eta: 1:01:47 time: 0.5772 data_time: 0.0017 memory: 15585 loss: 0.0088 +2023/06/06 07:40:08 - mmengine - INFO - Epoch(train) [7][1900/1969] lr: 9.2151e-05 eta: 1:00:45 time: 0.5744 data_time: 0.0016 memory: 15585 loss: 0.0146 +2023/06/06 07:40:46 - mmengine - INFO - Exp name: convnext_small_4xb256_IF_1m_lr4e-4_aug_5e-1_20230606_051640 +2023/06/06 07:40:46 - mmengine - INFO - Saving checkpoint at 7 epochs +2023/06/06 07:41:22 - mmengine - INFO - Epoch(val) [7][29/29] accuracy/top1: 99.1998 single-label/precision_classwise: [98.76304626464844, 99.74870300292969] single-label/recall_classwise: [99.7979965209961, 98.46511840820312] single-label/f1-score_classwise: [99.27783203125, 99.10275268554688] data_time: 0.2569 time: 0.6451 +2023/06/06 07:42:24 - mmengine - INFO - Epoch(train) [8][ 100/1969] lr: 8.7904e-05 eta: 0:59:01 time: 0.5269 data_time: 0.0016 memory: 15585 loss: 0.0151 +2023/06/06 07:43:23 - mmengine - INFO - Epoch(train) [8][ 200/1969] lr: 8.5432e-05 eta: 0:57:59 time: 0.6385 data_time: 0.0015 memory: 15585 loss: 0.0149 +2023/06/06 07:43:33 - mmengine - INFO - Exp name: convnext_small_4xb256_IF_1m_lr4e-4_aug_5e-1_20230606_051640 +2023/06/06 07:44:21 - mmengine - INFO - Epoch(train) [8][ 300/1969] lr: 8.2989e-05 eta: 0:56:57 time: 0.6128 data_time: 0.0018 memory: 15585 loss: 0.0128 +2023/06/06 07:45:19 - mmengine - INFO - Epoch(train) [8][ 400/1969] lr: 8.0578e-05 eta: 0:55:55 time: 0.5707 data_time: 0.0016 memory: 15585 loss: 0.0105 +2023/06/06 07:46:19 - mmengine - INFO - Epoch(train) [8][ 500/1969] lr: 7.8198e-05 eta: 0:54:54 time: 0.5328 data_time: 0.0015 memory: 15585 loss: 0.0172 +2023/06/06 07:47:17 - mmengine - INFO - Epoch(train) [8][ 600/1969] lr: 7.5851e-05 eta: 0:53:52 time: 0.5547 data_time: 0.0017 memory: 15585 loss: 0.0139 +2023/06/06 07:48:17 - mmengine - INFO - Epoch(train) [8][ 700/1969] lr: 7.3536e-05 eta: 0:52:50 time: 0.5885 data_time: 0.0014 memory: 15585 loss: 0.0112 +2023/06/06 07:49:15 - mmengine - INFO - Epoch(train) [8][ 800/1969] lr: 7.1255e-05 eta: 0:51:49 time: 0.5895 data_time: 0.0016 memory: 15585 loss: 0.0147 +2023/06/06 07:50:13 - mmengine - INFO - Epoch(train) [8][ 900/1969] lr: 6.9008e-05 eta: 0:50:47 time: 0.6017 data_time: 0.0016 memory: 15585 loss: 0.0147 +2023/06/06 07:51:10 - mmengine - INFO - Epoch(train) [8][1000/1969] lr: 6.6796e-05 eta: 0:49:45 time: 0.5416 data_time: 0.0017 memory: 15585 loss: 0.0138 +2023/06/06 07:52:10 - mmengine - INFO - Epoch(train) [8][1100/1969] lr: 6.4618e-05 eta: 0:48:43 time: 0.6131 data_time: 0.0017 memory: 15585 loss: 0.0105 +2023/06/06 07:53:07 - mmengine - INFO - Epoch(train) [8][1200/1969] lr: 6.2477e-05 eta: 0:47:41 time: 0.5806 data_time: 0.0017 memory: 15585 loss: 0.0127 +2023/06/06 07:53:17 - mmengine - INFO - Exp name: convnext_small_4xb256_IF_1m_lr4e-4_aug_5e-1_20230606_051640 +2023/06/06 07:54:05 - mmengine - INFO - Epoch(train) [8][1300/1969] lr: 6.0372e-05 eta: 0:46:40 time: 0.5783 data_time: 0.0016 memory: 15585 loss: 0.0160 +2023/06/06 07:55:02 - mmengine - INFO - Epoch(train) [8][1400/1969] lr: 5.8303e-05 eta: 0:45:38 time: 0.5524 data_time: 0.0015 memory: 15585 loss: 0.0115 +2023/06/06 07:55:58 - mmengine - INFO - Epoch(train) [8][1500/1969] lr: 5.6272e-05 eta: 0:44:36 time: 0.5469 data_time: 0.0016 memory: 15585 loss: 0.0094 +2023/06/06 07:56:57 - mmengine - INFO - Epoch(train) [8][1600/1969] lr: 5.4279e-05 eta: 0:43:35 time: 0.5986 data_time: 0.0015 memory: 15585 loss: 0.0129 +2023/06/06 07:57:57 - mmengine - INFO - Epoch(train) [8][1700/1969] lr: 5.2324e-05 eta: 0:42:34 time: 0.7414 data_time: 0.0015 memory: 15585 loss: 0.0112 +2023/06/06 07:58:55 - mmengine - INFO - Epoch(train) [8][1800/1969] lr: 5.0408e-05 eta: 0:41:32 time: 0.5644 data_time: 0.0017 memory: 15585 loss: 0.0156 +2023/06/06 07:59:54 - mmengine - INFO - Epoch(train) [8][1900/1969] lr: 4.8532e-05 eta: 0:40:31 time: 0.5333 data_time: 0.0017 memory: 15585 loss: 0.0115 +2023/06/06 08:00:33 - mmengine - INFO - Exp name: convnext_small_4xb256_IF_1m_lr4e-4_aug_5e-1_20230606_051640 +2023/06/06 08:00:33 - mmengine - INFO - Saving checkpoint at 8 epochs +2023/06/06 08:01:08 - mmengine - INFO - Epoch(val) [8][29/29] accuracy/top1: 99.3528 single-label/precision_classwise: [99.016845703125, 99.77294158935547] single-label/recall_classwise: [99.81693267822266, 98.78294372558594] single-label/f1-score_classwise: [99.415283203125, 99.27548217773438] data_time: 0.2271 time: 0.6160 +2023/06/06 08:02:10 - mmengine - INFO - Epoch(train) [9][ 100/1969] lr: 4.5451e-05 eta: 0:38:48 time: 0.5428 data_time: 0.0097 memory: 15585 loss: 0.0150 +2023/06/06 08:03:06 - mmengine - INFO - Epoch(train) [9][ 200/1969] lr: 4.3683e-05 eta: 0:37:47 time: 0.5537 data_time: 0.0018 memory: 15585 loss: 0.0145 +2023/06/06 08:03:34 - mmengine - INFO - Exp name: convnext_small_4xb256_IF_1m_lr4e-4_aug_5e-1_20230606_051640 +2023/06/06 08:04:04 - mmengine - INFO - Epoch(train) [9][ 300/1969] lr: 4.1955e-05 eta: 0:36:45 time: 0.6048 data_time: 0.0018 memory: 15585 loss: 0.0176 +2023/06/06 08:04:59 - mmengine - INFO - Epoch(train) [9][ 400/1969] lr: 4.0269e-05 eta: 0:35:43 time: 0.5539 data_time: 0.0016 memory: 15585 loss: 0.0102 +2023/06/06 08:05:55 - mmengine - INFO - Epoch(train) [9][ 500/1969] lr: 3.8626e-05 eta: 0:34:42 time: 0.5768 data_time: 0.0016 memory: 15585 loss: 0.0150 +2023/06/06 08:06:52 - mmengine - INFO - Epoch(train) [9][ 600/1969] lr: 3.7024e-05 eta: 0:33:40 time: 0.5638 data_time: 0.0017 memory: 15585 loss: 0.0162 +2023/06/06 08:07:48 - mmengine - INFO - Epoch(train) [9][ 700/1969] lr: 3.5465e-05 eta: 0:32:39 time: 0.5511 data_time: 0.0018 memory: 15585 loss: 0.0131 +2023/06/06 08:08:44 - mmengine - INFO - Epoch(train) [9][ 800/1969] lr: 3.3950e-05 eta: 0:31:38 time: 0.5371 data_time: 0.0015 memory: 15585 loss: 0.0154 +2023/06/06 08:09:41 - mmengine - INFO - Epoch(train) [9][ 900/1969] lr: 3.2478e-05 eta: 0:30:37 time: 0.5488 data_time: 0.0017 memory: 15585 loss: 0.0070 +2023/06/06 08:10:38 - mmengine - INFO - Epoch(train) [9][1000/1969] lr: 3.1049e-05 eta: 0:29:36 time: 0.6144 data_time: 0.0017 memory: 15585 loss: 0.0131 +2023/06/06 08:11:36 - mmengine - INFO - Epoch(train) [9][1100/1969] lr: 2.9666e-05 eta: 0:28:35 time: 0.5919 data_time: 0.0019 memory: 15585 loss: 0.0121 +2023/06/06 08:12:32 - mmengine - INFO - Epoch(train) [9][1200/1969] lr: 2.8326e-05 eta: 0:27:34 time: 0.5686 data_time: 0.0018 memory: 15585 loss: 0.0113 +2023/06/06 08:13:00 - mmengine - INFO - Exp name: convnext_small_4xb256_IF_1m_lr4e-4_aug_5e-1_20230606_051640 +2023/06/06 08:13:30 - mmengine - INFO - Epoch(train) [9][1300/1969] lr: 2.7032e-05 eta: 0:26:33 time: 0.5744 data_time: 0.0018 memory: 15585 loss: 0.0135 +2023/06/06 08:14:26 - mmengine - INFO - Epoch(train) [9][1400/1969] lr: 2.5783e-05 eta: 0:25:32 time: 0.5824 data_time: 0.0017 memory: 15585 loss: 0.0134 +2023/06/06 08:15:24 - mmengine - INFO - Epoch(train) [9][1500/1969] lr: 2.4580e-05 eta: 0:24:31 time: 0.5801 data_time: 0.0017 memory: 15585 loss: 0.0147 +2023/06/06 08:16:23 - mmengine - INFO - Epoch(train) [9][1600/1969] lr: 2.3422e-05 eta: 0:23:30 time: 0.5650 data_time: 0.0017 memory: 15585 loss: 0.0157 +2023/06/06 08:17:20 - mmengine - INFO - Epoch(train) [9][1700/1969] lr: 2.2311e-05 eta: 0:22:30 time: 0.5444 data_time: 0.0015 memory: 15585 loss: 0.0141 +2023/06/06 08:18:16 - mmengine - INFO - Epoch(train) [9][1800/1969] lr: 2.1246e-05 eta: 0:21:29 time: 0.5492 data_time: 0.0016 memory: 15585 loss: 0.0111 +2023/06/06 08:19:11 - mmengine - INFO - Epoch(train) [9][1900/1969] lr: 2.0229e-05 eta: 0:20:28 time: 0.5414 data_time: 0.0019 memory: 15585 loss: 0.0082 +2023/06/06 08:19:50 - mmengine - INFO - Exp name: convnext_small_4xb256_IF_1m_lr4e-4_aug_5e-1_20230606_051640 +2023/06/06 08:19:50 - mmengine - INFO - Saving checkpoint at 9 epochs +2023/06/06 08:20:25 - mmengine - INFO - Epoch(val) [9][29/29] accuracy/top1: 99.4294 single-label/precision_classwise: [99.14728546142578, 99.78111267089844] single-label/recall_classwise: [99.8232421875, 98.94573974609375] single-label/f1-score_classwise: [99.48411560058594, 99.3616714477539] data_time: 0.2501 time: 0.6393 +2023/06/06 08:21:27 - mmengine - INFO - Epoch(train) [10][ 100/1969] lr: 1.8615e-05 eta: 0:18:46 time: 0.6204 data_time: 0.1034 memory: 15585 loss: 0.0098 +2023/06/06 08:22:24 - mmengine - INFO - Epoch(train) [10][ 200/1969] lr: 1.7725e-05 eta: 0:17:45 time: 0.5342 data_time: 0.0168 memory: 15585 loss: 0.0141 +2023/06/06 08:23:08 - mmengine - INFO - Exp name: convnext_small_4xb256_IF_1m_lr4e-4_aug_5e-1_20230606_051640 +2023/06/06 08:23:20 - mmengine - INFO - Epoch(train) [10][ 300/1969] lr: 1.6881e-05 eta: 0:16:45 time: 0.5726 data_time: 0.0014 memory: 15585 loss: 0.0077 +2023/06/06 08:24:17 - mmengine - INFO - Epoch(train) [10][ 400/1969] lr: 1.6086e-05 eta: 0:15:44 time: 0.5676 data_time: 0.0525 memory: 15585 loss: 0.0166 +2023/06/06 08:25:16 - mmengine - INFO - Epoch(train) [10][ 500/1969] lr: 1.5339e-05 eta: 0:14:44 time: 0.6158 data_time: 0.0996 memory: 15585 loss: 0.0131 +2023/06/06 08:26:13 - mmengine - INFO - Epoch(train) [10][ 600/1969] lr: 1.4640e-05 eta: 0:13:43 time: 0.5693 data_time: 0.0444 memory: 15585 loss: 0.0093 +2023/06/06 08:27:09 - mmengine - INFO - Epoch(train) [10][ 700/1969] lr: 1.3990e-05 eta: 0:12:43 time: 0.5392 data_time: 0.0248 memory: 15585 loss: 0.0223 +2023/06/06 08:28:07 - mmengine - INFO - Epoch(train) [10][ 800/1969] lr: 1.3388e-05 eta: 0:11:43 time: 0.6149 data_time: 0.0550 memory: 15585 loss: 0.0103 +2023/06/06 08:29:04 - mmengine - INFO - Epoch(train) [10][ 900/1969] lr: 1.2835e-05 eta: 0:10:42 time: 0.5793 data_time: 0.0016 memory: 15585 loss: 0.0098 +2023/06/06 08:30:00 - mmengine - INFO - Epoch(train) [10][1000/1969] lr: 1.2331e-05 eta: 0:09:42 time: 0.5574 data_time: 0.0015 memory: 15585 loss: 0.0161 +2023/06/06 08:31:02 - mmengine - INFO - Epoch(train) [10][1100/1969] lr: 1.1876e-05 eta: 0:08:42 time: 0.9886 data_time: 0.0016 memory: 15585 loss: 0.0103 +2023/06/06 08:31:58 - mmengine - INFO - Epoch(train) [10][1200/1969] lr: 1.1470e-05 eta: 0:07:42 time: 0.5682 data_time: 0.0018 memory: 15585 loss: 0.0150 +2023/06/06 08:32:42 - mmengine - INFO - Exp name: convnext_small_4xb256_IF_1m_lr4e-4_aug_5e-1_20230606_051640 +2023/06/06 08:32:54 - mmengine - INFO - Epoch(train) [10][1300/1969] lr: 1.1113e-05 eta: 0:06:41 time: 0.5546 data_time: 0.0017 memory: 15585 loss: 0.0123 +2023/06/06 08:33:51 - mmengine - INFO - Epoch(train) [10][1400/1969] lr: 1.0806e-05 eta: 0:05:41 time: 0.5908 data_time: 0.0017 memory: 15585 loss: 0.0085 +2023/06/06 08:34:48 - mmengine - INFO - Epoch(train) [10][1500/1969] lr: 1.0548e-05 eta: 0:04:41 time: 0.5833 data_time: 0.0016 memory: 15585 loss: 0.0115 +2023/06/06 08:35:44 - mmengine - INFO - Epoch(train) [10][1600/1969] lr: 1.0340e-05 eta: 0:03:41 time: 0.5510 data_time: 0.0016 memory: 15585 loss: 0.0117 +2023/06/06 08:36:41 - mmengine - INFO - Epoch(train) [10][1700/1969] lr: 1.0181e-05 eta: 0:02:41 time: 0.5503 data_time: 0.0019 memory: 15585 loss: 0.0106 +2023/06/06 08:37:37 - mmengine - INFO - Epoch(train) [10][1800/1969] lr: 1.0072e-05 eta: 0:01:41 time: 0.5223 data_time: 0.0017 memory: 15585 loss: 0.0187 +2023/06/06 08:38:33 - mmengine - INFO - Epoch(train) [10][1900/1969] lr: 1.0012e-05 eta: 0:00:41 time: 0.5330 data_time: 0.0016 memory: 15585 loss: 0.0096 +2023/06/06 08:39:17 - mmengine - INFO - Exp name: convnext_small_4xb256_IF_1m_lr4e-4_aug_5e-1_20230606_051640 +2023/06/06 08:39:17 - mmengine - INFO - Saving checkpoint at 10 epochs +2023/06/06 08:39:52 - mmengine - INFO - Epoch(val) [10][29/29] accuracy/top1: 99.4155 single-label/precision_classwise: [99.1964340209961, 99.68778991699219] single-label/recall_classwise: [99.74748992919922, 99.00775146484375] single-label/f1-score_classwise: [99.47119903564453, 99.34661102294922] data_time: 0.2342 time: 0.6225 diff --git a/convnext_small_4xb256_IF_1m_lr4e-4_aug_5e-1/20230606_051640/vis_data/20230606_051640.json b/convnext_small_4xb256_IF_1m_lr4e-4_aug_5e-1/20230606_051640/vis_data/20230606_051640.json new file mode 100644 index 0000000000000000000000000000000000000000..e53dabfd8305948b36969de30b70bcb9aeb4d3eb --- /dev/null +++ b/convnext_small_4xb256_IF_1m_lr4e-4_aug_5e-1/20230606_051640/vis_data/20230606_051640.json @@ -0,0 +1,200 @@ +{"lr": 0.0003999756737991541, "data_time": 0.0017441987991333007, "loss": 0.09458963386714458, "time": 0.6123632192611694, "epoch": 1, "memory": 15587, "step": 100} +{"lr": 0.00039990171605141573, "data_time": 0.029685473442077635, "loss": 0.058537983894348146, "time": 0.6956503629684448, "epoch": 1, "memory": 15587, "step": 200} +{"lr": 0.00039977814310179305, "data_time": 0.0017234563827514648, "loss": 0.058321688696742056, "time": 0.6685066938400268, "epoch": 1, "memory": 15587, "step": 300} +{"lr": 0.00039960498640766363, "data_time": 0.001700282096862793, "loss": 0.05933081284165383, "time": 0.6553350687026978, "epoch": 1, "memory": 15587, "step": 400} +{"lr": 0.00039938229004870185, "data_time": 0.0017362117767333984, "loss": 0.05389338620007038, "time": 0.638532018661499, "epoch": 1, "memory": 15587, "step": 500} +{"lr": 0.00039911011071566005, "data_time": 0.0016216278076171876, "loss": 0.05744111407548189, "time": 0.602908706665039, "epoch": 1, "memory": 15587, "step": 600} +{"lr": 0.0003987885176959352, "data_time": 0.0015788078308105469, "loss": 0.05633894167840481, "time": 1.1324015617370606, "epoch": 1, "memory": 15587, "step": 700} +{"lr": 0.0003984175928559308, "data_time": 0.0017818689346313476, "loss": 0.060062802582979205, "time": 0.6036136150360107, "epoch": 1, "memory": 15587, "step": 800} +{"lr": 0.00039799743062021714, "data_time": 0.001535654067993164, "loss": 0.05464814603328705, "time": 0.5759683847427368, "epoch": 1, "memory": 15587, "step": 900} +{"lr": 0.0003975281379474946, "data_time": 0.0016119003295898438, "loss": 0.04153641015291214, "time": 0.5958358764648437, "epoch": 1, "memory": 15587, "step": 1000} +{"lr": 0.00039700983430336476, "data_time": 0.001715397834777832, "loss": 0.04811144415289163, "time": 0.6263498544692994, "epoch": 1, "memory": 15587, "step": 1100} +{"lr": 0.0003964426516299177, "data_time": 0.0016070127487182618, "loss": 0.04214367736130953, "time": 0.6444735765457154, "epoch": 1, "memory": 15587, "step": 1200} +{"lr": 0.0003958267343121465, "data_time": 0.0014572381973266602, "loss": 0.04770536925643683, "time": 0.6142401456832886, "epoch": 1, "memory": 15587, "step": 1300} +{"lr": 0.00039516223914119037, "data_time": 0.001609635353088379, "loss": 0.03325816858559847, "time": 0.5964842557907104, "epoch": 1, "memory": 15587, "step": 1400} +{"lr": 0.0003944493352744212, "data_time": 0.0015526533126831055, "loss": 0.039781056623905894, "time": 0.6076323509216308, "epoch": 1, "memory": 15587, "step": 1500} +{"lr": 0.0003936882041923822, "data_time": 0.0015211105346679688, "loss": 0.04655163232237101, "time": 0.6154497861862183, "epoch": 1, "memory": 15587, "step": 1600} +{"lr": 0.0003928790396525886, "data_time": 0.0016469955444335938, "loss": 0.0350088344886899, "time": 0.6293362379074097, "epoch": 1, "memory": 15587, "step": 1700} +{"lr": 0.00039202204764020483, "data_time": 0.001549100875854492, "loss": 0.034160731546580794, "time": 1.3417204856872558, "epoch": 1, "memory": 15587, "step": 1800} +{"lr": 0.00039111744631560716, "data_time": 0.0015239238739013672, "loss": 0.04124070554971695, "time": 0.6438556909561157, "epoch": 1, "memory": 15587, "step": 1900} +{"accuracy/top1": 80.1259536743164, "data_time": 0.2369171668743265, "time": 0.700283930219453, "step": 1} +{"lr": 0.0003894811013774212, "data_time": 0.0016692638397216796, "loss": 0.03209449835121632, "time": 0.6675132513046265, "epoch": 2, "memory": 15587, "step": 2069} +{"lr": 0.00038844960758563737, "data_time": 0.001808333396911621, "loss": 0.026527286507189274, "time": 0.6054823875427247, "epoch": 2, "memory": 15585, "step": 2169} +{"lr": 0.00038737141390069504, "data_time": 0.0015696048736572265, "loss": 0.03775574136525393, "time": 0.6006609439849854, "epoch": 2, "memory": 15585, "step": 2269} +{"lr": 0.0003862467947932271, "data_time": 0.0015077829360961915, "loss": 0.03677110578864813, "time": 0.6380847215652465, "epoch": 2, "memory": 15585, "step": 2369} +{"lr": 0.000385076036552165, "data_time": 0.0015467166900634765, "loss": 0.03364796955138445, "time": 0.5798214197158813, "epoch": 2, "memory": 15585, "step": 2469} +{"lr": 0.00038385943721186026, "data_time": 0.002039456367492676, "loss": 0.03959393538534641, "time": 0.6175883054733277, "epoch": 2, "memory": 15585, "step": 2569} +{"lr": 0.0003825973064762139, "data_time": 0.0014548063278198241, "loss": 0.02873118189163506, "time": 0.6704680681228637, "epoch": 2, "memory": 15585, "step": 2669} +{"lr": 0.000381289965639839, "data_time": 0.001670694351196289, "loss": 0.03372421395033598, "time": 0.6355009078979492, "epoch": 2, "memory": 15585, "step": 2769} +{"lr": 0.00037993774750626663, "data_time": 0.001581096649169922, "loss": 0.03461470026522875, "time": 0.6274290561676026, "epoch": 2, "memory": 15585, "step": 2869} +{"lr": 0.00037854099630322966, "data_time": 0.0015678882598876953, "loss": 0.04298805464059115, "time": 0.6427930593490601, "epoch": 2, "memory": 15585, "step": 2969} +{"lr": 0.00037710006759503174, "data_time": 0.0014120817184448242, "loss": 0.03247109018266201, "time": 0.6416108846664429, "epoch": 2, "memory": 15585, "step": 3069} +{"lr": 0.00037561532819203284, "data_time": 0.0016913652420043946, "loss": 0.032137765083462, "time": 0.6324577569961548, "epoch": 2, "memory": 15585, "step": 3169} +{"lr": 0.00037408715605727376, "data_time": 0.0018703937530517578, "loss": 0.0294894490391016, "time": 0.6005995035171509, "epoch": 2, "memory": 15585, "step": 3269} +{"lr": 0.00037251594021025933, "data_time": 0.0016335248947143555, "loss": 0.02664182074368, "time": 0.6230206966400147, "epoch": 2, "memory": 15585, "step": 3369} +{"lr": 0.00037090208062792615, "data_time": 0.001602005958557129, "loss": 0.023595809610560535, "time": 0.5774687051773071, "epoch": 2, "memory": 15585, "step": 3469} +{"lr": 0.00036924598814282443, "data_time": 0.0014952659606933595, "loss": 0.029191960766911506, "time": 0.5890163660049439, "epoch": 2, "memory": 15585, "step": 3569} +{"lr": 0.00036754808433853343, "data_time": 0.001488971710205078, "loss": 0.027995210606604815, "time": 0.5980283260345459, "epoch": 2, "memory": 15585, "step": 3669} +{"lr": 0.00036580880144233954, "data_time": 0.0015398025512695312, "loss": 0.03276628088206053, "time": 0.6006958246231079, "epoch": 2, "memory": 15585, "step": 3769} +{"lr": 0.0003640285822152096, "data_time": 0.0015473365783691406, "loss": 0.03431772254407406, "time": 0.5462408304214478, "epoch": 2, "memory": 15585, "step": 3869} +{"accuracy/top1": 91.78873443603516, "data_time": 0.2112113316853841, "time": 0.5998448212941487, "step": 2} +{"lr": 0.0003609282333974528, "data_time": 0.03719151020050049, "loss": 0.024518429953604936, "time": 0.5542458295822144, "epoch": 3, "memory": 15585, "step": 4038} +{"lr": 0.0003590401733608027, "data_time": 0.08352704048156738, "loss": 0.03717135079205036, "time": 0.6016617774963379, "epoch": 3, "memory": 15585, "step": 4138} +{"lr": 0.00035711290005063985, "data_time": 0.05858302116394043, "loss": 0.027805393422022463, "time": 0.6124847650527954, "epoch": 3, "memory": 15585, "step": 4238} +{"lr": 0.0003551469040837546, "data_time": 0.0022240638732910155, "loss": 0.029554964695125818, "time": 0.6046499967575073, "epoch": 3, "memory": 15585, "step": 4338} +{"lr": 0.00035314268593437953, "data_time": 0.001641535758972168, "loss": 0.03180052498355508, "time": 0.5970683097839355, "epoch": 3, "memory": 15585, "step": 4438} +{"lr": 0.00035110075580678697, "data_time": 0.0016754627227783202, "loss": 0.028151841554790735, "time": 0.6520636796951294, "epoch": 3, "memory": 15585, "step": 4538} +{"lr": 0.00034902163350540724, "data_time": 0.0019218206405639648, "loss": 0.023087999410927297, "time": 0.5890148162841797, "epoch": 3, "memory": 15585, "step": 4638} +{"lr": 0.0003469058483025052, "data_time": 0.0017933368682861329, "loss": 0.026231523649767042, "time": 0.5728699684143066, "epoch": 3, "memory": 15585, "step": 4738} +{"lr": 0.0003447539388034461, "data_time": 0.001554727554321289, "loss": 0.02800388978794217, "time": 0.6035315036773682, "epoch": 3, "memory": 15585, "step": 4838} +{"lr": 0.0003425664528095857, "data_time": 0.0018976449966430664, "loss": 0.021152870915830135, "time": 0.6211369514465332, "epoch": 3, "memory": 15585, "step": 4938} +{"lr": 0.00034034394717881845, "data_time": 0.001685929298400879, "loss": 0.026131332106888294, "time": 0.5432857990264892, "epoch": 3, "memory": 15585, "step": 5038} +{"lr": 0.00033808698768382034, "data_time": 0.00146176815032959, "loss": 0.03089748863130808, "time": 0.5564919710159302, "epoch": 3, "memory": 15585, "step": 5138} +{"lr": 0.00033579614886802443, "data_time": 0.001519179344177246, "loss": 0.04191914033144713, "time": 0.6116936445236206, "epoch": 3, "memory": 15585, "step": 5238} +{"lr": 0.00033347201389936123, "data_time": 0.001765871047973633, "loss": 0.02352065369486809, "time": 0.5813561201095581, "epoch": 3, "memory": 15585, "step": 5338} +{"lr": 0.0003311151744218034, "data_time": 0.0015958070755004883, "loss": 0.030639497749507427, "time": 0.5732959032058715, "epoch": 3, "memory": 15585, "step": 5438} +{"lr": 0.0003287262304047557, "data_time": 0.0017245292663574218, "loss": 0.02399143176153302, "time": 0.6123759508132934, "epoch": 3, "memory": 15585, "step": 5538} +{"lr": 0.0003263057899903226, "data_time": 0.0014634370803833009, "loss": 0.023891198588535188, "time": 0.5682729482650757, "epoch": 3, "memory": 15585, "step": 5638} +{"lr": 0.00032385446933849547, "data_time": 0.0016894817352294921, "loss": 0.0306385925039649, "time": 0.6199266910552979, "epoch": 3, "memory": 15585, "step": 5738} +{"lr": 0.0003213728924703014, "data_time": 0.0017627477645874023, "loss": 0.02045824136584997, "time": 0.6045604705810547, "epoch": 3, "memory": 15585, "step": 5838} +{"accuracy/top1": 95.73084259033203, "data_time": 0.1199626604715983, "time": 0.5050847053527832, "step": 3} +{"lr": 0.00031711202351665616, "data_time": 0.0020610809326171873, "loss": 0.023080874979496003, "time": 0.5588046073913574, "epoch": 4, "memory": 15585, "step": 6007} +{"lr": 0.0003145522137064149, "data_time": 0.0015259742736816405, "loss": 0.02575897970236838, "time": 0.6027894258499146, "epoch": 4, "memory": 15585, "step": 6107} +{"lr": 0.0003119645157107334, "data_time": 0.0016039133071899414, "loss": 0.036812640447169545, "time": 0.6135179996490479, "epoch": 4, "memory": 15585, "step": 6207} +{"lr": 0.00030934958826756915, "data_time": 0.0015448331832885742, "loss": 0.017345844767987728, "time": 0.5793030261993408, "epoch": 4, "memory": 15585, "step": 6307} +{"lr": 0.0003067080970465515, "data_time": 0.001685023307800293, "loss": 0.02414605803787708, "time": 0.579094648361206, "epoch": 4, "memory": 15585, "step": 6407} +{"lr": 0.00030404071447952345, "data_time": 0.001581573486328125, "loss": 0.01597123914398253, "time": 0.6501397848129272, "epoch": 4, "memory": 15585, "step": 6507} +{"lr": 0.0003013481195893653, "data_time": 0.0017234325408935548, "loss": 0.024644845677539707, "time": 0.5737666130065918, "epoch": 4, "memory": 15585, "step": 6607} +{"lr": 0.0002986309978171385, "data_time": 0.0016404390335083008, "loss": 0.028770623495802283, "time": 0.6156988859176635, "epoch": 4, "memory": 15585, "step": 6707} +{"lr": 0.0002958900408475955, "data_time": 0.0016281843185424806, "loss": 0.019331029430031777, "time": 0.5952183485031128, "epoch": 4, "memory": 15585, "step": 6807} +{"lr": 0.0002931259464331015, "data_time": 0.001688098907470703, "loss": 0.023258120100945235, "time": 0.6249657869338989, "epoch": 4, "memory": 15585, "step": 6907} +{"lr": 0.00029033941821601176, "data_time": 0.0020334482192993163, "loss": 0.029206813685595988, "time": 0.5614066839218139, "epoch": 4, "memory": 15585, "step": 7007} +{"lr": 0.0002875311655495464, "data_time": 0.0014782428741455078, "loss": 0.024458643654361366, "time": 0.556343412399292, "epoch": 4, "memory": 15585, "step": 7107} +{"lr": 0.0002847019033172168, "data_time": 0.0016544580459594727, "loss": 0.021977328462526202, "time": 0.5464712381362915, "epoch": 4, "memory": 15585, "step": 7207} +{"lr": 0.0002818523517508384, "data_time": 0.001614999771118164, "loss": 0.02037540073506534, "time": 0.595465087890625, "epoch": 4, "memory": 15585, "step": 7307} +{"lr": 0.00027898323624718594, "data_time": 0.0017370462417602539, "loss": 0.022024381440132856, "time": 0.585851502418518, "epoch": 4, "memory": 15585, "step": 7407} +{"lr": 0.000276095287183333, "data_time": 0.0014989852905273437, "loss": 0.023259676760062575, "time": 0.6028532743453979, "epoch": 4, "memory": 15585, "step": 7507} +{"lr": 0.00027318923973072227, "data_time": 0.0017900705337524415, "loss": 0.028592138458043335, "time": 0.6054024696350098, "epoch": 4, "memory": 15585, "step": 7607} +{"lr": 0.00027026583366801727, "data_time": 0.0016146659851074218, "loss": 0.027045744750648738, "time": 0.576863956451416, "epoch": 4, "memory": 15585, "step": 7707} +{"lr": 0.0002673258131927808, "data_time": 0.001558542251586914, "loss": 0.02875189818441868, "time": 0.5726858139038086, "epoch": 4, "memory": 15585, "step": 7807} +{"accuracy/top1": 97.62012481689453, "data_time": 0.22566306591033936, "time": 0.6116198460261028, "step": 4} +{"lr": 0.0002623215076436499, "data_time": 0.04466564655303955, "loss": 0.028714340645819902, "time": 0.58972647190094, "epoch": 5, "memory": 15585, "step": 7976} +{"lr": 0.00025934052007928784, "data_time": 0.029674386978149413, "loss": 0.028971432987600566, "time": 0.5518029928207397, "epoch": 5, "memory": 15585, "step": 8076} +{"lr": 0.0002563456993074567, "data_time": 0.0015370607376098632, "loss": 0.028553877910599112, "time": 0.6341614961624146, "epoch": 5, "memory": 15585, "step": 8176} +{"lr": 0.00025333780770542177, "data_time": 0.0015366315841674806, "loss": 0.019479702226817608, "time": 0.5838659048080445, "epoch": 5, "memory": 15585, "step": 8276} +{"lr": 0.00025031761097782693, "data_time": 0.0017219066619873046, "loss": 0.02388954218477011, "time": 0.5737927675247192, "epoch": 5, "memory": 15585, "step": 8376} +{"lr": 0.0002472858779617731, "data_time": 0.001516866683959961, "loss": 0.022251126868650318, "time": 0.5874070167541504, "epoch": 5, "memory": 15585, "step": 8476} +{"lr": 0.0002442433804310988, "data_time": 0.0016100645065307618, "loss": 0.030083376727998257, "time": 0.6104545593261719, "epoch": 5, "memory": 15585, "step": 8576} +{"lr": 0.00024119089289991304, "data_time": 0.0015225887298583984, "loss": 0.01802371027879417, "time": 0.5405940294265748, "epoch": 5, "memory": 15585, "step": 8676} +{"lr": 0.00023812919242543494, "data_time": 0.0016010522842407227, "loss": 0.022229759022593498, "time": 0.5434931993484498, "epoch": 5, "memory": 15585, "step": 8776} +{"lr": 0.00023505905841017684, "data_time": 0.0015564203262329102, "loss": 0.023329756082966925, "time": 0.555145525932312, "epoch": 5, "memory": 15585, "step": 8876} +{"lr": 0.00023198127240353698, "data_time": 0.0016033649444580078, "loss": 0.02470110231079161, "time": 0.5642518520355224, "epoch": 5, "memory": 15585, "step": 8976} +{"lr": 0.00022889661790284532, "data_time": 0.0014160871505737305, "loss": 0.02083730003796518, "time": 0.6121579885482789, "epoch": 5, "memory": 15585, "step": 9076} +{"lr": 0.000225805880153911, "data_time": 0.0015762567520141602, "loss": 0.0174110543448478, "time": 0.5741606473922729, "epoch": 5, "memory": 15585, "step": 9176} +{"lr": 0.00022270984595112745, "data_time": 0.001848888397216797, "loss": 0.02076743058860302, "time": 0.5884270191192627, "epoch": 5, "memory": 15585, "step": 9276} +{"lr": 0.00021960930343718077, "data_time": 0.0014249563217163086, "loss": 0.0137361329048872, "time": 0.5783840894699097, "epoch": 5, "memory": 15585, "step": 9376} +{"lr": 0.00021650504190241605, "data_time": 0.0017305612564086914, "loss": 0.025541602773591875, "time": 0.5636526584625244, "epoch": 5, "memory": 15585, "step": 9476} +{"lr": 0.00021339785158391245, "data_time": 0.0019342660903930663, "loss": 0.02035396988503635, "time": 0.5900669813156127, "epoch": 5, "memory": 15585, "step": 9576} +{"lr": 0.00021028852346431534, "data_time": 0.0016449451446533202, "loss": 0.01910709673538804, "time": 0.5614126443862915, "epoch": 5, "memory": 15585, "step": 9676} +{"lr": 0.00020717784907047976, "data_time": 0.0016514778137207032, "loss": 0.021278704004362225, "time": 0.5583648204803466, "epoch": 5, "memory": 15585, "step": 9776} +{"accuracy/top1": 98.43081665039062, "data_time": 0.21341134707132975, "time": 0.6042571067810059, "step": 5} +{"lr": 0.0002019199632200586, "data_time": 0.0016721010208129883, "loss": 0.01492706169374287, "time": 0.6085896968841553, "epoch": 6, "memory": 15585, "step": 9945} +{"lr": 0.00019880959773413697, "data_time": 0.0015503883361816407, "loss": 0.017514523956924676, "time": 0.5879445791244506, "epoch": 6, "memory": 15585, "step": 10045} +{"lr": 0.00019570080810944632, "data_time": 0.0015748023986816406, "loss": 0.01872837315313518, "time": 0.6430665016174316, "epoch": 6, "memory": 15585, "step": 10145} +{"lr": 0.00019259438573575975, "data_time": 0.001423478126525879, "loss": 0.019388327561318874, "time": 0.5860364437103271, "epoch": 6, "memory": 15585, "step": 10245} +{"lr": 0.00018949112140023044, "data_time": 0.0015466451644897462, "loss": 0.027449066191911696, "time": 0.5448642015457154, "epoch": 6, "memory": 15585, "step": 10345} +{"lr": 0.00018639180508608498, "data_time": 0.0017944097518920899, "loss": 0.018055544793605806, "time": 0.5434005975723266, "epoch": 6, "memory": 15585, "step": 10445} +{"lr": 0.00018329722577152123, "data_time": 0.0015152931213378907, "loss": 0.012535664858296514, "time": 0.5646077156066894, "epoch": 6, "memory": 15585, "step": 10545} +{"lr": 0.0001802081712288607, "data_time": 0.0017061710357666015, "loss": 0.014948234194889665, "time": 0.5422788381576538, "epoch": 6, "memory": 15585, "step": 10645} +{"lr": 0.00017712542782401123, "data_time": 0.0015752077102661132, "loss": 0.024940438149496914, "time": 0.561505913734436, "epoch": 6, "memory": 15585, "step": 10745} +{"lr": 0.00017404978031628336, "data_time": 0.0017782926559448242, "loss": 0.013869933737441897, "time": 0.5385247468948364, "epoch": 6, "memory": 15585, "step": 10845} +{"lr": 0.00017098201165861962, "data_time": 0.001707601547241211, "loss": 0.015723573695868253, "time": 0.5790489912033081, "epoch": 6, "memory": 15585, "step": 10945} +{"lr": 0.00016792290279828072, "data_time": 0.0016601085662841797, "loss": 0.019538279995322226, "time": 0.5651997089385986, "epoch": 6, "memory": 15585, "step": 11045} +{"lr": 0.00016487323247804387, "data_time": 0.0014456510543823242, "loss": 0.02261862359009683, "time": 0.5819118976593017, "epoch": 6, "memory": 15585, "step": 11145} +{"lr": 0.00016183377703796263, "data_time": 0.002166295051574707, "loss": 0.01894620726816356, "time": 0.5798263788223267, "epoch": 6, "memory": 15585, "step": 11245} +{"lr": 0.00015880531021773676, "data_time": 0.001630687713623047, "loss": 0.015934361144900322, "time": 0.5687504529953002, "epoch": 6, "memory": 15585, "step": 11345} +{"lr": 0.00015578860295974566, "data_time": 0.0015659809112548828, "loss": 0.020451073767617345, "time": 0.5816017389297485, "epoch": 6, "memory": 15585, "step": 11445} +{"lr": 0.0001527844232127924, "data_time": 0.0018509864807128907, "loss": 0.016789604676887393, "time": 0.6481788158416748, "epoch": 6, "memory": 15585, "step": 11545} +{"lr": 0.000149793535736612, "data_time": 0.0017035961151123046, "loss": 0.027278015110641717, "time": 0.6013190984725952, "epoch": 6, "memory": 15585, "step": 11645} +{"lr": 0.00014681670190718876, "data_time": 0.001486515998840332, "loss": 0.020000103628262877, "time": 0.579273796081543, "epoch": 6, "memory": 15585, "step": 11745} +{"accuracy/top1": 98.83789825439453, "data_time": 0.2203480323155721, "time": 0.6101419369379679, "step": 6} +{"lr": 0.00014181991425636192, "data_time": 0.0015768051147460938, "loss": 0.013803625106811523, "time": 0.5759463548660279, "epoch": 7, "memory": 15585, "step": 11914} +{"lr": 0.00013888463509383338, "data_time": 0.0016977310180664063, "loss": 0.016744786640629172, "time": 0.5868659257888794, "epoch": 7, "memory": 15585, "step": 12014} +{"lr": 0.000135966186604959, "data_time": 0.0015201330184936523, "loss": 0.019931176863610744, "time": 0.5900272369384766, "epoch": 7, "memory": 15585, "step": 12114} +{"lr": 0.0001330653117252752, "data_time": 0.0016937494277954102, "loss": 0.013702026475220919, "time": 0.6056107759475708, "epoch": 7, "memory": 15585, "step": 12214} +{"lr": 0.00013018274891668863, "data_time": 0.0017450809478759765, "loss": 0.015689974883571267, "time": 0.5967775821685791, "epoch": 7, "memory": 15585, "step": 12314} +{"lr": 0.000127319231979489, "data_time": 0.001546025276184082, "loss": 0.022085619112476706, "time": 0.5849074840545654, "epoch": 7, "memory": 15585, "step": 12414} +{"lr": 0.0001244754898655487, "data_time": 0.0016199588775634766, "loss": 0.0158816359937191, "time": 0.618298864364624, "epoch": 7, "memory": 15585, "step": 12514} +{"lr": 0.0001216522464927582, "data_time": 0.0019073724746704102, "loss": 0.014288743631914258, "time": 0.5871162652969361, "epoch": 7, "memory": 15585, "step": 12614} +{"lr": 0.00011885022056074007, "data_time": 0.001526618003845215, "loss": 0.014623307110741734, "time": 0.5565303802490235, "epoch": 7, "memory": 15585, "step": 12714} +{"lr": 0.00011607012536789545, "data_time": 0.0016824960708618163, "loss": 0.01136747356504202, "time": 0.7399872303009033, "epoch": 7, "memory": 15585, "step": 12814} +{"lr": 0.00011331266862982092, "data_time": 0.0021506309509277343, "loss": 0.01907173115760088, "time": 0.5873501777648926, "epoch": 7, "memory": 15585, "step": 12914} +{"lr": 0.00011057855229914932, "data_time": 0.001657724380493164, "loss": 0.018510606605559587, "time": 0.5710991859436035, "epoch": 7, "memory": 15585, "step": 13014} +{"lr": 0.0001078684723868573, "data_time": 0.0015811681747436523, "loss": 0.020667529525235297, "time": 0.6221235036849976, "epoch": 7, "memory": 15585, "step": 13114} +{"lr": 0.00010518311878508484, "data_time": 0.001685357093811035, "loss": 0.013659366965293884, "time": 0.5702636480331421, "epoch": 7, "memory": 15585, "step": 13214} +{"lr": 0.00010252317509151276, "data_time": 0.001720738410949707, "loss": 0.01599430530332029, "time": 0.5902864694595337, "epoch": 7, "memory": 15585, "step": 13314} +{"lr": 9.98893184353425e-05, "data_time": 0.0015124082565307617, "loss": 0.009631693596020341, "time": 0.5261112451553345, "epoch": 7, "memory": 15585, "step": 13414} +{"lr": 9.728221930492262e-05, "data_time": 0.0018680810928344727, "loss": 0.01863910616375506, "time": 0.5468105792999267, "epoch": 7, "memory": 15585, "step": 13514} +{"lr": 9.470254137706575e-05, "data_time": 0.001739501953125, "loss": 0.008839649660512806, "time": 0.5772305727005005, "epoch": 7, "memory": 15585, "step": 13614} +{"lr": 9.215094134809975e-05, "data_time": 0.001617121696472168, "loss": 0.01457508341409266, "time": 0.5743704080581665, "epoch": 7, "memory": 15585, "step": 13714} +{"accuracy/top1": 99.19975280761719, "data_time": 0.25693716208140055, "time": 0.6450648625691732, "step": 7} +{"lr": 8.790437228679661e-05, "data_time": 0.0016317367553710938, "loss": 0.015079016098752618, "time": 0.5268658876419068, "epoch": 8, "memory": 15585, "step": 13883} +{"lr": 8.543150502338196e-05, "data_time": 0.0015479564666748048, "loss": 0.014931700145825743, "time": 0.6385215282440185, "epoch": 8, "memory": 15585, "step": 13983} +{"lr": 8.298907574244125e-05, "data_time": 0.0018334150314331054, "loss": 0.012764481781050562, "time": 0.6127574443817139, "epoch": 8, "memory": 15585, "step": 14083} +{"lr": 8.057770620156745e-05, "data_time": 0.0016407489776611329, "loss": 0.010467389039695263, "time": 0.5706641197204589, "epoch": 8, "memory": 15585, "step": 14183} +{"lr": 7.819801025162347e-05, "data_time": 0.0015373468399047852, "loss": 0.017235457338392733, "time": 0.5328397035598755, "epoch": 8, "memory": 15585, "step": 14283} +{"lr": 7.585059368047708e-05, "data_time": 0.001691913604736328, "loss": 0.013922725338488817, "time": 0.5547436714172364, "epoch": 8, "memory": 15585, "step": 14383} +{"lr": 7.353605405878812e-05, "data_time": 0.0014408588409423827, "loss": 0.011247835727408529, "time": 0.5884881019592285, "epoch": 8, "memory": 15585, "step": 14483} +{"lr": 7.125498058788827e-05, "data_time": 0.0016216516494750976, "loss": 0.014739425061270595, "time": 0.5894668340682984, "epoch": 8, "memory": 15585, "step": 14583} +{"lr": 6.900795394979036e-05, "data_time": 0.001628541946411133, "loss": 0.014728918811306358, "time": 0.6016840457916259, "epoch": 8, "memory": 15585, "step": 14683} +{"lr": 6.679554615936711e-05, "data_time": 0.001741790771484375, "loss": 0.0137552491389215, "time": 0.541572904586792, "epoch": 8, "memory": 15585, "step": 14783} +{"lr": 6.461832041873565e-05, "data_time": 0.0016823768615722655, "loss": 0.010548897786065936, "time": 0.6130940914154053, "epoch": 8, "memory": 15585, "step": 14883} +{"lr": 6.247683097388664e-05, "data_time": 0.001662135124206543, "loss": 0.012707919208332896, "time": 0.5806445360183716, "epoch": 8, "memory": 15585, "step": 14983} +{"lr": 6.037162297359161e-05, "data_time": 0.0016053438186645508, "loss": 0.016017212765291333, "time": 0.5783066034317017, "epoch": 8, "memory": 15585, "step": 15083} +{"lr": 5.830323233062794e-05, "data_time": 0.0015491962432861329, "loss": 0.011485350038856267, "time": 0.5523838520050048, "epoch": 8, "memory": 15585, "step": 15183} +{"lr": 5.6272185585353725e-05, "data_time": 0.0015683889389038086, "loss": 0.009434764878824354, "time": 0.5469341039657593, "epoch": 8, "memory": 15585, "step": 15283} +{"lr": 5.4278999771668655e-05, "data_time": 0.0015121936798095704, "loss": 0.012850778782740236, "time": 0.5985788106918335, "epoch": 8, "memory": 15585, "step": 15383} +{"lr": 5.2324182285395304e-05, "data_time": 0.001495218276977539, "loss": 0.011221066955477, "time": 0.7413820266723633, "epoch": 8, "memory": 15585, "step": 15483} +{"lr": 5.0408230755113716e-05, "data_time": 0.0017287254333496094, "loss": 0.015580749697983265, "time": 0.564391827583313, "epoch": 8, "memory": 15585, "step": 15583} +{"lr": 4.853163291548285e-05, "data_time": 0.0016614913940429688, "loss": 0.011525907646864652, "time": 0.5333415746688843, "epoch": 8, "memory": 15585, "step": 15683} +{"accuracy/top1": 99.35284423828125, "data_time": 0.22713043689727783, "time": 0.6159555673599243, "step": 8} +{"lr": 4.5450966211010106e-05, "data_time": 0.009714245796203613, "loss": 0.014958574995398521, "time": 0.5427632331848145, "epoch": 9, "memory": 15585, "step": 15852} +{"lr": 4.368257232393176e-05, "data_time": 0.0017705202102661134, "loss": 0.014493309147655963, "time": 0.5537340879440308, "epoch": 9, "memory": 15585, "step": 15952} +{"lr": 4.195524424634855e-05, "data_time": 0.0018399953842163086, "loss": 0.01762465056963265, "time": 0.6047696113586426, "epoch": 9, "memory": 15585, "step": 16052} +{"lr": 4.0269421695945e-05, "data_time": 0.0016439437866210937, "loss": 0.010192060191184283, "time": 0.553926968574524, "epoch": 9, "memory": 15585, "step": 16152} +{"lr": 3.862553382454128e-05, "data_time": 0.0016122817993164062, "loss": 0.014994328608736396, "time": 0.5768372297286988, "epoch": 9, "memory": 15585, "step": 16252} +{"lr": 3.702399910884571e-05, "data_time": 0.0017409086227416991, "loss": 0.01618064888752997, "time": 0.5637802839279175, "epoch": 9, "memory": 15585, "step": 16352} +{"lr": 3.5465225243925684e-05, "data_time": 0.0018478631973266602, "loss": 0.013086614571511746, "time": 0.5510546207427979, "epoch": 9, "memory": 15585, "step": 16452} +{"lr": 3.394960903942234e-05, "data_time": 0.0015231609344482423, "loss": 0.015384511649608612, "time": 0.5371349334716797, "epoch": 9, "memory": 15585, "step": 16552} +{"lr": 3.247753631853657e-05, "data_time": 0.001718449592590332, "loss": 0.007045535370707512, "time": 0.5488334655761719, "epoch": 9, "memory": 15585, "step": 16652} +{"lr": 3.104938181981207e-05, "data_time": 0.0016681909561157226, "loss": 0.01306831338442862, "time": 0.6143965721130371, "epoch": 9, "memory": 15585, "step": 16752} +{"lr": 2.966550910173992e-05, "data_time": 0.001856088638305664, "loss": 0.012065926007926464, "time": 0.5918649911880494, "epoch": 9, "memory": 15585, "step": 16852} +{"lr": 2.8326270450208946e-05, "data_time": 0.0017817497253417968, "loss": 0.011343830358237028, "time": 0.5685947656631469, "epoch": 9, "memory": 15585, "step": 16952} +{"lr": 2.7032006788826352e-05, "data_time": 0.0018084049224853516, "loss": 0.013476584572345018, "time": 0.5744309186935425, "epoch": 9, "memory": 15585, "step": 17052} +{"lr": 2.5783047592130032e-05, "data_time": 0.0016757011413574218, "loss": 0.01341892620548606, "time": 0.5823689699172974, "epoch": 9, "memory": 15585, "step": 17152} +{"lr": 2.4579710801716043e-05, "data_time": 0.0016843557357788086, "loss": 0.014690388226881623, "time": 0.5800591468811035, "epoch": 9, "memory": 15585, "step": 17252} +{"lr": 2.3422302745301813e-05, "data_time": 0.0016840696334838867, "loss": 0.01572784800082445, "time": 0.5649601221084595, "epoch": 9, "memory": 15585, "step": 17352} +{"lr": 2.2311118058745624e-05, "data_time": 0.0015390634536743164, "loss": 0.01409651730209589, "time": 0.5444443702697754, "epoch": 9, "memory": 15585, "step": 17452} +{"lr": 2.1246439611042613e-05, "data_time": 0.0016150236129760741, "loss": 0.011113508930429816, "time": 0.5491689443588257, "epoch": 9, "memory": 15585, "step": 17552} +{"lr": 2.0228538432315955e-05, "data_time": 0.0018887996673583984, "loss": 0.008217486878857016, "time": 0.5413547277450561, "epoch": 9, "memory": 15585, "step": 17652} +{"accuracy/top1": 99.42938995361328, "data_time": 0.25013980865478513, "time": 0.6392630577087403, "step": 9} +{"lr": 1.861533120597442e-05, "data_time": 0.10335335731506348, "loss": 0.009815009543672204, "time": 0.6204253435134888, "epoch": 10, "memory": 15585, "step": 17821} +{"lr": 1.7724513410024274e-05, "data_time": 0.016815757751464842, "loss": 0.014071066724136471, "time": 0.5341600179672241, "epoch": 10, "memory": 15585, "step": 17921} +{"lr": 1.6881369443034795e-05, "data_time": 0.001433730125427246, "loss": 0.00765361082740128, "time": 0.5726219892501831, "epoch": 10, "memory": 15585, "step": 18021} +{"lr": 1.6086113940151352e-05, "data_time": 0.05248734951019287, "loss": 0.016640963777899743, "time": 0.5675643444061279, "epoch": 10, "memory": 15585, "step": 18121} +{"lr": 1.5338949345780815e-05, "data_time": 0.09957523345947265, "loss": 0.013072047475725412, "time": 0.615776252746582, "epoch": 10, "memory": 15585, "step": 18221} +{"lr": 1.4640065862056652e-05, "data_time": 0.044413280487060544, "loss": 0.009331522695720195, "time": 0.5693021535873413, "epoch": 10, "memory": 15585, "step": 18321} +{"lr": 1.3989641400419702e-05, "data_time": 0.024750924110412596, "loss": 0.022307624062523245, "time": 0.5392184972763061, "epoch": 10, "memory": 15585, "step": 18421} +{"lr": 1.33878415363284e-05, "data_time": 0.055040931701660155, "loss": 0.010318919736891984, "time": 0.6149203062057496, "epoch": 10, "memory": 15585, "step": 18521} +{"lr": 1.2834819467108807e-05, "data_time": 0.0015883684158325196, "loss": 0.009763891529291868, "time": 0.5793373346328735, "epoch": 10, "memory": 15585, "step": 18621} +{"lr": 1.2330715972956002e-05, "data_time": 0.0015215873718261719, "loss": 0.016147507447749378, "time": 0.5573807716369629, "epoch": 10, "memory": 15585, "step": 18721} +{"lr": 1.187565938109641e-05, "data_time": 0.001622319221496582, "loss": 0.010331430146470666, "time": 0.9885760307312011, "epoch": 10, "memory": 15585, "step": 18821} +{"lr": 1.1469765533119946e-05, "data_time": 0.001847052574157715, "loss": 0.014957475429400801, "time": 0.56817147731781, "epoch": 10, "memory": 15585, "step": 18921} +{"lr": 1.1113137755490966e-05, "data_time": 0.0016748428344726563, "loss": 0.012262860126793385, "time": 0.5546363830566406, "epoch": 10, "memory": 15585, "step": 19021} +{"lr": 1.0805866833244799e-05, "data_time": 0.0017088890075683595, "loss": 0.008473808877170086, "time": 0.5907685041427613, "epoch": 10, "memory": 15585, "step": 19121} +{"lr": 1.0548030986877194e-05, "data_time": 0.0016157388687133788, "loss": 0.011519062705338002, "time": 0.5833476781845093, "epoch": 10, "memory": 15585, "step": 19221} +{"lr": 1.0339695852431895e-05, "data_time": 0.0015515327453613282, "loss": 0.011657098587602378, "time": 0.5510348320007324, "epoch": 10, "memory": 15585, "step": 19321} +{"lr": 1.0180914464792136e-05, "data_time": 0.0019409894943237306, "loss": 0.01063668834976852, "time": 0.5503006219863892, "epoch": 10, "memory": 15585, "step": 19421} +{"lr": 1.007172724417983e-05, "data_time": 0.0016666173934936524, "loss": 0.01871240623295307, "time": 0.5223334550857544, "epoch": 10, "memory": 15585, "step": 19521} +{"lr": 1.0012161985865822e-05, "data_time": 0.001567983627319336, "loss": 0.009617753140628339, "time": 0.5329965829849244, "epoch": 10, "memory": 15585, "step": 19621} +{"accuracy/top1": 99.41547393798828, "data_time": 0.2342461109161377, "time": 0.622505521774292, "step": 10} diff --git a/convnext_small_4xb256_IF_1m_lr4e-4_aug_5e-1/20230606_051640/vis_data/config.py b/convnext_small_4xb256_IF_1m_lr4e-4_aug_5e-1/20230606_051640/vis_data/config.py new file mode 100644 index 0000000000000000000000000000000000000000..387bb5b9606b514ed2e86ad5d7d57200ef9f0bd9 --- /dev/null +++ b/convnext_small_4xb256_IF_1m_lr4e-4_aug_5e-1/20230606_051640/vis_data/config.py @@ -0,0 +1,264 @@ +optim_wrapper = dict( + optimizer=dict( + type='AdamW', + lr=0.0004, + weight_decay=0.05, + eps=1e-08, + betas=(0.9, 0.999), + _scope_='mmpretrain'), + paramwise_cfg=dict( + norm_decay_mult=0.0, + bias_decay_mult=0.0, + flat_decay_mult=0.0, + custom_keys=dict({ + '.absolute_pos_embed': dict(decay_mult=0.0), + '.relative_position_bias_table': dict(decay_mult=0.0) + })), + type='AmpOptimWrapper', + dtype='bfloat16', + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=False, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=4096) +model = dict( + type='ImageClassifier', + backbone=dict( + frozen_stages=2, + type='ConvNeXt', + arch='small', + drop_path_rate=0.4, + init_cfg=dict( + type='Pretrained', + checkpoint= + 'https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth', + prefix='backbone')), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=768, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + init_cfg=None), + init_cfg=dict( + type='TruncNormal', layer=['Conv2d', 'Linear'], std=0.02, bias=0.0), + train_cfg=None) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/IF80w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/if-dpmsolver++-50-20w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/cc1m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/if-dpmsolver++-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/if-dpmsolver++-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/convnext_small_4xb256_IF_1m_lr4e-4_aug_5e-1' diff --git a/convnext_small_4xb256_IF_1m_lr4e-4_aug_5e-1/20230606_051640/vis_data/events.out.tfevents.1685999814.SH-IDC1-10-140-24-124.251653.0 b/convnext_small_4xb256_IF_1m_lr4e-4_aug_5e-1/20230606_051640/vis_data/events.out.tfevents.1685999814.SH-IDC1-10-140-24-124.251653.0 new file mode 100644 index 0000000000000000000000000000000000000000..8da2a9f3162feecb6a769c0e00ff4d97a738bf26 --- /dev/null +++ b/convnext_small_4xb256_IF_1m_lr4e-4_aug_5e-1/20230606_051640/vis_data/events.out.tfevents.1685999814.SH-IDC1-10-140-24-124.251653.0 @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:041761322f2bd950b55158a3f7bf6d6c52102638953755640ddf5be849a83752 +size 1951416 diff --git a/convnext_small_4xb256_IF_1m_lr4e-4_aug_5e-1/20230606_051640/vis_data/scalars.json b/convnext_small_4xb256_IF_1m_lr4e-4_aug_5e-1/20230606_051640/vis_data/scalars.json new file mode 100644 index 0000000000000000000000000000000000000000..e53dabfd8305948b36969de30b70bcb9aeb4d3eb --- /dev/null +++ b/convnext_small_4xb256_IF_1m_lr4e-4_aug_5e-1/20230606_051640/vis_data/scalars.json @@ -0,0 +1,200 @@ +{"lr": 0.0003999756737991541, "data_time": 0.0017441987991333007, "loss": 0.09458963386714458, "time": 0.6123632192611694, "epoch": 1, "memory": 15587, "step": 100} +{"lr": 0.00039990171605141573, "data_time": 0.029685473442077635, "loss": 0.058537983894348146, "time": 0.6956503629684448, "epoch": 1, "memory": 15587, "step": 200} +{"lr": 0.00039977814310179305, "data_time": 0.0017234563827514648, "loss": 0.058321688696742056, "time": 0.6685066938400268, "epoch": 1, "memory": 15587, "step": 300} +{"lr": 0.00039960498640766363, "data_time": 0.001700282096862793, "loss": 0.05933081284165383, "time": 0.6553350687026978, "epoch": 1, "memory": 15587, "step": 400} +{"lr": 0.00039938229004870185, "data_time": 0.0017362117767333984, "loss": 0.05389338620007038, "time": 0.638532018661499, "epoch": 1, "memory": 15587, "step": 500} +{"lr": 0.00039911011071566005, "data_time": 0.0016216278076171876, "loss": 0.05744111407548189, "time": 0.602908706665039, "epoch": 1, "memory": 15587, "step": 600} +{"lr": 0.0003987885176959352, "data_time": 0.0015788078308105469, "loss": 0.05633894167840481, "time": 1.1324015617370606, "epoch": 1, "memory": 15587, "step": 700} +{"lr": 0.0003984175928559308, "data_time": 0.0017818689346313476, "loss": 0.060062802582979205, "time": 0.6036136150360107, "epoch": 1, "memory": 15587, "step": 800} +{"lr": 0.00039799743062021714, "data_time": 0.001535654067993164, "loss": 0.05464814603328705, "time": 0.5759683847427368, "epoch": 1, "memory": 15587, "step": 900} +{"lr": 0.0003975281379474946, "data_time": 0.0016119003295898438, "loss": 0.04153641015291214, "time": 0.5958358764648437, "epoch": 1, "memory": 15587, "step": 1000} +{"lr": 0.00039700983430336476, "data_time": 0.001715397834777832, "loss": 0.04811144415289163, "time": 0.6263498544692994, "epoch": 1, "memory": 15587, "step": 1100} +{"lr": 0.0003964426516299177, "data_time": 0.0016070127487182618, "loss": 0.04214367736130953, "time": 0.6444735765457154, "epoch": 1, "memory": 15587, "step": 1200} +{"lr": 0.0003958267343121465, "data_time": 0.0014572381973266602, "loss": 0.04770536925643683, "time": 0.6142401456832886, "epoch": 1, "memory": 15587, "step": 1300} +{"lr": 0.00039516223914119037, "data_time": 0.001609635353088379, "loss": 0.03325816858559847, "time": 0.5964842557907104, "epoch": 1, "memory": 15587, "step": 1400} +{"lr": 0.0003944493352744212, "data_time": 0.0015526533126831055, "loss": 0.039781056623905894, "time": 0.6076323509216308, "epoch": 1, "memory": 15587, "step": 1500} +{"lr": 0.0003936882041923822, "data_time": 0.0015211105346679688, "loss": 0.04655163232237101, "time": 0.6154497861862183, "epoch": 1, "memory": 15587, "step": 1600} +{"lr": 0.0003928790396525886, "data_time": 0.0016469955444335938, "loss": 0.0350088344886899, "time": 0.6293362379074097, "epoch": 1, "memory": 15587, "step": 1700} +{"lr": 0.00039202204764020483, "data_time": 0.001549100875854492, "loss": 0.034160731546580794, "time": 1.3417204856872558, "epoch": 1, "memory": 15587, "step": 1800} +{"lr": 0.00039111744631560716, "data_time": 0.0015239238739013672, "loss": 0.04124070554971695, "time": 0.6438556909561157, "epoch": 1, "memory": 15587, "step": 1900} +{"accuracy/top1": 80.1259536743164, "data_time": 0.2369171668743265, "time": 0.700283930219453, "step": 1} +{"lr": 0.0003894811013774212, "data_time": 0.0016692638397216796, "loss": 0.03209449835121632, "time": 0.6675132513046265, "epoch": 2, "memory": 15587, "step": 2069} +{"lr": 0.00038844960758563737, "data_time": 0.001808333396911621, "loss": 0.026527286507189274, "time": 0.6054823875427247, "epoch": 2, "memory": 15585, "step": 2169} +{"lr": 0.00038737141390069504, "data_time": 0.0015696048736572265, "loss": 0.03775574136525393, "time": 0.6006609439849854, "epoch": 2, "memory": 15585, "step": 2269} +{"lr": 0.0003862467947932271, "data_time": 0.0015077829360961915, "loss": 0.03677110578864813, "time": 0.6380847215652465, "epoch": 2, "memory": 15585, "step": 2369} +{"lr": 0.000385076036552165, "data_time": 0.0015467166900634765, "loss": 0.03364796955138445, "time": 0.5798214197158813, "epoch": 2, "memory": 15585, "step": 2469} +{"lr": 0.00038385943721186026, "data_time": 0.002039456367492676, "loss": 0.03959393538534641, "time": 0.6175883054733277, "epoch": 2, "memory": 15585, "step": 2569} +{"lr": 0.0003825973064762139, "data_time": 0.0014548063278198241, "loss": 0.02873118189163506, "time": 0.6704680681228637, "epoch": 2, "memory": 15585, "step": 2669} +{"lr": 0.000381289965639839, "data_time": 0.001670694351196289, "loss": 0.03372421395033598, "time": 0.6355009078979492, "epoch": 2, "memory": 15585, "step": 2769} +{"lr": 0.00037993774750626663, "data_time": 0.001581096649169922, "loss": 0.03461470026522875, "time": 0.6274290561676026, "epoch": 2, "memory": 15585, "step": 2869} +{"lr": 0.00037854099630322966, "data_time": 0.0015678882598876953, "loss": 0.04298805464059115, "time": 0.6427930593490601, "epoch": 2, "memory": 15585, "step": 2969} +{"lr": 0.00037710006759503174, "data_time": 0.0014120817184448242, "loss": 0.03247109018266201, "time": 0.6416108846664429, "epoch": 2, "memory": 15585, "step": 3069} +{"lr": 0.00037561532819203284, "data_time": 0.0016913652420043946, "loss": 0.032137765083462, "time": 0.6324577569961548, "epoch": 2, "memory": 15585, "step": 3169} +{"lr": 0.00037408715605727376, "data_time": 0.0018703937530517578, "loss": 0.0294894490391016, "time": 0.6005995035171509, "epoch": 2, "memory": 15585, "step": 3269} +{"lr": 0.00037251594021025933, "data_time": 0.0016335248947143555, "loss": 0.02664182074368, "time": 0.6230206966400147, "epoch": 2, "memory": 15585, "step": 3369} +{"lr": 0.00037090208062792615, "data_time": 0.001602005958557129, "loss": 0.023595809610560535, "time": 0.5774687051773071, "epoch": 2, "memory": 15585, "step": 3469} +{"lr": 0.00036924598814282443, "data_time": 0.0014952659606933595, "loss": 0.029191960766911506, "time": 0.5890163660049439, "epoch": 2, "memory": 15585, "step": 3569} +{"lr": 0.00036754808433853343, "data_time": 0.001488971710205078, "loss": 0.027995210606604815, "time": 0.5980283260345459, "epoch": 2, "memory": 15585, "step": 3669} +{"lr": 0.00036580880144233954, "data_time": 0.0015398025512695312, "loss": 0.03276628088206053, "time": 0.6006958246231079, "epoch": 2, "memory": 15585, "step": 3769} +{"lr": 0.0003640285822152096, "data_time": 0.0015473365783691406, "loss": 0.03431772254407406, "time": 0.5462408304214478, "epoch": 2, "memory": 15585, "step": 3869} +{"accuracy/top1": 91.78873443603516, "data_time": 0.2112113316853841, "time": 0.5998448212941487, "step": 2} +{"lr": 0.0003609282333974528, "data_time": 0.03719151020050049, "loss": 0.024518429953604936, "time": 0.5542458295822144, "epoch": 3, "memory": 15585, "step": 4038} +{"lr": 0.0003590401733608027, "data_time": 0.08352704048156738, "loss": 0.03717135079205036, "time": 0.6016617774963379, "epoch": 3, "memory": 15585, "step": 4138} +{"lr": 0.00035711290005063985, "data_time": 0.05858302116394043, "loss": 0.027805393422022463, "time": 0.6124847650527954, "epoch": 3, "memory": 15585, "step": 4238} +{"lr": 0.0003551469040837546, "data_time": 0.0022240638732910155, "loss": 0.029554964695125818, "time": 0.6046499967575073, "epoch": 3, "memory": 15585, "step": 4338} +{"lr": 0.00035314268593437953, "data_time": 0.001641535758972168, "loss": 0.03180052498355508, "time": 0.5970683097839355, "epoch": 3, "memory": 15585, "step": 4438} +{"lr": 0.00035110075580678697, "data_time": 0.0016754627227783202, "loss": 0.028151841554790735, "time": 0.6520636796951294, "epoch": 3, "memory": 15585, "step": 4538} +{"lr": 0.00034902163350540724, "data_time": 0.0019218206405639648, "loss": 0.023087999410927297, "time": 0.5890148162841797, "epoch": 3, "memory": 15585, "step": 4638} +{"lr": 0.0003469058483025052, "data_time": 0.0017933368682861329, "loss": 0.026231523649767042, "time": 0.5728699684143066, "epoch": 3, "memory": 15585, "step": 4738} +{"lr": 0.0003447539388034461, "data_time": 0.001554727554321289, "loss": 0.02800388978794217, "time": 0.6035315036773682, "epoch": 3, "memory": 15585, "step": 4838} +{"lr": 0.0003425664528095857, "data_time": 0.0018976449966430664, "loss": 0.021152870915830135, "time": 0.6211369514465332, "epoch": 3, "memory": 15585, "step": 4938} +{"lr": 0.00034034394717881845, "data_time": 0.001685929298400879, "loss": 0.026131332106888294, "time": 0.5432857990264892, "epoch": 3, "memory": 15585, "step": 5038} +{"lr": 0.00033808698768382034, "data_time": 0.00146176815032959, "loss": 0.03089748863130808, "time": 0.5564919710159302, "epoch": 3, "memory": 15585, "step": 5138} +{"lr": 0.00033579614886802443, "data_time": 0.001519179344177246, "loss": 0.04191914033144713, "time": 0.6116936445236206, "epoch": 3, "memory": 15585, "step": 5238} +{"lr": 0.00033347201389936123, "data_time": 0.001765871047973633, "loss": 0.02352065369486809, "time": 0.5813561201095581, "epoch": 3, "memory": 15585, "step": 5338} +{"lr": 0.0003311151744218034, "data_time": 0.0015958070755004883, "loss": 0.030639497749507427, "time": 0.5732959032058715, "epoch": 3, "memory": 15585, "step": 5438} +{"lr": 0.0003287262304047557, "data_time": 0.0017245292663574218, "loss": 0.02399143176153302, "time": 0.6123759508132934, "epoch": 3, "memory": 15585, "step": 5538} +{"lr": 0.0003263057899903226, "data_time": 0.0014634370803833009, "loss": 0.023891198588535188, "time": 0.5682729482650757, "epoch": 3, "memory": 15585, "step": 5638} +{"lr": 0.00032385446933849547, "data_time": 0.0016894817352294921, "loss": 0.0306385925039649, "time": 0.6199266910552979, "epoch": 3, "memory": 15585, "step": 5738} +{"lr": 0.0003213728924703014, "data_time": 0.0017627477645874023, "loss": 0.02045824136584997, "time": 0.6045604705810547, "epoch": 3, "memory": 15585, "step": 5838} +{"accuracy/top1": 95.73084259033203, "data_time": 0.1199626604715983, "time": 0.5050847053527832, "step": 3} +{"lr": 0.00031711202351665616, "data_time": 0.0020610809326171873, "loss": 0.023080874979496003, "time": 0.5588046073913574, "epoch": 4, "memory": 15585, "step": 6007} +{"lr": 0.0003145522137064149, "data_time": 0.0015259742736816405, "loss": 0.02575897970236838, "time": 0.6027894258499146, "epoch": 4, "memory": 15585, "step": 6107} +{"lr": 0.0003119645157107334, "data_time": 0.0016039133071899414, "loss": 0.036812640447169545, "time": 0.6135179996490479, "epoch": 4, "memory": 15585, "step": 6207} +{"lr": 0.00030934958826756915, "data_time": 0.0015448331832885742, "loss": 0.017345844767987728, "time": 0.5793030261993408, "epoch": 4, "memory": 15585, "step": 6307} +{"lr": 0.0003067080970465515, "data_time": 0.001685023307800293, "loss": 0.02414605803787708, "time": 0.579094648361206, "epoch": 4, "memory": 15585, "step": 6407} +{"lr": 0.00030404071447952345, "data_time": 0.001581573486328125, "loss": 0.01597123914398253, "time": 0.6501397848129272, "epoch": 4, "memory": 15585, "step": 6507} +{"lr": 0.0003013481195893653, "data_time": 0.0017234325408935548, "loss": 0.024644845677539707, "time": 0.5737666130065918, "epoch": 4, "memory": 15585, "step": 6607} +{"lr": 0.0002986309978171385, "data_time": 0.0016404390335083008, "loss": 0.028770623495802283, "time": 0.6156988859176635, "epoch": 4, "memory": 15585, "step": 6707} +{"lr": 0.0002958900408475955, "data_time": 0.0016281843185424806, "loss": 0.019331029430031777, "time": 0.5952183485031128, "epoch": 4, "memory": 15585, "step": 6807} +{"lr": 0.0002931259464331015, "data_time": 0.001688098907470703, "loss": 0.023258120100945235, "time": 0.6249657869338989, "epoch": 4, "memory": 15585, "step": 6907} +{"lr": 0.00029033941821601176, "data_time": 0.0020334482192993163, "loss": 0.029206813685595988, "time": 0.5614066839218139, "epoch": 4, "memory": 15585, "step": 7007} +{"lr": 0.0002875311655495464, "data_time": 0.0014782428741455078, "loss": 0.024458643654361366, "time": 0.556343412399292, "epoch": 4, "memory": 15585, "step": 7107} +{"lr": 0.0002847019033172168, "data_time": 0.0016544580459594727, "loss": 0.021977328462526202, "time": 0.5464712381362915, "epoch": 4, "memory": 15585, "step": 7207} +{"lr": 0.0002818523517508384, "data_time": 0.001614999771118164, "loss": 0.02037540073506534, "time": 0.595465087890625, "epoch": 4, "memory": 15585, "step": 7307} +{"lr": 0.00027898323624718594, "data_time": 0.0017370462417602539, "loss": 0.022024381440132856, "time": 0.585851502418518, "epoch": 4, "memory": 15585, "step": 7407} +{"lr": 0.000276095287183333, "data_time": 0.0014989852905273437, "loss": 0.023259676760062575, "time": 0.6028532743453979, "epoch": 4, "memory": 15585, "step": 7507} +{"lr": 0.00027318923973072227, "data_time": 0.0017900705337524415, "loss": 0.028592138458043335, "time": 0.6054024696350098, "epoch": 4, "memory": 15585, "step": 7607} +{"lr": 0.00027026583366801727, "data_time": 0.0016146659851074218, "loss": 0.027045744750648738, "time": 0.576863956451416, "epoch": 4, "memory": 15585, "step": 7707} +{"lr": 0.0002673258131927808, "data_time": 0.001558542251586914, "loss": 0.02875189818441868, "time": 0.5726858139038086, "epoch": 4, "memory": 15585, "step": 7807} +{"accuracy/top1": 97.62012481689453, "data_time": 0.22566306591033936, "time": 0.6116198460261028, "step": 4} +{"lr": 0.0002623215076436499, "data_time": 0.04466564655303955, "loss": 0.028714340645819902, "time": 0.58972647190094, "epoch": 5, "memory": 15585, "step": 7976} +{"lr": 0.00025934052007928784, "data_time": 0.029674386978149413, "loss": 0.028971432987600566, "time": 0.5518029928207397, "epoch": 5, "memory": 15585, "step": 8076} +{"lr": 0.0002563456993074567, "data_time": 0.0015370607376098632, "loss": 0.028553877910599112, "time": 0.6341614961624146, "epoch": 5, "memory": 15585, "step": 8176} +{"lr": 0.00025333780770542177, "data_time": 0.0015366315841674806, "loss": 0.019479702226817608, "time": 0.5838659048080445, "epoch": 5, "memory": 15585, "step": 8276} +{"lr": 0.00025031761097782693, "data_time": 0.0017219066619873046, "loss": 0.02388954218477011, "time": 0.5737927675247192, "epoch": 5, "memory": 15585, "step": 8376} +{"lr": 0.0002472858779617731, "data_time": 0.001516866683959961, "loss": 0.022251126868650318, "time": 0.5874070167541504, "epoch": 5, "memory": 15585, "step": 8476} +{"lr": 0.0002442433804310988, "data_time": 0.0016100645065307618, "loss": 0.030083376727998257, "time": 0.6104545593261719, "epoch": 5, "memory": 15585, "step": 8576} +{"lr": 0.00024119089289991304, "data_time": 0.0015225887298583984, "loss": 0.01802371027879417, "time": 0.5405940294265748, "epoch": 5, "memory": 15585, "step": 8676} +{"lr": 0.00023812919242543494, "data_time": 0.0016010522842407227, "loss": 0.022229759022593498, "time": 0.5434931993484498, "epoch": 5, "memory": 15585, "step": 8776} +{"lr": 0.00023505905841017684, "data_time": 0.0015564203262329102, "loss": 0.023329756082966925, "time": 0.555145525932312, "epoch": 5, "memory": 15585, "step": 8876} +{"lr": 0.00023198127240353698, "data_time": 0.0016033649444580078, "loss": 0.02470110231079161, "time": 0.5642518520355224, "epoch": 5, "memory": 15585, "step": 8976} +{"lr": 0.00022889661790284532, "data_time": 0.0014160871505737305, "loss": 0.02083730003796518, "time": 0.6121579885482789, "epoch": 5, "memory": 15585, "step": 9076} +{"lr": 0.000225805880153911, "data_time": 0.0015762567520141602, "loss": 0.0174110543448478, "time": 0.5741606473922729, "epoch": 5, "memory": 15585, "step": 9176} +{"lr": 0.00022270984595112745, "data_time": 0.001848888397216797, "loss": 0.02076743058860302, "time": 0.5884270191192627, "epoch": 5, "memory": 15585, "step": 9276} +{"lr": 0.00021960930343718077, "data_time": 0.0014249563217163086, "loss": 0.0137361329048872, "time": 0.5783840894699097, "epoch": 5, "memory": 15585, "step": 9376} +{"lr": 0.00021650504190241605, "data_time": 0.0017305612564086914, "loss": 0.025541602773591875, "time": 0.5636526584625244, "epoch": 5, "memory": 15585, "step": 9476} +{"lr": 0.00021339785158391245, "data_time": 0.0019342660903930663, "loss": 0.02035396988503635, "time": 0.5900669813156127, "epoch": 5, "memory": 15585, "step": 9576} +{"lr": 0.00021028852346431534, "data_time": 0.0016449451446533202, "loss": 0.01910709673538804, "time": 0.5614126443862915, "epoch": 5, "memory": 15585, "step": 9676} +{"lr": 0.00020717784907047976, "data_time": 0.0016514778137207032, "loss": 0.021278704004362225, "time": 0.5583648204803466, "epoch": 5, "memory": 15585, "step": 9776} +{"accuracy/top1": 98.43081665039062, "data_time": 0.21341134707132975, "time": 0.6042571067810059, "step": 5} +{"lr": 0.0002019199632200586, "data_time": 0.0016721010208129883, "loss": 0.01492706169374287, "time": 0.6085896968841553, "epoch": 6, "memory": 15585, "step": 9945} +{"lr": 0.00019880959773413697, "data_time": 0.0015503883361816407, "loss": 0.017514523956924676, "time": 0.5879445791244506, "epoch": 6, "memory": 15585, "step": 10045} +{"lr": 0.00019570080810944632, "data_time": 0.0015748023986816406, "loss": 0.01872837315313518, "time": 0.6430665016174316, "epoch": 6, "memory": 15585, "step": 10145} +{"lr": 0.00019259438573575975, "data_time": 0.001423478126525879, "loss": 0.019388327561318874, "time": 0.5860364437103271, "epoch": 6, "memory": 15585, "step": 10245} +{"lr": 0.00018949112140023044, "data_time": 0.0015466451644897462, "loss": 0.027449066191911696, "time": 0.5448642015457154, "epoch": 6, "memory": 15585, "step": 10345} +{"lr": 0.00018639180508608498, "data_time": 0.0017944097518920899, "loss": 0.018055544793605806, "time": 0.5434005975723266, "epoch": 6, "memory": 15585, "step": 10445} +{"lr": 0.00018329722577152123, "data_time": 0.0015152931213378907, "loss": 0.012535664858296514, "time": 0.5646077156066894, "epoch": 6, "memory": 15585, "step": 10545} +{"lr": 0.0001802081712288607, "data_time": 0.0017061710357666015, "loss": 0.014948234194889665, "time": 0.5422788381576538, "epoch": 6, "memory": 15585, "step": 10645} +{"lr": 0.00017712542782401123, "data_time": 0.0015752077102661132, "loss": 0.024940438149496914, "time": 0.561505913734436, "epoch": 6, "memory": 15585, "step": 10745} +{"lr": 0.00017404978031628336, "data_time": 0.0017782926559448242, "loss": 0.013869933737441897, "time": 0.5385247468948364, "epoch": 6, "memory": 15585, "step": 10845} +{"lr": 0.00017098201165861962, "data_time": 0.001707601547241211, "loss": 0.015723573695868253, "time": 0.5790489912033081, "epoch": 6, "memory": 15585, "step": 10945} +{"lr": 0.00016792290279828072, "data_time": 0.0016601085662841797, "loss": 0.019538279995322226, "time": 0.5651997089385986, "epoch": 6, "memory": 15585, "step": 11045} +{"lr": 0.00016487323247804387, "data_time": 0.0014456510543823242, "loss": 0.02261862359009683, "time": 0.5819118976593017, "epoch": 6, "memory": 15585, "step": 11145} +{"lr": 0.00016183377703796263, "data_time": 0.002166295051574707, "loss": 0.01894620726816356, "time": 0.5798263788223267, "epoch": 6, "memory": 15585, "step": 11245} +{"lr": 0.00015880531021773676, "data_time": 0.001630687713623047, "loss": 0.015934361144900322, "time": 0.5687504529953002, "epoch": 6, "memory": 15585, "step": 11345} +{"lr": 0.00015578860295974566, "data_time": 0.0015659809112548828, "loss": 0.020451073767617345, "time": 0.5816017389297485, "epoch": 6, "memory": 15585, "step": 11445} +{"lr": 0.0001527844232127924, "data_time": 0.0018509864807128907, "loss": 0.016789604676887393, "time": 0.6481788158416748, "epoch": 6, "memory": 15585, "step": 11545} +{"lr": 0.000149793535736612, "data_time": 0.0017035961151123046, "loss": 0.027278015110641717, "time": 0.6013190984725952, "epoch": 6, "memory": 15585, "step": 11645} +{"lr": 0.00014681670190718876, "data_time": 0.001486515998840332, "loss": 0.020000103628262877, "time": 0.579273796081543, "epoch": 6, "memory": 15585, "step": 11745} +{"accuracy/top1": 98.83789825439453, "data_time": 0.2203480323155721, "time": 0.6101419369379679, "step": 6} +{"lr": 0.00014181991425636192, "data_time": 0.0015768051147460938, "loss": 0.013803625106811523, "time": 0.5759463548660279, "epoch": 7, "memory": 15585, "step": 11914} +{"lr": 0.00013888463509383338, "data_time": 0.0016977310180664063, "loss": 0.016744786640629172, "time": 0.5868659257888794, "epoch": 7, "memory": 15585, "step": 12014} +{"lr": 0.000135966186604959, "data_time": 0.0015201330184936523, "loss": 0.019931176863610744, "time": 0.5900272369384766, "epoch": 7, "memory": 15585, "step": 12114} +{"lr": 0.0001330653117252752, "data_time": 0.0016937494277954102, "loss": 0.013702026475220919, "time": 0.6056107759475708, "epoch": 7, "memory": 15585, "step": 12214} +{"lr": 0.00013018274891668863, "data_time": 0.0017450809478759765, "loss": 0.015689974883571267, "time": 0.5967775821685791, "epoch": 7, "memory": 15585, "step": 12314} +{"lr": 0.000127319231979489, "data_time": 0.001546025276184082, "loss": 0.022085619112476706, "time": 0.5849074840545654, "epoch": 7, "memory": 15585, "step": 12414} +{"lr": 0.0001244754898655487, "data_time": 0.0016199588775634766, "loss": 0.0158816359937191, "time": 0.618298864364624, "epoch": 7, "memory": 15585, "step": 12514} +{"lr": 0.0001216522464927582, "data_time": 0.0019073724746704102, "loss": 0.014288743631914258, "time": 0.5871162652969361, "epoch": 7, "memory": 15585, "step": 12614} +{"lr": 0.00011885022056074007, "data_time": 0.001526618003845215, "loss": 0.014623307110741734, "time": 0.5565303802490235, "epoch": 7, "memory": 15585, "step": 12714} +{"lr": 0.00011607012536789545, "data_time": 0.0016824960708618163, "loss": 0.01136747356504202, "time": 0.7399872303009033, "epoch": 7, "memory": 15585, "step": 12814} +{"lr": 0.00011331266862982092, "data_time": 0.0021506309509277343, "loss": 0.01907173115760088, "time": 0.5873501777648926, "epoch": 7, "memory": 15585, "step": 12914} +{"lr": 0.00011057855229914932, "data_time": 0.001657724380493164, "loss": 0.018510606605559587, "time": 0.5710991859436035, "epoch": 7, "memory": 15585, "step": 13014} +{"lr": 0.0001078684723868573, "data_time": 0.0015811681747436523, "loss": 0.020667529525235297, "time": 0.6221235036849976, "epoch": 7, "memory": 15585, "step": 13114} +{"lr": 0.00010518311878508484, "data_time": 0.001685357093811035, "loss": 0.013659366965293884, "time": 0.5702636480331421, "epoch": 7, "memory": 15585, "step": 13214} +{"lr": 0.00010252317509151276, "data_time": 0.001720738410949707, "loss": 0.01599430530332029, "time": 0.5902864694595337, "epoch": 7, "memory": 15585, "step": 13314} +{"lr": 9.98893184353425e-05, "data_time": 0.0015124082565307617, "loss": 0.009631693596020341, "time": 0.5261112451553345, "epoch": 7, "memory": 15585, "step": 13414} +{"lr": 9.728221930492262e-05, "data_time": 0.0018680810928344727, "loss": 0.01863910616375506, "time": 0.5468105792999267, "epoch": 7, "memory": 15585, "step": 13514} +{"lr": 9.470254137706575e-05, "data_time": 0.001739501953125, "loss": 0.008839649660512806, "time": 0.5772305727005005, "epoch": 7, "memory": 15585, "step": 13614} +{"lr": 9.215094134809975e-05, "data_time": 0.001617121696472168, "loss": 0.01457508341409266, "time": 0.5743704080581665, "epoch": 7, "memory": 15585, "step": 13714} +{"accuracy/top1": 99.19975280761719, "data_time": 0.25693716208140055, "time": 0.6450648625691732, "step": 7} +{"lr": 8.790437228679661e-05, "data_time": 0.0016317367553710938, "loss": 0.015079016098752618, "time": 0.5268658876419068, "epoch": 8, "memory": 15585, "step": 13883} +{"lr": 8.543150502338196e-05, "data_time": 0.0015479564666748048, "loss": 0.014931700145825743, "time": 0.6385215282440185, "epoch": 8, "memory": 15585, "step": 13983} +{"lr": 8.298907574244125e-05, "data_time": 0.0018334150314331054, "loss": 0.012764481781050562, "time": 0.6127574443817139, "epoch": 8, "memory": 15585, "step": 14083} +{"lr": 8.057770620156745e-05, "data_time": 0.0016407489776611329, "loss": 0.010467389039695263, "time": 0.5706641197204589, "epoch": 8, "memory": 15585, "step": 14183} +{"lr": 7.819801025162347e-05, "data_time": 0.0015373468399047852, "loss": 0.017235457338392733, "time": 0.5328397035598755, "epoch": 8, "memory": 15585, "step": 14283} +{"lr": 7.585059368047708e-05, "data_time": 0.001691913604736328, "loss": 0.013922725338488817, "time": 0.5547436714172364, "epoch": 8, "memory": 15585, "step": 14383} +{"lr": 7.353605405878812e-05, "data_time": 0.0014408588409423827, "loss": 0.011247835727408529, "time": 0.5884881019592285, "epoch": 8, "memory": 15585, "step": 14483} +{"lr": 7.125498058788827e-05, "data_time": 0.0016216516494750976, "loss": 0.014739425061270595, "time": 0.5894668340682984, "epoch": 8, "memory": 15585, "step": 14583} +{"lr": 6.900795394979036e-05, "data_time": 0.001628541946411133, "loss": 0.014728918811306358, "time": 0.6016840457916259, "epoch": 8, "memory": 15585, "step": 14683} +{"lr": 6.679554615936711e-05, "data_time": 0.001741790771484375, "loss": 0.0137552491389215, "time": 0.541572904586792, "epoch": 8, "memory": 15585, "step": 14783} +{"lr": 6.461832041873565e-05, "data_time": 0.0016823768615722655, "loss": 0.010548897786065936, "time": 0.6130940914154053, "epoch": 8, "memory": 15585, "step": 14883} +{"lr": 6.247683097388664e-05, "data_time": 0.001662135124206543, "loss": 0.012707919208332896, "time": 0.5806445360183716, "epoch": 8, "memory": 15585, "step": 14983} +{"lr": 6.037162297359161e-05, "data_time": 0.0016053438186645508, "loss": 0.016017212765291333, "time": 0.5783066034317017, "epoch": 8, "memory": 15585, "step": 15083} +{"lr": 5.830323233062794e-05, "data_time": 0.0015491962432861329, "loss": 0.011485350038856267, "time": 0.5523838520050048, "epoch": 8, "memory": 15585, "step": 15183} +{"lr": 5.6272185585353725e-05, "data_time": 0.0015683889389038086, "loss": 0.009434764878824354, "time": 0.5469341039657593, "epoch": 8, "memory": 15585, "step": 15283} +{"lr": 5.4278999771668655e-05, "data_time": 0.0015121936798095704, "loss": 0.012850778782740236, "time": 0.5985788106918335, "epoch": 8, "memory": 15585, "step": 15383} +{"lr": 5.2324182285395304e-05, "data_time": 0.001495218276977539, "loss": 0.011221066955477, "time": 0.7413820266723633, "epoch": 8, "memory": 15585, "step": 15483} +{"lr": 5.0408230755113716e-05, "data_time": 0.0017287254333496094, "loss": 0.015580749697983265, "time": 0.564391827583313, "epoch": 8, "memory": 15585, "step": 15583} +{"lr": 4.853163291548285e-05, "data_time": 0.0016614913940429688, "loss": 0.011525907646864652, "time": 0.5333415746688843, "epoch": 8, "memory": 15585, "step": 15683} +{"accuracy/top1": 99.35284423828125, "data_time": 0.22713043689727783, "time": 0.6159555673599243, "step": 8} +{"lr": 4.5450966211010106e-05, "data_time": 0.009714245796203613, "loss": 0.014958574995398521, "time": 0.5427632331848145, "epoch": 9, "memory": 15585, "step": 15852} +{"lr": 4.368257232393176e-05, "data_time": 0.0017705202102661134, "loss": 0.014493309147655963, "time": 0.5537340879440308, "epoch": 9, "memory": 15585, "step": 15952} +{"lr": 4.195524424634855e-05, "data_time": 0.0018399953842163086, "loss": 0.01762465056963265, "time": 0.6047696113586426, "epoch": 9, "memory": 15585, "step": 16052} +{"lr": 4.0269421695945e-05, "data_time": 0.0016439437866210937, "loss": 0.010192060191184283, "time": 0.553926968574524, "epoch": 9, "memory": 15585, "step": 16152} +{"lr": 3.862553382454128e-05, "data_time": 0.0016122817993164062, "loss": 0.014994328608736396, "time": 0.5768372297286988, "epoch": 9, "memory": 15585, "step": 16252} +{"lr": 3.702399910884571e-05, "data_time": 0.0017409086227416991, "loss": 0.01618064888752997, "time": 0.5637802839279175, "epoch": 9, "memory": 15585, "step": 16352} +{"lr": 3.5465225243925684e-05, "data_time": 0.0018478631973266602, "loss": 0.013086614571511746, "time": 0.5510546207427979, "epoch": 9, "memory": 15585, "step": 16452} +{"lr": 3.394960903942234e-05, "data_time": 0.0015231609344482423, "loss": 0.015384511649608612, "time": 0.5371349334716797, "epoch": 9, "memory": 15585, "step": 16552} +{"lr": 3.247753631853657e-05, "data_time": 0.001718449592590332, "loss": 0.007045535370707512, "time": 0.5488334655761719, "epoch": 9, "memory": 15585, "step": 16652} +{"lr": 3.104938181981207e-05, "data_time": 0.0016681909561157226, "loss": 0.01306831338442862, "time": 0.6143965721130371, "epoch": 9, "memory": 15585, "step": 16752} +{"lr": 2.966550910173992e-05, "data_time": 0.001856088638305664, "loss": 0.012065926007926464, "time": 0.5918649911880494, "epoch": 9, "memory": 15585, "step": 16852} +{"lr": 2.8326270450208946e-05, "data_time": 0.0017817497253417968, "loss": 0.011343830358237028, "time": 0.5685947656631469, "epoch": 9, "memory": 15585, "step": 16952} +{"lr": 2.7032006788826352e-05, "data_time": 0.0018084049224853516, "loss": 0.013476584572345018, "time": 0.5744309186935425, "epoch": 9, "memory": 15585, "step": 17052} +{"lr": 2.5783047592130032e-05, "data_time": 0.0016757011413574218, "loss": 0.01341892620548606, "time": 0.5823689699172974, "epoch": 9, "memory": 15585, "step": 17152} +{"lr": 2.4579710801716043e-05, "data_time": 0.0016843557357788086, "loss": 0.014690388226881623, "time": 0.5800591468811035, "epoch": 9, "memory": 15585, "step": 17252} +{"lr": 2.3422302745301813e-05, "data_time": 0.0016840696334838867, "loss": 0.01572784800082445, "time": 0.5649601221084595, "epoch": 9, "memory": 15585, "step": 17352} +{"lr": 2.2311118058745624e-05, "data_time": 0.0015390634536743164, "loss": 0.01409651730209589, "time": 0.5444443702697754, "epoch": 9, "memory": 15585, "step": 17452} +{"lr": 2.1246439611042613e-05, "data_time": 0.0016150236129760741, "loss": 0.011113508930429816, "time": 0.5491689443588257, "epoch": 9, "memory": 15585, "step": 17552} +{"lr": 2.0228538432315955e-05, "data_time": 0.0018887996673583984, "loss": 0.008217486878857016, "time": 0.5413547277450561, "epoch": 9, "memory": 15585, "step": 17652} +{"accuracy/top1": 99.42938995361328, "data_time": 0.25013980865478513, "time": 0.6392630577087403, "step": 9} +{"lr": 1.861533120597442e-05, "data_time": 0.10335335731506348, "loss": 0.009815009543672204, "time": 0.6204253435134888, "epoch": 10, "memory": 15585, "step": 17821} +{"lr": 1.7724513410024274e-05, "data_time": 0.016815757751464842, "loss": 0.014071066724136471, "time": 0.5341600179672241, "epoch": 10, "memory": 15585, "step": 17921} +{"lr": 1.6881369443034795e-05, "data_time": 0.001433730125427246, "loss": 0.00765361082740128, "time": 0.5726219892501831, "epoch": 10, "memory": 15585, "step": 18021} +{"lr": 1.6086113940151352e-05, "data_time": 0.05248734951019287, "loss": 0.016640963777899743, "time": 0.5675643444061279, "epoch": 10, "memory": 15585, "step": 18121} +{"lr": 1.5338949345780815e-05, "data_time": 0.09957523345947265, "loss": 0.013072047475725412, "time": 0.615776252746582, "epoch": 10, "memory": 15585, "step": 18221} +{"lr": 1.4640065862056652e-05, "data_time": 0.044413280487060544, "loss": 0.009331522695720195, "time": 0.5693021535873413, "epoch": 10, "memory": 15585, "step": 18321} +{"lr": 1.3989641400419702e-05, "data_time": 0.024750924110412596, "loss": 0.022307624062523245, "time": 0.5392184972763061, "epoch": 10, "memory": 15585, "step": 18421} +{"lr": 1.33878415363284e-05, "data_time": 0.055040931701660155, "loss": 0.010318919736891984, "time": 0.6149203062057496, "epoch": 10, "memory": 15585, "step": 18521} +{"lr": 1.2834819467108807e-05, "data_time": 0.0015883684158325196, "loss": 0.009763891529291868, "time": 0.5793373346328735, "epoch": 10, "memory": 15585, "step": 18621} +{"lr": 1.2330715972956002e-05, "data_time": 0.0015215873718261719, "loss": 0.016147507447749378, "time": 0.5573807716369629, "epoch": 10, "memory": 15585, "step": 18721} +{"lr": 1.187565938109641e-05, "data_time": 0.001622319221496582, "loss": 0.010331430146470666, "time": 0.9885760307312011, "epoch": 10, "memory": 15585, "step": 18821} +{"lr": 1.1469765533119946e-05, "data_time": 0.001847052574157715, "loss": 0.014957475429400801, "time": 0.56817147731781, "epoch": 10, "memory": 15585, "step": 18921} +{"lr": 1.1113137755490966e-05, "data_time": 0.0016748428344726563, "loss": 0.012262860126793385, "time": 0.5546363830566406, "epoch": 10, "memory": 15585, "step": 19021} +{"lr": 1.0805866833244799e-05, "data_time": 0.0017088890075683595, "loss": 0.008473808877170086, "time": 0.5907685041427613, "epoch": 10, "memory": 15585, "step": 19121} +{"lr": 1.0548030986877194e-05, "data_time": 0.0016157388687133788, "loss": 0.011519062705338002, "time": 0.5833476781845093, "epoch": 10, "memory": 15585, "step": 19221} +{"lr": 1.0339695852431895e-05, "data_time": 0.0015515327453613282, "loss": 0.011657098587602378, "time": 0.5510348320007324, "epoch": 10, "memory": 15585, "step": 19321} +{"lr": 1.0180914464792136e-05, "data_time": 0.0019409894943237306, "loss": 0.01063668834976852, "time": 0.5503006219863892, "epoch": 10, "memory": 15585, "step": 19421} +{"lr": 1.007172724417983e-05, "data_time": 0.0016666173934936524, "loss": 0.01871240623295307, "time": 0.5223334550857544, "epoch": 10, "memory": 15585, "step": 19521} +{"lr": 1.0012161985865822e-05, "data_time": 0.001567983627319336, "loss": 0.009617753140628339, "time": 0.5329965829849244, "epoch": 10, "memory": 15585, "step": 19621} +{"accuracy/top1": 99.41547393798828, "data_time": 0.2342461109161377, "time": 0.622505521774292, "step": 10} diff --git a/convnext_small_4xb256_IF_1m_lr4e-4_aug_5e-1/20230606_051640/vis_data/vis_image/0000000.png_1.png b/convnext_small_4xb256_IF_1m_lr4e-4_aug_5e-1/20230606_051640/vis_data/vis_image/0000000.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..98cd585336d810c0bce830d0a62ddffd19cdee9f Binary files /dev/null and b/convnext_small_4xb256_IF_1m_lr4e-4_aug_5e-1/20230606_051640/vis_data/vis_image/0000000.png_1.png differ diff --git a/convnext_small_4xb256_IF_1m_lr4e-4_aug_5e-1/20230606_051640/vis_data/vis_image/0000000.png_10.png b/convnext_small_4xb256_IF_1m_lr4e-4_aug_5e-1/20230606_051640/vis_data/vis_image/0000000.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..e4e1022d87040280176c5619aeb00e8e4989c904 Binary files /dev/null and b/convnext_small_4xb256_IF_1m_lr4e-4_aug_5e-1/20230606_051640/vis_data/vis_image/0000000.png_10.png differ diff --git a/convnext_small_4xb256_IF_1m_lr4e-4_aug_5e-1/20230606_051640/vis_data/vis_image/0000000.png_2.png b/convnext_small_4xb256_IF_1m_lr4e-4_aug_5e-1/20230606_051640/vis_data/vis_image/0000000.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..ed076f6075e827d0c64c4fc99d52d3bd8e0881a9 Binary files /dev/null and b/convnext_small_4xb256_IF_1m_lr4e-4_aug_5e-1/20230606_051640/vis_data/vis_image/0000000.png_2.png differ diff --git a/convnext_small_4xb256_IF_1m_lr4e-4_aug_5e-1/20230606_051640/vis_data/vis_image/0000000.png_3.png b/convnext_small_4xb256_IF_1m_lr4e-4_aug_5e-1/20230606_051640/vis_data/vis_image/0000000.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..271add509f6d07443d9ee975159de57054934900 Binary files /dev/null and b/convnext_small_4xb256_IF_1m_lr4e-4_aug_5e-1/20230606_051640/vis_data/vis_image/0000000.png_3.png differ diff --git a/convnext_small_4xb256_IF_1m_lr4e-4_aug_5e-1/20230606_051640/vis_data/vis_image/0000000.png_4.png b/convnext_small_4xb256_IF_1m_lr4e-4_aug_5e-1/20230606_051640/vis_data/vis_image/0000000.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..7097fd9b24526eeaf9c2676a4a086fa6f5cd75fb Binary files /dev/null and b/convnext_small_4xb256_IF_1m_lr4e-4_aug_5e-1/20230606_051640/vis_data/vis_image/0000000.png_4.png differ diff --git a/convnext_small_4xb256_IF_1m_lr4e-4_aug_5e-1/20230606_051640/vis_data/vis_image/0000000.png_5.png b/convnext_small_4xb256_IF_1m_lr4e-4_aug_5e-1/20230606_051640/vis_data/vis_image/0000000.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..b51672ddf099437ff151a08f7bcf2ec51e3cbbf0 Binary files /dev/null and b/convnext_small_4xb256_IF_1m_lr4e-4_aug_5e-1/20230606_051640/vis_data/vis_image/0000000.png_5.png differ diff --git a/convnext_small_4xb256_IF_1m_lr4e-4_aug_5e-1/20230606_051640/vis_data/vis_image/0000000.png_6.png b/convnext_small_4xb256_IF_1m_lr4e-4_aug_5e-1/20230606_051640/vis_data/vis_image/0000000.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..f8fa08b0a16fe4c2b26042ec6c738b653185aa4e Binary files /dev/null and b/convnext_small_4xb256_IF_1m_lr4e-4_aug_5e-1/20230606_051640/vis_data/vis_image/0000000.png_6.png differ diff --git a/convnext_small_4xb256_IF_1m_lr4e-4_aug_5e-1/20230606_051640/vis_data/vis_image/0000000.png_7.png b/convnext_small_4xb256_IF_1m_lr4e-4_aug_5e-1/20230606_051640/vis_data/vis_image/0000000.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..001dd82f0135a7b47a7c12b2923fa735b8f58e26 Binary files /dev/null and b/convnext_small_4xb256_IF_1m_lr4e-4_aug_5e-1/20230606_051640/vis_data/vis_image/0000000.png_7.png differ diff --git a/convnext_small_4xb256_IF_1m_lr4e-4_aug_5e-1/20230606_051640/vis_data/vis_image/0000000.png_8.png b/convnext_small_4xb256_IF_1m_lr4e-4_aug_5e-1/20230606_051640/vis_data/vis_image/0000000.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..0efd13c468b3ef731c59f49216a622c2dce5c465 Binary files /dev/null and b/convnext_small_4xb256_IF_1m_lr4e-4_aug_5e-1/20230606_051640/vis_data/vis_image/0000000.png_8.png differ diff --git a/convnext_small_4xb256_IF_1m_lr4e-4_aug_5e-1/20230606_051640/vis_data/vis_image/0000000.png_9.png b/convnext_small_4xb256_IF_1m_lr4e-4_aug_5e-1/20230606_051640/vis_data/vis_image/0000000.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..508706992e7bf84581439be6f2e3111bb8c84cf9 Binary files /dev/null and b/convnext_small_4xb256_IF_1m_lr4e-4_aug_5e-1/20230606_051640/vis_data/vis_image/0000000.png_9.png differ diff --git a/convnext_small_4xb256_IF_1m_lr4e-4_aug_5e-1/20230606_051640/vis_data/vis_image/000005174.jpg_1.png b/convnext_small_4xb256_IF_1m_lr4e-4_aug_5e-1/20230606_051640/vis_data/vis_image/000005174.jpg_1.png new file mode 100644 index 0000000000000000000000000000000000000000..231d4875944d0b5aa37ea11cf31e3dbd141960d2 Binary files /dev/null and b/convnext_small_4xb256_IF_1m_lr4e-4_aug_5e-1/20230606_051640/vis_data/vis_image/000005174.jpg_1.png differ diff --git a/convnext_small_4xb256_IF_1m_lr4e-4_aug_5e-1/20230606_051640/vis_data/vis_image/000005174.jpg_10.png b/convnext_small_4xb256_IF_1m_lr4e-4_aug_5e-1/20230606_051640/vis_data/vis_image/000005174.jpg_10.png new file mode 100644 index 0000000000000000000000000000000000000000..dd90b13709bff49a458f2ba6f885701c4d4388f8 Binary files /dev/null and b/convnext_small_4xb256_IF_1m_lr4e-4_aug_5e-1/20230606_051640/vis_data/vis_image/000005174.jpg_10.png differ diff --git a/convnext_small_4xb256_IF_1m_lr4e-4_aug_5e-1/20230606_051640/vis_data/vis_image/000005174.jpg_2.png b/convnext_small_4xb256_IF_1m_lr4e-4_aug_5e-1/20230606_051640/vis_data/vis_image/000005174.jpg_2.png new file mode 100644 index 0000000000000000000000000000000000000000..0f2ffd16f01a67fbfce8a2ac8668789c1273ed3d Binary files /dev/null and b/convnext_small_4xb256_IF_1m_lr4e-4_aug_5e-1/20230606_051640/vis_data/vis_image/000005174.jpg_2.png differ diff --git a/convnext_small_4xb256_IF_1m_lr4e-4_aug_5e-1/20230606_051640/vis_data/vis_image/000005174.jpg_3.png b/convnext_small_4xb256_IF_1m_lr4e-4_aug_5e-1/20230606_051640/vis_data/vis_image/000005174.jpg_3.png new file mode 100644 index 0000000000000000000000000000000000000000..fccd8fcf6267d2eb3a9157db20390518e2c32d55 Binary files /dev/null and b/convnext_small_4xb256_IF_1m_lr4e-4_aug_5e-1/20230606_051640/vis_data/vis_image/000005174.jpg_3.png differ diff --git a/convnext_small_4xb256_IF_1m_lr4e-4_aug_5e-1/20230606_051640/vis_data/vis_image/000005174.jpg_4.png b/convnext_small_4xb256_IF_1m_lr4e-4_aug_5e-1/20230606_051640/vis_data/vis_image/000005174.jpg_4.png new file mode 100644 index 0000000000000000000000000000000000000000..fac4757f1c44c672e206e4a6715d6175118fd18e Binary files /dev/null and b/convnext_small_4xb256_IF_1m_lr4e-4_aug_5e-1/20230606_051640/vis_data/vis_image/000005174.jpg_4.png differ diff --git a/convnext_small_4xb256_IF_1m_lr4e-4_aug_5e-1/20230606_051640/vis_data/vis_image/000005174.jpg_5.png b/convnext_small_4xb256_IF_1m_lr4e-4_aug_5e-1/20230606_051640/vis_data/vis_image/000005174.jpg_5.png new file mode 100644 index 0000000000000000000000000000000000000000..7b9e81e2128e454805ed331f6f963de7fa95d95c Binary files /dev/null and b/convnext_small_4xb256_IF_1m_lr4e-4_aug_5e-1/20230606_051640/vis_data/vis_image/000005174.jpg_5.png differ diff --git a/convnext_small_4xb256_IF_1m_lr4e-4_aug_5e-1/20230606_051640/vis_data/vis_image/000005174.jpg_6.png b/convnext_small_4xb256_IF_1m_lr4e-4_aug_5e-1/20230606_051640/vis_data/vis_image/000005174.jpg_6.png new file mode 100644 index 0000000000000000000000000000000000000000..e5b1da1762af7cd11e75397d89506a4cd0335626 Binary files /dev/null and b/convnext_small_4xb256_IF_1m_lr4e-4_aug_5e-1/20230606_051640/vis_data/vis_image/000005174.jpg_6.png differ diff --git a/convnext_small_4xb256_IF_1m_lr4e-4_aug_5e-1/20230606_051640/vis_data/vis_image/000005174.jpg_7.png b/convnext_small_4xb256_IF_1m_lr4e-4_aug_5e-1/20230606_051640/vis_data/vis_image/000005174.jpg_7.png new file mode 100644 index 0000000000000000000000000000000000000000..1a410631c065737501e3e3e1bec4ef09abe2051e Binary files /dev/null and b/convnext_small_4xb256_IF_1m_lr4e-4_aug_5e-1/20230606_051640/vis_data/vis_image/000005174.jpg_7.png differ diff --git a/convnext_small_4xb256_IF_1m_lr4e-4_aug_5e-1/20230606_051640/vis_data/vis_image/000005174.jpg_8.png b/convnext_small_4xb256_IF_1m_lr4e-4_aug_5e-1/20230606_051640/vis_data/vis_image/000005174.jpg_8.png new file mode 100644 index 0000000000000000000000000000000000000000..7367dd234a6c0c160da3c29236b768511b948522 Binary files /dev/null and b/convnext_small_4xb256_IF_1m_lr4e-4_aug_5e-1/20230606_051640/vis_data/vis_image/000005174.jpg_8.png differ diff --git a/convnext_small_4xb256_IF_1m_lr4e-4_aug_5e-1/20230606_051640/vis_data/vis_image/000005174.jpg_9.png b/convnext_small_4xb256_IF_1m_lr4e-4_aug_5e-1/20230606_051640/vis_data/vis_image/000005174.jpg_9.png new file mode 100644 index 0000000000000000000000000000000000000000..813b995bd3214345eb898aabd14002ad9efeb2c3 Binary files /dev/null and b/convnext_small_4xb256_IF_1m_lr4e-4_aug_5e-1/20230606_051640/vis_data/vis_image/000005174.jpg_9.png differ diff --git a/convnext_small_4xb256_IF_1m_lr4e-4_aug_5e-1/convnext_small_4xb256_IF_1m_lr4e-4_aug_5e-1.py b/convnext_small_4xb256_IF_1m_lr4e-4_aug_5e-1/convnext_small_4xb256_IF_1m_lr4e-4_aug_5e-1.py new file mode 100644 index 0000000000000000000000000000000000000000..387bb5b9606b514ed2e86ad5d7d57200ef9f0bd9 --- /dev/null +++ b/convnext_small_4xb256_IF_1m_lr4e-4_aug_5e-1/convnext_small_4xb256_IF_1m_lr4e-4_aug_5e-1.py @@ -0,0 +1,264 @@ +optim_wrapper = dict( + optimizer=dict( + type='AdamW', + lr=0.0004, + weight_decay=0.05, + eps=1e-08, + betas=(0.9, 0.999), + _scope_='mmpretrain'), + paramwise_cfg=dict( + norm_decay_mult=0.0, + bias_decay_mult=0.0, + flat_decay_mult=0.0, + custom_keys=dict({ + '.absolute_pos_embed': dict(decay_mult=0.0), + '.relative_position_bias_table': dict(decay_mult=0.0) + })), + type='AmpOptimWrapper', + dtype='bfloat16', + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=False, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=4096) +model = dict( + type='ImageClassifier', + backbone=dict( + frozen_stages=2, + type='ConvNeXt', + arch='small', + drop_path_rate=0.4, + init_cfg=dict( + type='Pretrained', + checkpoint= + 'https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth', + prefix='backbone')), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=768, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + init_cfg=None), + init_cfg=dict( + type='TruncNormal', layer=['Conv2d', 'Linear'], std=0.02, bias=0.0), + train_cfg=None) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/IF80w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/if-dpmsolver++-50-20w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/cc1m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/if-dpmsolver++-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/if-dpmsolver++-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/convnext_small_4xb256_IF_1m_lr4e-4_aug_5e-1' diff --git a/convnext_small_4xb256_IF_1m_lr4e-4_aug_5e-1/epoch_1.pth b/convnext_small_4xb256_IF_1m_lr4e-4_aug_5e-1/epoch_1.pth new file mode 100644 index 0000000000000000000000000000000000000000..a9eda82839116bbe3d5e8db24afaa5927dc09e6b --- /dev/null +++ b/convnext_small_4xb256_IF_1m_lr4e-4_aug_5e-1/epoch_1.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c51a1df71f811e35fa437965cb72861ce9035c1635a27d419e84aaf182dd68c7 +size 782113114 diff --git a/convnext_small_4xb256_IF_1m_lr4e-4_aug_5e-1/epoch_10.pth b/convnext_small_4xb256_IF_1m_lr4e-4_aug_5e-1/epoch_10.pth new file mode 100644 index 0000000000000000000000000000000000000000..b6dbc885086983c287ff5848363e08e461f78eba --- /dev/null +++ b/convnext_small_4xb256_IF_1m_lr4e-4_aug_5e-1/epoch_10.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:1731db71ef97c9422904d19500f2389e446a9e988c46f69572e6e754d1fac560 +size 783410586 diff --git a/convnext_small_4xb256_IF_1m_lr4e-4_aug_5e-1/epoch_2.pth b/convnext_small_4xb256_IF_1m_lr4e-4_aug_5e-1/epoch_2.pth new file mode 100644 index 0000000000000000000000000000000000000000..ac85714a60635e5a35e68336cc554bc2deda25ba --- /dev/null +++ b/convnext_small_4xb256_IF_1m_lr4e-4_aug_5e-1/epoch_2.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:8238c95020c43fa81e20e0bace913d3359e677ce8febb711690107bf3cd69f07 +size 782258074 diff --git a/convnext_small_4xb256_IF_1m_lr4e-4_aug_5e-1/epoch_3.pth b/convnext_small_4xb256_IF_1m_lr4e-4_aug_5e-1/epoch_3.pth new file mode 100644 index 0000000000000000000000000000000000000000..71c036e07b23c7bd3413c7bd07c524cafd3025da --- /dev/null +++ b/convnext_small_4xb256_IF_1m_lr4e-4_aug_5e-1/epoch_3.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:e85440175bb53fc6eab9769f81c0c4327e1144a2deb02262c776fb592bd8dc37 +size 782402074 diff --git a/convnext_small_4xb256_IF_1m_lr4e-4_aug_5e-1/epoch_4.pth b/convnext_small_4xb256_IF_1m_lr4e-4_aug_5e-1/epoch_4.pth new file mode 100644 index 0000000000000000000000000000000000000000..196af6d7f3467ce522323196a9919f995ca31124 --- /dev/null +++ b/convnext_small_4xb256_IF_1m_lr4e-4_aug_5e-1/epoch_4.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:532abc6fb36c142f2892e92204891fe08464ef3f40b212e4aa538204e3e2dba4 +size 782546138 diff --git a/convnext_small_4xb256_IF_1m_lr4e-4_aug_5e-1/epoch_5.pth b/convnext_small_4xb256_IF_1m_lr4e-4_aug_5e-1/epoch_5.pth new file mode 100644 index 0000000000000000000000000000000000000000..7424edf3b9dd4088df3a02a4f491b07feaf0e025 --- /dev/null +++ b/convnext_small_4xb256_IF_1m_lr4e-4_aug_5e-1/epoch_5.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:a67e6db22205baacc39fc00f5d9551a6e90cf12037b40d36d89bca1d52e0f5fa +size 782690074 diff --git a/convnext_small_4xb256_IF_1m_lr4e-4_aug_5e-1/epoch_6.pth b/convnext_small_4xb256_IF_1m_lr4e-4_aug_5e-1/epoch_6.pth new file mode 100644 index 0000000000000000000000000000000000000000..1f53039c23d923252a67f69fe4fcbd47b3bb442e --- /dev/null +++ b/convnext_small_4xb256_IF_1m_lr4e-4_aug_5e-1/epoch_6.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:341bc211107f0bda47f6d0dd7943d139918888de3431ca77a14bcfdd67258eaf +size 782833690 diff --git a/convnext_small_4xb256_IF_1m_lr4e-4_aug_5e-1/epoch_7.pth b/convnext_small_4xb256_IF_1m_lr4e-4_aug_5e-1/epoch_7.pth new file mode 100644 index 0000000000000000000000000000000000000000..e5f732f8222f591cd496d8dccffc490ee5e3bc40 --- /dev/null +++ b/convnext_small_4xb256_IF_1m_lr4e-4_aug_5e-1/epoch_7.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:97530b97b8c876b97f50a91dea2030f9ec1a41794441add50a08411f1173c55d +size 782977306 diff --git a/convnext_small_4xb256_IF_1m_lr4e-4_aug_5e-1/epoch_8.pth b/convnext_small_4xb256_IF_1m_lr4e-4_aug_5e-1/epoch_8.pth new file mode 100644 index 0000000000000000000000000000000000000000..7ef44dcf5a33d1f2e5c9409c12846123270f3155 --- /dev/null +++ b/convnext_small_4xb256_IF_1m_lr4e-4_aug_5e-1/epoch_8.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:f088215cfd2194547210b826b15ee6c911ec0dbeda1b17916a0bb7fa48544b13 +size 783120666 diff --git a/convnext_small_4xb256_IF_1m_lr4e-4_aug_5e-1/epoch_9.pth b/convnext_small_4xb256_IF_1m_lr4e-4_aug_5e-1/epoch_9.pth new file mode 100644 index 0000000000000000000000000000000000000000..b2422f83408578c936f5bcd8c428603946f1fdf5 --- /dev/null +++ b/convnext_small_4xb256_IF_1m_lr4e-4_aug_5e-1/epoch_9.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:acc7ae1703693c1a9457ef03fe5b660e82dbc0a6c2ad9a3e3cf2c4c8df6ca9d3 +size 783265306 diff --git a/convnext_small_4xb256_IF_1m_lr4e-4_aug_5e-1/last_checkpoint b/convnext_small_4xb256_IF_1m_lr4e-4_aug_5e-1/last_checkpoint new file mode 100644 index 0000000000000000000000000000000000000000..a2a6206586589df62259a34266e780bab3147a67 --- /dev/null +++ b/convnext_small_4xb256_IF_1m_lr4e-4_aug_5e-1/last_checkpoint @@ -0,0 +1 @@ +/mnt/petrelfs/luzeyu/workspace/fakebench/mmpretrain/workdir/convnext_small_4xb256_IF_1m_lr4e-4_aug_5e-1/epoch_10.pth \ No newline at end of file diff --git a/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1/20230606_005643/20230606_005643.log b/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1/20230606_005643/20230606_005643.log new file mode 100644 index 0000000000000000000000000000000000000000..b9e0addc5f97cb177130164694d586567edd850b --- /dev/null +++ b/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1/20230606_005643/20230606_005643.log @@ -0,0 +1,1970 @@ +2023/06/06 00:56:47 - mmengine - INFO - +------------------------------------------------------------ +System environment: + sys.platform: linux + Python: 3.10.9 (main, Mar 8 2023, 10:47:38) [GCC 11.2.0] + CUDA available: True + numpy_random_seed: 545223607 + GPU 0,1,2,3: NVIDIA A100-SXM4-80GB + CUDA_HOME: /mnt/petrelfs/share/cuda-11.6 + NVCC: Cuda compilation tools, release 11.6, V11.6.124 + GCC: gcc (GCC) 7.5.0 + PyTorch: 1.13.1 + PyTorch compiling details: PyTorch built with: + - GCC 9.3 + - C++ Version: 201402 + - Intel(R) oneAPI Math Kernel Library Version 2021.4-Product Build 20210904 for Intel(R) 64 architecture applications + - Intel(R) MKL-DNN v2.6.0 (Git Hash 52b5f107dd9cf10910aaa19cb47f3abf9b349815) + - OpenMP 201511 (a.k.a. OpenMP 4.5) + - LAPACK is enabled (usually provided by MKL) + - NNPACK is enabled + - CPU capability usage: AVX2 + - CUDA Runtime 11.6 + - NVCC architecture flags: -gencode;arch=compute_37,code=sm_37;-gencode;arch=compute_50,code=sm_50;-gencode;arch=compute_60,code=sm_60;-gencode;arch=compute_61,code=sm_61;-gencode;arch=compute_70,code=sm_70;-gencode;arch=compute_75,code=sm_75;-gencode;arch=compute_80,code=sm_80;-gencode;arch=compute_86,code=sm_86;-gencode;arch=compute_37,code=compute_37 + - CuDNN 8.3.2 (built against CUDA 11.5) + - Magma 2.6.1 + - Build settings: BLAS_INFO=mkl, BUILD_TYPE=Release, CUDA_VERSION=11.6, CUDNN_VERSION=8.3.2, CXX_COMPILER=/opt/rh/devtoolset-9/root/usr/bin/c++, CXX_FLAGS= -fabi-version=11 -Wno-deprecated -fvisibility-inlines-hidden -DUSE_PTHREADPOOL -fopenmp -DNDEBUG -DUSE_KINETO -DUSE_FBGEMM -DUSE_QNNPACK -DUSE_PYTORCH_QNNPACK -DUSE_XNNPACK -DSYMBOLICATE_MOBILE_DEBUG_HANDLE -DEDGE_PROFILER_USE_KINETO -O2 -fPIC -Wno-narrowing -Wall -Wextra -Werror=return-type -Werror=non-virtual-dtor -Wno-missing-field-initializers -Wno-type-limits -Wno-array-bounds -Wno-unknown-pragmas -Wunused-local-typedefs -Wno-unused-parameter -Wno-unused-function -Wno-unused-result -Wno-strict-overflow -Wno-strict-aliasing -Wno-error=deprecated-declarations -Wno-stringop-overflow -Wno-psabi -Wno-error=pedantic -Wno-error=redundant-decls -Wno-error=old-style-cast -fdiagnostics-color=always -faligned-new -Wno-unused-but-set-variable -Wno-maybe-uninitialized -fno-math-errno -fno-trapping-math -Werror=format -Werror=cast-function-type -Wno-stringop-overflow, LAPACK_INFO=mkl, PERF_WITH_AVX=1, PERF_WITH_AVX2=1, PERF_WITH_AVX512=1, TORCH_VERSION=1.13.1, USE_CUDA=ON, USE_CUDNN=ON, USE_EXCEPTION_PTR=1, USE_GFLAGS=OFF, USE_GLOG=OFF, USE_MKL=ON, USE_MKLDNN=ON, USE_MPI=OFF, USE_NCCL=ON, USE_NNPACK=ON, USE_OPENMP=ON, USE_ROCM=OFF, + + TorchVision: 0.14.1 + OpenCV: 4.7.0 + MMEngine: 0.7.3 + +Runtime environment: + cudnn_benchmark: True + mp_cfg: {'mp_start_method': 'fork', 'opencv_num_threads': 0} + dist_cfg: {'backend': 'nccl'} + seed: None + deterministic: False + Distributed launcher: slurm + Distributed training: True + GPU number: 4 +------------------------------------------------------------ + +2023/06/06 00:56:51 - mmengine - INFO - Config: +optim_wrapper = dict( + optimizer=dict( + type='AdamW', + lr=0.0004, + weight_decay=0.05, + eps=1e-08, + betas=(0.9, 0.999), + _scope_='mmpretrain'), + paramwise_cfg=dict( + norm_decay_mult=0.0, + bias_decay_mult=0.0, + flat_decay_mult=0.0, + custom_keys=dict({ + '.absolute_pos_embed': dict(decay_mult=0.0), + '.relative_position_bias_table': dict(decay_mult=0.0) + })), + type='AmpOptimWrapper', + dtype='bfloat16', + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=False, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=4096) +model = dict( + type='ImageClassifier', + backbone=dict( + frozen_stages=2, + type='ConvNeXt', + arch='small', + drop_path_rate=0.4, + init_cfg=dict( + type='Pretrained', + checkpoint= + 'https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth', + prefix='backbone')), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=768, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + init_cfg=None), + init_cfg=dict( + type='TruncNormal', layer=['Conv2d', 'Linear'], std=0.02, bias=0.0), + train_cfg=None) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/all_0.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/all_1.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3fake8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/cc1m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3real8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV2-1-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/if-dpmsolver++-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV2-1-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/if-dpmsolver++-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1' + +2023/06/06 00:57:03 - mmengine - INFO - Hooks will be executed in the following order: +before_run: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_load_checkpoint: +(ABOVE_NORMAL) EMAHook + -------------------- +before_train: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_train_epoch: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(NORMAL ) DistSamplerSeedHook + -------------------- +before_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook + -------------------- +after_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +after_train_epoch: +(NORMAL ) IterTimerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_val_epoch: +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook + -------------------- +before_val_iter: +(NORMAL ) IterTimerHook + -------------------- +after_val_iter: +(NORMAL ) IterTimerHook +(NORMAL ) VisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_val_epoch: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_save_checkpoint: +(ABOVE_NORMAL) EMAHook + -------------------- +after_train: +(VERY_LOW ) CheckpointHook + -------------------- +before_test_epoch: +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook + -------------------- +before_test_iter: +(NORMAL ) IterTimerHook + -------------------- +after_test_iter: +(NORMAL ) IterTimerHook +(NORMAL ) VisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_test_epoch: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_run: +(BELOW_NORMAL) LoggerHook + -------------------- +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.2.0.weight:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.2.0.bias:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.2.1.bias:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.3.0.weight:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.3.0.bias:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.3.1.bias:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.2.0.gamma:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.2.0.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.2.0.norm.weight:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.2.0.norm.bias:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.2.0.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.2.0.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.2.1.gamma:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.2.1.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.2.1.norm.weight:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.2.1.norm.bias:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.2.1.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.2.1.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.2.2.gamma:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.2.2.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.2.2.norm.weight:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.2.2.norm.bias:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.2.2.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.2.2.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.2.3.gamma:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.2.3.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.2.3.norm.weight:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.2.3.norm.bias:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.2.3.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.2.3.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.2.4.gamma:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.2.4.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.2.4.norm.weight:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.2.4.norm.bias:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.2.4.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.2.4.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.2.5.gamma:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.2.5.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.2.5.norm.weight:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.2.5.norm.bias:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.2.5.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.2.5.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.2.6.gamma:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.2.6.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.2.6.norm.weight:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.2.6.norm.bias:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.2.6.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.2.6.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.2.7.gamma:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.2.7.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.2.7.norm.weight:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.2.7.norm.bias:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.2.7.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.2.7.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.2.8.gamma:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.2.8.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.2.8.norm.weight:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.2.8.norm.bias:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.2.8.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.2.8.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.2.9.gamma:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.2.9.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.2.9.norm.weight:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.2.9.norm.bias:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.2.9.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.2.9.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.2.10.gamma:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.2.10.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.2.10.norm.weight:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.2.10.norm.bias:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.2.10.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.2.10.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.2.11.gamma:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.2.11.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.2.11.norm.weight:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.2.11.norm.bias:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.2.11.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.2.11.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.2.12.gamma:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.2.12.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.2.12.norm.weight:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.2.12.norm.bias:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.2.12.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.2.12.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.2.13.gamma:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.2.13.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.2.13.norm.weight:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.2.13.norm.bias:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.2.13.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.2.13.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.2.14.gamma:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.2.14.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.2.14.norm.weight:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.2.14.norm.bias:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.2.14.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.2.14.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.2.15.gamma:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.2.15.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.2.15.norm.weight:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.2.15.norm.bias:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.2.15.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.2.15.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.2.16.gamma:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.2.16.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.2.16.norm.weight:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.2.16.norm.bias:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.2.16.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.2.16.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.2.17.gamma:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.2.17.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.2.17.norm.weight:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.2.17.norm.bias:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.2.17.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.2.17.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.2.18.gamma:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.2.18.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.2.18.norm.weight:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.2.18.norm.bias:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.2.18.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.2.18.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.2.19.gamma:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.2.19.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.2.19.norm.weight:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.2.19.norm.bias:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.2.19.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.2.19.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.2.20.gamma:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.2.20.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.2.20.norm.weight:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.2.20.norm.bias:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.2.20.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.2.20.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.2.21.gamma:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.2.21.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.2.21.norm.weight:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.2.21.norm.bias:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.2.21.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.2.21.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.2.22.gamma:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.2.22.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.2.22.norm.weight:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.2.22.norm.bias:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.2.22.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.2.22.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.2.23.gamma:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.2.23.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.2.23.norm.weight:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.2.23.norm.bias:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.2.23.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.2.23.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.2.24.gamma:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.2.24.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.2.24.norm.weight:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.2.24.norm.bias:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.2.24.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.2.24.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.2.25.gamma:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.2.25.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.2.25.norm.weight:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.2.25.norm.bias:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.2.25.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.2.25.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.2.26.gamma:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.2.26.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.2.26.norm.weight:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.2.26.norm.bias:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.2.26.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.2.26.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.3.0.gamma:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.3.0.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.3.0.norm.weight:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.3.0.norm.bias:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.3.0.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.3.0.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.3.1.gamma:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.3.1.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.3.1.norm.weight:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.3.1.norm.bias:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.3.1.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.3.1.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.3.2.gamma:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.3.2.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.3.2.norm.weight:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.3.2.norm.bias:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.3.2.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.stages.3.2.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.norm3.weight:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- backbone.norm3.bias:weight_decay=0.0 +2023/06/06 00:57:17 - mmengine - INFO - paramwise_options -- head.fc.bias:weight_decay=0.0 +2023/06/06 00:57:22 - mmengine - INFO - load backbone in model from: https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth +Name of parameter - Initialization information + +backbone.downsample_layers.0.0.weight - torch.Size([96, 3, 4, 4]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.0.0.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.0.1.weight - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.0.1.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.1.0.weight - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.1.0.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.1.1.weight - torch.Size([192, 96, 2, 2]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.1.1.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.2.0.weight - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.2.0.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.2.1.weight - torch.Size([384, 192, 2, 2]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.2.1.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.3.0.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.3.0.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.3.1.weight - torch.Size([768, 384, 2, 2]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.3.1.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.0.gamma - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.0.depthwise_conv.weight - torch.Size([96, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.0.depthwise_conv.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.0.norm.weight - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.0.norm.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.0.pointwise_conv1.weight - torch.Size([384, 96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.0.pointwise_conv1.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.0.pointwise_conv2.weight - torch.Size([96, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.0.pointwise_conv2.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.1.gamma - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.1.depthwise_conv.weight - torch.Size([96, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.1.depthwise_conv.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.1.norm.weight - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.1.norm.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.1.pointwise_conv1.weight - torch.Size([384, 96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.1.pointwise_conv1.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.1.pointwise_conv2.weight - torch.Size([96, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.1.pointwise_conv2.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.2.gamma - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.2.depthwise_conv.weight - torch.Size([96, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.2.depthwise_conv.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.2.norm.weight - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.2.norm.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.2.pointwise_conv1.weight - torch.Size([384, 96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.2.pointwise_conv1.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.2.pointwise_conv2.weight - torch.Size([96, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.2.pointwise_conv2.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.0.gamma - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.0.depthwise_conv.weight - torch.Size([192, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.0.depthwise_conv.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.0.norm.weight - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.0.norm.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.0.pointwise_conv1.weight - torch.Size([768, 192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.0.pointwise_conv1.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.0.pointwise_conv2.weight - torch.Size([192, 768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.0.pointwise_conv2.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.1.gamma - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.1.depthwise_conv.weight - torch.Size([192, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.1.depthwise_conv.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.1.norm.weight - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.1.norm.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.1.pointwise_conv1.weight - torch.Size([768, 192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.1.pointwise_conv1.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.1.pointwise_conv2.weight - torch.Size([192, 768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.1.pointwise_conv2.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.2.gamma - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.2.depthwise_conv.weight - torch.Size([192, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.2.depthwise_conv.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.2.norm.weight - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.2.norm.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.2.pointwise_conv1.weight - torch.Size([768, 192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.2.pointwise_conv1.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.2.pointwise_conv2.weight - torch.Size([192, 768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.2.pointwise_conv2.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.0.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.0.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.0.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.0.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.0.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.0.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.0.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.0.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.0.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.1.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.1.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.1.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.1.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.1.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.1.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.1.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.1.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.1.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.2.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.2.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.2.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.2.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.2.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.2.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.2.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.2.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.2.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.3.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.3.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.3.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.3.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.3.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.3.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.3.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.3.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.3.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.4.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.4.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.4.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.4.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.4.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.4.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.4.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.4.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.4.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.5.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.5.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.5.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.5.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.5.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.5.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.5.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.5.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.5.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.6.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.6.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.6.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.6.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.6.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.6.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.6.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.6.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.6.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.7.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.7.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.7.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.7.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.7.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.7.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.7.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.7.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.7.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.8.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.8.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.8.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.8.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.8.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.8.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.8.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.8.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.8.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.9.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.9.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.9.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.9.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.9.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.9.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.9.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.9.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.9.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.10.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.10.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.10.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.10.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.10.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.10.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.10.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.10.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.10.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.11.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.11.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.11.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.11.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.11.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.11.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.11.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.11.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.11.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.12.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.12.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.12.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.12.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.12.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.12.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.12.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.12.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.12.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.13.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.13.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.13.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.13.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.13.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.13.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.13.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.13.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.13.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.14.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.14.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.14.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.14.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.14.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.14.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.14.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.14.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.14.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.15.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.15.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.15.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.15.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.15.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.15.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.15.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.15.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.15.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.16.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.16.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.16.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.16.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.16.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.16.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.16.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.16.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.16.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.17.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.17.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.17.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.17.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.17.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.17.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.17.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.17.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.17.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.18.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.18.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.18.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.18.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.18.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.18.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.18.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.18.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.18.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.19.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.19.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.19.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.19.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.19.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.19.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.19.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.19.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.19.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.20.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.20.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.20.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.20.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.20.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.20.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.20.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.20.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.20.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.21.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.21.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.21.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.21.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.21.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.21.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.21.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.21.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.21.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.22.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.22.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.22.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.22.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.22.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.22.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.22.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.22.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.22.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.23.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.23.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.23.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.23.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.23.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.23.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.23.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.23.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.23.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.24.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.24.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.24.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.24.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.24.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.24.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.24.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.24.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.24.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.25.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.25.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.25.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.25.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.25.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.25.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.25.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.25.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.25.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.26.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.26.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.26.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.26.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.26.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.26.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.26.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.26.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.26.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.0.gamma - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.0.depthwise_conv.weight - torch.Size([768, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.0.depthwise_conv.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.0.norm.weight - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.0.norm.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.0.pointwise_conv1.weight - torch.Size([3072, 768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.0.pointwise_conv1.bias - torch.Size([3072]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.0.pointwise_conv2.weight - torch.Size([768, 3072]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.0.pointwise_conv2.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.1.gamma - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.1.depthwise_conv.weight - torch.Size([768, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.1.depthwise_conv.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.1.norm.weight - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.1.norm.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.1.pointwise_conv1.weight - torch.Size([3072, 768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.1.pointwise_conv1.bias - torch.Size([3072]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.1.pointwise_conv2.weight - torch.Size([768, 3072]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.1.pointwise_conv2.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.2.gamma - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.2.depthwise_conv.weight - torch.Size([768, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.2.depthwise_conv.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.2.norm.weight - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.2.norm.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.2.pointwise_conv1.weight - torch.Size([3072, 768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.2.pointwise_conv1.bias - torch.Size([3072]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.2.pointwise_conv2.weight - torch.Size([768, 3072]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.2.pointwise_conv2.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.norm3.weight - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.norm3.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +head.fc.weight - torch.Size([2, 768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +head.fc.bias - torch.Size([2]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 +2023/06/06 00:57:23 - mmengine - WARNING - "FileClient" will be deprecated in future. Please use io functions in https://mmengine.readthedocs.io/en/latest/api/fileio.html#file-io +2023/06/06 00:57:23 - mmengine - WARNING - "HardDiskBackend" is the alias of "LocalBackend" and the former will be deprecated in future. +2023/06/06 00:57:23 - mmengine - INFO - Checkpoints will be saved to /mnt/petrelfs/luzeyu/workspace/fakebench/mmpretrain/workdir/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1. +2023/06/06 00:58:47 - mmengine - INFO - Epoch(train) [1][ 100/2046] lr: 3.9998e-04 eta: 4:47:46 time: 0.7797 data_time: 0.0017 memory: 15587 loss: 0.0458 +2023/06/06 01:00:07 - mmengine - INFO - Epoch(train) [1][ 200/2046] lr: 3.9991e-04 eta: 4:36:56 time: 0.7382 data_time: 0.0018 memory: 15587 loss: 0.0491 +2023/06/06 01:01:22 - mmengine - INFO - Epoch(train) [1][ 300/2046] lr: 3.9979e-04 eta: 4:27:45 time: 0.7866 data_time: 0.1939 memory: 15587 loss: 0.0337 +2023/06/06 01:02:41 - mmengine - INFO - Epoch(train) [1][ 400/2046] lr: 3.9963e-04 eta: 4:25:58 time: 0.8206 data_time: 0.0015 memory: 15587 loss: 0.0264 +2023/06/06 01:04:00 - mmengine - INFO - Epoch(train) [1][ 500/2046] lr: 3.9943e-04 eta: 4:24:12 time: 0.7349 data_time: 0.0017 memory: 15587 loss: 0.0323 +2023/06/06 01:05:18 - mmengine - INFO - Epoch(train) [1][ 600/2046] lr: 3.9918e-04 eta: 4:22:16 time: 0.8030 data_time: 0.0016 memory: 15587 loss: 0.0361 +2023/06/06 01:06:35 - mmengine - INFO - Epoch(train) [1][ 700/2046] lr: 3.9888e-04 eta: 4:19:53 time: 0.7368 data_time: 0.0016 memory: 15587 loss: 0.0309 +2023/06/06 01:07:53 - mmengine - INFO - Epoch(train) [1][ 800/2046] lr: 3.9853e-04 eta: 4:18:11 time: 0.7841 data_time: 0.0016 memory: 15587 loss: 0.0283 +2023/06/06 01:09:12 - mmengine - INFO - Epoch(train) [1][ 900/2046] lr: 3.9815e-04 eta: 4:16:49 time: 0.7561 data_time: 0.0017 memory: 15587 loss: 0.0323 +2023/06/06 01:10:30 - mmengine - INFO - Exp name: convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1_20230606_005643 +2023/06/06 01:10:30 - mmengine - INFO - Epoch(train) [1][1000/2046] lr: 3.9771e-04 eta: 4:15:16 time: 0.7418 data_time: 0.0015 memory: 15587 loss: 0.0242 +2023/06/06 01:11:47 - mmengine - INFO - Epoch(train) [1][1100/2046] lr: 3.9723e-04 eta: 4:13:33 time: 0.7945 data_time: 0.0016 memory: 15587 loss: 0.0282 +2023/06/06 01:13:03 - mmengine - INFO - Epoch(train) [1][1200/2046] lr: 3.9670e-04 eta: 4:11:29 time: 0.7638 data_time: 0.0016 memory: 15587 loss: 0.0250 +2023/06/06 01:14:19 - mmengine - INFO - Epoch(train) [1][1300/2046] lr: 3.9613e-04 eta: 4:09:32 time: 0.8598 data_time: 0.0024 memory: 15587 loss: 0.0172 +2023/06/06 01:15:36 - mmengine - INFO - Epoch(train) [1][1400/2046] lr: 3.9552e-04 eta: 4:07:58 time: 0.7743 data_time: 0.0015 memory: 15587 loss: 0.0270 +2023/06/06 01:16:52 - mmengine - INFO - Epoch(train) [1][1500/2046] lr: 3.9486e-04 eta: 4:06:23 time: 0.8124 data_time: 0.0017 memory: 15587 loss: 0.0190 +2023/06/06 01:18:10 - mmengine - INFO - Epoch(train) [1][1600/2046] lr: 3.9415e-04 eta: 4:04:58 time: 0.7806 data_time: 0.0014 memory: 15587 loss: 0.0246 +2023/06/06 01:19:24 - mmengine - INFO - Epoch(train) [1][1700/2046] lr: 3.9340e-04 eta: 4:03:02 time: 0.7742 data_time: 0.0014 memory: 15587 loss: 0.0195 +2023/06/06 01:20:41 - mmengine - INFO - Epoch(train) [1][1800/2046] lr: 3.9261e-04 eta: 4:01:40 time: 0.7816 data_time: 0.0014 memory: 15587 loss: 0.0189 +2023/06/06 01:21:57 - mmengine - INFO - Epoch(train) [1][1900/2046] lr: 3.9177e-04 eta: 4:00:02 time: 0.7651 data_time: 0.0014 memory: 15587 loss: 0.0173 +2023/06/06 01:23:13 - mmengine - INFO - Exp name: convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1_20230606_005643 +2023/06/06 01:23:13 - mmengine - INFO - Epoch(train) [1][2000/2046] lr: 3.9089e-04 eta: 3:58:28 time: 0.7751 data_time: 0.0015 memory: 15587 loss: 0.0145 +2023/06/06 01:23:45 - mmengine - INFO - Exp name: convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1_20230606_005643 +2023/06/06 01:23:45 - mmengine - INFO - Saving checkpoint at 1 epochs +2023/06/06 01:24:45 - mmengine - INFO - Epoch(val) [1][60/60] accuracy/top1: 50.6926 single-label/precision_classwise: [99.60828399658203, 30.150636672973633] single-label/recall_classwise: [37.45554733276367, 99.45736694335938] single-label/f1-score_classwise: [54.44007873535156, 46.27342224121094] data_time: 0.1255 time: 0.5631 +2023/06/06 01:26:03 - mmengine - INFO - Epoch(train) [2][ 100/2046] lr: 3.8952e-04 eta: 3:56:09 time: 0.7574 data_time: 0.0352 memory: 15587 loss: 0.0203 +2023/06/06 01:28:51 - mmengine - INFO - Epoch(train) [2][ 200/2046] lr: 3.8853e-04 eta: 4:07:02 time: 0.7844 data_time: 0.0185 memory: 15585 loss: 0.0246 +2023/06/06 01:30:06 - mmengine - INFO - Epoch(train) [2][ 300/2046] lr: 3.8750e-04 eta: 4:04:49 time: 0.7823 data_time: 0.0015 memory: 15585 loss: 0.0142 +2023/06/06 01:31:21 - mmengine - INFO - Epoch(train) [2][ 400/2046] lr: 3.8642e-04 eta: 4:02:48 time: 0.7561 data_time: 0.0019 memory: 15585 loss: 0.0222 +2023/06/06 01:32:37 - mmengine - INFO - Epoch(train) [2][ 500/2046] lr: 3.8530e-04 eta: 4:00:53 time: 0.7986 data_time: 0.0015 memory: 15585 loss: 0.0148 +2023/06/06 01:33:54 - mmengine - INFO - Epoch(train) [2][ 600/2046] lr: 3.8414e-04 eta: 3:59:06 time: 0.7966 data_time: 0.0015 memory: 15585 loss: 0.0155 +2023/06/06 01:35:09 - mmengine - INFO - Epoch(train) [2][ 700/2046] lr: 3.8293e-04 eta: 3:57:08 time: 0.7410 data_time: 0.0014 memory: 15585 loss: 0.0125 +2023/06/06 01:36:23 - mmengine - INFO - Epoch(train) [2][ 800/2046] lr: 3.8169e-04 eta: 3:55:07 time: 0.7098 data_time: 0.0016 memory: 15585 loss: 0.0159 +2023/06/06 01:37:36 - mmengine - INFO - Epoch(train) [2][ 900/2046] lr: 3.8040e-04 eta: 3:53:10 time: 0.7683 data_time: 0.0015 memory: 15585 loss: 0.0207 +2023/06/06 01:38:16 - mmengine - INFO - Exp name: convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1_20230606_005643 +2023/06/06 01:38:50 - mmengine - INFO - Epoch(train) [2][1000/2046] lr: 3.7907e-04 eta: 3:51:12 time: 0.7313 data_time: 0.0016 memory: 15585 loss: 0.0197 +2023/06/06 01:40:06 - mmengine - INFO - Epoch(train) [2][1100/2046] lr: 3.7770e-04 eta: 3:49:36 time: 0.7730 data_time: 0.0016 memory: 15585 loss: 0.0212 +2023/06/06 01:41:23 - mmengine - INFO - Epoch(train) [2][1200/2046] lr: 3.7629e-04 eta: 3:48:04 time: 0.7195 data_time: 0.0016 memory: 15585 loss: 0.0191 +2023/06/06 01:42:39 - mmengine - INFO - Epoch(train) [2][1300/2046] lr: 3.7484e-04 eta: 3:46:24 time: 0.7650 data_time: 0.0015 memory: 15585 loss: 0.0165 +2023/06/06 01:43:55 - mmengine - INFO - Epoch(train) [2][1400/2046] lr: 3.7335e-04 eta: 3:44:49 time: 0.7551 data_time: 0.0014 memory: 15585 loss: 0.0138 +2023/06/06 01:45:11 - mmengine - INFO - Epoch(train) [2][1500/2046] lr: 3.7182e-04 eta: 3:43:14 time: 0.7818 data_time: 0.0018 memory: 15585 loss: 0.0191 +2023/06/06 01:46:28 - mmengine - INFO - Epoch(train) [2][1600/2046] lr: 3.7025e-04 eta: 3:41:41 time: 0.8163 data_time: 0.0016 memory: 15585 loss: 0.0148 +2023/06/06 01:47:49 - mmengine - INFO - Epoch(train) [2][1700/2046] lr: 3.6864e-04 eta: 3:40:31 time: 0.7613 data_time: 0.0015 memory: 15585 loss: 0.0122 +2023/06/06 01:49:06 - mmengine - INFO - Epoch(train) [2][1800/2046] lr: 3.6699e-04 eta: 3:39:03 time: 0.7514 data_time: 0.0016 memory: 15585 loss: 0.0146 +2023/06/06 01:50:18 - mmengine - INFO - Epoch(train) [2][1900/2046] lr: 3.6531e-04 eta: 3:37:13 time: 0.6864 data_time: 0.0016 memory: 15585 loss: 0.0081 +2023/06/06 01:50:54 - mmengine - INFO - Exp name: convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1_20230606_005643 +2023/06/06 01:51:30 - mmengine - INFO - Epoch(train) [2][2000/2046] lr: 3.6358e-04 eta: 3:35:27 time: 0.7499 data_time: 0.0019 memory: 15585 loss: 0.0160 +2023/06/06 01:52:04 - mmengine - INFO - Exp name: convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1_20230606_005643 +2023/06/06 01:52:04 - mmengine - INFO - Saving checkpoint at 2 epochs +2023/06/06 01:53:01 - mmengine - INFO - Epoch(val) [2][60/60] accuracy/top1: 82.2932 single-label/precision_classwise: [99.82679748535156, 54.68876266479492] single-label/recall_classwise: [77.62136840820312, 99.50387573242188] single-label/f1-score_classwise: [87.334716796875, 70.58370208740234] data_time: 0.1093 time: 0.5076 +2023/06/06 01:54:19 - mmengine - INFO - Epoch(train) [3][ 100/2046] lr: 3.6100e-04 eta: 3:33:21 time: 0.7322 data_time: 0.0628 memory: 15585 loss: 0.0193 +2023/06/06 01:55:35 - mmengine - INFO - Epoch(train) [3][ 200/2046] lr: 3.5918e-04 eta: 3:31:51 time: 0.7675 data_time: 0.0019 memory: 15585 loss: 0.0135 +2023/06/06 01:56:48 - mmengine - INFO - Epoch(train) [3][ 300/2046] lr: 3.5733e-04 eta: 3:30:14 time: 0.7365 data_time: 0.0016 memory: 15585 loss: 0.0184 +2023/06/06 01:58:04 - mmengine - INFO - Epoch(train) [3][ 400/2046] lr: 3.5544e-04 eta: 3:28:45 time: 0.7499 data_time: 0.0016 memory: 15585 loss: 0.0132 +2023/06/06 01:59:20 - mmengine - INFO - Epoch(train) [3][ 500/2046] lr: 3.5352e-04 eta: 3:27:19 time: 0.7414 data_time: 0.0015 memory: 15585 loss: 0.0153 +2023/06/06 02:00:36 - mmengine - INFO - Epoch(train) [3][ 600/2046] lr: 3.5156e-04 eta: 3:25:50 time: 0.7465 data_time: 0.0015 memory: 15585 loss: 0.0100 +2023/06/06 02:01:50 - mmengine - INFO - Epoch(train) [3][ 700/2046] lr: 3.4957e-04 eta: 3:24:19 time: 0.7002 data_time: 0.0016 memory: 15585 loss: 0.0143 +2023/06/06 02:03:06 - mmengine - INFO - Epoch(train) [3][ 800/2046] lr: 3.4755e-04 eta: 3:22:53 time: 0.7270 data_time: 0.0016 memory: 15585 loss: 0.0122 +2023/06/06 02:04:23 - mmengine - INFO - Epoch(train) [3][ 900/2046] lr: 3.4549e-04 eta: 3:21:31 time: 0.7481 data_time: 0.0015 memory: 15585 loss: 0.0170 +2023/06/06 02:04:29 - mmengine - INFO - Exp name: convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1_20230606_005643 +2023/06/06 02:05:39 - mmengine - INFO - Epoch(train) [3][1000/2046] lr: 3.4339e-04 eta: 3:20:07 time: 0.7832 data_time: 0.0015 memory: 15585 loss: 0.0125 +2023/06/06 02:06:51 - mmengine - INFO - Epoch(train) [3][1100/2046] lr: 3.4127e-04 eta: 3:18:32 time: 0.7109 data_time: 0.0017 memory: 15585 loss: 0.0122 +2023/06/06 02:08:04 - mmengine - INFO - Epoch(train) [3][1200/2046] lr: 3.3911e-04 eta: 3:16:58 time: 0.7497 data_time: 0.0018 memory: 15585 loss: 0.0105 +2023/06/06 02:09:13 - mmengine - INFO - Epoch(train) [3][1300/2046] lr: 3.3692e-04 eta: 3:15:16 time: 0.7058 data_time: 0.0015 memory: 15585 loss: 0.0105 +2023/06/06 02:10:22 - mmengine - INFO - Epoch(train) [3][1400/2046] lr: 3.3470e-04 eta: 3:13:34 time: 0.6669 data_time: 0.0015 memory: 15585 loss: 0.0115 +2023/06/06 02:11:34 - mmengine - INFO - Epoch(train) [3][1500/2046] lr: 3.3245e-04 eta: 3:12:01 time: 0.7419 data_time: 0.0015 memory: 15585 loss: 0.0173 +2023/06/06 02:12:45 - mmengine - INFO - Epoch(train) [3][1600/2046] lr: 3.3017e-04 eta: 3:10:28 time: 0.7054 data_time: 0.0017 memory: 15585 loss: 0.0083 +2023/06/06 02:14:03 - mmengine - INFO - Epoch(train) [3][1700/2046] lr: 3.2786e-04 eta: 3:09:12 time: 0.8213 data_time: 0.0016 memory: 15585 loss: 0.0104 +2023/06/06 02:15:13 - mmengine - INFO - Epoch(train) [3][1800/2046] lr: 3.2552e-04 eta: 3:07:36 time: 0.7036 data_time: 0.0015 memory: 15585 loss: 0.0135 +2023/06/06 02:16:24 - mmengine - INFO - Epoch(train) [3][1900/2046] lr: 3.2315e-04 eta: 3:06:03 time: 0.7210 data_time: 0.0016 memory: 15585 loss: 0.0127 +2023/06/06 02:16:30 - mmengine - INFO - Exp name: convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1_20230606_005643 +2023/06/06 02:17:36 - mmengine - INFO - Epoch(train) [3][2000/2046] lr: 3.2075e-04 eta: 3:04:33 time: 0.7016 data_time: 0.0015 memory: 15585 loss: 0.0097 +2023/06/06 02:18:08 - mmengine - INFO - Exp name: convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1_20230606_005643 +2023/06/06 02:18:08 - mmengine - INFO - Saving checkpoint at 3 epochs +2023/06/06 02:19:06 - mmengine - INFO - Epoch(val) [3][60/60] accuracy/top1: 86.2817 single-label/precision_classwise: [99.9312744140625, 60.908443450927734] single-label/recall_classwise: [82.61473083496094, 99.79069519042969] single-label/f1-score_classwise: [90.45166015625, 75.6456527709961] data_time: 0.1113 time: 0.5069 +2023/06/06 02:20:24 - mmengine - INFO - Epoch(train) [4][ 100/2046] lr: 3.1721e-04 eta: 3:02:36 time: 0.7208 data_time: 0.2000 memory: 15585 loss: 0.0105 +2023/06/06 02:21:35 - mmengine - INFO - Epoch(train) [4][ 200/2046] lr: 3.1474e-04 eta: 3:01:05 time: 0.7366 data_time: 0.1850 memory: 15585 loss: 0.0112 +2023/06/06 02:22:48 - mmengine - INFO - Epoch(train) [4][ 300/2046] lr: 3.1226e-04 eta: 2:59:40 time: 0.7972 data_time: 0.0301 memory: 15585 loss: 0.0111 +2023/06/06 02:24:00 - mmengine - INFO - Epoch(train) [4][ 400/2046] lr: 3.0974e-04 eta: 2:58:14 time: 0.7285 data_time: 0.0015 memory: 15585 loss: 0.0200 +2023/06/06 02:25:11 - mmengine - INFO - Epoch(train) [4][ 500/2046] lr: 3.0721e-04 eta: 2:56:44 time: 0.7132 data_time: 0.0016 memory: 15585 loss: 0.0110 +2023/06/06 02:26:23 - mmengine - INFO - Epoch(train) [4][ 600/2046] lr: 3.0464e-04 eta: 2:55:18 time: 0.7079 data_time: 0.0017 memory: 15585 loss: 0.0242 +2023/06/06 02:27:35 - mmengine - INFO - Epoch(train) [4][ 700/2046] lr: 3.0206e-04 eta: 2:53:52 time: 0.7495 data_time: 0.0020 memory: 15585 loss: 0.0082 +2023/06/06 02:28:49 - mmengine - INFO - Epoch(train) [4][ 800/2046] lr: 2.9945e-04 eta: 2:52:29 time: 0.7136 data_time: 0.0015 memory: 15585 loss: 0.0105 +2023/06/06 02:29:35 - mmengine - INFO - Exp name: convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1_20230606_005643 +2023/06/06 02:30:00 - mmengine - INFO - Epoch(train) [4][ 900/2046] lr: 2.9682e-04 eta: 2:51:04 time: 0.7208 data_time: 0.0016 memory: 15585 loss: 0.0111 +2023/06/06 02:31:13 - mmengine - INFO - Epoch(train) [4][1000/2046] lr: 2.9417e-04 eta: 2:49:40 time: 0.7220 data_time: 0.0015 memory: 15585 loss: 0.0153 +2023/06/06 02:32:27 - mmengine - INFO - Epoch(train) [4][1100/2046] lr: 2.9149e-04 eta: 2:48:20 time: 0.7023 data_time: 0.0016 memory: 15585 loss: 0.0111 +2023/06/06 02:33:43 - mmengine - INFO - Epoch(train) [4][1200/2046] lr: 2.8880e-04 eta: 2:47:03 time: 0.7223 data_time: 0.0017 memory: 15585 loss: 0.0069 +2023/06/06 02:35:03 - mmengine - INFO - Epoch(train) [4][1300/2046] lr: 2.8609e-04 eta: 2:45:53 time: 0.7889 data_time: 0.0019 memory: 15585 loss: 0.0085 +2023/06/06 02:36:19 - mmengine - INFO - Epoch(train) [4][1400/2046] lr: 2.8336e-04 eta: 2:44:35 time: 0.9031 data_time: 0.0016 memory: 15585 loss: 0.0123 +2023/06/06 02:37:38 - mmengine - INFO - Epoch(train) [4][1500/2046] lr: 2.8060e-04 eta: 2:43:22 time: 0.7258 data_time: 0.0015 memory: 15585 loss: 0.0113 +2023/06/06 02:38:50 - mmengine - INFO - Epoch(train) [4][1600/2046] lr: 2.7784e-04 eta: 2:41:59 time: 0.7605 data_time: 0.0019 memory: 15585 loss: 0.0109 +2023/06/06 02:40:01 - mmengine - INFO - Epoch(train) [4][1700/2046] lr: 2.7505e-04 eta: 2:40:34 time: 0.6591 data_time: 0.0018 memory: 15585 loss: 0.0160 +2023/06/06 02:41:16 - mmengine - INFO - Epoch(train) [4][1800/2046] lr: 2.7225e-04 eta: 2:39:15 time: 0.7746 data_time: 0.0017 memory: 15585 loss: 0.0174 +2023/06/06 02:42:02 - mmengine - INFO - Exp name: convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1_20230606_005643 +2023/06/06 02:42:30 - mmengine - INFO - Epoch(train) [4][1900/2046] lr: 2.6943e-04 eta: 2:37:56 time: 0.7723 data_time: 0.0016 memory: 15585 loss: 0.0188 +2023/06/06 02:43:43 - mmengine - INFO - Epoch(train) [4][2000/2046] lr: 2.6660e-04 eta: 2:36:34 time: 0.7511 data_time: 0.0016 memory: 15585 loss: 0.0119 +2023/06/06 02:44:15 - mmengine - INFO - Exp name: convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1_20230606_005643 +2023/06/06 02:44:15 - mmengine - INFO - Saving checkpoint at 4 epochs +2023/06/06 02:45:15 - mmengine - INFO - Epoch(val) [4][60/60] accuracy/top1: 85.9673 single-label/precision_classwise: [99.93093872070312, 60.36576461791992] single-label/recall_classwise: [82.21492767333984, 99.79069519042969] single-label/f1-score_classwise: [90.21137237548828, 75.2257080078125] data_time: 0.1060 time: 0.5014 +2023/06/06 02:46:35 - mmengine - INFO - Epoch(train) [5][ 100/2046] lr: 2.6243e-04 eta: 2:34:45 time: 0.7169 data_time: 0.1993 memory: 15585 loss: 0.0108 +2023/06/06 02:47:47 - mmengine - INFO - Epoch(train) [5][ 200/2046] lr: 2.5956e-04 eta: 2:33:23 time: 0.6787 data_time: 0.1657 memory: 15585 loss: 0.0131 +2023/06/06 02:49:00 - mmengine - INFO - Epoch(train) [5][ 300/2046] lr: 2.5668e-04 eta: 2:32:01 time: 0.7334 data_time: 0.2105 memory: 15585 loss: 0.0080 +2023/06/06 02:50:12 - mmengine - INFO - Epoch(train) [5][ 400/2046] lr: 2.5379e-04 eta: 2:30:40 time: 0.7105 data_time: 0.1912 memory: 15585 loss: 0.0067 +2023/06/06 02:51:24 - mmengine - INFO - Epoch(train) [5][ 500/2046] lr: 2.5089e-04 eta: 2:29:18 time: 0.7173 data_time: 0.1859 memory: 15585 loss: 0.0128 +2023/06/06 02:52:36 - mmengine - INFO - Epoch(train) [5][ 600/2046] lr: 2.4797e-04 eta: 2:27:56 time: 0.7288 data_time: 0.2084 memory: 15585 loss: 0.0167 +2023/06/06 02:53:48 - mmengine - INFO - Epoch(train) [5][ 700/2046] lr: 2.4504e-04 eta: 2:26:34 time: 0.7263 data_time: 0.2089 memory: 15585 loss: 0.0111 +2023/06/06 02:55:01 - mmengine - INFO - Epoch(train) [5][ 800/2046] lr: 2.4211e-04 eta: 2:25:16 time: 0.6963 data_time: 0.0016 memory: 15585 loss: 0.0126 +2023/06/06 02:55:15 - mmengine - INFO - Exp name: convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1_20230606_005643 +2023/06/06 02:56:14 - mmengine - INFO - Epoch(train) [5][ 900/2046] lr: 2.3917e-04 eta: 2:23:55 time: 0.7081 data_time: 0.0676 memory: 15585 loss: 0.0116 +2023/06/06 02:57:26 - mmengine - INFO - Epoch(train) [5][1000/2046] lr: 2.3621e-04 eta: 2:22:35 time: 0.7398 data_time: 0.1405 memory: 15585 loss: 0.0085 +2023/06/06 02:58:44 - mmengine - INFO - Epoch(train) [5][1100/2046] lr: 2.3326e-04 eta: 2:21:21 time: 0.7699 data_time: 0.1513 memory: 15585 loss: 0.0118 +2023/06/06 02:59:56 - mmengine - INFO - Epoch(train) [5][1200/2046] lr: 2.3029e-04 eta: 2:20:01 time: 0.7450 data_time: 0.2086 memory: 15585 loss: 0.0078 +2023/06/06 03:01:10 - mmengine - INFO - Epoch(train) [5][1300/2046] lr: 2.2732e-04 eta: 2:18:43 time: 0.7762 data_time: 0.0874 memory: 15585 loss: 0.0093 +2023/06/06 03:02:18 - mmengine - INFO - Epoch(train) [5][1400/2046] lr: 2.2434e-04 eta: 2:17:18 time: 0.6896 data_time: 0.0015 memory: 15585 loss: 0.0123 +2023/06/06 03:03:30 - mmengine - INFO - Epoch(train) [5][1500/2046] lr: 2.2136e-04 eta: 2:15:58 time: 0.7268 data_time: 0.0016 memory: 15585 loss: 0.0073 +2023/06/06 03:04:44 - mmengine - INFO - Epoch(train) [5][1600/2046] lr: 2.1837e-04 eta: 2:14:41 time: 0.7389 data_time: 0.0015 memory: 15585 loss: 0.0089 +2023/06/06 03:06:00 - mmengine - INFO - Epoch(train) [5][1700/2046] lr: 2.1538e-04 eta: 2:13:26 time: 0.7554 data_time: 0.0017 memory: 15585 loss: 0.0123 +2023/06/06 03:07:17 - mmengine - INFO - Epoch(train) [5][1800/2046] lr: 2.1239e-04 eta: 2:12:11 time: 0.7416 data_time: 0.0016 memory: 15585 loss: 0.0074 +2023/06/06 03:07:28 - mmengine - INFO - Exp name: convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1_20230606_005643 +2023/06/06 03:08:34 - mmengine - INFO - Epoch(train) [5][1900/2046] lr: 2.0940e-04 eta: 2:10:57 time: 0.7387 data_time: 0.0016 memory: 15585 loss: 0.0103 +2023/06/06 03:09:50 - mmengine - INFO - Epoch(train) [5][2000/2046] lr: 2.0641e-04 eta: 2:09:41 time: 0.8017 data_time: 0.0016 memory: 15585 loss: 0.0129 +2023/06/06 03:10:24 - mmengine - INFO - Exp name: convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1_20230606_005643 +2023/06/06 03:10:24 - mmengine - INFO - Saving checkpoint at 5 epochs +2023/06/06 03:11:21 - mmengine - INFO - Epoch(val) [5][60/60] accuracy/top1: 85.7604 single-label/precision_classwise: [99.95892333984375, 60.003726959228516] single-label/recall_classwise: [81.92874908447266, 99.87596893310547] single-label/f1-score_classwise: [90.0501937866211, 74.9679946899414] data_time: 0.1039 time: 0.5004 +2023/06/06 03:12:41 - mmengine - INFO - Epoch(train) [6][ 100/2046] lr: 2.0204e-04 eta: 2:07:53 time: 0.6941 data_time: 0.0337 memory: 15585 loss: 0.0068 +2023/06/06 03:14:00 - mmengine - INFO - Epoch(train) [6][ 200/2046] lr: 1.9904e-04 eta: 2:06:41 time: 0.7269 data_time: 0.0016 memory: 15585 loss: 0.0091 +2023/06/06 03:15:16 - mmengine - INFO - Epoch(train) [6][ 300/2046] lr: 1.9605e-04 eta: 2:05:25 time: 0.7452 data_time: 0.0015 memory: 15585 loss: 0.0099 +2023/06/06 03:16:32 - mmengine - INFO - Epoch(train) [6][ 400/2046] lr: 1.9306e-04 eta: 2:04:10 time: 0.7591 data_time: 0.0015 memory: 15585 loss: 0.0139 +2023/06/06 03:17:49 - mmengine - INFO - Epoch(train) [6][ 500/2046] lr: 1.9007e-04 eta: 2:02:55 time: 0.7070 data_time: 0.0016 memory: 15585 loss: 0.0069 +2023/06/06 03:19:04 - mmengine - INFO - Epoch(train) [6][ 600/2046] lr: 1.8709e-04 eta: 2:01:39 time: 0.7734 data_time: 0.0017 memory: 15585 loss: 0.0077 +2023/06/06 03:20:20 - mmengine - INFO - Epoch(train) [6][ 700/2046] lr: 1.8411e-04 eta: 2:00:23 time: 0.7779 data_time: 0.0015 memory: 15585 loss: 0.0125 +2023/06/06 03:21:15 - mmengine - INFO - Exp name: convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1_20230606_005643 +2023/06/06 03:21:37 - mmengine - INFO - Epoch(train) [6][ 800/2046] lr: 1.8114e-04 eta: 1:59:09 time: 0.7920 data_time: 0.0018 memory: 15585 loss: 0.0078 +2023/06/06 03:22:56 - mmengine - INFO - Epoch(train) [6][ 900/2046] lr: 1.7817e-04 eta: 1:57:55 time: 0.7732 data_time: 0.0016 memory: 15585 loss: 0.0096 +2023/06/06 03:24:20 - mmengine - INFO - Epoch(train) [6][1000/2046] lr: 1.7521e-04 eta: 1:56:46 time: 0.7342 data_time: 0.0016 memory: 15585 loss: 0.0062 +2023/06/06 03:25:39 - mmengine - INFO - Epoch(train) [6][1100/2046] lr: 1.7225e-04 eta: 1:55:32 time: 0.8174 data_time: 0.0018 memory: 15585 loss: 0.0094 +2023/06/06 03:26:56 - mmengine - INFO - Epoch(train) [6][1200/2046] lr: 1.6930e-04 eta: 1:54:18 time: 0.7791 data_time: 0.0018 memory: 15585 loss: 0.0091 +2023/06/06 03:28:14 - mmengine - INFO - Epoch(train) [6][1300/2046] lr: 1.6636e-04 eta: 1:53:03 time: 0.8271 data_time: 0.0015 memory: 15585 loss: 0.0123 +2023/06/06 03:29:33 - mmengine - INFO - Epoch(train) [6][1400/2046] lr: 1.6343e-04 eta: 1:51:49 time: 0.6922 data_time: 0.0015 memory: 15585 loss: 0.0087 +2023/06/06 03:30:46 - mmengine - INFO - Epoch(train) [6][1500/2046] lr: 1.6051e-04 eta: 1:50:31 time: 0.7868 data_time: 0.0015 memory: 15585 loss: 0.0038 +2023/06/06 03:32:00 - mmengine - INFO - Epoch(train) [6][1600/2046] lr: 1.5760e-04 eta: 1:49:14 time: 0.7386 data_time: 0.0016 memory: 15585 loss: 0.0109 +2023/06/06 03:33:14 - mmengine - INFO - Epoch(train) [6][1700/2046] lr: 1.5470e-04 eta: 1:47:57 time: 0.7590 data_time: 0.0016 memory: 15585 loss: 0.0093 +2023/06/06 03:34:07 - mmengine - INFO - Exp name: convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1_20230606_005643 +2023/06/06 03:34:29 - mmengine - INFO - Epoch(train) [6][1800/2046] lr: 1.5182e-04 eta: 1:46:40 time: 0.6745 data_time: 0.0016 memory: 15585 loss: 0.0080 +2023/06/06 03:35:43 - mmengine - INFO - Epoch(train) [6][1900/2046] lr: 1.4894e-04 eta: 1:45:23 time: 0.7071 data_time: 0.0015 memory: 15585 loss: 0.0120 +2023/06/06 03:36:55 - mmengine - INFO - Epoch(train) [6][2000/2046] lr: 1.4608e-04 eta: 1:44:04 time: 0.6970 data_time: 0.0016 memory: 15585 loss: 0.0037 +2023/06/06 03:37:28 - mmengine - INFO - Exp name: convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1_20230606_005643 +2023/06/06 03:37:28 - mmengine - INFO - Saving checkpoint at 6 epochs +2023/06/06 03:38:26 - mmengine - INFO - Epoch(val) [6][60/60] accuracy/top1: 85.7836 single-label/precision_classwise: [99.95637512207031, 60.0438117980957] single-label/recall_classwise: [81.96031188964844, 99.86821746826172] single-label/f1-score_classwise: [90.06822204589844, 74.99708557128906] data_time: 0.1129 time: 0.5116 +2023/06/06 03:39:50 - mmengine - INFO - Epoch(train) [7][ 100/2046] lr: 1.4193e-04 eta: 1:42:17 time: 0.7381 data_time: 0.0916 memory: 15585 loss: 0.0052 +2023/06/06 03:41:03 - mmengine - INFO - Epoch(train) [7][ 200/2046] lr: 1.3910e-04 eta: 1:41:00 time: 0.7429 data_time: 0.2282 memory: 15585 loss: 0.0080 +2023/06/06 03:42:20 - mmengine - INFO - Epoch(train) [7][ 300/2046] lr: 1.3629e-04 eta: 1:39:45 time: 0.8034 data_time: 0.0015 memory: 15585 loss: 0.0072 +2023/06/06 03:43:37 - mmengine - INFO - Epoch(train) [7][ 400/2046] lr: 1.3350e-04 eta: 1:38:29 time: 0.7963 data_time: 0.0016 memory: 15585 loss: 0.0103 +2023/06/06 03:44:52 - mmengine - INFO - Epoch(train) [7][ 500/2046] lr: 1.3072e-04 eta: 1:37:13 time: 0.7438 data_time: 0.0017 memory: 15585 loss: 0.0059 +2023/06/06 03:46:05 - mmengine - INFO - Epoch(train) [7][ 600/2046] lr: 1.2796e-04 eta: 1:35:55 time: 0.7304 data_time: 0.0015 memory: 15585 loss: 0.0080 +2023/06/06 03:47:20 - mmengine - INFO - Epoch(train) [7][ 700/2046] lr: 1.2522e-04 eta: 1:34:39 time: 0.7707 data_time: 0.0017 memory: 15585 loss: 0.0085 +2023/06/06 03:47:38 - mmengine - INFO - Exp name: convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1_20230606_005643 +2023/06/06 03:48:36 - mmengine - INFO - Epoch(train) [7][ 800/2046] lr: 1.2250e-04 eta: 1:33:23 time: 0.7316 data_time: 0.0015 memory: 15585 loss: 0.0044 +2023/06/06 03:49:56 - mmengine - INFO - Epoch(train) [7][ 900/2046] lr: 1.1980e-04 eta: 1:32:09 time: 0.7846 data_time: 0.0015 memory: 15585 loss: 0.0054 +2023/06/06 03:51:09 - mmengine - INFO - Epoch(train) [7][1000/2046] lr: 1.1711e-04 eta: 1:30:52 time: 0.7386 data_time: 0.0015 memory: 15585 loss: 0.0100 +2023/06/06 03:52:24 - mmengine - INFO - Epoch(train) [7][1100/2046] lr: 1.1445e-04 eta: 1:29:36 time: 0.7163 data_time: 0.0016 memory: 15585 loss: 0.0089 +2023/06/06 03:53:39 - mmengine - INFO - Epoch(train) [7][1200/2046] lr: 1.1181e-04 eta: 1:28:19 time: 0.7818 data_time: 0.0016 memory: 15585 loss: 0.0036 +2023/06/06 03:54:52 - mmengine - INFO - Epoch(train) [7][1300/2046] lr: 1.0919e-04 eta: 1:27:02 time: 0.6835 data_time: 0.0016 memory: 15585 loss: 0.0053 +2023/06/06 03:56:06 - mmengine - INFO - Epoch(train) [7][1400/2046] lr: 1.0659e-04 eta: 1:25:45 time: 0.7692 data_time: 0.0018 memory: 15585 loss: 0.0058 +2023/06/06 03:57:24 - mmengine - INFO - Epoch(train) [7][1500/2046] lr: 1.0402e-04 eta: 1:24:30 time: 0.7656 data_time: 0.0017 memory: 15585 loss: 0.0096 +2023/06/06 03:58:38 - mmengine - INFO - Epoch(train) [7][1600/2046] lr: 1.0147e-04 eta: 1:23:13 time: 0.7026 data_time: 0.0016 memory: 15585 loss: 0.0060 +2023/06/06 03:59:51 - mmengine - INFO - Epoch(train) [7][1700/2046] lr: 9.8946e-05 eta: 1:21:56 time: 0.6843 data_time: 0.0017 memory: 15585 loss: 0.0073 +2023/06/06 04:00:08 - mmengine - INFO - Exp name: convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1_20230606_005643 +2023/06/06 04:01:04 - mmengine - INFO - Epoch(train) [7][1800/2046] lr: 9.6446e-05 eta: 1:20:39 time: 0.7048 data_time: 0.0017 memory: 15585 loss: 0.0078 +2023/06/06 04:02:15 - mmengine - INFO - Epoch(train) [7][1900/2046] lr: 9.3972e-05 eta: 1:19:21 time: 0.7080 data_time: 0.0015 memory: 15585 loss: 0.0089 +2023/06/06 04:03:28 - mmengine - INFO - Epoch(train) [7][2000/2046] lr: 9.1523e-05 eta: 1:18:04 time: 0.7416 data_time: 0.0017 memory: 15585 loss: 0.0046 +2023/06/06 04:03:59 - mmengine - INFO - Exp name: convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1_20230606_005643 +2023/06/06 04:03:59 - mmengine - INFO - Saving checkpoint at 7 epochs +2023/06/06 04:04:57 - mmengine - INFO - Epoch(val) [7][60/60] accuracy/top1: 86.3115 single-label/precision_classwise: [99.93638610839844, 60.958290100097656] single-label/recall_classwise: [82.64839935302734, 99.80619812011719] single-label/f1-score_classwise: [90.47393798828125, 75.68854522705078] data_time: 0.1092 time: 0.5069 +2023/06/06 04:06:15 - mmengine - INFO - Epoch(train) [8][ 100/2046] lr: 8.7997e-05 eta: 1:16:13 time: 0.7495 data_time: 0.1354 memory: 15585 loss: 0.0069 +2023/06/06 04:07:28 - mmengine - INFO - Epoch(train) [8][ 200/2046] lr: 8.5616e-05 eta: 1:14:56 time: 0.7030 data_time: 0.0343 memory: 15585 loss: 0.0072 +2023/06/06 04:08:42 - mmengine - INFO - Epoch(train) [8][ 300/2046] lr: 8.3262e-05 eta: 1:13:40 time: 0.7513 data_time: 0.0018 memory: 15585 loss: 0.0047 +2023/06/06 04:09:57 - mmengine - INFO - Epoch(train) [8][ 400/2046] lr: 8.0938e-05 eta: 1:12:24 time: 0.6823 data_time: 0.0016 memory: 15585 loss: 0.0036 +2023/06/06 04:11:11 - mmengine - INFO - Epoch(train) [8][ 500/2046] lr: 7.8642e-05 eta: 1:11:07 time: 0.7327 data_time: 0.0015 memory: 15585 loss: 0.0043 +2023/06/06 04:12:24 - mmengine - INFO - Epoch(train) [8][ 600/2046] lr: 7.6377e-05 eta: 1:09:51 time: 0.7514 data_time: 0.0018 memory: 15585 loss: 0.0091 +2023/06/06 04:13:21 - mmengine - INFO - Exp name: convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1_20230606_005643 +2023/06/06 04:13:39 - mmengine - INFO - Epoch(train) [8][ 700/2046] lr: 7.4142e-05 eta: 1:08:35 time: 0.9306 data_time: 0.0014 memory: 15585 loss: 0.0046 +2023/06/06 04:14:59 - mmengine - INFO - Epoch(train) [8][ 800/2046] lr: 7.1937e-05 eta: 1:07:21 time: 0.7215 data_time: 0.0017 memory: 15585 loss: 0.0048 +2023/06/06 04:16:13 - mmengine - INFO - Epoch(train) [8][ 900/2046] lr: 6.9764e-05 eta: 1:06:04 time: 0.7434 data_time: 0.0018 memory: 15585 loss: 0.0061 +2023/06/06 04:17:25 - mmengine - INFO - Epoch(train) [8][1000/2046] lr: 6.7623e-05 eta: 1:04:48 time: 0.7571 data_time: 0.0017 memory: 15585 loss: 0.0095 +2023/06/06 04:18:40 - mmengine - INFO - Epoch(train) [8][1100/2046] lr: 6.5515e-05 eta: 1:03:32 time: 0.7015 data_time: 0.0016 memory: 15585 loss: 0.0069 +2023/06/06 04:19:53 - mmengine - INFO - Epoch(train) [8][1200/2046] lr: 6.3439e-05 eta: 1:02:15 time: 0.7101 data_time: 0.0016 memory: 15585 loss: 0.0062 +2023/06/06 04:21:06 - mmengine - INFO - Epoch(train) [8][1300/2046] lr: 6.1396e-05 eta: 1:00:59 time: 0.7369 data_time: 0.0015 memory: 15585 loss: 0.0051 +2023/06/06 04:22:21 - mmengine - INFO - Epoch(train) [8][1400/2046] lr: 5.9388e-05 eta: 0:59:43 time: 0.7467 data_time: 0.0021 memory: 15585 loss: 0.0039 +2023/06/06 04:23:36 - mmengine - INFO - Epoch(train) [8][1500/2046] lr: 5.7413e-05 eta: 0:58:27 time: 0.7236 data_time: 0.0018 memory: 15585 loss: 0.0059 +2023/06/06 04:24:51 - mmengine - INFO - Epoch(train) [8][1600/2046] lr: 5.5474e-05 eta: 0:57:11 time: 0.6820 data_time: 0.0018 memory: 15585 loss: 0.0043 +2023/06/06 04:25:51 - mmengine - INFO - Exp name: convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1_20230606_005643 +2023/06/06 04:26:07 - mmengine - INFO - Epoch(train) [8][1700/2046] lr: 5.3570e-05 eta: 0:55:56 time: 0.7614 data_time: 0.0016 memory: 15585 loss: 0.0069 +2023/06/06 04:27:26 - mmengine - INFO - Epoch(train) [8][1800/2046] lr: 5.1701e-05 eta: 0:54:41 time: 0.8009 data_time: 0.0016 memory: 15585 loss: 0.0071 +2023/06/06 04:28:42 - mmengine - INFO - Epoch(train) [8][1900/2046] lr: 4.9869e-05 eta: 0:53:25 time: 0.7648 data_time: 0.0016 memory: 15585 loss: 0.0051 +2023/06/06 04:30:00 - mmengine - INFO - Epoch(train) [8][2000/2046] lr: 4.8073e-05 eta: 0:52:10 time: 0.7756 data_time: 0.0017 memory: 15585 loss: 0.0070 +2023/06/06 04:30:33 - mmengine - INFO - Exp name: convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1_20230606_005643 +2023/06/06 04:30:33 - mmengine - INFO - Saving checkpoint at 8 epochs +2023/06/06 04:31:31 - mmengine - INFO - Epoch(val) [8][60/60] accuracy/top1: 86.8858 single-label/precision_classwise: [99.95963287353516, 61.96614074707031] single-label/recall_classwise: [83.35963439941406, 99.87596893310547] single-label/f1-score_classwise: [90.90805053710938, 76.48106384277344] data_time: 0.1137 time: 0.5121 +2023/06/06 04:32:49 - mmengine - INFO - Epoch(train) [9][ 100/2046] lr: 4.5518e-05 eta: 0:50:20 time: 0.7079 data_time: 0.1924 memory: 15585 loss: 0.0042 +2023/06/06 04:34:07 - mmengine - INFO - Epoch(train) [9][ 200/2046] lr: 4.3814e-05 eta: 0:49:05 time: 0.8093 data_time: 0.0017 memory: 15585 loss: 0.0065 +2023/06/06 04:35:25 - mmengine - INFO - Epoch(train) [9][ 300/2046] lr: 4.2148e-05 eta: 0:47:50 time: 0.7897 data_time: 0.0015 memory: 15585 loss: 0.0056 +2023/06/06 04:36:54 - mmengine - INFO - Epoch(train) [9][ 400/2046] lr: 4.0520e-05 eta: 0:46:37 time: 0.7985 data_time: 0.0015 memory: 15585 loss: 0.0033 +2023/06/06 04:38:10 - mmengine - INFO - Epoch(train) [9][ 500/2046] lr: 3.8931e-05 eta: 0:45:21 time: 0.7752 data_time: 0.0016 memory: 15585 loss: 0.0058 +2023/06/06 04:39:27 - mmengine - INFO - Epoch(train) [9][ 600/2046] lr: 3.7381e-05 eta: 0:44:06 time: 0.7823 data_time: 0.0017 memory: 15585 loss: 0.0076 +2023/06/06 04:39:51 - mmengine - INFO - Exp name: convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1_20230606_005643 +2023/06/06 04:40:43 - mmengine - INFO - Epoch(train) [9][ 700/2046] lr: 3.5871e-05 eta: 0:42:50 time: 0.7845 data_time: 0.0017 memory: 15585 loss: 0.0065 +2023/06/06 04:41:59 - mmengine - INFO - Epoch(train) [9][ 800/2046] lr: 3.4401e-05 eta: 0:41:34 time: 0.7427 data_time: 0.0015 memory: 15585 loss: 0.0056 +2023/06/06 04:43:15 - mmengine - INFO - Epoch(train) [9][ 900/2046] lr: 3.2971e-05 eta: 0:40:19 time: 0.7652 data_time: 0.0015 memory: 15585 loss: 0.0051 +2023/06/06 04:44:32 - mmengine - INFO - Epoch(train) [9][1000/2046] lr: 3.1581e-05 eta: 0:39:03 time: 0.7513 data_time: 0.0016 memory: 15585 loss: 0.0048 +2023/06/06 04:45:48 - mmengine - INFO - Epoch(train) [9][1100/2046] lr: 3.0232e-05 eta: 0:37:47 time: 0.7822 data_time: 0.0015 memory: 15585 loss: 0.0049 +2023/06/06 04:47:05 - mmengine - INFO - Epoch(train) [9][1200/2046] lr: 2.8925e-05 eta: 0:36:32 time: 0.7657 data_time: 0.0015 memory: 15585 loss: 0.0055 +2023/06/06 04:48:24 - mmengine - INFO - Epoch(train) [9][1300/2046] lr: 2.7659e-05 eta: 0:35:16 time: 0.7962 data_time: 0.0016 memory: 15585 loss: 0.0071 +2023/06/06 04:49:40 - mmengine - INFO - Epoch(train) [9][1400/2046] lr: 2.6435e-05 eta: 0:34:00 time: 0.7714 data_time: 0.0016 memory: 15585 loss: 0.0060 +2023/06/06 04:50:55 - mmengine - INFO - Epoch(train) [9][1500/2046] lr: 2.5253e-05 eta: 0:32:45 time: 0.7584 data_time: 0.0017 memory: 15585 loss: 0.0090 +2023/06/06 04:52:11 - mmengine - INFO - Epoch(train) [9][1600/2046] lr: 2.4113e-05 eta: 0:31:29 time: 0.7131 data_time: 0.0017 memory: 15585 loss: 0.0073 +2023/06/06 04:52:34 - mmengine - INFO - Exp name: convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1_20230606_005643 +2023/06/06 04:53:27 - mmengine - INFO - Epoch(train) [9][1700/2046] lr: 2.3016e-05 eta: 0:30:13 time: 0.7824 data_time: 0.0016 memory: 15585 loss: 0.0069 +2023/06/06 04:54:49 - mmengine - INFO - Epoch(train) [9][1800/2046] lr: 2.1962e-05 eta: 0:28:58 time: 0.7645 data_time: 0.0016 memory: 15585 loss: 0.0031 +2023/06/06 04:56:05 - mmengine - INFO - Epoch(train) [9][1900/2046] lr: 2.0951e-05 eta: 0:27:42 time: 0.7386 data_time: 0.0015 memory: 15585 loss: 0.0032 +2023/06/06 04:57:21 - mmengine - INFO - Epoch(train) [9][2000/2046] lr: 1.9984e-05 eta: 0:26:26 time: 0.7436 data_time: 0.0016 memory: 15585 loss: 0.0059 +2023/06/06 04:57:54 - mmengine - INFO - Exp name: convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1_20230606_005643 +2023/06/06 04:57:54 - mmengine - INFO - Saving checkpoint at 9 epochs +2023/06/06 04:58:52 - mmengine - INFO - Epoch(val) [9][60/60] accuracy/top1: 87.1456 single-label/precision_classwise: [99.94973754882812, 62.44242858886719] single-label/recall_classwise: [83.69842529296875, 99.84496307373047] single-label/f1-score_classwise: [91.10503387451172, 76.83360290527344] data_time: 0.1032 time: 0.5024 +2023/06/06 05:00:10 - mmengine - INFO - Epoch(train) [10][ 100/2046] lr: 1.8649e-05 eta: 0:24:36 time: 0.7176 data_time: 0.0015 memory: 15585 loss: 0.0068 +2023/06/06 05:01:23 - mmengine - INFO - Epoch(train) [10][ 200/2046] lr: 1.7790e-05 eta: 0:23:20 time: 0.7045 data_time: 0.0017 memory: 15585 loss: 0.0076 +2023/06/06 05:02:43 - mmengine - INFO - Epoch(train) [10][ 300/2046] lr: 1.6974e-05 eta: 0:22:04 time: 0.7591 data_time: 0.0016 memory: 15585 loss: 0.0031 +2023/06/06 05:03:59 - mmengine - INFO - Epoch(train) [10][ 400/2046] lr: 1.6202e-05 eta: 0:20:48 time: 0.7902 data_time: 0.0016 memory: 15585 loss: 0.0018 +2023/06/06 05:05:15 - mmengine - INFO - Epoch(train) [10][ 500/2046] lr: 1.5476e-05 eta: 0:19:32 time: 0.7735 data_time: 0.0018 memory: 15585 loss: 0.0035 +2023/06/06 05:06:19 - mmengine - INFO - Exp name: convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1_20230606_005643 +2023/06/06 05:06:31 - mmengine - INFO - Epoch(train) [10][ 600/2046] lr: 1.4793e-05 eta: 0:18:17 time: 0.7880 data_time: 0.0015 memory: 15585 loss: 0.0075 +2023/06/06 05:07:47 - mmengine - INFO - Epoch(train) [10][ 700/2046] lr: 1.4156e-05 eta: 0:17:01 time: 0.7852 data_time: 0.0016 memory: 15585 loss: 0.0054 +2023/06/06 05:09:04 - mmengine - INFO - Epoch(train) [10][ 800/2046] lr: 1.3564e-05 eta: 0:15:45 time: 0.7876 data_time: 0.0016 memory: 15585 loss: 0.0046 +2023/06/06 05:10:21 - mmengine - INFO - Epoch(train) [10][ 900/2046] lr: 1.3016e-05 eta: 0:14:29 time: 0.8062 data_time: 0.0018 memory: 15585 loss: 0.0086 +2023/06/06 05:11:35 - mmengine - INFO - Epoch(train) [10][1000/2046] lr: 1.2514e-05 eta: 0:13:13 time: 0.8434 data_time: 0.0019 memory: 15585 loss: 0.0064 +2023/06/06 05:12:49 - mmengine - INFO - Epoch(train) [10][1100/2046] lr: 1.2058e-05 eta: 0:11:57 time: 0.7644 data_time: 0.0015 memory: 15585 loss: 0.0061 +2023/06/06 05:14:07 - mmengine - INFO - Epoch(train) [10][1200/2046] lr: 1.1647e-05 eta: 0:10:41 time: 0.8092 data_time: 0.0016 memory: 15585 loss: 0.0047 +2023/06/06 05:15:23 - mmengine - INFO - Epoch(train) [10][1300/2046] lr: 1.1281e-05 eta: 0:09:25 time: 0.7980 data_time: 0.0017 memory: 15585 loss: 0.0061 +2023/06/06 05:16:39 - mmengine - INFO - Epoch(train) [10][1400/2046] lr: 1.0961e-05 eta: 0:08:10 time: 0.7399 data_time: 0.0018 memory: 15585 loss: 0.0083 +2023/06/06 05:17:57 - mmengine - INFO - Epoch(train) [10][1500/2046] lr: 1.0687e-05 eta: 0:06:54 time: 0.7104 data_time: 0.0015 memory: 15585 loss: 0.0065 +2023/06/06 05:19:07 - mmengine - INFO - Exp name: convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1_20230606_005643 +2023/06/06 05:19:20 - mmengine - INFO - Epoch(train) [10][1600/2046] lr: 1.0459e-05 eta: 0:05:38 time: 0.8109 data_time: 0.0017 memory: 15585 loss: 0.0043 +2023/06/06 05:20:44 - mmengine - INFO - Epoch(train) [10][1700/2046] lr: 1.0277e-05 eta: 0:04:22 time: 0.7935 data_time: 0.0015 memory: 15585 loss: 0.0023 +2023/06/06 05:22:02 - mmengine - INFO - Epoch(train) [10][1800/2046] lr: 1.0140e-05 eta: 0:03:06 time: 0.7602 data_time: 0.0016 memory: 15585 loss: 0.0034 +2023/06/06 05:23:19 - mmengine - INFO - Epoch(train) [10][1900/2046] lr: 1.0050e-05 eta: 0:01:50 time: 0.7593 data_time: 0.0016 memory: 15585 loss: 0.0048 +2023/06/06 05:24:43 - mmengine - INFO - Epoch(train) [10][2000/2046] lr: 1.0005e-05 eta: 0:00:34 time: 0.7175 data_time: 0.0017 memory: 15585 loss: 0.0061 +2023/06/06 05:25:20 - mmengine - INFO - Exp name: convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1_20230606_005643 +2023/06/06 05:25:20 - mmengine - INFO - Saving checkpoint at 10 epochs +2023/06/06 05:26:18 - mmengine - INFO - Epoch(val) [10][60/60] accuracy/top1: 87.4832 single-label/precision_classwise: [99.93502044677734, 63.07383346557617] single-label/recall_classwise: [84.14031219482422, 99.79844665527344] single-label/f1-score_classwise: [91.36002349853516, 77.29578399658203] data_time: 0.1258 time: 0.5219 diff --git a/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1/20230606_005643/vis_data/20230606_005643.json b/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1/20230606_005643/vis_data/20230606_005643.json new file mode 100644 index 0000000000000000000000000000000000000000..92bb2340f18c97f6e63b3d86022d9069cd7b6822 --- /dev/null +++ b/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1/20230606_005643/vis_data/20230606_005643.json @@ -0,0 +1,210 @@ +{"lr": 0.00039997747031453974, "data_time": 0.0017314672470092774, "loss": 0.04582378547638655, "time": 0.7797232389450073, "epoch": 1, "memory": 15587, "step": 100} +{"lr": 0.0003999089739988646, "data_time": 0.001805877685546875, "loss": 0.04913739897310734, "time": 0.738232421875, "epoch": 1, "memory": 15587, "step": 200} +{"lr": 0.00039979452490327365, "data_time": 0.19394207000732422, "loss": 0.03367728535085916, "time": 0.7865641832351684, "epoch": 1, "memory": 15587, "step": 300} +{"lr": 0.0003996341500108966, "data_time": 0.0015164613723754883, "loss": 0.02638430558145046, "time": 0.8205552816390991, "epoch": 1, "memory": 15587, "step": 400} +{"lr": 0.00039942788713257383, "data_time": 0.001667618751525879, "loss": 0.03231051415205002, "time": 0.7348530769348145, "epoch": 1, "memory": 15587, "step": 500} +{"lr": 0.00039917578489794236, "data_time": 0.0015869855880737305, "loss": 0.03607837650924921, "time": 0.8029991388320923, "epoch": 1, "memory": 15587, "step": 600} +{"lr": 0.00039887790274397046, "data_time": 0.0016368865966796876, "loss": 0.030855814926326275, "time": 0.736802339553833, "epoch": 1, "memory": 15587, "step": 700} +{"lr": 0.00039853431090094445, "data_time": 0.001633024215698242, "loss": 0.028339687455445527, "time": 0.7840763330459595, "epoch": 1, "memory": 15587, "step": 800} +{"lr": 0.0003981450903759111, "data_time": 0.0016636610031127929, "loss": 0.032328211702406406, "time": 0.7560638904571533, "epoch": 1, "memory": 15587, "step": 900} +{"lr": 0.00039771033293357867, "data_time": 0.0015117883682250976, "loss": 0.024187591020017864, "time": 0.7417938470840454, "epoch": 1, "memory": 15587, "step": 1000} +{"lr": 0.0003972301410746815, "data_time": 0.00158998966217041, "loss": 0.028163632191717625, "time": 0.7944855690002441, "epoch": 1, "memory": 15587, "step": 1100} +{"lr": 0.00039670462801181557, "data_time": 0.0015922307968139649, "loss": 0.025045361928641797, "time": 0.7637933015823364, "epoch": 1, "memory": 15587, "step": 1200} +{"lr": 0.0003961339176427448, "data_time": 0.0023871898651123048, "loss": 0.017230405099689962, "time": 0.8597577810287476, "epoch": 1, "memory": 15587, "step": 1300} +{"lr": 0.00039551814452119214, "data_time": 0.0015121221542358399, "loss": 0.026982803316786885, "time": 0.7743359804153442, "epoch": 1, "memory": 15587, "step": 1400} +{"lr": 0.00039485745382511467, "data_time": 0.0016725778579711914, "loss": 0.019004460936412216, "time": 0.8123712062835693, "epoch": 1, "memory": 15587, "step": 1500} +{"lr": 0.0003941520013224769, "data_time": 0.0013941049575805664, "loss": 0.024567633122205734, "time": 0.7806166172027588, "epoch": 1, "memory": 15587, "step": 1600} +{"lr": 0.0003934019533345259, "data_time": 0.0014190196990966797, "loss": 0.019510003039613366, "time": 0.7741846323013306, "epoch": 1, "memory": 15587, "step": 1700} +{"lr": 0.00039260748669657817, "data_time": 0.001448655128479004, "loss": 0.018855373607948422, "time": 0.781554388999939, "epoch": 1, "memory": 15587, "step": 1800} +{"lr": 0.000391768788716328, "data_time": 0.0013930082321166992, "loss": 0.017342483764514326, "time": 0.7650550842285156, "epoch": 1, "memory": 15587, "step": 1900} +{"lr": 0.00039088605712968733, "data_time": 0.0015112400054931641, "loss": 0.014465322950854897, "time": 0.775104570388794, "epoch": 1, "memory": 15587, "step": 2000} +{"accuracy/top1": 50.69261932373047, "data_time": 0.12550911903381348, "time": 0.5630749146143595, "step": 1} +{"lr": 0.0003895186269690071, "data_time": 0.035175633430480954, "loss": 0.020252875657752155, "time": 0.7574407815933227, "epoch": 2, "memory": 15587, "step": 2146} +{"lr": 0.0003885284798052241, "data_time": 0.01851482391357422, "loss": 0.024551139771938325, "time": 0.7844219207763672, "epoch": 2, "memory": 15585, "step": 2246} +{"lr": 0.0003874950629875727, "data_time": 0.001478862762451172, "loss": 0.014201085036620497, "time": 0.7822738409042358, "epoch": 2, "memory": 15585, "step": 2346} +{"lr": 0.0003864186201599172, "data_time": 0.0019092321395874023, "loss": 0.022182260919362307, "time": 0.7561484575271606, "epoch": 2, "memory": 15585, "step": 2446} +{"lr": 0.0003852994051101632, "data_time": 0.0015421152114868165, "loss": 0.014758217940106988, "time": 0.7986104726791382, "epoch": 2, "memory": 15585, "step": 2546} +{"lr": 0.0003841376817104243, "data_time": 0.0015173673629760743, "loss": 0.015461743343621492, "time": 0.7966030597686767, "epoch": 2, "memory": 15585, "step": 2646} +{"lr": 0.0003829337238548097, "data_time": 0.001443791389465332, "loss": 0.012522991793230177, "time": 0.740956974029541, "epoch": 2, "memory": 15585, "step": 2746} +{"lr": 0.0003816878153948477, "data_time": 0.0015905141830444337, "loss": 0.015928221913054586, "time": 0.709771990776062, "epoch": 2, "memory": 15585, "step": 2846} +{"lr": 0.0003804002500725666, "data_time": 0.0014972925186157227, "loss": 0.020683481032028793, "time": 0.7683061361312866, "epoch": 2, "memory": 15585, "step": 2946} +{"lr": 0.0003790713314512378, "data_time": 0.0016095876693725587, "loss": 0.019671098375692964, "time": 0.7312532663345337, "epoch": 2, "memory": 15585, "step": 3046} +{"lr": 0.00037770137284380876, "data_time": 0.0016412973403930665, "loss": 0.02124711577780545, "time": 0.7730305433273316, "epoch": 2, "memory": 15585, "step": 3146} +{"lr": 0.00037629069723903126, "data_time": 0.001612401008605957, "loss": 0.019148432463407517, "time": 0.7195446729660034, "epoch": 2, "memory": 15585, "step": 3246} +{"lr": 0.0003748396372253147, "data_time": 0.001468658447265625, "loss": 0.01650288375094533, "time": 0.764950156211853, "epoch": 2, "memory": 15585, "step": 3346} +{"lr": 0.0003733485349123126, "data_time": 0.00144503116607666, "loss": 0.013821152597665786, "time": 0.7550675868988037, "epoch": 2, "memory": 15585, "step": 3446} +{"lr": 0.0003718177418502636, "data_time": 0.001757049560546875, "loss": 0.01906374916434288, "time": 0.7818350791931152, "epoch": 2, "memory": 15585, "step": 3546} +{"lr": 0.0003702476189471092, "data_time": 0.0016087532043457032, "loss": 0.01476140357553959, "time": 0.8163014650344849, "epoch": 2, "memory": 15585, "step": 3646} +{"lr": 0.00036863853638340543, "data_time": 0.0014557361602783204, "loss": 0.012234715931117534, "time": 0.7612666845321655, "epoch": 2, "memory": 15585, "step": 3746} +{"lr": 0.000366990873525045, "data_time": 0.0015973091125488282, "loss": 0.01461083865724504, "time": 0.7513570547103882, "epoch": 2, "memory": 15585, "step": 3846} +{"lr": 0.00036530501883381603, "data_time": 0.0015964746475219727, "loss": 0.008052535681053996, "time": 0.6864346742630005, "epoch": 2, "memory": 15585, "step": 3946} +{"lr": 0.00036358136977581845, "data_time": 0.0018983840942382812, "loss": 0.016030167089775206, "time": 0.7498691082000732, "epoch": 2, "memory": 15585, "step": 4046} +{"accuracy/top1": 82.29316711425781, "data_time": 0.1093459246588535, "time": 0.5076444539867464, "step": 2} +{"lr": 0.0003609978147988783, "data_time": 0.06281070709228516, "loss": 0.019261301029473543, "time": 0.7321836471557617, "epoch": 3, "memory": 15585, "step": 4192} +{"lr": 0.00035918293929014, "data_time": 0.001866316795349121, "loss": 0.013484583934769034, "time": 0.7675493717193603, "epoch": 3, "memory": 15585, "step": 4292} +{"lr": 0.0003573317127887786, "data_time": 0.0016434192657470703, "loss": 0.018350797798484565, "time": 0.7364666223526001, "epoch": 3, "memory": 15585, "step": 4392} +{"lr": 0.000355444571749834, "data_time": 0.0015516996383666993, "loss": 0.013195968139916659, "time": 0.7499399185180664, "epoch": 3, "memory": 15585, "step": 4492} +{"lr": 0.0003535219610957493, "data_time": 0.0015429258346557617, "loss": 0.015334970876574516, "time": 0.7414395809173584, "epoch": 3, "memory": 15585, "step": 4592} +{"lr": 0.00035156433411147255, "data_time": 0.0014833688735961914, "loss": 0.01002070913091302, "time": 0.7465383768081665, "epoch": 3, "memory": 15585, "step": 4692} +{"lr": 0.00034957215233758935, "data_time": 0.0015995025634765625, "loss": 0.014274924341589213, "time": 0.7002089977264404, "epoch": 3, "memory": 15585, "step": 4792} +{"lr": 0.00034754588546150636, "data_time": 0.001563286781311035, "loss": 0.012178977811709046, "time": 0.7269941806793213, "epoch": 3, "memory": 15585, "step": 4892} +{"lr": 0.0003454860112067171, "data_time": 0.0015238523483276367, "loss": 0.017030299128964543, "time": 0.7481244087219239, "epoch": 3, "memory": 15585, "step": 4992} +{"lr": 0.00034339301522016903, "data_time": 0.0015039920806884765, "loss": 0.012458660826086997, "time": 0.7832327604293823, "epoch": 3, "memory": 15585, "step": 5092} +{"lr": 0.0003412673909577655, "data_time": 0.0016904354095458984, "loss": 0.012210268201306463, "time": 0.7109328031539917, "epoch": 3, "memory": 15585, "step": 5192} +{"lr": 0.00033910963956802737, "data_time": 0.00177459716796875, "loss": 0.01046714880503714, "time": 0.7497158050537109, "epoch": 3, "memory": 15585, "step": 5292} +{"lr": 0.00033692026977393767, "data_time": 0.0015202760696411133, "loss": 0.010511100618168711, "time": 0.7057794332504272, "epoch": 3, "memory": 15585, "step": 5392} +{"lr": 0.0003346997977530036, "data_time": 0.0015427827835083007, "loss": 0.011450219480320812, "time": 0.666927170753479, "epoch": 3, "memory": 15585, "step": 5492} +{"lr": 0.0003324487470155584, "data_time": 0.0015279769897460938, "loss": 0.017318726144731046, "time": 0.7419221878051758, "epoch": 3, "memory": 15585, "step": 5592} +{"lr": 0.0003301676482813382, "data_time": 0.0017217874526977539, "loss": 0.008253357093781232, "time": 0.7053928375244141, "epoch": 3, "memory": 15585, "step": 5692} +{"lr": 0.00032785703935435476, "data_time": 0.0016490697860717773, "loss": 0.010431417217478155, "time": 0.8212690114974975, "epoch": 3, "memory": 15585, "step": 5792} +{"lr": 0.00032551746499609967, "data_time": 0.0015106201171875, "loss": 0.013492595544084907, "time": 0.7035637140274048, "epoch": 3, "memory": 15585, "step": 5892} +{"lr": 0.0003231494767971101, "data_time": 0.0015542745590209962, "loss": 0.012716319225728513, "time": 0.7209977626800537, "epoch": 3, "memory": 15585, "step": 5992} +{"lr": 0.0003207536330469211, "data_time": 0.001537013053894043, "loss": 0.009736620634794236, "time": 0.7016186952590943, "epoch": 3, "memory": 15585, "step": 6092} +{"accuracy/top1": 86.2817153930664, "data_time": 0.111266233881966, "time": 0.5068588843111133, "step": 3} +{"lr": 0.000317206849615549, "data_time": 0.20003046989440917, "loss": 0.010455530788749456, "time": 0.7208166360855103, "epoch": 4, "memory": 15585, "step": 6238} +{"lr": 0.00031474489842163173, "data_time": 0.1849517345428467, "loss": 0.01119880867190659, "time": 0.7366448879241944, "epoch": 4, "memory": 15585, "step": 6338} +{"lr": 0.0003122570731847668, "data_time": 0.030050158500671387, "loss": 0.011106752138584853, "time": 0.7972426414489746, "epoch": 4, "memory": 15585, "step": 6438} +{"lr": 0.00030974396044791, "data_time": 0.0014658451080322265, "loss": 0.020003657322376966, "time": 0.7284992456436157, "epoch": 4, "memory": 15585, "step": 6538} +{"lr": 0.0003072061527159319, "data_time": 0.001621842384338379, "loss": 0.011039762571454049, "time": 0.7132418394088745, "epoch": 4, "memory": 15585, "step": 6638} +{"lr": 0.0003046442483159274, "data_time": 0.0017466068267822266, "loss": 0.02424939088523388, "time": 0.7078982591629028, "epoch": 4, "memory": 15585, "step": 6738} +{"lr": 0.0003020588512561505, "data_time": 0.001969766616821289, "loss": 0.008156781597062946, "time": 0.7495105743408204, "epoch": 4, "memory": 15585, "step": 6838} +{"lr": 0.0002994505710836101, "data_time": 0.0015324831008911132, "loss": 0.010514680109918118, "time": 0.7136235237121582, "epoch": 4, "memory": 15585, "step": 6938} +{"lr": 0.0002968200227403596, "data_time": 0.0015955686569213868, "loss": 0.0111268509645015, "time": 0.7207844018936157, "epoch": 4, "memory": 15585, "step": 7038} +{"lr": 0.0002941678264185154, "data_time": 0.0014647245407104492, "loss": 0.015258364658802748, "time": 0.7219701528549194, "epoch": 4, "memory": 15585, "step": 7138} +{"lr": 0.0002914946074140367, "data_time": 0.0015655517578125, "loss": 0.011088904459029436, "time": 0.7023492813110351, "epoch": 4, "memory": 15585, "step": 7238} +{"lr": 0.00028880099597930275, "data_time": 0.0016550302505493164, "loss": 0.006905767740681768, "time": 0.7222923994064331, "epoch": 4, "memory": 15585, "step": 7338} +{"lr": 0.0002860876271745208, "data_time": 0.0018896341323852539, "loss": 0.0085047650616616, "time": 0.7889053821563721, "epoch": 4, "memory": 15585, "step": 7438} +{"lr": 0.0002833551407180016, "data_time": 0.0015702962875366211, "loss": 0.012262030225247144, "time": 0.9031259775161743, "epoch": 4, "memory": 15585, "step": 7538} +{"lr": 0.00028060418083533477, "data_time": 0.0015096664428710938, "loss": 0.011341116577386855, "time": 0.7258390188217163, "epoch": 4, "memory": 15585, "step": 7638} +{"lr": 0.0002778353961075036, "data_time": 0.0019012689590454102, "loss": 0.010915554920211434, "time": 0.7604675531387329, "epoch": 4, "memory": 15585, "step": 7738} +{"lr": 0.0002750494393179725, "data_time": 0.0017791986465454102, "loss": 0.015977052552625538, "time": 0.6591270446777344, "epoch": 4, "memory": 15585, "step": 7838} +{"lr": 0.00027224696729878206, "data_time": 0.0016847610473632812, "loss": 0.017442272650077938, "time": 0.77463698387146, "epoch": 4, "memory": 15585, "step": 7938} +{"lr": 0.00026942864077569254, "data_time": 0.0015532970428466797, "loss": 0.018840466765686868, "time": 0.7723486661911011, "epoch": 4, "memory": 15585, "step": 8038} +{"lr": 0.0002665951242124057, "data_time": 0.0015737295150756836, "loss": 0.011894524283707141, "time": 0.7510597467422485, "epoch": 4, "memory": 15585, "step": 8138} +{"accuracy/top1": 85.96726989746094, "data_time": 0.10596991367027407, "time": 0.5014197787300485, "step": 4} +{"lr": 0.0002624322962007581, "data_time": 0.19933886528015138, "loss": 0.010770977195352315, "time": 0.7168541193008423, "epoch": 5, "memory": 15585, "step": 8284} +{"lr": 0.0002595642622579449, "data_time": 0.16567800045013428, "loss": 0.013085513794794678, "time": 0.6787379264831543, "epoch": 5, "memory": 15585, "step": 8384} +{"lr": 0.0002566833639534589, "data_time": 0.21054401397705078, "loss": 0.008027076534926892, "time": 0.7334361553192139, "epoch": 5, "memory": 15585, "step": 8484} +{"lr": 0.0002537902805032591, "data_time": 0.191230845451355, "loss": 0.006691782409325242, "time": 0.7104598045349121, "epoch": 5, "memory": 15585, "step": 8584} +{"lr": 0.00025088569399614127, "data_time": 0.1858525514602661, "loss": 0.012757928250357509, "time": 0.7172849893569946, "epoch": 5, "memory": 15585, "step": 8684} +{"lr": 0.0002479702892329199, "data_time": 0.20844554901123047, "loss": 0.01673705712892115, "time": 0.7288325548171997, "epoch": 5, "memory": 15585, "step": 8784} +{"lr": 0.00024504475356498226, "data_time": 0.20890192985534667, "loss": 0.01106560449115932, "time": 0.7262905597686767, "epoch": 5, "memory": 15585, "step": 8884} +{"lr": 0.0002421097767322291, "data_time": 0.0015900135040283203, "loss": 0.012585304258391262, "time": 0.6963326454162597, "epoch": 5, "memory": 15585, "step": 8984} +{"lr": 0.00023916605070046265, "data_time": 0.06759204864501953, "loss": 0.01155054229311645, "time": 0.7081430673599243, "epoch": 5, "memory": 15585, "step": 9084} +{"lr": 0.00023621426949824308, "data_time": 0.14045076370239257, "loss": 0.00849416577257216, "time": 0.7397668123245239, "epoch": 5, "memory": 15585, "step": 9184} +{"lr": 0.0002332551290532596, "data_time": 0.15128490924835206, "loss": 0.011787590105086565, "time": 0.7698582649230957, "epoch": 5, "memory": 15585, "step": 9284} +{"lr": 0.00023028932702825637, "data_time": 0.20856213569641113, "loss": 0.007843662519007921, "time": 0.7450214385986328, "epoch": 5, "memory": 15585, "step": 9384} +{"lr": 0.0002273175626565469, "data_time": 0.0874218225479126, "loss": 0.00932695814408362, "time": 0.7761678218841552, "epoch": 5, "memory": 15585, "step": 9484} +{"lr": 0.00022434053657715923, "data_time": 0.0015324592590332032, "loss": 0.012277663219720125, "time": 0.6896156549453736, "epoch": 5, "memory": 15585, "step": 9584} +{"lr": 0.00022135895066965, "data_time": 0.0016330718994140626, "loss": 0.007330503175035119, "time": 0.7267601251602173, "epoch": 5, "memory": 15585, "step": 9684} +{"lr": 0.00021837350788862523, "data_time": 0.0014814615249633789, "loss": 0.008938904805108904, "time": 0.7388899326324463, "epoch": 5, "memory": 15585, "step": 9784} +{"lr": 0.00021538491209800788, "data_time": 0.0016603469848632812, "loss": 0.012299540638923644, "time": 0.7553982019424439, "epoch": 5, "memory": 15585, "step": 9884} +{"lr": 0.0002123938679050912, "data_time": 0.0015993833541870118, "loss": 0.007402639091014862, "time": 0.7416349649429321, "epoch": 5, "memory": 15585, "step": 9984} +{"lr": 0.0002094010804944171, "data_time": 0.0015699625015258788, "loss": 0.010342383664101363, "time": 0.7386821746826172, "epoch": 5, "memory": 15585, "step": 10084} +{"lr": 0.00020640725546151753, "data_time": 0.0015939712524414063, "loss": 0.0128705317620188, "time": 0.8017161369323731, "epoch": 5, "memory": 15585, "step": 10184} +{"accuracy/top1": 85.76039123535156, "data_time": 0.1038989004541616, "time": 0.500392565961744, "step": 5} +{"lr": 0.00020203586947950216, "data_time": 0.033733582496643065, "loss": 0.006772565050050616, "time": 0.6941255807876587, "epoch": 6, "memory": 15585, "step": 10330} +{"lr": 0.0001990424959328828, "data_time": 0.0016382217407226562, "loss": 0.0091314981225878, "time": 0.7268946647644043, "epoch": 6, "memory": 15585, "step": 10430} +{"lr": 0.00019605052695921703, "data_time": 0.0015198707580566407, "loss": 0.009885561093688012, "time": 0.7451726913452148, "epoch": 6, "memory": 15585, "step": 10530} +{"lr": 0.00019306066796107762, "data_time": 0.0014662027359008789, "loss": 0.013890283508226275, "time": 0.7591482639312744, "epoch": 6, "memory": 15585, "step": 10630} +{"lr": 0.00019007362384357827, "data_time": 0.0016084909439086914, "loss": 0.006854643998667598, "time": 0.7070081949234008, "epoch": 6, "memory": 15585, "step": 10730} +{"lr": 0.00018709009884818184, "data_time": 0.0017470359802246095, "loss": 0.0077122320421040055, "time": 0.7733983993530273, "epoch": 6, "memory": 15585, "step": 10830} +{"lr": 0.00018411079638666402, "data_time": 0.001527118682861328, "loss": 0.012490549311041831, "time": 0.7779302358627319, "epoch": 6, "memory": 15585, "step": 10930} +{"lr": 0.00018113641887527326, "data_time": 0.0017706632614135742, "loss": 0.007820055913180112, "time": 0.7920363664627075, "epoch": 6, "memory": 15585, "step": 11030} +{"lr": 0.00017816766756912557, "data_time": 0.0015843629837036133, "loss": 0.009647671738639474, "time": 0.7731965780258179, "epoch": 6, "memory": 15585, "step": 11130} +{"lr": 0.0001752052423968725, "data_time": 0.0015913963317871094, "loss": 0.0061654413118958475, "time": 0.734242582321167, "epoch": 6, "memory": 15585, "step": 11230} +{"lr": 0.00017224984179568312, "data_time": 0.001848125457763672, "loss": 0.009429567726328969, "time": 0.8173613548278809, "epoch": 6, "memory": 15585, "step": 11330} +{"lr": 0.00016930216254657575, "data_time": 0.0018263816833496093, "loss": 0.009123638924211263, "time": 0.7791075229644775, "epoch": 6, "memory": 15585, "step": 11430} +{"lr": 0.0001663628996101415, "data_time": 0.0014664888381958007, "loss": 0.012329932628199458, "time": 0.827071237564087, "epoch": 6, "memory": 15585, "step": 11530} +{"lr": 0.0001634327459626967, "data_time": 0.0015308380126953125, "loss": 0.00868935938924551, "time": 0.6922076225280762, "epoch": 6, "memory": 15585, "step": 11630} +{"lr": 0.0001605123924329032, "data_time": 0.0015259742736816405, "loss": 0.0037909636273980142, "time": 0.7867913961410522, "epoch": 6, "memory": 15585, "step": 11730} +{"lr": 0.0001576025275388945, "data_time": 0.0015814542770385743, "loss": 0.010859042685478926, "time": 0.7386335611343384, "epoch": 6, "memory": 15585, "step": 11830} +{"lr": 0.00015470383732594812, "data_time": 0.00157470703125, "loss": 0.00931333415210247, "time": 0.7589995384216308, "epoch": 6, "memory": 15585, "step": 11930} +{"lr": 0.00015181700520473825, "data_time": 0.0015536785125732423, "loss": 0.007989521836861968, "time": 0.6745124578475952, "epoch": 6, "memory": 15585, "step": 12030} +{"lr": 0.00014894271179021254, "data_time": 0.0015350103378295899, "loss": 0.012033420102670788, "time": 0.7070968151092529, "epoch": 6, "memory": 15585, "step": 12130} +{"lr": 0.0001460816347411264, "data_time": 0.001602506637573242, "loss": 0.003712561307474971, "time": 0.6969620704650878, "epoch": 6, "memory": 15585, "step": 12230} +{"accuracy/top1": 85.78356170654297, "data_time": 0.11287666930526984, "time": 0.5116247036417977, "step": 6} +{"lr": 0.00014192959250590055, "data_time": 0.09155445098876953, "loss": 0.005247314460575581, "time": 0.7381344795227051, "epoch": 7, "memory": 15585, "step": 12376} +{"lr": 0.00013910389161427854, "data_time": 0.22820212841033935, "loss": 0.007978016650304199, "time": 0.7429003477096557, "epoch": 7, "memory": 15585, "step": 12476} +{"lr": 0.00013629372674084914, "data_time": 0.0015171289443969727, "loss": 0.007210482098162174, "time": 0.8034107923507691, "epoch": 7, "memory": 15585, "step": 12576} +{"lr": 0.00013349976042508036, "data_time": 0.001597285270690918, "loss": 0.010318087646737695, "time": 0.7963253736495972, "epoch": 7, "memory": 15585, "step": 12676} +{"lr": 0.0001307226513873823, "data_time": 0.001737833023071289, "loss": 0.005943114683032036, "time": 0.7438418149948121, "epoch": 7, "memory": 15585, "step": 12776} +{"lr": 0.00012796305437380283, "data_time": 0.0014502048492431641, "loss": 0.007989374967291952, "time": 0.7304174184799195, "epoch": 7, "memory": 15585, "step": 12876} +{"lr": 0.00012522162000166114, "data_time": 0.0017261266708374023, "loss": 0.00851035611703992, "time": 0.7707362651824952, "epoch": 7, "memory": 15585, "step": 12976} +{"lr": 0.00012249899460615546, "data_time": 0.001514577865600586, "loss": 0.004354821005836129, "time": 0.7316250085830689, "epoch": 7, "memory": 15585, "step": 13076} +{"lr": 0.00011979582008797885, "data_time": 0.0014783859252929688, "loss": 0.0053947197273373606, "time": 0.7846344709396362, "epoch": 7, "memory": 15585, "step": 13176} +{"lr": 0.00011711273376198264, "data_time": 0.001546788215637207, "loss": 0.009958622278645635, "time": 0.738587212562561, "epoch": 7, "memory": 15585, "step": 13276} +{"lr": 0.00011445036820691838, "data_time": 0.001619863510131836, "loss": 0.008890557801350951, "time": 0.7163378000259399, "epoch": 7, "memory": 15585, "step": 13376} +{"lr": 0.00011180935111629823, "data_time": 0.0015972375869750977, "loss": 0.0036014584358781574, "time": 0.7818014860153198, "epoch": 7, "memory": 15585, "step": 13476} +{"lr": 0.00010919030515040621, "data_time": 0.0016465902328491211, "loss": 0.00534128830768168, "time": 0.6835239648818969, "epoch": 7, "memory": 15585, "step": 13576} +{"lr": 0.00010659384778949681, "data_time": 0.0017879962921142577, "loss": 0.005843141861259937, "time": 0.7692352294921875, "epoch": 7, "memory": 15585, "step": 13676} +{"lr": 0.00010402059118821447, "data_time": 0.001724576950073242, "loss": 0.009580270061269403, "time": 0.7655837059020996, "epoch": 7, "memory": 15585, "step": 13776} +{"lr": 0.00010147114203126853, "data_time": 0.0016311168670654296, "loss": 0.006007488537579775, "time": 0.7026114940643311, "epoch": 7, "memory": 15585, "step": 13876} +{"lr": 9.89461013903984e-05, "data_time": 0.0016609668731689454, "loss": 0.007340662693604827, "time": 0.684268856048584, "epoch": 7, "memory": 15585, "step": 13976} +{"lr": 9.644606458266133e-05, "data_time": 0.0016548633575439453, "loss": 0.007790833944454789, "time": 0.7047852039337158, "epoch": 7, "memory": 15585, "step": 14076} +{"lr": 9.397162103007764e-05, "data_time": 0.001506471633911133, "loss": 0.008929860172793269, "time": 0.7079825878143311, "epoch": 7, "memory": 15585, "step": 14176} +{"lr": 9.152335412066514e-05, "data_time": 0.0016981601715087891, "loss": 0.00461966791190207, "time": 0.7415994882583619, "epoch": 7, "memory": 15585, "step": 14276} +{"accuracy/top1": 86.31150817871094, "data_time": 0.10922900184256132, "time": 0.5069334819668629, "step": 7} +{"lr": 8.79970864551918e-05, "data_time": 0.1353994846343994, "loss": 0.006859015021473169, "time": 0.7495360136032104, "epoch": 8, "memory": 15585, "step": 14422} +{"lr": 8.56156575096546e-05, "data_time": 0.03434908390045166, "loss": 0.007221652613952756, "time": 0.7029934883117676, "epoch": 8, "memory": 15585, "step": 14522} +{"lr": 8.326237525384291e-05, "data_time": 0.0017881393432617188, "loss": 0.004737551975995302, "time": 0.7513375759124756, "epoch": 8, "memory": 15585, "step": 14622} +{"lr": 8.093779451014054e-05, "data_time": 0.0015610218048095702, "loss": 0.003590864920988679, "time": 0.6822865247726441, "epoch": 8, "memory": 15585, "step": 14722} +{"lr": 7.86424633341094e-05, "data_time": 0.001518583297729492, "loss": 0.004272003518417477, "time": 0.7326555967330932, "epoch": 8, "memory": 15585, "step": 14822} +{"lr": 7.637692288527767e-05, "data_time": 0.0017592191696166992, "loss": 0.00909584192559123, "time": 0.7514014959335327, "epoch": 8, "memory": 15585, "step": 14922} +{"lr": 7.414170729955236e-05, "data_time": 0.0013817548751831055, "loss": 0.004582835081964731, "time": 0.9306327342987061, "epoch": 8, "memory": 15585, "step": 15022} +{"lr": 7.193734356328913e-05, "data_time": 0.0016582489013671875, "loss": 0.004775876412168145, "time": 0.7215326070785523, "epoch": 8, "memory": 15585, "step": 15122} +{"lr": 6.976435138904706e-05, "data_time": 0.0018262863159179688, "loss": 0.006084925401955843, "time": 0.7434223413467407, "epoch": 8, "memory": 15585, "step": 15222} +{"lr": 6.762324309305816e-05, "data_time": 0.0017258405685424804, "loss": 0.009475041879341006, "time": 0.7570581197738647, "epoch": 8, "memory": 15585, "step": 15322} +{"lr": 6.551452347444078e-05, "data_time": 0.0016386032104492188, "loss": 0.006913653155788779, "time": 0.701457405090332, "epoch": 8, "memory": 15585, "step": 15422} +{"lr": 6.34386896961863e-05, "data_time": 0.0016444921493530273, "loss": 0.006224154215306044, "time": 0.7100944280624389, "epoch": 8, "memory": 15585, "step": 15522} +{"lr": 6.139623116794487e-05, "data_time": 0.001488494873046875, "loss": 0.0051321182399988174, "time": 0.7368664026260376, "epoch": 8, "memory": 15585, "step": 15622} +{"lr": 5.938762943063905e-05, "data_time": 0.0021057605743408205, "loss": 0.003891888679936528, "time": 0.7467297792434693, "epoch": 8, "memory": 15585, "step": 15722} +{"lr": 5.741335804293427e-05, "data_time": 0.0017569780349731446, "loss": 0.005946611613035202, "time": 0.7236199855804444, "epoch": 8, "memory": 15585, "step": 15822} +{"lr": 5.5473882469589504e-05, "data_time": 0.0018040895462036132, "loss": 0.0042647217400372025, "time": 0.6820027589797973, "epoch": 8, "memory": 15585, "step": 15922} +{"lr": 5.356965997171673e-05, "data_time": 0.0015529632568359376, "loss": 0.006887713773176074, "time": 0.7614014625549317, "epoch": 8, "memory": 15585, "step": 16022} +{"lr": 5.170113949897451e-05, "data_time": 0.001645064353942871, "loss": 0.007057383051142096, "time": 0.8008774518966675, "epoch": 8, "memory": 15585, "step": 16122} +{"lr": 4.986876158372164e-05, "data_time": 0.0015783309936523438, "loss": 0.00506384982727468, "time": 0.764797592163086, "epoch": 8, "memory": 15585, "step": 16222} +{"lr": 4.8072958237154565e-05, "data_time": 0.0016556501388549805, "loss": 0.007003710838034749, "time": 0.7755881547927856, "epoch": 8, "memory": 15585, "step": 16322} +{"accuracy/top1": 86.88578796386719, "data_time": 0.11372942611819407, "time": 0.5120906868918997, "step": 8} +{"lr": 4.551764078948187e-05, "data_time": 0.19244694709777832, "loss": 0.004175967955961824, "time": 0.7079405546188354, "epoch": 9, "memory": 15585, "step": 16468} +{"lr": 4.38135946476099e-05, "data_time": 0.0016949892044067383, "loss": 0.006547352997586131, "time": 0.8092899799346924, "epoch": 9, "memory": 15585, "step": 16568} +{"lr": 4.214755067289632e-05, "data_time": 0.001466655731201172, "loss": 0.005554694728925824, "time": 0.7897008657455444, "epoch": 9, "memory": 15585, "step": 16668} +{"lr": 4.051990166076003e-05, "data_time": 0.0015451431274414063, "loss": 0.0033359632827341555, "time": 0.7985238313674927, "epoch": 9, "memory": 15585, "step": 16768} +{"lr": 3.893103135441892e-05, "data_time": 0.001585078239440918, "loss": 0.0057896914891898636, "time": 0.7752117872238159, "epoch": 9, "memory": 15585, "step": 16868} +{"lr": 3.7381314354415884e-05, "data_time": 0.0016515254974365234, "loss": 0.007572964346036315, "time": 0.7823155879974365, "epoch": 9, "memory": 15585, "step": 16968} +{"lr": 3.587111603030147e-05, "data_time": 0.0016986370086669923, "loss": 0.006491952342912555, "time": 0.7844523668289185, "epoch": 9, "memory": 15585, "step": 17068} +{"lr": 3.440079243449231e-05, "data_time": 0.0015486717224121095, "loss": 0.005592259066179395, "time": 0.7427172660827637, "epoch": 9, "memory": 15585, "step": 17168} +{"lr": 3.297069021832627e-05, "data_time": 0.001545095443725586, "loss": 0.005130115430802107, "time": 0.7652437210083007, "epoch": 9, "memory": 15585, "step": 17268} +{"lr": 3.158114655033408e-05, "data_time": 0.0015884637832641602, "loss": 0.004831411922350526, "time": 0.7513247966766358, "epoch": 9, "memory": 15585, "step": 17368} +{"lr": 3.023248903674686e-05, "data_time": 0.0014780521392822265, "loss": 0.004853937122970819, "time": 0.7821897745132447, "epoch": 9, "memory": 15585, "step": 17468} +{"lr": 2.8925035644257496e-05, "data_time": 0.0015176773071289063, "loss": 0.005496340431272983, "time": 0.7657095670700074, "epoch": 9, "memory": 15585, "step": 17568} +{"lr": 2.76590946250556e-05, "data_time": 0.0015669822692871093, "loss": 0.00706896954216063, "time": 0.796151065826416, "epoch": 9, "memory": 15585, "step": 17668} +{"lr": 2.6434964444152048e-05, "data_time": 0.0015767335891723633, "loss": 0.005961532332003116, "time": 0.7714215517044067, "epoch": 9, "memory": 15585, "step": 17768} +{"lr": 2.525293370901126e-05, "data_time": 0.0017023086547851562, "loss": 0.008977956883609296, "time": 0.758408522605896, "epoch": 9, "memory": 15585, "step": 17868} +{"lr": 2.411328110150761e-05, "data_time": 0.0017127037048339845, "loss": 0.0072656759526580576, "time": 0.7131024837493897, "epoch": 9, "memory": 15585, "step": 17968} +{"lr": 2.3016275312222135e-05, "data_time": 0.0016339778900146484, "loss": 0.006877522775903344, "time": 0.7824469566345215, "epoch": 9, "memory": 15585, "step": 18068} +{"lr": 2.1962174977094375e-05, "data_time": 0.0015943288803100587, "loss": 0.0030747639015316964, "time": 0.7644901752471924, "epoch": 9, "memory": 15585, "step": 18168} +{"lr": 2.095122861644491e-05, "data_time": 0.0015036344528198242, "loss": 0.003204282792285085, "time": 0.7385929346084594, "epoch": 9, "memory": 15585, "step": 18268} +{"lr": 1.9983674576383036e-05, "data_time": 0.0015795469284057618, "loss": 0.00590423010289669, "time": 0.7436267614364624, "epoch": 9, "memory": 15585, "step": 18368} +{"accuracy/top1": 87.1456298828125, "data_time": 0.10318008016367428, "time": 0.5023639632053063, "step": 9} +{"lr": 1.8649439622232705e-05, "data_time": 0.0014722347259521484, "loss": 0.006848942209035158, "time": 0.7175880908966065, "epoch": 10, "memory": 15585, "step": 18514} +{"lr": 1.7789580193179967e-05, "data_time": 0.0017381668090820312, "loss": 0.007593048457056284, "time": 0.7045023679733277, "epoch": 10, "memory": 15585, "step": 18614} +{"lr": 1.697385849182042e-05, "data_time": 0.0016040563583374023, "loss": 0.003128493158146739, "time": 0.7591366291046142, "epoch": 10, "memory": 15585, "step": 18714} +{"lr": 1.620246683705586e-05, "data_time": 0.001618337631225586, "loss": 0.001843567006289959, "time": 0.7901627540588378, "epoch": 10, "memory": 15585, "step": 18814} +{"lr": 1.547558709629951e-05, "data_time": 0.0017681360244750977, "loss": 0.0034913192968815563, "time": 0.7734734058380127, "epoch": 10, "memory": 15585, "step": 18914} +{"lr": 1.479339064259843e-05, "data_time": 0.0015447378158569337, "loss": 0.0075127596501261, "time": 0.7880001783370971, "epoch": 10, "memory": 15585, "step": 19014} +{"lr": 1.4156038314229283e-05, "data_time": 0.0016036748886108399, "loss": 0.005412493133917451, "time": 0.7852174520492554, "epoch": 10, "memory": 15585, "step": 19114} +{"lr": 1.3563680376778489e-05, "data_time": 0.001610541343688965, "loss": 0.00456148344092071, "time": 0.7876395463943482, "epoch": 10, "memory": 15585, "step": 19214} +{"lr": 1.3016456487714728e-05, "data_time": 0.0017944574356079102, "loss": 0.00857179998420179, "time": 0.8061561822891236, "epoch": 10, "memory": 15585, "step": 19314} +{"lr": 1.2514495663462562e-05, "data_time": 0.0019151687622070313, "loss": 0.006411237828433514, "time": 0.8434019088745117, "epoch": 10, "memory": 15585, "step": 19414} +{"lr": 1.205791624898479e-05, "data_time": 0.001524066925048828, "loss": 0.006115010101348161, "time": 0.7644287824630738, "epoch": 10, "memory": 15585, "step": 19514} +{"lr": 1.1646825889880944e-05, "data_time": 0.0016330718994140626, "loss": 0.004738573217764497, "time": 0.8091838598251343, "epoch": 10, "memory": 15585, "step": 19614} +{"lr": 1.128132150700797e-05, "data_time": 0.0016618490219116211, "loss": 0.0060673046857118605, "time": 0.7980396986007691, "epoch": 10, "memory": 15585, "step": 19714} +{"lr": 1.0961489273629872e-05, "data_time": 0.0017602920532226562, "loss": 0.008298689546063543, "time": 0.7398819923400879, "epoch": 10, "memory": 15585, "step": 19814} +{"lr": 1.0687404595100842e-05, "data_time": 0.0014931678771972657, "loss": 0.00652552735991776, "time": 0.7104421615600586, "epoch": 10, "memory": 15585, "step": 19914} +{"lr": 1.0459132091087482e-05, "data_time": 0.0016926050186157227, "loss": 0.004293518140912056, "time": 0.8109429121017456, "epoch": 10, "memory": 15585, "step": 20014} +{"lr": 1.0276725580333608e-05, "data_time": 0.001483583450317383, "loss": 0.0022668967954814433, "time": 0.7934521198272705, "epoch": 10, "memory": 15585, "step": 20114} +{"lr": 1.0140228067971661e-05, "data_time": 0.0016131401062011719, "loss": 0.0033864120952785013, "time": 0.7601506471633911, "epoch": 10, "memory": 15585, "step": 20214} +{"lr": 1.0049671735383602e-05, "data_time": 0.0016006708145141601, "loss": 0.004824089026078582, "time": 0.7592834234237671, "epoch": 10, "memory": 15585, "step": 20314} +{"lr": 1.000507793261375e-05, "data_time": 0.0016960620880126954, "loss": 0.006116680335253477, "time": 0.7174782752990723, "epoch": 10, "memory": 15585, "step": 20414} +{"accuracy/top1": 87.48324584960938, "data_time": 0.12582454525056433, "time": 0.5218902220491504, "step": 10} diff --git a/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1/20230606_005643/vis_data/config.py b/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1/20230606_005643/vis_data/config.py new file mode 100644 index 0000000000000000000000000000000000000000..f4842b8fcf8c13cf9f7a8b857c19151fda618735 --- /dev/null +++ b/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1/20230606_005643/vis_data/config.py @@ -0,0 +1,366 @@ +optim_wrapper = dict( + optimizer=dict( + type='AdamW', + lr=0.0004, + weight_decay=0.05, + eps=1e-08, + betas=(0.9, 0.999), + _scope_='mmpretrain'), + paramwise_cfg=dict( + norm_decay_mult=0.0, + bias_decay_mult=0.0, + flat_decay_mult=0.0, + custom_keys=dict({ + '.absolute_pos_embed': dict(decay_mult=0.0), + '.relative_position_bias_table': dict(decay_mult=0.0) + })), + type='AmpOptimWrapper', + dtype='bfloat16', + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=False, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=4096) +model = dict( + type='ImageClassifier', + backbone=dict( + frozen_stages=2, + type='ConvNeXt', + arch='small', + drop_path_rate=0.4, + init_cfg=dict( + type='Pretrained', + checkpoint= + 'https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth', + prefix='backbone')), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=768, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + init_cfg=None), + init_cfg=dict( + type='TruncNormal', layer=['Conv2d', 'Linear'], std=0.02, bias=0.0), + train_cfg=None) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/all_0.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/all_1.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3fake8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/cc1m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3real8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV2-1-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/if-dpmsolver++-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV2-1-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/if-dpmsolver++-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1' diff --git a/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1/20230606_005643/vis_data/events.out.tfevents.1685984217.SH-IDC1-10-140-24-103.207843.0 b/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1/20230606_005643/vis_data/events.out.tfevents.1685984217.SH-IDC1-10-140-24-103.207843.0 new file mode 100644 index 0000000000000000000000000000000000000000..2bba5052cf33e0e41224d10f6d6cf86c11a21e41 --- /dev/null +++ b/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1/20230606_005643/vis_data/events.out.tfevents.1685984217.SH-IDC1-10-140-24-103.207843.0 @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:5f025d95a1032b865b2eaf9aeb39d5829727869c5856833effd64dcb0bbadd6d +size 2986390 diff --git a/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1/20230606_005643/vis_data/scalars.json b/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1/20230606_005643/vis_data/scalars.json new file mode 100644 index 0000000000000000000000000000000000000000..92bb2340f18c97f6e63b3d86022d9069cd7b6822 --- /dev/null +++ b/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1/20230606_005643/vis_data/scalars.json @@ -0,0 +1,210 @@ +{"lr": 0.00039997747031453974, "data_time": 0.0017314672470092774, "loss": 0.04582378547638655, "time": 0.7797232389450073, "epoch": 1, "memory": 15587, "step": 100} +{"lr": 0.0003999089739988646, "data_time": 0.001805877685546875, "loss": 0.04913739897310734, "time": 0.738232421875, "epoch": 1, "memory": 15587, "step": 200} +{"lr": 0.00039979452490327365, "data_time": 0.19394207000732422, "loss": 0.03367728535085916, "time": 0.7865641832351684, "epoch": 1, "memory": 15587, "step": 300} +{"lr": 0.0003996341500108966, "data_time": 0.0015164613723754883, "loss": 0.02638430558145046, "time": 0.8205552816390991, "epoch": 1, "memory": 15587, "step": 400} +{"lr": 0.00039942788713257383, "data_time": 0.001667618751525879, "loss": 0.03231051415205002, "time": 0.7348530769348145, "epoch": 1, "memory": 15587, "step": 500} +{"lr": 0.00039917578489794236, "data_time": 0.0015869855880737305, "loss": 0.03607837650924921, "time": 0.8029991388320923, "epoch": 1, "memory": 15587, "step": 600} +{"lr": 0.00039887790274397046, "data_time": 0.0016368865966796876, "loss": 0.030855814926326275, "time": 0.736802339553833, "epoch": 1, "memory": 15587, "step": 700} +{"lr": 0.00039853431090094445, "data_time": 0.001633024215698242, "loss": 0.028339687455445527, "time": 0.7840763330459595, "epoch": 1, "memory": 15587, "step": 800} +{"lr": 0.0003981450903759111, "data_time": 0.0016636610031127929, "loss": 0.032328211702406406, "time": 0.7560638904571533, "epoch": 1, "memory": 15587, "step": 900} +{"lr": 0.00039771033293357867, "data_time": 0.0015117883682250976, "loss": 0.024187591020017864, "time": 0.7417938470840454, "epoch": 1, "memory": 15587, "step": 1000} +{"lr": 0.0003972301410746815, "data_time": 0.00158998966217041, "loss": 0.028163632191717625, "time": 0.7944855690002441, "epoch": 1, "memory": 15587, "step": 1100} +{"lr": 0.00039670462801181557, "data_time": 0.0015922307968139649, "loss": 0.025045361928641797, "time": 0.7637933015823364, "epoch": 1, "memory": 15587, "step": 1200} +{"lr": 0.0003961339176427448, "data_time": 0.0023871898651123048, "loss": 0.017230405099689962, "time": 0.8597577810287476, "epoch": 1, "memory": 15587, "step": 1300} +{"lr": 0.00039551814452119214, "data_time": 0.0015121221542358399, "loss": 0.026982803316786885, "time": 0.7743359804153442, "epoch": 1, "memory": 15587, "step": 1400} +{"lr": 0.00039485745382511467, "data_time": 0.0016725778579711914, "loss": 0.019004460936412216, "time": 0.8123712062835693, "epoch": 1, "memory": 15587, "step": 1500} +{"lr": 0.0003941520013224769, "data_time": 0.0013941049575805664, "loss": 0.024567633122205734, "time": 0.7806166172027588, "epoch": 1, "memory": 15587, "step": 1600} +{"lr": 0.0003934019533345259, "data_time": 0.0014190196990966797, "loss": 0.019510003039613366, "time": 0.7741846323013306, "epoch": 1, "memory": 15587, "step": 1700} +{"lr": 0.00039260748669657817, "data_time": 0.001448655128479004, "loss": 0.018855373607948422, "time": 0.781554388999939, "epoch": 1, "memory": 15587, "step": 1800} +{"lr": 0.000391768788716328, "data_time": 0.0013930082321166992, "loss": 0.017342483764514326, "time": 0.7650550842285156, "epoch": 1, "memory": 15587, "step": 1900} +{"lr": 0.00039088605712968733, "data_time": 0.0015112400054931641, "loss": 0.014465322950854897, "time": 0.775104570388794, "epoch": 1, "memory": 15587, "step": 2000} +{"accuracy/top1": 50.69261932373047, "data_time": 0.12550911903381348, "time": 0.5630749146143595, "step": 1} +{"lr": 0.0003895186269690071, "data_time": 0.035175633430480954, "loss": 0.020252875657752155, "time": 0.7574407815933227, "epoch": 2, "memory": 15587, "step": 2146} +{"lr": 0.0003885284798052241, "data_time": 0.01851482391357422, "loss": 0.024551139771938325, "time": 0.7844219207763672, "epoch": 2, "memory": 15585, "step": 2246} +{"lr": 0.0003874950629875727, "data_time": 0.001478862762451172, "loss": 0.014201085036620497, "time": 0.7822738409042358, "epoch": 2, "memory": 15585, "step": 2346} +{"lr": 0.0003864186201599172, "data_time": 0.0019092321395874023, "loss": 0.022182260919362307, "time": 0.7561484575271606, "epoch": 2, "memory": 15585, "step": 2446} +{"lr": 0.0003852994051101632, "data_time": 0.0015421152114868165, "loss": 0.014758217940106988, "time": 0.7986104726791382, "epoch": 2, "memory": 15585, "step": 2546} +{"lr": 0.0003841376817104243, "data_time": 0.0015173673629760743, "loss": 0.015461743343621492, "time": 0.7966030597686767, "epoch": 2, "memory": 15585, "step": 2646} +{"lr": 0.0003829337238548097, "data_time": 0.001443791389465332, "loss": 0.012522991793230177, "time": 0.740956974029541, "epoch": 2, "memory": 15585, "step": 2746} +{"lr": 0.0003816878153948477, "data_time": 0.0015905141830444337, "loss": 0.015928221913054586, "time": 0.709771990776062, "epoch": 2, "memory": 15585, "step": 2846} +{"lr": 0.0003804002500725666, "data_time": 0.0014972925186157227, "loss": 0.020683481032028793, "time": 0.7683061361312866, "epoch": 2, "memory": 15585, "step": 2946} +{"lr": 0.0003790713314512378, "data_time": 0.0016095876693725587, "loss": 0.019671098375692964, "time": 0.7312532663345337, "epoch": 2, "memory": 15585, "step": 3046} +{"lr": 0.00037770137284380876, "data_time": 0.0016412973403930665, "loss": 0.02124711577780545, "time": 0.7730305433273316, "epoch": 2, "memory": 15585, "step": 3146} +{"lr": 0.00037629069723903126, "data_time": 0.001612401008605957, "loss": 0.019148432463407517, "time": 0.7195446729660034, "epoch": 2, "memory": 15585, "step": 3246} +{"lr": 0.0003748396372253147, "data_time": 0.001468658447265625, "loss": 0.01650288375094533, "time": 0.764950156211853, "epoch": 2, "memory": 15585, "step": 3346} +{"lr": 0.0003733485349123126, "data_time": 0.00144503116607666, "loss": 0.013821152597665786, "time": 0.7550675868988037, "epoch": 2, "memory": 15585, "step": 3446} +{"lr": 0.0003718177418502636, "data_time": 0.001757049560546875, "loss": 0.01906374916434288, "time": 0.7818350791931152, "epoch": 2, "memory": 15585, "step": 3546} +{"lr": 0.0003702476189471092, "data_time": 0.0016087532043457032, "loss": 0.01476140357553959, "time": 0.8163014650344849, "epoch": 2, "memory": 15585, "step": 3646} +{"lr": 0.00036863853638340543, "data_time": 0.0014557361602783204, "loss": 0.012234715931117534, "time": 0.7612666845321655, "epoch": 2, "memory": 15585, "step": 3746} +{"lr": 0.000366990873525045, "data_time": 0.0015973091125488282, "loss": 0.01461083865724504, "time": 0.7513570547103882, "epoch": 2, "memory": 15585, "step": 3846} +{"lr": 0.00036530501883381603, "data_time": 0.0015964746475219727, "loss": 0.008052535681053996, "time": 0.6864346742630005, "epoch": 2, "memory": 15585, "step": 3946} +{"lr": 0.00036358136977581845, "data_time": 0.0018983840942382812, "loss": 0.016030167089775206, "time": 0.7498691082000732, "epoch": 2, "memory": 15585, "step": 4046} +{"accuracy/top1": 82.29316711425781, "data_time": 0.1093459246588535, "time": 0.5076444539867464, "step": 2} +{"lr": 0.0003609978147988783, "data_time": 0.06281070709228516, "loss": 0.019261301029473543, "time": 0.7321836471557617, "epoch": 3, "memory": 15585, "step": 4192} +{"lr": 0.00035918293929014, "data_time": 0.001866316795349121, "loss": 0.013484583934769034, "time": 0.7675493717193603, "epoch": 3, "memory": 15585, "step": 4292} +{"lr": 0.0003573317127887786, "data_time": 0.0016434192657470703, "loss": 0.018350797798484565, "time": 0.7364666223526001, "epoch": 3, "memory": 15585, "step": 4392} +{"lr": 0.000355444571749834, "data_time": 0.0015516996383666993, "loss": 0.013195968139916659, "time": 0.7499399185180664, "epoch": 3, "memory": 15585, "step": 4492} +{"lr": 0.0003535219610957493, "data_time": 0.0015429258346557617, "loss": 0.015334970876574516, "time": 0.7414395809173584, "epoch": 3, "memory": 15585, "step": 4592} +{"lr": 0.00035156433411147255, "data_time": 0.0014833688735961914, "loss": 0.01002070913091302, "time": 0.7465383768081665, "epoch": 3, "memory": 15585, "step": 4692} +{"lr": 0.00034957215233758935, "data_time": 0.0015995025634765625, "loss": 0.014274924341589213, "time": 0.7002089977264404, "epoch": 3, "memory": 15585, "step": 4792} +{"lr": 0.00034754588546150636, "data_time": 0.001563286781311035, "loss": 0.012178977811709046, "time": 0.7269941806793213, "epoch": 3, "memory": 15585, "step": 4892} +{"lr": 0.0003454860112067171, "data_time": 0.0015238523483276367, "loss": 0.017030299128964543, "time": 0.7481244087219239, "epoch": 3, "memory": 15585, "step": 4992} +{"lr": 0.00034339301522016903, "data_time": 0.0015039920806884765, "loss": 0.012458660826086997, "time": 0.7832327604293823, "epoch": 3, "memory": 15585, "step": 5092} +{"lr": 0.0003412673909577655, "data_time": 0.0016904354095458984, "loss": 0.012210268201306463, "time": 0.7109328031539917, "epoch": 3, "memory": 15585, "step": 5192} +{"lr": 0.00033910963956802737, "data_time": 0.00177459716796875, "loss": 0.01046714880503714, "time": 0.7497158050537109, "epoch": 3, "memory": 15585, "step": 5292} +{"lr": 0.00033692026977393767, "data_time": 0.0015202760696411133, "loss": 0.010511100618168711, "time": 0.7057794332504272, "epoch": 3, "memory": 15585, "step": 5392} +{"lr": 0.0003346997977530036, "data_time": 0.0015427827835083007, "loss": 0.011450219480320812, "time": 0.666927170753479, "epoch": 3, "memory": 15585, "step": 5492} +{"lr": 0.0003324487470155584, "data_time": 0.0015279769897460938, "loss": 0.017318726144731046, "time": 0.7419221878051758, "epoch": 3, "memory": 15585, "step": 5592} +{"lr": 0.0003301676482813382, "data_time": 0.0017217874526977539, "loss": 0.008253357093781232, "time": 0.7053928375244141, "epoch": 3, "memory": 15585, "step": 5692} +{"lr": 0.00032785703935435476, "data_time": 0.0016490697860717773, "loss": 0.010431417217478155, "time": 0.8212690114974975, "epoch": 3, "memory": 15585, "step": 5792} +{"lr": 0.00032551746499609967, "data_time": 0.0015106201171875, "loss": 0.013492595544084907, "time": 0.7035637140274048, "epoch": 3, "memory": 15585, "step": 5892} +{"lr": 0.0003231494767971101, "data_time": 0.0015542745590209962, "loss": 0.012716319225728513, "time": 0.7209977626800537, "epoch": 3, "memory": 15585, "step": 5992} +{"lr": 0.0003207536330469211, "data_time": 0.001537013053894043, "loss": 0.009736620634794236, "time": 0.7016186952590943, "epoch": 3, "memory": 15585, "step": 6092} +{"accuracy/top1": 86.2817153930664, "data_time": 0.111266233881966, "time": 0.5068588843111133, "step": 3} +{"lr": 0.000317206849615549, "data_time": 0.20003046989440917, "loss": 0.010455530788749456, "time": 0.7208166360855103, "epoch": 4, "memory": 15585, "step": 6238} +{"lr": 0.00031474489842163173, "data_time": 0.1849517345428467, "loss": 0.01119880867190659, "time": 0.7366448879241944, "epoch": 4, "memory": 15585, "step": 6338} +{"lr": 0.0003122570731847668, "data_time": 0.030050158500671387, "loss": 0.011106752138584853, "time": 0.7972426414489746, "epoch": 4, "memory": 15585, "step": 6438} +{"lr": 0.00030974396044791, "data_time": 0.0014658451080322265, "loss": 0.020003657322376966, "time": 0.7284992456436157, "epoch": 4, "memory": 15585, "step": 6538} +{"lr": 0.0003072061527159319, "data_time": 0.001621842384338379, "loss": 0.011039762571454049, "time": 0.7132418394088745, "epoch": 4, "memory": 15585, "step": 6638} +{"lr": 0.0003046442483159274, "data_time": 0.0017466068267822266, "loss": 0.02424939088523388, "time": 0.7078982591629028, "epoch": 4, "memory": 15585, "step": 6738} +{"lr": 0.0003020588512561505, "data_time": 0.001969766616821289, "loss": 0.008156781597062946, "time": 0.7495105743408204, "epoch": 4, "memory": 15585, "step": 6838} +{"lr": 0.0002994505710836101, "data_time": 0.0015324831008911132, "loss": 0.010514680109918118, "time": 0.7136235237121582, "epoch": 4, "memory": 15585, "step": 6938} +{"lr": 0.0002968200227403596, "data_time": 0.0015955686569213868, "loss": 0.0111268509645015, "time": 0.7207844018936157, "epoch": 4, "memory": 15585, "step": 7038} +{"lr": 0.0002941678264185154, "data_time": 0.0014647245407104492, "loss": 0.015258364658802748, "time": 0.7219701528549194, "epoch": 4, "memory": 15585, "step": 7138} +{"lr": 0.0002914946074140367, "data_time": 0.0015655517578125, "loss": 0.011088904459029436, "time": 0.7023492813110351, "epoch": 4, "memory": 15585, "step": 7238} +{"lr": 0.00028880099597930275, "data_time": 0.0016550302505493164, "loss": 0.006905767740681768, "time": 0.7222923994064331, "epoch": 4, "memory": 15585, "step": 7338} +{"lr": 0.0002860876271745208, "data_time": 0.0018896341323852539, "loss": 0.0085047650616616, "time": 0.7889053821563721, "epoch": 4, "memory": 15585, "step": 7438} +{"lr": 0.0002833551407180016, "data_time": 0.0015702962875366211, "loss": 0.012262030225247144, "time": 0.9031259775161743, "epoch": 4, "memory": 15585, "step": 7538} +{"lr": 0.00028060418083533477, "data_time": 0.0015096664428710938, "loss": 0.011341116577386855, "time": 0.7258390188217163, "epoch": 4, "memory": 15585, "step": 7638} +{"lr": 0.0002778353961075036, "data_time": 0.0019012689590454102, "loss": 0.010915554920211434, "time": 0.7604675531387329, "epoch": 4, "memory": 15585, "step": 7738} +{"lr": 0.0002750494393179725, "data_time": 0.0017791986465454102, "loss": 0.015977052552625538, "time": 0.6591270446777344, "epoch": 4, "memory": 15585, "step": 7838} +{"lr": 0.00027224696729878206, "data_time": 0.0016847610473632812, "loss": 0.017442272650077938, "time": 0.77463698387146, "epoch": 4, "memory": 15585, "step": 7938} +{"lr": 0.00026942864077569254, "data_time": 0.0015532970428466797, "loss": 0.018840466765686868, "time": 0.7723486661911011, "epoch": 4, "memory": 15585, "step": 8038} +{"lr": 0.0002665951242124057, "data_time": 0.0015737295150756836, "loss": 0.011894524283707141, "time": 0.7510597467422485, "epoch": 4, "memory": 15585, "step": 8138} +{"accuracy/top1": 85.96726989746094, "data_time": 0.10596991367027407, "time": 0.5014197787300485, "step": 4} +{"lr": 0.0002624322962007581, "data_time": 0.19933886528015138, "loss": 0.010770977195352315, "time": 0.7168541193008423, "epoch": 5, "memory": 15585, "step": 8284} +{"lr": 0.0002595642622579449, "data_time": 0.16567800045013428, "loss": 0.013085513794794678, "time": 0.6787379264831543, "epoch": 5, "memory": 15585, "step": 8384} +{"lr": 0.0002566833639534589, "data_time": 0.21054401397705078, "loss": 0.008027076534926892, "time": 0.7334361553192139, "epoch": 5, "memory": 15585, "step": 8484} +{"lr": 0.0002537902805032591, "data_time": 0.191230845451355, "loss": 0.006691782409325242, "time": 0.7104598045349121, "epoch": 5, "memory": 15585, "step": 8584} +{"lr": 0.00025088569399614127, "data_time": 0.1858525514602661, "loss": 0.012757928250357509, "time": 0.7172849893569946, "epoch": 5, "memory": 15585, "step": 8684} +{"lr": 0.0002479702892329199, "data_time": 0.20844554901123047, "loss": 0.01673705712892115, "time": 0.7288325548171997, "epoch": 5, "memory": 15585, "step": 8784} +{"lr": 0.00024504475356498226, "data_time": 0.20890192985534667, "loss": 0.01106560449115932, "time": 0.7262905597686767, "epoch": 5, "memory": 15585, "step": 8884} +{"lr": 0.0002421097767322291, "data_time": 0.0015900135040283203, "loss": 0.012585304258391262, "time": 0.6963326454162597, "epoch": 5, "memory": 15585, "step": 8984} +{"lr": 0.00023916605070046265, "data_time": 0.06759204864501953, "loss": 0.01155054229311645, "time": 0.7081430673599243, "epoch": 5, "memory": 15585, "step": 9084} +{"lr": 0.00023621426949824308, "data_time": 0.14045076370239257, "loss": 0.00849416577257216, "time": 0.7397668123245239, "epoch": 5, "memory": 15585, "step": 9184} +{"lr": 0.0002332551290532596, "data_time": 0.15128490924835206, "loss": 0.011787590105086565, "time": 0.7698582649230957, "epoch": 5, "memory": 15585, "step": 9284} +{"lr": 0.00023028932702825637, "data_time": 0.20856213569641113, "loss": 0.007843662519007921, "time": 0.7450214385986328, "epoch": 5, "memory": 15585, "step": 9384} +{"lr": 0.0002273175626565469, "data_time": 0.0874218225479126, "loss": 0.00932695814408362, "time": 0.7761678218841552, "epoch": 5, "memory": 15585, "step": 9484} +{"lr": 0.00022434053657715923, "data_time": 0.0015324592590332032, "loss": 0.012277663219720125, "time": 0.6896156549453736, "epoch": 5, "memory": 15585, "step": 9584} +{"lr": 0.00022135895066965, "data_time": 0.0016330718994140626, "loss": 0.007330503175035119, "time": 0.7267601251602173, "epoch": 5, "memory": 15585, "step": 9684} +{"lr": 0.00021837350788862523, "data_time": 0.0014814615249633789, "loss": 0.008938904805108904, "time": 0.7388899326324463, "epoch": 5, "memory": 15585, "step": 9784} +{"lr": 0.00021538491209800788, "data_time": 0.0016603469848632812, "loss": 0.012299540638923644, "time": 0.7553982019424439, "epoch": 5, "memory": 15585, "step": 9884} +{"lr": 0.0002123938679050912, "data_time": 0.0015993833541870118, "loss": 0.007402639091014862, "time": 0.7416349649429321, "epoch": 5, "memory": 15585, "step": 9984} +{"lr": 0.0002094010804944171, "data_time": 0.0015699625015258788, "loss": 0.010342383664101363, "time": 0.7386821746826172, "epoch": 5, "memory": 15585, "step": 10084} +{"lr": 0.00020640725546151753, "data_time": 0.0015939712524414063, "loss": 0.0128705317620188, "time": 0.8017161369323731, "epoch": 5, "memory": 15585, "step": 10184} +{"accuracy/top1": 85.76039123535156, "data_time": 0.1038989004541616, "time": 0.500392565961744, "step": 5} +{"lr": 0.00020203586947950216, "data_time": 0.033733582496643065, "loss": 0.006772565050050616, "time": 0.6941255807876587, "epoch": 6, "memory": 15585, "step": 10330} +{"lr": 0.0001990424959328828, "data_time": 0.0016382217407226562, "loss": 0.0091314981225878, "time": 0.7268946647644043, "epoch": 6, "memory": 15585, "step": 10430} +{"lr": 0.00019605052695921703, "data_time": 0.0015198707580566407, "loss": 0.009885561093688012, "time": 0.7451726913452148, "epoch": 6, "memory": 15585, "step": 10530} +{"lr": 0.00019306066796107762, "data_time": 0.0014662027359008789, "loss": 0.013890283508226275, "time": 0.7591482639312744, "epoch": 6, "memory": 15585, "step": 10630} +{"lr": 0.00019007362384357827, "data_time": 0.0016084909439086914, "loss": 0.006854643998667598, "time": 0.7070081949234008, "epoch": 6, "memory": 15585, "step": 10730} +{"lr": 0.00018709009884818184, "data_time": 0.0017470359802246095, "loss": 0.0077122320421040055, "time": 0.7733983993530273, "epoch": 6, "memory": 15585, "step": 10830} +{"lr": 0.00018411079638666402, "data_time": 0.001527118682861328, "loss": 0.012490549311041831, "time": 0.7779302358627319, "epoch": 6, "memory": 15585, "step": 10930} +{"lr": 0.00018113641887527326, "data_time": 0.0017706632614135742, "loss": 0.007820055913180112, "time": 0.7920363664627075, "epoch": 6, "memory": 15585, "step": 11030} +{"lr": 0.00017816766756912557, "data_time": 0.0015843629837036133, "loss": 0.009647671738639474, "time": 0.7731965780258179, "epoch": 6, "memory": 15585, "step": 11130} +{"lr": 0.0001752052423968725, "data_time": 0.0015913963317871094, "loss": 0.0061654413118958475, "time": 0.734242582321167, "epoch": 6, "memory": 15585, "step": 11230} +{"lr": 0.00017224984179568312, "data_time": 0.001848125457763672, "loss": 0.009429567726328969, "time": 0.8173613548278809, "epoch": 6, "memory": 15585, "step": 11330} +{"lr": 0.00016930216254657575, "data_time": 0.0018263816833496093, "loss": 0.009123638924211263, "time": 0.7791075229644775, "epoch": 6, "memory": 15585, "step": 11430} +{"lr": 0.0001663628996101415, "data_time": 0.0014664888381958007, "loss": 0.012329932628199458, "time": 0.827071237564087, "epoch": 6, "memory": 15585, "step": 11530} +{"lr": 0.0001634327459626967, "data_time": 0.0015308380126953125, "loss": 0.00868935938924551, "time": 0.6922076225280762, "epoch": 6, "memory": 15585, "step": 11630} +{"lr": 0.0001605123924329032, "data_time": 0.0015259742736816405, "loss": 0.0037909636273980142, "time": 0.7867913961410522, "epoch": 6, "memory": 15585, "step": 11730} +{"lr": 0.0001576025275388945, "data_time": 0.0015814542770385743, "loss": 0.010859042685478926, "time": 0.7386335611343384, "epoch": 6, "memory": 15585, "step": 11830} +{"lr": 0.00015470383732594812, "data_time": 0.00157470703125, "loss": 0.00931333415210247, "time": 0.7589995384216308, "epoch": 6, "memory": 15585, "step": 11930} +{"lr": 0.00015181700520473825, "data_time": 0.0015536785125732423, "loss": 0.007989521836861968, "time": 0.6745124578475952, "epoch": 6, "memory": 15585, "step": 12030} +{"lr": 0.00014894271179021254, "data_time": 0.0015350103378295899, "loss": 0.012033420102670788, "time": 0.7070968151092529, "epoch": 6, "memory": 15585, "step": 12130} +{"lr": 0.0001460816347411264, "data_time": 0.001602506637573242, "loss": 0.003712561307474971, "time": 0.6969620704650878, "epoch": 6, "memory": 15585, "step": 12230} +{"accuracy/top1": 85.78356170654297, "data_time": 0.11287666930526984, "time": 0.5116247036417977, "step": 6} +{"lr": 0.00014192959250590055, "data_time": 0.09155445098876953, "loss": 0.005247314460575581, "time": 0.7381344795227051, "epoch": 7, "memory": 15585, "step": 12376} +{"lr": 0.00013910389161427854, "data_time": 0.22820212841033935, "loss": 0.007978016650304199, "time": 0.7429003477096557, "epoch": 7, "memory": 15585, "step": 12476} +{"lr": 0.00013629372674084914, "data_time": 0.0015171289443969727, "loss": 0.007210482098162174, "time": 0.8034107923507691, "epoch": 7, "memory": 15585, "step": 12576} +{"lr": 0.00013349976042508036, "data_time": 0.001597285270690918, "loss": 0.010318087646737695, "time": 0.7963253736495972, "epoch": 7, "memory": 15585, "step": 12676} +{"lr": 0.0001307226513873823, "data_time": 0.001737833023071289, "loss": 0.005943114683032036, "time": 0.7438418149948121, "epoch": 7, "memory": 15585, "step": 12776} +{"lr": 0.00012796305437380283, "data_time": 0.0014502048492431641, "loss": 0.007989374967291952, "time": 0.7304174184799195, "epoch": 7, "memory": 15585, "step": 12876} +{"lr": 0.00012522162000166114, "data_time": 0.0017261266708374023, "loss": 0.00851035611703992, "time": 0.7707362651824952, "epoch": 7, "memory": 15585, "step": 12976} +{"lr": 0.00012249899460615546, "data_time": 0.001514577865600586, "loss": 0.004354821005836129, "time": 0.7316250085830689, "epoch": 7, "memory": 15585, "step": 13076} +{"lr": 0.00011979582008797885, "data_time": 0.0014783859252929688, "loss": 0.0053947197273373606, "time": 0.7846344709396362, "epoch": 7, "memory": 15585, "step": 13176} +{"lr": 0.00011711273376198264, "data_time": 0.001546788215637207, "loss": 0.009958622278645635, "time": 0.738587212562561, "epoch": 7, "memory": 15585, "step": 13276} +{"lr": 0.00011445036820691838, "data_time": 0.001619863510131836, "loss": 0.008890557801350951, "time": 0.7163378000259399, "epoch": 7, "memory": 15585, "step": 13376} +{"lr": 0.00011180935111629823, "data_time": 0.0015972375869750977, "loss": 0.0036014584358781574, "time": 0.7818014860153198, "epoch": 7, "memory": 15585, "step": 13476} +{"lr": 0.00010919030515040621, "data_time": 0.0016465902328491211, "loss": 0.00534128830768168, "time": 0.6835239648818969, "epoch": 7, "memory": 15585, "step": 13576} +{"lr": 0.00010659384778949681, "data_time": 0.0017879962921142577, "loss": 0.005843141861259937, "time": 0.7692352294921875, "epoch": 7, "memory": 15585, "step": 13676} +{"lr": 0.00010402059118821447, "data_time": 0.001724576950073242, "loss": 0.009580270061269403, "time": 0.7655837059020996, "epoch": 7, "memory": 15585, "step": 13776} +{"lr": 0.00010147114203126853, "data_time": 0.0016311168670654296, "loss": 0.006007488537579775, "time": 0.7026114940643311, "epoch": 7, "memory": 15585, "step": 13876} +{"lr": 9.89461013903984e-05, "data_time": 0.0016609668731689454, "loss": 0.007340662693604827, "time": 0.684268856048584, "epoch": 7, "memory": 15585, "step": 13976} +{"lr": 9.644606458266133e-05, "data_time": 0.0016548633575439453, "loss": 0.007790833944454789, "time": 0.7047852039337158, "epoch": 7, "memory": 15585, "step": 14076} +{"lr": 9.397162103007764e-05, "data_time": 0.001506471633911133, "loss": 0.008929860172793269, "time": 0.7079825878143311, "epoch": 7, "memory": 15585, "step": 14176} +{"lr": 9.152335412066514e-05, "data_time": 0.0016981601715087891, "loss": 0.00461966791190207, "time": 0.7415994882583619, "epoch": 7, "memory": 15585, "step": 14276} +{"accuracy/top1": 86.31150817871094, "data_time": 0.10922900184256132, "time": 0.5069334819668629, "step": 7} +{"lr": 8.79970864551918e-05, "data_time": 0.1353994846343994, "loss": 0.006859015021473169, "time": 0.7495360136032104, "epoch": 8, "memory": 15585, "step": 14422} +{"lr": 8.56156575096546e-05, "data_time": 0.03434908390045166, "loss": 0.007221652613952756, "time": 0.7029934883117676, "epoch": 8, "memory": 15585, "step": 14522} +{"lr": 8.326237525384291e-05, "data_time": 0.0017881393432617188, "loss": 0.004737551975995302, "time": 0.7513375759124756, "epoch": 8, "memory": 15585, "step": 14622} +{"lr": 8.093779451014054e-05, "data_time": 0.0015610218048095702, "loss": 0.003590864920988679, "time": 0.6822865247726441, "epoch": 8, "memory": 15585, "step": 14722} +{"lr": 7.86424633341094e-05, "data_time": 0.001518583297729492, "loss": 0.004272003518417477, "time": 0.7326555967330932, "epoch": 8, "memory": 15585, "step": 14822} +{"lr": 7.637692288527767e-05, "data_time": 0.0017592191696166992, "loss": 0.00909584192559123, "time": 0.7514014959335327, "epoch": 8, "memory": 15585, "step": 14922} +{"lr": 7.414170729955236e-05, "data_time": 0.0013817548751831055, "loss": 0.004582835081964731, "time": 0.9306327342987061, "epoch": 8, "memory": 15585, "step": 15022} +{"lr": 7.193734356328913e-05, "data_time": 0.0016582489013671875, "loss": 0.004775876412168145, "time": 0.7215326070785523, "epoch": 8, "memory": 15585, "step": 15122} +{"lr": 6.976435138904706e-05, "data_time": 0.0018262863159179688, "loss": 0.006084925401955843, "time": 0.7434223413467407, "epoch": 8, "memory": 15585, "step": 15222} +{"lr": 6.762324309305816e-05, "data_time": 0.0017258405685424804, "loss": 0.009475041879341006, "time": 0.7570581197738647, "epoch": 8, "memory": 15585, "step": 15322} +{"lr": 6.551452347444078e-05, "data_time": 0.0016386032104492188, "loss": 0.006913653155788779, "time": 0.701457405090332, "epoch": 8, "memory": 15585, "step": 15422} +{"lr": 6.34386896961863e-05, "data_time": 0.0016444921493530273, "loss": 0.006224154215306044, "time": 0.7100944280624389, "epoch": 8, "memory": 15585, "step": 15522} +{"lr": 6.139623116794487e-05, "data_time": 0.001488494873046875, "loss": 0.0051321182399988174, "time": 0.7368664026260376, "epoch": 8, "memory": 15585, "step": 15622} +{"lr": 5.938762943063905e-05, "data_time": 0.0021057605743408205, "loss": 0.003891888679936528, "time": 0.7467297792434693, "epoch": 8, "memory": 15585, "step": 15722} +{"lr": 5.741335804293427e-05, "data_time": 0.0017569780349731446, "loss": 0.005946611613035202, "time": 0.7236199855804444, "epoch": 8, "memory": 15585, "step": 15822} +{"lr": 5.5473882469589504e-05, "data_time": 0.0018040895462036132, "loss": 0.0042647217400372025, "time": 0.6820027589797973, "epoch": 8, "memory": 15585, "step": 15922} +{"lr": 5.356965997171673e-05, "data_time": 0.0015529632568359376, "loss": 0.006887713773176074, "time": 0.7614014625549317, "epoch": 8, "memory": 15585, "step": 16022} +{"lr": 5.170113949897451e-05, "data_time": 0.001645064353942871, "loss": 0.007057383051142096, "time": 0.8008774518966675, "epoch": 8, "memory": 15585, "step": 16122} +{"lr": 4.986876158372164e-05, "data_time": 0.0015783309936523438, "loss": 0.00506384982727468, "time": 0.764797592163086, "epoch": 8, "memory": 15585, "step": 16222} +{"lr": 4.8072958237154565e-05, "data_time": 0.0016556501388549805, "loss": 0.007003710838034749, "time": 0.7755881547927856, "epoch": 8, "memory": 15585, "step": 16322} +{"accuracy/top1": 86.88578796386719, "data_time": 0.11372942611819407, "time": 0.5120906868918997, "step": 8} +{"lr": 4.551764078948187e-05, "data_time": 0.19244694709777832, "loss": 0.004175967955961824, "time": 0.7079405546188354, "epoch": 9, "memory": 15585, "step": 16468} +{"lr": 4.38135946476099e-05, "data_time": 0.0016949892044067383, "loss": 0.006547352997586131, "time": 0.8092899799346924, "epoch": 9, "memory": 15585, "step": 16568} +{"lr": 4.214755067289632e-05, "data_time": 0.001466655731201172, "loss": 0.005554694728925824, "time": 0.7897008657455444, "epoch": 9, "memory": 15585, "step": 16668} +{"lr": 4.051990166076003e-05, "data_time": 0.0015451431274414063, "loss": 0.0033359632827341555, "time": 0.7985238313674927, "epoch": 9, "memory": 15585, "step": 16768} +{"lr": 3.893103135441892e-05, "data_time": 0.001585078239440918, "loss": 0.0057896914891898636, "time": 0.7752117872238159, "epoch": 9, "memory": 15585, "step": 16868} +{"lr": 3.7381314354415884e-05, "data_time": 0.0016515254974365234, "loss": 0.007572964346036315, "time": 0.7823155879974365, "epoch": 9, "memory": 15585, "step": 16968} +{"lr": 3.587111603030147e-05, "data_time": 0.0016986370086669923, "loss": 0.006491952342912555, "time": 0.7844523668289185, "epoch": 9, "memory": 15585, "step": 17068} +{"lr": 3.440079243449231e-05, "data_time": 0.0015486717224121095, "loss": 0.005592259066179395, "time": 0.7427172660827637, "epoch": 9, "memory": 15585, "step": 17168} +{"lr": 3.297069021832627e-05, "data_time": 0.001545095443725586, "loss": 0.005130115430802107, "time": 0.7652437210083007, "epoch": 9, "memory": 15585, "step": 17268} +{"lr": 3.158114655033408e-05, "data_time": 0.0015884637832641602, "loss": 0.004831411922350526, "time": 0.7513247966766358, "epoch": 9, "memory": 15585, "step": 17368} +{"lr": 3.023248903674686e-05, "data_time": 0.0014780521392822265, "loss": 0.004853937122970819, "time": 0.7821897745132447, "epoch": 9, "memory": 15585, "step": 17468} +{"lr": 2.8925035644257496e-05, "data_time": 0.0015176773071289063, "loss": 0.005496340431272983, "time": 0.7657095670700074, "epoch": 9, "memory": 15585, "step": 17568} +{"lr": 2.76590946250556e-05, "data_time": 0.0015669822692871093, "loss": 0.00706896954216063, "time": 0.796151065826416, "epoch": 9, "memory": 15585, "step": 17668} +{"lr": 2.6434964444152048e-05, "data_time": 0.0015767335891723633, "loss": 0.005961532332003116, "time": 0.7714215517044067, "epoch": 9, "memory": 15585, "step": 17768} +{"lr": 2.525293370901126e-05, "data_time": 0.0017023086547851562, "loss": 0.008977956883609296, "time": 0.758408522605896, "epoch": 9, "memory": 15585, "step": 17868} +{"lr": 2.411328110150761e-05, "data_time": 0.0017127037048339845, "loss": 0.0072656759526580576, "time": 0.7131024837493897, "epoch": 9, "memory": 15585, "step": 17968} +{"lr": 2.3016275312222135e-05, "data_time": 0.0016339778900146484, "loss": 0.006877522775903344, "time": 0.7824469566345215, "epoch": 9, "memory": 15585, "step": 18068} +{"lr": 2.1962174977094375e-05, "data_time": 0.0015943288803100587, "loss": 0.0030747639015316964, "time": 0.7644901752471924, "epoch": 9, "memory": 15585, "step": 18168} +{"lr": 2.095122861644491e-05, "data_time": 0.0015036344528198242, "loss": 0.003204282792285085, "time": 0.7385929346084594, "epoch": 9, "memory": 15585, "step": 18268} +{"lr": 1.9983674576383036e-05, "data_time": 0.0015795469284057618, "loss": 0.00590423010289669, "time": 0.7436267614364624, "epoch": 9, "memory": 15585, "step": 18368} +{"accuracy/top1": 87.1456298828125, "data_time": 0.10318008016367428, "time": 0.5023639632053063, "step": 9} +{"lr": 1.8649439622232705e-05, "data_time": 0.0014722347259521484, "loss": 0.006848942209035158, "time": 0.7175880908966065, "epoch": 10, "memory": 15585, "step": 18514} +{"lr": 1.7789580193179967e-05, "data_time": 0.0017381668090820312, "loss": 0.007593048457056284, "time": 0.7045023679733277, "epoch": 10, "memory": 15585, "step": 18614} +{"lr": 1.697385849182042e-05, "data_time": 0.0016040563583374023, "loss": 0.003128493158146739, "time": 0.7591366291046142, "epoch": 10, "memory": 15585, "step": 18714} +{"lr": 1.620246683705586e-05, "data_time": 0.001618337631225586, "loss": 0.001843567006289959, "time": 0.7901627540588378, "epoch": 10, "memory": 15585, "step": 18814} +{"lr": 1.547558709629951e-05, "data_time": 0.0017681360244750977, "loss": 0.0034913192968815563, "time": 0.7734734058380127, "epoch": 10, "memory": 15585, "step": 18914} +{"lr": 1.479339064259843e-05, "data_time": 0.0015447378158569337, "loss": 0.0075127596501261, "time": 0.7880001783370971, "epoch": 10, "memory": 15585, "step": 19014} +{"lr": 1.4156038314229283e-05, "data_time": 0.0016036748886108399, "loss": 0.005412493133917451, "time": 0.7852174520492554, "epoch": 10, "memory": 15585, "step": 19114} +{"lr": 1.3563680376778489e-05, "data_time": 0.001610541343688965, "loss": 0.00456148344092071, "time": 0.7876395463943482, "epoch": 10, "memory": 15585, "step": 19214} +{"lr": 1.3016456487714728e-05, "data_time": 0.0017944574356079102, "loss": 0.00857179998420179, "time": 0.8061561822891236, "epoch": 10, "memory": 15585, "step": 19314} +{"lr": 1.2514495663462562e-05, "data_time": 0.0019151687622070313, "loss": 0.006411237828433514, "time": 0.8434019088745117, "epoch": 10, "memory": 15585, "step": 19414} +{"lr": 1.205791624898479e-05, "data_time": 0.001524066925048828, "loss": 0.006115010101348161, "time": 0.7644287824630738, "epoch": 10, "memory": 15585, "step": 19514} +{"lr": 1.1646825889880944e-05, "data_time": 0.0016330718994140626, "loss": 0.004738573217764497, "time": 0.8091838598251343, "epoch": 10, "memory": 15585, "step": 19614} +{"lr": 1.128132150700797e-05, "data_time": 0.0016618490219116211, "loss": 0.0060673046857118605, "time": 0.7980396986007691, "epoch": 10, "memory": 15585, "step": 19714} +{"lr": 1.0961489273629872e-05, "data_time": 0.0017602920532226562, "loss": 0.008298689546063543, "time": 0.7398819923400879, "epoch": 10, "memory": 15585, "step": 19814} +{"lr": 1.0687404595100842e-05, "data_time": 0.0014931678771972657, "loss": 0.00652552735991776, "time": 0.7104421615600586, "epoch": 10, "memory": 15585, "step": 19914} +{"lr": 1.0459132091087482e-05, "data_time": 0.0016926050186157227, "loss": 0.004293518140912056, "time": 0.8109429121017456, "epoch": 10, "memory": 15585, "step": 20014} +{"lr": 1.0276725580333608e-05, "data_time": 0.001483583450317383, "loss": 0.0022668967954814433, "time": 0.7934521198272705, "epoch": 10, "memory": 15585, "step": 20114} +{"lr": 1.0140228067971661e-05, "data_time": 0.0016131401062011719, "loss": 0.0033864120952785013, "time": 0.7601506471633911, "epoch": 10, "memory": 15585, "step": 20214} +{"lr": 1.0049671735383602e-05, "data_time": 0.0016006708145141601, "loss": 0.004824089026078582, "time": 0.7592834234237671, "epoch": 10, "memory": 15585, "step": 20314} +{"lr": 1.000507793261375e-05, "data_time": 0.0016960620880126954, "loss": 0.006116680335253477, "time": 0.7174782752990723, "epoch": 10, "memory": 15585, "step": 20414} +{"accuracy/top1": 87.48324584960938, "data_time": 0.12582454525056433, "time": 0.5218902220491504, "step": 10} diff --git a/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1/20230606_005643/vis_data/vis_image/0000000.png_1.png b/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1/20230606_005643/vis_data/vis_image/0000000.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..68d4cb32e393b3900bae33b6b5d1c27a2e986f70 Binary files /dev/null and b/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1/20230606_005643/vis_data/vis_image/0000000.png_1.png differ diff --git a/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1/20230606_005643/vis_data/vis_image/0000000.png_10.png b/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1/20230606_005643/vis_data/vis_image/0000000.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..9880a00928ef9395999eafef9b4b19bd31b5d3cd Binary files /dev/null and b/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1/20230606_005643/vis_data/vis_image/0000000.png_10.png differ diff --git a/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1/20230606_005643/vis_data/vis_image/0000000.png_2.png b/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1/20230606_005643/vis_data/vis_image/0000000.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..09533de24ec6be40fa19002412730448c62c1879 Binary files /dev/null and b/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1/20230606_005643/vis_data/vis_image/0000000.png_2.png differ diff --git a/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1/20230606_005643/vis_data/vis_image/0000000.png_3.png b/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1/20230606_005643/vis_data/vis_image/0000000.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..5d8cb7ec756ddd673015fdad10baf38511a625c6 Binary files /dev/null and b/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1/20230606_005643/vis_data/vis_image/0000000.png_3.png differ diff --git a/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1/20230606_005643/vis_data/vis_image/0000000.png_4.png b/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1/20230606_005643/vis_data/vis_image/0000000.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..ca2ab103c34934be342a15afc3b1e4717e5362bb Binary files /dev/null and b/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1/20230606_005643/vis_data/vis_image/0000000.png_4.png differ diff --git a/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1/20230606_005643/vis_data/vis_image/0000000.png_5.png b/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1/20230606_005643/vis_data/vis_image/0000000.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..8021c3974825cc673ed6afe93619f3bf75eb9581 Binary files /dev/null and b/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1/20230606_005643/vis_data/vis_image/0000000.png_5.png differ diff --git a/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1/20230606_005643/vis_data/vis_image/0000000.png_6.png b/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1/20230606_005643/vis_data/vis_image/0000000.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..f916f05e0d5268809f0820f15c578a9891e449de Binary files /dev/null and b/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1/20230606_005643/vis_data/vis_image/0000000.png_6.png differ diff --git a/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1/20230606_005643/vis_data/vis_image/0000000.png_7.png b/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1/20230606_005643/vis_data/vis_image/0000000.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..03ce84da34178aa615af2fdc7fcefe78f1405bc9 Binary files /dev/null and b/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1/20230606_005643/vis_data/vis_image/0000000.png_7.png differ diff --git a/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1/20230606_005643/vis_data/vis_image/0000000.png_8.png b/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1/20230606_005643/vis_data/vis_image/0000000.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..44970347e3b3037143cb30932728d05949b9a0e7 Binary files /dev/null and b/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1/20230606_005643/vis_data/vis_image/0000000.png_8.png differ diff --git a/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1/20230606_005643/vis_data/vis_image/0000000.png_9.png b/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1/20230606_005643/vis_data/vis_image/0000000.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..ef4c88160fb0adcce0b0e5cd5b8fc34dbe8a0771 Binary files /dev/null and b/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1/20230606_005643/vis_data/vis_image/0000000.png_9.png differ diff --git a/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1/20230606_005643/vis_data/vis_image/000015337.jpg_1.png b/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1/20230606_005643/vis_data/vis_image/000015337.jpg_1.png new file mode 100644 index 0000000000000000000000000000000000000000..ed4167a7f6c0c58db81c114f9fe147cf6e5a6407 Binary files /dev/null and b/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1/20230606_005643/vis_data/vis_image/000015337.jpg_1.png differ diff --git a/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1/20230606_005643/vis_data/vis_image/000015337.jpg_10.png b/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1/20230606_005643/vis_data/vis_image/000015337.jpg_10.png new file mode 100644 index 0000000000000000000000000000000000000000..b1b94b265a86f23784bee0e82e4b54539f60988a Binary files /dev/null and b/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1/20230606_005643/vis_data/vis_image/000015337.jpg_10.png differ diff --git a/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1/20230606_005643/vis_data/vis_image/000015337.jpg_2.png b/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1/20230606_005643/vis_data/vis_image/000015337.jpg_2.png new file mode 100644 index 0000000000000000000000000000000000000000..8ed0071f3588206ed532f9fb76100e9dc18115a9 Binary files /dev/null and b/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1/20230606_005643/vis_data/vis_image/000015337.jpg_2.png differ diff --git a/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1/20230606_005643/vis_data/vis_image/000015337.jpg_3.png b/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1/20230606_005643/vis_data/vis_image/000015337.jpg_3.png new file mode 100644 index 0000000000000000000000000000000000000000..502a4c5ea9839b4192a0ac4095ec594fa50e295e Binary files /dev/null and b/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1/20230606_005643/vis_data/vis_image/000015337.jpg_3.png differ diff --git a/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1/20230606_005643/vis_data/vis_image/000015337.jpg_4.png b/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1/20230606_005643/vis_data/vis_image/000015337.jpg_4.png new file mode 100644 index 0000000000000000000000000000000000000000..84f079f2d46ced8bb8434d6544258a1df18a9ea2 Binary files /dev/null and b/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1/20230606_005643/vis_data/vis_image/000015337.jpg_4.png differ diff --git a/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1/20230606_005643/vis_data/vis_image/000015337.jpg_5.png b/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1/20230606_005643/vis_data/vis_image/000015337.jpg_5.png new file mode 100644 index 0000000000000000000000000000000000000000..6948bb289648466ac744f1d5bd6eb5c5972a52f6 Binary files /dev/null and b/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1/20230606_005643/vis_data/vis_image/000015337.jpg_5.png differ diff --git a/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1/20230606_005643/vis_data/vis_image/000015337.jpg_6.png b/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1/20230606_005643/vis_data/vis_image/000015337.jpg_6.png new file mode 100644 index 0000000000000000000000000000000000000000..4b40ce6ba0392dd9be663419f920312a7c8425d6 Binary files /dev/null and b/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1/20230606_005643/vis_data/vis_image/000015337.jpg_6.png differ diff --git a/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1/20230606_005643/vis_data/vis_image/000015337.jpg_7.png b/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1/20230606_005643/vis_data/vis_image/000015337.jpg_7.png new file mode 100644 index 0000000000000000000000000000000000000000..e2d01310adb1ed499f21e6624a5687a8b8b0fabc Binary files /dev/null and b/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1/20230606_005643/vis_data/vis_image/000015337.jpg_7.png differ diff --git a/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1/20230606_005643/vis_data/vis_image/000015337.jpg_8.png b/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1/20230606_005643/vis_data/vis_image/000015337.jpg_8.png new file mode 100644 index 0000000000000000000000000000000000000000..017f8701590f0d187253fadf6ea7115a788a49dc Binary files /dev/null and b/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1/20230606_005643/vis_data/vis_image/000015337.jpg_8.png differ diff --git a/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1/20230606_005643/vis_data/vis_image/000015337.jpg_9.png b/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1/20230606_005643/vis_data/vis_image/000015337.jpg_9.png new file mode 100644 index 0000000000000000000000000000000000000000..6a076734d898b01c38f01107868da97b994c67d3 Binary files /dev/null and b/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1/20230606_005643/vis_data/vis_image/000015337.jpg_9.png differ diff --git a/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1/20230606_005643/vis_data/vis_image/0004159.png_1.png b/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1/20230606_005643/vis_data/vis_image/0004159.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..1ca6f688e4b367e290eade509ee50690c105cae6 Binary files /dev/null and b/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1/20230606_005643/vis_data/vis_image/0004159.png_1.png differ diff --git a/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1/20230606_005643/vis_data/vis_image/0004159.png_10.png b/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1/20230606_005643/vis_data/vis_image/0004159.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..08fbf1cb3ee530979e82a53fdd07e0a1293e99c4 Binary files /dev/null and b/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1/20230606_005643/vis_data/vis_image/0004159.png_10.png differ diff --git a/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1/20230606_005643/vis_data/vis_image/0004159.png_2.png b/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1/20230606_005643/vis_data/vis_image/0004159.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..afdd65d7719aaab8db8af29171e1893339143a50 Binary files /dev/null and b/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1/20230606_005643/vis_data/vis_image/0004159.png_2.png differ diff --git a/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1/20230606_005643/vis_data/vis_image/0004159.png_3.png b/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1/20230606_005643/vis_data/vis_image/0004159.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..7852d0a0022de835e9d5812cb60abdc5f6a26a47 Binary files /dev/null and b/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1/20230606_005643/vis_data/vis_image/0004159.png_3.png differ diff --git a/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1/20230606_005643/vis_data/vis_image/0004159.png_4.png b/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1/20230606_005643/vis_data/vis_image/0004159.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..cdeeedc072e1b89d48484243c21918aa965c61ac Binary files /dev/null and b/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1/20230606_005643/vis_data/vis_image/0004159.png_4.png differ diff --git a/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1/20230606_005643/vis_data/vis_image/0004159.png_5.png b/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1/20230606_005643/vis_data/vis_image/0004159.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..d76dc38baf47b10f689411a638fa9747ba6081cc Binary files /dev/null and b/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1/20230606_005643/vis_data/vis_image/0004159.png_5.png differ diff --git a/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1/20230606_005643/vis_data/vis_image/0004159.png_6.png b/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1/20230606_005643/vis_data/vis_image/0004159.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..9bbcd5bd98e2150e7186301bf0389ab7e75195a2 Binary files /dev/null and b/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1/20230606_005643/vis_data/vis_image/0004159.png_6.png differ diff --git a/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1/20230606_005643/vis_data/vis_image/0004159.png_7.png b/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1/20230606_005643/vis_data/vis_image/0004159.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..6189bc5b10dd5fc27862ee5fceb34ccce2de1ea6 Binary files /dev/null and b/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1/20230606_005643/vis_data/vis_image/0004159.png_7.png differ diff --git a/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1/20230606_005643/vis_data/vis_image/0004159.png_8.png b/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1/20230606_005643/vis_data/vis_image/0004159.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..1d67bb6a2fb7e881d3d2519700da712804b0d1f4 Binary files /dev/null and b/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1/20230606_005643/vis_data/vis_image/0004159.png_8.png differ diff --git a/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1/20230606_005643/vis_data/vis_image/0004159.png_9.png b/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1/20230606_005643/vis_data/vis_image/0004159.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..1d18e543413d8fc9d6c0dd1d2bfa73cfa285dd5b Binary files /dev/null and b/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1/20230606_005643/vis_data/vis_image/0004159.png_9.png differ diff --git a/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1/20230606_005643/vis_data/vis_image/0008318.png_1.png b/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1/20230606_005643/vis_data/vis_image/0008318.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..dc73345b5e070b889fff04a680bfd0e5829773eb Binary files /dev/null and b/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1/20230606_005643/vis_data/vis_image/0008318.png_1.png differ diff --git a/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1/20230606_005643/vis_data/vis_image/0008318.png_10.png b/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1/20230606_005643/vis_data/vis_image/0008318.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..4f5e56cfcbcce2c348d3d21d82f74a34fa4feb4b Binary files /dev/null and b/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1/20230606_005643/vis_data/vis_image/0008318.png_10.png differ diff --git a/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1/20230606_005643/vis_data/vis_image/0008318.png_2.png b/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1/20230606_005643/vis_data/vis_image/0008318.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..561708b780869997f6b21beea9d5de78414e4dca Binary files /dev/null and b/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1/20230606_005643/vis_data/vis_image/0008318.png_2.png differ diff --git a/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1/20230606_005643/vis_data/vis_image/0008318.png_3.png b/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1/20230606_005643/vis_data/vis_image/0008318.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..965febd3045e3e711fe767232ca7a9633ed17429 Binary files /dev/null and b/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1/20230606_005643/vis_data/vis_image/0008318.png_3.png differ diff --git a/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1/20230606_005643/vis_data/vis_image/0008318.png_4.png b/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1/20230606_005643/vis_data/vis_image/0008318.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..17a8c1ef87d1d63264ed5ce34eb9d60686472602 Binary files /dev/null and b/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1/20230606_005643/vis_data/vis_image/0008318.png_4.png differ diff --git a/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1/20230606_005643/vis_data/vis_image/0008318.png_5.png b/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1/20230606_005643/vis_data/vis_image/0008318.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..ddbd182150b3f3349349c5afdd39673ca0b8c07f Binary files /dev/null and b/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1/20230606_005643/vis_data/vis_image/0008318.png_5.png differ diff --git a/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1/20230606_005643/vis_data/vis_image/0008318.png_6.png b/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1/20230606_005643/vis_data/vis_image/0008318.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..898aa481efa50fd256f713a27c45476be92bf389 Binary files /dev/null and b/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1/20230606_005643/vis_data/vis_image/0008318.png_6.png differ diff --git a/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1/20230606_005643/vis_data/vis_image/0008318.png_7.png b/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1/20230606_005643/vis_data/vis_image/0008318.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..b64391ac1e81efd3ca377eb7c79ed3b15eb85133 Binary files /dev/null and b/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1/20230606_005643/vis_data/vis_image/0008318.png_7.png differ diff --git a/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1/20230606_005643/vis_data/vis_image/0008318.png_8.png b/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1/20230606_005643/vis_data/vis_image/0008318.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..ff0ef16e83cf27b566411a6a9aeeeee1c7a85f06 Binary files /dev/null and b/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1/20230606_005643/vis_data/vis_image/0008318.png_8.png differ diff --git a/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1/20230606_005643/vis_data/vis_image/0008318.png_9.png b/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1/20230606_005643/vis_data/vis_image/0008318.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..2a9b75e9552d19826d4d5c6bff4d7db42c70d836 Binary files /dev/null and b/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1/20230606_005643/vis_data/vis_image/0008318.png_9.png differ diff --git a/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1.py b/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1.py new file mode 100644 index 0000000000000000000000000000000000000000..f4842b8fcf8c13cf9f7a8b857c19151fda618735 --- /dev/null +++ b/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1.py @@ -0,0 +1,366 @@ +optim_wrapper = dict( + optimizer=dict( + type='AdamW', + lr=0.0004, + weight_decay=0.05, + eps=1e-08, + betas=(0.9, 0.999), + _scope_='mmpretrain'), + paramwise_cfg=dict( + norm_decay_mult=0.0, + bias_decay_mult=0.0, + flat_decay_mult=0.0, + custom_keys=dict({ + '.absolute_pos_embed': dict(decay_mult=0.0), + '.relative_position_bias_table': dict(decay_mult=0.0) + })), + type='AmpOptimWrapper', + dtype='bfloat16', + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=False, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=4096) +model = dict( + type='ImageClassifier', + backbone=dict( + frozen_stages=2, + type='ConvNeXt', + arch='small', + drop_path_rate=0.4, + init_cfg=dict( + type='Pretrained', + checkpoint= + 'https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth', + prefix='backbone')), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=768, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + init_cfg=None), + init_cfg=dict( + type='TruncNormal', layer=['Conv2d', 'Linear'], std=0.02, bias=0.0), + train_cfg=None) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/all_0.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/all_1.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3fake8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/cc1m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3real8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV2-1-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/if-dpmsolver++-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV2-1-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/if-dpmsolver++-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1' diff --git a/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1/epoch_1.pth b/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1/epoch_1.pth new file mode 100644 index 0000000000000000000000000000000000000000..62176504645523d60bc07b619b287a4eea4aa3ca --- /dev/null +++ b/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1/epoch_1.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:eb0cf1b4161cda7434fe73d311d6f481562c6eb16406b51da64493d1fbf96dd9 +size 782128090 diff --git a/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1/epoch_10.pth b/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1/epoch_10.pth new file mode 100644 index 0000000000000000000000000000000000000000..9bf99b9c91d6b01af9efe36a9b09dae46eb6c307 --- /dev/null +++ b/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1/epoch_10.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:32953a35d245c21d734eb5eb1829c6660a8ed1efc26f26a58d79ad70aa3b1e10 +size 783475674 diff --git a/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1/epoch_2.pth b/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1/epoch_2.pth new file mode 100644 index 0000000000000000000000000000000000000000..239a0d9c5ee4f5127bff2c8260248e81b4fd9cdb --- /dev/null +++ b/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1/epoch_2.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:b49eca0b4dbd6568ab33a9545f82ba67001354efa252ab68002188a76f540dd9 +size 782279002 diff --git a/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1/epoch_3.pth b/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1/epoch_3.pth new file mode 100644 index 0000000000000000000000000000000000000000..fa1d520445e9449841439b98725d19ca5d44165b --- /dev/null +++ b/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1/epoch_3.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:e8b5d5df00ffad135957d528d858b4ec271a1b06995b2d08fdc89174bbb14a17 +size 782428826 diff --git a/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1/epoch_4.pth b/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1/epoch_4.pth new file mode 100644 index 0000000000000000000000000000000000000000..645f9fc7634411489caede27fee9ec106a3d9343 --- /dev/null +++ b/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1/epoch_4.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:2c51746275105fe701d23b93565e4463703e5203236049aae7fe30bfb6931a21 +size 782578330 diff --git a/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1/epoch_5.pth b/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1/epoch_5.pth new file mode 100644 index 0000000000000000000000000000000000000000..311589a90e44d87c075508a932f4e4aa18a8852a --- /dev/null +++ b/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1/epoch_5.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:abeb610034e2207eb5607c7f28ce617ae3ed00959c1dbbc1a7d21aeb10cec9f7 +size 782727898 diff --git a/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1/epoch_6.pth b/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1/epoch_6.pth new file mode 100644 index 0000000000000000000000000000000000000000..1d9f7b0aae6dbd9a5d3f9fc778b042d77df6888c --- /dev/null +++ b/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1/epoch_6.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:1c695da817e0094a7717522e228441128e63ea4860ed2a18af1f0b86061f5f98 +size 782877338 diff --git a/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1/epoch_7.pth b/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1/epoch_7.pth new file mode 100644 index 0000000000000000000000000000000000000000..4fdb59edd5ed5df4fc260e88d0124a616e360470 --- /dev/null +++ b/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1/epoch_7.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:9a25f64c52d418b26fde1bdd2132d6406e50266a3dac54a022f311a17313d1d2 +size 783026458 diff --git a/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1/epoch_8.pth b/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1/epoch_8.pth new file mode 100644 index 0000000000000000000000000000000000000000..3c71e72ee7178552b80f08138a21fe0eb1f9e7fa --- /dev/null +++ b/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1/epoch_8.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:df47a9967b1f3ce5a2375060d813f3e10096ff1c1d1805c1a95b73742e327903 +size 783175322 diff --git a/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1/epoch_9.pth b/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1/epoch_9.pth new file mode 100644 index 0000000000000000000000000000000000000000..838e010e958010675bb68170fd80bfa69eb9cca6 --- /dev/null +++ b/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1/epoch_9.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:d4f9bf1510c8d187e3693a9252ed0fef725cca9702c1dee72e62535ae566f9a2 +size 783325082 diff --git a/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1/last_checkpoint b/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1/last_checkpoint new file mode 100644 index 0000000000000000000000000000000000000000..7f08ca69ae96a079fcb325e2598f43bf3893ebaf --- /dev/null +++ b/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1/last_checkpoint @@ -0,0 +1 @@ +/mnt/petrelfs/luzeyu/workspace/fakebench/mmpretrain/workdir/convnext_small_4xb256_all2_1m_lr4e-4_aug_1e-1/epoch_10.pth \ No newline at end of file diff --git a/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1/20230606_010416/20230606_010416.log b/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1/20230606_010416/20230606_010416.log new file mode 100644 index 0000000000000000000000000000000000000000..bf6ba2753cc51092eeb914704ed56c15bef00800 --- /dev/null +++ b/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1/20230606_010416/20230606_010416.log @@ -0,0 +1,1970 @@ +2023/06/06 01:04:20 - mmengine - INFO - +------------------------------------------------------------ +System environment: + sys.platform: linux + Python: 3.10.9 (main, Mar 8 2023, 10:47:38) [GCC 11.2.0] + CUDA available: True + numpy_random_seed: 1894902388 + GPU 0,1,2,3: NVIDIA A100-SXM4-80GB + CUDA_HOME: /mnt/petrelfs/share/cuda-11.6 + NVCC: Cuda compilation tools, release 11.6, V11.6.124 + GCC: gcc (GCC) 7.5.0 + PyTorch: 1.13.1 + PyTorch compiling details: PyTorch built with: + - GCC 9.3 + - C++ Version: 201402 + - Intel(R) oneAPI Math Kernel Library Version 2021.4-Product Build 20210904 for Intel(R) 64 architecture applications + - Intel(R) MKL-DNN v2.6.0 (Git Hash 52b5f107dd9cf10910aaa19cb47f3abf9b349815) + - OpenMP 201511 (a.k.a. OpenMP 4.5) + - LAPACK is enabled (usually provided by MKL) + - NNPACK is enabled + - CPU capability usage: AVX2 + - CUDA Runtime 11.6 + - NVCC architecture flags: -gencode;arch=compute_37,code=sm_37;-gencode;arch=compute_50,code=sm_50;-gencode;arch=compute_60,code=sm_60;-gencode;arch=compute_61,code=sm_61;-gencode;arch=compute_70,code=sm_70;-gencode;arch=compute_75,code=sm_75;-gencode;arch=compute_80,code=sm_80;-gencode;arch=compute_86,code=sm_86;-gencode;arch=compute_37,code=compute_37 + - CuDNN 8.3.2 (built against CUDA 11.5) + - Magma 2.6.1 + - Build settings: BLAS_INFO=mkl, BUILD_TYPE=Release, CUDA_VERSION=11.6, CUDNN_VERSION=8.3.2, CXX_COMPILER=/opt/rh/devtoolset-9/root/usr/bin/c++, CXX_FLAGS= -fabi-version=11 -Wno-deprecated -fvisibility-inlines-hidden -DUSE_PTHREADPOOL -fopenmp -DNDEBUG -DUSE_KINETO -DUSE_FBGEMM -DUSE_QNNPACK -DUSE_PYTORCH_QNNPACK -DUSE_XNNPACK -DSYMBOLICATE_MOBILE_DEBUG_HANDLE -DEDGE_PROFILER_USE_KINETO -O2 -fPIC -Wno-narrowing -Wall -Wextra -Werror=return-type -Werror=non-virtual-dtor -Wno-missing-field-initializers -Wno-type-limits -Wno-array-bounds -Wno-unknown-pragmas -Wunused-local-typedefs -Wno-unused-parameter -Wno-unused-function -Wno-unused-result -Wno-strict-overflow -Wno-strict-aliasing -Wno-error=deprecated-declarations -Wno-stringop-overflow -Wno-psabi -Wno-error=pedantic -Wno-error=redundant-decls -Wno-error=old-style-cast -fdiagnostics-color=always -faligned-new -Wno-unused-but-set-variable -Wno-maybe-uninitialized -fno-math-errno -fno-trapping-math -Werror=format -Werror=cast-function-type -Wno-stringop-overflow, LAPACK_INFO=mkl, PERF_WITH_AVX=1, PERF_WITH_AVX2=1, PERF_WITH_AVX512=1, TORCH_VERSION=1.13.1, USE_CUDA=ON, USE_CUDNN=ON, USE_EXCEPTION_PTR=1, USE_GFLAGS=OFF, USE_GLOG=OFF, USE_MKL=ON, USE_MKLDNN=ON, USE_MPI=OFF, USE_NCCL=ON, USE_NNPACK=ON, USE_OPENMP=ON, USE_ROCM=OFF, + + TorchVision: 0.14.1 + OpenCV: 4.7.0 + MMEngine: 0.7.3 + +Runtime environment: + cudnn_benchmark: True + mp_cfg: {'mp_start_method': 'fork', 'opencv_num_threads': 0} + dist_cfg: {'backend': 'nccl'} + seed: None + deterministic: False + Distributed launcher: slurm + Distributed training: True + GPU number: 4 +------------------------------------------------------------ + +2023/06/06 01:04:25 - mmengine - INFO - Config: +optim_wrapper = dict( + optimizer=dict( + type='AdamW', + lr=0.0004, + weight_decay=0.05, + eps=1e-08, + betas=(0.9, 0.999), + _scope_='mmpretrain'), + paramwise_cfg=dict( + norm_decay_mult=0.0, + bias_decay_mult=0.0, + flat_decay_mult=0.0, + custom_keys=dict({ + '.absolute_pos_embed': dict(decay_mult=0.0), + '.relative_position_bias_table': dict(decay_mult=0.0) + })), + type='AmpOptimWrapper', + dtype='bfloat16', + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=False, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=4096) +model = dict( + type='ImageClassifier', + backbone=dict( + frozen_stages=2, + type='ConvNeXt', + arch='small', + drop_path_rate=0.4, + init_cfg=dict( + type='Pretrained', + checkpoint= + 'https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth', + prefix='backbone')), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=768, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + init_cfg=None), + init_cfg=dict( + type='TruncNormal', layer=['Conv2d', 'Linear'], std=0.02, bias=0.0), + train_cfg=None) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/all_0.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/all_1.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3fake8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/cc1m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3real8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV2-1-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/if-dpmsolver++-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV2-1-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/if-dpmsolver++-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1' + +2023/06/06 01:04:35 - mmengine - INFO - Hooks will be executed in the following order: +before_run: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_load_checkpoint: +(ABOVE_NORMAL) EMAHook + -------------------- +before_train: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_train_epoch: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(NORMAL ) DistSamplerSeedHook + -------------------- +before_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook + -------------------- +after_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +after_train_epoch: +(NORMAL ) IterTimerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_val_epoch: +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook + -------------------- +before_val_iter: +(NORMAL ) IterTimerHook + -------------------- +after_val_iter: +(NORMAL ) IterTimerHook +(NORMAL ) VisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_val_epoch: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_save_checkpoint: +(ABOVE_NORMAL) EMAHook + -------------------- +after_train: +(VERY_LOW ) CheckpointHook + -------------------- +before_test_epoch: +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook + -------------------- +before_test_iter: +(NORMAL ) IterTimerHook + -------------------- +after_test_iter: +(NORMAL ) IterTimerHook +(NORMAL ) VisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_test_epoch: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_run: +(BELOW_NORMAL) LoggerHook + -------------------- +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.2.0.weight:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.2.0.bias:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.2.1.bias:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.3.0.weight:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.3.0.bias:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.3.1.bias:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.2.0.gamma:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.2.0.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.2.0.norm.weight:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.2.0.norm.bias:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.2.0.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.2.0.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.2.1.gamma:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.2.1.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.2.1.norm.weight:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.2.1.norm.bias:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.2.1.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.2.1.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.2.2.gamma:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.2.2.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.2.2.norm.weight:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.2.2.norm.bias:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.2.2.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.2.2.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.2.3.gamma:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.2.3.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.2.3.norm.weight:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.2.3.norm.bias:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.2.3.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.2.3.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.2.4.gamma:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.2.4.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.2.4.norm.weight:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.2.4.norm.bias:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.2.4.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.2.4.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.2.5.gamma:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.2.5.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.2.5.norm.weight:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.2.5.norm.bias:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.2.5.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.2.5.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.2.6.gamma:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.2.6.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.2.6.norm.weight:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.2.6.norm.bias:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.2.6.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.2.6.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.2.7.gamma:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.2.7.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.2.7.norm.weight:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.2.7.norm.bias:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.2.7.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.2.7.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.2.8.gamma:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.2.8.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.2.8.norm.weight:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.2.8.norm.bias:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.2.8.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.2.8.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.2.9.gamma:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.2.9.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.2.9.norm.weight:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.2.9.norm.bias:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.2.9.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.2.9.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.2.10.gamma:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.2.10.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.2.10.norm.weight:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.2.10.norm.bias:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.2.10.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.2.10.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.2.11.gamma:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.2.11.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.2.11.norm.weight:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.2.11.norm.bias:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.2.11.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.2.11.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.2.12.gamma:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.2.12.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.2.12.norm.weight:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.2.12.norm.bias:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.2.12.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.2.12.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.2.13.gamma:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.2.13.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.2.13.norm.weight:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.2.13.norm.bias:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.2.13.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.2.13.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.2.14.gamma:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.2.14.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.2.14.norm.weight:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.2.14.norm.bias:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.2.14.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.2.14.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.2.15.gamma:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.2.15.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.2.15.norm.weight:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.2.15.norm.bias:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.2.15.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.2.15.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.2.16.gamma:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.2.16.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.2.16.norm.weight:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.2.16.norm.bias:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.2.16.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.2.16.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.2.17.gamma:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.2.17.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.2.17.norm.weight:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.2.17.norm.bias:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.2.17.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.2.17.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.2.18.gamma:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.2.18.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.2.18.norm.weight:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.2.18.norm.bias:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.2.18.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.2.18.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.2.19.gamma:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.2.19.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.2.19.norm.weight:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.2.19.norm.bias:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.2.19.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.2.19.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.2.20.gamma:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.2.20.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.2.20.norm.weight:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.2.20.norm.bias:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.2.20.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.2.20.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.2.21.gamma:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.2.21.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.2.21.norm.weight:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.2.21.norm.bias:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.2.21.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.2.21.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.2.22.gamma:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.2.22.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.2.22.norm.weight:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.2.22.norm.bias:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.2.22.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.2.22.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.2.23.gamma:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.2.23.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.2.23.norm.weight:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.2.23.norm.bias:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.2.23.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.2.23.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.2.24.gamma:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.2.24.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.2.24.norm.weight:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.2.24.norm.bias:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.2.24.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.2.24.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.2.25.gamma:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.2.25.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.2.25.norm.weight:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.2.25.norm.bias:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.2.25.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.2.25.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.2.26.gamma:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.2.26.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.2.26.norm.weight:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.2.26.norm.bias:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.2.26.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.2.26.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.3.0.gamma:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.3.0.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.3.0.norm.weight:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.3.0.norm.bias:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.3.0.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.3.0.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.3.1.gamma:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.3.1.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.3.1.norm.weight:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.3.1.norm.bias:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.3.1.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.3.1.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.3.2.gamma:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.3.2.depthwise_conv.bias:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.3.2.norm.weight:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.3.2.norm.bias:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.3.2.pointwise_conv1.bias:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.stages.3.2.pointwise_conv2.bias:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.norm3.weight:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- backbone.norm3.bias:weight_decay=0.0 +2023/06/06 01:04:50 - mmengine - INFO - paramwise_options -- head.fc.bias:weight_decay=0.0 +2023/06/06 01:04:55 - mmengine - INFO - load backbone in model from: https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth +Name of parameter - Initialization information + +backbone.downsample_layers.0.0.weight - torch.Size([96, 3, 4, 4]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.0.0.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.0.1.weight - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.0.1.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.1.0.weight - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.1.0.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.1.1.weight - torch.Size([192, 96, 2, 2]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.1.1.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.2.0.weight - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.2.0.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.2.1.weight - torch.Size([384, 192, 2, 2]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.2.1.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.3.0.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.3.0.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.3.1.weight - torch.Size([768, 384, 2, 2]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.3.1.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.0.gamma - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.0.depthwise_conv.weight - torch.Size([96, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.0.depthwise_conv.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.0.norm.weight - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.0.norm.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.0.pointwise_conv1.weight - torch.Size([384, 96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.0.pointwise_conv1.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.0.pointwise_conv2.weight - torch.Size([96, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.0.pointwise_conv2.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.1.gamma - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.1.depthwise_conv.weight - torch.Size([96, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.1.depthwise_conv.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.1.norm.weight - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.1.norm.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.1.pointwise_conv1.weight - torch.Size([384, 96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.1.pointwise_conv1.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.1.pointwise_conv2.weight - torch.Size([96, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.1.pointwise_conv2.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.2.gamma - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.2.depthwise_conv.weight - torch.Size([96, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.2.depthwise_conv.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.2.norm.weight - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.2.norm.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.2.pointwise_conv1.weight - torch.Size([384, 96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.2.pointwise_conv1.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.2.pointwise_conv2.weight - torch.Size([96, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.2.pointwise_conv2.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.0.gamma - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.0.depthwise_conv.weight - torch.Size([192, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.0.depthwise_conv.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.0.norm.weight - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.0.norm.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.0.pointwise_conv1.weight - torch.Size([768, 192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.0.pointwise_conv1.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.0.pointwise_conv2.weight - torch.Size([192, 768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.0.pointwise_conv2.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.1.gamma - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.1.depthwise_conv.weight - torch.Size([192, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.1.depthwise_conv.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.1.norm.weight - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.1.norm.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.1.pointwise_conv1.weight - torch.Size([768, 192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.1.pointwise_conv1.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.1.pointwise_conv2.weight - torch.Size([192, 768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.1.pointwise_conv2.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.2.gamma - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.2.depthwise_conv.weight - torch.Size([192, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.2.depthwise_conv.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.2.norm.weight - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.2.norm.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.2.pointwise_conv1.weight - torch.Size([768, 192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.2.pointwise_conv1.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.2.pointwise_conv2.weight - torch.Size([192, 768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.2.pointwise_conv2.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.0.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.0.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.0.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.0.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.0.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.0.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.0.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.0.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.0.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.1.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.1.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.1.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.1.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.1.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.1.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.1.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.1.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.1.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.2.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.2.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.2.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.2.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.2.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.2.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.2.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.2.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.2.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.3.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.3.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.3.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.3.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.3.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.3.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.3.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.3.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.3.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.4.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.4.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.4.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.4.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.4.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.4.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.4.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.4.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.4.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.5.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.5.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.5.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.5.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.5.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.5.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.5.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.5.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.5.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.6.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.6.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.6.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.6.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.6.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.6.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.6.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.6.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.6.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.7.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.7.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.7.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.7.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.7.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.7.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.7.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.7.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.7.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.8.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.8.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.8.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.8.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.8.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.8.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.8.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.8.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.8.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.9.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.9.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.9.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.9.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.9.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.9.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.9.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.9.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.9.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.10.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.10.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.10.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.10.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.10.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.10.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.10.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.10.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.10.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.11.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.11.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.11.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.11.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.11.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.11.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.11.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.11.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.11.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.12.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.12.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.12.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.12.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.12.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.12.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.12.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.12.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.12.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.13.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.13.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.13.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.13.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.13.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.13.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.13.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.13.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.13.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.14.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.14.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.14.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.14.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.14.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.14.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.14.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.14.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.14.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.15.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.15.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.15.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.15.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.15.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.15.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.15.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.15.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.15.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.16.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.16.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.16.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.16.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.16.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.16.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.16.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.16.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.16.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.17.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.17.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.17.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.17.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.17.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.17.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.17.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.17.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.17.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.18.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.18.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.18.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.18.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.18.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.18.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.18.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.18.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.18.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.19.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.19.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.19.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.19.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.19.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.19.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.19.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.19.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.19.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.20.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.20.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.20.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.20.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.20.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.20.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.20.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.20.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.20.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.21.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.21.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.21.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.21.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.21.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.21.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.21.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.21.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.21.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.22.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.22.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.22.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.22.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.22.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.22.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.22.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.22.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.22.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.23.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.23.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.23.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.23.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.23.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.23.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.23.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.23.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.23.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.24.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.24.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.24.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.24.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.24.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.24.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.24.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.24.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.24.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.25.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.25.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.25.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.25.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.25.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.25.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.25.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.25.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.25.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.26.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.26.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.26.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.26.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.26.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.26.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.26.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.26.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.26.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.0.gamma - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.0.depthwise_conv.weight - torch.Size([768, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.0.depthwise_conv.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.0.norm.weight - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.0.norm.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.0.pointwise_conv1.weight - torch.Size([3072, 768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.0.pointwise_conv1.bias - torch.Size([3072]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.0.pointwise_conv2.weight - torch.Size([768, 3072]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.0.pointwise_conv2.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.1.gamma - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.1.depthwise_conv.weight - torch.Size([768, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.1.depthwise_conv.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.1.norm.weight - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.1.norm.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.1.pointwise_conv1.weight - torch.Size([3072, 768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.1.pointwise_conv1.bias - torch.Size([3072]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.1.pointwise_conv2.weight - torch.Size([768, 3072]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.1.pointwise_conv2.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.2.gamma - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.2.depthwise_conv.weight - torch.Size([768, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.2.depthwise_conv.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.2.norm.weight - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.2.norm.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.2.pointwise_conv1.weight - torch.Size([3072, 768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.2.pointwise_conv1.bias - torch.Size([3072]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.2.pointwise_conv2.weight - torch.Size([768, 3072]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.2.pointwise_conv2.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.norm3.weight - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.norm3.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +head.fc.weight - torch.Size([2, 768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +head.fc.bias - torch.Size([2]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 +2023/06/06 01:04:55 - mmengine - WARNING - "FileClient" will be deprecated in future. Please use io functions in https://mmengine.readthedocs.io/en/latest/api/fileio.html#file-io +2023/06/06 01:04:55 - mmengine - WARNING - "HardDiskBackend" is the alias of "LocalBackend" and the former will be deprecated in future. +2023/06/06 01:04:55 - mmengine - INFO - Checkpoints will be saved to /mnt/petrelfs/luzeyu/workspace/fakebench/mmpretrain/workdir/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1. +2023/06/06 01:06:26 - mmengine - INFO - Epoch(train) [1][ 100/2046] lr: 3.9998e-04 eta: 5:08:26 time: 0.7958 data_time: 0.2701 memory: 15587 loss: 0.1392 +2023/06/06 01:07:49 - mmengine - INFO - Epoch(train) [1][ 200/2046] lr: 3.9991e-04 eta: 4:54:26 time: 0.9024 data_time: 0.3738 memory: 15587 loss: 0.1375 +2023/06/06 01:09:10 - mmengine - INFO - Epoch(train) [1][ 300/2046] lr: 3.9979e-04 eta: 4:45:34 time: 0.9035 data_time: 0.3757 memory: 15587 loss: 0.0921 +2023/06/06 01:10:30 - mmengine - INFO - Epoch(train) [1][ 400/2046] lr: 3.9963e-04 eta: 4:40:23 time: 0.8388 data_time: 0.3155 memory: 15587 loss: 0.0873 +2023/06/06 01:11:50 - mmengine - INFO - Epoch(train) [1][ 500/2046] lr: 3.9943e-04 eta: 4:36:21 time: 0.8585 data_time: 0.3260 memory: 15587 loss: 0.0746 +2023/06/06 01:13:10 - mmengine - INFO - Epoch(train) [1][ 600/2046] lr: 3.9918e-04 eta: 4:33:04 time: 0.8172 data_time: 0.2896 memory: 15587 loss: 0.0638 +2023/06/06 01:14:30 - mmengine - INFO - Epoch(train) [1][ 700/2046] lr: 3.9888e-04 eta: 4:30:26 time: 0.8152 data_time: 0.2835 memory: 15587 loss: 0.0777 +2023/06/06 01:15:49 - mmengine - INFO - Epoch(train) [1][ 800/2046] lr: 3.9853e-04 eta: 4:28:03 time: 0.8127 data_time: 0.2877 memory: 15587 loss: 0.0541 +2023/06/06 01:17:10 - mmengine - INFO - Epoch(train) [1][ 900/2046] lr: 3.9815e-04 eta: 4:26:17 time: 0.7819 data_time: 0.2367 memory: 15587 loss: 0.0647 +2023/06/06 01:18:29 - mmengine - INFO - Exp name: convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1_20230606_010416 +2023/06/06 01:18:29 - mmengine - INFO - Epoch(train) [1][1000/2046] lr: 3.9771e-04 eta: 4:24:03 time: 0.8234 data_time: 0.3073 memory: 15587 loss: 0.0585 +2023/06/06 01:19:47 - mmengine - INFO - Epoch(train) [1][1100/2046] lr: 3.9723e-04 eta: 4:21:47 time: 0.7858 data_time: 0.2449 memory: 15587 loss: 0.0581 +2023/06/06 01:21:08 - mmengine - INFO - Epoch(train) [1][1200/2046] lr: 3.9670e-04 eta: 4:20:19 time: 0.8089 data_time: 0.2885 memory: 15587 loss: 0.0499 +2023/06/06 01:22:26 - mmengine - INFO - Epoch(train) [1][1300/2046] lr: 3.9613e-04 eta: 4:18:10 time: 0.8045 data_time: 0.2801 memory: 15587 loss: 0.0488 +2023/06/06 01:23:44 - mmengine - INFO - Epoch(train) [1][1400/2046] lr: 3.9552e-04 eta: 4:16:15 time: 0.7633 data_time: 0.2436 memory: 15587 loss: 0.0486 +2023/06/06 01:25:05 - mmengine - INFO - Epoch(train) [1][1500/2046] lr: 3.9486e-04 eta: 4:14:59 time: 0.7957 data_time: 0.2661 memory: 15587 loss: 0.0547 +2023/06/06 01:27:57 - mmengine - INFO - Epoch(train) [1][1600/2046] lr: 3.9415e-04 eta: 4:31:29 time: 0.7911 data_time: 0.2749 memory: 15587 loss: 0.0342 +2023/06/06 01:29:17 - mmengine - INFO - Epoch(train) [1][1700/2046] lr: 3.9340e-04 eta: 4:28:57 time: 0.7990 data_time: 0.2700 memory: 15587 loss: 0.0482 +2023/06/06 01:30:39 - mmengine - INFO - Epoch(train) [1][1800/2046] lr: 3.9261e-04 eta: 4:26:41 time: 0.7877 data_time: 0.2718 memory: 15587 loss: 0.0437 +2023/06/06 01:31:58 - mmengine - INFO - Epoch(train) [1][1900/2046] lr: 3.9177e-04 eta: 4:24:15 time: 0.7656 data_time: 0.2469 memory: 15587 loss: 0.0473 +2023/06/06 01:33:18 - mmengine - INFO - Exp name: convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1_20230606_010416 +2023/06/06 01:33:18 - mmengine - INFO - Epoch(train) [1][2000/2046] lr: 3.9089e-04 eta: 4:21:58 time: 0.8185 data_time: 0.3026 memory: 15587 loss: 0.0451 +2023/06/06 01:33:54 - mmengine - INFO - Exp name: convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1_20230606_010416 +2023/06/06 01:33:54 - mmengine - INFO - Saving checkpoint at 1 epochs +2023/06/06 01:34:56 - mmengine - INFO - Epoch(val) [1][60/60] accuracy/top1: 38.9868 single-label/precision_classwise: [99.67372131347656, 25.887393951416016] single-label/recall_classwise: [22.498579025268555, 99.72868347167969] single-label/f1-score_classwise: [36.71072769165039, 41.10486602783203] data_time: 0.1516 time: 0.5884 +2023/06/06 01:36:20 - mmengine - INFO - Epoch(train) [2][ 100/2046] lr: 3.8952e-04 eta: 4:19:24 time: 0.7479 data_time: 0.2298 memory: 15587 loss: 0.0458 +2023/06/06 01:37:40 - mmengine - INFO - Epoch(train) [2][ 200/2046] lr: 3.8853e-04 eta: 4:17:13 time: 0.8239 data_time: 0.3026 memory: 15585 loss: 0.0549 +2023/06/06 01:38:56 - mmengine - INFO - Epoch(train) [2][ 300/2046] lr: 3.8750e-04 eta: 4:14:42 time: 0.7945 data_time: 0.2779 memory: 15585 loss: 0.0457 +2023/06/06 01:40:15 - mmengine - INFO - Epoch(train) [2][ 400/2046] lr: 3.8642e-04 eta: 4:12:41 time: 0.8379 data_time: 0.3167 memory: 15585 loss: 0.0313 +2023/06/06 01:41:34 - mmengine - INFO - Epoch(train) [2][ 500/2046] lr: 3.8530e-04 eta: 4:10:40 time: 0.8049 data_time: 0.2866 memory: 15585 loss: 0.0441 +2023/06/06 01:42:56 - mmengine - INFO - Epoch(train) [2][ 600/2046] lr: 3.8414e-04 eta: 4:09:03 time: 0.8776 data_time: 0.3547 memory: 15585 loss: 0.0409 +2023/06/06 01:44:14 - mmengine - INFO - Epoch(train) [2][ 700/2046] lr: 3.8293e-04 eta: 4:07:03 time: 0.7312 data_time: 0.2153 memory: 15585 loss: 0.0416 +2023/06/06 01:45:35 - mmengine - INFO - Epoch(train) [2][ 800/2046] lr: 3.8169e-04 eta: 4:05:22 time: 0.7886 data_time: 0.2724 memory: 15585 loss: 0.0519 +2023/06/06 01:46:56 - mmengine - INFO - Epoch(train) [2][ 900/2046] lr: 3.8040e-04 eta: 4:03:41 time: 0.7950 data_time: 0.2791 memory: 15585 loss: 0.0410 +2023/06/06 01:47:43 - mmengine - INFO - Exp name: convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1_20230606_010416 +2023/06/06 01:48:18 - mmengine - INFO - Epoch(train) [2][1000/2046] lr: 3.7907e-04 eta: 4:02:10 time: 0.8115 data_time: 0.2835 memory: 15585 loss: 0.0353 +2023/06/06 01:49:40 - mmengine - INFO - Epoch(train) [2][1100/2046] lr: 3.7770e-04 eta: 4:00:37 time: 0.8389 data_time: 0.3216 memory: 15585 loss: 0.0425 +2023/06/06 01:51:01 - mmengine - INFO - Epoch(train) [2][1200/2046] lr: 3.7629e-04 eta: 3:59:00 time: 0.8844 data_time: 0.3698 memory: 15585 loss: 0.0341 +2023/06/06 01:52:22 - mmengine - INFO - Epoch(train) [2][1300/2046] lr: 3.7484e-04 eta: 3:57:24 time: 0.7872 data_time: 0.2738 memory: 15585 loss: 0.0327 +2023/06/06 01:53:41 - mmengine - INFO - Epoch(train) [2][1400/2046] lr: 3.7335e-04 eta: 3:55:45 time: 0.7798 data_time: 0.2642 memory: 15585 loss: 0.0389 +2023/06/06 01:54:59 - mmengine - INFO - Epoch(train) [2][1500/2046] lr: 3.7182e-04 eta: 3:53:55 time: 0.7733 data_time: 0.2552 memory: 15585 loss: 0.0428 +2023/06/06 01:56:19 - mmengine - INFO - Epoch(train) [2][1600/2046] lr: 3.7025e-04 eta: 3:52:19 time: 0.7788 data_time: 0.2633 memory: 15585 loss: 0.0354 +2023/06/06 01:57:40 - mmengine - INFO - Epoch(train) [2][1700/2046] lr: 3.6864e-04 eta: 3:50:47 time: 0.7819 data_time: 0.2623 memory: 15585 loss: 0.0395 +2023/06/06 01:58:59 - mmengine - INFO - Epoch(train) [2][1800/2046] lr: 3.6699e-04 eta: 3:49:07 time: 0.7705 data_time: 0.2537 memory: 15585 loss: 0.0275 +2023/06/06 02:00:19 - mmengine - INFO - Epoch(train) [2][1900/2046] lr: 3.6531e-04 eta: 3:47:31 time: 0.8427 data_time: 0.3259 memory: 15585 loss: 0.0341 +2023/06/06 02:01:03 - mmengine - INFO - Exp name: convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1_20230606_010416 +2023/06/06 02:01:37 - mmengine - INFO - Epoch(train) [2][2000/2046] lr: 3.6358e-04 eta: 3:45:50 time: 0.7408 data_time: 0.2269 memory: 15585 loss: 0.0423 +2023/06/06 02:02:11 - mmengine - INFO - Exp name: convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1_20230606_010416 +2023/06/06 02:02:11 - mmengine - INFO - Saving checkpoint at 2 epochs +2023/06/06 02:03:08 - mmengine - INFO - Epoch(val) [2][60/60] accuracy/top1: 72.2705 single-label/precision_classwise: [99.87356567382812, 43.48311233520508] single-label/recall_classwise: [64.82545471191406, 99.69766998291016] single-label/f1-score_classwise: [78.62036895751953, 60.55512237548828] data_time: 0.1178 time: 0.5150 +2023/06/06 02:04:34 - mmengine - INFO - Epoch(train) [3][ 100/2046] lr: 3.6100e-04 eta: 3:43:48 time: 0.8040 data_time: 0.2866 memory: 15585 loss: 0.0348 +2023/06/06 02:05:55 - mmengine - INFO - Epoch(train) [3][ 200/2046] lr: 3.5918e-04 eta: 3:42:20 time: 0.9055 data_time: 0.3887 memory: 15585 loss: 0.0368 +2023/06/06 02:07:16 - mmengine - INFO - Epoch(train) [3][ 300/2046] lr: 3.5733e-04 eta: 3:40:51 time: 0.8093 data_time: 0.2925 memory: 15585 loss: 0.0273 +2023/06/06 02:08:32 - mmengine - INFO - Epoch(train) [3][ 400/2046] lr: 3.5544e-04 eta: 3:39:07 time: 0.7472 data_time: 0.2207 memory: 15585 loss: 0.0323 +2023/06/06 02:09:49 - mmengine - INFO - Epoch(train) [3][ 500/2046] lr: 3.5352e-04 eta: 3:37:25 time: 0.7950 data_time: 0.2765 memory: 15585 loss: 0.0304 +2023/06/06 02:11:06 - mmengine - INFO - Epoch(train) [3][ 600/2046] lr: 3.5156e-04 eta: 3:35:47 time: 0.7997 data_time: 0.2783 memory: 15585 loss: 0.0358 +2023/06/06 02:12:24 - mmengine - INFO - Epoch(train) [3][ 700/2046] lr: 3.4957e-04 eta: 3:34:10 time: 0.8007 data_time: 0.2833 memory: 15585 loss: 0.0355 +2023/06/06 02:13:46 - mmengine - INFO - Epoch(train) [3][ 800/2046] lr: 3.4755e-04 eta: 3:32:50 time: 0.8184 data_time: 0.3011 memory: 15585 loss: 0.0334 +2023/06/06 02:15:07 - mmengine - INFO - Epoch(train) [3][ 900/2046] lr: 3.4549e-04 eta: 3:31:23 time: 0.8423 data_time: 0.3268 memory: 15585 loss: 0.0379 +2023/06/06 02:15:13 - mmengine - INFO - Exp name: convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1_20230606_010416 +2023/06/06 02:16:27 - mmengine - INFO - Epoch(train) [3][1000/2046] lr: 3.4339e-04 eta: 3:29:56 time: 0.8398 data_time: 0.3119 memory: 15585 loss: 0.0345 +2023/06/06 02:17:48 - mmengine - INFO - Epoch(train) [3][1100/2046] lr: 3.4127e-04 eta: 3:28:31 time: 0.8348 data_time: 0.3210 memory: 15585 loss: 0.0299 +2023/06/06 02:19:09 - mmengine - INFO - Epoch(train) [3][1200/2046] lr: 3.3911e-04 eta: 3:27:05 time: 0.7952 data_time: 0.2795 memory: 15585 loss: 0.0296 +2023/06/06 02:20:29 - mmengine - INFO - Epoch(train) [3][1300/2046] lr: 3.3692e-04 eta: 3:25:38 time: 0.8414 data_time: 0.3265 memory: 15585 loss: 0.0308 +2023/06/06 02:21:49 - mmengine - INFO - Epoch(train) [3][1400/2046] lr: 3.3470e-04 eta: 3:24:11 time: 0.8437 data_time: 0.3257 memory: 15585 loss: 0.0374 +2023/06/06 02:23:06 - mmengine - INFO - Epoch(train) [3][1500/2046] lr: 3.3245e-04 eta: 3:22:38 time: 0.7892 data_time: 0.2726 memory: 15585 loss: 0.0312 +2023/06/06 02:24:27 - mmengine - INFO - Epoch(train) [3][1600/2046] lr: 3.3017e-04 eta: 3:21:12 time: 0.7879 data_time: 0.2596 memory: 15585 loss: 0.0240 +2023/06/06 02:25:48 - mmengine - INFO - Epoch(train) [3][1700/2046] lr: 3.2786e-04 eta: 3:19:49 time: 0.8338 data_time: 0.3188 memory: 15585 loss: 0.0331 +2023/06/06 02:27:06 - mmengine - INFO - Epoch(train) [3][1800/2046] lr: 3.2552e-04 eta: 3:18:19 time: 0.7656 data_time: 0.2398 memory: 15585 loss: 0.0294 +2023/06/06 02:28:27 - mmengine - INFO - Epoch(train) [3][1900/2046] lr: 3.2315e-04 eta: 3:16:56 time: 0.8108 data_time: 0.2961 memory: 15585 loss: 0.0254 +2023/06/06 02:28:35 - mmengine - INFO - Exp name: convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1_20230606_010416 +2023/06/06 02:29:48 - mmengine - INFO - Epoch(train) [3][2000/2046] lr: 3.2075e-04 eta: 3:15:31 time: 0.7704 data_time: 0.2411 memory: 15585 loss: 0.0244 +2023/06/06 02:30:25 - mmengine - INFO - Exp name: convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1_20230606_010416 +2023/06/06 02:30:25 - mmengine - INFO - Saving checkpoint at 3 epochs +2023/06/06 02:31:22 - mmengine - INFO - Epoch(val) [3][60/60] accuracy/top1: 83.1654 single-label/precision_classwise: [99.92781829833984, 55.92579650878906] single-label/recall_classwise: [78.6524429321289, 99.79069519042969] single-label/f1-score_classwise: [88.02279663085938, 71.679931640625] data_time: 0.1182 time: 0.5168 +2023/06/06 02:32:47 - mmengine - INFO - Epoch(train) [4][ 100/2046] lr: 3.1721e-04 eta: 3:13:38 time: 0.7643 data_time: 0.2498 memory: 15585 loss: 0.0327 +2023/06/06 02:34:14 - mmengine - INFO - Epoch(train) [4][ 200/2046] lr: 3.1474e-04 eta: 3:12:28 time: 0.7231 data_time: 0.2062 memory: 15585 loss: 0.0294 +2023/06/06 02:35:33 - mmengine - INFO - Epoch(train) [4][ 300/2046] lr: 3.1226e-04 eta: 3:11:02 time: 0.7808 data_time: 0.2639 memory: 15585 loss: 0.0378 +2023/06/06 02:36:56 - mmengine - INFO - Epoch(train) [4][ 400/2046] lr: 3.0974e-04 eta: 3:09:43 time: 0.7860 data_time: 0.2659 memory: 15585 loss: 0.0281 +2023/06/06 02:38:22 - mmengine - INFO - Epoch(train) [4][ 500/2046] lr: 3.0721e-04 eta: 3:08:29 time: 0.7929 data_time: 0.2748 memory: 15585 loss: 0.0233 +2023/06/06 02:39:41 - mmengine - INFO - Epoch(train) [4][ 600/2046] lr: 3.0464e-04 eta: 3:07:02 time: 0.8242 data_time: 0.3018 memory: 15585 loss: 0.0314 +2023/06/06 02:41:04 - mmengine - INFO - Epoch(train) [4][ 700/2046] lr: 3.0206e-04 eta: 3:05:42 time: 0.8265 data_time: 0.2968 memory: 15585 loss: 0.0274 +2023/06/06 02:42:26 - mmengine - INFO - Epoch(train) [4][ 800/2046] lr: 2.9945e-04 eta: 3:04:21 time: 0.8951 data_time: 0.3722 memory: 15585 loss: 0.0256 +2023/06/06 02:43:18 - mmengine - INFO - Exp name: convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1_20230606_010416 +2023/06/06 02:43:46 - mmengine - INFO - Epoch(train) [4][ 900/2046] lr: 2.9682e-04 eta: 3:02:56 time: 0.7949 data_time: 0.2791 memory: 15585 loss: 0.0289 +2023/06/06 02:45:09 - mmengine - INFO - Epoch(train) [4][1000/2046] lr: 2.9417e-04 eta: 3:01:37 time: 0.7665 data_time: 0.2461 memory: 15585 loss: 0.0331 +2023/06/06 02:46:30 - mmengine - INFO - Epoch(train) [4][1100/2046] lr: 2.9149e-04 eta: 3:00:13 time: 0.8194 data_time: 0.2930 memory: 15585 loss: 0.0258 +2023/06/06 02:47:52 - mmengine - INFO - Epoch(train) [4][1200/2046] lr: 2.8880e-04 eta: 2:58:52 time: 0.7900 data_time: 0.2735 memory: 15585 loss: 0.0254 +2023/06/06 02:49:14 - mmengine - INFO - Epoch(train) [4][1300/2046] lr: 2.8609e-04 eta: 2:57:29 time: 0.8024 data_time: 0.2727 memory: 15585 loss: 0.0261 +2023/06/06 02:50:35 - mmengine - INFO - Epoch(train) [4][1400/2046] lr: 2.8336e-04 eta: 2:56:06 time: 0.7911 data_time: 0.2670 memory: 15585 loss: 0.0243 +2023/06/06 02:51:55 - mmengine - INFO - Epoch(train) [4][1500/2046] lr: 2.8060e-04 eta: 2:54:43 time: 0.8225 data_time: 0.3055 memory: 15585 loss: 0.0437 +2023/06/06 02:53:16 - mmengine - INFO - Epoch(train) [4][1600/2046] lr: 2.7784e-04 eta: 2:53:20 time: 0.7928 data_time: 0.2729 memory: 15585 loss: 0.0192 +2023/06/06 02:54:38 - mmengine - INFO - Epoch(train) [4][1700/2046] lr: 2.7505e-04 eta: 2:51:58 time: 0.8098 data_time: 0.2932 memory: 15585 loss: 0.0191 +2023/06/06 02:55:59 - mmengine - INFO - Epoch(train) [4][1800/2046] lr: 2.7225e-04 eta: 2:50:34 time: 0.8470 data_time: 0.3285 memory: 15585 loss: 0.0264 +2023/06/06 02:56:50 - mmengine - INFO - Exp name: convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1_20230606_010416 +2023/06/06 02:57:18 - mmengine - INFO - Epoch(train) [4][1900/2046] lr: 2.6943e-04 eta: 2:49:09 time: 0.7181 data_time: 0.1970 memory: 15585 loss: 0.0245 +2023/06/06 02:58:44 - mmengine - INFO - Epoch(train) [4][2000/2046] lr: 2.6660e-04 eta: 2:47:54 time: 0.7924 data_time: 0.2692 memory: 15585 loss: 0.0340 +2023/06/06 02:59:20 - mmengine - INFO - Exp name: convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1_20230606_010416 +2023/06/06 02:59:20 - mmengine - INFO - Saving checkpoint at 4 epochs +2023/06/06 03:00:17 - mmengine - INFO - Epoch(val) [4][60/60] accuracy/top1: 86.0947 single-label/precision_classwise: [99.94638061523438, 60.578548431396484] single-label/recall_classwise: [82.36432647705078, 99.83720397949219] single-label/f1-score_classwise: [90.30755615234375, 75.40397644042969] data_time: 0.1164 time: 0.5140 +2023/06/06 03:01:45 - mmengine - INFO - Epoch(train) [5][ 100/2046] lr: 2.6243e-04 eta: 2:46:01 time: 0.7494 data_time: 0.2231 memory: 15585 loss: 0.0271 +2023/06/06 03:03:03 - mmengine - INFO - Epoch(train) [5][ 200/2046] lr: 2.5956e-04 eta: 2:44:34 time: 0.7844 data_time: 0.2449 memory: 15585 loss: 0.0235 +2023/06/06 03:04:23 - mmengine - INFO - Epoch(train) [5][ 300/2046] lr: 2.5668e-04 eta: 2:43:10 time: 0.7958 data_time: 0.2802 memory: 15585 loss: 0.0338 +2023/06/06 03:05:43 - mmengine - INFO - Epoch(train) [5][ 400/2046] lr: 2.5379e-04 eta: 2:41:45 time: 0.8320 data_time: 0.3009 memory: 15585 loss: 0.0254 +2023/06/06 03:07:04 - mmengine - INFO - Epoch(train) [5][ 500/2046] lr: 2.5089e-04 eta: 2:40:22 time: 0.8451 data_time: 0.3283 memory: 15585 loss: 0.0314 +2023/06/06 03:08:25 - mmengine - INFO - Epoch(train) [5][ 600/2046] lr: 2.4797e-04 eta: 2:39:00 time: 0.8140 data_time: 0.2720 memory: 15585 loss: 0.0209 +2023/06/06 03:09:44 - mmengine - INFO - Epoch(train) [5][ 700/2046] lr: 2.4504e-04 eta: 2:37:35 time: 0.7953 data_time: 0.2794 memory: 15585 loss: 0.0227 +2023/06/06 03:11:06 - mmengine - INFO - Epoch(train) [5][ 800/2046] lr: 2.4211e-04 eta: 2:36:14 time: 0.8188 data_time: 0.2925 memory: 15585 loss: 0.0290 +2023/06/06 03:11:20 - mmengine - INFO - Exp name: convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1_20230606_010416 +2023/06/06 03:12:27 - mmengine - INFO - Epoch(train) [5][ 900/2046] lr: 2.3917e-04 eta: 2:34:51 time: 0.8167 data_time: 0.2998 memory: 15585 loss: 0.0213 +2023/06/06 03:13:51 - mmengine - INFO - Epoch(train) [5][1000/2046] lr: 2.3621e-04 eta: 2:33:32 time: 1.1002 data_time: 0.5579 memory: 15585 loss: 0.0216 +2023/06/06 03:15:11 - mmengine - INFO - Epoch(train) [5][1100/2046] lr: 2.3326e-04 eta: 2:32:08 time: 0.7832 data_time: 0.2640 memory: 15585 loss: 0.0205 +2023/06/06 03:16:31 - mmengine - INFO - Epoch(train) [5][1200/2046] lr: 2.3029e-04 eta: 2:30:45 time: 0.8034 data_time: 0.2760 memory: 15585 loss: 0.0274 +2023/06/06 03:17:51 - mmengine - INFO - Epoch(train) [5][1300/2046] lr: 2.2732e-04 eta: 2:29:21 time: 0.7480 data_time: 0.2330 memory: 15585 loss: 0.0258 +2023/06/06 03:19:10 - mmengine - INFO - Epoch(train) [5][1400/2046] lr: 2.2434e-04 eta: 2:27:57 time: 0.8174 data_time: 0.2954 memory: 15585 loss: 0.0216 +2023/06/06 03:20:31 - mmengine - INFO - Epoch(train) [5][1500/2046] lr: 2.2136e-04 eta: 2:26:34 time: 0.8232 data_time: 0.3069 memory: 15585 loss: 0.0252 +2023/06/06 03:21:52 - mmengine - INFO - Epoch(train) [5][1600/2046] lr: 2.1837e-04 eta: 2:25:12 time: 0.8073 data_time: 0.2897 memory: 15585 loss: 0.0268 +2023/06/06 03:23:21 - mmengine - INFO - Epoch(train) [5][1700/2046] lr: 2.1538e-04 eta: 2:23:58 time: 1.3032 data_time: 0.7872 memory: 15585 loss: 0.0236 +2023/06/06 03:24:41 - mmengine - INFO - Epoch(train) [5][1800/2046] lr: 2.1239e-04 eta: 2:22:34 time: 0.8175 data_time: 0.3025 memory: 15585 loss: 0.0230 +2023/06/06 03:24:55 - mmengine - INFO - Exp name: convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1_20230606_010416 +2023/06/06 03:26:02 - mmengine - INFO - Epoch(train) [5][1900/2046] lr: 2.0940e-04 eta: 2:21:12 time: 0.8159 data_time: 0.2892 memory: 15585 loss: 0.0178 +2023/06/06 03:27:23 - mmengine - INFO - Epoch(train) [5][2000/2046] lr: 2.0641e-04 eta: 2:19:50 time: 0.7444 data_time: 0.2281 memory: 15585 loss: 0.0213 +2023/06/06 03:28:01 - mmengine - INFO - Exp name: convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1_20230606_010416 +2023/06/06 03:28:01 - mmengine - INFO - Saving checkpoint at 5 epochs +2023/06/06 03:29:00 - mmengine - INFO - Epoch(val) [5][60/60] accuracy/top1: 87.5064 single-label/precision_classwise: [99.93753814697266, 63.115840911865234] single-label/recall_classwise: [84.16766357421875, 99.80619812011719] single-label/f1-score_classwise: [91.37720489501953, 77.32965087890625] data_time: 0.1380 time: 0.5391 +2023/06/06 03:30:26 - mmengine - INFO - Epoch(train) [6][ 100/2046] lr: 2.0204e-04 eta: 2:17:56 time: 0.7719 data_time: 0.2560 memory: 15585 loss: 0.0190 +2023/06/06 03:31:46 - mmengine - INFO - Epoch(train) [6][ 200/2046] lr: 1.9904e-04 eta: 2:16:32 time: 0.7724 data_time: 0.2547 memory: 15585 loss: 0.0280 +2023/06/06 03:33:08 - mmengine - INFO - Epoch(train) [6][ 300/2046] lr: 1.9605e-04 eta: 2:15:11 time: 0.8006 data_time: 0.2712 memory: 15585 loss: 0.0235 +2023/06/06 03:34:32 - mmengine - INFO - Epoch(train) [6][ 400/2046] lr: 1.9306e-04 eta: 2:13:51 time: 0.7995 data_time: 0.2791 memory: 15585 loss: 0.0286 +2023/06/06 03:35:53 - mmengine - INFO - Epoch(train) [6][ 500/2046] lr: 1.9007e-04 eta: 2:12:29 time: 0.7788 data_time: 0.2587 memory: 15585 loss: 0.0227 +2023/06/06 03:37:12 - mmengine - INFO - Epoch(train) [6][ 600/2046] lr: 1.8709e-04 eta: 2:11:05 time: 0.7657 data_time: 0.2397 memory: 15585 loss: 0.0214 +2023/06/06 03:38:33 - mmengine - INFO - Epoch(train) [6][ 700/2046] lr: 1.8411e-04 eta: 2:09:43 time: 0.8567 data_time: 0.3359 memory: 15585 loss: 0.0209 +2023/06/06 03:39:33 - mmengine - INFO - Exp name: convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1_20230606_010416 +2023/06/06 03:39:56 - mmengine - INFO - Epoch(train) [6][ 800/2046] lr: 1.8114e-04 eta: 2:08:22 time: 0.8103 data_time: 0.2874 memory: 15585 loss: 0.0126 +2023/06/06 03:41:20 - mmengine - INFO - Epoch(train) [6][ 900/2046] lr: 1.7817e-04 eta: 2:07:02 time: 0.8827 data_time: 0.3654 memory: 15585 loss: 0.0183 +2023/06/06 03:42:40 - mmengine - INFO - Epoch(train) [6][1000/2046] lr: 1.7521e-04 eta: 2:05:39 time: 0.8592 data_time: 0.3379 memory: 15585 loss: 0.0141 +2023/06/06 03:44:00 - mmengine - INFO - Epoch(train) [6][1100/2046] lr: 1.7225e-04 eta: 2:04:16 time: 0.7913 data_time: 0.2708 memory: 15585 loss: 0.0182 +2023/06/06 03:45:21 - mmengine - INFO - Epoch(train) [6][1200/2046] lr: 1.6930e-04 eta: 2:02:54 time: 0.8678 data_time: 0.3423 memory: 15585 loss: 0.0222 +2023/06/06 03:46:42 - mmengine - INFO - Epoch(train) [6][1300/2046] lr: 1.6636e-04 eta: 2:01:31 time: 0.7773 data_time: 0.2553 memory: 15585 loss: 0.0158 +2023/06/06 03:48:04 - mmengine - INFO - Epoch(train) [6][1400/2046] lr: 1.6343e-04 eta: 2:00:10 time: 0.8159 data_time: 0.2897 memory: 15585 loss: 0.0194 +2023/06/06 03:49:24 - mmengine - INFO - Epoch(train) [6][1500/2046] lr: 1.6051e-04 eta: 1:58:47 time: 0.8255 data_time: 0.3055 memory: 15585 loss: 0.0178 +2023/06/06 03:50:48 - mmengine - INFO - Epoch(train) [6][1600/2046] lr: 1.5760e-04 eta: 1:57:28 time: 0.8811 data_time: 0.3614 memory: 15585 loss: 0.0219 +2023/06/06 03:52:07 - mmengine - INFO - Epoch(train) [6][1700/2046] lr: 1.5470e-04 eta: 1:56:04 time: 0.7466 data_time: 0.2287 memory: 15585 loss: 0.0187 +2023/06/06 03:53:03 - mmengine - INFO - Exp name: convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1_20230606_010416 +2023/06/06 03:53:27 - mmengine - INFO - Epoch(train) [6][1800/2046] lr: 1.5182e-04 eta: 1:54:41 time: 0.8182 data_time: 0.2965 memory: 15585 loss: 0.0199 +2023/06/06 03:54:48 - mmengine - INFO - Epoch(train) [6][1900/2046] lr: 1.4894e-04 eta: 1:53:19 time: 0.8054 data_time: 0.2875 memory: 15585 loss: 0.0134 +2023/06/06 03:56:09 - mmengine - INFO - Epoch(train) [6][2000/2046] lr: 1.4608e-04 eta: 1:51:57 time: 0.8736 data_time: 0.3585 memory: 15585 loss: 0.0156 +2023/06/06 03:56:47 - mmengine - INFO - Exp name: convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1_20230606_010416 +2023/06/06 03:56:47 - mmengine - INFO - Saving checkpoint at 6 epochs +2023/06/06 03:57:45 - mmengine - INFO - Epoch(val) [6][60/60] accuracy/top1: 87.8821 single-label/precision_classwise: [99.91557312011719, 63.83844757080078] single-label/recall_classwise: [84.66426849365234, 99.73643493652344] single-label/f1-score_classwise: [91.65984344482422, 77.8483657836914] data_time: 0.1157 time: 0.5134 +2023/06/06 03:59:10 - mmengine - INFO - Epoch(train) [7][ 100/2046] lr: 1.4193e-04 eta: 1:50:01 time: 0.7785 data_time: 0.2619 memory: 15585 loss: 0.0215 +2023/06/06 04:00:31 - mmengine - INFO - Epoch(train) [7][ 200/2046] lr: 1.3910e-04 eta: 1:48:39 time: 0.7597 data_time: 0.2425 memory: 15585 loss: 0.0194 +2023/06/06 04:01:49 - mmengine - INFO - Epoch(train) [7][ 300/2046] lr: 1.3629e-04 eta: 1:47:14 time: 0.7638 data_time: 0.2488 memory: 15585 loss: 0.0135 +2023/06/06 04:03:09 - mmengine - INFO - Epoch(train) [7][ 400/2046] lr: 1.3350e-04 eta: 1:45:52 time: 0.7843 data_time: 0.2680 memory: 15585 loss: 0.0212 +2023/06/06 04:04:31 - mmengine - INFO - Epoch(train) [7][ 500/2046] lr: 1.3072e-04 eta: 1:44:31 time: 0.8909 data_time: 0.3747 memory: 15585 loss: 0.0172 +2023/06/06 04:05:51 - mmengine - INFO - Epoch(train) [7][ 600/2046] lr: 1.2796e-04 eta: 1:43:08 time: 0.8221 data_time: 0.3060 memory: 15585 loss: 0.0148 +2023/06/06 04:07:12 - mmengine - INFO - Epoch(train) [7][ 700/2046] lr: 1.2522e-04 eta: 1:41:46 time: 0.8083 data_time: 0.2801 memory: 15585 loss: 0.0141 +2023/06/06 04:07:31 - mmengine - INFO - Exp name: convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1_20230606_010416 +2023/06/06 04:08:32 - mmengine - INFO - Epoch(train) [7][ 800/2046] lr: 1.2250e-04 eta: 1:40:24 time: 0.7926 data_time: 0.2755 memory: 15585 loss: 0.0181 +2023/06/06 04:09:52 - mmengine - INFO - Epoch(train) [7][ 900/2046] lr: 1.1980e-04 eta: 1:39:01 time: 0.8414 data_time: 0.3056 memory: 15585 loss: 0.0163 +2023/06/06 04:11:13 - mmengine - INFO - Epoch(train) [7][1000/2046] lr: 1.1711e-04 eta: 1:37:39 time: 0.8151 data_time: 0.2962 memory: 15585 loss: 0.0166 +2023/06/06 04:12:33 - mmengine - INFO - Epoch(train) [7][1100/2046] lr: 1.1445e-04 eta: 1:36:17 time: 0.7785 data_time: 0.2505 memory: 15585 loss: 0.0188 +2023/06/06 04:13:57 - mmengine - INFO - Epoch(train) [7][1200/2046] lr: 1.1181e-04 eta: 1:34:56 time: 0.8088 data_time: 0.2909 memory: 15585 loss: 0.0121 +2023/06/06 04:15:25 - mmengine - INFO - Epoch(train) [7][1300/2046] lr: 1.0919e-04 eta: 1:33:38 time: 0.9457 data_time: 0.4291 memory: 15585 loss: 0.0189 +2023/06/06 04:16:44 - mmengine - INFO - Epoch(train) [7][1400/2046] lr: 1.0659e-04 eta: 1:32:15 time: 0.8051 data_time: 0.2921 memory: 15585 loss: 0.0139 +2023/06/06 04:18:05 - mmengine - INFO - Epoch(train) [7][1500/2046] lr: 1.0402e-04 eta: 1:30:53 time: 0.7975 data_time: 0.2597 memory: 15585 loss: 0.0093 +2023/06/06 04:19:26 - mmengine - INFO - Epoch(train) [7][1600/2046] lr: 1.0147e-04 eta: 1:29:31 time: 0.8147 data_time: 0.2983 memory: 15585 loss: 0.0112 +2023/06/06 04:20:45 - mmengine - INFO - Epoch(train) [7][1700/2046] lr: 9.8946e-05 eta: 1:28:09 time: 0.7830 data_time: 0.2527 memory: 15585 loss: 0.0214 +2023/06/06 04:21:06 - mmengine - INFO - Exp name: convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1_20230606_010416 +2023/06/06 04:22:04 - mmengine - INFO - Epoch(train) [7][1800/2046] lr: 9.6446e-05 eta: 1:26:46 time: 0.7633 data_time: 0.2507 memory: 15585 loss: 0.0178 +2023/06/06 04:23:24 - mmengine - INFO - Epoch(train) [7][1900/2046] lr: 9.3972e-05 eta: 1:25:24 time: 0.7969 data_time: 0.2761 memory: 15585 loss: 0.0222 +2023/06/06 04:24:45 - mmengine - INFO - Epoch(train) [7][2000/2046] lr: 9.1523e-05 eta: 1:24:02 time: 0.8107 data_time: 0.2922 memory: 15585 loss: 0.0176 +2023/06/06 04:25:22 - mmengine - INFO - Exp name: convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1_20230606_010416 +2023/06/06 04:25:22 - mmengine - INFO - Saving checkpoint at 7 epochs +2023/06/06 04:26:20 - mmengine - INFO - Epoch(val) [7][60/60] accuracy/top1: 88.1783 single-label/precision_classwise: [99.86663055419922, 64.4394302368164] single-label/recall_classwise: [85.08301544189453, 99.5813980102539] single-label/f1-score_classwise: [91.88397216796875, 78.24578094482422] data_time: 0.1255 time: 0.5257 +2023/06/06 04:27:47 - mmengine - INFO - Epoch(train) [8][ 100/2046] lr: 8.7997e-05 eta: 1:22:05 time: 0.8291 data_time: 0.3117 memory: 15585 loss: 0.0116 +2023/06/06 04:29:07 - mmengine - INFO - Epoch(train) [8][ 200/2046] lr: 8.5616e-05 eta: 1:20:43 time: 0.7554 data_time: 0.2356 memory: 15585 loss: 0.0141 +2023/06/06 04:30:27 - mmengine - INFO - Epoch(train) [8][ 300/2046] lr: 8.3262e-05 eta: 1:19:21 time: 0.8817 data_time: 0.3643 memory: 15585 loss: 0.0121 +2023/06/06 04:31:54 - mmengine - INFO - Epoch(train) [8][ 400/2046] lr: 8.0938e-05 eta: 1:18:01 time: 0.8485 data_time: 0.3289 memory: 15585 loss: 0.0089 +2023/06/06 04:33:14 - mmengine - INFO - Epoch(train) [8][ 500/2046] lr: 7.8642e-05 eta: 1:16:39 time: 0.8497 data_time: 0.3327 memory: 15585 loss: 0.0090 +2023/06/06 04:34:35 - mmengine - INFO - Epoch(train) [8][ 600/2046] lr: 7.6377e-05 eta: 1:15:17 time: 0.7938 data_time: 0.2752 memory: 15585 loss: 0.0114 +2023/06/06 04:35:39 - mmengine - INFO - Exp name: convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1_20230606_010416 +2023/06/06 04:35:56 - mmengine - INFO - Epoch(train) [8][ 700/2046] lr: 7.4142e-05 eta: 1:13:55 time: 0.8439 data_time: 0.3268 memory: 15585 loss: 0.0156 +2023/06/06 04:37:27 - mmengine - INFO - Epoch(train) [8][ 800/2046] lr: 7.1937e-05 eta: 1:12:37 time: 0.8573 data_time: 0.3329 memory: 15585 loss: 0.0137 +2023/06/06 04:38:48 - mmengine - INFO - Epoch(train) [8][ 900/2046] lr: 6.9764e-05 eta: 1:11:15 time: 0.8557 data_time: 0.3374 memory: 15585 loss: 0.0141 +2023/06/06 04:40:09 - mmengine - INFO - Epoch(train) [8][1000/2046] lr: 6.7623e-05 eta: 1:09:54 time: 0.7917 data_time: 0.2645 memory: 15585 loss: 0.0102 +2023/06/06 04:41:28 - mmengine - INFO - Epoch(train) [8][1100/2046] lr: 6.5515e-05 eta: 1:08:31 time: 0.7819 data_time: 0.2654 memory: 15585 loss: 0.0073 +2023/06/06 04:42:49 - mmengine - INFO - Epoch(train) [8][1200/2046] lr: 6.3439e-05 eta: 1:07:09 time: 0.7767 data_time: 0.2581 memory: 15585 loss: 0.0141 +2023/06/06 04:44:08 - mmengine - INFO - Epoch(train) [8][1300/2046] lr: 6.1396e-05 eta: 1:05:47 time: 0.7623 data_time: 0.2449 memory: 15585 loss: 0.0161 +2023/06/06 04:45:30 - mmengine - INFO - Epoch(train) [8][1400/2046] lr: 5.9388e-05 eta: 1:04:25 time: 0.7965 data_time: 0.2801 memory: 15585 loss: 0.0159 +2023/06/06 04:46:51 - mmengine - INFO - Epoch(train) [8][1500/2046] lr: 5.7413e-05 eta: 1:03:03 time: 0.8169 data_time: 0.2992 memory: 15585 loss: 0.0126 +2023/06/06 04:48:11 - mmengine - INFO - Epoch(train) [8][1600/2046] lr: 5.5474e-05 eta: 1:01:41 time: 0.7706 data_time: 0.2459 memory: 15585 loss: 0.0110 +2023/06/06 04:49:14 - mmengine - INFO - Exp name: convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1_20230606_010416 +2023/06/06 04:49:32 - mmengine - INFO - Epoch(train) [8][1700/2046] lr: 5.3570e-05 eta: 1:00:20 time: 0.8012 data_time: 0.2833 memory: 15585 loss: 0.0120 +2023/06/06 04:50:52 - mmengine - INFO - Epoch(train) [8][1800/2046] lr: 5.1701e-05 eta: 0:58:58 time: 0.8191 data_time: 0.2945 memory: 15585 loss: 0.0192 +2023/06/06 04:52:12 - mmengine - INFO - Epoch(train) [8][1900/2046] lr: 4.9869e-05 eta: 0:57:36 time: 0.7935 data_time: 0.2772 memory: 15585 loss: 0.0100 +2023/06/06 04:53:34 - mmengine - INFO - Epoch(train) [8][2000/2046] lr: 4.8073e-05 eta: 0:56:14 time: 0.7931 data_time: 0.2682 memory: 15585 loss: 0.0122 +2023/06/06 04:54:16 - mmengine - INFO - Exp name: convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1_20230606_010416 +2023/06/06 04:54:16 - mmengine - INFO - Saving checkpoint at 8 epochs +2023/06/06 04:55:12 - mmengine - INFO - Epoch(val) [8][60/60] accuracy/top1: 88.5027 single-label/precision_classwise: [99.83541870117188, 65.09917449951172] single-label/recall_classwise: [85.5228042602539, 99.48062133789062] single-label/f1-score_classwise: [92.12653350830078, 78.69867706298828] data_time: 0.1088 time: 0.5064 +2023/06/06 04:56:37 - mmengine - INFO - Epoch(train) [9][ 100/2046] lr: 4.5518e-05 eta: 0:54:17 time: 0.7418 data_time: 0.2235 memory: 15585 loss: 0.0112 +2023/06/06 04:57:57 - mmengine - INFO - Epoch(train) [9][ 200/2046] lr: 4.3814e-05 eta: 0:52:55 time: 0.7792 data_time: 0.2658 memory: 15585 loss: 0.0111 +2023/06/06 04:59:17 - mmengine - INFO - Epoch(train) [9][ 300/2046] lr: 4.2148e-05 eta: 0:51:33 time: 0.7733 data_time: 0.2576 memory: 15585 loss: 0.0114 +2023/06/06 05:00:36 - mmengine - INFO - Epoch(train) [9][ 400/2046] lr: 4.0520e-05 eta: 0:50:11 time: 0.7547 data_time: 0.2291 memory: 15585 loss: 0.0079 +2023/06/06 05:01:53 - mmengine - INFO - Epoch(train) [9][ 500/2046] lr: 3.8931e-05 eta: 0:48:48 time: 0.7559 data_time: 0.2270 memory: 15585 loss: 0.0186 +2023/06/06 05:03:20 - mmengine - INFO - Epoch(train) [9][ 600/2046] lr: 3.7381e-05 eta: 0:47:28 time: 0.7941 data_time: 0.2782 memory: 15585 loss: 0.0102 +2023/06/06 05:03:45 - mmengine - INFO - Exp name: convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1_20230606_010416 +2023/06/06 05:04:40 - mmengine - INFO - Epoch(train) [9][ 700/2046] lr: 3.5871e-05 eta: 0:46:06 time: 0.8424 data_time: 0.3135 memory: 15585 loss: 0.0087 +2023/06/06 05:06:00 - mmengine - INFO - Epoch(train) [9][ 800/2046] lr: 3.4401e-05 eta: 0:44:44 time: 0.8215 data_time: 0.2964 memory: 15585 loss: 0.0124 +2023/06/06 05:07:21 - mmengine - INFO - Epoch(train) [9][ 900/2046] lr: 3.2971e-05 eta: 0:43:23 time: 0.8374 data_time: 0.2991 memory: 15585 loss: 0.0134 +2023/06/06 05:08:41 - mmengine - INFO - Epoch(train) [9][1000/2046] lr: 3.1581e-05 eta: 0:42:01 time: 0.7520 data_time: 0.2319 memory: 15585 loss: 0.0116 +2023/06/06 05:10:03 - mmengine - INFO - Epoch(train) [9][1100/2046] lr: 3.0232e-05 eta: 0:40:39 time: 0.8407 data_time: 0.3095 memory: 15585 loss: 0.0135 +2023/06/06 05:11:24 - mmengine - INFO - Epoch(train) [9][1200/2046] lr: 2.8925e-05 eta: 0:39:18 time: 0.7950 data_time: 0.2786 memory: 15585 loss: 0.0172 +2023/06/06 05:12:46 - mmengine - INFO - Epoch(train) [9][1300/2046] lr: 2.7659e-05 eta: 0:37:56 time: 0.8147 data_time: 0.2874 memory: 15585 loss: 0.0119 +2023/06/06 05:14:07 - mmengine - INFO - Epoch(train) [9][1400/2046] lr: 2.6435e-05 eta: 0:36:35 time: 0.7527 data_time: 0.2351 memory: 15585 loss: 0.0142 +2023/06/06 05:15:29 - mmengine - INFO - Epoch(train) [9][1500/2046] lr: 2.5253e-05 eta: 0:35:13 time: 0.8142 data_time: 0.2828 memory: 15585 loss: 0.0148 +2023/06/06 05:16:49 - mmengine - INFO - Epoch(train) [9][1600/2046] lr: 2.4113e-05 eta: 0:33:51 time: 0.7912 data_time: 0.2741 memory: 15585 loss: 0.0068 +2023/06/06 05:17:15 - mmengine - INFO - Exp name: convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1_20230606_010416 +2023/06/06 05:18:09 - mmengine - INFO - Epoch(train) [9][1700/2046] lr: 2.3016e-05 eta: 0:32:30 time: 0.8068 data_time: 0.2768 memory: 15585 loss: 0.0106 +2023/06/06 05:19:37 - mmengine - INFO - Epoch(train) [9][1800/2046] lr: 2.1962e-05 eta: 0:31:09 time: 0.8448 data_time: 0.3263 memory: 15585 loss: 0.0109 +2023/06/06 05:21:03 - mmengine - INFO - Epoch(train) [9][1900/2046] lr: 2.0951e-05 eta: 0:29:48 time: 0.8141 data_time: 0.2914 memory: 15585 loss: 0.0108 +2023/06/06 05:22:24 - mmengine - INFO - Epoch(train) [9][2000/2046] lr: 1.9984e-05 eta: 0:28:26 time: 0.7526 data_time: 0.2341 memory: 15585 loss: 0.0098 +2023/06/06 05:23:01 - mmengine - INFO - Exp name: convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1_20230606_010416 +2023/06/06 05:23:01 - mmengine - INFO - Saving checkpoint at 9 epochs +2023/06/06 05:23:59 - mmengine - INFO - Epoch(val) [9][60/60] accuracy/top1: 88.9115 single-label/precision_classwise: [99.86563110351562, 65.90559387207031] single-label/recall_classwise: [86.01729583740234, 99.57363891601562] single-label/f1-score_classwise: [92.42561340332031, 79.31460571289062] data_time: 0.1418 time: 0.5377 +2023/06/06 05:25:35 - mmengine - INFO - Epoch(train) [10][ 100/2046] lr: 1.8649e-05 eta: 0:26:29 time: 0.7865 data_time: 0.2441 memory: 15585 loss: 0.0158 +2023/06/06 05:26:56 - mmengine - INFO - Epoch(train) [10][ 200/2046] lr: 1.7790e-05 eta: 0:25:07 time: 0.7988 data_time: 0.2783 memory: 15585 loss: 0.0056 +2023/06/06 05:28:20 - mmengine - INFO - Epoch(train) [10][ 300/2046] lr: 1.6974e-05 eta: 0:23:45 time: 0.8295 data_time: 0.3039 memory: 15585 loss: 0.0134 +2023/06/06 05:29:42 - mmengine - INFO - Epoch(train) [10][ 400/2046] lr: 1.6202e-05 eta: 0:22:24 time: 0.7845 data_time: 0.2657 memory: 15585 loss: 0.0075 +2023/06/06 05:31:05 - mmengine - INFO - Epoch(train) [10][ 500/2046] lr: 1.5476e-05 eta: 0:21:02 time: 0.8048 data_time: 0.2691 memory: 15585 loss: 0.0108 +2023/06/06 05:32:15 - mmengine - INFO - Exp name: convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1_20230606_010416 +2023/06/06 05:32:26 - mmengine - INFO - Epoch(train) [10][ 600/2046] lr: 1.4793e-05 eta: 0:19:40 time: 0.8739 data_time: 0.3581 memory: 15585 loss: 0.0121 +2023/06/06 05:33:45 - mmengine - INFO - Epoch(train) [10][ 700/2046] lr: 1.4156e-05 eta: 0:18:19 time: 0.7648 data_time: 0.2480 memory: 15585 loss: 0.0082 +2023/06/06 05:36:39 - mmengine - INFO - Epoch(train) [10][ 800/2046] lr: 1.3564e-05 eta: 0:17:03 time: 0.8906 data_time: 0.3683 memory: 15585 loss: 0.0104 +2023/06/06 05:38:00 - mmengine - INFO - Epoch(train) [10][ 900/2046] lr: 1.3016e-05 eta: 0:15:41 time: 0.8082 data_time: 0.2902 memory: 15585 loss: 0.0085 +2023/06/06 05:39:32 - mmengine - INFO - Epoch(train) [10][1000/2046] lr: 1.2514e-05 eta: 0:14:19 time: 0.8325 data_time: 0.2941 memory: 15585 loss: 0.0101 +2023/06/06 05:40:53 - mmengine - INFO - Epoch(train) [10][1100/2046] lr: 1.2058e-05 eta: 0:12:57 time: 0.7797 data_time: 0.2604 memory: 15585 loss: 0.0084 +2023/06/06 05:42:14 - mmengine - INFO - Epoch(train) [10][1200/2046] lr: 1.1647e-05 eta: 0:11:35 time: 0.8372 data_time: 0.3092 memory: 15585 loss: 0.0144 +2023/06/06 05:43:36 - mmengine - INFO - Epoch(train) [10][1300/2046] lr: 1.1281e-05 eta: 0:10:13 time: 0.8310 data_time: 0.3144 memory: 15585 loss: 0.0101 +2023/06/06 05:45:01 - mmengine - INFO - Epoch(train) [10][1400/2046] lr: 1.0961e-05 eta: 0:08:50 time: 0.7695 data_time: 0.2354 memory: 15585 loss: 0.0101 +2023/06/06 05:46:17 - mmengine - INFO - Epoch(train) [10][1500/2046] lr: 1.0687e-05 eta: 0:07:28 time: 0.8652 data_time: 0.3482 memory: 15585 loss: 0.0137 +2023/06/06 05:47:29 - mmengine - INFO - Exp name: convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1_20230606_010416 +2023/06/06 05:47:40 - mmengine - INFO - Epoch(train) [10][1600/2046] lr: 1.0459e-05 eta: 0:06:06 time: 0.8284 data_time: 0.3058 memory: 15585 loss: 0.0141 +2023/06/06 05:49:03 - mmengine - INFO - Epoch(train) [10][1700/2046] lr: 1.0277e-05 eta: 0:04:44 time: 0.8370 data_time: 0.3203 memory: 15585 loss: 0.0118 +2023/06/06 05:50:22 - mmengine - INFO - Epoch(train) [10][1800/2046] lr: 1.0140e-05 eta: 0:03:22 time: 0.8042 data_time: 0.2741 memory: 15585 loss: 0.0100 +2023/06/06 05:51:43 - mmengine - INFO - Epoch(train) [10][1900/2046] lr: 1.0050e-05 eta: 0:01:59 time: 0.8743 data_time: 0.3553 memory: 15585 loss: 0.0118 +2023/06/06 05:53:05 - mmengine - INFO - Epoch(train) [10][2000/2046] lr: 1.0005e-05 eta: 0:00:37 time: 0.8159 data_time: 0.2956 memory: 15585 loss: 0.0066 +2023/06/06 05:53:43 - mmengine - INFO - Exp name: convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1_20230606_010416 +2023/06/06 05:53:43 - mmengine - INFO - Saving checkpoint at 10 epochs +2023/06/06 05:54:40 - mmengine - INFO - Epoch(val) [10][60/60] accuracy/top1: 88.9032 single-label/precision_classwise: [99.84857177734375, 65.90010833740234] single-label/recall_classwise: [86.0215072631836, 99.51937866210938] single-label/f1-score_classwise: [92.42073822021484, 79.29341125488281] data_time: 0.1205 time: 0.5190 diff --git a/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1/20230606_010416/vis_data/20230606_010416.json b/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1/20230606_010416/vis_data/20230606_010416.json new file mode 100644 index 0000000000000000000000000000000000000000..05ecf8b06ced248974f9ee3965eeea68bbc27311 --- /dev/null +++ b/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1/20230606_010416/vis_data/20230606_010416.json @@ -0,0 +1,210 @@ +{"lr": 0.00039997747031453974, "data_time": 0.27008748054504395, "loss": 0.13923058956861495, "time": 0.7958432912826539, "epoch": 1, "memory": 15587, "step": 100} +{"lr": 0.0003999089739988646, "data_time": 0.3738239765167236, "loss": 0.13745130524039267, "time": 0.9023931503295899, "epoch": 1, "memory": 15587, "step": 200} +{"lr": 0.00039979452490327365, "data_time": 0.3756752014160156, "loss": 0.09208748266100883, "time": 0.9034847021102905, "epoch": 1, "memory": 15587, "step": 300} +{"lr": 0.0003996341500108966, "data_time": 0.31546545028686523, "loss": 0.08725760839879512, "time": 0.8387650728225708, "epoch": 1, "memory": 15587, "step": 400} +{"lr": 0.00039942788713257383, "data_time": 0.3259701728820801, "loss": 0.07456294409930705, "time": 0.8584913492202759, "epoch": 1, "memory": 15587, "step": 500} +{"lr": 0.00039917578489794236, "data_time": 0.28955626487731934, "loss": 0.06376053467392921, "time": 0.8172077417373658, "epoch": 1, "memory": 15587, "step": 600} +{"lr": 0.00039887790274397046, "data_time": 0.2834655284881592, "loss": 0.07767984047532081, "time": 0.8152337312698364, "epoch": 1, "memory": 15587, "step": 700} +{"lr": 0.00039853431090094445, "data_time": 0.2876667737960815, "loss": 0.05413009580224752, "time": 0.8127007961273194, "epoch": 1, "memory": 15587, "step": 800} +{"lr": 0.0003981450903759111, "data_time": 0.23674638271331788, "loss": 0.06468360386788845, "time": 0.7818946123123169, "epoch": 1, "memory": 15587, "step": 900} +{"lr": 0.00039771033293357867, "data_time": 0.3072946548461914, "loss": 0.058453100919723514, "time": 0.8234207153320312, "epoch": 1, "memory": 15587, "step": 1000} +{"lr": 0.0003972301410746815, "data_time": 0.24485464096069337, "loss": 0.058112204819917676, "time": 0.7857584953308105, "epoch": 1, "memory": 15587, "step": 1100} +{"lr": 0.00039670462801181557, "data_time": 0.2884711265563965, "loss": 0.0498960081487894, "time": 0.8088911056518555, "epoch": 1, "memory": 15587, "step": 1200} +{"lr": 0.0003961339176427448, "data_time": 0.28011374473571776, "loss": 0.04878234844654798, "time": 0.8044605731964112, "epoch": 1, "memory": 15587, "step": 1300} +{"lr": 0.00039551814452119214, "data_time": 0.24360334873199463, "loss": 0.048634478263556956, "time": 0.7632551670074463, "epoch": 1, "memory": 15587, "step": 1400} +{"lr": 0.00039485745382511467, "data_time": 0.2661212682723999, "loss": 0.05470843706279993, "time": 0.795650315284729, "epoch": 1, "memory": 15587, "step": 1500} +{"lr": 0.0003941520013224769, "data_time": 0.2748616933822632, "loss": 0.03421585243195295, "time": 0.791138482093811, "epoch": 1, "memory": 15587, "step": 1600} +{"lr": 0.0003934019533345259, "data_time": 0.2699565887451172, "loss": 0.048215757682919504, "time": 0.7989970207214355, "epoch": 1, "memory": 15587, "step": 1700} +{"lr": 0.00039260748669657817, "data_time": 0.27178153991699217, "loss": 0.04365649037063122, "time": 0.7876599311828614, "epoch": 1, "memory": 15587, "step": 1800} +{"lr": 0.000391768788716328, "data_time": 0.2469374418258667, "loss": 0.04733328409492969, "time": 0.7656367301940918, "epoch": 1, "memory": 15587, "step": 1900} +{"lr": 0.00039088605712968733, "data_time": 0.30262730121612547, "loss": 0.045094087906181815, "time": 0.8185220241546631, "epoch": 1, "memory": 15587, "step": 2000} +{"accuracy/top1": 38.986812591552734, "data_time": 0.15160495042800903, "time": 0.5883608341217041, "step": 1} +{"lr": 0.0003895186269690071, "data_time": 0.22976510524749755, "loss": 0.045768016949296, "time": 0.7478801965713501, "epoch": 2, "memory": 15587, "step": 2146} +{"lr": 0.0003885284798052241, "data_time": 0.30257401466369627, "loss": 0.0549089414998889, "time": 0.8239081382751465, "epoch": 2, "memory": 15585, "step": 2246} +{"lr": 0.0003874950629875727, "data_time": 0.2778758525848389, "loss": 0.04565986730158329, "time": 0.7945029020309449, "epoch": 2, "memory": 15585, "step": 2346} +{"lr": 0.0003864186201599172, "data_time": 0.3167431354522705, "loss": 0.03130431789904833, "time": 0.8378938436508179, "epoch": 2, "memory": 15585, "step": 2446} +{"lr": 0.0003852994051101632, "data_time": 0.2866102457046509, "loss": 0.04411281365901232, "time": 0.8049334526062012, "epoch": 2, "memory": 15585, "step": 2546} +{"lr": 0.0003841376817104243, "data_time": 0.35471739768981936, "loss": 0.040894638188183306, "time": 0.8776494026184082, "epoch": 2, "memory": 15585, "step": 2646} +{"lr": 0.0003829337238548097, "data_time": 0.2152803659439087, "loss": 0.04159433804452419, "time": 0.731187105178833, "epoch": 2, "memory": 15585, "step": 2746} +{"lr": 0.0003816878153948477, "data_time": 0.27244584560394286, "loss": 0.05190552137792111, "time": 0.7886195421218872, "epoch": 2, "memory": 15585, "step": 2846} +{"lr": 0.0003804002500725666, "data_time": 0.2790867805480957, "loss": 0.0409588823094964, "time": 0.7950185537338257, "epoch": 2, "memory": 15585, "step": 2946} +{"lr": 0.0003790713314512378, "data_time": 0.28353979587554934, "loss": 0.03531469237059355, "time": 0.811493992805481, "epoch": 2, "memory": 15585, "step": 3046} +{"lr": 0.00037770137284380876, "data_time": 0.3215766906738281, "loss": 0.04253164660185575, "time": 0.8388626098632812, "epoch": 2, "memory": 15585, "step": 3146} +{"lr": 0.00037629069723903126, "data_time": 0.3698327302932739, "loss": 0.03412234988063574, "time": 0.8844455480575562, "epoch": 2, "memory": 15585, "step": 3246} +{"lr": 0.0003748396372253147, "data_time": 0.27383334636688234, "loss": 0.03273127228021622, "time": 0.7871771574020385, "epoch": 2, "memory": 15585, "step": 3346} +{"lr": 0.0003733485349123126, "data_time": 0.26422815322875975, "loss": 0.03890739511698484, "time": 0.7798113107681275, "epoch": 2, "memory": 15585, "step": 3446} +{"lr": 0.0003718177418502636, "data_time": 0.25523183345794676, "loss": 0.04283151123672724, "time": 0.773280668258667, "epoch": 2, "memory": 15585, "step": 3546} +{"lr": 0.0003702476189471092, "data_time": 0.26333303451538087, "loss": 0.03535877075046301, "time": 0.7787991285324096, "epoch": 2, "memory": 15585, "step": 3646} +{"lr": 0.00036863853638340543, "data_time": 0.26228883266448977, "loss": 0.03946240935474634, "time": 0.7818804502487182, "epoch": 2, "memory": 15585, "step": 3746} +{"lr": 0.000366990873525045, "data_time": 0.2537489414215088, "loss": 0.027454120200127365, "time": 0.7705466032028199, "epoch": 2, "memory": 15585, "step": 3846} +{"lr": 0.00036530501883381603, "data_time": 0.3259198427200317, "loss": 0.03414789345115423, "time": 0.8427048683166504, "epoch": 2, "memory": 15585, "step": 3946} +{"lr": 0.00036358136977581845, "data_time": 0.22690415382385254, "loss": 0.042348215635865924, "time": 0.740772008895874, "epoch": 2, "memory": 15585, "step": 4046} +{"accuracy/top1": 72.27049255371094, "data_time": 0.11775830143787822, "time": 0.514974793449777, "step": 2} +{"lr": 0.0003609978147988783, "data_time": 0.2866313695907593, "loss": 0.03475107345730066, "time": 0.8039616346359253, "epoch": 3, "memory": 15585, "step": 4192} +{"lr": 0.00035918293929014, "data_time": 0.3887096643447876, "loss": 0.03676401153206825, "time": 0.9055249214172363, "epoch": 3, "memory": 15585, "step": 4292} +{"lr": 0.0003573317127887786, "data_time": 0.2925009489059448, "loss": 0.027336132153868674, "time": 0.8093053340911865, "epoch": 3, "memory": 15585, "step": 4392} +{"lr": 0.000355444571749834, "data_time": 0.22066619396209716, "loss": 0.032349448464810845, "time": 0.7472469568252563, "epoch": 3, "memory": 15585, "step": 4492} +{"lr": 0.0003535219610957493, "data_time": 0.2764780044555664, "loss": 0.030434053856879473, "time": 0.7949762105941772, "epoch": 3, "memory": 15585, "step": 4592} +{"lr": 0.00035156433411147255, "data_time": 0.27833940982818606, "loss": 0.035816552862524983, "time": 0.7996628046035766, "epoch": 3, "memory": 15585, "step": 4692} +{"lr": 0.00034957215233758935, "data_time": 0.28330082893371583, "loss": 0.03549886345863342, "time": 0.8007184743881226, "epoch": 3, "memory": 15585, "step": 4792} +{"lr": 0.00034754588546150636, "data_time": 0.3010856628417969, "loss": 0.03336919061839581, "time": 0.818373703956604, "epoch": 3, "memory": 15585, "step": 4892} +{"lr": 0.0003454860112067171, "data_time": 0.32679333686828616, "loss": 0.03792732730507851, "time": 0.8422648191452027, "epoch": 3, "memory": 15585, "step": 4992} +{"lr": 0.00034339301522016903, "data_time": 0.3118792772293091, "loss": 0.034485948644578455, "time": 0.8398421287536622, "epoch": 3, "memory": 15585, "step": 5092} +{"lr": 0.0003412673909577655, "data_time": 0.32096998691558837, "loss": 0.029889604449272154, "time": 0.8347985506057739, "epoch": 3, "memory": 15585, "step": 5192} +{"lr": 0.00033910963956802737, "data_time": 0.2795300245285034, "loss": 0.029570287559181452, "time": 0.7951908349990845, "epoch": 3, "memory": 15585, "step": 5292} +{"lr": 0.00033692026977393767, "data_time": 0.3264751434326172, "loss": 0.030787329375743865, "time": 0.8414042234420777, "epoch": 3, "memory": 15585, "step": 5392} +{"lr": 0.0003346997977530036, "data_time": 0.3256623029708862, "loss": 0.037439318187534806, "time": 0.8437143087387085, "epoch": 3, "memory": 15585, "step": 5492} +{"lr": 0.0003324487470155584, "data_time": 0.27264065742492677, "loss": 0.03120652474462986, "time": 0.7892414093017578, "epoch": 3, "memory": 15585, "step": 5592} +{"lr": 0.0003301676482813382, "data_time": 0.259604811668396, "loss": 0.023959132796153425, "time": 0.7879086017608643, "epoch": 3, "memory": 15585, "step": 5692} +{"lr": 0.00032785703935435476, "data_time": 0.31877496242523196, "loss": 0.0331362733617425, "time": 0.8338300228118897, "epoch": 3, "memory": 15585, "step": 5792} +{"lr": 0.00032551746499609967, "data_time": 0.23977253437042237, "loss": 0.029436669871211053, "time": 0.7656080007553101, "epoch": 3, "memory": 15585, "step": 5892} +{"lr": 0.0003231494767971101, "data_time": 0.29613943099975587, "loss": 0.025362644670531155, "time": 0.8107655048370361, "epoch": 3, "memory": 15585, "step": 5992} +{"lr": 0.0003207536330469211, "data_time": 0.24107058048248292, "loss": 0.024404334463179113, "time": 0.7703856229782104, "epoch": 3, "memory": 15585, "step": 6092} +{"accuracy/top1": 83.16535186767578, "data_time": 0.11824517953591268, "time": 0.5168199851864674, "step": 3} +{"lr": 0.000317206849615549, "data_time": 0.24976627826690673, "loss": 0.0327368400990963, "time": 0.7642760038375854, "epoch": 4, "memory": 15585, "step": 6238} +{"lr": 0.00031474489842163173, "data_time": 0.2061671018600464, "loss": 0.029396084044128655, "time": 0.7230918169021606, "epoch": 4, "memory": 15585, "step": 6338} +{"lr": 0.0003122570731847668, "data_time": 0.2639167785644531, "loss": 0.03776932386681438, "time": 0.7808402299880981, "epoch": 4, "memory": 15585, "step": 6438} +{"lr": 0.00030974396044791, "data_time": 0.26590380668640134, "loss": 0.0281455896794796, "time": 0.7859663248062134, "epoch": 4, "memory": 15585, "step": 6538} +{"lr": 0.0003072061527159319, "data_time": 0.27479000091552735, "loss": 0.02331636203452945, "time": 0.7929321765899658, "epoch": 4, "memory": 15585, "step": 6638} +{"lr": 0.0003046442483159274, "data_time": 0.3018221855163574, "loss": 0.03137182611972093, "time": 0.8242158174514771, "epoch": 4, "memory": 15585, "step": 6738} +{"lr": 0.0003020588512561505, "data_time": 0.2968027353286743, "loss": 0.02740856409072876, "time": 0.8264802694320679, "epoch": 4, "memory": 15585, "step": 6838} +{"lr": 0.0002994505710836101, "data_time": 0.37222654819488527, "loss": 0.02564994916319847, "time": 0.8950869083404541, "epoch": 4, "memory": 15585, "step": 6938} +{"lr": 0.0002968200227403596, "data_time": 0.2791316509246826, "loss": 0.028884059656411408, "time": 0.794876503944397, "epoch": 4, "memory": 15585, "step": 7038} +{"lr": 0.0002941678264185154, "data_time": 0.24613494873046876, "loss": 0.03310949513688684, "time": 0.7665436506271363, "epoch": 4, "memory": 15585, "step": 7138} +{"lr": 0.0002914946074140367, "data_time": 0.2930370569229126, "loss": 0.02582396026700735, "time": 0.8193508625030518, "epoch": 4, "memory": 15585, "step": 7238} +{"lr": 0.00028880099597930275, "data_time": 0.2735331058502197, "loss": 0.025366627145558596, "time": 0.7899934768676757, "epoch": 4, "memory": 15585, "step": 7338} +{"lr": 0.0002860876271745208, "data_time": 0.27272055149078367, "loss": 0.02609073668718338, "time": 0.8024159669876099, "epoch": 4, "memory": 15585, "step": 7438} +{"lr": 0.0002833551407180016, "data_time": 0.26704916954040525, "loss": 0.02427520966157317, "time": 0.7911346435546875, "epoch": 4, "memory": 15585, "step": 7538} +{"lr": 0.00028060418083533477, "data_time": 0.30552144050598146, "loss": 0.04366566501557827, "time": 0.8224530458450318, "epoch": 4, "memory": 15585, "step": 7638} +{"lr": 0.0002778353961075036, "data_time": 0.2729483604431152, "loss": 0.019242750527337194, "time": 0.7928371906280518, "epoch": 4, "memory": 15585, "step": 7738} +{"lr": 0.0002750494393179725, "data_time": 0.29315340518951416, "loss": 0.019132024236023427, "time": 0.8097799777984619, "epoch": 4, "memory": 15585, "step": 7838} +{"lr": 0.00027224696729878206, "data_time": 0.3284644603729248, "loss": 0.026389026921242474, "time": 0.8470310211181641, "epoch": 4, "memory": 15585, "step": 7938} +{"lr": 0.00026942864077569254, "data_time": 0.19699296951293946, "loss": 0.024535698350518942, "time": 0.7180789947509766, "epoch": 4, "memory": 15585, "step": 8038} +{"lr": 0.0002665951242124057, "data_time": 0.2692307949066162, "loss": 0.03399046678096056, "time": 0.7924437999725342, "epoch": 4, "memory": 15585, "step": 8138} +{"accuracy/top1": 86.0947036743164, "data_time": 0.11637008385580094, "time": 0.513966329762193, "step": 4} +{"lr": 0.0002624322962007581, "data_time": 0.22305374145507811, "loss": 0.02709553427994251, "time": 0.7494282007217408, "epoch": 5, "memory": 15585, "step": 8284} +{"lr": 0.0002595642622579449, "data_time": 0.2448746919631958, "loss": 0.023471019277349115, "time": 0.784416675567627, "epoch": 5, "memory": 15585, "step": 8384} +{"lr": 0.0002566833639534589, "data_time": 0.2801742792129517, "loss": 0.03382280953228474, "time": 0.7957663536071777, "epoch": 5, "memory": 15585, "step": 8484} +{"lr": 0.0002537902805032591, "data_time": 0.30092740058898926, "loss": 0.025358874769881367, "time": 0.832024359703064, "epoch": 5, "memory": 15585, "step": 8584} +{"lr": 0.00025088569399614127, "data_time": 0.32829761505126953, "loss": 0.0313843123614788, "time": 0.8450688362121582, "epoch": 5, "memory": 15585, "step": 8684} +{"lr": 0.0002479702892329199, "data_time": 0.272039008140564, "loss": 0.02091716518625617, "time": 0.8140338659286499, "epoch": 5, "memory": 15585, "step": 8784} +{"lr": 0.00024504475356498226, "data_time": 0.27944681644439695, "loss": 0.02268802421167493, "time": 0.7953030347824097, "epoch": 5, "memory": 15585, "step": 8884} +{"lr": 0.0002421097767322291, "data_time": 0.2924706697463989, "loss": 0.028954963386058807, "time": 0.8187745571136474, "epoch": 5, "memory": 15585, "step": 8984} +{"lr": 0.00023916605070046265, "data_time": 0.2997701644897461, "loss": 0.02127212667837739, "time": 0.8167422533035278, "epoch": 5, "memory": 15585, "step": 9084} +{"lr": 0.00023621426949824308, "data_time": 0.5579146146774292, "loss": 0.021598626906052232, "time": 1.1002458810806275, "epoch": 5, "memory": 15585, "step": 9184} +{"lr": 0.0002332551290532596, "data_time": 0.26402647495269777, "loss": 0.02053514327853918, "time": 0.7831912994384765, "epoch": 5, "memory": 15585, "step": 9284} +{"lr": 0.00023028932702825637, "data_time": 0.2759858131408691, "loss": 0.027401977218687534, "time": 0.8034228086471558, "epoch": 5, "memory": 15585, "step": 9384} +{"lr": 0.0002273175626565469, "data_time": 0.2330115795135498, "loss": 0.025782711850479246, "time": 0.748007345199585, "epoch": 5, "memory": 15585, "step": 9484} +{"lr": 0.00022434053657715923, "data_time": 0.29543793201446533, "loss": 0.02157797208055854, "time": 0.8174185037612915, "epoch": 5, "memory": 15585, "step": 9584} +{"lr": 0.00022135895066965, "data_time": 0.30692052841186523, "loss": 0.025195316784083842, "time": 0.823170280456543, "epoch": 5, "memory": 15585, "step": 9684} +{"lr": 0.00021837350788862523, "data_time": 0.2897050380706787, "loss": 0.02679475788027048, "time": 0.8073092460632324, "epoch": 5, "memory": 15585, "step": 9784} +{"lr": 0.00021538491209800788, "data_time": 0.7872004508972168, "loss": 0.02362949438393116, "time": 1.3031886339187622, "epoch": 5, "memory": 15585, "step": 9884} +{"lr": 0.0002123938679050912, "data_time": 0.3025233030319214, "loss": 0.022985117742791772, "time": 0.8174977540969849, "epoch": 5, "memory": 15585, "step": 9984} +{"lr": 0.0002094010804944171, "data_time": 0.28919928073883056, "loss": 0.017819868260994555, "time": 0.8158895969390869, "epoch": 5, "memory": 15585, "step": 10084} +{"lr": 0.00020640725546151753, "data_time": 0.22807033061981202, "loss": 0.021346819400787354, "time": 0.7443963766098023, "epoch": 5, "memory": 15585, "step": 10184} +{"accuracy/top1": 87.50641632080078, "data_time": 0.1380059250065538, "time": 0.5391156947026488, "step": 5} +{"lr": 0.00020203586947950216, "data_time": 0.2559981107711792, "loss": 0.01903561851941049, "time": 0.7719154834747315, "epoch": 6, "memory": 15585, "step": 10330} +{"lr": 0.0001990424959328828, "data_time": 0.2547386884689331, "loss": 0.027998182456940413, "time": 0.7724256753921509, "epoch": 6, "memory": 15585, "step": 10430} +{"lr": 0.00019605052695921703, "data_time": 0.2711981773376465, "loss": 0.023488518362864853, "time": 0.8005754947662354, "epoch": 6, "memory": 15585, "step": 10530} +{"lr": 0.00019306066796107762, "data_time": 0.27912838459014894, "loss": 0.02859131288714707, "time": 0.7995332241058349, "epoch": 6, "memory": 15585, "step": 10630} +{"lr": 0.00019007362384357827, "data_time": 0.2587214469909668, "loss": 0.022684884164482354, "time": 0.7788129329681397, "epoch": 6, "memory": 15585, "step": 10730} +{"lr": 0.00018709009884818184, "data_time": 0.2396928071975708, "loss": 0.021371140610426664, "time": 0.7657092571258545, "epoch": 6, "memory": 15585, "step": 10830} +{"lr": 0.00018411079638666402, "data_time": 0.3359281778335571, "loss": 0.020875070244073868, "time": 0.8567445278167725, "epoch": 6, "memory": 15585, "step": 10930} +{"lr": 0.00018113641887527326, "data_time": 0.2874326467514038, "loss": 0.01259036036208272, "time": 0.8102641582489014, "epoch": 6, "memory": 15585, "step": 11030} +{"lr": 0.00017816766756912557, "data_time": 0.3653801679611206, "loss": 0.01827344549819827, "time": 0.8827403068542481, "epoch": 6, "memory": 15585, "step": 11130} +{"lr": 0.0001752052423968725, "data_time": 0.3379427194595337, "loss": 0.014073522575199605, "time": 0.8592472076416016, "epoch": 6, "memory": 15585, "step": 11230} +{"lr": 0.00017224984179568312, "data_time": 0.2708329439163208, "loss": 0.01823663548566401, "time": 0.7913220643997192, "epoch": 6, "memory": 15585, "step": 11330} +{"lr": 0.00016930216254657575, "data_time": 0.3422927141189575, "loss": 0.022221644502133132, "time": 0.867781400680542, "epoch": 6, "memory": 15585, "step": 11430} +{"lr": 0.0001663628996101415, "data_time": 0.2553313970565796, "loss": 0.01576085495762527, "time": 0.7772531509399414, "epoch": 6, "memory": 15585, "step": 11530} +{"lr": 0.0001634327459626967, "data_time": 0.28970420360565186, "loss": 0.019388765934854744, "time": 0.8159053325653076, "epoch": 6, "memory": 15585, "step": 11630} +{"lr": 0.0001605123924329032, "data_time": 0.3055410861968994, "loss": 0.01775304921902716, "time": 0.8254775524139404, "epoch": 6, "memory": 15585, "step": 11730} +{"lr": 0.0001576025275388945, "data_time": 0.3613670825958252, "loss": 0.021886652009561657, "time": 0.8811241388320923, "epoch": 6, "memory": 15585, "step": 11830} +{"lr": 0.00015470383732594812, "data_time": 0.2286855936050415, "loss": 0.018671649461612105, "time": 0.7465840816497803, "epoch": 6, "memory": 15585, "step": 11930} +{"lr": 0.00015181700520473825, "data_time": 0.29651646614074706, "loss": 0.019918525079265236, "time": 0.8182250261306763, "epoch": 6, "memory": 15585, "step": 12030} +{"lr": 0.00014894271179021254, "data_time": 0.287540340423584, "loss": 0.013365720445290209, "time": 0.8053610801696778, "epoch": 6, "memory": 15585, "step": 12130} +{"lr": 0.0001460816347411264, "data_time": 0.35853650569915774, "loss": 0.015575483674183488, "time": 0.8735523700714112, "epoch": 6, "memory": 15585, "step": 12230} +{"accuracy/top1": 87.8821029663086, "data_time": 0.1156603000203117, "time": 0.5134156062954762, "step": 6} +{"lr": 0.00014192959250590055, "data_time": 0.2618546962738037, "loss": 0.021499492274597288, "time": 0.7784563779830933, "epoch": 7, "memory": 15585, "step": 12376} +{"lr": 0.00013910389161427854, "data_time": 0.24254865646362306, "loss": 0.01944695943966508, "time": 0.7597046613693237, "epoch": 7, "memory": 15585, "step": 12476} +{"lr": 0.00013629372674084914, "data_time": 0.2487579107284546, "loss": 0.013473668042570353, "time": 0.7638455629348755, "epoch": 7, "memory": 15585, "step": 12576} +{"lr": 0.00013349976042508036, "data_time": 0.26796531677246094, "loss": 0.02121005724184215, "time": 0.7842514276504516, "epoch": 7, "memory": 15585, "step": 12676} +{"lr": 0.0001307226513873823, "data_time": 0.37471370697021483, "loss": 0.01720991963520646, "time": 0.8909207582473755, "epoch": 7, "memory": 15585, "step": 12776} +{"lr": 0.00012796305437380283, "data_time": 0.3059714317321777, "loss": 0.014831477496773004, "time": 0.8221205234527588, "epoch": 7, "memory": 15585, "step": 12876} +{"lr": 0.00012522162000166114, "data_time": 0.2801429510116577, "loss": 0.014092972502112389, "time": 0.8082698106765747, "epoch": 7, "memory": 15585, "step": 12976} +{"lr": 0.00012249899460615546, "data_time": 0.27553768157958985, "loss": 0.01806235294789076, "time": 0.7926186561584473, "epoch": 7, "memory": 15585, "step": 13076} +{"lr": 0.00011979582008797885, "data_time": 0.30564844608306885, "loss": 0.016327786445617675, "time": 0.841417932510376, "epoch": 7, "memory": 15585, "step": 13176} +{"lr": 0.00011711273376198264, "data_time": 0.2962301015853882, "loss": 0.01663127182982862, "time": 0.8150619506835938, "epoch": 7, "memory": 15585, "step": 13276} +{"lr": 0.00011445036820691838, "data_time": 0.25047237873077394, "loss": 0.018808535439893602, "time": 0.7785188913345337, "epoch": 7, "memory": 15585, "step": 13376} +{"lr": 0.00011180935111629823, "data_time": 0.2908912181854248, "loss": 0.012106695957481862, "time": 0.8087669134140014, "epoch": 7, "memory": 15585, "step": 13476} +{"lr": 0.00010919030515040621, "data_time": 0.4291386604309082, "loss": 0.018903985805809497, "time": 0.9457412004470825, "epoch": 7, "memory": 15585, "step": 13576} +{"lr": 0.00010659384778949681, "data_time": 0.2921184778213501, "loss": 0.013892869511619211, "time": 0.8050588846206665, "epoch": 7, "memory": 15585, "step": 13676} +{"lr": 0.00010402059118821447, "data_time": 0.2596737384796143, "loss": 0.009289868036285042, "time": 0.7975058555603027, "epoch": 7, "memory": 15585, "step": 13776} +{"lr": 0.00010147114203126853, "data_time": 0.2982635498046875, "loss": 0.011215496947988868, "time": 0.814716386795044, "epoch": 7, "memory": 15585, "step": 13876} +{"lr": 9.89461013903984e-05, "data_time": 0.25265505313873293, "loss": 0.021423447458073498, "time": 0.7830265045166016, "epoch": 7, "memory": 15585, "step": 13976} +{"lr": 9.644606458266133e-05, "data_time": 0.2507262468338013, "loss": 0.01776902177371085, "time": 0.7632845163345336, "epoch": 7, "memory": 15585, "step": 14076} +{"lr": 9.397162103007764e-05, "data_time": 0.2760718107223511, "loss": 0.02215931466780603, "time": 0.7969234704971313, "epoch": 7, "memory": 15585, "step": 14176} +{"lr": 9.152335412066514e-05, "data_time": 0.2921916484832764, "loss": 0.017632154375314714, "time": 0.8106762170791626, "epoch": 7, "memory": 15585, "step": 14276} +{"accuracy/top1": 88.1783447265625, "data_time": 0.12549807204574834, "time": 0.5257302933051938, "step": 7} +{"lr": 8.79970864551918e-05, "data_time": 0.3116596221923828, "loss": 0.011558062583208083, "time": 0.8290964126586914, "epoch": 8, "memory": 15585, "step": 14422} +{"lr": 8.56156575096546e-05, "data_time": 0.23561570644378663, "loss": 0.014075045567005872, "time": 0.7553547620773315, "epoch": 8, "memory": 15585, "step": 14522} +{"lr": 8.326237525384291e-05, "data_time": 0.3643265962600708, "loss": 0.012113719852641225, "time": 0.8816712856292724, "epoch": 8, "memory": 15585, "step": 14622} +{"lr": 8.093779451014054e-05, "data_time": 0.32892110347747805, "loss": 0.008901950716972352, "time": 0.8484573841094971, "epoch": 8, "memory": 15585, "step": 14722} +{"lr": 7.86424633341094e-05, "data_time": 0.3327207326889038, "loss": 0.008992444211617111, "time": 0.8496640682220459, "epoch": 8, "memory": 15585, "step": 14822} +{"lr": 7.637692288527767e-05, "data_time": 0.27522692680358884, "loss": 0.011448814813047648, "time": 0.7937656164169311, "epoch": 8, "memory": 15585, "step": 14922} +{"lr": 7.414170729955236e-05, "data_time": 0.32682440280914304, "loss": 0.015614293608814478, "time": 0.8439353704452515, "epoch": 8, "memory": 15585, "step": 15022} +{"lr": 7.193734356328913e-05, "data_time": 0.33294880390167236, "loss": 0.013695200346410275, "time": 0.8572995901107788, "epoch": 8, "memory": 15585, "step": 15122} +{"lr": 6.976435138904706e-05, "data_time": 0.3373740196228027, "loss": 0.014129743725061417, "time": 0.8557215929031372, "epoch": 8, "memory": 15585, "step": 15222} +{"lr": 6.762324309305816e-05, "data_time": 0.26450512409210203, "loss": 0.01022324119694531, "time": 0.7916579484939575, "epoch": 8, "memory": 15585, "step": 15322} +{"lr": 6.551452347444078e-05, "data_time": 0.2653649091720581, "loss": 0.007250535348430276, "time": 0.7819453477859497, "epoch": 8, "memory": 15585, "step": 15422} +{"lr": 6.34386896961863e-05, "data_time": 0.2581397771835327, "loss": 0.014060096349567175, "time": 0.776689100265503, "epoch": 8, "memory": 15585, "step": 15522} +{"lr": 6.139623116794487e-05, "data_time": 0.244904088973999, "loss": 0.016063893772661687, "time": 0.7623092412948609, "epoch": 8, "memory": 15585, "step": 15622} +{"lr": 5.938762943063905e-05, "data_time": 0.28007290363311765, "loss": 0.015862515289336444, "time": 0.7965219974517822, "epoch": 8, "memory": 15585, "step": 15722} +{"lr": 5.741335804293427e-05, "data_time": 0.2992445707321167, "loss": 0.012573866778984666, "time": 0.8168627262115479, "epoch": 8, "memory": 15585, "step": 15822} +{"lr": 5.5473882469589504e-05, "data_time": 0.24587674140930177, "loss": 0.011037486558780073, "time": 0.7706165790557862, "epoch": 8, "memory": 15585, "step": 15922} +{"lr": 5.356965997171673e-05, "data_time": 0.2832584619522095, "loss": 0.011991947283968329, "time": 0.8011831760406494, "epoch": 8, "memory": 15585, "step": 16022} +{"lr": 5.170113949897451e-05, "data_time": 0.29445812702178953, "loss": 0.019166898727416993, "time": 0.8190693140029908, "epoch": 8, "memory": 15585, "step": 16122} +{"lr": 4.986876158372164e-05, "data_time": 0.2771552562713623, "loss": 0.010040620574727654, "time": 0.7935341358184814, "epoch": 8, "memory": 15585, "step": 16222} +{"lr": 4.8072958237154565e-05, "data_time": 0.26821324825286863, "loss": 0.012176671624183654, "time": 0.7931279897689819, "epoch": 8, "memory": 15585, "step": 16322} +{"accuracy/top1": 88.50272369384766, "data_time": 0.10883679546293665, "time": 0.5064185330125152, "step": 8} +{"lr": 4.551764078948187e-05, "data_time": 0.22350099086761474, "loss": 0.011239055823534728, "time": 0.7417830467224121, "epoch": 9, "memory": 15585, "step": 16468} +{"lr": 4.38135946476099e-05, "data_time": 0.2658128499984741, "loss": 0.011058948514983059, "time": 0.7792037487030029, "epoch": 9, "memory": 15585, "step": 16568} +{"lr": 4.214755067289632e-05, "data_time": 0.25759875774383545, "loss": 0.011372651671990752, "time": 0.7732922792434692, "epoch": 9, "memory": 15585, "step": 16668} +{"lr": 4.051990166076003e-05, "data_time": 0.22912235260009767, "loss": 0.007898762496188283, "time": 0.7547113418579101, "epoch": 9, "memory": 15585, "step": 16768} +{"lr": 3.893103135441892e-05, "data_time": 0.22703702449798585, "loss": 0.018630159180611373, "time": 0.7559427976608276, "epoch": 9, "memory": 15585, "step": 16868} +{"lr": 3.7381314354415884e-05, "data_time": 0.2782020330429077, "loss": 0.010206338297575713, "time": 0.794103455543518, "epoch": 9, "memory": 15585, "step": 16968} +{"lr": 3.587111603030147e-05, "data_time": 0.3134643077850342, "loss": 0.008711456693708897, "time": 0.8424233436584473, "epoch": 9, "memory": 15585, "step": 17068} +{"lr": 3.440079243449231e-05, "data_time": 0.2963653802871704, "loss": 0.012440914800390602, "time": 0.8215030908584595, "epoch": 9, "memory": 15585, "step": 17168} +{"lr": 3.297069021832627e-05, "data_time": 0.2990638971328735, "loss": 0.01341496268287301, "time": 0.8373693466186524, "epoch": 9, "memory": 15585, "step": 17268} +{"lr": 3.158114655033408e-05, "data_time": 0.23194804191589355, "loss": 0.011563332751393318, "time": 0.7520285844802856, "epoch": 9, "memory": 15585, "step": 17368} +{"lr": 3.023248903674686e-05, "data_time": 0.30950255393981935, "loss": 0.01345430500805378, "time": 0.840735650062561, "epoch": 9, "memory": 15585, "step": 17468} +{"lr": 2.8925035644257496e-05, "data_time": 0.2786473989486694, "loss": 0.017220094613730908, "time": 0.7949512004852295, "epoch": 9, "memory": 15585, "step": 17568} +{"lr": 2.76590946250556e-05, "data_time": 0.28743205070495603, "loss": 0.011918843863531947, "time": 0.8146660327911377, "epoch": 9, "memory": 15585, "step": 17668} +{"lr": 2.6434964444152048e-05, "data_time": 0.2351306676864624, "loss": 0.014188050897791982, "time": 0.7527170896530151, "epoch": 9, "memory": 15585, "step": 17768} +{"lr": 2.525293370901126e-05, "data_time": 0.2827725887298584, "loss": 0.0147943832911551, "time": 0.8142024755477906, "epoch": 9, "memory": 15585, "step": 17868} +{"lr": 2.411328110150761e-05, "data_time": 0.27406988143920896, "loss": 0.006789855239912867, "time": 0.7911711454391479, "epoch": 9, "memory": 15585, "step": 17968} +{"lr": 2.3016275312222135e-05, "data_time": 0.27676851749420167, "loss": 0.010608707321807742, "time": 0.8067718267440795, "epoch": 9, "memory": 15585, "step": 18068} +{"lr": 2.1962174977094375e-05, "data_time": 0.3263155221939087, "loss": 0.01089247171767056, "time": 0.8447614908218384, "epoch": 9, "memory": 15585, "step": 18168} +{"lr": 2.095122861644491e-05, "data_time": 0.29137680530548093, "loss": 0.01079562189988792, "time": 0.8140552759170532, "epoch": 9, "memory": 15585, "step": 18268} +{"lr": 1.9983674576383036e-05, "data_time": 0.2340712547302246, "loss": 0.009820595383644104, "time": 0.7525712013244629, "epoch": 9, "memory": 15585, "step": 18368} +{"accuracy/top1": 88.91150665283203, "data_time": 0.14181355570183427, "time": 0.5377230839651139, "step": 9} +{"lr": 1.8649439622232705e-05, "data_time": 0.24411065578460694, "loss": 0.015803095977753402, "time": 0.7865235805511475, "epoch": 10, "memory": 15585, "step": 18514} +{"lr": 1.7789580193179967e-05, "data_time": 0.2782934904098511, "loss": 0.005647242115810513, "time": 0.7988179445266723, "epoch": 10, "memory": 15585, "step": 18614} +{"lr": 1.697385849182042e-05, "data_time": 0.3038593292236328, "loss": 0.013444292452186346, "time": 0.8295249938964844, "epoch": 10, "memory": 15585, "step": 18714} +{"lr": 1.620246683705586e-05, "data_time": 0.265687084197998, "loss": 0.007501485291868448, "time": 0.7844733238220215, "epoch": 10, "memory": 15585, "step": 18814} +{"lr": 1.547558709629951e-05, "data_time": 0.26913881301879883, "loss": 0.01077024326659739, "time": 0.804829478263855, "epoch": 10, "memory": 15585, "step": 18914} +{"lr": 1.479339064259843e-05, "data_time": 0.35813217163085936, "loss": 0.01208462598733604, "time": 0.8739128828048706, "epoch": 10, "memory": 15585, "step": 19014} +{"lr": 1.4156038314229283e-05, "data_time": 0.24796745777130128, "loss": 0.008239454310387372, "time": 0.7647835493087769, "epoch": 10, "memory": 15585, "step": 19114} +{"lr": 1.3563680376778489e-05, "data_time": 0.3683037281036377, "loss": 0.010397197306156158, "time": 0.8906365633010864, "epoch": 10, "memory": 15585, "step": 19214} +{"lr": 1.3016456487714728e-05, "data_time": 0.29022114276885985, "loss": 0.008477146970108151, "time": 0.8082372903823852, "epoch": 10, "memory": 15585, "step": 19314} +{"lr": 1.2514495663462562e-05, "data_time": 0.29406323432922366, "loss": 0.010061830189079047, "time": 0.8324848175048828, "epoch": 10, "memory": 15585, "step": 19414} +{"lr": 1.205791624898479e-05, "data_time": 0.260433554649353, "loss": 0.008440747577697038, "time": 0.7797337770462036, "epoch": 10, "memory": 15585, "step": 19514} +{"lr": 1.1646825889880944e-05, "data_time": 0.3092293977737427, "loss": 0.014427883084863424, "time": 0.8371877431869507, "epoch": 10, "memory": 15585, "step": 19614} +{"lr": 1.128132150700797e-05, "data_time": 0.31436724662780763, "loss": 0.010066068498417735, "time": 0.8309544086456299, "epoch": 10, "memory": 15585, "step": 19714} +{"lr": 1.0961489273629872e-05, "data_time": 0.23543896675109863, "loss": 0.010142676578834654, "time": 0.7694511413574219, "epoch": 10, "memory": 15585, "step": 19814} +{"lr": 1.0687404595100842e-05, "data_time": 0.3482056140899658, "loss": 0.013703162875026464, "time": 0.8651807308197021, "epoch": 10, "memory": 15585, "step": 19914} +{"lr": 1.0459132091087482e-05, "data_time": 0.3057724475860596, "loss": 0.014058442646637559, "time": 0.8284258604049682, "epoch": 10, "memory": 15585, "step": 20014} +{"lr": 1.0276725580333608e-05, "data_time": 0.3203453779220581, "loss": 0.011783366790041327, "time": 0.837022066116333, "epoch": 10, "memory": 15585, "step": 20114} +{"lr": 1.0140228067971661e-05, "data_time": 0.2741419315338135, "loss": 0.009997434495016933, "time": 0.8041923761367797, "epoch": 10, "memory": 15585, "step": 20214} +{"lr": 1.0049671735383602e-05, "data_time": 0.3553286552429199, "loss": 0.011790684284642338, "time": 0.8742934226989746, "epoch": 10, "memory": 15585, "step": 20314} +{"lr": 1.000507793261375e-05, "data_time": 0.2955633640289307, "loss": 0.006616629473865032, "time": 0.8158665180206299, "epoch": 10, "memory": 15585, "step": 20414} +{"accuracy/top1": 88.90323638916016, "data_time": 0.12052183463925221, "time": 0.5190118688051818, "step": 10} diff --git a/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1/20230606_010416/vis_data/config.py b/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1/20230606_010416/vis_data/config.py new file mode 100644 index 0000000000000000000000000000000000000000..01d25c41d41ae62d6798eabf6e093ccc9c38588d --- /dev/null +++ b/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1/20230606_010416/vis_data/config.py @@ -0,0 +1,366 @@ +optim_wrapper = dict( + optimizer=dict( + type='AdamW', + lr=0.0004, + weight_decay=0.05, + eps=1e-08, + betas=(0.9, 0.999), + _scope_='mmpretrain'), + paramwise_cfg=dict( + norm_decay_mult=0.0, + bias_decay_mult=0.0, + flat_decay_mult=0.0, + custom_keys=dict({ + '.absolute_pos_embed': dict(decay_mult=0.0), + '.relative_position_bias_table': dict(decay_mult=0.0) + })), + type='AmpOptimWrapper', + dtype='bfloat16', + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=False, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=4096) +model = dict( + type='ImageClassifier', + backbone=dict( + frozen_stages=2, + type='ConvNeXt', + arch='small', + drop_path_rate=0.4, + init_cfg=dict( + type='Pretrained', + checkpoint= + 'https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth', + prefix='backbone')), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=768, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + init_cfg=None), + init_cfg=dict( + type='TruncNormal', layer=['Conv2d', 'Linear'], std=0.02, bias=0.0), + train_cfg=None) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/all_0.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/all_1.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3fake8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/cc1m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3real8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV2-1-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/if-dpmsolver++-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV2-1-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/if-dpmsolver++-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1' diff --git a/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1/20230606_010416/vis_data/events.out.tfevents.1685984670.SH-IDC1-10-140-24-133.253958.0 b/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1/20230606_010416/vis_data/events.out.tfevents.1685984670.SH-IDC1-10-140-24-133.253958.0 new file mode 100644 index 0000000000000000000000000000000000000000..553c5348ed3627ad2cf70611c13a7d0a7b3cc01a --- /dev/null +++ b/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1/20230606_010416/vis_data/events.out.tfevents.1685984670.SH-IDC1-10-140-24-133.253958.0 @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:214bafcf346b17dd48308ce8a83f1f0cb49d1879fd52472ea99d2498f8612185 +size 2838295 diff --git a/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1/20230606_010416/vis_data/scalars.json b/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1/20230606_010416/vis_data/scalars.json new file mode 100644 index 0000000000000000000000000000000000000000..05ecf8b06ced248974f9ee3965eeea68bbc27311 --- /dev/null +++ b/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1/20230606_010416/vis_data/scalars.json @@ -0,0 +1,210 @@ +{"lr": 0.00039997747031453974, "data_time": 0.27008748054504395, "loss": 0.13923058956861495, "time": 0.7958432912826539, "epoch": 1, "memory": 15587, "step": 100} +{"lr": 0.0003999089739988646, "data_time": 0.3738239765167236, "loss": 0.13745130524039267, "time": 0.9023931503295899, "epoch": 1, "memory": 15587, "step": 200} +{"lr": 0.00039979452490327365, "data_time": 0.3756752014160156, "loss": 0.09208748266100883, "time": 0.9034847021102905, "epoch": 1, "memory": 15587, "step": 300} +{"lr": 0.0003996341500108966, "data_time": 0.31546545028686523, "loss": 0.08725760839879512, "time": 0.8387650728225708, "epoch": 1, "memory": 15587, "step": 400} +{"lr": 0.00039942788713257383, "data_time": 0.3259701728820801, "loss": 0.07456294409930705, "time": 0.8584913492202759, "epoch": 1, "memory": 15587, "step": 500} +{"lr": 0.00039917578489794236, "data_time": 0.28955626487731934, "loss": 0.06376053467392921, "time": 0.8172077417373658, "epoch": 1, "memory": 15587, "step": 600} +{"lr": 0.00039887790274397046, "data_time": 0.2834655284881592, "loss": 0.07767984047532081, "time": 0.8152337312698364, "epoch": 1, "memory": 15587, "step": 700} +{"lr": 0.00039853431090094445, "data_time": 0.2876667737960815, "loss": 0.05413009580224752, "time": 0.8127007961273194, "epoch": 1, "memory": 15587, "step": 800} +{"lr": 0.0003981450903759111, "data_time": 0.23674638271331788, "loss": 0.06468360386788845, "time": 0.7818946123123169, "epoch": 1, "memory": 15587, "step": 900} +{"lr": 0.00039771033293357867, "data_time": 0.3072946548461914, "loss": 0.058453100919723514, "time": 0.8234207153320312, "epoch": 1, "memory": 15587, "step": 1000} +{"lr": 0.0003972301410746815, "data_time": 0.24485464096069337, "loss": 0.058112204819917676, "time": 0.7857584953308105, "epoch": 1, "memory": 15587, "step": 1100} +{"lr": 0.00039670462801181557, "data_time": 0.2884711265563965, "loss": 0.0498960081487894, "time": 0.8088911056518555, "epoch": 1, "memory": 15587, "step": 1200} +{"lr": 0.0003961339176427448, "data_time": 0.28011374473571776, "loss": 0.04878234844654798, "time": 0.8044605731964112, "epoch": 1, "memory": 15587, "step": 1300} +{"lr": 0.00039551814452119214, "data_time": 0.24360334873199463, "loss": 0.048634478263556956, "time": 0.7632551670074463, "epoch": 1, "memory": 15587, "step": 1400} +{"lr": 0.00039485745382511467, "data_time": 0.2661212682723999, "loss": 0.05470843706279993, "time": 0.795650315284729, "epoch": 1, "memory": 15587, "step": 1500} +{"lr": 0.0003941520013224769, "data_time": 0.2748616933822632, "loss": 0.03421585243195295, "time": 0.791138482093811, "epoch": 1, "memory": 15587, "step": 1600} +{"lr": 0.0003934019533345259, "data_time": 0.2699565887451172, "loss": 0.048215757682919504, "time": 0.7989970207214355, "epoch": 1, "memory": 15587, "step": 1700} +{"lr": 0.00039260748669657817, "data_time": 0.27178153991699217, "loss": 0.04365649037063122, "time": 0.7876599311828614, "epoch": 1, "memory": 15587, "step": 1800} +{"lr": 0.000391768788716328, "data_time": 0.2469374418258667, "loss": 0.04733328409492969, "time": 0.7656367301940918, "epoch": 1, "memory": 15587, "step": 1900} +{"lr": 0.00039088605712968733, "data_time": 0.30262730121612547, "loss": 0.045094087906181815, "time": 0.8185220241546631, "epoch": 1, "memory": 15587, "step": 2000} +{"accuracy/top1": 38.986812591552734, "data_time": 0.15160495042800903, "time": 0.5883608341217041, "step": 1} +{"lr": 0.0003895186269690071, "data_time": 0.22976510524749755, "loss": 0.045768016949296, "time": 0.7478801965713501, "epoch": 2, "memory": 15587, "step": 2146} +{"lr": 0.0003885284798052241, "data_time": 0.30257401466369627, "loss": 0.0549089414998889, "time": 0.8239081382751465, "epoch": 2, "memory": 15585, "step": 2246} +{"lr": 0.0003874950629875727, "data_time": 0.2778758525848389, "loss": 0.04565986730158329, "time": 0.7945029020309449, "epoch": 2, "memory": 15585, "step": 2346} +{"lr": 0.0003864186201599172, "data_time": 0.3167431354522705, "loss": 0.03130431789904833, "time": 0.8378938436508179, "epoch": 2, "memory": 15585, "step": 2446} +{"lr": 0.0003852994051101632, "data_time": 0.2866102457046509, "loss": 0.04411281365901232, "time": 0.8049334526062012, "epoch": 2, "memory": 15585, "step": 2546} +{"lr": 0.0003841376817104243, "data_time": 0.35471739768981936, "loss": 0.040894638188183306, "time": 0.8776494026184082, "epoch": 2, "memory": 15585, "step": 2646} +{"lr": 0.0003829337238548097, "data_time": 0.2152803659439087, "loss": 0.04159433804452419, "time": 0.731187105178833, "epoch": 2, "memory": 15585, "step": 2746} +{"lr": 0.0003816878153948477, "data_time": 0.27244584560394286, "loss": 0.05190552137792111, "time": 0.7886195421218872, "epoch": 2, "memory": 15585, "step": 2846} +{"lr": 0.0003804002500725666, "data_time": 0.2790867805480957, "loss": 0.0409588823094964, "time": 0.7950185537338257, "epoch": 2, "memory": 15585, "step": 2946} +{"lr": 0.0003790713314512378, "data_time": 0.28353979587554934, "loss": 0.03531469237059355, "time": 0.811493992805481, "epoch": 2, "memory": 15585, "step": 3046} +{"lr": 0.00037770137284380876, "data_time": 0.3215766906738281, "loss": 0.04253164660185575, "time": 0.8388626098632812, "epoch": 2, "memory": 15585, "step": 3146} +{"lr": 0.00037629069723903126, "data_time": 0.3698327302932739, "loss": 0.03412234988063574, "time": 0.8844455480575562, "epoch": 2, "memory": 15585, "step": 3246} +{"lr": 0.0003748396372253147, "data_time": 0.27383334636688234, "loss": 0.03273127228021622, "time": 0.7871771574020385, "epoch": 2, "memory": 15585, "step": 3346} +{"lr": 0.0003733485349123126, "data_time": 0.26422815322875975, "loss": 0.03890739511698484, "time": 0.7798113107681275, "epoch": 2, "memory": 15585, "step": 3446} +{"lr": 0.0003718177418502636, "data_time": 0.25523183345794676, "loss": 0.04283151123672724, "time": 0.773280668258667, "epoch": 2, "memory": 15585, "step": 3546} +{"lr": 0.0003702476189471092, "data_time": 0.26333303451538087, "loss": 0.03535877075046301, "time": 0.7787991285324096, "epoch": 2, "memory": 15585, "step": 3646} +{"lr": 0.00036863853638340543, "data_time": 0.26228883266448977, "loss": 0.03946240935474634, "time": 0.7818804502487182, "epoch": 2, "memory": 15585, "step": 3746} +{"lr": 0.000366990873525045, "data_time": 0.2537489414215088, "loss": 0.027454120200127365, "time": 0.7705466032028199, "epoch": 2, "memory": 15585, "step": 3846} +{"lr": 0.00036530501883381603, "data_time": 0.3259198427200317, "loss": 0.03414789345115423, "time": 0.8427048683166504, "epoch": 2, "memory": 15585, "step": 3946} +{"lr": 0.00036358136977581845, "data_time": 0.22690415382385254, "loss": 0.042348215635865924, "time": 0.740772008895874, "epoch": 2, "memory": 15585, "step": 4046} +{"accuracy/top1": 72.27049255371094, "data_time": 0.11775830143787822, "time": 0.514974793449777, "step": 2} +{"lr": 0.0003609978147988783, "data_time": 0.2866313695907593, "loss": 0.03475107345730066, "time": 0.8039616346359253, "epoch": 3, "memory": 15585, "step": 4192} +{"lr": 0.00035918293929014, "data_time": 0.3887096643447876, "loss": 0.03676401153206825, "time": 0.9055249214172363, "epoch": 3, "memory": 15585, "step": 4292} +{"lr": 0.0003573317127887786, "data_time": 0.2925009489059448, "loss": 0.027336132153868674, "time": 0.8093053340911865, "epoch": 3, "memory": 15585, "step": 4392} +{"lr": 0.000355444571749834, "data_time": 0.22066619396209716, "loss": 0.032349448464810845, "time": 0.7472469568252563, "epoch": 3, "memory": 15585, "step": 4492} +{"lr": 0.0003535219610957493, "data_time": 0.2764780044555664, "loss": 0.030434053856879473, "time": 0.7949762105941772, "epoch": 3, "memory": 15585, "step": 4592} +{"lr": 0.00035156433411147255, "data_time": 0.27833940982818606, "loss": 0.035816552862524983, "time": 0.7996628046035766, "epoch": 3, "memory": 15585, "step": 4692} +{"lr": 0.00034957215233758935, "data_time": 0.28330082893371583, "loss": 0.03549886345863342, "time": 0.8007184743881226, "epoch": 3, "memory": 15585, "step": 4792} +{"lr": 0.00034754588546150636, "data_time": 0.3010856628417969, "loss": 0.03336919061839581, "time": 0.818373703956604, "epoch": 3, "memory": 15585, "step": 4892} +{"lr": 0.0003454860112067171, "data_time": 0.32679333686828616, "loss": 0.03792732730507851, "time": 0.8422648191452027, "epoch": 3, "memory": 15585, "step": 4992} +{"lr": 0.00034339301522016903, "data_time": 0.3118792772293091, "loss": 0.034485948644578455, "time": 0.8398421287536622, "epoch": 3, "memory": 15585, "step": 5092} +{"lr": 0.0003412673909577655, "data_time": 0.32096998691558837, "loss": 0.029889604449272154, "time": 0.8347985506057739, "epoch": 3, "memory": 15585, "step": 5192} +{"lr": 0.00033910963956802737, "data_time": 0.2795300245285034, "loss": 0.029570287559181452, "time": 0.7951908349990845, "epoch": 3, "memory": 15585, "step": 5292} +{"lr": 0.00033692026977393767, "data_time": 0.3264751434326172, "loss": 0.030787329375743865, "time": 0.8414042234420777, "epoch": 3, "memory": 15585, "step": 5392} +{"lr": 0.0003346997977530036, "data_time": 0.3256623029708862, "loss": 0.037439318187534806, "time": 0.8437143087387085, "epoch": 3, "memory": 15585, "step": 5492} +{"lr": 0.0003324487470155584, "data_time": 0.27264065742492677, "loss": 0.03120652474462986, "time": 0.7892414093017578, "epoch": 3, "memory": 15585, "step": 5592} +{"lr": 0.0003301676482813382, "data_time": 0.259604811668396, "loss": 0.023959132796153425, "time": 0.7879086017608643, "epoch": 3, "memory": 15585, "step": 5692} +{"lr": 0.00032785703935435476, "data_time": 0.31877496242523196, "loss": 0.0331362733617425, "time": 0.8338300228118897, "epoch": 3, "memory": 15585, "step": 5792} +{"lr": 0.00032551746499609967, "data_time": 0.23977253437042237, "loss": 0.029436669871211053, "time": 0.7656080007553101, "epoch": 3, "memory": 15585, "step": 5892} +{"lr": 0.0003231494767971101, "data_time": 0.29613943099975587, "loss": 0.025362644670531155, "time": 0.8107655048370361, "epoch": 3, "memory": 15585, "step": 5992} +{"lr": 0.0003207536330469211, "data_time": 0.24107058048248292, "loss": 0.024404334463179113, "time": 0.7703856229782104, "epoch": 3, "memory": 15585, "step": 6092} +{"accuracy/top1": 83.16535186767578, "data_time": 0.11824517953591268, "time": 0.5168199851864674, "step": 3} +{"lr": 0.000317206849615549, "data_time": 0.24976627826690673, "loss": 0.0327368400990963, "time": 0.7642760038375854, "epoch": 4, "memory": 15585, "step": 6238} +{"lr": 0.00031474489842163173, "data_time": 0.2061671018600464, "loss": 0.029396084044128655, "time": 0.7230918169021606, "epoch": 4, "memory": 15585, "step": 6338} +{"lr": 0.0003122570731847668, "data_time": 0.2639167785644531, "loss": 0.03776932386681438, "time": 0.7808402299880981, "epoch": 4, "memory": 15585, "step": 6438} +{"lr": 0.00030974396044791, "data_time": 0.26590380668640134, "loss": 0.0281455896794796, "time": 0.7859663248062134, "epoch": 4, "memory": 15585, "step": 6538} +{"lr": 0.0003072061527159319, "data_time": 0.27479000091552735, "loss": 0.02331636203452945, "time": 0.7929321765899658, "epoch": 4, "memory": 15585, "step": 6638} +{"lr": 0.0003046442483159274, "data_time": 0.3018221855163574, "loss": 0.03137182611972093, "time": 0.8242158174514771, "epoch": 4, "memory": 15585, "step": 6738} +{"lr": 0.0003020588512561505, "data_time": 0.2968027353286743, "loss": 0.02740856409072876, "time": 0.8264802694320679, "epoch": 4, "memory": 15585, "step": 6838} +{"lr": 0.0002994505710836101, "data_time": 0.37222654819488527, "loss": 0.02564994916319847, "time": 0.8950869083404541, "epoch": 4, "memory": 15585, "step": 6938} +{"lr": 0.0002968200227403596, "data_time": 0.2791316509246826, "loss": 0.028884059656411408, "time": 0.794876503944397, "epoch": 4, "memory": 15585, "step": 7038} +{"lr": 0.0002941678264185154, "data_time": 0.24613494873046876, "loss": 0.03310949513688684, "time": 0.7665436506271363, "epoch": 4, "memory": 15585, "step": 7138} +{"lr": 0.0002914946074140367, "data_time": 0.2930370569229126, "loss": 0.02582396026700735, "time": 0.8193508625030518, "epoch": 4, "memory": 15585, "step": 7238} +{"lr": 0.00028880099597930275, "data_time": 0.2735331058502197, "loss": 0.025366627145558596, "time": 0.7899934768676757, "epoch": 4, "memory": 15585, "step": 7338} +{"lr": 0.0002860876271745208, "data_time": 0.27272055149078367, "loss": 0.02609073668718338, "time": 0.8024159669876099, "epoch": 4, "memory": 15585, "step": 7438} +{"lr": 0.0002833551407180016, "data_time": 0.26704916954040525, "loss": 0.02427520966157317, "time": 0.7911346435546875, "epoch": 4, "memory": 15585, "step": 7538} +{"lr": 0.00028060418083533477, "data_time": 0.30552144050598146, "loss": 0.04366566501557827, "time": 0.8224530458450318, "epoch": 4, "memory": 15585, "step": 7638} +{"lr": 0.0002778353961075036, "data_time": 0.2729483604431152, "loss": 0.019242750527337194, "time": 0.7928371906280518, "epoch": 4, "memory": 15585, "step": 7738} +{"lr": 0.0002750494393179725, "data_time": 0.29315340518951416, "loss": 0.019132024236023427, "time": 0.8097799777984619, "epoch": 4, "memory": 15585, "step": 7838} +{"lr": 0.00027224696729878206, "data_time": 0.3284644603729248, "loss": 0.026389026921242474, "time": 0.8470310211181641, "epoch": 4, "memory": 15585, "step": 7938} +{"lr": 0.00026942864077569254, "data_time": 0.19699296951293946, "loss": 0.024535698350518942, "time": 0.7180789947509766, "epoch": 4, "memory": 15585, "step": 8038} +{"lr": 0.0002665951242124057, "data_time": 0.2692307949066162, "loss": 0.03399046678096056, "time": 0.7924437999725342, "epoch": 4, "memory": 15585, "step": 8138} +{"accuracy/top1": 86.0947036743164, "data_time": 0.11637008385580094, "time": 0.513966329762193, "step": 4} +{"lr": 0.0002624322962007581, "data_time": 0.22305374145507811, "loss": 0.02709553427994251, "time": 0.7494282007217408, "epoch": 5, "memory": 15585, "step": 8284} +{"lr": 0.0002595642622579449, "data_time": 0.2448746919631958, "loss": 0.023471019277349115, "time": 0.784416675567627, "epoch": 5, "memory": 15585, "step": 8384} +{"lr": 0.0002566833639534589, "data_time": 0.2801742792129517, "loss": 0.03382280953228474, "time": 0.7957663536071777, "epoch": 5, "memory": 15585, "step": 8484} +{"lr": 0.0002537902805032591, "data_time": 0.30092740058898926, "loss": 0.025358874769881367, "time": 0.832024359703064, "epoch": 5, "memory": 15585, "step": 8584} +{"lr": 0.00025088569399614127, "data_time": 0.32829761505126953, "loss": 0.0313843123614788, "time": 0.8450688362121582, "epoch": 5, "memory": 15585, "step": 8684} +{"lr": 0.0002479702892329199, "data_time": 0.272039008140564, "loss": 0.02091716518625617, "time": 0.8140338659286499, "epoch": 5, "memory": 15585, "step": 8784} +{"lr": 0.00024504475356498226, "data_time": 0.27944681644439695, "loss": 0.02268802421167493, "time": 0.7953030347824097, "epoch": 5, "memory": 15585, "step": 8884} +{"lr": 0.0002421097767322291, "data_time": 0.2924706697463989, "loss": 0.028954963386058807, "time": 0.8187745571136474, "epoch": 5, "memory": 15585, "step": 8984} +{"lr": 0.00023916605070046265, "data_time": 0.2997701644897461, "loss": 0.02127212667837739, "time": 0.8167422533035278, "epoch": 5, "memory": 15585, "step": 9084} +{"lr": 0.00023621426949824308, "data_time": 0.5579146146774292, "loss": 0.021598626906052232, "time": 1.1002458810806275, "epoch": 5, "memory": 15585, "step": 9184} +{"lr": 0.0002332551290532596, "data_time": 0.26402647495269777, "loss": 0.02053514327853918, "time": 0.7831912994384765, "epoch": 5, "memory": 15585, "step": 9284} +{"lr": 0.00023028932702825637, "data_time": 0.2759858131408691, "loss": 0.027401977218687534, "time": 0.8034228086471558, "epoch": 5, "memory": 15585, "step": 9384} +{"lr": 0.0002273175626565469, "data_time": 0.2330115795135498, "loss": 0.025782711850479246, "time": 0.748007345199585, "epoch": 5, "memory": 15585, "step": 9484} +{"lr": 0.00022434053657715923, "data_time": 0.29543793201446533, "loss": 0.02157797208055854, "time": 0.8174185037612915, "epoch": 5, "memory": 15585, "step": 9584} +{"lr": 0.00022135895066965, "data_time": 0.30692052841186523, "loss": 0.025195316784083842, "time": 0.823170280456543, "epoch": 5, "memory": 15585, "step": 9684} +{"lr": 0.00021837350788862523, "data_time": 0.2897050380706787, "loss": 0.02679475788027048, "time": 0.8073092460632324, "epoch": 5, "memory": 15585, "step": 9784} +{"lr": 0.00021538491209800788, "data_time": 0.7872004508972168, "loss": 0.02362949438393116, "time": 1.3031886339187622, "epoch": 5, "memory": 15585, "step": 9884} +{"lr": 0.0002123938679050912, "data_time": 0.3025233030319214, "loss": 0.022985117742791772, "time": 0.8174977540969849, "epoch": 5, "memory": 15585, "step": 9984} +{"lr": 0.0002094010804944171, "data_time": 0.28919928073883056, "loss": 0.017819868260994555, "time": 0.8158895969390869, "epoch": 5, "memory": 15585, "step": 10084} +{"lr": 0.00020640725546151753, "data_time": 0.22807033061981202, "loss": 0.021346819400787354, "time": 0.7443963766098023, "epoch": 5, "memory": 15585, "step": 10184} +{"accuracy/top1": 87.50641632080078, "data_time": 0.1380059250065538, "time": 0.5391156947026488, "step": 5} +{"lr": 0.00020203586947950216, "data_time": 0.2559981107711792, "loss": 0.01903561851941049, "time": 0.7719154834747315, "epoch": 6, "memory": 15585, "step": 10330} +{"lr": 0.0001990424959328828, "data_time": 0.2547386884689331, "loss": 0.027998182456940413, "time": 0.7724256753921509, "epoch": 6, "memory": 15585, "step": 10430} +{"lr": 0.00019605052695921703, "data_time": 0.2711981773376465, "loss": 0.023488518362864853, "time": 0.8005754947662354, "epoch": 6, "memory": 15585, "step": 10530} +{"lr": 0.00019306066796107762, "data_time": 0.27912838459014894, "loss": 0.02859131288714707, "time": 0.7995332241058349, "epoch": 6, "memory": 15585, "step": 10630} +{"lr": 0.00019007362384357827, "data_time": 0.2587214469909668, "loss": 0.022684884164482354, "time": 0.7788129329681397, "epoch": 6, "memory": 15585, "step": 10730} +{"lr": 0.00018709009884818184, "data_time": 0.2396928071975708, "loss": 0.021371140610426664, "time": 0.7657092571258545, "epoch": 6, "memory": 15585, "step": 10830} +{"lr": 0.00018411079638666402, "data_time": 0.3359281778335571, "loss": 0.020875070244073868, "time": 0.8567445278167725, "epoch": 6, "memory": 15585, "step": 10930} +{"lr": 0.00018113641887527326, "data_time": 0.2874326467514038, "loss": 0.01259036036208272, "time": 0.8102641582489014, "epoch": 6, "memory": 15585, "step": 11030} +{"lr": 0.00017816766756912557, "data_time": 0.3653801679611206, "loss": 0.01827344549819827, "time": 0.8827403068542481, "epoch": 6, "memory": 15585, "step": 11130} +{"lr": 0.0001752052423968725, "data_time": 0.3379427194595337, "loss": 0.014073522575199605, "time": 0.8592472076416016, "epoch": 6, "memory": 15585, "step": 11230} +{"lr": 0.00017224984179568312, "data_time": 0.2708329439163208, "loss": 0.01823663548566401, "time": 0.7913220643997192, "epoch": 6, "memory": 15585, "step": 11330} +{"lr": 0.00016930216254657575, "data_time": 0.3422927141189575, "loss": 0.022221644502133132, "time": 0.867781400680542, "epoch": 6, "memory": 15585, "step": 11430} +{"lr": 0.0001663628996101415, "data_time": 0.2553313970565796, "loss": 0.01576085495762527, "time": 0.7772531509399414, "epoch": 6, "memory": 15585, "step": 11530} +{"lr": 0.0001634327459626967, "data_time": 0.28970420360565186, "loss": 0.019388765934854744, "time": 0.8159053325653076, "epoch": 6, "memory": 15585, "step": 11630} +{"lr": 0.0001605123924329032, "data_time": 0.3055410861968994, "loss": 0.01775304921902716, "time": 0.8254775524139404, "epoch": 6, "memory": 15585, "step": 11730} +{"lr": 0.0001576025275388945, "data_time": 0.3613670825958252, "loss": 0.021886652009561657, "time": 0.8811241388320923, "epoch": 6, "memory": 15585, "step": 11830} +{"lr": 0.00015470383732594812, "data_time": 0.2286855936050415, "loss": 0.018671649461612105, "time": 0.7465840816497803, "epoch": 6, "memory": 15585, "step": 11930} +{"lr": 0.00015181700520473825, "data_time": 0.29651646614074706, "loss": 0.019918525079265236, "time": 0.8182250261306763, "epoch": 6, "memory": 15585, "step": 12030} +{"lr": 0.00014894271179021254, "data_time": 0.287540340423584, "loss": 0.013365720445290209, "time": 0.8053610801696778, "epoch": 6, "memory": 15585, "step": 12130} +{"lr": 0.0001460816347411264, "data_time": 0.35853650569915774, "loss": 0.015575483674183488, "time": 0.8735523700714112, "epoch": 6, "memory": 15585, "step": 12230} +{"accuracy/top1": 87.8821029663086, "data_time": 0.1156603000203117, "time": 0.5134156062954762, "step": 6} +{"lr": 0.00014192959250590055, "data_time": 0.2618546962738037, "loss": 0.021499492274597288, "time": 0.7784563779830933, "epoch": 7, "memory": 15585, "step": 12376} +{"lr": 0.00013910389161427854, "data_time": 0.24254865646362306, "loss": 0.01944695943966508, "time": 0.7597046613693237, "epoch": 7, "memory": 15585, "step": 12476} +{"lr": 0.00013629372674084914, "data_time": 0.2487579107284546, "loss": 0.013473668042570353, "time": 0.7638455629348755, "epoch": 7, "memory": 15585, "step": 12576} +{"lr": 0.00013349976042508036, "data_time": 0.26796531677246094, "loss": 0.02121005724184215, "time": 0.7842514276504516, "epoch": 7, "memory": 15585, "step": 12676} +{"lr": 0.0001307226513873823, "data_time": 0.37471370697021483, "loss": 0.01720991963520646, "time": 0.8909207582473755, "epoch": 7, "memory": 15585, "step": 12776} +{"lr": 0.00012796305437380283, "data_time": 0.3059714317321777, "loss": 0.014831477496773004, "time": 0.8221205234527588, "epoch": 7, "memory": 15585, "step": 12876} +{"lr": 0.00012522162000166114, "data_time": 0.2801429510116577, "loss": 0.014092972502112389, "time": 0.8082698106765747, "epoch": 7, "memory": 15585, "step": 12976} +{"lr": 0.00012249899460615546, "data_time": 0.27553768157958985, "loss": 0.01806235294789076, "time": 0.7926186561584473, "epoch": 7, "memory": 15585, "step": 13076} +{"lr": 0.00011979582008797885, "data_time": 0.30564844608306885, "loss": 0.016327786445617675, "time": 0.841417932510376, "epoch": 7, "memory": 15585, "step": 13176} +{"lr": 0.00011711273376198264, "data_time": 0.2962301015853882, "loss": 0.01663127182982862, "time": 0.8150619506835938, "epoch": 7, "memory": 15585, "step": 13276} +{"lr": 0.00011445036820691838, "data_time": 0.25047237873077394, "loss": 0.018808535439893602, "time": 0.7785188913345337, "epoch": 7, "memory": 15585, "step": 13376} +{"lr": 0.00011180935111629823, "data_time": 0.2908912181854248, "loss": 0.012106695957481862, "time": 0.8087669134140014, "epoch": 7, "memory": 15585, "step": 13476} +{"lr": 0.00010919030515040621, "data_time": 0.4291386604309082, "loss": 0.018903985805809497, "time": 0.9457412004470825, "epoch": 7, "memory": 15585, "step": 13576} +{"lr": 0.00010659384778949681, "data_time": 0.2921184778213501, "loss": 0.013892869511619211, "time": 0.8050588846206665, "epoch": 7, "memory": 15585, "step": 13676} +{"lr": 0.00010402059118821447, "data_time": 0.2596737384796143, "loss": 0.009289868036285042, "time": 0.7975058555603027, "epoch": 7, "memory": 15585, "step": 13776} +{"lr": 0.00010147114203126853, "data_time": 0.2982635498046875, "loss": 0.011215496947988868, "time": 0.814716386795044, "epoch": 7, "memory": 15585, "step": 13876} +{"lr": 9.89461013903984e-05, "data_time": 0.25265505313873293, "loss": 0.021423447458073498, "time": 0.7830265045166016, "epoch": 7, "memory": 15585, "step": 13976} +{"lr": 9.644606458266133e-05, "data_time": 0.2507262468338013, "loss": 0.01776902177371085, "time": 0.7632845163345336, "epoch": 7, "memory": 15585, "step": 14076} +{"lr": 9.397162103007764e-05, "data_time": 0.2760718107223511, "loss": 0.02215931466780603, "time": 0.7969234704971313, "epoch": 7, "memory": 15585, "step": 14176} +{"lr": 9.152335412066514e-05, "data_time": 0.2921916484832764, "loss": 0.017632154375314714, "time": 0.8106762170791626, "epoch": 7, "memory": 15585, "step": 14276} +{"accuracy/top1": 88.1783447265625, "data_time": 0.12549807204574834, "time": 0.5257302933051938, "step": 7} +{"lr": 8.79970864551918e-05, "data_time": 0.3116596221923828, "loss": 0.011558062583208083, "time": 0.8290964126586914, "epoch": 8, "memory": 15585, "step": 14422} +{"lr": 8.56156575096546e-05, "data_time": 0.23561570644378663, "loss": 0.014075045567005872, "time": 0.7553547620773315, "epoch": 8, "memory": 15585, "step": 14522} +{"lr": 8.326237525384291e-05, "data_time": 0.3643265962600708, "loss": 0.012113719852641225, "time": 0.8816712856292724, "epoch": 8, "memory": 15585, "step": 14622} +{"lr": 8.093779451014054e-05, "data_time": 0.32892110347747805, "loss": 0.008901950716972352, "time": 0.8484573841094971, "epoch": 8, "memory": 15585, "step": 14722} +{"lr": 7.86424633341094e-05, "data_time": 0.3327207326889038, "loss": 0.008992444211617111, "time": 0.8496640682220459, "epoch": 8, "memory": 15585, "step": 14822} +{"lr": 7.637692288527767e-05, "data_time": 0.27522692680358884, "loss": 0.011448814813047648, "time": 0.7937656164169311, "epoch": 8, "memory": 15585, "step": 14922} +{"lr": 7.414170729955236e-05, "data_time": 0.32682440280914304, "loss": 0.015614293608814478, "time": 0.8439353704452515, "epoch": 8, "memory": 15585, "step": 15022} +{"lr": 7.193734356328913e-05, "data_time": 0.33294880390167236, "loss": 0.013695200346410275, "time": 0.8572995901107788, "epoch": 8, "memory": 15585, "step": 15122} +{"lr": 6.976435138904706e-05, "data_time": 0.3373740196228027, "loss": 0.014129743725061417, "time": 0.8557215929031372, "epoch": 8, "memory": 15585, "step": 15222} +{"lr": 6.762324309305816e-05, "data_time": 0.26450512409210203, "loss": 0.01022324119694531, "time": 0.7916579484939575, "epoch": 8, "memory": 15585, "step": 15322} +{"lr": 6.551452347444078e-05, "data_time": 0.2653649091720581, "loss": 0.007250535348430276, "time": 0.7819453477859497, "epoch": 8, "memory": 15585, "step": 15422} +{"lr": 6.34386896961863e-05, "data_time": 0.2581397771835327, "loss": 0.014060096349567175, "time": 0.776689100265503, "epoch": 8, "memory": 15585, "step": 15522} +{"lr": 6.139623116794487e-05, "data_time": 0.244904088973999, "loss": 0.016063893772661687, "time": 0.7623092412948609, "epoch": 8, "memory": 15585, "step": 15622} +{"lr": 5.938762943063905e-05, "data_time": 0.28007290363311765, "loss": 0.015862515289336444, "time": 0.7965219974517822, "epoch": 8, "memory": 15585, "step": 15722} +{"lr": 5.741335804293427e-05, "data_time": 0.2992445707321167, "loss": 0.012573866778984666, "time": 0.8168627262115479, "epoch": 8, "memory": 15585, "step": 15822} +{"lr": 5.5473882469589504e-05, "data_time": 0.24587674140930177, "loss": 0.011037486558780073, "time": 0.7706165790557862, "epoch": 8, "memory": 15585, "step": 15922} +{"lr": 5.356965997171673e-05, "data_time": 0.2832584619522095, "loss": 0.011991947283968329, "time": 0.8011831760406494, "epoch": 8, "memory": 15585, "step": 16022} +{"lr": 5.170113949897451e-05, "data_time": 0.29445812702178953, "loss": 0.019166898727416993, "time": 0.8190693140029908, "epoch": 8, "memory": 15585, "step": 16122} +{"lr": 4.986876158372164e-05, "data_time": 0.2771552562713623, "loss": 0.010040620574727654, "time": 0.7935341358184814, "epoch": 8, "memory": 15585, "step": 16222} +{"lr": 4.8072958237154565e-05, "data_time": 0.26821324825286863, "loss": 0.012176671624183654, "time": 0.7931279897689819, "epoch": 8, "memory": 15585, "step": 16322} +{"accuracy/top1": 88.50272369384766, "data_time": 0.10883679546293665, "time": 0.5064185330125152, "step": 8} +{"lr": 4.551764078948187e-05, "data_time": 0.22350099086761474, "loss": 0.011239055823534728, "time": 0.7417830467224121, "epoch": 9, "memory": 15585, "step": 16468} +{"lr": 4.38135946476099e-05, "data_time": 0.2658128499984741, "loss": 0.011058948514983059, "time": 0.7792037487030029, "epoch": 9, "memory": 15585, "step": 16568} +{"lr": 4.214755067289632e-05, "data_time": 0.25759875774383545, "loss": 0.011372651671990752, "time": 0.7732922792434692, "epoch": 9, "memory": 15585, "step": 16668} +{"lr": 4.051990166076003e-05, "data_time": 0.22912235260009767, "loss": 0.007898762496188283, "time": 0.7547113418579101, "epoch": 9, "memory": 15585, "step": 16768} +{"lr": 3.893103135441892e-05, "data_time": 0.22703702449798585, "loss": 0.018630159180611373, "time": 0.7559427976608276, "epoch": 9, "memory": 15585, "step": 16868} +{"lr": 3.7381314354415884e-05, "data_time": 0.2782020330429077, "loss": 0.010206338297575713, "time": 0.794103455543518, "epoch": 9, "memory": 15585, "step": 16968} +{"lr": 3.587111603030147e-05, "data_time": 0.3134643077850342, "loss": 0.008711456693708897, "time": 0.8424233436584473, "epoch": 9, "memory": 15585, "step": 17068} +{"lr": 3.440079243449231e-05, "data_time": 0.2963653802871704, "loss": 0.012440914800390602, "time": 0.8215030908584595, "epoch": 9, "memory": 15585, "step": 17168} +{"lr": 3.297069021832627e-05, "data_time": 0.2990638971328735, "loss": 0.01341496268287301, "time": 0.8373693466186524, "epoch": 9, "memory": 15585, "step": 17268} +{"lr": 3.158114655033408e-05, "data_time": 0.23194804191589355, "loss": 0.011563332751393318, "time": 0.7520285844802856, "epoch": 9, "memory": 15585, "step": 17368} +{"lr": 3.023248903674686e-05, "data_time": 0.30950255393981935, "loss": 0.01345430500805378, "time": 0.840735650062561, "epoch": 9, "memory": 15585, "step": 17468} +{"lr": 2.8925035644257496e-05, "data_time": 0.2786473989486694, "loss": 0.017220094613730908, "time": 0.7949512004852295, "epoch": 9, "memory": 15585, "step": 17568} +{"lr": 2.76590946250556e-05, "data_time": 0.28743205070495603, "loss": 0.011918843863531947, "time": 0.8146660327911377, "epoch": 9, "memory": 15585, "step": 17668} +{"lr": 2.6434964444152048e-05, "data_time": 0.2351306676864624, "loss": 0.014188050897791982, "time": 0.7527170896530151, "epoch": 9, "memory": 15585, "step": 17768} +{"lr": 2.525293370901126e-05, "data_time": 0.2827725887298584, "loss": 0.0147943832911551, "time": 0.8142024755477906, "epoch": 9, "memory": 15585, "step": 17868} +{"lr": 2.411328110150761e-05, "data_time": 0.27406988143920896, "loss": 0.006789855239912867, "time": 0.7911711454391479, "epoch": 9, "memory": 15585, "step": 17968} +{"lr": 2.3016275312222135e-05, "data_time": 0.27676851749420167, "loss": 0.010608707321807742, "time": 0.8067718267440795, "epoch": 9, "memory": 15585, "step": 18068} +{"lr": 2.1962174977094375e-05, "data_time": 0.3263155221939087, "loss": 0.01089247171767056, "time": 0.8447614908218384, "epoch": 9, "memory": 15585, "step": 18168} +{"lr": 2.095122861644491e-05, "data_time": 0.29137680530548093, "loss": 0.01079562189988792, "time": 0.8140552759170532, "epoch": 9, "memory": 15585, "step": 18268} +{"lr": 1.9983674576383036e-05, "data_time": 0.2340712547302246, "loss": 0.009820595383644104, "time": 0.7525712013244629, "epoch": 9, "memory": 15585, "step": 18368} +{"accuracy/top1": 88.91150665283203, "data_time": 0.14181355570183427, "time": 0.5377230839651139, "step": 9} +{"lr": 1.8649439622232705e-05, "data_time": 0.24411065578460694, "loss": 0.015803095977753402, "time": 0.7865235805511475, "epoch": 10, "memory": 15585, "step": 18514} +{"lr": 1.7789580193179967e-05, "data_time": 0.2782934904098511, "loss": 0.005647242115810513, "time": 0.7988179445266723, "epoch": 10, "memory": 15585, "step": 18614} +{"lr": 1.697385849182042e-05, "data_time": 0.3038593292236328, "loss": 0.013444292452186346, "time": 0.8295249938964844, "epoch": 10, "memory": 15585, "step": 18714} +{"lr": 1.620246683705586e-05, "data_time": 0.265687084197998, "loss": 0.007501485291868448, "time": 0.7844733238220215, "epoch": 10, "memory": 15585, "step": 18814} +{"lr": 1.547558709629951e-05, "data_time": 0.26913881301879883, "loss": 0.01077024326659739, "time": 0.804829478263855, "epoch": 10, "memory": 15585, "step": 18914} +{"lr": 1.479339064259843e-05, "data_time": 0.35813217163085936, "loss": 0.01208462598733604, "time": 0.8739128828048706, "epoch": 10, "memory": 15585, "step": 19014} +{"lr": 1.4156038314229283e-05, "data_time": 0.24796745777130128, "loss": 0.008239454310387372, "time": 0.7647835493087769, "epoch": 10, "memory": 15585, "step": 19114} +{"lr": 1.3563680376778489e-05, "data_time": 0.3683037281036377, "loss": 0.010397197306156158, "time": 0.8906365633010864, "epoch": 10, "memory": 15585, "step": 19214} +{"lr": 1.3016456487714728e-05, "data_time": 0.29022114276885985, "loss": 0.008477146970108151, "time": 0.8082372903823852, "epoch": 10, "memory": 15585, "step": 19314} +{"lr": 1.2514495663462562e-05, "data_time": 0.29406323432922366, "loss": 0.010061830189079047, "time": 0.8324848175048828, "epoch": 10, "memory": 15585, "step": 19414} +{"lr": 1.205791624898479e-05, "data_time": 0.260433554649353, "loss": 0.008440747577697038, "time": 0.7797337770462036, "epoch": 10, "memory": 15585, "step": 19514} +{"lr": 1.1646825889880944e-05, "data_time": 0.3092293977737427, "loss": 0.014427883084863424, "time": 0.8371877431869507, "epoch": 10, "memory": 15585, "step": 19614} +{"lr": 1.128132150700797e-05, "data_time": 0.31436724662780763, "loss": 0.010066068498417735, "time": 0.8309544086456299, "epoch": 10, "memory": 15585, "step": 19714} +{"lr": 1.0961489273629872e-05, "data_time": 0.23543896675109863, "loss": 0.010142676578834654, "time": 0.7694511413574219, "epoch": 10, "memory": 15585, "step": 19814} +{"lr": 1.0687404595100842e-05, "data_time": 0.3482056140899658, "loss": 0.013703162875026464, "time": 0.8651807308197021, "epoch": 10, "memory": 15585, "step": 19914} +{"lr": 1.0459132091087482e-05, "data_time": 0.3057724475860596, "loss": 0.014058442646637559, "time": 0.8284258604049682, "epoch": 10, "memory": 15585, "step": 20014} +{"lr": 1.0276725580333608e-05, "data_time": 0.3203453779220581, "loss": 0.011783366790041327, "time": 0.837022066116333, "epoch": 10, "memory": 15585, "step": 20114} +{"lr": 1.0140228067971661e-05, "data_time": 0.2741419315338135, "loss": 0.009997434495016933, "time": 0.8041923761367797, "epoch": 10, "memory": 15585, "step": 20214} +{"lr": 1.0049671735383602e-05, "data_time": 0.3553286552429199, "loss": 0.011790684284642338, "time": 0.8742934226989746, "epoch": 10, "memory": 15585, "step": 20314} +{"lr": 1.000507793261375e-05, "data_time": 0.2955633640289307, "loss": 0.006616629473865032, "time": 0.8158665180206299, "epoch": 10, "memory": 15585, "step": 20414} +{"accuracy/top1": 88.90323638916016, "data_time": 0.12052183463925221, "time": 0.5190118688051818, "step": 10} diff --git a/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1/20230606_010416/vis_data/vis_image/0000000.png_1.png b/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1/20230606_010416/vis_data/vis_image/0000000.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..26864ca6f4c5de0ff1fea02e8d46345c9cf70e0f Binary files /dev/null and b/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1/20230606_010416/vis_data/vis_image/0000000.png_1.png differ diff --git a/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1/20230606_010416/vis_data/vis_image/0000000.png_10.png b/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1/20230606_010416/vis_data/vis_image/0000000.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..b1177c9d139bba5d33d441829720dcb1cae14ee1 Binary files /dev/null and b/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1/20230606_010416/vis_data/vis_image/0000000.png_10.png differ diff --git a/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1/20230606_010416/vis_data/vis_image/0000000.png_2.png b/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1/20230606_010416/vis_data/vis_image/0000000.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..54881f37bf8553e9148d90067f34efcad5548bb8 Binary files /dev/null and b/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1/20230606_010416/vis_data/vis_image/0000000.png_2.png differ diff --git a/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1/20230606_010416/vis_data/vis_image/0000000.png_3.png b/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1/20230606_010416/vis_data/vis_image/0000000.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..b000ad7a7ef9d2e140c9d73135970fc9325a58af Binary files /dev/null and b/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1/20230606_010416/vis_data/vis_image/0000000.png_3.png differ diff --git a/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1/20230606_010416/vis_data/vis_image/0000000.png_4.png b/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1/20230606_010416/vis_data/vis_image/0000000.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..b73aab473afab68c211c87f1ab17e7b1bcd22703 Binary files /dev/null and b/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1/20230606_010416/vis_data/vis_image/0000000.png_4.png differ diff --git a/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1/20230606_010416/vis_data/vis_image/0000000.png_5.png b/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1/20230606_010416/vis_data/vis_image/0000000.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..4bf9415f5eb346d621e5156c61092fa01cf173b3 Binary files /dev/null and b/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1/20230606_010416/vis_data/vis_image/0000000.png_5.png differ diff --git a/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1/20230606_010416/vis_data/vis_image/0000000.png_6.png b/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1/20230606_010416/vis_data/vis_image/0000000.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..4919a4113f197823bcb3de587a6f84691b5e44f3 Binary files /dev/null and b/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1/20230606_010416/vis_data/vis_image/0000000.png_6.png differ diff --git a/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1/20230606_010416/vis_data/vis_image/0000000.png_7.png b/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1/20230606_010416/vis_data/vis_image/0000000.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..69a0d61f68d99627824fad2765d640bddf66133c Binary files /dev/null and b/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1/20230606_010416/vis_data/vis_image/0000000.png_7.png differ diff --git a/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1/20230606_010416/vis_data/vis_image/0000000.png_8.png b/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1/20230606_010416/vis_data/vis_image/0000000.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..90978c09f24e66204ffeb058fae9e996c29d33ac Binary files /dev/null and b/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1/20230606_010416/vis_data/vis_image/0000000.png_8.png differ diff --git a/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1/20230606_010416/vis_data/vis_image/0000000.png_9.png b/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1/20230606_010416/vis_data/vis_image/0000000.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..2a059e199756020db6545ec497733d9641d47a7d Binary files /dev/null and b/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1/20230606_010416/vis_data/vis_image/0000000.png_9.png differ diff --git a/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1/20230606_010416/vis_data/vis_image/000015337.jpg_1.png b/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1/20230606_010416/vis_data/vis_image/000015337.jpg_1.png new file mode 100644 index 0000000000000000000000000000000000000000..73a32bf366b2cad3a4b428e9de53f73883016676 Binary files /dev/null and b/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1/20230606_010416/vis_data/vis_image/000015337.jpg_1.png differ diff --git a/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1/20230606_010416/vis_data/vis_image/000015337.jpg_10.png b/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1/20230606_010416/vis_data/vis_image/000015337.jpg_10.png new file mode 100644 index 0000000000000000000000000000000000000000..3a6c28052ea88dec817a1bc243307b9c6eb6928c Binary files /dev/null and b/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1/20230606_010416/vis_data/vis_image/000015337.jpg_10.png differ diff --git a/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1/20230606_010416/vis_data/vis_image/000015337.jpg_2.png b/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1/20230606_010416/vis_data/vis_image/000015337.jpg_2.png new file mode 100644 index 0000000000000000000000000000000000000000..d19450541ceabbeb131abf610981e765b3d3e594 Binary files /dev/null and b/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1/20230606_010416/vis_data/vis_image/000015337.jpg_2.png differ diff --git a/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1/20230606_010416/vis_data/vis_image/000015337.jpg_3.png b/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1/20230606_010416/vis_data/vis_image/000015337.jpg_3.png new file mode 100644 index 0000000000000000000000000000000000000000..6613850df35d2a6285e2c17dc89226539fdf488d Binary files /dev/null and b/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1/20230606_010416/vis_data/vis_image/000015337.jpg_3.png differ diff --git a/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1/20230606_010416/vis_data/vis_image/000015337.jpg_4.png b/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1/20230606_010416/vis_data/vis_image/000015337.jpg_4.png new file mode 100644 index 0000000000000000000000000000000000000000..ba362d4146a9743410ae9af7ec8e5a12e82ff866 Binary files /dev/null and b/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1/20230606_010416/vis_data/vis_image/000015337.jpg_4.png differ diff --git a/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1/20230606_010416/vis_data/vis_image/000015337.jpg_5.png b/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1/20230606_010416/vis_data/vis_image/000015337.jpg_5.png new file mode 100644 index 0000000000000000000000000000000000000000..7e5ad80c81c6e4d16ff35ad28c73839f4fce8fad Binary files /dev/null and b/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1/20230606_010416/vis_data/vis_image/000015337.jpg_5.png differ diff --git a/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1/20230606_010416/vis_data/vis_image/000015337.jpg_6.png b/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1/20230606_010416/vis_data/vis_image/000015337.jpg_6.png new file mode 100644 index 0000000000000000000000000000000000000000..2c27e75894ed8e6e08d316e9ea4abc52bee8742f Binary files /dev/null and b/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1/20230606_010416/vis_data/vis_image/000015337.jpg_6.png differ diff --git a/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1/20230606_010416/vis_data/vis_image/000015337.jpg_7.png b/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1/20230606_010416/vis_data/vis_image/000015337.jpg_7.png new file mode 100644 index 0000000000000000000000000000000000000000..d3b23d2446f83083ac486168aeed73210fb54d92 Binary files /dev/null and b/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1/20230606_010416/vis_data/vis_image/000015337.jpg_7.png differ diff --git a/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1/20230606_010416/vis_data/vis_image/000015337.jpg_8.png b/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1/20230606_010416/vis_data/vis_image/000015337.jpg_8.png new file mode 100644 index 0000000000000000000000000000000000000000..22bb5023e8851e62ebae6294ac01c3bf500d5d3c Binary files /dev/null and b/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1/20230606_010416/vis_data/vis_image/000015337.jpg_8.png differ diff --git a/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1/20230606_010416/vis_data/vis_image/000015337.jpg_9.png b/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1/20230606_010416/vis_data/vis_image/000015337.jpg_9.png new file mode 100644 index 0000000000000000000000000000000000000000..b8a7cb218bb5ce0b02d34f23f112d0a9a11c4e55 Binary files /dev/null and b/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1/20230606_010416/vis_data/vis_image/000015337.jpg_9.png differ diff --git a/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1/20230606_010416/vis_data/vis_image/0004159.png_1.png b/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1/20230606_010416/vis_data/vis_image/0004159.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..1cf3809a76b2b3497a851c90408c068639fbf603 Binary files /dev/null and b/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1/20230606_010416/vis_data/vis_image/0004159.png_1.png differ diff --git a/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1/20230606_010416/vis_data/vis_image/0004159.png_10.png b/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1/20230606_010416/vis_data/vis_image/0004159.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..0b1b08031b7e894e432f23656e5ae97e203a4ec7 Binary files /dev/null and b/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1/20230606_010416/vis_data/vis_image/0004159.png_10.png differ diff --git a/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1/20230606_010416/vis_data/vis_image/0004159.png_2.png b/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1/20230606_010416/vis_data/vis_image/0004159.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..6f101df542974f562fc88edf62ef7c99a7416a95 Binary files /dev/null and b/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1/20230606_010416/vis_data/vis_image/0004159.png_2.png differ diff --git a/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1/20230606_010416/vis_data/vis_image/0004159.png_3.png b/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1/20230606_010416/vis_data/vis_image/0004159.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..5c6c827f59edbbb005a38ddad33ee874e275a2b6 Binary files /dev/null and b/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1/20230606_010416/vis_data/vis_image/0004159.png_3.png differ diff --git a/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1/20230606_010416/vis_data/vis_image/0004159.png_4.png b/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1/20230606_010416/vis_data/vis_image/0004159.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..6e3263cbad78fb7d26299c196a659026e47b1e49 Binary files /dev/null and b/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1/20230606_010416/vis_data/vis_image/0004159.png_4.png differ diff --git a/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1/20230606_010416/vis_data/vis_image/0004159.png_5.png b/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1/20230606_010416/vis_data/vis_image/0004159.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..4319f719434c638130d676496742db67be0f0870 Binary files /dev/null and b/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1/20230606_010416/vis_data/vis_image/0004159.png_5.png differ diff --git a/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1/20230606_010416/vis_data/vis_image/0004159.png_6.png b/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1/20230606_010416/vis_data/vis_image/0004159.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..ce07de0234ec5a545ed8cd6087b1bef7fb97524a Binary files /dev/null and b/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1/20230606_010416/vis_data/vis_image/0004159.png_6.png differ diff --git a/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1/20230606_010416/vis_data/vis_image/0004159.png_7.png b/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1/20230606_010416/vis_data/vis_image/0004159.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..c526fe5a72d70116a16e1aea36399893a072dc1a Binary files /dev/null and b/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1/20230606_010416/vis_data/vis_image/0004159.png_7.png differ diff --git a/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1/20230606_010416/vis_data/vis_image/0004159.png_8.png b/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1/20230606_010416/vis_data/vis_image/0004159.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..303e6d5eae8d351cf26bdfc64d8db1f53da00c3b Binary files /dev/null and b/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1/20230606_010416/vis_data/vis_image/0004159.png_8.png differ diff --git a/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1/20230606_010416/vis_data/vis_image/0004159.png_9.png b/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1/20230606_010416/vis_data/vis_image/0004159.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..257e6bc7f270eb11a96ca10c09c166628e33782a Binary files /dev/null and b/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1/20230606_010416/vis_data/vis_image/0004159.png_9.png differ diff --git a/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1/20230606_010416/vis_data/vis_image/0008318.png_1.png b/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1/20230606_010416/vis_data/vis_image/0008318.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..e42a766dbafee37661655c9cab66e26e089852e7 Binary files /dev/null and b/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1/20230606_010416/vis_data/vis_image/0008318.png_1.png differ diff --git a/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1/20230606_010416/vis_data/vis_image/0008318.png_10.png b/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1/20230606_010416/vis_data/vis_image/0008318.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..bc3574708106ae8284e77c83c961b1056f197d19 Binary files /dev/null and b/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1/20230606_010416/vis_data/vis_image/0008318.png_10.png differ diff --git a/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1/20230606_010416/vis_data/vis_image/0008318.png_2.png b/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1/20230606_010416/vis_data/vis_image/0008318.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..75267dd714cc08dc1ff80e9bbd8d0032a66ec3d7 Binary files /dev/null and b/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1/20230606_010416/vis_data/vis_image/0008318.png_2.png differ diff --git a/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1/20230606_010416/vis_data/vis_image/0008318.png_3.png b/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1/20230606_010416/vis_data/vis_image/0008318.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..2f52eeb64a004490750c74aeab85e210a6fa9b2a Binary files /dev/null and b/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1/20230606_010416/vis_data/vis_image/0008318.png_3.png differ diff --git a/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1/20230606_010416/vis_data/vis_image/0008318.png_4.png b/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1/20230606_010416/vis_data/vis_image/0008318.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..e8b476f2728d2695e355ce5114994f459fffbec6 Binary files /dev/null and b/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1/20230606_010416/vis_data/vis_image/0008318.png_4.png differ diff --git a/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1/20230606_010416/vis_data/vis_image/0008318.png_5.png b/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1/20230606_010416/vis_data/vis_image/0008318.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..98e9791ebdc315a3a44753f8f77f3587afe22ef6 Binary files /dev/null and b/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1/20230606_010416/vis_data/vis_image/0008318.png_5.png differ diff --git a/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1/20230606_010416/vis_data/vis_image/0008318.png_6.png b/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1/20230606_010416/vis_data/vis_image/0008318.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..304ff3d9fa62e0dfb8d1e862ae390b26b899c954 Binary files /dev/null and b/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1/20230606_010416/vis_data/vis_image/0008318.png_6.png differ diff --git a/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1/20230606_010416/vis_data/vis_image/0008318.png_7.png b/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1/20230606_010416/vis_data/vis_image/0008318.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..55573efe75238a80dc73597c49eab55b22f227af Binary files /dev/null and b/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1/20230606_010416/vis_data/vis_image/0008318.png_7.png differ diff --git a/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1/20230606_010416/vis_data/vis_image/0008318.png_8.png b/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1/20230606_010416/vis_data/vis_image/0008318.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..c2b590cc74f1501cb17a3999d21205016bd930dd Binary files /dev/null and b/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1/20230606_010416/vis_data/vis_image/0008318.png_8.png differ diff --git a/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1/20230606_010416/vis_data/vis_image/0008318.png_9.png b/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1/20230606_010416/vis_data/vis_image/0008318.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..0794e7ac90d4f4840485bcb9a62cffb69cda6e20 Binary files /dev/null and b/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1/20230606_010416/vis_data/vis_image/0008318.png_9.png differ diff --git a/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1.py b/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1.py new file mode 100644 index 0000000000000000000000000000000000000000..01d25c41d41ae62d6798eabf6e093ccc9c38588d --- /dev/null +++ b/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1.py @@ -0,0 +1,366 @@ +optim_wrapper = dict( + optimizer=dict( + type='AdamW', + lr=0.0004, + weight_decay=0.05, + eps=1e-08, + betas=(0.9, 0.999), + _scope_='mmpretrain'), + paramwise_cfg=dict( + norm_decay_mult=0.0, + bias_decay_mult=0.0, + flat_decay_mult=0.0, + custom_keys=dict({ + '.absolute_pos_embed': dict(decay_mult=0.0), + '.relative_position_bias_table': dict(decay_mult=0.0) + })), + type='AmpOptimWrapper', + dtype='bfloat16', + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=False, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=4096) +model = dict( + type='ImageClassifier', + backbone=dict( + frozen_stages=2, + type='ConvNeXt', + arch='small', + drop_path_rate=0.4, + init_cfg=dict( + type='Pretrained', + checkpoint= + 'https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth', + prefix='backbone')), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=768, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + init_cfg=None), + init_cfg=dict( + type='TruncNormal', layer=['Conv2d', 'Linear'], std=0.02, bias=0.0), + train_cfg=None) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/all_0.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/all_1.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3fake8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/cc1m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3real8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV2-1-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/if-dpmsolver++-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV2-1-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/if-dpmsolver++-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1' diff --git a/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1/epoch_1.pth b/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1/epoch_1.pth new file mode 100644 index 0000000000000000000000000000000000000000..88f52a4134f940b967214c1640fcaff24c42b25a --- /dev/null +++ b/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1/epoch_1.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:d541abc83258a86623e1de6e65a57485a5716d7c15c3c491b1895d69b484740d +size 782128282 diff --git a/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1/epoch_10.pth b/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1/epoch_10.pth new file mode 100644 index 0000000000000000000000000000000000000000..0fd270de0624a0d8e44a480b7100d80427f982c6 --- /dev/null +++ b/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1/epoch_10.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:b7ed40d8b5660d1df8c3f3dc10bc2b6f1aba068b43ff0258827c9a479233df4b +size 783483930 diff --git a/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1/epoch_2.pth b/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1/epoch_2.pth new file mode 100644 index 0000000000000000000000000000000000000000..1160a382eaf0b99ee429884c5235b7c8cffd8a58 --- /dev/null +++ b/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1/epoch_2.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:4208375e40cdfc9a01db2249b8c52bf03dd76351c42d49e38e4c5c7da85d641a +size 782280026 diff --git a/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1/epoch_3.pth b/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1/epoch_3.pth new file mode 100644 index 0000000000000000000000000000000000000000..e84a1f5420281c215ddc151471e6e23b9a842317 --- /dev/null +++ b/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1/epoch_3.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:d3d0ac49939508fd97247ef971197d5692956509331ffbd07f2893ad52f99551 +size 782430554 diff --git a/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1/epoch_4.pth b/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1/epoch_4.pth new file mode 100644 index 0000000000000000000000000000000000000000..05d3b0f737f70f18f7b6106870a35512dc82c756 --- /dev/null +++ b/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1/epoch_4.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:347592f453bb7090505fbe9a363337e9c039f287902645c51883166acb551cb0 +size 782581082 diff --git a/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1/epoch_5.pth b/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1/epoch_5.pth new file mode 100644 index 0000000000000000000000000000000000000000..f1e8b2a0f8425560336f85635446cc94d7f5cd17 --- /dev/null +++ b/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1/epoch_5.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:687c703e40d553a3141851c05ddf6475c8b62ba6c8d59f90e699d3de69eb9453 +size 782731098 diff --git a/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1/epoch_6.pth b/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1/epoch_6.pth new file mode 100644 index 0000000000000000000000000000000000000000..cea8d55efcc2fa3c7c10decff0dcaab8d2ba050e --- /dev/null +++ b/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1/epoch_6.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:a9d3472929b974a39f653bfddc17578cb17bc4f15767a54c394ce6e3be7b84ee +size 782881370 diff --git a/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1/epoch_7.pth b/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1/epoch_7.pth new file mode 100644 index 0000000000000000000000000000000000000000..f6991e509a3c3a0c51fc67693e5423ba6b6a4563 --- /dev/null +++ b/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1/epoch_7.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:b1bd4bd88685c24a108d15337dfbfdb86128364a1c802a367217051dd9b23c5d +size 783031514 diff --git a/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1/epoch_8.pth b/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1/epoch_8.pth new file mode 100644 index 0000000000000000000000000000000000000000..3783ee532a265ab7ba4bcafd076afca1692de2e6 --- /dev/null +++ b/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1/epoch_8.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:152028bb0f6d2bb365c7ba7df9f815a705e5ca6c3da0d64a6cae71d4059dfe21 +size 783181466 diff --git a/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1/epoch_9.pth b/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1/epoch_9.pth new file mode 100644 index 0000000000000000000000000000000000000000..ab2c3b10567b909da9ab1d5ec7eef80e6d5518da --- /dev/null +++ b/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1/epoch_9.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:12e5d3e9d251a0a0d285cdd20b99d25afa23e0cad4f5ed55c0dc7280ccc0e744 +size 783332378 diff --git a/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1/last_checkpoint b/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1/last_checkpoint new file mode 100644 index 0000000000000000000000000000000000000000..7057cf48f1962caa598d102825772a0a9bbe1df7 --- /dev/null +++ b/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1/last_checkpoint @@ -0,0 +1 @@ +/mnt/petrelfs/luzeyu/workspace/fakebench/mmpretrain/workdir/convnext_small_4xb256_all2_1m_lr4e-4_aug_5e-1/epoch_10.pth \ No newline at end of file diff --git a/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1/20230605_024312/20230605_024312.log b/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1/20230605_024312/20230605_024312.log new file mode 100644 index 0000000000000000000000000000000000000000..194f1aa886ebafb6b569fc8e41a895dc23237401 --- /dev/null +++ b/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1/20230605_024312/20230605_024312.log @@ -0,0 +1,1987 @@ +2023/06/05 02:43:17 - mmengine - INFO - +------------------------------------------------------------ +System environment: + sys.platform: linux + Python: 3.10.9 (main, Mar 8 2023, 10:47:38) [GCC 11.2.0] + CUDA available: True + numpy_random_seed: 589172170 + GPU 0,1,2,3: NVIDIA A100-SXM4-80GB + CUDA_HOME: /mnt/petrelfs/share/cuda-11.6 + NVCC: Cuda compilation tools, release 11.6, V11.6.124 + GCC: gcc (GCC) 7.5.0 + PyTorch: 1.13.1 + PyTorch compiling details: PyTorch built with: + - GCC 9.3 + - C++ Version: 201402 + - Intel(R) oneAPI Math Kernel Library Version 2021.4-Product Build 20210904 for Intel(R) 64 architecture applications + - Intel(R) MKL-DNN v2.6.0 (Git Hash 52b5f107dd9cf10910aaa19cb47f3abf9b349815) + - OpenMP 201511 (a.k.a. OpenMP 4.5) + - LAPACK is enabled (usually provided by MKL) + - NNPACK is enabled + - CPU capability usage: AVX2 + - CUDA Runtime 11.6 + - NVCC architecture flags: -gencode;arch=compute_37,code=sm_37;-gencode;arch=compute_50,code=sm_50;-gencode;arch=compute_60,code=sm_60;-gencode;arch=compute_61,code=sm_61;-gencode;arch=compute_70,code=sm_70;-gencode;arch=compute_75,code=sm_75;-gencode;arch=compute_80,code=sm_80;-gencode;arch=compute_86,code=sm_86;-gencode;arch=compute_37,code=compute_37 + - CuDNN 8.3.2 (built against CUDA 11.5) + - Magma 2.6.1 + - Build settings: BLAS_INFO=mkl, BUILD_TYPE=Release, CUDA_VERSION=11.6, CUDNN_VERSION=8.3.2, CXX_COMPILER=/opt/rh/devtoolset-9/root/usr/bin/c++, CXX_FLAGS= -fabi-version=11 -Wno-deprecated -fvisibility-inlines-hidden -DUSE_PTHREADPOOL -fopenmp -DNDEBUG -DUSE_KINETO -DUSE_FBGEMM -DUSE_QNNPACK -DUSE_PYTORCH_QNNPACK -DUSE_XNNPACK -DSYMBOLICATE_MOBILE_DEBUG_HANDLE -DEDGE_PROFILER_USE_KINETO -O2 -fPIC -Wno-narrowing -Wall -Wextra -Werror=return-type -Werror=non-virtual-dtor -Wno-missing-field-initializers -Wno-type-limits -Wno-array-bounds -Wno-unknown-pragmas -Wunused-local-typedefs -Wno-unused-parameter -Wno-unused-function -Wno-unused-result -Wno-strict-overflow -Wno-strict-aliasing -Wno-error=deprecated-declarations -Wno-stringop-overflow -Wno-psabi -Wno-error=pedantic -Wno-error=redundant-decls -Wno-error=old-style-cast -fdiagnostics-color=always -faligned-new -Wno-unused-but-set-variable -Wno-maybe-uninitialized -fno-math-errno -fno-trapping-math -Werror=format -Werror=cast-function-type -Wno-stringop-overflow, LAPACK_INFO=mkl, PERF_WITH_AVX=1, PERF_WITH_AVX2=1, PERF_WITH_AVX512=1, TORCH_VERSION=1.13.1, USE_CUDA=ON, USE_CUDNN=ON, USE_EXCEPTION_PTR=1, USE_GFLAGS=OFF, USE_GLOG=OFF, USE_MKL=ON, USE_MKLDNN=ON, USE_MPI=OFF, USE_NCCL=ON, USE_NNPACK=ON, USE_OPENMP=ON, USE_ROCM=OFF, + + TorchVision: 0.14.1 + OpenCV: 4.7.0 + MMEngine: 0.7.3 + +Runtime environment: + cudnn_benchmark: True + mp_cfg: {'mp_start_method': 'fork', 'opencv_num_threads': 0} + dist_cfg: {'backend': 'nccl'} + seed: None + deterministic: False + Distributed launcher: slurm + Distributed training: True + GPU number: 4 +------------------------------------------------------------ + +2023/06/05 02:43:21 - mmengine - INFO - Config: +optim_wrapper = dict( + optimizer=dict( + type='AdamW', + lr=0.0004, + weight_decay=0.05, + eps=1e-08, + betas=(0.9, 0.999), + _scope_='mmpretrain'), + paramwise_cfg=dict( + norm_decay_mult=0.0, + bias_decay_mult=0.0, + flat_decay_mult=0.0, + custom_keys=dict({ + '.absolute_pos_embed': dict(decay_mult=0.0), + '.relative_position_bias_table': dict(decay_mult=0.0) + })), + type='AmpOptimWrapper', + dtype='bfloat16', + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=False, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=4096) +model = dict( + type='ImageClassifier', + backbone=dict( + frozen_stages=2, + type='ConvNeXt', + arch='small', + drop_path_rate=0.4, + init_cfg=dict( + type='Pretrained', + checkpoint= + 'https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth', + prefix='backbone')), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=768, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + init_cfg=None), + init_cfg=dict( + type='TruncNormal', layer=['Conv2d', 'Linear'], std=0.02, bias=0.0), + train_cfg=None) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/all1.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/all2.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/all3.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3fake8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/cc1m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3real8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV2-1-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/if-dpmsolver++-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV2-1-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/if-dpmsolver++-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1' + +2023/06/05 02:43:32 - mmengine - INFO - Hooks will be executed in the following order: +before_run: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_load_checkpoint: +(ABOVE_NORMAL) EMAHook + -------------------- +before_train: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_train_epoch: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(NORMAL ) DistSamplerSeedHook + -------------------- +before_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook + -------------------- +after_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +after_train_epoch: +(NORMAL ) IterTimerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_val_epoch: +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook + -------------------- +before_val_iter: +(NORMAL ) IterTimerHook + -------------------- +after_val_iter: +(NORMAL ) IterTimerHook +(NORMAL ) VisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_val_epoch: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_save_checkpoint: +(ABOVE_NORMAL) EMAHook + -------------------- +after_train: +(VERY_LOW ) CheckpointHook + -------------------- +before_test_epoch: +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook + -------------------- +before_test_iter: +(NORMAL ) IterTimerHook + -------------------- +after_test_iter: +(NORMAL ) IterTimerHook +(NORMAL ) VisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_test_epoch: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_run: +(BELOW_NORMAL) LoggerHook + -------------------- +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.2.0.weight:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.2.0.bias:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.2.1.bias:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.3.0.weight:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.3.0.bias:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.3.1.bias:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.2.0.gamma:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.2.0.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.2.0.norm.weight:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.2.0.norm.bias:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.2.0.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.2.0.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.2.1.gamma:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.2.1.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.2.1.norm.weight:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.2.1.norm.bias:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.2.1.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.2.1.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.2.2.gamma:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.2.2.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.2.2.norm.weight:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.2.2.norm.bias:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.2.2.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.2.2.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.2.3.gamma:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.2.3.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.2.3.norm.weight:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.2.3.norm.bias:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.2.3.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.2.3.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.2.4.gamma:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.2.4.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.2.4.norm.weight:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.2.4.norm.bias:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.2.4.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.2.4.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.2.5.gamma:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.2.5.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.2.5.norm.weight:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.2.5.norm.bias:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.2.5.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.2.5.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.2.6.gamma:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.2.6.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.2.6.norm.weight:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.2.6.norm.bias:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.2.6.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.2.6.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.2.7.gamma:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.2.7.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.2.7.norm.weight:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.2.7.norm.bias:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.2.7.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.2.7.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.2.8.gamma:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.2.8.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.2.8.norm.weight:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.2.8.norm.bias:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.2.8.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.2.8.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.2.9.gamma:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.2.9.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.2.9.norm.weight:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.2.9.norm.bias:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.2.9.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.2.9.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.2.10.gamma:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.2.10.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.2.10.norm.weight:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.2.10.norm.bias:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.2.10.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.2.10.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.2.11.gamma:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.2.11.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.2.11.norm.weight:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.2.11.norm.bias:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.2.11.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.2.11.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.2.12.gamma:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.2.12.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.2.12.norm.weight:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.2.12.norm.bias:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.2.12.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.2.12.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.2.13.gamma:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.2.13.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.2.13.norm.weight:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.2.13.norm.bias:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.2.13.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.2.13.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.2.14.gamma:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.2.14.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.2.14.norm.weight:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.2.14.norm.bias:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.2.14.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.2.14.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.2.15.gamma:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.2.15.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.2.15.norm.weight:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.2.15.norm.bias:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.2.15.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.2.15.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.2.16.gamma:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.2.16.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.2.16.norm.weight:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.2.16.norm.bias:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.2.16.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.2.16.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.2.17.gamma:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.2.17.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.2.17.norm.weight:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.2.17.norm.bias:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.2.17.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.2.17.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.2.18.gamma:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.2.18.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.2.18.norm.weight:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.2.18.norm.bias:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.2.18.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.2.18.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.2.19.gamma:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.2.19.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.2.19.norm.weight:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.2.19.norm.bias:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.2.19.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.2.19.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.2.20.gamma:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.2.20.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.2.20.norm.weight:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.2.20.norm.bias:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.2.20.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.2.20.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.2.21.gamma:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.2.21.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.2.21.norm.weight:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.2.21.norm.bias:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.2.21.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.2.21.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.2.22.gamma:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.2.22.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.2.22.norm.weight:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.2.22.norm.bias:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.2.22.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.2.22.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.2.23.gamma:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.2.23.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.2.23.norm.weight:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.2.23.norm.bias:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.2.23.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.2.23.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.2.24.gamma:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.2.24.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.2.24.norm.weight:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.2.24.norm.bias:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.2.24.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.2.24.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.2.25.gamma:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.2.25.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.2.25.norm.weight:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.2.25.norm.bias:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.2.25.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.2.25.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.2.26.gamma:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.2.26.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.2.26.norm.weight:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.2.26.norm.bias:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.2.26.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.2.26.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.3.0.gamma:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.3.0.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.3.0.norm.weight:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.3.0.norm.bias:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.3.0.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.3.0.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.3.1.gamma:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.3.1.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.3.1.norm.weight:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.3.1.norm.bias:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.3.1.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.3.1.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.3.2.gamma:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.3.2.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.3.2.norm.weight:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.3.2.norm.bias:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.3.2.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.stages.3.2.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.norm3.weight:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- backbone.norm3.bias:weight_decay=0.0 +2023/06/05 02:43:46 - mmengine - INFO - paramwise_options -- head.fc.bias:weight_decay=0.0 +2023/06/05 02:43:51 - mmengine - INFO - load backbone in model from: https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth +Name of parameter - Initialization information + +backbone.downsample_layers.0.0.weight - torch.Size([96, 3, 4, 4]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.0.0.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.0.1.weight - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.0.1.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.1.0.weight - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.1.0.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.1.1.weight - torch.Size([192, 96, 2, 2]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.1.1.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.2.0.weight - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.2.0.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.2.1.weight - torch.Size([384, 192, 2, 2]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.2.1.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.3.0.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.3.0.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.3.1.weight - torch.Size([768, 384, 2, 2]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.3.1.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.0.gamma - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.0.depthwise_conv.weight - torch.Size([96, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.0.depthwise_conv.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.0.norm.weight - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.0.norm.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.0.pointwise_conv1.weight - torch.Size([384, 96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.0.pointwise_conv1.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.0.pointwise_conv2.weight - torch.Size([96, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.0.pointwise_conv2.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.1.gamma - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.1.depthwise_conv.weight - torch.Size([96, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.1.depthwise_conv.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.1.norm.weight - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.1.norm.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.1.pointwise_conv1.weight - torch.Size([384, 96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.1.pointwise_conv1.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.1.pointwise_conv2.weight - torch.Size([96, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.1.pointwise_conv2.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.2.gamma - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.2.depthwise_conv.weight - torch.Size([96, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.2.depthwise_conv.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.2.norm.weight - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.2.norm.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.2.pointwise_conv1.weight - torch.Size([384, 96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.2.pointwise_conv1.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.2.pointwise_conv2.weight - torch.Size([96, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.2.pointwise_conv2.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.0.gamma - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.0.depthwise_conv.weight - torch.Size([192, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.0.depthwise_conv.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.0.norm.weight - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.0.norm.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.0.pointwise_conv1.weight - torch.Size([768, 192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.0.pointwise_conv1.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.0.pointwise_conv2.weight - torch.Size([192, 768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.0.pointwise_conv2.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.1.gamma - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.1.depthwise_conv.weight - torch.Size([192, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.1.depthwise_conv.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.1.norm.weight - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.1.norm.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.1.pointwise_conv1.weight - torch.Size([768, 192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.1.pointwise_conv1.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.1.pointwise_conv2.weight - torch.Size([192, 768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.1.pointwise_conv2.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.2.gamma - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.2.depthwise_conv.weight - torch.Size([192, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.2.depthwise_conv.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.2.norm.weight - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.2.norm.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.2.pointwise_conv1.weight - torch.Size([768, 192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.2.pointwise_conv1.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.2.pointwise_conv2.weight - torch.Size([192, 768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.2.pointwise_conv2.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.0.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.0.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.0.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.0.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.0.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.0.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.0.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.0.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.0.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.1.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.1.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.1.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.1.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.1.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.1.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.1.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.1.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.1.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.2.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.2.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.2.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.2.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.2.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.2.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.2.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.2.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.2.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.3.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.3.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.3.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.3.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.3.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.3.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.3.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.3.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.3.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.4.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.4.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.4.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.4.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.4.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.4.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.4.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.4.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.4.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.5.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.5.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.5.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.5.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.5.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.5.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.5.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.5.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.5.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.6.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.6.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.6.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.6.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.6.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.6.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.6.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.6.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.6.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.7.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.7.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.7.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.7.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.7.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.7.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.7.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.7.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.7.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.8.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.8.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.8.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.8.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.8.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.8.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.8.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.8.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.8.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.9.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.9.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.9.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.9.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.9.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.9.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.9.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.9.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.9.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.10.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.10.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.10.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.10.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.10.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.10.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.10.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.10.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.10.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.11.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.11.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.11.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.11.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.11.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.11.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.11.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.11.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.11.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.12.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.12.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.12.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.12.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.12.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.12.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.12.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.12.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.12.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.13.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.13.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.13.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.13.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.13.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.13.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.13.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.13.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.13.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.14.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.14.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.14.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.14.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.14.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.14.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.14.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.14.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.14.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.15.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.15.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.15.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.15.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.15.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.15.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.15.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.15.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.15.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.16.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.16.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.16.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.16.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.16.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.16.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.16.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.16.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.16.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.17.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.17.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.17.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.17.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.17.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.17.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.17.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.17.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.17.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.18.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.18.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.18.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.18.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.18.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.18.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.18.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.18.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.18.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.19.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.19.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.19.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.19.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.19.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.19.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.19.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.19.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.19.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.20.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.20.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.20.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.20.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.20.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.20.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.20.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.20.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.20.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.21.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.21.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.21.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.21.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.21.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.21.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.21.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.21.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.21.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.22.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.22.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.22.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.22.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.22.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.22.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.22.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.22.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.22.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.23.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.23.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.23.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.23.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.23.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.23.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.23.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.23.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.23.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.24.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.24.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.24.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.24.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.24.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.24.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.24.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.24.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.24.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.25.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.25.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.25.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.25.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.25.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.25.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.25.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.25.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.25.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.26.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.26.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.26.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.26.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.26.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.26.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.26.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.26.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.26.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.0.gamma - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.0.depthwise_conv.weight - torch.Size([768, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.0.depthwise_conv.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.0.norm.weight - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.0.norm.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.0.pointwise_conv1.weight - torch.Size([3072, 768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.0.pointwise_conv1.bias - torch.Size([3072]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.0.pointwise_conv2.weight - torch.Size([768, 3072]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.0.pointwise_conv2.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.1.gamma - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.1.depthwise_conv.weight - torch.Size([768, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.1.depthwise_conv.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.1.norm.weight - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.1.norm.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.1.pointwise_conv1.weight - torch.Size([3072, 768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.1.pointwise_conv1.bias - torch.Size([3072]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.1.pointwise_conv2.weight - torch.Size([768, 3072]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.1.pointwise_conv2.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.2.gamma - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.2.depthwise_conv.weight - torch.Size([768, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.2.depthwise_conv.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.2.norm.weight - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.2.norm.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.2.pointwise_conv1.weight - torch.Size([3072, 768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.2.pointwise_conv1.bias - torch.Size([3072]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.2.pointwise_conv2.weight - torch.Size([768, 3072]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.2.pointwise_conv2.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.norm3.weight - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.norm3.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +head.fc.weight - torch.Size([2, 768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +head.fc.bias - torch.Size([2]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 +2023/06/05 02:43:51 - mmengine - WARNING - "FileClient" will be deprecated in future. Please use io functions in https://mmengine.readthedocs.io/en/latest/api/fileio.html#file-io +2023/06/05 02:43:51 - mmengine - WARNING - "HardDiskBackend" is the alias of "LocalBackend" and the former will be deprecated in future. +2023/06/05 02:43:51 - mmengine - INFO - Checkpoints will be saved to /mnt/petrelfs/luzeyu/workspace/fakebench/mmpretrain/workdir/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1. +2023/06/05 02:45:11 - mmengine - INFO - Epoch(train) [1][ 100/2056] lr: 3.9998e-04 eta: 4:31:12 time: 0.7631 data_time: 0.2428 memory: 15587 loss: 0.0623 +2023/06/05 02:46:25 - mmengine - INFO - Epoch(train) [1][ 200/2056] lr: 3.9991e-04 eta: 4:19:56 time: 0.8761 data_time: 0.1870 memory: 15587 loss: 0.0472 +2023/06/05 02:47:38 - mmengine - INFO - Epoch(train) [1][ 300/2056] lr: 3.9980e-04 eta: 4:14:45 time: 0.6199 data_time: 0.0017 memory: 15587 loss: 0.0453 +2023/06/05 02:48:52 - mmengine - INFO - Epoch(train) [1][ 400/2056] lr: 3.9964e-04 eta: 4:12:26 time: 0.8003 data_time: 0.0016 memory: 15587 loss: 0.0395 +2023/06/05 02:50:06 - mmengine - INFO - Epoch(train) [1][ 500/2056] lr: 3.9943e-04 eta: 4:10:20 time: 0.7467 data_time: 0.0015 memory: 15587 loss: 0.0337 +2023/06/05 02:51:19 - mmengine - INFO - Epoch(train) [1][ 600/2056] lr: 3.9918e-04 eta: 4:08:28 time: 0.7769 data_time: 0.0017 memory: 15587 loss: 0.0297 +2023/06/05 02:52:32 - mmengine - INFO - Epoch(train) [1][ 700/2056] lr: 3.9889e-04 eta: 4:06:22 time: 0.7020 data_time: 0.0015 memory: 15587 loss: 0.0250 +2023/06/05 02:53:44 - mmengine - INFO - Epoch(train) [1][ 800/2056] lr: 3.9855e-04 eta: 4:03:52 time: 0.7306 data_time: 0.0014 memory: 15587 loss: 0.0194 +2023/06/05 02:54:59 - mmengine - INFO - Epoch(train) [1][ 900/2056] lr: 3.9816e-04 eta: 4:03:00 time: 0.7105 data_time: 0.0018 memory: 15587 loss: 0.0278 +2023/06/05 02:56:11 - mmengine - INFO - Exp name: convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1_20230605_024312 +2023/06/05 02:56:11 - mmengine - INFO - Epoch(train) [1][1000/2056] lr: 3.9773e-04 eta: 4:01:11 time: 0.7125 data_time: 0.0018 memory: 15587 loss: 0.0308 +2023/06/05 02:57:25 - mmengine - INFO - Epoch(train) [1][1100/2056] lr: 3.9726e-04 eta: 3:59:48 time: 0.7369 data_time: 0.0016 memory: 15587 loss: 0.0199 +2023/06/05 02:58:38 - mmengine - INFO - Epoch(train) [1][1200/2056] lr: 3.9674e-04 eta: 3:58:30 time: 0.7870 data_time: 0.0019 memory: 15587 loss: 0.0224 +2023/06/05 02:59:53 - mmengine - INFO - Epoch(train) [1][1300/2056] lr: 3.9617e-04 eta: 3:57:23 time: 0.8033 data_time: 0.0020 memory: 15587 loss: 0.0198 +2023/06/05 03:01:05 - mmengine - INFO - Epoch(train) [1][1400/2056] lr: 3.9556e-04 eta: 3:55:51 time: 0.7527 data_time: 0.0018 memory: 15587 loss: 0.0241 +2023/06/05 03:02:20 - mmengine - INFO - Epoch(train) [1][1500/2056] lr: 3.9491e-04 eta: 3:54:44 time: 0.7369 data_time: 0.0017 memory: 15587 loss: 0.0207 +2023/06/05 03:03:34 - mmengine - INFO - Epoch(train) [1][1600/2056] lr: 3.9421e-04 eta: 3:53:29 time: 0.7223 data_time: 0.0015 memory: 15587 loss: 0.0189 +2023/06/05 03:04:49 - mmengine - INFO - Epoch(train) [1][1700/2056] lr: 3.9347e-04 eta: 3:52:29 time: 0.7528 data_time: 0.0016 memory: 15587 loss: 0.0199 +2023/06/05 03:06:07 - mmengine - INFO - Epoch(train) [1][1800/2056] lr: 3.9268e-04 eta: 3:52:03 time: 0.6903 data_time: 0.0015 memory: 15587 loss: 0.0257 +2023/06/05 03:07:21 - mmengine - INFO - Epoch(train) [1][1900/2056] lr: 3.9185e-04 eta: 3:50:41 time: 0.7485 data_time: 0.0022 memory: 15587 loss: 0.0288 +2023/06/05 03:08:34 - mmengine - INFO - Exp name: convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1_20230605_024312 +2023/06/05 03:08:34 - mmengine - INFO - Epoch(train) [1][2000/2056] lr: 3.9097e-04 eta: 3:49:22 time: 0.7453 data_time: 0.0015 memory: 15587 loss: 0.0209 +2023/06/05 03:09:16 - mmengine - INFO - Exp name: convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1_20230605_024312 +2023/06/05 03:09:16 - mmengine - INFO - Saving checkpoint at 1 epochs +2023/06/05 03:10:14 - mmengine - INFO - Epoch(val) [1][60/60] accuracy/top1: 77.5268 single-label/precision_classwise: [98.64710235595703, 48.67045211791992] single-label/recall_classwise: [72.41967010498047, 96.3410873413086] single-label/f1-score_classwise: [83.52283477783203, 64.67022705078125] data_time: 0.1075 time: 0.5419 +2023/06/05 03:11:33 - mmengine - INFO - Epoch(train) [2][ 100/2056] lr: 3.8952e-04 eta: 3:48:03 time: 0.7441 data_time: 0.0021 memory: 15587 loss: 0.0186 +2023/06/05 03:12:48 - mmengine - INFO - Epoch(train) [2][ 200/2056] lr: 3.8854e-04 eta: 3:46:57 time: 0.7279 data_time: 0.0020 memory: 15585 loss: 0.0196 +2023/06/05 03:14:01 - mmengine - INFO - Epoch(train) [2][ 300/2056] lr: 3.8751e-04 eta: 3:45:35 time: 0.7977 data_time: 0.0018 memory: 15585 loss: 0.0131 +2023/06/05 03:15:14 - mmengine - INFO - Epoch(train) [2][ 400/2056] lr: 3.8644e-04 eta: 3:44:10 time: 0.6740 data_time: 0.0018 memory: 15585 loss: 0.0201 +2023/06/05 03:16:27 - mmengine - INFO - Epoch(train) [2][ 500/2056] lr: 3.8533e-04 eta: 3:42:43 time: 0.7429 data_time: 0.0018 memory: 15585 loss: 0.0137 +2023/06/05 03:17:39 - mmengine - INFO - Epoch(train) [2][ 600/2056] lr: 3.8417e-04 eta: 3:41:14 time: 0.7237 data_time: 0.0016 memory: 15585 loss: 0.0254 +2023/06/05 03:18:52 - mmengine - INFO - Epoch(train) [2][ 700/2056] lr: 3.8298e-04 eta: 3:39:55 time: 0.7169 data_time: 0.0015 memory: 15585 loss: 0.0123 +2023/06/05 03:20:06 - mmengine - INFO - Epoch(train) [2][ 800/2056] lr: 3.8174e-04 eta: 3:38:39 time: 0.7130 data_time: 0.0017 memory: 15585 loss: 0.0218 +2023/06/05 03:21:19 - mmengine - INFO - Epoch(train) [2][ 900/2056] lr: 3.8046e-04 eta: 3:37:21 time: 0.7371 data_time: 0.0018 memory: 15585 loss: 0.0202 +2023/06/05 03:21:54 - mmengine - INFO - Exp name: convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1_20230605_024312 +2023/06/05 03:22:33 - mmengine - INFO - Epoch(train) [2][1000/2056] lr: 3.7914e-04 eta: 3:36:02 time: 0.7195 data_time: 0.0018 memory: 15585 loss: 0.0199 +2023/06/05 03:25:16 - mmengine - INFO - Epoch(train) [2][1100/2056] lr: 3.7778e-04 eta: 3:43:00 time: 0.7288 data_time: 0.0017 memory: 15585 loss: 0.0245 +2023/06/05 03:26:29 - mmengine - INFO - Epoch(train) [2][1200/2056] lr: 3.7637e-04 eta: 3:41:22 time: 0.6837 data_time: 0.0016 memory: 15585 loss: 0.0194 +2023/06/05 03:27:43 - mmengine - INFO - Epoch(train) [2][1300/2056] lr: 3.7493e-04 eta: 3:39:50 time: 0.6828 data_time: 0.0016 memory: 15585 loss: 0.0194 +2023/06/05 03:28:56 - mmengine - INFO - Epoch(train) [2][1400/2056] lr: 3.7345e-04 eta: 3:38:18 time: 0.7277 data_time: 0.0018 memory: 15585 loss: 0.0206 +2023/06/05 03:30:08 - mmengine - INFO - Epoch(train) [2][1500/2056] lr: 3.7193e-04 eta: 3:36:38 time: 0.7252 data_time: 0.0018 memory: 15585 loss: 0.0192 +2023/06/05 03:31:21 - mmengine - INFO - Epoch(train) [2][1600/2056] lr: 3.7037e-04 eta: 3:35:07 time: 0.6932 data_time: 0.0017 memory: 15585 loss: 0.0203 +2023/06/05 03:32:33 - mmengine - INFO - Epoch(train) [2][1700/2056] lr: 3.6877e-04 eta: 3:33:30 time: 0.6836 data_time: 0.0017 memory: 15585 loss: 0.0125 +2023/06/05 03:33:47 - mmengine - INFO - Epoch(train) [2][1800/2056] lr: 3.6714e-04 eta: 3:32:04 time: 0.6878 data_time: 0.0018 memory: 15585 loss: 0.0215 +2023/06/05 03:35:00 - mmengine - INFO - Epoch(train) [2][1900/2056] lr: 3.6546e-04 eta: 3:30:34 time: 0.7289 data_time: 0.0019 memory: 15585 loss: 0.0186 +2023/06/05 03:35:34 - mmengine - INFO - Exp name: convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1_20230605_024312 +2023/06/05 03:36:14 - mmengine - INFO - Epoch(train) [2][2000/2056] lr: 3.6375e-04 eta: 3:29:10 time: 0.7337 data_time: 0.0021 memory: 15585 loss: 0.0173 +2023/06/05 03:36:54 - mmengine - INFO - Exp name: convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1_20230605_024312 +2023/06/05 03:36:54 - mmengine - INFO - Saving checkpoint at 2 epochs +2023/06/05 03:37:49 - mmengine - INFO - Epoch(val) [2][60/60] accuracy/top1: 95.2253 single-label/precision_classwise: [99.51854705810547, 82.62002563476562] single-label/recall_classwise: [94.38587188720703, 98.31783294677734] single-label/f1-score_classwise: [96.88427734375, 89.78797149658203] data_time: 0.0984 time: 0.4947 +2023/06/05 03:39:08 - mmengine - INFO - Epoch(train) [3][ 100/2056] lr: 3.6101e-04 eta: 3:27:11 time: 0.7854 data_time: 0.1038 memory: 15585 loss: 0.0168 +2023/06/05 03:40:21 - mmengine - INFO - Epoch(train) [3][ 200/2056] lr: 3.5920e-04 eta: 3:25:46 time: 0.7197 data_time: 0.0016 memory: 15585 loss: 0.0186 +2023/06/05 03:41:34 - mmengine - INFO - Epoch(train) [3][ 300/2056] lr: 3.5736e-04 eta: 3:24:16 time: 0.7030 data_time: 0.0018 memory: 15585 loss: 0.0102 +2023/06/05 03:42:48 - mmengine - INFO - Epoch(train) [3][ 400/2056] lr: 3.5548e-04 eta: 3:22:54 time: 0.7280 data_time: 0.0016 memory: 15585 loss: 0.0191 +2023/06/05 03:44:01 - mmengine - INFO - Epoch(train) [3][ 500/2056] lr: 3.5357e-04 eta: 3:21:29 time: 0.6521 data_time: 0.0017 memory: 15585 loss: 0.0142 +2023/06/05 03:45:14 - mmengine - INFO - Epoch(train) [3][ 600/2056] lr: 3.5162e-04 eta: 3:20:03 time: 0.7260 data_time: 0.0017 memory: 15585 loss: 0.0174 +2023/06/05 03:46:26 - mmengine - INFO - Epoch(train) [3][ 700/2056] lr: 3.4964e-04 eta: 3:18:37 time: 0.7121 data_time: 0.0020 memory: 15585 loss: 0.0180 +2023/06/05 03:47:39 - mmengine - INFO - Epoch(train) [3][ 800/2056] lr: 3.4763e-04 eta: 3:17:10 time: 0.6918 data_time: 0.0016 memory: 15585 loss: 0.0129 +2023/06/05 03:48:41 - mmengine - INFO - Exp name: convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1_20230605_024312 +2023/06/05 03:48:49 - mmengine - INFO - Epoch(train) [3][ 900/2056] lr: 3.4558e-04 eta: 3:15:40 time: 0.7706 data_time: 0.0015 memory: 15585 loss: 0.0184 +2023/06/05 03:50:03 - mmengine - INFO - Epoch(train) [3][1000/2056] lr: 3.4350e-04 eta: 3:14:18 time: 0.7217 data_time: 0.0016 memory: 15585 loss: 0.0183 +2023/06/05 03:51:18 - mmengine - INFO - Epoch(train) [3][1100/2056] lr: 3.4138e-04 eta: 3:13:01 time: 0.7417 data_time: 0.0020 memory: 15585 loss: 0.0177 +2023/06/05 03:52:33 - mmengine - INFO - Epoch(train) [3][1200/2056] lr: 3.3924e-04 eta: 3:11:45 time: 0.7296 data_time: 0.0022 memory: 15585 loss: 0.0128 +2023/06/05 03:53:45 - mmengine - INFO - Epoch(train) [3][1300/2056] lr: 3.3706e-04 eta: 3:10:19 time: 0.7496 data_time: 0.0019 memory: 15585 loss: 0.0117 +2023/06/05 03:54:57 - mmengine - INFO - Epoch(train) [3][1400/2056] lr: 3.3485e-04 eta: 3:08:56 time: 0.7832 data_time: 0.0018 memory: 15585 loss: 0.0153 +2023/06/05 03:56:10 - mmengine - INFO - Epoch(train) [3][1500/2056] lr: 3.3261e-04 eta: 3:07:32 time: 0.7567 data_time: 0.0017 memory: 15585 loss: 0.0181 +2023/06/05 03:57:26 - mmengine - INFO - Epoch(train) [3][1600/2056] lr: 3.3035e-04 eta: 3:06:20 time: 0.8014 data_time: 0.0015 memory: 15585 loss: 0.0114 +2023/06/05 03:58:39 - mmengine - INFO - Epoch(train) [3][1700/2056] lr: 3.2805e-04 eta: 3:04:58 time: 0.7804 data_time: 0.0018 memory: 15585 loss: 0.0106 +2023/06/05 03:59:52 - mmengine - INFO - Epoch(train) [3][1800/2056] lr: 3.2572e-04 eta: 3:03:37 time: 0.7029 data_time: 0.0018 memory: 15585 loss: 0.0142 +2023/06/05 04:00:57 - mmengine - INFO - Exp name: convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1_20230605_024312 +2023/06/05 04:01:05 - mmengine - INFO - Epoch(train) [3][1900/2056] lr: 3.2337e-04 eta: 3:02:19 time: 0.7051 data_time: 0.0021 memory: 15585 loss: 0.0174 +2023/06/05 04:02:19 - mmengine - INFO - Epoch(train) [3][2000/2056] lr: 3.2099e-04 eta: 3:00:59 time: 0.7326 data_time: 0.0017 memory: 15585 loss: 0.0158 +2023/06/05 04:02:59 - mmengine - INFO - Exp name: convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1_20230605_024312 +2023/06/05 04:02:59 - mmengine - INFO - Saving checkpoint at 3 epochs +2023/06/05 04:03:54 - mmengine - INFO - Epoch(val) [3][60/60] accuracy/top1: 98.2358 single-label/precision_classwise: [99.87118530273438, 92.73436737060547] single-label/recall_classwise: [97.88313293457031, 99.53488159179688] single-label/f1-score_classwise: [98.8671646118164, 96.01435852050781] data_time: 0.0908 time: 0.4854 +2023/06/05 04:05:14 - mmengine - INFO - Epoch(train) [4][ 100/2056] lr: 3.1722e-04 eta: 2:59:08 time: 0.6741 data_time: 0.0016 memory: 15585 loss: 0.0164 +2023/06/05 04:06:26 - mmengine - INFO - Epoch(train) [4][ 200/2056] lr: 3.1477e-04 eta: 2:57:46 time: 0.7042 data_time: 0.0017 memory: 15585 loss: 0.0148 +2023/06/05 04:07:37 - mmengine - INFO - Epoch(train) [4][ 300/2056] lr: 3.1229e-04 eta: 2:56:21 time: 0.6967 data_time: 0.0018 memory: 15585 loss: 0.0138 +2023/06/05 04:08:48 - mmengine - INFO - Epoch(train) [4][ 400/2056] lr: 3.0979e-04 eta: 2:54:58 time: 0.7047 data_time: 0.0017 memory: 15585 loss: 0.0160 +2023/06/05 04:10:04 - mmengine - INFO - Epoch(train) [4][ 500/2056] lr: 3.0727e-04 eta: 2:53:45 time: 1.0374 data_time: 0.0018 memory: 15585 loss: 0.0213 +2023/06/05 04:11:23 - mmengine - INFO - Epoch(train) [4][ 600/2056] lr: 3.0472e-04 eta: 2:52:38 time: 1.1711 data_time: 0.0017 memory: 15585 loss: 0.0165 +2023/06/05 04:12:37 - mmengine - INFO - Epoch(train) [4][ 700/2056] lr: 3.0215e-04 eta: 2:51:20 time: 0.7086 data_time: 0.0016 memory: 15585 loss: 0.0179 +2023/06/05 04:13:49 - mmengine - INFO - Epoch(train) [4][ 800/2056] lr: 2.9955e-04 eta: 2:49:58 time: 0.6495 data_time: 0.0018 memory: 15585 loss: 0.0180 +2023/06/05 04:14:15 - mmengine - INFO - Exp name: convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1_20230605_024312 +2023/06/05 04:15:02 - mmengine - INFO - Epoch(train) [4][ 900/2056] lr: 2.9694e-04 eta: 2:48:39 time: 0.6842 data_time: 0.0018 memory: 15585 loss: 0.0100 +2023/06/05 04:16:16 - mmengine - INFO - Epoch(train) [4][1000/2056] lr: 2.9430e-04 eta: 2:47:23 time: 0.7165 data_time: 0.0017 memory: 15585 loss: 0.0150 +2023/06/05 04:17:30 - mmengine - INFO - Epoch(train) [4][1100/2056] lr: 2.9164e-04 eta: 2:46:06 time: 0.7719 data_time: 0.0017 memory: 15585 loss: 0.0150 +2023/06/05 04:18:44 - mmengine - INFO - Epoch(train) [4][1200/2056] lr: 2.8896e-04 eta: 2:44:49 time: 0.6913 data_time: 0.0019 memory: 15585 loss: 0.0131 +2023/06/05 04:19:58 - mmengine - INFO - Epoch(train) [4][1300/2056] lr: 2.8626e-04 eta: 2:43:33 time: 0.7019 data_time: 0.0017 memory: 15585 loss: 0.0114 +2023/06/05 04:21:12 - mmengine - INFO - Epoch(train) [4][1400/2056] lr: 2.8354e-04 eta: 2:42:16 time: 0.7164 data_time: 0.0015 memory: 15585 loss: 0.0117 +2023/06/05 04:22:26 - mmengine - INFO - Epoch(train) [4][1500/2056] lr: 2.8081e-04 eta: 2:40:59 time: 0.7726 data_time: 0.0017 memory: 15585 loss: 0.0108 +2023/06/05 04:23:37 - mmengine - INFO - Epoch(train) [4][1600/2056] lr: 2.7805e-04 eta: 2:39:38 time: 0.6858 data_time: 0.0015 memory: 15585 loss: 0.0195 +2023/06/05 04:24:50 - mmengine - INFO - Epoch(train) [4][1700/2056] lr: 2.7528e-04 eta: 2:38:21 time: 0.6736 data_time: 0.0016 memory: 15585 loss: 0.0167 +2023/06/05 04:27:38 - mmengine - INFO - Epoch(train) [4][1800/2056] lr: 2.7249e-04 eta: 2:39:32 time: 10.0168 data_time: 0.0017 memory: 15585 loss: 0.0075 +2023/06/05 04:28:09 - mmengine - INFO - Exp name: convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1_20230605_024312 +2023/06/05 04:28:57 - mmengine - INFO - Epoch(train) [4][1900/2056] lr: 2.6969e-04 eta: 2:38:21 time: 0.7568 data_time: 0.0019 memory: 15585 loss: 0.0099 +2023/06/05 04:30:10 - mmengine - INFO - Epoch(train) [4][2000/2056] lr: 2.6687e-04 eta: 2:37:00 time: 0.7775 data_time: 0.0019 memory: 15585 loss: 0.0210 +2023/06/05 04:30:51 - mmengine - INFO - Exp name: convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1_20230605_024312 +2023/06/05 04:30:51 - mmengine - INFO - Saving checkpoint at 4 epochs +2023/06/05 04:31:47 - mmengine - INFO - Epoch(val) [4][60/60] accuracy/top1: 98.9772 single-label/precision_classwise: [99.93186950683594, 95.64441680908203] single-label/recall_classwise: [98.76691436767578, 99.75193786621094] single-label/f1-score_classwise: [99.3459701538086, 97.65499877929688] data_time: 0.0911 time: 0.4895 +2023/06/05 04:33:04 - mmengine - INFO - Epoch(train) [5][ 100/2056] lr: 2.6245e-04 eta: 2:35:01 time: 0.7086 data_time: 0.0016 memory: 15585 loss: 0.0101 +2023/06/05 04:34:21 - mmengine - INFO - Epoch(train) [5][ 200/2056] lr: 2.5959e-04 eta: 2:33:46 time: 0.7159 data_time: 0.0015 memory: 15585 loss: 0.0099 +2023/06/05 04:35:31 - mmengine - INFO - Epoch(train) [5][ 300/2056] lr: 2.5673e-04 eta: 2:32:23 time: 0.6373 data_time: 0.0016 memory: 15585 loss: 0.0095 +2023/06/05 04:36:42 - mmengine - INFO - Epoch(train) [5][ 400/2056] lr: 2.5385e-04 eta: 2:31:00 time: 0.6685 data_time: 0.0015 memory: 15585 loss: 0.0178 +2023/06/05 04:37:53 - mmengine - INFO - Epoch(train) [5][ 500/2056] lr: 2.5096e-04 eta: 2:29:37 time: 0.6963 data_time: 0.0017 memory: 15585 loss: 0.0085 +2023/06/05 04:39:07 - mmengine - INFO - Epoch(train) [5][ 600/2056] lr: 2.4806e-04 eta: 2:28:19 time: 0.7227 data_time: 0.0019 memory: 15585 loss: 0.0159 +2023/06/05 04:40:21 - mmengine - INFO - Epoch(train) [5][ 700/2056] lr: 2.4514e-04 eta: 2:27:01 time: 0.7444 data_time: 0.0016 memory: 15585 loss: 0.0161 +2023/06/05 04:41:15 - mmengine - INFO - Exp name: convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1_20230605_024312 +2023/06/05 04:41:32 - mmengine - INFO - Epoch(train) [5][ 800/2056] lr: 2.4222e-04 eta: 2:25:39 time: 0.7544 data_time: 0.0015 memory: 15585 loss: 0.0059 +2023/06/05 04:42:43 - mmengine - INFO - Epoch(train) [5][ 900/2056] lr: 2.3929e-04 eta: 2:24:17 time: 0.7046 data_time: 0.0017 memory: 15585 loss: 0.0187 +2023/06/05 04:43:57 - mmengine - INFO - Epoch(train) [5][1000/2056] lr: 2.3636e-04 eta: 2:22:58 time: 0.7288 data_time: 0.0018 memory: 15585 loss: 0.0144 +2023/06/05 04:45:08 - mmengine - INFO - Epoch(train) [5][1100/2056] lr: 2.3341e-04 eta: 2:21:38 time: 0.6843 data_time: 0.0017 memory: 15585 loss: 0.0141 +2023/06/05 04:46:23 - mmengine - INFO - Epoch(train) [5][1200/2056] lr: 2.3046e-04 eta: 2:20:21 time: 0.7770 data_time: 0.0016 memory: 15585 loss: 0.0092 +2023/06/05 04:47:37 - mmengine - INFO - Epoch(train) [5][1300/2056] lr: 2.2751e-04 eta: 2:19:03 time: 0.7770 data_time: 0.0017 memory: 15585 loss: 0.0098 +2023/06/05 04:48:47 - mmengine - INFO - Epoch(train) [5][1400/2056] lr: 2.2454e-04 eta: 2:17:42 time: 0.7560 data_time: 0.0016 memory: 15585 loss: 0.0117 +2023/06/05 04:49:59 - mmengine - INFO - Epoch(train) [5][1500/2056] lr: 2.2158e-04 eta: 2:16:22 time: 0.7189 data_time: 0.0015 memory: 15585 loss: 0.0071 +2023/06/05 04:51:11 - mmengine - INFO - Epoch(train) [5][1600/2056] lr: 2.1861e-04 eta: 2:15:03 time: 0.7445 data_time: 0.0016 memory: 15585 loss: 0.0106 +2023/06/05 04:52:24 - mmengine - INFO - Epoch(train) [5][1700/2056] lr: 2.1563e-04 eta: 2:13:44 time: 0.6917 data_time: 0.0019 memory: 15585 loss: 0.0183 +2023/06/05 04:53:18 - mmengine - INFO - Exp name: convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1_20230605_024312 +2023/06/05 04:53:34 - mmengine - INFO - Epoch(train) [5][1800/2056] lr: 2.1266e-04 eta: 2:12:24 time: 0.6804 data_time: 0.0017 memory: 15585 loss: 0.0080 +2023/06/05 04:54:47 - mmengine - INFO - Epoch(train) [5][1900/2056] lr: 2.0968e-04 eta: 2:11:06 time: 0.7282 data_time: 0.0017 memory: 15585 loss: 0.0080 +2023/06/05 04:55:59 - mmengine - INFO - Epoch(train) [5][2000/2056] lr: 2.0670e-04 eta: 2:09:46 time: 0.7011 data_time: 0.0018 memory: 15585 loss: 0.0093 +2023/06/05 04:56:39 - mmengine - INFO - Exp name: convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1_20230605_024312 +2023/06/05 04:56:39 - mmengine - INFO - Saving checkpoint at 5 epochs +2023/06/05 04:57:36 - mmengine - INFO - Epoch(val) [5][60/60] accuracy/top1: 99.2718 single-label/precision_classwise: [99.94908142089844, 96.87030029296875] single-label/recall_classwise: [99.1246337890625, 99.81395721435547] single-label/f1-score_classwise: [99.53514862060547, 98.3200912475586] data_time: 0.0875 time: 0.4827 +2023/06/05 04:58:52 - mmengine - INFO - Epoch(train) [6][ 100/2056] lr: 2.0205e-04 eta: 2:07:47 time: 0.6573 data_time: 0.1427 memory: 15585 loss: 0.0078 +2023/06/05 05:00:03 - mmengine - INFO - Epoch(train) [6][ 200/2056] lr: 1.9907e-04 eta: 2:06:28 time: 0.6651 data_time: 0.0019 memory: 15585 loss: 0.0120 +2023/06/05 05:01:16 - mmengine - INFO - Epoch(train) [6][ 300/2056] lr: 1.9609e-04 eta: 2:05:10 time: 0.6592 data_time: 0.0017 memory: 15585 loss: 0.0103 +2023/06/05 05:02:29 - mmengine - INFO - Epoch(train) [6][ 400/2056] lr: 1.9312e-04 eta: 2:03:53 time: 0.6693 data_time: 0.0017 memory: 15585 loss: 0.0133 +2023/06/05 05:03:40 - mmengine - INFO - Epoch(train) [6][ 500/2056] lr: 1.9015e-04 eta: 2:02:34 time: 0.7153 data_time: 0.0018 memory: 15585 loss: 0.0079 +2023/06/05 05:04:52 - mmengine - INFO - Epoch(train) [6][ 600/2056] lr: 1.8718e-04 eta: 2:01:16 time: 0.7191 data_time: 0.0018 memory: 15585 loss: 0.0145 +2023/06/05 05:06:03 - mmengine - INFO - Epoch(train) [6][ 700/2056] lr: 1.8421e-04 eta: 1:59:57 time: 0.7072 data_time: 0.0018 memory: 15585 loss: 0.0086 +2023/06/05 05:06:18 - mmengine - INFO - Exp name: convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1_20230605_024312 +2023/06/05 05:07:16 - mmengine - INFO - Epoch(train) [6][ 800/2056] lr: 1.8125e-04 eta: 1:58:40 time: 0.6682 data_time: 0.0018 memory: 15585 loss: 0.0137 +2023/06/05 05:08:28 - mmengine - INFO - Epoch(train) [6][ 900/2056] lr: 1.7830e-04 eta: 1:57:22 time: 0.7239 data_time: 0.0019 memory: 15585 loss: 0.0055 +2023/06/05 05:09:37 - mmengine - INFO - Epoch(train) [6][1000/2056] lr: 1.7535e-04 eta: 1:56:03 time: 0.7384 data_time: 0.0016 memory: 15585 loss: 0.0075 +2023/06/05 05:10:46 - mmengine - INFO - Epoch(train) [6][1100/2056] lr: 1.7241e-04 eta: 1:54:42 time: 0.6851 data_time: 0.0021 memory: 15585 loss: 0.0084 +2023/06/05 05:11:55 - mmengine - INFO - Epoch(train) [6][1200/2056] lr: 1.6947e-04 eta: 1:53:23 time: 0.6774 data_time: 0.0017 memory: 15585 loss: 0.0068 +2023/06/05 05:13:04 - mmengine - INFO - Epoch(train) [6][1300/2056] lr: 1.6655e-04 eta: 1:52:03 time: 0.7411 data_time: 0.0590 memory: 15585 loss: 0.0122 +2023/06/05 05:14:14 - mmengine - INFO - Epoch(train) [6][1400/2056] lr: 1.6363e-04 eta: 1:50:45 time: 0.6791 data_time: 0.0017 memory: 15585 loss: 0.0083 +2023/06/05 05:15:27 - mmengine - INFO - Epoch(train) [6][1500/2056] lr: 1.6072e-04 eta: 1:49:28 time: 0.7193 data_time: 0.0017 memory: 15585 loss: 0.0076 +2023/06/05 05:16:38 - mmengine - INFO - Epoch(train) [6][1600/2056] lr: 1.5783e-04 eta: 1:48:11 time: 0.7224 data_time: 0.1958 memory: 15585 loss: 0.0125 +2023/06/05 05:17:51 - mmengine - INFO - Epoch(train) [6][1700/2056] lr: 1.5494e-04 eta: 1:46:54 time: 0.7076 data_time: 0.1786 memory: 15585 loss: 0.0075 +2023/06/05 05:18:05 - mmengine - INFO - Exp name: convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1_20230605_024312 +2023/06/05 05:19:01 - mmengine - INFO - Epoch(train) [6][1800/2056] lr: 1.5207e-04 eta: 1:45:37 time: 0.7575 data_time: 0.2287 memory: 15585 loss: 0.0121 +2023/06/05 05:20:12 - mmengine - INFO - Epoch(train) [6][1900/2056] lr: 1.4921e-04 eta: 1:44:19 time: 0.7313 data_time: 0.2036 memory: 15585 loss: 0.0073 +2023/06/05 05:21:23 - mmengine - INFO - Epoch(train) [6][2000/2056] lr: 1.4636e-04 eta: 1:43:02 time: 0.6938 data_time: 0.1664 memory: 15585 loss: 0.0115 +2023/06/05 05:22:00 - mmengine - INFO - Exp name: convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1_20230605_024312 +2023/06/05 05:22:00 - mmengine - INFO - Saving checkpoint at 6 epochs +2023/06/05 05:22:54 - mmengine - INFO - Epoch(val) [6][60/60] accuracy/top1: 99.4770 single-label/precision_classwise: [99.94709777832031, 97.78976440429688] single-label/recall_classwise: [99.38766479492188, 99.80619812011719] single-label/f1-score_classwise: [99.66659545898438, 98.78768920898438] data_time: 0.0809 time: 0.4790 +2023/06/05 05:24:10 - mmengine - INFO - Epoch(train) [7][ 100/2056] lr: 1.4194e-04 eta: 1:41:03 time: 0.6390 data_time: 0.1207 memory: 15585 loss: 0.0086 +2023/06/05 05:25:21 - mmengine - INFO - Epoch(train) [7][ 200/2056] lr: 1.3913e-04 eta: 1:39:46 time: 0.7095 data_time: 0.1933 memory: 15585 loss: 0.0065 +2023/06/05 05:26:33 - mmengine - INFO - Epoch(train) [7][ 300/2056] lr: 1.3633e-04 eta: 1:38:30 time: 0.7017 data_time: 0.1832 memory: 15585 loss: 0.0128 +2023/06/05 05:27:44 - mmengine - INFO - Epoch(train) [7][ 400/2056] lr: 1.3355e-04 eta: 1:37:13 time: 0.6942 data_time: 0.1770 memory: 15585 loss: 0.0083 +2023/06/05 05:28:55 - mmengine - INFO - Epoch(train) [7][ 500/2056] lr: 1.3079e-04 eta: 1:35:57 time: 0.6421 data_time: 0.1284 memory: 15585 loss: 0.0059 +2023/06/05 05:30:05 - mmengine - INFO - Epoch(train) [7][ 600/2056] lr: 1.2804e-04 eta: 1:34:40 time: 0.6475 data_time: 0.1340 memory: 15585 loss: 0.0063 +2023/06/05 05:30:52 - mmengine - INFO - Exp name: convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1_20230605_024312 +2023/06/05 05:31:17 - mmengine - INFO - Epoch(train) [7][ 700/2056] lr: 1.2531e-04 eta: 1:33:24 time: 0.7181 data_time: 0.1922 memory: 15585 loss: 0.0048 +2023/06/05 05:32:28 - mmengine - INFO - Epoch(train) [7][ 800/2056] lr: 1.2260e-04 eta: 1:32:07 time: 0.7137 data_time: 0.1985 memory: 15585 loss: 0.0072 +2023/06/05 05:33:39 - mmengine - INFO - Epoch(train) [7][ 900/2056] lr: 1.1991e-04 eta: 1:30:51 time: 0.7107 data_time: 0.1928 memory: 15585 loss: 0.0101 +2023/06/05 05:34:49 - mmengine - INFO - Epoch(train) [7][1000/2056] lr: 1.1724e-04 eta: 1:29:34 time: 0.6920 data_time: 0.1783 memory: 15585 loss: 0.0077 +2023/06/05 05:36:01 - mmengine - INFO - Epoch(train) [7][1100/2056] lr: 1.1459e-04 eta: 1:28:18 time: 0.6591 data_time: 0.1450 memory: 15585 loss: 0.0056 +2023/06/05 05:37:11 - mmengine - INFO - Epoch(train) [7][1200/2056] lr: 1.1196e-04 eta: 1:27:01 time: 0.7452 data_time: 0.2258 memory: 15585 loss: 0.0047 +2023/06/05 05:38:21 - mmengine - INFO - Epoch(train) [7][1300/2056] lr: 1.0936e-04 eta: 1:25:45 time: 0.6749 data_time: 0.1603 memory: 15585 loss: 0.0108 +2023/06/05 05:39:30 - mmengine - INFO - Epoch(train) [7][1400/2056] lr: 1.0677e-04 eta: 1:24:28 time: 0.7285 data_time: 0.2096 memory: 15585 loss: 0.0072 +2023/06/05 05:40:41 - mmengine - INFO - Epoch(train) [7][1500/2056] lr: 1.0421e-04 eta: 1:23:12 time: 0.7457 data_time: 0.2261 memory: 15585 loss: 0.0107 +2023/06/05 05:41:51 - mmengine - INFO - Epoch(train) [7][1600/2056] lr: 1.0167e-04 eta: 1:21:56 time: 0.7211 data_time: 0.2004 memory: 15585 loss: 0.0073 +2023/06/05 05:42:39 - mmengine - INFO - Exp name: convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1_20230605_024312 +2023/06/05 05:43:04 - mmengine - INFO - Epoch(train) [7][1700/2056] lr: 9.9154e-05 eta: 1:20:41 time: 0.7009 data_time: 0.0196 memory: 15585 loss: 0.0030 +2023/06/05 05:44:17 - mmengine - INFO - Epoch(train) [7][1800/2056] lr: 9.6664e-05 eta: 1:19:26 time: 0.7151 data_time: 0.0017 memory: 15585 loss: 0.0071 +2023/06/05 05:45:29 - mmengine - INFO - Epoch(train) [7][1900/2056] lr: 9.4199e-05 eta: 1:18:11 time: 0.6945 data_time: 0.0016 memory: 15585 loss: 0.0077 +2023/06/05 05:46:45 - mmengine - INFO - Epoch(train) [7][2000/2056] lr: 9.1760e-05 eta: 1:16:58 time: 0.7431 data_time: 0.0017 memory: 15585 loss: 0.0048 +2023/06/05 05:47:25 - mmengine - INFO - Exp name: convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1_20230605_024312 +2023/06/05 05:47:25 - mmengine - INFO - Saving checkpoint at 7 epochs +2023/06/05 05:48:23 - mmengine - INFO - Epoch(val) [7][60/60] accuracy/top1: 99.5945 single-label/precision_classwise: [99.95984649658203, 98.2757339477539] single-label/recall_classwise: [99.5244369506836, 99.85271453857422] single-label/f1-score_classwise: [99.74166107177734, 99.05794525146484] data_time: 0.0965 time: 0.4932 +2023/06/05 05:49:40 - mmengine - INFO - Epoch(train) [8][ 100/2056] lr: 8.8009e-05 eta: 1:15:02 time: 0.7576 data_time: 0.0016 memory: 15585 loss: 0.0141 +2023/06/05 05:50:53 - mmengine - INFO - Epoch(train) [8][ 200/2056] lr: 8.5639e-05 eta: 1:13:48 time: 0.6687 data_time: 0.0016 memory: 15585 loss: 0.0069 +2023/06/05 05:52:06 - mmengine - INFO - Epoch(train) [8][ 300/2056] lr: 8.3296e-05 eta: 1:12:33 time: 0.7806 data_time: 0.0016 memory: 15585 loss: 0.0034 +2023/06/05 05:53:19 - mmengine - INFO - Epoch(train) [8][ 400/2056] lr: 8.0983e-05 eta: 1:11:18 time: 0.6729 data_time: 0.0015 memory: 15585 loss: 0.0086 +2023/06/05 05:54:30 - mmengine - INFO - Epoch(train) [8][ 500/2056] lr: 7.8698e-05 eta: 1:10:03 time: 0.6439 data_time: 0.0015 memory: 15585 loss: 0.0089 +2023/06/05 05:55:41 - mmengine - INFO - Epoch(train) [8][ 600/2056] lr: 7.6443e-05 eta: 1:08:48 time: 0.6902 data_time: 0.0022 memory: 15585 loss: 0.0045 +2023/06/05 05:55:48 - mmengine - INFO - Exp name: convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1_20230605_024312 +2023/06/05 05:56:54 - mmengine - INFO - Epoch(train) [8][ 700/2056] lr: 7.4217e-05 eta: 1:07:33 time: 0.6785 data_time: 0.0017 memory: 15585 loss: 0.0083 +2023/06/05 05:58:08 - mmengine - INFO - Epoch(train) [8][ 800/2056] lr: 7.2022e-05 eta: 1:06:19 time: 0.7329 data_time: 0.0018 memory: 15585 loss: 0.0072 +2023/06/05 05:59:19 - mmengine - INFO - Epoch(train) [8][ 900/2056] lr: 6.9859e-05 eta: 1:05:03 time: 0.6910 data_time: 0.0019 memory: 15585 loss: 0.0079 +2023/06/05 06:00:31 - mmengine - INFO - Epoch(train) [8][1000/2056] lr: 6.7727e-05 eta: 1:03:49 time: 0.7238 data_time: 0.0018 memory: 15585 loss: 0.0041 +2023/06/05 06:01:45 - mmengine - INFO - Epoch(train) [8][1100/2056] lr: 6.5626e-05 eta: 1:02:35 time: 0.7481 data_time: 0.0019 memory: 15585 loss: 0.0037 +2023/06/05 06:02:58 - mmengine - INFO - Epoch(train) [8][1200/2056] lr: 6.3559e-05 eta: 1:01:20 time: 0.7303 data_time: 0.0017 memory: 15585 loss: 0.0046 +2023/06/05 06:04:12 - mmengine - INFO - Epoch(train) [8][1300/2056] lr: 6.1524e-05 eta: 1:00:06 time: 0.7296 data_time: 0.0018 memory: 15585 loss: 0.0043 +2023/06/05 06:05:23 - mmengine - INFO - Epoch(train) [8][1400/2056] lr: 5.9523e-05 eta: 0:58:51 time: 0.7323 data_time: 0.0016 memory: 15585 loss: 0.0062 +2023/06/05 06:06:33 - mmengine - INFO - Epoch(train) [8][1500/2056] lr: 5.7556e-05 eta: 0:57:36 time: 0.7765 data_time: 0.0017 memory: 15585 loss: 0.0041 +2023/06/05 06:07:43 - mmengine - INFO - Epoch(train) [8][1600/2056] lr: 5.5623e-05 eta: 0:56:21 time: 0.6560 data_time: 0.0018 memory: 15585 loss: 0.0062 +2023/06/05 06:07:49 - mmengine - INFO - Exp name: convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1_20230605_024312 +2023/06/05 06:08:56 - mmengine - INFO - Epoch(train) [8][1700/2056] lr: 5.3726e-05 eta: 0:55:06 time: 0.6898 data_time: 0.0017 memory: 15585 loss: 0.0065 +2023/06/05 06:10:04 - mmengine - INFO - Epoch(train) [8][1800/2056] lr: 5.1863e-05 eta: 0:53:51 time: 0.6894 data_time: 0.0017 memory: 15585 loss: 0.0092 +2023/06/05 06:11:14 - mmengine - INFO - Epoch(train) [8][1900/2056] lr: 5.0036e-05 eta: 0:52:36 time: 0.7110 data_time: 0.0016 memory: 15585 loss: 0.0081 +2023/06/05 06:12:22 - mmengine - INFO - Epoch(train) [8][2000/2056] lr: 4.8246e-05 eta: 0:51:20 time: 0.6360 data_time: 0.0017 memory: 15585 loss: 0.0029 +2023/06/05 06:13:02 - mmengine - INFO - Exp name: convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1_20230605_024312 +2023/06/05 06:13:02 - mmengine - INFO - Saving checkpoint at 8 epochs +2023/06/05 06:13:57 - mmengine - INFO - Epoch(val) [8][60/60] accuracy/top1: 99.6624 single-label/precision_classwise: [99.97676849365234, 98.52468872070312] single-label/recall_classwise: [99.59388732910156, 99.91472625732422] single-label/f1-score_classwise: [99.78496551513672, 99.21483612060547] data_time: 0.0967 time: 0.4956 +2023/06/05 06:15:19 - mmengine - INFO - Epoch(train) [9][ 100/2056] lr: 4.5526e-05 eta: 0:49:26 time: 0.6787 data_time: 0.1645 memory: 15585 loss: 0.0039 +2023/06/05 06:16:33 - mmengine - INFO - Epoch(train) [9][ 200/2056] lr: 4.3830e-05 eta: 0:48:12 time: 0.6841 data_time: 0.1662 memory: 15585 loss: 0.0063 +2023/06/05 06:17:46 - mmengine - INFO - Epoch(train) [9][ 300/2056] lr: 4.2172e-05 eta: 0:46:58 time: 0.6964 data_time: 0.1792 memory: 15585 loss: 0.0053 +2023/06/05 06:19:03 - mmengine - INFO - Epoch(train) [9][ 400/2056] lr: 4.0551e-05 eta: 0:45:45 time: 0.6942 data_time: 0.1786 memory: 15585 loss: 0.0092 +2023/06/05 06:20:15 - mmengine - INFO - Epoch(train) [9][ 500/2056] lr: 3.8969e-05 eta: 0:44:31 time: 0.7024 data_time: 0.1854 memory: 15585 loss: 0.0046 +2023/06/05 06:20:53 - mmengine - INFO - Exp name: convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1_20230605_024312 +2023/06/05 06:21:26 - mmengine - INFO - Epoch(train) [9][ 600/2056] lr: 3.7426e-05 eta: 0:43:16 time: 0.7193 data_time: 0.2014 memory: 15585 loss: 0.0077 +2023/06/05 06:22:37 - mmengine - INFO - Epoch(train) [9][ 700/2056] lr: 3.5922e-05 eta: 0:42:01 time: 0.6989 data_time: 0.1831 memory: 15585 loss: 0.0082 +2023/06/05 06:23:48 - mmengine - INFO - Epoch(train) [9][ 800/2056] lr: 3.4457e-05 eta: 0:40:47 time: 0.7060 data_time: 0.1893 memory: 15585 loss: 0.0062 +2023/06/05 06:24:58 - mmengine - INFO - Epoch(train) [9][ 900/2056] lr: 3.3032e-05 eta: 0:39:32 time: 0.6924 data_time: 0.1768 memory: 15585 loss: 0.0068 +2023/06/05 06:26:08 - mmengine - INFO - Epoch(train) [9][1000/2056] lr: 3.1648e-05 eta: 0:38:18 time: 0.7288 data_time: 0.2100 memory: 15585 loss: 0.0053 +2023/06/05 06:27:17 - mmengine - INFO - Epoch(train) [9][1100/2056] lr: 3.0304e-05 eta: 0:37:03 time: 0.7111 data_time: 0.1953 memory: 15585 loss: 0.0046 +2023/06/05 06:28:29 - mmengine - INFO - Epoch(train) [9][1200/2056] lr: 2.9000e-05 eta: 0:35:49 time: 0.7567 data_time: 0.2194 memory: 15585 loss: 0.0085 +2023/06/05 06:29:40 - mmengine - INFO - Epoch(train) [9][1300/2056] lr: 2.7738e-05 eta: 0:34:35 time: 0.6878 data_time: 0.1740 memory: 15585 loss: 0.0028 +2023/06/05 06:30:50 - mmengine - INFO - Epoch(train) [9][1400/2056] lr: 2.6517e-05 eta: 0:33:20 time: 0.7098 data_time: 0.1879 memory: 15585 loss: 0.0073 +2023/06/05 06:32:03 - mmengine - INFO - Epoch(train) [9][1500/2056] lr: 2.5338e-05 eta: 0:32:06 time: 0.7443 data_time: 0.2301 memory: 15585 loss: 0.0054 +2023/06/05 06:32:40 - mmengine - INFO - Exp name: convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1_20230605_024312 +2023/06/05 06:33:14 - mmengine - INFO - Epoch(train) [9][1600/2056] lr: 2.4200e-05 eta: 0:30:52 time: 0.7058 data_time: 0.1921 memory: 15585 loss: 0.0068 +2023/06/05 06:34:26 - mmengine - INFO - Epoch(train) [9][1700/2056] lr: 2.3105e-05 eta: 0:29:38 time: 0.7712 data_time: 0.2537 memory: 15585 loss: 0.0063 +2023/06/05 06:35:38 - mmengine - INFO - Epoch(train) [9][1800/2056] lr: 2.2053e-05 eta: 0:28:24 time: 0.6791 data_time: 0.1619 memory: 15585 loss: 0.0043 +2023/06/05 06:36:48 - mmengine - INFO - Epoch(train) [9][1900/2056] lr: 2.1043e-05 eta: 0:27:10 time: 0.6678 data_time: 0.1542 memory: 15585 loss: 0.0050 +2023/06/05 06:38:03 - mmengine - INFO - Epoch(train) [9][2000/2056] lr: 2.0076e-05 eta: 0:25:57 time: 1.1885 data_time: 0.6659 memory: 15585 loss: 0.0088 +2023/06/05 06:38:45 - mmengine - INFO - Exp name: convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1_20230605_024312 +2023/06/05 06:38:45 - mmengine - INFO - Saving checkpoint at 9 epochs +2023/06/05 06:39:40 - mmengine - INFO - Epoch(val) [9][60/60] accuracy/top1: 99.7402 single-label/precision_classwise: [99.95992279052734, 98.94001007080078] single-label/recall_classwise: [99.70960998535156, 99.85271453857422] single-label/f1-score_classwise: [99.83460998535156, 99.39427185058594] data_time: 0.0835 time: 0.4795 +2023/06/05 06:40:56 - mmengine - INFO - Epoch(train) [10][ 100/2056] lr: 1.8654e-05 eta: 0:24:02 time: 0.6964 data_time: 0.1825 memory: 15585 loss: 0.0049 +2023/06/05 06:42:06 - mmengine - INFO - Epoch(train) [10][ 200/2056] lr: 1.7798e-05 eta: 0:22:48 time: 0.6968 data_time: 0.1789 memory: 15585 loss: 0.0037 +2023/06/05 06:43:18 - mmengine - INFO - Epoch(train) [10][ 300/2056] lr: 1.6985e-05 eta: 0:21:34 time: 0.7106 data_time: 0.1934 memory: 15585 loss: 0.0044 +2023/06/05 06:44:29 - mmengine - INFO - Epoch(train) [10][ 400/2056] lr: 1.6217e-05 eta: 0:20:20 time: 0.7314 data_time: 0.2154 memory: 15585 loss: 0.0030 +2023/06/05 06:45:39 - mmengine - INFO - Exp name: convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1_20230605_024312 +2023/06/05 06:45:41 - mmengine - INFO - Epoch(train) [10][ 500/2056] lr: 1.5493e-05 eta: 0:19:06 time: 0.7328 data_time: 0.2149 memory: 15585 loss: 0.0067 +2023/06/05 06:46:53 - mmengine - INFO - Epoch(train) [10][ 600/2056] lr: 1.4813e-05 eta: 0:17:52 time: 0.6693 data_time: 0.1554 memory: 15585 loss: 0.0061 +2023/06/05 06:48:04 - mmengine - INFO - Epoch(train) [10][ 700/2056] lr: 1.4177e-05 eta: 0:16:38 time: 0.7060 data_time: 0.1883 memory: 15585 loss: 0.0040 +2023/06/05 06:49:15 - mmengine - INFO - Epoch(train) [10][ 800/2056] lr: 1.3586e-05 eta: 0:15:25 time: 0.6780 data_time: 0.1643 memory: 15585 loss: 0.0039 +2023/06/05 06:50:27 - mmengine - INFO - Epoch(train) [10][ 900/2056] lr: 1.3039e-05 eta: 0:14:11 time: 0.7067 data_time: 0.1888 memory: 15585 loss: 0.0056 +2023/06/05 06:51:39 - mmengine - INFO - Epoch(train) [10][1000/2056] lr: 1.2538e-05 eta: 0:12:57 time: 0.7427 data_time: 0.2254 memory: 15585 loss: 0.0048 +2023/06/05 06:52:49 - mmengine - INFO - Epoch(train) [10][1100/2056] lr: 1.2081e-05 eta: 0:11:43 time: 0.7287 data_time: 0.2116 memory: 15585 loss: 0.0048 +2023/06/05 06:54:00 - mmengine - INFO - Epoch(train) [10][1200/2056] lr: 1.1670e-05 eta: 0:10:30 time: 0.7002 data_time: 0.1833 memory: 15585 loss: 0.0048 +2023/06/05 06:55:11 - mmengine - INFO - Epoch(train) [10][1300/2056] lr: 1.1303e-05 eta: 0:09:16 time: 0.7107 data_time: 0.1948 memory: 15585 loss: 0.0050 +2023/06/05 06:56:22 - mmengine - INFO - Epoch(train) [10][1400/2056] lr: 1.0982e-05 eta: 0:08:02 time: 0.7014 data_time: 0.1840 memory: 15585 loss: 0.0054 +2023/06/05 06:57:31 - mmengine - INFO - Exp name: convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1_20230605_024312 +2023/06/05 06:57:33 - mmengine - INFO - Epoch(train) [10][1500/2056] lr: 1.0706e-05 eta: 0:06:49 time: 0.7881 data_time: 0.2710 memory: 15585 loss: 0.0054 +2023/06/05 06:58:44 - mmengine - INFO - Epoch(train) [10][1600/2056] lr: 1.0475e-05 eta: 0:05:35 time: 0.6616 data_time: 0.1448 memory: 15585 loss: 0.0063 +2023/06/05 06:59:53 - mmengine - INFO - Epoch(train) [10][1700/2056] lr: 1.0290e-05 eta: 0:04:21 time: 0.6839 data_time: 0.1701 memory: 15585 loss: 0.0037 +2023/06/05 07:01:03 - mmengine - INFO - Epoch(train) [10][1800/2056] lr: 1.0150e-05 eta: 0:03:08 time: 0.7083 data_time: 0.1896 memory: 15585 loss: 0.0066 +2023/06/05 07:02:14 - mmengine - INFO - Epoch(train) [10][1900/2056] lr: 1.0056e-05 eta: 0:01:54 time: 0.7005 data_time: 0.1833 memory: 15585 loss: 0.0042 +2023/06/05 07:03:29 - mmengine - INFO - Epoch(train) [10][2000/2056] lr: 1.0007e-05 eta: 0:00:41 time: 0.6809 data_time: 0.1671 memory: 15585 loss: 0.0035 +2023/06/05 07:04:08 - mmengine - INFO - Exp name: convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1_20230605_024312 +2023/06/05 07:04:08 - mmengine - INFO - Saving checkpoint at 10 epochs +2023/06/05 07:05:04 - mmengine - INFO - Epoch(val) [10][60/60] accuracy/top1: 99.7600 single-label/precision_classwise: [99.94307708740234, 99.09168243408203] single-label/recall_classwise: [99.75170135498047, 99.79069519042969] single-label/f1-score_classwise: [99.84729766845703, 99.43995666503906] data_time: 0.0943 time: 0.4889 diff --git a/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1/20230605_024312/vis_data/20230605_024312.json b/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1/20230605_024312/vis_data/20230605_024312.json new file mode 100644 index 0000000000000000000000000000000000000000..8879f019d58e05add43b7c40e73d5fd92a24b907 --- /dev/null +++ b/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1/20230605_024312/vis_data/20230605_024312.json @@ -0,0 +1,210 @@ +{"lr": 0.00039997768893775784, "data_time": 0.2427647352218628, "loss": 0.0622634295374155, "time": 0.7631019592285156, "epoch": 1, "memory": 15587, "step": 100} +{"lr": 0.00039990985724436777, "data_time": 0.18695836067199706, "loss": 0.047248711064457896, "time": 0.8761033773422241, "epoch": 1, "memory": 15587, "step": 200} +{"lr": 0.00039979651848055537, "data_time": 0.0017068624496459962, "loss": 0.04533916190266609, "time": 0.619895601272583, "epoch": 1, "memory": 15587, "step": 300} +{"lr": 0.0003996376991083742, "data_time": 0.0016076803207397462, "loss": 0.03946216888725758, "time": 0.8002885818481446, "epoch": 1, "memory": 15587, "step": 400} +{"lr": 0.00039943343620857754, "data_time": 0.0015278339385986328, "loss": 0.03365235310047865, "time": 0.7466793298721314, "epoch": 1, "memory": 15587, "step": 500} +{"lr": 0.0003991837774719618, "data_time": 0.001695871353149414, "loss": 0.029732736758887768, "time": 0.7768775939941406, "epoch": 1, "memory": 15587, "step": 600} +{"lr": 0.00039888878118822876, "data_time": 0.0015054941177368164, "loss": 0.025014087837189437, "time": 0.7019802570343018, "epoch": 1, "memory": 15587, "step": 700} +{"lr": 0.0003985485162323803, "data_time": 0.0014446496963500977, "loss": 0.019376942422240973, "time": 0.7305632829666138, "epoch": 1, "memory": 15587, "step": 800} +{"lr": 0.0003981630620486352, "data_time": 0.001797652244567871, "loss": 0.027765481360256673, "time": 0.710541844367981, "epoch": 1, "memory": 15587, "step": 900} +{"lr": 0.0003977325086318812, "data_time": 0.0017927885055541992, "loss": 0.030781422276049852, "time": 0.7124827623367309, "epoch": 1, "memory": 15587, "step": 1000} +{"lr": 0.00039725695650666204, "data_time": 0.0015607357025146484, "loss": 0.019887801771983505, "time": 0.7369407653808594, "epoch": 1, "memory": 15587, "step": 1100} +{"lr": 0.0003967365167037102, "data_time": 0.0018869400024414062, "loss": 0.02237267531454563, "time": 0.7870367765426636, "epoch": 1, "memory": 15587, "step": 1200} +{"lr": 0.0003961713107340207, "data_time": 0.0019826889038085938, "loss": 0.01975932056084275, "time": 0.8032874345779419, "epoch": 1, "memory": 15587, "step": 1300} +{"lr": 0.0003955614705604829, "data_time": 0.0018016815185546875, "loss": 0.02405152888968587, "time": 0.7527009725570679, "epoch": 1, "memory": 15587, "step": 1400} +{"lr": 0.00039490713856706923, "data_time": 0.0017481088638305665, "loss": 0.02071991842240095, "time": 0.7369158744812012, "epoch": 1, "memory": 15587, "step": 1500} +{"lr": 0.00039420846752559095, "data_time": 0.0014647483825683595, "loss": 0.01888023940846324, "time": 0.7223272562026978, "epoch": 1, "memory": 15587, "step": 1600} +{"lr": 0.0003934656205600326, "data_time": 0.0015542268753051757, "loss": 0.019930011453107, "time": 0.7527501344680786, "epoch": 1, "memory": 15587, "step": 1700} +{"lr": 0.0003926787711084616, "data_time": 0.0015392780303955078, "loss": 0.025704196002334357, "time": 0.6902652263641358, "epoch": 1, "memory": 15587, "step": 1800} +{"lr": 0.0003918481028825385, "data_time": 0.002183270454406738, "loss": 0.028769377805292606, "time": 0.7484632015228272, "epoch": 1, "memory": 15587, "step": 1900} +{"lr": 0.0003909738098246218, "data_time": 0.001527857780456543, "loss": 0.020901503786444665, "time": 0.7452915906906128, "epoch": 1, "memory": 15587, "step": 2000} +{"accuracy/top1": 77.52677154541016, "data_time": 0.10747306744257609, "time": 0.5418768803278605, "step": 1} +{"lr": 0.0003895232896423554, "data_time": 0.0020531415939331055, "loss": 0.018552379263564945, "time": 0.744081974029541, "epoch": 2, "memory": 15587, "step": 2156} +{"lr": 0.00038853827118924355, "data_time": 0.0020484209060668947, "loss": 0.01960562001913786, "time": 0.7278545379638672, "epoch": 2, "memory": 15585, "step": 2256} +{"lr": 0.0003875104006754391, "data_time": 0.0018478155136108399, "loss": 0.013127893302589655, "time": 0.7977020740509033, "epoch": 2, "memory": 15585, "step": 2356} +{"lr": 0.00038643991808560204, "data_time": 0.0018390893936157226, "loss": 0.020102136302739382, "time": 0.6740249633789063, "epoch": 2, "memory": 15585, "step": 2456} +{"lr": 0.00038532707335335494, "data_time": 0.0018429517745971679, "loss": 0.013678281661123037, "time": 0.7428551435470581, "epoch": 2, "memory": 15585, "step": 2556} +{"lr": 0.00038417212630292743, "data_time": 0.0015572309494018555, "loss": 0.02540783192962408, "time": 0.7237333059310913, "epoch": 2, "memory": 15585, "step": 2656} +{"lr": 0.00038297534658849464, "data_time": 0.0014730215072631836, "loss": 0.012263161223381757, "time": 0.7169277191162109, "epoch": 2, "memory": 15585, "step": 2756} +{"lr": 0.00038173701363121816, "data_time": 0.0016788482666015626, "loss": 0.021818548487499357, "time": 0.7130337715148926, "epoch": 2, "memory": 15585, "step": 2856} +{"lr": 0.00038045741655400836, "data_time": 0.0018208980560302734, "loss": 0.020187804708257318, "time": 0.7371300458908081, "epoch": 2, "memory": 15585, "step": 2956} +{"lr": 0.00037913685411401933, "data_time": 0.0018122196197509766, "loss": 0.019882525131106377, "time": 0.7195440292358398, "epoch": 2, "memory": 15585, "step": 3056} +{"lr": 0.00037777563463289686, "data_time": 0.0016786575317382813, "loss": 0.02451257212087512, "time": 0.7288497686386108, "epoch": 2, "memory": 15585, "step": 3156} +{"lr": 0.0003763740759247917, "data_time": 0.0015790224075317382, "loss": 0.01938260458409786, "time": 0.6837079524993896, "epoch": 2, "memory": 15585, "step": 3256} +{"lr": 0.0003749325052221578, "data_time": 0.0016111135482788086, "loss": 0.01937106898985803, "time": 0.6827808141708374, "epoch": 2, "memory": 15585, "step": 3356} +{"lr": 0.00037345125909934995, "data_time": 0.0018295764923095704, "loss": 0.020618916163221, "time": 0.727691388130188, "epoch": 2, "memory": 15585, "step": 3456} +{"lr": 0.00037193068339404196, "data_time": 0.0017955303192138672, "loss": 0.019244227232411502, "time": 0.7252246618270874, "epoch": 2, "memory": 15585, "step": 3556} +{"lr": 0.0003703711331264808, "data_time": 0.0016932725906372071, "loss": 0.020278052147477865, "time": 0.6932339906692505, "epoch": 2, "memory": 15585, "step": 3656} +{"lr": 0.0003687729724165983, "data_time": 0.0017278194427490234, "loss": 0.01249537207186222, "time": 0.6835992813110352, "epoch": 2, "memory": 15585, "step": 3756} +{"lr": 0.000367136574398996, "data_time": 0.001810908317565918, "loss": 0.021494934195652605, "time": 0.6877713680267334, "epoch": 2, "memory": 15585, "step": 3856} +{"lr": 0.00036546232113582794, "data_time": 0.0018505811691284179, "loss": 0.01864777714945376, "time": 0.7289227247238159, "epoch": 2, "memory": 15585, "step": 3956} +{"lr": 0.0003637506035275962, "data_time": 0.002080535888671875, "loss": 0.017337575368583203, "time": 0.7337353229522705, "epoch": 2, "memory": 15585, "step": 4056} +{"accuracy/top1": 95.22533416748047, "data_time": 0.09835075159541896, "time": 0.4946637935325748, "step": 2} +{"lr": 0.00036100646510740043, "data_time": 0.10377991199493408, "loss": 0.016776063246652484, "time": 0.7853656768798828, "epoch": 3, "memory": 15585, "step": 4212} +{"lr": 0.0003592006803637847, "data_time": 0.001594686508178711, "loss": 0.01856973674148321, "time": 0.7196727514266967, "epoch": 3, "memory": 15585, "step": 4312} +{"lr": 0.00035735889322747414, "data_time": 0.0017902612686157226, "loss": 0.010173213155940174, "time": 0.7030227422714234, "epoch": 3, "memory": 15585, "step": 4412} +{"lr": 0.0003554815337143657, "data_time": 0.001588726043701172, "loss": 0.01909800935536623, "time": 0.7280056238174438, "epoch": 3, "memory": 15585, "step": 4512} +{"lr": 0.00035356904014570526, "data_time": 0.0017295122146606446, "loss": 0.014161157002672554, "time": 0.6520989656448364, "epoch": 3, "memory": 15585, "step": 4612} +{"lr": 0.00035162185904575157, "data_time": 0.0017443418502807618, "loss": 0.01739039127714932, "time": 0.7259718418121338, "epoch": 3, "memory": 15585, "step": 4712} +{"lr": 0.0003496404450375223, "data_time": 0.0019626379013061523, "loss": 0.01800237870775163, "time": 0.7121296167373657, "epoch": 3, "memory": 15585, "step": 4812} +{"lr": 0.0003476252607366498, "data_time": 0.0016444921493530273, "loss": 0.012891631200909614, "time": 0.6917529344558716, "epoch": 3, "memory": 15585, "step": 4912} +{"lr": 0.0003455767766433698, "data_time": 0.0015131235122680664, "loss": 0.018378550466150046, "time": 0.7705854415893555, "epoch": 3, "memory": 15585, "step": 5012} +{"lr": 0.0003434954710326717, "data_time": 0.0016249895095825195, "loss": 0.018253946397453546, "time": 0.7216732501983643, "epoch": 3, "memory": 15585, "step": 5112} +{"lr": 0.00034138182984263203, "data_time": 0.0019607782363891602, "loss": 0.01774289016611874, "time": 0.7416847467422485, "epoch": 3, "memory": 15585, "step": 5212} +{"lr": 0.0003392363465609565, "data_time": 0.0021759748458862306, "loss": 0.012799293221905827, "time": 0.7296148538589478, "epoch": 3, "memory": 15585, "step": 5312} +{"lr": 0.0003370595221097656, "data_time": 0.0018844366073608398, "loss": 0.011717123771086335, "time": 0.7495620965957641, "epoch": 3, "memory": 15585, "step": 5412} +{"lr": 0.00033485186472863657, "data_time": 0.0017618656158447266, "loss": 0.01528772497549653, "time": 0.7832259416580201, "epoch": 3, "memory": 15585, "step": 5512} +{"lr": 0.00033261388985594397, "data_time": 0.001717352867126465, "loss": 0.01808881820179522, "time": 0.7567421436309815, "epoch": 3, "memory": 15585, "step": 5612} +{"lr": 0.0003303461200085145, "data_time": 0.0015334367752075195, "loss": 0.011359138414263725, "time": 0.8013696193695068, "epoch": 3, "memory": 15585, "step": 5712} +{"lr": 0.0003280490846596322, "data_time": 0.001840519905090332, "loss": 0.010607956117019058, "time": 0.7803977012634278, "epoch": 3, "memory": 15585, "step": 5812} +{"lr": 0.00032572332011541707, "data_time": 0.0018265247344970703, "loss": 0.014171441504731774, "time": 0.7029386997222901, "epoch": 3, "memory": 15585, "step": 5912} +{"lr": 0.00032336936938961124, "data_time": 0.002108311653137207, "loss": 0.017399783618748187, "time": 0.7051195621490478, "epoch": 3, "memory": 15585, "step": 6012} +{"lr": 0.00032098778207679645, "data_time": 0.0016778945922851563, "loss": 0.01578828217461705, "time": 0.7325539112091064, "epoch": 3, "memory": 15585, "step": 6112} +{"accuracy/top1": 98.23577117919922, "data_time": 0.09079766273498535, "time": 0.485427731373271, "step": 3} +{"lr": 0.0003172186408067781, "data_time": 0.0015857219696044922, "loss": 0.016427195025607945, "time": 0.6740639925003051, "epoch": 4, "memory": 15585, "step": 6268} +{"lr": 0.00031476885256500535, "data_time": 0.001706075668334961, "loss": 0.014819549396634102, "time": 0.7042318344116211, "epoch": 4, "memory": 15585, "step": 6368} +{"lr": 0.00031229343576357664, "data_time": 0.0018445730209350586, "loss": 0.013786209328100086, "time": 0.6967251777648926, "epoch": 4, "memory": 15585, "step": 6468} +{"lr": 0.00030979296835667217, "data_time": 0.0017379045486450196, "loss": 0.015972312819212676, "time": 0.7047409772872925, "epoch": 4, "memory": 15585, "step": 6568} +{"lr": 0.000307268034147225, "data_time": 0.0017672538757324218, "loss": 0.02125865938141942, "time": 1.0374451398849487, "epoch": 4, "memory": 15585, "step": 6668} +{"lr": 0.00030471922265061713, "data_time": 0.0017290353775024415, "loss": 0.016504886886104943, "time": 1.171081256866455, "epoch": 4, "memory": 15585, "step": 6768} +{"lr": 0.00030214712895704036, "data_time": 0.001592707633972168, "loss": 0.017919448390603065, "time": 0.7085912466049195, "epoch": 4, "memory": 15585, "step": 6868} +{"lr": 0.00029955235359255527, "data_time": 0.0018253564834594727, "loss": 0.01797847356647253, "time": 0.6495321273803711, "epoch": 4, "memory": 15585, "step": 6968} +{"lr": 0.00029693550237888314, "data_time": 0.0018332242965698243, "loss": 0.010003110906109215, "time": 0.6841523170471191, "epoch": 4, "memory": 15585, "step": 7068} +{"lr": 0.0002942971862919601, "data_time": 0.0017138004302978515, "loss": 0.014989589992910623, "time": 0.7165249824523926, "epoch": 4, "memory": 15585, "step": 7168} +{"lr": 0.0002916380213192881, "data_time": 0.0016910552978515625, "loss": 0.014970519207417965, "time": 0.7719467639923095, "epoch": 4, "memory": 15585, "step": 7268} +{"lr": 0.00028895862831611433, "data_time": 0.001915311813354492, "loss": 0.013057804573327303, "time": 0.6912996053695679, "epoch": 4, "memory": 15585, "step": 7368} +{"lr": 0.0002862596328604766, "data_time": 0.0016709566116333008, "loss": 0.011388324014842511, "time": 0.7018782615661621, "epoch": 4, "memory": 15585, "step": 7468} +{"lr": 0.0002835416651071449, "data_time": 0.0015023708343505859, "loss": 0.01167175085283816, "time": 0.7163537025451661, "epoch": 4, "memory": 15585, "step": 7568} +{"lr": 0.00028080535964049476, "data_time": 0.0016505002975463867, "loss": 0.010760634113103152, "time": 0.7725522518157959, "epoch": 4, "memory": 15585, "step": 7668} +{"lr": 0.0002780513553263448, "data_time": 0.0015471935272216796, "loss": 0.019454200752079487, "time": 0.6858330011367798, "epoch": 4, "memory": 15585, "step": 7768} +{"lr": 0.0002752802951627973, "data_time": 0.0016010522842407227, "loss": 0.016746526444330813, "time": 0.6736471652984619, "epoch": 4, "memory": 15585, "step": 7868} +{"lr": 0.0002724928261301117, "data_time": 0.0016916275024414062, "loss": 0.007524235686287284, "time": 10.016830921173096, "epoch": 4, "memory": 15585, "step": 7968} +{"lr": 0.00026968959903964925, "data_time": 0.0019032955169677734, "loss": 0.009913359256461262, "time": 0.7568388223648072, "epoch": 4, "memory": 15585, "step": 8068} +{"lr": 0.0002668712683819236, "data_time": 0.0018506050109863281, "loss": 0.02095145294442773, "time": 0.777468228340149, "epoch": 4, "memory": 15585, "step": 8168} +{"accuracy/top1": 98.97721099853516, "data_time": 0.09110732547572402, "time": 0.48946628414216586, "step": 4} +{"lr": 0.00026244607407074216, "data_time": 0.0015883445739746094, "loss": 0.010050296876579523, "time": 0.7085708618164063, "epoch": 5, "memory": 15585, "step": 8324} +{"lr": 0.00025959208467259544, "data_time": 0.0014809131622314452, "loss": 0.009908492537215352, "time": 0.7159083366394043, "epoch": 5, "memory": 15585, "step": 8424} +{"lr": 0.0002567253492498155, "data_time": 0.001568150520324707, "loss": 0.00951842861250043, "time": 0.6372730970382691, "epoch": 5, "memory": 15585, "step": 8524} +{"lr": 0.00025384653712068393, "data_time": 0.0015228509902954102, "loss": 0.01781184640713036, "time": 0.6685082912445068, "epoch": 5, "memory": 15585, "step": 8624} +{"lr": 0.00025095632042312373, "data_time": 0.0016996145248413086, "loss": 0.008547591418027878, "time": 0.6963433742523193, "epoch": 5, "memory": 15585, "step": 8724} +{"lr": 0.00024805537395776515, "data_time": 0.0018937826156616212, "loss": 0.015852055372670294, "time": 0.7226897001266479, "epoch": 5, "memory": 15585, "step": 8824} +{"lr": 0.00024514437503039924, "data_time": 0.0015663623809814454, "loss": 0.016126701282337307, "time": 0.7444435119628906, "epoch": 5, "memory": 15585, "step": 8924} +{"lr": 0.0002422240032938408, "data_time": 0.0014767408370971679, "loss": 0.005876592779532075, "time": 0.7543826580047608, "epoch": 5, "memory": 15585, "step": 9024} +{"lr": 0.00023929494058924557, "data_time": 0.0016627073287963866, "loss": 0.01868175994604826, "time": 0.7045737981796265, "epoch": 5, "memory": 15585, "step": 9124} +{"lr": 0.00023635787078691464, "data_time": 0.001768183708190918, "loss": 0.01440737503580749, "time": 0.7288059949874878, "epoch": 5, "memory": 15585, "step": 9224} +{"lr": 0.0002334134796266267, "data_time": 0.0017325162887573242, "loss": 0.014109491789713501, "time": 0.6842771053314209, "epoch": 5, "memory": 15585, "step": 9324} +{"lr": 0.00023046245455753318, "data_time": 0.0016141653060913086, "loss": 0.009199622645974159, "time": 0.7769855260848999, "epoch": 5, "memory": 15585, "step": 9424} +{"lr": 0.0002275054845776535, "data_time": 0.0016528606414794923, "loss": 0.009800035832449793, "time": 0.7770039796829223, "epoch": 5, "memory": 15585, "step": 9524} +{"lr": 0.00022454326007301025, "data_time": 0.0016128301620483398, "loss": 0.011652785446494817, "time": 0.7559764385223389, "epoch": 5, "memory": 15585, "step": 9624} +{"lr": 0.00022157647265643937, "data_time": 0.0014563560485839843, "loss": 0.0071499592158943415, "time": 0.7189206838607788, "epoch": 5, "memory": 15585, "step": 9724} +{"lr": 0.00021860581500611435, "data_time": 0.0016391515731811524, "loss": 0.010576095199212432, "time": 0.7444949626922608, "epoch": 5, "memory": 15585, "step": 9824} +{"lr": 0.00021563198070382117, "data_time": 0.0018813133239746094, "loss": 0.018329182965680958, "time": 0.691676664352417, "epoch": 5, "memory": 15585, "step": 9924} +{"lr": 0.00021265566407302255, "data_time": 0.0016747713088989258, "loss": 0.00799864111468196, "time": 0.680380892753601, "epoch": 5, "memory": 15585, "step": 10024} +{"lr": 0.00020967756001674922, "data_time": 0.0017102956771850586, "loss": 0.007989173568785191, "time": 0.7281992435455322, "epoch": 5, "memory": 15585, "step": 10124} +{"lr": 0.0002066983638553557, "data_time": 0.0018079519271850587, "loss": 0.00930230151861906, "time": 0.701107406616211, "epoch": 5, "memory": 15585, "step": 10224} +{"accuracy/top1": 99.27180480957031, "data_time": 0.08750685316617371, "time": 0.4827237637316594, "step": 5} +{"lr": 0.0002020502853543317, "data_time": 0.1426548957824707, "loss": 0.007771449815481901, "time": 0.6573257684707642, "epoch": 6, "memory": 15585, "step": 10380} +{"lr": 0.000199071463167015, "data_time": 0.0019295215606689453, "loss": 0.011988806491717695, "time": 0.6651087522506713, "epoch": 6, "memory": 15585, "step": 10480} +{"lr": 0.00019609402515978532, "data_time": 0.0016627550125122071, "loss": 0.010287653375416993, "time": 0.6591985702514649, "epoch": 6, "memory": 15585, "step": 10580} +{"lr": 0.00019311866649748968, "data_time": 0.0017017126083374023, "loss": 0.013306788494810462, "time": 0.669314694404602, "epoch": 6, "memory": 15585, "step": 10680} +{"lr": 0.00019014608185949442, "data_time": 0.0018292903900146485, "loss": 0.007914037490263582, "time": 0.7153408288955688, "epoch": 6, "memory": 15585, "step": 10780} +{"lr": 0.00018717696527749397, "data_time": 0.0017592430114746094, "loss": 0.014519574679434299, "time": 0.7190943479537963, "epoch": 6, "memory": 15585, "step": 10880} +{"lr": 0.00018421200997346973, "data_time": 0.0017786502838134765, "loss": 0.00859321649186313, "time": 0.7071747541427612, "epoch": 6, "memory": 15585, "step": 10980} +{"lr": 0.00018125190819783787, "data_time": 0.0018130779266357423, "loss": 0.01373229855671525, "time": 0.6682259321212769, "epoch": 6, "memory": 15585, "step": 11080} +{"lr": 0.00017829735106782464, "data_time": 0.001851797103881836, "loss": 0.005507562635466456, "time": 0.723945951461792, "epoch": 6, "memory": 15585, "step": 11180} +{"lr": 0.00017534902840610667, "data_time": 0.0016392230987548827, "loss": 0.0074813324492424725, "time": 0.7383609294891358, "epoch": 6, "memory": 15585, "step": 11280} +{"lr": 0.00017240762857975229, "data_time": 0.0020828723907470705, "loss": 0.008436280023306608, "time": 0.685112452507019, "epoch": 6, "memory": 15585, "step": 11380} +{"lr": 0.00016947383833950319, "data_time": 0.001681375503540039, "loss": 0.006760358670726418, "time": 0.6774173974990845, "epoch": 6, "memory": 15585, "step": 11480} +{"lr": 0.0001665483426594338, "data_time": 0.05896878242492676, "loss": 0.012217931030318142, "time": 0.7410712718963623, "epoch": 6, "memory": 15585, "step": 11580} +{"lr": 0.00016363182457702495, "data_time": 0.0017159461975097656, "loss": 0.008303981134667992, "time": 0.6790645360946655, "epoch": 6, "memory": 15585, "step": 11680} +{"lr": 0.0001607249650336909, "data_time": 0.0017436742782592773, "loss": 0.007574423681944609, "time": 0.719298267364502, "epoch": 6, "memory": 15585, "step": 11780} +{"lr": 0.00015782844271579344, "data_time": 0.19577276706695557, "loss": 0.012513563968241215, "time": 0.7223661422729493, "epoch": 6, "memory": 15585, "step": 11880} +{"lr": 0.0001549429338961849, "data_time": 0.17857692241668702, "loss": 0.007475029630586505, "time": 0.7075620412826538, "epoch": 6, "memory": 15585, "step": 11980} +{"lr": 0.00015206911227631342, "data_time": 0.22871408462524415, "loss": 0.012053997442126273, "time": 0.7575007200241088, "epoch": 6, "memory": 15585, "step": 12080} +{"lr": 0.00014920764882892865, "data_time": 0.20356442928314208, "loss": 0.007269883435219527, "time": 0.7312607049942017, "epoch": 6, "memory": 15585, "step": 12180} +{"lr": 0.00014635921164142459, "data_time": 0.16637322902679444, "loss": 0.011504888720810413, "time": 0.6937987804412842, "epoch": 6, "memory": 15585, "step": 12280} +{"accuracy/top1": 99.47702026367188, "data_time": 0.08086108379676694, "time": 0.47901404099386247, "step": 6} +{"lr": 0.00014194323525930575, "data_time": 0.12070648670196533, "loss": 0.008615901321172714, "time": 0.6390037536621094, "epoch": 7, "memory": 15585, "step": 12436} +{"lr": 0.0001391311681531885, "data_time": 0.19333932399749756, "loss": 0.006534418370574713, "time": 0.7094549655914306, "epoch": 7, "memory": 15585, "step": 12536} +{"lr": 0.00013633447993868745, "data_time": 0.183227014541626, "loss": 0.012832779809832573, "time": 0.7016785621643067, "epoch": 7, "memory": 15585, "step": 12636} +{"lr": 0.00013355382357963566, "data_time": 0.17696599960327147, "loss": 0.008349302085116505, "time": 0.6941949129104614, "epoch": 7, "memory": 15585, "step": 12736} +{"lr": 0.0001307898482967891, "data_time": 0.1283672571182251, "loss": 0.005931426817551255, "time": 0.6420610189437866, "epoch": 7, "memory": 15585, "step": 12836} +{"lr": 0.00012804319941624686, "data_time": 0.13395004272460936, "loss": 0.006299674604088068, "time": 0.6475434303283691, "epoch": 7, "memory": 15585, "step": 12936} +{"lr": 0.00012531451821878227, "data_time": 0.1922152042388916, "loss": 0.004753225343301892, "time": 0.7181339263916016, "epoch": 7, "memory": 15585, "step": 13036} +{"lr": 0.00012260444179011874, "data_time": 0.19852476119995116, "loss": 0.0071558387950062755, "time": 0.7137287616729736, "epoch": 7, "memory": 15585, "step": 13136} +{"lr": 0.00011991360287218363, "data_time": 0.19275262355804443, "loss": 0.010148387541994453, "time": 0.7107342004776, "epoch": 7, "memory": 15585, "step": 13236} +{"lr": 0.00011724262971537941, "data_time": 0.17831885814666748, "loss": 0.0077144072856754065, "time": 0.6920490741729737, "epoch": 7, "memory": 15585, "step": 13336} +{"lr": 0.00011459214593189938, "data_time": 0.1449887275695801, "loss": 0.005640890542417764, "time": 0.6591064453125, "epoch": 7, "memory": 15585, "step": 13436} +{"lr": 0.00011196277035012846, "data_time": 0.2257547378540039, "loss": 0.004714374570176006, "time": 0.7451757669448853, "epoch": 7, "memory": 15585, "step": 13536} +{"lr": 0.0001093551168701611, "data_time": 0.16027560234069824, "loss": 0.01077058119699359, "time": 0.6749431848526001, "epoch": 7, "memory": 15585, "step": 13636} +{"lr": 0.00010676979432046926, "data_time": 0.20957491397857667, "loss": 0.007215554779395461, "time": 0.7285349369049072, "epoch": 7, "memory": 15585, "step": 13736} +{"lr": 0.00010420740631575464, "data_time": 0.22608597278594972, "loss": 0.010731688188388943, "time": 0.7457456111907959, "epoch": 7, "memory": 15585, "step": 13836} +{"lr": 0.00010166855111601829, "data_time": 0.20041840076446532, "loss": 0.007292342046275735, "time": 0.7210818529129028, "epoch": 7, "memory": 15585, "step": 13936} +{"lr": 9.91538214868799e-05, "data_time": 0.019598913192749024, "loss": 0.0029966373927891254, "time": 0.7008754014968872, "epoch": 7, "memory": 15585, "step": 14036} +{"lr": 9.666380456118125e-05, "data_time": 0.0016842126846313477, "loss": 0.007061452884227037, "time": 0.7150977373123169, "epoch": 7, "memory": 15585, "step": 14136} +{"lr": 9.419908170190298e-05, "data_time": 0.001611781120300293, "loss": 0.007744133938103914, "time": 0.6945077180862427, "epoch": 7, "memory": 15585, "step": 14236} +{"lr": 9.176022836643002e-05, "data_time": 0.0016968727111816406, "loss": 0.004769612848758697, "time": 0.7431050062179565, "epoch": 7, "memory": 15585, "step": 14336} +{"accuracy/top1": 99.59452819824219, "data_time": 0.09648423507565358, "time": 0.4932242768709777, "step": 7} +{"lr": 8.80086206394144e-05, "data_time": 0.0015840768814086915, "loss": 0.01414155294187367, "time": 0.7576480627059936, "epoch": 8, "memory": 15585, "step": 14492} +{"lr": 8.563857333566659e-05, "data_time": 0.00155029296875, "loss": 0.0069256947375833985, "time": 0.6686500549316406, "epoch": 8, "memory": 15585, "step": 14592} +{"lr": 8.329639424200186e-05, "data_time": 0.0015752553939819337, "loss": 0.0034400981850922106, "time": 0.7806374073028565, "epoch": 8, "memory": 15585, "step": 14692} +{"lr": 8.098263020459022e-05, "data_time": 0.0015067815780639648, "loss": 0.008589015481993556, "time": 0.6729009866714477, "epoch": 8, "memory": 15585, "step": 14792} +{"lr": 7.869782143532516e-05, "data_time": 0.0015240192413330078, "loss": 0.008856590837240219, "time": 0.6439080715179444, "epoch": 8, "memory": 15585, "step": 14892} +{"lr": 7.644250138569548e-05, "data_time": 0.0021954774856567383, "loss": 0.004469246277585626, "time": 0.6901609182357789, "epoch": 8, "memory": 15585, "step": 14992} +{"lr": 7.4217196622237e-05, "data_time": 0.0017419815063476562, "loss": 0.008263613749295472, "time": 0.678468656539917, "epoch": 8, "memory": 15585, "step": 15092} +{"lr": 7.202242670359085e-05, "data_time": 0.0017836332321166993, "loss": 0.00724510052241385, "time": 0.7328895092010498, "epoch": 8, "memory": 15585, "step": 15192} +{"lr": 6.98587040591985e-05, "data_time": 0.0019298315048217774, "loss": 0.007897790055721998, "time": 0.6909587860107422, "epoch": 8, "memory": 15585, "step": 15292} +{"lr": 6.772653386966062e-05, "data_time": 0.0017855167388916016, "loss": 0.0041174809448421, "time": 0.7238125324249267, "epoch": 8, "memory": 15585, "step": 15392} +{"lr": 6.562641394878866e-05, "data_time": 0.0018667936325073241, "loss": 0.0037497988902032374, "time": 0.7481331586837768, "epoch": 8, "memory": 15585, "step": 15492} +{"lr": 6.35588346273771e-05, "data_time": 0.0016849994659423827, "loss": 0.004643295519053936, "time": 0.7302852153778077, "epoch": 8, "memory": 15585, "step": 15592} +{"lr": 6.152427863872164e-05, "data_time": 0.0017604827880859375, "loss": 0.004337272793054581, "time": 0.7296119213104248, "epoch": 8, "memory": 15585, "step": 15692} +{"lr": 5.952322100591216e-05, "data_time": 0.0016360759735107421, "loss": 0.006246315967291594, "time": 0.7323440790176392, "epoch": 8, "memory": 15585, "step": 15792} +{"lr": 5.755612893092559e-05, "data_time": 0.0016683340072631836, "loss": 0.004080366250127554, "time": 0.7764551639556885, "epoch": 8, "memory": 15585, "step": 15892} +{"lr": 5.5623461685544754e-05, "data_time": 0.0018137454986572265, "loss": 0.00618846919387579, "time": 0.6559581518173218, "epoch": 8, "memory": 15585, "step": 15992} +{"lr": 5.37256705041282e-05, "data_time": 0.0016515731811523437, "loss": 0.006456415681168437, "time": 0.6898155212402344, "epoch": 8, "memory": 15585, "step": 16092} +{"lr": 5.186319847825775e-05, "data_time": 0.0017484426498413086, "loss": 0.009180582035332919, "time": 0.6893758296966552, "epoch": 8, "memory": 15585, "step": 16192} +{"lr": 5.0036480453286144e-05, "data_time": 0.0016029834747314452, "loss": 0.008056677551940084, "time": 0.7109777450561523, "epoch": 8, "memory": 15585, "step": 16292} +{"lr": 4.8245942926810745e-05, "data_time": 0.001706099510192871, "loss": 0.0028826996218413114, "time": 0.6360068798065186, "epoch": 8, "memory": 15585, "step": 16392} +{"accuracy/top1": 99.66238403320312, "data_time": 0.09673175655427527, "time": 0.49555478330518377, "step": 8} +{"lr": 4.5525937358206823e-05, "data_time": 0.16450128555297852, "loss": 0.003932453924790025, "time": 0.6787249088287354, "epoch": 9, "memory": 15585, "step": 16548} +{"lr": 4.382990660001016e-05, "data_time": 0.16617732048034667, "loss": 0.006312030553817749, "time": 0.6840532541275024, "epoch": 9, "memory": 15585, "step": 16648} +{"lr": 4.217150543579113e-05, "data_time": 0.1792240858078003, "loss": 0.005301159713417292, "time": 0.6963784217834472, "epoch": 9, "memory": 15585, "step": 16748} +{"lr": 4.0551121064943564e-05, "data_time": 0.1785569190979004, "loss": 0.009223576216027141, "time": 0.6942339658737182, "epoch": 9, "memory": 15585, "step": 16848} +{"lr": 3.896913181079464e-05, "data_time": 0.18540642261505128, "loss": 0.00460140579380095, "time": 0.70242919921875, "epoch": 9, "memory": 15585, "step": 16948} +{"lr": 3.742590703227469e-05, "data_time": 0.20136003494262694, "loss": 0.007680822629481554, "time": 0.7193239688873291, "epoch": 9, "memory": 15585, "step": 17048} +{"lr": 3.592180703768032e-05, "data_time": 0.18309106826782226, "loss": 0.008242048975080251, "time": 0.6988625764846802, "epoch": 9, "memory": 15585, "step": 17148} +{"lr": 3.445718300055071e-05, "data_time": 0.18934359550476074, "loss": 0.006236546812579036, "time": 0.7060253858566284, "epoch": 9, "memory": 15585, "step": 17248} +{"lr": 3.3032376877676173e-05, "data_time": 0.17684197425842285, "loss": 0.00680283117108047, "time": 0.6924414873123169, "epoch": 9, "memory": 15585, "step": 17348} +{"lr": 3.1647721329259246e-05, "data_time": 0.21000983715057372, "loss": 0.005297743622213602, "time": 0.7288161277770996, "epoch": 9, "memory": 15585, "step": 17448} +{"lr": 3.0303539641245756e-05, "data_time": 0.19528508186340332, "loss": 0.004597390862181782, "time": 0.7111314296722412, "epoch": 9, "memory": 15585, "step": 17548} +{"lr": 2.9000145649844975e-05, "data_time": 0.21944591999053956, "loss": 0.008540700143203139, "time": 0.7567103624343872, "epoch": 9, "memory": 15585, "step": 17648} +{"lr": 2.7737843668255695e-05, "data_time": 0.17396090030670167, "loss": 0.002788296481594443, "time": 0.6878469944000244, "epoch": 9, "memory": 15585, "step": 17748} +{"lr": 2.6516928415616184e-05, "data_time": 0.18790993690490723, "loss": 0.007269242499023676, "time": 0.7097814798355102, "epoch": 9, "memory": 15585, "step": 17848} +{"lr": 2.5337684948193946e-05, "data_time": 0.2300894021987915, "loss": 0.005430571408942342, "time": 0.744309401512146, "epoch": 9, "memory": 15585, "step": 17948} +{"lr": 2.4200388592830947e-05, "data_time": 0.19214770793914795, "loss": 0.006849743146449328, "time": 0.7057928800582886, "epoch": 9, "memory": 15585, "step": 18048} +{"lr": 2.3105304882661933e-05, "data_time": 0.25373048782348634, "loss": 0.006313639925792813, "time": 0.7711594104766846, "epoch": 9, "memory": 15585, "step": 18148} +{"lr": 2.20526894951175e-05, "data_time": 0.16188671588897705, "loss": 0.004326836299151182, "time": 0.6790915250778198, "epoch": 9, "memory": 15585, "step": 18248} +{"lr": 2.1042788192229725e-05, "data_time": 0.1542069673538208, "loss": 0.005015872605144978, "time": 0.667760705947876, "epoch": 9, "memory": 15585, "step": 18348} +{"lr": 2.007583676325189e-05, "data_time": 0.6659237384796143, "loss": 0.008820148324593902, "time": 1.188455843925476, "epoch": 9, "memory": 15585, "step": 18448} +{"accuracy/top1": 99.74016571044922, "data_time": 0.08351332633221736, "time": 0.4794860472444628, "step": 9} +{"lr": 1.8653686449507784e-05, "data_time": 0.18245489597320558, "loss": 0.004861492803320289, "time": 0.6963652849197388, "epoch": 10, "memory": 15585, "step": 18604} +{"lr": 1.7797691544415563e-05, "data_time": 0.1789107084274292, "loss": 0.003677115123718977, "time": 0.696782398223877, "epoch": 10, "memory": 15585, "step": 18704} +{"lr": 1.6985404170356326e-05, "data_time": 0.1933534860610962, "loss": 0.0043910601641982796, "time": 0.7105748653411865, "epoch": 10, "memory": 15585, "step": 18804} +{"lr": 1.6217013978172993e-05, "data_time": 0.21540849208831786, "loss": 0.003004590841010213, "time": 0.7314217567443848, "epoch": 10, "memory": 15585, "step": 18904} +{"lr": 1.549270036970338e-05, "data_time": 0.21487078666687012, "loss": 0.006738194124773144, "time": 0.7328275918960572, "epoch": 10, "memory": 15585, "step": 19004} +{"lr": 1.4812632455893675e-05, "data_time": 0.15540411472320556, "loss": 0.006134961266070604, "time": 0.6692819595336914, "epoch": 10, "memory": 15585, "step": 19104} +{"lr": 1.4176969017315005e-05, "data_time": 0.18832159042358398, "loss": 0.004016678966581822, "time": 0.7060202121734619, "epoch": 10, "memory": 15585, "step": 19204} +{"lr": 1.358585846709161e-05, "data_time": 0.16427857875823976, "loss": 0.00385310766287148, "time": 0.6779886960983277, "epoch": 10, "memory": 15585, "step": 19304} +{"lr": 1.3039438816249855e-05, "data_time": 0.18878569602966308, "loss": 0.005551433609798551, "time": 0.7067301750183106, "epoch": 10, "memory": 15585, "step": 19404} +{"lr": 1.2537837641495561e-05, "data_time": 0.22544901371002196, "loss": 0.004782603727653623, "time": 0.7426913499832153, "epoch": 10, "memory": 15585, "step": 19504} +{"lr": 1.2081172055427927e-05, "data_time": 0.21164205074310302, "loss": 0.004750721715390682, "time": 0.7287100076675415, "epoch": 10, "memory": 15585, "step": 19604} +{"lr": 1.1669548679196218e-05, "data_time": 0.18330190181732178, "loss": 0.004795972025021911, "time": 0.7001710653305053, "epoch": 10, "memory": 15585, "step": 19704} +{"lr": 1.130306361760621e-05, "data_time": 0.19480786323547364, "loss": 0.004976117424666882, "time": 0.7107309579849244, "epoch": 10, "memory": 15585, "step": 19804} +{"lr": 1.0981802436681767e-05, "data_time": 0.18398218154907225, "loss": 0.005396238854154944, "time": 0.701444411277771, "epoch": 10, "memory": 15585, "step": 19904} +{"lr": 1.0705840143687269e-05, "data_time": 0.2710393190383911, "loss": 0.005424016620963812, "time": 0.7881027221679687, "epoch": 10, "memory": 15585, "step": 20004} +{"lr": 1.0475241169614925e-05, "data_time": 0.1448136329650879, "loss": 0.006270120851695538, "time": 0.6615827798843383, "epoch": 10, "memory": 15585, "step": 20104} +{"lr": 1.0290059354141742e-05, "data_time": 0.17011544704437256, "loss": 0.0036644930019974708, "time": 0.6838934183120727, "epoch": 10, "memory": 15585, "step": 20204} +{"lr": 1.015033793305906e-05, "data_time": 0.18959639072418213, "loss": 0.0066305665299296376, "time": 0.7083418846130372, "epoch": 10, "memory": 15585, "step": 20304} +{"lr": 1.0056109528177973e-05, "data_time": 0.18329384326934814, "loss": 0.004180255625396967, "time": 0.7005026340484619, "epoch": 10, "memory": 15585, "step": 20404} +{"lr": 1.0007396139712939e-05, "data_time": 0.1671433687210083, "loss": 0.0035097086802124977, "time": 0.6808896064758301, "epoch": 10, "memory": 15585, "step": 20504} +{"accuracy/top1": 99.76002502441406, "data_time": 0.09428098944366956, "time": 0.48892652011308513, "step": 10} diff --git a/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1/20230605_024312/vis_data/config.py b/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1/20230605_024312/vis_data/config.py new file mode 100644 index 0000000000000000000000000000000000000000..c1d367f5d6e5dc2c7a25ab22cbbe5fda918650bd --- /dev/null +++ b/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1/20230605_024312/vis_data/config.py @@ -0,0 +1,383 @@ +optim_wrapper = dict( + optimizer=dict( + type='AdamW', + lr=0.0004, + weight_decay=0.05, + eps=1e-08, + betas=(0.9, 0.999), + _scope_='mmpretrain'), + paramwise_cfg=dict( + norm_decay_mult=0.0, + bias_decay_mult=0.0, + flat_decay_mult=0.0, + custom_keys=dict({ + '.absolute_pos_embed': dict(decay_mult=0.0), + '.relative_position_bias_table': dict(decay_mult=0.0) + })), + type='AmpOptimWrapper', + dtype='bfloat16', + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=False, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=4096) +model = dict( + type='ImageClassifier', + backbone=dict( + frozen_stages=2, + type='ConvNeXt', + arch='small', + drop_path_rate=0.4, + init_cfg=dict( + type='Pretrained', + checkpoint= + 'https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth', + prefix='backbone')), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=768, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + init_cfg=None), + init_cfg=dict( + type='TruncNormal', layer=['Conv2d', 'Linear'], std=0.02, bias=0.0), + train_cfg=None) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/all1.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/all2.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/all3.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3fake8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/cc1m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3real8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV2-1-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/if-dpmsolver++-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV2-1-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/if-dpmsolver++-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1' diff --git a/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1/20230605_024312/vis_data/events.out.tfevents.1685904207.SH-IDC1-10-140-24-15.115896.0 b/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1/20230605_024312/vis_data/events.out.tfevents.1685904207.SH-IDC1-10-140-24-15.115896.0 new file mode 100644 index 0000000000000000000000000000000000000000..6b4518e9fb2d016ed6c717383ac4aef1bc6432fd --- /dev/null +++ b/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1/20230605_024312/vis_data/events.out.tfevents.1685904207.SH-IDC1-10-140-24-15.115896.0 @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:54854e321bb232300b9f96d0a6bd6b106a16225666633a82e4b3ce113a26e4ee +size 2960046 diff --git a/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1/20230605_024312/vis_data/scalars.json b/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1/20230605_024312/vis_data/scalars.json new file mode 100644 index 0000000000000000000000000000000000000000..8879f019d58e05add43b7c40e73d5fd92a24b907 --- /dev/null +++ b/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1/20230605_024312/vis_data/scalars.json @@ -0,0 +1,210 @@ +{"lr": 0.00039997768893775784, "data_time": 0.2427647352218628, "loss": 0.0622634295374155, "time": 0.7631019592285156, "epoch": 1, "memory": 15587, "step": 100} +{"lr": 0.00039990985724436777, "data_time": 0.18695836067199706, "loss": 0.047248711064457896, "time": 0.8761033773422241, "epoch": 1, "memory": 15587, "step": 200} +{"lr": 0.00039979651848055537, "data_time": 0.0017068624496459962, "loss": 0.04533916190266609, "time": 0.619895601272583, "epoch": 1, "memory": 15587, "step": 300} +{"lr": 0.0003996376991083742, "data_time": 0.0016076803207397462, "loss": 0.03946216888725758, "time": 0.8002885818481446, "epoch": 1, "memory": 15587, "step": 400} +{"lr": 0.00039943343620857754, "data_time": 0.0015278339385986328, "loss": 0.03365235310047865, "time": 0.7466793298721314, "epoch": 1, "memory": 15587, "step": 500} +{"lr": 0.0003991837774719618, "data_time": 0.001695871353149414, "loss": 0.029732736758887768, "time": 0.7768775939941406, "epoch": 1, "memory": 15587, "step": 600} +{"lr": 0.00039888878118822876, "data_time": 0.0015054941177368164, "loss": 0.025014087837189437, "time": 0.7019802570343018, "epoch": 1, "memory": 15587, "step": 700} +{"lr": 0.0003985485162323803, "data_time": 0.0014446496963500977, "loss": 0.019376942422240973, "time": 0.7305632829666138, "epoch": 1, "memory": 15587, "step": 800} +{"lr": 0.0003981630620486352, "data_time": 0.001797652244567871, "loss": 0.027765481360256673, "time": 0.710541844367981, "epoch": 1, "memory": 15587, "step": 900} +{"lr": 0.0003977325086318812, "data_time": 0.0017927885055541992, "loss": 0.030781422276049852, "time": 0.7124827623367309, "epoch": 1, "memory": 15587, "step": 1000} +{"lr": 0.00039725695650666204, "data_time": 0.0015607357025146484, "loss": 0.019887801771983505, "time": 0.7369407653808594, "epoch": 1, "memory": 15587, "step": 1100} +{"lr": 0.0003967365167037102, "data_time": 0.0018869400024414062, "loss": 0.02237267531454563, "time": 0.7870367765426636, "epoch": 1, "memory": 15587, "step": 1200} +{"lr": 0.0003961713107340207, "data_time": 0.0019826889038085938, "loss": 0.01975932056084275, "time": 0.8032874345779419, "epoch": 1, "memory": 15587, "step": 1300} +{"lr": 0.0003955614705604829, "data_time": 0.0018016815185546875, "loss": 0.02405152888968587, "time": 0.7527009725570679, "epoch": 1, "memory": 15587, "step": 1400} +{"lr": 0.00039490713856706923, "data_time": 0.0017481088638305665, "loss": 0.02071991842240095, "time": 0.7369158744812012, "epoch": 1, "memory": 15587, "step": 1500} +{"lr": 0.00039420846752559095, "data_time": 0.0014647483825683595, "loss": 0.01888023940846324, "time": 0.7223272562026978, "epoch": 1, "memory": 15587, "step": 1600} +{"lr": 0.0003934656205600326, "data_time": 0.0015542268753051757, "loss": 0.019930011453107, "time": 0.7527501344680786, "epoch": 1, "memory": 15587, "step": 1700} +{"lr": 0.0003926787711084616, "data_time": 0.0015392780303955078, "loss": 0.025704196002334357, "time": 0.6902652263641358, "epoch": 1, "memory": 15587, "step": 1800} +{"lr": 0.0003918481028825385, "data_time": 0.002183270454406738, "loss": 0.028769377805292606, "time": 0.7484632015228272, "epoch": 1, "memory": 15587, "step": 1900} +{"lr": 0.0003909738098246218, "data_time": 0.001527857780456543, "loss": 0.020901503786444665, "time": 0.7452915906906128, "epoch": 1, "memory": 15587, "step": 2000} +{"accuracy/top1": 77.52677154541016, "data_time": 0.10747306744257609, "time": 0.5418768803278605, "step": 1} +{"lr": 0.0003895232896423554, "data_time": 0.0020531415939331055, "loss": 0.018552379263564945, "time": 0.744081974029541, "epoch": 2, "memory": 15587, "step": 2156} +{"lr": 0.00038853827118924355, "data_time": 0.0020484209060668947, "loss": 0.01960562001913786, "time": 0.7278545379638672, "epoch": 2, "memory": 15585, "step": 2256} +{"lr": 0.0003875104006754391, "data_time": 0.0018478155136108399, "loss": 0.013127893302589655, "time": 0.7977020740509033, "epoch": 2, "memory": 15585, "step": 2356} +{"lr": 0.00038643991808560204, "data_time": 0.0018390893936157226, "loss": 0.020102136302739382, "time": 0.6740249633789063, "epoch": 2, "memory": 15585, "step": 2456} +{"lr": 0.00038532707335335494, "data_time": 0.0018429517745971679, "loss": 0.013678281661123037, "time": 0.7428551435470581, "epoch": 2, "memory": 15585, "step": 2556} +{"lr": 0.00038417212630292743, "data_time": 0.0015572309494018555, "loss": 0.02540783192962408, "time": 0.7237333059310913, "epoch": 2, "memory": 15585, "step": 2656} +{"lr": 0.00038297534658849464, "data_time": 0.0014730215072631836, "loss": 0.012263161223381757, "time": 0.7169277191162109, "epoch": 2, "memory": 15585, "step": 2756} +{"lr": 0.00038173701363121816, "data_time": 0.0016788482666015626, "loss": 0.021818548487499357, "time": 0.7130337715148926, "epoch": 2, "memory": 15585, "step": 2856} +{"lr": 0.00038045741655400836, "data_time": 0.0018208980560302734, "loss": 0.020187804708257318, "time": 0.7371300458908081, "epoch": 2, "memory": 15585, "step": 2956} +{"lr": 0.00037913685411401933, "data_time": 0.0018122196197509766, "loss": 0.019882525131106377, "time": 0.7195440292358398, "epoch": 2, "memory": 15585, "step": 3056} +{"lr": 0.00037777563463289686, "data_time": 0.0016786575317382813, "loss": 0.02451257212087512, "time": 0.7288497686386108, "epoch": 2, "memory": 15585, "step": 3156} +{"lr": 0.0003763740759247917, "data_time": 0.0015790224075317382, "loss": 0.01938260458409786, "time": 0.6837079524993896, "epoch": 2, "memory": 15585, "step": 3256} +{"lr": 0.0003749325052221578, "data_time": 0.0016111135482788086, "loss": 0.01937106898985803, "time": 0.6827808141708374, "epoch": 2, "memory": 15585, "step": 3356} +{"lr": 0.00037345125909934995, "data_time": 0.0018295764923095704, "loss": 0.020618916163221, "time": 0.727691388130188, "epoch": 2, "memory": 15585, "step": 3456} +{"lr": 0.00037193068339404196, "data_time": 0.0017955303192138672, "loss": 0.019244227232411502, "time": 0.7252246618270874, "epoch": 2, "memory": 15585, "step": 3556} +{"lr": 0.0003703711331264808, "data_time": 0.0016932725906372071, "loss": 0.020278052147477865, "time": 0.6932339906692505, "epoch": 2, "memory": 15585, "step": 3656} +{"lr": 0.0003687729724165983, "data_time": 0.0017278194427490234, "loss": 0.01249537207186222, "time": 0.6835992813110352, "epoch": 2, "memory": 15585, "step": 3756} +{"lr": 0.000367136574398996, "data_time": 0.001810908317565918, "loss": 0.021494934195652605, "time": 0.6877713680267334, "epoch": 2, "memory": 15585, "step": 3856} +{"lr": 0.00036546232113582794, "data_time": 0.0018505811691284179, "loss": 0.01864777714945376, "time": 0.7289227247238159, "epoch": 2, "memory": 15585, "step": 3956} +{"lr": 0.0003637506035275962, "data_time": 0.002080535888671875, "loss": 0.017337575368583203, "time": 0.7337353229522705, "epoch": 2, "memory": 15585, "step": 4056} +{"accuracy/top1": 95.22533416748047, "data_time": 0.09835075159541896, "time": 0.4946637935325748, "step": 2} +{"lr": 0.00036100646510740043, "data_time": 0.10377991199493408, "loss": 0.016776063246652484, "time": 0.7853656768798828, "epoch": 3, "memory": 15585, "step": 4212} +{"lr": 0.0003592006803637847, "data_time": 0.001594686508178711, "loss": 0.01856973674148321, "time": 0.7196727514266967, "epoch": 3, "memory": 15585, "step": 4312} +{"lr": 0.00035735889322747414, "data_time": 0.0017902612686157226, "loss": 0.010173213155940174, "time": 0.7030227422714234, "epoch": 3, "memory": 15585, "step": 4412} +{"lr": 0.0003554815337143657, "data_time": 0.001588726043701172, "loss": 0.01909800935536623, "time": 0.7280056238174438, "epoch": 3, "memory": 15585, "step": 4512} +{"lr": 0.00035356904014570526, "data_time": 0.0017295122146606446, "loss": 0.014161157002672554, "time": 0.6520989656448364, "epoch": 3, "memory": 15585, "step": 4612} +{"lr": 0.00035162185904575157, "data_time": 0.0017443418502807618, "loss": 0.01739039127714932, "time": 0.7259718418121338, "epoch": 3, "memory": 15585, "step": 4712} +{"lr": 0.0003496404450375223, "data_time": 0.0019626379013061523, "loss": 0.01800237870775163, "time": 0.7121296167373657, "epoch": 3, "memory": 15585, "step": 4812} +{"lr": 0.0003476252607366498, "data_time": 0.0016444921493530273, "loss": 0.012891631200909614, "time": 0.6917529344558716, "epoch": 3, "memory": 15585, "step": 4912} +{"lr": 0.0003455767766433698, "data_time": 0.0015131235122680664, "loss": 0.018378550466150046, "time": 0.7705854415893555, "epoch": 3, "memory": 15585, "step": 5012} +{"lr": 0.0003434954710326717, "data_time": 0.0016249895095825195, "loss": 0.018253946397453546, "time": 0.7216732501983643, "epoch": 3, "memory": 15585, "step": 5112} +{"lr": 0.00034138182984263203, "data_time": 0.0019607782363891602, "loss": 0.01774289016611874, "time": 0.7416847467422485, "epoch": 3, "memory": 15585, "step": 5212} +{"lr": 0.0003392363465609565, "data_time": 0.0021759748458862306, "loss": 0.012799293221905827, "time": 0.7296148538589478, "epoch": 3, "memory": 15585, "step": 5312} +{"lr": 0.0003370595221097656, "data_time": 0.0018844366073608398, "loss": 0.011717123771086335, "time": 0.7495620965957641, "epoch": 3, "memory": 15585, "step": 5412} +{"lr": 0.00033485186472863657, "data_time": 0.0017618656158447266, "loss": 0.01528772497549653, "time": 0.7832259416580201, "epoch": 3, "memory": 15585, "step": 5512} +{"lr": 0.00033261388985594397, "data_time": 0.001717352867126465, "loss": 0.01808881820179522, "time": 0.7567421436309815, "epoch": 3, "memory": 15585, "step": 5612} +{"lr": 0.0003303461200085145, "data_time": 0.0015334367752075195, "loss": 0.011359138414263725, "time": 0.8013696193695068, "epoch": 3, "memory": 15585, "step": 5712} +{"lr": 0.0003280490846596322, "data_time": 0.001840519905090332, "loss": 0.010607956117019058, "time": 0.7803977012634278, "epoch": 3, "memory": 15585, "step": 5812} +{"lr": 0.00032572332011541707, "data_time": 0.0018265247344970703, "loss": 0.014171441504731774, "time": 0.7029386997222901, "epoch": 3, "memory": 15585, "step": 5912} +{"lr": 0.00032336936938961124, "data_time": 0.002108311653137207, "loss": 0.017399783618748187, "time": 0.7051195621490478, "epoch": 3, "memory": 15585, "step": 6012} +{"lr": 0.00032098778207679645, "data_time": 0.0016778945922851563, "loss": 0.01578828217461705, "time": 0.7325539112091064, "epoch": 3, "memory": 15585, "step": 6112} +{"accuracy/top1": 98.23577117919922, "data_time": 0.09079766273498535, "time": 0.485427731373271, "step": 3} +{"lr": 0.0003172186408067781, "data_time": 0.0015857219696044922, "loss": 0.016427195025607945, "time": 0.6740639925003051, "epoch": 4, "memory": 15585, "step": 6268} +{"lr": 0.00031476885256500535, "data_time": 0.001706075668334961, "loss": 0.014819549396634102, "time": 0.7042318344116211, "epoch": 4, "memory": 15585, "step": 6368} +{"lr": 0.00031229343576357664, "data_time": 0.0018445730209350586, "loss": 0.013786209328100086, "time": 0.6967251777648926, "epoch": 4, "memory": 15585, "step": 6468} +{"lr": 0.00030979296835667217, "data_time": 0.0017379045486450196, "loss": 0.015972312819212676, "time": 0.7047409772872925, "epoch": 4, "memory": 15585, "step": 6568} +{"lr": 0.000307268034147225, "data_time": 0.0017672538757324218, "loss": 0.02125865938141942, "time": 1.0374451398849487, "epoch": 4, "memory": 15585, "step": 6668} +{"lr": 0.00030471922265061713, "data_time": 0.0017290353775024415, "loss": 0.016504886886104943, "time": 1.171081256866455, "epoch": 4, "memory": 15585, "step": 6768} +{"lr": 0.00030214712895704036, "data_time": 0.001592707633972168, "loss": 0.017919448390603065, "time": 0.7085912466049195, "epoch": 4, "memory": 15585, "step": 6868} +{"lr": 0.00029955235359255527, "data_time": 0.0018253564834594727, "loss": 0.01797847356647253, "time": 0.6495321273803711, "epoch": 4, "memory": 15585, "step": 6968} +{"lr": 0.00029693550237888314, "data_time": 0.0018332242965698243, "loss": 0.010003110906109215, "time": 0.6841523170471191, "epoch": 4, "memory": 15585, "step": 7068} +{"lr": 0.0002942971862919601, "data_time": 0.0017138004302978515, "loss": 0.014989589992910623, "time": 0.7165249824523926, "epoch": 4, "memory": 15585, "step": 7168} +{"lr": 0.0002916380213192881, "data_time": 0.0016910552978515625, "loss": 0.014970519207417965, "time": 0.7719467639923095, "epoch": 4, "memory": 15585, "step": 7268} +{"lr": 0.00028895862831611433, "data_time": 0.001915311813354492, "loss": 0.013057804573327303, "time": 0.6912996053695679, "epoch": 4, "memory": 15585, "step": 7368} +{"lr": 0.0002862596328604766, "data_time": 0.0016709566116333008, "loss": 0.011388324014842511, "time": 0.7018782615661621, "epoch": 4, "memory": 15585, "step": 7468} +{"lr": 0.0002835416651071449, "data_time": 0.0015023708343505859, "loss": 0.01167175085283816, "time": 0.7163537025451661, "epoch": 4, "memory": 15585, "step": 7568} +{"lr": 0.00028080535964049476, "data_time": 0.0016505002975463867, "loss": 0.010760634113103152, "time": 0.7725522518157959, "epoch": 4, "memory": 15585, "step": 7668} +{"lr": 0.0002780513553263448, "data_time": 0.0015471935272216796, "loss": 0.019454200752079487, "time": 0.6858330011367798, "epoch": 4, "memory": 15585, "step": 7768} +{"lr": 0.0002752802951627973, "data_time": 0.0016010522842407227, "loss": 0.016746526444330813, "time": 0.6736471652984619, "epoch": 4, "memory": 15585, "step": 7868} +{"lr": 0.0002724928261301117, "data_time": 0.0016916275024414062, "loss": 0.007524235686287284, "time": 10.016830921173096, "epoch": 4, "memory": 15585, "step": 7968} +{"lr": 0.00026968959903964925, "data_time": 0.0019032955169677734, "loss": 0.009913359256461262, "time": 0.7568388223648072, "epoch": 4, "memory": 15585, "step": 8068} +{"lr": 0.0002668712683819236, "data_time": 0.0018506050109863281, "loss": 0.02095145294442773, "time": 0.777468228340149, "epoch": 4, "memory": 15585, "step": 8168} +{"accuracy/top1": 98.97721099853516, "data_time": 0.09110732547572402, "time": 0.48946628414216586, "step": 4} +{"lr": 0.00026244607407074216, "data_time": 0.0015883445739746094, "loss": 0.010050296876579523, "time": 0.7085708618164063, "epoch": 5, "memory": 15585, "step": 8324} +{"lr": 0.00025959208467259544, "data_time": 0.0014809131622314452, "loss": 0.009908492537215352, "time": 0.7159083366394043, "epoch": 5, "memory": 15585, "step": 8424} +{"lr": 0.0002567253492498155, "data_time": 0.001568150520324707, "loss": 0.00951842861250043, "time": 0.6372730970382691, "epoch": 5, "memory": 15585, "step": 8524} +{"lr": 0.00025384653712068393, "data_time": 0.0015228509902954102, "loss": 0.01781184640713036, "time": 0.6685082912445068, "epoch": 5, "memory": 15585, "step": 8624} +{"lr": 0.00025095632042312373, "data_time": 0.0016996145248413086, "loss": 0.008547591418027878, "time": 0.6963433742523193, "epoch": 5, "memory": 15585, "step": 8724} +{"lr": 0.00024805537395776515, "data_time": 0.0018937826156616212, "loss": 0.015852055372670294, "time": 0.7226897001266479, "epoch": 5, "memory": 15585, "step": 8824} +{"lr": 0.00024514437503039924, "data_time": 0.0015663623809814454, "loss": 0.016126701282337307, "time": 0.7444435119628906, "epoch": 5, "memory": 15585, "step": 8924} +{"lr": 0.0002422240032938408, "data_time": 0.0014767408370971679, "loss": 0.005876592779532075, "time": 0.7543826580047608, "epoch": 5, "memory": 15585, "step": 9024} +{"lr": 0.00023929494058924557, "data_time": 0.0016627073287963866, "loss": 0.01868175994604826, "time": 0.7045737981796265, "epoch": 5, "memory": 15585, "step": 9124} +{"lr": 0.00023635787078691464, "data_time": 0.001768183708190918, "loss": 0.01440737503580749, "time": 0.7288059949874878, "epoch": 5, "memory": 15585, "step": 9224} +{"lr": 0.0002334134796266267, "data_time": 0.0017325162887573242, "loss": 0.014109491789713501, "time": 0.6842771053314209, "epoch": 5, "memory": 15585, "step": 9324} +{"lr": 0.00023046245455753318, "data_time": 0.0016141653060913086, "loss": 0.009199622645974159, "time": 0.7769855260848999, "epoch": 5, "memory": 15585, "step": 9424} +{"lr": 0.0002275054845776535, "data_time": 0.0016528606414794923, "loss": 0.009800035832449793, "time": 0.7770039796829223, "epoch": 5, "memory": 15585, "step": 9524} +{"lr": 0.00022454326007301025, "data_time": 0.0016128301620483398, "loss": 0.011652785446494817, "time": 0.7559764385223389, "epoch": 5, "memory": 15585, "step": 9624} +{"lr": 0.00022157647265643937, "data_time": 0.0014563560485839843, "loss": 0.0071499592158943415, "time": 0.7189206838607788, "epoch": 5, "memory": 15585, "step": 9724} +{"lr": 0.00021860581500611435, "data_time": 0.0016391515731811524, "loss": 0.010576095199212432, "time": 0.7444949626922608, "epoch": 5, "memory": 15585, "step": 9824} +{"lr": 0.00021563198070382117, "data_time": 0.0018813133239746094, "loss": 0.018329182965680958, "time": 0.691676664352417, "epoch": 5, "memory": 15585, "step": 9924} +{"lr": 0.00021265566407302255, "data_time": 0.0016747713088989258, "loss": 0.00799864111468196, "time": 0.680380892753601, "epoch": 5, "memory": 15585, "step": 10024} +{"lr": 0.00020967756001674922, "data_time": 0.0017102956771850586, "loss": 0.007989173568785191, "time": 0.7281992435455322, "epoch": 5, "memory": 15585, "step": 10124} +{"lr": 0.0002066983638553557, "data_time": 0.0018079519271850587, "loss": 0.00930230151861906, "time": 0.701107406616211, "epoch": 5, "memory": 15585, "step": 10224} +{"accuracy/top1": 99.27180480957031, "data_time": 0.08750685316617371, "time": 0.4827237637316594, "step": 5} +{"lr": 0.0002020502853543317, "data_time": 0.1426548957824707, "loss": 0.007771449815481901, "time": 0.6573257684707642, "epoch": 6, "memory": 15585, "step": 10380} +{"lr": 0.000199071463167015, "data_time": 0.0019295215606689453, "loss": 0.011988806491717695, "time": 0.6651087522506713, "epoch": 6, "memory": 15585, "step": 10480} +{"lr": 0.00019609402515978532, "data_time": 0.0016627550125122071, "loss": 0.010287653375416993, "time": 0.6591985702514649, "epoch": 6, "memory": 15585, "step": 10580} +{"lr": 0.00019311866649748968, "data_time": 0.0017017126083374023, "loss": 0.013306788494810462, "time": 0.669314694404602, "epoch": 6, "memory": 15585, "step": 10680} +{"lr": 0.00019014608185949442, "data_time": 0.0018292903900146485, "loss": 0.007914037490263582, "time": 0.7153408288955688, "epoch": 6, "memory": 15585, "step": 10780} +{"lr": 0.00018717696527749397, "data_time": 0.0017592430114746094, "loss": 0.014519574679434299, "time": 0.7190943479537963, "epoch": 6, "memory": 15585, "step": 10880} +{"lr": 0.00018421200997346973, "data_time": 0.0017786502838134765, "loss": 0.00859321649186313, "time": 0.7071747541427612, "epoch": 6, "memory": 15585, "step": 10980} +{"lr": 0.00018125190819783787, "data_time": 0.0018130779266357423, "loss": 0.01373229855671525, "time": 0.6682259321212769, "epoch": 6, "memory": 15585, "step": 11080} +{"lr": 0.00017829735106782464, "data_time": 0.001851797103881836, "loss": 0.005507562635466456, "time": 0.723945951461792, "epoch": 6, "memory": 15585, "step": 11180} +{"lr": 0.00017534902840610667, "data_time": 0.0016392230987548827, "loss": 0.0074813324492424725, "time": 0.7383609294891358, "epoch": 6, "memory": 15585, "step": 11280} +{"lr": 0.00017240762857975229, "data_time": 0.0020828723907470705, "loss": 0.008436280023306608, "time": 0.685112452507019, "epoch": 6, "memory": 15585, "step": 11380} +{"lr": 0.00016947383833950319, "data_time": 0.001681375503540039, "loss": 0.006760358670726418, "time": 0.6774173974990845, "epoch": 6, "memory": 15585, "step": 11480} +{"lr": 0.0001665483426594338, "data_time": 0.05896878242492676, "loss": 0.012217931030318142, "time": 0.7410712718963623, "epoch": 6, "memory": 15585, "step": 11580} +{"lr": 0.00016363182457702495, "data_time": 0.0017159461975097656, "loss": 0.008303981134667992, "time": 0.6790645360946655, "epoch": 6, "memory": 15585, "step": 11680} +{"lr": 0.0001607249650336909, "data_time": 0.0017436742782592773, "loss": 0.007574423681944609, "time": 0.719298267364502, "epoch": 6, "memory": 15585, "step": 11780} +{"lr": 0.00015782844271579344, "data_time": 0.19577276706695557, "loss": 0.012513563968241215, "time": 0.7223661422729493, "epoch": 6, "memory": 15585, "step": 11880} +{"lr": 0.0001549429338961849, "data_time": 0.17857692241668702, "loss": 0.007475029630586505, "time": 0.7075620412826538, "epoch": 6, "memory": 15585, "step": 11980} +{"lr": 0.00015206911227631342, "data_time": 0.22871408462524415, "loss": 0.012053997442126273, "time": 0.7575007200241088, "epoch": 6, "memory": 15585, "step": 12080} +{"lr": 0.00014920764882892865, "data_time": 0.20356442928314208, "loss": 0.007269883435219527, "time": 0.7312607049942017, "epoch": 6, "memory": 15585, "step": 12180} +{"lr": 0.00014635921164142459, "data_time": 0.16637322902679444, "loss": 0.011504888720810413, "time": 0.6937987804412842, "epoch": 6, "memory": 15585, "step": 12280} +{"accuracy/top1": 99.47702026367188, "data_time": 0.08086108379676694, "time": 0.47901404099386247, "step": 6} +{"lr": 0.00014194323525930575, "data_time": 0.12070648670196533, "loss": 0.008615901321172714, "time": 0.6390037536621094, "epoch": 7, "memory": 15585, "step": 12436} +{"lr": 0.0001391311681531885, "data_time": 0.19333932399749756, "loss": 0.006534418370574713, "time": 0.7094549655914306, "epoch": 7, "memory": 15585, "step": 12536} +{"lr": 0.00013633447993868745, "data_time": 0.183227014541626, "loss": 0.012832779809832573, "time": 0.7016785621643067, "epoch": 7, "memory": 15585, "step": 12636} +{"lr": 0.00013355382357963566, "data_time": 0.17696599960327147, "loss": 0.008349302085116505, "time": 0.6941949129104614, "epoch": 7, "memory": 15585, "step": 12736} +{"lr": 0.0001307898482967891, "data_time": 0.1283672571182251, "loss": 0.005931426817551255, "time": 0.6420610189437866, "epoch": 7, "memory": 15585, "step": 12836} +{"lr": 0.00012804319941624686, "data_time": 0.13395004272460936, "loss": 0.006299674604088068, "time": 0.6475434303283691, "epoch": 7, "memory": 15585, "step": 12936} +{"lr": 0.00012531451821878227, "data_time": 0.1922152042388916, "loss": 0.004753225343301892, "time": 0.7181339263916016, "epoch": 7, "memory": 15585, "step": 13036} +{"lr": 0.00012260444179011874, "data_time": 0.19852476119995116, "loss": 0.0071558387950062755, "time": 0.7137287616729736, "epoch": 7, "memory": 15585, "step": 13136} +{"lr": 0.00011991360287218363, "data_time": 0.19275262355804443, "loss": 0.010148387541994453, "time": 0.7107342004776, "epoch": 7, "memory": 15585, "step": 13236} +{"lr": 0.00011724262971537941, "data_time": 0.17831885814666748, "loss": 0.0077144072856754065, "time": 0.6920490741729737, "epoch": 7, "memory": 15585, "step": 13336} +{"lr": 0.00011459214593189938, "data_time": 0.1449887275695801, "loss": 0.005640890542417764, "time": 0.6591064453125, "epoch": 7, "memory": 15585, "step": 13436} +{"lr": 0.00011196277035012846, "data_time": 0.2257547378540039, "loss": 0.004714374570176006, "time": 0.7451757669448853, "epoch": 7, "memory": 15585, "step": 13536} +{"lr": 0.0001093551168701611, "data_time": 0.16027560234069824, "loss": 0.01077058119699359, "time": 0.6749431848526001, "epoch": 7, "memory": 15585, "step": 13636} +{"lr": 0.00010676979432046926, "data_time": 0.20957491397857667, "loss": 0.007215554779395461, "time": 0.7285349369049072, "epoch": 7, "memory": 15585, "step": 13736} +{"lr": 0.00010420740631575464, "data_time": 0.22608597278594972, "loss": 0.010731688188388943, "time": 0.7457456111907959, "epoch": 7, "memory": 15585, "step": 13836} +{"lr": 0.00010166855111601829, "data_time": 0.20041840076446532, "loss": 0.007292342046275735, "time": 0.7210818529129028, "epoch": 7, "memory": 15585, "step": 13936} +{"lr": 9.91538214868799e-05, "data_time": 0.019598913192749024, "loss": 0.0029966373927891254, "time": 0.7008754014968872, "epoch": 7, "memory": 15585, "step": 14036} +{"lr": 9.666380456118125e-05, "data_time": 0.0016842126846313477, "loss": 0.007061452884227037, "time": 0.7150977373123169, "epoch": 7, "memory": 15585, "step": 14136} +{"lr": 9.419908170190298e-05, "data_time": 0.001611781120300293, "loss": 0.007744133938103914, "time": 0.6945077180862427, "epoch": 7, "memory": 15585, "step": 14236} +{"lr": 9.176022836643002e-05, "data_time": 0.0016968727111816406, "loss": 0.004769612848758697, "time": 0.7431050062179565, "epoch": 7, "memory": 15585, "step": 14336} +{"accuracy/top1": 99.59452819824219, "data_time": 0.09648423507565358, "time": 0.4932242768709777, "step": 7} +{"lr": 8.80086206394144e-05, "data_time": 0.0015840768814086915, "loss": 0.01414155294187367, "time": 0.7576480627059936, "epoch": 8, "memory": 15585, "step": 14492} +{"lr": 8.563857333566659e-05, "data_time": 0.00155029296875, "loss": 0.0069256947375833985, "time": 0.6686500549316406, "epoch": 8, "memory": 15585, "step": 14592} +{"lr": 8.329639424200186e-05, "data_time": 0.0015752553939819337, "loss": 0.0034400981850922106, "time": 0.7806374073028565, "epoch": 8, "memory": 15585, "step": 14692} +{"lr": 8.098263020459022e-05, "data_time": 0.0015067815780639648, "loss": 0.008589015481993556, "time": 0.6729009866714477, "epoch": 8, "memory": 15585, "step": 14792} +{"lr": 7.869782143532516e-05, "data_time": 0.0015240192413330078, "loss": 0.008856590837240219, "time": 0.6439080715179444, "epoch": 8, "memory": 15585, "step": 14892} +{"lr": 7.644250138569548e-05, "data_time": 0.0021954774856567383, "loss": 0.004469246277585626, "time": 0.6901609182357789, "epoch": 8, "memory": 15585, "step": 14992} +{"lr": 7.4217196622237e-05, "data_time": 0.0017419815063476562, "loss": 0.008263613749295472, "time": 0.678468656539917, "epoch": 8, "memory": 15585, "step": 15092} +{"lr": 7.202242670359085e-05, "data_time": 0.0017836332321166993, "loss": 0.00724510052241385, "time": 0.7328895092010498, "epoch": 8, "memory": 15585, "step": 15192} +{"lr": 6.98587040591985e-05, "data_time": 0.0019298315048217774, "loss": 0.007897790055721998, "time": 0.6909587860107422, "epoch": 8, "memory": 15585, "step": 15292} +{"lr": 6.772653386966062e-05, "data_time": 0.0017855167388916016, "loss": 0.0041174809448421, "time": 0.7238125324249267, "epoch": 8, "memory": 15585, "step": 15392} +{"lr": 6.562641394878866e-05, "data_time": 0.0018667936325073241, "loss": 0.0037497988902032374, "time": 0.7481331586837768, "epoch": 8, "memory": 15585, "step": 15492} +{"lr": 6.35588346273771e-05, "data_time": 0.0016849994659423827, "loss": 0.004643295519053936, "time": 0.7302852153778077, "epoch": 8, "memory": 15585, "step": 15592} +{"lr": 6.152427863872164e-05, "data_time": 0.0017604827880859375, "loss": 0.004337272793054581, "time": 0.7296119213104248, "epoch": 8, "memory": 15585, "step": 15692} +{"lr": 5.952322100591216e-05, "data_time": 0.0016360759735107421, "loss": 0.006246315967291594, "time": 0.7323440790176392, "epoch": 8, "memory": 15585, "step": 15792} +{"lr": 5.755612893092559e-05, "data_time": 0.0016683340072631836, "loss": 0.004080366250127554, "time": 0.7764551639556885, "epoch": 8, "memory": 15585, "step": 15892} +{"lr": 5.5623461685544754e-05, "data_time": 0.0018137454986572265, "loss": 0.00618846919387579, "time": 0.6559581518173218, "epoch": 8, "memory": 15585, "step": 15992} +{"lr": 5.37256705041282e-05, "data_time": 0.0016515731811523437, "loss": 0.006456415681168437, "time": 0.6898155212402344, "epoch": 8, "memory": 15585, "step": 16092} +{"lr": 5.186319847825775e-05, "data_time": 0.0017484426498413086, "loss": 0.009180582035332919, "time": 0.6893758296966552, "epoch": 8, "memory": 15585, "step": 16192} +{"lr": 5.0036480453286144e-05, "data_time": 0.0016029834747314452, "loss": 0.008056677551940084, "time": 0.7109777450561523, "epoch": 8, "memory": 15585, "step": 16292} +{"lr": 4.8245942926810745e-05, "data_time": 0.001706099510192871, "loss": 0.0028826996218413114, "time": 0.6360068798065186, "epoch": 8, "memory": 15585, "step": 16392} +{"accuracy/top1": 99.66238403320312, "data_time": 0.09673175655427527, "time": 0.49555478330518377, "step": 8} +{"lr": 4.5525937358206823e-05, "data_time": 0.16450128555297852, "loss": 0.003932453924790025, "time": 0.6787249088287354, "epoch": 9, "memory": 15585, "step": 16548} +{"lr": 4.382990660001016e-05, "data_time": 0.16617732048034667, "loss": 0.006312030553817749, "time": 0.6840532541275024, "epoch": 9, "memory": 15585, "step": 16648} +{"lr": 4.217150543579113e-05, "data_time": 0.1792240858078003, "loss": 0.005301159713417292, "time": 0.6963784217834472, "epoch": 9, "memory": 15585, "step": 16748} +{"lr": 4.0551121064943564e-05, "data_time": 0.1785569190979004, "loss": 0.009223576216027141, "time": 0.6942339658737182, "epoch": 9, "memory": 15585, "step": 16848} +{"lr": 3.896913181079464e-05, "data_time": 0.18540642261505128, "loss": 0.00460140579380095, "time": 0.70242919921875, "epoch": 9, "memory": 15585, "step": 16948} +{"lr": 3.742590703227469e-05, "data_time": 0.20136003494262694, "loss": 0.007680822629481554, "time": 0.7193239688873291, "epoch": 9, "memory": 15585, "step": 17048} +{"lr": 3.592180703768032e-05, "data_time": 0.18309106826782226, "loss": 0.008242048975080251, "time": 0.6988625764846802, "epoch": 9, "memory": 15585, "step": 17148} +{"lr": 3.445718300055071e-05, "data_time": 0.18934359550476074, "loss": 0.006236546812579036, "time": 0.7060253858566284, "epoch": 9, "memory": 15585, "step": 17248} +{"lr": 3.3032376877676173e-05, "data_time": 0.17684197425842285, "loss": 0.00680283117108047, "time": 0.6924414873123169, "epoch": 9, "memory": 15585, "step": 17348} +{"lr": 3.1647721329259246e-05, "data_time": 0.21000983715057372, "loss": 0.005297743622213602, "time": 0.7288161277770996, "epoch": 9, "memory": 15585, "step": 17448} +{"lr": 3.0303539641245756e-05, "data_time": 0.19528508186340332, "loss": 0.004597390862181782, "time": 0.7111314296722412, "epoch": 9, "memory": 15585, "step": 17548} +{"lr": 2.9000145649844975e-05, "data_time": 0.21944591999053956, "loss": 0.008540700143203139, "time": 0.7567103624343872, "epoch": 9, "memory": 15585, "step": 17648} +{"lr": 2.7737843668255695e-05, "data_time": 0.17396090030670167, "loss": 0.002788296481594443, "time": 0.6878469944000244, "epoch": 9, "memory": 15585, "step": 17748} +{"lr": 2.6516928415616184e-05, "data_time": 0.18790993690490723, "loss": 0.007269242499023676, "time": 0.7097814798355102, "epoch": 9, "memory": 15585, "step": 17848} +{"lr": 2.5337684948193946e-05, "data_time": 0.2300894021987915, "loss": 0.005430571408942342, "time": 0.744309401512146, "epoch": 9, "memory": 15585, "step": 17948} +{"lr": 2.4200388592830947e-05, "data_time": 0.19214770793914795, "loss": 0.006849743146449328, "time": 0.7057928800582886, "epoch": 9, "memory": 15585, "step": 18048} +{"lr": 2.3105304882661933e-05, "data_time": 0.25373048782348634, "loss": 0.006313639925792813, "time": 0.7711594104766846, "epoch": 9, "memory": 15585, "step": 18148} +{"lr": 2.20526894951175e-05, "data_time": 0.16188671588897705, "loss": 0.004326836299151182, "time": 0.6790915250778198, "epoch": 9, "memory": 15585, "step": 18248} +{"lr": 2.1042788192229725e-05, "data_time": 0.1542069673538208, "loss": 0.005015872605144978, "time": 0.667760705947876, "epoch": 9, "memory": 15585, "step": 18348} +{"lr": 2.007583676325189e-05, "data_time": 0.6659237384796143, "loss": 0.008820148324593902, "time": 1.188455843925476, "epoch": 9, "memory": 15585, "step": 18448} +{"accuracy/top1": 99.74016571044922, "data_time": 0.08351332633221736, "time": 0.4794860472444628, "step": 9} +{"lr": 1.8653686449507784e-05, "data_time": 0.18245489597320558, "loss": 0.004861492803320289, "time": 0.6963652849197388, "epoch": 10, "memory": 15585, "step": 18604} +{"lr": 1.7797691544415563e-05, "data_time": 0.1789107084274292, "loss": 0.003677115123718977, "time": 0.696782398223877, "epoch": 10, "memory": 15585, "step": 18704} +{"lr": 1.6985404170356326e-05, "data_time": 0.1933534860610962, "loss": 0.0043910601641982796, "time": 0.7105748653411865, "epoch": 10, "memory": 15585, "step": 18804} +{"lr": 1.6217013978172993e-05, "data_time": 0.21540849208831786, "loss": 0.003004590841010213, "time": 0.7314217567443848, "epoch": 10, "memory": 15585, "step": 18904} +{"lr": 1.549270036970338e-05, "data_time": 0.21487078666687012, "loss": 0.006738194124773144, "time": 0.7328275918960572, "epoch": 10, "memory": 15585, "step": 19004} +{"lr": 1.4812632455893675e-05, "data_time": 0.15540411472320556, "loss": 0.006134961266070604, "time": 0.6692819595336914, "epoch": 10, "memory": 15585, "step": 19104} +{"lr": 1.4176969017315005e-05, "data_time": 0.18832159042358398, "loss": 0.004016678966581822, "time": 0.7060202121734619, "epoch": 10, "memory": 15585, "step": 19204} +{"lr": 1.358585846709161e-05, "data_time": 0.16427857875823976, "loss": 0.00385310766287148, "time": 0.6779886960983277, "epoch": 10, "memory": 15585, "step": 19304} +{"lr": 1.3039438816249855e-05, "data_time": 0.18878569602966308, "loss": 0.005551433609798551, "time": 0.7067301750183106, "epoch": 10, "memory": 15585, "step": 19404} +{"lr": 1.2537837641495561e-05, "data_time": 0.22544901371002196, "loss": 0.004782603727653623, "time": 0.7426913499832153, "epoch": 10, "memory": 15585, "step": 19504} +{"lr": 1.2081172055427927e-05, "data_time": 0.21164205074310302, "loss": 0.004750721715390682, "time": 0.7287100076675415, "epoch": 10, "memory": 15585, "step": 19604} +{"lr": 1.1669548679196218e-05, "data_time": 0.18330190181732178, "loss": 0.004795972025021911, "time": 0.7001710653305053, "epoch": 10, "memory": 15585, "step": 19704} +{"lr": 1.130306361760621e-05, "data_time": 0.19480786323547364, "loss": 0.004976117424666882, "time": 0.7107309579849244, "epoch": 10, "memory": 15585, "step": 19804} +{"lr": 1.0981802436681767e-05, "data_time": 0.18398218154907225, "loss": 0.005396238854154944, "time": 0.701444411277771, "epoch": 10, "memory": 15585, "step": 19904} +{"lr": 1.0705840143687269e-05, "data_time": 0.2710393190383911, "loss": 0.005424016620963812, "time": 0.7881027221679687, "epoch": 10, "memory": 15585, "step": 20004} +{"lr": 1.0475241169614925e-05, "data_time": 0.1448136329650879, "loss": 0.006270120851695538, "time": 0.6615827798843383, "epoch": 10, "memory": 15585, "step": 20104} +{"lr": 1.0290059354141742e-05, "data_time": 0.17011544704437256, "loss": 0.0036644930019974708, "time": 0.6838934183120727, "epoch": 10, "memory": 15585, "step": 20204} +{"lr": 1.015033793305906e-05, "data_time": 0.18959639072418213, "loss": 0.0066305665299296376, "time": 0.7083418846130372, "epoch": 10, "memory": 15585, "step": 20304} +{"lr": 1.0056109528177973e-05, "data_time": 0.18329384326934814, "loss": 0.004180255625396967, "time": 0.7005026340484619, "epoch": 10, "memory": 15585, "step": 20404} +{"lr": 1.0007396139712939e-05, "data_time": 0.1671433687210083, "loss": 0.0035097086802124977, "time": 0.6808896064758301, "epoch": 10, "memory": 15585, "step": 20504} +{"accuracy/top1": 99.76002502441406, "data_time": 0.09428098944366956, "time": 0.48892652011308513, "step": 10} diff --git a/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1/20230605_024312/vis_data/vis_image/0000000.png_1.png b/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1/20230605_024312/vis_data/vis_image/0000000.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..5567c0d70ccfa2468ccadc0d554edec146ac6e28 Binary files /dev/null and b/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1/20230605_024312/vis_data/vis_image/0000000.png_1.png differ diff --git a/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1/20230605_024312/vis_data/vis_image/0000000.png_10.png b/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1/20230605_024312/vis_data/vis_image/0000000.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..c82fd6e60100bb990ed3a355185b54fcc4015307 Binary files /dev/null and b/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1/20230605_024312/vis_data/vis_image/0000000.png_10.png differ diff --git a/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1/20230605_024312/vis_data/vis_image/0000000.png_2.png b/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1/20230605_024312/vis_data/vis_image/0000000.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..e41dd59bd7d2750959202b92f15d2113f3dff74f Binary files /dev/null and b/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1/20230605_024312/vis_data/vis_image/0000000.png_2.png differ diff --git a/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1/20230605_024312/vis_data/vis_image/0000000.png_3.png b/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1/20230605_024312/vis_data/vis_image/0000000.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..a8dcb073269182adadc577c21a55776a94b742a2 Binary files /dev/null and b/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1/20230605_024312/vis_data/vis_image/0000000.png_3.png differ diff --git a/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1/20230605_024312/vis_data/vis_image/0000000.png_4.png b/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1/20230605_024312/vis_data/vis_image/0000000.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..8a663f407ef34f5219ece375ed1fee151bf8c402 Binary files /dev/null and b/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1/20230605_024312/vis_data/vis_image/0000000.png_4.png differ diff --git a/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1/20230605_024312/vis_data/vis_image/0000000.png_5.png b/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1/20230605_024312/vis_data/vis_image/0000000.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..fcd99b44aacd1eabc862c2ff285e2bf403227776 Binary files /dev/null and b/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1/20230605_024312/vis_data/vis_image/0000000.png_5.png differ diff --git a/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1/20230605_024312/vis_data/vis_image/0000000.png_6.png b/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1/20230605_024312/vis_data/vis_image/0000000.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..b3df00232632ffd18b2326baad17ac47e0355d75 Binary files /dev/null and b/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1/20230605_024312/vis_data/vis_image/0000000.png_6.png differ diff --git a/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1/20230605_024312/vis_data/vis_image/0000000.png_7.png b/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1/20230605_024312/vis_data/vis_image/0000000.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..e5fd21f1b15dc5e93b8f215c4b34a6c67868e0e0 Binary files /dev/null and b/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1/20230605_024312/vis_data/vis_image/0000000.png_7.png differ diff --git a/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1/20230605_024312/vis_data/vis_image/0000000.png_8.png b/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1/20230605_024312/vis_data/vis_image/0000000.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..13921ae169ce41329c3d4ebfa8271298a2a52cdb Binary files /dev/null and b/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1/20230605_024312/vis_data/vis_image/0000000.png_8.png differ diff --git a/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1/20230605_024312/vis_data/vis_image/0000000.png_9.png b/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1/20230605_024312/vis_data/vis_image/0000000.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..11765e803b3907e5633072aedfbf51d85a062f8d Binary files /dev/null and b/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1/20230605_024312/vis_data/vis_image/0000000.png_9.png differ diff --git a/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1/20230605_024312/vis_data/vis_image/000015337.jpg_1.png b/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1/20230605_024312/vis_data/vis_image/000015337.jpg_1.png new file mode 100644 index 0000000000000000000000000000000000000000..06c973d00ef5a554560abb879437ddad27a294a2 Binary files /dev/null and b/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1/20230605_024312/vis_data/vis_image/000015337.jpg_1.png differ diff --git a/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1/20230605_024312/vis_data/vis_image/000015337.jpg_10.png b/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1/20230605_024312/vis_data/vis_image/000015337.jpg_10.png new file mode 100644 index 0000000000000000000000000000000000000000..e5e06f2c55bd78cfc781c7b00e0bb3bcc79c55be Binary files /dev/null and b/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1/20230605_024312/vis_data/vis_image/000015337.jpg_10.png differ diff --git a/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1/20230605_024312/vis_data/vis_image/000015337.jpg_2.png b/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1/20230605_024312/vis_data/vis_image/000015337.jpg_2.png new file mode 100644 index 0000000000000000000000000000000000000000..63d0a6e0f001f0240b83a170a877ec9a8d3982b5 Binary files /dev/null and b/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1/20230605_024312/vis_data/vis_image/000015337.jpg_2.png differ diff --git a/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1/20230605_024312/vis_data/vis_image/000015337.jpg_3.png b/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1/20230605_024312/vis_data/vis_image/000015337.jpg_3.png new file mode 100644 index 0000000000000000000000000000000000000000..ac52f7ccb2492251769bc0cc583c0bb398539460 Binary files /dev/null and b/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1/20230605_024312/vis_data/vis_image/000015337.jpg_3.png differ diff --git a/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1/20230605_024312/vis_data/vis_image/000015337.jpg_4.png b/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1/20230605_024312/vis_data/vis_image/000015337.jpg_4.png new file mode 100644 index 0000000000000000000000000000000000000000..443af6ed4727770050bae7be157b17668a76342c Binary files /dev/null and b/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1/20230605_024312/vis_data/vis_image/000015337.jpg_4.png differ diff --git a/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1/20230605_024312/vis_data/vis_image/000015337.jpg_5.png b/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1/20230605_024312/vis_data/vis_image/000015337.jpg_5.png new file mode 100644 index 0000000000000000000000000000000000000000..b23194430f22112ed0263ef032c12a17113fc5e4 Binary files /dev/null and b/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1/20230605_024312/vis_data/vis_image/000015337.jpg_5.png differ diff --git a/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1/20230605_024312/vis_data/vis_image/000015337.jpg_6.png b/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1/20230605_024312/vis_data/vis_image/000015337.jpg_6.png new file mode 100644 index 0000000000000000000000000000000000000000..f43787b349fcd0b56b3010fcb606c096bf8b7799 Binary files /dev/null and b/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1/20230605_024312/vis_data/vis_image/000015337.jpg_6.png differ diff --git a/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1/20230605_024312/vis_data/vis_image/000015337.jpg_7.png b/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1/20230605_024312/vis_data/vis_image/000015337.jpg_7.png new file mode 100644 index 0000000000000000000000000000000000000000..d7aee6575387edc0353cd61dd8ada580f0b26e4b Binary files /dev/null and b/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1/20230605_024312/vis_data/vis_image/000015337.jpg_7.png differ diff --git a/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1/20230605_024312/vis_data/vis_image/000015337.jpg_8.png b/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1/20230605_024312/vis_data/vis_image/000015337.jpg_8.png new file mode 100644 index 0000000000000000000000000000000000000000..7672dac3e084bf16a2af7cd0f1446fd3bbfe32ed Binary files /dev/null and b/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1/20230605_024312/vis_data/vis_image/000015337.jpg_8.png differ diff --git a/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1/20230605_024312/vis_data/vis_image/000015337.jpg_9.png b/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1/20230605_024312/vis_data/vis_image/000015337.jpg_9.png new file mode 100644 index 0000000000000000000000000000000000000000..490702c9a08704b8a633d71531a0b1ce851b8d21 Binary files /dev/null and b/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1/20230605_024312/vis_data/vis_image/000015337.jpg_9.png differ diff --git a/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1/20230605_024312/vis_data/vis_image/0004159.png_1.png b/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1/20230605_024312/vis_data/vis_image/0004159.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..e4b6cebe9b9e8981670204a359e4abf8b0b2de30 Binary files /dev/null and b/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1/20230605_024312/vis_data/vis_image/0004159.png_1.png differ diff --git a/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1/20230605_024312/vis_data/vis_image/0004159.png_10.png b/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1/20230605_024312/vis_data/vis_image/0004159.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..7630ee67b9bd3e65f294b6772d84768b0014dcf3 Binary files /dev/null and b/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1/20230605_024312/vis_data/vis_image/0004159.png_10.png differ diff --git a/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1/20230605_024312/vis_data/vis_image/0004159.png_2.png b/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1/20230605_024312/vis_data/vis_image/0004159.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..1bdbfd43f352faee3796e9b63654df30b2ed2628 Binary files /dev/null and b/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1/20230605_024312/vis_data/vis_image/0004159.png_2.png differ diff --git a/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1/20230605_024312/vis_data/vis_image/0004159.png_3.png b/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1/20230605_024312/vis_data/vis_image/0004159.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..35055a838842391254752e17c12431ebd9af9a4c Binary files /dev/null and b/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1/20230605_024312/vis_data/vis_image/0004159.png_3.png differ diff --git a/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1/20230605_024312/vis_data/vis_image/0004159.png_4.png b/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1/20230605_024312/vis_data/vis_image/0004159.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..a97ba3c263f259a69cf0dd40c2d708e4120a49d0 Binary files /dev/null and b/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1/20230605_024312/vis_data/vis_image/0004159.png_4.png differ diff --git a/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1/20230605_024312/vis_data/vis_image/0004159.png_5.png b/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1/20230605_024312/vis_data/vis_image/0004159.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..c7ca96931a45f64c58b9dc4452a9601499e63558 Binary files /dev/null and b/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1/20230605_024312/vis_data/vis_image/0004159.png_5.png differ diff --git a/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1/20230605_024312/vis_data/vis_image/0004159.png_6.png b/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1/20230605_024312/vis_data/vis_image/0004159.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..1a81a968d546754e435b8ddfbb447715c271117b Binary files /dev/null and b/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1/20230605_024312/vis_data/vis_image/0004159.png_6.png differ diff --git a/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1/20230605_024312/vis_data/vis_image/0004159.png_7.png b/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1/20230605_024312/vis_data/vis_image/0004159.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..ff33a4b4be545f4f31984bc8ebc5bdd9b73185c7 Binary files /dev/null and b/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1/20230605_024312/vis_data/vis_image/0004159.png_7.png differ diff --git a/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1/20230605_024312/vis_data/vis_image/0004159.png_8.png b/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1/20230605_024312/vis_data/vis_image/0004159.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..ec90354595ec26ba6c286e54427bbae9f5b2962d Binary files /dev/null and b/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1/20230605_024312/vis_data/vis_image/0004159.png_8.png differ diff --git a/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1/20230605_024312/vis_data/vis_image/0004159.png_9.png b/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1/20230605_024312/vis_data/vis_image/0004159.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..2d62e529cdc06365344874cf61e93a1fa2c54392 Binary files /dev/null and b/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1/20230605_024312/vis_data/vis_image/0004159.png_9.png differ diff --git a/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1/20230605_024312/vis_data/vis_image/0008318.png_1.png b/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1/20230605_024312/vis_data/vis_image/0008318.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..b06b919c62836463430ad3a48e061470c4087db0 Binary files /dev/null and b/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1/20230605_024312/vis_data/vis_image/0008318.png_1.png differ diff --git a/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1/20230605_024312/vis_data/vis_image/0008318.png_10.png b/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1/20230605_024312/vis_data/vis_image/0008318.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..6c61e3e43a9aaed73835fb5a37dcdcdd7b9b2354 Binary files /dev/null and b/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1/20230605_024312/vis_data/vis_image/0008318.png_10.png differ diff --git a/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1/20230605_024312/vis_data/vis_image/0008318.png_2.png b/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1/20230605_024312/vis_data/vis_image/0008318.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..6e233146aa3818bc8ea92b85566102ebd5a22073 Binary files /dev/null and b/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1/20230605_024312/vis_data/vis_image/0008318.png_2.png differ diff --git a/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1/20230605_024312/vis_data/vis_image/0008318.png_3.png b/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1/20230605_024312/vis_data/vis_image/0008318.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..fc0375d853f66b8aae779a5e003b694581069ccc Binary files /dev/null and b/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1/20230605_024312/vis_data/vis_image/0008318.png_3.png differ diff --git a/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1/20230605_024312/vis_data/vis_image/0008318.png_4.png b/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1/20230605_024312/vis_data/vis_image/0008318.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..1be059d0136153c2ac58fc6d9d0ac35ebbded33e Binary files /dev/null and b/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1/20230605_024312/vis_data/vis_image/0008318.png_4.png differ diff --git a/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1/20230605_024312/vis_data/vis_image/0008318.png_5.png b/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1/20230605_024312/vis_data/vis_image/0008318.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..7a322fe96542dcfc3d4384412d8e5061f2985a99 Binary files /dev/null and b/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1/20230605_024312/vis_data/vis_image/0008318.png_5.png differ diff --git a/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1/20230605_024312/vis_data/vis_image/0008318.png_6.png b/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1/20230605_024312/vis_data/vis_image/0008318.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..9da772a569e792b010f3ceff489e9766f179c66c Binary files /dev/null and b/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1/20230605_024312/vis_data/vis_image/0008318.png_6.png differ diff --git a/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1/20230605_024312/vis_data/vis_image/0008318.png_7.png b/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1/20230605_024312/vis_data/vis_image/0008318.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..e2b2fa82eb0ccf1203282ccbb51183513237ed32 Binary files /dev/null and b/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1/20230605_024312/vis_data/vis_image/0008318.png_7.png differ diff --git a/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1/20230605_024312/vis_data/vis_image/0008318.png_8.png b/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1/20230605_024312/vis_data/vis_image/0008318.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..5a9675330688f5222e1bd6054c67a3e501250ed0 Binary files /dev/null and b/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1/20230605_024312/vis_data/vis_image/0008318.png_8.png differ diff --git a/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1/20230605_024312/vis_data/vis_image/0008318.png_9.png b/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1/20230605_024312/vis_data/vis_image/0008318.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..43fd9afeecb0c18e8fdf1b38dedc00e497b61422 Binary files /dev/null and b/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1/20230605_024312/vis_data/vis_image/0008318.png_9.png differ diff --git a/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1.py b/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1.py new file mode 100644 index 0000000000000000000000000000000000000000..c1d367f5d6e5dc2c7a25ab22cbbe5fda918650bd --- /dev/null +++ b/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1.py @@ -0,0 +1,383 @@ +optim_wrapper = dict( + optimizer=dict( + type='AdamW', + lr=0.0004, + weight_decay=0.05, + eps=1e-08, + betas=(0.9, 0.999), + _scope_='mmpretrain'), + paramwise_cfg=dict( + norm_decay_mult=0.0, + bias_decay_mult=0.0, + flat_decay_mult=0.0, + custom_keys=dict({ + '.absolute_pos_embed': dict(decay_mult=0.0), + '.relative_position_bias_table': dict(decay_mult=0.0) + })), + type='AmpOptimWrapper', + dtype='bfloat16', + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=False, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=4096) +model = dict( + type='ImageClassifier', + backbone=dict( + frozen_stages=2, + type='ConvNeXt', + arch='small', + drop_path_rate=0.4, + init_cfg=dict( + type='Pretrained', + checkpoint= + 'https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth', + prefix='backbone')), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=768, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + init_cfg=None), + init_cfg=dict( + type='TruncNormal', layer=['Conv2d', 'Linear'], std=0.02, bias=0.0), + train_cfg=None) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/all1.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/all2.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/all3.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3fake8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/cc1m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3real8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV2-1-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/if-dpmsolver++-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV2-1-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/if-dpmsolver++-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1' diff --git a/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1/epoch_1.pth b/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1/epoch_1.pth new file mode 100644 index 0000000000000000000000000000000000000000..1e24b1390e96a048e512588f6b31eac367cab5b1 --- /dev/null +++ b/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1/epoch_1.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:43555f64a9a1bac1ea88efbdc78cff83b17074972fe5534ccdb29c41afab4686 +size 782130778 diff --git a/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1/epoch_10.pth b/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1/epoch_10.pth new file mode 100644 index 0000000000000000000000000000000000000000..380d8a4ef843456cd53e2c291c948bb229d53d82 --- /dev/null +++ b/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1/epoch_10.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:d5ebc583b7fb0d5b265c357b41348fe7fda04d291150cc49d5fba60436dfd859 +size 783486234 diff --git a/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1/epoch_2.pth b/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1/epoch_2.pth new file mode 100644 index 0000000000000000000000000000000000000000..a56d1cf5db7ff6b0e304be45f5d82b5c1c72a549 --- /dev/null +++ b/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1/epoch_2.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:e805269ef2c73dd25ae3915d456f38033351ddd582d6d736ecd8899022f08472 +size 782282714 diff --git a/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1/epoch_3.pth b/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1/epoch_3.pth new file mode 100644 index 0000000000000000000000000000000000000000..27c6b7816a857cb2343b356b01bb8b4e56146ede --- /dev/null +++ b/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1/epoch_3.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:30ce832ce7467754ca83cf072768bf774d274baabbd93062fb910209229ebcdb +size 782433370 diff --git a/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1/epoch_4.pth b/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1/epoch_4.pth new file mode 100644 index 0000000000000000000000000000000000000000..03996b906cc6aee021fcd697f6988a86e23ba7b5 --- /dev/null +++ b/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1/epoch_4.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:e39400488d8a81433fc5430da7ff3a910689e58116080218833eefbc8c40c7e6 +size 782583962 diff --git a/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1/epoch_5.pth b/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1/epoch_5.pth new file mode 100644 index 0000000000000000000000000000000000000000..4b2e55a966a8078873130aa1f05844db939d462b --- /dev/null +++ b/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1/epoch_5.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:fc7afa677ee038724751445ddc880dc050e70ac5dddb27806b03d548f69fda78 +size 782733978 diff --git a/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1/epoch_6.pth b/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1/epoch_6.pth new file mode 100644 index 0000000000000000000000000000000000000000..0870973b61ba66ef926d7a034b0d6e846e29c634 --- /dev/null +++ b/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1/epoch_6.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:e71b4cda1a3d01ecae673cfcdfb1032a7e9e03ebc9540a151c9ca39fd2cbbe0a +size 782884250 diff --git a/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1/epoch_7.pth b/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1/epoch_7.pth new file mode 100644 index 0000000000000000000000000000000000000000..2a91a3054cd953f33de57885659f5e69d274fc6e --- /dev/null +++ b/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1/epoch_7.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:1b58fac86c3ff811bacf81e42eb183fc009621ee588dfeace3722c771348cd75 +size 783034330 diff --git a/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1/epoch_8.pth b/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1/epoch_8.pth new file mode 100644 index 0000000000000000000000000000000000000000..6c86ebf4872b9ec5b490f9a06fe951245c5ae748 --- /dev/null +++ b/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1/epoch_8.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:3b4e44b43b4d1a32d730f0cf015bf172018b5b0d838dd9c6d91f215cd738eb2f +size 783183962 diff --git a/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1/epoch_9.pth b/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1/epoch_9.pth new file mode 100644 index 0000000000000000000000000000000000000000..2e65a165874c7d385dac5c8ae80ec26a91bfae0c --- /dev/null +++ b/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1/epoch_9.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:81bcf272e74e988b88af9346ebdf0b010fc6030a7154e0e08728fc9ac8dad844 +size 783334874 diff --git a/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1/last_checkpoint b/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1/last_checkpoint new file mode 100644 index 0000000000000000000000000000000000000000..f836d61ab108e8a82245e8176f3466ed99457755 --- /dev/null +++ b/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1/last_checkpoint @@ -0,0 +1 @@ +/mnt/petrelfs/luzeyu/workspace/fakebench/mmpretrain/workdir/convnext_small_4xb256_all_1m_lr4e-4_aug_1e-1/epoch_10.pth \ No newline at end of file diff --git a/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1/20230605_024744/20230605_024744.log b/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1/20230605_024744/20230605_024744.log new file mode 100644 index 0000000000000000000000000000000000000000..4a7471b9601967a5096def43d9a7c8fa2bb846c7 --- /dev/null +++ b/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1/20230605_024744/20230605_024744.log @@ -0,0 +1,1987 @@ +2023/06/05 02:47:49 - mmengine - INFO - +------------------------------------------------------------ +System environment: + sys.platform: linux + Python: 3.10.9 (main, Mar 8 2023, 10:47:38) [GCC 11.2.0] + CUDA available: True + numpy_random_seed: 914339228 + GPU 0,1,2,3: NVIDIA A100-SXM4-80GB + CUDA_HOME: /mnt/petrelfs/share/cuda-11.6 + NVCC: Cuda compilation tools, release 11.6, V11.6.124 + GCC: gcc (GCC) 7.5.0 + PyTorch: 1.13.1 + PyTorch compiling details: PyTorch built with: + - GCC 9.3 + - C++ Version: 201402 + - Intel(R) oneAPI Math Kernel Library Version 2021.4-Product Build 20210904 for Intel(R) 64 architecture applications + - Intel(R) MKL-DNN v2.6.0 (Git Hash 52b5f107dd9cf10910aaa19cb47f3abf9b349815) + - OpenMP 201511 (a.k.a. OpenMP 4.5) + - LAPACK is enabled (usually provided by MKL) + - NNPACK is enabled + - CPU capability usage: AVX2 + - CUDA Runtime 11.6 + - NVCC architecture flags: -gencode;arch=compute_37,code=sm_37;-gencode;arch=compute_50,code=sm_50;-gencode;arch=compute_60,code=sm_60;-gencode;arch=compute_61,code=sm_61;-gencode;arch=compute_70,code=sm_70;-gencode;arch=compute_75,code=sm_75;-gencode;arch=compute_80,code=sm_80;-gencode;arch=compute_86,code=sm_86;-gencode;arch=compute_37,code=compute_37 + - CuDNN 8.3.2 (built against CUDA 11.5) + - Magma 2.6.1 + - Build settings: BLAS_INFO=mkl, BUILD_TYPE=Release, CUDA_VERSION=11.6, CUDNN_VERSION=8.3.2, CXX_COMPILER=/opt/rh/devtoolset-9/root/usr/bin/c++, CXX_FLAGS= -fabi-version=11 -Wno-deprecated -fvisibility-inlines-hidden -DUSE_PTHREADPOOL -fopenmp -DNDEBUG -DUSE_KINETO -DUSE_FBGEMM -DUSE_QNNPACK -DUSE_PYTORCH_QNNPACK -DUSE_XNNPACK -DSYMBOLICATE_MOBILE_DEBUG_HANDLE -DEDGE_PROFILER_USE_KINETO -O2 -fPIC -Wno-narrowing -Wall -Wextra -Werror=return-type -Werror=non-virtual-dtor -Wno-missing-field-initializers -Wno-type-limits -Wno-array-bounds -Wno-unknown-pragmas -Wunused-local-typedefs -Wno-unused-parameter -Wno-unused-function -Wno-unused-result -Wno-strict-overflow -Wno-strict-aliasing -Wno-error=deprecated-declarations -Wno-stringop-overflow -Wno-psabi -Wno-error=pedantic -Wno-error=redundant-decls -Wno-error=old-style-cast -fdiagnostics-color=always -faligned-new -Wno-unused-but-set-variable -Wno-maybe-uninitialized -fno-math-errno -fno-trapping-math -Werror=format -Werror=cast-function-type -Wno-stringop-overflow, LAPACK_INFO=mkl, PERF_WITH_AVX=1, PERF_WITH_AVX2=1, PERF_WITH_AVX512=1, TORCH_VERSION=1.13.1, USE_CUDA=ON, USE_CUDNN=ON, USE_EXCEPTION_PTR=1, USE_GFLAGS=OFF, USE_GLOG=OFF, USE_MKL=ON, USE_MKLDNN=ON, USE_MPI=OFF, USE_NCCL=ON, USE_NNPACK=ON, USE_OPENMP=ON, USE_ROCM=OFF, + + TorchVision: 0.14.1 + OpenCV: 4.7.0 + MMEngine: 0.7.3 + +Runtime environment: + cudnn_benchmark: True + mp_cfg: {'mp_start_method': 'fork', 'opencv_num_threads': 0} + dist_cfg: {'backend': 'nccl'} + seed: None + deterministic: False + Distributed launcher: slurm + Distributed training: True + GPU number: 4 +------------------------------------------------------------ + +2023/06/05 02:47:53 - mmengine - INFO - Config: +optim_wrapper = dict( + optimizer=dict( + type='AdamW', + lr=0.0004, + weight_decay=0.05, + eps=1e-08, + betas=(0.9, 0.999), + _scope_='mmpretrain'), + paramwise_cfg=dict( + norm_decay_mult=0.0, + bias_decay_mult=0.0, + flat_decay_mult=0.0, + custom_keys=dict({ + '.absolute_pos_embed': dict(decay_mult=0.0), + '.relative_position_bias_table': dict(decay_mult=0.0) + })), + type='AmpOptimWrapper', + dtype='bfloat16', + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=False, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=4096) +model = dict( + type='ImageClassifier', + backbone=dict( + frozen_stages=2, + type='ConvNeXt', + arch='small', + drop_path_rate=0.4, + init_cfg=dict( + type='Pretrained', + checkpoint= + 'https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth', + prefix='backbone')), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=768, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + init_cfg=None), + init_cfg=dict( + type='TruncNormal', layer=['Conv2d', 'Linear'], std=0.02, bias=0.0), + train_cfg=None) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/all1.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/all2.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/all3.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3fake8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/cc1m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3real8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV2-1-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/if-dpmsolver++-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV2-1-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/if-dpmsolver++-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1' + +2023/06/05 02:48:06 - mmengine - INFO - Hooks will be executed in the following order: +before_run: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_load_checkpoint: +(ABOVE_NORMAL) EMAHook + -------------------- +before_train: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_train_epoch: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(NORMAL ) DistSamplerSeedHook + -------------------- +before_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook + -------------------- +after_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +after_train_epoch: +(NORMAL ) IterTimerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_val_epoch: +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook + -------------------- +before_val_iter: +(NORMAL ) IterTimerHook + -------------------- +after_val_iter: +(NORMAL ) IterTimerHook +(NORMAL ) VisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_val_epoch: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_save_checkpoint: +(ABOVE_NORMAL) EMAHook + -------------------- +after_train: +(VERY_LOW ) CheckpointHook + -------------------- +before_test_epoch: +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook + -------------------- +before_test_iter: +(NORMAL ) IterTimerHook + -------------------- +after_test_iter: +(NORMAL ) IterTimerHook +(NORMAL ) VisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_test_epoch: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_run: +(BELOW_NORMAL) LoggerHook + -------------------- +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.2.0.weight:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.2.0.bias:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.2.1.bias:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.3.0.weight:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.3.0.bias:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.3.1.bias:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.2.0.gamma:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.2.0.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.2.0.norm.weight:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.2.0.norm.bias:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.2.0.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.2.0.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.2.1.gamma:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.2.1.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.2.1.norm.weight:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.2.1.norm.bias:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.2.1.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.2.1.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.2.2.gamma:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.2.2.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.2.2.norm.weight:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.2.2.norm.bias:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.2.2.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.2.2.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.2.3.gamma:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.2.3.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.2.3.norm.weight:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.2.3.norm.bias:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.2.3.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.2.3.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.2.4.gamma:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.2.4.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.2.4.norm.weight:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.2.4.norm.bias:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.2.4.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.2.4.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.2.5.gamma:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.2.5.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.2.5.norm.weight:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.2.5.norm.bias:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.2.5.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.2.5.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.2.6.gamma:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.2.6.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.2.6.norm.weight:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.2.6.norm.bias:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.2.6.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.2.6.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.2.7.gamma:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.2.7.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.2.7.norm.weight:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.2.7.norm.bias:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.2.7.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.2.7.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.2.8.gamma:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.2.8.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.2.8.norm.weight:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.2.8.norm.bias:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.2.8.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.2.8.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.2.9.gamma:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.2.9.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.2.9.norm.weight:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.2.9.norm.bias:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.2.9.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.2.9.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.2.10.gamma:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.2.10.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.2.10.norm.weight:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.2.10.norm.bias:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.2.10.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.2.10.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.2.11.gamma:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.2.11.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.2.11.norm.weight:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.2.11.norm.bias:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.2.11.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.2.11.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.2.12.gamma:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.2.12.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.2.12.norm.weight:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.2.12.norm.bias:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.2.12.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.2.12.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.2.13.gamma:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.2.13.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.2.13.norm.weight:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.2.13.norm.bias:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.2.13.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.2.13.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.2.14.gamma:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.2.14.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.2.14.norm.weight:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.2.14.norm.bias:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.2.14.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.2.14.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.2.15.gamma:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.2.15.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.2.15.norm.weight:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.2.15.norm.bias:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.2.15.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.2.15.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.2.16.gamma:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.2.16.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.2.16.norm.weight:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.2.16.norm.bias:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.2.16.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.2.16.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.2.17.gamma:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.2.17.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.2.17.norm.weight:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.2.17.norm.bias:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.2.17.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.2.17.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.2.18.gamma:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.2.18.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.2.18.norm.weight:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.2.18.norm.bias:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.2.18.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.2.18.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.2.19.gamma:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.2.19.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.2.19.norm.weight:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.2.19.norm.bias:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.2.19.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.2.19.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.2.20.gamma:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.2.20.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.2.20.norm.weight:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.2.20.norm.bias:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.2.20.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.2.20.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.2.21.gamma:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.2.21.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.2.21.norm.weight:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.2.21.norm.bias:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.2.21.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.2.21.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.2.22.gamma:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.2.22.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.2.22.norm.weight:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.2.22.norm.bias:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.2.22.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.2.22.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.2.23.gamma:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.2.23.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.2.23.norm.weight:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.2.23.norm.bias:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.2.23.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.2.23.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.2.24.gamma:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.2.24.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.2.24.norm.weight:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.2.24.norm.bias:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.2.24.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.2.24.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.2.25.gamma:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.2.25.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.2.25.norm.weight:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.2.25.norm.bias:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.2.25.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.2.25.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.2.26.gamma:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.2.26.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.2.26.norm.weight:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.2.26.norm.bias:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.2.26.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.2.26.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.3.0.gamma:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.3.0.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.3.0.norm.weight:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.3.0.norm.bias:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.3.0.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.3.0.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.3.1.gamma:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.3.1.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.3.1.norm.weight:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.3.1.norm.bias:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.3.1.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.3.1.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.3.2.gamma:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.3.2.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.3.2.norm.weight:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.3.2.norm.bias:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.3.2.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.stages.3.2.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.norm3.weight:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- backbone.norm3.bias:weight_decay=0.0 +2023/06/05 02:48:22 - mmengine - INFO - paramwise_options -- head.fc.bias:weight_decay=0.0 +2023/06/05 02:48:28 - mmengine - INFO - load backbone in model from: https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth +Name of parameter - Initialization information + +backbone.downsample_layers.0.0.weight - torch.Size([96, 3, 4, 4]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.0.0.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.0.1.weight - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.0.1.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.1.0.weight - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.1.0.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.1.1.weight - torch.Size([192, 96, 2, 2]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.1.1.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.2.0.weight - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.2.0.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.2.1.weight - torch.Size([384, 192, 2, 2]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.2.1.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.3.0.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.3.0.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.3.1.weight - torch.Size([768, 384, 2, 2]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.3.1.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.0.gamma - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.0.depthwise_conv.weight - torch.Size([96, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.0.depthwise_conv.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.0.norm.weight - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.0.norm.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.0.pointwise_conv1.weight - torch.Size([384, 96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.0.pointwise_conv1.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.0.pointwise_conv2.weight - torch.Size([96, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.0.pointwise_conv2.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.1.gamma - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.1.depthwise_conv.weight - torch.Size([96, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.1.depthwise_conv.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.1.norm.weight - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.1.norm.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.1.pointwise_conv1.weight - torch.Size([384, 96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.1.pointwise_conv1.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.1.pointwise_conv2.weight - torch.Size([96, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.1.pointwise_conv2.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.2.gamma - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.2.depthwise_conv.weight - torch.Size([96, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.2.depthwise_conv.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.2.norm.weight - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.2.norm.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.2.pointwise_conv1.weight - torch.Size([384, 96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.2.pointwise_conv1.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.2.pointwise_conv2.weight - torch.Size([96, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.2.pointwise_conv2.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.0.gamma - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.0.depthwise_conv.weight - torch.Size([192, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.0.depthwise_conv.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.0.norm.weight - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.0.norm.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.0.pointwise_conv1.weight - torch.Size([768, 192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.0.pointwise_conv1.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.0.pointwise_conv2.weight - torch.Size([192, 768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.0.pointwise_conv2.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.1.gamma - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.1.depthwise_conv.weight - torch.Size([192, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.1.depthwise_conv.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.1.norm.weight - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.1.norm.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.1.pointwise_conv1.weight - torch.Size([768, 192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.1.pointwise_conv1.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.1.pointwise_conv2.weight - torch.Size([192, 768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.1.pointwise_conv2.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.2.gamma - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.2.depthwise_conv.weight - torch.Size([192, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.2.depthwise_conv.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.2.norm.weight - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.2.norm.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.2.pointwise_conv1.weight - torch.Size([768, 192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.2.pointwise_conv1.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.2.pointwise_conv2.weight - torch.Size([192, 768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.2.pointwise_conv2.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.0.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.0.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.0.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.0.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.0.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.0.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.0.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.0.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.0.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.1.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.1.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.1.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.1.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.1.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.1.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.1.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.1.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.1.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.2.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.2.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.2.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.2.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.2.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.2.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.2.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.2.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.2.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.3.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.3.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.3.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.3.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.3.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.3.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.3.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.3.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.3.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.4.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.4.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.4.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.4.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.4.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.4.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.4.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.4.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.4.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.5.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.5.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.5.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.5.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.5.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.5.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.5.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.5.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.5.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.6.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.6.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.6.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.6.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.6.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.6.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.6.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.6.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.6.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.7.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.7.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.7.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.7.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.7.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.7.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.7.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.7.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.7.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.8.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.8.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.8.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.8.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.8.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.8.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.8.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.8.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.8.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.9.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.9.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.9.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.9.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.9.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.9.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.9.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.9.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.9.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.10.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.10.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.10.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.10.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.10.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.10.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.10.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.10.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.10.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.11.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.11.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.11.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.11.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.11.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.11.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.11.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.11.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.11.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.12.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.12.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.12.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.12.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.12.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.12.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.12.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.12.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.12.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.13.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.13.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.13.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.13.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.13.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.13.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.13.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.13.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.13.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.14.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.14.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.14.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.14.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.14.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.14.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.14.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.14.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.14.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.15.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.15.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.15.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.15.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.15.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.15.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.15.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.15.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.15.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.16.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.16.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.16.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.16.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.16.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.16.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.16.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.16.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.16.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.17.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.17.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.17.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.17.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.17.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.17.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.17.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.17.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.17.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.18.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.18.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.18.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.18.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.18.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.18.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.18.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.18.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.18.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.19.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.19.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.19.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.19.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.19.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.19.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.19.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.19.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.19.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.20.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.20.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.20.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.20.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.20.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.20.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.20.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.20.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.20.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.21.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.21.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.21.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.21.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.21.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.21.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.21.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.21.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.21.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.22.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.22.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.22.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.22.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.22.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.22.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.22.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.22.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.22.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.23.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.23.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.23.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.23.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.23.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.23.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.23.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.23.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.23.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.24.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.24.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.24.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.24.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.24.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.24.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.24.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.24.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.24.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.25.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.25.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.25.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.25.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.25.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.25.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.25.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.25.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.25.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.26.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.26.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.26.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.26.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.26.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.26.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.26.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.26.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.26.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.0.gamma - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.0.depthwise_conv.weight - torch.Size([768, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.0.depthwise_conv.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.0.norm.weight - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.0.norm.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.0.pointwise_conv1.weight - torch.Size([3072, 768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.0.pointwise_conv1.bias - torch.Size([3072]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.0.pointwise_conv2.weight - torch.Size([768, 3072]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.0.pointwise_conv2.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.1.gamma - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.1.depthwise_conv.weight - torch.Size([768, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.1.depthwise_conv.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.1.norm.weight - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.1.norm.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.1.pointwise_conv1.weight - torch.Size([3072, 768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.1.pointwise_conv1.bias - torch.Size([3072]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.1.pointwise_conv2.weight - torch.Size([768, 3072]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.1.pointwise_conv2.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.2.gamma - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.2.depthwise_conv.weight - torch.Size([768, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.2.depthwise_conv.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.2.norm.weight - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.2.norm.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.2.pointwise_conv1.weight - torch.Size([3072, 768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.2.pointwise_conv1.bias - torch.Size([3072]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.2.pointwise_conv2.weight - torch.Size([768, 3072]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.2.pointwise_conv2.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.norm3.weight - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.norm3.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +head.fc.weight - torch.Size([2, 768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +head.fc.bias - torch.Size([2]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 +2023/06/05 02:48:28 - mmengine - WARNING - "FileClient" will be deprecated in future. Please use io functions in https://mmengine.readthedocs.io/en/latest/api/fileio.html#file-io +2023/06/05 02:48:28 - mmengine - WARNING - "HardDiskBackend" is the alias of "LocalBackend" and the former will be deprecated in future. +2023/06/05 02:48:28 - mmengine - INFO - Checkpoints will be saved to /mnt/petrelfs/luzeyu/workspace/fakebench/mmpretrain/workdir/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1. +2023/06/05 02:49:59 - mmengine - INFO - Epoch(train) [1][ 100/2056] lr: 3.9998e-04 eta: 5:09:53 time: 0.8093 data_time: 0.2831 memory: 15587 loss: 0.1379 +2023/06/05 02:51:19 - mmengine - INFO - Epoch(train) [1][ 200/2056] lr: 3.9991e-04 eta: 4:50:14 time: 0.8239 data_time: 0.3029 memory: 15587 loss: 0.1014 +2023/06/05 02:52:40 - mmengine - INFO - Epoch(train) [1][ 300/2056] lr: 3.9980e-04 eta: 4:43:18 time: 0.8115 data_time: 0.2871 memory: 15587 loss: 0.1143 +2023/06/05 02:53:59 - mmengine - INFO - Epoch(train) [1][ 400/2056] lr: 3.9964e-04 eta: 4:38:10 time: 0.8054 data_time: 0.2840 memory: 15587 loss: 0.0828 +2023/06/05 02:55:21 - mmengine - INFO - Epoch(train) [1][ 500/2056] lr: 3.9943e-04 eta: 4:36:22 time: 0.8163 data_time: 0.3001 memory: 15587 loss: 0.0818 +2023/06/05 02:56:41 - mmengine - INFO - Epoch(train) [1][ 600/2056] lr: 3.9918e-04 eta: 4:33:27 time: 0.8103 data_time: 0.2885 memory: 15587 loss: 0.0753 +2023/06/05 02:58:02 - mmengine - INFO - Epoch(train) [1][ 700/2056] lr: 3.9889e-04 eta: 4:31:24 time: 0.8154 data_time: 0.2856 memory: 15587 loss: 0.0730 +2023/06/05 02:59:23 - mmengine - INFO - Epoch(train) [1][ 800/2056] lr: 3.9855e-04 eta: 4:29:28 time: 0.8638 data_time: 0.3365 memory: 15587 loss: 0.0625 +2023/06/05 03:01:59 - mmengine - INFO - Epoch(train) [1][ 900/2056] lr: 3.9816e-04 eta: 4:55:17 time: 0.8841 data_time: 0.0017 memory: 15587 loss: 0.0653 +2023/06/05 03:03:19 - mmengine - INFO - Exp name: convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1_20230605_024744 +2023/06/05 03:03:19 - mmengine - INFO - Epoch(train) [1][1000/2056] lr: 3.9773e-04 eta: 4:50:31 time: 0.7775 data_time: 0.0019 memory: 15587 loss: 0.0784 +2023/06/05 03:04:37 - mmengine - INFO - Epoch(train) [1][1100/2056] lr: 3.9726e-04 eta: 4:45:47 time: 0.7850 data_time: 0.0030 memory: 15587 loss: 0.0714 +2023/06/05 03:06:07 - mmengine - INFO - Epoch(train) [1][1200/2056] lr: 3.9674e-04 eta: 4:44:40 time: 0.7213 data_time: 0.0018 memory: 15587 loss: 0.0590 +2023/06/05 03:07:24 - mmengine - INFO - Epoch(train) [1][1300/2056] lr: 3.9617e-04 eta: 4:40:26 time: 0.8026 data_time: 0.1319 memory: 15587 loss: 0.0675 +2023/06/05 03:08:43 - mmengine - INFO - Epoch(train) [1][1400/2056] lr: 3.9556e-04 eta: 4:37:09 time: 0.8237 data_time: 0.3057 memory: 15587 loss: 0.0647 +2023/06/05 03:10:04 - mmengine - INFO - Epoch(train) [1][1500/2056] lr: 3.9491e-04 eta: 4:34:21 time: 0.7852 data_time: 0.2062 memory: 15587 loss: 0.0612 +2023/06/05 03:11:26 - mmengine - INFO - Epoch(train) [1][1600/2056] lr: 3.9421e-04 eta: 4:32:13 time: 0.7951 data_time: 0.0304 memory: 15587 loss: 0.0509 +2023/06/05 03:12:47 - mmengine - INFO - Epoch(train) [1][1700/2056] lr: 3.9347e-04 eta: 4:29:49 time: 0.8705 data_time: 0.3520 memory: 15587 loss: 0.0588 +2023/06/05 03:14:10 - mmengine - INFO - Epoch(train) [1][1800/2056] lr: 3.9268e-04 eta: 4:27:48 time: 0.8031 data_time: 0.2821 memory: 15587 loss: 0.0506 +2023/06/05 03:15:30 - mmengine - INFO - Epoch(train) [1][1900/2056] lr: 3.9185e-04 eta: 4:25:32 time: 0.7872 data_time: 0.2564 memory: 15587 loss: 0.0507 +2023/06/05 03:16:50 - mmengine - INFO - Exp name: convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1_20230605_024744 +2023/06/05 03:16:50 - mmengine - INFO - Epoch(train) [1][2000/2056] lr: 3.9097e-04 eta: 4:23:16 time: 0.7734 data_time: 0.2422 memory: 15587 loss: 0.0577 +2023/06/05 03:17:35 - mmengine - INFO - Exp name: convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1_20230605_024744 +2023/06/05 03:17:35 - mmengine - INFO - Saving checkpoint at 1 epochs +2023/06/05 03:18:41 - mmengine - INFO - Epoch(val) [1][60/60] accuracy/top1: 81.1016 single-label/precision_classwise: [96.97371673583984, 53.366981506347656] single-label/recall_classwise: [78.41886901855469, 90.9844970703125] single-label/f1-score_classwise: [86.71483612060547, 67.27423858642578] data_time: 0.1895 time: 0.6321 +2023/06/05 03:20:08 - mmengine - INFO - Epoch(train) [2][ 100/2056] lr: 3.8952e-04 eta: 4:20:52 time: 0.8160 data_time: 0.2979 memory: 15587 loss: 0.0412 +2023/06/05 03:21:29 - mmengine - INFO - Epoch(train) [2][ 200/2056] lr: 3.8854e-04 eta: 4:18:49 time: 0.8290 data_time: 0.2896 memory: 15585 loss: 0.0497 +2023/06/05 03:22:49 - mmengine - INFO - Epoch(train) [2][ 300/2056] lr: 3.8751e-04 eta: 4:16:49 time: 0.7842 data_time: 0.2659 memory: 15585 loss: 0.0456 +2023/06/05 03:25:40 - mmengine - INFO - Epoch(train) [2][ 400/2056] lr: 3.8644e-04 eta: 4:26:04 time: 0.7427 data_time: 0.0158 memory: 15585 loss: 0.0440 +2023/06/05 03:27:00 - mmengine - INFO - Epoch(train) [2][ 500/2056] lr: 3.8533e-04 eta: 4:23:32 time: 0.8405 data_time: 0.1786 memory: 15585 loss: 0.0451 +2023/06/05 03:28:19 - mmengine - INFO - Epoch(train) [2][ 600/2056] lr: 3.8417e-04 eta: 4:21:09 time: 0.8419 data_time: 0.0019 memory: 15585 loss: 0.0430 +2023/06/05 03:29:40 - mmengine - INFO - Epoch(train) [2][ 700/2056] lr: 3.8298e-04 eta: 4:18:59 time: 0.7799 data_time: 0.0024 memory: 15585 loss: 0.0341 +2023/06/05 03:31:01 - mmengine - INFO - Epoch(train) [2][ 800/2056] lr: 3.8174e-04 eta: 4:16:52 time: 0.8440 data_time: 0.0017 memory: 15585 loss: 0.0491 +2023/06/05 03:32:23 - mmengine - INFO - Epoch(train) [2][ 900/2056] lr: 3.8046e-04 eta: 4:14:56 time: 0.8752 data_time: 0.2605 memory: 15585 loss: 0.0416 +2023/06/05 03:32:59 - mmengine - INFO - Exp name: convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1_20230605_024744 +2023/06/05 03:33:46 - mmengine - INFO - Epoch(train) [2][1000/2056] lr: 3.7914e-04 eta: 4:13:04 time: 0.8177 data_time: 0.2993 memory: 15585 loss: 0.0359 +2023/06/05 03:35:04 - mmengine - INFO - Epoch(train) [2][1100/2056] lr: 3.7778e-04 eta: 4:10:51 time: 0.7665 data_time: 0.2422 memory: 15585 loss: 0.0365 +2023/06/05 03:36:23 - mmengine - INFO - Epoch(train) [2][1200/2056] lr: 3.7637e-04 eta: 4:08:46 time: 0.7523 data_time: 0.2286 memory: 15585 loss: 0.0444 +2023/06/05 03:37:44 - mmengine - INFO - Epoch(train) [2][1300/2056] lr: 3.7493e-04 eta: 4:06:52 time: 0.7945 data_time: 0.2676 memory: 15585 loss: 0.0457 +2023/06/05 03:39:03 - mmengine - INFO - Epoch(train) [2][1400/2056] lr: 3.7345e-04 eta: 4:04:51 time: 0.7683 data_time: 0.2472 memory: 15585 loss: 0.0404 +2023/06/05 03:40:24 - mmengine - INFO - Epoch(train) [2][1500/2056] lr: 3.7193e-04 eta: 4:03:02 time: 0.7875 data_time: 0.2686 memory: 15585 loss: 0.0537 +2023/06/05 03:41:44 - mmengine - INFO - Epoch(train) [2][1600/2056] lr: 3.7037e-04 eta: 4:01:06 time: 0.7754 data_time: 0.2571 memory: 15585 loss: 0.0268 +2023/06/05 03:43:05 - mmengine - INFO - Epoch(train) [2][1700/2056] lr: 3.6877e-04 eta: 3:59:22 time: 0.7572 data_time: 0.2341 memory: 15585 loss: 0.0450 +2023/06/05 03:44:26 - mmengine - INFO - Epoch(train) [2][1800/2056] lr: 3.6714e-04 eta: 3:57:36 time: 0.8512 data_time: 0.3279 memory: 15585 loss: 0.0426 +2023/06/05 03:45:46 - mmengine - INFO - Epoch(train) [2][1900/2056] lr: 3.6546e-04 eta: 3:55:51 time: 0.7758 data_time: 0.2527 memory: 15585 loss: 0.0440 +2023/06/05 03:46:23 - mmengine - INFO - Exp name: convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1_20230605_024744 +2023/06/05 03:47:07 - mmengine - INFO - Epoch(train) [2][2000/2056] lr: 3.6375e-04 eta: 3:54:07 time: 0.7712 data_time: 0.2568 memory: 15585 loss: 0.0445 +2023/06/05 03:47:50 - mmengine - INFO - Exp name: convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1_20230605_024744 +2023/06/05 03:47:50 - mmengine - INFO - Saving checkpoint at 2 epochs +2023/06/05 03:48:53 - mmengine - INFO - Epoch(val) [2][60/60] accuracy/top1: 91.1226 single-label/precision_classwise: [99.34224700927734, 71.28572845458984] single-label/recall_classwise: [89.30412292480469, 97.82170867919922] single-label/f1-score_classwise: [94.05611419677734, 82.47173309326172] data_time: 0.1708 time: 0.5689 +2023/06/05 03:50:21 - mmengine - INFO - Epoch(train) [3][ 100/2056] lr: 3.6101e-04 eta: 3:51:47 time: 0.9223 data_time: 0.0859 memory: 15585 loss: 0.0411 +2023/06/05 03:51:41 - mmengine - INFO - Epoch(train) [3][ 200/2056] lr: 3.5920e-04 eta: 3:50:04 time: 0.8254 data_time: 0.0017 memory: 15585 loss: 0.0461 +2023/06/05 03:53:02 - mmengine - INFO - Epoch(train) [3][ 300/2056] lr: 3.5736e-04 eta: 3:48:24 time: 0.7564 data_time: 0.0017 memory: 15585 loss: 0.0399 +2023/06/05 03:54:21 - mmengine - INFO - Epoch(train) [3][ 400/2056] lr: 3.5548e-04 eta: 3:46:38 time: 0.8286 data_time: 0.0997 memory: 15585 loss: 0.0289 +2023/06/05 03:55:41 - mmengine - INFO - Epoch(train) [3][ 500/2056] lr: 3.5357e-04 eta: 3:44:55 time: 0.7644 data_time: 0.0158 memory: 15585 loss: 0.0338 +2023/06/05 03:57:04 - mmengine - INFO - Epoch(train) [3][ 600/2056] lr: 3.5162e-04 eta: 3:43:25 time: 0.8293 data_time: 0.3038 memory: 15585 loss: 0.0360 +2023/06/05 03:58:24 - mmengine - INFO - Epoch(train) [3][ 700/2056] lr: 3.4964e-04 eta: 3:41:47 time: 0.8274 data_time: 0.2917 memory: 15585 loss: 0.0386 +2023/06/05 03:59:47 - mmengine - INFO - Epoch(train) [3][ 800/2056] lr: 3.4763e-04 eta: 3:40:17 time: 0.8628 data_time: 0.3372 memory: 15585 loss: 0.0351 +2023/06/05 04:00:57 - mmengine - INFO - Exp name: convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1_20230605_024744 +2023/06/05 04:01:06 - mmengine - INFO - Epoch(train) [3][ 900/2056] lr: 3.4558e-04 eta: 3:38:37 time: 0.7998 data_time: 0.2681 memory: 15585 loss: 0.0442 +2023/06/05 04:02:25 - mmengine - INFO - Epoch(train) [3][1000/2056] lr: 3.4350e-04 eta: 3:36:54 time: 0.7846 data_time: 0.2597 memory: 15585 loss: 0.0317 +2023/06/05 04:03:45 - mmengine - INFO - Epoch(train) [3][1100/2056] lr: 3.4138e-04 eta: 3:35:19 time: 0.8333 data_time: 0.3108 memory: 15585 loss: 0.0382 +2023/06/05 04:05:07 - mmengine - INFO - Epoch(train) [3][1200/2056] lr: 3.3924e-04 eta: 3:33:49 time: 0.7544 data_time: 0.2329 memory: 15585 loss: 0.0368 +2023/06/05 04:06:26 - mmengine - INFO - Epoch(train) [3][1300/2056] lr: 3.3706e-04 eta: 3:32:09 time: 0.7744 data_time: 0.2456 memory: 15585 loss: 0.0307 +2023/06/05 04:07:46 - mmengine - INFO - Epoch(train) [3][1400/2056] lr: 3.3485e-04 eta: 3:30:36 time: 0.7751 data_time: 0.2515 memory: 15585 loss: 0.0302 +2023/06/05 04:09:06 - mmengine - INFO - Epoch(train) [3][1500/2056] lr: 3.3261e-04 eta: 3:28:59 time: 0.7380 data_time: 0.2201 memory: 15585 loss: 0.0414 +2023/06/05 04:10:28 - mmengine - INFO - Epoch(train) [3][1600/2056] lr: 3.3035e-04 eta: 3:27:32 time: 0.7552 data_time: 0.2110 memory: 15585 loss: 0.0333 +2023/06/05 04:11:54 - mmengine - INFO - Epoch(train) [3][1700/2056] lr: 3.2805e-04 eta: 3:26:14 time: 0.7050 data_time: 0.1883 memory: 15585 loss: 0.0361 +2023/06/05 04:13:13 - mmengine - INFO - Epoch(train) [3][1800/2056] lr: 3.2572e-04 eta: 3:24:37 time: 0.8197 data_time: 0.2859 memory: 15585 loss: 0.0405 +2023/06/05 04:14:25 - mmengine - INFO - Exp name: convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1_20230605_024744 +2023/06/05 04:14:34 - mmengine - INFO - Epoch(train) [3][1900/2056] lr: 3.2337e-04 eta: 3:23:06 time: 0.7869 data_time: 0.2520 memory: 15585 loss: 0.0232 +2023/06/05 04:15:54 - mmengine - INFO - Epoch(train) [3][2000/2056] lr: 3.2099e-04 eta: 3:21:33 time: 0.8069 data_time: 0.2728 memory: 15585 loss: 0.0356 +2023/06/05 04:16:39 - mmengine - INFO - Exp name: convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1_20230605_024744 +2023/06/05 04:16:39 - mmengine - INFO - Saving checkpoint at 3 epochs +2023/06/05 04:17:44 - mmengine - INFO - Epoch(val) [3][60/60] accuracy/top1: 94.3531 single-label/precision_classwise: [99.77432250976562, 79.44389343261719] single-label/recall_classwise: [93.03074645996094, 99.22480773925781] single-label/f1-score_classwise: [96.28460693359375, 88.23934936523438] data_time: 0.1858 time: 0.5866 +2023/06/05 04:19:11 - mmengine - INFO - Epoch(train) [4][ 100/2056] lr: 3.1722e-04 eta: 3:19:27 time: 0.8348 data_time: 0.3116 memory: 15585 loss: 0.0353 +2023/06/05 04:20:32 - mmengine - INFO - Epoch(train) [4][ 200/2056] lr: 3.1477e-04 eta: 3:17:57 time: 0.8264 data_time: 0.3095 memory: 15585 loss: 0.0333 +2023/06/05 04:21:53 - mmengine - INFO - Epoch(train) [4][ 300/2056] lr: 3.1229e-04 eta: 3:16:27 time: 0.7942 data_time: 0.2794 memory: 15585 loss: 0.0325 +2023/06/05 04:23:14 - mmengine - INFO - Epoch(train) [4][ 400/2056] lr: 3.0979e-04 eta: 3:14:58 time: 0.8265 data_time: 0.3086 memory: 15585 loss: 0.0239 +2023/06/05 04:24:35 - mmengine - INFO - Epoch(train) [4][ 500/2056] lr: 3.0727e-04 eta: 3:13:30 time: 0.8145 data_time: 0.2922 memory: 15585 loss: 0.0395 +2023/06/05 04:25:55 - mmengine - INFO - Epoch(train) [4][ 600/2056] lr: 3.0472e-04 eta: 3:11:59 time: 0.7668 data_time: 0.2447 memory: 15585 loss: 0.0367 +2023/06/05 04:28:54 - mmengine - INFO - Epoch(train) [4][ 700/2056] lr: 3.0215e-04 eta: 3:13:45 time: 0.8646 data_time: 0.3433 memory: 15585 loss: 0.0258 +2023/06/05 04:30:13 - mmengine - INFO - Epoch(train) [4][ 800/2056] lr: 2.9955e-04 eta: 3:12:09 time: 0.8239 data_time: 0.2991 memory: 15585 loss: 0.0185 +2023/06/05 04:30:39 - mmengine - INFO - Exp name: convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1_20230605_024744 +2023/06/05 04:31:33 - mmengine - INFO - Epoch(train) [4][ 900/2056] lr: 2.9694e-04 eta: 3:10:35 time: 0.8184 data_time: 0.3048 memory: 15585 loss: 0.0298 +2023/06/05 04:32:52 - mmengine - INFO - Epoch(train) [4][1000/2056] lr: 2.9430e-04 eta: 3:08:59 time: 0.7564 data_time: 0.2362 memory: 15585 loss: 0.0240 +2023/06/05 04:34:15 - mmengine - INFO - Epoch(train) [4][1100/2056] lr: 2.9164e-04 eta: 3:07:32 time: 0.7578 data_time: 0.2384 memory: 15585 loss: 0.0304 +2023/06/05 04:35:34 - mmengine - INFO - Epoch(train) [4][1200/2056] lr: 2.8896e-04 eta: 3:05:56 time: 0.8321 data_time: 0.3101 memory: 15585 loss: 0.0248 +2023/06/05 04:36:53 - mmengine - INFO - Epoch(train) [4][1300/2056] lr: 2.8626e-04 eta: 3:04:23 time: 0.7995 data_time: 0.2189 memory: 15585 loss: 0.0334 +2023/06/05 04:38:10 - mmengine - INFO - Epoch(train) [4][1400/2056] lr: 2.8354e-04 eta: 3:02:45 time: 0.7395 data_time: 0.2061 memory: 15585 loss: 0.0256 +2023/06/05 04:39:29 - mmengine - INFO - Epoch(train) [4][1500/2056] lr: 2.8081e-04 eta: 3:01:12 time: 0.7484 data_time: 0.2268 memory: 15585 loss: 0.0233 +2023/06/05 04:40:53 - mmengine - INFO - Epoch(train) [4][1600/2056] lr: 2.7805e-04 eta: 2:59:47 time: 0.9600 data_time: 0.0260 memory: 15585 loss: 0.0361 +2023/06/05 04:42:16 - mmengine - INFO - Epoch(train) [4][1700/2056] lr: 2.7528e-04 eta: 2:58:21 time: 0.8151 data_time: 0.0019 memory: 15585 loss: 0.0302 +2023/06/05 04:43:41 - mmengine - INFO - Epoch(train) [4][1800/2056] lr: 2.7249e-04 eta: 2:56:57 time: 0.9153 data_time: 0.0017 memory: 15585 loss: 0.0235 +2023/06/05 04:44:05 - mmengine - INFO - Exp name: convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1_20230605_024744 +2023/06/05 04:44:58 - mmengine - INFO - Epoch(train) [4][1900/2056] lr: 2.6969e-04 eta: 2:55:22 time: 0.8375 data_time: 0.0017 memory: 15585 loss: 0.0291 +2023/06/05 04:46:27 - mmengine - INFO - Epoch(train) [4][2000/2056] lr: 2.6687e-04 eta: 2:54:04 time: 0.8518 data_time: 0.0016 memory: 15585 loss: 0.0238 +2023/06/05 04:47:06 - mmengine - INFO - Exp name: convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1_20230605_024744 +2023/06/05 04:47:06 - mmengine - INFO - Saving checkpoint at 4 epochs +2023/06/05 04:48:09 - mmengine - INFO - Epoch(val) [4][60/60] accuracy/top1: 96.1687 single-label/precision_classwise: [99.9006576538086, 84.99173736572266] single-label/recall_classwise: [95.2233657836914, 99.65116119384766] single-label/f1-score_classwise: [97.50595092773438, 91.73951721191406] data_time: 0.1729 time: 0.5709 +2023/06/05 04:49:34 - mmengine - INFO - Epoch(train) [5][ 100/2056] lr: 2.6245e-04 eta: 2:51:41 time: 0.7771 data_time: 0.2220 memory: 15585 loss: 0.0245 +2023/06/05 04:50:54 - mmengine - INFO - Epoch(train) [5][ 200/2056] lr: 2.5959e-04 eta: 2:50:11 time: 0.7902 data_time: 0.2059 memory: 15585 loss: 0.0267 +2023/06/05 04:52:13 - mmengine - INFO - Epoch(train) [5][ 300/2056] lr: 2.5673e-04 eta: 2:48:40 time: 0.7884 data_time: 0.2714 memory: 15585 loss: 0.0205 +2023/06/05 04:53:34 - mmengine - INFO - Epoch(train) [5][ 400/2056] lr: 2.5385e-04 eta: 2:47:11 time: 0.7576 data_time: 0.2412 memory: 15585 loss: 0.0259 +2023/06/05 04:54:53 - mmengine - INFO - Epoch(train) [5][ 500/2056] lr: 2.5096e-04 eta: 2:45:40 time: 0.8023 data_time: 0.2309 memory: 15585 loss: 0.0237 +2023/06/05 04:56:13 - mmengine - INFO - Epoch(train) [5][ 600/2056] lr: 2.4806e-04 eta: 2:44:12 time: 0.7826 data_time: 0.0018 memory: 15585 loss: 0.0279 +2023/06/05 04:57:34 - mmengine - INFO - Epoch(train) [5][ 700/2056] lr: 2.4514e-04 eta: 2:42:43 time: 0.7857 data_time: 0.1974 memory: 15585 loss: 0.0294 +2023/06/05 04:58:34 - mmengine - INFO - Exp name: convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1_20230605_024744 +2023/06/05 04:58:53 - mmengine - INFO - Epoch(train) [5][ 800/2056] lr: 2.4222e-04 eta: 2:41:13 time: 0.8777 data_time: 0.1455 memory: 15585 loss: 0.0258 +2023/06/05 05:00:12 - mmengine - INFO - Epoch(train) [5][ 900/2056] lr: 2.3929e-04 eta: 2:39:44 time: 0.8500 data_time: 0.2705 memory: 15585 loss: 0.0223 +2023/06/05 05:01:32 - mmengine - INFO - Epoch(train) [5][1000/2056] lr: 2.3636e-04 eta: 2:38:15 time: 0.7864 data_time: 0.2676 memory: 15585 loss: 0.0221 +2023/06/05 05:02:51 - mmengine - INFO - Epoch(train) [5][1100/2056] lr: 2.3341e-04 eta: 2:36:46 time: 0.7152 data_time: 0.0018 memory: 15585 loss: 0.0202 +2023/06/05 05:04:09 - mmengine - INFO - Epoch(train) [5][1200/2056] lr: 2.3046e-04 eta: 2:35:14 time: 0.8243 data_time: 0.0483 memory: 15585 loss: 0.0244 +2023/06/05 05:05:27 - mmengine - INFO - Epoch(train) [5][1300/2056] lr: 2.2751e-04 eta: 2:33:45 time: 0.7456 data_time: 0.2275 memory: 15585 loss: 0.0238 +2023/06/05 05:06:48 - mmengine - INFO - Epoch(train) [5][1400/2056] lr: 2.2454e-04 eta: 2:32:18 time: 0.8006 data_time: 0.2769 memory: 15585 loss: 0.0291 +2023/06/05 05:08:07 - mmengine - INFO - Epoch(train) [5][1500/2056] lr: 2.2158e-04 eta: 2:30:49 time: 0.7813 data_time: 0.2587 memory: 15585 loss: 0.0200 +2023/06/05 05:09:27 - mmengine - INFO - Epoch(train) [5][1600/2056] lr: 2.1861e-04 eta: 2:29:22 time: 0.7763 data_time: 0.2558 memory: 15585 loss: 0.0267 +2023/06/05 05:10:45 - mmengine - INFO - Epoch(train) [5][1700/2056] lr: 2.1563e-04 eta: 2:27:52 time: 0.8034 data_time: 0.2812 memory: 15585 loss: 0.0225 +2023/06/05 05:11:46 - mmengine - INFO - Exp name: convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1_20230605_024744 +2023/06/05 05:12:04 - mmengine - INFO - Epoch(train) [5][1800/2056] lr: 2.1266e-04 eta: 2:26:25 time: 0.7831 data_time: 0.2560 memory: 15585 loss: 0.0271 +2023/06/05 05:13:24 - mmengine - INFO - Epoch(train) [5][1900/2056] lr: 2.0968e-04 eta: 2:24:58 time: 0.8185 data_time: 0.2918 memory: 15585 loss: 0.0284 +2023/06/05 05:14:43 - mmengine - INFO - Epoch(train) [5][2000/2056] lr: 2.0670e-04 eta: 2:23:30 time: 0.8062 data_time: 0.2809 memory: 15585 loss: 0.0233 +2023/06/05 05:15:32 - mmengine - INFO - Exp name: convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1_20230605_024744 +2023/06/05 05:15:32 - mmengine - INFO - Saving checkpoint at 5 epochs +2023/06/05 05:16:35 - mmengine - INFO - Epoch(val) [5][60/60] accuracy/top1: 97.3570 single-label/precision_classwise: [99.910888671875, 89.2056884765625] single-label/recall_classwise: [96.72579956054688, 99.68216705322266] single-label/f1-score_classwise: [98.29254913330078, 94.15340423583984] data_time: 0.1691 time: 0.5697 +2023/06/05 05:18:01 - mmengine - INFO - Epoch(train) [6][ 100/2056] lr: 2.0205e-04 eta: 2:21:25 time: 0.8112 data_time: 0.2811 memory: 15585 loss: 0.0272 +2023/06/05 05:19:23 - mmengine - INFO - Epoch(train) [6][ 200/2056] lr: 1.9907e-04 eta: 2:20:00 time: 0.7849 data_time: 0.2631 memory: 15585 loss: 0.0200 +2023/06/05 05:20:42 - mmengine - INFO - Epoch(train) [6][ 300/2056] lr: 1.9609e-04 eta: 2:18:33 time: 0.7870 data_time: 0.2310 memory: 15585 loss: 0.0266 +2023/06/05 05:22:00 - mmengine - INFO - Epoch(train) [6][ 400/2056] lr: 1.9312e-04 eta: 2:17:05 time: 0.7678 data_time: 0.2131 memory: 15585 loss: 0.0261 +2023/06/05 05:23:21 - mmengine - INFO - Epoch(train) [6][ 500/2056] lr: 1.9015e-04 eta: 2:15:40 time: 0.8316 data_time: 0.2908 memory: 15585 loss: 0.0238 +2023/06/05 05:24:41 - mmengine - INFO - Epoch(train) [6][ 600/2056] lr: 1.8718e-04 eta: 2:14:13 time: 0.7920 data_time: 0.2684 memory: 15585 loss: 0.0251 +2023/06/05 05:26:01 - mmengine - INFO - Epoch(train) [6][ 700/2056] lr: 1.8421e-04 eta: 2:12:47 time: 0.7676 data_time: 0.2469 memory: 15585 loss: 0.0241 +2023/06/05 05:26:17 - mmengine - INFO - Exp name: convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1_20230605_024744 +2023/06/05 05:27:20 - mmengine - INFO - Epoch(train) [6][ 800/2056] lr: 1.8125e-04 eta: 2:11:20 time: 0.7901 data_time: 0.2715 memory: 15585 loss: 0.0225 +2023/06/05 05:28:39 - mmengine - INFO - Epoch(train) [6][ 900/2056] lr: 1.7830e-04 eta: 2:09:54 time: 0.7876 data_time: 0.2463 memory: 15585 loss: 0.0118 +2023/06/05 05:29:59 - mmengine - INFO - Epoch(train) [6][1000/2056] lr: 1.7535e-04 eta: 2:08:29 time: 0.8193 data_time: 0.3008 memory: 15585 loss: 0.0276 +2023/06/05 05:31:19 - mmengine - INFO - Epoch(train) [6][1100/2056] lr: 1.7241e-04 eta: 2:07:02 time: 0.7586 data_time: 0.2408 memory: 15585 loss: 0.0228 +2023/06/05 05:32:37 - mmengine - INFO - Epoch(train) [6][1200/2056] lr: 1.6947e-04 eta: 2:05:36 time: 0.7496 data_time: 0.2290 memory: 15585 loss: 0.0175 +2023/06/05 05:33:55 - mmengine - INFO - Epoch(train) [6][1300/2056] lr: 1.6655e-04 eta: 2:04:09 time: 0.7467 data_time: 0.2172 memory: 15585 loss: 0.0183 +2023/06/05 05:35:13 - mmengine - INFO - Epoch(train) [6][1400/2056] lr: 1.6363e-04 eta: 2:02:42 time: 0.7547 data_time: 0.2350 memory: 15585 loss: 0.0276 +2023/06/05 05:36:32 - mmengine - INFO - Epoch(train) [6][1500/2056] lr: 1.6072e-04 eta: 2:01:16 time: 0.8136 data_time: 0.2802 memory: 15585 loss: 0.0198 +2023/06/05 05:37:51 - mmengine - INFO - Epoch(train) [6][1600/2056] lr: 1.5783e-04 eta: 1:59:50 time: 0.7779 data_time: 0.2150 memory: 15585 loss: 0.0237 +2023/06/05 05:39:09 - mmengine - INFO - Epoch(train) [6][1700/2056] lr: 1.5494e-04 eta: 1:58:25 time: 0.7947 data_time: 0.2023 memory: 15585 loss: 0.0158 +2023/06/05 05:39:25 - mmengine - INFO - Exp name: convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1_20230605_024744 +2023/06/05 05:40:29 - mmengine - INFO - Epoch(train) [6][1800/2056] lr: 1.5207e-04 eta: 1:56:59 time: 0.7699 data_time: 0.1757 memory: 15585 loss: 0.0199 +2023/06/05 05:41:47 - mmengine - INFO - Epoch(train) [6][1900/2056] lr: 1.4921e-04 eta: 1:55:33 time: 0.7960 data_time: 0.2741 memory: 15585 loss: 0.0242 +2023/06/05 05:43:07 - mmengine - INFO - Epoch(train) [6][2000/2056] lr: 1.4636e-04 eta: 1:54:09 time: 0.7756 data_time: 0.1780 memory: 15585 loss: 0.0150 +2023/06/05 05:43:52 - mmengine - INFO - Exp name: convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1_20230605_024744 +2023/06/05 05:43:52 - mmengine - INFO - Saving checkpoint at 6 epochs +2023/06/05 05:44:55 - mmengine - INFO - Epoch(val) [6][60/60] accuracy/top1: 98.2606 single-label/precision_classwise: [99.89692687988281, 92.76073455810547] single-label/recall_classwise: [97.88944244384766, 99.6279067993164] single-label/f1-score_classwise: [98.88299560546875, 96.07176208496094] data_time: 0.1731 time: 0.5702 +2023/06/05 05:46:22 - mmengine - INFO - Epoch(train) [7][ 100/2056] lr: 1.4194e-04 eta: 1:52:02 time: 0.8414 data_time: 0.2266 memory: 15585 loss: 0.0250 +2023/06/05 05:47:42 - mmengine - INFO - Epoch(train) [7][ 200/2056] lr: 1.3913e-04 eta: 1:50:37 time: 0.7997 data_time: 0.2701 memory: 15585 loss: 0.0241 +2023/06/05 05:49:02 - mmengine - INFO - Epoch(train) [7][ 300/2056] lr: 1.3633e-04 eta: 1:49:13 time: 0.8095 data_time: 0.2745 memory: 15585 loss: 0.0157 +2023/06/05 05:50:22 - mmengine - INFO - Epoch(train) [7][ 400/2056] lr: 1.3355e-04 eta: 1:47:48 time: 0.7694 data_time: 0.2377 memory: 15585 loss: 0.0151 +2023/06/05 05:51:41 - mmengine - INFO - Epoch(train) [7][ 500/2056] lr: 1.3079e-04 eta: 1:46:24 time: 0.7464 data_time: 0.2304 memory: 15585 loss: 0.0183 +2023/06/05 05:52:59 - mmengine - INFO - Epoch(train) [7][ 600/2056] lr: 1.2804e-04 eta: 1:44:58 time: 0.7953 data_time: 0.2669 memory: 15585 loss: 0.0170 +2023/06/05 05:53:52 - mmengine - INFO - Exp name: convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1_20230605_024744 +2023/06/05 05:54:19 - mmengine - INFO - Epoch(train) [7][ 700/2056] lr: 1.2531e-04 eta: 1:43:34 time: 0.7800 data_time: 0.2572 memory: 15585 loss: 0.0176 +2023/06/05 05:55:37 - mmengine - INFO - Epoch(train) [7][ 800/2056] lr: 1.2260e-04 eta: 1:42:09 time: 0.8133 data_time: 0.2900 memory: 15585 loss: 0.0214 +2023/06/05 05:56:58 - mmengine - INFO - Epoch(train) [7][ 900/2056] lr: 1.1991e-04 eta: 1:40:45 time: 0.8051 data_time: 0.0720 memory: 15585 loss: 0.0208 +2023/06/05 05:58:17 - mmengine - INFO - Epoch(train) [7][1000/2056] lr: 1.1724e-04 eta: 1:39:21 time: 0.8733 data_time: 0.1212 memory: 15585 loss: 0.0162 +2023/06/05 05:59:36 - mmengine - INFO - Epoch(train) [7][1100/2056] lr: 1.1459e-04 eta: 1:37:57 time: 0.8096 data_time: 0.0877 memory: 15585 loss: 0.0261 +2023/06/05 06:00:57 - mmengine - INFO - Epoch(train) [7][1200/2056] lr: 1.1196e-04 eta: 1:36:33 time: 0.7523 data_time: 0.0020 memory: 15585 loss: 0.0160 +2023/06/05 06:02:16 - mmengine - INFO - Epoch(train) [7][1300/2056] lr: 1.0936e-04 eta: 1:35:09 time: 0.8146 data_time: 0.0716 memory: 15585 loss: 0.0158 +2023/06/05 06:03:38 - mmengine - INFO - Epoch(train) [7][1400/2056] lr: 1.0677e-04 eta: 1:33:47 time: 0.8054 data_time: 0.0018 memory: 15585 loss: 0.0217 +2023/06/05 06:04:58 - mmengine - INFO - Epoch(train) [7][1500/2056] lr: 1.0421e-04 eta: 1:32:23 time: 0.8385 data_time: 0.1228 memory: 15585 loss: 0.0123 +2023/06/05 06:06:17 - mmengine - INFO - Epoch(train) [7][1600/2056] lr: 1.0167e-04 eta: 1:30:58 time: 0.8114 data_time: 0.1254 memory: 15585 loss: 0.0283 +2023/06/05 06:07:08 - mmengine - INFO - Exp name: convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1_20230605_024744 +2023/06/05 06:07:36 - mmengine - INFO - Epoch(train) [7][1700/2056] lr: 9.9154e-05 eta: 1:29:35 time: 0.7894 data_time: 0.2400 memory: 15585 loss: 0.0095 +2023/06/05 06:08:57 - mmengine - INFO - Epoch(train) [7][1800/2056] lr: 9.6664e-05 eta: 1:28:12 time: 0.7629 data_time: 0.2127 memory: 15585 loss: 0.0206 +2023/06/05 06:10:17 - mmengine - INFO - Epoch(train) [7][1900/2056] lr: 9.4199e-05 eta: 1:26:48 time: 0.7785 data_time: 0.2547 memory: 15585 loss: 0.0154 +2023/06/05 06:11:36 - mmengine - INFO - Epoch(train) [7][2000/2056] lr: 9.1760e-05 eta: 1:25:24 time: 0.8137 data_time: 0.2521 memory: 15585 loss: 0.0187 +2023/06/05 06:12:21 - mmengine - INFO - Exp name: convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1_20230605_024744 +2023/06/05 06:12:21 - mmengine - INFO - Saving checkpoint at 7 epochs +2023/06/05 06:13:24 - mmengine - INFO - Epoch(val) [7][60/60] accuracy/top1: 98.7819 single-label/precision_classwise: [99.90400695800781, 94.89886474609375] single-label/recall_classwise: [98.54596710205078, 99.65116119384766] single-label/f1-score_classwise: [99.2203369140625, 97.21697235107422] data_time: 0.1721 time: 0.5732 +2023/06/05 06:15:00 - mmengine - INFO - Epoch(train) [8][ 100/2056] lr: 8.8009e-05 eta: 1:23:21 time: 0.7470 data_time: 0.2292 memory: 15585 loss: 0.0174 +2023/06/05 06:16:23 - mmengine - INFO - Epoch(train) [8][ 200/2056] lr: 8.5639e-05 eta: 1:21:59 time: 0.8033 data_time: 0.2748 memory: 15585 loss: 0.0183 +2023/06/05 06:17:46 - mmengine - INFO - Epoch(train) [8][ 300/2056] lr: 8.3296e-05 eta: 1:20:37 time: 0.7825 data_time: 0.2507 memory: 15585 loss: 0.0166 +2023/06/05 06:19:13 - mmengine - INFO - Epoch(train) [8][ 400/2056] lr: 8.0983e-05 eta: 1:19:16 time: 0.8123 data_time: 0.2932 memory: 15585 loss: 0.0140 +2023/06/05 06:20:33 - mmengine - INFO - Epoch(train) [8][ 500/2056] lr: 7.8698e-05 eta: 1:17:53 time: 0.7944 data_time: 0.2682 memory: 15585 loss: 0.0149 +2023/06/05 06:21:53 - mmengine - INFO - Epoch(train) [8][ 600/2056] lr: 7.6443e-05 eta: 1:16:29 time: 0.7864 data_time: 0.2670 memory: 15585 loss: 0.0165 +2023/06/05 06:21:59 - mmengine - INFO - Exp name: convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1_20230605_024744 +2023/06/05 06:23:10 - mmengine - INFO - Epoch(train) [8][ 700/2056] lr: 7.4217e-05 eta: 1:15:05 time: 0.7936 data_time: 0.2543 memory: 15585 loss: 0.0125 +2023/06/05 06:24:29 - mmengine - INFO - Epoch(train) [8][ 800/2056] lr: 7.2022e-05 eta: 1:13:41 time: 0.8688 data_time: 0.3485 memory: 15585 loss: 0.0158 +2023/06/05 06:25:48 - mmengine - INFO - Epoch(train) [8][ 900/2056] lr: 6.9859e-05 eta: 1:12:18 time: 0.7784 data_time: 0.2589 memory: 15585 loss: 0.0173 +2023/06/05 06:27:07 - mmengine - INFO - Epoch(train) [8][1000/2056] lr: 6.7727e-05 eta: 1:10:54 time: 0.8168 data_time: 0.2986 memory: 15585 loss: 0.0161 +2023/06/05 06:28:26 - mmengine - INFO - Epoch(train) [8][1100/2056] lr: 6.5626e-05 eta: 1:09:31 time: 0.8004 data_time: 0.1944 memory: 15585 loss: 0.0170 +2023/06/05 06:29:46 - mmengine - INFO - Epoch(train) [8][1200/2056] lr: 6.3559e-05 eta: 1:08:08 time: 0.8158 data_time: 0.2590 memory: 15585 loss: 0.0195 +2023/06/05 06:31:04 - mmengine - INFO - Epoch(train) [8][1300/2056] lr: 6.1524e-05 eta: 1:06:44 time: 0.7625 data_time: 0.2264 memory: 15585 loss: 0.0109 +2023/06/05 06:32:25 - mmengine - INFO - Epoch(train) [8][1400/2056] lr: 5.9523e-05 eta: 1:05:21 time: 0.7905 data_time: 0.2740 memory: 15585 loss: 0.0105 +2023/06/05 06:33:43 - mmengine - INFO - Epoch(train) [8][1500/2056] lr: 5.7556e-05 eta: 1:03:58 time: 0.7921 data_time: 0.2661 memory: 15585 loss: 0.0144 +2023/06/05 06:35:04 - mmengine - INFO - Epoch(train) [8][1600/2056] lr: 5.5623e-05 eta: 1:02:35 time: 0.8158 data_time: 0.2971 memory: 15585 loss: 0.0146 +2023/06/05 06:35:10 - mmengine - INFO - Exp name: convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1_20230605_024744 +2023/06/05 06:36:24 - mmengine - INFO - Epoch(train) [8][1700/2056] lr: 5.3726e-05 eta: 1:01:13 time: 0.8255 data_time: 0.3045 memory: 15585 loss: 0.0172 +2023/06/05 06:37:44 - mmengine - INFO - Epoch(train) [8][1800/2056] lr: 5.1863e-05 eta: 0:59:50 time: 0.7780 data_time: 0.2567 memory: 15585 loss: 0.0195 +2023/06/05 06:39:12 - mmengine - INFO - Epoch(train) [8][1900/2056] lr: 5.0036e-05 eta: 0:58:29 time: 0.7729 data_time: 0.2469 memory: 15585 loss: 0.0102 +2023/06/05 06:40:30 - mmengine - INFO - Epoch(train) [8][2000/2056] lr: 4.8246e-05 eta: 0:57:06 time: 0.7714 data_time: 0.2501 memory: 15585 loss: 0.0173 +2023/06/05 06:41:15 - mmengine - INFO - Exp name: convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1_20230605_024744 +2023/06/05 06:41:15 - mmengine - INFO - Saving checkpoint at 8 epochs +2023/06/05 06:42:18 - mmengine - INFO - Epoch(val) [8][60/60] accuracy/top1: 99.1344 single-label/precision_classwise: [99.89384460449219, 96.44975280761719] single-label/recall_classwise: [99.00469207763672, 99.6124038696289] single-label/f1-score_classwise: [99.44728088378906, 98.00556945800781] data_time: 0.1758 time: 0.5721 +2023/06/05 06:43:44 - mmengine - INFO - Epoch(train) [9][ 100/2056] lr: 4.5526e-05 eta: 0:54:58 time: 0.7902 data_time: 0.0962 memory: 15585 loss: 0.0107 +2023/06/05 06:45:06 - mmengine - INFO - Epoch(train) [9][ 200/2056] lr: 4.3830e-05 eta: 0:53:36 time: 0.7944 data_time: 0.2691 memory: 15585 loss: 0.0129 +2023/06/05 06:46:24 - mmengine - INFO - Epoch(train) [9][ 300/2056] lr: 4.2172e-05 eta: 0:52:13 time: 0.7631 data_time: 0.2382 memory: 15585 loss: 0.0162 +2023/06/05 06:47:43 - mmengine - INFO - Epoch(train) [9][ 400/2056] lr: 4.0551e-05 eta: 0:50:50 time: 0.7814 data_time: 0.2549 memory: 15585 loss: 0.0122 +2023/06/05 06:49:03 - mmengine - INFO - Epoch(train) [9][ 500/2056] lr: 3.8969e-05 eta: 0:49:27 time: 0.7537 data_time: 0.2330 memory: 15585 loss: 0.0141 +2023/06/05 06:49:46 - mmengine - INFO - Exp name: convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1_20230605_024744 +2023/06/05 06:50:22 - mmengine - INFO - Epoch(train) [9][ 600/2056] lr: 3.7426e-05 eta: 0:48:04 time: 0.7691 data_time: 0.2440 memory: 15585 loss: 0.0113 +2023/06/05 06:51:41 - mmengine - INFO - Epoch(train) [9][ 700/2056] lr: 3.5922e-05 eta: 0:46:42 time: 0.7857 data_time: 0.2666 memory: 15585 loss: 0.0098 +2023/06/05 06:52:59 - mmengine - INFO - Epoch(train) [9][ 800/2056] lr: 3.4457e-05 eta: 0:45:19 time: 0.7662 data_time: 0.2273 memory: 15585 loss: 0.0162 +2023/06/05 06:54:20 - mmengine - INFO - Epoch(train) [9][ 900/2056] lr: 3.3032e-05 eta: 0:43:56 time: 0.8337 data_time: 0.1884 memory: 15585 loss: 0.0143 +2023/06/05 06:55:54 - mmengine - INFO - Epoch(train) [9][1000/2056] lr: 3.1648e-05 eta: 0:42:36 time: 1.1181 data_time: 0.0019 memory: 15585 loss: 0.0107 +2023/06/05 06:57:19 - mmengine - INFO - Epoch(train) [9][1100/2056] lr: 3.0304e-05 eta: 0:41:15 time: 0.9032 data_time: 0.0022 memory: 15585 loss: 0.0209 +2023/06/05 06:59:24 - mmengine - INFO - Epoch(train) [9][1200/2056] lr: 2.9000e-05 eta: 0:40:00 time: 0.8974 data_time: 0.0025 memory: 15585 loss: 0.0085 +2023/06/05 07:00:56 - mmengine - INFO - Epoch(train) [9][1300/2056] lr: 2.7738e-05 eta: 0:38:39 time: 1.0430 data_time: 0.0043 memory: 15585 loss: 0.0084 +2023/06/05 07:02:14 - mmengine - INFO - Epoch(train) [9][1400/2056] lr: 2.6517e-05 eta: 0:37:16 time: 0.8371 data_time: 0.0017 memory: 15585 loss: 0.0205 +2023/06/05 07:03:35 - mmengine - INFO - Epoch(train) [9][1500/2056] lr: 2.5338e-05 eta: 0:35:53 time: 0.7816 data_time: 0.0019 memory: 15585 loss: 0.0097 +2023/06/05 07:04:26 - mmengine - INFO - Exp name: convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1_20230605_024744 +2023/06/05 07:05:10 - mmengine - INFO - Epoch(train) [9][1600/2056] lr: 2.4200e-05 eta: 0:34:32 time: 0.9838 data_time: 0.0016 memory: 15585 loss: 0.0118 +2023/06/05 07:07:04 - mmengine - INFO - Epoch(train) [9][1700/2056] lr: 2.3105e-05 eta: 0:33:14 time: 1.4381 data_time: 0.0018 memory: 15585 loss: 0.0097 +2023/06/05 07:08:23 - mmengine - INFO - Epoch(train) [9][1800/2056] lr: 2.2053e-05 eta: 0:31:51 time: 0.7416 data_time: 0.0016 memory: 15585 loss: 0.0128 +2023/06/05 07:10:00 - mmengine - INFO - Epoch(train) [9][1900/2056] lr: 2.1043e-05 eta: 0:30:30 time: 0.8639 data_time: 0.0015 memory: 15585 loss: 0.0155 +2023/06/05 07:11:22 - mmengine - INFO - Epoch(train) [9][2000/2056] lr: 2.0076e-05 eta: 0:29:07 time: 0.7752 data_time: 0.0019 memory: 15585 loss: 0.0127 +2023/06/05 07:12:04 - mmengine - INFO - Exp name: convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1_20230605_024744 +2023/06/05 07:12:04 - mmengine - INFO - Saving checkpoint at 9 epochs +2023/06/05 07:13:07 - mmengine - INFO - Epoch(val) [9][60/60] accuracy/top1: 99.2850 single-label/precision_classwise: [99.88558959960938, 97.14155578613281] single-label/recall_classwise: [99.20459747314453, 99.5813980102539] single-label/f1-score_classwise: [99.54393005371094, 98.34634399414062] data_time: 0.1779 time: 0.5769 +2023/06/05 07:14:34 - mmengine - INFO - Epoch(train) [10][ 100/2056] lr: 1.8654e-05 eta: 0:26:58 time: 0.8109 data_time: 0.2893 memory: 15585 loss: 0.0071 +2023/06/05 07:15:56 - mmengine - INFO - Epoch(train) [10][ 200/2056] lr: 1.7798e-05 eta: 0:25:35 time: 0.8038 data_time: 0.2820 memory: 15585 loss: 0.0104 +2023/06/05 07:17:20 - mmengine - INFO - Epoch(train) [10][ 300/2056] lr: 1.6985e-05 eta: 0:24:12 time: 0.7790 data_time: 0.2602 memory: 15585 loss: 0.0068 +2023/06/05 07:18:41 - mmengine - INFO - Epoch(train) [10][ 400/2056] lr: 1.6217e-05 eta: 0:22:49 time: 0.8380 data_time: 0.2240 memory: 15585 loss: 0.0107 +2023/06/05 07:20:04 - mmengine - INFO - Exp name: convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1_20230605_024744 +2023/06/05 07:20:07 - mmengine - INFO - Epoch(train) [10][ 500/2056] lr: 1.5493e-05 eta: 0:21:27 time: 0.7857 data_time: 0.2620 memory: 15585 loss: 0.0094 +2023/06/05 07:21:27 - mmengine - INFO - Epoch(train) [10][ 600/2056] lr: 1.4813e-05 eta: 0:20:04 time: 0.8414 data_time: 0.3159 memory: 15585 loss: 0.0158 +2023/06/05 07:22:53 - mmengine - INFO - Epoch(train) [10][ 700/2056] lr: 1.4177e-05 eta: 0:18:42 time: 0.7514 data_time: 0.2331 memory: 15585 loss: 0.0099 +2023/06/05 07:24:14 - mmengine - INFO - Epoch(train) [10][ 800/2056] lr: 1.3586e-05 eta: 0:17:19 time: 0.8071 data_time: 0.2902 memory: 15585 loss: 0.0091 +2023/06/05 07:25:32 - mmengine - INFO - Epoch(train) [10][ 900/2056] lr: 1.3039e-05 eta: 0:15:56 time: 0.8341 data_time: 0.3169 memory: 15585 loss: 0.0111 +2023/06/05 07:26:55 - mmengine - INFO - Epoch(train) [10][1000/2056] lr: 1.2538e-05 eta: 0:14:33 time: 0.7921 data_time: 0.2673 memory: 15585 loss: 0.0189 +2023/06/05 07:28:19 - mmengine - INFO - Epoch(train) [10][1100/2056] lr: 1.2081e-05 eta: 0:13:10 time: 0.8007 data_time: 0.2791 memory: 15585 loss: 0.0098 +2023/06/05 07:29:38 - mmengine - INFO - Epoch(train) [10][1200/2056] lr: 1.1670e-05 eta: 0:11:47 time: 0.8460 data_time: 0.3270 memory: 15585 loss: 0.0103 +2023/06/05 07:31:00 - mmengine - INFO - Epoch(train) [10][1300/2056] lr: 1.1303e-05 eta: 0:10:25 time: 0.8323 data_time: 0.3109 memory: 15585 loss: 0.0094 +2023/06/05 07:32:20 - mmengine - INFO - Epoch(train) [10][1400/2056] lr: 1.0982e-05 eta: 0:09:02 time: 0.8083 data_time: 0.2901 memory: 15585 loss: 0.0148 +2023/06/05 07:33:37 - mmengine - INFO - Exp name: convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1_20230605_024744 +2023/06/05 07:33:39 - mmengine - INFO - Epoch(train) [10][1500/2056] lr: 1.0706e-05 eta: 0:07:39 time: 0.8064 data_time: 0.2735 memory: 15585 loss: 0.0069 +2023/06/05 07:35:07 - mmengine - INFO - Epoch(train) [10][1600/2056] lr: 1.0475e-05 eta: 0:06:17 time: 0.7850 data_time: 0.0018 memory: 15585 loss: 0.0101 +2023/06/05 07:36:29 - mmengine - INFO - Epoch(train) [10][1700/2056] lr: 1.0290e-05 eta: 0:04:54 time: 0.8510 data_time: 0.1622 memory: 15585 loss: 0.0088 +2023/06/05 07:37:47 - mmengine - INFO - Epoch(train) [10][1800/2056] lr: 1.0150e-05 eta: 0:03:31 time: 0.7695 data_time: 0.1050 memory: 15585 loss: 0.0107 +2023/06/05 07:39:09 - mmengine - INFO - Epoch(train) [10][1900/2056] lr: 1.0056e-05 eta: 0:02:08 time: 0.8865 data_time: 0.3398 memory: 15585 loss: 0.0108 +2023/06/05 07:40:31 - mmengine - INFO - Epoch(train) [10][2000/2056] lr: 1.0007e-05 eta: 0:00:46 time: 0.8584 data_time: 0.3367 memory: 15585 loss: 0.0096 +2023/06/05 07:41:16 - mmengine - INFO - Exp name: convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1_20230605_024744 +2023/06/05 07:41:16 - mmengine - INFO - Saving checkpoint at 10 epochs +2023/06/05 07:42:18 - mmengine - INFO - Epoch(val) [10][60/60] accuracy/top1: 99.4141 single-label/precision_classwise: [99.85625457763672, 97.81978607177734] single-label/recall_classwise: [99.39818572998047, 99.47286987304688] single-label/f1-score_classwise: [99.62669372558594, 98.639404296875] data_time: 0.1702 time: 0.5692 diff --git a/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1/20230605_024744/vis_data/20230605_024744.json b/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1/20230605_024744/vis_data/20230605_024744.json new file mode 100644 index 0000000000000000000000000000000000000000..e456628e7f4166d68e77fe1aa393d4c8ed1b4dfd --- /dev/null +++ b/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1/20230605_024744/vis_data/20230605_024744.json @@ -0,0 +1,210 @@ +{"lr": 0.00039997768893775784, "data_time": 0.2831082820892334, "loss": 0.1379379466176033, "time": 0.8092506408691407, "epoch": 1, "memory": 15587, "step": 100} +{"lr": 0.00039990985724436777, "data_time": 0.3028762102127075, "loss": 0.1013808086514473, "time": 0.8239328861236572, "epoch": 1, "memory": 15587, "step": 200} +{"lr": 0.00039979651848055537, "data_time": 0.28708322048187257, "loss": 0.11428516358137131, "time": 0.8114557981491088, "epoch": 1, "memory": 15587, "step": 300} +{"lr": 0.0003996376991083742, "data_time": 0.2839618444442749, "loss": 0.0828025195747614, "time": 0.805398964881897, "epoch": 1, "memory": 15587, "step": 400} +{"lr": 0.00039943343620857754, "data_time": 0.3001241207122803, "loss": 0.08183570690453053, "time": 0.8163015127182007, "epoch": 1, "memory": 15587, "step": 500} +{"lr": 0.0003991837774719618, "data_time": 0.2885365724563599, "loss": 0.07526886537671089, "time": 0.8103415012359619, "epoch": 1, "memory": 15587, "step": 600} +{"lr": 0.00039888878118822876, "data_time": 0.285597825050354, "loss": 0.07299174517393112, "time": 0.8153990745544434, "epoch": 1, "memory": 15587, "step": 700} +{"lr": 0.0003985485162323803, "data_time": 0.33648607730865476, "loss": 0.06248037666082382, "time": 0.8638121843338012, "epoch": 1, "memory": 15587, "step": 800} +{"lr": 0.0003981630620486352, "data_time": 0.001656317710876465, "loss": 0.06533353365957736, "time": 0.8840745449066162, "epoch": 1, "memory": 15587, "step": 900} +{"lr": 0.0003977325086318812, "data_time": 0.0019359350204467773, "loss": 0.07843562215566635, "time": 0.7775065422058105, "epoch": 1, "memory": 15587, "step": 1000} +{"lr": 0.00039725695650666204, "data_time": 0.0030431509017944335, "loss": 0.0714233446866274, "time": 0.7849505424499512, "epoch": 1, "memory": 15587, "step": 1100} +{"lr": 0.0003967365167037102, "data_time": 0.001819944381713867, "loss": 0.05900231897830963, "time": 0.7213258743286133, "epoch": 1, "memory": 15587, "step": 1200} +{"lr": 0.0003961713107340207, "data_time": 0.1318582534790039, "loss": 0.06752421371638775, "time": 0.802603816986084, "epoch": 1, "memory": 15587, "step": 1300} +{"lr": 0.0003955614705604829, "data_time": 0.30567810535430906, "loss": 0.06470905505120754, "time": 0.8236953973770141, "epoch": 1, "memory": 15587, "step": 1400} +{"lr": 0.00039490713856706923, "data_time": 0.206199312210083, "loss": 0.06124845277518034, "time": 0.7851998805999756, "epoch": 1, "memory": 15587, "step": 1500} +{"lr": 0.00039420846752559095, "data_time": 0.030395698547363282, "loss": 0.05094659887254238, "time": 0.7950512170791626, "epoch": 1, "memory": 15587, "step": 1600} +{"lr": 0.0003934656205600326, "data_time": 0.3519749164581299, "loss": 0.05878002867102623, "time": 0.8704967021942138, "epoch": 1, "memory": 15587, "step": 1700} +{"lr": 0.0003926787711084616, "data_time": 0.28209857940673827, "loss": 0.050562608428299426, "time": 0.8030772924423217, "epoch": 1, "memory": 15587, "step": 1800} +{"lr": 0.0003918481028825385, "data_time": 0.25635926723480223, "loss": 0.05066502410918474, "time": 0.7872023820877075, "epoch": 1, "memory": 15587, "step": 1900} +{"lr": 0.0003909738098246218, "data_time": 0.2422402620315552, "loss": 0.05767842046916485, "time": 0.7734029531478882, "epoch": 1, "memory": 15587, "step": 2000} +{"accuracy/top1": 81.10157012939453, "data_time": 0.1894524892171224, "time": 0.6320564905802409, "step": 1} +{"lr": 0.0003895232896423554, "data_time": 0.29793739318847656, "loss": 0.04124173987656832, "time": 0.8159654140472412, "epoch": 2, "memory": 15587, "step": 2156} +{"lr": 0.00038853827118924355, "data_time": 0.28956103324890137, "loss": 0.04970041885972023, "time": 0.8290251016616821, "epoch": 2, "memory": 15585, "step": 2256} +{"lr": 0.0003875104006754391, "data_time": 0.26585290431976316, "loss": 0.04555199816823006, "time": 0.7842116832733155, "epoch": 2, "memory": 15585, "step": 2356} +{"lr": 0.00038643991808560204, "data_time": 0.015763568878173827, "loss": 0.04397035613656044, "time": 0.7426855325698852, "epoch": 2, "memory": 15585, "step": 2456} +{"lr": 0.00038532707335335494, "data_time": 0.17856643199920655, "loss": 0.04511376321315765, "time": 0.8405045509338379, "epoch": 2, "memory": 15585, "step": 2556} +{"lr": 0.00038417212630292743, "data_time": 0.0018572568893432616, "loss": 0.04297624509781599, "time": 0.8418952941894531, "epoch": 2, "memory": 15585, "step": 2656} +{"lr": 0.00038297534658849464, "data_time": 0.0023754596710205077, "loss": 0.034093549475073814, "time": 0.779910683631897, "epoch": 2, "memory": 15585, "step": 2756} +{"lr": 0.00038173701363121816, "data_time": 0.0017132520675659179, "loss": 0.049136029183864595, "time": 0.8440181016921997, "epoch": 2, "memory": 15585, "step": 2856} +{"lr": 0.00038045741655400836, "data_time": 0.26050171852111814, "loss": 0.04158437177538872, "time": 0.8751527070999146, "epoch": 2, "memory": 15585, "step": 2956} +{"lr": 0.00037913685411401933, "data_time": 0.29927406311035154, "loss": 0.03589790295809507, "time": 0.817652678489685, "epoch": 2, "memory": 15585, "step": 3056} +{"lr": 0.00037777563463289686, "data_time": 0.24219095706939697, "loss": 0.03650339599698782, "time": 0.7664620161056519, "epoch": 2, "memory": 15585, "step": 3156} +{"lr": 0.0003763740759247917, "data_time": 0.22861981391906738, "loss": 0.044386313296854496, "time": 0.7522628784179688, "epoch": 2, "memory": 15585, "step": 3256} +{"lr": 0.0003749325052221578, "data_time": 0.26761460304260254, "loss": 0.045708251930773255, "time": 0.7945474863052369, "epoch": 2, "memory": 15585, "step": 3356} +{"lr": 0.00037345125909934995, "data_time": 0.2472461938858032, "loss": 0.04041445031762123, "time": 0.7682995796203613, "epoch": 2, "memory": 15585, "step": 3456} +{"lr": 0.00037193068339404196, "data_time": 0.26855196952819826, "loss": 0.05374577697366476, "time": 0.7874505519866943, "epoch": 2, "memory": 15585, "step": 3556} +{"lr": 0.0003703711331264808, "data_time": 0.2570746183395386, "loss": 0.02681981734931469, "time": 0.775376558303833, "epoch": 2, "memory": 15585, "step": 3656} +{"lr": 0.0003687729724165983, "data_time": 0.2340846538543701, "loss": 0.04499406293034554, "time": 0.7571647644042969, "epoch": 2, "memory": 15585, "step": 3756} +{"lr": 0.000367136574398996, "data_time": 0.32793877124786375, "loss": 0.042623497731983664, "time": 0.8512455224990845, "epoch": 2, "memory": 15585, "step": 3856} +{"lr": 0.00036546232113582794, "data_time": 0.2526634454727173, "loss": 0.04397753719240427, "time": 0.7758017778396606, "epoch": 2, "memory": 15585, "step": 3956} +{"lr": 0.0003637506035275962, "data_time": 0.2568434476852417, "loss": 0.044520477391779424, "time": 0.7711585283279419, "epoch": 2, "memory": 15585, "step": 4056} +{"accuracy/top1": 91.12258911132812, "data_time": 0.17079986118879475, "time": 0.56885121298618, "step": 2} +{"lr": 0.00036100646510740043, "data_time": 0.08590762615203858, "loss": 0.041065428592264655, "time": 0.9222569227218628, "epoch": 3, "memory": 15585, "step": 4212} +{"lr": 0.0003592006803637847, "data_time": 0.001749420166015625, "loss": 0.04609261769801378, "time": 0.8253833532333374, "epoch": 3, "memory": 15585, "step": 4312} +{"lr": 0.00035735889322747414, "data_time": 0.0016744613647460937, "loss": 0.03991765761747956, "time": 0.7563973188400268, "epoch": 3, "memory": 15585, "step": 4412} +{"lr": 0.0003554815337143657, "data_time": 0.09968936443328857, "loss": 0.028875134605914354, "time": 0.8285691976547241, "epoch": 3, "memory": 15585, "step": 4512} +{"lr": 0.00035356904014570526, "data_time": 0.015761756896972658, "loss": 0.03384460154920817, "time": 0.7644056558609009, "epoch": 3, "memory": 15585, "step": 4612} +{"lr": 0.00035162185904575157, "data_time": 0.3038105249404907, "loss": 0.035961291566491124, "time": 0.8292667865753174, "epoch": 3, "memory": 15585, "step": 4712} +{"lr": 0.0003496404450375223, "data_time": 0.2916796922683716, "loss": 0.03861950431019068, "time": 0.8274330377578736, "epoch": 3, "memory": 15585, "step": 4812} +{"lr": 0.0003476252607366498, "data_time": 0.3371662855148315, "loss": 0.035083943791687486, "time": 0.8627967834472656, "epoch": 3, "memory": 15585, "step": 4912} +{"lr": 0.0003455767766433698, "data_time": 0.2681027889251709, "loss": 0.04419847317039967, "time": 0.7997649192810059, "epoch": 3, "memory": 15585, "step": 5012} +{"lr": 0.0003434954710326717, "data_time": 0.2596555233001709, "loss": 0.03167481794953346, "time": 0.7845546007156372, "epoch": 3, "memory": 15585, "step": 5112} +{"lr": 0.00034138182984263203, "data_time": 0.3108095169067383, "loss": 0.03817869685590267, "time": 0.8332990407943726, "epoch": 3, "memory": 15585, "step": 5212} +{"lr": 0.0003392363465609565, "data_time": 0.23287680149078369, "loss": 0.03682389818131924, "time": 0.7543991804122925, "epoch": 3, "memory": 15585, "step": 5312} +{"lr": 0.0003370595221097656, "data_time": 0.24555511474609376, "loss": 0.03068116307258606, "time": 0.7744240999221802, "epoch": 3, "memory": 15585, "step": 5412} +{"lr": 0.00033485186472863657, "data_time": 0.2514724493026733, "loss": 0.03019562130793929, "time": 0.7751284599304199, "epoch": 3, "memory": 15585, "step": 5512} +{"lr": 0.00033261388985594397, "data_time": 0.22008254528045654, "loss": 0.04136306568980217, "time": 0.7380250453948974, "epoch": 3, "memory": 15585, "step": 5612} +{"lr": 0.0003303461200085145, "data_time": 0.21104114055633544, "loss": 0.0332688445225358, "time": 0.7552468776702881, "epoch": 3, "memory": 15585, "step": 5712} +{"lr": 0.0003280490846596322, "data_time": 0.18828837871551513, "loss": 0.036070290487259626, "time": 0.7050377130508423, "epoch": 3, "memory": 15585, "step": 5812} +{"lr": 0.00032572332011541707, "data_time": 0.2858557224273682, "loss": 0.04054839164018631, "time": 0.8197050809860229, "epoch": 3, "memory": 15585, "step": 5912} +{"lr": 0.00032336936938961124, "data_time": 0.2519676923751831, "loss": 0.023237418197095395, "time": 0.7868886470794678, "epoch": 3, "memory": 15585, "step": 6012} +{"lr": 0.00032098778207679645, "data_time": 0.2727648735046387, "loss": 0.03557647168636322, "time": 0.8068763256072998, "epoch": 3, "memory": 15585, "step": 6112} +{"accuracy/top1": 94.3531494140625, "data_time": 0.18577750393601714, "time": 0.5866230706699559, "step": 3} +{"lr": 0.0003172186408067781, "data_time": 0.31164615154266356, "loss": 0.035311220400035384, "time": 0.8347751379013062, "epoch": 4, "memory": 15585, "step": 6268} +{"lr": 0.00031476885256500535, "data_time": 0.30954408645629883, "loss": 0.03334829155355692, "time": 0.8263779640197754, "epoch": 4, "memory": 15585, "step": 6368} +{"lr": 0.00031229343576357664, "data_time": 0.27940547466278076, "loss": 0.03248629802837968, "time": 0.7942147016525268, "epoch": 4, "memory": 15585, "step": 6468} +{"lr": 0.00030979296835667217, "data_time": 0.3086205244064331, "loss": 0.023935234546661376, "time": 0.8264507055282593, "epoch": 4, "memory": 15585, "step": 6568} +{"lr": 0.000307268034147225, "data_time": 0.2922163248062134, "loss": 0.03954189233481884, "time": 0.8144617557525635, "epoch": 4, "memory": 15585, "step": 6668} +{"lr": 0.00030471922265061713, "data_time": 0.24466795921325685, "loss": 0.0366573590785265, "time": 0.7668239593505859, "epoch": 4, "memory": 15585, "step": 6768} +{"lr": 0.00030214712895704036, "data_time": 0.34331541061401366, "loss": 0.02575256349518895, "time": 0.8645653009414673, "epoch": 4, "memory": 15585, "step": 6868} +{"lr": 0.00029955235359255527, "data_time": 0.29909601211547854, "loss": 0.018522901460528374, "time": 0.8239031553268432, "epoch": 4, "memory": 15585, "step": 6968} +{"lr": 0.00029693550237888314, "data_time": 0.30478610992431643, "loss": 0.029821417992934585, "time": 0.8184063911437989, "epoch": 4, "memory": 15585, "step": 7068} +{"lr": 0.0002942971862919601, "data_time": 0.2361917734146118, "loss": 0.02400244167074561, "time": 0.7563504934310913, "epoch": 4, "memory": 15585, "step": 7168} +{"lr": 0.0002916380213192881, "data_time": 0.23835356235504152, "loss": 0.0304020157083869, "time": 0.7577677488327026, "epoch": 4, "memory": 15585, "step": 7268} +{"lr": 0.00028895862831611433, "data_time": 0.3100778579711914, "loss": 0.02480067331343889, "time": 0.8320559501647949, "epoch": 4, "memory": 15585, "step": 7368} +{"lr": 0.0002862596328604766, "data_time": 0.2189112901687622, "loss": 0.033387305587530135, "time": 0.7994933128356934, "epoch": 4, "memory": 15585, "step": 7468} +{"lr": 0.0002835416651071449, "data_time": 0.20606186389923095, "loss": 0.025601667817682028, "time": 0.7394556760787964, "epoch": 4, "memory": 15585, "step": 7568} +{"lr": 0.00028080535964049476, "data_time": 0.22684597969055176, "loss": 0.023305664584040642, "time": 0.7484435558319091, "epoch": 4, "memory": 15585, "step": 7668} +{"lr": 0.0002780513553263448, "data_time": 0.02597343921661377, "loss": 0.03611135305836797, "time": 0.9599784851074219, "epoch": 4, "memory": 15585, "step": 7768} +{"lr": 0.0002752802951627973, "data_time": 0.0018501520156860352, "loss": 0.03020003205165267, "time": 0.815102481842041, "epoch": 4, "memory": 15585, "step": 7868} +{"lr": 0.0002724928261301117, "data_time": 0.0017300844192504883, "loss": 0.023491135472431778, "time": 0.9153478860855102, "epoch": 4, "memory": 15585, "step": 7968} +{"lr": 0.00026968959903964925, "data_time": 0.0017294168472290039, "loss": 0.02911972114816308, "time": 0.8375027894973754, "epoch": 4, "memory": 15585, "step": 8068} +{"lr": 0.0002668712683819236, "data_time": 0.0016197681427001954, "loss": 0.023819428961724044, "time": 0.8518417596817016, "epoch": 4, "memory": 15585, "step": 8168} +{"accuracy/top1": 96.1686782836914, "data_time": 0.17289388766054248, "time": 0.570914119970603, "step": 4} +{"lr": 0.00026244607407074216, "data_time": 0.22203283309936522, "loss": 0.024537878390401602, "time": 0.7770836353302002, "epoch": 5, "memory": 15585, "step": 8324} +{"lr": 0.00025959208467259544, "data_time": 0.20585229396820068, "loss": 0.02669526357203722, "time": 0.7901898145675659, "epoch": 5, "memory": 15585, "step": 8424} +{"lr": 0.0002567253492498155, "data_time": 0.27142164707183836, "loss": 0.02051792209967971, "time": 0.7884191989898681, "epoch": 5, "memory": 15585, "step": 8524} +{"lr": 0.00025384653712068393, "data_time": 0.2411658525466919, "loss": 0.025866320729255675, "time": 0.757617998123169, "epoch": 5, "memory": 15585, "step": 8624} +{"lr": 0.00025095632042312373, "data_time": 0.2308708667755127, "loss": 0.02373034837655723, "time": 0.8023012399673461, "epoch": 5, "memory": 15585, "step": 8724} +{"lr": 0.00024805537395776515, "data_time": 0.0018132925033569336, "loss": 0.027906367275863885, "time": 0.7825748920440674, "epoch": 5, "memory": 15585, "step": 8824} +{"lr": 0.00024514437503039924, "data_time": 0.19741315841674806, "loss": 0.02937898728996515, "time": 0.7856703281402588, "epoch": 5, "memory": 15585, "step": 8924} +{"lr": 0.0002422240032938408, "data_time": 0.14554126262664796, "loss": 0.02580722509883344, "time": 0.8777301788330079, "epoch": 5, "memory": 15585, "step": 9024} +{"lr": 0.00023929494058924557, "data_time": 0.270503830909729, "loss": 0.02230552537366748, "time": 0.8500177383422851, "epoch": 5, "memory": 15585, "step": 9124} +{"lr": 0.00023635787078691464, "data_time": 0.26755404472351074, "loss": 0.02209522812627256, "time": 0.7864372968673706, "epoch": 5, "memory": 15585, "step": 9224} +{"lr": 0.0002334134796266267, "data_time": 0.001841259002685547, "loss": 0.02020230023190379, "time": 0.7152056694030762, "epoch": 5, "memory": 15585, "step": 9324} +{"lr": 0.00023046245455753318, "data_time": 0.04827768802642822, "loss": 0.024352390971034765, "time": 0.824299955368042, "epoch": 5, "memory": 15585, "step": 9424} +{"lr": 0.0002275054845776535, "data_time": 0.2274636745452881, "loss": 0.023816074430942535, "time": 0.7456465244293213, "epoch": 5, "memory": 15585, "step": 9524} +{"lr": 0.00022454326007301025, "data_time": 0.27688424587249755, "loss": 0.029075198993086814, "time": 0.8006130695343018, "epoch": 5, "memory": 15585, "step": 9624} +{"lr": 0.00022157647265643937, "data_time": 0.25865755081176756, "loss": 0.020016076928004622, "time": 0.7812500238418579, "epoch": 5, "memory": 15585, "step": 9724} +{"lr": 0.00021860581500611435, "data_time": 0.2557729005813599, "loss": 0.02665470065549016, "time": 0.7762583732604981, "epoch": 5, "memory": 15585, "step": 9824} +{"lr": 0.00021563198070382117, "data_time": 0.28117234706878663, "loss": 0.022489734273403884, "time": 0.8034007310867309, "epoch": 5, "memory": 15585, "step": 9924} +{"lr": 0.00021265566407302255, "data_time": 0.25601351261138916, "loss": 0.02712933961302042, "time": 0.7830763101577759, "epoch": 5, "memory": 15585, "step": 10024} +{"lr": 0.00020967756001674922, "data_time": 0.29178638458251954, "loss": 0.02838323647156358, "time": 0.8185221195220947, "epoch": 5, "memory": 15585, "step": 10124} +{"lr": 0.0002066983638553557, "data_time": 0.2809009075164795, "loss": 0.023266431502997877, "time": 0.8061702013015747, "epoch": 5, "memory": 15585, "step": 10224} +{"accuracy/top1": 97.35697174072266, "data_time": 0.16911720838703093, "time": 0.5697144015890653, "step": 5} +{"lr": 0.0002020502853543317, "data_time": 0.2810657978057861, "loss": 0.027179038245230915, "time": 0.8111610174179077, "epoch": 6, "memory": 15585, "step": 10380} +{"lr": 0.000199071463167015, "data_time": 0.26308317184448243, "loss": 0.020010809879750013, "time": 0.7848746538162231, "epoch": 6, "memory": 15585, "step": 10480} +{"lr": 0.00019609402515978532, "data_time": 0.23099541664123535, "loss": 0.026591809932142496, "time": 0.7869767427444458, "epoch": 6, "memory": 15585, "step": 10580} +{"lr": 0.00019311866649748968, "data_time": 0.21308507919311523, "loss": 0.026091505959630012, "time": 0.7677872180938721, "epoch": 6, "memory": 15585, "step": 10680} +{"lr": 0.00019014608185949442, "data_time": 0.2908239126205444, "loss": 0.02376266596838832, "time": 0.8316193342208862, "epoch": 6, "memory": 15585, "step": 10780} +{"lr": 0.00018717696527749397, "data_time": 0.26842200756073, "loss": 0.02511770874261856, "time": 0.7919795751571655, "epoch": 6, "memory": 15585, "step": 10880} +{"lr": 0.00018421200997346973, "data_time": 0.24688453674316407, "loss": 0.024106958508491518, "time": 0.7676408767700196, "epoch": 6, "memory": 15585, "step": 10980} +{"lr": 0.00018125190819783787, "data_time": 0.27145895957946775, "loss": 0.02252453090623021, "time": 0.7901381492614746, "epoch": 6, "memory": 15585, "step": 11080} +{"lr": 0.00017829735106782464, "data_time": 0.24628117084503173, "loss": 0.011761882016435266, "time": 0.7876333713531494, "epoch": 6, "memory": 15585, "step": 11180} +{"lr": 0.00017534902840610667, "data_time": 0.30084273815155027, "loss": 0.027595242159441112, "time": 0.8193407535552979, "epoch": 6, "memory": 15585, "step": 11280} +{"lr": 0.00017240762857975229, "data_time": 0.240846848487854, "loss": 0.022845309879630803, "time": 0.7586073875427246, "epoch": 6, "memory": 15585, "step": 11380} +{"lr": 0.00016947383833950319, "data_time": 0.22904131412506104, "loss": 0.01753133866004646, "time": 0.749635124206543, "epoch": 6, "memory": 15585, "step": 11480} +{"lr": 0.0001665483426594338, "data_time": 0.2171614646911621, "loss": 0.01834913678467274, "time": 0.7466540575027466, "epoch": 6, "memory": 15585, "step": 11580} +{"lr": 0.00016363182457702495, "data_time": 0.23504273891448973, "loss": 0.027551061613485216, "time": 0.7546716928482056, "epoch": 6, "memory": 15585, "step": 11680} +{"lr": 0.0001607249650336909, "data_time": 0.28019123077392577, "loss": 0.019759564893320203, "time": 0.813611102104187, "epoch": 6, "memory": 15585, "step": 11780} +{"lr": 0.00015782844271579344, "data_time": 0.21503834724426268, "loss": 0.023741032090038063, "time": 0.7778896808624267, "epoch": 6, "memory": 15585, "step": 11880} +{"lr": 0.0001549429338961849, "data_time": 0.20234980583190917, "loss": 0.01582119045779109, "time": 0.7946874380111695, "epoch": 6, "memory": 15585, "step": 11980} +{"lr": 0.00015206911227631342, "data_time": 0.17572021484375, "loss": 0.019931396888568997, "time": 0.7698912382125854, "epoch": 6, "memory": 15585, "step": 12080} +{"lr": 0.00014920764882892865, "data_time": 0.2740559816360474, "loss": 0.024159587547183038, "time": 0.7960219383239746, "epoch": 6, "memory": 15585, "step": 12180} +{"lr": 0.00014635921164142459, "data_time": 0.17802643775939941, "loss": 0.014993937406688928, "time": 0.7756404638290405, "epoch": 6, "memory": 15585, "step": 12280} +{"accuracy/top1": 98.2605972290039, "data_time": 0.17307630523306425, "time": 0.5701735488703994, "step": 6} +{"lr": 0.00014194323525930575, "data_time": 0.22661159038543702, "loss": 0.024994338117539882, "time": 0.8414340734481811, "epoch": 7, "memory": 15585, "step": 12436} +{"lr": 0.0001391311681531885, "data_time": 0.2701252460479736, "loss": 0.024148925440385938, "time": 0.7996945858001709, "epoch": 7, "memory": 15585, "step": 12536} +{"lr": 0.00013633447993868745, "data_time": 0.2745425462722778, "loss": 0.015681801736354826, "time": 0.8095317840576172, "epoch": 7, "memory": 15585, "step": 12636} +{"lr": 0.00013355382357963566, "data_time": 0.23766775131225587, "loss": 0.015086093964055181, "time": 0.7693504810333252, "epoch": 7, "memory": 15585, "step": 12736} +{"lr": 0.0001307898482967891, "data_time": 0.23037829399108886, "loss": 0.01834228038787842, "time": 0.7463788986206055, "epoch": 7, "memory": 15585, "step": 12836} +{"lr": 0.00012804319941624686, "data_time": 0.2668689012527466, "loss": 0.016991642070934176, "time": 0.7953432083129883, "epoch": 7, "memory": 15585, "step": 12936} +{"lr": 0.00012531451821878227, "data_time": 0.25720701217651365, "loss": 0.017566187493503093, "time": 0.7799966335296631, "epoch": 7, "memory": 15585, "step": 13036} +{"lr": 0.00012260444179011874, "data_time": 0.29000751972198485, "loss": 0.02144675310701132, "time": 0.813347053527832, "epoch": 7, "memory": 15585, "step": 13136} +{"lr": 0.00011991360287218363, "data_time": 0.07197637557983398, "loss": 0.020820771483704446, "time": 0.8050727844238281, "epoch": 7, "memory": 15585, "step": 13236} +{"lr": 0.00011724262971537941, "data_time": 0.12122166156768799, "loss": 0.01619745516218245, "time": 0.873307180404663, "epoch": 7, "memory": 15585, "step": 13336} +{"lr": 0.00011459214593189938, "data_time": 0.087725830078125, "loss": 0.02611830420792103, "time": 0.8096433639526367, "epoch": 7, "memory": 15585, "step": 13436} +{"lr": 0.00011196277035012846, "data_time": 0.0019665956497192383, "loss": 0.016010493924841284, "time": 0.7523234844207763, "epoch": 7, "memory": 15585, "step": 13536} +{"lr": 0.0001093551168701611, "data_time": 0.07163176536560059, "loss": 0.015815067756921052, "time": 0.814582347869873, "epoch": 7, "memory": 15585, "step": 13636} +{"lr": 0.00010676979432046926, "data_time": 0.0018159151077270508, "loss": 0.021654323022812606, "time": 0.8054304838180542, "epoch": 7, "memory": 15585, "step": 13736} +{"lr": 0.00010420740631575464, "data_time": 0.12277102470397949, "loss": 0.012254316126927733, "time": 0.8384614706039428, "epoch": 7, "memory": 15585, "step": 13836} +{"lr": 0.00010166855111601829, "data_time": 0.12544896602630615, "loss": 0.0282936405390501, "time": 0.8113990068435669, "epoch": 7, "memory": 15585, "step": 13936} +{"lr": 9.91538214868799e-05, "data_time": 0.2400291919708252, "loss": 0.009547821804881096, "time": 0.7894332647323609, "epoch": 7, "memory": 15585, "step": 14036} +{"lr": 9.666380456118125e-05, "data_time": 0.21268165111541748, "loss": 0.020644811866804957, "time": 0.7628713846206665, "epoch": 7, "memory": 15585, "step": 14136} +{"lr": 9.419908170190298e-05, "data_time": 0.25474069118499754, "loss": 0.015355687029659748, "time": 0.7784512996673584, "epoch": 7, "memory": 15585, "step": 14236} +{"lr": 9.176022836643002e-05, "data_time": 0.2521061420440674, "loss": 0.018681674776598812, "time": 0.8136514902114869, "epoch": 7, "memory": 15585, "step": 14336} +{"accuracy/top1": 98.78192138671875, "data_time": 0.17212986164405697, "time": 0.5731900207331924, "step": 7} +{"lr": 8.80086206394144e-05, "data_time": 0.22919075489044188, "loss": 0.017404032032936813, "time": 0.7469552040100098, "epoch": 8, "memory": 15585, "step": 14492} +{"lr": 8.563857333566659e-05, "data_time": 0.2747959613800049, "loss": 0.018328665662556885, "time": 0.8032639503479004, "epoch": 8, "memory": 15585, "step": 14592} +{"lr": 8.329639424200186e-05, "data_time": 0.2506705284118652, "loss": 0.016642919182777403, "time": 0.7824928522109985, "epoch": 8, "memory": 15585, "step": 14692} +{"lr": 8.098263020459022e-05, "data_time": 0.29321334362030027, "loss": 0.014005721732974052, "time": 0.8123298406600952, "epoch": 8, "memory": 15585, "step": 14792} +{"lr": 7.869782143532516e-05, "data_time": 0.268187952041626, "loss": 0.014867176348343492, "time": 0.7943655729293824, "epoch": 8, "memory": 15585, "step": 14892} +{"lr": 7.644250138569548e-05, "data_time": 0.26698756217956543, "loss": 0.016485038213431835, "time": 0.7864221811294556, "epoch": 8, "memory": 15585, "step": 14992} +{"lr": 7.4217196622237e-05, "data_time": 0.25427148342132566, "loss": 0.012455874122679234, "time": 0.7935800790786743, "epoch": 8, "memory": 15585, "step": 15092} +{"lr": 7.202242670359085e-05, "data_time": 0.348530650138855, "loss": 0.01578704100102186, "time": 0.8687826633453369, "epoch": 8, "memory": 15585, "step": 15192} +{"lr": 6.98587040591985e-05, "data_time": 0.25888495445251464, "loss": 0.017278297105804086, "time": 0.7783713340759277, "epoch": 8, "memory": 15585, "step": 15292} +{"lr": 6.772653386966062e-05, "data_time": 0.29857597351074217, "loss": 0.016061631310731173, "time": 0.8167670011520386, "epoch": 8, "memory": 15585, "step": 15392} +{"lr": 6.562641394878866e-05, "data_time": 0.19436044692993165, "loss": 0.017046904563903807, "time": 0.800396728515625, "epoch": 8, "memory": 15585, "step": 15492} +{"lr": 6.35588346273771e-05, "data_time": 0.25895874500274657, "loss": 0.019470716919749977, "time": 0.8157904624938965, "epoch": 8, "memory": 15585, "step": 15592} +{"lr": 6.152427863872164e-05, "data_time": 0.22635767459869385, "loss": 0.010894187493249774, "time": 0.7625266790390015, "epoch": 8, "memory": 15585, "step": 15692} +{"lr": 5.952322100591216e-05, "data_time": 0.27396225929260254, "loss": 0.01048377575352788, "time": 0.7904689311981201, "epoch": 8, "memory": 15585, "step": 15792} +{"lr": 5.755612893092559e-05, "data_time": 0.26614985466003416, "loss": 0.014424818428233266, "time": 0.7920905590057373, "epoch": 8, "memory": 15585, "step": 15892} +{"lr": 5.5623461685544754e-05, "data_time": 0.29710395336151124, "loss": 0.01462442851625383, "time": 0.8157669305801392, "epoch": 8, "memory": 15585, "step": 15992} +{"lr": 5.37256705041282e-05, "data_time": 0.3045217752456665, "loss": 0.017198205180466174, "time": 0.8254985094070435, "epoch": 8, "memory": 15585, "step": 16092} +{"lr": 5.186319847825775e-05, "data_time": 0.25671370029449464, "loss": 0.019532240414991974, "time": 0.7780438423156738, "epoch": 8, "memory": 15585, "step": 16192} +{"lr": 5.0036480453286144e-05, "data_time": 0.24693162441253663, "loss": 0.010180334467440844, "time": 0.772924542427063, "epoch": 8, "memory": 15585, "step": 16292} +{"lr": 4.8245942926810745e-05, "data_time": 0.250138521194458, "loss": 0.017302094027400017, "time": 0.7713725805282593, "epoch": 8, "memory": 15585, "step": 16392} +{"accuracy/top1": 99.13443756103516, "data_time": 0.17577528171851986, "time": 0.5720648648308926, "step": 8} +{"lr": 4.5525937358206823e-05, "data_time": 0.09623749256134033, "loss": 0.010653944639489054, "time": 0.7902414798736572, "epoch": 9, "memory": 15585, "step": 16548} +{"lr": 4.382990660001016e-05, "data_time": 0.26913387775421144, "loss": 0.01291049225255847, "time": 0.7944161415100097, "epoch": 9, "memory": 15585, "step": 16648} +{"lr": 4.217150543579113e-05, "data_time": 0.23822276592254638, "loss": 0.01621076031588018, "time": 0.7630682945251465, "epoch": 9, "memory": 15585, "step": 16748} +{"lr": 4.0551121064943564e-05, "data_time": 0.2548978328704834, "loss": 0.012189733469858766, "time": 0.7813825845718384, "epoch": 9, "memory": 15585, "step": 16848} +{"lr": 3.896913181079464e-05, "data_time": 0.2330319404602051, "loss": 0.014061120757833123, "time": 0.753749418258667, "epoch": 9, "memory": 15585, "step": 16948} +{"lr": 3.742590703227469e-05, "data_time": 0.24396300315856934, "loss": 0.011295050429180265, "time": 0.769128966331482, "epoch": 9, "memory": 15585, "step": 17048} +{"lr": 3.592180703768032e-05, "data_time": 0.26656978130340575, "loss": 0.009772074222564698, "time": 0.7856827735900879, "epoch": 9, "memory": 15585, "step": 17148} +{"lr": 3.445718300055071e-05, "data_time": 0.2272563695907593, "loss": 0.016224468871951104, "time": 0.7661733865737915, "epoch": 9, "memory": 15585, "step": 17248} +{"lr": 3.3032376877676173e-05, "data_time": 0.18842051029205323, "loss": 0.014255373179912567, "time": 0.8337079524993897, "epoch": 9, "memory": 15585, "step": 17348} +{"lr": 3.1647721329259246e-05, "data_time": 0.0018663883209228515, "loss": 0.010739016812294722, "time": 1.1180885314941407, "epoch": 9, "memory": 15585, "step": 17448} +{"lr": 3.0303539641245756e-05, "data_time": 0.0022369861602783204, "loss": 0.02085434915497899, "time": 0.903159236907959, "epoch": 9, "memory": 15585, "step": 17548} +{"lr": 2.9000145649844975e-05, "data_time": 0.0024661779403686523, "loss": 0.008549377089366317, "time": 0.8973897218704223, "epoch": 9, "memory": 15585, "step": 17648} +{"lr": 2.7737843668255695e-05, "data_time": 0.004348850250244141, "loss": 0.00835349946282804, "time": 1.0430068731307984, "epoch": 9, "memory": 15585, "step": 17748} +{"lr": 2.6516928415616184e-05, "data_time": 0.0016827583312988281, "loss": 0.02047748272307217, "time": 0.8370586633682251, "epoch": 9, "memory": 15585, "step": 17848} +{"lr": 2.5337684948193946e-05, "data_time": 0.0019025564193725585, "loss": 0.009703512722626328, "time": 0.7816456317901611, "epoch": 9, "memory": 15585, "step": 17948} +{"lr": 2.4200388592830947e-05, "data_time": 0.001636672019958496, "loss": 0.011804910842329264, "time": 0.9837682008743286, "epoch": 9, "memory": 15585, "step": 18048} +{"lr": 2.3105304882661933e-05, "data_time": 0.0017518281936645507, "loss": 0.009747504582628607, "time": 1.4381043195724488, "epoch": 9, "memory": 15585, "step": 18148} +{"lr": 2.20526894951175e-05, "data_time": 0.0016247987747192382, "loss": 0.012831246200948953, "time": 0.7416228771209716, "epoch": 9, "memory": 15585, "step": 18248} +{"lr": 2.1042788192229725e-05, "data_time": 0.0015455007553100586, "loss": 0.01550100278109312, "time": 0.8639036893844605, "epoch": 9, "memory": 15585, "step": 18348} +{"lr": 2.007583676325189e-05, "data_time": 0.0018747568130493164, "loss": 0.012653633020818233, "time": 0.7752285480499268, "epoch": 9, "memory": 15585, "step": 18448} +{"accuracy/top1": 99.28504180908203, "data_time": 0.17789622603869829, "time": 0.5768623117540703, "step": 9} +{"lr": 1.8653686449507784e-05, "data_time": 0.2892716884613037, "loss": 0.007131221424788237, "time": 0.8109148025512696, "epoch": 10, "memory": 15585, "step": 18604} +{"lr": 1.7797691544415563e-05, "data_time": 0.28204195499420165, "loss": 0.010444499505683779, "time": 0.8037574291229248, "epoch": 10, "memory": 15585, "step": 18704} +{"lr": 1.6985404170356326e-05, "data_time": 0.26016457080841066, "loss": 0.00676915175281465, "time": 0.7789924383163452, "epoch": 10, "memory": 15585, "step": 18804} +{"lr": 1.6217013978172993e-05, "data_time": 0.22398149967193604, "loss": 0.010695388168096542, "time": 0.8379882097244262, "epoch": 10, "memory": 15585, "step": 18904} +{"lr": 1.549270036970338e-05, "data_time": 0.26202614307403566, "loss": 0.009390882449224591, "time": 0.7857458353042602, "epoch": 10, "memory": 15585, "step": 19004} +{"lr": 1.4812632455893675e-05, "data_time": 0.315905237197876, "loss": 0.015783053589984775, "time": 0.8413533210754395, "epoch": 10, "memory": 15585, "step": 19104} +{"lr": 1.4176969017315005e-05, "data_time": 0.23308274745941163, "loss": 0.009905860992148519, "time": 0.7513970851898193, "epoch": 10, "memory": 15585, "step": 19204} +{"lr": 1.358585846709161e-05, "data_time": 0.29018981456756593, "loss": 0.00909892893396318, "time": 0.8070831060409546, "epoch": 10, "memory": 15585, "step": 19304} +{"lr": 1.3039438816249855e-05, "data_time": 0.31691243648529055, "loss": 0.011139605101197958, "time": 0.8341225147247314, "epoch": 10, "memory": 15585, "step": 19404} +{"lr": 1.2537837641495561e-05, "data_time": 0.2673105955123901, "loss": 0.01893265973776579, "time": 0.7921293258666993, "epoch": 10, "memory": 15585, "step": 19504} +{"lr": 1.2081172055427927e-05, "data_time": 0.27910146713256834, "loss": 0.009846163867041468, "time": 0.8006896734237671, "epoch": 10, "memory": 15585, "step": 19604} +{"lr": 1.1669548679196218e-05, "data_time": 0.32702112197875977, "loss": 0.010256294440478087, "time": 0.8460445880889893, "epoch": 10, "memory": 15585, "step": 19704} +{"lr": 1.130306361760621e-05, "data_time": 0.3108600378036499, "loss": 0.009392891265451907, "time": 0.8323075294494628, "epoch": 10, "memory": 15585, "step": 19804} +{"lr": 1.0981802436681767e-05, "data_time": 0.29009621143341063, "loss": 0.014776562433689832, "time": 0.8083298921585083, "epoch": 10, "memory": 15585, "step": 19904} +{"lr": 1.0705840143687269e-05, "data_time": 0.2735370874404907, "loss": 0.006929418072104454, "time": 0.8063861131668091, "epoch": 10, "memory": 15585, "step": 20004} +{"lr": 1.0475241169614925e-05, "data_time": 0.001825571060180664, "loss": 0.01014872114174068, "time": 0.7849845886230469, "epoch": 10, "memory": 15585, "step": 20104} +{"lr": 1.0290059354141742e-05, "data_time": 0.16223702430725098, "loss": 0.008769241999834776, "time": 0.8510101079940796, "epoch": 10, "memory": 15585, "step": 20204} +{"lr": 1.015033793305906e-05, "data_time": 0.10497753620147705, "loss": 0.010651576286181807, "time": 0.7695328950881958, "epoch": 10, "memory": 15585, "step": 20304} +{"lr": 1.0056109528177973e-05, "data_time": 0.3397828578948975, "loss": 0.010774973314255476, "time": 0.8865358829498291, "epoch": 10, "memory": 15585, "step": 20404} +{"lr": 1.0007396139712939e-05, "data_time": 0.3367367506027222, "loss": 0.009603068884462117, "time": 0.8584039449691773, "epoch": 10, "memory": 15585, "step": 20504} +{"accuracy/top1": 99.41413116455078, "data_time": 0.1701847607972192, "time": 0.5691666055898197, "step": 10} diff --git a/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1/20230605_024744/vis_data/config.py b/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1/20230605_024744/vis_data/config.py new file mode 100644 index 0000000000000000000000000000000000000000..9611f05a56d42d078592c667e78da62dde067cb2 --- /dev/null +++ b/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1/20230605_024744/vis_data/config.py @@ -0,0 +1,383 @@ +optim_wrapper = dict( + optimizer=dict( + type='AdamW', + lr=0.0004, + weight_decay=0.05, + eps=1e-08, + betas=(0.9, 0.999), + _scope_='mmpretrain'), + paramwise_cfg=dict( + norm_decay_mult=0.0, + bias_decay_mult=0.0, + flat_decay_mult=0.0, + custom_keys=dict({ + '.absolute_pos_embed': dict(decay_mult=0.0), + '.relative_position_bias_table': dict(decay_mult=0.0) + })), + type='AmpOptimWrapper', + dtype='bfloat16', + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=False, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=4096) +model = dict( + type='ImageClassifier', + backbone=dict( + frozen_stages=2, + type='ConvNeXt', + arch='small', + drop_path_rate=0.4, + init_cfg=dict( + type='Pretrained', + checkpoint= + 'https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth', + prefix='backbone')), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=768, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + init_cfg=None), + init_cfg=dict( + type='TruncNormal', layer=['Conv2d', 'Linear'], std=0.02, bias=0.0), + train_cfg=None) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/all1.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/all2.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/all3.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3fake8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/cc1m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3real8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV2-1-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/if-dpmsolver++-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV2-1-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/if-dpmsolver++-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1' diff --git a/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1/20230605_024744/vis_data/events.out.tfevents.1685904480.SH-IDC1-10-140-24-130.196043.0 b/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1/20230605_024744/vis_data/events.out.tfevents.1685904480.SH-IDC1-10-140-24-130.196043.0 new file mode 100644 index 0000000000000000000000000000000000000000..cb7929a297828cbf859c4116722c732b580f5f81 --- /dev/null +++ b/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1/20230605_024744/vis_data/events.out.tfevents.1685904480.SH-IDC1-10-140-24-130.196043.0 @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:263b06e0020486906f06e0737e4dfefc53d92b7fe9e230311810bb15f1bd0453 +size 2692778 diff --git a/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1/20230605_024744/vis_data/scalars.json b/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1/20230605_024744/vis_data/scalars.json new file mode 100644 index 0000000000000000000000000000000000000000..e456628e7f4166d68e77fe1aa393d4c8ed1b4dfd --- /dev/null +++ b/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1/20230605_024744/vis_data/scalars.json @@ -0,0 +1,210 @@ +{"lr": 0.00039997768893775784, "data_time": 0.2831082820892334, "loss": 0.1379379466176033, "time": 0.8092506408691407, "epoch": 1, "memory": 15587, "step": 100} +{"lr": 0.00039990985724436777, "data_time": 0.3028762102127075, "loss": 0.1013808086514473, "time": 0.8239328861236572, "epoch": 1, "memory": 15587, "step": 200} +{"lr": 0.00039979651848055537, "data_time": 0.28708322048187257, "loss": 0.11428516358137131, "time": 0.8114557981491088, "epoch": 1, "memory": 15587, "step": 300} +{"lr": 0.0003996376991083742, "data_time": 0.2839618444442749, "loss": 0.0828025195747614, "time": 0.805398964881897, "epoch": 1, "memory": 15587, "step": 400} +{"lr": 0.00039943343620857754, "data_time": 0.3001241207122803, "loss": 0.08183570690453053, "time": 0.8163015127182007, "epoch": 1, "memory": 15587, "step": 500} +{"lr": 0.0003991837774719618, "data_time": 0.2885365724563599, "loss": 0.07526886537671089, "time": 0.8103415012359619, "epoch": 1, "memory": 15587, "step": 600} +{"lr": 0.00039888878118822876, "data_time": 0.285597825050354, "loss": 0.07299174517393112, "time": 0.8153990745544434, "epoch": 1, "memory": 15587, "step": 700} +{"lr": 0.0003985485162323803, "data_time": 0.33648607730865476, "loss": 0.06248037666082382, "time": 0.8638121843338012, "epoch": 1, "memory": 15587, "step": 800} +{"lr": 0.0003981630620486352, "data_time": 0.001656317710876465, "loss": 0.06533353365957736, "time": 0.8840745449066162, "epoch": 1, "memory": 15587, "step": 900} +{"lr": 0.0003977325086318812, "data_time": 0.0019359350204467773, "loss": 0.07843562215566635, "time": 0.7775065422058105, "epoch": 1, "memory": 15587, "step": 1000} +{"lr": 0.00039725695650666204, "data_time": 0.0030431509017944335, "loss": 0.0714233446866274, "time": 0.7849505424499512, "epoch": 1, "memory": 15587, "step": 1100} +{"lr": 0.0003967365167037102, "data_time": 0.001819944381713867, "loss": 0.05900231897830963, "time": 0.7213258743286133, "epoch": 1, "memory": 15587, "step": 1200} +{"lr": 0.0003961713107340207, "data_time": 0.1318582534790039, "loss": 0.06752421371638775, "time": 0.802603816986084, "epoch": 1, "memory": 15587, "step": 1300} +{"lr": 0.0003955614705604829, "data_time": 0.30567810535430906, "loss": 0.06470905505120754, "time": 0.8236953973770141, "epoch": 1, "memory": 15587, "step": 1400} +{"lr": 0.00039490713856706923, "data_time": 0.206199312210083, "loss": 0.06124845277518034, "time": 0.7851998805999756, "epoch": 1, "memory": 15587, "step": 1500} +{"lr": 0.00039420846752559095, "data_time": 0.030395698547363282, "loss": 0.05094659887254238, "time": 0.7950512170791626, "epoch": 1, "memory": 15587, "step": 1600} +{"lr": 0.0003934656205600326, "data_time": 0.3519749164581299, "loss": 0.05878002867102623, "time": 0.8704967021942138, "epoch": 1, "memory": 15587, "step": 1700} +{"lr": 0.0003926787711084616, "data_time": 0.28209857940673827, "loss": 0.050562608428299426, "time": 0.8030772924423217, "epoch": 1, "memory": 15587, "step": 1800} +{"lr": 0.0003918481028825385, "data_time": 0.25635926723480223, "loss": 0.05066502410918474, "time": 0.7872023820877075, "epoch": 1, "memory": 15587, "step": 1900} +{"lr": 0.0003909738098246218, "data_time": 0.2422402620315552, "loss": 0.05767842046916485, "time": 0.7734029531478882, "epoch": 1, "memory": 15587, "step": 2000} +{"accuracy/top1": 81.10157012939453, "data_time": 0.1894524892171224, "time": 0.6320564905802409, "step": 1} +{"lr": 0.0003895232896423554, "data_time": 0.29793739318847656, "loss": 0.04124173987656832, "time": 0.8159654140472412, "epoch": 2, "memory": 15587, "step": 2156} +{"lr": 0.00038853827118924355, "data_time": 0.28956103324890137, "loss": 0.04970041885972023, "time": 0.8290251016616821, "epoch": 2, "memory": 15585, "step": 2256} +{"lr": 0.0003875104006754391, "data_time": 0.26585290431976316, "loss": 0.04555199816823006, "time": 0.7842116832733155, "epoch": 2, "memory": 15585, "step": 2356} +{"lr": 0.00038643991808560204, "data_time": 0.015763568878173827, "loss": 0.04397035613656044, "time": 0.7426855325698852, "epoch": 2, "memory": 15585, "step": 2456} +{"lr": 0.00038532707335335494, "data_time": 0.17856643199920655, "loss": 0.04511376321315765, "time": 0.8405045509338379, "epoch": 2, "memory": 15585, "step": 2556} +{"lr": 0.00038417212630292743, "data_time": 0.0018572568893432616, "loss": 0.04297624509781599, "time": 0.8418952941894531, "epoch": 2, "memory": 15585, "step": 2656} +{"lr": 0.00038297534658849464, "data_time": 0.0023754596710205077, "loss": 0.034093549475073814, "time": 0.779910683631897, "epoch": 2, "memory": 15585, "step": 2756} +{"lr": 0.00038173701363121816, "data_time": 0.0017132520675659179, "loss": 0.049136029183864595, "time": 0.8440181016921997, "epoch": 2, "memory": 15585, "step": 2856} +{"lr": 0.00038045741655400836, "data_time": 0.26050171852111814, "loss": 0.04158437177538872, "time": 0.8751527070999146, "epoch": 2, "memory": 15585, "step": 2956} +{"lr": 0.00037913685411401933, "data_time": 0.29927406311035154, "loss": 0.03589790295809507, "time": 0.817652678489685, "epoch": 2, "memory": 15585, "step": 3056} +{"lr": 0.00037777563463289686, "data_time": 0.24219095706939697, "loss": 0.03650339599698782, "time": 0.7664620161056519, "epoch": 2, "memory": 15585, "step": 3156} +{"lr": 0.0003763740759247917, "data_time": 0.22861981391906738, "loss": 0.044386313296854496, "time": 0.7522628784179688, "epoch": 2, "memory": 15585, "step": 3256} +{"lr": 0.0003749325052221578, "data_time": 0.26761460304260254, "loss": 0.045708251930773255, "time": 0.7945474863052369, "epoch": 2, "memory": 15585, "step": 3356} +{"lr": 0.00037345125909934995, "data_time": 0.2472461938858032, "loss": 0.04041445031762123, "time": 0.7682995796203613, "epoch": 2, "memory": 15585, "step": 3456} +{"lr": 0.00037193068339404196, "data_time": 0.26855196952819826, "loss": 0.05374577697366476, "time": 0.7874505519866943, "epoch": 2, "memory": 15585, "step": 3556} +{"lr": 0.0003703711331264808, "data_time": 0.2570746183395386, "loss": 0.02681981734931469, "time": 0.775376558303833, "epoch": 2, "memory": 15585, "step": 3656} +{"lr": 0.0003687729724165983, "data_time": 0.2340846538543701, "loss": 0.04499406293034554, "time": 0.7571647644042969, "epoch": 2, "memory": 15585, "step": 3756} +{"lr": 0.000367136574398996, "data_time": 0.32793877124786375, "loss": 0.042623497731983664, "time": 0.8512455224990845, "epoch": 2, "memory": 15585, "step": 3856} +{"lr": 0.00036546232113582794, "data_time": 0.2526634454727173, "loss": 0.04397753719240427, "time": 0.7758017778396606, "epoch": 2, "memory": 15585, "step": 3956} +{"lr": 0.0003637506035275962, "data_time": 0.2568434476852417, "loss": 0.044520477391779424, "time": 0.7711585283279419, "epoch": 2, "memory": 15585, "step": 4056} +{"accuracy/top1": 91.12258911132812, "data_time": 0.17079986118879475, "time": 0.56885121298618, "step": 2} +{"lr": 0.00036100646510740043, "data_time": 0.08590762615203858, "loss": 0.041065428592264655, "time": 0.9222569227218628, "epoch": 3, "memory": 15585, "step": 4212} +{"lr": 0.0003592006803637847, "data_time": 0.001749420166015625, "loss": 0.04609261769801378, "time": 0.8253833532333374, "epoch": 3, "memory": 15585, "step": 4312} +{"lr": 0.00035735889322747414, "data_time": 0.0016744613647460937, "loss": 0.03991765761747956, "time": 0.7563973188400268, "epoch": 3, "memory": 15585, "step": 4412} +{"lr": 0.0003554815337143657, "data_time": 0.09968936443328857, "loss": 0.028875134605914354, "time": 0.8285691976547241, "epoch": 3, "memory": 15585, "step": 4512} +{"lr": 0.00035356904014570526, "data_time": 0.015761756896972658, "loss": 0.03384460154920817, "time": 0.7644056558609009, "epoch": 3, "memory": 15585, "step": 4612} +{"lr": 0.00035162185904575157, "data_time": 0.3038105249404907, "loss": 0.035961291566491124, "time": 0.8292667865753174, "epoch": 3, "memory": 15585, "step": 4712} +{"lr": 0.0003496404450375223, "data_time": 0.2916796922683716, "loss": 0.03861950431019068, "time": 0.8274330377578736, "epoch": 3, "memory": 15585, "step": 4812} +{"lr": 0.0003476252607366498, "data_time": 0.3371662855148315, "loss": 0.035083943791687486, "time": 0.8627967834472656, "epoch": 3, "memory": 15585, "step": 4912} +{"lr": 0.0003455767766433698, "data_time": 0.2681027889251709, "loss": 0.04419847317039967, "time": 0.7997649192810059, "epoch": 3, "memory": 15585, "step": 5012} +{"lr": 0.0003434954710326717, "data_time": 0.2596555233001709, "loss": 0.03167481794953346, "time": 0.7845546007156372, "epoch": 3, "memory": 15585, "step": 5112} +{"lr": 0.00034138182984263203, "data_time": 0.3108095169067383, "loss": 0.03817869685590267, "time": 0.8332990407943726, "epoch": 3, "memory": 15585, "step": 5212} +{"lr": 0.0003392363465609565, "data_time": 0.23287680149078369, "loss": 0.03682389818131924, "time": 0.7543991804122925, "epoch": 3, "memory": 15585, "step": 5312} +{"lr": 0.0003370595221097656, "data_time": 0.24555511474609376, "loss": 0.03068116307258606, "time": 0.7744240999221802, "epoch": 3, "memory": 15585, "step": 5412} +{"lr": 0.00033485186472863657, "data_time": 0.2514724493026733, "loss": 0.03019562130793929, "time": 0.7751284599304199, "epoch": 3, "memory": 15585, "step": 5512} +{"lr": 0.00033261388985594397, "data_time": 0.22008254528045654, "loss": 0.04136306568980217, "time": 0.7380250453948974, "epoch": 3, "memory": 15585, "step": 5612} +{"lr": 0.0003303461200085145, "data_time": 0.21104114055633544, "loss": 0.0332688445225358, "time": 0.7552468776702881, "epoch": 3, "memory": 15585, "step": 5712} +{"lr": 0.0003280490846596322, "data_time": 0.18828837871551513, "loss": 0.036070290487259626, "time": 0.7050377130508423, "epoch": 3, "memory": 15585, "step": 5812} +{"lr": 0.00032572332011541707, "data_time": 0.2858557224273682, "loss": 0.04054839164018631, "time": 0.8197050809860229, "epoch": 3, "memory": 15585, "step": 5912} +{"lr": 0.00032336936938961124, "data_time": 0.2519676923751831, "loss": 0.023237418197095395, "time": 0.7868886470794678, "epoch": 3, "memory": 15585, "step": 6012} +{"lr": 0.00032098778207679645, "data_time": 0.2727648735046387, "loss": 0.03557647168636322, "time": 0.8068763256072998, "epoch": 3, "memory": 15585, "step": 6112} +{"accuracy/top1": 94.3531494140625, "data_time": 0.18577750393601714, "time": 0.5866230706699559, "step": 3} +{"lr": 0.0003172186408067781, "data_time": 0.31164615154266356, "loss": 0.035311220400035384, "time": 0.8347751379013062, "epoch": 4, "memory": 15585, "step": 6268} +{"lr": 0.00031476885256500535, "data_time": 0.30954408645629883, "loss": 0.03334829155355692, "time": 0.8263779640197754, "epoch": 4, "memory": 15585, "step": 6368} +{"lr": 0.00031229343576357664, "data_time": 0.27940547466278076, "loss": 0.03248629802837968, "time": 0.7942147016525268, "epoch": 4, "memory": 15585, "step": 6468} +{"lr": 0.00030979296835667217, "data_time": 0.3086205244064331, "loss": 0.023935234546661376, "time": 0.8264507055282593, "epoch": 4, "memory": 15585, "step": 6568} +{"lr": 0.000307268034147225, "data_time": 0.2922163248062134, "loss": 0.03954189233481884, "time": 0.8144617557525635, "epoch": 4, "memory": 15585, "step": 6668} +{"lr": 0.00030471922265061713, "data_time": 0.24466795921325685, "loss": 0.0366573590785265, "time": 0.7668239593505859, "epoch": 4, "memory": 15585, "step": 6768} +{"lr": 0.00030214712895704036, "data_time": 0.34331541061401366, "loss": 0.02575256349518895, "time": 0.8645653009414673, "epoch": 4, "memory": 15585, "step": 6868} +{"lr": 0.00029955235359255527, "data_time": 0.29909601211547854, "loss": 0.018522901460528374, "time": 0.8239031553268432, "epoch": 4, "memory": 15585, "step": 6968} +{"lr": 0.00029693550237888314, "data_time": 0.30478610992431643, "loss": 0.029821417992934585, "time": 0.8184063911437989, "epoch": 4, "memory": 15585, "step": 7068} +{"lr": 0.0002942971862919601, "data_time": 0.2361917734146118, "loss": 0.02400244167074561, "time": 0.7563504934310913, "epoch": 4, "memory": 15585, "step": 7168} +{"lr": 0.0002916380213192881, "data_time": 0.23835356235504152, "loss": 0.0304020157083869, "time": 0.7577677488327026, "epoch": 4, "memory": 15585, "step": 7268} +{"lr": 0.00028895862831611433, "data_time": 0.3100778579711914, "loss": 0.02480067331343889, "time": 0.8320559501647949, "epoch": 4, "memory": 15585, "step": 7368} +{"lr": 0.0002862596328604766, "data_time": 0.2189112901687622, "loss": 0.033387305587530135, "time": 0.7994933128356934, "epoch": 4, "memory": 15585, "step": 7468} +{"lr": 0.0002835416651071449, "data_time": 0.20606186389923095, "loss": 0.025601667817682028, "time": 0.7394556760787964, "epoch": 4, "memory": 15585, "step": 7568} +{"lr": 0.00028080535964049476, "data_time": 0.22684597969055176, "loss": 0.023305664584040642, "time": 0.7484435558319091, "epoch": 4, "memory": 15585, "step": 7668} +{"lr": 0.0002780513553263448, "data_time": 0.02597343921661377, "loss": 0.03611135305836797, "time": 0.9599784851074219, "epoch": 4, "memory": 15585, "step": 7768} +{"lr": 0.0002752802951627973, "data_time": 0.0018501520156860352, "loss": 0.03020003205165267, "time": 0.815102481842041, "epoch": 4, "memory": 15585, "step": 7868} +{"lr": 0.0002724928261301117, "data_time": 0.0017300844192504883, "loss": 0.023491135472431778, "time": 0.9153478860855102, "epoch": 4, "memory": 15585, "step": 7968} +{"lr": 0.00026968959903964925, "data_time": 0.0017294168472290039, "loss": 0.02911972114816308, "time": 0.8375027894973754, "epoch": 4, "memory": 15585, "step": 8068} +{"lr": 0.0002668712683819236, "data_time": 0.0016197681427001954, "loss": 0.023819428961724044, "time": 0.8518417596817016, "epoch": 4, "memory": 15585, "step": 8168} +{"accuracy/top1": 96.1686782836914, "data_time": 0.17289388766054248, "time": 0.570914119970603, "step": 4} +{"lr": 0.00026244607407074216, "data_time": 0.22203283309936522, "loss": 0.024537878390401602, "time": 0.7770836353302002, "epoch": 5, "memory": 15585, "step": 8324} +{"lr": 0.00025959208467259544, "data_time": 0.20585229396820068, "loss": 0.02669526357203722, "time": 0.7901898145675659, "epoch": 5, "memory": 15585, "step": 8424} +{"lr": 0.0002567253492498155, "data_time": 0.27142164707183836, "loss": 0.02051792209967971, "time": 0.7884191989898681, "epoch": 5, "memory": 15585, "step": 8524} +{"lr": 0.00025384653712068393, "data_time": 0.2411658525466919, "loss": 0.025866320729255675, "time": 0.757617998123169, "epoch": 5, "memory": 15585, "step": 8624} +{"lr": 0.00025095632042312373, "data_time": 0.2308708667755127, "loss": 0.02373034837655723, "time": 0.8023012399673461, "epoch": 5, "memory": 15585, "step": 8724} +{"lr": 0.00024805537395776515, "data_time": 0.0018132925033569336, "loss": 0.027906367275863885, "time": 0.7825748920440674, "epoch": 5, "memory": 15585, "step": 8824} +{"lr": 0.00024514437503039924, "data_time": 0.19741315841674806, "loss": 0.02937898728996515, "time": 0.7856703281402588, "epoch": 5, "memory": 15585, "step": 8924} +{"lr": 0.0002422240032938408, "data_time": 0.14554126262664796, "loss": 0.02580722509883344, "time": 0.8777301788330079, "epoch": 5, "memory": 15585, "step": 9024} +{"lr": 0.00023929494058924557, "data_time": 0.270503830909729, "loss": 0.02230552537366748, "time": 0.8500177383422851, "epoch": 5, "memory": 15585, "step": 9124} +{"lr": 0.00023635787078691464, "data_time": 0.26755404472351074, "loss": 0.02209522812627256, "time": 0.7864372968673706, "epoch": 5, "memory": 15585, "step": 9224} +{"lr": 0.0002334134796266267, "data_time": 0.001841259002685547, "loss": 0.02020230023190379, "time": 0.7152056694030762, "epoch": 5, "memory": 15585, "step": 9324} +{"lr": 0.00023046245455753318, "data_time": 0.04827768802642822, "loss": 0.024352390971034765, "time": 0.824299955368042, "epoch": 5, "memory": 15585, "step": 9424} +{"lr": 0.0002275054845776535, "data_time": 0.2274636745452881, "loss": 0.023816074430942535, "time": 0.7456465244293213, "epoch": 5, "memory": 15585, "step": 9524} +{"lr": 0.00022454326007301025, "data_time": 0.27688424587249755, "loss": 0.029075198993086814, "time": 0.8006130695343018, "epoch": 5, "memory": 15585, "step": 9624} +{"lr": 0.00022157647265643937, "data_time": 0.25865755081176756, "loss": 0.020016076928004622, "time": 0.7812500238418579, "epoch": 5, "memory": 15585, "step": 9724} +{"lr": 0.00021860581500611435, "data_time": 0.2557729005813599, "loss": 0.02665470065549016, "time": 0.7762583732604981, "epoch": 5, "memory": 15585, "step": 9824} +{"lr": 0.00021563198070382117, "data_time": 0.28117234706878663, "loss": 0.022489734273403884, "time": 0.8034007310867309, "epoch": 5, "memory": 15585, "step": 9924} +{"lr": 0.00021265566407302255, "data_time": 0.25601351261138916, "loss": 0.02712933961302042, "time": 0.7830763101577759, "epoch": 5, "memory": 15585, "step": 10024} +{"lr": 0.00020967756001674922, "data_time": 0.29178638458251954, "loss": 0.02838323647156358, "time": 0.8185221195220947, "epoch": 5, "memory": 15585, "step": 10124} +{"lr": 0.0002066983638553557, "data_time": 0.2809009075164795, "loss": 0.023266431502997877, "time": 0.8061702013015747, "epoch": 5, "memory": 15585, "step": 10224} +{"accuracy/top1": 97.35697174072266, "data_time": 0.16911720838703093, "time": 0.5697144015890653, "step": 5} +{"lr": 0.0002020502853543317, "data_time": 0.2810657978057861, "loss": 0.027179038245230915, "time": 0.8111610174179077, "epoch": 6, "memory": 15585, "step": 10380} +{"lr": 0.000199071463167015, "data_time": 0.26308317184448243, "loss": 0.020010809879750013, "time": 0.7848746538162231, "epoch": 6, "memory": 15585, "step": 10480} +{"lr": 0.00019609402515978532, "data_time": 0.23099541664123535, "loss": 0.026591809932142496, "time": 0.7869767427444458, "epoch": 6, "memory": 15585, "step": 10580} +{"lr": 0.00019311866649748968, "data_time": 0.21308507919311523, "loss": 0.026091505959630012, "time": 0.7677872180938721, "epoch": 6, "memory": 15585, "step": 10680} +{"lr": 0.00019014608185949442, "data_time": 0.2908239126205444, "loss": 0.02376266596838832, "time": 0.8316193342208862, "epoch": 6, "memory": 15585, "step": 10780} +{"lr": 0.00018717696527749397, "data_time": 0.26842200756073, "loss": 0.02511770874261856, "time": 0.7919795751571655, "epoch": 6, "memory": 15585, "step": 10880} +{"lr": 0.00018421200997346973, "data_time": 0.24688453674316407, "loss": 0.024106958508491518, "time": 0.7676408767700196, "epoch": 6, "memory": 15585, "step": 10980} +{"lr": 0.00018125190819783787, "data_time": 0.27145895957946775, "loss": 0.02252453090623021, "time": 0.7901381492614746, "epoch": 6, "memory": 15585, "step": 11080} +{"lr": 0.00017829735106782464, "data_time": 0.24628117084503173, "loss": 0.011761882016435266, "time": 0.7876333713531494, "epoch": 6, "memory": 15585, "step": 11180} +{"lr": 0.00017534902840610667, "data_time": 0.30084273815155027, "loss": 0.027595242159441112, "time": 0.8193407535552979, "epoch": 6, "memory": 15585, "step": 11280} +{"lr": 0.00017240762857975229, "data_time": 0.240846848487854, "loss": 0.022845309879630803, "time": 0.7586073875427246, "epoch": 6, "memory": 15585, "step": 11380} +{"lr": 0.00016947383833950319, "data_time": 0.22904131412506104, "loss": 0.01753133866004646, "time": 0.749635124206543, "epoch": 6, "memory": 15585, "step": 11480} +{"lr": 0.0001665483426594338, "data_time": 0.2171614646911621, "loss": 0.01834913678467274, "time": 0.7466540575027466, "epoch": 6, "memory": 15585, "step": 11580} +{"lr": 0.00016363182457702495, "data_time": 0.23504273891448973, "loss": 0.027551061613485216, "time": 0.7546716928482056, "epoch": 6, "memory": 15585, "step": 11680} +{"lr": 0.0001607249650336909, "data_time": 0.28019123077392577, "loss": 0.019759564893320203, "time": 0.813611102104187, "epoch": 6, "memory": 15585, "step": 11780} +{"lr": 0.00015782844271579344, "data_time": 0.21503834724426268, "loss": 0.023741032090038063, "time": 0.7778896808624267, "epoch": 6, "memory": 15585, "step": 11880} +{"lr": 0.0001549429338961849, "data_time": 0.20234980583190917, "loss": 0.01582119045779109, "time": 0.7946874380111695, "epoch": 6, "memory": 15585, "step": 11980} +{"lr": 0.00015206911227631342, "data_time": 0.17572021484375, "loss": 0.019931396888568997, "time": 0.7698912382125854, "epoch": 6, "memory": 15585, "step": 12080} +{"lr": 0.00014920764882892865, "data_time": 0.2740559816360474, "loss": 0.024159587547183038, "time": 0.7960219383239746, "epoch": 6, "memory": 15585, "step": 12180} +{"lr": 0.00014635921164142459, "data_time": 0.17802643775939941, "loss": 0.014993937406688928, "time": 0.7756404638290405, "epoch": 6, "memory": 15585, "step": 12280} +{"accuracy/top1": 98.2605972290039, "data_time": 0.17307630523306425, "time": 0.5701735488703994, "step": 6} +{"lr": 0.00014194323525930575, "data_time": 0.22661159038543702, "loss": 0.024994338117539882, "time": 0.8414340734481811, "epoch": 7, "memory": 15585, "step": 12436} +{"lr": 0.0001391311681531885, "data_time": 0.2701252460479736, "loss": 0.024148925440385938, "time": 0.7996945858001709, "epoch": 7, "memory": 15585, "step": 12536} +{"lr": 0.00013633447993868745, "data_time": 0.2745425462722778, "loss": 0.015681801736354826, "time": 0.8095317840576172, "epoch": 7, "memory": 15585, "step": 12636} +{"lr": 0.00013355382357963566, "data_time": 0.23766775131225587, "loss": 0.015086093964055181, "time": 0.7693504810333252, "epoch": 7, "memory": 15585, "step": 12736} +{"lr": 0.0001307898482967891, "data_time": 0.23037829399108886, "loss": 0.01834228038787842, "time": 0.7463788986206055, "epoch": 7, "memory": 15585, "step": 12836} +{"lr": 0.00012804319941624686, "data_time": 0.2668689012527466, "loss": 0.016991642070934176, "time": 0.7953432083129883, "epoch": 7, "memory": 15585, "step": 12936} +{"lr": 0.00012531451821878227, "data_time": 0.25720701217651365, "loss": 0.017566187493503093, "time": 0.7799966335296631, "epoch": 7, "memory": 15585, "step": 13036} +{"lr": 0.00012260444179011874, "data_time": 0.29000751972198485, "loss": 0.02144675310701132, "time": 0.813347053527832, "epoch": 7, "memory": 15585, "step": 13136} +{"lr": 0.00011991360287218363, "data_time": 0.07197637557983398, "loss": 0.020820771483704446, "time": 0.8050727844238281, "epoch": 7, "memory": 15585, "step": 13236} +{"lr": 0.00011724262971537941, "data_time": 0.12122166156768799, "loss": 0.01619745516218245, "time": 0.873307180404663, "epoch": 7, "memory": 15585, "step": 13336} +{"lr": 0.00011459214593189938, "data_time": 0.087725830078125, "loss": 0.02611830420792103, "time": 0.8096433639526367, "epoch": 7, "memory": 15585, "step": 13436} +{"lr": 0.00011196277035012846, "data_time": 0.0019665956497192383, "loss": 0.016010493924841284, "time": 0.7523234844207763, "epoch": 7, "memory": 15585, "step": 13536} +{"lr": 0.0001093551168701611, "data_time": 0.07163176536560059, "loss": 0.015815067756921052, "time": 0.814582347869873, "epoch": 7, "memory": 15585, "step": 13636} +{"lr": 0.00010676979432046926, "data_time": 0.0018159151077270508, "loss": 0.021654323022812606, "time": 0.8054304838180542, "epoch": 7, "memory": 15585, "step": 13736} +{"lr": 0.00010420740631575464, "data_time": 0.12277102470397949, "loss": 0.012254316126927733, "time": 0.8384614706039428, "epoch": 7, "memory": 15585, "step": 13836} +{"lr": 0.00010166855111601829, "data_time": 0.12544896602630615, "loss": 0.0282936405390501, "time": 0.8113990068435669, "epoch": 7, "memory": 15585, "step": 13936} +{"lr": 9.91538214868799e-05, "data_time": 0.2400291919708252, "loss": 0.009547821804881096, "time": 0.7894332647323609, "epoch": 7, "memory": 15585, "step": 14036} +{"lr": 9.666380456118125e-05, "data_time": 0.21268165111541748, "loss": 0.020644811866804957, "time": 0.7628713846206665, "epoch": 7, "memory": 15585, "step": 14136} +{"lr": 9.419908170190298e-05, "data_time": 0.25474069118499754, "loss": 0.015355687029659748, "time": 0.7784512996673584, "epoch": 7, "memory": 15585, "step": 14236} +{"lr": 9.176022836643002e-05, "data_time": 0.2521061420440674, "loss": 0.018681674776598812, "time": 0.8136514902114869, "epoch": 7, "memory": 15585, "step": 14336} +{"accuracy/top1": 98.78192138671875, "data_time": 0.17212986164405697, "time": 0.5731900207331924, "step": 7} +{"lr": 8.80086206394144e-05, "data_time": 0.22919075489044188, "loss": 0.017404032032936813, "time": 0.7469552040100098, "epoch": 8, "memory": 15585, "step": 14492} +{"lr": 8.563857333566659e-05, "data_time": 0.2747959613800049, "loss": 0.018328665662556885, "time": 0.8032639503479004, "epoch": 8, "memory": 15585, "step": 14592} +{"lr": 8.329639424200186e-05, "data_time": 0.2506705284118652, "loss": 0.016642919182777403, "time": 0.7824928522109985, "epoch": 8, "memory": 15585, "step": 14692} +{"lr": 8.098263020459022e-05, "data_time": 0.29321334362030027, "loss": 0.014005721732974052, "time": 0.8123298406600952, "epoch": 8, "memory": 15585, "step": 14792} +{"lr": 7.869782143532516e-05, "data_time": 0.268187952041626, "loss": 0.014867176348343492, "time": 0.7943655729293824, "epoch": 8, "memory": 15585, "step": 14892} +{"lr": 7.644250138569548e-05, "data_time": 0.26698756217956543, "loss": 0.016485038213431835, "time": 0.7864221811294556, "epoch": 8, "memory": 15585, "step": 14992} +{"lr": 7.4217196622237e-05, "data_time": 0.25427148342132566, "loss": 0.012455874122679234, "time": 0.7935800790786743, "epoch": 8, "memory": 15585, "step": 15092} +{"lr": 7.202242670359085e-05, "data_time": 0.348530650138855, "loss": 0.01578704100102186, "time": 0.8687826633453369, "epoch": 8, "memory": 15585, "step": 15192} +{"lr": 6.98587040591985e-05, "data_time": 0.25888495445251464, "loss": 0.017278297105804086, "time": 0.7783713340759277, "epoch": 8, "memory": 15585, "step": 15292} +{"lr": 6.772653386966062e-05, "data_time": 0.29857597351074217, "loss": 0.016061631310731173, "time": 0.8167670011520386, "epoch": 8, "memory": 15585, "step": 15392} +{"lr": 6.562641394878866e-05, "data_time": 0.19436044692993165, "loss": 0.017046904563903807, "time": 0.800396728515625, "epoch": 8, "memory": 15585, "step": 15492} +{"lr": 6.35588346273771e-05, "data_time": 0.25895874500274657, "loss": 0.019470716919749977, "time": 0.8157904624938965, "epoch": 8, "memory": 15585, "step": 15592} +{"lr": 6.152427863872164e-05, "data_time": 0.22635767459869385, "loss": 0.010894187493249774, "time": 0.7625266790390015, "epoch": 8, "memory": 15585, "step": 15692} +{"lr": 5.952322100591216e-05, "data_time": 0.27396225929260254, "loss": 0.01048377575352788, "time": 0.7904689311981201, "epoch": 8, "memory": 15585, "step": 15792} +{"lr": 5.755612893092559e-05, "data_time": 0.26614985466003416, "loss": 0.014424818428233266, "time": 0.7920905590057373, "epoch": 8, "memory": 15585, "step": 15892} +{"lr": 5.5623461685544754e-05, "data_time": 0.29710395336151124, "loss": 0.01462442851625383, "time": 0.8157669305801392, "epoch": 8, "memory": 15585, "step": 15992} +{"lr": 5.37256705041282e-05, "data_time": 0.3045217752456665, "loss": 0.017198205180466174, "time": 0.8254985094070435, "epoch": 8, "memory": 15585, "step": 16092} +{"lr": 5.186319847825775e-05, "data_time": 0.25671370029449464, "loss": 0.019532240414991974, "time": 0.7780438423156738, "epoch": 8, "memory": 15585, "step": 16192} +{"lr": 5.0036480453286144e-05, "data_time": 0.24693162441253663, "loss": 0.010180334467440844, "time": 0.772924542427063, "epoch": 8, "memory": 15585, "step": 16292} +{"lr": 4.8245942926810745e-05, "data_time": 0.250138521194458, "loss": 0.017302094027400017, "time": 0.7713725805282593, "epoch": 8, "memory": 15585, "step": 16392} +{"accuracy/top1": 99.13443756103516, "data_time": 0.17577528171851986, "time": 0.5720648648308926, "step": 8} +{"lr": 4.5525937358206823e-05, "data_time": 0.09623749256134033, "loss": 0.010653944639489054, "time": 0.7902414798736572, "epoch": 9, "memory": 15585, "step": 16548} +{"lr": 4.382990660001016e-05, "data_time": 0.26913387775421144, "loss": 0.01291049225255847, "time": 0.7944161415100097, "epoch": 9, "memory": 15585, "step": 16648} +{"lr": 4.217150543579113e-05, "data_time": 0.23822276592254638, "loss": 0.01621076031588018, "time": 0.7630682945251465, "epoch": 9, "memory": 15585, "step": 16748} +{"lr": 4.0551121064943564e-05, "data_time": 0.2548978328704834, "loss": 0.012189733469858766, "time": 0.7813825845718384, "epoch": 9, "memory": 15585, "step": 16848} +{"lr": 3.896913181079464e-05, "data_time": 0.2330319404602051, "loss": 0.014061120757833123, "time": 0.753749418258667, "epoch": 9, "memory": 15585, "step": 16948} +{"lr": 3.742590703227469e-05, "data_time": 0.24396300315856934, "loss": 0.011295050429180265, "time": 0.769128966331482, "epoch": 9, "memory": 15585, "step": 17048} +{"lr": 3.592180703768032e-05, "data_time": 0.26656978130340575, "loss": 0.009772074222564698, "time": 0.7856827735900879, "epoch": 9, "memory": 15585, "step": 17148} +{"lr": 3.445718300055071e-05, "data_time": 0.2272563695907593, "loss": 0.016224468871951104, "time": 0.7661733865737915, "epoch": 9, "memory": 15585, "step": 17248} +{"lr": 3.3032376877676173e-05, "data_time": 0.18842051029205323, "loss": 0.014255373179912567, "time": 0.8337079524993897, "epoch": 9, "memory": 15585, "step": 17348} +{"lr": 3.1647721329259246e-05, "data_time": 0.0018663883209228515, "loss": 0.010739016812294722, "time": 1.1180885314941407, "epoch": 9, "memory": 15585, "step": 17448} +{"lr": 3.0303539641245756e-05, "data_time": 0.0022369861602783204, "loss": 0.02085434915497899, "time": 0.903159236907959, "epoch": 9, "memory": 15585, "step": 17548} +{"lr": 2.9000145649844975e-05, "data_time": 0.0024661779403686523, "loss": 0.008549377089366317, "time": 0.8973897218704223, "epoch": 9, "memory": 15585, "step": 17648} +{"lr": 2.7737843668255695e-05, "data_time": 0.004348850250244141, "loss": 0.00835349946282804, "time": 1.0430068731307984, "epoch": 9, "memory": 15585, "step": 17748} +{"lr": 2.6516928415616184e-05, "data_time": 0.0016827583312988281, "loss": 0.02047748272307217, "time": 0.8370586633682251, "epoch": 9, "memory": 15585, "step": 17848} +{"lr": 2.5337684948193946e-05, "data_time": 0.0019025564193725585, "loss": 0.009703512722626328, "time": 0.7816456317901611, "epoch": 9, "memory": 15585, "step": 17948} +{"lr": 2.4200388592830947e-05, "data_time": 0.001636672019958496, "loss": 0.011804910842329264, "time": 0.9837682008743286, "epoch": 9, "memory": 15585, "step": 18048} +{"lr": 2.3105304882661933e-05, "data_time": 0.0017518281936645507, "loss": 0.009747504582628607, "time": 1.4381043195724488, "epoch": 9, "memory": 15585, "step": 18148} +{"lr": 2.20526894951175e-05, "data_time": 0.0016247987747192382, "loss": 0.012831246200948953, "time": 0.7416228771209716, "epoch": 9, "memory": 15585, "step": 18248} +{"lr": 2.1042788192229725e-05, "data_time": 0.0015455007553100586, "loss": 0.01550100278109312, "time": 0.8639036893844605, "epoch": 9, "memory": 15585, "step": 18348} +{"lr": 2.007583676325189e-05, "data_time": 0.0018747568130493164, "loss": 0.012653633020818233, "time": 0.7752285480499268, "epoch": 9, "memory": 15585, "step": 18448} +{"accuracy/top1": 99.28504180908203, "data_time": 0.17789622603869829, "time": 0.5768623117540703, "step": 9} +{"lr": 1.8653686449507784e-05, "data_time": 0.2892716884613037, "loss": 0.007131221424788237, "time": 0.8109148025512696, "epoch": 10, "memory": 15585, "step": 18604} +{"lr": 1.7797691544415563e-05, "data_time": 0.28204195499420165, "loss": 0.010444499505683779, "time": 0.8037574291229248, "epoch": 10, "memory": 15585, "step": 18704} +{"lr": 1.6985404170356326e-05, "data_time": 0.26016457080841066, "loss": 0.00676915175281465, "time": 0.7789924383163452, "epoch": 10, "memory": 15585, "step": 18804} +{"lr": 1.6217013978172993e-05, "data_time": 0.22398149967193604, "loss": 0.010695388168096542, "time": 0.8379882097244262, "epoch": 10, "memory": 15585, "step": 18904} +{"lr": 1.549270036970338e-05, "data_time": 0.26202614307403566, "loss": 0.009390882449224591, "time": 0.7857458353042602, "epoch": 10, "memory": 15585, "step": 19004} +{"lr": 1.4812632455893675e-05, "data_time": 0.315905237197876, "loss": 0.015783053589984775, "time": 0.8413533210754395, "epoch": 10, "memory": 15585, "step": 19104} +{"lr": 1.4176969017315005e-05, "data_time": 0.23308274745941163, "loss": 0.009905860992148519, "time": 0.7513970851898193, "epoch": 10, "memory": 15585, "step": 19204} +{"lr": 1.358585846709161e-05, "data_time": 0.29018981456756593, "loss": 0.00909892893396318, "time": 0.8070831060409546, "epoch": 10, "memory": 15585, "step": 19304} +{"lr": 1.3039438816249855e-05, "data_time": 0.31691243648529055, "loss": 0.011139605101197958, "time": 0.8341225147247314, "epoch": 10, "memory": 15585, "step": 19404} +{"lr": 1.2537837641495561e-05, "data_time": 0.2673105955123901, "loss": 0.01893265973776579, "time": 0.7921293258666993, "epoch": 10, "memory": 15585, "step": 19504} +{"lr": 1.2081172055427927e-05, "data_time": 0.27910146713256834, "loss": 0.009846163867041468, "time": 0.8006896734237671, "epoch": 10, "memory": 15585, "step": 19604} +{"lr": 1.1669548679196218e-05, "data_time": 0.32702112197875977, "loss": 0.010256294440478087, "time": 0.8460445880889893, "epoch": 10, "memory": 15585, "step": 19704} +{"lr": 1.130306361760621e-05, "data_time": 0.3108600378036499, "loss": 0.009392891265451907, "time": 0.8323075294494628, "epoch": 10, "memory": 15585, "step": 19804} +{"lr": 1.0981802436681767e-05, "data_time": 0.29009621143341063, "loss": 0.014776562433689832, "time": 0.8083298921585083, "epoch": 10, "memory": 15585, "step": 19904} +{"lr": 1.0705840143687269e-05, "data_time": 0.2735370874404907, "loss": 0.006929418072104454, "time": 0.8063861131668091, "epoch": 10, "memory": 15585, "step": 20004} +{"lr": 1.0475241169614925e-05, "data_time": 0.001825571060180664, "loss": 0.01014872114174068, "time": 0.7849845886230469, "epoch": 10, "memory": 15585, "step": 20104} +{"lr": 1.0290059354141742e-05, "data_time": 0.16223702430725098, "loss": 0.008769241999834776, "time": 0.8510101079940796, "epoch": 10, "memory": 15585, "step": 20204} +{"lr": 1.015033793305906e-05, "data_time": 0.10497753620147705, "loss": 0.010651576286181807, "time": 0.7695328950881958, "epoch": 10, "memory": 15585, "step": 20304} +{"lr": 1.0056109528177973e-05, "data_time": 0.3397828578948975, "loss": 0.010774973314255476, "time": 0.8865358829498291, "epoch": 10, "memory": 15585, "step": 20404} +{"lr": 1.0007396139712939e-05, "data_time": 0.3367367506027222, "loss": 0.009603068884462117, "time": 0.8584039449691773, "epoch": 10, "memory": 15585, "step": 20504} +{"accuracy/top1": 99.41413116455078, "data_time": 0.1701847607972192, "time": 0.5691666055898197, "step": 10} diff --git a/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1/20230605_024744/vis_data/vis_image/0000000.png_1.png b/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1/20230605_024744/vis_data/vis_image/0000000.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..0a77604ea25207cb2ff76654d347f12da1eeb2e3 Binary files /dev/null and b/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1/20230605_024744/vis_data/vis_image/0000000.png_1.png differ diff --git a/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1/20230605_024744/vis_data/vis_image/0000000.png_10.png b/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1/20230605_024744/vis_data/vis_image/0000000.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..80530a28e6fb25692841d68a3e2860040b0e25d3 Binary files /dev/null and b/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1/20230605_024744/vis_data/vis_image/0000000.png_10.png differ diff --git a/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1/20230605_024744/vis_data/vis_image/0000000.png_2.png b/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1/20230605_024744/vis_data/vis_image/0000000.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..66aa002bd93f4cedbf7b728a3a149a9507622474 Binary files /dev/null and b/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1/20230605_024744/vis_data/vis_image/0000000.png_2.png differ diff --git a/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1/20230605_024744/vis_data/vis_image/0000000.png_3.png b/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1/20230605_024744/vis_data/vis_image/0000000.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..9394e88f68b6b74559ad45a92dc40036ce24ab2f Binary files /dev/null and b/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1/20230605_024744/vis_data/vis_image/0000000.png_3.png differ diff --git a/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1/20230605_024744/vis_data/vis_image/0000000.png_4.png b/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1/20230605_024744/vis_data/vis_image/0000000.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..3fa7a00a27fe7bd48e1648ead08f3e46c500e648 Binary files /dev/null and b/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1/20230605_024744/vis_data/vis_image/0000000.png_4.png differ diff --git a/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1/20230605_024744/vis_data/vis_image/0000000.png_5.png b/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1/20230605_024744/vis_data/vis_image/0000000.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..ac70c50625552b138013677e15a72fc6d513ffac Binary files /dev/null and b/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1/20230605_024744/vis_data/vis_image/0000000.png_5.png differ diff --git a/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1/20230605_024744/vis_data/vis_image/0000000.png_6.png b/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1/20230605_024744/vis_data/vis_image/0000000.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..af1747cafb5979af651a2d1b5a692d2be058d785 Binary files /dev/null and b/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1/20230605_024744/vis_data/vis_image/0000000.png_6.png differ diff --git a/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1/20230605_024744/vis_data/vis_image/0000000.png_7.png b/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1/20230605_024744/vis_data/vis_image/0000000.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..f4cc7efec85b5f07ffe4486bdf524af24d29e583 Binary files /dev/null and b/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1/20230605_024744/vis_data/vis_image/0000000.png_7.png differ diff --git a/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1/20230605_024744/vis_data/vis_image/0000000.png_8.png b/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1/20230605_024744/vis_data/vis_image/0000000.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..ec6caac0f10e42d0658f3ab56241e04d1fce54a2 Binary files /dev/null and b/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1/20230605_024744/vis_data/vis_image/0000000.png_8.png differ diff --git a/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1/20230605_024744/vis_data/vis_image/0000000.png_9.png b/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1/20230605_024744/vis_data/vis_image/0000000.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..0c919567fa35425e3e1c0a5f90e5534e803b3e1d Binary files /dev/null and b/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1/20230605_024744/vis_data/vis_image/0000000.png_9.png differ diff --git a/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1/20230605_024744/vis_data/vis_image/000015337.jpg_1.png b/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1/20230605_024744/vis_data/vis_image/000015337.jpg_1.png new file mode 100644 index 0000000000000000000000000000000000000000..e5fe525ed6f4e563d01f59ecbbc127eee6f9a4f6 Binary files /dev/null and b/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1/20230605_024744/vis_data/vis_image/000015337.jpg_1.png differ diff --git a/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1/20230605_024744/vis_data/vis_image/000015337.jpg_10.png b/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1/20230605_024744/vis_data/vis_image/000015337.jpg_10.png new file mode 100644 index 0000000000000000000000000000000000000000..3c4563fd434c0a5e2a6e99582cfba3d00f4a1f53 Binary files /dev/null and b/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1/20230605_024744/vis_data/vis_image/000015337.jpg_10.png differ diff --git a/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1/20230605_024744/vis_data/vis_image/000015337.jpg_2.png b/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1/20230605_024744/vis_data/vis_image/000015337.jpg_2.png new file mode 100644 index 0000000000000000000000000000000000000000..9752ab696b2f70cf05ef55ca1e02ebf6363cf7d2 Binary files /dev/null and b/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1/20230605_024744/vis_data/vis_image/000015337.jpg_2.png differ diff --git a/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1/20230605_024744/vis_data/vis_image/000015337.jpg_3.png b/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1/20230605_024744/vis_data/vis_image/000015337.jpg_3.png new file mode 100644 index 0000000000000000000000000000000000000000..f49e6773ed0744e5287e87e8849625de72e7d949 Binary files /dev/null and b/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1/20230605_024744/vis_data/vis_image/000015337.jpg_3.png differ diff --git a/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1/20230605_024744/vis_data/vis_image/000015337.jpg_4.png b/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1/20230605_024744/vis_data/vis_image/000015337.jpg_4.png new file mode 100644 index 0000000000000000000000000000000000000000..51ac142df797381764c29255b1882d146a52ae54 Binary files /dev/null and b/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1/20230605_024744/vis_data/vis_image/000015337.jpg_4.png differ diff --git a/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1/20230605_024744/vis_data/vis_image/000015337.jpg_5.png b/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1/20230605_024744/vis_data/vis_image/000015337.jpg_5.png new file mode 100644 index 0000000000000000000000000000000000000000..a68c9334c4178de1ac933c98df433c5ebba941de Binary files /dev/null and b/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1/20230605_024744/vis_data/vis_image/000015337.jpg_5.png differ diff --git a/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1/20230605_024744/vis_data/vis_image/000015337.jpg_6.png b/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1/20230605_024744/vis_data/vis_image/000015337.jpg_6.png new file mode 100644 index 0000000000000000000000000000000000000000..7cef615cf42982bf2fef2eff61a4972968a18f4c Binary files /dev/null and b/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1/20230605_024744/vis_data/vis_image/000015337.jpg_6.png differ diff --git a/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1/20230605_024744/vis_data/vis_image/000015337.jpg_7.png b/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1/20230605_024744/vis_data/vis_image/000015337.jpg_7.png new file mode 100644 index 0000000000000000000000000000000000000000..b888284c94684defe2640b6fafa2b0a8a3e5874b Binary files /dev/null and b/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1/20230605_024744/vis_data/vis_image/000015337.jpg_7.png differ diff --git a/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1/20230605_024744/vis_data/vis_image/000015337.jpg_8.png b/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1/20230605_024744/vis_data/vis_image/000015337.jpg_8.png new file mode 100644 index 0000000000000000000000000000000000000000..32cd29112f9162d0f2e6e642c5d40fac0a7744da Binary files /dev/null and b/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1/20230605_024744/vis_data/vis_image/000015337.jpg_8.png differ diff --git a/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1/20230605_024744/vis_data/vis_image/000015337.jpg_9.png b/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1/20230605_024744/vis_data/vis_image/000015337.jpg_9.png new file mode 100644 index 0000000000000000000000000000000000000000..244c47a5eb8e4ff72eaac43f143d25ec3e05d6ea Binary files /dev/null and b/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1/20230605_024744/vis_data/vis_image/000015337.jpg_9.png differ diff --git a/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1/20230605_024744/vis_data/vis_image/0004159.png_1.png b/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1/20230605_024744/vis_data/vis_image/0004159.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..a0953e571e2d840cb7cec8b8e4a105f5eb0e732b Binary files /dev/null and b/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1/20230605_024744/vis_data/vis_image/0004159.png_1.png differ diff --git a/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1/20230605_024744/vis_data/vis_image/0004159.png_10.png b/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1/20230605_024744/vis_data/vis_image/0004159.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..9f1adb361044f696c555a879aaef2d3efece5c36 Binary files /dev/null and b/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1/20230605_024744/vis_data/vis_image/0004159.png_10.png differ diff --git a/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1/20230605_024744/vis_data/vis_image/0004159.png_2.png b/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1/20230605_024744/vis_data/vis_image/0004159.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..890789c234468842d665acdb060a5b5884b62b2f Binary files /dev/null and b/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1/20230605_024744/vis_data/vis_image/0004159.png_2.png differ diff --git a/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1/20230605_024744/vis_data/vis_image/0004159.png_3.png b/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1/20230605_024744/vis_data/vis_image/0004159.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..7d10a2c8b4062ca673d96de00112fed4824a195e Binary files /dev/null and b/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1/20230605_024744/vis_data/vis_image/0004159.png_3.png differ diff --git a/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1/20230605_024744/vis_data/vis_image/0004159.png_4.png b/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1/20230605_024744/vis_data/vis_image/0004159.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..c1a191957b4afe9f898821841f3e52d2622f28c5 Binary files /dev/null and b/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1/20230605_024744/vis_data/vis_image/0004159.png_4.png differ diff --git a/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1/20230605_024744/vis_data/vis_image/0004159.png_5.png b/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1/20230605_024744/vis_data/vis_image/0004159.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..5cd6fab36ce4108ce42b5331c02d3c6f7a1c6fdf Binary files /dev/null and b/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1/20230605_024744/vis_data/vis_image/0004159.png_5.png differ diff --git a/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1/20230605_024744/vis_data/vis_image/0004159.png_6.png b/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1/20230605_024744/vis_data/vis_image/0004159.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..1b56c506724e8d551d425776a18228871b35f183 Binary files /dev/null and b/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1/20230605_024744/vis_data/vis_image/0004159.png_6.png differ diff --git a/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1/20230605_024744/vis_data/vis_image/0004159.png_7.png b/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1/20230605_024744/vis_data/vis_image/0004159.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..88dfbe18b914eb43f8057770fe03213b75072075 Binary files /dev/null and b/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1/20230605_024744/vis_data/vis_image/0004159.png_7.png differ diff --git a/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1/20230605_024744/vis_data/vis_image/0004159.png_8.png b/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1/20230605_024744/vis_data/vis_image/0004159.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..a0f63d39c2a98ab4953f59b20ea2bfc5277d56c8 Binary files /dev/null and b/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1/20230605_024744/vis_data/vis_image/0004159.png_8.png differ diff --git a/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1/20230605_024744/vis_data/vis_image/0004159.png_9.png b/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1/20230605_024744/vis_data/vis_image/0004159.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..17c8f1040bd03f24dcf9dfd7480b669ba72eb49b Binary files /dev/null and b/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1/20230605_024744/vis_data/vis_image/0004159.png_9.png differ diff --git a/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1/20230605_024744/vis_data/vis_image/0008318.png_1.png b/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1/20230605_024744/vis_data/vis_image/0008318.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..864a568b1c52f51bbe06610b34edaf27c99f663f Binary files /dev/null and b/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1/20230605_024744/vis_data/vis_image/0008318.png_1.png differ diff --git a/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1/20230605_024744/vis_data/vis_image/0008318.png_10.png b/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1/20230605_024744/vis_data/vis_image/0008318.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..7246cccf7676bae646bfafba741f380e602ed5b0 Binary files /dev/null and b/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1/20230605_024744/vis_data/vis_image/0008318.png_10.png differ diff --git a/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1/20230605_024744/vis_data/vis_image/0008318.png_2.png b/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1/20230605_024744/vis_data/vis_image/0008318.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..67468755faa03683de47fcf6b68773dd8c4a5f3f Binary files /dev/null and b/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1/20230605_024744/vis_data/vis_image/0008318.png_2.png differ diff --git a/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1/20230605_024744/vis_data/vis_image/0008318.png_3.png b/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1/20230605_024744/vis_data/vis_image/0008318.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..15fc33b35bf743911414418317864943ddbf1115 Binary files /dev/null and b/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1/20230605_024744/vis_data/vis_image/0008318.png_3.png differ diff --git a/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1/20230605_024744/vis_data/vis_image/0008318.png_4.png b/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1/20230605_024744/vis_data/vis_image/0008318.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..af93b283081299db65411cbd4a94b64d0472dcfc Binary files /dev/null and b/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1/20230605_024744/vis_data/vis_image/0008318.png_4.png differ diff --git a/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1/20230605_024744/vis_data/vis_image/0008318.png_5.png b/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1/20230605_024744/vis_data/vis_image/0008318.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..55656011fde5a9a5f54174add788578ca128683d Binary files /dev/null and b/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1/20230605_024744/vis_data/vis_image/0008318.png_5.png differ diff --git a/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1/20230605_024744/vis_data/vis_image/0008318.png_6.png b/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1/20230605_024744/vis_data/vis_image/0008318.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..7f0bc98b22388bf61b8fc66e8d715b07d9ed7228 Binary files /dev/null and b/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1/20230605_024744/vis_data/vis_image/0008318.png_6.png differ diff --git a/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1/20230605_024744/vis_data/vis_image/0008318.png_7.png b/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1/20230605_024744/vis_data/vis_image/0008318.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..d0d24b07e39c4797bfed854643377247918d7e33 Binary files /dev/null and b/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1/20230605_024744/vis_data/vis_image/0008318.png_7.png differ diff --git a/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1/20230605_024744/vis_data/vis_image/0008318.png_8.png b/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1/20230605_024744/vis_data/vis_image/0008318.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..3e1b2d9932be33f3b10707df6ff8e9fe7b2cd8e3 Binary files /dev/null and b/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1/20230605_024744/vis_data/vis_image/0008318.png_8.png differ diff --git a/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1/20230605_024744/vis_data/vis_image/0008318.png_9.png b/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1/20230605_024744/vis_data/vis_image/0008318.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..30c68bda13024a523eb3ecb089dd4839f6bbeb62 Binary files /dev/null and b/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1/20230605_024744/vis_data/vis_image/0008318.png_9.png differ diff --git a/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1.py b/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1.py new file mode 100644 index 0000000000000000000000000000000000000000..9611f05a56d42d078592c667e78da62dde067cb2 --- /dev/null +++ b/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1.py @@ -0,0 +1,383 @@ +optim_wrapper = dict( + optimizer=dict( + type='AdamW', + lr=0.0004, + weight_decay=0.05, + eps=1e-08, + betas=(0.9, 0.999), + _scope_='mmpretrain'), + paramwise_cfg=dict( + norm_decay_mult=0.0, + bias_decay_mult=0.0, + flat_decay_mult=0.0, + custom_keys=dict({ + '.absolute_pos_embed': dict(decay_mult=0.0), + '.relative_position_bias_table': dict(decay_mult=0.0) + })), + type='AmpOptimWrapper', + dtype='bfloat16', + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=False, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=4096) +model = dict( + type='ImageClassifier', + backbone=dict( + frozen_stages=2, + type='ConvNeXt', + arch='small', + drop_path_rate=0.4, + init_cfg=dict( + type='Pretrained', + checkpoint= + 'https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth', + prefix='backbone')), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=768, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + init_cfg=None), + init_cfg=dict( + type='TruncNormal', layer=['Conv2d', 'Linear'], std=0.02, bias=0.0), + train_cfg=None) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/all1.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/all2.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/all3.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3fake8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/cc1m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3real8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV2-1-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/if-dpmsolver++-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV2-1-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/if-dpmsolver++-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1' diff --git a/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1/epoch_1.pth b/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1/epoch_1.pth new file mode 100644 index 0000000000000000000000000000000000000000..f78ed283456caa2a3a629ed7b0ebfb6b71041ffe --- /dev/null +++ b/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1/epoch_1.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ba140df1dbaf34d5867bee42ef8e87bc6d345e5ba6e9bfee9ab2f9b18dbc231c +size 782131226 diff --git a/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1/epoch_10.pth b/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1/epoch_10.pth new file mode 100644 index 0000000000000000000000000000000000000000..1198d5b8dd8c702327e336117d110af255b89888 --- /dev/null +++ b/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1/epoch_10.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:39fc293100cc572829399a5568f697e2094555acfdd37eb00eb72aa0e3df34a8 +size 783496410 diff --git a/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1/epoch_2.pth b/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1/epoch_2.pth new file mode 100644 index 0000000000000000000000000000000000000000..18eae28c6b277f456ac89b1b241ba8c8407d40a4 --- /dev/null +++ b/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1/epoch_2.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:856d8626d0faeff28fef7067410c9dbbf1defbf1bb6c8abbd88d338999cb4c2e +size 782283866 diff --git a/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1/epoch_3.pth b/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1/epoch_3.pth new file mode 100644 index 0000000000000000000000000000000000000000..2e12396acb42e9d5817667db951047d4c9c234f4 --- /dev/null +++ b/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1/epoch_3.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:22b34bf5dd0b446e5d7064bb158ad7d60cc4d48c5e02334388feb4a5c4813d21 +size 782435290 diff --git a/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1/epoch_4.pth b/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1/epoch_4.pth new file mode 100644 index 0000000000000000000000000000000000000000..5df52d23abf52eba9730eb5926483b7c57c2e059 --- /dev/null +++ b/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1/epoch_4.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:e9b5dafd0398ef1dd13b510aeaceb75ada41415c021e3ddefde6d04418960253 +size 782587098 diff --git a/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1/epoch_5.pth b/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1/epoch_5.pth new file mode 100644 index 0000000000000000000000000000000000000000..22aa693a952799ffc798b6a9dcc71c428342e4fe --- /dev/null +++ b/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1/epoch_5.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:3cfbb84f1e4f42a758f67e25ca0758b17bac94d3675fd3d22e83120c6c52e58a +size 782738650 diff --git a/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1/epoch_6.pth b/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1/epoch_6.pth new file mode 100644 index 0000000000000000000000000000000000000000..634e0f5ea497e92354fdf25945cf0d3fa627d3b9 --- /dev/null +++ b/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1/epoch_6.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:1b11ed0bd2d0a02b1eedb094a4936ba1a4bf0a337f2bf7eaa6a4065eae6dddff +size 782889946 diff --git a/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1/epoch_7.pth b/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1/epoch_7.pth new file mode 100644 index 0000000000000000000000000000000000000000..f3d9eb09ad67aede5688f53b026b1e86ec9d1f5e --- /dev/null +++ b/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1/epoch_7.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:dfc0e2a1e12e2249cb05468d57fb32c10d627db22d619e64f41401495b49cbdf +size 783040986 diff --git a/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1/epoch_8.pth b/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1/epoch_8.pth new file mode 100644 index 0000000000000000000000000000000000000000..c40fa240c64a37f80fb198ac81e16de0ac182089 --- /dev/null +++ b/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1/epoch_8.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:e0b6221f25eea69048f0ae586653473f44f8643557dd07dbd00d44cc88cfb05c +size 783192090 diff --git a/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1/epoch_9.pth b/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1/epoch_9.pth new file mode 100644 index 0000000000000000000000000000000000000000..b73d33023dc1990040abde8e6e45d86ed6aca673 --- /dev/null +++ b/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1/epoch_9.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:cc1c01468148fdb3840bf91e377829dc74002a286a930ec1fe8268065d5d5832 +size 783343706 diff --git a/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1/last_checkpoint b/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1/last_checkpoint new file mode 100644 index 0000000000000000000000000000000000000000..d7ca65c3f81b66f0dc29f5488280e83fb1ee0a02 --- /dev/null +++ b/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1/last_checkpoint @@ -0,0 +1 @@ +/mnt/petrelfs/luzeyu/workspace/fakebench/mmpretrain/workdir/convnext_small_4xb256_all_1m_lr4e-4_aug_5e-1/epoch_10.pth \ No newline at end of file diff --git a/convnext_small_4xb256_fake5m-lr4e-4/20230601_025223/20230601_025223.log b/convnext_small_4xb256_fake5m-lr4e-4/20230601_025223/20230601_025223.log new file mode 100644 index 0000000000000000000000000000000000000000..5d7b93844fa7f9e600303e95a413ce86d571d786 --- /dev/null +++ b/convnext_small_4xb256_fake5m-lr4e-4/20230601_025223/20230601_025223.log @@ -0,0 +1,2591 @@ +2023/06/01 02:52:27 - mmengine - INFO - +------------------------------------------------------------ +System environment: + sys.platform: linux + Python: 3.10.9 (main, Mar 8 2023, 10:47:38) [GCC 11.2.0] + CUDA available: True + numpy_random_seed: 630854578 + GPU 0,1,2,3: NVIDIA A100-SXM4-80GB + CUDA_HOME: /mnt/petrelfs/share/cuda-11.6 + NVCC: Cuda compilation tools, release 11.6, V11.6.124 + GCC: gcc (GCC) 4.8.5 20150623 (Red Hat 4.8.5-44) + PyTorch: 1.13.1 + PyTorch compiling details: PyTorch built with: + - GCC 9.3 + - C++ Version: 201402 + - Intel(R) oneAPI Math Kernel Library Version 2021.4-Product Build 20210904 for Intel(R) 64 architecture applications + - Intel(R) MKL-DNN v2.6.0 (Git Hash 52b5f107dd9cf10910aaa19cb47f3abf9b349815) + - OpenMP 201511 (a.k.a. OpenMP 4.5) + - LAPACK is enabled (usually provided by MKL) + - NNPACK is enabled + - CPU capability usage: AVX2 + - CUDA Runtime 11.6 + - NVCC architecture flags: -gencode;arch=compute_37,code=sm_37;-gencode;arch=compute_50,code=sm_50;-gencode;arch=compute_60,code=sm_60;-gencode;arch=compute_61,code=sm_61;-gencode;arch=compute_70,code=sm_70;-gencode;arch=compute_75,code=sm_75;-gencode;arch=compute_80,code=sm_80;-gencode;arch=compute_86,code=sm_86;-gencode;arch=compute_37,code=compute_37 + - CuDNN 8.3.2 (built against CUDA 11.5) + - Magma 2.6.1 + - Build settings: BLAS_INFO=mkl, BUILD_TYPE=Release, CUDA_VERSION=11.6, CUDNN_VERSION=8.3.2, CXX_COMPILER=/opt/rh/devtoolset-9/root/usr/bin/c++, CXX_FLAGS= -fabi-version=11 -Wno-deprecated -fvisibility-inlines-hidden -DUSE_PTHREADPOOL -fopenmp -DNDEBUG -DUSE_KINETO -DUSE_FBGEMM -DUSE_QNNPACK -DUSE_PYTORCH_QNNPACK -DUSE_XNNPACK -DSYMBOLICATE_MOBILE_DEBUG_HANDLE -DEDGE_PROFILER_USE_KINETO -O2 -fPIC -Wno-narrowing -Wall -Wextra -Werror=return-type -Werror=non-virtual-dtor -Wno-missing-field-initializers -Wno-type-limits -Wno-array-bounds -Wno-unknown-pragmas -Wunused-local-typedefs -Wno-unused-parameter -Wno-unused-function -Wno-unused-result -Wno-strict-overflow -Wno-strict-aliasing -Wno-error=deprecated-declarations -Wno-stringop-overflow -Wno-psabi -Wno-error=pedantic -Wno-error=redundant-decls -Wno-error=old-style-cast -fdiagnostics-color=always -faligned-new -Wno-unused-but-set-variable -Wno-maybe-uninitialized -fno-math-errno -fno-trapping-math -Werror=format -Werror=cast-function-type -Wno-stringop-overflow, LAPACK_INFO=mkl, PERF_WITH_AVX=1, PERF_WITH_AVX2=1, PERF_WITH_AVX512=1, TORCH_VERSION=1.13.1, USE_CUDA=ON, USE_CUDNN=ON, USE_EXCEPTION_PTR=1, USE_GFLAGS=OFF, USE_GLOG=OFF, USE_MKL=ON, USE_MKLDNN=ON, USE_MPI=OFF, USE_NCCL=ON, USE_NNPACK=ON, USE_OPENMP=ON, USE_ROCM=OFF, + + TorchVision: 0.14.1 + OpenCV: 4.7.0 + MMEngine: 0.7.3 + +Runtime environment: + cudnn_benchmark: True + mp_cfg: {'mp_start_method': 'fork', 'opencv_num_threads': 0} + dist_cfg: {'backend': 'nccl'} + seed: None + deterministic: False + Distributed launcher: slurm + Distributed training: True + GPU number: 4 +------------------------------------------------------------ + +2023/06/01 02:52:27 - mmengine - INFO - Config: +optim_wrapper = dict( + optimizer=dict( + type='AdamW', + lr=0.0004, + weight_decay=0.05, + eps=1e-08, + betas=(0.9, 0.999), + _scope_='mmpretrain'), + paramwise_cfg=dict( + norm_decay_mult=0.0, + bias_decay_mult=0.0, + flat_decay_mult=0.0, + custom_keys=dict({ + '.absolute_pos_embed': dict(decay_mult=0.0), + '.relative_position_bias_table': dict(decay_mult=0.0) + })), + type='AmpOptimWrapper', + dtype='bfloat16', + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=True, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=20, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=4096) +model = dict( + type='ImageClassifier', + backbone=dict(type='ConvNeXt', arch='small', drop_path_rate=0.4), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=768, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + init_cfg=None), + init_cfg=dict( + type='TruncNormal', layer=['Conv2d', 'Linear'], std=0.02, bias=0.0), + train_cfg=None) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stablediffusionV1-5R2-dpmsolver-25-5m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/convnext_small_4xb256_fake5m-lr4e-4' + +2023/06/01 02:52:31 - mmengine - INFO - Hooks will be executed in the following order: +before_run: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_load_checkpoint: +(ABOVE_NORMAL) EMAHook + -------------------- +before_train: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_train_epoch: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(NORMAL ) DistSamplerSeedHook + -------------------- +before_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook + -------------------- +after_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +after_train_epoch: +(NORMAL ) IterTimerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_val_epoch: +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook + -------------------- +before_val_iter: +(NORMAL ) IterTimerHook + -------------------- +after_val_iter: +(NORMAL ) IterTimerHook +(NORMAL ) VisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_val_epoch: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_save_checkpoint: +(ABOVE_NORMAL) EMAHook + -------------------- +after_train: +(VERY_LOW ) CheckpointHook + -------------------- +before_test_epoch: +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook + -------------------- +before_test_iter: +(NORMAL ) IterTimerHook + -------------------- +after_test_iter: +(NORMAL ) IterTimerHook +(NORMAL ) VisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_test_epoch: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_run: +(BELOW_NORMAL) LoggerHook + -------------------- +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.0.0.bias:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.0.1.weight:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.0.1.bias:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.1.0.weight:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.1.0.bias:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.1.1.bias:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.2.0.weight:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.2.0.bias:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.2.1.bias:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.3.0.weight:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.3.0.bias:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.3.1.bias:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.0.0.gamma:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.0.0.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.0.0.norm.weight:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.0.0.norm.bias:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.0.0.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.0.0.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.0.1.gamma:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.0.1.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.0.1.norm.weight:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.0.1.norm.bias:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.0.1.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.0.1.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.0.2.gamma:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.0.2.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.0.2.norm.weight:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.0.2.norm.bias:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.0.2.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.0.2.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.1.0.gamma:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.1.0.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.1.0.norm.weight:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.1.0.norm.bias:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.1.0.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.1.0.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.1.1.gamma:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.1.1.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.1.1.norm.weight:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.1.1.norm.bias:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.1.1.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.1.1.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.1.2.gamma:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.1.2.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.1.2.norm.weight:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.1.2.norm.bias:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.1.2.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.1.2.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.2.0.gamma:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.2.0.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.2.0.norm.weight:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.2.0.norm.bias:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.2.0.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.2.0.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.2.1.gamma:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.2.1.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.2.1.norm.weight:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.2.1.norm.bias:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.2.1.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.2.1.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.2.2.gamma:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.2.2.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.2.2.norm.weight:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.2.2.norm.bias:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.2.2.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.2.2.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.2.3.gamma:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.2.3.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.2.3.norm.weight:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.2.3.norm.bias:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.2.3.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.2.3.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.2.4.gamma:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.2.4.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.2.4.norm.weight:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.2.4.norm.bias:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.2.4.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.2.4.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.2.5.gamma:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.2.5.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.2.5.norm.weight:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.2.5.norm.bias:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.2.5.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.2.5.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.2.6.gamma:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.2.6.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.2.6.norm.weight:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.2.6.norm.bias:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.2.6.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.2.6.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.2.7.gamma:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.2.7.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.2.7.norm.weight:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.2.7.norm.bias:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.2.7.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.2.7.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.2.8.gamma:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.2.8.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.2.8.norm.weight:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.2.8.norm.bias:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.2.8.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.2.8.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.2.9.gamma:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.2.9.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.2.9.norm.weight:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.2.9.norm.bias:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.2.9.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.2.9.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.2.10.gamma:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.2.10.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.2.10.norm.weight:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.2.10.norm.bias:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.2.10.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.2.10.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.2.11.gamma:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.2.11.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.2.11.norm.weight:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.2.11.norm.bias:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.2.11.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.2.11.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.2.12.gamma:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.2.12.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.2.12.norm.weight:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.2.12.norm.bias:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.2.12.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.2.12.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.2.13.gamma:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.2.13.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.2.13.norm.weight:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.2.13.norm.bias:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.2.13.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.2.13.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.2.14.gamma:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.2.14.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.2.14.norm.weight:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.2.14.norm.bias:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.2.14.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.2.14.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.2.15.gamma:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.2.15.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.2.15.norm.weight:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.2.15.norm.bias:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.2.15.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.2.15.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.2.16.gamma:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.2.16.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.2.16.norm.weight:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.2.16.norm.bias:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.2.16.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.2.16.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.2.17.gamma:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.2.17.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.2.17.norm.weight:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.2.17.norm.bias:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.2.17.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.2.17.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.2.18.gamma:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.2.18.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.2.18.norm.weight:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.2.18.norm.bias:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.2.18.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.2.18.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.2.19.gamma:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.2.19.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.2.19.norm.weight:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.2.19.norm.bias:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.2.19.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.2.19.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.2.20.gamma:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.2.20.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.2.20.norm.weight:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.2.20.norm.bias:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.2.20.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.2.20.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.2.21.gamma:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.2.21.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.2.21.norm.weight:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.2.21.norm.bias:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.2.21.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.2.21.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.2.22.gamma:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.2.22.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.2.22.norm.weight:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.2.22.norm.bias:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.2.22.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.2.22.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.2.23.gamma:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.2.23.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.2.23.norm.weight:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.2.23.norm.bias:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.2.23.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.2.23.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.2.24.gamma:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.2.24.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.2.24.norm.weight:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.2.24.norm.bias:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.2.24.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.2.24.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.2.25.gamma:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.2.25.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.2.25.norm.weight:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.2.25.norm.bias:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.2.25.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.2.25.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.2.26.gamma:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.2.26.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.2.26.norm.weight:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.2.26.norm.bias:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.2.26.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.2.26.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.3.0.gamma:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.3.0.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.3.0.norm.weight:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.3.0.norm.bias:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.3.0.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.3.0.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.3.1.gamma:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.3.1.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.3.1.norm.weight:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.3.1.norm.bias:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.3.1.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.3.1.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.3.2.gamma:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.3.2.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.3.2.norm.weight:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.3.2.norm.bias:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.3.2.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.stages.3.2.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.norm3.weight:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- backbone.norm3.bias:weight_decay=0.0 +2023/06/01 02:52:58 - mmengine - INFO - paramwise_options -- head.fc.bias:weight_decay=0.0 +Name of parameter - Initialization information + +backbone.downsample_layers.0.0.weight - torch.Size([96, 3, 4, 4]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.downsample_layers.0.0.bias - torch.Size([96]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.downsample_layers.0.1.weight - torch.Size([96]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.downsample_layers.0.1.bias - torch.Size([96]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.downsample_layers.1.0.weight - torch.Size([96]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.downsample_layers.1.0.bias - torch.Size([96]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.downsample_layers.1.1.weight - torch.Size([192, 96, 2, 2]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.downsample_layers.1.1.bias - torch.Size([192]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.downsample_layers.2.0.weight - torch.Size([192]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.downsample_layers.2.0.bias - torch.Size([192]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.downsample_layers.2.1.weight - torch.Size([384, 192, 2, 2]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.downsample_layers.2.1.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.downsample_layers.3.0.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.downsample_layers.3.0.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.downsample_layers.3.1.weight - torch.Size([768, 384, 2, 2]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.downsample_layers.3.1.bias - torch.Size([768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.0.gamma - torch.Size([96]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.0.0.depthwise_conv.weight - torch.Size([96, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.0.depthwise_conv.bias - torch.Size([96]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.0.norm.weight - torch.Size([96]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.0.0.norm.bias - torch.Size([96]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.0.0.pointwise_conv1.weight - torch.Size([384, 96]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.0.pointwise_conv1.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.0.pointwise_conv2.weight - torch.Size([96, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.0.pointwise_conv2.bias - torch.Size([96]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.1.gamma - torch.Size([96]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.0.1.depthwise_conv.weight - torch.Size([96, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.1.depthwise_conv.bias - torch.Size([96]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.1.norm.weight - torch.Size([96]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.0.1.norm.bias - torch.Size([96]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.0.1.pointwise_conv1.weight - torch.Size([384, 96]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.1.pointwise_conv1.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.1.pointwise_conv2.weight - torch.Size([96, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.1.pointwise_conv2.bias - torch.Size([96]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.2.gamma - torch.Size([96]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.0.2.depthwise_conv.weight - torch.Size([96, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.2.depthwise_conv.bias - torch.Size([96]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.2.norm.weight - torch.Size([96]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.0.2.norm.bias - torch.Size([96]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.0.2.pointwise_conv1.weight - torch.Size([384, 96]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.2.pointwise_conv1.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.2.pointwise_conv2.weight - torch.Size([96, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.2.pointwise_conv2.bias - torch.Size([96]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.0.gamma - torch.Size([192]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.1.0.depthwise_conv.weight - torch.Size([192, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.0.depthwise_conv.bias - torch.Size([192]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.0.norm.weight - torch.Size([192]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.1.0.norm.bias - torch.Size([192]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.1.0.pointwise_conv1.weight - torch.Size([768, 192]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.0.pointwise_conv1.bias - torch.Size([768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.0.pointwise_conv2.weight - torch.Size([192, 768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.0.pointwise_conv2.bias - torch.Size([192]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.1.gamma - torch.Size([192]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.1.1.depthwise_conv.weight - torch.Size([192, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.1.depthwise_conv.bias - torch.Size([192]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.1.norm.weight - torch.Size([192]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.1.1.norm.bias - torch.Size([192]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.1.1.pointwise_conv1.weight - torch.Size([768, 192]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.1.pointwise_conv1.bias - torch.Size([768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.1.pointwise_conv2.weight - torch.Size([192, 768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.1.pointwise_conv2.bias - torch.Size([192]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.2.gamma - torch.Size([192]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.1.2.depthwise_conv.weight - torch.Size([192, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.2.depthwise_conv.bias - torch.Size([192]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.2.norm.weight - torch.Size([192]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.1.2.norm.bias - torch.Size([192]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.1.2.pointwise_conv1.weight - torch.Size([768, 192]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.2.pointwise_conv1.bias - torch.Size([768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.2.pointwise_conv2.weight - torch.Size([192, 768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.2.pointwise_conv2.bias - torch.Size([192]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.0.gamma - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.0.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.0.depthwise_conv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.0.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.0.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.0.pointwise_conv1.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.0.pointwise_conv1.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.0.pointwise_conv2.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.0.pointwise_conv2.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.1.gamma - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.1.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.1.depthwise_conv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.1.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.1.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.1.pointwise_conv1.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.1.pointwise_conv1.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.1.pointwise_conv2.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.1.pointwise_conv2.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.2.gamma - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.2.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.2.depthwise_conv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.2.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.2.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.2.pointwise_conv1.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.2.pointwise_conv1.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.2.pointwise_conv2.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.2.pointwise_conv2.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.3.gamma - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.3.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.3.depthwise_conv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.3.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.3.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.3.pointwise_conv1.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.3.pointwise_conv1.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.3.pointwise_conv2.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.3.pointwise_conv2.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.4.gamma - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.4.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.4.depthwise_conv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.4.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.4.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.4.pointwise_conv1.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.4.pointwise_conv1.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.4.pointwise_conv2.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.4.pointwise_conv2.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.5.gamma - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.5.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.5.depthwise_conv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.5.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.5.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.5.pointwise_conv1.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.5.pointwise_conv1.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.5.pointwise_conv2.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.5.pointwise_conv2.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.6.gamma - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.6.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.6.depthwise_conv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.6.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.6.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.6.pointwise_conv1.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.6.pointwise_conv1.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.6.pointwise_conv2.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.6.pointwise_conv2.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.7.gamma - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.7.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.7.depthwise_conv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.7.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.7.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.7.pointwise_conv1.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.7.pointwise_conv1.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.7.pointwise_conv2.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.7.pointwise_conv2.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.8.gamma - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.8.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.8.depthwise_conv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.8.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.8.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.8.pointwise_conv1.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.8.pointwise_conv1.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.8.pointwise_conv2.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.8.pointwise_conv2.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.9.gamma - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.9.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.9.depthwise_conv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.9.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.9.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.9.pointwise_conv1.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.9.pointwise_conv1.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.9.pointwise_conv2.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.9.pointwise_conv2.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.10.gamma - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.10.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.10.depthwise_conv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.10.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.10.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.10.pointwise_conv1.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.10.pointwise_conv1.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.10.pointwise_conv2.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.10.pointwise_conv2.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.11.gamma - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.11.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.11.depthwise_conv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.11.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.11.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.11.pointwise_conv1.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.11.pointwise_conv1.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.11.pointwise_conv2.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.11.pointwise_conv2.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.12.gamma - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.12.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.12.depthwise_conv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.12.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.12.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.12.pointwise_conv1.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.12.pointwise_conv1.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.12.pointwise_conv2.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.12.pointwise_conv2.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.13.gamma - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.13.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.13.depthwise_conv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.13.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.13.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.13.pointwise_conv1.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.13.pointwise_conv1.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.13.pointwise_conv2.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.13.pointwise_conv2.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.14.gamma - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.14.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.14.depthwise_conv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.14.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.14.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.14.pointwise_conv1.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.14.pointwise_conv1.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.14.pointwise_conv2.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.14.pointwise_conv2.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.15.gamma - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.15.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.15.depthwise_conv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.15.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.15.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.15.pointwise_conv1.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.15.pointwise_conv1.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.15.pointwise_conv2.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.15.pointwise_conv2.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.16.gamma - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.16.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.16.depthwise_conv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.16.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.16.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.16.pointwise_conv1.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.16.pointwise_conv1.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.16.pointwise_conv2.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.16.pointwise_conv2.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.17.gamma - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.17.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.17.depthwise_conv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.17.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.17.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.17.pointwise_conv1.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.17.pointwise_conv1.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.17.pointwise_conv2.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.17.pointwise_conv2.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.18.gamma - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.18.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.18.depthwise_conv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.18.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.18.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.18.pointwise_conv1.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.18.pointwise_conv1.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.18.pointwise_conv2.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.18.pointwise_conv2.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.19.gamma - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.19.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.19.depthwise_conv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.19.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.19.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.19.pointwise_conv1.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.19.pointwise_conv1.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.19.pointwise_conv2.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.19.pointwise_conv2.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.20.gamma - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.20.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.20.depthwise_conv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.20.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.20.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.20.pointwise_conv1.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.20.pointwise_conv1.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.20.pointwise_conv2.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.20.pointwise_conv2.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.21.gamma - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.21.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.21.depthwise_conv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.21.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.21.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.21.pointwise_conv1.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.21.pointwise_conv1.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.21.pointwise_conv2.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.21.pointwise_conv2.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.22.gamma - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.22.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.22.depthwise_conv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.22.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.22.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.22.pointwise_conv1.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.22.pointwise_conv1.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.22.pointwise_conv2.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.22.pointwise_conv2.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.23.gamma - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.23.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.23.depthwise_conv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.23.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.23.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.23.pointwise_conv1.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.23.pointwise_conv1.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.23.pointwise_conv2.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.23.pointwise_conv2.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.24.gamma - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.24.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.24.depthwise_conv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.24.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.24.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.24.pointwise_conv1.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.24.pointwise_conv1.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.24.pointwise_conv2.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.24.pointwise_conv2.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.25.gamma - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.25.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.25.depthwise_conv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.25.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.25.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.25.pointwise_conv1.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.25.pointwise_conv1.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.25.pointwise_conv2.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.25.pointwise_conv2.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.26.gamma - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.26.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.26.depthwise_conv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.26.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.26.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.26.pointwise_conv1.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.26.pointwise_conv1.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.26.pointwise_conv2.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.26.pointwise_conv2.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.0.gamma - torch.Size([768]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.3.0.depthwise_conv.weight - torch.Size([768, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.0.depthwise_conv.bias - torch.Size([768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.0.norm.weight - torch.Size([768]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.3.0.norm.bias - torch.Size([768]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.3.0.pointwise_conv1.weight - torch.Size([3072, 768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.0.pointwise_conv1.bias - torch.Size([3072]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.0.pointwise_conv2.weight - torch.Size([768, 3072]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.0.pointwise_conv2.bias - torch.Size([768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.1.gamma - torch.Size([768]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.3.1.depthwise_conv.weight - torch.Size([768, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.1.depthwise_conv.bias - torch.Size([768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.1.norm.weight - torch.Size([768]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.3.1.norm.bias - torch.Size([768]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.3.1.pointwise_conv1.weight - torch.Size([3072, 768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.1.pointwise_conv1.bias - torch.Size([3072]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.1.pointwise_conv2.weight - torch.Size([768, 3072]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.1.pointwise_conv2.bias - torch.Size([768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.2.gamma - torch.Size([768]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.3.2.depthwise_conv.weight - torch.Size([768, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.2.depthwise_conv.bias - torch.Size([768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.2.norm.weight - torch.Size([768]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.3.2.norm.bias - torch.Size([768]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.3.2.pointwise_conv1.weight - torch.Size([3072, 768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.2.pointwise_conv1.bias - torch.Size([3072]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.2.pointwise_conv2.weight - torch.Size([768, 3072]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.2.pointwise_conv2.bias - torch.Size([768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.norm3.weight - torch.Size([768]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.norm3.bias - torch.Size([768]): +The value is the same before and after calling `init_weights` of ImageClassifier + +head.fc.weight - torch.Size([2, 768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +head.fc.bias - torch.Size([2]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 +2023/06/01 02:52:59 - mmengine - WARNING - "FileClient" will be deprecated in future. Please use io functions in https://mmengine.readthedocs.io/en/latest/api/fileio.html#file-io +2023/06/01 02:52:59 - mmengine - WARNING - "HardDiskBackend" is the alias of "LocalBackend" and the former will be deprecated in future. +2023/06/01 02:52:59 - mmengine - INFO - Checkpoints will be saved to /mnt/petrelfs/luzeyu/workspace/fakebench/mmpretrain/workdir/convnext_small_4xb256_fake5m-lr4e-4. +2023/06/01 02:54:24 - mmengine - INFO - Epoch(train) [1][ 100/5758] lr: 4.0000e-04 eta: 1 day, 3:13:12 time: 0.7355 data_time: 0.0235 memory: 25071 loss: 0.5784 +2023/06/01 02:55:40 - mmengine - INFO - Epoch(train) [1][ 200/5758] lr: 4.0000e-04 eta: 1 day, 1:44:07 time: 0.7451 data_time: 0.0013 memory: 25071 loss: 0.5478 +2023/06/01 02:56:56 - mmengine - INFO - Epoch(train) [1][ 300/5758] lr: 4.0000e-04 eta: 1 day, 1:13:14 time: 0.8194 data_time: 0.0012 memory: 25071 loss: 0.5147 +2023/06/01 02:59:12 - mmengine - INFO - Epoch(train) [1][ 400/5758] lr: 4.0000e-04 eta: 1 day, 5:42:10 time: 0.8708 data_time: 0.0013 memory: 25071 loss: 0.5042 +2023/06/01 03:00:26 - mmengine - INFO - Epoch(train) [1][ 500/5758] lr: 4.0000e-04 eta: 1 day, 4:27:36 time: 0.7508 data_time: 0.0015 memory: 25071 loss: 0.4645 +2023/06/01 03:01:42 - mmengine - INFO - Epoch(train) [1][ 600/5758] lr: 4.0000e-04 eta: 1 day, 3:44:54 time: 0.7683 data_time: 0.0013 memory: 25071 loss: 0.4523 +2023/06/01 03:02:57 - mmengine - INFO - Epoch(train) [1][ 700/5758] lr: 4.0000e-04 eta: 1 day, 3:10:32 time: 0.7672 data_time: 0.0013 memory: 25071 loss: 0.4443 +2023/06/01 03:04:12 - mmengine - INFO - Epoch(train) [1][ 800/5758] lr: 4.0000e-04 eta: 1 day, 2:44:45 time: 0.7105 data_time: 0.0012 memory: 25071 loss: 0.4505 +2023/06/01 03:05:27 - mmengine - INFO - Epoch(train) [1][ 900/5758] lr: 4.0000e-04 eta: 1 day, 2:24:02 time: 0.7121 data_time: 0.0011 memory: 25071 loss: 0.4024 +2023/06/01 03:06:43 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m-lr4e-4_20230601_025223 +2023/06/01 03:06:43 - mmengine - INFO - Epoch(train) [1][1000/5758] lr: 4.0000e-04 eta: 1 day, 2:08:24 time: 0.7268 data_time: 0.0012 memory: 25071 loss: 0.4003 +2023/06/01 03:08:01 - mmengine - INFO - Epoch(train) [1][1100/5758] lr: 4.0000e-04 eta: 1 day, 1:58:43 time: 0.7350 data_time: 0.0013 memory: 25071 loss: 0.4231 +2023/06/01 03:09:15 - mmengine - INFO - Epoch(train) [1][1200/5758] lr: 4.0000e-04 eta: 1 day, 1:44:59 time: 0.7074 data_time: 0.0013 memory: 25071 loss: 0.4058 +2023/06/01 03:10:30 - mmengine - INFO - Epoch(train) [1][1300/5758] lr: 4.0000e-04 eta: 1 day, 1:33:57 time: 0.7718 data_time: 0.0015 memory: 25071 loss: 0.4136 +2023/06/01 03:11:47 - mmengine - INFO - Epoch(train) [1][1400/5758] lr: 4.0000e-04 eta: 1 day, 1:27:35 time: 0.7975 data_time: 0.0015 memory: 25071 loss: 0.3822 +2023/06/01 03:13:04 - mmengine - INFO - Epoch(train) [1][1500/5758] lr: 4.0000e-04 eta: 1 day, 1:22:11 time: 0.7651 data_time: 0.0015 memory: 25071 loss: 0.3885 +2023/06/01 03:14:19 - mmengine - INFO - Epoch(train) [1][1600/5758] lr: 4.0000e-04 eta: 1 day, 1:14:51 time: 0.7081 data_time: 0.0016 memory: 25071 loss: 0.3577 +2023/06/01 03:15:35 - mmengine - INFO - Epoch(train) [1][1700/5758] lr: 4.0000e-04 eta: 1 day, 1:07:59 time: 0.7145 data_time: 0.0013 memory: 25071 loss: 0.3665 +2023/06/01 03:16:50 - mmengine - INFO - Epoch(train) [1][1800/5758] lr: 4.0000e-04 eta: 1 day, 1:02:10 time: 0.7648 data_time: 0.0019 memory: 25071 loss: 0.3575 +2023/06/01 03:18:07 - mmengine - INFO - Epoch(train) [1][1900/5758] lr: 4.0000e-04 eta: 1 day, 0:58:07 time: 0.7963 data_time: 0.0014 memory: 25071 loss: 0.3466 +2023/06/01 03:19:24 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m-lr4e-4_20230601_025223 +2023/06/01 03:19:24 - mmengine - INFO - Epoch(train) [1][2000/5758] lr: 4.0000e-04 eta: 1 day, 0:55:15 time: 0.7751 data_time: 0.0014 memory: 25071 loss: 0.3712 +2023/06/01 03:20:39 - mmengine - INFO - Epoch(train) [1][2100/5758] lr: 4.0000e-04 eta: 1 day, 0:50:00 time: 0.7594 data_time: 0.0013 memory: 25071 loss: 0.3137 +2023/06/01 03:21:55 - mmengine - INFO - Epoch(train) [1][2200/5758] lr: 4.0000e-04 eta: 1 day, 0:45:54 time: 0.7413 data_time: 0.0012 memory: 25071 loss: 0.3413 +2023/06/01 03:23:10 - mmengine - INFO - Epoch(train) [1][2300/5758] lr: 4.0000e-04 eta: 1 day, 0:41:08 time: 0.7055 data_time: 0.0014 memory: 25071 loss: 0.3255 +2023/06/01 03:24:25 - mmengine - INFO - Epoch(train) [1][2400/5758] lr: 4.0000e-04 eta: 1 day, 0:37:02 time: 0.8326 data_time: 0.0012 memory: 25071 loss: 0.3191 +2023/06/01 03:25:42 - mmengine - INFO - Epoch(train) [1][2500/5758] lr: 4.0000e-04 eta: 1 day, 0:34:20 time: 0.7385 data_time: 0.0015 memory: 25071 loss: 0.3149 +2023/06/01 03:26:56 - mmengine - INFO - Epoch(train) [1][2600/5758] lr: 4.0000e-04 eta: 1 day, 0:29:47 time: 0.7294 data_time: 0.0013 memory: 25071 loss: 0.3039 +2023/06/01 03:28:14 - mmengine - INFO - Epoch(train) [1][2700/5758] lr: 4.0000e-04 eta: 1 day, 0:28:12 time: 0.7716 data_time: 0.0013 memory: 25071 loss: 0.2762 +2023/06/01 03:29:30 - mmengine - INFO - Epoch(train) [1][2800/5758] lr: 4.0000e-04 eta: 1 day, 0:25:29 time: 0.7249 data_time: 0.0013 memory: 25071 loss: 0.2939 +2023/06/01 03:30:47 - mmengine - INFO - Epoch(train) [1][2900/5758] lr: 4.0000e-04 eta: 1 day, 0:23:20 time: 0.8226 data_time: 0.0013 memory: 25071 loss: 0.2997 +2023/06/01 03:32:03 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m-lr4e-4_20230601_025223 +2023/06/01 03:32:03 - mmengine - INFO - Epoch(train) [1][3000/5758] lr: 4.0000e-04 eta: 1 day, 0:20:53 time: 0.7454 data_time: 0.0012 memory: 25071 loss: 0.2650 +2023/06/01 03:33:18 - mmengine - INFO - Epoch(train) [1][3100/5758] lr: 4.0000e-04 eta: 1 day, 0:17:21 time: 0.7389 data_time: 0.0015 memory: 25071 loss: 0.2935 +2023/06/01 03:34:31 - mmengine - INFO - Epoch(train) [1][3200/5758] lr: 4.0000e-04 eta: 1 day, 0:13:17 time: 0.7104 data_time: 0.0013 memory: 25071 loss: 0.2718 +2023/06/01 03:35:48 - mmengine - INFO - Epoch(train) [1][3300/5758] lr: 4.0000e-04 eta: 1 day, 0:11:21 time: 0.7612 data_time: 0.0013 memory: 25071 loss: 0.2757 +2023/06/01 03:37:04 - mmengine - INFO - Epoch(train) [1][3400/5758] lr: 4.0000e-04 eta: 1 day, 0:09:06 time: 0.7747 data_time: 0.0012 memory: 25071 loss: 0.2523 +2023/06/01 03:38:18 - mmengine - INFO - Epoch(train) [1][3500/5758] lr: 4.0000e-04 eta: 1 day, 0:05:58 time: 0.7460 data_time: 0.0013 memory: 25071 loss: 0.2789 +2023/06/01 03:39:35 - mmengine - INFO - Epoch(train) [1][3600/5758] lr: 4.0000e-04 eta: 1 day, 0:04:05 time: 0.7365 data_time: 0.0015 memory: 25071 loss: 0.2533 +2023/06/01 03:40:51 - mmengine - INFO - Epoch(train) [1][3700/5758] lr: 4.0000e-04 eta: 1 day, 0:01:47 time: 0.7437 data_time: 0.0015 memory: 25071 loss: 0.2760 +2023/06/01 03:42:05 - mmengine - INFO - Epoch(train) [1][3800/5758] lr: 4.0000e-04 eta: 23:58:55 time: 0.7767 data_time: 0.0013 memory: 25071 loss: 0.2452 +2023/06/01 03:43:19 - mmengine - INFO - Epoch(train) [1][3900/5758] lr: 4.0000e-04 eta: 23:55:52 time: 0.7031 data_time: 0.0014 memory: 25071 loss: 0.2408 +2023/06/01 03:44:35 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m-lr4e-4_20230601_025223 +2023/06/01 03:44:35 - mmengine - INFO - Epoch(train) [1][4000/5758] lr: 4.0000e-04 eta: 23:54:06 time: 0.7432 data_time: 0.0015 memory: 25071 loss: 0.2521 +2023/06/01 03:45:51 - mmengine - INFO - Epoch(train) [1][4100/5758] lr: 4.0000e-04 eta: 23:51:56 time: 0.8582 data_time: 0.0013 memory: 25071 loss: 0.2462 +2023/06/01 03:47:06 - mmengine - INFO - Epoch(train) [1][4200/5758] lr: 4.0000e-04 eta: 23:49:36 time: 0.7725 data_time: 0.0014 memory: 25071 loss: 0.2331 +2023/06/01 03:48:21 - mmengine - INFO - Epoch(train) [1][4300/5758] lr: 4.0000e-04 eta: 23:47:22 time: 0.8162 data_time: 0.0014 memory: 25071 loss: 0.2292 +2023/06/01 03:49:37 - mmengine - INFO - Epoch(train) [1][4400/5758] lr: 4.0000e-04 eta: 23:45:41 time: 0.7541 data_time: 0.0013 memory: 25071 loss: 0.2369 +2023/06/01 03:50:52 - mmengine - INFO - Epoch(train) [1][4500/5758] lr: 4.0000e-04 eta: 23:43:37 time: 0.7501 data_time: 0.0013 memory: 25071 loss: 0.2367 +2023/06/01 03:52:08 - mmengine - INFO - Epoch(train) [1][4600/5758] lr: 4.0000e-04 eta: 23:41:42 time: 0.7481 data_time: 0.0014 memory: 25071 loss: 0.2545 +2023/06/01 03:53:23 - mmengine - INFO - Epoch(train) [1][4700/5758] lr: 4.0000e-04 eta: 23:39:30 time: 0.7297 data_time: 0.0013 memory: 25071 loss: 0.2158 +2023/06/01 03:54:39 - mmengine - INFO - Epoch(train) [1][4800/5758] lr: 4.0000e-04 eta: 23:37:50 time: 0.7454 data_time: 0.0014 memory: 25071 loss: 0.2174 +2023/06/01 03:55:55 - mmengine - INFO - Epoch(train) [1][4900/5758] lr: 4.0000e-04 eta: 23:36:12 time: 0.7197 data_time: 0.0013 memory: 25071 loss: 0.2253 +2023/06/01 03:57:09 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m-lr4e-4_20230601_025223 +2023/06/01 03:57:09 - mmengine - INFO - Epoch(train) [1][5000/5758] lr: 4.0000e-04 eta: 23:33:54 time: 0.7516 data_time: 0.0014 memory: 25071 loss: 0.2269 +2023/06/01 03:58:28 - mmengine - INFO - Epoch(train) [1][5100/5758] lr: 4.0000e-04 eta: 23:33:06 time: 0.7508 data_time: 0.0015 memory: 25071 loss: 0.1993 +2023/06/01 03:59:44 - mmengine - INFO - Epoch(train) [1][5200/5758] lr: 4.0000e-04 eta: 23:31:34 time: 0.7859 data_time: 0.0011 memory: 25071 loss: 0.2098 +2023/06/01 04:00:59 - mmengine - INFO - Epoch(train) [1][5300/5758] lr: 4.0000e-04 eta: 23:29:40 time: 0.7765 data_time: 0.0015 memory: 25071 loss: 0.1995 +2023/06/01 04:02:15 - mmengine - INFO - Epoch(train) [1][5400/5758] lr: 4.0000e-04 eta: 23:27:59 time: 0.7908 data_time: 0.0013 memory: 25071 loss: 0.1939 +2023/06/01 04:03:31 - mmengine - INFO - Epoch(train) [1][5500/5758] lr: 4.0000e-04 eta: 23:26:18 time: 0.8352 data_time: 0.0015 memory: 25071 loss: 0.1963 +2023/06/01 04:04:47 - mmengine - INFO - Epoch(train) [1][5600/5758] lr: 4.0000e-04 eta: 23:24:51 time: 0.7716 data_time: 0.0012 memory: 25071 loss: 0.2095 +2023/06/01 04:06:04 - mmengine - INFO - Epoch(train) [1][5700/5758] lr: 4.0000e-04 eta: 23:23:29 time: 0.7181 data_time: 0.0013 memory: 25071 loss: 0.1822 +2023/06/01 04:06:48 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m-lr4e-4_20230601_025223 +2023/06/01 04:06:48 - mmengine - INFO - Saving checkpoint at 1 epochs +2023/06/01 04:07:09 - mmengine - INFO - Epoch(val) [1][16/16] accuracy/top1: 91.8061 single-label/precision_classwise: [100.0, 0.0] single-label/recall_classwise: [91.80606842041016, 0.0] single-label/f1-score_classwise: [95.72801208496094, 0.0] data_time: 0.2737 time: 0.7940 +2023/06/01 04:08:33 - mmengine - INFO - Epoch(train) [2][ 100/5758] lr: 3.9760e-04 eta: 23:23:24 time: 0.7775 data_time: 0.0565 memory: 25074 loss: 0.1782 +2023/06/01 04:09:47 - mmengine - INFO - Epoch(train) [2][ 200/5758] lr: 3.9760e-04 eta: 23:21:06 time: 0.7055 data_time: 0.0014 memory: 25074 loss: 0.1800 +2023/06/01 04:10:19 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m-lr4e-4_20230601_025223 +2023/06/01 04:11:04 - mmengine - INFO - Epoch(train) [2][ 300/5758] lr: 3.9760e-04 eta: 23:19:48 time: 0.7790 data_time: 0.0016 memory: 25074 loss: 0.1910 +2023/06/01 04:12:20 - mmengine - INFO - Epoch(train) [2][ 400/5758] lr: 3.9760e-04 eta: 23:18:11 time: 0.8070 data_time: 0.0014 memory: 25074 loss: 0.1830 +2023/06/01 04:13:35 - mmengine - INFO - Epoch(train) [2][ 500/5758] lr: 3.9760e-04 eta: 23:16:21 time: 0.7681 data_time: 0.0015 memory: 25074 loss: 0.1616 +2023/06/01 04:14:52 - mmengine - INFO - Epoch(train) [2][ 600/5758] lr: 3.9760e-04 eta: 23:15:09 time: 0.7869 data_time: 0.0014 memory: 25074 loss: 0.1782 +2023/06/01 04:16:08 - mmengine - INFO - Epoch(train) [2][ 700/5758] lr: 3.9760e-04 eta: 23:13:36 time: 0.7201 data_time: 0.0015 memory: 25074 loss: 0.1541 +2023/06/01 04:17:25 - mmengine - INFO - Epoch(train) [2][ 800/5758] lr: 3.9760e-04 eta: 23:12:32 time: 0.7324 data_time: 0.0016 memory: 25074 loss: 0.2027 +2023/06/01 04:18:42 - mmengine - INFO - Epoch(train) [2][ 900/5758] lr: 3.9760e-04 eta: 23:11:00 time: 0.7485 data_time: 0.0012 memory: 25074 loss: 0.1709 +2023/06/01 04:19:56 - mmengine - INFO - Epoch(train) [2][1000/5758] lr: 3.9760e-04 eta: 23:09:04 time: 0.7299 data_time: 0.0014 memory: 25074 loss: 0.1712 +2023/06/01 04:21:12 - mmengine - INFO - Epoch(train) [2][1100/5758] lr: 3.9760e-04 eta: 23:07:38 time: 0.7633 data_time: 0.0021 memory: 25074 loss: 0.1817 +2023/06/01 04:22:27 - mmengine - INFO - Epoch(train) [2][1200/5758] lr: 3.9760e-04 eta: 23:05:54 time: 0.7628 data_time: 0.0014 memory: 25074 loss: 0.1518 +2023/06/01 04:22:59 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m-lr4e-4_20230601_025223 +2023/06/01 04:23:44 - mmengine - INFO - Epoch(train) [2][1300/5758] lr: 3.9760e-04 eta: 23:04:29 time: 0.8086 data_time: 0.0015 memory: 25074 loss: 0.1438 +2023/06/01 04:24:59 - mmengine - INFO - Epoch(train) [2][1400/5758] lr: 3.9760e-04 eta: 23:02:42 time: 0.7716 data_time: 0.0011 memory: 25074 loss: 0.1663 +2023/06/01 04:26:15 - mmengine - INFO - Epoch(train) [2][1500/5758] lr: 3.9760e-04 eta: 23:01:13 time: 0.7307 data_time: 0.0012 memory: 25074 loss: 0.1608 +2023/06/01 04:27:31 - mmengine - INFO - Epoch(train) [2][1600/5758] lr: 3.9760e-04 eta: 22:59:51 time: 0.7474 data_time: 0.0014 memory: 25074 loss: 0.1664 +2023/06/01 04:28:47 - mmengine - INFO - Epoch(train) [2][1700/5758] lr: 3.9760e-04 eta: 22:58:19 time: 0.7901 data_time: 0.0013 memory: 25074 loss: 0.2015 +2023/06/01 04:30:12 - mmengine - INFO - Epoch(train) [2][1800/5758] lr: 3.9760e-04 eta: 22:59:07 time: 0.8205 data_time: 0.0016 memory: 25074 loss: 0.1418 +2023/06/01 04:31:31 - mmengine - INFO - Epoch(train) [2][1900/5758] lr: 3.9760e-04 eta: 22:58:09 time: 0.7210 data_time: 0.0015 memory: 25074 loss: 0.1437 +2023/06/01 04:32:46 - mmengine - INFO - Epoch(train) [2][2000/5758] lr: 3.9760e-04 eta: 22:56:33 time: 0.7201 data_time: 0.0012 memory: 25074 loss: 0.1375 +2023/06/01 04:34:01 - mmengine - INFO - Epoch(train) [2][2100/5758] lr: 3.9760e-04 eta: 22:54:53 time: 0.7130 data_time: 0.0012 memory: 25074 loss: 0.1309 +2023/06/01 04:35:17 - mmengine - INFO - Epoch(train) [2][2200/5758] lr: 3.9760e-04 eta: 22:53:13 time: 0.7774 data_time: 0.0017 memory: 25074 loss: 0.1502 +2023/06/01 04:35:48 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m-lr4e-4_20230601_025223 +2023/06/01 04:36:31 - mmengine - INFO - Epoch(train) [2][2300/5758] lr: 3.9760e-04 eta: 22:51:27 time: 0.7510 data_time: 0.0013 memory: 25074 loss: 0.1180 +2023/06/01 04:37:45 - mmengine - INFO - Epoch(train) [2][2400/5758] lr: 3.9760e-04 eta: 22:49:32 time: 0.7999 data_time: 0.0017 memory: 25074 loss: 0.1314 +2023/06/01 04:39:01 - mmengine - INFO - Epoch(train) [2][2500/5758] lr: 3.9760e-04 eta: 22:48:06 time: 0.8065 data_time: 0.0013 memory: 25074 loss: 0.1316 +2023/06/01 04:40:17 - mmengine - INFO - Epoch(train) [2][2600/5758] lr: 3.9760e-04 eta: 22:46:38 time: 0.8055 data_time: 0.0013 memory: 25074 loss: 0.1340 +2023/06/01 04:41:33 - mmengine - INFO - Epoch(train) [2][2700/5758] lr: 3.9760e-04 eta: 22:45:11 time: 0.7320 data_time: 0.0013 memory: 25074 loss: 0.1468 +2023/06/01 04:42:48 - mmengine - INFO - Epoch(train) [2][2800/5758] lr: 3.9760e-04 eta: 22:43:32 time: 0.7309 data_time: 0.0217 memory: 25074 loss: 0.1461 +2023/06/01 04:44:03 - mmengine - INFO - Epoch(train) [2][2900/5758] lr: 3.9760e-04 eta: 22:41:56 time: 0.7167 data_time: 0.0013 memory: 25074 loss: 0.1379 +2023/06/01 04:45:21 - mmengine - INFO - Epoch(train) [2][3000/5758] lr: 3.9760e-04 eta: 22:40:49 time: 0.8110 data_time: 0.0013 memory: 25074 loss: 0.1202 +2023/06/01 04:46:39 - mmengine - INFO - Epoch(train) [2][3100/5758] lr: 3.9760e-04 eta: 22:39:51 time: 0.7776 data_time: 0.0014 memory: 25074 loss: 0.1139 +2023/06/01 04:47:54 - mmengine - INFO - Epoch(train) [2][3200/5758] lr: 3.9760e-04 eta: 22:38:14 time: 0.7573 data_time: 0.0015 memory: 25074 loss: 0.1117 +2023/06/01 04:48:25 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m-lr4e-4_20230601_025223 +2023/06/01 04:49:11 - mmengine - INFO - Epoch(train) [2][3300/5758] lr: 3.9760e-04 eta: 22:36:54 time: 0.8192 data_time: 0.0015 memory: 25074 loss: 0.1004 +2023/06/01 04:50:26 - mmengine - INFO - Epoch(train) [2][3400/5758] lr: 3.9760e-04 eta: 22:35:18 time: 0.7964 data_time: 0.0012 memory: 25074 loss: 0.1033 +2023/06/01 04:51:41 - mmengine - INFO - Epoch(train) [2][3500/5758] lr: 3.9760e-04 eta: 22:33:47 time: 0.7701 data_time: 0.0019 memory: 25074 loss: 0.1100 +2023/06/01 04:52:56 - mmengine - INFO - Epoch(train) [2][3600/5758] lr: 3.9760e-04 eta: 22:32:10 time: 0.7280 data_time: 0.0015 memory: 25074 loss: 0.1168 +2023/06/01 04:54:12 - mmengine - INFO - Epoch(train) [2][3700/5758] lr: 3.9760e-04 eta: 22:30:43 time: 0.7469 data_time: 0.0015 memory: 25074 loss: 0.1131 +2023/06/01 04:55:27 - mmengine - INFO - Epoch(train) [2][3800/5758] lr: 3.9760e-04 eta: 22:29:11 time: 0.7641 data_time: 0.0013 memory: 25074 loss: 0.1271 +2023/06/01 04:56:42 - mmengine - INFO - Epoch(train) [2][3900/5758] lr: 3.9760e-04 eta: 22:27:34 time: 0.7198 data_time: 0.0013 memory: 25074 loss: 0.1022 +2023/06/01 04:57:58 - mmengine - INFO - Epoch(train) [2][4000/5758] lr: 3.9760e-04 eta: 22:26:09 time: 0.7557 data_time: 0.0015 memory: 25074 loss: 0.1157 +2023/06/01 04:59:13 - mmengine - INFO - Epoch(train) [2][4100/5758] lr: 3.9760e-04 eta: 22:24:34 time: 0.7587 data_time: 0.0013 memory: 25074 loss: 0.0872 +2023/06/01 05:00:28 - mmengine - INFO - Epoch(train) [2][4200/5758] lr: 3.9760e-04 eta: 22:23:08 time: 0.7446 data_time: 0.0014 memory: 25074 loss: 0.1129 +2023/06/01 05:01:00 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m-lr4e-4_20230601_025223 +2023/06/01 05:01:44 - mmengine - INFO - Epoch(train) [2][4300/5758] lr: 3.9760e-04 eta: 22:21:44 time: 0.7774 data_time: 0.0014 memory: 25074 loss: 0.0923 +2023/06/01 05:03:02 - mmengine - INFO - Epoch(train) [2][4400/5758] lr: 3.9760e-04 eta: 22:20:40 time: 0.7308 data_time: 0.0012 memory: 25074 loss: 0.0991 +2023/06/01 05:04:21 - mmengine - INFO - Epoch(train) [2][4500/5758] lr: 3.9760e-04 eta: 22:19:42 time: 0.8337 data_time: 0.0014 memory: 25074 loss: 0.0925 +2023/06/01 05:05:37 - mmengine - INFO - Epoch(train) [2][4600/5758] lr: 3.9760e-04 eta: 22:18:20 time: 0.7156 data_time: 0.0014 memory: 25074 loss: 0.0899 +2023/06/01 05:06:52 - mmengine - INFO - Epoch(train) [2][4700/5758] lr: 3.9760e-04 eta: 22:16:50 time: 0.7531 data_time: 0.0014 memory: 25074 loss: 0.0911 +2023/06/01 05:08:09 - mmengine - INFO - Epoch(train) [2][4800/5758] lr: 3.9760e-04 eta: 22:15:35 time: 0.7037 data_time: 0.0011 memory: 25074 loss: 0.0824 +2023/06/01 05:09:24 - mmengine - INFO - Epoch(train) [2][4900/5758] lr: 3.9760e-04 eta: 22:14:10 time: 0.7221 data_time: 0.0013 memory: 25074 loss: 0.0937 +2023/06/01 05:10:39 - mmengine - INFO - Epoch(train) [2][5000/5758] lr: 3.9760e-04 eta: 22:12:36 time: 0.7704 data_time: 0.0014 memory: 25074 loss: 0.0817 +2023/06/01 05:11:55 - mmengine - INFO - Epoch(train) [2][5100/5758] lr: 3.9760e-04 eta: 22:11:08 time: 0.7601 data_time: 0.0013 memory: 25074 loss: 0.0884 +2023/06/01 05:13:10 - mmengine - INFO - Epoch(train) [2][5200/5758] lr: 3.9760e-04 eta: 22:09:35 time: 0.7458 data_time: 0.0013 memory: 25074 loss: 0.0916 +2023/06/01 05:13:40 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m-lr4e-4_20230601_025223 +2023/06/01 05:14:24 - mmengine - INFO - Epoch(train) [2][5300/5758] lr: 3.9760e-04 eta: 22:07:58 time: 0.7738 data_time: 0.0015 memory: 25074 loss: 0.0801 +2023/06/01 05:15:41 - mmengine - INFO - Epoch(train) [2][5400/5758] lr: 3.9760e-04 eta: 22:06:45 time: 0.7339 data_time: 0.0016 memory: 25074 loss: 0.0808 +2023/06/01 05:16:57 - mmengine - INFO - Epoch(train) [2][5500/5758] lr: 3.9760e-04 eta: 22:05:23 time: 0.7791 data_time: 0.0013 memory: 25074 loss: 0.0726 +2023/06/01 05:18:12 - mmengine - INFO - Epoch(train) [2][5600/5758] lr: 3.9760e-04 eta: 22:03:51 time: 0.7348 data_time: 0.0012 memory: 25074 loss: 0.0649 +2023/06/01 05:19:28 - mmengine - INFO - Epoch(train) [2][5700/5758] lr: 3.9760e-04 eta: 22:02:32 time: 0.7162 data_time: 0.0018 memory: 25074 loss: 0.0762 +2023/06/01 05:20:12 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m-lr4e-4_20230601_025223 +2023/06/01 05:20:12 - mmengine - INFO - Saving checkpoint at 2 epochs +2023/06/01 05:20:31 - mmengine - INFO - Epoch(val) [2][16/16] accuracy/top1: 98.9016 single-label/precision_classwise: [100.0, 0.0] single-label/recall_classwise: [98.90158081054688, 0.0] single-label/f1-score_classwise: [99.44776153564453, 0.0] data_time: 0.2710 time: 0.6596 +2023/06/01 05:21:54 - mmengine - INFO - Epoch(train) [3][ 100/5758] lr: 3.9046e-04 eta: 22:01:26 time: 0.7157 data_time: 0.0011 memory: 25074 loss: 0.0745 +2023/06/01 05:23:11 - mmengine - INFO - Epoch(train) [3][ 200/5758] lr: 3.9046e-04 eta: 22:00:13 time: 0.7966 data_time: 0.0012 memory: 25074 loss: 0.0684 +2023/06/01 05:24:25 - mmengine - INFO - Epoch(train) [3][ 300/5758] lr: 3.9046e-04 eta: 21:58:34 time: 0.7046 data_time: 0.0017 memory: 25074 loss: 0.0831 +2023/06/01 05:25:40 - mmengine - INFO - Epoch(train) [3][ 400/5758] lr: 3.9046e-04 eta: 21:57:00 time: 0.7289 data_time: 0.0014 memory: 25074 loss: 0.0760 +2023/06/01 05:26:44 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m-lr4e-4_20230601_025223 +2023/06/01 05:26:56 - mmengine - INFO - Epoch(train) [3][ 500/5758] lr: 3.9046e-04 eta: 21:55:41 time: 0.7726 data_time: 0.0016 memory: 25074 loss: 0.0670 +2023/06/01 05:28:10 - mmengine - INFO - Epoch(train) [3][ 600/5758] lr: 3.9046e-04 eta: 21:54:07 time: 0.7358 data_time: 0.0012 memory: 25074 loss: 0.0660 +2023/06/01 05:29:27 - mmengine - INFO - Epoch(train) [3][ 700/5758] lr: 3.9046e-04 eta: 21:52:55 time: 0.7129 data_time: 0.0013 memory: 25074 loss: 0.0705 +2023/06/01 05:30:43 - mmengine - INFO - Epoch(train) [3][ 800/5758] lr: 3.9046e-04 eta: 21:51:30 time: 0.7293 data_time: 0.0013 memory: 25074 loss: 0.0651 +2023/06/01 05:31:57 - mmengine - INFO - Epoch(train) [3][ 900/5758] lr: 3.9046e-04 eta: 21:49:51 time: 0.7054 data_time: 0.0015 memory: 25074 loss: 0.0749 +2023/06/01 05:33:12 - mmengine - INFO - Epoch(train) [3][1000/5758] lr: 3.9046e-04 eta: 21:48:22 time: 0.7707 data_time: 0.0013 memory: 25074 loss: 0.0564 +2023/06/01 05:34:27 - mmengine - INFO - Epoch(train) [3][1100/5758] lr: 3.9046e-04 eta: 21:46:56 time: 0.7297 data_time: 0.0015 memory: 25074 loss: 0.0704 +2023/06/01 05:35:45 - mmengine - INFO - Epoch(train) [3][1200/5758] lr: 3.9046e-04 eta: 21:45:50 time: 0.7406 data_time: 0.0019 memory: 25074 loss: 0.0760 +2023/06/01 05:37:00 - mmengine - INFO - Epoch(train) [3][1300/5758] lr: 3.9046e-04 eta: 21:44:24 time: 0.7762 data_time: 0.0014 memory: 25074 loss: 0.0773 +2023/06/01 05:38:14 - mmengine - INFO - Epoch(train) [3][1400/5758] lr: 3.9046e-04 eta: 21:42:45 time: 0.7468 data_time: 0.0015 memory: 25074 loss: 0.0650 +2023/06/01 05:39:18 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m-lr4e-4_20230601_025223 +2023/06/01 05:39:29 - mmengine - INFO - Epoch(train) [3][1500/5758] lr: 3.9046e-04 eta: 21:41:22 time: 0.7174 data_time: 0.0013 memory: 25074 loss: 0.0588 +2023/06/01 05:40:44 - mmengine - INFO - Epoch(train) [3][1600/5758] lr: 3.9046e-04 eta: 21:39:49 time: 0.7174 data_time: 0.0012 memory: 25074 loss: 0.0569 +2023/06/01 05:41:59 - mmengine - INFO - Epoch(train) [3][1700/5758] lr: 3.9046e-04 eta: 21:38:21 time: 0.7747 data_time: 0.0014 memory: 25074 loss: 0.0567 +2023/06/01 05:43:13 - mmengine - INFO - Epoch(train) [3][1800/5758] lr: 3.9046e-04 eta: 21:36:47 time: 0.7548 data_time: 0.0011 memory: 25074 loss: 0.0581 +2023/06/01 05:44:29 - mmengine - INFO - Epoch(train) [3][1900/5758] lr: 3.9046e-04 eta: 21:35:27 time: 0.7594 data_time: 0.0014 memory: 25074 loss: 0.0690 +2023/06/01 05:45:44 - mmengine - INFO - Epoch(train) [3][2000/5758] lr: 3.9046e-04 eta: 21:34:00 time: 0.7768 data_time: 0.0017 memory: 25074 loss: 0.0605 +2023/06/01 05:46:58 - mmengine - INFO - Epoch(train) [3][2100/5758] lr: 3.9046e-04 eta: 21:32:26 time: 0.7126 data_time: 0.0015 memory: 25074 loss: 0.0567 +2023/06/01 05:48:14 - mmengine - INFO - Epoch(train) [3][2200/5758] lr: 3.9046e-04 eta: 21:31:07 time: 0.7780 data_time: 0.0017 memory: 25074 loss: 0.0515 +2023/06/01 05:49:31 - mmengine - INFO - Epoch(train) [3][2300/5758] lr: 3.9046e-04 eta: 21:29:55 time: 0.7747 data_time: 0.0014 memory: 25074 loss: 0.0475 +2023/06/01 05:50:46 - mmengine - INFO - Epoch(train) [3][2400/5758] lr: 3.9046e-04 eta: 21:28:30 time: 0.7782 data_time: 0.0015 memory: 25074 loss: 0.0676 +2023/06/01 05:51:50 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m-lr4e-4_20230601_025223 +2023/06/01 05:52:02 - mmengine - INFO - Epoch(train) [3][2500/5758] lr: 3.9046e-04 eta: 21:27:15 time: 0.8000 data_time: 0.0015 memory: 25074 loss: 0.0591 +2023/06/01 05:53:19 - mmengine - INFO - Epoch(train) [3][2600/5758] lr: 3.9046e-04 eta: 21:25:57 time: 0.7755 data_time: 0.0014 memory: 25074 loss: 0.0595 +2023/06/01 05:54:33 - mmengine - INFO - Epoch(train) [3][2700/5758] lr: 3.9046e-04 eta: 21:24:28 time: 0.7597 data_time: 0.0021 memory: 25074 loss: 0.0491 +2023/06/01 05:55:47 - mmengine - INFO - Epoch(train) [3][2800/5758] lr: 3.9046e-04 eta: 21:22:55 time: 0.7481 data_time: 0.0011 memory: 25074 loss: 0.0764 +2023/06/01 05:57:04 - mmengine - INFO - Epoch(train) [3][2900/5758] lr: 3.9046e-04 eta: 21:21:43 time: 0.7465 data_time: 0.0016 memory: 25074 loss: 0.0751 +2023/06/01 05:58:22 - mmengine - INFO - Epoch(train) [3][3000/5758] lr: 3.9046e-04 eta: 21:20:38 time: 0.7644 data_time: 0.0017 memory: 25074 loss: 0.0550 +2023/06/01 05:59:37 - mmengine - INFO - Epoch(train) [3][3100/5758] lr: 3.9046e-04 eta: 21:19:11 time: 0.7086 data_time: 0.0014 memory: 25074 loss: 0.0431 +2023/06/01 06:00:53 - mmengine - INFO - Epoch(train) [3][3200/5758] lr: 3.9046e-04 eta: 21:17:54 time: 0.7651 data_time: 0.0020 memory: 25074 loss: 0.0482 +2023/06/01 06:02:07 - mmengine - INFO - Epoch(train) [3][3300/5758] lr: 3.9046e-04 eta: 21:16:23 time: 0.7305 data_time: 0.0016 memory: 25074 loss: 0.0450 +2023/06/01 06:03:23 - mmengine - INFO - Epoch(train) [3][3400/5758] lr: 3.9046e-04 eta: 21:15:00 time: 0.7621 data_time: 0.0014 memory: 25074 loss: 0.0523 +2023/06/01 06:04:27 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m-lr4e-4_20230601_025223 +2023/06/01 06:04:39 - mmengine - INFO - Epoch(train) [3][3500/5758] lr: 3.9046e-04 eta: 21:13:42 time: 0.7406 data_time: 0.0020 memory: 25074 loss: 0.0451 +2023/06/01 06:05:54 - mmengine - INFO - Epoch(train) [3][3600/5758] lr: 3.9046e-04 eta: 21:12:22 time: 0.7896 data_time: 0.0014 memory: 25074 loss: 0.0478 +2023/06/01 06:07:10 - mmengine - INFO - Epoch(train) [3][3700/5758] lr: 3.9046e-04 eta: 21:10:59 time: 0.7914 data_time: 0.0014 memory: 25074 loss: 0.0440 +2023/06/01 06:08:26 - mmengine - INFO - Epoch(train) [3][3800/5758] lr: 3.9046e-04 eta: 21:09:44 time: 0.8261 data_time: 0.0018 memory: 25074 loss: 0.0476 +2023/06/01 06:09:44 - mmengine - INFO - Epoch(train) [3][3900/5758] lr: 3.9046e-04 eta: 21:08:37 time: 0.7511 data_time: 0.0013 memory: 25074 loss: 0.0440 +2023/06/01 06:10:59 - mmengine - INFO - Epoch(train) [3][4000/5758] lr: 3.9046e-04 eta: 21:07:10 time: 0.7318 data_time: 0.0017 memory: 25074 loss: 0.0345 +2023/06/01 06:12:15 - mmengine - INFO - Epoch(train) [3][4100/5758] lr: 3.9046e-04 eta: 21:05:56 time: 0.7794 data_time: 0.0020 memory: 25074 loss: 0.0410 +2023/06/01 06:13:31 - mmengine - INFO - Epoch(train) [3][4200/5758] lr: 3.9046e-04 eta: 21:04:37 time: 0.7607 data_time: 0.0013 memory: 25074 loss: 0.0409 +2023/06/01 06:14:47 - mmengine - INFO - Epoch(train) [3][4300/5758] lr: 3.9046e-04 eta: 21:03:16 time: 0.7232 data_time: 0.0014 memory: 25074 loss: 0.0485 +2023/06/01 06:16:02 - mmengine - INFO - Epoch(train) [3][4400/5758] lr: 3.9046e-04 eta: 21:01:55 time: 0.7389 data_time: 0.0016 memory: 25074 loss: 0.0453 +2023/06/01 06:17:06 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m-lr4e-4_20230601_025223 +2023/06/01 06:17:20 - mmengine - INFO - Epoch(train) [3][4500/5758] lr: 3.9046e-04 eta: 21:00:45 time: 0.8390 data_time: 0.0013 memory: 25074 loss: 0.0387 +2023/06/01 06:18:35 - mmengine - INFO - Epoch(train) [3][4600/5758] lr: 3.9046e-04 eta: 20:59:24 time: 0.7243 data_time: 0.0013 memory: 25074 loss: 0.0395 +2023/06/01 06:19:53 - mmengine - INFO - Epoch(train) [3][4700/5758] lr: 3.9046e-04 eta: 20:58:15 time: 0.7853 data_time: 0.0016 memory: 25074 loss: 0.0503 +2023/06/01 06:21:10 - mmengine - INFO - Epoch(train) [3][4800/5758] lr: 3.9046e-04 eta: 20:57:03 time: 0.7573 data_time: 0.0016 memory: 25074 loss: 0.0491 +2023/06/01 06:22:28 - mmengine - INFO - Epoch(train) [3][4900/5758] lr: 3.9046e-04 eta: 20:55:58 time: 0.8332 data_time: 0.0013 memory: 25074 loss: 0.0407 +2023/06/01 06:23:46 - mmengine - INFO - Epoch(train) [3][5000/5758] lr: 3.9046e-04 eta: 20:54:51 time: 0.7584 data_time: 0.0013 memory: 25074 loss: 0.0441 +2023/06/01 06:25:02 - mmengine - INFO - Epoch(train) [3][5100/5758] lr: 3.9046e-04 eta: 20:53:37 time: 0.7815 data_time: 0.0015 memory: 25074 loss: 0.0359 +2023/06/01 06:26:20 - mmengine - INFO - Epoch(train) [3][5200/5758] lr: 3.9046e-04 eta: 20:52:28 time: 0.7429 data_time: 0.0015 memory: 25074 loss: 0.0568 +2023/06/01 06:27:36 - mmengine - INFO - Epoch(train) [3][5300/5758] lr: 3.9046e-04 eta: 20:51:12 time: 0.8077 data_time: 0.0013 memory: 25074 loss: 0.0534 +2023/06/01 06:28:53 - mmengine - INFO - Epoch(train) [3][5400/5758] lr: 3.9046e-04 eta: 20:50:00 time: 0.7623 data_time: 0.0014 memory: 25074 loss: 0.0455 +2023/06/01 06:29:58 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m-lr4e-4_20230601_025223 +2023/06/01 06:30:10 - mmengine - INFO - Epoch(train) [3][5500/5758] lr: 3.9046e-04 eta: 20:48:46 time: 0.7359 data_time: 0.0013 memory: 25074 loss: 0.0399 +2023/06/01 06:31:27 - mmengine - INFO - Epoch(train) [3][5600/5758] lr: 3.9046e-04 eta: 20:47:33 time: 0.7623 data_time: 0.0015 memory: 25074 loss: 0.0338 +2023/06/01 06:32:42 - mmengine - INFO - Epoch(train) [3][5700/5758] lr: 3.9046e-04 eta: 20:46:07 time: 0.7115 data_time: 0.0014 memory: 25074 loss: 0.0365 +2023/06/01 06:33:25 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m-lr4e-4_20230601_025223 +2023/06/01 06:33:25 - mmengine - INFO - Saving checkpoint at 3 epochs +2023/06/01 06:33:44 - mmengine - INFO - Epoch(val) [3][16/16] accuracy/top1: 99.3624 single-label/precision_classwise: [100.0, 0.0] single-label/recall_classwise: [99.36241149902344, 0.0] single-label/f1-score_classwise: [99.68018341064453, 0.0] data_time: 0.2469 time: 0.6325 +2023/06/01 06:35:09 - mmengine - INFO - Epoch(train) [4][ 100/5758] lr: 3.7875e-04 eta: 20:44:44 time: 0.7851 data_time: 0.0387 memory: 25074 loss: 0.0398 +2023/06/01 06:36:25 - mmengine - INFO - Epoch(train) [4][ 200/5758] lr: 3.7875e-04 eta: 20:43:30 time: 0.8119 data_time: 0.0572 memory: 25074 loss: 0.0475 +2023/06/01 06:37:42 - mmengine - INFO - Epoch(train) [4][ 300/5758] lr: 3.7875e-04 eta: 20:42:13 time: 0.7516 data_time: 0.0015 memory: 25074 loss: 0.0324 +2023/06/01 06:38:58 - mmengine - INFO - Epoch(train) [4][ 400/5758] lr: 3.7875e-04 eta: 20:40:55 time: 0.7947 data_time: 0.0818 memory: 25074 loss: 0.0329 +2023/06/01 06:40:12 - mmengine - INFO - Epoch(train) [4][ 500/5758] lr: 3.7875e-04 eta: 20:39:27 time: 0.7455 data_time: 0.0014 memory: 25074 loss: 0.0316 +2023/06/01 06:41:29 - mmengine - INFO - Epoch(train) [4][ 600/5758] lr: 3.7875e-04 eta: 20:38:14 time: 0.7151 data_time: 0.0016 memory: 25074 loss: 0.0315 +2023/06/01 06:42:46 - mmengine - INFO - Epoch(train) [4][ 700/5758] lr: 3.7875e-04 eta: 20:36:59 time: 0.7612 data_time: 0.0015 memory: 25074 loss: 0.0327 +2023/06/01 06:43:06 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m-lr4e-4_20230601_025223 +2023/06/01 06:44:01 - mmengine - INFO - Epoch(train) [4][ 800/5758] lr: 3.7875e-04 eta: 20:35:39 time: 0.7357 data_time: 0.0013 memory: 25074 loss: 0.0304 +2023/06/01 06:45:17 - mmengine - INFO - Epoch(train) [4][ 900/5758] lr: 3.7875e-04 eta: 20:34:16 time: 0.7143 data_time: 0.0012 memory: 25074 loss: 0.0428 +2023/06/01 06:46:36 - mmengine - INFO - Epoch(train) [4][1000/5758] lr: 3.7875e-04 eta: 20:33:13 time: 0.8544 data_time: 0.0014 memory: 25074 loss: 0.0318 +2023/06/01 06:47:49 - mmengine - INFO - Epoch(train) [4][1100/5758] lr: 3.7875e-04 eta: 20:31:39 time: 0.7077 data_time: 0.0014 memory: 25074 loss: 0.0471 +2023/06/01 06:49:03 - mmengine - INFO - Epoch(train) [4][1200/5758] lr: 3.7875e-04 eta: 20:30:13 time: 0.7490 data_time: 0.0015 memory: 25074 loss: 0.0331 +2023/06/01 06:50:19 - mmengine - INFO - Epoch(train) [4][1300/5758] lr: 3.7875e-04 eta: 20:28:57 time: 0.7809 data_time: 0.0016 memory: 25074 loss: 0.0279 +2023/06/01 06:51:37 - mmengine - INFO - Epoch(train) [4][1400/5758] lr: 3.7875e-04 eta: 20:27:48 time: 0.7931 data_time: 0.0013 memory: 25074 loss: 0.0257 +2023/06/01 06:52:53 - mmengine - INFO - Epoch(train) [4][1500/5758] lr: 3.7875e-04 eta: 20:26:30 time: 0.8923 data_time: 0.0021 memory: 25074 loss: 0.0240 +2023/06/01 06:54:09 - mmengine - INFO - Epoch(train) [4][1600/5758] lr: 3.7875e-04 eta: 20:25:10 time: 0.7380 data_time: 0.0013 memory: 25074 loss: 0.0191 +2023/06/01 06:55:26 - mmengine - INFO - Epoch(train) [4][1700/5758] lr: 3.7875e-04 eta: 20:23:58 time: 0.7897 data_time: 0.0020 memory: 25074 loss: 0.0287 +2023/06/01 06:55:47 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m-lr4e-4_20230601_025223 +2023/06/01 06:56:43 - mmengine - INFO - Epoch(train) [4][1800/5758] lr: 3.7875e-04 eta: 20:22:42 time: 0.7566 data_time: 0.0016 memory: 25074 loss: 0.0266 +2023/06/01 06:57:59 - mmengine - INFO - Epoch(train) [4][1900/5758] lr: 3.7875e-04 eta: 20:21:26 time: 0.7687 data_time: 0.0013 memory: 25074 loss: 0.0292 +2023/06/01 06:59:16 - mmengine - INFO - Epoch(train) [4][2000/5758] lr: 3.7875e-04 eta: 20:20:14 time: 0.7645 data_time: 0.0015 memory: 25074 loss: 0.0306 +2023/06/01 07:00:32 - mmengine - INFO - Epoch(train) [4][2100/5758] lr: 3.7875e-04 eta: 20:18:57 time: 0.8134 data_time: 0.0014 memory: 25074 loss: 0.0293 +2023/06/01 07:01:48 - mmengine - INFO - Epoch(train) [4][2200/5758] lr: 3.7875e-04 eta: 20:17:35 time: 0.7550 data_time: 0.0014 memory: 25074 loss: 0.0297 +2023/06/01 07:03:04 - mmengine - INFO - Epoch(train) [4][2300/5758] lr: 3.7875e-04 eta: 20:16:20 time: 0.7044 data_time: 0.0014 memory: 25074 loss: 0.0353 +2023/06/01 07:04:20 - mmengine - INFO - Epoch(train) [4][2400/5758] lr: 3.7875e-04 eta: 20:14:59 time: 0.7185 data_time: 0.0014 memory: 25074 loss: 0.0218 +2023/06/01 07:05:34 - mmengine - INFO - Epoch(train) [4][2500/5758] lr: 3.7875e-04 eta: 20:13:33 time: 0.7235 data_time: 0.0017 memory: 25074 loss: 0.0347 +2023/06/01 07:06:49 - mmengine - INFO - Epoch(train) [4][2600/5758] lr: 3.7875e-04 eta: 20:12:09 time: 0.7608 data_time: 0.0013 memory: 25074 loss: 0.0320 +2023/06/01 07:08:06 - mmengine - INFO - Epoch(train) [4][2700/5758] lr: 3.7875e-04 eta: 20:10:58 time: 0.7250 data_time: 0.0016 memory: 25074 loss: 0.0285 +2023/06/01 07:08:26 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m-lr4e-4_20230601_025223 +2023/06/01 07:09:24 - mmengine - INFO - Epoch(train) [4][2800/5758] lr: 3.7875e-04 eta: 20:09:51 time: 0.7813 data_time: 0.0016 memory: 25074 loss: 0.0179 +2023/06/01 07:10:41 - mmengine - INFO - Epoch(train) [4][2900/5758] lr: 3.7875e-04 eta: 20:08:35 time: 0.7683 data_time: 0.0013 memory: 25074 loss: 0.0183 +2023/06/01 07:11:58 - mmengine - INFO - Epoch(train) [4][3000/5758] lr: 3.7875e-04 eta: 20:07:21 time: 0.8128 data_time: 0.0014 memory: 25074 loss: 0.0250 +2023/06/01 07:13:14 - mmengine - INFO - Epoch(train) [4][3100/5758] lr: 3.7875e-04 eta: 20:06:02 time: 0.7945 data_time: 0.0015 memory: 25074 loss: 0.0252 +2023/06/01 07:14:28 - mmengine - INFO - Epoch(train) [4][3200/5758] lr: 3.7875e-04 eta: 20:04:38 time: 0.7190 data_time: 0.0013 memory: 25074 loss: 0.0204 +2023/06/01 07:15:42 - mmengine - INFO - Epoch(train) [4][3300/5758] lr: 3.7875e-04 eta: 20:03:12 time: 0.8481 data_time: 0.0013 memory: 25074 loss: 0.0245 +2023/06/01 07:16:59 - mmengine - INFO - Epoch(train) [4][3400/5758] lr: 3.7875e-04 eta: 20:01:54 time: 0.8405 data_time: 0.0012 memory: 25074 loss: 0.0327 +2023/06/01 07:18:15 - mmengine - INFO - Epoch(train) [4][3500/5758] lr: 3.7875e-04 eta: 20:00:37 time: 0.8200 data_time: 0.0015 memory: 25074 loss: 0.0209 +2023/06/01 07:19:31 - mmengine - INFO - Epoch(train) [4][3600/5758] lr: 3.7875e-04 eta: 19:59:19 time: 0.7642 data_time: 0.0013 memory: 25074 loss: 0.0159 +2023/06/01 07:20:46 - mmengine - INFO - Epoch(train) [4][3700/5758] lr: 3.7875e-04 eta: 19:57:58 time: 0.7679 data_time: 0.0013 memory: 25074 loss: 0.0206 +2023/06/01 07:21:06 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m-lr4e-4_20230601_025223 +2023/06/01 07:22:01 - mmengine - INFO - Epoch(train) [4][3800/5758] lr: 3.7875e-04 eta: 19:56:38 time: 0.7576 data_time: 0.0013 memory: 25074 loss: 0.0178 +2023/06/01 07:23:19 - mmengine - INFO - Epoch(train) [4][3900/5758] lr: 3.7875e-04 eta: 19:55:29 time: 0.7413 data_time: 0.0013 memory: 25074 loss: 0.0279 +2023/06/01 07:24:35 - mmengine - INFO - Epoch(train) [4][4000/5758] lr: 3.7875e-04 eta: 19:54:09 time: 0.8110 data_time: 0.0016 memory: 25074 loss: 0.0188 +2023/06/01 07:25:50 - mmengine - INFO - Epoch(train) [4][4100/5758] lr: 3.7875e-04 eta: 19:52:48 time: 0.8231 data_time: 0.0025 memory: 25074 loss: 0.0184 +2023/06/01 07:27:05 - mmengine - INFO - Epoch(train) [4][4200/5758] lr: 3.7875e-04 eta: 19:51:25 time: 0.7696 data_time: 0.0015 memory: 25074 loss: 0.0222 +2023/06/01 07:28:22 - mmengine - INFO - Epoch(train) [4][4300/5758] lr: 3.7875e-04 eta: 19:50:11 time: 0.8069 data_time: 0.0016 memory: 25074 loss: 0.0289 +2023/06/01 07:29:39 - mmengine - INFO - Epoch(train) [4][4400/5758] lr: 3.7875e-04 eta: 19:48:59 time: 0.7560 data_time: 0.0016 memory: 25074 loss: 0.0210 +2023/06/01 07:30:55 - mmengine - INFO - Epoch(train) [4][4500/5758] lr: 3.7875e-04 eta: 19:47:43 time: 0.7347 data_time: 0.0016 memory: 25074 loss: 0.0169 +2023/06/01 07:32:11 - mmengine - INFO - Epoch(train) [4][4600/5758] lr: 3.7875e-04 eta: 19:46:22 time: 0.7053 data_time: 0.0014 memory: 25074 loss: 0.0155 +2023/06/01 07:33:27 - mmengine - INFO - Epoch(train) [4][4700/5758] lr: 3.7875e-04 eta: 19:45:07 time: 0.7833 data_time: 0.0014 memory: 25074 loss: 0.0194 +2023/06/01 07:33:46 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m-lr4e-4_20230601_025223 +2023/06/01 07:34:44 - mmengine - INFO - Epoch(train) [4][4800/5758] lr: 3.7875e-04 eta: 19:43:51 time: 0.8173 data_time: 0.0014 memory: 25074 loss: 0.0162 +2023/06/01 07:35:59 - mmengine - INFO - Epoch(train) [4][4900/5758] lr: 3.7875e-04 eta: 19:42:33 time: 0.7589 data_time: 0.0016 memory: 25074 loss: 0.0136 +2023/06/01 07:37:16 - mmengine - INFO - Epoch(train) [4][5000/5758] lr: 3.7875e-04 eta: 19:41:18 time: 0.8116 data_time: 0.0013 memory: 25074 loss: 0.0116 +2023/06/01 07:38:32 - mmengine - INFO - Epoch(train) [4][5100/5758] lr: 3.7875e-04 eta: 19:40:02 time: 0.7267 data_time: 0.0014 memory: 25074 loss: 0.0289 +2023/06/01 07:39:48 - mmengine - INFO - Epoch(train) [4][5200/5758] lr: 3.7875e-04 eta: 19:38:40 time: 0.7530 data_time: 0.0012 memory: 25074 loss: 0.0255 +2023/06/01 07:41:04 - mmengine - INFO - Epoch(train) [4][5300/5758] lr: 3.7875e-04 eta: 19:37:23 time: 0.7513 data_time: 0.0015 memory: 25074 loss: 0.0160 +2023/06/01 07:42:21 - mmengine - INFO - Epoch(train) [4][5400/5758] lr: 3.7875e-04 eta: 19:36:12 time: 0.8408 data_time: 0.0015 memory: 25074 loss: 0.0174 +2023/06/01 07:43:38 - mmengine - INFO - Epoch(train) [4][5500/5758] lr: 3.7875e-04 eta: 19:34:57 time: 0.7366 data_time: 0.0014 memory: 25074 loss: 0.0140 +2023/06/01 07:44:52 - mmengine - INFO - Epoch(train) [4][5600/5758] lr: 3.7875e-04 eta: 19:33:34 time: 0.7249 data_time: 0.0016 memory: 25074 loss: 0.0167 +2023/06/01 07:46:11 - mmengine - INFO - Epoch(train) [4][5700/5758] lr: 3.7875e-04 eta: 19:32:25 time: 0.7902 data_time: 0.0014 memory: 25074 loss: 0.0204 +2023/06/01 07:46:30 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m-lr4e-4_20230601_025223 +2023/06/01 07:46:55 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m-lr4e-4_20230601_025223 +2023/06/01 07:46:55 - mmengine - INFO - Saving checkpoint at 4 epochs +2023/06/01 07:47:15 - mmengine - INFO - Epoch(val) [4][16/16] accuracy/top1: 99.2677 single-label/precision_classwise: [100.0, 0.0] single-label/recall_classwise: [99.2677230834961, 0.0] single-label/f1-score_classwise: [99.63251495361328, 0.0] data_time: 0.2552 time: 0.6417 +2023/06/01 07:48:38 - mmengine - INFO - Epoch(train) [5][ 100/5758] lr: 3.6276e-04 eta: 19:30:54 time: 0.7256 data_time: 0.0015 memory: 25074 loss: 0.0130 +2023/06/01 07:49:54 - mmengine - INFO - Epoch(train) [5][ 200/5758] lr: 3.6276e-04 eta: 19:29:35 time: 0.7325 data_time: 0.0300 memory: 25074 loss: 0.0180 +2023/06/01 07:51:08 - mmengine - INFO - Epoch(train) [5][ 300/5758] lr: 3.6276e-04 eta: 19:28:11 time: 0.7051 data_time: 0.0012 memory: 25074 loss: 0.0222 +2023/06/01 07:52:24 - mmengine - INFO - Epoch(train) [5][ 400/5758] lr: 3.6276e-04 eta: 19:26:53 time: 0.7431 data_time: 0.0016 memory: 25074 loss: 0.0120 +2023/06/01 07:53:42 - mmengine - INFO - Epoch(train) [5][ 500/5758] lr: 3.6276e-04 eta: 19:25:41 time: 0.8131 data_time: 0.0013 memory: 25074 loss: 0.0139 +2023/06/01 07:54:56 - mmengine - INFO - Epoch(train) [5][ 600/5758] lr: 3.6276e-04 eta: 19:24:16 time: 0.7184 data_time: 0.0015 memory: 25074 loss: 0.0184 +2023/06/01 07:56:11 - mmengine - INFO - Epoch(train) [5][ 700/5758] lr: 3.6276e-04 eta: 19:22:55 time: 0.7740 data_time: 0.0011 memory: 25074 loss: 0.0210 +2023/06/01 07:57:28 - mmengine - INFO - Epoch(train) [5][ 800/5758] lr: 3.6276e-04 eta: 19:21:40 time: 0.7629 data_time: 0.0014 memory: 25074 loss: 0.0190 +2023/06/01 07:58:46 - mmengine - INFO - Epoch(train) [5][ 900/5758] lr: 3.6276e-04 eta: 19:20:33 time: 0.8346 data_time: 0.0013 memory: 25074 loss: 0.0123 +2023/06/01 07:59:40 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m-lr4e-4_20230601_025223 +2023/06/01 08:00:05 - mmengine - INFO - Epoch(train) [5][1000/5758] lr: 3.6276e-04 eta: 19:19:24 time: 0.8216 data_time: 0.0015 memory: 25074 loss: 0.0159 +2023/06/01 08:01:22 - mmengine - INFO - Epoch(train) [5][1100/5758] lr: 3.6276e-04 eta: 19:18:11 time: 0.7645 data_time: 0.0016 memory: 25074 loss: 0.0115 +2023/06/01 08:02:39 - mmengine - INFO - Epoch(train) [5][1200/5758] lr: 3.6276e-04 eta: 19:16:59 time: 0.7817 data_time: 0.0014 memory: 25074 loss: 0.0269 +2023/06/01 08:03:57 - mmengine - INFO - Epoch(train) [5][1300/5758] lr: 3.6276e-04 eta: 19:15:48 time: 0.7818 data_time: 0.0015 memory: 25074 loss: 0.0165 +2023/06/01 08:05:16 - mmengine - INFO - Epoch(train) [5][1400/5758] lr: 3.6276e-04 eta: 19:14:40 time: 0.7890 data_time: 0.0012 memory: 25074 loss: 0.0137 +2023/06/01 08:06:34 - mmengine - INFO - Epoch(train) [5][1500/5758] lr: 3.6276e-04 eta: 19:13:32 time: 0.7883 data_time: 0.0017 memory: 25074 loss: 0.0200 +2023/06/01 08:07:51 - mmengine - INFO - Epoch(train) [5][1600/5758] lr: 3.6276e-04 eta: 19:12:18 time: 0.7443 data_time: 0.0014 memory: 25074 loss: 0.0287 +2023/06/01 08:09:10 - mmengine - INFO - Epoch(train) [5][1700/5758] lr: 3.6276e-04 eta: 19:11:10 time: 0.7988 data_time: 0.0013 memory: 25074 loss: 0.0165 +2023/06/01 08:10:29 - mmengine - INFO - Epoch(train) [5][1800/5758] lr: 3.6276e-04 eta: 19:10:03 time: 0.8010 data_time: 0.0015 memory: 25074 loss: 0.0125 +2023/06/01 08:11:48 - mmengine - INFO - Epoch(train) [5][1900/5758] lr: 3.6276e-04 eta: 19:08:57 time: 0.7862 data_time: 0.0014 memory: 25074 loss: 0.0173 +2023/06/01 08:12:41 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m-lr4e-4_20230601_025223 +2023/06/01 08:13:05 - mmengine - INFO - Epoch(train) [5][2000/5758] lr: 3.6276e-04 eta: 19:07:42 time: 0.7393 data_time: 0.0014 memory: 25074 loss: 0.0116 +2023/06/01 08:14:23 - mmengine - INFO - Epoch(train) [5][2100/5758] lr: 3.6276e-04 eta: 19:06:30 time: 0.7982 data_time: 0.0760 memory: 25074 loss: 0.0146 +2023/06/01 08:15:40 - mmengine - INFO - Epoch(train) [5][2200/5758] lr: 3.6276e-04 eta: 19:05:15 time: 0.7996 data_time: 0.0013 memory: 25074 loss: 0.0147 +2023/06/01 08:16:58 - mmengine - INFO - Epoch(train) [5][2300/5758] lr: 3.6276e-04 eta: 19:04:04 time: 0.7735 data_time: 0.0012 memory: 25074 loss: 0.0132 +2023/06/01 08:18:16 - mmengine - INFO - Epoch(train) [5][2400/5758] lr: 3.6276e-04 eta: 19:02:54 time: 0.8141 data_time: 0.0571 memory: 25074 loss: 0.0112 +2023/06/01 08:19:33 - mmengine - INFO - Epoch(train) [5][2500/5758] lr: 3.6276e-04 eta: 19:01:39 time: 0.7679 data_time: 0.0076 memory: 25074 loss: 0.0139 +2023/06/01 08:20:51 - mmengine - INFO - Epoch(train) [5][2600/5758] lr: 3.6276e-04 eta: 19:00:30 time: 0.7801 data_time: 0.0771 memory: 25074 loss: 0.0154 +2023/06/01 08:22:10 - mmengine - INFO - Epoch(train) [5][2700/5758] lr: 3.6276e-04 eta: 18:59:23 time: 0.8040 data_time: 0.0901 memory: 25074 loss: 0.0218 +2023/06/01 08:23:31 - mmengine - INFO - Epoch(train) [5][2800/5758] lr: 3.6276e-04 eta: 18:58:20 time: 0.7368 data_time: 0.0336 memory: 25074 loss: 0.0107 +2023/06/01 08:24:47 - mmengine - INFO - Epoch(train) [5][2900/5758] lr: 3.6276e-04 eta: 18:57:03 time: 0.7857 data_time: 0.0824 memory: 25074 loss: 0.0140 +2023/06/01 08:25:41 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m-lr4e-4_20230601_025223 +2023/06/01 08:26:06 - mmengine - INFO - Epoch(train) [5][3000/5758] lr: 3.6276e-04 eta: 18:55:54 time: 0.7980 data_time: 0.0930 memory: 25074 loss: 0.0195 +2023/06/01 08:27:22 - mmengine - INFO - Epoch(train) [5][3100/5758] lr: 3.6276e-04 eta: 18:54:39 time: 0.7569 data_time: 0.0451 memory: 25074 loss: 0.0200 +2023/06/01 08:28:40 - mmengine - INFO - Epoch(train) [5][3200/5758] lr: 3.6276e-04 eta: 18:53:27 time: 0.7980 data_time: 0.0013 memory: 25074 loss: 0.0166 +2023/06/01 08:29:57 - mmengine - INFO - Epoch(train) [5][3300/5758] lr: 3.6276e-04 eta: 18:52:10 time: 0.7643 data_time: 0.0502 memory: 25074 loss: 0.0188 +2023/06/01 08:31:15 - mmengine - INFO - Epoch(train) [5][3400/5758] lr: 3.6276e-04 eta: 18:51:00 time: 0.7907 data_time: 0.0842 memory: 25074 loss: 0.0146 +2023/06/01 08:32:32 - mmengine - INFO - Epoch(train) [5][3500/5758] lr: 3.6276e-04 eta: 18:49:43 time: 0.7201 data_time: 0.0106 memory: 25074 loss: 0.0137 +2023/06/01 08:33:51 - mmengine - INFO - Epoch(train) [5][3600/5758] lr: 3.6276e-04 eta: 18:48:35 time: 0.7611 data_time: 0.0013 memory: 25074 loss: 0.0100 +2023/06/01 08:35:10 - mmengine - INFO - Epoch(train) [5][3700/5758] lr: 3.6276e-04 eta: 18:47:27 time: 0.7848 data_time: 0.0347 memory: 25074 loss: 0.0161 +2023/06/01 08:36:29 - mmengine - INFO - Epoch(train) [5][3800/5758] lr: 3.6276e-04 eta: 18:46:20 time: 0.7835 data_time: 0.0017 memory: 25074 loss: 0.0160 +2023/06/01 08:37:47 - mmengine - INFO - Epoch(train) [5][3900/5758] lr: 3.6276e-04 eta: 18:45:07 time: 0.7884 data_time: 0.0014 memory: 25074 loss: 0.0177 +2023/06/01 08:38:40 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m-lr4e-4_20230601_025223 +2023/06/01 08:39:06 - mmengine - INFO - Epoch(train) [5][4000/5758] lr: 3.6276e-04 eta: 18:43:59 time: 0.7635 data_time: 0.0013 memory: 25074 loss: 0.0127 +2023/06/01 08:40:25 - mmengine - INFO - Epoch(train) [5][4100/5758] lr: 3.6276e-04 eta: 18:42:51 time: 0.8115 data_time: 0.0013 memory: 25074 loss: 0.0102 +2023/06/01 08:41:44 - mmengine - INFO - Epoch(train) [5][4200/5758] lr: 3.6276e-04 eta: 18:41:43 time: 0.8063 data_time: 0.0013 memory: 25074 loss: 0.0117 +2023/06/01 08:43:02 - mmengine - INFO - Epoch(train) [5][4300/5758] lr: 3.6276e-04 eta: 18:40:31 time: 0.7690 data_time: 0.0012 memory: 25074 loss: 0.0113 +2023/06/01 08:44:22 - mmengine - INFO - Epoch(train) [5][4400/5758] lr: 3.6276e-04 eta: 18:39:26 time: 0.8084 data_time: 0.0013 memory: 25074 loss: 0.0116 +2023/06/01 08:45:40 - mmengine - INFO - Epoch(train) [5][4500/5758] lr: 3.6276e-04 eta: 18:38:15 time: 0.7783 data_time: 0.0014 memory: 25074 loss: 0.0118 +2023/06/01 08:46:59 - mmengine - INFO - Epoch(train) [5][4600/5758] lr: 3.6276e-04 eta: 18:37:03 time: 0.7517 data_time: 0.0014 memory: 25074 loss: 0.0138 +2023/06/01 08:48:17 - mmengine - INFO - Epoch(train) [5][4700/5758] lr: 3.6276e-04 eta: 18:35:51 time: 0.8466 data_time: 0.0033 memory: 25074 loss: 0.0074 +2023/06/01 08:49:36 - mmengine - INFO - Epoch(train) [5][4800/5758] lr: 3.6276e-04 eta: 18:34:45 time: 0.8229 data_time: 0.0014 memory: 25074 loss: 0.0186 +2023/06/01 08:50:55 - mmengine - INFO - Epoch(train) [5][4900/5758] lr: 3.6276e-04 eta: 18:33:36 time: 0.8427 data_time: 0.0012 memory: 25074 loss: 0.0084 +2023/06/01 08:51:49 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m-lr4e-4_20230601_025223 +2023/06/01 08:52:15 - mmengine - INFO - Epoch(train) [5][5000/5758] lr: 3.6276e-04 eta: 18:32:28 time: 0.7991 data_time: 0.0014 memory: 25074 loss: 0.0097 +2023/06/01 08:53:34 - mmengine - INFO - Epoch(train) [5][5100/5758] lr: 3.6276e-04 eta: 18:31:19 time: 0.8169 data_time: 0.0014 memory: 25074 loss: 0.0104 +2023/06/01 08:54:52 - mmengine - INFO - Epoch(train) [5][5200/5758] lr: 3.6276e-04 eta: 18:30:08 time: 0.8135 data_time: 0.0012 memory: 25074 loss: 0.0090 +2023/06/01 08:56:13 - mmengine - INFO - Epoch(train) [5][5300/5758] lr: 3.6276e-04 eta: 18:29:05 time: 0.8370 data_time: 0.0013 memory: 25074 loss: 0.0217 +2023/06/01 08:57:33 - mmengine - INFO - Epoch(train) [5][5400/5758] lr: 3.6276e-04 eta: 18:27:56 time: 0.7648 data_time: 0.0014 memory: 25074 loss: 0.0124 +2023/06/01 08:58:51 - mmengine - INFO - Epoch(train) [5][5500/5758] lr: 3.6276e-04 eta: 18:26:45 time: 0.8361 data_time: 0.0015 memory: 25074 loss: 0.0114 +2023/06/01 09:00:12 - mmengine - INFO - Epoch(train) [5][5600/5758] lr: 3.6276e-04 eta: 18:25:41 time: 0.8075 data_time: 0.0013 memory: 25074 loss: 0.0079 +2023/06/01 09:01:30 - mmengine - INFO - Epoch(train) [5][5700/5758] lr: 3.6276e-04 eta: 18:24:29 time: 0.8034 data_time: 0.0014 memory: 25074 loss: 0.0182 +2023/06/01 09:02:14 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m-lr4e-4_20230601_025223 +2023/06/01 09:02:14 - mmengine - INFO - Saving checkpoint at 5 epochs +2023/06/01 09:02:33 - mmengine - INFO - Epoch(val) [5][16/16] accuracy/top1: 99.4445 single-label/precision_classwise: [100.0, 0.0] single-label/recall_classwise: [99.4444808959961, 0.0] single-label/f1-score_classwise: [99.72146606445312, 0.0] data_time: 0.2366 time: 0.6223 +2023/06/01 09:04:01 - mmengine - INFO - Epoch(train) [6][ 100/5758] lr: 3.4289e-04 eta: 18:23:00 time: 0.8483 data_time: 0.0012 memory: 25074 loss: 0.0147 +2023/06/01 09:05:19 - mmengine - INFO - Epoch(train) [6][ 200/5758] lr: 3.4289e-04 eta: 18:21:49 time: 0.7660 data_time: 0.0014 memory: 25074 loss: 0.0131 +2023/06/01 09:05:27 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m-lr4e-4_20230601_025223 +2023/06/01 09:06:40 - mmengine - INFO - Epoch(train) [6][ 300/5758] lr: 3.4289e-04 eta: 18:20:43 time: 0.7825 data_time: 0.0015 memory: 25074 loss: 0.0064 +2023/06/01 09:07:59 - mmengine - INFO - Epoch(train) [6][ 400/5758] lr: 3.4289e-04 eta: 18:19:34 time: 0.7980 data_time: 0.0017 memory: 25074 loss: 0.0081 +2023/06/01 09:09:19 - mmengine - INFO - Epoch(train) [6][ 500/5758] lr: 3.4289e-04 eta: 18:18:26 time: 0.8616 data_time: 0.0015 memory: 25074 loss: 0.0135 +2023/06/01 09:10:39 - mmengine - INFO - Epoch(train) [6][ 600/5758] lr: 3.4289e-04 eta: 18:17:19 time: 0.7762 data_time: 0.0015 memory: 25074 loss: 0.0168 +2023/06/01 09:11:57 - mmengine - INFO - Epoch(train) [6][ 700/5758] lr: 3.4289e-04 eta: 18:16:05 time: 0.7673 data_time: 0.0013 memory: 25074 loss: 0.0144 +2023/06/01 09:13:50 - mmengine - INFO - Epoch(train) [6][ 800/5758] lr: 3.4289e-04 eta: 18:16:35 time: 1.3448 data_time: 0.0016 memory: 25074 loss: 0.0112 +2023/06/01 09:15:09 - mmengine - INFO - Epoch(train) [6][ 900/5758] lr: 3.4289e-04 eta: 18:15:22 time: 0.7338 data_time: 0.0013 memory: 25074 loss: 0.0133 +2023/06/01 09:16:27 - mmengine - INFO - Epoch(train) [6][1000/5758] lr: 3.4289e-04 eta: 18:14:10 time: 0.7987 data_time: 0.0016 memory: 25074 loss: 0.0086 +2023/06/01 09:17:47 - mmengine - INFO - Epoch(train) [6][1100/5758] lr: 3.4289e-04 eta: 18:13:01 time: 0.7991 data_time: 0.0014 memory: 25074 loss: 0.0103 +2023/06/01 09:19:06 - mmengine - INFO - Epoch(train) [6][1200/5758] lr: 3.4289e-04 eta: 18:11:51 time: 0.7866 data_time: 0.0014 memory: 25074 loss: 0.0070 +2023/06/01 09:19:14 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m-lr4e-4_20230601_025223 +2023/06/01 09:20:26 - mmengine - INFO - Epoch(train) [6][1300/5758] lr: 3.4289e-04 eta: 18:10:43 time: 0.7602 data_time: 0.0521 memory: 25074 loss: 0.0136 +2023/06/01 09:21:43 - mmengine - INFO - Epoch(train) [6][1400/5758] lr: 3.4289e-04 eta: 18:09:27 time: 0.7739 data_time: 0.0592 memory: 25074 loss: 0.0082 +2023/06/01 09:23:03 - mmengine - INFO - Epoch(train) [6][1500/5758] lr: 3.4289e-04 eta: 18:08:17 time: 0.7616 data_time: 0.0013 memory: 25074 loss: 0.0085 +2023/06/01 09:24:21 - mmengine - INFO - Epoch(train) [6][1600/5758] lr: 3.4289e-04 eta: 18:07:04 time: 0.7735 data_time: 0.0014 memory: 25074 loss: 0.0073 +2023/06/01 09:25:39 - mmengine - INFO - Epoch(train) [6][1700/5758] lr: 3.4289e-04 eta: 18:05:49 time: 0.8029 data_time: 0.0014 memory: 25074 loss: 0.0086 +2023/06/01 09:26:59 - mmengine - INFO - Epoch(train) [6][1800/5758] lr: 3.4289e-04 eta: 18:04:41 time: 0.8209 data_time: 0.0015 memory: 25074 loss: 0.0105 +2023/06/01 09:28:19 - mmengine - INFO - Epoch(train) [6][1900/5758] lr: 3.4289e-04 eta: 18:03:31 time: 0.8046 data_time: 0.0017 memory: 25074 loss: 0.0057 +2023/06/01 09:29:36 - mmengine - INFO - Epoch(train) [6][2000/5758] lr: 3.4289e-04 eta: 18:02:16 time: 0.7910 data_time: 0.0014 memory: 25074 loss: 0.0104 +2023/06/01 09:30:55 - mmengine - INFO - Epoch(train) [6][2100/5758] lr: 3.4289e-04 eta: 18:01:04 time: 0.7808 data_time: 0.0014 memory: 25074 loss: 0.0125 +2023/06/01 09:32:12 - mmengine - INFO - Epoch(train) [6][2200/5758] lr: 3.4289e-04 eta: 17:59:48 time: 0.8098 data_time: 0.0015 memory: 25074 loss: 0.0093 +2023/06/01 09:32:20 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m-lr4e-4_20230601_025223 +2023/06/01 09:33:46 - mmengine - INFO - Epoch(train) [6][2300/5758] lr: 3.4289e-04 eta: 17:59:15 time: 0.7603 data_time: 0.0015 memory: 25074 loss: 0.0093 +2023/06/01 09:35:03 - mmengine - INFO - Epoch(train) [6][2400/5758] lr: 3.4289e-04 eta: 17:58:00 time: 0.7537 data_time: 0.0013 memory: 25074 loss: 0.0072 +2023/06/01 09:36:21 - mmengine - INFO - Epoch(train) [6][2500/5758] lr: 3.4289e-04 eta: 17:56:44 time: 0.7763 data_time: 0.0013 memory: 25074 loss: 0.0117 +2023/06/01 09:37:40 - mmengine - INFO - Epoch(train) [6][2600/5758] lr: 3.4289e-04 eta: 17:55:32 time: 0.7822 data_time: 0.0011 memory: 25074 loss: 0.0084 +2023/06/01 09:38:56 - mmengine - INFO - Epoch(train) [6][2700/5758] lr: 3.4289e-04 eta: 17:54:14 time: 0.7625 data_time: 0.0013 memory: 25074 loss: 0.0118 +2023/06/01 09:40:16 - mmengine - INFO - Epoch(train) [6][2800/5758] lr: 3.4289e-04 eta: 17:53:04 time: 0.8257 data_time: 0.0013 memory: 25074 loss: 0.0087 +2023/06/01 09:41:35 - mmengine - INFO - Epoch(train) [6][2900/5758] lr: 3.4289e-04 eta: 17:51:51 time: 0.8246 data_time: 0.0011 memory: 25074 loss: 0.0075 +2023/06/01 09:42:52 - mmengine - INFO - Epoch(train) [6][3000/5758] lr: 3.4289e-04 eta: 17:50:35 time: 0.8015 data_time: 0.0014 memory: 25074 loss: 0.0089 +2023/06/01 09:44:12 - mmengine - INFO - Epoch(train) [6][3100/5758] lr: 3.4289e-04 eta: 17:49:25 time: 0.7954 data_time: 0.0015 memory: 25074 loss: 0.0082 +2023/06/01 09:45:32 - mmengine - INFO - Epoch(train) [6][3200/5758] lr: 3.4289e-04 eta: 17:48:17 time: 0.7549 data_time: 0.0012 memory: 25074 loss: 0.0075 +2023/06/01 09:45:41 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m-lr4e-4_20230601_025223 +2023/06/01 09:46:50 - mmengine - INFO - Epoch(train) [6][3300/5758] lr: 3.4289e-04 eta: 17:47:02 time: 0.8027 data_time: 0.0014 memory: 25074 loss: 0.0051 +2023/06/01 09:48:10 - mmengine - INFO - Epoch(train) [6][3400/5758] lr: 3.4289e-04 eta: 17:45:51 time: 0.8169 data_time: 0.0013 memory: 25074 loss: 0.0134 +2023/06/01 09:49:29 - mmengine - INFO - Epoch(train) [6][3500/5758] lr: 3.4289e-04 eta: 17:44:39 time: 0.8337 data_time: 0.0013 memory: 25074 loss: 0.0092 +2023/06/01 09:50:49 - mmengine - INFO - Epoch(train) [6][3600/5758] lr: 3.4289e-04 eta: 17:43:29 time: 0.7926 data_time: 0.0017 memory: 25074 loss: 0.0108 +2023/06/01 09:52:07 - mmengine - INFO - Epoch(train) [6][3700/5758] lr: 3.4289e-04 eta: 17:42:14 time: 0.7743 data_time: 0.0016 memory: 25074 loss: 0.0069 +2023/06/01 09:53:24 - mmengine - INFO - Epoch(train) [6][3800/5758] lr: 3.4289e-04 eta: 17:40:59 time: 0.7711 data_time: 0.0013 memory: 25074 loss: 0.0087 +2023/06/01 09:54:42 - mmengine - INFO - Epoch(train) [6][3900/5758] lr: 3.4289e-04 eta: 17:39:43 time: 0.7581 data_time: 0.0012 memory: 25074 loss: 0.0042 +2023/06/01 09:56:01 - mmengine - INFO - Epoch(train) [6][4000/5758] lr: 3.4289e-04 eta: 17:38:31 time: 0.7414 data_time: 0.0012 memory: 25074 loss: 0.0148 +2023/06/01 09:57:21 - mmengine - INFO - Epoch(train) [6][4100/5758] lr: 3.4289e-04 eta: 17:37:20 time: 0.7420 data_time: 0.0015 memory: 25074 loss: 0.0107 +2023/06/01 09:58:40 - mmengine - INFO - Epoch(train) [6][4200/5758] lr: 3.4289e-04 eta: 17:36:09 time: 0.8104 data_time: 0.0015 memory: 25074 loss: 0.0048 +2023/06/01 09:58:49 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m-lr4e-4_20230601_025223 +2023/06/01 10:00:00 - mmengine - INFO - Epoch(train) [6][4300/5758] lr: 3.4289e-04 eta: 17:34:59 time: 0.7592 data_time: 0.0012 memory: 25074 loss: 0.0131 +2023/06/01 10:01:20 - mmengine - INFO - Epoch(train) [6][4400/5758] lr: 3.4289e-04 eta: 17:33:48 time: 0.7964 data_time: 0.0013 memory: 25074 loss: 0.0086 +2023/06/01 10:02:38 - mmengine - INFO - Epoch(train) [6][4500/5758] lr: 3.4289e-04 eta: 17:32:33 time: 0.7747 data_time: 0.0589 memory: 25074 loss: 0.0130 +2023/06/01 10:03:57 - mmengine - INFO - Epoch(train) [6][4600/5758] lr: 3.4289e-04 eta: 17:31:21 time: 0.7957 data_time: 0.0031 memory: 25074 loss: 0.0125 +2023/06/01 10:05:17 - mmengine - INFO - Epoch(train) [6][4700/5758] lr: 3.4289e-04 eta: 17:30:09 time: 0.7807 data_time: 0.0765 memory: 25074 loss: 0.0061 +2023/06/01 10:06:37 - mmengine - INFO - Epoch(train) [6][4800/5758] lr: 3.4289e-04 eta: 17:29:00 time: 0.8368 data_time: 0.0555 memory: 25074 loss: 0.0106 +2023/06/01 10:07:56 - mmengine - INFO - Epoch(train) [6][4900/5758] lr: 3.4289e-04 eta: 17:27:47 time: 0.7715 data_time: 0.0018 memory: 25074 loss: 0.0095 +2023/06/01 10:09:15 - mmengine - INFO - Epoch(train) [6][5000/5758] lr: 3.4289e-04 eta: 17:26:34 time: 0.8249 data_time: 0.0265 memory: 25074 loss: 0.0177 +2023/06/01 10:10:34 - mmengine - INFO - Epoch(train) [6][5100/5758] lr: 3.4289e-04 eta: 17:25:21 time: 0.7611 data_time: 0.0569 memory: 25074 loss: 0.0058 +2023/06/01 10:11:52 - mmengine - INFO - Epoch(train) [6][5200/5758] lr: 3.4289e-04 eta: 17:24:07 time: 0.8272 data_time: 0.1221 memory: 25074 loss: 0.0086 +2023/06/01 10:12:00 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m-lr4e-4_20230601_025223 +2023/06/01 10:13:11 - mmengine - INFO - Epoch(train) [6][5300/5758] lr: 3.4289e-04 eta: 17:22:54 time: 0.7965 data_time: 0.0487 memory: 25074 loss: 0.0053 +2023/06/01 10:14:29 - mmengine - INFO - Epoch(train) [6][5400/5758] lr: 3.4289e-04 eta: 17:21:39 time: 0.8365 data_time: 0.1121 memory: 25074 loss: 0.0125 +2023/06/01 10:15:49 - mmengine - INFO - Epoch(train) [6][5500/5758] lr: 3.4289e-04 eta: 17:20:29 time: 0.7568 data_time: 0.0017 memory: 25074 loss: 0.0122 +2023/06/01 10:17:10 - mmengine - INFO - Epoch(train) [6][5600/5758] lr: 3.4289e-04 eta: 17:19:19 time: 0.8497 data_time: 0.0012 memory: 25074 loss: 0.0104 +2023/06/01 10:18:32 - mmengine - INFO - Epoch(train) [6][5700/5758] lr: 3.4289e-04 eta: 17:18:12 time: 0.8505 data_time: 0.0014 memory: 25074 loss: 0.0135 +2023/06/01 10:19:19 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m-lr4e-4_20230601_025223 +2023/06/01 10:19:19 - mmengine - INFO - Saving checkpoint at 6 epochs +2023/06/01 10:19:37 - mmengine - INFO - Epoch(val) [6][16/16] accuracy/top1: 99.7601 single-label/precision_classwise: [100.0, 0.0] single-label/recall_classwise: [99.76011657714844, 0.0] single-label/f1-score_classwise: [99.87991333007812, 0.0] data_time: 0.2468 time: 0.6383 +2023/06/01 10:21:06 - mmengine - INFO - Epoch(train) [7][ 100/5758] lr: 3.1962e-04 eta: 17:16:41 time: 0.7950 data_time: 0.0014 memory: 25074 loss: 0.0066 +2023/06/01 10:22:25 - mmengine - INFO - Epoch(train) [7][ 200/5758] lr: 3.1962e-04 eta: 17:15:28 time: 0.7778 data_time: 0.0012 memory: 25074 loss: 0.0059 +2023/06/01 10:23:44 - mmengine - INFO - Epoch(train) [7][ 300/5758] lr: 3.1962e-04 eta: 17:14:16 time: 0.8078 data_time: 0.0014 memory: 25074 loss: 0.0058 +2023/06/01 10:25:05 - mmengine - INFO - Epoch(train) [7][ 400/5758] lr: 3.1962e-04 eta: 17:13:06 time: 0.8209 data_time: 0.0014 memory: 25074 loss: 0.0102 +2023/06/01 10:25:45 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m-lr4e-4_20230601_025223 +2023/06/01 10:26:22 - mmengine - INFO - Epoch(train) [7][ 500/5758] lr: 3.1962e-04 eta: 17:11:49 time: 0.7721 data_time: 0.0014 memory: 25074 loss: 0.0055 +2023/06/01 10:27:42 - mmengine - INFO - Epoch(train) [7][ 600/5758] lr: 3.1962e-04 eta: 17:10:36 time: 0.7303 data_time: 0.0015 memory: 25074 loss: 0.0074 +2023/06/01 10:29:02 - mmengine - INFO - Epoch(train) [7][ 700/5758] lr: 3.1962e-04 eta: 17:09:25 time: 0.8308 data_time: 0.0011 memory: 25074 loss: 0.0085 +2023/06/01 10:30:23 - mmengine - INFO - Epoch(train) [7][ 800/5758] lr: 3.1962e-04 eta: 17:08:17 time: 0.8447 data_time: 0.0015 memory: 25074 loss: 0.0074 +2023/06/01 10:31:42 - mmengine - INFO - Epoch(train) [7][ 900/5758] lr: 3.1962e-04 eta: 17:07:05 time: 0.7620 data_time: 0.0014 memory: 25074 loss: 0.0020 +2023/06/01 10:33:02 - mmengine - INFO - Epoch(train) [7][1000/5758] lr: 3.1962e-04 eta: 17:05:52 time: 0.7677 data_time: 0.0014 memory: 25074 loss: 0.0069 +2023/06/01 10:34:24 - mmengine - INFO - Epoch(train) [7][1100/5758] lr: 3.1962e-04 eta: 17:04:45 time: 0.8134 data_time: 0.0015 memory: 25074 loss: 0.0059 +2023/06/01 10:35:45 - mmengine - INFO - Epoch(train) [7][1200/5758] lr: 3.1962e-04 eta: 17:03:37 time: 0.8411 data_time: 0.0017 memory: 25074 loss: 0.0079 +2023/06/01 10:37:04 - mmengine - INFO - Epoch(train) [7][1300/5758] lr: 3.1962e-04 eta: 17:02:24 time: 0.7843 data_time: 0.0014 memory: 25074 loss: 0.0067 +2023/06/01 10:38:26 - mmengine - INFO - Epoch(train) [7][1400/5758] lr: 3.1962e-04 eta: 17:01:15 time: 0.7981 data_time: 0.0015 memory: 25074 loss: 0.0134 +2023/06/01 10:39:08 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m-lr4e-4_20230601_025223 +2023/06/01 10:39:46 - mmengine - INFO - Epoch(train) [7][1500/5758] lr: 3.1962e-04 eta: 17:00:04 time: 0.7964 data_time: 0.0013 memory: 25074 loss: 0.0093 +2023/06/01 10:41:06 - mmengine - INFO - Epoch(train) [7][1600/5758] lr: 3.1962e-04 eta: 16:58:52 time: 0.7851 data_time: 0.0011 memory: 25074 loss: 0.0061 +2023/06/01 10:42:26 - mmengine - INFO - Epoch(train) [7][1700/5758] lr: 3.1962e-04 eta: 16:57:41 time: 0.7795 data_time: 0.0013 memory: 25074 loss: 0.0084 +2023/06/01 10:43:45 - mmengine - INFO - Epoch(train) [7][1800/5758] lr: 3.1962e-04 eta: 16:56:27 time: 0.8162 data_time: 0.0014 memory: 25074 loss: 0.0073 +2023/06/01 10:45:04 - mmengine - INFO - Epoch(train) [7][1900/5758] lr: 3.1962e-04 eta: 16:55:13 time: 0.7840 data_time: 0.0014 memory: 25074 loss: 0.0131 +2023/06/01 10:46:23 - mmengine - INFO - Epoch(train) [7][2000/5758] lr: 3.1962e-04 eta: 16:53:58 time: 0.7779 data_time: 0.0013 memory: 25074 loss: 0.0076 +2023/06/01 10:47:42 - mmengine - INFO - Epoch(train) [7][2100/5758] lr: 3.1962e-04 eta: 16:52:46 time: 0.7800 data_time: 0.0014 memory: 25074 loss: 0.0059 +2023/06/01 10:49:04 - mmengine - INFO - Epoch(train) [7][2200/5758] lr: 3.1962e-04 eta: 16:51:37 time: 0.8129 data_time: 0.0014 memory: 25074 loss: 0.0108 +2023/06/01 10:50:25 - mmengine - INFO - Epoch(train) [7][2300/5758] lr: 3.1962e-04 eta: 16:50:27 time: 0.7828 data_time: 0.0012 memory: 25074 loss: 0.0060 +2023/06/01 10:51:44 - mmengine - INFO - Epoch(train) [7][2400/5758] lr: 3.1962e-04 eta: 16:49:14 time: 0.7703 data_time: 0.0011 memory: 25074 loss: 0.0060 +2023/06/01 10:52:27 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m-lr4e-4_20230601_025223 +2023/06/01 10:53:06 - mmengine - INFO - Epoch(train) [7][2500/5758] lr: 3.1962e-04 eta: 16:48:06 time: 0.8260 data_time: 0.0015 memory: 25074 loss: 0.0099 +2023/06/01 10:54:30 - mmengine - INFO - Epoch(train) [7][2600/5758] lr: 3.1962e-04 eta: 16:47:02 time: 0.9324 data_time: 0.0013 memory: 25074 loss: 0.0091 +2023/06/01 10:55:48 - mmengine - INFO - Epoch(train) [7][2700/5758] lr: 3.1962e-04 eta: 16:45:46 time: 0.8190 data_time: 0.0015 memory: 25074 loss: 0.0088 +2023/06/01 10:57:08 - mmengine - INFO - Epoch(train) [7][2800/5758] lr: 3.1962e-04 eta: 16:44:33 time: 0.7778 data_time: 0.0012 memory: 25074 loss: 0.0059 +2023/06/01 10:58:30 - mmengine - INFO - Epoch(train) [7][2900/5758] lr: 3.1962e-04 eta: 16:43:27 time: 0.8132 data_time: 0.0013 memory: 25074 loss: 0.0077 +2023/06/01 10:59:53 - mmengine - INFO - Epoch(train) [7][3000/5758] lr: 3.1962e-04 eta: 16:42:20 time: 0.8165 data_time: 0.0012 memory: 25074 loss: 0.0085 +2023/06/01 11:01:16 - mmengine - INFO - Epoch(train) [7][3100/5758] lr: 3.1962e-04 eta: 16:41:12 time: 0.8407 data_time: 0.0013 memory: 25074 loss: 0.0079 +2023/06/01 11:02:34 - mmengine - INFO - Epoch(train) [7][3200/5758] lr: 3.1962e-04 eta: 16:39:57 time: 0.7709 data_time: 0.0015 memory: 25074 loss: 0.0118 +2023/06/01 11:03:53 - mmengine - INFO - Epoch(train) [7][3300/5758] lr: 3.1962e-04 eta: 16:38:43 time: 0.8400 data_time: 0.0016 memory: 25074 loss: 0.0052 +2023/06/01 11:05:15 - mmengine - INFO - Epoch(train) [7][3400/5758] lr: 3.1962e-04 eta: 16:37:33 time: 0.7972 data_time: 0.0012 memory: 25074 loss: 0.0075 +2023/06/01 11:05:56 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m-lr4e-4_20230601_025223 +2023/06/01 11:06:35 - mmengine - INFO - Epoch(train) [7][3500/5758] lr: 3.1962e-04 eta: 16:36:21 time: 0.8534 data_time: 0.0014 memory: 25074 loss: 0.0136 +2023/06/01 11:07:57 - mmengine - INFO - Epoch(train) [7][3600/5758] lr: 3.1962e-04 eta: 16:35:13 time: 0.8338 data_time: 0.0014 memory: 25074 loss: 0.0069 +2023/06/01 11:09:17 - mmengine - INFO - Epoch(train) [7][3700/5758] lr: 3.1962e-04 eta: 16:34:01 time: 0.7883 data_time: 0.0011 memory: 25074 loss: 0.0074 +2023/06/01 11:10:38 - mmengine - INFO - Epoch(train) [7][3800/5758] lr: 3.1962e-04 eta: 16:32:50 time: 0.8009 data_time: 0.0012 memory: 25074 loss: 0.0036 +2023/06/01 11:12:01 - mmengine - INFO - Epoch(train) [7][3900/5758] lr: 3.1962e-04 eta: 16:31:42 time: 0.7946 data_time: 0.0014 memory: 25074 loss: 0.0042 +2023/06/01 11:13:23 - mmengine - INFO - Epoch(train) [7][4000/5758] lr: 3.1962e-04 eta: 16:30:35 time: 0.8424 data_time: 0.0011 memory: 25074 loss: 0.0034 +2023/06/01 11:14:50 - mmengine - INFO - Epoch(train) [7][4100/5758] lr: 3.1962e-04 eta: 16:29:36 time: 0.8726 data_time: 0.0012 memory: 25074 loss: 0.0052 +2023/06/01 11:16:13 - mmengine - INFO - Epoch(train) [7][4200/5758] lr: 3.1962e-04 eta: 16:28:29 time: 0.8247 data_time: 0.0013 memory: 25074 loss: 0.0097 +2023/06/01 11:17:36 - mmengine - INFO - Epoch(train) [7][4300/5758] lr: 3.1962e-04 eta: 16:27:21 time: 0.8297 data_time: 0.0013 memory: 25074 loss: 0.0065 +2023/06/01 11:18:55 - mmengine - INFO - Epoch(train) [7][4400/5758] lr: 3.1962e-04 eta: 16:26:07 time: 0.8276 data_time: 0.0014 memory: 25074 loss: 0.0039 +2023/06/01 11:19:38 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m-lr4e-4_20230601_025223 +2023/06/01 11:20:16 - mmengine - INFO - Epoch(train) [7][4500/5758] lr: 3.1962e-04 eta: 16:24:56 time: 0.8222 data_time: 0.0017 memory: 25074 loss: 0.0031 +2023/06/01 11:21:38 - mmengine - INFO - Epoch(train) [7][4600/5758] lr: 3.1962e-04 eta: 16:23:47 time: 0.8391 data_time: 0.0016 memory: 25074 loss: 0.0092 +2023/06/01 11:23:00 - mmengine - INFO - Epoch(train) [7][4700/5758] lr: 3.1962e-04 eta: 16:22:37 time: 0.8018 data_time: 0.0012 memory: 25074 loss: 0.0083 +2023/06/01 11:24:23 - mmengine - INFO - Epoch(train) [7][4800/5758] lr: 3.1962e-04 eta: 16:21:29 time: 0.8394 data_time: 0.0014 memory: 25074 loss: 0.0026 +2023/06/01 11:25:44 - mmengine - INFO - Epoch(train) [7][4900/5758] lr: 3.1962e-04 eta: 16:20:18 time: 0.7875 data_time: 0.0013 memory: 25074 loss: 0.0031 +2023/06/01 11:27:05 - mmengine - INFO - Epoch(train) [7][5000/5758] lr: 3.1962e-04 eta: 16:19:06 time: 0.8163 data_time: 0.0015 memory: 25074 loss: 0.0071 +2023/06/01 11:28:25 - mmengine - INFO - Epoch(train) [7][5100/5758] lr: 3.1962e-04 eta: 16:17:54 time: 0.7525 data_time: 0.0014 memory: 25074 loss: 0.0053 +2023/06/01 11:29:45 - mmengine - INFO - Epoch(train) [7][5200/5758] lr: 3.1962e-04 eta: 16:16:40 time: 0.8108 data_time: 0.0014 memory: 25074 loss: 0.0089 +2023/06/01 11:31:06 - mmengine - INFO - Epoch(train) [7][5300/5758] lr: 3.1962e-04 eta: 16:15:28 time: 0.7976 data_time: 0.0014 memory: 25074 loss: 0.0081 +2023/06/01 11:32:27 - mmengine - INFO - Epoch(train) [7][5400/5758] lr: 3.1962e-04 eta: 16:14:17 time: 0.7771 data_time: 0.0013 memory: 25074 loss: 0.0042 +2023/06/01 11:33:10 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m-lr4e-4_20230601_025223 +2023/06/01 11:33:50 - mmengine - INFO - Epoch(train) [7][5500/5758] lr: 3.1962e-04 eta: 16:13:09 time: 0.8004 data_time: 0.0013 memory: 25074 loss: 0.0068 +2023/06/01 11:35:12 - mmengine - INFO - Epoch(train) [7][5600/5758] lr: 3.1962e-04 eta: 16:11:59 time: 0.7888 data_time: 0.0017 memory: 25074 loss: 0.0011 +2023/06/01 11:36:35 - mmengine - INFO - Epoch(train) [7][5700/5758] lr: 3.1962e-04 eta: 16:10:51 time: 0.8345 data_time: 0.0013 memory: 25074 loss: 0.0052 +2023/06/01 11:37:21 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m-lr4e-4_20230601_025223 +2023/06/01 11:37:21 - mmengine - INFO - Saving checkpoint at 7 epochs +2023/06/01 11:37:39 - mmengine - INFO - Epoch(val) [7][16/16] accuracy/top1: 99.8927 single-label/precision_classwise: [100.0, 0.0] single-label/recall_classwise: [99.89268493652344, 0.0] single-label/f1-score_classwise: [99.9463119506836, 0.0] data_time: 0.2596 time: 0.6451 +2023/06/01 11:39:11 - mmengine - INFO - Epoch(train) [8][ 100/5758] lr: 2.9353e-04 eta: 16:09:15 time: 0.8854 data_time: 0.0817 memory: 25074 loss: 0.0045 +2023/06/01 11:40:34 - mmengine - INFO - Epoch(train) [8][ 200/5758] lr: 2.9353e-04 eta: 16:08:07 time: 0.7808 data_time: 0.0013 memory: 25074 loss: 0.0086 +2023/06/01 11:41:56 - mmengine - INFO - Epoch(train) [8][ 300/5758] lr: 2.9353e-04 eta: 16:06:56 time: 0.8090 data_time: 0.0012 memory: 25074 loss: 0.0075 +2023/06/01 11:43:13 - mmengine - INFO - Epoch(train) [8][ 400/5758] lr: 2.9353e-04 eta: 16:05:38 time: 0.7352 data_time: 0.0013 memory: 25074 loss: 0.0061 +2023/06/01 11:44:31 - mmengine - INFO - Epoch(train) [8][ 500/5758] lr: 2.9353e-04 eta: 16:04:19 time: 0.7547 data_time: 0.0015 memory: 25074 loss: 0.0053 +2023/06/01 11:45:49 - mmengine - INFO - Epoch(train) [8][ 600/5758] lr: 2.9353e-04 eta: 16:03:03 time: 0.7743 data_time: 0.0016 memory: 25074 loss: 0.0062 +2023/06/01 11:47:02 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m-lr4e-4_20230601_025223 +2023/06/01 11:47:06 - mmengine - INFO - Epoch(train) [8][ 700/5758] lr: 2.9353e-04 eta: 16:01:43 time: 0.7527 data_time: 0.0016 memory: 25074 loss: 0.0047 +2023/06/01 11:48:21 - mmengine - INFO - Epoch(train) [8][ 800/5758] lr: 2.9353e-04 eta: 16:00:19 time: 0.7300 data_time: 0.0015 memory: 25074 loss: 0.0073 +2023/06/01 11:49:37 - mmengine - INFO - Epoch(train) [8][ 900/5758] lr: 2.9353e-04 eta: 15:58:59 time: 0.7418 data_time: 0.0016 memory: 25074 loss: 0.0056 +2023/06/01 11:50:53 - mmengine - INFO - Epoch(train) [8][1000/5758] lr: 2.9353e-04 eta: 15:57:38 time: 0.7793 data_time: 0.0013 memory: 25074 loss: 0.0064 +2023/06/01 11:52:10 - mmengine - INFO - Epoch(train) [8][1100/5758] lr: 2.9353e-04 eta: 15:56:19 time: 0.8202 data_time: 0.0015 memory: 25074 loss: 0.0044 +2023/06/01 11:53:28 - mmengine - INFO - Epoch(train) [8][1200/5758] lr: 2.9353e-04 eta: 15:55:02 time: 0.8394 data_time: 0.0014 memory: 25074 loss: 0.0040 +2023/06/01 11:54:45 - mmengine - INFO - Epoch(train) [8][1300/5758] lr: 2.9353e-04 eta: 15:53:42 time: 0.7523 data_time: 0.0016 memory: 25074 loss: 0.0036 +2023/06/01 11:56:00 - mmengine - INFO - Epoch(train) [8][1400/5758] lr: 2.9353e-04 eta: 15:52:20 time: 0.7828 data_time: 0.0017 memory: 25074 loss: 0.0090 +2023/06/01 11:57:17 - mmengine - INFO - Epoch(train) [8][1500/5758] lr: 2.9353e-04 eta: 15:51:00 time: 0.7775 data_time: 0.0016 memory: 25074 loss: 0.0035 +2023/06/01 11:58:34 - mmengine - INFO - Epoch(train) [8][1600/5758] lr: 2.9353e-04 eta: 15:49:41 time: 0.8056 data_time: 0.0013 memory: 25074 loss: 0.0103 +2023/06/01 11:59:44 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m-lr4e-4_20230601_025223 +2023/06/01 11:59:48 - mmengine - INFO - Epoch(train) [8][1700/5758] lr: 2.9353e-04 eta: 15:48:17 time: 0.7767 data_time: 0.0015 memory: 25074 loss: 0.0047 +2023/06/01 12:01:02 - mmengine - INFO - Epoch(train) [8][1800/5758] lr: 2.9353e-04 eta: 15:46:53 time: 0.7426 data_time: 0.0013 memory: 25074 loss: 0.0044 +2023/06/01 12:02:19 - mmengine - INFO - Epoch(train) [8][1900/5758] lr: 2.9353e-04 eta: 15:45:33 time: 0.7938 data_time: 0.0013 memory: 25074 loss: 0.0066 +2023/06/01 12:03:35 - mmengine - INFO - Epoch(train) [8][2000/5758] lr: 2.9353e-04 eta: 15:44:12 time: 0.7268 data_time: 0.0015 memory: 25074 loss: 0.0077 +2023/06/01 12:04:52 - mmengine - INFO - Epoch(train) [8][2100/5758] lr: 2.9353e-04 eta: 15:42:53 time: 0.8245 data_time: 0.0015 memory: 25074 loss: 0.0076 +2023/06/01 12:06:09 - mmengine - INFO - Epoch(train) [8][2200/5758] lr: 2.9353e-04 eta: 15:41:34 time: 0.8492 data_time: 0.0012 memory: 25074 loss: 0.0074 +2023/06/01 12:07:27 - mmengine - INFO - Epoch(train) [8][2300/5758] lr: 2.9353e-04 eta: 15:40:16 time: 0.7583 data_time: 0.0016 memory: 25074 loss: 0.0032 +2023/06/01 12:08:40 - mmengine - INFO - Epoch(train) [8][2400/5758] lr: 2.9353e-04 eta: 15:38:52 time: 0.7175 data_time: 0.0016 memory: 25074 loss: 0.0064 +2023/06/01 12:09:56 - mmengine - INFO - Epoch(train) [8][2500/5758] lr: 2.9353e-04 eta: 15:37:31 time: 0.7521 data_time: 0.0015 memory: 25074 loss: 0.0073 +2023/06/01 12:11:12 - mmengine - INFO - Epoch(train) [8][2600/5758] lr: 2.9353e-04 eta: 15:36:10 time: 0.7041 data_time: 0.0013 memory: 25074 loss: 0.0073 +2023/06/01 12:12:22 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m-lr4e-4_20230601_025223 +2023/06/01 12:12:26 - mmengine - INFO - Epoch(train) [8][2700/5758] lr: 2.9353e-04 eta: 15:34:46 time: 0.7087 data_time: 0.0015 memory: 25074 loss: 0.0043 +2023/06/01 12:13:43 - mmengine - INFO - Epoch(train) [8][2800/5758] lr: 2.9353e-04 eta: 15:33:27 time: 0.8373 data_time: 0.0729 memory: 25074 loss: 0.0055 +2023/06/01 12:15:07 - mmengine - INFO - Epoch(train) [8][2900/5758] lr: 2.9353e-04 eta: 15:32:20 time: 0.8191 data_time: 0.1148 memory: 25074 loss: 0.0123 +2023/06/01 12:16:24 - mmengine - INFO - Epoch(train) [8][3000/5758] lr: 2.9353e-04 eta: 15:31:00 time: 0.7130 data_time: 0.0096 memory: 25074 loss: 0.0080 +2023/06/01 12:17:45 - mmengine - INFO - Epoch(train) [8][3100/5758] lr: 2.9353e-04 eta: 15:29:48 time: 0.8460 data_time: 0.1411 memory: 25074 loss: 0.0039 +2023/06/01 12:19:09 - mmengine - INFO - Epoch(train) [8][3200/5758] lr: 2.9353e-04 eta: 15:28:41 time: 0.8490 data_time: 0.0016 memory: 25074 loss: 0.0062 +2023/06/01 12:20:34 - mmengine - INFO - Epoch(train) [8][3300/5758] lr: 2.9353e-04 eta: 15:27:34 time: 0.8788 data_time: 0.0013 memory: 25074 loss: 0.0054 +2023/06/01 12:21:57 - mmengine - INFO - Epoch(train) [8][3400/5758] lr: 2.9353e-04 eta: 15:26:26 time: 0.7774 data_time: 0.0013 memory: 25074 loss: 0.0054 +2023/06/01 12:23:14 - mmengine - INFO - Epoch(train) [8][3500/5758] lr: 2.9353e-04 eta: 15:25:06 time: 0.7341 data_time: 0.0017 memory: 25074 loss: 0.0044 +2023/06/01 12:24:31 - mmengine - INFO - Epoch(train) [8][3600/5758] lr: 2.9353e-04 eta: 15:23:47 time: 0.7962 data_time: 0.0014 memory: 25074 loss: 0.0027 +2023/06/01 12:25:43 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m-lr4e-4_20230601_025223 +2023/06/01 12:25:47 - mmengine - INFO - Epoch(train) [8][3700/5758] lr: 2.9353e-04 eta: 15:22:27 time: 0.7138 data_time: 0.0019 memory: 25074 loss: 0.0056 +2023/06/01 12:27:03 - mmengine - INFO - Epoch(train) [8][3800/5758] lr: 2.9353e-04 eta: 15:21:06 time: 0.7424 data_time: 0.0014 memory: 25074 loss: 0.0084 +2023/06/01 12:28:26 - mmengine - INFO - Epoch(train) [8][3900/5758] lr: 2.9353e-04 eta: 15:19:56 time: 0.8164 data_time: 0.0014 memory: 25074 loss: 0.0064 +2023/06/01 12:29:46 - mmengine - INFO - Epoch(train) [8][4000/5758] lr: 2.9353e-04 eta: 15:18:41 time: 0.7837 data_time: 0.0017 memory: 25074 loss: 0.0046 +2023/06/01 12:31:05 - mmengine - INFO - Epoch(train) [8][4100/5758] lr: 2.9353e-04 eta: 15:17:27 time: 0.8224 data_time: 0.0016 memory: 25074 loss: 0.0052 +2023/06/01 12:32:26 - mmengine - INFO - Epoch(train) [8][4200/5758] lr: 2.9353e-04 eta: 15:16:13 time: 0.8299 data_time: 0.0013 memory: 25074 loss: 0.0067 +2023/06/01 12:33:43 - mmengine - INFO - Epoch(train) [8][4300/5758] lr: 2.9353e-04 eta: 15:14:54 time: 0.7622 data_time: 0.0014 memory: 25074 loss: 0.0068 +2023/06/01 12:35:04 - mmengine - INFO - Epoch(train) [8][4400/5758] lr: 2.9353e-04 eta: 15:13:41 time: 0.7808 data_time: 0.0015 memory: 25074 loss: 0.0102 +2023/06/01 12:36:23 - mmengine - INFO - Epoch(train) [8][4500/5758] lr: 2.9353e-04 eta: 15:12:25 time: 0.7792 data_time: 0.0015 memory: 25074 loss: 0.0022 +2023/06/01 12:37:43 - mmengine - INFO - Epoch(train) [8][4600/5758] lr: 2.9353e-04 eta: 15:11:11 time: 0.8004 data_time: 0.0012 memory: 25074 loss: 0.0053 +2023/06/01 12:38:56 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m-lr4e-4_20230601_025223 +2023/06/01 12:39:00 - mmengine - INFO - Epoch(train) [8][4700/5758] lr: 2.9353e-04 eta: 15:09:52 time: 0.7794 data_time: 0.0015 memory: 25074 loss: 0.0034 +2023/06/01 12:40:22 - mmengine - INFO - Epoch(train) [8][4800/5758] lr: 2.9353e-04 eta: 15:08:41 time: 0.8116 data_time: 0.0014 memory: 25074 loss: 0.0076 +2023/06/01 12:41:42 - mmengine - INFO - Epoch(train) [8][4900/5758] lr: 2.9353e-04 eta: 15:07:26 time: 0.7434 data_time: 0.0017 memory: 25074 loss: 0.0080 +2023/06/01 12:42:57 - mmengine - INFO - Epoch(train) [8][5000/5758] lr: 2.9353e-04 eta: 15:06:04 time: 0.7443 data_time: 0.0013 memory: 25074 loss: 0.0047 +2023/06/01 12:44:16 - mmengine - INFO - Epoch(train) [8][5100/5758] lr: 2.9353e-04 eta: 15:04:48 time: 0.7896 data_time: 0.0012 memory: 25074 loss: 0.0046 +2023/06/01 12:45:37 - mmengine - INFO - Epoch(train) [8][5200/5758] lr: 2.9353e-04 eta: 15:03:34 time: 0.8348 data_time: 0.0015 memory: 25074 loss: 0.0067 +2023/06/01 12:46:56 - mmengine - INFO - Epoch(train) [8][5300/5758] lr: 2.9353e-04 eta: 15:02:19 time: 0.7869 data_time: 0.0014 memory: 25074 loss: 0.0047 +2023/06/01 12:48:12 - mmengine - INFO - Epoch(train) [8][5400/5758] lr: 2.9353e-04 eta: 15:00:58 time: 0.7504 data_time: 0.0018 memory: 25074 loss: 0.0049 +2023/06/01 12:49:28 - mmengine - INFO - Epoch(train) [8][5500/5758] lr: 2.9353e-04 eta: 14:59:38 time: 0.7846 data_time: 0.0016 memory: 25074 loss: 0.0090 +2023/06/01 12:50:47 - mmengine - INFO - Epoch(train) [8][5600/5758] lr: 2.9353e-04 eta: 14:58:22 time: 0.8250 data_time: 0.0017 memory: 25074 loss: 0.0035 +2023/06/01 12:52:00 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m-lr4e-4_20230601_025223 +2023/06/01 12:52:04 - mmengine - INFO - Epoch(train) [8][5700/5758] lr: 2.9353e-04 eta: 14:57:03 time: 0.7800 data_time: 0.0014 memory: 25074 loss: 0.0037 +2023/06/01 12:52:48 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m-lr4e-4_20230601_025223 +2023/06/01 12:52:48 - mmengine - INFO - Saving checkpoint at 8 epochs +2023/06/01 12:53:08 - mmengine - INFO - Epoch(val) [8][16/16] accuracy/top1: 99.9684 single-label/precision_classwise: [100.0, 0.0] single-label/recall_classwise: [99.96842956542969, 0.0] single-label/f1-score_classwise: [99.98420715332031, 0.0] data_time: 0.2517 time: 0.6379 +2023/06/01 12:54:36 - mmengine - INFO - Epoch(train) [9][ 100/5758] lr: 2.6526e-04 eta: 14:55:13 time: 0.8176 data_time: 0.0016 memory: 25074 loss: 0.0101 +2023/06/01 12:55:52 - mmengine - INFO - Epoch(train) [9][ 200/5758] lr: 2.6526e-04 eta: 14:53:53 time: 0.7844 data_time: 0.0015 memory: 25074 loss: 0.0030 +2023/06/01 12:57:07 - mmengine - INFO - Epoch(train) [9][ 300/5758] lr: 2.6526e-04 eta: 14:52:30 time: 0.7494 data_time: 0.0013 memory: 25074 loss: 0.0091 +2023/06/01 12:58:25 - mmengine - INFO - Epoch(train) [9][ 400/5758] lr: 2.6526e-04 eta: 14:51:13 time: 0.8462 data_time: 0.0014 memory: 25074 loss: 0.0043 +2023/06/01 12:59:43 - mmengine - INFO - Epoch(train) [9][ 500/5758] lr: 2.6526e-04 eta: 14:49:55 time: 0.7555 data_time: 0.0013 memory: 25074 loss: 0.0032 +2023/06/01 13:01:02 - mmengine - INFO - Epoch(train) [9][ 600/5758] lr: 2.6526e-04 eta: 14:48:38 time: 0.8164 data_time: 0.0017 memory: 25074 loss: 0.0028 +2023/06/01 13:02:18 - mmengine - INFO - Epoch(train) [9][ 700/5758] lr: 2.6526e-04 eta: 14:47:19 time: 0.7282 data_time: 0.0013 memory: 25074 loss: 0.0048 +2023/06/01 13:03:36 - mmengine - INFO - Epoch(train) [9][ 800/5758] lr: 2.6526e-04 eta: 14:46:01 time: 0.7747 data_time: 0.0015 memory: 25074 loss: 0.0038 +2023/06/01 13:04:56 - mmengine - INFO - Epoch(train) [9][ 900/5758] lr: 2.6526e-04 eta: 14:44:46 time: 0.8504 data_time: 0.0016 memory: 25074 loss: 0.0038 +2023/06/01 13:05:24 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m-lr4e-4_20230601_025223 +2023/06/01 13:06:12 - mmengine - INFO - Epoch(train) [9][1000/5758] lr: 2.6526e-04 eta: 14:43:25 time: 0.7096 data_time: 0.0015 memory: 25074 loss: 0.0064 +2023/06/01 13:07:28 - mmengine - INFO - Epoch(train) [9][1100/5758] lr: 2.6526e-04 eta: 14:42:05 time: 0.7514 data_time: 0.0015 memory: 25074 loss: 0.0016 +2023/06/01 13:08:45 - mmengine - INFO - Epoch(train) [9][1200/5758] lr: 2.6526e-04 eta: 14:40:46 time: 0.7462 data_time: 0.0016 memory: 25074 loss: 0.0038 +2023/06/01 13:10:02 - mmengine - INFO - Epoch(train) [9][1300/5758] lr: 2.6526e-04 eta: 14:39:27 time: 0.7379 data_time: 0.0327 memory: 25074 loss: 0.0029 +2023/06/01 13:11:20 - mmengine - INFO - Epoch(train) [9][1400/5758] lr: 2.6526e-04 eta: 14:38:09 time: 0.8054 data_time: 0.1002 memory: 25074 loss: 0.0033 +2023/06/01 13:12:38 - mmengine - INFO - Epoch(train) [9][1500/5758] lr: 2.6526e-04 eta: 14:36:52 time: 0.7718 data_time: 0.0016 memory: 25074 loss: 0.0048 +2023/06/01 13:13:57 - mmengine - INFO - Epoch(train) [9][1600/5758] lr: 2.6526e-04 eta: 14:35:35 time: 0.7555 data_time: 0.0014 memory: 25074 loss: 0.0036 +2023/06/01 13:15:17 - mmengine - INFO - Epoch(train) [9][1700/5758] lr: 2.6526e-04 eta: 14:34:21 time: 0.8240 data_time: 0.0014 memory: 25074 loss: 0.0026 +2023/06/01 13:16:44 - mmengine - INFO - Epoch(train) [9][1800/5758] lr: 2.6526e-04 eta: 14:33:16 time: 1.0330 data_time: 0.0013 memory: 25074 loss: 0.0051 +2023/06/01 13:18:11 - mmengine - INFO - Epoch(train) [9][1900/5758] lr: 2.6526e-04 eta: 14:32:11 time: 0.7988 data_time: 0.0014 memory: 25074 loss: 0.0022 +2023/06/01 13:18:38 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m-lr4e-4_20230601_025223 +2023/06/01 13:19:30 - mmengine - INFO - Epoch(train) [9][2000/5758] lr: 2.6526e-04 eta: 14:30:55 time: 0.7403 data_time: 0.0016 memory: 25074 loss: 0.0046 +2023/06/01 13:20:49 - mmengine - INFO - Epoch(train) [9][2100/5758] lr: 2.6526e-04 eta: 14:29:38 time: 0.8200 data_time: 0.0014 memory: 25074 loss: 0.0057 +2023/06/01 13:22:07 - mmengine - INFO - Epoch(train) [9][2200/5758] lr: 2.6526e-04 eta: 14:28:20 time: 0.7848 data_time: 0.0015 memory: 25074 loss: 0.0060 +2023/06/01 13:23:24 - mmengine - INFO - Epoch(train) [9][2300/5758] lr: 2.6526e-04 eta: 14:27:02 time: 0.8107 data_time: 0.0015 memory: 25074 loss: 0.0053 +2023/06/01 13:24:43 - mmengine - INFO - Epoch(train) [9][2400/5758] lr: 2.6526e-04 eta: 14:25:45 time: 0.7425 data_time: 0.0014 memory: 25074 loss: 0.0053 +2023/06/01 13:26:02 - mmengine - INFO - Epoch(train) [9][2500/5758] lr: 2.6526e-04 eta: 14:24:29 time: 0.7792 data_time: 0.0021 memory: 25074 loss: 0.0018 +2023/06/01 13:27:24 - mmengine - INFO - Epoch(train) [9][2600/5758] lr: 2.6526e-04 eta: 14:23:16 time: 0.7862 data_time: 0.0014 memory: 25074 loss: 0.0045 +2023/06/01 13:28:46 - mmengine - INFO - Epoch(train) [9][2700/5758] lr: 2.6526e-04 eta: 14:22:03 time: 0.7917 data_time: 0.0015 memory: 25074 loss: 0.0036 +2023/06/01 13:30:06 - mmengine - INFO - Epoch(train) [9][2800/5758] lr: 2.6526e-04 eta: 14:20:49 time: 0.7571 data_time: 0.0020 memory: 25074 loss: 0.0105 +2023/06/01 13:31:25 - mmengine - INFO - Epoch(train) [9][2900/5758] lr: 2.6526e-04 eta: 14:19:32 time: 0.7906 data_time: 0.0016 memory: 25074 loss: 0.0051 +2023/06/01 13:31:52 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m-lr4e-4_20230601_025223 +2023/06/01 13:32:43 - mmengine - INFO - Epoch(train) [9][3000/5758] lr: 2.6526e-04 eta: 14:18:14 time: 0.8172 data_time: 0.0014 memory: 25074 loss: 0.0055 +2023/06/01 13:33:59 - mmengine - INFO - Epoch(train) [9][3100/5758] lr: 2.6526e-04 eta: 14:16:55 time: 0.7779 data_time: 0.0741 memory: 25074 loss: 0.0048 +2023/06/01 13:35:18 - mmengine - INFO - Epoch(train) [9][3200/5758] lr: 2.6526e-04 eta: 14:15:38 time: 0.8225 data_time: 0.1179 memory: 25074 loss: 0.0079 +2023/06/01 13:36:38 - mmengine - INFO - Epoch(train) [9][3300/5758] lr: 2.6526e-04 eta: 14:14:22 time: 0.7911 data_time: 0.0813 memory: 25074 loss: 0.0027 +2023/06/01 13:37:57 - mmengine - INFO - Epoch(train) [9][3400/5758] lr: 2.6526e-04 eta: 14:13:06 time: 0.8399 data_time: 0.1306 memory: 25074 loss: 0.0021 +2023/06/01 13:39:16 - mmengine - INFO - Epoch(train) [9][3500/5758] lr: 2.6526e-04 eta: 14:11:49 time: 0.7852 data_time: 0.0814 memory: 25074 loss: 0.0068 +2023/06/01 13:40:34 - mmengine - INFO - Epoch(train) [9][3600/5758] lr: 2.6526e-04 eta: 14:10:32 time: 0.8148 data_time: 0.1111 memory: 25074 loss: 0.0015 +2023/06/01 13:41:51 - mmengine - INFO - Epoch(train) [9][3700/5758] lr: 2.6526e-04 eta: 14:09:12 time: 0.8466 data_time: 0.1385 memory: 25074 loss: 0.0039 +2023/06/01 13:43:11 - mmengine - INFO - Epoch(train) [9][3800/5758] lr: 2.6526e-04 eta: 14:07:57 time: 0.7784 data_time: 0.0731 memory: 25074 loss: 0.0051 +2023/06/01 13:44:28 - mmengine - INFO - Epoch(train) [9][3900/5758] lr: 2.6526e-04 eta: 14:06:38 time: 0.7899 data_time: 0.0748 memory: 25074 loss: 0.0036 +2023/06/01 13:44:55 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m-lr4e-4_20230601_025223 +2023/06/01 13:45:47 - mmengine - INFO - Epoch(train) [9][4000/5758] lr: 2.6526e-04 eta: 14:05:22 time: 0.7897 data_time: 0.0846 memory: 25074 loss: 0.0045 +2023/06/01 13:47:04 - mmengine - INFO - Epoch(train) [9][4100/5758] lr: 2.6526e-04 eta: 14:04:02 time: 0.7821 data_time: 0.0786 memory: 25074 loss: 0.0054 +2023/06/01 13:48:23 - mmengine - INFO - Epoch(train) [9][4200/5758] lr: 2.6526e-04 eta: 14:02:46 time: 0.7515 data_time: 0.0484 memory: 25074 loss: 0.0035 +2023/06/01 13:49:41 - mmengine - INFO - Epoch(train) [9][4300/5758] lr: 2.6526e-04 eta: 14:01:29 time: 0.7980 data_time: 0.0017 memory: 25074 loss: 0.0038 +2023/06/01 13:50:59 - mmengine - INFO - Epoch(train) [9][4400/5758] lr: 2.6526e-04 eta: 14:00:11 time: 0.8107 data_time: 0.0020 memory: 25074 loss: 0.0020 +2023/06/01 13:52:15 - mmengine - INFO - Epoch(train) [9][4500/5758] lr: 2.6526e-04 eta: 13:58:50 time: 0.7461 data_time: 0.0014 memory: 25074 loss: 0.0038 +2023/06/01 13:53:32 - mmengine - INFO - Epoch(train) [9][4600/5758] lr: 2.6526e-04 eta: 13:57:31 time: 0.7556 data_time: 0.0014 memory: 25074 loss: 0.0014 +2023/06/01 13:54:49 - mmengine - INFO - Epoch(train) [9][4700/5758] lr: 2.6526e-04 eta: 13:56:12 time: 0.7805 data_time: 0.0016 memory: 25074 loss: 0.0049 +2023/06/01 13:56:08 - mmengine - INFO - Epoch(train) [9][4800/5758] lr: 2.6526e-04 eta: 13:54:55 time: 0.8374 data_time: 0.0014 memory: 25074 loss: 0.0046 +2023/06/01 13:57:25 - mmengine - INFO - Epoch(train) [9][4900/5758] lr: 2.6526e-04 eta: 13:53:36 time: 0.7419 data_time: 0.0049 memory: 25074 loss: 0.0032 +2023/06/01 13:57:55 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m-lr4e-4_20230601_025223 +2023/06/01 13:58:44 - mmengine - INFO - Epoch(train) [9][5000/5758] lr: 2.6526e-04 eta: 13:52:19 time: 0.7634 data_time: 0.0015 memory: 25074 loss: 0.0032 +2023/06/01 14:00:02 - mmengine - INFO - Epoch(train) [9][5100/5758] lr: 2.6526e-04 eta: 13:51:02 time: 0.7813 data_time: 0.0015 memory: 25074 loss: 0.0044 +2023/06/01 14:01:19 - mmengine - INFO - Epoch(train) [9][5200/5758] lr: 2.6526e-04 eta: 13:49:43 time: 0.7531 data_time: 0.0013 memory: 25074 loss: 0.0035 +2023/06/01 14:02:37 - mmengine - INFO - Epoch(train) [9][5300/5758] lr: 2.6526e-04 eta: 13:48:25 time: 0.7890 data_time: 0.0015 memory: 25074 loss: 0.0029 +2023/06/01 14:03:58 - mmengine - INFO - Epoch(train) [9][5400/5758] lr: 2.6526e-04 eta: 13:47:10 time: 0.8217 data_time: 0.0013 memory: 25074 loss: 0.0049 +2023/06/01 14:05:15 - mmengine - INFO - Epoch(train) [9][5500/5758] lr: 2.6526e-04 eta: 13:45:52 time: 0.7716 data_time: 0.0014 memory: 25074 loss: 0.0020 +2023/06/01 14:06:33 - mmengine - INFO - Epoch(train) [9][5600/5758] lr: 2.6526e-04 eta: 13:44:33 time: 0.7720 data_time: 0.0016 memory: 25074 loss: 0.0039 +2023/06/01 14:07:49 - mmengine - INFO - Epoch(train) [9][5700/5758] lr: 2.6526e-04 eta: 13:43:14 time: 0.7533 data_time: 0.0016 memory: 25074 loss: 0.0032 +2023/06/01 14:08:34 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m-lr4e-4_20230601_025223 +2023/06/01 14:08:34 - mmengine - INFO - Saving checkpoint at 9 epochs +2023/06/01 14:08:54 - mmengine - INFO - Epoch(val) [9][16/16] accuracy/top1: 99.9747 single-label/precision_classwise: [100.0, 0.0] single-label/recall_classwise: [99.9747543334961, 0.0] single-label/f1-score_classwise: [99.98737335205078, 0.0] data_time: 0.2696 time: 0.6554 +2023/06/01 14:10:21 - mmengine - INFO - Epoch(train) [10][ 100/5758] lr: 2.3550e-04 eta: 13:41:21 time: 0.7609 data_time: 0.0572 memory: 25074 loss: 0.0045 +2023/06/01 14:11:22 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m-lr4e-4_20230601_025223 +2023/06/01 14:11:39 - mmengine - INFO - Epoch(train) [10][ 200/5758] lr: 2.3550e-04 eta: 13:40:04 time: 0.8048 data_time: 0.0889 memory: 25074 loss: 0.0012 +2023/06/01 14:12:55 - mmengine - INFO - Epoch(train) [10][ 300/5758] lr: 2.3550e-04 eta: 13:38:44 time: 0.7699 data_time: 0.0137 memory: 25074 loss: 0.0074 +2023/06/01 14:14:14 - mmengine - INFO - Epoch(train) [10][ 400/5758] lr: 2.3550e-04 eta: 13:37:27 time: 0.8420 data_time: 0.1002 memory: 25074 loss: 0.0019 +2023/06/01 14:15:43 - mmengine - INFO - Epoch(train) [10][ 500/5758] lr: 2.3550e-04 eta: 13:36:22 time: 0.7677 data_time: 0.0015 memory: 25074 loss: 0.0027 +2023/06/01 14:16:59 - mmengine - INFO - Epoch(train) [10][ 600/5758] lr: 2.3550e-04 eta: 13:35:02 time: 0.7667 data_time: 0.0078 memory: 25074 loss: 0.0050 +2023/06/01 14:18:20 - mmengine - INFO - Epoch(train) [10][ 700/5758] lr: 2.3550e-04 eta: 13:33:48 time: 0.8272 data_time: 0.0014 memory: 25074 loss: 0.0033 +2023/06/01 14:19:37 - mmengine - INFO - Epoch(train) [10][ 800/5758] lr: 2.3550e-04 eta: 13:32:28 time: 0.7318 data_time: 0.0013 memory: 25074 loss: 0.0024 +2023/06/01 14:20:56 - mmengine - INFO - Epoch(train) [10][ 900/5758] lr: 2.3550e-04 eta: 13:31:12 time: 0.7616 data_time: 0.0016 memory: 25074 loss: 0.0018 +2023/06/01 14:22:14 - mmengine - INFO - Epoch(train) [10][1000/5758] lr: 2.3550e-04 eta: 13:29:54 time: 0.7791 data_time: 0.0718 memory: 25074 loss: 0.0065 +2023/06/01 14:23:36 - mmengine - INFO - Epoch(train) [10][1100/5758] lr: 2.3550e-04 eta: 13:28:41 time: 0.7737 data_time: 0.0016 memory: 25074 loss: 0.0037 +2023/06/01 14:24:36 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m-lr4e-4_20230601_025223 +2023/06/01 14:24:53 - mmengine - INFO - Epoch(train) [10][1200/5758] lr: 2.3550e-04 eta: 13:27:22 time: 0.8046 data_time: 0.0014 memory: 25074 loss: 0.0053 +2023/06/01 14:26:12 - mmengine - INFO - Epoch(train) [10][1300/5758] lr: 2.3550e-04 eta: 13:26:05 time: 0.8024 data_time: 0.0015 memory: 25074 loss: 0.0061 +2023/06/01 14:27:32 - mmengine - INFO - Epoch(train) [10][1400/5758] lr: 2.3550e-04 eta: 13:24:49 time: 0.7712 data_time: 0.0013 memory: 25074 loss: 0.0053 +2023/06/01 14:28:54 - mmengine - INFO - Epoch(train) [10][1500/5758] lr: 2.3550e-04 eta: 13:23:35 time: 0.8612 data_time: 0.0015 memory: 25074 loss: 0.0026 +2023/06/01 14:30:18 - mmengine - INFO - Epoch(train) [10][1600/5758] lr: 2.3550e-04 eta: 13:22:25 time: 0.8964 data_time: 0.0012 memory: 25074 loss: 0.0027 +2023/06/01 14:31:51 - mmengine - INFO - Epoch(train) [10][1700/5758] lr: 2.3550e-04 eta: 13:21:24 time: 0.8743 data_time: 0.0016 memory: 25074 loss: 0.0019 +2023/06/01 14:33:21 - mmengine - INFO - Epoch(train) [10][1800/5758] lr: 2.3550e-04 eta: 13:20:20 time: 1.4330 data_time: 0.0067 memory: 25074 loss: 0.0024 +2023/06/01 14:34:56 - mmengine - INFO - Epoch(train) [10][1900/5758] lr: 2.3550e-04 eta: 13:19:21 time: 0.8864 data_time: 0.0020 memory: 25074 loss: 0.0045 +2023/06/01 14:36:19 - mmengine - INFO - Epoch(train) [10][2000/5758] lr: 2.3550e-04 eta: 13:18:09 time: 0.8816 data_time: 0.0015 memory: 25074 loss: 0.0043 +2023/06/01 14:37:39 - mmengine - INFO - Epoch(train) [10][2100/5758] lr: 2.3550e-04 eta: 13:16:53 time: 0.7687 data_time: 0.0013 memory: 25074 loss: 0.0040 +2023/06/01 14:38:41 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m-lr4e-4_20230601_025223 +2023/06/01 14:38:59 - mmengine - INFO - Epoch(train) [10][2200/5758] lr: 2.3550e-04 eta: 13:15:37 time: 0.8162 data_time: 0.0013 memory: 25074 loss: 0.0012 +2023/06/01 14:40:18 - mmengine - INFO - Epoch(train) [10][2300/5758] lr: 2.3550e-04 eta: 13:14:20 time: 0.7689 data_time: 0.0015 memory: 25074 loss: 0.0039 +2023/06/01 14:41:38 - mmengine - INFO - Epoch(train) [10][2400/5758] lr: 2.3550e-04 eta: 13:13:04 time: 0.7375 data_time: 0.0013 memory: 25074 loss: 0.0055 +2023/06/01 14:42:57 - mmengine - INFO - Epoch(train) [10][2500/5758] lr: 2.3550e-04 eta: 13:11:47 time: 0.7793 data_time: 0.0028 memory: 25074 loss: 0.0078 +2023/06/01 14:44:15 - mmengine - INFO - Epoch(train) [10][2600/5758] lr: 2.3550e-04 eta: 13:10:29 time: 0.8095 data_time: 0.0026 memory: 25074 loss: 0.0033 +2023/06/01 14:45:32 - mmengine - INFO - Epoch(train) [10][2700/5758] lr: 2.3550e-04 eta: 13:09:10 time: 0.7800 data_time: 0.0015 memory: 25074 loss: 0.0033 +2023/06/01 14:46:49 - mmengine - INFO - Epoch(train) [10][2800/5758] lr: 2.3550e-04 eta: 13:07:50 time: 0.7883 data_time: 0.0013 memory: 25074 loss: 0.0042 +2023/06/01 14:48:06 - mmengine - INFO - Epoch(train) [10][2900/5758] lr: 2.3550e-04 eta: 13:06:31 time: 0.7512 data_time: 0.0014 memory: 25074 loss: 0.0017 +2023/06/01 14:49:25 - mmengine - INFO - Epoch(train) [10][3000/5758] lr: 2.3550e-04 eta: 13:05:14 time: 0.7922 data_time: 0.0015 memory: 25074 loss: 0.0034 +2023/06/01 14:50:44 - mmengine - INFO - Epoch(train) [10][3100/5758] lr: 2.3550e-04 eta: 13:03:56 time: 0.7578 data_time: 0.0018 memory: 25074 loss: 0.0065 +2023/06/01 14:51:40 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m-lr4e-4_20230601_025223 +2023/06/01 14:51:56 - mmengine - INFO - Epoch(train) [10][3200/5758] lr: 2.3550e-04 eta: 13:02:32 time: 0.7229 data_time: 0.0016 memory: 25074 loss: 0.0022 +2023/06/01 14:53:11 - mmengine - INFO - Epoch(train) [10][3300/5758] lr: 2.3550e-04 eta: 13:01:10 time: 0.7517 data_time: 0.0024 memory: 25074 loss: 0.0020 +2023/06/01 14:54:25 - mmengine - INFO - Epoch(train) [10][3400/5758] lr: 2.3550e-04 eta: 12:59:48 time: 0.7662 data_time: 0.0016 memory: 25074 loss: 0.0023 +2023/06/01 14:55:40 - mmengine - INFO - Epoch(train) [10][3500/5758] lr: 2.3550e-04 eta: 12:58:27 time: 0.7472 data_time: 0.0015 memory: 25074 loss: 0.0019 +2023/06/01 14:56:55 - mmengine - INFO - Epoch(train) [10][3600/5758] lr: 2.3550e-04 eta: 12:57:05 time: 0.7647 data_time: 0.0021 memory: 25074 loss: 0.0027 +2023/06/01 14:58:16 - mmengine - INFO - Epoch(train) [10][3700/5758] lr: 2.3550e-04 eta: 12:55:50 time: 0.7772 data_time: 0.0014 memory: 25074 loss: 0.0033 +2023/06/01 14:59:31 - mmengine - INFO - Epoch(train) [10][3800/5758] lr: 2.3550e-04 eta: 12:54:29 time: 0.7240 data_time: 0.0016 memory: 25074 loss: 0.0039 +2023/06/01 15:00:46 - mmengine - INFO - Epoch(train) [10][3900/5758] lr: 2.3550e-04 eta: 12:53:08 time: 0.8107 data_time: 0.0015 memory: 25074 loss: 0.0053 +2023/06/01 15:02:00 - mmengine - INFO - Epoch(train) [10][4000/5758] lr: 2.3550e-04 eta: 12:51:45 time: 0.7875 data_time: 0.0015 memory: 25074 loss: 0.0056 +2023/06/01 15:03:15 - mmengine - INFO - Epoch(train) [10][4100/5758] lr: 2.3550e-04 eta: 12:50:24 time: 0.7610 data_time: 0.0015 memory: 25074 loss: 0.0019 +2023/06/01 15:04:12 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m-lr4e-4_20230601_025223 +2023/06/01 15:04:29 - mmengine - INFO - Epoch(train) [10][4200/5758] lr: 2.3550e-04 eta: 12:49:02 time: 0.7405 data_time: 0.0014 memory: 25074 loss: 0.0056 +2023/06/01 15:05:44 - mmengine - INFO - Epoch(train) [10][4300/5758] lr: 2.3550e-04 eta: 12:47:41 time: 0.7767 data_time: 0.0015 memory: 25074 loss: 0.0042 +2023/06/01 15:06:58 - mmengine - INFO - Epoch(train) [10][4400/5758] lr: 2.3550e-04 eta: 12:46:19 time: 0.7357 data_time: 0.0013 memory: 25074 loss: 0.0028 +2023/06/01 15:08:12 - mmengine - INFO - Epoch(train) [10][4500/5758] lr: 2.3550e-04 eta: 12:44:56 time: 0.7194 data_time: 0.0013 memory: 25074 loss: 0.0034 +2023/06/01 15:09:25 - mmengine - INFO - Epoch(train) [10][4600/5758] lr: 2.3550e-04 eta: 12:43:33 time: 0.7068 data_time: 0.0014 memory: 25074 loss: 0.0025 +2023/06/01 15:10:44 - mmengine - INFO - Epoch(train) [10][4700/5758] lr: 2.3550e-04 eta: 12:42:15 time: 0.8126 data_time: 0.0015 memory: 25074 loss: 0.0070 +2023/06/01 15:12:01 - mmengine - INFO - Epoch(train) [10][4800/5758] lr: 2.3550e-04 eta: 12:40:57 time: 0.7594 data_time: 0.0016 memory: 25074 loss: 0.0059 +2023/06/01 15:13:19 - mmengine - INFO - Epoch(train) [10][4900/5758] lr: 2.3550e-04 eta: 12:39:38 time: 0.7697 data_time: 0.0022 memory: 25074 loss: 0.0046 +2023/06/01 15:14:40 - mmengine - INFO - Epoch(train) [10][5000/5758] lr: 2.3550e-04 eta: 12:38:24 time: 0.9141 data_time: 0.0014 memory: 25074 loss: 0.0018 +2023/06/01 15:16:01 - mmengine - INFO - Epoch(train) [10][5100/5758] lr: 2.3550e-04 eta: 12:37:09 time: 0.8386 data_time: 0.0013 memory: 25074 loss: 0.0042 +2023/06/01 15:17:03 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m-lr4e-4_20230601_025223 +2023/06/01 15:17:22 - mmengine - INFO - Epoch(train) [10][5200/5758] lr: 2.3550e-04 eta: 12:35:54 time: 0.8170 data_time: 0.0016 memory: 25074 loss: 0.0047 +2023/06/01 15:18:45 - mmengine - INFO - Epoch(train) [10][5300/5758] lr: 2.3550e-04 eta: 12:34:41 time: 0.8308 data_time: 0.0015 memory: 25074 loss: 0.0010 +2023/06/01 15:20:05 - mmengine - INFO - Epoch(train) [10][5400/5758] lr: 2.3550e-04 eta: 12:33:25 time: 0.7947 data_time: 0.0016 memory: 25074 loss: 0.0022 +2023/06/01 15:21:23 - mmengine - INFO - Epoch(train) [10][5500/5758] lr: 2.3550e-04 eta: 12:32:07 time: 0.7589 data_time: 0.0012 memory: 25074 loss: 0.0040 +2023/06/01 15:22:42 - mmengine - INFO - Epoch(train) [10][5600/5758] lr: 2.3550e-04 eta: 12:30:50 time: 0.7645 data_time: 0.0014 memory: 25074 loss: 0.0010 +2023/06/01 15:23:59 - mmengine - INFO - Epoch(train) [10][5700/5758] lr: 2.3550e-04 eta: 12:29:31 time: 0.7397 data_time: 0.0015 memory: 25074 loss: 0.0005 +2023/06/01 15:24:42 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m-lr4e-4_20230601_025223 +2023/06/01 15:24:42 - mmengine - INFO - Saving checkpoint at 10 epochs +2023/06/01 15:25:01 - mmengine - INFO - Epoch(val) [10][16/16] accuracy/top1: 99.9874 single-label/precision_classwise: [100.0, 0.0] single-label/recall_classwise: [99.98737335205078, 0.0] single-label/f1-score_classwise: [99.99369049072266, 0.0] data_time: 0.2496 time: 0.6368 +2023/06/01 15:26:27 - mmengine - INFO - Epoch(train) [11][ 100/5758] lr: 2.0500e-04 eta: 12:27:33 time: 0.7271 data_time: 0.0018 memory: 25074 loss: 0.0015 +2023/06/01 15:27:46 - mmengine - INFO - Epoch(train) [11][ 200/5758] lr: 2.0500e-04 eta: 12:26:16 time: 0.8801 data_time: 0.0451 memory: 25074 loss: 0.0034 +2023/06/01 15:29:10 - mmengine - INFO - Epoch(train) [11][ 300/5758] lr: 2.0500e-04 eta: 12:25:04 time: 0.9065 data_time: 0.0510 memory: 25074 loss: 0.0065 +2023/06/01 15:30:29 - mmengine - INFO - Epoch(train) [11][ 400/5758] lr: 2.0500e-04 eta: 12:23:47 time: 0.8374 data_time: 0.0016 memory: 25074 loss: 0.0008 +2023/06/01 15:30:45 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m-lr4e-4_20230601_025223 +2023/06/01 15:31:52 - mmengine - INFO - Epoch(train) [11][ 500/5758] lr: 2.0500e-04 eta: 12:22:34 time: 0.9274 data_time: 0.0015 memory: 25074 loss: 0.0057 +2023/06/01 15:33:15 - mmengine - INFO - Epoch(train) [11][ 600/5758] lr: 2.0500e-04 eta: 12:21:20 time: 0.9980 data_time: 0.0018 memory: 25074 loss: 0.0028 +2023/06/01 15:34:33 - mmengine - INFO - Epoch(train) [11][ 700/5758] lr: 2.0500e-04 eta: 12:20:02 time: 0.7386 data_time: 0.0014 memory: 25074 loss: 0.0018 +2023/06/01 15:35:49 - mmengine - INFO - Epoch(train) [11][ 800/5758] lr: 2.0500e-04 eta: 12:18:42 time: 0.7587 data_time: 0.0017 memory: 25074 loss: 0.0031 +2023/06/01 15:37:10 - mmengine - INFO - Epoch(train) [11][ 900/5758] lr: 2.0500e-04 eta: 12:17:27 time: 0.8805 data_time: 0.0019 memory: 25074 loss: 0.0029 +2023/06/01 15:38:28 - mmengine - INFO - Epoch(train) [11][1000/5758] lr: 2.0500e-04 eta: 12:16:09 time: 0.7420 data_time: 0.0015 memory: 25074 loss: 0.0012 +2023/06/01 15:39:45 - mmengine - INFO - Epoch(train) [11][1100/5758] lr: 2.0500e-04 eta: 12:14:50 time: 0.7582 data_time: 0.0015 memory: 25074 loss: 0.0029 +2023/06/01 15:41:02 - mmengine - INFO - Epoch(train) [11][1200/5758] lr: 2.0500e-04 eta: 12:13:31 time: 0.7900 data_time: 0.0014 memory: 25074 loss: 0.0015 +2023/06/01 15:42:18 - mmengine - INFO - Epoch(train) [11][1300/5758] lr: 2.0500e-04 eta: 12:12:11 time: 0.8209 data_time: 0.0015 memory: 25074 loss: 0.0032 +2023/06/01 15:43:36 - mmengine - INFO - Epoch(train) [11][1400/5758] lr: 2.0500e-04 eta: 12:10:53 time: 0.7465 data_time: 0.0017 memory: 25074 loss: 0.0054 +2023/06/01 15:43:52 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m-lr4e-4_20230601_025223 +2023/06/01 15:44:55 - mmengine - INFO - Epoch(train) [11][1500/5758] lr: 2.0500e-04 eta: 12:09:35 time: 0.7894 data_time: 0.0816 memory: 25074 loss: 0.0050 +2023/06/01 15:46:15 - mmengine - INFO - Epoch(train) [11][1600/5758] lr: 2.0500e-04 eta: 12:08:20 time: 0.7492 data_time: 0.0022 memory: 25074 loss: 0.0013 +2023/06/01 15:47:33 - mmengine - INFO - Epoch(train) [11][1700/5758] lr: 2.0500e-04 eta: 12:07:01 time: 0.7897 data_time: 0.0020 memory: 25074 loss: 0.0032 +2023/06/01 15:48:49 - mmengine - INFO - Epoch(train) [11][1800/5758] lr: 2.0500e-04 eta: 12:05:41 time: 0.7466 data_time: 0.0014 memory: 25074 loss: 0.0034 +2023/06/01 15:50:09 - mmengine - INFO - Epoch(train) [11][1900/5758] lr: 2.0500e-04 eta: 12:04:24 time: 0.8003 data_time: 0.0018 memory: 25074 loss: 0.0024 +2023/06/01 15:51:26 - mmengine - INFO - Epoch(train) [11][2000/5758] lr: 2.0500e-04 eta: 12:03:05 time: 0.7389 data_time: 0.0017 memory: 25074 loss: 0.0057 +2023/06/01 15:52:42 - mmengine - INFO - Epoch(train) [11][2100/5758] lr: 2.0500e-04 eta: 12:01:46 time: 0.8425 data_time: 0.0019 memory: 25074 loss: 0.0072 +2023/06/01 15:53:58 - mmengine - INFO - Epoch(train) [11][2200/5758] lr: 2.0500e-04 eta: 12:00:25 time: 0.7815 data_time: 0.0017 memory: 25074 loss: 0.0024 +2023/06/01 15:55:12 - mmengine - INFO - Epoch(train) [11][2300/5758] lr: 2.0500e-04 eta: 11:59:04 time: 0.7463 data_time: 0.0013 memory: 25074 loss: 0.0050 +2023/06/01 15:56:30 - mmengine - INFO - Epoch(train) [11][2400/5758] lr: 2.0500e-04 eta: 11:57:45 time: 0.7921 data_time: 0.0014 memory: 25074 loss: 0.0045 +2023/06/01 15:56:45 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m-lr4e-4_20230601_025223 +2023/06/01 15:57:47 - mmengine - INFO - Epoch(train) [11][2500/5758] lr: 2.0500e-04 eta: 11:56:27 time: 0.7467 data_time: 0.0014 memory: 25074 loss: 0.0026 +2023/06/01 15:59:06 - mmengine - INFO - Epoch(train) [11][2600/5758] lr: 2.0500e-04 eta: 11:55:10 time: 0.7708 data_time: 0.0016 memory: 25074 loss: 0.0015 +2023/06/01 16:00:27 - mmengine - INFO - Epoch(train) [11][2700/5758] lr: 2.0500e-04 eta: 11:53:54 time: 0.7862 data_time: 0.0015 memory: 25074 loss: 0.0024 +2023/06/01 16:01:44 - mmengine - INFO - Epoch(train) [11][2800/5758] lr: 2.0500e-04 eta: 11:52:35 time: 0.8156 data_time: 0.0018 memory: 25074 loss: 0.0020 +2023/06/01 16:03:04 - mmengine - INFO - Epoch(train) [11][2900/5758] lr: 2.0500e-04 eta: 11:51:18 time: 0.8130 data_time: 0.0013 memory: 25074 loss: 0.0033 +2023/06/01 16:04:22 - mmengine - INFO - Epoch(train) [11][3000/5758] lr: 2.0500e-04 eta: 11:50:00 time: 0.7654 data_time: 0.0015 memory: 25074 loss: 0.0024 +2023/06/01 16:05:40 - mmengine - INFO - Epoch(train) [11][3100/5758] lr: 2.0500e-04 eta: 11:48:42 time: 0.7790 data_time: 0.0015 memory: 25074 loss: 0.0013 +2023/06/01 16:06:59 - mmengine - INFO - Epoch(train) [11][3200/5758] lr: 2.0500e-04 eta: 11:47:26 time: 0.7409 data_time: 0.0015 memory: 25074 loss: 0.0044 +2023/06/01 16:08:15 - mmengine - INFO - Epoch(train) [11][3300/5758] lr: 2.0500e-04 eta: 11:46:06 time: 0.7407 data_time: 0.0017 memory: 25074 loss: 0.0013 +2023/06/01 16:09:32 - mmengine - INFO - Epoch(train) [11][3400/5758] lr: 2.0500e-04 eta: 11:44:47 time: 0.7501 data_time: 0.0014 memory: 25074 loss: 0.0015 +2023/06/01 16:09:48 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m-lr4e-4_20230601_025223 +2023/06/01 16:10:50 - mmengine - INFO - Epoch(train) [11][3500/5758] lr: 2.0500e-04 eta: 11:43:28 time: 0.7902 data_time: 0.0014 memory: 25074 loss: 0.0011 +2023/06/01 16:12:07 - mmengine - INFO - Epoch(train) [11][3600/5758] lr: 2.0500e-04 eta: 11:42:09 time: 0.7723 data_time: 0.0014 memory: 25074 loss: 0.0016 +2023/06/01 16:13:25 - mmengine - INFO - Epoch(train) [11][3700/5758] lr: 2.0500e-04 eta: 11:40:51 time: 0.7486 data_time: 0.0018 memory: 25074 loss: 0.0022 +2023/06/01 16:14:43 - mmengine - INFO - Epoch(train) [11][3800/5758] lr: 2.0500e-04 eta: 11:39:33 time: 0.7520 data_time: 0.0016 memory: 25074 loss: 0.0042 +2023/06/01 16:15:59 - mmengine - INFO - Epoch(train) [11][3900/5758] lr: 2.0500e-04 eta: 11:38:13 time: 0.7831 data_time: 0.0015 memory: 25074 loss: 0.0076 +2023/06/01 16:17:15 - mmengine - INFO - Epoch(train) [11][4000/5758] lr: 2.0500e-04 eta: 11:36:53 time: 0.7707 data_time: 0.0017 memory: 25074 loss: 0.0030 +2023/06/01 16:18:32 - mmengine - INFO - Epoch(train) [11][4100/5758] lr: 2.0500e-04 eta: 11:35:34 time: 0.7245 data_time: 0.0015 memory: 25074 loss: 0.0006 +2023/06/01 16:19:49 - mmengine - INFO - Epoch(train) [11][4200/5758] lr: 2.0500e-04 eta: 11:34:15 time: 0.7571 data_time: 0.0014 memory: 25074 loss: 0.0015 +2023/06/01 16:21:06 - mmengine - INFO - Epoch(train) [11][4300/5758] lr: 2.0500e-04 eta: 11:32:57 time: 0.7757 data_time: 0.0017 memory: 25074 loss: 0.0057 +2023/06/01 16:22:22 - mmengine - INFO - Epoch(train) [11][4400/5758] lr: 2.0500e-04 eta: 11:31:37 time: 0.7608 data_time: 0.0014 memory: 25074 loss: 0.0038 +2023/06/01 16:22:37 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m-lr4e-4_20230601_025223 +2023/06/01 16:23:40 - mmengine - INFO - Epoch(train) [11][4500/5758] lr: 2.0500e-04 eta: 11:30:19 time: 0.7441 data_time: 0.0015 memory: 25074 loss: 0.0013 +2023/06/01 16:24:57 - mmengine - INFO - Epoch(train) [11][4600/5758] lr: 2.0500e-04 eta: 11:29:00 time: 0.7970 data_time: 0.0014 memory: 25074 loss: 0.0009 +2023/06/01 16:26:12 - mmengine - INFO - Epoch(train) [11][4700/5758] lr: 2.0500e-04 eta: 11:27:39 time: 0.7249 data_time: 0.0013 memory: 25074 loss: 0.0046 +2023/06/01 16:27:30 - mmengine - INFO - Epoch(train) [11][4800/5758] lr: 2.0500e-04 eta: 11:26:21 time: 0.7977 data_time: 0.0014 memory: 25074 loss: 0.0057 +2023/06/01 16:28:46 - mmengine - INFO - Epoch(train) [11][4900/5758] lr: 2.0500e-04 eta: 11:25:01 time: 0.7627 data_time: 0.0014 memory: 25074 loss: 0.0024 +2023/06/01 16:30:05 - mmengine - INFO - Epoch(train) [11][5000/5758] lr: 2.0500e-04 eta: 11:23:44 time: 0.8184 data_time: 0.0013 memory: 25074 loss: 0.0018 +2023/06/01 16:31:24 - mmengine - INFO - Epoch(train) [11][5100/5758] lr: 2.0500e-04 eta: 11:22:27 time: 0.8354 data_time: 0.0015 memory: 25074 loss: 0.0025 +2023/06/01 16:32:42 - mmengine - INFO - Epoch(train) [11][5200/5758] lr: 2.0500e-04 eta: 11:21:09 time: 0.7609 data_time: 0.0017 memory: 25074 loss: 0.0015 +2023/06/01 16:33:58 - mmengine - INFO - Epoch(train) [11][5300/5758] lr: 2.0500e-04 eta: 11:19:49 time: 0.7798 data_time: 0.0017 memory: 25074 loss: 0.0011 +2023/06/01 16:35:16 - mmengine - INFO - Epoch(train) [11][5400/5758] lr: 2.0500e-04 eta: 11:18:31 time: 0.7652 data_time: 0.0015 memory: 25074 loss: 0.0006 +2023/06/01 16:35:31 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m-lr4e-4_20230601_025223 +2023/06/01 16:36:34 - mmengine - INFO - Epoch(train) [11][5500/5758] lr: 2.0500e-04 eta: 11:17:13 time: 0.7664 data_time: 0.0015 memory: 25074 loss: 0.0052 +2023/06/01 16:37:49 - mmengine - INFO - Epoch(train) [11][5600/5758] lr: 2.0500e-04 eta: 11:15:53 time: 0.7844 data_time: 0.0014 memory: 25074 loss: 0.0014 +2023/06/01 16:39:08 - mmengine - INFO - Epoch(train) [11][5700/5758] lr: 2.0500e-04 eta: 11:14:35 time: 0.7780 data_time: 0.0014 memory: 25074 loss: 0.0015 +2023/06/01 16:39:51 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m-lr4e-4_20230601_025223 +2023/06/01 16:39:51 - mmengine - INFO - Saving checkpoint at 11 epochs +2023/06/01 16:40:10 - mmengine - INFO - Epoch(val) [11][16/16] accuracy/top1: 99.9874 single-label/precision_classwise: [100.0, 0.0] single-label/recall_classwise: [99.98737335205078, 0.0] single-label/f1-score_classwise: [99.99369049072266, 0.0] data_time: 0.2366 time: 0.6225 +2023/06/01 16:41:35 - mmengine - INFO - Epoch(train) [12][ 100/5758] lr: 1.7450e-04 eta: 11:12:37 time: 0.7773 data_time: 0.0014 memory: 25074 loss: 0.0006 +2023/06/01 16:42:54 - mmengine - INFO - Epoch(train) [12][ 200/5758] lr: 1.7450e-04 eta: 11:11:19 time: 0.7790 data_time: 0.0018 memory: 25074 loss: 0.0028 +2023/06/01 16:44:11 - mmengine - INFO - Epoch(train) [12][ 300/5758] lr: 1.7450e-04 eta: 11:10:00 time: 0.7833 data_time: 0.0540 memory: 25074 loss: 0.0004 +2023/06/01 16:45:29 - mmengine - INFO - Epoch(train) [12][ 400/5758] lr: 1.7450e-04 eta: 11:08:42 time: 0.8103 data_time: 0.1059 memory: 25074 loss: 0.0022 +2023/06/01 16:46:45 - mmengine - INFO - Epoch(train) [12][ 500/5758] lr: 1.7450e-04 eta: 11:07:23 time: 0.7739 data_time: 0.0559 memory: 25074 loss: 0.0029 +2023/06/01 16:48:01 - mmengine - INFO - Epoch(train) [12][ 600/5758] lr: 1.7450e-04 eta: 11:06:03 time: 0.7308 data_time: 0.0249 memory: 25074 loss: 0.0027 +2023/06/01 16:48:50 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m-lr4e-4_20230601_025223 +2023/06/01 16:49:19 - mmengine - INFO - Epoch(train) [12][ 700/5758] lr: 1.7450e-04 eta: 11:04:45 time: 0.7611 data_time: 0.0534 memory: 25074 loss: 0.0006 +2023/06/01 16:50:36 - mmengine - INFO - Epoch(train) [12][ 800/5758] lr: 1.7450e-04 eta: 11:03:26 time: 0.7511 data_time: 0.0467 memory: 25074 loss: 0.0009 +2023/06/01 16:51:53 - mmengine - INFO - Epoch(train) [12][ 900/5758] lr: 1.7450e-04 eta: 11:02:07 time: 0.7555 data_time: 0.0375 memory: 25074 loss: 0.0013 +2023/06/01 16:53:11 - mmengine - INFO - Epoch(train) [12][1000/5758] lr: 1.7450e-04 eta: 11:00:49 time: 0.7859 data_time: 0.0784 memory: 25074 loss: 0.0029 +2023/06/01 16:54:28 - mmengine - INFO - Epoch(train) [12][1100/5758] lr: 1.7450e-04 eta: 10:59:30 time: 0.7688 data_time: 0.0529 memory: 25074 loss: 0.0016 +2023/06/01 16:55:44 - mmengine - INFO - Epoch(train) [12][1200/5758] lr: 1.7450e-04 eta: 10:58:11 time: 0.7536 data_time: 0.0493 memory: 25074 loss: 0.0030 +2023/06/01 16:57:03 - mmengine - INFO - Epoch(train) [12][1300/5758] lr: 1.7450e-04 eta: 10:56:53 time: 0.7556 data_time: 0.0014 memory: 25074 loss: 0.0011 +2023/06/01 16:58:20 - mmengine - INFO - Epoch(train) [12][1400/5758] lr: 1.7450e-04 eta: 10:55:35 time: 0.7934 data_time: 0.0015 memory: 25074 loss: 0.0008 +2023/06/01 16:59:39 - mmengine - INFO - Epoch(train) [12][1500/5758] lr: 1.7450e-04 eta: 10:54:18 time: 0.7839 data_time: 0.0014 memory: 25074 loss: 0.0023 +2023/06/01 17:00:59 - mmengine - INFO - Epoch(train) [12][1600/5758] lr: 1.7450e-04 eta: 10:53:01 time: 0.7486 data_time: 0.0015 memory: 25074 loss: 0.0002 +2023/06/01 17:01:46 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m-lr4e-4_20230601_025223 +2023/06/01 17:02:16 - mmengine - INFO - Epoch(train) [12][1700/5758] lr: 1.7450e-04 eta: 10:51:42 time: 0.8463 data_time: 0.0020 memory: 25074 loss: 0.0014 +2023/06/01 17:03:35 - mmengine - INFO - Epoch(train) [12][1800/5758] lr: 1.7450e-04 eta: 10:50:25 time: 0.7678 data_time: 0.0015 memory: 25074 loss: 0.0006 +2023/06/01 17:04:53 - mmengine - INFO - Epoch(train) [12][1900/5758] lr: 1.7450e-04 eta: 10:49:07 time: 0.7692 data_time: 0.0015 memory: 25074 loss: 0.0034 +2023/06/01 17:06:10 - mmengine - INFO - Epoch(train) [12][2000/5758] lr: 1.7450e-04 eta: 10:47:48 time: 0.8049 data_time: 0.0016 memory: 25074 loss: 0.0039 +2023/06/01 17:07:26 - mmengine - INFO - Epoch(train) [12][2100/5758] lr: 1.7450e-04 eta: 10:46:29 time: 0.7503 data_time: 0.0014 memory: 25074 loss: 0.0023 +2023/06/01 17:08:44 - mmengine - INFO - Epoch(train) [12][2200/5758] lr: 1.7450e-04 eta: 10:45:10 time: 0.7604 data_time: 0.0014 memory: 25074 loss: 0.0012 +2023/06/01 17:10:01 - mmengine - INFO - Epoch(train) [12][2300/5758] lr: 1.7450e-04 eta: 10:43:52 time: 0.7409 data_time: 0.0014 memory: 25074 loss: 0.0016 +2023/06/01 17:11:20 - mmengine - INFO - Epoch(train) [12][2400/5758] lr: 1.7450e-04 eta: 10:42:34 time: 0.8177 data_time: 0.0016 memory: 25074 loss: 0.0014 +2023/06/01 17:12:36 - mmengine - INFO - Epoch(train) [12][2500/5758] lr: 1.7450e-04 eta: 10:41:15 time: 0.7445 data_time: 0.0019 memory: 25074 loss: 0.0037 +2023/06/01 17:13:54 - mmengine - INFO - Epoch(train) [12][2600/5758] lr: 1.7450e-04 eta: 10:39:57 time: 0.7504 data_time: 0.0017 memory: 25074 loss: 0.0014 +2023/06/01 17:14:42 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m-lr4e-4_20230601_025223 +2023/06/01 17:15:11 - mmengine - INFO - Epoch(train) [12][2700/5758] lr: 1.7450e-04 eta: 10:38:38 time: 0.7381 data_time: 0.0016 memory: 25074 loss: 0.0013 +2023/06/01 17:16:27 - mmengine - INFO - Epoch(train) [12][2800/5758] lr: 1.7450e-04 eta: 10:37:18 time: 0.7505 data_time: 0.0014 memory: 25074 loss: 0.0027 +2023/06/01 17:17:43 - mmengine - INFO - Epoch(train) [12][2900/5758] lr: 1.7450e-04 eta: 10:35:59 time: 0.7413 data_time: 0.0013 memory: 25074 loss: 0.0036 +2023/06/01 17:19:00 - mmengine - INFO - Epoch(train) [12][3000/5758] lr: 1.7450e-04 eta: 10:34:40 time: 0.8202 data_time: 0.0014 memory: 25074 loss: 0.0002 +2023/06/01 17:20:19 - mmengine - INFO - Epoch(train) [12][3100/5758] lr: 1.7450e-04 eta: 10:33:23 time: 0.8542 data_time: 0.0017 memory: 25074 loss: 0.0040 +2023/06/01 17:21:35 - mmengine - INFO - Epoch(train) [12][3200/5758] lr: 1.7450e-04 eta: 10:32:04 time: 0.7686 data_time: 0.0015 memory: 25074 loss: 0.0009 +2023/06/01 17:22:53 - mmengine - INFO - Epoch(train) [12][3300/5758] lr: 1.7450e-04 eta: 10:30:45 time: 0.7899 data_time: 0.0013 memory: 25074 loss: 0.0009 +2023/06/01 17:24:11 - mmengine - INFO - Epoch(train) [12][3400/5758] lr: 1.7450e-04 eta: 10:29:27 time: 0.8040 data_time: 0.0015 memory: 25074 loss: 0.0004 +2023/06/01 17:25:27 - mmengine - INFO - Epoch(train) [12][3500/5758] lr: 1.7450e-04 eta: 10:28:08 time: 0.7677 data_time: 0.0014 memory: 25074 loss: 0.0051 +2023/06/01 17:26:45 - mmengine - INFO - Epoch(train) [12][3600/5758] lr: 1.7450e-04 eta: 10:26:50 time: 0.7757 data_time: 0.0013 memory: 25074 loss: 0.0045 +2023/06/01 17:27:36 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m-lr4e-4_20230601_025223 +2023/06/01 17:28:06 - mmengine - INFO - Epoch(train) [12][3700/5758] lr: 1.7450e-04 eta: 10:25:34 time: 0.7820 data_time: 0.0015 memory: 25074 loss: 0.0024 +2023/06/01 17:29:24 - mmengine - INFO - Epoch(train) [12][3800/5758] lr: 1.7450e-04 eta: 10:24:17 time: 0.8002 data_time: 0.0015 memory: 25074 loss: 0.0045 +2023/06/01 17:30:44 - mmengine - INFO - Epoch(train) [12][3900/5758] lr: 1.7450e-04 eta: 10:23:00 time: 0.8138 data_time: 0.0015 memory: 25074 loss: 0.0089 +2023/06/01 17:32:03 - mmengine - INFO - Epoch(train) [12][4000/5758] lr: 1.7450e-04 eta: 10:21:43 time: 0.7602 data_time: 0.0017 memory: 25074 loss: 0.0016 +2023/06/01 17:33:24 - mmengine - INFO - Epoch(train) [12][4100/5758] lr: 1.7450e-04 eta: 10:20:27 time: 0.8197 data_time: 0.0017 memory: 25074 loss: 0.0015 +2023/06/01 17:34:42 - mmengine - INFO - Epoch(train) [12][4200/5758] lr: 1.7450e-04 eta: 10:19:08 time: 0.7985 data_time: 0.0015 memory: 25074 loss: 0.0026 +2023/06/01 17:36:01 - mmengine - INFO - Epoch(train) [12][4300/5758] lr: 1.7450e-04 eta: 10:17:51 time: 0.7773 data_time: 0.0014 memory: 25074 loss: 0.0041 +2023/06/01 17:37:21 - mmengine - INFO - Epoch(train) [12][4400/5758] lr: 1.7450e-04 eta: 10:16:35 time: 0.7709 data_time: 0.0020 memory: 25074 loss: 0.0006 +2023/06/01 17:38:39 - mmengine - INFO - Epoch(train) [12][4500/5758] lr: 1.7450e-04 eta: 10:15:17 time: 0.7850 data_time: 0.0017 memory: 25074 loss: 0.0020 +2023/06/01 17:39:58 - mmengine - INFO - Epoch(train) [12][4600/5758] lr: 1.7450e-04 eta: 10:13:59 time: 0.7553 data_time: 0.0013 memory: 25074 loss: 0.0010 +2023/06/01 17:40:47 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m-lr4e-4_20230601_025223 +2023/06/01 17:41:15 - mmengine - INFO - Epoch(train) [12][4700/5758] lr: 1.7450e-04 eta: 10:12:41 time: 0.7943 data_time: 0.0015 memory: 25074 loss: 0.0018 +2023/06/01 17:42:34 - mmengine - INFO - Epoch(train) [12][4800/5758] lr: 1.7450e-04 eta: 10:11:23 time: 0.7821 data_time: 0.0015 memory: 25074 loss: 0.0019 +2023/06/01 17:43:51 - mmengine - INFO - Epoch(train) [12][4900/5758] lr: 1.7450e-04 eta: 10:10:04 time: 0.7713 data_time: 0.0014 memory: 25074 loss: 0.0037 +2023/06/01 17:45:10 - mmengine - INFO - Epoch(train) [12][5000/5758] lr: 1.7450e-04 eta: 10:08:47 time: 0.8314 data_time: 0.0013 memory: 25074 loss: 0.0011 +2023/06/01 17:46:31 - mmengine - INFO - Epoch(train) [12][5100/5758] lr: 1.7450e-04 eta: 10:07:31 time: 0.8069 data_time: 0.0013 memory: 25074 loss: 0.0035 +2023/06/01 17:47:53 - mmengine - INFO - Epoch(train) [12][5200/5758] lr: 1.7450e-04 eta: 10:06:16 time: 0.8398 data_time: 0.0022 memory: 25074 loss: 0.0011 +2023/06/01 17:49:15 - mmengine - INFO - Epoch(train) [12][5300/5758] lr: 1.7450e-04 eta: 10:05:00 time: 0.7985 data_time: 0.0017 memory: 25074 loss: 0.0003 +2023/06/01 17:50:35 - mmengine - INFO - Epoch(train) [12][5400/5758] lr: 1.7450e-04 eta: 10:03:44 time: 0.8258 data_time: 0.0015 memory: 25074 loss: 0.0013 +2023/06/01 17:51:53 - mmengine - INFO - Epoch(train) [12][5500/5758] lr: 1.7450e-04 eta: 10:02:26 time: 0.7689 data_time: 0.0016 memory: 25074 loss: 0.0009 +2023/06/01 17:53:14 - mmengine - INFO - Epoch(train) [12][5600/5758] lr: 1.7450e-04 eta: 10:01:10 time: 0.7946 data_time: 0.0018 memory: 25074 loss: 0.0008 +2023/06/01 17:54:03 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m-lr4e-4_20230601_025223 +2023/06/01 17:54:33 - mmengine - INFO - Epoch(train) [12][5700/5758] lr: 1.7450e-04 eta: 9:59:52 time: 0.7336 data_time: 0.0019 memory: 25074 loss: 0.0023 +2023/06/01 17:55:20 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m-lr4e-4_20230601_025223 +2023/06/01 17:55:20 - mmengine - INFO - Saving checkpoint at 12 epochs +2023/06/01 17:55:39 - mmengine - INFO - Epoch(val) [12][16/16] accuracy/top1: 99.9937 single-label/precision_classwise: [100.0, 0.0] single-label/recall_classwise: [99.99369049072266, 0.0] single-label/f1-score_classwise: [99.9968490600586, 0.0] data_time: 0.2499 time: 0.6355 +2023/06/01 17:57:04 - mmengine - INFO - Epoch(train) [13][ 100/5758] lr: 1.4474e-04 eta: 9:57:55 time: 0.8509 data_time: 0.0728 memory: 25074 loss: 0.0009 +2023/06/01 17:58:21 - mmengine - INFO - Epoch(train) [13][ 200/5758] lr: 1.4474e-04 eta: 9:56:36 time: 0.7632 data_time: 0.0024 memory: 25074 loss: 0.0006 +2023/06/01 17:59:42 - mmengine - INFO - Epoch(train) [13][ 300/5758] lr: 1.4474e-04 eta: 9:55:20 time: 0.7639 data_time: 0.0500 memory: 25074 loss: 0.0003 +2023/06/01 18:00:59 - mmengine - INFO - Epoch(train) [13][ 400/5758] lr: 1.4474e-04 eta: 9:54:01 time: 0.7331 data_time: 0.0250 memory: 25074 loss: 0.0013 +2023/06/01 18:02:20 - mmengine - INFO - Epoch(train) [13][ 500/5758] lr: 1.4474e-04 eta: 9:52:45 time: 0.8057 data_time: 0.1014 memory: 25074 loss: 0.0023 +2023/06/01 18:03:33 - mmengine - INFO - Epoch(train) [13][ 600/5758] lr: 1.4474e-04 eta: 9:51:24 time: 0.7540 data_time: 0.0015 memory: 25074 loss: 0.0023 +2023/06/01 18:04:48 - mmengine - INFO - Epoch(train) [13][ 700/5758] lr: 1.4474e-04 eta: 9:50:04 time: 0.7368 data_time: 0.0016 memory: 25074 loss: 0.0020 +2023/06/01 18:06:03 - mmengine - INFO - Epoch(train) [13][ 800/5758] lr: 1.4474e-04 eta: 9:48:44 time: 0.7242 data_time: 0.0017 memory: 25074 loss: 0.0014 +2023/06/01 18:07:17 - mmengine - INFO - Epoch(train) [13][ 900/5758] lr: 1.4474e-04 eta: 9:47:23 time: 0.7239 data_time: 0.0014 memory: 25074 loss: 0.0004 +2023/06/01 18:07:20 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m-lr4e-4_20230601_025223 +2023/06/01 18:08:32 - mmengine - INFO - Epoch(train) [13][1000/5758] lr: 1.4474e-04 eta: 9:46:03 time: 0.7724 data_time: 0.0018 memory: 25074 loss: 0.0017 +2023/06/01 18:09:49 - mmengine - INFO - Epoch(train) [13][1100/5758] lr: 1.4474e-04 eta: 9:44:44 time: 0.7634 data_time: 0.0018 memory: 25074 loss: 0.0008 +2023/06/01 18:11:06 - mmengine - INFO - Epoch(train) [13][1200/5758] lr: 1.4474e-04 eta: 9:43:26 time: 0.7473 data_time: 0.0023 memory: 25074 loss: 0.0024 +2023/06/01 18:12:24 - mmengine - INFO - Epoch(train) [13][1300/5758] lr: 1.4474e-04 eta: 9:42:08 time: 0.7653 data_time: 0.0021 memory: 25074 loss: 0.0040 +2023/06/01 18:13:40 - mmengine - INFO - Epoch(train) [13][1400/5758] lr: 1.4474e-04 eta: 9:40:49 time: 0.7782 data_time: 0.0019 memory: 25074 loss: 0.0015 +2023/06/01 18:14:56 - mmengine - INFO - Epoch(train) [13][1500/5758] lr: 1.4474e-04 eta: 9:39:29 time: 0.7193 data_time: 0.0021 memory: 25074 loss: 0.0017 +2023/06/01 18:16:16 - mmengine - INFO - Epoch(train) [13][1600/5758] lr: 1.4474e-04 eta: 9:38:12 time: 0.7994 data_time: 0.0018 memory: 25074 loss: 0.0042 +2023/06/01 18:17:30 - mmengine - INFO - Epoch(train) [13][1700/5758] lr: 1.4474e-04 eta: 9:36:52 time: 0.7312 data_time: 0.0025 memory: 25074 loss: 0.0013 +2023/06/01 18:18:45 - mmengine - INFO - Epoch(train) [13][1800/5758] lr: 1.4474e-04 eta: 9:35:32 time: 0.7531 data_time: 0.0018 memory: 25074 loss: 0.0006 +2023/06/01 18:20:01 - mmengine - INFO - Epoch(train) [13][1900/5758] lr: 1.4474e-04 eta: 9:34:13 time: 0.7860 data_time: 0.0024 memory: 25074 loss: 0.0020 +2023/06/01 18:20:05 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m-lr4e-4_20230601_025223 +2023/06/01 18:21:17 - mmengine - INFO - Epoch(train) [13][2000/5758] lr: 1.4474e-04 eta: 9:32:53 time: 0.7634 data_time: 0.0018 memory: 25074 loss: 0.0008 +2023/06/01 18:22:33 - mmengine - INFO - Epoch(train) [13][2100/5758] lr: 1.4474e-04 eta: 9:31:34 time: 0.7482 data_time: 0.0023 memory: 25074 loss: 0.0016 +2023/06/01 18:23:49 - mmengine - INFO - Epoch(train) [13][2200/5758] lr: 1.4474e-04 eta: 9:30:15 time: 0.7663 data_time: 0.0018 memory: 25074 loss: 0.0003 +2023/06/01 18:25:08 - mmengine - INFO - Epoch(train) [13][2300/5758] lr: 1.4474e-04 eta: 9:28:57 time: 0.7855 data_time: 0.0014 memory: 25074 loss: 0.0008 +2023/06/01 18:26:24 - mmengine - INFO - Epoch(train) [13][2400/5758] lr: 1.4474e-04 eta: 9:27:38 time: 0.7426 data_time: 0.0014 memory: 25074 loss: 0.0012 +2023/06/01 18:27:40 - mmengine - INFO - Epoch(train) [13][2500/5758] lr: 1.4474e-04 eta: 9:26:19 time: 0.7954 data_time: 0.0013 memory: 25074 loss: 0.0031 +2023/06/01 18:28:59 - mmengine - INFO - Epoch(train) [13][2600/5758] lr: 1.4474e-04 eta: 9:25:01 time: 0.7913 data_time: 0.0018 memory: 25074 loss: 0.0013 +2023/06/01 18:30:16 - mmengine - INFO - Epoch(train) [13][2700/5758] lr: 1.4474e-04 eta: 9:23:43 time: 0.7682 data_time: 0.0015 memory: 25074 loss: 0.0010 +2023/06/01 18:31:32 - mmengine - INFO - Epoch(train) [13][2800/5758] lr: 1.4474e-04 eta: 9:22:24 time: 0.7579 data_time: 0.0017 memory: 25074 loss: 0.0006 +2023/06/01 18:32:49 - mmengine - INFO - Epoch(train) [13][2900/5758] lr: 1.4474e-04 eta: 9:21:05 time: 0.7687 data_time: 0.0017 memory: 25074 loss: 0.0026 +2023/06/01 18:32:52 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m-lr4e-4_20230601_025223 +2023/06/01 18:34:05 - mmengine - INFO - Epoch(train) [13][3000/5758] lr: 1.4474e-04 eta: 9:19:46 time: 0.7707 data_time: 0.0016 memory: 25074 loss: 0.0031 +2023/06/01 18:35:24 - mmengine - INFO - Epoch(train) [13][3100/5758] lr: 1.4474e-04 eta: 9:18:28 time: 0.7957 data_time: 0.0013 memory: 25074 loss: 0.0025 +2023/06/01 18:36:40 - mmengine - INFO - Epoch(train) [13][3200/5758] lr: 1.4474e-04 eta: 9:17:09 time: 0.7365 data_time: 0.0013 memory: 25074 loss: 0.0011 +2023/06/01 18:37:55 - mmengine - INFO - Epoch(train) [13][3300/5758] lr: 1.4474e-04 eta: 9:15:49 time: 0.7628 data_time: 0.0016 memory: 25074 loss: 0.0019 +2023/06/01 18:39:10 - mmengine - INFO - Epoch(train) [13][3400/5758] lr: 1.4474e-04 eta: 9:14:29 time: 0.7674 data_time: 0.0017 memory: 25074 loss: 0.0021 +2023/06/01 18:40:25 - mmengine - INFO - Epoch(train) [13][3500/5758] lr: 1.4474e-04 eta: 9:13:10 time: 0.7265 data_time: 0.0015 memory: 25074 loss: 0.0008 +2023/06/01 18:41:41 - mmengine - INFO - Epoch(train) [13][3600/5758] lr: 1.4474e-04 eta: 9:11:51 time: 0.7296 data_time: 0.0014 memory: 25074 loss: 0.0026 +2023/06/01 18:42:54 - mmengine - INFO - Epoch(train) [13][3700/5758] lr: 1.4474e-04 eta: 9:10:30 time: 0.7435 data_time: 0.0016 memory: 25074 loss: 0.0018 +2023/06/01 18:44:07 - mmengine - INFO - Epoch(train) [13][3800/5758] lr: 1.4474e-04 eta: 9:09:09 time: 0.7049 data_time: 0.0013 memory: 25074 loss: 0.0005 +2023/06/01 18:45:20 - mmengine - INFO - Epoch(train) [13][3900/5758] lr: 1.4474e-04 eta: 9:07:48 time: 0.7305 data_time: 0.0014 memory: 25074 loss: 0.0013 +2023/06/01 18:45:23 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m-lr4e-4_20230601_025223 +2023/06/01 18:46:33 - mmengine - INFO - Epoch(train) [13][4000/5758] lr: 1.4474e-04 eta: 9:06:27 time: 0.7381 data_time: 0.0015 memory: 25074 loss: 0.0009 +2023/06/01 18:47:46 - mmengine - INFO - Epoch(train) [13][4100/5758] lr: 1.4474e-04 eta: 9:05:07 time: 0.7095 data_time: 0.0015 memory: 25074 loss: 0.0038 +2023/06/01 18:49:01 - mmengine - INFO - Epoch(train) [13][4200/5758] lr: 1.4474e-04 eta: 9:03:47 time: 0.7096 data_time: 0.0014 memory: 25074 loss: 0.0012 +2023/06/01 18:50:14 - mmengine - INFO - Epoch(train) [13][4300/5758] lr: 1.4474e-04 eta: 9:02:26 time: 0.7412 data_time: 0.0015 memory: 25074 loss: 0.0007 +2023/06/01 18:51:28 - mmengine - INFO - Epoch(train) [13][4400/5758] lr: 1.4474e-04 eta: 9:01:06 time: 0.7486 data_time: 0.0017 memory: 25074 loss: 0.0051 +2023/06/01 18:52:42 - mmengine - INFO - Epoch(train) [13][4500/5758] lr: 1.4474e-04 eta: 8:59:46 time: 0.7732 data_time: 0.0016 memory: 25074 loss: 0.0001 +2023/06/01 18:54:00 - mmengine - INFO - Epoch(train) [13][4600/5758] lr: 1.4474e-04 eta: 8:58:28 time: 0.7479 data_time: 0.0015 memory: 25074 loss: 0.0020 +2023/06/01 18:55:17 - mmengine - INFO - Epoch(train) [13][4700/5758] lr: 1.4474e-04 eta: 8:57:09 time: 0.7430 data_time: 0.0013 memory: 25074 loss: 0.0010 +2023/06/01 18:56:30 - mmengine - INFO - Epoch(train) [13][4800/5758] lr: 1.4474e-04 eta: 8:55:49 time: 0.7274 data_time: 0.0016 memory: 25074 loss: 0.0010 +2023/06/01 18:57:45 - mmengine - INFO - Epoch(train) [13][4900/5758] lr: 1.4474e-04 eta: 8:54:30 time: 0.7867 data_time: 0.0015 memory: 25074 loss: 0.0011 +2023/06/01 18:57:48 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m-lr4e-4_20230601_025223 +2023/06/01 18:58:57 - mmengine - INFO - Epoch(train) [13][5000/5758] lr: 1.4474e-04 eta: 8:53:08 time: 0.7084 data_time: 0.0013 memory: 25074 loss: 0.0027 +2023/06/01 19:00:11 - mmengine - INFO - Epoch(train) [13][5100/5758] lr: 1.4474e-04 eta: 8:51:48 time: 0.8714 data_time: 0.0017 memory: 25074 loss: 0.0033 +2023/06/01 19:01:23 - mmengine - INFO - Epoch(train) [13][5200/5758] lr: 1.4474e-04 eta: 8:50:27 time: 0.7066 data_time: 0.0015 memory: 25074 loss: 0.0031 +2023/06/01 19:02:38 - mmengine - INFO - Epoch(train) [13][5300/5758] lr: 1.4474e-04 eta: 8:49:08 time: 0.7848 data_time: 0.0012 memory: 25074 loss: 0.0007 +2023/06/01 19:03:53 - mmengine - INFO - Epoch(train) [13][5400/5758] lr: 1.4474e-04 eta: 8:47:48 time: 0.7287 data_time: 0.0016 memory: 25074 loss: 0.0004 +2023/06/01 19:05:06 - mmengine - INFO - Epoch(train) [13][5500/5758] lr: 1.4474e-04 eta: 8:46:27 time: 0.7446 data_time: 0.0017 memory: 25074 loss: 0.0009 +2023/06/01 19:06:20 - mmengine - INFO - Epoch(train) [13][5600/5758] lr: 1.4474e-04 eta: 8:45:08 time: 0.7054 data_time: 0.0016 memory: 25074 loss: 0.0058 +2023/06/01 19:07:36 - mmengine - INFO - Epoch(train) [13][5700/5758] lr: 1.4474e-04 eta: 8:43:49 time: 0.7207 data_time: 0.0015 memory: 25074 loss: 0.0018 +2023/06/01 19:08:18 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m-lr4e-4_20230601_025223 +2023/06/01 19:08:18 - mmengine - INFO - Saving checkpoint at 13 epochs +2023/06/01 19:08:37 - mmengine - INFO - Epoch(val) [13][16/16] accuracy/top1: 99.9937 single-label/precision_classwise: [100.0, 0.0] single-label/recall_classwise: [99.99369049072266, 0.0] single-label/f1-score_classwise: [99.9968490600586, 0.0] data_time: 0.2429 time: 0.6289 +2023/06/01 19:10:02 - mmengine - INFO - Epoch(train) [14][ 100/5758] lr: 1.1647e-04 eta: 8:41:48 time: 0.7784 data_time: 0.0233 memory: 25074 loss: 0.0022 +2023/06/01 19:10:37 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m-lr4e-4_20230601_025223 +2023/06/01 19:11:17 - mmengine - INFO - Epoch(train) [14][ 200/5758] lr: 1.1647e-04 eta: 8:40:29 time: 0.7372 data_time: 0.0015 memory: 25074 loss: 0.0024 +2023/06/01 19:12:33 - mmengine - INFO - Epoch(train) [14][ 300/5758] lr: 1.1647e-04 eta: 8:39:09 time: 0.7058 data_time: 0.0015 memory: 25074 loss: 0.0009 +2023/06/01 19:13:47 - mmengine - INFO - Epoch(train) [14][ 400/5758] lr: 1.1647e-04 eta: 8:37:50 time: 0.7190 data_time: 0.0019 memory: 25074 loss: 0.0012 +2023/06/01 19:15:01 - mmengine - INFO - Epoch(train) [14][ 500/5758] lr: 1.1647e-04 eta: 8:36:30 time: 0.7497 data_time: 0.0456 memory: 25074 loss: 0.0014 +2023/06/01 19:16:13 - mmengine - INFO - Epoch(train) [14][ 600/5758] lr: 1.1647e-04 eta: 8:35:09 time: 0.7059 data_time: 0.0013 memory: 25074 loss: 0.0030 +2023/06/01 19:17:25 - mmengine - INFO - Epoch(train) [14][ 700/5758] lr: 1.1647e-04 eta: 8:33:48 time: 0.7181 data_time: 0.0144 memory: 25074 loss: 0.0014 +2023/06/01 19:18:37 - mmengine - INFO - Epoch(train) [14][ 800/5758] lr: 1.1647e-04 eta: 8:32:27 time: 0.7468 data_time: 0.0424 memory: 25074 loss: 0.0012 +2023/06/01 19:19:51 - mmengine - INFO - Epoch(train) [14][ 900/5758] lr: 1.1647e-04 eta: 8:31:07 time: 0.7151 data_time: 0.0016 memory: 25074 loss: 0.0001 +2023/06/01 19:21:05 - mmengine - INFO - Epoch(train) [14][1000/5758] lr: 1.1647e-04 eta: 8:29:47 time: 0.7630 data_time: 0.0015 memory: 25074 loss: 0.0037 +2023/06/01 19:22:18 - mmengine - INFO - Epoch(train) [14][1100/5758] lr: 1.1647e-04 eta: 8:28:27 time: 0.7464 data_time: 0.0018 memory: 25074 loss: 0.0008 +2023/06/01 19:22:53 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m-lr4e-4_20230601_025223 +2023/06/01 19:23:34 - mmengine - INFO - Epoch(train) [14][1200/5758] lr: 1.1647e-04 eta: 8:27:08 time: 0.7901 data_time: 0.0020 memory: 25074 loss: 0.0010 +2023/06/01 19:24:46 - mmengine - INFO - Epoch(train) [14][1300/5758] lr: 1.1647e-04 eta: 8:25:48 time: 0.7513 data_time: 0.0014 memory: 25074 loss: 0.0006 +2023/06/01 19:25:59 - mmengine - INFO - Epoch(train) [14][1400/5758] lr: 1.1647e-04 eta: 8:24:27 time: 0.7353 data_time: 0.0014 memory: 25074 loss: 0.0002 +2023/06/01 19:27:12 - mmengine - INFO - Epoch(train) [14][1500/5758] lr: 1.1647e-04 eta: 8:23:07 time: 0.7191 data_time: 0.0016 memory: 25074 loss: 0.0027 +2023/06/01 19:28:27 - mmengine - INFO - Epoch(train) [14][1600/5758] lr: 1.1647e-04 eta: 8:21:48 time: 0.7148 data_time: 0.0014 memory: 25074 loss: 0.0001 +2023/06/01 19:29:41 - mmengine - INFO - Epoch(train) [14][1700/5758] lr: 1.1647e-04 eta: 8:20:28 time: 0.7152 data_time: 0.0015 memory: 25074 loss: 0.0006 +2023/06/01 19:30:56 - mmengine - INFO - Epoch(train) [14][1800/5758] lr: 1.1647e-04 eta: 8:19:09 time: 0.7560 data_time: 0.0017 memory: 25074 loss: 0.0005 +2023/06/01 19:32:10 - mmengine - INFO - Epoch(train) [14][1900/5758] lr: 1.1647e-04 eta: 8:17:49 time: 0.7273 data_time: 0.0015 memory: 25074 loss: 0.0007 +2023/06/01 19:33:25 - mmengine - INFO - Epoch(train) [14][2000/5758] lr: 1.1647e-04 eta: 8:16:30 time: 0.7825 data_time: 0.0014 memory: 25074 loss: 0.0017 +2023/06/01 19:34:41 - mmengine - INFO - Epoch(train) [14][2100/5758] lr: 1.1647e-04 eta: 8:15:11 time: 0.7625 data_time: 0.0014 memory: 25074 loss: 0.0003 +2023/06/01 19:35:16 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m-lr4e-4_20230601_025223 +2023/06/01 19:35:56 - mmengine - INFO - Epoch(train) [14][2200/5758] lr: 1.1647e-04 eta: 8:13:52 time: 0.8063 data_time: 0.0014 memory: 25074 loss: 0.0029 +2023/06/01 19:37:14 - mmengine - INFO - Epoch(train) [14][2300/5758] lr: 1.1647e-04 eta: 8:12:34 time: 0.7869 data_time: 0.0014 memory: 25074 loss: 0.0007 +2023/06/01 19:38:32 - mmengine - INFO - Epoch(train) [14][2400/5758] lr: 1.1647e-04 eta: 8:11:17 time: 0.8502 data_time: 0.0014 memory: 25074 loss: 0.0013 +2023/06/01 19:39:50 - mmengine - INFO - Epoch(train) [14][2500/5758] lr: 1.1647e-04 eta: 8:09:59 time: 0.7570 data_time: 0.0015 memory: 25074 loss: 0.0014 +2023/06/01 19:41:09 - mmengine - INFO - Epoch(train) [14][2600/5758] lr: 1.1647e-04 eta: 8:08:42 time: 0.8025 data_time: 0.0013 memory: 25074 loss: 0.0017 +2023/06/01 19:42:28 - mmengine - INFO - Epoch(train) [14][2700/5758] lr: 1.1647e-04 eta: 8:07:25 time: 0.8707 data_time: 0.0014 memory: 25074 loss: 0.0019 +2023/06/01 19:43:48 - mmengine - INFO - Epoch(train) [14][2800/5758] lr: 1.1647e-04 eta: 8:06:08 time: 0.8228 data_time: 0.0014 memory: 25074 loss: 0.0015 +2023/06/01 19:45:08 - mmengine - INFO - Epoch(train) [14][2900/5758] lr: 1.1647e-04 eta: 8:04:51 time: 0.7924 data_time: 0.0014 memory: 25074 loss: 0.0032 +2023/06/01 19:46:27 - mmengine - INFO - Epoch(train) [14][3000/5758] lr: 1.1647e-04 eta: 8:03:34 time: 0.7878 data_time: 0.0013 memory: 25074 loss: 0.0011 +2023/06/01 19:47:46 - mmengine - INFO - Epoch(train) [14][3100/5758] lr: 1.1647e-04 eta: 8:02:17 time: 0.7608 data_time: 0.0014 memory: 25074 loss: 0.0008 +2023/06/01 19:48:24 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m-lr4e-4_20230601_025223 +2023/06/01 19:49:07 - mmengine - INFO - Epoch(train) [14][3200/5758] lr: 1.1647e-04 eta: 8:01:01 time: 0.8227 data_time: 0.0014 memory: 25074 loss: 0.0009 +2023/06/01 19:50:25 - mmengine - INFO - Epoch(train) [14][3300/5758] lr: 1.1647e-04 eta: 7:59:43 time: 0.7711 data_time: 0.0017 memory: 25074 loss: 0.0001 +2023/06/01 19:51:44 - mmengine - INFO - Epoch(train) [14][3400/5758] lr: 1.1647e-04 eta: 7:58:26 time: 0.7489 data_time: 0.0015 memory: 25074 loss: 0.0063 +2023/06/01 19:53:02 - mmengine - INFO - Epoch(train) [14][3500/5758] lr: 1.1647e-04 eta: 7:57:08 time: 0.7752 data_time: 0.0014 memory: 25074 loss: 0.0017 +2023/06/01 19:54:20 - mmengine - INFO - Epoch(train) [14][3600/5758] lr: 1.1647e-04 eta: 7:55:50 time: 0.7770 data_time: 0.0015 memory: 25074 loss: 0.0027 +2023/06/01 19:55:35 - mmengine - INFO - Epoch(train) [14][3700/5758] lr: 1.1647e-04 eta: 7:54:31 time: 0.7678 data_time: 0.0013 memory: 25074 loss: 0.0002 +2023/06/01 19:56:52 - mmengine - INFO - Epoch(train) [14][3800/5758] lr: 1.1647e-04 eta: 7:53:14 time: 0.7268 data_time: 0.0017 memory: 25074 loss: 0.0027 +2023/06/01 19:58:09 - mmengine - INFO - Epoch(train) [14][3900/5758] lr: 1.1647e-04 eta: 7:51:55 time: 0.7710 data_time: 0.0015 memory: 25074 loss: 0.0023 +2023/06/01 19:59:27 - mmengine - INFO - Epoch(train) [14][4000/5758] lr: 1.1647e-04 eta: 7:50:38 time: 0.8276 data_time: 0.0015 memory: 25074 loss: 0.0016 +2023/06/01 20:00:43 - mmengine - INFO - Epoch(train) [14][4100/5758] lr: 1.1647e-04 eta: 7:49:19 time: 0.7827 data_time: 0.0015 memory: 25074 loss: 0.0010 +2023/06/01 20:01:17 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m-lr4e-4_20230601_025223 +2023/06/01 20:01:58 - mmengine - INFO - Epoch(train) [14][4200/5758] lr: 1.1647e-04 eta: 7:48:00 time: 0.7341 data_time: 0.0014 memory: 25074 loss: 0.0021 +2023/06/01 20:03:15 - mmengine - INFO - Epoch(train) [14][4300/5758] lr: 1.1647e-04 eta: 7:46:42 time: 0.7725 data_time: 0.0015 memory: 25074 loss: 0.0016 +2023/06/01 20:04:33 - mmengine - INFO - Epoch(train) [14][4400/5758] lr: 1.1647e-04 eta: 7:45:24 time: 0.7204 data_time: 0.0013 memory: 25074 loss: 0.0032 +2023/06/01 20:05:51 - mmengine - INFO - Epoch(train) [14][4500/5758] lr: 1.1647e-04 eta: 7:44:07 time: 0.7621 data_time: 0.0015 memory: 25074 loss: 0.0015 +2023/06/01 20:07:07 - mmengine - INFO - Epoch(train) [14][4600/5758] lr: 1.1647e-04 eta: 7:42:48 time: 0.7450 data_time: 0.0015 memory: 25074 loss: 0.0010 +2023/06/01 20:08:24 - mmengine - INFO - Epoch(train) [14][4700/5758] lr: 1.1647e-04 eta: 7:41:30 time: 0.7573 data_time: 0.0015 memory: 25074 loss: 0.0004 +2023/06/01 20:09:40 - mmengine - INFO - Epoch(train) [14][4800/5758] lr: 1.1647e-04 eta: 7:40:11 time: 0.7113 data_time: 0.0014 memory: 25074 loss: 0.0004 +2023/06/01 20:10:56 - mmengine - INFO - Epoch(train) [14][4900/5758] lr: 1.1647e-04 eta: 7:38:52 time: 0.7391 data_time: 0.0014 memory: 25074 loss: 0.0017 +2023/06/01 20:12:11 - mmengine - INFO - Epoch(train) [14][5000/5758] lr: 1.1647e-04 eta: 7:37:34 time: 0.8220 data_time: 0.0014 memory: 25074 loss: 0.0005 +2023/06/01 20:13:27 - mmengine - INFO - Epoch(train) [14][5100/5758] lr: 1.1647e-04 eta: 7:36:15 time: 0.7243 data_time: 0.0014 memory: 25074 loss: 0.0003 +2023/06/01 20:14:02 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m-lr4e-4_20230601_025223 +2023/06/01 20:14:42 - mmengine - INFO - Epoch(train) [14][5200/5758] lr: 1.1647e-04 eta: 7:34:56 time: 0.7185 data_time: 0.0013 memory: 25074 loss: 0.0007 +2023/06/01 20:15:56 - mmengine - INFO - Epoch(train) [14][5300/5758] lr: 1.1647e-04 eta: 7:33:37 time: 0.7506 data_time: 0.0014 memory: 25074 loss: 0.0002 +2023/06/01 20:17:15 - mmengine - INFO - Epoch(train) [14][5400/5758] lr: 1.1647e-04 eta: 7:32:19 time: 0.7409 data_time: 0.0014 memory: 25074 loss: 0.0004 +2023/06/01 20:18:30 - mmengine - INFO - Epoch(train) [14][5500/5758] lr: 1.1647e-04 eta: 7:31:01 time: 0.7215 data_time: 0.0019 memory: 25074 loss: 0.0003 +2023/06/01 20:19:48 - mmengine - INFO - Epoch(train) [14][5600/5758] lr: 1.1647e-04 eta: 7:29:43 time: 0.7352 data_time: 0.0014 memory: 25074 loss: 0.0037 +2023/06/01 20:21:07 - mmengine - INFO - Epoch(train) [14][5700/5758] lr: 1.1647e-04 eta: 7:28:26 time: 0.7899 data_time: 0.0016 memory: 25074 loss: 0.0006 +2023/06/01 20:21:52 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m-lr4e-4_20230601_025223 +2023/06/01 20:21:52 - mmengine - INFO - Saving checkpoint at 14 epochs +2023/06/01 20:22:11 - mmengine - INFO - Epoch(val) [14][16/16] accuracy/top1: 99.9937 single-label/precision_classwise: [100.0, 0.0] single-label/recall_classwise: [99.99369049072266, 0.0] single-label/f1-score_classwise: [99.9968490600586, 0.0] data_time: 0.2443 time: 0.6311 +2023/06/01 20:23:39 - mmengine - INFO - Epoch(train) [15][ 100/5758] lr: 9.0382e-05 eta: 7:26:27 time: 0.8083 data_time: 0.0013 memory: 25074 loss: 0.0010 +2023/06/01 20:24:58 - mmengine - INFO - Epoch(train) [15][ 200/5758] lr: 9.0382e-05 eta: 7:25:10 time: 0.7608 data_time: 0.0014 memory: 25074 loss: 0.0025 +2023/06/01 20:26:17 - mmengine - INFO - Epoch(train) [15][ 300/5758] lr: 9.0382e-05 eta: 7:23:53 time: 0.7824 data_time: 0.0013 memory: 25074 loss: 0.0006 +2023/06/01 20:27:27 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m-lr4e-4_20230601_025223 +2023/06/01 20:27:36 - mmengine - INFO - Epoch(train) [15][ 400/5758] lr: 9.0382e-05 eta: 7:22:36 time: 0.7232 data_time: 0.0014 memory: 25074 loss: 0.0004 +2023/06/01 20:28:54 - mmengine - INFO - Epoch(train) [15][ 500/5758] lr: 9.0382e-05 eta: 7:21:18 time: 0.8211 data_time: 0.0014 memory: 25074 loss: 0.0018 +2023/06/01 20:30:18 - mmengine - INFO - Epoch(train) [15][ 600/5758] lr: 9.0382e-05 eta: 7:20:02 time: 0.8222 data_time: 0.0016 memory: 25074 loss: 0.0016 +2023/06/01 20:31:40 - mmengine - INFO - Epoch(train) [15][ 700/5758] lr: 9.0382e-05 eta: 7:18:47 time: 0.8281 data_time: 0.0014 memory: 25074 loss: 0.0007 +2023/06/01 20:32:59 - mmengine - INFO - Epoch(train) [15][ 800/5758] lr: 9.0382e-05 eta: 7:17:29 time: 0.7273 data_time: 0.0015 memory: 25074 loss: 0.0007 +2023/06/01 20:34:15 - mmengine - INFO - Epoch(train) [15][ 900/5758] lr: 9.0382e-05 eta: 7:16:11 time: 0.7592 data_time: 0.0018 memory: 25074 loss: 0.0014 +2023/06/01 20:35:31 - mmengine - INFO - Epoch(train) [15][1000/5758] lr: 9.0382e-05 eta: 7:14:52 time: 0.7900 data_time: 0.0032 memory: 25074 loss: 0.0018 +2023/06/01 20:36:48 - mmengine - INFO - Epoch(train) [15][1100/5758] lr: 9.0382e-05 eta: 7:13:34 time: 0.7625 data_time: 0.0021 memory: 25074 loss: 0.0004 +2023/06/01 20:38:07 - mmengine - INFO - Epoch(train) [15][1200/5758] lr: 9.0382e-05 eta: 7:12:17 time: 0.7359 data_time: 0.0016 memory: 25074 loss: 0.0010 +2023/06/01 20:39:26 - mmengine - INFO - Epoch(train) [15][1300/5758] lr: 9.0382e-05 eta: 7:11:00 time: 0.8281 data_time: 0.0015 memory: 25074 loss: 0.0021 +2023/06/01 20:40:38 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m-lr4e-4_20230601_025223 +2023/06/01 20:40:47 - mmengine - INFO - Epoch(train) [15][1400/5758] lr: 9.0382e-05 eta: 7:09:43 time: 0.7969 data_time: 0.0018 memory: 25074 loss: 0.0020 +2023/06/01 20:42:06 - mmengine - INFO - Epoch(train) [15][1500/5758] lr: 9.0382e-05 eta: 7:08:26 time: 0.7719 data_time: 0.0016 memory: 25074 loss: 0.0009 +2023/06/01 20:43:25 - mmengine - INFO - Epoch(train) [15][1600/5758] lr: 9.0382e-05 eta: 7:07:08 time: 0.8478 data_time: 0.0015 memory: 25074 loss: 0.0022 +2023/06/01 20:44:43 - mmengine - INFO - Epoch(train) [15][1700/5758] lr: 9.0382e-05 eta: 7:05:51 time: 0.8147 data_time: 0.0017 memory: 25074 loss: 0.0001 +2023/06/01 20:46:00 - mmengine - INFO - Epoch(train) [15][1800/5758] lr: 9.0382e-05 eta: 7:04:33 time: 0.8911 data_time: 0.0017 memory: 25074 loss: 0.0000 +2023/06/01 20:47:14 - mmengine - INFO - Epoch(train) [15][1900/5758] lr: 9.0382e-05 eta: 7:03:13 time: 0.7292 data_time: 0.0019 memory: 25074 loss: 0.0001 +2023/06/01 20:48:29 - mmengine - INFO - Epoch(train) [15][2000/5758] lr: 9.0382e-05 eta: 7:01:55 time: 0.8028 data_time: 0.0017 memory: 25074 loss: 0.0007 +2023/06/01 20:49:45 - mmengine - INFO - Epoch(train) [15][2100/5758] lr: 9.0382e-05 eta: 7:00:36 time: 0.8323 data_time: 0.0016 memory: 25074 loss: 0.0014 +2023/06/01 20:51:07 - mmengine - INFO - Epoch(train) [15][2200/5758] lr: 9.0382e-05 eta: 6:59:20 time: 0.8938 data_time: 0.0014 memory: 25074 loss: 0.0004 +2023/06/01 20:52:37 - mmengine - INFO - Epoch(train) [15][2300/5758] lr: 9.0382e-05 eta: 6:58:07 time: 0.7980 data_time: 0.0015 memory: 25074 loss: 0.0006 +2023/06/01 20:53:58 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m-lr4e-4_20230601_025223 +2023/06/01 20:54:10 - mmengine - INFO - Epoch(train) [15][2400/5758] lr: 9.0382e-05 eta: 6:56:55 time: 1.0595 data_time: 0.0016 memory: 25074 loss: 0.0004 +2023/06/01 20:55:53 - mmengine - INFO - Epoch(train) [15][2500/5758] lr: 9.0382e-05 eta: 6:55:47 time: 0.9078 data_time: 0.0015 memory: 25074 loss: 0.0009 +2023/06/01 20:57:33 - mmengine - INFO - Epoch(train) [15][2600/5758] lr: 9.0382e-05 eta: 6:54:37 time: 0.9692 data_time: 0.0015 memory: 25074 loss: 0.0019 +2023/06/01 20:59:06 - mmengine - INFO - Epoch(train) [15][2700/5758] lr: 9.0382e-05 eta: 6:53:25 time: 0.7200 data_time: 0.0019 memory: 25074 loss: 0.0014 +2023/06/01 21:00:26 - mmengine - INFO - Epoch(train) [15][2800/5758] lr: 9.0382e-05 eta: 6:52:08 time: 0.8423 data_time: 0.0015 memory: 25074 loss: 0.0001 +2023/06/01 21:01:48 - mmengine - INFO - Epoch(train) [15][2900/5758] lr: 9.0382e-05 eta: 6:50:52 time: 0.8672 data_time: 0.0014 memory: 25074 loss: 0.0026 +2023/06/01 21:03:15 - mmengine - INFO - Epoch(train) [15][3000/5758] lr: 9.0382e-05 eta: 6:49:37 time: 0.8385 data_time: 0.0018 memory: 25074 loss: 0.0006 +2023/06/01 21:04:41 - mmengine - INFO - Epoch(train) [15][3100/5758] lr: 9.0382e-05 eta: 6:48:23 time: 0.8059 data_time: 0.0015 memory: 25074 loss: 0.0010 +2023/06/01 21:06:23 - mmengine - INFO - Epoch(train) [15][3200/5758] lr: 9.0382e-05 eta: 6:47:14 time: 1.0917 data_time: 0.0015 memory: 25074 loss: 0.0004 +2023/06/01 21:08:11 - mmengine - INFO - Epoch(train) [15][3300/5758] lr: 9.0382e-05 eta: 6:46:07 time: 1.1890 data_time: 0.0017 memory: 25074 loss: 0.0016 +2023/06/01 21:09:45 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m-lr4e-4_20230601_025223 +2023/06/01 21:09:57 - mmengine - INFO - Epoch(train) [15][3400/5758] lr: 9.0382e-05 eta: 6:44:59 time: 1.0003 data_time: 0.0013 memory: 25074 loss: 0.0009 +2023/06/01 21:11:56 - mmengine - INFO - Epoch(train) [15][3500/5758] lr: 9.0382e-05 eta: 6:43:57 time: 2.5094 data_time: 0.0032 memory: 25074 loss: 0.0032 +2023/06/01 21:14:00 - mmengine - INFO - Epoch(train) [15][3600/5758] lr: 9.0382e-05 eta: 6:42:55 time: 0.9590 data_time: 0.0014 memory: 25074 loss: 0.0006 +2023/06/01 21:15:51 - mmengine - INFO - Epoch(train) [15][3700/5758] lr: 9.0382e-05 eta: 6:41:49 time: 1.0497 data_time: 0.0015 memory: 25074 loss: 0.0009 +2023/06/01 21:17:20 - mmengine - INFO - Epoch(train) [15][3800/5758] lr: 9.0382e-05 eta: 6:40:35 time: 0.7711 data_time: 0.0019 memory: 25074 loss: 0.0003 +2023/06/01 21:18:32 - mmengine - INFO - Epoch(train) [15][3900/5758] lr: 9.0382e-05 eta: 6:39:15 time: 0.7106 data_time: 0.0013 memory: 25074 loss: 0.0008 +2023/06/01 21:19:45 - mmengine - INFO - Epoch(train) [15][4000/5758] lr: 9.0382e-05 eta: 6:37:55 time: 0.7041 data_time: 0.0014 memory: 25074 loss: 0.0018 +2023/06/01 21:20:57 - mmengine - INFO - Epoch(train) [15][4100/5758] lr: 9.0382e-05 eta: 6:36:34 time: 0.7279 data_time: 0.0013 memory: 25074 loss: 0.0021 +2023/06/01 21:22:13 - mmengine - INFO - Epoch(train) [15][4200/5758] lr: 9.0382e-05 eta: 6:35:15 time: 0.7523 data_time: 0.0016 memory: 25074 loss: 0.0017 +2023/06/01 21:23:27 - mmengine - INFO - Epoch(train) [15][4300/5758] lr: 9.0382e-05 eta: 6:33:56 time: 0.7270 data_time: 0.0013 memory: 25074 loss: 0.0007 +2023/06/01 21:24:32 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m-lr4e-4_20230601_025223 +2023/06/01 21:24:40 - mmengine - INFO - Epoch(train) [15][4400/5758] lr: 9.0382e-05 eta: 6:32:36 time: 0.7073 data_time: 0.0015 memory: 25074 loss: 0.0003 +2023/06/01 21:25:56 - mmengine - INFO - Epoch(train) [15][4500/5758] lr: 9.0382e-05 eta: 6:31:17 time: 0.7589 data_time: 0.0017 memory: 25074 loss: 0.0006 +2023/06/01 21:27:10 - mmengine - INFO - Epoch(train) [15][4600/5758] lr: 9.0382e-05 eta: 6:29:57 time: 0.7193 data_time: 0.0017 memory: 25074 loss: 0.0001 +2023/06/01 21:28:25 - mmengine - INFO - Epoch(train) [15][4700/5758] lr: 9.0382e-05 eta: 6:28:38 time: 0.7302 data_time: 0.0020 memory: 25074 loss: 0.0004 +2023/06/01 21:29:44 - mmengine - INFO - Epoch(train) [15][4800/5758] lr: 9.0382e-05 eta: 6:27:20 time: 0.8181 data_time: 0.0015 memory: 25074 loss: 0.0003 +2023/06/01 21:30:59 - mmengine - INFO - Epoch(train) [15][4900/5758] lr: 9.0382e-05 eta: 6:26:01 time: 0.7277 data_time: 0.0012 memory: 25074 loss: 0.0016 +2023/06/01 21:32:13 - mmengine - INFO - Epoch(train) [15][5000/5758] lr: 9.0382e-05 eta: 6:24:41 time: 0.7537 data_time: 0.0014 memory: 25074 loss: 0.0007 +2023/06/01 21:33:30 - mmengine - INFO - Epoch(train) [15][5100/5758] lr: 9.0382e-05 eta: 6:23:23 time: 0.7465 data_time: 0.0014 memory: 25074 loss: 0.0006 +2023/06/01 21:34:45 - mmengine - INFO - Epoch(train) [15][5200/5758] lr: 9.0382e-05 eta: 6:22:04 time: 0.7062 data_time: 0.0015 memory: 25074 loss: 0.0022 +2023/06/01 21:36:00 - mmengine - INFO - Epoch(train) [15][5300/5758] lr: 9.0382e-05 eta: 6:20:44 time: 0.7733 data_time: 0.0015 memory: 25074 loss: 0.0003 +2023/06/01 21:37:05 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m-lr4e-4_20230601_025223 +2023/06/01 21:37:15 - mmengine - INFO - Epoch(train) [15][5400/5758] lr: 9.0382e-05 eta: 6:19:25 time: 0.8091 data_time: 0.0025 memory: 25074 loss: 0.0028 +2023/06/01 21:38:30 - mmengine - INFO - Epoch(train) [15][5500/5758] lr: 9.0382e-05 eta: 6:18:06 time: 0.7728 data_time: 0.0022 memory: 25074 loss: 0.0002 +2023/06/01 21:39:45 - mmengine - INFO - Epoch(train) [15][5600/5758] lr: 9.0382e-05 eta: 6:16:47 time: 0.7776 data_time: 0.0018 memory: 25074 loss: 0.0020 +2023/06/01 21:41:02 - mmengine - INFO - Epoch(train) [15][5700/5758] lr: 9.0382e-05 eta: 6:15:29 time: 0.7203 data_time: 0.0012 memory: 25074 loss: 0.0012 +2023/06/01 21:41:45 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m-lr4e-4_20230601_025223 +2023/06/01 21:41:45 - mmengine - INFO - Saving checkpoint at 15 epochs +2023/06/01 21:42:04 - mmengine - INFO - Epoch(val) [15][16/16] accuracy/top1: 99.9937 single-label/precision_classwise: [100.0, 0.0] single-label/recall_classwise: [99.99369049072266, 0.0] single-label/f1-score_classwise: [99.9968490600586, 0.0] data_time: 0.2504 time: 0.6420 +2023/06/01 21:43:28 - mmengine - INFO - Epoch(train) [16][ 100/5758] lr: 6.7114e-05 eta: 6:13:27 time: 0.7708 data_time: 0.0014 memory: 25074 loss: 0.0009 +2023/06/01 21:44:45 - mmengine - INFO - Epoch(train) [16][ 200/5758] lr: 6.7114e-05 eta: 6:12:08 time: 0.7555 data_time: 0.0016 memory: 25074 loss: 0.0005 +2023/06/01 21:46:01 - mmengine - INFO - Epoch(train) [16][ 300/5758] lr: 6.7114e-05 eta: 6:10:49 time: 0.8286 data_time: 0.0642 memory: 25074 loss: 0.0002 +2023/06/01 21:47:17 - mmengine - INFO - Epoch(train) [16][ 400/5758] lr: 6.7114e-05 eta: 6:09:30 time: 0.7918 data_time: 0.0886 memory: 25074 loss: 0.0006 +2023/06/01 21:48:35 - mmengine - INFO - Epoch(train) [16][ 500/5758] lr: 6.7114e-05 eta: 6:08:12 time: 0.7812 data_time: 0.0623 memory: 25074 loss: 0.0006 +2023/06/01 21:49:50 - mmengine - INFO - Epoch(train) [16][ 600/5758] lr: 6.7114e-05 eta: 6:06:53 time: 0.7860 data_time: 0.0809 memory: 25074 loss: 0.0003 +2023/06/01 21:50:14 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m-lr4e-4_20230601_025223 +2023/06/01 21:51:07 - mmengine - INFO - Epoch(train) [16][ 700/5758] lr: 6.7114e-05 eta: 6:05:35 time: 0.7530 data_time: 0.0347 memory: 25074 loss: 0.0006 +2023/06/01 21:52:24 - mmengine - INFO - Epoch(train) [16][ 800/5758] lr: 6.7114e-05 eta: 6:04:16 time: 0.7798 data_time: 0.0054 memory: 25074 loss: 0.0026 +2023/06/01 21:53:40 - mmengine - INFO - Epoch(train) [16][ 900/5758] lr: 6.7114e-05 eta: 6:02:57 time: 0.7396 data_time: 0.0015 memory: 25074 loss: 0.0001 +2023/06/01 21:54:53 - mmengine - INFO - Epoch(train) [16][1000/5758] lr: 6.7114e-05 eta: 6:01:38 time: 0.7725 data_time: 0.0014 memory: 25074 loss: 0.0007 +2023/06/01 21:56:08 - mmengine - INFO - Epoch(train) [16][1100/5758] lr: 6.7114e-05 eta: 6:00:19 time: 0.7387 data_time: 0.0016 memory: 25074 loss: 0.0003 +2023/06/01 21:57:29 - mmengine - INFO - Epoch(train) [16][1200/5758] lr: 6.7114e-05 eta: 5:59:02 time: 0.7246 data_time: 0.0015 memory: 25074 loss: 0.0022 +2023/06/01 21:58:43 - mmengine - INFO - Epoch(train) [16][1300/5758] lr: 6.7114e-05 eta: 5:57:42 time: 0.7315 data_time: 0.0016 memory: 25074 loss: 0.0012 +2023/06/01 21:59:57 - mmengine - INFO - Epoch(train) [16][1400/5758] lr: 6.7114e-05 eta: 5:56:23 time: 0.7534 data_time: 0.0017 memory: 25074 loss: 0.0002 +2023/06/01 22:01:11 - mmengine - INFO - Epoch(train) [16][1500/5758] lr: 6.7114e-05 eta: 5:55:03 time: 0.7060 data_time: 0.0013 memory: 25074 loss: 0.0006 +2023/06/01 22:02:22 - mmengine - INFO - Epoch(train) [16][1600/5758] lr: 6.7114e-05 eta: 5:53:43 time: 0.7124 data_time: 0.0018 memory: 25074 loss: 0.0003 +2023/06/01 22:02:44 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m-lr4e-4_20230601_025223 +2023/06/01 22:03:36 - mmengine - INFO - Epoch(train) [16][1700/5758] lr: 6.7114e-05 eta: 5:52:24 time: 0.7276 data_time: 0.0015 memory: 25074 loss: 0.0006 +2023/06/01 22:04:50 - mmengine - INFO - Epoch(train) [16][1800/5758] lr: 6.7114e-05 eta: 5:51:05 time: 0.7885 data_time: 0.0017 memory: 25074 loss: 0.0001 +2023/06/01 22:06:03 - mmengine - INFO - Epoch(train) [16][1900/5758] lr: 6.7114e-05 eta: 5:49:45 time: 0.7216 data_time: 0.0013 memory: 25074 loss: 0.0014 +2023/06/01 22:07:16 - mmengine - INFO - Epoch(train) [16][2000/5758] lr: 6.7114e-05 eta: 5:48:26 time: 0.7748 data_time: 0.0013 memory: 25074 loss: 0.0011 +2023/06/01 22:08:32 - mmengine - INFO - Epoch(train) [16][2100/5758] lr: 6.7114e-05 eta: 5:47:07 time: 0.7620 data_time: 0.0014 memory: 25074 loss: 0.0002 +2023/06/01 22:09:46 - mmengine - INFO - Epoch(train) [16][2200/5758] lr: 6.7114e-05 eta: 5:45:48 time: 0.7522 data_time: 0.0015 memory: 25074 loss: 0.0001 +2023/06/01 22:11:00 - mmengine - INFO - Epoch(train) [16][2300/5758] lr: 6.7114e-05 eta: 5:44:29 time: 0.7594 data_time: 0.0016 memory: 25074 loss: 0.0003 +2023/06/01 22:12:16 - mmengine - INFO - Epoch(train) [16][2400/5758] lr: 6.7114e-05 eta: 5:43:10 time: 0.8098 data_time: 0.0015 memory: 25074 loss: 0.0021 +2023/06/01 22:13:32 - mmengine - INFO - Epoch(train) [16][2500/5758] lr: 6.7114e-05 eta: 5:41:51 time: 0.7633 data_time: 0.0017 memory: 25074 loss: 0.0000 +2023/06/01 22:14:46 - mmengine - INFO - Epoch(train) [16][2600/5758] lr: 6.7114e-05 eta: 5:40:32 time: 0.7777 data_time: 0.0013 memory: 25074 loss: 0.0002 +2023/06/01 22:15:10 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m-lr4e-4_20230601_025223 +2023/06/01 22:16:01 - mmengine - INFO - Epoch(train) [16][2700/5758] lr: 6.7114e-05 eta: 5:39:13 time: 0.7150 data_time: 0.0015 memory: 25074 loss: 0.0032 +2023/06/01 22:17:16 - mmengine - INFO - Epoch(train) [16][2800/5758] lr: 6.7114e-05 eta: 5:37:54 time: 0.7699 data_time: 0.0014 memory: 25074 loss: 0.0001 +2023/06/01 22:18:30 - mmengine - INFO - Epoch(train) [16][2900/5758] lr: 6.7114e-05 eta: 5:36:35 time: 0.7072 data_time: 0.0013 memory: 25074 loss: 0.0006 +2023/06/01 22:19:47 - mmengine - INFO - Epoch(train) [16][3000/5758] lr: 6.7114e-05 eta: 5:35:17 time: 0.8838 data_time: 0.0018 memory: 25074 loss: 0.0055 +2023/06/01 22:21:05 - mmengine - INFO - Epoch(train) [16][3100/5758] lr: 6.7114e-05 eta: 5:33:59 time: 0.8784 data_time: 0.0019 memory: 25074 loss: 0.0030 +2023/06/01 22:22:24 - mmengine - INFO - Epoch(train) [16][3200/5758] lr: 6.7114e-05 eta: 5:32:41 time: 0.8018 data_time: 0.0021 memory: 25074 loss: 0.0012 +2023/06/01 22:23:42 - mmengine - INFO - Epoch(train) [16][3300/5758] lr: 6.7114e-05 eta: 5:31:23 time: 0.8086 data_time: 0.0016 memory: 25074 loss: 0.0001 +2023/06/01 22:24:59 - mmengine - INFO - Epoch(train) [16][3400/5758] lr: 6.7114e-05 eta: 5:30:05 time: 0.7591 data_time: 0.0013 memory: 25074 loss: 0.0007 +2023/06/01 22:26:17 - mmengine - INFO - Epoch(train) [16][3500/5758] lr: 6.7114e-05 eta: 5:28:47 time: 0.7344 data_time: 0.0014 memory: 25074 loss: 0.0016 +2023/06/01 22:27:34 - mmengine - INFO - Epoch(train) [16][3600/5758] lr: 6.7114e-05 eta: 5:27:28 time: 0.7476 data_time: 0.0016 memory: 25074 loss: 0.0038 +2023/06/01 22:27:59 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m-lr4e-4_20230601_025223 +2023/06/01 22:28:50 - mmengine - INFO - Epoch(train) [16][3700/5758] lr: 6.7114e-05 eta: 5:26:10 time: 0.7130 data_time: 0.0014 memory: 25074 loss: 0.0011 diff --git a/convnext_small_4xb256_fake5m-lr4e-4/20230601_025223/vis_data/20230601_025223.json b/convnext_small_4xb256_fake5m-lr4e-4/20230601_025223/vis_data/20230601_025223.json new file mode 100644 index 0000000000000000000000000000000000000000..acd46c6dd81ae719bc575ff43f547118a20deac1 --- /dev/null +++ b/convnext_small_4xb256_fake5m-lr4e-4/20230601_025223/vis_data/20230601_025223.json @@ -0,0 +1,907 @@ +{"lr": 0.0004, "data_time": 0.023465752601623535, "loss": 0.5783761262893676, "time": 0.7354685544967652, "epoch": 1, "memory": 25071, "step": 100} +{"lr": 0.0004, "data_time": 0.0012981891632080078, "loss": 0.5477651119232178, "time": 0.7451019525527954, "epoch": 1, "memory": 25071, "step": 200} +{"lr": 0.0004, "data_time": 0.0011882543563842773, "loss": 0.5147063314914704, "time": 0.8193954229354858, "epoch": 1, "memory": 25071, "step": 300} +{"lr": 0.0004, "data_time": 0.0012914657592773438, "loss": 0.5042121767997741, "time": 0.8707980155944824, "epoch": 1, "memory": 25071, "step": 400} +{"lr": 0.0004, "data_time": 0.0014605045318603516, "loss": 0.46452565789222716, "time": 0.7508379697799683, "epoch": 1, "memory": 25071, "step": 500} +{"lr": 0.0004, "data_time": 0.0013184547424316406, "loss": 0.45228410363197324, "time": 0.7682911157608032, "epoch": 1, "memory": 25071, "step": 600} +{"lr": 0.0004, "data_time": 0.001284956932067871, "loss": 0.44432016015052794, "time": 0.7671506881713868, "epoch": 1, "memory": 25071, "step": 700} +{"lr": 0.0004, "data_time": 0.0012042999267578125, "loss": 0.45052569210529325, "time": 0.710477876663208, "epoch": 1, "memory": 25071, "step": 800} +{"lr": 0.0004, "data_time": 0.0011302709579467773, "loss": 0.40240114033222196, "time": 0.7120702743530274, "epoch": 1, "memory": 25071, "step": 900} +{"lr": 0.0004, "data_time": 0.0011893749237060548, "loss": 0.4003148555755615, "time": 0.7268030405044555, "epoch": 1, "memory": 25071, "step": 1000} +{"lr": 0.0004, "data_time": 0.001346588134765625, "loss": 0.4231216698884964, "time": 0.73503258228302, "epoch": 1, "memory": 25071, "step": 1100} +{"lr": 0.0004, "data_time": 0.0012927055358886719, "loss": 0.4057676702737808, "time": 0.7073668003082275, "epoch": 1, "memory": 25071, "step": 1200} +{"lr": 0.0004, "data_time": 0.0015211105346679688, "loss": 0.4136209189891815, "time": 0.7717820167541504, "epoch": 1, "memory": 25071, "step": 1300} +{"lr": 0.0004, "data_time": 0.0015099525451660156, "loss": 0.3821967035531998, "time": 0.7974962711334228, "epoch": 1, "memory": 25071, "step": 1400} +{"lr": 0.0004, "data_time": 0.0014603614807128906, "loss": 0.3885293424129486, "time": 0.7651231765747071, "epoch": 1, "memory": 25071, "step": 1500} +{"lr": 0.0004, "data_time": 0.0015938997268676758, "loss": 0.35765465497970583, "time": 0.7080837488174438, "epoch": 1, "memory": 25071, "step": 1600} +{"lr": 0.0004, "data_time": 0.0012611865997314454, "loss": 0.36647533178329467, "time": 0.7144793033599853, "epoch": 1, "memory": 25071, "step": 1700} +{"lr": 0.0004, "data_time": 0.001858091354370117, "loss": 0.3575074225664139, "time": 0.7647769451141357, "epoch": 1, "memory": 25071, "step": 1800} +{"lr": 0.0004, "data_time": 0.001350712776184082, "loss": 0.34662758111953734, "time": 0.7962718725204467, "epoch": 1, "memory": 25071, "step": 1900} +{"lr": 0.0004, "data_time": 0.0013519287109375, "loss": 0.37120504677295685, "time": 0.7751044273376465, "epoch": 1, "memory": 25071, "step": 2000} +{"lr": 0.0004, "data_time": 0.001309347152709961, "loss": 0.31365267783403394, "time": 0.759409499168396, "epoch": 1, "memory": 25071, "step": 2100} +{"lr": 0.0004, "data_time": 0.001217794418334961, "loss": 0.3412786304950714, "time": 0.7413238286972046, "epoch": 1, "memory": 25071, "step": 2200} +{"lr": 0.0004, "data_time": 0.0013586759567260742, "loss": 0.3255023032426834, "time": 0.7054993152618408, "epoch": 1, "memory": 25071, "step": 2300} +{"lr": 0.0004, "data_time": 0.0011933565139770508, "loss": 0.31905866116285325, "time": 0.8325782299041748, "epoch": 1, "memory": 25071, "step": 2400} +{"lr": 0.0004, "data_time": 0.0014669656753540038, "loss": 0.31494944393634794, "time": 0.7384720563888549, "epoch": 1, "memory": 25071, "step": 2500} +{"lr": 0.0004, "data_time": 0.0012603282928466796, "loss": 0.30391877591609956, "time": 0.7293519020080567, "epoch": 1, "memory": 25071, "step": 2600} +{"lr": 0.0004, "data_time": 0.0012967586517333984, "loss": 0.27619979083538054, "time": 0.7716137170791626, "epoch": 1, "memory": 25071, "step": 2700} +{"lr": 0.0004, "data_time": 0.0012556791305541992, "loss": 0.29391306936740874, "time": 0.7249278068542481, "epoch": 1, "memory": 25071, "step": 2800} +{"lr": 0.0004, "data_time": 0.0013440847396850586, "loss": 0.29973573088645933, "time": 0.8226109266281127, "epoch": 1, "memory": 25071, "step": 2900} +{"lr": 0.0004, "data_time": 0.0011576175689697265, "loss": 0.2650411456823349, "time": 0.7453691244125367, "epoch": 1, "memory": 25071, "step": 3000} +{"lr": 0.0004, "data_time": 0.0015404701232910156, "loss": 0.29347882568836214, "time": 0.7388896703720093, "epoch": 1, "memory": 25071, "step": 3100} +{"lr": 0.0004, "data_time": 0.0012520074844360352, "loss": 0.27183108031749725, "time": 0.710351824760437, "epoch": 1, "memory": 25071, "step": 3200} +{"lr": 0.0004, "data_time": 0.0012935638427734376, "loss": 0.2756800249218941, "time": 0.7611816167831421, "epoch": 1, "memory": 25071, "step": 3300} +{"lr": 0.0004, "data_time": 0.0012006044387817382, "loss": 0.2523490786552429, "time": 0.7746668577194213, "epoch": 1, "memory": 25071, "step": 3400} +{"lr": 0.0004, "data_time": 0.0012663602828979492, "loss": 0.2788736060261726, "time": 0.7459849834442138, "epoch": 1, "memory": 25071, "step": 3500} +{"lr": 0.0004, "data_time": 0.0014612436294555663, "loss": 0.2532954692840576, "time": 0.7365252733230591, "epoch": 1, "memory": 25071, "step": 3600} +{"lr": 0.0004, "data_time": 0.0015092849731445312, "loss": 0.2760276019573212, "time": 0.7436590194702148, "epoch": 1, "memory": 25071, "step": 3700} +{"lr": 0.0004, "data_time": 0.0013444185256958007, "loss": 0.24519322961568832, "time": 0.7767281055450439, "epoch": 1, "memory": 25071, "step": 3800} +{"lr": 0.0004, "data_time": 0.0013512611389160157, "loss": 0.24075794219970703, "time": 0.7031121253967285, "epoch": 1, "memory": 25071, "step": 3900} +{"lr": 0.0004, "data_time": 0.0014681577682495116, "loss": 0.2520707279443741, "time": 0.7432272434234619, "epoch": 1, "memory": 25071, "step": 4000} +{"lr": 0.0004, "data_time": 0.0013287782669067383, "loss": 0.24621658474206926, "time": 0.858167290687561, "epoch": 1, "memory": 25071, "step": 4100} +{"lr": 0.0004, "data_time": 0.0014040946960449218, "loss": 0.23313160091638566, "time": 0.7725039958953858, "epoch": 1, "memory": 25071, "step": 4200} +{"lr": 0.0004, "data_time": 0.0013962030410766602, "loss": 0.22922349125146865, "time": 0.8162384033203125, "epoch": 1, "memory": 25071, "step": 4300} +{"lr": 0.0004, "data_time": 0.0013297557830810546, "loss": 0.23687607645988465, "time": 0.754087257385254, "epoch": 1, "memory": 25071, "step": 4400} +{"lr": 0.0004, "data_time": 0.001346278190612793, "loss": 0.23668272644281388, "time": 0.7501036882400512, "epoch": 1, "memory": 25071, "step": 4500} +{"lr": 0.0004, "data_time": 0.0013803482055664063, "loss": 0.2544979602098465, "time": 0.7481486558914184, "epoch": 1, "memory": 25071, "step": 4600} +{"lr": 0.0004, "data_time": 0.0012869119644165039, "loss": 0.21583334356546402, "time": 0.7296990156173706, "epoch": 1, "memory": 25071, "step": 4700} +{"lr": 0.0004, "data_time": 0.001413273811340332, "loss": 0.2173559471964836, "time": 0.7453904151916504, "epoch": 1, "memory": 25071, "step": 4800} +{"lr": 0.0004, "data_time": 0.0012706995010375976, "loss": 0.22525127977132797, "time": 0.719694972038269, "epoch": 1, "memory": 25071, "step": 4900} +{"lr": 0.0004, "data_time": 0.001439213752746582, "loss": 0.22691788226366044, "time": 0.7515617609024048, "epoch": 1, "memory": 25071, "step": 5000} +{"lr": 0.0004, "data_time": 0.0014858007431030273, "loss": 0.19926073253154755, "time": 0.7507566690444947, "epoch": 1, "memory": 25071, "step": 5100} +{"lr": 0.0004, "data_time": 0.00108184814453125, "loss": 0.20982834398746492, "time": 0.7858760356903076, "epoch": 1, "memory": 25071, "step": 5200} +{"lr": 0.0004, "data_time": 0.0014892339706420899, "loss": 0.19950440973043443, "time": 0.7765029191970825, "epoch": 1, "memory": 25071, "step": 5300} +{"lr": 0.0004, "data_time": 0.0012821674346923829, "loss": 0.19394015818834304, "time": 0.7908424854278564, "epoch": 1, "memory": 25071, "step": 5400} +{"lr": 0.0004, "data_time": 0.0014799118041992187, "loss": 0.19626849442720412, "time": 0.8351553916931153, "epoch": 1, "memory": 25071, "step": 5500} +{"lr": 0.0004, "data_time": 0.001165318489074707, "loss": 0.20947481840848922, "time": 0.7716299772262574, "epoch": 1, "memory": 25071, "step": 5600} +{"lr": 0.0004, "data_time": 0.0012734174728393556, "loss": 0.1822401225566864, "time": 0.7181431293487549, "epoch": 1, "memory": 25071, "step": 5700} +{"accuracy/top1": 91.80606842041016, "data_time": 0.2736958712339401, "time": 0.7940136343240738, "step": 1} +{"lr": 0.00039759922641605187, "data_time": 0.056543922424316405, "loss": 0.178238408267498, "time": 0.7774782419204712, "epoch": 2, "memory": 25074, "step": 5858} +{"lr": 0.00039759922641605187, "data_time": 0.0014495372772216797, "loss": 0.17998147308826445, "time": 0.7055389881134033, "epoch": 2, "memory": 25074, "step": 5958} +{"lr": 0.00039759922641605187, "data_time": 0.0016408443450927734, "loss": 0.19104122668504714, "time": 0.7790418148040772, "epoch": 2, "memory": 25074, "step": 6058} +{"lr": 0.00039759922641605187, "data_time": 0.0013885498046875, "loss": 0.18300386071205138, "time": 0.8070295333862305, "epoch": 2, "memory": 25074, "step": 6158} +{"lr": 0.00039759922641605187, "data_time": 0.001452493667602539, "loss": 0.16158113032579421, "time": 0.7680969715118409, "epoch": 2, "memory": 25074, "step": 6258} +{"lr": 0.00039759922641605187, "data_time": 0.0013581037521362305, "loss": 0.1782495841383934, "time": 0.7868995189666748, "epoch": 2, "memory": 25074, "step": 6358} +{"lr": 0.00039759922641605187, "data_time": 0.0014699459075927734, "loss": 0.15414375811815262, "time": 0.7201312065124512, "epoch": 2, "memory": 25074, "step": 6458} +{"lr": 0.00039759922641605187, "data_time": 0.0016120672225952148, "loss": 0.2027489274740219, "time": 0.732387924194336, "epoch": 2, "memory": 25074, "step": 6558} +{"lr": 0.00039759922641605187, "data_time": 0.0012201547622680664, "loss": 0.17093847543001175, "time": 0.7485415458679199, "epoch": 2, "memory": 25074, "step": 6658} +{"lr": 0.00039759922641605187, "data_time": 0.0014460086822509766, "loss": 0.17120749801397322, "time": 0.7299433469772338, "epoch": 2, "memory": 25074, "step": 6758} +{"lr": 0.00039759922641605187, "data_time": 0.0021276235580444335, "loss": 0.1816856697201729, "time": 0.7633233308792114, "epoch": 2, "memory": 25074, "step": 6858} +{"lr": 0.00039759922641605187, "data_time": 0.0014348983764648437, "loss": 0.151786420494318, "time": 0.7627900600433349, "epoch": 2, "memory": 25074, "step": 6958} +{"lr": 0.00039759922641605187, "data_time": 0.0014765024185180663, "loss": 0.14380268454551698, "time": 0.8085668087005615, "epoch": 2, "memory": 25074, "step": 7058} +{"lr": 0.00039759922641605187, "data_time": 0.0010685682296752929, "loss": 0.1663478970527649, "time": 0.7716030836105346, "epoch": 2, "memory": 25074, "step": 7158} +{"lr": 0.00039759922641605187, "data_time": 0.0012255191802978515, "loss": 0.1608097583055496, "time": 0.7307199478149414, "epoch": 2, "memory": 25074, "step": 7258} +{"lr": 0.00039759922641605187, "data_time": 0.0013592004776000976, "loss": 0.16642693132162095, "time": 0.7474058628082275, "epoch": 2, "memory": 25074, "step": 7358} +{"lr": 0.00039759922641605187, "data_time": 0.001287078857421875, "loss": 0.20146722197532654, "time": 0.7900904655456543, "epoch": 2, "memory": 25074, "step": 7458} +{"lr": 0.00039759922641605187, "data_time": 0.0015846729278564454, "loss": 0.14182349368929864, "time": 0.8205183982849121, "epoch": 2, "memory": 25074, "step": 7558} +{"lr": 0.00039759922641605187, "data_time": 0.0014862298965454101, "loss": 0.1436505876481533, "time": 0.7209877490997314, "epoch": 2, "memory": 25074, "step": 7658} +{"lr": 0.00039759922641605187, "data_time": 0.001214313507080078, "loss": 0.1374882437288761, "time": 0.7201198101043701, "epoch": 2, "memory": 25074, "step": 7758} +{"lr": 0.00039759922641605187, "data_time": 0.0012128829956054687, "loss": 0.13091749101877212, "time": 0.7129969596862793, "epoch": 2, "memory": 25074, "step": 7858} +{"lr": 0.00039759922641605187, "data_time": 0.0016506671905517577, "loss": 0.15016361474990844, "time": 0.7774442911148072, "epoch": 2, "memory": 25074, "step": 7958} +{"lr": 0.00039759922641605187, "data_time": 0.001264667510986328, "loss": 0.11804806515574455, "time": 0.7509546518325806, "epoch": 2, "memory": 25074, "step": 8058} +{"lr": 0.00039759922641605187, "data_time": 0.0017354726791381837, "loss": 0.13144950568675995, "time": 0.7998786926269531, "epoch": 2, "memory": 25074, "step": 8158} +{"lr": 0.00039759922641605187, "data_time": 0.0013310432434082032, "loss": 0.13164022490382193, "time": 0.8064572811126709, "epoch": 2, "memory": 25074, "step": 8258} +{"lr": 0.00039759922641605187, "data_time": 0.0012734174728393556, "loss": 0.13403431996703147, "time": 0.805510687828064, "epoch": 2, "memory": 25074, "step": 8358} +{"lr": 0.00039759922641605187, "data_time": 0.0013266801834106445, "loss": 0.1468445010483265, "time": 0.7319570064544678, "epoch": 2, "memory": 25074, "step": 8458} +{"lr": 0.00039759922641605187, "data_time": 0.02174842357635498, "loss": 0.14606546387076377, "time": 0.7308906316757202, "epoch": 2, "memory": 25074, "step": 8558} +{"lr": 0.00039759922641605187, "data_time": 0.0012790203094482423, "loss": 0.13792258203029634, "time": 0.7167191982269288, "epoch": 2, "memory": 25074, "step": 8658} +{"lr": 0.00039759922641605187, "data_time": 0.001343703269958496, "loss": 0.12017014175653458, "time": 0.8110358238220214, "epoch": 2, "memory": 25074, "step": 8758} +{"lr": 0.00039759922641605187, "data_time": 0.001419210433959961, "loss": 0.1138884425163269, "time": 0.7775670289993286, "epoch": 2, "memory": 25074, "step": 8858} +{"lr": 0.00039759922641605187, "data_time": 0.001520252227783203, "loss": 0.11171320900321006, "time": 0.7573474407196045, "epoch": 2, "memory": 25074, "step": 8958} +{"lr": 0.00039759922641605187, "data_time": 0.0015263319015502929, "loss": 0.10042026489973069, "time": 0.8191678762435913, "epoch": 2, "memory": 25074, "step": 9058} +{"lr": 0.00039759922641605187, "data_time": 0.0011941909790039063, "loss": 0.10328052565455437, "time": 0.7963718175888062, "epoch": 2, "memory": 25074, "step": 9158} +{"lr": 0.00039759922641605187, "data_time": 0.0018949270248413085, "loss": 0.11000730395317078, "time": 0.7700808048248291, "epoch": 2, "memory": 25074, "step": 9258} +{"lr": 0.00039759922641605187, "data_time": 0.0014587879180908204, "loss": 0.11679540202021599, "time": 0.7279536008834839, "epoch": 2, "memory": 25074, "step": 9358} +{"lr": 0.00039759922641605187, "data_time": 0.0014569282531738282, "loss": 0.11313460543751716, "time": 0.7468536138534546, "epoch": 2, "memory": 25074, "step": 9458} +{"lr": 0.00039759922641605187, "data_time": 0.0012717247009277344, "loss": 0.1270844042301178, "time": 0.7640789031982422, "epoch": 2, "memory": 25074, "step": 9558} +{"lr": 0.00039759922641605187, "data_time": 0.001318812370300293, "loss": 0.10224962010979652, "time": 0.7197729825973511, "epoch": 2, "memory": 25074, "step": 9658} +{"lr": 0.00039759922641605187, "data_time": 0.0014826059341430664, "loss": 0.1157186646014452, "time": 0.7557222366333007, "epoch": 2, "memory": 25074, "step": 9758} +{"lr": 0.00039759922641605187, "data_time": 0.001317000389099121, "loss": 0.08724111206829548, "time": 0.7586980819702148, "epoch": 2, "memory": 25074, "step": 9858} +{"lr": 0.00039759922641605187, "data_time": 0.0013677358627319336, "loss": 0.11286000311374664, "time": 0.7446044921875, "epoch": 2, "memory": 25074, "step": 9958} +{"lr": 0.00039759922641605187, "data_time": 0.0014358758926391602, "loss": 0.09231222979724407, "time": 0.7773568153381347, "epoch": 2, "memory": 25074, "step": 10058} +{"lr": 0.00039759922641605187, "data_time": 0.0011890411376953125, "loss": 0.0991213247179985, "time": 0.7307825565338135, "epoch": 2, "memory": 25074, "step": 10158} +{"lr": 0.00039759922641605187, "data_time": 0.0013567209243774414, "loss": 0.09247121214866638, "time": 0.8336795330047607, "epoch": 2, "memory": 25074, "step": 10258} +{"lr": 0.00039759922641605187, "data_time": 0.0014424324035644531, "loss": 0.08989433161914348, "time": 0.7155681610107422, "epoch": 2, "memory": 25074, "step": 10358} +{"lr": 0.00039759922641605187, "data_time": 0.0013794898986816406, "loss": 0.0911391343921423, "time": 0.7531449317932128, "epoch": 2, "memory": 25074, "step": 10458} +{"lr": 0.00039759922641605187, "data_time": 0.0010863780975341798, "loss": 0.08242987990379333, "time": 0.7036821603775024, "epoch": 2, "memory": 25074, "step": 10558} +{"lr": 0.00039759922641605187, "data_time": 0.001341533660888672, "loss": 0.09365952722728252, "time": 0.7220810174942016, "epoch": 2, "memory": 25074, "step": 10658} +{"lr": 0.00039759922641605187, "data_time": 0.0013874292373657227, "loss": 0.08166426531970501, "time": 0.7704447507858276, "epoch": 2, "memory": 25074, "step": 10758} +{"lr": 0.00039759922641605187, "data_time": 0.0013012170791625976, "loss": 0.08841171264648437, "time": 0.7600651741027832, "epoch": 2, "memory": 25074, "step": 10858} +{"lr": 0.00039759922641605187, "data_time": 0.0013210058212280273, "loss": 0.0915576845407486, "time": 0.7458028316497802, "epoch": 2, "memory": 25074, "step": 10958} +{"lr": 0.00039759922641605187, "data_time": 0.00149383544921875, "loss": 0.08014451824128628, "time": 0.7738240957260132, "epoch": 2, "memory": 25074, "step": 11058} +{"lr": 0.00039759922641605187, "data_time": 0.0016082763671875, "loss": 0.08079681731760502, "time": 0.733919644355774, "epoch": 2, "memory": 25074, "step": 11158} +{"lr": 0.00039759922641605187, "data_time": 0.0012682199478149414, "loss": 0.07264478243887425, "time": 0.779146695137024, "epoch": 2, "memory": 25074, "step": 11258} +{"lr": 0.00039759922641605187, "data_time": 0.00124969482421875, "loss": 0.06492337491363287, "time": 0.7348495483398437, "epoch": 2, "memory": 25074, "step": 11358} +{"lr": 0.00039759922641605187, "data_time": 0.00177154541015625, "loss": 0.07624929696321488, "time": 0.7161550760269165, "epoch": 2, "memory": 25074, "step": 11458} +{"accuracy/top1": 98.90158081054688, "data_time": 0.27104833546806784, "time": 0.65962955530952, "step": 2} +{"lr": 0.00039045602067755495, "data_time": 0.0011338472366333007, "loss": 0.07454138360917568, "time": 0.715718674659729, "epoch": 3, "memory": 25074, "step": 11616} +{"lr": 0.00039045602067755495, "data_time": 0.0012253999710083007, "loss": 0.06838808674365282, "time": 0.7966464519500732, "epoch": 3, "memory": 25074, "step": 11716} +{"lr": 0.00039045602067755495, "data_time": 0.0017017126083374023, "loss": 0.08313975371420383, "time": 0.7046485662460327, "epoch": 3, "memory": 25074, "step": 11816} +{"lr": 0.00039045602067755495, "data_time": 0.0014360427856445312, "loss": 0.0759969051927328, "time": 0.7288727045059205, "epoch": 3, "memory": 25074, "step": 11916} +{"lr": 0.00039045602067755495, "data_time": 0.001629948616027832, "loss": 0.06696641705930233, "time": 0.7726234197616577, "epoch": 3, "memory": 25074, "step": 12016} +{"lr": 0.00039045602067755495, "data_time": 0.0011850833892822266, "loss": 0.06603091154247523, "time": 0.7357964515686035, "epoch": 3, "memory": 25074, "step": 12116} +{"lr": 0.00039045602067755495, "data_time": 0.0013424873352050782, "loss": 0.07045400068163872, "time": 0.7128629446029663, "epoch": 3, "memory": 25074, "step": 12216} +{"lr": 0.00039045602067755495, "data_time": 0.001307225227355957, "loss": 0.06513373404741288, "time": 0.7293326139450074, "epoch": 3, "memory": 25074, "step": 12316} +{"lr": 0.00039045602067755495, "data_time": 0.0014556407928466796, "loss": 0.07489310540258884, "time": 0.7053848743438721, "epoch": 3, "memory": 25074, "step": 12416} +{"lr": 0.00039045602067755495, "data_time": 0.0012600421905517578, "loss": 0.056421241350471975, "time": 0.7707218170166016, "epoch": 3, "memory": 25074, "step": 12516} +{"lr": 0.00039045602067755495, "data_time": 0.001521897315979004, "loss": 0.07037061303853989, "time": 0.729718804359436, "epoch": 3, "memory": 25074, "step": 12616} +{"lr": 0.00039045602067755495, "data_time": 0.001941537857055664, "loss": 0.07602909170091152, "time": 0.7406081914901733, "epoch": 3, "memory": 25074, "step": 12716} +{"lr": 0.00039045602067755495, "data_time": 0.0014235734939575194, "loss": 0.07726535201072693, "time": 0.7761637449264527, "epoch": 3, "memory": 25074, "step": 12816} +{"lr": 0.00039045602067755495, "data_time": 0.0014698266983032226, "loss": 0.06499647535383701, "time": 0.7467527866363526, "epoch": 3, "memory": 25074, "step": 12916} +{"lr": 0.00039045602067755495, "data_time": 0.0013027667999267578, "loss": 0.058831796050071716, "time": 0.717350435256958, "epoch": 3, "memory": 25074, "step": 13016} +{"lr": 0.00039045602067755495, "data_time": 0.0011768579483032227, "loss": 0.056864219903945926, "time": 0.7174227476119995, "epoch": 3, "memory": 25074, "step": 13116} +{"lr": 0.00039045602067755495, "data_time": 0.0013763904571533203, "loss": 0.056697160378098486, "time": 0.774661374092102, "epoch": 3, "memory": 25074, "step": 13216} +{"lr": 0.00039045602067755495, "data_time": 0.0010921478271484375, "loss": 0.05814656335860491, "time": 0.7548272848129273, "epoch": 3, "memory": 25074, "step": 13316} +{"lr": 0.00039045602067755495, "data_time": 0.001402592658996582, "loss": 0.0690337635576725, "time": 0.7593602657318115, "epoch": 3, "memory": 25074, "step": 13416} +{"lr": 0.00039045602067755495, "data_time": 0.0016754388809204102, "loss": 0.06053033582866192, "time": 0.776843523979187, "epoch": 3, "memory": 25074, "step": 13516} +{"lr": 0.00039045602067755495, "data_time": 0.001482534408569336, "loss": 0.05673934854567051, "time": 0.7125611305236816, "epoch": 3, "memory": 25074, "step": 13616} +{"lr": 0.00039045602067755495, "data_time": 0.0016509294509887695, "loss": 0.051472777128219606, "time": 0.7779937744140625, "epoch": 3, "memory": 25074, "step": 13716} +{"lr": 0.00039045602067755495, "data_time": 0.001448345184326172, "loss": 0.047471163980662825, "time": 0.774700665473938, "epoch": 3, "memory": 25074, "step": 13816} +{"lr": 0.00039045602067755495, "data_time": 0.0014891862869262696, "loss": 0.06760014072060586, "time": 0.7781532526016235, "epoch": 3, "memory": 25074, "step": 13916} +{"lr": 0.00039045602067755495, "data_time": 0.0015434503555297851, "loss": 0.05911597870290279, "time": 0.799963402748108, "epoch": 3, "memory": 25074, "step": 14016} +{"lr": 0.00039045602067755495, "data_time": 0.0014227867126464845, "loss": 0.05952814538031816, "time": 0.7754967451095581, "epoch": 3, "memory": 25074, "step": 14116} +{"lr": 0.00039045602067755495, "data_time": 0.0020697355270385743, "loss": 0.049110569432377814, "time": 0.7596653938293457, "epoch": 3, "memory": 25074, "step": 14216} +{"lr": 0.00039045602067755495, "data_time": 0.0010865688323974609, "loss": 0.07636752352118492, "time": 0.7480664014816284, "epoch": 3, "memory": 25074, "step": 14316} +{"lr": 0.00039045602067755495, "data_time": 0.0016338348388671875, "loss": 0.07510394230484962, "time": 0.7465034484863281, "epoch": 3, "memory": 25074, "step": 14416} +{"lr": 0.00039045602067755495, "data_time": 0.0017137527465820312, "loss": 0.055041104555130005, "time": 0.7643941402435303, "epoch": 3, "memory": 25074, "step": 14516} +{"lr": 0.00039045602067755495, "data_time": 0.0013677358627319336, "loss": 0.043075740337371826, "time": 0.7085863351821899, "epoch": 3, "memory": 25074, "step": 14616} +{"lr": 0.00039045602067755495, "data_time": 0.0019581794738769533, "loss": 0.04821114744991064, "time": 0.7650643348693847, "epoch": 3, "memory": 25074, "step": 14716} +{"lr": 0.00039045602067755495, "data_time": 0.0015894174575805664, "loss": 0.04504678547382355, "time": 0.7304926156997681, "epoch": 3, "memory": 25074, "step": 14816} +{"lr": 0.00039045602067755495, "data_time": 0.0013896465301513673, "loss": 0.05232441704720259, "time": 0.7621388673782349, "epoch": 3, "memory": 25074, "step": 14916} +{"lr": 0.00039045602067755495, "data_time": 0.001965498924255371, "loss": 0.04509794358164072, "time": 0.7406379461288453, "epoch": 3, "memory": 25074, "step": 15016} +{"lr": 0.00039045602067755495, "data_time": 0.0014034509658813477, "loss": 0.0478059820830822, "time": 0.7896098375320435, "epoch": 3, "memory": 25074, "step": 15116} +{"lr": 0.00039045602067755495, "data_time": 0.0014190196990966797, "loss": 0.044007493555545805, "time": 0.7913527488708496, "epoch": 3, "memory": 25074, "step": 15216} +{"lr": 0.00039045602067755495, "data_time": 0.0017563581466674804, "loss": 0.04758833684027195, "time": 0.8260623455047608, "epoch": 3, "memory": 25074, "step": 15316} +{"lr": 0.00039045602067755495, "data_time": 0.001270627975463867, "loss": 0.0439551105722785, "time": 0.7511175394058227, "epoch": 3, "memory": 25074, "step": 15416} +{"lr": 0.00039045602067755495, "data_time": 0.001655721664428711, "loss": 0.03452812507748604, "time": 0.731798243522644, "epoch": 3, "memory": 25074, "step": 15516} +{"lr": 0.00039045602067755495, "data_time": 0.0020056724548339843, "loss": 0.04098694976419211, "time": 0.779358172416687, "epoch": 3, "memory": 25074, "step": 15616} +{"lr": 0.00039045602067755495, "data_time": 0.001270151138305664, "loss": 0.04094848036766052, "time": 0.760709261894226, "epoch": 3, "memory": 25074, "step": 15716} +{"lr": 0.00039045602067755495, "data_time": 0.0014201641082763673, "loss": 0.048485515639185905, "time": 0.7232496023178101, "epoch": 3, "memory": 25074, "step": 15816} +{"lr": 0.00039045602067755495, "data_time": 0.001566171646118164, "loss": 0.04530941024422645, "time": 0.7389329433441162, "epoch": 3, "memory": 25074, "step": 15916} +{"lr": 0.00039045602067755495, "data_time": 0.001257467269897461, "loss": 0.03874981384724378, "time": 0.8389924764633179, "epoch": 3, "memory": 25074, "step": 16016} +{"lr": 0.00039045602067755495, "data_time": 0.0012781620025634766, "loss": 0.03951667342334986, "time": 0.7242990016937256, "epoch": 3, "memory": 25074, "step": 16116} +{"lr": 0.00039045602067755495, "data_time": 0.0015552520751953125, "loss": 0.050310324877500534, "time": 0.7852558374404908, "epoch": 3, "memory": 25074, "step": 16216} +{"lr": 0.00039045602067755495, "data_time": 0.0015773773193359375, "loss": 0.049104707688093184, "time": 0.7572763919830322, "epoch": 3, "memory": 25074, "step": 16316} +{"lr": 0.00039045602067755495, "data_time": 0.0013127803802490234, "loss": 0.040728192590177056, "time": 0.8331624269485474, "epoch": 3, "memory": 25074, "step": 16416} +{"lr": 0.00039045602067755495, "data_time": 0.0013446569442749023, "loss": 0.0441434271633625, "time": 0.7584138870239258, "epoch": 3, "memory": 25074, "step": 16516} +{"lr": 0.00039045602067755495, "data_time": 0.0014935016632080078, "loss": 0.03594810180366039, "time": 0.7814659118652344, "epoch": 3, "memory": 25074, "step": 16616} +{"lr": 0.00039045602067755495, "data_time": 0.0014584779739379883, "loss": 0.05678151026368141, "time": 0.742890739440918, "epoch": 3, "memory": 25074, "step": 16716} +{"lr": 0.00039045602067755495, "data_time": 0.0012686729431152343, "loss": 0.05335999634116888, "time": 0.8077174186706543, "epoch": 3, "memory": 25074, "step": 16816} +{"lr": 0.00039045602067755495, "data_time": 0.0013684511184692382, "loss": 0.045536359772086143, "time": 0.7622642040252685, "epoch": 3, "memory": 25074, "step": 16916} +{"lr": 0.00039045602067755495, "data_time": 0.0012946367263793946, "loss": 0.039878766797482965, "time": 0.735869288444519, "epoch": 3, "memory": 25074, "step": 17016} +{"lr": 0.00039045602067755495, "data_time": 0.0014983654022216798, "loss": 0.033761563571169974, "time": 0.7622643947601319, "epoch": 3, "memory": 25074, "step": 17116} +{"lr": 0.00039045602067755495, "data_time": 0.0013837099075317382, "loss": 0.036536894738674164, "time": 0.7115382432937623, "epoch": 3, "memory": 25074, "step": 17216} +{"accuracy/top1": 99.36241149902344, "data_time": 0.2469493080587948, "time": 0.6325257665970746, "step": 3} +{"lr": 0.0003787462722167317, "data_time": 0.03865253925323486, "loss": 0.039829705841839314, "time": 0.7850553512573242, "epoch": 4, "memory": 25074, "step": 17374} +{"lr": 0.0003787462722167317, "data_time": 0.05717322826385498, "loss": 0.0474745761603117, "time": 0.8118953227996826, "epoch": 4, "memory": 25074, "step": 17474} +{"lr": 0.0003787462722167317, "data_time": 0.0014620780944824218, "loss": 0.032372019998729226, "time": 0.7516115665435791, "epoch": 4, "memory": 25074, "step": 17574} +{"lr": 0.0003787462722167317, "data_time": 0.08182368278503419, "loss": 0.0328549487516284, "time": 0.7947424650192261, "epoch": 4, "memory": 25074, "step": 17674} +{"lr": 0.0003787462722167317, "data_time": 0.001350259780883789, "loss": 0.03159841159358621, "time": 0.7455472469329834, "epoch": 4, "memory": 25074, "step": 17774} +{"lr": 0.0003787462722167317, "data_time": 0.0016098976135253905, "loss": 0.03146227048709989, "time": 0.7151310682296753, "epoch": 4, "memory": 25074, "step": 17874} +{"lr": 0.0003787462722167317, "data_time": 0.0015421152114868165, "loss": 0.03269414212554693, "time": 0.7611994504928589, "epoch": 4, "memory": 25074, "step": 17974} +{"lr": 0.0003787462722167317, "data_time": 0.0012829303741455078, "loss": 0.030372516997158528, "time": 0.7356581211090087, "epoch": 4, "memory": 25074, "step": 18074} +{"lr": 0.0003787462722167317, "data_time": 0.0012136936187744141, "loss": 0.04282185938209295, "time": 0.7142583131790161, "epoch": 4, "memory": 25074, "step": 18174} +{"lr": 0.0003787462722167317, "data_time": 0.0013745784759521484, "loss": 0.0318029697984457, "time": 0.8543846607208252, "epoch": 4, "memory": 25074, "step": 18274} +{"lr": 0.0003787462722167317, "data_time": 0.0014328718185424804, "loss": 0.04707674775272608, "time": 0.7077360153198242, "epoch": 4, "memory": 25074, "step": 18374} +{"lr": 0.0003787462722167317, "data_time": 0.0014774799346923828, "loss": 0.03311942107975483, "time": 0.7490308523178101, "epoch": 4, "memory": 25074, "step": 18474} +{"lr": 0.0003787462722167317, "data_time": 0.001595759391784668, "loss": 0.02794723752886057, "time": 0.7808529376983643, "epoch": 4, "memory": 25074, "step": 18574} +{"lr": 0.0003787462722167317, "data_time": 0.001329803466796875, "loss": 0.025707329018041493, "time": 0.7930867671966553, "epoch": 4, "memory": 25074, "step": 18674} +{"lr": 0.0003787462722167317, "data_time": 0.0021139860153198244, "loss": 0.02396458266302943, "time": 0.8923398017883301, "epoch": 4, "memory": 25074, "step": 18774} +{"lr": 0.0003787462722167317, "data_time": 0.0012781143188476563, "loss": 0.019072595937177538, "time": 0.7380176544189453, "epoch": 4, "memory": 25074, "step": 18874} +{"lr": 0.0003787462722167317, "data_time": 0.002022719383239746, "loss": 0.028676498495042325, "time": 0.7897438526153564, "epoch": 4, "memory": 25074, "step": 18974} +{"lr": 0.0003787462722167317, "data_time": 0.0016179084777832031, "loss": 0.02655111365020275, "time": 0.7565790176391601, "epoch": 4, "memory": 25074, "step": 19074} +{"lr": 0.0003787462722167317, "data_time": 0.0013271093368530274, "loss": 0.02916183224879205, "time": 0.7686939239501953, "epoch": 4, "memory": 25074, "step": 19174} +{"lr": 0.0003787462722167317, "data_time": 0.0014940023422241211, "loss": 0.030577187798917292, "time": 0.7645267724990845, "epoch": 4, "memory": 25074, "step": 19274} +{"lr": 0.0003787462722167317, "data_time": 0.00135495662689209, "loss": 0.029290075413882732, "time": 0.8134437322616577, "epoch": 4, "memory": 25074, "step": 19374} +{"lr": 0.0003787462722167317, "data_time": 0.0014248132705688477, "loss": 0.029681845940649508, "time": 0.7550102472305298, "epoch": 4, "memory": 25074, "step": 19474} +{"lr": 0.0003787462722167317, "data_time": 0.0013571977615356445, "loss": 0.03532544211484492, "time": 0.7044047355651856, "epoch": 4, "memory": 25074, "step": 19574} +{"lr": 0.0003787462722167317, "data_time": 0.001385045051574707, "loss": 0.02183720823377371, "time": 0.7184542417526245, "epoch": 4, "memory": 25074, "step": 19674} +{"lr": 0.0003787462722167317, "data_time": 0.0016919612884521485, "loss": 0.034702480956912043, "time": 0.723512840270996, "epoch": 4, "memory": 25074, "step": 19774} +{"lr": 0.0003787462722167317, "data_time": 0.001338648796081543, "loss": 0.03200682578608394, "time": 0.7607735872268677, "epoch": 4, "memory": 25074, "step": 19874} +{"lr": 0.0003787462722167317, "data_time": 0.0015716075897216797, "loss": 0.028486664686352015, "time": 0.7250436305999756, "epoch": 4, "memory": 25074, "step": 19974} +{"lr": 0.0003787462722167317, "data_time": 0.0015590429306030274, "loss": 0.017885612230747937, "time": 0.7812924385070801, "epoch": 4, "memory": 25074, "step": 20074} +{"lr": 0.0003787462722167317, "data_time": 0.0012953758239746093, "loss": 0.018333968101069333, "time": 0.7683312892913818, "epoch": 4, "memory": 25074, "step": 20174} +{"lr": 0.0003787462722167317, "data_time": 0.0013596534729003907, "loss": 0.024994514510035516, "time": 0.8128173589706421, "epoch": 4, "memory": 25074, "step": 20274} +{"lr": 0.0003787462722167317, "data_time": 0.0014835119247436524, "loss": 0.025219791987910866, "time": 0.7945241689682007, "epoch": 4, "memory": 25074, "step": 20374} +{"lr": 0.0003787462722167317, "data_time": 0.0013244867324829102, "loss": 0.02037292649038136, "time": 0.7190348148345947, "epoch": 4, "memory": 25074, "step": 20474} +{"lr": 0.0003787462722167317, "data_time": 0.001303553581237793, "loss": 0.024501336086541415, "time": 0.8481273174285888, "epoch": 4, "memory": 25074, "step": 20574} +{"lr": 0.0003787462722167317, "data_time": 0.001217031478881836, "loss": 0.03265264169313013, "time": 0.8404972791671753, "epoch": 4, "memory": 25074, "step": 20674} +{"lr": 0.0003787462722167317, "data_time": 0.0015292167663574219, "loss": 0.02086063758470118, "time": 0.8199644804000854, "epoch": 4, "memory": 25074, "step": 20774} +{"lr": 0.0003787462722167317, "data_time": 0.0013418912887573243, "loss": 0.015881731593981385, "time": 0.7641673564910889, "epoch": 4, "memory": 25074, "step": 20874} +{"lr": 0.0003787462722167317, "data_time": 0.0012826919555664062, "loss": 0.02060639802366495, "time": 0.7679383039474488, "epoch": 4, "memory": 25074, "step": 20974} +{"lr": 0.0003787462722167317, "data_time": 0.0013321876525878907, "loss": 0.017834770819172265, "time": 0.7576194524765014, "epoch": 4, "memory": 25074, "step": 21074} +{"lr": 0.0003787462722167317, "data_time": 0.0013240814208984376, "loss": 0.027906512189656497, "time": 0.7413241147994996, "epoch": 4, "memory": 25074, "step": 21174} +{"lr": 0.0003787462722167317, "data_time": 0.001589059829711914, "loss": 0.01876042317599058, "time": 0.811037015914917, "epoch": 4, "memory": 25074, "step": 21274} +{"lr": 0.0003787462722167317, "data_time": 0.0024838924407958986, "loss": 0.018446039967238904, "time": 0.8231079816818238, "epoch": 4, "memory": 25074, "step": 21374} +{"lr": 0.0003787462722167317, "data_time": 0.001452183723449707, "loss": 0.02220039302483201, "time": 0.7696118116378784, "epoch": 4, "memory": 25074, "step": 21474} +{"lr": 0.0003787462722167317, "data_time": 0.001588726043701172, "loss": 0.028869006037712096, "time": 0.8068956136703491, "epoch": 4, "memory": 25074, "step": 21574} +{"lr": 0.0003787462722167317, "data_time": 0.001605987548828125, "loss": 0.02095076614059508, "time": 0.75597984790802, "epoch": 4, "memory": 25074, "step": 21674} +{"lr": 0.0003787462722167317, "data_time": 0.001576685905456543, "loss": 0.016870320495218038, "time": 0.7346962928771973, "epoch": 4, "memory": 25074, "step": 21774} +{"lr": 0.0003787462722167317, "data_time": 0.0014455080032348632, "loss": 0.015548450499773025, "time": 0.7052653551101684, "epoch": 4, "memory": 25074, "step": 21874} +{"lr": 0.0003787462722167317, "data_time": 0.0014158725738525391, "loss": 0.019401692179962994, "time": 0.7833166599273682, "epoch": 4, "memory": 25074, "step": 21974} +{"lr": 0.0003787462722167317, "data_time": 0.001394367218017578, "loss": 0.0162011640612036, "time": 0.8173018217086792, "epoch": 4, "memory": 25074, "step": 22074} +{"lr": 0.0003787462722167317, "data_time": 0.001615428924560547, "loss": 0.013591721281409263, "time": 0.758896803855896, "epoch": 4, "memory": 25074, "step": 22174} +{"lr": 0.0003787462722167317, "data_time": 0.0013357639312744141, "loss": 0.01157363336533308, "time": 0.811647915840149, "epoch": 4, "memory": 25074, "step": 22274} +{"lr": 0.0003787462722167317, "data_time": 0.0014384984970092773, "loss": 0.02888364139944315, "time": 0.7266649007797241, "epoch": 4, "memory": 25074, "step": 22374} +{"lr": 0.0003787462722167317, "data_time": 0.0012154817581176759, "loss": 0.025532242748886345, "time": 0.7530094861984253, "epoch": 4, "memory": 25074, "step": 22474} +{"lr": 0.0003787462722167317, "data_time": 0.0015397071838378906, "loss": 0.01601006994023919, "time": 0.7512917518615723, "epoch": 4, "memory": 25074, "step": 22574} +{"lr": 0.0003787462722167317, "data_time": 0.0015405654907226563, "loss": 0.017409794451668858, "time": 0.8407786607742309, "epoch": 4, "memory": 25074, "step": 22674} +{"lr": 0.0003787462722167317, "data_time": 0.0013744115829467773, "loss": 0.013993551721796393, "time": 0.7365664720535279, "epoch": 4, "memory": 25074, "step": 22774} +{"lr": 0.0003787462722167317, "data_time": 0.0016142606735229492, "loss": 0.016717789648100733, "time": 0.7249078512191772, "epoch": 4, "memory": 25074, "step": 22874} +{"lr": 0.0003787462722167317, "data_time": 0.0013981103897094727, "loss": 0.02044197516515851, "time": 0.7902284145355225, "epoch": 4, "memory": 25074, "step": 22974} +{"accuracy/top1": 99.2677230834961, "data_time": 0.25519415911506205, "time": 0.6416875614839441, "step": 4} +{"lr": 0.00036275831390311474, "data_time": 0.001486825942993164, "loss": 0.01302544167265296, "time": 0.7256197690963745, "epoch": 5, "memory": 25074, "step": 23132} +{"lr": 0.00036275831390311474, "data_time": 0.030049228668212892, "loss": 0.017989942478016017, "time": 0.7324834346771241, "epoch": 5, "memory": 25074, "step": 23232} +{"lr": 0.00036275831390311474, "data_time": 0.0012367963790893555, "loss": 0.02218693825416267, "time": 0.7051385164260864, "epoch": 5, "memory": 25074, "step": 23332} +{"lr": 0.00036275831390311474, "data_time": 0.0015694379806518554, "loss": 0.012043322250247002, "time": 0.7431371688842774, "epoch": 5, "memory": 25074, "step": 23432} +{"lr": 0.00036275831390311474, "data_time": 0.0012540340423583985, "loss": 0.01390527212060988, "time": 0.8130687713623047, "epoch": 5, "memory": 25074, "step": 23532} +{"lr": 0.00036275831390311474, "data_time": 0.0015015363693237304, "loss": 0.018388553336262704, "time": 0.7184257507324219, "epoch": 5, "memory": 25074, "step": 23632} +{"lr": 0.00036275831390311474, "data_time": 0.0011070966720581055, "loss": 0.020958791440352797, "time": 0.7739711046218872, "epoch": 5, "memory": 25074, "step": 23732} +{"lr": 0.00036275831390311474, "data_time": 0.0013579845428466797, "loss": 0.019040547451004385, "time": 0.7628731012344361, "epoch": 5, "memory": 25074, "step": 23832} +{"lr": 0.00036275831390311474, "data_time": 0.0012732505798339843, "loss": 0.01234361007809639, "time": 0.8346306324005127, "epoch": 5, "memory": 25074, "step": 23932} +{"lr": 0.00036275831390311474, "data_time": 0.0015250205993652343, "loss": 0.015864039305597544, "time": 0.8216228723526001, "epoch": 5, "memory": 25074, "step": 24032} +{"lr": 0.00036275831390311474, "data_time": 0.0015598535537719727, "loss": 0.011547359358519315, "time": 0.7645201683044434, "epoch": 5, "memory": 25074, "step": 24132} +{"lr": 0.00036275831390311474, "data_time": 0.0014086246490478515, "loss": 0.026875092508271338, "time": 0.7816853761672974, "epoch": 5, "memory": 25074, "step": 24232} +{"lr": 0.00036275831390311474, "data_time": 0.0015123605728149414, "loss": 0.016534141125157477, "time": 0.7818456411361694, "epoch": 5, "memory": 25074, "step": 24332} +{"lr": 0.00036275831390311474, "data_time": 0.001155686378479004, "loss": 0.013718874007463456, "time": 0.7890047073364258, "epoch": 5, "memory": 25074, "step": 24432} +{"lr": 0.00036275831390311474, "data_time": 0.001723313331604004, "loss": 0.02003337643109262, "time": 0.7882824182510376, "epoch": 5, "memory": 25074, "step": 24532} +{"lr": 0.00036275831390311474, "data_time": 0.0013726234436035156, "loss": 0.028694703336805107, "time": 0.7442945003509521, "epoch": 5, "memory": 25074, "step": 24632} +{"lr": 0.00036275831390311474, "data_time": 0.0013441324234008789, "loss": 0.016452176216989755, "time": 0.7988262414932251, "epoch": 5, "memory": 25074, "step": 24732} +{"lr": 0.00036275831390311474, "data_time": 0.0014655351638793944, "loss": 0.012544103804975748, "time": 0.8010499000549316, "epoch": 5, "memory": 25074, "step": 24832} +{"lr": 0.00036275831390311474, "data_time": 0.001410841941833496, "loss": 0.017276473995298147, "time": 0.7862449407577514, "epoch": 5, "memory": 25074, "step": 24932} +{"lr": 0.00036275831390311474, "data_time": 0.0013671875, "loss": 0.011573469080030919, "time": 0.7392837285995484, "epoch": 5, "memory": 25074, "step": 25032} +{"lr": 0.00036275831390311474, "data_time": 0.07601547241210938, "loss": 0.014642014773562551, "time": 0.7981736898422241, "epoch": 5, "memory": 25074, "step": 25132} +{"lr": 0.00036275831390311474, "data_time": 0.001341557502746582, "loss": 0.014687849627807736, "time": 0.7996220111846923, "epoch": 5, "memory": 25074, "step": 25232} +{"lr": 0.00036275831390311474, "data_time": 0.0012368202209472657, "loss": 0.01322974874638021, "time": 0.7734551429748535, "epoch": 5, "memory": 25074, "step": 25332} +{"lr": 0.00036275831390311474, "data_time": 0.057136940956115725, "loss": 0.01116472678259015, "time": 0.8140712976455688, "epoch": 5, "memory": 25074, "step": 25432} +{"lr": 0.00036275831390311474, "data_time": 0.00757746696472168, "loss": 0.013869006372988223, "time": 0.7679396152496338, "epoch": 5, "memory": 25074, "step": 25532} +{"lr": 0.00036275831390311474, "data_time": 0.07710325717926025, "loss": 0.015432554762810469, "time": 0.7800628423690796, "epoch": 5, "memory": 25074, "step": 25632} +{"lr": 0.00036275831390311474, "data_time": 0.09012775421142578, "loss": 0.021824515145272016, "time": 0.8040318012237548, "epoch": 5, "memory": 25074, "step": 25732} +{"lr": 0.00036275831390311474, "data_time": 0.033565640449523926, "loss": 0.010669211437925697, "time": 0.7368215560913086, "epoch": 5, "memory": 25074, "step": 25832} +{"lr": 0.00036275831390311474, "data_time": 0.08241913318634034, "loss": 0.013968339376151561, "time": 0.7856771230697632, "epoch": 5, "memory": 25074, "step": 25932} +{"lr": 0.00036275831390311474, "data_time": 0.09301645755767822, "loss": 0.019537716545164584, "time": 0.7979531526565552, "epoch": 5, "memory": 25074, "step": 26032} +{"lr": 0.00036275831390311474, "data_time": 0.045124602317810056, "loss": 0.019960994319990277, "time": 0.7569398641586303, "epoch": 5, "memory": 25074, "step": 26132} +{"lr": 0.00036275831390311474, "data_time": 0.001261758804321289, "loss": 0.0166315873619169, "time": 0.7980321645736694, "epoch": 5, "memory": 25074, "step": 26232} +{"lr": 0.00036275831390311474, "data_time": 0.05023837089538574, "loss": 0.018772314256057143, "time": 0.7643418788909913, "epoch": 5, "memory": 25074, "step": 26332} +{"lr": 0.00036275831390311474, "data_time": 0.08424079418182373, "loss": 0.0145819500554353, "time": 0.7906779527664185, "epoch": 5, "memory": 25074, "step": 26432} +{"lr": 0.00036275831390311474, "data_time": 0.0105849027633667, "loss": 0.01371259642764926, "time": 0.7200708627700806, "epoch": 5, "memory": 25074, "step": 26532} +{"lr": 0.00036275831390311474, "data_time": 0.001291799545288086, "loss": 0.009983195271342993, "time": 0.7610705852508545, "epoch": 5, "memory": 25074, "step": 26632} +{"lr": 0.00036275831390311474, "data_time": 0.03467512130737305, "loss": 0.01607303861528635, "time": 0.7847788810729981, "epoch": 5, "memory": 25074, "step": 26732} +{"lr": 0.00036275831390311474, "data_time": 0.0016903877258300781, "loss": 0.01597420508041978, "time": 0.7835286378860473, "epoch": 5, "memory": 25074, "step": 26832} +{"lr": 0.00036275831390311474, "data_time": 0.0014004230499267579, "loss": 0.017677761614322662, "time": 0.7884481906890869, "epoch": 5, "memory": 25074, "step": 26932} +{"lr": 0.00036275831390311474, "data_time": 0.0013092994689941407, "loss": 0.012730128457769751, "time": 0.763460350036621, "epoch": 5, "memory": 25074, "step": 27032} +{"lr": 0.00036275831390311474, "data_time": 0.0012837886810302735, "loss": 0.010222951741889119, "time": 0.8114837169647217, "epoch": 5, "memory": 25074, "step": 27132} +{"lr": 0.00036275831390311474, "data_time": 0.0012880086898803711, "loss": 0.011691733077168465, "time": 0.8062796115875244, "epoch": 5, "memory": 25074, "step": 27232} +{"lr": 0.00036275831390311474, "data_time": 0.0011825799942016602, "loss": 0.01129258768633008, "time": 0.7690437078475952, "epoch": 5, "memory": 25074, "step": 27332} +{"lr": 0.00036275831390311474, "data_time": 0.0012610912322998046, "loss": 0.011570308823138475, "time": 0.8083793640136718, "epoch": 5, "memory": 25074, "step": 27432} +{"lr": 0.00036275831390311474, "data_time": 0.001442551612854004, "loss": 0.011827259371057153, "time": 0.7783132553100586, "epoch": 5, "memory": 25074, "step": 27532} +{"lr": 0.00036275831390311474, "data_time": 0.0014040946960449218, "loss": 0.013755986001342535, "time": 0.7516528844833374, "epoch": 5, "memory": 25074, "step": 27632} +{"lr": 0.00036275831390311474, "data_time": 0.003322935104370117, "loss": 0.007357860030606389, "time": 0.8465522289276123, "epoch": 5, "memory": 25074, "step": 27732} +{"lr": 0.00036275831390311474, "data_time": 0.0014224290847778321, "loss": 0.01856976402923465, "time": 0.8229195356369019, "epoch": 5, "memory": 25074, "step": 27832} +{"lr": 0.00036275831390311474, "data_time": 0.0012489557266235352, "loss": 0.008409297186881303, "time": 0.8426750421524047, "epoch": 5, "memory": 25074, "step": 27932} +{"lr": 0.00036275831390311474, "data_time": 0.0014356851577758789, "loss": 0.009741560509428381, "time": 0.7991224527359009, "epoch": 5, "memory": 25074, "step": 28032} +{"lr": 0.00036275831390311474, "data_time": 0.0013633966445922852, "loss": 0.010366294672712684, "time": 0.8169411897659302, "epoch": 5, "memory": 25074, "step": 28132} +{"lr": 0.00036275831390311474, "data_time": 0.0012496471405029296, "loss": 0.009001431753858924, "time": 0.81349618434906, "epoch": 5, "memory": 25074, "step": 28232} +{"lr": 0.00036275831390311474, "data_time": 0.0013092756271362305, "loss": 0.021743905264884233, "time": 0.8370378494262696, "epoch": 5, "memory": 25074, "step": 28332} +{"lr": 0.00036275831390311474, "data_time": 0.0013876914978027343, "loss": 0.01237932126969099, "time": 0.7648314237594604, "epoch": 5, "memory": 25074, "step": 28432} +{"lr": 0.00036275831390311474, "data_time": 0.0015105724334716797, "loss": 0.01142486920580268, "time": 0.8360764026641846, "epoch": 5, "memory": 25074, "step": 28532} +{"lr": 0.00036275831390311474, "data_time": 0.001252460479736328, "loss": 0.007908255234360695, "time": 0.8074623107910156, "epoch": 5, "memory": 25074, "step": 28632} +{"lr": 0.00036275831390311474, "data_time": 0.0013581275939941405, "loss": 0.018199953250586987, "time": 0.803422212600708, "epoch": 5, "memory": 25074, "step": 28732} +{"accuracy/top1": 99.44447326660156, "data_time": 0.2366017874549417, "time": 0.622307160321404, "step": 5} +{"lr": 0.00034288582233137673, "data_time": 0.0012431144714355469, "loss": 0.01467427983880043, "time": 0.8483107328414917, "epoch": 6, "memory": 25074, "step": 28890} +{"lr": 0.00034288582233137673, "data_time": 0.0013927221298217773, "loss": 0.013072089897468686, "time": 0.7660422325134277, "epoch": 6, "memory": 25074, "step": 28990} +{"lr": 0.00034288582233137673, "data_time": 0.0014702320098876954, "loss": 0.006401233281940222, "time": 0.7825122356414795, "epoch": 6, "memory": 25074, "step": 29090} +{"lr": 0.00034288582233137673, "data_time": 0.001685357093811035, "loss": 0.00809799530543387, "time": 0.7979695558547973, "epoch": 6, "memory": 25074, "step": 29190} +{"lr": 0.00034288582233137673, "data_time": 0.0014662742614746094, "loss": 0.013530059484764934, "time": 0.8616000890731812, "epoch": 6, "memory": 25074, "step": 29290} +{"lr": 0.00034288582233137673, "data_time": 0.0015248775482177735, "loss": 0.01676182746887207, "time": 0.7762357234954834, "epoch": 6, "memory": 25074, "step": 29390} +{"lr": 0.00034288582233137673, "data_time": 0.0013289928436279296, "loss": 0.014427924575284124, "time": 0.7673114776611328, "epoch": 6, "memory": 25074, "step": 29490} +{"lr": 0.00034288582233137673, "data_time": 0.001611328125, "loss": 0.011207863595336676, "time": 1.3448450803756713, "epoch": 6, "memory": 25074, "step": 29590} +{"lr": 0.00034288582233137673, "data_time": 0.0012563467025756836, "loss": 0.013308819988742471, "time": 0.7337980508804322, "epoch": 6, "memory": 25074, "step": 29690} +{"lr": 0.00034288582233137673, "data_time": 0.001601409912109375, "loss": 0.008573074685409665, "time": 0.7987351894378663, "epoch": 6, "memory": 25074, "step": 29790} +{"lr": 0.00034288582233137673, "data_time": 0.0014267444610595703, "loss": 0.010339982621371747, "time": 0.7991379022598266, "epoch": 6, "memory": 25074, "step": 29890} +{"lr": 0.00034288582233137673, "data_time": 0.0014389276504516602, "loss": 0.006984920892864466, "time": 0.7865674257278442, "epoch": 6, "memory": 25074, "step": 29990} +{"lr": 0.00034288582233137673, "data_time": 0.052128958702087405, "loss": 0.013638529181480407, "time": 0.7602356195449829, "epoch": 6, "memory": 25074, "step": 30090} +{"lr": 0.00034288582233137673, "data_time": 0.059166574478149415, "loss": 0.008216159651055932, "time": 0.7738542079925537, "epoch": 6, "memory": 25074, "step": 30190} +{"lr": 0.00034288582233137673, "data_time": 0.001326298713684082, "loss": 0.008545483695343137, "time": 0.7616467952728272, "epoch": 6, "memory": 25074, "step": 30290} +{"lr": 0.00034288582233137673, "data_time": 0.0013590335845947265, "loss": 0.00730477231554687, "time": 0.7734612703323365, "epoch": 6, "memory": 25074, "step": 30390} +{"lr": 0.00034288582233137673, "data_time": 0.00142977237701416, "loss": 0.008598646288737655, "time": 0.8029321432113647, "epoch": 6, "memory": 25074, "step": 30490} +{"lr": 0.00034288582233137673, "data_time": 0.0014775753021240234, "loss": 0.01053528799675405, "time": 0.8208680391311646, "epoch": 6, "memory": 25074, "step": 30590} +{"lr": 0.00034288582233137673, "data_time": 0.0016880273818969727, "loss": 0.005710984487086535, "time": 0.8045543432235718, "epoch": 6, "memory": 25074, "step": 30690} +{"lr": 0.00034288582233137673, "data_time": 0.001369786262512207, "loss": 0.010369472578167916, "time": 0.7910224437713623, "epoch": 6, "memory": 25074, "step": 30790} +{"lr": 0.00034288582233137673, "data_time": 0.0013910055160522461, "loss": 0.012486135913059115, "time": 0.7808441877365112, "epoch": 6, "memory": 25074, "step": 30890} +{"lr": 0.00034288582233137673, "data_time": 0.0015449047088623047, "loss": 0.009330012183636428, "time": 0.8098372220993042, "epoch": 6, "memory": 25074, "step": 30990} +{"lr": 0.00034288582233137673, "data_time": 0.0015043497085571289, "loss": 0.009312838781625032, "time": 0.7602557420730591, "epoch": 6, "memory": 25074, "step": 31090} +{"lr": 0.00034288582233137673, "data_time": 0.0013005495071411132, "loss": 0.00718923076055944, "time": 0.7536764860153198, "epoch": 6, "memory": 25074, "step": 31190} +{"lr": 0.00034288582233137673, "data_time": 0.0012532949447631836, "loss": 0.011695928080007434, "time": 0.776336407661438, "epoch": 6, "memory": 25074, "step": 31290} +{"lr": 0.00034288582233137673, "data_time": 0.0011270284652709962, "loss": 0.008433237066492438, "time": 0.7822027921676635, "epoch": 6, "memory": 25074, "step": 31390} +{"lr": 0.00034288582233137673, "data_time": 0.0012879371643066406, "loss": 0.011809056997299195, "time": 0.7624987840652466, "epoch": 6, "memory": 25074, "step": 31490} +{"lr": 0.00034288582233137673, "data_time": 0.001293039321899414, "loss": 0.008680937206372618, "time": 0.8256975173950195, "epoch": 6, "memory": 25074, "step": 31590} +{"lr": 0.00034288582233137673, "data_time": 0.0011420726776123046, "loss": 0.007531129196286202, "time": 0.8245503425598144, "epoch": 6, "memory": 25074, "step": 31690} +{"lr": 0.00034288582233137673, "data_time": 0.0014070034027099609, "loss": 0.008921292377635837, "time": 0.8015362977981567, "epoch": 6, "memory": 25074, "step": 31790} +{"lr": 0.00034288582233137673, "data_time": 0.001501154899597168, "loss": 0.008194751432165503, "time": 0.7953985929489136, "epoch": 6, "memory": 25074, "step": 31890} +{"lr": 0.00034288582233137673, "data_time": 0.0012296915054321289, "loss": 0.007451539672911167, "time": 0.754914116859436, "epoch": 6, "memory": 25074, "step": 31990} +{"lr": 0.00034288582233137673, "data_time": 0.001436305046081543, "loss": 0.005120963882654905, "time": 0.8026779651641845, "epoch": 6, "memory": 25074, "step": 32090} +{"lr": 0.00034288582233137673, "data_time": 0.0012905120849609375, "loss": 0.01344825685955584, "time": 0.8168651342391968, "epoch": 6, "memory": 25074, "step": 32190} +{"lr": 0.00034288582233137673, "data_time": 0.0013339757919311524, "loss": 0.009167814254760742, "time": 0.8336549282073975, "epoch": 6, "memory": 25074, "step": 32290} +{"lr": 0.00034288582233137673, "data_time": 0.0016960382461547851, "loss": 0.010822345362976193, "time": 0.7926252603530883, "epoch": 6, "memory": 25074, "step": 32390} +{"lr": 0.00034288582233137673, "data_time": 0.001642751693725586, "loss": 0.006898094387724996, "time": 0.7743218898773193, "epoch": 6, "memory": 25074, "step": 32490} +{"lr": 0.00034288582233137673, "data_time": 0.0012653112411499024, "loss": 0.008719425462186337, "time": 0.7710724830627441, "epoch": 6, "memory": 25074, "step": 32590} +{"lr": 0.00034288582233137673, "data_time": 0.0011759519577026368, "loss": 0.00417350516654551, "time": 0.7580895185470581, "epoch": 6, "memory": 25074, "step": 32690} +{"lr": 0.00034288582233137673, "data_time": 0.0012284517288208008, "loss": 0.014815912302583455, "time": 0.7413803100585937, "epoch": 6, "memory": 25074, "step": 32790} +{"lr": 0.00034288582233137673, "data_time": 0.0014820098876953125, "loss": 0.010735587356612087, "time": 0.7420071363449097, "epoch": 6, "memory": 25074, "step": 32890} +{"lr": 0.00034288582233137673, "data_time": 0.0014560222625732422, "loss": 0.004798201657831669, "time": 0.8103760004043579, "epoch": 6, "memory": 25074, "step": 32990} +{"lr": 0.00034288582233137673, "data_time": 0.001189279556274414, "loss": 0.013060365384444595, "time": 0.7591584682464599, "epoch": 6, "memory": 25074, "step": 33090} +{"lr": 0.00034288582233137673, "data_time": 0.0012809276580810548, "loss": 0.008555329404771328, "time": 0.7963660717010498, "epoch": 6, "memory": 25074, "step": 33190} +{"lr": 0.00034288582233137673, "data_time": 0.05887458324432373, "loss": 0.013045204104855657, "time": 0.7747400522232055, "epoch": 6, "memory": 25074, "step": 33290} +{"lr": 0.00034288582233137673, "data_time": 0.0030554771423339845, "loss": 0.012456198781728744, "time": 0.7957233428955078, "epoch": 6, "memory": 25074, "step": 33390} +{"lr": 0.00034288582233137673, "data_time": 0.07650580406188964, "loss": 0.006107565620914102, "time": 0.780707859992981, "epoch": 6, "memory": 25074, "step": 33490} +{"lr": 0.00034288582233137673, "data_time": 0.055533742904663085, "loss": 0.010568537842482328, "time": 0.8367711067199707, "epoch": 6, "memory": 25074, "step": 33590} +{"lr": 0.00034288582233137673, "data_time": 0.0018268585205078124, "loss": 0.009524564957246184, "time": 0.7714697599411011, "epoch": 6, "memory": 25074, "step": 33690} +{"lr": 0.00034288582233137673, "data_time": 0.026483893394470215, "loss": 0.017681375658139587, "time": 0.8249334335327149, "epoch": 6, "memory": 25074, "step": 33790} +{"lr": 0.00034288582233137673, "data_time": 0.05687577724456787, "loss": 0.005821938579902053, "time": 0.7611339092254639, "epoch": 6, "memory": 25074, "step": 33890} +{"lr": 0.00034288582233137673, "data_time": 0.12212176322937011, "loss": 0.008629760798066854, "time": 0.8271709680557251, "epoch": 6, "memory": 25074, "step": 33990} +{"lr": 0.00034288582233137673, "data_time": 0.04865329265594483, "loss": 0.00526374327018857, "time": 0.7964529037475586, "epoch": 6, "memory": 25074, "step": 34090} +{"lr": 0.00034288582233137673, "data_time": 0.11214594841003418, "loss": 0.012475757859647274, "time": 0.8364870309829712, "epoch": 6, "memory": 25074, "step": 34190} +{"lr": 0.00034288582233137673, "data_time": 0.001747274398803711, "loss": 0.012195754889398813, "time": 0.7567744016647339, "epoch": 6, "memory": 25074, "step": 34290} +{"lr": 0.00034288582233137673, "data_time": 0.0012288808822631836, "loss": 0.010368771757930518, "time": 0.8497467041015625, "epoch": 6, "memory": 25074, "step": 34390} +{"lr": 0.00034288582233137673, "data_time": 0.0013932228088378907, "loss": 0.013538908492773772, "time": 0.8504750967025757, "epoch": 6, "memory": 25074, "step": 34490} +{"accuracy/top1": 99.76011657714844, "data_time": 0.2468103521010455, "time": 0.6382620755363914, "step": 6} +{"lr": 0.00031961812419703224, "data_time": 0.0013899564743041991, "loss": 0.006582147255539894, "time": 0.7950494766235352, "epoch": 7, "memory": 25074, "step": 34648} +{"lr": 0.00031961812419703224, "data_time": 0.0011787176132202148, "loss": 0.005933204246684909, "time": 0.7777802228927613, "epoch": 7, "memory": 25074, "step": 34748} +{"lr": 0.00031961812419703224, "data_time": 0.0014272689819335937, "loss": 0.00582629581913352, "time": 0.8077687501907349, "epoch": 7, "memory": 25074, "step": 34848} +{"lr": 0.00031961812419703224, "data_time": 0.0014104366302490235, "loss": 0.010212561720982194, "time": 0.8208668470382691, "epoch": 7, "memory": 25074, "step": 34948} +{"lr": 0.00031961812419703224, "data_time": 0.0013582706451416016, "loss": 0.005470056552439928, "time": 0.7720654726028442, "epoch": 7, "memory": 25074, "step": 35048} +{"lr": 0.00031961812419703224, "data_time": 0.0014965534210205078, "loss": 0.007361783506348729, "time": 0.7302881956100464, "epoch": 7, "memory": 25074, "step": 35148} +{"lr": 0.00031961812419703224, "data_time": 0.0010710477828979493, "loss": 0.008483129786327481, "time": 0.8307805299758911, "epoch": 7, "memory": 25074, "step": 35248} +{"lr": 0.00031961812419703224, "data_time": 0.0015103578567504882, "loss": 0.007372635323554278, "time": 0.8447330236434937, "epoch": 7, "memory": 25074, "step": 35348} +{"lr": 0.00031961812419703224, "data_time": 0.0013512372970581055, "loss": 0.0020218949764966966, "time": 0.7620120525360108, "epoch": 7, "memory": 25074, "step": 35448} +{"lr": 0.00031961812419703224, "data_time": 0.0013833284378051759, "loss": 0.006898512272164226, "time": 0.7676534175872802, "epoch": 7, "memory": 25074, "step": 35548} +{"lr": 0.00031961812419703224, "data_time": 0.0015387773513793946, "loss": 0.005877008941024542, "time": 0.8133733987808227, "epoch": 7, "memory": 25074, "step": 35648} +{"lr": 0.00031961812419703224, "data_time": 0.0016529321670532226, "loss": 0.007860430562868715, "time": 0.8411382913589478, "epoch": 7, "memory": 25074, "step": 35748} +{"lr": 0.00031961812419703224, "data_time": 0.001423478126525879, "loss": 0.006672559911385179, "time": 0.784315538406372, "epoch": 7, "memory": 25074, "step": 35848} +{"lr": 0.00031961812419703224, "data_time": 0.0014984607696533203, "loss": 0.013350946130231023, "time": 0.7980647325515747, "epoch": 7, "memory": 25074, "step": 35948} +{"lr": 0.00031961812419703224, "data_time": 0.0013253211975097657, "loss": 0.009304246306419373, "time": 0.7963850498199463, "epoch": 7, "memory": 25074, "step": 36048} +{"lr": 0.00031961812419703224, "data_time": 0.0010668039321899414, "loss": 0.006125065824016928, "time": 0.7851140022277832, "epoch": 7, "memory": 25074, "step": 36148} +{"lr": 0.00031961812419703224, "data_time": 0.0012668371200561523, "loss": 0.008404319267719984, "time": 0.7794710397720337, "epoch": 7, "memory": 25074, "step": 36248} +{"lr": 0.00031961812419703224, "data_time": 0.0014218568801879882, "loss": 0.007274479186162352, "time": 0.8161937475204468, "epoch": 7, "memory": 25074, "step": 36348} +{"lr": 0.00031961812419703224, "data_time": 0.0014168262481689454, "loss": 0.013077496737241744, "time": 0.7840482950210571, "epoch": 7, "memory": 25074, "step": 36448} +{"lr": 0.00031961812419703224, "data_time": 0.0012868642807006836, "loss": 0.007608062401413917, "time": 0.7778895854949951, "epoch": 7, "memory": 25074, "step": 36548} +{"lr": 0.00031961812419703224, "data_time": 0.001390552520751953, "loss": 0.005919932993128896, "time": 0.7799853801727294, "epoch": 7, "memory": 25074, "step": 36648} +{"lr": 0.00031961812419703224, "data_time": 0.001351642608642578, "loss": 0.010841946955770254, "time": 0.8129358768463135, "epoch": 7, "memory": 25074, "step": 36748} +{"lr": 0.00031961812419703224, "data_time": 0.0011922121047973633, "loss": 0.0059931542258709666, "time": 0.7827545166015625, "epoch": 7, "memory": 25074, "step": 36848} +{"lr": 0.00031961812419703224, "data_time": 0.0011208772659301758, "loss": 0.006005952227860689, "time": 0.7702524900436402, "epoch": 7, "memory": 25074, "step": 36948} +{"lr": 0.00031961812419703224, "data_time": 0.0014847517013549805, "loss": 0.009921358013525606, "time": 0.8259676218032836, "epoch": 7, "memory": 25074, "step": 37048} +{"lr": 0.00031961812419703224, "data_time": 0.0012612104415893554, "loss": 0.009072218835353852, "time": 0.9323811769485474, "epoch": 7, "memory": 25074, "step": 37148} +{"lr": 0.00031961812419703224, "data_time": 0.001452803611755371, "loss": 0.008785020234063268, "time": 0.818962574005127, "epoch": 7, "memory": 25074, "step": 37248} +{"lr": 0.00031961812419703224, "data_time": 0.001206374168395996, "loss": 0.005893416702747345, "time": 0.7777831554412842, "epoch": 7, "memory": 25074, "step": 37348} +{"lr": 0.00031961812419703224, "data_time": 0.0013034582138061524, "loss": 0.007719710515812039, "time": 0.8131859540939331, "epoch": 7, "memory": 25074, "step": 37448} +{"lr": 0.00031961812419703224, "data_time": 0.0012425422668457032, "loss": 0.008503195736557246, "time": 0.816457986831665, "epoch": 7, "memory": 25074, "step": 37548} +{"lr": 0.00031961812419703224, "data_time": 0.0013104915618896485, "loss": 0.007905058888718485, "time": 0.8407413005828858, "epoch": 7, "memory": 25074, "step": 37648} +{"lr": 0.00031961812419703224, "data_time": 0.0015394449234008788, "loss": 0.011772033665329218, "time": 0.7709179639816284, "epoch": 7, "memory": 25074, "step": 37748} +{"lr": 0.00031961812419703224, "data_time": 0.0015858650207519532, "loss": 0.005198361305519939, "time": 0.8399975538253784, "epoch": 7, "memory": 25074, "step": 37848} +{"lr": 0.00031961812419703224, "data_time": 0.0011867761611938476, "loss": 0.007491279020905494, "time": 0.7971672058105469, "epoch": 7, "memory": 25074, "step": 37948} +{"lr": 0.00031961812419703224, "data_time": 0.0013853788375854492, "loss": 0.013604060234501957, "time": 0.8534063100814819, "epoch": 7, "memory": 25074, "step": 38048} +{"lr": 0.00031961812419703224, "data_time": 0.0014464855194091797, "loss": 0.006931758066639304, "time": 0.8338172674179077, "epoch": 7, "memory": 25074, "step": 38148} +{"lr": 0.00031961812419703224, "data_time": 0.0011494159698486328, "loss": 0.007359489006921649, "time": 0.7883419036865235, "epoch": 7, "memory": 25074, "step": 38248} +{"lr": 0.00031961812419703224, "data_time": 0.0012499570846557617, "loss": 0.0036069298163056373, "time": 0.8009266376495361, "epoch": 7, "memory": 25074, "step": 38348} +{"lr": 0.00031961812419703224, "data_time": 0.0014397859573364257, "loss": 0.004166057799011469, "time": 0.7946410655975342, "epoch": 7, "memory": 25074, "step": 38448} +{"lr": 0.00031961812419703224, "data_time": 0.0011008024215698243, "loss": 0.0033724519424140455, "time": 0.8424299955368042, "epoch": 7, "memory": 25074, "step": 38548} +{"lr": 0.00031961812419703224, "data_time": 0.0012166738510131837, "loss": 0.005197452334687114, "time": 0.8725511312484742, "epoch": 7, "memory": 25074, "step": 38648} +{"lr": 0.00031961812419703224, "data_time": 0.0013074398040771485, "loss": 0.009734292468056082, "time": 0.8246922492980957, "epoch": 7, "memory": 25074, "step": 38748} +{"lr": 0.00031961812419703224, "data_time": 0.0012614965438842774, "loss": 0.006522056041285395, "time": 0.8296808719635009, "epoch": 7, "memory": 25074, "step": 38848} +{"lr": 0.00031961812419703224, "data_time": 0.0013681650161743164, "loss": 0.003907531034201384, "time": 0.8276172637939453, "epoch": 7, "memory": 25074, "step": 38948} +{"lr": 0.00031961812419703224, "data_time": 0.0017262697219848633, "loss": 0.0030798387713730337, "time": 0.8221835374832154, "epoch": 7, "memory": 25074, "step": 39048} +{"lr": 0.00031961812419703224, "data_time": 0.0016393423080444335, "loss": 0.009174630232155322, "time": 0.8391355514526367, "epoch": 7, "memory": 25074, "step": 39148} +{"lr": 0.00031961812419703224, "data_time": 0.0012131690979003905, "loss": 0.008284668670967221, "time": 0.8018346309661866, "epoch": 7, "memory": 25074, "step": 39248} +{"lr": 0.00031961812419703224, "data_time": 0.001375102996826172, "loss": 0.0025517712347209454, "time": 0.839412784576416, "epoch": 7, "memory": 25074, "step": 39348} +{"lr": 0.00031961812419703224, "data_time": 0.0013039588928222656, "loss": 0.0031253993045538666, "time": 0.7875226736068726, "epoch": 7, "memory": 25074, "step": 39448} +{"lr": 0.00031961812419703224, "data_time": 0.0014575958251953126, "loss": 0.007066106610000134, "time": 0.8162870168685913, "epoch": 7, "memory": 25074, "step": 39548} +{"lr": 0.00031961812419703224, "data_time": 0.0014191627502441405, "loss": 0.0052933010738343, "time": 0.7524826765060425, "epoch": 7, "memory": 25074, "step": 39648} +{"lr": 0.00031961812419703224, "data_time": 0.001366734504699707, "loss": 0.008860271843150257, "time": 0.8108349323272706, "epoch": 7, "memory": 25074, "step": 39748} +{"lr": 0.00031961812419703224, "data_time": 0.0013921737670898437, "loss": 0.008120916970074176, "time": 0.7975910186767579, "epoch": 7, "memory": 25074, "step": 39848} +{"lr": 0.00031961812419703224, "data_time": 0.0012505292892456056, "loss": 0.004192754300311208, "time": 0.7771164655685425, "epoch": 7, "memory": 25074, "step": 39948} +{"lr": 0.00031961812419703224, "data_time": 0.0012868642807006836, "loss": 0.006793348211795091, "time": 0.8004083156585693, "epoch": 7, "memory": 25074, "step": 40048} +{"lr": 0.00031961812419703224, "data_time": 0.0016983270645141602, "loss": 0.001058782683685422, "time": 0.7887581348419189, "epoch": 7, "memory": 25074, "step": 40148} +{"lr": 0.00031961812419703224, "data_time": 0.0013393878936767579, "loss": 0.005151572031900286, "time": 0.8344828367233277, "epoch": 7, "memory": 25074, "step": 40248} +{"accuracy/top1": 99.8926773071289, "data_time": 0.2596237098469454, "time": 0.645106596105239, "step": 7} +{"lr": 0.0002935281474492116, "data_time": 0.08174433708190917, "loss": 0.004533440480008722, "time": 0.8853747844696045, "epoch": 8, "memory": 25074, "step": 40406} +{"lr": 0.0002935281474492116, "data_time": 0.0012613534927368164, "loss": 0.008634087070822716, "time": 0.7807964324951172, "epoch": 8, "memory": 25074, "step": 40506} +{"lr": 0.0002935281474492116, "data_time": 0.0012382030487060546, "loss": 0.007455760054290295, "time": 0.8089942455291748, "epoch": 8, "memory": 25074, "step": 40606} +{"lr": 0.0002935281474492116, "data_time": 0.001336359977722168, "loss": 0.006103390548378229, "time": 0.7352396726608277, "epoch": 8, "memory": 25074, "step": 40706} +{"lr": 0.0002935281474492116, "data_time": 0.0014557600021362304, "loss": 0.005343703227117658, "time": 0.754692792892456, "epoch": 8, "memory": 25074, "step": 40806} +{"lr": 0.0002935281474492116, "data_time": 0.0016048431396484375, "loss": 0.006173395831137895, "time": 0.7743478059768677, "epoch": 8, "memory": 25074, "step": 40906} +{"lr": 0.0002935281474492116, "data_time": 0.0016127586364746093, "loss": 0.004697187058627606, "time": 0.7527307987213134, "epoch": 8, "memory": 25074, "step": 41006} +{"lr": 0.0002935281474492116, "data_time": 0.0014994144439697266, "loss": 0.007261675829067826, "time": 0.7300341606140137, "epoch": 8, "memory": 25074, "step": 41106} +{"lr": 0.0002935281474492116, "data_time": 0.0015994071960449218, "loss": 0.005552244512364269, "time": 0.7418447256088256, "epoch": 8, "memory": 25074, "step": 41206} +{"lr": 0.0002935281474492116, "data_time": 0.0012565135955810546, "loss": 0.006357346661388874, "time": 0.7793103218078613, "epoch": 8, "memory": 25074, "step": 41306} +{"lr": 0.0002935281474492116, "data_time": 0.0014730453491210937, "loss": 0.004397300910204649, "time": 0.8202110767364502, "epoch": 8, "memory": 25074, "step": 41406} +{"lr": 0.0002935281474492116, "data_time": 0.0014173030853271485, "loss": 0.003983988706022501, "time": 0.8394148111343384, "epoch": 8, "memory": 25074, "step": 41506} +{"lr": 0.0002935281474492116, "data_time": 0.0016449213027954102, "loss": 0.003592493990436196, "time": 0.752266526222229, "epoch": 8, "memory": 25074, "step": 41606} +{"lr": 0.0002935281474492116, "data_time": 0.001744985580444336, "loss": 0.008995352173224091, "time": 0.7827673435211182, "epoch": 8, "memory": 25074, "step": 41706} +{"lr": 0.0002935281474492116, "data_time": 0.0015886068344116212, "loss": 0.0035457221325486898, "time": 0.777490234375, "epoch": 8, "memory": 25074, "step": 41806} +{"lr": 0.0002935281474492116, "data_time": 0.0012670516967773437, "loss": 0.010279138339683414, "time": 0.8055679082870484, "epoch": 8, "memory": 25074, "step": 41906} +{"lr": 0.0002935281474492116, "data_time": 0.0014852285385131836, "loss": 0.004720044881105423, "time": 0.7767008304595947, "epoch": 8, "memory": 25074, "step": 42006} +{"lr": 0.0002935281474492116, "data_time": 0.0012957334518432617, "loss": 0.004443504381924868, "time": 0.7426398515701294, "epoch": 8, "memory": 25074, "step": 42106} +{"lr": 0.0002935281474492116, "data_time": 0.0012934446334838868, "loss": 0.006596829975023866, "time": 0.7938202857971192, "epoch": 8, "memory": 25074, "step": 42206} +{"lr": 0.0002935281474492116, "data_time": 0.001536989212036133, "loss": 0.007685122918337584, "time": 0.7267671585083008, "epoch": 8, "memory": 25074, "step": 42306} +{"lr": 0.0002935281474492116, "data_time": 0.0014831304550170898, "loss": 0.0075868109706789255, "time": 0.8245455503463746, "epoch": 8, "memory": 25074, "step": 42406} +{"lr": 0.0002935281474492116, "data_time": 0.0012464523315429688, "loss": 0.007385736098513007, "time": 0.8491693735122681, "epoch": 8, "memory": 25074, "step": 42506} +{"lr": 0.0002935281474492116, "data_time": 0.0015519380569458008, "loss": 0.003152378322556615, "time": 0.7582871913909912, "epoch": 8, "memory": 25074, "step": 42606} +{"lr": 0.0002935281474492116, "data_time": 0.0015825271606445313, "loss": 0.0063883363734930755, "time": 0.7175159454345703, "epoch": 8, "memory": 25074, "step": 42706} +{"lr": 0.0002935281474492116, "data_time": 0.0015311479568481446, "loss": 0.007287848042324185, "time": 0.7520819902420044, "epoch": 8, "memory": 25074, "step": 42806} +{"lr": 0.0002935281474492116, "data_time": 0.0012787580490112305, "loss": 0.007251392956823111, "time": 0.7041292190551758, "epoch": 8, "memory": 25074, "step": 42906} +{"lr": 0.0002935281474492116, "data_time": 0.0015406370162963866, "loss": 0.004316490795463324, "time": 0.7086874961853027, "epoch": 8, "memory": 25074, "step": 43006} +{"lr": 0.0002935281474492116, "data_time": 0.07285220623016357, "loss": 0.005532010411843658, "time": 0.8373117446899414, "epoch": 8, "memory": 25074, "step": 43106} +{"lr": 0.0002935281474492116, "data_time": 0.11478307247161865, "loss": 0.012309360085055232, "time": 0.8191475391387939, "epoch": 8, "memory": 25074, "step": 43206} +{"lr": 0.0002935281474492116, "data_time": 0.009643650054931641, "loss": 0.008047929825261235, "time": 0.7130385160446167, "epoch": 8, "memory": 25074, "step": 43306} +{"lr": 0.0002935281474492116, "data_time": 0.141137957572937, "loss": 0.0039288499392569065, "time": 0.8460032224655152, "epoch": 8, "memory": 25074, "step": 43406} +{"lr": 0.0002935281474492116, "data_time": 0.0016232013702392578, "loss": 0.00623535355553031, "time": 0.849001145362854, "epoch": 8, "memory": 25074, "step": 43506} +{"lr": 0.0002935281474492116, "data_time": 0.0012703657150268555, "loss": 0.00541509916074574, "time": 0.8788488388061524, "epoch": 8, "memory": 25074, "step": 43606} +{"lr": 0.0002935281474492116, "data_time": 0.0012674331665039062, "loss": 0.005426371656358242, "time": 0.7773834705352783, "epoch": 8, "memory": 25074, "step": 43706} +{"lr": 0.0002935281474492116, "data_time": 0.0017203807830810547, "loss": 0.004386982461437583, "time": 0.7341472864151001, "epoch": 8, "memory": 25074, "step": 43806} +{"lr": 0.0002935281474492116, "data_time": 0.0014246225357055664, "loss": 0.0026717221364378927, "time": 0.7962167263031006, "epoch": 8, "memory": 25074, "step": 43906} +{"lr": 0.0002935281474492116, "data_time": 0.0019036293029785155, "loss": 0.005585245974361897, "time": 0.7138374090194702, "epoch": 8, "memory": 25074, "step": 44006} +{"lr": 0.0002935281474492116, "data_time": 0.001444530487060547, "loss": 0.008353070076555013, "time": 0.7424083948135376, "epoch": 8, "memory": 25074, "step": 44106} +{"lr": 0.0002935281474492116, "data_time": 0.0014462471008300781, "loss": 0.00635327473282814, "time": 0.8163536548614502, "epoch": 8, "memory": 25074, "step": 44206} +{"lr": 0.0002935281474492116, "data_time": 0.0016684532165527344, "loss": 0.004578802362084388, "time": 0.7836791753768921, "epoch": 8, "memory": 25074, "step": 44306} +{"lr": 0.0002935281474492116, "data_time": 0.0016208410263061524, "loss": 0.005181445740163326, "time": 0.8223507881164551, "epoch": 8, "memory": 25074, "step": 44406} +{"lr": 0.0002935281474492116, "data_time": 0.0013053178787231444, "loss": 0.006681210361421108, "time": 0.8298809766769409, "epoch": 8, "memory": 25074, "step": 44506} +{"lr": 0.0002935281474492116, "data_time": 0.0013679265975952148, "loss": 0.006784602254629135, "time": 0.7622411489486695, "epoch": 8, "memory": 25074, "step": 44606} +{"lr": 0.0002935281474492116, "data_time": 0.0014540195465087891, "loss": 0.01024253643117845, "time": 0.7808483839035034, "epoch": 8, "memory": 25074, "step": 44706} +{"lr": 0.0002935281474492116, "data_time": 0.0015398025512695312, "loss": 0.002162778237834573, "time": 0.7791775941848755, "epoch": 8, "memory": 25074, "step": 44806} +{"lr": 0.0002935281474492116, "data_time": 0.0012364864349365234, "loss": 0.0053187865298241375, "time": 0.8003702163696289, "epoch": 8, "memory": 25074, "step": 44906} +{"lr": 0.0002935281474492116, "data_time": 0.0015358448028564454, "loss": 0.003356165671721101, "time": 0.77942054271698, "epoch": 8, "memory": 25074, "step": 45006} +{"lr": 0.0002935281474492116, "data_time": 0.0014313936233520507, "loss": 0.007574144750833511, "time": 0.8116400718688965, "epoch": 8, "memory": 25074, "step": 45106} +{"lr": 0.0002935281474492116, "data_time": 0.0016705751419067382, "loss": 0.007972662895917892, "time": 0.7434316158294678, "epoch": 8, "memory": 25074, "step": 45206} +{"lr": 0.0002935281474492116, "data_time": 0.0012651920318603516, "loss": 0.004711715783923864, "time": 0.7442546606063842, "epoch": 8, "memory": 25074, "step": 45306} +{"lr": 0.0002935281474492116, "data_time": 0.0011793613433837891, "loss": 0.004612048994749785, "time": 0.789636492729187, "epoch": 8, "memory": 25074, "step": 45406} +{"lr": 0.0002935281474492116, "data_time": 0.0014587879180908204, "loss": 0.006722743064165116, "time": 0.8347979307174682, "epoch": 8, "memory": 25074, "step": 45506} +{"lr": 0.0002935281474492116, "data_time": 0.0014156579971313476, "loss": 0.004672659328207374, "time": 0.78687002658844, "epoch": 8, "memory": 25074, "step": 45606} +{"lr": 0.0002935281474492116, "data_time": 0.001810598373413086, "loss": 0.004899133229628205, "time": 0.7504078865051269, "epoch": 8, "memory": 25074, "step": 45706} +{"lr": 0.0002935281474492116, "data_time": 0.0016098260879516602, "loss": 0.008960438240319491, "time": 0.7846177101135254, "epoch": 8, "memory": 25074, "step": 45806} +{"lr": 0.0002935281474492116, "data_time": 0.0016513347625732421, "loss": 0.003535037161782384, "time": 0.825016188621521, "epoch": 8, "memory": 25074, "step": 45906} +{"lr": 0.0002935281474492116, "data_time": 0.0014484167098999024, "loss": 0.0037020920775830747, "time": 0.7800180196762085, "epoch": 8, "memory": 25074, "step": 46006} +{"accuracy/top1": 99.96842956542969, "data_time": 0.25169263166539807, "time": 0.6378555718590232, "step": 8} +{"lr": 0.00026525831390311475, "data_time": 0.0016158819198608398, "loss": 0.010060214251279832, "time": 0.8175989151000976, "epoch": 9, "memory": 25074, "step": 46164} +{"lr": 0.00026525831390311475, "data_time": 0.001458597183227539, "loss": 0.003008700720965862, "time": 0.784386134147644, "epoch": 9, "memory": 25074, "step": 46264} +{"lr": 0.00026525831390311475, "data_time": 0.0013403892517089844, "loss": 0.009089087136089802, "time": 0.749416732788086, "epoch": 9, "memory": 25074, "step": 46364} +{"lr": 0.00026525831390311475, "data_time": 0.0013660907745361329, "loss": 0.004336386872455478, "time": 0.8461895704269409, "epoch": 9, "memory": 25074, "step": 46464} +{"lr": 0.00026525831390311475, "data_time": 0.0013083219528198242, "loss": 0.0032400793861597776, "time": 0.7555406332015991, "epoch": 9, "memory": 25074, "step": 46564} +{"lr": 0.00026525831390311475, "data_time": 0.00168917179107666, "loss": 0.002776416391134262, "time": 0.8164075613021851, "epoch": 9, "memory": 25074, "step": 46664} +{"lr": 0.00026525831390311475, "data_time": 0.0012827157974243165, "loss": 0.004847720824182033, "time": 0.7282079458236694, "epoch": 9, "memory": 25074, "step": 46764} +{"lr": 0.00026525831390311475, "data_time": 0.001482677459716797, "loss": 0.0037592971231788395, "time": 0.7747260808944703, "epoch": 9, "memory": 25074, "step": 46864} +{"lr": 0.00026525831390311475, "data_time": 0.0016307830810546875, "loss": 0.003753243526443839, "time": 0.8503926515579223, "epoch": 9, "memory": 25074, "step": 46964} +{"lr": 0.00026525831390311475, "data_time": 0.0015087604522705078, "loss": 0.006353098386898637, "time": 0.7096498966217041, "epoch": 9, "memory": 25074, "step": 47064} +{"lr": 0.00026525831390311475, "data_time": 0.0014767169952392579, "loss": 0.0016077717766165734, "time": 0.7514087438583374, "epoch": 9, "memory": 25074, "step": 47164} +{"lr": 0.00026525831390311475, "data_time": 0.001598668098449707, "loss": 0.003843969479203224, "time": 0.7461911201477051, "epoch": 9, "memory": 25074, "step": 47264} +{"lr": 0.00026525831390311475, "data_time": 0.032735586166381836, "loss": 0.0028642588295042515, "time": 0.7378857135772705, "epoch": 9, "memory": 25074, "step": 47364} +{"lr": 0.00026525831390311475, "data_time": 0.10020203590393066, "loss": 0.0032532909885048867, "time": 0.8054174184799194, "epoch": 9, "memory": 25074, "step": 47464} +{"lr": 0.00026525831390311475, "data_time": 0.0016108512878417968, "loss": 0.0048339446540921925, "time": 0.7717607498168946, "epoch": 9, "memory": 25074, "step": 47564} +{"lr": 0.00026525831390311475, "data_time": 0.0014272451400756836, "loss": 0.003592148795723915, "time": 0.7555326223373413, "epoch": 9, "memory": 25074, "step": 47664} +{"lr": 0.00026525831390311475, "data_time": 0.0014083385467529297, "loss": 0.002575997915118933, "time": 0.8240430593490601, "epoch": 9, "memory": 25074, "step": 47764} +{"lr": 0.00026525831390311475, "data_time": 0.001306009292602539, "loss": 0.005089926393702626, "time": 1.0330263137817384, "epoch": 9, "memory": 25074, "step": 47864} +{"lr": 0.00026525831390311475, "data_time": 0.001449275016784668, "loss": 0.0021736639086157084, "time": 0.798820972442627, "epoch": 9, "memory": 25074, "step": 47964} +{"lr": 0.00026525831390311475, "data_time": 0.001599431037902832, "loss": 0.004608659353107214, "time": 0.7402681112289429, "epoch": 9, "memory": 25074, "step": 48064} +{"lr": 0.00026525831390311475, "data_time": 0.0013656139373779297, "loss": 0.005686013540253043, "time": 0.8199795007705688, "epoch": 9, "memory": 25074, "step": 48164} +{"lr": 0.00026525831390311475, "data_time": 0.0015366792678833008, "loss": 0.006041695084422827, "time": 0.7848222494125366, "epoch": 9, "memory": 25074, "step": 48264} +{"lr": 0.00026525831390311475, "data_time": 0.0015355348587036133, "loss": 0.005315634049475193, "time": 0.810681676864624, "epoch": 9, "memory": 25074, "step": 48364} +{"lr": 0.00026525831390311475, "data_time": 0.0013628721237182618, "loss": 0.005327724944800138, "time": 0.7424980163574219, "epoch": 9, "memory": 25074, "step": 48464} +{"lr": 0.00026525831390311475, "data_time": 0.0020877361297607423, "loss": 0.0017982026096433402, "time": 0.7792024612426758, "epoch": 9, "memory": 25074, "step": 48564} +{"lr": 0.00026525831390311475, "data_time": 0.0013993740081787109, "loss": 0.0045343706849962475, "time": 0.7861818075180054, "epoch": 9, "memory": 25074, "step": 48664} +{"lr": 0.00026525831390311475, "data_time": 0.001527690887451172, "loss": 0.003606828721240163, "time": 0.7916712522506714, "epoch": 9, "memory": 25074, "step": 48764} +{"lr": 0.00026525831390311475, "data_time": 0.0020267486572265623, "loss": 0.010490537900477648, "time": 0.7570760726928711, "epoch": 9, "memory": 25074, "step": 48864} +{"lr": 0.00026525831390311475, "data_time": 0.0016240119934082032, "loss": 0.005129858665168285, "time": 0.7905623197555542, "epoch": 9, "memory": 25074, "step": 48964} +{"lr": 0.00026525831390311475, "data_time": 0.0014204263687133789, "loss": 0.005493970308452845, "time": 0.8172428131103515, "epoch": 9, "memory": 25074, "step": 49064} +{"lr": 0.00026525831390311475, "data_time": 0.0740964651107788, "loss": 0.004783406946808099, "time": 0.7778555154800415, "epoch": 9, "memory": 25074, "step": 49164} +{"lr": 0.00026525831390311475, "data_time": 0.11791810989379883, "loss": 0.007851413358002901, "time": 0.8225356101989746, "epoch": 9, "memory": 25074, "step": 49264} +{"lr": 0.00026525831390311475, "data_time": 0.08130881786346436, "loss": 0.0026842295192182064, "time": 0.7910864591598511, "epoch": 9, "memory": 25074, "step": 49364} +{"lr": 0.00026525831390311475, "data_time": 0.13063509464263917, "loss": 0.002057286910712719, "time": 0.8399464607238769, "epoch": 9, "memory": 25074, "step": 49464} +{"lr": 0.00026525831390311475, "data_time": 0.08142657279968261, "loss": 0.006802480015903711, "time": 0.7852344751358032, "epoch": 9, "memory": 25074, "step": 49564} +{"lr": 0.00026525831390311475, "data_time": 0.11105425357818603, "loss": 0.0015038976445794106, "time": 0.8148003578186035, "epoch": 9, "memory": 25074, "step": 49664} +{"lr": 0.00026525831390311475, "data_time": 0.13849332332611083, "loss": 0.003886723518371582, "time": 0.8466254949569703, "epoch": 9, "memory": 25074, "step": 49764} +{"lr": 0.00026525831390311475, "data_time": 0.07313673496246338, "loss": 0.005093358037993312, "time": 0.7784403800964356, "epoch": 9, "memory": 25074, "step": 49864} +{"lr": 0.00026525831390311475, "data_time": 0.07475545406341552, "loss": 0.003619781881570816, "time": 0.7899048566818238, "epoch": 9, "memory": 25074, "step": 49964} +{"lr": 0.00026525831390311475, "data_time": 0.08455250263214112, "loss": 0.0045316462405025956, "time": 0.7896649360656738, "epoch": 9, "memory": 25074, "step": 50064} +{"lr": 0.00026525831390311475, "data_time": 0.07857251167297363, "loss": 0.005448000272735953, "time": 0.7821491241455079, "epoch": 9, "memory": 25074, "step": 50164} +{"lr": 0.00026525831390311475, "data_time": 0.048411202430725095, "loss": 0.0034753140062093735, "time": 0.751510214805603, "epoch": 9, "memory": 25074, "step": 50264} +{"lr": 0.00026525831390311475, "data_time": 0.0017418384552001954, "loss": 0.0038141491357237102, "time": 0.7980393171310425, "epoch": 9, "memory": 25074, "step": 50364} +{"lr": 0.00026525831390311475, "data_time": 0.0020066976547241213, "loss": 0.0020103800110518933, "time": 0.8106767177581787, "epoch": 9, "memory": 25074, "step": 50464} +{"lr": 0.00026525831390311475, "data_time": 0.0013964176177978516, "loss": 0.0038177080918103457, "time": 0.7461270332336426, "epoch": 9, "memory": 25074, "step": 50564} +{"lr": 0.00026525831390311475, "data_time": 0.0014374732971191406, "loss": 0.001423835102468729, "time": 0.7556110858917237, "epoch": 9, "memory": 25074, "step": 50664} +{"lr": 0.00026525831390311475, "data_time": 0.0015955209732055665, "loss": 0.00488781756721437, "time": 0.7805282831192016, "epoch": 9, "memory": 25074, "step": 50764} +{"lr": 0.00026525831390311475, "data_time": 0.0013822555541992188, "loss": 0.004550968669354916, "time": 0.8374075174331665, "epoch": 9, "memory": 25074, "step": 50864} +{"lr": 0.00026525831390311475, "data_time": 0.004905128479003906, "loss": 0.0032081414479762315, "time": 0.7419009685516358, "epoch": 9, "memory": 25074, "step": 50964} +{"lr": 0.00026525831390311475, "data_time": 0.0015484809875488282, "loss": 0.0032443346455693243, "time": 0.7633739471435547, "epoch": 9, "memory": 25074, "step": 51064} +{"lr": 0.00026525831390311475, "data_time": 0.001518845558166504, "loss": 0.004434935515746474, "time": 0.7812606573104859, "epoch": 9, "memory": 25074, "step": 51164} +{"lr": 0.00026525831390311475, "data_time": 0.00128173828125, "loss": 0.003473523911088705, "time": 0.7531224489212036, "epoch": 9, "memory": 25074, "step": 51264} +{"lr": 0.00026525831390311475, "data_time": 0.001495671272277832, "loss": 0.0029246626887470483, "time": 0.7889619588851928, "epoch": 9, "memory": 25074, "step": 51364} +{"lr": 0.00026525831390311475, "data_time": 0.0013351917266845703, "loss": 0.004868371831253171, "time": 0.8216519594192505, "epoch": 9, "memory": 25074, "step": 51464} +{"lr": 0.00026525831390311475, "data_time": 0.0014025449752807617, "loss": 0.0019766067154705523, "time": 0.7716431379318237, "epoch": 9, "memory": 25074, "step": 51564} +{"lr": 0.00026525831390311475, "data_time": 0.0015723228454589844, "loss": 0.0039232814684510234, "time": 0.7720461845397949, "epoch": 9, "memory": 25074, "step": 51664} +{"lr": 0.00026525831390311475, "data_time": 0.0016120195388793946, "loss": 0.003156856168061495, "time": 0.7533194303512574, "epoch": 9, "memory": 25074, "step": 51764} +{"accuracy/top1": 99.97474670410156, "data_time": 0.2695574620190789, "time": 0.6553976676043343, "step": 9} +{"lr": 0.000235504720682845, "data_time": 0.057193541526794435, "loss": 0.004519202141091227, "time": 0.7609281539916992, "epoch": 10, "memory": 25074, "step": 51922} +{"lr": 0.000235504720682845, "data_time": 0.08888649940490723, "loss": 0.0011984863318502902, "time": 0.804810619354248, "epoch": 10, "memory": 25074, "step": 52022} +{"lr": 0.000235504720682845, "data_time": 0.013669371604919434, "loss": 0.007437628647312522, "time": 0.7698734521865844, "epoch": 10, "memory": 25074, "step": 52122} +{"lr": 0.000235504720682845, "data_time": 0.10023550987243653, "loss": 0.0019167819991707802, "time": 0.8420120716094971, "epoch": 10, "memory": 25074, "step": 52222} +{"lr": 0.000235504720682845, "data_time": 0.0015046834945678712, "loss": 0.0026889451313763858, "time": 0.7677485942840576, "epoch": 10, "memory": 25074, "step": 52322} +{"lr": 0.000235504720682845, "data_time": 0.007773685455322266, "loss": 0.0049743682611733675, "time": 0.7666850328445435, "epoch": 10, "memory": 25074, "step": 52422} +{"lr": 0.000235504720682845, "data_time": 0.0013694286346435547, "loss": 0.003320180857554078, "time": 0.8272189855575561, "epoch": 10, "memory": 25074, "step": 52522} +{"lr": 0.000235504720682845, "data_time": 0.0013002634048461914, "loss": 0.0023750689812004565, "time": 0.7317907094955445, "epoch": 10, "memory": 25074, "step": 52622} +{"lr": 0.000235504720682845, "data_time": 0.001642465591430664, "loss": 0.0017983930185437203, "time": 0.761565375328064, "epoch": 10, "memory": 25074, "step": 52722} +{"lr": 0.000235504720682845, "data_time": 0.07177071571350098, "loss": 0.006485605705529451, "time": 0.7790808916091919, "epoch": 10, "memory": 25074, "step": 52822} +{"lr": 0.000235504720682845, "data_time": 0.0016214847564697266, "loss": 0.003673615679144859, "time": 0.7736930608749389, "epoch": 10, "memory": 25074, "step": 52922} +{"lr": 0.000235504720682845, "data_time": 0.001363515853881836, "loss": 0.00527961584739387, "time": 0.8046427965164185, "epoch": 10, "memory": 25074, "step": 53022} +{"lr": 0.000235504720682845, "data_time": 0.0014634132385253906, "loss": 0.006082437140867114, "time": 0.802378797531128, "epoch": 10, "memory": 25074, "step": 53122} +{"lr": 0.000235504720682845, "data_time": 0.0012588262557983398, "loss": 0.005271776346489787, "time": 0.7711512565612793, "epoch": 10, "memory": 25074, "step": 53222} +{"lr": 0.000235504720682845, "data_time": 0.0014651060104370118, "loss": 0.002576451888307929, "time": 0.8612143516540527, "epoch": 10, "memory": 25074, "step": 53322} +{"lr": 0.000235504720682845, "data_time": 0.0011648416519165039, "loss": 0.0026614618953317406, "time": 0.8963905334472656, "epoch": 10, "memory": 25074, "step": 53422} +{"lr": 0.000235504720682845, "data_time": 0.0015602827072143555, "loss": 0.001886838860809803, "time": 0.8743463039398194, "epoch": 10, "memory": 25074, "step": 53522} +{"lr": 0.000235504720682845, "data_time": 0.006681203842163086, "loss": 0.002380321780219674, "time": 1.4329699516296386, "epoch": 10, "memory": 25074, "step": 53622} +{"lr": 0.000235504720682845, "data_time": 0.0020303964614868165, "loss": 0.004485356388613582, "time": 0.8864237070083618, "epoch": 10, "memory": 25074, "step": 53722} +{"lr": 0.000235504720682845, "data_time": 0.0014768362045288086, "loss": 0.004296805337071418, "time": 0.8816102743148804, "epoch": 10, "memory": 25074, "step": 53822} +{"lr": 0.000235504720682845, "data_time": 0.0013443470001220704, "loss": 0.004021440260112285, "time": 0.7687443494796753, "epoch": 10, "memory": 25074, "step": 53922} +{"lr": 0.000235504720682845, "data_time": 0.0012895584106445313, "loss": 0.0012011222075670957, "time": 0.8162369012832642, "epoch": 10, "memory": 25074, "step": 54022} +{"lr": 0.000235504720682845, "data_time": 0.0015340805053710937, "loss": 0.003927409090101719, "time": 0.7689162492752075, "epoch": 10, "memory": 25074, "step": 54122} +{"lr": 0.000235504720682845, "data_time": 0.001264500617980957, "loss": 0.005482113501057029, "time": 0.7374703407287597, "epoch": 10, "memory": 25074, "step": 54222} +{"lr": 0.000235504720682845, "data_time": 0.0028434514999389647, "loss": 0.007824982004240155, "time": 0.7792750835418701, "epoch": 10, "memory": 25074, "step": 54322} +{"lr": 0.000235504720682845, "data_time": 0.0025654077529907227, "loss": 0.003261878388002515, "time": 0.8095210552215576, "epoch": 10, "memory": 25074, "step": 54422} +{"lr": 0.000235504720682845, "data_time": 0.0015420198440551757, "loss": 0.0032570686656981707, "time": 0.7800433397293091, "epoch": 10, "memory": 25074, "step": 54522} +{"lr": 0.000235504720682845, "data_time": 0.0013351678848266602, "loss": 0.004245009552687406, "time": 0.788340425491333, "epoch": 10, "memory": 25074, "step": 54622} +{"lr": 0.000235504720682845, "data_time": 0.001420736312866211, "loss": 0.0016847546212375164, "time": 0.7512306451797486, "epoch": 10, "memory": 25074, "step": 54722} +{"lr": 0.000235504720682845, "data_time": 0.0014606714248657227, "loss": 0.0033814623951911927, "time": 0.7921860933303833, "epoch": 10, "memory": 25074, "step": 54822} +{"lr": 0.000235504720682845, "data_time": 0.0017631053924560547, "loss": 0.006519189290702343, "time": 0.7578413486480713, "epoch": 10, "memory": 25074, "step": 54922} +{"lr": 0.000235504720682845, "data_time": 0.001624131202697754, "loss": 0.0022362173069268463, "time": 0.7228867769241333, "epoch": 10, "memory": 25074, "step": 55022} +{"lr": 0.000235504720682845, "data_time": 0.0024099826812744142, "loss": 0.002010176237672567, "time": 0.7517282724380493, "epoch": 10, "memory": 25074, "step": 55122} +{"lr": 0.000235504720682845, "data_time": 0.001647639274597168, "loss": 0.0022532119881361725, "time": 0.7662103652954102, "epoch": 10, "memory": 25074, "step": 55222} +{"lr": 0.000235504720682845, "data_time": 0.001513671875, "loss": 0.0018771516624838114, "time": 0.7472053289413452, "epoch": 10, "memory": 25074, "step": 55322} +{"lr": 0.000235504720682845, "data_time": 0.0020645618438720702, "loss": 0.0027181121520698072, "time": 0.7646674156188965, "epoch": 10, "memory": 25074, "step": 55422} +{"lr": 0.000235504720682845, "data_time": 0.001380157470703125, "loss": 0.003250989643856883, "time": 0.7771836996078492, "epoch": 10, "memory": 25074, "step": 55522} +{"lr": 0.000235504720682845, "data_time": 0.0016100406646728516, "loss": 0.0038707809057086707, "time": 0.7240278482437134, "epoch": 10, "memory": 25074, "step": 55622} +{"lr": 0.000235504720682845, "data_time": 0.001523876190185547, "loss": 0.005304170772433281, "time": 0.8107309341430664, "epoch": 10, "memory": 25074, "step": 55722} +{"lr": 0.000235504720682845, "data_time": 0.0015041828155517578, "loss": 0.005562584102153778, "time": 0.787463927268982, "epoch": 10, "memory": 25074, "step": 55822} +{"lr": 0.000235504720682845, "data_time": 0.001501917839050293, "loss": 0.0019271073397248983, "time": 0.761007022857666, "epoch": 10, "memory": 25074, "step": 55922} +{"lr": 0.000235504720682845, "data_time": 0.0014229297637939453, "loss": 0.005638011125847697, "time": 0.7404680252075195, "epoch": 10, "memory": 25074, "step": 56022} +{"lr": 0.000235504720682845, "data_time": 0.0015123128890991212, "loss": 0.00421171523630619, "time": 0.7766730785369873, "epoch": 10, "memory": 25074, "step": 56122} +{"lr": 0.000235504720682845, "data_time": 0.0013295888900756836, "loss": 0.0028375647030770778, "time": 0.7357343196868896, "epoch": 10, "memory": 25074, "step": 56222} +{"lr": 0.000235504720682845, "data_time": 0.0013447284698486328, "loss": 0.00337574053555727, "time": 0.7193809509277344, "epoch": 10, "memory": 25074, "step": 56322} +{"lr": 0.000235504720682845, "data_time": 0.001387786865234375, "loss": 0.0025291111785918476, "time": 0.706843900680542, "epoch": 10, "memory": 25074, "step": 56422} +{"lr": 0.000235504720682845, "data_time": 0.001521921157836914, "loss": 0.0069944093003869055, "time": 0.8126255512237549, "epoch": 10, "memory": 25074, "step": 56522} +{"lr": 0.000235504720682845, "data_time": 0.0016293525695800781, "loss": 0.005933578871190548, "time": 0.7594385385513306, "epoch": 10, "memory": 25074, "step": 56622} +{"lr": 0.000235504720682845, "data_time": 0.002179074287414551, "loss": 0.004614521749317646, "time": 0.769674563407898, "epoch": 10, "memory": 25074, "step": 56722} +{"lr": 0.000235504720682845, "data_time": 0.001364898681640625, "loss": 0.0018319363705813884, "time": 0.9140798568725585, "epoch": 10, "memory": 25074, "step": 56822} +{"lr": 0.000235504720682845, "data_time": 0.0012913227081298828, "loss": 0.004197343206033111, "time": 0.8385738372802735, "epoch": 10, "memory": 25074, "step": 56922} +{"lr": 0.000235504720682845, "data_time": 0.0015700340270996093, "loss": 0.004743618937209248, "time": 0.8169739723205567, "epoch": 10, "memory": 25074, "step": 57022} +{"lr": 0.000235504720682845, "data_time": 0.0015157222747802734, "loss": 0.0010100162122398614, "time": 0.8307600975036621, "epoch": 10, "memory": 25074, "step": 57122} +{"lr": 0.000235504720682845, "data_time": 0.0016313314437866211, "loss": 0.002245061891153455, "time": 0.7947349309921264, "epoch": 10, "memory": 25074, "step": 57222} +{"lr": 0.000235504720682845, "data_time": 0.0012349367141723632, "loss": 0.003983878530561924, "time": 0.7588669776916503, "epoch": 10, "memory": 25074, "step": 57322} +{"lr": 0.000235504720682845, "data_time": 0.0013663291931152344, "loss": 0.0010076561942696572, "time": 0.7645234823226928, "epoch": 10, "memory": 25074, "step": 57422} +{"lr": 0.000235504720682845, "data_time": 0.001473379135131836, "loss": 0.0005049189552664757, "time": 0.7397166967391968, "epoch": 10, "memory": 25074, "step": 57522} +{"accuracy/top1": 99.98737335205078, "data_time": 0.24964468619402716, "time": 0.636816235149608, "step": 10} +{"lr": 0.00020499999999999997, "data_time": 0.0017685174942016601, "loss": 0.001499880850315094, "time": 0.727084732055664, "epoch": 11, "memory": 25074, "step": 57680} +{"lr": 0.00020499999999999997, "data_time": 0.04508950710296631, "loss": 0.003448608098551631, "time": 0.8800601005554199, "epoch": 11, "memory": 25074, "step": 57780} +{"lr": 0.00020499999999999997, "data_time": 0.05101258754730224, "loss": 0.00649166819639504, "time": 0.9064756870269776, "epoch": 11, "memory": 25074, "step": 57880} +{"lr": 0.00020499999999999997, "data_time": 0.0015633106231689453, "loss": 0.000826117442920804, "time": 0.8373608112335205, "epoch": 11, "memory": 25074, "step": 57980} +{"lr": 0.00020499999999999997, "data_time": 0.0015419483184814452, "loss": 0.005671384185552597, "time": 0.9274006366729737, "epoch": 11, "memory": 25074, "step": 58080} +{"lr": 0.00020499999999999997, "data_time": 0.0017696142196655274, "loss": 0.002768180798739195, "time": 0.9979973793029785, "epoch": 11, "memory": 25074, "step": 58180} +{"lr": 0.00020499999999999997, "data_time": 0.0013934850692749023, "loss": 0.0017746952828019858, "time": 0.7385730981826782, "epoch": 11, "memory": 25074, "step": 58280} +{"lr": 0.00020499999999999997, "data_time": 0.0016549110412597656, "loss": 0.003073259023949504, "time": 0.7587138175964355, "epoch": 11, "memory": 25074, "step": 58380} +{"lr": 0.00020499999999999997, "data_time": 0.0019368410110473632, "loss": 0.002911191340535879, "time": 0.880544376373291, "epoch": 11, "memory": 25074, "step": 58480} +{"lr": 0.00020499999999999997, "data_time": 0.0015068769454956055, "loss": 0.0012459274381399154, "time": 0.7419887065887452, "epoch": 11, "memory": 25074, "step": 58580} +{"lr": 0.00020499999999999997, "data_time": 0.0014966487884521484, "loss": 0.0028540372848510744, "time": 0.7582054615020752, "epoch": 11, "memory": 25074, "step": 58680} +{"lr": 0.00020499999999999997, "data_time": 0.0013843059539794921, "loss": 0.0014779695309698581, "time": 0.7900362730026245, "epoch": 11, "memory": 25074, "step": 58780} +{"lr": 0.00020499999999999997, "data_time": 0.0014931201934814454, "loss": 0.003163971332833171, "time": 0.8209036827087403, "epoch": 11, "memory": 25074, "step": 58880} +{"lr": 0.00020499999999999997, "data_time": 0.0016663551330566406, "loss": 0.005382882244884968, "time": 0.7464561223983764, "epoch": 11, "memory": 25074, "step": 58980} +{"lr": 0.00020499999999999997, "data_time": 0.0816460371017456, "loss": 0.005045345053076744, "time": 0.7894214630126953, "epoch": 11, "memory": 25074, "step": 59080} +{"lr": 0.00020499999999999997, "data_time": 0.0022483587265014647, "loss": 0.0013307729735970498, "time": 0.7491605520248413, "epoch": 11, "memory": 25074, "step": 59180} +{"lr": 0.00020499999999999997, "data_time": 0.0019892215728759765, "loss": 0.0032491808757185935, "time": 0.7896609783172608, "epoch": 11, "memory": 25074, "step": 59280} +{"lr": 0.00020499999999999997, "data_time": 0.001386404037475586, "loss": 0.003408097615465522, "time": 0.7466486930847168, "epoch": 11, "memory": 25074, "step": 59380} +{"lr": 0.00020499999999999997, "data_time": 0.001789402961730957, "loss": 0.002397431433200836, "time": 0.8003311395645142, "epoch": 11, "memory": 25074, "step": 59480} +{"lr": 0.00020499999999999997, "data_time": 0.0017194509506225585, "loss": 0.005724278744310141, "time": 0.7389395952224731, "epoch": 11, "memory": 25074, "step": 59580} +{"lr": 0.00020499999999999997, "data_time": 0.0019245624542236328, "loss": 0.007220759894698859, "time": 0.8425361633300781, "epoch": 11, "memory": 25074, "step": 59680} +{"lr": 0.00020499999999999997, "data_time": 0.0016591072082519532, "loss": 0.002435768349096179, "time": 0.781492805480957, "epoch": 11, "memory": 25074, "step": 59780} +{"lr": 0.00020499999999999997, "data_time": 0.0012711048126220702, "loss": 0.0049605985637754205, "time": 0.746262526512146, "epoch": 11, "memory": 25074, "step": 59880} +{"lr": 0.00020499999999999997, "data_time": 0.0014362573623657227, "loss": 0.00446456135250628, "time": 0.7921473264694214, "epoch": 11, "memory": 25074, "step": 59980} +{"lr": 0.00020499999999999997, "data_time": 0.0013808250427246095, "loss": 0.0026375993620604277, "time": 0.7466942548751831, "epoch": 11, "memory": 25074, "step": 60080} +{"lr": 0.00020499999999999997, "data_time": 0.0015541315078735352, "loss": 0.0015348814893513917, "time": 0.7708494424819946, "epoch": 11, "memory": 25074, "step": 60180} +{"lr": 0.00020499999999999997, "data_time": 0.0014572620391845703, "loss": 0.0024017376359552144, "time": 0.7861558437347412, "epoch": 11, "memory": 25074, "step": 60280} +{"lr": 0.00020499999999999997, "data_time": 0.001779484748840332, "loss": 0.0019975043367594482, "time": 0.8155534029006958, "epoch": 11, "memory": 25074, "step": 60380} +{"lr": 0.00020499999999999997, "data_time": 0.0013263702392578125, "loss": 0.003345109848305583, "time": 0.8130066633224488, "epoch": 11, "memory": 25074, "step": 60480} +{"lr": 0.00020499999999999997, "data_time": 0.001529693603515625, "loss": 0.0024079356342554094, "time": 0.7653929471969605, "epoch": 11, "memory": 25074, "step": 60580} +{"lr": 0.00020499999999999997, "data_time": 0.0014828205108642577, "loss": 0.0012781056109815836, "time": 0.779039716720581, "epoch": 11, "memory": 25074, "step": 60680} +{"lr": 0.00020499999999999997, "data_time": 0.0014978647232055664, "loss": 0.004390385374426842, "time": 0.7408694744110107, "epoch": 11, "memory": 25074, "step": 60780} +{"lr": 0.00020499999999999997, "data_time": 0.0017479658126831055, "loss": 0.001297913258895278, "time": 0.7407306432723999, "epoch": 11, "memory": 25074, "step": 60880} +{"lr": 0.00020499999999999997, "data_time": 0.0013545989990234376, "loss": 0.0014874675776809453, "time": 0.7500781774520874, "epoch": 11, "memory": 25074, "step": 60980} +{"lr": 0.00020499999999999997, "data_time": 0.0013677597045898438, "loss": 0.0011331463232636453, "time": 0.7901705741882324, "epoch": 11, "memory": 25074, "step": 61080} +{"lr": 0.00020499999999999997, "data_time": 0.0014270782470703126, "loss": 0.001596615230664611, "time": 0.7723257541656494, "epoch": 11, "memory": 25074, "step": 61180} +{"lr": 0.00020499999999999997, "data_time": 0.0018324851989746094, "loss": 0.0022148349322378635, "time": 0.7486151218414306, "epoch": 11, "memory": 25074, "step": 61280} +{"lr": 0.00020499999999999997, "data_time": 0.001596689224243164, "loss": 0.004162867739796639, "time": 0.7519829511642456, "epoch": 11, "memory": 25074, "step": 61380} +{"lr": 0.00020499999999999997, "data_time": 0.001479935646057129, "loss": 0.007598197925835848, "time": 0.7830942630767822, "epoch": 11, "memory": 25074, "step": 61480} +{"lr": 0.00020499999999999997, "data_time": 0.001686239242553711, "loss": 0.0030011987779289483, "time": 0.7706879854202271, "epoch": 11, "memory": 25074, "step": 61580} +{"lr": 0.00020499999999999997, "data_time": 0.001537013053894043, "loss": 0.0005841518752276898, "time": 0.7245378255844116, "epoch": 11, "memory": 25074, "step": 61680} +{"lr": 0.00020499999999999997, "data_time": 0.0013563394546508788, "loss": 0.0015193565282970666, "time": 0.7570759534835816, "epoch": 11, "memory": 25074, "step": 61780} +{"lr": 0.00020499999999999997, "data_time": 0.0016640663146972657, "loss": 0.005667509092018008, "time": 0.7757426023483276, "epoch": 11, "memory": 25074, "step": 61880} +{"lr": 0.00020499999999999997, "data_time": 0.0014312028884887694, "loss": 0.003807988530024886, "time": 0.760835337638855, "epoch": 11, "memory": 25074, "step": 61980} +{"lr": 0.00020499999999999997, "data_time": 0.0015012502670288086, "loss": 0.0013010187074542045, "time": 0.7440502166748046, "epoch": 11, "memory": 25074, "step": 62080} +{"lr": 0.00020499999999999997, "data_time": 0.0013770580291748047, "loss": 0.0009279805701225996, "time": 0.797015643119812, "epoch": 11, "memory": 25074, "step": 62180} +{"lr": 0.00020499999999999997, "data_time": 0.0012821435928344726, "loss": 0.004593444196507335, "time": 0.7248758316040039, "epoch": 11, "memory": 25074, "step": 62280} +{"lr": 0.00020499999999999997, "data_time": 0.0013997793197631837, "loss": 0.005748032545670867, "time": 0.7976506471633911, "epoch": 11, "memory": 25074, "step": 62380} +{"lr": 0.00020499999999999997, "data_time": 0.0014078378677368163, "loss": 0.002443343959748745, "time": 0.762683629989624, "epoch": 11, "memory": 25074, "step": 62480} +{"lr": 0.00020499999999999997, "data_time": 0.0013005256652832032, "loss": 0.001835272926837206, "time": 0.8184296369552613, "epoch": 11, "memory": 25074, "step": 62580} +{"lr": 0.00020499999999999997, "data_time": 0.0015009164810180664, "loss": 0.002549597527831793, "time": 0.8354214906692505, "epoch": 11, "memory": 25074, "step": 62680} +{"lr": 0.00020499999999999997, "data_time": 0.0017477989196777344, "loss": 0.001508195186033845, "time": 0.760907506942749, "epoch": 11, "memory": 25074, "step": 62780} +{"lr": 0.00020499999999999997, "data_time": 0.0017365694046020507, "loss": 0.0010644513182342052, "time": 0.7797685861587524, "epoch": 11, "memory": 25074, "step": 62880} +{"lr": 0.00020499999999999997, "data_time": 0.0014548778533935546, "loss": 0.0006180366966873407, "time": 0.7651755094528199, "epoch": 11, "memory": 25074, "step": 62980} +{"lr": 0.00020499999999999997, "data_time": 0.0014676332473754882, "loss": 0.005244012922048569, "time": 0.7664060354232788, "epoch": 11, "memory": 25074, "step": 63080} +{"lr": 0.00020499999999999997, "data_time": 0.00142364501953125, "loss": 0.0013581722043454647, "time": 0.7844285726547241, "epoch": 11, "memory": 25074, "step": 63180} +{"lr": 0.00020499999999999997, "data_time": 0.0013538122177124024, "loss": 0.0014689499512314796, "time": 0.7780123233795166, "epoch": 11, "memory": 25074, "step": 63280} +{"accuracy/top1": 99.98737335205078, "data_time": 0.23662527869729436, "time": 0.6224999007056741, "step": 11} +{"lr": 0.000174495279317155, "data_time": 0.0013837575912475585, "loss": 0.0005901526659727096, "time": 0.7773324966430664, "epoch": 12, "memory": 25074, "step": 63438} +{"lr": 0.000174495279317155, "data_time": 0.0018453359603881835, "loss": 0.0027551088016480206, "time": 0.7789901971817017, "epoch": 12, "memory": 25074, "step": 63538} +{"lr": 0.000174495279317155, "data_time": 0.05396201610565186, "loss": 0.0004269802011549473, "time": 0.7833281755447388, "epoch": 12, "memory": 25074, "step": 63638} +{"lr": 0.000174495279317155, "data_time": 0.10586466789245605, "loss": 0.002240213146433234, "time": 0.8102561473846436, "epoch": 12, "memory": 25074, "step": 63738} +{"lr": 0.000174495279317155, "data_time": 0.05591053962707519, "loss": 0.002945806086063385, "time": 0.7739051580429077, "epoch": 12, "memory": 25074, "step": 63838} +{"lr": 0.000174495279317155, "data_time": 0.0248889684677124, "loss": 0.0026704743038862945, "time": 0.7308327198028565, "epoch": 12, "memory": 25074, "step": 63938} +{"lr": 0.000174495279317155, "data_time": 0.053395533561706544, "loss": 0.0005782907828688622, "time": 0.7611489295959473, "epoch": 12, "memory": 25074, "step": 64038} +{"lr": 0.000174495279317155, "data_time": 0.04669842720031738, "loss": 0.0008660111576318741, "time": 0.751051378250122, "epoch": 12, "memory": 25074, "step": 64138} +{"lr": 0.000174495279317155, "data_time": 0.037481832504272464, "loss": 0.0012844897340983153, "time": 0.7554550170898438, "epoch": 12, "memory": 25074, "step": 64238} +{"lr": 0.000174495279317155, "data_time": 0.07840168476104736, "loss": 0.0029215611051768065, "time": 0.7859464406967163, "epoch": 12, "memory": 25074, "step": 64338} +{"lr": 0.000174495279317155, "data_time": 0.0528986930847168, "loss": 0.0016341653652489185, "time": 0.768789267539978, "epoch": 12, "memory": 25074, "step": 64438} +{"lr": 0.000174495279317155, "data_time": 0.04934244155883789, "loss": 0.0030007160268723963, "time": 0.7535741090774536, "epoch": 12, "memory": 25074, "step": 64538} +{"lr": 0.000174495279317155, "data_time": 0.0014331817626953125, "loss": 0.0011381905525922776, "time": 0.7555559158325196, "epoch": 12, "memory": 25074, "step": 64638} +{"lr": 0.000174495279317155, "data_time": 0.001530933380126953, "loss": 0.0007664205972105264, "time": 0.7934005498886109, "epoch": 12, "memory": 25074, "step": 64738} +{"lr": 0.000174495279317155, "data_time": 0.001383829116821289, "loss": 0.0023317751474678516, "time": 0.7839012861251831, "epoch": 12, "memory": 25074, "step": 64838} +{"lr": 0.000174495279317155, "data_time": 0.0014744281768798828, "loss": 0.00024874787777662277, "time": 0.748563265800476, "epoch": 12, "memory": 25074, "step": 64938} +{"lr": 0.000174495279317155, "data_time": 0.001970839500427246, "loss": 0.0013773555867373944, "time": 0.8462858915328979, "epoch": 12, "memory": 25074, "step": 65038} +{"lr": 0.000174495279317155, "data_time": 0.0015425682067871094, "loss": 0.0005728702992200852, "time": 0.767841124534607, "epoch": 12, "memory": 25074, "step": 65138} +{"lr": 0.000174495279317155, "data_time": 0.0014516592025756836, "loss": 0.0033644781913608314, "time": 0.7692043781280518, "epoch": 12, "memory": 25074, "step": 65238} +{"lr": 0.000174495279317155, "data_time": 0.0015937566757202148, "loss": 0.0038938840385526418, "time": 0.8048548936843872, "epoch": 12, "memory": 25074, "step": 65338} +{"lr": 0.000174495279317155, "data_time": 0.0014206886291503907, "loss": 0.0023420528043061493, "time": 0.7502812623977662, "epoch": 12, "memory": 25074, "step": 65438} +{"lr": 0.000174495279317155, "data_time": 0.0013887882232666016, "loss": 0.0012345291674137116, "time": 0.7603619337081909, "epoch": 12, "memory": 25074, "step": 65538} +{"lr": 0.000174495279317155, "data_time": 0.0014254331588745117, "loss": 0.0016018415335565805, "time": 0.7409027338027954, "epoch": 12, "memory": 25074, "step": 65638} +{"lr": 0.000174495279317155, "data_time": 0.001643204689025879, "loss": 0.0014056430663913488, "time": 0.817690110206604, "epoch": 12, "memory": 25074, "step": 65738} +{"lr": 0.000174495279317155, "data_time": 0.0018610000610351563, "loss": 0.0036975953727960585, "time": 0.7444644927978515, "epoch": 12, "memory": 25074, "step": 65838} +{"lr": 0.000174495279317155, "data_time": 0.0016638755798339844, "loss": 0.0013773952610790729, "time": 0.7503877639770508, "epoch": 12, "memory": 25074, "step": 65938} +{"lr": 0.000174495279317155, "data_time": 0.0016218662261962891, "loss": 0.0012711997609585524, "time": 0.7381119728088379, "epoch": 12, "memory": 25074, "step": 66038} +{"lr": 0.000174495279317155, "data_time": 0.0013531923294067382, "loss": 0.0026980101596564055, "time": 0.7505352973937989, "epoch": 12, "memory": 25074, "step": 66138} +{"lr": 0.000174495279317155, "data_time": 0.0013225317001342774, "loss": 0.003637139638885856, "time": 0.7412928581237793, "epoch": 12, "memory": 25074, "step": 66238} +{"lr": 0.000174495279317155, "data_time": 0.0013952970504760743, "loss": 0.000229179160669446, "time": 0.8201618909835815, "epoch": 12, "memory": 25074, "step": 66338} +{"lr": 0.000174495279317155, "data_time": 0.0016584396362304688, "loss": 0.004031705809757114, "time": 0.854230260848999, "epoch": 12, "memory": 25074, "step": 66438} +{"lr": 0.000174495279317155, "data_time": 0.0015131711959838867, "loss": 0.0008632448501884937, "time": 0.7685657739639282, "epoch": 12, "memory": 25074, "step": 66538} +{"lr": 0.000174495279317155, "data_time": 0.0013129234313964844, "loss": 0.000890235137194395, "time": 0.7899316787719727, "epoch": 12, "memory": 25074, "step": 66638} +{"lr": 0.000174495279317155, "data_time": 0.0015313148498535157, "loss": 0.00043141678906977177, "time": 0.8040109872817993, "epoch": 12, "memory": 25074, "step": 66738} +{"lr": 0.000174495279317155, "data_time": 0.0014068365097045898, "loss": 0.0051310168579220775, "time": 0.767658805847168, "epoch": 12, "memory": 25074, "step": 66838} +{"lr": 0.000174495279317155, "data_time": 0.0013263225555419922, "loss": 0.004541805619373918, "time": 0.7757488489151001, "epoch": 12, "memory": 25074, "step": 66938} +{"lr": 0.000174495279317155, "data_time": 0.0014766693115234376, "loss": 0.0023615763522684572, "time": 0.781972599029541, "epoch": 12, "memory": 25074, "step": 67038} +{"lr": 0.000174495279317155, "data_time": 0.0014945268630981445, "loss": 0.004515993222594261, "time": 0.8002325534820557, "epoch": 12, "memory": 25074, "step": 67138} +{"lr": 0.000174495279317155, "data_time": 0.0015024900436401366, "loss": 0.008942172070965171, "time": 0.8138460636138916, "epoch": 12, "memory": 25074, "step": 67238} +{"lr": 0.000174495279317155, "data_time": 0.001655268669128418, "loss": 0.00155441677197814, "time": 0.7601854801177979, "epoch": 12, "memory": 25074, "step": 67338} +{"lr": 0.000174495279317155, "data_time": 0.0016769647598266601, "loss": 0.0015118430368602277, "time": 0.8197365522384643, "epoch": 12, "memory": 25074, "step": 67438} +{"lr": 0.000174495279317155, "data_time": 0.0015154123306274415, "loss": 0.0025519861839711664, "time": 0.7984761953353882, "epoch": 12, "memory": 25074, "step": 67538} +{"lr": 0.000174495279317155, "data_time": 0.0014311313629150391, "loss": 0.004107937542721629, "time": 0.7772931098937989, "epoch": 12, "memory": 25074, "step": 67638} +{"lr": 0.000174495279317155, "data_time": 0.0019978761672973635, "loss": 0.0005500425584614276, "time": 0.7709338903427124, "epoch": 12, "memory": 25074, "step": 67738} +{"lr": 0.000174495279317155, "data_time": 0.001685023307800293, "loss": 0.002001170115545392, "time": 0.7850199937820435, "epoch": 12, "memory": 25074, "step": 67838} +{"lr": 0.000174495279317155, "data_time": 0.0013102054595947265, "loss": 0.0009745461866259575, "time": 0.7552722215652465, "epoch": 12, "memory": 25074, "step": 67938} +{"lr": 0.000174495279317155, "data_time": 0.0014890670776367188, "loss": 0.0018088762648403645, "time": 0.7943161964416504, "epoch": 12, "memory": 25074, "step": 68038} +{"lr": 0.000174495279317155, "data_time": 0.0015288829803466798, "loss": 0.0018784672021865845, "time": 0.782120680809021, "epoch": 12, "memory": 25074, "step": 68138} +{"lr": 0.000174495279317155, "data_time": 0.001427030563354492, "loss": 0.003732118383049965, "time": 0.7712553977966309, "epoch": 12, "memory": 25074, "step": 68238} +{"lr": 0.000174495279317155, "data_time": 0.0013137578964233399, "loss": 0.0010923946276307106, "time": 0.8313997745513916, "epoch": 12, "memory": 25074, "step": 68338} +{"lr": 0.000174495279317155, "data_time": 0.001332569122314453, "loss": 0.0035476938355714083, "time": 0.8069141387939454, "epoch": 12, "memory": 25074, "step": 68438} +{"lr": 0.000174495279317155, "data_time": 0.0022151947021484377, "loss": 0.0010505390353500843, "time": 0.839823317527771, "epoch": 12, "memory": 25074, "step": 68538} +{"lr": 0.000174495279317155, "data_time": 0.0016594409942626953, "loss": 0.0003214818425476551, "time": 0.798504638671875, "epoch": 12, "memory": 25074, "step": 68638} +{"lr": 0.000174495279317155, "data_time": 0.0015073060989379884, "loss": 0.0012518554460257292, "time": 0.8257957696914673, "epoch": 12, "memory": 25074, "step": 68738} +{"lr": 0.000174495279317155, "data_time": 0.0016289472579956055, "loss": 0.0008868366945534945, "time": 0.7688867568969726, "epoch": 12, "memory": 25074, "step": 68838} +{"lr": 0.000174495279317155, "data_time": 0.0017962932586669921, "loss": 0.0007607958279550076, "time": 0.7945826768875122, "epoch": 12, "memory": 25074, "step": 68938} +{"lr": 0.000174495279317155, "data_time": 0.0019196033477783202, "loss": 0.002332293661311269, "time": 0.7335556507110595, "epoch": 12, "memory": 25074, "step": 69038} +{"accuracy/top1": 99.99368286132812, "data_time": 0.24987386254703298, "time": 0.6355440616607666, "step": 12} +{"lr": 0.00014474168609688524, "data_time": 0.07275240421295166, "loss": 0.0009204788133502006, "time": 0.8508753061294556, "epoch": 13, "memory": 25074, "step": 69196} +{"lr": 0.00014474168609688524, "data_time": 0.002440643310546875, "loss": 0.0006284591276198625, "time": 0.763164472579956, "epoch": 13, "memory": 25074, "step": 69296} +{"lr": 0.00014474168609688524, "data_time": 0.05001986026763916, "loss": 0.00031299563124775887, "time": 0.7638823509216308, "epoch": 13, "memory": 25074, "step": 69396} +{"lr": 0.00014474168609688524, "data_time": 0.024952101707458495, "loss": 0.0012822882272303104, "time": 0.7330520391464234, "epoch": 13, "memory": 25074, "step": 69496} +{"lr": 0.00014474168609688524, "data_time": 0.10140235424041748, "loss": 0.002310385042801499, "time": 0.8057371854782105, "epoch": 13, "memory": 25074, "step": 69596} +{"lr": 0.00014474168609688524, "data_time": 0.001494145393371582, "loss": 0.0022555193398147823, "time": 0.7540436029434204, "epoch": 13, "memory": 25074, "step": 69696} +{"lr": 0.00014474168609688524, "data_time": 0.001599574089050293, "loss": 0.0020067887380719185, "time": 0.7368262767791748, "epoch": 13, "memory": 25074, "step": 69796} +{"lr": 0.00014474168609688524, "data_time": 0.0016867399215698242, "loss": 0.001377742551267147, "time": 0.7241564750671386, "epoch": 13, "memory": 25074, "step": 69896} +{"lr": 0.00014474168609688524, "data_time": 0.0014216899871826172, "loss": 0.00044514876790344713, "time": 0.7239116907119751, "epoch": 13, "memory": 25074, "step": 69996} +{"lr": 0.00014474168609688524, "data_time": 0.0018394708633422852, "loss": 0.001726783299818635, "time": 0.7723653793334961, "epoch": 13, "memory": 25074, "step": 70096} +{"lr": 0.00014474168609688524, "data_time": 0.0017693996429443359, "loss": 0.0007506433874368667, "time": 0.7634341955184937, "epoch": 13, "memory": 25074, "step": 70196} +{"lr": 0.00014474168609688524, "data_time": 0.0022705554962158202, "loss": 0.0024152600672096014, "time": 0.7472733497619629, "epoch": 13, "memory": 25074, "step": 70296} +{"lr": 0.00014474168609688524, "data_time": 0.0021012544631958006, "loss": 0.003978089103475213, "time": 0.7653134346008301, "epoch": 13, "memory": 25074, "step": 70396} +{"lr": 0.00014474168609688524, "data_time": 0.0019349336624145507, "loss": 0.001471764873713255, "time": 0.7781517028808593, "epoch": 13, "memory": 25074, "step": 70496} +{"lr": 0.00014474168609688524, "data_time": 0.0021084070205688475, "loss": 0.0016972984187304974, "time": 0.7192719936370849, "epoch": 13, "memory": 25074, "step": 70596} +{"lr": 0.00014474168609688524, "data_time": 0.0018044710159301758, "loss": 0.004171270085498691, "time": 0.7994264841079712, "epoch": 13, "memory": 25074, "step": 70696} +{"lr": 0.00014474168609688524, "data_time": 0.0025376796722412108, "loss": 0.0013212434481829406, "time": 0.7311803102493286, "epoch": 13, "memory": 25074, "step": 70796} +{"lr": 0.00014474168609688524, "data_time": 0.001797652244567871, "loss": 0.0005765300709754228, "time": 0.7530529737472534, "epoch": 13, "memory": 25074, "step": 70896} +{"lr": 0.00014474168609688524, "data_time": 0.002409029006958008, "loss": 0.0020185355097055435, "time": 0.7860418796539307, "epoch": 13, "memory": 25074, "step": 70996} +{"lr": 0.00014474168609688524, "data_time": 0.0017685890197753906, "loss": 0.0007795645389705897, "time": 0.7633878469467164, "epoch": 13, "memory": 25074, "step": 71096} +{"lr": 0.00014474168609688524, "data_time": 0.0022638320922851564, "loss": 0.0016027627978473901, "time": 0.7482191562652588, "epoch": 13, "memory": 25074, "step": 71196} +{"lr": 0.00014474168609688524, "data_time": 0.0017683744430541993, "loss": 0.0003004889469593763, "time": 0.7662945032119751, "epoch": 13, "memory": 25074, "step": 71296} +{"lr": 0.00014474168609688524, "data_time": 0.0013980627059936523, "loss": 0.0008117327932268381, "time": 0.785521125793457, "epoch": 13, "memory": 25074, "step": 71396} +{"lr": 0.00014474168609688524, "data_time": 0.0014436960220336914, "loss": 0.001178226713091135, "time": 0.7426433801651001, "epoch": 13, "memory": 25074, "step": 71496} +{"lr": 0.00014474168609688524, "data_time": 0.0012624263763427734, "loss": 0.0030744282994419336, "time": 0.795448112487793, "epoch": 13, "memory": 25074, "step": 71596} +{"lr": 0.00014474168609688524, "data_time": 0.0018224477767944335, "loss": 0.0013457435183227063, "time": 0.7912577390670776, "epoch": 13, "memory": 25074, "step": 71696} +{"lr": 0.00014474168609688524, "data_time": 0.001501178741455078, "loss": 0.0009950317442417144, "time": 0.7682091474533081, "epoch": 13, "memory": 25074, "step": 71796} +{"lr": 0.00014474168609688524, "data_time": 0.0017308712005615235, "loss": 0.0005844507366418838, "time": 0.7579143285751343, "epoch": 13, "memory": 25074, "step": 71896} +{"lr": 0.00014474168609688524, "data_time": 0.00167236328125, "loss": 0.0026319696102291346, "time": 0.7687328815460205, "epoch": 13, "memory": 25074, "step": 71996} +{"lr": 0.00014474168609688524, "data_time": 0.0015536308288574218, "loss": 0.003094658488407731, "time": 0.7707023620605469, "epoch": 13, "memory": 25074, "step": 72096} +{"lr": 0.00014474168609688524, "data_time": 0.0013475418090820312, "loss": 0.002535417955368757, "time": 0.7957300186157227, "epoch": 13, "memory": 25074, "step": 72196} +{"lr": 0.00014474168609688524, "data_time": 0.0013477325439453125, "loss": 0.001134424051269889, "time": 0.7364935398101806, "epoch": 13, "memory": 25074, "step": 72296} +{"lr": 0.00014474168609688524, "data_time": 0.0015645503997802734, "loss": 0.0019105600658804178, "time": 0.7628180265426636, "epoch": 13, "memory": 25074, "step": 72396} +{"lr": 0.00014474168609688524, "data_time": 0.0017191171646118164, "loss": 0.0021224543917924164, "time": 0.7673990488052368, "epoch": 13, "memory": 25074, "step": 72496} +{"lr": 0.00014474168609688524, "data_time": 0.0015280008316040038, "loss": 0.0007678441237658262, "time": 0.7265167236328125, "epoch": 13, "memory": 25074, "step": 72596} +{"lr": 0.00014474168609688524, "data_time": 0.0014240503311157226, "loss": 0.0026364439632743596, "time": 0.7296411275863648, "epoch": 13, "memory": 25074, "step": 72696} +{"lr": 0.00014474168609688524, "data_time": 0.001578044891357422, "loss": 0.001819831971079111, "time": 0.7435076236724854, "epoch": 13, "memory": 25074, "step": 72796} +{"lr": 0.00014474168609688524, "data_time": 0.0013093948364257812, "loss": 0.0004806566052138805, "time": 0.7049421310424805, "epoch": 13, "memory": 25074, "step": 72896} +{"lr": 0.00014474168609688524, "data_time": 0.0013599395751953125, "loss": 0.001344196079298854, "time": 0.7304930210113525, "epoch": 13, "memory": 25074, "step": 72996} +{"lr": 0.00014474168609688524, "data_time": 0.0015346765518188476, "loss": 0.0008602182846516371, "time": 0.7381244897842407, "epoch": 13, "memory": 25074, "step": 73096} +{"lr": 0.00014474168609688524, "data_time": 0.0015476226806640625, "loss": 0.0038446308579295875, "time": 0.7094938516616821, "epoch": 13, "memory": 25074, "step": 73196} +{"lr": 0.00014474168609688524, "data_time": 0.0014039039611816405, "loss": 0.0012205264065414667, "time": 0.7096496820449829, "epoch": 13, "memory": 25074, "step": 73296} +{"lr": 0.00014474168609688524, "data_time": 0.0014803409576416016, "loss": 0.0006892472971230746, "time": 0.7412248611450195, "epoch": 13, "memory": 25074, "step": 73396} +{"lr": 0.00014474168609688524, "data_time": 0.0016997575759887696, "loss": 0.005097674299031496, "time": 0.7486302375793457, "epoch": 13, "memory": 25074, "step": 73496} +{"lr": 0.00014474168609688524, "data_time": 0.001591658592224121, "loss": 0.000147221889346838, "time": 0.7732123136520386, "epoch": 13, "memory": 25074, "step": 73596} +{"lr": 0.00014474168609688524, "data_time": 0.0015129804611206054, "loss": 0.002005761070176959, "time": 0.7478807449340821, "epoch": 13, "memory": 25074, "step": 73696} +{"lr": 0.00014474168609688524, "data_time": 0.0013300657272338867, "loss": 0.0010326614137738943, "time": 0.7430232763290405, "epoch": 13, "memory": 25074, "step": 73796} +{"lr": 0.00014474168609688524, "data_time": 0.0015828371047973632, "loss": 0.0009551784489303828, "time": 0.7274349689483642, "epoch": 13, "memory": 25074, "step": 73896} +{"lr": 0.00014474168609688524, "data_time": 0.0015261173248291016, "loss": 0.001108011044561863, "time": 0.7866940259933471, "epoch": 13, "memory": 25074, "step": 73996} +{"lr": 0.00014474168609688524, "data_time": 0.0013470888137817384, "loss": 0.0027155133429914713, "time": 0.7083929061889649, "epoch": 13, "memory": 25074, "step": 74096} +{"lr": 0.00014474168609688524, "data_time": 0.0017130374908447266, "loss": 0.003259146912023425, "time": 0.8713862895965576, "epoch": 13, "memory": 25074, "step": 74196} +{"lr": 0.00014474168609688524, "data_time": 0.0014954566955566405, "loss": 0.003147553373128176, "time": 0.7065548181533814, "epoch": 13, "memory": 25074, "step": 74296} +{"lr": 0.00014474168609688524, "data_time": 0.0012180089950561523, "loss": 0.0007164672948420048, "time": 0.7848201036453247, "epoch": 13, "memory": 25074, "step": 74396} +{"lr": 0.00014474168609688524, "data_time": 0.0015608549118041992, "loss": 0.00037650628946721553, "time": 0.7286721706390381, "epoch": 13, "memory": 25074, "step": 74496} +{"lr": 0.00014474168609688524, "data_time": 0.0016849040985107422, "loss": 0.0008910450153052807, "time": 0.7445698022842407, "epoch": 13, "memory": 25074, "step": 74596} +{"lr": 0.00014474168609688524, "data_time": 0.0016117095947265625, "loss": 0.005756586650386452, "time": 0.7053746461868287, "epoch": 13, "memory": 25074, "step": 74696} +{"lr": 0.00014474168609688524, "data_time": 0.0014872550964355469, "loss": 0.0018001841846853495, "time": 0.7206930875778198, "epoch": 13, "memory": 25074, "step": 74796} +{"accuracy/top1": 99.99368286132812, "data_time": 0.2428590409895953, "time": 0.6288780184353099, "step": 13} +{"lr": 0.00011647185255078837, "data_time": 0.023251724243164063, "loss": 0.0021779678761959078, "time": 0.77837233543396, "epoch": 14, "memory": 25074, "step": 74954} +{"lr": 0.00011647185255078837, "data_time": 0.0015166044235229493, "loss": 0.0023673285264521836, "time": 0.7371764421463013, "epoch": 14, "memory": 25074, "step": 75054} +{"lr": 0.00011647185255078837, "data_time": 0.0015041828155517578, "loss": 0.0009057887364178896, "time": 0.7058263063430786, "epoch": 14, "memory": 25074, "step": 75154} +{"lr": 0.00011647185255078837, "data_time": 0.0018524408340454101, "loss": 0.0011784181464463473, "time": 0.718956470489502, "epoch": 14, "memory": 25074, "step": 75254} +{"lr": 0.00011647185255078837, "data_time": 0.04562981128692627, "loss": 0.001378782605752349, "time": 0.749661135673523, "epoch": 14, "memory": 25074, "step": 75354} +{"lr": 0.00011647185255078837, "data_time": 0.001342034339904785, "loss": 0.0030419050250202417, "time": 0.7059260845184326, "epoch": 14, "memory": 25074, "step": 75454} +{"lr": 0.00011647185255078837, "data_time": 0.014382219314575196, "loss": 0.0014209263026714325, "time": 0.7180840253829956, "epoch": 14, "memory": 25074, "step": 75554} +{"lr": 0.00011647185255078837, "data_time": 0.042363214492797854, "loss": 0.001219785725697875, "time": 0.7468163251876831, "epoch": 14, "memory": 25074, "step": 75654} +{"lr": 0.00011647185255078837, "data_time": 0.0016139745712280273, "loss": 9.78053081780672e-05, "time": 0.71507728099823, "epoch": 14, "memory": 25074, "step": 75754} +{"lr": 0.00011647185255078837, "data_time": 0.0015370845794677734, "loss": 0.0037169348448514937, "time": 0.7629705905914307, "epoch": 14, "memory": 25074, "step": 75854} +{"lr": 0.00011647185255078837, "data_time": 0.0018133878707885741, "loss": 0.0007813195697963238, "time": 0.7463514328002929, "epoch": 14, "memory": 25074, "step": 75954} +{"lr": 0.00011647185255078837, "data_time": 0.002047610282897949, "loss": 0.0010452068410813809, "time": 0.7900663137435913, "epoch": 14, "memory": 25074, "step": 76054} +{"lr": 0.00011647185255078837, "data_time": 0.001435399055480957, "loss": 0.000550618302077055, "time": 0.7513041019439697, "epoch": 14, "memory": 25074, "step": 76154} +{"lr": 0.00011647185255078837, "data_time": 0.0014093160629272462, "loss": 0.0002473832108080387, "time": 0.7352958440780639, "epoch": 14, "memory": 25074, "step": 76254} +{"lr": 0.00011647185255078837, "data_time": 0.0015857219696044922, "loss": 0.002685864083468914, "time": 0.7190948486328125, "epoch": 14, "memory": 25074, "step": 76354} +{"lr": 0.00011647185255078837, "data_time": 0.0014364242553710938, "loss": 0.00013651251792907716, "time": 0.7147719383239746, "epoch": 14, "memory": 25074, "step": 76454} +{"lr": 0.00011647185255078837, "data_time": 0.0014735937118530273, "loss": 0.00056815380230546, "time": 0.715182113647461, "epoch": 14, "memory": 25074, "step": 76554} +{"lr": 0.00011647185255078837, "data_time": 0.0016825437545776368, "loss": 0.0004540197551250458, "time": 0.7560132741928101, "epoch": 14, "memory": 25074, "step": 76654} +{"lr": 0.00011647185255078837, "data_time": 0.0014857769012451172, "loss": 0.0006996762473136187, "time": 0.727306318283081, "epoch": 14, "memory": 25074, "step": 76754} +{"lr": 0.00011647185255078837, "data_time": 0.0013864278793334962, "loss": 0.0016852287575602532, "time": 0.7824569702148437, "epoch": 14, "memory": 25074, "step": 76854} +{"lr": 0.00011647185255078837, "data_time": 0.0014194726943969726, "loss": 0.000279172183945775, "time": 0.7624621152877807, "epoch": 14, "memory": 25074, "step": 76954} +{"lr": 0.00011647185255078837, "data_time": 0.0013602256774902343, "loss": 0.00285165524110198, "time": 0.806310510635376, "epoch": 14, "memory": 25074, "step": 77054} +{"lr": 0.00011647185255078837, "data_time": 0.0013950347900390625, "loss": 0.0007428525947034359, "time": 0.786906099319458, "epoch": 14, "memory": 25074, "step": 77154} +{"lr": 0.00011647185255078837, "data_time": 0.0013521194458007812, "loss": 0.0013474670238792896, "time": 0.8502303838729859, "epoch": 14, "memory": 25074, "step": 77254} +{"lr": 0.00011647185255078837, "data_time": 0.0014919996261596679, "loss": 0.0014406180940568446, "time": 0.7569593906402587, "epoch": 14, "memory": 25074, "step": 77354} +{"lr": 0.00011647185255078837, "data_time": 0.0013175487518310546, "loss": 0.0016626224853098393, "time": 0.8024927616119385, "epoch": 14, "memory": 25074, "step": 77454} +{"lr": 0.00011647185255078837, "data_time": 0.0014471292495727539, "loss": 0.0018643232993781566, "time": 0.8707294225692749, "epoch": 14, "memory": 25074, "step": 77554} +{"lr": 0.00011647185255078837, "data_time": 0.001388382911682129, "loss": 0.001479786215350032, "time": 0.8227628469467163, "epoch": 14, "memory": 25074, "step": 77654} +{"lr": 0.00011647185255078837, "data_time": 0.0014124631881713868, "loss": 0.0031971388030797242, "time": 0.7924062252044678, "epoch": 14, "memory": 25074, "step": 77754} +{"lr": 0.00011647185255078837, "data_time": 0.0013155460357666016, "loss": 0.001103545445948839, "time": 0.7877817392349243, "epoch": 14, "memory": 25074, "step": 77854} +{"lr": 0.00011647185255078837, "data_time": 0.001440286636352539, "loss": 0.0007647598162293434, "time": 0.7607779026031494, "epoch": 14, "memory": 25074, "step": 77954} +{"lr": 0.00011647185255078837, "data_time": 0.0014202356338500976, "loss": 0.0008859530556946993, "time": 0.8226583242416382, "epoch": 14, "memory": 25074, "step": 78054} +{"lr": 0.00011647185255078837, "data_time": 0.0016968488693237306, "loss": 0.00011167400516569615, "time": 0.7710731029510498, "epoch": 14, "memory": 25074, "step": 78154} +{"lr": 0.00011647185255078837, "data_time": 0.0014856338500976562, "loss": 0.006312939804047346, "time": 0.7489454030990601, "epoch": 14, "memory": 25074, "step": 78254} +{"lr": 0.00011647185255078837, "data_time": 0.0014198780059814452, "loss": 0.0017056366894394159, "time": 0.7752131462097168, "epoch": 14, "memory": 25074, "step": 78354} +{"lr": 0.00011647185255078837, "data_time": 0.0014912605285644532, "loss": 0.002734039630740881, "time": 0.7770315408706665, "epoch": 14, "memory": 25074, "step": 78454} +{"lr": 0.00011647185255078837, "data_time": 0.0012785911560058594, "loss": 0.00021743196994066237, "time": 0.7677789211273194, "epoch": 14, "memory": 25074, "step": 78554} +{"lr": 0.00011647185255078837, "data_time": 0.0016590356826782227, "loss": 0.0027144406456500294, "time": 0.7268472194671631, "epoch": 14, "memory": 25074, "step": 78654} +{"lr": 0.00011647185255078837, "data_time": 0.001490473747253418, "loss": 0.0022927312646061184, "time": 0.7710299730300904, "epoch": 14, "memory": 25074, "step": 78754} +{"lr": 0.00011647185255078837, "data_time": 0.001482844352722168, "loss": 0.0015905157662928105, "time": 0.8276148557662963, "epoch": 14, "memory": 25074, "step": 78854} +{"lr": 0.00011647185255078837, "data_time": 0.0015105962753295898, "loss": 0.0009809607174247504, "time": 0.782722020149231, "epoch": 14, "memory": 25074, "step": 78954} +{"lr": 0.00011647185255078837, "data_time": 0.0013812780380249023, "loss": 0.0021457645576447248, "time": 0.7340906620025635, "epoch": 14, "memory": 25074, "step": 79054} +{"lr": 0.00011647185255078837, "data_time": 0.0014567136764526366, "loss": 0.0016096735373139382, "time": 0.7725087642669678, "epoch": 14, "memory": 25074, "step": 79154} +{"lr": 0.00011647185255078837, "data_time": 0.0012729644775390625, "loss": 0.003232082212343812, "time": 0.7204482316970825, "epoch": 14, "memory": 25074, "step": 79254} +{"lr": 0.00011647185255078837, "data_time": 0.0015396833419799804, "loss": 0.0014629330486059188, "time": 0.762050724029541, "epoch": 14, "memory": 25074, "step": 79354} +{"lr": 0.00011647185255078837, "data_time": 0.0014620542526245118, "loss": 0.0010259154718369246, "time": 0.7450391292572022, "epoch": 14, "memory": 25074, "step": 79454} +{"lr": 0.00011647185255078837, "data_time": 0.0015154123306274415, "loss": 0.00039701308123767375, "time": 0.757326078414917, "epoch": 14, "memory": 25074, "step": 79554} +{"lr": 0.00011647185255078837, "data_time": 0.0014201641082763673, "loss": 0.00040809502825140954, "time": 0.7113478899002075, "epoch": 14, "memory": 25074, "step": 79654} +{"lr": 0.00011647185255078837, "data_time": 0.0014184474945068359, "loss": 0.0016616976354271173, "time": 0.7391294717788697, "epoch": 14, "memory": 25074, "step": 79754} +{"lr": 0.00011647185255078837, "data_time": 0.0013500213623046874, "loss": 0.0005124249029904604, "time": 0.8219952344894409, "epoch": 14, "memory": 25074, "step": 79854} +{"lr": 0.00011647185255078837, "data_time": 0.0013647079467773438, "loss": 0.00026878686621785165, "time": 0.7242864847183228, "epoch": 14, "memory": 25074, "step": 79954} +{"lr": 0.00011647185255078837, "data_time": 0.0013128995895385742, "loss": 0.0007196391932666301, "time": 0.7184814453125, "epoch": 14, "memory": 25074, "step": 80054} +{"lr": 0.00011647185255078837, "data_time": 0.0013649463653564453, "loss": 0.00015480811707675458, "time": 0.7506291389465332, "epoch": 14, "memory": 25074, "step": 80154} +{"lr": 0.00011647185255078837, "data_time": 0.0014166116714477539, "loss": 0.00039838752709329126, "time": 0.740925645828247, "epoch": 14, "memory": 25074, "step": 80254} +{"lr": 0.00011647185255078837, "data_time": 0.0018710851669311523, "loss": 0.0003164768684655428, "time": 0.7214632987976074, "epoch": 14, "memory": 25074, "step": 80354} +{"lr": 0.00011647185255078837, "data_time": 0.0014228105545043945, "loss": 0.0037071477621793748, "time": 0.7352063894271851, "epoch": 14, "memory": 25074, "step": 80454} +{"lr": 0.00011647185255078837, "data_time": 0.0016200542449951172, "loss": 0.0006053172051906586, "time": 0.7899404525756836, "epoch": 14, "memory": 25074, "step": 80554} +{"accuracy/top1": 99.99368286132812, "data_time": 0.24431033695445342, "time": 0.631132588667028, "step": 14} +{"lr": 9.038187580296775e-05, "data_time": 0.001281404495239258, "loss": 0.0009693463332951068, "time": 0.8083181619644165, "epoch": 15, "memory": 25074, "step": 80712} +{"lr": 9.038187580296775e-05, "data_time": 0.0013797521591186524, "loss": 0.0024553865659981966, "time": 0.7607942581176758, "epoch": 15, "memory": 25074, "step": 80812} +{"lr": 9.038187580296775e-05, "data_time": 0.001313185691833496, "loss": 0.0005513661075383425, "time": 0.782365870475769, "epoch": 15, "memory": 25074, "step": 80912} +{"lr": 9.038187580296775e-05, "data_time": 0.001375126838684082, "loss": 0.0003903736360371113, "time": 0.7232391595840454, "epoch": 15, "memory": 25074, "step": 81012} +{"lr": 9.038187580296775e-05, "data_time": 0.0014269590377807618, "loss": 0.0017582374624907971, "time": 0.8211493015289306, "epoch": 15, "memory": 25074, "step": 81112} +{"lr": 9.038187580296775e-05, "data_time": 0.0016207218170166016, "loss": 0.0016284781508147717, "time": 0.8221693515777588, "epoch": 15, "memory": 25074, "step": 81212} +{"lr": 9.038187580296775e-05, "data_time": 0.0014461755752563476, "loss": 0.0007385155651718379, "time": 0.8281132459640503, "epoch": 15, "memory": 25074, "step": 81312} +{"lr": 9.038187580296775e-05, "data_time": 0.0014560937881469727, "loss": 0.0007029476575553418, "time": 0.7273197174072266, "epoch": 15, "memory": 25074, "step": 81412} +{"lr": 9.038187580296775e-05, "data_time": 0.0018094778060913086, "loss": 0.0013593892101198435, "time": 0.7591881513595581, "epoch": 15, "memory": 25074, "step": 81512} +{"lr": 9.038187580296775e-05, "data_time": 0.003216552734375, "loss": 0.0018323538359254598, "time": 0.7899890184402466, "epoch": 15, "memory": 25074, "step": 81612} +{"lr": 9.038187580296775e-05, "data_time": 0.0020852088928222656, "loss": 0.00044703842140734194, "time": 0.762487530708313, "epoch": 15, "memory": 25074, "step": 81712} +{"lr": 9.038187580296775e-05, "data_time": 0.0015711069107055664, "loss": 0.001037484547123313, "time": 0.7359006643295288, "epoch": 15, "memory": 25074, "step": 81812} +{"lr": 9.038187580296775e-05, "data_time": 0.0015401840209960938, "loss": 0.0020602371543645857, "time": 0.8280802488327026, "epoch": 15, "memory": 25074, "step": 81912} +{"lr": 9.038187580296775e-05, "data_time": 0.0017678499221801757, "loss": 0.0019632921554148196, "time": 0.7969156980514527, "epoch": 15, "memory": 25074, "step": 82012} +{"lr": 9.038187580296775e-05, "data_time": 0.0015585184097290038, "loss": 0.0008575992193073035, "time": 0.7718917846679687, "epoch": 15, "memory": 25074, "step": 82112} +{"lr": 9.038187580296775e-05, "data_time": 0.0015117645263671875, "loss": 0.0021742898039519786, "time": 0.8478286981582641, "epoch": 15, "memory": 25074, "step": 82212} +{"lr": 9.038187580296775e-05, "data_time": 0.0016597747802734376, "loss": 0.00014989529736340046, "time": 0.8146865606307984, "epoch": 15, "memory": 25074, "step": 82312} +{"lr": 9.038187580296775e-05, "data_time": 0.0017172574996948242, "loss": 3.949040547013283e-05, "time": 0.8910597801208496, "epoch": 15, "memory": 25074, "step": 82412} +{"lr": 9.038187580296775e-05, "data_time": 0.0019261360168457032, "loss": 0.00012080157175660133, "time": 0.729236364364624, "epoch": 15, "memory": 25074, "step": 82512} +{"lr": 9.038187580296775e-05, "data_time": 0.0016622304916381835, "loss": 0.0007462635636329651, "time": 0.8027929782867431, "epoch": 15, "memory": 25074, "step": 82612} +{"lr": 9.038187580296775e-05, "data_time": 0.001559901237487793, "loss": 0.001427473733201623, "time": 0.8323099136352539, "epoch": 15, "memory": 25074, "step": 82712} +{"lr": 9.038187580296775e-05, "data_time": 0.0014462471008300781, "loss": 0.0003619640599936247, "time": 0.8937884330749511, "epoch": 15, "memory": 25074, "step": 82812} +{"lr": 9.038187580296775e-05, "data_time": 0.0015408992767333984, "loss": 0.0006396289449185133, "time": 0.7980130910873413, "epoch": 15, "memory": 25074, "step": 82912} +{"lr": 9.038187580296775e-05, "data_time": 0.0015927553176879883, "loss": 0.0003920118324458599, "time": 1.0595411777496337, "epoch": 15, "memory": 25074, "step": 83012} +{"lr": 9.038187580296775e-05, "data_time": 0.0015401601791381835, "loss": 0.0009338007308542729, "time": 0.9077517271041871, "epoch": 15, "memory": 25074, "step": 83112} +{"lr": 9.038187580296775e-05, "data_time": 0.0015286922454833985, "loss": 0.0019225267693400382, "time": 0.9691908359527588, "epoch": 15, "memory": 25074, "step": 83212} +{"lr": 9.038187580296775e-05, "data_time": 0.0018684148788452148, "loss": 0.0013671924360096454, "time": 0.7199590682983399, "epoch": 15, "memory": 25074, "step": 83312} +{"lr": 9.038187580296775e-05, "data_time": 0.0014957904815673829, "loss": 0.00010000835172832012, "time": 0.8423433065414428, "epoch": 15, "memory": 25074, "step": 83412} +{"lr": 9.038187580296775e-05, "data_time": 0.0013504505157470703, "loss": 0.002599572343751788, "time": 0.8672428607940674, "epoch": 15, "memory": 25074, "step": 83512} +{"lr": 9.038187580296775e-05, "data_time": 0.0018329143524169922, "loss": 0.0005764301866292953, "time": 0.8385485410690308, "epoch": 15, "memory": 25074, "step": 83612} +{"lr": 9.038187580296775e-05, "data_time": 0.0014724969863891602, "loss": 0.0009647830855101347, "time": 0.8059179306030273, "epoch": 15, "memory": 25074, "step": 83712} +{"lr": 9.038187580296775e-05, "data_time": 0.0014623641967773438, "loss": 0.0003822332713752985, "time": 1.0916605472564698, "epoch": 15, "memory": 25074, "step": 83812} +{"lr": 9.038187580296775e-05, "data_time": 0.0016545772552490235, "loss": 0.0016348369419574738, "time": 1.1889729022979736, "epoch": 15, "memory": 25074, "step": 83912} +{"lr": 9.038187580296775e-05, "data_time": 0.0013372182846069336, "loss": 0.0008744071703404189, "time": 1.000340986251831, "epoch": 15, "memory": 25074, "step": 84012} +{"lr": 9.038187580296775e-05, "data_time": 0.0031939983367919923, "loss": 0.003211942035704851, "time": 2.5094349861145018, "epoch": 15, "memory": 25074, "step": 84112} +{"lr": 9.038187580296775e-05, "data_time": 0.0013994932174682616, "loss": 0.0005596747156232596, "time": 0.9590139627456665, "epoch": 15, "memory": 25074, "step": 84212} +{"lr": 9.038187580296775e-05, "data_time": 0.0014598846435546875, "loss": 0.0009359739255160093, "time": 1.0496535539627074, "epoch": 15, "memory": 25074, "step": 84312} +{"lr": 9.038187580296775e-05, "data_time": 0.001909017562866211, "loss": 0.00027056271210312843, "time": 0.7710723161697388, "epoch": 15, "memory": 25074, "step": 84412} +{"lr": 9.038187580296775e-05, "data_time": 0.0012775182723999024, "loss": 0.0008024506270885468, "time": 0.7106168985366821, "epoch": 15, "memory": 25074, "step": 84512} +{"lr": 9.038187580296775e-05, "data_time": 0.0013846158981323242, "loss": 0.0017863199580460786, "time": 0.7040702104568481, "epoch": 15, "memory": 25074, "step": 84612} +{"lr": 9.038187580296775e-05, "data_time": 0.0012623786926269532, "loss": 0.0021234650630503893, "time": 0.7278531312942504, "epoch": 15, "memory": 25074, "step": 84712} +{"lr": 9.038187580296775e-05, "data_time": 0.001620650291442871, "loss": 0.0017398543655872344, "time": 0.7522794008255005, "epoch": 15, "memory": 25074, "step": 84812} +{"lr": 9.038187580296775e-05, "data_time": 0.0013030290603637696, "loss": 0.0006522384937852621, "time": 0.7269864320755005, "epoch": 15, "memory": 25074, "step": 84912} +{"lr": 9.038187580296775e-05, "data_time": 0.0014772415161132812, "loss": 0.00027401009574532507, "time": 0.7073343992233276, "epoch": 15, "memory": 25074, "step": 85012} +{"lr": 9.038187580296775e-05, "data_time": 0.0016780614852905274, "loss": 0.000629105232656002, "time": 0.7588959693908691, "epoch": 15, "memory": 25074, "step": 85112} +{"lr": 9.038187580296775e-05, "data_time": 0.001724386215209961, "loss": 0.0001358284614980221, "time": 0.719288420677185, "epoch": 15, "memory": 25074, "step": 85212} +{"lr": 9.038187580296775e-05, "data_time": 0.002022528648376465, "loss": 0.0004343196284025908, "time": 0.7302109241485596, "epoch": 15, "memory": 25074, "step": 85312} +{"lr": 9.038187580296775e-05, "data_time": 0.0014947891235351563, "loss": 0.0002664435654878616, "time": 0.8181253671646118, "epoch": 15, "memory": 25074, "step": 85412} +{"lr": 9.038187580296775e-05, "data_time": 0.001249384880065918, "loss": 0.0016120796091854573, "time": 0.7277424812316895, "epoch": 15, "memory": 25074, "step": 85512} +{"lr": 9.038187580296775e-05, "data_time": 0.001448678970336914, "loss": 0.000722819659858942, "time": 0.7536602020263672, "epoch": 15, "memory": 25074, "step": 85612} +{"lr": 9.038187580296775e-05, "data_time": 0.0014109134674072266, "loss": 0.0005820733960717917, "time": 0.7464721441268921, "epoch": 15, "memory": 25074, "step": 85712} +{"lr": 9.038187580296775e-05, "data_time": 0.0015386819839477539, "loss": 0.0021876913495361807, "time": 0.7061995267868042, "epoch": 15, "memory": 25074, "step": 85812} +{"lr": 9.038187580296775e-05, "data_time": 0.0014644384384155274, "loss": 0.0002771410159766674, "time": 0.7733000755310059, "epoch": 15, "memory": 25074, "step": 85912} +{"lr": 9.038187580296775e-05, "data_time": 0.002461814880371094, "loss": 0.002774343313649297, "time": 0.8090592861175537, "epoch": 15, "memory": 25074, "step": 86012} +{"lr": 9.038187580296775e-05, "data_time": 0.002170419692993164, "loss": 0.00018875827081501483, "time": 0.7727634906768799, "epoch": 15, "memory": 25074, "step": 86112} +{"lr": 9.038187580296775e-05, "data_time": 0.0017713546752929688, "loss": 0.0019611213356256485, "time": 0.7775627374649048, "epoch": 15, "memory": 25074, "step": 86212} +{"lr": 9.038187580296775e-05, "data_time": 0.0011821746826171874, "loss": 0.0011772952042520047, "time": 0.720346188545227, "epoch": 15, "memory": 25074, "step": 86312} +{"accuracy/top1": 99.99368286132812, "data_time": 0.2504125482895795, "time": 0.6420064393211814, "step": 15} +{"lr": 6.711417766862323e-05, "data_time": 0.001391148567199707, "loss": 0.0008806531317532062, "time": 0.770824646949768, "epoch": 16, "memory": 25074, "step": 86470} +{"lr": 6.711417766862323e-05, "data_time": 0.001569509506225586, "loss": 0.0004885589703917503, "time": 0.7555091857910157, "epoch": 16, "memory": 25074, "step": 86570} +{"lr": 6.711417766862323e-05, "data_time": 0.0642164945602417, "loss": 0.00020497986115515233, "time": 0.8285637617111206, "epoch": 16, "memory": 25074, "step": 86670} +{"lr": 6.711417766862323e-05, "data_time": 0.08859062194824219, "loss": 0.0005509410053491593, "time": 0.7917944908142089, "epoch": 16, "memory": 25074, "step": 86770} +{"lr": 6.711417766862323e-05, "data_time": 0.06232173442840576, "loss": 0.0005546720698475837, "time": 0.7812191009521484, "epoch": 16, "memory": 25074, "step": 86870} +{"lr": 6.711417766862323e-05, "data_time": 0.08091220855712891, "loss": 0.0002881082706153393, "time": 0.7860160589218139, "epoch": 16, "memory": 25074, "step": 86970} +{"lr": 6.711417766862323e-05, "data_time": 0.034747886657714847, "loss": 0.0006381961517035961, "time": 0.7529673099517822, "epoch": 16, "memory": 25074, "step": 87070} +{"lr": 6.711417766862323e-05, "data_time": 0.005439329147338867, "loss": 0.0026498822029680015, "time": 0.7797904968261719, "epoch": 16, "memory": 25074, "step": 87170} +{"lr": 6.711417766862323e-05, "data_time": 0.0014828920364379882, "loss": 0.0001123073510825634, "time": 0.739583158493042, "epoch": 16, "memory": 25074, "step": 87270} +{"lr": 6.711417766862323e-05, "data_time": 0.0014094114303588867, "loss": 0.0006772896274924278, "time": 0.7724883794784546, "epoch": 16, "memory": 25074, "step": 87370} +{"lr": 6.711417766862323e-05, "data_time": 0.0015820026397705077, "loss": 0.0003269531764090061, "time": 0.7386952638626099, "epoch": 16, "memory": 25074, "step": 87470} +{"lr": 6.711417766862323e-05, "data_time": 0.0014675140380859375, "loss": 0.0021864949725568295, "time": 0.724583911895752, "epoch": 16, "memory": 25074, "step": 87570} +{"lr": 6.711417766862323e-05, "data_time": 0.0015862464904785156, "loss": 0.001155404606834054, "time": 0.7314566612243653, "epoch": 16, "memory": 25074, "step": 87670} +{"lr": 6.711417766862323e-05, "data_time": 0.0016598939895629884, "loss": 0.00015166825614869593, "time": 0.7534103155136108, "epoch": 16, "memory": 25074, "step": 87770} +{"lr": 6.711417766862323e-05, "data_time": 0.0013062477111816407, "loss": 0.000558520620688796, "time": 0.7059796810150146, "epoch": 16, "memory": 25074, "step": 87870} +{"lr": 6.711417766862323e-05, "data_time": 0.0017880916595458985, "loss": 0.0003085790202021599, "time": 0.7124033689498901, "epoch": 16, "memory": 25074, "step": 87970} +{"lr": 6.711417766862323e-05, "data_time": 0.0014652729034423829, "loss": 0.0005887267179787159, "time": 0.7275777578353881, "epoch": 16, "memory": 25074, "step": 88070} +{"lr": 6.711417766862323e-05, "data_time": 0.0017225265502929688, "loss": 8.797864429652691e-05, "time": 0.7884787797927857, "epoch": 16, "memory": 25074, "step": 88170} +{"lr": 6.711417766862323e-05, "data_time": 0.0012642860412597657, "loss": 0.0013945603277534246, "time": 0.7215691804885864, "epoch": 16, "memory": 25074, "step": 88270} +{"lr": 6.711417766862323e-05, "data_time": 0.0012665033340454102, "loss": 0.001149605493992567, "time": 0.7747556209564209, "epoch": 16, "memory": 25074, "step": 88370} +{"lr": 6.711417766862323e-05, "data_time": 0.0013764619827270508, "loss": 0.00020167063921689988, "time": 0.7620030164718627, "epoch": 16, "memory": 25074, "step": 88470} +{"lr": 6.711417766862323e-05, "data_time": 0.001493072509765625, "loss": 7.639559917151927e-05, "time": 0.7522417306900024, "epoch": 16, "memory": 25074, "step": 88570} +{"lr": 6.711417766862323e-05, "data_time": 0.0015863418579101563, "loss": 0.0002899395301938057, "time": 0.7593885660171509, "epoch": 16, "memory": 25074, "step": 88670} +{"lr": 6.711417766862323e-05, "data_time": 0.0014772891998291015, "loss": 0.0021343606058508158, "time": 0.8097762823104858, "epoch": 16, "memory": 25074, "step": 88770} +{"lr": 6.711417766862323e-05, "data_time": 0.0017124176025390624, "loss": 2.8182798996567725e-05, "time": 0.7633267164230346, "epoch": 16, "memory": 25074, "step": 88870} +{"lr": 6.711417766862323e-05, "data_time": 0.0013273954391479492, "loss": 0.00016270913183689118, "time": 0.7777122497558594, "epoch": 16, "memory": 25074, "step": 88970} +{"lr": 6.711417766862323e-05, "data_time": 0.0015081167221069336, "loss": 0.0032006279565393926, "time": 0.7150451898574829, "epoch": 16, "memory": 25074, "step": 89070} +{"lr": 6.711417766862323e-05, "data_time": 0.0014369487762451172, "loss": 7.198280654847622e-05, "time": 0.769907283782959, "epoch": 16, "memory": 25074, "step": 89170} +{"lr": 6.711417766862323e-05, "data_time": 0.0012654304504394532, "loss": 0.0006047617178410291, "time": 0.7071659326553345, "epoch": 16, "memory": 25074, "step": 89270} +{"lr": 6.711417766862323e-05, "data_time": 0.001788759231567383, "loss": 0.005545500945299864, "time": 0.8838212490081787, "epoch": 16, "memory": 25074, "step": 89370} +{"lr": 6.711417766862323e-05, "data_time": 0.0018588066101074218, "loss": 0.003012413950636983, "time": 0.8784438848495484, "epoch": 16, "memory": 25074, "step": 89470} +{"lr": 6.711417766862323e-05, "data_time": 0.002134108543395996, "loss": 0.001166795054450631, "time": 0.8018247604370117, "epoch": 16, "memory": 25074, "step": 89570} +{"lr": 6.711417766862323e-05, "data_time": 0.001578354835510254, "loss": 0.0001045859418809414, "time": 0.8085804224014282, "epoch": 16, "memory": 25074, "step": 89670} +{"lr": 6.711417766862323e-05, "data_time": 0.0012562990188598633, "loss": 0.0007325883023440837, "time": 0.7591107368469239, "epoch": 16, "memory": 25074, "step": 89770} +{"lr": 6.711417766862323e-05, "data_time": 0.0014090776443481446, "loss": 0.0015640189405530692, "time": 0.7343702554702759, "epoch": 16, "memory": 25074, "step": 89870} +{"lr": 6.711417766862323e-05, "data_time": 0.0015744686126708985, "loss": 0.003829053184017539, "time": 0.7476104259490967, "epoch": 16, "memory": 25074, "step": 89970} +{"lr": 6.711417766862323e-05, "data_time": 0.001384592056274414, "loss": 0.0010552929248660802, "time": 0.7129765033721924, "epoch": 16, "memory": 25074, "step": 90070} diff --git a/convnext_small_4xb256_fake5m-lr4e-4/20230601_025223/vis_data/config.py b/convnext_small_4xb256_fake5m-lr4e-4/20230601_025223/vis_data/config.py new file mode 100644 index 0000000000000000000000000000000000000000..ac1ce3ab72a9def3059afc183203855eae50d3ab --- /dev/null +++ b/convnext_small_4xb256_fake5m-lr4e-4/20230601_025223/vis_data/config.py @@ -0,0 +1,169 @@ +optim_wrapper = dict( + optimizer=dict( + type='AdamW', + lr=0.0004, + weight_decay=0.05, + eps=1e-08, + betas=(0.9, 0.999), + _scope_='mmpretrain'), + paramwise_cfg=dict( + norm_decay_mult=0.0, + bias_decay_mult=0.0, + flat_decay_mult=0.0, + custom_keys=dict({ + '.absolute_pos_embed': dict(decay_mult=0.0), + '.relative_position_bias_table': dict(decay_mult=0.0) + })), + type='AmpOptimWrapper', + dtype='bfloat16', + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=True, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=20, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=4096) +model = dict( + type='ImageClassifier', + backbone=dict(type='ConvNeXt', arch='small', drop_path_rate=0.4), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=768, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + init_cfg=None), + init_cfg=dict( + type='TruncNormal', layer=['Conv2d', 'Linear'], std=0.02, bias=0.0), + train_cfg=None) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stablediffusionV1-5R2-dpmsolver-25-5m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/convnext_small_4xb256_fake5m-lr4e-4' diff --git a/convnext_small_4xb256_fake5m-lr4e-4/20230601_025223/vis_data/events.out.tfevents.1685559150.SH-IDC1-10-140-24-113.62911.0 b/convnext_small_4xb256_fake5m-lr4e-4/20230601_025223/vis_data/events.out.tfevents.1685559150.SH-IDC1-10-140-24-113.62911.0 new file mode 100644 index 0000000000000000000000000000000000000000..c58ffd58e232688d4a1cc52a68b3695ab0bb1f8e --- /dev/null +++ b/convnext_small_4xb256_fake5m-lr4e-4/20230601_025223/vis_data/events.out.tfevents.1685559150.SH-IDC1-10-140-24-113.62911.0 @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:4ef223fa3c0b9012cb692e8a2e5a1318321fa73b4c48853ccab4b9412fc93fd4 +size 1310307 diff --git a/convnext_small_4xb256_fake5m-lr4e-4/20230601_025223/vis_data/scalars.json b/convnext_small_4xb256_fake5m-lr4e-4/20230601_025223/vis_data/scalars.json new file mode 100644 index 0000000000000000000000000000000000000000..acd46c6dd81ae719bc575ff43f547118a20deac1 --- /dev/null +++ b/convnext_small_4xb256_fake5m-lr4e-4/20230601_025223/vis_data/scalars.json @@ -0,0 +1,907 @@ +{"lr": 0.0004, "data_time": 0.023465752601623535, "loss": 0.5783761262893676, "time": 0.7354685544967652, "epoch": 1, "memory": 25071, "step": 100} +{"lr": 0.0004, "data_time": 0.0012981891632080078, "loss": 0.5477651119232178, "time": 0.7451019525527954, "epoch": 1, "memory": 25071, "step": 200} +{"lr": 0.0004, "data_time": 0.0011882543563842773, "loss": 0.5147063314914704, "time": 0.8193954229354858, "epoch": 1, "memory": 25071, "step": 300} +{"lr": 0.0004, "data_time": 0.0012914657592773438, "loss": 0.5042121767997741, "time": 0.8707980155944824, "epoch": 1, "memory": 25071, "step": 400} +{"lr": 0.0004, "data_time": 0.0014605045318603516, "loss": 0.46452565789222716, "time": 0.7508379697799683, "epoch": 1, "memory": 25071, "step": 500} +{"lr": 0.0004, "data_time": 0.0013184547424316406, "loss": 0.45228410363197324, "time": 0.7682911157608032, "epoch": 1, "memory": 25071, "step": 600} +{"lr": 0.0004, "data_time": 0.001284956932067871, "loss": 0.44432016015052794, "time": 0.7671506881713868, "epoch": 1, "memory": 25071, "step": 700} +{"lr": 0.0004, "data_time": 0.0012042999267578125, "loss": 0.45052569210529325, "time": 0.710477876663208, "epoch": 1, "memory": 25071, "step": 800} +{"lr": 0.0004, "data_time": 0.0011302709579467773, "loss": 0.40240114033222196, "time": 0.7120702743530274, "epoch": 1, "memory": 25071, "step": 900} +{"lr": 0.0004, "data_time": 0.0011893749237060548, "loss": 0.4003148555755615, "time": 0.7268030405044555, "epoch": 1, "memory": 25071, "step": 1000} +{"lr": 0.0004, "data_time": 0.001346588134765625, "loss": 0.4231216698884964, "time": 0.73503258228302, "epoch": 1, "memory": 25071, "step": 1100} +{"lr": 0.0004, "data_time": 0.0012927055358886719, "loss": 0.4057676702737808, "time": 0.7073668003082275, "epoch": 1, "memory": 25071, "step": 1200} +{"lr": 0.0004, "data_time": 0.0015211105346679688, "loss": 0.4136209189891815, "time": 0.7717820167541504, "epoch": 1, "memory": 25071, "step": 1300} +{"lr": 0.0004, "data_time": 0.0015099525451660156, "loss": 0.3821967035531998, "time": 0.7974962711334228, "epoch": 1, "memory": 25071, "step": 1400} +{"lr": 0.0004, "data_time": 0.0014603614807128906, "loss": 0.3885293424129486, "time": 0.7651231765747071, "epoch": 1, "memory": 25071, "step": 1500} +{"lr": 0.0004, "data_time": 0.0015938997268676758, "loss": 0.35765465497970583, "time": 0.7080837488174438, "epoch": 1, "memory": 25071, "step": 1600} +{"lr": 0.0004, "data_time": 0.0012611865997314454, "loss": 0.36647533178329467, "time": 0.7144793033599853, "epoch": 1, "memory": 25071, "step": 1700} +{"lr": 0.0004, "data_time": 0.001858091354370117, "loss": 0.3575074225664139, "time": 0.7647769451141357, "epoch": 1, "memory": 25071, "step": 1800} +{"lr": 0.0004, "data_time": 0.001350712776184082, "loss": 0.34662758111953734, "time": 0.7962718725204467, "epoch": 1, "memory": 25071, "step": 1900} +{"lr": 0.0004, "data_time": 0.0013519287109375, "loss": 0.37120504677295685, "time": 0.7751044273376465, "epoch": 1, "memory": 25071, "step": 2000} +{"lr": 0.0004, "data_time": 0.001309347152709961, "loss": 0.31365267783403394, "time": 0.759409499168396, "epoch": 1, "memory": 25071, "step": 2100} +{"lr": 0.0004, "data_time": 0.001217794418334961, "loss": 0.3412786304950714, "time": 0.7413238286972046, "epoch": 1, "memory": 25071, "step": 2200} +{"lr": 0.0004, "data_time": 0.0013586759567260742, "loss": 0.3255023032426834, "time": 0.7054993152618408, "epoch": 1, "memory": 25071, "step": 2300} +{"lr": 0.0004, "data_time": 0.0011933565139770508, "loss": 0.31905866116285325, "time": 0.8325782299041748, "epoch": 1, "memory": 25071, "step": 2400} +{"lr": 0.0004, "data_time": 0.0014669656753540038, "loss": 0.31494944393634794, "time": 0.7384720563888549, "epoch": 1, "memory": 25071, "step": 2500} +{"lr": 0.0004, "data_time": 0.0012603282928466796, "loss": 0.30391877591609956, "time": 0.7293519020080567, "epoch": 1, "memory": 25071, "step": 2600} +{"lr": 0.0004, "data_time": 0.0012967586517333984, "loss": 0.27619979083538054, "time": 0.7716137170791626, "epoch": 1, "memory": 25071, "step": 2700} +{"lr": 0.0004, "data_time": 0.0012556791305541992, "loss": 0.29391306936740874, "time": 0.7249278068542481, "epoch": 1, "memory": 25071, "step": 2800} +{"lr": 0.0004, "data_time": 0.0013440847396850586, "loss": 0.29973573088645933, "time": 0.8226109266281127, "epoch": 1, "memory": 25071, "step": 2900} +{"lr": 0.0004, "data_time": 0.0011576175689697265, "loss": 0.2650411456823349, "time": 0.7453691244125367, "epoch": 1, "memory": 25071, "step": 3000} +{"lr": 0.0004, "data_time": 0.0015404701232910156, "loss": 0.29347882568836214, "time": 0.7388896703720093, "epoch": 1, "memory": 25071, "step": 3100} +{"lr": 0.0004, "data_time": 0.0012520074844360352, "loss": 0.27183108031749725, "time": 0.710351824760437, "epoch": 1, "memory": 25071, "step": 3200} +{"lr": 0.0004, "data_time": 0.0012935638427734376, "loss": 0.2756800249218941, "time": 0.7611816167831421, "epoch": 1, "memory": 25071, "step": 3300} +{"lr": 0.0004, "data_time": 0.0012006044387817382, "loss": 0.2523490786552429, "time": 0.7746668577194213, "epoch": 1, "memory": 25071, "step": 3400} +{"lr": 0.0004, "data_time": 0.0012663602828979492, "loss": 0.2788736060261726, "time": 0.7459849834442138, "epoch": 1, "memory": 25071, "step": 3500} +{"lr": 0.0004, "data_time": 0.0014612436294555663, "loss": 0.2532954692840576, "time": 0.7365252733230591, "epoch": 1, "memory": 25071, "step": 3600} +{"lr": 0.0004, "data_time": 0.0015092849731445312, "loss": 0.2760276019573212, "time": 0.7436590194702148, "epoch": 1, "memory": 25071, "step": 3700} +{"lr": 0.0004, "data_time": 0.0013444185256958007, "loss": 0.24519322961568832, "time": 0.7767281055450439, "epoch": 1, "memory": 25071, "step": 3800} +{"lr": 0.0004, "data_time": 0.0013512611389160157, "loss": 0.24075794219970703, "time": 0.7031121253967285, "epoch": 1, "memory": 25071, "step": 3900} +{"lr": 0.0004, "data_time": 0.0014681577682495116, "loss": 0.2520707279443741, "time": 0.7432272434234619, "epoch": 1, "memory": 25071, "step": 4000} +{"lr": 0.0004, "data_time": 0.0013287782669067383, "loss": 0.24621658474206926, "time": 0.858167290687561, "epoch": 1, "memory": 25071, "step": 4100} +{"lr": 0.0004, "data_time": 0.0014040946960449218, "loss": 0.23313160091638566, "time": 0.7725039958953858, "epoch": 1, "memory": 25071, "step": 4200} +{"lr": 0.0004, "data_time": 0.0013962030410766602, "loss": 0.22922349125146865, "time": 0.8162384033203125, "epoch": 1, "memory": 25071, "step": 4300} +{"lr": 0.0004, "data_time": 0.0013297557830810546, "loss": 0.23687607645988465, "time": 0.754087257385254, "epoch": 1, "memory": 25071, "step": 4400} +{"lr": 0.0004, "data_time": 0.001346278190612793, "loss": 0.23668272644281388, "time": 0.7501036882400512, "epoch": 1, "memory": 25071, "step": 4500} +{"lr": 0.0004, "data_time": 0.0013803482055664063, "loss": 0.2544979602098465, "time": 0.7481486558914184, "epoch": 1, "memory": 25071, "step": 4600} +{"lr": 0.0004, "data_time": 0.0012869119644165039, "loss": 0.21583334356546402, "time": 0.7296990156173706, "epoch": 1, "memory": 25071, "step": 4700} +{"lr": 0.0004, "data_time": 0.001413273811340332, "loss": 0.2173559471964836, "time": 0.7453904151916504, "epoch": 1, "memory": 25071, "step": 4800} +{"lr": 0.0004, "data_time": 0.0012706995010375976, "loss": 0.22525127977132797, "time": 0.719694972038269, "epoch": 1, "memory": 25071, "step": 4900} +{"lr": 0.0004, "data_time": 0.001439213752746582, "loss": 0.22691788226366044, "time": 0.7515617609024048, "epoch": 1, "memory": 25071, "step": 5000} +{"lr": 0.0004, "data_time": 0.0014858007431030273, "loss": 0.19926073253154755, "time": 0.7507566690444947, "epoch": 1, "memory": 25071, "step": 5100} +{"lr": 0.0004, "data_time": 0.00108184814453125, "loss": 0.20982834398746492, "time": 0.7858760356903076, "epoch": 1, "memory": 25071, "step": 5200} +{"lr": 0.0004, "data_time": 0.0014892339706420899, "loss": 0.19950440973043443, "time": 0.7765029191970825, "epoch": 1, "memory": 25071, "step": 5300} +{"lr": 0.0004, "data_time": 0.0012821674346923829, "loss": 0.19394015818834304, "time": 0.7908424854278564, "epoch": 1, "memory": 25071, "step": 5400} +{"lr": 0.0004, "data_time": 0.0014799118041992187, "loss": 0.19626849442720412, "time": 0.8351553916931153, "epoch": 1, "memory": 25071, "step": 5500} +{"lr": 0.0004, "data_time": 0.001165318489074707, "loss": 0.20947481840848922, "time": 0.7716299772262574, "epoch": 1, "memory": 25071, "step": 5600} +{"lr": 0.0004, "data_time": 0.0012734174728393556, "loss": 0.1822401225566864, "time": 0.7181431293487549, "epoch": 1, "memory": 25071, "step": 5700} +{"accuracy/top1": 91.80606842041016, "data_time": 0.2736958712339401, "time": 0.7940136343240738, "step": 1} +{"lr": 0.00039759922641605187, "data_time": 0.056543922424316405, "loss": 0.178238408267498, "time": 0.7774782419204712, "epoch": 2, "memory": 25074, "step": 5858} +{"lr": 0.00039759922641605187, "data_time": 0.0014495372772216797, "loss": 0.17998147308826445, "time": 0.7055389881134033, "epoch": 2, "memory": 25074, "step": 5958} +{"lr": 0.00039759922641605187, "data_time": 0.0016408443450927734, "loss": 0.19104122668504714, "time": 0.7790418148040772, "epoch": 2, "memory": 25074, "step": 6058} +{"lr": 0.00039759922641605187, "data_time": 0.0013885498046875, "loss": 0.18300386071205138, "time": 0.8070295333862305, "epoch": 2, "memory": 25074, "step": 6158} +{"lr": 0.00039759922641605187, "data_time": 0.001452493667602539, "loss": 0.16158113032579421, "time": 0.7680969715118409, "epoch": 2, "memory": 25074, "step": 6258} +{"lr": 0.00039759922641605187, "data_time": 0.0013581037521362305, "loss": 0.1782495841383934, "time": 0.7868995189666748, "epoch": 2, "memory": 25074, "step": 6358} +{"lr": 0.00039759922641605187, "data_time": 0.0014699459075927734, "loss": 0.15414375811815262, "time": 0.7201312065124512, "epoch": 2, "memory": 25074, "step": 6458} +{"lr": 0.00039759922641605187, "data_time": 0.0016120672225952148, "loss": 0.2027489274740219, "time": 0.732387924194336, "epoch": 2, "memory": 25074, "step": 6558} +{"lr": 0.00039759922641605187, "data_time": 0.0012201547622680664, "loss": 0.17093847543001175, "time": 0.7485415458679199, "epoch": 2, "memory": 25074, "step": 6658} +{"lr": 0.00039759922641605187, "data_time": 0.0014460086822509766, "loss": 0.17120749801397322, "time": 0.7299433469772338, "epoch": 2, "memory": 25074, "step": 6758} +{"lr": 0.00039759922641605187, "data_time": 0.0021276235580444335, "loss": 0.1816856697201729, "time": 0.7633233308792114, "epoch": 2, "memory": 25074, "step": 6858} +{"lr": 0.00039759922641605187, "data_time": 0.0014348983764648437, "loss": 0.151786420494318, "time": 0.7627900600433349, "epoch": 2, "memory": 25074, "step": 6958} +{"lr": 0.00039759922641605187, "data_time": 0.0014765024185180663, "loss": 0.14380268454551698, "time": 0.8085668087005615, "epoch": 2, "memory": 25074, "step": 7058} +{"lr": 0.00039759922641605187, "data_time": 0.0010685682296752929, "loss": 0.1663478970527649, "time": 0.7716030836105346, "epoch": 2, "memory": 25074, "step": 7158} +{"lr": 0.00039759922641605187, "data_time": 0.0012255191802978515, "loss": 0.1608097583055496, "time": 0.7307199478149414, "epoch": 2, "memory": 25074, "step": 7258} +{"lr": 0.00039759922641605187, "data_time": 0.0013592004776000976, "loss": 0.16642693132162095, "time": 0.7474058628082275, "epoch": 2, "memory": 25074, "step": 7358} +{"lr": 0.00039759922641605187, "data_time": 0.001287078857421875, "loss": 0.20146722197532654, "time": 0.7900904655456543, "epoch": 2, "memory": 25074, "step": 7458} +{"lr": 0.00039759922641605187, "data_time": 0.0015846729278564454, "loss": 0.14182349368929864, "time": 0.8205183982849121, "epoch": 2, "memory": 25074, "step": 7558} +{"lr": 0.00039759922641605187, "data_time": 0.0014862298965454101, "loss": 0.1436505876481533, "time": 0.7209877490997314, "epoch": 2, "memory": 25074, "step": 7658} +{"lr": 0.00039759922641605187, "data_time": 0.001214313507080078, "loss": 0.1374882437288761, "time": 0.7201198101043701, "epoch": 2, "memory": 25074, "step": 7758} +{"lr": 0.00039759922641605187, "data_time": 0.0012128829956054687, "loss": 0.13091749101877212, "time": 0.7129969596862793, "epoch": 2, "memory": 25074, "step": 7858} +{"lr": 0.00039759922641605187, "data_time": 0.0016506671905517577, "loss": 0.15016361474990844, "time": 0.7774442911148072, "epoch": 2, "memory": 25074, "step": 7958} +{"lr": 0.00039759922641605187, "data_time": 0.001264667510986328, "loss": 0.11804806515574455, "time": 0.7509546518325806, "epoch": 2, "memory": 25074, "step": 8058} +{"lr": 0.00039759922641605187, "data_time": 0.0017354726791381837, "loss": 0.13144950568675995, "time": 0.7998786926269531, "epoch": 2, "memory": 25074, "step": 8158} +{"lr": 0.00039759922641605187, "data_time": 0.0013310432434082032, "loss": 0.13164022490382193, "time": 0.8064572811126709, "epoch": 2, "memory": 25074, "step": 8258} +{"lr": 0.00039759922641605187, "data_time": 0.0012734174728393556, "loss": 0.13403431996703147, "time": 0.805510687828064, "epoch": 2, "memory": 25074, "step": 8358} +{"lr": 0.00039759922641605187, "data_time": 0.0013266801834106445, "loss": 0.1468445010483265, "time": 0.7319570064544678, "epoch": 2, "memory": 25074, "step": 8458} +{"lr": 0.00039759922641605187, "data_time": 0.02174842357635498, "loss": 0.14606546387076377, "time": 0.7308906316757202, "epoch": 2, "memory": 25074, "step": 8558} +{"lr": 0.00039759922641605187, "data_time": 0.0012790203094482423, "loss": 0.13792258203029634, "time": 0.7167191982269288, "epoch": 2, "memory": 25074, "step": 8658} +{"lr": 0.00039759922641605187, "data_time": 0.001343703269958496, "loss": 0.12017014175653458, "time": 0.8110358238220214, "epoch": 2, "memory": 25074, "step": 8758} +{"lr": 0.00039759922641605187, "data_time": 0.001419210433959961, "loss": 0.1138884425163269, "time": 0.7775670289993286, "epoch": 2, "memory": 25074, "step": 8858} +{"lr": 0.00039759922641605187, "data_time": 0.001520252227783203, "loss": 0.11171320900321006, "time": 0.7573474407196045, "epoch": 2, "memory": 25074, "step": 8958} +{"lr": 0.00039759922641605187, "data_time": 0.0015263319015502929, "loss": 0.10042026489973069, "time": 0.8191678762435913, "epoch": 2, "memory": 25074, "step": 9058} +{"lr": 0.00039759922641605187, "data_time": 0.0011941909790039063, "loss": 0.10328052565455437, "time": 0.7963718175888062, "epoch": 2, "memory": 25074, "step": 9158} +{"lr": 0.00039759922641605187, "data_time": 0.0018949270248413085, "loss": 0.11000730395317078, "time": 0.7700808048248291, "epoch": 2, "memory": 25074, "step": 9258} +{"lr": 0.00039759922641605187, "data_time": 0.0014587879180908204, "loss": 0.11679540202021599, "time": 0.7279536008834839, "epoch": 2, "memory": 25074, "step": 9358} +{"lr": 0.00039759922641605187, "data_time": 0.0014569282531738282, "loss": 0.11313460543751716, "time": 0.7468536138534546, "epoch": 2, "memory": 25074, "step": 9458} +{"lr": 0.00039759922641605187, "data_time": 0.0012717247009277344, "loss": 0.1270844042301178, "time": 0.7640789031982422, "epoch": 2, "memory": 25074, "step": 9558} +{"lr": 0.00039759922641605187, "data_time": 0.001318812370300293, "loss": 0.10224962010979652, "time": 0.7197729825973511, "epoch": 2, "memory": 25074, "step": 9658} +{"lr": 0.00039759922641605187, "data_time": 0.0014826059341430664, "loss": 0.1157186646014452, "time": 0.7557222366333007, "epoch": 2, "memory": 25074, "step": 9758} +{"lr": 0.00039759922641605187, "data_time": 0.001317000389099121, "loss": 0.08724111206829548, "time": 0.7586980819702148, "epoch": 2, "memory": 25074, "step": 9858} +{"lr": 0.00039759922641605187, "data_time": 0.0013677358627319336, "loss": 0.11286000311374664, "time": 0.7446044921875, "epoch": 2, "memory": 25074, "step": 9958} +{"lr": 0.00039759922641605187, "data_time": 0.0014358758926391602, "loss": 0.09231222979724407, "time": 0.7773568153381347, "epoch": 2, "memory": 25074, "step": 10058} +{"lr": 0.00039759922641605187, "data_time": 0.0011890411376953125, "loss": 0.0991213247179985, "time": 0.7307825565338135, "epoch": 2, "memory": 25074, "step": 10158} +{"lr": 0.00039759922641605187, "data_time": 0.0013567209243774414, "loss": 0.09247121214866638, "time": 0.8336795330047607, "epoch": 2, "memory": 25074, "step": 10258} +{"lr": 0.00039759922641605187, "data_time": 0.0014424324035644531, "loss": 0.08989433161914348, "time": 0.7155681610107422, "epoch": 2, "memory": 25074, "step": 10358} +{"lr": 0.00039759922641605187, "data_time": 0.0013794898986816406, "loss": 0.0911391343921423, "time": 0.7531449317932128, "epoch": 2, "memory": 25074, "step": 10458} +{"lr": 0.00039759922641605187, "data_time": 0.0010863780975341798, "loss": 0.08242987990379333, "time": 0.7036821603775024, "epoch": 2, "memory": 25074, "step": 10558} +{"lr": 0.00039759922641605187, "data_time": 0.001341533660888672, "loss": 0.09365952722728252, "time": 0.7220810174942016, "epoch": 2, "memory": 25074, "step": 10658} +{"lr": 0.00039759922641605187, "data_time": 0.0013874292373657227, "loss": 0.08166426531970501, "time": 0.7704447507858276, "epoch": 2, "memory": 25074, "step": 10758} +{"lr": 0.00039759922641605187, "data_time": 0.0013012170791625976, "loss": 0.08841171264648437, "time": 0.7600651741027832, "epoch": 2, "memory": 25074, "step": 10858} +{"lr": 0.00039759922641605187, "data_time": 0.0013210058212280273, "loss": 0.0915576845407486, "time": 0.7458028316497802, "epoch": 2, "memory": 25074, "step": 10958} +{"lr": 0.00039759922641605187, "data_time": 0.00149383544921875, "loss": 0.08014451824128628, "time": 0.7738240957260132, "epoch": 2, "memory": 25074, "step": 11058} +{"lr": 0.00039759922641605187, "data_time": 0.0016082763671875, "loss": 0.08079681731760502, "time": 0.733919644355774, "epoch": 2, "memory": 25074, "step": 11158} +{"lr": 0.00039759922641605187, "data_time": 0.0012682199478149414, "loss": 0.07264478243887425, "time": 0.779146695137024, "epoch": 2, "memory": 25074, "step": 11258} +{"lr": 0.00039759922641605187, "data_time": 0.00124969482421875, "loss": 0.06492337491363287, "time": 0.7348495483398437, "epoch": 2, "memory": 25074, "step": 11358} +{"lr": 0.00039759922641605187, "data_time": 0.00177154541015625, "loss": 0.07624929696321488, "time": 0.7161550760269165, "epoch": 2, "memory": 25074, "step": 11458} +{"accuracy/top1": 98.90158081054688, "data_time": 0.27104833546806784, "time": 0.65962955530952, "step": 2} +{"lr": 0.00039045602067755495, "data_time": 0.0011338472366333007, "loss": 0.07454138360917568, "time": 0.715718674659729, "epoch": 3, "memory": 25074, "step": 11616} +{"lr": 0.00039045602067755495, "data_time": 0.0012253999710083007, "loss": 0.06838808674365282, "time": 0.7966464519500732, "epoch": 3, "memory": 25074, "step": 11716} +{"lr": 0.00039045602067755495, "data_time": 0.0017017126083374023, "loss": 0.08313975371420383, "time": 0.7046485662460327, "epoch": 3, "memory": 25074, "step": 11816} +{"lr": 0.00039045602067755495, "data_time": 0.0014360427856445312, "loss": 0.0759969051927328, "time": 0.7288727045059205, "epoch": 3, "memory": 25074, "step": 11916} +{"lr": 0.00039045602067755495, "data_time": 0.001629948616027832, "loss": 0.06696641705930233, "time": 0.7726234197616577, "epoch": 3, "memory": 25074, "step": 12016} +{"lr": 0.00039045602067755495, "data_time": 0.0011850833892822266, "loss": 0.06603091154247523, "time": 0.7357964515686035, "epoch": 3, "memory": 25074, "step": 12116} +{"lr": 0.00039045602067755495, "data_time": 0.0013424873352050782, "loss": 0.07045400068163872, "time": 0.7128629446029663, "epoch": 3, "memory": 25074, "step": 12216} +{"lr": 0.00039045602067755495, "data_time": 0.001307225227355957, "loss": 0.06513373404741288, "time": 0.7293326139450074, "epoch": 3, "memory": 25074, "step": 12316} +{"lr": 0.00039045602067755495, "data_time": 0.0014556407928466796, "loss": 0.07489310540258884, "time": 0.7053848743438721, "epoch": 3, "memory": 25074, "step": 12416} +{"lr": 0.00039045602067755495, "data_time": 0.0012600421905517578, "loss": 0.056421241350471975, "time": 0.7707218170166016, "epoch": 3, "memory": 25074, "step": 12516} +{"lr": 0.00039045602067755495, "data_time": 0.001521897315979004, "loss": 0.07037061303853989, "time": 0.729718804359436, "epoch": 3, "memory": 25074, "step": 12616} +{"lr": 0.00039045602067755495, "data_time": 0.001941537857055664, "loss": 0.07602909170091152, "time": 0.7406081914901733, "epoch": 3, "memory": 25074, "step": 12716} +{"lr": 0.00039045602067755495, "data_time": 0.0014235734939575194, "loss": 0.07726535201072693, "time": 0.7761637449264527, "epoch": 3, "memory": 25074, "step": 12816} +{"lr": 0.00039045602067755495, "data_time": 0.0014698266983032226, "loss": 0.06499647535383701, "time": 0.7467527866363526, "epoch": 3, "memory": 25074, "step": 12916} +{"lr": 0.00039045602067755495, "data_time": 0.0013027667999267578, "loss": 0.058831796050071716, "time": 0.717350435256958, "epoch": 3, "memory": 25074, "step": 13016} +{"lr": 0.00039045602067755495, "data_time": 0.0011768579483032227, "loss": 0.056864219903945926, "time": 0.7174227476119995, "epoch": 3, "memory": 25074, "step": 13116} +{"lr": 0.00039045602067755495, "data_time": 0.0013763904571533203, "loss": 0.056697160378098486, "time": 0.774661374092102, "epoch": 3, "memory": 25074, "step": 13216} +{"lr": 0.00039045602067755495, "data_time": 0.0010921478271484375, "loss": 0.05814656335860491, "time": 0.7548272848129273, "epoch": 3, "memory": 25074, "step": 13316} +{"lr": 0.00039045602067755495, "data_time": 0.001402592658996582, "loss": 0.0690337635576725, "time": 0.7593602657318115, "epoch": 3, "memory": 25074, "step": 13416} +{"lr": 0.00039045602067755495, "data_time": 0.0016754388809204102, "loss": 0.06053033582866192, "time": 0.776843523979187, "epoch": 3, "memory": 25074, "step": 13516} +{"lr": 0.00039045602067755495, "data_time": 0.001482534408569336, "loss": 0.05673934854567051, "time": 0.7125611305236816, "epoch": 3, "memory": 25074, "step": 13616} +{"lr": 0.00039045602067755495, "data_time": 0.0016509294509887695, "loss": 0.051472777128219606, "time": 0.7779937744140625, "epoch": 3, "memory": 25074, "step": 13716} +{"lr": 0.00039045602067755495, "data_time": 0.001448345184326172, "loss": 0.047471163980662825, "time": 0.774700665473938, "epoch": 3, "memory": 25074, "step": 13816} +{"lr": 0.00039045602067755495, "data_time": 0.0014891862869262696, "loss": 0.06760014072060586, "time": 0.7781532526016235, "epoch": 3, "memory": 25074, "step": 13916} +{"lr": 0.00039045602067755495, "data_time": 0.0015434503555297851, "loss": 0.05911597870290279, "time": 0.799963402748108, "epoch": 3, "memory": 25074, "step": 14016} +{"lr": 0.00039045602067755495, "data_time": 0.0014227867126464845, "loss": 0.05952814538031816, "time": 0.7754967451095581, "epoch": 3, "memory": 25074, "step": 14116} +{"lr": 0.00039045602067755495, "data_time": 0.0020697355270385743, "loss": 0.049110569432377814, "time": 0.7596653938293457, "epoch": 3, "memory": 25074, "step": 14216} +{"lr": 0.00039045602067755495, "data_time": 0.0010865688323974609, "loss": 0.07636752352118492, "time": 0.7480664014816284, "epoch": 3, "memory": 25074, "step": 14316} +{"lr": 0.00039045602067755495, "data_time": 0.0016338348388671875, "loss": 0.07510394230484962, "time": 0.7465034484863281, "epoch": 3, "memory": 25074, "step": 14416} +{"lr": 0.00039045602067755495, "data_time": 0.0017137527465820312, "loss": 0.055041104555130005, "time": 0.7643941402435303, "epoch": 3, "memory": 25074, "step": 14516} +{"lr": 0.00039045602067755495, "data_time": 0.0013677358627319336, "loss": 0.043075740337371826, "time": 0.7085863351821899, "epoch": 3, "memory": 25074, "step": 14616} +{"lr": 0.00039045602067755495, "data_time": 0.0019581794738769533, "loss": 0.04821114744991064, "time": 0.7650643348693847, "epoch": 3, "memory": 25074, "step": 14716} +{"lr": 0.00039045602067755495, "data_time": 0.0015894174575805664, "loss": 0.04504678547382355, "time": 0.7304926156997681, "epoch": 3, "memory": 25074, "step": 14816} +{"lr": 0.00039045602067755495, "data_time": 0.0013896465301513673, "loss": 0.05232441704720259, "time": 0.7621388673782349, "epoch": 3, "memory": 25074, "step": 14916} +{"lr": 0.00039045602067755495, "data_time": 0.001965498924255371, "loss": 0.04509794358164072, "time": 0.7406379461288453, "epoch": 3, "memory": 25074, "step": 15016} +{"lr": 0.00039045602067755495, "data_time": 0.0014034509658813477, "loss": 0.0478059820830822, "time": 0.7896098375320435, "epoch": 3, "memory": 25074, "step": 15116} +{"lr": 0.00039045602067755495, "data_time": 0.0014190196990966797, "loss": 0.044007493555545805, "time": 0.7913527488708496, "epoch": 3, "memory": 25074, "step": 15216} +{"lr": 0.00039045602067755495, "data_time": 0.0017563581466674804, "loss": 0.04758833684027195, "time": 0.8260623455047608, "epoch": 3, "memory": 25074, "step": 15316} +{"lr": 0.00039045602067755495, "data_time": 0.001270627975463867, "loss": 0.0439551105722785, "time": 0.7511175394058227, "epoch": 3, "memory": 25074, "step": 15416} +{"lr": 0.00039045602067755495, "data_time": 0.001655721664428711, "loss": 0.03452812507748604, "time": 0.731798243522644, "epoch": 3, "memory": 25074, "step": 15516} +{"lr": 0.00039045602067755495, "data_time": 0.0020056724548339843, "loss": 0.04098694976419211, "time": 0.779358172416687, "epoch": 3, "memory": 25074, "step": 15616} +{"lr": 0.00039045602067755495, "data_time": 0.001270151138305664, "loss": 0.04094848036766052, "time": 0.760709261894226, "epoch": 3, "memory": 25074, "step": 15716} +{"lr": 0.00039045602067755495, "data_time": 0.0014201641082763673, "loss": 0.048485515639185905, "time": 0.7232496023178101, "epoch": 3, "memory": 25074, "step": 15816} +{"lr": 0.00039045602067755495, "data_time": 0.001566171646118164, "loss": 0.04530941024422645, "time": 0.7389329433441162, "epoch": 3, "memory": 25074, "step": 15916} +{"lr": 0.00039045602067755495, "data_time": 0.001257467269897461, "loss": 0.03874981384724378, "time": 0.8389924764633179, "epoch": 3, "memory": 25074, "step": 16016} +{"lr": 0.00039045602067755495, "data_time": 0.0012781620025634766, "loss": 0.03951667342334986, "time": 0.7242990016937256, "epoch": 3, "memory": 25074, "step": 16116} +{"lr": 0.00039045602067755495, "data_time": 0.0015552520751953125, "loss": 0.050310324877500534, "time": 0.7852558374404908, "epoch": 3, "memory": 25074, "step": 16216} +{"lr": 0.00039045602067755495, "data_time": 0.0015773773193359375, "loss": 0.049104707688093184, "time": 0.7572763919830322, "epoch": 3, "memory": 25074, "step": 16316} +{"lr": 0.00039045602067755495, "data_time": 0.0013127803802490234, "loss": 0.040728192590177056, "time": 0.8331624269485474, "epoch": 3, "memory": 25074, "step": 16416} +{"lr": 0.00039045602067755495, "data_time": 0.0013446569442749023, "loss": 0.0441434271633625, "time": 0.7584138870239258, "epoch": 3, "memory": 25074, "step": 16516} +{"lr": 0.00039045602067755495, "data_time": 0.0014935016632080078, "loss": 0.03594810180366039, "time": 0.7814659118652344, "epoch": 3, "memory": 25074, "step": 16616} +{"lr": 0.00039045602067755495, "data_time": 0.0014584779739379883, "loss": 0.05678151026368141, "time": 0.742890739440918, "epoch": 3, "memory": 25074, "step": 16716} +{"lr": 0.00039045602067755495, "data_time": 0.0012686729431152343, "loss": 0.05335999634116888, "time": 0.8077174186706543, "epoch": 3, "memory": 25074, "step": 16816} +{"lr": 0.00039045602067755495, "data_time": 0.0013684511184692382, "loss": 0.045536359772086143, "time": 0.7622642040252685, "epoch": 3, "memory": 25074, "step": 16916} +{"lr": 0.00039045602067755495, "data_time": 0.0012946367263793946, "loss": 0.039878766797482965, "time": 0.735869288444519, "epoch": 3, "memory": 25074, "step": 17016} +{"lr": 0.00039045602067755495, "data_time": 0.0014983654022216798, "loss": 0.033761563571169974, "time": 0.7622643947601319, "epoch": 3, "memory": 25074, "step": 17116} +{"lr": 0.00039045602067755495, "data_time": 0.0013837099075317382, "loss": 0.036536894738674164, "time": 0.7115382432937623, "epoch": 3, "memory": 25074, "step": 17216} +{"accuracy/top1": 99.36241149902344, "data_time": 0.2469493080587948, "time": 0.6325257665970746, "step": 3} +{"lr": 0.0003787462722167317, "data_time": 0.03865253925323486, "loss": 0.039829705841839314, "time": 0.7850553512573242, "epoch": 4, "memory": 25074, "step": 17374} +{"lr": 0.0003787462722167317, "data_time": 0.05717322826385498, "loss": 0.0474745761603117, "time": 0.8118953227996826, "epoch": 4, "memory": 25074, "step": 17474} +{"lr": 0.0003787462722167317, "data_time": 0.0014620780944824218, "loss": 0.032372019998729226, "time": 0.7516115665435791, "epoch": 4, "memory": 25074, "step": 17574} +{"lr": 0.0003787462722167317, "data_time": 0.08182368278503419, "loss": 0.0328549487516284, "time": 0.7947424650192261, "epoch": 4, "memory": 25074, "step": 17674} +{"lr": 0.0003787462722167317, "data_time": 0.001350259780883789, "loss": 0.03159841159358621, "time": 0.7455472469329834, "epoch": 4, "memory": 25074, "step": 17774} +{"lr": 0.0003787462722167317, "data_time": 0.0016098976135253905, "loss": 0.03146227048709989, "time": 0.7151310682296753, "epoch": 4, "memory": 25074, "step": 17874} +{"lr": 0.0003787462722167317, "data_time": 0.0015421152114868165, "loss": 0.03269414212554693, "time": 0.7611994504928589, "epoch": 4, "memory": 25074, "step": 17974} +{"lr": 0.0003787462722167317, "data_time": 0.0012829303741455078, "loss": 0.030372516997158528, "time": 0.7356581211090087, "epoch": 4, "memory": 25074, "step": 18074} +{"lr": 0.0003787462722167317, "data_time": 0.0012136936187744141, "loss": 0.04282185938209295, "time": 0.7142583131790161, "epoch": 4, "memory": 25074, "step": 18174} +{"lr": 0.0003787462722167317, "data_time": 0.0013745784759521484, "loss": 0.0318029697984457, "time": 0.8543846607208252, "epoch": 4, "memory": 25074, "step": 18274} +{"lr": 0.0003787462722167317, "data_time": 0.0014328718185424804, "loss": 0.04707674775272608, "time": 0.7077360153198242, "epoch": 4, "memory": 25074, "step": 18374} +{"lr": 0.0003787462722167317, "data_time": 0.0014774799346923828, "loss": 0.03311942107975483, "time": 0.7490308523178101, "epoch": 4, "memory": 25074, "step": 18474} +{"lr": 0.0003787462722167317, "data_time": 0.001595759391784668, "loss": 0.02794723752886057, "time": 0.7808529376983643, "epoch": 4, "memory": 25074, "step": 18574} +{"lr": 0.0003787462722167317, "data_time": 0.001329803466796875, "loss": 0.025707329018041493, "time": 0.7930867671966553, "epoch": 4, "memory": 25074, "step": 18674} +{"lr": 0.0003787462722167317, "data_time": 0.0021139860153198244, "loss": 0.02396458266302943, "time": 0.8923398017883301, "epoch": 4, "memory": 25074, "step": 18774} +{"lr": 0.0003787462722167317, "data_time": 0.0012781143188476563, "loss": 0.019072595937177538, "time": 0.7380176544189453, "epoch": 4, "memory": 25074, "step": 18874} +{"lr": 0.0003787462722167317, "data_time": 0.002022719383239746, "loss": 0.028676498495042325, "time": 0.7897438526153564, "epoch": 4, "memory": 25074, "step": 18974} +{"lr": 0.0003787462722167317, "data_time": 0.0016179084777832031, "loss": 0.02655111365020275, "time": 0.7565790176391601, "epoch": 4, "memory": 25074, "step": 19074} +{"lr": 0.0003787462722167317, "data_time": 0.0013271093368530274, "loss": 0.02916183224879205, "time": 0.7686939239501953, "epoch": 4, "memory": 25074, "step": 19174} +{"lr": 0.0003787462722167317, "data_time": 0.0014940023422241211, "loss": 0.030577187798917292, "time": 0.7645267724990845, "epoch": 4, "memory": 25074, "step": 19274} +{"lr": 0.0003787462722167317, "data_time": 0.00135495662689209, "loss": 0.029290075413882732, "time": 0.8134437322616577, "epoch": 4, "memory": 25074, "step": 19374} +{"lr": 0.0003787462722167317, "data_time": 0.0014248132705688477, "loss": 0.029681845940649508, "time": 0.7550102472305298, "epoch": 4, "memory": 25074, "step": 19474} +{"lr": 0.0003787462722167317, "data_time": 0.0013571977615356445, "loss": 0.03532544211484492, "time": 0.7044047355651856, "epoch": 4, "memory": 25074, "step": 19574} +{"lr": 0.0003787462722167317, "data_time": 0.001385045051574707, "loss": 0.02183720823377371, "time": 0.7184542417526245, "epoch": 4, "memory": 25074, "step": 19674} +{"lr": 0.0003787462722167317, "data_time": 0.0016919612884521485, "loss": 0.034702480956912043, "time": 0.723512840270996, "epoch": 4, "memory": 25074, "step": 19774} +{"lr": 0.0003787462722167317, "data_time": 0.001338648796081543, "loss": 0.03200682578608394, "time": 0.7607735872268677, "epoch": 4, "memory": 25074, "step": 19874} +{"lr": 0.0003787462722167317, "data_time": 0.0015716075897216797, "loss": 0.028486664686352015, "time": 0.7250436305999756, "epoch": 4, "memory": 25074, "step": 19974} +{"lr": 0.0003787462722167317, "data_time": 0.0015590429306030274, "loss": 0.017885612230747937, "time": 0.7812924385070801, "epoch": 4, "memory": 25074, "step": 20074} +{"lr": 0.0003787462722167317, "data_time": 0.0012953758239746093, "loss": 0.018333968101069333, "time": 0.7683312892913818, "epoch": 4, "memory": 25074, "step": 20174} +{"lr": 0.0003787462722167317, "data_time": 0.0013596534729003907, "loss": 0.024994514510035516, "time": 0.8128173589706421, "epoch": 4, "memory": 25074, "step": 20274} +{"lr": 0.0003787462722167317, "data_time": 0.0014835119247436524, "loss": 0.025219791987910866, "time": 0.7945241689682007, "epoch": 4, "memory": 25074, "step": 20374} +{"lr": 0.0003787462722167317, "data_time": 0.0013244867324829102, "loss": 0.02037292649038136, "time": 0.7190348148345947, "epoch": 4, "memory": 25074, "step": 20474} +{"lr": 0.0003787462722167317, "data_time": 0.001303553581237793, "loss": 0.024501336086541415, "time": 0.8481273174285888, "epoch": 4, "memory": 25074, "step": 20574} +{"lr": 0.0003787462722167317, "data_time": 0.001217031478881836, "loss": 0.03265264169313013, "time": 0.8404972791671753, "epoch": 4, "memory": 25074, "step": 20674} +{"lr": 0.0003787462722167317, "data_time": 0.0015292167663574219, "loss": 0.02086063758470118, "time": 0.8199644804000854, "epoch": 4, "memory": 25074, "step": 20774} +{"lr": 0.0003787462722167317, "data_time": 0.0013418912887573243, "loss": 0.015881731593981385, "time": 0.7641673564910889, "epoch": 4, "memory": 25074, "step": 20874} +{"lr": 0.0003787462722167317, "data_time": 0.0012826919555664062, "loss": 0.02060639802366495, "time": 0.7679383039474488, "epoch": 4, "memory": 25074, "step": 20974} +{"lr": 0.0003787462722167317, "data_time": 0.0013321876525878907, "loss": 0.017834770819172265, "time": 0.7576194524765014, "epoch": 4, "memory": 25074, "step": 21074} +{"lr": 0.0003787462722167317, "data_time": 0.0013240814208984376, "loss": 0.027906512189656497, "time": 0.7413241147994996, "epoch": 4, "memory": 25074, "step": 21174} +{"lr": 0.0003787462722167317, "data_time": 0.001589059829711914, "loss": 0.01876042317599058, "time": 0.811037015914917, "epoch": 4, "memory": 25074, "step": 21274} +{"lr": 0.0003787462722167317, "data_time": 0.0024838924407958986, "loss": 0.018446039967238904, "time": 0.8231079816818238, "epoch": 4, "memory": 25074, "step": 21374} +{"lr": 0.0003787462722167317, "data_time": 0.001452183723449707, "loss": 0.02220039302483201, "time": 0.7696118116378784, "epoch": 4, "memory": 25074, "step": 21474} +{"lr": 0.0003787462722167317, "data_time": 0.001588726043701172, "loss": 0.028869006037712096, "time": 0.8068956136703491, "epoch": 4, "memory": 25074, "step": 21574} +{"lr": 0.0003787462722167317, "data_time": 0.001605987548828125, "loss": 0.02095076614059508, "time": 0.75597984790802, "epoch": 4, "memory": 25074, "step": 21674} +{"lr": 0.0003787462722167317, "data_time": 0.001576685905456543, "loss": 0.016870320495218038, "time": 0.7346962928771973, "epoch": 4, "memory": 25074, "step": 21774} +{"lr": 0.0003787462722167317, "data_time": 0.0014455080032348632, "loss": 0.015548450499773025, "time": 0.7052653551101684, "epoch": 4, "memory": 25074, "step": 21874} +{"lr": 0.0003787462722167317, "data_time": 0.0014158725738525391, "loss": 0.019401692179962994, "time": 0.7833166599273682, "epoch": 4, "memory": 25074, "step": 21974} +{"lr": 0.0003787462722167317, "data_time": 0.001394367218017578, "loss": 0.0162011640612036, "time": 0.8173018217086792, "epoch": 4, "memory": 25074, "step": 22074} +{"lr": 0.0003787462722167317, "data_time": 0.001615428924560547, "loss": 0.013591721281409263, "time": 0.758896803855896, "epoch": 4, "memory": 25074, "step": 22174} +{"lr": 0.0003787462722167317, "data_time": 0.0013357639312744141, "loss": 0.01157363336533308, "time": 0.811647915840149, "epoch": 4, "memory": 25074, "step": 22274} +{"lr": 0.0003787462722167317, "data_time": 0.0014384984970092773, "loss": 0.02888364139944315, "time": 0.7266649007797241, "epoch": 4, "memory": 25074, "step": 22374} +{"lr": 0.0003787462722167317, "data_time": 0.0012154817581176759, "loss": 0.025532242748886345, "time": 0.7530094861984253, "epoch": 4, "memory": 25074, "step": 22474} +{"lr": 0.0003787462722167317, "data_time": 0.0015397071838378906, "loss": 0.01601006994023919, "time": 0.7512917518615723, "epoch": 4, "memory": 25074, "step": 22574} +{"lr": 0.0003787462722167317, "data_time": 0.0015405654907226563, "loss": 0.017409794451668858, "time": 0.8407786607742309, "epoch": 4, "memory": 25074, "step": 22674} +{"lr": 0.0003787462722167317, "data_time": 0.0013744115829467773, "loss": 0.013993551721796393, "time": 0.7365664720535279, "epoch": 4, "memory": 25074, "step": 22774} +{"lr": 0.0003787462722167317, "data_time": 0.0016142606735229492, "loss": 0.016717789648100733, "time": 0.7249078512191772, "epoch": 4, "memory": 25074, "step": 22874} +{"lr": 0.0003787462722167317, "data_time": 0.0013981103897094727, "loss": 0.02044197516515851, "time": 0.7902284145355225, "epoch": 4, "memory": 25074, "step": 22974} +{"accuracy/top1": 99.2677230834961, "data_time": 0.25519415911506205, "time": 0.6416875614839441, "step": 4} +{"lr": 0.00036275831390311474, "data_time": 0.001486825942993164, "loss": 0.01302544167265296, "time": 0.7256197690963745, "epoch": 5, "memory": 25074, "step": 23132} +{"lr": 0.00036275831390311474, "data_time": 0.030049228668212892, "loss": 0.017989942478016017, "time": 0.7324834346771241, "epoch": 5, "memory": 25074, "step": 23232} +{"lr": 0.00036275831390311474, "data_time": 0.0012367963790893555, "loss": 0.02218693825416267, "time": 0.7051385164260864, "epoch": 5, "memory": 25074, "step": 23332} +{"lr": 0.00036275831390311474, "data_time": 0.0015694379806518554, "loss": 0.012043322250247002, "time": 0.7431371688842774, "epoch": 5, "memory": 25074, "step": 23432} +{"lr": 0.00036275831390311474, "data_time": 0.0012540340423583985, "loss": 0.01390527212060988, "time": 0.8130687713623047, "epoch": 5, "memory": 25074, "step": 23532} +{"lr": 0.00036275831390311474, "data_time": 0.0015015363693237304, "loss": 0.018388553336262704, "time": 0.7184257507324219, "epoch": 5, "memory": 25074, "step": 23632} +{"lr": 0.00036275831390311474, "data_time": 0.0011070966720581055, "loss": 0.020958791440352797, "time": 0.7739711046218872, "epoch": 5, "memory": 25074, "step": 23732} +{"lr": 0.00036275831390311474, "data_time": 0.0013579845428466797, "loss": 0.019040547451004385, "time": 0.7628731012344361, "epoch": 5, "memory": 25074, "step": 23832} +{"lr": 0.00036275831390311474, "data_time": 0.0012732505798339843, "loss": 0.01234361007809639, "time": 0.8346306324005127, "epoch": 5, "memory": 25074, "step": 23932} +{"lr": 0.00036275831390311474, "data_time": 0.0015250205993652343, "loss": 0.015864039305597544, "time": 0.8216228723526001, "epoch": 5, "memory": 25074, "step": 24032} +{"lr": 0.00036275831390311474, "data_time": 0.0015598535537719727, "loss": 0.011547359358519315, "time": 0.7645201683044434, "epoch": 5, "memory": 25074, "step": 24132} +{"lr": 0.00036275831390311474, "data_time": 0.0014086246490478515, "loss": 0.026875092508271338, "time": 0.7816853761672974, "epoch": 5, "memory": 25074, "step": 24232} +{"lr": 0.00036275831390311474, "data_time": 0.0015123605728149414, "loss": 0.016534141125157477, "time": 0.7818456411361694, "epoch": 5, "memory": 25074, "step": 24332} +{"lr": 0.00036275831390311474, "data_time": 0.001155686378479004, "loss": 0.013718874007463456, "time": 0.7890047073364258, "epoch": 5, "memory": 25074, "step": 24432} +{"lr": 0.00036275831390311474, "data_time": 0.001723313331604004, "loss": 0.02003337643109262, "time": 0.7882824182510376, "epoch": 5, "memory": 25074, "step": 24532} +{"lr": 0.00036275831390311474, "data_time": 0.0013726234436035156, "loss": 0.028694703336805107, "time": 0.7442945003509521, "epoch": 5, "memory": 25074, "step": 24632} +{"lr": 0.00036275831390311474, "data_time": 0.0013441324234008789, "loss": 0.016452176216989755, "time": 0.7988262414932251, "epoch": 5, "memory": 25074, "step": 24732} +{"lr": 0.00036275831390311474, "data_time": 0.0014655351638793944, "loss": 0.012544103804975748, "time": 0.8010499000549316, "epoch": 5, "memory": 25074, "step": 24832} +{"lr": 0.00036275831390311474, "data_time": 0.001410841941833496, "loss": 0.017276473995298147, "time": 0.7862449407577514, "epoch": 5, "memory": 25074, "step": 24932} +{"lr": 0.00036275831390311474, "data_time": 0.0013671875, "loss": 0.011573469080030919, "time": 0.7392837285995484, "epoch": 5, "memory": 25074, "step": 25032} +{"lr": 0.00036275831390311474, "data_time": 0.07601547241210938, "loss": 0.014642014773562551, "time": 0.7981736898422241, "epoch": 5, "memory": 25074, "step": 25132} +{"lr": 0.00036275831390311474, "data_time": 0.001341557502746582, "loss": 0.014687849627807736, "time": 0.7996220111846923, "epoch": 5, "memory": 25074, "step": 25232} +{"lr": 0.00036275831390311474, "data_time": 0.0012368202209472657, "loss": 0.01322974874638021, "time": 0.7734551429748535, "epoch": 5, "memory": 25074, "step": 25332} +{"lr": 0.00036275831390311474, "data_time": 0.057136940956115725, "loss": 0.01116472678259015, "time": 0.8140712976455688, "epoch": 5, "memory": 25074, "step": 25432} +{"lr": 0.00036275831390311474, "data_time": 0.00757746696472168, "loss": 0.013869006372988223, "time": 0.7679396152496338, "epoch": 5, "memory": 25074, "step": 25532} +{"lr": 0.00036275831390311474, "data_time": 0.07710325717926025, "loss": 0.015432554762810469, "time": 0.7800628423690796, "epoch": 5, "memory": 25074, "step": 25632} +{"lr": 0.00036275831390311474, "data_time": 0.09012775421142578, "loss": 0.021824515145272016, "time": 0.8040318012237548, "epoch": 5, "memory": 25074, "step": 25732} +{"lr": 0.00036275831390311474, "data_time": 0.033565640449523926, "loss": 0.010669211437925697, "time": 0.7368215560913086, "epoch": 5, "memory": 25074, "step": 25832} +{"lr": 0.00036275831390311474, "data_time": 0.08241913318634034, "loss": 0.013968339376151561, "time": 0.7856771230697632, "epoch": 5, "memory": 25074, "step": 25932} +{"lr": 0.00036275831390311474, "data_time": 0.09301645755767822, "loss": 0.019537716545164584, "time": 0.7979531526565552, "epoch": 5, "memory": 25074, "step": 26032} +{"lr": 0.00036275831390311474, "data_time": 0.045124602317810056, "loss": 0.019960994319990277, "time": 0.7569398641586303, "epoch": 5, "memory": 25074, "step": 26132} +{"lr": 0.00036275831390311474, "data_time": 0.001261758804321289, "loss": 0.0166315873619169, "time": 0.7980321645736694, "epoch": 5, "memory": 25074, "step": 26232} +{"lr": 0.00036275831390311474, "data_time": 0.05023837089538574, "loss": 0.018772314256057143, "time": 0.7643418788909913, "epoch": 5, "memory": 25074, "step": 26332} +{"lr": 0.00036275831390311474, "data_time": 0.08424079418182373, "loss": 0.0145819500554353, "time": 0.7906779527664185, "epoch": 5, "memory": 25074, "step": 26432} +{"lr": 0.00036275831390311474, "data_time": 0.0105849027633667, "loss": 0.01371259642764926, "time": 0.7200708627700806, "epoch": 5, "memory": 25074, "step": 26532} +{"lr": 0.00036275831390311474, "data_time": 0.001291799545288086, "loss": 0.009983195271342993, "time": 0.7610705852508545, "epoch": 5, "memory": 25074, "step": 26632} +{"lr": 0.00036275831390311474, "data_time": 0.03467512130737305, "loss": 0.01607303861528635, "time": 0.7847788810729981, "epoch": 5, "memory": 25074, "step": 26732} +{"lr": 0.00036275831390311474, "data_time": 0.0016903877258300781, "loss": 0.01597420508041978, "time": 0.7835286378860473, "epoch": 5, "memory": 25074, "step": 26832} +{"lr": 0.00036275831390311474, "data_time": 0.0014004230499267579, "loss": 0.017677761614322662, "time": 0.7884481906890869, "epoch": 5, "memory": 25074, "step": 26932} +{"lr": 0.00036275831390311474, "data_time": 0.0013092994689941407, "loss": 0.012730128457769751, "time": 0.763460350036621, "epoch": 5, "memory": 25074, "step": 27032} +{"lr": 0.00036275831390311474, "data_time": 0.0012837886810302735, "loss": 0.010222951741889119, "time": 0.8114837169647217, "epoch": 5, "memory": 25074, "step": 27132} +{"lr": 0.00036275831390311474, "data_time": 0.0012880086898803711, "loss": 0.011691733077168465, "time": 0.8062796115875244, "epoch": 5, "memory": 25074, "step": 27232} +{"lr": 0.00036275831390311474, "data_time": 0.0011825799942016602, "loss": 0.01129258768633008, "time": 0.7690437078475952, "epoch": 5, "memory": 25074, "step": 27332} +{"lr": 0.00036275831390311474, "data_time": 0.0012610912322998046, "loss": 0.011570308823138475, "time": 0.8083793640136718, "epoch": 5, "memory": 25074, "step": 27432} +{"lr": 0.00036275831390311474, "data_time": 0.001442551612854004, "loss": 0.011827259371057153, "time": 0.7783132553100586, "epoch": 5, "memory": 25074, "step": 27532} +{"lr": 0.00036275831390311474, "data_time": 0.0014040946960449218, "loss": 0.013755986001342535, "time": 0.7516528844833374, "epoch": 5, "memory": 25074, "step": 27632} +{"lr": 0.00036275831390311474, "data_time": 0.003322935104370117, "loss": 0.007357860030606389, "time": 0.8465522289276123, "epoch": 5, "memory": 25074, "step": 27732} +{"lr": 0.00036275831390311474, "data_time": 0.0014224290847778321, "loss": 0.01856976402923465, "time": 0.8229195356369019, "epoch": 5, "memory": 25074, "step": 27832} +{"lr": 0.00036275831390311474, "data_time": 0.0012489557266235352, "loss": 0.008409297186881303, "time": 0.8426750421524047, "epoch": 5, "memory": 25074, "step": 27932} +{"lr": 0.00036275831390311474, "data_time": 0.0014356851577758789, "loss": 0.009741560509428381, "time": 0.7991224527359009, "epoch": 5, "memory": 25074, "step": 28032} +{"lr": 0.00036275831390311474, "data_time": 0.0013633966445922852, "loss": 0.010366294672712684, "time": 0.8169411897659302, "epoch": 5, "memory": 25074, "step": 28132} +{"lr": 0.00036275831390311474, "data_time": 0.0012496471405029296, "loss": 0.009001431753858924, "time": 0.81349618434906, "epoch": 5, "memory": 25074, "step": 28232} +{"lr": 0.00036275831390311474, "data_time": 0.0013092756271362305, "loss": 0.021743905264884233, "time": 0.8370378494262696, "epoch": 5, "memory": 25074, "step": 28332} +{"lr": 0.00036275831390311474, "data_time": 0.0013876914978027343, "loss": 0.01237932126969099, "time": 0.7648314237594604, "epoch": 5, "memory": 25074, "step": 28432} +{"lr": 0.00036275831390311474, "data_time": 0.0015105724334716797, "loss": 0.01142486920580268, "time": 0.8360764026641846, "epoch": 5, "memory": 25074, "step": 28532} +{"lr": 0.00036275831390311474, "data_time": 0.001252460479736328, "loss": 0.007908255234360695, "time": 0.8074623107910156, "epoch": 5, "memory": 25074, "step": 28632} +{"lr": 0.00036275831390311474, "data_time": 0.0013581275939941405, "loss": 0.018199953250586987, "time": 0.803422212600708, "epoch": 5, "memory": 25074, "step": 28732} +{"accuracy/top1": 99.44447326660156, "data_time": 0.2366017874549417, "time": 0.622307160321404, "step": 5} +{"lr": 0.00034288582233137673, "data_time": 0.0012431144714355469, "loss": 0.01467427983880043, "time": 0.8483107328414917, "epoch": 6, "memory": 25074, "step": 28890} +{"lr": 0.00034288582233137673, "data_time": 0.0013927221298217773, "loss": 0.013072089897468686, "time": 0.7660422325134277, "epoch": 6, "memory": 25074, "step": 28990} +{"lr": 0.00034288582233137673, "data_time": 0.0014702320098876954, "loss": 0.006401233281940222, "time": 0.7825122356414795, "epoch": 6, "memory": 25074, "step": 29090} +{"lr": 0.00034288582233137673, "data_time": 0.001685357093811035, "loss": 0.00809799530543387, "time": 0.7979695558547973, "epoch": 6, "memory": 25074, "step": 29190} +{"lr": 0.00034288582233137673, "data_time": 0.0014662742614746094, "loss": 0.013530059484764934, "time": 0.8616000890731812, "epoch": 6, "memory": 25074, "step": 29290} +{"lr": 0.00034288582233137673, "data_time": 0.0015248775482177735, "loss": 0.01676182746887207, "time": 0.7762357234954834, "epoch": 6, "memory": 25074, "step": 29390} +{"lr": 0.00034288582233137673, "data_time": 0.0013289928436279296, "loss": 0.014427924575284124, "time": 0.7673114776611328, "epoch": 6, "memory": 25074, "step": 29490} +{"lr": 0.00034288582233137673, "data_time": 0.001611328125, "loss": 0.011207863595336676, "time": 1.3448450803756713, "epoch": 6, "memory": 25074, "step": 29590} +{"lr": 0.00034288582233137673, "data_time": 0.0012563467025756836, "loss": 0.013308819988742471, "time": 0.7337980508804322, "epoch": 6, "memory": 25074, "step": 29690} +{"lr": 0.00034288582233137673, "data_time": 0.001601409912109375, "loss": 0.008573074685409665, "time": 0.7987351894378663, "epoch": 6, "memory": 25074, "step": 29790} +{"lr": 0.00034288582233137673, "data_time": 0.0014267444610595703, "loss": 0.010339982621371747, "time": 0.7991379022598266, "epoch": 6, "memory": 25074, "step": 29890} +{"lr": 0.00034288582233137673, "data_time": 0.0014389276504516602, "loss": 0.006984920892864466, "time": 0.7865674257278442, "epoch": 6, "memory": 25074, "step": 29990} +{"lr": 0.00034288582233137673, "data_time": 0.052128958702087405, "loss": 0.013638529181480407, "time": 0.7602356195449829, "epoch": 6, "memory": 25074, "step": 30090} +{"lr": 0.00034288582233137673, "data_time": 0.059166574478149415, "loss": 0.008216159651055932, "time": 0.7738542079925537, "epoch": 6, "memory": 25074, "step": 30190} +{"lr": 0.00034288582233137673, "data_time": 0.001326298713684082, "loss": 0.008545483695343137, "time": 0.7616467952728272, "epoch": 6, "memory": 25074, "step": 30290} +{"lr": 0.00034288582233137673, "data_time": 0.0013590335845947265, "loss": 0.00730477231554687, "time": 0.7734612703323365, "epoch": 6, "memory": 25074, "step": 30390} +{"lr": 0.00034288582233137673, "data_time": 0.00142977237701416, "loss": 0.008598646288737655, "time": 0.8029321432113647, "epoch": 6, "memory": 25074, "step": 30490} +{"lr": 0.00034288582233137673, "data_time": 0.0014775753021240234, "loss": 0.01053528799675405, "time": 0.8208680391311646, "epoch": 6, "memory": 25074, "step": 30590} +{"lr": 0.00034288582233137673, "data_time": 0.0016880273818969727, "loss": 0.005710984487086535, "time": 0.8045543432235718, "epoch": 6, "memory": 25074, "step": 30690} +{"lr": 0.00034288582233137673, "data_time": 0.001369786262512207, "loss": 0.010369472578167916, "time": 0.7910224437713623, "epoch": 6, "memory": 25074, "step": 30790} +{"lr": 0.00034288582233137673, "data_time": 0.0013910055160522461, "loss": 0.012486135913059115, "time": 0.7808441877365112, "epoch": 6, "memory": 25074, "step": 30890} +{"lr": 0.00034288582233137673, "data_time": 0.0015449047088623047, "loss": 0.009330012183636428, "time": 0.8098372220993042, "epoch": 6, "memory": 25074, "step": 30990} +{"lr": 0.00034288582233137673, "data_time": 0.0015043497085571289, "loss": 0.009312838781625032, "time": 0.7602557420730591, "epoch": 6, "memory": 25074, "step": 31090} +{"lr": 0.00034288582233137673, "data_time": 0.0013005495071411132, "loss": 0.00718923076055944, "time": 0.7536764860153198, "epoch": 6, "memory": 25074, "step": 31190} +{"lr": 0.00034288582233137673, "data_time": 0.0012532949447631836, "loss": 0.011695928080007434, "time": 0.776336407661438, "epoch": 6, "memory": 25074, "step": 31290} +{"lr": 0.00034288582233137673, "data_time": 0.0011270284652709962, "loss": 0.008433237066492438, "time": 0.7822027921676635, "epoch": 6, "memory": 25074, "step": 31390} +{"lr": 0.00034288582233137673, "data_time": 0.0012879371643066406, "loss": 0.011809056997299195, "time": 0.7624987840652466, "epoch": 6, "memory": 25074, "step": 31490} +{"lr": 0.00034288582233137673, "data_time": 0.001293039321899414, "loss": 0.008680937206372618, "time": 0.8256975173950195, "epoch": 6, "memory": 25074, "step": 31590} +{"lr": 0.00034288582233137673, "data_time": 0.0011420726776123046, "loss": 0.007531129196286202, "time": 0.8245503425598144, "epoch": 6, "memory": 25074, "step": 31690} +{"lr": 0.00034288582233137673, "data_time": 0.0014070034027099609, "loss": 0.008921292377635837, "time": 0.8015362977981567, "epoch": 6, "memory": 25074, "step": 31790} +{"lr": 0.00034288582233137673, "data_time": 0.001501154899597168, "loss": 0.008194751432165503, "time": 0.7953985929489136, "epoch": 6, "memory": 25074, "step": 31890} +{"lr": 0.00034288582233137673, "data_time": 0.0012296915054321289, "loss": 0.007451539672911167, "time": 0.754914116859436, "epoch": 6, "memory": 25074, "step": 31990} +{"lr": 0.00034288582233137673, "data_time": 0.001436305046081543, "loss": 0.005120963882654905, "time": 0.8026779651641845, "epoch": 6, "memory": 25074, "step": 32090} +{"lr": 0.00034288582233137673, "data_time": 0.0012905120849609375, "loss": 0.01344825685955584, "time": 0.8168651342391968, "epoch": 6, "memory": 25074, "step": 32190} +{"lr": 0.00034288582233137673, "data_time": 0.0013339757919311524, "loss": 0.009167814254760742, "time": 0.8336549282073975, "epoch": 6, "memory": 25074, "step": 32290} +{"lr": 0.00034288582233137673, "data_time": 0.0016960382461547851, "loss": 0.010822345362976193, "time": 0.7926252603530883, "epoch": 6, "memory": 25074, "step": 32390} +{"lr": 0.00034288582233137673, "data_time": 0.001642751693725586, "loss": 0.006898094387724996, "time": 0.7743218898773193, "epoch": 6, "memory": 25074, "step": 32490} +{"lr": 0.00034288582233137673, "data_time": 0.0012653112411499024, "loss": 0.008719425462186337, "time": 0.7710724830627441, "epoch": 6, "memory": 25074, "step": 32590} +{"lr": 0.00034288582233137673, "data_time": 0.0011759519577026368, "loss": 0.00417350516654551, "time": 0.7580895185470581, "epoch": 6, "memory": 25074, "step": 32690} +{"lr": 0.00034288582233137673, "data_time": 0.0012284517288208008, "loss": 0.014815912302583455, "time": 0.7413803100585937, "epoch": 6, "memory": 25074, "step": 32790} +{"lr": 0.00034288582233137673, "data_time": 0.0014820098876953125, "loss": 0.010735587356612087, "time": 0.7420071363449097, "epoch": 6, "memory": 25074, "step": 32890} +{"lr": 0.00034288582233137673, "data_time": 0.0014560222625732422, "loss": 0.004798201657831669, "time": 0.8103760004043579, "epoch": 6, "memory": 25074, "step": 32990} +{"lr": 0.00034288582233137673, "data_time": 0.001189279556274414, "loss": 0.013060365384444595, "time": 0.7591584682464599, "epoch": 6, "memory": 25074, "step": 33090} +{"lr": 0.00034288582233137673, "data_time": 0.0012809276580810548, "loss": 0.008555329404771328, "time": 0.7963660717010498, "epoch": 6, "memory": 25074, "step": 33190} +{"lr": 0.00034288582233137673, "data_time": 0.05887458324432373, "loss": 0.013045204104855657, "time": 0.7747400522232055, "epoch": 6, "memory": 25074, "step": 33290} +{"lr": 0.00034288582233137673, "data_time": 0.0030554771423339845, "loss": 0.012456198781728744, "time": 0.7957233428955078, "epoch": 6, "memory": 25074, "step": 33390} +{"lr": 0.00034288582233137673, "data_time": 0.07650580406188964, "loss": 0.006107565620914102, "time": 0.780707859992981, "epoch": 6, "memory": 25074, "step": 33490} +{"lr": 0.00034288582233137673, "data_time": 0.055533742904663085, "loss": 0.010568537842482328, "time": 0.8367711067199707, "epoch": 6, "memory": 25074, "step": 33590} +{"lr": 0.00034288582233137673, "data_time": 0.0018268585205078124, "loss": 0.009524564957246184, "time": 0.7714697599411011, "epoch": 6, "memory": 25074, "step": 33690} +{"lr": 0.00034288582233137673, "data_time": 0.026483893394470215, "loss": 0.017681375658139587, "time": 0.8249334335327149, "epoch": 6, "memory": 25074, "step": 33790} +{"lr": 0.00034288582233137673, "data_time": 0.05687577724456787, "loss": 0.005821938579902053, "time": 0.7611339092254639, "epoch": 6, "memory": 25074, "step": 33890} +{"lr": 0.00034288582233137673, "data_time": 0.12212176322937011, "loss": 0.008629760798066854, "time": 0.8271709680557251, "epoch": 6, "memory": 25074, "step": 33990} +{"lr": 0.00034288582233137673, "data_time": 0.04865329265594483, "loss": 0.00526374327018857, "time": 0.7964529037475586, "epoch": 6, "memory": 25074, "step": 34090} +{"lr": 0.00034288582233137673, "data_time": 0.11214594841003418, "loss": 0.012475757859647274, "time": 0.8364870309829712, "epoch": 6, "memory": 25074, "step": 34190} +{"lr": 0.00034288582233137673, "data_time": 0.001747274398803711, "loss": 0.012195754889398813, "time": 0.7567744016647339, "epoch": 6, "memory": 25074, "step": 34290} +{"lr": 0.00034288582233137673, "data_time": 0.0012288808822631836, "loss": 0.010368771757930518, "time": 0.8497467041015625, "epoch": 6, "memory": 25074, "step": 34390} +{"lr": 0.00034288582233137673, "data_time": 0.0013932228088378907, "loss": 0.013538908492773772, "time": 0.8504750967025757, "epoch": 6, "memory": 25074, "step": 34490} +{"accuracy/top1": 99.76011657714844, "data_time": 0.2468103521010455, "time": 0.6382620755363914, "step": 6} +{"lr": 0.00031961812419703224, "data_time": 0.0013899564743041991, "loss": 0.006582147255539894, "time": 0.7950494766235352, "epoch": 7, "memory": 25074, "step": 34648} +{"lr": 0.00031961812419703224, "data_time": 0.0011787176132202148, "loss": 0.005933204246684909, "time": 0.7777802228927613, "epoch": 7, "memory": 25074, "step": 34748} +{"lr": 0.00031961812419703224, "data_time": 0.0014272689819335937, "loss": 0.00582629581913352, "time": 0.8077687501907349, "epoch": 7, "memory": 25074, "step": 34848} +{"lr": 0.00031961812419703224, "data_time": 0.0014104366302490235, "loss": 0.010212561720982194, "time": 0.8208668470382691, "epoch": 7, "memory": 25074, "step": 34948} +{"lr": 0.00031961812419703224, "data_time": 0.0013582706451416016, "loss": 0.005470056552439928, "time": 0.7720654726028442, "epoch": 7, "memory": 25074, "step": 35048} +{"lr": 0.00031961812419703224, "data_time": 0.0014965534210205078, "loss": 0.007361783506348729, "time": 0.7302881956100464, "epoch": 7, "memory": 25074, "step": 35148} +{"lr": 0.00031961812419703224, "data_time": 0.0010710477828979493, "loss": 0.008483129786327481, "time": 0.8307805299758911, "epoch": 7, "memory": 25074, "step": 35248} +{"lr": 0.00031961812419703224, "data_time": 0.0015103578567504882, "loss": 0.007372635323554278, "time": 0.8447330236434937, "epoch": 7, "memory": 25074, "step": 35348} +{"lr": 0.00031961812419703224, "data_time": 0.0013512372970581055, "loss": 0.0020218949764966966, "time": 0.7620120525360108, "epoch": 7, "memory": 25074, "step": 35448} +{"lr": 0.00031961812419703224, "data_time": 0.0013833284378051759, "loss": 0.006898512272164226, "time": 0.7676534175872802, "epoch": 7, "memory": 25074, "step": 35548} +{"lr": 0.00031961812419703224, "data_time": 0.0015387773513793946, "loss": 0.005877008941024542, "time": 0.8133733987808227, "epoch": 7, "memory": 25074, "step": 35648} +{"lr": 0.00031961812419703224, "data_time": 0.0016529321670532226, "loss": 0.007860430562868715, "time": 0.8411382913589478, "epoch": 7, "memory": 25074, "step": 35748} +{"lr": 0.00031961812419703224, "data_time": 0.001423478126525879, "loss": 0.006672559911385179, "time": 0.784315538406372, "epoch": 7, "memory": 25074, "step": 35848} +{"lr": 0.00031961812419703224, "data_time": 0.0014984607696533203, "loss": 0.013350946130231023, "time": 0.7980647325515747, "epoch": 7, "memory": 25074, "step": 35948} +{"lr": 0.00031961812419703224, "data_time": 0.0013253211975097657, "loss": 0.009304246306419373, "time": 0.7963850498199463, "epoch": 7, "memory": 25074, "step": 36048} +{"lr": 0.00031961812419703224, "data_time": 0.0010668039321899414, "loss": 0.006125065824016928, "time": 0.7851140022277832, "epoch": 7, "memory": 25074, "step": 36148} +{"lr": 0.00031961812419703224, "data_time": 0.0012668371200561523, "loss": 0.008404319267719984, "time": 0.7794710397720337, "epoch": 7, "memory": 25074, "step": 36248} +{"lr": 0.00031961812419703224, "data_time": 0.0014218568801879882, "loss": 0.007274479186162352, "time": 0.8161937475204468, "epoch": 7, "memory": 25074, "step": 36348} +{"lr": 0.00031961812419703224, "data_time": 0.0014168262481689454, "loss": 0.013077496737241744, "time": 0.7840482950210571, "epoch": 7, "memory": 25074, "step": 36448} +{"lr": 0.00031961812419703224, "data_time": 0.0012868642807006836, "loss": 0.007608062401413917, "time": 0.7778895854949951, "epoch": 7, "memory": 25074, "step": 36548} +{"lr": 0.00031961812419703224, "data_time": 0.001390552520751953, "loss": 0.005919932993128896, "time": 0.7799853801727294, "epoch": 7, "memory": 25074, "step": 36648} +{"lr": 0.00031961812419703224, "data_time": 0.001351642608642578, "loss": 0.010841946955770254, "time": 0.8129358768463135, "epoch": 7, "memory": 25074, "step": 36748} +{"lr": 0.00031961812419703224, "data_time": 0.0011922121047973633, "loss": 0.0059931542258709666, "time": 0.7827545166015625, "epoch": 7, "memory": 25074, "step": 36848} +{"lr": 0.00031961812419703224, "data_time": 0.0011208772659301758, "loss": 0.006005952227860689, "time": 0.7702524900436402, "epoch": 7, "memory": 25074, "step": 36948} +{"lr": 0.00031961812419703224, "data_time": 0.0014847517013549805, "loss": 0.009921358013525606, "time": 0.8259676218032836, "epoch": 7, "memory": 25074, "step": 37048} +{"lr": 0.00031961812419703224, "data_time": 0.0012612104415893554, "loss": 0.009072218835353852, "time": 0.9323811769485474, "epoch": 7, "memory": 25074, "step": 37148} +{"lr": 0.00031961812419703224, "data_time": 0.001452803611755371, "loss": 0.008785020234063268, "time": 0.818962574005127, "epoch": 7, "memory": 25074, "step": 37248} +{"lr": 0.00031961812419703224, "data_time": 0.001206374168395996, "loss": 0.005893416702747345, "time": 0.7777831554412842, "epoch": 7, "memory": 25074, "step": 37348} +{"lr": 0.00031961812419703224, "data_time": 0.0013034582138061524, "loss": 0.007719710515812039, "time": 0.8131859540939331, "epoch": 7, "memory": 25074, "step": 37448} +{"lr": 0.00031961812419703224, "data_time": 0.0012425422668457032, "loss": 0.008503195736557246, "time": 0.816457986831665, "epoch": 7, "memory": 25074, "step": 37548} +{"lr": 0.00031961812419703224, "data_time": 0.0013104915618896485, "loss": 0.007905058888718485, "time": 0.8407413005828858, "epoch": 7, "memory": 25074, "step": 37648} +{"lr": 0.00031961812419703224, "data_time": 0.0015394449234008788, "loss": 0.011772033665329218, "time": 0.7709179639816284, "epoch": 7, "memory": 25074, "step": 37748} +{"lr": 0.00031961812419703224, "data_time": 0.0015858650207519532, "loss": 0.005198361305519939, "time": 0.8399975538253784, "epoch": 7, "memory": 25074, "step": 37848} +{"lr": 0.00031961812419703224, "data_time": 0.0011867761611938476, "loss": 0.007491279020905494, "time": 0.7971672058105469, "epoch": 7, "memory": 25074, "step": 37948} +{"lr": 0.00031961812419703224, "data_time": 0.0013853788375854492, "loss": 0.013604060234501957, "time": 0.8534063100814819, "epoch": 7, "memory": 25074, "step": 38048} +{"lr": 0.00031961812419703224, "data_time": 0.0014464855194091797, "loss": 0.006931758066639304, "time": 0.8338172674179077, "epoch": 7, "memory": 25074, "step": 38148} +{"lr": 0.00031961812419703224, "data_time": 0.0011494159698486328, "loss": 0.007359489006921649, "time": 0.7883419036865235, "epoch": 7, "memory": 25074, "step": 38248} +{"lr": 0.00031961812419703224, "data_time": 0.0012499570846557617, "loss": 0.0036069298163056373, "time": 0.8009266376495361, "epoch": 7, "memory": 25074, "step": 38348} +{"lr": 0.00031961812419703224, "data_time": 0.0014397859573364257, "loss": 0.004166057799011469, "time": 0.7946410655975342, "epoch": 7, "memory": 25074, "step": 38448} +{"lr": 0.00031961812419703224, "data_time": 0.0011008024215698243, "loss": 0.0033724519424140455, "time": 0.8424299955368042, "epoch": 7, "memory": 25074, "step": 38548} +{"lr": 0.00031961812419703224, "data_time": 0.0012166738510131837, "loss": 0.005197452334687114, "time": 0.8725511312484742, "epoch": 7, "memory": 25074, "step": 38648} +{"lr": 0.00031961812419703224, "data_time": 0.0013074398040771485, "loss": 0.009734292468056082, "time": 0.8246922492980957, "epoch": 7, "memory": 25074, "step": 38748} +{"lr": 0.00031961812419703224, "data_time": 0.0012614965438842774, "loss": 0.006522056041285395, "time": 0.8296808719635009, "epoch": 7, "memory": 25074, "step": 38848} +{"lr": 0.00031961812419703224, "data_time": 0.0013681650161743164, "loss": 0.003907531034201384, "time": 0.8276172637939453, "epoch": 7, "memory": 25074, "step": 38948} +{"lr": 0.00031961812419703224, "data_time": 0.0017262697219848633, "loss": 0.0030798387713730337, "time": 0.8221835374832154, "epoch": 7, "memory": 25074, "step": 39048} +{"lr": 0.00031961812419703224, "data_time": 0.0016393423080444335, "loss": 0.009174630232155322, "time": 0.8391355514526367, "epoch": 7, "memory": 25074, "step": 39148} +{"lr": 0.00031961812419703224, "data_time": 0.0012131690979003905, "loss": 0.008284668670967221, "time": 0.8018346309661866, "epoch": 7, "memory": 25074, "step": 39248} +{"lr": 0.00031961812419703224, "data_time": 0.001375102996826172, "loss": 0.0025517712347209454, "time": 0.839412784576416, "epoch": 7, "memory": 25074, "step": 39348} +{"lr": 0.00031961812419703224, "data_time": 0.0013039588928222656, "loss": 0.0031253993045538666, "time": 0.7875226736068726, "epoch": 7, "memory": 25074, "step": 39448} +{"lr": 0.00031961812419703224, "data_time": 0.0014575958251953126, "loss": 0.007066106610000134, "time": 0.8162870168685913, "epoch": 7, "memory": 25074, "step": 39548} +{"lr": 0.00031961812419703224, "data_time": 0.0014191627502441405, "loss": 0.0052933010738343, "time": 0.7524826765060425, "epoch": 7, "memory": 25074, "step": 39648} +{"lr": 0.00031961812419703224, "data_time": 0.001366734504699707, "loss": 0.008860271843150257, "time": 0.8108349323272706, "epoch": 7, "memory": 25074, "step": 39748} +{"lr": 0.00031961812419703224, "data_time": 0.0013921737670898437, "loss": 0.008120916970074176, "time": 0.7975910186767579, "epoch": 7, "memory": 25074, "step": 39848} +{"lr": 0.00031961812419703224, "data_time": 0.0012505292892456056, "loss": 0.004192754300311208, "time": 0.7771164655685425, "epoch": 7, "memory": 25074, "step": 39948} +{"lr": 0.00031961812419703224, "data_time": 0.0012868642807006836, "loss": 0.006793348211795091, "time": 0.8004083156585693, "epoch": 7, "memory": 25074, "step": 40048} +{"lr": 0.00031961812419703224, "data_time": 0.0016983270645141602, "loss": 0.001058782683685422, "time": 0.7887581348419189, "epoch": 7, "memory": 25074, "step": 40148} +{"lr": 0.00031961812419703224, "data_time": 0.0013393878936767579, "loss": 0.005151572031900286, "time": 0.8344828367233277, "epoch": 7, "memory": 25074, "step": 40248} +{"accuracy/top1": 99.8926773071289, "data_time": 0.2596237098469454, "time": 0.645106596105239, "step": 7} +{"lr": 0.0002935281474492116, "data_time": 0.08174433708190917, "loss": 0.004533440480008722, "time": 0.8853747844696045, "epoch": 8, "memory": 25074, "step": 40406} +{"lr": 0.0002935281474492116, "data_time": 0.0012613534927368164, "loss": 0.008634087070822716, "time": 0.7807964324951172, "epoch": 8, "memory": 25074, "step": 40506} +{"lr": 0.0002935281474492116, "data_time": 0.0012382030487060546, "loss": 0.007455760054290295, "time": 0.8089942455291748, "epoch": 8, "memory": 25074, "step": 40606} +{"lr": 0.0002935281474492116, "data_time": 0.001336359977722168, "loss": 0.006103390548378229, "time": 0.7352396726608277, "epoch": 8, "memory": 25074, "step": 40706} +{"lr": 0.0002935281474492116, "data_time": 0.0014557600021362304, "loss": 0.005343703227117658, "time": 0.754692792892456, "epoch": 8, "memory": 25074, "step": 40806} +{"lr": 0.0002935281474492116, "data_time": 0.0016048431396484375, "loss": 0.006173395831137895, "time": 0.7743478059768677, "epoch": 8, "memory": 25074, "step": 40906} +{"lr": 0.0002935281474492116, "data_time": 0.0016127586364746093, "loss": 0.004697187058627606, "time": 0.7527307987213134, "epoch": 8, "memory": 25074, "step": 41006} +{"lr": 0.0002935281474492116, "data_time": 0.0014994144439697266, "loss": 0.007261675829067826, "time": 0.7300341606140137, "epoch": 8, "memory": 25074, "step": 41106} +{"lr": 0.0002935281474492116, "data_time": 0.0015994071960449218, "loss": 0.005552244512364269, "time": 0.7418447256088256, "epoch": 8, "memory": 25074, "step": 41206} +{"lr": 0.0002935281474492116, "data_time": 0.0012565135955810546, "loss": 0.006357346661388874, "time": 0.7793103218078613, "epoch": 8, "memory": 25074, "step": 41306} +{"lr": 0.0002935281474492116, "data_time": 0.0014730453491210937, "loss": 0.004397300910204649, "time": 0.8202110767364502, "epoch": 8, "memory": 25074, "step": 41406} +{"lr": 0.0002935281474492116, "data_time": 0.0014173030853271485, "loss": 0.003983988706022501, "time": 0.8394148111343384, "epoch": 8, "memory": 25074, "step": 41506} +{"lr": 0.0002935281474492116, "data_time": 0.0016449213027954102, "loss": 0.003592493990436196, "time": 0.752266526222229, "epoch": 8, "memory": 25074, "step": 41606} +{"lr": 0.0002935281474492116, "data_time": 0.001744985580444336, "loss": 0.008995352173224091, "time": 0.7827673435211182, "epoch": 8, "memory": 25074, "step": 41706} +{"lr": 0.0002935281474492116, "data_time": 0.0015886068344116212, "loss": 0.0035457221325486898, "time": 0.777490234375, "epoch": 8, "memory": 25074, "step": 41806} +{"lr": 0.0002935281474492116, "data_time": 0.0012670516967773437, "loss": 0.010279138339683414, "time": 0.8055679082870484, "epoch": 8, "memory": 25074, "step": 41906} +{"lr": 0.0002935281474492116, "data_time": 0.0014852285385131836, "loss": 0.004720044881105423, "time": 0.7767008304595947, "epoch": 8, "memory": 25074, "step": 42006} +{"lr": 0.0002935281474492116, "data_time": 0.0012957334518432617, "loss": 0.004443504381924868, "time": 0.7426398515701294, "epoch": 8, "memory": 25074, "step": 42106} +{"lr": 0.0002935281474492116, "data_time": 0.0012934446334838868, "loss": 0.006596829975023866, "time": 0.7938202857971192, "epoch": 8, "memory": 25074, "step": 42206} +{"lr": 0.0002935281474492116, "data_time": 0.001536989212036133, "loss": 0.007685122918337584, "time": 0.7267671585083008, "epoch": 8, "memory": 25074, "step": 42306} +{"lr": 0.0002935281474492116, "data_time": 0.0014831304550170898, "loss": 0.0075868109706789255, "time": 0.8245455503463746, "epoch": 8, "memory": 25074, "step": 42406} +{"lr": 0.0002935281474492116, "data_time": 0.0012464523315429688, "loss": 0.007385736098513007, "time": 0.8491693735122681, "epoch": 8, "memory": 25074, "step": 42506} +{"lr": 0.0002935281474492116, "data_time": 0.0015519380569458008, "loss": 0.003152378322556615, "time": 0.7582871913909912, "epoch": 8, "memory": 25074, "step": 42606} +{"lr": 0.0002935281474492116, "data_time": 0.0015825271606445313, "loss": 0.0063883363734930755, "time": 0.7175159454345703, "epoch": 8, "memory": 25074, "step": 42706} +{"lr": 0.0002935281474492116, "data_time": 0.0015311479568481446, "loss": 0.007287848042324185, "time": 0.7520819902420044, "epoch": 8, "memory": 25074, "step": 42806} +{"lr": 0.0002935281474492116, "data_time": 0.0012787580490112305, "loss": 0.007251392956823111, "time": 0.7041292190551758, "epoch": 8, "memory": 25074, "step": 42906} +{"lr": 0.0002935281474492116, "data_time": 0.0015406370162963866, "loss": 0.004316490795463324, "time": 0.7086874961853027, "epoch": 8, "memory": 25074, "step": 43006} +{"lr": 0.0002935281474492116, "data_time": 0.07285220623016357, "loss": 0.005532010411843658, "time": 0.8373117446899414, "epoch": 8, "memory": 25074, "step": 43106} +{"lr": 0.0002935281474492116, "data_time": 0.11478307247161865, "loss": 0.012309360085055232, "time": 0.8191475391387939, "epoch": 8, "memory": 25074, "step": 43206} +{"lr": 0.0002935281474492116, "data_time": 0.009643650054931641, "loss": 0.008047929825261235, "time": 0.7130385160446167, "epoch": 8, "memory": 25074, "step": 43306} +{"lr": 0.0002935281474492116, "data_time": 0.141137957572937, "loss": 0.0039288499392569065, "time": 0.8460032224655152, "epoch": 8, "memory": 25074, "step": 43406} +{"lr": 0.0002935281474492116, "data_time": 0.0016232013702392578, "loss": 0.00623535355553031, "time": 0.849001145362854, "epoch": 8, "memory": 25074, "step": 43506} +{"lr": 0.0002935281474492116, "data_time": 0.0012703657150268555, "loss": 0.00541509916074574, "time": 0.8788488388061524, "epoch": 8, "memory": 25074, "step": 43606} +{"lr": 0.0002935281474492116, "data_time": 0.0012674331665039062, "loss": 0.005426371656358242, "time": 0.7773834705352783, "epoch": 8, "memory": 25074, "step": 43706} +{"lr": 0.0002935281474492116, "data_time": 0.0017203807830810547, "loss": 0.004386982461437583, "time": 0.7341472864151001, "epoch": 8, "memory": 25074, "step": 43806} +{"lr": 0.0002935281474492116, "data_time": 0.0014246225357055664, "loss": 0.0026717221364378927, "time": 0.7962167263031006, "epoch": 8, "memory": 25074, "step": 43906} +{"lr": 0.0002935281474492116, "data_time": 0.0019036293029785155, "loss": 0.005585245974361897, "time": 0.7138374090194702, "epoch": 8, "memory": 25074, "step": 44006} +{"lr": 0.0002935281474492116, "data_time": 0.001444530487060547, "loss": 0.008353070076555013, "time": 0.7424083948135376, "epoch": 8, "memory": 25074, "step": 44106} +{"lr": 0.0002935281474492116, "data_time": 0.0014462471008300781, "loss": 0.00635327473282814, "time": 0.8163536548614502, "epoch": 8, "memory": 25074, "step": 44206} +{"lr": 0.0002935281474492116, "data_time": 0.0016684532165527344, "loss": 0.004578802362084388, "time": 0.7836791753768921, "epoch": 8, "memory": 25074, "step": 44306} +{"lr": 0.0002935281474492116, "data_time": 0.0016208410263061524, "loss": 0.005181445740163326, "time": 0.8223507881164551, "epoch": 8, "memory": 25074, "step": 44406} +{"lr": 0.0002935281474492116, "data_time": 0.0013053178787231444, "loss": 0.006681210361421108, "time": 0.8298809766769409, "epoch": 8, "memory": 25074, "step": 44506} +{"lr": 0.0002935281474492116, "data_time": 0.0013679265975952148, "loss": 0.006784602254629135, "time": 0.7622411489486695, "epoch": 8, "memory": 25074, "step": 44606} +{"lr": 0.0002935281474492116, "data_time": 0.0014540195465087891, "loss": 0.01024253643117845, "time": 0.7808483839035034, "epoch": 8, "memory": 25074, "step": 44706} +{"lr": 0.0002935281474492116, "data_time": 0.0015398025512695312, "loss": 0.002162778237834573, "time": 0.7791775941848755, "epoch": 8, "memory": 25074, "step": 44806} +{"lr": 0.0002935281474492116, "data_time": 0.0012364864349365234, "loss": 0.0053187865298241375, "time": 0.8003702163696289, "epoch": 8, "memory": 25074, "step": 44906} +{"lr": 0.0002935281474492116, "data_time": 0.0015358448028564454, "loss": 0.003356165671721101, "time": 0.77942054271698, "epoch": 8, "memory": 25074, "step": 45006} +{"lr": 0.0002935281474492116, "data_time": 0.0014313936233520507, "loss": 0.007574144750833511, "time": 0.8116400718688965, "epoch": 8, "memory": 25074, "step": 45106} +{"lr": 0.0002935281474492116, "data_time": 0.0016705751419067382, "loss": 0.007972662895917892, "time": 0.7434316158294678, "epoch": 8, "memory": 25074, "step": 45206} +{"lr": 0.0002935281474492116, "data_time": 0.0012651920318603516, "loss": 0.004711715783923864, "time": 0.7442546606063842, "epoch": 8, "memory": 25074, "step": 45306} +{"lr": 0.0002935281474492116, "data_time": 0.0011793613433837891, "loss": 0.004612048994749785, "time": 0.789636492729187, "epoch": 8, "memory": 25074, "step": 45406} +{"lr": 0.0002935281474492116, "data_time": 0.0014587879180908204, "loss": 0.006722743064165116, "time": 0.8347979307174682, "epoch": 8, "memory": 25074, "step": 45506} +{"lr": 0.0002935281474492116, "data_time": 0.0014156579971313476, "loss": 0.004672659328207374, "time": 0.78687002658844, "epoch": 8, "memory": 25074, "step": 45606} +{"lr": 0.0002935281474492116, "data_time": 0.001810598373413086, "loss": 0.004899133229628205, "time": 0.7504078865051269, "epoch": 8, "memory": 25074, "step": 45706} +{"lr": 0.0002935281474492116, "data_time": 0.0016098260879516602, "loss": 0.008960438240319491, "time": 0.7846177101135254, "epoch": 8, "memory": 25074, "step": 45806} +{"lr": 0.0002935281474492116, "data_time": 0.0016513347625732421, "loss": 0.003535037161782384, "time": 0.825016188621521, "epoch": 8, "memory": 25074, "step": 45906} +{"lr": 0.0002935281474492116, "data_time": 0.0014484167098999024, "loss": 0.0037020920775830747, "time": 0.7800180196762085, "epoch": 8, "memory": 25074, "step": 46006} +{"accuracy/top1": 99.96842956542969, "data_time": 0.25169263166539807, "time": 0.6378555718590232, "step": 8} +{"lr": 0.00026525831390311475, "data_time": 0.0016158819198608398, "loss": 0.010060214251279832, "time": 0.8175989151000976, "epoch": 9, "memory": 25074, "step": 46164} +{"lr": 0.00026525831390311475, "data_time": 0.001458597183227539, "loss": 0.003008700720965862, "time": 0.784386134147644, "epoch": 9, "memory": 25074, "step": 46264} +{"lr": 0.00026525831390311475, "data_time": 0.0013403892517089844, "loss": 0.009089087136089802, "time": 0.749416732788086, "epoch": 9, "memory": 25074, "step": 46364} +{"lr": 0.00026525831390311475, "data_time": 0.0013660907745361329, "loss": 0.004336386872455478, "time": 0.8461895704269409, "epoch": 9, "memory": 25074, "step": 46464} +{"lr": 0.00026525831390311475, "data_time": 0.0013083219528198242, "loss": 0.0032400793861597776, "time": 0.7555406332015991, "epoch": 9, "memory": 25074, "step": 46564} +{"lr": 0.00026525831390311475, "data_time": 0.00168917179107666, "loss": 0.002776416391134262, "time": 0.8164075613021851, "epoch": 9, "memory": 25074, "step": 46664} +{"lr": 0.00026525831390311475, "data_time": 0.0012827157974243165, "loss": 0.004847720824182033, "time": 0.7282079458236694, "epoch": 9, "memory": 25074, "step": 46764} +{"lr": 0.00026525831390311475, "data_time": 0.001482677459716797, "loss": 0.0037592971231788395, "time": 0.7747260808944703, "epoch": 9, "memory": 25074, "step": 46864} +{"lr": 0.00026525831390311475, "data_time": 0.0016307830810546875, "loss": 0.003753243526443839, "time": 0.8503926515579223, "epoch": 9, "memory": 25074, "step": 46964} +{"lr": 0.00026525831390311475, "data_time": 0.0015087604522705078, "loss": 0.006353098386898637, "time": 0.7096498966217041, "epoch": 9, "memory": 25074, "step": 47064} +{"lr": 0.00026525831390311475, "data_time": 0.0014767169952392579, "loss": 0.0016077717766165734, "time": 0.7514087438583374, "epoch": 9, "memory": 25074, "step": 47164} +{"lr": 0.00026525831390311475, "data_time": 0.001598668098449707, "loss": 0.003843969479203224, "time": 0.7461911201477051, "epoch": 9, "memory": 25074, "step": 47264} +{"lr": 0.00026525831390311475, "data_time": 0.032735586166381836, "loss": 0.0028642588295042515, "time": 0.7378857135772705, "epoch": 9, "memory": 25074, "step": 47364} +{"lr": 0.00026525831390311475, "data_time": 0.10020203590393066, "loss": 0.0032532909885048867, "time": 0.8054174184799194, "epoch": 9, "memory": 25074, "step": 47464} +{"lr": 0.00026525831390311475, "data_time": 0.0016108512878417968, "loss": 0.0048339446540921925, "time": 0.7717607498168946, "epoch": 9, "memory": 25074, "step": 47564} +{"lr": 0.00026525831390311475, "data_time": 0.0014272451400756836, "loss": 0.003592148795723915, "time": 0.7555326223373413, "epoch": 9, "memory": 25074, "step": 47664} +{"lr": 0.00026525831390311475, "data_time": 0.0014083385467529297, "loss": 0.002575997915118933, "time": 0.8240430593490601, "epoch": 9, "memory": 25074, "step": 47764} +{"lr": 0.00026525831390311475, "data_time": 0.001306009292602539, "loss": 0.005089926393702626, "time": 1.0330263137817384, "epoch": 9, "memory": 25074, "step": 47864} +{"lr": 0.00026525831390311475, "data_time": 0.001449275016784668, "loss": 0.0021736639086157084, "time": 0.798820972442627, "epoch": 9, "memory": 25074, "step": 47964} +{"lr": 0.00026525831390311475, "data_time": 0.001599431037902832, "loss": 0.004608659353107214, "time": 0.7402681112289429, "epoch": 9, "memory": 25074, "step": 48064} +{"lr": 0.00026525831390311475, "data_time": 0.0013656139373779297, "loss": 0.005686013540253043, "time": 0.8199795007705688, "epoch": 9, "memory": 25074, "step": 48164} +{"lr": 0.00026525831390311475, "data_time": 0.0015366792678833008, "loss": 0.006041695084422827, "time": 0.7848222494125366, "epoch": 9, "memory": 25074, "step": 48264} +{"lr": 0.00026525831390311475, "data_time": 0.0015355348587036133, "loss": 0.005315634049475193, "time": 0.810681676864624, "epoch": 9, "memory": 25074, "step": 48364} +{"lr": 0.00026525831390311475, "data_time": 0.0013628721237182618, "loss": 0.005327724944800138, "time": 0.7424980163574219, "epoch": 9, "memory": 25074, "step": 48464} +{"lr": 0.00026525831390311475, "data_time": 0.0020877361297607423, "loss": 0.0017982026096433402, "time": 0.7792024612426758, "epoch": 9, "memory": 25074, "step": 48564} +{"lr": 0.00026525831390311475, "data_time": 0.0013993740081787109, "loss": 0.0045343706849962475, "time": 0.7861818075180054, "epoch": 9, "memory": 25074, "step": 48664} +{"lr": 0.00026525831390311475, "data_time": 0.001527690887451172, "loss": 0.003606828721240163, "time": 0.7916712522506714, "epoch": 9, "memory": 25074, "step": 48764} +{"lr": 0.00026525831390311475, "data_time": 0.0020267486572265623, "loss": 0.010490537900477648, "time": 0.7570760726928711, "epoch": 9, "memory": 25074, "step": 48864} +{"lr": 0.00026525831390311475, "data_time": 0.0016240119934082032, "loss": 0.005129858665168285, "time": 0.7905623197555542, "epoch": 9, "memory": 25074, "step": 48964} +{"lr": 0.00026525831390311475, "data_time": 0.0014204263687133789, "loss": 0.005493970308452845, "time": 0.8172428131103515, "epoch": 9, "memory": 25074, "step": 49064} +{"lr": 0.00026525831390311475, "data_time": 0.0740964651107788, "loss": 0.004783406946808099, "time": 0.7778555154800415, "epoch": 9, "memory": 25074, "step": 49164} +{"lr": 0.00026525831390311475, "data_time": 0.11791810989379883, "loss": 0.007851413358002901, "time": 0.8225356101989746, "epoch": 9, "memory": 25074, "step": 49264} +{"lr": 0.00026525831390311475, "data_time": 0.08130881786346436, "loss": 0.0026842295192182064, "time": 0.7910864591598511, "epoch": 9, "memory": 25074, "step": 49364} +{"lr": 0.00026525831390311475, "data_time": 0.13063509464263917, "loss": 0.002057286910712719, "time": 0.8399464607238769, "epoch": 9, "memory": 25074, "step": 49464} +{"lr": 0.00026525831390311475, "data_time": 0.08142657279968261, "loss": 0.006802480015903711, "time": 0.7852344751358032, "epoch": 9, "memory": 25074, "step": 49564} +{"lr": 0.00026525831390311475, "data_time": 0.11105425357818603, "loss": 0.0015038976445794106, "time": 0.8148003578186035, "epoch": 9, "memory": 25074, "step": 49664} +{"lr": 0.00026525831390311475, "data_time": 0.13849332332611083, "loss": 0.003886723518371582, "time": 0.8466254949569703, "epoch": 9, "memory": 25074, "step": 49764} +{"lr": 0.00026525831390311475, "data_time": 0.07313673496246338, "loss": 0.005093358037993312, "time": 0.7784403800964356, "epoch": 9, "memory": 25074, "step": 49864} +{"lr": 0.00026525831390311475, "data_time": 0.07475545406341552, "loss": 0.003619781881570816, "time": 0.7899048566818238, "epoch": 9, "memory": 25074, "step": 49964} +{"lr": 0.00026525831390311475, "data_time": 0.08455250263214112, "loss": 0.0045316462405025956, "time": 0.7896649360656738, "epoch": 9, "memory": 25074, "step": 50064} +{"lr": 0.00026525831390311475, "data_time": 0.07857251167297363, "loss": 0.005448000272735953, "time": 0.7821491241455079, "epoch": 9, "memory": 25074, "step": 50164} +{"lr": 0.00026525831390311475, "data_time": 0.048411202430725095, "loss": 0.0034753140062093735, "time": 0.751510214805603, "epoch": 9, "memory": 25074, "step": 50264} +{"lr": 0.00026525831390311475, "data_time": 0.0017418384552001954, "loss": 0.0038141491357237102, "time": 0.7980393171310425, "epoch": 9, "memory": 25074, "step": 50364} +{"lr": 0.00026525831390311475, "data_time": 0.0020066976547241213, "loss": 0.0020103800110518933, "time": 0.8106767177581787, "epoch": 9, "memory": 25074, "step": 50464} +{"lr": 0.00026525831390311475, "data_time": 0.0013964176177978516, "loss": 0.0038177080918103457, "time": 0.7461270332336426, "epoch": 9, "memory": 25074, "step": 50564} +{"lr": 0.00026525831390311475, "data_time": 0.0014374732971191406, "loss": 0.001423835102468729, "time": 0.7556110858917237, "epoch": 9, "memory": 25074, "step": 50664} +{"lr": 0.00026525831390311475, "data_time": 0.0015955209732055665, "loss": 0.00488781756721437, "time": 0.7805282831192016, "epoch": 9, "memory": 25074, "step": 50764} +{"lr": 0.00026525831390311475, "data_time": 0.0013822555541992188, "loss": 0.004550968669354916, "time": 0.8374075174331665, "epoch": 9, "memory": 25074, "step": 50864} +{"lr": 0.00026525831390311475, "data_time": 0.004905128479003906, "loss": 0.0032081414479762315, "time": 0.7419009685516358, "epoch": 9, "memory": 25074, "step": 50964} +{"lr": 0.00026525831390311475, "data_time": 0.0015484809875488282, "loss": 0.0032443346455693243, "time": 0.7633739471435547, "epoch": 9, "memory": 25074, "step": 51064} +{"lr": 0.00026525831390311475, "data_time": 0.001518845558166504, "loss": 0.004434935515746474, "time": 0.7812606573104859, "epoch": 9, "memory": 25074, "step": 51164} +{"lr": 0.00026525831390311475, "data_time": 0.00128173828125, "loss": 0.003473523911088705, "time": 0.7531224489212036, "epoch": 9, "memory": 25074, "step": 51264} +{"lr": 0.00026525831390311475, "data_time": 0.001495671272277832, "loss": 0.0029246626887470483, "time": 0.7889619588851928, "epoch": 9, "memory": 25074, "step": 51364} +{"lr": 0.00026525831390311475, "data_time": 0.0013351917266845703, "loss": 0.004868371831253171, "time": 0.8216519594192505, "epoch": 9, "memory": 25074, "step": 51464} +{"lr": 0.00026525831390311475, "data_time": 0.0014025449752807617, "loss": 0.0019766067154705523, "time": 0.7716431379318237, "epoch": 9, "memory": 25074, "step": 51564} +{"lr": 0.00026525831390311475, "data_time": 0.0015723228454589844, "loss": 0.0039232814684510234, "time": 0.7720461845397949, "epoch": 9, "memory": 25074, "step": 51664} +{"lr": 0.00026525831390311475, "data_time": 0.0016120195388793946, "loss": 0.003156856168061495, "time": 0.7533194303512574, "epoch": 9, "memory": 25074, "step": 51764} +{"accuracy/top1": 99.97474670410156, "data_time": 0.2695574620190789, "time": 0.6553976676043343, "step": 9} +{"lr": 0.000235504720682845, "data_time": 0.057193541526794435, "loss": 0.004519202141091227, "time": 0.7609281539916992, "epoch": 10, "memory": 25074, "step": 51922} +{"lr": 0.000235504720682845, "data_time": 0.08888649940490723, "loss": 0.0011984863318502902, "time": 0.804810619354248, "epoch": 10, "memory": 25074, "step": 52022} +{"lr": 0.000235504720682845, "data_time": 0.013669371604919434, "loss": 0.007437628647312522, "time": 0.7698734521865844, "epoch": 10, "memory": 25074, "step": 52122} +{"lr": 0.000235504720682845, "data_time": 0.10023550987243653, "loss": 0.0019167819991707802, "time": 0.8420120716094971, "epoch": 10, "memory": 25074, "step": 52222} +{"lr": 0.000235504720682845, "data_time": 0.0015046834945678712, "loss": 0.0026889451313763858, "time": 0.7677485942840576, "epoch": 10, "memory": 25074, "step": 52322} +{"lr": 0.000235504720682845, "data_time": 0.007773685455322266, "loss": 0.0049743682611733675, "time": 0.7666850328445435, "epoch": 10, "memory": 25074, "step": 52422} +{"lr": 0.000235504720682845, "data_time": 0.0013694286346435547, "loss": 0.003320180857554078, "time": 0.8272189855575561, "epoch": 10, "memory": 25074, "step": 52522} +{"lr": 0.000235504720682845, "data_time": 0.0013002634048461914, "loss": 0.0023750689812004565, "time": 0.7317907094955445, "epoch": 10, "memory": 25074, "step": 52622} +{"lr": 0.000235504720682845, "data_time": 0.001642465591430664, "loss": 0.0017983930185437203, "time": 0.761565375328064, "epoch": 10, "memory": 25074, "step": 52722} +{"lr": 0.000235504720682845, "data_time": 0.07177071571350098, "loss": 0.006485605705529451, "time": 0.7790808916091919, "epoch": 10, "memory": 25074, "step": 52822} +{"lr": 0.000235504720682845, "data_time": 0.0016214847564697266, "loss": 0.003673615679144859, "time": 0.7736930608749389, "epoch": 10, "memory": 25074, "step": 52922} +{"lr": 0.000235504720682845, "data_time": 0.001363515853881836, "loss": 0.00527961584739387, "time": 0.8046427965164185, "epoch": 10, "memory": 25074, "step": 53022} +{"lr": 0.000235504720682845, "data_time": 0.0014634132385253906, "loss": 0.006082437140867114, "time": 0.802378797531128, "epoch": 10, "memory": 25074, "step": 53122} +{"lr": 0.000235504720682845, "data_time": 0.0012588262557983398, "loss": 0.005271776346489787, "time": 0.7711512565612793, "epoch": 10, "memory": 25074, "step": 53222} +{"lr": 0.000235504720682845, "data_time": 0.0014651060104370118, "loss": 0.002576451888307929, "time": 0.8612143516540527, "epoch": 10, "memory": 25074, "step": 53322} +{"lr": 0.000235504720682845, "data_time": 0.0011648416519165039, "loss": 0.0026614618953317406, "time": 0.8963905334472656, "epoch": 10, "memory": 25074, "step": 53422} +{"lr": 0.000235504720682845, "data_time": 0.0015602827072143555, "loss": 0.001886838860809803, "time": 0.8743463039398194, "epoch": 10, "memory": 25074, "step": 53522} +{"lr": 0.000235504720682845, "data_time": 0.006681203842163086, "loss": 0.002380321780219674, "time": 1.4329699516296386, "epoch": 10, "memory": 25074, "step": 53622} +{"lr": 0.000235504720682845, "data_time": 0.0020303964614868165, "loss": 0.004485356388613582, "time": 0.8864237070083618, "epoch": 10, "memory": 25074, "step": 53722} +{"lr": 0.000235504720682845, "data_time": 0.0014768362045288086, "loss": 0.004296805337071418, "time": 0.8816102743148804, "epoch": 10, "memory": 25074, "step": 53822} +{"lr": 0.000235504720682845, "data_time": 0.0013443470001220704, "loss": 0.004021440260112285, "time": 0.7687443494796753, "epoch": 10, "memory": 25074, "step": 53922} +{"lr": 0.000235504720682845, "data_time": 0.0012895584106445313, "loss": 0.0012011222075670957, "time": 0.8162369012832642, "epoch": 10, "memory": 25074, "step": 54022} +{"lr": 0.000235504720682845, "data_time": 0.0015340805053710937, "loss": 0.003927409090101719, "time": 0.7689162492752075, "epoch": 10, "memory": 25074, "step": 54122} +{"lr": 0.000235504720682845, "data_time": 0.001264500617980957, "loss": 0.005482113501057029, "time": 0.7374703407287597, "epoch": 10, "memory": 25074, "step": 54222} +{"lr": 0.000235504720682845, "data_time": 0.0028434514999389647, "loss": 0.007824982004240155, "time": 0.7792750835418701, "epoch": 10, "memory": 25074, "step": 54322} +{"lr": 0.000235504720682845, "data_time": 0.0025654077529907227, "loss": 0.003261878388002515, "time": 0.8095210552215576, "epoch": 10, "memory": 25074, "step": 54422} +{"lr": 0.000235504720682845, "data_time": 0.0015420198440551757, "loss": 0.0032570686656981707, "time": 0.7800433397293091, "epoch": 10, "memory": 25074, "step": 54522} +{"lr": 0.000235504720682845, "data_time": 0.0013351678848266602, "loss": 0.004245009552687406, "time": 0.788340425491333, "epoch": 10, "memory": 25074, "step": 54622} +{"lr": 0.000235504720682845, "data_time": 0.001420736312866211, "loss": 0.0016847546212375164, "time": 0.7512306451797486, "epoch": 10, "memory": 25074, "step": 54722} +{"lr": 0.000235504720682845, "data_time": 0.0014606714248657227, "loss": 0.0033814623951911927, "time": 0.7921860933303833, "epoch": 10, "memory": 25074, "step": 54822} +{"lr": 0.000235504720682845, "data_time": 0.0017631053924560547, "loss": 0.006519189290702343, "time": 0.7578413486480713, "epoch": 10, "memory": 25074, "step": 54922} +{"lr": 0.000235504720682845, "data_time": 0.001624131202697754, "loss": 0.0022362173069268463, "time": 0.7228867769241333, "epoch": 10, "memory": 25074, "step": 55022} +{"lr": 0.000235504720682845, "data_time": 0.0024099826812744142, "loss": 0.002010176237672567, "time": 0.7517282724380493, "epoch": 10, "memory": 25074, "step": 55122} +{"lr": 0.000235504720682845, "data_time": 0.001647639274597168, "loss": 0.0022532119881361725, "time": 0.7662103652954102, "epoch": 10, "memory": 25074, "step": 55222} +{"lr": 0.000235504720682845, "data_time": 0.001513671875, "loss": 0.0018771516624838114, "time": 0.7472053289413452, "epoch": 10, "memory": 25074, "step": 55322} +{"lr": 0.000235504720682845, "data_time": 0.0020645618438720702, "loss": 0.0027181121520698072, "time": 0.7646674156188965, "epoch": 10, "memory": 25074, "step": 55422} +{"lr": 0.000235504720682845, "data_time": 0.001380157470703125, "loss": 0.003250989643856883, "time": 0.7771836996078492, "epoch": 10, "memory": 25074, "step": 55522} +{"lr": 0.000235504720682845, "data_time": 0.0016100406646728516, "loss": 0.0038707809057086707, "time": 0.7240278482437134, "epoch": 10, "memory": 25074, "step": 55622} +{"lr": 0.000235504720682845, "data_time": 0.001523876190185547, "loss": 0.005304170772433281, "time": 0.8107309341430664, "epoch": 10, "memory": 25074, "step": 55722} +{"lr": 0.000235504720682845, "data_time": 0.0015041828155517578, "loss": 0.005562584102153778, "time": 0.787463927268982, "epoch": 10, "memory": 25074, "step": 55822} +{"lr": 0.000235504720682845, "data_time": 0.001501917839050293, "loss": 0.0019271073397248983, "time": 0.761007022857666, "epoch": 10, "memory": 25074, "step": 55922} +{"lr": 0.000235504720682845, "data_time": 0.0014229297637939453, "loss": 0.005638011125847697, "time": 0.7404680252075195, "epoch": 10, "memory": 25074, "step": 56022} +{"lr": 0.000235504720682845, "data_time": 0.0015123128890991212, "loss": 0.00421171523630619, "time": 0.7766730785369873, "epoch": 10, "memory": 25074, "step": 56122} +{"lr": 0.000235504720682845, "data_time": 0.0013295888900756836, "loss": 0.0028375647030770778, "time": 0.7357343196868896, "epoch": 10, "memory": 25074, "step": 56222} +{"lr": 0.000235504720682845, "data_time": 0.0013447284698486328, "loss": 0.00337574053555727, "time": 0.7193809509277344, "epoch": 10, "memory": 25074, "step": 56322} +{"lr": 0.000235504720682845, "data_time": 0.001387786865234375, "loss": 0.0025291111785918476, "time": 0.706843900680542, "epoch": 10, "memory": 25074, "step": 56422} +{"lr": 0.000235504720682845, "data_time": 0.001521921157836914, "loss": 0.0069944093003869055, "time": 0.8126255512237549, "epoch": 10, "memory": 25074, "step": 56522} +{"lr": 0.000235504720682845, "data_time": 0.0016293525695800781, "loss": 0.005933578871190548, "time": 0.7594385385513306, "epoch": 10, "memory": 25074, "step": 56622} +{"lr": 0.000235504720682845, "data_time": 0.002179074287414551, "loss": 0.004614521749317646, "time": 0.769674563407898, "epoch": 10, "memory": 25074, "step": 56722} +{"lr": 0.000235504720682845, "data_time": 0.001364898681640625, "loss": 0.0018319363705813884, "time": 0.9140798568725585, "epoch": 10, "memory": 25074, "step": 56822} +{"lr": 0.000235504720682845, "data_time": 0.0012913227081298828, "loss": 0.004197343206033111, "time": 0.8385738372802735, "epoch": 10, "memory": 25074, "step": 56922} +{"lr": 0.000235504720682845, "data_time": 0.0015700340270996093, "loss": 0.004743618937209248, "time": 0.8169739723205567, "epoch": 10, "memory": 25074, "step": 57022} +{"lr": 0.000235504720682845, "data_time": 0.0015157222747802734, "loss": 0.0010100162122398614, "time": 0.8307600975036621, "epoch": 10, "memory": 25074, "step": 57122} +{"lr": 0.000235504720682845, "data_time": 0.0016313314437866211, "loss": 0.002245061891153455, "time": 0.7947349309921264, "epoch": 10, "memory": 25074, "step": 57222} +{"lr": 0.000235504720682845, "data_time": 0.0012349367141723632, "loss": 0.003983878530561924, "time": 0.7588669776916503, "epoch": 10, "memory": 25074, "step": 57322} +{"lr": 0.000235504720682845, "data_time": 0.0013663291931152344, "loss": 0.0010076561942696572, "time": 0.7645234823226928, "epoch": 10, "memory": 25074, "step": 57422} +{"lr": 0.000235504720682845, "data_time": 0.001473379135131836, "loss": 0.0005049189552664757, "time": 0.7397166967391968, "epoch": 10, "memory": 25074, "step": 57522} +{"accuracy/top1": 99.98737335205078, "data_time": 0.24964468619402716, "time": 0.636816235149608, "step": 10} +{"lr": 0.00020499999999999997, "data_time": 0.0017685174942016601, "loss": 0.001499880850315094, "time": 0.727084732055664, "epoch": 11, "memory": 25074, "step": 57680} +{"lr": 0.00020499999999999997, "data_time": 0.04508950710296631, "loss": 0.003448608098551631, "time": 0.8800601005554199, "epoch": 11, "memory": 25074, "step": 57780} +{"lr": 0.00020499999999999997, "data_time": 0.05101258754730224, "loss": 0.00649166819639504, "time": 0.9064756870269776, "epoch": 11, "memory": 25074, "step": 57880} +{"lr": 0.00020499999999999997, "data_time": 0.0015633106231689453, "loss": 0.000826117442920804, "time": 0.8373608112335205, "epoch": 11, "memory": 25074, "step": 57980} +{"lr": 0.00020499999999999997, "data_time": 0.0015419483184814452, "loss": 0.005671384185552597, "time": 0.9274006366729737, "epoch": 11, "memory": 25074, "step": 58080} +{"lr": 0.00020499999999999997, "data_time": 0.0017696142196655274, "loss": 0.002768180798739195, "time": 0.9979973793029785, "epoch": 11, "memory": 25074, "step": 58180} +{"lr": 0.00020499999999999997, "data_time": 0.0013934850692749023, "loss": 0.0017746952828019858, "time": 0.7385730981826782, "epoch": 11, "memory": 25074, "step": 58280} +{"lr": 0.00020499999999999997, "data_time": 0.0016549110412597656, "loss": 0.003073259023949504, "time": 0.7587138175964355, "epoch": 11, "memory": 25074, "step": 58380} +{"lr": 0.00020499999999999997, "data_time": 0.0019368410110473632, "loss": 0.002911191340535879, "time": 0.880544376373291, "epoch": 11, "memory": 25074, "step": 58480} +{"lr": 0.00020499999999999997, "data_time": 0.0015068769454956055, "loss": 0.0012459274381399154, "time": 0.7419887065887452, "epoch": 11, "memory": 25074, "step": 58580} +{"lr": 0.00020499999999999997, "data_time": 0.0014966487884521484, "loss": 0.0028540372848510744, "time": 0.7582054615020752, "epoch": 11, "memory": 25074, "step": 58680} +{"lr": 0.00020499999999999997, "data_time": 0.0013843059539794921, "loss": 0.0014779695309698581, "time": 0.7900362730026245, "epoch": 11, "memory": 25074, "step": 58780} +{"lr": 0.00020499999999999997, "data_time": 0.0014931201934814454, "loss": 0.003163971332833171, "time": 0.8209036827087403, "epoch": 11, "memory": 25074, "step": 58880} +{"lr": 0.00020499999999999997, "data_time": 0.0016663551330566406, "loss": 0.005382882244884968, "time": 0.7464561223983764, "epoch": 11, "memory": 25074, "step": 58980} +{"lr": 0.00020499999999999997, "data_time": 0.0816460371017456, "loss": 0.005045345053076744, "time": 0.7894214630126953, "epoch": 11, "memory": 25074, "step": 59080} +{"lr": 0.00020499999999999997, "data_time": 0.0022483587265014647, "loss": 0.0013307729735970498, "time": 0.7491605520248413, "epoch": 11, "memory": 25074, "step": 59180} +{"lr": 0.00020499999999999997, "data_time": 0.0019892215728759765, "loss": 0.0032491808757185935, "time": 0.7896609783172608, "epoch": 11, "memory": 25074, "step": 59280} +{"lr": 0.00020499999999999997, "data_time": 0.001386404037475586, "loss": 0.003408097615465522, "time": 0.7466486930847168, "epoch": 11, "memory": 25074, "step": 59380} +{"lr": 0.00020499999999999997, "data_time": 0.001789402961730957, "loss": 0.002397431433200836, "time": 0.8003311395645142, "epoch": 11, "memory": 25074, "step": 59480} +{"lr": 0.00020499999999999997, "data_time": 0.0017194509506225585, "loss": 0.005724278744310141, "time": 0.7389395952224731, "epoch": 11, "memory": 25074, "step": 59580} +{"lr": 0.00020499999999999997, "data_time": 0.0019245624542236328, "loss": 0.007220759894698859, "time": 0.8425361633300781, "epoch": 11, "memory": 25074, "step": 59680} +{"lr": 0.00020499999999999997, "data_time": 0.0016591072082519532, "loss": 0.002435768349096179, "time": 0.781492805480957, "epoch": 11, "memory": 25074, "step": 59780} +{"lr": 0.00020499999999999997, "data_time": 0.0012711048126220702, "loss": 0.0049605985637754205, "time": 0.746262526512146, "epoch": 11, "memory": 25074, "step": 59880} +{"lr": 0.00020499999999999997, "data_time": 0.0014362573623657227, "loss": 0.00446456135250628, "time": 0.7921473264694214, "epoch": 11, "memory": 25074, "step": 59980} +{"lr": 0.00020499999999999997, "data_time": 0.0013808250427246095, "loss": 0.0026375993620604277, "time": 0.7466942548751831, "epoch": 11, "memory": 25074, "step": 60080} +{"lr": 0.00020499999999999997, "data_time": 0.0015541315078735352, "loss": 0.0015348814893513917, "time": 0.7708494424819946, "epoch": 11, "memory": 25074, "step": 60180} +{"lr": 0.00020499999999999997, "data_time": 0.0014572620391845703, "loss": 0.0024017376359552144, "time": 0.7861558437347412, "epoch": 11, "memory": 25074, "step": 60280} +{"lr": 0.00020499999999999997, "data_time": 0.001779484748840332, "loss": 0.0019975043367594482, "time": 0.8155534029006958, "epoch": 11, "memory": 25074, "step": 60380} +{"lr": 0.00020499999999999997, "data_time": 0.0013263702392578125, "loss": 0.003345109848305583, "time": 0.8130066633224488, "epoch": 11, "memory": 25074, "step": 60480} +{"lr": 0.00020499999999999997, "data_time": 0.001529693603515625, "loss": 0.0024079356342554094, "time": 0.7653929471969605, "epoch": 11, "memory": 25074, "step": 60580} +{"lr": 0.00020499999999999997, "data_time": 0.0014828205108642577, "loss": 0.0012781056109815836, "time": 0.779039716720581, "epoch": 11, "memory": 25074, "step": 60680} +{"lr": 0.00020499999999999997, "data_time": 0.0014978647232055664, "loss": 0.004390385374426842, "time": 0.7408694744110107, "epoch": 11, "memory": 25074, "step": 60780} +{"lr": 0.00020499999999999997, "data_time": 0.0017479658126831055, "loss": 0.001297913258895278, "time": 0.7407306432723999, "epoch": 11, "memory": 25074, "step": 60880} +{"lr": 0.00020499999999999997, "data_time": 0.0013545989990234376, "loss": 0.0014874675776809453, "time": 0.7500781774520874, "epoch": 11, "memory": 25074, "step": 60980} +{"lr": 0.00020499999999999997, "data_time": 0.0013677597045898438, "loss": 0.0011331463232636453, "time": 0.7901705741882324, "epoch": 11, "memory": 25074, "step": 61080} +{"lr": 0.00020499999999999997, "data_time": 0.0014270782470703126, "loss": 0.001596615230664611, "time": 0.7723257541656494, "epoch": 11, "memory": 25074, "step": 61180} +{"lr": 0.00020499999999999997, "data_time": 0.0018324851989746094, "loss": 0.0022148349322378635, "time": 0.7486151218414306, "epoch": 11, "memory": 25074, "step": 61280} +{"lr": 0.00020499999999999997, "data_time": 0.001596689224243164, "loss": 0.004162867739796639, "time": 0.7519829511642456, "epoch": 11, "memory": 25074, "step": 61380} +{"lr": 0.00020499999999999997, "data_time": 0.001479935646057129, "loss": 0.007598197925835848, "time": 0.7830942630767822, "epoch": 11, "memory": 25074, "step": 61480} +{"lr": 0.00020499999999999997, "data_time": 0.001686239242553711, "loss": 0.0030011987779289483, "time": 0.7706879854202271, "epoch": 11, "memory": 25074, "step": 61580} +{"lr": 0.00020499999999999997, "data_time": 0.001537013053894043, "loss": 0.0005841518752276898, "time": 0.7245378255844116, "epoch": 11, "memory": 25074, "step": 61680} +{"lr": 0.00020499999999999997, "data_time": 0.0013563394546508788, "loss": 0.0015193565282970666, "time": 0.7570759534835816, "epoch": 11, "memory": 25074, "step": 61780} +{"lr": 0.00020499999999999997, "data_time": 0.0016640663146972657, "loss": 0.005667509092018008, "time": 0.7757426023483276, "epoch": 11, "memory": 25074, "step": 61880} +{"lr": 0.00020499999999999997, "data_time": 0.0014312028884887694, "loss": 0.003807988530024886, "time": 0.760835337638855, "epoch": 11, "memory": 25074, "step": 61980} +{"lr": 0.00020499999999999997, "data_time": 0.0015012502670288086, "loss": 0.0013010187074542045, "time": 0.7440502166748046, "epoch": 11, "memory": 25074, "step": 62080} +{"lr": 0.00020499999999999997, "data_time": 0.0013770580291748047, "loss": 0.0009279805701225996, "time": 0.797015643119812, "epoch": 11, "memory": 25074, "step": 62180} +{"lr": 0.00020499999999999997, "data_time": 0.0012821435928344726, "loss": 0.004593444196507335, "time": 0.7248758316040039, "epoch": 11, "memory": 25074, "step": 62280} +{"lr": 0.00020499999999999997, "data_time": 0.0013997793197631837, "loss": 0.005748032545670867, "time": 0.7976506471633911, "epoch": 11, "memory": 25074, "step": 62380} +{"lr": 0.00020499999999999997, "data_time": 0.0014078378677368163, "loss": 0.002443343959748745, "time": 0.762683629989624, "epoch": 11, "memory": 25074, "step": 62480} +{"lr": 0.00020499999999999997, "data_time": 0.0013005256652832032, "loss": 0.001835272926837206, "time": 0.8184296369552613, "epoch": 11, "memory": 25074, "step": 62580} +{"lr": 0.00020499999999999997, "data_time": 0.0015009164810180664, "loss": 0.002549597527831793, "time": 0.8354214906692505, "epoch": 11, "memory": 25074, "step": 62680} +{"lr": 0.00020499999999999997, "data_time": 0.0017477989196777344, "loss": 0.001508195186033845, "time": 0.760907506942749, "epoch": 11, "memory": 25074, "step": 62780} +{"lr": 0.00020499999999999997, "data_time": 0.0017365694046020507, "loss": 0.0010644513182342052, "time": 0.7797685861587524, "epoch": 11, "memory": 25074, "step": 62880} +{"lr": 0.00020499999999999997, "data_time": 0.0014548778533935546, "loss": 0.0006180366966873407, "time": 0.7651755094528199, "epoch": 11, "memory": 25074, "step": 62980} +{"lr": 0.00020499999999999997, "data_time": 0.0014676332473754882, "loss": 0.005244012922048569, "time": 0.7664060354232788, "epoch": 11, "memory": 25074, "step": 63080} +{"lr": 0.00020499999999999997, "data_time": 0.00142364501953125, "loss": 0.0013581722043454647, "time": 0.7844285726547241, "epoch": 11, "memory": 25074, "step": 63180} +{"lr": 0.00020499999999999997, "data_time": 0.0013538122177124024, "loss": 0.0014689499512314796, "time": 0.7780123233795166, "epoch": 11, "memory": 25074, "step": 63280} +{"accuracy/top1": 99.98737335205078, "data_time": 0.23662527869729436, "time": 0.6224999007056741, "step": 11} +{"lr": 0.000174495279317155, "data_time": 0.0013837575912475585, "loss": 0.0005901526659727096, "time": 0.7773324966430664, "epoch": 12, "memory": 25074, "step": 63438} +{"lr": 0.000174495279317155, "data_time": 0.0018453359603881835, "loss": 0.0027551088016480206, "time": 0.7789901971817017, "epoch": 12, "memory": 25074, "step": 63538} +{"lr": 0.000174495279317155, "data_time": 0.05396201610565186, "loss": 0.0004269802011549473, "time": 0.7833281755447388, "epoch": 12, "memory": 25074, "step": 63638} +{"lr": 0.000174495279317155, "data_time": 0.10586466789245605, "loss": 0.002240213146433234, "time": 0.8102561473846436, "epoch": 12, "memory": 25074, "step": 63738} +{"lr": 0.000174495279317155, "data_time": 0.05591053962707519, "loss": 0.002945806086063385, "time": 0.7739051580429077, "epoch": 12, "memory": 25074, "step": 63838} +{"lr": 0.000174495279317155, "data_time": 0.0248889684677124, "loss": 0.0026704743038862945, "time": 0.7308327198028565, "epoch": 12, "memory": 25074, "step": 63938} +{"lr": 0.000174495279317155, "data_time": 0.053395533561706544, "loss": 0.0005782907828688622, "time": 0.7611489295959473, "epoch": 12, "memory": 25074, "step": 64038} +{"lr": 0.000174495279317155, "data_time": 0.04669842720031738, "loss": 0.0008660111576318741, "time": 0.751051378250122, "epoch": 12, "memory": 25074, "step": 64138} +{"lr": 0.000174495279317155, "data_time": 0.037481832504272464, "loss": 0.0012844897340983153, "time": 0.7554550170898438, "epoch": 12, "memory": 25074, "step": 64238} +{"lr": 0.000174495279317155, "data_time": 0.07840168476104736, "loss": 0.0029215611051768065, "time": 0.7859464406967163, "epoch": 12, "memory": 25074, "step": 64338} +{"lr": 0.000174495279317155, "data_time": 0.0528986930847168, "loss": 0.0016341653652489185, "time": 0.768789267539978, "epoch": 12, "memory": 25074, "step": 64438} +{"lr": 0.000174495279317155, "data_time": 0.04934244155883789, "loss": 0.0030007160268723963, "time": 0.7535741090774536, "epoch": 12, "memory": 25074, "step": 64538} +{"lr": 0.000174495279317155, "data_time": 0.0014331817626953125, "loss": 0.0011381905525922776, "time": 0.7555559158325196, "epoch": 12, "memory": 25074, "step": 64638} +{"lr": 0.000174495279317155, "data_time": 0.001530933380126953, "loss": 0.0007664205972105264, "time": 0.7934005498886109, "epoch": 12, "memory": 25074, "step": 64738} +{"lr": 0.000174495279317155, "data_time": 0.001383829116821289, "loss": 0.0023317751474678516, "time": 0.7839012861251831, "epoch": 12, "memory": 25074, "step": 64838} +{"lr": 0.000174495279317155, "data_time": 0.0014744281768798828, "loss": 0.00024874787777662277, "time": 0.748563265800476, "epoch": 12, "memory": 25074, "step": 64938} +{"lr": 0.000174495279317155, "data_time": 0.001970839500427246, "loss": 0.0013773555867373944, "time": 0.8462858915328979, "epoch": 12, "memory": 25074, "step": 65038} +{"lr": 0.000174495279317155, "data_time": 0.0015425682067871094, "loss": 0.0005728702992200852, "time": 0.767841124534607, "epoch": 12, "memory": 25074, "step": 65138} +{"lr": 0.000174495279317155, "data_time": 0.0014516592025756836, "loss": 0.0033644781913608314, "time": 0.7692043781280518, "epoch": 12, "memory": 25074, "step": 65238} +{"lr": 0.000174495279317155, "data_time": 0.0015937566757202148, "loss": 0.0038938840385526418, "time": 0.8048548936843872, "epoch": 12, "memory": 25074, "step": 65338} +{"lr": 0.000174495279317155, "data_time": 0.0014206886291503907, "loss": 0.0023420528043061493, "time": 0.7502812623977662, "epoch": 12, "memory": 25074, "step": 65438} +{"lr": 0.000174495279317155, "data_time": 0.0013887882232666016, "loss": 0.0012345291674137116, "time": 0.7603619337081909, "epoch": 12, "memory": 25074, "step": 65538} +{"lr": 0.000174495279317155, "data_time": 0.0014254331588745117, "loss": 0.0016018415335565805, "time": 0.7409027338027954, "epoch": 12, "memory": 25074, "step": 65638} +{"lr": 0.000174495279317155, "data_time": 0.001643204689025879, "loss": 0.0014056430663913488, "time": 0.817690110206604, "epoch": 12, "memory": 25074, "step": 65738} +{"lr": 0.000174495279317155, "data_time": 0.0018610000610351563, "loss": 0.0036975953727960585, "time": 0.7444644927978515, "epoch": 12, "memory": 25074, "step": 65838} +{"lr": 0.000174495279317155, "data_time": 0.0016638755798339844, "loss": 0.0013773952610790729, "time": 0.7503877639770508, "epoch": 12, "memory": 25074, "step": 65938} +{"lr": 0.000174495279317155, "data_time": 0.0016218662261962891, "loss": 0.0012711997609585524, "time": 0.7381119728088379, "epoch": 12, "memory": 25074, "step": 66038} +{"lr": 0.000174495279317155, "data_time": 0.0013531923294067382, "loss": 0.0026980101596564055, "time": 0.7505352973937989, "epoch": 12, "memory": 25074, "step": 66138} +{"lr": 0.000174495279317155, "data_time": 0.0013225317001342774, "loss": 0.003637139638885856, "time": 0.7412928581237793, "epoch": 12, "memory": 25074, "step": 66238} +{"lr": 0.000174495279317155, "data_time": 0.0013952970504760743, "loss": 0.000229179160669446, "time": 0.8201618909835815, "epoch": 12, "memory": 25074, "step": 66338} +{"lr": 0.000174495279317155, "data_time": 0.0016584396362304688, "loss": 0.004031705809757114, "time": 0.854230260848999, "epoch": 12, "memory": 25074, "step": 66438} +{"lr": 0.000174495279317155, "data_time": 0.0015131711959838867, "loss": 0.0008632448501884937, "time": 0.7685657739639282, "epoch": 12, "memory": 25074, "step": 66538} +{"lr": 0.000174495279317155, "data_time": 0.0013129234313964844, "loss": 0.000890235137194395, "time": 0.7899316787719727, "epoch": 12, "memory": 25074, "step": 66638} +{"lr": 0.000174495279317155, "data_time": 0.0015313148498535157, "loss": 0.00043141678906977177, "time": 0.8040109872817993, "epoch": 12, "memory": 25074, "step": 66738} +{"lr": 0.000174495279317155, "data_time": 0.0014068365097045898, "loss": 0.0051310168579220775, "time": 0.767658805847168, "epoch": 12, "memory": 25074, "step": 66838} +{"lr": 0.000174495279317155, "data_time": 0.0013263225555419922, "loss": 0.004541805619373918, "time": 0.7757488489151001, "epoch": 12, "memory": 25074, "step": 66938} +{"lr": 0.000174495279317155, "data_time": 0.0014766693115234376, "loss": 0.0023615763522684572, "time": 0.781972599029541, "epoch": 12, "memory": 25074, "step": 67038} +{"lr": 0.000174495279317155, "data_time": 0.0014945268630981445, "loss": 0.004515993222594261, "time": 0.8002325534820557, "epoch": 12, "memory": 25074, "step": 67138} +{"lr": 0.000174495279317155, "data_time": 0.0015024900436401366, "loss": 0.008942172070965171, "time": 0.8138460636138916, "epoch": 12, "memory": 25074, "step": 67238} +{"lr": 0.000174495279317155, "data_time": 0.001655268669128418, "loss": 0.00155441677197814, "time": 0.7601854801177979, "epoch": 12, "memory": 25074, "step": 67338} +{"lr": 0.000174495279317155, "data_time": 0.0016769647598266601, "loss": 0.0015118430368602277, "time": 0.8197365522384643, "epoch": 12, "memory": 25074, "step": 67438} +{"lr": 0.000174495279317155, "data_time": 0.0015154123306274415, "loss": 0.0025519861839711664, "time": 0.7984761953353882, "epoch": 12, "memory": 25074, "step": 67538} +{"lr": 0.000174495279317155, "data_time": 0.0014311313629150391, "loss": 0.004107937542721629, "time": 0.7772931098937989, "epoch": 12, "memory": 25074, "step": 67638} +{"lr": 0.000174495279317155, "data_time": 0.0019978761672973635, "loss": 0.0005500425584614276, "time": 0.7709338903427124, "epoch": 12, "memory": 25074, "step": 67738} +{"lr": 0.000174495279317155, "data_time": 0.001685023307800293, "loss": 0.002001170115545392, "time": 0.7850199937820435, "epoch": 12, "memory": 25074, "step": 67838} +{"lr": 0.000174495279317155, "data_time": 0.0013102054595947265, "loss": 0.0009745461866259575, "time": 0.7552722215652465, "epoch": 12, "memory": 25074, "step": 67938} +{"lr": 0.000174495279317155, "data_time": 0.0014890670776367188, "loss": 0.0018088762648403645, "time": 0.7943161964416504, "epoch": 12, "memory": 25074, "step": 68038} +{"lr": 0.000174495279317155, "data_time": 0.0015288829803466798, "loss": 0.0018784672021865845, "time": 0.782120680809021, "epoch": 12, "memory": 25074, "step": 68138} +{"lr": 0.000174495279317155, "data_time": 0.001427030563354492, "loss": 0.003732118383049965, "time": 0.7712553977966309, "epoch": 12, "memory": 25074, "step": 68238} +{"lr": 0.000174495279317155, "data_time": 0.0013137578964233399, "loss": 0.0010923946276307106, "time": 0.8313997745513916, "epoch": 12, "memory": 25074, "step": 68338} +{"lr": 0.000174495279317155, "data_time": 0.001332569122314453, "loss": 0.0035476938355714083, "time": 0.8069141387939454, "epoch": 12, "memory": 25074, "step": 68438} +{"lr": 0.000174495279317155, "data_time": 0.0022151947021484377, "loss": 0.0010505390353500843, "time": 0.839823317527771, "epoch": 12, "memory": 25074, "step": 68538} +{"lr": 0.000174495279317155, "data_time": 0.0016594409942626953, "loss": 0.0003214818425476551, "time": 0.798504638671875, "epoch": 12, "memory": 25074, "step": 68638} +{"lr": 0.000174495279317155, "data_time": 0.0015073060989379884, "loss": 0.0012518554460257292, "time": 0.8257957696914673, "epoch": 12, "memory": 25074, "step": 68738} +{"lr": 0.000174495279317155, "data_time": 0.0016289472579956055, "loss": 0.0008868366945534945, "time": 0.7688867568969726, "epoch": 12, "memory": 25074, "step": 68838} +{"lr": 0.000174495279317155, "data_time": 0.0017962932586669921, "loss": 0.0007607958279550076, "time": 0.7945826768875122, "epoch": 12, "memory": 25074, "step": 68938} +{"lr": 0.000174495279317155, "data_time": 0.0019196033477783202, "loss": 0.002332293661311269, "time": 0.7335556507110595, "epoch": 12, "memory": 25074, "step": 69038} +{"accuracy/top1": 99.99368286132812, "data_time": 0.24987386254703298, "time": 0.6355440616607666, "step": 12} +{"lr": 0.00014474168609688524, "data_time": 0.07275240421295166, "loss": 0.0009204788133502006, "time": 0.8508753061294556, "epoch": 13, "memory": 25074, "step": 69196} +{"lr": 0.00014474168609688524, "data_time": 0.002440643310546875, "loss": 0.0006284591276198625, "time": 0.763164472579956, "epoch": 13, "memory": 25074, "step": 69296} +{"lr": 0.00014474168609688524, "data_time": 0.05001986026763916, "loss": 0.00031299563124775887, "time": 0.7638823509216308, "epoch": 13, "memory": 25074, "step": 69396} +{"lr": 0.00014474168609688524, "data_time": 0.024952101707458495, "loss": 0.0012822882272303104, "time": 0.7330520391464234, "epoch": 13, "memory": 25074, "step": 69496} +{"lr": 0.00014474168609688524, "data_time": 0.10140235424041748, "loss": 0.002310385042801499, "time": 0.8057371854782105, "epoch": 13, "memory": 25074, "step": 69596} +{"lr": 0.00014474168609688524, "data_time": 0.001494145393371582, "loss": 0.0022555193398147823, "time": 0.7540436029434204, "epoch": 13, "memory": 25074, "step": 69696} +{"lr": 0.00014474168609688524, "data_time": 0.001599574089050293, "loss": 0.0020067887380719185, "time": 0.7368262767791748, "epoch": 13, "memory": 25074, "step": 69796} +{"lr": 0.00014474168609688524, "data_time": 0.0016867399215698242, "loss": 0.001377742551267147, "time": 0.7241564750671386, "epoch": 13, "memory": 25074, "step": 69896} +{"lr": 0.00014474168609688524, "data_time": 0.0014216899871826172, "loss": 0.00044514876790344713, "time": 0.7239116907119751, "epoch": 13, "memory": 25074, "step": 69996} +{"lr": 0.00014474168609688524, "data_time": 0.0018394708633422852, "loss": 0.001726783299818635, "time": 0.7723653793334961, "epoch": 13, "memory": 25074, "step": 70096} +{"lr": 0.00014474168609688524, "data_time": 0.0017693996429443359, "loss": 0.0007506433874368667, "time": 0.7634341955184937, "epoch": 13, "memory": 25074, "step": 70196} +{"lr": 0.00014474168609688524, "data_time": 0.0022705554962158202, "loss": 0.0024152600672096014, "time": 0.7472733497619629, "epoch": 13, "memory": 25074, "step": 70296} +{"lr": 0.00014474168609688524, "data_time": 0.0021012544631958006, "loss": 0.003978089103475213, "time": 0.7653134346008301, "epoch": 13, "memory": 25074, "step": 70396} +{"lr": 0.00014474168609688524, "data_time": 0.0019349336624145507, "loss": 0.001471764873713255, "time": 0.7781517028808593, "epoch": 13, "memory": 25074, "step": 70496} +{"lr": 0.00014474168609688524, "data_time": 0.0021084070205688475, "loss": 0.0016972984187304974, "time": 0.7192719936370849, "epoch": 13, "memory": 25074, "step": 70596} +{"lr": 0.00014474168609688524, "data_time": 0.0018044710159301758, "loss": 0.004171270085498691, "time": 0.7994264841079712, "epoch": 13, "memory": 25074, "step": 70696} +{"lr": 0.00014474168609688524, "data_time": 0.0025376796722412108, "loss": 0.0013212434481829406, "time": 0.7311803102493286, "epoch": 13, "memory": 25074, "step": 70796} +{"lr": 0.00014474168609688524, "data_time": 0.001797652244567871, "loss": 0.0005765300709754228, "time": 0.7530529737472534, "epoch": 13, "memory": 25074, "step": 70896} +{"lr": 0.00014474168609688524, "data_time": 0.002409029006958008, "loss": 0.0020185355097055435, "time": 0.7860418796539307, "epoch": 13, "memory": 25074, "step": 70996} +{"lr": 0.00014474168609688524, "data_time": 0.0017685890197753906, "loss": 0.0007795645389705897, "time": 0.7633878469467164, "epoch": 13, "memory": 25074, "step": 71096} +{"lr": 0.00014474168609688524, "data_time": 0.0022638320922851564, "loss": 0.0016027627978473901, "time": 0.7482191562652588, "epoch": 13, "memory": 25074, "step": 71196} +{"lr": 0.00014474168609688524, "data_time": 0.0017683744430541993, "loss": 0.0003004889469593763, "time": 0.7662945032119751, "epoch": 13, "memory": 25074, "step": 71296} +{"lr": 0.00014474168609688524, "data_time": 0.0013980627059936523, "loss": 0.0008117327932268381, "time": 0.785521125793457, "epoch": 13, "memory": 25074, "step": 71396} +{"lr": 0.00014474168609688524, "data_time": 0.0014436960220336914, "loss": 0.001178226713091135, "time": 0.7426433801651001, "epoch": 13, "memory": 25074, "step": 71496} +{"lr": 0.00014474168609688524, "data_time": 0.0012624263763427734, "loss": 0.0030744282994419336, "time": 0.795448112487793, "epoch": 13, "memory": 25074, "step": 71596} +{"lr": 0.00014474168609688524, "data_time": 0.0018224477767944335, "loss": 0.0013457435183227063, "time": 0.7912577390670776, "epoch": 13, "memory": 25074, "step": 71696} +{"lr": 0.00014474168609688524, "data_time": 0.001501178741455078, "loss": 0.0009950317442417144, "time": 0.7682091474533081, "epoch": 13, "memory": 25074, "step": 71796} +{"lr": 0.00014474168609688524, "data_time": 0.0017308712005615235, "loss": 0.0005844507366418838, "time": 0.7579143285751343, "epoch": 13, "memory": 25074, "step": 71896} +{"lr": 0.00014474168609688524, "data_time": 0.00167236328125, "loss": 0.0026319696102291346, "time": 0.7687328815460205, "epoch": 13, "memory": 25074, "step": 71996} +{"lr": 0.00014474168609688524, "data_time": 0.0015536308288574218, "loss": 0.003094658488407731, "time": 0.7707023620605469, "epoch": 13, "memory": 25074, "step": 72096} +{"lr": 0.00014474168609688524, "data_time": 0.0013475418090820312, "loss": 0.002535417955368757, "time": 0.7957300186157227, "epoch": 13, "memory": 25074, "step": 72196} +{"lr": 0.00014474168609688524, "data_time": 0.0013477325439453125, "loss": 0.001134424051269889, "time": 0.7364935398101806, "epoch": 13, "memory": 25074, "step": 72296} +{"lr": 0.00014474168609688524, "data_time": 0.0015645503997802734, "loss": 0.0019105600658804178, "time": 0.7628180265426636, "epoch": 13, "memory": 25074, "step": 72396} +{"lr": 0.00014474168609688524, "data_time": 0.0017191171646118164, "loss": 0.0021224543917924164, "time": 0.7673990488052368, "epoch": 13, "memory": 25074, "step": 72496} +{"lr": 0.00014474168609688524, "data_time": 0.0015280008316040038, "loss": 0.0007678441237658262, "time": 0.7265167236328125, "epoch": 13, "memory": 25074, "step": 72596} +{"lr": 0.00014474168609688524, "data_time": 0.0014240503311157226, "loss": 0.0026364439632743596, "time": 0.7296411275863648, "epoch": 13, "memory": 25074, "step": 72696} +{"lr": 0.00014474168609688524, "data_time": 0.001578044891357422, "loss": 0.001819831971079111, "time": 0.7435076236724854, "epoch": 13, "memory": 25074, "step": 72796} +{"lr": 0.00014474168609688524, "data_time": 0.0013093948364257812, "loss": 0.0004806566052138805, "time": 0.7049421310424805, "epoch": 13, "memory": 25074, "step": 72896} +{"lr": 0.00014474168609688524, "data_time": 0.0013599395751953125, "loss": 0.001344196079298854, "time": 0.7304930210113525, "epoch": 13, "memory": 25074, "step": 72996} +{"lr": 0.00014474168609688524, "data_time": 0.0015346765518188476, "loss": 0.0008602182846516371, "time": 0.7381244897842407, "epoch": 13, "memory": 25074, "step": 73096} +{"lr": 0.00014474168609688524, "data_time": 0.0015476226806640625, "loss": 0.0038446308579295875, "time": 0.7094938516616821, "epoch": 13, "memory": 25074, "step": 73196} +{"lr": 0.00014474168609688524, "data_time": 0.0014039039611816405, "loss": 0.0012205264065414667, "time": 0.7096496820449829, "epoch": 13, "memory": 25074, "step": 73296} +{"lr": 0.00014474168609688524, "data_time": 0.0014803409576416016, "loss": 0.0006892472971230746, "time": 0.7412248611450195, "epoch": 13, "memory": 25074, "step": 73396} +{"lr": 0.00014474168609688524, "data_time": 0.0016997575759887696, "loss": 0.005097674299031496, "time": 0.7486302375793457, "epoch": 13, "memory": 25074, "step": 73496} +{"lr": 0.00014474168609688524, "data_time": 0.001591658592224121, "loss": 0.000147221889346838, "time": 0.7732123136520386, "epoch": 13, "memory": 25074, "step": 73596} +{"lr": 0.00014474168609688524, "data_time": 0.0015129804611206054, "loss": 0.002005761070176959, "time": 0.7478807449340821, "epoch": 13, "memory": 25074, "step": 73696} +{"lr": 0.00014474168609688524, "data_time": 0.0013300657272338867, "loss": 0.0010326614137738943, "time": 0.7430232763290405, "epoch": 13, "memory": 25074, "step": 73796} +{"lr": 0.00014474168609688524, "data_time": 0.0015828371047973632, "loss": 0.0009551784489303828, "time": 0.7274349689483642, "epoch": 13, "memory": 25074, "step": 73896} +{"lr": 0.00014474168609688524, "data_time": 0.0015261173248291016, "loss": 0.001108011044561863, "time": 0.7866940259933471, "epoch": 13, "memory": 25074, "step": 73996} +{"lr": 0.00014474168609688524, "data_time": 0.0013470888137817384, "loss": 0.0027155133429914713, "time": 0.7083929061889649, "epoch": 13, "memory": 25074, "step": 74096} +{"lr": 0.00014474168609688524, "data_time": 0.0017130374908447266, "loss": 0.003259146912023425, "time": 0.8713862895965576, "epoch": 13, "memory": 25074, "step": 74196} +{"lr": 0.00014474168609688524, "data_time": 0.0014954566955566405, "loss": 0.003147553373128176, "time": 0.7065548181533814, "epoch": 13, "memory": 25074, "step": 74296} +{"lr": 0.00014474168609688524, "data_time": 0.0012180089950561523, "loss": 0.0007164672948420048, "time": 0.7848201036453247, "epoch": 13, "memory": 25074, "step": 74396} +{"lr": 0.00014474168609688524, "data_time": 0.0015608549118041992, "loss": 0.00037650628946721553, "time": 0.7286721706390381, "epoch": 13, "memory": 25074, "step": 74496} +{"lr": 0.00014474168609688524, "data_time": 0.0016849040985107422, "loss": 0.0008910450153052807, "time": 0.7445698022842407, "epoch": 13, "memory": 25074, "step": 74596} +{"lr": 0.00014474168609688524, "data_time": 0.0016117095947265625, "loss": 0.005756586650386452, "time": 0.7053746461868287, "epoch": 13, "memory": 25074, "step": 74696} +{"lr": 0.00014474168609688524, "data_time": 0.0014872550964355469, "loss": 0.0018001841846853495, "time": 0.7206930875778198, "epoch": 13, "memory": 25074, "step": 74796} +{"accuracy/top1": 99.99368286132812, "data_time": 0.2428590409895953, "time": 0.6288780184353099, "step": 13} +{"lr": 0.00011647185255078837, "data_time": 0.023251724243164063, "loss": 0.0021779678761959078, "time": 0.77837233543396, "epoch": 14, "memory": 25074, "step": 74954} +{"lr": 0.00011647185255078837, "data_time": 0.0015166044235229493, "loss": 0.0023673285264521836, "time": 0.7371764421463013, "epoch": 14, "memory": 25074, "step": 75054} +{"lr": 0.00011647185255078837, "data_time": 0.0015041828155517578, "loss": 0.0009057887364178896, "time": 0.7058263063430786, "epoch": 14, "memory": 25074, "step": 75154} +{"lr": 0.00011647185255078837, "data_time": 0.0018524408340454101, "loss": 0.0011784181464463473, "time": 0.718956470489502, "epoch": 14, "memory": 25074, "step": 75254} +{"lr": 0.00011647185255078837, "data_time": 0.04562981128692627, "loss": 0.001378782605752349, "time": 0.749661135673523, "epoch": 14, "memory": 25074, "step": 75354} +{"lr": 0.00011647185255078837, "data_time": 0.001342034339904785, "loss": 0.0030419050250202417, "time": 0.7059260845184326, "epoch": 14, "memory": 25074, "step": 75454} +{"lr": 0.00011647185255078837, "data_time": 0.014382219314575196, "loss": 0.0014209263026714325, "time": 0.7180840253829956, "epoch": 14, "memory": 25074, "step": 75554} +{"lr": 0.00011647185255078837, "data_time": 0.042363214492797854, "loss": 0.001219785725697875, "time": 0.7468163251876831, "epoch": 14, "memory": 25074, "step": 75654} +{"lr": 0.00011647185255078837, "data_time": 0.0016139745712280273, "loss": 9.78053081780672e-05, "time": 0.71507728099823, "epoch": 14, "memory": 25074, "step": 75754} +{"lr": 0.00011647185255078837, "data_time": 0.0015370845794677734, "loss": 0.0037169348448514937, "time": 0.7629705905914307, "epoch": 14, "memory": 25074, "step": 75854} +{"lr": 0.00011647185255078837, "data_time": 0.0018133878707885741, "loss": 0.0007813195697963238, "time": 0.7463514328002929, "epoch": 14, "memory": 25074, "step": 75954} +{"lr": 0.00011647185255078837, "data_time": 0.002047610282897949, "loss": 0.0010452068410813809, "time": 0.7900663137435913, "epoch": 14, "memory": 25074, "step": 76054} +{"lr": 0.00011647185255078837, "data_time": 0.001435399055480957, "loss": 0.000550618302077055, "time": 0.7513041019439697, "epoch": 14, "memory": 25074, "step": 76154} +{"lr": 0.00011647185255078837, "data_time": 0.0014093160629272462, "loss": 0.0002473832108080387, "time": 0.7352958440780639, "epoch": 14, "memory": 25074, "step": 76254} +{"lr": 0.00011647185255078837, "data_time": 0.0015857219696044922, "loss": 0.002685864083468914, "time": 0.7190948486328125, "epoch": 14, "memory": 25074, "step": 76354} +{"lr": 0.00011647185255078837, "data_time": 0.0014364242553710938, "loss": 0.00013651251792907716, "time": 0.7147719383239746, "epoch": 14, "memory": 25074, "step": 76454} +{"lr": 0.00011647185255078837, "data_time": 0.0014735937118530273, "loss": 0.00056815380230546, "time": 0.715182113647461, "epoch": 14, "memory": 25074, "step": 76554} +{"lr": 0.00011647185255078837, "data_time": 0.0016825437545776368, "loss": 0.0004540197551250458, "time": 0.7560132741928101, "epoch": 14, "memory": 25074, "step": 76654} +{"lr": 0.00011647185255078837, "data_time": 0.0014857769012451172, "loss": 0.0006996762473136187, "time": 0.727306318283081, "epoch": 14, "memory": 25074, "step": 76754} +{"lr": 0.00011647185255078837, "data_time": 0.0013864278793334962, "loss": 0.0016852287575602532, "time": 0.7824569702148437, "epoch": 14, "memory": 25074, "step": 76854} +{"lr": 0.00011647185255078837, "data_time": 0.0014194726943969726, "loss": 0.000279172183945775, "time": 0.7624621152877807, "epoch": 14, "memory": 25074, "step": 76954} +{"lr": 0.00011647185255078837, "data_time": 0.0013602256774902343, "loss": 0.00285165524110198, "time": 0.806310510635376, "epoch": 14, "memory": 25074, "step": 77054} +{"lr": 0.00011647185255078837, "data_time": 0.0013950347900390625, "loss": 0.0007428525947034359, "time": 0.786906099319458, "epoch": 14, "memory": 25074, "step": 77154} +{"lr": 0.00011647185255078837, "data_time": 0.0013521194458007812, "loss": 0.0013474670238792896, "time": 0.8502303838729859, "epoch": 14, "memory": 25074, "step": 77254} +{"lr": 0.00011647185255078837, "data_time": 0.0014919996261596679, "loss": 0.0014406180940568446, "time": 0.7569593906402587, "epoch": 14, "memory": 25074, "step": 77354} +{"lr": 0.00011647185255078837, "data_time": 0.0013175487518310546, "loss": 0.0016626224853098393, "time": 0.8024927616119385, "epoch": 14, "memory": 25074, "step": 77454} +{"lr": 0.00011647185255078837, "data_time": 0.0014471292495727539, "loss": 0.0018643232993781566, "time": 0.8707294225692749, "epoch": 14, "memory": 25074, "step": 77554} +{"lr": 0.00011647185255078837, "data_time": 0.001388382911682129, "loss": 0.001479786215350032, "time": 0.8227628469467163, "epoch": 14, "memory": 25074, "step": 77654} +{"lr": 0.00011647185255078837, "data_time": 0.0014124631881713868, "loss": 0.0031971388030797242, "time": 0.7924062252044678, "epoch": 14, "memory": 25074, "step": 77754} +{"lr": 0.00011647185255078837, "data_time": 0.0013155460357666016, "loss": 0.001103545445948839, "time": 0.7877817392349243, "epoch": 14, "memory": 25074, "step": 77854} +{"lr": 0.00011647185255078837, "data_time": 0.001440286636352539, "loss": 0.0007647598162293434, "time": 0.7607779026031494, "epoch": 14, "memory": 25074, "step": 77954} +{"lr": 0.00011647185255078837, "data_time": 0.0014202356338500976, "loss": 0.0008859530556946993, "time": 0.8226583242416382, "epoch": 14, "memory": 25074, "step": 78054} +{"lr": 0.00011647185255078837, "data_time": 0.0016968488693237306, "loss": 0.00011167400516569615, "time": 0.7710731029510498, "epoch": 14, "memory": 25074, "step": 78154} +{"lr": 0.00011647185255078837, "data_time": 0.0014856338500976562, "loss": 0.006312939804047346, "time": 0.7489454030990601, "epoch": 14, "memory": 25074, "step": 78254} +{"lr": 0.00011647185255078837, "data_time": 0.0014198780059814452, "loss": 0.0017056366894394159, "time": 0.7752131462097168, "epoch": 14, "memory": 25074, "step": 78354} +{"lr": 0.00011647185255078837, "data_time": 0.0014912605285644532, "loss": 0.002734039630740881, "time": 0.7770315408706665, "epoch": 14, "memory": 25074, "step": 78454} +{"lr": 0.00011647185255078837, "data_time": 0.0012785911560058594, "loss": 0.00021743196994066237, "time": 0.7677789211273194, "epoch": 14, "memory": 25074, "step": 78554} +{"lr": 0.00011647185255078837, "data_time": 0.0016590356826782227, "loss": 0.0027144406456500294, "time": 0.7268472194671631, "epoch": 14, "memory": 25074, "step": 78654} +{"lr": 0.00011647185255078837, "data_time": 0.001490473747253418, "loss": 0.0022927312646061184, "time": 0.7710299730300904, "epoch": 14, "memory": 25074, "step": 78754} +{"lr": 0.00011647185255078837, "data_time": 0.001482844352722168, "loss": 0.0015905157662928105, "time": 0.8276148557662963, "epoch": 14, "memory": 25074, "step": 78854} +{"lr": 0.00011647185255078837, "data_time": 0.0015105962753295898, "loss": 0.0009809607174247504, "time": 0.782722020149231, "epoch": 14, "memory": 25074, "step": 78954} +{"lr": 0.00011647185255078837, "data_time": 0.0013812780380249023, "loss": 0.0021457645576447248, "time": 0.7340906620025635, "epoch": 14, "memory": 25074, "step": 79054} +{"lr": 0.00011647185255078837, "data_time": 0.0014567136764526366, "loss": 0.0016096735373139382, "time": 0.7725087642669678, "epoch": 14, "memory": 25074, "step": 79154} +{"lr": 0.00011647185255078837, "data_time": 0.0012729644775390625, "loss": 0.003232082212343812, "time": 0.7204482316970825, "epoch": 14, "memory": 25074, "step": 79254} +{"lr": 0.00011647185255078837, "data_time": 0.0015396833419799804, "loss": 0.0014629330486059188, "time": 0.762050724029541, "epoch": 14, "memory": 25074, "step": 79354} +{"lr": 0.00011647185255078837, "data_time": 0.0014620542526245118, "loss": 0.0010259154718369246, "time": 0.7450391292572022, "epoch": 14, "memory": 25074, "step": 79454} +{"lr": 0.00011647185255078837, "data_time": 0.0015154123306274415, "loss": 0.00039701308123767375, "time": 0.757326078414917, "epoch": 14, "memory": 25074, "step": 79554} +{"lr": 0.00011647185255078837, "data_time": 0.0014201641082763673, "loss": 0.00040809502825140954, "time": 0.7113478899002075, "epoch": 14, "memory": 25074, "step": 79654} +{"lr": 0.00011647185255078837, "data_time": 0.0014184474945068359, "loss": 0.0016616976354271173, "time": 0.7391294717788697, "epoch": 14, "memory": 25074, "step": 79754} +{"lr": 0.00011647185255078837, "data_time": 0.0013500213623046874, "loss": 0.0005124249029904604, "time": 0.8219952344894409, "epoch": 14, "memory": 25074, "step": 79854} +{"lr": 0.00011647185255078837, "data_time": 0.0013647079467773438, "loss": 0.00026878686621785165, "time": 0.7242864847183228, "epoch": 14, "memory": 25074, "step": 79954} +{"lr": 0.00011647185255078837, "data_time": 0.0013128995895385742, "loss": 0.0007196391932666301, "time": 0.7184814453125, "epoch": 14, "memory": 25074, "step": 80054} +{"lr": 0.00011647185255078837, "data_time": 0.0013649463653564453, "loss": 0.00015480811707675458, "time": 0.7506291389465332, "epoch": 14, "memory": 25074, "step": 80154} +{"lr": 0.00011647185255078837, "data_time": 0.0014166116714477539, "loss": 0.00039838752709329126, "time": 0.740925645828247, "epoch": 14, "memory": 25074, "step": 80254} +{"lr": 0.00011647185255078837, "data_time": 0.0018710851669311523, "loss": 0.0003164768684655428, "time": 0.7214632987976074, "epoch": 14, "memory": 25074, "step": 80354} +{"lr": 0.00011647185255078837, "data_time": 0.0014228105545043945, "loss": 0.0037071477621793748, "time": 0.7352063894271851, "epoch": 14, "memory": 25074, "step": 80454} +{"lr": 0.00011647185255078837, "data_time": 0.0016200542449951172, "loss": 0.0006053172051906586, "time": 0.7899404525756836, "epoch": 14, "memory": 25074, "step": 80554} +{"accuracy/top1": 99.99368286132812, "data_time": 0.24431033695445342, "time": 0.631132588667028, "step": 14} +{"lr": 9.038187580296775e-05, "data_time": 0.001281404495239258, "loss": 0.0009693463332951068, "time": 0.8083181619644165, "epoch": 15, "memory": 25074, "step": 80712} +{"lr": 9.038187580296775e-05, "data_time": 0.0013797521591186524, "loss": 0.0024553865659981966, "time": 0.7607942581176758, "epoch": 15, "memory": 25074, "step": 80812} +{"lr": 9.038187580296775e-05, "data_time": 0.001313185691833496, "loss": 0.0005513661075383425, "time": 0.782365870475769, "epoch": 15, "memory": 25074, "step": 80912} +{"lr": 9.038187580296775e-05, "data_time": 0.001375126838684082, "loss": 0.0003903736360371113, "time": 0.7232391595840454, "epoch": 15, "memory": 25074, "step": 81012} +{"lr": 9.038187580296775e-05, "data_time": 0.0014269590377807618, "loss": 0.0017582374624907971, "time": 0.8211493015289306, "epoch": 15, "memory": 25074, "step": 81112} +{"lr": 9.038187580296775e-05, "data_time": 0.0016207218170166016, "loss": 0.0016284781508147717, "time": 0.8221693515777588, "epoch": 15, "memory": 25074, "step": 81212} +{"lr": 9.038187580296775e-05, "data_time": 0.0014461755752563476, "loss": 0.0007385155651718379, "time": 0.8281132459640503, "epoch": 15, "memory": 25074, "step": 81312} +{"lr": 9.038187580296775e-05, "data_time": 0.0014560937881469727, "loss": 0.0007029476575553418, "time": 0.7273197174072266, "epoch": 15, "memory": 25074, "step": 81412} +{"lr": 9.038187580296775e-05, "data_time": 0.0018094778060913086, "loss": 0.0013593892101198435, "time": 0.7591881513595581, "epoch": 15, "memory": 25074, "step": 81512} +{"lr": 9.038187580296775e-05, "data_time": 0.003216552734375, "loss": 0.0018323538359254598, "time": 0.7899890184402466, "epoch": 15, "memory": 25074, "step": 81612} +{"lr": 9.038187580296775e-05, "data_time": 0.0020852088928222656, "loss": 0.00044703842140734194, "time": 0.762487530708313, "epoch": 15, "memory": 25074, "step": 81712} +{"lr": 9.038187580296775e-05, "data_time": 0.0015711069107055664, "loss": 0.001037484547123313, "time": 0.7359006643295288, "epoch": 15, "memory": 25074, "step": 81812} +{"lr": 9.038187580296775e-05, "data_time": 0.0015401840209960938, "loss": 0.0020602371543645857, "time": 0.8280802488327026, "epoch": 15, "memory": 25074, "step": 81912} +{"lr": 9.038187580296775e-05, "data_time": 0.0017678499221801757, "loss": 0.0019632921554148196, "time": 0.7969156980514527, "epoch": 15, "memory": 25074, "step": 82012} +{"lr": 9.038187580296775e-05, "data_time": 0.0015585184097290038, "loss": 0.0008575992193073035, "time": 0.7718917846679687, "epoch": 15, "memory": 25074, "step": 82112} +{"lr": 9.038187580296775e-05, "data_time": 0.0015117645263671875, "loss": 0.0021742898039519786, "time": 0.8478286981582641, "epoch": 15, "memory": 25074, "step": 82212} +{"lr": 9.038187580296775e-05, "data_time": 0.0016597747802734376, "loss": 0.00014989529736340046, "time": 0.8146865606307984, "epoch": 15, "memory": 25074, "step": 82312} +{"lr": 9.038187580296775e-05, "data_time": 0.0017172574996948242, "loss": 3.949040547013283e-05, "time": 0.8910597801208496, "epoch": 15, "memory": 25074, "step": 82412} +{"lr": 9.038187580296775e-05, "data_time": 0.0019261360168457032, "loss": 0.00012080157175660133, "time": 0.729236364364624, "epoch": 15, "memory": 25074, "step": 82512} +{"lr": 9.038187580296775e-05, "data_time": 0.0016622304916381835, "loss": 0.0007462635636329651, "time": 0.8027929782867431, "epoch": 15, "memory": 25074, "step": 82612} +{"lr": 9.038187580296775e-05, "data_time": 0.001559901237487793, "loss": 0.001427473733201623, "time": 0.8323099136352539, "epoch": 15, "memory": 25074, "step": 82712} +{"lr": 9.038187580296775e-05, "data_time": 0.0014462471008300781, "loss": 0.0003619640599936247, "time": 0.8937884330749511, "epoch": 15, "memory": 25074, "step": 82812} +{"lr": 9.038187580296775e-05, "data_time": 0.0015408992767333984, "loss": 0.0006396289449185133, "time": 0.7980130910873413, "epoch": 15, "memory": 25074, "step": 82912} +{"lr": 9.038187580296775e-05, "data_time": 0.0015927553176879883, "loss": 0.0003920118324458599, "time": 1.0595411777496337, "epoch": 15, "memory": 25074, "step": 83012} +{"lr": 9.038187580296775e-05, "data_time": 0.0015401601791381835, "loss": 0.0009338007308542729, "time": 0.9077517271041871, "epoch": 15, "memory": 25074, "step": 83112} +{"lr": 9.038187580296775e-05, "data_time": 0.0015286922454833985, "loss": 0.0019225267693400382, "time": 0.9691908359527588, "epoch": 15, "memory": 25074, "step": 83212} +{"lr": 9.038187580296775e-05, "data_time": 0.0018684148788452148, "loss": 0.0013671924360096454, "time": 0.7199590682983399, "epoch": 15, "memory": 25074, "step": 83312} +{"lr": 9.038187580296775e-05, "data_time": 0.0014957904815673829, "loss": 0.00010000835172832012, "time": 0.8423433065414428, "epoch": 15, "memory": 25074, "step": 83412} +{"lr": 9.038187580296775e-05, "data_time": 0.0013504505157470703, "loss": 0.002599572343751788, "time": 0.8672428607940674, "epoch": 15, "memory": 25074, "step": 83512} +{"lr": 9.038187580296775e-05, "data_time": 0.0018329143524169922, "loss": 0.0005764301866292953, "time": 0.8385485410690308, "epoch": 15, "memory": 25074, "step": 83612} +{"lr": 9.038187580296775e-05, "data_time": 0.0014724969863891602, "loss": 0.0009647830855101347, "time": 0.8059179306030273, "epoch": 15, "memory": 25074, "step": 83712} +{"lr": 9.038187580296775e-05, "data_time": 0.0014623641967773438, "loss": 0.0003822332713752985, "time": 1.0916605472564698, "epoch": 15, "memory": 25074, "step": 83812} +{"lr": 9.038187580296775e-05, "data_time": 0.0016545772552490235, "loss": 0.0016348369419574738, "time": 1.1889729022979736, "epoch": 15, "memory": 25074, "step": 83912} +{"lr": 9.038187580296775e-05, "data_time": 0.0013372182846069336, "loss": 0.0008744071703404189, "time": 1.000340986251831, "epoch": 15, "memory": 25074, "step": 84012} +{"lr": 9.038187580296775e-05, "data_time": 0.0031939983367919923, "loss": 0.003211942035704851, "time": 2.5094349861145018, "epoch": 15, "memory": 25074, "step": 84112} +{"lr": 9.038187580296775e-05, "data_time": 0.0013994932174682616, "loss": 0.0005596747156232596, "time": 0.9590139627456665, "epoch": 15, "memory": 25074, "step": 84212} +{"lr": 9.038187580296775e-05, "data_time": 0.0014598846435546875, "loss": 0.0009359739255160093, "time": 1.0496535539627074, "epoch": 15, "memory": 25074, "step": 84312} +{"lr": 9.038187580296775e-05, "data_time": 0.001909017562866211, "loss": 0.00027056271210312843, "time": 0.7710723161697388, "epoch": 15, "memory": 25074, "step": 84412} +{"lr": 9.038187580296775e-05, "data_time": 0.0012775182723999024, "loss": 0.0008024506270885468, "time": 0.7106168985366821, "epoch": 15, "memory": 25074, "step": 84512} +{"lr": 9.038187580296775e-05, "data_time": 0.0013846158981323242, "loss": 0.0017863199580460786, "time": 0.7040702104568481, "epoch": 15, "memory": 25074, "step": 84612} +{"lr": 9.038187580296775e-05, "data_time": 0.0012623786926269532, "loss": 0.0021234650630503893, "time": 0.7278531312942504, "epoch": 15, "memory": 25074, "step": 84712} +{"lr": 9.038187580296775e-05, "data_time": 0.001620650291442871, "loss": 0.0017398543655872344, "time": 0.7522794008255005, "epoch": 15, "memory": 25074, "step": 84812} +{"lr": 9.038187580296775e-05, "data_time": 0.0013030290603637696, "loss": 0.0006522384937852621, "time": 0.7269864320755005, "epoch": 15, "memory": 25074, "step": 84912} +{"lr": 9.038187580296775e-05, "data_time": 0.0014772415161132812, "loss": 0.00027401009574532507, "time": 0.7073343992233276, "epoch": 15, "memory": 25074, "step": 85012} +{"lr": 9.038187580296775e-05, "data_time": 0.0016780614852905274, "loss": 0.000629105232656002, "time": 0.7588959693908691, "epoch": 15, "memory": 25074, "step": 85112} +{"lr": 9.038187580296775e-05, "data_time": 0.001724386215209961, "loss": 0.0001358284614980221, "time": 0.719288420677185, "epoch": 15, "memory": 25074, "step": 85212} +{"lr": 9.038187580296775e-05, "data_time": 0.002022528648376465, "loss": 0.0004343196284025908, "time": 0.7302109241485596, "epoch": 15, "memory": 25074, "step": 85312} +{"lr": 9.038187580296775e-05, "data_time": 0.0014947891235351563, "loss": 0.0002664435654878616, "time": 0.8181253671646118, "epoch": 15, "memory": 25074, "step": 85412} +{"lr": 9.038187580296775e-05, "data_time": 0.001249384880065918, "loss": 0.0016120796091854573, "time": 0.7277424812316895, "epoch": 15, "memory": 25074, "step": 85512} +{"lr": 9.038187580296775e-05, "data_time": 0.001448678970336914, "loss": 0.000722819659858942, "time": 0.7536602020263672, "epoch": 15, "memory": 25074, "step": 85612} +{"lr": 9.038187580296775e-05, "data_time": 0.0014109134674072266, "loss": 0.0005820733960717917, "time": 0.7464721441268921, "epoch": 15, "memory": 25074, "step": 85712} +{"lr": 9.038187580296775e-05, "data_time": 0.0015386819839477539, "loss": 0.0021876913495361807, "time": 0.7061995267868042, "epoch": 15, "memory": 25074, "step": 85812} +{"lr": 9.038187580296775e-05, "data_time": 0.0014644384384155274, "loss": 0.0002771410159766674, "time": 0.7733000755310059, "epoch": 15, "memory": 25074, "step": 85912} +{"lr": 9.038187580296775e-05, "data_time": 0.002461814880371094, "loss": 0.002774343313649297, "time": 0.8090592861175537, "epoch": 15, "memory": 25074, "step": 86012} +{"lr": 9.038187580296775e-05, "data_time": 0.002170419692993164, "loss": 0.00018875827081501483, "time": 0.7727634906768799, "epoch": 15, "memory": 25074, "step": 86112} +{"lr": 9.038187580296775e-05, "data_time": 0.0017713546752929688, "loss": 0.0019611213356256485, "time": 0.7775627374649048, "epoch": 15, "memory": 25074, "step": 86212} +{"lr": 9.038187580296775e-05, "data_time": 0.0011821746826171874, "loss": 0.0011772952042520047, "time": 0.720346188545227, "epoch": 15, "memory": 25074, "step": 86312} +{"accuracy/top1": 99.99368286132812, "data_time": 0.2504125482895795, "time": 0.6420064393211814, "step": 15} +{"lr": 6.711417766862323e-05, "data_time": 0.001391148567199707, "loss": 0.0008806531317532062, "time": 0.770824646949768, "epoch": 16, "memory": 25074, "step": 86470} +{"lr": 6.711417766862323e-05, "data_time": 0.001569509506225586, "loss": 0.0004885589703917503, "time": 0.7555091857910157, "epoch": 16, "memory": 25074, "step": 86570} +{"lr": 6.711417766862323e-05, "data_time": 0.0642164945602417, "loss": 0.00020497986115515233, "time": 0.8285637617111206, "epoch": 16, "memory": 25074, "step": 86670} +{"lr": 6.711417766862323e-05, "data_time": 0.08859062194824219, "loss": 0.0005509410053491593, "time": 0.7917944908142089, "epoch": 16, "memory": 25074, "step": 86770} +{"lr": 6.711417766862323e-05, "data_time": 0.06232173442840576, "loss": 0.0005546720698475837, "time": 0.7812191009521484, "epoch": 16, "memory": 25074, "step": 86870} +{"lr": 6.711417766862323e-05, "data_time": 0.08091220855712891, "loss": 0.0002881082706153393, "time": 0.7860160589218139, "epoch": 16, "memory": 25074, "step": 86970} +{"lr": 6.711417766862323e-05, "data_time": 0.034747886657714847, "loss": 0.0006381961517035961, "time": 0.7529673099517822, "epoch": 16, "memory": 25074, "step": 87070} +{"lr": 6.711417766862323e-05, "data_time": 0.005439329147338867, "loss": 0.0026498822029680015, "time": 0.7797904968261719, "epoch": 16, "memory": 25074, "step": 87170} +{"lr": 6.711417766862323e-05, "data_time": 0.0014828920364379882, "loss": 0.0001123073510825634, "time": 0.739583158493042, "epoch": 16, "memory": 25074, "step": 87270} +{"lr": 6.711417766862323e-05, "data_time": 0.0014094114303588867, "loss": 0.0006772896274924278, "time": 0.7724883794784546, "epoch": 16, "memory": 25074, "step": 87370} +{"lr": 6.711417766862323e-05, "data_time": 0.0015820026397705077, "loss": 0.0003269531764090061, "time": 0.7386952638626099, "epoch": 16, "memory": 25074, "step": 87470} +{"lr": 6.711417766862323e-05, "data_time": 0.0014675140380859375, "loss": 0.0021864949725568295, "time": 0.724583911895752, "epoch": 16, "memory": 25074, "step": 87570} +{"lr": 6.711417766862323e-05, "data_time": 0.0015862464904785156, "loss": 0.001155404606834054, "time": 0.7314566612243653, "epoch": 16, "memory": 25074, "step": 87670} +{"lr": 6.711417766862323e-05, "data_time": 0.0016598939895629884, "loss": 0.00015166825614869593, "time": 0.7534103155136108, "epoch": 16, "memory": 25074, "step": 87770} +{"lr": 6.711417766862323e-05, "data_time": 0.0013062477111816407, "loss": 0.000558520620688796, "time": 0.7059796810150146, "epoch": 16, "memory": 25074, "step": 87870} +{"lr": 6.711417766862323e-05, "data_time": 0.0017880916595458985, "loss": 0.0003085790202021599, "time": 0.7124033689498901, "epoch": 16, "memory": 25074, "step": 87970} +{"lr": 6.711417766862323e-05, "data_time": 0.0014652729034423829, "loss": 0.0005887267179787159, "time": 0.7275777578353881, "epoch": 16, "memory": 25074, "step": 88070} +{"lr": 6.711417766862323e-05, "data_time": 0.0017225265502929688, "loss": 8.797864429652691e-05, "time": 0.7884787797927857, "epoch": 16, "memory": 25074, "step": 88170} +{"lr": 6.711417766862323e-05, "data_time": 0.0012642860412597657, "loss": 0.0013945603277534246, "time": 0.7215691804885864, "epoch": 16, "memory": 25074, "step": 88270} +{"lr": 6.711417766862323e-05, "data_time": 0.0012665033340454102, "loss": 0.001149605493992567, "time": 0.7747556209564209, "epoch": 16, "memory": 25074, "step": 88370} +{"lr": 6.711417766862323e-05, "data_time": 0.0013764619827270508, "loss": 0.00020167063921689988, "time": 0.7620030164718627, "epoch": 16, "memory": 25074, "step": 88470} +{"lr": 6.711417766862323e-05, "data_time": 0.001493072509765625, "loss": 7.639559917151927e-05, "time": 0.7522417306900024, "epoch": 16, "memory": 25074, "step": 88570} +{"lr": 6.711417766862323e-05, "data_time": 0.0015863418579101563, "loss": 0.0002899395301938057, "time": 0.7593885660171509, "epoch": 16, "memory": 25074, "step": 88670} +{"lr": 6.711417766862323e-05, "data_time": 0.0014772891998291015, "loss": 0.0021343606058508158, "time": 0.8097762823104858, "epoch": 16, "memory": 25074, "step": 88770} +{"lr": 6.711417766862323e-05, "data_time": 0.0017124176025390624, "loss": 2.8182798996567725e-05, "time": 0.7633267164230346, "epoch": 16, "memory": 25074, "step": 88870} +{"lr": 6.711417766862323e-05, "data_time": 0.0013273954391479492, "loss": 0.00016270913183689118, "time": 0.7777122497558594, "epoch": 16, "memory": 25074, "step": 88970} +{"lr": 6.711417766862323e-05, "data_time": 0.0015081167221069336, "loss": 0.0032006279565393926, "time": 0.7150451898574829, "epoch": 16, "memory": 25074, "step": 89070} +{"lr": 6.711417766862323e-05, "data_time": 0.0014369487762451172, "loss": 7.198280654847622e-05, "time": 0.769907283782959, "epoch": 16, "memory": 25074, "step": 89170} +{"lr": 6.711417766862323e-05, "data_time": 0.0012654304504394532, "loss": 0.0006047617178410291, "time": 0.7071659326553345, "epoch": 16, "memory": 25074, "step": 89270} +{"lr": 6.711417766862323e-05, "data_time": 0.001788759231567383, "loss": 0.005545500945299864, "time": 0.8838212490081787, "epoch": 16, "memory": 25074, "step": 89370} +{"lr": 6.711417766862323e-05, "data_time": 0.0018588066101074218, "loss": 0.003012413950636983, "time": 0.8784438848495484, "epoch": 16, "memory": 25074, "step": 89470} +{"lr": 6.711417766862323e-05, "data_time": 0.002134108543395996, "loss": 0.001166795054450631, "time": 0.8018247604370117, "epoch": 16, "memory": 25074, "step": 89570} +{"lr": 6.711417766862323e-05, "data_time": 0.001578354835510254, "loss": 0.0001045859418809414, "time": 0.8085804224014282, "epoch": 16, "memory": 25074, "step": 89670} +{"lr": 6.711417766862323e-05, "data_time": 0.0012562990188598633, "loss": 0.0007325883023440837, "time": 0.7591107368469239, "epoch": 16, "memory": 25074, "step": 89770} +{"lr": 6.711417766862323e-05, "data_time": 0.0014090776443481446, "loss": 0.0015640189405530692, "time": 0.7343702554702759, "epoch": 16, "memory": 25074, "step": 89870} +{"lr": 6.711417766862323e-05, "data_time": 0.0015744686126708985, "loss": 0.003829053184017539, "time": 0.7476104259490967, "epoch": 16, "memory": 25074, "step": 89970} +{"lr": 6.711417766862323e-05, "data_time": 0.001384592056274414, "loss": 0.0010552929248660802, "time": 0.7129765033721924, "epoch": 16, "memory": 25074, "step": 90070} diff --git a/convnext_small_4xb256_fake5m-lr4e-4/20230601_025223/vis_data/vis_image/0000000.png_1.png b/convnext_small_4xb256_fake5m-lr4e-4/20230601_025223/vis_data/vis_image/0000000.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..b2aa1aaaf9f97ce8b19f3bbce6b7024e76a7ba4c Binary files /dev/null and b/convnext_small_4xb256_fake5m-lr4e-4/20230601_025223/vis_data/vis_image/0000000.png_1.png differ diff --git a/convnext_small_4xb256_fake5m-lr4e-4/20230601_025223/vis_data/vis_image/0000000.png_10.png b/convnext_small_4xb256_fake5m-lr4e-4/20230601_025223/vis_data/vis_image/0000000.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..6f264824565adf0630fdfc2e331f1ad6f2495172 Binary files /dev/null and b/convnext_small_4xb256_fake5m-lr4e-4/20230601_025223/vis_data/vis_image/0000000.png_10.png differ diff --git a/convnext_small_4xb256_fake5m-lr4e-4/20230601_025223/vis_data/vis_image/0000000.png_11.png b/convnext_small_4xb256_fake5m-lr4e-4/20230601_025223/vis_data/vis_image/0000000.png_11.png new file mode 100644 index 0000000000000000000000000000000000000000..6f264824565adf0630fdfc2e331f1ad6f2495172 Binary files /dev/null and b/convnext_small_4xb256_fake5m-lr4e-4/20230601_025223/vis_data/vis_image/0000000.png_11.png differ diff --git a/convnext_small_4xb256_fake5m-lr4e-4/20230601_025223/vis_data/vis_image/0000000.png_12.png b/convnext_small_4xb256_fake5m-lr4e-4/20230601_025223/vis_data/vis_image/0000000.png_12.png new file mode 100644 index 0000000000000000000000000000000000000000..6f264824565adf0630fdfc2e331f1ad6f2495172 Binary files /dev/null and b/convnext_small_4xb256_fake5m-lr4e-4/20230601_025223/vis_data/vis_image/0000000.png_12.png differ diff --git a/convnext_small_4xb256_fake5m-lr4e-4/20230601_025223/vis_data/vis_image/0000000.png_13.png b/convnext_small_4xb256_fake5m-lr4e-4/20230601_025223/vis_data/vis_image/0000000.png_13.png new file mode 100644 index 0000000000000000000000000000000000000000..6f264824565adf0630fdfc2e331f1ad6f2495172 Binary files /dev/null and b/convnext_small_4xb256_fake5m-lr4e-4/20230601_025223/vis_data/vis_image/0000000.png_13.png differ diff --git a/convnext_small_4xb256_fake5m-lr4e-4/20230601_025223/vis_data/vis_image/0000000.png_14.png b/convnext_small_4xb256_fake5m-lr4e-4/20230601_025223/vis_data/vis_image/0000000.png_14.png new file mode 100644 index 0000000000000000000000000000000000000000..6f264824565adf0630fdfc2e331f1ad6f2495172 Binary files /dev/null and b/convnext_small_4xb256_fake5m-lr4e-4/20230601_025223/vis_data/vis_image/0000000.png_14.png differ diff --git a/convnext_small_4xb256_fake5m-lr4e-4/20230601_025223/vis_data/vis_image/0000000.png_15.png b/convnext_small_4xb256_fake5m-lr4e-4/20230601_025223/vis_data/vis_image/0000000.png_15.png new file mode 100644 index 0000000000000000000000000000000000000000..6f264824565adf0630fdfc2e331f1ad6f2495172 Binary files /dev/null and b/convnext_small_4xb256_fake5m-lr4e-4/20230601_025223/vis_data/vis_image/0000000.png_15.png differ diff --git a/convnext_small_4xb256_fake5m-lr4e-4/20230601_025223/vis_data/vis_image/0000000.png_2.png b/convnext_small_4xb256_fake5m-lr4e-4/20230601_025223/vis_data/vis_image/0000000.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..6f264824565adf0630fdfc2e331f1ad6f2495172 Binary files /dev/null and b/convnext_small_4xb256_fake5m-lr4e-4/20230601_025223/vis_data/vis_image/0000000.png_2.png differ diff --git a/convnext_small_4xb256_fake5m-lr4e-4/20230601_025223/vis_data/vis_image/0000000.png_3.png b/convnext_small_4xb256_fake5m-lr4e-4/20230601_025223/vis_data/vis_image/0000000.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..6f264824565adf0630fdfc2e331f1ad6f2495172 Binary files /dev/null and b/convnext_small_4xb256_fake5m-lr4e-4/20230601_025223/vis_data/vis_image/0000000.png_3.png differ diff --git a/convnext_small_4xb256_fake5m-lr4e-4/20230601_025223/vis_data/vis_image/0000000.png_4.png b/convnext_small_4xb256_fake5m-lr4e-4/20230601_025223/vis_data/vis_image/0000000.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..6f264824565adf0630fdfc2e331f1ad6f2495172 Binary files /dev/null and b/convnext_small_4xb256_fake5m-lr4e-4/20230601_025223/vis_data/vis_image/0000000.png_4.png differ diff --git a/convnext_small_4xb256_fake5m-lr4e-4/20230601_025223/vis_data/vis_image/0000000.png_5.png b/convnext_small_4xb256_fake5m-lr4e-4/20230601_025223/vis_data/vis_image/0000000.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..6f264824565adf0630fdfc2e331f1ad6f2495172 Binary files /dev/null and b/convnext_small_4xb256_fake5m-lr4e-4/20230601_025223/vis_data/vis_image/0000000.png_5.png differ diff --git a/convnext_small_4xb256_fake5m-lr4e-4/20230601_025223/vis_data/vis_image/0000000.png_6.png b/convnext_small_4xb256_fake5m-lr4e-4/20230601_025223/vis_data/vis_image/0000000.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..6f264824565adf0630fdfc2e331f1ad6f2495172 Binary files /dev/null and b/convnext_small_4xb256_fake5m-lr4e-4/20230601_025223/vis_data/vis_image/0000000.png_6.png differ diff --git a/convnext_small_4xb256_fake5m-lr4e-4/20230601_025223/vis_data/vis_image/0000000.png_7.png b/convnext_small_4xb256_fake5m-lr4e-4/20230601_025223/vis_data/vis_image/0000000.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..6f264824565adf0630fdfc2e331f1ad6f2495172 Binary files /dev/null and b/convnext_small_4xb256_fake5m-lr4e-4/20230601_025223/vis_data/vis_image/0000000.png_7.png differ diff --git a/convnext_small_4xb256_fake5m-lr4e-4/20230601_025223/vis_data/vis_image/0000000.png_8.png b/convnext_small_4xb256_fake5m-lr4e-4/20230601_025223/vis_data/vis_image/0000000.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..6f264824565adf0630fdfc2e331f1ad6f2495172 Binary files /dev/null and b/convnext_small_4xb256_fake5m-lr4e-4/20230601_025223/vis_data/vis_image/0000000.png_8.png differ diff --git a/convnext_small_4xb256_fake5m-lr4e-4/20230601_025223/vis_data/vis_image/0000000.png_9.png b/convnext_small_4xb256_fake5m-lr4e-4/20230601_025223/vis_data/vis_image/0000000.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..6f264824565adf0630fdfc2e331f1ad6f2495172 Binary files /dev/null and b/convnext_small_4xb256_fake5m-lr4e-4/20230601_025223/vis_data/vis_image/0000000.png_9.png differ diff --git a/convnext_small_4xb256_fake5m-lr4e-4/convnext_small_4xb256_fake5m-lr4e-4.py b/convnext_small_4xb256_fake5m-lr4e-4/convnext_small_4xb256_fake5m-lr4e-4.py new file mode 100644 index 0000000000000000000000000000000000000000..ac1ce3ab72a9def3059afc183203855eae50d3ab --- /dev/null +++ b/convnext_small_4xb256_fake5m-lr4e-4/convnext_small_4xb256_fake5m-lr4e-4.py @@ -0,0 +1,169 @@ +optim_wrapper = dict( + optimizer=dict( + type='AdamW', + lr=0.0004, + weight_decay=0.05, + eps=1e-08, + betas=(0.9, 0.999), + _scope_='mmpretrain'), + paramwise_cfg=dict( + norm_decay_mult=0.0, + bias_decay_mult=0.0, + flat_decay_mult=0.0, + custom_keys=dict({ + '.absolute_pos_embed': dict(decay_mult=0.0), + '.relative_position_bias_table': dict(decay_mult=0.0) + })), + type='AmpOptimWrapper', + dtype='bfloat16', + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=True, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=20, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=4096) +model = dict( + type='ImageClassifier', + backbone=dict(type='ConvNeXt', arch='small', drop_path_rate=0.4), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=768, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + init_cfg=None), + init_cfg=dict( + type='TruncNormal', layer=['Conv2d', 'Linear'], std=0.02, bias=0.0), + train_cfg=None) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stablediffusionV1-5R2-dpmsolver-25-5m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/convnext_small_4xb256_fake5m-lr4e-4' diff --git a/convnext_small_4xb256_fake5m-lr4e-4/epoch_1.pth b/convnext_small_4xb256_fake5m-lr4e-4/epoch_1.pth new file mode 100644 index 0000000000000000000000000000000000000000..104309bd5bb54de5ceace5e7fa1a43eed2995978 --- /dev/null +++ b/convnext_small_4xb256_fake5m-lr4e-4/epoch_1.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c699a68fd0f070c0046ad6dadcb8953189b730b9d8015bff70256c4a7752580e +size 792305888 diff --git a/convnext_small_4xb256_fake5m-lr4e-4/epoch_10.pth b/convnext_small_4xb256_fake5m-lr4e-4/epoch_10.pth new file mode 100644 index 0000000000000000000000000000000000000000..a3ec3dcb7aa233e425c630c3269a31dcffd39d20 --- /dev/null +++ b/convnext_small_4xb256_fake5m-lr4e-4/epoch_10.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:829d618dc001743bb14c77d34d55b4842f8dfac603ce08fedf70be9264b3b1ed +size 796063456 diff --git a/convnext_small_4xb256_fake5m-lr4e-4/epoch_11.pth b/convnext_small_4xb256_fake5m-lr4e-4/epoch_11.pth new file mode 100644 index 0000000000000000000000000000000000000000..dd40bf1ccbd936eaba28e8f853fe69f1e1a53bf1 --- /dev/null +++ b/convnext_small_4xb256_fake5m-lr4e-4/epoch_11.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:29dd6cbae2550ed7bb00155bd9ba40ce673d4174947884935a0ea8129ec3fcb8 +size 796477536 diff --git a/convnext_small_4xb256_fake5m-lr4e-4/epoch_12.pth b/convnext_small_4xb256_fake5m-lr4e-4/epoch_12.pth new file mode 100644 index 0000000000000000000000000000000000000000..4c2fd0ad7a77692a5ed61ded614d10d07e35c4c3 --- /dev/null +++ b/convnext_small_4xb256_fake5m-lr4e-4/epoch_12.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:436911ea15d44cd1afc32327d4526803624a3de70a27541ba3499268d740cc3e +size 796891360 diff --git a/convnext_small_4xb256_fake5m-lr4e-4/epoch_13.pth b/convnext_small_4xb256_fake5m-lr4e-4/epoch_13.pth new file mode 100644 index 0000000000000000000000000000000000000000..2b1d6145acaa2017ae42397e0e5e4ff1de5ff084 --- /dev/null +++ b/convnext_small_4xb256_fake5m-lr4e-4/epoch_13.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:f72c0523e0ca3cd25c80ea08cd2fbeab05665735240d19b3bf4f9384c9138c87 +size 797311264 diff --git a/convnext_small_4xb256_fake5m-lr4e-4/epoch_14.pth b/convnext_small_4xb256_fake5m-lr4e-4/epoch_14.pth new file mode 100644 index 0000000000000000000000000000000000000000..16ed01c235a96f0990202e79a31d364c7e19510f --- /dev/null +++ b/convnext_small_4xb256_fake5m-lr4e-4/epoch_14.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:5e8a907edf1fc07839cd06f8cc68d36fdc6575b933b92581e2dc2581e5db2d61 +size 797725216 diff --git a/convnext_small_4xb256_fake5m-lr4e-4/epoch_15.pth b/convnext_small_4xb256_fake5m-lr4e-4/epoch_15.pth new file mode 100644 index 0000000000000000000000000000000000000000..56c13fd545a09336093942f08ab52a3a4fef175c --- /dev/null +++ b/convnext_small_4xb256_fake5m-lr4e-4/epoch_15.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:881567eae26008aae030e869c75d4d56b774d7e4e69d5b196496044b10357649 +size 798151136 diff --git a/convnext_small_4xb256_fake5m-lr4e-4/epoch_2.pth b/convnext_small_4xb256_fake5m-lr4e-4/epoch_2.pth new file mode 100644 index 0000000000000000000000000000000000000000..8c79c9ff2f25b0f9e713e48b575fdaacfdb7badf --- /dev/null +++ b/convnext_small_4xb256_fake5m-lr4e-4/epoch_2.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:e14b9b99237114a1824570e7d49d56cd0ec8605dea43a729727b4b80fa784e0b +size 792732960 diff --git a/convnext_small_4xb256_fake5m-lr4e-4/epoch_3.pth b/convnext_small_4xb256_fake5m-lr4e-4/epoch_3.pth new file mode 100644 index 0000000000000000000000000000000000000000..7dd8ac393c716f5174cdfa4484bf18ca66e28857 --- /dev/null +++ b/convnext_small_4xb256_fake5m-lr4e-4/epoch_3.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:a82dc1a54adffbe7f356278496431f21fb871a8c7e12be4eb1dcd1eedddb7d5a +size 793164576 diff --git a/convnext_small_4xb256_fake5m-lr4e-4/epoch_4.pth b/convnext_small_4xb256_fake5m-lr4e-4/epoch_4.pth new file mode 100644 index 0000000000000000000000000000000000000000..45eb4fe3b58a8e580e20ad0947251045ab830570 --- /dev/null +++ b/convnext_small_4xb256_fake5m-lr4e-4/epoch_4.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:52e5852a423994711a13c3fd13b2e2daf8ca6c0ff7b5f53087069485a439cff9 +size 793584544 diff --git a/convnext_small_4xb256_fake5m-lr4e-4/epoch_5.pth b/convnext_small_4xb256_fake5m-lr4e-4/epoch_5.pth new file mode 100644 index 0000000000000000000000000000000000000000..fa23249c22320ac5bf553de0c2df8c4a07b59c8e --- /dev/null +++ b/convnext_small_4xb256_fake5m-lr4e-4/epoch_5.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:e8e236c6985af87e764c0179156d6c9f047608f9e15a63cf50bb6e064adf4f85 +size 793997344 diff --git a/convnext_small_4xb256_fake5m-lr4e-4/epoch_6.pth b/convnext_small_4xb256_fake5m-lr4e-4/epoch_6.pth new file mode 100644 index 0000000000000000000000000000000000000000..6f0c717d7b1c7f8f794fd74b5dd4e13053df3487 --- /dev/null +++ b/convnext_small_4xb256_fake5m-lr4e-4/epoch_6.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:5f8b1b191740caae7eaf8b988ff47d2f1e35dba2733e82b79cdb4f8d76358eec +size 794408864 diff --git a/convnext_small_4xb256_fake5m-lr4e-4/epoch_7.pth b/convnext_small_4xb256_fake5m-lr4e-4/epoch_7.pth new file mode 100644 index 0000000000000000000000000000000000000000..f2daa46e5eb5e13d3248c2c4e3489b2e479fea79 --- /dev/null +++ b/convnext_small_4xb256_fake5m-lr4e-4/epoch_7.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:679f208eddea9614f160287cf536a085bbddee8d6bf45a516c439a939b07b968 +size 794824864 diff --git a/convnext_small_4xb256_fake5m-lr4e-4/epoch_8.pth b/convnext_small_4xb256_fake5m-lr4e-4/epoch_8.pth new file mode 100644 index 0000000000000000000000000000000000000000..1df5a77074cecb613df6db5e29f99b95510f7880 --- /dev/null +++ b/convnext_small_4xb256_fake5m-lr4e-4/epoch_8.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:f7f3fa42556eca7e50b1cb48bc0426c2662515bb7572c7a27082b1e9d061df33 +size 795234400 diff --git a/convnext_small_4xb256_fake5m-lr4e-4/epoch_9.pth b/convnext_small_4xb256_fake5m-lr4e-4/epoch_9.pth new file mode 100644 index 0000000000000000000000000000000000000000..b71e7ce5d09ec1f37bb67e09ef0a256c607346b5 --- /dev/null +++ b/convnext_small_4xb256_fake5m-lr4e-4/epoch_9.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:a7159dac0b563d9524e5fa51699542608262f82d7b7e4c4144655709b0cbbafb +size 795649184 diff --git a/convnext_small_4xb256_fake5m-lr4e-4/last_checkpoint b/convnext_small_4xb256_fake5m-lr4e-4/last_checkpoint new file mode 100644 index 0000000000000000000000000000000000000000..bec7202a2a8ecb646f1fa2271d8832988dc3af4a --- /dev/null +++ b/convnext_small_4xb256_fake5m-lr4e-4/last_checkpoint @@ -0,0 +1 @@ +/mnt/petrelfs/luzeyu/workspace/fakebench/mmpretrain/workdir/convnext_small_4xb256_fake5m-lr4e-4/epoch_15.pth \ No newline at end of file diff --git a/convnext_small_4xb256_fake5m_lr6e-4_10epoch/20230601_151930/20230601_151930.log b/convnext_small_4xb256_fake5m_lr6e-4_10epoch/20230601_151930/20230601_151930.log new file mode 100644 index 0000000000000000000000000000000000000000..73bd037cbcc1a90016a018c7b762419588145bfa --- /dev/null +++ b/convnext_small_4xb256_fake5m_lr6e-4_10epoch/20230601_151930/20230601_151930.log @@ -0,0 +1,1909 @@ +2023/06/01 15:19:34 - mmengine - INFO - +------------------------------------------------------------ +System environment: + sys.platform: linux + Python: 3.10.9 (main, Mar 8 2023, 10:47:38) [GCC 11.2.0] + CUDA available: True + numpy_random_seed: 233860104 + GPU 0,1,2,3: NVIDIA A100-SXM4-80GB + CUDA_HOME: /mnt/petrelfs/share/cuda-11.6 + NVCC: Cuda compilation tools, release 11.6, V11.6.124 + GCC: gcc (GCC) 4.8.5 20150623 (Red Hat 4.8.5-44) + PyTorch: 1.13.1 + PyTorch compiling details: PyTorch built with: + - GCC 9.3 + - C++ Version: 201402 + - Intel(R) oneAPI Math Kernel Library Version 2021.4-Product Build 20210904 for Intel(R) 64 architecture applications + - Intel(R) MKL-DNN v2.6.0 (Git Hash 52b5f107dd9cf10910aaa19cb47f3abf9b349815) + - OpenMP 201511 (a.k.a. OpenMP 4.5) + - LAPACK is enabled (usually provided by MKL) + - NNPACK is enabled + - CPU capability usage: AVX2 + - CUDA Runtime 11.6 + - NVCC architecture flags: -gencode;arch=compute_37,code=sm_37;-gencode;arch=compute_50,code=sm_50;-gencode;arch=compute_60,code=sm_60;-gencode;arch=compute_61,code=sm_61;-gencode;arch=compute_70,code=sm_70;-gencode;arch=compute_75,code=sm_75;-gencode;arch=compute_80,code=sm_80;-gencode;arch=compute_86,code=sm_86;-gencode;arch=compute_37,code=compute_37 + - CuDNN 8.3.2 (built against CUDA 11.5) + - Magma 2.6.1 + - Build settings: BLAS_INFO=mkl, BUILD_TYPE=Release, CUDA_VERSION=11.6, CUDNN_VERSION=8.3.2, CXX_COMPILER=/opt/rh/devtoolset-9/root/usr/bin/c++, CXX_FLAGS= -fabi-version=11 -Wno-deprecated -fvisibility-inlines-hidden -DUSE_PTHREADPOOL -fopenmp -DNDEBUG -DUSE_KINETO -DUSE_FBGEMM -DUSE_QNNPACK -DUSE_PYTORCH_QNNPACK -DUSE_XNNPACK -DSYMBOLICATE_MOBILE_DEBUG_HANDLE -DEDGE_PROFILER_USE_KINETO -O2 -fPIC -Wno-narrowing -Wall -Wextra -Werror=return-type -Werror=non-virtual-dtor -Wno-missing-field-initializers -Wno-type-limits -Wno-array-bounds -Wno-unknown-pragmas -Wunused-local-typedefs -Wno-unused-parameter -Wno-unused-function -Wno-unused-result -Wno-strict-overflow -Wno-strict-aliasing -Wno-error=deprecated-declarations -Wno-stringop-overflow -Wno-psabi -Wno-error=pedantic -Wno-error=redundant-decls -Wno-error=old-style-cast -fdiagnostics-color=always -faligned-new -Wno-unused-but-set-variable -Wno-maybe-uninitialized -fno-math-errno -fno-trapping-math -Werror=format -Werror=cast-function-type -Wno-stringop-overflow, LAPACK_INFO=mkl, PERF_WITH_AVX=1, PERF_WITH_AVX2=1, PERF_WITH_AVX512=1, TORCH_VERSION=1.13.1, USE_CUDA=ON, USE_CUDNN=ON, USE_EXCEPTION_PTR=1, USE_GFLAGS=OFF, USE_GLOG=OFF, USE_MKL=ON, USE_MKLDNN=ON, USE_MPI=OFF, USE_NCCL=ON, USE_NNPACK=ON, USE_OPENMP=ON, USE_ROCM=OFF, + + TorchVision: 0.14.1 + OpenCV: 4.7.0 + MMEngine: 0.7.3 + +Runtime environment: + cudnn_benchmark: True + mp_cfg: {'mp_start_method': 'fork', 'opencv_num_threads': 0} + dist_cfg: {'backend': 'nccl'} + seed: None + deterministic: False + Distributed launcher: slurm + Distributed training: True + GPU number: 4 +------------------------------------------------------------ + +2023/06/01 15:19:34 - mmengine - INFO - Config: +optim_wrapper = dict( + optimizer=dict( + type='AdamW', + lr=0.0006, + weight_decay=0.05, + eps=1e-08, + betas=(0.9, 0.999), + _scope_='mmpretrain'), + paramwise_cfg=dict( + norm_decay_mult=0.0, + bias_decay_mult=0.0, + flat_decay_mult=0.0, + custom_keys=dict({ + '.absolute_pos_embed': dict(decay_mult=0.0), + '.relative_position_bias_table': dict(decay_mult=0.0) + })), + type='AmpOptimWrapper', + dtype='bfloat16', + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=True, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=4096) +model = dict( + type='ImageClassifier', + backbone=dict(type='ConvNeXt', arch='small', drop_path_rate=0.4), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=768, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + init_cfg=None), + init_cfg=dict( + type='TruncNormal', layer=['Conv2d', 'Linear'], std=0.02, bias=0.0), + train_cfg=None) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stablediffusionV1-5R2-dpmsolver-25-5m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/convnext_small_4xb256_fake5m_lr6e-4_10epoch' + +2023/06/01 15:19:39 - mmengine - INFO - Hooks will be executed in the following order: +before_run: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_load_checkpoint: +(ABOVE_NORMAL) EMAHook + -------------------- +before_train: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_train_epoch: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(NORMAL ) DistSamplerSeedHook + -------------------- +before_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook + -------------------- +after_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +after_train_epoch: +(NORMAL ) IterTimerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_val_epoch: +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook + -------------------- +before_val_iter: +(NORMAL ) IterTimerHook + -------------------- +after_val_iter: +(NORMAL ) IterTimerHook +(NORMAL ) VisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_val_epoch: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_save_checkpoint: +(ABOVE_NORMAL) EMAHook + -------------------- +after_train: +(VERY_LOW ) CheckpointHook + -------------------- +before_test_epoch: +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook + -------------------- +before_test_iter: +(NORMAL ) IterTimerHook + -------------------- +after_test_iter: +(NORMAL ) IterTimerHook +(NORMAL ) VisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_test_epoch: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_run: +(BELOW_NORMAL) LoggerHook + -------------------- +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.0.0.bias:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.0.1.weight:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.0.1.bias:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.1.0.weight:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.1.0.bias:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.1.1.bias:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.2.0.weight:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.2.0.bias:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.2.1.bias:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.3.0.weight:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.3.0.bias:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.3.1.bias:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.0.0.gamma:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.0.0.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.0.0.norm.weight:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.0.0.norm.bias:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.0.0.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.0.0.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.0.1.gamma:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.0.1.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.0.1.norm.weight:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.0.1.norm.bias:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.0.1.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.0.1.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.0.2.gamma:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.0.2.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.0.2.norm.weight:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.0.2.norm.bias:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.0.2.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.0.2.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.1.0.gamma:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.1.0.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.1.0.norm.weight:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.1.0.norm.bias:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.1.0.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.1.0.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.1.1.gamma:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.1.1.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.1.1.norm.weight:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.1.1.norm.bias:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.1.1.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.1.1.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.1.2.gamma:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.1.2.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.1.2.norm.weight:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.1.2.norm.bias:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.1.2.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.1.2.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.2.0.gamma:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.2.0.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.2.0.norm.weight:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.2.0.norm.bias:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.2.0.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.2.0.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.2.1.gamma:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.2.1.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.2.1.norm.weight:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.2.1.norm.bias:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.2.1.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.2.1.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.2.2.gamma:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.2.2.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.2.2.norm.weight:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.2.2.norm.bias:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.2.2.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.2.2.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.2.3.gamma:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.2.3.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.2.3.norm.weight:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.2.3.norm.bias:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.2.3.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.2.3.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.2.4.gamma:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.2.4.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.2.4.norm.weight:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.2.4.norm.bias:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.2.4.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.2.4.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.2.5.gamma:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.2.5.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.2.5.norm.weight:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.2.5.norm.bias:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.2.5.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.2.5.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.2.6.gamma:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.2.6.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.2.6.norm.weight:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.2.6.norm.bias:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.2.6.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.2.6.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.2.7.gamma:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.2.7.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.2.7.norm.weight:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.2.7.norm.bias:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.2.7.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.2.7.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.2.8.gamma:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.2.8.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.2.8.norm.weight:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.2.8.norm.bias:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.2.8.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.2.8.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.2.9.gamma:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.2.9.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.2.9.norm.weight:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.2.9.norm.bias:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.2.9.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.2.9.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.2.10.gamma:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.2.10.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.2.10.norm.weight:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.2.10.norm.bias:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.2.10.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.2.10.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.2.11.gamma:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.2.11.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.2.11.norm.weight:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.2.11.norm.bias:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.2.11.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.2.11.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.2.12.gamma:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.2.12.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.2.12.norm.weight:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.2.12.norm.bias:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.2.12.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.2.12.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.2.13.gamma:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.2.13.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.2.13.norm.weight:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.2.13.norm.bias:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.2.13.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.2.13.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.2.14.gamma:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.2.14.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.2.14.norm.weight:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.2.14.norm.bias:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.2.14.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.2.14.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.2.15.gamma:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.2.15.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.2.15.norm.weight:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.2.15.norm.bias:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.2.15.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.2.15.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.2.16.gamma:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.2.16.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.2.16.norm.weight:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.2.16.norm.bias:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.2.16.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.2.16.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.2.17.gamma:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.2.17.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.2.17.norm.weight:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.2.17.norm.bias:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.2.17.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.2.17.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.2.18.gamma:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.2.18.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.2.18.norm.weight:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.2.18.norm.bias:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.2.18.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.2.18.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.2.19.gamma:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.2.19.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.2.19.norm.weight:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.2.19.norm.bias:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.2.19.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.2.19.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.2.20.gamma:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.2.20.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.2.20.norm.weight:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.2.20.norm.bias:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.2.20.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.2.20.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.2.21.gamma:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.2.21.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.2.21.norm.weight:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.2.21.norm.bias:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.2.21.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.2.21.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.2.22.gamma:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.2.22.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.2.22.norm.weight:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.2.22.norm.bias:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.2.22.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.2.22.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.2.23.gamma:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.2.23.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.2.23.norm.weight:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.2.23.norm.bias:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.2.23.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.2.23.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.2.24.gamma:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.2.24.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.2.24.norm.weight:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.2.24.norm.bias:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.2.24.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.2.24.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.2.25.gamma:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.2.25.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.2.25.norm.weight:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.2.25.norm.bias:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.2.25.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.2.25.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.2.26.gamma:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.2.26.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.2.26.norm.weight:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.2.26.norm.bias:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.2.26.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.2.26.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.3.0.gamma:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.3.0.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.3.0.norm.weight:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.3.0.norm.bias:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.3.0.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.3.0.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.3.1.gamma:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.3.1.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.3.1.norm.weight:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.3.1.norm.bias:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.3.1.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.3.1.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.3.2.gamma:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.3.2.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.3.2.norm.weight:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.3.2.norm.bias:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.3.2.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.stages.3.2.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.norm3.weight:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- backbone.norm3.bias:weight_decay=0.0 +2023/06/01 15:20:05 - mmengine - INFO - paramwise_options -- head.fc.bias:weight_decay=0.0 +Name of parameter - Initialization information + +backbone.downsample_layers.0.0.weight - torch.Size([96, 3, 4, 4]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.downsample_layers.0.0.bias - torch.Size([96]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.downsample_layers.0.1.weight - torch.Size([96]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.downsample_layers.0.1.bias - torch.Size([96]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.downsample_layers.1.0.weight - torch.Size([96]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.downsample_layers.1.0.bias - torch.Size([96]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.downsample_layers.1.1.weight - torch.Size([192, 96, 2, 2]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.downsample_layers.1.1.bias - torch.Size([192]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.downsample_layers.2.0.weight - torch.Size([192]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.downsample_layers.2.0.bias - torch.Size([192]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.downsample_layers.2.1.weight - torch.Size([384, 192, 2, 2]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.downsample_layers.2.1.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.downsample_layers.3.0.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.downsample_layers.3.0.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.downsample_layers.3.1.weight - torch.Size([768, 384, 2, 2]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.downsample_layers.3.1.bias - torch.Size([768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.0.gamma - torch.Size([96]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.0.0.depthwise_conv.weight - torch.Size([96, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.0.depthwise_conv.bias - torch.Size([96]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.0.norm.weight - torch.Size([96]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.0.0.norm.bias - torch.Size([96]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.0.0.pointwise_conv1.weight - torch.Size([384, 96]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.0.pointwise_conv1.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.0.pointwise_conv2.weight - torch.Size([96, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.0.pointwise_conv2.bias - torch.Size([96]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.1.gamma - torch.Size([96]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.0.1.depthwise_conv.weight - torch.Size([96, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.1.depthwise_conv.bias - torch.Size([96]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.1.norm.weight - torch.Size([96]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.0.1.norm.bias - torch.Size([96]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.0.1.pointwise_conv1.weight - torch.Size([384, 96]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.1.pointwise_conv1.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.1.pointwise_conv2.weight - torch.Size([96, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.1.pointwise_conv2.bias - torch.Size([96]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.2.gamma - torch.Size([96]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.0.2.depthwise_conv.weight - torch.Size([96, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.2.depthwise_conv.bias - torch.Size([96]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.2.norm.weight - torch.Size([96]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.0.2.norm.bias - torch.Size([96]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.0.2.pointwise_conv1.weight - torch.Size([384, 96]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.2.pointwise_conv1.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.2.pointwise_conv2.weight - torch.Size([96, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.2.pointwise_conv2.bias - torch.Size([96]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.0.gamma - torch.Size([192]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.1.0.depthwise_conv.weight - torch.Size([192, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.0.depthwise_conv.bias - torch.Size([192]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.0.norm.weight - torch.Size([192]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.1.0.norm.bias - torch.Size([192]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.1.0.pointwise_conv1.weight - torch.Size([768, 192]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.0.pointwise_conv1.bias - torch.Size([768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.0.pointwise_conv2.weight - torch.Size([192, 768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.0.pointwise_conv2.bias - torch.Size([192]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.1.gamma - torch.Size([192]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.1.1.depthwise_conv.weight - torch.Size([192, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.1.depthwise_conv.bias - torch.Size([192]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.1.norm.weight - torch.Size([192]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.1.1.norm.bias - torch.Size([192]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.1.1.pointwise_conv1.weight - torch.Size([768, 192]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.1.pointwise_conv1.bias - torch.Size([768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.1.pointwise_conv2.weight - torch.Size([192, 768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.1.pointwise_conv2.bias - torch.Size([192]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.2.gamma - torch.Size([192]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.1.2.depthwise_conv.weight - torch.Size([192, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.2.depthwise_conv.bias - torch.Size([192]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.2.norm.weight - torch.Size([192]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.1.2.norm.bias - torch.Size([192]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.1.2.pointwise_conv1.weight - torch.Size([768, 192]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.2.pointwise_conv1.bias - torch.Size([768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.2.pointwise_conv2.weight - torch.Size([192, 768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.2.pointwise_conv2.bias - torch.Size([192]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.0.gamma - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.0.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.0.depthwise_conv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.0.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.0.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.0.pointwise_conv1.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.0.pointwise_conv1.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.0.pointwise_conv2.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.0.pointwise_conv2.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.1.gamma - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.1.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.1.depthwise_conv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.1.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.1.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.1.pointwise_conv1.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.1.pointwise_conv1.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.1.pointwise_conv2.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.1.pointwise_conv2.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.2.gamma - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.2.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.2.depthwise_conv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.2.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.2.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.2.pointwise_conv1.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.2.pointwise_conv1.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.2.pointwise_conv2.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.2.pointwise_conv2.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.3.gamma - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.3.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.3.depthwise_conv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.3.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.3.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.3.pointwise_conv1.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.3.pointwise_conv1.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.3.pointwise_conv2.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.3.pointwise_conv2.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.4.gamma - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.4.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.4.depthwise_conv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.4.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.4.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.4.pointwise_conv1.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.4.pointwise_conv1.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.4.pointwise_conv2.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.4.pointwise_conv2.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.5.gamma - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.5.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.5.depthwise_conv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.5.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.5.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.5.pointwise_conv1.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.5.pointwise_conv1.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.5.pointwise_conv2.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.5.pointwise_conv2.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.6.gamma - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.6.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.6.depthwise_conv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.6.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.6.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.6.pointwise_conv1.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.6.pointwise_conv1.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.6.pointwise_conv2.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.6.pointwise_conv2.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.7.gamma - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.7.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.7.depthwise_conv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.7.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.7.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.7.pointwise_conv1.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.7.pointwise_conv1.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.7.pointwise_conv2.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.7.pointwise_conv2.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.8.gamma - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.8.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.8.depthwise_conv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.8.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.8.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.8.pointwise_conv1.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.8.pointwise_conv1.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.8.pointwise_conv2.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.8.pointwise_conv2.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.9.gamma - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.9.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.9.depthwise_conv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.9.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.9.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.9.pointwise_conv1.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.9.pointwise_conv1.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.9.pointwise_conv2.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.9.pointwise_conv2.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.10.gamma - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.10.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.10.depthwise_conv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.10.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.10.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.10.pointwise_conv1.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.10.pointwise_conv1.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.10.pointwise_conv2.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.10.pointwise_conv2.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.11.gamma - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.11.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.11.depthwise_conv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.11.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.11.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.11.pointwise_conv1.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.11.pointwise_conv1.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.11.pointwise_conv2.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.11.pointwise_conv2.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.12.gamma - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.12.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.12.depthwise_conv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.12.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.12.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.12.pointwise_conv1.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.12.pointwise_conv1.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.12.pointwise_conv2.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.12.pointwise_conv2.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.13.gamma - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.13.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.13.depthwise_conv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.13.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.13.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.13.pointwise_conv1.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.13.pointwise_conv1.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.13.pointwise_conv2.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.13.pointwise_conv2.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.14.gamma - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.14.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.14.depthwise_conv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.14.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.14.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.14.pointwise_conv1.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.14.pointwise_conv1.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.14.pointwise_conv2.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.14.pointwise_conv2.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.15.gamma - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.15.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.15.depthwise_conv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.15.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.15.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.15.pointwise_conv1.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.15.pointwise_conv1.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.15.pointwise_conv2.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.15.pointwise_conv2.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.16.gamma - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.16.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.16.depthwise_conv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.16.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.16.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.16.pointwise_conv1.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.16.pointwise_conv1.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.16.pointwise_conv2.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.16.pointwise_conv2.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.17.gamma - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.17.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.17.depthwise_conv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.17.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.17.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.17.pointwise_conv1.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.17.pointwise_conv1.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.17.pointwise_conv2.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.17.pointwise_conv2.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.18.gamma - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.18.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.18.depthwise_conv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.18.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.18.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.18.pointwise_conv1.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.18.pointwise_conv1.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.18.pointwise_conv2.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.18.pointwise_conv2.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.19.gamma - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.19.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.19.depthwise_conv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.19.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.19.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.19.pointwise_conv1.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.19.pointwise_conv1.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.19.pointwise_conv2.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.19.pointwise_conv2.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.20.gamma - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.20.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.20.depthwise_conv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.20.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.20.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.20.pointwise_conv1.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.20.pointwise_conv1.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.20.pointwise_conv2.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.20.pointwise_conv2.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.21.gamma - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.21.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.21.depthwise_conv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.21.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.21.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.21.pointwise_conv1.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.21.pointwise_conv1.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.21.pointwise_conv2.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.21.pointwise_conv2.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.22.gamma - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.22.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.22.depthwise_conv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.22.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.22.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.22.pointwise_conv1.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.22.pointwise_conv1.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.22.pointwise_conv2.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.22.pointwise_conv2.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.23.gamma - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.23.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.23.depthwise_conv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.23.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.23.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.23.pointwise_conv1.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.23.pointwise_conv1.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.23.pointwise_conv2.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.23.pointwise_conv2.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.24.gamma - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.24.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.24.depthwise_conv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.24.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.24.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.24.pointwise_conv1.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.24.pointwise_conv1.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.24.pointwise_conv2.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.24.pointwise_conv2.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.25.gamma - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.25.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.25.depthwise_conv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.25.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.25.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.25.pointwise_conv1.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.25.pointwise_conv1.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.25.pointwise_conv2.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.25.pointwise_conv2.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.26.gamma - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.26.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.26.depthwise_conv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.26.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.26.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.26.pointwise_conv1.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.26.pointwise_conv1.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.26.pointwise_conv2.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.26.pointwise_conv2.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.0.gamma - torch.Size([768]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.3.0.depthwise_conv.weight - torch.Size([768, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.0.depthwise_conv.bias - torch.Size([768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.0.norm.weight - torch.Size([768]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.3.0.norm.bias - torch.Size([768]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.3.0.pointwise_conv1.weight - torch.Size([3072, 768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.0.pointwise_conv1.bias - torch.Size([3072]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.0.pointwise_conv2.weight - torch.Size([768, 3072]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.0.pointwise_conv2.bias - torch.Size([768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.1.gamma - torch.Size([768]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.3.1.depthwise_conv.weight - torch.Size([768, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.1.depthwise_conv.bias - torch.Size([768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.1.norm.weight - torch.Size([768]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.3.1.norm.bias - torch.Size([768]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.3.1.pointwise_conv1.weight - torch.Size([3072, 768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.1.pointwise_conv1.bias - torch.Size([3072]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.1.pointwise_conv2.weight - torch.Size([768, 3072]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.1.pointwise_conv2.bias - torch.Size([768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.2.gamma - torch.Size([768]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.3.2.depthwise_conv.weight - torch.Size([768, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.2.depthwise_conv.bias - torch.Size([768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.2.norm.weight - torch.Size([768]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.3.2.norm.bias - torch.Size([768]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.3.2.pointwise_conv1.weight - torch.Size([3072, 768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.2.pointwise_conv1.bias - torch.Size([3072]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.2.pointwise_conv2.weight - torch.Size([768, 3072]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.2.pointwise_conv2.bias - torch.Size([768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.norm3.weight - torch.Size([768]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.norm3.bias - torch.Size([768]): +The value is the same before and after calling `init_weights` of ImageClassifier + +head.fc.weight - torch.Size([2, 768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +head.fc.bias - torch.Size([2]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 +2023/06/01 15:20:06 - mmengine - WARNING - "FileClient" will be deprecated in future. Please use io functions in https://mmengine.readthedocs.io/en/latest/api/fileio.html#file-io +2023/06/01 15:20:06 - mmengine - WARNING - "HardDiskBackend" is the alias of "LocalBackend" and the former will be deprecated in future. +2023/06/01 15:20:06 - mmengine - INFO - Checkpoints will be saved to /mnt/petrelfs/luzeyu/workspace/fakebench/mmpretrain/workdir/convnext_small_4xb256_fake5m_lr6e-4_10epoch. +2023/06/01 15:21:40 - mmengine - INFO - Epoch(train) [1][ 100/5758] lr: 6.0000e-04 eta: 15:01:30 time: 0.8301 data_time: 0.0016 memory: 25062 loss: 0.6136 +2023/06/01 15:23:07 - mmengine - INFO - Epoch(train) [1][ 200/5758] lr: 6.0000e-04 eta: 14:24:36 time: 0.8468 data_time: 0.1306 memory: 25062 loss: 0.5753 +2023/06/01 15:24:30 - mmengine - INFO - Epoch(train) [1][ 300/5758] lr: 6.0000e-04 eta: 13:59:46 time: 0.8305 data_time: 0.1261 memory: 25062 loss: 0.5316 +2023/06/01 15:25:54 - mmengine - INFO - Epoch(train) [1][ 400/5758] lr: 6.0000e-04 eta: 13:50:00 time: 0.8568 data_time: 0.1325 memory: 25062 loss: 0.5086 +2023/06/01 15:27:16 - mmengine - INFO - Epoch(train) [1][ 500/5758] lr: 6.0000e-04 eta: 13:38:52 time: 0.8660 data_time: 0.1639 memory: 25062 loss: 0.5010 +2023/06/01 15:28:42 - mmengine - INFO - Epoch(train) [1][ 600/5758] lr: 6.0000e-04 eta: 13:37:24 time: 0.8801 data_time: 0.1425 memory: 25062 loss: 0.4692 +2023/06/01 15:30:08 - mmengine - INFO - Epoch(train) [1][ 700/5758] lr: 6.0000e-04 eta: 13:35:41 time: 0.9139 data_time: 0.2090 memory: 25062 loss: 0.4822 +2023/06/01 15:31:32 - mmengine - INFO - Epoch(train) [1][ 800/5758] lr: 6.0000e-04 eta: 13:31:52 time: 0.8784 data_time: 0.1478 memory: 25062 loss: 0.4323 +2023/06/01 15:32:58 - mmengine - INFO - Epoch(train) [1][ 900/5758] lr: 6.0000e-04 eta: 13:30:16 time: 0.8333 data_time: 0.1298 memory: 25062 loss: 0.4235 +2023/06/01 15:34:23 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_lr6e-4_10epoch_20230601_151930 +2023/06/01 15:34:23 - mmengine - INFO - Epoch(train) [1][1000/5758] lr: 6.0000e-04 eta: 13:28:07 time: 0.8239 data_time: 0.1030 memory: 25062 loss: 0.4348 +2023/06/01 15:35:45 - mmengine - INFO - Epoch(train) [1][1100/5758] lr: 6.0000e-04 eta: 13:23:14 time: 0.8161 data_time: 0.1142 memory: 25062 loss: 0.4187 +2023/06/01 15:37:12 - mmengine - INFO - Epoch(train) [1][1200/5758] lr: 6.0000e-04 eta: 13:23:21 time: 0.8774 data_time: 0.1601 memory: 25062 loss: 0.4102 +2023/06/01 15:38:33 - mmengine - INFO - Epoch(train) [1][1300/5758] lr: 6.0000e-04 eta: 13:18:53 time: 0.7686 data_time: 0.0665 memory: 25062 loss: 0.4231 +2023/06/01 15:39:57 - mmengine - INFO - Epoch(train) [1][1400/5758] lr: 6.0000e-04 eta: 13:16:36 time: 0.8020 data_time: 0.0882 memory: 25062 loss: 0.3834 +2023/06/01 15:41:20 - mmengine - INFO - Epoch(train) [1][1500/5758] lr: 6.0000e-04 eta: 13:13:38 time: 0.8428 data_time: 0.1409 memory: 25062 loss: 0.3752 +2023/06/01 15:42:43 - mmengine - INFO - Epoch(train) [1][1600/5758] lr: 6.0000e-04 eta: 13:11:30 time: 0.8140 data_time: 0.1000 memory: 25062 loss: 0.3995 +2023/06/01 15:44:06 - mmengine - INFO - Epoch(train) [1][1700/5758] lr: 6.0000e-04 eta: 13:09:06 time: 0.8324 data_time: 0.1297 memory: 25062 loss: 0.3621 +2023/06/01 15:45:32 - mmengine - INFO - Epoch(train) [1][1800/5758] lr: 6.0000e-04 eta: 13:08:07 time: 0.7991 data_time: 0.0887 memory: 25062 loss: 0.3500 +2023/06/01 15:46:56 - mmengine - INFO - Epoch(train) [1][1900/5758] lr: 6.0000e-04 eta: 13:06:30 time: 0.8130 data_time: 0.1101 memory: 25062 loss: 0.3260 +2023/06/01 15:48:19 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_lr6e-4_10epoch_20230601_151930 +2023/06/01 15:48:19 - mmengine - INFO - Epoch(train) [1][2000/5758] lr: 6.0000e-04 eta: 13:04:15 time: 0.8048 data_time: 0.0940 memory: 25062 loss: 0.3501 +2023/06/01 15:49:41 - mmengine - INFO - Epoch(train) [1][2100/5758] lr: 6.0000e-04 eta: 13:01:27 time: 0.8261 data_time: 0.1207 memory: 25062 loss: 0.3447 +2023/06/01 15:51:04 - mmengine - INFO - Epoch(train) [1][2200/5758] lr: 6.0000e-04 eta: 12:59:35 time: 0.8569 data_time: 0.1428 memory: 25062 loss: 0.3121 +2023/06/01 15:52:26 - mmengine - INFO - Epoch(train) [1][2300/5758] lr: 6.0000e-04 eta: 12:57:14 time: 0.8037 data_time: 0.1016 memory: 25062 loss: 0.2937 +2023/06/01 15:53:49 - mmengine - INFO - Epoch(train) [1][2400/5758] lr: 6.0000e-04 eta: 12:55:18 time: 0.7807 data_time: 0.0642 memory: 25062 loss: 0.3212 +2023/06/01 15:55:12 - mmengine - INFO - Epoch(train) [1][2500/5758] lr: 6.0000e-04 eta: 12:53:14 time: 0.8198 data_time: 0.1138 memory: 25062 loss: 0.2847 +2023/06/01 15:56:34 - mmengine - INFO - Epoch(train) [1][2600/5758] lr: 6.0000e-04 eta: 12:51:14 time: 0.8528 data_time: 0.1508 memory: 25062 loss: 0.2840 +2023/06/01 15:57:59 - mmengine - INFO - Epoch(train) [1][2700/5758] lr: 6.0000e-04 eta: 12:49:56 time: 0.8648 data_time: 0.1616 memory: 25062 loss: 0.2992 +2023/06/01 15:59:23 - mmengine - INFO - Epoch(train) [1][2800/5758] lr: 6.0000e-04 eta: 12:48:32 time: 0.8262 data_time: 0.1241 memory: 25062 loss: 0.2753 +2023/06/01 16:00:48 - mmengine - INFO - Epoch(train) [1][2900/5758] lr: 6.0000e-04 eta: 12:47:26 time: 0.8611 data_time: 0.1581 memory: 25062 loss: 0.2809 +2023/06/01 16:02:12 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_lr6e-4_10epoch_20230601_151930 +2023/06/01 16:02:12 - mmengine - INFO - Epoch(train) [1][3000/5758] lr: 6.0000e-04 eta: 12:45:54 time: 0.8031 data_time: 0.1000 memory: 25062 loss: 0.2707 +2023/06/01 16:03:37 - mmengine - INFO - Epoch(train) [1][3100/5758] lr: 6.0000e-04 eta: 12:44:40 time: 0.8376 data_time: 0.1345 memory: 25062 loss: 0.2056 +2023/06/01 16:05:00 - mmengine - INFO - Epoch(train) [1][3200/5758] lr: 6.0000e-04 eta: 12:42:57 time: 0.8092 data_time: 0.1070 memory: 25062 loss: 0.2621 +2023/06/01 16:06:22 - mmengine - INFO - Epoch(train) [1][3300/5758] lr: 6.0000e-04 eta: 12:40:59 time: 0.8065 data_time: 0.1042 memory: 25062 loss: 0.2359 +2023/06/01 16:07:45 - mmengine - INFO - Epoch(train) [1][3400/5758] lr: 6.0000e-04 eta: 12:39:12 time: 0.8174 data_time: 0.1120 memory: 25062 loss: 0.2258 +2023/06/01 16:09:08 - mmengine - INFO - Epoch(train) [1][3500/5758] lr: 6.0000e-04 eta: 12:37:41 time: 0.7992 data_time: 0.0951 memory: 25062 loss: 0.2161 +2023/06/01 16:10:30 - mmengine - INFO - Epoch(train) [1][3600/5758] lr: 6.0000e-04 eta: 12:35:36 time: 0.8226 data_time: 0.1195 memory: 25062 loss: 0.2434 +2023/06/01 16:11:53 - mmengine - INFO - Epoch(train) [1][3700/5758] lr: 6.0000e-04 eta: 12:34:01 time: 0.8323 data_time: 0.1300 memory: 25062 loss: 0.2233 +2023/06/01 16:13:16 - mmengine - INFO - Epoch(train) [1][3800/5758] lr: 6.0000e-04 eta: 12:32:21 time: 0.8118 data_time: 0.1066 memory: 25062 loss: 0.2265 +2023/06/01 16:14:38 - mmengine - INFO - Epoch(train) [1][3900/5758] lr: 6.0000e-04 eta: 12:30:31 time: 0.8279 data_time: 0.1231 memory: 25062 loss: 0.1960 +2023/06/01 16:16:00 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_lr6e-4_10epoch_20230601_151930 +2023/06/01 16:16:00 - mmengine - INFO - Epoch(train) [1][4000/5758] lr: 6.0000e-04 eta: 12:28:45 time: 0.7732 data_time: 0.0716 memory: 25062 loss: 0.2081 +2023/06/01 16:17:24 - mmengine - INFO - Epoch(train) [1][4100/5758] lr: 6.0000e-04 eta: 12:27:23 time: 0.8011 data_time: 0.0996 memory: 25062 loss: 0.2198 +2023/06/01 16:18:46 - mmengine - INFO - Epoch(train) [1][4200/5758] lr: 6.0000e-04 eta: 12:25:33 time: 0.7929 data_time: 0.0895 memory: 25062 loss: 0.2044 +2023/06/01 16:20:09 - mmengine - INFO - Epoch(train) [1][4300/5758] lr: 6.0000e-04 eta: 12:23:59 time: 0.8623 data_time: 0.1586 memory: 25062 loss: 0.1742 +2023/06/01 16:21:31 - mmengine - INFO - Epoch(train) [1][4400/5758] lr: 6.0000e-04 eta: 12:22:17 time: 0.8375 data_time: 0.1323 memory: 25062 loss: 0.1845 +2023/06/01 16:22:53 - mmengine - INFO - Epoch(train) [1][4500/5758] lr: 6.0000e-04 eta: 12:20:33 time: 0.8227 data_time: 0.1210 memory: 25062 loss: 0.1781 +2023/06/01 16:24:17 - mmengine - INFO - Epoch(train) [1][4600/5758] lr: 6.0000e-04 eta: 12:19:13 time: 0.8568 data_time: 0.1541 memory: 25062 loss: 0.3168 +2023/06/01 16:25:39 - mmengine - INFO - Epoch(train) [1][4700/5758] lr: 6.0000e-04 eta: 12:17:27 time: 0.8568 data_time: 0.1554 memory: 25062 loss: 0.1928 +2023/06/01 16:27:02 - mmengine - INFO - Epoch(train) [1][4800/5758] lr: 6.0000e-04 eta: 12:15:56 time: 0.8851 data_time: 0.1811 memory: 25062 loss: 0.2053 +2023/06/01 16:28:26 - mmengine - INFO - Epoch(train) [1][4900/5758] lr: 6.0000e-04 eta: 12:14:35 time: 0.8319 data_time: 0.1261 memory: 25062 loss: 0.1515 +2023/06/01 16:29:50 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_lr6e-4_10epoch_20230601_151930 +2023/06/01 16:29:50 - mmengine - INFO - Epoch(train) [1][5000/5758] lr: 6.0000e-04 eta: 12:13:17 time: 0.8575 data_time: 0.1549 memory: 25062 loss: 0.1652 +2023/06/01 16:31:13 - mmengine - INFO - Epoch(train) [1][5100/5758] lr: 6.0000e-04 eta: 12:11:49 time: 0.8082 data_time: 0.1015 memory: 25062 loss: 0.1640 +2023/06/01 16:32:37 - mmengine - INFO - Epoch(train) [1][5200/5758] lr: 6.0000e-04 eta: 12:10:22 time: 0.8294 data_time: 0.1258 memory: 25062 loss: 0.1739 +2023/06/01 16:34:00 - mmengine - INFO - Epoch(train) [1][5300/5758] lr: 6.0000e-04 eta: 12:08:58 time: 0.8580 data_time: 0.1554 memory: 25062 loss: 0.1747 +2023/06/01 16:35:25 - mmengine - INFO - Epoch(train) [1][5400/5758] lr: 6.0000e-04 eta: 12:07:43 time: 0.8157 data_time: 0.1134 memory: 25062 loss: 0.1565 +2023/06/01 16:36:48 - mmengine - INFO - Epoch(train) [1][5500/5758] lr: 6.0000e-04 eta: 12:06:19 time: 0.8382 data_time: 0.1363 memory: 25062 loss: 0.1694 +2023/06/01 16:38:12 - mmengine - INFO - Epoch(train) [1][5600/5758] lr: 6.0000e-04 eta: 12:04:57 time: 0.8262 data_time: 0.1222 memory: 25062 loss: 0.1364 +2023/06/01 16:39:34 - mmengine - INFO - Epoch(train) [1][5700/5758] lr: 6.0000e-04 eta: 12:03:17 time: 0.8234 data_time: 0.1201 memory: 25062 loss: 0.1674 +2023/06/01 16:40:24 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_lr6e-4_10epoch_20230601_151930 +2023/06/01 16:40:24 - mmengine - INFO - Saving checkpoint at 1 epochs +2023/06/01 16:40:45 - mmengine - INFO - Epoch(val) [1][16/16] accuracy/top1: 86.5223 single-label/precision_classwise: [100.0, 0.0] single-label/recall_classwise: [86.5223159790039, 0.0] single-label/f1-score_classwise: [92.77422332763672, 0.0] data_time: 0.3321 time: 0.8401 +2023/06/01 16:42:16 - mmengine - INFO - Epoch(train) [2][ 100/5758] lr: 5.8556e-04 eta: 12:02:16 time: 0.8460 data_time: 0.1329 memory: 25066 loss: 0.1236 +2023/06/01 16:43:40 - mmengine - INFO - Epoch(train) [2][ 200/5758] lr: 5.8556e-04 eta: 12:00:51 time: 0.7929 data_time: 0.0889 memory: 25066 loss: 0.1282 +2023/06/01 16:44:16 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_lr6e-4_10epoch_20230601_151930 +2023/06/01 16:45:04 - mmengine - INFO - Epoch(train) [2][ 300/5758] lr: 5.8556e-04 eta: 11:59:32 time: 0.8224 data_time: 0.0942 memory: 25066 loss: 0.1569 +2023/06/01 16:46:26 - mmengine - INFO - Epoch(train) [2][ 400/5758] lr: 5.8556e-04 eta: 11:57:55 time: 0.8277 data_time: 0.1253 memory: 25066 loss: 0.1417 +2023/06/01 16:47:51 - mmengine - INFO - Epoch(train) [2][ 500/5758] lr: 5.8556e-04 eta: 11:56:41 time: 0.8733 data_time: 0.1405 memory: 25066 loss: 0.1335 +2023/06/01 16:49:15 - mmengine - INFO - Epoch(train) [2][ 600/5758] lr: 5.8556e-04 eta: 11:55:18 time: 0.8575 data_time: 0.1550 memory: 25066 loss: 0.1063 +2023/06/01 16:50:39 - mmengine - INFO - Epoch(train) [2][ 700/5758] lr: 5.8556e-04 eta: 11:53:53 time: 0.9077 data_time: 0.1908 memory: 25066 loss: 0.1296 +2023/06/01 16:52:02 - mmengine - INFO - Epoch(train) [2][ 800/5758] lr: 5.8556e-04 eta: 11:52:23 time: 0.7918 data_time: 0.0883 memory: 25066 loss: 0.1345 +2023/06/01 16:53:27 - mmengine - INFO - Epoch(train) [2][ 900/5758] lr: 5.8556e-04 eta: 11:51:11 time: 0.8432 data_time: 0.1277 memory: 25066 loss: 0.1296 +2023/06/01 16:54:50 - mmengine - INFO - Epoch(train) [2][1000/5758] lr: 5.8556e-04 eta: 11:49:45 time: 0.8065 data_time: 0.1037 memory: 25066 loss: 0.1163 +2023/06/01 16:56:15 - mmengine - INFO - Epoch(train) [2][1100/5758] lr: 5.8556e-04 eta: 11:48:28 time: 0.8996 data_time: 0.1836 memory: 25066 loss: 0.1069 +2023/06/01 16:57:39 - mmengine - INFO - Epoch(train) [2][1200/5758] lr: 5.8556e-04 eta: 11:47:07 time: 0.8483 data_time: 0.1438 memory: 25066 loss: 0.1431 +2023/06/01 16:58:17 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_lr6e-4_10epoch_20230601_151930 +2023/06/01 16:59:05 - mmengine - INFO - Epoch(train) [2][1300/5758] lr: 5.8556e-04 eta: 11:45:55 time: 0.8595 data_time: 0.1570 memory: 25066 loss: 0.1091 +2023/06/01 17:00:28 - mmengine - INFO - Epoch(train) [2][1400/5758] lr: 5.8556e-04 eta: 11:44:25 time: 0.8275 data_time: 0.1254 memory: 25066 loss: 0.1010 +2023/06/01 17:01:50 - mmengine - INFO - Epoch(train) [2][1500/5758] lr: 5.8556e-04 eta: 11:42:49 time: 0.8394 data_time: 0.1369 memory: 25066 loss: 0.0947 +2023/06/01 17:03:13 - mmengine - INFO - Epoch(train) [2][1600/5758] lr: 5.8556e-04 eta: 11:41:21 time: 0.8770 data_time: 0.1731 memory: 25066 loss: 0.1035 +2023/06/01 17:04:35 - mmengine - INFO - Epoch(train) [2][1700/5758] lr: 5.8556e-04 eta: 11:39:42 time: 0.7741 data_time: 0.0710 memory: 25066 loss: 0.0897 +2023/06/01 17:05:57 - mmengine - INFO - Epoch(train) [2][1800/5758] lr: 5.8556e-04 eta: 11:38:08 time: 0.8122 data_time: 0.1096 memory: 25066 loss: 0.1070 +2023/06/01 17:07:20 - mmengine - INFO - Epoch(train) [2][1900/5758] lr: 5.8556e-04 eta: 11:36:41 time: 0.8073 data_time: 0.1047 memory: 25066 loss: 0.0824 +2023/06/01 17:08:44 - mmengine - INFO - Epoch(train) [2][2000/5758] lr: 5.8556e-04 eta: 11:35:21 time: 0.8282 data_time: 0.1264 memory: 25066 loss: 0.0888 +2023/06/01 17:10:09 - mmengine - INFO - Epoch(train) [2][2100/5758] lr: 5.8556e-04 eta: 11:34:05 time: 0.8363 data_time: 0.1342 memory: 25066 loss: 0.0893 +2023/06/01 17:11:33 - mmengine - INFO - Epoch(train) [2][2200/5758] lr: 5.8556e-04 eta: 11:32:41 time: 0.8177 data_time: 0.1141 memory: 25066 loss: 0.0837 +2023/06/01 17:12:08 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_lr6e-4_10epoch_20230601_151930 +2023/06/01 17:12:56 - mmengine - INFO - Epoch(train) [2][2300/5758] lr: 5.8556e-04 eta: 11:31:09 time: 0.8371 data_time: 0.1340 memory: 25066 loss: 0.0904 +2023/06/01 17:14:19 - mmengine - INFO - Epoch(train) [2][2400/5758] lr: 5.8556e-04 eta: 11:29:42 time: 0.8227 data_time: 0.1211 memory: 25066 loss: 0.0925 +2023/06/01 17:15:41 - mmengine - INFO - Epoch(train) [2][2500/5758] lr: 5.8556e-04 eta: 11:28:08 time: 0.8003 data_time: 0.0956 memory: 25066 loss: 0.0765 +2023/06/01 17:17:04 - mmengine - INFO - Epoch(train) [2][2600/5758] lr: 5.8556e-04 eta: 11:26:42 time: 0.8467 data_time: 0.1418 memory: 25066 loss: 0.0910 +2023/06/01 17:18:28 - mmengine - INFO - Epoch(train) [2][2700/5758] lr: 5.8556e-04 eta: 11:25:20 time: 0.8406 data_time: 0.1208 memory: 25066 loss: 0.0805 +2023/06/01 17:19:53 - mmengine - INFO - Epoch(train) [2][2800/5758] lr: 5.8556e-04 eta: 11:24:03 time: 0.8358 data_time: 0.1329 memory: 25066 loss: 0.0921 +2023/06/01 17:21:16 - mmengine - INFO - Epoch(train) [2][2900/5758] lr: 5.8556e-04 eta: 11:22:36 time: 0.8334 data_time: 0.1187 memory: 25066 loss: 0.0565 +2023/06/01 17:22:39 - mmengine - INFO - Epoch(train) [2][3000/5758] lr: 5.8556e-04 eta: 11:21:10 time: 0.8442 data_time: 0.1376 memory: 25066 loss: 0.0610 +2023/06/01 17:24:02 - mmengine - INFO - Epoch(train) [2][3100/5758] lr: 5.8556e-04 eta: 11:19:42 time: 0.8432 data_time: 0.1374 memory: 25066 loss: 0.0686 +2023/06/01 17:25:26 - mmengine - INFO - Epoch(train) [2][3200/5758] lr: 5.8556e-04 eta: 11:18:16 time: 0.8548 data_time: 0.1522 memory: 25066 loss: 0.0688 +2023/06/01 17:26:01 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_lr6e-4_10epoch_20230601_151930 +2023/06/01 17:26:49 - mmengine - INFO - Epoch(train) [2][3300/5758] lr: 5.8556e-04 eta: 11:16:53 time: 0.8283 data_time: 0.1184 memory: 25066 loss: 0.0735 +2023/06/01 17:28:14 - mmengine - INFO - Epoch(train) [2][3400/5758] lr: 5.8556e-04 eta: 11:15:35 time: 0.8156 data_time: 0.1136 memory: 25066 loss: 0.0726 +2023/06/01 17:29:38 - mmengine - INFO - Epoch(train) [2][3500/5758] lr: 5.8556e-04 eta: 11:14:10 time: 0.8736 data_time: 0.1600 memory: 25066 loss: 0.0768 +2023/06/01 17:31:01 - mmengine - INFO - Epoch(train) [2][3600/5758] lr: 5.8556e-04 eta: 11:12:45 time: 0.8076 data_time: 0.1054 memory: 25066 loss: 0.0873 +2023/06/01 17:32:25 - mmengine - INFO - Epoch(train) [2][3700/5758] lr: 5.8556e-04 eta: 11:11:21 time: 0.8640 data_time: 0.1572 memory: 25066 loss: 0.0549 +2023/06/01 17:33:51 - mmengine - INFO - Epoch(train) [2][3800/5758] lr: 5.8556e-04 eta: 11:10:08 time: 0.8968 data_time: 0.1907 memory: 25066 loss: 0.0641 +2023/06/01 17:35:14 - mmengine - INFO - Epoch(train) [2][3900/5758] lr: 5.8556e-04 eta: 11:08:45 time: 0.8787 data_time: 0.1615 memory: 25066 loss: 0.0804 +2023/06/01 17:36:38 - mmengine - INFO - Epoch(train) [2][4000/5758] lr: 5.8556e-04 eta: 11:07:20 time: 0.8570 data_time: 0.1535 memory: 25066 loss: 0.0699 +2023/06/01 17:38:01 - mmengine - INFO - Epoch(train) [2][4100/5758] lr: 5.8556e-04 eta: 11:05:54 time: 0.8303 data_time: 0.1127 memory: 25066 loss: 0.0612 +2023/06/01 17:39:26 - mmengine - INFO - Epoch(train) [2][4200/5758] lr: 5.8556e-04 eta: 11:04:36 time: 0.8912 data_time: 0.1861 memory: 25066 loss: 0.0609 +2023/06/01 17:40:02 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_lr6e-4_10epoch_20230601_151930 +2023/06/01 17:40:49 - mmengine - INFO - Epoch(train) [2][4300/5758] lr: 5.8556e-04 eta: 11:03:07 time: 0.8470 data_time: 0.1399 memory: 25066 loss: 0.0564 +2023/06/01 17:42:12 - mmengine - INFO - Epoch(train) [2][4400/5758] lr: 5.8556e-04 eta: 11:01:43 time: 0.9081 data_time: 0.2032 memory: 25066 loss: 0.0500 +2023/06/01 17:43:38 - mmengine - INFO - Epoch(train) [2][4500/5758] lr: 5.8556e-04 eta: 11:00:25 time: 0.8254 data_time: 0.1145 memory: 25066 loss: 0.0777 +2023/06/01 17:45:01 - mmengine - INFO - Epoch(train) [2][4600/5758] lr: 5.8556e-04 eta: 10:59:00 time: 0.8228 data_time: 0.1176 memory: 25066 loss: 0.0523 +2023/06/01 17:46:24 - mmengine - INFO - Epoch(train) [2][4700/5758] lr: 5.8556e-04 eta: 10:57:32 time: 0.7838 data_time: 0.0803 memory: 25066 loss: 0.0400 +2023/06/01 17:47:48 - mmengine - INFO - Epoch(train) [2][4800/5758] lr: 5.8556e-04 eta: 10:56:10 time: 0.8601 data_time: 0.1581 memory: 25066 loss: 0.0565 +2023/06/01 17:49:12 - mmengine - INFO - Epoch(train) [2][4900/5758] lr: 5.8556e-04 eta: 10:54:48 time: 0.8648 data_time: 0.1515 memory: 25066 loss: 0.0502 +2023/06/01 17:50:38 - mmengine - INFO - Epoch(train) [2][5000/5758] lr: 5.8556e-04 eta: 10:53:36 time: 0.8466 data_time: 0.1417 memory: 25066 loss: 0.0420 +2023/06/01 17:52:04 - mmengine - INFO - Epoch(train) [2][5100/5758] lr: 5.8556e-04 eta: 10:52:21 time: 0.8379 data_time: 0.1261 memory: 25066 loss: 0.0404 +2023/06/01 17:53:31 - mmengine - INFO - Epoch(train) [2][5200/5758] lr: 5.8556e-04 eta: 10:51:09 time: 0.8608 data_time: 0.1584 memory: 25066 loss: 0.0502 +2023/06/01 17:54:08 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_lr6e-4_10epoch_20230601_151930 +2023/06/01 17:54:56 - mmengine - INFO - Epoch(train) [2][5300/5758] lr: 5.8556e-04 eta: 10:49:52 time: 0.8829 data_time: 0.1704 memory: 25066 loss: 0.0507 +2023/06/01 17:56:20 - mmengine - INFO - Epoch(train) [2][5400/5758] lr: 5.8556e-04 eta: 10:48:26 time: 0.7996 data_time: 0.0969 memory: 25066 loss: 0.0393 +2023/06/01 17:57:44 - mmengine - INFO - Epoch(train) [2][5500/5758] lr: 5.8556e-04 eta: 10:47:03 time: 0.8025 data_time: 0.0999 memory: 25066 loss: 0.0495 +2023/06/01 17:59:08 - mmengine - INFO - Epoch(train) [2][5600/5758] lr: 5.8556e-04 eta: 10:45:44 time: 0.8860 data_time: 0.1823 memory: 25066 loss: 0.0388 +2023/06/01 18:00:30 - mmengine - INFO - Epoch(train) [2][5700/5758] lr: 5.8556e-04 eta: 10:44:10 time: 0.8243 data_time: 0.1130 memory: 25066 loss: 0.0503 +2023/06/01 18:01:17 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_lr6e-4_10epoch_20230601_151930 +2023/06/01 18:01:17 - mmengine - INFO - Saving checkpoint at 2 epochs +2023/06/01 18:01:35 - mmengine - INFO - Epoch(val) [2][16/16] accuracy/top1: 96.8310 single-label/precision_classwise: [100.0, 0.0] single-label/recall_classwise: [96.83100891113281, 0.0] single-label/f1-score_classwise: [98.3899917602539, 0.0] data_time: 0.2141 time: 0.6079 +2023/06/01 18:03:05 - mmengine - INFO - Epoch(train) [3][ 100/5758] lr: 5.4366e-04 eta: 10:42:20 time: 0.8666 data_time: 0.1303 memory: 25066 loss: 0.0332 +2023/06/01 18:04:27 - mmengine - INFO - Epoch(train) [3][ 200/5758] lr: 5.4366e-04 eta: 10:40:49 time: 0.8071 data_time: 0.1039 memory: 25066 loss: 0.0452 +2023/06/01 18:05:48 - mmengine - INFO - Epoch(train) [3][ 300/5758] lr: 5.4366e-04 eta: 10:39:15 time: 0.8046 data_time: 0.0973 memory: 25066 loss: 0.0478 +2023/06/01 18:07:09 - mmengine - INFO - Epoch(train) [3][ 400/5758] lr: 5.4366e-04 eta: 10:37:40 time: 0.8091 data_time: 0.1069 memory: 25066 loss: 0.0556 +2023/06/01 18:08:19 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_lr6e-4_10epoch_20230601_151930 +2023/06/01 18:08:32 - mmengine - INFO - Epoch(train) [3][ 500/5758] lr: 5.4366e-04 eta: 10:36:11 time: 0.8456 data_time: 0.1284 memory: 25066 loss: 0.0415 +2023/06/01 18:09:54 - mmengine - INFO - Epoch(train) [3][ 600/5758] lr: 5.4366e-04 eta: 10:34:41 time: 0.8118 data_time: 0.1078 memory: 25066 loss: 0.0395 +2023/06/01 18:11:17 - mmengine - INFO - Epoch(train) [3][ 700/5758] lr: 5.4366e-04 eta: 10:33:14 time: 0.8189 data_time: 0.1092 memory: 25066 loss: 0.0326 +2023/06/01 18:12:40 - mmengine - INFO - Epoch(train) [3][ 800/5758] lr: 5.4366e-04 eta: 10:31:48 time: 0.7846 data_time: 0.0818 memory: 25066 loss: 0.0232 +2023/06/01 18:14:04 - mmengine - INFO - Epoch(train) [3][ 900/5758] lr: 5.4366e-04 eta: 10:30:24 time: 0.8178 data_time: 0.0969 memory: 25066 loss: 0.0411 +2023/06/01 18:15:26 - mmengine - INFO - Epoch(train) [3][1000/5758] lr: 5.4366e-04 eta: 10:28:56 time: 0.8281 data_time: 0.1249 memory: 25066 loss: 0.0390 +2023/06/01 18:16:50 - mmengine - INFO - Epoch(train) [3][1100/5758] lr: 5.4366e-04 eta: 10:27:33 time: 0.8391 data_time: 0.1359 memory: 25066 loss: 0.0323 +2023/06/01 18:18:12 - mmengine - INFO - Epoch(train) [3][1200/5758] lr: 5.4366e-04 eta: 10:26:04 time: 0.9035 data_time: 0.1980 memory: 25066 loss: 0.0405 +2023/06/01 18:19:34 - mmengine - INFO - Epoch(train) [3][1300/5758] lr: 5.4366e-04 eta: 10:24:31 time: 0.8079 data_time: 0.1058 memory: 25066 loss: 0.0312 +2023/06/01 18:20:58 - mmengine - INFO - Epoch(train) [3][1400/5758] lr: 5.4366e-04 eta: 10:23:11 time: 0.8532 data_time: 0.1514 memory: 25066 loss: 0.0297 +2023/06/01 18:22:09 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_lr6e-4_10epoch_20230601_151930 +2023/06/01 18:22:21 - mmengine - INFO - Epoch(train) [3][1500/5758] lr: 5.4366e-04 eta: 10:21:43 time: 0.7585 data_time: 0.0560 memory: 25066 loss: 0.0340 +2023/06/01 18:23:43 - mmengine - INFO - Epoch(train) [3][1600/5758] lr: 5.4366e-04 eta: 10:20:14 time: 0.7842 data_time: 0.0829 memory: 25066 loss: 0.0368 +2023/06/01 18:25:07 - mmengine - INFO - Epoch(train) [3][1700/5758] lr: 5.4366e-04 eta: 10:18:50 time: 0.8231 data_time: 0.1196 memory: 25066 loss: 0.0488 +2023/06/01 18:26:30 - mmengine - INFO - Epoch(train) [3][1800/5758] lr: 5.4366e-04 eta: 10:17:24 time: 0.8201 data_time: 0.1072 memory: 25066 loss: 0.0294 +2023/06/01 18:27:52 - mmengine - INFO - Epoch(train) [3][1900/5758] lr: 5.4366e-04 eta: 10:15:54 time: 0.8074 data_time: 0.1044 memory: 25066 loss: 0.0255 +2023/06/01 18:29:16 - mmengine - INFO - Epoch(train) [3][2000/5758] lr: 5.4366e-04 eta: 10:14:32 time: 0.8549 data_time: 0.1520 memory: 25066 loss: 0.0351 +2023/06/01 18:30:40 - mmengine - INFO - Epoch(train) [3][2100/5758] lr: 5.4366e-04 eta: 10:13:09 time: 0.8145 data_time: 0.1107 memory: 25066 loss: 0.0320 +2023/06/01 18:32:03 - mmengine - INFO - Epoch(train) [3][2200/5758] lr: 5.4366e-04 eta: 10:11:44 time: 0.8259 data_time: 0.1212 memory: 25066 loss: 0.0195 +2023/06/01 18:33:25 - mmengine - INFO - Epoch(train) [3][2300/5758] lr: 5.4366e-04 eta: 10:10:17 time: 0.8362 data_time: 0.1334 memory: 25066 loss: 0.0317 +2023/06/01 18:34:47 - mmengine - INFO - Epoch(train) [3][2400/5758] lr: 5.4366e-04 eta: 10:08:48 time: 0.8608 data_time: 0.1493 memory: 25066 loss: 0.1196 +2023/06/01 18:35:59 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_lr6e-4_10epoch_20230601_151930 +2023/06/01 18:36:11 - mmengine - INFO - Epoch(train) [3][2500/5758] lr: 5.4366e-04 eta: 10:07:24 time: 0.8154 data_time: 0.1119 memory: 25066 loss: 0.0289 +2023/06/01 18:37:35 - mmengine - INFO - Epoch(train) [3][2600/5758] lr: 5.4366e-04 eta: 10:06:00 time: 0.8033 data_time: 0.0971 memory: 25066 loss: 0.0295 +2023/06/01 18:38:57 - mmengine - INFO - Epoch(train) [3][2700/5758] lr: 5.4366e-04 eta: 10:04:33 time: 0.8177 data_time: 0.1154 memory: 25066 loss: 0.0292 +2023/06/01 18:40:22 - mmengine - INFO - Epoch(train) [3][2800/5758] lr: 5.4366e-04 eta: 10:03:12 time: 0.9236 data_time: 0.2145 memory: 25066 loss: 0.0235 +2023/06/01 18:41:41 - mmengine - INFO - Epoch(train) [3][2900/5758] lr: 5.4366e-04 eta: 10:01:36 time: 0.7762 data_time: 0.0730 memory: 25066 loss: 0.0228 +2023/06/01 18:43:01 - mmengine - INFO - Epoch(train) [3][3000/5758] lr: 5.4366e-04 eta: 10:00:01 time: 0.7907 data_time: 0.0686 memory: 25066 loss: 0.0269 +2023/06/01 18:44:20 - mmengine - INFO - Epoch(train) [3][3100/5758] lr: 5.4366e-04 eta: 9:58:25 time: 0.8204 data_time: 0.1178 memory: 25066 loss: 0.0312 +2023/06/01 18:45:43 - mmengine - INFO - Epoch(train) [3][3200/5758] lr: 5.4366e-04 eta: 9:56:57 time: 0.7979 data_time: 0.0773 memory: 25066 loss: 0.0229 +2023/06/01 18:47:02 - mmengine - INFO - Epoch(train) [3][3300/5758] lr: 5.4366e-04 eta: 9:55:23 time: 0.7741 data_time: 0.0695 memory: 25066 loss: 0.0267 +2023/06/01 18:48:23 - mmengine - INFO - Epoch(train) [3][3400/5758] lr: 5.4366e-04 eta: 9:53:51 time: 0.8562 data_time: 0.1434 memory: 25066 loss: 0.0226 +2023/06/01 18:49:30 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_lr6e-4_10epoch_20230601_151930 +2023/06/01 18:49:42 - mmengine - INFO - Epoch(train) [3][3500/5758] lr: 5.4366e-04 eta: 9:52:15 time: 0.7817 data_time: 0.0794 memory: 25066 loss: 0.0282 +2023/06/01 18:51:00 - mmengine - INFO - Epoch(train) [3][3600/5758] lr: 5.4366e-04 eta: 9:50:37 time: 0.7718 data_time: 0.0611 memory: 25066 loss: 0.0231 +2023/06/01 18:52:19 - mmengine - INFO - Epoch(train) [3][3700/5758] lr: 5.4366e-04 eta: 9:48:59 time: 0.7899 data_time: 0.0868 memory: 25066 loss: 0.0299 +2023/06/01 18:53:38 - mmengine - INFO - Epoch(train) [3][3800/5758] lr: 5.4366e-04 eta: 9:47:25 time: 0.8594 data_time: 0.1504 memory: 25066 loss: 0.0297 +2023/06/01 18:54:58 - mmengine - INFO - Epoch(train) [3][3900/5758] lr: 5.4366e-04 eta: 9:45:52 time: 0.7856 data_time: 0.0828 memory: 25066 loss: 0.0262 +2023/06/01 18:56:19 - mmengine - INFO - Epoch(train) [3][4000/5758] lr: 5.4366e-04 eta: 9:44:21 time: 0.8091 data_time: 0.0846 memory: 25066 loss: 0.0214 +2023/06/01 18:57:39 - mmengine - INFO - Epoch(train) [3][4100/5758] lr: 5.4366e-04 eta: 9:42:50 time: 0.8409 data_time: 0.1375 memory: 25066 loss: 0.0231 +2023/06/01 18:59:00 - mmengine - INFO - Epoch(train) [3][4200/5758] lr: 5.4366e-04 eta: 9:41:21 time: 0.8298 data_time: 0.1235 memory: 25066 loss: 0.0274 +2023/06/01 19:00:20 - mmengine - INFO - Epoch(train) [3][4300/5758] lr: 5.4366e-04 eta: 9:39:47 time: 0.7840 data_time: 0.0806 memory: 25066 loss: 0.0249 +2023/06/01 19:01:40 - mmengine - INFO - Epoch(train) [3][4400/5758] lr: 5.4366e-04 eta: 9:38:17 time: 0.7858 data_time: 0.0803 memory: 25066 loss: 0.0181 +2023/06/01 19:02:50 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_lr6e-4_10epoch_20230601_151930 +2023/06/01 19:03:02 - mmengine - INFO - Epoch(train) [3][4500/5758] lr: 5.4366e-04 eta: 9:36:49 time: 0.8037 data_time: 0.1003 memory: 25066 loss: 0.0230 +2023/06/01 19:04:22 - mmengine - INFO - Epoch(train) [3][4600/5758] lr: 5.4366e-04 eta: 9:35:17 time: 0.7827 data_time: 0.0726 memory: 25066 loss: 0.0357 +2023/06/01 19:05:43 - mmengine - INFO - Epoch(train) [3][4700/5758] lr: 5.4366e-04 eta: 9:33:48 time: 0.8066 data_time: 0.1039 memory: 25066 loss: 0.0255 +2023/06/01 19:07:06 - mmengine - INFO - Epoch(train) [3][4800/5758] lr: 5.4366e-04 eta: 9:32:24 time: 0.8146 data_time: 0.1090 memory: 25066 loss: 0.0293 +2023/06/01 19:08:27 - mmengine - INFO - Epoch(train) [3][4900/5758] lr: 5.4366e-04 eta: 9:30:55 time: 0.8810 data_time: 0.1746 memory: 25066 loss: 0.0250 +2023/06/01 19:09:49 - mmengine - INFO - Epoch(train) [3][5000/5758] lr: 5.4366e-04 eta: 9:29:31 time: 0.8818 data_time: 0.1664 memory: 25066 loss: 0.0233 +2023/06/01 19:11:12 - mmengine - INFO - Epoch(train) [3][5100/5758] lr: 5.4366e-04 eta: 9:28:05 time: 0.8650 data_time: 0.1624 memory: 25066 loss: 0.0237 +2023/06/01 19:12:33 - mmengine - INFO - Epoch(train) [3][5200/5758] lr: 5.4366e-04 eta: 9:26:38 time: 0.8561 data_time: 0.1537 memory: 25066 loss: 0.0245 +2023/06/01 19:13:52 - mmengine - INFO - Epoch(train) [3][5300/5758] lr: 5.4366e-04 eta: 9:25:03 time: 0.8227 data_time: 0.1207 memory: 25066 loss: 0.0256 +2023/06/01 19:15:11 - mmengine - INFO - Epoch(train) [3][5400/5758] lr: 5.4366e-04 eta: 9:23:32 time: 0.7978 data_time: 0.0842 memory: 25066 loss: 0.0151 +2023/06/01 19:16:19 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_lr6e-4_10epoch_20230601_151930 +2023/06/01 19:16:31 - mmengine - INFO - Epoch(train) [3][5500/5758] lr: 5.4366e-04 eta: 9:22:01 time: 0.7742 data_time: 0.0689 memory: 25066 loss: 0.0265 +2023/06/01 19:17:50 - mmengine - INFO - Epoch(train) [3][5600/5758] lr: 5.4366e-04 eta: 9:20:28 time: 0.7887 data_time: 0.0831 memory: 25066 loss: 0.0193 +2023/06/01 19:19:10 - mmengine - INFO - Epoch(train) [3][5700/5758] lr: 5.4366e-04 eta: 9:18:57 time: 0.7398 data_time: 0.0376 memory: 25066 loss: 0.0174 +2023/06/01 19:19:56 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_lr6e-4_10epoch_20230601_151930 +2023/06/01 19:19:56 - mmengine - INFO - Saving checkpoint at 3 epochs +2023/06/01 19:20:14 - mmengine - INFO - Epoch(val) [3][16/16] accuracy/top1: 99.2614 single-label/precision_classwise: [100.0, 0.0] single-label/recall_classwise: [99.26140594482422, 0.0] single-label/f1-score_classwise: [99.62933349609375, 0.0] data_time: 0.2029 time: 0.5869 +2023/06/01 19:21:43 - mmengine - INFO - Epoch(train) [4][ 100/5758] lr: 4.7840e-04 eta: 9:16:54 time: 0.7977 data_time: 0.0880 memory: 25066 loss: 0.0183 +2023/06/01 19:23:03 - mmengine - INFO - Epoch(train) [4][ 200/5758] lr: 4.7840e-04 eta: 9:15:24 time: 0.7744 data_time: 0.0697 memory: 25066 loss: 0.0174 +2023/06/01 19:24:22 - mmengine - INFO - Epoch(train) [4][ 300/5758] lr: 4.7840e-04 eta: 9:13:52 time: 0.8122 data_time: 0.1092 memory: 25066 loss: 0.0190 +2023/06/01 19:25:43 - mmengine - INFO - Epoch(train) [4][ 400/5758] lr: 4.7840e-04 eta: 9:12:24 time: 0.7974 data_time: 0.0936 memory: 25066 loss: 0.0144 +2023/06/01 19:27:04 - mmengine - INFO - Epoch(train) [4][ 500/5758] lr: 4.7840e-04 eta: 9:10:57 time: 0.8005 data_time: 0.0910 memory: 25066 loss: 0.0114 +2023/06/01 19:28:25 - mmengine - INFO - Epoch(train) [4][ 600/5758] lr: 4.7840e-04 eta: 9:09:28 time: 0.8131 data_time: 0.1103 memory: 25066 loss: 0.0186 +2023/06/01 19:29:45 - mmengine - INFO - Epoch(train) [4][ 700/5758] lr: 4.7840e-04 eta: 9:07:58 time: 0.8075 data_time: 0.1023 memory: 25066 loss: 0.0180 +2023/06/01 19:30:06 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_lr6e-4_10epoch_20230601_151930 +2023/06/01 19:31:07 - mmengine - INFO - Epoch(train) [4][ 800/5758] lr: 4.7840e-04 eta: 9:06:33 time: 0.8489 data_time: 0.1443 memory: 25066 loss: 0.0145 +2023/06/01 19:32:26 - mmengine - INFO - Epoch(train) [4][ 900/5758] lr: 4.7840e-04 eta: 9:05:03 time: 0.7504 data_time: 0.0373 memory: 25066 loss: 0.0171 +2023/06/01 19:33:43 - mmengine - INFO - Epoch(train) [4][1000/5758] lr: 4.7840e-04 eta: 9:03:27 time: 0.8076 data_time: 0.1041 memory: 25066 loss: 0.0106 +2023/06/01 19:35:00 - mmengine - INFO - Epoch(train) [4][1100/5758] lr: 4.7840e-04 eta: 9:01:50 time: 0.7448 data_time: 0.0423 memory: 25066 loss: 0.0142 +2023/06/01 19:36:17 - mmengine - INFO - Epoch(train) [4][1200/5758] lr: 4.7840e-04 eta: 9:00:15 time: 0.8690 data_time: 0.1645 memory: 25066 loss: 0.0214 +2023/06/01 19:37:33 - mmengine - INFO - Epoch(train) [4][1300/5758] lr: 4.7840e-04 eta: 8:58:38 time: 0.7868 data_time: 0.0844 memory: 25066 loss: 0.0449 +2023/06/01 19:38:50 - mmengine - INFO - Epoch(train) [4][1400/5758] lr: 4.7840e-04 eta: 8:57:02 time: 0.7427 data_time: 0.0397 memory: 25066 loss: 0.0130 +2023/06/01 19:40:10 - mmengine - INFO - Epoch(train) [4][1500/5758] lr: 4.7840e-04 eta: 8:55:34 time: 0.7940 data_time: 0.0899 memory: 25066 loss: 0.0171 +2023/06/01 19:41:30 - mmengine - INFO - Epoch(train) [4][1600/5758] lr: 4.7840e-04 eta: 8:54:06 time: 0.7753 data_time: 0.0719 memory: 25066 loss: 0.0158 +2023/06/01 19:42:52 - mmengine - INFO - Epoch(train) [4][1700/5758] lr: 4.7840e-04 eta: 8:52:40 time: 0.8125 data_time: 0.1094 memory: 25066 loss: 0.0137 +2023/06/01 19:43:12 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_lr6e-4_10epoch_20230601_151930 +2023/06/01 19:44:13 - mmengine - INFO - Epoch(train) [4][1800/5758] lr: 4.7840e-04 eta: 8:51:14 time: 0.8262 data_time: 0.1217 memory: 25066 loss: 0.0137 +2023/06/01 19:45:34 - mmengine - INFO - Epoch(train) [4][1900/5758] lr: 4.7840e-04 eta: 8:49:49 time: 0.8281 data_time: 0.1258 memory: 25066 loss: 0.0109 +2023/06/01 19:46:53 - mmengine - INFO - Epoch(train) [4][2000/5758] lr: 4.7840e-04 eta: 8:48:18 time: 0.7801 data_time: 0.0404 memory: 25066 loss: 0.0128 +2023/06/01 19:48:15 - mmengine - INFO - Epoch(train) [4][2100/5758] lr: 4.7840e-04 eta: 8:46:54 time: 0.9109 data_time: 0.2039 memory: 25066 loss: 0.0141 +2023/06/01 19:49:36 - mmengine - INFO - Epoch(train) [4][2200/5758] lr: 4.7840e-04 eta: 8:45:27 time: 0.8495 data_time: 0.1468 memory: 25066 loss: 0.0352 +2023/06/01 19:50:57 - mmengine - INFO - Epoch(train) [4][2300/5758] lr: 4.7840e-04 eta: 8:44:00 time: 0.9044 data_time: 0.1897 memory: 25066 loss: 0.0136 +2023/06/01 19:52:17 - mmengine - INFO - Epoch(train) [4][2400/5758] lr: 4.7840e-04 eta: 8:42:33 time: 0.8408 data_time: 0.1303 memory: 25066 loss: 0.0113 +2023/06/01 19:53:36 - mmengine - INFO - Epoch(train) [4][2500/5758] lr: 4.7840e-04 eta: 8:41:04 time: 0.8577 data_time: 0.1554 memory: 25066 loss: 0.0236 +2023/06/01 19:54:53 - mmengine - INFO - Epoch(train) [4][2600/5758] lr: 4.7840e-04 eta: 8:39:30 time: 0.7522 data_time: 0.0396 memory: 25066 loss: 0.0146 +2023/06/01 19:56:12 - mmengine - INFO - Epoch(train) [4][2700/5758] lr: 4.7840e-04 eta: 8:37:59 time: 0.7691 data_time: 0.0670 memory: 25066 loss: 0.0159 +2023/06/01 19:56:31 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_lr6e-4_10epoch_20230601_151930 +2023/06/01 19:57:29 - mmengine - INFO - Epoch(train) [4][2800/5758] lr: 4.7840e-04 eta: 8:36:26 time: 0.7620 data_time: 0.0601 memory: 25066 loss: 0.0214 +2023/06/01 19:58:46 - mmengine - INFO - Epoch(train) [4][2900/5758] lr: 4.7840e-04 eta: 8:34:53 time: 0.8197 data_time: 0.1169 memory: 25066 loss: 0.0170 +2023/06/01 20:00:05 - mmengine - INFO - Epoch(train) [4][3000/5758] lr: 4.7840e-04 eta: 8:33:25 time: 0.8164 data_time: 0.1140 memory: 25066 loss: 0.0128 +2023/06/01 20:01:23 - mmengine - INFO - Epoch(train) [4][3100/5758] lr: 4.7840e-04 eta: 8:31:54 time: 0.7618 data_time: 0.0592 memory: 25066 loss: 0.0138 +2023/06/01 20:02:41 - mmengine - INFO - Epoch(train) [4][3200/5758] lr: 4.7840e-04 eta: 8:30:23 time: 0.8307 data_time: 0.1147 memory: 25066 loss: 0.0149 +2023/06/01 20:03:59 - mmengine - INFO - Epoch(train) [4][3300/5758] lr: 4.7840e-04 eta: 8:28:53 time: 0.7617 data_time: 0.0590 memory: 25066 loss: 0.0139 +2023/06/01 20:05:19 - mmengine - INFO - Epoch(train) [4][3400/5758] lr: 4.7840e-04 eta: 8:27:25 time: 0.7684 data_time: 0.0579 memory: 25066 loss: 0.0107 +2023/06/01 20:06:37 - mmengine - INFO - Epoch(train) [4][3500/5758] lr: 4.7840e-04 eta: 8:25:55 time: 0.8562 data_time: 0.1374 memory: 25066 loss: 0.0160 +2023/06/01 20:07:53 - mmengine - INFO - Epoch(train) [4][3600/5758] lr: 4.7840e-04 eta: 8:24:20 time: 0.7535 data_time: 0.0444 memory: 25066 loss: 0.0116 +2023/06/01 20:09:10 - mmengine - INFO - Epoch(train) [4][3700/5758] lr: 4.7840e-04 eta: 8:22:48 time: 0.7964 data_time: 0.0803 memory: 25066 loss: 0.0204 +2023/06/01 20:09:29 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_lr6e-4_10epoch_20230601_151930 +2023/06/01 20:10:27 - mmengine - INFO - Epoch(train) [4][3800/5758] lr: 4.7840e-04 eta: 8:21:16 time: 0.7654 data_time: 0.0586 memory: 25066 loss: 0.0104 +2023/06/01 20:11:44 - mmengine - INFO - Epoch(train) [4][3900/5758] lr: 4.7840e-04 eta: 8:19:45 time: 0.7723 data_time: 0.0686 memory: 25066 loss: 0.0090 +2023/06/01 20:13:02 - mmengine - INFO - Epoch(train) [4][4000/5758] lr: 4.7840e-04 eta: 8:18:16 time: 0.7557 data_time: 0.0459 memory: 25066 loss: 0.0172 +2023/06/01 20:14:19 - mmengine - INFO - Epoch(train) [4][4100/5758] lr: 4.7840e-04 eta: 8:16:44 time: 0.7566 data_time: 0.0481 memory: 25066 loss: 0.0155 +2023/06/01 20:15:35 - mmengine - INFO - Epoch(train) [4][4200/5758] lr: 4.7840e-04 eta: 8:15:11 time: 0.7266 data_time: 0.0138 memory: 25066 loss: 0.0147 +2023/06/01 20:16:53 - mmengine - INFO - Epoch(train) [4][4300/5758] lr: 4.7840e-04 eta: 8:13:42 time: 0.7723 data_time: 0.0567 memory: 25066 loss: 0.0101 +2023/06/01 20:18:10 - mmengine - INFO - Epoch(train) [4][4400/5758] lr: 4.7840e-04 eta: 8:12:11 time: 0.7344 data_time: 0.0318 memory: 25066 loss: 0.0158 +2023/06/01 20:19:28 - mmengine - INFO - Epoch(train) [4][4500/5758] lr: 4.7840e-04 eta: 8:10:41 time: 0.8160 data_time: 0.1009 memory: 25066 loss: 0.0076 +2023/06/01 20:20:47 - mmengine - INFO - Epoch(train) [4][4600/5758] lr: 4.7840e-04 eta: 8:09:15 time: 0.7799 data_time: 0.0775 memory: 25066 loss: 0.0258 +2023/06/01 20:22:07 - mmengine - INFO - Epoch(train) [4][4700/5758] lr: 4.7840e-04 eta: 8:07:48 time: 0.7870 data_time: 0.0748 memory: 25066 loss: 0.0096 +2023/06/01 20:22:27 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_lr6e-4_10epoch_20230601_151930 +2023/06/01 20:23:28 - mmengine - INFO - Epoch(train) [4][4800/5758] lr: 4.7840e-04 eta: 8:06:24 time: 0.8041 data_time: 0.1015 memory: 25066 loss: 0.0156 +2023/06/01 20:24:48 - mmengine - INFO - Epoch(train) [4][4900/5758] lr: 4.7840e-04 eta: 8:04:58 time: 0.7937 data_time: 0.0779 memory: 25066 loss: 0.0119 +2023/06/01 20:26:07 - mmengine - INFO - Epoch(train) [4][5000/5758] lr: 4.7840e-04 eta: 8:03:31 time: 0.7734 data_time: 0.0716 memory: 25066 loss: 0.0130 +2023/06/01 20:27:26 - mmengine - INFO - Epoch(train) [4][5100/5758] lr: 4.7840e-04 eta: 8:02:04 time: 0.7727 data_time: 0.0516 memory: 25066 loss: 0.0142 +2023/06/01 20:28:45 - mmengine - INFO - Epoch(train) [4][5200/5758] lr: 4.7840e-04 eta: 8:00:37 time: 0.7742 data_time: 0.0695 memory: 25066 loss: 0.0112 +2023/06/01 20:30:07 - mmengine - INFO - Epoch(train) [4][5300/5758] lr: 4.7840e-04 eta: 7:59:14 time: 0.8512 data_time: 0.1366 memory: 25066 loss: 0.0075 +2023/06/01 20:31:28 - mmengine - INFO - Epoch(train) [4][5400/5758] lr: 4.7840e-04 eta: 7:57:50 time: 0.8221 data_time: 0.1193 memory: 25066 loss: 0.0140 +2023/06/01 20:32:45 - mmengine - INFO - Epoch(train) [4][5500/5758] lr: 4.7840e-04 eta: 7:56:20 time: 0.7843 data_time: 0.0811 memory: 25066 loss: 0.0095 +2023/06/01 20:34:02 - mmengine - INFO - Epoch(train) [4][5600/5758] lr: 4.7840e-04 eta: 7:54:51 time: 0.7721 data_time: 0.0693 memory: 25066 loss: 0.0076 +2023/06/01 20:35:19 - mmengine - INFO - Epoch(train) [4][5700/5758] lr: 4.7840e-04 eta: 7:53:21 time: 0.7076 data_time: 0.0047 memory: 25066 loss: 0.0113 +2023/06/01 20:35:38 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_lr6e-4_10epoch_20230601_151930 +2023/06/01 20:36:03 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_lr6e-4_10epoch_20230601_151930 +2023/06/01 20:36:03 - mmengine - INFO - Saving checkpoint at 4 epochs +2023/06/01 20:36:21 - mmengine - INFO - Epoch(val) [4][16/16] accuracy/top1: 99.4445 single-label/precision_classwise: [100.0, 0.0] single-label/recall_classwise: [99.4444808959961, 0.0] single-label/f1-score_classwise: [99.72146606445312, 0.0] data_time: 0.2146 time: 0.6036 +2023/06/01 20:37:45 - mmengine - INFO - Epoch(train) [5][ 100/5758] lr: 3.9616e-04 eta: 7:51:09 time: 0.7603 data_time: 0.0226 memory: 25066 loss: 0.0041 +2023/06/01 20:39:05 - mmengine - INFO - Epoch(train) [5][ 200/5758] lr: 3.9616e-04 eta: 7:49:43 time: 0.8631 data_time: 0.1466 memory: 25066 loss: 0.0046 +2023/06/01 20:40:26 - mmengine - INFO - Epoch(train) [5][ 300/5758] lr: 3.9616e-04 eta: 7:48:20 time: 0.8112 data_time: 0.0918 memory: 25066 loss: 0.0160 +2023/06/01 20:41:44 - mmengine - INFO - Epoch(train) [5][ 400/5758] lr: 3.9616e-04 eta: 7:46:52 time: 0.7571 data_time: 0.0521 memory: 25066 loss: 0.0093 +2023/06/01 20:43:00 - mmengine - INFO - Epoch(train) [5][ 500/5758] lr: 3.9616e-04 eta: 7:45:22 time: 0.7910 data_time: 0.0658 memory: 25066 loss: 0.0051 +2023/06/01 20:44:17 - mmengine - INFO - Epoch(train) [5][ 600/5758] lr: 3.9616e-04 eta: 7:43:51 time: 0.7848 data_time: 0.0795 memory: 25066 loss: 0.0066 +2023/06/01 20:45:34 - mmengine - INFO - Epoch(train) [5][ 700/5758] lr: 3.9616e-04 eta: 7:42:22 time: 0.7471 data_time: 0.0404 memory: 25066 loss: 0.0131 +2023/06/01 20:46:48 - mmengine - INFO - Epoch(train) [5][ 800/5758] lr: 3.9616e-04 eta: 7:40:50 time: 0.7847 data_time: 0.0820 memory: 25066 loss: 0.0137 +2023/06/01 20:48:08 - mmengine - INFO - Epoch(train) [5][ 900/5758] lr: 3.9616e-04 eta: 7:39:24 time: 0.8649 data_time: 0.1619 memory: 25066 loss: 0.0061 +2023/06/01 20:49:03 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_lr6e-4_10epoch_20230601_151930 +2023/06/01 20:49:28 - mmengine - INFO - Epoch(train) [5][1000/5758] lr: 3.9616e-04 eta: 7:38:01 time: 0.7684 data_time: 0.0016 memory: 25066 loss: 0.0131 +2023/06/01 20:50:54 - mmengine - INFO - Epoch(train) [5][1100/5758] lr: 3.9616e-04 eta: 7:36:44 time: 0.8580 data_time: 0.0013 memory: 25066 loss: 0.0064 +2023/06/01 20:52:32 - mmengine - INFO - Epoch(train) [5][1200/5758] lr: 3.9616e-04 eta: 7:35:44 time: 0.8787 data_time: 0.0012 memory: 25066 loss: 0.0151 +2023/06/01 20:54:10 - mmengine - INFO - Epoch(train) [5][1300/5758] lr: 3.9616e-04 eta: 7:34:45 time: 1.0647 data_time: 0.0012 memory: 25066 loss: 0.0063 +2023/06/01 20:55:55 - mmengine - INFO - Epoch(train) [5][1400/5758] lr: 3.9616e-04 eta: 7:33:54 time: 1.0390 data_time: 0.1527 memory: 25066 loss: 0.0054 +2023/06/01 20:57:42 - mmengine - INFO - Epoch(train) [5][1500/5758] lr: 3.9616e-04 eta: 7:33:05 time: 1.2326 data_time: 0.2088 memory: 25066 loss: 0.0046 +2023/06/01 20:59:18 - mmengine - INFO - Epoch(train) [5][1600/5758] lr: 3.9616e-04 eta: 7:32:01 time: 0.7947 data_time: 0.0014 memory: 25066 loss: 0.0112 +2023/06/01 21:00:45 - mmengine - INFO - Epoch(train) [5][1700/5758] lr: 3.9616e-04 eta: 7:30:45 time: 0.8645 data_time: 0.0251 memory: 25066 loss: 0.0104 +2023/06/01 21:02:17 - mmengine - INFO - Epoch(train) [5][1800/5758] lr: 3.9616e-04 eta: 7:29:35 time: 0.8959 data_time: 0.0014 memory: 25066 loss: 0.0071 +2023/06/01 21:03:51 - mmengine - INFO - Epoch(train) [5][1900/5758] lr: 3.9616e-04 eta: 7:28:28 time: 1.0799 data_time: 0.0014 memory: 25066 loss: 0.0115 +2023/06/01 21:04:55 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_lr6e-4_10epoch_20230601_151930 +2023/06/01 21:05:23 - mmengine - INFO - Epoch(train) [5][2000/5758] lr: 3.9616e-04 eta: 7:27:18 time: 0.9674 data_time: 0.0013 memory: 25066 loss: 0.0123 +2023/06/01 21:07:13 - mmengine - INFO - Epoch(train) [5][2100/5758] lr: 3.9616e-04 eta: 7:26:31 time: 1.0896 data_time: 0.0014 memory: 25066 loss: 0.0064 +2023/06/01 21:09:08 - mmengine - INFO - Epoch(train) [5][2200/5758] lr: 3.9616e-04 eta: 7:25:51 time: 1.0883 data_time: 0.0016 memory: 25066 loss: 0.0020 +2023/06/01 21:10:58 - mmengine - INFO - Epoch(train) [5][2300/5758] lr: 3.9616e-04 eta: 7:25:02 time: 1.1327 data_time: 0.0015 memory: 25066 loss: 0.0078 +2023/06/01 21:13:09 - mmengine - INFO - Epoch(train) [5][2400/5758] lr: 3.9616e-04 eta: 7:24:40 time: 0.9786 data_time: 0.2644 memory: 25066 loss: 0.0088 +2023/06/01 21:14:55 - mmengine - INFO - Epoch(train) [5][2500/5758] lr: 3.9616e-04 eta: 7:23:46 time: 1.0893 data_time: 0.3752 memory: 25066 loss: 0.0072 +2023/06/01 21:16:45 - mmengine - INFO - Epoch(train) [5][2600/5758] lr: 3.9616e-04 eta: 7:22:57 time: 1.0284 data_time: 0.0013 memory: 25066 loss: 0.0122 +2023/06/01 21:18:06 - mmengine - INFO - Epoch(train) [5][2700/5758] lr: 3.9616e-04 eta: 7:21:31 time: 0.8187 data_time: 0.0014 memory: 25066 loss: 0.0119 +2023/06/01 21:19:25 - mmengine - INFO - Epoch(train) [5][2800/5758] lr: 3.9616e-04 eta: 7:20:02 time: 0.7864 data_time: 0.0015 memory: 25066 loss: 0.0098 +2023/06/01 21:20:45 - mmengine - INFO - Epoch(train) [5][2900/5758] lr: 3.9616e-04 eta: 7:18:36 time: 0.7898 data_time: 0.0013 memory: 25066 loss: 0.0045 +2023/06/01 21:21:40 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_lr6e-4_10epoch_20230601_151930 +2023/06/01 21:22:06 - mmengine - INFO - Epoch(train) [5][3000/5758] lr: 3.9616e-04 eta: 7:17:10 time: 0.8629 data_time: 0.0015 memory: 25066 loss: 0.0101 +2023/06/01 21:23:28 - mmengine - INFO - Epoch(train) [5][3100/5758] lr: 3.9616e-04 eta: 7:15:45 time: 0.8258 data_time: 0.0014 memory: 25066 loss: 0.0121 +2023/06/01 21:24:48 - mmengine - INFO - Epoch(train) [5][3200/5758] lr: 3.9616e-04 eta: 7:14:18 time: 0.8656 data_time: 0.0017 memory: 25066 loss: 0.0073 +2023/06/01 21:26:09 - mmengine - INFO - Epoch(train) [5][3300/5758] lr: 3.9616e-04 eta: 7:12:53 time: 0.7962 data_time: 0.0012 memory: 25066 loss: 0.0106 +2023/06/01 21:27:32 - mmengine - INFO - Epoch(train) [5][3400/5758] lr: 3.9616e-04 eta: 7:11:29 time: 0.8071 data_time: 0.0018 memory: 25066 loss: 0.0103 +2023/06/01 21:28:55 - mmengine - INFO - Epoch(train) [5][3500/5758] lr: 3.9616e-04 eta: 7:10:06 time: 0.8289 data_time: 0.0013 memory: 25066 loss: 0.0116 +2023/06/01 21:30:20 - mmengine - INFO - Epoch(train) [5][3600/5758] lr: 3.9616e-04 eta: 7:08:45 time: 0.7845 data_time: 0.0013 memory: 25066 loss: 0.0075 +2023/06/01 21:31:43 - mmengine - INFO - Epoch(train) [5][3700/5758] lr: 3.9616e-04 eta: 7:07:21 time: 0.8107 data_time: 0.0014 memory: 25066 loss: 0.0074 +2023/06/01 21:33:07 - mmengine - INFO - Epoch(train) [5][3800/5758] lr: 3.9616e-04 eta: 7:05:59 time: 0.8171 data_time: 0.0014 memory: 25066 loss: 0.0090 +2023/06/01 21:34:29 - mmengine - INFO - Epoch(train) [5][3900/5758] lr: 3.9616e-04 eta: 7:04:35 time: 0.7969 data_time: 0.0013 memory: 25066 loss: 0.0063 +2023/06/01 21:35:26 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_lr6e-4_10epoch_20230601_151930 +2023/06/01 21:35:52 - mmengine - INFO - Epoch(train) [5][4000/5758] lr: 3.9616e-04 eta: 7:03:12 time: 0.7924 data_time: 0.0015 memory: 25066 loss: 0.0065 +2023/06/01 21:37:15 - mmengine - INFO - Epoch(train) [5][4100/5758] lr: 3.9616e-04 eta: 7:01:49 time: 0.8515 data_time: 0.0014 memory: 25066 loss: 0.0109 +2023/06/01 21:38:39 - mmengine - INFO - Epoch(train) [5][4200/5758] lr: 3.9616e-04 eta: 7:00:26 time: 0.8348 data_time: 0.0014 memory: 25066 loss: 0.0048 +2023/06/01 21:40:02 - mmengine - INFO - Epoch(train) [5][4300/5758] lr: 3.9616e-04 eta: 6:59:03 time: 0.8007 data_time: 0.0014 memory: 25066 loss: 0.0115 +2023/06/01 21:41:25 - mmengine - INFO - Epoch(train) [5][4400/5758] lr: 3.9616e-04 eta: 6:57:40 time: 0.8031 data_time: 0.0012 memory: 25066 loss: 0.0048 +2023/06/01 21:42:50 - mmengine - INFO - Epoch(train) [5][4500/5758] lr: 3.9616e-04 eta: 6:56:19 time: 0.8247 data_time: 0.0013 memory: 25066 loss: 0.0016 +2023/06/01 21:44:12 - mmengine - INFO - Epoch(train) [5][4600/5758] lr: 3.9616e-04 eta: 6:54:55 time: 0.8476 data_time: 0.0020 memory: 25066 loss: 0.0078 +2023/06/01 21:45:35 - mmengine - INFO - Epoch(train) [5][4700/5758] lr: 3.9616e-04 eta: 6:53:31 time: 0.8104 data_time: 0.0013 memory: 25066 loss: 0.0076 +2023/06/01 21:46:58 - mmengine - INFO - Epoch(train) [5][4800/5758] lr: 3.9616e-04 eta: 6:52:08 time: 0.7907 data_time: 0.0012 memory: 25066 loss: 0.0078 +2023/06/01 21:48:21 - mmengine - INFO - Epoch(train) [5][4900/5758] lr: 3.9616e-04 eta: 6:50:45 time: 0.8167 data_time: 0.0014 memory: 25066 loss: 0.0071 +2023/06/01 21:49:18 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_lr6e-4_10epoch_20230601_151930 +2023/06/01 21:49:44 - mmengine - INFO - Epoch(train) [5][5000/5758] lr: 3.9616e-04 eta: 6:49:21 time: 0.8198 data_time: 0.0014 memory: 25066 loss: 0.0073 +2023/06/01 21:51:07 - mmengine - INFO - Epoch(train) [5][5100/5758] lr: 3.9616e-04 eta: 6:47:58 time: 0.8231 data_time: 0.0013 memory: 25066 loss: 0.0047 +2023/06/01 21:52:32 - mmengine - INFO - Epoch(train) [5][5200/5758] lr: 3.9616e-04 eta: 6:46:37 time: 0.8077 data_time: 0.0012 memory: 25066 loss: 0.0076 +2023/06/01 21:53:56 - mmengine - INFO - Epoch(train) [5][5300/5758] lr: 3.9616e-04 eta: 6:45:14 time: 0.8082 data_time: 0.0014 memory: 25066 loss: 0.0110 +2023/06/01 21:55:18 - mmengine - INFO - Epoch(train) [5][5400/5758] lr: 3.9616e-04 eta: 6:43:51 time: 0.8239 data_time: 0.0013 memory: 25066 loss: 0.0092 +2023/06/01 21:56:41 - mmengine - INFO - Epoch(train) [5][5500/5758] lr: 3.9616e-04 eta: 6:42:27 time: 0.8434 data_time: 0.0014 memory: 25066 loss: 0.0116 +2023/06/01 21:58:04 - mmengine - INFO - Epoch(train) [5][5600/5758] lr: 3.9616e-04 eta: 6:41:04 time: 0.8207 data_time: 0.0013 memory: 25066 loss: 0.0051 +2023/06/01 21:59:25 - mmengine - INFO - Epoch(train) [5][5700/5758] lr: 3.9616e-04 eta: 6:39:38 time: 0.8371 data_time: 0.0015 memory: 25066 loss: 0.0090 +2023/06/01 22:00:13 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_lr6e-4_10epoch_20230601_151930 +2023/06/01 22:00:13 - mmengine - INFO - Saving checkpoint at 5 epochs +2023/06/01 22:00:31 - mmengine - INFO - Epoch(val) [5][16/16] accuracy/top1: 99.7033 single-label/precision_classwise: [100.0, 0.0] single-label/recall_classwise: [99.70330047607422, 0.0] single-label/f1-score_classwise: [99.85142517089844, 0.0] data_time: 0.2072 time: 0.5988 +2023/06/01 22:01:59 - mmengine - INFO - Epoch(train) [6][ 100/5758] lr: 3.0500e-04 eta: 6:37:32 time: 0.8239 data_time: 0.0015 memory: 25066 loss: 0.0039 +2023/06/01 22:03:21 - mmengine - INFO - Epoch(train) [6][ 200/5758] lr: 3.0500e-04 eta: 6:36:08 time: 0.7924 data_time: 0.0012 memory: 25066 loss: 0.0027 +2023/06/01 22:03:28 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_lr6e-4_10epoch_20230601_151930 +2023/06/01 22:04:41 - mmengine - INFO - Epoch(train) [6][ 300/5758] lr: 3.0500e-04 eta: 6:34:42 time: 0.7972 data_time: 0.0014 memory: 25066 loss: 0.0057 +2023/06/01 22:06:03 - mmengine - INFO - Epoch(train) [6][ 400/5758] lr: 3.0500e-04 eta: 6:33:18 time: 0.7973 data_time: 0.0014 memory: 25066 loss: 0.0198 +2023/06/01 22:07:23 - mmengine - INFO - Epoch(train) [6][ 500/5758] lr: 3.0500e-04 eta: 6:31:52 time: 0.7932 data_time: 0.0012 memory: 25066 loss: 0.0064 +2023/06/01 22:08:45 - mmengine - INFO - Epoch(train) [6][ 600/5758] lr: 3.0500e-04 eta: 6:30:27 time: 0.8728 data_time: 0.0013 memory: 25066 loss: 0.0043 +2023/06/01 22:10:07 - mmengine - INFO - Epoch(train) [6][ 700/5758] lr: 3.0500e-04 eta: 6:29:03 time: 0.8821 data_time: 0.0012 memory: 25066 loss: 0.0060 +2023/06/01 22:11:30 - mmengine - INFO - Epoch(train) [6][ 800/5758] lr: 3.0500e-04 eta: 6:27:40 time: 0.7848 data_time: 0.0013 memory: 25066 loss: 0.0047 +2023/06/01 22:12:53 - mmengine - INFO - Epoch(train) [6][ 900/5758] lr: 3.0500e-04 eta: 6:26:17 time: 0.8502 data_time: 0.0016 memory: 25066 loss: 0.0055 +2023/06/01 22:14:15 - mmengine - INFO - Epoch(train) [6][1000/5758] lr: 3.0500e-04 eta: 6:24:52 time: 0.8248 data_time: 0.0012 memory: 25066 loss: 0.0079 +2023/06/01 22:15:38 - mmengine - INFO - Epoch(train) [6][1100/5758] lr: 3.0500e-04 eta: 6:23:29 time: 0.8590 data_time: 0.0126 memory: 25066 loss: 0.0032 +2023/06/01 22:17:01 - mmengine - INFO - Epoch(train) [6][1200/5758] lr: 3.0500e-04 eta: 6:22:06 time: 0.8617 data_time: 0.0013 memory: 25066 loss: 0.0093 +2023/06/01 22:17:09 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_lr6e-4_10epoch_20230601_151930 +2023/06/01 22:18:26 - mmengine - INFO - Epoch(train) [6][1300/5758] lr: 3.0500e-04 eta: 6:20:44 time: 0.8198 data_time: 0.0013 memory: 25066 loss: 0.0082 +2023/06/01 22:19:50 - mmengine - INFO - Epoch(train) [6][1400/5758] lr: 3.0500e-04 eta: 6:19:22 time: 0.9328 data_time: 0.0014 memory: 25066 loss: 0.0068 +2023/06/01 22:21:13 - mmengine - INFO - Epoch(train) [6][1500/5758] lr: 3.0500e-04 eta: 6:17:59 time: 0.8414 data_time: 0.0013 memory: 25066 loss: 0.0017 diff --git a/convnext_small_4xb256_fake5m_lr6e-4_10epoch/20230601_151930/vis_data/20230601_151930.json b/convnext_small_4xb256_fake5m_lr6e-4_10epoch/20230601_151930/vis_data/20230601_151930.json new file mode 100644 index 0000000000000000000000000000000000000000..82523aacdc675345c3e3955bc1b05e5a05b8a5c6 --- /dev/null +++ b/convnext_small_4xb256_fake5m_lr6e-4_10epoch/20230601_151930/vis_data/20230601_151930.json @@ -0,0 +1,305 @@ +{"lr": 0.0006, "data_time": 0.0015582561492919922, "loss": 0.6135746002197265, "time": 0.830146074295044, "epoch": 1, "memory": 25062, "step": 100} +{"lr": 0.0006, "data_time": 0.13062400817871095, "loss": 0.5753461420536041, "time": 0.846795630455017, "epoch": 1, "memory": 25062, "step": 200} +{"lr": 0.0006, "data_time": 0.12613627910614014, "loss": 0.5315729141235351, "time": 0.8305403709411621, "epoch": 1, "memory": 25062, "step": 300} +{"lr": 0.0006, "data_time": 0.13247416019439698, "loss": 0.5085993051528931, "time": 0.8568308115005493, "epoch": 1, "memory": 25062, "step": 400} +{"lr": 0.0006, "data_time": 0.16386404037475585, "loss": 0.500954395532608, "time": 0.8659586191177369, "epoch": 1, "memory": 25062, "step": 500} +{"lr": 0.0006, "data_time": 0.14253461360931396, "loss": 0.4692378222942352, "time": 0.8801122665405273, "epoch": 1, "memory": 25062, "step": 600} +{"lr": 0.0006, "data_time": 0.20899367332458496, "loss": 0.4822313070297241, "time": 0.913886308670044, "epoch": 1, "memory": 25062, "step": 700} +{"lr": 0.0006, "data_time": 0.1478217601776123, "loss": 0.43232129216194154, "time": 0.8784266710281372, "epoch": 1, "memory": 25062, "step": 800} +{"lr": 0.0006, "data_time": 0.12980871200561522, "loss": 0.4234696596860886, "time": 0.8333329200744629, "epoch": 1, "memory": 25062, "step": 900} +{"lr": 0.0006, "data_time": 0.10303521156311035, "loss": 0.43477452397346494, "time": 0.8239191770553589, "epoch": 1, "memory": 25062, "step": 1000} +{"lr": 0.0006, "data_time": 0.11415770053863525, "loss": 0.4187207818031311, "time": 0.8161101341247559, "epoch": 1, "memory": 25062, "step": 1100} +{"lr": 0.0006, "data_time": 0.160141658782959, "loss": 0.41021077036857606, "time": 0.8773677349090576, "epoch": 1, "memory": 25062, "step": 1200} +{"lr": 0.0006, "data_time": 0.06650912761688232, "loss": 0.42308527827262876, "time": 0.7686495304107666, "epoch": 1, "memory": 25062, "step": 1300} +{"lr": 0.0006, "data_time": 0.08822498321533204, "loss": 0.38337395489215853, "time": 0.8020293474197387, "epoch": 1, "memory": 25062, "step": 1400} +{"lr": 0.0006, "data_time": 0.14092085361480713, "loss": 0.3752002418041229, "time": 0.8428231716156006, "epoch": 1, "memory": 25062, "step": 1500} +{"lr": 0.0006, "data_time": 0.09997437000274659, "loss": 0.3994987905025482, "time": 0.8140286207199097, "epoch": 1, "memory": 25062, "step": 1600} +{"lr": 0.0006, "data_time": 0.12974863052368163, "loss": 0.3620677664875984, "time": 0.8324440002441407, "epoch": 1, "memory": 25062, "step": 1700} +{"lr": 0.0006, "data_time": 0.08871958255767823, "loss": 0.349988853931427, "time": 0.7991168022155761, "epoch": 1, "memory": 25062, "step": 1800} +{"lr": 0.0006, "data_time": 0.1101177453994751, "loss": 0.32603266537189485, "time": 0.813023567199707, "epoch": 1, "memory": 25062, "step": 1900} +{"lr": 0.0006, "data_time": 0.09395744800567626, "loss": 0.35006090998649597, "time": 0.8047962665557862, "epoch": 1, "memory": 25062, "step": 2000} +{"lr": 0.0006, "data_time": 0.12074847221374511, "loss": 0.344668909907341, "time": 0.8260648012161255, "epoch": 1, "memory": 25062, "step": 2100} +{"lr": 0.0006, "data_time": 0.14276158809661865, "loss": 0.3121231347322464, "time": 0.8568590641021728, "epoch": 1, "memory": 25062, "step": 2200} +{"lr": 0.0006, "data_time": 0.10163240432739258, "loss": 0.2936995327472687, "time": 0.8037424802780151, "epoch": 1, "memory": 25062, "step": 2300} +{"lr": 0.0006, "data_time": 0.06423301696777343, "loss": 0.321243605017662, "time": 0.780709195137024, "epoch": 1, "memory": 25062, "step": 2400} +{"lr": 0.0006, "data_time": 0.11378779411315917, "loss": 0.28465537130832674, "time": 0.8197810888290405, "epoch": 1, "memory": 25062, "step": 2500} +{"lr": 0.0006, "data_time": 0.15081541538238524, "loss": 0.28395094722509384, "time": 0.852768349647522, "epoch": 1, "memory": 25062, "step": 2600} +{"lr": 0.0006, "data_time": 0.16163480281829834, "loss": 0.2992450505495071, "time": 0.8647766828536987, "epoch": 1, "memory": 25062, "step": 2700} +{"lr": 0.0006, "data_time": 0.12410805225372315, "loss": 0.2752879589796066, "time": 0.8262264490127563, "epoch": 1, "memory": 25062, "step": 2800} +{"lr": 0.0006, "data_time": 0.15808987617492676, "loss": 0.2809392124414444, "time": 0.8610882043838501, "epoch": 1, "memory": 25062, "step": 2900} +{"lr": 0.0006, "data_time": 0.10003814697265626, "loss": 0.2706796482205391, "time": 0.8030795574188232, "epoch": 1, "memory": 25062, "step": 3000} +{"lr": 0.0006, "data_time": 0.1345454216003418, "loss": 0.2056187316775322, "time": 0.8375606536865234, "epoch": 1, "memory": 25062, "step": 3100} +{"lr": 0.0006, "data_time": 0.10702877044677735, "loss": 0.2620606556534767, "time": 0.8091813564300537, "epoch": 1, "memory": 25062, "step": 3200} +{"lr": 0.0006, "data_time": 0.10420176982879639, "loss": 0.23590899109840394, "time": 0.8064955234527588, "epoch": 1, "memory": 25062, "step": 3300} +{"lr": 0.0006, "data_time": 0.11196503639221192, "loss": 0.2258203849196434, "time": 0.8173755168914795, "epoch": 1, "memory": 25062, "step": 3400} +{"lr": 0.0006, "data_time": 0.09510846138000488, "loss": 0.2160615622997284, "time": 0.7991591930389405, "epoch": 1, "memory": 25062, "step": 3500} +{"lr": 0.0006, "data_time": 0.11947128772735596, "loss": 0.24335542768239976, "time": 0.822614598274231, "epoch": 1, "memory": 25062, "step": 3600} +{"lr": 0.0006, "data_time": 0.13002800941467285, "loss": 0.22328522503376008, "time": 0.8323217630386353, "epoch": 1, "memory": 25062, "step": 3700} +{"lr": 0.0006, "data_time": 0.10663254261016845, "loss": 0.22647091150283813, "time": 0.8117939949035644, "epoch": 1, "memory": 25062, "step": 3800} +{"lr": 0.0006, "data_time": 0.12309346199035645, "loss": 0.19595374017953873, "time": 0.8279386043548584, "epoch": 1, "memory": 25062, "step": 3900} +{"lr": 0.0006, "data_time": 0.07163856029510499, "loss": 0.20814566165208817, "time": 0.7732212781906128, "epoch": 1, "memory": 25062, "step": 4000} +{"lr": 0.0006, "data_time": 0.09958779811859131, "loss": 0.21983536332845688, "time": 0.8011039018630981, "epoch": 1, "memory": 25062, "step": 4100} +{"lr": 0.0006, "data_time": 0.0894810438156128, "loss": 0.2043963760137558, "time": 0.7928557157516479, "epoch": 1, "memory": 25062, "step": 4200} +{"lr": 0.0006, "data_time": 0.15857112407684326, "loss": 0.17422495484352113, "time": 0.8622824192047119, "epoch": 1, "memory": 25062, "step": 4300} +{"lr": 0.0006, "data_time": 0.13226869106292724, "loss": 0.18446631729602814, "time": 0.8374736547470093, "epoch": 1, "memory": 25062, "step": 4400} +{"lr": 0.0006, "data_time": 0.12099769115447997, "loss": 0.17812515795230865, "time": 0.8227338790893555, "epoch": 1, "memory": 25062, "step": 4500} +{"lr": 0.0006, "data_time": 0.15409488677978517, "loss": 0.31683850288391113, "time": 0.8568028450012207, "epoch": 1, "memory": 25062, "step": 4600} +{"lr": 0.0006, "data_time": 0.1554493188858032, "loss": 0.19284236431121826, "time": 0.8568015813827514, "epoch": 1, "memory": 25062, "step": 4700} +{"lr": 0.0006, "data_time": 0.1810582160949707, "loss": 0.20528918504714966, "time": 0.8851372718811035, "epoch": 1, "memory": 25062, "step": 4800} +{"lr": 0.0006, "data_time": 0.12607507705688475, "loss": 0.15152284875512123, "time": 0.8319072723388672, "epoch": 1, "memory": 25062, "step": 4900} +{"lr": 0.0006, "data_time": 0.15493619441986084, "loss": 0.16518262326717376, "time": 0.8574823617935181, "epoch": 1, "memory": 25062, "step": 5000} +{"lr": 0.0006, "data_time": 0.1015016794204712, "loss": 0.16402522027492522, "time": 0.8082464933395386, "epoch": 1, "memory": 25062, "step": 5100} +{"lr": 0.0006, "data_time": 0.12581818103790282, "loss": 0.17387931197881698, "time": 0.8294331312179566, "epoch": 1, "memory": 25062, "step": 5200} +{"lr": 0.0006, "data_time": 0.15540964603424073, "loss": 0.17468418180942535, "time": 0.8579760789871216, "epoch": 1, "memory": 25062, "step": 5300} +{"lr": 0.0006, "data_time": 0.1133528470993042, "loss": 0.15653457567095758, "time": 0.8156604290008544, "epoch": 1, "memory": 25062, "step": 5400} +{"lr": 0.0006, "data_time": 0.13631539344787597, "loss": 0.16939338743686677, "time": 0.8382399320602417, "epoch": 1, "memory": 25062, "step": 5500} +{"lr": 0.0006, "data_time": 0.12221121788024902, "loss": 0.13643198758363723, "time": 0.8262169361114502, "epoch": 1, "memory": 25062, "step": 5600} +{"lr": 0.0006, "data_time": 0.12007019519805909, "loss": 0.1673684224486351, "time": 0.8233999013900757, "epoch": 1, "memory": 25062, "step": 5700} +{"accuracy/top1": 86.5223159790039, "data_time": 0.332147479057312, "time": 0.8400849550962448, "step": 1} +{"lr": 0.0005855616723070703, "data_time": 0.1329329013824463, "loss": 0.12358986958861351, "time": 0.8460269212722779, "epoch": 2, "memory": 25066, "step": 5858} +{"lr": 0.0005855616723070703, "data_time": 0.08893582820892335, "loss": 0.1282133489847183, "time": 0.7928776741027832, "epoch": 2, "memory": 25066, "step": 5958} +{"lr": 0.0005855616723070703, "data_time": 0.09415514469146728, "loss": 0.1568968787789345, "time": 0.8223574161529541, "epoch": 2, "memory": 25066, "step": 6058} +{"lr": 0.0005855616723070703, "data_time": 0.12533230781555177, "loss": 0.14167584106326103, "time": 0.8277125358581543, "epoch": 2, "memory": 25066, "step": 6158} +{"lr": 0.0005855616723070703, "data_time": 0.1404815435409546, "loss": 0.1335474856197834, "time": 0.8733075857162476, "epoch": 2, "memory": 25066, "step": 6258} +{"lr": 0.0005855616723070703, "data_time": 0.15498428344726561, "loss": 0.1062768466770649, "time": 0.8575185537338257, "epoch": 2, "memory": 25066, "step": 6358} +{"lr": 0.0005855616723070703, "data_time": 0.19075372219085693, "loss": 0.12959541007876396, "time": 0.9077394723892211, "epoch": 2, "memory": 25066, "step": 6458} +{"lr": 0.0005855616723070703, "data_time": 0.08825895786285401, "loss": 0.13451345264911652, "time": 0.7918497323989868, "epoch": 2, "memory": 25066, "step": 6558} +{"lr": 0.0005855616723070703, "data_time": 0.12768311500549318, "loss": 0.12959048748016358, "time": 0.8432443618774415, "epoch": 2, "memory": 25066, "step": 6658} +{"lr": 0.0005855616723070703, "data_time": 0.10369279384613037, "loss": 0.11627456694841384, "time": 0.8065032243728638, "epoch": 2, "memory": 25066, "step": 6758} +{"lr": 0.0005855616723070703, "data_time": 0.18359854221343994, "loss": 0.1069206453859806, "time": 0.8995799779891968, "epoch": 2, "memory": 25066, "step": 6858} +{"lr": 0.0005855616723070703, "data_time": 0.1437976360321045, "loss": 0.1431097500026226, "time": 0.8483048439025879, "epoch": 2, "memory": 25066, "step": 6958} +{"lr": 0.0005855616723070703, "data_time": 0.15699679851531984, "loss": 0.10914113745093346, "time": 0.8594900846481324, "epoch": 2, "memory": 25066, "step": 7058} +{"lr": 0.0005855616723070703, "data_time": 0.12544469833374022, "loss": 0.10099955797195434, "time": 0.8274879693984986, "epoch": 2, "memory": 25066, "step": 7158} +{"lr": 0.0005855616723070703, "data_time": 0.1368650197982788, "loss": 0.09467699006199837, "time": 0.8394469976425171, "epoch": 2, "memory": 25066, "step": 7258} +{"lr": 0.0005855616723070703, "data_time": 0.17312967777252197, "loss": 0.10345427542924882, "time": 0.8770464181900024, "epoch": 2, "memory": 25066, "step": 7358} +{"lr": 0.0005855616723070703, "data_time": 0.07103266716003417, "loss": 0.08968858756124973, "time": 0.7741149187088012, "epoch": 2, "memory": 25066, "step": 7458} +{"lr": 0.0005855616723070703, "data_time": 0.10962119102478027, "loss": 0.10697086453437805, "time": 0.8122239589691163, "epoch": 2, "memory": 25066, "step": 7558} +{"lr": 0.0005855616723070703, "data_time": 0.1046525478363037, "loss": 0.08235031291842461, "time": 0.8072562932968139, "epoch": 2, "memory": 25066, "step": 7658} +{"lr": 0.0005855616723070703, "data_time": 0.1263685941696167, "loss": 0.08875132799148559, "time": 0.8282488107681274, "epoch": 2, "memory": 25066, "step": 7758} +{"lr": 0.0005855616723070703, "data_time": 0.1342289686203003, "loss": 0.08926248177886009, "time": 0.83630530834198, "epoch": 2, "memory": 25066, "step": 7858} +{"lr": 0.0005855616723070703, "data_time": 0.11405665874481201, "loss": 0.0836502704769373, "time": 0.8176812887191772, "epoch": 2, "memory": 25066, "step": 7958} +{"lr": 0.0005855616723070703, "data_time": 0.13397748470306398, "loss": 0.09036344364285469, "time": 0.8371359586715699, "epoch": 2, "memory": 25066, "step": 8058} +{"lr": 0.0005855616723070703, "data_time": 0.12112700939178467, "loss": 0.09246006608009338, "time": 0.8227143049240112, "epoch": 2, "memory": 25066, "step": 8158} +{"lr": 0.0005855616723070703, "data_time": 0.09559087753295899, "loss": 0.076456942781806, "time": 0.8003002643585205, "epoch": 2, "memory": 25066, "step": 8258} +{"lr": 0.0005855616723070703, "data_time": 0.14178693294525146, "loss": 0.09103170521557331, "time": 0.8466941118240356, "epoch": 2, "memory": 25066, "step": 8358} +{"lr": 0.0005855616723070703, "data_time": 0.1208219051361084, "loss": 0.08048920668661594, "time": 0.8405956506729126, "epoch": 2, "memory": 25066, "step": 8458} +{"lr": 0.0005855616723070703, "data_time": 0.13285036087036134, "loss": 0.09214386157691479, "time": 0.8357965469360351, "epoch": 2, "memory": 25066, "step": 8558} +{"lr": 0.0005855616723070703, "data_time": 0.11866247653961182, "loss": 0.05654298588633537, "time": 0.8333797693252564, "epoch": 2, "memory": 25066, "step": 8658} +{"lr": 0.0005855616723070703, "data_time": 0.137612247467041, "loss": 0.061009171232581136, "time": 0.8441805124282837, "epoch": 2, "memory": 25066, "step": 8758} +{"lr": 0.0005855616723070703, "data_time": 0.13744451999664306, "loss": 0.06859854981303215, "time": 0.8431969404220581, "epoch": 2, "memory": 25066, "step": 8858} +{"lr": 0.0005855616723070703, "data_time": 0.15220577716827394, "loss": 0.06884837374091149, "time": 0.8548434734344482, "epoch": 2, "memory": 25066, "step": 8958} +{"lr": 0.0005855616723070703, "data_time": 0.11841278076171875, "loss": 0.07345729060471058, "time": 0.8282695770263672, "epoch": 2, "memory": 25066, "step": 9058} +{"lr": 0.0005855616723070703, "data_time": 0.11359946727752686, "loss": 0.07261206656694412, "time": 0.8156155586242676, "epoch": 2, "memory": 25066, "step": 9158} +{"lr": 0.0005855616723070703, "data_time": 0.16002295017242432, "loss": 0.07682518362998962, "time": 0.8735629081726074, "epoch": 2, "memory": 25066, "step": 9258} +{"lr": 0.0005855616723070703, "data_time": 0.1053574800491333, "loss": 0.0873219683766365, "time": 0.8075915098190307, "epoch": 2, "memory": 25066, "step": 9358} +{"lr": 0.0005855616723070703, "data_time": 0.15722007751464845, "loss": 0.054900694824755195, "time": 0.864040207862854, "epoch": 2, "memory": 25066, "step": 9458} +{"lr": 0.0005855616723070703, "data_time": 0.1907418489456177, "loss": 0.06413311734795571, "time": 0.8967849969863891, "epoch": 2, "memory": 25066, "step": 9558} +{"lr": 0.0005855616723070703, "data_time": 0.16151020526885987, "loss": 0.08038142397999763, "time": 0.8787475824356079, "epoch": 2, "memory": 25066, "step": 9658} +{"lr": 0.0005855616723070703, "data_time": 0.15345432758331298, "loss": 0.06994653046131134, "time": 0.8569607973098755, "epoch": 2, "memory": 25066, "step": 9758} +{"lr": 0.0005855616723070703, "data_time": 0.11273493766784667, "loss": 0.06123212277889252, "time": 0.8303402900695801, "epoch": 2, "memory": 25066, "step": 9858} +{"lr": 0.0005855616723070703, "data_time": 0.1861478328704834, "loss": 0.06094014495611191, "time": 0.891200065612793, "epoch": 2, "memory": 25066, "step": 9958} +{"lr": 0.0005855616723070703, "data_time": 0.13991703987121581, "loss": 0.05642457380890846, "time": 0.8469839334487915, "epoch": 2, "memory": 25066, "step": 10058} +{"lr": 0.0005855616723070703, "data_time": 0.203238844871521, "loss": 0.04999357163906097, "time": 0.908148193359375, "epoch": 2, "memory": 25066, "step": 10158} +{"lr": 0.0005855616723070703, "data_time": 0.11449542045593261, "loss": 0.07773915491998196, "time": 0.8254468202590942, "epoch": 2, "memory": 25066, "step": 10258} +{"lr": 0.0005855616723070703, "data_time": 0.11759653091430664, "loss": 0.05232198443263769, "time": 0.8228446960449218, "epoch": 2, "memory": 25066, "step": 10358} +{"lr": 0.0005855616723070703, "data_time": 0.08028585910797119, "loss": 0.04003726141527295, "time": 0.7837674856185913, "epoch": 2, "memory": 25066, "step": 10458} +{"lr": 0.0005855616723070703, "data_time": 0.1580594778060913, "loss": 0.056496366113424304, "time": 0.8600718736648559, "epoch": 2, "memory": 25066, "step": 10558} +{"lr": 0.0005855616723070703, "data_time": 0.15153591632843016, "loss": 0.05019565653055906, "time": 0.8647865056991577, "epoch": 2, "memory": 25066, "step": 10658} +{"lr": 0.0005855616723070703, "data_time": 0.14174971580505372, "loss": 0.04199137855321169, "time": 0.8465540170669555, "epoch": 2, "memory": 25066, "step": 10758} +{"lr": 0.0005855616723070703, "data_time": 0.12608003616333008, "loss": 0.040434060338884595, "time": 0.8379182577133178, "epoch": 2, "memory": 25066, "step": 10858} +{"lr": 0.0005855616723070703, "data_time": 0.15844414234161378, "loss": 0.05015103369951248, "time": 0.8607777118682861, "epoch": 2, "memory": 25066, "step": 10958} +{"lr": 0.0005855616723070703, "data_time": 0.17037336826324462, "loss": 0.05067151039838791, "time": 0.8828636169433594, "epoch": 2, "memory": 25066, "step": 11058} +{"lr": 0.0005855616723070703, "data_time": 0.09685750007629394, "loss": 0.039331309497356415, "time": 0.7996238470077515, "epoch": 2, "memory": 25066, "step": 11158} +{"lr": 0.0005855616723070703, "data_time": 0.099875807762146, "loss": 0.04945165850222111, "time": 0.8025046825408936, "epoch": 2, "memory": 25066, "step": 11258} +{"lr": 0.0005855616723070703, "data_time": 0.18228275775909425, "loss": 0.03875251393765211, "time": 0.8860106945037842, "epoch": 2, "memory": 25066, "step": 11358} +{"lr": 0.0005855616723070703, "data_time": 0.11300158500671387, "loss": 0.050302263163030145, "time": 0.8242648363113403, "epoch": 2, "memory": 25066, "step": 11458} +{"accuracy/top1": 96.83100891113281, "data_time": 0.2140779214746812, "time": 0.6079331706551945, "step": 2} +{"lr": 0.0005436600133406095, "data_time": 0.13033041954040528, "loss": 0.033195352740585805, "time": 0.8666026592254639, "epoch": 3, "memory": 25066, "step": 11616} +{"lr": 0.0005436600133406095, "data_time": 0.10392272472381592, "loss": 0.04524232670664787, "time": 0.8071052074432373, "epoch": 3, "memory": 25066, "step": 11716} +{"lr": 0.0005436600133406095, "data_time": 0.09729797840118408, "loss": 0.04777164980769157, "time": 0.80458664894104, "epoch": 3, "memory": 25066, "step": 11816} +{"lr": 0.0005436600133406095, "data_time": 0.10693371295928955, "loss": 0.05555416196584702, "time": 0.8090775728225708, "epoch": 3, "memory": 25066, "step": 11916} +{"lr": 0.0005436600133406095, "data_time": 0.12842955589294433, "loss": 0.0414990421384573, "time": 0.8455898284912109, "epoch": 3, "memory": 25066, "step": 12016} +{"lr": 0.0005436600133406095, "data_time": 0.10779645442962646, "loss": 0.0394628276117146, "time": 0.8118354082107544, "epoch": 3, "memory": 25066, "step": 12116} +{"lr": 0.0005436600133406095, "data_time": 0.10915007591247558, "loss": 0.03261984065175057, "time": 0.8189421415328979, "epoch": 3, "memory": 25066, "step": 12216} +{"lr": 0.0005436600133406095, "data_time": 0.08182039260864257, "loss": 0.023212652560323475, "time": 0.7845738649368286, "epoch": 3, "memory": 25066, "step": 12316} +{"lr": 0.0005436600133406095, "data_time": 0.09693608283996583, "loss": 0.04113286808133125, "time": 0.8177855253219605, "epoch": 3, "memory": 25066, "step": 12416} +{"lr": 0.0005436600133406095, "data_time": 0.12489337921142578, "loss": 0.03896365687251091, "time": 0.8280580043792725, "epoch": 3, "memory": 25066, "step": 12516} +{"lr": 0.0005436600133406095, "data_time": 0.13593392372131347, "loss": 0.032323870621621606, "time": 0.8391335487365723, "epoch": 3, "memory": 25066, "step": 12616} +{"lr": 0.0005436600133406095, "data_time": 0.19797844886779786, "loss": 0.04052624525502324, "time": 0.9034525632858277, "epoch": 3, "memory": 25066, "step": 12716} +{"lr": 0.0005436600133406095, "data_time": 0.10575785636901855, "loss": 0.031234129518270492, "time": 0.8078648328781128, "epoch": 3, "memory": 25066, "step": 12816} +{"lr": 0.0005436600133406095, "data_time": 0.15137965679168702, "loss": 0.02967866063117981, "time": 0.853244137763977, "epoch": 3, "memory": 25066, "step": 12916} +{"lr": 0.0005436600133406095, "data_time": 0.05598795413970947, "loss": 0.03397217188030481, "time": 0.7584885597229004, "epoch": 3, "memory": 25066, "step": 13016} +{"lr": 0.0005436600133406095, "data_time": 0.08288066387176514, "loss": 0.0368267472833395, "time": 0.7842445850372315, "epoch": 3, "memory": 25066, "step": 13116} +{"lr": 0.0005436600133406095, "data_time": 0.11959879398345948, "loss": 0.04880100060254335, "time": 0.8230604887008667, "epoch": 3, "memory": 25066, "step": 13216} +{"lr": 0.0005436600133406095, "data_time": 0.10724380016326904, "loss": 0.029367662314325572, "time": 0.820131516456604, "epoch": 3, "memory": 25066, "step": 13316} +{"lr": 0.0005436600133406095, "data_time": 0.10439128875732422, "loss": 0.02547192620113492, "time": 0.8074216842651367, "epoch": 3, "memory": 25066, "step": 13416} +{"lr": 0.0005436600133406095, "data_time": 0.151987886428833, "loss": 0.03506448734551668, "time": 0.8548566579818726, "epoch": 3, "memory": 25066, "step": 13516} +{"lr": 0.0005436600133406095, "data_time": 0.11074981689453126, "loss": 0.032025845162570475, "time": 0.8144988536834716, "epoch": 3, "memory": 25066, "step": 13616} +{"lr": 0.0005436600133406095, "data_time": 0.12121062278747559, "loss": 0.019510540971532465, "time": 0.8259487390518189, "epoch": 3, "memory": 25066, "step": 13716} +{"lr": 0.0005436600133406095, "data_time": 0.13341965675354003, "loss": 0.03174030315130949, "time": 0.8362045288085938, "epoch": 3, "memory": 25066, "step": 13816} +{"lr": 0.0005436600133406095, "data_time": 0.1493135929107666, "loss": 0.11955727338790893, "time": 0.8607951641082764, "epoch": 3, "memory": 25066, "step": 13916} +{"lr": 0.0005436600133406095, "data_time": 0.11190319061279297, "loss": 0.0288596716709435, "time": 0.8154298543930054, "epoch": 3, "memory": 25066, "step": 14016} +{"lr": 0.0005436600133406095, "data_time": 0.09707696437835693, "loss": 0.02947291750460863, "time": 0.8033147335052491, "epoch": 3, "memory": 25066, "step": 14116} +{"lr": 0.0005436600133406095, "data_time": 0.11541984081268311, "loss": 0.02915971577167511, "time": 0.817680811882019, "epoch": 3, "memory": 25066, "step": 14216} +{"lr": 0.0005436600133406095, "data_time": 0.21454989910125732, "loss": 0.023450619913637637, "time": 0.9236043930053711, "epoch": 3, "memory": 25066, "step": 14316} +{"lr": 0.0005436600133406095, "data_time": 0.07299714088439942, "loss": 0.022828979371115565, "time": 0.7761953353881836, "epoch": 3, "memory": 25066, "step": 14416} +{"lr": 0.0005436600133406095, "data_time": 0.068560791015625, "loss": 0.026883899234235287, "time": 0.790733814239502, "epoch": 3, "memory": 25066, "step": 14516} +{"lr": 0.0005436600133406095, "data_time": 0.11776745319366455, "loss": 0.031237490102648736, "time": 0.8204161405563355, "epoch": 3, "memory": 25066, "step": 14616} +{"lr": 0.0005436600133406095, "data_time": 0.07734129428863526, "loss": 0.022913479153066872, "time": 0.7978680372238159, "epoch": 3, "memory": 25066, "step": 14716} +{"lr": 0.0005436600133406095, "data_time": 0.06951143741607665, "loss": 0.026734100049361588, "time": 0.7740558624267578, "epoch": 3, "memory": 25066, "step": 14816} +{"lr": 0.0005436600133406095, "data_time": 0.14341337680816652, "loss": 0.022646971326321364, "time": 0.8562366008758545, "epoch": 3, "memory": 25066, "step": 14916} +{"lr": 0.0005436600133406095, "data_time": 0.07941281795501709, "loss": 0.028175718523561953, "time": 0.7817277431488037, "epoch": 3, "memory": 25066, "step": 15016} +{"lr": 0.0005436600133406095, "data_time": 0.0610595703125, "loss": 0.02311288514174521, "time": 0.7718296051025391, "epoch": 3, "memory": 25066, "step": 15116} +{"lr": 0.0005436600133406095, "data_time": 0.08676931858062745, "loss": 0.029936926439404488, "time": 0.7898554563522339, "epoch": 3, "memory": 25066, "step": 15216} +{"lr": 0.0005436600133406095, "data_time": 0.15036725997924805, "loss": 0.02967368345707655, "time": 0.8594315528869629, "epoch": 3, "memory": 25066, "step": 15316} +{"lr": 0.0005436600133406095, "data_time": 0.08284909725189209, "loss": 0.026235038973391057, "time": 0.7856340169906616, "epoch": 3, "memory": 25066, "step": 15416} +{"lr": 0.0005436600133406095, "data_time": 0.08458776473999023, "loss": 0.02139403447508812, "time": 0.8090615510940552, "epoch": 3, "memory": 25066, "step": 15516} +{"lr": 0.0005436600133406095, "data_time": 0.13745789527893065, "loss": 0.023093489464372396, "time": 0.840945029258728, "epoch": 3, "memory": 25066, "step": 15616} +{"lr": 0.0005436600133406095, "data_time": 0.12347440719604492, "loss": 0.02739778677932918, "time": 0.829805850982666, "epoch": 3, "memory": 25066, "step": 15716} +{"lr": 0.0005436600133406095, "data_time": 0.08062148094177246, "loss": 0.024879414681345226, "time": 0.7840144872665405, "epoch": 3, "memory": 25066, "step": 15816} +{"lr": 0.0005436600133406095, "data_time": 0.08028206825256348, "loss": 0.018084383476525544, "time": 0.7857820272445679, "epoch": 3, "memory": 25066, "step": 15916} +{"lr": 0.0005436600133406095, "data_time": 0.10027310848236085, "loss": 0.023039410542696714, "time": 0.8037139654159546, "epoch": 3, "memory": 25066, "step": 16016} +{"lr": 0.0005436600133406095, "data_time": 0.07260229587554931, "loss": 0.035664135776460174, "time": 0.7827414751052857, "epoch": 3, "memory": 25066, "step": 16116} +{"lr": 0.0005436600133406095, "data_time": 0.10394108295440674, "loss": 0.025487054558470845, "time": 0.8065945148468018, "epoch": 3, "memory": 25066, "step": 16216} +{"lr": 0.0005436600133406095, "data_time": 0.10904536247253419, "loss": 0.02932376405224204, "time": 0.8145534992218018, "epoch": 3, "memory": 25066, "step": 16316} +{"lr": 0.0005436600133406095, "data_time": 0.1746471643447876, "loss": 0.025024187564849854, "time": 0.8810407161712647, "epoch": 3, "memory": 25066, "step": 16416} +{"lr": 0.0005436600133406095, "data_time": 0.1664111852645874, "loss": 0.023270398005843164, "time": 0.8818182468414306, "epoch": 3, "memory": 25066, "step": 16516} +{"lr": 0.0005436600133406095, "data_time": 0.1623560905456543, "loss": 0.023703093780204652, "time": 0.8650493383407593, "epoch": 3, "memory": 25066, "step": 16616} +{"lr": 0.0005436600133406095, "data_time": 0.15368180274963378, "loss": 0.024507660139352083, "time": 0.8561034202575684, "epoch": 3, "memory": 25066, "step": 16716} +{"lr": 0.0005436600133406095, "data_time": 0.12068026065826416, "loss": 0.02562913470901549, "time": 0.8227468490600586, "epoch": 3, "memory": 25066, "step": 16816} +{"lr": 0.0005436600133406095, "data_time": 0.08420326709747314, "loss": 0.015059865079820156, "time": 0.7977602243423462, "epoch": 3, "memory": 25066, "step": 16916} +{"lr": 0.0005436600133406095, "data_time": 0.0689265251159668, "loss": 0.02647847877815366, "time": 0.7741671800613403, "epoch": 3, "memory": 25066, "step": 17016} +{"lr": 0.0005436600133406095, "data_time": 0.08307995796203613, "loss": 0.01928577646613121, "time": 0.7887391567230224, "epoch": 3, "memory": 25066, "step": 17116} +{"lr": 0.0005436600133406095, "data_time": 0.03762664794921875, "loss": 0.017405051505193114, "time": 0.7398056745529175, "epoch": 3, "memory": 25066, "step": 17216} +{"accuracy/top1": 99.26140594482422, "data_time": 0.20285582542419434, "time": 0.586881721720976, "step": 3} +{"lr": 0.0004783966494262796, "data_time": 0.08802928924560546, "loss": 0.018302278080955147, "time": 0.7976623773574829, "epoch": 4, "memory": 25066, "step": 17374} +{"lr": 0.0004783966494262796, "data_time": 0.06969025135040283, "loss": 0.01742699109017849, "time": 0.774363660812378, "epoch": 4, "memory": 25066, "step": 17474} +{"lr": 0.0004783966494262796, "data_time": 0.10924909114837647, "loss": 0.018951719999313353, "time": 0.8121732234954834, "epoch": 4, "memory": 25066, "step": 17574} +{"lr": 0.0004783966494262796, "data_time": 0.093589186668396, "loss": 0.0143913384526968, "time": 0.7974068403244019, "epoch": 4, "memory": 25066, "step": 17674} +{"lr": 0.0004783966494262796, "data_time": 0.09095463752746583, "loss": 0.011356176808476448, "time": 0.8005338430404663, "epoch": 4, "memory": 25066, "step": 17774} +{"lr": 0.0004783966494262796, "data_time": 0.11034657955169677, "loss": 0.018610521452501416, "time": 0.813131594657898, "epoch": 4, "memory": 25066, "step": 17874} +{"lr": 0.0004783966494262796, "data_time": 0.10225954055786132, "loss": 0.01797268958762288, "time": 0.8075380086898803, "epoch": 4, "memory": 25066, "step": 17974} +{"lr": 0.0004783966494262796, "data_time": 0.14429073333740233, "loss": 0.01446473179385066, "time": 0.8488998413085938, "epoch": 4, "memory": 25066, "step": 18074} +{"lr": 0.0004783966494262796, "data_time": 0.03730018138885498, "loss": 0.017122942581772806, "time": 0.7503591537475586, "epoch": 4, "memory": 25066, "step": 18174} +{"lr": 0.0004783966494262796, "data_time": 0.10412406921386719, "loss": 0.010631949035450816, "time": 0.8075726270675659, "epoch": 4, "memory": 25066, "step": 18274} +{"lr": 0.0004783966494262796, "data_time": 0.042281508445739746, "loss": 0.014166658092290164, "time": 0.7447845935821533, "epoch": 4, "memory": 25066, "step": 18374} +{"lr": 0.0004783966494262796, "data_time": 0.16450142860412598, "loss": 0.021440746309235693, "time": 0.8690449476242066, "epoch": 4, "memory": 25066, "step": 18474} +{"lr": 0.0004783966494262796, "data_time": 0.08441944122314453, "loss": 0.04489952577278018, "time": 0.7867696762084961, "epoch": 4, "memory": 25066, "step": 18574} +{"lr": 0.0004783966494262796, "data_time": 0.03972434997558594, "loss": 0.013026199210435152, "time": 0.7427253246307373, "epoch": 4, "memory": 25066, "step": 18674} +{"lr": 0.0004783966494262796, "data_time": 0.08989651203155517, "loss": 0.017057787952944638, "time": 0.7940354108810425, "epoch": 4, "memory": 25066, "step": 18774} +{"lr": 0.0004783966494262796, "data_time": 0.07190797328948975, "loss": 0.015776285761967303, "time": 0.7752673149108886, "epoch": 4, "memory": 25066, "step": 18874} +{"lr": 0.0004783966494262796, "data_time": 0.10939245223999024, "loss": 0.013684185827150941, "time": 0.812546181678772, "epoch": 4, "memory": 25066, "step": 18974} +{"lr": 0.0004783966494262796, "data_time": 0.12172412872314453, "loss": 0.01371868490241468, "time": 0.8262057065963745, "epoch": 4, "memory": 25066, "step": 19074} +{"lr": 0.0004783966494262796, "data_time": 0.1257690668106079, "loss": 0.010920207994058729, "time": 0.8281398296356202, "epoch": 4, "memory": 25066, "step": 19174} +{"lr": 0.0004783966494262796, "data_time": 0.040434837341308594, "loss": 0.012805491732433439, "time": 0.7801449298858643, "epoch": 4, "memory": 25066, "step": 19274} +{"lr": 0.0004783966494262796, "data_time": 0.20392210483551027, "loss": 0.01414217851124704, "time": 0.9109137535095215, "epoch": 4, "memory": 25066, "step": 19374} +{"lr": 0.0004783966494262796, "data_time": 0.1468379020690918, "loss": 0.03519926955923438, "time": 0.8494501352310181, "epoch": 4, "memory": 25066, "step": 19474} +{"lr": 0.0004783966494262796, "data_time": 0.18970537185668945, "loss": 0.01362181599251926, "time": 0.9044387102127075, "epoch": 4, "memory": 25066, "step": 19574} +{"lr": 0.0004783966494262796, "data_time": 0.13027243614196776, "loss": 0.011255398346111178, "time": 0.8407919645309448, "epoch": 4, "memory": 25066, "step": 19674} +{"lr": 0.0004783966494262796, "data_time": 0.15535576343536378, "loss": 0.02364458399824798, "time": 0.8576542139053345, "epoch": 4, "memory": 25066, "step": 19774} +{"lr": 0.0004783966494262796, "data_time": 0.03958296775817871, "loss": 0.014590098150074482, "time": 0.7521777391433716, "epoch": 4, "memory": 25066, "step": 19874} +{"lr": 0.0004783966494262796, "data_time": 0.06697421073913574, "loss": 0.015909790294244885, "time": 0.7691185235977173, "epoch": 4, "memory": 25066, "step": 19974} +{"lr": 0.0004783966494262796, "data_time": 0.06014094352722168, "loss": 0.02140692276880145, "time": 0.7619815111160279, "epoch": 4, "memory": 25066, "step": 20074} +{"lr": 0.0004783966494262796, "data_time": 0.11686837673187256, "loss": 0.01703349258750677, "time": 0.819719386100769, "epoch": 4, "memory": 25066, "step": 20174} +{"lr": 0.0004783966494262796, "data_time": 0.11401152610778809, "loss": 0.012818874884396791, "time": 0.8164216995239257, "epoch": 4, "memory": 25066, "step": 20274} +{"lr": 0.0004783966494262796, "data_time": 0.05924406051635742, "loss": 0.013755055237561464, "time": 0.7617597103118896, "epoch": 4, "memory": 25066, "step": 20374} +{"lr": 0.0004783966494262796, "data_time": 0.11467101573944091, "loss": 0.014929664228111506, "time": 0.8307372570037842, "epoch": 4, "memory": 25066, "step": 20474} +{"lr": 0.0004783966494262796, "data_time": 0.05902366638183594, "loss": 0.01392804174683988, "time": 0.7617453336715698, "epoch": 4, "memory": 25066, "step": 20574} +{"lr": 0.0004783966494262796, "data_time": 0.05788445472717285, "loss": 0.010660858545452356, "time": 0.7684449672698974, "epoch": 4, "memory": 25066, "step": 20674} +{"lr": 0.0004783966494262796, "data_time": 0.13744583129882812, "loss": 0.01604933012276888, "time": 0.8561843156814575, "epoch": 4, "memory": 25066, "step": 20774} +{"lr": 0.0004783966494262796, "data_time": 0.044433426856994626, "loss": 0.011551185231655835, "time": 0.7534519672393799, "epoch": 4, "memory": 25066, "step": 20874} +{"lr": 0.0004783966494262796, "data_time": 0.08031361103057862, "loss": 0.02041889582760632, "time": 0.7964224576950073, "epoch": 4, "memory": 25066, "step": 20974} +{"lr": 0.0004783966494262796, "data_time": 0.058609390258789064, "loss": 0.010382734099403023, "time": 0.7653856992721557, "epoch": 4, "memory": 25066, "step": 21074} +{"lr": 0.0004783966494262796, "data_time": 0.06863110065460205, "loss": 0.008980284724384547, "time": 0.772297477722168, "epoch": 4, "memory": 25066, "step": 21174} +{"lr": 0.0004783966494262796, "data_time": 0.04592280387878418, "loss": 0.01715405434370041, "time": 0.7556788921356201, "epoch": 4, "memory": 25066, "step": 21274} +{"lr": 0.0004783966494262796, "data_time": 0.0480712890625, "loss": 0.015487766405567528, "time": 0.7565844535827637, "epoch": 4, "memory": 25066, "step": 21374} +{"lr": 0.0004783966494262796, "data_time": 0.013774728775024414, "loss": 0.01465374231338501, "time": 0.7266249418258667, "epoch": 4, "memory": 25066, "step": 21474} +{"lr": 0.0004783966494262796, "data_time": 0.05666790008544922, "loss": 0.010056767472997307, "time": 0.7723235130310059, "epoch": 4, "memory": 25066, "step": 21574} +{"lr": 0.0004783966494262796, "data_time": 0.03183979988098144, "loss": 0.015812888741493225, "time": 0.7344065427780151, "epoch": 4, "memory": 25066, "step": 21674} +{"lr": 0.0004783966494262796, "data_time": 0.10094289779663086, "loss": 0.0076159191317856315, "time": 0.8159659147262573, "epoch": 4, "memory": 25066, "step": 21774} +{"lr": 0.0004783966494262796, "data_time": 0.0774773359298706, "loss": 0.025784523971378804, "time": 0.7799385547637939, "epoch": 4, "memory": 25066, "step": 21874} +{"lr": 0.0004783966494262796, "data_time": 0.07479071617126465, "loss": 0.009562530554831028, "time": 0.7869988203048706, "epoch": 4, "memory": 25066, "step": 21974} +{"lr": 0.0004783966494262796, "data_time": 0.10151920318603516, "loss": 0.01564459167420864, "time": 0.8041109800338745, "epoch": 4, "memory": 25066, "step": 22074} +{"lr": 0.0004783966494262796, "data_time": 0.0779494047164917, "loss": 0.011913094436749816, "time": 0.7937377691268921, "epoch": 4, "memory": 25066, "step": 22174} +{"lr": 0.0004783966494262796, "data_time": 0.07155652046203613, "loss": 0.013006499083712698, "time": 0.773443603515625, "epoch": 4, "memory": 25066, "step": 22274} +{"lr": 0.0004783966494262796, "data_time": 0.05156948566436768, "loss": 0.01421361225657165, "time": 0.7727417945861816, "epoch": 4, "memory": 25066, "step": 22374} +{"lr": 0.0004783966494262796, "data_time": 0.06946351528167724, "loss": 0.01124355997890234, "time": 0.7742018461227417, "epoch": 4, "memory": 25066, "step": 22474} +{"lr": 0.0004783966494262796, "data_time": 0.1365509510040283, "loss": 0.007517954707145691, "time": 0.8512271642684937, "epoch": 4, "memory": 25066, "step": 22574} +{"lr": 0.0004783966494262796, "data_time": 0.11934657096862793, "loss": 0.01400508708320558, "time": 0.8221124887466431, "epoch": 4, "memory": 25066, "step": 22674} +{"lr": 0.0004783966494262796, "data_time": 0.08114609718322754, "loss": 0.009509793762117625, "time": 0.7843251943588256, "epoch": 4, "memory": 25066, "step": 22774} +{"lr": 0.0004783966494262796, "data_time": 0.06932976245880126, "loss": 0.007604291290044784, "time": 0.772094988822937, "epoch": 4, "memory": 25066, "step": 22874} +{"lr": 0.0004783966494262796, "data_time": 0.004683208465576172, "loss": 0.011296708742156625, "time": 0.7076105117797852, "epoch": 4, "memory": 25066, "step": 22974} +{"accuracy/top1": 99.44447326660156, "data_time": 0.2145690497230081, "time": 0.6036080612855799, "step": 4} +{"lr": 0.0003961600133406095, "data_time": 0.022608208656311034, "loss": 0.004102718085050583, "time": 0.7602856159210205, "epoch": 5, "memory": 25066, "step": 23132} +{"lr": 0.0003961600133406095, "data_time": 0.14658496379852295, "loss": 0.004557293513789773, "time": 0.8631012916564942, "epoch": 5, "memory": 25066, "step": 23232} +{"lr": 0.0003961600133406095, "data_time": 0.09179384708404541, "loss": 0.01601978885009885, "time": 0.8112494468688964, "epoch": 5, "memory": 25066, "step": 23332} +{"lr": 0.0003961600133406095, "data_time": 0.05205504894256592, "loss": 0.009309962298721075, "time": 0.7570523977279663, "epoch": 5, "memory": 25066, "step": 23432} +{"lr": 0.0003961600133406095, "data_time": 0.06578717231750489, "loss": 0.005065946886315942, "time": 0.7910119771957398, "epoch": 5, "memory": 25066, "step": 23532} +{"lr": 0.0003961600133406095, "data_time": 0.07953751087188721, "loss": 0.006550627620890736, "time": 0.7847600936889648, "epoch": 5, "memory": 25066, "step": 23632} +{"lr": 0.0003961600133406095, "data_time": 0.040402793884277345, "loss": 0.01310245501808822, "time": 0.747068190574646, "epoch": 5, "memory": 25066, "step": 23732} +{"lr": 0.0003961600133406095, "data_time": 0.08198513984680175, "loss": 0.013690729532390833, "time": 0.7846913814544678, "epoch": 5, "memory": 25066, "step": 23832} +{"lr": 0.0003961600133406095, "data_time": 0.16190268993377685, "loss": 0.006088630622252822, "time": 0.8648688077926636, "epoch": 5, "memory": 25066, "step": 23932} +{"lr": 0.0003961600133406095, "data_time": 0.0015821456909179688, "loss": 0.013069759169593454, "time": 0.7684478044509888, "epoch": 5, "memory": 25066, "step": 24032} +{"lr": 0.0003961600133406095, "data_time": 0.0012533187866210938, "loss": 0.006365784769877791, "time": 0.8579529523849487, "epoch": 5, "memory": 25066, "step": 24132} +{"lr": 0.0003961600133406095, "data_time": 0.001192927360534668, "loss": 0.015119435312226415, "time": 0.8786704063415527, "epoch": 5, "memory": 25066, "step": 24232} +{"lr": 0.0003961600133406095, "data_time": 0.0012086153030395508, "loss": 0.006258977996185422, "time": 1.0646855592727662, "epoch": 5, "memory": 25066, "step": 24332} +{"lr": 0.0003961600133406095, "data_time": 0.15270371437072755, "loss": 0.0053798640612512825, "time": 1.0390119552612305, "epoch": 5, "memory": 25066, "step": 24432} +{"lr": 0.0003961600133406095, "data_time": 0.20881047248840331, "loss": 0.00460607442073524, "time": 1.2325568437576293, "epoch": 5, "memory": 25066, "step": 24532} +{"lr": 0.0003961600133406095, "data_time": 0.0013935327529907226, "loss": 0.0112472013104707, "time": 0.7947478771209717, "epoch": 5, "memory": 25066, "step": 24632} +{"lr": 0.0003961600133406095, "data_time": 0.025103092193603516, "loss": 0.010350334830582142, "time": 0.864455270767212, "epoch": 5, "memory": 25066, "step": 24732} +{"lr": 0.0003961600133406095, "data_time": 0.0013593196868896484, "loss": 0.007137893233448267, "time": 0.8959303140640259, "epoch": 5, "memory": 25066, "step": 24832} +{"lr": 0.0003961600133406095, "data_time": 0.0014004945755004884, "loss": 0.011522663012146949, "time": 1.0799331188201904, "epoch": 5, "memory": 25066, "step": 24932} +{"lr": 0.0003961600133406095, "data_time": 0.0013248205184936523, "loss": 0.012295640585944057, "time": 0.9674221754074097, "epoch": 5, "memory": 25066, "step": 25032} +{"lr": 0.0003961600133406095, "data_time": 0.001428365707397461, "loss": 0.006364683760330081, "time": 1.0895797729492187, "epoch": 5, "memory": 25066, "step": 25132} +{"lr": 0.0003961600133406095, "data_time": 0.001578354835510254, "loss": 0.0019763521384447815, "time": 1.088343048095703, "epoch": 5, "memory": 25066, "step": 25232} +{"lr": 0.0003961600133406095, "data_time": 0.0014547109603881836, "loss": 0.007758986670523882, "time": 1.1326516151428223, "epoch": 5, "memory": 25066, "step": 25332} +{"lr": 0.0003961600133406095, "data_time": 0.2643946886062622, "loss": 0.008848836505785584, "time": 0.9786058902740479, "epoch": 5, "memory": 25066, "step": 25432} +{"lr": 0.0003961600133406095, "data_time": 0.37523105144500735, "loss": 0.007231908617541194, "time": 1.0892903089523316, "epoch": 5, "memory": 25066, "step": 25532} +{"lr": 0.0003961600133406095, "data_time": 0.0012932300567626953, "loss": 0.012182714557275176, "time": 1.02841956615448, "epoch": 5, "memory": 25066, "step": 25632} +{"lr": 0.0003961600133406095, "data_time": 0.001392364501953125, "loss": 0.01186219952069223, "time": 0.8187432527542114, "epoch": 5, "memory": 25066, "step": 25732} +{"lr": 0.0003961600133406095, "data_time": 0.0015081644058227539, "loss": 0.009799295058473944, "time": 0.7864231586456298, "epoch": 5, "memory": 25066, "step": 25832} +{"lr": 0.0003961600133406095, "data_time": 0.00128173828125, "loss": 0.004450850514695048, "time": 0.7898314237594605, "epoch": 5, "memory": 25066, "step": 25932} +{"lr": 0.0003961600133406095, "data_time": 0.0014527797698974609, "loss": 0.010132176661863923, "time": 0.8628645181655884, "epoch": 5, "memory": 25066, "step": 26032} +{"lr": 0.0003961600133406095, "data_time": 0.0013979673385620117, "loss": 0.012101198825985194, "time": 0.8257627725601197, "epoch": 5, "memory": 25066, "step": 26132} +{"lr": 0.0003961600133406095, "data_time": 0.0016843557357788086, "loss": 0.007321288902312517, "time": 0.8655916452407837, "epoch": 5, "memory": 25066, "step": 26232} +{"lr": 0.0003961600133406095, "data_time": 0.001238250732421875, "loss": 0.010624748328700662, "time": 0.796226692199707, "epoch": 5, "memory": 25066, "step": 26332} +{"lr": 0.0003961600133406095, "data_time": 0.0017630815505981444, "loss": 0.010346449492499233, "time": 0.8071348667144775, "epoch": 5, "memory": 25066, "step": 26432} +{"lr": 0.0003961600133406095, "data_time": 0.0013033390045166016, "loss": 0.01157821766100824, "time": 0.8289038181304932, "epoch": 5, "memory": 25066, "step": 26532} +{"lr": 0.0003961600133406095, "data_time": 0.001332259178161621, "loss": 0.0075154932215809826, "time": 0.784517765045166, "epoch": 5, "memory": 25066, "step": 26632} +{"lr": 0.0003961600133406095, "data_time": 0.0014123201370239257, "loss": 0.007396357506513596, "time": 0.8106993675231934, "epoch": 5, "memory": 25066, "step": 26732} +{"lr": 0.0003961600133406095, "data_time": 0.0013507604598999023, "loss": 0.009038366330787539, "time": 0.817111873626709, "epoch": 5, "memory": 25066, "step": 26832} +{"lr": 0.0003961600133406095, "data_time": 0.001336216926574707, "loss": 0.0063036609906703235, "time": 0.7969225168228149, "epoch": 5, "memory": 25066, "step": 26932} +{"lr": 0.0003961600133406095, "data_time": 0.0014711141586303711, "loss": 0.006457010982558131, "time": 0.7924376964569092, "epoch": 5, "memory": 25066, "step": 27032} +{"lr": 0.0003961600133406095, "data_time": 0.0013987302780151367, "loss": 0.010920880734920502, "time": 0.8514780521392822, "epoch": 5, "memory": 25066, "step": 27132} +{"lr": 0.0003961600133406095, "data_time": 0.001439666748046875, "loss": 0.004842010932043195, "time": 0.8348291158676148, "epoch": 5, "memory": 25066, "step": 27232} +{"lr": 0.0003961600133406095, "data_time": 0.0014387130737304687, "loss": 0.011458777915686369, "time": 0.8007415056228637, "epoch": 5, "memory": 25066, "step": 27332} +{"lr": 0.0003961600133406095, "data_time": 0.001244974136352539, "loss": 0.004842579970136285, "time": 0.8030724287033081, "epoch": 5, "memory": 25066, "step": 27432} +{"lr": 0.0003961600133406095, "data_time": 0.0013063669204711914, "loss": 0.0015838440507650376, "time": 0.8247030973434448, "epoch": 5, "memory": 25066, "step": 27532} +{"lr": 0.0003961600133406095, "data_time": 0.002011609077453613, "loss": 0.007811467070132494, "time": 0.8476327180862426, "epoch": 5, "memory": 25066, "step": 27632} +{"lr": 0.0003961600133406095, "data_time": 0.0012860536575317384, "loss": 0.007589854020625353, "time": 0.8103592395782471, "epoch": 5, "memory": 25066, "step": 27732} +{"lr": 0.0003961600133406095, "data_time": 0.0011636018753051758, "loss": 0.007840817188844084, "time": 0.7907381057739258, "epoch": 5, "memory": 25066, "step": 27832} +{"lr": 0.0003961600133406095, "data_time": 0.0013646125793457032, "loss": 0.007127438858151436, "time": 0.816729211807251, "epoch": 5, "memory": 25066, "step": 27932} +{"lr": 0.0003961600133406095, "data_time": 0.0014326095581054688, "loss": 0.00728184673935175, "time": 0.8198399782180786, "epoch": 5, "memory": 25066, "step": 28032} +{"lr": 0.0003961600133406095, "data_time": 0.0013283252716064452, "loss": 0.004735107207670808, "time": 0.8231123447418213, "epoch": 5, "memory": 25066, "step": 28132} +{"lr": 0.0003961600133406095, "data_time": 0.0012480497360229492, "loss": 0.007624412514269352, "time": 0.8076583862304687, "epoch": 5, "memory": 25066, "step": 28232} +{"lr": 0.0003961600133406095, "data_time": 0.001441216468811035, "loss": 0.010951170464977622, "time": 0.8081752538681031, "epoch": 5, "memory": 25066, "step": 28332} +{"lr": 0.0003961600133406095, "data_time": 0.0013387203216552734, "loss": 0.009171189460903405, "time": 0.8238954067230224, "epoch": 5, "memory": 25066, "step": 28432} +{"lr": 0.0003961600133406095, "data_time": 0.0013531208038330077, "loss": 0.011602256866171955, "time": 0.8434428453445435, "epoch": 5, "memory": 25066, "step": 28532} +{"lr": 0.0003961600133406095, "data_time": 0.0013370275497436523, "loss": 0.005083956802263856, "time": 0.8207385540008545, "epoch": 5, "memory": 25066, "step": 28632} +{"lr": 0.0003961600133406095, "data_time": 0.001479792594909668, "loss": 0.009026247914880514, "time": 0.8370945453643799, "epoch": 5, "memory": 25066, "step": 28732} +{"accuracy/top1": 99.70330047607422, "data_time": 0.20719463684979608, "time": 0.5988076574662152, "step": 5} +{"lr": 0.000305, "data_time": 0.0014534473419189453, "loss": 0.003872048668563366, "time": 0.8239111423492431, "epoch": 6, "memory": 25066, "step": 28890} +{"lr": 0.000305, "data_time": 0.0012253999710083007, "loss": 0.0026534893549978733, "time": 0.7924069166183472, "epoch": 6, "memory": 25066, "step": 28990} +{"lr": 0.000305, "data_time": 0.0014041900634765626, "loss": 0.005696064792573452, "time": 0.7971612453460694, "epoch": 6, "memory": 25066, "step": 29090} +{"lr": 0.000305, "data_time": 0.001421213150024414, "loss": 0.019828826701268554, "time": 0.7973195314407349, "epoch": 6, "memory": 25066, "step": 29190} +{"lr": 0.000305, "data_time": 0.0012470245361328124, "loss": 0.006403261562809348, "time": 0.7931866884231568, "epoch": 6, "memory": 25066, "step": 29290} +{"lr": 0.000305, "data_time": 0.0012543201446533203, "loss": 0.004346122406423092, "time": 0.8728467941284179, "epoch": 6, "memory": 25066, "step": 29390} +{"lr": 0.000305, "data_time": 0.0012321949005126953, "loss": 0.005958028463646769, "time": 0.882062578201294, "epoch": 6, "memory": 25066, "step": 29490} +{"lr": 0.000305, "data_time": 0.0013159990310668945, "loss": 0.0047013307455927135, "time": 0.7847998380661011, "epoch": 6, "memory": 25066, "step": 29590} +{"lr": 0.000305, "data_time": 0.0016243457794189453, "loss": 0.005479770759120583, "time": 0.8502360820770264, "epoch": 6, "memory": 25066, "step": 29690} +{"lr": 0.000305, "data_time": 0.0012400150299072266, "loss": 0.00786884999834001, "time": 0.8248006343841553, "epoch": 6, "memory": 25066, "step": 29790} +{"lr": 0.000305, "data_time": 0.01263282299041748, "loss": 0.0031838535331189634, "time": 0.8590489387512207, "epoch": 6, "memory": 25066, "step": 29890} +{"lr": 0.000305, "data_time": 0.0013402938842773438, "loss": 0.009283325169235468, "time": 0.861748218536377, "epoch": 6, "memory": 25066, "step": 29990} +{"lr": 0.000305, "data_time": 0.0013037443161010743, "loss": 0.008197297761216759, "time": 0.8198406934738159, "epoch": 6, "memory": 25066, "step": 30090} +{"lr": 0.000305, "data_time": 0.0014167308807373046, "loss": 0.0068174193147569895, "time": 0.932839322090149, "epoch": 6, "memory": 25066, "step": 30190} +{"lr": 0.000305, "data_time": 0.0013034820556640625, "loss": 0.0017456692177802325, "time": 0.8413915157318115, "epoch": 6, "memory": 25066, "step": 30290} diff --git a/convnext_small_4xb256_fake5m_lr6e-4_10epoch/20230601_151930/vis_data/config.py b/convnext_small_4xb256_fake5m_lr6e-4_10epoch/20230601_151930/vis_data/config.py new file mode 100644 index 0000000000000000000000000000000000000000..dce5e76162f46d520159aeda7be0d77c281cf731 --- /dev/null +++ b/convnext_small_4xb256_fake5m_lr6e-4_10epoch/20230601_151930/vis_data/config.py @@ -0,0 +1,169 @@ +optim_wrapper = dict( + optimizer=dict( + type='AdamW', + lr=0.0006, + weight_decay=0.05, + eps=1e-08, + betas=(0.9, 0.999), + _scope_='mmpretrain'), + paramwise_cfg=dict( + norm_decay_mult=0.0, + bias_decay_mult=0.0, + flat_decay_mult=0.0, + custom_keys=dict({ + '.absolute_pos_embed': dict(decay_mult=0.0), + '.relative_position_bias_table': dict(decay_mult=0.0) + })), + type='AmpOptimWrapper', + dtype='bfloat16', + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=True, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=4096) +model = dict( + type='ImageClassifier', + backbone=dict(type='ConvNeXt', arch='small', drop_path_rate=0.4), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=768, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + init_cfg=None), + init_cfg=dict( + type='TruncNormal', layer=['Conv2d', 'Linear'], std=0.02, bias=0.0), + train_cfg=None) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stablediffusionV1-5R2-dpmsolver-25-5m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/convnext_small_4xb256_fake5m_lr6e-4_10epoch' diff --git a/convnext_small_4xb256_fake5m_lr6e-4_10epoch/20230601_151930/vis_data/events.out.tfevents.1685603978.SH-IDC1-10-140-24-134.143946.0 b/convnext_small_4xb256_fake5m_lr6e-4_10epoch/20230601_151930/vis_data/events.out.tfevents.1685603978.SH-IDC1-10-140-24-134.143946.0 new file mode 100644 index 0000000000000000000000000000000000000000..3369476bbed7f01922e7fde036cf48425f278c61 --- /dev/null +++ b/convnext_small_4xb256_fake5m_lr6e-4_10epoch/20230601_151930/vis_data/events.out.tfevents.1685603978.SH-IDC1-10-140-24-134.143946.0 @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ce4d07130f67a4cf1c728859e790ba0e487d4e6cda3130020dee4b35ee0a2f87 +size 440706 diff --git a/convnext_small_4xb256_fake5m_lr6e-4_10epoch/20230601_151930/vis_data/scalars.json b/convnext_small_4xb256_fake5m_lr6e-4_10epoch/20230601_151930/vis_data/scalars.json new file mode 100644 index 0000000000000000000000000000000000000000..82523aacdc675345c3e3955bc1b05e5a05b8a5c6 --- /dev/null +++ b/convnext_small_4xb256_fake5m_lr6e-4_10epoch/20230601_151930/vis_data/scalars.json @@ -0,0 +1,305 @@ +{"lr": 0.0006, "data_time": 0.0015582561492919922, "loss": 0.6135746002197265, "time": 0.830146074295044, "epoch": 1, "memory": 25062, "step": 100} +{"lr": 0.0006, "data_time": 0.13062400817871095, "loss": 0.5753461420536041, "time": 0.846795630455017, "epoch": 1, "memory": 25062, "step": 200} +{"lr": 0.0006, "data_time": 0.12613627910614014, "loss": 0.5315729141235351, "time": 0.8305403709411621, "epoch": 1, "memory": 25062, "step": 300} +{"lr": 0.0006, "data_time": 0.13247416019439698, "loss": 0.5085993051528931, "time": 0.8568308115005493, "epoch": 1, "memory": 25062, "step": 400} +{"lr": 0.0006, "data_time": 0.16386404037475585, "loss": 0.500954395532608, "time": 0.8659586191177369, "epoch": 1, "memory": 25062, "step": 500} +{"lr": 0.0006, "data_time": 0.14253461360931396, "loss": 0.4692378222942352, "time": 0.8801122665405273, "epoch": 1, "memory": 25062, "step": 600} +{"lr": 0.0006, "data_time": 0.20899367332458496, "loss": 0.4822313070297241, "time": 0.913886308670044, "epoch": 1, "memory": 25062, "step": 700} +{"lr": 0.0006, "data_time": 0.1478217601776123, "loss": 0.43232129216194154, "time": 0.8784266710281372, "epoch": 1, "memory": 25062, "step": 800} +{"lr": 0.0006, "data_time": 0.12980871200561522, "loss": 0.4234696596860886, "time": 0.8333329200744629, "epoch": 1, "memory": 25062, "step": 900} +{"lr": 0.0006, "data_time": 0.10303521156311035, "loss": 0.43477452397346494, "time": 0.8239191770553589, "epoch": 1, "memory": 25062, "step": 1000} +{"lr": 0.0006, "data_time": 0.11415770053863525, "loss": 0.4187207818031311, "time": 0.8161101341247559, "epoch": 1, "memory": 25062, "step": 1100} +{"lr": 0.0006, "data_time": 0.160141658782959, "loss": 0.41021077036857606, "time": 0.8773677349090576, "epoch": 1, "memory": 25062, "step": 1200} +{"lr": 0.0006, "data_time": 0.06650912761688232, "loss": 0.42308527827262876, "time": 0.7686495304107666, "epoch": 1, "memory": 25062, "step": 1300} +{"lr": 0.0006, "data_time": 0.08822498321533204, "loss": 0.38337395489215853, "time": 0.8020293474197387, "epoch": 1, "memory": 25062, "step": 1400} +{"lr": 0.0006, "data_time": 0.14092085361480713, "loss": 0.3752002418041229, "time": 0.8428231716156006, "epoch": 1, "memory": 25062, "step": 1500} +{"lr": 0.0006, "data_time": 0.09997437000274659, "loss": 0.3994987905025482, "time": 0.8140286207199097, "epoch": 1, "memory": 25062, "step": 1600} +{"lr": 0.0006, "data_time": 0.12974863052368163, "loss": 0.3620677664875984, "time": 0.8324440002441407, "epoch": 1, "memory": 25062, "step": 1700} +{"lr": 0.0006, "data_time": 0.08871958255767823, "loss": 0.349988853931427, "time": 0.7991168022155761, "epoch": 1, "memory": 25062, "step": 1800} +{"lr": 0.0006, "data_time": 0.1101177453994751, "loss": 0.32603266537189485, "time": 0.813023567199707, "epoch": 1, "memory": 25062, "step": 1900} +{"lr": 0.0006, "data_time": 0.09395744800567626, "loss": 0.35006090998649597, "time": 0.8047962665557862, "epoch": 1, "memory": 25062, "step": 2000} +{"lr": 0.0006, "data_time": 0.12074847221374511, "loss": 0.344668909907341, "time": 0.8260648012161255, "epoch": 1, "memory": 25062, "step": 2100} +{"lr": 0.0006, "data_time": 0.14276158809661865, "loss": 0.3121231347322464, "time": 0.8568590641021728, "epoch": 1, "memory": 25062, "step": 2200} +{"lr": 0.0006, "data_time": 0.10163240432739258, "loss": 0.2936995327472687, "time": 0.8037424802780151, "epoch": 1, "memory": 25062, "step": 2300} +{"lr": 0.0006, "data_time": 0.06423301696777343, "loss": 0.321243605017662, "time": 0.780709195137024, "epoch": 1, "memory": 25062, "step": 2400} +{"lr": 0.0006, "data_time": 0.11378779411315917, "loss": 0.28465537130832674, "time": 0.8197810888290405, "epoch": 1, "memory": 25062, "step": 2500} +{"lr": 0.0006, "data_time": 0.15081541538238524, "loss": 0.28395094722509384, "time": 0.852768349647522, "epoch": 1, "memory": 25062, "step": 2600} +{"lr": 0.0006, "data_time": 0.16163480281829834, "loss": 0.2992450505495071, "time": 0.8647766828536987, "epoch": 1, "memory": 25062, "step": 2700} +{"lr": 0.0006, "data_time": 0.12410805225372315, "loss": 0.2752879589796066, "time": 0.8262264490127563, "epoch": 1, "memory": 25062, "step": 2800} +{"lr": 0.0006, "data_time": 0.15808987617492676, "loss": 0.2809392124414444, "time": 0.8610882043838501, "epoch": 1, "memory": 25062, "step": 2900} +{"lr": 0.0006, "data_time": 0.10003814697265626, "loss": 0.2706796482205391, "time": 0.8030795574188232, "epoch": 1, "memory": 25062, "step": 3000} +{"lr": 0.0006, "data_time": 0.1345454216003418, "loss": 0.2056187316775322, "time": 0.8375606536865234, "epoch": 1, "memory": 25062, "step": 3100} +{"lr": 0.0006, "data_time": 0.10702877044677735, "loss": 0.2620606556534767, "time": 0.8091813564300537, "epoch": 1, "memory": 25062, "step": 3200} +{"lr": 0.0006, "data_time": 0.10420176982879639, "loss": 0.23590899109840394, "time": 0.8064955234527588, "epoch": 1, "memory": 25062, "step": 3300} +{"lr": 0.0006, "data_time": 0.11196503639221192, "loss": 0.2258203849196434, "time": 0.8173755168914795, "epoch": 1, "memory": 25062, "step": 3400} +{"lr": 0.0006, "data_time": 0.09510846138000488, "loss": 0.2160615622997284, "time": 0.7991591930389405, "epoch": 1, "memory": 25062, "step": 3500} +{"lr": 0.0006, "data_time": 0.11947128772735596, "loss": 0.24335542768239976, "time": 0.822614598274231, "epoch": 1, "memory": 25062, "step": 3600} +{"lr": 0.0006, "data_time": 0.13002800941467285, "loss": 0.22328522503376008, "time": 0.8323217630386353, "epoch": 1, "memory": 25062, "step": 3700} +{"lr": 0.0006, "data_time": 0.10663254261016845, "loss": 0.22647091150283813, "time": 0.8117939949035644, "epoch": 1, "memory": 25062, "step": 3800} +{"lr": 0.0006, "data_time": 0.12309346199035645, "loss": 0.19595374017953873, "time": 0.8279386043548584, "epoch": 1, "memory": 25062, "step": 3900} +{"lr": 0.0006, "data_time": 0.07163856029510499, "loss": 0.20814566165208817, "time": 0.7732212781906128, "epoch": 1, "memory": 25062, "step": 4000} +{"lr": 0.0006, "data_time": 0.09958779811859131, "loss": 0.21983536332845688, "time": 0.8011039018630981, "epoch": 1, "memory": 25062, "step": 4100} +{"lr": 0.0006, "data_time": 0.0894810438156128, "loss": 0.2043963760137558, "time": 0.7928557157516479, "epoch": 1, "memory": 25062, "step": 4200} +{"lr": 0.0006, "data_time": 0.15857112407684326, "loss": 0.17422495484352113, "time": 0.8622824192047119, "epoch": 1, "memory": 25062, "step": 4300} +{"lr": 0.0006, "data_time": 0.13226869106292724, "loss": 0.18446631729602814, "time": 0.8374736547470093, "epoch": 1, "memory": 25062, "step": 4400} +{"lr": 0.0006, "data_time": 0.12099769115447997, "loss": 0.17812515795230865, "time": 0.8227338790893555, "epoch": 1, "memory": 25062, "step": 4500} +{"lr": 0.0006, "data_time": 0.15409488677978517, "loss": 0.31683850288391113, "time": 0.8568028450012207, "epoch": 1, "memory": 25062, "step": 4600} +{"lr": 0.0006, "data_time": 0.1554493188858032, "loss": 0.19284236431121826, "time": 0.8568015813827514, "epoch": 1, "memory": 25062, "step": 4700} +{"lr": 0.0006, "data_time": 0.1810582160949707, "loss": 0.20528918504714966, "time": 0.8851372718811035, "epoch": 1, "memory": 25062, "step": 4800} +{"lr": 0.0006, "data_time": 0.12607507705688475, "loss": 0.15152284875512123, "time": 0.8319072723388672, "epoch": 1, "memory": 25062, "step": 4900} +{"lr": 0.0006, "data_time": 0.15493619441986084, "loss": 0.16518262326717376, "time": 0.8574823617935181, "epoch": 1, "memory": 25062, "step": 5000} +{"lr": 0.0006, "data_time": 0.1015016794204712, "loss": 0.16402522027492522, "time": 0.8082464933395386, "epoch": 1, "memory": 25062, "step": 5100} +{"lr": 0.0006, "data_time": 0.12581818103790282, "loss": 0.17387931197881698, "time": 0.8294331312179566, "epoch": 1, "memory": 25062, "step": 5200} +{"lr": 0.0006, "data_time": 0.15540964603424073, "loss": 0.17468418180942535, "time": 0.8579760789871216, "epoch": 1, "memory": 25062, "step": 5300} +{"lr": 0.0006, "data_time": 0.1133528470993042, "loss": 0.15653457567095758, "time": 0.8156604290008544, "epoch": 1, "memory": 25062, "step": 5400} +{"lr": 0.0006, "data_time": 0.13631539344787597, "loss": 0.16939338743686677, "time": 0.8382399320602417, "epoch": 1, "memory": 25062, "step": 5500} +{"lr": 0.0006, "data_time": 0.12221121788024902, "loss": 0.13643198758363723, "time": 0.8262169361114502, "epoch": 1, "memory": 25062, "step": 5600} +{"lr": 0.0006, "data_time": 0.12007019519805909, "loss": 0.1673684224486351, "time": 0.8233999013900757, "epoch": 1, "memory": 25062, "step": 5700} +{"accuracy/top1": 86.5223159790039, "data_time": 0.332147479057312, "time": 0.8400849550962448, "step": 1} +{"lr": 0.0005855616723070703, "data_time": 0.1329329013824463, "loss": 0.12358986958861351, "time": 0.8460269212722779, "epoch": 2, "memory": 25066, "step": 5858} +{"lr": 0.0005855616723070703, "data_time": 0.08893582820892335, "loss": 0.1282133489847183, "time": 0.7928776741027832, "epoch": 2, "memory": 25066, "step": 5958} +{"lr": 0.0005855616723070703, "data_time": 0.09415514469146728, "loss": 0.1568968787789345, "time": 0.8223574161529541, "epoch": 2, "memory": 25066, "step": 6058} +{"lr": 0.0005855616723070703, "data_time": 0.12533230781555177, "loss": 0.14167584106326103, "time": 0.8277125358581543, "epoch": 2, "memory": 25066, "step": 6158} +{"lr": 0.0005855616723070703, "data_time": 0.1404815435409546, "loss": 0.1335474856197834, "time": 0.8733075857162476, "epoch": 2, "memory": 25066, "step": 6258} +{"lr": 0.0005855616723070703, "data_time": 0.15498428344726561, "loss": 0.1062768466770649, "time": 0.8575185537338257, "epoch": 2, "memory": 25066, "step": 6358} +{"lr": 0.0005855616723070703, "data_time": 0.19075372219085693, "loss": 0.12959541007876396, "time": 0.9077394723892211, "epoch": 2, "memory": 25066, "step": 6458} +{"lr": 0.0005855616723070703, "data_time": 0.08825895786285401, "loss": 0.13451345264911652, "time": 0.7918497323989868, "epoch": 2, "memory": 25066, "step": 6558} +{"lr": 0.0005855616723070703, "data_time": 0.12768311500549318, "loss": 0.12959048748016358, "time": 0.8432443618774415, "epoch": 2, "memory": 25066, "step": 6658} +{"lr": 0.0005855616723070703, "data_time": 0.10369279384613037, "loss": 0.11627456694841384, "time": 0.8065032243728638, "epoch": 2, "memory": 25066, "step": 6758} +{"lr": 0.0005855616723070703, "data_time": 0.18359854221343994, "loss": 0.1069206453859806, "time": 0.8995799779891968, "epoch": 2, "memory": 25066, "step": 6858} +{"lr": 0.0005855616723070703, "data_time": 0.1437976360321045, "loss": 0.1431097500026226, "time": 0.8483048439025879, "epoch": 2, "memory": 25066, "step": 6958} +{"lr": 0.0005855616723070703, "data_time": 0.15699679851531984, "loss": 0.10914113745093346, "time": 0.8594900846481324, "epoch": 2, "memory": 25066, "step": 7058} +{"lr": 0.0005855616723070703, "data_time": 0.12544469833374022, "loss": 0.10099955797195434, "time": 0.8274879693984986, "epoch": 2, "memory": 25066, "step": 7158} +{"lr": 0.0005855616723070703, "data_time": 0.1368650197982788, "loss": 0.09467699006199837, "time": 0.8394469976425171, "epoch": 2, "memory": 25066, "step": 7258} +{"lr": 0.0005855616723070703, "data_time": 0.17312967777252197, "loss": 0.10345427542924882, "time": 0.8770464181900024, "epoch": 2, "memory": 25066, "step": 7358} +{"lr": 0.0005855616723070703, "data_time": 0.07103266716003417, "loss": 0.08968858756124973, "time": 0.7741149187088012, "epoch": 2, "memory": 25066, "step": 7458} +{"lr": 0.0005855616723070703, "data_time": 0.10962119102478027, "loss": 0.10697086453437805, "time": 0.8122239589691163, "epoch": 2, "memory": 25066, "step": 7558} +{"lr": 0.0005855616723070703, "data_time": 0.1046525478363037, "loss": 0.08235031291842461, "time": 0.8072562932968139, "epoch": 2, "memory": 25066, "step": 7658} +{"lr": 0.0005855616723070703, "data_time": 0.1263685941696167, "loss": 0.08875132799148559, "time": 0.8282488107681274, "epoch": 2, "memory": 25066, "step": 7758} +{"lr": 0.0005855616723070703, "data_time": 0.1342289686203003, "loss": 0.08926248177886009, "time": 0.83630530834198, "epoch": 2, "memory": 25066, "step": 7858} +{"lr": 0.0005855616723070703, "data_time": 0.11405665874481201, "loss": 0.0836502704769373, "time": 0.8176812887191772, "epoch": 2, "memory": 25066, "step": 7958} +{"lr": 0.0005855616723070703, "data_time": 0.13397748470306398, "loss": 0.09036344364285469, "time": 0.8371359586715699, "epoch": 2, "memory": 25066, "step": 8058} +{"lr": 0.0005855616723070703, "data_time": 0.12112700939178467, "loss": 0.09246006608009338, "time": 0.8227143049240112, "epoch": 2, "memory": 25066, "step": 8158} +{"lr": 0.0005855616723070703, "data_time": 0.09559087753295899, "loss": 0.076456942781806, "time": 0.8003002643585205, "epoch": 2, "memory": 25066, "step": 8258} +{"lr": 0.0005855616723070703, "data_time": 0.14178693294525146, "loss": 0.09103170521557331, "time": 0.8466941118240356, "epoch": 2, "memory": 25066, "step": 8358} +{"lr": 0.0005855616723070703, "data_time": 0.1208219051361084, "loss": 0.08048920668661594, "time": 0.8405956506729126, "epoch": 2, "memory": 25066, "step": 8458} +{"lr": 0.0005855616723070703, "data_time": 0.13285036087036134, "loss": 0.09214386157691479, "time": 0.8357965469360351, "epoch": 2, "memory": 25066, "step": 8558} +{"lr": 0.0005855616723070703, "data_time": 0.11866247653961182, "loss": 0.05654298588633537, "time": 0.8333797693252564, "epoch": 2, "memory": 25066, "step": 8658} +{"lr": 0.0005855616723070703, "data_time": 0.137612247467041, "loss": 0.061009171232581136, "time": 0.8441805124282837, "epoch": 2, "memory": 25066, "step": 8758} +{"lr": 0.0005855616723070703, "data_time": 0.13744451999664306, "loss": 0.06859854981303215, "time": 0.8431969404220581, "epoch": 2, "memory": 25066, "step": 8858} +{"lr": 0.0005855616723070703, "data_time": 0.15220577716827394, "loss": 0.06884837374091149, "time": 0.8548434734344482, "epoch": 2, "memory": 25066, "step": 8958} +{"lr": 0.0005855616723070703, "data_time": 0.11841278076171875, "loss": 0.07345729060471058, "time": 0.8282695770263672, "epoch": 2, "memory": 25066, "step": 9058} +{"lr": 0.0005855616723070703, "data_time": 0.11359946727752686, "loss": 0.07261206656694412, "time": 0.8156155586242676, "epoch": 2, "memory": 25066, "step": 9158} +{"lr": 0.0005855616723070703, "data_time": 0.16002295017242432, "loss": 0.07682518362998962, "time": 0.8735629081726074, "epoch": 2, "memory": 25066, "step": 9258} +{"lr": 0.0005855616723070703, "data_time": 0.1053574800491333, "loss": 0.0873219683766365, "time": 0.8075915098190307, "epoch": 2, "memory": 25066, "step": 9358} +{"lr": 0.0005855616723070703, "data_time": 0.15722007751464845, "loss": 0.054900694824755195, "time": 0.864040207862854, "epoch": 2, "memory": 25066, "step": 9458} +{"lr": 0.0005855616723070703, "data_time": 0.1907418489456177, "loss": 0.06413311734795571, "time": 0.8967849969863891, "epoch": 2, "memory": 25066, "step": 9558} +{"lr": 0.0005855616723070703, "data_time": 0.16151020526885987, "loss": 0.08038142397999763, "time": 0.8787475824356079, "epoch": 2, "memory": 25066, "step": 9658} +{"lr": 0.0005855616723070703, "data_time": 0.15345432758331298, "loss": 0.06994653046131134, "time": 0.8569607973098755, "epoch": 2, "memory": 25066, "step": 9758} +{"lr": 0.0005855616723070703, "data_time": 0.11273493766784667, "loss": 0.06123212277889252, "time": 0.8303402900695801, "epoch": 2, "memory": 25066, "step": 9858} +{"lr": 0.0005855616723070703, "data_time": 0.1861478328704834, "loss": 0.06094014495611191, "time": 0.891200065612793, "epoch": 2, "memory": 25066, "step": 9958} +{"lr": 0.0005855616723070703, "data_time": 0.13991703987121581, "loss": 0.05642457380890846, "time": 0.8469839334487915, "epoch": 2, "memory": 25066, "step": 10058} +{"lr": 0.0005855616723070703, "data_time": 0.203238844871521, "loss": 0.04999357163906097, "time": 0.908148193359375, "epoch": 2, "memory": 25066, "step": 10158} +{"lr": 0.0005855616723070703, "data_time": 0.11449542045593261, "loss": 0.07773915491998196, "time": 0.8254468202590942, "epoch": 2, "memory": 25066, "step": 10258} +{"lr": 0.0005855616723070703, "data_time": 0.11759653091430664, "loss": 0.05232198443263769, "time": 0.8228446960449218, "epoch": 2, "memory": 25066, "step": 10358} +{"lr": 0.0005855616723070703, "data_time": 0.08028585910797119, "loss": 0.04003726141527295, "time": 0.7837674856185913, "epoch": 2, "memory": 25066, "step": 10458} +{"lr": 0.0005855616723070703, "data_time": 0.1580594778060913, "loss": 0.056496366113424304, "time": 0.8600718736648559, "epoch": 2, "memory": 25066, "step": 10558} +{"lr": 0.0005855616723070703, "data_time": 0.15153591632843016, "loss": 0.05019565653055906, "time": 0.8647865056991577, "epoch": 2, "memory": 25066, "step": 10658} +{"lr": 0.0005855616723070703, "data_time": 0.14174971580505372, "loss": 0.04199137855321169, "time": 0.8465540170669555, "epoch": 2, "memory": 25066, "step": 10758} +{"lr": 0.0005855616723070703, "data_time": 0.12608003616333008, "loss": 0.040434060338884595, "time": 0.8379182577133178, "epoch": 2, "memory": 25066, "step": 10858} +{"lr": 0.0005855616723070703, "data_time": 0.15844414234161378, "loss": 0.05015103369951248, "time": 0.8607777118682861, "epoch": 2, "memory": 25066, "step": 10958} +{"lr": 0.0005855616723070703, "data_time": 0.17037336826324462, "loss": 0.05067151039838791, "time": 0.8828636169433594, "epoch": 2, "memory": 25066, "step": 11058} +{"lr": 0.0005855616723070703, "data_time": 0.09685750007629394, "loss": 0.039331309497356415, "time": 0.7996238470077515, "epoch": 2, "memory": 25066, "step": 11158} +{"lr": 0.0005855616723070703, "data_time": 0.099875807762146, "loss": 0.04945165850222111, "time": 0.8025046825408936, "epoch": 2, "memory": 25066, "step": 11258} +{"lr": 0.0005855616723070703, "data_time": 0.18228275775909425, "loss": 0.03875251393765211, "time": 0.8860106945037842, "epoch": 2, "memory": 25066, "step": 11358} +{"lr": 0.0005855616723070703, "data_time": 0.11300158500671387, "loss": 0.050302263163030145, "time": 0.8242648363113403, "epoch": 2, "memory": 25066, "step": 11458} +{"accuracy/top1": 96.83100891113281, "data_time": 0.2140779214746812, "time": 0.6079331706551945, "step": 2} +{"lr": 0.0005436600133406095, "data_time": 0.13033041954040528, "loss": 0.033195352740585805, "time": 0.8666026592254639, "epoch": 3, "memory": 25066, "step": 11616} +{"lr": 0.0005436600133406095, "data_time": 0.10392272472381592, "loss": 0.04524232670664787, "time": 0.8071052074432373, "epoch": 3, "memory": 25066, "step": 11716} +{"lr": 0.0005436600133406095, "data_time": 0.09729797840118408, "loss": 0.04777164980769157, "time": 0.80458664894104, "epoch": 3, "memory": 25066, "step": 11816} +{"lr": 0.0005436600133406095, "data_time": 0.10693371295928955, "loss": 0.05555416196584702, "time": 0.8090775728225708, "epoch": 3, "memory": 25066, "step": 11916} +{"lr": 0.0005436600133406095, "data_time": 0.12842955589294433, "loss": 0.0414990421384573, "time": 0.8455898284912109, "epoch": 3, "memory": 25066, "step": 12016} +{"lr": 0.0005436600133406095, "data_time": 0.10779645442962646, "loss": 0.0394628276117146, "time": 0.8118354082107544, "epoch": 3, "memory": 25066, "step": 12116} +{"lr": 0.0005436600133406095, "data_time": 0.10915007591247558, "loss": 0.03261984065175057, "time": 0.8189421415328979, "epoch": 3, "memory": 25066, "step": 12216} +{"lr": 0.0005436600133406095, "data_time": 0.08182039260864257, "loss": 0.023212652560323475, "time": 0.7845738649368286, "epoch": 3, "memory": 25066, "step": 12316} +{"lr": 0.0005436600133406095, "data_time": 0.09693608283996583, "loss": 0.04113286808133125, "time": 0.8177855253219605, "epoch": 3, "memory": 25066, "step": 12416} +{"lr": 0.0005436600133406095, "data_time": 0.12489337921142578, "loss": 0.03896365687251091, "time": 0.8280580043792725, "epoch": 3, "memory": 25066, "step": 12516} +{"lr": 0.0005436600133406095, "data_time": 0.13593392372131347, "loss": 0.032323870621621606, "time": 0.8391335487365723, "epoch": 3, "memory": 25066, "step": 12616} +{"lr": 0.0005436600133406095, "data_time": 0.19797844886779786, "loss": 0.04052624525502324, "time": 0.9034525632858277, "epoch": 3, "memory": 25066, "step": 12716} +{"lr": 0.0005436600133406095, "data_time": 0.10575785636901855, "loss": 0.031234129518270492, "time": 0.8078648328781128, "epoch": 3, "memory": 25066, "step": 12816} +{"lr": 0.0005436600133406095, "data_time": 0.15137965679168702, "loss": 0.02967866063117981, "time": 0.853244137763977, "epoch": 3, "memory": 25066, "step": 12916} +{"lr": 0.0005436600133406095, "data_time": 0.05598795413970947, "loss": 0.03397217188030481, "time": 0.7584885597229004, "epoch": 3, "memory": 25066, "step": 13016} +{"lr": 0.0005436600133406095, "data_time": 0.08288066387176514, "loss": 0.0368267472833395, "time": 0.7842445850372315, "epoch": 3, "memory": 25066, "step": 13116} +{"lr": 0.0005436600133406095, "data_time": 0.11959879398345948, "loss": 0.04880100060254335, "time": 0.8230604887008667, "epoch": 3, "memory": 25066, "step": 13216} +{"lr": 0.0005436600133406095, "data_time": 0.10724380016326904, "loss": 0.029367662314325572, "time": 0.820131516456604, "epoch": 3, "memory": 25066, "step": 13316} +{"lr": 0.0005436600133406095, "data_time": 0.10439128875732422, "loss": 0.02547192620113492, "time": 0.8074216842651367, "epoch": 3, "memory": 25066, "step": 13416} +{"lr": 0.0005436600133406095, "data_time": 0.151987886428833, "loss": 0.03506448734551668, "time": 0.8548566579818726, "epoch": 3, "memory": 25066, "step": 13516} +{"lr": 0.0005436600133406095, "data_time": 0.11074981689453126, "loss": 0.032025845162570475, "time": 0.8144988536834716, "epoch": 3, "memory": 25066, "step": 13616} +{"lr": 0.0005436600133406095, "data_time": 0.12121062278747559, "loss": 0.019510540971532465, "time": 0.8259487390518189, "epoch": 3, "memory": 25066, "step": 13716} +{"lr": 0.0005436600133406095, "data_time": 0.13341965675354003, "loss": 0.03174030315130949, "time": 0.8362045288085938, "epoch": 3, "memory": 25066, "step": 13816} +{"lr": 0.0005436600133406095, "data_time": 0.1493135929107666, "loss": 0.11955727338790893, "time": 0.8607951641082764, "epoch": 3, "memory": 25066, "step": 13916} +{"lr": 0.0005436600133406095, "data_time": 0.11190319061279297, "loss": 0.0288596716709435, "time": 0.8154298543930054, "epoch": 3, "memory": 25066, "step": 14016} +{"lr": 0.0005436600133406095, "data_time": 0.09707696437835693, "loss": 0.02947291750460863, "time": 0.8033147335052491, "epoch": 3, "memory": 25066, "step": 14116} +{"lr": 0.0005436600133406095, "data_time": 0.11541984081268311, "loss": 0.02915971577167511, "time": 0.817680811882019, "epoch": 3, "memory": 25066, "step": 14216} +{"lr": 0.0005436600133406095, "data_time": 0.21454989910125732, "loss": 0.023450619913637637, "time": 0.9236043930053711, "epoch": 3, "memory": 25066, "step": 14316} +{"lr": 0.0005436600133406095, "data_time": 0.07299714088439942, "loss": 0.022828979371115565, "time": 0.7761953353881836, "epoch": 3, "memory": 25066, "step": 14416} +{"lr": 0.0005436600133406095, "data_time": 0.068560791015625, "loss": 0.026883899234235287, "time": 0.790733814239502, "epoch": 3, "memory": 25066, "step": 14516} +{"lr": 0.0005436600133406095, "data_time": 0.11776745319366455, "loss": 0.031237490102648736, "time": 0.8204161405563355, "epoch": 3, "memory": 25066, "step": 14616} +{"lr": 0.0005436600133406095, "data_time": 0.07734129428863526, "loss": 0.022913479153066872, "time": 0.7978680372238159, "epoch": 3, "memory": 25066, "step": 14716} +{"lr": 0.0005436600133406095, "data_time": 0.06951143741607665, "loss": 0.026734100049361588, "time": 0.7740558624267578, "epoch": 3, "memory": 25066, "step": 14816} +{"lr": 0.0005436600133406095, "data_time": 0.14341337680816652, "loss": 0.022646971326321364, "time": 0.8562366008758545, "epoch": 3, "memory": 25066, "step": 14916} +{"lr": 0.0005436600133406095, "data_time": 0.07941281795501709, "loss": 0.028175718523561953, "time": 0.7817277431488037, "epoch": 3, "memory": 25066, "step": 15016} +{"lr": 0.0005436600133406095, "data_time": 0.0610595703125, "loss": 0.02311288514174521, "time": 0.7718296051025391, "epoch": 3, "memory": 25066, "step": 15116} +{"lr": 0.0005436600133406095, "data_time": 0.08676931858062745, "loss": 0.029936926439404488, "time": 0.7898554563522339, "epoch": 3, "memory": 25066, "step": 15216} +{"lr": 0.0005436600133406095, "data_time": 0.15036725997924805, "loss": 0.02967368345707655, "time": 0.8594315528869629, "epoch": 3, "memory": 25066, "step": 15316} +{"lr": 0.0005436600133406095, "data_time": 0.08284909725189209, "loss": 0.026235038973391057, "time": 0.7856340169906616, "epoch": 3, "memory": 25066, "step": 15416} +{"lr": 0.0005436600133406095, "data_time": 0.08458776473999023, "loss": 0.02139403447508812, "time": 0.8090615510940552, "epoch": 3, "memory": 25066, "step": 15516} +{"lr": 0.0005436600133406095, "data_time": 0.13745789527893065, "loss": 0.023093489464372396, "time": 0.840945029258728, "epoch": 3, "memory": 25066, "step": 15616} +{"lr": 0.0005436600133406095, "data_time": 0.12347440719604492, "loss": 0.02739778677932918, "time": 0.829805850982666, "epoch": 3, "memory": 25066, "step": 15716} +{"lr": 0.0005436600133406095, "data_time": 0.08062148094177246, "loss": 0.024879414681345226, "time": 0.7840144872665405, "epoch": 3, "memory": 25066, "step": 15816} +{"lr": 0.0005436600133406095, "data_time": 0.08028206825256348, "loss": 0.018084383476525544, "time": 0.7857820272445679, "epoch": 3, "memory": 25066, "step": 15916} +{"lr": 0.0005436600133406095, "data_time": 0.10027310848236085, "loss": 0.023039410542696714, "time": 0.8037139654159546, "epoch": 3, "memory": 25066, "step": 16016} +{"lr": 0.0005436600133406095, "data_time": 0.07260229587554931, "loss": 0.035664135776460174, "time": 0.7827414751052857, "epoch": 3, "memory": 25066, "step": 16116} +{"lr": 0.0005436600133406095, "data_time": 0.10394108295440674, "loss": 0.025487054558470845, "time": 0.8065945148468018, "epoch": 3, "memory": 25066, "step": 16216} +{"lr": 0.0005436600133406095, "data_time": 0.10904536247253419, "loss": 0.02932376405224204, "time": 0.8145534992218018, "epoch": 3, "memory": 25066, "step": 16316} +{"lr": 0.0005436600133406095, "data_time": 0.1746471643447876, "loss": 0.025024187564849854, "time": 0.8810407161712647, "epoch": 3, "memory": 25066, "step": 16416} +{"lr": 0.0005436600133406095, "data_time": 0.1664111852645874, "loss": 0.023270398005843164, "time": 0.8818182468414306, "epoch": 3, "memory": 25066, "step": 16516} +{"lr": 0.0005436600133406095, "data_time": 0.1623560905456543, "loss": 0.023703093780204652, "time": 0.8650493383407593, "epoch": 3, "memory": 25066, "step": 16616} +{"lr": 0.0005436600133406095, "data_time": 0.15368180274963378, "loss": 0.024507660139352083, "time": 0.8561034202575684, "epoch": 3, "memory": 25066, "step": 16716} +{"lr": 0.0005436600133406095, "data_time": 0.12068026065826416, "loss": 0.02562913470901549, "time": 0.8227468490600586, "epoch": 3, "memory": 25066, "step": 16816} +{"lr": 0.0005436600133406095, "data_time": 0.08420326709747314, "loss": 0.015059865079820156, "time": 0.7977602243423462, "epoch": 3, "memory": 25066, "step": 16916} +{"lr": 0.0005436600133406095, "data_time": 0.0689265251159668, "loss": 0.02647847877815366, "time": 0.7741671800613403, "epoch": 3, "memory": 25066, "step": 17016} +{"lr": 0.0005436600133406095, "data_time": 0.08307995796203613, "loss": 0.01928577646613121, "time": 0.7887391567230224, "epoch": 3, "memory": 25066, "step": 17116} +{"lr": 0.0005436600133406095, "data_time": 0.03762664794921875, "loss": 0.017405051505193114, "time": 0.7398056745529175, "epoch": 3, "memory": 25066, "step": 17216} +{"accuracy/top1": 99.26140594482422, "data_time": 0.20285582542419434, "time": 0.586881721720976, "step": 3} +{"lr": 0.0004783966494262796, "data_time": 0.08802928924560546, "loss": 0.018302278080955147, "time": 0.7976623773574829, "epoch": 4, "memory": 25066, "step": 17374} +{"lr": 0.0004783966494262796, "data_time": 0.06969025135040283, "loss": 0.01742699109017849, "time": 0.774363660812378, "epoch": 4, "memory": 25066, "step": 17474} +{"lr": 0.0004783966494262796, "data_time": 0.10924909114837647, "loss": 0.018951719999313353, "time": 0.8121732234954834, "epoch": 4, "memory": 25066, "step": 17574} +{"lr": 0.0004783966494262796, "data_time": 0.093589186668396, "loss": 0.0143913384526968, "time": 0.7974068403244019, "epoch": 4, "memory": 25066, "step": 17674} +{"lr": 0.0004783966494262796, "data_time": 0.09095463752746583, "loss": 0.011356176808476448, "time": 0.8005338430404663, "epoch": 4, "memory": 25066, "step": 17774} +{"lr": 0.0004783966494262796, "data_time": 0.11034657955169677, "loss": 0.018610521452501416, "time": 0.813131594657898, "epoch": 4, "memory": 25066, "step": 17874} +{"lr": 0.0004783966494262796, "data_time": 0.10225954055786132, "loss": 0.01797268958762288, "time": 0.8075380086898803, "epoch": 4, "memory": 25066, "step": 17974} +{"lr": 0.0004783966494262796, "data_time": 0.14429073333740233, "loss": 0.01446473179385066, "time": 0.8488998413085938, "epoch": 4, "memory": 25066, "step": 18074} +{"lr": 0.0004783966494262796, "data_time": 0.03730018138885498, "loss": 0.017122942581772806, "time": 0.7503591537475586, "epoch": 4, "memory": 25066, "step": 18174} +{"lr": 0.0004783966494262796, "data_time": 0.10412406921386719, "loss": 0.010631949035450816, "time": 0.8075726270675659, "epoch": 4, "memory": 25066, "step": 18274} +{"lr": 0.0004783966494262796, "data_time": 0.042281508445739746, "loss": 0.014166658092290164, "time": 0.7447845935821533, "epoch": 4, "memory": 25066, "step": 18374} +{"lr": 0.0004783966494262796, "data_time": 0.16450142860412598, "loss": 0.021440746309235693, "time": 0.8690449476242066, "epoch": 4, "memory": 25066, "step": 18474} +{"lr": 0.0004783966494262796, "data_time": 0.08441944122314453, "loss": 0.04489952577278018, "time": 0.7867696762084961, "epoch": 4, "memory": 25066, "step": 18574} +{"lr": 0.0004783966494262796, "data_time": 0.03972434997558594, "loss": 0.013026199210435152, "time": 0.7427253246307373, "epoch": 4, "memory": 25066, "step": 18674} +{"lr": 0.0004783966494262796, "data_time": 0.08989651203155517, "loss": 0.017057787952944638, "time": 0.7940354108810425, "epoch": 4, "memory": 25066, "step": 18774} +{"lr": 0.0004783966494262796, "data_time": 0.07190797328948975, "loss": 0.015776285761967303, "time": 0.7752673149108886, "epoch": 4, "memory": 25066, "step": 18874} +{"lr": 0.0004783966494262796, "data_time": 0.10939245223999024, "loss": 0.013684185827150941, "time": 0.812546181678772, "epoch": 4, "memory": 25066, "step": 18974} +{"lr": 0.0004783966494262796, "data_time": 0.12172412872314453, "loss": 0.01371868490241468, "time": 0.8262057065963745, "epoch": 4, "memory": 25066, "step": 19074} +{"lr": 0.0004783966494262796, "data_time": 0.1257690668106079, "loss": 0.010920207994058729, "time": 0.8281398296356202, "epoch": 4, "memory": 25066, "step": 19174} +{"lr": 0.0004783966494262796, "data_time": 0.040434837341308594, "loss": 0.012805491732433439, "time": 0.7801449298858643, "epoch": 4, "memory": 25066, "step": 19274} +{"lr": 0.0004783966494262796, "data_time": 0.20392210483551027, "loss": 0.01414217851124704, "time": 0.9109137535095215, "epoch": 4, "memory": 25066, "step": 19374} +{"lr": 0.0004783966494262796, "data_time": 0.1468379020690918, "loss": 0.03519926955923438, "time": 0.8494501352310181, "epoch": 4, "memory": 25066, "step": 19474} +{"lr": 0.0004783966494262796, "data_time": 0.18970537185668945, "loss": 0.01362181599251926, "time": 0.9044387102127075, "epoch": 4, "memory": 25066, "step": 19574} +{"lr": 0.0004783966494262796, "data_time": 0.13027243614196776, "loss": 0.011255398346111178, "time": 0.8407919645309448, "epoch": 4, "memory": 25066, "step": 19674} +{"lr": 0.0004783966494262796, "data_time": 0.15535576343536378, "loss": 0.02364458399824798, "time": 0.8576542139053345, "epoch": 4, "memory": 25066, "step": 19774} +{"lr": 0.0004783966494262796, "data_time": 0.03958296775817871, "loss": 0.014590098150074482, "time": 0.7521777391433716, "epoch": 4, "memory": 25066, "step": 19874} +{"lr": 0.0004783966494262796, "data_time": 0.06697421073913574, "loss": 0.015909790294244885, "time": 0.7691185235977173, "epoch": 4, "memory": 25066, "step": 19974} +{"lr": 0.0004783966494262796, "data_time": 0.06014094352722168, "loss": 0.02140692276880145, "time": 0.7619815111160279, "epoch": 4, "memory": 25066, "step": 20074} +{"lr": 0.0004783966494262796, "data_time": 0.11686837673187256, "loss": 0.01703349258750677, "time": 0.819719386100769, "epoch": 4, "memory": 25066, "step": 20174} +{"lr": 0.0004783966494262796, "data_time": 0.11401152610778809, "loss": 0.012818874884396791, "time": 0.8164216995239257, "epoch": 4, "memory": 25066, "step": 20274} +{"lr": 0.0004783966494262796, "data_time": 0.05924406051635742, "loss": 0.013755055237561464, "time": 0.7617597103118896, "epoch": 4, "memory": 25066, "step": 20374} +{"lr": 0.0004783966494262796, "data_time": 0.11467101573944091, "loss": 0.014929664228111506, "time": 0.8307372570037842, "epoch": 4, "memory": 25066, "step": 20474} +{"lr": 0.0004783966494262796, "data_time": 0.05902366638183594, "loss": 0.01392804174683988, "time": 0.7617453336715698, "epoch": 4, "memory": 25066, "step": 20574} +{"lr": 0.0004783966494262796, "data_time": 0.05788445472717285, "loss": 0.010660858545452356, "time": 0.7684449672698974, "epoch": 4, "memory": 25066, "step": 20674} +{"lr": 0.0004783966494262796, "data_time": 0.13744583129882812, "loss": 0.01604933012276888, "time": 0.8561843156814575, "epoch": 4, "memory": 25066, "step": 20774} +{"lr": 0.0004783966494262796, "data_time": 0.044433426856994626, "loss": 0.011551185231655835, "time": 0.7534519672393799, "epoch": 4, "memory": 25066, "step": 20874} +{"lr": 0.0004783966494262796, "data_time": 0.08031361103057862, "loss": 0.02041889582760632, "time": 0.7964224576950073, "epoch": 4, "memory": 25066, "step": 20974} +{"lr": 0.0004783966494262796, "data_time": 0.058609390258789064, "loss": 0.010382734099403023, "time": 0.7653856992721557, "epoch": 4, "memory": 25066, "step": 21074} +{"lr": 0.0004783966494262796, "data_time": 0.06863110065460205, "loss": 0.008980284724384547, "time": 0.772297477722168, "epoch": 4, "memory": 25066, "step": 21174} +{"lr": 0.0004783966494262796, "data_time": 0.04592280387878418, "loss": 0.01715405434370041, "time": 0.7556788921356201, "epoch": 4, "memory": 25066, "step": 21274} +{"lr": 0.0004783966494262796, "data_time": 0.0480712890625, "loss": 0.015487766405567528, "time": 0.7565844535827637, "epoch": 4, "memory": 25066, "step": 21374} +{"lr": 0.0004783966494262796, "data_time": 0.013774728775024414, "loss": 0.01465374231338501, "time": 0.7266249418258667, "epoch": 4, "memory": 25066, "step": 21474} +{"lr": 0.0004783966494262796, "data_time": 0.05666790008544922, "loss": 0.010056767472997307, "time": 0.7723235130310059, "epoch": 4, "memory": 25066, "step": 21574} +{"lr": 0.0004783966494262796, "data_time": 0.03183979988098144, "loss": 0.015812888741493225, "time": 0.7344065427780151, "epoch": 4, "memory": 25066, "step": 21674} +{"lr": 0.0004783966494262796, "data_time": 0.10094289779663086, "loss": 0.0076159191317856315, "time": 0.8159659147262573, "epoch": 4, "memory": 25066, "step": 21774} +{"lr": 0.0004783966494262796, "data_time": 0.0774773359298706, "loss": 0.025784523971378804, "time": 0.7799385547637939, "epoch": 4, "memory": 25066, "step": 21874} +{"lr": 0.0004783966494262796, "data_time": 0.07479071617126465, "loss": 0.009562530554831028, "time": 0.7869988203048706, "epoch": 4, "memory": 25066, "step": 21974} +{"lr": 0.0004783966494262796, "data_time": 0.10151920318603516, "loss": 0.01564459167420864, "time": 0.8041109800338745, "epoch": 4, "memory": 25066, "step": 22074} +{"lr": 0.0004783966494262796, "data_time": 0.0779494047164917, "loss": 0.011913094436749816, "time": 0.7937377691268921, "epoch": 4, "memory": 25066, "step": 22174} +{"lr": 0.0004783966494262796, "data_time": 0.07155652046203613, "loss": 0.013006499083712698, "time": 0.773443603515625, "epoch": 4, "memory": 25066, "step": 22274} +{"lr": 0.0004783966494262796, "data_time": 0.05156948566436768, "loss": 0.01421361225657165, "time": 0.7727417945861816, "epoch": 4, "memory": 25066, "step": 22374} +{"lr": 0.0004783966494262796, "data_time": 0.06946351528167724, "loss": 0.01124355997890234, "time": 0.7742018461227417, "epoch": 4, "memory": 25066, "step": 22474} +{"lr": 0.0004783966494262796, "data_time": 0.1365509510040283, "loss": 0.007517954707145691, "time": 0.8512271642684937, "epoch": 4, "memory": 25066, "step": 22574} +{"lr": 0.0004783966494262796, "data_time": 0.11934657096862793, "loss": 0.01400508708320558, "time": 0.8221124887466431, "epoch": 4, "memory": 25066, "step": 22674} +{"lr": 0.0004783966494262796, "data_time": 0.08114609718322754, "loss": 0.009509793762117625, "time": 0.7843251943588256, "epoch": 4, "memory": 25066, "step": 22774} +{"lr": 0.0004783966494262796, "data_time": 0.06932976245880126, "loss": 0.007604291290044784, "time": 0.772094988822937, "epoch": 4, "memory": 25066, "step": 22874} +{"lr": 0.0004783966494262796, "data_time": 0.004683208465576172, "loss": 0.011296708742156625, "time": 0.7076105117797852, "epoch": 4, "memory": 25066, "step": 22974} +{"accuracy/top1": 99.44447326660156, "data_time": 0.2145690497230081, "time": 0.6036080612855799, "step": 4} +{"lr": 0.0003961600133406095, "data_time": 0.022608208656311034, "loss": 0.004102718085050583, "time": 0.7602856159210205, "epoch": 5, "memory": 25066, "step": 23132} +{"lr": 0.0003961600133406095, "data_time": 0.14658496379852295, "loss": 0.004557293513789773, "time": 0.8631012916564942, "epoch": 5, "memory": 25066, "step": 23232} +{"lr": 0.0003961600133406095, "data_time": 0.09179384708404541, "loss": 0.01601978885009885, "time": 0.8112494468688964, "epoch": 5, "memory": 25066, "step": 23332} +{"lr": 0.0003961600133406095, "data_time": 0.05205504894256592, "loss": 0.009309962298721075, "time": 0.7570523977279663, "epoch": 5, "memory": 25066, "step": 23432} +{"lr": 0.0003961600133406095, "data_time": 0.06578717231750489, "loss": 0.005065946886315942, "time": 0.7910119771957398, "epoch": 5, "memory": 25066, "step": 23532} +{"lr": 0.0003961600133406095, "data_time": 0.07953751087188721, "loss": 0.006550627620890736, "time": 0.7847600936889648, "epoch": 5, "memory": 25066, "step": 23632} +{"lr": 0.0003961600133406095, "data_time": 0.040402793884277345, "loss": 0.01310245501808822, "time": 0.747068190574646, "epoch": 5, "memory": 25066, "step": 23732} +{"lr": 0.0003961600133406095, "data_time": 0.08198513984680175, "loss": 0.013690729532390833, "time": 0.7846913814544678, "epoch": 5, "memory": 25066, "step": 23832} +{"lr": 0.0003961600133406095, "data_time": 0.16190268993377685, "loss": 0.006088630622252822, "time": 0.8648688077926636, "epoch": 5, "memory": 25066, "step": 23932} +{"lr": 0.0003961600133406095, "data_time": 0.0015821456909179688, "loss": 0.013069759169593454, "time": 0.7684478044509888, "epoch": 5, "memory": 25066, "step": 24032} +{"lr": 0.0003961600133406095, "data_time": 0.0012533187866210938, "loss": 0.006365784769877791, "time": 0.8579529523849487, "epoch": 5, "memory": 25066, "step": 24132} +{"lr": 0.0003961600133406095, "data_time": 0.001192927360534668, "loss": 0.015119435312226415, "time": 0.8786704063415527, "epoch": 5, "memory": 25066, "step": 24232} +{"lr": 0.0003961600133406095, "data_time": 0.0012086153030395508, "loss": 0.006258977996185422, "time": 1.0646855592727662, "epoch": 5, "memory": 25066, "step": 24332} +{"lr": 0.0003961600133406095, "data_time": 0.15270371437072755, "loss": 0.0053798640612512825, "time": 1.0390119552612305, "epoch": 5, "memory": 25066, "step": 24432} +{"lr": 0.0003961600133406095, "data_time": 0.20881047248840331, "loss": 0.00460607442073524, "time": 1.2325568437576293, "epoch": 5, "memory": 25066, "step": 24532} +{"lr": 0.0003961600133406095, "data_time": 0.0013935327529907226, "loss": 0.0112472013104707, "time": 0.7947478771209717, "epoch": 5, "memory": 25066, "step": 24632} +{"lr": 0.0003961600133406095, "data_time": 0.025103092193603516, "loss": 0.010350334830582142, "time": 0.864455270767212, "epoch": 5, "memory": 25066, "step": 24732} +{"lr": 0.0003961600133406095, "data_time": 0.0013593196868896484, "loss": 0.007137893233448267, "time": 0.8959303140640259, "epoch": 5, "memory": 25066, "step": 24832} +{"lr": 0.0003961600133406095, "data_time": 0.0014004945755004884, "loss": 0.011522663012146949, "time": 1.0799331188201904, "epoch": 5, "memory": 25066, "step": 24932} +{"lr": 0.0003961600133406095, "data_time": 0.0013248205184936523, "loss": 0.012295640585944057, "time": 0.9674221754074097, "epoch": 5, "memory": 25066, "step": 25032} +{"lr": 0.0003961600133406095, "data_time": 0.001428365707397461, "loss": 0.006364683760330081, "time": 1.0895797729492187, "epoch": 5, "memory": 25066, "step": 25132} +{"lr": 0.0003961600133406095, "data_time": 0.001578354835510254, "loss": 0.0019763521384447815, "time": 1.088343048095703, "epoch": 5, "memory": 25066, "step": 25232} +{"lr": 0.0003961600133406095, "data_time": 0.0014547109603881836, "loss": 0.007758986670523882, "time": 1.1326516151428223, "epoch": 5, "memory": 25066, "step": 25332} +{"lr": 0.0003961600133406095, "data_time": 0.2643946886062622, "loss": 0.008848836505785584, "time": 0.9786058902740479, "epoch": 5, "memory": 25066, "step": 25432} +{"lr": 0.0003961600133406095, "data_time": 0.37523105144500735, "loss": 0.007231908617541194, "time": 1.0892903089523316, "epoch": 5, "memory": 25066, "step": 25532} +{"lr": 0.0003961600133406095, "data_time": 0.0012932300567626953, "loss": 0.012182714557275176, "time": 1.02841956615448, "epoch": 5, "memory": 25066, "step": 25632} +{"lr": 0.0003961600133406095, "data_time": 0.001392364501953125, "loss": 0.01186219952069223, "time": 0.8187432527542114, "epoch": 5, "memory": 25066, "step": 25732} +{"lr": 0.0003961600133406095, "data_time": 0.0015081644058227539, "loss": 0.009799295058473944, "time": 0.7864231586456298, "epoch": 5, "memory": 25066, "step": 25832} +{"lr": 0.0003961600133406095, "data_time": 0.00128173828125, "loss": 0.004450850514695048, "time": 0.7898314237594605, "epoch": 5, "memory": 25066, "step": 25932} +{"lr": 0.0003961600133406095, "data_time": 0.0014527797698974609, "loss": 0.010132176661863923, "time": 0.8628645181655884, "epoch": 5, "memory": 25066, "step": 26032} +{"lr": 0.0003961600133406095, "data_time": 0.0013979673385620117, "loss": 0.012101198825985194, "time": 0.8257627725601197, "epoch": 5, "memory": 25066, "step": 26132} +{"lr": 0.0003961600133406095, "data_time": 0.0016843557357788086, "loss": 0.007321288902312517, "time": 0.8655916452407837, "epoch": 5, "memory": 25066, "step": 26232} +{"lr": 0.0003961600133406095, "data_time": 0.001238250732421875, "loss": 0.010624748328700662, "time": 0.796226692199707, "epoch": 5, "memory": 25066, "step": 26332} +{"lr": 0.0003961600133406095, "data_time": 0.0017630815505981444, "loss": 0.010346449492499233, "time": 0.8071348667144775, "epoch": 5, "memory": 25066, "step": 26432} +{"lr": 0.0003961600133406095, "data_time": 0.0013033390045166016, "loss": 0.01157821766100824, "time": 0.8289038181304932, "epoch": 5, "memory": 25066, "step": 26532} +{"lr": 0.0003961600133406095, "data_time": 0.001332259178161621, "loss": 0.0075154932215809826, "time": 0.784517765045166, "epoch": 5, "memory": 25066, "step": 26632} +{"lr": 0.0003961600133406095, "data_time": 0.0014123201370239257, "loss": 0.007396357506513596, "time": 0.8106993675231934, "epoch": 5, "memory": 25066, "step": 26732} +{"lr": 0.0003961600133406095, "data_time": 0.0013507604598999023, "loss": 0.009038366330787539, "time": 0.817111873626709, "epoch": 5, "memory": 25066, "step": 26832} +{"lr": 0.0003961600133406095, "data_time": 0.001336216926574707, "loss": 0.0063036609906703235, "time": 0.7969225168228149, "epoch": 5, "memory": 25066, "step": 26932} +{"lr": 0.0003961600133406095, "data_time": 0.0014711141586303711, "loss": 0.006457010982558131, "time": 0.7924376964569092, "epoch": 5, "memory": 25066, "step": 27032} +{"lr": 0.0003961600133406095, "data_time": 0.0013987302780151367, "loss": 0.010920880734920502, "time": 0.8514780521392822, "epoch": 5, "memory": 25066, "step": 27132} +{"lr": 0.0003961600133406095, "data_time": 0.001439666748046875, "loss": 0.004842010932043195, "time": 0.8348291158676148, "epoch": 5, "memory": 25066, "step": 27232} +{"lr": 0.0003961600133406095, "data_time": 0.0014387130737304687, "loss": 0.011458777915686369, "time": 0.8007415056228637, "epoch": 5, "memory": 25066, "step": 27332} +{"lr": 0.0003961600133406095, "data_time": 0.001244974136352539, "loss": 0.004842579970136285, "time": 0.8030724287033081, "epoch": 5, "memory": 25066, "step": 27432} +{"lr": 0.0003961600133406095, "data_time": 0.0013063669204711914, "loss": 0.0015838440507650376, "time": 0.8247030973434448, "epoch": 5, "memory": 25066, "step": 27532} +{"lr": 0.0003961600133406095, "data_time": 0.002011609077453613, "loss": 0.007811467070132494, "time": 0.8476327180862426, "epoch": 5, "memory": 25066, "step": 27632} +{"lr": 0.0003961600133406095, "data_time": 0.0012860536575317384, "loss": 0.007589854020625353, "time": 0.8103592395782471, "epoch": 5, "memory": 25066, "step": 27732} +{"lr": 0.0003961600133406095, "data_time": 0.0011636018753051758, "loss": 0.007840817188844084, "time": 0.7907381057739258, "epoch": 5, "memory": 25066, "step": 27832} +{"lr": 0.0003961600133406095, "data_time": 0.0013646125793457032, "loss": 0.007127438858151436, "time": 0.816729211807251, "epoch": 5, "memory": 25066, "step": 27932} +{"lr": 0.0003961600133406095, "data_time": 0.0014326095581054688, "loss": 0.00728184673935175, "time": 0.8198399782180786, "epoch": 5, "memory": 25066, "step": 28032} +{"lr": 0.0003961600133406095, "data_time": 0.0013283252716064452, "loss": 0.004735107207670808, "time": 0.8231123447418213, "epoch": 5, "memory": 25066, "step": 28132} +{"lr": 0.0003961600133406095, "data_time": 0.0012480497360229492, "loss": 0.007624412514269352, "time": 0.8076583862304687, "epoch": 5, "memory": 25066, "step": 28232} +{"lr": 0.0003961600133406095, "data_time": 0.001441216468811035, "loss": 0.010951170464977622, "time": 0.8081752538681031, "epoch": 5, "memory": 25066, "step": 28332} +{"lr": 0.0003961600133406095, "data_time": 0.0013387203216552734, "loss": 0.009171189460903405, "time": 0.8238954067230224, "epoch": 5, "memory": 25066, "step": 28432} +{"lr": 0.0003961600133406095, "data_time": 0.0013531208038330077, "loss": 0.011602256866171955, "time": 0.8434428453445435, "epoch": 5, "memory": 25066, "step": 28532} +{"lr": 0.0003961600133406095, "data_time": 0.0013370275497436523, "loss": 0.005083956802263856, "time": 0.8207385540008545, "epoch": 5, "memory": 25066, "step": 28632} +{"lr": 0.0003961600133406095, "data_time": 0.001479792594909668, "loss": 0.009026247914880514, "time": 0.8370945453643799, "epoch": 5, "memory": 25066, "step": 28732} +{"accuracy/top1": 99.70330047607422, "data_time": 0.20719463684979608, "time": 0.5988076574662152, "step": 5} +{"lr": 0.000305, "data_time": 0.0014534473419189453, "loss": 0.003872048668563366, "time": 0.8239111423492431, "epoch": 6, "memory": 25066, "step": 28890} +{"lr": 0.000305, "data_time": 0.0012253999710083007, "loss": 0.0026534893549978733, "time": 0.7924069166183472, "epoch": 6, "memory": 25066, "step": 28990} +{"lr": 0.000305, "data_time": 0.0014041900634765626, "loss": 0.005696064792573452, "time": 0.7971612453460694, "epoch": 6, "memory": 25066, "step": 29090} +{"lr": 0.000305, "data_time": 0.001421213150024414, "loss": 0.019828826701268554, "time": 0.7973195314407349, "epoch": 6, "memory": 25066, "step": 29190} +{"lr": 0.000305, "data_time": 0.0012470245361328124, "loss": 0.006403261562809348, "time": 0.7931866884231568, "epoch": 6, "memory": 25066, "step": 29290} +{"lr": 0.000305, "data_time": 0.0012543201446533203, "loss": 0.004346122406423092, "time": 0.8728467941284179, "epoch": 6, "memory": 25066, "step": 29390} +{"lr": 0.000305, "data_time": 0.0012321949005126953, "loss": 0.005958028463646769, "time": 0.882062578201294, "epoch": 6, "memory": 25066, "step": 29490} +{"lr": 0.000305, "data_time": 0.0013159990310668945, "loss": 0.0047013307455927135, "time": 0.7847998380661011, "epoch": 6, "memory": 25066, "step": 29590} +{"lr": 0.000305, "data_time": 0.0016243457794189453, "loss": 0.005479770759120583, "time": 0.8502360820770264, "epoch": 6, "memory": 25066, "step": 29690} +{"lr": 0.000305, "data_time": 0.0012400150299072266, "loss": 0.00786884999834001, "time": 0.8248006343841553, "epoch": 6, "memory": 25066, "step": 29790} +{"lr": 0.000305, "data_time": 0.01263282299041748, "loss": 0.0031838535331189634, "time": 0.8590489387512207, "epoch": 6, "memory": 25066, "step": 29890} +{"lr": 0.000305, "data_time": 0.0013402938842773438, "loss": 0.009283325169235468, "time": 0.861748218536377, "epoch": 6, "memory": 25066, "step": 29990} +{"lr": 0.000305, "data_time": 0.0013037443161010743, "loss": 0.008197297761216759, "time": 0.8198406934738159, "epoch": 6, "memory": 25066, "step": 30090} +{"lr": 0.000305, "data_time": 0.0014167308807373046, "loss": 0.0068174193147569895, "time": 0.932839322090149, "epoch": 6, "memory": 25066, "step": 30190} +{"lr": 0.000305, "data_time": 0.0013034820556640625, "loss": 0.0017456692177802325, "time": 0.8413915157318115, "epoch": 6, "memory": 25066, "step": 30290} diff --git a/convnext_small_4xb256_fake5m_lr6e-4_10epoch/20230601_151930/vis_data/vis_image/0000000.png_1.png b/convnext_small_4xb256_fake5m_lr6e-4_10epoch/20230601_151930/vis_data/vis_image/0000000.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..2c16504fc568a1704c637bf4583d882b94e52a7a Binary files /dev/null and b/convnext_small_4xb256_fake5m_lr6e-4_10epoch/20230601_151930/vis_data/vis_image/0000000.png_1.png differ diff --git a/convnext_small_4xb256_fake5m_lr6e-4_10epoch/20230601_151930/vis_data/vis_image/0000000.png_2.png b/convnext_small_4xb256_fake5m_lr6e-4_10epoch/20230601_151930/vis_data/vis_image/0000000.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..e1606975577487c3b0be6dd199d6f550e6fa62c5 Binary files /dev/null and b/convnext_small_4xb256_fake5m_lr6e-4_10epoch/20230601_151930/vis_data/vis_image/0000000.png_2.png differ diff --git a/convnext_small_4xb256_fake5m_lr6e-4_10epoch/20230601_151930/vis_data/vis_image/0000000.png_3.png b/convnext_small_4xb256_fake5m_lr6e-4_10epoch/20230601_151930/vis_data/vis_image/0000000.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..6f264824565adf0630fdfc2e331f1ad6f2495172 Binary files /dev/null and b/convnext_small_4xb256_fake5m_lr6e-4_10epoch/20230601_151930/vis_data/vis_image/0000000.png_3.png differ diff --git a/convnext_small_4xb256_fake5m_lr6e-4_10epoch/20230601_151930/vis_data/vis_image/0000000.png_4.png b/convnext_small_4xb256_fake5m_lr6e-4_10epoch/20230601_151930/vis_data/vis_image/0000000.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..6f264824565adf0630fdfc2e331f1ad6f2495172 Binary files /dev/null and b/convnext_small_4xb256_fake5m_lr6e-4_10epoch/20230601_151930/vis_data/vis_image/0000000.png_4.png differ diff --git a/convnext_small_4xb256_fake5m_lr6e-4_10epoch/20230601_151930/vis_data/vis_image/0000000.png_5.png b/convnext_small_4xb256_fake5m_lr6e-4_10epoch/20230601_151930/vis_data/vis_image/0000000.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..6f264824565adf0630fdfc2e331f1ad6f2495172 Binary files /dev/null and b/convnext_small_4xb256_fake5m_lr6e-4_10epoch/20230601_151930/vis_data/vis_image/0000000.png_5.png differ diff --git a/convnext_small_4xb256_fake5m_lr6e-4_10epoch/convnext_small_4xb256_fake5m_lr6e-4_10epoch.py b/convnext_small_4xb256_fake5m_lr6e-4_10epoch/convnext_small_4xb256_fake5m_lr6e-4_10epoch.py new file mode 100644 index 0000000000000000000000000000000000000000..dce5e76162f46d520159aeda7be0d77c281cf731 --- /dev/null +++ b/convnext_small_4xb256_fake5m_lr6e-4_10epoch/convnext_small_4xb256_fake5m_lr6e-4_10epoch.py @@ -0,0 +1,169 @@ +optim_wrapper = dict( + optimizer=dict( + type='AdamW', + lr=0.0006, + weight_decay=0.05, + eps=1e-08, + betas=(0.9, 0.999), + _scope_='mmpretrain'), + paramwise_cfg=dict( + norm_decay_mult=0.0, + bias_decay_mult=0.0, + flat_decay_mult=0.0, + custom_keys=dict({ + '.absolute_pos_embed': dict(decay_mult=0.0), + '.relative_position_bias_table': dict(decay_mult=0.0) + })), + type='AmpOptimWrapper', + dtype='bfloat16', + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=True, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=4096) +model = dict( + type='ImageClassifier', + backbone=dict(type='ConvNeXt', arch='small', drop_path_rate=0.4), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=768, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + init_cfg=None), + init_cfg=dict( + type='TruncNormal', layer=['Conv2d', 'Linear'], std=0.02, bias=0.0), + train_cfg=None) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stablediffusionV1-5R2-dpmsolver-25-5m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/convnext_small_4xb256_fake5m_lr6e-4_10epoch' diff --git a/convnext_small_4xb256_fake5m_lr6e-4_10epoch/epoch_1.pth b/convnext_small_4xb256_fake5m_lr6e-4_10epoch/epoch_1.pth new file mode 100644 index 0000000000000000000000000000000000000000..5855d453f009f1a447ef76e90a18017a5b6e8e20 --- /dev/null +++ b/convnext_small_4xb256_fake5m_lr6e-4_10epoch/epoch_1.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:2ebe9cde03dc7d7eb000d782fd139eae98c27afc5adaf654758199fe56095453 +size 792300192 diff --git a/convnext_small_4xb256_fake5m_lr6e-4_10epoch/epoch_2.pth b/convnext_small_4xb256_fake5m_lr6e-4_10epoch/epoch_2.pth new file mode 100644 index 0000000000000000000000000000000000000000..e9e1d09c2e7f369089200f5af292211e9f0b3a68 --- /dev/null +++ b/convnext_small_4xb256_fake5m_lr6e-4_10epoch/epoch_2.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:66e753bf9bf17f837babf349379e1d852e27e315de7f0f5aeadec2561664aafc +size 792721760 diff --git a/convnext_small_4xb256_fake5m_lr6e-4_10epoch/epoch_3.pth b/convnext_small_4xb256_fake5m_lr6e-4_10epoch/epoch_3.pth new file mode 100644 index 0000000000000000000000000000000000000000..2323749b502a2f6b6b4b02281d25cc9d7fff5a3a --- /dev/null +++ b/convnext_small_4xb256_fake5m_lr6e-4_10epoch/epoch_3.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:4cf9f5953a81598bd479e2700f90086c48d4bedef8679c34f6b6e0ecb788b4f0 +size 793147744 diff --git a/convnext_small_4xb256_fake5m_lr6e-4_10epoch/epoch_4.pth b/convnext_small_4xb256_fake5m_lr6e-4_10epoch/epoch_4.pth new file mode 100644 index 0000000000000000000000000000000000000000..2730edbdcdc909bdc98979acd3fea62095d4ef76 --- /dev/null +++ b/convnext_small_4xb256_fake5m_lr6e-4_10epoch/epoch_4.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:1b303804fd98cd9edc74bcd9a7836ed5daf018bd53b8be50b0e11d85924c3418 +size 793572512 diff --git a/convnext_small_4xb256_fake5m_lr6e-4_10epoch/epoch_5.pth b/convnext_small_4xb256_fake5m_lr6e-4_10epoch/epoch_5.pth new file mode 100644 index 0000000000000000000000000000000000000000..45ccd72ed123a6b8583991bf2c75a228473cb5f0 --- /dev/null +++ b/convnext_small_4xb256_fake5m_lr6e-4_10epoch/epoch_5.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:d0f5f8d2b8fb9e5f6c63832b4ab264ab01d4f1a5390d288c29d97b55d08c4950 +size 793987872 diff --git a/convnext_small_4xb256_fake5m_lr6e-4_10epoch/last_checkpoint b/convnext_small_4xb256_fake5m_lr6e-4_10epoch/last_checkpoint new file mode 100644 index 0000000000000000000000000000000000000000..c86fca4dfde292ce905a4dee2bdfdb23fa434b0b --- /dev/null +++ b/convnext_small_4xb256_fake5m_lr6e-4_10epoch/last_checkpoint @@ -0,0 +1 @@ +/mnt/petrelfs/luzeyu/workspace/fakebench/mmpretrain/workdir/convnext_small_4xb256_fake5m_lr6e-4_10epoch/epoch_5.pth \ No newline at end of file diff --git a/convnext_small_4xb256_fake5m_lr8e-4_10epoch/20230601_151933/20230601_151933.log b/convnext_small_4xb256_fake5m_lr8e-4_10epoch/20230601_151933/20230601_151933.log new file mode 100644 index 0000000000000000000000000000000000000000..adbc6a3f7009efcc3c96586f9b51af4ac26d75fc --- /dev/null +++ b/convnext_small_4xb256_fake5m_lr8e-4_10epoch/20230601_151933/20230601_151933.log @@ -0,0 +1,1931 @@ +2023/06/01 15:19:37 - mmengine - INFO - +------------------------------------------------------------ +System environment: + sys.platform: linux + Python: 3.10.9 (main, Mar 8 2023, 10:47:38) [GCC 11.2.0] + CUDA available: True + numpy_random_seed: 1978382300 + GPU 0,1,2,3: NVIDIA A100-SXM4-80GB + CUDA_HOME: /mnt/petrelfs/share/cuda-11.6 + NVCC: Cuda compilation tools, release 11.6, V11.6.124 + GCC: gcc (GCC) 4.8.5 20150623 (Red Hat 4.8.5-44) + PyTorch: 1.13.1 + PyTorch compiling details: PyTorch built with: + - GCC 9.3 + - C++ Version: 201402 + - Intel(R) oneAPI Math Kernel Library Version 2021.4-Product Build 20210904 for Intel(R) 64 architecture applications + - Intel(R) MKL-DNN v2.6.0 (Git Hash 52b5f107dd9cf10910aaa19cb47f3abf9b349815) + - OpenMP 201511 (a.k.a. OpenMP 4.5) + - LAPACK is enabled (usually provided by MKL) + - NNPACK is enabled + - CPU capability usage: AVX2 + - CUDA Runtime 11.6 + - NVCC architecture flags: -gencode;arch=compute_37,code=sm_37;-gencode;arch=compute_50,code=sm_50;-gencode;arch=compute_60,code=sm_60;-gencode;arch=compute_61,code=sm_61;-gencode;arch=compute_70,code=sm_70;-gencode;arch=compute_75,code=sm_75;-gencode;arch=compute_80,code=sm_80;-gencode;arch=compute_86,code=sm_86;-gencode;arch=compute_37,code=compute_37 + - CuDNN 8.3.2 (built against CUDA 11.5) + - Magma 2.6.1 + - Build settings: BLAS_INFO=mkl, BUILD_TYPE=Release, CUDA_VERSION=11.6, CUDNN_VERSION=8.3.2, CXX_COMPILER=/opt/rh/devtoolset-9/root/usr/bin/c++, CXX_FLAGS= -fabi-version=11 -Wno-deprecated -fvisibility-inlines-hidden -DUSE_PTHREADPOOL -fopenmp -DNDEBUG -DUSE_KINETO -DUSE_FBGEMM -DUSE_QNNPACK -DUSE_PYTORCH_QNNPACK -DUSE_XNNPACK -DSYMBOLICATE_MOBILE_DEBUG_HANDLE -DEDGE_PROFILER_USE_KINETO -O2 -fPIC -Wno-narrowing -Wall -Wextra -Werror=return-type -Werror=non-virtual-dtor -Wno-missing-field-initializers -Wno-type-limits -Wno-array-bounds -Wno-unknown-pragmas -Wunused-local-typedefs -Wno-unused-parameter -Wno-unused-function -Wno-unused-result -Wno-strict-overflow -Wno-strict-aliasing -Wno-error=deprecated-declarations -Wno-stringop-overflow -Wno-psabi -Wno-error=pedantic -Wno-error=redundant-decls -Wno-error=old-style-cast -fdiagnostics-color=always -faligned-new -Wno-unused-but-set-variable -Wno-maybe-uninitialized -fno-math-errno -fno-trapping-math -Werror=format -Werror=cast-function-type -Wno-stringop-overflow, LAPACK_INFO=mkl, PERF_WITH_AVX=1, PERF_WITH_AVX2=1, PERF_WITH_AVX512=1, TORCH_VERSION=1.13.1, USE_CUDA=ON, USE_CUDNN=ON, USE_EXCEPTION_PTR=1, USE_GFLAGS=OFF, USE_GLOG=OFF, USE_MKL=ON, USE_MKLDNN=ON, USE_MPI=OFF, USE_NCCL=ON, USE_NNPACK=ON, USE_OPENMP=ON, USE_ROCM=OFF, + + TorchVision: 0.14.1 + OpenCV: 4.7.0 + MMEngine: 0.7.3 + +Runtime environment: + cudnn_benchmark: True + mp_cfg: {'mp_start_method': 'fork', 'opencv_num_threads': 0} + dist_cfg: {'backend': 'nccl'} + seed: None + deterministic: False + Distributed launcher: slurm + Distributed training: True + GPU number: 4 +------------------------------------------------------------ + +2023/06/01 15:19:38 - mmengine - INFO - Config: +optim_wrapper = dict( + optimizer=dict( + type='AdamW', + lr=0.0008, + weight_decay=0.05, + eps=1e-08, + betas=(0.9, 0.999), + _scope_='mmpretrain'), + paramwise_cfg=dict( + norm_decay_mult=0.0, + bias_decay_mult=0.0, + flat_decay_mult=0.0, + custom_keys=dict({ + '.absolute_pos_embed': dict(decay_mult=0.0), + '.relative_position_bias_table': dict(decay_mult=0.0) + })), + type='AmpOptimWrapper', + dtype='bfloat16', + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=True, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=4096) +model = dict( + type='ImageClassifier', + backbone=dict(type='ConvNeXt', arch='small', drop_path_rate=0.4), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=768, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + init_cfg=None), + init_cfg=dict( + type='TruncNormal', layer=['Conv2d', 'Linear'], std=0.02, bias=0.0), + train_cfg=None) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stablediffusionV1-5R2-dpmsolver-25-5m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/convnext_small_4xb256_fake5m_lr8e-4_10epoch' + +2023/06/01 15:19:42 - mmengine - INFO - Hooks will be executed in the following order: +before_run: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_load_checkpoint: +(ABOVE_NORMAL) EMAHook + -------------------- +before_train: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_train_epoch: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(NORMAL ) DistSamplerSeedHook + -------------------- +before_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook + -------------------- +after_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +after_train_epoch: +(NORMAL ) IterTimerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_val_epoch: +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook + -------------------- +before_val_iter: +(NORMAL ) IterTimerHook + -------------------- +after_val_iter: +(NORMAL ) IterTimerHook +(NORMAL ) VisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_val_epoch: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_save_checkpoint: +(ABOVE_NORMAL) EMAHook + -------------------- +after_train: +(VERY_LOW ) CheckpointHook + -------------------- +before_test_epoch: +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook + -------------------- +before_test_iter: +(NORMAL ) IterTimerHook + -------------------- +after_test_iter: +(NORMAL ) IterTimerHook +(NORMAL ) VisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_test_epoch: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_run: +(BELOW_NORMAL) LoggerHook + -------------------- +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.0.0.bias:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.0.1.weight:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.0.1.bias:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.1.0.weight:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.1.0.bias:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.1.1.bias:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.2.0.weight:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.2.0.bias:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.2.1.bias:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.3.0.weight:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.3.0.bias:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.3.1.bias:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.0.0.gamma:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.0.0.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.0.0.norm.weight:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.0.0.norm.bias:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.0.0.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.0.0.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.0.1.gamma:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.0.1.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.0.1.norm.weight:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.0.1.norm.bias:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.0.1.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.0.1.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.0.2.gamma:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.0.2.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.0.2.norm.weight:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.0.2.norm.bias:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.0.2.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.0.2.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.1.0.gamma:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.1.0.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.1.0.norm.weight:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.1.0.norm.bias:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.1.0.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.1.0.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.1.1.gamma:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.1.1.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.1.1.norm.weight:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.1.1.norm.bias:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.1.1.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.1.1.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.1.2.gamma:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.1.2.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.1.2.norm.weight:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.1.2.norm.bias:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.1.2.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.1.2.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.2.0.gamma:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.2.0.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.2.0.norm.weight:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.2.0.norm.bias:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.2.0.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.2.0.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.2.1.gamma:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.2.1.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.2.1.norm.weight:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.2.1.norm.bias:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.2.1.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.2.1.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.2.2.gamma:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.2.2.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.2.2.norm.weight:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.2.2.norm.bias:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.2.2.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.2.2.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.2.3.gamma:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.2.3.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.2.3.norm.weight:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.2.3.norm.bias:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.2.3.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.2.3.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.2.4.gamma:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.2.4.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.2.4.norm.weight:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.2.4.norm.bias:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.2.4.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.2.4.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.2.5.gamma:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.2.5.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.2.5.norm.weight:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.2.5.norm.bias:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.2.5.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.2.5.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.2.6.gamma:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.2.6.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.2.6.norm.weight:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.2.6.norm.bias:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.2.6.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.2.6.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.2.7.gamma:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.2.7.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.2.7.norm.weight:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.2.7.norm.bias:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.2.7.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.2.7.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.2.8.gamma:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.2.8.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.2.8.norm.weight:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.2.8.norm.bias:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.2.8.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.2.8.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.2.9.gamma:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.2.9.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.2.9.norm.weight:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.2.9.norm.bias:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.2.9.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.2.9.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.2.10.gamma:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.2.10.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.2.10.norm.weight:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.2.10.norm.bias:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.2.10.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.2.10.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.2.11.gamma:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.2.11.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.2.11.norm.weight:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.2.11.norm.bias:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.2.11.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.2.11.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.2.12.gamma:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.2.12.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.2.12.norm.weight:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.2.12.norm.bias:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.2.12.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.2.12.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.2.13.gamma:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.2.13.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.2.13.norm.weight:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.2.13.norm.bias:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.2.13.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.2.13.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.2.14.gamma:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.2.14.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.2.14.norm.weight:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.2.14.norm.bias:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.2.14.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.2.14.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.2.15.gamma:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.2.15.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.2.15.norm.weight:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.2.15.norm.bias:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.2.15.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.2.15.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.2.16.gamma:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.2.16.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.2.16.norm.weight:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.2.16.norm.bias:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.2.16.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.2.16.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.2.17.gamma:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.2.17.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.2.17.norm.weight:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.2.17.norm.bias:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.2.17.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.2.17.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.2.18.gamma:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.2.18.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.2.18.norm.weight:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.2.18.norm.bias:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.2.18.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.2.18.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.2.19.gamma:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.2.19.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.2.19.norm.weight:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.2.19.norm.bias:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.2.19.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.2.19.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.2.20.gamma:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.2.20.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.2.20.norm.weight:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.2.20.norm.bias:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.2.20.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.2.20.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.2.21.gamma:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.2.21.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.2.21.norm.weight:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.2.21.norm.bias:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.2.21.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.2.21.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.2.22.gamma:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.2.22.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.2.22.norm.weight:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.2.22.norm.bias:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.2.22.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.2.22.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.2.23.gamma:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.2.23.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.2.23.norm.weight:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.2.23.norm.bias:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.2.23.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.2.23.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.2.24.gamma:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.2.24.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.2.24.norm.weight:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.2.24.norm.bias:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.2.24.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.2.24.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.2.25.gamma:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.2.25.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.2.25.norm.weight:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.2.25.norm.bias:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.2.25.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.2.25.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.2.26.gamma:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.2.26.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.2.26.norm.weight:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.2.26.norm.bias:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.2.26.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.2.26.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.3.0.gamma:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.3.0.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 15:20:09 - mmengine - INFO - paramwise_options -- backbone.stages.3.0.norm.weight:weight_decay=0.0 +2023/06/01 15:20:10 - mmengine - INFO - paramwise_options -- backbone.stages.3.0.norm.bias:weight_decay=0.0 +2023/06/01 15:20:10 - mmengine - INFO - paramwise_options -- backbone.stages.3.0.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 15:20:10 - mmengine - INFO - paramwise_options -- backbone.stages.3.0.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 15:20:10 - mmengine - INFO - paramwise_options -- backbone.stages.3.1.gamma:weight_decay=0.0 +2023/06/01 15:20:10 - mmengine - INFO - paramwise_options -- backbone.stages.3.1.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 15:20:10 - mmengine - INFO - paramwise_options -- backbone.stages.3.1.norm.weight:weight_decay=0.0 +2023/06/01 15:20:10 - mmengine - INFO - paramwise_options -- backbone.stages.3.1.norm.bias:weight_decay=0.0 +2023/06/01 15:20:10 - mmengine - INFO - paramwise_options -- backbone.stages.3.1.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 15:20:10 - mmengine - INFO - paramwise_options -- backbone.stages.3.1.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 15:20:10 - mmengine - INFO - paramwise_options -- backbone.stages.3.2.gamma:weight_decay=0.0 +2023/06/01 15:20:10 - mmengine - INFO - paramwise_options -- backbone.stages.3.2.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 15:20:10 - mmengine - INFO - paramwise_options -- backbone.stages.3.2.norm.weight:weight_decay=0.0 +2023/06/01 15:20:10 - mmengine - INFO - paramwise_options -- backbone.stages.3.2.norm.bias:weight_decay=0.0 +2023/06/01 15:20:10 - mmengine - INFO - paramwise_options -- backbone.stages.3.2.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 15:20:10 - mmengine - INFO - paramwise_options -- backbone.stages.3.2.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 15:20:10 - mmengine - INFO - paramwise_options -- backbone.norm3.weight:weight_decay=0.0 +2023/06/01 15:20:10 - mmengine - INFO - paramwise_options -- backbone.norm3.bias:weight_decay=0.0 +2023/06/01 15:20:10 - mmengine - INFO - paramwise_options -- head.fc.bias:weight_decay=0.0 +Name of parameter - Initialization information + +backbone.downsample_layers.0.0.weight - torch.Size([96, 3, 4, 4]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.downsample_layers.0.0.bias - torch.Size([96]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.downsample_layers.0.1.weight - torch.Size([96]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.downsample_layers.0.1.bias - torch.Size([96]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.downsample_layers.1.0.weight - torch.Size([96]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.downsample_layers.1.0.bias - torch.Size([96]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.downsample_layers.1.1.weight - torch.Size([192, 96, 2, 2]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.downsample_layers.1.1.bias - torch.Size([192]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.downsample_layers.2.0.weight - torch.Size([192]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.downsample_layers.2.0.bias - torch.Size([192]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.downsample_layers.2.1.weight - torch.Size([384, 192, 2, 2]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.downsample_layers.2.1.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.downsample_layers.3.0.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.downsample_layers.3.0.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.downsample_layers.3.1.weight - torch.Size([768, 384, 2, 2]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.downsample_layers.3.1.bias - torch.Size([768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.0.gamma - torch.Size([96]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.0.0.depthwise_conv.weight - torch.Size([96, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.0.depthwise_conv.bias - torch.Size([96]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.0.norm.weight - torch.Size([96]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.0.0.norm.bias - torch.Size([96]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.0.0.pointwise_conv1.weight - torch.Size([384, 96]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.0.pointwise_conv1.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.0.pointwise_conv2.weight - torch.Size([96, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.0.pointwise_conv2.bias - torch.Size([96]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.1.gamma - torch.Size([96]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.0.1.depthwise_conv.weight - torch.Size([96, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.1.depthwise_conv.bias - torch.Size([96]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.1.norm.weight - torch.Size([96]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.0.1.norm.bias - torch.Size([96]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.0.1.pointwise_conv1.weight - torch.Size([384, 96]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.1.pointwise_conv1.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.1.pointwise_conv2.weight - torch.Size([96, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.1.pointwise_conv2.bias - torch.Size([96]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.2.gamma - torch.Size([96]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.0.2.depthwise_conv.weight - torch.Size([96, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.2.depthwise_conv.bias - torch.Size([96]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.2.norm.weight - torch.Size([96]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.0.2.norm.bias - torch.Size([96]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.0.2.pointwise_conv1.weight - torch.Size([384, 96]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.2.pointwise_conv1.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.2.pointwise_conv2.weight - torch.Size([96, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.2.pointwise_conv2.bias - torch.Size([96]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.0.gamma - torch.Size([192]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.1.0.depthwise_conv.weight - torch.Size([192, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.0.depthwise_conv.bias - torch.Size([192]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.0.norm.weight - torch.Size([192]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.1.0.norm.bias - torch.Size([192]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.1.0.pointwise_conv1.weight - torch.Size([768, 192]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.0.pointwise_conv1.bias - torch.Size([768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.0.pointwise_conv2.weight - torch.Size([192, 768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.0.pointwise_conv2.bias - torch.Size([192]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.1.gamma - torch.Size([192]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.1.1.depthwise_conv.weight - torch.Size([192, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.1.depthwise_conv.bias - torch.Size([192]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.1.norm.weight - torch.Size([192]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.1.1.norm.bias - torch.Size([192]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.1.1.pointwise_conv1.weight - torch.Size([768, 192]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.1.pointwise_conv1.bias - torch.Size([768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.1.pointwise_conv2.weight - torch.Size([192, 768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.1.pointwise_conv2.bias - torch.Size([192]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.2.gamma - torch.Size([192]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.1.2.depthwise_conv.weight - torch.Size([192, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.2.depthwise_conv.bias - torch.Size([192]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.2.norm.weight - torch.Size([192]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.1.2.norm.bias - torch.Size([192]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.1.2.pointwise_conv1.weight - torch.Size([768, 192]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.2.pointwise_conv1.bias - torch.Size([768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.2.pointwise_conv2.weight - torch.Size([192, 768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.2.pointwise_conv2.bias - torch.Size([192]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.0.gamma - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.0.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.0.depthwise_conv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.0.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.0.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.0.pointwise_conv1.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.0.pointwise_conv1.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.0.pointwise_conv2.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.0.pointwise_conv2.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.1.gamma - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.1.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.1.depthwise_conv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.1.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.1.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.1.pointwise_conv1.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.1.pointwise_conv1.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.1.pointwise_conv2.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.1.pointwise_conv2.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.2.gamma - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.2.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.2.depthwise_conv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.2.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.2.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.2.pointwise_conv1.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.2.pointwise_conv1.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.2.pointwise_conv2.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.2.pointwise_conv2.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.3.gamma - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.3.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.3.depthwise_conv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.3.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.3.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.3.pointwise_conv1.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.3.pointwise_conv1.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.3.pointwise_conv2.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.3.pointwise_conv2.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.4.gamma - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.4.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.4.depthwise_conv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.4.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.4.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.4.pointwise_conv1.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.4.pointwise_conv1.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.4.pointwise_conv2.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.4.pointwise_conv2.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.5.gamma - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.5.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.5.depthwise_conv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.5.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.5.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.5.pointwise_conv1.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.5.pointwise_conv1.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.5.pointwise_conv2.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.5.pointwise_conv2.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.6.gamma - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.6.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.6.depthwise_conv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.6.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.6.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.6.pointwise_conv1.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.6.pointwise_conv1.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.6.pointwise_conv2.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.6.pointwise_conv2.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.7.gamma - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.7.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.7.depthwise_conv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.7.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.7.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.7.pointwise_conv1.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.7.pointwise_conv1.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.7.pointwise_conv2.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.7.pointwise_conv2.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.8.gamma - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.8.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.8.depthwise_conv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.8.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.8.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.8.pointwise_conv1.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.8.pointwise_conv1.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.8.pointwise_conv2.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.8.pointwise_conv2.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.9.gamma - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.9.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.9.depthwise_conv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.9.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.9.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.9.pointwise_conv1.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.9.pointwise_conv1.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.9.pointwise_conv2.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.9.pointwise_conv2.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.10.gamma - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.10.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.10.depthwise_conv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.10.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.10.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.10.pointwise_conv1.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.10.pointwise_conv1.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.10.pointwise_conv2.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.10.pointwise_conv2.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.11.gamma - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.11.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.11.depthwise_conv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.11.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.11.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.11.pointwise_conv1.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.11.pointwise_conv1.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.11.pointwise_conv2.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.11.pointwise_conv2.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.12.gamma - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.12.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.12.depthwise_conv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.12.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.12.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.12.pointwise_conv1.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.12.pointwise_conv1.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.12.pointwise_conv2.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.12.pointwise_conv2.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.13.gamma - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.13.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.13.depthwise_conv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.13.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.13.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.13.pointwise_conv1.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.13.pointwise_conv1.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.13.pointwise_conv2.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.13.pointwise_conv2.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.14.gamma - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.14.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.14.depthwise_conv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.14.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.14.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.14.pointwise_conv1.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.14.pointwise_conv1.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.14.pointwise_conv2.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.14.pointwise_conv2.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.15.gamma - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.15.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.15.depthwise_conv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.15.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.15.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.15.pointwise_conv1.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.15.pointwise_conv1.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.15.pointwise_conv2.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.15.pointwise_conv2.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.16.gamma - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.16.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.16.depthwise_conv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.16.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.16.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.16.pointwise_conv1.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.16.pointwise_conv1.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.16.pointwise_conv2.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.16.pointwise_conv2.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.17.gamma - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.17.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.17.depthwise_conv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.17.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.17.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.17.pointwise_conv1.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.17.pointwise_conv1.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.17.pointwise_conv2.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.17.pointwise_conv2.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.18.gamma - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.18.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.18.depthwise_conv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.18.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.18.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.18.pointwise_conv1.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.18.pointwise_conv1.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.18.pointwise_conv2.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.18.pointwise_conv2.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.19.gamma - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.19.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.19.depthwise_conv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.19.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.19.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.19.pointwise_conv1.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.19.pointwise_conv1.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.19.pointwise_conv2.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.19.pointwise_conv2.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.20.gamma - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.20.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.20.depthwise_conv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.20.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.20.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.20.pointwise_conv1.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.20.pointwise_conv1.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.20.pointwise_conv2.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.20.pointwise_conv2.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.21.gamma - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.21.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.21.depthwise_conv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.21.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.21.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.21.pointwise_conv1.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.21.pointwise_conv1.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.21.pointwise_conv2.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.21.pointwise_conv2.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.22.gamma - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.22.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.22.depthwise_conv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.22.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.22.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.22.pointwise_conv1.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.22.pointwise_conv1.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.22.pointwise_conv2.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.22.pointwise_conv2.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.23.gamma - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.23.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.23.depthwise_conv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.23.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.23.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.23.pointwise_conv1.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.23.pointwise_conv1.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.23.pointwise_conv2.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.23.pointwise_conv2.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.24.gamma - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.24.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.24.depthwise_conv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.24.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.24.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.24.pointwise_conv1.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.24.pointwise_conv1.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.24.pointwise_conv2.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.24.pointwise_conv2.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.25.gamma - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.25.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.25.depthwise_conv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.25.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.25.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.25.pointwise_conv1.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.25.pointwise_conv1.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.25.pointwise_conv2.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.25.pointwise_conv2.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.26.gamma - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.26.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.26.depthwise_conv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.26.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.26.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.26.pointwise_conv1.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.26.pointwise_conv1.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.26.pointwise_conv2.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.26.pointwise_conv2.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.0.gamma - torch.Size([768]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.3.0.depthwise_conv.weight - torch.Size([768, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.0.depthwise_conv.bias - torch.Size([768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.0.norm.weight - torch.Size([768]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.3.0.norm.bias - torch.Size([768]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.3.0.pointwise_conv1.weight - torch.Size([3072, 768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.0.pointwise_conv1.bias - torch.Size([3072]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.0.pointwise_conv2.weight - torch.Size([768, 3072]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.0.pointwise_conv2.bias - torch.Size([768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.1.gamma - torch.Size([768]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.3.1.depthwise_conv.weight - torch.Size([768, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.1.depthwise_conv.bias - torch.Size([768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.1.norm.weight - torch.Size([768]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.3.1.norm.bias - torch.Size([768]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.3.1.pointwise_conv1.weight - torch.Size([3072, 768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.1.pointwise_conv1.bias - torch.Size([3072]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.1.pointwise_conv2.weight - torch.Size([768, 3072]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.1.pointwise_conv2.bias - torch.Size([768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.2.gamma - torch.Size([768]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.3.2.depthwise_conv.weight - torch.Size([768, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.2.depthwise_conv.bias - torch.Size([768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.2.norm.weight - torch.Size([768]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.3.2.norm.bias - torch.Size([768]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.3.2.pointwise_conv1.weight - torch.Size([3072, 768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.2.pointwise_conv1.bias - torch.Size([3072]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.2.pointwise_conv2.weight - torch.Size([768, 3072]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.2.pointwise_conv2.bias - torch.Size([768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.norm3.weight - torch.Size([768]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.norm3.bias - torch.Size([768]): +The value is the same before and after calling `init_weights` of ImageClassifier + +head.fc.weight - torch.Size([2, 768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +head.fc.bias - torch.Size([2]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 +2023/06/01 15:20:10 - mmengine - WARNING - "FileClient" will be deprecated in future. Please use io functions in https://mmengine.readthedocs.io/en/latest/api/fileio.html#file-io +2023/06/01 15:20:10 - mmengine - WARNING - "HardDiskBackend" is the alias of "LocalBackend" and the former will be deprecated in future. +2023/06/01 15:20:10 - mmengine - INFO - Checkpoints will be saved to /mnt/petrelfs/luzeyu/workspace/fakebench/mmpretrain/workdir/convnext_small_4xb256_fake5m_lr8e-4_10epoch. +2023/06/01 15:21:43 - mmengine - INFO - Epoch(train) [1][ 100/5758] lr: 8.0000e-04 eta: 14:44:08 time: 0.7506 data_time: 0.0014 memory: 25071 loss: 0.6051 +2023/06/01 15:23:02 - mmengine - INFO - Epoch(train) [1][ 200/5758] lr: 8.0000e-04 eta: 13:40:24 time: 0.7945 data_time: 0.0014 memory: 25071 loss: 0.5874 +2023/06/01 15:24:20 - mmengine - INFO - Epoch(train) [1][ 300/5758] lr: 8.0000e-04 eta: 13:14:27 time: 0.7569 data_time: 0.0013 memory: 25071 loss: 0.5443 +2023/06/01 15:25:36 - mmengine - INFO - Epoch(train) [1][ 400/5758] lr: 8.0000e-04 eta: 12:56:10 time: 0.7552 data_time: 0.0013 memory: 25071 loss: 0.5240 +2023/06/01 15:26:54 - mmengine - INFO - Epoch(train) [1][ 500/5758] lr: 8.0000e-04 eta: 12:47:16 time: 0.8014 data_time: 0.0946 memory: 25071 loss: 0.4859 +2023/06/01 15:28:14 - mmengine - INFO - Epoch(train) [1][ 600/5758] lr: 8.0000e-04 eta: 12:45:34 time: 0.8219 data_time: 0.1175 memory: 25071 loss: 0.4861 +2023/06/01 15:29:36 - mmengine - INFO - Epoch(train) [1][ 700/5758] lr: 8.0000e-04 eta: 12:46:05 time: 0.8456 data_time: 0.1403 memory: 25071 loss: 0.4459 +2023/06/01 15:30:54 - mmengine - INFO - Epoch(train) [1][ 800/5758] lr: 8.0000e-04 eta: 12:41:14 time: 0.7260 data_time: 0.0016 memory: 25071 loss: 0.4626 +2023/06/01 15:32:13 - mmengine - INFO - Epoch(train) [1][ 900/5758] lr: 8.0000e-04 eta: 12:38:35 time: 0.7729 data_time: 0.0016 memory: 25071 loss: 0.4264 +2023/06/01 15:33:35 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_lr8e-4_10epoch_20230601_151933 +2023/06/01 15:33:35 - mmengine - INFO - Epoch(train) [1][1000/5758] lr: 8.0000e-04 eta: 12:38:47 time: 0.8703 data_time: 0.1511 memory: 25071 loss: 0.4033 +2023/06/01 15:34:52 - mmengine - INFO - Epoch(train) [1][1100/5758] lr: 8.0000e-04 eta: 12:34:38 time: 0.7977 data_time: 0.0891 memory: 25071 loss: 0.4077 +2023/06/01 15:36:13 - mmengine - INFO - Epoch(train) [1][1200/5758] lr: 8.0000e-04 eta: 12:33:26 time: 0.7865 data_time: 0.0679 memory: 25071 loss: 0.3849 +2023/06/01 15:37:33 - mmengine - INFO - Epoch(train) [1][1300/5758] lr: 8.0000e-04 eta: 12:32:38 time: 0.7082 data_time: 0.0017 memory: 25071 loss: 0.4117 +2023/06/01 15:38:47 - mmengine - INFO - Epoch(train) [1][1400/5758] lr: 8.0000e-04 eta: 12:27:02 time: 0.7876 data_time: 0.0656 memory: 25071 loss: 0.3786 +2023/06/01 15:40:06 - mmengine - INFO - Epoch(train) [1][1500/5758] lr: 8.0000e-04 eta: 12:24:54 time: 0.7545 data_time: 0.0498 memory: 25071 loss: 0.3571 +2023/06/01 15:41:22 - mmengine - INFO - Epoch(train) [1][1600/5758] lr: 8.0000e-04 eta: 12:21:23 time: 0.7991 data_time: 0.0770 memory: 25071 loss: 0.3465 +2023/06/01 15:42:39 - mmengine - INFO - Epoch(train) [1][1700/5758] lr: 8.0000e-04 eta: 12:19:06 time: 0.7818 data_time: 0.0023 memory: 25071 loss: 0.3371 +2023/06/01 15:43:56 - mmengine - INFO - Epoch(train) [1][1800/5758] lr: 8.0000e-04 eta: 12:16:14 time: 0.7372 data_time: 0.0143 memory: 25071 loss: 0.3621 +2023/06/01 15:45:13 - mmengine - INFO - Epoch(train) [1][1900/5758] lr: 8.0000e-04 eta: 12:14:07 time: 0.7871 data_time: 0.0021 memory: 25071 loss: 0.3515 +2023/06/01 15:46:30 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_lr8e-4_10epoch_20230601_151933 +2023/06/01 15:46:30 - mmengine - INFO - Epoch(train) [1][2000/5758] lr: 8.0000e-04 eta: 12:11:40 time: 0.7590 data_time: 0.0019 memory: 25071 loss: 0.3616 +2023/06/01 15:47:45 - mmengine - INFO - Epoch(train) [1][2100/5758] lr: 8.0000e-04 eta: 12:08:43 time: 0.7730 data_time: 0.0013 memory: 25071 loss: 0.3253 +2023/06/01 15:49:00 - mmengine - INFO - Epoch(train) [1][2200/5758] lr: 8.0000e-04 eta: 12:05:44 time: 0.7480 data_time: 0.0018 memory: 25071 loss: 0.3194 +2023/06/01 15:50:15 - mmengine - INFO - Epoch(train) [1][2300/5758] lr: 8.0000e-04 eta: 12:03:04 time: 0.7312 data_time: 0.0015 memory: 25071 loss: 0.3168 +2023/06/01 15:51:30 - mmengine - INFO - Epoch(train) [1][2400/5758] lr: 8.0000e-04 eta: 12:00:16 time: 0.7680 data_time: 0.0021 memory: 25071 loss: 0.3000 +2023/06/01 15:52:46 - mmengine - INFO - Epoch(train) [1][2500/5758] lr: 8.0000e-04 eta: 11:58:11 time: 0.7476 data_time: 0.0015 memory: 25071 loss: 0.3226 +2023/06/01 15:54:00 - mmengine - INFO - Epoch(train) [1][2600/5758] lr: 8.0000e-04 eta: 11:55:28 time: 0.7302 data_time: 0.0014 memory: 25071 loss: 0.3157 +2023/06/01 15:55:16 - mmengine - INFO - Epoch(train) [1][2700/5758] lr: 8.0000e-04 eta: 11:53:09 time: 0.7981 data_time: 0.0014 memory: 25071 loss: 0.2700 +2023/06/01 15:56:31 - mmengine - INFO - Epoch(train) [1][2800/5758] lr: 8.0000e-04 eta: 11:50:57 time: 0.7392 data_time: 0.0013 memory: 25071 loss: 0.2793 +2023/06/01 15:57:47 - mmengine - INFO - Epoch(train) [1][2900/5758] lr: 8.0000e-04 eta: 11:49:10 time: 0.7565 data_time: 0.0019 memory: 25071 loss: 0.2973 +2023/06/01 15:59:05 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_lr8e-4_10epoch_20230601_151933 +2023/06/01 15:59:05 - mmengine - INFO - Epoch(train) [1][3000/5758] lr: 8.0000e-04 eta: 11:47:54 time: 0.7988 data_time: 0.0014 memory: 25071 loss: 0.2973 +2023/06/01 16:00:23 - mmengine - INFO - Epoch(train) [1][3100/5758] lr: 8.0000e-04 eta: 11:46:42 time: 0.7925 data_time: 0.0014 memory: 25071 loss: 0.2567 +2023/06/01 16:01:39 - mmengine - INFO - Epoch(train) [1][3200/5758] lr: 8.0000e-04 eta: 11:44:44 time: 0.7815 data_time: 0.0017 memory: 25071 loss: 0.2280 +2023/06/01 16:02:56 - mmengine - INFO - Epoch(train) [1][3300/5758] lr: 8.0000e-04 eta: 11:43:14 time: 0.7388 data_time: 0.0013 memory: 25071 loss: 0.2463 +2023/06/01 16:04:12 - mmengine - INFO - Epoch(train) [1][3400/5758] lr: 8.0000e-04 eta: 11:41:39 time: 0.7859 data_time: 0.0804 memory: 25071 loss: 0.2533 +2023/06/01 16:05:30 - mmengine - INFO - Epoch(train) [1][3500/5758] lr: 8.0000e-04 eta: 11:40:22 time: 0.8053 data_time: 0.0992 memory: 25071 loss: 0.2439 +2023/06/01 16:06:46 - mmengine - INFO - Epoch(train) [1][3600/5758] lr: 8.0000e-04 eta: 11:38:40 time: 0.7468 data_time: 0.0412 memory: 25071 loss: 0.2317 +2023/06/01 16:08:02 - mmengine - INFO - Epoch(train) [1][3700/5758] lr: 8.0000e-04 eta: 11:36:49 time: 0.7561 data_time: 0.0494 memory: 25071 loss: 0.2219 +2023/06/01 16:09:18 - mmengine - INFO - Epoch(train) [1][3800/5758] lr: 8.0000e-04 eta: 11:35:11 time: 0.7989 data_time: 0.0016 memory: 25071 loss: 0.2318 +2023/06/01 16:10:34 - mmengine - INFO - Epoch(train) [1][3900/5758] lr: 8.0000e-04 eta: 11:33:38 time: 0.7792 data_time: 0.0018 memory: 25071 loss: 0.2203 +2023/06/01 16:11:49 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_lr8e-4_10epoch_20230601_151933 +2023/06/01 16:11:49 - mmengine - INFO - Epoch(train) [1][4000/5758] lr: 8.0000e-04 eta: 11:31:51 time: 0.7416 data_time: 0.0014 memory: 25071 loss: 0.2366 +2023/06/01 16:13:05 - mmengine - INFO - Epoch(train) [1][4100/5758] lr: 8.0000e-04 eta: 11:30:13 time: 0.7655 data_time: 0.0054 memory: 25071 loss: 0.2334 +2023/06/01 16:14:21 - mmengine - INFO - Epoch(train) [1][4200/5758] lr: 8.0000e-04 eta: 11:28:29 time: 0.7289 data_time: 0.0012 memory: 25071 loss: 0.1918 +2023/06/01 16:15:38 - mmengine - INFO - Epoch(train) [1][4300/5758] lr: 8.0000e-04 eta: 11:27:11 time: 0.7665 data_time: 0.0531 memory: 25071 loss: 0.2151 +2023/06/01 16:16:55 - mmengine - INFO - Epoch(train) [1][4400/5758] lr: 8.0000e-04 eta: 11:25:51 time: 0.7700 data_time: 0.0014 memory: 25071 loss: 0.2013 +2023/06/01 16:18:11 - mmengine - INFO - Epoch(train) [1][4500/5758] lr: 8.0000e-04 eta: 11:24:15 time: 0.7384 data_time: 0.0282 memory: 25071 loss: 0.1992 +2023/06/01 16:19:28 - mmengine - INFO - Epoch(train) [1][4600/5758] lr: 8.0000e-04 eta: 11:22:52 time: 0.7599 data_time: 0.0518 memory: 25071 loss: 0.1944 +2023/06/01 16:20:44 - mmengine - INFO - Epoch(train) [1][4700/5758] lr: 8.0000e-04 eta: 11:21:17 time: 0.7711 data_time: 0.0487 memory: 25071 loss: 0.2022 +2023/06/01 16:21:59 - mmengine - INFO - Epoch(train) [1][4800/5758] lr: 8.0000e-04 eta: 11:19:33 time: 0.7907 data_time: 0.0859 memory: 25071 loss: 0.1862 +2023/06/01 16:23:16 - mmengine - INFO - Epoch(train) [1][4900/5758] lr: 8.0000e-04 eta: 11:18:13 time: 0.7507 data_time: 0.0014 memory: 25071 loss: 0.1750 +2023/06/01 16:24:30 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_lr8e-4_10epoch_20230601_151933 +2023/06/01 16:24:30 - mmengine - INFO - Epoch(train) [1][5000/5758] lr: 8.0000e-04 eta: 11:16:22 time: 0.7253 data_time: 0.0190 memory: 25071 loss: 0.1673 +2023/06/01 16:25:44 - mmengine - INFO - Epoch(train) [1][5100/5758] lr: 8.0000e-04 eta: 11:14:40 time: 0.7871 data_time: 0.0087 memory: 25071 loss: 0.1804 +2023/06/01 16:27:01 - mmengine - INFO - Epoch(train) [1][5200/5758] lr: 8.0000e-04 eta: 11:13:13 time: 0.7414 data_time: 0.0014 memory: 25071 loss: 0.1654 +2023/06/01 16:28:19 - mmengine - INFO - Epoch(train) [1][5300/5758] lr: 8.0000e-04 eta: 11:12:09 time: 0.7748 data_time: 0.0014 memory: 25071 loss: 0.1610 +2023/06/01 16:29:37 - mmengine - INFO - Epoch(train) [1][5400/5758] lr: 8.0000e-04 eta: 11:10:56 time: 0.8160 data_time: 0.1113 memory: 25071 loss: 0.1691 +2023/06/01 16:30:54 - mmengine - INFO - Epoch(train) [1][5500/5758] lr: 8.0000e-04 eta: 11:09:43 time: 0.7451 data_time: 0.0311 memory: 25071 loss: 0.1555 +2023/06/01 16:32:09 - mmengine - INFO - Epoch(train) [1][5600/5758] lr: 8.0000e-04 eta: 11:08:05 time: 0.7679 data_time: 0.0627 memory: 25071 loss: 0.1784 +2023/06/01 16:33:26 - mmengine - INFO - Epoch(train) [1][5700/5758] lr: 8.0000e-04 eta: 11:06:43 time: 0.7157 data_time: 0.0016 memory: 25071 loss: 0.1679 +2023/06/01 16:34:11 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_lr8e-4_10epoch_20230601_151933 +2023/06/01 16:34:11 - mmengine - INFO - Saving checkpoint at 1 epochs +2023/06/01 16:34:32 - mmengine - INFO - Epoch(val) [1][16/16] accuracy/top1: 86.1183 single-label/precision_classwise: [100.0, 0.0] single-label/recall_classwise: [86.11830139160156, 0.0] single-label/f1-score_classwise: [92.54146575927734, 0.0] data_time: 0.2746 time: 0.7907 +2023/06/01 16:35:55 - mmengine - INFO - Epoch(train) [2][ 100/5758] lr: 7.8067e-04 eta: 11:05:41 time: 0.7470 data_time: 0.0013 memory: 25074 loss: 0.1396 +2023/06/01 16:37:13 - mmengine - INFO - Epoch(train) [2][ 200/5758] lr: 7.8067e-04 eta: 11:04:25 time: 0.7680 data_time: 0.0015 memory: 25074 loss: 0.1422 +2023/06/01 16:37:45 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_lr8e-4_10epoch_20230601_151933 +2023/06/01 16:38:31 - mmengine - INFO - Epoch(train) [2][ 300/5758] lr: 7.8067e-04 eta: 11:03:20 time: 0.8220 data_time: 0.1153 memory: 25074 loss: 0.1474 +2023/06/01 16:39:49 - mmengine - INFO - Epoch(train) [2][ 400/5758] lr: 7.8067e-04 eta: 11:02:08 time: 0.7834 data_time: 0.0786 memory: 25074 loss: 0.1410 +2023/06/01 16:41:08 - mmengine - INFO - Epoch(train) [2][ 500/5758] lr: 7.8067e-04 eta: 11:01:04 time: 0.7688 data_time: 0.0511 memory: 25074 loss: 0.1275 +2023/06/01 16:42:26 - mmengine - INFO - Epoch(train) [2][ 600/5758] lr: 7.8067e-04 eta: 10:59:48 time: 0.7567 data_time: 0.0519 memory: 25074 loss: 0.1334 +2023/06/01 16:43:44 - mmengine - INFO - Epoch(train) [2][ 700/5758] lr: 7.8067e-04 eta: 10:58:41 time: 0.8402 data_time: 0.1231 memory: 25074 loss: 0.1633 +2023/06/01 16:45:01 - mmengine - INFO - Epoch(train) [2][ 800/5758] lr: 7.8067e-04 eta: 10:57:19 time: 0.7651 data_time: 0.0600 memory: 25074 loss: 0.1396 +2023/06/01 16:46:18 - mmengine - INFO - Epoch(train) [2][ 900/5758] lr: 7.8067e-04 eta: 10:55:57 time: 0.7572 data_time: 0.0315 memory: 25074 loss: 0.1194 +2023/06/01 16:47:32 - mmengine - INFO - Epoch(train) [2][1000/5758] lr: 7.8067e-04 eta: 10:54:22 time: 0.7635 data_time: 0.0570 memory: 25074 loss: 0.1214 +2023/06/01 16:48:50 - mmengine - INFO - Epoch(train) [2][1100/5758] lr: 7.8067e-04 eta: 10:53:04 time: 0.8109 data_time: 0.0870 memory: 25074 loss: 0.1099 +2023/06/01 16:50:07 - mmengine - INFO - Epoch(train) [2][1200/5758] lr: 7.8067e-04 eta: 10:51:49 time: 0.7557 data_time: 0.0502 memory: 25074 loss: 0.1467 +2023/06/01 16:50:41 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_lr8e-4_10epoch_20230601_151933 +2023/06/01 16:51:26 - mmengine - INFO - Epoch(train) [2][1300/5758] lr: 7.8067e-04 eta: 10:50:39 time: 0.7816 data_time: 0.0497 memory: 25074 loss: 0.1172 +2023/06/01 16:52:42 - mmengine - INFO - Epoch(train) [2][1400/5758] lr: 7.8067e-04 eta: 10:49:13 time: 0.7602 data_time: 0.0514 memory: 25074 loss: 0.1109 +2023/06/01 16:53:59 - mmengine - INFO - Epoch(train) [2][1500/5758] lr: 7.8067e-04 eta: 10:47:57 time: 0.8090 data_time: 0.0885 memory: 25074 loss: 0.1281 +2023/06/01 16:55:17 - mmengine - INFO - Epoch(train) [2][1600/5758] lr: 7.8067e-04 eta: 10:46:42 time: 0.7568 data_time: 0.0515 memory: 25074 loss: 0.0891 +2023/06/01 16:56:35 - mmengine - INFO - Epoch(train) [2][1700/5758] lr: 7.8067e-04 eta: 10:45:31 time: 0.8123 data_time: 0.0896 memory: 25074 loss: 0.1266 +2023/06/01 16:57:51 - mmengine - INFO - Epoch(train) [2][1800/5758] lr: 7.8067e-04 eta: 10:44:10 time: 0.7753 data_time: 0.0679 memory: 25074 loss: 0.0912 +2023/06/01 16:59:11 - mmengine - INFO - Epoch(train) [2][1900/5758] lr: 7.8067e-04 eta: 10:43:10 time: 0.8053 data_time: 0.0634 memory: 25074 loss: 0.0897 +2023/06/01 17:00:28 - mmengine - INFO - Epoch(train) [2][2000/5758] lr: 7.8067e-04 eta: 10:41:50 time: 0.8057 data_time: 0.0998 memory: 25074 loss: 0.0953 +2023/06/01 17:01:46 - mmengine - INFO - Epoch(train) [2][2100/5758] lr: 7.8067e-04 eta: 10:40:38 time: 0.7976 data_time: 0.0831 memory: 25074 loss: 0.1029 +2023/06/01 17:03:03 - mmengine - INFO - Epoch(train) [2][2200/5758] lr: 7.8067e-04 eta: 10:39:17 time: 0.7596 data_time: 0.0537 memory: 25074 loss: 0.1154 +2023/06/01 17:03:37 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_lr8e-4_10epoch_20230601_151933 +2023/06/01 17:04:23 - mmengine - INFO - Epoch(train) [2][2300/5758] lr: 7.8067e-04 eta: 10:38:13 time: 0.8365 data_time: 0.1179 memory: 25074 loss: 0.0848 +2023/06/01 17:05:38 - mmengine - INFO - Epoch(train) [2][2400/5758] lr: 7.8067e-04 eta: 10:36:45 time: 0.7549 data_time: 0.0503 memory: 25074 loss: 0.0939 +2023/06/01 17:06:56 - mmengine - INFO - Epoch(train) [2][2500/5758] lr: 7.8067e-04 eta: 10:35:28 time: 0.7556 data_time: 0.0352 memory: 25074 loss: 0.0870 +2023/06/01 17:08:13 - mmengine - INFO - Epoch(train) [2][2600/5758] lr: 7.8067e-04 eta: 10:34:10 time: 0.8121 data_time: 0.1060 memory: 25074 loss: 0.0941 +2023/06/01 17:09:30 - mmengine - INFO - Epoch(train) [2][2700/5758] lr: 7.8067e-04 eta: 10:32:54 time: 0.7826 data_time: 0.0612 memory: 25074 loss: 0.0978 +2023/06/01 17:10:45 - mmengine - INFO - Epoch(train) [2][2800/5758] lr: 7.8067e-04 eta: 10:31:25 time: 0.7326 data_time: 0.0283 memory: 25074 loss: 0.1292 +2023/06/01 17:12:04 - mmengine - INFO - Epoch(train) [2][2900/5758] lr: 7.8067e-04 eta: 10:30:16 time: 0.8347 data_time: 0.1272 memory: 25074 loss: 0.0820 +2023/06/01 17:13:19 - mmengine - INFO - Epoch(train) [2][3000/5758] lr: 7.8067e-04 eta: 10:28:45 time: 0.7352 data_time: 0.0299 memory: 25074 loss: 0.0940 +2023/06/01 17:14:35 - mmengine - INFO - Epoch(train) [2][3100/5758] lr: 7.8067e-04 eta: 10:27:21 time: 0.7568 data_time: 0.0243 memory: 25074 loss: 0.0730 +2023/06/01 17:15:51 - mmengine - INFO - Epoch(train) [2][3200/5758] lr: 7.8067e-04 eta: 10:25:56 time: 0.7521 data_time: 0.0463 memory: 25074 loss: 0.0723 +2023/06/01 17:16:22 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_lr8e-4_10epoch_20230601_151933 +2023/06/01 17:17:06 - mmengine - INFO - Epoch(train) [2][3300/5758] lr: 7.8067e-04 eta: 10:24:28 time: 0.7702 data_time: 0.0651 memory: 25074 loss: 0.0547 +2023/06/01 17:18:24 - mmengine - INFO - Epoch(train) [2][3400/5758] lr: 7.8067e-04 eta: 10:23:11 time: 0.7728 data_time: 0.0680 memory: 25074 loss: 0.1019 +2023/06/01 17:19:40 - mmengine - INFO - Epoch(train) [2][3500/5758] lr: 7.8067e-04 eta: 10:21:50 time: 0.7219 data_time: 0.0107 memory: 25074 loss: 0.0605 +2023/06/01 17:20:54 - mmengine - INFO - Epoch(train) [2][3600/5758] lr: 7.8067e-04 eta: 10:20:17 time: 0.7418 data_time: 0.0377 memory: 25074 loss: 0.0874 +2023/06/01 17:22:11 - mmengine - INFO - Epoch(train) [2][3700/5758] lr: 7.8067e-04 eta: 10:18:56 time: 0.7827 data_time: 0.0012 memory: 25074 loss: 0.0697 +2023/06/01 17:23:28 - mmengine - INFO - Epoch(train) [2][3800/5758] lr: 7.8067e-04 eta: 10:17:42 time: 0.7893 data_time: 0.0012 memory: 25074 loss: 0.0612 +2023/06/01 17:24:45 - mmengine - INFO - Epoch(train) [2][3900/5758] lr: 7.8067e-04 eta: 10:16:20 time: 0.7257 data_time: 0.0011 memory: 25074 loss: 0.0676 +2023/06/01 17:26:01 - mmengine - INFO - Epoch(train) [2][4000/5758] lr: 7.8067e-04 eta: 10:15:00 time: 0.7165 data_time: 0.0016 memory: 25074 loss: 0.0679 +2023/06/01 17:27:19 - mmengine - INFO - Epoch(train) [2][4100/5758] lr: 7.8067e-04 eta: 10:13:45 time: 0.8063 data_time: 0.0011 memory: 25074 loss: 0.0613 +2023/06/01 17:28:37 - mmengine - INFO - Epoch(train) [2][4200/5758] lr: 7.8067e-04 eta: 10:12:32 time: 0.7596 data_time: 0.0013 memory: 25074 loss: 0.0501 +2023/06/01 17:29:09 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_lr8e-4_10epoch_20230601_151933 +2023/06/01 17:29:55 - mmengine - INFO - Epoch(train) [2][4300/5758] lr: 7.8067e-04 eta: 10:11:20 time: 0.7571 data_time: 0.0012 memory: 25074 loss: 0.0541 +2023/06/01 17:31:12 - mmengine - INFO - Epoch(train) [2][4400/5758] lr: 7.8067e-04 eta: 10:10:00 time: 0.7774 data_time: 0.0014 memory: 25074 loss: 0.0563 +2023/06/01 17:32:29 - mmengine - INFO - Epoch(train) [2][4500/5758] lr: 7.8067e-04 eta: 10:08:43 time: 0.8590 data_time: 0.0012 memory: 25074 loss: 0.0726 +2023/06/01 17:33:46 - mmengine - INFO - Epoch(train) [2][4600/5758] lr: 7.8067e-04 eta: 10:07:23 time: 0.8274 data_time: 0.0011 memory: 25074 loss: 0.0685 +2023/06/01 17:35:04 - mmengine - INFO - Epoch(train) [2][4700/5758] lr: 7.8067e-04 eta: 10:06:11 time: 0.7579 data_time: 0.0013 memory: 25074 loss: 0.0565 +2023/06/01 17:36:21 - mmengine - INFO - Epoch(train) [2][4800/5758] lr: 7.8067e-04 eta: 10:04:53 time: 0.7490 data_time: 0.0017 memory: 25074 loss: 0.0513 +2023/06/01 17:37:38 - mmengine - INFO - Epoch(train) [2][4900/5758] lr: 7.8067e-04 eta: 10:03:35 time: 0.7642 data_time: 0.0013 memory: 25074 loss: 0.0576 +2023/06/01 17:38:55 - mmengine - INFO - Epoch(train) [2][5000/5758] lr: 7.8067e-04 eta: 10:02:17 time: 0.8177 data_time: 0.0013 memory: 25074 loss: 0.0806 +2023/06/01 17:40:11 - mmengine - INFO - Epoch(train) [2][5100/5758] lr: 7.8067e-04 eta: 10:00:54 time: 0.8291 data_time: 0.0291 memory: 25074 loss: 0.0415 +2023/06/01 17:41:29 - mmengine - INFO - Epoch(train) [2][5200/5758] lr: 7.8067e-04 eta: 9:59:42 time: 0.7862 data_time: 0.0011 memory: 25074 loss: 0.0887 +2023/06/01 17:42:03 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_lr8e-4_10epoch_20230601_151933 +2023/06/01 17:42:50 - mmengine - INFO - Epoch(train) [2][5300/5758] lr: 7.8067e-04 eta: 9:58:40 time: 0.8146 data_time: 0.0011 memory: 25074 loss: 0.0636 +2023/06/01 17:44:10 - mmengine - INFO - Epoch(train) [2][5400/5758] lr: 7.8067e-04 eta: 9:57:33 time: 0.7690 data_time: 0.0013 memory: 25074 loss: 0.0432 +2023/06/01 17:45:28 - mmengine - INFO - Epoch(train) [2][5500/5758] lr: 7.8067e-04 eta: 9:56:21 time: 0.7919 data_time: 0.0025 memory: 25074 loss: 0.0397 +2023/06/01 17:46:46 - mmengine - INFO - Epoch(train) [2][5600/5758] lr: 7.8067e-04 eta: 9:55:08 time: 0.7723 data_time: 0.0011 memory: 25074 loss: 0.0308 +2023/06/01 17:48:05 - mmengine - INFO - Epoch(train) [2][5700/5758] lr: 7.8067e-04 eta: 9:53:57 time: 0.7432 data_time: 0.0012 memory: 25074 loss: 0.0377 +2023/06/01 17:48:51 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_lr8e-4_10epoch_20230601_151933 +2023/06/01 17:48:51 - mmengine - INFO - Saving checkpoint at 2 epochs +2023/06/01 17:49:10 - mmengine - INFO - Epoch(val) [2][16/16] accuracy/top1: 76.7755 single-label/precision_classwise: [100.0, 0.0] single-label/recall_classwise: [76.77545166015625, 0.0] single-label/f1-score_classwise: [86.86211395263672, 0.0] data_time: 0.2141 time: 0.6064 +2023/06/01 17:50:37 - mmengine - INFO - Epoch(train) [3][ 100/5758] lr: 7.2456e-04 eta: 9:52:39 time: 0.8642 data_time: 0.1599 memory: 25074 loss: 0.0409 +2023/06/01 17:51:57 - mmengine - INFO - Epoch(train) [3][ 200/5758] lr: 7.2456e-04 eta: 9:51:31 time: 0.8657 data_time: 0.0013 memory: 25074 loss: 0.0408 +2023/06/01 17:53:16 - mmengine - INFO - Epoch(train) [3][ 300/5758] lr: 7.2456e-04 eta: 9:50:21 time: 0.7799 data_time: 0.0012 memory: 25074 loss: 0.0489 +2023/06/01 17:54:37 - mmengine - INFO - Epoch(train) [3][ 400/5758] lr: 7.2456e-04 eta: 9:49:16 time: 0.8322 data_time: 0.0014 memory: 25074 loss: 0.0387 +2023/06/01 17:55:43 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_lr8e-4_10epoch_20230601_151933 +2023/06/01 17:55:54 - mmengine - INFO - Epoch(train) [3][ 500/5758] lr: 7.2456e-04 eta: 9:47:59 time: 0.7364 data_time: 0.0014 memory: 25074 loss: 0.0478 +2023/06/01 17:57:12 - mmengine - INFO - Epoch(train) [3][ 600/5758] lr: 7.2456e-04 eta: 9:46:42 time: 0.8261 data_time: 0.0013 memory: 25074 loss: 0.0476 +2023/06/01 17:58:31 - mmengine - INFO - Epoch(train) [3][ 700/5758] lr: 7.2456e-04 eta: 9:45:33 time: 0.7853 data_time: 0.0014 memory: 25074 loss: 0.0371 +2023/06/01 17:59:51 - mmengine - INFO - Epoch(train) [3][ 800/5758] lr: 7.2456e-04 eta: 9:44:23 time: 0.7424 data_time: 0.0013 memory: 25074 loss: 0.0399 +2023/06/01 18:01:07 - mmengine - INFO - Epoch(train) [3][ 900/5758] lr: 7.2456e-04 eta: 9:43:02 time: 0.7254 data_time: 0.0015 memory: 25074 loss: 0.0339 +2023/06/01 18:02:25 - mmengine - INFO - Epoch(train) [3][1000/5758] lr: 7.2456e-04 eta: 9:41:44 time: 0.7534 data_time: 0.0012 memory: 25074 loss: 0.0367 +2023/06/01 18:03:40 - mmengine - INFO - Epoch(train) [3][1100/5758] lr: 7.2456e-04 eta: 9:40:19 time: 0.7488 data_time: 0.0013 memory: 25074 loss: 0.0403 +2023/06/01 18:04:54 - mmengine - INFO - Epoch(train) [3][1200/5758] lr: 7.2456e-04 eta: 9:38:50 time: 0.7519 data_time: 0.0013 memory: 25074 loss: 0.0316 +2023/06/01 18:06:08 - mmengine - INFO - Epoch(train) [3][1300/5758] lr: 7.2456e-04 eta: 9:37:19 time: 0.7207 data_time: 0.0012 memory: 25074 loss: 0.0451 +2023/06/01 18:07:20 - mmengine - INFO - Epoch(train) [3][1400/5758] lr: 7.2456e-04 eta: 9:35:45 time: 0.7175 data_time: 0.0014 memory: 25074 loss: 0.0269 +2023/06/01 18:08:24 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_lr8e-4_10epoch_20230601_151933 +2023/06/01 18:08:37 - mmengine - INFO - Epoch(train) [3][1500/5758] lr: 7.2456e-04 eta: 9:34:25 time: 0.7748 data_time: 0.0012 memory: 25074 loss: 0.0473 +2023/06/01 18:09:54 - mmengine - INFO - Epoch(train) [3][1600/5758] lr: 7.2456e-04 eta: 9:33:06 time: 0.7289 data_time: 0.0012 memory: 25074 loss: 0.0291 +2023/06/01 18:11:10 - mmengine - INFO - Epoch(train) [3][1700/5758] lr: 7.2456e-04 eta: 9:31:46 time: 0.7666 data_time: 0.0015 memory: 25074 loss: 0.0304 +2023/06/01 18:12:29 - mmengine - INFO - Epoch(train) [3][1800/5758] lr: 7.2456e-04 eta: 9:30:32 time: 0.8025 data_time: 0.0013 memory: 25074 loss: 0.0270 +2023/06/01 18:13:46 - mmengine - INFO - Epoch(train) [3][1900/5758] lr: 7.2456e-04 eta: 9:29:16 time: 0.7735 data_time: 0.0012 memory: 25074 loss: 0.0490 +2023/06/01 18:15:04 - mmengine - INFO - Epoch(train) [3][2000/5758] lr: 7.2456e-04 eta: 9:27:59 time: 0.7427 data_time: 0.0013 memory: 25074 loss: 0.0306 +2023/06/01 18:16:23 - mmengine - INFO - Epoch(train) [3][2100/5758] lr: 7.2456e-04 eta: 9:26:48 time: 0.7914 data_time: 0.0013 memory: 25074 loss: 0.0331 +2023/06/01 18:17:42 - mmengine - INFO - Epoch(train) [3][2200/5758] lr: 7.2456e-04 eta: 9:25:34 time: 0.7605 data_time: 0.0014 memory: 25074 loss: 0.0283 +2023/06/01 18:18:58 - mmengine - INFO - Epoch(train) [3][2300/5758] lr: 7.2456e-04 eta: 9:24:12 time: 0.7679 data_time: 0.0018 memory: 25074 loss: 0.0346 +2023/06/01 18:20:16 - mmengine - INFO - Epoch(train) [3][2400/5758] lr: 7.2456e-04 eta: 9:22:58 time: 0.7837 data_time: 0.0014 memory: 25074 loss: 0.0287 +2023/06/01 18:21:22 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_lr8e-4_10epoch_20230601_151933 +2023/06/01 18:21:34 - mmengine - INFO - Epoch(train) [3][2500/5758] lr: 7.2456e-04 eta: 9:21:43 time: 0.7716 data_time: 0.0012 memory: 25074 loss: 0.0242 +2023/06/01 18:22:53 - mmengine - INFO - Epoch(train) [3][2600/5758] lr: 7.2456e-04 eta: 9:20:30 time: 0.8137 data_time: 0.0013 memory: 25074 loss: 0.0366 +2023/06/01 18:24:12 - mmengine - INFO - Epoch(train) [3][2700/5758] lr: 7.2456e-04 eta: 9:19:17 time: 0.7664 data_time: 0.0013 memory: 25074 loss: 0.0303 +2023/06/01 18:25:32 - mmengine - INFO - Epoch(train) [3][2800/5758] lr: 7.2456e-04 eta: 9:18:08 time: 0.8259 data_time: 0.0014 memory: 25074 loss: 0.0319 +2023/06/01 18:26:48 - mmengine - INFO - Epoch(train) [3][2900/5758] lr: 7.2456e-04 eta: 9:16:47 time: 0.7290 data_time: 0.0012 memory: 25074 loss: 0.0267 +2023/06/01 18:28:07 - mmengine - INFO - Epoch(train) [3][3000/5758] lr: 7.2456e-04 eta: 9:15:35 time: 0.7738 data_time: 0.0014 memory: 25074 loss: 0.0375 +2023/06/01 18:29:24 - mmengine - INFO - Epoch(train) [3][3100/5758] lr: 7.2456e-04 eta: 9:14:15 time: 0.7594 data_time: 0.0011 memory: 25074 loss: 0.0331 +2023/06/01 18:30:42 - mmengine - INFO - Epoch(train) [3][3200/5758] lr: 7.2456e-04 eta: 9:12:59 time: 0.8288 data_time: 0.0013 memory: 25074 loss: 0.0310 +2023/06/01 18:32:00 - mmengine - INFO - Epoch(train) [3][3300/5758] lr: 7.2456e-04 eta: 9:11:43 time: 0.7697 data_time: 0.0013 memory: 25074 loss: 0.0347 +2023/06/01 18:33:18 - mmengine - INFO - Epoch(train) [3][3400/5758] lr: 7.2456e-04 eta: 9:10:28 time: 0.7653 data_time: 0.0014 memory: 25074 loss: 0.0253 +2023/06/01 18:34:22 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_lr8e-4_10epoch_20230601_151933 +2023/06/01 18:34:34 - mmengine - INFO - Epoch(train) [3][3500/5758] lr: 7.2456e-04 eta: 9:09:08 time: 0.7613 data_time: 0.0018 memory: 25074 loss: 0.0301 +2023/06/01 18:35:51 - mmengine - INFO - Epoch(train) [3][3600/5758] lr: 7.2456e-04 eta: 9:07:50 time: 0.7686 data_time: 0.0013 memory: 25074 loss: 0.0211 +2023/06/01 18:37:09 - mmengine - INFO - Epoch(train) [3][3700/5758] lr: 7.2456e-04 eta: 9:06:32 time: 0.7396 data_time: 0.0012 memory: 25074 loss: 0.0180 +2023/06/01 18:38:24 - mmengine - INFO - Epoch(train) [3][3800/5758] lr: 7.2456e-04 eta: 9:05:09 time: 0.8045 data_time: 0.0014 memory: 25074 loss: 0.0234 +2023/06/01 18:39:42 - mmengine - INFO - Epoch(train) [3][3900/5758] lr: 7.2456e-04 eta: 9:03:52 time: 0.7762 data_time: 0.0013 memory: 25074 loss: 0.0326 +2023/06/01 18:40:58 - mmengine - INFO - Epoch(train) [3][4000/5758] lr: 7.2456e-04 eta: 9:02:31 time: 0.7058 data_time: 0.0012 memory: 25074 loss: 0.0238 +2023/06/01 18:42:14 - mmengine - INFO - Epoch(train) [3][4100/5758] lr: 7.2456e-04 eta: 9:01:10 time: 0.7490 data_time: 0.0011 memory: 25074 loss: 0.0344 +2023/06/01 18:43:29 - mmengine - INFO - Epoch(train) [3][4200/5758] lr: 7.2456e-04 eta: 8:59:47 time: 0.7419 data_time: 0.0012 memory: 25074 loss: 0.0182 +2023/06/01 18:44:43 - mmengine - INFO - Epoch(train) [3][4300/5758] lr: 7.2456e-04 eta: 8:58:21 time: 0.7350 data_time: 0.0020 memory: 25074 loss: 0.0187 +2023/06/01 18:45:58 - mmengine - INFO - Epoch(train) [3][4400/5758] lr: 7.2456e-04 eta: 8:56:57 time: 0.7215 data_time: 0.0013 memory: 25074 loss: 0.0175 +2023/06/01 18:47:00 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_lr8e-4_10epoch_20230601_151933 +2023/06/01 18:47:12 - mmengine - INFO - Epoch(train) [3][4500/5758] lr: 7.2456e-04 eta: 8:55:32 time: 0.7292 data_time: 0.0013 memory: 25074 loss: 0.0283 +2023/06/01 18:48:26 - mmengine - INFO - Epoch(train) [3][4600/5758] lr: 7.2456e-04 eta: 8:54:06 time: 0.7865 data_time: 0.0013 memory: 25074 loss: 0.0258 +2023/06/01 18:49:41 - mmengine - INFO - Epoch(train) [3][4700/5758] lr: 7.2456e-04 eta: 8:52:42 time: 0.7103 data_time: 0.0014 memory: 25074 loss: 0.0276 +2023/06/01 18:50:55 - mmengine - INFO - Epoch(train) [3][4800/5758] lr: 7.2456e-04 eta: 8:51:17 time: 0.7225 data_time: 0.0012 memory: 25074 loss: 0.0262 +2023/06/01 18:52:10 - mmengine - INFO - Epoch(train) [3][4900/5758] lr: 7.2456e-04 eta: 8:49:54 time: 0.7503 data_time: 0.0014 memory: 25074 loss: 0.0211 +2023/06/01 18:53:24 - mmengine - INFO - Epoch(train) [3][5000/5758] lr: 7.2456e-04 eta: 8:48:28 time: 0.7485 data_time: 0.0012 memory: 25074 loss: 0.0249 +2023/06/01 18:54:39 - mmengine - INFO - Epoch(train) [3][5100/5758] lr: 7.2456e-04 eta: 8:47:06 time: 0.8163 data_time: 0.0013 memory: 25074 loss: 0.0288 +2023/06/01 18:55:54 - mmengine - INFO - Epoch(train) [3][5200/5758] lr: 7.2456e-04 eta: 8:45:42 time: 0.7064 data_time: 0.0011 memory: 25074 loss: 0.0171 +2023/06/01 18:57:06 - mmengine - INFO - Epoch(train) [3][5300/5758] lr: 7.2456e-04 eta: 8:44:14 time: 0.7820 data_time: 0.0014 memory: 25074 loss: 0.0233 +2023/06/01 18:58:18 - mmengine - INFO - Epoch(train) [3][5400/5758] lr: 7.2456e-04 eta: 8:42:43 time: 0.7306 data_time: 0.0013 memory: 25074 loss: 0.0241 +2023/06/01 18:59:20 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_lr8e-4_10epoch_20230601_151933 +2023/06/01 18:59:31 - mmengine - INFO - Epoch(train) [3][5500/5758] lr: 7.2456e-04 eta: 8:41:17 time: 0.7203 data_time: 0.0014 memory: 25074 loss: 0.0157 +2023/06/01 19:00:43 - mmengine - INFO - Epoch(train) [3][5600/5758] lr: 7.2456e-04 eta: 8:39:47 time: 0.7511 data_time: 0.0016 memory: 25074 loss: 0.0279 +2023/06/01 19:01:57 - mmengine - INFO - Epoch(train) [3][5700/5758] lr: 7.2456e-04 eta: 8:38:24 time: 0.7102 data_time: 0.0012 memory: 25074 loss: 0.0210 +2023/06/01 19:02:40 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_lr8e-4_10epoch_20230601_151933 +2023/06/01 19:02:40 - mmengine - INFO - Saving checkpoint at 3 epochs +2023/06/01 19:02:59 - mmengine - INFO - Epoch(val) [3][16/16] accuracy/top1: 96.7742 single-label/precision_classwise: [100.0, 0.0] single-label/recall_classwise: [96.7741928100586, 0.0] single-label/f1-score_classwise: [98.36064910888672, 0.0] data_time: 0.2160 time: 0.6017 +2023/06/01 19:04:21 - mmengine - INFO - Epoch(train) [4][ 100/5758] lr: 6.3718e-04 eta: 8:36:29 time: 0.7935 data_time: 0.0017 memory: 25074 loss: 0.0258 +2023/06/01 19:05:38 - mmengine - INFO - Epoch(train) [4][ 200/5758] lr: 6.3718e-04 eta: 8:35:11 time: 0.7570 data_time: 0.0014 memory: 25074 loss: 0.0159 +2023/06/01 19:06:52 - mmengine - INFO - Epoch(train) [4][ 300/5758] lr: 6.3718e-04 eta: 8:33:47 time: 0.7929 data_time: 0.0012 memory: 25074 loss: 0.0225 +2023/06/01 19:08:07 - mmengine - INFO - Epoch(train) [4][ 400/5758] lr: 6.3718e-04 eta: 8:32:26 time: 0.7393 data_time: 0.0013 memory: 25074 loss: 0.0277 +2023/06/01 19:09:21 - mmengine - INFO - Epoch(train) [4][ 500/5758] lr: 6.3718e-04 eta: 8:31:03 time: 0.8157 data_time: 0.0014 memory: 25074 loss: 0.0181 +2023/06/01 19:10:37 - mmengine - INFO - Epoch(train) [4][ 600/5758] lr: 6.3718e-04 eta: 8:29:43 time: 0.7535 data_time: 0.0014 memory: 25074 loss: 0.0164 +2023/06/01 19:11:53 - mmengine - INFO - Epoch(train) [4][ 700/5758] lr: 6.3718e-04 eta: 8:28:23 time: 0.7648 data_time: 0.0014 memory: 25074 loss: 0.0167 +2023/06/01 19:12:12 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_lr8e-4_10epoch_20230601_151933 +2023/06/01 19:13:07 - mmengine - INFO - Epoch(train) [4][ 800/5758] lr: 6.3718e-04 eta: 8:27:00 time: 0.7416 data_time: 0.0016 memory: 25074 loss: 0.0215 +2023/06/01 19:14:22 - mmengine - INFO - Epoch(train) [4][ 900/5758] lr: 6.3718e-04 eta: 8:25:39 time: 0.7078 data_time: 0.0014 memory: 25074 loss: 0.0232 +2023/06/01 19:15:37 - mmengine - INFO - Epoch(train) [4][1000/5758] lr: 6.3718e-04 eta: 8:24:17 time: 0.7725 data_time: 0.0014 memory: 25074 loss: 0.0196 +2023/06/01 19:16:51 - mmengine - INFO - Epoch(train) [4][1100/5758] lr: 6.3718e-04 eta: 8:22:54 time: 0.7734 data_time: 0.0013 memory: 25074 loss: 0.0169 +2023/06/01 19:18:04 - mmengine - INFO - Epoch(train) [4][1200/5758] lr: 6.3718e-04 eta: 8:21:29 time: 0.7418 data_time: 0.0014 memory: 25074 loss: 0.0227 +2023/06/01 19:19:18 - mmengine - INFO - Epoch(train) [4][1300/5758] lr: 6.3718e-04 eta: 8:20:06 time: 0.7640 data_time: 0.0013 memory: 25074 loss: 0.0130 +2023/06/01 19:20:33 - mmengine - INFO - Epoch(train) [4][1400/5758] lr: 6.3718e-04 eta: 8:18:44 time: 0.7582 data_time: 0.0013 memory: 25074 loss: 0.0220 +2023/06/01 19:21:47 - mmengine - INFO - Epoch(train) [4][1500/5758] lr: 6.3718e-04 eta: 8:17:21 time: 0.7138 data_time: 0.0013 memory: 25074 loss: 0.0128 +2023/06/01 19:23:00 - mmengine - INFO - Epoch(train) [4][1600/5758] lr: 6.3718e-04 eta: 8:15:57 time: 0.7485 data_time: 0.0013 memory: 25074 loss: 0.0154 +2023/06/01 19:24:13 - mmengine - INFO - Epoch(train) [4][1700/5758] lr: 6.3718e-04 eta: 8:14:31 time: 0.7062 data_time: 0.0014 memory: 25074 loss: 0.0123 +2023/06/01 19:24:32 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_lr8e-4_10epoch_20230601_151933 +2023/06/01 19:25:28 - mmengine - INFO - Epoch(train) [4][1800/5758] lr: 6.3718e-04 eta: 8:13:11 time: 0.7450 data_time: 0.0014 memory: 25074 loss: 0.0102 +2023/06/01 19:26:42 - mmengine - INFO - Epoch(train) [4][1900/5758] lr: 6.3718e-04 eta: 8:11:49 time: 0.7606 data_time: 0.0015 memory: 25074 loss: 0.0224 +2023/06/01 19:27:56 - mmengine - INFO - Epoch(train) [4][2000/5758] lr: 6.3718e-04 eta: 8:10:27 time: 0.7531 data_time: 0.0012 memory: 25074 loss: 0.0173 +2023/06/01 19:29:13 - mmengine - INFO - Epoch(train) [4][2100/5758] lr: 6.3718e-04 eta: 8:09:10 time: 0.7509 data_time: 0.0013 memory: 25074 loss: 0.0196 +2023/06/01 19:30:27 - mmengine - INFO - Epoch(train) [4][2200/5758] lr: 6.3718e-04 eta: 8:07:47 time: 0.7145 data_time: 0.0014 memory: 25074 loss: 0.0238 +2023/06/01 19:31:43 - mmengine - INFO - Epoch(train) [4][2300/5758] lr: 6.3718e-04 eta: 8:06:29 time: 0.7425 data_time: 0.0013 memory: 25074 loss: 0.0136 +2023/06/01 19:32:57 - mmengine - INFO - Epoch(train) [4][2400/5758] lr: 6.3718e-04 eta: 8:05:08 time: 0.7495 data_time: 0.0013 memory: 25074 loss: 0.0176 +2023/06/01 19:34:13 - mmengine - INFO - Epoch(train) [4][2500/5758] lr: 6.3718e-04 eta: 8:03:49 time: 0.7634 data_time: 0.0014 memory: 25074 loss: 0.0263 +2023/06/01 19:35:28 - mmengine - INFO - Epoch(train) [4][2600/5758] lr: 6.3718e-04 eta: 8:02:28 time: 0.7888 data_time: 0.0014 memory: 25074 loss: 0.0170 +2023/06/01 19:36:43 - mmengine - INFO - Epoch(train) [4][2700/5758] lr: 6.3718e-04 eta: 8:01:09 time: 0.7309 data_time: 0.0012 memory: 25074 loss: 0.0184 +2023/06/01 19:37:03 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_lr8e-4_10epoch_20230601_151933 +2023/06/01 19:37:59 - mmengine - INFO - Epoch(train) [4][2800/5758] lr: 6.3718e-04 eta: 7:59:51 time: 0.7832 data_time: 0.0014 memory: 25074 loss: 0.0176 +2023/06/01 19:39:17 - mmengine - INFO - Epoch(train) [4][2900/5758] lr: 6.3718e-04 eta: 7:58:36 time: 0.7578 data_time: 0.0013 memory: 25074 loss: 0.0150 +2023/06/01 19:40:38 - mmengine - INFO - Epoch(train) [4][3000/5758] lr: 6.3718e-04 eta: 7:57:27 time: 0.9324 data_time: 0.0014 memory: 25074 loss: 0.0153 +2023/06/01 19:41:57 - mmengine - INFO - Epoch(train) [4][3100/5758] lr: 6.3718e-04 eta: 7:56:13 time: 0.8992 data_time: 0.0015 memory: 25074 loss: 0.0159 +2023/06/01 19:43:17 - mmengine - INFO - Epoch(train) [4][3200/5758] lr: 6.3718e-04 eta: 7:55:02 time: 0.8718 data_time: 0.0012 memory: 25074 loss: 0.0119 +2023/06/01 19:44:39 - mmengine - INFO - Epoch(train) [4][3300/5758] lr: 6.3718e-04 eta: 7:53:56 time: 0.8446 data_time: 0.0012 memory: 25074 loss: 0.0200 +2023/06/01 19:45:58 - mmengine - INFO - Epoch(train) [4][3400/5758] lr: 6.3718e-04 eta: 7:52:42 time: 0.7645 data_time: 0.0012 memory: 25074 loss: 0.0125 +2023/06/01 19:47:20 - mmengine - INFO - Epoch(train) [4][3500/5758] lr: 6.3718e-04 eta: 7:51:34 time: 0.8373 data_time: 0.0013 memory: 25074 loss: 0.0086 +2023/06/01 19:48:43 - mmengine - INFO - Epoch(train) [4][3600/5758] lr: 6.3718e-04 eta: 7:50:29 time: 0.8577 data_time: 0.0016 memory: 25074 loss: 0.0102 +2023/06/01 19:50:04 - mmengine - INFO - Epoch(train) [4][3700/5758] lr: 6.3718e-04 eta: 7:49:19 time: 0.7834 data_time: 0.0014 memory: 25074 loss: 0.0200 +2023/06/01 19:50:24 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_lr8e-4_10epoch_20230601_151933 +2023/06/01 19:51:23 - mmengine - INFO - Epoch(train) [4][3800/5758] lr: 6.3718e-04 eta: 7:48:06 time: 0.7836 data_time: 0.0013 memory: 25074 loss: 0.0177 +2023/06/01 19:52:44 - mmengine - INFO - Epoch(train) [4][3900/5758] lr: 6.3718e-04 eta: 7:46:55 time: 0.8408 data_time: 0.0013 memory: 25074 loss: 0.0173 +2023/06/01 19:54:03 - mmengine - INFO - Epoch(train) [4][4000/5758] lr: 6.3718e-04 eta: 7:45:42 time: 0.8116 data_time: 0.0014 memory: 25074 loss: 0.0121 +2023/06/01 19:55:20 - mmengine - INFO - Epoch(train) [4][4100/5758] lr: 6.3718e-04 eta: 7:44:25 time: 0.7603 data_time: 0.0015 memory: 25074 loss: 0.0125 +2023/06/01 19:56:37 - mmengine - INFO - Epoch(train) [4][4200/5758] lr: 6.3718e-04 eta: 7:43:08 time: 0.7567 data_time: 0.0013 memory: 25074 loss: 0.0108 +2023/06/01 19:57:53 - mmengine - INFO - Epoch(train) [4][4300/5758] lr: 6.3718e-04 eta: 7:41:49 time: 0.7464 data_time: 0.0012 memory: 25074 loss: 0.0148 +2023/06/01 19:59:12 - mmengine - INFO - Epoch(train) [4][4400/5758] lr: 6.3718e-04 eta: 7:40:36 time: 0.7918 data_time: 0.0015 memory: 25074 loss: 0.0146 +2023/06/01 20:00:32 - mmengine - INFO - Epoch(train) [4][4500/5758] lr: 6.3718e-04 eta: 7:39:24 time: 0.7790 data_time: 0.0013 memory: 25074 loss: 0.0110 +2023/06/01 20:01:50 - mmengine - INFO - Epoch(train) [4][4600/5758] lr: 6.3718e-04 eta: 7:38:08 time: 0.7412 data_time: 0.0013 memory: 25074 loss: 0.0235 +2023/06/01 20:03:09 - mmengine - INFO - Epoch(train) [4][4700/5758] lr: 6.3718e-04 eta: 7:36:55 time: 0.7579 data_time: 0.0013 memory: 25074 loss: 0.0132 +2023/06/01 20:03:29 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_lr8e-4_10epoch_20230601_151933 +2023/06/01 20:04:26 - mmengine - INFO - Epoch(train) [4][4800/5758] lr: 6.3718e-04 eta: 7:35:38 time: 0.8052 data_time: 0.0013 memory: 25074 loss: 0.0113 +2023/06/01 20:05:45 - mmengine - INFO - Epoch(train) [4][4900/5758] lr: 6.3718e-04 eta: 7:34:24 time: 0.7334 data_time: 0.0015 memory: 25074 loss: 0.0158 +2023/06/01 20:07:00 - mmengine - INFO - Epoch(train) [4][5000/5758] lr: 6.3718e-04 eta: 7:33:03 time: 0.7601 data_time: 0.0014 memory: 25074 loss: 0.0120 +2023/06/01 20:08:15 - mmengine - INFO - Epoch(train) [4][5100/5758] lr: 6.3718e-04 eta: 7:31:43 time: 0.7546 data_time: 0.0014 memory: 25074 loss: 0.0113 +2023/06/01 20:09:33 - mmengine - INFO - Epoch(train) [4][5200/5758] lr: 6.3718e-04 eta: 7:30:28 time: 0.7790 data_time: 0.0012 memory: 25074 loss: 0.0093 +2023/06/01 20:10:51 - mmengine - INFO - Epoch(train) [4][5300/5758] lr: 6.3718e-04 eta: 7:29:12 time: 0.8177 data_time: 0.0014 memory: 25074 loss: 0.0046 +2023/06/01 20:12:05 - mmengine - INFO - Epoch(train) [4][5400/5758] lr: 6.3718e-04 eta: 7:27:51 time: 0.7535 data_time: 0.0016 memory: 25074 loss: 0.0150 +2023/06/01 20:13:20 - mmengine - INFO - Epoch(train) [4][5500/5758] lr: 6.3718e-04 eta: 7:26:31 time: 0.8566 data_time: 0.0012 memory: 25074 loss: 0.0181 +2023/06/01 20:14:38 - mmengine - INFO - Epoch(train) [4][5600/5758] lr: 6.3718e-04 eta: 7:25:16 time: 0.7794 data_time: 0.0012 memory: 25074 loss: 0.0124 +2023/06/01 20:15:54 - mmengine - INFO - Epoch(train) [4][5700/5758] lr: 6.3718e-04 eta: 7:23:57 time: 0.7683 data_time: 0.0015 memory: 25074 loss: 0.0172 +2023/06/01 20:16:13 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_lr8e-4_10epoch_20230601_151933 +2023/06/01 20:16:38 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_lr8e-4_10epoch_20230601_151933 +2023/06/01 20:16:38 - mmengine - INFO - Saving checkpoint at 4 epochs +2023/06/01 20:16:56 - mmengine - INFO - Epoch(val) [4][16/16] accuracy/top1: 99.0089 single-label/precision_classwise: [100.0, 0.0] single-label/recall_classwise: [99.00890350341797, 0.0] single-label/f1-score_classwise: [99.50198364257812, 0.0] data_time: 0.2015 time: 0.5868 +2023/06/01 20:18:20 - mmengine - INFO - Epoch(train) [5][ 100/5758] lr: 5.2706e-04 eta: 7:22:05 time: 0.7147 data_time: 0.0013 memory: 25074 loss: 0.0129 +2023/06/01 20:19:38 - mmengine - INFO - Epoch(train) [5][ 200/5758] lr: 5.2706e-04 eta: 7:20:50 time: 0.7590 data_time: 0.0013 memory: 25074 loss: 0.0067 +2023/06/01 20:20:56 - mmengine - INFO - Epoch(train) [5][ 300/5758] lr: 5.2706e-04 eta: 7:19:33 time: 0.7968 data_time: 0.0013 memory: 25074 loss: 0.0094 +2023/06/01 20:22:17 - mmengine - INFO - Epoch(train) [5][ 400/5758] lr: 5.2706e-04 eta: 7:18:22 time: 0.8214 data_time: 0.0013 memory: 25074 loss: 0.0155 +2023/06/01 20:23:36 - mmengine - INFO - Epoch(train) [5][ 500/5758] lr: 5.2706e-04 eta: 7:17:09 time: 0.8269 data_time: 0.0014 memory: 25074 loss: 0.0063 +2023/06/01 20:24:55 - mmengine - INFO - Epoch(train) [5][ 600/5758] lr: 5.2706e-04 eta: 7:15:55 time: 0.7884 data_time: 0.0012 memory: 25074 loss: 0.0113 +2023/06/01 20:26:17 - mmengine - INFO - Epoch(train) [5][ 700/5758] lr: 5.2706e-04 eta: 7:14:44 time: 0.8699 data_time: 0.0015 memory: 25074 loss: 0.0225 +2023/06/01 20:27:37 - mmengine - INFO - Epoch(train) [5][ 800/5758] lr: 5.2706e-04 eta: 7:13:31 time: 0.8240 data_time: 0.0014 memory: 25074 loss: 0.0139 +2023/06/01 20:28:54 - mmengine - INFO - Epoch(train) [5][ 900/5758] lr: 5.2706e-04 eta: 7:12:14 time: 0.7652 data_time: 0.0013 memory: 25074 loss: 0.0084 +2023/06/01 20:29:49 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_lr8e-4_10epoch_20230601_151933 +2023/06/01 20:30:16 - mmengine - INFO - Epoch(train) [5][1000/5758] lr: 5.2706e-04 eta: 7:11:04 time: 0.9262 data_time: 0.0018 memory: 25074 loss: 0.0099 +2023/06/01 20:31:38 - mmengine - INFO - Epoch(train) [5][1100/5758] lr: 5.2706e-04 eta: 7:09:54 time: 0.8112 data_time: 0.0014 memory: 25074 loss: 0.0060 +2023/06/01 20:32:59 - mmengine - INFO - Epoch(train) [5][1200/5758] lr: 5.2706e-04 eta: 7:08:41 time: 0.8130 data_time: 0.0021 memory: 25074 loss: 0.0144 +2023/06/01 20:34:15 - mmengine - INFO - Epoch(train) [5][1300/5758] lr: 5.2706e-04 eta: 7:07:23 time: 0.7553 data_time: 0.0016 memory: 25074 loss: 0.0104 +2023/06/01 20:35:32 - mmengine - INFO - Epoch(train) [5][1400/5758] lr: 5.2706e-04 eta: 7:06:05 time: 0.7635 data_time: 0.0015 memory: 25074 loss: 0.0051 +2023/06/01 20:36:52 - mmengine - INFO - Epoch(train) [5][1500/5758] lr: 5.2706e-04 eta: 7:04:53 time: 0.9195 data_time: 0.0013 memory: 25074 loss: 0.0063 +2023/06/01 20:38:09 - mmengine - INFO - Epoch(train) [5][1600/5758] lr: 5.2706e-04 eta: 7:03:36 time: 0.7294 data_time: 0.0015 memory: 25074 loss: 0.0092 +2023/06/01 20:39:28 - mmengine - INFO - Epoch(train) [5][1700/5758] lr: 5.2706e-04 eta: 7:02:20 time: 0.7862 data_time: 0.0015 memory: 25074 loss: 0.0117 +2023/06/01 20:40:46 - mmengine - INFO - Epoch(train) [5][1800/5758] lr: 5.2706e-04 eta: 7:01:05 time: 0.7528 data_time: 0.0014 memory: 25074 loss: 0.0096 +2023/06/01 20:42:03 - mmengine - INFO - Epoch(train) [5][1900/5758] lr: 5.2706e-04 eta: 6:59:48 time: 0.8107 data_time: 0.0014 memory: 25074 loss: 0.0092 +2023/06/01 20:42:56 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_lr8e-4_10epoch_20230601_151933 +2023/06/01 20:43:19 - mmengine - INFO - Epoch(train) [5][2000/5758] lr: 5.2706e-04 eta: 6:58:29 time: 0.7297 data_time: 0.0012 memory: 25074 loss: 0.0036 +2023/06/01 20:44:37 - mmengine - INFO - Epoch(train) [5][2100/5758] lr: 5.2706e-04 eta: 6:57:12 time: 0.7547 data_time: 0.0013 memory: 25074 loss: 0.0079 +2023/06/01 20:45:52 - mmengine - INFO - Epoch(train) [5][2200/5758] lr: 5.2706e-04 eta: 6:55:53 time: 0.7851 data_time: 0.0018 memory: 25074 loss: 0.0123 +2023/06/01 20:47:07 - mmengine - INFO - Epoch(train) [5][2300/5758] lr: 5.2706e-04 eta: 6:54:33 time: 0.7426 data_time: 0.0013 memory: 25074 loss: 0.0095 +2023/06/01 20:48:24 - mmengine - INFO - Epoch(train) [5][2400/5758] lr: 5.2706e-04 eta: 6:53:16 time: 0.7632 data_time: 0.0013 memory: 25074 loss: 0.0094 +2023/06/01 20:49:39 - mmengine - INFO - Epoch(train) [5][2500/5758] lr: 5.2706e-04 eta: 6:51:56 time: 0.7866 data_time: 0.0015 memory: 25074 loss: 0.0083 +2023/06/01 20:50:59 - mmengine - INFO - Epoch(train) [5][2600/5758] lr: 5.2706e-04 eta: 6:50:42 time: 0.7794 data_time: 0.0016 memory: 25074 loss: 0.0123 +2023/06/01 20:52:30 - mmengine - INFO - Epoch(train) [5][2700/5758] lr: 5.2706e-04 eta: 6:49:42 time: 0.8221 data_time: 0.0014 memory: 25074 loss: 0.0102 +2023/06/01 20:54:02 - mmengine - INFO - Epoch(train) [5][2800/5758] lr: 5.2706e-04 eta: 6:48:44 time: 1.0886 data_time: 0.0013 memory: 25074 loss: 0.0081 +2023/06/01 20:55:45 - mmengine - INFO - Epoch(train) [5][2900/5758] lr: 5.2706e-04 eta: 6:47:58 time: 1.1644 data_time: 0.0012 memory: 25074 loss: 0.0100 +2023/06/01 20:56:56 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_lr8e-4_10epoch_20230601_151933 +2023/06/01 20:57:28 - mmengine - INFO - Epoch(train) [5][3000/5758] lr: 5.2706e-04 eta: 6:47:11 time: 1.0126 data_time: 0.0013 memory: 25074 loss: 0.0131 +2023/06/01 20:59:03 - mmengine - INFO - Epoch(train) [5][3100/5758] lr: 5.2706e-04 eta: 6:46:15 time: 0.7859 data_time: 0.0012 memory: 25074 loss: 0.0130 +2023/06/01 21:00:23 - mmengine - INFO - Epoch(train) [5][3200/5758] lr: 5.2706e-04 eta: 6:45:00 time: 0.8908 data_time: 0.0014 memory: 25074 loss: 0.0082 +2023/06/01 21:01:45 - mmengine - INFO - Epoch(train) [5][3300/5758] lr: 5.2706e-04 eta: 6:43:48 time: 0.7799 data_time: 0.0011 memory: 25074 loss: 0.0135 +2023/06/01 21:03:11 - mmengine - INFO - Epoch(train) [5][3400/5758] lr: 5.2706e-04 eta: 6:42:41 time: 0.8007 data_time: 0.0011 memory: 25074 loss: 0.0117 +2023/06/01 21:04:37 - mmengine - INFO - Epoch(train) [5][3500/5758] lr: 5.2706e-04 eta: 6:41:33 time: 0.8350 data_time: 0.0016 memory: 25074 loss: 0.0143 +2023/06/01 21:06:16 - mmengine - INFO - Epoch(train) [5][3600/5758] lr: 5.2706e-04 eta: 6:40:40 time: 1.1066 data_time: 0.0015 memory: 25074 loss: 0.0078 +2023/06/01 21:08:01 - mmengine - INFO - Epoch(train) [5][3700/5758] lr: 5.2706e-04 eta: 6:39:54 time: 1.1303 data_time: 0.0013 memory: 25074 loss: 0.0064 +2023/06/01 21:09:48 - mmengine - INFO - Epoch(train) [5][3800/5758] lr: 5.2706e-04 eta: 6:39:09 time: 1.0235 data_time: 0.0016 memory: 25074 loss: 0.0089 +2023/06/01 21:11:39 - mmengine - INFO - Epoch(train) [5][3900/5758] lr: 5.2706e-04 eta: 6:38:30 time: 1.0116 data_time: 0.0016 memory: 25074 loss: 0.0067 +2023/06/01 21:13:07 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_lr8e-4_10epoch_20230601_151933 +2023/06/01 21:13:41 - mmengine - INFO - Epoch(train) [5][4000/5758] lr: 5.2706e-04 eta: 6:38:01 time: 1.0733 data_time: 0.0013 memory: 25074 loss: 0.0073 +2023/06/01 21:15:28 - mmengine - INFO - Epoch(train) [5][4100/5758] lr: 5.2706e-04 eta: 6:37:16 time: 1.1214 data_time: 0.0012 memory: 25074 loss: 0.0074 +2023/06/01 21:17:05 - mmengine - INFO - Epoch(train) [5][4200/5758] lr: 5.2706e-04 eta: 6:36:17 time: 0.7100 data_time: 0.0014 memory: 25074 loss: 0.0105 +2023/06/01 21:18:17 - mmengine - INFO - Epoch(train) [5][4300/5758] lr: 5.2706e-04 eta: 6:34:53 time: 0.7223 data_time: 0.0013 memory: 25074 loss: 0.0061 +2023/06/01 21:19:29 - mmengine - INFO - Epoch(train) [5][4400/5758] lr: 5.2706e-04 eta: 6:33:27 time: 0.7129 data_time: 0.0014 memory: 25074 loss: 0.0081 +2023/06/01 21:20:41 - mmengine - INFO - Epoch(train) [5][4500/5758] lr: 5.2706e-04 eta: 6:32:02 time: 0.7286 data_time: 0.0014 memory: 25074 loss: 0.0101 +2023/06/01 21:21:55 - mmengine - INFO - Epoch(train) [5][4600/5758] lr: 5.2706e-04 eta: 6:30:39 time: 0.7627 data_time: 0.0017 memory: 25074 loss: 0.0092 +2023/06/01 21:23:09 - mmengine - INFO - Epoch(train) [5][4700/5758] lr: 5.2706e-04 eta: 6:29:16 time: 0.7391 data_time: 0.0014 memory: 25074 loss: 0.0088 +2023/06/01 21:24:23 - mmengine - INFO - Epoch(train) [5][4800/5758] lr: 5.2706e-04 eta: 6:27:53 time: 0.7088 data_time: 0.0012 memory: 25074 loss: 0.0085 +2023/06/01 21:25:37 - mmengine - INFO - Epoch(train) [5][4900/5758] lr: 5.2706e-04 eta: 6:26:30 time: 0.7325 data_time: 0.0017 memory: 25074 loss: 0.0092 +2023/06/01 21:26:27 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_lr8e-4_10epoch_20230601_151933 +2023/06/01 21:26:50 - mmengine - INFO - Epoch(train) [5][5000/5758] lr: 5.2706e-04 eta: 6:25:07 time: 0.7610 data_time: 0.0015 memory: 25074 loss: 0.0099 +2023/06/01 21:28:04 - mmengine - INFO - Epoch(train) [5][5100/5758] lr: 5.2706e-04 eta: 6:23:44 time: 0.7463 data_time: 0.0015 memory: 25074 loss: 0.0040 +2023/06/01 21:29:19 - mmengine - INFO - Epoch(train) [5][5200/5758] lr: 5.2706e-04 eta: 6:22:23 time: 0.7690 data_time: 0.0014 memory: 25074 loss: 0.0138 +2023/06/01 21:30:34 - mmengine - INFO - Epoch(train) [5][5300/5758] lr: 5.2706e-04 eta: 6:21:02 time: 0.7084 data_time: 0.0013 memory: 25074 loss: 0.0067 +2023/06/01 21:31:49 - mmengine - INFO - Epoch(train) [5][5400/5758] lr: 5.2706e-04 eta: 6:19:40 time: 0.7608 data_time: 0.0015 memory: 25074 loss: 0.0052 +2023/06/01 21:33:06 - mmengine - INFO - Epoch(train) [5][5500/5758] lr: 5.2706e-04 eta: 6:18:21 time: 0.7732 data_time: 0.0015 memory: 25074 loss: 0.0118 +2023/06/01 21:34:21 - mmengine - INFO - Epoch(train) [5][5600/5758] lr: 5.2706e-04 eta: 6:17:00 time: 0.7315 data_time: 0.0015 memory: 25074 loss: 0.0055 +2023/06/01 21:35:38 - mmengine - INFO - Epoch(train) [5][5700/5758] lr: 5.2706e-04 eta: 6:15:40 time: 0.8020 data_time: 0.0019 memory: 25074 loss: 0.0111 +2023/06/01 21:36:21 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_lr8e-4_10epoch_20230601_151933 +2023/06/01 21:36:21 - mmengine - INFO - Saving checkpoint at 5 epochs +2023/06/01 21:36:40 - mmengine - INFO - Epoch(val) [5][16/16] accuracy/top1: 99.5518 single-label/precision_classwise: [100.0, 0.0] single-label/recall_classwise: [99.55179595947266, 0.0] single-label/f1-score_classwise: [99.77539825439453, 0.0] data_time: 0.2146 time: 0.6045 +2023/06/01 21:38:04 - mmengine - INFO - Epoch(train) [6][ 100/5758] lr: 4.0500e-04 eta: 6:13:40 time: 0.7455 data_time: 0.0014 memory: 25074 loss: 0.0051 +2023/06/01 21:39:21 - mmengine - INFO - Epoch(train) [6][ 200/5758] lr: 4.0500e-04 eta: 6:12:20 time: 0.8215 data_time: 0.0017 memory: 25074 loss: 0.0084 +2023/06/01 21:39:29 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_lr8e-4_10epoch_20230601_151933 +2023/06/01 21:40:37 - mmengine - INFO - Epoch(train) [6][ 300/5758] lr: 4.0500e-04 eta: 6:11:00 time: 0.7772 data_time: 0.0014 memory: 25074 loss: 0.0034 +2023/06/01 21:41:55 - mmengine - INFO - Epoch(train) [6][ 400/5758] lr: 4.0500e-04 eta: 6:09:42 time: 0.8253 data_time: 0.0012 memory: 25074 loss: 0.0066 +2023/06/01 21:43:12 - mmengine - INFO - Epoch(train) [6][ 500/5758] lr: 4.0500e-04 eta: 6:08:22 time: 0.7902 data_time: 0.0011 memory: 25074 loss: 0.0078 +2023/06/01 21:44:27 - mmengine - INFO - Epoch(train) [6][ 600/5758] lr: 4.0500e-04 eta: 6:07:01 time: 0.8013 data_time: 0.0015 memory: 25074 loss: 0.0142 +2023/06/01 21:45:44 - mmengine - INFO - Epoch(train) [6][ 700/5758] lr: 4.0500e-04 eta: 6:05:42 time: 0.7547 data_time: 0.0017 memory: 25074 loss: 0.0048 +2023/06/01 21:47:01 - mmengine - INFO - Epoch(train) [6][ 800/5758] lr: 4.0500e-04 eta: 6:04:23 time: 0.7394 data_time: 0.0011 memory: 25074 loss: 0.0054 +2023/06/01 21:48:16 - mmengine - INFO - Epoch(train) [6][ 900/5758] lr: 4.0500e-04 eta: 6:03:02 time: 0.7421 data_time: 0.0012 memory: 25074 loss: 0.0057 +2023/06/01 21:49:31 - mmengine - INFO - Epoch(train) [6][1000/5758] lr: 4.0500e-04 eta: 6:01:41 time: 0.8012 data_time: 0.0014 memory: 25074 loss: 0.0064 +2023/06/01 21:50:45 - mmengine - INFO - Epoch(train) [6][1100/5758] lr: 4.0500e-04 eta: 6:00:19 time: 0.7989 data_time: 0.0012 memory: 25074 loss: 0.0105 +2023/06/01 21:52:00 - mmengine - INFO - Epoch(train) [6][1200/5758] lr: 4.0500e-04 eta: 5:58:59 time: 0.7886 data_time: 0.0012 memory: 25074 loss: 0.0087 +2023/06/01 21:52:08 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_lr8e-4_10epoch_20230601_151933 +2023/06/01 21:53:18 - mmengine - INFO - Epoch(train) [6][1300/5758] lr: 4.0500e-04 eta: 5:57:40 time: 0.7340 data_time: 0.0013 memory: 25074 loss: 0.0102 +2023/06/01 21:54:33 - mmengine - INFO - Epoch(train) [6][1400/5758] lr: 4.0500e-04 eta: 5:56:20 time: 0.7762 data_time: 0.0015 memory: 25074 loss: 0.0079 +2023/06/01 21:55:47 - mmengine - INFO - Epoch(train) [6][1500/5758] lr: 4.0500e-04 eta: 5:54:57 time: 0.7268 data_time: 0.0017 memory: 25074 loss: 0.0067 +2023/06/01 21:57:03 - mmengine - INFO - Epoch(train) [6][1600/5758] lr: 4.0500e-04 eta: 5:53:38 time: 0.7981 data_time: 0.0011 memory: 25074 loss: 0.0078 +2023/06/01 21:58:19 - mmengine - INFO - Epoch(train) [6][1700/5758] lr: 4.0500e-04 eta: 5:52:18 time: 0.7192 data_time: 0.0017 memory: 25074 loss: 0.0042 +2023/06/01 21:59:32 - mmengine - INFO - Epoch(train) [6][1800/5758] lr: 4.0500e-04 eta: 5:50:55 time: 0.7095 data_time: 0.0012 memory: 25074 loss: 0.0039 +2023/06/01 22:00:47 - mmengine - INFO - Epoch(train) [6][1900/5758] lr: 4.0500e-04 eta: 5:49:35 time: 0.7100 data_time: 0.0017 memory: 25074 loss: 0.0069 +2023/06/01 22:01:59 - mmengine - INFO - Epoch(train) [6][2000/5758] lr: 4.0500e-04 eta: 5:48:12 time: 0.7125 data_time: 0.0013 memory: 25074 loss: 0.0052 +2023/06/01 22:03:12 - mmengine - INFO - Epoch(train) [6][2100/5758] lr: 4.0500e-04 eta: 5:46:49 time: 0.7057 data_time: 0.0011 memory: 25074 loss: 0.0087 +2023/06/01 22:04:25 - mmengine - INFO - Epoch(train) [6][2200/5758] lr: 4.0500e-04 eta: 5:45:27 time: 0.7326 data_time: 0.0011 memory: 25074 loss: 0.0027 +2023/06/01 22:04:32 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_lr8e-4_10epoch_20230601_151933 +2023/06/01 22:05:39 - mmengine - INFO - Epoch(train) [6][2300/5758] lr: 4.0500e-04 eta: 5:44:05 time: 0.7253 data_time: 0.0013 memory: 25074 loss: 0.0045 +2023/06/01 22:06:52 - mmengine - INFO - Epoch(train) [6][2400/5758] lr: 4.0500e-04 eta: 5:42:44 time: 0.7326 data_time: 0.0019 memory: 25074 loss: 0.0023 +2023/06/01 22:08:06 - mmengine - INFO - Epoch(train) [6][2500/5758] lr: 4.0500e-04 eta: 5:41:23 time: 0.7439 data_time: 0.0014 memory: 25074 loss: 0.0074 +2023/06/01 22:09:20 - mmengine - INFO - Epoch(train) [6][2600/5758] lr: 4.0500e-04 eta: 5:40:01 time: 0.7132 data_time: 0.0015 memory: 25074 loss: 0.0057 +2023/06/01 22:10:35 - mmengine - INFO - Epoch(train) [6][2700/5758] lr: 4.0500e-04 eta: 5:38:41 time: 0.7745 data_time: 0.0015 memory: 25074 loss: 0.0102 +2023/06/01 22:11:51 - mmengine - INFO - Epoch(train) [6][2800/5758] lr: 4.0500e-04 eta: 5:37:21 time: 0.7077 data_time: 0.0012 memory: 25074 loss: 0.0038 +2023/06/01 22:13:08 - mmengine - INFO - Epoch(train) [6][2900/5758] lr: 4.0500e-04 eta: 5:36:03 time: 0.7656 data_time: 0.0013 memory: 25074 loss: 0.0050 +2023/06/01 22:14:24 - mmengine - INFO - Epoch(train) [6][3000/5758] lr: 4.0500e-04 eta: 5:34:44 time: 0.7475 data_time: 0.0012 memory: 25074 loss: 0.0060 +2023/06/01 22:15:41 - mmengine - INFO - Epoch(train) [6][3100/5758] lr: 4.0500e-04 eta: 5:33:25 time: 0.7584 data_time: 0.0015 memory: 25074 loss: 0.0122 +2023/06/01 22:16:57 - mmengine - INFO - Epoch(train) [6][3200/5758] lr: 4.0500e-04 eta: 5:32:05 time: 0.8306 data_time: 0.0016 memory: 25074 loss: 0.0040 +2023/06/01 22:17:04 - mmengine - INFO - Exp name: convnext_small_4xb256_fake5m_lr8e-4_10epoch_20230601_151933 +2023/06/01 22:18:11 - mmengine - INFO - Epoch(train) [6][3300/5758] lr: 4.0500e-04 eta: 5:30:45 time: 0.7689 data_time: 0.0016 memory: 25074 loss: 0.0070 +2023/06/01 22:19:27 - mmengine - INFO - Epoch(train) [6][3400/5758] lr: 4.0500e-04 eta: 5:29:25 time: 0.7551 data_time: 0.0012 memory: 25074 loss: 0.0103 +2023/06/01 22:20:44 - mmengine - INFO - Epoch(train) [6][3500/5758] lr: 4.0500e-04 eta: 5:28:07 time: 0.7272 data_time: 0.0013 memory: 25074 loss: 0.0054 diff --git a/convnext_small_4xb256_fake5m_lr8e-4_10epoch/20230601_151933/vis_data/20230601_151933.json b/convnext_small_4xb256_fake5m_lr8e-4_10epoch/20230601_151933/vis_data/20230601_151933.json new file mode 100644 index 0000000000000000000000000000000000000000..fdca5ba5a049858db6a574c4772f2095e861f390 --- /dev/null +++ b/convnext_small_4xb256_fake5m_lr8e-4_10epoch/20230601_151933/vis_data/20230601_151933.json @@ -0,0 +1,325 @@ +{"lr": 0.0008, "data_time": 0.0013764381408691406, "loss": 0.6050899505615235, "time": 0.7506237506866456, "epoch": 1, "memory": 25071, "step": 100} +{"lr": 0.0008, "data_time": 0.0013804197311401366, "loss": 0.5873622417449951, "time": 0.7945192337036133, "epoch": 1, "memory": 25071, "step": 200} +{"lr": 0.0008, "data_time": 0.0013028383255004883, "loss": 0.5443078756332398, "time": 0.7569274187088013, "epoch": 1, "memory": 25071, "step": 300} +{"lr": 0.0008, "data_time": 0.0012797832489013672, "loss": 0.523950707912445, "time": 0.7552407264709473, "epoch": 1, "memory": 25071, "step": 400} +{"lr": 0.0008, "data_time": 0.09460806846618652, "loss": 0.4858797013759613, "time": 0.8013857364654541, "epoch": 1, "memory": 25071, "step": 500} +{"lr": 0.0008, "data_time": 0.1175036907196045, "loss": 0.48607711791992186, "time": 0.8218770980834961, "epoch": 1, "memory": 25071, "step": 600} +{"lr": 0.0008, "data_time": 0.14032084941864015, "loss": 0.4458679795265198, "time": 0.8456096649169922, "epoch": 1, "memory": 25071, "step": 700} +{"lr": 0.0008, "data_time": 0.00155184268951416, "loss": 0.46259907484054563, "time": 0.7259519338607788, "epoch": 1, "memory": 25071, "step": 800} +{"lr": 0.0008, "data_time": 0.001605057716369629, "loss": 0.42642366886138916, "time": 0.7728655576705933, "epoch": 1, "memory": 25071, "step": 900} +{"lr": 0.0008, "data_time": 0.15106136798858644, "loss": 0.4032673418521881, "time": 0.8703392505645752, "epoch": 1, "memory": 25071, "step": 1000} +{"lr": 0.0008, "data_time": 0.08912608623504639, "loss": 0.40774006843566896, "time": 0.7977385759353638, "epoch": 1, "memory": 25071, "step": 1100} +{"lr": 0.0008, "data_time": 0.06785049438476562, "loss": 0.38488052487373353, "time": 0.7864893436431885, "epoch": 1, "memory": 25071, "step": 1200} +{"lr": 0.0008, "data_time": 0.0016669750213623047, "loss": 0.41173230707645414, "time": 0.7081609010696411, "epoch": 1, "memory": 25071, "step": 1300} +{"lr": 0.0008, "data_time": 0.06563777923583984, "loss": 0.37856956124305724, "time": 0.7875759124755859, "epoch": 1, "memory": 25071, "step": 1400} +{"lr": 0.0008, "data_time": 0.04978952407836914, "loss": 0.35707377791404726, "time": 0.7544897794723511, "epoch": 1, "memory": 25071, "step": 1500} +{"lr": 0.0008, "data_time": 0.07696795463562012, "loss": 0.3464987128973007, "time": 0.7990949869155883, "epoch": 1, "memory": 25071, "step": 1600} +{"lr": 0.0008, "data_time": 0.0022781848907470702, "loss": 0.3371317148208618, "time": 0.7818030834197998, "epoch": 1, "memory": 25071, "step": 1700} +{"lr": 0.0008, "data_time": 0.014308905601501465, "loss": 0.36208291053771974, "time": 0.7372144937515259, "epoch": 1, "memory": 25071, "step": 1800} +{"lr": 0.0008, "data_time": 0.002062225341796875, "loss": 0.3514672130346298, "time": 0.7870503187179565, "epoch": 1, "memory": 25071, "step": 1900} +{"lr": 0.0008, "data_time": 0.0019228935241699218, "loss": 0.36156936585903166, "time": 0.7589655876159668, "epoch": 1, "memory": 25071, "step": 2000} +{"lr": 0.0008, "data_time": 0.0012876033782958985, "loss": 0.32534425854682925, "time": 0.7730360507965088, "epoch": 1, "memory": 25071, "step": 2100} +{"lr": 0.0008, "data_time": 0.001839447021484375, "loss": 0.31942436695098875, "time": 0.7479748249053955, "epoch": 1, "memory": 25071, "step": 2200} +{"lr": 0.0008, "data_time": 0.0014813661575317383, "loss": 0.31683720350265504, "time": 0.7312279462814331, "epoch": 1, "memory": 25071, "step": 2300} +{"lr": 0.0008, "data_time": 0.0021446704864501952, "loss": 0.30001647770404816, "time": 0.7679721593856812, "epoch": 1, "memory": 25071, "step": 2400} +{"lr": 0.0008, "data_time": 0.0015379667282104492, "loss": 0.3225553035736084, "time": 0.7475534915924072, "epoch": 1, "memory": 25071, "step": 2500} +{"lr": 0.0008, "data_time": 0.0014340877532958984, "loss": 0.3157380446791649, "time": 0.7302091360092163, "epoch": 1, "memory": 25071, "step": 2600} +{"lr": 0.0008, "data_time": 0.001423811912536621, "loss": 0.26995800733566283, "time": 0.798098087310791, "epoch": 1, "memory": 25071, "step": 2700} +{"lr": 0.0008, "data_time": 0.0012891769409179687, "loss": 0.2793251246213913, "time": 0.7392258167266845, "epoch": 1, "memory": 25071, "step": 2800} +{"lr": 0.0008, "data_time": 0.0019237756729125976, "loss": 0.29729904383420946, "time": 0.7564581871032715, "epoch": 1, "memory": 25071, "step": 2900} +{"lr": 0.0008, "data_time": 0.0013764381408691406, "loss": 0.2973321720957756, "time": 0.7987517356872559, "epoch": 1, "memory": 25071, "step": 3000} +{"lr": 0.0008, "data_time": 0.0014077186584472655, "loss": 0.25674239099025725, "time": 0.7925383806228637, "epoch": 1, "memory": 25071, "step": 3100} +{"lr": 0.0008, "data_time": 0.0017213106155395507, "loss": 0.22798804491758345, "time": 0.7815066576004028, "epoch": 1, "memory": 25071, "step": 3200} +{"lr": 0.0008, "data_time": 0.0013444662094116212, "loss": 0.24631092995405196, "time": 0.738797926902771, "epoch": 1, "memory": 25071, "step": 3300} +{"lr": 0.0008, "data_time": 0.08044974803924561, "loss": 0.2532701015472412, "time": 0.7858518123626709, "epoch": 1, "memory": 25071, "step": 3400} +{"lr": 0.0008, "data_time": 0.09921200275421142, "loss": 0.2438528597354889, "time": 0.8053362131118774, "epoch": 1, "memory": 25071, "step": 3500} +{"lr": 0.0008, "data_time": 0.04121699333190918, "loss": 0.23169133365154265, "time": 0.7468256711959839, "epoch": 1, "memory": 25071, "step": 3600} +{"lr": 0.0008, "data_time": 0.049419617652893065, "loss": 0.22192611992359162, "time": 0.756129240989685, "epoch": 1, "memory": 25071, "step": 3700} +{"lr": 0.0008, "data_time": 0.0016196489334106446, "loss": 0.23181588351726531, "time": 0.7988584041595459, "epoch": 1, "memory": 25071, "step": 3800} +{"lr": 0.0008, "data_time": 0.0018055438995361328, "loss": 0.22034767866134644, "time": 0.7792217016220093, "epoch": 1, "memory": 25071, "step": 3900} +{"lr": 0.0008, "data_time": 0.0013665199279785157, "loss": 0.23661839216947556, "time": 0.7415711879730225, "epoch": 1, "memory": 25071, "step": 4000} +{"lr": 0.0008, "data_time": 0.0054359912872314455, "loss": 0.2334280788898468, "time": 0.7654765367507934, "epoch": 1, "memory": 25071, "step": 4100} +{"lr": 0.0008, "data_time": 0.0012468576431274414, "loss": 0.19178823679685592, "time": 0.7288877487182617, "epoch": 1, "memory": 25071, "step": 4200} +{"lr": 0.0008, "data_time": 0.0530895471572876, "loss": 0.21510777920484542, "time": 0.7665005922317505, "epoch": 1, "memory": 25071, "step": 4300} +{"lr": 0.0008, "data_time": 0.00139617919921875, "loss": 0.20133443921804428, "time": 0.7699970483779908, "epoch": 1, "memory": 25071, "step": 4400} +{"lr": 0.0008, "data_time": 0.028180885314941406, "loss": 0.19917598515748977, "time": 0.7384316682815552, "epoch": 1, "memory": 25071, "step": 4500} +{"lr": 0.0008, "data_time": 0.05180978775024414, "loss": 0.19444433748722076, "time": 0.7599361658096313, "epoch": 1, "memory": 25071, "step": 4600} +{"lr": 0.0008, "data_time": 0.048706555366516115, "loss": 0.20219548791646957, "time": 0.7711160182952881, "epoch": 1, "memory": 25071, "step": 4700} +{"lr": 0.0008, "data_time": 0.08586246967315674, "loss": 0.1862374499440193, "time": 0.7907384872436524, "epoch": 1, "memory": 25071, "step": 4800} +{"lr": 0.0008, "data_time": 0.0014488935470581056, "loss": 0.1750187024474144, "time": 0.7507224798202514, "epoch": 1, "memory": 25071, "step": 4900} +{"lr": 0.0008, "data_time": 0.01899867057800293, "loss": 0.16733268201351165, "time": 0.7253173828125, "epoch": 1, "memory": 25071, "step": 5000} +{"lr": 0.0008, "data_time": 0.008738946914672852, "loss": 0.18044504523277283, "time": 0.7870842218399048, "epoch": 1, "memory": 25071, "step": 5100} +{"lr": 0.0008, "data_time": 0.001414036750793457, "loss": 0.165419402718544, "time": 0.7413724422454834, "epoch": 1, "memory": 25071, "step": 5200} +{"lr": 0.0008, "data_time": 0.0014182567596435548, "loss": 0.1610153131186962, "time": 0.7748068571090698, "epoch": 1, "memory": 25071, "step": 5300} +{"lr": 0.0008, "data_time": 0.11126623153686524, "loss": 0.16909749209880828, "time": 0.8160391569137573, "epoch": 1, "memory": 25071, "step": 5400} +{"lr": 0.0008, "data_time": 0.031126642227172853, "loss": 0.15554608553647994, "time": 0.745137619972229, "epoch": 1, "memory": 25071, "step": 5500} +{"lr": 0.0008, "data_time": 0.06271398067474365, "loss": 0.17844616919755935, "time": 0.7679157257080078, "epoch": 1, "memory": 25071, "step": 5600} +{"lr": 0.0008, "data_time": 0.0016042947769165038, "loss": 0.16791197061538696, "time": 0.7157271862030029, "epoch": 1, "memory": 25071, "step": 5700} +{"accuracy/top1": 86.11830139160156, "data_time": 0.27462947368621826, "time": 0.7906721234321594, "step": 1} +{"lr": 0.0007806673239365858, "data_time": 0.0012798070907592773, "loss": 0.1395976021885872, "time": 0.746979832649231, "epoch": 2, "memory": 25074, "step": 5858} +{"lr": 0.0007806673239365858, "data_time": 0.0015470743179321288, "loss": 0.14222779422998427, "time": 0.7680478811264038, "epoch": 2, "memory": 25074, "step": 5958} +{"lr": 0.0007806673239365858, "data_time": 0.11526215076446533, "loss": 0.14738767072558404, "time": 0.8219593048095704, "epoch": 2, "memory": 25074, "step": 6058} +{"lr": 0.0007806673239365858, "data_time": 0.0786397933959961, "loss": 0.14101561307907104, "time": 0.7834032535552978, "epoch": 2, "memory": 25074, "step": 6158} +{"lr": 0.0007806673239365858, "data_time": 0.05114147663116455, "loss": 0.127545128762722, "time": 0.7688297510147095, "epoch": 2, "memory": 25074, "step": 6258} +{"lr": 0.0007806673239365858, "data_time": 0.051855945587158205, "loss": 0.13338923081755638, "time": 0.7567293405532837, "epoch": 2, "memory": 25074, "step": 6358} +{"lr": 0.0007806673239365858, "data_time": 0.12313151359558105, "loss": 0.16329557150602342, "time": 0.8402329683303833, "epoch": 2, "memory": 25074, "step": 6458} +{"lr": 0.0007806673239365858, "data_time": 0.05999410152435303, "loss": 0.13960904777050018, "time": 0.7650542736053467, "epoch": 2, "memory": 25074, "step": 6558} +{"lr": 0.0007806673239365858, "data_time": 0.03151988983154297, "loss": 0.11943572163581848, "time": 0.757231879234314, "epoch": 2, "memory": 25074, "step": 6658} +{"lr": 0.0007806673239365858, "data_time": 0.05698392391204834, "loss": 0.1213602788746357, "time": 0.7635266304016113, "epoch": 2, "memory": 25074, "step": 6758} +{"lr": 0.0007806673239365858, "data_time": 0.08698782920837403, "loss": 0.10986629128456116, "time": 0.8109496116638184, "epoch": 2, "memory": 25074, "step": 6858} +{"lr": 0.0007806673239365858, "data_time": 0.050164651870727536, "loss": 0.14670346677303314, "time": 0.7557155847549438, "epoch": 2, "memory": 25074, "step": 6958} +{"lr": 0.0007806673239365858, "data_time": 0.049664640426635744, "loss": 0.1171759121119976, "time": 0.7815929174423217, "epoch": 2, "memory": 25074, "step": 7058} +{"lr": 0.0007806673239365858, "data_time": 0.0514491081237793, "loss": 0.11087181940674781, "time": 0.760224986076355, "epoch": 2, "memory": 25074, "step": 7158} +{"lr": 0.0007806673239365858, "data_time": 0.08852627277374267, "loss": 0.1281448319554329, "time": 0.8089758872985839, "epoch": 2, "memory": 25074, "step": 7258} +{"lr": 0.0007806673239365858, "data_time": 0.051537489891052245, "loss": 0.08914934396743775, "time": 0.7567667484283447, "epoch": 2, "memory": 25074, "step": 7358} +{"lr": 0.0007806673239365858, "data_time": 0.0895998477935791, "loss": 0.1266442857682705, "time": 0.81229567527771, "epoch": 2, "memory": 25074, "step": 7458} +{"lr": 0.0007806673239365858, "data_time": 0.06785578727722168, "loss": 0.09124645218253136, "time": 0.7752837896347046, "epoch": 2, "memory": 25074, "step": 7558} +{"lr": 0.0007806673239365858, "data_time": 0.06337199211120606, "loss": 0.08968169614672661, "time": 0.8053130388259888, "epoch": 2, "memory": 25074, "step": 7658} +{"lr": 0.0007806673239365858, "data_time": 0.09980368614196777, "loss": 0.09525418058037757, "time": 0.80571448802948, "epoch": 2, "memory": 25074, "step": 7758} +{"lr": 0.0007806673239365858, "data_time": 0.0830761432647705, "loss": 0.10288258828222752, "time": 0.7975805282592774, "epoch": 2, "memory": 25074, "step": 7858} +{"lr": 0.0007806673239365858, "data_time": 0.05366334915161133, "loss": 0.11542944610118866, "time": 0.7596214056015015, "epoch": 2, "memory": 25074, "step": 7958} +{"lr": 0.0007806673239365858, "data_time": 0.11788253784179688, "loss": 0.08477553986012935, "time": 0.8365232229232789, "epoch": 2, "memory": 25074, "step": 8058} +{"lr": 0.0007806673239365858, "data_time": 0.0503159761428833, "loss": 0.09394867718219757, "time": 0.7549262523651123, "epoch": 2, "memory": 25074, "step": 8158} +{"lr": 0.0007806673239365858, "data_time": 0.035231494903564455, "loss": 0.08703736923635005, "time": 0.7555581569671631, "epoch": 2, "memory": 25074, "step": 8258} +{"lr": 0.0007806673239365858, "data_time": 0.10597724914550781, "loss": 0.09405033588409424, "time": 0.8121357917785644, "epoch": 2, "memory": 25074, "step": 8358} +{"lr": 0.0007806673239365858, "data_time": 0.061185598373413086, "loss": 0.09777909182012082, "time": 0.7826323509216309, "epoch": 2, "memory": 25074, "step": 8458} +{"lr": 0.0007806673239365858, "data_time": 0.028299427032470702, "loss": 0.12923009917140008, "time": 0.7326319694519043, "epoch": 2, "memory": 25074, "step": 8558} +{"lr": 0.0007806673239365858, "data_time": 0.12724955081939698, "loss": 0.08197196312248707, "time": 0.8346829175949096, "epoch": 2, "memory": 25074, "step": 8658} +{"lr": 0.0007806673239365858, "data_time": 0.029903173446655273, "loss": 0.09399201795458793, "time": 0.73518807888031, "epoch": 2, "memory": 25074, "step": 8758} +{"lr": 0.0007806673239365858, "data_time": 0.024277424812316893, "loss": 0.07302029766142368, "time": 0.7567739963531495, "epoch": 2, "memory": 25074, "step": 8858} +{"lr": 0.0007806673239365858, "data_time": 0.04628999233245849, "loss": 0.07228384427726268, "time": 0.7520631551742554, "epoch": 2, "memory": 25074, "step": 8958} +{"lr": 0.0007806673239365858, "data_time": 0.06505131721496582, "loss": 0.05472337082028389, "time": 0.770245361328125, "epoch": 2, "memory": 25074, "step": 9058} +{"lr": 0.0007806673239365858, "data_time": 0.06804196834564209, "loss": 0.10189352296292782, "time": 0.7728426933288575, "epoch": 2, "memory": 25074, "step": 9158} +{"lr": 0.0007806673239365858, "data_time": 0.010705995559692382, "loss": 0.06046130508184433, "time": 0.7218541622161865, "epoch": 2, "memory": 25074, "step": 9258} +{"lr": 0.0007806673239365858, "data_time": 0.037660741806030275, "loss": 0.08741049394011498, "time": 0.7417712450027466, "epoch": 2, "memory": 25074, "step": 9358} +{"lr": 0.0007806673239365858, "data_time": 0.0012258291244506836, "loss": 0.06973813138902188, "time": 0.7826901912689209, "epoch": 2, "memory": 25074, "step": 9458} +{"lr": 0.0007806673239365858, "data_time": 0.0012144088745117188, "loss": 0.061193672940135, "time": 0.7893070936203003, "epoch": 2, "memory": 25074, "step": 9558} +{"lr": 0.0007806673239365858, "data_time": 0.0010945558547973632, "loss": 0.06761947125196457, "time": 0.7257037162780762, "epoch": 2, "memory": 25074, "step": 9658} +{"lr": 0.0007806673239365858, "data_time": 0.0016387939453125, "loss": 0.06786825507879257, "time": 0.7165427684783936, "epoch": 2, "memory": 25074, "step": 9758} +{"lr": 0.0007806673239365858, "data_time": 0.0010945320129394532, "loss": 0.06125165410339832, "time": 0.8063461065292359, "epoch": 2, "memory": 25074, "step": 9858} +{"lr": 0.0007806673239365858, "data_time": 0.001276540756225586, "loss": 0.050140045769512656, "time": 0.7595997810363769, "epoch": 2, "memory": 25074, "step": 9958} +{"lr": 0.0007806673239365858, "data_time": 0.0011738300323486327, "loss": 0.05405156258493662, "time": 0.7570838928222656, "epoch": 2, "memory": 25074, "step": 10058} +{"lr": 0.0007806673239365858, "data_time": 0.001383829116821289, "loss": 0.05630311444401741, "time": 0.777427077293396, "epoch": 2, "memory": 25074, "step": 10158} +{"lr": 0.0007806673239365858, "data_time": 0.001216292381286621, "loss": 0.0726040281355381, "time": 0.8589967489242554, "epoch": 2, "memory": 25074, "step": 10258} +{"lr": 0.0007806673239365858, "data_time": 0.0011072874069213868, "loss": 0.06845370680093765, "time": 0.8274120330810547, "epoch": 2, "memory": 25074, "step": 10358} +{"lr": 0.0007806673239365858, "data_time": 0.0013106584548950196, "loss": 0.05654991827905178, "time": 0.7579174995422363, "epoch": 2, "memory": 25074, "step": 10458} +{"lr": 0.0007806673239365858, "data_time": 0.0016683578491210938, "loss": 0.05133899617940187, "time": 0.7490403890609741, "epoch": 2, "memory": 25074, "step": 10558} +{"lr": 0.0007806673239365858, "data_time": 0.001333904266357422, "loss": 0.057648342847824094, "time": 0.764176607131958, "epoch": 2, "memory": 25074, "step": 10658} +{"lr": 0.0007806673239365858, "data_time": 0.0013494014739990235, "loss": 0.08061864003539085, "time": 0.8177014827728272, "epoch": 2, "memory": 25074, "step": 10758} +{"lr": 0.0007806673239365858, "data_time": 0.029112839698791505, "loss": 0.04146339111030102, "time": 0.8291461229324341, "epoch": 2, "memory": 25074, "step": 10858} +{"lr": 0.0007806673239365858, "data_time": 0.0011493921279907226, "loss": 0.0886738296598196, "time": 0.7861827373504638, "epoch": 2, "memory": 25074, "step": 10958} +{"lr": 0.0007806673239365858, "data_time": 0.0011094570159912109, "loss": 0.06363021060824395, "time": 0.8145726919174194, "epoch": 2, "memory": 25074, "step": 11058} +{"lr": 0.0007806673239365858, "data_time": 0.001296830177307129, "loss": 0.04318107292056084, "time": 0.7690038681030273, "epoch": 2, "memory": 25074, "step": 11158} +{"lr": 0.0007806673239365858, "data_time": 0.0024667978286743164, "loss": 0.039654660411179064, "time": 0.7918517112731933, "epoch": 2, "memory": 25074, "step": 11258} +{"lr": 0.0007806673239365858, "data_time": 0.0011334180831909179, "loss": 0.030768837593495846, "time": 0.7723324537277222, "epoch": 2, "memory": 25074, "step": 11358} +{"lr": 0.0007806673239365858, "data_time": 0.001248335838317871, "loss": 0.03774199243634939, "time": 0.7432373762130737, "epoch": 2, "memory": 25074, "step": 11458} +{"accuracy/top1": 76.77545166015625, "data_time": 0.21405748759998994, "time": 0.6063573220196892, "step": 2} +{"lr": 0.0007245617127781043, "data_time": 0.1599264144897461, "loss": 0.04088939521461725, "time": 0.8642370462417602, "epoch": 3, "memory": 25074, "step": 11616} +{"lr": 0.0007245617127781043, "data_time": 0.001267719268798828, "loss": 0.040835813619196415, "time": 0.8656621456146241, "epoch": 3, "memory": 25074, "step": 11716} +{"lr": 0.0007245617127781043, "data_time": 0.001159811019897461, "loss": 0.048896664567291734, "time": 0.7798550128936768, "epoch": 3, "memory": 25074, "step": 11816} +{"lr": 0.0007245617127781043, "data_time": 0.001351618766784668, "loss": 0.03867869433015585, "time": 0.8321730613708496, "epoch": 3, "memory": 25074, "step": 11916} +{"lr": 0.0007245617127781043, "data_time": 0.0014122247695922852, "loss": 0.04783026725053787, "time": 0.7364333868026733, "epoch": 3, "memory": 25074, "step": 12016} +{"lr": 0.0007245617127781043, "data_time": 0.001270461082458496, "loss": 0.04759179260581732, "time": 0.8261472225189209, "epoch": 3, "memory": 25074, "step": 12116} +{"lr": 0.0007245617127781043, "data_time": 0.001415562629699707, "loss": 0.037141728401184085, "time": 0.7853118658065796, "epoch": 3, "memory": 25074, "step": 12216} +{"lr": 0.0007245617127781043, "data_time": 0.0012857437133789063, "loss": 0.039891570992767814, "time": 0.7424207925796509, "epoch": 3, "memory": 25074, "step": 12316} +{"lr": 0.0007245617127781043, "data_time": 0.0014500379562377929, "loss": 0.03392054084688425, "time": 0.7254342555999755, "epoch": 3, "memory": 25074, "step": 12416} +{"lr": 0.0007245617127781043, "data_time": 0.0012304544448852538, "loss": 0.03665917366743088, "time": 0.7534410715103149, "epoch": 3, "memory": 25074, "step": 12516} +{"lr": 0.0007245617127781043, "data_time": 0.0012962818145751953, "loss": 0.040328314155340196, "time": 0.7487728595733643, "epoch": 3, "memory": 25074, "step": 12616} +{"lr": 0.0007245617127781043, "data_time": 0.0013105392456054688, "loss": 0.03161096451804042, "time": 0.7518738031387329, "epoch": 3, "memory": 25074, "step": 12716} +{"lr": 0.0007245617127781043, "data_time": 0.0012244939804077148, "loss": 0.04508720170706511, "time": 0.7207002878189087, "epoch": 3, "memory": 25074, "step": 12816} +{"lr": 0.0007245617127781043, "data_time": 0.0013757705688476562, "loss": 0.026851135678589343, "time": 0.717484426498413, "epoch": 3, "memory": 25074, "step": 12916} +{"lr": 0.0007245617127781043, "data_time": 0.0012349367141723632, "loss": 0.04733375571668148, "time": 0.7747965097427368, "epoch": 3, "memory": 25074, "step": 13016} +{"lr": 0.0007245617127781043, "data_time": 0.0012233495712280274, "loss": 0.02905960977077484, "time": 0.7289293766021728, "epoch": 3, "memory": 25074, "step": 13116} +{"lr": 0.0007245617127781043, "data_time": 0.0015436887741088867, "loss": 0.030410025874152778, "time": 0.7665880441665649, "epoch": 3, "memory": 25074, "step": 13216} +{"lr": 0.0007245617127781043, "data_time": 0.0012766122817993164, "loss": 0.027036966988816857, "time": 0.8024507284164428, "epoch": 3, "memory": 25074, "step": 13316} +{"lr": 0.0007245617127781043, "data_time": 0.0011659622192382812, "loss": 0.048966099973767996, "time": 0.7735217332839965, "epoch": 3, "memory": 25074, "step": 13416} +{"lr": 0.0007245617127781043, "data_time": 0.0013410091400146485, "loss": 0.030625978391617535, "time": 0.7426556587219239, "epoch": 3, "memory": 25074, "step": 13516} +{"lr": 0.0007245617127781043, "data_time": 0.0013001203536987306, "loss": 0.0330763032194227, "time": 0.7913923263549805, "epoch": 3, "memory": 25074, "step": 13616} +{"lr": 0.0007245617127781043, "data_time": 0.0013531684875488282, "loss": 0.028288222942501308, "time": 0.7605397462844848, "epoch": 3, "memory": 25074, "step": 13716} +{"lr": 0.0007245617127781043, "data_time": 0.0018346309661865234, "loss": 0.03461533999070525, "time": 0.7678704500198364, "epoch": 3, "memory": 25074, "step": 13816} +{"lr": 0.0007245617127781043, "data_time": 0.0014458656311035155, "loss": 0.02867770232260227, "time": 0.7836845397949219, "epoch": 3, "memory": 25074, "step": 13916} +{"lr": 0.0007245617127781043, "data_time": 0.0011763334274291991, "loss": 0.024216140899807215, "time": 0.7715628147125244, "epoch": 3, "memory": 25074, "step": 14016} +{"lr": 0.0007245617127781043, "data_time": 0.0012571573257446288, "loss": 0.03660787008702755, "time": 0.8137492656707763, "epoch": 3, "memory": 25074, "step": 14116} +{"lr": 0.0007245617127781043, "data_time": 0.001326131820678711, "loss": 0.0303189299069345, "time": 0.7664277076721191, "epoch": 3, "memory": 25074, "step": 14216} +{"lr": 0.0007245617127781043, "data_time": 0.001418900489807129, "loss": 0.03185373768210411, "time": 0.8259416103363038, "epoch": 3, "memory": 25074, "step": 14316} +{"lr": 0.0007245617127781043, "data_time": 0.001248311996459961, "loss": 0.02666334230452776, "time": 0.7289786815643311, "epoch": 3, "memory": 25074, "step": 14416} +{"lr": 0.0007245617127781043, "data_time": 0.0013849496841430663, "loss": 0.037510369904339314, "time": 0.7737699747085571, "epoch": 3, "memory": 25074, "step": 14516} +{"lr": 0.0007245617127781043, "data_time": 0.0010577678680419923, "loss": 0.03314679367467761, "time": 0.7594129323959351, "epoch": 3, "memory": 25074, "step": 14616} +{"lr": 0.0007245617127781043, "data_time": 0.0012985706329345704, "loss": 0.031005522515624762, "time": 0.8288192987442017, "epoch": 3, "memory": 25074, "step": 14716} +{"lr": 0.0007245617127781043, "data_time": 0.0012567996978759765, "loss": 0.034702819958329204, "time": 0.7696936845779419, "epoch": 3, "memory": 25074, "step": 14816} +{"lr": 0.0007245617127781043, "data_time": 0.0014369964599609375, "loss": 0.025255820620805024, "time": 0.7652963399887085, "epoch": 3, "memory": 25074, "step": 14916} +{"lr": 0.0007245617127781043, "data_time": 0.0017720937728881836, "loss": 0.030143201537430286, "time": 0.761293911933899, "epoch": 3, "memory": 25074, "step": 15016} +{"lr": 0.0007245617127781043, "data_time": 0.0013030290603637696, "loss": 0.021141273621469735, "time": 0.7685921907424926, "epoch": 3, "memory": 25074, "step": 15116} +{"lr": 0.0007245617127781043, "data_time": 0.0012258768081665039, "loss": 0.017992442473769188, "time": 0.7396396636962891, "epoch": 3, "memory": 25074, "step": 15216} +{"lr": 0.0007245617127781043, "data_time": 0.001431441307067871, "loss": 0.023397974809631704, "time": 0.8045117616653442, "epoch": 3, "memory": 25074, "step": 15316} +{"lr": 0.0007245617127781043, "data_time": 0.0012537240982055664, "loss": 0.03260563835501671, "time": 0.7762276887893677, "epoch": 3, "memory": 25074, "step": 15416} +{"lr": 0.0007245617127781043, "data_time": 0.0012106895446777344, "loss": 0.023786334693431853, "time": 0.7057984113693238, "epoch": 3, "memory": 25074, "step": 15516} +{"lr": 0.0007245617127781043, "data_time": 0.0011429548263549804, "loss": 0.034362148866057395, "time": 0.7490205287933349, "epoch": 3, "memory": 25074, "step": 15616} +{"lr": 0.0007245617127781043, "data_time": 0.0012099027633666992, "loss": 0.018229505466297268, "time": 0.7419039726257324, "epoch": 3, "memory": 25074, "step": 15716} +{"lr": 0.0007245617127781043, "data_time": 0.0019709110260009766, "loss": 0.018690613843500615, "time": 0.7349676370620728, "epoch": 3, "memory": 25074, "step": 15816} +{"lr": 0.0007245617127781043, "data_time": 0.0013330459594726562, "loss": 0.017513964604586364, "time": 0.7214994668960572, "epoch": 3, "memory": 25074, "step": 15916} +{"lr": 0.0007245617127781043, "data_time": 0.001301121711730957, "loss": 0.028252078453078866, "time": 0.7291983604431153, "epoch": 3, "memory": 25074, "step": 16016} +{"lr": 0.0007245617127781043, "data_time": 0.001300644874572754, "loss": 0.02576492396183312, "time": 0.7864515542984009, "epoch": 3, "memory": 25074, "step": 16116} +{"lr": 0.0007245617127781043, "data_time": 0.001355290412902832, "loss": 0.027555054612457754, "time": 0.710329246520996, "epoch": 3, "memory": 25074, "step": 16216} +{"lr": 0.0007245617127781043, "data_time": 0.0011693954467773438, "loss": 0.026239687763154506, "time": 0.7224598169326782, "epoch": 3, "memory": 25074, "step": 16316} +{"lr": 0.0007245617127781043, "data_time": 0.0013732671737670898, "loss": 0.021136733843013643, "time": 0.7503315448760987, "epoch": 3, "memory": 25074, "step": 16416} +{"lr": 0.0007245617127781043, "data_time": 0.001243901252746582, "loss": 0.024915443174540995, "time": 0.7485300302505493, "epoch": 3, "memory": 25074, "step": 16516} +{"lr": 0.0007245617127781043, "data_time": 0.0013096094131469726, "loss": 0.02882102965377271, "time": 0.816267991065979, "epoch": 3, "memory": 25074, "step": 16616} +{"lr": 0.0007245617127781043, "data_time": 0.0010734796524047852, "loss": 0.01705039106309414, "time": 0.7063563823699951, "epoch": 3, "memory": 25074, "step": 16716} +{"lr": 0.0007245617127781043, "data_time": 0.001378631591796875, "loss": 0.02329266774468124, "time": 0.781953239440918, "epoch": 3, "memory": 25074, "step": 16816} +{"lr": 0.0007245617127781043, "data_time": 0.0013356924057006836, "loss": 0.024129189271479844, "time": 0.7305824756622314, "epoch": 3, "memory": 25074, "step": 16916} +{"lr": 0.0007245617127781043, "data_time": 0.0014432430267333984, "loss": 0.015659291297197342, "time": 0.7203398704528808, "epoch": 3, "memory": 25074, "step": 17016} +{"lr": 0.0007245617127781043, "data_time": 0.0016174554824829102, "loss": 0.02788719804957509, "time": 0.7511308193206787, "epoch": 3, "memory": 25074, "step": 17116} +{"lr": 0.0007245617127781043, "data_time": 0.0011986494064331055, "loss": 0.021045492310076953, "time": 0.7102416753768921, "epoch": 3, "memory": 25074, "step": 17216} +{"accuracy/top1": 96.7741928100586, "data_time": 0.2159565196317785, "time": 0.6016581899979535, "step": 3} +{"lr": 0.0006371751746555269, "data_time": 0.0016854047775268556, "loss": 0.02584357662126422, "time": 0.7935037851333618, "epoch": 4, "memory": 25074, "step": 17374} +{"lr": 0.0006371751746555269, "data_time": 0.0013775825500488281, "loss": 0.015903302934020758, "time": 0.7569707632064819, "epoch": 4, "memory": 25074, "step": 17474} +{"lr": 0.0006371751746555269, "data_time": 0.0012466907501220703, "loss": 0.022549204388633372, "time": 0.7929486989974975, "epoch": 4, "memory": 25074, "step": 17574} +{"lr": 0.0006371751746555269, "data_time": 0.0013414859771728516, "loss": 0.027700148802250624, "time": 0.739317512512207, "epoch": 4, "memory": 25074, "step": 17674} +{"lr": 0.0006371751746555269, "data_time": 0.0014379501342773437, "loss": 0.018095119623467326, "time": 0.815676498413086, "epoch": 4, "memory": 25074, "step": 17774} +{"lr": 0.0006371751746555269, "data_time": 0.0014359712600708007, "loss": 0.01641491292975843, "time": 0.7535361051559448, "epoch": 4, "memory": 25074, "step": 17874} +{"lr": 0.0006371751746555269, "data_time": 0.001419830322265625, "loss": 0.016665436886250974, "time": 0.7647857666015625, "epoch": 4, "memory": 25074, "step": 17974} +{"lr": 0.0006371751746555269, "data_time": 0.0015505552291870117, "loss": 0.021457855915650727, "time": 0.7416374444961548, "epoch": 4, "memory": 25074, "step": 18074} +{"lr": 0.0006371751746555269, "data_time": 0.0013824701309204102, "loss": 0.023184484895318748, "time": 0.7078312635421753, "epoch": 4, "memory": 25074, "step": 18174} +{"lr": 0.0006371751746555269, "data_time": 0.0014255046844482422, "loss": 0.019590003369376063, "time": 0.7725302457809449, "epoch": 4, "memory": 25074, "step": 18274} +{"lr": 0.0006371751746555269, "data_time": 0.0013128757476806641, "loss": 0.016926329070702194, "time": 0.7734216690063477, "epoch": 4, "memory": 25074, "step": 18374} +{"lr": 0.0006371751746555269, "data_time": 0.0014129638671875, "loss": 0.022653862461447717, "time": 0.741803240776062, "epoch": 4, "memory": 25074, "step": 18474} +{"lr": 0.0006371751746555269, "data_time": 0.0012583017349243164, "loss": 0.012973507773131132, "time": 0.7640147924423217, "epoch": 4, "memory": 25074, "step": 18574} +{"lr": 0.0006371751746555269, "data_time": 0.0012765645980834961, "loss": 0.022019442077726124, "time": 0.7581893920898437, "epoch": 4, "memory": 25074, "step": 18674} +{"lr": 0.0006371751746555269, "data_time": 0.0012765884399414062, "loss": 0.012811435712501406, "time": 0.7138432025909424, "epoch": 4, "memory": 25074, "step": 18774} +{"lr": 0.0006371751746555269, "data_time": 0.0012908458709716796, "loss": 0.01544500607997179, "time": 0.7485158920288086, "epoch": 4, "memory": 25074, "step": 18874} +{"lr": 0.0006371751746555269, "data_time": 0.0013615846633911132, "loss": 0.01228244355879724, "time": 0.7062382221221923, "epoch": 4, "memory": 25074, "step": 18974} +{"lr": 0.0006371751746555269, "data_time": 0.0014231443405151368, "loss": 0.010163441905751825, "time": 0.7450262308120728, "epoch": 4, "memory": 25074, "step": 19074} +{"lr": 0.0006371751746555269, "data_time": 0.0014815807342529296, "loss": 0.022428280021995307, "time": 0.7605623006820679, "epoch": 4, "memory": 25074, "step": 19174} +{"lr": 0.0006371751746555269, "data_time": 0.0011575698852539062, "loss": 0.017260070750489832, "time": 0.7530862808227539, "epoch": 4, "memory": 25074, "step": 19274} +{"lr": 0.0006371751746555269, "data_time": 0.0013218402862548828, "loss": 0.019637471809983254, "time": 0.7508957862854004, "epoch": 4, "memory": 25074, "step": 19374} +{"lr": 0.0006371751746555269, "data_time": 0.0013967275619506836, "loss": 0.023825343092903496, "time": 0.714510703086853, "epoch": 4, "memory": 25074, "step": 19474} +{"lr": 0.0006371751746555269, "data_time": 0.0012618541717529298, "loss": 0.013557658018544316, "time": 0.7425036907196045, "epoch": 4, "memory": 25074, "step": 19574} +{"lr": 0.0006371751746555269, "data_time": 0.001282811164855957, "loss": 0.017627335991710423, "time": 0.7495125293731689, "epoch": 4, "memory": 25074, "step": 19674} +{"lr": 0.0006371751746555269, "data_time": 0.0013806343078613282, "loss": 0.026325806882232426, "time": 0.7633819818496704, "epoch": 4, "memory": 25074, "step": 19774} +{"lr": 0.0006371751746555269, "data_time": 0.0014236927032470702, "loss": 0.01704902113415301, "time": 0.7887636423110962, "epoch": 4, "memory": 25074, "step": 19874} +{"lr": 0.0006371751746555269, "data_time": 0.0011818408966064453, "loss": 0.018378399033099412, "time": 0.7308815956115723, "epoch": 4, "memory": 25074, "step": 19974} +{"lr": 0.0006371751746555269, "data_time": 0.0013622999191284179, "loss": 0.01764199361205101, "time": 0.7831684827804566, "epoch": 4, "memory": 25074, "step": 20074} +{"lr": 0.0006371751746555269, "data_time": 0.001314258575439453, "loss": 0.014964620536193251, "time": 0.7577922582626343, "epoch": 4, "memory": 25074, "step": 20174} +{"lr": 0.0006371751746555269, "data_time": 0.0014181137084960938, "loss": 0.015263845445588231, "time": 0.9323982238769531, "epoch": 4, "memory": 25074, "step": 20274} +{"lr": 0.0006371751746555269, "data_time": 0.0015181303024291992, "loss": 0.015902172727510332, "time": 0.8992016553878784, "epoch": 4, "memory": 25074, "step": 20374} +{"lr": 0.0006371751746555269, "data_time": 0.001190924644470215, "loss": 0.01186499735340476, "time": 0.871763277053833, "epoch": 4, "memory": 25074, "step": 20474} +{"lr": 0.0006371751746555269, "data_time": 0.0012467384338378906, "loss": 0.0199515497777611, "time": 0.8446225643157959, "epoch": 4, "memory": 25074, "step": 20574} +{"lr": 0.0006371751746555269, "data_time": 0.0011667966842651366, "loss": 0.012517213681712746, "time": 0.7645492553710938, "epoch": 4, "memory": 25074, "step": 20674} +{"lr": 0.0006371751746555269, "data_time": 0.0013451814651489259, "loss": 0.008589001651853324, "time": 0.8372746229171752, "epoch": 4, "memory": 25074, "step": 20774} +{"lr": 0.0006371751746555269, "data_time": 0.0015634775161743164, "loss": 0.010175559064373374, "time": 0.8577198505401611, "epoch": 4, "memory": 25074, "step": 20874} +{"lr": 0.0006371751746555269, "data_time": 0.0013834714889526367, "loss": 0.019995456002652644, "time": 0.7834474802017212, "epoch": 4, "memory": 25074, "step": 20974} +{"lr": 0.0006371751746555269, "data_time": 0.001290607452392578, "loss": 0.01769702658057213, "time": 0.7835808277130127, "epoch": 4, "memory": 25074, "step": 21074} +{"lr": 0.0006371751746555269, "data_time": 0.0013242483139038086, "loss": 0.0172896021977067, "time": 0.8407667875289917, "epoch": 4, "memory": 25074, "step": 21174} +{"lr": 0.0006371751746555269, "data_time": 0.001394939422607422, "loss": 0.012087896093726158, "time": 0.8116055488586426, "epoch": 4, "memory": 25074, "step": 21274} +{"lr": 0.0006371751746555269, "data_time": 0.0015437602996826172, "loss": 0.012474996922537684, "time": 0.7602555751800537, "epoch": 4, "memory": 25074, "step": 21374} +{"lr": 0.0006371751746555269, "data_time": 0.0013099193572998046, "loss": 0.010769471200183034, "time": 0.7567335367202759, "epoch": 4, "memory": 25074, "step": 21474} +{"lr": 0.0006371751746555269, "data_time": 0.0011894941329956056, "loss": 0.014800886064767838, "time": 0.7463791608810425, "epoch": 4, "memory": 25074, "step": 21574} +{"lr": 0.0006371751746555269, "data_time": 0.001492786407470703, "loss": 0.014644334558397532, "time": 0.7918062686920166, "epoch": 4, "memory": 25074, "step": 21674} +{"lr": 0.0006371751746555269, "data_time": 0.0012953996658325196, "loss": 0.011028381902724505, "time": 0.7790021657943725, "epoch": 4, "memory": 25074, "step": 21774} +{"lr": 0.0006371751746555269, "data_time": 0.0012609720230102538, "loss": 0.023543927865102888, "time": 0.7411756038665771, "epoch": 4, "memory": 25074, "step": 21874} +{"lr": 0.0006371751746555269, "data_time": 0.0013468265533447266, "loss": 0.01321612591855228, "time": 0.7579327344894409, "epoch": 4, "memory": 25074, "step": 21974} +{"lr": 0.0006371751746555269, "data_time": 0.0012727022171020509, "loss": 0.011263295263051986, "time": 0.8052274942398071, "epoch": 4, "memory": 25074, "step": 22074} +{"lr": 0.0006371751746555269, "data_time": 0.001506948471069336, "loss": 0.015829464327543973, "time": 0.733405590057373, "epoch": 4, "memory": 25074, "step": 22174} +{"lr": 0.0006371751746555269, "data_time": 0.0014414310455322266, "loss": 0.01200642492622137, "time": 0.7600989580154419, "epoch": 4, "memory": 25074, "step": 22274} +{"lr": 0.0006371751746555269, "data_time": 0.00143740177154541, "loss": 0.01129843876697123, "time": 0.7545924425125122, "epoch": 4, "memory": 25074, "step": 22374} +{"lr": 0.0006371751746555269, "data_time": 0.001237034797668457, "loss": 0.009275696938857436, "time": 0.7790245294570923, "epoch": 4, "memory": 25074, "step": 22474} +{"lr": 0.0006371751746555269, "data_time": 0.001354074478149414, "loss": 0.00458734123967588, "time": 0.8176832437515259, "epoch": 4, "memory": 25074, "step": 22574} +{"lr": 0.0006371751746555269, "data_time": 0.0015718460083007813, "loss": 0.015020606759935617, "time": 0.753489899635315, "epoch": 4, "memory": 25074, "step": 22674} +{"lr": 0.0006371751746555269, "data_time": 0.0012067079544067383, "loss": 0.018114811880514024, "time": 0.8565834283828735, "epoch": 4, "memory": 25074, "step": 22774} +{"lr": 0.0006371751746555269, "data_time": 0.0011778831481933593, "loss": 0.012367798015475273, "time": 0.7794310569763183, "epoch": 4, "memory": 25074, "step": 22874} +{"lr": 0.0006371751746555269, "data_time": 0.0015181779861450195, "loss": 0.017175087658688425, "time": 0.7682755470275879, "epoch": 4, "memory": 25074, "step": 22974} +{"accuracy/top1": 99.00889587402344, "data_time": 0.20151387943941004, "time": 0.5868225658641142, "step": 4} +{"lr": 0.0005270617127781042, "data_time": 0.0012909412384033204, "loss": 0.012921292847022415, "time": 0.7147369384765625, "epoch": 5, "memory": 25074, "step": 23132} +{"lr": 0.0005270617127781042, "data_time": 0.0012772560119628906, "loss": 0.006702626030892134, "time": 0.7589564800262452, "epoch": 5, "memory": 25074, "step": 23232} +{"lr": 0.0005270617127781042, "data_time": 0.001299881935119629, "loss": 0.009381459048017859, "time": 0.7967705726623535, "epoch": 5, "memory": 25074, "step": 23332} +{"lr": 0.0005270617127781042, "data_time": 0.0012833833694458007, "loss": 0.015548907918855548, "time": 0.8214048385620117, "epoch": 5, "memory": 25074, "step": 23432} +{"lr": 0.0005270617127781042, "data_time": 0.0013784646987915039, "loss": 0.006287212623283267, "time": 0.8268746376037598, "epoch": 5, "memory": 25074, "step": 23532} +{"lr": 0.0005270617127781042, "data_time": 0.0012119770050048827, "loss": 0.011349517572671174, "time": 0.7883623838424683, "epoch": 5, "memory": 25074, "step": 23632} +{"lr": 0.0005270617127781042, "data_time": 0.0015107154846191405, "loss": 0.022485012747347356, "time": 0.8699362277984619, "epoch": 5, "memory": 25074, "step": 23732} +{"lr": 0.0005270617127781042, "data_time": 0.001415252685546875, "loss": 0.013932380033656955, "time": 0.8240361213684082, "epoch": 5, "memory": 25074, "step": 23832} +{"lr": 0.0005270617127781042, "data_time": 0.0013387203216552734, "loss": 0.00835984256118536, "time": 0.7652427434921265, "epoch": 5, "memory": 25074, "step": 23932} +{"lr": 0.0005270617127781042, "data_time": 0.0018189191818237306, "loss": 0.00987613033503294, "time": 0.9262356996536255, "epoch": 5, "memory": 25074, "step": 24032} +{"lr": 0.0005270617127781042, "data_time": 0.0014084815979003907, "loss": 0.0059687618631869555, "time": 0.811207628250122, "epoch": 5, "memory": 25074, "step": 24132} +{"lr": 0.0005270617127781042, "data_time": 0.0021492481231689454, "loss": 0.014398828661069275, "time": 0.8130000352859497, "epoch": 5, "memory": 25074, "step": 24232} +{"lr": 0.0005270617127781042, "data_time": 0.0015762805938720702, "loss": 0.010351484594866633, "time": 0.7553416967391968, "epoch": 5, "memory": 25074, "step": 24332} +{"lr": 0.0005270617127781042, "data_time": 0.0014958620071411134, "loss": 0.005105896294116974, "time": 0.7634966850280762, "epoch": 5, "memory": 25074, "step": 24432} +{"lr": 0.0005270617127781042, "data_time": 0.0012889623641967774, "loss": 0.006283015664666891, "time": 0.9195183515548706, "epoch": 5, "memory": 25074, "step": 24532} +{"lr": 0.0005270617127781042, "data_time": 0.0015323638916015625, "loss": 0.009210485918447375, "time": 0.7294370889663696, "epoch": 5, "memory": 25074, "step": 24632} +{"lr": 0.0005270617127781042, "data_time": 0.0014931440353393554, "loss": 0.011729899188503623, "time": 0.7861579656600952, "epoch": 5, "memory": 25074, "step": 24732} +{"lr": 0.0005270617127781042, "data_time": 0.001440143585205078, "loss": 0.009647990157827736, "time": 0.7527569770812989, "epoch": 5, "memory": 25074, "step": 24832} +{"lr": 0.0005270617127781042, "data_time": 0.0013882875442504882, "loss": 0.00918117887340486, "time": 0.8107163906097412, "epoch": 5, "memory": 25074, "step": 24932} +{"lr": 0.0005270617127781042, "data_time": 0.0012422561645507812, "loss": 0.0036405269987881184, "time": 0.7296783208847046, "epoch": 5, "memory": 25074, "step": 25032} +{"lr": 0.0005270617127781042, "data_time": 0.001269960403442383, "loss": 0.00785787650384009, "time": 0.7547079801559449, "epoch": 5, "memory": 25074, "step": 25132} +{"lr": 0.0005270617127781042, "data_time": 0.0017754793167114257, "loss": 0.012282479414716363, "time": 0.7850768089294433, "epoch": 5, "memory": 25074, "step": 25232} +{"lr": 0.0005270617127781042, "data_time": 0.0013004064559936524, "loss": 0.009536367142572999, "time": 0.7426051616668701, "epoch": 5, "memory": 25074, "step": 25332} +{"lr": 0.0005270617127781042, "data_time": 0.0012665987014770508, "loss": 0.00943656899034977, "time": 0.7631567239761352, "epoch": 5, "memory": 25074, "step": 25432} +{"lr": 0.0005270617127781042, "data_time": 0.0015227556228637694, "loss": 0.008347800001502037, "time": 0.786600136756897, "epoch": 5, "memory": 25074, "step": 25532} +{"lr": 0.0005270617127781042, "data_time": 0.0015579462051391602, "loss": 0.012324441689997911, "time": 0.7793980360031127, "epoch": 5, "memory": 25074, "step": 25632} +{"lr": 0.0005270617127781042, "data_time": 0.0014023303985595704, "loss": 0.010182300163432956, "time": 0.822082805633545, "epoch": 5, "memory": 25074, "step": 25732} +{"lr": 0.0005270617127781042, "data_time": 0.0012905359268188476, "loss": 0.008146683033555747, "time": 1.0886083364486694, "epoch": 5, "memory": 25074, "step": 25832} +{"lr": 0.0005270617127781042, "data_time": 0.0012215375900268555, "loss": 0.010026113595813513, "time": 1.1644296884536742, "epoch": 5, "memory": 25074, "step": 25932} +{"lr": 0.0005270617127781042, "data_time": 0.0013471126556396484, "loss": 0.013122609863057733, "time": 1.0125898599624634, "epoch": 5, "memory": 25074, "step": 26032} +{"lr": 0.0005270617127781042, "data_time": 0.0011795997619628907, "loss": 0.012971840659156441, "time": 0.7858582973480225, "epoch": 5, "memory": 25074, "step": 26132} +{"lr": 0.0005270617127781042, "data_time": 0.0013992071151733398, "loss": 0.008178239269182087, "time": 0.8907784700393677, "epoch": 5, "memory": 25074, "step": 26232} +{"lr": 0.0005270617127781042, "data_time": 0.0011201620101928711, "loss": 0.013482456980273127, "time": 0.7799110651016236, "epoch": 5, "memory": 25074, "step": 26332} +{"lr": 0.0005270617127781042, "data_time": 0.0011100530624389648, "loss": 0.011699210526421666, "time": 0.8007047414779663, "epoch": 5, "memory": 25074, "step": 26432} +{"lr": 0.0005270617127781042, "data_time": 0.0015693426132202149, "loss": 0.01425934825092554, "time": 0.8350407600402832, "epoch": 5, "memory": 25074, "step": 26532} +{"lr": 0.0005270617127781042, "data_time": 0.001519489288330078, "loss": 0.00782128032296896, "time": 1.1066365718841553, "epoch": 5, "memory": 25074, "step": 26632} +{"lr": 0.0005270617127781042, "data_time": 0.0012603759765625, "loss": 0.006382914027199149, "time": 1.1303036212921143, "epoch": 5, "memory": 25074, "step": 26732} +{"lr": 0.0005270617127781042, "data_time": 0.001593303680419922, "loss": 0.008885769452899695, "time": 1.0235493898391723, "epoch": 5, "memory": 25074, "step": 26832} +{"lr": 0.0005270617127781042, "data_time": 0.0015812397003173828, "loss": 0.006711181439459324, "time": 1.011578345298767, "epoch": 5, "memory": 25074, "step": 26932} +{"lr": 0.0005270617127781042, "data_time": 0.001276254653930664, "loss": 0.007281673047691584, "time": 1.0732971906661988, "epoch": 5, "memory": 25074, "step": 27032} +{"lr": 0.0005270617127781042, "data_time": 0.0012431859970092774, "loss": 0.007350308913737535, "time": 1.12139573097229, "epoch": 5, "memory": 25074, "step": 27132} +{"lr": 0.0005270617127781042, "data_time": 0.0014218330383300782, "loss": 0.010497211245819926, "time": 0.7100059986114502, "epoch": 5, "memory": 25074, "step": 27232} +{"lr": 0.0005270617127781042, "data_time": 0.0013143062591552735, "loss": 0.006091742077842355, "time": 0.7222891092300415, "epoch": 5, "memory": 25074, "step": 27332} +{"lr": 0.0005270617127781042, "data_time": 0.0014330625534057617, "loss": 0.008059976575896143, "time": 0.7129441738128662, "epoch": 5, "memory": 25074, "step": 27432} +{"lr": 0.0005270617127781042, "data_time": 0.0013603448867797851, "loss": 0.010062522627413273, "time": 0.7286438703536987, "epoch": 5, "memory": 25074, "step": 27532} +{"lr": 0.0005270617127781042, "data_time": 0.001739501953125, "loss": 0.00918293297290802, "time": 0.762654972076416, "epoch": 5, "memory": 25074, "step": 27632} +{"lr": 0.0005270617127781042, "data_time": 0.0014391899108886718, "loss": 0.008797229966148734, "time": 0.7391265630722046, "epoch": 5, "memory": 25074, "step": 27732} +{"lr": 0.0005270617127781042, "data_time": 0.001235818862915039, "loss": 0.008473049057647586, "time": 0.7087532758712769, "epoch": 5, "memory": 25074, "step": 27832} +{"lr": 0.0005270617127781042, "data_time": 0.0016701698303222656, "loss": 0.009205130394548178, "time": 0.7324969530105591, "epoch": 5, "memory": 25074, "step": 27932} +{"lr": 0.0005270617127781042, "data_time": 0.0014713048934936524, "loss": 0.009877536445856094, "time": 0.7609650135040283, "epoch": 5, "memory": 25074, "step": 28032} +{"lr": 0.0005270617127781042, "data_time": 0.001546931266784668, "loss": 0.004014493990689516, "time": 0.7463459014892578, "epoch": 5, "memory": 25074, "step": 28132} +{"lr": 0.0005270617127781042, "data_time": 0.0013782501220703125, "loss": 0.013779939617961645, "time": 0.7689538478851319, "epoch": 5, "memory": 25074, "step": 28232} +{"lr": 0.0005270617127781042, "data_time": 0.0012537002563476562, "loss": 0.006694731302559376, "time": 0.7084328651428222, "epoch": 5, "memory": 25074, "step": 28332} +{"lr": 0.0005270617127781042, "data_time": 0.0015495538711547852, "loss": 0.005180006474256515, "time": 0.7607593536376953, "epoch": 5, "memory": 25074, "step": 28432} +{"lr": 0.0005270617127781042, "data_time": 0.0015459537506103515, "loss": 0.01184030957520008, "time": 0.7731788396835327, "epoch": 5, "memory": 25074, "step": 28532} +{"lr": 0.0005270617127781042, "data_time": 0.0015075922012329102, "loss": 0.005504630692303181, "time": 0.7314815044403076, "epoch": 5, "memory": 25074, "step": 28632} +{"lr": 0.0005270617127781042, "data_time": 0.0018773078918457031, "loss": 0.011145124398171901, "time": 0.802031421661377, "epoch": 5, "memory": 25074, "step": 28732} +{"accuracy/top1": 99.55179595947266, "data_time": 0.2146015587974997, "time": 0.6045234764323515, "step": 5} +{"lr": 0.000405, "data_time": 0.0013513803482055665, "loss": 0.0050857753027230505, "time": 0.7455167293548584, "epoch": 6, "memory": 25074, "step": 28890} +{"lr": 0.000405, "data_time": 0.0016866445541381836, "loss": 0.008379572164267301, "time": 0.8214875936508179, "epoch": 6, "memory": 25074, "step": 28990} +{"lr": 0.000405, "data_time": 0.0013658523559570313, "loss": 0.003400090243667364, "time": 0.7772329568862915, "epoch": 6, "memory": 25074, "step": 29090} +{"lr": 0.000405, "data_time": 0.001150059700012207, "loss": 0.006579025229439139, "time": 0.8252673864364624, "epoch": 6, "memory": 25074, "step": 29190} +{"lr": 0.000405, "data_time": 0.0011378765106201173, "loss": 0.007819813350215555, "time": 0.7901975870132446, "epoch": 6, "memory": 25074, "step": 29290} +{"lr": 0.000405, "data_time": 0.0014597177505493164, "loss": 0.0142287143971771, "time": 0.8013340473175049, "epoch": 6, "memory": 25074, "step": 29390} +{"lr": 0.000405, "data_time": 0.0017203569412231445, "loss": 0.0047938835341483355, "time": 0.7546528577804565, "epoch": 6, "memory": 25074, "step": 29490} +{"lr": 0.000405, "data_time": 0.0010878324508666992, "loss": 0.005435280269011855, "time": 0.7393974542617798, "epoch": 6, "memory": 25074, "step": 29590} +{"lr": 0.000405, "data_time": 0.0012293338775634765, "loss": 0.005714353220537305, "time": 0.7420917272567749, "epoch": 6, "memory": 25074, "step": 29690} +{"lr": 0.000405, "data_time": 0.001352977752685547, "loss": 0.006395082967355847, "time": 0.8012165784835815, "epoch": 6, "memory": 25074, "step": 29790} +{"lr": 0.000405, "data_time": 0.0012380838394165038, "loss": 0.010534848691895604, "time": 0.798853087425232, "epoch": 6, "memory": 25074, "step": 29890} +{"lr": 0.000405, "data_time": 0.0012102603912353515, "loss": 0.008739868039265275, "time": 0.7886104822158814, "epoch": 6, "memory": 25074, "step": 29990} +{"lr": 0.000405, "data_time": 0.0013063907623291015, "loss": 0.010199566651135684, "time": 0.7340371370315552, "epoch": 6, "memory": 25074, "step": 30090} +{"lr": 0.000405, "data_time": 0.001451396942138672, "loss": 0.007939047180116177, "time": 0.7761954069137573, "epoch": 6, "memory": 25074, "step": 30190} +{"lr": 0.000405, "data_time": 0.0017353296279907227, "loss": 0.006712845666334033, "time": 0.7268011331558227, "epoch": 6, "memory": 25074, "step": 30290} +{"lr": 0.000405, "data_time": 0.0010803461074829102, "loss": 0.007841747719794512, "time": 0.7981168031692505, "epoch": 6, "memory": 25074, "step": 30390} +{"lr": 0.000405, "data_time": 0.0016908884048461915, "loss": 0.0041880821343511345, "time": 0.7192023277282715, "epoch": 6, "memory": 25074, "step": 30490} +{"lr": 0.000405, "data_time": 0.0012493371963500977, "loss": 0.003917154716327786, "time": 0.7094663381576538, "epoch": 6, "memory": 25074, "step": 30590} +{"lr": 0.000405, "data_time": 0.0016638755798339844, "loss": 0.0069235051982104775, "time": 0.7100032091140747, "epoch": 6, "memory": 25074, "step": 30690} +{"lr": 0.000405, "data_time": 0.0013436079025268555, "loss": 0.0052233261056244375, "time": 0.7124608516693115, "epoch": 6, "memory": 25074, "step": 30790} +{"lr": 0.000405, "data_time": 0.0011367321014404297, "loss": 0.00867383647710085, "time": 0.7056768417358399, "epoch": 6, "memory": 25074, "step": 30890} +{"lr": 0.000405, "data_time": 0.001108694076538086, "loss": 0.0026851602364331485, "time": 0.7325651168823242, "epoch": 6, "memory": 25074, "step": 30990} +{"lr": 0.000405, "data_time": 0.0013169288635253907, "loss": 0.004504090035334229, "time": 0.725252103805542, "epoch": 6, "memory": 25074, "step": 31090} +{"lr": 0.000405, "data_time": 0.00191497802734375, "loss": 0.0022620852570980786, "time": 0.7325668811798096, "epoch": 6, "memory": 25074, "step": 31190} +{"lr": 0.000405, "data_time": 0.0014155387878417968, "loss": 0.007406285963952541, "time": 0.7438627481460571, "epoch": 6, "memory": 25074, "step": 31290} +{"lr": 0.000405, "data_time": 0.001490020751953125, "loss": 0.005735225137323141, "time": 0.7131868362426758, "epoch": 6, "memory": 25074, "step": 31390} +{"lr": 0.000405, "data_time": 0.0015232324600219726, "loss": 0.010248028067871928, "time": 0.7744721889495849, "epoch": 6, "memory": 25074, "step": 31490} +{"lr": 0.000405, "data_time": 0.0012341976165771485, "loss": 0.0037938281893730164, "time": 0.707684850692749, "epoch": 6, "memory": 25074, "step": 31590} +{"lr": 0.000405, "data_time": 0.0012534379959106446, "loss": 0.005043638637289405, "time": 0.7655620336532593, "epoch": 6, "memory": 25074, "step": 31690} +{"lr": 0.000405, "data_time": 0.0011670112609863282, "loss": 0.005976723181083799, "time": 0.7475017070770263, "epoch": 6, "memory": 25074, "step": 31790} +{"lr": 0.000405, "data_time": 0.0014744997024536133, "loss": 0.012173730600625276, "time": 0.7583656311035156, "epoch": 6, "memory": 25074, "step": 31890} +{"lr": 0.000405, "data_time": 0.0016011953353881835, "loss": 0.003958803694695234, "time": 0.8305601596832275, "epoch": 6, "memory": 25074, "step": 31990} +{"lr": 0.000405, "data_time": 0.0016053676605224609, "loss": 0.0069731857627630235, "time": 0.768882942199707, "epoch": 6, "memory": 25074, "step": 32090} +{"lr": 0.000405, "data_time": 0.0012295246124267578, "loss": 0.010267635434865951, "time": 0.7551312446594238, "epoch": 6, "memory": 25074, "step": 32190} +{"lr": 0.000405, "data_time": 0.0012998104095458985, "loss": 0.005389460688456893, "time": 0.7272004127502442, "epoch": 6, "memory": 25074, "step": 32290} diff --git a/convnext_small_4xb256_fake5m_lr8e-4_10epoch/20230601_151933/vis_data/config.py b/convnext_small_4xb256_fake5m_lr8e-4_10epoch/20230601_151933/vis_data/config.py new file mode 100644 index 0000000000000000000000000000000000000000..e9462c7bdf98df98a9c3d600ff4a02772382f3d3 --- /dev/null +++ b/convnext_small_4xb256_fake5m_lr8e-4_10epoch/20230601_151933/vis_data/config.py @@ -0,0 +1,169 @@ +optim_wrapper = dict( + optimizer=dict( + type='AdamW', + lr=0.0008, + weight_decay=0.05, + eps=1e-08, + betas=(0.9, 0.999), + _scope_='mmpretrain'), + paramwise_cfg=dict( + norm_decay_mult=0.0, + bias_decay_mult=0.0, + flat_decay_mult=0.0, + custom_keys=dict({ + '.absolute_pos_embed': dict(decay_mult=0.0), + '.relative_position_bias_table': dict(decay_mult=0.0) + })), + type='AmpOptimWrapper', + dtype='bfloat16', + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=True, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=4096) +model = dict( + type='ImageClassifier', + backbone=dict(type='ConvNeXt', arch='small', drop_path_rate=0.4), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=768, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + init_cfg=None), + init_cfg=dict( + type='TruncNormal', layer=['Conv2d', 'Linear'], std=0.02, bias=0.0), + train_cfg=None) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stablediffusionV1-5R2-dpmsolver-25-5m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/convnext_small_4xb256_fake5m_lr8e-4_10epoch' diff --git a/convnext_small_4xb256_fake5m_lr8e-4_10epoch/20230601_151933/vis_data/events.out.tfevents.1685603980.SH-IDC1-10-140-24-124.4960.0 b/convnext_small_4xb256_fake5m_lr8e-4_10epoch/20230601_151933/vis_data/events.out.tfevents.1685603980.SH-IDC1-10-140-24-124.4960.0 new file mode 100644 index 0000000000000000000000000000000000000000..5656a99001baa84c9b10ebfa15c572e6dc442e12 --- /dev/null +++ b/convnext_small_4xb256_fake5m_lr8e-4_10epoch/20230601_151933/vis_data/events.out.tfevents.1685603980.SH-IDC1-10-140-24-124.4960.0 @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:37e249088424c5916988d9c114aaa016c85533e1b3907db000e128e964f4d7a4 +size 446154 diff --git a/convnext_small_4xb256_fake5m_lr8e-4_10epoch/20230601_151933/vis_data/scalars.json b/convnext_small_4xb256_fake5m_lr8e-4_10epoch/20230601_151933/vis_data/scalars.json new file mode 100644 index 0000000000000000000000000000000000000000..fdca5ba5a049858db6a574c4772f2095e861f390 --- /dev/null +++ b/convnext_small_4xb256_fake5m_lr8e-4_10epoch/20230601_151933/vis_data/scalars.json @@ -0,0 +1,325 @@ +{"lr": 0.0008, "data_time": 0.0013764381408691406, "loss": 0.6050899505615235, "time": 0.7506237506866456, "epoch": 1, "memory": 25071, "step": 100} +{"lr": 0.0008, "data_time": 0.0013804197311401366, "loss": 0.5873622417449951, "time": 0.7945192337036133, "epoch": 1, "memory": 25071, "step": 200} +{"lr": 0.0008, "data_time": 0.0013028383255004883, "loss": 0.5443078756332398, "time": 0.7569274187088013, "epoch": 1, "memory": 25071, "step": 300} +{"lr": 0.0008, "data_time": 0.0012797832489013672, "loss": 0.523950707912445, "time": 0.7552407264709473, "epoch": 1, "memory": 25071, "step": 400} +{"lr": 0.0008, "data_time": 0.09460806846618652, "loss": 0.4858797013759613, "time": 0.8013857364654541, "epoch": 1, "memory": 25071, "step": 500} +{"lr": 0.0008, "data_time": 0.1175036907196045, "loss": 0.48607711791992186, "time": 0.8218770980834961, "epoch": 1, "memory": 25071, "step": 600} +{"lr": 0.0008, "data_time": 0.14032084941864015, "loss": 0.4458679795265198, "time": 0.8456096649169922, "epoch": 1, "memory": 25071, "step": 700} +{"lr": 0.0008, "data_time": 0.00155184268951416, "loss": 0.46259907484054563, "time": 0.7259519338607788, "epoch": 1, "memory": 25071, "step": 800} +{"lr": 0.0008, "data_time": 0.001605057716369629, "loss": 0.42642366886138916, "time": 0.7728655576705933, "epoch": 1, "memory": 25071, "step": 900} +{"lr": 0.0008, "data_time": 0.15106136798858644, "loss": 0.4032673418521881, "time": 0.8703392505645752, "epoch": 1, "memory": 25071, "step": 1000} +{"lr": 0.0008, "data_time": 0.08912608623504639, "loss": 0.40774006843566896, "time": 0.7977385759353638, "epoch": 1, "memory": 25071, "step": 1100} +{"lr": 0.0008, "data_time": 0.06785049438476562, "loss": 0.38488052487373353, "time": 0.7864893436431885, "epoch": 1, "memory": 25071, "step": 1200} +{"lr": 0.0008, "data_time": 0.0016669750213623047, "loss": 0.41173230707645414, "time": 0.7081609010696411, "epoch": 1, "memory": 25071, "step": 1300} +{"lr": 0.0008, "data_time": 0.06563777923583984, "loss": 0.37856956124305724, "time": 0.7875759124755859, "epoch": 1, "memory": 25071, "step": 1400} +{"lr": 0.0008, "data_time": 0.04978952407836914, "loss": 0.35707377791404726, "time": 0.7544897794723511, "epoch": 1, "memory": 25071, "step": 1500} +{"lr": 0.0008, "data_time": 0.07696795463562012, "loss": 0.3464987128973007, "time": 0.7990949869155883, "epoch": 1, "memory": 25071, "step": 1600} +{"lr": 0.0008, "data_time": 0.0022781848907470702, "loss": 0.3371317148208618, "time": 0.7818030834197998, "epoch": 1, "memory": 25071, "step": 1700} +{"lr": 0.0008, "data_time": 0.014308905601501465, "loss": 0.36208291053771974, "time": 0.7372144937515259, "epoch": 1, "memory": 25071, "step": 1800} +{"lr": 0.0008, "data_time": 0.002062225341796875, "loss": 0.3514672130346298, "time": 0.7870503187179565, "epoch": 1, "memory": 25071, "step": 1900} +{"lr": 0.0008, "data_time": 0.0019228935241699218, "loss": 0.36156936585903166, "time": 0.7589655876159668, "epoch": 1, "memory": 25071, "step": 2000} +{"lr": 0.0008, "data_time": 0.0012876033782958985, "loss": 0.32534425854682925, "time": 0.7730360507965088, "epoch": 1, "memory": 25071, "step": 2100} +{"lr": 0.0008, "data_time": 0.001839447021484375, "loss": 0.31942436695098875, "time": 0.7479748249053955, "epoch": 1, "memory": 25071, "step": 2200} +{"lr": 0.0008, "data_time": 0.0014813661575317383, "loss": 0.31683720350265504, "time": 0.7312279462814331, "epoch": 1, "memory": 25071, "step": 2300} +{"lr": 0.0008, "data_time": 0.0021446704864501952, "loss": 0.30001647770404816, "time": 0.7679721593856812, "epoch": 1, "memory": 25071, "step": 2400} +{"lr": 0.0008, "data_time": 0.0015379667282104492, "loss": 0.3225553035736084, "time": 0.7475534915924072, "epoch": 1, "memory": 25071, "step": 2500} +{"lr": 0.0008, "data_time": 0.0014340877532958984, "loss": 0.3157380446791649, "time": 0.7302091360092163, "epoch": 1, "memory": 25071, "step": 2600} +{"lr": 0.0008, "data_time": 0.001423811912536621, "loss": 0.26995800733566283, "time": 0.798098087310791, "epoch": 1, "memory": 25071, "step": 2700} +{"lr": 0.0008, "data_time": 0.0012891769409179687, "loss": 0.2793251246213913, "time": 0.7392258167266845, "epoch": 1, "memory": 25071, "step": 2800} +{"lr": 0.0008, "data_time": 0.0019237756729125976, "loss": 0.29729904383420946, "time": 0.7564581871032715, "epoch": 1, "memory": 25071, "step": 2900} +{"lr": 0.0008, "data_time": 0.0013764381408691406, "loss": 0.2973321720957756, "time": 0.7987517356872559, "epoch": 1, "memory": 25071, "step": 3000} +{"lr": 0.0008, "data_time": 0.0014077186584472655, "loss": 0.25674239099025725, "time": 0.7925383806228637, "epoch": 1, "memory": 25071, "step": 3100} +{"lr": 0.0008, "data_time": 0.0017213106155395507, "loss": 0.22798804491758345, "time": 0.7815066576004028, "epoch": 1, "memory": 25071, "step": 3200} +{"lr": 0.0008, "data_time": 0.0013444662094116212, "loss": 0.24631092995405196, "time": 0.738797926902771, "epoch": 1, "memory": 25071, "step": 3300} +{"lr": 0.0008, "data_time": 0.08044974803924561, "loss": 0.2532701015472412, "time": 0.7858518123626709, "epoch": 1, "memory": 25071, "step": 3400} +{"lr": 0.0008, "data_time": 0.09921200275421142, "loss": 0.2438528597354889, "time": 0.8053362131118774, "epoch": 1, "memory": 25071, "step": 3500} +{"lr": 0.0008, "data_time": 0.04121699333190918, "loss": 0.23169133365154265, "time": 0.7468256711959839, "epoch": 1, "memory": 25071, "step": 3600} +{"lr": 0.0008, "data_time": 0.049419617652893065, "loss": 0.22192611992359162, "time": 0.756129240989685, "epoch": 1, "memory": 25071, "step": 3700} +{"lr": 0.0008, "data_time": 0.0016196489334106446, "loss": 0.23181588351726531, "time": 0.7988584041595459, "epoch": 1, "memory": 25071, "step": 3800} +{"lr": 0.0008, "data_time": 0.0018055438995361328, "loss": 0.22034767866134644, "time": 0.7792217016220093, "epoch": 1, "memory": 25071, "step": 3900} +{"lr": 0.0008, "data_time": 0.0013665199279785157, "loss": 0.23661839216947556, "time": 0.7415711879730225, "epoch": 1, "memory": 25071, "step": 4000} +{"lr": 0.0008, "data_time": 0.0054359912872314455, "loss": 0.2334280788898468, "time": 0.7654765367507934, "epoch": 1, "memory": 25071, "step": 4100} +{"lr": 0.0008, "data_time": 0.0012468576431274414, "loss": 0.19178823679685592, "time": 0.7288877487182617, "epoch": 1, "memory": 25071, "step": 4200} +{"lr": 0.0008, "data_time": 0.0530895471572876, "loss": 0.21510777920484542, "time": 0.7665005922317505, "epoch": 1, "memory": 25071, "step": 4300} +{"lr": 0.0008, "data_time": 0.00139617919921875, "loss": 0.20133443921804428, "time": 0.7699970483779908, "epoch": 1, "memory": 25071, "step": 4400} +{"lr": 0.0008, "data_time": 0.028180885314941406, "loss": 0.19917598515748977, "time": 0.7384316682815552, "epoch": 1, "memory": 25071, "step": 4500} +{"lr": 0.0008, "data_time": 0.05180978775024414, "loss": 0.19444433748722076, "time": 0.7599361658096313, "epoch": 1, "memory": 25071, "step": 4600} +{"lr": 0.0008, "data_time": 0.048706555366516115, "loss": 0.20219548791646957, "time": 0.7711160182952881, "epoch": 1, "memory": 25071, "step": 4700} +{"lr": 0.0008, "data_time": 0.08586246967315674, "loss": 0.1862374499440193, "time": 0.7907384872436524, "epoch": 1, "memory": 25071, "step": 4800} +{"lr": 0.0008, "data_time": 0.0014488935470581056, "loss": 0.1750187024474144, "time": 0.7507224798202514, "epoch": 1, "memory": 25071, "step": 4900} +{"lr": 0.0008, "data_time": 0.01899867057800293, "loss": 0.16733268201351165, "time": 0.7253173828125, "epoch": 1, "memory": 25071, "step": 5000} +{"lr": 0.0008, "data_time": 0.008738946914672852, "loss": 0.18044504523277283, "time": 0.7870842218399048, "epoch": 1, "memory": 25071, "step": 5100} +{"lr": 0.0008, "data_time": 0.001414036750793457, "loss": 0.165419402718544, "time": 0.7413724422454834, "epoch": 1, "memory": 25071, "step": 5200} +{"lr": 0.0008, "data_time": 0.0014182567596435548, "loss": 0.1610153131186962, "time": 0.7748068571090698, "epoch": 1, "memory": 25071, "step": 5300} +{"lr": 0.0008, "data_time": 0.11126623153686524, "loss": 0.16909749209880828, "time": 0.8160391569137573, "epoch": 1, "memory": 25071, "step": 5400} +{"lr": 0.0008, "data_time": 0.031126642227172853, "loss": 0.15554608553647994, "time": 0.745137619972229, "epoch": 1, "memory": 25071, "step": 5500} +{"lr": 0.0008, "data_time": 0.06271398067474365, "loss": 0.17844616919755935, "time": 0.7679157257080078, "epoch": 1, "memory": 25071, "step": 5600} +{"lr": 0.0008, "data_time": 0.0016042947769165038, "loss": 0.16791197061538696, "time": 0.7157271862030029, "epoch": 1, "memory": 25071, "step": 5700} +{"accuracy/top1": 86.11830139160156, "data_time": 0.27462947368621826, "time": 0.7906721234321594, "step": 1} +{"lr": 0.0007806673239365858, "data_time": 0.0012798070907592773, "loss": 0.1395976021885872, "time": 0.746979832649231, "epoch": 2, "memory": 25074, "step": 5858} +{"lr": 0.0007806673239365858, "data_time": 0.0015470743179321288, "loss": 0.14222779422998427, "time": 0.7680478811264038, "epoch": 2, "memory": 25074, "step": 5958} +{"lr": 0.0007806673239365858, "data_time": 0.11526215076446533, "loss": 0.14738767072558404, "time": 0.8219593048095704, "epoch": 2, "memory": 25074, "step": 6058} +{"lr": 0.0007806673239365858, "data_time": 0.0786397933959961, "loss": 0.14101561307907104, "time": 0.7834032535552978, "epoch": 2, "memory": 25074, "step": 6158} +{"lr": 0.0007806673239365858, "data_time": 0.05114147663116455, "loss": 0.127545128762722, "time": 0.7688297510147095, "epoch": 2, "memory": 25074, "step": 6258} +{"lr": 0.0007806673239365858, "data_time": 0.051855945587158205, "loss": 0.13338923081755638, "time": 0.7567293405532837, "epoch": 2, "memory": 25074, "step": 6358} +{"lr": 0.0007806673239365858, "data_time": 0.12313151359558105, "loss": 0.16329557150602342, "time": 0.8402329683303833, "epoch": 2, "memory": 25074, "step": 6458} +{"lr": 0.0007806673239365858, "data_time": 0.05999410152435303, "loss": 0.13960904777050018, "time": 0.7650542736053467, "epoch": 2, "memory": 25074, "step": 6558} +{"lr": 0.0007806673239365858, "data_time": 0.03151988983154297, "loss": 0.11943572163581848, "time": 0.757231879234314, "epoch": 2, "memory": 25074, "step": 6658} +{"lr": 0.0007806673239365858, "data_time": 0.05698392391204834, "loss": 0.1213602788746357, "time": 0.7635266304016113, "epoch": 2, "memory": 25074, "step": 6758} +{"lr": 0.0007806673239365858, "data_time": 0.08698782920837403, "loss": 0.10986629128456116, "time": 0.8109496116638184, "epoch": 2, "memory": 25074, "step": 6858} +{"lr": 0.0007806673239365858, "data_time": 0.050164651870727536, "loss": 0.14670346677303314, "time": 0.7557155847549438, "epoch": 2, "memory": 25074, "step": 6958} +{"lr": 0.0007806673239365858, "data_time": 0.049664640426635744, "loss": 0.1171759121119976, "time": 0.7815929174423217, "epoch": 2, "memory": 25074, "step": 7058} +{"lr": 0.0007806673239365858, "data_time": 0.0514491081237793, "loss": 0.11087181940674781, "time": 0.760224986076355, "epoch": 2, "memory": 25074, "step": 7158} +{"lr": 0.0007806673239365858, "data_time": 0.08852627277374267, "loss": 0.1281448319554329, "time": 0.8089758872985839, "epoch": 2, "memory": 25074, "step": 7258} +{"lr": 0.0007806673239365858, "data_time": 0.051537489891052245, "loss": 0.08914934396743775, "time": 0.7567667484283447, "epoch": 2, "memory": 25074, "step": 7358} +{"lr": 0.0007806673239365858, "data_time": 0.0895998477935791, "loss": 0.1266442857682705, "time": 0.81229567527771, "epoch": 2, "memory": 25074, "step": 7458} +{"lr": 0.0007806673239365858, "data_time": 0.06785578727722168, "loss": 0.09124645218253136, "time": 0.7752837896347046, "epoch": 2, "memory": 25074, "step": 7558} +{"lr": 0.0007806673239365858, "data_time": 0.06337199211120606, "loss": 0.08968169614672661, "time": 0.8053130388259888, "epoch": 2, "memory": 25074, "step": 7658} +{"lr": 0.0007806673239365858, "data_time": 0.09980368614196777, "loss": 0.09525418058037757, "time": 0.80571448802948, "epoch": 2, "memory": 25074, "step": 7758} +{"lr": 0.0007806673239365858, "data_time": 0.0830761432647705, "loss": 0.10288258828222752, "time": 0.7975805282592774, "epoch": 2, "memory": 25074, "step": 7858} +{"lr": 0.0007806673239365858, "data_time": 0.05366334915161133, "loss": 0.11542944610118866, "time": 0.7596214056015015, "epoch": 2, "memory": 25074, "step": 7958} +{"lr": 0.0007806673239365858, "data_time": 0.11788253784179688, "loss": 0.08477553986012935, "time": 0.8365232229232789, "epoch": 2, "memory": 25074, "step": 8058} +{"lr": 0.0007806673239365858, "data_time": 0.0503159761428833, "loss": 0.09394867718219757, "time": 0.7549262523651123, "epoch": 2, "memory": 25074, "step": 8158} +{"lr": 0.0007806673239365858, "data_time": 0.035231494903564455, "loss": 0.08703736923635005, "time": 0.7555581569671631, "epoch": 2, "memory": 25074, "step": 8258} +{"lr": 0.0007806673239365858, "data_time": 0.10597724914550781, "loss": 0.09405033588409424, "time": 0.8121357917785644, "epoch": 2, "memory": 25074, "step": 8358} +{"lr": 0.0007806673239365858, "data_time": 0.061185598373413086, "loss": 0.09777909182012082, "time": 0.7826323509216309, "epoch": 2, "memory": 25074, "step": 8458} +{"lr": 0.0007806673239365858, "data_time": 0.028299427032470702, "loss": 0.12923009917140008, "time": 0.7326319694519043, "epoch": 2, "memory": 25074, "step": 8558} +{"lr": 0.0007806673239365858, "data_time": 0.12724955081939698, "loss": 0.08197196312248707, "time": 0.8346829175949096, "epoch": 2, "memory": 25074, "step": 8658} +{"lr": 0.0007806673239365858, "data_time": 0.029903173446655273, "loss": 0.09399201795458793, "time": 0.73518807888031, "epoch": 2, "memory": 25074, "step": 8758} +{"lr": 0.0007806673239365858, "data_time": 0.024277424812316893, "loss": 0.07302029766142368, "time": 0.7567739963531495, "epoch": 2, "memory": 25074, "step": 8858} +{"lr": 0.0007806673239365858, "data_time": 0.04628999233245849, "loss": 0.07228384427726268, "time": 0.7520631551742554, "epoch": 2, "memory": 25074, "step": 8958} +{"lr": 0.0007806673239365858, "data_time": 0.06505131721496582, "loss": 0.05472337082028389, "time": 0.770245361328125, "epoch": 2, "memory": 25074, "step": 9058} +{"lr": 0.0007806673239365858, "data_time": 0.06804196834564209, "loss": 0.10189352296292782, "time": 0.7728426933288575, "epoch": 2, "memory": 25074, "step": 9158} +{"lr": 0.0007806673239365858, "data_time": 0.010705995559692382, "loss": 0.06046130508184433, "time": 0.7218541622161865, "epoch": 2, "memory": 25074, "step": 9258} +{"lr": 0.0007806673239365858, "data_time": 0.037660741806030275, "loss": 0.08741049394011498, "time": 0.7417712450027466, "epoch": 2, "memory": 25074, "step": 9358} +{"lr": 0.0007806673239365858, "data_time": 0.0012258291244506836, "loss": 0.06973813138902188, "time": 0.7826901912689209, "epoch": 2, "memory": 25074, "step": 9458} +{"lr": 0.0007806673239365858, "data_time": 0.0012144088745117188, "loss": 0.061193672940135, "time": 0.7893070936203003, "epoch": 2, "memory": 25074, "step": 9558} +{"lr": 0.0007806673239365858, "data_time": 0.0010945558547973632, "loss": 0.06761947125196457, "time": 0.7257037162780762, "epoch": 2, "memory": 25074, "step": 9658} +{"lr": 0.0007806673239365858, "data_time": 0.0016387939453125, "loss": 0.06786825507879257, "time": 0.7165427684783936, "epoch": 2, "memory": 25074, "step": 9758} +{"lr": 0.0007806673239365858, "data_time": 0.0010945320129394532, "loss": 0.06125165410339832, "time": 0.8063461065292359, "epoch": 2, "memory": 25074, "step": 9858} +{"lr": 0.0007806673239365858, "data_time": 0.001276540756225586, "loss": 0.050140045769512656, "time": 0.7595997810363769, "epoch": 2, "memory": 25074, "step": 9958} +{"lr": 0.0007806673239365858, "data_time": 0.0011738300323486327, "loss": 0.05405156258493662, "time": 0.7570838928222656, "epoch": 2, "memory": 25074, "step": 10058} +{"lr": 0.0007806673239365858, "data_time": 0.001383829116821289, "loss": 0.05630311444401741, "time": 0.777427077293396, "epoch": 2, "memory": 25074, "step": 10158} +{"lr": 0.0007806673239365858, "data_time": 0.001216292381286621, "loss": 0.0726040281355381, "time": 0.8589967489242554, "epoch": 2, "memory": 25074, "step": 10258} +{"lr": 0.0007806673239365858, "data_time": 0.0011072874069213868, "loss": 0.06845370680093765, "time": 0.8274120330810547, "epoch": 2, "memory": 25074, "step": 10358} +{"lr": 0.0007806673239365858, "data_time": 0.0013106584548950196, "loss": 0.05654991827905178, "time": 0.7579174995422363, "epoch": 2, "memory": 25074, "step": 10458} +{"lr": 0.0007806673239365858, "data_time": 0.0016683578491210938, "loss": 0.05133899617940187, "time": 0.7490403890609741, "epoch": 2, "memory": 25074, "step": 10558} +{"lr": 0.0007806673239365858, "data_time": 0.001333904266357422, "loss": 0.057648342847824094, "time": 0.764176607131958, "epoch": 2, "memory": 25074, "step": 10658} +{"lr": 0.0007806673239365858, "data_time": 0.0013494014739990235, "loss": 0.08061864003539085, "time": 0.8177014827728272, "epoch": 2, "memory": 25074, "step": 10758} +{"lr": 0.0007806673239365858, "data_time": 0.029112839698791505, "loss": 0.04146339111030102, "time": 0.8291461229324341, "epoch": 2, "memory": 25074, "step": 10858} +{"lr": 0.0007806673239365858, "data_time": 0.0011493921279907226, "loss": 0.0886738296598196, "time": 0.7861827373504638, "epoch": 2, "memory": 25074, "step": 10958} +{"lr": 0.0007806673239365858, "data_time": 0.0011094570159912109, "loss": 0.06363021060824395, "time": 0.8145726919174194, "epoch": 2, "memory": 25074, "step": 11058} +{"lr": 0.0007806673239365858, "data_time": 0.001296830177307129, "loss": 0.04318107292056084, "time": 0.7690038681030273, "epoch": 2, "memory": 25074, "step": 11158} +{"lr": 0.0007806673239365858, "data_time": 0.0024667978286743164, "loss": 0.039654660411179064, "time": 0.7918517112731933, "epoch": 2, "memory": 25074, "step": 11258} +{"lr": 0.0007806673239365858, "data_time": 0.0011334180831909179, "loss": 0.030768837593495846, "time": 0.7723324537277222, "epoch": 2, "memory": 25074, "step": 11358} +{"lr": 0.0007806673239365858, "data_time": 0.001248335838317871, "loss": 0.03774199243634939, "time": 0.7432373762130737, "epoch": 2, "memory": 25074, "step": 11458} +{"accuracy/top1": 76.77545166015625, "data_time": 0.21405748759998994, "time": 0.6063573220196892, "step": 2} +{"lr": 0.0007245617127781043, "data_time": 0.1599264144897461, "loss": 0.04088939521461725, "time": 0.8642370462417602, "epoch": 3, "memory": 25074, "step": 11616} +{"lr": 0.0007245617127781043, "data_time": 0.001267719268798828, "loss": 0.040835813619196415, "time": 0.8656621456146241, "epoch": 3, "memory": 25074, "step": 11716} +{"lr": 0.0007245617127781043, "data_time": 0.001159811019897461, "loss": 0.048896664567291734, "time": 0.7798550128936768, "epoch": 3, "memory": 25074, "step": 11816} +{"lr": 0.0007245617127781043, "data_time": 0.001351618766784668, "loss": 0.03867869433015585, "time": 0.8321730613708496, "epoch": 3, "memory": 25074, "step": 11916} +{"lr": 0.0007245617127781043, "data_time": 0.0014122247695922852, "loss": 0.04783026725053787, "time": 0.7364333868026733, "epoch": 3, "memory": 25074, "step": 12016} +{"lr": 0.0007245617127781043, "data_time": 0.001270461082458496, "loss": 0.04759179260581732, "time": 0.8261472225189209, "epoch": 3, "memory": 25074, "step": 12116} +{"lr": 0.0007245617127781043, "data_time": 0.001415562629699707, "loss": 0.037141728401184085, "time": 0.7853118658065796, "epoch": 3, "memory": 25074, "step": 12216} +{"lr": 0.0007245617127781043, "data_time": 0.0012857437133789063, "loss": 0.039891570992767814, "time": 0.7424207925796509, "epoch": 3, "memory": 25074, "step": 12316} +{"lr": 0.0007245617127781043, "data_time": 0.0014500379562377929, "loss": 0.03392054084688425, "time": 0.7254342555999755, "epoch": 3, "memory": 25074, "step": 12416} +{"lr": 0.0007245617127781043, "data_time": 0.0012304544448852538, "loss": 0.03665917366743088, "time": 0.7534410715103149, "epoch": 3, "memory": 25074, "step": 12516} +{"lr": 0.0007245617127781043, "data_time": 0.0012962818145751953, "loss": 0.040328314155340196, "time": 0.7487728595733643, "epoch": 3, "memory": 25074, "step": 12616} +{"lr": 0.0007245617127781043, "data_time": 0.0013105392456054688, "loss": 0.03161096451804042, "time": 0.7518738031387329, "epoch": 3, "memory": 25074, "step": 12716} +{"lr": 0.0007245617127781043, "data_time": 0.0012244939804077148, "loss": 0.04508720170706511, "time": 0.7207002878189087, "epoch": 3, "memory": 25074, "step": 12816} +{"lr": 0.0007245617127781043, "data_time": 0.0013757705688476562, "loss": 0.026851135678589343, "time": 0.717484426498413, "epoch": 3, "memory": 25074, "step": 12916} +{"lr": 0.0007245617127781043, "data_time": 0.0012349367141723632, "loss": 0.04733375571668148, "time": 0.7747965097427368, "epoch": 3, "memory": 25074, "step": 13016} +{"lr": 0.0007245617127781043, "data_time": 0.0012233495712280274, "loss": 0.02905960977077484, "time": 0.7289293766021728, "epoch": 3, "memory": 25074, "step": 13116} +{"lr": 0.0007245617127781043, "data_time": 0.0015436887741088867, "loss": 0.030410025874152778, "time": 0.7665880441665649, "epoch": 3, "memory": 25074, "step": 13216} +{"lr": 0.0007245617127781043, "data_time": 0.0012766122817993164, "loss": 0.027036966988816857, "time": 0.8024507284164428, "epoch": 3, "memory": 25074, "step": 13316} +{"lr": 0.0007245617127781043, "data_time": 0.0011659622192382812, "loss": 0.048966099973767996, "time": 0.7735217332839965, "epoch": 3, "memory": 25074, "step": 13416} +{"lr": 0.0007245617127781043, "data_time": 0.0013410091400146485, "loss": 0.030625978391617535, "time": 0.7426556587219239, "epoch": 3, "memory": 25074, "step": 13516} +{"lr": 0.0007245617127781043, "data_time": 0.0013001203536987306, "loss": 0.0330763032194227, "time": 0.7913923263549805, "epoch": 3, "memory": 25074, "step": 13616} +{"lr": 0.0007245617127781043, "data_time": 0.0013531684875488282, "loss": 0.028288222942501308, "time": 0.7605397462844848, "epoch": 3, "memory": 25074, "step": 13716} +{"lr": 0.0007245617127781043, "data_time": 0.0018346309661865234, "loss": 0.03461533999070525, "time": 0.7678704500198364, "epoch": 3, "memory": 25074, "step": 13816} +{"lr": 0.0007245617127781043, "data_time": 0.0014458656311035155, "loss": 0.02867770232260227, "time": 0.7836845397949219, "epoch": 3, "memory": 25074, "step": 13916} +{"lr": 0.0007245617127781043, "data_time": 0.0011763334274291991, "loss": 0.024216140899807215, "time": 0.7715628147125244, "epoch": 3, "memory": 25074, "step": 14016} +{"lr": 0.0007245617127781043, "data_time": 0.0012571573257446288, "loss": 0.03660787008702755, "time": 0.8137492656707763, "epoch": 3, "memory": 25074, "step": 14116} +{"lr": 0.0007245617127781043, "data_time": 0.001326131820678711, "loss": 0.0303189299069345, "time": 0.7664277076721191, "epoch": 3, "memory": 25074, "step": 14216} +{"lr": 0.0007245617127781043, "data_time": 0.001418900489807129, "loss": 0.03185373768210411, "time": 0.8259416103363038, "epoch": 3, "memory": 25074, "step": 14316} +{"lr": 0.0007245617127781043, "data_time": 0.001248311996459961, "loss": 0.02666334230452776, "time": 0.7289786815643311, "epoch": 3, "memory": 25074, "step": 14416} +{"lr": 0.0007245617127781043, "data_time": 0.0013849496841430663, "loss": 0.037510369904339314, "time": 0.7737699747085571, "epoch": 3, "memory": 25074, "step": 14516} +{"lr": 0.0007245617127781043, "data_time": 0.0010577678680419923, "loss": 0.03314679367467761, "time": 0.7594129323959351, "epoch": 3, "memory": 25074, "step": 14616} +{"lr": 0.0007245617127781043, "data_time": 0.0012985706329345704, "loss": 0.031005522515624762, "time": 0.8288192987442017, "epoch": 3, "memory": 25074, "step": 14716} +{"lr": 0.0007245617127781043, "data_time": 0.0012567996978759765, "loss": 0.034702819958329204, "time": 0.7696936845779419, "epoch": 3, "memory": 25074, "step": 14816} +{"lr": 0.0007245617127781043, "data_time": 0.0014369964599609375, "loss": 0.025255820620805024, "time": 0.7652963399887085, "epoch": 3, "memory": 25074, "step": 14916} +{"lr": 0.0007245617127781043, "data_time": 0.0017720937728881836, "loss": 0.030143201537430286, "time": 0.761293911933899, "epoch": 3, "memory": 25074, "step": 15016} +{"lr": 0.0007245617127781043, "data_time": 0.0013030290603637696, "loss": 0.021141273621469735, "time": 0.7685921907424926, "epoch": 3, "memory": 25074, "step": 15116} +{"lr": 0.0007245617127781043, "data_time": 0.0012258768081665039, "loss": 0.017992442473769188, "time": 0.7396396636962891, "epoch": 3, "memory": 25074, "step": 15216} +{"lr": 0.0007245617127781043, "data_time": 0.001431441307067871, "loss": 0.023397974809631704, "time": 0.8045117616653442, "epoch": 3, "memory": 25074, "step": 15316} +{"lr": 0.0007245617127781043, "data_time": 0.0012537240982055664, "loss": 0.03260563835501671, "time": 0.7762276887893677, "epoch": 3, "memory": 25074, "step": 15416} +{"lr": 0.0007245617127781043, "data_time": 0.0012106895446777344, "loss": 0.023786334693431853, "time": 0.7057984113693238, "epoch": 3, "memory": 25074, "step": 15516} +{"lr": 0.0007245617127781043, "data_time": 0.0011429548263549804, "loss": 0.034362148866057395, "time": 0.7490205287933349, "epoch": 3, "memory": 25074, "step": 15616} +{"lr": 0.0007245617127781043, "data_time": 0.0012099027633666992, "loss": 0.018229505466297268, "time": 0.7419039726257324, "epoch": 3, "memory": 25074, "step": 15716} +{"lr": 0.0007245617127781043, "data_time": 0.0019709110260009766, "loss": 0.018690613843500615, "time": 0.7349676370620728, "epoch": 3, "memory": 25074, "step": 15816} +{"lr": 0.0007245617127781043, "data_time": 0.0013330459594726562, "loss": 0.017513964604586364, "time": 0.7214994668960572, "epoch": 3, "memory": 25074, "step": 15916} +{"lr": 0.0007245617127781043, "data_time": 0.001301121711730957, "loss": 0.028252078453078866, "time": 0.7291983604431153, "epoch": 3, "memory": 25074, "step": 16016} +{"lr": 0.0007245617127781043, "data_time": 0.001300644874572754, "loss": 0.02576492396183312, "time": 0.7864515542984009, "epoch": 3, "memory": 25074, "step": 16116} +{"lr": 0.0007245617127781043, "data_time": 0.001355290412902832, "loss": 0.027555054612457754, "time": 0.710329246520996, "epoch": 3, "memory": 25074, "step": 16216} +{"lr": 0.0007245617127781043, "data_time": 0.0011693954467773438, "loss": 0.026239687763154506, "time": 0.7224598169326782, "epoch": 3, "memory": 25074, "step": 16316} +{"lr": 0.0007245617127781043, "data_time": 0.0013732671737670898, "loss": 0.021136733843013643, "time": 0.7503315448760987, "epoch": 3, "memory": 25074, "step": 16416} +{"lr": 0.0007245617127781043, "data_time": 0.001243901252746582, "loss": 0.024915443174540995, "time": 0.7485300302505493, "epoch": 3, "memory": 25074, "step": 16516} +{"lr": 0.0007245617127781043, "data_time": 0.0013096094131469726, "loss": 0.02882102965377271, "time": 0.816267991065979, "epoch": 3, "memory": 25074, "step": 16616} +{"lr": 0.0007245617127781043, "data_time": 0.0010734796524047852, "loss": 0.01705039106309414, "time": 0.7063563823699951, "epoch": 3, "memory": 25074, "step": 16716} +{"lr": 0.0007245617127781043, "data_time": 0.001378631591796875, "loss": 0.02329266774468124, "time": 0.781953239440918, "epoch": 3, "memory": 25074, "step": 16816} +{"lr": 0.0007245617127781043, "data_time": 0.0013356924057006836, "loss": 0.024129189271479844, "time": 0.7305824756622314, "epoch": 3, "memory": 25074, "step": 16916} +{"lr": 0.0007245617127781043, "data_time": 0.0014432430267333984, "loss": 0.015659291297197342, "time": 0.7203398704528808, "epoch": 3, "memory": 25074, "step": 17016} +{"lr": 0.0007245617127781043, "data_time": 0.0016174554824829102, "loss": 0.02788719804957509, "time": 0.7511308193206787, "epoch": 3, "memory": 25074, "step": 17116} +{"lr": 0.0007245617127781043, "data_time": 0.0011986494064331055, "loss": 0.021045492310076953, "time": 0.7102416753768921, "epoch": 3, "memory": 25074, "step": 17216} +{"accuracy/top1": 96.7741928100586, "data_time": 0.2159565196317785, "time": 0.6016581899979535, "step": 3} +{"lr": 0.0006371751746555269, "data_time": 0.0016854047775268556, "loss": 0.02584357662126422, "time": 0.7935037851333618, "epoch": 4, "memory": 25074, "step": 17374} +{"lr": 0.0006371751746555269, "data_time": 0.0013775825500488281, "loss": 0.015903302934020758, "time": 0.7569707632064819, "epoch": 4, "memory": 25074, "step": 17474} +{"lr": 0.0006371751746555269, "data_time": 0.0012466907501220703, "loss": 0.022549204388633372, "time": 0.7929486989974975, "epoch": 4, "memory": 25074, "step": 17574} +{"lr": 0.0006371751746555269, "data_time": 0.0013414859771728516, "loss": 0.027700148802250624, "time": 0.739317512512207, "epoch": 4, "memory": 25074, "step": 17674} +{"lr": 0.0006371751746555269, "data_time": 0.0014379501342773437, "loss": 0.018095119623467326, "time": 0.815676498413086, "epoch": 4, "memory": 25074, "step": 17774} +{"lr": 0.0006371751746555269, "data_time": 0.0014359712600708007, "loss": 0.01641491292975843, "time": 0.7535361051559448, "epoch": 4, "memory": 25074, "step": 17874} +{"lr": 0.0006371751746555269, "data_time": 0.001419830322265625, "loss": 0.016665436886250974, "time": 0.7647857666015625, "epoch": 4, "memory": 25074, "step": 17974} +{"lr": 0.0006371751746555269, "data_time": 0.0015505552291870117, "loss": 0.021457855915650727, "time": 0.7416374444961548, "epoch": 4, "memory": 25074, "step": 18074} +{"lr": 0.0006371751746555269, "data_time": 0.0013824701309204102, "loss": 0.023184484895318748, "time": 0.7078312635421753, "epoch": 4, "memory": 25074, "step": 18174} +{"lr": 0.0006371751746555269, "data_time": 0.0014255046844482422, "loss": 0.019590003369376063, "time": 0.7725302457809449, "epoch": 4, "memory": 25074, "step": 18274} +{"lr": 0.0006371751746555269, "data_time": 0.0013128757476806641, "loss": 0.016926329070702194, "time": 0.7734216690063477, "epoch": 4, "memory": 25074, "step": 18374} +{"lr": 0.0006371751746555269, "data_time": 0.0014129638671875, "loss": 0.022653862461447717, "time": 0.741803240776062, "epoch": 4, "memory": 25074, "step": 18474} +{"lr": 0.0006371751746555269, "data_time": 0.0012583017349243164, "loss": 0.012973507773131132, "time": 0.7640147924423217, "epoch": 4, "memory": 25074, "step": 18574} +{"lr": 0.0006371751746555269, "data_time": 0.0012765645980834961, "loss": 0.022019442077726124, "time": 0.7581893920898437, "epoch": 4, "memory": 25074, "step": 18674} +{"lr": 0.0006371751746555269, "data_time": 0.0012765884399414062, "loss": 0.012811435712501406, "time": 0.7138432025909424, "epoch": 4, "memory": 25074, "step": 18774} +{"lr": 0.0006371751746555269, "data_time": 0.0012908458709716796, "loss": 0.01544500607997179, "time": 0.7485158920288086, "epoch": 4, "memory": 25074, "step": 18874} +{"lr": 0.0006371751746555269, "data_time": 0.0013615846633911132, "loss": 0.01228244355879724, "time": 0.7062382221221923, "epoch": 4, "memory": 25074, "step": 18974} +{"lr": 0.0006371751746555269, "data_time": 0.0014231443405151368, "loss": 0.010163441905751825, "time": 0.7450262308120728, "epoch": 4, "memory": 25074, "step": 19074} +{"lr": 0.0006371751746555269, "data_time": 0.0014815807342529296, "loss": 0.022428280021995307, "time": 0.7605623006820679, "epoch": 4, "memory": 25074, "step": 19174} +{"lr": 0.0006371751746555269, "data_time": 0.0011575698852539062, "loss": 0.017260070750489832, "time": 0.7530862808227539, "epoch": 4, "memory": 25074, "step": 19274} +{"lr": 0.0006371751746555269, "data_time": 0.0013218402862548828, "loss": 0.019637471809983254, "time": 0.7508957862854004, "epoch": 4, "memory": 25074, "step": 19374} +{"lr": 0.0006371751746555269, "data_time": 0.0013967275619506836, "loss": 0.023825343092903496, "time": 0.714510703086853, "epoch": 4, "memory": 25074, "step": 19474} +{"lr": 0.0006371751746555269, "data_time": 0.0012618541717529298, "loss": 0.013557658018544316, "time": 0.7425036907196045, "epoch": 4, "memory": 25074, "step": 19574} +{"lr": 0.0006371751746555269, "data_time": 0.001282811164855957, "loss": 0.017627335991710423, "time": 0.7495125293731689, "epoch": 4, "memory": 25074, "step": 19674} +{"lr": 0.0006371751746555269, "data_time": 0.0013806343078613282, "loss": 0.026325806882232426, "time": 0.7633819818496704, "epoch": 4, "memory": 25074, "step": 19774} +{"lr": 0.0006371751746555269, "data_time": 0.0014236927032470702, "loss": 0.01704902113415301, "time": 0.7887636423110962, "epoch": 4, "memory": 25074, "step": 19874} +{"lr": 0.0006371751746555269, "data_time": 0.0011818408966064453, "loss": 0.018378399033099412, "time": 0.7308815956115723, "epoch": 4, "memory": 25074, "step": 19974} +{"lr": 0.0006371751746555269, "data_time": 0.0013622999191284179, "loss": 0.01764199361205101, "time": 0.7831684827804566, "epoch": 4, "memory": 25074, "step": 20074} +{"lr": 0.0006371751746555269, "data_time": 0.001314258575439453, "loss": 0.014964620536193251, "time": 0.7577922582626343, "epoch": 4, "memory": 25074, "step": 20174} +{"lr": 0.0006371751746555269, "data_time": 0.0014181137084960938, "loss": 0.015263845445588231, "time": 0.9323982238769531, "epoch": 4, "memory": 25074, "step": 20274} +{"lr": 0.0006371751746555269, "data_time": 0.0015181303024291992, "loss": 0.015902172727510332, "time": 0.8992016553878784, "epoch": 4, "memory": 25074, "step": 20374} +{"lr": 0.0006371751746555269, "data_time": 0.001190924644470215, "loss": 0.01186499735340476, "time": 0.871763277053833, "epoch": 4, "memory": 25074, "step": 20474} +{"lr": 0.0006371751746555269, "data_time": 0.0012467384338378906, "loss": 0.0199515497777611, "time": 0.8446225643157959, "epoch": 4, "memory": 25074, "step": 20574} +{"lr": 0.0006371751746555269, "data_time": 0.0011667966842651366, "loss": 0.012517213681712746, "time": 0.7645492553710938, "epoch": 4, "memory": 25074, "step": 20674} +{"lr": 0.0006371751746555269, "data_time": 0.0013451814651489259, "loss": 0.008589001651853324, "time": 0.8372746229171752, "epoch": 4, "memory": 25074, "step": 20774} +{"lr": 0.0006371751746555269, "data_time": 0.0015634775161743164, "loss": 0.010175559064373374, "time": 0.8577198505401611, "epoch": 4, "memory": 25074, "step": 20874} +{"lr": 0.0006371751746555269, "data_time": 0.0013834714889526367, "loss": 0.019995456002652644, "time": 0.7834474802017212, "epoch": 4, "memory": 25074, "step": 20974} +{"lr": 0.0006371751746555269, "data_time": 0.001290607452392578, "loss": 0.01769702658057213, "time": 0.7835808277130127, "epoch": 4, "memory": 25074, "step": 21074} +{"lr": 0.0006371751746555269, "data_time": 0.0013242483139038086, "loss": 0.0172896021977067, "time": 0.8407667875289917, "epoch": 4, "memory": 25074, "step": 21174} +{"lr": 0.0006371751746555269, "data_time": 0.001394939422607422, "loss": 0.012087896093726158, "time": 0.8116055488586426, "epoch": 4, "memory": 25074, "step": 21274} +{"lr": 0.0006371751746555269, "data_time": 0.0015437602996826172, "loss": 0.012474996922537684, "time": 0.7602555751800537, "epoch": 4, "memory": 25074, "step": 21374} +{"lr": 0.0006371751746555269, "data_time": 0.0013099193572998046, "loss": 0.010769471200183034, "time": 0.7567335367202759, "epoch": 4, "memory": 25074, "step": 21474} +{"lr": 0.0006371751746555269, "data_time": 0.0011894941329956056, "loss": 0.014800886064767838, "time": 0.7463791608810425, "epoch": 4, "memory": 25074, "step": 21574} +{"lr": 0.0006371751746555269, "data_time": 0.001492786407470703, "loss": 0.014644334558397532, "time": 0.7918062686920166, "epoch": 4, "memory": 25074, "step": 21674} +{"lr": 0.0006371751746555269, "data_time": 0.0012953996658325196, "loss": 0.011028381902724505, "time": 0.7790021657943725, "epoch": 4, "memory": 25074, "step": 21774} +{"lr": 0.0006371751746555269, "data_time": 0.0012609720230102538, "loss": 0.023543927865102888, "time": 0.7411756038665771, "epoch": 4, "memory": 25074, "step": 21874} +{"lr": 0.0006371751746555269, "data_time": 0.0013468265533447266, "loss": 0.01321612591855228, "time": 0.7579327344894409, "epoch": 4, "memory": 25074, "step": 21974} +{"lr": 0.0006371751746555269, "data_time": 0.0012727022171020509, "loss": 0.011263295263051986, "time": 0.8052274942398071, "epoch": 4, "memory": 25074, "step": 22074} +{"lr": 0.0006371751746555269, "data_time": 0.001506948471069336, "loss": 0.015829464327543973, "time": 0.733405590057373, "epoch": 4, "memory": 25074, "step": 22174} +{"lr": 0.0006371751746555269, "data_time": 0.0014414310455322266, "loss": 0.01200642492622137, "time": 0.7600989580154419, "epoch": 4, "memory": 25074, "step": 22274} +{"lr": 0.0006371751746555269, "data_time": 0.00143740177154541, "loss": 0.01129843876697123, "time": 0.7545924425125122, "epoch": 4, "memory": 25074, "step": 22374} +{"lr": 0.0006371751746555269, "data_time": 0.001237034797668457, "loss": 0.009275696938857436, "time": 0.7790245294570923, "epoch": 4, "memory": 25074, "step": 22474} +{"lr": 0.0006371751746555269, "data_time": 0.001354074478149414, "loss": 0.00458734123967588, "time": 0.8176832437515259, "epoch": 4, "memory": 25074, "step": 22574} +{"lr": 0.0006371751746555269, "data_time": 0.0015718460083007813, "loss": 0.015020606759935617, "time": 0.753489899635315, "epoch": 4, "memory": 25074, "step": 22674} +{"lr": 0.0006371751746555269, "data_time": 0.0012067079544067383, "loss": 0.018114811880514024, "time": 0.8565834283828735, "epoch": 4, "memory": 25074, "step": 22774} +{"lr": 0.0006371751746555269, "data_time": 0.0011778831481933593, "loss": 0.012367798015475273, "time": 0.7794310569763183, "epoch": 4, "memory": 25074, "step": 22874} +{"lr": 0.0006371751746555269, "data_time": 0.0015181779861450195, "loss": 0.017175087658688425, "time": 0.7682755470275879, "epoch": 4, "memory": 25074, "step": 22974} +{"accuracy/top1": 99.00889587402344, "data_time": 0.20151387943941004, "time": 0.5868225658641142, "step": 4} +{"lr": 0.0005270617127781042, "data_time": 0.0012909412384033204, "loss": 0.012921292847022415, "time": 0.7147369384765625, "epoch": 5, "memory": 25074, "step": 23132} +{"lr": 0.0005270617127781042, "data_time": 0.0012772560119628906, "loss": 0.006702626030892134, "time": 0.7589564800262452, "epoch": 5, "memory": 25074, "step": 23232} +{"lr": 0.0005270617127781042, "data_time": 0.001299881935119629, "loss": 0.009381459048017859, "time": 0.7967705726623535, "epoch": 5, "memory": 25074, "step": 23332} +{"lr": 0.0005270617127781042, "data_time": 0.0012833833694458007, "loss": 0.015548907918855548, "time": 0.8214048385620117, "epoch": 5, "memory": 25074, "step": 23432} +{"lr": 0.0005270617127781042, "data_time": 0.0013784646987915039, "loss": 0.006287212623283267, "time": 0.8268746376037598, "epoch": 5, "memory": 25074, "step": 23532} +{"lr": 0.0005270617127781042, "data_time": 0.0012119770050048827, "loss": 0.011349517572671174, "time": 0.7883623838424683, "epoch": 5, "memory": 25074, "step": 23632} +{"lr": 0.0005270617127781042, "data_time": 0.0015107154846191405, "loss": 0.022485012747347356, "time": 0.8699362277984619, "epoch": 5, "memory": 25074, "step": 23732} +{"lr": 0.0005270617127781042, "data_time": 0.001415252685546875, "loss": 0.013932380033656955, "time": 0.8240361213684082, "epoch": 5, "memory": 25074, "step": 23832} +{"lr": 0.0005270617127781042, "data_time": 0.0013387203216552734, "loss": 0.00835984256118536, "time": 0.7652427434921265, "epoch": 5, "memory": 25074, "step": 23932} +{"lr": 0.0005270617127781042, "data_time": 0.0018189191818237306, "loss": 0.00987613033503294, "time": 0.9262356996536255, "epoch": 5, "memory": 25074, "step": 24032} +{"lr": 0.0005270617127781042, "data_time": 0.0014084815979003907, "loss": 0.0059687618631869555, "time": 0.811207628250122, "epoch": 5, "memory": 25074, "step": 24132} +{"lr": 0.0005270617127781042, "data_time": 0.0021492481231689454, "loss": 0.014398828661069275, "time": 0.8130000352859497, "epoch": 5, "memory": 25074, "step": 24232} +{"lr": 0.0005270617127781042, "data_time": 0.0015762805938720702, "loss": 0.010351484594866633, "time": 0.7553416967391968, "epoch": 5, "memory": 25074, "step": 24332} +{"lr": 0.0005270617127781042, "data_time": 0.0014958620071411134, "loss": 0.005105896294116974, "time": 0.7634966850280762, "epoch": 5, "memory": 25074, "step": 24432} +{"lr": 0.0005270617127781042, "data_time": 0.0012889623641967774, "loss": 0.006283015664666891, "time": 0.9195183515548706, "epoch": 5, "memory": 25074, "step": 24532} +{"lr": 0.0005270617127781042, "data_time": 0.0015323638916015625, "loss": 0.009210485918447375, "time": 0.7294370889663696, "epoch": 5, "memory": 25074, "step": 24632} +{"lr": 0.0005270617127781042, "data_time": 0.0014931440353393554, "loss": 0.011729899188503623, "time": 0.7861579656600952, "epoch": 5, "memory": 25074, "step": 24732} +{"lr": 0.0005270617127781042, "data_time": 0.001440143585205078, "loss": 0.009647990157827736, "time": 0.7527569770812989, "epoch": 5, "memory": 25074, "step": 24832} +{"lr": 0.0005270617127781042, "data_time": 0.0013882875442504882, "loss": 0.00918117887340486, "time": 0.8107163906097412, "epoch": 5, "memory": 25074, "step": 24932} +{"lr": 0.0005270617127781042, "data_time": 0.0012422561645507812, "loss": 0.0036405269987881184, "time": 0.7296783208847046, "epoch": 5, "memory": 25074, "step": 25032} +{"lr": 0.0005270617127781042, "data_time": 0.001269960403442383, "loss": 0.00785787650384009, "time": 0.7547079801559449, "epoch": 5, "memory": 25074, "step": 25132} +{"lr": 0.0005270617127781042, "data_time": 0.0017754793167114257, "loss": 0.012282479414716363, "time": 0.7850768089294433, "epoch": 5, "memory": 25074, "step": 25232} +{"lr": 0.0005270617127781042, "data_time": 0.0013004064559936524, "loss": 0.009536367142572999, "time": 0.7426051616668701, "epoch": 5, "memory": 25074, "step": 25332} +{"lr": 0.0005270617127781042, "data_time": 0.0012665987014770508, "loss": 0.00943656899034977, "time": 0.7631567239761352, "epoch": 5, "memory": 25074, "step": 25432} +{"lr": 0.0005270617127781042, "data_time": 0.0015227556228637694, "loss": 0.008347800001502037, "time": 0.786600136756897, "epoch": 5, "memory": 25074, "step": 25532} +{"lr": 0.0005270617127781042, "data_time": 0.0015579462051391602, "loss": 0.012324441689997911, "time": 0.7793980360031127, "epoch": 5, "memory": 25074, "step": 25632} +{"lr": 0.0005270617127781042, "data_time": 0.0014023303985595704, "loss": 0.010182300163432956, "time": 0.822082805633545, "epoch": 5, "memory": 25074, "step": 25732} +{"lr": 0.0005270617127781042, "data_time": 0.0012905359268188476, "loss": 0.008146683033555747, "time": 1.0886083364486694, "epoch": 5, "memory": 25074, "step": 25832} +{"lr": 0.0005270617127781042, "data_time": 0.0012215375900268555, "loss": 0.010026113595813513, "time": 1.1644296884536742, "epoch": 5, "memory": 25074, "step": 25932} +{"lr": 0.0005270617127781042, "data_time": 0.0013471126556396484, "loss": 0.013122609863057733, "time": 1.0125898599624634, "epoch": 5, "memory": 25074, "step": 26032} +{"lr": 0.0005270617127781042, "data_time": 0.0011795997619628907, "loss": 0.012971840659156441, "time": 0.7858582973480225, "epoch": 5, "memory": 25074, "step": 26132} +{"lr": 0.0005270617127781042, "data_time": 0.0013992071151733398, "loss": 0.008178239269182087, "time": 0.8907784700393677, "epoch": 5, "memory": 25074, "step": 26232} +{"lr": 0.0005270617127781042, "data_time": 0.0011201620101928711, "loss": 0.013482456980273127, "time": 0.7799110651016236, "epoch": 5, "memory": 25074, "step": 26332} +{"lr": 0.0005270617127781042, "data_time": 0.0011100530624389648, "loss": 0.011699210526421666, "time": 0.8007047414779663, "epoch": 5, "memory": 25074, "step": 26432} +{"lr": 0.0005270617127781042, "data_time": 0.0015693426132202149, "loss": 0.01425934825092554, "time": 0.8350407600402832, "epoch": 5, "memory": 25074, "step": 26532} +{"lr": 0.0005270617127781042, "data_time": 0.001519489288330078, "loss": 0.00782128032296896, "time": 1.1066365718841553, "epoch": 5, "memory": 25074, "step": 26632} +{"lr": 0.0005270617127781042, "data_time": 0.0012603759765625, "loss": 0.006382914027199149, "time": 1.1303036212921143, "epoch": 5, "memory": 25074, "step": 26732} +{"lr": 0.0005270617127781042, "data_time": 0.001593303680419922, "loss": 0.008885769452899695, "time": 1.0235493898391723, "epoch": 5, "memory": 25074, "step": 26832} +{"lr": 0.0005270617127781042, "data_time": 0.0015812397003173828, "loss": 0.006711181439459324, "time": 1.011578345298767, "epoch": 5, "memory": 25074, "step": 26932} +{"lr": 0.0005270617127781042, "data_time": 0.001276254653930664, "loss": 0.007281673047691584, "time": 1.0732971906661988, "epoch": 5, "memory": 25074, "step": 27032} +{"lr": 0.0005270617127781042, "data_time": 0.0012431859970092774, "loss": 0.007350308913737535, "time": 1.12139573097229, "epoch": 5, "memory": 25074, "step": 27132} +{"lr": 0.0005270617127781042, "data_time": 0.0014218330383300782, "loss": 0.010497211245819926, "time": 0.7100059986114502, "epoch": 5, "memory": 25074, "step": 27232} +{"lr": 0.0005270617127781042, "data_time": 0.0013143062591552735, "loss": 0.006091742077842355, "time": 0.7222891092300415, "epoch": 5, "memory": 25074, "step": 27332} +{"lr": 0.0005270617127781042, "data_time": 0.0014330625534057617, "loss": 0.008059976575896143, "time": 0.7129441738128662, "epoch": 5, "memory": 25074, "step": 27432} +{"lr": 0.0005270617127781042, "data_time": 0.0013603448867797851, "loss": 0.010062522627413273, "time": 0.7286438703536987, "epoch": 5, "memory": 25074, "step": 27532} +{"lr": 0.0005270617127781042, "data_time": 0.001739501953125, "loss": 0.00918293297290802, "time": 0.762654972076416, "epoch": 5, "memory": 25074, "step": 27632} +{"lr": 0.0005270617127781042, "data_time": 0.0014391899108886718, "loss": 0.008797229966148734, "time": 0.7391265630722046, "epoch": 5, "memory": 25074, "step": 27732} +{"lr": 0.0005270617127781042, "data_time": 0.001235818862915039, "loss": 0.008473049057647586, "time": 0.7087532758712769, "epoch": 5, "memory": 25074, "step": 27832} +{"lr": 0.0005270617127781042, "data_time": 0.0016701698303222656, "loss": 0.009205130394548178, "time": 0.7324969530105591, "epoch": 5, "memory": 25074, "step": 27932} +{"lr": 0.0005270617127781042, "data_time": 0.0014713048934936524, "loss": 0.009877536445856094, "time": 0.7609650135040283, "epoch": 5, "memory": 25074, "step": 28032} +{"lr": 0.0005270617127781042, "data_time": 0.001546931266784668, "loss": 0.004014493990689516, "time": 0.7463459014892578, "epoch": 5, "memory": 25074, "step": 28132} +{"lr": 0.0005270617127781042, "data_time": 0.0013782501220703125, "loss": 0.013779939617961645, "time": 0.7689538478851319, "epoch": 5, "memory": 25074, "step": 28232} +{"lr": 0.0005270617127781042, "data_time": 0.0012537002563476562, "loss": 0.006694731302559376, "time": 0.7084328651428222, "epoch": 5, "memory": 25074, "step": 28332} +{"lr": 0.0005270617127781042, "data_time": 0.0015495538711547852, "loss": 0.005180006474256515, "time": 0.7607593536376953, "epoch": 5, "memory": 25074, "step": 28432} +{"lr": 0.0005270617127781042, "data_time": 0.0015459537506103515, "loss": 0.01184030957520008, "time": 0.7731788396835327, "epoch": 5, "memory": 25074, "step": 28532} +{"lr": 0.0005270617127781042, "data_time": 0.0015075922012329102, "loss": 0.005504630692303181, "time": 0.7314815044403076, "epoch": 5, "memory": 25074, "step": 28632} +{"lr": 0.0005270617127781042, "data_time": 0.0018773078918457031, "loss": 0.011145124398171901, "time": 0.802031421661377, "epoch": 5, "memory": 25074, "step": 28732} +{"accuracy/top1": 99.55179595947266, "data_time": 0.2146015587974997, "time": 0.6045234764323515, "step": 5} +{"lr": 0.000405, "data_time": 0.0013513803482055665, "loss": 0.0050857753027230505, "time": 0.7455167293548584, "epoch": 6, "memory": 25074, "step": 28890} +{"lr": 0.000405, "data_time": 0.0016866445541381836, "loss": 0.008379572164267301, "time": 0.8214875936508179, "epoch": 6, "memory": 25074, "step": 28990} +{"lr": 0.000405, "data_time": 0.0013658523559570313, "loss": 0.003400090243667364, "time": 0.7772329568862915, "epoch": 6, "memory": 25074, "step": 29090} +{"lr": 0.000405, "data_time": 0.001150059700012207, "loss": 0.006579025229439139, "time": 0.8252673864364624, "epoch": 6, "memory": 25074, "step": 29190} +{"lr": 0.000405, "data_time": 0.0011378765106201173, "loss": 0.007819813350215555, "time": 0.7901975870132446, "epoch": 6, "memory": 25074, "step": 29290} +{"lr": 0.000405, "data_time": 0.0014597177505493164, "loss": 0.0142287143971771, "time": 0.8013340473175049, "epoch": 6, "memory": 25074, "step": 29390} +{"lr": 0.000405, "data_time": 0.0017203569412231445, "loss": 0.0047938835341483355, "time": 0.7546528577804565, "epoch": 6, "memory": 25074, "step": 29490} +{"lr": 0.000405, "data_time": 0.0010878324508666992, "loss": 0.005435280269011855, "time": 0.7393974542617798, "epoch": 6, "memory": 25074, "step": 29590} +{"lr": 0.000405, "data_time": 0.0012293338775634765, "loss": 0.005714353220537305, "time": 0.7420917272567749, "epoch": 6, "memory": 25074, "step": 29690} +{"lr": 0.000405, "data_time": 0.001352977752685547, "loss": 0.006395082967355847, "time": 0.8012165784835815, "epoch": 6, "memory": 25074, "step": 29790} +{"lr": 0.000405, "data_time": 0.0012380838394165038, "loss": 0.010534848691895604, "time": 0.798853087425232, "epoch": 6, "memory": 25074, "step": 29890} +{"lr": 0.000405, "data_time": 0.0012102603912353515, "loss": 0.008739868039265275, "time": 0.7886104822158814, "epoch": 6, "memory": 25074, "step": 29990} +{"lr": 0.000405, "data_time": 0.0013063907623291015, "loss": 0.010199566651135684, "time": 0.7340371370315552, "epoch": 6, "memory": 25074, "step": 30090} +{"lr": 0.000405, "data_time": 0.001451396942138672, "loss": 0.007939047180116177, "time": 0.7761954069137573, "epoch": 6, "memory": 25074, "step": 30190} +{"lr": 0.000405, "data_time": 0.0017353296279907227, "loss": 0.006712845666334033, "time": 0.7268011331558227, "epoch": 6, "memory": 25074, "step": 30290} +{"lr": 0.000405, "data_time": 0.0010803461074829102, "loss": 0.007841747719794512, "time": 0.7981168031692505, "epoch": 6, "memory": 25074, "step": 30390} +{"lr": 0.000405, "data_time": 0.0016908884048461915, "loss": 0.0041880821343511345, "time": 0.7192023277282715, "epoch": 6, "memory": 25074, "step": 30490} +{"lr": 0.000405, "data_time": 0.0012493371963500977, "loss": 0.003917154716327786, "time": 0.7094663381576538, "epoch": 6, "memory": 25074, "step": 30590} +{"lr": 0.000405, "data_time": 0.0016638755798339844, "loss": 0.0069235051982104775, "time": 0.7100032091140747, "epoch": 6, "memory": 25074, "step": 30690} +{"lr": 0.000405, "data_time": 0.0013436079025268555, "loss": 0.0052233261056244375, "time": 0.7124608516693115, "epoch": 6, "memory": 25074, "step": 30790} +{"lr": 0.000405, "data_time": 0.0011367321014404297, "loss": 0.00867383647710085, "time": 0.7056768417358399, "epoch": 6, "memory": 25074, "step": 30890} +{"lr": 0.000405, "data_time": 0.001108694076538086, "loss": 0.0026851602364331485, "time": 0.7325651168823242, "epoch": 6, "memory": 25074, "step": 30990} +{"lr": 0.000405, "data_time": 0.0013169288635253907, "loss": 0.004504090035334229, "time": 0.725252103805542, "epoch": 6, "memory": 25074, "step": 31090} +{"lr": 0.000405, "data_time": 0.00191497802734375, "loss": 0.0022620852570980786, "time": 0.7325668811798096, "epoch": 6, "memory": 25074, "step": 31190} +{"lr": 0.000405, "data_time": 0.0014155387878417968, "loss": 0.007406285963952541, "time": 0.7438627481460571, "epoch": 6, "memory": 25074, "step": 31290} +{"lr": 0.000405, "data_time": 0.001490020751953125, "loss": 0.005735225137323141, "time": 0.7131868362426758, "epoch": 6, "memory": 25074, "step": 31390} +{"lr": 0.000405, "data_time": 0.0015232324600219726, "loss": 0.010248028067871928, "time": 0.7744721889495849, "epoch": 6, "memory": 25074, "step": 31490} +{"lr": 0.000405, "data_time": 0.0012341976165771485, "loss": 0.0037938281893730164, "time": 0.707684850692749, "epoch": 6, "memory": 25074, "step": 31590} +{"lr": 0.000405, "data_time": 0.0012534379959106446, "loss": 0.005043638637289405, "time": 0.7655620336532593, "epoch": 6, "memory": 25074, "step": 31690} +{"lr": 0.000405, "data_time": 0.0011670112609863282, "loss": 0.005976723181083799, "time": 0.7475017070770263, "epoch": 6, "memory": 25074, "step": 31790} +{"lr": 0.000405, "data_time": 0.0014744997024536133, "loss": 0.012173730600625276, "time": 0.7583656311035156, "epoch": 6, "memory": 25074, "step": 31890} +{"lr": 0.000405, "data_time": 0.0016011953353881835, "loss": 0.003958803694695234, "time": 0.8305601596832275, "epoch": 6, "memory": 25074, "step": 31990} +{"lr": 0.000405, "data_time": 0.0016053676605224609, "loss": 0.0069731857627630235, "time": 0.768882942199707, "epoch": 6, "memory": 25074, "step": 32090} +{"lr": 0.000405, "data_time": 0.0012295246124267578, "loss": 0.010267635434865951, "time": 0.7551312446594238, "epoch": 6, "memory": 25074, "step": 32190} +{"lr": 0.000405, "data_time": 0.0012998104095458985, "loss": 0.005389460688456893, "time": 0.7272004127502442, "epoch": 6, "memory": 25074, "step": 32290} diff --git a/convnext_small_4xb256_fake5m_lr8e-4_10epoch/20230601_151933/vis_data/vis_image/0000000.png_1.png b/convnext_small_4xb256_fake5m_lr8e-4_10epoch/20230601_151933/vis_data/vis_image/0000000.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..b2aa1aaaf9f97ce8b19f3bbce6b7024e76a7ba4c Binary files /dev/null and b/convnext_small_4xb256_fake5m_lr8e-4_10epoch/20230601_151933/vis_data/vis_image/0000000.png_1.png differ diff --git a/convnext_small_4xb256_fake5m_lr8e-4_10epoch/20230601_151933/vis_data/vis_image/0000000.png_2.png b/convnext_small_4xb256_fake5m_lr8e-4_10epoch/20230601_151933/vis_data/vis_image/0000000.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..3068bccfa8ed5f3ad7e0110060b2f984e12a99fb Binary files /dev/null and b/convnext_small_4xb256_fake5m_lr8e-4_10epoch/20230601_151933/vis_data/vis_image/0000000.png_2.png differ diff --git a/convnext_small_4xb256_fake5m_lr8e-4_10epoch/20230601_151933/vis_data/vis_image/0000000.png_3.png b/convnext_small_4xb256_fake5m_lr8e-4_10epoch/20230601_151933/vis_data/vis_image/0000000.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..6f264824565adf0630fdfc2e331f1ad6f2495172 Binary files /dev/null and b/convnext_small_4xb256_fake5m_lr8e-4_10epoch/20230601_151933/vis_data/vis_image/0000000.png_3.png differ diff --git a/convnext_small_4xb256_fake5m_lr8e-4_10epoch/20230601_151933/vis_data/vis_image/0000000.png_4.png b/convnext_small_4xb256_fake5m_lr8e-4_10epoch/20230601_151933/vis_data/vis_image/0000000.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..6f264824565adf0630fdfc2e331f1ad6f2495172 Binary files /dev/null and b/convnext_small_4xb256_fake5m_lr8e-4_10epoch/20230601_151933/vis_data/vis_image/0000000.png_4.png differ diff --git a/convnext_small_4xb256_fake5m_lr8e-4_10epoch/20230601_151933/vis_data/vis_image/0000000.png_5.png b/convnext_small_4xb256_fake5m_lr8e-4_10epoch/20230601_151933/vis_data/vis_image/0000000.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..6f264824565adf0630fdfc2e331f1ad6f2495172 Binary files /dev/null and b/convnext_small_4xb256_fake5m_lr8e-4_10epoch/20230601_151933/vis_data/vis_image/0000000.png_5.png differ diff --git a/convnext_small_4xb256_fake5m_lr8e-4_10epoch/convnext_small_4xb256_fake5m_lr8e-4_10epoch.py b/convnext_small_4xb256_fake5m_lr8e-4_10epoch/convnext_small_4xb256_fake5m_lr8e-4_10epoch.py new file mode 100644 index 0000000000000000000000000000000000000000..e9462c7bdf98df98a9c3d600ff4a02772382f3d3 --- /dev/null +++ b/convnext_small_4xb256_fake5m_lr8e-4_10epoch/convnext_small_4xb256_fake5m_lr8e-4_10epoch.py @@ -0,0 +1,169 @@ +optim_wrapper = dict( + optimizer=dict( + type='AdamW', + lr=0.0008, + weight_decay=0.05, + eps=1e-08, + betas=(0.9, 0.999), + _scope_='mmpretrain'), + paramwise_cfg=dict( + norm_decay_mult=0.0, + bias_decay_mult=0.0, + flat_decay_mult=0.0, + custom_keys=dict({ + '.absolute_pos_embed': dict(decay_mult=0.0), + '.relative_position_bias_table': dict(decay_mult=0.0) + })), + type='AmpOptimWrapper', + dtype='bfloat16', + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=True, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=4096) +model = dict( + type='ImageClassifier', + backbone=dict(type='ConvNeXt', arch='small', drop_path_rate=0.4), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=768, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + init_cfg=None), + init_cfg=dict( + type='TruncNormal', layer=['Conv2d', 'Linear'], std=0.02, bias=0.0), + train_cfg=None) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stablediffusionV1-5R2-dpmsolver-25-5m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/convnext_small_4xb256_fake5m_lr8e-4_10epoch' diff --git a/convnext_small_4xb256_fake5m_lr8e-4_10epoch/epoch_1.pth b/convnext_small_4xb256_fake5m_lr8e-4_10epoch/epoch_1.pth new file mode 100644 index 0000000000000000000000000000000000000000..781f0f044abfc76171ec1e4bbef377ff6cf23635 --- /dev/null +++ b/convnext_small_4xb256_fake5m_lr8e-4_10epoch/epoch_1.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:eefde457282cb46bb571bb509474a9f393cdfd0cd60b70b91e44c2a2d6d5a1dc +size 792306720 diff --git a/convnext_small_4xb256_fake5m_lr8e-4_10epoch/epoch_2.pth b/convnext_small_4xb256_fake5m_lr8e-4_10epoch/epoch_2.pth new file mode 100644 index 0000000000000000000000000000000000000000..19e9a95675796a540af7557b783a12f6cda53a76 --- /dev/null +++ b/convnext_small_4xb256_fake5m_lr8e-4_10epoch/epoch_2.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:df436c06ab7f08d7b90bbbdd5794366251f42e3fea219844baf472330a2e439f +size 792728992 diff --git a/convnext_small_4xb256_fake5m_lr8e-4_10epoch/epoch_3.pth b/convnext_small_4xb256_fake5m_lr8e-4_10epoch/epoch_3.pth new file mode 100644 index 0000000000000000000000000000000000000000..e83995da496bcb0a90c6559bb59864e0f2dbc9d0 --- /dev/null +++ b/convnext_small_4xb256_fake5m_lr8e-4_10epoch/epoch_3.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:a55f26de0c80dd5411fb7916dff033cc75ca3848085bf916b06cc474f8cfdff6 +size 793143456 diff --git a/convnext_small_4xb256_fake5m_lr8e-4_10epoch/epoch_4.pth b/convnext_small_4xb256_fake5m_lr8e-4_10epoch/epoch_4.pth new file mode 100644 index 0000000000000000000000000000000000000000..7c91199c7ca5b5f6a2bddadd51ac7137adbf8171 --- /dev/null +++ b/convnext_small_4xb256_fake5m_lr8e-4_10epoch/epoch_4.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:6b70b9b1aab3541a4032fc645199202d3c19ae8ce9abc41f31806ae1369ef28f +size 793562208 diff --git a/convnext_small_4xb256_fake5m_lr8e-4_10epoch/epoch_5.pth b/convnext_small_4xb256_fake5m_lr8e-4_10epoch/epoch_5.pth new file mode 100644 index 0000000000000000000000000000000000000000..8e8b168584ecf14a3beb41974ec9988b51b0d943 --- /dev/null +++ b/convnext_small_4xb256_fake5m_lr8e-4_10epoch/epoch_5.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:8c0e2489686e49101324e407640ad9db980b8173c1cbb0a2de0d4c19a51ad516 +size 793979104 diff --git a/convnext_small_4xb256_fake5m_lr8e-4_10epoch/last_checkpoint b/convnext_small_4xb256_fake5m_lr8e-4_10epoch/last_checkpoint new file mode 100644 index 0000000000000000000000000000000000000000..03f119273886cbfee29887fe609138133014f887 --- /dev/null +++ b/convnext_small_4xb256_fake5m_lr8e-4_10epoch/last_checkpoint @@ -0,0 +1 @@ +/mnt/petrelfs/luzeyu/workspace/fakebench/mmpretrain/workdir/convnext_small_4xb256_fake5m_lr8e-4_10epoch/epoch_5.pth \ No newline at end of file diff --git a/convnext_small_4xb256_fake7m-lr4e-4/20230601_161522/20230601_161522.log b/convnext_small_4xb256_fake7m-lr4e-4/20230601_161522/20230601_161522.log new file mode 100644 index 0000000000000000000000000000000000000000..f3b5899b2580c3411802d676b3012f95e5695502 --- /dev/null +++ b/convnext_small_4xb256_fake7m-lr4e-4/20230601_161522/20230601_161522.log @@ -0,0 +1,1602 @@ +2023/06/01 16:15:27 - mmengine - INFO - +------------------------------------------------------------ +System environment: + sys.platform: linux + Python: 3.10.9 (main, Mar 8 2023, 10:47:38) [GCC 11.2.0] + CUDA available: True + numpy_random_seed: 1120907987 + GPU 0,1,2,3: NVIDIA A100-SXM4-80GB + CUDA_HOME: /mnt/petrelfs/share/cuda-11.6 + NVCC: Cuda compilation tools, release 11.6, V11.6.124 + GCC: gcc (GCC) 4.8.5 20150623 (Red Hat 4.8.5-44) + PyTorch: 1.13.1 + PyTorch compiling details: PyTorch built with: + - GCC 9.3 + - C++ Version: 201402 + - Intel(R) oneAPI Math Kernel Library Version 2021.4-Product Build 20210904 for Intel(R) 64 architecture applications + - Intel(R) MKL-DNN v2.6.0 (Git Hash 52b5f107dd9cf10910aaa19cb47f3abf9b349815) + - OpenMP 201511 (a.k.a. OpenMP 4.5) + - LAPACK is enabled (usually provided by MKL) + - NNPACK is enabled + - CPU capability usage: AVX2 + - CUDA Runtime 11.6 + - NVCC architecture flags: -gencode;arch=compute_37,code=sm_37;-gencode;arch=compute_50,code=sm_50;-gencode;arch=compute_60,code=sm_60;-gencode;arch=compute_61,code=sm_61;-gencode;arch=compute_70,code=sm_70;-gencode;arch=compute_75,code=sm_75;-gencode;arch=compute_80,code=sm_80;-gencode;arch=compute_86,code=sm_86;-gencode;arch=compute_37,code=compute_37 + - CuDNN 8.3.2 (built against CUDA 11.5) + - Magma 2.6.1 + - Build settings: BLAS_INFO=mkl, BUILD_TYPE=Release, CUDA_VERSION=11.6, CUDNN_VERSION=8.3.2, CXX_COMPILER=/opt/rh/devtoolset-9/root/usr/bin/c++, CXX_FLAGS= -fabi-version=11 -Wno-deprecated -fvisibility-inlines-hidden -DUSE_PTHREADPOOL -fopenmp -DNDEBUG -DUSE_KINETO -DUSE_FBGEMM -DUSE_QNNPACK -DUSE_PYTORCH_QNNPACK -DUSE_XNNPACK -DSYMBOLICATE_MOBILE_DEBUG_HANDLE -DEDGE_PROFILER_USE_KINETO -O2 -fPIC -Wno-narrowing -Wall -Wextra -Werror=return-type -Werror=non-virtual-dtor -Wno-missing-field-initializers -Wno-type-limits -Wno-array-bounds -Wno-unknown-pragmas -Wunused-local-typedefs -Wno-unused-parameter -Wno-unused-function -Wno-unused-result -Wno-strict-overflow -Wno-strict-aliasing -Wno-error=deprecated-declarations -Wno-stringop-overflow -Wno-psabi -Wno-error=pedantic -Wno-error=redundant-decls -Wno-error=old-style-cast -fdiagnostics-color=always -faligned-new -Wno-unused-but-set-variable -Wno-maybe-uninitialized -fno-math-errno -fno-trapping-math -Werror=format -Werror=cast-function-type -Wno-stringop-overflow, LAPACK_INFO=mkl, PERF_WITH_AVX=1, PERF_WITH_AVX2=1, PERF_WITH_AVX512=1, TORCH_VERSION=1.13.1, USE_CUDA=ON, USE_CUDNN=ON, USE_EXCEPTION_PTR=1, USE_GFLAGS=OFF, USE_GLOG=OFF, USE_MKL=ON, USE_MKLDNN=ON, USE_MPI=OFF, USE_NCCL=ON, USE_NNPACK=ON, USE_OPENMP=ON, USE_ROCM=OFF, + + TorchVision: 0.14.1 + OpenCV: 4.7.0 + MMEngine: 0.7.3 + +Runtime environment: + cudnn_benchmark: True + mp_cfg: {'mp_start_method': 'fork', 'opencv_num_threads': 0} + dist_cfg: {'backend': 'nccl'} + seed: None + deterministic: False + Distributed launcher: slurm + Distributed training: True + GPU number: 4 +------------------------------------------------------------ + +2023/06/01 16:15:29 - mmengine - INFO - Config: +optim_wrapper = dict( + optimizer=dict( + type='AdamW', + lr=0.0004, + weight_decay=0.05, + eps=1e-08, + betas=(0.9, 0.999), + _scope_='mmpretrain'), + paramwise_cfg=dict( + norm_decay_mult=0.0, + bias_decay_mult=0.0, + flat_decay_mult=0.0, + custom_keys=dict({ + '.absolute_pos_embed': dict(decay_mult=0.0), + '.relative_position_bias_table': dict(decay_mult=0.0) + })), + type='AmpOptimWrapper', + dtype='bfloat16', + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=True, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=4096) +model = dict( + type='ImageClassifier', + backbone=dict(type='ConvNeXt', arch='small', drop_path_rate=0.4), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=768, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + init_cfg=None), + init_cfg=dict( + type='TruncNormal', layer=['Conv2d', 'Linear'], std=0.02, bias=0.0), + train_cfg=None) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stablediffusionV1-5R2-dpmsolver-25-5m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/if-dpmsolver++-50-20w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stablediffusionV2-1-dpmsolver-25-190w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/convnext_small_4xb256_fake7m-lr4e-4' + +2023/06/01 16:15:35 - mmengine - INFO - Hooks will be executed in the following order: +before_run: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_load_checkpoint: +(ABOVE_NORMAL) EMAHook + -------------------- +before_train: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_train_epoch: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(NORMAL ) DistSamplerSeedHook + -------------------- +before_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook + -------------------- +after_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +after_train_epoch: +(NORMAL ) IterTimerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_val_epoch: +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook + -------------------- +before_val_iter: +(NORMAL ) IterTimerHook + -------------------- +after_val_iter: +(NORMAL ) IterTimerHook +(NORMAL ) VisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_val_epoch: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_save_checkpoint: +(ABOVE_NORMAL) EMAHook + -------------------- +after_train: +(VERY_LOW ) CheckpointHook + -------------------- +before_test_epoch: +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook + -------------------- +before_test_iter: +(NORMAL ) IterTimerHook + -------------------- +after_test_iter: +(NORMAL ) IterTimerHook +(NORMAL ) VisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_test_epoch: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_run: +(BELOW_NORMAL) LoggerHook + -------------------- +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.0.0.bias:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.0.1.weight:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.0.1.bias:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.1.0.weight:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.1.0.bias:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.1.1.bias:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.2.0.weight:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.2.0.bias:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.2.1.bias:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.3.0.weight:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.3.0.bias:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.3.1.bias:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.0.0.gamma:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.0.0.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.0.0.norm.weight:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.0.0.norm.bias:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.0.0.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.0.0.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.0.1.gamma:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.0.1.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.0.1.norm.weight:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.0.1.norm.bias:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.0.1.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.0.1.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.0.2.gamma:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.0.2.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.0.2.norm.weight:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.0.2.norm.bias:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.0.2.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.0.2.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.1.0.gamma:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.1.0.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.1.0.norm.weight:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.1.0.norm.bias:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.1.0.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.1.0.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.1.1.gamma:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.1.1.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.1.1.norm.weight:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.1.1.norm.bias:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.1.1.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.1.1.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.1.2.gamma:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.1.2.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.1.2.norm.weight:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.1.2.norm.bias:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.1.2.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.1.2.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.0.gamma:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.0.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.0.norm.weight:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.0.norm.bias:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.0.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.0.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.1.gamma:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.1.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.1.norm.weight:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.1.norm.bias:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.1.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.1.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.2.gamma:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.2.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.2.norm.weight:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.2.norm.bias:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.2.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.2.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.3.gamma:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.3.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.3.norm.weight:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.3.norm.bias:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.3.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.3.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.4.gamma:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.4.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.4.norm.weight:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.4.norm.bias:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.4.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.4.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.5.gamma:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.5.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.5.norm.weight:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.5.norm.bias:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.5.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.5.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.6.gamma:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.6.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.6.norm.weight:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.6.norm.bias:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.6.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.6.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.7.gamma:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.7.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.7.norm.weight:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.7.norm.bias:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.7.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.7.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.8.gamma:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.8.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.8.norm.weight:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.8.norm.bias:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.8.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.8.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.9.gamma:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.9.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.9.norm.weight:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.9.norm.bias:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.9.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.9.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.10.gamma:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.10.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.10.norm.weight:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.10.norm.bias:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.10.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.10.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.11.gamma:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.11.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.11.norm.weight:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.11.norm.bias:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.11.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.11.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.12.gamma:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.12.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.12.norm.weight:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.12.norm.bias:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.12.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.12.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.13.gamma:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.13.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.13.norm.weight:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.13.norm.bias:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.13.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.13.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.14.gamma:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.14.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.14.norm.weight:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.14.norm.bias:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.14.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.14.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.15.gamma:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.15.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.15.norm.weight:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.15.norm.bias:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.15.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.15.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.16.gamma:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.16.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.16.norm.weight:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.16.norm.bias:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.16.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.16.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.17.gamma:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.17.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.17.norm.weight:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.17.norm.bias:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.17.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.17.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.18.gamma:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.18.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.18.norm.weight:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.18.norm.bias:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.18.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.18.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.19.gamma:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.19.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.19.norm.weight:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.19.norm.bias:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.19.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.19.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.20.gamma:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.20.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.20.norm.weight:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.20.norm.bias:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.20.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.20.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.21.gamma:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.21.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.21.norm.weight:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.21.norm.bias:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.21.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.21.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.22.gamma:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.22.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.22.norm.weight:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.22.norm.bias:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.22.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.22.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.23.gamma:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.23.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.23.norm.weight:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.23.norm.bias:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.23.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.23.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.24.gamma:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.24.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.24.norm.weight:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.24.norm.bias:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.24.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.24.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.25.gamma:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.25.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.25.norm.weight:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.25.norm.bias:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.25.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.25.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.26.gamma:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.26.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.26.norm.weight:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.26.norm.bias:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.26.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.2.26.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.3.0.gamma:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.3.0.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.3.0.norm.weight:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.3.0.norm.bias:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.3.0.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.3.0.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.3.1.gamma:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.3.1.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.3.1.norm.weight:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.3.1.norm.bias:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.3.1.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.3.1.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.3.2.gamma:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.3.2.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.3.2.norm.weight:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.3.2.norm.bias:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.3.2.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.stages.3.2.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.norm3.weight:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- backbone.norm3.bias:weight_decay=0.0 +2023/06/01 16:16:14 - mmengine - INFO - paramwise_options -- head.fc.bias:weight_decay=0.0 +Name of parameter - Initialization information + +backbone.downsample_layers.0.0.weight - torch.Size([96, 3, 4, 4]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.downsample_layers.0.0.bias - torch.Size([96]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.downsample_layers.0.1.weight - torch.Size([96]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.downsample_layers.0.1.bias - torch.Size([96]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.downsample_layers.1.0.weight - torch.Size([96]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.downsample_layers.1.0.bias - torch.Size([96]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.downsample_layers.1.1.weight - torch.Size([192, 96, 2, 2]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.downsample_layers.1.1.bias - torch.Size([192]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.downsample_layers.2.0.weight - torch.Size([192]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.downsample_layers.2.0.bias - torch.Size([192]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.downsample_layers.2.1.weight - torch.Size([384, 192, 2, 2]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.downsample_layers.2.1.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.downsample_layers.3.0.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.downsample_layers.3.0.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.downsample_layers.3.1.weight - torch.Size([768, 384, 2, 2]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.downsample_layers.3.1.bias - torch.Size([768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.0.gamma - torch.Size([96]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.0.0.depthwise_conv.weight - torch.Size([96, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.0.depthwise_conv.bias - torch.Size([96]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.0.norm.weight - torch.Size([96]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.0.0.norm.bias - torch.Size([96]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.0.0.pointwise_conv1.weight - torch.Size([384, 96]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.0.pointwise_conv1.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.0.pointwise_conv2.weight - torch.Size([96, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.0.pointwise_conv2.bias - torch.Size([96]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.1.gamma - torch.Size([96]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.0.1.depthwise_conv.weight - torch.Size([96, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.1.depthwise_conv.bias - torch.Size([96]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.1.norm.weight - torch.Size([96]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.0.1.norm.bias - torch.Size([96]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.0.1.pointwise_conv1.weight - torch.Size([384, 96]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.1.pointwise_conv1.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.1.pointwise_conv2.weight - torch.Size([96, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.1.pointwise_conv2.bias - torch.Size([96]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.2.gamma - torch.Size([96]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.0.2.depthwise_conv.weight - torch.Size([96, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.2.depthwise_conv.bias - torch.Size([96]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.2.norm.weight - torch.Size([96]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.0.2.norm.bias - torch.Size([96]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.0.2.pointwise_conv1.weight - torch.Size([384, 96]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.2.pointwise_conv1.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.2.pointwise_conv2.weight - torch.Size([96, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.2.pointwise_conv2.bias - torch.Size([96]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.0.gamma - torch.Size([192]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.1.0.depthwise_conv.weight - torch.Size([192, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.0.depthwise_conv.bias - torch.Size([192]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.0.norm.weight - torch.Size([192]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.1.0.norm.bias - torch.Size([192]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.1.0.pointwise_conv1.weight - torch.Size([768, 192]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.0.pointwise_conv1.bias - torch.Size([768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.0.pointwise_conv2.weight - torch.Size([192, 768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.0.pointwise_conv2.bias - torch.Size([192]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.1.gamma - torch.Size([192]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.1.1.depthwise_conv.weight - torch.Size([192, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.1.depthwise_conv.bias - torch.Size([192]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.1.norm.weight - torch.Size([192]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.1.1.norm.bias - torch.Size([192]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.1.1.pointwise_conv1.weight - torch.Size([768, 192]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.1.pointwise_conv1.bias - torch.Size([768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.1.pointwise_conv2.weight - torch.Size([192, 768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.1.pointwise_conv2.bias - torch.Size([192]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.2.gamma - torch.Size([192]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.1.2.depthwise_conv.weight - torch.Size([192, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.2.depthwise_conv.bias - torch.Size([192]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.2.norm.weight - torch.Size([192]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.1.2.norm.bias - torch.Size([192]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.1.2.pointwise_conv1.weight - torch.Size([768, 192]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.2.pointwise_conv1.bias - torch.Size([768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.2.pointwise_conv2.weight - torch.Size([192, 768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.2.pointwise_conv2.bias - torch.Size([192]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.0.gamma - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.0.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.0.depthwise_conv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.0.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.0.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.0.pointwise_conv1.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.0.pointwise_conv1.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.0.pointwise_conv2.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.0.pointwise_conv2.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.1.gamma - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.1.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.1.depthwise_conv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.1.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.1.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.1.pointwise_conv1.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.1.pointwise_conv1.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.1.pointwise_conv2.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.1.pointwise_conv2.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.2.gamma - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.2.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.2.depthwise_conv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.2.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.2.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.2.pointwise_conv1.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.2.pointwise_conv1.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.2.pointwise_conv2.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.2.pointwise_conv2.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.3.gamma - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.3.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.3.depthwise_conv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.3.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.3.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.3.pointwise_conv1.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.3.pointwise_conv1.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.3.pointwise_conv2.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.3.pointwise_conv2.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.4.gamma - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.4.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.4.depthwise_conv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.4.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.4.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.4.pointwise_conv1.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.4.pointwise_conv1.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.4.pointwise_conv2.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.4.pointwise_conv2.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.5.gamma - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.5.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.5.depthwise_conv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.5.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.5.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.5.pointwise_conv1.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.5.pointwise_conv1.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.5.pointwise_conv2.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.5.pointwise_conv2.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.6.gamma - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.6.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.6.depthwise_conv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.6.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.6.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.6.pointwise_conv1.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.6.pointwise_conv1.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.6.pointwise_conv2.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.6.pointwise_conv2.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.7.gamma - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.7.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.7.depthwise_conv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.7.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.7.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.7.pointwise_conv1.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.7.pointwise_conv1.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.7.pointwise_conv2.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.7.pointwise_conv2.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.8.gamma - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.8.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.8.depthwise_conv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.8.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.8.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.8.pointwise_conv1.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.8.pointwise_conv1.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.8.pointwise_conv2.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.8.pointwise_conv2.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.9.gamma - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.9.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.9.depthwise_conv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.9.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.9.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.9.pointwise_conv1.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.9.pointwise_conv1.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.9.pointwise_conv2.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.9.pointwise_conv2.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.10.gamma - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.10.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.10.depthwise_conv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.10.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.10.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.10.pointwise_conv1.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.10.pointwise_conv1.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.10.pointwise_conv2.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.10.pointwise_conv2.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.11.gamma - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.11.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.11.depthwise_conv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.11.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.11.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.11.pointwise_conv1.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.11.pointwise_conv1.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.11.pointwise_conv2.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.11.pointwise_conv2.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.12.gamma - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.12.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.12.depthwise_conv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.12.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.12.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.12.pointwise_conv1.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.12.pointwise_conv1.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.12.pointwise_conv2.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.12.pointwise_conv2.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.13.gamma - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.13.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.13.depthwise_conv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.13.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.13.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.13.pointwise_conv1.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.13.pointwise_conv1.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.13.pointwise_conv2.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.13.pointwise_conv2.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.14.gamma - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.14.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.14.depthwise_conv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.14.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.14.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.14.pointwise_conv1.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.14.pointwise_conv1.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.14.pointwise_conv2.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.14.pointwise_conv2.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.15.gamma - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.15.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.15.depthwise_conv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.15.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.15.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.15.pointwise_conv1.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.15.pointwise_conv1.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.15.pointwise_conv2.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.15.pointwise_conv2.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.16.gamma - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.16.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.16.depthwise_conv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.16.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.16.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.16.pointwise_conv1.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.16.pointwise_conv1.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.16.pointwise_conv2.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.16.pointwise_conv2.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.17.gamma - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.17.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.17.depthwise_conv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.17.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.17.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.17.pointwise_conv1.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.17.pointwise_conv1.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.17.pointwise_conv2.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.17.pointwise_conv2.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.18.gamma - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.18.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.18.depthwise_conv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.18.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.18.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.18.pointwise_conv1.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.18.pointwise_conv1.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.18.pointwise_conv2.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.18.pointwise_conv2.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.19.gamma - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.19.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.19.depthwise_conv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.19.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.19.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.19.pointwise_conv1.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.19.pointwise_conv1.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.19.pointwise_conv2.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.19.pointwise_conv2.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.20.gamma - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.20.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.20.depthwise_conv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.20.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.20.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.20.pointwise_conv1.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.20.pointwise_conv1.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.20.pointwise_conv2.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.20.pointwise_conv2.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.21.gamma - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.21.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.21.depthwise_conv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.21.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.21.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.21.pointwise_conv1.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.21.pointwise_conv1.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.21.pointwise_conv2.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.21.pointwise_conv2.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.22.gamma - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.22.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.22.depthwise_conv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.22.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.22.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.22.pointwise_conv1.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.22.pointwise_conv1.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.22.pointwise_conv2.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.22.pointwise_conv2.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.23.gamma - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.23.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.23.depthwise_conv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.23.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.23.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.23.pointwise_conv1.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.23.pointwise_conv1.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.23.pointwise_conv2.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.23.pointwise_conv2.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.24.gamma - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.24.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.24.depthwise_conv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.24.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.24.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.24.pointwise_conv1.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.24.pointwise_conv1.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.24.pointwise_conv2.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.24.pointwise_conv2.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.25.gamma - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.25.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.25.depthwise_conv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.25.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.25.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.25.pointwise_conv1.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.25.pointwise_conv1.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.25.pointwise_conv2.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.25.pointwise_conv2.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.26.gamma - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.26.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.26.depthwise_conv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.26.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.26.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.26.pointwise_conv1.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.26.pointwise_conv1.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.26.pointwise_conv2.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.26.pointwise_conv2.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.0.gamma - torch.Size([768]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.3.0.depthwise_conv.weight - torch.Size([768, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.0.depthwise_conv.bias - torch.Size([768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.0.norm.weight - torch.Size([768]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.3.0.norm.bias - torch.Size([768]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.3.0.pointwise_conv1.weight - torch.Size([3072, 768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.0.pointwise_conv1.bias - torch.Size([3072]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.0.pointwise_conv2.weight - torch.Size([768, 3072]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.0.pointwise_conv2.bias - torch.Size([768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.1.gamma - torch.Size([768]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.3.1.depthwise_conv.weight - torch.Size([768, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.1.depthwise_conv.bias - torch.Size([768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.1.norm.weight - torch.Size([768]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.3.1.norm.bias - torch.Size([768]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.3.1.pointwise_conv1.weight - torch.Size([3072, 768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.1.pointwise_conv1.bias - torch.Size([3072]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.1.pointwise_conv2.weight - torch.Size([768, 3072]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.1.pointwise_conv2.bias - torch.Size([768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.2.gamma - torch.Size([768]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.3.2.depthwise_conv.weight - torch.Size([768, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.2.depthwise_conv.bias - torch.Size([768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.2.norm.weight - torch.Size([768]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.3.2.norm.bias - torch.Size([768]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.3.2.pointwise_conv1.weight - torch.Size([3072, 768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.2.pointwise_conv1.bias - torch.Size([3072]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.2.pointwise_conv2.weight - torch.Size([768, 3072]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.2.pointwise_conv2.bias - torch.Size([768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.norm3.weight - torch.Size([768]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.norm3.bias - torch.Size([768]): +The value is the same before and after calling `init_weights` of ImageClassifier + +head.fc.weight - torch.Size([2, 768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +head.fc.bias - torch.Size([2]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 +2023/06/01 16:16:16 - mmengine - WARNING - "FileClient" will be deprecated in future. Please use io functions in https://mmengine.readthedocs.io/en/latest/api/fileio.html#file-io +2023/06/01 16:16:16 - mmengine - WARNING - "HardDiskBackend" is the alias of "LocalBackend" and the former will be deprecated in future. +2023/06/01 16:16:16 - mmengine - INFO - Checkpoints will be saved to /mnt/petrelfs/luzeyu/workspace/fakebench/mmpretrain/workdir/convnext_small_4xb256_fake7m-lr4e-4. +2023/06/01 16:17:39 - mmengine - INFO - Epoch(train) [1][ 100/7800] lr: 4.0000e-04 eta: 17:58:30 time: 0.7479 data_time: 0.0016 memory: 25071 loss: 0.5570 +2023/06/01 16:18:50 - mmengine - INFO - Epoch(train) [1][ 200/7800] lr: 4.0000e-04 eta: 16:40:25 time: 0.7111 data_time: 0.0013 memory: 25071 loss: 0.5156 +2023/06/01 16:20:02 - mmengine - INFO - Epoch(train) [1][ 300/7800] lr: 4.0000e-04 eta: 16:15:16 time: 0.7357 data_time: 0.0014 memory: 25071 loss: 0.4904 +2023/06/01 16:21:13 - mmengine - INFO - Epoch(train) [1][ 400/7800] lr: 4.0000e-04 eta: 16:00:42 time: 0.7196 data_time: 0.0015 memory: 25071 loss: 0.4712 diff --git a/convnext_small_4xb256_fake7m-lr4e-4/20230601_161522/vis_data/20230601_161522.json b/convnext_small_4xb256_fake7m-lr4e-4/20230601_161522/vis_data/20230601_161522.json new file mode 100644 index 0000000000000000000000000000000000000000..4f5ccee2641ba4dd65311a6ec58d32d6e7385b34 --- /dev/null +++ b/convnext_small_4xb256_fake7m-lr4e-4/20230601_161522/vis_data/20230601_161522.json @@ -0,0 +1,4 @@ +{"lr": 0.0004, "data_time": 0.0016201496124267577, "loss": 0.5570417404174804, "time": 0.7478580951690674, "epoch": 1, "memory": 25071, "step": 100} +{"lr": 0.0004, "data_time": 0.0013155698776245116, "loss": 0.5156107902526855, "time": 0.7111196279525757, "epoch": 1, "memory": 25071, "step": 200} +{"lr": 0.0004, "data_time": 0.0014447212219238282, "loss": 0.4904367089271545, "time": 0.7357485294342041, "epoch": 1, "memory": 25071, "step": 300} +{"lr": 0.0004, "data_time": 0.0014618158340454102, "loss": 0.4712392330169678, "time": 0.7196192026138306, "epoch": 1, "memory": 25071, "step": 400} diff --git a/convnext_small_4xb256_fake7m-lr4e-4/20230601_161522/vis_data/config.py b/convnext_small_4xb256_fake7m-lr4e-4/20230601_161522/vis_data/config.py new file mode 100644 index 0000000000000000000000000000000000000000..0e6e101fa3314a1c359ce105454f400d187c5c3d --- /dev/null +++ b/convnext_small_4xb256_fake7m-lr4e-4/20230601_161522/vis_data/config.py @@ -0,0 +1,203 @@ +optim_wrapper = dict( + optimizer=dict( + type='AdamW', + lr=0.0004, + weight_decay=0.05, + eps=1e-08, + betas=(0.9, 0.999), + _scope_='mmpretrain'), + paramwise_cfg=dict( + norm_decay_mult=0.0, + bias_decay_mult=0.0, + flat_decay_mult=0.0, + custom_keys=dict({ + '.absolute_pos_embed': dict(decay_mult=0.0), + '.relative_position_bias_table': dict(decay_mult=0.0) + })), + type='AmpOptimWrapper', + dtype='bfloat16', + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=True, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=4096) +model = dict( + type='ImageClassifier', + backbone=dict(type='ConvNeXt', arch='small', drop_path_rate=0.4), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=768, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + init_cfg=None), + init_cfg=dict( + type='TruncNormal', layer=['Conv2d', 'Linear'], std=0.02, bias=0.0), + train_cfg=None) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stablediffusionV1-5R2-dpmsolver-25-5m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/if-dpmsolver++-50-20w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stablediffusionV2-1-dpmsolver-25-190w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/convnext_small_4xb256_fake7m-lr4e-4' diff --git a/convnext_small_4xb256_fake7m-lr4e-4/20230601_161522/vis_data/events.out.tfevents.1685607332.SH-IDC1-10-140-24-78.177218.0 b/convnext_small_4xb256_fake7m-lr4e-4/20230601_161522/vis_data/events.out.tfevents.1685607332.SH-IDC1-10-140-24-78.177218.0 new file mode 100644 index 0000000000000000000000000000000000000000..6f1b0a193469952d25b202bff6c03650a6a4837f --- /dev/null +++ b/convnext_small_4xb256_fake7m-lr4e-4/20230601_161522/vis_data/events.out.tfevents.1685607332.SH-IDC1-10-140-24-78.177218.0 @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:d0f9471956652d02bc6f1b3e36ecb87375f42112a95aca045d3ffd28009cad1a +size 8299 diff --git a/convnext_small_4xb256_fake7m-lr4e-4/20230601_161522/vis_data/scalars.json b/convnext_small_4xb256_fake7m-lr4e-4/20230601_161522/vis_data/scalars.json new file mode 100644 index 0000000000000000000000000000000000000000..4f5ccee2641ba4dd65311a6ec58d32d6e7385b34 --- /dev/null +++ b/convnext_small_4xb256_fake7m-lr4e-4/20230601_161522/vis_data/scalars.json @@ -0,0 +1,4 @@ +{"lr": 0.0004, "data_time": 0.0016201496124267577, "loss": 0.5570417404174804, "time": 0.7478580951690674, "epoch": 1, "memory": 25071, "step": 100} +{"lr": 0.0004, "data_time": 0.0013155698776245116, "loss": 0.5156107902526855, "time": 0.7111196279525757, "epoch": 1, "memory": 25071, "step": 200} +{"lr": 0.0004, "data_time": 0.0014447212219238282, "loss": 0.4904367089271545, "time": 0.7357485294342041, "epoch": 1, "memory": 25071, "step": 300} +{"lr": 0.0004, "data_time": 0.0014618158340454102, "loss": 0.4712392330169678, "time": 0.7196192026138306, "epoch": 1, "memory": 25071, "step": 400} diff --git a/convnext_small_4xb256_fake7m-lr4e-4/20230601_193207/20230601_193207.log b/convnext_small_4xb256_fake7m-lr4e-4/20230601_193207/20230601_193207.log new file mode 100644 index 0000000000000000000000000000000000000000..188ac4f0570606808a2db23edb6579eab8c1aef3 --- /dev/null +++ b/convnext_small_4xb256_fake7m-lr4e-4/20230601_193207/20230601_193207.log @@ -0,0 +1,2484 @@ +2023/06/01 19:32:11 - mmengine - INFO - +------------------------------------------------------------ +System environment: + sys.platform: linux + Python: 3.10.9 (main, Mar 8 2023, 10:47:38) [GCC 11.2.0] + CUDA available: True + numpy_random_seed: 1782260821 + GPU 0,1,2,3: NVIDIA A100-SXM4-80GB + CUDA_HOME: /mnt/petrelfs/share/cuda-11.6 + NVCC: Cuda compilation tools, release 11.6, V11.6.124 + GCC: gcc (GCC) 4.8.5 20150623 (Red Hat 4.8.5-44) + PyTorch: 1.13.1 + PyTorch compiling details: PyTorch built with: + - GCC 9.3 + - C++ Version: 201402 + - Intel(R) oneAPI Math Kernel Library Version 2021.4-Product Build 20210904 for Intel(R) 64 architecture applications + - Intel(R) MKL-DNN v2.6.0 (Git Hash 52b5f107dd9cf10910aaa19cb47f3abf9b349815) + - OpenMP 201511 (a.k.a. OpenMP 4.5) + - LAPACK is enabled (usually provided by MKL) + - NNPACK is enabled + - CPU capability usage: AVX2 + - CUDA Runtime 11.6 + - NVCC architecture flags: -gencode;arch=compute_37,code=sm_37;-gencode;arch=compute_50,code=sm_50;-gencode;arch=compute_60,code=sm_60;-gencode;arch=compute_61,code=sm_61;-gencode;arch=compute_70,code=sm_70;-gencode;arch=compute_75,code=sm_75;-gencode;arch=compute_80,code=sm_80;-gencode;arch=compute_86,code=sm_86;-gencode;arch=compute_37,code=compute_37 + - CuDNN 8.3.2 (built against CUDA 11.5) + - Magma 2.6.1 + - Build settings: BLAS_INFO=mkl, BUILD_TYPE=Release, CUDA_VERSION=11.6, CUDNN_VERSION=8.3.2, CXX_COMPILER=/opt/rh/devtoolset-9/root/usr/bin/c++, CXX_FLAGS= -fabi-version=11 -Wno-deprecated -fvisibility-inlines-hidden -DUSE_PTHREADPOOL -fopenmp -DNDEBUG -DUSE_KINETO -DUSE_FBGEMM -DUSE_QNNPACK -DUSE_PYTORCH_QNNPACK -DUSE_XNNPACK -DSYMBOLICATE_MOBILE_DEBUG_HANDLE -DEDGE_PROFILER_USE_KINETO -O2 -fPIC -Wno-narrowing -Wall -Wextra -Werror=return-type -Werror=non-virtual-dtor -Wno-missing-field-initializers -Wno-type-limits -Wno-array-bounds -Wno-unknown-pragmas -Wunused-local-typedefs -Wno-unused-parameter -Wno-unused-function -Wno-unused-result -Wno-strict-overflow -Wno-strict-aliasing -Wno-error=deprecated-declarations -Wno-stringop-overflow -Wno-psabi -Wno-error=pedantic -Wno-error=redundant-decls -Wno-error=old-style-cast -fdiagnostics-color=always -faligned-new -Wno-unused-but-set-variable -Wno-maybe-uninitialized -fno-math-errno -fno-trapping-math -Werror=format -Werror=cast-function-type -Wno-stringop-overflow, LAPACK_INFO=mkl, PERF_WITH_AVX=1, PERF_WITH_AVX2=1, PERF_WITH_AVX512=1, TORCH_VERSION=1.13.1, USE_CUDA=ON, USE_CUDNN=ON, USE_EXCEPTION_PTR=1, USE_GFLAGS=OFF, USE_GLOG=OFF, USE_MKL=ON, USE_MKLDNN=ON, USE_MPI=OFF, USE_NCCL=ON, USE_NNPACK=ON, USE_OPENMP=ON, USE_ROCM=OFF, + + TorchVision: 0.14.1 + OpenCV: 4.7.0 + MMEngine: 0.7.3 + +Runtime environment: + cudnn_benchmark: True + mp_cfg: {'mp_start_method': 'fork', 'opencv_num_threads': 0} + dist_cfg: {'backend': 'nccl'} + seed: None + deterministic: False + Distributed launcher: slurm + Distributed training: True + GPU number: 4 +------------------------------------------------------------ + +2023/06/01 19:32:13 - mmengine - INFO - Config: +optim_wrapper = dict( + optimizer=dict( + type='AdamW', + lr=0.0004, + weight_decay=0.05, + eps=1e-08, + betas=(0.9, 0.999), + _scope_='mmpretrain'), + paramwise_cfg=dict( + norm_decay_mult=0.0, + bias_decay_mult=0.0, + flat_decay_mult=0.0, + custom_keys=dict({ + '.absolute_pos_embed': dict(decay_mult=0.0), + '.relative_position_bias_table': dict(decay_mult=0.0) + })), + type='AmpOptimWrapper', + dtype='bfloat16', + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=True, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=4096) +model = dict( + type='ImageClassifier', + backbone=dict(type='ConvNeXt', arch='small', drop_path_rate=0.4), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=768, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + init_cfg=None), + init_cfg=dict( + type='TruncNormal', layer=['Conv2d', 'Linear'], std=0.02, bias=0.0), + train_cfg=None) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stablediffusionV1-5R2-dpmsolver-25-5m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/if-dpmsolver++-50-20w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stablediffusionV2-1-dpmsolver-25-190w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/convnext_small_4xb256_fake7m-lr4e-4' + +2023/06/01 19:32:20 - mmengine - INFO - Hooks will be executed in the following order: +before_run: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_load_checkpoint: +(ABOVE_NORMAL) EMAHook + -------------------- +before_train: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_train_epoch: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(NORMAL ) DistSamplerSeedHook + -------------------- +before_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook + -------------------- +after_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +after_train_epoch: +(NORMAL ) IterTimerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_val_epoch: +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook + -------------------- +before_val_iter: +(NORMAL ) IterTimerHook + -------------------- +after_val_iter: +(NORMAL ) IterTimerHook +(NORMAL ) VisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_val_epoch: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_save_checkpoint: +(ABOVE_NORMAL) EMAHook + -------------------- +after_train: +(VERY_LOW ) CheckpointHook + -------------------- +before_test_epoch: +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook + -------------------- +before_test_iter: +(NORMAL ) IterTimerHook + -------------------- +after_test_iter: +(NORMAL ) IterTimerHook +(NORMAL ) VisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_test_epoch: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_run: +(BELOW_NORMAL) LoggerHook + -------------------- +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.0.0.bias:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.0.1.weight:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.0.1.bias:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.1.0.weight:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.1.0.bias:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.1.1.bias:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.2.0.weight:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.2.0.bias:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.2.1.bias:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.3.0.weight:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.3.0.bias:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.3.1.bias:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.0.0.gamma:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.0.0.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.0.0.norm.weight:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.0.0.norm.bias:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.0.0.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.0.0.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.0.1.gamma:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.0.1.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.0.1.norm.weight:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.0.1.norm.bias:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.0.1.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.0.1.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.0.2.gamma:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.0.2.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.0.2.norm.weight:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.0.2.norm.bias:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.0.2.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.0.2.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.1.0.gamma:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.1.0.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.1.0.norm.weight:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.1.0.norm.bias:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.1.0.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.1.0.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.1.1.gamma:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.1.1.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.1.1.norm.weight:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.1.1.norm.bias:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.1.1.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.1.1.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.1.2.gamma:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.1.2.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.1.2.norm.weight:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.1.2.norm.bias:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.1.2.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.1.2.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.2.0.gamma:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.2.0.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.2.0.norm.weight:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.2.0.norm.bias:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.2.0.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.2.0.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.2.1.gamma:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.2.1.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.2.1.norm.weight:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.2.1.norm.bias:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.2.1.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.2.1.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.2.2.gamma:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.2.2.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.2.2.norm.weight:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.2.2.norm.bias:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.2.2.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.2.2.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.2.3.gamma:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.2.3.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.2.3.norm.weight:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.2.3.norm.bias:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.2.3.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.2.3.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.2.4.gamma:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.2.4.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.2.4.norm.weight:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.2.4.norm.bias:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.2.4.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.2.4.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.2.5.gamma:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.2.5.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.2.5.norm.weight:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.2.5.norm.bias:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.2.5.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.2.5.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.2.6.gamma:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.2.6.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.2.6.norm.weight:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.2.6.norm.bias:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.2.6.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.2.6.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.2.7.gamma:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.2.7.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.2.7.norm.weight:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.2.7.norm.bias:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.2.7.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.2.7.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.2.8.gamma:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.2.8.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.2.8.norm.weight:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.2.8.norm.bias:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.2.8.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.2.8.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.2.9.gamma:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.2.9.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.2.9.norm.weight:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.2.9.norm.bias:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.2.9.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.2.9.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.2.10.gamma:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.2.10.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.2.10.norm.weight:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.2.10.norm.bias:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.2.10.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.2.10.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.2.11.gamma:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.2.11.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.2.11.norm.weight:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.2.11.norm.bias:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.2.11.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.2.11.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.2.12.gamma:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.2.12.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.2.12.norm.weight:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.2.12.norm.bias:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.2.12.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.2.12.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.2.13.gamma:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.2.13.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.2.13.norm.weight:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.2.13.norm.bias:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.2.13.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.2.13.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.2.14.gamma:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.2.14.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.2.14.norm.weight:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.2.14.norm.bias:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.2.14.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.2.14.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.2.15.gamma:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.2.15.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.2.15.norm.weight:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.2.15.norm.bias:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.2.15.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.2.15.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.2.16.gamma:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.2.16.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.2.16.norm.weight:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.2.16.norm.bias:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.2.16.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.2.16.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.2.17.gamma:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.2.17.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.2.17.norm.weight:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.2.17.norm.bias:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.2.17.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.2.17.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.2.18.gamma:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.2.18.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.2.18.norm.weight:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.2.18.norm.bias:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.2.18.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.2.18.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.2.19.gamma:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.2.19.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.2.19.norm.weight:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.2.19.norm.bias:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.2.19.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.2.19.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.2.20.gamma:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.2.20.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.2.20.norm.weight:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.2.20.norm.bias:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.2.20.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.2.20.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.2.21.gamma:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.2.21.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.2.21.norm.weight:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.2.21.norm.bias:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.2.21.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.2.21.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.2.22.gamma:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.2.22.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.2.22.norm.weight:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.2.22.norm.bias:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.2.22.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.2.22.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.2.23.gamma:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.2.23.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.2.23.norm.weight:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.2.23.norm.bias:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.2.23.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.2.23.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.2.24.gamma:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.2.24.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.2.24.norm.weight:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.2.24.norm.bias:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.2.24.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.2.24.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.2.25.gamma:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.2.25.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.2.25.norm.weight:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.2.25.norm.bias:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.2.25.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.2.25.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.2.26.gamma:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.2.26.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.2.26.norm.weight:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.2.26.norm.bias:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.2.26.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.2.26.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.3.0.gamma:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.3.0.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.3.0.norm.weight:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.3.0.norm.bias:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.3.0.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.3.0.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.3.1.gamma:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.3.1.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.3.1.norm.weight:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.3.1.norm.bias:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.3.1.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.3.1.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.3.2.gamma:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.3.2.depthwise_conv.bias:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.3.2.norm.weight:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.3.2.norm.bias:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.3.2.pointwise_conv1.bias:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.stages.3.2.pointwise_conv2.bias:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.norm3.weight:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- backbone.norm3.bias:weight_decay=0.0 +2023/06/01 19:32:57 - mmengine - INFO - paramwise_options -- head.fc.bias:weight_decay=0.0 +Name of parameter - Initialization information + +backbone.downsample_layers.0.0.weight - torch.Size([96, 3, 4, 4]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.downsample_layers.0.0.bias - torch.Size([96]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.downsample_layers.0.1.weight - torch.Size([96]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.downsample_layers.0.1.bias - torch.Size([96]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.downsample_layers.1.0.weight - torch.Size([96]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.downsample_layers.1.0.bias - torch.Size([96]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.downsample_layers.1.1.weight - torch.Size([192, 96, 2, 2]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.downsample_layers.1.1.bias - torch.Size([192]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.downsample_layers.2.0.weight - torch.Size([192]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.downsample_layers.2.0.bias - torch.Size([192]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.downsample_layers.2.1.weight - torch.Size([384, 192, 2, 2]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.downsample_layers.2.1.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.downsample_layers.3.0.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.downsample_layers.3.0.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.downsample_layers.3.1.weight - torch.Size([768, 384, 2, 2]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.downsample_layers.3.1.bias - torch.Size([768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.0.gamma - torch.Size([96]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.0.0.depthwise_conv.weight - torch.Size([96, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.0.depthwise_conv.bias - torch.Size([96]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.0.norm.weight - torch.Size([96]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.0.0.norm.bias - torch.Size([96]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.0.0.pointwise_conv1.weight - torch.Size([384, 96]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.0.pointwise_conv1.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.0.pointwise_conv2.weight - torch.Size([96, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.0.pointwise_conv2.bias - torch.Size([96]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.1.gamma - torch.Size([96]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.0.1.depthwise_conv.weight - torch.Size([96, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.1.depthwise_conv.bias - torch.Size([96]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.1.norm.weight - torch.Size([96]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.0.1.norm.bias - torch.Size([96]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.0.1.pointwise_conv1.weight - torch.Size([384, 96]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.1.pointwise_conv1.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.1.pointwise_conv2.weight - torch.Size([96, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.1.pointwise_conv2.bias - torch.Size([96]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.2.gamma - torch.Size([96]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.0.2.depthwise_conv.weight - torch.Size([96, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.2.depthwise_conv.bias - torch.Size([96]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.2.norm.weight - torch.Size([96]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.0.2.norm.bias - torch.Size([96]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.0.2.pointwise_conv1.weight - torch.Size([384, 96]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.2.pointwise_conv1.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.2.pointwise_conv2.weight - torch.Size([96, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.2.pointwise_conv2.bias - torch.Size([96]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.0.gamma - torch.Size([192]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.1.0.depthwise_conv.weight - torch.Size([192, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.0.depthwise_conv.bias - torch.Size([192]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.0.norm.weight - torch.Size([192]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.1.0.norm.bias - torch.Size([192]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.1.0.pointwise_conv1.weight - torch.Size([768, 192]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.0.pointwise_conv1.bias - torch.Size([768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.0.pointwise_conv2.weight - torch.Size([192, 768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.0.pointwise_conv2.bias - torch.Size([192]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.1.gamma - torch.Size([192]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.1.1.depthwise_conv.weight - torch.Size([192, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.1.depthwise_conv.bias - torch.Size([192]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.1.norm.weight - torch.Size([192]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.1.1.norm.bias - torch.Size([192]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.1.1.pointwise_conv1.weight - torch.Size([768, 192]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.1.pointwise_conv1.bias - torch.Size([768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.1.pointwise_conv2.weight - torch.Size([192, 768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.1.pointwise_conv2.bias - torch.Size([192]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.2.gamma - torch.Size([192]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.1.2.depthwise_conv.weight - torch.Size([192, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.2.depthwise_conv.bias - torch.Size([192]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.2.norm.weight - torch.Size([192]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.1.2.norm.bias - torch.Size([192]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.1.2.pointwise_conv1.weight - torch.Size([768, 192]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.2.pointwise_conv1.bias - torch.Size([768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.2.pointwise_conv2.weight - torch.Size([192, 768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.2.pointwise_conv2.bias - torch.Size([192]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.0.gamma - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.0.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.0.depthwise_conv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.0.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.0.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.0.pointwise_conv1.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.0.pointwise_conv1.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.0.pointwise_conv2.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.0.pointwise_conv2.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.1.gamma - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.1.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.1.depthwise_conv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.1.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.1.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.1.pointwise_conv1.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.1.pointwise_conv1.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.1.pointwise_conv2.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.1.pointwise_conv2.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.2.gamma - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.2.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.2.depthwise_conv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.2.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.2.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.2.pointwise_conv1.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.2.pointwise_conv1.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.2.pointwise_conv2.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.2.pointwise_conv2.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.3.gamma - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.3.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.3.depthwise_conv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.3.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.3.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.3.pointwise_conv1.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.3.pointwise_conv1.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.3.pointwise_conv2.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.3.pointwise_conv2.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.4.gamma - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.4.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.4.depthwise_conv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.4.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.4.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.4.pointwise_conv1.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.4.pointwise_conv1.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.4.pointwise_conv2.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.4.pointwise_conv2.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.5.gamma - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.5.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.5.depthwise_conv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.5.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.5.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.5.pointwise_conv1.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.5.pointwise_conv1.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.5.pointwise_conv2.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.5.pointwise_conv2.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.6.gamma - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.6.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.6.depthwise_conv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.6.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.6.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.6.pointwise_conv1.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.6.pointwise_conv1.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.6.pointwise_conv2.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.6.pointwise_conv2.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.7.gamma - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.7.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.7.depthwise_conv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.7.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.7.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.7.pointwise_conv1.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.7.pointwise_conv1.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.7.pointwise_conv2.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.7.pointwise_conv2.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.8.gamma - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.8.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.8.depthwise_conv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.8.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.8.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.8.pointwise_conv1.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.8.pointwise_conv1.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.8.pointwise_conv2.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.8.pointwise_conv2.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.9.gamma - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.9.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.9.depthwise_conv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.9.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.9.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.9.pointwise_conv1.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.9.pointwise_conv1.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.9.pointwise_conv2.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.9.pointwise_conv2.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.10.gamma - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.10.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.10.depthwise_conv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.10.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.10.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.10.pointwise_conv1.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.10.pointwise_conv1.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.10.pointwise_conv2.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.10.pointwise_conv2.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.11.gamma - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.11.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.11.depthwise_conv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.11.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.11.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.11.pointwise_conv1.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.11.pointwise_conv1.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.11.pointwise_conv2.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.11.pointwise_conv2.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.12.gamma - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.12.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.12.depthwise_conv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.12.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.12.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.12.pointwise_conv1.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.12.pointwise_conv1.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.12.pointwise_conv2.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.12.pointwise_conv2.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.13.gamma - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.13.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.13.depthwise_conv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.13.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.13.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.13.pointwise_conv1.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.13.pointwise_conv1.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.13.pointwise_conv2.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.13.pointwise_conv2.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.14.gamma - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.14.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.14.depthwise_conv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.14.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.14.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.14.pointwise_conv1.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.14.pointwise_conv1.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.14.pointwise_conv2.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.14.pointwise_conv2.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.15.gamma - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.15.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.15.depthwise_conv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.15.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.15.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.15.pointwise_conv1.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.15.pointwise_conv1.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.15.pointwise_conv2.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.15.pointwise_conv2.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.16.gamma - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.16.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.16.depthwise_conv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.16.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.16.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.16.pointwise_conv1.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.16.pointwise_conv1.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.16.pointwise_conv2.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.16.pointwise_conv2.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.17.gamma - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.17.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.17.depthwise_conv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.17.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.17.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.17.pointwise_conv1.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.17.pointwise_conv1.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.17.pointwise_conv2.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.17.pointwise_conv2.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.18.gamma - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.18.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.18.depthwise_conv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.18.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.18.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.18.pointwise_conv1.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.18.pointwise_conv1.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.18.pointwise_conv2.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.18.pointwise_conv2.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.19.gamma - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.19.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.19.depthwise_conv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.19.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.19.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.19.pointwise_conv1.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.19.pointwise_conv1.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.19.pointwise_conv2.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.19.pointwise_conv2.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.20.gamma - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.20.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.20.depthwise_conv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.20.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.20.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.20.pointwise_conv1.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.20.pointwise_conv1.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.20.pointwise_conv2.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.20.pointwise_conv2.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.21.gamma - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.21.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.21.depthwise_conv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.21.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.21.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.21.pointwise_conv1.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.21.pointwise_conv1.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.21.pointwise_conv2.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.21.pointwise_conv2.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.22.gamma - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.22.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.22.depthwise_conv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.22.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.22.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.22.pointwise_conv1.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.22.pointwise_conv1.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.22.pointwise_conv2.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.22.pointwise_conv2.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.23.gamma - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.23.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.23.depthwise_conv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.23.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.23.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.23.pointwise_conv1.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.23.pointwise_conv1.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.23.pointwise_conv2.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.23.pointwise_conv2.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.24.gamma - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.24.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.24.depthwise_conv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.24.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.24.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.24.pointwise_conv1.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.24.pointwise_conv1.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.24.pointwise_conv2.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.24.pointwise_conv2.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.25.gamma - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.25.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.25.depthwise_conv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.25.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.25.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.25.pointwise_conv1.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.25.pointwise_conv1.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.25.pointwise_conv2.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.25.pointwise_conv2.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.26.gamma - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.26.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.26.depthwise_conv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.26.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.26.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.26.pointwise_conv1.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.26.pointwise_conv1.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.26.pointwise_conv2.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.26.pointwise_conv2.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.0.gamma - torch.Size([768]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.3.0.depthwise_conv.weight - torch.Size([768, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.0.depthwise_conv.bias - torch.Size([768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.0.norm.weight - torch.Size([768]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.3.0.norm.bias - torch.Size([768]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.3.0.pointwise_conv1.weight - torch.Size([3072, 768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.0.pointwise_conv1.bias - torch.Size([3072]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.0.pointwise_conv2.weight - torch.Size([768, 3072]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.0.pointwise_conv2.bias - torch.Size([768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.1.gamma - torch.Size([768]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.3.1.depthwise_conv.weight - torch.Size([768, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.1.depthwise_conv.bias - torch.Size([768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.1.norm.weight - torch.Size([768]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.3.1.norm.bias - torch.Size([768]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.3.1.pointwise_conv1.weight - torch.Size([3072, 768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.1.pointwise_conv1.bias - torch.Size([3072]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.1.pointwise_conv2.weight - torch.Size([768, 3072]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.1.pointwise_conv2.bias - torch.Size([768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.2.gamma - torch.Size([768]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.3.2.depthwise_conv.weight - torch.Size([768, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.2.depthwise_conv.bias - torch.Size([768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.2.norm.weight - torch.Size([768]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.3.2.norm.bias - torch.Size([768]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.3.2.pointwise_conv1.weight - torch.Size([3072, 768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.2.pointwise_conv1.bias - torch.Size([3072]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.2.pointwise_conv2.weight - torch.Size([768, 3072]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.2.pointwise_conv2.bias - torch.Size([768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.norm3.weight - torch.Size([768]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.norm3.bias - torch.Size([768]): +The value is the same before and after calling `init_weights` of ImageClassifier + +head.fc.weight - torch.Size([2, 768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +head.fc.bias - torch.Size([2]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 +2023/06/01 19:32:59 - mmengine - WARNING - "FileClient" will be deprecated in future. Please use io functions in https://mmengine.readthedocs.io/en/latest/api/fileio.html#file-io +2023/06/01 19:32:59 - mmengine - WARNING - "HardDiskBackend" is the alias of "LocalBackend" and the former will be deprecated in future. +2023/06/01 19:32:59 - mmengine - INFO - Checkpoints will be saved to /mnt/petrelfs/luzeyu/workspace/fakebench/mmpretrain/workdir/convnext_small_4xb256_fake7m-lr4e-4. +2023/06/01 19:39:44 - mmengine - INFO - Epoch(train) [1][ 100/7800] lr: 4.0000e-04 eta: 3 days, 15:39:13 time: 4.0286 data_time: 2.6907 memory: 25071 loss: 0.5509 +2023/06/01 19:46:42 - mmengine - INFO - Epoch(train) [1][ 200/7800] lr: 4.0000e-04 eta: 3 days, 16:59:07 time: 4.6966 data_time: 3.9033 memory: 25071 loss: 0.5123 +2023/06/01 19:54:37 - mmengine - INFO - Epoch(train) [1][ 300/7800] lr: 4.0000e-04 eta: 3 days, 21:21:59 time: 4.2892 data_time: 3.5818 memory: 25071 loss: 0.4802 +2023/06/01 20:02:27 - mmengine - INFO - Epoch(train) [1][ 400/7800] lr: 4.0000e-04 eta: 3 days, 23:16:16 time: 4.4667 data_time: 3.3730 memory: 25071 loss: 0.4763 +2023/06/01 20:10:01 - mmengine - INFO - Epoch(train) [1][ 500/7800] lr: 4.0000e-04 eta: 3 days, 23:39:59 time: 4.4804 data_time: 3.1529 memory: 25071 loss: 0.4614 +2023/06/01 20:17:23 - mmengine - INFO - Epoch(train) [1][ 600/7800] lr: 4.0000e-04 eta: 3 days, 23:28:03 time: 4.4408 data_time: 3.7333 memory: 25071 loss: 0.4546 +2023/06/01 20:24:37 - mmengine - INFO - Epoch(train) [1][ 700/7800] lr: 4.0000e-04 eta: 3 days, 23:02:36 time: 4.4552 data_time: 3.7477 memory: 25071 loss: 0.4439 +2023/06/01 20:33:23 - mmengine - INFO - Epoch(train) [1][ 800/7800] lr: 4.0000e-04 eta: 4 days, 1:09:24 time: 3.7072 data_time: 2.9975 memory: 25071 loss: 0.4616 +2023/06/01 20:40:08 - mmengine - INFO - Epoch(train) [1][ 900/7800] lr: 4.0000e-04 eta: 3 days, 23:52:27 time: 4.0192 data_time: 3.3110 memory: 25071 loss: 0.4083 +2023/06/01 20:46:39 - mmengine - INFO - Exp name: convnext_small_4xb256_fake7m-lr4e-4_20230601_193207 +2023/06/01 20:46:39 - mmengine - INFO - Epoch(train) [1][1000/7800] lr: 4.0000e-04 eta: 3 days, 22:32:21 time: 0.7083 data_time: 0.0013 memory: 25071 loss: 0.4357 +2023/06/01 20:47:51 - mmengine - INFO - Epoch(train) [1][1100/7800] lr: 4.0000e-04 eta: 3 days, 15:14:16 time: 0.7554 data_time: 0.0460 memory: 25071 loss: 0.3937 +2023/06/01 20:49:04 - mmengine - INFO - Epoch(train) [1][1200/7800] lr: 4.0000e-04 eta: 3 days, 9:09:26 time: 0.7048 data_time: 0.0014 memory: 25071 loss: 0.4059 +2023/06/01 20:50:17 - mmengine - INFO - Epoch(train) [1][1300/7800] lr: 4.0000e-04 eta: 3 days, 4:00:30 time: 0.7289 data_time: 0.0013 memory: 25071 loss: 0.4099 +2023/06/01 20:51:34 - mmengine - INFO - Epoch(train) [1][1400/7800] lr: 4.0000e-04 eta: 2 days, 23:39:33 time: 0.8211 data_time: 0.0014 memory: 25071 loss: 0.4109 +2023/06/01 20:52:52 - mmengine - INFO - Epoch(train) [1][1500/7800] lr: 4.0000e-04 eta: 2 days, 19:54:09 time: 0.7272 data_time: 0.0014 memory: 25071 loss: 0.3993 +2023/06/01 20:54:23 - mmengine - INFO - Epoch(train) [1][1600/7800] lr: 4.0000e-04 eta: 2 days, 16:46:51 time: 0.8960 data_time: 0.1898 memory: 25071 loss: 0.3987 +2023/06/01 20:55:54 - mmengine - INFO - Epoch(train) [1][1700/7800] lr: 4.0000e-04 eta: 2 days, 14:01:44 time: 0.8547 data_time: 0.0013 memory: 25071 loss: 0.3719 +2023/06/01 20:57:27 - mmengine - INFO - Epoch(train) [1][1800/7800] lr: 4.0000e-04 eta: 2 days, 11:36:09 time: 0.8932 data_time: 0.0015 memory: 25071 loss: 0.3637 +2023/06/01 20:58:54 - mmengine - INFO - Epoch(train) [1][1900/7800] lr: 4.0000e-04 eta: 2 days, 9:21:08 time: 0.7350 data_time: 0.0014 memory: 25071 loss: 0.3782 +2023/06/01 21:00:06 - mmengine - INFO - Exp name: convnext_small_4xb256_fake7m-lr4e-4_20230601_193207 +2023/06/01 21:00:06 - mmengine - INFO - Epoch(train) [1][2000/7800] lr: 4.0000e-04 eta: 2 days, 7:10:38 time: 0.7688 data_time: 0.0014 memory: 25071 loss: 0.3699 +2023/06/01 21:01:23 - mmengine - INFO - Epoch(train) [1][2100/7800] lr: 4.0000e-04 eta: 2 days, 5:15:06 time: 0.7926 data_time: 0.0013 memory: 25071 loss: 0.3760 +2023/06/01 21:02:41 - mmengine - INFO - Epoch(train) [1][2200/7800] lr: 4.0000e-04 eta: 2 days, 3:30:31 time: 0.7186 data_time: 0.0014 memory: 25071 loss: 0.3778 +2023/06/01 21:04:01 - mmengine - INFO - Epoch(train) [1][2300/7800] lr: 4.0000e-04 eta: 2 days, 1:56:19 time: 0.7635 data_time: 0.0015 memory: 25071 loss: 0.3432 +2023/06/01 21:05:19 - mmengine - INFO - Epoch(train) [1][2400/7800] lr: 4.0000e-04 eta: 2 days, 0:28:40 time: 0.7809 data_time: 0.0012 memory: 25071 loss: 0.3328 +2023/06/01 21:06:53 - mmengine - INFO - Epoch(train) [1][2500/7800] lr: 4.0000e-04 eta: 1 day, 23:15:37 time: 0.8116 data_time: 0.0013 memory: 25071 loss: 0.3289 +2023/06/01 21:08:26 - mmengine - INFO - Epoch(train) [1][2600/7800] lr: 4.0000e-04 eta: 1 day, 22:07:53 time: 1.0313 data_time: 0.0013 memory: 25071 loss: 0.3313 +2023/06/01 21:09:58 - mmengine - INFO - Epoch(train) [1][2700/7800] lr: 4.0000e-04 eta: 1 day, 21:04:45 time: 1.0104 data_time: 0.0016 memory: 25071 loss: 0.3416 +2023/06/01 21:11:33 - mmengine - INFO - Epoch(train) [1][2800/7800] lr: 4.0000e-04 eta: 1 day, 20:07:27 time: 0.9495 data_time: 0.0014 memory: 25071 loss: 0.3161 +2023/06/01 21:13:30 - mmengine - INFO - Epoch(train) [1][2900/7800] lr: 4.0000e-04 eta: 1 day, 19:23:01 time: 0.7670 data_time: 0.0013 memory: 25071 loss: 0.3132 +2023/06/01 21:15:04 - mmengine - INFO - Exp name: convnext_small_4xb256_fake7m-lr4e-4_20230601_193207 +2023/06/01 21:15:04 - mmengine - INFO - Epoch(train) [1][3000/7800] lr: 4.0000e-04 eta: 1 day, 18:31:53 time: 1.0443 data_time: 0.0013 memory: 25071 loss: 0.3302 +2023/06/01 21:16:36 - mmengine - INFO - Epoch(train) [1][3100/7800] lr: 4.0000e-04 eta: 1 day, 17:43:26 time: 0.8310 data_time: 0.0272 memory: 25071 loss: 0.3198 +2023/06/01 21:17:49 - mmengine - INFO - Epoch(train) [1][3200/7800] lr: 4.0000e-04 eta: 1 day, 16:50:27 time: 0.7057 data_time: 0.0015 memory: 25071 loss: 0.2964 +2023/06/01 21:19:00 - mmengine - INFO - Epoch(train) [1][3300/7800] lr: 4.0000e-04 eta: 1 day, 16:00:00 time: 0.7190 data_time: 0.0013 memory: 25071 loss: 0.3205 +2023/06/01 21:20:11 - mmengine - INFO - Epoch(train) [1][3400/7800] lr: 4.0000e-04 eta: 1 day, 15:12:16 time: 0.7054 data_time: 0.0014 memory: 25071 loss: 0.3121 +2023/06/01 21:21:23 - mmengine - INFO - Epoch(train) [1][3500/7800] lr: 4.0000e-04 eta: 1 day, 14:27:13 time: 0.7053 data_time: 0.0017 memory: 25071 loss: 0.3116 +2023/06/01 21:22:34 - mmengine - INFO - Epoch(train) [1][3600/7800] lr: 4.0000e-04 eta: 1 day, 13:44:34 time: 0.7053 data_time: 0.0013 memory: 25071 loss: 0.2981 +2023/06/01 21:23:45 - mmengine - INFO - Epoch(train) [1][3700/7800] lr: 4.0000e-04 eta: 1 day, 13:04:18 time: 0.7148 data_time: 0.0014 memory: 25071 loss: 0.3141 +2023/06/01 21:24:56 - mmengine - INFO - Epoch(train) [1][3800/7800] lr: 4.0000e-04 eta: 1 day, 12:25:57 time: 0.7061 data_time: 0.0015 memory: 25071 loss: 0.2956 +2023/06/01 21:26:07 - mmengine - INFO - Epoch(train) [1][3900/7800] lr: 4.0000e-04 eta: 1 day, 11:49:35 time: 0.7245 data_time: 0.0014 memory: 25071 loss: 0.2936 +2023/06/01 21:27:18 - mmengine - INFO - Exp name: convnext_small_4xb256_fake7m-lr4e-4_20230601_193207 +2023/06/01 21:27:18 - mmengine - INFO - Epoch(train) [1][4000/7800] lr: 4.0000e-04 eta: 1 day, 11:14:57 time: 0.7071 data_time: 0.0015 memory: 25071 loss: 0.2547 +2023/06/01 21:28:29 - mmengine - INFO - Epoch(train) [1][4100/7800] lr: 4.0000e-04 eta: 1 day, 10:41:57 time: 0.7236 data_time: 0.0014 memory: 25071 loss: 0.2920 +2023/06/01 21:29:41 - mmengine - INFO - Epoch(train) [1][4200/7800] lr: 4.0000e-04 eta: 1 day, 10:10:28 time: 0.7080 data_time: 0.0013 memory: 25071 loss: 0.2884 +2023/06/01 21:30:52 - mmengine - INFO - Epoch(train) [1][4300/7800] lr: 4.0000e-04 eta: 1 day, 9:40:28 time: 0.7135 data_time: 0.0014 memory: 25071 loss: 0.2842 +2023/06/01 21:32:03 - mmengine - INFO - Epoch(train) [1][4400/7800] lr: 4.0000e-04 eta: 1 day, 9:11:39 time: 0.7044 data_time: 0.0015 memory: 25071 loss: 0.2772 +2023/06/01 21:33:14 - mmengine - INFO - Epoch(train) [1][4500/7800] lr: 4.0000e-04 eta: 1 day, 8:44:09 time: 0.7201 data_time: 0.0014 memory: 25071 loss: 0.2899 +2023/06/01 21:34:25 - mmengine - INFO - Epoch(train) [1][4600/7800] lr: 4.0000e-04 eta: 1 day, 8:17:43 time: 0.7056 data_time: 0.0013 memory: 25071 loss: 0.2694 +2023/06/01 21:35:38 - mmengine - INFO - Epoch(train) [1][4700/7800] lr: 4.0000e-04 eta: 1 day, 7:52:43 time: 0.7550 data_time: 0.0014 memory: 25071 loss: 0.2595 +2023/06/01 21:36:51 - mmengine - INFO - Epoch(train) [1][4800/7800] lr: 4.0000e-04 eta: 1 day, 7:29:01 time: 0.7813 data_time: 0.0012 memory: 25071 loss: 0.2693 +2023/06/01 21:38:04 - mmengine - INFO - Epoch(train) [1][4900/7800] lr: 4.0000e-04 eta: 1 day, 7:06:04 time: 0.7182 data_time: 0.0014 memory: 25071 loss: 0.2865 +2023/06/01 21:39:16 - mmengine - INFO - Exp name: convnext_small_4xb256_fake7m-lr4e-4_20230601_193207 +2023/06/01 21:39:16 - mmengine - INFO - Epoch(train) [1][5000/7800] lr: 4.0000e-04 eta: 1 day, 6:43:36 time: 0.7069 data_time: 0.0014 memory: 25071 loss: 0.2608 +2023/06/01 21:40:28 - mmengine - INFO - Epoch(train) [1][5100/7800] lr: 4.0000e-04 eta: 1 day, 6:22:18 time: 0.7125 data_time: 0.0012 memory: 25071 loss: 0.2788 +2023/06/01 21:41:41 - mmengine - INFO - Epoch(train) [1][5200/7800] lr: 4.0000e-04 eta: 1 day, 6:01:47 time: 0.7055 data_time: 0.0014 memory: 25071 loss: 0.2631 +2023/06/01 21:42:55 - mmengine - INFO - Epoch(train) [1][5300/7800] lr: 4.0000e-04 eta: 1 day, 5:42:18 time: 0.7458 data_time: 0.0013 memory: 25071 loss: 0.2580 +2023/06/01 21:44:08 - mmengine - INFO - Epoch(train) [1][5400/7800] lr: 4.0000e-04 eta: 1 day, 5:23:10 time: 0.7410 data_time: 0.0014 memory: 25071 loss: 0.2387 +2023/06/01 21:45:24 - mmengine - INFO - Epoch(train) [1][5500/7800] lr: 4.0000e-04 eta: 1 day, 5:05:22 time: 0.7156 data_time: 0.0013 memory: 25071 loss: 0.2532 +2023/06/01 21:46:36 - mmengine - INFO - Epoch(train) [1][5600/7800] lr: 4.0000e-04 eta: 1 day, 4:47:29 time: 0.7894 data_time: 0.0015 memory: 25071 loss: 0.2202 +2023/06/01 21:47:50 - mmengine - INFO - Epoch(train) [1][5700/7800] lr: 4.0000e-04 eta: 1 day, 4:30:28 time: 0.7222 data_time: 0.0014 memory: 25071 loss: 0.2580 +2023/06/01 21:49:02 - mmengine - INFO - Epoch(train) [1][5800/7800] lr: 4.0000e-04 eta: 1 day, 4:13:32 time: 0.7052 data_time: 0.0015 memory: 25071 loss: 0.2619 +2023/06/01 21:50:16 - mmengine - INFO - Epoch(train) [1][5900/7800] lr: 4.0000e-04 eta: 1 day, 3:57:42 time: 0.7065 data_time: 0.0015 memory: 25071 loss: 0.2398 +2023/06/01 21:51:28 - mmengine - INFO - Exp name: convnext_small_4xb256_fake7m-lr4e-4_20230601_193207 +2023/06/01 21:51:28 - mmengine - INFO - Epoch(train) [1][6000/7800] lr: 4.0000e-04 eta: 1 day, 3:41:50 time: 0.7039 data_time: 0.0018 memory: 25071 loss: 0.2715 +2023/06/01 21:52:41 - mmengine - INFO - Epoch(train) [1][6100/7800] lr: 4.0000e-04 eta: 1 day, 3:26:34 time: 0.7212 data_time: 0.0015 memory: 25071 loss: 0.2308 +2023/06/01 21:53:53 - mmengine - INFO - Epoch(train) [1][6200/7800] lr: 4.0000e-04 eta: 1 day, 3:11:37 time: 0.7204 data_time: 0.0013 memory: 25071 loss: 0.2449 +2023/06/01 21:55:04 - mmengine - INFO - Epoch(train) [1][6300/7800] lr: 4.0000e-04 eta: 1 day, 2:57:06 time: 0.7057 data_time: 0.0012 memory: 25071 loss: 0.2394 +2023/06/01 21:56:15 - mmengine - INFO - Epoch(train) [1][6400/7800] lr: 4.0000e-04 eta: 1 day, 2:42:51 time: 0.7073 data_time: 0.0014 memory: 25071 loss: 0.2350 +2023/06/01 21:57:27 - mmengine - INFO - Epoch(train) [1][6500/7800] lr: 4.0000e-04 eta: 1 day, 2:29:03 time: 0.7066 data_time: 0.0014 memory: 25071 loss: 0.2142 +2023/06/01 21:58:39 - mmengine - INFO - Epoch(train) [1][6600/7800] lr: 4.0000e-04 eta: 1 day, 2:15:49 time: 0.7030 data_time: 0.0014 memory: 25071 loss: 0.2427 +2023/06/01 21:59:51 - mmengine - INFO - Epoch(train) [1][6700/7800] lr: 4.0000e-04 eta: 1 day, 2:02:49 time: 0.7081 data_time: 0.0015 memory: 25071 loss: 0.2333 +2023/06/01 22:01:02 - mmengine - INFO - Epoch(train) [1][6800/7800] lr: 4.0000e-04 eta: 1 day, 1:50:05 time: 0.7077 data_time: 0.0014 memory: 25071 loss: 0.2395 +2023/06/01 22:02:13 - mmengine - INFO - Epoch(train) [1][6900/7800] lr: 4.0000e-04 eta: 1 day, 1:37:46 time: 0.7037 data_time: 0.0014 memory: 25071 loss: 0.2277 +2023/06/01 22:03:24 - mmengine - INFO - Exp name: convnext_small_4xb256_fake7m-lr4e-4_20230601_193207 +2023/06/01 22:03:24 - mmengine - INFO - Epoch(train) [1][7000/7800] lr: 4.0000e-04 eta: 1 day, 1:25:42 time: 0.7160 data_time: 0.0015 memory: 25071 loss: 0.2067 +2023/06/01 22:04:36 - mmengine - INFO - Epoch(train) [1][7100/7800] lr: 4.0000e-04 eta: 1 day, 1:13:56 time: 0.7053 data_time: 0.0013 memory: 25071 loss: 0.2090 +2023/06/01 22:05:47 - mmengine - INFO - Epoch(train) [1][7200/7800] lr: 4.0000e-04 eta: 1 day, 1:02:27 time: 0.7126 data_time: 0.0013 memory: 25071 loss: 0.2177 +2023/06/01 22:06:58 - mmengine - INFO - Epoch(train) [1][7300/7800] lr: 4.0000e-04 eta: 1 day, 0:51:17 time: 0.7069 data_time: 0.0015 memory: 25071 loss: 0.2149 +2023/06/01 22:08:09 - mmengine - INFO - Epoch(train) [1][7400/7800] lr: 4.0000e-04 eta: 1 day, 0:40:22 time: 0.7383 data_time: 0.0016 memory: 25071 loss: 0.2121 +2023/06/01 22:09:22 - mmengine - INFO - Epoch(train) [1][7500/7800] lr: 4.0000e-04 eta: 1 day, 0:29:54 time: 0.7065 data_time: 0.0014 memory: 25071 loss: 0.2146 +2023/06/01 22:10:33 - mmengine - INFO - Epoch(train) [1][7600/7800] lr: 4.0000e-04 eta: 1 day, 0:19:29 time: 0.7192 data_time: 0.0017 memory: 25071 loss: 0.1986 +2023/06/01 22:11:45 - mmengine - INFO - Epoch(train) [1][7700/7800] lr: 4.0000e-04 eta: 1 day, 0:09:28 time: 0.7051 data_time: 0.0014 memory: 25071 loss: 0.2252 +2023/06/01 22:12:57 - mmengine - INFO - Exp name: convnext_small_4xb256_fake7m-lr4e-4_20230601_193207 +2023/06/01 22:12:57 - mmengine - INFO - Epoch(train) [1][7800/7800] lr: 4.0000e-04 eta: 23:59:42 time: 0.6595 data_time: 0.0011 memory: 25071 loss: 0.2027 +2023/06/01 22:12:57 - mmengine - INFO - Saving checkpoint at 1 epochs +2023/06/01 22:13:19 - mmengine - INFO - Epoch(val) [1][16/16] accuracy/top1: 82.7473 single-label/precision_classwise: [100.0, 0.0] single-label/recall_classwise: [82.74729919433594, 0.0] single-label/f1-score_classwise: [90.55924987792969, 0.0] data_time: 0.2619 time: 0.7811 +2023/06/01 22:14:39 - mmengine - INFO - Epoch(train) [2][ 100/7800] lr: 3.9046e-04 eta: 23:51:12 time: 0.7229 data_time: 0.0014 memory: 25074 loss: 0.1985 +2023/06/01 22:15:51 - mmengine - INFO - Exp name: convnext_small_4xb256_fake7m-lr4e-4_20230601_193207 +2023/06/01 22:15:51 - mmengine - INFO - Epoch(train) [2][ 200/7800] lr: 3.9046e-04 eta: 23:41:50 time: 0.7411 data_time: 0.0018 memory: 25074 loss: 0.1917 +2023/06/01 22:17:04 - mmengine - INFO - Epoch(train) [2][ 300/7800] lr: 3.9046e-04 eta: 23:32:48 time: 0.7229 data_time: 0.0012 memory: 25074 loss: 0.1983 +2023/06/01 22:18:17 - mmengine - INFO - Epoch(train) [2][ 400/7800] lr: 3.9046e-04 eta: 23:23:55 time: 0.7558 data_time: 0.0013 memory: 25074 loss: 0.1937 +2023/06/01 22:19:28 - mmengine - INFO - Epoch(train) [2][ 500/7800] lr: 3.9046e-04 eta: 23:15:00 time: 0.7217 data_time: 0.0014 memory: 25074 loss: 0.1959 +2023/06/01 22:20:42 - mmengine - INFO - Epoch(train) [2][ 600/7800] lr: 3.9046e-04 eta: 23:06:37 time: 0.8182 data_time: 0.0013 memory: 25074 loss: 0.1858 +2023/06/01 22:21:55 - mmengine - INFO - Epoch(train) [2][ 700/7800] lr: 3.9046e-04 eta: 22:58:14 time: 0.7238 data_time: 0.0016 memory: 25074 loss: 0.2049 +2023/06/01 22:23:08 - mmengine - INFO - Epoch(train) [2][ 800/7800] lr: 3.9046e-04 eta: 22:50:02 time: 0.7289 data_time: 0.0167 memory: 25074 loss: 0.1949 +2023/06/01 22:24:22 - mmengine - INFO - Epoch(train) [2][ 900/7800] lr: 3.9046e-04 eta: 22:42:16 time: 0.7226 data_time: 0.0015 memory: 25074 loss: 0.1710 +2023/06/01 22:25:34 - mmengine - INFO - Epoch(train) [2][1000/7800] lr: 3.9046e-04 eta: 22:34:14 time: 0.7215 data_time: 0.0018 memory: 25074 loss: 0.1994 +2023/06/01 22:26:46 - mmengine - INFO - Epoch(train) [2][1100/7800] lr: 3.9046e-04 eta: 22:26:24 time: 0.7400 data_time: 0.0017 memory: 25074 loss: 0.1912 +2023/06/01 22:27:58 - mmengine - INFO - Exp name: convnext_small_4xb256_fake7m-lr4e-4_20230601_193207 +2023/06/01 22:27:58 - mmengine - INFO - Epoch(train) [2][1200/7800] lr: 3.9046e-04 eta: 22:18:40 time: 0.7093 data_time: 0.0019 memory: 25074 loss: 0.1867 +2023/06/01 22:29:09 - mmengine - INFO - Epoch(train) [2][1300/7800] lr: 3.9046e-04 eta: 22:11:02 time: 0.7357 data_time: 0.0020 memory: 25074 loss: 0.1699 +2023/06/01 22:30:22 - mmengine - INFO - Epoch(train) [2][1400/7800] lr: 3.9046e-04 eta: 22:03:40 time: 0.7638 data_time: 0.0525 memory: 25074 loss: 0.1863 +2023/06/01 22:31:36 - mmengine - INFO - Epoch(train) [2][1500/7800] lr: 3.9046e-04 eta: 21:56:43 time: 0.7948 data_time: 0.0463 memory: 25074 loss: 0.1805 +2023/06/01 22:32:49 - mmengine - INFO - Epoch(train) [2][1600/7800] lr: 3.9046e-04 eta: 21:49:41 time: 0.7715 data_time: 0.0505 memory: 25074 loss: 0.1714 +2023/06/01 22:34:03 - mmengine - INFO - Epoch(train) [2][1700/7800] lr: 3.9046e-04 eta: 21:42:54 time: 0.7826 data_time: 0.0510 memory: 25074 loss: 0.2019 +2023/06/01 22:35:18 - mmengine - INFO - Epoch(train) [2][1800/7800] lr: 3.9046e-04 eta: 21:36:19 time: 0.7894 data_time: 0.0732 memory: 25074 loss: 0.1585 +2023/06/01 22:36:33 - mmengine - INFO - Epoch(train) [2][1900/7800] lr: 3.9046e-04 eta: 21:29:54 time: 0.7229 data_time: 0.0016 memory: 25074 loss: 0.1712 +2023/06/01 22:37:46 - mmengine - INFO - Epoch(train) [2][2000/7800] lr: 3.9046e-04 eta: 21:23:21 time: 0.7708 data_time: 0.0506 memory: 25074 loss: 0.1745 +2023/06/01 22:39:00 - mmengine - INFO - Epoch(train) [2][2100/7800] lr: 3.9046e-04 eta: 21:17:02 time: 0.7293 data_time: 0.0055 memory: 25074 loss: 0.1704 +2023/06/01 22:40:12 - mmengine - INFO - Exp name: convnext_small_4xb256_fake7m-lr4e-4_20230601_193207 +2023/06/01 22:40:12 - mmengine - INFO - Epoch(train) [2][2200/7800] lr: 3.9046e-04 eta: 21:10:31 time: 0.7221 data_time: 0.0014 memory: 25074 loss: 0.1926 +2023/06/01 22:41:24 - mmengine - INFO - Epoch(train) [2][2300/7800] lr: 3.9046e-04 eta: 21:04:13 time: 0.7274 data_time: 0.0014 memory: 25074 loss: 0.1505 +2023/06/01 22:42:37 - mmengine - INFO - Epoch(train) [2][2400/7800] lr: 3.9046e-04 eta: 20:57:59 time: 0.7262 data_time: 0.0014 memory: 25074 loss: 0.1697 +2023/06/01 22:43:50 - mmengine - INFO - Epoch(train) [2][2500/7800] lr: 3.9046e-04 eta: 20:51:56 time: 0.7154 data_time: 0.0018 memory: 25074 loss: 0.1554 +2023/06/01 22:45:04 - mmengine - INFO - Epoch(train) [2][2600/7800] lr: 3.9046e-04 eta: 20:46:05 time: 0.7603 data_time: 0.0247 memory: 25074 loss: 0.1719 +2023/06/01 22:46:18 - mmengine - INFO - Epoch(train) [2][2700/7800] lr: 3.9046e-04 eta: 20:40:19 time: 0.7367 data_time: 0.0014 memory: 25074 loss: 0.1694 +2023/06/01 22:47:32 - mmengine - INFO - Epoch(train) [2][2800/7800] lr: 3.9046e-04 eta: 20:34:38 time: 0.7340 data_time: 0.0018 memory: 25074 loss: 0.1552 +2023/06/01 22:48:46 - mmengine - INFO - Epoch(train) [2][2900/7800] lr: 3.9046e-04 eta: 20:29:03 time: 0.7191 data_time: 0.0013 memory: 25074 loss: 0.1589 +2023/06/01 22:50:01 - mmengine - INFO - Epoch(train) [2][3000/7800] lr: 3.9046e-04 eta: 20:23:38 time: 0.7508 data_time: 0.0468 memory: 25074 loss: 0.1378 +2023/06/01 22:51:15 - mmengine - INFO - Epoch(train) [2][3100/7800] lr: 3.9046e-04 eta: 20:18:11 time: 0.7291 data_time: 0.0017 memory: 25074 loss: 0.1549 +2023/06/01 22:52:30 - mmengine - INFO - Exp name: convnext_small_4xb256_fake7m-lr4e-4_20230601_193207 +2023/06/01 22:52:30 - mmengine - INFO - Epoch(train) [2][3200/7800] lr: 3.9046e-04 eta: 20:12:57 time: 0.7854 data_time: 0.0016 memory: 25074 loss: 0.1439 +2023/06/01 22:53:45 - mmengine - INFO - Epoch(train) [2][3300/7800] lr: 3.9046e-04 eta: 20:07:48 time: 0.7321 data_time: 0.0024 memory: 25074 loss: 0.1316 +2023/06/01 22:54:59 - mmengine - INFO - Epoch(train) [2][3400/7800] lr: 3.9046e-04 eta: 20:02:36 time: 0.7056 data_time: 0.0016 memory: 25074 loss: 0.1438 +2023/06/01 22:56:15 - mmengine - INFO - Epoch(train) [2][3500/7800] lr: 3.9046e-04 eta: 19:57:35 time: 0.7379 data_time: 0.0020 memory: 25074 loss: 0.1425 +2023/06/01 22:57:30 - mmengine - INFO - Epoch(train) [2][3600/7800] lr: 3.9046e-04 eta: 19:52:35 time: 0.7063 data_time: 0.0018 memory: 25074 loss: 0.1623 +2023/06/01 22:58:42 - mmengine - INFO - Epoch(train) [2][3700/7800] lr: 3.9046e-04 eta: 19:47:24 time: 0.7041 data_time: 0.0015 memory: 25074 loss: 0.1341 +2023/06/01 22:59:58 - mmengine - INFO - Epoch(train) [2][3800/7800] lr: 3.9046e-04 eta: 19:42:38 time: 0.7398 data_time: 0.0017 memory: 25074 loss: 0.1327 +2023/06/01 23:01:12 - mmengine - INFO - Epoch(train) [2][3900/7800] lr: 3.9046e-04 eta: 19:37:47 time: 0.7337 data_time: 0.0017 memory: 25074 loss: 0.1444 +2023/06/01 23:02:25 - mmengine - INFO - Epoch(train) [2][4000/7800] lr: 3.9046e-04 eta: 19:32:49 time: 0.7078 data_time: 0.0018 memory: 25074 loss: 0.1457 +2023/06/01 23:03:40 - mmengine - INFO - Epoch(train) [2][4100/7800] lr: 3.9046e-04 eta: 19:28:13 time: 0.7532 data_time: 0.0021 memory: 25074 loss: 0.1403 +2023/06/01 23:04:57 - mmengine - INFO - Exp name: convnext_small_4xb256_fake7m-lr4e-4_20230601_193207 +2023/06/01 23:04:57 - mmengine - INFO - Epoch(train) [2][4200/7800] lr: 3.9046e-04 eta: 19:23:43 time: 0.7085 data_time: 0.0019 memory: 25074 loss: 0.1420 +2023/06/01 23:06:12 - mmengine - INFO - Epoch(train) [2][4300/7800] lr: 3.9046e-04 eta: 19:19:13 time: 0.7066 data_time: 0.0017 memory: 25074 loss: 0.1451 +2023/06/01 23:07:28 - mmengine - INFO - Epoch(train) [2][4400/7800] lr: 3.9046e-04 eta: 19:14:47 time: 0.7116 data_time: 0.0019 memory: 25074 loss: 0.1489 +2023/06/01 23:08:43 - mmengine - INFO - Epoch(train) [2][4500/7800] lr: 3.9046e-04 eta: 19:10:22 time: 0.7098 data_time: 0.0022 memory: 25074 loss: 0.1328 +2023/06/01 23:10:02 - mmengine - INFO - Epoch(train) [2][4600/7800] lr: 3.9046e-04 eta: 19:06:15 time: 0.7055 data_time: 0.0016 memory: 25074 loss: 0.1407 +2023/06/01 23:11:14 - mmengine - INFO - Epoch(train) [2][4700/7800] lr: 3.9046e-04 eta: 19:01:41 time: 0.7075 data_time: 0.0016 memory: 25074 loss: 0.1413 +2023/06/01 23:12:28 - mmengine - INFO - Epoch(train) [2][4800/7800] lr: 3.9046e-04 eta: 18:57:19 time: 0.7152 data_time: 0.0013 memory: 25074 loss: 0.1367 +2023/06/01 23:13:40 - mmengine - INFO - Epoch(train) [2][4900/7800] lr: 3.9046e-04 eta: 18:52:47 time: 0.7070 data_time: 0.0019 memory: 25074 loss: 0.1440 +2023/06/01 23:14:59 - mmengine - INFO - Epoch(train) [2][5000/7800] lr: 3.9046e-04 eta: 18:48:52 time: 0.7081 data_time: 0.0014 memory: 25074 loss: 0.1091 +2023/06/01 23:16:10 - mmengine - INFO - Epoch(train) [2][5100/7800] lr: 3.9046e-04 eta: 18:44:22 time: 0.7043 data_time: 0.0013 memory: 25074 loss: 0.1354 +2023/06/01 23:17:22 - mmengine - INFO - Exp name: convnext_small_4xb256_fake7m-lr4e-4_20230601_193207 +2023/06/01 23:17:22 - mmengine - INFO - Epoch(train) [2][5200/7800] lr: 3.9046e-04 eta: 18:40:00 time: 0.7107 data_time: 0.0020 memory: 25074 loss: 0.1416 +2023/06/01 23:18:33 - mmengine - INFO - Epoch(train) [2][5300/7800] lr: 3.9046e-04 eta: 18:35:40 time: 0.7087 data_time: 0.0019 memory: 25074 loss: 0.1293 +2023/06/01 23:19:45 - mmengine - INFO - Epoch(train) [2][5400/7800] lr: 3.9046e-04 eta: 18:31:22 time: 0.7078 data_time: 0.0014 memory: 25074 loss: 0.1384 +2023/06/01 23:21:00 - mmengine - INFO - Epoch(train) [2][5500/7800] lr: 3.9046e-04 eta: 18:27:22 time: 0.7970 data_time: 0.0018 memory: 25074 loss: 0.1351 +2023/06/01 23:22:14 - mmengine - INFO - Epoch(train) [2][5600/7800] lr: 3.9046e-04 eta: 18:23:21 time: 0.7428 data_time: 0.0025 memory: 25074 loss: 0.1224 +2023/06/01 23:23:26 - mmengine - INFO - Epoch(train) [2][5700/7800] lr: 3.9046e-04 eta: 18:19:14 time: 0.7067 data_time: 0.0023 memory: 25074 loss: 0.1189 +2023/06/01 23:24:37 - mmengine - INFO - Epoch(train) [2][5800/7800] lr: 3.9046e-04 eta: 18:15:05 time: 0.7213 data_time: 0.0012 memory: 25074 loss: 0.1204 +2023/06/01 23:25:50 - mmengine - INFO - Epoch(train) [2][5900/7800] lr: 3.9046e-04 eta: 18:11:05 time: 0.7418 data_time: 0.0020 memory: 25074 loss: 0.1280 +2023/06/01 23:27:06 - mmengine - INFO - Epoch(train) [2][6000/7800] lr: 3.9046e-04 eta: 18:07:24 time: 0.7354 data_time: 0.0017 memory: 25074 loss: 0.1093 +2023/06/01 23:28:21 - mmengine - INFO - Epoch(train) [2][6100/7800] lr: 3.9046e-04 eta: 18:03:39 time: 0.8256 data_time: 0.0022 memory: 25074 loss: 0.1093 +2023/06/01 23:29:36 - mmengine - INFO - Exp name: convnext_small_4xb256_fake7m-lr4e-4_20230601_193207 +2023/06/01 23:29:36 - mmengine - INFO - Epoch(train) [2][6200/7800] lr: 3.9046e-04 eta: 17:59:57 time: 0.7257 data_time: 0.0018 memory: 25074 loss: 0.1156 +2023/06/01 23:30:50 - mmengine - INFO - Epoch(train) [2][6300/7800] lr: 3.9046e-04 eta: 17:56:13 time: 0.7052 data_time: 0.0018 memory: 25074 loss: 0.1120 +2023/06/01 23:32:07 - mmengine - INFO - Epoch(train) [2][6400/7800] lr: 3.9046e-04 eta: 17:52:41 time: 0.7792 data_time: 0.0018 memory: 25074 loss: 0.1105 +2023/06/01 23:33:25 - mmengine - INFO - Epoch(train) [2][6500/7800] lr: 3.9046e-04 eta: 17:49:19 time: 0.9102 data_time: 0.0033 memory: 25074 loss: 0.0977 +2023/06/01 23:34:44 - mmengine - INFO - Epoch(train) [2][6600/7800] lr: 3.9046e-04 eta: 17:46:02 time: 0.8740 data_time: 0.0042 memory: 25074 loss: 0.1228 +2023/06/01 23:36:01 - mmengine - INFO - Epoch(train) [2][6700/7800] lr: 3.9046e-04 eta: 17:42:38 time: 0.7378 data_time: 0.0013 memory: 25074 loss: 0.1226 +2023/06/01 23:37:22 - mmengine - INFO - Epoch(train) [2][6800/7800] lr: 3.9046e-04 eta: 17:39:36 time: 0.7553 data_time: 0.0012 memory: 25074 loss: 0.1168 +2023/06/01 23:38:41 - mmengine - INFO - Epoch(train) [2][6900/7800] lr: 3.9046e-04 eta: 17:36:24 time: 0.8950 data_time: 0.0032 memory: 25074 loss: 0.1075 +2023/06/01 23:39:59 - mmengine - INFO - Epoch(train) [2][7000/7800] lr: 3.9046e-04 eta: 17:33:07 time: 0.7629 data_time: 0.0013 memory: 25074 loss: 0.1148 +2023/06/01 23:41:21 - mmengine - INFO - Epoch(train) [2][7100/7800] lr: 3.9046e-04 eta: 17:30:13 time: 0.7381 data_time: 0.0014 memory: 25074 loss: 0.1204 +2023/06/01 23:42:42 - mmengine - INFO - Exp name: convnext_small_4xb256_fake7m-lr4e-4_20230601_193207 +2023/06/01 23:42:42 - mmengine - INFO - Epoch(train) [2][7200/7800] lr: 3.9046e-04 eta: 17:27:13 time: 0.7545 data_time: 0.0017 memory: 25074 loss: 0.0947 +2023/06/01 23:43:57 - mmengine - INFO - Epoch(train) [2][7300/7800] lr: 3.9046e-04 eta: 17:23:50 time: 0.8110 data_time: 0.0022 memory: 25074 loss: 0.0919 +2023/06/01 23:45:14 - mmengine - INFO - Epoch(train) [2][7400/7800] lr: 3.9046e-04 eta: 17:20:35 time: 0.7215 data_time: 0.0014 memory: 25074 loss: 0.1084 +2023/06/01 23:46:32 - mmengine - INFO - Epoch(train) [2][7500/7800] lr: 3.9046e-04 eta: 17:17:30 time: 0.7215 data_time: 0.0013 memory: 25074 loss: 0.0946 +2023/06/01 23:47:50 - mmengine - INFO - Epoch(train) [2][7600/7800] lr: 3.9046e-04 eta: 17:14:22 time: 0.7616 data_time: 0.0013 memory: 25074 loss: 0.0957 +2023/06/01 23:49:06 - mmengine - INFO - Epoch(train) [2][7700/7800] lr: 3.9046e-04 eta: 17:11:11 time: 0.8587 data_time: 0.0026 memory: 25074 loss: 0.1187 +2023/06/01 23:50:19 - mmengine - INFO - Exp name: convnext_small_4xb256_fake7m-lr4e-4_20230601_193207 +2023/06/01 23:50:19 - mmengine - INFO - Epoch(train) [2][7800/7800] lr: 3.9046e-04 eta: 17:07:50 time: 0.6593 data_time: 0.0011 memory: 25074 loss: 0.1062 +2023/06/01 23:50:19 - mmengine - INFO - Saving checkpoint at 2 epochs +2023/06/01 23:50:41 - mmengine - INFO - Epoch(val) [2][16/16] accuracy/top1: 96.6101 single-label/precision_classwise: [100.0, 0.0] single-label/recall_classwise: [96.61006164550781, 0.0] single-label/f1-score_classwise: [98.27580261230469, 0.0] data_time: 0.2034 time: 0.5981 +2023/06/01 23:52:05 - mmengine - INFO - Epoch(train) [3][ 100/7800] lr: 3.6276e-04 eta: 17:05:10 time: 0.7277 data_time: 0.0013 memory: 25074 loss: 0.1031 +2023/06/01 23:53:23 - mmengine - INFO - Epoch(train) [3][ 200/7800] lr: 3.6276e-04 eta: 17:02:11 time: 0.7492 data_time: 0.0465 memory: 25074 loss: 0.0878 +2023/06/01 23:54:40 - mmengine - INFO - Epoch(train) [3][ 300/7800] lr: 3.6276e-04 eta: 16:59:09 time: 0.7634 data_time: 0.0429 memory: 25074 loss: 0.1111 +2023/06/01 23:55:59 - mmengine - INFO - Exp name: convnext_small_4xb256_fake7m-lr4e-4_20230601_193207 +2023/06/01 23:55:59 - mmengine - INFO - Epoch(train) [3][ 400/7800] lr: 3.6276e-04 eta: 16:56:15 time: 0.7561 data_time: 0.0538 memory: 25074 loss: 0.1005 +2023/06/01 23:57:15 - mmengine - INFO - Epoch(train) [3][ 500/7800] lr: 3.6276e-04 eta: 16:53:08 time: 0.8451 data_time: 0.0944 memory: 25074 loss: 0.0835 +2023/06/01 23:58:31 - mmengine - INFO - Epoch(train) [3][ 600/7800] lr: 3.6276e-04 eta: 16:50:06 time: 0.7252 data_time: 0.0198 memory: 25074 loss: 0.1135 +2023/06/01 23:59:49 - mmengine - INFO - Epoch(train) [3][ 700/7800] lr: 3.6276e-04 eta: 16:47:11 time: 0.7392 data_time: 0.0363 memory: 25074 loss: 0.1013 +2023/06/02 00:01:06 - mmengine - INFO - Epoch(train) [3][ 800/7800] lr: 3.6276e-04 eta: 16:44:14 time: 0.7498 data_time: 0.0459 memory: 25074 loss: 0.1023 +2023/06/02 00:02:24 - mmengine - INFO - Epoch(train) [3][ 900/7800] lr: 3.6276e-04 eta: 16:41:23 time: 0.7870 data_time: 0.0714 memory: 25074 loss: 0.1002 +2023/06/02 00:03:40 - mmengine - INFO - Epoch(train) [3][1000/7800] lr: 3.6276e-04 eta: 16:38:25 time: 0.7834 data_time: 0.0785 memory: 25074 loss: 0.1016 +2023/06/02 00:04:57 - mmengine - INFO - Epoch(train) [3][1100/7800] lr: 3.6276e-04 eta: 16:35:33 time: 0.7039 data_time: 0.0013 memory: 25074 loss: 0.0844 +2023/06/02 00:06:15 - mmengine - INFO - Epoch(train) [3][1200/7800] lr: 3.6276e-04 eta: 16:32:43 time: 0.7562 data_time: 0.0531 memory: 25074 loss: 0.0883 +2023/06/02 00:07:34 - mmengine - INFO - Epoch(train) [3][1300/7800] lr: 3.6276e-04 eta: 16:30:00 time: 0.7354 data_time: 0.0328 memory: 25074 loss: 0.0958 +2023/06/02 00:08:50 - mmengine - INFO - Exp name: convnext_small_4xb256_fake7m-lr4e-4_20230601_193207 +2023/06/02 00:08:50 - mmengine - INFO - Epoch(train) [3][1400/7800] lr: 3.6276e-04 eta: 16:27:07 time: 0.9363 data_time: 0.1775 memory: 25074 loss: 0.0794 +2023/06/02 00:10:07 - mmengine - INFO - Epoch(train) [3][1500/7800] lr: 3.6276e-04 eta: 16:24:19 time: 0.7688 data_time: 0.0654 memory: 25074 loss: 0.0948 +2023/06/02 00:11:24 - mmengine - INFO - Epoch(train) [3][1600/7800] lr: 3.6276e-04 eta: 16:21:31 time: 0.7122 data_time: 0.0094 memory: 25074 loss: 0.0884 +2023/06/02 00:12:42 - mmengine - INFO - Epoch(train) [3][1700/7800] lr: 3.6276e-04 eta: 16:18:48 time: 0.7304 data_time: 0.0274 memory: 25074 loss: 0.0909 +2023/06/02 00:14:01 - mmengine - INFO - Epoch(train) [3][1800/7800] lr: 3.6276e-04 eta: 16:16:07 time: 0.7273 data_time: 0.0248 memory: 25074 loss: 0.0816 +2023/06/02 00:15:17 - mmengine - INFO - Epoch(train) [3][1900/7800] lr: 3.6276e-04 eta: 16:13:21 time: 0.7882 data_time: 0.0505 memory: 25074 loss: 0.0820 +2023/06/02 00:16:33 - mmengine - INFO - Epoch(train) [3][2000/7800] lr: 3.6276e-04 eta: 16:10:35 time: 0.7208 data_time: 0.0166 memory: 25074 loss: 0.0842 +2023/06/02 00:17:52 - mmengine - INFO - Epoch(train) [3][2100/7800] lr: 3.6276e-04 eta: 16:07:57 time: 0.8150 data_time: 0.0942 memory: 25074 loss: 0.0952 +2023/06/02 00:19:09 - mmengine - INFO - Epoch(train) [3][2200/7800] lr: 3.6276e-04 eta: 16:05:16 time: 0.7977 data_time: 0.0938 memory: 25074 loss: 0.1101 +2023/06/02 00:20:26 - mmengine - INFO - Epoch(train) [3][2300/7800] lr: 3.6276e-04 eta: 16:02:35 time: 0.9638 data_time: 0.2113 memory: 25074 loss: 0.0964 +2023/06/02 00:21:42 - mmengine - INFO - Exp name: convnext_small_4xb256_fake7m-lr4e-4_20230601_193207 +2023/06/02 00:21:42 - mmengine - INFO - Epoch(train) [3][2400/7800] lr: 3.6276e-04 eta: 15:59:52 time: 0.7268 data_time: 0.0207 memory: 25074 loss: 0.0952 +2023/06/02 00:23:01 - mmengine - INFO - Epoch(train) [3][2500/7800] lr: 3.6276e-04 eta: 15:57:21 time: 0.7533 data_time: 0.0420 memory: 25074 loss: 0.0861 +2023/06/02 00:24:19 - mmengine - INFO - Epoch(train) [3][2600/7800] lr: 3.6276e-04 eta: 15:54:46 time: 0.7038 data_time: 0.0014 memory: 25074 loss: 0.0895 +2023/06/02 00:25:37 - mmengine - INFO - Epoch(train) [3][2700/7800] lr: 3.6276e-04 eta: 15:52:12 time: 0.7630 data_time: 0.0442 memory: 25074 loss: 0.0815 +2023/06/02 00:26:52 - mmengine - INFO - Epoch(train) [3][2800/7800] lr: 3.6276e-04 eta: 15:49:28 time: 0.8042 data_time: 0.0557 memory: 25074 loss: 0.0898 +2023/06/02 00:28:07 - mmengine - INFO - Epoch(train) [3][2900/7800] lr: 3.6276e-04 eta: 15:46:46 time: 0.7311 data_time: 0.0014 memory: 25074 loss: 0.0885 +2023/06/02 00:29:22 - mmengine - INFO - Epoch(train) [3][3000/7800] lr: 3.6276e-04 eta: 15:44:06 time: 0.7047 data_time: 0.0013 memory: 25074 loss: 0.0714 +2023/06/02 00:30:39 - mmengine - INFO - Epoch(train) [3][3100/7800] lr: 3.6276e-04 eta: 15:41:34 time: 0.7873 data_time: 0.0014 memory: 25074 loss: 0.0924 +2023/06/02 00:31:57 - mmengine - INFO - Epoch(train) [3][3200/7800] lr: 3.6276e-04 eta: 15:39:04 time: 0.8584 data_time: 0.0024 memory: 25074 loss: 0.0801 +2023/06/02 00:33:12 - mmengine - INFO - Epoch(train) [3][3300/7800] lr: 3.6276e-04 eta: 15:36:25 time: 0.7364 data_time: 0.0013 memory: 25074 loss: 0.0935 +2023/06/02 00:34:29 - mmengine - INFO - Exp name: convnext_small_4xb256_fake7m-lr4e-4_20230601_193207 +2023/06/02 00:34:29 - mmengine - INFO - Epoch(train) [3][3400/7800] lr: 3.6276e-04 eta: 15:33:54 time: 0.7544 data_time: 0.0014 memory: 25074 loss: 0.0754 +2023/06/02 00:35:45 - mmengine - INFO - Epoch(train) [3][3500/7800] lr: 3.6276e-04 eta: 15:31:20 time: 0.7403 data_time: 0.0018 memory: 25074 loss: 0.0880 +2023/06/02 00:37:04 - mmengine - INFO - Epoch(train) [3][3600/7800] lr: 3.6276e-04 eta: 15:28:58 time: 0.7126 data_time: 0.0013 memory: 25074 loss: 0.0794 +2023/06/02 00:38:23 - mmengine - INFO - Epoch(train) [3][3700/7800] lr: 3.6276e-04 eta: 15:26:33 time: 0.8992 data_time: 0.0043 memory: 25074 loss: 0.0750 +2023/06/02 00:39:38 - mmengine - INFO - Epoch(train) [3][3800/7800] lr: 3.6276e-04 eta: 15:24:00 time: 0.7162 data_time: 0.0013 memory: 25074 loss: 0.0925 +2023/06/02 00:40:56 - mmengine - INFO - Epoch(train) [3][3900/7800] lr: 3.6276e-04 eta: 15:21:33 time: 0.7639 data_time: 0.0014 memory: 25074 loss: 0.0705 +2023/06/02 00:42:16 - mmengine - INFO - Epoch(train) [3][4000/7800] lr: 3.6276e-04 eta: 15:19:18 time: 0.7643 data_time: 0.0021 memory: 25074 loss: 0.0739 +2023/06/02 00:43:33 - mmengine - INFO - Epoch(train) [3][4100/7800] lr: 3.6276e-04 eta: 15:16:51 time: 0.7534 data_time: 0.0013 memory: 25074 loss: 0.0771 +2023/06/02 00:44:47 - mmengine - INFO - Epoch(train) [3][4200/7800] lr: 3.6276e-04 eta: 15:14:18 time: 0.8158 data_time: 0.0023 memory: 25074 loss: 0.0668 +2023/06/02 00:46:03 - mmengine - INFO - Epoch(train) [3][4300/7800] lr: 3.6276e-04 eta: 15:11:49 time: 0.7153 data_time: 0.0019 memory: 25074 loss: 0.0910 +2023/06/02 00:47:20 - mmengine - INFO - Exp name: convnext_small_4xb256_fake7m-lr4e-4_20230601_193207 +2023/06/02 00:47:20 - mmengine - INFO - Epoch(train) [3][4400/7800] lr: 3.6276e-04 eta: 15:09:24 time: 0.7268 data_time: 0.0013 memory: 25074 loss: 0.0727 +2023/06/02 00:48:36 - mmengine - INFO - Epoch(train) [3][4500/7800] lr: 3.6276e-04 eta: 15:06:59 time: 0.7686 data_time: 0.0017 memory: 25074 loss: 0.0823 +2023/06/02 00:49:55 - mmengine - INFO - Epoch(train) [3][4600/7800] lr: 3.6276e-04 eta: 15:04:43 time: 0.7393 data_time: 0.0014 memory: 25074 loss: 0.0660 +2023/06/02 00:51:08 - mmengine - INFO - Epoch(train) [3][4700/7800] lr: 3.6276e-04 eta: 15:02:09 time: 0.7038 data_time: 0.0014 memory: 25074 loss: 0.0788 +2023/06/02 00:52:25 - mmengine - INFO - Epoch(train) [3][4800/7800] lr: 3.6276e-04 eta: 14:59:48 time: 0.7664 data_time: 0.0013 memory: 25074 loss: 0.0639 +2023/06/02 00:53:44 - mmengine - INFO - Epoch(train) [3][4900/7800] lr: 3.6276e-04 eta: 14:57:31 time: 0.7615 data_time: 0.0014 memory: 25074 loss: 0.0645 +2023/06/02 00:55:01 - mmengine - INFO - Epoch(train) [3][5000/7800] lr: 3.6276e-04 eta: 14:55:11 time: 0.7718 data_time: 0.0013 memory: 25074 loss: 0.0709 +2023/06/02 00:56:17 - mmengine - INFO - Epoch(train) [3][5100/7800] lr: 3.6276e-04 eta: 14:52:50 time: 0.8924 data_time: 0.0029 memory: 25074 loss: 0.0807 +2023/06/02 00:57:32 - mmengine - INFO - Epoch(train) [3][5200/7800] lr: 3.6276e-04 eta: 14:50:25 time: 0.7351 data_time: 0.0014 memory: 25074 loss: 0.0633 +2023/06/02 00:58:49 - mmengine - INFO - Epoch(train) [3][5300/7800] lr: 3.6276e-04 eta: 14:48:07 time: 0.7061 data_time: 0.0013 memory: 25074 loss: 0.0720 +2023/06/02 01:00:07 - mmengine - INFO - Exp name: convnext_small_4xb256_fake7m-lr4e-4_20230601_193207 +2023/06/02 01:00:07 - mmengine - INFO - Epoch(train) [3][5400/7800] lr: 3.6276e-04 eta: 14:45:53 time: 0.7415 data_time: 0.0015 memory: 25074 loss: 0.0693 +2023/06/02 01:01:25 - mmengine - INFO - Epoch(train) [3][5500/7800] lr: 3.6276e-04 eta: 14:43:38 time: 0.7101 data_time: 0.0013 memory: 25074 loss: 0.0728 +2023/06/02 01:02:41 - mmengine - INFO - Epoch(train) [3][5600/7800] lr: 3.6276e-04 eta: 14:41:19 time: 0.8412 data_time: 0.0025 memory: 25074 loss: 0.0712 +2023/06/02 01:03:58 - mmengine - INFO - Epoch(train) [3][5700/7800] lr: 3.6276e-04 eta: 14:39:03 time: 0.7576 data_time: 0.0012 memory: 25074 loss: 0.0705 +2023/06/02 01:05:16 - mmengine - INFO - Epoch(train) [3][5800/7800] lr: 3.6276e-04 eta: 14:36:49 time: 0.7439 data_time: 0.0013 memory: 25074 loss: 0.0619 +2023/06/02 01:06:32 - mmengine - INFO - Epoch(train) [3][5900/7800] lr: 3.6276e-04 eta: 14:34:32 time: 0.7911 data_time: 0.0014 memory: 25074 loss: 0.0657 +2023/06/02 01:07:50 - mmengine - INFO - Epoch(train) [3][6000/7800] lr: 3.6276e-04 eta: 14:32:21 time: 0.8509 data_time: 0.0026 memory: 25074 loss: 0.0577 +2023/06/02 01:09:08 - mmengine - INFO - Epoch(train) [3][6100/7800] lr: 3.6276e-04 eta: 14:30:09 time: 0.7543 data_time: 0.0015 memory: 25074 loss: 0.0676 +2023/06/02 01:10:25 - mmengine - INFO - Epoch(train) [3][6200/7800] lr: 3.6276e-04 eta: 14:27:56 time: 0.7568 data_time: 0.0012 memory: 25074 loss: 0.0743 +2023/06/02 01:11:43 - mmengine - INFO - Epoch(train) [3][6300/7800] lr: 3.6276e-04 eta: 14:25:47 time: 0.7757 data_time: 0.0014 memory: 25074 loss: 0.0724 +2023/06/02 01:13:01 - mmengine - INFO - Exp name: convnext_small_4xb256_fake7m-lr4e-4_20230601_193207 +2023/06/02 01:13:01 - mmengine - INFO - Epoch(train) [3][6400/7800] lr: 3.6276e-04 eta: 14:23:36 time: 0.7807 data_time: 0.0014 memory: 25074 loss: 0.0723 +2023/06/02 01:14:17 - mmengine - INFO - Epoch(train) [3][6500/7800] lr: 3.6276e-04 eta: 14:21:22 time: 0.8099 data_time: 0.0046 memory: 25074 loss: 0.0550 +2023/06/02 01:15:34 - mmengine - INFO - Epoch(train) [3][6600/7800] lr: 3.6276e-04 eta: 14:19:10 time: 0.7357 data_time: 0.0016 memory: 25074 loss: 0.0614 +2023/06/02 01:16:51 - mmengine - INFO - Epoch(train) [3][6700/7800] lr: 3.6276e-04 eta: 14:17:00 time: 0.7164 data_time: 0.0013 memory: 25074 loss: 0.0563 +2023/06/02 01:18:06 - mmengine - INFO - Epoch(train) [3][6800/7800] lr: 3.6276e-04 eta: 14:14:45 time: 0.7336 data_time: 0.0012 memory: 25074 loss: 0.0602 +2023/06/02 01:19:23 - mmengine - INFO - Epoch(train) [3][6900/7800] lr: 3.6276e-04 eta: 14:12:34 time: 0.7601 data_time: 0.0020 memory: 25074 loss: 0.0672 +2023/06/02 01:20:37 - mmengine - INFO - Epoch(train) [3][7000/7800] lr: 3.6276e-04 eta: 14:10:17 time: 0.8072 data_time: 0.0013 memory: 25074 loss: 0.0679 +2023/06/02 01:21:53 - mmengine - INFO - Epoch(train) [3][7100/7800] lr: 3.6276e-04 eta: 14:08:06 time: 0.7624 data_time: 0.0014 memory: 25074 loss: 0.0546 +2023/06/02 01:23:09 - mmengine - INFO - Epoch(train) [3][7200/7800] lr: 3.6276e-04 eta: 14:05:56 time: 0.7364 data_time: 0.0015 memory: 25074 loss: 0.0602 +2023/06/02 01:24:26 - mmengine - INFO - Epoch(train) [3][7300/7800] lr: 3.6276e-04 eta: 14:03:47 time: 0.7205 data_time: 0.0177 memory: 25074 loss: 0.0597 +2023/06/02 01:25:40 - mmengine - INFO - Exp name: convnext_small_4xb256_fake7m-lr4e-4_20230601_193207 +2023/06/02 01:25:40 - mmengine - INFO - Epoch(train) [3][7400/7800] lr: 3.6276e-04 eta: 14:01:32 time: 0.8036 data_time: 0.0394 memory: 25074 loss: 0.0582 +2023/06/02 01:26:59 - mmengine - INFO - Epoch(train) [3][7500/7800] lr: 3.6276e-04 eta: 13:59:30 time: 0.8376 data_time: 0.1330 memory: 25074 loss: 0.0549 +2023/06/02 01:28:15 - mmengine - INFO - Epoch(train) [3][7600/7800] lr: 3.6276e-04 eta: 13:57:22 time: 0.7252 data_time: 0.0084 memory: 25074 loss: 0.0659 +2023/06/02 01:29:32 - mmengine - INFO - Epoch(train) [3][7700/7800] lr: 3.6276e-04 eta: 13:55:15 time: 0.7339 data_time: 0.0290 memory: 25074 loss: 0.0624 +2023/06/02 01:30:46 - mmengine - INFO - Exp name: convnext_small_4xb256_fake7m-lr4e-4_20230601_193207 +2023/06/02 01:30:46 - mmengine - INFO - Epoch(train) [3][7800/7800] lr: 3.6276e-04 eta: 13:53:03 time: 0.8273 data_time: 0.1349 memory: 25074 loss: 0.0593 +2023/06/02 01:30:46 - mmengine - INFO - Saving checkpoint at 3 epochs +2023/06/02 01:31:07 - mmengine - INFO - Epoch(val) [3][16/16] accuracy/top1: 97.8915 single-label/precision_classwise: [100.0, 0.0] single-label/recall_classwise: [97.89154815673828, 0.0] single-label/f1-score_classwise: [98.9345474243164, 0.0] data_time: 0.1995 time: 0.5852 +2023/06/02 01:32:32 - mmengine - INFO - Epoch(train) [4][ 100/7800] lr: 3.1962e-04 eta: 13:51:17 time: 0.8813 data_time: 0.0035 memory: 25074 loss: 0.0452 +2023/06/02 01:33:47 - mmengine - INFO - Epoch(train) [4][ 200/7800] lr: 3.1962e-04 eta: 13:49:08 time: 0.7198 data_time: 0.0013 memory: 25074 loss: 0.0451 +2023/06/02 01:35:04 - mmengine - INFO - Epoch(train) [4][ 300/7800] lr: 3.1962e-04 eta: 13:47:02 time: 0.7505 data_time: 0.0014 memory: 25074 loss: 0.0567 +2023/06/02 01:36:19 - mmengine - INFO - Epoch(train) [4][ 400/7800] lr: 3.1962e-04 eta: 13:44:54 time: 0.7433 data_time: 0.0012 memory: 25074 loss: 0.0512 +2023/06/02 01:37:34 - mmengine - INFO - Epoch(train) [4][ 500/7800] lr: 3.1962e-04 eta: 13:42:45 time: 0.8092 data_time: 0.0025 memory: 25074 loss: 0.0510 +2023/06/02 01:38:47 - mmengine - INFO - Exp name: convnext_small_4xb256_fake7m-lr4e-4_20230601_193207 +2023/06/02 01:38:47 - mmengine - INFO - Epoch(train) [4][ 600/7800] lr: 3.1962e-04 eta: 13:40:35 time: 0.7508 data_time: 0.0014 memory: 25074 loss: 0.0508 +2023/06/02 01:40:06 - mmengine - INFO - Epoch(train) [4][ 700/7800] lr: 3.1962e-04 eta: 13:38:35 time: 0.8642 data_time: 0.0013 memory: 25074 loss: 0.0506 +2023/06/02 01:41:23 - mmengine - INFO - Epoch(train) [4][ 800/7800] lr: 3.1962e-04 eta: 13:36:34 time: 0.7549 data_time: 0.0014 memory: 25074 loss: 0.0605 +2023/06/02 01:42:42 - mmengine - INFO - Epoch(train) [4][ 900/7800] lr: 3.1962e-04 eta: 13:34:36 time: 0.7465 data_time: 0.0014 memory: 25074 loss: 0.0472 +2023/06/02 01:44:00 - mmengine - INFO - Epoch(train) [4][1000/7800] lr: 3.1962e-04 eta: 13:32:36 time: 0.8669 data_time: 0.0053 memory: 25074 loss: 0.0533 +2023/06/02 01:45:16 - mmengine - INFO - Epoch(train) [4][1100/7800] lr: 3.1962e-04 eta: 13:30:31 time: 0.7893 data_time: 0.0013 memory: 25074 loss: 0.0478 +2023/06/02 01:46:33 - mmengine - INFO - Epoch(train) [4][1200/7800] lr: 3.1962e-04 eta: 13:28:31 time: 0.7030 data_time: 0.0013 memory: 25074 loss: 0.0473 +2023/06/02 01:47:47 - mmengine - INFO - Epoch(train) [4][1300/7800] lr: 3.1962e-04 eta: 13:26:24 time: 0.7806 data_time: 0.0013 memory: 25074 loss: 0.0477 +2023/06/02 01:49:05 - mmengine - INFO - Epoch(train) [4][1400/7800] lr: 3.1962e-04 eta: 13:24:25 time: 0.8386 data_time: 0.0025 memory: 25074 loss: 0.0517 +2023/06/02 01:50:22 - mmengine - INFO - Epoch(train) [4][1500/7800] lr: 3.1962e-04 eta: 13:22:25 time: 0.7871 data_time: 0.0014 memory: 25074 loss: 0.0527 +2023/06/02 01:51:40 - mmengine - INFO - Exp name: convnext_small_4xb256_fake7m-lr4e-4_20230601_193207 +2023/06/02 01:51:40 - mmengine - INFO - Epoch(train) [4][1600/7800] lr: 3.1962e-04 eta: 13:20:28 time: 0.7482 data_time: 0.0014 memory: 25074 loss: 0.0437 +2023/06/02 01:52:59 - mmengine - INFO - Epoch(train) [4][1700/7800] lr: 3.1962e-04 eta: 13:18:34 time: 0.7583 data_time: 0.0014 memory: 25074 loss: 0.0427 +2023/06/02 01:54:17 - mmengine - INFO - Epoch(train) [4][1800/7800] lr: 3.1962e-04 eta: 13:16:36 time: 0.8144 data_time: 0.0021 memory: 25074 loss: 0.0424 +2023/06/02 01:55:34 - mmengine - INFO - Epoch(train) [4][1900/7800] lr: 3.1962e-04 eta: 13:14:37 time: 0.8977 data_time: 0.0045 memory: 25074 loss: 0.0543 +2023/06/02 01:56:47 - mmengine - INFO - Epoch(train) [4][2000/7800] lr: 3.1962e-04 eta: 13:12:31 time: 0.7170 data_time: 0.0013 memory: 25074 loss: 0.0469 +2023/06/02 01:58:06 - mmengine - INFO - Epoch(train) [4][2100/7800] lr: 3.1962e-04 eta: 13:10:36 time: 0.7366 data_time: 0.0013 memory: 25074 loss: 0.0492 +2023/06/02 01:59:22 - mmengine - INFO - Epoch(train) [4][2200/7800] lr: 3.1962e-04 eta: 13:08:36 time: 0.7409 data_time: 0.0012 memory: 25074 loss: 0.0500 +2023/06/02 02:00:41 - mmengine - INFO - Epoch(train) [4][2300/7800] lr: 3.1962e-04 eta: 13:06:42 time: 0.7745 data_time: 0.0014 memory: 25074 loss: 0.0506 +2023/06/02 02:01:56 - mmengine - INFO - Epoch(train) [4][2400/7800] lr: 3.1962e-04 eta: 13:04:43 time: 0.7800 data_time: 0.0030 memory: 25074 loss: 0.0369 +2023/06/02 02:03:13 - mmengine - INFO - Epoch(train) [4][2500/7800] lr: 3.1962e-04 eta: 13:02:45 time: 0.7255 data_time: 0.0014 memory: 25074 loss: 0.0309 +2023/06/02 02:04:30 - mmengine - INFO - Exp name: convnext_small_4xb256_fake7m-lr4e-4_20230601_193207 +2023/06/02 02:04:30 - mmengine - INFO - Epoch(train) [4][2600/7800] lr: 3.1962e-04 eta: 13:00:48 time: 0.7262 data_time: 0.0013 memory: 25074 loss: 0.0335 +2023/06/02 02:05:47 - mmengine - INFO - Epoch(train) [4][2700/7800] lr: 3.1962e-04 eta: 12:58:53 time: 0.7928 data_time: 0.0013 memory: 25074 loss: 0.0502 +2023/06/02 02:07:04 - mmengine - INFO - Epoch(train) [4][2800/7800] lr: 3.1962e-04 eta: 12:56:58 time: 1.0026 data_time: 0.0027 memory: 25074 loss: 0.0473 +2023/06/02 02:08:21 - mmengine - INFO - Epoch(train) [4][2900/7800] lr: 3.1962e-04 eta: 12:55:02 time: 0.7895 data_time: 0.0012 memory: 25074 loss: 0.0544 +2023/06/02 02:09:39 - mmengine - INFO - Epoch(train) [4][3000/7800] lr: 3.1962e-04 eta: 12:53:07 time: 0.7531 data_time: 0.0014 memory: 25074 loss: 0.0429 +2023/06/02 02:10:56 - mmengine - INFO - Epoch(train) [4][3100/7800] lr: 3.1962e-04 eta: 12:51:13 time: 0.7631 data_time: 0.0014 memory: 25074 loss: 0.0377 +2023/06/02 02:12:13 - mmengine - INFO - Epoch(train) [4][3200/7800] lr: 3.1962e-04 eta: 12:49:18 time: 0.7831 data_time: 0.0013 memory: 25074 loss: 0.0510 +2023/06/02 02:13:28 - mmengine - INFO - Epoch(train) [4][3300/7800] lr: 3.1962e-04 eta: 12:47:21 time: 0.7721 data_time: 0.0015 memory: 25074 loss: 0.0481 +2023/06/02 02:14:47 - mmengine - INFO - Epoch(train) [4][3400/7800] lr: 3.1962e-04 eta: 12:45:30 time: 0.7106 data_time: 0.0012 memory: 25074 loss: 0.0439 +2023/06/02 02:16:04 - mmengine - INFO - Epoch(train) [4][3500/7800] lr: 3.1962e-04 eta: 12:43:36 time: 0.7423 data_time: 0.0014 memory: 25074 loss: 0.0322 +2023/06/02 02:17:22 - mmengine - INFO - Exp name: convnext_small_4xb256_fake7m-lr4e-4_20230601_193207 +2023/06/02 02:17:22 - mmengine - INFO - Epoch(train) [4][3600/7800] lr: 3.1962e-04 eta: 12:41:44 time: 0.7716 data_time: 0.0013 memory: 25074 loss: 0.0452 +2023/06/02 02:18:38 - mmengine - INFO - Epoch(train) [4][3700/7800] lr: 3.1962e-04 eta: 12:39:50 time: 0.8700 data_time: 0.0040 memory: 25074 loss: 0.0379 +2023/06/02 02:19:55 - mmengine - INFO - Epoch(train) [4][3800/7800] lr: 3.1962e-04 eta: 12:37:56 time: 0.7465 data_time: 0.0012 memory: 25074 loss: 0.0414 +2023/06/02 02:21:12 - mmengine - INFO - Epoch(train) [4][3900/7800] lr: 3.1962e-04 eta: 12:36:03 time: 0.7334 data_time: 0.0012 memory: 25074 loss: 0.0318 +2023/06/02 02:22:30 - mmengine - INFO - Epoch(train) [4][4000/7800] lr: 3.1962e-04 eta: 12:34:13 time: 0.7902 data_time: 0.0680 memory: 25074 loss: 0.0386 +2023/06/02 02:23:47 - mmengine - INFO - Epoch(train) [4][4100/7800] lr: 3.1962e-04 eta: 12:32:20 time: 0.7996 data_time: 0.0013 memory: 25074 loss: 0.0369 +2023/06/02 02:25:04 - mmengine - INFO - Epoch(train) [4][4200/7800] lr: 3.1962e-04 eta: 12:30:28 time: 0.8273 data_time: 0.0018 memory: 25074 loss: 0.0491 +2023/06/02 02:26:20 - mmengine - INFO - Epoch(train) [4][4300/7800] lr: 3.1962e-04 eta: 12:28:35 time: 0.7333 data_time: 0.0014 memory: 25074 loss: 0.0331 +2023/06/02 02:27:38 - mmengine - INFO - Epoch(train) [4][4400/7800] lr: 3.1962e-04 eta: 12:26:45 time: 0.7441 data_time: 0.0016 memory: 25074 loss: 0.0357 +2023/06/02 02:28:55 - mmengine - INFO - Epoch(train) [4][4500/7800] lr: 3.1962e-04 eta: 12:24:54 time: 0.7311 data_time: 0.0013 memory: 25074 loss: 0.0313 +2023/06/02 02:30:11 - mmengine - INFO - Exp name: convnext_small_4xb256_fake7m-lr4e-4_20230601_193207 +2023/06/02 02:30:11 - mmengine - INFO - Epoch(train) [4][4600/7800] lr: 3.1962e-04 eta: 12:23:01 time: 0.8084 data_time: 0.0037 memory: 25074 loss: 0.0349 +2023/06/02 02:31:25 - mmengine - INFO - Epoch(train) [4][4700/7800] lr: 3.1962e-04 eta: 12:21:05 time: 0.7614 data_time: 0.0013 memory: 25074 loss: 0.0390 +2023/06/02 02:32:42 - mmengine - INFO - Epoch(train) [4][4800/7800] lr: 3.1962e-04 eta: 12:19:14 time: 0.7503 data_time: 0.0015 memory: 25074 loss: 0.0287 +2023/06/02 02:33:58 - mmengine - INFO - Epoch(train) [4][4900/7800] lr: 3.1962e-04 eta: 12:17:22 time: 0.7055 data_time: 0.0014 memory: 25074 loss: 0.0424 +2023/06/02 02:35:13 - mmengine - INFO - Epoch(train) [4][5000/7800] lr: 3.1962e-04 eta: 12:15:29 time: 0.8059 data_time: 0.0029 memory: 25074 loss: 0.0364 +2023/06/02 02:36:28 - mmengine - INFO - Epoch(train) [4][5100/7800] lr: 3.1962e-04 eta: 12:13:35 time: 0.7404 data_time: 0.0013 memory: 25074 loss: 0.0338 +2023/06/02 02:37:45 - mmengine - INFO - Epoch(train) [4][5200/7800] lr: 3.1962e-04 eta: 12:11:45 time: 0.7099 data_time: 0.0014 memory: 25074 loss: 0.0460 +2023/06/02 02:39:00 - mmengine - INFO - Epoch(train) [4][5300/7800] lr: 3.1962e-04 eta: 12:09:53 time: 0.7666 data_time: 0.0015 memory: 25074 loss: 0.0366 +2023/06/02 02:40:16 - mmengine - INFO - Epoch(train) [4][5400/7800] lr: 3.1962e-04 eta: 12:08:03 time: 0.7370 data_time: 0.0013 memory: 25074 loss: 0.0313 +2023/06/02 02:41:29 - mmengine - INFO - Epoch(train) [4][5500/7800] lr: 3.1962e-04 eta: 12:06:07 time: 0.7406 data_time: 0.0014 memory: 25074 loss: 0.0284 +2023/06/02 02:42:46 - mmengine - INFO - Exp name: convnext_small_4xb256_fake7m-lr4e-4_20230601_193207 +2023/06/02 02:42:46 - mmengine - INFO - Epoch(train) [4][5600/7800] lr: 3.1962e-04 eta: 12:04:19 time: 0.7131 data_time: 0.0013 memory: 25074 loss: 0.0277 +2023/06/02 02:44:02 - mmengine - INFO - Epoch(train) [4][5700/7800] lr: 3.1962e-04 eta: 12:02:28 time: 0.7044 data_time: 0.0014 memory: 25074 loss: 0.0293 +2023/06/02 02:45:18 - mmengine - INFO - Epoch(train) [4][5800/7800] lr: 3.1962e-04 eta: 12:00:38 time: 0.7160 data_time: 0.0013 memory: 25074 loss: 0.0368 +2023/06/02 02:46:32 - mmengine - INFO - Epoch(train) [4][5900/7800] lr: 3.1962e-04 eta: 11:58:46 time: 0.8058 data_time: 0.0018 memory: 25074 loss: 0.0364 +2023/06/02 02:47:49 - mmengine - INFO - Epoch(train) [4][6000/7800] lr: 3.1962e-04 eta: 11:56:58 time: 0.7128 data_time: 0.0015 memory: 25074 loss: 0.0253 +2023/06/02 02:49:05 - mmengine - INFO - Epoch(train) [4][6100/7800] lr: 3.1962e-04 eta: 11:55:09 time: 0.7350 data_time: 0.0013 memory: 25074 loss: 0.0473 +2023/06/02 02:50:21 - mmengine - INFO - Epoch(train) [4][6200/7800] lr: 3.1962e-04 eta: 11:53:19 time: 0.7089 data_time: 0.0013 memory: 25074 loss: 0.0364 +2023/06/02 02:51:34 - mmengine - INFO - Epoch(train) [4][6300/7800] lr: 3.1962e-04 eta: 11:51:27 time: 0.8332 data_time: 0.0033 memory: 25074 loss: 0.0305 +2023/06/02 02:52:50 - mmengine - INFO - Epoch(train) [4][6400/7800] lr: 3.1962e-04 eta: 11:49:38 time: 0.7499 data_time: 0.0012 memory: 25074 loss: 0.0299 +2023/06/02 02:54:07 - mmengine - INFO - Epoch(train) [4][6500/7800] lr: 3.1962e-04 eta: 11:47:51 time: 0.7201 data_time: 0.0014 memory: 25074 loss: 0.0347 +2023/06/02 02:55:26 - mmengine - INFO - Exp name: convnext_small_4xb256_fake7m-lr4e-4_20230601_193207 +2023/06/02 02:55:26 - mmengine - INFO - Epoch(train) [4][6600/7800] lr: 3.1962e-04 eta: 11:46:09 time: 0.7682 data_time: 0.0016 memory: 25074 loss: 0.0322 +2023/06/02 02:56:43 - mmengine - INFO - Epoch(train) [4][6700/7800] lr: 3.1962e-04 eta: 11:44:21 time: 0.8383 data_time: 0.0017 memory: 25074 loss: 0.0274 +2023/06/02 02:57:56 - mmengine - INFO - Epoch(train) [4][6800/7800] lr: 3.1962e-04 eta: 11:42:30 time: 0.7305 data_time: 0.0012 memory: 25074 loss: 0.0389 +2023/06/02 02:59:14 - mmengine - INFO - Epoch(train) [4][6900/7800] lr: 3.1962e-04 eta: 11:40:46 time: 0.7839 data_time: 0.0013 memory: 25074 loss: 0.0288 +2023/06/02 03:00:31 - mmengine - INFO - Epoch(train) [4][7000/7800] lr: 3.1962e-04 eta: 11:39:00 time: 0.7049 data_time: 0.0013 memory: 25074 loss: 0.0222 +2023/06/02 03:01:48 - mmengine - INFO - Epoch(train) [4][7100/7800] lr: 3.1962e-04 eta: 11:37:14 time: 0.7344 data_time: 0.0012 memory: 25074 loss: 0.0238 +2023/06/02 03:03:04 - mmengine - INFO - Epoch(train) [4][7200/7800] lr: 3.1962e-04 eta: 11:35:27 time: 0.9068 data_time: 0.0052 memory: 25074 loss: 0.0258 +2023/06/02 03:04:21 - mmengine - INFO - Epoch(train) [4][7300/7800] lr: 3.1962e-04 eta: 11:33:42 time: 0.7572 data_time: 0.0012 memory: 25074 loss: 0.0271 +2023/06/02 03:05:39 - mmengine - INFO - Epoch(train) [4][7400/7800] lr: 3.1962e-04 eta: 11:31:58 time: 0.7048 data_time: 0.0013 memory: 25074 loss: 0.0342 +2023/06/02 03:06:54 - mmengine - INFO - Epoch(train) [4][7500/7800] lr: 3.1962e-04 eta: 11:30:12 time: 0.7596 data_time: 0.0012 memory: 25074 loss: 0.0318 +2023/06/02 03:08:10 - mmengine - INFO - Exp name: convnext_small_4xb256_fake7m-lr4e-4_20230601_193207 +2023/06/02 03:08:10 - mmengine - INFO - Epoch(train) [4][7600/7800] lr: 3.1962e-04 eta: 11:28:25 time: 0.7889 data_time: 0.0021 memory: 25074 loss: 0.0216 +2023/06/02 03:09:24 - mmengine - INFO - Epoch(train) [4][7700/7800] lr: 3.1962e-04 eta: 11:26:37 time: 0.7223 data_time: 0.0013 memory: 25074 loss: 0.0249 +2023/06/02 03:10:39 - mmengine - INFO - Exp name: convnext_small_4xb256_fake7m-lr4e-4_20230601_193207 +2023/06/02 03:10:39 - mmengine - INFO - Epoch(train) [4][7800/7800] lr: 3.1962e-04 eta: 11:24:50 time: 0.6581 data_time: 0.0010 memory: 25074 loss: 0.0262 +2023/06/02 03:10:39 - mmengine - INFO - Saving checkpoint at 4 epochs +2023/06/02 03:10:58 - mmengine - INFO - Epoch(val) [4][16/16] accuracy/top1: 99.1288 single-label/precision_classwise: [100.0, 0.0] single-label/recall_classwise: [99.12884521484375, 0.0] single-label/f1-score_classwise: [99.5625228881836, 0.0] data_time: 0.1988 time: 0.5838 +2023/06/02 03:12:21 - mmengine - INFO - Epoch(train) [5][ 100/7800] lr: 2.6526e-04 eta: 11:23:15 time: 0.7176 data_time: 0.0015 memory: 25074 loss: 0.0277 +2023/06/02 03:13:37 - mmengine - INFO - Epoch(train) [5][ 200/7800] lr: 2.6526e-04 eta: 11:21:30 time: 0.7168 data_time: 0.0013 memory: 25074 loss: 0.0280 +2023/06/02 03:14:52 - mmengine - INFO - Epoch(train) [5][ 300/7800] lr: 2.6526e-04 eta: 11:19:43 time: 0.7129 data_time: 0.0013 memory: 25074 loss: 0.0251 +2023/06/02 03:16:08 - mmengine - INFO - Epoch(train) [5][ 400/7800] lr: 2.6526e-04 eta: 11:17:57 time: 0.9069 data_time: 0.0040 memory: 25074 loss: 0.0200 +2023/06/02 03:17:24 - mmengine - INFO - Epoch(train) [5][ 500/7800] lr: 2.6526e-04 eta: 11:16:13 time: 0.7485 data_time: 0.0014 memory: 25074 loss: 0.0353 +2023/06/02 03:18:41 - mmengine - INFO - Epoch(train) [5][ 600/7800] lr: 2.6526e-04 eta: 11:14:30 time: 0.7763 data_time: 0.0018 memory: 25074 loss: 0.0238 +2023/06/02 03:19:58 - mmengine - INFO - Epoch(train) [5][ 700/7800] lr: 2.6526e-04 eta: 11:12:47 time: 0.7267 data_time: 0.0014 memory: 25074 loss: 0.0185 +2023/06/02 03:21:16 - mmengine - INFO - Exp name: convnext_small_4xb256_fake7m-lr4e-4_20230601_193207 +2023/06/02 03:21:16 - mmengine - INFO - Epoch(train) [5][ 800/7800] lr: 2.6526e-04 eta: 11:11:05 time: 0.9306 data_time: 0.0022 memory: 25074 loss: 0.0222 +2023/06/02 03:22:30 - mmengine - INFO - Epoch(train) [5][ 900/7800] lr: 2.6526e-04 eta: 11:09:19 time: 0.7062 data_time: 0.0014 memory: 25074 loss: 0.0292 +2023/06/02 03:23:45 - mmengine - INFO - Epoch(train) [5][1000/7800] lr: 2.6526e-04 eta: 11:07:34 time: 0.7170 data_time: 0.0015 memory: 25074 loss: 0.0264 +2023/06/02 03:24:58 - mmengine - INFO - Epoch(train) [5][1100/7800] lr: 2.6526e-04 eta: 11:05:46 time: 0.7059 data_time: 0.0013 memory: 25074 loss: 0.0232 +2023/06/02 03:26:14 - mmengine - INFO - Epoch(train) [5][1200/7800] lr: 2.6526e-04 eta: 11:04:03 time: 0.8432 data_time: 0.0017 memory: 25074 loss: 0.0203 +2023/06/02 03:27:27 - mmengine - INFO - Epoch(train) [5][1300/7800] lr: 2.6526e-04 eta: 11:02:14 time: 0.7056 data_time: 0.0014 memory: 25074 loss: 0.0210 +2023/06/02 03:28:44 - mmengine - INFO - Epoch(train) [5][1400/7800] lr: 2.6526e-04 eta: 11:00:33 time: 0.8002 data_time: 0.0014 memory: 25074 loss: 0.0299 +2023/06/02 03:30:01 - mmengine - INFO - Epoch(train) [5][1500/7800] lr: 2.6526e-04 eta: 10:58:51 time: 0.7579 data_time: 0.0013 memory: 25074 loss: 0.0168 +2023/06/02 03:31:19 - mmengine - INFO - Epoch(train) [5][1600/7800] lr: 2.6526e-04 eta: 10:57:11 time: 0.7537 data_time: 0.0013 memory: 25074 loss: 0.0234 +2023/06/02 03:32:34 - mmengine - INFO - Epoch(train) [5][1700/7800] lr: 2.6526e-04 eta: 10:55:27 time: 0.7972 data_time: 0.0030 memory: 25074 loss: 0.0295 +2023/06/02 03:33:52 - mmengine - INFO - Exp name: convnext_small_4xb256_fake7m-lr4e-4_20230601_193207 +2023/06/02 03:33:52 - mmengine - INFO - Epoch(train) [5][1800/7800] lr: 2.6526e-04 eta: 10:53:47 time: 0.7700 data_time: 0.0014 memory: 25074 loss: 0.0247 +2023/06/02 03:35:09 - mmengine - INFO - Epoch(train) [5][1900/7800] lr: 2.6526e-04 eta: 10:52:06 time: 0.7092 data_time: 0.0015 memory: 25074 loss: 0.0253 +2023/06/02 03:36:26 - mmengine - INFO - Epoch(train) [5][2000/7800] lr: 2.6526e-04 eta: 10:50:26 time: 0.8257 data_time: 0.0019 memory: 25074 loss: 0.0204 +2023/06/02 03:37:44 - mmengine - INFO - Epoch(train) [5][2100/7800] lr: 2.6526e-04 eta: 10:48:46 time: 0.7289 data_time: 0.0014 memory: 25074 loss: 0.0347 +2023/06/02 03:38:59 - mmengine - INFO - Epoch(train) [5][2200/7800] lr: 2.6526e-04 eta: 10:47:03 time: 0.8441 data_time: 0.0029 memory: 25074 loss: 0.0173 +2023/06/02 03:40:17 - mmengine - INFO - Epoch(train) [5][2300/7800] lr: 2.6526e-04 eta: 10:45:24 time: 0.7313 data_time: 0.0015 memory: 25074 loss: 0.0200 +2023/06/02 03:41:37 - mmengine - INFO - Epoch(train) [5][2400/7800] lr: 2.6526e-04 eta: 10:43:48 time: 0.7822 data_time: 0.0015 memory: 25074 loss: 0.0236 +2023/06/02 03:42:56 - mmengine - INFO - Epoch(train) [5][2500/7800] lr: 2.6526e-04 eta: 10:42:10 time: 0.7080 data_time: 0.0015 memory: 25074 loss: 0.0168 +2023/06/02 03:44:11 - mmengine - INFO - Epoch(train) [5][2600/7800] lr: 2.6526e-04 eta: 10:40:27 time: 0.8000 data_time: 0.0021 memory: 25074 loss: 0.0176 +2023/06/02 03:45:24 - mmengine - INFO - Epoch(train) [5][2700/7800] lr: 2.6526e-04 eta: 10:38:42 time: 0.7039 data_time: 0.0013 memory: 25074 loss: 0.0159 +2023/06/02 03:46:40 - mmengine - INFO - Exp name: convnext_small_4xb256_fake7m-lr4e-4_20230601_193207 +2023/06/02 03:46:40 - mmengine - INFO - Epoch(train) [5][2800/7800] lr: 2.6526e-04 eta: 10:37:02 time: 0.7533 data_time: 0.0013 memory: 25074 loss: 0.0238 +2023/06/02 03:47:57 - mmengine - INFO - Epoch(train) [5][2900/7800] lr: 2.6526e-04 eta: 10:35:22 time: 0.7325 data_time: 0.0012 memory: 25074 loss: 0.0187 +2023/06/02 03:49:14 - mmengine - INFO - Epoch(train) [5][3000/7800] lr: 2.6526e-04 eta: 10:33:42 time: 0.7467 data_time: 0.0013 memory: 25074 loss: 0.0173 +2023/06/02 03:50:26 - mmengine - INFO - Epoch(train) [5][3100/7800] lr: 2.6526e-04 eta: 10:31:57 time: 0.7075 data_time: 0.0012 memory: 25074 loss: 0.0178 +2023/06/02 03:51:42 - mmengine - INFO - Epoch(train) [5][3200/7800] lr: 2.6526e-04 eta: 10:30:16 time: 0.7232 data_time: 0.0014 memory: 25074 loss: 0.0222 +2023/06/02 03:52:58 - mmengine - INFO - Epoch(train) [5][3300/7800] lr: 2.6526e-04 eta: 10:28:36 time: 0.7038 data_time: 0.0013 memory: 25074 loss: 0.0235 +2023/06/02 03:54:12 - mmengine - INFO - Epoch(train) [5][3400/7800] lr: 2.6526e-04 eta: 10:26:54 time: 0.7553 data_time: 0.0013 memory: 25074 loss: 0.0202 +2023/06/02 03:55:27 - mmengine - INFO - Epoch(train) [5][3500/7800] lr: 2.6526e-04 eta: 10:25:13 time: 0.8633 data_time: 0.0024 memory: 25074 loss: 0.0261 +2023/06/02 03:56:42 - mmengine - INFO - Epoch(train) [5][3600/7800] lr: 2.6526e-04 eta: 10:23:31 time: 0.7161 data_time: 0.0015 memory: 25074 loss: 0.0204 +2023/06/02 03:57:58 - mmengine - INFO - Epoch(train) [5][3700/7800] lr: 2.6526e-04 eta: 10:21:52 time: 0.7410 data_time: 0.0013 memory: 25074 loss: 0.0241 +2023/06/02 03:59:14 - mmengine - INFO - Exp name: convnext_small_4xb256_fake7m-lr4e-4_20230601_193207 +2023/06/02 03:59:14 - mmengine - INFO - Epoch(train) [5][3800/7800] lr: 2.6526e-04 eta: 10:20:12 time: 0.7227 data_time: 0.0014 memory: 25074 loss: 0.0229 +2023/06/02 04:00:28 - mmengine - INFO - Epoch(train) [5][3900/7800] lr: 2.6526e-04 eta: 10:18:30 time: 0.8395 data_time: 0.0022 memory: 25074 loss: 0.0289 +2023/06/02 04:01:43 - mmengine - INFO - Epoch(train) [5][4000/7800] lr: 2.6526e-04 eta: 10:16:50 time: 0.7871 data_time: 0.0014 memory: 25074 loss: 0.0215 +2023/06/02 04:03:00 - mmengine - INFO - Epoch(train) [5][4100/7800] lr: 2.6526e-04 eta: 10:15:11 time: 0.7504 data_time: 0.0012 memory: 25074 loss: 0.0211 +2023/06/02 04:04:16 - mmengine - INFO - Epoch(train) [5][4200/7800] lr: 2.6526e-04 eta: 10:13:33 time: 0.8191 data_time: 0.0014 memory: 25074 loss: 0.0202 +2023/06/02 04:05:32 - mmengine - INFO - Epoch(train) [5][4300/7800] lr: 2.6526e-04 eta: 10:11:53 time: 0.8432 data_time: 0.0044 memory: 25074 loss: 0.0225 +2023/06/02 04:06:47 - mmengine - INFO - Epoch(train) [5][4400/7800] lr: 2.6526e-04 eta: 10:10:13 time: 0.7227 data_time: 0.0016 memory: 25074 loss: 0.0193 +2023/06/02 04:08:05 - mmengine - INFO - Epoch(train) [5][4500/7800] lr: 2.6526e-04 eta: 10:08:37 time: 0.7565 data_time: 0.0015 memory: 25074 loss: 0.0206 +2023/06/02 04:09:23 - mmengine - INFO - Epoch(train) [5][4600/7800] lr: 2.6526e-04 eta: 10:07:01 time: 0.7280 data_time: 0.0013 memory: 25074 loss: 0.0103 +2023/06/02 04:10:42 - mmengine - INFO - Epoch(train) [5][4700/7800] lr: 2.6526e-04 eta: 10:05:26 time: 0.8212 data_time: 0.0014 memory: 25074 loss: 0.0215 +2023/06/02 04:12:00 - mmengine - INFO - Exp name: convnext_small_4xb256_fake7m-lr4e-4_20230601_193207 +2023/06/02 04:12:00 - mmengine - INFO - Epoch(train) [5][4800/7800] lr: 2.6526e-04 eta: 10:03:51 time: 0.8626 data_time: 0.0033 memory: 25074 loss: 0.0248 +2023/06/02 04:13:19 - mmengine - INFO - Epoch(train) [5][4900/7800] lr: 2.6526e-04 eta: 10:02:15 time: 0.7454 data_time: 0.0012 memory: 25074 loss: 0.0237 +2023/06/02 04:14:38 - mmengine - INFO - Epoch(train) [5][5000/7800] lr: 2.6526e-04 eta: 10:00:41 time: 0.7424 data_time: 0.0012 memory: 25074 loss: 0.0118 +2023/06/02 04:15:56 - mmengine - INFO - Epoch(train) [5][5100/7800] lr: 2.6526e-04 eta: 9:59:06 time: 0.7848 data_time: 0.0013 memory: 25074 loss: 0.0179 +2023/06/02 04:17:15 - mmengine - INFO - Epoch(train) [5][5200/7800] lr: 2.6526e-04 eta: 9:57:31 time: 0.7393 data_time: 0.0013 memory: 25074 loss: 0.0205 +2023/06/02 04:18:31 - mmengine - INFO - Epoch(train) [5][5300/7800] lr: 2.6526e-04 eta: 9:55:54 time: 0.7775 data_time: 0.0026 memory: 25074 loss: 0.0251 +2023/06/02 04:19:48 - mmengine - INFO - Epoch(train) [5][5400/7800] lr: 2.6526e-04 eta: 9:54:17 time: 0.7563 data_time: 0.0013 memory: 25074 loss: 0.0214 +2023/06/02 04:21:05 - mmengine - INFO - Epoch(train) [5][5500/7800] lr: 2.6526e-04 eta: 9:52:40 time: 0.7428 data_time: 0.0012 memory: 25074 loss: 0.0172 +2023/06/02 04:22:21 - mmengine - INFO - Epoch(train) [5][5600/7800] lr: 2.6526e-04 eta: 9:51:03 time: 0.8290 data_time: 0.0012 memory: 25074 loss: 0.0118 +2023/06/02 04:23:39 - mmengine - INFO - Epoch(train) [5][5700/7800] lr: 2.6526e-04 eta: 9:49:28 time: 0.8662 data_time: 0.0026 memory: 25074 loss: 0.0161 +2023/06/02 04:24:55 - mmengine - INFO - Exp name: convnext_small_4xb256_fake7m-lr4e-4_20230601_193207 +2023/06/02 04:24:55 - mmengine - INFO - Epoch(train) [5][5800/7800] lr: 2.6526e-04 eta: 9:47:51 time: 0.7516 data_time: 0.0015 memory: 25074 loss: 0.0144 +2023/06/02 04:26:14 - mmengine - INFO - Epoch(train) [5][5900/7800] lr: 2.6526e-04 eta: 9:46:17 time: 0.7796 data_time: 0.0018 memory: 25074 loss: 0.0236 +2023/06/02 04:27:28 - mmengine - INFO - Epoch(train) [5][6000/7800] lr: 2.6526e-04 eta: 9:44:38 time: 0.7682 data_time: 0.0016 memory: 25074 loss: 0.0186 +2023/06/02 04:28:44 - mmengine - INFO - Epoch(train) [5][6100/7800] lr: 2.6526e-04 eta: 9:43:01 time: 0.7692 data_time: 0.0016 memory: 25074 loss: 0.0128 +2023/06/02 04:29:59 - mmengine - INFO - Epoch(train) [5][6200/7800] lr: 2.6526e-04 eta: 9:41:23 time: 0.8142 data_time: 0.0018 memory: 25074 loss: 0.0192 +2023/06/02 04:31:11 - mmengine - INFO - Epoch(train) [5][6300/7800] lr: 2.6526e-04 eta: 9:39:42 time: 0.7051 data_time: 0.0017 memory: 25074 loss: 0.0169 +2023/06/02 04:32:28 - mmengine - INFO - Epoch(train) [5][6400/7800] lr: 2.6526e-04 eta: 9:38:06 time: 0.7511 data_time: 0.0014 memory: 25074 loss: 0.0186 +2023/06/02 04:33:46 - mmengine - INFO - Epoch(train) [5][6500/7800] lr: 2.6526e-04 eta: 9:36:32 time: 0.7846 data_time: 0.0014 memory: 25074 loss: 0.0145 +2023/06/02 04:35:03 - mmengine - INFO - Epoch(train) [5][6600/7800] lr: 2.6526e-04 eta: 9:34:57 time: 0.8151 data_time: 0.0013 memory: 25074 loss: 0.0166 +2023/06/02 04:36:19 - mmengine - INFO - Epoch(train) [5][6700/7800] lr: 2.6526e-04 eta: 9:33:21 time: 0.8038 data_time: 0.0029 memory: 25074 loss: 0.0215 +2023/06/02 04:37:35 - mmengine - INFO - Exp name: convnext_small_4xb256_fake7m-lr4e-4_20230601_193207 +2023/06/02 04:37:35 - mmengine - INFO - Epoch(train) [5][6800/7800] lr: 2.6526e-04 eta: 9:31:45 time: 0.7133 data_time: 0.0014 memory: 25074 loss: 0.0212 +2023/06/02 04:38:52 - mmengine - INFO - Epoch(train) [5][6900/7800] lr: 2.6526e-04 eta: 9:30:09 time: 0.7230 data_time: 0.0015 memory: 25074 loss: 0.0165 +2023/06/02 04:40:10 - mmengine - INFO - Epoch(train) [5][7000/7800] lr: 2.6526e-04 eta: 9:28:36 time: 0.7170 data_time: 0.0014 memory: 25074 loss: 0.0256 +2023/06/02 04:41:25 - mmengine - INFO - Epoch(train) [5][7100/7800] lr: 2.6526e-04 eta: 9:26:59 time: 0.8092 data_time: 0.0023 memory: 25074 loss: 0.0162 +2023/06/02 04:42:41 - mmengine - INFO - Epoch(train) [5][7200/7800] lr: 2.6526e-04 eta: 9:25:23 time: 0.7202 data_time: 0.0012 memory: 25074 loss: 0.0110 +2023/06/02 04:43:59 - mmengine - INFO - Epoch(train) [5][7300/7800] lr: 2.6526e-04 eta: 9:23:49 time: 0.7683 data_time: 0.0015 memory: 25074 loss: 0.0215 +2023/06/02 04:45:15 - mmengine - INFO - Epoch(train) [5][7400/7800] lr: 2.6526e-04 eta: 9:22:15 time: 0.7448 data_time: 0.0016 memory: 25074 loss: 0.0148 +2023/06/02 04:46:32 - mmengine - INFO - Epoch(train) [5][7500/7800] lr: 2.6526e-04 eta: 9:20:40 time: 0.7692 data_time: 0.0013 memory: 25074 loss: 0.0205 +2023/06/02 04:47:49 - mmengine - INFO - Epoch(train) [5][7600/7800] lr: 2.6526e-04 eta: 9:19:05 time: 0.8574 data_time: 0.0036 memory: 25074 loss: 0.0180 +2023/06/02 04:49:04 - mmengine - INFO - Epoch(train) [5][7700/7800] lr: 2.6526e-04 eta: 9:17:29 time: 0.7245 data_time: 0.0014 memory: 25074 loss: 0.0179 +2023/06/02 04:50:20 - mmengine - INFO - Exp name: convnext_small_4xb256_fake7m-lr4e-4_20230601_193207 +2023/06/02 04:50:20 - mmengine - INFO - Epoch(train) [5][7800/7800] lr: 2.6526e-04 eta: 9:15:54 time: 0.6578 data_time: 0.0009 memory: 25074 loss: 0.0144 +2023/06/02 04:50:20 - mmengine - INFO - Saving checkpoint at 5 epochs +2023/06/02 04:50:39 - mmengine - INFO - Epoch(val) [5][16/16] accuracy/top1: 99.5518 single-label/precision_classwise: [100.0, 0.0] single-label/recall_classwise: [99.55179595947266, 0.0] single-label/f1-score_classwise: [99.77539825439453, 0.0] data_time: 0.1989 time: 0.5841 +2023/06/02 04:52:03 - mmengine - INFO - Epoch(train) [6][ 100/7800] lr: 2.0500e-04 eta: 9:14:27 time: 0.7040 data_time: 0.0018 memory: 25074 loss: 0.0192 +2023/06/02 04:53:21 - mmengine - INFO - Epoch(train) [6][ 200/7800] lr: 2.0500e-04 eta: 9:12:53 time: 0.7180 data_time: 0.0014 memory: 25074 loss: 0.0132 +2023/06/02 04:54:35 - mmengine - INFO - Epoch(train) [6][ 300/7800] lr: 2.0500e-04 eta: 9:11:17 time: 0.7914 data_time: 0.0265 memory: 25074 loss: 0.0116 +2023/06/02 04:55:51 - mmengine - INFO - Epoch(train) [6][ 400/7800] lr: 2.0500e-04 eta: 9:09:42 time: 0.7774 data_time: 0.0015 memory: 25074 loss: 0.0129 +2023/06/02 04:57:06 - mmengine - INFO - Epoch(train) [6][ 500/7800] lr: 2.0500e-04 eta: 9:08:06 time: 0.7367 data_time: 0.0012 memory: 25074 loss: 0.0138 +2023/06/02 04:58:22 - mmengine - INFO - Epoch(train) [6][ 600/7800] lr: 2.0500e-04 eta: 9:06:32 time: 0.7232 data_time: 0.0016 memory: 25074 loss: 0.0137 +2023/06/02 04:59:35 - mmengine - INFO - Epoch(train) [6][ 700/7800] lr: 2.0500e-04 eta: 9:04:54 time: 0.7332 data_time: 0.0015 memory: 25074 loss: 0.0121 +2023/06/02 05:00:51 - mmengine - INFO - Epoch(train) [6][ 800/7800] lr: 2.0500e-04 eta: 9:03:20 time: 0.8491 data_time: 0.0042 memory: 25074 loss: 0.0172 +2023/06/02 05:02:06 - mmengine - INFO - Epoch(train) [6][ 900/7800] lr: 2.0500e-04 eta: 9:01:45 time: 0.7486 data_time: 0.0013 memory: 25074 loss: 0.0087 +2023/06/02 05:03:21 - mmengine - INFO - Exp name: convnext_small_4xb256_fake7m-lr4e-4_20230601_193207 +2023/06/02 05:03:21 - mmengine - INFO - Epoch(train) [6][1000/7800] lr: 2.0500e-04 eta: 9:00:10 time: 0.7175 data_time: 0.0013 memory: 25074 loss: 0.0071 +2023/06/02 05:04:36 - mmengine - INFO - Epoch(train) [6][1100/7800] lr: 2.0500e-04 eta: 8:58:35 time: 0.7293 data_time: 0.0013 memory: 25074 loss: 0.0144 +2023/06/02 05:05:53 - mmengine - INFO - Epoch(train) [6][1200/7800] lr: 2.0500e-04 eta: 8:57:01 time: 0.8143 data_time: 0.0036 memory: 25074 loss: 0.0161 +2023/06/02 05:07:06 - mmengine - INFO - Epoch(train) [6][1300/7800] lr: 2.0500e-04 eta: 8:55:25 time: 0.7256 data_time: 0.0013 memory: 25074 loss: 0.0196 +2023/06/02 05:08:21 - mmengine - INFO - Epoch(train) [6][1400/7800] lr: 2.0500e-04 eta: 8:53:50 time: 0.7181 data_time: 0.0016 memory: 25074 loss: 0.0086 +2023/06/02 05:09:39 - mmengine - INFO - Epoch(train) [6][1500/7800] lr: 2.0500e-04 eta: 8:52:18 time: 0.7049 data_time: 0.0014 memory: 25074 loss: 0.0151 +2023/06/02 05:10:54 - mmengine - INFO - Epoch(train) [6][1600/7800] lr: 2.0500e-04 eta: 8:50:43 time: 0.7559 data_time: 0.0013 memory: 25074 loss: 0.0176 +2023/06/02 05:12:09 - mmengine - INFO - Epoch(train) [6][1700/7800] lr: 2.0500e-04 eta: 8:49:10 time: 0.7320 data_time: 0.0016 memory: 25074 loss: 0.0141 +2023/06/02 05:13:25 - mmengine - INFO - Epoch(train) [6][1800/7800] lr: 2.0500e-04 eta: 8:47:36 time: 0.7044 data_time: 0.0014 memory: 25074 loss: 0.0172 +2023/06/02 05:14:39 - mmengine - INFO - Epoch(train) [6][1900/7800] lr: 2.0500e-04 eta: 8:46:01 time: 0.7042 data_time: 0.0015 memory: 25074 loss: 0.0128 +2023/06/02 05:15:53 - mmengine - INFO - Exp name: convnext_small_4xb256_fake7m-lr4e-4_20230601_193207 +2023/06/02 05:15:53 - mmengine - INFO - Epoch(train) [6][2000/7800] lr: 2.0500e-04 eta: 8:44:26 time: 0.7031 data_time: 0.0013 memory: 25074 loss: 0.0176 +2023/06/02 05:17:05 - mmengine - INFO - Epoch(train) [6][2100/7800] lr: 2.0500e-04 eta: 8:42:49 time: 0.7236 data_time: 0.0020 memory: 25074 loss: 0.0095 +2023/06/02 05:18:19 - mmengine - INFO - Epoch(train) [6][2200/7800] lr: 2.0500e-04 eta: 8:41:14 time: 0.7041 data_time: 0.0013 memory: 25074 loss: 0.0158 +2023/06/02 05:19:34 - mmengine - INFO - Epoch(train) [6][2300/7800] lr: 2.0500e-04 eta: 8:39:40 time: 0.7404 data_time: 0.0015 memory: 25074 loss: 0.0109 +2023/06/02 05:20:50 - mmengine - INFO - Epoch(train) [6][2400/7800] lr: 2.0500e-04 eta: 8:38:07 time: 0.7050 data_time: 0.0015 memory: 25074 loss: 0.0096 +2023/06/02 05:22:04 - mmengine - INFO - Epoch(train) [6][2500/7800] lr: 2.0500e-04 eta: 8:36:32 time: 0.7942 data_time: 0.0026 memory: 25074 loss: 0.0085 +2023/06/02 05:23:18 - mmengine - INFO - Epoch(train) [6][2600/7800] lr: 2.0500e-04 eta: 8:34:58 time: 0.7253 data_time: 0.0014 memory: 25074 loss: 0.0143 +2023/06/02 05:24:32 - mmengine - INFO - Epoch(train) [6][2700/7800] lr: 2.0500e-04 eta: 8:33:24 time: 0.7215 data_time: 0.0014 memory: 25074 loss: 0.0163 +2023/06/02 05:25:46 - mmengine - INFO - Epoch(train) [6][2800/7800] lr: 2.0500e-04 eta: 8:31:49 time: 0.7048 data_time: 0.0012 memory: 25074 loss: 0.0130 +2023/06/02 05:27:01 - mmengine - INFO - Epoch(train) [6][2900/7800] lr: 2.0500e-04 eta: 8:30:16 time: 0.8043 data_time: 0.0025 memory: 25074 loss: 0.0122 +2023/06/02 05:28:15 - mmengine - INFO - Exp name: convnext_small_4xb256_fake7m-lr4e-4_20230601_193207 +2023/06/02 05:28:15 - mmengine - INFO - Epoch(train) [6][3000/7800] lr: 2.0500e-04 eta: 8:28:42 time: 0.7057 data_time: 0.0017 memory: 25074 loss: 0.0126 +2023/06/02 05:29:29 - mmengine - INFO - Epoch(train) [6][3100/7800] lr: 2.0500e-04 eta: 8:27:08 time: 0.7048 data_time: 0.0014 memory: 25074 loss: 0.0113 +2023/06/02 05:30:44 - mmengine - INFO - Epoch(train) [6][3200/7800] lr: 2.0500e-04 eta: 8:25:35 time: 0.7044 data_time: 0.0015 memory: 25074 loss: 0.0145 +2023/06/02 05:31:58 - mmengine - INFO - Epoch(train) [6][3300/7800] lr: 2.0500e-04 eta: 8:24:01 time: 0.8029 data_time: 0.0026 memory: 25074 loss: 0.0125 +2023/06/02 05:33:10 - mmengine - INFO - Epoch(train) [6][3400/7800] lr: 2.0500e-04 eta: 8:22:26 time: 0.7167 data_time: 0.0013 memory: 25074 loss: 0.0096 +2023/06/02 05:34:26 - mmengine - INFO - Epoch(train) [6][3500/7800] lr: 2.0500e-04 eta: 8:20:53 time: 0.7345 data_time: 0.0014 memory: 25074 loss: 0.0152 +2023/06/02 05:35:43 - mmengine - INFO - Epoch(train) [6][3600/7800] lr: 2.0500e-04 eta: 8:19:23 time: 0.8146 data_time: 0.0014 memory: 25074 loss: 0.0115 +2023/06/02 05:37:00 - mmengine - INFO - Epoch(train) [6][3700/7800] lr: 2.0500e-04 eta: 8:17:52 time: 0.8299 data_time: 0.0021 memory: 25074 loss: 0.0089 +2023/06/02 05:38:14 - mmengine - INFO - Epoch(train) [6][3800/7800] lr: 2.0500e-04 eta: 8:16:19 time: 0.7325 data_time: 0.0014 memory: 25074 loss: 0.0151 +2023/06/02 05:39:32 - mmengine - INFO - Epoch(train) [6][3900/7800] lr: 2.0500e-04 eta: 8:14:48 time: 0.7424 data_time: 0.0013 memory: 25074 loss: 0.0092 +2023/06/02 05:40:50 - mmengine - INFO - Exp name: convnext_small_4xb256_fake7m-lr4e-4_20230601_193207 +2023/06/02 05:40:50 - mmengine - INFO - Epoch(train) [6][4000/7800] lr: 2.0500e-04 eta: 8:13:19 time: 0.7807 data_time: 0.0016 memory: 25074 loss: 0.0135 +2023/06/02 05:42:08 - mmengine - INFO - Epoch(train) [6][4100/7800] lr: 2.0500e-04 eta: 8:11:48 time: 0.7461 data_time: 0.0016 memory: 25074 loss: 0.0199 +2023/06/02 05:43:25 - mmengine - INFO - Epoch(train) [6][4200/7800] lr: 2.0500e-04 eta: 8:10:18 time: 0.9208 data_time: 0.0048 memory: 25074 loss: 0.0126 +2023/06/02 05:44:41 - mmengine - INFO - Epoch(train) [6][4300/7800] lr: 2.0500e-04 eta: 8:08:46 time: 0.7424 data_time: 0.0015 memory: 25074 loss: 0.0106 +2023/06/02 05:45:58 - mmengine - INFO - Epoch(train) [6][4400/7800] lr: 2.0500e-04 eta: 8:07:16 time: 0.7358 data_time: 0.0014 memory: 25074 loss: 0.0142 +2023/06/02 05:47:14 - mmengine - INFO - Epoch(train) [6][4500/7800] lr: 2.0500e-04 eta: 8:05:45 time: 0.7345 data_time: 0.0013 memory: 25074 loss: 0.0112 +2023/06/02 05:48:33 - mmengine - INFO - Epoch(train) [6][4600/7800] lr: 2.0500e-04 eta: 8:04:16 time: 0.7354 data_time: 0.0194 memory: 25074 loss: 0.0140 +2023/06/02 05:49:45 - mmengine - INFO - Epoch(train) [6][4700/7800] lr: 2.0500e-04 eta: 8:02:42 time: 0.7458 data_time: 0.0236 memory: 25074 loss: 0.0153 +2023/06/02 05:51:02 - mmengine - INFO - Epoch(train) [6][4800/7800] lr: 2.0500e-04 eta: 8:01:12 time: 0.7644 data_time: 0.0495 memory: 25074 loss: 0.0135 +2023/06/02 05:52:20 - mmengine - INFO - Epoch(train) [6][4900/7800] lr: 2.0500e-04 eta: 7:59:42 time: 0.7958 data_time: 0.0844 memory: 25074 loss: 0.0091 +2023/06/02 05:53:37 - mmengine - INFO - Exp name: convnext_small_4xb256_fake7m-lr4e-4_20230601_193207 +2023/06/02 05:53:37 - mmengine - INFO - Epoch(train) [6][5000/7800] lr: 2.0500e-04 eta: 7:58:12 time: 0.7159 data_time: 0.0013 memory: 25074 loss: 0.0113 +2023/06/02 05:54:54 - mmengine - INFO - Epoch(train) [6][5100/7800] lr: 2.0500e-04 eta: 7:56:42 time: 0.9396 data_time: 0.1972 memory: 25074 loss: 0.0144 +2023/06/02 05:56:07 - mmengine - INFO - Epoch(train) [6][5200/7800] lr: 2.0500e-04 eta: 7:55:09 time: 0.7914 data_time: 0.0821 memory: 25074 loss: 0.0129 +2023/06/02 05:57:27 - mmengine - INFO - Epoch(train) [6][5300/7800] lr: 2.0500e-04 eta: 7:53:41 time: 0.7634 data_time: 0.0503 memory: 25074 loss: 0.0093 +2023/06/02 05:58:43 - mmengine - INFO - Epoch(train) [6][5400/7800] lr: 2.0500e-04 eta: 7:52:10 time: 0.7753 data_time: 0.0563 memory: 25074 loss: 0.0082 +2023/06/02 06:00:00 - mmengine - INFO - Epoch(train) [6][5500/7800] lr: 2.0500e-04 eta: 7:50:41 time: 0.7634 data_time: 0.0608 memory: 25074 loss: 0.0099 +2023/06/02 06:01:13 - mmengine - INFO - Epoch(train) [6][5600/7800] lr: 2.0500e-04 eta: 7:49:08 time: 0.7847 data_time: 0.0189 memory: 25074 loss: 0.0076 +2023/06/02 06:02:27 - mmengine - INFO - Epoch(train) [6][5700/7800] lr: 2.0500e-04 eta: 7:47:36 time: 0.7342 data_time: 0.0263 memory: 25074 loss: 0.0119 +2023/06/02 06:03:46 - mmengine - INFO - Epoch(train) [6][5800/7800] lr: 2.0500e-04 eta: 7:46:08 time: 0.7233 data_time: 0.0013 memory: 25074 loss: 0.0120 +2023/06/02 06:05:03 - mmengine - INFO - Epoch(train) [6][5900/7800] lr: 2.0500e-04 eta: 7:44:38 time: 0.8072 data_time: 0.0925 memory: 25074 loss: 0.0099 +2023/06/02 06:06:18 - mmengine - INFO - Exp name: convnext_small_4xb256_fake7m-lr4e-4_20230601_193207 +2023/06/02 06:06:18 - mmengine - INFO - Epoch(train) [6][6000/7800] lr: 2.0500e-04 eta: 7:43:07 time: 0.8687 data_time: 0.1281 memory: 25074 loss: 0.0146 +2023/06/02 06:07:31 - mmengine - INFO - Epoch(train) [6][6100/7800] lr: 2.0500e-04 eta: 7:41:35 time: 0.7035 data_time: 0.0011 memory: 25074 loss: 0.0074 +2023/06/02 06:08:46 - mmengine - INFO - Epoch(train) [6][6200/7800] lr: 2.0500e-04 eta: 7:40:04 time: 0.7305 data_time: 0.0013 memory: 25074 loss: 0.0104 +2023/06/02 06:10:04 - mmengine - INFO - Epoch(train) [6][6300/7800] lr: 2.0500e-04 eta: 7:38:35 time: 0.7492 data_time: 0.0014 memory: 25074 loss: 0.0144 +2023/06/02 06:11:21 - mmengine - INFO - Epoch(train) [6][6400/7800] lr: 2.0500e-04 eta: 7:37:06 time: 0.7857 data_time: 0.0014 memory: 25074 loss: 0.0153 +2023/06/02 06:12:35 - mmengine - INFO - Epoch(train) [6][6500/7800] lr: 2.0500e-04 eta: 7:35:35 time: 0.7276 data_time: 0.0016 memory: 25074 loss: 0.0144 +2023/06/02 06:13:51 - mmengine - INFO - Epoch(train) [6][6600/7800] lr: 2.0500e-04 eta: 7:34:05 time: 0.7304 data_time: 0.0015 memory: 25074 loss: 0.0111 +2023/06/02 06:15:11 - mmengine - INFO - Epoch(train) [6][6700/7800] lr: 2.0500e-04 eta: 7:32:38 time: 0.8665 data_time: 0.0015 memory: 25074 loss: 0.0100 +2023/06/02 06:16:30 - mmengine - INFO - Epoch(train) [6][6800/7800] lr: 2.0500e-04 eta: 7:31:10 time: 0.8030 data_time: 0.0013 memory: 25074 loss: 0.0105 +2023/06/02 06:17:47 - mmengine - INFO - Epoch(train) [6][6900/7800] lr: 2.0500e-04 eta: 7:29:41 time: 0.7231 data_time: 0.0021 memory: 25074 loss: 0.0120 +2023/06/02 06:19:04 - mmengine - INFO - Exp name: convnext_small_4xb256_fake7m-lr4e-4_20230601_193207 +2023/06/02 06:19:04 - mmengine - INFO - Epoch(train) [6][7000/7800] lr: 2.0500e-04 eta: 7:28:12 time: 0.8995 data_time: 0.0035 memory: 25074 loss: 0.0104 +2023/06/02 06:20:23 - mmengine - INFO - Epoch(train) [6][7100/7800] lr: 2.0500e-04 eta: 7:26:45 time: 0.7369 data_time: 0.0014 memory: 25074 loss: 0.0120 +2023/06/02 06:21:43 - mmengine - INFO - Epoch(train) [6][7200/7800] lr: 2.0500e-04 eta: 7:25:18 time: 0.7986 data_time: 0.0014 memory: 25074 loss: 0.0090 +2023/06/02 06:23:04 - mmengine - INFO - Epoch(train) [6][7300/7800] lr: 2.0500e-04 eta: 7:23:52 time: 0.7629 data_time: 0.0016 memory: 25074 loss: 0.0095 +2023/06/02 06:24:23 - mmengine - INFO - Epoch(train) [6][7400/7800] lr: 2.0500e-04 eta: 7:22:24 time: 0.7718 data_time: 0.0013 memory: 25074 loss: 0.0160 +2023/06/02 06:25:42 - mmengine - INFO - Epoch(train) [6][7500/7800] lr: 2.0500e-04 eta: 7:20:57 time: 0.8350 data_time: 0.0026 memory: 25074 loss: 0.0193 +2023/06/02 06:26:58 - mmengine - INFO - Epoch(train) [6][7600/7800] lr: 2.0500e-04 eta: 7:19:27 time: 0.7859 data_time: 0.0015 memory: 25074 loss: 0.0140 +2023/06/02 06:28:17 - mmengine - INFO - Epoch(train) [6][7700/7800] lr: 2.0500e-04 eta: 7:18:00 time: 0.7804 data_time: 0.0013 memory: 25074 loss: 0.0082 +2023/06/02 06:29:32 - mmengine - INFO - Exp name: convnext_small_4xb256_fake7m-lr4e-4_20230601_193207 +2023/06/02 06:29:32 - mmengine - INFO - Epoch(train) [6][7800/7800] lr: 2.0500e-04 eta: 7:16:30 time: 0.6597 data_time: 0.0015 memory: 25074 loss: 0.0099 +2023/06/02 06:29:32 - mmengine - INFO - Saving checkpoint at 6 epochs +2023/06/02 06:29:52 - mmengine - INFO - Epoch(val) [6][16/16] accuracy/top1: 99.8674 single-label/precision_classwise: [100.0, 0.0] single-label/recall_classwise: [99.867431640625, 0.0] single-label/f1-score_classwise: [99.93367004394531, 0.0] data_time: 0.2034 time: 0.5902 +2023/06/02 06:31:18 - mmengine - INFO - Epoch(train) [7][ 100/7800] lr: 1.4474e-04 eta: 7:15:08 time: 0.8041 data_time: 0.0020 memory: 25074 loss: 0.0089 +2023/06/02 06:32:34 - mmengine - INFO - Exp name: convnext_small_4xb256_fake7m-lr4e-4_20230601_193207 +2023/06/02 06:32:34 - mmengine - INFO - Epoch(train) [7][ 200/7800] lr: 1.4474e-04 eta: 7:13:38 time: 0.7048 data_time: 0.0019 memory: 25074 loss: 0.0100 +2023/06/02 06:33:51 - mmengine - INFO - Epoch(train) [7][ 300/7800] lr: 1.4474e-04 eta: 7:12:10 time: 0.9699 data_time: 0.0019 memory: 25074 loss: 0.0072 +2023/06/02 06:35:04 - mmengine - INFO - Epoch(train) [7][ 400/7800] lr: 1.4474e-04 eta: 7:10:39 time: 0.7037 data_time: 0.0014 memory: 25074 loss: 0.0062 +2023/06/02 06:36:20 - mmengine - INFO - Epoch(train) [7][ 500/7800] lr: 1.4474e-04 eta: 7:09:10 time: 0.7363 data_time: 0.0014 memory: 25074 loss: 0.0094 +2023/06/02 06:37:37 - mmengine - INFO - Epoch(train) [7][ 600/7800] lr: 1.4474e-04 eta: 7:07:41 time: 0.7047 data_time: 0.0014 memory: 25074 loss: 0.0082 +2023/06/02 06:38:52 - mmengine - INFO - Epoch(train) [7][ 700/7800] lr: 1.4474e-04 eta: 7:06:12 time: 0.8040 data_time: 0.0013 memory: 25074 loss: 0.0058 +2023/06/02 06:40:07 - mmengine - INFO - Epoch(train) [7][ 800/7800] lr: 1.4474e-04 eta: 7:04:42 time: 0.7182 data_time: 0.0019 memory: 25074 loss: 0.0103 +2023/06/02 06:41:21 - mmengine - INFO - Epoch(train) [7][ 900/7800] lr: 1.4474e-04 eta: 7:03:13 time: 0.8044 data_time: 0.0016 memory: 25074 loss: 0.0081 +2023/06/02 06:42:37 - mmengine - INFO - Epoch(train) [7][1000/7800] lr: 1.4474e-04 eta: 7:01:44 time: 0.7166 data_time: 0.0013 memory: 25074 loss: 0.0073 +2023/06/02 06:43:54 - mmengine - INFO - Epoch(train) [7][1100/7800] lr: 1.4474e-04 eta: 7:00:15 time: 0.7935 data_time: 0.0015 memory: 25074 loss: 0.0035 +2023/06/02 06:45:08 - mmengine - INFO - Exp name: convnext_small_4xb256_fake7m-lr4e-4_20230601_193207 +2023/06/02 06:45:08 - mmengine - INFO - Epoch(train) [7][1200/7800] lr: 1.4474e-04 eta: 6:58:46 time: 0.8661 data_time: 0.0037 memory: 25074 loss: 0.0094 +2023/06/02 06:46:23 - mmengine - INFO - Epoch(train) [7][1300/7800] lr: 1.4474e-04 eta: 6:57:17 time: 0.7574 data_time: 0.0014 memory: 25074 loss: 0.0121 +2023/06/02 06:47:41 - mmengine - INFO - Epoch(train) [7][1400/7800] lr: 1.4474e-04 eta: 6:55:49 time: 0.7831 data_time: 0.0016 memory: 25074 loss: 0.0082 +2023/06/02 06:48:59 - mmengine - INFO - Epoch(train) [7][1500/7800] lr: 1.4474e-04 eta: 6:54:22 time: 0.7489 data_time: 0.0012 memory: 25074 loss: 0.0087 +2023/06/02 06:50:14 - mmengine - INFO - Epoch(train) [7][1600/7800] lr: 1.4474e-04 eta: 6:52:53 time: 0.8024 data_time: 0.0014 memory: 25074 loss: 0.0090 +2023/06/02 06:51:33 - mmengine - INFO - Epoch(train) [7][1700/7800] lr: 1.4474e-04 eta: 6:51:26 time: 0.8814 data_time: 0.0026 memory: 25074 loss: 0.0047 +2023/06/02 06:52:51 - mmengine - INFO - Epoch(train) [7][1800/7800] lr: 1.4474e-04 eta: 6:49:59 time: 0.7099 data_time: 0.0014 memory: 25074 loss: 0.0095 +2023/06/02 06:54:08 - mmengine - INFO - Epoch(train) [7][1900/7800] lr: 1.4474e-04 eta: 6:48:32 time: 0.7453 data_time: 0.0013 memory: 25074 loss: 0.0112 +2023/06/02 06:55:27 - mmengine - INFO - Epoch(train) [7][2000/7800] lr: 1.4474e-04 eta: 6:47:05 time: 0.7479 data_time: 0.0014 memory: 25074 loss: 0.0060 +2023/06/02 06:56:41 - mmengine - INFO - Epoch(train) [7][2100/7800] lr: 1.4474e-04 eta: 6:45:36 time: 0.8155 data_time: 0.0022 memory: 25074 loss: 0.0088 +2023/06/02 06:57:56 - mmengine - INFO - Exp name: convnext_small_4xb256_fake7m-lr4e-4_20230601_193207 +2023/06/02 06:57:56 - mmengine - INFO - Epoch(train) [7][2200/7800] lr: 1.4474e-04 eta: 6:44:07 time: 0.7294 data_time: 0.0013 memory: 25074 loss: 0.0091 +2023/06/02 06:59:16 - mmengine - INFO - Epoch(train) [7][2300/7800] lr: 1.4474e-04 eta: 6:42:41 time: 0.8101 data_time: 0.0015 memory: 25074 loss: 0.0098 +2023/06/02 07:00:34 - mmengine - INFO - Epoch(train) [7][2400/7800] lr: 1.4474e-04 eta: 6:41:15 time: 0.8001 data_time: 0.0013 memory: 25074 loss: 0.0080 +2023/06/02 07:01:53 - mmengine - INFO - Epoch(train) [7][2500/7800] lr: 1.4474e-04 eta: 6:39:48 time: 0.7050 data_time: 0.0013 memory: 25074 loss: 0.0085 +2023/06/02 07:03:10 - mmengine - INFO - Epoch(train) [7][2600/7800] lr: 1.4474e-04 eta: 6:38:21 time: 0.9377 data_time: 0.0032 memory: 25074 loss: 0.0069 +2023/06/02 07:04:25 - mmengine - INFO - Epoch(train) [7][2700/7800] lr: 1.4474e-04 eta: 6:36:52 time: 0.7347 data_time: 0.0014 memory: 25074 loss: 0.0041 +2023/06/02 07:05:41 - mmengine - INFO - Epoch(train) [7][2800/7800] lr: 1.4474e-04 eta: 6:35:24 time: 0.7203 data_time: 0.0013 memory: 25074 loss: 0.0047 +2023/06/02 07:06:55 - mmengine - INFO - Epoch(train) [7][2900/7800] lr: 1.4474e-04 eta: 6:33:55 time: 0.7055 data_time: 0.0016 memory: 25074 loss: 0.0125 +2023/06/02 07:08:11 - mmengine - INFO - Epoch(train) [7][3000/7800] lr: 1.4474e-04 eta: 6:32:28 time: 0.8159 data_time: 0.0013 memory: 25074 loss: 0.0108 +2023/06/02 07:09:27 - mmengine - INFO - Epoch(train) [7][3100/7800] lr: 1.4474e-04 eta: 6:31:00 time: 0.7243 data_time: 0.0016 memory: 25074 loss: 0.0031 +2023/06/02 07:10:44 - mmengine - INFO - Exp name: convnext_small_4xb256_fake7m-lr4e-4_20230601_193207 +2023/06/02 07:10:44 - mmengine - INFO - Epoch(train) [7][3200/7800] lr: 1.4474e-04 eta: 6:29:33 time: 0.7496 data_time: 0.0015 memory: 25074 loss: 0.0074 +2023/06/02 07:11:59 - mmengine - INFO - Epoch(train) [7][3300/7800] lr: 1.4474e-04 eta: 6:28:05 time: 0.7317 data_time: 0.0012 memory: 25074 loss: 0.0064 +2023/06/02 07:13:14 - mmengine - INFO - Epoch(train) [7][3400/7800] lr: 1.4474e-04 eta: 6:26:36 time: 0.7060 data_time: 0.0013 memory: 25074 loss: 0.0056 +2023/06/02 07:14:27 - mmengine - INFO - Epoch(train) [7][3500/7800] lr: 1.4474e-04 eta: 6:25:07 time: 0.8345 data_time: 0.0035 memory: 25074 loss: 0.0099 +2023/06/02 07:15:41 - mmengine - INFO - Epoch(train) [7][3600/7800] lr: 1.4474e-04 eta: 6:23:39 time: 0.7681 data_time: 0.0016 memory: 25074 loss: 0.0106 +2023/06/02 07:16:57 - mmengine - INFO - Epoch(train) [7][3700/7800] lr: 1.4474e-04 eta: 6:22:11 time: 0.7229 data_time: 0.0015 memory: 25074 loss: 0.0074 +2023/06/02 07:18:13 - mmengine - INFO - Epoch(train) [7][3800/7800] lr: 1.4474e-04 eta: 6:20:44 time: 0.7126 data_time: 0.0014 memory: 25074 loss: 0.0065 +2023/06/02 07:19:29 - mmengine - INFO - Epoch(train) [7][3900/7800] lr: 1.4474e-04 eta: 6:19:17 time: 0.8828 data_time: 0.0032 memory: 25074 loss: 0.0071 +2023/06/02 07:20:44 - mmengine - INFO - Epoch(train) [7][4000/7800] lr: 1.4474e-04 eta: 6:17:49 time: 0.7371 data_time: 0.0015 memory: 25074 loss: 0.0068 +2023/06/02 07:22:00 - mmengine - INFO - Epoch(train) [7][4100/7800] lr: 1.4474e-04 eta: 6:16:21 time: 0.7462 data_time: 0.0014 memory: 25074 loss: 0.0075 +2023/06/02 07:23:17 - mmengine - INFO - Exp name: convnext_small_4xb256_fake7m-lr4e-4_20230601_193207 +2023/06/02 07:23:17 - mmengine - INFO - Epoch(train) [7][4200/7800] lr: 1.4474e-04 eta: 6:14:55 time: 0.7587 data_time: 0.0015 memory: 25074 loss: 0.0060 +2023/06/02 07:24:35 - mmengine - INFO - Epoch(train) [7][4300/7800] lr: 1.4474e-04 eta: 6:13:29 time: 0.8089 data_time: 0.0013 memory: 25074 loss: 0.0076 +2023/06/02 07:25:52 - mmengine - INFO - Epoch(train) [7][4400/7800] lr: 1.4474e-04 eta: 6:12:02 time: 0.8470 data_time: 0.0029 memory: 25074 loss: 0.0110 +2023/06/02 07:27:09 - mmengine - INFO - Epoch(train) [7][4500/7800] lr: 1.4474e-04 eta: 6:10:35 time: 0.7581 data_time: 0.0014 memory: 25074 loss: 0.0158 +2023/06/02 07:28:26 - mmengine - INFO - Epoch(train) [7][4600/7800] lr: 1.4474e-04 eta: 6:09:09 time: 0.7477 data_time: 0.0013 memory: 25074 loss: 0.0081 +2023/06/02 07:29:44 - mmengine - INFO - Epoch(train) [7][4700/7800] lr: 1.4474e-04 eta: 6:07:43 time: 0.7334 data_time: 0.0013 memory: 25074 loss: 0.0095 +2023/06/02 07:31:00 - mmengine - INFO - Epoch(train) [7][4800/7800] lr: 1.4474e-04 eta: 6:06:16 time: 0.8429 data_time: 0.0023 memory: 25074 loss: 0.0079 +2023/06/02 07:32:15 - mmengine - INFO - Epoch(train) [7][4900/7800] lr: 1.4474e-04 eta: 6:04:49 time: 0.7119 data_time: 0.0013 memory: 25074 loss: 0.0077 +2023/06/02 07:33:34 - mmengine - INFO - Epoch(train) [7][5000/7800] lr: 1.4474e-04 eta: 6:03:23 time: 0.7642 data_time: 0.0014 memory: 25074 loss: 0.0098 +2023/06/02 07:34:52 - mmengine - INFO - Epoch(train) [7][5100/7800] lr: 1.4474e-04 eta: 6:01:57 time: 0.7312 data_time: 0.0015 memory: 25074 loss: 0.0060 +2023/06/02 07:36:10 - mmengine - INFO - Exp name: convnext_small_4xb256_fake7m-lr4e-4_20230601_193207 +2023/06/02 07:36:10 - mmengine - INFO - Epoch(train) [7][5200/7800] lr: 1.4474e-04 eta: 6:00:31 time: 0.7900 data_time: 0.0015 memory: 25074 loss: 0.0065 +2023/06/02 07:37:25 - mmengine - INFO - Epoch(train) [7][5300/7800] lr: 1.4474e-04 eta: 5:59:04 time: 0.7965 data_time: 0.0032 memory: 25074 loss: 0.0106 +2023/06/02 07:38:37 - mmengine - INFO - Epoch(train) [7][5400/7800] lr: 1.4474e-04 eta: 5:57:35 time: 0.7052 data_time: 0.0013 memory: 25074 loss: 0.0103 +2023/06/02 07:39:52 - mmengine - INFO - Epoch(train) [7][5500/7800] lr: 1.4474e-04 eta: 5:56:08 time: 0.7626 data_time: 0.0015 memory: 25074 loss: 0.0064 +2023/06/02 07:41:55 - mmengine - INFO - Epoch(train) [7][5600/7800] lr: 1.4474e-04 eta: 5:55:05 time: 0.7141 data_time: 0.0014 memory: 25074 loss: 0.0102 +2023/06/02 07:43:07 - mmengine - INFO - Epoch(train) [7][5700/7800] lr: 1.4474e-04 eta: 5:53:36 time: 0.7040 data_time: 0.0015 memory: 25074 loss: 0.0071 +2023/06/02 07:44:22 - mmengine - INFO - Epoch(train) [7][5800/7800] lr: 1.4474e-04 eta: 5:52:09 time: 0.7820 data_time: 0.0014 memory: 25074 loss: 0.0074 +2023/06/02 07:45:37 - mmengine - INFO - Epoch(train) [7][5900/7800] lr: 1.4474e-04 eta: 5:50:42 time: 0.9154 data_time: 0.0040 memory: 25074 loss: 0.0071 +2023/06/02 07:46:48 - mmengine - INFO - Epoch(train) [7][6000/7800] lr: 1.4474e-04 eta: 5:49:13 time: 0.7208 data_time: 0.0012 memory: 25074 loss: 0.0055 +2023/06/02 07:48:03 - mmengine - INFO - Epoch(train) [7][6100/7800] lr: 1.4474e-04 eta: 5:47:45 time: 0.7074 data_time: 0.0012 memory: 25074 loss: 0.0056 +2023/06/02 07:49:17 - mmengine - INFO - Exp name: convnext_small_4xb256_fake7m-lr4e-4_20230601_193207 +2023/06/02 07:49:17 - mmengine - INFO - Epoch(train) [7][6200/7800] lr: 1.4474e-04 eta: 5:46:18 time: 0.7069 data_time: 0.0013 memory: 25074 loss: 0.0038 +2023/06/02 07:50:34 - mmengine - INFO - Epoch(train) [7][6300/7800] lr: 1.4474e-04 eta: 5:44:52 time: 0.7342 data_time: 0.0014 memory: 25074 loss: 0.0063 +2023/06/02 07:51:46 - mmengine - INFO - Epoch(train) [7][6400/7800] lr: 1.4474e-04 eta: 5:43:24 time: 0.7185 data_time: 0.0021 memory: 25074 loss: 0.0078 +2023/06/02 07:53:00 - mmengine - INFO - Epoch(train) [7][6500/7800] lr: 1.4474e-04 eta: 5:41:57 time: 0.7051 data_time: 0.0015 memory: 25074 loss: 0.0062 +2023/06/02 07:54:15 - mmengine - INFO - Epoch(train) [7][6600/7800] lr: 1.4474e-04 eta: 5:40:30 time: 0.7160 data_time: 0.0013 memory: 25074 loss: 0.0063 +2023/06/02 07:55:31 - mmengine - INFO - Epoch(train) [7][6700/7800] lr: 1.4474e-04 eta: 5:39:03 time: 0.7321 data_time: 0.0013 memory: 25074 loss: 0.0080 +2023/06/02 07:56:44 - mmengine - INFO - Epoch(train) [7][6800/7800] lr: 1.4474e-04 eta: 5:37:36 time: 0.7774 data_time: 0.0017 memory: 25074 loss: 0.0085 +2023/06/02 07:58:01 - mmengine - INFO - Epoch(train) [7][6900/7800] lr: 1.4474e-04 eta: 5:36:10 time: 0.7956 data_time: 0.0015 memory: 25074 loss: 0.0053 +2023/06/02 07:59:17 - mmengine - INFO - Epoch(train) [7][7000/7800] lr: 1.4474e-04 eta: 5:34:44 time: 0.7151 data_time: 0.0014 memory: 25074 loss: 0.0082 +2023/06/02 08:00:32 - mmengine - INFO - Epoch(train) [7][7100/7800] lr: 1.4474e-04 eta: 5:33:17 time: 0.7039 data_time: 0.0014 memory: 25074 loss: 0.0052 +2023/06/02 08:01:46 - mmengine - INFO - Exp name: convnext_small_4xb256_fake7m-lr4e-4_20230601_193207 +2023/06/02 08:01:46 - mmengine - INFO - Epoch(train) [7][7200/7800] lr: 1.4474e-04 eta: 5:31:50 time: 0.9040 data_time: 0.0037 memory: 25074 loss: 0.0045 +2023/06/02 08:03:02 - mmengine - INFO - Epoch(train) [7][7300/7800] lr: 1.4474e-04 eta: 5:30:24 time: 0.7274 data_time: 0.0016 memory: 25074 loss: 0.0166 +2023/06/02 08:04:20 - mmengine - INFO - Epoch(train) [7][7400/7800] lr: 1.4474e-04 eta: 5:28:59 time: 0.7064 data_time: 0.0015 memory: 25074 loss: 0.0097 +2023/06/02 08:05:39 - mmengine - INFO - Epoch(train) [7][7500/7800] lr: 1.4474e-04 eta: 5:27:34 time: 0.7352 data_time: 0.0014 memory: 25074 loss: 0.0060 +2023/06/02 08:06:55 - mmengine - INFO - Epoch(train) [7][7600/7800] lr: 1.4474e-04 eta: 5:26:09 time: 0.7424 data_time: 0.0014 memory: 25074 loss: 0.0065 +2023/06/02 08:08:11 - mmengine - INFO - Epoch(train) [7][7700/7800] lr: 1.4474e-04 eta: 5:24:43 time: 0.7895 data_time: 0.0029 memory: 25074 loss: 0.0140 +2023/06/02 08:09:29 - mmengine - INFO - Exp name: convnext_small_4xb256_fake7m-lr4e-4_20230601_193207 +2023/06/02 08:09:29 - mmengine - INFO - Epoch(train) [7][7800/7800] lr: 1.4474e-04 eta: 5:23:18 time: 0.6711 data_time: 0.0013 memory: 25074 loss: 0.0079 +2023/06/02 08:09:29 - mmengine - INFO - Saving checkpoint at 7 epochs +2023/06/02 08:09:51 - mmengine - INFO - Epoch(val) [7][16/16] accuracy/top1: 99.9495 single-label/precision_classwise: [100.0, 0.0] single-label/recall_classwise: [99.94949340820312, 0.0] single-label/f1-score_classwise: [99.97473907470703, 0.0] data_time: 0.2005 time: 0.5844 +2023/06/02 08:11:14 - mmengine - INFO - Epoch(train) [8][ 100/7800] lr: 9.0382e-05 eta: 5:21:55 time: 0.7649 data_time: 0.0015 memory: 25074 loss: 0.0023 +2023/06/02 08:12:31 - mmengine - INFO - Epoch(train) [8][ 200/7800] lr: 9.0382e-05 eta: 5:20:29 time: 0.7621 data_time: 0.0015 memory: 25074 loss: 0.0030 +2023/06/02 08:13:49 - mmengine - INFO - Epoch(train) [8][ 300/7800] lr: 9.0382e-05 eta: 5:19:04 time: 0.7448 data_time: 0.0014 memory: 25074 loss: 0.0043 +2023/06/02 08:15:06 - mmengine - INFO - Exp name: convnext_small_4xb256_fake7m-lr4e-4_20230601_193207 +2023/06/02 08:15:06 - mmengine - INFO - Epoch(train) [8][ 400/7800] lr: 9.0382e-05 eta: 5:17:39 time: 0.7148 data_time: 0.0016 memory: 25074 loss: 0.0104 +2023/06/02 08:16:19 - mmengine - INFO - Epoch(train) [8][ 500/7800] lr: 9.0382e-05 eta: 5:16:12 time: 0.7642 data_time: 0.0025 memory: 25074 loss: 0.0066 +2023/06/02 08:17:33 - mmengine - INFO - Epoch(train) [8][ 600/7800] lr: 9.0382e-05 eta: 5:14:46 time: 0.7108 data_time: 0.0014 memory: 25074 loss: 0.0040 +2023/06/02 08:18:51 - mmengine - INFO - Epoch(train) [8][ 700/7800] lr: 9.0382e-05 eta: 5:13:21 time: 0.7742 data_time: 0.0015 memory: 25074 loss: 0.0043 +2023/06/02 08:20:10 - mmengine - INFO - Epoch(train) [8][ 800/7800] lr: 9.0382e-05 eta: 5:11:56 time: 0.7537 data_time: 0.0014 memory: 25074 loss: 0.0032 +2023/06/02 08:21:27 - mmengine - INFO - Epoch(train) [8][ 900/7800] lr: 9.0382e-05 eta: 5:10:31 time: 0.8996 data_time: 0.0021 memory: 25074 loss: 0.0043 +2023/06/02 08:22:41 - mmengine - INFO - Epoch(train) [8][1000/7800] lr: 9.0382e-05 eta: 5:09:05 time: 0.7374 data_time: 0.0014 memory: 25074 loss: 0.0067 +2023/06/02 08:24:00 - mmengine - INFO - Epoch(train) [8][1100/7800] lr: 9.0382e-05 eta: 5:07:41 time: 0.7287 data_time: 0.0012 memory: 25074 loss: 0.0130 +2023/06/02 08:25:16 - mmengine - INFO - Epoch(train) [8][1200/7800] lr: 9.0382e-05 eta: 5:06:15 time: 0.7369 data_time: 0.0015 memory: 25074 loss: 0.0107 +2023/06/02 08:26:34 - mmengine - INFO - Epoch(train) [8][1300/7800] lr: 9.0382e-05 eta: 5:04:50 time: 0.7150 data_time: 0.0013 memory: 25074 loss: 0.0108 +2023/06/02 08:27:48 - mmengine - INFO - Exp name: convnext_small_4xb256_fake7m-lr4e-4_20230601_193207 +2023/06/02 08:27:48 - mmengine - INFO - Epoch(train) [8][1400/7800] lr: 9.0382e-05 eta: 5:03:24 time: 0.7505 data_time: 0.0041 memory: 25074 loss: 0.0046 +2023/06/02 08:29:02 - mmengine - INFO - Epoch(train) [8][1500/7800] lr: 9.0382e-05 eta: 5:01:58 time: 0.7120 data_time: 0.0013 memory: 25074 loss: 0.0060 +2023/06/02 08:30:19 - mmengine - INFO - Epoch(train) [8][1600/7800] lr: 9.0382e-05 eta: 5:00:33 time: 0.7306 data_time: 0.0015 memory: 25074 loss: 0.0074 +2023/06/02 08:31:36 - mmengine - INFO - Epoch(train) [8][1700/7800] lr: 9.0382e-05 eta: 4:59:08 time: 0.7193 data_time: 0.0016 memory: 25074 loss: 0.0057 +2023/06/02 08:32:47 - mmengine - INFO - Epoch(train) [8][1800/7800] lr: 9.0382e-05 eta: 4:57:41 time: 0.7835 data_time: 0.0030 memory: 25074 loss: 0.0056 +2023/06/02 08:34:02 - mmengine - INFO - Epoch(train) [8][1900/7800] lr: 9.0382e-05 eta: 4:56:15 time: 0.7550 data_time: 0.0016 memory: 25074 loss: 0.0046 +2023/06/02 08:35:19 - mmengine - INFO - Epoch(train) [8][2000/7800] lr: 9.0382e-05 eta: 4:54:50 time: 0.7496 data_time: 0.0015 memory: 25074 loss: 0.0051 +2023/06/02 08:36:35 - mmengine - INFO - Epoch(train) [8][2100/7800] lr: 9.0382e-05 eta: 4:53:25 time: 0.7202 data_time: 0.0014 memory: 25074 loss: 0.0050 +2023/06/02 08:37:51 - mmengine - INFO - Epoch(train) [8][2200/7800] lr: 9.0382e-05 eta: 4:52:00 time: 0.7533 data_time: 0.0016 memory: 25074 loss: 0.0062 +2023/06/02 08:39:06 - mmengine - INFO - Epoch(train) [8][2300/7800] lr: 9.0382e-05 eta: 4:50:35 time: 0.7272 data_time: 0.0014 memory: 25074 loss: 0.0046 +2023/06/02 08:40:25 - mmengine - INFO - Exp name: convnext_small_4xb256_fake7m-lr4e-4_20230601_193207 +2023/06/02 08:40:25 - mmengine - INFO - Epoch(train) [8][2400/7800] lr: 9.0382e-05 eta: 4:49:11 time: 0.7921 data_time: 0.0016 memory: 25074 loss: 0.0067 +2023/06/02 08:41:41 - mmengine - INFO - Epoch(train) [8][2500/7800] lr: 9.0382e-05 eta: 4:47:46 time: 0.7397 data_time: 0.0014 memory: 25074 loss: 0.0018 +2023/06/02 08:42:55 - mmengine - INFO - Epoch(train) [8][2600/7800] lr: 9.0382e-05 eta: 4:46:20 time: 0.7160 data_time: 0.0013 memory: 25074 loss: 0.0054 +2023/06/02 08:44:11 - mmengine - INFO - Epoch(train) [8][2700/7800] lr: 9.0382e-05 eta: 4:44:55 time: 0.7304 data_time: 0.0015 memory: 25074 loss: 0.0055 +2023/06/02 08:45:27 - mmengine - INFO - Epoch(train) [8][2800/7800] lr: 9.0382e-05 eta: 4:43:30 time: 0.7393 data_time: 0.0015 memory: 25074 loss: 0.0061 +2023/06/02 08:46:45 - mmengine - INFO - Epoch(train) [8][2900/7800] lr: 9.0382e-05 eta: 4:42:06 time: 0.8159 data_time: 0.0016 memory: 25074 loss: 0.0031 +2023/06/02 08:48:01 - mmengine - INFO - Epoch(train) [8][3000/7800] lr: 9.0382e-05 eta: 4:40:41 time: 0.7586 data_time: 0.0014 memory: 25074 loss: 0.0055 +2023/06/02 08:49:20 - mmengine - INFO - Epoch(train) [8][3100/7800] lr: 9.0382e-05 eta: 4:39:17 time: 0.7385 data_time: 0.0013 memory: 25074 loss: 0.0079 +2023/06/02 08:50:34 - mmengine - INFO - Epoch(train) [8][3200/7800] lr: 9.0382e-05 eta: 4:37:51 time: 0.7191 data_time: 0.0014 memory: 25074 loss: 0.0073 +2023/06/02 08:51:48 - mmengine - INFO - Epoch(train) [8][3300/7800] lr: 9.0382e-05 eta: 4:36:26 time: 0.7039 data_time: 0.0013 memory: 25074 loss: 0.0096 +2023/06/02 08:52:59 - mmengine - INFO - Exp name: convnext_small_4xb256_fake7m-lr4e-4_20230601_193207 +2023/06/02 08:52:59 - mmengine - INFO - Epoch(train) [8][3400/7800] lr: 9.0382e-05 eta: 4:35:00 time: 0.7197 data_time: 0.0013 memory: 25074 loss: 0.0082 +2023/06/02 08:54:11 - mmengine - INFO - Epoch(train) [8][3500/7800] lr: 9.0382e-05 eta: 4:33:33 time: 0.7056 data_time: 0.0015 memory: 25074 loss: 0.0016 +2023/06/02 08:55:23 - mmengine - INFO - Epoch(train) [8][3600/7800] lr: 9.0382e-05 eta: 4:32:07 time: 0.7270 data_time: 0.0013 memory: 25074 loss: 0.0081 +2023/06/02 08:56:37 - mmengine - INFO - Epoch(train) [8][3700/7800] lr: 9.0382e-05 eta: 4:30:42 time: 0.7339 data_time: 0.0013 memory: 25074 loss: 0.0041 +2023/06/02 08:57:51 - mmengine - INFO - Epoch(train) [8][3800/7800] lr: 9.0382e-05 eta: 4:29:17 time: 0.7326 data_time: 0.0167 memory: 25074 loss: 0.0035 +2023/06/02 08:59:05 - mmengine - INFO - Epoch(train) [8][3900/7800] lr: 9.0382e-05 eta: 4:27:52 time: 0.7155 data_time: 0.0131 memory: 25074 loss: 0.0037 +2023/06/02 09:00:19 - mmengine - INFO - Epoch(train) [8][4000/7800] lr: 9.0382e-05 eta: 4:26:26 time: 0.7241 data_time: 0.0013 memory: 25074 loss: 0.0060 +2023/06/02 09:01:33 - mmengine - INFO - Epoch(train) [8][4100/7800] lr: 9.0382e-05 eta: 4:25:01 time: 0.7035 data_time: 0.0013 memory: 25074 loss: 0.0076 +2023/06/02 09:02:47 - mmengine - INFO - Epoch(train) [8][4200/7800] lr: 9.0382e-05 eta: 4:23:36 time: 0.7352 data_time: 0.0167 memory: 25074 loss: 0.0110 +2023/06/02 09:03:59 - mmengine - INFO - Epoch(train) [8][4300/7800] lr: 9.0382e-05 eta: 4:22:10 time: 0.7043 data_time: 0.0016 memory: 25074 loss: 0.0044 +2023/06/02 09:05:13 - mmengine - INFO - Exp name: convnext_small_4xb256_fake7m-lr4e-4_20230601_193207 +2023/06/02 09:05:13 - mmengine - INFO - Epoch(train) [8][4400/7800] lr: 9.0382e-05 eta: 4:20:45 time: 0.7339 data_time: 0.0013 memory: 25074 loss: 0.0039 +2023/06/02 09:06:27 - mmengine - INFO - Epoch(train) [8][4500/7800] lr: 9.0382e-05 eta: 4:19:20 time: 0.7037 data_time: 0.0013 memory: 25074 loss: 0.0057 +2023/06/02 09:07:40 - mmengine - INFO - Epoch(train) [8][4600/7800] lr: 9.0382e-05 eta: 4:17:55 time: 0.7101 data_time: 0.0017 memory: 25074 loss: 0.0076 +2023/06/02 09:08:52 - mmengine - INFO - Epoch(train) [8][4700/7800] lr: 9.0382e-05 eta: 4:16:29 time: 0.7043 data_time: 0.0014 memory: 25074 loss: 0.0038 +2023/06/02 09:10:04 - mmengine - INFO - Epoch(train) [8][4800/7800] lr: 9.0382e-05 eta: 4:15:04 time: 0.7907 data_time: 0.0014 memory: 25074 loss: 0.0073 +2023/06/02 09:11:17 - mmengine - INFO - Epoch(train) [8][4900/7800] lr: 9.0382e-05 eta: 4:13:39 time: 0.7490 data_time: 0.0014 memory: 25074 loss: 0.0033 +2023/06/02 09:12:31 - mmengine - INFO - Epoch(train) [8][5000/7800] lr: 9.0382e-05 eta: 4:12:14 time: 0.7490 data_time: 0.0012 memory: 25074 loss: 0.0069 +2023/06/02 09:13:44 - mmengine - INFO - Epoch(train) [8][5100/7800] lr: 9.0382e-05 eta: 4:10:49 time: 0.8165 data_time: 0.0014 memory: 25074 loss: 0.0108 +2023/06/02 09:14:59 - mmengine - INFO - Epoch(train) [8][5200/7800] lr: 9.0382e-05 eta: 4:09:24 time: 0.7553 data_time: 0.0014 memory: 25074 loss: 0.0103 +2023/06/02 09:16:13 - mmengine - INFO - Epoch(train) [8][5300/7800] lr: 9.0382e-05 eta: 4:08:00 time: 0.7752 data_time: 0.0013 memory: 25074 loss: 0.0027 +2023/06/02 09:17:29 - mmengine - INFO - Exp name: convnext_small_4xb256_fake7m-lr4e-4_20230601_193207 +2023/06/02 09:17:29 - mmengine - INFO - Epoch(train) [8][5400/7800] lr: 9.0382e-05 eta: 4:06:35 time: 0.7999 data_time: 0.0016 memory: 25074 loss: 0.0041 +2023/06/02 09:18:43 - mmengine - INFO - Epoch(train) [8][5500/7800] lr: 9.0382e-05 eta: 4:05:11 time: 0.7048 data_time: 0.0017 memory: 25074 loss: 0.0023 +2023/06/02 09:19:59 - mmengine - INFO - Epoch(train) [8][5600/7800] lr: 9.0382e-05 eta: 4:03:47 time: 0.7879 data_time: 0.0646 memory: 25074 loss: 0.0038 +2023/06/02 09:21:14 - mmengine - INFO - Epoch(train) [8][5700/7800] lr: 9.0382e-05 eta: 4:02:22 time: 0.7234 data_time: 0.0208 memory: 25074 loss: 0.0047 +2023/06/02 09:22:30 - mmengine - INFO - Epoch(train) [8][5800/7800] lr: 9.0382e-05 eta: 4:00:59 time: 0.7716 data_time: 0.0535 memory: 25074 loss: 0.0084 +2023/06/02 09:23:45 - mmengine - INFO - Epoch(train) [8][5900/7800] lr: 9.0382e-05 eta: 3:59:34 time: 0.7763 data_time: 0.0731 memory: 25074 loss: 0.0053 +2023/06/02 09:25:00 - mmengine - INFO - Epoch(train) [8][6000/7800] lr: 9.0382e-05 eta: 3:58:10 time: 0.7316 data_time: 0.0063 memory: 25074 loss: 0.0033 +2023/06/02 09:26:15 - mmengine - INFO - Epoch(train) [8][6100/7800] lr: 9.0382e-05 eta: 3:56:46 time: 0.7293 data_time: 0.0266 memory: 25074 loss: 0.0037 +2023/06/02 09:27:30 - mmengine - INFO - Epoch(train) [8][6200/7800] lr: 9.0382e-05 eta: 3:55:22 time: 0.7237 data_time: 0.0220 memory: 25074 loss: 0.0074 +2023/06/02 09:28:45 - mmengine - INFO - Epoch(train) [8][6300/7800] lr: 9.0382e-05 eta: 3:53:58 time: 0.7157 data_time: 0.0112 memory: 25074 loss: 0.0051 +2023/06/02 09:30:00 - mmengine - INFO - Exp name: convnext_small_4xb256_fake7m-lr4e-4_20230601_193207 +2023/06/02 09:30:00 - mmengine - INFO - Epoch(train) [8][6400/7800] lr: 9.0382e-05 eta: 3:52:33 time: 0.7175 data_time: 0.0143 memory: 25074 loss: 0.0064 +2023/06/02 09:31:14 - mmengine - INFO - Epoch(train) [8][6500/7800] lr: 9.0382e-05 eta: 3:51:09 time: 0.7317 data_time: 0.0296 memory: 25074 loss: 0.0033 +2023/06/02 09:32:28 - mmengine - INFO - Epoch(train) [8][6600/7800] lr: 9.0382e-05 eta: 3:49:45 time: 0.7188 data_time: 0.0014 memory: 25074 loss: 0.0050 +2023/06/02 09:33:40 - mmengine - INFO - Epoch(train) [8][6700/7800] lr: 9.0382e-05 eta: 3:48:20 time: 0.7051 data_time: 0.0014 memory: 25074 loss: 0.0110 +2023/06/02 09:34:54 - mmengine - INFO - Epoch(train) [8][6800/7800] lr: 9.0382e-05 eta: 3:46:56 time: 0.7302 data_time: 0.0183 memory: 25074 loss: 0.0056 +2023/06/02 09:36:07 - mmengine - INFO - Epoch(train) [8][6900/7800] lr: 9.0382e-05 eta: 3:45:32 time: 0.7976 data_time: 0.0952 memory: 25074 loss: 0.0090 +2023/06/02 09:37:22 - mmengine - INFO - Epoch(train) [8][7000/7800] lr: 9.0382e-05 eta: 3:44:08 time: 0.7058 data_time: 0.0030 memory: 25074 loss: 0.0030 +2023/06/02 09:38:34 - mmengine - INFO - Epoch(train) [8][7100/7800] lr: 9.0382e-05 eta: 3:42:43 time: 0.7666 data_time: 0.0647 memory: 25074 loss: 0.0067 +2023/06/02 09:39:49 - mmengine - INFO - Epoch(train) [8][7200/7800] lr: 9.0382e-05 eta: 3:41:19 time: 0.7343 data_time: 0.0318 memory: 25074 loss: 0.0023 +2023/06/02 09:41:02 - mmengine - INFO - Epoch(train) [8][7300/7800] lr: 9.0382e-05 eta: 3:39:55 time: 0.7442 data_time: 0.0420 memory: 25074 loss: 0.0068 +2023/06/02 09:42:16 - mmengine - INFO - Exp name: convnext_small_4xb256_fake7m-lr4e-4_20230601_193207 +2023/06/02 09:42:16 - mmengine - INFO - Epoch(train) [8][7400/7800] lr: 9.0382e-05 eta: 3:38:31 time: 0.7542 data_time: 0.0516 memory: 25074 loss: 0.0074 +2023/06/02 09:43:30 - mmengine - INFO - Epoch(train) [8][7500/7800] lr: 9.0382e-05 eta: 3:37:07 time: 0.7354 data_time: 0.0319 memory: 25074 loss: 0.0039 +2023/06/02 09:44:44 - mmengine - INFO - Epoch(train) [8][7600/7800] lr: 9.0382e-05 eta: 3:35:43 time: 0.7340 data_time: 0.0313 memory: 25074 loss: 0.0027 +2023/06/02 09:45:57 - mmengine - INFO - Epoch(train) [8][7700/7800] lr: 9.0382e-05 eta: 3:34:19 time: 0.7514 data_time: 0.0490 memory: 25074 loss: 0.0020 +2023/06/02 09:47:12 - mmengine - INFO - Exp name: convnext_small_4xb256_fake7m-lr4e-4_20230601_193207 +2023/06/02 09:47:12 - mmengine - INFO - Epoch(train) [8][7800/7800] lr: 9.0382e-05 eta: 3:32:55 time: 0.6614 data_time: 0.0038 memory: 25074 loss: 0.0039 +2023/06/02 09:47:12 - mmengine - INFO - Saving checkpoint at 8 epochs +2023/06/02 09:47:31 - mmengine - INFO - Epoch(val) [8][16/16] accuracy/top1: 99.9684 single-label/precision_classwise: [100.0, 0.0] single-label/recall_classwise: [99.96842956542969, 0.0] single-label/f1-score_classwise: [99.98420715332031, 0.0] data_time: 0.2067 time: 0.5926 +2023/06/02 09:48:52 - mmengine - INFO - Epoch(train) [9][ 100/7800] lr: 4.7242e-05 eta: 3:31:33 time: 0.7392 data_time: 0.0013 memory: 25074 loss: 0.0036 +2023/06/02 09:50:07 - mmengine - INFO - Epoch(train) [9][ 200/7800] lr: 4.7242e-05 eta: 3:30:09 time: 0.7699 data_time: 0.0014 memory: 25074 loss: 0.0064 +2023/06/02 09:51:22 - mmengine - INFO - Epoch(train) [9][ 300/7800] lr: 4.7242e-05 eta: 3:28:46 time: 0.8053 data_time: 0.0014 memory: 25074 loss: 0.0016 +2023/06/02 09:52:37 - mmengine - INFO - Epoch(train) [9][ 400/7800] lr: 4.7242e-05 eta: 3:27:22 time: 0.7831 data_time: 0.0014 memory: 25074 loss: 0.0043 +2023/06/02 09:53:50 - mmengine - INFO - Epoch(train) [9][ 500/7800] lr: 4.7242e-05 eta: 3:25:58 time: 0.7325 data_time: 0.0013 memory: 25074 loss: 0.0062 +2023/06/02 09:55:05 - mmengine - INFO - Exp name: convnext_small_4xb256_fake7m-lr4e-4_20230601_193207 +2023/06/02 09:55:05 - mmengine - INFO - Epoch(train) [9][ 600/7800] lr: 4.7242e-05 eta: 3:24:35 time: 0.7303 data_time: 0.0012 memory: 25074 loss: 0.0019 +2023/06/02 09:56:18 - mmengine - INFO - Epoch(train) [9][ 700/7800] lr: 4.7242e-05 eta: 3:23:11 time: 0.7313 data_time: 0.0015 memory: 25074 loss: 0.0027 +2023/06/02 09:57:31 - mmengine - INFO - Epoch(train) [9][ 800/7800] lr: 4.7242e-05 eta: 3:21:47 time: 0.7365 data_time: 0.0015 memory: 25074 loss: 0.0041 +2023/06/02 09:58:43 - mmengine - INFO - Epoch(train) [9][ 900/7800] lr: 4.7242e-05 eta: 3:20:23 time: 0.7074 data_time: 0.0015 memory: 25074 loss: 0.0030 +2023/06/02 09:59:55 - mmengine - INFO - Epoch(train) [9][1000/7800] lr: 4.7242e-05 eta: 3:18:59 time: 0.7042 data_time: 0.0012 memory: 25074 loss: 0.0050 +2023/06/02 10:01:07 - mmengine - INFO - Epoch(train) [9][1100/7800] lr: 4.7242e-05 eta: 3:17:35 time: 0.7203 data_time: 0.0015 memory: 25074 loss: 0.0040 +2023/06/02 10:02:19 - mmengine - INFO - Epoch(train) [9][1200/7800] lr: 4.7242e-05 eta: 3:16:11 time: 0.7256 data_time: 0.0014 memory: 25074 loss: 0.0027 +2023/06/02 10:03:33 - mmengine - INFO - Epoch(train) [9][1300/7800] lr: 4.7242e-05 eta: 3:14:47 time: 0.7170 data_time: 0.0015 memory: 25074 loss: 0.0023 +2023/06/02 10:04:46 - mmengine - INFO - Epoch(train) [9][1400/7800] lr: 4.7242e-05 eta: 3:13:24 time: 0.7808 data_time: 0.0017 memory: 25074 loss: 0.0019 +2023/06/02 10:06:00 - mmengine - INFO - Epoch(train) [9][1500/7800] lr: 4.7242e-05 eta: 3:12:00 time: 0.7450 data_time: 0.0014 memory: 25074 loss: 0.0047 +2023/06/02 10:07:13 - mmengine - INFO - Exp name: convnext_small_4xb256_fake7m-lr4e-4_20230601_193207 +2023/06/02 10:07:13 - mmengine - INFO - Epoch(train) [9][1600/7800] lr: 4.7242e-05 eta: 3:10:37 time: 0.7035 data_time: 0.0016 memory: 25074 loss: 0.0043 +2023/06/02 10:08:27 - mmengine - INFO - Epoch(train) [9][1700/7800] lr: 4.7242e-05 eta: 3:09:13 time: 0.7848 data_time: 0.0014 memory: 25074 loss: 0.0053 +2023/06/02 10:09:41 - mmengine - INFO - Epoch(train) [9][1800/7800] lr: 4.7242e-05 eta: 3:07:50 time: 0.7080 data_time: 0.0015 memory: 25074 loss: 0.0027 +2023/06/02 10:10:54 - mmengine - INFO - Epoch(train) [9][1900/7800] lr: 4.7242e-05 eta: 3:06:26 time: 0.7791 data_time: 0.0013 memory: 25074 loss: 0.0075 +2023/06/02 10:12:06 - mmengine - INFO - Epoch(train) [9][2000/7800] lr: 4.7242e-05 eta: 3:05:03 time: 0.7049 data_time: 0.0016 memory: 25074 loss: 0.0035 +2023/06/02 10:13:18 - mmengine - INFO - Epoch(train) [9][2100/7800] lr: 4.7242e-05 eta: 3:03:39 time: 0.7286 data_time: 0.0016 memory: 25074 loss: 0.0033 +2023/06/02 10:14:33 - mmengine - INFO - Epoch(train) [9][2200/7800] lr: 4.7242e-05 eta: 3:02:16 time: 0.7335 data_time: 0.0013 memory: 25074 loss: 0.0033 +2023/06/02 10:15:47 - mmengine - INFO - Epoch(train) [9][2300/7800] lr: 4.7242e-05 eta: 3:00:53 time: 0.7037 data_time: 0.0013 memory: 25074 loss: 0.0057 +2023/06/02 10:17:01 - mmengine - INFO - Epoch(train) [9][2400/7800] lr: 4.7242e-05 eta: 2:59:30 time: 0.7045 data_time: 0.0014 memory: 25074 loss: 0.0052 +2023/06/02 10:18:17 - mmengine - INFO - Epoch(train) [9][2500/7800] lr: 4.7242e-05 eta: 2:58:07 time: 0.7237 data_time: 0.0014 memory: 25074 loss: 0.0022 +2023/06/02 10:19:34 - mmengine - INFO - Exp name: convnext_small_4xb256_fake7m-lr4e-4_20230601_193207 +2023/06/02 10:19:34 - mmengine - INFO - Epoch(train) [9][2600/7800] lr: 4.7242e-05 eta: 2:56:44 time: 0.7207 data_time: 0.0014 memory: 25074 loss: 0.0024 +2023/06/02 10:20:48 - mmengine - INFO - Epoch(train) [9][2700/7800] lr: 4.7242e-05 eta: 2:55:21 time: 0.7151 data_time: 0.0014 memory: 25074 loss: 0.0038 +2023/06/02 10:22:01 - mmengine - INFO - Epoch(train) [9][2800/7800] lr: 4.7242e-05 eta: 2:53:58 time: 0.7044 data_time: 0.0017 memory: 25074 loss: 0.0054 +2023/06/02 10:23:15 - mmengine - INFO - Epoch(train) [9][2900/7800] lr: 4.7242e-05 eta: 2:52:35 time: 0.7301 data_time: 0.0015 memory: 25074 loss: 0.0026 +2023/06/02 10:24:28 - mmengine - INFO - Epoch(train) [9][3000/7800] lr: 4.7242e-05 eta: 2:51:12 time: 0.7076 data_time: 0.0014 memory: 25074 loss: 0.0038 +2023/06/02 10:25:42 - mmengine - INFO - Epoch(train) [9][3100/7800] lr: 4.7242e-05 eta: 2:49:49 time: 0.8134 data_time: 0.0015 memory: 25074 loss: 0.0023 +2023/06/02 10:26:56 - mmengine - INFO - Epoch(train) [9][3200/7800] lr: 4.7242e-05 eta: 2:48:26 time: 0.7075 data_time: 0.0014 memory: 25074 loss: 0.0012 +2023/06/02 10:28:08 - mmengine - INFO - Epoch(train) [9][3300/7800] lr: 4.7242e-05 eta: 2:47:03 time: 0.7072 data_time: 0.0017 memory: 25074 loss: 0.0032 +2023/06/02 10:29:22 - mmengine - INFO - Epoch(train) [9][3400/7800] lr: 4.7242e-05 eta: 2:45:40 time: 0.7887 data_time: 0.0014 memory: 25074 loss: 0.0061 +2023/06/02 10:30:36 - mmengine - INFO - Epoch(train) [9][3500/7800] lr: 4.7242e-05 eta: 2:44:17 time: 0.7290 data_time: 0.0014 memory: 25074 loss: 0.0025 +2023/06/02 10:31:50 - mmengine - INFO - Exp name: convnext_small_4xb256_fake7m-lr4e-4_20230601_193207 +2023/06/02 10:31:50 - mmengine - INFO - Epoch(train) [9][3600/7800] lr: 4.7242e-05 eta: 2:42:54 time: 0.7047 data_time: 0.0015 memory: 25074 loss: 0.0050 +2023/06/02 10:33:04 - mmengine - INFO - Epoch(train) [9][3700/7800] lr: 4.7242e-05 eta: 2:41:31 time: 0.7327 data_time: 0.0013 memory: 25074 loss: 0.0030 +2023/06/02 10:34:18 - mmengine - INFO - Epoch(train) [9][3800/7800] lr: 4.7242e-05 eta: 2:40:09 time: 0.7578 data_time: 0.0013 memory: 25074 loss: 0.0041 +2023/06/02 10:35:32 - mmengine - INFO - Epoch(train) [9][3900/7800] lr: 4.7242e-05 eta: 2:38:46 time: 0.7433 data_time: 0.0015 memory: 25074 loss: 0.0022 +2023/06/02 10:36:48 - mmengine - INFO - Epoch(train) [9][4000/7800] lr: 4.7242e-05 eta: 2:37:23 time: 0.7208 data_time: 0.0017 memory: 25074 loss: 0.0047 +2023/06/02 10:38:03 - mmengine - INFO - Epoch(train) [9][4100/7800] lr: 4.7242e-05 eta: 2:36:01 time: 0.7343 data_time: 0.0014 memory: 25074 loss: 0.0054 +2023/06/02 10:39:17 - mmengine - INFO - Epoch(train) [9][4200/7800] lr: 4.7242e-05 eta: 2:34:38 time: 0.7817 data_time: 0.0014 memory: 25074 loss: 0.0041 +2023/06/02 10:40:32 - mmengine - INFO - Epoch(train) [9][4300/7800] lr: 4.7242e-05 eta: 2:33:16 time: 0.8030 data_time: 0.0014 memory: 25074 loss: 0.0022 +2023/06/02 10:41:48 - mmengine - INFO - Epoch(train) [9][4400/7800] lr: 4.7242e-05 eta: 2:31:53 time: 0.7102 data_time: 0.0014 memory: 25074 loss: 0.0057 +2023/06/02 10:43:05 - mmengine - INFO - Epoch(train) [9][4500/7800] lr: 4.7242e-05 eta: 2:30:31 time: 0.7785 data_time: 0.0018 memory: 25074 loss: 0.0027 +2023/06/02 10:44:21 - mmengine - INFO - Exp name: convnext_small_4xb256_fake7m-lr4e-4_20230601_193207 +2023/06/02 10:44:21 - mmengine - INFO - Epoch(train) [9][4600/7800] lr: 4.7242e-05 eta: 2:29:09 time: 0.7824 data_time: 0.0015 memory: 25074 loss: 0.0013 +2023/06/02 10:45:38 - mmengine - INFO - Epoch(train) [9][4700/7800] lr: 4.7242e-05 eta: 2:27:47 time: 0.7595 data_time: 0.0014 memory: 25074 loss: 0.0026 +2023/06/02 10:46:54 - mmengine - INFO - Epoch(train) [9][4800/7800] lr: 4.7242e-05 eta: 2:26:25 time: 0.7793 data_time: 0.0015 memory: 25074 loss: 0.0024 +2023/06/02 10:48:09 - mmengine - INFO - Epoch(train) [9][4900/7800] lr: 4.7242e-05 eta: 2:25:03 time: 0.7778 data_time: 0.0015 memory: 25074 loss: 0.0012 +2023/06/02 10:49:27 - mmengine - INFO - Epoch(train) [9][5000/7800] lr: 4.7242e-05 eta: 2:23:41 time: 0.8095 data_time: 0.0015 memory: 25074 loss: 0.0014 +2023/06/02 10:50:43 - mmengine - INFO - Epoch(train) [9][5100/7800] lr: 4.7242e-05 eta: 2:22:18 time: 0.7228 data_time: 0.0014 memory: 25074 loss: 0.0024 +2023/06/02 10:51:59 - mmengine - INFO - Epoch(train) [9][5200/7800] lr: 4.7242e-05 eta: 2:20:56 time: 0.7303 data_time: 0.0018 memory: 25074 loss: 0.0035 +2023/06/02 10:53:16 - mmengine - INFO - Epoch(train) [9][5300/7800] lr: 4.7242e-05 eta: 2:19:34 time: 0.7654 data_time: 0.0014 memory: 25074 loss: 0.0018 +2023/06/02 10:54:31 - mmengine - INFO - Epoch(train) [9][5400/7800] lr: 4.7242e-05 eta: 2:18:12 time: 0.7296 data_time: 0.0015 memory: 25074 loss: 0.0038 +2023/06/02 10:55:48 - mmengine - INFO - Epoch(train) [9][5500/7800] lr: 4.7242e-05 eta: 2:16:50 time: 0.7504 data_time: 0.0013 memory: 25074 loss: 0.0027 +2023/06/02 10:57:02 - mmengine - INFO - Exp name: convnext_small_4xb256_fake7m-lr4e-4_20230601_193207 +2023/06/02 10:57:02 - mmengine - INFO - Epoch(train) [9][5600/7800] lr: 4.7242e-05 eta: 2:15:28 time: 0.7779 data_time: 0.0013 memory: 25074 loss: 0.0044 +2023/06/02 10:58:17 - mmengine - INFO - Epoch(train) [9][5700/7800] lr: 4.7242e-05 eta: 2:14:06 time: 0.7397 data_time: 0.0015 memory: 25074 loss: 0.0013 +2023/06/02 10:59:33 - mmengine - INFO - Epoch(train) [9][5800/7800] lr: 4.7242e-05 eta: 2:12:44 time: 0.7474 data_time: 0.0017 memory: 25074 loss: 0.0071 +2023/06/02 11:00:48 - mmengine - INFO - Epoch(train) [9][5900/7800] lr: 4.7242e-05 eta: 2:11:21 time: 0.7402 data_time: 0.0013 memory: 25074 loss: 0.0025 +2023/06/02 11:02:01 - mmengine - INFO - Epoch(train) [9][6000/7800] lr: 4.7242e-05 eta: 2:09:59 time: 0.7041 data_time: 0.0013 memory: 25074 loss: 0.0037 +2023/06/02 11:03:14 - mmengine - INFO - Epoch(train) [9][6100/7800] lr: 4.7242e-05 eta: 2:08:37 time: 0.7304 data_time: 0.0014 memory: 25074 loss: 0.0023 +2023/06/02 11:04:28 - mmengine - INFO - Epoch(train) [9][6200/7800] lr: 4.7242e-05 eta: 2:07:14 time: 0.7496 data_time: 0.0419 memory: 25074 loss: 0.0030 +2023/06/02 11:05:47 - mmengine - INFO - Epoch(train) [9][6300/7800] lr: 4.7242e-05 eta: 2:05:53 time: 0.7571 data_time: 0.0539 memory: 25074 loss: 0.0065 +2023/06/02 11:07:02 - mmengine - INFO - Epoch(train) [9][6400/7800] lr: 4.7242e-05 eta: 2:04:31 time: 0.7949 data_time: 0.0827 memory: 25074 loss: 0.0013 +2023/06/02 11:08:16 - mmengine - INFO - Epoch(train) [9][6500/7800] lr: 4.7242e-05 eta: 2:03:09 time: 0.7289 data_time: 0.0260 memory: 25074 loss: 0.0047 +2023/06/02 11:09:29 - mmengine - INFO - Exp name: convnext_small_4xb256_fake7m-lr4e-4_20230601_193207 +2023/06/02 11:09:29 - mmengine - INFO - Epoch(train) [9][6600/7800] lr: 4.7242e-05 eta: 2:01:46 time: 0.7207 data_time: 0.0014 memory: 25074 loss: 0.0014 +2023/06/02 11:10:42 - mmengine - INFO - Epoch(train) [9][6700/7800] lr: 4.7242e-05 eta: 2:00:24 time: 0.7669 data_time: 0.0645 memory: 25074 loss: 0.0018 +2023/06/02 11:11:56 - mmengine - INFO - Epoch(train) [9][6800/7800] lr: 4.7242e-05 eta: 1:59:02 time: 0.7626 data_time: 0.0585 memory: 25074 loss: 0.0042 +2023/06/02 11:13:12 - mmengine - INFO - Epoch(train) [9][6900/7800] lr: 4.7242e-05 eta: 1:57:40 time: 0.7868 data_time: 0.0844 memory: 25074 loss: 0.0009 +2023/06/02 11:14:28 - mmengine - INFO - Epoch(train) [9][7000/7800] lr: 4.7242e-05 eta: 1:56:18 time: 0.7711 data_time: 0.0635 memory: 25074 loss: 0.0031 +2023/06/02 11:15:41 - mmengine - INFO - Epoch(train) [9][7100/7800] lr: 4.7242e-05 eta: 1:54:56 time: 0.7045 data_time: 0.0013 memory: 25074 loss: 0.0028 +2023/06/02 11:16:55 - mmengine - INFO - Epoch(train) [9][7200/7800] lr: 4.7242e-05 eta: 1:53:34 time: 0.7519 data_time: 0.0261 memory: 25074 loss: 0.0032 +2023/06/02 11:18:09 - mmengine - INFO - Epoch(train) [9][7300/7800] lr: 4.7242e-05 eta: 1:52:12 time: 0.7501 data_time: 0.0404 memory: 25074 loss: 0.0059 +2023/06/02 11:19:24 - mmengine - INFO - Epoch(train) [9][7400/7800] lr: 4.7242e-05 eta: 1:50:50 time: 0.7557 data_time: 0.0409 memory: 25074 loss: 0.0035 +2023/06/02 11:20:40 - mmengine - INFO - Epoch(train) [9][7500/7800] lr: 4.7242e-05 eta: 1:49:29 time: 0.7449 data_time: 0.0380 memory: 25074 loss: 0.0016 +2023/06/02 11:21:56 - mmengine - INFO - Exp name: convnext_small_4xb256_fake7m-lr4e-4_20230601_193207 +2023/06/02 11:21:56 - mmengine - INFO - Epoch(train) [9][7600/7800] lr: 4.7242e-05 eta: 1:48:07 time: 0.7677 data_time: 0.0546 memory: 25074 loss: 0.0021 +2023/06/02 11:23:12 - mmengine - INFO - Epoch(train) [9][7700/7800] lr: 4.7242e-05 eta: 1:46:45 time: 0.7734 data_time: 0.0566 memory: 25074 loss: 0.0044 +2023/06/02 11:24:26 - mmengine - INFO - Exp name: convnext_small_4xb256_fake7m-lr4e-4_20230601_193207 +2023/06/02 11:24:26 - mmengine - INFO - Epoch(train) [9][7800/7800] lr: 4.7242e-05 eta: 1:45:23 time: 0.7081 data_time: 0.0487 memory: 25074 loss: 0.0038 +2023/06/02 11:24:26 - mmengine - INFO - Saving checkpoint at 9 epochs +2023/06/02 11:24:45 - mmengine - INFO - Epoch(val) [9][16/16] accuracy/top1: 99.9747 single-label/precision_classwise: [100.0, 0.0] single-label/recall_classwise: [99.9747543334961, 0.0] single-label/f1-score_classwise: [99.98737335205078, 0.0] data_time: 0.2134 time: 0.5989 +2023/06/02 11:26:09 - mmengine - INFO - Epoch(train) [10][ 100/7800] lr: 1.9544e-05 eta: 1:44:03 time: 0.7957 data_time: 0.0015 memory: 25074 loss: 0.0037 +2023/06/02 11:27:25 - mmengine - INFO - Epoch(train) [10][ 200/7800] lr: 1.9544e-05 eta: 1:42:41 time: 0.7895 data_time: 0.0014 memory: 25074 loss: 0.0018 +2023/06/02 11:28:40 - mmengine - INFO - Epoch(train) [10][ 300/7800] lr: 1.9544e-05 eta: 1:41:19 time: 0.7389 data_time: 0.0015 memory: 25074 loss: 0.0021 +2023/06/02 11:29:54 - mmengine - INFO - Epoch(train) [10][ 400/7800] lr: 1.9544e-05 eta: 1:39:58 time: 0.7773 data_time: 0.0356 memory: 25074 loss: 0.0028 +2023/06/02 11:31:11 - mmengine - INFO - Epoch(train) [10][ 500/7800] lr: 1.9544e-05 eta: 1:38:36 time: 0.7678 data_time: 0.0516 memory: 25074 loss: 0.0027 +2023/06/02 11:32:25 - mmengine - INFO - Epoch(train) [10][ 600/7800] lr: 1.9544e-05 eta: 1:37:14 time: 0.7584 data_time: 0.0569 memory: 25074 loss: 0.0006 +2023/06/02 11:33:41 - mmengine - INFO - Epoch(train) [10][ 700/7800] lr: 1.9544e-05 eta: 1:35:53 time: 0.7109 data_time: 0.0085 memory: 25074 loss: 0.0004 +2023/06/02 11:34:55 - mmengine - INFO - Exp name: convnext_small_4xb256_fake7m-lr4e-4_20230601_193207 +2023/06/02 11:34:55 - mmengine - INFO - Epoch(train) [10][ 800/7800] lr: 1.9544e-05 eta: 1:34:31 time: 0.7505 data_time: 0.0382 memory: 25074 loss: 0.0020 +2023/06/02 11:36:09 - mmengine - INFO - Epoch(train) [10][ 900/7800] lr: 1.9544e-05 eta: 1:33:09 time: 0.7037 data_time: 0.0013 memory: 25074 loss: 0.0040 +2023/06/02 11:37:22 - mmengine - INFO - Epoch(train) [10][1000/7800] lr: 1.9544e-05 eta: 1:31:47 time: 0.7170 data_time: 0.0016 memory: 25074 loss: 0.0019 +2023/06/02 11:38:35 - mmengine - INFO - Epoch(train) [10][1100/7800] lr: 1.9544e-05 eta: 1:30:26 time: 0.7402 data_time: 0.0376 memory: 25074 loss: 0.0011 +2023/06/02 11:39:50 - mmengine - INFO - Epoch(train) [10][1200/7800] lr: 1.9544e-05 eta: 1:29:04 time: 0.7538 data_time: 0.0367 memory: 25074 loss: 0.0014 +2023/06/02 11:41:06 - mmengine - INFO - Epoch(train) [10][1300/7800] lr: 1.9544e-05 eta: 1:27:43 time: 0.7104 data_time: 0.0081 memory: 25074 loss: 0.0042 +2023/06/02 11:42:21 - mmengine - INFO - Epoch(train) [10][1400/7800] lr: 1.9544e-05 eta: 1:26:21 time: 0.7429 data_time: 0.0276 memory: 25074 loss: 0.0007 +2023/06/02 11:43:36 - mmengine - INFO - Epoch(train) [10][1500/7800] lr: 1.9544e-05 eta: 1:25:00 time: 0.7376 data_time: 0.0203 memory: 25074 loss: 0.0026 +2023/06/02 11:44:51 - mmengine - INFO - Epoch(train) [10][1600/7800] lr: 1.9544e-05 eta: 1:23:38 time: 0.7761 data_time: 0.0719 memory: 25074 loss: 0.0043 +2023/06/02 11:46:06 - mmengine - INFO - Epoch(train) [10][1700/7800] lr: 1.9544e-05 eta: 1:22:17 time: 0.7320 data_time: 0.0301 memory: 25074 loss: 0.0021 +2023/06/02 11:47:21 - mmengine - INFO - Exp name: convnext_small_4xb256_fake7m-lr4e-4_20230601_193207 +2023/06/02 11:47:21 - mmengine - INFO - Epoch(train) [10][1800/7800] lr: 1.9544e-05 eta: 1:20:55 time: 0.7585 data_time: 0.0560 memory: 25074 loss: 0.0013 +2023/06/02 11:48:37 - mmengine - INFO - Epoch(train) [10][1900/7800] lr: 1.9544e-05 eta: 1:19:34 time: 0.7960 data_time: 0.0789 memory: 25074 loss: 0.0033 +2023/06/02 11:49:54 - mmengine - INFO - Epoch(train) [10][2000/7800] lr: 1.9544e-05 eta: 1:18:13 time: 0.7630 data_time: 0.0606 memory: 25074 loss: 0.0031 +2023/06/02 11:51:09 - mmengine - INFO - Epoch(train) [10][2100/7800] lr: 1.9544e-05 eta: 1:16:51 time: 0.7575 data_time: 0.0373 memory: 25074 loss: 0.0020 +2023/06/02 11:52:27 - mmengine - INFO - Epoch(train) [10][2200/7800] lr: 1.9544e-05 eta: 1:15:30 time: 0.7565 data_time: 0.0544 memory: 25074 loss: 0.0017 +2023/06/02 11:53:44 - mmengine - INFO - Epoch(train) [10][2300/7800] lr: 1.9544e-05 eta: 1:14:09 time: 0.7713 data_time: 0.0577 memory: 25074 loss: 0.0025 +2023/06/02 11:55:01 - mmengine - INFO - Epoch(train) [10][2400/7800] lr: 1.9544e-05 eta: 1:12:48 time: 0.7408 data_time: 0.0379 memory: 25074 loss: 0.0026 +2023/06/02 11:56:18 - mmengine - INFO - Epoch(train) [10][2500/7800] lr: 1.9544e-05 eta: 1:11:27 time: 0.8322 data_time: 0.1215 memory: 25074 loss: 0.0025 +2023/06/02 11:57:37 - mmengine - INFO - Epoch(train) [10][2600/7800] lr: 1.9544e-05 eta: 1:10:06 time: 0.7813 data_time: 0.0793 memory: 25074 loss: 0.0046 +2023/06/02 11:58:53 - mmengine - INFO - Epoch(train) [10][2700/7800] lr: 1.9544e-05 eta: 1:08:44 time: 0.7358 data_time: 0.0160 memory: 25074 loss: 0.0006 +2023/06/02 12:00:09 - mmengine - INFO - Exp name: convnext_small_4xb256_fake7m-lr4e-4_20230601_193207 +2023/06/02 12:00:09 - mmengine - INFO - Epoch(train) [10][2800/7800] lr: 1.9544e-05 eta: 1:07:23 time: 0.7702 data_time: 0.0662 memory: 25074 loss: 0.0026 +2023/06/02 12:01:25 - mmengine - INFO - Epoch(train) [10][2900/7800] lr: 1.9544e-05 eta: 1:06:02 time: 0.7679 data_time: 0.0572 memory: 25074 loss: 0.0029 +2023/06/02 12:02:42 - mmengine - INFO - Epoch(train) [10][3000/7800] lr: 1.9544e-05 eta: 1:04:41 time: 0.7612 data_time: 0.0588 memory: 25074 loss: 0.0033 +2023/06/02 12:04:00 - mmengine - INFO - Epoch(train) [10][3100/7800] lr: 1.9544e-05 eta: 1:03:20 time: 0.8150 data_time: 0.1026 memory: 25074 loss: 0.0013 +2023/06/02 12:05:18 - mmengine - INFO - Epoch(train) [10][3200/7800] lr: 1.9544e-05 eta: 1:01:59 time: 0.7916 data_time: 0.0894 memory: 25074 loss: 0.0026 +2023/06/02 12:06:35 - mmengine - INFO - Epoch(train) [10][3300/7800] lr: 1.9544e-05 eta: 1:00:38 time: 0.8216 data_time: 0.1191 memory: 25074 loss: 0.0019 +2023/06/02 12:07:51 - mmengine - INFO - Epoch(train) [10][3400/7800] lr: 1.9544e-05 eta: 0:59:17 time: 0.7418 data_time: 0.0399 memory: 25074 loss: 0.0024 +2023/06/02 12:09:08 - mmengine - INFO - Epoch(train) [10][3500/7800] lr: 1.9544e-05 eta: 0:57:56 time: 0.7923 data_time: 0.0881 memory: 25074 loss: 0.0004 +2023/06/02 12:10:24 - mmengine - INFO - Epoch(train) [10][3600/7800] lr: 1.9544e-05 eta: 0:56:34 time: 0.7810 data_time: 0.0782 memory: 25074 loss: 0.0022 +2023/06/02 12:11:39 - mmengine - INFO - Epoch(train) [10][3700/7800] lr: 1.9544e-05 eta: 0:55:13 time: 0.7199 data_time: 0.0028 memory: 25074 loss: 0.0020 +2023/06/02 12:12:55 - mmengine - INFO - Exp name: convnext_small_4xb256_fake7m-lr4e-4_20230601_193207 +2023/06/02 12:12:55 - mmengine - INFO - Epoch(train) [10][3800/7800] lr: 1.9544e-05 eta: 0:53:52 time: 0.7285 data_time: 0.0263 memory: 25074 loss: 0.0041 +2023/06/02 12:14:11 - mmengine - INFO - Epoch(train) [10][3900/7800] lr: 1.9544e-05 eta: 0:52:31 time: 0.8004 data_time: 0.0972 memory: 25074 loss: 0.0021 +2023/06/02 12:15:27 - mmengine - INFO - Epoch(train) [10][4000/7800] lr: 1.9544e-05 eta: 0:51:10 time: 0.7129 data_time: 0.0107 memory: 25074 loss: 0.0014 +2023/06/02 12:16:42 - mmengine - INFO - Epoch(train) [10][4100/7800] lr: 1.9544e-05 eta: 0:49:49 time: 0.7433 data_time: 0.0411 memory: 25074 loss: 0.0006 +2023/06/02 12:17:57 - mmengine - INFO - Epoch(train) [10][4200/7800] lr: 1.9544e-05 eta: 0:48:28 time: 0.7423 data_time: 0.0392 memory: 25074 loss: 0.0011 +2023/06/02 12:19:14 - mmengine - INFO - Epoch(train) [10][4300/7800] lr: 1.9544e-05 eta: 0:47:07 time: 0.7847 data_time: 0.0818 memory: 25074 loss: 0.0018 +2023/06/02 12:20:28 - mmengine - INFO - Epoch(train) [10][4400/7800] lr: 1.9544e-05 eta: 0:45:46 time: 0.7334 data_time: 0.0296 memory: 25074 loss: 0.0020 +2023/06/02 12:21:44 - mmengine - INFO - Epoch(train) [10][4500/7800] lr: 1.9544e-05 eta: 0:44:25 time: 0.7362 data_time: 0.0333 memory: 25074 loss: 0.0043 +2023/06/02 12:23:00 - mmengine - INFO - Epoch(train) [10][4600/7800] lr: 1.9544e-05 eta: 0:43:04 time: 0.7836 data_time: 0.0812 memory: 25074 loss: 0.0014 +2023/06/02 12:24:15 - mmengine - INFO - Epoch(train) [10][4700/7800] lr: 1.9544e-05 eta: 0:41:43 time: 0.7360 data_time: 0.0321 memory: 25074 loss: 0.0009 +2023/06/02 12:25:31 - mmengine - INFO - Exp name: convnext_small_4xb256_fake7m-lr4e-4_20230601_193207 +2023/06/02 12:25:31 - mmengine - INFO - Epoch(train) [10][4800/7800] lr: 1.9544e-05 eta: 0:40:22 time: 0.7759 data_time: 0.0725 memory: 25074 loss: 0.0019 +2023/06/02 12:26:47 - mmengine - INFO - Epoch(train) [10][4900/7800] lr: 1.9544e-05 eta: 0:39:01 time: 0.7111 data_time: 0.0085 memory: 25074 loss: 0.0051 +2023/06/02 12:28:02 - mmengine - INFO - Epoch(train) [10][5000/7800] lr: 1.9544e-05 eta: 0:37:40 time: 0.7879 data_time: 0.0854 memory: 25074 loss: 0.0021 +2023/06/02 12:29:19 - mmengine - INFO - Epoch(train) [10][5100/7800] lr: 1.9544e-05 eta: 0:36:19 time: 0.7608 data_time: 0.0570 memory: 25074 loss: 0.0019 +2023/06/02 12:30:34 - mmengine - INFO - Epoch(train) [10][5200/7800] lr: 1.9544e-05 eta: 0:34:58 time: 0.7179 data_time: 0.0149 memory: 25074 loss: 0.0033 +2023/06/02 12:31:52 - mmengine - INFO - Epoch(train) [10][5300/7800] lr: 1.9544e-05 eta: 0:33:37 time: 0.7500 data_time: 0.0480 memory: 25074 loss: 0.0014 +2023/06/02 12:33:08 - mmengine - INFO - Epoch(train) [10][5400/7800] lr: 1.9544e-05 eta: 0:32:17 time: 0.7263 data_time: 0.0238 memory: 25074 loss: 0.0015 +2023/06/02 12:34:25 - mmengine - INFO - Epoch(train) [10][5500/7800] lr: 1.9544e-05 eta: 0:30:56 time: 0.7382 data_time: 0.0358 memory: 25074 loss: 0.0047 +2023/06/02 12:35:42 - mmengine - INFO - Epoch(train) [10][5600/7800] lr: 1.9544e-05 eta: 0:29:35 time: 0.7676 data_time: 0.0588 memory: 25074 loss: 0.0012 +2023/06/02 12:36:59 - mmengine - INFO - Epoch(train) [10][5700/7800] lr: 1.9544e-05 eta: 0:28:14 time: 0.7253 data_time: 0.0217 memory: 25074 loss: 0.0004 +2023/06/02 12:38:16 - mmengine - INFO - Exp name: convnext_small_4xb256_fake7m-lr4e-4_20230601_193207 +2023/06/02 12:38:16 - mmengine - INFO - Epoch(train) [10][5800/7800] lr: 1.9544e-05 eta: 0:26:53 time: 0.7731 data_time: 0.0633 memory: 25074 loss: 0.0035 +2023/06/02 12:39:37 - mmengine - INFO - Epoch(train) [10][5900/7800] lr: 1.9544e-05 eta: 0:25:33 time: 0.9304 data_time: 0.2107 memory: 25074 loss: 0.0013 +2023/06/02 12:41:05 - mmengine - INFO - Epoch(train) [10][6000/7800] lr: 1.9544e-05 eta: 0:24:12 time: 0.8475 data_time: 0.1305 memory: 25074 loss: 0.0021 +2023/06/02 12:42:34 - mmengine - INFO - Epoch(train) [10][6100/7800] lr: 1.9544e-05 eta: 0:22:52 time: 0.9318 data_time: 0.2094 memory: 25074 loss: 0.0021 +2023/06/02 12:43:59 - mmengine - INFO - Epoch(train) [10][6200/7800] lr: 1.9544e-05 eta: 0:21:31 time: 0.8268 data_time: 0.0996 memory: 25074 loss: 0.0015 +2023/06/02 12:45:16 - mmengine - INFO - Epoch(train) [10][6300/7800] lr: 1.9544e-05 eta: 0:20:10 time: 0.7307 data_time: 0.0278 memory: 25074 loss: 0.0020 +2023/06/02 12:46:44 - mmengine - INFO - Epoch(train) [10][6400/7800] lr: 1.9544e-05 eta: 0:18:50 time: 0.8450 data_time: 0.1008 memory: 25074 loss: 0.0020 +2023/06/02 12:48:10 - mmengine - INFO - Epoch(train) [10][6500/7800] lr: 1.9544e-05 eta: 0:17:29 time: 0.8716 data_time: 0.1534 memory: 25074 loss: 0.0009 +2023/06/02 12:49:37 - mmengine - INFO - Epoch(train) [10][6600/7800] lr: 1.9544e-05 eta: 0:16:08 time: 0.8197 data_time: 0.0954 memory: 25074 loss: 0.0008 +2023/06/02 12:50:54 - mmengine - INFO - Epoch(train) [10][6700/7800] lr: 1.9544e-05 eta: 0:14:48 time: 0.7565 data_time: 0.0529 memory: 25074 loss: 0.0034 +2023/06/02 12:52:12 - mmengine - INFO - Exp name: convnext_small_4xb256_fake7m-lr4e-4_20230601_193207 +2023/06/02 12:52:12 - mmengine - INFO - Epoch(train) [10][6800/7800] lr: 1.9544e-05 eta: 0:13:27 time: 0.7693 data_time: 0.0416 memory: 25074 loss: 0.0025 +2023/06/02 12:53:28 - mmengine - INFO - Epoch(train) [10][6900/7800] lr: 1.9544e-05 eta: 0:12:06 time: 0.7205 data_time: 0.0184 memory: 25074 loss: 0.0011 +2023/06/02 12:54:45 - mmengine - INFO - Epoch(train) [10][7000/7800] lr: 1.9544e-05 eta: 0:10:45 time: 0.8058 data_time: 0.1022 memory: 25074 loss: 0.0039 +2023/06/02 12:56:10 - mmengine - INFO - Epoch(train) [10][7100/7800] lr: 1.9544e-05 eta: 0:09:25 time: 0.8737 data_time: 0.1591 memory: 25074 loss: 0.0012 +2023/06/02 12:57:37 - mmengine - INFO - Epoch(train) [10][7200/7800] lr: 1.9544e-05 eta: 0:08:04 time: 0.9030 data_time: 0.1643 memory: 25074 loss: 0.0019 +2023/06/02 12:59:03 - mmengine - INFO - Epoch(train) [10][7300/7800] lr: 1.9544e-05 eta: 0:06:43 time: 0.7804 data_time: 0.0746 memory: 25074 loss: 0.0015 +2023/06/02 13:00:19 - mmengine - INFO - Epoch(train) [10][7400/7800] lr: 1.9544e-05 eta: 0:05:22 time: 0.7538 data_time: 0.0314 memory: 25074 loss: 0.0009 +2023/06/02 13:01:36 - mmengine - INFO - Epoch(train) [10][7500/7800] lr: 1.9544e-05 eta: 0:04:02 time: 0.7621 data_time: 0.0595 memory: 25074 loss: 0.0024 +2023/06/02 13:02:54 - mmengine - INFO - Epoch(train) [10][7600/7800] lr: 1.9544e-05 eta: 0:02:41 time: 0.9070 data_time: 0.1417 memory: 25074 loss: 0.0040 +2023/06/02 13:04:21 - mmengine - INFO - Epoch(train) [10][7700/7800] lr: 1.9544e-05 eta: 0:01:20 time: 0.8927 data_time: 0.1705 memory: 25074 loss: 0.0023 +2023/06/02 13:05:47 - mmengine - INFO - Exp name: convnext_small_4xb256_fake7m-lr4e-4_20230601_193207 +2023/06/02 13:05:47 - mmengine - INFO - Epoch(train) [10][7800/7800] lr: 1.9544e-05 eta: 0:00:00 time: 0.8018 data_time: 0.1295 memory: 25074 loss: 0.0029 +2023/06/02 13:05:47 - mmengine - INFO - Saving checkpoint at 10 epochs +2023/06/02 13:06:10 - mmengine - INFO - Epoch(val) [10][16/16] accuracy/top1: 99.9874 single-label/precision_classwise: [100.0, 0.0] single-label/recall_classwise: [99.98737335205078, 0.0] single-label/f1-score_classwise: [99.99369049072266, 0.0] data_time: 0.3145 time: 0.7018 diff --git a/convnext_small_4xb256_fake7m-lr4e-4/20230601_193207/vis_data/20230601_193207.json b/convnext_small_4xb256_fake7m-lr4e-4/20230601_193207/vis_data/20230601_193207.json new file mode 100644 index 0000000000000000000000000000000000000000..be42ea66000a05cca1a5afe3e0bd445ce348ebc2 --- /dev/null +++ b/convnext_small_4xb256_fake7m-lr4e-4/20230601_193207/vis_data/20230601_193207.json @@ -0,0 +1,790 @@ +{"lr": 0.0004, "data_time": 2.6907129287719727, "loss": 0.5509223461151123, "time": 4.028603577613831, "epoch": 1, "memory": 25071, "step": 100} +{"lr": 0.0004, "data_time": 3.903346085548401, "loss": 0.5122920751571656, "time": 4.696592187881469, "epoch": 1, "memory": 25071, "step": 200} +{"lr": 0.0004, "data_time": 3.581833267211914, "loss": 0.4802056074142456, "time": 4.289185667037964, "epoch": 1, "memory": 25071, "step": 300} +{"lr": 0.0004, "data_time": 3.3729814291000366, "loss": 0.47626489102840425, "time": 4.466727089881897, "epoch": 1, "memory": 25071, "step": 400} +{"lr": 0.0004, "data_time": 3.152855086326599, "loss": 0.46141456365585326, "time": 4.480378150939941, "epoch": 1, "memory": 25071, "step": 500} +{"lr": 0.0004, "data_time": 3.7332908153533935, "loss": 0.45464659929275514, "time": 4.440839076042176, "epoch": 1, "memory": 25071, "step": 600} +{"lr": 0.0004, "data_time": 3.7476879358291626, "loss": 0.4438972115516663, "time": 4.455186796188355, "epoch": 1, "memory": 25071, "step": 700} +{"lr": 0.0004, "data_time": 2.997472071647644, "loss": 0.46163980662822723, "time": 3.7071758985519407, "epoch": 1, "memory": 25071, "step": 800} +{"lr": 0.0004, "data_time": 3.3110134601593018, "loss": 0.4082850754261017, "time": 4.019241809844971, "epoch": 1, "memory": 25071, "step": 900} +{"lr": 0.0004, "data_time": 0.0013388633728027345, "loss": 0.4357276916503906, "time": 0.7083222389221191, "epoch": 1, "memory": 25071, "step": 1000} +{"lr": 0.0004, "data_time": 0.04599974155426025, "loss": 0.393719083070755, "time": 0.7553951740264893, "epoch": 1, "memory": 25071, "step": 1100} +{"lr": 0.0004, "data_time": 0.0013998746871948242, "loss": 0.40593518018722535, "time": 0.7048080205917359, "epoch": 1, "memory": 25071, "step": 1200} +{"lr": 0.0004, "data_time": 0.0012827157974243165, "loss": 0.4099356472492218, "time": 0.7289187192916871, "epoch": 1, "memory": 25071, "step": 1300} +{"lr": 0.0004, "data_time": 0.0013564825057983398, "loss": 0.4108930230140686, "time": 0.8211391925811767, "epoch": 1, "memory": 25071, "step": 1400} +{"lr": 0.0004, "data_time": 0.0013655424118041992, "loss": 0.39929743111133575, "time": 0.7272112131118774, "epoch": 1, "memory": 25071, "step": 1500} +{"lr": 0.0004, "data_time": 0.18983705043792726, "loss": 0.39867955446243286, "time": 0.8959766387939453, "epoch": 1, "memory": 25071, "step": 1600} +{"lr": 0.0004, "data_time": 0.0012964248657226563, "loss": 0.3719035655260086, "time": 0.854664397239685, "epoch": 1, "memory": 25071, "step": 1700} +{"lr": 0.0004, "data_time": 0.0014933109283447265, "loss": 0.3636783301830292, "time": 0.8932012319564819, "epoch": 1, "memory": 25071, "step": 1800} +{"lr": 0.0004, "data_time": 0.0014470577239990234, "loss": 0.37819368243217466, "time": 0.7350067853927612, "epoch": 1, "memory": 25071, "step": 1900} +{"lr": 0.0004, "data_time": 0.0013707876205444336, "loss": 0.3699472784996033, "time": 0.7687848091125489, "epoch": 1, "memory": 25071, "step": 2000} +{"lr": 0.0004, "data_time": 0.001299571990966797, "loss": 0.3759961068630219, "time": 0.792644739151001, "epoch": 1, "memory": 25071, "step": 2100} +{"lr": 0.0004, "data_time": 0.0014440059661865235, "loss": 0.37775363922119143, "time": 0.7185564517974854, "epoch": 1, "memory": 25071, "step": 2200} +{"lr": 0.0004, "data_time": 0.0014520883560180664, "loss": 0.3431844234466553, "time": 0.7634948253631592, "epoch": 1, "memory": 25071, "step": 2300} +{"lr": 0.0004, "data_time": 0.0012256860733032226, "loss": 0.3327546685934067, "time": 0.7808838129043579, "epoch": 1, "memory": 25071, "step": 2400} +{"lr": 0.0004, "data_time": 0.0012555122375488281, "loss": 0.3289307564496994, "time": 0.8116376638412476, "epoch": 1, "memory": 25071, "step": 2500} +{"lr": 0.0004, "data_time": 0.0012950897216796875, "loss": 0.33132781088352203, "time": 1.0313302040100099, "epoch": 1, "memory": 25071, "step": 2600} +{"lr": 0.0004, "data_time": 0.001563262939453125, "loss": 0.34159435629844664, "time": 1.0104379415512086, "epoch": 1, "memory": 25071, "step": 2700} +{"lr": 0.0004, "data_time": 0.0014277935028076173, "loss": 0.3161256194114685, "time": 0.9495459079742432, "epoch": 1, "memory": 25071, "step": 2800} +{"lr": 0.0004, "data_time": 0.001340341567993164, "loss": 0.3131725758314133, "time": 0.7670013427734375, "epoch": 1, "memory": 25071, "step": 2900} +{"lr": 0.0004, "data_time": 0.0013190746307373048, "loss": 0.33015253841876985, "time": 1.0442669153213502, "epoch": 1, "memory": 25071, "step": 3000} +{"lr": 0.0004, "data_time": 0.027217388153076172, "loss": 0.3198246330022812, "time": 0.8309589385986328, "epoch": 1, "memory": 25071, "step": 3100} +{"lr": 0.0004, "data_time": 0.0014668941497802735, "loss": 0.29642116725444795, "time": 0.7057172536849976, "epoch": 1, "memory": 25071, "step": 3200} +{"lr": 0.0004, "data_time": 0.001323866844177246, "loss": 0.3204584240913391, "time": 0.719047999382019, "epoch": 1, "memory": 25071, "step": 3300} +{"lr": 0.0004, "data_time": 0.0013828277587890625, "loss": 0.3120553225278854, "time": 0.7054277658462524, "epoch": 1, "memory": 25071, "step": 3400} +{"lr": 0.0004, "data_time": 0.0016828298568725586, "loss": 0.3116033345460892, "time": 0.7053396701812744, "epoch": 1, "memory": 25071, "step": 3500} +{"lr": 0.0004, "data_time": 0.0012962818145751953, "loss": 0.2981396973133087, "time": 0.7053417921066284, "epoch": 1, "memory": 25071, "step": 3600} +{"lr": 0.0004, "data_time": 0.0013994693756103516, "loss": 0.3141171634197235, "time": 0.7148031711578369, "epoch": 1, "memory": 25071, "step": 3700} +{"lr": 0.0004, "data_time": 0.001499032974243164, "loss": 0.29559786021709444, "time": 0.7061149120330811, "epoch": 1, "memory": 25071, "step": 3800} +{"lr": 0.0004, "data_time": 0.0014240741729736328, "loss": 0.29358674883842467, "time": 0.7244684934616089, "epoch": 1, "memory": 25071, "step": 3900} +{"lr": 0.0004, "data_time": 0.001511836051940918, "loss": 0.25467085242271426, "time": 0.7070822238922119, "epoch": 1, "memory": 25071, "step": 4000} +{"lr": 0.0004, "data_time": 0.0014264106750488282, "loss": 0.2919668465852737, "time": 0.7236125946044922, "epoch": 1, "memory": 25071, "step": 4100} +{"lr": 0.0004, "data_time": 0.001343989372253418, "loss": 0.2884243577718735, "time": 0.7079657316207886, "epoch": 1, "memory": 25071, "step": 4200} +{"lr": 0.0004, "data_time": 0.0014359235763549804, "loss": 0.2842044696211815, "time": 0.7134523868560791, "epoch": 1, "memory": 25071, "step": 4300} +{"lr": 0.0004, "data_time": 0.0015362739562988282, "loss": 0.2771912842988968, "time": 0.7043992519378662, "epoch": 1, "memory": 25071, "step": 4400} +{"lr": 0.0004, "data_time": 0.0013992071151733398, "loss": 0.2899408534169197, "time": 0.7200767517089843, "epoch": 1, "memory": 25071, "step": 4500} +{"lr": 0.0004, "data_time": 0.0013010025024414063, "loss": 0.26940833032131195, "time": 0.7055598020553588, "epoch": 1, "memory": 25071, "step": 4600} +{"lr": 0.0004, "data_time": 0.0014023542404174804, "loss": 0.2595094949007034, "time": 0.7550057649612427, "epoch": 1, "memory": 25071, "step": 4700} +{"lr": 0.0004, "data_time": 0.0012409687042236328, "loss": 0.26932478696107864, "time": 0.7812519311904907, "epoch": 1, "memory": 25071, "step": 4800} +{"lr": 0.0004, "data_time": 0.0014226198196411132, "loss": 0.2865415155887604, "time": 0.718199610710144, "epoch": 1, "memory": 25071, "step": 4900} +{"lr": 0.0004, "data_time": 0.0013959169387817382, "loss": 0.26076784282922744, "time": 0.7069357872009278, "epoch": 1, "memory": 25071, "step": 5000} +{"lr": 0.0004, "data_time": 0.0012188196182250977, "loss": 0.27880939841270447, "time": 0.712481164932251, "epoch": 1, "memory": 25071, "step": 5100} +{"lr": 0.0004, "data_time": 0.001408553123474121, "loss": 0.26313952952623365, "time": 0.7054989099502563, "epoch": 1, "memory": 25071, "step": 5200} +{"lr": 0.0004, "data_time": 0.0013321876525878907, "loss": 0.257980714738369, "time": 0.7458206415176392, "epoch": 1, "memory": 25071, "step": 5300} +{"lr": 0.0004, "data_time": 0.0014032602310180664, "loss": 0.23871318995952606, "time": 0.7410083770751953, "epoch": 1, "memory": 25071, "step": 5400} +{"lr": 0.0004, "data_time": 0.0012554407119750976, "loss": 0.25315780490636824, "time": 0.715627908706665, "epoch": 1, "memory": 25071, "step": 5500} +{"lr": 0.0004, "data_time": 0.0015192270278930665, "loss": 0.22020687609910966, "time": 0.7893864393234253, "epoch": 1, "memory": 25071, "step": 5600} +{"lr": 0.0004, "data_time": 0.0013898849487304688, "loss": 0.25799562931060793, "time": 0.7222212553024292, "epoch": 1, "memory": 25071, "step": 5700} +{"lr": 0.0004, "data_time": 0.0014517545700073243, "loss": 0.26193687617778777, "time": 0.705207109451294, "epoch": 1, "memory": 25071, "step": 5800} +{"lr": 0.0004, "data_time": 0.0014998435974121094, "loss": 0.23978979438543319, "time": 0.7064939975738526, "epoch": 1, "memory": 25071, "step": 5900} +{"lr": 0.0004, "data_time": 0.0018176555633544922, "loss": 0.27151983976364136, "time": 0.7038760662078858, "epoch": 1, "memory": 25071, "step": 6000} +{"lr": 0.0004, "data_time": 0.0015011310577392578, "loss": 0.23080928325653077, "time": 0.7212326526641846, "epoch": 1, "memory": 25071, "step": 6100} +{"lr": 0.0004, "data_time": 0.0012937307357788086, "loss": 0.24490288645029068, "time": 0.720430064201355, "epoch": 1, "memory": 25071, "step": 6200} +{"lr": 0.0004, "data_time": 0.0011920690536499023, "loss": 0.23940146416425706, "time": 0.7056973457336426, "epoch": 1, "memory": 25071, "step": 6300} +{"lr": 0.0004, "data_time": 0.001397109031677246, "loss": 0.2349616140127182, "time": 0.7072972536087037, "epoch": 1, "memory": 25071, "step": 6400} +{"lr": 0.0004, "data_time": 0.0014444112777709961, "loss": 0.21422109603881836, "time": 0.7065572261810302, "epoch": 1, "memory": 25071, "step": 6500} +{"lr": 0.0004, "data_time": 0.0013858795166015625, "loss": 0.24266654700040818, "time": 0.7030148029327392, "epoch": 1, "memory": 25071, "step": 6600} +{"lr": 0.0004, "data_time": 0.0014795541763305664, "loss": 0.23327862322330475, "time": 0.7081075668334961, "epoch": 1, "memory": 25071, "step": 6700} +{"lr": 0.0004, "data_time": 0.0014479637145996093, "loss": 0.23950548470020294, "time": 0.7077049016952515, "epoch": 1, "memory": 25071, "step": 6800} +{"lr": 0.0004, "data_time": 0.0014442205429077148, "loss": 0.2277327612042427, "time": 0.7036959648132324, "epoch": 1, "memory": 25071, "step": 6900} +{"lr": 0.0004, "data_time": 0.0014896869659423828, "loss": 0.2067130535840988, "time": 0.7160272836685181, "epoch": 1, "memory": 25071, "step": 7000} +{"lr": 0.0004, "data_time": 0.0013399362564086915, "loss": 0.2089599072933197, "time": 0.7053474426269531, "epoch": 1, "memory": 25071, "step": 7100} +{"lr": 0.0004, "data_time": 0.0013405799865722657, "loss": 0.21773237138986587, "time": 0.7126405477523804, "epoch": 1, "memory": 25071, "step": 7200} +{"lr": 0.0004, "data_time": 0.0014854192733764649, "loss": 0.21485509276390075, "time": 0.7069435834884643, "epoch": 1, "memory": 25071, "step": 7300} +{"lr": 0.0004, "data_time": 0.001581096649169922, "loss": 0.21211579144001008, "time": 0.7383034706115723, "epoch": 1, "memory": 25071, "step": 7400} +{"lr": 0.0004, "data_time": 0.0014016866683959962, "loss": 0.21462307274341583, "time": 0.7065046787261963, "epoch": 1, "memory": 25071, "step": 7500} +{"lr": 0.0004, "data_time": 0.0016673803329467773, "loss": 0.19861916899681092, "time": 0.7191998958587646, "epoch": 1, "memory": 25071, "step": 7600} +{"lr": 0.0004, "data_time": 0.0014283895492553712, "loss": 0.22524501979351044, "time": 0.7050751447677612, "epoch": 1, "memory": 25071, "step": 7700} +{"lr": 0.0004, "data_time": 0.0010645389556884766, "loss": 0.2027382418513298, "time": 0.6594531059265136, "epoch": 1, "memory": 25071, "step": 7800} +{"accuracy/top1": 82.74729919433594, "data_time": 0.26189135015010834, "time": 0.7810759395360947, "step": 1} +{"lr": 0.00039045602067755495, "data_time": 0.001395416259765625, "loss": 0.1985473230481148, "time": 0.7228825330734253, "epoch": 2, "memory": 25074, "step": 7900} +{"lr": 0.00039045602067755495, "data_time": 0.00175323486328125, "loss": 0.19168051928281785, "time": 0.7411270856857299, "epoch": 2, "memory": 25074, "step": 8000} +{"lr": 0.00039045602067755495, "data_time": 0.0011951208114624023, "loss": 0.19834990948438644, "time": 0.722902488708496, "epoch": 2, "memory": 25074, "step": 8100} +{"lr": 0.00039045602067755495, "data_time": 0.0012741327285766602, "loss": 0.19371471256017686, "time": 0.7558099746704101, "epoch": 2, "memory": 25074, "step": 8200} +{"lr": 0.00039045602067755495, "data_time": 0.0013913631439208985, "loss": 0.1958928033709526, "time": 0.7217024803161621, "epoch": 2, "memory": 25074, "step": 8300} +{"lr": 0.00039045602067755495, "data_time": 0.0012528419494628907, "loss": 0.18579413443803788, "time": 0.8182249307632447, "epoch": 2, "memory": 25074, "step": 8400} +{"lr": 0.00039045602067755495, "data_time": 0.001584482192993164, "loss": 0.20487104803323747, "time": 0.7238365411758423, "epoch": 2, "memory": 25074, "step": 8500} +{"lr": 0.00039045602067755495, "data_time": 0.01669943332672119, "loss": 0.19487895220518112, "time": 0.7289236545562744, "epoch": 2, "memory": 25074, "step": 8600} +{"lr": 0.00039045602067755495, "data_time": 0.0015350580215454102, "loss": 0.17102978229522706, "time": 0.7226077556610108, "epoch": 2, "memory": 25074, "step": 8700} +{"lr": 0.00039045602067755495, "data_time": 0.0017715930938720704, "loss": 0.19943539649248124, "time": 0.7214926242828369, "epoch": 2, "memory": 25074, "step": 8800} +{"lr": 0.00039045602067755495, "data_time": 0.0017403364181518555, "loss": 0.19120719283819199, "time": 0.739957618713379, "epoch": 2, "memory": 25074, "step": 8900} +{"lr": 0.00039045602067755495, "data_time": 0.0018882274627685547, "loss": 0.1867499127984047, "time": 0.7092678785324097, "epoch": 2, "memory": 25074, "step": 9000} +{"lr": 0.00039045602067755495, "data_time": 0.0019953250885009766, "loss": 0.16988205015659333, "time": 0.7357144355773926, "epoch": 2, "memory": 25074, "step": 9100} +{"lr": 0.00039045602067755495, "data_time": 0.05246288776397705, "loss": 0.18630445897579193, "time": 0.7638375520706177, "epoch": 2, "memory": 25074, "step": 9200} +{"lr": 0.00039045602067755495, "data_time": 0.046308159828186035, "loss": 0.18053615242242813, "time": 0.7948039054870606, "epoch": 2, "memory": 25074, "step": 9300} +{"lr": 0.00039045602067755495, "data_time": 0.05054788589477539, "loss": 0.17141434997320176, "time": 0.7714882373809815, "epoch": 2, "memory": 25074, "step": 9400} +{"lr": 0.00039045602067755495, "data_time": 0.05095982551574707, "loss": 0.20186308026313782, "time": 0.7825620412826538, "epoch": 2, "memory": 25074, "step": 9500} +{"lr": 0.00039045602067755495, "data_time": 0.07320964336395264, "loss": 0.15854080840945245, "time": 0.7893943071365357, "epoch": 2, "memory": 25074, "step": 9600} +{"lr": 0.00039045602067755495, "data_time": 0.00156402587890625, "loss": 0.17121621072292328, "time": 0.7229015350341796, "epoch": 2, "memory": 25074, "step": 9700} +{"lr": 0.00039045602067755495, "data_time": 0.05058057308197021, "loss": 0.17450278401374816, "time": 0.7707611083984375, "epoch": 2, "memory": 25074, "step": 9800} +{"lr": 0.00039045602067755495, "data_time": 0.0055065155029296875, "loss": 0.17037517130374907, "time": 0.7292756557464599, "epoch": 2, "memory": 25074, "step": 9900} +{"lr": 0.00039045602067755495, "data_time": 0.0013831138610839843, "loss": 0.19259472638368608, "time": 0.7221044778823853, "epoch": 2, "memory": 25074, "step": 10000} +{"lr": 0.00039045602067755495, "data_time": 0.0013656139373779297, "loss": 0.15048861056566237, "time": 0.7273980379104614, "epoch": 2, "memory": 25074, "step": 10100} +{"lr": 0.00039045602067755495, "data_time": 0.0013884544372558595, "loss": 0.16968381106853486, "time": 0.7262316465377807, "epoch": 2, "memory": 25074, "step": 10200} +{"lr": 0.00039045602067755495, "data_time": 0.0018139362335205078, "loss": 0.15536620318889618, "time": 0.7153719186782836, "epoch": 2, "memory": 25074, "step": 10300} +{"lr": 0.00039045602067755495, "data_time": 0.024681353569030763, "loss": 0.1719048336148262, "time": 0.7603081464767456, "epoch": 2, "memory": 25074, "step": 10400} +{"lr": 0.00039045602067755495, "data_time": 0.0014130353927612304, "loss": 0.169435116648674, "time": 0.7366525411605835, "epoch": 2, "memory": 25074, "step": 10500} +{"lr": 0.00039045602067755495, "data_time": 0.001836705207824707, "loss": 0.15519988983869554, "time": 0.734037184715271, "epoch": 2, "memory": 25074, "step": 10600} +{"lr": 0.00039045602067755495, "data_time": 0.001296854019165039, "loss": 0.15888091921806335, "time": 0.7190788507461547, "epoch": 2, "memory": 25074, "step": 10700} +{"lr": 0.00039045602067755495, "data_time": 0.04676048755645752, "loss": 0.13778819739818574, "time": 0.7508063077926636, "epoch": 2, "memory": 25074, "step": 10800} +{"lr": 0.00039045602067755495, "data_time": 0.0017165422439575195, "loss": 0.15494614839553833, "time": 0.7291338205337524, "epoch": 2, "memory": 25074, "step": 10900} +{"lr": 0.00039045602067755495, "data_time": 0.0015732526779174804, "loss": 0.1438566505908966, "time": 0.7853857040405273, "epoch": 2, "memory": 25074, "step": 11000} +{"lr": 0.00039045602067755495, "data_time": 0.0023783445358276367, "loss": 0.1316233515739441, "time": 0.7321438550949096, "epoch": 2, "memory": 25074, "step": 11100} +{"lr": 0.00039045602067755495, "data_time": 0.0015875339508056641, "loss": 0.14384893029928209, "time": 0.7056066751480102, "epoch": 2, "memory": 25074, "step": 11200} +{"lr": 0.00039045602067755495, "data_time": 0.0020194053649902344, "loss": 0.14245496317744255, "time": 0.7379435539245606, "epoch": 2, "memory": 25074, "step": 11300} +{"lr": 0.00039045602067755495, "data_time": 0.0017960786819458008, "loss": 0.1622574359178543, "time": 0.7063355207443237, "epoch": 2, "memory": 25074, "step": 11400} +{"lr": 0.00039045602067755495, "data_time": 0.0014561891555786132, "loss": 0.13409009650349618, "time": 0.7040745735168457, "epoch": 2, "memory": 25074, "step": 11500} +{"lr": 0.00039045602067755495, "data_time": 0.0016580820083618164, "loss": 0.13272096514701842, "time": 0.7397542715072631, "epoch": 2, "memory": 25074, "step": 11600} +{"lr": 0.00039045602067755495, "data_time": 0.001663970947265625, "loss": 0.14436589553952217, "time": 0.733658742904663, "epoch": 2, "memory": 25074, "step": 11700} +{"lr": 0.00039045602067755495, "data_time": 0.0017907142639160157, "loss": 0.14573181942105293, "time": 0.7077654123306274, "epoch": 2, "memory": 25074, "step": 11800} +{"lr": 0.00039045602067755495, "data_time": 0.002051091194152832, "loss": 0.1403046302497387, "time": 0.7531724214553833, "epoch": 2, "memory": 25074, "step": 11900} +{"lr": 0.00039045602067755495, "data_time": 0.0019072294235229492, "loss": 0.14200572595000266, "time": 0.7084569692611694, "epoch": 2, "memory": 25074, "step": 12000} +{"lr": 0.00039045602067755495, "data_time": 0.001704263687133789, "loss": 0.14508214965462685, "time": 0.70661940574646, "epoch": 2, "memory": 25074, "step": 12100} +{"lr": 0.00039045602067755495, "data_time": 0.0018909454345703125, "loss": 0.14885776340961457, "time": 0.7115938425064087, "epoch": 2, "memory": 25074, "step": 12200} +{"lr": 0.00039045602067755495, "data_time": 0.0021780967712402345, "loss": 0.13284775018692016, "time": 0.7098190546035766, "epoch": 2, "memory": 25074, "step": 12300} +{"lr": 0.00039045602067755495, "data_time": 0.0016410112380981444, "loss": 0.14073834717273712, "time": 0.7054736852645874, "epoch": 2, "memory": 25074, "step": 12400} +{"lr": 0.00039045602067755495, "data_time": 0.001635122299194336, "loss": 0.14130793288350105, "time": 0.7074922323226929, "epoch": 2, "memory": 25074, "step": 12500} +{"lr": 0.00039045602067755495, "data_time": 0.001258993148803711, "loss": 0.1366831324994564, "time": 0.7151970863342285, "epoch": 2, "memory": 25074, "step": 12600} +{"lr": 0.00039045602067755495, "data_time": 0.0018751382827758788, "loss": 0.14397881776094437, "time": 0.7069556713104248, "epoch": 2, "memory": 25074, "step": 12700} +{"lr": 0.00039045602067755495, "data_time": 0.001420283317565918, "loss": 0.109129748493433, "time": 0.7081318140029907, "epoch": 2, "memory": 25074, "step": 12800} +{"lr": 0.00039045602067755495, "data_time": 0.0013454914093017577, "loss": 0.13538127541542053, "time": 0.704267144203186, "epoch": 2, "memory": 25074, "step": 12900} +{"lr": 0.00039045602067755495, "data_time": 0.002026271820068359, "loss": 0.14159722924232482, "time": 0.7106671333312988, "epoch": 2, "memory": 25074, "step": 13000} +{"lr": 0.00039045602067755495, "data_time": 0.0019367456436157227, "loss": 0.12928728386759758, "time": 0.7087173700332642, "epoch": 2, "memory": 25074, "step": 13100} +{"lr": 0.00039045602067755495, "data_time": 0.0013850927352905273, "loss": 0.13836505636572838, "time": 0.707839035987854, "epoch": 2, "memory": 25074, "step": 13200} +{"lr": 0.00039045602067755495, "data_time": 0.0018125057220458984, "loss": 0.13511562943458558, "time": 0.797005581855774, "epoch": 2, "memory": 25074, "step": 13300} +{"lr": 0.00039045602067755495, "data_time": 0.002506875991821289, "loss": 0.122445597499609, "time": 0.7427887439727783, "epoch": 2, "memory": 25074, "step": 13400} +{"lr": 0.00039045602067755495, "data_time": 0.002304267883300781, "loss": 0.11888976395130157, "time": 0.7066573143005371, "epoch": 2, "memory": 25074, "step": 13500} +{"lr": 0.00039045602067755495, "data_time": 0.0012183666229248046, "loss": 0.12036007642745972, "time": 0.7212828636169434, "epoch": 2, "memory": 25074, "step": 13600} +{"lr": 0.00039045602067755495, "data_time": 0.0019524097442626953, "loss": 0.1279669962823391, "time": 0.7417988061904908, "epoch": 2, "memory": 25074, "step": 13700} +{"lr": 0.00039045602067755495, "data_time": 0.001661539077758789, "loss": 0.1092663660645485, "time": 0.7353540658950806, "epoch": 2, "memory": 25074, "step": 13800} +{"lr": 0.00039045602067755495, "data_time": 0.0022333383560180662, "loss": 0.10931268632411957, "time": 0.8255977392196655, "epoch": 2, "memory": 25074, "step": 13900} +{"lr": 0.00039045602067755495, "data_time": 0.0017616033554077148, "loss": 0.11564795449376106, "time": 0.725689172744751, "epoch": 2, "memory": 25074, "step": 14000} +{"lr": 0.00039045602067755495, "data_time": 0.0017904520034790039, "loss": 0.11199476048350335, "time": 0.7052001476287841, "epoch": 2, "memory": 25074, "step": 14100} +{"lr": 0.00039045602067755495, "data_time": 0.0017853736877441405, "loss": 0.11048664525151253, "time": 0.7792333364486694, "epoch": 2, "memory": 25074, "step": 14200} +{"lr": 0.00039045602067755495, "data_time": 0.0033494710922241213, "loss": 0.09773226976394653, "time": 0.910241675376892, "epoch": 2, "memory": 25074, "step": 14300} +{"lr": 0.00039045602067755495, "data_time": 0.004221343994140625, "loss": 0.12284593433141708, "time": 0.8739740371704101, "epoch": 2, "memory": 25074, "step": 14400} +{"lr": 0.00039045602067755495, "data_time": 0.0012976646423339844, "loss": 0.12258741408586502, "time": 0.737842607498169, "epoch": 2, "memory": 25074, "step": 14500} +{"lr": 0.00039045602067755495, "data_time": 0.0012325048446655273, "loss": 0.1167706497013569, "time": 0.7552871465682983, "epoch": 2, "memory": 25074, "step": 14600} +{"lr": 0.00039045602067755495, "data_time": 0.0031726598739624024, "loss": 0.10753279253840446, "time": 0.8949979305267334, "epoch": 2, "memory": 25074, "step": 14700} +{"lr": 0.00039045602067755495, "data_time": 0.0013030529022216796, "loss": 0.1147630363702774, "time": 0.7629199266433716, "epoch": 2, "memory": 25074, "step": 14800} +{"lr": 0.00039045602067755495, "data_time": 0.0014464855194091797, "loss": 0.12037307247519494, "time": 0.738120698928833, "epoch": 2, "memory": 25074, "step": 14900} +{"lr": 0.00039045602067755495, "data_time": 0.0017142295837402344, "loss": 0.09465071111917496, "time": 0.7545335531234741, "epoch": 2, "memory": 25074, "step": 15000} +{"lr": 0.00039045602067755495, "data_time": 0.002235269546508789, "loss": 0.09188344329595566, "time": 0.8110369920730591, "epoch": 2, "memory": 25074, "step": 15100} +{"lr": 0.00039045602067755495, "data_time": 0.0013940095901489257, "loss": 0.10843180418014527, "time": 0.7215148448944092, "epoch": 2, "memory": 25074, "step": 15200} +{"lr": 0.00039045602067755495, "data_time": 0.0013376474380493164, "loss": 0.09459175616502762, "time": 0.7215344905853271, "epoch": 2, "memory": 25074, "step": 15300} +{"lr": 0.00039045602067755495, "data_time": 0.001319742202758789, "loss": 0.09571233578026295, "time": 0.7615896701812744, "epoch": 2, "memory": 25074, "step": 15400} +{"lr": 0.00039045602067755495, "data_time": 0.002578878402709961, "loss": 0.11871992759406566, "time": 0.8586896181106567, "epoch": 2, "memory": 25074, "step": 15500} +{"lr": 0.00039045602067755495, "data_time": 0.0011219501495361329, "loss": 0.10624518673866987, "time": 0.6593152284622192, "epoch": 2, "memory": 25074, "step": 15600} +{"accuracy/top1": 96.61006164550781, "data_time": 0.20340758211472454, "time": 0.5980817289913402, "step": 2} +{"lr": 0.00036275831390311474, "data_time": 0.0013336896896362304, "loss": 0.10310088619589805, "time": 0.7276587009429931, "epoch": 3, "memory": 25074, "step": 15700} +{"lr": 0.00036275831390311474, "data_time": 0.04652788639068604, "loss": 0.08784256875514984, "time": 0.7491942882537842, "epoch": 3, "memory": 25074, "step": 15800} +{"lr": 0.00036275831390311474, "data_time": 0.042873191833496097, "loss": 0.11111270263791084, "time": 0.7634331941604614, "epoch": 3, "memory": 25074, "step": 15900} +{"lr": 0.00036275831390311474, "data_time": 0.05382053852081299, "loss": 0.10052410736680031, "time": 0.7560703992843628, "epoch": 3, "memory": 25074, "step": 16000} +{"lr": 0.00036275831390311474, "data_time": 0.09436025619506835, "loss": 0.0834791749715805, "time": 0.8450967788696289, "epoch": 3, "memory": 25074, "step": 16100} +{"lr": 0.00036275831390311474, "data_time": 0.019767117500305176, "loss": 0.11351828798651695, "time": 0.7252115726470947, "epoch": 3, "memory": 25074, "step": 16200} +{"lr": 0.00036275831390311474, "data_time": 0.036263322830200194, "loss": 0.10130115523934365, "time": 0.7392185211181641, "epoch": 3, "memory": 25074, "step": 16300} +{"lr": 0.00036275831390311474, "data_time": 0.04586970806121826, "loss": 0.10232554823160171, "time": 0.7497789621353149, "epoch": 3, "memory": 25074, "step": 16400} +{"lr": 0.00036275831390311474, "data_time": 0.07142379283905029, "loss": 0.10023172087967396, "time": 0.7870445966720581, "epoch": 3, "memory": 25074, "step": 16500} +{"lr": 0.00036275831390311474, "data_time": 0.0784637451171875, "loss": 0.10155723989009857, "time": 0.7833858251571655, "epoch": 3, "memory": 25074, "step": 16600} +{"lr": 0.00036275831390311474, "data_time": 0.001291179656982422, "loss": 0.0844054602086544, "time": 0.7038801193237305, "epoch": 3, "memory": 25074, "step": 16700} +{"lr": 0.00036275831390311474, "data_time": 0.05306949615478516, "loss": 0.0883020780980587, "time": 0.7562115907669067, "epoch": 3, "memory": 25074, "step": 16800} +{"lr": 0.00036275831390311474, "data_time": 0.032765316963195804, "loss": 0.09582230374217034, "time": 0.7353592634201049, "epoch": 3, "memory": 25074, "step": 16900} +{"lr": 0.00036275831390311474, "data_time": 0.17749545574188233, "loss": 0.07940515093505382, "time": 0.9362649202346802, "epoch": 3, "memory": 25074, "step": 17000} +{"lr": 0.00036275831390311474, "data_time": 0.06539099216461182, "loss": 0.09481642134487629, "time": 0.7688310861587524, "epoch": 3, "memory": 25074, "step": 17100} +{"lr": 0.00036275831390311474, "data_time": 0.00936436653137207, "loss": 0.08838141858577728, "time": 0.7122403144836426, "epoch": 3, "memory": 25074, "step": 17200} +{"lr": 0.00036275831390311474, "data_time": 0.027420973777770995, "loss": 0.09090999439358712, "time": 0.7304120779037475, "epoch": 3, "memory": 25074, "step": 17300} +{"lr": 0.00036275831390311474, "data_time": 0.02483525276184082, "loss": 0.08162333928048611, "time": 0.7273419618606567, "epoch": 3, "memory": 25074, "step": 17400} +{"lr": 0.00036275831390311474, "data_time": 0.05052340030670166, "loss": 0.08197577185928821, "time": 0.7882192611694336, "epoch": 3, "memory": 25074, "step": 17500} +{"lr": 0.00036275831390311474, "data_time": 0.016635370254516602, "loss": 0.08423399589955807, "time": 0.72077476978302, "epoch": 3, "memory": 25074, "step": 17600} +{"lr": 0.00036275831390311474, "data_time": 0.09424901008605957, "loss": 0.09517066031694413, "time": 0.815043830871582, "epoch": 3, "memory": 25074, "step": 17700} +{"lr": 0.00036275831390311474, "data_time": 0.09383594989776611, "loss": 0.11006402745842933, "time": 0.7976555347442627, "epoch": 3, "memory": 25074, "step": 17800} +{"lr": 0.00036275831390311474, "data_time": 0.2113175392150879, "loss": 0.09635679423809052, "time": 0.9638019323348999, "epoch": 3, "memory": 25074, "step": 17900} +{"lr": 0.00036275831390311474, "data_time": 0.020669436454772948, "loss": 0.09522832930088043, "time": 0.7267727136611939, "epoch": 3, "memory": 25074, "step": 18000} +{"lr": 0.00036275831390311474, "data_time": 0.041952657699584964, "loss": 0.08613555580377578, "time": 0.7532625436782837, "epoch": 3, "memory": 25074, "step": 18100} +{"lr": 0.00036275831390311474, "data_time": 0.0014177799224853516, "loss": 0.08952978923916817, "time": 0.7037564754486084, "epoch": 3, "memory": 25074, "step": 18200} +{"lr": 0.00036275831390311474, "data_time": 0.04420890808105469, "loss": 0.08152703158557414, "time": 0.763043737411499, "epoch": 3, "memory": 25074, "step": 18300} +{"lr": 0.00036275831390311474, "data_time": 0.05574164390563965, "loss": 0.08979608416557312, "time": 0.8042138576507568, "epoch": 3, "memory": 25074, "step": 18400} +{"lr": 0.00036275831390311474, "data_time": 0.0014020919799804688, "loss": 0.08850124888122082, "time": 0.731146764755249, "epoch": 3, "memory": 25074, "step": 18500} +{"lr": 0.00036275831390311474, "data_time": 0.0012506008148193359, "loss": 0.0713675782084465, "time": 0.7047372341156006, "epoch": 3, "memory": 25074, "step": 18600} +{"lr": 0.00036275831390311474, "data_time": 0.0013912677764892577, "loss": 0.09244537055492401, "time": 0.7872651815414429, "epoch": 3, "memory": 25074, "step": 18700} +{"lr": 0.00036275831390311474, "data_time": 0.002373361587524414, "loss": 0.08013321980834007, "time": 0.8584137678146362, "epoch": 3, "memory": 25074, "step": 18800} +{"lr": 0.00036275831390311474, "data_time": 0.0012579679489135743, "loss": 0.09345643445849419, "time": 0.7364373683929444, "epoch": 3, "memory": 25074, "step": 18900} +{"lr": 0.00036275831390311474, "data_time": 0.001368880271911621, "loss": 0.07542329467833042, "time": 0.7544003248214721, "epoch": 3, "memory": 25074, "step": 19000} +{"lr": 0.00036275831390311474, "data_time": 0.001826310157775879, "loss": 0.08801689743995667, "time": 0.740264105796814, "epoch": 3, "memory": 25074, "step": 19100} +{"lr": 0.00036275831390311474, "data_time": 0.0012601375579833984, "loss": 0.0794067095965147, "time": 0.7126410245895386, "epoch": 3, "memory": 25074, "step": 19200} +{"lr": 0.00036275831390311474, "data_time": 0.004342913627624512, "loss": 0.07495554648339749, "time": 0.8991502046585083, "epoch": 3, "memory": 25074, "step": 19300} +{"lr": 0.00036275831390311474, "data_time": 0.001290130615234375, "loss": 0.0925452709197998, "time": 0.7162141561508178, "epoch": 3, "memory": 25074, "step": 19400} +{"lr": 0.00036275831390311474, "data_time": 0.0013996124267578124, "loss": 0.07054197788238525, "time": 0.7638662099838257, "epoch": 3, "memory": 25074, "step": 19500} +{"lr": 0.00036275831390311474, "data_time": 0.0020952939987182615, "loss": 0.07390230745077134, "time": 0.7642737865447998, "epoch": 3, "memory": 25074, "step": 19600} +{"lr": 0.00036275831390311474, "data_time": 0.0013412952423095704, "loss": 0.07707810625433922, "time": 0.7533664226531982, "epoch": 3, "memory": 25074, "step": 19700} +{"lr": 0.00036275831390311474, "data_time": 0.002320408821105957, "loss": 0.06675496846437454, "time": 0.8157778263092041, "epoch": 3, "memory": 25074, "step": 19800} +{"lr": 0.00036275831390311474, "data_time": 0.0018549203872680665, "loss": 0.09099142514169216, "time": 0.7152692794799804, "epoch": 3, "memory": 25074, "step": 19900} +{"lr": 0.00036275831390311474, "data_time": 0.0013274669647216797, "loss": 0.07269903682172299, "time": 0.7267545700073242, "epoch": 3, "memory": 25074, "step": 20000} +{"lr": 0.00036275831390311474, "data_time": 0.0017048835754394532, "loss": 0.08231619521975517, "time": 0.7686270713806153, "epoch": 3, "memory": 25074, "step": 20100} +{"lr": 0.00036275831390311474, "data_time": 0.001417827606201172, "loss": 0.06602434609085321, "time": 0.7393132209777832, "epoch": 3, "memory": 25074, "step": 20200} +{"lr": 0.00036275831390311474, "data_time": 0.0013814210891723634, "loss": 0.07883532866835594, "time": 0.7038391351699829, "epoch": 3, "memory": 25074, "step": 20300} +{"lr": 0.00036275831390311474, "data_time": 0.0012912511825561523, "loss": 0.06388877741992474, "time": 0.7664240837097168, "epoch": 3, "memory": 25074, "step": 20400} +{"lr": 0.00036275831390311474, "data_time": 0.0014119863510131836, "loss": 0.06450840719044208, "time": 0.761465334892273, "epoch": 3, "memory": 25074, "step": 20500} +{"lr": 0.00036275831390311474, "data_time": 0.0012936115264892579, "loss": 0.07086011171340942, "time": 0.7718260288238525, "epoch": 3, "memory": 25074, "step": 20600} +{"lr": 0.00036275831390311474, "data_time": 0.002866697311401367, "loss": 0.0806809864938259, "time": 0.8923632383346558, "epoch": 3, "memory": 25074, "step": 20700} +{"lr": 0.00036275831390311474, "data_time": 0.0014344215393066405, "loss": 0.06330490708351136, "time": 0.7350774765014648, "epoch": 3, "memory": 25074, "step": 20800} +{"lr": 0.00036275831390311474, "data_time": 0.0013097524642944336, "loss": 0.07199015617370605, "time": 0.7060601234436035, "epoch": 3, "memory": 25074, "step": 20900} +{"lr": 0.00036275831390311474, "data_time": 0.0014673233032226562, "loss": 0.06927086636424065, "time": 0.7415284872055053, "epoch": 3, "memory": 25074, "step": 21000} +{"lr": 0.00036275831390311474, "data_time": 0.0013409137725830078, "loss": 0.07275770232081413, "time": 0.7100635528564453, "epoch": 3, "memory": 25074, "step": 21100} +{"lr": 0.00036275831390311474, "data_time": 0.0024672508239746093, "loss": 0.071163310110569, "time": 0.8411956548690795, "epoch": 3, "memory": 25074, "step": 21200} +{"lr": 0.00036275831390311474, "data_time": 0.0012351512908935548, "loss": 0.07052629329264164, "time": 0.757619071006775, "epoch": 3, "memory": 25074, "step": 21300} +{"lr": 0.00036275831390311474, "data_time": 0.0012850999832153321, "loss": 0.06190228536725044, "time": 0.7438790559768677, "epoch": 3, "memory": 25074, "step": 21400} +{"lr": 0.00036275831390311474, "data_time": 0.0013517618179321289, "loss": 0.06569394581019879, "time": 0.7910687446594238, "epoch": 3, "memory": 25074, "step": 21500} +{"lr": 0.00036275831390311474, "data_time": 0.0026489973068237306, "loss": 0.057651686668396, "time": 0.8509131669998169, "epoch": 3, "memory": 25074, "step": 21600} +{"lr": 0.00036275831390311474, "data_time": 0.0014965295791625976, "loss": 0.06758101843297482, "time": 0.754273247718811, "epoch": 3, "memory": 25074, "step": 21700} +{"lr": 0.00036275831390311474, "data_time": 0.0012147903442382812, "loss": 0.07426468916237354, "time": 0.7568423032760621, "epoch": 3, "memory": 25074, "step": 21800} +{"lr": 0.00036275831390311474, "data_time": 0.0013579130172729492, "loss": 0.07236175537109375, "time": 0.7756822586059571, "epoch": 3, "memory": 25074, "step": 21900} +{"lr": 0.00036275831390311474, "data_time": 0.001392507553100586, "loss": 0.07234665043652058, "time": 0.7807016372680664, "epoch": 3, "memory": 25074, "step": 22000} +{"lr": 0.00036275831390311474, "data_time": 0.00457148551940918, "loss": 0.054981644824147226, "time": 0.8098761796951294, "epoch": 3, "memory": 25074, "step": 22100} +{"lr": 0.00036275831390311474, "data_time": 0.0015793323516845703, "loss": 0.06135090030729771, "time": 0.735714316368103, "epoch": 3, "memory": 25074, "step": 22200} +{"lr": 0.00036275831390311474, "data_time": 0.0013195276260375977, "loss": 0.05634091794490814, "time": 0.7164009809494019, "epoch": 3, "memory": 25074, "step": 22300} +{"lr": 0.00036275831390311474, "data_time": 0.0012403011322021484, "loss": 0.06019404791295528, "time": 0.7335800170898438, "epoch": 3, "memory": 25074, "step": 22400} +{"lr": 0.00036275831390311474, "data_time": 0.0020447731018066405, "loss": 0.06717828437685966, "time": 0.7600839614868165, "epoch": 3, "memory": 25074, "step": 22500} +{"lr": 0.00036275831390311474, "data_time": 0.0013044118881225587, "loss": 0.06792790107429028, "time": 0.8072243928909302, "epoch": 3, "memory": 25074, "step": 22600} +{"lr": 0.00036275831390311474, "data_time": 0.0013843059539794921, "loss": 0.05459876991808414, "time": 0.7623852729797364, "epoch": 3, "memory": 25074, "step": 22700} +{"lr": 0.00036275831390311474, "data_time": 0.001479363441467285, "loss": 0.060161656513810155, "time": 0.7363945245742798, "epoch": 3, "memory": 25074, "step": 22800} +{"lr": 0.00036275831390311474, "data_time": 0.01769576072692871, "loss": 0.05969017185270786, "time": 0.7204919815063476, "epoch": 3, "memory": 25074, "step": 22900} +{"lr": 0.00036275831390311474, "data_time": 0.03938698768615723, "loss": 0.058220787532627584, "time": 0.8035845756530762, "epoch": 3, "memory": 25074, "step": 23000} +{"lr": 0.00036275831390311474, "data_time": 0.13300104141235353, "loss": 0.054922381229698655, "time": 0.8375943899154663, "epoch": 3, "memory": 25074, "step": 23100} +{"lr": 0.00036275831390311474, "data_time": 0.00842437744140625, "loss": 0.06590216085314751, "time": 0.7251642227172852, "epoch": 3, "memory": 25074, "step": 23200} +{"lr": 0.00036275831390311474, "data_time": 0.02899308204650879, "loss": 0.06242698859423399, "time": 0.733867883682251, "epoch": 3, "memory": 25074, "step": 23300} +{"lr": 0.00036275831390311474, "data_time": 0.13492271900177003, "loss": 0.05929134879261255, "time": 0.8273138523101806, "epoch": 3, "memory": 25074, "step": 23400} +{"accuracy/top1": 97.89154052734375, "data_time": 0.1995232806486242, "time": 0.5852146148681641, "step": 3} +{"lr": 0.00031961812419703224, "data_time": 0.0034704208374023438, "loss": 0.04517615139484406, "time": 0.8812820911407471, "epoch": 4, "memory": 25074, "step": 23500} +{"lr": 0.00031961812419703224, "data_time": 0.00133514404296875, "loss": 0.045108597725629807, "time": 0.7198319435119629, "epoch": 4, "memory": 25074, "step": 23600} +{"lr": 0.00031961812419703224, "data_time": 0.0014042139053344726, "loss": 0.05669744834303856, "time": 0.7504631519317627, "epoch": 4, "memory": 25074, "step": 23700} +{"lr": 0.00031961812419703224, "data_time": 0.001220846176147461, "loss": 0.0512423537671566, "time": 0.7433174133300782, "epoch": 4, "memory": 25074, "step": 23800} +{"lr": 0.00031961812419703224, "data_time": 0.00250704288482666, "loss": 0.05096057280898094, "time": 0.8091708421707153, "epoch": 4, "memory": 25074, "step": 23900} +{"lr": 0.00031961812419703224, "data_time": 0.0014014959335327149, "loss": 0.0507582375779748, "time": 0.7508286714553833, "epoch": 4, "memory": 25074, "step": 24000} +{"lr": 0.00031961812419703224, "data_time": 0.0013149261474609375, "loss": 0.050636058859527114, "time": 0.8642278194427491, "epoch": 4, "memory": 25074, "step": 24100} +{"lr": 0.00031961812419703224, "data_time": 0.0013562917709350586, "loss": 0.06051886379718781, "time": 0.7549211978912354, "epoch": 4, "memory": 25074, "step": 24200} +{"lr": 0.00031961812419703224, "data_time": 0.0013926506042480468, "loss": 0.047161291167140004, "time": 0.7464770078659058, "epoch": 4, "memory": 25074, "step": 24300} +{"lr": 0.00031961812419703224, "data_time": 0.0053446292877197266, "loss": 0.05329375602304935, "time": 0.8669121026992798, "epoch": 4, "memory": 25074, "step": 24400} +{"lr": 0.00031961812419703224, "data_time": 0.0012766599655151367, "loss": 0.0478238308802247, "time": 0.7892888069152832, "epoch": 4, "memory": 25074, "step": 24500} +{"lr": 0.00031961812419703224, "data_time": 0.0013270854949951171, "loss": 0.04734180849045515, "time": 0.7029707670211792, "epoch": 4, "memory": 25074, "step": 24600} +{"lr": 0.00031961812419703224, "data_time": 0.0012899875640869141, "loss": 0.0476755864918232, "time": 0.7805631399154663, "epoch": 4, "memory": 25074, "step": 24700} +{"lr": 0.00031961812419703224, "data_time": 0.002505064010620117, "loss": 0.0517438736744225, "time": 0.8386474132537842, "epoch": 4, "memory": 25074, "step": 24800} +{"lr": 0.00031961812419703224, "data_time": 0.0013689994812011719, "loss": 0.05274440515786409, "time": 0.7870511770248413, "epoch": 4, "memory": 25074, "step": 24900} +{"lr": 0.00031961812419703224, "data_time": 0.0013701677322387694, "loss": 0.043704770132899286, "time": 0.7481824159622192, "epoch": 4, "memory": 25074, "step": 25000} +{"lr": 0.00031961812419703224, "data_time": 0.0014128446578979491, "loss": 0.04268152825534344, "time": 0.7583313465118409, "epoch": 4, "memory": 25074, "step": 25100} +{"lr": 0.00031961812419703224, "data_time": 0.00205385684967041, "loss": 0.04238433614373207, "time": 0.8143611192703247, "epoch": 4, "memory": 25074, "step": 25200} +{"lr": 0.00031961812419703224, "data_time": 0.0045460939407348635, "loss": 0.05427344739437103, "time": 0.8976642847061157, "epoch": 4, "memory": 25074, "step": 25300} +{"lr": 0.00031961812419703224, "data_time": 0.0012501716613769532, "loss": 0.04692898578941822, "time": 0.7169560194015503, "epoch": 4, "memory": 25074, "step": 25400} +{"lr": 0.00031961812419703224, "data_time": 0.001289534568786621, "loss": 0.04919785112142563, "time": 0.7365872859954834, "epoch": 4, "memory": 25074, "step": 25500} +{"lr": 0.00031961812419703224, "data_time": 0.0012316465377807616, "loss": 0.05004159957170486, "time": 0.7409481048583985, "epoch": 4, "memory": 25074, "step": 25600} +{"lr": 0.00031961812419703224, "data_time": 0.0014320135116577149, "loss": 0.05063360538333654, "time": 0.7744909286499023, "epoch": 4, "memory": 25074, "step": 25700} +{"lr": 0.00031961812419703224, "data_time": 0.0029634952545166014, "loss": 0.03686757292598486, "time": 0.7800127267837524, "epoch": 4, "memory": 25074, "step": 25800} +{"lr": 0.00031961812419703224, "data_time": 0.001425027847290039, "loss": 0.030946124531328677, "time": 0.725481915473938, "epoch": 4, "memory": 25074, "step": 25900} +{"lr": 0.00031961812419703224, "data_time": 0.001331329345703125, "loss": 0.03346645049750805, "time": 0.7261993408203125, "epoch": 4, "memory": 25074, "step": 26000} +{"lr": 0.00031961812419703224, "data_time": 0.001289200782775879, "loss": 0.050221163965761664, "time": 0.7927984952926636, "epoch": 4, "memory": 25074, "step": 26100} +{"lr": 0.00031961812419703224, "data_time": 0.0026758193969726564, "loss": 0.047254554182291034, "time": 1.0025804042816162, "epoch": 4, "memory": 25074, "step": 26200} +{"lr": 0.00031961812419703224, "data_time": 0.001240110397338867, "loss": 0.05443720631301403, "time": 0.7894600868225098, "epoch": 4, "memory": 25074, "step": 26300} +{"lr": 0.00031961812419703224, "data_time": 0.0013683557510375977, "loss": 0.04285751581192017, "time": 0.7531031370162964, "epoch": 4, "memory": 25074, "step": 26400} +{"lr": 0.00031961812419703224, "data_time": 0.0013829469680786133, "loss": 0.037723523750901225, "time": 0.7630849599838256, "epoch": 4, "memory": 25074, "step": 26500} +{"lr": 0.00031961812419703224, "data_time": 0.001319551467895508, "loss": 0.051023753546178344, "time": 0.783145809173584, "epoch": 4, "memory": 25074, "step": 26600} +{"lr": 0.00031961812419703224, "data_time": 0.0015150547027587891, "loss": 0.04813426043838263, "time": 0.7720998525619507, "epoch": 4, "memory": 25074, "step": 26700} +{"lr": 0.00031961812419703224, "data_time": 0.001169872283935547, "loss": 0.043939481303095815, "time": 0.7106137037277221, "epoch": 4, "memory": 25074, "step": 26800} +{"lr": 0.00031961812419703224, "data_time": 0.0014029979705810548, "loss": 0.03220619978383184, "time": 0.7422962427139282, "epoch": 4, "memory": 25074, "step": 26900} +{"lr": 0.00031961812419703224, "data_time": 0.001316666603088379, "loss": 0.04515266828238964, "time": 0.7716180086135864, "epoch": 4, "memory": 25074, "step": 27000} +{"lr": 0.00031961812419703224, "data_time": 0.004015541076660157, "loss": 0.037893311493098734, "time": 0.8699528932571411, "epoch": 4, "memory": 25074, "step": 27100} +{"lr": 0.00031961812419703224, "data_time": 0.0012329578399658202, "loss": 0.041449266765266654, "time": 0.7464729309082031, "epoch": 4, "memory": 25074, "step": 27200} +{"lr": 0.00031961812419703224, "data_time": 0.0012342691421508788, "loss": 0.03178658131510019, "time": 0.733365535736084, "epoch": 4, "memory": 25074, "step": 27300} +{"lr": 0.00031961812419703224, "data_time": 0.06804072856903076, "loss": 0.03859858289361, "time": 0.7901822566986084, "epoch": 4, "memory": 25074, "step": 27400} +{"lr": 0.00031961812419703224, "data_time": 0.0012955904006958009, "loss": 0.03694054465740919, "time": 0.7995924711227417, "epoch": 4, "memory": 25074, "step": 27500} +{"lr": 0.00031961812419703224, "data_time": 0.0018056631088256836, "loss": 0.04908970184624195, "time": 0.8272752046585083, "epoch": 4, "memory": 25074, "step": 27600} +{"lr": 0.00031961812419703224, "data_time": 0.0014280080795288086, "loss": 0.033122124476358296, "time": 0.7333185195922851, "epoch": 4, "memory": 25074, "step": 27700} +{"lr": 0.00031961812419703224, "data_time": 0.0016160726547241211, "loss": 0.03568306379020214, "time": 0.7440671920776367, "epoch": 4, "memory": 25074, "step": 27800} +{"lr": 0.00031961812419703224, "data_time": 0.0012700796127319337, "loss": 0.031327463872730735, "time": 0.7310517787933349, "epoch": 4, "memory": 25074, "step": 27900} +{"lr": 0.00031961812419703224, "data_time": 0.003715372085571289, "loss": 0.03493130672723055, "time": 0.8083968877792358, "epoch": 4, "memory": 25074, "step": 28000} +{"lr": 0.00031961812419703224, "data_time": 0.0012696266174316406, "loss": 0.03904465064406395, "time": 0.7613831043243409, "epoch": 4, "memory": 25074, "step": 28100} +{"lr": 0.00031961812419703224, "data_time": 0.001513051986694336, "loss": 0.02872838657349348, "time": 0.7503234624862671, "epoch": 4, "memory": 25074, "step": 28200} +{"lr": 0.00031961812419703224, "data_time": 0.0014272451400756836, "loss": 0.042362000048160556, "time": 0.7054878234863281, "epoch": 4, "memory": 25074, "step": 28300} +{"lr": 0.00031961812419703224, "data_time": 0.0029218912124633787, "loss": 0.03637932613492012, "time": 0.8058889627456665, "epoch": 4, "memory": 25074, "step": 28400} +{"lr": 0.00031961812419703224, "data_time": 0.001334834098815918, "loss": 0.03377866544760764, "time": 0.7404050350189209, "epoch": 4, "memory": 25074, "step": 28500} +{"lr": 0.00031961812419703224, "data_time": 0.0013770341873168945, "loss": 0.04600885380059481, "time": 0.7099129438400269, "epoch": 4, "memory": 25074, "step": 28600} +{"lr": 0.00031961812419703224, "data_time": 0.0014901399612426759, "loss": 0.03663482228294015, "time": 0.7665527105331421, "epoch": 4, "memory": 25074, "step": 28700} +{"lr": 0.00031961812419703224, "data_time": 0.001254701614379883, "loss": 0.03127280483022332, "time": 0.737017560005188, "epoch": 4, "memory": 25074, "step": 28800} +{"lr": 0.00031961812419703224, "data_time": 0.0013598918914794922, "loss": 0.028428874118253587, "time": 0.7405951738357544, "epoch": 4, "memory": 25074, "step": 28900} +{"lr": 0.00031961812419703224, "data_time": 0.0012872934341430664, "loss": 0.027683245483785868, "time": 0.7130789995193482, "epoch": 4, "memory": 25074, "step": 29000} +{"lr": 0.00031961812419703224, "data_time": 0.0013851404190063476, "loss": 0.029278316907584666, "time": 0.7043890237808228, "epoch": 4, "memory": 25074, "step": 29100} +{"lr": 0.00031961812419703224, "data_time": 0.0013159751892089845, "loss": 0.03678624513559044, "time": 0.7159635066986084, "epoch": 4, "memory": 25074, "step": 29200} +{"lr": 0.00031961812419703224, "data_time": 0.001799941062927246, "loss": 0.0363719992339611, "time": 0.805773115158081, "epoch": 4, "memory": 25074, "step": 29300} +{"lr": 0.00031961812419703224, "data_time": 0.0014745950698852538, "loss": 0.02529519619420171, "time": 0.7127826452255249, "epoch": 4, "memory": 25074, "step": 29400} +{"lr": 0.00031961812419703224, "data_time": 0.0012825727462768555, "loss": 0.047313752584159374, "time": 0.73499436378479, "epoch": 4, "memory": 25074, "step": 29500} +{"lr": 0.00031961812419703224, "data_time": 0.0012951374053955078, "loss": 0.03643022384494543, "time": 0.7089319705963135, "epoch": 4, "memory": 25074, "step": 29600} +{"lr": 0.00031961812419703224, "data_time": 0.0033378839492797852, "loss": 0.03052303325384855, "time": 0.8332024812698364, "epoch": 4, "memory": 25074, "step": 29700} +{"lr": 0.00031961812419703224, "data_time": 0.0012300968170166015, "loss": 0.029925617575645446, "time": 0.749910855293274, "epoch": 4, "memory": 25074, "step": 29800} +{"lr": 0.00031961812419703224, "data_time": 0.001368236541748047, "loss": 0.03468024590983987, "time": 0.7201395273208618, "epoch": 4, "memory": 25074, "step": 29900} +{"lr": 0.00031961812419703224, "data_time": 0.0015823125839233398, "loss": 0.03215574417263269, "time": 0.7682300806045532, "epoch": 4, "memory": 25074, "step": 30000} +{"lr": 0.00031961812419703224, "data_time": 0.0017320871353149413, "loss": 0.02737377109006047, "time": 0.8383288860321045, "epoch": 4, "memory": 25074, "step": 30100} +{"lr": 0.00031961812419703224, "data_time": 0.0012495756149291993, "loss": 0.03885923847556114, "time": 0.730503249168396, "epoch": 4, "memory": 25074, "step": 30200} +{"lr": 0.00031961812419703224, "data_time": 0.0013231277465820313, "loss": 0.028759025596082212, "time": 0.7839333772659302, "epoch": 4, "memory": 25074, "step": 30300} +{"lr": 0.00031961812419703224, "data_time": 0.0013317108154296876, "loss": 0.022157855425029994, "time": 0.7049121379852294, "epoch": 4, "memory": 25074, "step": 30400} +{"lr": 0.00031961812419703224, "data_time": 0.0012486934661865234, "loss": 0.023827042430639267, "time": 0.7343915700912476, "epoch": 4, "memory": 25074, "step": 30500} +{"lr": 0.00031961812419703224, "data_time": 0.005153155326843262, "loss": 0.025807593390345572, "time": 0.9068384647369385, "epoch": 4, "memory": 25074, "step": 30600} +{"lr": 0.00031961812419703224, "data_time": 0.0012485027313232423, "loss": 0.027053234493359925, "time": 0.7571660995483398, "epoch": 4, "memory": 25074, "step": 30700} +{"lr": 0.00031961812419703224, "data_time": 0.001289057731628418, "loss": 0.03416409445926547, "time": 0.7048086643218994, "epoch": 4, "memory": 25074, "step": 30800} +{"lr": 0.00031961812419703224, "data_time": 0.0011974573135375977, "loss": 0.031753899343311784, "time": 0.7595767259597779, "epoch": 4, "memory": 25074, "step": 30900} +{"lr": 0.00031961812419703224, "data_time": 0.0021053552627563477, "loss": 0.021635157894343136, "time": 0.7889343976974488, "epoch": 4, "memory": 25074, "step": 31000} +{"lr": 0.00031961812419703224, "data_time": 0.0013131141662597657, "loss": 0.0249127839691937, "time": 0.7223089218139649, "epoch": 4, "memory": 25074, "step": 31100} +{"lr": 0.00031961812419703224, "data_time": 0.001041245460510254, "loss": 0.026153938472270967, "time": 0.6581466913223266, "epoch": 4, "memory": 25074, "step": 31200} +{"accuracy/top1": 99.12883758544922, "data_time": 0.1987632583169376, "time": 0.5838278321658864, "step": 4} +{"lr": 0.00026525831390311475, "data_time": 0.0015133380889892577, "loss": 0.027698755217716097, "time": 0.7176483631134033, "epoch": 5, "memory": 25074, "step": 31300} +{"lr": 0.00026525831390311475, "data_time": 0.0013219594955444336, "loss": 0.028043771721422674, "time": 0.716804838180542, "epoch": 5, "memory": 25074, "step": 31400} +{"lr": 0.00026525831390311475, "data_time": 0.0012701034545898437, "loss": 0.025052245054394007, "time": 0.7129191160202026, "epoch": 5, "memory": 25074, "step": 31500} +{"lr": 0.00026525831390311475, "data_time": 0.00399935245513916, "loss": 0.020011105202138425, "time": 0.9069459438323975, "epoch": 5, "memory": 25074, "step": 31600} +{"lr": 0.00026525831390311475, "data_time": 0.0014359474182128907, "loss": 0.035278898850083354, "time": 0.7484704494476319, "epoch": 5, "memory": 25074, "step": 31700} +{"lr": 0.00026525831390311475, "data_time": 0.0018388986587524413, "loss": 0.023777909763157367, "time": 0.776302981376648, "epoch": 5, "memory": 25074, "step": 31800} +{"lr": 0.00026525831390311475, "data_time": 0.0014034032821655274, "loss": 0.018504769867286087, "time": 0.7267004966735839, "epoch": 5, "memory": 25074, "step": 31900} +{"lr": 0.00026525831390311475, "data_time": 0.0022325754165649415, "loss": 0.022173112258315086, "time": 0.930619478225708, "epoch": 5, "memory": 25074, "step": 32000} +{"lr": 0.00026525831390311475, "data_time": 0.0014310359954833984, "loss": 0.029171309247612954, "time": 0.7062459230422974, "epoch": 5, "memory": 25074, "step": 32100} +{"lr": 0.00026525831390311475, "data_time": 0.0014700651168823241, "loss": 0.026411455031484364, "time": 0.7170152187347412, "epoch": 5, "memory": 25074, "step": 32200} +{"lr": 0.00026525831390311475, "data_time": 0.0013317346572875976, "loss": 0.0231924450956285, "time": 0.7058538675308228, "epoch": 5, "memory": 25074, "step": 32300} +{"lr": 0.00026525831390311475, "data_time": 0.00167388916015625, "loss": 0.02030365206301212, "time": 0.8431793928146363, "epoch": 5, "memory": 25074, "step": 32400} +{"lr": 0.00026525831390311475, "data_time": 0.001358366012573242, "loss": 0.021000511338934304, "time": 0.7055831909179687, "epoch": 5, "memory": 25074, "step": 32500} +{"lr": 0.00026525831390311475, "data_time": 0.0013675928115844727, "loss": 0.02990973894484341, "time": 0.8002129554748535, "epoch": 5, "memory": 25074, "step": 32600} +{"lr": 0.00026525831390311475, "data_time": 0.0013486146926879883, "loss": 0.016818862548097967, "time": 0.7578875064849854, "epoch": 5, "memory": 25074, "step": 32700} +{"lr": 0.00026525831390311475, "data_time": 0.0012736082077026366, "loss": 0.0233812834136188, "time": 0.7536785125732421, "epoch": 5, "memory": 25074, "step": 32800} +{"lr": 0.00026525831390311475, "data_time": 0.0030343294143676757, "loss": 0.02946585286408663, "time": 0.7971512079238892, "epoch": 5, "memory": 25074, "step": 32900} +{"lr": 0.00026525831390311475, "data_time": 0.0013983011245727538, "loss": 0.024748912872746588, "time": 0.7699547052383423, "epoch": 5, "memory": 25074, "step": 33000} +{"lr": 0.00026525831390311475, "data_time": 0.0014682292938232421, "loss": 0.025327199790626764, "time": 0.7092276573181152, "epoch": 5, "memory": 25074, "step": 33100} +{"lr": 0.00026525831390311475, "data_time": 0.0018947601318359375, "loss": 0.020371937844902276, "time": 0.825655174255371, "epoch": 5, "memory": 25074, "step": 33200} +{"lr": 0.00026525831390311475, "data_time": 0.0014052391052246094, "loss": 0.0347462797537446, "time": 0.7288672924041748, "epoch": 5, "memory": 25074, "step": 33300} +{"lr": 0.00026525831390311475, "data_time": 0.002937626838684082, "loss": 0.01729632602073252, "time": 0.8441120147705078, "epoch": 5, "memory": 25074, "step": 33400} +{"lr": 0.00026525831390311475, "data_time": 0.001480698585510254, "loss": 0.019980670278891922, "time": 0.7312592029571533, "epoch": 5, "memory": 25074, "step": 33500} +{"lr": 0.00026525831390311475, "data_time": 0.0014867782592773438, "loss": 0.023634103778749704, "time": 0.7822185516357422, "epoch": 5, "memory": 25074, "step": 33600} +{"lr": 0.00026525831390311475, "data_time": 0.0015033721923828126, "loss": 0.016829533502459526, "time": 0.7079687595367432, "epoch": 5, "memory": 25074, "step": 33700} +{"lr": 0.00026525831390311475, "data_time": 0.0021425724029541016, "loss": 0.017635055258870123, "time": 0.8000155210494995, "epoch": 5, "memory": 25074, "step": 33800} +{"lr": 0.00026525831390311475, "data_time": 0.0012916326522827148, "loss": 0.01589393466711044, "time": 0.7038563966751099, "epoch": 5, "memory": 25074, "step": 33900} +{"lr": 0.00026525831390311475, "data_time": 0.0013196229934692382, "loss": 0.0237630317453295, "time": 0.7533472776412964, "epoch": 5, "memory": 25074, "step": 34000} +{"lr": 0.00026525831390311475, "data_time": 0.001246190071105957, "loss": 0.018666595220565796, "time": 0.7325097560882569, "epoch": 5, "memory": 25074, "step": 34100} +{"lr": 0.00026525831390311475, "data_time": 0.0012932062149047852, "loss": 0.017278934270143508, "time": 0.7466542959213257, "epoch": 5, "memory": 25074, "step": 34200} +{"lr": 0.00026525831390311475, "data_time": 0.0011881113052368164, "loss": 0.017761203926056623, "time": 0.7075449228286743, "epoch": 5, "memory": 25074, "step": 34300} +{"lr": 0.00026525831390311475, "data_time": 0.001402568817138672, "loss": 0.022226954717189072, "time": 0.723229956626892, "epoch": 5, "memory": 25074, "step": 34400} +{"lr": 0.00026525831390311475, "data_time": 0.0012682676315307617, "loss": 0.02354100691154599, "time": 0.7038027048110962, "epoch": 5, "memory": 25074, "step": 34500} +{"lr": 0.00026525831390311475, "data_time": 0.0012714624404907226, "loss": 0.020151436701416968, "time": 0.7552997827529907, "epoch": 5, "memory": 25074, "step": 34600} +{"lr": 0.00026525831390311475, "data_time": 0.0024079084396362305, "loss": 0.02605678914114833, "time": 0.8632853031158447, "epoch": 5, "memory": 25074, "step": 34700} +{"lr": 0.00026525831390311475, "data_time": 0.0014836788177490234, "loss": 0.020395143935456873, "time": 0.7161465406417846, "epoch": 5, "memory": 25074, "step": 34800} +{"lr": 0.00026525831390311475, "data_time": 0.00125274658203125, "loss": 0.024147900752723217, "time": 0.7409570932388305, "epoch": 5, "memory": 25074, "step": 34900} +{"lr": 0.00026525831390311475, "data_time": 0.0014492034912109374, "loss": 0.02286131465807557, "time": 0.7227227687835693, "epoch": 5, "memory": 25074, "step": 35000} +{"lr": 0.00026525831390311475, "data_time": 0.0022057533264160157, "loss": 0.02891218983568251, "time": 0.8394655466079712, "epoch": 5, "memory": 25074, "step": 35100} +{"lr": 0.00026525831390311475, "data_time": 0.0014395475387573241, "loss": 0.02146008899435401, "time": 0.7871113061904907, "epoch": 5, "memory": 25074, "step": 35200} +{"lr": 0.00026525831390311475, "data_time": 0.0012011766433715821, "loss": 0.021086252108216286, "time": 0.7503572940826416, "epoch": 5, "memory": 25074, "step": 35300} +{"lr": 0.00026525831390311475, "data_time": 0.001401805877685547, "loss": 0.020239979634061454, "time": 0.8190633058547974, "epoch": 5, "memory": 25074, "step": 35400} +{"lr": 0.00026525831390311475, "data_time": 0.004416513442993164, "loss": 0.022512492537498475, "time": 0.8431733608245849, "epoch": 5, "memory": 25074, "step": 35500} +{"lr": 0.00026525831390311475, "data_time": 0.0015740633010864259, "loss": 0.019313552416861057, "time": 0.7226876735687255, "epoch": 5, "memory": 25074, "step": 35600} +{"lr": 0.00026525831390311475, "data_time": 0.001483774185180664, "loss": 0.020557713508605958, "time": 0.7565377712249756, "epoch": 5, "memory": 25074, "step": 35700} +{"lr": 0.00026525831390311475, "data_time": 0.0013116836547851563, "loss": 0.010310977790504694, "time": 0.7280407428741456, "epoch": 5, "memory": 25074, "step": 35800} +{"lr": 0.00026525831390311475, "data_time": 0.0013847112655639648, "loss": 0.021481232857331632, "time": 0.8211503028869629, "epoch": 5, "memory": 25074, "step": 35900} +{"lr": 0.00026525831390311475, "data_time": 0.003250861167907715, "loss": 0.024764028284698723, "time": 0.8625778913497925, "epoch": 5, "memory": 25074, "step": 36000} +{"lr": 0.00026525831390311475, "data_time": 0.0012320518493652345, "loss": 0.02365124863572419, "time": 0.7453600883483886, "epoch": 5, "memory": 25074, "step": 36100} +{"lr": 0.00026525831390311475, "data_time": 0.0012217998504638673, "loss": 0.011829232051968574, "time": 0.742449951171875, "epoch": 5, "memory": 25074, "step": 36200} +{"lr": 0.00026525831390311475, "data_time": 0.0012722015380859375, "loss": 0.017892223224043845, "time": 0.7847510576248169, "epoch": 5, "memory": 25074, "step": 36300} +{"lr": 0.00026525831390311475, "data_time": 0.0013031482696533204, "loss": 0.020463507436215878, "time": 0.7392638683319092, "epoch": 5, "memory": 25074, "step": 36400} +{"lr": 0.00026525831390311475, "data_time": 0.002644181251525879, "loss": 0.02505156071856618, "time": 0.7775050878524781, "epoch": 5, "memory": 25074, "step": 36500} +{"lr": 0.00026525831390311475, "data_time": 0.0013347625732421874, "loss": 0.021378376707434654, "time": 0.7562998294830322, "epoch": 5, "memory": 25074, "step": 36600} +{"lr": 0.00026525831390311475, "data_time": 0.001228165626525879, "loss": 0.01717157321982086, "time": 0.7427967548370361, "epoch": 5, "memory": 25074, "step": 36700} +{"lr": 0.00026525831390311475, "data_time": 0.0012204170227050782, "loss": 0.011767417564988137, "time": 0.8289592027664184, "epoch": 5, "memory": 25074, "step": 36800} +{"lr": 0.00026525831390311475, "data_time": 0.002592658996582031, "loss": 0.016141739999875428, "time": 0.8661892890930176, "epoch": 5, "memory": 25074, "step": 36900} +{"lr": 0.00026525831390311475, "data_time": 0.0015212535858154298, "loss": 0.014416280947625638, "time": 0.7515527248382569, "epoch": 5, "memory": 25074, "step": 37000} +{"lr": 0.00026525831390311475, "data_time": 0.0018460512161254882, "loss": 0.023629384580999613, "time": 0.7796048879623413, "epoch": 5, "memory": 25074, "step": 37100} +{"lr": 0.00026525831390311475, "data_time": 0.0016292810440063476, "loss": 0.018551683891564606, "time": 0.7681916952133179, "epoch": 5, "memory": 25074, "step": 37200} +{"lr": 0.00026525831390311475, "data_time": 0.001635289192199707, "loss": 0.012822458054870367, "time": 0.7692230463027954, "epoch": 5, "memory": 25074, "step": 37300} +{"lr": 0.00026525831390311475, "data_time": 0.0017686128616333009, "loss": 0.01919706165790558, "time": 0.8141758441925049, "epoch": 5, "memory": 25074, "step": 37400} +{"lr": 0.00026525831390311475, "data_time": 0.001650524139404297, "loss": 0.01689677583053708, "time": 0.7051492691040039, "epoch": 5, "memory": 25074, "step": 37500} +{"lr": 0.00026525831390311475, "data_time": 0.001420307159423828, "loss": 0.018556368444114922, "time": 0.7511433362960815, "epoch": 5, "memory": 25074, "step": 37600} +{"lr": 0.00026525831390311475, "data_time": 0.001360917091369629, "loss": 0.014470644388347864, "time": 0.7846235990524292, "epoch": 5, "memory": 25074, "step": 37700} +{"lr": 0.00026525831390311475, "data_time": 0.001263904571533203, "loss": 0.016627713898196818, "time": 0.8151473760604858, "epoch": 5, "memory": 25074, "step": 37800} +{"lr": 0.00026525831390311475, "data_time": 0.0029468536376953125, "loss": 0.02148309852927923, "time": 0.8038137912750244, "epoch": 5, "memory": 25074, "step": 37900} +{"lr": 0.00026525831390311475, "data_time": 0.0013540267944335937, "loss": 0.02115346337668598, "time": 0.7132618188858032, "epoch": 5, "memory": 25074, "step": 38000} +{"lr": 0.00026525831390311475, "data_time": 0.0014809370040893555, "loss": 0.016541927028447388, "time": 0.7229875326156616, "epoch": 5, "memory": 25074, "step": 38100} +{"lr": 0.00026525831390311475, "data_time": 0.0013556718826293946, "loss": 0.025569619284942745, "time": 0.7170409440994263, "epoch": 5, "memory": 25074, "step": 38200} +{"lr": 0.00026525831390311475, "data_time": 0.002307558059692383, "loss": 0.01624313648790121, "time": 0.8092431783676147, "epoch": 5, "memory": 25074, "step": 38300} +{"lr": 0.00026525831390311475, "data_time": 0.001160883903503418, "loss": 0.011005748156458139, "time": 0.720248532295227, "epoch": 5, "memory": 25074, "step": 38400} +{"lr": 0.00026525831390311475, "data_time": 0.0014744281768798828, "loss": 0.021540593169629575, "time": 0.7683113098144532, "epoch": 5, "memory": 25074, "step": 38500} +{"lr": 0.00026525831390311475, "data_time": 0.0015828609466552734, "loss": 0.014804975502192974, "time": 0.7448345184326172, "epoch": 5, "memory": 25074, "step": 38600} +{"lr": 0.00026525831390311475, "data_time": 0.0012781620025634766, "loss": 0.020508582051843405, "time": 0.769232201576233, "epoch": 5, "memory": 25074, "step": 38700} +{"lr": 0.00026525831390311475, "data_time": 0.0035619020462036135, "loss": 0.01795935081318021, "time": 0.8573519229888916, "epoch": 5, "memory": 25074, "step": 38800} +{"lr": 0.00026525831390311475, "data_time": 0.0013607978820800782, "loss": 0.01790966363623738, "time": 0.7245409488677979, "epoch": 5, "memory": 25074, "step": 38900} +{"lr": 0.00026525831390311475, "data_time": 0.000868844985961914, "loss": 0.014371182140894233, "time": 0.6578127861022949, "epoch": 5, "memory": 25074, "step": 39000} +{"accuracy/top1": 99.55179595947266, "data_time": 0.1989019478068632, "time": 0.5841261919806985, "step": 5} +{"lr": 0.000205, "data_time": 0.0018460750579833984, "loss": 0.019231055956333877, "time": 0.7040271759033203, "epoch": 6, "memory": 25074, "step": 39100} +{"lr": 0.000205, "data_time": 0.001373624801635742, "loss": 0.013181134220212698, "time": 0.7180349349975585, "epoch": 6, "memory": 25074, "step": 39200} +{"lr": 0.000205, "data_time": 0.026488566398620607, "loss": 0.011611545691266656, "time": 0.7913643598556519, "epoch": 6, "memory": 25074, "step": 39300} +{"lr": 0.000205, "data_time": 0.0014906883239746093, "loss": 0.012931501772254706, "time": 0.7774279356002808, "epoch": 6, "memory": 25074, "step": 39400} +{"lr": 0.000205, "data_time": 0.0012401580810546876, "loss": 0.013847041875123978, "time": 0.7366981744766236, "epoch": 6, "memory": 25074, "step": 39500} +{"lr": 0.000205, "data_time": 0.0015658140182495117, "loss": 0.013702572509646415, "time": 0.7231918334960937, "epoch": 6, "memory": 25074, "step": 39600} +{"lr": 0.000205, "data_time": 0.001517486572265625, "loss": 0.012082140566781164, "time": 0.7332303762435913, "epoch": 6, "memory": 25074, "step": 39700} +{"lr": 0.000205, "data_time": 0.004176282882690429, "loss": 0.01720128827728331, "time": 0.8490522861480713, "epoch": 6, "memory": 25074, "step": 39800} +{"lr": 0.000205, "data_time": 0.0012623786926269532, "loss": 0.008695256197825074, "time": 0.7486298322677613, "epoch": 6, "memory": 25074, "step": 39900} +{"lr": 0.000205, "data_time": 0.0013205051422119141, "loss": 0.007066844450309873, "time": 0.7174515724182129, "epoch": 6, "memory": 25074, "step": 40000} +{"lr": 0.000205, "data_time": 0.0012791872024536133, "loss": 0.014360112557187676, "time": 0.7292573928833008, "epoch": 6, "memory": 25074, "step": 40100} +{"lr": 0.000205, "data_time": 0.003551483154296875, "loss": 0.016115556936711073, "time": 0.8143055677413941, "epoch": 6, "memory": 25074, "step": 40200} +{"lr": 0.000205, "data_time": 0.0013048410415649413, "loss": 0.0195564454421401, "time": 0.7255980014801026, "epoch": 6, "memory": 25074, "step": 40300} +{"lr": 0.000205, "data_time": 0.0015650749206542968, "loss": 0.008563556848093868, "time": 0.7180584669113159, "epoch": 6, "memory": 25074, "step": 40400} +{"lr": 0.000205, "data_time": 0.0014289617538452148, "loss": 0.015089165093377233, "time": 0.7049487590789795, "epoch": 6, "memory": 25074, "step": 40500} +{"lr": 0.000205, "data_time": 0.0012910842895507812, "loss": 0.01764496141113341, "time": 0.7559063673019409, "epoch": 6, "memory": 25074, "step": 40600} +{"lr": 0.000205, "data_time": 0.0015976667404174806, "loss": 0.014122887561097742, "time": 0.7320395708084106, "epoch": 6, "memory": 25074, "step": 40700} +{"lr": 0.000205, "data_time": 0.0014389753341674805, "loss": 0.017195624113082886, "time": 0.7043787002563476, "epoch": 6, "memory": 25074, "step": 40800} +{"lr": 0.000205, "data_time": 0.0014748096466064454, "loss": 0.012815610086545349, "time": 0.7041952848434448, "epoch": 6, "memory": 25074, "step": 40900} +{"lr": 0.000205, "data_time": 0.0013394355773925781, "loss": 0.017642981838434935, "time": 0.7030986309051513, "epoch": 6, "memory": 25074, "step": 41000} +{"lr": 0.000205, "data_time": 0.0019710779190063475, "loss": 0.009535460453480481, "time": 0.7235670328140259, "epoch": 6, "memory": 25074, "step": 41100} +{"lr": 0.000205, "data_time": 0.0012816190719604492, "loss": 0.015751146338880063, "time": 0.704103946685791, "epoch": 6, "memory": 25074, "step": 41200} +{"lr": 0.000205, "data_time": 0.0014745235443115235, "loss": 0.010906191961839794, "time": 0.7403566122055054, "epoch": 6, "memory": 25074, "step": 41300} +{"lr": 0.000205, "data_time": 0.0014739274978637696, "loss": 0.009607540257275105, "time": 0.7050373792648316, "epoch": 6, "memory": 25074, "step": 41400} +{"lr": 0.000205, "data_time": 0.0025877952575683594, "loss": 0.008496478525921702, "time": 0.7942193508148193, "epoch": 6, "memory": 25074, "step": 41500} +{"lr": 0.000205, "data_time": 0.0013800621032714843, "loss": 0.01434514089487493, "time": 0.7253471612930298, "epoch": 6, "memory": 25074, "step": 41600} +{"lr": 0.000205, "data_time": 0.0013710260391235352, "loss": 0.016336551727727057, "time": 0.7215386152267456, "epoch": 6, "memory": 25074, "step": 41700} +{"lr": 0.000205, "data_time": 0.0012293577194213868, "loss": 0.013037927681580186, "time": 0.7048011779785156, "epoch": 6, "memory": 25074, "step": 41800} +{"lr": 0.000205, "data_time": 0.002528524398803711, "loss": 0.012209090078249574, "time": 0.8043434858322144, "epoch": 6, "memory": 25074, "step": 41900} +{"lr": 0.000205, "data_time": 0.0016994237899780273, "loss": 0.012641871068626642, "time": 0.705679988861084, "epoch": 6, "memory": 25074, "step": 42000} +{"lr": 0.000205, "data_time": 0.0014030218124389648, "loss": 0.01131478725001216, "time": 0.7047741174697876, "epoch": 6, "memory": 25074, "step": 42100} +{"lr": 0.000205, "data_time": 0.0014711141586303711, "loss": 0.01450472166761756, "time": 0.7043525695800781, "epoch": 6, "memory": 25074, "step": 42200} +{"lr": 0.000205, "data_time": 0.0026247262954711913, "loss": 0.012451252155005931, "time": 0.802859902381897, "epoch": 6, "memory": 25074, "step": 42300} +{"lr": 0.000205, "data_time": 0.0013098955154418946, "loss": 0.009639228554442525, "time": 0.7166960716247559, "epoch": 6, "memory": 25074, "step": 42400} +{"lr": 0.000205, "data_time": 0.0013704538345336915, "loss": 0.015238004876300693, "time": 0.7345313310623169, "epoch": 6, "memory": 25074, "step": 42500} +{"lr": 0.000205, "data_time": 0.0014132976531982422, "loss": 0.011543279327452183, "time": 0.8145903587341309, "epoch": 6, "memory": 25074, "step": 42600} +{"lr": 0.000205, "data_time": 0.0020676136016845705, "loss": 0.008894957602024078, "time": 0.829864764213562, "epoch": 6, "memory": 25074, "step": 42700} +{"lr": 0.000205, "data_time": 0.0014184951782226563, "loss": 0.015117054292932152, "time": 0.7325458526611328, "epoch": 6, "memory": 25074, "step": 42800} +{"lr": 0.000205, "data_time": 0.0013321161270141602, "loss": 0.009235883736982942, "time": 0.7424086809158326, "epoch": 6, "memory": 25074, "step": 42900} +{"lr": 0.000205, "data_time": 0.0015939950942993164, "loss": 0.013493990153074264, "time": 0.7807001113891602, "epoch": 6, "memory": 25074, "step": 43000} +{"lr": 0.000205, "data_time": 0.001594853401184082, "loss": 0.01986273843795061, "time": 0.7461189031600952, "epoch": 6, "memory": 25074, "step": 43100} +{"lr": 0.000205, "data_time": 0.004760479927062989, "loss": 0.012629456259310246, "time": 0.9208427190780639, "epoch": 6, "memory": 25074, "step": 43200} +{"lr": 0.000205, "data_time": 0.0015197992324829102, "loss": 0.010611542407423258, "time": 0.7424284934997558, "epoch": 6, "memory": 25074, "step": 43300} +{"lr": 0.000205, "data_time": 0.0013847827911376953, "loss": 0.014244463993236423, "time": 0.7358451128005982, "epoch": 6, "memory": 25074, "step": 43400} +{"lr": 0.000205, "data_time": 0.0013418912887573243, "loss": 0.011222891882061959, "time": 0.7344607591629029, "epoch": 6, "memory": 25074, "step": 43500} +{"lr": 0.000205, "data_time": 0.019410061836242675, "loss": 0.01397267752327025, "time": 0.7354110717773438, "epoch": 6, "memory": 25074, "step": 43600} +{"lr": 0.000205, "data_time": 0.023637723922729493, "loss": 0.015269479947164655, "time": 0.7457617521286011, "epoch": 6, "memory": 25074, "step": 43700} +{"lr": 0.000205, "data_time": 0.04947209358215332, "loss": 0.013520948542281985, "time": 0.7643749952316284, "epoch": 6, "memory": 25074, "step": 43800} +{"lr": 0.000205, "data_time": 0.08439290523529053, "loss": 0.009138938132673502, "time": 0.7958259820938111, "epoch": 6, "memory": 25074, "step": 43900} +{"lr": 0.000205, "data_time": 0.0013115644454956056, "loss": 0.011309412447735668, "time": 0.7158514261245728, "epoch": 6, "memory": 25074, "step": 44000} +{"lr": 0.000205, "data_time": 0.19720866680145263, "loss": 0.0143601362593472, "time": 0.9396367073059082, "epoch": 6, "memory": 25074, "step": 44100} +{"lr": 0.000205, "data_time": 0.08205864429473878, "loss": 0.012898083496838807, "time": 0.7914128303527832, "epoch": 6, "memory": 25074, "step": 44200} +{"lr": 0.000205, "data_time": 0.050267791748046874, "loss": 0.009320208802819252, "time": 0.7634162664413452, "epoch": 6, "memory": 25074, "step": 44300} +{"lr": 0.000205, "data_time": 0.05628128051757812, "loss": 0.008199787745252252, "time": 0.7752848386764526, "epoch": 6, "memory": 25074, "step": 44400} +{"lr": 0.000205, "data_time": 0.06082310676574707, "loss": 0.009938874375075102, "time": 0.7634033918380737, "epoch": 6, "memory": 25074, "step": 44500} +{"lr": 0.000205, "data_time": 0.018892621994018553, "loss": 0.007618839293718338, "time": 0.784738826751709, "epoch": 6, "memory": 25074, "step": 44600} +{"lr": 0.000205, "data_time": 0.026316905021667482, "loss": 0.011888827150687575, "time": 0.7341613054275513, "epoch": 6, "memory": 25074, "step": 44700} +{"lr": 0.000205, "data_time": 0.0013403892517089844, "loss": 0.012039583222940565, "time": 0.7232708692550659, "epoch": 6, "memory": 25074, "step": 44800} +{"lr": 0.000205, "data_time": 0.09245812892913818, "loss": 0.00985512938350439, "time": 0.8072475910186767, "epoch": 6, "memory": 25074, "step": 44900} +{"lr": 0.000205, "data_time": 0.1281057596206665, "loss": 0.014563016546890139, "time": 0.8686982870101929, "epoch": 6, "memory": 25074, "step": 45000} +{"lr": 0.000205, "data_time": 0.0011036157608032226, "loss": 0.007399301510304213, "time": 0.703536868095398, "epoch": 6, "memory": 25074, "step": 45100} +{"lr": 0.000205, "data_time": 0.0012583732604980469, "loss": 0.010445272270590066, "time": 0.7305407047271728, "epoch": 6, "memory": 25074, "step": 45200} +{"lr": 0.000205, "data_time": 0.0014368534088134766, "loss": 0.0144415148999542, "time": 0.749235200881958, "epoch": 6, "memory": 25074, "step": 45300} +{"lr": 0.000205, "data_time": 0.0014327049255371093, "loss": 0.015263443626463414, "time": 0.7857434749603271, "epoch": 6, "memory": 25074, "step": 45400} +{"lr": 0.000205, "data_time": 0.0015762567520141602, "loss": 0.01437802640721202, "time": 0.7276469945907593, "epoch": 6, "memory": 25074, "step": 45500} +{"lr": 0.000205, "data_time": 0.001499176025390625, "loss": 0.011055087158456444, "time": 0.7303604364395142, "epoch": 6, "memory": 25074, "step": 45600} +{"lr": 0.000205, "data_time": 0.001480841636657715, "loss": 0.010005244286730886, "time": 0.8664628028869629, "epoch": 6, "memory": 25074, "step": 45700} +{"lr": 0.000205, "data_time": 0.001304936408996582, "loss": 0.010457220673561095, "time": 0.8030084848403931, "epoch": 6, "memory": 25074, "step": 45800} +{"lr": 0.000205, "data_time": 0.002091646194458008, "loss": 0.012011106358841062, "time": 0.7231206893920898, "epoch": 6, "memory": 25074, "step": 45900} +{"lr": 0.000205, "data_time": 0.0034818410873413085, "loss": 0.010440809139981865, "time": 0.8994972467422485, "epoch": 6, "memory": 25074, "step": 46000} +{"lr": 0.000205, "data_time": 0.001363849639892578, "loss": 0.012036335561424494, "time": 0.7369155406951904, "epoch": 6, "memory": 25074, "step": 46100} +{"lr": 0.000205, "data_time": 0.0013537883758544921, "loss": 0.009035256505012513, "time": 0.7985840797424316, "epoch": 6, "memory": 25074, "step": 46200} +{"lr": 0.000205, "data_time": 0.0016448259353637694, "loss": 0.009535867674276232, "time": 0.7629476547241211, "epoch": 6, "memory": 25074, "step": 46300} +{"lr": 0.000205, "data_time": 0.0013123035430908203, "loss": 0.01597025040537119, "time": 0.7717717885971069, "epoch": 6, "memory": 25074, "step": 46400} +{"lr": 0.000205, "data_time": 0.002623105049133301, "loss": 0.01929229013621807, "time": 0.8350199222564697, "epoch": 6, "memory": 25074, "step": 46500} +{"lr": 0.000205, "data_time": 0.0015046358108520507, "loss": 0.014013120532035827, "time": 0.7858851432800293, "epoch": 6, "memory": 25074, "step": 46600} +{"lr": 0.000205, "data_time": 0.0012997627258300782, "loss": 0.008184753078967332, "time": 0.7804155826568604, "epoch": 6, "memory": 25074, "step": 46700} +{"lr": 0.000205, "data_time": 0.0014567613601684571, "loss": 0.009931951481848954, "time": 0.6597428798675538, "epoch": 6, "memory": 25074, "step": 46800} +{"accuracy/top1": 99.867431640625, "data_time": 0.2033681448768167, "time": 0.5902450505424949, "step": 6} +{"lr": 0.00014474168609688527, "data_time": 0.002039980888366699, "loss": 0.008903268352150917, "time": 0.8040812969207763, "epoch": 7, "memory": 25074, "step": 46900} +{"lr": 0.00014474168609688527, "data_time": 0.0019177913665771485, "loss": 0.009964527282863855, "time": 0.7048364162445069, "epoch": 7, "memory": 25074, "step": 47000} +{"lr": 0.00014474168609688527, "data_time": 0.0019239187240600586, "loss": 0.0071931109763681885, "time": 0.9699006319046021, "epoch": 7, "memory": 25074, "step": 47100} +{"lr": 0.00014474168609688527, "data_time": 0.0014249086380004883, "loss": 0.0061600552406162025, "time": 0.7036574125289917, "epoch": 7, "memory": 25074, "step": 47200} +{"lr": 0.00014474168609688527, "data_time": 0.0013582229614257813, "loss": 0.009402553550899029, "time": 0.7363144636154175, "epoch": 7, "memory": 25074, "step": 47300} +{"lr": 0.00014474168609688527, "data_time": 0.0013524293899536133, "loss": 0.008217450184747576, "time": 0.7047011137008667, "epoch": 7, "memory": 25074, "step": 47400} +{"lr": 0.00014474168609688527, "data_time": 0.0012930870056152345, "loss": 0.0057783856522291895, "time": 0.8039856195449829, "epoch": 7, "memory": 25074, "step": 47500} +{"lr": 0.00014474168609688527, "data_time": 0.001946091651916504, "loss": 0.010270766355097294, "time": 0.7182083606719971, "epoch": 7, "memory": 25074, "step": 47600} +{"lr": 0.00014474168609688527, "data_time": 0.0015686273574829102, "loss": 0.008132395148277283, "time": 0.8044255256652832, "epoch": 7, "memory": 25074, "step": 47700} +{"lr": 0.00014474168609688527, "data_time": 0.0013362407684326173, "loss": 0.007332683773711324, "time": 0.7166149854660034, "epoch": 7, "memory": 25074, "step": 47800} +{"lr": 0.00014474168609688527, "data_time": 0.0015433788299560546, "loss": 0.0035177916288375854, "time": 0.7934905767440796, "epoch": 7, "memory": 25074, "step": 47900} +{"lr": 0.00014474168609688527, "data_time": 0.0037358283996582033, "loss": 0.009405344678089022, "time": 0.8660728454589843, "epoch": 7, "memory": 25074, "step": 48000} +{"lr": 0.00014474168609688527, "data_time": 0.001355457305908203, "loss": 0.012066180864349008, "time": 0.7574211120605469, "epoch": 7, "memory": 25074, "step": 48100} +{"lr": 0.00014474168609688527, "data_time": 0.0015769481658935546, "loss": 0.008158619981259108, "time": 0.7830992698669433, "epoch": 7, "memory": 25074, "step": 48200} +{"lr": 0.00014474168609688527, "data_time": 0.0012317657470703124, "loss": 0.008703110832720995, "time": 0.7489234209060669, "epoch": 7, "memory": 25074, "step": 48300} +{"lr": 0.00014474168609688527, "data_time": 0.0014250516891479493, "loss": 0.00903176930733025, "time": 0.8023950815200805, "epoch": 7, "memory": 25074, "step": 48400} +{"lr": 0.00014474168609688527, "data_time": 0.0025968074798583983, "loss": 0.004749373253434897, "time": 0.8814101219177246, "epoch": 7, "memory": 25074, "step": 48500} +{"lr": 0.00014474168609688527, "data_time": 0.0013814449310302734, "loss": 0.009541194513440132, "time": 0.7099181175231933, "epoch": 7, "memory": 25074, "step": 48600} +{"lr": 0.00014474168609688527, "data_time": 0.00133209228515625, "loss": 0.01119645982980728, "time": 0.7453160285949707, "epoch": 7, "memory": 25074, "step": 48700} +{"lr": 0.00014474168609688527, "data_time": 0.0013719797134399414, "loss": 0.005983489099889994, "time": 0.7479058742523194, "epoch": 7, "memory": 25074, "step": 48800} +{"lr": 0.00014474168609688527, "data_time": 0.0022298812866210936, "loss": 0.008830545051023365, "time": 0.8155020713806153, "epoch": 7, "memory": 25074, "step": 48900} +{"lr": 0.00014474168609688527, "data_time": 0.0013176918029785157, "loss": 0.009136131405830384, "time": 0.7293948173522949, "epoch": 7, "memory": 25074, "step": 49000} +{"lr": 0.00014474168609688527, "data_time": 0.001520228385925293, "loss": 0.009767140028998256, "time": 0.8100643157958984, "epoch": 7, "memory": 25074, "step": 49100} +{"lr": 0.00014474168609688527, "data_time": 0.0013205051422119141, "loss": 0.007960773445665837, "time": 0.8000875949859619, "epoch": 7, "memory": 25074, "step": 49200} +{"lr": 0.00014474168609688527, "data_time": 0.0013315200805664063, "loss": 0.008458681823685765, "time": 0.7050417423248291, "epoch": 7, "memory": 25074, "step": 49300} +{"lr": 0.00014474168609688527, "data_time": 0.0031881809234619142, "loss": 0.006933752307668328, "time": 0.9377498865127564, "epoch": 7, "memory": 25074, "step": 49400} +{"lr": 0.00014474168609688527, "data_time": 0.0013633966445922852, "loss": 0.004100177623331547, "time": 0.7347213983535766, "epoch": 7, "memory": 25074, "step": 49500} +{"lr": 0.00014474168609688527, "data_time": 0.0013150691986083985, "loss": 0.0046753640286624435, "time": 0.7202700138092041, "epoch": 7, "memory": 25074, "step": 49600} +{"lr": 0.00014474168609688527, "data_time": 0.0015794992446899413, "loss": 0.012539870711043477, "time": 0.7055185079574585, "epoch": 7, "memory": 25074, "step": 49700} +{"lr": 0.00014474168609688527, "data_time": 0.0012950897216796875, "loss": 0.010754825919866562, "time": 0.815887188911438, "epoch": 7, "memory": 25074, "step": 49800} +{"lr": 0.00014474168609688527, "data_time": 0.0016344070434570312, "loss": 0.003147457307204604, "time": 0.7243025064468384, "epoch": 7, "memory": 25074, "step": 49900} +{"lr": 0.00014474168609688527, "data_time": 0.0014695405960083007, "loss": 0.007378797419369221, "time": 0.7496392250061035, "epoch": 7, "memory": 25074, "step": 50000} +{"lr": 0.00014474168609688527, "data_time": 0.0012321233749389648, "loss": 0.006423063715919853, "time": 0.7316712141036987, "epoch": 7, "memory": 25074, "step": 50100} +{"lr": 0.00014474168609688527, "data_time": 0.0013420820236206056, "loss": 0.005607046559453011, "time": 0.7059974908828736, "epoch": 7, "memory": 25074, "step": 50200} +{"lr": 0.00014474168609688527, "data_time": 0.003515005111694336, "loss": 0.009945366438478232, "time": 0.8344825744628906, "epoch": 7, "memory": 25074, "step": 50300} +{"lr": 0.00014474168609688527, "data_time": 0.0015527963638305663, "loss": 0.010599404806271195, "time": 0.7681415557861329, "epoch": 7, "memory": 25074, "step": 50400} +{"lr": 0.00014474168609688527, "data_time": 0.001502251625061035, "loss": 0.007424869574606419, "time": 0.722867465019226, "epoch": 7, "memory": 25074, "step": 50500} +{"lr": 0.00014474168609688527, "data_time": 0.001420450210571289, "loss": 0.006481426581740379, "time": 0.7126387357711792, "epoch": 7, "memory": 25074, "step": 50600} +{"lr": 0.00014474168609688527, "data_time": 0.003195929527282715, "loss": 0.007120138918980956, "time": 0.8828189373016357, "epoch": 7, "memory": 25074, "step": 50700} +{"lr": 0.00014474168609688527, "data_time": 0.0014522075653076172, "loss": 0.006790314661338925, "time": 0.7370819568634033, "epoch": 7, "memory": 25074, "step": 50800} +{"lr": 0.00014474168609688527, "data_time": 0.001359868049621582, "loss": 0.007473695790395141, "time": 0.7462273120880127, "epoch": 7, "memory": 25074, "step": 50900} +{"lr": 0.00014474168609688527, "data_time": 0.0014990568161010742, "loss": 0.0059722459875047205, "time": 0.7586614608764648, "epoch": 7, "memory": 25074, "step": 51000} +{"lr": 0.00014474168609688527, "data_time": 0.0012622833251953124, "loss": 0.007621635589748621, "time": 0.808885145187378, "epoch": 7, "memory": 25074, "step": 51100} +{"lr": 0.00014474168609688527, "data_time": 0.0029109716415405273, "loss": 0.010983528615906835, "time": 0.8470036745071411, "epoch": 7, "memory": 25074, "step": 51200} +{"lr": 0.00014474168609688527, "data_time": 0.0014332294464111327, "loss": 0.01582004497759044, "time": 0.7581027507781982, "epoch": 7, "memory": 25074, "step": 51300} +{"lr": 0.00014474168609688527, "data_time": 0.0013092041015625, "loss": 0.008110878337174655, "time": 0.747739315032959, "epoch": 7, "memory": 25074, "step": 51400} +{"lr": 0.00014474168609688527, "data_time": 0.0013184070587158204, "loss": 0.009489529207348824, "time": 0.7333887577056885, "epoch": 7, "memory": 25074, "step": 51500} +{"lr": 0.00014474168609688527, "data_time": 0.0023162364959716797, "loss": 0.007917167106643318, "time": 0.8428509473800659, "epoch": 7, "memory": 25074, "step": 51600} +{"lr": 0.00014474168609688527, "data_time": 0.0012856245040893555, "loss": 0.007682484900578856, "time": 0.7118677139282227, "epoch": 7, "memory": 25074, "step": 51700} +{"lr": 0.00014474168609688527, "data_time": 0.0013965845108032226, "loss": 0.009817048953846098, "time": 0.7641555547714234, "epoch": 7, "memory": 25074, "step": 51800} +{"lr": 0.00014474168609688527, "data_time": 0.0015497207641601562, "loss": 0.005965145118534565, "time": 0.7311853170394897, "epoch": 7, "memory": 25074, "step": 51900} +{"lr": 0.00014474168609688527, "data_time": 0.0015300989151000976, "loss": 0.006508117076009512, "time": 0.7900309562683105, "epoch": 7, "memory": 25074, "step": 52000} +{"lr": 0.00014474168609688527, "data_time": 0.003158879280090332, "loss": 0.010603367071598768, "time": 0.7965076923370361, "epoch": 7, "memory": 25074, "step": 52100} +{"lr": 0.00014474168609688527, "data_time": 0.0012988567352294922, "loss": 0.010281467298045754, "time": 0.7051944017410279, "epoch": 7, "memory": 25074, "step": 52200} +{"lr": 0.00014474168609688527, "data_time": 0.0014504671096801757, "loss": 0.006418878398835659, "time": 0.7626439094543457, "epoch": 7, "memory": 25074, "step": 52300} +{"lr": 0.00014474168609688527, "data_time": 0.0013737916946411134, "loss": 0.010212240228429436, "time": 0.7140905618667602, "epoch": 7, "memory": 25074, "step": 52400} +{"lr": 0.00014474168609688527, "data_time": 0.0014587640762329102, "loss": 0.007123590679839253, "time": 0.7039916038513183, "epoch": 7, "memory": 25074, "step": 52500} +{"lr": 0.00014474168609688527, "data_time": 0.0014328956604003906, "loss": 0.0073842604644596575, "time": 0.7820403099060058, "epoch": 7, "memory": 25074, "step": 52600} +{"lr": 0.00014474168609688527, "data_time": 0.003953814506530762, "loss": 0.007126175705343485, "time": 0.9153799057006836, "epoch": 7, "memory": 25074, "step": 52700} +{"lr": 0.00014474168609688527, "data_time": 0.0012418746948242188, "loss": 0.005464016925543547, "time": 0.7207589387893677, "epoch": 7, "memory": 25074, "step": 52800} +{"lr": 0.00014474168609688527, "data_time": 0.001244354248046875, "loss": 0.005571501096710563, "time": 0.7073694944381714, "epoch": 7, "memory": 25074, "step": 52900} +{"lr": 0.00014474168609688527, "data_time": 0.0013068437576293946, "loss": 0.003804550599306822, "time": 0.7069416046142578, "epoch": 7, "memory": 25074, "step": 53000} +{"lr": 0.00014474168609688527, "data_time": 0.00139310359954834, "loss": 0.006326936697587371, "time": 0.7342127561569214, "epoch": 7, "memory": 25074, "step": 53100} +{"lr": 0.00014474168609688527, "data_time": 0.002085375785827637, "loss": 0.007772488426417113, "time": 0.7185160398483277, "epoch": 7, "memory": 25074, "step": 53200} +{"lr": 0.00014474168609688527, "data_time": 0.0014556169509887696, "loss": 0.006153023708611727, "time": 0.7050572395324707, "epoch": 7, "memory": 25074, "step": 53300} +{"lr": 0.00014474168609688527, "data_time": 0.0013092279434204102, "loss": 0.006335545424371957, "time": 0.7160191535949707, "epoch": 7, "memory": 25074, "step": 53400} +{"lr": 0.00014474168609688527, "data_time": 0.001264190673828125, "loss": 0.008011970529332757, "time": 0.7320989847183228, "epoch": 7, "memory": 25074, "step": 53500} +{"lr": 0.00014474168609688527, "data_time": 0.0017467737197875977, "loss": 0.008451334619894623, "time": 0.7773605346679687, "epoch": 7, "memory": 25074, "step": 53600} +{"lr": 0.00014474168609688527, "data_time": 0.0014884233474731444, "loss": 0.005294105829671025, "time": 0.7955523014068604, "epoch": 7, "memory": 25074, "step": 53700} +{"lr": 0.00014474168609688527, "data_time": 0.0014178752899169922, "loss": 0.008219748688861727, "time": 0.7150989532470703, "epoch": 7, "memory": 25074, "step": 53800} +{"lr": 0.00014474168609688527, "data_time": 0.0013798236846923827, "loss": 0.005182515317574144, "time": 0.7039018630981445, "epoch": 7, "memory": 25074, "step": 53900} +{"lr": 0.00014474168609688527, "data_time": 0.0037261962890625, "loss": 0.0045313382055610415, "time": 0.90395667552948, "epoch": 7, "memory": 25074, "step": 54000} +{"lr": 0.00014474168609688527, "data_time": 0.0015595197677612306, "loss": 0.016572838090360163, "time": 0.7274059295654297, "epoch": 7, "memory": 25074, "step": 54100} +{"lr": 0.00014474168609688527, "data_time": 0.0014645814895629882, "loss": 0.0096655139233917, "time": 0.7063659429550171, "epoch": 7, "memory": 25074, "step": 54200} +{"lr": 0.00014474168609688527, "data_time": 0.0014026403427124024, "loss": 0.006036846432834864, "time": 0.735233473777771, "epoch": 7, "memory": 25074, "step": 54300} +{"lr": 0.00014474168609688527, "data_time": 0.0014257431030273438, "loss": 0.006519056530669332, "time": 0.7424371480941773, "epoch": 7, "memory": 25074, "step": 54400} +{"lr": 0.00014474168609688527, "data_time": 0.0028959989547729494, "loss": 0.014014235092326998, "time": 0.7895010232925415, "epoch": 7, "memory": 25074, "step": 54500} +{"lr": 0.00014474168609688527, "data_time": 0.0012877225875854493, "loss": 0.007901851180940866, "time": 0.6710731506347656, "epoch": 7, "memory": 25074, "step": 54600} +{"accuracy/top1": 99.94949340820312, "data_time": 0.20048201785368078, "time": 0.5843884664423326, "step": 7} +{"lr": 9.038187580296776e-05, "data_time": 0.0014898061752319336, "loss": 0.002261943044140935, "time": 0.7649091720581055, "epoch": 8, "memory": 25074, "step": 54700} +{"lr": 9.038187580296776e-05, "data_time": 0.0014862775802612304, "loss": 0.0029871654231101276, "time": 0.7620866298675537, "epoch": 8, "memory": 25074, "step": 54800} +{"lr": 9.038187580296776e-05, "data_time": 0.0014408349990844727, "loss": 0.004256152547895908, "time": 0.744843864440918, "epoch": 8, "memory": 25074, "step": 54900} +{"lr": 9.038187580296776e-05, "data_time": 0.0016133308410644532, "loss": 0.010352508304640651, "time": 0.7147590637207031, "epoch": 8, "memory": 25074, "step": 55000} +{"lr": 9.038187580296776e-05, "data_time": 0.0025226354598999025, "loss": 0.006565147684887052, "time": 0.7641940832138061, "epoch": 8, "memory": 25074, "step": 55100} +{"lr": 9.038187580296776e-05, "data_time": 0.0014429092407226562, "loss": 0.004011364700272679, "time": 0.7107864141464233, "epoch": 8, "memory": 25074, "step": 55200} +{"lr": 9.038187580296776e-05, "data_time": 0.0015347719192504883, "loss": 0.00426431018859148, "time": 0.7742004632949829, "epoch": 8, "memory": 25074, "step": 55300} +{"lr": 9.038187580296776e-05, "data_time": 0.0013683557510375977, "loss": 0.0031507344916462897, "time": 0.7537154436111451, "epoch": 8, "memory": 25074, "step": 55400} +{"lr": 9.038187580296776e-05, "data_time": 0.002106904983520508, "loss": 0.004265059903264046, "time": 0.8996085882186889, "epoch": 8, "memory": 25074, "step": 55500} +{"lr": 9.038187580296776e-05, "data_time": 0.0014028549194335938, "loss": 0.006714503979310393, "time": 0.7374112844467163, "epoch": 8, "memory": 25074, "step": 55600} +{"lr": 9.038187580296776e-05, "data_time": 0.001227593421936035, "loss": 0.013006808469071984, "time": 0.7286698341369628, "epoch": 8, "memory": 25074, "step": 55700} +{"lr": 9.038187580296776e-05, "data_time": 0.0014836549758911132, "loss": 0.010702064307406545, "time": 0.7369499444961548, "epoch": 8, "memory": 25074, "step": 55800} +{"lr": 9.038187580296776e-05, "data_time": 0.0013474702835083007, "loss": 0.01078363386914134, "time": 0.7149914026260376, "epoch": 8, "memory": 25074, "step": 55900} +{"lr": 9.038187580296776e-05, "data_time": 0.00410001277923584, "loss": 0.0045760229229927065, "time": 0.7505262851715088, "epoch": 8, "memory": 25074, "step": 56000} +{"lr": 9.038187580296776e-05, "data_time": 0.0012885332107543945, "loss": 0.0060073901433497666, "time": 0.7120464324951172, "epoch": 8, "memory": 25074, "step": 56100} +{"lr": 9.038187580296776e-05, "data_time": 0.0015423059463500976, "loss": 0.007365959184244275, "time": 0.7305982828140258, "epoch": 8, "memory": 25074, "step": 56200} +{"lr": 9.038187580296776e-05, "data_time": 0.001609945297241211, "loss": 0.005678521329537034, "time": 0.7192543268203735, "epoch": 8, "memory": 25074, "step": 56300} +{"lr": 9.038187580296776e-05, "data_time": 0.002950239181518555, "loss": 0.005616268655285239, "time": 0.7835325717926025, "epoch": 8, "memory": 25074, "step": 56400} +{"lr": 9.038187580296776e-05, "data_time": 0.001552271842956543, "loss": 0.004612666042521596, "time": 0.7550269365310669, "epoch": 8, "memory": 25074, "step": 56500} +{"lr": 9.038187580296776e-05, "data_time": 0.001545858383178711, "loss": 0.005139764491468668, "time": 0.7496322631835938, "epoch": 8, "memory": 25074, "step": 56600} +{"lr": 9.038187580296776e-05, "data_time": 0.0013528108596801759, "loss": 0.005035810358822346, "time": 0.7202007293701171, "epoch": 8, "memory": 25074, "step": 56700} +{"lr": 9.038187580296776e-05, "data_time": 0.0015584230422973633, "loss": 0.006232081679627299, "time": 0.7533228635787964, "epoch": 8, "memory": 25074, "step": 56800} +{"lr": 9.038187580296776e-05, "data_time": 0.0013693571090698242, "loss": 0.004599780263379216, "time": 0.7272397994995117, "epoch": 8, "memory": 25074, "step": 56900} +{"lr": 9.038187580296776e-05, "data_time": 0.0016028404235839844, "loss": 0.006653623189777136, "time": 0.7921180248260498, "epoch": 8, "memory": 25074, "step": 57000} +{"lr": 9.038187580296776e-05, "data_time": 0.0014243125915527344, "loss": 0.0017732460051774978, "time": 0.739703369140625, "epoch": 8, "memory": 25074, "step": 57100} +{"lr": 9.038187580296776e-05, "data_time": 0.001279759407043457, "loss": 0.005414340365678072, "time": 0.7160408020019531, "epoch": 8, "memory": 25074, "step": 57200} +{"lr": 9.038187580296776e-05, "data_time": 0.001488471031188965, "loss": 0.005453924834728241, "time": 0.7304423570632934, "epoch": 8, "memory": 25074, "step": 57300} +{"lr": 9.038187580296776e-05, "data_time": 0.0015074729919433594, "loss": 0.0060550330206751825, "time": 0.7392881631851196, "epoch": 8, "memory": 25074, "step": 57400} +{"lr": 9.038187580296776e-05, "data_time": 0.0016181468963623047, "loss": 0.0031339862383902074, "time": 0.8159468412399292, "epoch": 8, "memory": 25074, "step": 57500} +{"lr": 9.038187580296776e-05, "data_time": 0.0014441967010498046, "loss": 0.005521983653306961, "time": 0.758598780632019, "epoch": 8, "memory": 25074, "step": 57600} +{"lr": 9.038187580296776e-05, "data_time": 0.0013397455215454102, "loss": 0.00791026996448636, "time": 0.7384653091430664, "epoch": 8, "memory": 25074, "step": 57700} +{"lr": 9.038187580296776e-05, "data_time": 0.001361536979675293, "loss": 0.007311791973188519, "time": 0.7191442966461181, "epoch": 8, "memory": 25074, "step": 57800} +{"lr": 9.038187580296776e-05, "data_time": 0.001282072067260742, "loss": 0.009632058395072817, "time": 0.7039193868637085, "epoch": 8, "memory": 25074, "step": 57900} +{"lr": 9.038187580296776e-05, "data_time": 0.0013043403625488282, "loss": 0.008204174460843205, "time": 0.7197263240814209, "epoch": 8, "memory": 25074, "step": 58000} +{"lr": 9.038187580296776e-05, "data_time": 0.001505422592163086, "loss": 0.0016403344925493002, "time": 0.7055933952331543, "epoch": 8, "memory": 25074, "step": 58100} +{"lr": 9.038187580296776e-05, "data_time": 0.0013222694396972656, "loss": 0.008053093031048775, "time": 0.7270200490951538, "epoch": 8, "memory": 25074, "step": 58200} +{"lr": 9.038187580296776e-05, "data_time": 0.0013428688049316405, "loss": 0.004103085165843368, "time": 0.7339093923568726, "epoch": 8, "memory": 25074, "step": 58300} +{"lr": 9.038187580296776e-05, "data_time": 0.01665949821472168, "loss": 0.0034529171418398617, "time": 0.7325861930847168, "epoch": 8, "memory": 25074, "step": 58400} +{"lr": 9.038187580296776e-05, "data_time": 0.013081240653991699, "loss": 0.0036817141342908146, "time": 0.7154910802841187, "epoch": 8, "memory": 25074, "step": 58500} +{"lr": 9.038187580296776e-05, "data_time": 0.0013049125671386718, "loss": 0.005961047857999802, "time": 0.7240694522857666, "epoch": 8, "memory": 25074, "step": 58600} +{"lr": 9.038187580296776e-05, "data_time": 0.0013455629348754882, "loss": 0.007555626705288887, "time": 0.7034820556640625, "epoch": 8, "memory": 25074, "step": 58700} +{"lr": 9.038187580296776e-05, "data_time": 0.016745781898498534, "loss": 0.011000830959528685, "time": 0.7352453708648682, "epoch": 8, "memory": 25074, "step": 58800} +{"lr": 9.038187580296776e-05, "data_time": 0.0015687942504882812, "loss": 0.004430231545120478, "time": 0.7043278932571411, "epoch": 8, "memory": 25074, "step": 58900} +{"lr": 9.038187580296776e-05, "data_time": 0.0013176918029785157, "loss": 0.003918724087998271, "time": 0.7339490413665771, "epoch": 8, "memory": 25074, "step": 59000} +{"lr": 9.038187580296776e-05, "data_time": 0.0012527942657470704, "loss": 0.005688633117824793, "time": 0.7037309885025025, "epoch": 8, "memory": 25074, "step": 59100} +{"lr": 9.038187580296776e-05, "data_time": 0.0016512632369995116, "loss": 0.007635505963116884, "time": 0.7101470708847046, "epoch": 8, "memory": 25074, "step": 59200} +{"lr": 9.038187580296776e-05, "data_time": 0.001399683952331543, "loss": 0.0037501564249396322, "time": 0.7043220520019531, "epoch": 8, "memory": 25074, "step": 59300} +{"lr": 9.038187580296776e-05, "data_time": 0.0013820171356201173, "loss": 0.007279509259387851, "time": 0.790700364112854, "epoch": 8, "memory": 25074, "step": 59400} +{"lr": 9.038187580296776e-05, "data_time": 0.001435232162475586, "loss": 0.0032900267280638216, "time": 0.7489850521087646, "epoch": 8, "memory": 25074, "step": 59500} +{"lr": 9.038187580296776e-05, "data_time": 0.0012386560440063477, "loss": 0.006933847209438682, "time": 0.7489638805389405, "epoch": 8, "memory": 25074, "step": 59600} +{"lr": 9.038187580296776e-05, "data_time": 0.001439070701599121, "loss": 0.010818727081641555, "time": 0.8164791822433471, "epoch": 8, "memory": 25074, "step": 59700} +{"lr": 9.038187580296776e-05, "data_time": 0.0014067888259887695, "loss": 0.010326781868934631, "time": 0.7552563905715942, "epoch": 8, "memory": 25074, "step": 59800} +{"lr": 9.038187580296776e-05, "data_time": 0.0012593030929565429, "loss": 0.0027001675684005023, "time": 0.7751612663269043, "epoch": 8, "memory": 25074, "step": 59900} +{"lr": 9.038187580296776e-05, "data_time": 0.0016253471374511718, "loss": 0.004097736719995737, "time": 0.7999375343322754, "epoch": 8, "memory": 25074, "step": 60000} +{"lr": 9.038187580296776e-05, "data_time": 0.0016825199127197266, "loss": 0.0023204955738037824, "time": 0.7047548770904541, "epoch": 8, "memory": 25074, "step": 60100} +{"lr": 9.038187580296776e-05, "data_time": 0.06462223529815674, "loss": 0.0038289979100227358, "time": 0.7878780364990234, "epoch": 8, "memory": 25074, "step": 60200} +{"lr": 9.038187580296776e-05, "data_time": 0.02077195644378662, "loss": 0.00469278828240931, "time": 0.723414421081543, "epoch": 8, "memory": 25074, "step": 60300} +{"lr": 9.038187580296776e-05, "data_time": 0.053536486625671384, "loss": 0.00837267953902483, "time": 0.7716482639312744, "epoch": 8, "memory": 25074, "step": 60400} +{"lr": 9.038187580296776e-05, "data_time": 0.07309715747833252, "loss": 0.0052996464539319275, "time": 0.7762666463851928, "epoch": 8, "memory": 25074, "step": 60500} +{"lr": 9.038187580296776e-05, "data_time": 0.006291007995605469, "loss": 0.003254408948123455, "time": 0.7316439628601075, "epoch": 8, "memory": 25074, "step": 60600} +{"lr": 9.038187580296776e-05, "data_time": 0.026555657386779785, "loss": 0.0037077505141496657, "time": 0.729297137260437, "epoch": 8, "memory": 25074, "step": 60700} +{"lr": 9.038187580296776e-05, "data_time": 0.022004365921020508, "loss": 0.007412866549566388, "time": 0.7237049818038941, "epoch": 8, "memory": 25074, "step": 60800} +{"lr": 9.038187580296776e-05, "data_time": 0.011216974258422852, "loss": 0.005148720880970359, "time": 0.7157186269760132, "epoch": 8, "memory": 25074, "step": 60900} +{"lr": 9.038187580296776e-05, "data_time": 0.014313530921936036, "loss": 0.006422832049429417, "time": 0.7175225257873535, "epoch": 8, "memory": 25074, "step": 61000} +{"lr": 9.038187580296776e-05, "data_time": 0.029600811004638673, "loss": 0.0033322285860776903, "time": 0.7317031860351563, "epoch": 8, "memory": 25074, "step": 61100} +{"lr": 9.038187580296776e-05, "data_time": 0.0013681888580322266, "loss": 0.00500296363607049, "time": 0.7187703371047973, "epoch": 8, "memory": 25074, "step": 61200} +{"lr": 9.038187580296776e-05, "data_time": 0.0014237403869628907, "loss": 0.011014320328831673, "time": 0.7050602436065674, "epoch": 8, "memory": 25074, "step": 61300} +{"lr": 9.038187580296776e-05, "data_time": 0.018259286880493164, "loss": 0.0056420756503939625, "time": 0.7301955938339233, "epoch": 8, "memory": 25074, "step": 61400} +{"lr": 9.038187580296776e-05, "data_time": 0.09516351222991944, "loss": 0.008991885185241699, "time": 0.7975855588912963, "epoch": 8, "memory": 25074, "step": 61500} +{"lr": 9.038187580296776e-05, "data_time": 0.003044319152832031, "loss": 0.003039657976478338, "time": 0.7058028221130371, "epoch": 8, "memory": 25074, "step": 61600} +{"lr": 9.038187580296776e-05, "data_time": 0.06466200351715087, "loss": 0.006659262720495463, "time": 0.7665573358535767, "epoch": 8, "memory": 25074, "step": 61700} +{"lr": 9.038187580296776e-05, "data_time": 0.03183677196502686, "loss": 0.0022524781059473755, "time": 0.7343443155288696, "epoch": 8, "memory": 25074, "step": 61800} +{"lr": 9.038187580296776e-05, "data_time": 0.04197757244110108, "loss": 0.006802659714594483, "time": 0.7442480087280273, "epoch": 8, "memory": 25074, "step": 61900} +{"lr": 9.038187580296776e-05, "data_time": 0.05157251358032226, "loss": 0.00735700074583292, "time": 0.7542294502258301, "epoch": 8, "memory": 25074, "step": 62000} +{"lr": 9.038187580296776e-05, "data_time": 0.03191332817077637, "loss": 0.003927370300516486, "time": 0.7353523015975952, "epoch": 8, "memory": 25074, "step": 62100} +{"lr": 9.038187580296776e-05, "data_time": 0.031256580352783205, "loss": 0.0026940954849123955, "time": 0.7339596509933471, "epoch": 8, "memory": 25074, "step": 62200} +{"lr": 9.038187580296776e-05, "data_time": 0.049047088623046874, "loss": 0.0020412841346114873, "time": 0.7513560771942138, "epoch": 8, "memory": 25074, "step": 62300} +{"lr": 9.038187580296776e-05, "data_time": 0.003758692741394043, "loss": 0.0038932988536544144, "time": 0.6614237546920776, "epoch": 8, "memory": 25074, "step": 62400} +{"accuracy/top1": 99.96842956542969, "data_time": 0.20667296297409954, "time": 0.5925739793216481, "step": 8} +{"lr": 4.724168609688527e-05, "data_time": 0.0012686729431152343, "loss": 0.0035854040179401636, "time": 0.7392109632492065, "epoch": 9, "memory": 25074, "step": 62500} +{"lr": 4.724168609688527e-05, "data_time": 0.0013946294784545898, "loss": 0.006433201255276799, "time": 0.7699014663696289, "epoch": 9, "memory": 25074, "step": 62600} +{"lr": 4.724168609688527e-05, "data_time": 0.0014280557632446289, "loss": 0.0015518635977059604, "time": 0.8053184509277344, "epoch": 9, "memory": 25074, "step": 62700} +{"lr": 4.724168609688527e-05, "data_time": 0.0014133930206298828, "loss": 0.004280343279242515, "time": 0.7830590009689331, "epoch": 9, "memory": 25074, "step": 62800} +{"lr": 4.724168609688527e-05, "data_time": 0.0012753725051879883, "loss": 0.006203908612951637, "time": 0.7324862480163574, "epoch": 9, "memory": 25074, "step": 62900} +{"lr": 4.724168609688527e-05, "data_time": 0.0012497186660766601, "loss": 0.0019470603205263616, "time": 0.7303324699401855, "epoch": 9, "memory": 25074, "step": 63000} +{"lr": 4.724168609688527e-05, "data_time": 0.0015161514282226562, "loss": 0.0026734769344329834, "time": 0.7312972784042359, "epoch": 9, "memory": 25074, "step": 63100} +{"lr": 4.724168609688527e-05, "data_time": 0.0015453100204467773, "loss": 0.004109944170340896, "time": 0.736474084854126, "epoch": 9, "memory": 25074, "step": 63200} +{"lr": 4.724168609688527e-05, "data_time": 0.001528310775756836, "loss": 0.0030402639880776406, "time": 0.707416033744812, "epoch": 9, "memory": 25074, "step": 63300} +{"lr": 4.724168609688527e-05, "data_time": 0.0012206077575683595, "loss": 0.0050008496269583706, "time": 0.7041665315628052, "epoch": 9, "memory": 25074, "step": 63400} +{"lr": 4.724168609688527e-05, "data_time": 0.0014792680740356445, "loss": 0.003973063547164202, "time": 0.7202898740768433, "epoch": 9, "memory": 25074, "step": 63500} +{"lr": 4.724168609688527e-05, "data_time": 0.0013941287994384765, "loss": 0.0026609358843415975, "time": 0.7256098747253418, "epoch": 9, "memory": 25074, "step": 63600} +{"lr": 4.724168609688527e-05, "data_time": 0.001529383659362793, "loss": 0.0023342340253293516, "time": 0.7169832468032837, "epoch": 9, "memory": 25074, "step": 63700} +{"lr": 4.724168609688527e-05, "data_time": 0.0016998529434204101, "loss": 0.0019028247334063053, "time": 0.7807982683181762, "epoch": 9, "memory": 25074, "step": 63800} +{"lr": 4.724168609688527e-05, "data_time": 0.001434159278869629, "loss": 0.00470067341811955, "time": 0.74498610496521, "epoch": 9, "memory": 25074, "step": 63900} +{"lr": 4.724168609688527e-05, "data_time": 0.0016046285629272461, "loss": 0.004311484005302191, "time": 0.7034945249557495, "epoch": 9, "memory": 25074, "step": 64000} +{"lr": 4.724168609688527e-05, "data_time": 0.0013657093048095703, "loss": 0.005308238137513399, "time": 0.7847606897354126, "epoch": 9, "memory": 25074, "step": 64100} +{"lr": 4.724168609688527e-05, "data_time": 0.001500105857849121, "loss": 0.00266368156298995, "time": 0.7079743385314942, "epoch": 9, "memory": 25074, "step": 64200} +{"lr": 4.724168609688527e-05, "data_time": 0.0012812852859497071, "loss": 0.007512034056708217, "time": 0.7790673971176147, "epoch": 9, "memory": 25074, "step": 64300} +{"lr": 4.724168609688527e-05, "data_time": 0.0015714168548583984, "loss": 0.003466403903439641, "time": 0.7049143552780152, "epoch": 9, "memory": 25074, "step": 64400} +{"lr": 4.724168609688527e-05, "data_time": 0.001565098762512207, "loss": 0.003334543481469154, "time": 0.7285926342010498, "epoch": 9, "memory": 25074, "step": 64500} +{"lr": 4.724168609688527e-05, "data_time": 0.0013058185577392578, "loss": 0.0033261864446103573, "time": 0.73345046043396, "epoch": 9, "memory": 25074, "step": 64600} +{"lr": 4.724168609688527e-05, "data_time": 0.0012994527816772461, "loss": 0.005671628518030047, "time": 0.7037026882171631, "epoch": 9, "memory": 25074, "step": 64700} +{"lr": 4.724168609688527e-05, "data_time": 0.0014406442642211914, "loss": 0.00519427671097219, "time": 0.7044807434082031, "epoch": 9, "memory": 25074, "step": 64800} +{"lr": 4.724168609688527e-05, "data_time": 0.0014109373092651366, "loss": 0.0022163213696330787, "time": 0.723729419708252, "epoch": 9, "memory": 25074, "step": 64900} +{"lr": 4.724168609688527e-05, "data_time": 0.0013831853866577148, "loss": 0.002377808466553688, "time": 0.720691180229187, "epoch": 9, "memory": 25074, "step": 65000} +{"lr": 4.724168609688527e-05, "data_time": 0.0014168977737426757, "loss": 0.0038187152706086634, "time": 0.7150551080703735, "epoch": 9, "memory": 25074, "step": 65100} +{"lr": 4.724168609688527e-05, "data_time": 0.0016774892807006835, "loss": 0.005426485743373633, "time": 0.7044049978256226, "epoch": 9, "memory": 25074, "step": 65200} +{"lr": 4.724168609688527e-05, "data_time": 0.0014931440353393554, "loss": 0.0025714052841067313, "time": 0.7300945520401001, "epoch": 9, "memory": 25074, "step": 65300} +{"lr": 4.724168609688527e-05, "data_time": 0.0013509035110473634, "loss": 0.00381183628924191, "time": 0.7075841426849365, "epoch": 9, "memory": 25074, "step": 65400} +{"lr": 4.724168609688527e-05, "data_time": 0.0014547586441040039, "loss": 0.002330287452787161, "time": 0.8134341478347779, "epoch": 9, "memory": 25074, "step": 65500} +{"lr": 4.724168609688527e-05, "data_time": 0.0013962745666503905, "loss": 0.0011964922305196524, "time": 0.7074713230133056, "epoch": 9, "memory": 25074, "step": 65600} +{"lr": 4.724168609688527e-05, "data_time": 0.0016634702682495118, "loss": 0.003165269223973155, "time": 0.7071994781494141, "epoch": 9, "memory": 25074, "step": 65700} +{"lr": 4.724168609688527e-05, "data_time": 0.00136110782623291, "loss": 0.006068702321499586, "time": 0.7887085676193237, "epoch": 9, "memory": 25074, "step": 65800} +{"lr": 4.724168609688527e-05, "data_time": 0.0014057397842407227, "loss": 0.002544358605518937, "time": 0.7290219068527222, "epoch": 9, "memory": 25074, "step": 65900} +{"lr": 4.724168609688527e-05, "data_time": 0.001516866683959961, "loss": 0.005004653427749872, "time": 0.7046590566635131, "epoch": 9, "memory": 25074, "step": 66000} +{"lr": 4.724168609688527e-05, "data_time": 0.0012632369995117187, "loss": 0.0029656046070158483, "time": 0.7326936483383178, "epoch": 9, "memory": 25074, "step": 66100} +{"lr": 4.724168609688527e-05, "data_time": 0.0013437986373901368, "loss": 0.004099781811237335, "time": 0.7578452825546265, "epoch": 9, "memory": 25074, "step": 66200} +{"lr": 4.724168609688527e-05, "data_time": 0.0014946937561035156, "loss": 0.002212268952280283, "time": 0.7432986497879028, "epoch": 9, "memory": 25074, "step": 66300} +{"lr": 4.724168609688527e-05, "data_time": 0.0017410993576049804, "loss": 0.0047306973487138745, "time": 0.7208121061325073, "epoch": 9, "memory": 25074, "step": 66400} +{"lr": 4.724168609688527e-05, "data_time": 0.0014394998550415039, "loss": 0.005374382436275482, "time": 0.7342538356781005, "epoch": 9, "memory": 25074, "step": 66500} +{"lr": 4.724168609688527e-05, "data_time": 0.0014314889907836915, "loss": 0.004135999083518982, "time": 0.7817184209823609, "epoch": 9, "memory": 25074, "step": 66600} +{"lr": 4.724168609688527e-05, "data_time": 0.0013975381851196289, "loss": 0.002245303150266409, "time": 0.802988600730896, "epoch": 9, "memory": 25074, "step": 66700} +{"lr": 4.724168609688527e-05, "data_time": 0.001355743408203125, "loss": 0.00570099544711411, "time": 0.7101654529571533, "epoch": 9, "memory": 25074, "step": 66800} +{"lr": 4.724168609688527e-05, "data_time": 0.001800203323364258, "loss": 0.002721928060054779, "time": 0.778480052947998, "epoch": 9, "memory": 25074, "step": 66900} +{"lr": 4.724168609688527e-05, "data_time": 0.0014678955078125, "loss": 0.0013445299584418536, "time": 0.7823601961135864, "epoch": 9, "memory": 25074, "step": 67000} +{"lr": 4.724168609688527e-05, "data_time": 0.0014457464218139648, "loss": 0.002576215984299779, "time": 0.7595079421997071, "epoch": 9, "memory": 25074, "step": 67100} +{"lr": 4.724168609688527e-05, "data_time": 0.0014696836471557618, "loss": 0.0024327327962964775, "time": 0.7793433189392089, "epoch": 9, "memory": 25074, "step": 67200} +{"lr": 4.724168609688527e-05, "data_time": 0.0014782428741455078, "loss": 0.0012204103637486695, "time": 0.7778377771377564, "epoch": 9, "memory": 25074, "step": 67300} +{"lr": 4.724168609688527e-05, "data_time": 0.0015259504318237305, "loss": 0.0014354664366692305, "time": 0.8095024824142456, "epoch": 9, "memory": 25074, "step": 67400} +{"lr": 4.724168609688527e-05, "data_time": 0.0014205455780029296, "loss": 0.0023573593702167273, "time": 0.7228098630905151, "epoch": 9, "memory": 25074, "step": 67500} +{"lr": 4.724168609688527e-05, "data_time": 0.001766538619995117, "loss": 0.0034583337604999543, "time": 0.730274248123169, "epoch": 9, "memory": 25074, "step": 67600} +{"lr": 4.724168609688527e-05, "data_time": 0.001429271697998047, "loss": 0.001797714876011014, "time": 0.7654394149780274, "epoch": 9, "memory": 25074, "step": 67700} +{"lr": 4.724168609688527e-05, "data_time": 0.001492762565612793, "loss": 0.003790541971102357, "time": 0.7295792579650879, "epoch": 9, "memory": 25074, "step": 67800} +{"lr": 4.724168609688527e-05, "data_time": 0.001260519027709961, "loss": 0.002734592417255044, "time": 0.750445818901062, "epoch": 9, "memory": 25074, "step": 67900} +{"lr": 4.724168609688527e-05, "data_time": 0.0013107776641845704, "loss": 0.004440054949373007, "time": 0.7778757095336915, "epoch": 9, "memory": 25074, "step": 68000} +{"lr": 4.724168609688527e-05, "data_time": 0.0014533281326293945, "loss": 0.0013329595327377318, "time": 0.7396915912628174, "epoch": 9, "memory": 25074, "step": 68100} +{"lr": 4.724168609688527e-05, "data_time": 0.0017024993896484375, "loss": 0.007108057430014014, "time": 0.7474076986312866, "epoch": 9, "memory": 25074, "step": 68200} +{"lr": 4.724168609688527e-05, "data_time": 0.001279926300048828, "loss": 0.0025010320357978346, "time": 0.7402063846588135, "epoch": 9, "memory": 25074, "step": 68300} +{"lr": 4.724168609688527e-05, "data_time": 0.0012903690338134765, "loss": 0.003660020511597395, "time": 0.7040968656539917, "epoch": 9, "memory": 25074, "step": 68400} +{"lr": 4.724168609688527e-05, "data_time": 0.0013795137405395509, "loss": 0.0022671422455459835, "time": 0.7303619384765625, "epoch": 9, "memory": 25074, "step": 68500} +{"lr": 4.724168609688527e-05, "data_time": 0.04188542366027832, "loss": 0.003003792883828282, "time": 0.7496346950531005, "epoch": 9, "memory": 25074, "step": 68600} +{"lr": 4.724168609688527e-05, "data_time": 0.05385987758636475, "loss": 0.006541739497333765, "time": 0.7570548057556152, "epoch": 9, "memory": 25074, "step": 68700} +{"lr": 4.724168609688527e-05, "data_time": 0.08272285461425781, "loss": 0.001273703435435891, "time": 0.7949224948883057, "epoch": 9, "memory": 25074, "step": 68800} +{"lr": 4.724168609688527e-05, "data_time": 0.026029586791992188, "loss": 0.004725271742790938, "time": 0.7288677930831909, "epoch": 9, "memory": 25074, "step": 68900} +{"lr": 4.724168609688527e-05, "data_time": 0.00137176513671875, "loss": 0.0013643792364746333, "time": 0.7207146883010864, "epoch": 9, "memory": 25074, "step": 69000} +{"lr": 4.724168609688527e-05, "data_time": 0.06450567245483399, "loss": 0.0017586033325642347, "time": 0.766940426826477, "epoch": 9, "memory": 25074, "step": 69100} +{"lr": 4.724168609688527e-05, "data_time": 0.05853328704833984, "loss": 0.004154441645368934, "time": 0.7625686645507812, "epoch": 9, "memory": 25074, "step": 69200} +{"lr": 4.724168609688527e-05, "data_time": 0.08435139656066895, "loss": 0.0008559535257518291, "time": 0.7867570638656616, "epoch": 9, "memory": 25074, "step": 69300} +{"lr": 4.724168609688527e-05, "data_time": 0.06345672607421875, "loss": 0.003086626483127475, "time": 0.7710541486740112, "epoch": 9, "memory": 25074, "step": 69400} +{"lr": 4.724168609688527e-05, "data_time": 0.0013145208358764648, "loss": 0.0027724782936275004, "time": 0.7045375823974609, "epoch": 9, "memory": 25074, "step": 69500} +{"lr": 4.724168609688527e-05, "data_time": 0.0261336088180542, "loss": 0.003240892989560962, "time": 0.7518714189529419, "epoch": 9, "memory": 25074, "step": 69600} +{"lr": 4.724168609688527e-05, "data_time": 0.0404421329498291, "loss": 0.005900827376171947, "time": 0.7500909090042114, "epoch": 9, "memory": 25074, "step": 69700} +{"lr": 4.724168609688527e-05, "data_time": 0.04092690944671631, "loss": 0.0035092404112219812, "time": 0.7556733131408692, "epoch": 9, "memory": 25074, "step": 69800} +{"lr": 4.724168609688527e-05, "data_time": 0.03795361518859863, "loss": 0.001649703411385417, "time": 0.7448919296264649, "epoch": 9, "memory": 25074, "step": 69900} +{"lr": 4.724168609688527e-05, "data_time": 0.05460622310638428, "loss": 0.002138024941086769, "time": 0.7676746368408203, "epoch": 9, "memory": 25074, "step": 70000} +{"lr": 4.724168609688527e-05, "data_time": 0.05661880970001221, "loss": 0.004421891504898667, "time": 0.7733850955963135, "epoch": 9, "memory": 25074, "step": 70100} +{"lr": 4.724168609688527e-05, "data_time": 0.04865431785583496, "loss": 0.0037510984839173034, "time": 0.7081318616867065, "epoch": 9, "memory": 25074, "step": 70200} +{"accuracy/top1": 99.97474670410156, "data_time": 0.21337213235742905, "time": 0.5989095042733585, "step": 9} +{"lr": 1.9543979322445064e-05, "data_time": 0.001468658447265625, "loss": 0.003716249298304319, "time": 0.7956549644470214, "epoch": 10, "memory": 25074, "step": 70300} +{"lr": 1.9543979322445064e-05, "data_time": 0.001401972770690918, "loss": 0.0018167675007134676, "time": 0.7895112991333008, "epoch": 10, "memory": 25074, "step": 70400} +{"lr": 1.9543979322445064e-05, "data_time": 0.001549363136291504, "loss": 0.002099432609975338, "time": 0.7388676881790162, "epoch": 10, "memory": 25074, "step": 70500} +{"lr": 1.9543979322445064e-05, "data_time": 0.035631775856018066, "loss": 0.002846449287608266, "time": 0.7772918462753295, "epoch": 10, "memory": 25074, "step": 70600} +{"lr": 1.9543979322445064e-05, "data_time": 0.05164930820465088, "loss": 0.0027175822760909796, "time": 0.7677915811538696, "epoch": 10, "memory": 25074, "step": 70700} +{"lr": 1.9543979322445064e-05, "data_time": 0.056874418258666994, "loss": 0.0006234078668057918, "time": 0.7584271430969238, "epoch": 10, "memory": 25074, "step": 70800} +{"lr": 1.9543979322445064e-05, "data_time": 0.00845353603363037, "loss": 0.00038216873072087764, "time": 0.710853123664856, "epoch": 10, "memory": 25074, "step": 70900} +{"lr": 1.9543979322445064e-05, "data_time": 0.03818886280059815, "loss": 0.001954852370545268, "time": 0.7504744052886962, "epoch": 10, "memory": 25074, "step": 71000} +{"lr": 1.9543979322445064e-05, "data_time": 0.0012757539749145507, "loss": 0.004034789325669408, "time": 0.7036790370941162, "epoch": 10, "memory": 25074, "step": 71100} +{"lr": 1.9543979322445064e-05, "data_time": 0.0016345977783203125, "loss": 0.0018768099136650563, "time": 0.7170355558395386, "epoch": 10, "memory": 25074, "step": 71200} +{"lr": 1.9543979322445064e-05, "data_time": 0.03760805130004883, "loss": 0.0011197004001587628, "time": 0.7402360439300537, "epoch": 10, "memory": 25074, "step": 71300} +{"lr": 1.9543979322445064e-05, "data_time": 0.036709022521972653, "loss": 0.001394212432205677, "time": 0.7538413286209107, "epoch": 10, "memory": 25074, "step": 71400} +{"lr": 1.9543979322445064e-05, "data_time": 0.008089041709899903, "loss": 0.00416420167312026, "time": 0.7103933572769165, "epoch": 10, "memory": 25074, "step": 71500} +{"lr": 1.9543979322445064e-05, "data_time": 0.027568650245666505, "loss": 0.0006730372086167335, "time": 0.7429495096206665, "epoch": 10, "memory": 25074, "step": 71600} +{"lr": 1.9543979322445064e-05, "data_time": 0.020343494415283204, "loss": 0.00261286492459476, "time": 0.7376342296600342, "epoch": 10, "memory": 25074, "step": 71700} +{"lr": 1.9543979322445064e-05, "data_time": 0.07191858291625977, "loss": 0.004302518954500556, "time": 0.776107668876648, "epoch": 10, "memory": 25074, "step": 71800} +{"lr": 1.9543979322445064e-05, "data_time": 0.030131149291992187, "loss": 0.002136178920045495, "time": 0.7319931983947754, "epoch": 10, "memory": 25074, "step": 71900} +{"lr": 1.9543979322445064e-05, "data_time": 0.05597836971282959, "loss": 0.001320331171154976, "time": 0.758454155921936, "epoch": 10, "memory": 25074, "step": 72000} +{"lr": 1.9543979322445064e-05, "data_time": 0.07893037796020508, "loss": 0.0033421942498534918, "time": 0.7960315227508545, "epoch": 10, "memory": 25074, "step": 72100} +{"lr": 1.9543979322445064e-05, "data_time": 0.06062870025634766, "loss": 0.003074402594938874, "time": 0.7629699468612671, "epoch": 10, "memory": 25074, "step": 72200} +{"lr": 1.9543979322445064e-05, "data_time": 0.0372807502746582, "loss": 0.001983436569571495, "time": 0.7575109004974365, "epoch": 10, "memory": 25074, "step": 72300} +{"lr": 1.9543979322445064e-05, "data_time": 0.054410362243652345, "loss": 0.0016810684464871883, "time": 0.7565388202667236, "epoch": 10, "memory": 25074, "step": 72400} +{"lr": 1.9543979322445064e-05, "data_time": 0.057735109329223634, "loss": 0.0025253639556467532, "time": 0.7713340520858765, "epoch": 10, "memory": 25074, "step": 72500} +{"lr": 1.9543979322445064e-05, "data_time": 0.03789963722229004, "loss": 0.0026211796794086695, "time": 0.7408131837844849, "epoch": 10, "memory": 25074, "step": 72600} +{"lr": 1.9543979322445064e-05, "data_time": 0.12151339054107665, "loss": 0.00252187754958868, "time": 0.8321881532669068, "epoch": 10, "memory": 25074, "step": 72700} +{"lr": 1.9543979322445064e-05, "data_time": 0.07931854724884033, "loss": 0.004591597151011229, "time": 0.7812834739685058, "epoch": 10, "memory": 25074, "step": 72800} +{"lr": 1.9543979322445064e-05, "data_time": 0.015970945358276367, "loss": 0.0006349658593535423, "time": 0.7358031034469604, "epoch": 10, "memory": 25074, "step": 72900} +{"lr": 1.9543979322445064e-05, "data_time": 0.06624820232391357, "loss": 0.0025834666099399327, "time": 0.7701605796813965, "epoch": 10, "memory": 25074, "step": 73000} +{"lr": 1.9543979322445064e-05, "data_time": 0.057169413566589354, "loss": 0.002863296028226614, "time": 0.7678876876831054, "epoch": 10, "memory": 25074, "step": 73100} +{"lr": 1.9543979322445064e-05, "data_time": 0.058753585815429686, "loss": 0.0033069149125367404, "time": 0.761167311668396, "epoch": 10, "memory": 25074, "step": 73200} +{"lr": 1.9543979322445064e-05, "data_time": 0.10259363651275635, "loss": 0.0012726620770990849, "time": 0.8150048732757569, "epoch": 10, "memory": 25074, "step": 73300} +{"lr": 1.9543979322445064e-05, "data_time": 0.08943588733673095, "loss": 0.0025894547812640666, "time": 0.7915759325027466, "epoch": 10, "memory": 25074, "step": 73400} +{"lr": 1.9543979322445064e-05, "data_time": 0.11913619041442872, "loss": 0.001922428421676159, "time": 0.8215875148773193, "epoch": 10, "memory": 25074, "step": 73500} +{"lr": 1.9543979322445064e-05, "data_time": 0.03992674350738525, "loss": 0.0023916376288980246, "time": 0.7417946338653565, "epoch": 10, "memory": 25074, "step": 73600} +{"lr": 1.9543979322445064e-05, "data_time": 0.08811109066009522, "loss": 0.0004283975809812546, "time": 0.7922694683074951, "epoch": 10, "memory": 25074, "step": 73700} +{"lr": 1.9543979322445064e-05, "data_time": 0.07816421985626221, "loss": 0.002233076561242342, "time": 0.7810037851333618, "epoch": 10, "memory": 25074, "step": 73800} +{"lr": 1.9543979322445064e-05, "data_time": 0.0027591943740844726, "loss": 0.001962736435234547, "time": 0.7198577642440795, "epoch": 10, "memory": 25074, "step": 73900} +{"lr": 1.9543979322445064e-05, "data_time": 0.026323866844177247, "loss": 0.004134145192801952, "time": 0.7285466909408569, "epoch": 10, "memory": 25074, "step": 74000} +{"lr": 1.9543979322445064e-05, "data_time": 0.09715259075164795, "loss": 0.002139729354530573, "time": 0.8003833055496216, "epoch": 10, "memory": 25074, "step": 74100} +{"lr": 1.9543979322445064e-05, "data_time": 0.010705947875976562, "loss": 0.0014401759952306747, "time": 0.7128584146499634, "epoch": 10, "memory": 25074, "step": 74200} +{"lr": 1.9543979322445064e-05, "data_time": 0.04110572338104248, "loss": 0.0005889842286705971, "time": 0.7432945489883422, "epoch": 10, "memory": 25074, "step": 74300} +{"lr": 1.9543979322445064e-05, "data_time": 0.039203739166259764, "loss": 0.001075806003063917, "time": 0.7423350572586059, "epoch": 10, "memory": 25074, "step": 74400} +{"lr": 1.9543979322445064e-05, "data_time": 0.08183917999267579, "loss": 0.0017675799317657948, "time": 0.7847230434417725, "epoch": 10, "memory": 25074, "step": 74500} +{"lr": 1.9543979322445064e-05, "data_time": 0.02958788871765137, "loss": 0.0019665206782519817, "time": 0.7333899736404419, "epoch": 10, "memory": 25074, "step": 74600} +{"lr": 1.9543979322445064e-05, "data_time": 0.033285999298095705, "loss": 0.004259195365011692, "time": 0.7361727237701416, "epoch": 10, "memory": 25074, "step": 74700} +{"lr": 1.9543979322445064e-05, "data_time": 0.08124017715454102, "loss": 0.001425787154585123, "time": 0.7836496114730835, "epoch": 10, "memory": 25074, "step": 74800} +{"lr": 1.9543979322445064e-05, "data_time": 0.03210391998291016, "loss": 0.0009439696557819843, "time": 0.7359946966171265, "epoch": 10, "memory": 25074, "step": 74900} +{"lr": 1.9543979322445064e-05, "data_time": 0.07245752811431885, "loss": 0.0018797555938363075, "time": 0.7759171962738037, "epoch": 10, "memory": 25074, "step": 75000} +{"lr": 1.9543979322445064e-05, "data_time": 0.008467388153076173, "loss": 0.005092905554920435, "time": 0.7110937356948852, "epoch": 10, "memory": 25074, "step": 75100} +{"lr": 1.9543979322445064e-05, "data_time": 0.08542578220367432, "loss": 0.0020634511951357127, "time": 0.7878558397293091, "epoch": 10, "memory": 25074, "step": 75200} +{"lr": 1.9543979322445064e-05, "data_time": 0.05697500705718994, "loss": 0.0018645947333425284, "time": 0.7608028173446655, "epoch": 10, "memory": 25074, "step": 75300} +{"lr": 1.9543979322445064e-05, "data_time": 0.014886569976806641, "loss": 0.0032790504395961762, "time": 0.7179263353347778, "epoch": 10, "memory": 25074, "step": 75400} +{"lr": 1.9543979322445064e-05, "data_time": 0.0480043888092041, "loss": 0.00141446515917778, "time": 0.7499783039093018, "epoch": 10, "memory": 25074, "step": 75500} +{"lr": 1.9543979322445064e-05, "data_time": 0.023751306533813476, "loss": 0.0014892212580889463, "time": 0.7263204336166382, "epoch": 10, "memory": 25074, "step": 75600} +{"lr": 1.9543979322445064e-05, "data_time": 0.03580174446105957, "loss": 0.004743988672271371, "time": 0.7381531715393066, "epoch": 10, "memory": 25074, "step": 75700} +{"lr": 1.9543979322445064e-05, "data_time": 0.05883114337921143, "loss": 0.001213197922334075, "time": 0.7676050424575805, "epoch": 10, "memory": 25074, "step": 75800} +{"lr": 1.9543979322445064e-05, "data_time": 0.021689653396606445, "loss": 0.00039984621107578275, "time": 0.7252687931060791, "epoch": 10, "memory": 25074, "step": 75900} +{"lr": 1.9543979322445064e-05, "data_time": 0.06327095031738281, "loss": 0.0034788088873028756, "time": 0.7730864524841309, "epoch": 10, "memory": 25074, "step": 76000} +{"lr": 1.9543979322445064e-05, "data_time": 0.21071991920471192, "loss": 0.00129951941780746, "time": 0.9303645133972168, "epoch": 10, "memory": 25074, "step": 76100} +{"lr": 1.9543979322445064e-05, "data_time": 0.13052523136138916, "loss": 0.0020625774282962083, "time": 0.8474969387054443, "epoch": 10, "memory": 25074, "step": 76200} +{"lr": 1.9543979322445064e-05, "data_time": 0.20942361354827882, "loss": 0.002098038839176297, "time": 0.9318359613418579, "epoch": 10, "memory": 25074, "step": 76300} +{"lr": 1.9543979322445064e-05, "data_time": 0.09960784912109374, "loss": 0.0014654813334345818, "time": 0.8268126964569091, "epoch": 10, "memory": 25074, "step": 76400} +{"lr": 1.9543979322445064e-05, "data_time": 0.02778441905975342, "loss": 0.0019784611184149982, "time": 0.7307383298873902, "epoch": 10, "memory": 25074, "step": 76500} +{"lr": 1.9543979322445064e-05, "data_time": 0.10081713199615479, "loss": 0.0020369043573737144, "time": 0.8450204133987427, "epoch": 10, "memory": 25074, "step": 76600} +{"lr": 1.9543979322445064e-05, "data_time": 0.1534421443939209, "loss": 0.000898544630035758, "time": 0.871632170677185, "epoch": 10, "memory": 25074, "step": 76700} +{"lr": 1.9543979322445064e-05, "data_time": 0.09541425704956055, "loss": 0.0007972294464707374, "time": 0.8196889877319335, "epoch": 10, "memory": 25074, "step": 76800} +{"lr": 1.9543979322445064e-05, "data_time": 0.052926373481750486, "loss": 0.0033644529525190593, "time": 0.7564990282058716, "epoch": 10, "memory": 25074, "step": 76900} +{"lr": 1.9543979322445064e-05, "data_time": 0.041558361053466795, "loss": 0.002528955042362213, "time": 0.7693127393722534, "epoch": 10, "memory": 25074, "step": 77000} +{"lr": 1.9543979322445064e-05, "data_time": 0.01842055320739746, "loss": 0.0011274866294115782, "time": 0.7204957723617553, "epoch": 10, "memory": 25074, "step": 77100} +{"lr": 1.9543979322445064e-05, "data_time": 0.10217568874359131, "loss": 0.00392243810929358, "time": 0.805843448638916, "epoch": 10, "memory": 25074, "step": 77200} +{"lr": 1.9543979322445064e-05, "data_time": 0.159128999710083, "loss": 0.001156916655600071, "time": 0.8737040758132935, "epoch": 10, "memory": 25074, "step": 77300} +{"lr": 1.9543979322445064e-05, "data_time": 0.16430633068084716, "loss": 0.0018640908412635327, "time": 0.9029998779296875, "epoch": 10, "memory": 25074, "step": 77400} +{"lr": 1.9543979322445064e-05, "data_time": 0.07460556030273438, "loss": 0.0014721516519784928, "time": 0.7803693056106568, "epoch": 10, "memory": 25074, "step": 77500} +{"lr": 1.9543979322445064e-05, "data_time": 0.03137850761413574, "loss": 0.0008515995927155018, "time": 0.753780722618103, "epoch": 10, "memory": 25074, "step": 77600} +{"lr": 1.9543979322445064e-05, "data_time": 0.05947234630584717, "loss": 0.00241722147911787, "time": 0.7621008634567261, "epoch": 10, "memory": 25074, "step": 77700} +{"lr": 1.9543979322445064e-05, "data_time": 0.14171135425567627, "loss": 0.004039214039221406, "time": 0.9070079088211059, "epoch": 10, "memory": 25074, "step": 77800} +{"lr": 1.9543979322445064e-05, "data_time": 0.17052626609802246, "loss": 0.002315824432298541, "time": 0.8927216768264771, "epoch": 10, "memory": 25074, "step": 77900} +{"lr": 1.9543979322445064e-05, "data_time": 0.12948236465454102, "loss": 0.0029011948667175603, "time": 0.8018013715744019, "epoch": 10, "memory": 25074, "step": 78000} +{"accuracy/top1": 99.98737335205078, "data_time": 0.31446392395917105, "time": 0.7018048202290255, "step": 10} diff --git a/convnext_small_4xb256_fake7m-lr4e-4/20230601_193207/vis_data/config.py b/convnext_small_4xb256_fake7m-lr4e-4/20230601_193207/vis_data/config.py new file mode 100644 index 0000000000000000000000000000000000000000..0e6e101fa3314a1c359ce105454f400d187c5c3d --- /dev/null +++ b/convnext_small_4xb256_fake7m-lr4e-4/20230601_193207/vis_data/config.py @@ -0,0 +1,203 @@ +optim_wrapper = dict( + optimizer=dict( + type='AdamW', + lr=0.0004, + weight_decay=0.05, + eps=1e-08, + betas=(0.9, 0.999), + _scope_='mmpretrain'), + paramwise_cfg=dict( + norm_decay_mult=0.0, + bias_decay_mult=0.0, + flat_decay_mult=0.0, + custom_keys=dict({ + '.absolute_pos_embed': dict(decay_mult=0.0), + '.relative_position_bias_table': dict(decay_mult=0.0) + })), + type='AmpOptimWrapper', + dtype='bfloat16', + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=True, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=4096) +model = dict( + type='ImageClassifier', + backbone=dict(type='ConvNeXt', arch='small', drop_path_rate=0.4), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=768, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + init_cfg=None), + init_cfg=dict( + type='TruncNormal', layer=['Conv2d', 'Linear'], std=0.02, bias=0.0), + train_cfg=None) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stablediffusionV1-5R2-dpmsolver-25-5m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/if-dpmsolver++-50-20w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stablediffusionV2-1-dpmsolver-25-190w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/convnext_small_4xb256_fake7m-lr4e-4' diff --git a/convnext_small_4xb256_fake7m-lr4e-4/20230601_193207/vis_data/events.out.tfevents.1685619137.SH-IDC1-10-140-24-134.106918.0 b/convnext_small_4xb256_fake7m-lr4e-4/20230601_193207/vis_data/events.out.tfevents.1685619137.SH-IDC1-10-140-24-134.106918.0 new file mode 100644 index 0000000000000000000000000000000000000000..a9bcb7aa4f039655d1b8c449e52c0ed9a76622d5 --- /dev/null +++ b/convnext_small_4xb256_fake7m-lr4e-4/20230601_193207/vis_data/events.out.tfevents.1685619137.SH-IDC1-10-140-24-134.106918.0 @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:3b42feb095043b8cc6033e267f3e9272cf480806acd8e4d63ed67c1d511e9976 +size 926714 diff --git a/convnext_small_4xb256_fake7m-lr4e-4/20230601_193207/vis_data/scalars.json b/convnext_small_4xb256_fake7m-lr4e-4/20230601_193207/vis_data/scalars.json new file mode 100644 index 0000000000000000000000000000000000000000..be42ea66000a05cca1a5afe3e0bd445ce348ebc2 --- /dev/null +++ b/convnext_small_4xb256_fake7m-lr4e-4/20230601_193207/vis_data/scalars.json @@ -0,0 +1,790 @@ +{"lr": 0.0004, "data_time": 2.6907129287719727, "loss": 0.5509223461151123, "time": 4.028603577613831, "epoch": 1, "memory": 25071, "step": 100} +{"lr": 0.0004, "data_time": 3.903346085548401, "loss": 0.5122920751571656, "time": 4.696592187881469, "epoch": 1, "memory": 25071, "step": 200} +{"lr": 0.0004, "data_time": 3.581833267211914, "loss": 0.4802056074142456, "time": 4.289185667037964, "epoch": 1, "memory": 25071, "step": 300} +{"lr": 0.0004, "data_time": 3.3729814291000366, "loss": 0.47626489102840425, "time": 4.466727089881897, "epoch": 1, "memory": 25071, "step": 400} +{"lr": 0.0004, "data_time": 3.152855086326599, "loss": 0.46141456365585326, "time": 4.480378150939941, "epoch": 1, "memory": 25071, "step": 500} +{"lr": 0.0004, "data_time": 3.7332908153533935, "loss": 0.45464659929275514, "time": 4.440839076042176, "epoch": 1, "memory": 25071, "step": 600} +{"lr": 0.0004, "data_time": 3.7476879358291626, "loss": 0.4438972115516663, "time": 4.455186796188355, "epoch": 1, "memory": 25071, "step": 700} +{"lr": 0.0004, "data_time": 2.997472071647644, "loss": 0.46163980662822723, "time": 3.7071758985519407, "epoch": 1, "memory": 25071, "step": 800} +{"lr": 0.0004, "data_time": 3.3110134601593018, "loss": 0.4082850754261017, "time": 4.019241809844971, "epoch": 1, "memory": 25071, "step": 900} +{"lr": 0.0004, "data_time": 0.0013388633728027345, "loss": 0.4357276916503906, "time": 0.7083222389221191, "epoch": 1, "memory": 25071, "step": 1000} +{"lr": 0.0004, "data_time": 0.04599974155426025, "loss": 0.393719083070755, "time": 0.7553951740264893, "epoch": 1, "memory": 25071, "step": 1100} +{"lr": 0.0004, "data_time": 0.0013998746871948242, "loss": 0.40593518018722535, "time": 0.7048080205917359, "epoch": 1, "memory": 25071, "step": 1200} +{"lr": 0.0004, "data_time": 0.0012827157974243165, "loss": 0.4099356472492218, "time": 0.7289187192916871, "epoch": 1, "memory": 25071, "step": 1300} +{"lr": 0.0004, "data_time": 0.0013564825057983398, "loss": 0.4108930230140686, "time": 0.8211391925811767, "epoch": 1, "memory": 25071, "step": 1400} +{"lr": 0.0004, "data_time": 0.0013655424118041992, "loss": 0.39929743111133575, "time": 0.7272112131118774, "epoch": 1, "memory": 25071, "step": 1500} +{"lr": 0.0004, "data_time": 0.18983705043792726, "loss": 0.39867955446243286, "time": 0.8959766387939453, "epoch": 1, "memory": 25071, "step": 1600} +{"lr": 0.0004, "data_time": 0.0012964248657226563, "loss": 0.3719035655260086, "time": 0.854664397239685, "epoch": 1, "memory": 25071, "step": 1700} +{"lr": 0.0004, "data_time": 0.0014933109283447265, "loss": 0.3636783301830292, "time": 0.8932012319564819, "epoch": 1, "memory": 25071, "step": 1800} +{"lr": 0.0004, "data_time": 0.0014470577239990234, "loss": 0.37819368243217466, "time": 0.7350067853927612, "epoch": 1, "memory": 25071, "step": 1900} +{"lr": 0.0004, "data_time": 0.0013707876205444336, "loss": 0.3699472784996033, "time": 0.7687848091125489, "epoch": 1, "memory": 25071, "step": 2000} +{"lr": 0.0004, "data_time": 0.001299571990966797, "loss": 0.3759961068630219, "time": 0.792644739151001, "epoch": 1, "memory": 25071, "step": 2100} +{"lr": 0.0004, "data_time": 0.0014440059661865235, "loss": 0.37775363922119143, "time": 0.7185564517974854, "epoch": 1, "memory": 25071, "step": 2200} +{"lr": 0.0004, "data_time": 0.0014520883560180664, "loss": 0.3431844234466553, "time": 0.7634948253631592, "epoch": 1, "memory": 25071, "step": 2300} +{"lr": 0.0004, "data_time": 0.0012256860733032226, "loss": 0.3327546685934067, "time": 0.7808838129043579, "epoch": 1, "memory": 25071, "step": 2400} +{"lr": 0.0004, "data_time": 0.0012555122375488281, "loss": 0.3289307564496994, "time": 0.8116376638412476, "epoch": 1, "memory": 25071, "step": 2500} +{"lr": 0.0004, "data_time": 0.0012950897216796875, "loss": 0.33132781088352203, "time": 1.0313302040100099, "epoch": 1, "memory": 25071, "step": 2600} +{"lr": 0.0004, "data_time": 0.001563262939453125, "loss": 0.34159435629844664, "time": 1.0104379415512086, "epoch": 1, "memory": 25071, "step": 2700} +{"lr": 0.0004, "data_time": 0.0014277935028076173, "loss": 0.3161256194114685, "time": 0.9495459079742432, "epoch": 1, "memory": 25071, "step": 2800} +{"lr": 0.0004, "data_time": 0.001340341567993164, "loss": 0.3131725758314133, "time": 0.7670013427734375, "epoch": 1, "memory": 25071, "step": 2900} +{"lr": 0.0004, "data_time": 0.0013190746307373048, "loss": 0.33015253841876985, "time": 1.0442669153213502, "epoch": 1, "memory": 25071, "step": 3000} +{"lr": 0.0004, "data_time": 0.027217388153076172, "loss": 0.3198246330022812, "time": 0.8309589385986328, "epoch": 1, "memory": 25071, "step": 3100} +{"lr": 0.0004, "data_time": 0.0014668941497802735, "loss": 0.29642116725444795, "time": 0.7057172536849976, "epoch": 1, "memory": 25071, "step": 3200} +{"lr": 0.0004, "data_time": 0.001323866844177246, "loss": 0.3204584240913391, "time": 0.719047999382019, "epoch": 1, "memory": 25071, "step": 3300} +{"lr": 0.0004, "data_time": 0.0013828277587890625, "loss": 0.3120553225278854, "time": 0.7054277658462524, "epoch": 1, "memory": 25071, "step": 3400} +{"lr": 0.0004, "data_time": 0.0016828298568725586, "loss": 0.3116033345460892, "time": 0.7053396701812744, "epoch": 1, "memory": 25071, "step": 3500} +{"lr": 0.0004, "data_time": 0.0012962818145751953, "loss": 0.2981396973133087, "time": 0.7053417921066284, "epoch": 1, "memory": 25071, "step": 3600} +{"lr": 0.0004, "data_time": 0.0013994693756103516, "loss": 0.3141171634197235, "time": 0.7148031711578369, "epoch": 1, "memory": 25071, "step": 3700} +{"lr": 0.0004, "data_time": 0.001499032974243164, "loss": 0.29559786021709444, "time": 0.7061149120330811, "epoch": 1, "memory": 25071, "step": 3800} +{"lr": 0.0004, "data_time": 0.0014240741729736328, "loss": 0.29358674883842467, "time": 0.7244684934616089, "epoch": 1, "memory": 25071, "step": 3900} +{"lr": 0.0004, "data_time": 0.001511836051940918, "loss": 0.25467085242271426, "time": 0.7070822238922119, "epoch": 1, "memory": 25071, "step": 4000} +{"lr": 0.0004, "data_time": 0.0014264106750488282, "loss": 0.2919668465852737, "time": 0.7236125946044922, "epoch": 1, "memory": 25071, "step": 4100} +{"lr": 0.0004, "data_time": 0.001343989372253418, "loss": 0.2884243577718735, "time": 0.7079657316207886, "epoch": 1, "memory": 25071, "step": 4200} +{"lr": 0.0004, "data_time": 0.0014359235763549804, "loss": 0.2842044696211815, "time": 0.7134523868560791, "epoch": 1, "memory": 25071, "step": 4300} +{"lr": 0.0004, "data_time": 0.0015362739562988282, "loss": 0.2771912842988968, "time": 0.7043992519378662, "epoch": 1, "memory": 25071, "step": 4400} +{"lr": 0.0004, "data_time": 0.0013992071151733398, "loss": 0.2899408534169197, "time": 0.7200767517089843, "epoch": 1, "memory": 25071, "step": 4500} +{"lr": 0.0004, "data_time": 0.0013010025024414063, "loss": 0.26940833032131195, "time": 0.7055598020553588, "epoch": 1, "memory": 25071, "step": 4600} +{"lr": 0.0004, "data_time": 0.0014023542404174804, "loss": 0.2595094949007034, "time": 0.7550057649612427, "epoch": 1, "memory": 25071, "step": 4700} +{"lr": 0.0004, "data_time": 0.0012409687042236328, "loss": 0.26932478696107864, "time": 0.7812519311904907, "epoch": 1, "memory": 25071, "step": 4800} +{"lr": 0.0004, "data_time": 0.0014226198196411132, "loss": 0.2865415155887604, "time": 0.718199610710144, "epoch": 1, "memory": 25071, "step": 4900} +{"lr": 0.0004, "data_time": 0.0013959169387817382, "loss": 0.26076784282922744, "time": 0.7069357872009278, "epoch": 1, "memory": 25071, "step": 5000} +{"lr": 0.0004, "data_time": 0.0012188196182250977, "loss": 0.27880939841270447, "time": 0.712481164932251, "epoch": 1, "memory": 25071, "step": 5100} +{"lr": 0.0004, "data_time": 0.001408553123474121, "loss": 0.26313952952623365, "time": 0.7054989099502563, "epoch": 1, "memory": 25071, "step": 5200} +{"lr": 0.0004, "data_time": 0.0013321876525878907, "loss": 0.257980714738369, "time": 0.7458206415176392, "epoch": 1, "memory": 25071, "step": 5300} +{"lr": 0.0004, "data_time": 0.0014032602310180664, "loss": 0.23871318995952606, "time": 0.7410083770751953, "epoch": 1, "memory": 25071, "step": 5400} +{"lr": 0.0004, "data_time": 0.0012554407119750976, "loss": 0.25315780490636824, "time": 0.715627908706665, "epoch": 1, "memory": 25071, "step": 5500} +{"lr": 0.0004, "data_time": 0.0015192270278930665, "loss": 0.22020687609910966, "time": 0.7893864393234253, "epoch": 1, "memory": 25071, "step": 5600} +{"lr": 0.0004, "data_time": 0.0013898849487304688, "loss": 0.25799562931060793, "time": 0.7222212553024292, "epoch": 1, "memory": 25071, "step": 5700} +{"lr": 0.0004, "data_time": 0.0014517545700073243, "loss": 0.26193687617778777, "time": 0.705207109451294, "epoch": 1, "memory": 25071, "step": 5800} +{"lr": 0.0004, "data_time": 0.0014998435974121094, "loss": 0.23978979438543319, "time": 0.7064939975738526, "epoch": 1, "memory": 25071, "step": 5900} +{"lr": 0.0004, "data_time": 0.0018176555633544922, "loss": 0.27151983976364136, "time": 0.7038760662078858, "epoch": 1, "memory": 25071, "step": 6000} +{"lr": 0.0004, "data_time": 0.0015011310577392578, "loss": 0.23080928325653077, "time": 0.7212326526641846, "epoch": 1, "memory": 25071, "step": 6100} +{"lr": 0.0004, "data_time": 0.0012937307357788086, "loss": 0.24490288645029068, "time": 0.720430064201355, "epoch": 1, "memory": 25071, "step": 6200} +{"lr": 0.0004, "data_time": 0.0011920690536499023, "loss": 0.23940146416425706, "time": 0.7056973457336426, "epoch": 1, "memory": 25071, "step": 6300} +{"lr": 0.0004, "data_time": 0.001397109031677246, "loss": 0.2349616140127182, "time": 0.7072972536087037, "epoch": 1, "memory": 25071, "step": 6400} +{"lr": 0.0004, "data_time": 0.0014444112777709961, "loss": 0.21422109603881836, "time": 0.7065572261810302, "epoch": 1, "memory": 25071, "step": 6500} +{"lr": 0.0004, "data_time": 0.0013858795166015625, "loss": 0.24266654700040818, "time": 0.7030148029327392, "epoch": 1, "memory": 25071, "step": 6600} +{"lr": 0.0004, "data_time": 0.0014795541763305664, "loss": 0.23327862322330475, "time": 0.7081075668334961, "epoch": 1, "memory": 25071, "step": 6700} +{"lr": 0.0004, "data_time": 0.0014479637145996093, "loss": 0.23950548470020294, "time": 0.7077049016952515, "epoch": 1, "memory": 25071, "step": 6800} +{"lr": 0.0004, "data_time": 0.0014442205429077148, "loss": 0.2277327612042427, "time": 0.7036959648132324, "epoch": 1, "memory": 25071, "step": 6900} +{"lr": 0.0004, "data_time": 0.0014896869659423828, "loss": 0.2067130535840988, "time": 0.7160272836685181, "epoch": 1, "memory": 25071, "step": 7000} +{"lr": 0.0004, "data_time": 0.0013399362564086915, "loss": 0.2089599072933197, "time": 0.7053474426269531, "epoch": 1, "memory": 25071, "step": 7100} +{"lr": 0.0004, "data_time": 0.0013405799865722657, "loss": 0.21773237138986587, "time": 0.7126405477523804, "epoch": 1, "memory": 25071, "step": 7200} +{"lr": 0.0004, "data_time": 0.0014854192733764649, "loss": 0.21485509276390075, "time": 0.7069435834884643, "epoch": 1, "memory": 25071, "step": 7300} +{"lr": 0.0004, "data_time": 0.001581096649169922, "loss": 0.21211579144001008, "time": 0.7383034706115723, "epoch": 1, "memory": 25071, "step": 7400} +{"lr": 0.0004, "data_time": 0.0014016866683959962, "loss": 0.21462307274341583, "time": 0.7065046787261963, "epoch": 1, "memory": 25071, "step": 7500} +{"lr": 0.0004, "data_time": 0.0016673803329467773, "loss": 0.19861916899681092, "time": 0.7191998958587646, "epoch": 1, "memory": 25071, "step": 7600} +{"lr": 0.0004, "data_time": 0.0014283895492553712, "loss": 0.22524501979351044, "time": 0.7050751447677612, "epoch": 1, "memory": 25071, "step": 7700} +{"lr": 0.0004, "data_time": 0.0010645389556884766, "loss": 0.2027382418513298, "time": 0.6594531059265136, "epoch": 1, "memory": 25071, "step": 7800} +{"accuracy/top1": 82.74729919433594, "data_time": 0.26189135015010834, "time": 0.7810759395360947, "step": 1} +{"lr": 0.00039045602067755495, "data_time": 0.001395416259765625, "loss": 0.1985473230481148, "time": 0.7228825330734253, "epoch": 2, "memory": 25074, "step": 7900} +{"lr": 0.00039045602067755495, "data_time": 0.00175323486328125, "loss": 0.19168051928281785, "time": 0.7411270856857299, "epoch": 2, "memory": 25074, "step": 8000} +{"lr": 0.00039045602067755495, "data_time": 0.0011951208114624023, "loss": 0.19834990948438644, "time": 0.722902488708496, "epoch": 2, "memory": 25074, "step": 8100} +{"lr": 0.00039045602067755495, "data_time": 0.0012741327285766602, "loss": 0.19371471256017686, "time": 0.7558099746704101, "epoch": 2, "memory": 25074, "step": 8200} +{"lr": 0.00039045602067755495, "data_time": 0.0013913631439208985, "loss": 0.1958928033709526, "time": 0.7217024803161621, "epoch": 2, "memory": 25074, "step": 8300} +{"lr": 0.00039045602067755495, "data_time": 0.0012528419494628907, "loss": 0.18579413443803788, "time": 0.8182249307632447, "epoch": 2, "memory": 25074, "step": 8400} +{"lr": 0.00039045602067755495, "data_time": 0.001584482192993164, "loss": 0.20487104803323747, "time": 0.7238365411758423, "epoch": 2, "memory": 25074, "step": 8500} +{"lr": 0.00039045602067755495, "data_time": 0.01669943332672119, "loss": 0.19487895220518112, "time": 0.7289236545562744, "epoch": 2, "memory": 25074, "step": 8600} +{"lr": 0.00039045602067755495, "data_time": 0.0015350580215454102, "loss": 0.17102978229522706, "time": 0.7226077556610108, "epoch": 2, "memory": 25074, "step": 8700} +{"lr": 0.00039045602067755495, "data_time": 0.0017715930938720704, "loss": 0.19943539649248124, "time": 0.7214926242828369, "epoch": 2, "memory": 25074, "step": 8800} +{"lr": 0.00039045602067755495, "data_time": 0.0017403364181518555, "loss": 0.19120719283819199, "time": 0.739957618713379, "epoch": 2, "memory": 25074, "step": 8900} +{"lr": 0.00039045602067755495, "data_time": 0.0018882274627685547, "loss": 0.1867499127984047, "time": 0.7092678785324097, "epoch": 2, "memory": 25074, "step": 9000} +{"lr": 0.00039045602067755495, "data_time": 0.0019953250885009766, "loss": 0.16988205015659333, "time": 0.7357144355773926, "epoch": 2, "memory": 25074, "step": 9100} +{"lr": 0.00039045602067755495, "data_time": 0.05246288776397705, "loss": 0.18630445897579193, "time": 0.7638375520706177, "epoch": 2, "memory": 25074, "step": 9200} +{"lr": 0.00039045602067755495, "data_time": 0.046308159828186035, "loss": 0.18053615242242813, "time": 0.7948039054870606, "epoch": 2, "memory": 25074, "step": 9300} +{"lr": 0.00039045602067755495, "data_time": 0.05054788589477539, "loss": 0.17141434997320176, "time": 0.7714882373809815, "epoch": 2, "memory": 25074, "step": 9400} +{"lr": 0.00039045602067755495, "data_time": 0.05095982551574707, "loss": 0.20186308026313782, "time": 0.7825620412826538, "epoch": 2, "memory": 25074, "step": 9500} +{"lr": 0.00039045602067755495, "data_time": 0.07320964336395264, "loss": 0.15854080840945245, "time": 0.7893943071365357, "epoch": 2, "memory": 25074, "step": 9600} +{"lr": 0.00039045602067755495, "data_time": 0.00156402587890625, "loss": 0.17121621072292328, "time": 0.7229015350341796, "epoch": 2, "memory": 25074, "step": 9700} +{"lr": 0.00039045602067755495, "data_time": 0.05058057308197021, "loss": 0.17450278401374816, "time": 0.7707611083984375, "epoch": 2, "memory": 25074, "step": 9800} +{"lr": 0.00039045602067755495, "data_time": 0.0055065155029296875, "loss": 0.17037517130374907, "time": 0.7292756557464599, "epoch": 2, "memory": 25074, "step": 9900} +{"lr": 0.00039045602067755495, "data_time": 0.0013831138610839843, "loss": 0.19259472638368608, "time": 0.7221044778823853, "epoch": 2, "memory": 25074, "step": 10000} +{"lr": 0.00039045602067755495, "data_time": 0.0013656139373779297, "loss": 0.15048861056566237, "time": 0.7273980379104614, "epoch": 2, "memory": 25074, "step": 10100} +{"lr": 0.00039045602067755495, "data_time": 0.0013884544372558595, "loss": 0.16968381106853486, "time": 0.7262316465377807, "epoch": 2, "memory": 25074, "step": 10200} +{"lr": 0.00039045602067755495, "data_time": 0.0018139362335205078, "loss": 0.15536620318889618, "time": 0.7153719186782836, "epoch": 2, "memory": 25074, "step": 10300} +{"lr": 0.00039045602067755495, "data_time": 0.024681353569030763, "loss": 0.1719048336148262, "time": 0.7603081464767456, "epoch": 2, "memory": 25074, "step": 10400} +{"lr": 0.00039045602067755495, "data_time": 0.0014130353927612304, "loss": 0.169435116648674, "time": 0.7366525411605835, "epoch": 2, "memory": 25074, "step": 10500} +{"lr": 0.00039045602067755495, "data_time": 0.001836705207824707, "loss": 0.15519988983869554, "time": 0.734037184715271, "epoch": 2, "memory": 25074, "step": 10600} +{"lr": 0.00039045602067755495, "data_time": 0.001296854019165039, "loss": 0.15888091921806335, "time": 0.7190788507461547, "epoch": 2, "memory": 25074, "step": 10700} +{"lr": 0.00039045602067755495, "data_time": 0.04676048755645752, "loss": 0.13778819739818574, "time": 0.7508063077926636, "epoch": 2, "memory": 25074, "step": 10800} +{"lr": 0.00039045602067755495, "data_time": 0.0017165422439575195, "loss": 0.15494614839553833, "time": 0.7291338205337524, "epoch": 2, "memory": 25074, "step": 10900} +{"lr": 0.00039045602067755495, "data_time": 0.0015732526779174804, "loss": 0.1438566505908966, "time": 0.7853857040405273, "epoch": 2, "memory": 25074, "step": 11000} +{"lr": 0.00039045602067755495, "data_time": 0.0023783445358276367, "loss": 0.1316233515739441, "time": 0.7321438550949096, "epoch": 2, "memory": 25074, "step": 11100} +{"lr": 0.00039045602067755495, "data_time": 0.0015875339508056641, "loss": 0.14384893029928209, "time": 0.7056066751480102, "epoch": 2, "memory": 25074, "step": 11200} +{"lr": 0.00039045602067755495, "data_time": 0.0020194053649902344, "loss": 0.14245496317744255, "time": 0.7379435539245606, "epoch": 2, "memory": 25074, "step": 11300} +{"lr": 0.00039045602067755495, "data_time": 0.0017960786819458008, "loss": 0.1622574359178543, "time": 0.7063355207443237, "epoch": 2, "memory": 25074, "step": 11400} +{"lr": 0.00039045602067755495, "data_time": 0.0014561891555786132, "loss": 0.13409009650349618, "time": 0.7040745735168457, "epoch": 2, "memory": 25074, "step": 11500} +{"lr": 0.00039045602067755495, "data_time": 0.0016580820083618164, "loss": 0.13272096514701842, "time": 0.7397542715072631, "epoch": 2, "memory": 25074, "step": 11600} +{"lr": 0.00039045602067755495, "data_time": 0.001663970947265625, "loss": 0.14436589553952217, "time": 0.733658742904663, "epoch": 2, "memory": 25074, "step": 11700} +{"lr": 0.00039045602067755495, "data_time": 0.0017907142639160157, "loss": 0.14573181942105293, "time": 0.7077654123306274, "epoch": 2, "memory": 25074, "step": 11800} +{"lr": 0.00039045602067755495, "data_time": 0.002051091194152832, "loss": 0.1403046302497387, "time": 0.7531724214553833, "epoch": 2, "memory": 25074, "step": 11900} +{"lr": 0.00039045602067755495, "data_time": 0.0019072294235229492, "loss": 0.14200572595000266, "time": 0.7084569692611694, "epoch": 2, "memory": 25074, "step": 12000} +{"lr": 0.00039045602067755495, "data_time": 0.001704263687133789, "loss": 0.14508214965462685, "time": 0.70661940574646, "epoch": 2, "memory": 25074, "step": 12100} +{"lr": 0.00039045602067755495, "data_time": 0.0018909454345703125, "loss": 0.14885776340961457, "time": 0.7115938425064087, "epoch": 2, "memory": 25074, "step": 12200} +{"lr": 0.00039045602067755495, "data_time": 0.0021780967712402345, "loss": 0.13284775018692016, "time": 0.7098190546035766, "epoch": 2, "memory": 25074, "step": 12300} +{"lr": 0.00039045602067755495, "data_time": 0.0016410112380981444, "loss": 0.14073834717273712, "time": 0.7054736852645874, "epoch": 2, "memory": 25074, "step": 12400} +{"lr": 0.00039045602067755495, "data_time": 0.001635122299194336, "loss": 0.14130793288350105, "time": 0.7074922323226929, "epoch": 2, "memory": 25074, "step": 12500} +{"lr": 0.00039045602067755495, "data_time": 0.001258993148803711, "loss": 0.1366831324994564, "time": 0.7151970863342285, "epoch": 2, "memory": 25074, "step": 12600} +{"lr": 0.00039045602067755495, "data_time": 0.0018751382827758788, "loss": 0.14397881776094437, "time": 0.7069556713104248, "epoch": 2, "memory": 25074, "step": 12700} +{"lr": 0.00039045602067755495, "data_time": 0.001420283317565918, "loss": 0.109129748493433, "time": 0.7081318140029907, "epoch": 2, "memory": 25074, "step": 12800} +{"lr": 0.00039045602067755495, "data_time": 0.0013454914093017577, "loss": 0.13538127541542053, "time": 0.704267144203186, "epoch": 2, "memory": 25074, "step": 12900} +{"lr": 0.00039045602067755495, "data_time": 0.002026271820068359, "loss": 0.14159722924232482, "time": 0.7106671333312988, "epoch": 2, "memory": 25074, "step": 13000} +{"lr": 0.00039045602067755495, "data_time": 0.0019367456436157227, "loss": 0.12928728386759758, "time": 0.7087173700332642, "epoch": 2, "memory": 25074, "step": 13100} +{"lr": 0.00039045602067755495, "data_time": 0.0013850927352905273, "loss": 0.13836505636572838, "time": 0.707839035987854, "epoch": 2, "memory": 25074, "step": 13200} +{"lr": 0.00039045602067755495, "data_time": 0.0018125057220458984, "loss": 0.13511562943458558, "time": 0.797005581855774, "epoch": 2, "memory": 25074, "step": 13300} +{"lr": 0.00039045602067755495, "data_time": 0.002506875991821289, "loss": 0.122445597499609, "time": 0.7427887439727783, "epoch": 2, "memory": 25074, "step": 13400} +{"lr": 0.00039045602067755495, "data_time": 0.002304267883300781, "loss": 0.11888976395130157, "time": 0.7066573143005371, "epoch": 2, "memory": 25074, "step": 13500} +{"lr": 0.00039045602067755495, "data_time": 0.0012183666229248046, "loss": 0.12036007642745972, "time": 0.7212828636169434, "epoch": 2, "memory": 25074, "step": 13600} +{"lr": 0.00039045602067755495, "data_time": 0.0019524097442626953, "loss": 0.1279669962823391, "time": 0.7417988061904908, "epoch": 2, "memory": 25074, "step": 13700} +{"lr": 0.00039045602067755495, "data_time": 0.001661539077758789, "loss": 0.1092663660645485, "time": 0.7353540658950806, "epoch": 2, "memory": 25074, "step": 13800} +{"lr": 0.00039045602067755495, "data_time": 0.0022333383560180662, "loss": 0.10931268632411957, "time": 0.8255977392196655, "epoch": 2, "memory": 25074, "step": 13900} +{"lr": 0.00039045602067755495, "data_time": 0.0017616033554077148, "loss": 0.11564795449376106, "time": 0.725689172744751, "epoch": 2, "memory": 25074, "step": 14000} +{"lr": 0.00039045602067755495, "data_time": 0.0017904520034790039, "loss": 0.11199476048350335, "time": 0.7052001476287841, "epoch": 2, "memory": 25074, "step": 14100} +{"lr": 0.00039045602067755495, "data_time": 0.0017853736877441405, "loss": 0.11048664525151253, "time": 0.7792333364486694, "epoch": 2, "memory": 25074, "step": 14200} +{"lr": 0.00039045602067755495, "data_time": 0.0033494710922241213, "loss": 0.09773226976394653, "time": 0.910241675376892, "epoch": 2, "memory": 25074, "step": 14300} +{"lr": 0.00039045602067755495, "data_time": 0.004221343994140625, "loss": 0.12284593433141708, "time": 0.8739740371704101, "epoch": 2, "memory": 25074, "step": 14400} +{"lr": 0.00039045602067755495, "data_time": 0.0012976646423339844, "loss": 0.12258741408586502, "time": 0.737842607498169, "epoch": 2, "memory": 25074, "step": 14500} +{"lr": 0.00039045602067755495, "data_time": 0.0012325048446655273, "loss": 0.1167706497013569, "time": 0.7552871465682983, "epoch": 2, "memory": 25074, "step": 14600} +{"lr": 0.00039045602067755495, "data_time": 0.0031726598739624024, "loss": 0.10753279253840446, "time": 0.8949979305267334, "epoch": 2, "memory": 25074, "step": 14700} +{"lr": 0.00039045602067755495, "data_time": 0.0013030529022216796, "loss": 0.1147630363702774, "time": 0.7629199266433716, "epoch": 2, "memory": 25074, "step": 14800} +{"lr": 0.00039045602067755495, "data_time": 0.0014464855194091797, "loss": 0.12037307247519494, "time": 0.738120698928833, "epoch": 2, "memory": 25074, "step": 14900} +{"lr": 0.00039045602067755495, "data_time": 0.0017142295837402344, "loss": 0.09465071111917496, "time": 0.7545335531234741, "epoch": 2, "memory": 25074, "step": 15000} +{"lr": 0.00039045602067755495, "data_time": 0.002235269546508789, "loss": 0.09188344329595566, "time": 0.8110369920730591, "epoch": 2, "memory": 25074, "step": 15100} +{"lr": 0.00039045602067755495, "data_time": 0.0013940095901489257, "loss": 0.10843180418014527, "time": 0.7215148448944092, "epoch": 2, "memory": 25074, "step": 15200} +{"lr": 0.00039045602067755495, "data_time": 0.0013376474380493164, "loss": 0.09459175616502762, "time": 0.7215344905853271, "epoch": 2, "memory": 25074, "step": 15300} +{"lr": 0.00039045602067755495, "data_time": 0.001319742202758789, "loss": 0.09571233578026295, "time": 0.7615896701812744, "epoch": 2, "memory": 25074, "step": 15400} +{"lr": 0.00039045602067755495, "data_time": 0.002578878402709961, "loss": 0.11871992759406566, "time": 0.8586896181106567, "epoch": 2, "memory": 25074, "step": 15500} +{"lr": 0.00039045602067755495, "data_time": 0.0011219501495361329, "loss": 0.10624518673866987, "time": 0.6593152284622192, "epoch": 2, "memory": 25074, "step": 15600} +{"accuracy/top1": 96.61006164550781, "data_time": 0.20340758211472454, "time": 0.5980817289913402, "step": 2} +{"lr": 0.00036275831390311474, "data_time": 0.0013336896896362304, "loss": 0.10310088619589805, "time": 0.7276587009429931, "epoch": 3, "memory": 25074, "step": 15700} +{"lr": 0.00036275831390311474, "data_time": 0.04652788639068604, "loss": 0.08784256875514984, "time": 0.7491942882537842, "epoch": 3, "memory": 25074, "step": 15800} +{"lr": 0.00036275831390311474, "data_time": 0.042873191833496097, "loss": 0.11111270263791084, "time": 0.7634331941604614, "epoch": 3, "memory": 25074, "step": 15900} +{"lr": 0.00036275831390311474, "data_time": 0.05382053852081299, "loss": 0.10052410736680031, "time": 0.7560703992843628, "epoch": 3, "memory": 25074, "step": 16000} +{"lr": 0.00036275831390311474, "data_time": 0.09436025619506835, "loss": 0.0834791749715805, "time": 0.8450967788696289, "epoch": 3, "memory": 25074, "step": 16100} +{"lr": 0.00036275831390311474, "data_time": 0.019767117500305176, "loss": 0.11351828798651695, "time": 0.7252115726470947, "epoch": 3, "memory": 25074, "step": 16200} +{"lr": 0.00036275831390311474, "data_time": 0.036263322830200194, "loss": 0.10130115523934365, "time": 0.7392185211181641, "epoch": 3, "memory": 25074, "step": 16300} +{"lr": 0.00036275831390311474, "data_time": 0.04586970806121826, "loss": 0.10232554823160171, "time": 0.7497789621353149, "epoch": 3, "memory": 25074, "step": 16400} +{"lr": 0.00036275831390311474, "data_time": 0.07142379283905029, "loss": 0.10023172087967396, "time": 0.7870445966720581, "epoch": 3, "memory": 25074, "step": 16500} +{"lr": 0.00036275831390311474, "data_time": 0.0784637451171875, "loss": 0.10155723989009857, "time": 0.7833858251571655, "epoch": 3, "memory": 25074, "step": 16600} +{"lr": 0.00036275831390311474, "data_time": 0.001291179656982422, "loss": 0.0844054602086544, "time": 0.7038801193237305, "epoch": 3, "memory": 25074, "step": 16700} +{"lr": 0.00036275831390311474, "data_time": 0.05306949615478516, "loss": 0.0883020780980587, "time": 0.7562115907669067, "epoch": 3, "memory": 25074, "step": 16800} +{"lr": 0.00036275831390311474, "data_time": 0.032765316963195804, "loss": 0.09582230374217034, "time": 0.7353592634201049, "epoch": 3, "memory": 25074, "step": 16900} +{"lr": 0.00036275831390311474, "data_time": 0.17749545574188233, "loss": 0.07940515093505382, "time": 0.9362649202346802, "epoch": 3, "memory": 25074, "step": 17000} +{"lr": 0.00036275831390311474, "data_time": 0.06539099216461182, "loss": 0.09481642134487629, "time": 0.7688310861587524, "epoch": 3, "memory": 25074, "step": 17100} +{"lr": 0.00036275831390311474, "data_time": 0.00936436653137207, "loss": 0.08838141858577728, "time": 0.7122403144836426, "epoch": 3, "memory": 25074, "step": 17200} +{"lr": 0.00036275831390311474, "data_time": 0.027420973777770995, "loss": 0.09090999439358712, "time": 0.7304120779037475, "epoch": 3, "memory": 25074, "step": 17300} +{"lr": 0.00036275831390311474, "data_time": 0.02483525276184082, "loss": 0.08162333928048611, "time": 0.7273419618606567, "epoch": 3, "memory": 25074, "step": 17400} +{"lr": 0.00036275831390311474, "data_time": 0.05052340030670166, "loss": 0.08197577185928821, "time": 0.7882192611694336, "epoch": 3, "memory": 25074, "step": 17500} +{"lr": 0.00036275831390311474, "data_time": 0.016635370254516602, "loss": 0.08423399589955807, "time": 0.72077476978302, "epoch": 3, "memory": 25074, "step": 17600} +{"lr": 0.00036275831390311474, "data_time": 0.09424901008605957, "loss": 0.09517066031694413, "time": 0.815043830871582, "epoch": 3, "memory": 25074, "step": 17700} +{"lr": 0.00036275831390311474, "data_time": 0.09383594989776611, "loss": 0.11006402745842933, "time": 0.7976555347442627, "epoch": 3, "memory": 25074, "step": 17800} +{"lr": 0.00036275831390311474, "data_time": 0.2113175392150879, "loss": 0.09635679423809052, "time": 0.9638019323348999, "epoch": 3, "memory": 25074, "step": 17900} +{"lr": 0.00036275831390311474, "data_time": 0.020669436454772948, "loss": 0.09522832930088043, "time": 0.7267727136611939, "epoch": 3, "memory": 25074, "step": 18000} +{"lr": 0.00036275831390311474, "data_time": 0.041952657699584964, "loss": 0.08613555580377578, "time": 0.7532625436782837, "epoch": 3, "memory": 25074, "step": 18100} +{"lr": 0.00036275831390311474, "data_time": 0.0014177799224853516, "loss": 0.08952978923916817, "time": 0.7037564754486084, "epoch": 3, "memory": 25074, "step": 18200} +{"lr": 0.00036275831390311474, "data_time": 0.04420890808105469, "loss": 0.08152703158557414, "time": 0.763043737411499, "epoch": 3, "memory": 25074, "step": 18300} +{"lr": 0.00036275831390311474, "data_time": 0.05574164390563965, "loss": 0.08979608416557312, "time": 0.8042138576507568, "epoch": 3, "memory": 25074, "step": 18400} +{"lr": 0.00036275831390311474, "data_time": 0.0014020919799804688, "loss": 0.08850124888122082, "time": 0.731146764755249, "epoch": 3, "memory": 25074, "step": 18500} +{"lr": 0.00036275831390311474, "data_time": 0.0012506008148193359, "loss": 0.0713675782084465, "time": 0.7047372341156006, "epoch": 3, "memory": 25074, "step": 18600} +{"lr": 0.00036275831390311474, "data_time": 0.0013912677764892577, "loss": 0.09244537055492401, "time": 0.7872651815414429, "epoch": 3, "memory": 25074, "step": 18700} +{"lr": 0.00036275831390311474, "data_time": 0.002373361587524414, "loss": 0.08013321980834007, "time": 0.8584137678146362, "epoch": 3, "memory": 25074, "step": 18800} +{"lr": 0.00036275831390311474, "data_time": 0.0012579679489135743, "loss": 0.09345643445849419, "time": 0.7364373683929444, "epoch": 3, "memory": 25074, "step": 18900} +{"lr": 0.00036275831390311474, "data_time": 0.001368880271911621, "loss": 0.07542329467833042, "time": 0.7544003248214721, "epoch": 3, "memory": 25074, "step": 19000} +{"lr": 0.00036275831390311474, "data_time": 0.001826310157775879, "loss": 0.08801689743995667, "time": 0.740264105796814, "epoch": 3, "memory": 25074, "step": 19100} +{"lr": 0.00036275831390311474, "data_time": 0.0012601375579833984, "loss": 0.0794067095965147, "time": 0.7126410245895386, "epoch": 3, "memory": 25074, "step": 19200} +{"lr": 0.00036275831390311474, "data_time": 0.004342913627624512, "loss": 0.07495554648339749, "time": 0.8991502046585083, "epoch": 3, "memory": 25074, "step": 19300} +{"lr": 0.00036275831390311474, "data_time": 0.001290130615234375, "loss": 0.0925452709197998, "time": 0.7162141561508178, "epoch": 3, "memory": 25074, "step": 19400} +{"lr": 0.00036275831390311474, "data_time": 0.0013996124267578124, "loss": 0.07054197788238525, "time": 0.7638662099838257, "epoch": 3, "memory": 25074, "step": 19500} +{"lr": 0.00036275831390311474, "data_time": 0.0020952939987182615, "loss": 0.07390230745077134, "time": 0.7642737865447998, "epoch": 3, "memory": 25074, "step": 19600} +{"lr": 0.00036275831390311474, "data_time": 0.0013412952423095704, "loss": 0.07707810625433922, "time": 0.7533664226531982, "epoch": 3, "memory": 25074, "step": 19700} +{"lr": 0.00036275831390311474, "data_time": 0.002320408821105957, "loss": 0.06675496846437454, "time": 0.8157778263092041, "epoch": 3, "memory": 25074, "step": 19800} +{"lr": 0.00036275831390311474, "data_time": 0.0018549203872680665, "loss": 0.09099142514169216, "time": 0.7152692794799804, "epoch": 3, "memory": 25074, "step": 19900} +{"lr": 0.00036275831390311474, "data_time": 0.0013274669647216797, "loss": 0.07269903682172299, "time": 0.7267545700073242, "epoch": 3, "memory": 25074, "step": 20000} +{"lr": 0.00036275831390311474, "data_time": 0.0017048835754394532, "loss": 0.08231619521975517, "time": 0.7686270713806153, "epoch": 3, "memory": 25074, "step": 20100} +{"lr": 0.00036275831390311474, "data_time": 0.001417827606201172, "loss": 0.06602434609085321, "time": 0.7393132209777832, "epoch": 3, "memory": 25074, "step": 20200} +{"lr": 0.00036275831390311474, "data_time": 0.0013814210891723634, "loss": 0.07883532866835594, "time": 0.7038391351699829, "epoch": 3, "memory": 25074, "step": 20300} +{"lr": 0.00036275831390311474, "data_time": 0.0012912511825561523, "loss": 0.06388877741992474, "time": 0.7664240837097168, "epoch": 3, "memory": 25074, "step": 20400} +{"lr": 0.00036275831390311474, "data_time": 0.0014119863510131836, "loss": 0.06450840719044208, "time": 0.761465334892273, "epoch": 3, "memory": 25074, "step": 20500} +{"lr": 0.00036275831390311474, "data_time": 0.0012936115264892579, "loss": 0.07086011171340942, "time": 0.7718260288238525, "epoch": 3, "memory": 25074, "step": 20600} +{"lr": 0.00036275831390311474, "data_time": 0.002866697311401367, "loss": 0.0806809864938259, "time": 0.8923632383346558, "epoch": 3, "memory": 25074, "step": 20700} +{"lr": 0.00036275831390311474, "data_time": 0.0014344215393066405, "loss": 0.06330490708351136, "time": 0.7350774765014648, "epoch": 3, "memory": 25074, "step": 20800} +{"lr": 0.00036275831390311474, "data_time": 0.0013097524642944336, "loss": 0.07199015617370605, "time": 0.7060601234436035, "epoch": 3, "memory": 25074, "step": 20900} +{"lr": 0.00036275831390311474, "data_time": 0.0014673233032226562, "loss": 0.06927086636424065, "time": 0.7415284872055053, "epoch": 3, "memory": 25074, "step": 21000} +{"lr": 0.00036275831390311474, "data_time": 0.0013409137725830078, "loss": 0.07275770232081413, "time": 0.7100635528564453, "epoch": 3, "memory": 25074, "step": 21100} +{"lr": 0.00036275831390311474, "data_time": 0.0024672508239746093, "loss": 0.071163310110569, "time": 0.8411956548690795, "epoch": 3, "memory": 25074, "step": 21200} +{"lr": 0.00036275831390311474, "data_time": 0.0012351512908935548, "loss": 0.07052629329264164, "time": 0.757619071006775, "epoch": 3, "memory": 25074, "step": 21300} +{"lr": 0.00036275831390311474, "data_time": 0.0012850999832153321, "loss": 0.06190228536725044, "time": 0.7438790559768677, "epoch": 3, "memory": 25074, "step": 21400} +{"lr": 0.00036275831390311474, "data_time": 0.0013517618179321289, "loss": 0.06569394581019879, "time": 0.7910687446594238, "epoch": 3, "memory": 25074, "step": 21500} +{"lr": 0.00036275831390311474, "data_time": 0.0026489973068237306, "loss": 0.057651686668396, "time": 0.8509131669998169, "epoch": 3, "memory": 25074, "step": 21600} +{"lr": 0.00036275831390311474, "data_time": 0.0014965295791625976, "loss": 0.06758101843297482, "time": 0.754273247718811, "epoch": 3, "memory": 25074, "step": 21700} +{"lr": 0.00036275831390311474, "data_time": 0.0012147903442382812, "loss": 0.07426468916237354, "time": 0.7568423032760621, "epoch": 3, "memory": 25074, "step": 21800} +{"lr": 0.00036275831390311474, "data_time": 0.0013579130172729492, "loss": 0.07236175537109375, "time": 0.7756822586059571, "epoch": 3, "memory": 25074, "step": 21900} +{"lr": 0.00036275831390311474, "data_time": 0.001392507553100586, "loss": 0.07234665043652058, "time": 0.7807016372680664, "epoch": 3, "memory": 25074, "step": 22000} +{"lr": 0.00036275831390311474, "data_time": 0.00457148551940918, "loss": 0.054981644824147226, "time": 0.8098761796951294, "epoch": 3, "memory": 25074, "step": 22100} +{"lr": 0.00036275831390311474, "data_time": 0.0015793323516845703, "loss": 0.06135090030729771, "time": 0.735714316368103, "epoch": 3, "memory": 25074, "step": 22200} +{"lr": 0.00036275831390311474, "data_time": 0.0013195276260375977, "loss": 0.05634091794490814, "time": 0.7164009809494019, "epoch": 3, "memory": 25074, "step": 22300} +{"lr": 0.00036275831390311474, "data_time": 0.0012403011322021484, "loss": 0.06019404791295528, "time": 0.7335800170898438, "epoch": 3, "memory": 25074, "step": 22400} +{"lr": 0.00036275831390311474, "data_time": 0.0020447731018066405, "loss": 0.06717828437685966, "time": 0.7600839614868165, "epoch": 3, "memory": 25074, "step": 22500} +{"lr": 0.00036275831390311474, "data_time": 0.0013044118881225587, "loss": 0.06792790107429028, "time": 0.8072243928909302, "epoch": 3, "memory": 25074, "step": 22600} +{"lr": 0.00036275831390311474, "data_time": 0.0013843059539794921, "loss": 0.05459876991808414, "time": 0.7623852729797364, "epoch": 3, "memory": 25074, "step": 22700} +{"lr": 0.00036275831390311474, "data_time": 0.001479363441467285, "loss": 0.060161656513810155, "time": 0.7363945245742798, "epoch": 3, "memory": 25074, "step": 22800} +{"lr": 0.00036275831390311474, "data_time": 0.01769576072692871, "loss": 0.05969017185270786, "time": 0.7204919815063476, "epoch": 3, "memory": 25074, "step": 22900} +{"lr": 0.00036275831390311474, "data_time": 0.03938698768615723, "loss": 0.058220787532627584, "time": 0.8035845756530762, "epoch": 3, "memory": 25074, "step": 23000} +{"lr": 0.00036275831390311474, "data_time": 0.13300104141235353, "loss": 0.054922381229698655, "time": 0.8375943899154663, "epoch": 3, "memory": 25074, "step": 23100} +{"lr": 0.00036275831390311474, "data_time": 0.00842437744140625, "loss": 0.06590216085314751, "time": 0.7251642227172852, "epoch": 3, "memory": 25074, "step": 23200} +{"lr": 0.00036275831390311474, "data_time": 0.02899308204650879, "loss": 0.06242698859423399, "time": 0.733867883682251, "epoch": 3, "memory": 25074, "step": 23300} +{"lr": 0.00036275831390311474, "data_time": 0.13492271900177003, "loss": 0.05929134879261255, "time": 0.8273138523101806, "epoch": 3, "memory": 25074, "step": 23400} +{"accuracy/top1": 97.89154052734375, "data_time": 0.1995232806486242, "time": 0.5852146148681641, "step": 3} +{"lr": 0.00031961812419703224, "data_time": 0.0034704208374023438, "loss": 0.04517615139484406, "time": 0.8812820911407471, "epoch": 4, "memory": 25074, "step": 23500} +{"lr": 0.00031961812419703224, "data_time": 0.00133514404296875, "loss": 0.045108597725629807, "time": 0.7198319435119629, "epoch": 4, "memory": 25074, "step": 23600} +{"lr": 0.00031961812419703224, "data_time": 0.0014042139053344726, "loss": 0.05669744834303856, "time": 0.7504631519317627, "epoch": 4, "memory": 25074, "step": 23700} +{"lr": 0.00031961812419703224, "data_time": 0.001220846176147461, "loss": 0.0512423537671566, "time": 0.7433174133300782, "epoch": 4, "memory": 25074, "step": 23800} +{"lr": 0.00031961812419703224, "data_time": 0.00250704288482666, "loss": 0.05096057280898094, "time": 0.8091708421707153, "epoch": 4, "memory": 25074, "step": 23900} +{"lr": 0.00031961812419703224, "data_time": 0.0014014959335327149, "loss": 0.0507582375779748, "time": 0.7508286714553833, "epoch": 4, "memory": 25074, "step": 24000} +{"lr": 0.00031961812419703224, "data_time": 0.0013149261474609375, "loss": 0.050636058859527114, "time": 0.8642278194427491, "epoch": 4, "memory": 25074, "step": 24100} +{"lr": 0.00031961812419703224, "data_time": 0.0013562917709350586, "loss": 0.06051886379718781, "time": 0.7549211978912354, "epoch": 4, "memory": 25074, "step": 24200} +{"lr": 0.00031961812419703224, "data_time": 0.0013926506042480468, "loss": 0.047161291167140004, "time": 0.7464770078659058, "epoch": 4, "memory": 25074, "step": 24300} +{"lr": 0.00031961812419703224, "data_time": 0.0053446292877197266, "loss": 0.05329375602304935, "time": 0.8669121026992798, "epoch": 4, "memory": 25074, "step": 24400} +{"lr": 0.00031961812419703224, "data_time": 0.0012766599655151367, "loss": 0.0478238308802247, "time": 0.7892888069152832, "epoch": 4, "memory": 25074, "step": 24500} +{"lr": 0.00031961812419703224, "data_time": 0.0013270854949951171, "loss": 0.04734180849045515, "time": 0.7029707670211792, "epoch": 4, "memory": 25074, "step": 24600} +{"lr": 0.00031961812419703224, "data_time": 0.0012899875640869141, "loss": 0.0476755864918232, "time": 0.7805631399154663, "epoch": 4, "memory": 25074, "step": 24700} +{"lr": 0.00031961812419703224, "data_time": 0.002505064010620117, "loss": 0.0517438736744225, "time": 0.8386474132537842, "epoch": 4, "memory": 25074, "step": 24800} +{"lr": 0.00031961812419703224, "data_time": 0.0013689994812011719, "loss": 0.05274440515786409, "time": 0.7870511770248413, "epoch": 4, "memory": 25074, "step": 24900} +{"lr": 0.00031961812419703224, "data_time": 0.0013701677322387694, "loss": 0.043704770132899286, "time": 0.7481824159622192, "epoch": 4, "memory": 25074, "step": 25000} +{"lr": 0.00031961812419703224, "data_time": 0.0014128446578979491, "loss": 0.04268152825534344, "time": 0.7583313465118409, "epoch": 4, "memory": 25074, "step": 25100} +{"lr": 0.00031961812419703224, "data_time": 0.00205385684967041, "loss": 0.04238433614373207, "time": 0.8143611192703247, "epoch": 4, "memory": 25074, "step": 25200} +{"lr": 0.00031961812419703224, "data_time": 0.0045460939407348635, "loss": 0.05427344739437103, "time": 0.8976642847061157, "epoch": 4, "memory": 25074, "step": 25300} +{"lr": 0.00031961812419703224, "data_time": 0.0012501716613769532, "loss": 0.04692898578941822, "time": 0.7169560194015503, "epoch": 4, "memory": 25074, "step": 25400} +{"lr": 0.00031961812419703224, "data_time": 0.001289534568786621, "loss": 0.04919785112142563, "time": 0.7365872859954834, "epoch": 4, "memory": 25074, "step": 25500} +{"lr": 0.00031961812419703224, "data_time": 0.0012316465377807616, "loss": 0.05004159957170486, "time": 0.7409481048583985, "epoch": 4, "memory": 25074, "step": 25600} +{"lr": 0.00031961812419703224, "data_time": 0.0014320135116577149, "loss": 0.05063360538333654, "time": 0.7744909286499023, "epoch": 4, "memory": 25074, "step": 25700} +{"lr": 0.00031961812419703224, "data_time": 0.0029634952545166014, "loss": 0.03686757292598486, "time": 0.7800127267837524, "epoch": 4, "memory": 25074, "step": 25800} +{"lr": 0.00031961812419703224, "data_time": 0.001425027847290039, "loss": 0.030946124531328677, "time": 0.725481915473938, "epoch": 4, "memory": 25074, "step": 25900} +{"lr": 0.00031961812419703224, "data_time": 0.001331329345703125, "loss": 0.03346645049750805, "time": 0.7261993408203125, "epoch": 4, "memory": 25074, "step": 26000} +{"lr": 0.00031961812419703224, "data_time": 0.001289200782775879, "loss": 0.050221163965761664, "time": 0.7927984952926636, "epoch": 4, "memory": 25074, "step": 26100} +{"lr": 0.00031961812419703224, "data_time": 0.0026758193969726564, "loss": 0.047254554182291034, "time": 1.0025804042816162, "epoch": 4, "memory": 25074, "step": 26200} +{"lr": 0.00031961812419703224, "data_time": 0.001240110397338867, "loss": 0.05443720631301403, "time": 0.7894600868225098, "epoch": 4, "memory": 25074, "step": 26300} +{"lr": 0.00031961812419703224, "data_time": 0.0013683557510375977, "loss": 0.04285751581192017, "time": 0.7531031370162964, "epoch": 4, "memory": 25074, "step": 26400} +{"lr": 0.00031961812419703224, "data_time": 0.0013829469680786133, "loss": 0.037723523750901225, "time": 0.7630849599838256, "epoch": 4, "memory": 25074, "step": 26500} +{"lr": 0.00031961812419703224, "data_time": 0.001319551467895508, "loss": 0.051023753546178344, "time": 0.783145809173584, "epoch": 4, "memory": 25074, "step": 26600} +{"lr": 0.00031961812419703224, "data_time": 0.0015150547027587891, "loss": 0.04813426043838263, "time": 0.7720998525619507, "epoch": 4, "memory": 25074, "step": 26700} +{"lr": 0.00031961812419703224, "data_time": 0.001169872283935547, "loss": 0.043939481303095815, "time": 0.7106137037277221, "epoch": 4, "memory": 25074, "step": 26800} +{"lr": 0.00031961812419703224, "data_time": 0.0014029979705810548, "loss": 0.03220619978383184, "time": 0.7422962427139282, "epoch": 4, "memory": 25074, "step": 26900} +{"lr": 0.00031961812419703224, "data_time": 0.001316666603088379, "loss": 0.04515266828238964, "time": 0.7716180086135864, "epoch": 4, "memory": 25074, "step": 27000} +{"lr": 0.00031961812419703224, "data_time": 0.004015541076660157, "loss": 0.037893311493098734, "time": 0.8699528932571411, "epoch": 4, "memory": 25074, "step": 27100} +{"lr": 0.00031961812419703224, "data_time": 0.0012329578399658202, "loss": 0.041449266765266654, "time": 0.7464729309082031, "epoch": 4, "memory": 25074, "step": 27200} +{"lr": 0.00031961812419703224, "data_time": 0.0012342691421508788, "loss": 0.03178658131510019, "time": 0.733365535736084, "epoch": 4, "memory": 25074, "step": 27300} +{"lr": 0.00031961812419703224, "data_time": 0.06804072856903076, "loss": 0.03859858289361, "time": 0.7901822566986084, "epoch": 4, "memory": 25074, "step": 27400} +{"lr": 0.00031961812419703224, "data_time": 0.0012955904006958009, "loss": 0.03694054465740919, "time": 0.7995924711227417, "epoch": 4, "memory": 25074, "step": 27500} +{"lr": 0.00031961812419703224, "data_time": 0.0018056631088256836, "loss": 0.04908970184624195, "time": 0.8272752046585083, "epoch": 4, "memory": 25074, "step": 27600} +{"lr": 0.00031961812419703224, "data_time": 0.0014280080795288086, "loss": 0.033122124476358296, "time": 0.7333185195922851, "epoch": 4, "memory": 25074, "step": 27700} +{"lr": 0.00031961812419703224, "data_time": 0.0016160726547241211, "loss": 0.03568306379020214, "time": 0.7440671920776367, "epoch": 4, "memory": 25074, "step": 27800} +{"lr": 0.00031961812419703224, "data_time": 0.0012700796127319337, "loss": 0.031327463872730735, "time": 0.7310517787933349, "epoch": 4, "memory": 25074, "step": 27900} +{"lr": 0.00031961812419703224, "data_time": 0.003715372085571289, "loss": 0.03493130672723055, "time": 0.8083968877792358, "epoch": 4, "memory": 25074, "step": 28000} +{"lr": 0.00031961812419703224, "data_time": 0.0012696266174316406, "loss": 0.03904465064406395, "time": 0.7613831043243409, "epoch": 4, "memory": 25074, "step": 28100} +{"lr": 0.00031961812419703224, "data_time": 0.001513051986694336, "loss": 0.02872838657349348, "time": 0.7503234624862671, "epoch": 4, "memory": 25074, "step": 28200} +{"lr": 0.00031961812419703224, "data_time": 0.0014272451400756836, "loss": 0.042362000048160556, "time": 0.7054878234863281, "epoch": 4, "memory": 25074, "step": 28300} +{"lr": 0.00031961812419703224, "data_time": 0.0029218912124633787, "loss": 0.03637932613492012, "time": 0.8058889627456665, "epoch": 4, "memory": 25074, "step": 28400} +{"lr": 0.00031961812419703224, "data_time": 0.001334834098815918, "loss": 0.03377866544760764, "time": 0.7404050350189209, "epoch": 4, "memory": 25074, "step": 28500} +{"lr": 0.00031961812419703224, "data_time": 0.0013770341873168945, "loss": 0.04600885380059481, "time": 0.7099129438400269, "epoch": 4, "memory": 25074, "step": 28600} +{"lr": 0.00031961812419703224, "data_time": 0.0014901399612426759, "loss": 0.03663482228294015, "time": 0.7665527105331421, "epoch": 4, "memory": 25074, "step": 28700} +{"lr": 0.00031961812419703224, "data_time": 0.001254701614379883, "loss": 0.03127280483022332, "time": 0.737017560005188, "epoch": 4, "memory": 25074, "step": 28800} +{"lr": 0.00031961812419703224, "data_time": 0.0013598918914794922, "loss": 0.028428874118253587, "time": 0.7405951738357544, "epoch": 4, "memory": 25074, "step": 28900} +{"lr": 0.00031961812419703224, "data_time": 0.0012872934341430664, "loss": 0.027683245483785868, "time": 0.7130789995193482, "epoch": 4, "memory": 25074, "step": 29000} +{"lr": 0.00031961812419703224, "data_time": 0.0013851404190063476, "loss": 0.029278316907584666, "time": 0.7043890237808228, "epoch": 4, "memory": 25074, "step": 29100} +{"lr": 0.00031961812419703224, "data_time": 0.0013159751892089845, "loss": 0.03678624513559044, "time": 0.7159635066986084, "epoch": 4, "memory": 25074, "step": 29200} +{"lr": 0.00031961812419703224, "data_time": 0.001799941062927246, "loss": 0.0363719992339611, "time": 0.805773115158081, "epoch": 4, "memory": 25074, "step": 29300} +{"lr": 0.00031961812419703224, "data_time": 0.0014745950698852538, "loss": 0.02529519619420171, "time": 0.7127826452255249, "epoch": 4, "memory": 25074, "step": 29400} +{"lr": 0.00031961812419703224, "data_time": 0.0012825727462768555, "loss": 0.047313752584159374, "time": 0.73499436378479, "epoch": 4, "memory": 25074, "step": 29500} +{"lr": 0.00031961812419703224, "data_time": 0.0012951374053955078, "loss": 0.03643022384494543, "time": 0.7089319705963135, "epoch": 4, "memory": 25074, "step": 29600} +{"lr": 0.00031961812419703224, "data_time": 0.0033378839492797852, "loss": 0.03052303325384855, "time": 0.8332024812698364, "epoch": 4, "memory": 25074, "step": 29700} +{"lr": 0.00031961812419703224, "data_time": 0.0012300968170166015, "loss": 0.029925617575645446, "time": 0.749910855293274, "epoch": 4, "memory": 25074, "step": 29800} +{"lr": 0.00031961812419703224, "data_time": 0.001368236541748047, "loss": 0.03468024590983987, "time": 0.7201395273208618, "epoch": 4, "memory": 25074, "step": 29900} +{"lr": 0.00031961812419703224, "data_time": 0.0015823125839233398, "loss": 0.03215574417263269, "time": 0.7682300806045532, "epoch": 4, "memory": 25074, "step": 30000} +{"lr": 0.00031961812419703224, "data_time": 0.0017320871353149413, "loss": 0.02737377109006047, "time": 0.8383288860321045, "epoch": 4, "memory": 25074, "step": 30100} +{"lr": 0.00031961812419703224, "data_time": 0.0012495756149291993, "loss": 0.03885923847556114, "time": 0.730503249168396, "epoch": 4, "memory": 25074, "step": 30200} +{"lr": 0.00031961812419703224, "data_time": 0.0013231277465820313, "loss": 0.028759025596082212, "time": 0.7839333772659302, "epoch": 4, "memory": 25074, "step": 30300} +{"lr": 0.00031961812419703224, "data_time": 0.0013317108154296876, "loss": 0.022157855425029994, "time": 0.7049121379852294, "epoch": 4, "memory": 25074, "step": 30400} +{"lr": 0.00031961812419703224, "data_time": 0.0012486934661865234, "loss": 0.023827042430639267, "time": 0.7343915700912476, "epoch": 4, "memory": 25074, "step": 30500} +{"lr": 0.00031961812419703224, "data_time": 0.005153155326843262, "loss": 0.025807593390345572, "time": 0.9068384647369385, "epoch": 4, "memory": 25074, "step": 30600} +{"lr": 0.00031961812419703224, "data_time": 0.0012485027313232423, "loss": 0.027053234493359925, "time": 0.7571660995483398, "epoch": 4, "memory": 25074, "step": 30700} +{"lr": 0.00031961812419703224, "data_time": 0.001289057731628418, "loss": 0.03416409445926547, "time": 0.7048086643218994, "epoch": 4, "memory": 25074, "step": 30800} +{"lr": 0.00031961812419703224, "data_time": 0.0011974573135375977, "loss": 0.031753899343311784, "time": 0.7595767259597779, "epoch": 4, "memory": 25074, "step": 30900} +{"lr": 0.00031961812419703224, "data_time": 0.0021053552627563477, "loss": 0.021635157894343136, "time": 0.7889343976974488, "epoch": 4, "memory": 25074, "step": 31000} +{"lr": 0.00031961812419703224, "data_time": 0.0013131141662597657, "loss": 0.0249127839691937, "time": 0.7223089218139649, "epoch": 4, "memory": 25074, "step": 31100} +{"lr": 0.00031961812419703224, "data_time": 0.001041245460510254, "loss": 0.026153938472270967, "time": 0.6581466913223266, "epoch": 4, "memory": 25074, "step": 31200} +{"accuracy/top1": 99.12883758544922, "data_time": 0.1987632583169376, "time": 0.5838278321658864, "step": 4} +{"lr": 0.00026525831390311475, "data_time": 0.0015133380889892577, "loss": 0.027698755217716097, "time": 0.7176483631134033, "epoch": 5, "memory": 25074, "step": 31300} +{"lr": 0.00026525831390311475, "data_time": 0.0013219594955444336, "loss": 0.028043771721422674, "time": 0.716804838180542, "epoch": 5, "memory": 25074, "step": 31400} +{"lr": 0.00026525831390311475, "data_time": 0.0012701034545898437, "loss": 0.025052245054394007, "time": 0.7129191160202026, "epoch": 5, "memory": 25074, "step": 31500} +{"lr": 0.00026525831390311475, "data_time": 0.00399935245513916, "loss": 0.020011105202138425, "time": 0.9069459438323975, "epoch": 5, "memory": 25074, "step": 31600} +{"lr": 0.00026525831390311475, "data_time": 0.0014359474182128907, "loss": 0.035278898850083354, "time": 0.7484704494476319, "epoch": 5, "memory": 25074, "step": 31700} +{"lr": 0.00026525831390311475, "data_time": 0.0018388986587524413, "loss": 0.023777909763157367, "time": 0.776302981376648, "epoch": 5, "memory": 25074, "step": 31800} +{"lr": 0.00026525831390311475, "data_time": 0.0014034032821655274, "loss": 0.018504769867286087, "time": 0.7267004966735839, "epoch": 5, "memory": 25074, "step": 31900} +{"lr": 0.00026525831390311475, "data_time": 0.0022325754165649415, "loss": 0.022173112258315086, "time": 0.930619478225708, "epoch": 5, "memory": 25074, "step": 32000} +{"lr": 0.00026525831390311475, "data_time": 0.0014310359954833984, "loss": 0.029171309247612954, "time": 0.7062459230422974, "epoch": 5, "memory": 25074, "step": 32100} +{"lr": 0.00026525831390311475, "data_time": 0.0014700651168823241, "loss": 0.026411455031484364, "time": 0.7170152187347412, "epoch": 5, "memory": 25074, "step": 32200} +{"lr": 0.00026525831390311475, "data_time": 0.0013317346572875976, "loss": 0.0231924450956285, "time": 0.7058538675308228, "epoch": 5, "memory": 25074, "step": 32300} +{"lr": 0.00026525831390311475, "data_time": 0.00167388916015625, "loss": 0.02030365206301212, "time": 0.8431793928146363, "epoch": 5, "memory": 25074, "step": 32400} +{"lr": 0.00026525831390311475, "data_time": 0.001358366012573242, "loss": 0.021000511338934304, "time": 0.7055831909179687, "epoch": 5, "memory": 25074, "step": 32500} +{"lr": 0.00026525831390311475, "data_time": 0.0013675928115844727, "loss": 0.02990973894484341, "time": 0.8002129554748535, "epoch": 5, "memory": 25074, "step": 32600} +{"lr": 0.00026525831390311475, "data_time": 0.0013486146926879883, "loss": 0.016818862548097967, "time": 0.7578875064849854, "epoch": 5, "memory": 25074, "step": 32700} +{"lr": 0.00026525831390311475, "data_time": 0.0012736082077026366, "loss": 0.0233812834136188, "time": 0.7536785125732421, "epoch": 5, "memory": 25074, "step": 32800} +{"lr": 0.00026525831390311475, "data_time": 0.0030343294143676757, "loss": 0.02946585286408663, "time": 0.7971512079238892, "epoch": 5, "memory": 25074, "step": 32900} +{"lr": 0.00026525831390311475, "data_time": 0.0013983011245727538, "loss": 0.024748912872746588, "time": 0.7699547052383423, "epoch": 5, "memory": 25074, "step": 33000} +{"lr": 0.00026525831390311475, "data_time": 0.0014682292938232421, "loss": 0.025327199790626764, "time": 0.7092276573181152, "epoch": 5, "memory": 25074, "step": 33100} +{"lr": 0.00026525831390311475, "data_time": 0.0018947601318359375, "loss": 0.020371937844902276, "time": 0.825655174255371, "epoch": 5, "memory": 25074, "step": 33200} +{"lr": 0.00026525831390311475, "data_time": 0.0014052391052246094, "loss": 0.0347462797537446, "time": 0.7288672924041748, "epoch": 5, "memory": 25074, "step": 33300} +{"lr": 0.00026525831390311475, "data_time": 0.002937626838684082, "loss": 0.01729632602073252, "time": 0.8441120147705078, "epoch": 5, "memory": 25074, "step": 33400} +{"lr": 0.00026525831390311475, "data_time": 0.001480698585510254, "loss": 0.019980670278891922, "time": 0.7312592029571533, "epoch": 5, "memory": 25074, "step": 33500} +{"lr": 0.00026525831390311475, "data_time": 0.0014867782592773438, "loss": 0.023634103778749704, "time": 0.7822185516357422, "epoch": 5, "memory": 25074, "step": 33600} +{"lr": 0.00026525831390311475, "data_time": 0.0015033721923828126, "loss": 0.016829533502459526, "time": 0.7079687595367432, "epoch": 5, "memory": 25074, "step": 33700} +{"lr": 0.00026525831390311475, "data_time": 0.0021425724029541016, "loss": 0.017635055258870123, "time": 0.8000155210494995, "epoch": 5, "memory": 25074, "step": 33800} +{"lr": 0.00026525831390311475, "data_time": 0.0012916326522827148, "loss": 0.01589393466711044, "time": 0.7038563966751099, "epoch": 5, "memory": 25074, "step": 33900} +{"lr": 0.00026525831390311475, "data_time": 0.0013196229934692382, "loss": 0.0237630317453295, "time": 0.7533472776412964, "epoch": 5, "memory": 25074, "step": 34000} +{"lr": 0.00026525831390311475, "data_time": 0.001246190071105957, "loss": 0.018666595220565796, "time": 0.7325097560882569, "epoch": 5, "memory": 25074, "step": 34100} +{"lr": 0.00026525831390311475, "data_time": 0.0012932062149047852, "loss": 0.017278934270143508, "time": 0.7466542959213257, "epoch": 5, "memory": 25074, "step": 34200} +{"lr": 0.00026525831390311475, "data_time": 0.0011881113052368164, "loss": 0.017761203926056623, "time": 0.7075449228286743, "epoch": 5, "memory": 25074, "step": 34300} +{"lr": 0.00026525831390311475, "data_time": 0.001402568817138672, "loss": 0.022226954717189072, "time": 0.723229956626892, "epoch": 5, "memory": 25074, "step": 34400} +{"lr": 0.00026525831390311475, "data_time": 0.0012682676315307617, "loss": 0.02354100691154599, "time": 0.7038027048110962, "epoch": 5, "memory": 25074, "step": 34500} +{"lr": 0.00026525831390311475, "data_time": 0.0012714624404907226, "loss": 0.020151436701416968, "time": 0.7552997827529907, "epoch": 5, "memory": 25074, "step": 34600} +{"lr": 0.00026525831390311475, "data_time": 0.0024079084396362305, "loss": 0.02605678914114833, "time": 0.8632853031158447, "epoch": 5, "memory": 25074, "step": 34700} +{"lr": 0.00026525831390311475, "data_time": 0.0014836788177490234, "loss": 0.020395143935456873, "time": 0.7161465406417846, "epoch": 5, "memory": 25074, "step": 34800} +{"lr": 0.00026525831390311475, "data_time": 0.00125274658203125, "loss": 0.024147900752723217, "time": 0.7409570932388305, "epoch": 5, "memory": 25074, "step": 34900} +{"lr": 0.00026525831390311475, "data_time": 0.0014492034912109374, "loss": 0.02286131465807557, "time": 0.7227227687835693, "epoch": 5, "memory": 25074, "step": 35000} +{"lr": 0.00026525831390311475, "data_time": 0.0022057533264160157, "loss": 0.02891218983568251, "time": 0.8394655466079712, "epoch": 5, "memory": 25074, "step": 35100} +{"lr": 0.00026525831390311475, "data_time": 0.0014395475387573241, "loss": 0.02146008899435401, "time": 0.7871113061904907, "epoch": 5, "memory": 25074, "step": 35200} +{"lr": 0.00026525831390311475, "data_time": 0.0012011766433715821, "loss": 0.021086252108216286, "time": 0.7503572940826416, "epoch": 5, "memory": 25074, "step": 35300} +{"lr": 0.00026525831390311475, "data_time": 0.001401805877685547, "loss": 0.020239979634061454, "time": 0.8190633058547974, "epoch": 5, "memory": 25074, "step": 35400} +{"lr": 0.00026525831390311475, "data_time": 0.004416513442993164, "loss": 0.022512492537498475, "time": 0.8431733608245849, "epoch": 5, "memory": 25074, "step": 35500} +{"lr": 0.00026525831390311475, "data_time": 0.0015740633010864259, "loss": 0.019313552416861057, "time": 0.7226876735687255, "epoch": 5, "memory": 25074, "step": 35600} +{"lr": 0.00026525831390311475, "data_time": 0.001483774185180664, "loss": 0.020557713508605958, "time": 0.7565377712249756, "epoch": 5, "memory": 25074, "step": 35700} +{"lr": 0.00026525831390311475, "data_time": 0.0013116836547851563, "loss": 0.010310977790504694, "time": 0.7280407428741456, "epoch": 5, "memory": 25074, "step": 35800} +{"lr": 0.00026525831390311475, "data_time": 0.0013847112655639648, "loss": 0.021481232857331632, "time": 0.8211503028869629, "epoch": 5, "memory": 25074, "step": 35900} +{"lr": 0.00026525831390311475, "data_time": 0.003250861167907715, "loss": 0.024764028284698723, "time": 0.8625778913497925, "epoch": 5, "memory": 25074, "step": 36000} +{"lr": 0.00026525831390311475, "data_time": 0.0012320518493652345, "loss": 0.02365124863572419, "time": 0.7453600883483886, "epoch": 5, "memory": 25074, "step": 36100} +{"lr": 0.00026525831390311475, "data_time": 0.0012217998504638673, "loss": 0.011829232051968574, "time": 0.742449951171875, "epoch": 5, "memory": 25074, "step": 36200} +{"lr": 0.00026525831390311475, "data_time": 0.0012722015380859375, "loss": 0.017892223224043845, "time": 0.7847510576248169, "epoch": 5, "memory": 25074, "step": 36300} +{"lr": 0.00026525831390311475, "data_time": 0.0013031482696533204, "loss": 0.020463507436215878, "time": 0.7392638683319092, "epoch": 5, "memory": 25074, "step": 36400} +{"lr": 0.00026525831390311475, "data_time": 0.002644181251525879, "loss": 0.02505156071856618, "time": 0.7775050878524781, "epoch": 5, "memory": 25074, "step": 36500} +{"lr": 0.00026525831390311475, "data_time": 0.0013347625732421874, "loss": 0.021378376707434654, "time": 0.7562998294830322, "epoch": 5, "memory": 25074, "step": 36600} +{"lr": 0.00026525831390311475, "data_time": 0.001228165626525879, "loss": 0.01717157321982086, "time": 0.7427967548370361, "epoch": 5, "memory": 25074, "step": 36700} +{"lr": 0.00026525831390311475, "data_time": 0.0012204170227050782, "loss": 0.011767417564988137, "time": 0.8289592027664184, "epoch": 5, "memory": 25074, "step": 36800} +{"lr": 0.00026525831390311475, "data_time": 0.002592658996582031, "loss": 0.016141739999875428, "time": 0.8661892890930176, "epoch": 5, "memory": 25074, "step": 36900} +{"lr": 0.00026525831390311475, "data_time": 0.0015212535858154298, "loss": 0.014416280947625638, "time": 0.7515527248382569, "epoch": 5, "memory": 25074, "step": 37000} +{"lr": 0.00026525831390311475, "data_time": 0.0018460512161254882, "loss": 0.023629384580999613, "time": 0.7796048879623413, "epoch": 5, "memory": 25074, "step": 37100} +{"lr": 0.00026525831390311475, "data_time": 0.0016292810440063476, "loss": 0.018551683891564606, "time": 0.7681916952133179, "epoch": 5, "memory": 25074, "step": 37200} +{"lr": 0.00026525831390311475, "data_time": 0.001635289192199707, "loss": 0.012822458054870367, "time": 0.7692230463027954, "epoch": 5, "memory": 25074, "step": 37300} +{"lr": 0.00026525831390311475, "data_time": 0.0017686128616333009, "loss": 0.01919706165790558, "time": 0.8141758441925049, "epoch": 5, "memory": 25074, "step": 37400} +{"lr": 0.00026525831390311475, "data_time": 0.001650524139404297, "loss": 0.01689677583053708, "time": 0.7051492691040039, "epoch": 5, "memory": 25074, "step": 37500} +{"lr": 0.00026525831390311475, "data_time": 0.001420307159423828, "loss": 0.018556368444114922, "time": 0.7511433362960815, "epoch": 5, "memory": 25074, "step": 37600} +{"lr": 0.00026525831390311475, "data_time": 0.001360917091369629, "loss": 0.014470644388347864, "time": 0.7846235990524292, "epoch": 5, "memory": 25074, "step": 37700} +{"lr": 0.00026525831390311475, "data_time": 0.001263904571533203, "loss": 0.016627713898196818, "time": 0.8151473760604858, "epoch": 5, "memory": 25074, "step": 37800} +{"lr": 0.00026525831390311475, "data_time": 0.0029468536376953125, "loss": 0.02148309852927923, "time": 0.8038137912750244, "epoch": 5, "memory": 25074, "step": 37900} +{"lr": 0.00026525831390311475, "data_time": 0.0013540267944335937, "loss": 0.02115346337668598, "time": 0.7132618188858032, "epoch": 5, "memory": 25074, "step": 38000} +{"lr": 0.00026525831390311475, "data_time": 0.0014809370040893555, "loss": 0.016541927028447388, "time": 0.7229875326156616, "epoch": 5, "memory": 25074, "step": 38100} +{"lr": 0.00026525831390311475, "data_time": 0.0013556718826293946, "loss": 0.025569619284942745, "time": 0.7170409440994263, "epoch": 5, "memory": 25074, "step": 38200} +{"lr": 0.00026525831390311475, "data_time": 0.002307558059692383, "loss": 0.01624313648790121, "time": 0.8092431783676147, "epoch": 5, "memory": 25074, "step": 38300} +{"lr": 0.00026525831390311475, "data_time": 0.001160883903503418, "loss": 0.011005748156458139, "time": 0.720248532295227, "epoch": 5, "memory": 25074, "step": 38400} +{"lr": 0.00026525831390311475, "data_time": 0.0014744281768798828, "loss": 0.021540593169629575, "time": 0.7683113098144532, "epoch": 5, "memory": 25074, "step": 38500} +{"lr": 0.00026525831390311475, "data_time": 0.0015828609466552734, "loss": 0.014804975502192974, "time": 0.7448345184326172, "epoch": 5, "memory": 25074, "step": 38600} +{"lr": 0.00026525831390311475, "data_time": 0.0012781620025634766, "loss": 0.020508582051843405, "time": 0.769232201576233, "epoch": 5, "memory": 25074, "step": 38700} +{"lr": 0.00026525831390311475, "data_time": 0.0035619020462036135, "loss": 0.01795935081318021, "time": 0.8573519229888916, "epoch": 5, "memory": 25074, "step": 38800} +{"lr": 0.00026525831390311475, "data_time": 0.0013607978820800782, "loss": 0.01790966363623738, "time": 0.7245409488677979, "epoch": 5, "memory": 25074, "step": 38900} +{"lr": 0.00026525831390311475, "data_time": 0.000868844985961914, "loss": 0.014371182140894233, "time": 0.6578127861022949, "epoch": 5, "memory": 25074, "step": 39000} +{"accuracy/top1": 99.55179595947266, "data_time": 0.1989019478068632, "time": 0.5841261919806985, "step": 5} +{"lr": 0.000205, "data_time": 0.0018460750579833984, "loss": 0.019231055956333877, "time": 0.7040271759033203, "epoch": 6, "memory": 25074, "step": 39100} +{"lr": 0.000205, "data_time": 0.001373624801635742, "loss": 0.013181134220212698, "time": 0.7180349349975585, "epoch": 6, "memory": 25074, "step": 39200} +{"lr": 0.000205, "data_time": 0.026488566398620607, "loss": 0.011611545691266656, "time": 0.7913643598556519, "epoch": 6, "memory": 25074, "step": 39300} +{"lr": 0.000205, "data_time": 0.0014906883239746093, "loss": 0.012931501772254706, "time": 0.7774279356002808, "epoch": 6, "memory": 25074, "step": 39400} +{"lr": 0.000205, "data_time": 0.0012401580810546876, "loss": 0.013847041875123978, "time": 0.7366981744766236, "epoch": 6, "memory": 25074, "step": 39500} +{"lr": 0.000205, "data_time": 0.0015658140182495117, "loss": 0.013702572509646415, "time": 0.7231918334960937, "epoch": 6, "memory": 25074, "step": 39600} +{"lr": 0.000205, "data_time": 0.001517486572265625, "loss": 0.012082140566781164, "time": 0.7332303762435913, "epoch": 6, "memory": 25074, "step": 39700} +{"lr": 0.000205, "data_time": 0.004176282882690429, "loss": 0.01720128827728331, "time": 0.8490522861480713, "epoch": 6, "memory": 25074, "step": 39800} +{"lr": 0.000205, "data_time": 0.0012623786926269532, "loss": 0.008695256197825074, "time": 0.7486298322677613, "epoch": 6, "memory": 25074, "step": 39900} +{"lr": 0.000205, "data_time": 0.0013205051422119141, "loss": 0.007066844450309873, "time": 0.7174515724182129, "epoch": 6, "memory": 25074, "step": 40000} +{"lr": 0.000205, "data_time": 0.0012791872024536133, "loss": 0.014360112557187676, "time": 0.7292573928833008, "epoch": 6, "memory": 25074, "step": 40100} +{"lr": 0.000205, "data_time": 0.003551483154296875, "loss": 0.016115556936711073, "time": 0.8143055677413941, "epoch": 6, "memory": 25074, "step": 40200} +{"lr": 0.000205, "data_time": 0.0013048410415649413, "loss": 0.0195564454421401, "time": 0.7255980014801026, "epoch": 6, "memory": 25074, "step": 40300} +{"lr": 0.000205, "data_time": 0.0015650749206542968, "loss": 0.008563556848093868, "time": 0.7180584669113159, "epoch": 6, "memory": 25074, "step": 40400} +{"lr": 0.000205, "data_time": 0.0014289617538452148, "loss": 0.015089165093377233, "time": 0.7049487590789795, "epoch": 6, "memory": 25074, "step": 40500} +{"lr": 0.000205, "data_time": 0.0012910842895507812, "loss": 0.01764496141113341, "time": 0.7559063673019409, "epoch": 6, "memory": 25074, "step": 40600} +{"lr": 0.000205, "data_time": 0.0015976667404174806, "loss": 0.014122887561097742, "time": 0.7320395708084106, "epoch": 6, "memory": 25074, "step": 40700} +{"lr": 0.000205, "data_time": 0.0014389753341674805, "loss": 0.017195624113082886, "time": 0.7043787002563476, "epoch": 6, "memory": 25074, "step": 40800} +{"lr": 0.000205, "data_time": 0.0014748096466064454, "loss": 0.012815610086545349, "time": 0.7041952848434448, "epoch": 6, "memory": 25074, "step": 40900} +{"lr": 0.000205, "data_time": 0.0013394355773925781, "loss": 0.017642981838434935, "time": 0.7030986309051513, "epoch": 6, "memory": 25074, "step": 41000} +{"lr": 0.000205, "data_time": 0.0019710779190063475, "loss": 0.009535460453480481, "time": 0.7235670328140259, "epoch": 6, "memory": 25074, "step": 41100} +{"lr": 0.000205, "data_time": 0.0012816190719604492, "loss": 0.015751146338880063, "time": 0.704103946685791, "epoch": 6, "memory": 25074, "step": 41200} +{"lr": 0.000205, "data_time": 0.0014745235443115235, "loss": 0.010906191961839794, "time": 0.7403566122055054, "epoch": 6, "memory": 25074, "step": 41300} +{"lr": 0.000205, "data_time": 0.0014739274978637696, "loss": 0.009607540257275105, "time": 0.7050373792648316, "epoch": 6, "memory": 25074, "step": 41400} +{"lr": 0.000205, "data_time": 0.0025877952575683594, "loss": 0.008496478525921702, "time": 0.7942193508148193, "epoch": 6, "memory": 25074, "step": 41500} +{"lr": 0.000205, "data_time": 0.0013800621032714843, "loss": 0.01434514089487493, "time": 0.7253471612930298, "epoch": 6, "memory": 25074, "step": 41600} +{"lr": 0.000205, "data_time": 0.0013710260391235352, "loss": 0.016336551727727057, "time": 0.7215386152267456, "epoch": 6, "memory": 25074, "step": 41700} +{"lr": 0.000205, "data_time": 0.0012293577194213868, "loss": 0.013037927681580186, "time": 0.7048011779785156, "epoch": 6, "memory": 25074, "step": 41800} +{"lr": 0.000205, "data_time": 0.002528524398803711, "loss": 0.012209090078249574, "time": 0.8043434858322144, "epoch": 6, "memory": 25074, "step": 41900} +{"lr": 0.000205, "data_time": 0.0016994237899780273, "loss": 0.012641871068626642, "time": 0.705679988861084, "epoch": 6, "memory": 25074, "step": 42000} +{"lr": 0.000205, "data_time": 0.0014030218124389648, "loss": 0.01131478725001216, "time": 0.7047741174697876, "epoch": 6, "memory": 25074, "step": 42100} +{"lr": 0.000205, "data_time": 0.0014711141586303711, "loss": 0.01450472166761756, "time": 0.7043525695800781, "epoch": 6, "memory": 25074, "step": 42200} +{"lr": 0.000205, "data_time": 0.0026247262954711913, "loss": 0.012451252155005931, "time": 0.802859902381897, "epoch": 6, "memory": 25074, "step": 42300} +{"lr": 0.000205, "data_time": 0.0013098955154418946, "loss": 0.009639228554442525, "time": 0.7166960716247559, "epoch": 6, "memory": 25074, "step": 42400} +{"lr": 0.000205, "data_time": 0.0013704538345336915, "loss": 0.015238004876300693, "time": 0.7345313310623169, "epoch": 6, "memory": 25074, "step": 42500} +{"lr": 0.000205, "data_time": 0.0014132976531982422, "loss": 0.011543279327452183, "time": 0.8145903587341309, "epoch": 6, "memory": 25074, "step": 42600} +{"lr": 0.000205, "data_time": 0.0020676136016845705, "loss": 0.008894957602024078, "time": 0.829864764213562, "epoch": 6, "memory": 25074, "step": 42700} +{"lr": 0.000205, "data_time": 0.0014184951782226563, "loss": 0.015117054292932152, "time": 0.7325458526611328, "epoch": 6, "memory": 25074, "step": 42800} +{"lr": 0.000205, "data_time": 0.0013321161270141602, "loss": 0.009235883736982942, "time": 0.7424086809158326, "epoch": 6, "memory": 25074, "step": 42900} +{"lr": 0.000205, "data_time": 0.0015939950942993164, "loss": 0.013493990153074264, "time": 0.7807001113891602, "epoch": 6, "memory": 25074, "step": 43000} +{"lr": 0.000205, "data_time": 0.001594853401184082, "loss": 0.01986273843795061, "time": 0.7461189031600952, "epoch": 6, "memory": 25074, "step": 43100} +{"lr": 0.000205, "data_time": 0.004760479927062989, "loss": 0.012629456259310246, "time": 0.9208427190780639, "epoch": 6, "memory": 25074, "step": 43200} +{"lr": 0.000205, "data_time": 0.0015197992324829102, "loss": 0.010611542407423258, "time": 0.7424284934997558, "epoch": 6, "memory": 25074, "step": 43300} +{"lr": 0.000205, "data_time": 0.0013847827911376953, "loss": 0.014244463993236423, "time": 0.7358451128005982, "epoch": 6, "memory": 25074, "step": 43400} +{"lr": 0.000205, "data_time": 0.0013418912887573243, "loss": 0.011222891882061959, "time": 0.7344607591629029, "epoch": 6, "memory": 25074, "step": 43500} +{"lr": 0.000205, "data_time": 0.019410061836242675, "loss": 0.01397267752327025, "time": 0.7354110717773438, "epoch": 6, "memory": 25074, "step": 43600} +{"lr": 0.000205, "data_time": 0.023637723922729493, "loss": 0.015269479947164655, "time": 0.7457617521286011, "epoch": 6, "memory": 25074, "step": 43700} +{"lr": 0.000205, "data_time": 0.04947209358215332, "loss": 0.013520948542281985, "time": 0.7643749952316284, "epoch": 6, "memory": 25074, "step": 43800} +{"lr": 0.000205, "data_time": 0.08439290523529053, "loss": 0.009138938132673502, "time": 0.7958259820938111, "epoch": 6, "memory": 25074, "step": 43900} +{"lr": 0.000205, "data_time": 0.0013115644454956056, "loss": 0.011309412447735668, "time": 0.7158514261245728, "epoch": 6, "memory": 25074, "step": 44000} +{"lr": 0.000205, "data_time": 0.19720866680145263, "loss": 0.0143601362593472, "time": 0.9396367073059082, "epoch": 6, "memory": 25074, "step": 44100} +{"lr": 0.000205, "data_time": 0.08205864429473878, "loss": 0.012898083496838807, "time": 0.7914128303527832, "epoch": 6, "memory": 25074, "step": 44200} +{"lr": 0.000205, "data_time": 0.050267791748046874, "loss": 0.009320208802819252, "time": 0.7634162664413452, "epoch": 6, "memory": 25074, "step": 44300} +{"lr": 0.000205, "data_time": 0.05628128051757812, "loss": 0.008199787745252252, "time": 0.7752848386764526, "epoch": 6, "memory": 25074, "step": 44400} +{"lr": 0.000205, "data_time": 0.06082310676574707, "loss": 0.009938874375075102, "time": 0.7634033918380737, "epoch": 6, "memory": 25074, "step": 44500} +{"lr": 0.000205, "data_time": 0.018892621994018553, "loss": 0.007618839293718338, "time": 0.784738826751709, "epoch": 6, "memory": 25074, "step": 44600} +{"lr": 0.000205, "data_time": 0.026316905021667482, "loss": 0.011888827150687575, "time": 0.7341613054275513, "epoch": 6, "memory": 25074, "step": 44700} +{"lr": 0.000205, "data_time": 0.0013403892517089844, "loss": 0.012039583222940565, "time": 0.7232708692550659, "epoch": 6, "memory": 25074, "step": 44800} +{"lr": 0.000205, "data_time": 0.09245812892913818, "loss": 0.00985512938350439, "time": 0.8072475910186767, "epoch": 6, "memory": 25074, "step": 44900} +{"lr": 0.000205, "data_time": 0.1281057596206665, "loss": 0.014563016546890139, "time": 0.8686982870101929, "epoch": 6, "memory": 25074, "step": 45000} +{"lr": 0.000205, "data_time": 0.0011036157608032226, "loss": 0.007399301510304213, "time": 0.703536868095398, "epoch": 6, "memory": 25074, "step": 45100} +{"lr": 0.000205, "data_time": 0.0012583732604980469, "loss": 0.010445272270590066, "time": 0.7305407047271728, "epoch": 6, "memory": 25074, "step": 45200} +{"lr": 0.000205, "data_time": 0.0014368534088134766, "loss": 0.0144415148999542, "time": 0.749235200881958, "epoch": 6, "memory": 25074, "step": 45300} +{"lr": 0.000205, "data_time": 0.0014327049255371093, "loss": 0.015263443626463414, "time": 0.7857434749603271, "epoch": 6, "memory": 25074, "step": 45400} +{"lr": 0.000205, "data_time": 0.0015762567520141602, "loss": 0.01437802640721202, "time": 0.7276469945907593, "epoch": 6, "memory": 25074, "step": 45500} +{"lr": 0.000205, "data_time": 0.001499176025390625, "loss": 0.011055087158456444, "time": 0.7303604364395142, "epoch": 6, "memory": 25074, "step": 45600} +{"lr": 0.000205, "data_time": 0.001480841636657715, "loss": 0.010005244286730886, "time": 0.8664628028869629, "epoch": 6, "memory": 25074, "step": 45700} +{"lr": 0.000205, "data_time": 0.001304936408996582, "loss": 0.010457220673561095, "time": 0.8030084848403931, "epoch": 6, "memory": 25074, "step": 45800} +{"lr": 0.000205, "data_time": 0.002091646194458008, "loss": 0.012011106358841062, "time": 0.7231206893920898, "epoch": 6, "memory": 25074, "step": 45900} +{"lr": 0.000205, "data_time": 0.0034818410873413085, "loss": 0.010440809139981865, "time": 0.8994972467422485, "epoch": 6, "memory": 25074, "step": 46000} +{"lr": 0.000205, "data_time": 0.001363849639892578, "loss": 0.012036335561424494, "time": 0.7369155406951904, "epoch": 6, "memory": 25074, "step": 46100} +{"lr": 0.000205, "data_time": 0.0013537883758544921, "loss": 0.009035256505012513, "time": 0.7985840797424316, "epoch": 6, "memory": 25074, "step": 46200} +{"lr": 0.000205, "data_time": 0.0016448259353637694, "loss": 0.009535867674276232, "time": 0.7629476547241211, "epoch": 6, "memory": 25074, "step": 46300} +{"lr": 0.000205, "data_time": 0.0013123035430908203, "loss": 0.01597025040537119, "time": 0.7717717885971069, "epoch": 6, "memory": 25074, "step": 46400} +{"lr": 0.000205, "data_time": 0.002623105049133301, "loss": 0.01929229013621807, "time": 0.8350199222564697, "epoch": 6, "memory": 25074, "step": 46500} +{"lr": 0.000205, "data_time": 0.0015046358108520507, "loss": 0.014013120532035827, "time": 0.7858851432800293, "epoch": 6, "memory": 25074, "step": 46600} +{"lr": 0.000205, "data_time": 0.0012997627258300782, "loss": 0.008184753078967332, "time": 0.7804155826568604, "epoch": 6, "memory": 25074, "step": 46700} +{"lr": 0.000205, "data_time": 0.0014567613601684571, "loss": 0.009931951481848954, "time": 0.6597428798675538, "epoch": 6, "memory": 25074, "step": 46800} +{"accuracy/top1": 99.867431640625, "data_time": 0.2033681448768167, "time": 0.5902450505424949, "step": 6} +{"lr": 0.00014474168609688527, "data_time": 0.002039980888366699, "loss": 0.008903268352150917, "time": 0.8040812969207763, "epoch": 7, "memory": 25074, "step": 46900} +{"lr": 0.00014474168609688527, "data_time": 0.0019177913665771485, "loss": 0.009964527282863855, "time": 0.7048364162445069, "epoch": 7, "memory": 25074, "step": 47000} +{"lr": 0.00014474168609688527, "data_time": 0.0019239187240600586, "loss": 0.0071931109763681885, "time": 0.9699006319046021, "epoch": 7, "memory": 25074, "step": 47100} +{"lr": 0.00014474168609688527, "data_time": 0.0014249086380004883, "loss": 0.0061600552406162025, "time": 0.7036574125289917, "epoch": 7, "memory": 25074, "step": 47200} +{"lr": 0.00014474168609688527, "data_time": 0.0013582229614257813, "loss": 0.009402553550899029, "time": 0.7363144636154175, "epoch": 7, "memory": 25074, "step": 47300} +{"lr": 0.00014474168609688527, "data_time": 0.0013524293899536133, "loss": 0.008217450184747576, "time": 0.7047011137008667, "epoch": 7, "memory": 25074, "step": 47400} +{"lr": 0.00014474168609688527, "data_time": 0.0012930870056152345, "loss": 0.0057783856522291895, "time": 0.8039856195449829, "epoch": 7, "memory": 25074, "step": 47500} +{"lr": 0.00014474168609688527, "data_time": 0.001946091651916504, "loss": 0.010270766355097294, "time": 0.7182083606719971, "epoch": 7, "memory": 25074, "step": 47600} +{"lr": 0.00014474168609688527, "data_time": 0.0015686273574829102, "loss": 0.008132395148277283, "time": 0.8044255256652832, "epoch": 7, "memory": 25074, "step": 47700} +{"lr": 0.00014474168609688527, "data_time": 0.0013362407684326173, "loss": 0.007332683773711324, "time": 0.7166149854660034, "epoch": 7, "memory": 25074, "step": 47800} +{"lr": 0.00014474168609688527, "data_time": 0.0015433788299560546, "loss": 0.0035177916288375854, "time": 0.7934905767440796, "epoch": 7, "memory": 25074, "step": 47900} +{"lr": 0.00014474168609688527, "data_time": 0.0037358283996582033, "loss": 0.009405344678089022, "time": 0.8660728454589843, "epoch": 7, "memory": 25074, "step": 48000} +{"lr": 0.00014474168609688527, "data_time": 0.001355457305908203, "loss": 0.012066180864349008, "time": 0.7574211120605469, "epoch": 7, "memory": 25074, "step": 48100} +{"lr": 0.00014474168609688527, "data_time": 0.0015769481658935546, "loss": 0.008158619981259108, "time": 0.7830992698669433, "epoch": 7, "memory": 25074, "step": 48200} +{"lr": 0.00014474168609688527, "data_time": 0.0012317657470703124, "loss": 0.008703110832720995, "time": 0.7489234209060669, "epoch": 7, "memory": 25074, "step": 48300} +{"lr": 0.00014474168609688527, "data_time": 0.0014250516891479493, "loss": 0.00903176930733025, "time": 0.8023950815200805, "epoch": 7, "memory": 25074, "step": 48400} +{"lr": 0.00014474168609688527, "data_time": 0.0025968074798583983, "loss": 0.004749373253434897, "time": 0.8814101219177246, "epoch": 7, "memory": 25074, "step": 48500} +{"lr": 0.00014474168609688527, "data_time": 0.0013814449310302734, "loss": 0.009541194513440132, "time": 0.7099181175231933, "epoch": 7, "memory": 25074, "step": 48600} +{"lr": 0.00014474168609688527, "data_time": 0.00133209228515625, "loss": 0.01119645982980728, "time": 0.7453160285949707, "epoch": 7, "memory": 25074, "step": 48700} +{"lr": 0.00014474168609688527, "data_time": 0.0013719797134399414, "loss": 0.005983489099889994, "time": 0.7479058742523194, "epoch": 7, "memory": 25074, "step": 48800} +{"lr": 0.00014474168609688527, "data_time": 0.0022298812866210936, "loss": 0.008830545051023365, "time": 0.8155020713806153, "epoch": 7, "memory": 25074, "step": 48900} +{"lr": 0.00014474168609688527, "data_time": 0.0013176918029785157, "loss": 0.009136131405830384, "time": 0.7293948173522949, "epoch": 7, "memory": 25074, "step": 49000} +{"lr": 0.00014474168609688527, "data_time": 0.001520228385925293, "loss": 0.009767140028998256, "time": 0.8100643157958984, "epoch": 7, "memory": 25074, "step": 49100} +{"lr": 0.00014474168609688527, "data_time": 0.0013205051422119141, "loss": 0.007960773445665837, "time": 0.8000875949859619, "epoch": 7, "memory": 25074, "step": 49200} +{"lr": 0.00014474168609688527, "data_time": 0.0013315200805664063, "loss": 0.008458681823685765, "time": 0.7050417423248291, "epoch": 7, "memory": 25074, "step": 49300} +{"lr": 0.00014474168609688527, "data_time": 0.0031881809234619142, "loss": 0.006933752307668328, "time": 0.9377498865127564, "epoch": 7, "memory": 25074, "step": 49400} +{"lr": 0.00014474168609688527, "data_time": 0.0013633966445922852, "loss": 0.004100177623331547, "time": 0.7347213983535766, "epoch": 7, "memory": 25074, "step": 49500} +{"lr": 0.00014474168609688527, "data_time": 0.0013150691986083985, "loss": 0.0046753640286624435, "time": 0.7202700138092041, "epoch": 7, "memory": 25074, "step": 49600} +{"lr": 0.00014474168609688527, "data_time": 0.0015794992446899413, "loss": 0.012539870711043477, "time": 0.7055185079574585, "epoch": 7, "memory": 25074, "step": 49700} +{"lr": 0.00014474168609688527, "data_time": 0.0012950897216796875, "loss": 0.010754825919866562, "time": 0.815887188911438, "epoch": 7, "memory": 25074, "step": 49800} +{"lr": 0.00014474168609688527, "data_time": 0.0016344070434570312, "loss": 0.003147457307204604, "time": 0.7243025064468384, "epoch": 7, "memory": 25074, "step": 49900} +{"lr": 0.00014474168609688527, "data_time": 0.0014695405960083007, "loss": 0.007378797419369221, "time": 0.7496392250061035, "epoch": 7, "memory": 25074, "step": 50000} +{"lr": 0.00014474168609688527, "data_time": 0.0012321233749389648, "loss": 0.006423063715919853, "time": 0.7316712141036987, "epoch": 7, "memory": 25074, "step": 50100} +{"lr": 0.00014474168609688527, "data_time": 0.0013420820236206056, "loss": 0.005607046559453011, "time": 0.7059974908828736, "epoch": 7, "memory": 25074, "step": 50200} +{"lr": 0.00014474168609688527, "data_time": 0.003515005111694336, "loss": 0.009945366438478232, "time": 0.8344825744628906, "epoch": 7, "memory": 25074, "step": 50300} +{"lr": 0.00014474168609688527, "data_time": 0.0015527963638305663, "loss": 0.010599404806271195, "time": 0.7681415557861329, "epoch": 7, "memory": 25074, "step": 50400} +{"lr": 0.00014474168609688527, "data_time": 0.001502251625061035, "loss": 0.007424869574606419, "time": 0.722867465019226, "epoch": 7, "memory": 25074, "step": 50500} +{"lr": 0.00014474168609688527, "data_time": 0.001420450210571289, "loss": 0.006481426581740379, "time": 0.7126387357711792, "epoch": 7, "memory": 25074, "step": 50600} +{"lr": 0.00014474168609688527, "data_time": 0.003195929527282715, "loss": 0.007120138918980956, "time": 0.8828189373016357, "epoch": 7, "memory": 25074, "step": 50700} +{"lr": 0.00014474168609688527, "data_time": 0.0014522075653076172, "loss": 0.006790314661338925, "time": 0.7370819568634033, "epoch": 7, "memory": 25074, "step": 50800} +{"lr": 0.00014474168609688527, "data_time": 0.001359868049621582, "loss": 0.007473695790395141, "time": 0.7462273120880127, "epoch": 7, "memory": 25074, "step": 50900} +{"lr": 0.00014474168609688527, "data_time": 0.0014990568161010742, "loss": 0.0059722459875047205, "time": 0.7586614608764648, "epoch": 7, "memory": 25074, "step": 51000} +{"lr": 0.00014474168609688527, "data_time": 0.0012622833251953124, "loss": 0.007621635589748621, "time": 0.808885145187378, "epoch": 7, "memory": 25074, "step": 51100} +{"lr": 0.00014474168609688527, "data_time": 0.0029109716415405273, "loss": 0.010983528615906835, "time": 0.8470036745071411, "epoch": 7, "memory": 25074, "step": 51200} +{"lr": 0.00014474168609688527, "data_time": 0.0014332294464111327, "loss": 0.01582004497759044, "time": 0.7581027507781982, "epoch": 7, "memory": 25074, "step": 51300} +{"lr": 0.00014474168609688527, "data_time": 0.0013092041015625, "loss": 0.008110878337174655, "time": 0.747739315032959, "epoch": 7, "memory": 25074, "step": 51400} +{"lr": 0.00014474168609688527, "data_time": 0.0013184070587158204, "loss": 0.009489529207348824, "time": 0.7333887577056885, "epoch": 7, "memory": 25074, "step": 51500} +{"lr": 0.00014474168609688527, "data_time": 0.0023162364959716797, "loss": 0.007917167106643318, "time": 0.8428509473800659, "epoch": 7, "memory": 25074, "step": 51600} +{"lr": 0.00014474168609688527, "data_time": 0.0012856245040893555, "loss": 0.007682484900578856, "time": 0.7118677139282227, "epoch": 7, "memory": 25074, "step": 51700} +{"lr": 0.00014474168609688527, "data_time": 0.0013965845108032226, "loss": 0.009817048953846098, "time": 0.7641555547714234, "epoch": 7, "memory": 25074, "step": 51800} +{"lr": 0.00014474168609688527, "data_time": 0.0015497207641601562, "loss": 0.005965145118534565, "time": 0.7311853170394897, "epoch": 7, "memory": 25074, "step": 51900} +{"lr": 0.00014474168609688527, "data_time": 0.0015300989151000976, "loss": 0.006508117076009512, "time": 0.7900309562683105, "epoch": 7, "memory": 25074, "step": 52000} +{"lr": 0.00014474168609688527, "data_time": 0.003158879280090332, "loss": 0.010603367071598768, "time": 0.7965076923370361, "epoch": 7, "memory": 25074, "step": 52100} +{"lr": 0.00014474168609688527, "data_time": 0.0012988567352294922, "loss": 0.010281467298045754, "time": 0.7051944017410279, "epoch": 7, "memory": 25074, "step": 52200} +{"lr": 0.00014474168609688527, "data_time": 0.0014504671096801757, "loss": 0.006418878398835659, "time": 0.7626439094543457, "epoch": 7, "memory": 25074, "step": 52300} +{"lr": 0.00014474168609688527, "data_time": 0.0013737916946411134, "loss": 0.010212240228429436, "time": 0.7140905618667602, "epoch": 7, "memory": 25074, "step": 52400} +{"lr": 0.00014474168609688527, "data_time": 0.0014587640762329102, "loss": 0.007123590679839253, "time": 0.7039916038513183, "epoch": 7, "memory": 25074, "step": 52500} +{"lr": 0.00014474168609688527, "data_time": 0.0014328956604003906, "loss": 0.0073842604644596575, "time": 0.7820403099060058, "epoch": 7, "memory": 25074, "step": 52600} +{"lr": 0.00014474168609688527, "data_time": 0.003953814506530762, "loss": 0.007126175705343485, "time": 0.9153799057006836, "epoch": 7, "memory": 25074, "step": 52700} +{"lr": 0.00014474168609688527, "data_time": 0.0012418746948242188, "loss": 0.005464016925543547, "time": 0.7207589387893677, "epoch": 7, "memory": 25074, "step": 52800} +{"lr": 0.00014474168609688527, "data_time": 0.001244354248046875, "loss": 0.005571501096710563, "time": 0.7073694944381714, "epoch": 7, "memory": 25074, "step": 52900} +{"lr": 0.00014474168609688527, "data_time": 0.0013068437576293946, "loss": 0.003804550599306822, "time": 0.7069416046142578, "epoch": 7, "memory": 25074, "step": 53000} +{"lr": 0.00014474168609688527, "data_time": 0.00139310359954834, "loss": 0.006326936697587371, "time": 0.7342127561569214, "epoch": 7, "memory": 25074, "step": 53100} +{"lr": 0.00014474168609688527, "data_time": 0.002085375785827637, "loss": 0.007772488426417113, "time": 0.7185160398483277, "epoch": 7, "memory": 25074, "step": 53200} +{"lr": 0.00014474168609688527, "data_time": 0.0014556169509887696, "loss": 0.006153023708611727, "time": 0.7050572395324707, "epoch": 7, "memory": 25074, "step": 53300} +{"lr": 0.00014474168609688527, "data_time": 0.0013092279434204102, "loss": 0.006335545424371957, "time": 0.7160191535949707, "epoch": 7, "memory": 25074, "step": 53400} +{"lr": 0.00014474168609688527, "data_time": 0.001264190673828125, "loss": 0.008011970529332757, "time": 0.7320989847183228, "epoch": 7, "memory": 25074, "step": 53500} +{"lr": 0.00014474168609688527, "data_time": 0.0017467737197875977, "loss": 0.008451334619894623, "time": 0.7773605346679687, "epoch": 7, "memory": 25074, "step": 53600} +{"lr": 0.00014474168609688527, "data_time": 0.0014884233474731444, "loss": 0.005294105829671025, "time": 0.7955523014068604, "epoch": 7, "memory": 25074, "step": 53700} +{"lr": 0.00014474168609688527, "data_time": 0.0014178752899169922, "loss": 0.008219748688861727, "time": 0.7150989532470703, "epoch": 7, "memory": 25074, "step": 53800} +{"lr": 0.00014474168609688527, "data_time": 0.0013798236846923827, "loss": 0.005182515317574144, "time": 0.7039018630981445, "epoch": 7, "memory": 25074, "step": 53900} +{"lr": 0.00014474168609688527, "data_time": 0.0037261962890625, "loss": 0.0045313382055610415, "time": 0.90395667552948, "epoch": 7, "memory": 25074, "step": 54000} +{"lr": 0.00014474168609688527, "data_time": 0.0015595197677612306, "loss": 0.016572838090360163, "time": 0.7274059295654297, "epoch": 7, "memory": 25074, "step": 54100} +{"lr": 0.00014474168609688527, "data_time": 0.0014645814895629882, "loss": 0.0096655139233917, "time": 0.7063659429550171, "epoch": 7, "memory": 25074, "step": 54200} +{"lr": 0.00014474168609688527, "data_time": 0.0014026403427124024, "loss": 0.006036846432834864, "time": 0.735233473777771, "epoch": 7, "memory": 25074, "step": 54300} +{"lr": 0.00014474168609688527, "data_time": 0.0014257431030273438, "loss": 0.006519056530669332, "time": 0.7424371480941773, "epoch": 7, "memory": 25074, "step": 54400} +{"lr": 0.00014474168609688527, "data_time": 0.0028959989547729494, "loss": 0.014014235092326998, "time": 0.7895010232925415, "epoch": 7, "memory": 25074, "step": 54500} +{"lr": 0.00014474168609688527, "data_time": 0.0012877225875854493, "loss": 0.007901851180940866, "time": 0.6710731506347656, "epoch": 7, "memory": 25074, "step": 54600} +{"accuracy/top1": 99.94949340820312, "data_time": 0.20048201785368078, "time": 0.5843884664423326, "step": 7} +{"lr": 9.038187580296776e-05, "data_time": 0.0014898061752319336, "loss": 0.002261943044140935, "time": 0.7649091720581055, "epoch": 8, "memory": 25074, "step": 54700} +{"lr": 9.038187580296776e-05, "data_time": 0.0014862775802612304, "loss": 0.0029871654231101276, "time": 0.7620866298675537, "epoch": 8, "memory": 25074, "step": 54800} +{"lr": 9.038187580296776e-05, "data_time": 0.0014408349990844727, "loss": 0.004256152547895908, "time": 0.744843864440918, "epoch": 8, "memory": 25074, "step": 54900} +{"lr": 9.038187580296776e-05, "data_time": 0.0016133308410644532, "loss": 0.010352508304640651, "time": 0.7147590637207031, "epoch": 8, "memory": 25074, "step": 55000} +{"lr": 9.038187580296776e-05, "data_time": 0.0025226354598999025, "loss": 0.006565147684887052, "time": 0.7641940832138061, "epoch": 8, "memory": 25074, "step": 55100} +{"lr": 9.038187580296776e-05, "data_time": 0.0014429092407226562, "loss": 0.004011364700272679, "time": 0.7107864141464233, "epoch": 8, "memory": 25074, "step": 55200} +{"lr": 9.038187580296776e-05, "data_time": 0.0015347719192504883, "loss": 0.00426431018859148, "time": 0.7742004632949829, "epoch": 8, "memory": 25074, "step": 55300} +{"lr": 9.038187580296776e-05, "data_time": 0.0013683557510375977, "loss": 0.0031507344916462897, "time": 0.7537154436111451, "epoch": 8, "memory": 25074, "step": 55400} +{"lr": 9.038187580296776e-05, "data_time": 0.002106904983520508, "loss": 0.004265059903264046, "time": 0.8996085882186889, "epoch": 8, "memory": 25074, "step": 55500} +{"lr": 9.038187580296776e-05, "data_time": 0.0014028549194335938, "loss": 0.006714503979310393, "time": 0.7374112844467163, "epoch": 8, "memory": 25074, "step": 55600} +{"lr": 9.038187580296776e-05, "data_time": 0.001227593421936035, "loss": 0.013006808469071984, "time": 0.7286698341369628, "epoch": 8, "memory": 25074, "step": 55700} +{"lr": 9.038187580296776e-05, "data_time": 0.0014836549758911132, "loss": 0.010702064307406545, "time": 0.7369499444961548, "epoch": 8, "memory": 25074, "step": 55800} +{"lr": 9.038187580296776e-05, "data_time": 0.0013474702835083007, "loss": 0.01078363386914134, "time": 0.7149914026260376, "epoch": 8, "memory": 25074, "step": 55900} +{"lr": 9.038187580296776e-05, "data_time": 0.00410001277923584, "loss": 0.0045760229229927065, "time": 0.7505262851715088, "epoch": 8, "memory": 25074, "step": 56000} +{"lr": 9.038187580296776e-05, "data_time": 0.0012885332107543945, "loss": 0.0060073901433497666, "time": 0.7120464324951172, "epoch": 8, "memory": 25074, "step": 56100} +{"lr": 9.038187580296776e-05, "data_time": 0.0015423059463500976, "loss": 0.007365959184244275, "time": 0.7305982828140258, "epoch": 8, "memory": 25074, "step": 56200} +{"lr": 9.038187580296776e-05, "data_time": 0.001609945297241211, "loss": 0.005678521329537034, "time": 0.7192543268203735, "epoch": 8, "memory": 25074, "step": 56300} +{"lr": 9.038187580296776e-05, "data_time": 0.002950239181518555, "loss": 0.005616268655285239, "time": 0.7835325717926025, "epoch": 8, "memory": 25074, "step": 56400} +{"lr": 9.038187580296776e-05, "data_time": 0.001552271842956543, "loss": 0.004612666042521596, "time": 0.7550269365310669, "epoch": 8, "memory": 25074, "step": 56500} +{"lr": 9.038187580296776e-05, "data_time": 0.001545858383178711, "loss": 0.005139764491468668, "time": 0.7496322631835938, "epoch": 8, "memory": 25074, "step": 56600} +{"lr": 9.038187580296776e-05, "data_time": 0.0013528108596801759, "loss": 0.005035810358822346, "time": 0.7202007293701171, "epoch": 8, "memory": 25074, "step": 56700} +{"lr": 9.038187580296776e-05, "data_time": 0.0015584230422973633, "loss": 0.006232081679627299, "time": 0.7533228635787964, "epoch": 8, "memory": 25074, "step": 56800} +{"lr": 9.038187580296776e-05, "data_time": 0.0013693571090698242, "loss": 0.004599780263379216, "time": 0.7272397994995117, "epoch": 8, "memory": 25074, "step": 56900} +{"lr": 9.038187580296776e-05, "data_time": 0.0016028404235839844, "loss": 0.006653623189777136, "time": 0.7921180248260498, "epoch": 8, "memory": 25074, "step": 57000} +{"lr": 9.038187580296776e-05, "data_time": 0.0014243125915527344, "loss": 0.0017732460051774978, "time": 0.739703369140625, "epoch": 8, "memory": 25074, "step": 57100} +{"lr": 9.038187580296776e-05, "data_time": 0.001279759407043457, "loss": 0.005414340365678072, "time": 0.7160408020019531, "epoch": 8, "memory": 25074, "step": 57200} +{"lr": 9.038187580296776e-05, "data_time": 0.001488471031188965, "loss": 0.005453924834728241, "time": 0.7304423570632934, "epoch": 8, "memory": 25074, "step": 57300} +{"lr": 9.038187580296776e-05, "data_time": 0.0015074729919433594, "loss": 0.0060550330206751825, "time": 0.7392881631851196, "epoch": 8, "memory": 25074, "step": 57400} +{"lr": 9.038187580296776e-05, "data_time": 0.0016181468963623047, "loss": 0.0031339862383902074, "time": 0.8159468412399292, "epoch": 8, "memory": 25074, "step": 57500} +{"lr": 9.038187580296776e-05, "data_time": 0.0014441967010498046, "loss": 0.005521983653306961, "time": 0.758598780632019, "epoch": 8, "memory": 25074, "step": 57600} +{"lr": 9.038187580296776e-05, "data_time": 0.0013397455215454102, "loss": 0.00791026996448636, "time": 0.7384653091430664, "epoch": 8, "memory": 25074, "step": 57700} +{"lr": 9.038187580296776e-05, "data_time": 0.001361536979675293, "loss": 0.007311791973188519, "time": 0.7191442966461181, "epoch": 8, "memory": 25074, "step": 57800} +{"lr": 9.038187580296776e-05, "data_time": 0.001282072067260742, "loss": 0.009632058395072817, "time": 0.7039193868637085, "epoch": 8, "memory": 25074, "step": 57900} +{"lr": 9.038187580296776e-05, "data_time": 0.0013043403625488282, "loss": 0.008204174460843205, "time": 0.7197263240814209, "epoch": 8, "memory": 25074, "step": 58000} +{"lr": 9.038187580296776e-05, "data_time": 0.001505422592163086, "loss": 0.0016403344925493002, "time": 0.7055933952331543, "epoch": 8, "memory": 25074, "step": 58100} +{"lr": 9.038187580296776e-05, "data_time": 0.0013222694396972656, "loss": 0.008053093031048775, "time": 0.7270200490951538, "epoch": 8, "memory": 25074, "step": 58200} +{"lr": 9.038187580296776e-05, "data_time": 0.0013428688049316405, "loss": 0.004103085165843368, "time": 0.7339093923568726, "epoch": 8, "memory": 25074, "step": 58300} +{"lr": 9.038187580296776e-05, "data_time": 0.01665949821472168, "loss": 0.0034529171418398617, "time": 0.7325861930847168, "epoch": 8, "memory": 25074, "step": 58400} +{"lr": 9.038187580296776e-05, "data_time": 0.013081240653991699, "loss": 0.0036817141342908146, "time": 0.7154910802841187, "epoch": 8, "memory": 25074, "step": 58500} +{"lr": 9.038187580296776e-05, "data_time": 0.0013049125671386718, "loss": 0.005961047857999802, "time": 0.7240694522857666, "epoch": 8, "memory": 25074, "step": 58600} +{"lr": 9.038187580296776e-05, "data_time": 0.0013455629348754882, "loss": 0.007555626705288887, "time": 0.7034820556640625, "epoch": 8, "memory": 25074, "step": 58700} +{"lr": 9.038187580296776e-05, "data_time": 0.016745781898498534, "loss": 0.011000830959528685, "time": 0.7352453708648682, "epoch": 8, "memory": 25074, "step": 58800} +{"lr": 9.038187580296776e-05, "data_time": 0.0015687942504882812, "loss": 0.004430231545120478, "time": 0.7043278932571411, "epoch": 8, "memory": 25074, "step": 58900} +{"lr": 9.038187580296776e-05, "data_time": 0.0013176918029785157, "loss": 0.003918724087998271, "time": 0.7339490413665771, "epoch": 8, "memory": 25074, "step": 59000} +{"lr": 9.038187580296776e-05, "data_time": 0.0012527942657470704, "loss": 0.005688633117824793, "time": 0.7037309885025025, "epoch": 8, "memory": 25074, "step": 59100} +{"lr": 9.038187580296776e-05, "data_time": 0.0016512632369995116, "loss": 0.007635505963116884, "time": 0.7101470708847046, "epoch": 8, "memory": 25074, "step": 59200} +{"lr": 9.038187580296776e-05, "data_time": 0.001399683952331543, "loss": 0.0037501564249396322, "time": 0.7043220520019531, "epoch": 8, "memory": 25074, "step": 59300} +{"lr": 9.038187580296776e-05, "data_time": 0.0013820171356201173, "loss": 0.007279509259387851, "time": 0.790700364112854, "epoch": 8, "memory": 25074, "step": 59400} +{"lr": 9.038187580296776e-05, "data_time": 0.001435232162475586, "loss": 0.0032900267280638216, "time": 0.7489850521087646, "epoch": 8, "memory": 25074, "step": 59500} +{"lr": 9.038187580296776e-05, "data_time": 0.0012386560440063477, "loss": 0.006933847209438682, "time": 0.7489638805389405, "epoch": 8, "memory": 25074, "step": 59600} +{"lr": 9.038187580296776e-05, "data_time": 0.001439070701599121, "loss": 0.010818727081641555, "time": 0.8164791822433471, "epoch": 8, "memory": 25074, "step": 59700} +{"lr": 9.038187580296776e-05, "data_time": 0.0014067888259887695, "loss": 0.010326781868934631, "time": 0.7552563905715942, "epoch": 8, "memory": 25074, "step": 59800} +{"lr": 9.038187580296776e-05, "data_time": 0.0012593030929565429, "loss": 0.0027001675684005023, "time": 0.7751612663269043, "epoch": 8, "memory": 25074, "step": 59900} +{"lr": 9.038187580296776e-05, "data_time": 0.0016253471374511718, "loss": 0.004097736719995737, "time": 0.7999375343322754, "epoch": 8, "memory": 25074, "step": 60000} +{"lr": 9.038187580296776e-05, "data_time": 0.0016825199127197266, "loss": 0.0023204955738037824, "time": 0.7047548770904541, "epoch": 8, "memory": 25074, "step": 60100} +{"lr": 9.038187580296776e-05, "data_time": 0.06462223529815674, "loss": 0.0038289979100227358, "time": 0.7878780364990234, "epoch": 8, "memory": 25074, "step": 60200} +{"lr": 9.038187580296776e-05, "data_time": 0.02077195644378662, "loss": 0.00469278828240931, "time": 0.723414421081543, "epoch": 8, "memory": 25074, "step": 60300} +{"lr": 9.038187580296776e-05, "data_time": 0.053536486625671384, "loss": 0.00837267953902483, "time": 0.7716482639312744, "epoch": 8, "memory": 25074, "step": 60400} +{"lr": 9.038187580296776e-05, "data_time": 0.07309715747833252, "loss": 0.0052996464539319275, "time": 0.7762666463851928, "epoch": 8, "memory": 25074, "step": 60500} +{"lr": 9.038187580296776e-05, "data_time": 0.006291007995605469, "loss": 0.003254408948123455, "time": 0.7316439628601075, "epoch": 8, "memory": 25074, "step": 60600} +{"lr": 9.038187580296776e-05, "data_time": 0.026555657386779785, "loss": 0.0037077505141496657, "time": 0.729297137260437, "epoch": 8, "memory": 25074, "step": 60700} +{"lr": 9.038187580296776e-05, "data_time": 0.022004365921020508, "loss": 0.007412866549566388, "time": 0.7237049818038941, "epoch": 8, "memory": 25074, "step": 60800} +{"lr": 9.038187580296776e-05, "data_time": 0.011216974258422852, "loss": 0.005148720880970359, "time": 0.7157186269760132, "epoch": 8, "memory": 25074, "step": 60900} +{"lr": 9.038187580296776e-05, "data_time": 0.014313530921936036, "loss": 0.006422832049429417, "time": 0.7175225257873535, "epoch": 8, "memory": 25074, "step": 61000} +{"lr": 9.038187580296776e-05, "data_time": 0.029600811004638673, "loss": 0.0033322285860776903, "time": 0.7317031860351563, "epoch": 8, "memory": 25074, "step": 61100} +{"lr": 9.038187580296776e-05, "data_time": 0.0013681888580322266, "loss": 0.00500296363607049, "time": 0.7187703371047973, "epoch": 8, "memory": 25074, "step": 61200} +{"lr": 9.038187580296776e-05, "data_time": 0.0014237403869628907, "loss": 0.011014320328831673, "time": 0.7050602436065674, "epoch": 8, "memory": 25074, "step": 61300} +{"lr": 9.038187580296776e-05, "data_time": 0.018259286880493164, "loss": 0.0056420756503939625, "time": 0.7301955938339233, "epoch": 8, "memory": 25074, "step": 61400} +{"lr": 9.038187580296776e-05, "data_time": 0.09516351222991944, "loss": 0.008991885185241699, "time": 0.7975855588912963, "epoch": 8, "memory": 25074, "step": 61500} +{"lr": 9.038187580296776e-05, "data_time": 0.003044319152832031, "loss": 0.003039657976478338, "time": 0.7058028221130371, "epoch": 8, "memory": 25074, "step": 61600} +{"lr": 9.038187580296776e-05, "data_time": 0.06466200351715087, "loss": 0.006659262720495463, "time": 0.7665573358535767, "epoch": 8, "memory": 25074, "step": 61700} +{"lr": 9.038187580296776e-05, "data_time": 0.03183677196502686, "loss": 0.0022524781059473755, "time": 0.7343443155288696, "epoch": 8, "memory": 25074, "step": 61800} +{"lr": 9.038187580296776e-05, "data_time": 0.04197757244110108, "loss": 0.006802659714594483, "time": 0.7442480087280273, "epoch": 8, "memory": 25074, "step": 61900} +{"lr": 9.038187580296776e-05, "data_time": 0.05157251358032226, "loss": 0.00735700074583292, "time": 0.7542294502258301, "epoch": 8, "memory": 25074, "step": 62000} +{"lr": 9.038187580296776e-05, "data_time": 0.03191332817077637, "loss": 0.003927370300516486, "time": 0.7353523015975952, "epoch": 8, "memory": 25074, "step": 62100} +{"lr": 9.038187580296776e-05, "data_time": 0.031256580352783205, "loss": 0.0026940954849123955, "time": 0.7339596509933471, "epoch": 8, "memory": 25074, "step": 62200} +{"lr": 9.038187580296776e-05, "data_time": 0.049047088623046874, "loss": 0.0020412841346114873, "time": 0.7513560771942138, "epoch": 8, "memory": 25074, "step": 62300} +{"lr": 9.038187580296776e-05, "data_time": 0.003758692741394043, "loss": 0.0038932988536544144, "time": 0.6614237546920776, "epoch": 8, "memory": 25074, "step": 62400} +{"accuracy/top1": 99.96842956542969, "data_time": 0.20667296297409954, "time": 0.5925739793216481, "step": 8} +{"lr": 4.724168609688527e-05, "data_time": 0.0012686729431152343, "loss": 0.0035854040179401636, "time": 0.7392109632492065, "epoch": 9, "memory": 25074, "step": 62500} +{"lr": 4.724168609688527e-05, "data_time": 0.0013946294784545898, "loss": 0.006433201255276799, "time": 0.7699014663696289, "epoch": 9, "memory": 25074, "step": 62600} +{"lr": 4.724168609688527e-05, "data_time": 0.0014280557632446289, "loss": 0.0015518635977059604, "time": 0.8053184509277344, "epoch": 9, "memory": 25074, "step": 62700} +{"lr": 4.724168609688527e-05, "data_time": 0.0014133930206298828, "loss": 0.004280343279242515, "time": 0.7830590009689331, "epoch": 9, "memory": 25074, "step": 62800} +{"lr": 4.724168609688527e-05, "data_time": 0.0012753725051879883, "loss": 0.006203908612951637, "time": 0.7324862480163574, "epoch": 9, "memory": 25074, "step": 62900} +{"lr": 4.724168609688527e-05, "data_time": 0.0012497186660766601, "loss": 0.0019470603205263616, "time": 0.7303324699401855, "epoch": 9, "memory": 25074, "step": 63000} +{"lr": 4.724168609688527e-05, "data_time": 0.0015161514282226562, "loss": 0.0026734769344329834, "time": 0.7312972784042359, "epoch": 9, "memory": 25074, "step": 63100} +{"lr": 4.724168609688527e-05, "data_time": 0.0015453100204467773, "loss": 0.004109944170340896, "time": 0.736474084854126, "epoch": 9, "memory": 25074, "step": 63200} +{"lr": 4.724168609688527e-05, "data_time": 0.001528310775756836, "loss": 0.0030402639880776406, "time": 0.707416033744812, "epoch": 9, "memory": 25074, "step": 63300} +{"lr": 4.724168609688527e-05, "data_time": 0.0012206077575683595, "loss": 0.0050008496269583706, "time": 0.7041665315628052, "epoch": 9, "memory": 25074, "step": 63400} +{"lr": 4.724168609688527e-05, "data_time": 0.0014792680740356445, "loss": 0.003973063547164202, "time": 0.7202898740768433, "epoch": 9, "memory": 25074, "step": 63500} +{"lr": 4.724168609688527e-05, "data_time": 0.0013941287994384765, "loss": 0.0026609358843415975, "time": 0.7256098747253418, "epoch": 9, "memory": 25074, "step": 63600} +{"lr": 4.724168609688527e-05, "data_time": 0.001529383659362793, "loss": 0.0023342340253293516, "time": 0.7169832468032837, "epoch": 9, "memory": 25074, "step": 63700} +{"lr": 4.724168609688527e-05, "data_time": 0.0016998529434204101, "loss": 0.0019028247334063053, "time": 0.7807982683181762, "epoch": 9, "memory": 25074, "step": 63800} +{"lr": 4.724168609688527e-05, "data_time": 0.001434159278869629, "loss": 0.00470067341811955, "time": 0.74498610496521, "epoch": 9, "memory": 25074, "step": 63900} +{"lr": 4.724168609688527e-05, "data_time": 0.0016046285629272461, "loss": 0.004311484005302191, "time": 0.7034945249557495, "epoch": 9, "memory": 25074, "step": 64000} +{"lr": 4.724168609688527e-05, "data_time": 0.0013657093048095703, "loss": 0.005308238137513399, "time": 0.7847606897354126, "epoch": 9, "memory": 25074, "step": 64100} +{"lr": 4.724168609688527e-05, "data_time": 0.001500105857849121, "loss": 0.00266368156298995, "time": 0.7079743385314942, "epoch": 9, "memory": 25074, "step": 64200} +{"lr": 4.724168609688527e-05, "data_time": 0.0012812852859497071, "loss": 0.007512034056708217, "time": 0.7790673971176147, "epoch": 9, "memory": 25074, "step": 64300} +{"lr": 4.724168609688527e-05, "data_time": 0.0015714168548583984, "loss": 0.003466403903439641, "time": 0.7049143552780152, "epoch": 9, "memory": 25074, "step": 64400} +{"lr": 4.724168609688527e-05, "data_time": 0.001565098762512207, "loss": 0.003334543481469154, "time": 0.7285926342010498, "epoch": 9, "memory": 25074, "step": 64500} +{"lr": 4.724168609688527e-05, "data_time": 0.0013058185577392578, "loss": 0.0033261864446103573, "time": 0.73345046043396, "epoch": 9, "memory": 25074, "step": 64600} +{"lr": 4.724168609688527e-05, "data_time": 0.0012994527816772461, "loss": 0.005671628518030047, "time": 0.7037026882171631, "epoch": 9, "memory": 25074, "step": 64700} +{"lr": 4.724168609688527e-05, "data_time": 0.0014406442642211914, "loss": 0.00519427671097219, "time": 0.7044807434082031, "epoch": 9, "memory": 25074, "step": 64800} +{"lr": 4.724168609688527e-05, "data_time": 0.0014109373092651366, "loss": 0.0022163213696330787, "time": 0.723729419708252, "epoch": 9, "memory": 25074, "step": 64900} +{"lr": 4.724168609688527e-05, "data_time": 0.0013831853866577148, "loss": 0.002377808466553688, "time": 0.720691180229187, "epoch": 9, "memory": 25074, "step": 65000} +{"lr": 4.724168609688527e-05, "data_time": 0.0014168977737426757, "loss": 0.0038187152706086634, "time": 0.7150551080703735, "epoch": 9, "memory": 25074, "step": 65100} +{"lr": 4.724168609688527e-05, "data_time": 0.0016774892807006835, "loss": 0.005426485743373633, "time": 0.7044049978256226, "epoch": 9, "memory": 25074, "step": 65200} +{"lr": 4.724168609688527e-05, "data_time": 0.0014931440353393554, "loss": 0.0025714052841067313, "time": 0.7300945520401001, "epoch": 9, "memory": 25074, "step": 65300} +{"lr": 4.724168609688527e-05, "data_time": 0.0013509035110473634, "loss": 0.00381183628924191, "time": 0.7075841426849365, "epoch": 9, "memory": 25074, "step": 65400} +{"lr": 4.724168609688527e-05, "data_time": 0.0014547586441040039, "loss": 0.002330287452787161, "time": 0.8134341478347779, "epoch": 9, "memory": 25074, "step": 65500} +{"lr": 4.724168609688527e-05, "data_time": 0.0013962745666503905, "loss": 0.0011964922305196524, "time": 0.7074713230133056, "epoch": 9, "memory": 25074, "step": 65600} +{"lr": 4.724168609688527e-05, "data_time": 0.0016634702682495118, "loss": 0.003165269223973155, "time": 0.7071994781494141, "epoch": 9, "memory": 25074, "step": 65700} +{"lr": 4.724168609688527e-05, "data_time": 0.00136110782623291, "loss": 0.006068702321499586, "time": 0.7887085676193237, "epoch": 9, "memory": 25074, "step": 65800} +{"lr": 4.724168609688527e-05, "data_time": 0.0014057397842407227, "loss": 0.002544358605518937, "time": 0.7290219068527222, "epoch": 9, "memory": 25074, "step": 65900} +{"lr": 4.724168609688527e-05, "data_time": 0.001516866683959961, "loss": 0.005004653427749872, "time": 0.7046590566635131, "epoch": 9, "memory": 25074, "step": 66000} +{"lr": 4.724168609688527e-05, "data_time": 0.0012632369995117187, "loss": 0.0029656046070158483, "time": 0.7326936483383178, "epoch": 9, "memory": 25074, "step": 66100} +{"lr": 4.724168609688527e-05, "data_time": 0.0013437986373901368, "loss": 0.004099781811237335, "time": 0.7578452825546265, "epoch": 9, "memory": 25074, "step": 66200} +{"lr": 4.724168609688527e-05, "data_time": 0.0014946937561035156, "loss": 0.002212268952280283, "time": 0.7432986497879028, "epoch": 9, "memory": 25074, "step": 66300} +{"lr": 4.724168609688527e-05, "data_time": 0.0017410993576049804, "loss": 0.0047306973487138745, "time": 0.7208121061325073, "epoch": 9, "memory": 25074, "step": 66400} +{"lr": 4.724168609688527e-05, "data_time": 0.0014394998550415039, "loss": 0.005374382436275482, "time": 0.7342538356781005, "epoch": 9, "memory": 25074, "step": 66500} +{"lr": 4.724168609688527e-05, "data_time": 0.0014314889907836915, "loss": 0.004135999083518982, "time": 0.7817184209823609, "epoch": 9, "memory": 25074, "step": 66600} +{"lr": 4.724168609688527e-05, "data_time": 0.0013975381851196289, "loss": 0.002245303150266409, "time": 0.802988600730896, "epoch": 9, "memory": 25074, "step": 66700} +{"lr": 4.724168609688527e-05, "data_time": 0.001355743408203125, "loss": 0.00570099544711411, "time": 0.7101654529571533, "epoch": 9, "memory": 25074, "step": 66800} +{"lr": 4.724168609688527e-05, "data_time": 0.001800203323364258, "loss": 0.002721928060054779, "time": 0.778480052947998, "epoch": 9, "memory": 25074, "step": 66900} +{"lr": 4.724168609688527e-05, "data_time": 0.0014678955078125, "loss": 0.0013445299584418536, "time": 0.7823601961135864, "epoch": 9, "memory": 25074, "step": 67000} +{"lr": 4.724168609688527e-05, "data_time": 0.0014457464218139648, "loss": 0.002576215984299779, "time": 0.7595079421997071, "epoch": 9, "memory": 25074, "step": 67100} +{"lr": 4.724168609688527e-05, "data_time": 0.0014696836471557618, "loss": 0.0024327327962964775, "time": 0.7793433189392089, "epoch": 9, "memory": 25074, "step": 67200} +{"lr": 4.724168609688527e-05, "data_time": 0.0014782428741455078, "loss": 0.0012204103637486695, "time": 0.7778377771377564, "epoch": 9, "memory": 25074, "step": 67300} +{"lr": 4.724168609688527e-05, "data_time": 0.0015259504318237305, "loss": 0.0014354664366692305, "time": 0.8095024824142456, "epoch": 9, "memory": 25074, "step": 67400} +{"lr": 4.724168609688527e-05, "data_time": 0.0014205455780029296, "loss": 0.0023573593702167273, "time": 0.7228098630905151, "epoch": 9, "memory": 25074, "step": 67500} +{"lr": 4.724168609688527e-05, "data_time": 0.001766538619995117, "loss": 0.0034583337604999543, "time": 0.730274248123169, "epoch": 9, "memory": 25074, "step": 67600} +{"lr": 4.724168609688527e-05, "data_time": 0.001429271697998047, "loss": 0.001797714876011014, "time": 0.7654394149780274, "epoch": 9, "memory": 25074, "step": 67700} +{"lr": 4.724168609688527e-05, "data_time": 0.001492762565612793, "loss": 0.003790541971102357, "time": 0.7295792579650879, "epoch": 9, "memory": 25074, "step": 67800} +{"lr": 4.724168609688527e-05, "data_time": 0.001260519027709961, "loss": 0.002734592417255044, "time": 0.750445818901062, "epoch": 9, "memory": 25074, "step": 67900} +{"lr": 4.724168609688527e-05, "data_time": 0.0013107776641845704, "loss": 0.004440054949373007, "time": 0.7778757095336915, "epoch": 9, "memory": 25074, "step": 68000} +{"lr": 4.724168609688527e-05, "data_time": 0.0014533281326293945, "loss": 0.0013329595327377318, "time": 0.7396915912628174, "epoch": 9, "memory": 25074, "step": 68100} +{"lr": 4.724168609688527e-05, "data_time": 0.0017024993896484375, "loss": 0.007108057430014014, "time": 0.7474076986312866, "epoch": 9, "memory": 25074, "step": 68200} +{"lr": 4.724168609688527e-05, "data_time": 0.001279926300048828, "loss": 0.0025010320357978346, "time": 0.7402063846588135, "epoch": 9, "memory": 25074, "step": 68300} +{"lr": 4.724168609688527e-05, "data_time": 0.0012903690338134765, "loss": 0.003660020511597395, "time": 0.7040968656539917, "epoch": 9, "memory": 25074, "step": 68400} +{"lr": 4.724168609688527e-05, "data_time": 0.0013795137405395509, "loss": 0.0022671422455459835, "time": 0.7303619384765625, "epoch": 9, "memory": 25074, "step": 68500} +{"lr": 4.724168609688527e-05, "data_time": 0.04188542366027832, "loss": 0.003003792883828282, "time": 0.7496346950531005, "epoch": 9, "memory": 25074, "step": 68600} +{"lr": 4.724168609688527e-05, "data_time": 0.05385987758636475, "loss": 0.006541739497333765, "time": 0.7570548057556152, "epoch": 9, "memory": 25074, "step": 68700} +{"lr": 4.724168609688527e-05, "data_time": 0.08272285461425781, "loss": 0.001273703435435891, "time": 0.7949224948883057, "epoch": 9, "memory": 25074, "step": 68800} +{"lr": 4.724168609688527e-05, "data_time": 0.026029586791992188, "loss": 0.004725271742790938, "time": 0.7288677930831909, "epoch": 9, "memory": 25074, "step": 68900} +{"lr": 4.724168609688527e-05, "data_time": 0.00137176513671875, "loss": 0.0013643792364746333, "time": 0.7207146883010864, "epoch": 9, "memory": 25074, "step": 69000} +{"lr": 4.724168609688527e-05, "data_time": 0.06450567245483399, "loss": 0.0017586033325642347, "time": 0.766940426826477, "epoch": 9, "memory": 25074, "step": 69100} +{"lr": 4.724168609688527e-05, "data_time": 0.05853328704833984, "loss": 0.004154441645368934, "time": 0.7625686645507812, "epoch": 9, "memory": 25074, "step": 69200} +{"lr": 4.724168609688527e-05, "data_time": 0.08435139656066895, "loss": 0.0008559535257518291, "time": 0.7867570638656616, "epoch": 9, "memory": 25074, "step": 69300} +{"lr": 4.724168609688527e-05, "data_time": 0.06345672607421875, "loss": 0.003086626483127475, "time": 0.7710541486740112, "epoch": 9, "memory": 25074, "step": 69400} +{"lr": 4.724168609688527e-05, "data_time": 0.0013145208358764648, "loss": 0.0027724782936275004, "time": 0.7045375823974609, "epoch": 9, "memory": 25074, "step": 69500} +{"lr": 4.724168609688527e-05, "data_time": 0.0261336088180542, "loss": 0.003240892989560962, "time": 0.7518714189529419, "epoch": 9, "memory": 25074, "step": 69600} +{"lr": 4.724168609688527e-05, "data_time": 0.0404421329498291, "loss": 0.005900827376171947, "time": 0.7500909090042114, "epoch": 9, "memory": 25074, "step": 69700} +{"lr": 4.724168609688527e-05, "data_time": 0.04092690944671631, "loss": 0.0035092404112219812, "time": 0.7556733131408692, "epoch": 9, "memory": 25074, "step": 69800} +{"lr": 4.724168609688527e-05, "data_time": 0.03795361518859863, "loss": 0.001649703411385417, "time": 0.7448919296264649, "epoch": 9, "memory": 25074, "step": 69900} +{"lr": 4.724168609688527e-05, "data_time": 0.05460622310638428, "loss": 0.002138024941086769, "time": 0.7676746368408203, "epoch": 9, "memory": 25074, "step": 70000} +{"lr": 4.724168609688527e-05, "data_time": 0.05661880970001221, "loss": 0.004421891504898667, "time": 0.7733850955963135, "epoch": 9, "memory": 25074, "step": 70100} +{"lr": 4.724168609688527e-05, "data_time": 0.04865431785583496, "loss": 0.0037510984839173034, "time": 0.7081318616867065, "epoch": 9, "memory": 25074, "step": 70200} +{"accuracy/top1": 99.97474670410156, "data_time": 0.21337213235742905, "time": 0.5989095042733585, "step": 9} +{"lr": 1.9543979322445064e-05, "data_time": 0.001468658447265625, "loss": 0.003716249298304319, "time": 0.7956549644470214, "epoch": 10, "memory": 25074, "step": 70300} +{"lr": 1.9543979322445064e-05, "data_time": 0.001401972770690918, "loss": 0.0018167675007134676, "time": 0.7895112991333008, "epoch": 10, "memory": 25074, "step": 70400} +{"lr": 1.9543979322445064e-05, "data_time": 0.001549363136291504, "loss": 0.002099432609975338, "time": 0.7388676881790162, "epoch": 10, "memory": 25074, "step": 70500} +{"lr": 1.9543979322445064e-05, "data_time": 0.035631775856018066, "loss": 0.002846449287608266, "time": 0.7772918462753295, "epoch": 10, "memory": 25074, "step": 70600} +{"lr": 1.9543979322445064e-05, "data_time": 0.05164930820465088, "loss": 0.0027175822760909796, "time": 0.7677915811538696, "epoch": 10, "memory": 25074, "step": 70700} +{"lr": 1.9543979322445064e-05, "data_time": 0.056874418258666994, "loss": 0.0006234078668057918, "time": 0.7584271430969238, "epoch": 10, "memory": 25074, "step": 70800} +{"lr": 1.9543979322445064e-05, "data_time": 0.00845353603363037, "loss": 0.00038216873072087764, "time": 0.710853123664856, "epoch": 10, "memory": 25074, "step": 70900} +{"lr": 1.9543979322445064e-05, "data_time": 0.03818886280059815, "loss": 0.001954852370545268, "time": 0.7504744052886962, "epoch": 10, "memory": 25074, "step": 71000} +{"lr": 1.9543979322445064e-05, "data_time": 0.0012757539749145507, "loss": 0.004034789325669408, "time": 0.7036790370941162, "epoch": 10, "memory": 25074, "step": 71100} +{"lr": 1.9543979322445064e-05, "data_time": 0.0016345977783203125, "loss": 0.0018768099136650563, "time": 0.7170355558395386, "epoch": 10, "memory": 25074, "step": 71200} +{"lr": 1.9543979322445064e-05, "data_time": 0.03760805130004883, "loss": 0.0011197004001587628, "time": 0.7402360439300537, "epoch": 10, "memory": 25074, "step": 71300} +{"lr": 1.9543979322445064e-05, "data_time": 0.036709022521972653, "loss": 0.001394212432205677, "time": 0.7538413286209107, "epoch": 10, "memory": 25074, "step": 71400} +{"lr": 1.9543979322445064e-05, "data_time": 0.008089041709899903, "loss": 0.00416420167312026, "time": 0.7103933572769165, "epoch": 10, "memory": 25074, "step": 71500} +{"lr": 1.9543979322445064e-05, "data_time": 0.027568650245666505, "loss": 0.0006730372086167335, "time": 0.7429495096206665, "epoch": 10, "memory": 25074, "step": 71600} +{"lr": 1.9543979322445064e-05, "data_time": 0.020343494415283204, "loss": 0.00261286492459476, "time": 0.7376342296600342, "epoch": 10, "memory": 25074, "step": 71700} +{"lr": 1.9543979322445064e-05, "data_time": 0.07191858291625977, "loss": 0.004302518954500556, "time": 0.776107668876648, "epoch": 10, "memory": 25074, "step": 71800} +{"lr": 1.9543979322445064e-05, "data_time": 0.030131149291992187, "loss": 0.002136178920045495, "time": 0.7319931983947754, "epoch": 10, "memory": 25074, "step": 71900} +{"lr": 1.9543979322445064e-05, "data_time": 0.05597836971282959, "loss": 0.001320331171154976, "time": 0.758454155921936, "epoch": 10, "memory": 25074, "step": 72000} +{"lr": 1.9543979322445064e-05, "data_time": 0.07893037796020508, "loss": 0.0033421942498534918, "time": 0.7960315227508545, "epoch": 10, "memory": 25074, "step": 72100} +{"lr": 1.9543979322445064e-05, "data_time": 0.06062870025634766, "loss": 0.003074402594938874, "time": 0.7629699468612671, "epoch": 10, "memory": 25074, "step": 72200} +{"lr": 1.9543979322445064e-05, "data_time": 0.0372807502746582, "loss": 0.001983436569571495, "time": 0.7575109004974365, "epoch": 10, "memory": 25074, "step": 72300} +{"lr": 1.9543979322445064e-05, "data_time": 0.054410362243652345, "loss": 0.0016810684464871883, "time": 0.7565388202667236, "epoch": 10, "memory": 25074, "step": 72400} +{"lr": 1.9543979322445064e-05, "data_time": 0.057735109329223634, "loss": 0.0025253639556467532, "time": 0.7713340520858765, "epoch": 10, "memory": 25074, "step": 72500} +{"lr": 1.9543979322445064e-05, "data_time": 0.03789963722229004, "loss": 0.0026211796794086695, "time": 0.7408131837844849, "epoch": 10, "memory": 25074, "step": 72600} +{"lr": 1.9543979322445064e-05, "data_time": 0.12151339054107665, "loss": 0.00252187754958868, "time": 0.8321881532669068, "epoch": 10, "memory": 25074, "step": 72700} +{"lr": 1.9543979322445064e-05, "data_time": 0.07931854724884033, "loss": 0.004591597151011229, "time": 0.7812834739685058, "epoch": 10, "memory": 25074, "step": 72800} +{"lr": 1.9543979322445064e-05, "data_time": 0.015970945358276367, "loss": 0.0006349658593535423, "time": 0.7358031034469604, "epoch": 10, "memory": 25074, "step": 72900} +{"lr": 1.9543979322445064e-05, "data_time": 0.06624820232391357, "loss": 0.0025834666099399327, "time": 0.7701605796813965, "epoch": 10, "memory": 25074, "step": 73000} +{"lr": 1.9543979322445064e-05, "data_time": 0.057169413566589354, "loss": 0.002863296028226614, "time": 0.7678876876831054, "epoch": 10, "memory": 25074, "step": 73100} +{"lr": 1.9543979322445064e-05, "data_time": 0.058753585815429686, "loss": 0.0033069149125367404, "time": 0.761167311668396, "epoch": 10, "memory": 25074, "step": 73200} +{"lr": 1.9543979322445064e-05, "data_time": 0.10259363651275635, "loss": 0.0012726620770990849, "time": 0.8150048732757569, "epoch": 10, "memory": 25074, "step": 73300} +{"lr": 1.9543979322445064e-05, "data_time": 0.08943588733673095, "loss": 0.0025894547812640666, "time": 0.7915759325027466, "epoch": 10, "memory": 25074, "step": 73400} +{"lr": 1.9543979322445064e-05, "data_time": 0.11913619041442872, "loss": 0.001922428421676159, "time": 0.8215875148773193, "epoch": 10, "memory": 25074, "step": 73500} +{"lr": 1.9543979322445064e-05, "data_time": 0.03992674350738525, "loss": 0.0023916376288980246, "time": 0.7417946338653565, "epoch": 10, "memory": 25074, "step": 73600} +{"lr": 1.9543979322445064e-05, "data_time": 0.08811109066009522, "loss": 0.0004283975809812546, "time": 0.7922694683074951, "epoch": 10, "memory": 25074, "step": 73700} +{"lr": 1.9543979322445064e-05, "data_time": 0.07816421985626221, "loss": 0.002233076561242342, "time": 0.7810037851333618, "epoch": 10, "memory": 25074, "step": 73800} +{"lr": 1.9543979322445064e-05, "data_time": 0.0027591943740844726, "loss": 0.001962736435234547, "time": 0.7198577642440795, "epoch": 10, "memory": 25074, "step": 73900} +{"lr": 1.9543979322445064e-05, "data_time": 0.026323866844177247, "loss": 0.004134145192801952, "time": 0.7285466909408569, "epoch": 10, "memory": 25074, "step": 74000} +{"lr": 1.9543979322445064e-05, "data_time": 0.09715259075164795, "loss": 0.002139729354530573, "time": 0.8003833055496216, "epoch": 10, "memory": 25074, "step": 74100} +{"lr": 1.9543979322445064e-05, "data_time": 0.010705947875976562, "loss": 0.0014401759952306747, "time": 0.7128584146499634, "epoch": 10, "memory": 25074, "step": 74200} +{"lr": 1.9543979322445064e-05, "data_time": 0.04110572338104248, "loss": 0.0005889842286705971, "time": 0.7432945489883422, "epoch": 10, "memory": 25074, "step": 74300} +{"lr": 1.9543979322445064e-05, "data_time": 0.039203739166259764, "loss": 0.001075806003063917, "time": 0.7423350572586059, "epoch": 10, "memory": 25074, "step": 74400} +{"lr": 1.9543979322445064e-05, "data_time": 0.08183917999267579, "loss": 0.0017675799317657948, "time": 0.7847230434417725, "epoch": 10, "memory": 25074, "step": 74500} +{"lr": 1.9543979322445064e-05, "data_time": 0.02958788871765137, "loss": 0.0019665206782519817, "time": 0.7333899736404419, "epoch": 10, "memory": 25074, "step": 74600} +{"lr": 1.9543979322445064e-05, "data_time": 0.033285999298095705, "loss": 0.004259195365011692, "time": 0.7361727237701416, "epoch": 10, "memory": 25074, "step": 74700} +{"lr": 1.9543979322445064e-05, "data_time": 0.08124017715454102, "loss": 0.001425787154585123, "time": 0.7836496114730835, "epoch": 10, "memory": 25074, "step": 74800} +{"lr": 1.9543979322445064e-05, "data_time": 0.03210391998291016, "loss": 0.0009439696557819843, "time": 0.7359946966171265, "epoch": 10, "memory": 25074, "step": 74900} +{"lr": 1.9543979322445064e-05, "data_time": 0.07245752811431885, "loss": 0.0018797555938363075, "time": 0.7759171962738037, "epoch": 10, "memory": 25074, "step": 75000} +{"lr": 1.9543979322445064e-05, "data_time": 0.008467388153076173, "loss": 0.005092905554920435, "time": 0.7110937356948852, "epoch": 10, "memory": 25074, "step": 75100} +{"lr": 1.9543979322445064e-05, "data_time": 0.08542578220367432, "loss": 0.0020634511951357127, "time": 0.7878558397293091, "epoch": 10, "memory": 25074, "step": 75200} +{"lr": 1.9543979322445064e-05, "data_time": 0.05697500705718994, "loss": 0.0018645947333425284, "time": 0.7608028173446655, "epoch": 10, "memory": 25074, "step": 75300} +{"lr": 1.9543979322445064e-05, "data_time": 0.014886569976806641, "loss": 0.0032790504395961762, "time": 0.7179263353347778, "epoch": 10, "memory": 25074, "step": 75400} +{"lr": 1.9543979322445064e-05, "data_time": 0.0480043888092041, "loss": 0.00141446515917778, "time": 0.7499783039093018, "epoch": 10, "memory": 25074, "step": 75500} +{"lr": 1.9543979322445064e-05, "data_time": 0.023751306533813476, "loss": 0.0014892212580889463, "time": 0.7263204336166382, "epoch": 10, "memory": 25074, "step": 75600} +{"lr": 1.9543979322445064e-05, "data_time": 0.03580174446105957, "loss": 0.004743988672271371, "time": 0.7381531715393066, "epoch": 10, "memory": 25074, "step": 75700} +{"lr": 1.9543979322445064e-05, "data_time": 0.05883114337921143, "loss": 0.001213197922334075, "time": 0.7676050424575805, "epoch": 10, "memory": 25074, "step": 75800} +{"lr": 1.9543979322445064e-05, "data_time": 0.021689653396606445, "loss": 0.00039984621107578275, "time": 0.7252687931060791, "epoch": 10, "memory": 25074, "step": 75900} +{"lr": 1.9543979322445064e-05, "data_time": 0.06327095031738281, "loss": 0.0034788088873028756, "time": 0.7730864524841309, "epoch": 10, "memory": 25074, "step": 76000} +{"lr": 1.9543979322445064e-05, "data_time": 0.21071991920471192, "loss": 0.00129951941780746, "time": 0.9303645133972168, "epoch": 10, "memory": 25074, "step": 76100} +{"lr": 1.9543979322445064e-05, "data_time": 0.13052523136138916, "loss": 0.0020625774282962083, "time": 0.8474969387054443, "epoch": 10, "memory": 25074, "step": 76200} +{"lr": 1.9543979322445064e-05, "data_time": 0.20942361354827882, "loss": 0.002098038839176297, "time": 0.9318359613418579, "epoch": 10, "memory": 25074, "step": 76300} +{"lr": 1.9543979322445064e-05, "data_time": 0.09960784912109374, "loss": 0.0014654813334345818, "time": 0.8268126964569091, "epoch": 10, "memory": 25074, "step": 76400} +{"lr": 1.9543979322445064e-05, "data_time": 0.02778441905975342, "loss": 0.0019784611184149982, "time": 0.7307383298873902, "epoch": 10, "memory": 25074, "step": 76500} +{"lr": 1.9543979322445064e-05, "data_time": 0.10081713199615479, "loss": 0.0020369043573737144, "time": 0.8450204133987427, "epoch": 10, "memory": 25074, "step": 76600} +{"lr": 1.9543979322445064e-05, "data_time": 0.1534421443939209, "loss": 0.000898544630035758, "time": 0.871632170677185, "epoch": 10, "memory": 25074, "step": 76700} +{"lr": 1.9543979322445064e-05, "data_time": 0.09541425704956055, "loss": 0.0007972294464707374, "time": 0.8196889877319335, "epoch": 10, "memory": 25074, "step": 76800} +{"lr": 1.9543979322445064e-05, "data_time": 0.052926373481750486, "loss": 0.0033644529525190593, "time": 0.7564990282058716, "epoch": 10, "memory": 25074, "step": 76900} +{"lr": 1.9543979322445064e-05, "data_time": 0.041558361053466795, "loss": 0.002528955042362213, "time": 0.7693127393722534, "epoch": 10, "memory": 25074, "step": 77000} +{"lr": 1.9543979322445064e-05, "data_time": 0.01842055320739746, "loss": 0.0011274866294115782, "time": 0.7204957723617553, "epoch": 10, "memory": 25074, "step": 77100} +{"lr": 1.9543979322445064e-05, "data_time": 0.10217568874359131, "loss": 0.00392243810929358, "time": 0.805843448638916, "epoch": 10, "memory": 25074, "step": 77200} +{"lr": 1.9543979322445064e-05, "data_time": 0.159128999710083, "loss": 0.001156916655600071, "time": 0.8737040758132935, "epoch": 10, "memory": 25074, "step": 77300} +{"lr": 1.9543979322445064e-05, "data_time": 0.16430633068084716, "loss": 0.0018640908412635327, "time": 0.9029998779296875, "epoch": 10, "memory": 25074, "step": 77400} +{"lr": 1.9543979322445064e-05, "data_time": 0.07460556030273438, "loss": 0.0014721516519784928, "time": 0.7803693056106568, "epoch": 10, "memory": 25074, "step": 77500} +{"lr": 1.9543979322445064e-05, "data_time": 0.03137850761413574, "loss": 0.0008515995927155018, "time": 0.753780722618103, "epoch": 10, "memory": 25074, "step": 77600} +{"lr": 1.9543979322445064e-05, "data_time": 0.05947234630584717, "loss": 0.00241722147911787, "time": 0.7621008634567261, "epoch": 10, "memory": 25074, "step": 77700} +{"lr": 1.9543979322445064e-05, "data_time": 0.14171135425567627, "loss": 0.004039214039221406, "time": 0.9070079088211059, "epoch": 10, "memory": 25074, "step": 77800} +{"lr": 1.9543979322445064e-05, "data_time": 0.17052626609802246, "loss": 0.002315824432298541, "time": 0.8927216768264771, "epoch": 10, "memory": 25074, "step": 77900} +{"lr": 1.9543979322445064e-05, "data_time": 0.12948236465454102, "loss": 0.0029011948667175603, "time": 0.8018013715744019, "epoch": 10, "memory": 25074, "step": 78000} +{"accuracy/top1": 99.98737335205078, "data_time": 0.31446392395917105, "time": 0.7018048202290255, "step": 10} diff --git a/convnext_small_4xb256_fake7m-lr4e-4/20230601_193207/vis_data/vis_image/0000000.png_1.png b/convnext_small_4xb256_fake7m-lr4e-4/20230601_193207/vis_data/vis_image/0000000.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..78086a4ac43d9c8da460aa6d9b24e940253194a5 Binary files /dev/null and b/convnext_small_4xb256_fake7m-lr4e-4/20230601_193207/vis_data/vis_image/0000000.png_1.png differ diff --git a/convnext_small_4xb256_fake7m-lr4e-4/20230601_193207/vis_data/vis_image/0000000.png_10.png b/convnext_small_4xb256_fake7m-lr4e-4/20230601_193207/vis_data/vis_image/0000000.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..6f264824565adf0630fdfc2e331f1ad6f2495172 Binary files /dev/null and b/convnext_small_4xb256_fake7m-lr4e-4/20230601_193207/vis_data/vis_image/0000000.png_10.png differ diff --git a/convnext_small_4xb256_fake7m-lr4e-4/20230601_193207/vis_data/vis_image/0000000.png_2.png b/convnext_small_4xb256_fake7m-lr4e-4/20230601_193207/vis_data/vis_image/0000000.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..6f264824565adf0630fdfc2e331f1ad6f2495172 Binary files /dev/null and b/convnext_small_4xb256_fake7m-lr4e-4/20230601_193207/vis_data/vis_image/0000000.png_2.png differ diff --git a/convnext_small_4xb256_fake7m-lr4e-4/20230601_193207/vis_data/vis_image/0000000.png_3.png b/convnext_small_4xb256_fake7m-lr4e-4/20230601_193207/vis_data/vis_image/0000000.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..6f264824565adf0630fdfc2e331f1ad6f2495172 Binary files /dev/null and b/convnext_small_4xb256_fake7m-lr4e-4/20230601_193207/vis_data/vis_image/0000000.png_3.png differ diff --git a/convnext_small_4xb256_fake7m-lr4e-4/20230601_193207/vis_data/vis_image/0000000.png_4.png b/convnext_small_4xb256_fake7m-lr4e-4/20230601_193207/vis_data/vis_image/0000000.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..6f264824565adf0630fdfc2e331f1ad6f2495172 Binary files /dev/null and b/convnext_small_4xb256_fake7m-lr4e-4/20230601_193207/vis_data/vis_image/0000000.png_4.png differ diff --git a/convnext_small_4xb256_fake7m-lr4e-4/20230601_193207/vis_data/vis_image/0000000.png_5.png b/convnext_small_4xb256_fake7m-lr4e-4/20230601_193207/vis_data/vis_image/0000000.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..6f264824565adf0630fdfc2e331f1ad6f2495172 Binary files /dev/null and b/convnext_small_4xb256_fake7m-lr4e-4/20230601_193207/vis_data/vis_image/0000000.png_5.png differ diff --git a/convnext_small_4xb256_fake7m-lr4e-4/20230601_193207/vis_data/vis_image/0000000.png_6.png b/convnext_small_4xb256_fake7m-lr4e-4/20230601_193207/vis_data/vis_image/0000000.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..6f264824565adf0630fdfc2e331f1ad6f2495172 Binary files /dev/null and b/convnext_small_4xb256_fake7m-lr4e-4/20230601_193207/vis_data/vis_image/0000000.png_6.png differ diff --git a/convnext_small_4xb256_fake7m-lr4e-4/20230601_193207/vis_data/vis_image/0000000.png_7.png b/convnext_small_4xb256_fake7m-lr4e-4/20230601_193207/vis_data/vis_image/0000000.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..6f264824565adf0630fdfc2e331f1ad6f2495172 Binary files /dev/null and b/convnext_small_4xb256_fake7m-lr4e-4/20230601_193207/vis_data/vis_image/0000000.png_7.png differ diff --git a/convnext_small_4xb256_fake7m-lr4e-4/20230601_193207/vis_data/vis_image/0000000.png_8.png b/convnext_small_4xb256_fake7m-lr4e-4/20230601_193207/vis_data/vis_image/0000000.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..6f264824565adf0630fdfc2e331f1ad6f2495172 Binary files /dev/null and b/convnext_small_4xb256_fake7m-lr4e-4/20230601_193207/vis_data/vis_image/0000000.png_8.png differ diff --git a/convnext_small_4xb256_fake7m-lr4e-4/20230601_193207/vis_data/vis_image/0000000.png_9.png b/convnext_small_4xb256_fake7m-lr4e-4/20230601_193207/vis_data/vis_image/0000000.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..6f264824565adf0630fdfc2e331f1ad6f2495172 Binary files /dev/null and b/convnext_small_4xb256_fake7m-lr4e-4/20230601_193207/vis_data/vis_image/0000000.png_9.png differ diff --git a/convnext_small_4xb256_fake7m-lr4e-4/convnext_small_4xb256_fake7m-lr4e-4.py b/convnext_small_4xb256_fake7m-lr4e-4/convnext_small_4xb256_fake7m-lr4e-4.py new file mode 100644 index 0000000000000000000000000000000000000000..0e6e101fa3314a1c359ce105454f400d187c5c3d --- /dev/null +++ b/convnext_small_4xb256_fake7m-lr4e-4/convnext_small_4xb256_fake7m-lr4e-4.py @@ -0,0 +1,203 @@ +optim_wrapper = dict( + optimizer=dict( + type='AdamW', + lr=0.0004, + weight_decay=0.05, + eps=1e-08, + betas=(0.9, 0.999), + _scope_='mmpretrain'), + paramwise_cfg=dict( + norm_decay_mult=0.0, + bias_decay_mult=0.0, + flat_decay_mult=0.0, + custom_keys=dict({ + '.absolute_pos_embed': dict(decay_mult=0.0), + '.relative_position_bias_table': dict(decay_mult=0.0) + })), + type='AmpOptimWrapper', + dtype='bfloat16', + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=True, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=4096) +model = dict( + type='ImageClassifier', + backbone=dict(type='ConvNeXt', arch='small', drop_path_rate=0.4), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=768, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + init_cfg=None), + init_cfg=dict( + type='TruncNormal', layer=['Conv2d', 'Linear'], std=0.02, bias=0.0), + train_cfg=None) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stablediffusionV1-5R2-dpmsolver-25-5m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/if-dpmsolver++-50-20w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stablediffusionV2-1-dpmsolver-25-190w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/convnext_small_4xb256_fake7m-lr4e-4' diff --git a/convnext_small_4xb256_fake7m-lr4e-4/epoch_1.pth b/convnext_small_4xb256_fake7m-lr4e-4/epoch_1.pth new file mode 100644 index 0000000000000000000000000000000000000000..5083af3d7e3c5800173dba72f93027bd9debda48 --- /dev/null +++ b/convnext_small_4xb256_fake7m-lr4e-4/epoch_1.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:79bebd1fd691ec35f0eb24cbdc80d7c1313cedf70157bf61719c8961e97a77ce +size 792454752 diff --git a/convnext_small_4xb256_fake7m-lr4e-4/epoch_10.pth b/convnext_small_4xb256_fake7m-lr4e-4/epoch_10.pth new file mode 100644 index 0000000000000000000000000000000000000000..aea3e4c6dbc1d18c2560fc899e28f07c503ab631 --- /dev/null +++ b/convnext_small_4xb256_fake7m-lr4e-4/epoch_10.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:8508b91f8eeca485930858ab10e0d5b4c940602ca05a29468d7393b72cbb6f82 +size 797567072 diff --git a/convnext_small_4xb256_fake7m-lr4e-4/epoch_2.pth b/convnext_small_4xb256_fake7m-lr4e-4/epoch_2.pth new file mode 100644 index 0000000000000000000000000000000000000000..656b4de8a47f183ff99d6268f48de5ced3dfe954 --- /dev/null +++ b/convnext_small_4xb256_fake7m-lr4e-4/epoch_2.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:25e33a79b1bc1f1fa5f3da581f301273edabcaff8528fe367247913cf0ee1483 +size 793040992 diff --git a/convnext_small_4xb256_fake7m-lr4e-4/epoch_3.pth b/convnext_small_4xb256_fake7m-lr4e-4/epoch_3.pth new file mode 100644 index 0000000000000000000000000000000000000000..067be0f1c3b7dadd690dbb8a649e5524f7131dbb --- /dev/null +++ b/convnext_small_4xb256_fake7m-lr4e-4/epoch_3.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:b660d1f5f91c8253481d1f437dd82daa8451a8095bcf9f56027e6a09e4dfa632 +size 793601568 diff --git a/convnext_small_4xb256_fake7m-lr4e-4/epoch_4.pth b/convnext_small_4xb256_fake7m-lr4e-4/epoch_4.pth new file mode 100644 index 0000000000000000000000000000000000000000..fba7c301b8a13d7cfd47a6e0749c749232ac88ef --- /dev/null +++ b/convnext_small_4xb256_fake7m-lr4e-4/epoch_4.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:a6209e494e9d3e7263561033d5fca0db1095e47e9505b541173ae1d9577973bf +size 794169312 diff --git a/convnext_small_4xb256_fake7m-lr4e-4/epoch_5.pth b/convnext_small_4xb256_fake7m-lr4e-4/epoch_5.pth new file mode 100644 index 0000000000000000000000000000000000000000..6c5aeb6f0419d2ca375664628537ceddd537e88b --- /dev/null +++ b/convnext_small_4xb256_fake7m-lr4e-4/epoch_5.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:4180b60bdc8cc1612e3f45fd56cb33620d3b39738fb7aff86d32caf2d2997bd8 +size 794736032 diff --git a/convnext_small_4xb256_fake7m-lr4e-4/epoch_6.pth b/convnext_small_4xb256_fake7m-lr4e-4/epoch_6.pth new file mode 100644 index 0000000000000000000000000000000000000000..4269524857e9f5af691b4ccdbde69a2b25db835b --- /dev/null +++ b/convnext_small_4xb256_fake7m-lr4e-4/epoch_6.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:7663847254c9ed3b8dc315d368706f90b18e3aa3d91cfe96a9944b954f8764c5 +size 795300832 diff --git a/convnext_small_4xb256_fake7m-lr4e-4/epoch_7.pth b/convnext_small_4xb256_fake7m-lr4e-4/epoch_7.pth new file mode 100644 index 0000000000000000000000000000000000000000..8b60af6bafc1c20a8a3f4970d552f3c49d6d5d1f --- /dev/null +++ b/convnext_small_4xb256_fake7m-lr4e-4/epoch_7.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:7a3cf831b9fcb7a8a048ec7a11d82dbfe80a3e2e0a8b774ecf2a43a3a0a9cda7 +size 795871136 diff --git a/convnext_small_4xb256_fake7m-lr4e-4/epoch_8.pth b/convnext_small_4xb256_fake7m-lr4e-4/epoch_8.pth new file mode 100644 index 0000000000000000000000000000000000000000..76ec35b7e1863ec6d943af9a4360b9d0615c7d02 --- /dev/null +++ b/convnext_small_4xb256_fake7m-lr4e-4/epoch_8.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c61a421dd9612d9ba1df27f79f2c49946d111e7c3a67fba53575e42f52eea310 +size 796447584 diff --git a/convnext_small_4xb256_fake7m-lr4e-4/epoch_9.pth b/convnext_small_4xb256_fake7m-lr4e-4/epoch_9.pth new file mode 100644 index 0000000000000000000000000000000000000000..d66c82b1b89997c1eb1b831742bbf3c699a2d821 --- /dev/null +++ b/convnext_small_4xb256_fake7m-lr4e-4/epoch_9.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:f5438c8f201923592bb525ff07740415c2fc6058933db78a4451a8778cfbbc97 +size 797015072 diff --git a/convnext_small_4xb256_fake7m-lr4e-4/last_checkpoint b/convnext_small_4xb256_fake7m-lr4e-4/last_checkpoint new file mode 100644 index 0000000000000000000000000000000000000000..7b9793febb9c78fdf423ccfacf85a7bc0b8a4d45 --- /dev/null +++ b/convnext_small_4xb256_fake7m-lr4e-4/last_checkpoint @@ -0,0 +1 @@ +/mnt/petrelfs/luzeyu/workspace/fakebench/mmpretrain/workdir/convnext_small_4xb256_fake7m-lr4e-4/epoch_10.pth \ No newline at end of file diff --git a/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_1e-1/20230604_164846/20230604_164846.log b/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_1e-1/20230604_164846/20230604_164846.log new file mode 100644 index 0000000000000000000000000000000000000000..9f7be1c9db12c0af928cbcf781bd61658c5c382d --- /dev/null +++ b/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_1e-1/20230604_164846/20230604_164846.log @@ -0,0 +1,291 @@ +2023/06/04 16:48:51 - mmengine - INFO - +------------------------------------------------------------ +System environment: + sys.platform: linux + Python: 3.10.9 (main, Mar 8 2023, 10:47:38) [GCC 11.2.0] + CUDA available: True + numpy_random_seed: 1960642301 + GPU 0,1,2,3: NVIDIA A100-SXM4-80GB + CUDA_HOME: /mnt/petrelfs/share/cuda-11.6 + NVCC: Cuda compilation tools, release 11.6, V11.6.124 + GCC: gcc (GCC) 7.5.0 + PyTorch: 1.13.1 + PyTorch compiling details: PyTorch built with: + - GCC 9.3 + - C++ Version: 201402 + - Intel(R) oneAPI Math Kernel Library Version 2021.4-Product Build 20210904 for Intel(R) 64 architecture applications + - Intel(R) MKL-DNN v2.6.0 (Git Hash 52b5f107dd9cf10910aaa19cb47f3abf9b349815) + - OpenMP 201511 (a.k.a. OpenMP 4.5) + - LAPACK is enabled (usually provided by MKL) + - NNPACK is enabled + - CPU capability usage: AVX2 + - CUDA Runtime 11.6 + - NVCC architecture flags: -gencode;arch=compute_37,code=sm_37;-gencode;arch=compute_50,code=sm_50;-gencode;arch=compute_60,code=sm_60;-gencode;arch=compute_61,code=sm_61;-gencode;arch=compute_70,code=sm_70;-gencode;arch=compute_75,code=sm_75;-gencode;arch=compute_80,code=sm_80;-gencode;arch=compute_86,code=sm_86;-gencode;arch=compute_37,code=compute_37 + - CuDNN 8.3.2 (built against CUDA 11.5) + - Magma 2.6.1 + - Build settings: BLAS_INFO=mkl, BUILD_TYPE=Release, CUDA_VERSION=11.6, CUDNN_VERSION=8.3.2, CXX_COMPILER=/opt/rh/devtoolset-9/root/usr/bin/c++, CXX_FLAGS= -fabi-version=11 -Wno-deprecated -fvisibility-inlines-hidden -DUSE_PTHREADPOOL -fopenmp -DNDEBUG -DUSE_KINETO -DUSE_FBGEMM -DUSE_QNNPACK -DUSE_PYTORCH_QNNPACK -DUSE_XNNPACK -DSYMBOLICATE_MOBILE_DEBUG_HANDLE -DEDGE_PROFILER_USE_KINETO -O2 -fPIC -Wno-narrowing -Wall -Wextra -Werror=return-type -Werror=non-virtual-dtor -Wno-missing-field-initializers -Wno-type-limits -Wno-array-bounds -Wno-unknown-pragmas -Wunused-local-typedefs -Wno-unused-parameter -Wno-unused-function -Wno-unused-result -Wno-strict-overflow -Wno-strict-aliasing -Wno-error=deprecated-declarations -Wno-stringop-overflow -Wno-psabi -Wno-error=pedantic -Wno-error=redundant-decls -Wno-error=old-style-cast -fdiagnostics-color=always -faligned-new -Wno-unused-but-set-variable -Wno-maybe-uninitialized -fno-math-errno -fno-trapping-math -Werror=format -Werror=cast-function-type -Wno-stringop-overflow, LAPACK_INFO=mkl, PERF_WITH_AVX=1, PERF_WITH_AVX2=1, PERF_WITH_AVX512=1, TORCH_VERSION=1.13.1, USE_CUDA=ON, USE_CUDNN=ON, USE_EXCEPTION_PTR=1, USE_GFLAGS=OFF, USE_GLOG=OFF, USE_MKL=ON, USE_MKLDNN=ON, USE_MPI=OFF, USE_NCCL=ON, USE_NNPACK=ON, USE_OPENMP=ON, USE_ROCM=OFF, + + TorchVision: 0.14.1 + OpenCV: 4.7.0 + MMEngine: 0.7.3 + +Runtime environment: + cudnn_benchmark: True + mp_cfg: {'mp_start_method': 'fork', 'opencv_num_threads': 0} + dist_cfg: {'backend': 'nccl'} + seed: None + deterministic: False + Distributed launcher: slurm + Distributed training: True + GPU number: 4 +------------------------------------------------------------ + +2023/06/04 16:48:56 - mmengine - INFO - Config: +optim_wrapper = dict( + optimizer=dict( + type='AdamW', + lr=0.0004, + weight_decay=0.05, + eps=1e-08, + betas=(0.9, 0.999), + _scope_='mmpretrain'), + paramwise_cfg=dict( + norm_decay_mult=0.0, + bias_decay_mult=0.0, + flat_decay_mult=0.0, + custom_keys=dict({ + '.absolute_pos_embed': dict(decay_mult=0.0), + '.relative_position_bias_table': dict(decay_mult=0.0) + })), + type='AmpOptimWrapper', + dtype='bfloat16', + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=False, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=4096) +model = dict( + type='ImageClassifier', + backbone=dict( + frozen_stages=2, + type='ConvNeXt', + arch='small', + drop_path_rate=0.4, + init_cfg=dict( + type='Pretrained', + checkpoint= + 'https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth', + prefix='backbone')), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=768, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + init_cfg=None), + init_cfg=dict( + type='TruncNormal', layer=['Conv2d', 'Linear'], std=0.02, bias=0.0), + train_cfg=None) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stablediffusionV1-5R2-dpmsolver-25-1m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/cc1m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_1e-1' + diff --git a/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_1e-1/20230604_170206/20230604_170206.log b/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_1e-1/20230604_170206/20230604_170206.log new file mode 100644 index 0000000000000000000000000000000000000000..49c7393f541a86c4caf049f14985183b491e95b3 --- /dev/null +++ b/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_1e-1/20230604_170206/20230604_170206.log @@ -0,0 +1,1840 @@ +2023/06/04 17:02:10 - mmengine - INFO - +------------------------------------------------------------ +System environment: + sys.platform: linux + Python: 3.10.9 (main, Mar 8 2023, 10:47:38) [GCC 11.2.0] + CUDA available: True + numpy_random_seed: 70114618 + GPU 0,1,2,3: NVIDIA A100-SXM4-80GB + CUDA_HOME: /mnt/petrelfs/share/cuda-11.6 + NVCC: Cuda compilation tools, release 11.6, V11.6.124 + GCC: gcc (GCC) 7.5.0 + PyTorch: 1.13.1 + PyTorch compiling details: PyTorch built with: + - GCC 9.3 + - C++ Version: 201402 + - Intel(R) oneAPI Math Kernel Library Version 2021.4-Product Build 20210904 for Intel(R) 64 architecture applications + - Intel(R) MKL-DNN v2.6.0 (Git Hash 52b5f107dd9cf10910aaa19cb47f3abf9b349815) + - OpenMP 201511 (a.k.a. OpenMP 4.5) + - LAPACK is enabled (usually provided by MKL) + - NNPACK is enabled + - CPU capability usage: AVX2 + - CUDA Runtime 11.6 + - NVCC architecture flags: -gencode;arch=compute_37,code=sm_37;-gencode;arch=compute_50,code=sm_50;-gencode;arch=compute_60,code=sm_60;-gencode;arch=compute_61,code=sm_61;-gencode;arch=compute_70,code=sm_70;-gencode;arch=compute_75,code=sm_75;-gencode;arch=compute_80,code=sm_80;-gencode;arch=compute_86,code=sm_86;-gencode;arch=compute_37,code=compute_37 + - CuDNN 8.3.2 (built against CUDA 11.5) + - Magma 2.6.1 + - Build settings: BLAS_INFO=mkl, BUILD_TYPE=Release, CUDA_VERSION=11.6, CUDNN_VERSION=8.3.2, CXX_COMPILER=/opt/rh/devtoolset-9/root/usr/bin/c++, CXX_FLAGS= -fabi-version=11 -Wno-deprecated -fvisibility-inlines-hidden -DUSE_PTHREADPOOL -fopenmp -DNDEBUG -DUSE_KINETO -DUSE_FBGEMM -DUSE_QNNPACK -DUSE_PYTORCH_QNNPACK -DUSE_XNNPACK -DSYMBOLICATE_MOBILE_DEBUG_HANDLE -DEDGE_PROFILER_USE_KINETO -O2 -fPIC -Wno-narrowing -Wall -Wextra -Werror=return-type -Werror=non-virtual-dtor -Wno-missing-field-initializers -Wno-type-limits -Wno-array-bounds -Wno-unknown-pragmas -Wunused-local-typedefs -Wno-unused-parameter -Wno-unused-function -Wno-unused-result -Wno-strict-overflow -Wno-strict-aliasing -Wno-error=deprecated-declarations -Wno-stringop-overflow -Wno-psabi -Wno-error=pedantic -Wno-error=redundant-decls -Wno-error=old-style-cast -fdiagnostics-color=always -faligned-new -Wno-unused-but-set-variable -Wno-maybe-uninitialized -fno-math-errno -fno-trapping-math -Werror=format -Werror=cast-function-type -Wno-stringop-overflow, LAPACK_INFO=mkl, PERF_WITH_AVX=1, PERF_WITH_AVX2=1, PERF_WITH_AVX512=1, TORCH_VERSION=1.13.1, USE_CUDA=ON, USE_CUDNN=ON, USE_EXCEPTION_PTR=1, USE_GFLAGS=OFF, USE_GLOG=OFF, USE_MKL=ON, USE_MKLDNN=ON, USE_MPI=OFF, USE_NCCL=ON, USE_NNPACK=ON, USE_OPENMP=ON, USE_ROCM=OFF, + + TorchVision: 0.14.1 + OpenCV: 4.7.0 + MMEngine: 0.7.3 + +Runtime environment: + cudnn_benchmark: True + mp_cfg: {'mp_start_method': 'fork', 'opencv_num_threads': 0} + dist_cfg: {'backend': 'nccl'} + seed: None + deterministic: False + Distributed launcher: slurm + Distributed training: True + GPU number: 4 +------------------------------------------------------------ + +2023/06/04 17:02:15 - mmengine - INFO - Config: +optim_wrapper = dict( + optimizer=dict( + type='AdamW', + lr=0.0004, + weight_decay=0.05, + eps=1e-08, + betas=(0.9, 0.999), + _scope_='mmpretrain'), + paramwise_cfg=dict( + norm_decay_mult=0.0, + bias_decay_mult=0.0, + flat_decay_mult=0.0, + custom_keys=dict({ + '.absolute_pos_embed': dict(decay_mult=0.0), + '.relative_position_bias_table': dict(decay_mult=0.0) + })), + type='AmpOptimWrapper', + dtype='bfloat16', + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=False, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=4096) +model = dict( + type='ImageClassifier', + backbone=dict( + frozen_stages=2, + type='ConvNeXt', + arch='small', + drop_path_rate=0.4, + init_cfg=dict( + type='Pretrained', + checkpoint= + 'https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth', + prefix='backbone')), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=768, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + init_cfg=None), + init_cfg=dict( + type='TruncNormal', layer=['Conv2d', 'Linear'], std=0.02, bias=0.0), + train_cfg=None) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stablediffusionV1-5R2-dpmsolver-25-1m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/cc1m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_1e-1' + +2023/06/04 17:02:27 - mmengine - INFO - Hooks will be executed in the following order: +before_run: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_load_checkpoint: +(ABOVE_NORMAL) EMAHook + -------------------- +before_train: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_train_epoch: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(NORMAL ) DistSamplerSeedHook + -------------------- +before_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook + -------------------- +after_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +after_train_epoch: +(NORMAL ) IterTimerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_val_epoch: +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook + -------------------- +before_val_iter: +(NORMAL ) IterTimerHook + -------------------- +after_val_iter: +(NORMAL ) IterTimerHook +(NORMAL ) VisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_val_epoch: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_save_checkpoint: +(ABOVE_NORMAL) EMAHook + -------------------- +after_train: +(VERY_LOW ) CheckpointHook + -------------------- +before_test_epoch: +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook + -------------------- +before_test_iter: +(NORMAL ) IterTimerHook + -------------------- +after_test_iter: +(NORMAL ) IterTimerHook +(NORMAL ) VisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_test_epoch: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_run: +(BELOW_NORMAL) LoggerHook + -------------------- +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.2.0.weight:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.2.0.bias:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.2.1.bias:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.3.0.weight:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.3.0.bias:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.3.1.bias:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.2.0.gamma:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.2.0.depthwise_conv.bias:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.2.0.norm.weight:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.2.0.norm.bias:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.2.0.pointwise_conv1.bias:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.2.0.pointwise_conv2.bias:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.2.1.gamma:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.2.1.depthwise_conv.bias:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.2.1.norm.weight:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.2.1.norm.bias:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.2.1.pointwise_conv1.bias:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.2.1.pointwise_conv2.bias:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.2.2.gamma:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.2.2.depthwise_conv.bias:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.2.2.norm.weight:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.2.2.norm.bias:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.2.2.pointwise_conv1.bias:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.2.2.pointwise_conv2.bias:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.2.3.gamma:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.2.3.depthwise_conv.bias:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.2.3.norm.weight:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.2.3.norm.bias:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.2.3.pointwise_conv1.bias:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.2.3.pointwise_conv2.bias:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.2.4.gamma:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.2.4.depthwise_conv.bias:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.2.4.norm.weight:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.2.4.norm.bias:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.2.4.pointwise_conv1.bias:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.2.4.pointwise_conv2.bias:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.2.5.gamma:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.2.5.depthwise_conv.bias:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.2.5.norm.weight:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.2.5.norm.bias:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.2.5.pointwise_conv1.bias:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.2.5.pointwise_conv2.bias:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.2.6.gamma:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.2.6.depthwise_conv.bias:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.2.6.norm.weight:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.2.6.norm.bias:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.2.6.pointwise_conv1.bias:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.2.6.pointwise_conv2.bias:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.2.7.gamma:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.2.7.depthwise_conv.bias:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.2.7.norm.weight:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.2.7.norm.bias:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.2.7.pointwise_conv1.bias:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.2.7.pointwise_conv2.bias:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.2.8.gamma:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.2.8.depthwise_conv.bias:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.2.8.norm.weight:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.2.8.norm.bias:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.2.8.pointwise_conv1.bias:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.2.8.pointwise_conv2.bias:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.2.9.gamma:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.2.9.depthwise_conv.bias:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.2.9.norm.weight:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.2.9.norm.bias:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.2.9.pointwise_conv1.bias:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.2.9.pointwise_conv2.bias:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.2.10.gamma:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.2.10.depthwise_conv.bias:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.2.10.norm.weight:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.2.10.norm.bias:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.2.10.pointwise_conv1.bias:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.2.10.pointwise_conv2.bias:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.2.11.gamma:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.2.11.depthwise_conv.bias:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.2.11.norm.weight:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.2.11.norm.bias:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.2.11.pointwise_conv1.bias:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.2.11.pointwise_conv2.bias:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.2.12.gamma:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.2.12.depthwise_conv.bias:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.2.12.norm.weight:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.2.12.norm.bias:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.2.12.pointwise_conv1.bias:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.2.12.pointwise_conv2.bias:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.2.13.gamma:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.2.13.depthwise_conv.bias:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.2.13.norm.weight:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.2.13.norm.bias:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.2.13.pointwise_conv1.bias:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.2.13.pointwise_conv2.bias:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.2.14.gamma:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.2.14.depthwise_conv.bias:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.2.14.norm.weight:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.2.14.norm.bias:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.2.14.pointwise_conv1.bias:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.2.14.pointwise_conv2.bias:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.2.15.gamma:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.2.15.depthwise_conv.bias:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.2.15.norm.weight:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.2.15.norm.bias:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.2.15.pointwise_conv1.bias:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.2.15.pointwise_conv2.bias:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.2.16.gamma:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.2.16.depthwise_conv.bias:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.2.16.norm.weight:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.2.16.norm.bias:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.2.16.pointwise_conv1.bias:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.2.16.pointwise_conv2.bias:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.2.17.gamma:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.2.17.depthwise_conv.bias:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.2.17.norm.weight:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.2.17.norm.bias:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.2.17.pointwise_conv1.bias:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.2.17.pointwise_conv2.bias:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.2.18.gamma:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.2.18.depthwise_conv.bias:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.2.18.norm.weight:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.2.18.norm.bias:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.2.18.pointwise_conv1.bias:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.2.18.pointwise_conv2.bias:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.2.19.gamma:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.2.19.depthwise_conv.bias:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.2.19.norm.weight:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.2.19.norm.bias:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.2.19.pointwise_conv1.bias:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.2.19.pointwise_conv2.bias:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.2.20.gamma:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.2.20.depthwise_conv.bias:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.2.20.norm.weight:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.2.20.norm.bias:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.2.20.pointwise_conv1.bias:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.2.20.pointwise_conv2.bias:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.2.21.gamma:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.2.21.depthwise_conv.bias:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.2.21.norm.weight:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.2.21.norm.bias:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.2.21.pointwise_conv1.bias:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.2.21.pointwise_conv2.bias:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.2.22.gamma:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.2.22.depthwise_conv.bias:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.2.22.norm.weight:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.2.22.norm.bias:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.2.22.pointwise_conv1.bias:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.2.22.pointwise_conv2.bias:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.2.23.gamma:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.2.23.depthwise_conv.bias:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.2.23.norm.weight:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.2.23.norm.bias:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.2.23.pointwise_conv1.bias:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.2.23.pointwise_conv2.bias:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.2.24.gamma:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.2.24.depthwise_conv.bias:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.2.24.norm.weight:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.2.24.norm.bias:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.2.24.pointwise_conv1.bias:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.2.24.pointwise_conv2.bias:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.2.25.gamma:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.2.25.depthwise_conv.bias:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.2.25.norm.weight:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.2.25.norm.bias:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.2.25.pointwise_conv1.bias:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.2.25.pointwise_conv2.bias:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.2.26.gamma:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.2.26.depthwise_conv.bias:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.2.26.norm.weight:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.2.26.norm.bias:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.2.26.pointwise_conv1.bias:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.2.26.pointwise_conv2.bias:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.3.0.gamma:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.3.0.depthwise_conv.bias:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.3.0.norm.weight:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.3.0.norm.bias:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.3.0.pointwise_conv1.bias:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.3.0.pointwise_conv2.bias:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.3.1.gamma:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.3.1.depthwise_conv.bias:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.3.1.norm.weight:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.3.1.norm.bias:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.3.1.pointwise_conv1.bias:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.3.1.pointwise_conv2.bias:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.3.2.gamma:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.3.2.depthwise_conv.bias:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.3.2.norm.weight:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.3.2.norm.bias:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.3.2.pointwise_conv1.bias:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.stages.3.2.pointwise_conv2.bias:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.norm3.weight:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- backbone.norm3.bias:weight_decay=0.0 +2023/06/04 17:02:41 - mmengine - INFO - paramwise_options -- head.fc.bias:weight_decay=0.0 +2023/06/04 17:02:47 - mmengine - INFO - load backbone in model from: https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth +Name of parameter - Initialization information + +backbone.downsample_layers.0.0.weight - torch.Size([96, 3, 4, 4]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.0.0.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.0.1.weight - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.0.1.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.1.0.weight - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.1.0.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.1.1.weight - torch.Size([192, 96, 2, 2]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.1.1.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.2.0.weight - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.2.0.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.2.1.weight - torch.Size([384, 192, 2, 2]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.2.1.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.3.0.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.3.0.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.3.1.weight - torch.Size([768, 384, 2, 2]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.3.1.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.0.gamma - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.0.depthwise_conv.weight - torch.Size([96, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.0.depthwise_conv.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.0.norm.weight - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.0.norm.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.0.pointwise_conv1.weight - torch.Size([384, 96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.0.pointwise_conv1.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.0.pointwise_conv2.weight - torch.Size([96, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.0.pointwise_conv2.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.1.gamma - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.1.depthwise_conv.weight - torch.Size([96, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.1.depthwise_conv.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.1.norm.weight - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.1.norm.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.1.pointwise_conv1.weight - torch.Size([384, 96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.1.pointwise_conv1.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.1.pointwise_conv2.weight - torch.Size([96, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.1.pointwise_conv2.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.2.gamma - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.2.depthwise_conv.weight - torch.Size([96, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.2.depthwise_conv.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.2.norm.weight - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.2.norm.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.2.pointwise_conv1.weight - torch.Size([384, 96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.2.pointwise_conv1.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.2.pointwise_conv2.weight - torch.Size([96, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.2.pointwise_conv2.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.0.gamma - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.0.depthwise_conv.weight - torch.Size([192, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.0.depthwise_conv.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.0.norm.weight - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.0.norm.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.0.pointwise_conv1.weight - torch.Size([768, 192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.0.pointwise_conv1.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.0.pointwise_conv2.weight - torch.Size([192, 768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.0.pointwise_conv2.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.1.gamma - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.1.depthwise_conv.weight - torch.Size([192, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.1.depthwise_conv.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.1.norm.weight - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.1.norm.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.1.pointwise_conv1.weight - torch.Size([768, 192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.1.pointwise_conv1.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.1.pointwise_conv2.weight - torch.Size([192, 768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.1.pointwise_conv2.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.2.gamma - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.2.depthwise_conv.weight - torch.Size([192, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.2.depthwise_conv.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.2.norm.weight - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.2.norm.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.2.pointwise_conv1.weight - torch.Size([768, 192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.2.pointwise_conv1.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.2.pointwise_conv2.weight - torch.Size([192, 768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.2.pointwise_conv2.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.0.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.0.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.0.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.0.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.0.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.0.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.0.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.0.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.0.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.1.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.1.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.1.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.1.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.1.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.1.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.1.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.1.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.1.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.2.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.2.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.2.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.2.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.2.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.2.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.2.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.2.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.2.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.3.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.3.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.3.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.3.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.3.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.3.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.3.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.3.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.3.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.4.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.4.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.4.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.4.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.4.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.4.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.4.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.4.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.4.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.5.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.5.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.5.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.5.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.5.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.5.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.5.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.5.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.5.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.6.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.6.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.6.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.6.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.6.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.6.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.6.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.6.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.6.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.7.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.7.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.7.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.7.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.7.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.7.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.7.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.7.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.7.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.8.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.8.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.8.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.8.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.8.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.8.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.8.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.8.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.8.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.9.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.9.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.9.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.9.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.9.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.9.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.9.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.9.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.9.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.10.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.10.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.10.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.10.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.10.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.10.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.10.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.10.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.10.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.11.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.11.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.11.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.11.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.11.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.11.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.11.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.11.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.11.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.12.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.12.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.12.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.12.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.12.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.12.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.12.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.12.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.12.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.13.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.13.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.13.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.13.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.13.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.13.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.13.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.13.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.13.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.14.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.14.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.14.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.14.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.14.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.14.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.14.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.14.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.14.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.15.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.15.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.15.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.15.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.15.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.15.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.15.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.15.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.15.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.16.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.16.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.16.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.16.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.16.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.16.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.16.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.16.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.16.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.17.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.17.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.17.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.17.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.17.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.17.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.17.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.17.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.17.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.18.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.18.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.18.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.18.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.18.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.18.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.18.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.18.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.18.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.19.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.19.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.19.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.19.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.19.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.19.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.19.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.19.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.19.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.20.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.20.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.20.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.20.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.20.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.20.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.20.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.20.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.20.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.21.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.21.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.21.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.21.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.21.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.21.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.21.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.21.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.21.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.22.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.22.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.22.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.22.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.22.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.22.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.22.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.22.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.22.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.23.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.23.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.23.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.23.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.23.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.23.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.23.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.23.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.23.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.24.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.24.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.24.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.24.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.24.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.24.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.24.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.24.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.24.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.25.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.25.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.25.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.25.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.25.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.25.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.25.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.25.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.25.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.26.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.26.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.26.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.26.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.26.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.26.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.26.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.26.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.26.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.0.gamma - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.0.depthwise_conv.weight - torch.Size([768, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.0.depthwise_conv.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.0.norm.weight - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.0.norm.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.0.pointwise_conv1.weight - torch.Size([3072, 768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.0.pointwise_conv1.bias - torch.Size([3072]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.0.pointwise_conv2.weight - torch.Size([768, 3072]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.0.pointwise_conv2.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.1.gamma - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.1.depthwise_conv.weight - torch.Size([768, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.1.depthwise_conv.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.1.norm.weight - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.1.norm.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.1.pointwise_conv1.weight - torch.Size([3072, 768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.1.pointwise_conv1.bias - torch.Size([3072]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.1.pointwise_conv2.weight - torch.Size([768, 3072]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.1.pointwise_conv2.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.2.gamma - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.2.depthwise_conv.weight - torch.Size([768, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.2.depthwise_conv.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.2.norm.weight - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.2.norm.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.2.pointwise_conv1.weight - torch.Size([3072, 768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.2.pointwise_conv1.bias - torch.Size([3072]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.2.pointwise_conv2.weight - torch.Size([768, 3072]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.2.pointwise_conv2.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.norm3.weight - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.norm3.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +head.fc.weight - torch.Size([2, 768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +head.fc.bias - torch.Size([2]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 +2023/06/04 17:02:47 - mmengine - WARNING - "FileClient" will be deprecated in future. Please use io functions in https://mmengine.readthedocs.io/en/latest/api/fileio.html#file-io +2023/06/04 17:02:47 - mmengine - WARNING - "HardDiskBackend" is the alias of "LocalBackend" and the former will be deprecated in future. +2023/06/04 17:02:47 - mmengine - INFO - Checkpoints will be saved to /mnt/petrelfs/luzeyu/workspace/fakebench/mmpretrain/workdir/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_1e-1. +2023/06/04 17:04:45 - mmengine - INFO - Epoch(train) [1][ 100/1954] lr: 3.9998e-04 eta: 6:23:00 time: 1.0514 data_time: 0.4075 memory: 15587 loss: 0.0172 +2023/06/04 17:06:31 - mmengine - INFO - Epoch(train) [1][ 200/1954] lr: 3.9990e-04 eta: 6:01:41 time: 1.0496 data_time: 0.3037 memory: 15587 loss: 0.0104 +2023/06/04 17:08:15 - mmengine - INFO - Epoch(train) [1][ 300/1954] lr: 3.9977e-04 eta: 5:50:13 time: 1.0234 data_time: 0.4091 memory: 15587 loss: 0.0237 +2023/06/04 17:09:57 - mmengine - INFO - Epoch(train) [1][ 400/1954] lr: 3.9960e-04 eta: 5:42:44 time: 1.0230 data_time: 0.1444 memory: 15587 loss: 0.0210 +2023/06/04 17:11:39 - mmengine - INFO - Epoch(train) [1][ 500/1954] lr: 3.9937e-04 eta: 5:37:35 time: 0.9647 data_time: 0.0018 memory: 15587 loss: 0.0105 +2023/06/04 17:13:17 - mmengine - INFO - Epoch(train) [1][ 600/1954] lr: 3.9910e-04 eta: 5:31:29 time: 0.9304 data_time: 0.0013 memory: 15587 loss: 0.0091 +2023/06/04 17:14:54 - mmengine - INFO - Epoch(train) [1][ 700/1954] lr: 3.9877e-04 eta: 5:26:10 time: 0.9239 data_time: 0.0015 memory: 15587 loss: 0.0079 +2023/06/04 17:16:28 - mmengine - INFO - Epoch(train) [1][ 800/1954] lr: 3.9839e-04 eta: 5:20:23 time: 0.9333 data_time: 0.0014 memory: 15587 loss: 0.0130 +2023/06/04 17:18:01 - mmengine - INFO - Epoch(train) [1][ 900/1954] lr: 3.9797e-04 eta: 5:15:40 time: 0.9555 data_time: 0.0015 memory: 15587 loss: 0.0046 +2023/06/04 17:19:32 - mmengine - INFO - Exp name: convnext_small_4xb256_sdv1_1m_lr4e-4_aug_1e-1_20230604_170206 +2023/06/04 17:19:32 - mmengine - INFO - Epoch(train) [1][1000/1954] lr: 3.9749e-04 eta: 5:10:34 time: 0.8725 data_time: 0.0016 memory: 15587 loss: 0.0147 +2023/06/04 17:21:00 - mmengine - INFO - Epoch(train) [1][1100/1954] lr: 3.9696e-04 eta: 5:05:27 time: 0.8296 data_time: 0.0015 memory: 15587 loss: 0.0109 +2023/06/04 17:22:27 - mmengine - INFO - Epoch(train) [1][1200/1954] lr: 3.9639e-04 eta: 5:00:38 time: 0.8381 data_time: 0.0016 memory: 15587 loss: 0.0034 +2023/06/04 17:23:54 - mmengine - INFO - Epoch(train) [1][1300/1954] lr: 3.9576e-04 eta: 4:56:14 time: 0.8229 data_time: 0.0016 memory: 15587 loss: 0.0065 +2023/06/04 17:25:18 - mmengine - INFO - Epoch(train) [1][1400/1954] lr: 3.9509e-04 eta: 4:51:51 time: 0.8561 data_time: 0.0015 memory: 15587 loss: 0.0117 +2023/06/04 17:26:43 - mmengine - INFO - Epoch(train) [1][1500/1954] lr: 3.9436e-04 eta: 4:47:52 time: 0.8288 data_time: 0.0017 memory: 15587 loss: 0.0098 +2023/06/04 17:28:09 - mmengine - INFO - Epoch(train) [1][1600/1954] lr: 3.9359e-04 eta: 4:44:31 time: 0.8656 data_time: 0.0016 memory: 15587 loss: 0.0156 +2023/06/04 17:29:33 - mmengine - INFO - Epoch(train) [1][1700/1954] lr: 3.9277e-04 eta: 4:40:52 time: 0.7898 data_time: 0.0015 memory: 15587 loss: 0.0042 +2023/06/04 17:30:56 - mmengine - INFO - Epoch(train) [1][1800/1954] lr: 3.9190e-04 eta: 4:37:24 time: 0.8096 data_time: 0.0016 memory: 15587 loss: 0.0065 +2023/06/04 17:32:18 - mmengine - INFO - Epoch(train) [1][1900/1954] lr: 3.9098e-04 eta: 4:34:03 time: 0.9014 data_time: 0.0019 memory: 15587 loss: 0.0062 +2023/06/04 17:33:03 - mmengine - INFO - Exp name: convnext_small_4xb256_sdv1_1m_lr4e-4_aug_1e-1_20230604_170206 +2023/06/04 17:33:03 - mmengine - INFO - Saving checkpoint at 1 epochs +2023/06/04 17:33:40 - mmengine - INFO - Epoch(val) [1][29/29] accuracy/top1: 90.4666 single-label/precision_classwise: [95.42333984375, 85.47486114501953] single-label/recall_classwise: [86.86951446533203, 94.88372039794922] single-label/f1-score_classwise: [90.94573974609375, 89.93387603759766] data_time: 0.2133 time: 0.6839 +2023/06/04 17:34:27 - mmengine - INFO - Exp name: convnext_small_4xb256_sdv1_1m_lr4e-4_aug_1e-1_20230604_170206 +2023/06/04 17:35:11 - mmengine - INFO - Epoch(train) [2][ 100/1954] lr: 3.8947e-04 eta: 4:30:28 time: 0.8505 data_time: 0.1753 memory: 15587 loss: 0.0030 +2023/06/04 17:36:39 - mmengine - INFO - Epoch(train) [2][ 200/1954] lr: 3.8843e-04 eta: 4:28:20 time: 0.8441 data_time: 0.0022 memory: 15585 loss: 0.0063 +2023/06/04 17:38:05 - mmengine - INFO - Epoch(train) [2][ 300/1954] lr: 3.8735e-04 eta: 4:25:54 time: 0.8124 data_time: 0.0022 memory: 15585 loss: 0.0033 +2023/06/04 17:39:29 - mmengine - INFO - Epoch(train) [2][ 400/1954] lr: 3.8621e-04 eta: 4:23:19 time: 0.8673 data_time: 0.0025 memory: 15585 loss: 0.0095 +2023/06/04 17:40:51 - mmengine - INFO - Epoch(train) [2][ 500/1954] lr: 3.8503e-04 eta: 4:20:43 time: 0.7940 data_time: 0.0016 memory: 15585 loss: 0.0057 +2023/06/04 17:42:18 - mmengine - INFO - Epoch(train) [2][ 600/1954] lr: 3.8380e-04 eta: 4:18:35 time: 0.8247 data_time: 0.0017 memory: 15585 loss: 0.0085 +2023/06/04 17:43:43 - mmengine - INFO - Epoch(train) [2][ 700/1954] lr: 3.8253e-04 eta: 4:16:29 time: 0.8231 data_time: 0.0015 memory: 15585 loss: 0.0046 +2023/06/04 17:45:08 - mmengine - INFO - Epoch(train) [2][ 800/1954] lr: 3.8121e-04 eta: 4:14:18 time: 0.8181 data_time: 0.0016 memory: 15585 loss: 0.0082 +2023/06/04 17:46:34 - mmengine - INFO - Epoch(train) [2][ 900/1954] lr: 3.7984e-04 eta: 4:12:16 time: 0.8724 data_time: 0.0017 memory: 15585 loss: 0.0080 +2023/06/04 17:48:00 - mmengine - INFO - Epoch(train) [2][1000/1954] lr: 3.7843e-04 eta: 4:10:21 time: 0.8331 data_time: 0.0020 memory: 15585 loss: 0.0071 +2023/06/04 17:48:39 - mmengine - INFO - Exp name: convnext_small_4xb256_sdv1_1m_lr4e-4_aug_1e-1_20230604_170206 +2023/06/04 17:49:25 - mmengine - INFO - Epoch(train) [2][1100/1954] lr: 3.7698e-04 eta: 4:08:23 time: 0.8688 data_time: 0.0025 memory: 15585 loss: 0.0100 +2023/06/04 17:50:50 - mmengine - INFO - Epoch(train) [2][1200/1954] lr: 3.7548e-04 eta: 4:06:22 time: 0.8495 data_time: 0.0023 memory: 15585 loss: 0.0043 +2023/06/04 17:52:17 - mmengine - INFO - Epoch(train) [2][1300/1954] lr: 3.7393e-04 eta: 4:04:34 time: 0.9242 data_time: 0.0023 memory: 15585 loss: 0.0035 +2023/06/04 17:53:42 - mmengine - INFO - Epoch(train) [2][1400/1954] lr: 3.7234e-04 eta: 4:02:41 time: 0.8863 data_time: 0.0025 memory: 15585 loss: 0.0034 +2023/06/04 17:55:06 - mmengine - INFO - Epoch(train) [2][1500/1954] lr: 3.7071e-04 eta: 4:00:44 time: 0.8628 data_time: 0.0022 memory: 15585 loss: 0.0102 +2023/06/04 17:56:29 - mmengine - INFO - Epoch(train) [2][1600/1954] lr: 3.6904e-04 eta: 3:58:43 time: 0.8236 data_time: 0.1245 memory: 15585 loss: 0.0058 +2023/06/04 17:57:53 - mmengine - INFO - Epoch(train) [2][1700/1954] lr: 3.6732e-04 eta: 3:56:49 time: 0.8245 data_time: 0.0766 memory: 15585 loss: 0.0099 +2023/06/04 17:59:20 - mmengine - INFO - Epoch(train) [2][1800/1954] lr: 3.6557e-04 eta: 3:55:07 time: 0.8435 data_time: 0.0017 memory: 15585 loss: 0.0092 +2023/06/04 18:00:44 - mmengine - INFO - Epoch(train) [2][1900/1954] lr: 3.6377e-04 eta: 3:53:20 time: 0.8315 data_time: 0.0016 memory: 15585 loss: 0.0033 +2023/06/04 18:01:28 - mmengine - INFO - Exp name: convnext_small_4xb256_sdv1_1m_lr4e-4_aug_1e-1_20230604_170206 +2023/06/04 18:01:28 - mmengine - INFO - Saving checkpoint at 2 epochs +2023/06/04 18:02:04 - mmengine - INFO - Epoch(val) [2][29/29] accuracy/top1: 98.3717 single-label/precision_classwise: [98.14594268798828, 98.65372467041016] single-label/recall_classwise: [98.91421508789062, 97.7054214477539] single-label/f1-score_classwise: [98.5285873413086, 98.17728424072266] data_time: 0.2504 time: 0.6426 +2023/06/04 18:03:29 - mmengine - INFO - Exp name: convnext_small_4xb256_sdv1_1m_lr4e-4_aug_1e-1_20230604_170206 +2023/06/04 18:03:34 - mmengine - INFO - Epoch(train) [3][ 100/1954] lr: 3.6091e-04 eta: 3:50:46 time: 0.8195 data_time: 0.1533 memory: 15585 loss: 0.0037 +2023/06/04 18:05:00 - mmengine - INFO - Epoch(train) [3][ 200/1954] lr: 3.5901e-04 eta: 3:49:08 time: 0.9181 data_time: 0.0023 memory: 15585 loss: 0.0056 +2023/06/04 18:06:26 - mmengine - INFO - Epoch(train) [3][ 300/1954] lr: 3.5707e-04 eta: 3:47:24 time: 0.8385 data_time: 0.0023 memory: 15585 loss: 0.0063 +2023/06/04 18:07:54 - mmengine - INFO - Epoch(train) [3][ 400/1954] lr: 3.5509e-04 eta: 3:45:52 time: 0.8655 data_time: 0.0273 memory: 15585 loss: 0.0042 +2023/06/04 18:09:19 - mmengine - INFO - Epoch(train) [3][ 500/1954] lr: 3.5307e-04 eta: 3:44:10 time: 0.8887 data_time: 0.1688 memory: 15585 loss: 0.0037 +2023/06/04 18:10:46 - mmengine - INFO - Epoch(train) [3][ 600/1954] lr: 3.5101e-04 eta: 3:42:37 time: 0.8659 data_time: 0.3456 memory: 15585 loss: 0.0037 +2023/06/04 18:12:10 - mmengine - INFO - Epoch(train) [3][ 700/1954] lr: 3.4891e-04 eta: 3:40:50 time: 0.8163 data_time: 0.2988 memory: 15585 loss: 0.0057 +2023/06/04 18:13:35 - mmengine - INFO - Epoch(train) [3][ 800/1954] lr: 3.4677e-04 eta: 3:39:11 time: 0.8540 data_time: 0.0027 memory: 15585 loss: 0.0068 +2023/06/04 18:15:02 - mmengine - INFO - Epoch(train) [3][ 900/1954] lr: 3.4460e-04 eta: 3:37:36 time: 0.8360 data_time: 0.0024 memory: 15585 loss: 0.0072 +2023/06/04 18:16:28 - mmengine - INFO - Epoch(train) [3][1000/1954] lr: 3.4240e-04 eta: 3:36:01 time: 0.7947 data_time: 0.0023 memory: 15585 loss: 0.0106 +2023/06/04 18:17:47 - mmengine - INFO - Exp name: convnext_small_4xb256_sdv1_1m_lr4e-4_aug_1e-1_20230604_170206 +2023/06/04 18:17:51 - mmengine - INFO - Epoch(train) [3][1100/1954] lr: 3.4015e-04 eta: 3:34:15 time: 0.8103 data_time: 0.0020 memory: 15585 loss: 0.0050 +2023/06/04 18:19:16 - mmengine - INFO - Epoch(train) [3][1200/1954] lr: 3.3788e-04 eta: 3:32:37 time: 0.8360 data_time: 0.0016 memory: 15585 loss: 0.0077 +2023/06/04 18:20:42 - mmengine - INFO - Epoch(train) [3][1300/1954] lr: 3.3557e-04 eta: 3:31:01 time: 0.8929 data_time: 0.0016 memory: 15585 loss: 0.0054 +2023/06/04 18:22:07 - mmengine - INFO - Epoch(train) [3][1400/1954] lr: 3.3322e-04 eta: 3:29:25 time: 0.8350 data_time: 0.0024 memory: 15585 loss: 0.0067 +2023/06/04 18:23:32 - mmengine - INFO - Epoch(train) [3][1500/1954] lr: 3.3084e-04 eta: 3:27:49 time: 0.8931 data_time: 0.0019 memory: 15585 loss: 0.0037 +2023/06/04 18:24:57 - mmengine - INFO - Epoch(train) [3][1600/1954] lr: 3.2843e-04 eta: 3:26:12 time: 0.8709 data_time: 0.0020 memory: 15585 loss: 0.0053 +2023/06/04 18:26:22 - mmengine - INFO - Epoch(train) [3][1700/1954] lr: 3.2599e-04 eta: 3:24:35 time: 0.8444 data_time: 0.0017 memory: 15585 loss: 0.0032 +2023/06/04 18:27:48 - mmengine - INFO - Epoch(train) [3][1800/1954] lr: 3.2351e-04 eta: 3:23:02 time: 0.8666 data_time: 0.0018 memory: 15585 loss: 0.0033 +2023/06/04 18:29:13 - mmengine - INFO - Epoch(train) [3][1900/1954] lr: 3.2101e-04 eta: 3:21:27 time: 0.8863 data_time: 0.0023 memory: 15585 loss: 0.0049 +2023/06/04 18:30:00 - mmengine - INFO - Exp name: convnext_small_4xb256_sdv1_1m_lr4e-4_aug_1e-1_20230604_170206 +2023/06/04 18:30:00 - mmengine - INFO - Saving checkpoint at 3 epochs +2023/06/04 18:30:36 - mmengine - INFO - Epoch(val) [3][29/29] accuracy/top1: 99.4572 single-label/precision_classwise: [99.23410034179688, 99.734619140625] single-label/recall_classwise: [99.78536987304688, 99.05426788330078] single-label/f1-score_classwise: [99.50897216796875, 99.39328002929688] data_time: 0.2422 time: 0.6293 +2023/06/04 18:32:06 - mmengine - INFO - Epoch(train) [4][ 100/1954] lr: 3.1709e-04 eta: 3:19:15 time: 0.8580 data_time: 0.3000 memory: 15585 loss: 0.0048 +2023/06/04 18:32:40 - mmengine - INFO - Exp name: convnext_small_4xb256_sdv1_1m_lr4e-4_aug_1e-1_20230604_170206 +2023/06/04 18:33:31 - mmengine - INFO - Epoch(train) [4][ 200/1954] lr: 3.1451e-04 eta: 3:17:39 time: 0.7942 data_time: 0.0438 memory: 15585 loss: 0.0059 +2023/06/04 18:34:54 - mmengine - INFO - Epoch(train) [4][ 300/1954] lr: 3.1190e-04 eta: 3:16:01 time: 0.8303 data_time: 0.1321 memory: 15585 loss: 0.0015 +2023/06/04 18:36:19 - mmengine - INFO - Epoch(train) [4][ 400/1954] lr: 3.0927e-04 eta: 3:14:28 time: 0.8560 data_time: 0.0363 memory: 15585 loss: 0.0050 +2023/06/04 18:37:43 - mmengine - INFO - Epoch(train) [4][ 500/1954] lr: 3.0661e-04 eta: 3:12:51 time: 0.8093 data_time: 0.0677 memory: 15585 loss: 0.0058 +2023/06/04 18:39:09 - mmengine - INFO - Epoch(train) [4][ 600/1954] lr: 3.0392e-04 eta: 3:11:20 time: 0.8326 data_time: 0.0015 memory: 15585 loss: 0.0060 +2023/06/04 18:40:33 - mmengine - INFO - Epoch(train) [4][ 700/1954] lr: 3.0120e-04 eta: 3:09:45 time: 0.7895 data_time: 0.0016 memory: 15585 loss: 0.0097 +2023/06/04 18:41:57 - mmengine - INFO - Epoch(train) [4][ 800/1954] lr: 2.9846e-04 eta: 3:08:09 time: 0.8677 data_time: 0.2796 memory: 15585 loss: 0.0061 +2023/06/04 18:43:21 - mmengine - INFO - Epoch(train) [4][ 900/1954] lr: 2.9570e-04 eta: 3:06:35 time: 0.8668 data_time: 0.3513 memory: 15585 loss: 0.0040 +2023/06/04 18:44:46 - mmengine - INFO - Epoch(train) [4][1000/1954] lr: 2.9291e-04 eta: 3:05:03 time: 0.8544 data_time: 0.3394 memory: 15585 loss: 0.0084 +2023/06/04 18:46:12 - mmengine - INFO - Epoch(train) [4][1100/1954] lr: 2.9010e-04 eta: 3:03:31 time: 0.8620 data_time: 0.1649 memory: 15585 loss: 0.0063 +2023/06/04 18:46:47 - mmengine - INFO - Exp name: convnext_small_4xb256_sdv1_1m_lr4e-4_aug_1e-1_20230604_170206 +2023/06/04 18:47:40 - mmengine - INFO - Epoch(train) [4][1200/1954] lr: 2.8727e-04 eta: 3:02:06 time: 0.8409 data_time: 0.0015 memory: 15585 loss: 0.0066 +2023/06/04 18:49:07 - mmengine - INFO - Epoch(train) [4][1300/1954] lr: 2.8442e-04 eta: 3:00:36 time: 0.7618 data_time: 0.0016 memory: 15585 loss: 0.0050 +2023/06/04 18:50:30 - mmengine - INFO - Epoch(train) [4][1400/1954] lr: 2.8155e-04 eta: 2:59:01 time: 0.7866 data_time: 0.0015 memory: 15585 loss: 0.0054 +2023/06/04 18:51:53 - mmengine - INFO - Epoch(train) [4][1500/1954] lr: 2.7865e-04 eta: 2:57:26 time: 0.8116 data_time: 0.0018 memory: 15585 loss: 0.0049 +2023/06/04 18:53:18 - mmengine - INFO - Epoch(train) [4][1600/1954] lr: 2.7574e-04 eta: 2:55:55 time: 0.7899 data_time: 0.0016 memory: 15585 loss: 0.0044 +2023/06/04 18:54:44 - mmengine - INFO - Epoch(train) [4][1700/1954] lr: 2.7281e-04 eta: 2:54:25 time: 0.8063 data_time: 0.1027 memory: 15585 loss: 0.0094 +2023/06/04 18:56:12 - mmengine - INFO - Epoch(train) [4][1800/1954] lr: 2.6986e-04 eta: 2:52:58 time: 0.9091 data_time: 0.2126 memory: 15585 loss: 0.0053 +2023/06/04 18:57:40 - mmengine - INFO - Epoch(train) [4][1900/1954] lr: 2.6690e-04 eta: 2:51:32 time: 0.7987 data_time: 0.2828 memory: 15585 loss: 0.0040 +2023/06/04 18:58:23 - mmengine - INFO - Exp name: convnext_small_4xb256_sdv1_1m_lr4e-4_aug_1e-1_20230604_170206 +2023/06/04 18:58:23 - mmengine - INFO - Saving checkpoint at 4 epochs +2023/06/04 18:59:00 - mmengine - INFO - Epoch(val) [4][29/29] accuracy/top1: 99.7773 single-label/precision_classwise: [99.66630554199219, 99.9144515991211] single-label/recall_classwise: [99.93055725097656, 99.58914184570312] single-label/f1-score_classwise: [99.79824829101562, 99.75152587890625] data_time: 0.2668 time: 0.6511 +2023/06/04 19:00:29 - mmengine - INFO - Epoch(train) [5][ 100/1954] lr: 2.6230e-04 eta: 2:49:15 time: 0.8457 data_time: 0.2097 memory: 15585 loss: 0.0012 +2023/06/04 19:01:44 - mmengine - INFO - Exp name: convnext_small_4xb256_sdv1_1m_lr4e-4_aug_1e-1_20230604_170206 +2023/06/04 19:01:56 - mmengine - INFO - Epoch(train) [5][ 200/1954] lr: 2.5929e-04 eta: 2:47:47 time: 0.8890 data_time: 0.0246 memory: 15585 loss: 0.0033 +2023/06/04 19:03:22 - mmengine - INFO - Epoch(train) [5][ 300/1954] lr: 2.5628e-04 eta: 2:46:17 time: 0.8516 data_time: 0.0018 memory: 15585 loss: 0.0019 +2023/06/04 19:04:50 - mmengine - INFO - Epoch(train) [5][ 400/1954] lr: 2.5325e-04 eta: 2:44:50 time: 0.8578 data_time: 0.0020 memory: 15585 loss: 0.0064 +2023/06/04 19:06:17 - mmengine - INFO - Epoch(train) [5][ 500/1954] lr: 2.5020e-04 eta: 2:43:23 time: 0.9104 data_time: 0.0015 memory: 15585 loss: 0.0039 +2023/06/04 19:07:44 - mmengine - INFO - Epoch(train) [5][ 600/1954] lr: 2.4715e-04 eta: 2:41:56 time: 0.8483 data_time: 0.0015 memory: 15585 loss: 0.0016 +2023/06/04 19:09:12 - mmengine - INFO - Epoch(train) [5][ 700/1954] lr: 2.4408e-04 eta: 2:40:28 time: 0.8450 data_time: 0.0018 memory: 15585 loss: 0.0040 +2023/06/04 19:10:39 - mmengine - INFO - Epoch(train) [5][ 800/1954] lr: 2.4100e-04 eta: 2:39:01 time: 0.8575 data_time: 0.0016 memory: 15585 loss: 0.0020 +2023/06/04 19:12:06 - mmengine - INFO - Epoch(train) [5][ 900/1954] lr: 2.3792e-04 eta: 2:37:33 time: 0.9226 data_time: 0.0016 memory: 15585 loss: 0.0085 +2023/06/04 19:13:32 - mmengine - INFO - Epoch(train) [5][1000/1954] lr: 2.3482e-04 eta: 2:36:04 time: 0.8550 data_time: 0.0017 memory: 15585 loss: 0.0074 +2023/06/04 19:14:59 - mmengine - INFO - Epoch(train) [5][1100/1954] lr: 2.3172e-04 eta: 2:34:36 time: 0.8735 data_time: 0.0016 memory: 15585 loss: 0.0028 +2023/06/04 19:16:10 - mmengine - INFO - Exp name: convnext_small_4xb256_sdv1_1m_lr4e-4_aug_1e-1_20230604_170206 +2023/06/04 19:16:26 - mmengine - INFO - Epoch(train) [5][1200/1954] lr: 2.2861e-04 eta: 2:33:08 time: 0.8721 data_time: 0.0017 memory: 15585 loss: 0.0055 +2023/06/04 19:17:51 - mmengine - INFO - Epoch(train) [5][1300/1954] lr: 2.2550e-04 eta: 2:31:39 time: 0.9107 data_time: 0.0019 memory: 15585 loss: 0.0029 +2023/06/04 19:19:20 - mmengine - INFO - Epoch(train) [5][1400/1954] lr: 2.2238e-04 eta: 2:30:13 time: 0.8399 data_time: 0.0016 memory: 15585 loss: 0.0038 +2023/06/04 19:20:45 - mmengine - INFO - Epoch(train) [5][1500/1954] lr: 2.1925e-04 eta: 2:28:44 time: 0.9287 data_time: 0.0016 memory: 15585 loss: 0.0014 +2023/06/04 19:22:12 - mmengine - INFO - Epoch(train) [5][1600/1954] lr: 2.1612e-04 eta: 2:27:16 time: 0.8815 data_time: 0.0016 memory: 15585 loss: 0.0025 +2023/06/04 19:23:37 - mmengine - INFO - Epoch(train) [5][1700/1954] lr: 2.1299e-04 eta: 2:25:46 time: 0.8672 data_time: 0.0018 memory: 15585 loss: 0.0035 +2023/06/04 19:25:06 - mmengine - INFO - Epoch(train) [5][1800/1954] lr: 2.0986e-04 eta: 2:24:21 time: 0.8194 data_time: 0.0018 memory: 15585 loss: 0.0036 +2023/06/04 19:26:34 - mmengine - INFO - Epoch(train) [5][1900/1954] lr: 2.0672e-04 eta: 2:22:54 time: 0.8633 data_time: 0.0017 memory: 15585 loss: 0.0037 +2023/06/04 19:27:17 - mmengine - INFO - Exp name: convnext_small_4xb256_sdv1_1m_lr4e-4_aug_1e-1_20230604_170206 +2023/06/04 19:27:17 - mmengine - INFO - Saving checkpoint at 5 epochs +2023/06/04 19:27:54 - mmengine - INFO - Epoch(val) [5][29/29] accuracy/top1: 99.8434 single-label/precision_classwise: [99.75431823730469, 99.953369140625] single-label/recall_classwise: [99.96212005615234, 99.69766998291016] single-label/f1-score_classwise: [99.85811614990234, 99.82536315917969] data_time: 0.2372 time: 0.6240 +2023/06/04 19:29:25 - mmengine - INFO - Epoch(train) [6][ 100/1954] lr: 2.0190e-04 eta: 2:20:40 time: 0.8291 data_time: 0.3125 memory: 15585 loss: 0.0024 +2023/06/04 19:30:50 - mmengine - INFO - Epoch(train) [6][ 200/1954] lr: 1.9876e-04 eta: 2:19:10 time: 0.8245 data_time: 0.3061 memory: 15585 loss: 0.0024 +2023/06/04 19:31:15 - mmengine - INFO - Exp name: convnext_small_4xb256_sdv1_1m_lr4e-4_aug_1e-1_20230604_170206 +2023/06/04 19:32:13 - mmengine - INFO - Epoch(train) [6][ 300/1954] lr: 1.9563e-04 eta: 2:17:39 time: 0.8461 data_time: 0.2992 memory: 15585 loss: 0.0011 +2023/06/04 19:33:37 - mmengine - INFO - Epoch(train) [6][ 400/1954] lr: 1.9250e-04 eta: 2:16:09 time: 0.8310 data_time: 0.2791 memory: 15585 loss: 0.0015 +2023/06/04 19:35:07 - mmengine - INFO - Epoch(train) [6][ 500/1954] lr: 1.8937e-04 eta: 2:14:44 time: 0.9613 data_time: 0.0015 memory: 15585 loss: 0.0056 +2023/06/04 19:36:37 - mmengine - INFO - Epoch(train) [6][ 600/1954] lr: 1.8625e-04 eta: 2:13:19 time: 0.9406 data_time: 0.0016 memory: 15585 loss: 0.0036 +2023/06/04 19:38:03 - mmengine - INFO - Epoch(train) [6][ 700/1954] lr: 1.8313e-04 eta: 2:11:52 time: 0.8658 data_time: 0.0020 memory: 15585 loss: 0.0054 +2023/06/04 19:39:31 - mmengine - INFO - Epoch(train) [6][ 800/1954] lr: 1.8002e-04 eta: 2:10:24 time: 0.8828 data_time: 0.0016 memory: 15585 loss: 0.0042 +2023/06/04 19:40:59 - mmengine - INFO - Epoch(train) [6][ 900/1954] lr: 1.7691e-04 eta: 2:08:58 time: 0.9321 data_time: 0.0017 memory: 15585 loss: 0.0021 +2023/06/04 19:42:26 - mmengine - INFO - Epoch(train) [6][1000/1954] lr: 1.7381e-04 eta: 2:07:30 time: 0.9093 data_time: 0.0016 memory: 15585 loss: 0.0020 +2023/06/04 19:43:52 - mmengine - INFO - Epoch(train) [6][1100/1954] lr: 1.7072e-04 eta: 2:06:03 time: 0.8919 data_time: 0.0016 memory: 15585 loss: 0.0021 +2023/06/04 19:45:17 - mmengine - INFO - Epoch(train) [6][1200/1954] lr: 1.6764e-04 eta: 2:04:33 time: 0.8498 data_time: 0.0021 memory: 15585 loss: 0.0033 +2023/06/04 19:45:42 - mmengine - INFO - Exp name: convnext_small_4xb256_sdv1_1m_lr4e-4_aug_1e-1_20230604_170206 +2023/06/04 19:46:39 - mmengine - INFO - Epoch(train) [6][1300/1954] lr: 1.6457e-04 eta: 2:03:02 time: 0.8278 data_time: 0.0017 memory: 15585 loss: 0.0008 +2023/06/04 19:48:01 - mmengine - INFO - Epoch(train) [6][1400/1954] lr: 1.6151e-04 eta: 2:01:32 time: 0.8337 data_time: 0.0016 memory: 15585 loss: 0.0033 +2023/06/04 19:49:28 - mmengine - INFO - Epoch(train) [6][1500/1954] lr: 1.5846e-04 eta: 2:00:04 time: 0.8705 data_time: 0.0018 memory: 15585 loss: 0.0008 +2023/06/04 19:50:54 - mmengine - INFO - Epoch(train) [6][1600/1954] lr: 1.5542e-04 eta: 1:58:36 time: 0.8746 data_time: 0.0015 memory: 15585 loss: 0.0025 +2023/06/04 19:52:16 - mmengine - INFO - Epoch(train) [6][1700/1954] lr: 1.5239e-04 eta: 1:57:06 time: 0.8247 data_time: 0.0016 memory: 15585 loss: 0.0024 +2023/06/04 19:53:41 - mmengine - INFO - Epoch(train) [6][1800/1954] lr: 1.4938e-04 eta: 1:55:37 time: 0.8379 data_time: 0.0017 memory: 15585 loss: 0.0022 +2023/06/04 19:55:05 - mmengine - INFO - Epoch(train) [6][1900/1954] lr: 1.4638e-04 eta: 1:54:08 time: 0.8156 data_time: 0.0016 memory: 15585 loss: 0.0031 +2023/06/04 19:55:50 - mmengine - INFO - Exp name: convnext_small_4xb256_sdv1_1m_lr4e-4_aug_1e-1_20230604_170206 +2023/06/04 19:55:50 - mmengine - INFO - Saving checkpoint at 6 epochs +2023/06/04 19:56:27 - mmengine - INFO - Epoch(val) [6][29/29] accuracy/top1: 99.8747 single-label/precision_classwise: [99.80462646484375, 99.9611587524414] single-label/recall_classwise: [99.96842956542969, 99.75968933105469] single-label/f1-score_classwise: [99.88645935058594, 99.86031341552734] data_time: 0.2480 time: 0.6316 +2023/06/04 19:57:58 - mmengine - INFO - Epoch(train) [7][ 100/1954] lr: 1.4180e-04 eta: 1:51:55 time: 0.8383 data_time: 0.3229 memory: 15585 loss: 0.0004 +2023/06/04 19:59:25 - mmengine - INFO - Epoch(train) [7][ 200/1954] lr: 1.3884e-04 eta: 1:50:28 time: 0.8227 data_time: 0.0017 memory: 15585 loss: 0.0021 +2023/06/04 20:00:30 - mmengine - INFO - Exp name: convnext_small_4xb256_sdv1_1m_lr4e-4_aug_1e-1_20230604_170206 +2023/06/04 20:00:49 - mmengine - INFO - Epoch(train) [7][ 300/1954] lr: 1.3590e-04 eta: 1:48:59 time: 0.8198 data_time: 0.0016 memory: 15585 loss: 0.0034 +2023/06/04 20:02:14 - mmengine - INFO - Epoch(train) [7][ 400/1954] lr: 1.3298e-04 eta: 1:47:31 time: 0.8423 data_time: 0.0016 memory: 15585 loss: 0.0017 +2023/06/04 20:03:38 - mmengine - INFO - Epoch(train) [7][ 500/1954] lr: 1.3007e-04 eta: 1:46:02 time: 0.8357 data_time: 0.0019 memory: 15585 loss: 0.0038 +2023/06/04 20:05:03 - mmengine - INFO - Epoch(train) [7][ 600/1954] lr: 1.2719e-04 eta: 1:44:34 time: 0.8166 data_time: 0.0019 memory: 15585 loss: 0.0010 +2023/06/04 20:06:30 - mmengine - INFO - Epoch(train) [7][ 700/1954] lr: 1.2432e-04 eta: 1:43:07 time: 0.8601 data_time: 0.0016 memory: 15585 loss: 0.0013 +2023/06/04 20:07:56 - mmengine - INFO - Epoch(train) [7][ 800/1954] lr: 1.2148e-04 eta: 1:41:39 time: 0.8406 data_time: 0.0016 memory: 15585 loss: 0.0015 +2023/06/04 20:09:21 - mmengine - INFO - Epoch(train) [7][ 900/1954] lr: 1.1866e-04 eta: 1:40:12 time: 0.8938 data_time: 0.0018 memory: 15585 loss: 0.0039 +2023/06/04 20:10:47 - mmengine - INFO - Epoch(train) [7][1000/1954] lr: 1.1586e-04 eta: 1:38:44 time: 0.8393 data_time: 0.0016 memory: 15585 loss: 0.0029 +2023/06/04 20:12:11 - mmengine - INFO - Epoch(train) [7][1100/1954] lr: 1.1308e-04 eta: 1:37:16 time: 0.8667 data_time: 0.0016 memory: 15585 loss: 0.0026 +2023/06/04 20:13:36 - mmengine - INFO - Epoch(train) [7][1200/1954] lr: 1.1033e-04 eta: 1:35:48 time: 0.8391 data_time: 0.0016 memory: 15585 loss: 0.0006 +2023/06/04 20:14:44 - mmengine - INFO - Exp name: convnext_small_4xb256_sdv1_1m_lr4e-4_aug_1e-1_20230604_170206 +2023/06/04 20:15:04 - mmengine - INFO - Epoch(train) [7][1300/1954] lr: 1.0760e-04 eta: 1:34:21 time: 0.9049 data_time: 0.0018 memory: 15585 loss: 0.0019 +2023/06/04 20:16:28 - mmengine - INFO - Epoch(train) [7][1400/1954] lr: 1.0490e-04 eta: 1:32:53 time: 0.8195 data_time: 0.0019 memory: 15585 loss: 0.0023 +2023/06/04 20:17:53 - mmengine - INFO - Epoch(train) [7][1500/1954] lr: 1.0222e-04 eta: 1:31:25 time: 0.8479 data_time: 0.0016 memory: 15585 loss: 0.0046 +2023/06/04 20:19:20 - mmengine - INFO - Epoch(train) [7][1600/1954] lr: 9.9568e-05 eta: 1:29:58 time: 0.8420 data_time: 0.0015 memory: 15585 loss: 0.0008 +2023/06/04 20:20:50 - mmengine - INFO - Epoch(train) [7][1700/1954] lr: 9.6944e-05 eta: 1:28:33 time: 0.8856 data_time: 0.0020 memory: 15585 loss: 0.0009 +2023/06/04 20:22:13 - mmengine - INFO - Epoch(train) [7][1800/1954] lr: 9.4348e-05 eta: 1:27:05 time: 0.8360 data_time: 0.0021 memory: 15585 loss: 0.0006 +2023/06/04 20:23:37 - mmengine - INFO - Epoch(train) [7][1900/1954] lr: 9.1781e-05 eta: 1:25:36 time: 0.8575 data_time: 0.0018 memory: 15585 loss: 0.0041 +2023/06/04 20:24:21 - mmengine - INFO - Exp name: convnext_small_4xb256_sdv1_1m_lr4e-4_aug_1e-1_20230604_170206 +2023/06/04 20:24:21 - mmengine - INFO - Saving checkpoint at 7 epochs +2023/06/04 20:24:57 - mmengine - INFO - Epoch(val) [7][29/29] accuracy/top1: 99.8921 single-label/precision_classwise: [99.82352447509766, 99.97669982910156] single-label/recall_classwise: [99.98106384277344, 99.78294372558594] single-label/f1-score_classwise: [99.90222930908203, 99.87973022460938] data_time: 0.2443 time: 0.6280 +2023/06/04 20:26:31 - mmengine - INFO - Epoch(train) [8][ 100/1954] lr: 8.7885e-05 eta: 1:23:24 time: 0.8846 data_time: 0.0018 memory: 15585 loss: 0.0017 +2023/06/04 20:27:56 - mmengine - INFO - Epoch(train) [8][ 200/1954] lr: 8.5394e-05 eta: 1:21:57 time: 0.8424 data_time: 0.0015 memory: 15585 loss: 0.0024 +2023/06/04 20:29:23 - mmengine - INFO - Epoch(train) [8][ 300/1954] lr: 8.2933e-05 eta: 1:20:30 time: 0.8495 data_time: 0.0015 memory: 15585 loss: 0.0002 +2023/06/04 20:29:46 - mmengine - INFO - Exp name: convnext_small_4xb256_sdv1_1m_lr4e-4_aug_1e-1_20230604_170206 +2023/06/04 20:30:48 - mmengine - INFO - Epoch(train) [8][ 400/1954] lr: 8.0504e-05 eta: 1:19:02 time: 0.8393 data_time: 0.0016 memory: 15585 loss: 0.0004 +2023/06/04 20:32:12 - mmengine - INFO - Epoch(train) [8][ 500/1954] lr: 7.8107e-05 eta: 1:17:34 time: 0.8204 data_time: 0.0017 memory: 15585 loss: 0.0004 +2023/06/04 20:33:36 - mmengine - INFO - Epoch(train) [8][ 600/1954] lr: 7.5743e-05 eta: 1:16:07 time: 0.8491 data_time: 0.0175 memory: 15585 loss: 0.0003 +2023/06/04 20:34:59 - mmengine - INFO - Epoch(train) [8][ 700/1954] lr: 7.3413e-05 eta: 1:14:38 time: 0.7804 data_time: 0.0844 memory: 15585 loss: 0.0019 +2023/06/04 20:36:23 - mmengine - INFO - Epoch(train) [8][ 800/1954] lr: 7.1116e-05 eta: 1:13:10 time: 0.8845 data_time: 0.1173 memory: 15585 loss: 0.0004 +2023/06/04 20:37:48 - mmengine - INFO - Epoch(train) [8][ 900/1954] lr: 6.8854e-05 eta: 1:11:43 time: 0.8761 data_time: 0.3358 memory: 15585 loss: 0.0009 +2023/06/04 20:39:28 - mmengine - INFO - Epoch(train) [8][1000/1954] lr: 6.6627e-05 eta: 1:10:21 time: 1.1074 data_time: 0.5895 memory: 15585 loss: 0.0014 +2023/06/04 20:41:01 - mmengine - INFO - Epoch(train) [8][1100/1954] lr: 6.4436e-05 eta: 1:08:56 time: 0.7262 data_time: 0.2097 memory: 15585 loss: 0.0008 +2023/06/04 20:42:16 - mmengine - INFO - Epoch(train) [8][1200/1954] lr: 6.2282e-05 eta: 1:07:25 time: 0.7830 data_time: 0.2032 memory: 15585 loss: 0.0012 +2023/06/04 20:43:36 - mmengine - INFO - Epoch(train) [8][1300/1954] lr: 6.0164e-05 eta: 1:05:57 time: 0.7770 data_time: 0.0015 memory: 15585 loss: 0.0018 +2023/06/04 20:43:55 - mmengine - INFO - Exp name: convnext_small_4xb256_sdv1_1m_lr4e-4_aug_1e-1_20230604_170206 +2023/06/04 20:45:00 - mmengine - INFO - Epoch(train) [8][1400/1954] lr: 5.8083e-05 eta: 1:04:29 time: 0.8211 data_time: 0.0326 memory: 15585 loss: 0.0026 +2023/06/04 20:46:23 - mmengine - INFO - Epoch(train) [8][1500/1954] lr: 5.6041e-05 eta: 1:03:01 time: 0.8388 data_time: 0.1047 memory: 15585 loss: 0.0008 +2023/06/04 20:47:48 - mmengine - INFO - Epoch(train) [8][1600/1954] lr: 5.4037e-05 eta: 1:01:34 time: 0.8542 data_time: 0.1727 memory: 15585 loss: 0.0008 +2023/06/04 20:49:15 - mmengine - INFO - Epoch(train) [8][1700/1954] lr: 5.2072e-05 eta: 1:00:08 time: 0.8717 data_time: 0.3553 memory: 15585 loss: 0.0013 +2023/06/04 20:50:39 - mmengine - INFO - Epoch(train) [8][1800/1954] lr: 5.0147e-05 eta: 0:58:40 time: 0.8721 data_time: 0.3573 memory: 15585 loss: 0.0007 +2023/06/04 20:52:04 - mmengine - INFO - Epoch(train) [8][1900/1954] lr: 4.8261e-05 eta: 0:57:13 time: 0.8718 data_time: 0.3548 memory: 15585 loss: 0.0060 +2023/06/04 20:52:48 - mmengine - INFO - Exp name: convnext_small_4xb256_sdv1_1m_lr4e-4_aug_1e-1_20230604_170206 +2023/06/04 20:52:48 - mmengine - INFO - Saving checkpoint at 8 epochs +2023/06/04 20:53:25 - mmengine - INFO - Epoch(val) [8][29/29] accuracy/top1: 99.9235 single-label/precision_classwise: [99.88018035888672, 99.97671508789062] single-label/recall_classwise: [99.98106384277344, 99.85271453857422] single-label/f1-score_classwise: [99.93060302734375, 99.9146728515625] data_time: 0.2632 time: 0.6462 +2023/06/04 20:54:57 - mmengine - INFO - Epoch(train) [9][ 100/1954] lr: 4.5437e-05 eta: 0:55:00 time: 0.8541 data_time: 0.1202 memory: 15585 loss: 0.0011 +2023/06/04 20:56:23 - mmengine - INFO - Epoch(train) [9][ 200/1954] lr: 4.3656e-05 eta: 0:53:33 time: 0.7904 data_time: 0.0016 memory: 15585 loss: 0.0006 +2023/06/04 20:57:51 - mmengine - INFO - Epoch(train) [9][ 300/1954] lr: 4.1916e-05 eta: 0:52:07 time: 0.8504 data_time: 0.0015 memory: 15585 loss: 0.0006 +2023/06/04 20:58:51 - mmengine - INFO - Exp name: convnext_small_4xb256_sdv1_1m_lr4e-4_aug_1e-1_20230604_170206 +2023/06/04 20:59:18 - mmengine - INFO - Epoch(train) [9][ 400/1954] lr: 4.0218e-05 eta: 0:50:40 time: 0.8378 data_time: 0.0016 memory: 15585 loss: 0.0007 +2023/06/04 21:00:47 - mmengine - INFO - Epoch(train) [9][ 500/1954] lr: 3.8563e-05 eta: 0:49:14 time: 0.8297 data_time: 0.0018 memory: 15585 loss: 0.0013 +2023/06/04 21:02:13 - mmengine - INFO - Epoch(train) [9][ 600/1954] lr: 3.6951e-05 eta: 0:47:47 time: 0.8879 data_time: 0.0018 memory: 15585 loss: 0.0045 +2023/06/04 21:03:39 - mmengine - INFO - Epoch(train) [9][ 700/1954] lr: 3.5383e-05 eta: 0:46:20 time: 0.8603 data_time: 0.0015 memory: 15585 loss: 0.0007 +2023/06/04 21:05:05 - mmengine - INFO - Epoch(train) [9][ 800/1954] lr: 3.3858e-05 eta: 0:44:54 time: 0.9071 data_time: 0.0017 memory: 15585 loss: 0.0047 +2023/06/04 21:06:31 - mmengine - INFO - Epoch(train) [9][ 900/1954] lr: 3.2378e-05 eta: 0:43:27 time: 0.8494 data_time: 0.0843 memory: 15585 loss: 0.0008 +2023/06/04 21:07:55 - mmengine - INFO - Epoch(train) [9][1000/1954] lr: 3.0942e-05 eta: 0:42:00 time: 0.8001 data_time: 0.0247 memory: 15585 loss: 0.0004 +2023/06/04 21:09:22 - mmengine - INFO - Epoch(train) [9][1100/1954] lr: 2.9551e-05 eta: 0:40:33 time: 0.9163 data_time: 0.0022 memory: 15585 loss: 0.0003 +2023/06/04 21:10:49 - mmengine - INFO - Epoch(train) [9][1200/1954] lr: 2.8205e-05 eta: 0:39:07 time: 0.9418 data_time: 0.0019 memory: 15585 loss: 0.0014 +2023/06/04 21:12:12 - mmengine - INFO - Epoch(train) [9][1300/1954] lr: 2.6905e-05 eta: 0:37:39 time: 0.8361 data_time: 0.0017 memory: 15585 loss: 0.0013 +2023/06/04 21:13:08 - mmengine - INFO - Exp name: convnext_small_4xb256_sdv1_1m_lr4e-4_aug_1e-1_20230604_170206 +2023/06/04 21:13:34 - mmengine - INFO - Epoch(train) [9][1400/1954] lr: 2.5652e-05 eta: 0:36:12 time: 0.8759 data_time: 0.0018 memory: 15585 loss: 0.0027 +2023/06/04 21:14:57 - mmengine - INFO - Epoch(train) [9][1500/1954] lr: 2.4444e-05 eta: 0:34:45 time: 0.8527 data_time: 0.0017 memory: 15585 loss: 0.0007 +2023/06/04 21:16:23 - mmengine - INFO - Epoch(train) [9][1600/1954] lr: 2.3283e-05 eta: 0:33:18 time: 0.8638 data_time: 0.0020 memory: 15585 loss: 0.0000 +2023/06/04 21:17:46 - mmengine - INFO - Epoch(train) [9][1700/1954] lr: 2.2170e-05 eta: 0:31:51 time: 0.8282 data_time: 0.0017 memory: 15585 loss: 0.0012 +2023/06/04 21:19:13 - mmengine - INFO - Epoch(train) [9][1800/1954] lr: 2.1103e-05 eta: 0:30:25 time: 0.9069 data_time: 0.0017 memory: 15585 loss: 0.0033 +2023/06/04 21:20:38 - mmengine - INFO - Epoch(train) [9][1900/1954] lr: 2.0084e-05 eta: 0:28:58 time: 0.8268 data_time: 0.0017 memory: 15585 loss: 0.0002 +2023/06/04 21:21:25 - mmengine - INFO - Exp name: convnext_small_4xb256_sdv1_1m_lr4e-4_aug_1e-1_20230604_170206 +2023/06/04 21:21:25 - mmengine - INFO - Saving checkpoint at 9 epochs +2023/06/04 21:22:01 - mmengine - INFO - Epoch(val) [9][29/29] accuracy/top1: 99.9478 single-label/precision_classwise: [99.91169738769531, 99.99224090576172] single-label/recall_classwise: [99.99369049072266, 99.89147186279297] single-label/f1-score_classwise: [99.95266723632812, 99.94183349609375] data_time: 0.2645 time: 0.6500 +2023/06/04 21:23:31 - mmengine - INFO - Epoch(train) [10][ 100/1954] lr: 1.8608e-05 eta: 0:26:45 time: 0.9019 data_time: 0.2041 memory: 15585 loss: 0.0012 +2023/06/04 21:24:57 - mmengine - INFO - Epoch(train) [10][ 200/1954] lr: 1.7711e-05 eta: 0:25:18 time: 0.8680 data_time: 0.0016 memory: 15585 loss: 0.0020 +2023/06/04 21:26:22 - mmengine - INFO - Epoch(train) [10][ 300/1954] lr: 1.6863e-05 eta: 0:23:51 time: 0.8528 data_time: 0.0017 memory: 15585 loss: 0.0015 +2023/06/04 21:27:50 - mmengine - INFO - Epoch(train) [10][ 400/1954] lr: 1.6063e-05 eta: 0:22:25 time: 0.8909 data_time: 0.0016 memory: 15585 loss: 0.0014 +2023/06/04 21:28:00 - mmengine - INFO - Exp name: convnext_small_4xb256_sdv1_1m_lr4e-4_aug_1e-1_20230604_170206 +2023/06/04 21:29:17 - mmengine - INFO - Epoch(train) [10][ 500/1954] lr: 1.5311e-05 eta: 0:20:58 time: 0.8886 data_time: 0.0016 memory: 15585 loss: 0.0003 +2023/06/04 21:30:43 - mmengine - INFO - Epoch(train) [10][ 600/1954] lr: 1.4609e-05 eta: 0:19:32 time: 0.8753 data_time: 0.0015 memory: 15585 loss: 0.0002 +2023/06/04 21:32:06 - mmengine - INFO - Epoch(train) [10][ 700/1954] lr: 1.3956e-05 eta: 0:18:05 time: 0.8747 data_time: 0.0019 memory: 15585 loss: 0.0004 +2023/06/04 21:33:30 - mmengine - INFO - Epoch(train) [10][ 800/1954] lr: 1.3353e-05 eta: 0:16:38 time: 0.8568 data_time: 0.0016 memory: 15585 loss: 0.0007 +2023/06/04 21:34:58 - mmengine - INFO - Epoch(train) [10][ 900/1954] lr: 1.2798e-05 eta: 0:15:12 time: 0.8618 data_time: 0.0018 memory: 15585 loss: 0.0006 +2023/06/04 21:36:22 - mmengine - INFO - Epoch(train) [10][1000/1954] lr: 1.2294e-05 eta: 0:13:45 time: 0.8307 data_time: 0.0015 memory: 15585 loss: 0.0028 +2023/06/04 21:37:48 - mmengine - INFO - Epoch(train) [10][1100/1954] lr: 1.1840e-05 eta: 0:12:19 time: 0.8169 data_time: 0.0016 memory: 15585 loss: 0.0003 +2023/06/04 21:39:11 - mmengine - INFO - Epoch(train) [10][1200/1954] lr: 1.1435e-05 eta: 0:10:52 time: 0.8205 data_time: 0.0018 memory: 15585 loss: 0.0001 +2023/06/04 21:40:36 - mmengine - INFO - Epoch(train) [10][1300/1954] lr: 1.1080e-05 eta: 0:09:25 time: 0.8638 data_time: 0.0016 memory: 15585 loss: 0.0003 +2023/06/04 21:41:59 - mmengine - INFO - Epoch(train) [10][1400/1954] lr: 1.0776e-05 eta: 0:07:59 time: 0.8085 data_time: 0.0016 memory: 15585 loss: 0.0006 +2023/06/04 21:42:09 - mmengine - INFO - Exp name: convnext_small_4xb256_sdv1_1m_lr4e-4_aug_1e-1_20230604_170206 +2023/06/04 21:43:24 - mmengine - INFO - Epoch(train) [10][1500/1954] lr: 1.0522e-05 eta: 0:06:32 time: 0.8373 data_time: 0.0018 memory: 15585 loss: 0.0005 +2023/06/04 21:44:51 - mmengine - INFO - Epoch(train) [10][1600/1954] lr: 1.0318e-05 eta: 0:05:06 time: 0.8920 data_time: 0.0016 memory: 15585 loss: 0.0011 +2023/06/04 21:46:16 - mmengine - INFO - Epoch(train) [10][1700/1954] lr: 1.0164e-05 eta: 0:03:39 time: 0.9105 data_time: 0.0017 memory: 15585 loss: 0.0002 +2023/06/04 21:47:42 - mmengine - INFO - Epoch(train) [10][1800/1954] lr: 1.0061e-05 eta: 0:02:13 time: 0.8588 data_time: 0.0016 memory: 15585 loss: 0.0003 +2023/06/04 21:49:09 - mmengine - INFO - Epoch(train) [10][1900/1954] lr: 1.0008e-05 eta: 0:00:46 time: 0.8324 data_time: 0.0019 memory: 15585 loss: 0.0003 +2023/06/04 21:49:53 - mmengine - INFO - Exp name: convnext_small_4xb256_sdv1_1m_lr4e-4_aug_1e-1_20230604_170206 +2023/06/04 21:49:53 - mmengine - INFO - Saving checkpoint at 10 epochs +2023/06/04 21:50:29 - mmengine - INFO - Epoch(val) [10][29/29] accuracy/top1: 99.9478 single-label/precision_classwise: [99.9242935180664, 99.97673034667969] single-label/recall_classwise: [99.98106384277344, 99.90697479248047] single-label/f1-score_classwise: [99.95266723632812, 99.94184875488281] data_time: 0.2430 time: 0.6269 diff --git a/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_1e-1/20230604_170206/vis_data/20230604_170206.json b/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_1e-1/20230604_170206/vis_data/20230604_170206.json new file mode 100644 index 0000000000000000000000000000000000000000..f582b81fd6d5c4c25f6d53ef4717195f47703e37 --- /dev/null +++ b/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_1e-1/20230604_170206/vis_data/20230604_170206.json @@ -0,0 +1,200 @@ +{"lr": 0.00039997529889041205, "data_time": 0.4074542999267578, "loss": 0.017248274432495235, "time": 1.0513962745666503, "epoch": 1, "memory": 15587, "step": 100} +{"lr": 0.0003999002014233684, "data_time": 0.30373153686523435, "loss": 0.010412958264350892, "time": 1.0495609760284423, "epoch": 1, "memory": 15587, "step": 200} +{"lr": 0.0003997747244903853, "data_time": 0.40909771919250487, "loss": 0.023654562700539828, "time": 1.0233510255813598, "epoch": 1, "memory": 15587, "step": 300} +{"lr": 0.00039959890052580966, "data_time": 0.14441344738006592, "loss": 0.021022671554237603, "time": 1.0230309247970581, "epoch": 1, "memory": 15587, "step": 400} +{"lr": 0.0003993727749781197, "data_time": 0.0018305540084838866, "loss": 0.010535456100478768, "time": 0.9646939039230347, "epoch": 1, "memory": 15587, "step": 500} +{"lr": 0.000399096406298175, "data_time": 0.0012839555740356446, "loss": 0.009106617327779532, "time": 0.930386734008789, "epoch": 1, "memory": 15587, "step": 600} +{"lr": 0.0003987698659241082, "data_time": 0.0014708280563354493, "loss": 0.007947701634839177, "time": 0.9239459276199341, "epoch": 1, "memory": 15587, "step": 700} +{"lr": 0.0003983932382628598, "data_time": 0.0013887882232666016, "loss": 0.013005655445158481, "time": 0.9333152294158935, "epoch": 1, "memory": 15587, "step": 800} +{"lr": 0.00039796662066835946, "data_time": 0.0014635324478149414, "loss": 0.0046491074375808236, "time": 0.9554619789123535, "epoch": 1, "memory": 15587, "step": 900} +{"lr": 0.00039749012341636, "data_time": 0.0016210317611694337, "loss": 0.014733085688203574, "time": 0.872484827041626, "epoch": 1, "memory": 15587, "step": 1000} +{"lr": 0.00039696386967593436, "data_time": 0.001524949073791504, "loss": 0.010879722936078907, "time": 0.8295624732971192, "epoch": 1, "memory": 15587, "step": 1100} +{"lr": 0.00039638799547763706, "data_time": 0.00156097412109375, "loss": 0.003423507371917367, "time": 0.8381370782852173, "epoch": 1, "memory": 15587, "step": 1200} +{"lr": 0.0003957626496783407, "data_time": 0.001627635955810547, "loss": 0.0065059004817157986, "time": 0.8229460716247559, "epoch": 1, "memory": 15587, "step": 1300} +{"lr": 0.0003950879939227595, "data_time": 0.0015116214752197265, "loss": 0.011681228922680021, "time": 0.8561073780059815, "epoch": 1, "memory": 15587, "step": 1400} +{"lr": 0.0003943642026016663, "data_time": 0.0016747474670410155, "loss": 0.009761184407398104, "time": 0.8287646055221558, "epoch": 1, "memory": 15587, "step": 1500} +{"lr": 0.0003935914628068133, "data_time": 0.001641058921813965, "loss": 0.01562610026448965, "time": 0.8656338930130005, "epoch": 1, "memory": 15587, "step": 1600} +{"lr": 0.00039276997428257205, "data_time": 0.001541733741760254, "loss": 0.004199318075552583, "time": 0.7898478507995605, "epoch": 1, "memory": 15587, "step": 1700} +{"lr": 0.0003918999493743001, "data_time": 0.0015683174133300781, "loss": 0.0064587795175611975, "time": 0.8095824718475342, "epoch": 1, "memory": 15587, "step": 1800} +{"lr": 0.00039098161297345475, "data_time": 0.0018651485443115234, "loss": 0.00619882894679904, "time": 0.9013674259185791, "epoch": 1, "memory": 15587, "step": 1900} +{"accuracy/top1": 90.46658325195312, "data_time": 0.2132631416978507, "time": 0.6838534618246144, "step": 1} +{"lr": 0.0003894734390071047, "data_time": 0.17534224987030028, "loss": 0.0029863059520721436, "time": 0.850521731376648, "epoch": 2, "memory": 15587, "step": 2054} +{"lr": 0.0003884334872841784, "data_time": 0.0022087574005126955, "loss": 0.006332764867693186, "time": 0.8440931558609008, "epoch": 2, "memory": 15585, "step": 2154} +{"lr": 0.00038734612010935703, "data_time": 0.0021940231323242187, "loss": 0.0032624615821987392, "time": 0.8123685121536255, "epoch": 2, "memory": 15585, "step": 2254} +{"lr": 0.00038621161855458035, "data_time": 0.0024696826934814454, "loss": 0.009495723713189363, "time": 0.8672922611236572, "epoch": 2, "memory": 15585, "step": 2354} +{"lr": 0.0003850302758754841, "data_time": 0.0016459226608276367, "loss": 0.005710620945319533, "time": 0.7940369129180909, "epoch": 2, "memory": 15585, "step": 2454} +{"lr": 0.00038380239743559646, "data_time": 0.0017111539840698243, "loss": 0.008501151436939835, "time": 0.8246843338012695, "epoch": 2, "memory": 15585, "step": 2554} +{"lr": 0.00038252830062740647, "data_time": 0.0015415906906127929, "loss": 0.0045797074213624, "time": 0.8230563640594483, "epoch": 2, "memory": 15585, "step": 2654} +{"lr": 0.00038120831479032175, "data_time": 0.0016323328018188477, "loss": 0.008235360216349363, "time": 0.8180981636047363, "epoch": 2, "memory": 15585, "step": 2754} +{"lr": 0.0003798427811255361, "data_time": 0.0016580343246459961, "loss": 0.008007403230294585, "time": 0.872388505935669, "epoch": 2, "memory": 15585, "step": 2854} +{"lr": 0.0003784320526078357, "data_time": 0.001970386505126953, "loss": 0.00711355684325099, "time": 0.8330523490905761, "epoch": 2, "memory": 15585, "step": 2954} +{"lr": 0.0003769764938943556, "data_time": 0.00246584415435791, "loss": 0.010007486958056689, "time": 0.8688191652297974, "epoch": 2, "memory": 15585, "step": 3054} +{"lr": 0.0003754764812303242, "data_time": 0.0023420572280883787, "loss": 0.0043166371062397955, "time": 0.849473237991333, "epoch": 2, "memory": 15585, "step": 3154} +{"lr": 0.0003739324023518048, "data_time": 0.002280259132385254, "loss": 0.0035292053129523993, "time": 0.9241914033889771, "epoch": 2, "memory": 15585, "step": 3254} +{"lr": 0.0003723446563854724, "data_time": 0.0024772405624389647, "loss": 0.0033644698560237885, "time": 0.8862639427185058, "epoch": 2, "memory": 15585, "step": 3354} +{"lr": 0.0003707136537454424, "data_time": 0.002196335792541504, "loss": 0.010240558814257384, "time": 0.8627858638763428, "epoch": 2, "memory": 15585, "step": 3454} +{"lr": 0.0003690398160271841, "data_time": 0.12454843521118164, "loss": 0.005815344536677003, "time": 0.823633337020874, "epoch": 2, "memory": 15585, "step": 3554} +{"lr": 0.00036732357589854426, "data_time": 0.076629376411438, "loss": 0.00993875446729362, "time": 0.8245397567749023, "epoch": 2, "memory": 15585, "step": 3654} +{"lr": 0.00036556537698790475, "data_time": 0.0016868829727172852, "loss": 0.009170089801773428, "time": 0.843467116355896, "epoch": 2, "memory": 15585, "step": 3754} +{"lr": 0.0003637656737695118, "data_time": 0.0016482353210449218, "loss": 0.003345610899850726, "time": 0.8315291404724121, "epoch": 2, "memory": 15585, "step": 3854} +{"accuracy/top1": 98.37166595458984, "data_time": 0.2504310051600138, "time": 0.6425877412160238, "step": 2} +{"lr": 0.00036091403361175644, "data_time": 0.1533118486404419, "loss": 0.0037349598947912454, "time": 0.8195040702819825, "epoch": 3, "memory": 15585, "step": 4008} +{"lr": 0.00035901102535335625, "data_time": 0.002349710464477539, "loss": 0.00564337195828557, "time": 0.9181422472000123, "epoch": 3, "memory": 15585, "step": 4108} +{"lr": 0.00035706820701190153, "data_time": 0.0022867679595947265, "loss": 0.00629440750926733, "time": 0.8384639978408813, "epoch": 3, "memory": 15585, "step": 4208} +{"lr": 0.00035508608078364215, "data_time": 0.02729918956756592, "loss": 0.004229048639535904, "time": 0.8654555320739746, "epoch": 3, "memory": 15585, "step": 4308} +{"lr": 0.00035306515902546755, "data_time": 0.16883337497711182, "loss": 0.0037390483543276788, "time": 0.8887279510498047, "epoch": 3, "memory": 15585, "step": 4408} +{"lr": 0.0003510059641224655, "data_time": 0.3455545902252197, "loss": 0.0037441214080899954, "time": 0.8658688306808472, "epoch": 3, "memory": 15585, "step": 4508} +{"lr": 0.00034890902835289306, "data_time": 0.2988360643386841, "loss": 0.005673976615071297, "time": 0.8163290739059448, "epoch": 3, "memory": 15585, "step": 4608} +{"lr": 0.00034677489375058753, "data_time": 0.0026842594146728516, "loss": 0.006808177474886179, "time": 0.8540040731430054, "epoch": 3, "memory": 15585, "step": 4708} +{"lr": 0.00034460411196485875, "data_time": 0.0023552656173706056, "loss": 0.007155083771795035, "time": 0.8360354661941528, "epoch": 3, "memory": 15585, "step": 4808} +{"lr": 0.0003423972441178926, "data_time": 0.002347731590270996, "loss": 0.010615699458867311, "time": 0.7946882963180542, "epoch": 3, "memory": 15585, "step": 4908} +{"lr": 0.0003401548606597075, "data_time": 0.0019759416580200197, "loss": 0.004952302295714616, "time": 0.8103042840957642, "epoch": 3, "memory": 15585, "step": 5008} +{"lr": 0.0003378775412207003, "data_time": 0.0015921354293823241, "loss": 0.007694493606686592, "time": 0.8359559535980224, "epoch": 3, "memory": 15585, "step": 5108} +{"lr": 0.0003355658744618178, "data_time": 0.0015553712844848632, "loss": 0.0053688473999500275, "time": 0.8929475784301758, "epoch": 3, "memory": 15585, "step": 5208} +{"lr": 0.0003332204579223944, "data_time": 0.002374744415283203, "loss": 0.006741747166961432, "time": 0.8350327014923096, "epoch": 3, "memory": 15585, "step": 5308} +{"lr": 0.0003308418978656968, "data_time": 0.0018707990646362304, "loss": 0.0036798327229917048, "time": 0.8931418657302856, "epoch": 3, "memory": 15585, "step": 5408} +{"lr": 0.0003284308091222099, "data_time": 0.001961803436279297, "loss": 0.005295302253216505, "time": 0.8708714008331299, "epoch": 3, "memory": 15585, "step": 5508} +{"lr": 0.0003259878149307108, "data_time": 0.0017261028289794922, "loss": 0.0031720946542918684, "time": 0.8444495916366577, "epoch": 3, "memory": 15585, "step": 5608} +{"lr": 0.0003235135467771688, "data_time": 0.0017846107482910156, "loss": 0.0033446666318923237, "time": 0.8666119575500488, "epoch": 3, "memory": 15585, "step": 5708} +{"lr": 0.0003210086442315141, "data_time": 0.0023134469985961912, "loss": 0.004942983482033014, "time": 0.8863066911697388, "epoch": 3, "memory": 15585, "step": 5808} +{"accuracy/top1": 99.45722198486328, "data_time": 0.24220484097798664, "time": 0.6292988936106364, "step": 3} +{"lr": 0.00031709267628953966, "data_time": 0.30004100799560546, "loss": 0.004815851850435138, "time": 0.8579655408859252, "epoch": 4, "memory": 15585, "step": 5962} +{"lr": 0.00031451289120303685, "data_time": 0.04376761913299561, "loss": 0.0059142212849110365, "time": 0.7941560268402099, "epoch": 4, "memory": 15585, "step": 6062} +{"lr": 0.000311904798290621, "data_time": 0.13209106922149658, "loss": 0.0015460836701095104, "time": 0.8303063392639161, "epoch": 4, "memory": 15585, "step": 6162} +{"lr": 0.00030926907171438685, "data_time": 0.036271977424621585, "loss": 0.005030934326350689, "time": 0.8559933662414551, "epoch": 4, "memory": 15585, "step": 6262} +{"lr": 0.000306606392779414, "data_time": 0.06773190498352051, "loss": 0.005842992011457682, "time": 0.8093076944351196, "epoch": 4, "memory": 15585, "step": 6362} +{"lr": 0.0003039174497576579, "data_time": 0.001483154296875, "loss": 0.00600668704137206, "time": 0.8325612545013428, "epoch": 4, "memory": 15585, "step": 6462} +{"lr": 0.0003012029377100387, "data_time": 0.0016164064407348632, "loss": 0.009730385523289442, "time": 0.78953857421875, "epoch": 4, "memory": 15585, "step": 6562} +{"lr": 0.00029846355830677674, "data_time": 0.2795774698257446, "loss": 0.006102950312197208, "time": 0.8676823377609253, "epoch": 4, "memory": 15585, "step": 6662} +{"lr": 0.0002957000196460183, "data_time": 0.3513307571411133, "loss": 0.003987257648259401, "time": 0.8667858600616455, "epoch": 4, "memory": 15585, "step": 6762} +{"lr": 0.0002929130360708003, "data_time": 0.3393963098526001, "loss": 0.008418860659003258, "time": 0.8544230699539185, "epoch": 4, "memory": 15585, "step": 6862} +{"lr": 0.0002901033279844016, "data_time": 0.16490800380706788, "loss": 0.006323257321491837, "time": 0.8619724750518799, "epoch": 4, "memory": 15585, "step": 6962} +{"lr": 0.00028727162166412623, "data_time": 0.0015342950820922852, "loss": 0.006643430422991514, "time": 0.8409230232238769, "epoch": 4, "memory": 15585, "step": 7062} +{"lr": 0.000284418649073569, "data_time": 0.001609468460083008, "loss": 0.004987420374527573, "time": 0.7618176698684692, "epoch": 4, "memory": 15585, "step": 7162} +{"lr": 0.00028154514767341203, "data_time": 0.0014576673507690429, "loss": 0.005437305709347129, "time": 0.7865698337554932, "epoch": 4, "memory": 15585, "step": 7262} +{"lr": 0.0002786518602307988, "data_time": 0.001808929443359375, "loss": 0.004921631002798676, "time": 0.8115873098373413, "epoch": 4, "memory": 15585, "step": 7362} +{"lr": 0.0002757395346273384, "data_time": 0.0015999794006347657, "loss": 0.00435516182333231, "time": 0.7899288892745971, "epoch": 4, "memory": 15585, "step": 7462} +{"lr": 0.0002728089236657853, "data_time": 0.10272691249847413, "loss": 0.009449858544394374, "time": 0.80626060962677, "epoch": 4, "memory": 15585, "step": 7562} +{"lr": 0.0002698607848754497, "data_time": 0.21255826950073242, "loss": 0.005273900879547, "time": 0.9091211080551147, "epoch": 4, "memory": 15585, "step": 7662} +{"lr": 0.0002668958803163847, "data_time": 0.28281707763671876, "loss": 0.004004256753250957, "time": 0.7986958503723145, "epoch": 4, "memory": 15585, "step": 7762} +{"accuracy/top1": 99.7773208618164, "data_time": 0.26677437623341876, "time": 0.6510602235794067, "step": 4} +{"lr": 0.00026229890681650394, "data_time": 0.2097378969192505, "loss": 0.0011739041656255722, "time": 0.8457151174545288, "epoch": 5, "memory": 15585, "step": 7916} +{"lr": 0.00025929487224058475, "data_time": 0.024641084671020507, "loss": 0.003347337245941162, "time": 0.8889694213867188, "epoch": 5, "memory": 15585, "step": 8016} +{"lr": 0.0002562768030631272, "data_time": 0.0017786502838134765, "loss": 0.0018761730752885342, "time": 0.8516000747680664, "epoch": 5, "memory": 15585, "step": 8116} +{"lr": 0.0002532454794203871, "data_time": 0.0019945383071899416, "loss": 0.006383917247876525, "time": 0.8578239202499389, "epoch": 5, "memory": 15585, "step": 8216} +{"lr": 0.0002502016848747476, "data_time": 0.001480436325073242, "loss": 0.003877262445166707, "time": 0.9104105472564697, "epoch": 5, "memory": 15585, "step": 8316} +{"lr": 0.0002471462062121758, "data_time": 0.0015180349349975587, "loss": 0.0016422943212091924, "time": 0.8483364820480347, "epoch": 5, "memory": 15585, "step": 8416} +{"lr": 0.00024407983323884952, "data_time": 0.0017547130584716797, "loss": 0.003953368682414293, "time": 0.8449782848358154, "epoch": 5, "memory": 15585, "step": 8516} +{"lr": 0.0002410033585770006, "data_time": 0.0016140937805175781, "loss": 0.0019638130441308023, "time": 0.8575214385986328, "epoch": 5, "memory": 15585, "step": 8616} +{"lr": 0.00023791757746003292, "data_time": 0.0015656232833862304, "loss": 0.008501792373135687, "time": 0.9225797414779663, "epoch": 5, "memory": 15585, "step": 8716} +{"lr": 0.00023482328752696182, "data_time": 0.0017169713973999023, "loss": 0.007375417789444327, "time": 0.8549912691116333, "epoch": 5, "memory": 15585, "step": 8816} +{"lr": 0.00023172128861623345, "data_time": 0.00161895751953125, "loss": 0.0027543580159544944, "time": 0.8735415220260621, "epoch": 5, "memory": 15585, "step": 8916} +{"lr": 0.000228612382558976, "data_time": 0.001691722869873047, "loss": 0.005506934691220522, "time": 0.8721412420272827, "epoch": 5, "memory": 15585, "step": 9016} +{"lr": 0.00022549737297173692, "data_time": 0.0018543481826782226, "loss": 0.0029317505192011594, "time": 0.9107064008712769, "epoch": 5, "memory": 15585, "step": 9116} +{"lr": 0.00022237706504875477, "data_time": 0.0015569210052490234, "loss": 0.003759072907269001, "time": 0.8398517370223999, "epoch": 5, "memory": 15585, "step": 9216} +{"lr": 0.00021925226535382818, "data_time": 0.0015794992446899413, "loss": 0.0014174559153616428, "time": 0.9286900997161865, "epoch": 5, "memory": 15585, "step": 9316} +{"lr": 0.0002161237816118266, "data_time": 0.0015759944915771484, "loss": 0.0025255458895117044, "time": 0.8815348863601684, "epoch": 5, "memory": 15585, "step": 9416} +{"lr": 0.00021299242249990383, "data_time": 0.001840829849243164, "loss": 0.003546580672264099, "time": 0.8672035217285157, "epoch": 5, "memory": 15585, "step": 9516} +{"lr": 0.00020985899743846363, "data_time": 0.0017792940139770507, "loss": 0.003579519363120198, "time": 0.8193649530410767, "epoch": 5, "memory": 15585, "step": 9616} +{"lr": 0.00020672431638193363, "data_time": 0.0017078876495361327, "loss": 0.0037071458995342256, "time": 0.8633199453353881, "epoch": 5, "memory": 15585, "step": 9716} +{"accuracy/top1": 99.84342956542969, "data_time": 0.23723048369089764, "time": 0.6239738384882609, "step": 5} +{"lr": 0.00020189632111931318, "data_time": 0.3125102996826172, "loss": 0.002428907435387373, "time": 0.8290535688400269, "epoch": 6, "memory": 15585, "step": 9870} +{"lr": 0.00019876209288860824, "data_time": 0.30610611438751223, "loss": 0.00241771056316793, "time": 0.8245126485824585, "epoch": 6, "memory": 15585, "step": 9970} +{"lr": 0.00019562947708532286, "data_time": 0.29919991493225095, "loss": 0.00111872018314898, "time": 0.8460508823394776, "epoch": 6, "memory": 15585, "step": 10070} +{"lr": 0.00019249928345470104, "data_time": 0.2790926456451416, "loss": 0.0014760125428438186, "time": 0.8309964656829834, "epoch": 6, "memory": 15585, "step": 10170} +{"lr": 0.00018937232111588298, "data_time": 0.0015459060668945312, "loss": 0.005606566648930311, "time": 0.9613274812698365, "epoch": 6, "memory": 15585, "step": 10270} +{"lr": 0.000186249398352757, "data_time": 0.001607060432434082, "loss": 0.0035888387355953455, "time": 0.9406477689743042, "epoch": 6, "memory": 15585, "step": 10370} +{"lr": 0.0001831313224050277, "data_time": 0.0019884109497070312, "loss": 0.005360460001975298, "time": 0.8657693147659302, "epoch": 6, "memory": 15585, "step": 10470} +{"lr": 0.00018001889925955291, "data_time": 0.001595926284790039, "loss": 0.004190133512020111, "time": 0.8827555894851684, "epoch": 6, "memory": 15585, "step": 10570} +{"lr": 0.00017691293344200646, "data_time": 0.0016616582870483398, "loss": 0.0020628038328140974, "time": 0.9320860147476197, "epoch": 6, "memory": 15585, "step": 10670} +{"lr": 0.00017381422780891689, "data_time": 0.0015587806701660156, "loss": 0.002030350873246789, "time": 0.909303092956543, "epoch": 6, "memory": 15585, "step": 10770} +{"lr": 0.00017072358334013806, "data_time": 0.0015881061553955078, "loss": 0.0020942374598234893, "time": 0.8919180393218994, "epoch": 6, "memory": 15585, "step": 10870} +{"lr": 0.00016764179893180575, "data_time": 0.002107691764831543, "loss": 0.0032632900401949883, "time": 0.8498372077941895, "epoch": 6, "memory": 15585, "step": 10970} +{"lr": 0.00016456967118983176, "data_time": 0.001712965965270996, "loss": 0.0008120507467538118, "time": 0.8277901411056519, "epoch": 6, "memory": 15585, "step": 11070} +{"lr": 0.00016150799422398984, "data_time": 0.0016379594802856446, "loss": 0.003254414862021804, "time": 0.8336925745010376, "epoch": 6, "memory": 15585, "step": 11170} +{"lr": 0.0001584575594426489, "data_time": 0.0017516136169433594, "loss": 0.0008459640201181173, "time": 0.8705430507659913, "epoch": 6, "memory": 15585, "step": 11270} +{"lr": 0.00015541915534820135, "data_time": 0.0015369653701782227, "loss": 0.0025393597315996885, "time": 0.8745918273925781, "epoch": 6, "memory": 15585, "step": 11370} +{"lr": 0.00015239356733324542, "data_time": 0.0015781164169311524, "loss": 0.002424947964027524, "time": 0.8247071504592896, "epoch": 6, "memory": 15585, "step": 11470} +{"lr": 0.00014938157747756997, "data_time": 0.001655721664428711, "loss": 0.0021597898565232755, "time": 0.8379346847534179, "epoch": 6, "memory": 15585, "step": 11570} +{"lr": 0.00014638396434599565, "data_time": 0.0015805721282958984, "loss": 0.003143156925216317, "time": 0.8156458377838135, "epoch": 6, "memory": 15585, "step": 11670} +{"accuracy/top1": 99.87474822998047, "data_time": 0.24801313877105713, "time": 0.6315874497095744, "step": 6} +{"lr": 0.00014179754513556176, "data_time": 0.3228867769241333, "loss": 0.0003779163584113121, "time": 0.838274073600769, "epoch": 7, "memory": 15585, "step": 11824} +{"lr": 0.00013883992334674444, "data_time": 0.0016718864440917968, "loss": 0.002053354913368821, "time": 0.8226800203323364, "epoch": 7, "memory": 15585, "step": 11924} +{"lr": 0.00013589940317867467, "data_time": 0.0015621185302734375, "loss": 0.003381268586963415, "time": 0.8198431015014649, "epoch": 7, "memory": 15585, "step": 12024} +{"lr": 0.0001329767447220783, "data_time": 0.0016316890716552735, "loss": 0.0016960741020739078, "time": 0.8422715425491333, "epoch": 7, "memory": 15585, "step": 12124} +{"lr": 0.00013007270345063488, "data_time": 0.0018516063690185546, "loss": 0.0038290629163384438, "time": 0.8356634378433228, "epoch": 7, "memory": 15585, "step": 12224} +{"lr": 0.00012718803002569412, "data_time": 0.0019460439682006836, "loss": 0.0010147742927074433, "time": 0.8165646314620971, "epoch": 7, "memory": 15585, "step": 12324} +{"lr": 0.0001243234701022397, "data_time": 0.0016362905502319337, "loss": 0.0013078055344522, "time": 0.8601309537887574, "epoch": 7, "memory": 15585, "step": 12424} +{"lr": 0.00012147976413614597, "data_time": 0.0015713691711425782, "loss": 0.0014702303800731898, "time": 0.8406438589096069, "epoch": 7, "memory": 15585, "step": 12524} +{"lr": 0.0001186576471927784, "data_time": 0.0017952442169189454, "loss": 0.0038851252757012844, "time": 0.8938199043273926, "epoch": 7, "memory": 15585, "step": 12624} +{"lr": 0.00011585784875698867, "data_time": 0.0016071557998657226, "loss": 0.0028708187863230707, "time": 0.8393218278884887, "epoch": 7, "memory": 15585, "step": 12724} +{"lr": 0.00011308109254454993, "data_time": 0.0016457319259643554, "loss": 0.002554943924769759, "time": 0.8666845083236694, "epoch": 7, "memory": 15585, "step": 12824} +{"lr": 0.00011032809631508466, "data_time": 0.0015531063079833984, "loss": 0.0005809028167277575, "time": 0.8390853643417359, "epoch": 7, "memory": 15585, "step": 12924} +{"lr": 0.00010759957168653253, "data_time": 0.0018207550048828125, "loss": 0.0018609888385981322, "time": 0.9048506259918213, "epoch": 7, "memory": 15585, "step": 13024} +{"lr": 0.0001048962239512054, "data_time": 0.0019021749496459961, "loss": 0.0022601188626140355, "time": 0.819468641281128, "epoch": 7, "memory": 15585, "step": 13124} +{"lr": 0.00010221875189347644, "data_time": 0.001586747169494629, "loss": 0.004573145974427461, "time": 0.847865629196167, "epoch": 7, "memory": 15585, "step": 13224} +{"lr": 9.95678476091537e-05, "data_time": 0.0014557838439941406, "loss": 0.0008064919617027045, "time": 0.8420084714889526, "epoch": 7, "memory": 15585, "step": 13324} +{"lr": 9.694419632657999e-05, "data_time": 0.0020059823989868166, "loss": 0.0009036433417350054, "time": 0.8855749130249023, "epoch": 7, "memory": 15585, "step": 13424} +{"lr": 9.434847622951059e-05, "data_time": 0.002084207534790039, "loss": 0.000638015754520893, "time": 0.8360074758529663, "epoch": 7, "memory": 15585, "step": 13524} +{"lr": 9.178135828180905e-05, "data_time": 0.0018365859985351562, "loss": 0.004058591555804014, "time": 0.8574733018875123, "epoch": 7, "memory": 15585, "step": 13624} +{"accuracy/top1": 99.89214324951172, "data_time": 0.24434186617533366, "time": 0.6280344963073731, "step": 7} +{"lr": 8.788546579134055e-05, "data_time": 0.001804018020629883, "loss": 0.0016758719924837352, "time": 0.8845922231674195, "epoch": 8, "memory": 15585, "step": 13778} +{"lr": 8.539396307206318e-05, "data_time": 0.0015428543090820312, "loss": 0.0024170370772480963, "time": 0.8424384832382202, "epoch": 8, "memory": 15585, "step": 13878} +{"lr": 8.293337714106816e-05, "data_time": 0.001485586166381836, "loss": 0.00015728422440588474, "time": 0.8494835138320923, "epoch": 8, "memory": 15585, "step": 13978} +{"lr": 8.05043440315878e-05, "data_time": 0.0015645027160644531, "loss": 0.0004042779561132193, "time": 0.8392947673797607, "epoch": 8, "memory": 15585, "step": 14078} +{"lr": 7.810749162081216e-05, "data_time": 0.0016747236251831055, "loss": 0.0004208637401461601, "time": 0.8204168558120728, "epoch": 8, "memory": 15585, "step": 14178} +{"lr": 7.574343946758993e-05, "data_time": 0.017456459999084472, "loss": 0.0002836901228874922, "time": 0.8490535974502563, "epoch": 8, "memory": 15585, "step": 14278} +{"lr": 7.34127986522793e-05, "data_time": 0.08441815376281739, "loss": 0.0018746765796095132, "time": 0.7804399013519288, "epoch": 8, "memory": 15585, "step": 14378} +{"lr": 7.111617161879136e-05, "data_time": 0.11728715896606445, "loss": 0.0003813054412603378, "time": 0.884506344795227, "epoch": 8, "memory": 15585, "step": 14478} +{"lr": 6.885415201886485e-05, "data_time": 0.33581278324127195, "loss": 0.0009060120210051537, "time": 0.8761226177215576, "epoch": 8, "memory": 15585, "step": 14578} +{"lr": 6.662732455861337e-05, "data_time": 0.5895336389541626, "loss": 0.0013958164490759373, "time": 1.1073842525482178, "epoch": 8, "memory": 15585, "step": 14678} +{"lr": 6.443626484738661e-05, "data_time": 0.2097240924835205, "loss": 0.000790273305028677, "time": 0.726179552078247, "epoch": 8, "memory": 15585, "step": 14778} +{"lr": 6.22815392489815e-05, "data_time": 0.2031512975692749, "loss": 0.0011591968592256307, "time": 0.7829782724380493, "epoch": 8, "memory": 15585, "step": 14878} +{"lr": 6.016370473524328e-05, "data_time": 0.0015236854553222657, "loss": 0.0017688107211142779, "time": 0.7770345687866211, "epoch": 8, "memory": 15585, "step": 14978} +{"lr": 5.80833087420956e-05, "data_time": 0.03259367942810058, "loss": 0.002623250987380743, "time": 0.8210611343383789, "epoch": 8, "memory": 15585, "step": 15078} +{"lr": 5.604088902803397e-05, "data_time": 0.10467522144317627, "loss": 0.00075977579690516, "time": 0.8388400077819824, "epoch": 8, "memory": 15585, "step": 15178} +{"lr": 5.403697353512148e-05, "data_time": 0.1727466106414795, "loss": 0.0008391457144170999, "time": 0.8542494058609009, "epoch": 8, "memory": 15585, "step": 15278} +{"lr": 5.207208025252174e-05, "data_time": 0.3552724838256836, "loss": 0.001340480847284198, "time": 0.8716519355773926, "epoch": 8, "memory": 15585, "step": 15378} +{"lr": 5.014671708260458e-05, "data_time": 0.35734920501708983, "loss": 0.0007238555233925581, "time": 0.8721176624298096, "epoch": 8, "memory": 15585, "step": 15478} +{"lr": 4.826138170965952e-05, "data_time": 0.3547613859176636, "loss": 0.006013083131983876, "time": 0.8718185186386108, "epoch": 8, "memory": 15585, "step": 15578} +{"accuracy/top1": 99.92345428466797, "data_time": 0.2632083098093669, "time": 0.6461691697438557, "step": 8} +{"lr": 4.5437373040402475e-05, "data_time": 0.12021346092224121, "loss": 0.0011159366462379695, "time": 0.8541103363037109, "epoch": 9, "memory": 15585, "step": 15732} +{"lr": 4.3655875036148156e-05, "data_time": 0.0016057968139648437, "loss": 0.0005931212566792965, "time": 0.7904397964477539, "epoch": 9, "memory": 15585, "step": 15832} +{"lr": 4.19160826370641e-05, "data_time": 0.0015247821807861327, "loss": 0.0005626421887427568, "time": 0.8504166841506958, "epoch": 9, "memory": 15585, "step": 15932} +{"lr": 4.021844555952455e-05, "data_time": 0.0016149759292602539, "loss": 0.0007392247207462788, "time": 0.8377806425094605, "epoch": 9, "memory": 15585, "step": 16032} +{"lr": 3.8563402623236864e-05, "data_time": 0.0017754316329956054, "loss": 0.0013027803506702185, "time": 0.829702639579773, "epoch": 9, "memory": 15585, "step": 16132} +{"lr": 3.695138163781135e-05, "data_time": 0.001830744743347168, "loss": 0.004492873838171363, "time": 0.8878957986831665, "epoch": 9, "memory": 15585, "step": 16232} +{"lr": 3.538279929217756e-05, "data_time": 0.0014868974685668945, "loss": 0.0006998658180236816, "time": 0.8603259801864624, "epoch": 9, "memory": 15585, "step": 16332} +{"lr": 3.385806104687531e-05, "data_time": 0.0017010688781738282, "loss": 0.004667029203847051, "time": 0.9070610284805298, "epoch": 9, "memory": 15585, "step": 16432} +{"lr": 3.237756102924734e-05, "data_time": 0.0843231201171875, "loss": 0.0008170569315552711, "time": 0.849350380897522, "epoch": 9, "memory": 15585, "step": 16532} +{"lr": 3.094168193156213e-05, "data_time": 0.024666380882263184, "loss": 0.0003779273945838213, "time": 0.8001426696777344, "epoch": 9, "memory": 15585, "step": 16632} +{"lr": 2.95507949120924e-05, "data_time": 0.0021928787231445313, "loss": 0.0003065277356654406, "time": 0.9163157939910889, "epoch": 9, "memory": 15585, "step": 16732} +{"lr": 2.8205259499174602e-05, "data_time": 0.001944732666015625, "loss": 0.0014087852090597154, "time": 0.941752028465271, "epoch": 9, "memory": 15585, "step": 16832} +{"lr": 2.690542349827549e-05, "data_time": 0.0017181396484375, "loss": 0.001308317668735981, "time": 0.83608078956604, "epoch": 9, "memory": 15585, "step": 16932} +{"lr": 2.565162290208795e-05, "data_time": 0.0017790794372558594, "loss": 0.0026592658367007973, "time": 0.8758543491363525, "epoch": 9, "memory": 15585, "step": 17032} +{"lr": 2.4444181803680975e-05, "data_time": 0.0016588926315307616, "loss": 0.0006535890977829694, "time": 0.8526934385299683, "epoch": 9, "memory": 15585, "step": 17132} +{"lr": 2.32834123127253e-05, "data_time": 0.0020235300064086912, "loss": 2.9822345823049545e-05, "time": 0.8637571334838867, "epoch": 9, "memory": 15585, "step": 17232} +{"lr": 2.2169614474816763e-05, "data_time": 0.0017161130905151366, "loss": 0.001203312911093235, "time": 0.8281512975692749, "epoch": 9, "memory": 15585, "step": 17332} +{"lr": 2.110307619391749e-05, "data_time": 0.0017165899276733398, "loss": 0.0033061815425753593, "time": 0.9068790197372436, "epoch": 9, "memory": 15585, "step": 17432} +{"lr": 2.008407315793661e-05, "data_time": 0.0017263174057006836, "loss": 0.00022503635846078397, "time": 0.8267875194549561, "epoch": 9, "memory": 15585, "step": 17532} +{"accuracy/top1": 99.94781494140625, "data_time": 0.2644737482070923, "time": 0.6500085671742757, "step": 9} +{"lr": 1.8608381954463387e-05, "data_time": 0.20412657260894776, "loss": 0.0011625594459474086, "time": 0.9019122123718262, "epoch": 10, "memory": 15585, "step": 17686} +{"lr": 1.771127410231484e-05, "data_time": 0.0016240119934082032, "loss": 0.0019862732850015165, "time": 0.8680342197418213, "epoch": 10, "memory": 15585, "step": 17786} +{"lr": 1.6862578237010542e-05, "data_time": 0.001675558090209961, "loss": 0.001506176684051752, "time": 0.8527591228485107, "epoch": 10, "memory": 15585, "step": 17886} +{"lr": 1.6062513736694383e-05, "data_time": 0.0016326665878295898, "loss": 0.0014381874352693558, "time": 0.8908598184585571, "epoch": 10, "memory": 15585, "step": 17986} +{"lr": 1.531128740886036e-05, "data_time": 0.0015998125076293946, "loss": 0.00033911271020770073, "time": 0.8886195898056031, "epoch": 10, "memory": 15585, "step": 18086} +{"lr": 1.4609093436895328e-05, "data_time": 0.0015250444412231445, "loss": 0.0001683767419308424, "time": 0.8753059148788452, "epoch": 10, "memory": 15585, "step": 18186} +{"lr": 1.3956113329884566e-05, "data_time": 0.0019235849380493165, "loss": 0.00039821406826376913, "time": 0.8746515035629272, "epoch": 10, "memory": 15585, "step": 18286} +{"lr": 1.335251587569401e-05, "data_time": 0.0015807151794433594, "loss": 0.0007395950146019459, "time": 0.8567647933959961, "epoch": 10, "memory": 15585, "step": 18386} +{"lr": 1.2798457097340337e-05, "data_time": 0.0018355131149291992, "loss": 0.0005796691868454218, "time": 0.861804461479187, "epoch": 10, "memory": 15585, "step": 18486} +{"lr": 1.229408021266094e-05, "data_time": 0.001537156105041504, "loss": 0.002770001208409667, "time": 0.830748462677002, "epoch": 10, "memory": 15585, "step": 18586} +{"lr": 1.1839515597293723e-05, "data_time": 0.001573944091796875, "loss": 0.0003068405669182539, "time": 0.8169232606887817, "epoch": 10, "memory": 15585, "step": 18686} +{"lr": 1.1434880750976406e-05, "data_time": 0.0017671585083007812, "loss": 6.0982629656791686e-05, "time": 0.8205039978027344, "epoch": 10, "memory": 15585, "step": 18786} +{"lr": 1.108028026717431e-05, "data_time": 0.0015892982482910156, "loss": 0.00025234329514205455, "time": 0.8637696266174316, "epoch": 10, "memory": 15585, "step": 18886} +{"lr": 1.0775805806044138e-05, "data_time": 0.0015877008438110352, "loss": 0.0005985628813505173, "time": 0.8084832191467285, "epoch": 10, "memory": 15585, "step": 18986} +{"lr": 1.0521536070740776e-05, "data_time": 0.0017740964889526368, "loss": 0.0004730156622827053, "time": 0.8373129367828369, "epoch": 10, "memory": 15585, "step": 19086} +{"lr": 1.03175367870735e-05, "data_time": 0.0015578746795654296, "loss": 0.0011395436245948075, "time": 0.8919585704803467, "epoch": 10, "memory": 15585, "step": 19186} +{"lr": 1.0163860686516628e-05, "data_time": 0.0016566753387451173, "loss": 0.00020440882071852683, "time": 0.9104677200317383, "epoch": 10, "memory": 15585, "step": 19286} +{"lr": 1.0060547492578997e-05, "data_time": 0.0015944719314575195, "loss": 0.0002673800103366375, "time": 0.8587796926498413, "epoch": 10, "memory": 15585, "step": 19386} +{"lr": 1.0007623910535892e-05, "data_time": 0.0019279956817626954, "loss": 0.0002512660808861256, "time": 0.8323871612548828, "epoch": 10, "memory": 15585, "step": 19486} +{"accuracy/top1": 99.94781494140625, "data_time": 0.24300116697947186, "time": 0.6269446690877278, "step": 10} diff --git a/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_1e-1/20230604_170206/vis_data/config.py b/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_1e-1/20230604_170206/vis_data/config.py new file mode 100644 index 0000000000000000000000000000000000000000..77eb3de7b15a1fc93b06c70f847eec37b0754a05 --- /dev/null +++ b/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_1e-1/20230604_170206/vis_data/config.py @@ -0,0 +1,247 @@ +optim_wrapper = dict( + optimizer=dict( + type='AdamW', + lr=0.0004, + weight_decay=0.05, + eps=1e-08, + betas=(0.9, 0.999), + _scope_='mmpretrain'), + paramwise_cfg=dict( + norm_decay_mult=0.0, + bias_decay_mult=0.0, + flat_decay_mult=0.0, + custom_keys=dict({ + '.absolute_pos_embed': dict(decay_mult=0.0), + '.relative_position_bias_table': dict(decay_mult=0.0) + })), + type='AmpOptimWrapper', + dtype='bfloat16', + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=False, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=4096) +model = dict( + type='ImageClassifier', + backbone=dict( + frozen_stages=2, + type='ConvNeXt', + arch='small', + drop_path_rate=0.4, + init_cfg=dict( + type='Pretrained', + checkpoint= + 'https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth', + prefix='backbone')), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=768, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + init_cfg=None), + init_cfg=dict( + type='TruncNormal', layer=['Conv2d', 'Linear'], std=0.02, bias=0.0), + train_cfg=None) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stablediffusionV1-5R2-dpmsolver-25-1m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/cc1m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_1e-1' diff --git a/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_1e-1/20230604_170206/vis_data/events.out.tfevents.1685869341.SH-IDC1-10-140-24-128.42884.0 b/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_1e-1/20230604_170206/vis_data/events.out.tfevents.1685869341.SH-IDC1-10-140-24-128.42884.0 new file mode 100644 index 0000000000000000000000000000000000000000..2617d275eddf550cf195e6c84b10089bfb1e874a --- /dev/null +++ b/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_1e-1/20230604_170206/vis_data/events.out.tfevents.1685869341.SH-IDC1-10-140-24-128.42884.0 @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:036920b8b6e27d8b3db94a0585c5bc969ff507ed31bd15879fac4114fc9d2a90 +size 1503481 diff --git a/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_1e-1/20230604_170206/vis_data/scalars.json b/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_1e-1/20230604_170206/vis_data/scalars.json new file mode 100644 index 0000000000000000000000000000000000000000..f582b81fd6d5c4c25f6d53ef4717195f47703e37 --- /dev/null +++ b/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_1e-1/20230604_170206/vis_data/scalars.json @@ -0,0 +1,200 @@ +{"lr": 0.00039997529889041205, "data_time": 0.4074542999267578, "loss": 0.017248274432495235, "time": 1.0513962745666503, "epoch": 1, "memory": 15587, "step": 100} +{"lr": 0.0003999002014233684, "data_time": 0.30373153686523435, "loss": 0.010412958264350892, "time": 1.0495609760284423, "epoch": 1, "memory": 15587, "step": 200} +{"lr": 0.0003997747244903853, "data_time": 0.40909771919250487, "loss": 0.023654562700539828, "time": 1.0233510255813598, "epoch": 1, "memory": 15587, "step": 300} +{"lr": 0.00039959890052580966, "data_time": 0.14441344738006592, "loss": 0.021022671554237603, "time": 1.0230309247970581, "epoch": 1, "memory": 15587, "step": 400} +{"lr": 0.0003993727749781197, "data_time": 0.0018305540084838866, "loss": 0.010535456100478768, "time": 0.9646939039230347, "epoch": 1, "memory": 15587, "step": 500} +{"lr": 0.000399096406298175, "data_time": 0.0012839555740356446, "loss": 0.009106617327779532, "time": 0.930386734008789, "epoch": 1, "memory": 15587, "step": 600} +{"lr": 0.0003987698659241082, "data_time": 0.0014708280563354493, "loss": 0.007947701634839177, "time": 0.9239459276199341, "epoch": 1, "memory": 15587, "step": 700} +{"lr": 0.0003983932382628598, "data_time": 0.0013887882232666016, "loss": 0.013005655445158481, "time": 0.9333152294158935, "epoch": 1, "memory": 15587, "step": 800} +{"lr": 0.00039796662066835946, "data_time": 0.0014635324478149414, "loss": 0.0046491074375808236, "time": 0.9554619789123535, "epoch": 1, "memory": 15587, "step": 900} +{"lr": 0.00039749012341636, "data_time": 0.0016210317611694337, "loss": 0.014733085688203574, "time": 0.872484827041626, "epoch": 1, "memory": 15587, "step": 1000} +{"lr": 0.00039696386967593436, "data_time": 0.001524949073791504, "loss": 0.010879722936078907, "time": 0.8295624732971192, "epoch": 1, "memory": 15587, "step": 1100} +{"lr": 0.00039638799547763706, "data_time": 0.00156097412109375, "loss": 0.003423507371917367, "time": 0.8381370782852173, "epoch": 1, "memory": 15587, "step": 1200} +{"lr": 0.0003957626496783407, "data_time": 0.001627635955810547, "loss": 0.0065059004817157986, "time": 0.8229460716247559, "epoch": 1, "memory": 15587, "step": 1300} +{"lr": 0.0003950879939227595, "data_time": 0.0015116214752197265, "loss": 0.011681228922680021, "time": 0.8561073780059815, "epoch": 1, "memory": 15587, "step": 1400} +{"lr": 0.0003943642026016663, "data_time": 0.0016747474670410155, "loss": 0.009761184407398104, "time": 0.8287646055221558, "epoch": 1, "memory": 15587, "step": 1500} +{"lr": 0.0003935914628068133, "data_time": 0.001641058921813965, "loss": 0.01562610026448965, "time": 0.8656338930130005, "epoch": 1, "memory": 15587, "step": 1600} +{"lr": 0.00039276997428257205, "data_time": 0.001541733741760254, "loss": 0.004199318075552583, "time": 0.7898478507995605, "epoch": 1, "memory": 15587, "step": 1700} +{"lr": 0.0003918999493743001, "data_time": 0.0015683174133300781, "loss": 0.0064587795175611975, "time": 0.8095824718475342, "epoch": 1, "memory": 15587, "step": 1800} +{"lr": 0.00039098161297345475, "data_time": 0.0018651485443115234, "loss": 0.00619882894679904, "time": 0.9013674259185791, "epoch": 1, "memory": 15587, "step": 1900} +{"accuracy/top1": 90.46658325195312, "data_time": 0.2132631416978507, "time": 0.6838534618246144, "step": 1} +{"lr": 0.0003894734390071047, "data_time": 0.17534224987030028, "loss": 0.0029863059520721436, "time": 0.850521731376648, "epoch": 2, "memory": 15587, "step": 2054} +{"lr": 0.0003884334872841784, "data_time": 0.0022087574005126955, "loss": 0.006332764867693186, "time": 0.8440931558609008, "epoch": 2, "memory": 15585, "step": 2154} +{"lr": 0.00038734612010935703, "data_time": 0.0021940231323242187, "loss": 0.0032624615821987392, "time": 0.8123685121536255, "epoch": 2, "memory": 15585, "step": 2254} +{"lr": 0.00038621161855458035, "data_time": 0.0024696826934814454, "loss": 0.009495723713189363, "time": 0.8672922611236572, "epoch": 2, "memory": 15585, "step": 2354} +{"lr": 0.0003850302758754841, "data_time": 0.0016459226608276367, "loss": 0.005710620945319533, "time": 0.7940369129180909, "epoch": 2, "memory": 15585, "step": 2454} +{"lr": 0.00038380239743559646, "data_time": 0.0017111539840698243, "loss": 0.008501151436939835, "time": 0.8246843338012695, "epoch": 2, "memory": 15585, "step": 2554} +{"lr": 0.00038252830062740647, "data_time": 0.0015415906906127929, "loss": 0.0045797074213624, "time": 0.8230563640594483, "epoch": 2, "memory": 15585, "step": 2654} +{"lr": 0.00038120831479032175, "data_time": 0.0016323328018188477, "loss": 0.008235360216349363, "time": 0.8180981636047363, "epoch": 2, "memory": 15585, "step": 2754} +{"lr": 0.0003798427811255361, "data_time": 0.0016580343246459961, "loss": 0.008007403230294585, "time": 0.872388505935669, "epoch": 2, "memory": 15585, "step": 2854} +{"lr": 0.0003784320526078357, "data_time": 0.001970386505126953, "loss": 0.00711355684325099, "time": 0.8330523490905761, "epoch": 2, "memory": 15585, "step": 2954} +{"lr": 0.0003769764938943556, "data_time": 0.00246584415435791, "loss": 0.010007486958056689, "time": 0.8688191652297974, "epoch": 2, "memory": 15585, "step": 3054} +{"lr": 0.0003754764812303242, "data_time": 0.0023420572280883787, "loss": 0.0043166371062397955, "time": 0.849473237991333, "epoch": 2, "memory": 15585, "step": 3154} +{"lr": 0.0003739324023518048, "data_time": 0.002280259132385254, "loss": 0.0035292053129523993, "time": 0.9241914033889771, "epoch": 2, "memory": 15585, "step": 3254} +{"lr": 0.0003723446563854724, "data_time": 0.0024772405624389647, "loss": 0.0033644698560237885, "time": 0.8862639427185058, "epoch": 2, "memory": 15585, "step": 3354} +{"lr": 0.0003707136537454424, "data_time": 0.002196335792541504, "loss": 0.010240558814257384, "time": 0.8627858638763428, "epoch": 2, "memory": 15585, "step": 3454} +{"lr": 0.0003690398160271841, "data_time": 0.12454843521118164, "loss": 0.005815344536677003, "time": 0.823633337020874, "epoch": 2, "memory": 15585, "step": 3554} +{"lr": 0.00036732357589854426, "data_time": 0.076629376411438, "loss": 0.00993875446729362, "time": 0.8245397567749023, "epoch": 2, "memory": 15585, "step": 3654} +{"lr": 0.00036556537698790475, "data_time": 0.0016868829727172852, "loss": 0.009170089801773428, "time": 0.843467116355896, "epoch": 2, "memory": 15585, "step": 3754} +{"lr": 0.0003637656737695118, "data_time": 0.0016482353210449218, "loss": 0.003345610899850726, "time": 0.8315291404724121, "epoch": 2, "memory": 15585, "step": 3854} +{"accuracy/top1": 98.37166595458984, "data_time": 0.2504310051600138, "time": 0.6425877412160238, "step": 2} +{"lr": 0.00036091403361175644, "data_time": 0.1533118486404419, "loss": 0.0037349598947912454, "time": 0.8195040702819825, "epoch": 3, "memory": 15585, "step": 4008} +{"lr": 0.00035901102535335625, "data_time": 0.002349710464477539, "loss": 0.00564337195828557, "time": 0.9181422472000123, "epoch": 3, "memory": 15585, "step": 4108} +{"lr": 0.00035706820701190153, "data_time": 0.0022867679595947265, "loss": 0.00629440750926733, "time": 0.8384639978408813, "epoch": 3, "memory": 15585, "step": 4208} +{"lr": 0.00035508608078364215, "data_time": 0.02729918956756592, "loss": 0.004229048639535904, "time": 0.8654555320739746, "epoch": 3, "memory": 15585, "step": 4308} +{"lr": 0.00035306515902546755, "data_time": 0.16883337497711182, "loss": 0.0037390483543276788, "time": 0.8887279510498047, "epoch": 3, "memory": 15585, "step": 4408} +{"lr": 0.0003510059641224655, "data_time": 0.3455545902252197, "loss": 0.0037441214080899954, "time": 0.8658688306808472, "epoch": 3, "memory": 15585, "step": 4508} +{"lr": 0.00034890902835289306, "data_time": 0.2988360643386841, "loss": 0.005673976615071297, "time": 0.8163290739059448, "epoch": 3, "memory": 15585, "step": 4608} +{"lr": 0.00034677489375058753, "data_time": 0.0026842594146728516, "loss": 0.006808177474886179, "time": 0.8540040731430054, "epoch": 3, "memory": 15585, "step": 4708} +{"lr": 0.00034460411196485875, "data_time": 0.0023552656173706056, "loss": 0.007155083771795035, "time": 0.8360354661941528, "epoch": 3, "memory": 15585, "step": 4808} +{"lr": 0.0003423972441178926, "data_time": 0.002347731590270996, "loss": 0.010615699458867311, "time": 0.7946882963180542, "epoch": 3, "memory": 15585, "step": 4908} +{"lr": 0.0003401548606597075, "data_time": 0.0019759416580200197, "loss": 0.004952302295714616, "time": 0.8103042840957642, "epoch": 3, "memory": 15585, "step": 5008} +{"lr": 0.0003378775412207003, "data_time": 0.0015921354293823241, "loss": 0.007694493606686592, "time": 0.8359559535980224, "epoch": 3, "memory": 15585, "step": 5108} +{"lr": 0.0003355658744618178, "data_time": 0.0015553712844848632, "loss": 0.0053688473999500275, "time": 0.8929475784301758, "epoch": 3, "memory": 15585, "step": 5208} +{"lr": 0.0003332204579223944, "data_time": 0.002374744415283203, "loss": 0.006741747166961432, "time": 0.8350327014923096, "epoch": 3, "memory": 15585, "step": 5308} +{"lr": 0.0003308418978656968, "data_time": 0.0018707990646362304, "loss": 0.0036798327229917048, "time": 0.8931418657302856, "epoch": 3, "memory": 15585, "step": 5408} +{"lr": 0.0003284308091222099, "data_time": 0.001961803436279297, "loss": 0.005295302253216505, "time": 0.8708714008331299, "epoch": 3, "memory": 15585, "step": 5508} +{"lr": 0.0003259878149307108, "data_time": 0.0017261028289794922, "loss": 0.0031720946542918684, "time": 0.8444495916366577, "epoch": 3, "memory": 15585, "step": 5608} +{"lr": 0.0003235135467771688, "data_time": 0.0017846107482910156, "loss": 0.0033446666318923237, "time": 0.8666119575500488, "epoch": 3, "memory": 15585, "step": 5708} +{"lr": 0.0003210086442315141, "data_time": 0.0023134469985961912, "loss": 0.004942983482033014, "time": 0.8863066911697388, "epoch": 3, "memory": 15585, "step": 5808} +{"accuracy/top1": 99.45722198486328, "data_time": 0.24220484097798664, "time": 0.6292988936106364, "step": 3} +{"lr": 0.00031709267628953966, "data_time": 0.30004100799560546, "loss": 0.004815851850435138, "time": 0.8579655408859252, "epoch": 4, "memory": 15585, "step": 5962} +{"lr": 0.00031451289120303685, "data_time": 0.04376761913299561, "loss": 0.0059142212849110365, "time": 0.7941560268402099, "epoch": 4, "memory": 15585, "step": 6062} +{"lr": 0.000311904798290621, "data_time": 0.13209106922149658, "loss": 0.0015460836701095104, "time": 0.8303063392639161, "epoch": 4, "memory": 15585, "step": 6162} +{"lr": 0.00030926907171438685, "data_time": 0.036271977424621585, "loss": 0.005030934326350689, "time": 0.8559933662414551, "epoch": 4, "memory": 15585, "step": 6262} +{"lr": 0.000306606392779414, "data_time": 0.06773190498352051, "loss": 0.005842992011457682, "time": 0.8093076944351196, "epoch": 4, "memory": 15585, "step": 6362} +{"lr": 0.0003039174497576579, "data_time": 0.001483154296875, "loss": 0.00600668704137206, "time": 0.8325612545013428, "epoch": 4, "memory": 15585, "step": 6462} +{"lr": 0.0003012029377100387, "data_time": 0.0016164064407348632, "loss": 0.009730385523289442, "time": 0.78953857421875, "epoch": 4, "memory": 15585, "step": 6562} +{"lr": 0.00029846355830677674, "data_time": 0.2795774698257446, "loss": 0.006102950312197208, "time": 0.8676823377609253, "epoch": 4, "memory": 15585, "step": 6662} +{"lr": 0.0002957000196460183, "data_time": 0.3513307571411133, "loss": 0.003987257648259401, "time": 0.8667858600616455, "epoch": 4, "memory": 15585, "step": 6762} +{"lr": 0.0002929130360708003, "data_time": 0.3393963098526001, "loss": 0.008418860659003258, "time": 0.8544230699539185, "epoch": 4, "memory": 15585, "step": 6862} +{"lr": 0.0002901033279844016, "data_time": 0.16490800380706788, "loss": 0.006323257321491837, "time": 0.8619724750518799, "epoch": 4, "memory": 15585, "step": 6962} +{"lr": 0.00028727162166412623, "data_time": 0.0015342950820922852, "loss": 0.006643430422991514, "time": 0.8409230232238769, "epoch": 4, "memory": 15585, "step": 7062} +{"lr": 0.000284418649073569, "data_time": 0.001609468460083008, "loss": 0.004987420374527573, "time": 0.7618176698684692, "epoch": 4, "memory": 15585, "step": 7162} +{"lr": 0.00028154514767341203, "data_time": 0.0014576673507690429, "loss": 0.005437305709347129, "time": 0.7865698337554932, "epoch": 4, "memory": 15585, "step": 7262} +{"lr": 0.0002786518602307988, "data_time": 0.001808929443359375, "loss": 0.004921631002798676, "time": 0.8115873098373413, "epoch": 4, "memory": 15585, "step": 7362} +{"lr": 0.0002757395346273384, "data_time": 0.0015999794006347657, "loss": 0.00435516182333231, "time": 0.7899288892745971, "epoch": 4, "memory": 15585, "step": 7462} +{"lr": 0.0002728089236657853, "data_time": 0.10272691249847413, "loss": 0.009449858544394374, "time": 0.80626060962677, "epoch": 4, "memory": 15585, "step": 7562} +{"lr": 0.0002698607848754497, "data_time": 0.21255826950073242, "loss": 0.005273900879547, "time": 0.9091211080551147, "epoch": 4, "memory": 15585, "step": 7662} +{"lr": 0.0002668958803163847, "data_time": 0.28281707763671876, "loss": 0.004004256753250957, "time": 0.7986958503723145, "epoch": 4, "memory": 15585, "step": 7762} +{"accuracy/top1": 99.7773208618164, "data_time": 0.26677437623341876, "time": 0.6510602235794067, "step": 4} +{"lr": 0.00026229890681650394, "data_time": 0.2097378969192505, "loss": 0.0011739041656255722, "time": 0.8457151174545288, "epoch": 5, "memory": 15585, "step": 7916} +{"lr": 0.00025929487224058475, "data_time": 0.024641084671020507, "loss": 0.003347337245941162, "time": 0.8889694213867188, "epoch": 5, "memory": 15585, "step": 8016} +{"lr": 0.0002562768030631272, "data_time": 0.0017786502838134765, "loss": 0.0018761730752885342, "time": 0.8516000747680664, "epoch": 5, "memory": 15585, "step": 8116} +{"lr": 0.0002532454794203871, "data_time": 0.0019945383071899416, "loss": 0.006383917247876525, "time": 0.8578239202499389, "epoch": 5, "memory": 15585, "step": 8216} +{"lr": 0.0002502016848747476, "data_time": 0.001480436325073242, "loss": 0.003877262445166707, "time": 0.9104105472564697, "epoch": 5, "memory": 15585, "step": 8316} +{"lr": 0.0002471462062121758, "data_time": 0.0015180349349975587, "loss": 0.0016422943212091924, "time": 0.8483364820480347, "epoch": 5, "memory": 15585, "step": 8416} +{"lr": 0.00024407983323884952, "data_time": 0.0017547130584716797, "loss": 0.003953368682414293, "time": 0.8449782848358154, "epoch": 5, "memory": 15585, "step": 8516} +{"lr": 0.0002410033585770006, "data_time": 0.0016140937805175781, "loss": 0.0019638130441308023, "time": 0.8575214385986328, "epoch": 5, "memory": 15585, "step": 8616} +{"lr": 0.00023791757746003292, "data_time": 0.0015656232833862304, "loss": 0.008501792373135687, "time": 0.9225797414779663, "epoch": 5, "memory": 15585, "step": 8716} +{"lr": 0.00023482328752696182, "data_time": 0.0017169713973999023, "loss": 0.007375417789444327, "time": 0.8549912691116333, "epoch": 5, "memory": 15585, "step": 8816} +{"lr": 0.00023172128861623345, "data_time": 0.00161895751953125, "loss": 0.0027543580159544944, "time": 0.8735415220260621, "epoch": 5, "memory": 15585, "step": 8916} +{"lr": 0.000228612382558976, "data_time": 0.001691722869873047, "loss": 0.005506934691220522, "time": 0.8721412420272827, "epoch": 5, "memory": 15585, "step": 9016} +{"lr": 0.00022549737297173692, "data_time": 0.0018543481826782226, "loss": 0.0029317505192011594, "time": 0.9107064008712769, "epoch": 5, "memory": 15585, "step": 9116} +{"lr": 0.00022237706504875477, "data_time": 0.0015569210052490234, "loss": 0.003759072907269001, "time": 0.8398517370223999, "epoch": 5, "memory": 15585, "step": 9216} +{"lr": 0.00021925226535382818, "data_time": 0.0015794992446899413, "loss": 0.0014174559153616428, "time": 0.9286900997161865, "epoch": 5, "memory": 15585, "step": 9316} +{"lr": 0.0002161237816118266, "data_time": 0.0015759944915771484, "loss": 0.0025255458895117044, "time": 0.8815348863601684, "epoch": 5, "memory": 15585, "step": 9416} +{"lr": 0.00021299242249990383, "data_time": 0.001840829849243164, "loss": 0.003546580672264099, "time": 0.8672035217285157, "epoch": 5, "memory": 15585, "step": 9516} +{"lr": 0.00020985899743846363, "data_time": 0.0017792940139770507, "loss": 0.003579519363120198, "time": 0.8193649530410767, "epoch": 5, "memory": 15585, "step": 9616} +{"lr": 0.00020672431638193363, "data_time": 0.0017078876495361327, "loss": 0.0037071458995342256, "time": 0.8633199453353881, "epoch": 5, "memory": 15585, "step": 9716} +{"accuracy/top1": 99.84342956542969, "data_time": 0.23723048369089764, "time": 0.6239738384882609, "step": 5} +{"lr": 0.00020189632111931318, "data_time": 0.3125102996826172, "loss": 0.002428907435387373, "time": 0.8290535688400269, "epoch": 6, "memory": 15585, "step": 9870} +{"lr": 0.00019876209288860824, "data_time": 0.30610611438751223, "loss": 0.00241771056316793, "time": 0.8245126485824585, "epoch": 6, "memory": 15585, "step": 9970} +{"lr": 0.00019562947708532286, "data_time": 0.29919991493225095, "loss": 0.00111872018314898, "time": 0.8460508823394776, "epoch": 6, "memory": 15585, "step": 10070} +{"lr": 0.00019249928345470104, "data_time": 0.2790926456451416, "loss": 0.0014760125428438186, "time": 0.8309964656829834, "epoch": 6, "memory": 15585, "step": 10170} +{"lr": 0.00018937232111588298, "data_time": 0.0015459060668945312, "loss": 0.005606566648930311, "time": 0.9613274812698365, "epoch": 6, "memory": 15585, "step": 10270} +{"lr": 0.000186249398352757, "data_time": 0.001607060432434082, "loss": 0.0035888387355953455, "time": 0.9406477689743042, "epoch": 6, "memory": 15585, "step": 10370} +{"lr": 0.0001831313224050277, "data_time": 0.0019884109497070312, "loss": 0.005360460001975298, "time": 0.8657693147659302, "epoch": 6, "memory": 15585, "step": 10470} +{"lr": 0.00018001889925955291, "data_time": 0.001595926284790039, "loss": 0.004190133512020111, "time": 0.8827555894851684, "epoch": 6, "memory": 15585, "step": 10570} +{"lr": 0.00017691293344200646, "data_time": 0.0016616582870483398, "loss": 0.0020628038328140974, "time": 0.9320860147476197, "epoch": 6, "memory": 15585, "step": 10670} +{"lr": 0.00017381422780891689, "data_time": 0.0015587806701660156, "loss": 0.002030350873246789, "time": 0.909303092956543, "epoch": 6, "memory": 15585, "step": 10770} +{"lr": 0.00017072358334013806, "data_time": 0.0015881061553955078, "loss": 0.0020942374598234893, "time": 0.8919180393218994, "epoch": 6, "memory": 15585, "step": 10870} +{"lr": 0.00016764179893180575, "data_time": 0.002107691764831543, "loss": 0.0032632900401949883, "time": 0.8498372077941895, "epoch": 6, "memory": 15585, "step": 10970} +{"lr": 0.00016456967118983176, "data_time": 0.001712965965270996, "loss": 0.0008120507467538118, "time": 0.8277901411056519, "epoch": 6, "memory": 15585, "step": 11070} +{"lr": 0.00016150799422398984, "data_time": 0.0016379594802856446, "loss": 0.003254414862021804, "time": 0.8336925745010376, "epoch": 6, "memory": 15585, "step": 11170} +{"lr": 0.0001584575594426489, "data_time": 0.0017516136169433594, "loss": 0.0008459640201181173, "time": 0.8705430507659913, "epoch": 6, "memory": 15585, "step": 11270} +{"lr": 0.00015541915534820135, "data_time": 0.0015369653701782227, "loss": 0.0025393597315996885, "time": 0.8745918273925781, "epoch": 6, "memory": 15585, "step": 11370} +{"lr": 0.00015239356733324542, "data_time": 0.0015781164169311524, "loss": 0.002424947964027524, "time": 0.8247071504592896, "epoch": 6, "memory": 15585, "step": 11470} +{"lr": 0.00014938157747756997, "data_time": 0.001655721664428711, "loss": 0.0021597898565232755, "time": 0.8379346847534179, "epoch": 6, "memory": 15585, "step": 11570} +{"lr": 0.00014638396434599565, "data_time": 0.0015805721282958984, "loss": 0.003143156925216317, "time": 0.8156458377838135, "epoch": 6, "memory": 15585, "step": 11670} +{"accuracy/top1": 99.87474822998047, "data_time": 0.24801313877105713, "time": 0.6315874497095744, "step": 6} +{"lr": 0.00014179754513556176, "data_time": 0.3228867769241333, "loss": 0.0003779163584113121, "time": 0.838274073600769, "epoch": 7, "memory": 15585, "step": 11824} +{"lr": 0.00013883992334674444, "data_time": 0.0016718864440917968, "loss": 0.002053354913368821, "time": 0.8226800203323364, "epoch": 7, "memory": 15585, "step": 11924} +{"lr": 0.00013589940317867467, "data_time": 0.0015621185302734375, "loss": 0.003381268586963415, "time": 0.8198431015014649, "epoch": 7, "memory": 15585, "step": 12024} +{"lr": 0.0001329767447220783, "data_time": 0.0016316890716552735, "loss": 0.0016960741020739078, "time": 0.8422715425491333, "epoch": 7, "memory": 15585, "step": 12124} +{"lr": 0.00013007270345063488, "data_time": 0.0018516063690185546, "loss": 0.0038290629163384438, "time": 0.8356634378433228, "epoch": 7, "memory": 15585, "step": 12224} +{"lr": 0.00012718803002569412, "data_time": 0.0019460439682006836, "loss": 0.0010147742927074433, "time": 0.8165646314620971, "epoch": 7, "memory": 15585, "step": 12324} +{"lr": 0.0001243234701022397, "data_time": 0.0016362905502319337, "loss": 0.0013078055344522, "time": 0.8601309537887574, "epoch": 7, "memory": 15585, "step": 12424} +{"lr": 0.00012147976413614597, "data_time": 0.0015713691711425782, "loss": 0.0014702303800731898, "time": 0.8406438589096069, "epoch": 7, "memory": 15585, "step": 12524} +{"lr": 0.0001186576471927784, "data_time": 0.0017952442169189454, "loss": 0.0038851252757012844, "time": 0.8938199043273926, "epoch": 7, "memory": 15585, "step": 12624} +{"lr": 0.00011585784875698867, "data_time": 0.0016071557998657226, "loss": 0.0028708187863230707, "time": 0.8393218278884887, "epoch": 7, "memory": 15585, "step": 12724} +{"lr": 0.00011308109254454993, "data_time": 0.0016457319259643554, "loss": 0.002554943924769759, "time": 0.8666845083236694, "epoch": 7, "memory": 15585, "step": 12824} +{"lr": 0.00011032809631508466, "data_time": 0.0015531063079833984, "loss": 0.0005809028167277575, "time": 0.8390853643417359, "epoch": 7, "memory": 15585, "step": 12924} +{"lr": 0.00010759957168653253, "data_time": 0.0018207550048828125, "loss": 0.0018609888385981322, "time": 0.9048506259918213, "epoch": 7, "memory": 15585, "step": 13024} +{"lr": 0.0001048962239512054, "data_time": 0.0019021749496459961, "loss": 0.0022601188626140355, "time": 0.819468641281128, "epoch": 7, "memory": 15585, "step": 13124} +{"lr": 0.00010221875189347644, "data_time": 0.001586747169494629, "loss": 0.004573145974427461, "time": 0.847865629196167, "epoch": 7, "memory": 15585, "step": 13224} +{"lr": 9.95678476091537e-05, "data_time": 0.0014557838439941406, "loss": 0.0008064919617027045, "time": 0.8420084714889526, "epoch": 7, "memory": 15585, "step": 13324} +{"lr": 9.694419632657999e-05, "data_time": 0.0020059823989868166, "loss": 0.0009036433417350054, "time": 0.8855749130249023, "epoch": 7, "memory": 15585, "step": 13424} +{"lr": 9.434847622951059e-05, "data_time": 0.002084207534790039, "loss": 0.000638015754520893, "time": 0.8360074758529663, "epoch": 7, "memory": 15585, "step": 13524} +{"lr": 9.178135828180905e-05, "data_time": 0.0018365859985351562, "loss": 0.004058591555804014, "time": 0.8574733018875123, "epoch": 7, "memory": 15585, "step": 13624} +{"accuracy/top1": 99.89214324951172, "data_time": 0.24434186617533366, "time": 0.6280344963073731, "step": 7} +{"lr": 8.788546579134055e-05, "data_time": 0.001804018020629883, "loss": 0.0016758719924837352, "time": 0.8845922231674195, "epoch": 8, "memory": 15585, "step": 13778} +{"lr": 8.539396307206318e-05, "data_time": 0.0015428543090820312, "loss": 0.0024170370772480963, "time": 0.8424384832382202, "epoch": 8, "memory": 15585, "step": 13878} +{"lr": 8.293337714106816e-05, "data_time": 0.001485586166381836, "loss": 0.00015728422440588474, "time": 0.8494835138320923, "epoch": 8, "memory": 15585, "step": 13978} +{"lr": 8.05043440315878e-05, "data_time": 0.0015645027160644531, "loss": 0.0004042779561132193, "time": 0.8392947673797607, "epoch": 8, "memory": 15585, "step": 14078} +{"lr": 7.810749162081216e-05, "data_time": 0.0016747236251831055, "loss": 0.0004208637401461601, "time": 0.8204168558120728, "epoch": 8, "memory": 15585, "step": 14178} +{"lr": 7.574343946758993e-05, "data_time": 0.017456459999084472, "loss": 0.0002836901228874922, "time": 0.8490535974502563, "epoch": 8, "memory": 15585, "step": 14278} +{"lr": 7.34127986522793e-05, "data_time": 0.08441815376281739, "loss": 0.0018746765796095132, "time": 0.7804399013519288, "epoch": 8, "memory": 15585, "step": 14378} +{"lr": 7.111617161879136e-05, "data_time": 0.11728715896606445, "loss": 0.0003813054412603378, "time": 0.884506344795227, "epoch": 8, "memory": 15585, "step": 14478} +{"lr": 6.885415201886485e-05, "data_time": 0.33581278324127195, "loss": 0.0009060120210051537, "time": 0.8761226177215576, "epoch": 8, "memory": 15585, "step": 14578} +{"lr": 6.662732455861337e-05, "data_time": 0.5895336389541626, "loss": 0.0013958164490759373, "time": 1.1073842525482178, "epoch": 8, "memory": 15585, "step": 14678} +{"lr": 6.443626484738661e-05, "data_time": 0.2097240924835205, "loss": 0.000790273305028677, "time": 0.726179552078247, "epoch": 8, "memory": 15585, "step": 14778} +{"lr": 6.22815392489815e-05, "data_time": 0.2031512975692749, "loss": 0.0011591968592256307, "time": 0.7829782724380493, "epoch": 8, "memory": 15585, "step": 14878} +{"lr": 6.016370473524328e-05, "data_time": 0.0015236854553222657, "loss": 0.0017688107211142779, "time": 0.7770345687866211, "epoch": 8, "memory": 15585, "step": 14978} +{"lr": 5.80833087420956e-05, "data_time": 0.03259367942810058, "loss": 0.002623250987380743, "time": 0.8210611343383789, "epoch": 8, "memory": 15585, "step": 15078} +{"lr": 5.604088902803397e-05, "data_time": 0.10467522144317627, "loss": 0.00075977579690516, "time": 0.8388400077819824, "epoch": 8, "memory": 15585, "step": 15178} +{"lr": 5.403697353512148e-05, "data_time": 0.1727466106414795, "loss": 0.0008391457144170999, "time": 0.8542494058609009, "epoch": 8, "memory": 15585, "step": 15278} +{"lr": 5.207208025252174e-05, "data_time": 0.3552724838256836, "loss": 0.001340480847284198, "time": 0.8716519355773926, "epoch": 8, "memory": 15585, "step": 15378} +{"lr": 5.014671708260458e-05, "data_time": 0.35734920501708983, "loss": 0.0007238555233925581, "time": 0.8721176624298096, "epoch": 8, "memory": 15585, "step": 15478} +{"lr": 4.826138170965952e-05, "data_time": 0.3547613859176636, "loss": 0.006013083131983876, "time": 0.8718185186386108, "epoch": 8, "memory": 15585, "step": 15578} +{"accuracy/top1": 99.92345428466797, "data_time": 0.2632083098093669, "time": 0.6461691697438557, "step": 8} +{"lr": 4.5437373040402475e-05, "data_time": 0.12021346092224121, "loss": 0.0011159366462379695, "time": 0.8541103363037109, "epoch": 9, "memory": 15585, "step": 15732} +{"lr": 4.3655875036148156e-05, "data_time": 0.0016057968139648437, "loss": 0.0005931212566792965, "time": 0.7904397964477539, "epoch": 9, "memory": 15585, "step": 15832} +{"lr": 4.19160826370641e-05, "data_time": 0.0015247821807861327, "loss": 0.0005626421887427568, "time": 0.8504166841506958, "epoch": 9, "memory": 15585, "step": 15932} +{"lr": 4.021844555952455e-05, "data_time": 0.0016149759292602539, "loss": 0.0007392247207462788, "time": 0.8377806425094605, "epoch": 9, "memory": 15585, "step": 16032} +{"lr": 3.8563402623236864e-05, "data_time": 0.0017754316329956054, "loss": 0.0013027803506702185, "time": 0.829702639579773, "epoch": 9, "memory": 15585, "step": 16132} +{"lr": 3.695138163781135e-05, "data_time": 0.001830744743347168, "loss": 0.004492873838171363, "time": 0.8878957986831665, "epoch": 9, "memory": 15585, "step": 16232} +{"lr": 3.538279929217756e-05, "data_time": 0.0014868974685668945, "loss": 0.0006998658180236816, "time": 0.8603259801864624, "epoch": 9, "memory": 15585, "step": 16332} +{"lr": 3.385806104687531e-05, "data_time": 0.0017010688781738282, "loss": 0.004667029203847051, "time": 0.9070610284805298, "epoch": 9, "memory": 15585, "step": 16432} +{"lr": 3.237756102924734e-05, "data_time": 0.0843231201171875, "loss": 0.0008170569315552711, "time": 0.849350380897522, "epoch": 9, "memory": 15585, "step": 16532} +{"lr": 3.094168193156213e-05, "data_time": 0.024666380882263184, "loss": 0.0003779273945838213, "time": 0.8001426696777344, "epoch": 9, "memory": 15585, "step": 16632} +{"lr": 2.95507949120924e-05, "data_time": 0.0021928787231445313, "loss": 0.0003065277356654406, "time": 0.9163157939910889, "epoch": 9, "memory": 15585, "step": 16732} +{"lr": 2.8205259499174602e-05, "data_time": 0.001944732666015625, "loss": 0.0014087852090597154, "time": 0.941752028465271, "epoch": 9, "memory": 15585, "step": 16832} +{"lr": 2.690542349827549e-05, "data_time": 0.0017181396484375, "loss": 0.001308317668735981, "time": 0.83608078956604, "epoch": 9, "memory": 15585, "step": 16932} +{"lr": 2.565162290208795e-05, "data_time": 0.0017790794372558594, "loss": 0.0026592658367007973, "time": 0.8758543491363525, "epoch": 9, "memory": 15585, "step": 17032} +{"lr": 2.4444181803680975e-05, "data_time": 0.0016588926315307616, "loss": 0.0006535890977829694, "time": 0.8526934385299683, "epoch": 9, "memory": 15585, "step": 17132} +{"lr": 2.32834123127253e-05, "data_time": 0.0020235300064086912, "loss": 2.9822345823049545e-05, "time": 0.8637571334838867, "epoch": 9, "memory": 15585, "step": 17232} +{"lr": 2.2169614474816763e-05, "data_time": 0.0017161130905151366, "loss": 0.001203312911093235, "time": 0.8281512975692749, "epoch": 9, "memory": 15585, "step": 17332} +{"lr": 2.110307619391749e-05, "data_time": 0.0017165899276733398, "loss": 0.0033061815425753593, "time": 0.9068790197372436, "epoch": 9, "memory": 15585, "step": 17432} +{"lr": 2.008407315793661e-05, "data_time": 0.0017263174057006836, "loss": 0.00022503635846078397, "time": 0.8267875194549561, "epoch": 9, "memory": 15585, "step": 17532} +{"accuracy/top1": 99.94781494140625, "data_time": 0.2644737482070923, "time": 0.6500085671742757, "step": 9} +{"lr": 1.8608381954463387e-05, "data_time": 0.20412657260894776, "loss": 0.0011625594459474086, "time": 0.9019122123718262, "epoch": 10, "memory": 15585, "step": 17686} +{"lr": 1.771127410231484e-05, "data_time": 0.0016240119934082032, "loss": 0.0019862732850015165, "time": 0.8680342197418213, "epoch": 10, "memory": 15585, "step": 17786} +{"lr": 1.6862578237010542e-05, "data_time": 0.001675558090209961, "loss": 0.001506176684051752, "time": 0.8527591228485107, "epoch": 10, "memory": 15585, "step": 17886} +{"lr": 1.6062513736694383e-05, "data_time": 0.0016326665878295898, "loss": 0.0014381874352693558, "time": 0.8908598184585571, "epoch": 10, "memory": 15585, "step": 17986} +{"lr": 1.531128740886036e-05, "data_time": 0.0015998125076293946, "loss": 0.00033911271020770073, "time": 0.8886195898056031, "epoch": 10, "memory": 15585, "step": 18086} +{"lr": 1.4609093436895328e-05, "data_time": 0.0015250444412231445, "loss": 0.0001683767419308424, "time": 0.8753059148788452, "epoch": 10, "memory": 15585, "step": 18186} +{"lr": 1.3956113329884566e-05, "data_time": 0.0019235849380493165, "loss": 0.00039821406826376913, "time": 0.8746515035629272, "epoch": 10, "memory": 15585, "step": 18286} +{"lr": 1.335251587569401e-05, "data_time": 0.0015807151794433594, "loss": 0.0007395950146019459, "time": 0.8567647933959961, "epoch": 10, "memory": 15585, "step": 18386} +{"lr": 1.2798457097340337e-05, "data_time": 0.0018355131149291992, "loss": 0.0005796691868454218, "time": 0.861804461479187, "epoch": 10, "memory": 15585, "step": 18486} +{"lr": 1.229408021266094e-05, "data_time": 0.001537156105041504, "loss": 0.002770001208409667, "time": 0.830748462677002, "epoch": 10, "memory": 15585, "step": 18586} +{"lr": 1.1839515597293723e-05, "data_time": 0.001573944091796875, "loss": 0.0003068405669182539, "time": 0.8169232606887817, "epoch": 10, "memory": 15585, "step": 18686} +{"lr": 1.1434880750976406e-05, "data_time": 0.0017671585083007812, "loss": 6.0982629656791686e-05, "time": 0.8205039978027344, "epoch": 10, "memory": 15585, "step": 18786} +{"lr": 1.108028026717431e-05, "data_time": 0.0015892982482910156, "loss": 0.00025234329514205455, "time": 0.8637696266174316, "epoch": 10, "memory": 15585, "step": 18886} +{"lr": 1.0775805806044138e-05, "data_time": 0.0015877008438110352, "loss": 0.0005985628813505173, "time": 0.8084832191467285, "epoch": 10, "memory": 15585, "step": 18986} +{"lr": 1.0521536070740776e-05, "data_time": 0.0017740964889526368, "loss": 0.0004730156622827053, "time": 0.8373129367828369, "epoch": 10, "memory": 15585, "step": 19086} +{"lr": 1.03175367870735e-05, "data_time": 0.0015578746795654296, "loss": 0.0011395436245948075, "time": 0.8919585704803467, "epoch": 10, "memory": 15585, "step": 19186} +{"lr": 1.0163860686516628e-05, "data_time": 0.0016566753387451173, "loss": 0.00020440882071852683, "time": 0.9104677200317383, "epoch": 10, "memory": 15585, "step": 19286} +{"lr": 1.0060547492578997e-05, "data_time": 0.0015944719314575195, "loss": 0.0002673800103366375, "time": 0.8587796926498413, "epoch": 10, "memory": 15585, "step": 19386} +{"lr": 1.0007623910535892e-05, "data_time": 0.0019279956817626954, "loss": 0.0002512660808861256, "time": 0.8323871612548828, "epoch": 10, "memory": 15585, "step": 19486} +{"accuracy/top1": 99.94781494140625, "data_time": 0.24300116697947186, "time": 0.6269446690877278, "step": 10} diff --git a/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_1e-1/20230604_170206/vis_data/vis_image/0000000.png_1.png b/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_1e-1/20230604_170206/vis_data/vis_image/0000000.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..f536bbe218286711dc37b252b6874e242bf6d111 Binary files /dev/null and b/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_1e-1/20230604_170206/vis_data/vis_image/0000000.png_1.png differ diff --git a/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_1e-1/20230604_170206/vis_data/vis_image/0000000.png_10.png b/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_1e-1/20230604_170206/vis_data/vis_image/0000000.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..6494eacdacb29b92ec46b33529911b31feefac3e Binary files /dev/null and b/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_1e-1/20230604_170206/vis_data/vis_image/0000000.png_10.png differ diff --git a/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_1e-1/20230604_170206/vis_data/vis_image/0000000.png_2.png b/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_1e-1/20230604_170206/vis_data/vis_image/0000000.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..a0db9a1bb7ac6774b15b6173cf7b9527ca2b7cad Binary files /dev/null and b/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_1e-1/20230604_170206/vis_data/vis_image/0000000.png_2.png differ diff --git a/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_1e-1/20230604_170206/vis_data/vis_image/0000000.png_3.png b/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_1e-1/20230604_170206/vis_data/vis_image/0000000.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..95ebbee17f4c017ea874a3281dcdc7d2f7b44f15 Binary files /dev/null and b/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_1e-1/20230604_170206/vis_data/vis_image/0000000.png_3.png differ diff --git a/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_1e-1/20230604_170206/vis_data/vis_image/0000000.png_4.png b/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_1e-1/20230604_170206/vis_data/vis_image/0000000.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..78889de81ef6c4a1866b8315faf361474a6a9567 Binary files /dev/null and b/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_1e-1/20230604_170206/vis_data/vis_image/0000000.png_4.png differ diff --git a/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_1e-1/20230604_170206/vis_data/vis_image/0000000.png_5.png b/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_1e-1/20230604_170206/vis_data/vis_image/0000000.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..a7833312092f4f0b1bb06c9439eeb2122f310b22 Binary files /dev/null and b/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_1e-1/20230604_170206/vis_data/vis_image/0000000.png_5.png differ diff --git a/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_1e-1/20230604_170206/vis_data/vis_image/0000000.png_6.png b/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_1e-1/20230604_170206/vis_data/vis_image/0000000.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..2f56a8022a880b0d03c1df90aa7e6f016b146784 Binary files /dev/null and b/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_1e-1/20230604_170206/vis_data/vis_image/0000000.png_6.png differ diff --git a/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_1e-1/20230604_170206/vis_data/vis_image/0000000.png_7.png b/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_1e-1/20230604_170206/vis_data/vis_image/0000000.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..197ce1a484cdc7579e82eb28318849b489b3efed Binary files /dev/null and b/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_1e-1/20230604_170206/vis_data/vis_image/0000000.png_7.png differ diff --git a/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_1e-1/20230604_170206/vis_data/vis_image/0000000.png_8.png b/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_1e-1/20230604_170206/vis_data/vis_image/0000000.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..dc5843e3f078aff8b3beb72ac660123b5b83d57c Binary files /dev/null and b/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_1e-1/20230604_170206/vis_data/vis_image/0000000.png_8.png differ diff --git a/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_1e-1/20230604_170206/vis_data/vis_image/0000000.png_9.png b/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_1e-1/20230604_170206/vis_data/vis_image/0000000.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..c3d435a790ead861f3dbfb12694176da9ee2c35f Binary files /dev/null and b/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_1e-1/20230604_170206/vis_data/vis_image/0000000.png_9.png differ diff --git a/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_1e-1/20230604_170206/vis_data/vis_image/000005174.jpg_1.png b/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_1e-1/20230604_170206/vis_data/vis_image/000005174.jpg_1.png new file mode 100644 index 0000000000000000000000000000000000000000..62367c4b933bf3533ebdf5b59e2beafc44f257f2 Binary files /dev/null and b/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_1e-1/20230604_170206/vis_data/vis_image/000005174.jpg_1.png differ diff --git a/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_1e-1/20230604_170206/vis_data/vis_image/000005174.jpg_10.png b/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_1e-1/20230604_170206/vis_data/vis_image/000005174.jpg_10.png new file mode 100644 index 0000000000000000000000000000000000000000..2df58975202dd41fe518443b53503cae75cd1bd6 Binary files /dev/null and b/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_1e-1/20230604_170206/vis_data/vis_image/000005174.jpg_10.png differ diff --git a/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_1e-1/20230604_170206/vis_data/vis_image/000005174.jpg_2.png b/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_1e-1/20230604_170206/vis_data/vis_image/000005174.jpg_2.png new file mode 100644 index 0000000000000000000000000000000000000000..53a326afec426013c9484dc64a6d07f62efb8762 Binary files /dev/null and b/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_1e-1/20230604_170206/vis_data/vis_image/000005174.jpg_2.png differ diff --git a/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_1e-1/20230604_170206/vis_data/vis_image/000005174.jpg_3.png b/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_1e-1/20230604_170206/vis_data/vis_image/000005174.jpg_3.png new file mode 100644 index 0000000000000000000000000000000000000000..113b864d1ea87ba7392ca57bb4acfb2ce0bae788 Binary files /dev/null and b/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_1e-1/20230604_170206/vis_data/vis_image/000005174.jpg_3.png differ diff --git a/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_1e-1/20230604_170206/vis_data/vis_image/000005174.jpg_4.png b/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_1e-1/20230604_170206/vis_data/vis_image/000005174.jpg_4.png new file mode 100644 index 0000000000000000000000000000000000000000..d146f1b62c34ab82d017d969b434413d3decfdd7 Binary files /dev/null and b/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_1e-1/20230604_170206/vis_data/vis_image/000005174.jpg_4.png differ diff --git a/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_1e-1/20230604_170206/vis_data/vis_image/000005174.jpg_5.png b/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_1e-1/20230604_170206/vis_data/vis_image/000005174.jpg_5.png new file mode 100644 index 0000000000000000000000000000000000000000..faef76019cefc3299997a922a79ed8bebf7c1e16 Binary files /dev/null and b/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_1e-1/20230604_170206/vis_data/vis_image/000005174.jpg_5.png differ diff --git a/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_1e-1/20230604_170206/vis_data/vis_image/000005174.jpg_6.png b/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_1e-1/20230604_170206/vis_data/vis_image/000005174.jpg_6.png new file mode 100644 index 0000000000000000000000000000000000000000..a54727d4f12e9210cdb4122b861e4aeb6473e691 Binary files /dev/null and b/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_1e-1/20230604_170206/vis_data/vis_image/000005174.jpg_6.png differ diff --git a/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_1e-1/20230604_170206/vis_data/vis_image/000005174.jpg_7.png b/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_1e-1/20230604_170206/vis_data/vis_image/000005174.jpg_7.png new file mode 100644 index 0000000000000000000000000000000000000000..94e1f91e42de9812f78cfbae37d1c0d0aacc1d67 Binary files /dev/null and b/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_1e-1/20230604_170206/vis_data/vis_image/000005174.jpg_7.png differ diff --git a/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_1e-1/20230604_170206/vis_data/vis_image/000005174.jpg_8.png b/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_1e-1/20230604_170206/vis_data/vis_image/000005174.jpg_8.png new file mode 100644 index 0000000000000000000000000000000000000000..9eb3ccaa4ca3408664479ef5fe0c789403909255 Binary files /dev/null and b/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_1e-1/20230604_170206/vis_data/vis_image/000005174.jpg_8.png differ diff --git a/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_1e-1/20230604_170206/vis_data/vis_image/000005174.jpg_9.png b/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_1e-1/20230604_170206/vis_data/vis_image/000005174.jpg_9.png new file mode 100644 index 0000000000000000000000000000000000000000..e60d4a03f0e96cbcec37dc95672f19fd7d30577a Binary files /dev/null and b/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_1e-1/20230604_170206/vis_data/vis_image/000005174.jpg_9.png differ diff --git a/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_1e-1/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_1e-1.py b/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_1e-1/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_1e-1.py new file mode 100644 index 0000000000000000000000000000000000000000..77eb3de7b15a1fc93b06c70f847eec37b0754a05 --- /dev/null +++ b/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_1e-1/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_1e-1.py @@ -0,0 +1,247 @@ +optim_wrapper = dict( + optimizer=dict( + type='AdamW', + lr=0.0004, + weight_decay=0.05, + eps=1e-08, + betas=(0.9, 0.999), + _scope_='mmpretrain'), + paramwise_cfg=dict( + norm_decay_mult=0.0, + bias_decay_mult=0.0, + flat_decay_mult=0.0, + custom_keys=dict({ + '.absolute_pos_embed': dict(decay_mult=0.0), + '.relative_position_bias_table': dict(decay_mult=0.0) + })), + type='AmpOptimWrapper', + dtype='bfloat16', + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=False, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=4096) +model = dict( + type='ImageClassifier', + backbone=dict( + frozen_stages=2, + type='ConvNeXt', + arch='small', + drop_path_rate=0.4, + init_cfg=dict( + type='Pretrained', + checkpoint= + 'https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth', + prefix='backbone')), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=768, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + init_cfg=None), + init_cfg=dict( + type='TruncNormal', layer=['Conv2d', 'Linear'], std=0.02, bias=0.0), + train_cfg=None) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stablediffusionV1-5R2-dpmsolver-25-1m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/cc1m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_1e-1' diff --git a/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_1e-1/epoch_1.pth b/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_1e-1/epoch_1.pth new file mode 100644 index 0000000000000000000000000000000000000000..ba56558492852317ed325cc6ee80978234bd06df --- /dev/null +++ b/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_1e-1/epoch_1.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:8ecbc7125f6c955853999253516747a47a5a0cf12cf560867e4aebcf703aae74 +size 782108890 diff --git a/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_1e-1/epoch_10.pth b/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_1e-1/epoch_10.pth new file mode 100644 index 0000000000000000000000000000000000000000..2aeb1111d313b5b6ca16ebbe4c3d0d8114e6fba1 --- /dev/null +++ b/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_1e-1/epoch_10.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:2a21a8cc29da5e4fb32fd7129b8969111471a5dcaa5736c93d22ff84c478dcae +size 783377946 diff --git a/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_1e-1/epoch_2.pth b/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_1e-1/epoch_2.pth new file mode 100644 index 0000000000000000000000000000000000000000..be90d137b5fd431896050b1e399719c9734c645e --- /dev/null +++ b/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_1e-1/epoch_2.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:1408d6c42d18e5b07e3ea6ee0a8b8a25ce4bb841ac5662ab291efdbb8230354d +size 782251290 diff --git a/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_1e-1/epoch_3.pth b/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_1e-1/epoch_3.pth new file mode 100644 index 0000000000000000000000000000000000000000..710367841e0514db40803a7d8f8f3ae576df5f63 --- /dev/null +++ b/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_1e-1/epoch_3.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:7b8a6353eb2000624c06573ea644e9d7f38e9244bea4161a75c79bd361528485 +size 782392282 diff --git a/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_1e-1/epoch_4.pth b/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_1e-1/epoch_4.pth new file mode 100644 index 0000000000000000000000000000000000000000..3fa5880eb03735533f360d46e0fcf9beaf7e5683 --- /dev/null +++ b/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_1e-1/epoch_4.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:906498bac874f3489ee74e36b882e22e2444cffa1e186b214c87fcfec68adf91 +size 782533146 diff --git a/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_1e-1/epoch_5.pth b/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_1e-1/epoch_5.pth new file mode 100644 index 0000000000000000000000000000000000000000..abeb8bd06017bec372a12e910b8b83a975720917 --- /dev/null +++ b/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_1e-1/epoch_5.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:43b5a0f0bfb72d6848ba7ea18c176e0e93d46fb1d0ad752b1dc4a0070a100509 +size 782673370 diff --git a/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_1e-1/epoch_6.pth b/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_1e-1/epoch_6.pth new file mode 100644 index 0000000000000000000000000000000000000000..c4d1617d98cee9dce40f1412608ba56d3e7e23e2 --- /dev/null +++ b/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_1e-1/epoch_6.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:40f2c806141c98bfc70bf5406eab1a65d0aaad18aab5cb5b37ad977a124ed5be +size 782813530 diff --git a/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_1e-1/epoch_7.pth b/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_1e-1/epoch_7.pth new file mode 100644 index 0000000000000000000000000000000000000000..c4eceba712b0315bf445f9c13c905e074262ab42 --- /dev/null +++ b/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_1e-1/epoch_7.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:54d446a176265671cd38aacae622f98d28b29d108fce6a65eed8b956fddc619e +size 782953754 diff --git a/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_1e-1/epoch_8.pth b/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_1e-1/epoch_8.pth new file mode 100644 index 0000000000000000000000000000000000000000..275da3742ff66c10ec1069cc433b50617b4f9a48 --- /dev/null +++ b/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_1e-1/epoch_8.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c77c7c52b50fc8523c5a6c7dfa7256eb62fbf7065e70b94bd787be74b6aced35 +size 783094234 diff --git a/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_1e-1/epoch_9.pth b/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_1e-1/epoch_9.pth new file mode 100644 index 0000000000000000000000000000000000000000..99567be1f350850af78818c8336d7004fa5f75f4 --- /dev/null +++ b/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_1e-1/epoch_9.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:05427385b5a141d56d35aa9d712591b9e4c2a795a5da6cff34ffead18f5a1a72 +size 783235610 diff --git a/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_1e-1/last_checkpoint b/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_1e-1/last_checkpoint new file mode 100644 index 0000000000000000000000000000000000000000..1c3a50e08c5c6ccf792314c9230ab5423f79add2 --- /dev/null +++ b/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_1e-1/last_checkpoint @@ -0,0 +1 @@ +/mnt/petrelfs/luzeyu/workspace/fakebench/mmpretrain/workdir/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_1e-1/epoch_10.pth \ No newline at end of file diff --git a/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_5e-1/20230604_164906/20230604_164906.log b/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_5e-1/20230604_164906/20230604_164906.log new file mode 100644 index 0000000000000000000000000000000000000000..c7a6289ba73c5f85cb206a315bc80d16656d83b9 --- /dev/null +++ b/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_5e-1/20230604_164906/20230604_164906.log @@ -0,0 +1,1840 @@ +2023/06/04 16:49:10 - mmengine - INFO - +------------------------------------------------------------ +System environment: + sys.platform: linux + Python: 3.10.9 (main, Mar 8 2023, 10:47:38) [GCC 11.2.0] + CUDA available: True + numpy_random_seed: 1640872635 + GPU 0,1,2,3: NVIDIA A100-SXM4-80GB + CUDA_HOME: /mnt/petrelfs/share/cuda-11.6 + NVCC: Cuda compilation tools, release 11.6, V11.6.124 + GCC: gcc (GCC) 7.5.0 + PyTorch: 1.13.1 + PyTorch compiling details: PyTorch built with: + - GCC 9.3 + - C++ Version: 201402 + - Intel(R) oneAPI Math Kernel Library Version 2021.4-Product Build 20210904 for Intel(R) 64 architecture applications + - Intel(R) MKL-DNN v2.6.0 (Git Hash 52b5f107dd9cf10910aaa19cb47f3abf9b349815) + - OpenMP 201511 (a.k.a. OpenMP 4.5) + - LAPACK is enabled (usually provided by MKL) + - NNPACK is enabled + - CPU capability usage: AVX2 + - CUDA Runtime 11.6 + - NVCC architecture flags: -gencode;arch=compute_37,code=sm_37;-gencode;arch=compute_50,code=sm_50;-gencode;arch=compute_60,code=sm_60;-gencode;arch=compute_61,code=sm_61;-gencode;arch=compute_70,code=sm_70;-gencode;arch=compute_75,code=sm_75;-gencode;arch=compute_80,code=sm_80;-gencode;arch=compute_86,code=sm_86;-gencode;arch=compute_37,code=compute_37 + - CuDNN 8.3.2 (built against CUDA 11.5) + - Magma 2.6.1 + - Build settings: BLAS_INFO=mkl, BUILD_TYPE=Release, CUDA_VERSION=11.6, CUDNN_VERSION=8.3.2, CXX_COMPILER=/opt/rh/devtoolset-9/root/usr/bin/c++, CXX_FLAGS= -fabi-version=11 -Wno-deprecated -fvisibility-inlines-hidden -DUSE_PTHREADPOOL -fopenmp -DNDEBUG -DUSE_KINETO -DUSE_FBGEMM -DUSE_QNNPACK -DUSE_PYTORCH_QNNPACK -DUSE_XNNPACK -DSYMBOLICATE_MOBILE_DEBUG_HANDLE -DEDGE_PROFILER_USE_KINETO -O2 -fPIC -Wno-narrowing -Wall -Wextra -Werror=return-type -Werror=non-virtual-dtor -Wno-missing-field-initializers -Wno-type-limits -Wno-array-bounds -Wno-unknown-pragmas -Wunused-local-typedefs -Wno-unused-parameter -Wno-unused-function -Wno-unused-result -Wno-strict-overflow -Wno-strict-aliasing -Wno-error=deprecated-declarations -Wno-stringop-overflow -Wno-psabi -Wno-error=pedantic -Wno-error=redundant-decls -Wno-error=old-style-cast -fdiagnostics-color=always -faligned-new -Wno-unused-but-set-variable -Wno-maybe-uninitialized -fno-math-errno -fno-trapping-math -Werror=format -Werror=cast-function-type -Wno-stringop-overflow, LAPACK_INFO=mkl, PERF_WITH_AVX=1, PERF_WITH_AVX2=1, PERF_WITH_AVX512=1, TORCH_VERSION=1.13.1, USE_CUDA=ON, USE_CUDNN=ON, USE_EXCEPTION_PTR=1, USE_GFLAGS=OFF, USE_GLOG=OFF, USE_MKL=ON, USE_MKLDNN=ON, USE_MPI=OFF, USE_NCCL=ON, USE_NNPACK=ON, USE_OPENMP=ON, USE_ROCM=OFF, + + TorchVision: 0.14.1 + OpenCV: 4.7.0 + MMEngine: 0.7.3 + +Runtime environment: + cudnn_benchmark: True + mp_cfg: {'mp_start_method': 'fork', 'opencv_num_threads': 0} + dist_cfg: {'backend': 'nccl'} + seed: None + deterministic: False + Distributed launcher: slurm + Distributed training: True + GPU number: 4 +------------------------------------------------------------ + +2023/06/04 16:49:15 - mmengine - INFO - Config: +optim_wrapper = dict( + optimizer=dict( + type='AdamW', + lr=0.0004, + weight_decay=0.05, + eps=1e-08, + betas=(0.9, 0.999), + _scope_='mmpretrain'), + paramwise_cfg=dict( + norm_decay_mult=0.0, + bias_decay_mult=0.0, + flat_decay_mult=0.0, + custom_keys=dict({ + '.absolute_pos_embed': dict(decay_mult=0.0), + '.relative_position_bias_table': dict(decay_mult=0.0) + })), + type='AmpOptimWrapper', + dtype='bfloat16', + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=False, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=4096) +model = dict( + type='ImageClassifier', + backbone=dict( + frozen_stages=2, + type='ConvNeXt', + arch='small', + drop_path_rate=0.4, + init_cfg=dict( + type='Pretrained', + checkpoint= + 'https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth', + prefix='backbone')), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=768, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + init_cfg=None), + init_cfg=dict( + type='TruncNormal', layer=['Conv2d', 'Linear'], std=0.02, bias=0.0), + train_cfg=None) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stablediffusionV1-5R2-dpmsolver-25-1m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/cc1m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_5e-1' + +2023/06/04 16:49:28 - mmengine - INFO - Hooks will be executed in the following order: +before_run: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_load_checkpoint: +(ABOVE_NORMAL) EMAHook + -------------------- +before_train: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_train_epoch: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(NORMAL ) DistSamplerSeedHook + -------------------- +before_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook + -------------------- +after_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +after_train_epoch: +(NORMAL ) IterTimerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_val_epoch: +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook + -------------------- +before_val_iter: +(NORMAL ) IterTimerHook + -------------------- +after_val_iter: +(NORMAL ) IterTimerHook +(NORMAL ) VisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_val_epoch: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_save_checkpoint: +(ABOVE_NORMAL) EMAHook + -------------------- +after_train: +(VERY_LOW ) CheckpointHook + -------------------- +before_test_epoch: +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook + -------------------- +before_test_iter: +(NORMAL ) IterTimerHook + -------------------- +after_test_iter: +(NORMAL ) IterTimerHook +(NORMAL ) VisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_test_epoch: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_run: +(BELOW_NORMAL) LoggerHook + -------------------- +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.2.0.weight:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.2.0.bias:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.2.1.bias:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.3.0.weight:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.3.0.bias:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.3.1.bias:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.2.0.gamma:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.2.0.depthwise_conv.bias:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.2.0.norm.weight:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.2.0.norm.bias:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.2.0.pointwise_conv1.bias:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.2.0.pointwise_conv2.bias:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.2.1.gamma:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.2.1.depthwise_conv.bias:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.2.1.norm.weight:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.2.1.norm.bias:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.2.1.pointwise_conv1.bias:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.2.1.pointwise_conv2.bias:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.2.2.gamma:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.2.2.depthwise_conv.bias:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.2.2.norm.weight:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.2.2.norm.bias:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.2.2.pointwise_conv1.bias:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.2.2.pointwise_conv2.bias:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.2.3.gamma:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.2.3.depthwise_conv.bias:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.2.3.norm.weight:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.2.3.norm.bias:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.2.3.pointwise_conv1.bias:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.2.3.pointwise_conv2.bias:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.2.4.gamma:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.2.4.depthwise_conv.bias:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.2.4.norm.weight:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.2.4.norm.bias:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.2.4.pointwise_conv1.bias:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.2.4.pointwise_conv2.bias:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.2.5.gamma:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.2.5.depthwise_conv.bias:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.2.5.norm.weight:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.2.5.norm.bias:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.2.5.pointwise_conv1.bias:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.2.5.pointwise_conv2.bias:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.2.6.gamma:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.2.6.depthwise_conv.bias:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.2.6.norm.weight:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.2.6.norm.bias:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.2.6.pointwise_conv1.bias:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.2.6.pointwise_conv2.bias:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.2.7.gamma:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.2.7.depthwise_conv.bias:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.2.7.norm.weight:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.2.7.norm.bias:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.2.7.pointwise_conv1.bias:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.2.7.pointwise_conv2.bias:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.2.8.gamma:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.2.8.depthwise_conv.bias:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.2.8.norm.weight:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.2.8.norm.bias:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.2.8.pointwise_conv1.bias:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.2.8.pointwise_conv2.bias:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.2.9.gamma:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.2.9.depthwise_conv.bias:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.2.9.norm.weight:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.2.9.norm.bias:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.2.9.pointwise_conv1.bias:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.2.9.pointwise_conv2.bias:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.2.10.gamma:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.2.10.depthwise_conv.bias:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.2.10.norm.weight:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.2.10.norm.bias:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.2.10.pointwise_conv1.bias:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.2.10.pointwise_conv2.bias:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.2.11.gamma:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.2.11.depthwise_conv.bias:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.2.11.norm.weight:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.2.11.norm.bias:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.2.11.pointwise_conv1.bias:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.2.11.pointwise_conv2.bias:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.2.12.gamma:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.2.12.depthwise_conv.bias:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.2.12.norm.weight:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.2.12.norm.bias:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.2.12.pointwise_conv1.bias:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.2.12.pointwise_conv2.bias:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.2.13.gamma:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.2.13.depthwise_conv.bias:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.2.13.norm.weight:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.2.13.norm.bias:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.2.13.pointwise_conv1.bias:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.2.13.pointwise_conv2.bias:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.2.14.gamma:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.2.14.depthwise_conv.bias:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.2.14.norm.weight:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.2.14.norm.bias:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.2.14.pointwise_conv1.bias:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.2.14.pointwise_conv2.bias:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.2.15.gamma:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.2.15.depthwise_conv.bias:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.2.15.norm.weight:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.2.15.norm.bias:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.2.15.pointwise_conv1.bias:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.2.15.pointwise_conv2.bias:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.2.16.gamma:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.2.16.depthwise_conv.bias:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.2.16.norm.weight:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.2.16.norm.bias:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.2.16.pointwise_conv1.bias:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.2.16.pointwise_conv2.bias:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.2.17.gamma:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.2.17.depthwise_conv.bias:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.2.17.norm.weight:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.2.17.norm.bias:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.2.17.pointwise_conv1.bias:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.2.17.pointwise_conv2.bias:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.2.18.gamma:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.2.18.depthwise_conv.bias:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.2.18.norm.weight:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.2.18.norm.bias:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.2.18.pointwise_conv1.bias:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.2.18.pointwise_conv2.bias:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.2.19.gamma:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.2.19.depthwise_conv.bias:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.2.19.norm.weight:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.2.19.norm.bias:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.2.19.pointwise_conv1.bias:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.2.19.pointwise_conv2.bias:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.2.20.gamma:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.2.20.depthwise_conv.bias:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.2.20.norm.weight:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.2.20.norm.bias:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.2.20.pointwise_conv1.bias:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.2.20.pointwise_conv2.bias:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.2.21.gamma:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.2.21.depthwise_conv.bias:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.2.21.norm.weight:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.2.21.norm.bias:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.2.21.pointwise_conv1.bias:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.2.21.pointwise_conv2.bias:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.2.22.gamma:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.2.22.depthwise_conv.bias:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.2.22.norm.weight:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.2.22.norm.bias:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.2.22.pointwise_conv1.bias:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.2.22.pointwise_conv2.bias:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.2.23.gamma:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.2.23.depthwise_conv.bias:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.2.23.norm.weight:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.2.23.norm.bias:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.2.23.pointwise_conv1.bias:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.2.23.pointwise_conv2.bias:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.2.24.gamma:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.2.24.depthwise_conv.bias:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.2.24.norm.weight:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.2.24.norm.bias:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.2.24.pointwise_conv1.bias:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.2.24.pointwise_conv2.bias:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.2.25.gamma:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.2.25.depthwise_conv.bias:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.2.25.norm.weight:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.2.25.norm.bias:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.2.25.pointwise_conv1.bias:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.2.25.pointwise_conv2.bias:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.2.26.gamma:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.2.26.depthwise_conv.bias:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.2.26.norm.weight:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.2.26.norm.bias:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.2.26.pointwise_conv1.bias:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.2.26.pointwise_conv2.bias:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.3.0.gamma:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.3.0.depthwise_conv.bias:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.3.0.norm.weight:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.3.0.norm.bias:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.3.0.pointwise_conv1.bias:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.3.0.pointwise_conv2.bias:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.3.1.gamma:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.3.1.depthwise_conv.bias:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.3.1.norm.weight:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.3.1.norm.bias:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.3.1.pointwise_conv1.bias:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.3.1.pointwise_conv2.bias:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.3.2.gamma:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.3.2.depthwise_conv.bias:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.3.2.norm.weight:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.3.2.norm.bias:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.3.2.pointwise_conv1.bias:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.stages.3.2.pointwise_conv2.bias:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.norm3.weight:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- backbone.norm3.bias:weight_decay=0.0 +2023/06/04 16:49:43 - mmengine - INFO - paramwise_options -- head.fc.bias:weight_decay=0.0 +2023/06/04 16:49:49 - mmengine - INFO - load backbone in model from: https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth +Name of parameter - Initialization information + +backbone.downsample_layers.0.0.weight - torch.Size([96, 3, 4, 4]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.0.0.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.0.1.weight - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.0.1.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.1.0.weight - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.1.0.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.1.1.weight - torch.Size([192, 96, 2, 2]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.1.1.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.2.0.weight - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.2.0.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.2.1.weight - torch.Size([384, 192, 2, 2]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.2.1.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.3.0.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.3.0.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.3.1.weight - torch.Size([768, 384, 2, 2]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.3.1.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.0.gamma - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.0.depthwise_conv.weight - torch.Size([96, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.0.depthwise_conv.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.0.norm.weight - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.0.norm.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.0.pointwise_conv1.weight - torch.Size([384, 96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.0.pointwise_conv1.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.0.pointwise_conv2.weight - torch.Size([96, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.0.pointwise_conv2.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.1.gamma - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.1.depthwise_conv.weight - torch.Size([96, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.1.depthwise_conv.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.1.norm.weight - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.1.norm.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.1.pointwise_conv1.weight - torch.Size([384, 96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.1.pointwise_conv1.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.1.pointwise_conv2.weight - torch.Size([96, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.1.pointwise_conv2.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.2.gamma - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.2.depthwise_conv.weight - torch.Size([96, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.2.depthwise_conv.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.2.norm.weight - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.2.norm.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.2.pointwise_conv1.weight - torch.Size([384, 96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.2.pointwise_conv1.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.2.pointwise_conv2.weight - torch.Size([96, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.2.pointwise_conv2.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.0.gamma - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.0.depthwise_conv.weight - torch.Size([192, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.0.depthwise_conv.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.0.norm.weight - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.0.norm.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.0.pointwise_conv1.weight - torch.Size([768, 192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.0.pointwise_conv1.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.0.pointwise_conv2.weight - torch.Size([192, 768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.0.pointwise_conv2.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.1.gamma - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.1.depthwise_conv.weight - torch.Size([192, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.1.depthwise_conv.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.1.norm.weight - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.1.norm.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.1.pointwise_conv1.weight - torch.Size([768, 192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.1.pointwise_conv1.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.1.pointwise_conv2.weight - torch.Size([192, 768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.1.pointwise_conv2.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.2.gamma - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.2.depthwise_conv.weight - torch.Size([192, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.2.depthwise_conv.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.2.norm.weight - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.2.norm.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.2.pointwise_conv1.weight - torch.Size([768, 192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.2.pointwise_conv1.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.2.pointwise_conv2.weight - torch.Size([192, 768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.2.pointwise_conv2.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.0.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.0.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.0.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.0.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.0.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.0.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.0.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.0.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.0.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.1.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.1.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.1.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.1.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.1.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.1.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.1.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.1.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.1.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.2.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.2.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.2.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.2.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.2.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.2.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.2.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.2.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.2.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.3.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.3.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.3.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.3.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.3.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.3.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.3.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.3.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.3.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.4.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.4.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.4.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.4.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.4.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.4.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.4.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.4.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.4.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.5.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.5.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.5.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.5.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.5.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.5.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.5.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.5.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.5.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.6.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.6.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.6.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.6.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.6.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.6.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.6.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.6.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.6.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.7.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.7.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.7.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.7.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.7.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.7.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.7.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.7.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.7.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.8.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.8.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.8.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.8.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.8.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.8.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.8.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.8.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.8.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.9.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.9.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.9.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.9.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.9.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.9.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.9.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.9.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.9.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.10.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.10.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.10.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.10.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.10.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.10.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.10.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.10.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.10.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.11.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.11.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.11.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.11.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.11.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.11.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.11.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.11.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.11.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.12.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.12.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.12.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.12.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.12.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.12.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.12.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.12.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.12.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.13.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.13.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.13.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.13.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.13.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.13.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.13.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.13.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.13.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.14.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.14.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.14.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.14.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.14.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.14.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.14.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.14.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.14.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.15.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.15.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.15.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.15.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.15.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.15.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.15.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.15.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.15.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.16.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.16.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.16.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.16.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.16.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.16.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.16.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.16.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.16.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.17.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.17.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.17.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.17.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.17.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.17.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.17.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.17.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.17.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.18.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.18.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.18.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.18.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.18.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.18.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.18.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.18.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.18.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.19.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.19.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.19.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.19.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.19.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.19.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.19.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.19.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.19.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.20.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.20.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.20.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.20.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.20.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.20.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.20.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.20.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.20.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.21.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.21.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.21.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.21.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.21.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.21.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.21.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.21.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.21.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.22.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.22.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.22.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.22.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.22.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.22.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.22.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.22.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.22.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.23.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.23.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.23.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.23.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.23.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.23.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.23.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.23.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.23.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.24.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.24.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.24.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.24.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.24.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.24.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.24.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.24.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.24.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.25.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.25.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.25.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.25.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.25.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.25.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.25.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.25.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.25.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.26.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.26.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.26.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.26.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.26.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.26.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.26.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.26.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.26.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.0.gamma - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.0.depthwise_conv.weight - torch.Size([768, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.0.depthwise_conv.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.0.norm.weight - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.0.norm.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.0.pointwise_conv1.weight - torch.Size([3072, 768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.0.pointwise_conv1.bias - torch.Size([3072]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.0.pointwise_conv2.weight - torch.Size([768, 3072]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.0.pointwise_conv2.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.1.gamma - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.1.depthwise_conv.weight - torch.Size([768, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.1.depthwise_conv.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.1.norm.weight - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.1.norm.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.1.pointwise_conv1.weight - torch.Size([3072, 768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.1.pointwise_conv1.bias - torch.Size([3072]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.1.pointwise_conv2.weight - torch.Size([768, 3072]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.1.pointwise_conv2.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.2.gamma - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.2.depthwise_conv.weight - torch.Size([768, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.2.depthwise_conv.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.2.norm.weight - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.2.norm.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.2.pointwise_conv1.weight - torch.Size([3072, 768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.2.pointwise_conv1.bias - torch.Size([3072]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.2.pointwise_conv2.weight - torch.Size([768, 3072]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.2.pointwise_conv2.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.norm3.weight - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.norm3.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +head.fc.weight - torch.Size([2, 768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +head.fc.bias - torch.Size([2]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 +2023/06/04 16:49:49 - mmengine - WARNING - "FileClient" will be deprecated in future. Please use io functions in https://mmengine.readthedocs.io/en/latest/api/fileio.html#file-io +2023/06/04 16:49:49 - mmengine - WARNING - "HardDiskBackend" is the alias of "LocalBackend" and the former will be deprecated in future. +2023/06/04 16:49:49 - mmengine - INFO - Checkpoints will be saved to /mnt/petrelfs/luzeyu/workspace/fakebench/mmpretrain/workdir/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_5e-1. +2023/06/04 16:52:12 - mmengine - INFO - Epoch(train) [1][ 100/1954] lr: 3.9998e-04 eta: 7:42:56 time: 1.3830 data_time: 0.2749 memory: 15587 loss: 0.0448 +2023/06/04 16:54:24 - mmengine - INFO - Epoch(train) [1][ 200/1954] lr: 3.9990e-04 eta: 7:22:31 time: 1.3254 data_time: 0.0025 memory: 15587 loss: 0.0391 +2023/06/04 16:56:38 - mmengine - INFO - Epoch(train) [1][ 300/1954] lr: 3.9977e-04 eta: 7:16:23 time: 1.2638 data_time: 0.0018 memory: 15587 loss: 0.0306 +2023/06/04 16:58:54 - mmengine - INFO - Epoch(train) [1][ 400/1954] lr: 3.9960e-04 eta: 7:14:21 time: 1.4662 data_time: 0.0019 memory: 15587 loss: 0.0226 +2023/06/04 17:01:08 - mmengine - INFO - Epoch(train) [1][ 500/1954] lr: 3.9937e-04 eta: 7:10:56 time: 1.3434 data_time: 0.0018 memory: 15587 loss: 0.0164 +2023/06/04 17:03:20 - mmengine - INFO - Epoch(train) [1][ 600/1954] lr: 3.9910e-04 eta: 7:06:23 time: 1.3009 data_time: 0.0018 memory: 15587 loss: 0.0166 +2023/06/04 17:05:26 - mmengine - INFO - Epoch(train) [1][ 700/1954] lr: 3.9877e-04 eta: 7:00:10 time: 1.2781 data_time: 0.0016 memory: 15587 loss: 0.0195 +2023/06/04 17:07:29 - mmengine - INFO - Epoch(train) [1][ 800/1954] lr: 3.9839e-04 eta: 6:53:48 time: 1.1779 data_time: 0.0017 memory: 15587 loss: 0.0162 +2023/06/04 17:09:29 - mmengine - INFO - Epoch(train) [1][ 900/1954] lr: 3.9797e-04 eta: 6:47:03 time: 1.1197 data_time: 0.0018 memory: 15587 loss: 0.0111 +2023/06/04 17:11:25 - mmengine - INFO - Exp name: convnext_small_4xb256_sdv1_1m_lr4e-4_aug_5e-1_20230604_164906 +2023/06/04 17:11:25 - mmengine - INFO - Epoch(train) [1][1000/1954] lr: 3.9749e-04 eta: 6:40:17 time: 1.1227 data_time: 0.0018 memory: 15587 loss: 0.0145 +2023/06/04 17:13:19 - mmengine - INFO - Epoch(train) [1][1100/1954] lr: 3.9696e-04 eta: 6:33:45 time: 1.1131 data_time: 0.0019 memory: 15587 loss: 0.0093 +2023/06/04 17:15:11 - mmengine - INFO - Epoch(train) [1][1200/1954] lr: 3.9639e-04 eta: 6:27:28 time: 1.0852 data_time: 0.0309 memory: 15587 loss: 0.0169 +2023/06/04 17:16:58 - mmengine - INFO - Epoch(train) [1][1300/1954] lr: 3.9576e-04 eta: 6:20:53 time: 1.0916 data_time: 0.0361 memory: 15587 loss: 0.0206 +2023/06/04 17:18:45 - mmengine - INFO - Epoch(train) [1][1400/1954] lr: 3.9509e-04 eta: 6:14:43 time: 1.0548 data_time: 0.0018 memory: 15587 loss: 0.0208 +2023/06/04 17:20:29 - mmengine - INFO - Epoch(train) [1][1500/1954] lr: 3.9436e-04 eta: 6:08:42 time: 1.0407 data_time: 0.0025 memory: 15587 loss: 0.0154 +2023/06/04 17:22:12 - mmengine - INFO - Epoch(train) [1][1600/1954] lr: 3.9359e-04 eta: 6:03:04 time: 1.0215 data_time: 0.0016 memory: 15587 loss: 0.0101 +2023/06/04 17:23:54 - mmengine - INFO - Epoch(train) [1][1700/1954] lr: 3.9277e-04 eta: 5:57:36 time: 1.0217 data_time: 0.0023 memory: 15587 loss: 0.0230 +2023/06/04 17:25:33 - mmengine - INFO - Epoch(train) [1][1800/1954] lr: 3.9190e-04 eta: 5:52:08 time: 0.9366 data_time: 0.0017 memory: 15587 loss: 0.0221 +2023/06/04 17:27:09 - mmengine - INFO - Epoch(train) [1][1900/1954] lr: 3.9098e-04 eta: 5:46:36 time: 0.9413 data_time: 0.0018 memory: 15587 loss: 0.0135 +2023/06/04 17:28:03 - mmengine - INFO - Exp name: convnext_small_4xb256_sdv1_1m_lr4e-4_aug_5e-1_20230604_164906 +2023/06/04 17:28:03 - mmengine - INFO - Saving checkpoint at 1 epochs +2023/06/04 17:28:45 - mmengine - INFO - Epoch(val) [1][29/29] accuracy/top1: 88.1737 single-label/precision_classwise: [85.58110046386719, 92.20040893554688] single-label/recall_classwise: [94.4574203491211, 80.45736694335938] single-label/f1-score_classwise: [89.80045318603516, 85.9295425415039] data_time: 0.3702 time: 0.8396 +2023/06/04 17:29:36 - mmengine - INFO - Exp name: convnext_small_4xb256_sdv1_1m_lr4e-4_aug_5e-1_20230604_164906 +2023/06/04 17:30:25 - mmengine - INFO - Epoch(train) [2][ 100/1954] lr: 3.8947e-04 eta: 5:39:39 time: 0.9578 data_time: 0.0188 memory: 15587 loss: 0.0151 +2023/06/04 17:32:00 - mmengine - INFO - Epoch(train) [2][ 200/1954] lr: 3.8843e-04 eta: 5:34:50 time: 0.9782 data_time: 0.1783 memory: 15585 loss: 0.0091 +2023/06/04 17:33:35 - mmengine - INFO - Epoch(train) [2][ 300/1954] lr: 3.8735e-04 eta: 5:30:17 time: 0.9456 data_time: 0.1785 memory: 15585 loss: 0.0179 +2023/06/04 17:35:10 - mmengine - INFO - Epoch(train) [2][ 400/1954] lr: 3.8621e-04 eta: 5:25:55 time: 0.9511 data_time: 0.2252 memory: 15585 loss: 0.0132 +2023/06/04 17:36:45 - mmengine - INFO - Epoch(train) [2][ 500/1954] lr: 3.8503e-04 eta: 5:21:48 time: 0.9682 data_time: 0.1775 memory: 15585 loss: 0.0086 +2023/06/04 17:38:20 - mmengine - INFO - Epoch(train) [2][ 600/1954] lr: 3.8380e-04 eta: 5:17:55 time: 0.9981 data_time: 0.2474 memory: 15585 loss: 0.0148 +2023/06/04 17:39:54 - mmengine - INFO - Epoch(train) [2][ 700/1954] lr: 3.8253e-04 eta: 5:14:08 time: 0.9332 data_time: 0.4145 memory: 15585 loss: 0.0142 +2023/06/04 17:41:29 - mmengine - INFO - Epoch(train) [2][ 800/1954] lr: 3.8121e-04 eta: 5:10:38 time: 0.9716 data_time: 0.3266 memory: 15585 loss: 0.0129 +2023/06/04 17:43:05 - mmengine - INFO - Epoch(train) [2][ 900/1954] lr: 3.7984e-04 eta: 5:07:16 time: 0.9721 data_time: 0.3410 memory: 15585 loss: 0.0095 +2023/06/04 17:44:42 - mmengine - INFO - Epoch(train) [2][1000/1954] lr: 3.7843e-04 eta: 5:04:11 time: 0.9364 data_time: 0.4131 memory: 15585 loss: 0.0109 +2023/06/04 17:45:27 - mmengine - INFO - Exp name: convnext_small_4xb256_sdv1_1m_lr4e-4_aug_5e-1_20230604_164906 +2023/06/04 17:46:16 - mmengine - INFO - Epoch(train) [2][1100/1954] lr: 3.7698e-04 eta: 5:00:56 time: 0.9517 data_time: 0.4178 memory: 15585 loss: 0.0100 +2023/06/04 17:47:52 - mmengine - INFO - Epoch(train) [2][1200/1954] lr: 3.7548e-04 eta: 4:57:54 time: 0.9602 data_time: 0.4261 memory: 15585 loss: 0.0100 +2023/06/04 17:49:29 - mmengine - INFO - Epoch(train) [2][1300/1954] lr: 3.7393e-04 eta: 4:55:04 time: 1.0086 data_time: 0.4900 memory: 15585 loss: 0.0109 +2023/06/04 17:51:04 - mmengine - INFO - Epoch(train) [2][1400/1954] lr: 3.7234e-04 eta: 4:52:11 time: 0.9537 data_time: 0.4330 memory: 15585 loss: 0.0158 +2023/06/04 17:52:42 - mmengine - INFO - Epoch(train) [2][1500/1954] lr: 3.7071e-04 eta: 4:49:35 time: 1.0087 data_time: 0.4909 memory: 15585 loss: 0.0152 +2023/06/04 17:54:18 - mmengine - INFO - Epoch(train) [2][1600/1954] lr: 3.6904e-04 eta: 4:46:51 time: 0.9542 data_time: 0.4243 memory: 15585 loss: 0.0145 +2023/06/04 17:55:54 - mmengine - INFO - Epoch(train) [2][1700/1954] lr: 3.6732e-04 eta: 4:44:11 time: 0.9569 data_time: 0.4355 memory: 15585 loss: 0.0077 +2023/06/04 17:57:30 - mmengine - INFO - Epoch(train) [2][1800/1954] lr: 3.6557e-04 eta: 4:41:40 time: 0.9421 data_time: 0.4241 memory: 15585 loss: 0.0131 +2023/06/04 17:59:04 - mmengine - INFO - Epoch(train) [2][1900/1954] lr: 3.6377e-04 eta: 4:38:58 time: 0.9942 data_time: 0.4752 memory: 15585 loss: 0.0216 +2023/06/04 17:59:58 - mmengine - INFO - Exp name: convnext_small_4xb256_sdv1_1m_lr4e-4_aug_5e-1_20230604_164906 +2023/06/04 17:59:58 - mmengine - INFO - Saving checkpoint at 2 epochs +2023/06/04 18:00:37 - mmengine - INFO - Epoch(val) [2][29/29] accuracy/top1: 94.7671 single-label/precision_classwise: [91.81102752685547, 99.13763427734375] single-label/recall_classwise: [99.36872863769531, 89.11627960205078] single-label/f1-score_classwise: [95.44049835205078, 93.8602294921875] data_time: 0.3209 time: 0.7104 +2023/06/04 18:02:14 - mmengine - INFO - Exp name: convnext_small_4xb256_sdv1_1m_lr4e-4_aug_5e-1_20230604_164906 +2023/06/04 18:02:18 - mmengine - INFO - Epoch(train) [3][ 100/1954] lr: 3.6091e-04 eta: 4:35:37 time: 1.0242 data_time: 0.0026 memory: 15585 loss: 0.0092 +2023/06/04 18:03:52 - mmengine - INFO - Epoch(train) [3][ 200/1954] lr: 3.5901e-04 eta: 4:33:05 time: 0.9418 data_time: 0.0063 memory: 15585 loss: 0.0046 +2023/06/04 18:05:28 - mmengine - INFO - Epoch(train) [3][ 300/1954] lr: 3.5707e-04 eta: 4:30:41 time: 0.9086 data_time: 0.0020 memory: 15585 loss: 0.0128 +2023/06/04 18:07:04 - mmengine - INFO - Epoch(train) [3][ 400/1954] lr: 3.5509e-04 eta: 4:28:19 time: 1.0007 data_time: 0.0035 memory: 15585 loss: 0.0132 +2023/06/04 18:08:41 - mmengine - INFO - Epoch(train) [3][ 500/1954] lr: 3.5307e-04 eta: 4:26:05 time: 0.9657 data_time: 0.0026 memory: 15585 loss: 0.0109 +2023/06/04 18:10:17 - mmengine - INFO - Epoch(train) [3][ 600/1954] lr: 3.5101e-04 eta: 4:23:47 time: 0.9249 data_time: 0.0023 memory: 15585 loss: 0.0055 +2023/06/04 18:11:51 - mmengine - INFO - Epoch(train) [3][ 700/1954] lr: 3.4891e-04 eta: 4:21:26 time: 0.9141 data_time: 0.0017 memory: 15585 loss: 0.0058 +2023/06/04 18:13:29 - mmengine - INFO - Epoch(train) [3][ 800/1954] lr: 3.4677e-04 eta: 4:19:18 time: 0.9485 data_time: 0.0020 memory: 15585 loss: 0.0103 +2023/06/04 18:15:05 - mmengine - INFO - Epoch(train) [3][ 900/1954] lr: 3.4460e-04 eta: 4:17:06 time: 0.9205 data_time: 0.0020 memory: 15585 loss: 0.0176 +2023/06/04 18:16:41 - mmengine - INFO - Epoch(train) [3][1000/1954] lr: 3.4240e-04 eta: 4:14:55 time: 0.9256 data_time: 0.0105 memory: 15585 loss: 0.0117 +2023/06/04 18:18:12 - mmengine - INFO - Exp name: convnext_small_4xb256_sdv1_1m_lr4e-4_aug_5e-1_20230604_164906 +2023/06/04 18:18:18 - mmengine - INFO - Epoch(train) [3][1100/1954] lr: 3.4015e-04 eta: 4:12:47 time: 0.9667 data_time: 0.1481 memory: 15585 loss: 0.0096 +2023/06/04 18:19:52 - mmengine - INFO - Epoch(train) [3][1200/1954] lr: 3.3788e-04 eta: 4:10:35 time: 0.9566 data_time: 0.0562 memory: 15585 loss: 0.0107 +2023/06/04 18:21:28 - mmengine - INFO - Epoch(train) [3][1300/1954] lr: 3.3557e-04 eta: 4:08:27 time: 0.9531 data_time: 0.1323 memory: 15585 loss: 0.0093 +2023/06/04 18:23:04 - mmengine - INFO - Epoch(train) [3][1400/1954] lr: 3.3322e-04 eta: 4:06:21 time: 0.9326 data_time: 0.0349 memory: 15585 loss: 0.0097 +2023/06/04 18:24:40 - mmengine - INFO - Epoch(train) [3][1500/1954] lr: 3.3084e-04 eta: 4:04:17 time: 0.9965 data_time: 0.0429 memory: 15585 loss: 0.0156 +2023/06/04 18:26:15 - mmengine - INFO - Epoch(train) [3][1600/1954] lr: 3.2843e-04 eta: 4:02:11 time: 0.9147 data_time: 0.1175 memory: 15585 loss: 0.0100 +2023/06/04 18:27:50 - mmengine - INFO - Epoch(train) [3][1700/1954] lr: 3.2599e-04 eta: 4:00:06 time: 0.9154 data_time: 0.1914 memory: 15585 loss: 0.0054 +2023/06/04 18:29:26 - mmengine - INFO - Epoch(train) [3][1800/1954] lr: 3.2351e-04 eta: 3:58:05 time: 0.9449 data_time: 0.2983 memory: 15585 loss: 0.0070 +2023/06/04 18:31:01 - mmengine - INFO - Epoch(train) [3][1900/1954] lr: 3.2101e-04 eta: 3:56:02 time: 0.9196 data_time: 0.1253 memory: 15585 loss: 0.0095 +2023/06/04 18:31:52 - mmengine - INFO - Exp name: convnext_small_4xb256_sdv1_1m_lr4e-4_aug_5e-1_20230604_164906 +2023/06/04 18:31:52 - mmengine - INFO - Saving checkpoint at 3 epochs +2023/06/04 18:32:31 - mmengine - INFO - Epoch(val) [3][29/29] accuracy/top1: 97.4009 single-label/precision_classwise: [95.66743469238281, 99.7462158203125] single-label/recall_classwise: [99.80430603027344, 94.4496078491211] single-label/f1-score_classwise: [97.69209289550781, 97.02568054199219] data_time: 0.2821 time: 0.6703 +2023/06/04 18:34:11 - mmengine - INFO - Epoch(train) [4][ 100/1954] lr: 3.1709e-04 eta: 3:53:08 time: 0.9506 data_time: 0.1117 memory: 15585 loss: 0.0065 +2023/06/04 18:34:49 - mmengine - INFO - Exp name: convnext_small_4xb256_sdv1_1m_lr4e-4_aug_5e-1_20230604_164906 +2023/06/04 18:35:48 - mmengine - INFO - Epoch(train) [4][ 200/1954] lr: 3.1451e-04 eta: 3:51:10 time: 0.9671 data_time: 0.0018 memory: 15585 loss: 0.0076 +2023/06/04 18:37:23 - mmengine - INFO - Epoch(train) [4][ 300/1954] lr: 3.1190e-04 eta: 3:49:11 time: 1.0033 data_time: 0.0015 memory: 15585 loss: 0.0116 +2023/06/04 18:38:58 - mmengine - INFO - Epoch(train) [4][ 400/1954] lr: 3.0927e-04 eta: 3:47:10 time: 0.9540 data_time: 0.0017 memory: 15585 loss: 0.0067 +2023/06/04 18:40:35 - mmengine - INFO - Epoch(train) [4][ 500/1954] lr: 3.0661e-04 eta: 3:45:18 time: 0.9688 data_time: 0.0026 memory: 15585 loss: 0.0111 +2023/06/04 18:42:12 - mmengine - INFO - Epoch(train) [4][ 600/1954] lr: 3.0392e-04 eta: 3:43:23 time: 0.9817 data_time: 0.0019 memory: 15585 loss: 0.0084 +2023/06/04 18:43:47 - mmengine - INFO - Epoch(train) [4][ 700/1954] lr: 3.0120e-04 eta: 3:41:27 time: 0.9767 data_time: 0.0018 memory: 15585 loss: 0.0097 +2023/06/04 18:45:23 - mmengine - INFO - Epoch(train) [4][ 800/1954] lr: 2.9846e-04 eta: 3:39:30 time: 0.9105 data_time: 0.0356 memory: 15585 loss: 0.0037 +2023/06/04 18:46:59 - mmengine - INFO - Epoch(train) [4][ 900/1954] lr: 2.9570e-04 eta: 3:37:37 time: 0.9805 data_time: 0.1314 memory: 15585 loss: 0.0077 +2023/06/04 18:48:37 - mmengine - INFO - Epoch(train) [4][1000/1954] lr: 2.9291e-04 eta: 3:35:47 time: 1.0860 data_time: 0.5424 memory: 15585 loss: 0.0054 +2023/06/04 18:50:12 - mmengine - INFO - Epoch(train) [4][1100/1954] lr: 2.9010e-04 eta: 3:33:53 time: 0.9708 data_time: 0.4517 memory: 15585 loss: 0.0079 +2023/06/04 18:50:46 - mmengine - INFO - Exp name: convnext_small_4xb256_sdv1_1m_lr4e-4_aug_5e-1_20230604_164906 +2023/06/04 18:51:49 - mmengine - INFO - Epoch(train) [4][1200/1954] lr: 2.8727e-04 eta: 3:32:02 time: 0.9507 data_time: 0.4189 memory: 15585 loss: 0.0056 +2023/06/04 18:53:26 - mmengine - INFO - Epoch(train) [4][1300/1954] lr: 2.8442e-04 eta: 3:30:10 time: 0.9366 data_time: 0.4176 memory: 15585 loss: 0.0062 +2023/06/04 18:55:04 - mmengine - INFO - Epoch(train) [4][1400/1954] lr: 2.8155e-04 eta: 3:28:22 time: 1.0263 data_time: 0.4970 memory: 15585 loss: 0.0049 +2023/06/04 18:56:40 - mmengine - INFO - Epoch(train) [4][1500/1954] lr: 2.7865e-04 eta: 3:26:31 time: 0.9628 data_time: 0.4418 memory: 15585 loss: 0.0049 +2023/06/04 18:58:18 - mmengine - INFO - Epoch(train) [4][1600/1954] lr: 2.7574e-04 eta: 3:24:43 time: 1.0450 data_time: 0.5250 memory: 15585 loss: 0.0073 +2023/06/04 18:59:55 - mmengine - INFO - Epoch(train) [4][1700/1954] lr: 2.7281e-04 eta: 3:22:53 time: 0.9490 data_time: 0.4268 memory: 15585 loss: 0.0124 +2023/06/04 19:01:30 - mmengine - INFO - Epoch(train) [4][1800/1954] lr: 2.6986e-04 eta: 3:21:02 time: 0.9444 data_time: 0.4164 memory: 15585 loss: 0.0067 +2023/06/04 19:03:08 - mmengine - INFO - Epoch(train) [4][1900/1954] lr: 2.6690e-04 eta: 3:19:14 time: 0.9947 data_time: 0.4746 memory: 15585 loss: 0.0075 +2023/06/04 19:03:57 - mmengine - INFO - Exp name: convnext_small_4xb256_sdv1_1m_lr4e-4_aug_5e-1_20230604_164906 +2023/06/04 19:03:57 - mmengine - INFO - Saving checkpoint at 4 epochs +2023/06/04 19:04:36 - mmengine - INFO - Epoch(val) [4][29/29] accuracy/top1: 98.7927 single-label/precision_classwise: [97.927490234375, 99.90458679199219] single-label/recall_classwise: [99.92424774169922, 97.40310668945312] single-label/f1-score_classwise: [98.9157943725586, 98.63799285888672] data_time: 0.3012 time: 0.6896 +2023/06/04 19:06:18 - mmengine - INFO - Epoch(train) [5][ 100/1954] lr: 2.6230e-04 eta: 3:16:31 time: 0.9828 data_time: 0.0769 memory: 15585 loss: 0.0053 +2023/06/04 19:07:42 - mmengine - INFO - Exp name: convnext_small_4xb256_sdv1_1m_lr4e-4_aug_5e-1_20230604_164906 +2023/06/04 19:07:55 - mmengine - INFO - Epoch(train) [5][ 200/1954] lr: 2.5929e-04 eta: 3:14:43 time: 0.9394 data_time: 0.1465 memory: 15585 loss: 0.0072 +2023/06/04 19:09:31 - mmengine - INFO - Epoch(train) [5][ 300/1954] lr: 2.5628e-04 eta: 3:12:55 time: 0.9928 data_time: 0.0019 memory: 15585 loss: 0.0047 +2023/06/04 19:11:08 - mmengine - INFO - Epoch(train) [5][ 400/1954] lr: 2.5325e-04 eta: 3:11:07 time: 0.9198 data_time: 0.0263 memory: 15585 loss: 0.0052 +2023/06/04 19:12:44 - mmengine - INFO - Epoch(train) [5][ 500/1954] lr: 2.5020e-04 eta: 3:09:19 time: 0.9698 data_time: 0.0562 memory: 15585 loss: 0.0049 +2023/06/04 19:14:24 - mmengine - INFO - Epoch(train) [5][ 600/1954] lr: 2.4715e-04 eta: 3:07:36 time: 0.9476 data_time: 0.1659 memory: 15585 loss: 0.0043 +2023/06/04 19:16:00 - mmengine - INFO - Epoch(train) [5][ 700/1954] lr: 2.4408e-04 eta: 3:05:48 time: 0.9713 data_time: 0.2657 memory: 15585 loss: 0.0052 +2023/06/04 19:17:36 - mmengine - INFO - Epoch(train) [5][ 800/1954] lr: 2.4100e-04 eta: 3:04:01 time: 0.9430 data_time: 0.3135 memory: 15585 loss: 0.0069 +2023/06/04 19:19:13 - mmengine - INFO - Epoch(train) [5][ 900/1954] lr: 2.3792e-04 eta: 3:02:15 time: 0.9764 data_time: 0.4483 memory: 15585 loss: 0.0043 +2023/06/04 19:20:50 - mmengine - INFO - Epoch(train) [5][1000/1954] lr: 2.3482e-04 eta: 3:00:29 time: 0.9895 data_time: 0.4581 memory: 15585 loss: 0.0067 +2023/06/04 19:22:27 - mmengine - INFO - Epoch(train) [5][1100/1954] lr: 2.3172e-04 eta: 2:58:43 time: 0.9701 data_time: 0.4457 memory: 15585 loss: 0.0059 +2023/06/04 19:23:46 - mmengine - INFO - Exp name: convnext_small_4xb256_sdv1_1m_lr4e-4_aug_5e-1_20230604_164906 +2023/06/04 19:24:04 - mmengine - INFO - Epoch(train) [5][1200/1954] lr: 2.2861e-04 eta: 2:56:57 time: 1.0005 data_time: 0.4592 memory: 15585 loss: 0.0056 +2023/06/04 19:25:40 - mmengine - INFO - Epoch(train) [5][1300/1954] lr: 2.2550e-04 eta: 2:55:11 time: 0.9330 data_time: 0.4081 memory: 15585 loss: 0.0056 +2023/06/04 19:27:17 - mmengine - INFO - Epoch(train) [5][1400/1954] lr: 2.2238e-04 eta: 2:53:26 time: 0.9705 data_time: 0.4501 memory: 15585 loss: 0.0035 +2023/06/04 19:28:53 - mmengine - INFO - Epoch(train) [5][1500/1954] lr: 2.1925e-04 eta: 2:51:40 time: 0.9473 data_time: 0.4204 memory: 15585 loss: 0.0069 +2023/06/04 19:30:28 - mmengine - INFO - Epoch(train) [5][1600/1954] lr: 2.1612e-04 eta: 2:49:53 time: 0.9537 data_time: 0.3882 memory: 15585 loss: 0.0060 +2023/06/04 19:32:05 - mmengine - INFO - Epoch(train) [5][1700/1954] lr: 2.1299e-04 eta: 2:48:08 time: 0.9413 data_time: 0.4218 memory: 15585 loss: 0.0039 +2023/06/04 19:33:39 - mmengine - INFO - Epoch(train) [5][1800/1954] lr: 2.0986e-04 eta: 2:46:21 time: 0.9768 data_time: 0.4113 memory: 15585 loss: 0.0051 +2023/06/04 19:35:14 - mmengine - INFO - Epoch(train) [5][1900/1954] lr: 2.0672e-04 eta: 2:44:35 time: 0.9532 data_time: 0.3408 memory: 15585 loss: 0.0052 +2023/06/04 19:36:04 - mmengine - INFO - Exp name: convnext_small_4xb256_sdv1_1m_lr4e-4_aug_5e-1_20230604_164906 +2023/06/04 19:36:04 - mmengine - INFO - Saving checkpoint at 5 epochs +2023/06/04 19:36:42 - mmengine - INFO - Epoch(val) [5][29/29] accuracy/top1: 99.4050 single-label/precision_classwise: [98.97486877441406, 99.945068359375] single-label/recall_classwise: [99.955810546875, 98.72868347167969] single-label/f1-score_classwise: [99.46292114257812, 99.3331527709961] data_time: 0.2663 time: 0.6586 +2023/06/04 19:38:22 - mmengine - INFO - Epoch(train) [6][ 100/1954] lr: 2.0190e-04 eta: 2:41:55 time: 0.9475 data_time: 0.0017 memory: 15585 loss: 0.0035 +2023/06/04 19:39:58 - mmengine - INFO - Epoch(train) [6][ 200/1954] lr: 1.9876e-04 eta: 2:40:10 time: 0.9519 data_time: 0.0024 memory: 15585 loss: 0.0059 +2023/06/04 19:40:27 - mmengine - INFO - Exp name: convnext_small_4xb256_sdv1_1m_lr4e-4_aug_5e-1_20230604_164906 +2023/06/04 19:41:33 - mmengine - INFO - Epoch(train) [6][ 300/1954] lr: 1.9563e-04 eta: 2:38:25 time: 0.9515 data_time: 0.0030 memory: 15585 loss: 0.0071 +2023/06/04 19:43:07 - mmengine - INFO - Epoch(train) [6][ 400/1954] lr: 1.9250e-04 eta: 2:36:38 time: 0.9630 data_time: 0.0027 memory: 15585 loss: 0.0045 +2023/06/04 19:44:42 - mmengine - INFO - Epoch(train) [6][ 500/1954] lr: 1.8937e-04 eta: 2:34:53 time: 0.9069 data_time: 0.0019 memory: 15585 loss: 0.0040 +2023/06/04 19:46:17 - mmengine - INFO - Epoch(train) [6][ 600/1954] lr: 1.8625e-04 eta: 2:33:08 time: 0.9343 data_time: 0.0016 memory: 15585 loss: 0.0085 +2023/06/04 19:47:52 - mmengine - INFO - Epoch(train) [6][ 700/1954] lr: 1.8313e-04 eta: 2:31:24 time: 0.9232 data_time: 0.4035 memory: 15585 loss: 0.0044 +2023/06/04 19:49:28 - mmengine - INFO - Epoch(train) [6][ 800/1954] lr: 1.8002e-04 eta: 2:29:40 time: 0.9168 data_time: 0.3982 memory: 15585 loss: 0.0029 +2023/06/04 19:51:01 - mmengine - INFO - Epoch(train) [6][ 900/1954] lr: 1.7691e-04 eta: 2:27:54 time: 0.9349 data_time: 0.4151 memory: 15585 loss: 0.0043 +2023/06/04 19:52:35 - mmengine - INFO - Epoch(train) [6][1000/1954] lr: 1.7381e-04 eta: 2:26:09 time: 0.9523 data_time: 0.4240 memory: 15585 loss: 0.0050 +2023/06/04 19:54:11 - mmengine - INFO - Epoch(train) [6][1100/1954] lr: 1.7072e-04 eta: 2:24:26 time: 0.9581 data_time: 0.4349 memory: 15585 loss: 0.0037 +2023/06/04 19:55:48 - mmengine - INFO - Epoch(train) [6][1200/1954] lr: 1.6764e-04 eta: 2:22:43 time: 0.9355 data_time: 0.4164 memory: 15585 loss: 0.0036 +2023/06/04 19:56:17 - mmengine - INFO - Exp name: convnext_small_4xb256_sdv1_1m_lr4e-4_aug_5e-1_20230604_164906 +2023/06/04 19:57:24 - mmengine - INFO - Epoch(train) [6][1300/1954] lr: 1.6457e-04 eta: 2:21:00 time: 0.9170 data_time: 0.3972 memory: 15585 loss: 0.0022 +2023/06/04 19:59:00 - mmengine - INFO - Epoch(train) [6][1400/1954] lr: 1.6151e-04 eta: 2:19:18 time: 0.9628 data_time: 0.4432 memory: 15585 loss: 0.0022 +2023/06/04 20:00:35 - mmengine - INFO - Epoch(train) [6][1500/1954] lr: 1.5846e-04 eta: 2:17:35 time: 0.9256 data_time: 0.4077 memory: 15585 loss: 0.0041 +2023/06/04 20:02:09 - mmengine - INFO - Epoch(train) [6][1600/1954] lr: 1.5542e-04 eta: 2:15:51 time: 0.9689 data_time: 0.4408 memory: 15585 loss: 0.0022 +2023/06/04 20:03:45 - mmengine - INFO - Epoch(train) [6][1700/1954] lr: 1.5239e-04 eta: 2:14:08 time: 0.9478 data_time: 0.4262 memory: 15585 loss: 0.0066 +2023/06/04 20:05:21 - mmengine - INFO - Epoch(train) [6][1800/1954] lr: 1.4938e-04 eta: 2:12:25 time: 0.9390 data_time: 0.4163 memory: 15585 loss: 0.0034 +2023/06/04 20:06:56 - mmengine - INFO - Epoch(train) [6][1900/1954] lr: 1.4638e-04 eta: 2:10:43 time: 0.9382 data_time: 0.4195 memory: 15585 loss: 0.0010 +2023/06/04 20:07:51 - mmengine - INFO - Exp name: convnext_small_4xb256_sdv1_1m_lr4e-4_aug_5e-1_20230604_164906 +2023/06/04 20:07:51 - mmengine - INFO - Saving checkpoint at 6 epochs +2023/06/04 20:08:28 - mmengine - INFO - Epoch(val) [6][29/29] accuracy/top1: 99.6208 single-label/precision_classwise: [99.37853240966797, 99.92194366455078] single-label/recall_classwise: [99.93687438964844, 99.23255920410156] single-label/f1-score_classwise: [99.65692901611328, 99.5760498046875] data_time: 0.2607 time: 0.6484 +2023/06/04 20:10:07 - mmengine - INFO - Epoch(train) [7][ 100/1954] lr: 1.4180e-04 eta: 2:08:09 time: 0.9159 data_time: 0.0017 memory: 15585 loss: 0.0055 +2023/06/04 20:11:43 - mmengine - INFO - Epoch(train) [7][ 200/1954] lr: 1.3884e-04 eta: 2:06:28 time: 0.9376 data_time: 0.0019 memory: 15585 loss: 0.0062 +2023/06/04 20:12:58 - mmengine - INFO - Exp name: convnext_small_4xb256_sdv1_1m_lr4e-4_aug_5e-1_20230604_164906 +2023/06/04 20:13:24 - mmengine - INFO - Epoch(train) [7][ 300/1954] lr: 1.3590e-04 eta: 2:04:49 time: 1.4418 data_time: 0.0028 memory: 15585 loss: 0.0022 +2023/06/04 20:15:01 - mmengine - INFO - Epoch(train) [7][ 400/1954] lr: 1.3298e-04 eta: 2:03:07 time: 0.9647 data_time: 0.0027 memory: 15585 loss: 0.0024 +2023/06/04 20:16:34 - mmengine - INFO - Epoch(train) [7][ 500/1954] lr: 1.3007e-04 eta: 2:01:24 time: 0.9379 data_time: 0.0024 memory: 15585 loss: 0.0024 +2023/06/04 20:18:09 - mmengine - INFO - Epoch(train) [7][ 600/1954] lr: 1.2719e-04 eta: 1:59:41 time: 0.9883 data_time: 0.0018 memory: 15585 loss: 0.0034 +2023/06/04 20:19:42 - mmengine - INFO - Epoch(train) [7][ 700/1954] lr: 1.2432e-04 eta: 1:57:59 time: 0.9463 data_time: 0.0026 memory: 15585 loss: 0.0051 +2023/06/04 20:21:19 - mmengine - INFO - Epoch(train) [7][ 800/1954] lr: 1.2148e-04 eta: 1:56:17 time: 0.9885 data_time: 0.0019 memory: 15585 loss: 0.0032 +2023/06/04 20:22:57 - mmengine - INFO - Epoch(train) [7][ 900/1954] lr: 1.1866e-04 eta: 1:54:37 time: 0.9857 data_time: 0.0017 memory: 15585 loss: 0.0028 +2023/06/04 20:24:32 - mmengine - INFO - Epoch(train) [7][1000/1954] lr: 1.1586e-04 eta: 1:52:55 time: 0.9341 data_time: 0.0205 memory: 15585 loss: 0.0008 +2023/06/04 20:26:08 - mmengine - INFO - Epoch(train) [7][1100/1954] lr: 1.1308e-04 eta: 1:51:15 time: 0.9964 data_time: 0.1337 memory: 15585 loss: 0.0050 +2023/06/04 20:27:44 - mmengine - INFO - Epoch(train) [7][1200/1954] lr: 1.1033e-04 eta: 1:49:33 time: 0.9258 data_time: 0.0095 memory: 15585 loss: 0.0036 +2023/06/04 20:28:58 - mmengine - INFO - Exp name: convnext_small_4xb256_sdv1_1m_lr4e-4_aug_5e-1_20230604_164906 +2023/06/04 20:29:20 - mmengine - INFO - Epoch(train) [7][1300/1954] lr: 1.0760e-04 eta: 1:47:52 time: 0.9202 data_time: 0.0786 memory: 15585 loss: 0.0016 +2023/06/04 20:30:57 - mmengine - INFO - Epoch(train) [7][1400/1954] lr: 1.0490e-04 eta: 1:46:12 time: 1.0145 data_time: 0.2402 memory: 15585 loss: 0.0069 +2023/06/04 20:32:35 - mmengine - INFO - Epoch(train) [7][1500/1954] lr: 1.0222e-04 eta: 1:44:32 time: 0.9803 data_time: 0.4605 memory: 15585 loss: 0.0052 +2023/06/04 20:34:12 - mmengine - INFO - Epoch(train) [7][1600/1954] lr: 9.9568e-05 eta: 1:42:51 time: 0.9267 data_time: 0.3975 memory: 15585 loss: 0.0025 +2023/06/04 20:35:50 - mmengine - INFO - Epoch(train) [7][1700/1954] lr: 9.6944e-05 eta: 1:41:11 time: 0.9405 data_time: 0.4211 memory: 15585 loss: 0.0022 +2023/06/04 20:37:25 - mmengine - INFO - Epoch(train) [7][1800/1954] lr: 9.4348e-05 eta: 1:39:30 time: 0.8859 data_time: 0.3621 memory: 15585 loss: 0.0046 +2023/06/04 20:39:09 - mmengine - INFO - Epoch(train) [7][1900/1954] lr: 9.1781e-05 eta: 1:37:53 time: 1.1055 data_time: 0.5840 memory: 15585 loss: 0.0020 +2023/06/04 20:40:04 - mmengine - INFO - Exp name: convnext_small_4xb256_sdv1_1m_lr4e-4_aug_5e-1_20230604_164906 +2023/06/04 20:40:04 - mmengine - INFO - Saving checkpoint at 7 epochs +2023/06/04 20:40:47 - mmengine - INFO - Epoch(val) [7][29/29] accuracy/top1: 99.7182 single-label/precision_classwise: [99.57846069335938, 99.89102172851562] single-label/recall_classwise: [99.91162109375, 99.48062133789062] single-label/f1-score_classwise: [99.74476623535156, 99.68539428710938] data_time: 0.4317 time: 0.8270 +2023/06/04 20:42:21 - mmengine - INFO - Epoch(train) [8][ 100/1954] lr: 8.7885e-05 eta: 1:35:18 time: 0.9320 data_time: 0.0018 memory: 15585 loss: 0.0020 +2023/06/04 20:43:53 - mmengine - INFO - Epoch(train) [8][ 200/1954] lr: 8.5394e-05 eta: 1:33:36 time: 0.9013 data_time: 0.0019 memory: 15585 loss: 0.0039 +2023/06/04 20:45:26 - mmengine - INFO - Epoch(train) [8][ 300/1954] lr: 8.2933e-05 eta: 1:31:54 time: 0.9606 data_time: 0.0020 memory: 15585 loss: 0.0035 +2023/06/04 20:45:50 - mmengine - INFO - Exp name: convnext_small_4xb256_sdv1_1m_lr4e-4_aug_5e-1_20230604_164906 +2023/06/04 20:47:01 - mmengine - INFO - Epoch(train) [8][ 400/1954] lr: 8.0504e-05 eta: 1:30:14 time: 0.9543 data_time: 0.0017 memory: 15585 loss: 0.0014 +2023/06/04 20:48:37 - mmengine - INFO - Epoch(train) [8][ 500/1954] lr: 7.8107e-05 eta: 1:28:34 time: 0.9691 data_time: 0.0016 memory: 15585 loss: 0.0009 +2023/06/04 20:50:13 - mmengine - INFO - Epoch(train) [8][ 600/1954] lr: 7.5743e-05 eta: 1:26:53 time: 1.0130 data_time: 0.0021 memory: 15585 loss: 0.0040 +2023/06/04 20:51:49 - mmengine - INFO - Epoch(train) [8][ 700/1954] lr: 7.3413e-05 eta: 1:25:13 time: 0.9794 data_time: 0.0030 memory: 15585 loss: 0.0080 +2023/06/04 20:53:27 - mmengine - INFO - Epoch(train) [8][ 800/1954] lr: 7.1116e-05 eta: 1:23:33 time: 0.9529 data_time: 0.0026 memory: 15585 loss: 0.0022 +2023/06/04 20:55:04 - mmengine - INFO - Epoch(train) [8][ 900/1954] lr: 6.8854e-05 eta: 1:21:54 time: 0.9731 data_time: 0.2287 memory: 15585 loss: 0.0026 +2023/06/04 20:56:43 - mmengine - INFO - Epoch(train) [8][1000/1954] lr: 6.6627e-05 eta: 1:20:15 time: 0.9697 data_time: 0.4028 memory: 15585 loss: 0.0026 +2023/06/04 20:58:20 - mmengine - INFO - Epoch(train) [8][1100/1954] lr: 6.4436e-05 eta: 1:18:35 time: 0.9482 data_time: 0.4285 memory: 15585 loss: 0.0008 +2023/06/04 20:59:57 - mmengine - INFO - Epoch(train) [8][1200/1954] lr: 6.2282e-05 eta: 1:16:55 time: 0.9466 data_time: 0.4181 memory: 15585 loss: 0.0023 +2023/06/04 21:01:33 - mmengine - INFO - Epoch(train) [8][1300/1954] lr: 6.0164e-05 eta: 1:15:15 time: 0.9452 data_time: 0.4267 memory: 15585 loss: 0.0039 +2023/06/04 21:01:53 - mmengine - INFO - Exp name: convnext_small_4xb256_sdv1_1m_lr4e-4_aug_5e-1_20230604_164906 +2023/06/04 21:03:11 - mmengine - INFO - Epoch(train) [8][1400/1954] lr: 5.8083e-05 eta: 1:13:36 time: 0.9513 data_time: 0.4155 memory: 15585 loss: 0.0023 +2023/06/04 21:04:48 - mmengine - INFO - Epoch(train) [8][1500/1954] lr: 5.6041e-05 eta: 1:11:57 time: 0.9896 data_time: 0.4702 memory: 15585 loss: 0.0015 +2023/06/04 21:06:24 - mmengine - INFO - Epoch(train) [8][1600/1954] lr: 5.4037e-05 eta: 1:10:17 time: 0.9795 data_time: 0.4400 memory: 15585 loss: 0.0047 +2023/06/04 21:08:02 - mmengine - INFO - Epoch(train) [8][1700/1954] lr: 5.2072e-05 eta: 1:08:38 time: 0.9804 data_time: 0.4571 memory: 15585 loss: 0.0036 +2023/06/04 21:09:39 - mmengine - INFO - Epoch(train) [8][1800/1954] lr: 5.0147e-05 eta: 1:06:58 time: 0.9743 data_time: 0.4313 memory: 15585 loss: 0.0015 +2023/06/04 21:11:17 - mmengine - INFO - Epoch(train) [8][1900/1954] lr: 4.8261e-05 eta: 1:05:19 time: 0.9832 data_time: 0.4610 memory: 15585 loss: 0.0055 +2023/06/04 21:12:05 - mmengine - INFO - Exp name: convnext_small_4xb256_sdv1_1m_lr4e-4_aug_5e-1_20230604_164906 +2023/06/04 21:12:05 - mmengine - INFO - Saving checkpoint at 8 epochs +2023/06/04 21:12:43 - mmengine - INFO - Epoch(val) [8][29/29] accuracy/top1: 99.7773 single-label/precision_classwise: [99.65380096435547, 99.92998504638672] single-label/recall_classwise: [99.94318389892578, 99.57363891601562] single-label/f1-score_classwise: [99.79827880859375, 99.75149536132812] data_time: 0.2709 time: 0.6686 +2023/06/04 21:14:24 - mmengine - INFO - Epoch(train) [9][ 100/1954] lr: 4.5437e-05 eta: 1:02:46 time: 0.9479 data_time: 0.0024 memory: 15585 loss: 0.0003 +2023/06/04 21:16:00 - mmengine - INFO - Epoch(train) [9][ 200/1954] lr: 4.3656e-05 eta: 1:01:06 time: 0.9706 data_time: 0.0026 memory: 15585 loss: 0.0056 +2023/06/04 21:17:34 - mmengine - INFO - Epoch(train) [9][ 300/1954] lr: 4.1916e-05 eta: 0:59:26 time: 0.9527 data_time: 0.0028 memory: 15585 loss: 0.0010 +2023/06/04 21:18:39 - mmengine - INFO - Exp name: convnext_small_4xb256_sdv1_1m_lr4e-4_aug_5e-1_20230604_164906 +2023/06/04 21:19:08 - mmengine - INFO - Epoch(train) [9][ 400/1954] lr: 4.0218e-05 eta: 0:57:46 time: 0.9323 data_time: 0.0020 memory: 15585 loss: 0.0037 +2023/06/04 21:20:44 - mmengine - INFO - Epoch(train) [9][ 500/1954] lr: 3.8563e-05 eta: 0:56:07 time: 0.9459 data_time: 0.0025 memory: 15585 loss: 0.0023 +2023/06/04 21:22:20 - mmengine - INFO - Epoch(train) [9][ 600/1954] lr: 3.6951e-05 eta: 0:54:28 time: 0.9122 data_time: 0.0021 memory: 15585 loss: 0.0025 +2023/06/04 21:23:55 - mmengine - INFO - Epoch(train) [9][ 700/1954] lr: 3.5383e-05 eta: 0:52:48 time: 0.9328 data_time: 0.0022 memory: 15585 loss: 0.0004 +2023/06/04 21:25:32 - mmengine - INFO - Epoch(train) [9][ 800/1954] lr: 3.3858e-05 eta: 0:51:09 time: 0.9693 data_time: 0.0023 memory: 15585 loss: 0.0031 +2023/06/04 21:27:07 - mmengine - INFO - Epoch(train) [9][ 900/1954] lr: 3.2378e-05 eta: 0:49:29 time: 0.9079 data_time: 0.0023 memory: 15585 loss: 0.0016 +2023/06/04 21:28:43 - mmengine - INFO - Epoch(train) [9][1000/1954] lr: 3.0942e-05 eta: 0:47:50 time: 0.9517 data_time: 0.0624 memory: 15585 loss: 0.0037 +2023/06/04 21:30:19 - mmengine - INFO - Epoch(train) [9][1100/1954] lr: 2.9551e-05 eta: 0:46:11 time: 0.9905 data_time: 0.0483 memory: 15585 loss: 0.0030 +2023/06/04 21:31:55 - mmengine - INFO - Epoch(train) [9][1200/1954] lr: 2.8205e-05 eta: 0:44:32 time: 0.9796 data_time: 0.0020 memory: 15585 loss: 0.0017 +2023/06/04 21:33:30 - mmengine - INFO - Epoch(train) [9][1300/1954] lr: 2.6905e-05 eta: 0:42:53 time: 0.9274 data_time: 0.0233 memory: 15585 loss: 0.0021 +2023/06/04 21:34:36 - mmengine - INFO - Exp name: convnext_small_4xb256_sdv1_1m_lr4e-4_aug_5e-1_20230604_164906 +2023/06/04 21:35:06 - mmengine - INFO - Epoch(train) [9][1400/1954] lr: 2.5652e-05 eta: 0:41:14 time: 0.9994 data_time: 0.0024 memory: 15585 loss: 0.0033 +2023/06/04 21:36:43 - mmengine - INFO - Epoch(train) [9][1500/1954] lr: 2.4444e-05 eta: 0:39:35 time: 1.0504 data_time: 0.2069 memory: 15585 loss: 0.0006 +2023/06/04 21:38:21 - mmengine - INFO - Epoch(train) [9][1600/1954] lr: 2.3283e-05 eta: 0:37:56 time: 0.9768 data_time: 0.4368 memory: 15585 loss: 0.0036 +2023/06/04 21:39:57 - mmengine - INFO - Epoch(train) [9][1700/1954] lr: 2.2170e-05 eta: 0:36:17 time: 0.9541 data_time: 0.4295 memory: 15585 loss: 0.0007 +2023/06/04 21:41:33 - mmengine - INFO - Epoch(train) [9][1800/1954] lr: 2.1103e-05 eta: 0:34:38 time: 0.9119 data_time: 0.3858 memory: 15585 loss: 0.0017 +2023/06/04 21:43:09 - mmengine - INFO - Epoch(train) [9][1900/1954] lr: 2.0084e-05 eta: 0:32:59 time: 0.9981 data_time: 0.4807 memory: 15585 loss: 0.0002 +2023/06/04 21:43:57 - mmengine - INFO - Exp name: convnext_small_4xb256_sdv1_1m_lr4e-4_aug_5e-1_20230604_164906 +2023/06/04 21:43:57 - mmengine - INFO - Saving checkpoint at 9 epochs +2023/06/04 21:44:36 - mmengine - INFO - Epoch(val) [9][29/29] accuracy/top1: 99.8400 single-label/precision_classwise: [99.76683807373047, 99.93008422851562] single-label/recall_classwise: [99.94318389892578, 99.71317291259766] single-label/f1-score_classwise: [99.85493469238281, 99.8215103149414] data_time: 0.2824 time: 0.6790 +2023/06/04 21:46:15 - mmengine - INFO - Epoch(train) [10][ 100/1954] lr: 1.8608e-05 eta: 0:30:27 time: 0.9121 data_time: 0.3652 memory: 15585 loss: 0.0005 +2023/06/04 21:47:51 - mmengine - INFO - Epoch(train) [10][ 200/1954] lr: 1.7711e-05 eta: 0:28:48 time: 0.9657 data_time: 0.3726 memory: 15585 loss: 0.0006 +2023/06/04 21:49:24 - mmengine - INFO - Epoch(train) [10][ 300/1954] lr: 1.6863e-05 eta: 0:27:09 time: 0.9508 data_time: 0.2415 memory: 15585 loss: 0.0034 +2023/06/04 21:50:59 - mmengine - INFO - Epoch(train) [10][ 400/1954] lr: 1.6063e-05 eta: 0:25:30 time: 1.0000 data_time: 0.1849 memory: 15585 loss: 0.0015 +2023/06/04 21:51:15 - mmengine - INFO - Exp name: convnext_small_4xb256_sdv1_1m_lr4e-4_aug_5e-1_20230604_164906 +2023/06/04 21:52:36 - mmengine - INFO - Epoch(train) [10][ 500/1954] lr: 1.5311e-05 eta: 0:23:52 time: 1.0022 data_time: 0.3870 memory: 15585 loss: 0.0021 +2023/06/04 21:54:11 - mmengine - INFO - Epoch(train) [10][ 600/1954] lr: 1.4609e-05 eta: 0:22:13 time: 0.9471 data_time: 0.4147 memory: 15585 loss: 0.0006 +2023/06/04 21:55:45 - mmengine - INFO - Epoch(train) [10][ 700/1954] lr: 1.3956e-05 eta: 0:20:34 time: 0.9055 data_time: 0.3815 memory: 15585 loss: 0.0023 +2023/06/04 21:57:21 - mmengine - INFO - Epoch(train) [10][ 800/1954] lr: 1.3353e-05 eta: 0:18:55 time: 0.9655 data_time: 0.4071 memory: 15585 loss: 0.0024 +2023/06/04 21:58:57 - mmengine - INFO - Epoch(train) [10][ 900/1954] lr: 1.2798e-05 eta: 0:17:17 time: 0.9454 data_time: 0.4162 memory: 15585 loss: 0.0008 +2023/06/04 22:00:30 - mmengine - INFO - Epoch(train) [10][1000/1954] lr: 1.2294e-05 eta: 0:15:38 time: 0.9057 data_time: 0.3698 memory: 15585 loss: 0.0018 +2023/06/04 22:02:05 - mmengine - INFO - Epoch(train) [10][1100/1954] lr: 1.1840e-05 eta: 0:14:00 time: 0.9476 data_time: 0.3754 memory: 15585 loss: 0.0026 +2023/06/04 22:03:42 - mmengine - INFO - Epoch(train) [10][1200/1954] lr: 1.1435e-05 eta: 0:12:21 time: 0.9871 data_time: 0.4041 memory: 15585 loss: 0.0028 +2023/06/04 22:05:18 - mmengine - INFO - Epoch(train) [10][1300/1954] lr: 1.1080e-05 eta: 0:10:43 time: 0.9315 data_time: 0.4022 memory: 15585 loss: 0.0009 +2023/06/04 22:06:55 - mmengine - INFO - Epoch(train) [10][1400/1954] lr: 1.0776e-05 eta: 0:09:04 time: 0.9844 data_time: 0.4565 memory: 15585 loss: 0.0003 +2023/06/04 22:07:06 - mmengine - INFO - Exp name: convnext_small_4xb256_sdv1_1m_lr4e-4_aug_5e-1_20230604_164906 +2023/06/04 22:08:31 - mmengine - INFO - Epoch(train) [10][1500/1954] lr: 1.0522e-05 eta: 0:07:26 time: 0.9550 data_time: 0.4273 memory: 15585 loss: 0.0016 +2023/06/04 22:10:06 - mmengine - INFO - Epoch(train) [10][1600/1954] lr: 1.0318e-05 eta: 0:05:48 time: 0.8936 data_time: 0.3768 memory: 15585 loss: 0.0011 +2023/06/04 22:11:42 - mmengine - INFO - Epoch(train) [10][1700/1954] lr: 1.0164e-05 eta: 0:04:09 time: 0.9295 data_time: 0.3927 memory: 15585 loss: 0.0010 +2023/06/04 22:13:21 - mmengine - INFO - Epoch(train) [10][1800/1954] lr: 1.0061e-05 eta: 0:02:31 time: 1.0009 data_time: 0.4797 memory: 15585 loss: 0.0006 +2023/06/04 22:14:59 - mmengine - INFO - Epoch(train) [10][1900/1954] lr: 1.0008e-05 eta: 0:00:53 time: 0.9594 data_time: 0.4416 memory: 15585 loss: 0.0069 +2023/06/04 22:15:48 - mmengine - INFO - Exp name: convnext_small_4xb256_sdv1_1m_lr4e-4_aug_5e-1_20230604_164906 +2023/06/04 22:15:48 - mmengine - INFO - Saving checkpoint at 10 epochs +2023/06/04 22:16:25 - mmengine - INFO - Epoch(val) [10][29/29] accuracy/top1: 99.8887 single-label/precision_classwise: [99.83607482910156, 99.95342254638672] single-label/recall_classwise: [99.96212005615234, 99.79844665527344] single-label/f1-score_classwise: [99.89906311035156, 99.8758773803711] data_time: 0.2708 time: 0.6578 diff --git a/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_5e-1/20230604_164906/vis_data/20230604_164906.json b/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_5e-1/20230604_164906/vis_data/20230604_164906.json new file mode 100644 index 0000000000000000000000000000000000000000..1efad0b7b9cd2247948c01f5d24821f53f5ecb62 --- /dev/null +++ b/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_5e-1/20230604_164906/vis_data/20230604_164906.json @@ -0,0 +1,200 @@ +{"lr": 0.00039997529889041205, "data_time": 0.27488863468170166, "loss": 0.044811129383742807, "time": 1.3829522132873535, "epoch": 1, "memory": 15587, "step": 100} +{"lr": 0.0003999002014233684, "data_time": 0.0025069713592529297, "loss": 0.03910655211657286, "time": 1.3254430055618287, "epoch": 1, "memory": 15587, "step": 200} +{"lr": 0.0003997747244903853, "data_time": 0.0017630100250244141, "loss": 0.030557064851745964, "time": 1.2637983798980712, "epoch": 1, "memory": 15587, "step": 300} +{"lr": 0.00039959890052580966, "data_time": 0.0018989086151123048, "loss": 0.022639472782611848, "time": 1.4661691665649415, "epoch": 1, "memory": 15587, "step": 400} +{"lr": 0.0003993727749781197, "data_time": 0.0017950057983398438, "loss": 0.016433806577697396, "time": 1.3433789491653443, "epoch": 1, "memory": 15587, "step": 500} +{"lr": 0.000399096406298175, "data_time": 0.0018087387084960937, "loss": 0.01657052696682513, "time": 1.3008520364761353, "epoch": 1, "memory": 15587, "step": 600} +{"lr": 0.0003987698659241082, "data_time": 0.0015877723693847657, "loss": 0.01950347237288952, "time": 1.2781208276748657, "epoch": 1, "memory": 15587, "step": 700} +{"lr": 0.0003983932382628598, "data_time": 0.001738572120666504, "loss": 0.0162202721927315, "time": 1.1779246568679809, "epoch": 1, "memory": 15587, "step": 800} +{"lr": 0.00039796662066835946, "data_time": 0.001804184913635254, "loss": 0.01106295855715871, "time": 1.1196962594985962, "epoch": 1, "memory": 15587, "step": 900} +{"lr": 0.00039749012341636, "data_time": 0.0017992734909057617, "loss": 0.014533220883458852, "time": 1.1227311372756958, "epoch": 1, "memory": 15587, "step": 1000} +{"lr": 0.00039696386967593436, "data_time": 0.0019058942794799804, "loss": 0.009314351435750722, "time": 1.113140082359314, "epoch": 1, "memory": 15587, "step": 1100} +{"lr": 0.00039638799547763706, "data_time": 0.030929255485534667, "loss": 0.016852918826043606, "time": 1.0852095127105712, "epoch": 1, "memory": 15587, "step": 1200} +{"lr": 0.0003957626496783407, "data_time": 0.03611502647399902, "loss": 0.02064823657274246, "time": 1.09158034324646, "epoch": 1, "memory": 15587, "step": 1300} +{"lr": 0.0003950879939227595, "data_time": 0.001755666732788086, "loss": 0.020769246947020294, "time": 1.0547627449035644, "epoch": 1, "memory": 15587, "step": 1400} +{"lr": 0.0003943642026016663, "data_time": 0.0024544477462768556, "loss": 0.015398759581148624, "time": 1.0406849384307861, "epoch": 1, "memory": 15587, "step": 1500} +{"lr": 0.0003935914628068133, "data_time": 0.0016436100006103516, "loss": 0.010116389906033873, "time": 1.02152578830719, "epoch": 1, "memory": 15587, "step": 1600} +{"lr": 0.00039276997428257205, "data_time": 0.002260255813598633, "loss": 0.022971110278740527, "time": 1.0216881513595581, "epoch": 1, "memory": 15587, "step": 1700} +{"lr": 0.0003918999493743001, "data_time": 0.0017485857009887696, "loss": 0.02214900734834373, "time": 0.9366251945495605, "epoch": 1, "memory": 15587, "step": 1800} +{"lr": 0.00039098161297345475, "data_time": 0.0018000125885009766, "loss": 0.013530656974762679, "time": 0.9412976026535034, "epoch": 1, "memory": 15587, "step": 1900} +{"accuracy/top1": 88.17369079589844, "data_time": 0.3702001982721789, "time": 0.839593139188043, "step": 1} +{"lr": 0.0003894734390071047, "data_time": 0.01877894401550293, "loss": 0.015119136963039636, "time": 0.9577568769454956, "epoch": 2, "memory": 15587, "step": 2054} +{"lr": 0.0003884334872841784, "data_time": 0.17825229167938234, "loss": 0.009091089013963938, "time": 0.9782303810119629, "epoch": 2, "memory": 15585, "step": 2154} +{"lr": 0.00038734612010935703, "data_time": 0.17850701808929442, "loss": 0.017891387827694415, "time": 0.945639705657959, "epoch": 2, "memory": 15585, "step": 2254} +{"lr": 0.00038621161855458035, "data_time": 0.22520716190338136, "loss": 0.01324383676983416, "time": 0.9511027812957764, "epoch": 2, "memory": 15585, "step": 2354} +{"lr": 0.0003850302758754841, "data_time": 0.17745506763458252, "loss": 0.008567770197987556, "time": 0.9681565523147583, "epoch": 2, "memory": 15585, "step": 2454} +{"lr": 0.00038380239743559646, "data_time": 0.24744791984558107, "loss": 0.014836873067542911, "time": 0.9981346607208252, "epoch": 2, "memory": 15585, "step": 2554} +{"lr": 0.00038252830062740647, "data_time": 0.4144535779953003, "loss": 0.014212223328649997, "time": 0.9331673145294189, "epoch": 2, "memory": 15585, "step": 2654} +{"lr": 0.00038120831479032175, "data_time": 0.32656862735748293, "loss": 0.012860441068187355, "time": 0.9716259717941285, "epoch": 2, "memory": 15585, "step": 2754} +{"lr": 0.0003798427811255361, "data_time": 0.3410447835922241, "loss": 0.009475871594622732, "time": 0.9721494436264038, "epoch": 2, "memory": 15585, "step": 2854} +{"lr": 0.0003784320526078357, "data_time": 0.4131464004516602, "loss": 0.010852901544421912, "time": 0.9363669872283935, "epoch": 2, "memory": 15585, "step": 2954} +{"lr": 0.0003769764938943556, "data_time": 0.41784465312957764, "loss": 0.010025016637519002, "time": 0.9517346858978272, "epoch": 2, "memory": 15585, "step": 3054} +{"lr": 0.0003754764812303242, "data_time": 0.4260796308517456, "loss": 0.010024727461859584, "time": 0.9602128028869629, "epoch": 2, "memory": 15585, "step": 3154} +{"lr": 0.0003739324023518048, "data_time": 0.48997960090637205, "loss": 0.010866807214915753, "time": 1.0086214065551757, "epoch": 2, "memory": 15585, "step": 3254} +{"lr": 0.0003723446563854724, "data_time": 0.4330077886581421, "loss": 0.015846089040860534, "time": 0.9536540269851684, "epoch": 2, "memory": 15585, "step": 3354} +{"lr": 0.0003707136537454424, "data_time": 0.4909473657608032, "loss": 0.015231353789567947, "time": 1.0086801052093506, "epoch": 2, "memory": 15585, "step": 3454} +{"lr": 0.0003690398160271841, "data_time": 0.42433533668518064, "loss": 0.014469049172475934, "time": 0.9541850328445435, "epoch": 2, "memory": 15585, "step": 3554} +{"lr": 0.00036732357589854426, "data_time": 0.4355283260345459, "loss": 0.007716640597209334, "time": 0.9569052457809448, "epoch": 2, "memory": 15585, "step": 3654} +{"lr": 0.00036556537698790475, "data_time": 0.4240563869476318, "loss": 0.013134542619809508, "time": 0.9421380519866943, "epoch": 2, "memory": 15585, "step": 3754} +{"lr": 0.0003637656737695118, "data_time": 0.4751794099807739, "loss": 0.021589742554351688, "time": 0.9941707134246827, "epoch": 2, "memory": 15585, "step": 3854} +{"accuracy/top1": 94.76705932617188, "data_time": 0.3208686908086141, "time": 0.7103821833928426, "step": 2} +{"lr": 0.00036091403361175644, "data_time": 0.0025679349899291994, "loss": 0.009165691211819649, "time": 1.0242275953292848, "epoch": 3, "memory": 15585, "step": 4008} +{"lr": 0.00035901102535335625, "data_time": 0.006280827522277832, "loss": 0.004569197678938508, "time": 0.9417719841003418, "epoch": 3, "memory": 15585, "step": 4108} +{"lr": 0.00035706820701190153, "data_time": 0.0019710779190063475, "loss": 0.012781585427001118, "time": 0.9085953235626221, "epoch": 3, "memory": 15585, "step": 4208} +{"lr": 0.00035508608078364215, "data_time": 0.0034751653671264648, "loss": 0.013173878146335482, "time": 1.0007141351699829, "epoch": 3, "memory": 15585, "step": 4308} +{"lr": 0.00035306515902546755, "data_time": 0.002584242820739746, "loss": 0.0109096250962466, "time": 0.96571843624115, "epoch": 3, "memory": 15585, "step": 4408} +{"lr": 0.0003510059641224655, "data_time": 0.0023252487182617186, "loss": 0.0054708153009414675, "time": 0.9249018907546998, "epoch": 3, "memory": 15585, "step": 4508} +{"lr": 0.00034890902835289306, "data_time": 0.0017351150512695313, "loss": 0.005799282202497124, "time": 0.9141477346420288, "epoch": 3, "memory": 15585, "step": 4608} +{"lr": 0.00034677489375058753, "data_time": 0.001969265937805176, "loss": 0.010284630768001079, "time": 0.9485032558441162, "epoch": 3, "memory": 15585, "step": 4708} +{"lr": 0.00034460411196485875, "data_time": 0.0019747734069824217, "loss": 0.01762869395315647, "time": 0.9205304384231567, "epoch": 3, "memory": 15585, "step": 4808} +{"lr": 0.0003423972441178926, "data_time": 0.010460853576660156, "loss": 0.011661151982843876, "time": 0.9256266117095947, "epoch": 3, "memory": 15585, "step": 4908} +{"lr": 0.0003401548606597075, "data_time": 0.14806385040283204, "loss": 0.009646823955699801, "time": 0.9667104721069336, "epoch": 3, "memory": 15585, "step": 5008} +{"lr": 0.0003378775412207003, "data_time": 0.05621922016143799, "loss": 0.010691530350595713, "time": 0.9566173791885376, "epoch": 3, "memory": 15585, "step": 5108} +{"lr": 0.0003355658744618178, "data_time": 0.1323199987411499, "loss": 0.009256822476163506, "time": 0.9530836343765259, "epoch": 3, "memory": 15585, "step": 5208} +{"lr": 0.0003332204579223944, "data_time": 0.0348684549331665, "loss": 0.009746877616271377, "time": 0.9325660705566406, "epoch": 3, "memory": 15585, "step": 5308} +{"lr": 0.0003308418978656968, "data_time": 0.04287197589874268, "loss": 0.015595878195017576, "time": 0.996539568901062, "epoch": 3, "memory": 15585, "step": 5408} +{"lr": 0.0003284308091222099, "data_time": 0.11746449470520019, "loss": 0.009972090134397149, "time": 0.9147012948989868, "epoch": 3, "memory": 15585, "step": 5508} +{"lr": 0.0003259878149307108, "data_time": 0.19142618179321289, "loss": 0.0053694368805736305, "time": 0.9154141187667847, "epoch": 3, "memory": 15585, "step": 5608} +{"lr": 0.0003235135467771688, "data_time": 0.29831852912902834, "loss": 0.007030118349939584, "time": 0.9449497938156128, "epoch": 3, "memory": 15585, "step": 5708} +{"lr": 0.0003210086442315141, "data_time": 0.12527620792388916, "loss": 0.009467535465955735, "time": 0.9195688724517822, "epoch": 3, "memory": 15585, "step": 5808} +{"accuracy/top1": 97.40092468261719, "data_time": 0.28207352956136067, "time": 0.6702582120895386, "step": 3} +{"lr": 0.00031709267628953966, "data_time": 0.11172959804534913, "loss": 0.00649794265627861, "time": 0.9505893230438233, "epoch": 4, "memory": 15585, "step": 5962} +{"lr": 0.00031451289120303685, "data_time": 0.0017980337142944336, "loss": 0.007613994600251317, "time": 0.9671101331710815, "epoch": 4, "memory": 15585, "step": 6062} +{"lr": 0.000311904798290621, "data_time": 0.0014580249786376952, "loss": 0.011619580676779151, "time": 1.003280806541443, "epoch": 4, "memory": 15585, "step": 6162} +{"lr": 0.00030926907171438685, "data_time": 0.0017405271530151368, "loss": 0.00666549876332283, "time": 0.9539995670318604, "epoch": 4, "memory": 15585, "step": 6262} +{"lr": 0.000306606392779414, "data_time": 0.002644658088684082, "loss": 0.011141629936173559, "time": 0.968813157081604, "epoch": 4, "memory": 15585, "step": 6362} +{"lr": 0.0003039174497576579, "data_time": 0.0018522262573242188, "loss": 0.008385690348222851, "time": 0.9816806793212891, "epoch": 4, "memory": 15585, "step": 6462} +{"lr": 0.0003012029377100387, "data_time": 0.0018495559692382813, "loss": 0.009659701539203525, "time": 0.9766746759414673, "epoch": 4, "memory": 15585, "step": 6562} +{"lr": 0.00029846355830677674, "data_time": 0.03559055328369141, "loss": 0.0036875926423817875, "time": 0.9105247259140015, "epoch": 4, "memory": 15585, "step": 6662} +{"lr": 0.0002957000196460183, "data_time": 0.13139536380767822, "loss": 0.007654532138258219, "time": 0.9805450916290284, "epoch": 4, "memory": 15585, "step": 6762} +{"lr": 0.0002929130360708003, "data_time": 0.5423965692520142, "loss": 0.0054247718770056965, "time": 1.0859703302383423, "epoch": 4, "memory": 15585, "step": 6862} +{"lr": 0.0002901033279844016, "data_time": 0.45169754028320314, "loss": 0.007872724533081054, "time": 0.9707653522491455, "epoch": 4, "memory": 15585, "step": 6962} +{"lr": 0.00028727162166412623, "data_time": 0.41893854141235354, "loss": 0.005585976736620068, "time": 0.9506667137145997, "epoch": 4, "memory": 15585, "step": 7062} +{"lr": 0.000284418649073569, "data_time": 0.41764733791351316, "loss": 0.0061735509429126974, "time": 0.9365954399108887, "epoch": 4, "memory": 15585, "step": 7162} +{"lr": 0.00028154514767341203, "data_time": 0.49697954654693605, "loss": 0.0049189492128789425, "time": 1.0262855052948, "epoch": 4, "memory": 15585, "step": 7262} +{"lr": 0.0002786518602307988, "data_time": 0.441841721534729, "loss": 0.0049474481493234634, "time": 0.9627981424331665, "epoch": 4, "memory": 15585, "step": 7362} +{"lr": 0.0002757395346273384, "data_time": 0.5250383377075195, "loss": 0.007277003536000848, "time": 1.0449779272079467, "epoch": 4, "memory": 15585, "step": 7462} +{"lr": 0.0002728089236657853, "data_time": 0.4268195390701294, "loss": 0.01237349109724164, "time": 0.9489672899246215, "epoch": 4, "memory": 15585, "step": 7562} +{"lr": 0.0002698607848754497, "data_time": 0.4164048433303833, "loss": 0.006725517986342311, "time": 0.9444144248962403, "epoch": 4, "memory": 15585, "step": 7662} +{"lr": 0.0002668958803163847, "data_time": 0.4745892286300659, "loss": 0.007470806688070297, "time": 0.9946723699569702, "epoch": 4, "memory": 15585, "step": 7762} +{"accuracy/top1": 98.79267120361328, "data_time": 0.3012315114339193, "time": 0.6895922422409058, "step": 4} +{"lr": 0.00026229890681650394, "data_time": 0.0768885612487793, "loss": 0.00528542841784656, "time": 0.9828031301498413, "epoch": 5, "memory": 15585, "step": 7916} +{"lr": 0.00025929487224058475, "data_time": 0.1465217113494873, "loss": 0.0072232729755342005, "time": 0.9394273281097412, "epoch": 5, "memory": 15585, "step": 8016} +{"lr": 0.0002562768030631272, "data_time": 0.001854109764099121, "loss": 0.004652506532147526, "time": 0.9927838325500489, "epoch": 5, "memory": 15585, "step": 8116} +{"lr": 0.0002532454794203871, "data_time": 0.02627689838409424, "loss": 0.0052061258815228936, "time": 0.9198421239852905, "epoch": 5, "memory": 15585, "step": 8216} +{"lr": 0.0002502016848747476, "data_time": 0.056214523315429685, "loss": 0.004920996772125363, "time": 0.969770073890686, "epoch": 5, "memory": 15585, "step": 8316} +{"lr": 0.0002471462062121758, "data_time": 0.16587018966674805, "loss": 0.0043186724185943605, "time": 0.9476407766342163, "epoch": 5, "memory": 15585, "step": 8416} +{"lr": 0.00024407983323884952, "data_time": 0.2657217502593994, "loss": 0.005234378017485141, "time": 0.9712835073471069, "epoch": 5, "memory": 15585, "step": 8516} +{"lr": 0.0002410033585770006, "data_time": 0.3135234355926514, "loss": 0.0069424071814864876, "time": 0.9429662942886352, "epoch": 5, "memory": 15585, "step": 8616} +{"lr": 0.00023791757746003292, "data_time": 0.448268723487854, "loss": 0.00430152346380055, "time": 0.976430082321167, "epoch": 5, "memory": 15585, "step": 8716} +{"lr": 0.00023482328752696182, "data_time": 0.4581224679946899, "loss": 0.006737135397270322, "time": 0.9894840717315674, "epoch": 5, "memory": 15585, "step": 8816} +{"lr": 0.00023172128861623345, "data_time": 0.44569220542907717, "loss": 0.0058538383338600395, "time": 0.9701451778411865, "epoch": 5, "memory": 15585, "step": 8916} +{"lr": 0.000228612382558976, "data_time": 0.459169602394104, "loss": 0.005569694563746452, "time": 1.0004676580429077, "epoch": 5, "memory": 15585, "step": 9016} +{"lr": 0.00022549737297173692, "data_time": 0.40810461044311525, "loss": 0.005624811025336384, "time": 0.932976484298706, "epoch": 5, "memory": 15585, "step": 9116} +{"lr": 0.00022237706504875477, "data_time": 0.4501085996627808, "loss": 0.003529950091615319, "time": 0.9704623937606811, "epoch": 5, "memory": 15585, "step": 9216} +{"lr": 0.00021925226535382818, "data_time": 0.4203763246536255, "loss": 0.006921368557959795, "time": 0.947316837310791, "epoch": 5, "memory": 15585, "step": 9316} +{"lr": 0.0002161237816118266, "data_time": 0.38816003799438475, "loss": 0.006047765770927071, "time": 0.9536850214004516, "epoch": 5, "memory": 15585, "step": 9416} +{"lr": 0.00021299242249990383, "data_time": 0.4218069314956665, "loss": 0.0039038790855556726, "time": 0.9412720441818238, "epoch": 5, "memory": 15585, "step": 9516} +{"lr": 0.00020985899743846363, "data_time": 0.41132094860076907, "loss": 0.005127867497503757, "time": 0.9767580270767212, "epoch": 5, "memory": 15585, "step": 9616} +{"lr": 0.00020672431638193363, "data_time": 0.3408376693725586, "loss": 0.005221713194623589, "time": 0.9531792163848877, "epoch": 5, "memory": 15585, "step": 9716} +{"accuracy/top1": 99.405029296875, "data_time": 0.2662598927815755, "time": 0.6586158990859985, "step": 5} +{"lr": 0.00020189632111931318, "data_time": 0.001739192008972168, "loss": 0.003452608222141862, "time": 0.9475012302398682, "epoch": 6, "memory": 15585, "step": 9870} +{"lr": 0.00019876209288860824, "data_time": 0.0024026155471801756, "loss": 0.005942829232662916, "time": 0.9519016027450562, "epoch": 6, "memory": 15585, "step": 9970} +{"lr": 0.00019562947708532286, "data_time": 0.0029773950576782227, "loss": 0.0071106948889791965, "time": 0.951548433303833, "epoch": 6, "memory": 15585, "step": 10070} +{"lr": 0.00019249928345470104, "data_time": 0.0027435302734375, "loss": 0.004475785465911031, "time": 0.9630252361297608, "epoch": 6, "memory": 15585, "step": 10170} +{"lr": 0.00018937232111588298, "data_time": 0.0018990755081176758, "loss": 0.003966807201504707, "time": 0.9069278717041016, "epoch": 6, "memory": 15585, "step": 10270} +{"lr": 0.000186249398352757, "data_time": 0.0015522956848144532, "loss": 0.008469361485913396, "time": 0.9342559576034546, "epoch": 6, "memory": 15585, "step": 10370} +{"lr": 0.0001831313224050277, "data_time": 0.40353171825408934, "loss": 0.004387742327526212, "time": 0.9232488870620728, "epoch": 6, "memory": 15585, "step": 10470} +{"lr": 0.00018001889925955291, "data_time": 0.39816813468933104, "loss": 0.0028657491318881513, "time": 0.9168015956878662, "epoch": 6, "memory": 15585, "step": 10570} +{"lr": 0.00017691293344200646, "data_time": 0.4150885820388794, "loss": 0.004340462340041995, "time": 0.9349114656448364, "epoch": 6, "memory": 15585, "step": 10670} +{"lr": 0.00017381422780891689, "data_time": 0.42396697998046873, "loss": 0.005025573493912816, "time": 0.9522564172744751, "epoch": 6, "memory": 15585, "step": 10770} +{"lr": 0.00017072358334013806, "data_time": 0.434910249710083, "loss": 0.0037409603130072354, "time": 0.9580995798110962, "epoch": 6, "memory": 15585, "step": 10870} +{"lr": 0.00016764179893180575, "data_time": 0.41643226146698, "loss": 0.00355774681083858, "time": 0.9355035066604614, "epoch": 6, "memory": 15585, "step": 10970} +{"lr": 0.00016456967118983176, "data_time": 0.3971834659576416, "loss": 0.0021645229775458575, "time": 0.916974139213562, "epoch": 6, "memory": 15585, "step": 11070} +{"lr": 0.00016150799422398984, "data_time": 0.44319169521331786, "loss": 0.0022221977822482588, "time": 0.9627624034881592, "epoch": 6, "memory": 15585, "step": 11170} +{"lr": 0.0001584575594426489, "data_time": 0.40769870281219484, "loss": 0.004099268745630979, "time": 0.9255609512329102, "epoch": 6, "memory": 15585, "step": 11270} +{"lr": 0.00015541915534820135, "data_time": 0.44083538055419924, "loss": 0.002200006926432252, "time": 0.9688750028610229, "epoch": 6, "memory": 15585, "step": 11370} +{"lr": 0.00015239356733324542, "data_time": 0.4261681795120239, "loss": 0.006635595066472888, "time": 0.9478034973144531, "epoch": 6, "memory": 15585, "step": 11470} +{"lr": 0.00014938157747756997, "data_time": 0.4163465738296509, "loss": 0.003441391373053193, "time": 0.9390012264251709, "epoch": 6, "memory": 15585, "step": 11570} +{"lr": 0.00014638396434599565, "data_time": 0.4195083141326904, "loss": 0.000967968488112092, "time": 0.9382271766662598, "epoch": 6, "memory": 15585, "step": 11670} +{"accuracy/top1": 99.6207504272461, "data_time": 0.26066848436991374, "time": 0.6483653624852498, "step": 6} +{"lr": 0.00014179754513556176, "data_time": 0.0016521453857421876, "loss": 0.005475955689325929, "time": 0.9159265518188476, "epoch": 7, "memory": 15585, "step": 11824} +{"lr": 0.00013883992334674444, "data_time": 0.0019083738327026368, "loss": 0.006218972522765398, "time": 0.9376081705093384, "epoch": 7, "memory": 15585, "step": 11924} +{"lr": 0.00013589940317867467, "data_time": 0.002809476852416992, "loss": 0.002176971593871713, "time": 1.4418408155441285, "epoch": 7, "memory": 15585, "step": 12024} +{"lr": 0.0001329767447220783, "data_time": 0.0026750564575195312, "loss": 0.002354126889258623, "time": 0.9647247791290283, "epoch": 7, "memory": 15585, "step": 12124} +{"lr": 0.00013007270345063488, "data_time": 0.002393651008605957, "loss": 0.0024485097732394935, "time": 0.9378865480422973, "epoch": 7, "memory": 15585, "step": 12224} +{"lr": 0.00012718803002569412, "data_time": 0.0017752647399902344, "loss": 0.0034284644760191442, "time": 0.9883497476577758, "epoch": 7, "memory": 15585, "step": 12324} +{"lr": 0.0001243234701022397, "data_time": 0.0025546073913574217, "loss": 0.005110947135835886, "time": 0.9463302373886109, "epoch": 7, "memory": 15585, "step": 12424} +{"lr": 0.00012147976413614597, "data_time": 0.0019263744354248047, "loss": 0.0031996916048228742, "time": 0.9885228395462036, "epoch": 7, "memory": 15585, "step": 12524} +{"lr": 0.0001186576471927784, "data_time": 0.001650834083557129, "loss": 0.002845912938937545, "time": 0.9857007265090942, "epoch": 7, "memory": 15585, "step": 12624} +{"lr": 0.00011585784875698867, "data_time": 0.020549511909484862, "loss": 0.0007590523455291986, "time": 0.9341190338134766, "epoch": 7, "memory": 15585, "step": 12724} +{"lr": 0.00011308109254454993, "data_time": 0.13373348712921143, "loss": 0.005037216423079372, "time": 0.9964096069335937, "epoch": 7, "memory": 15585, "step": 12824} +{"lr": 0.00011032809631508466, "data_time": 0.009465742111206054, "loss": 0.0036153506953269243, "time": 0.9257693767547608, "epoch": 7, "memory": 15585, "step": 12924} +{"lr": 0.00010759957168653253, "data_time": 0.07855479717254639, "loss": 0.001611655903980136, "time": 0.9202292680740356, "epoch": 7, "memory": 15585, "step": 13024} +{"lr": 0.0001048962239512054, "data_time": 0.24015531539916993, "loss": 0.006936500268056989, "time": 1.0145259618759155, "epoch": 7, "memory": 15585, "step": 13124} +{"lr": 0.00010221875189347644, "data_time": 0.46049938201904295, "loss": 0.005218543158844113, "time": 0.9803165197372437, "epoch": 7, "memory": 15585, "step": 13224} +{"lr": 9.95678476091537e-05, "data_time": 0.3974884510040283, "loss": 0.002453421102836728, "time": 0.9266520977020264, "epoch": 7, "memory": 15585, "step": 13324} +{"lr": 9.694419632657999e-05, "data_time": 0.4210886716842651, "loss": 0.0021715907845646144, "time": 0.9405019521713257, "epoch": 7, "memory": 15585, "step": 13424} +{"lr": 9.434847622951059e-05, "data_time": 0.3621471405029297, "loss": 0.004599348129704595, "time": 0.8858633041381836, "epoch": 7, "memory": 15585, "step": 13524} +{"lr": 9.178135828180905e-05, "data_time": 0.5840172529220581, "loss": 0.001961578754708171, "time": 1.105517578125, "epoch": 7, "memory": 15585, "step": 13624} +{"accuracy/top1": 99.71817779541016, "data_time": 0.4316521247227987, "time": 0.8269522666931153, "step": 7} +{"lr": 8.788546579134055e-05, "data_time": 0.0017841815948486327, "loss": 0.0019957524724304675, "time": 0.931977367401123, "epoch": 8, "memory": 15585, "step": 13778} +{"lr": 8.539396307206318e-05, "data_time": 0.0018756628036499024, "loss": 0.0039055104367434978, "time": 0.901304030418396, "epoch": 8, "memory": 15585, "step": 13878} +{"lr": 8.293337714106816e-05, "data_time": 0.002025413513183594, "loss": 0.0034841256681829693, "time": 0.9605952978134156, "epoch": 8, "memory": 15585, "step": 13978} +{"lr": 8.05043440315878e-05, "data_time": 0.0017398834228515626, "loss": 0.0014356670435518026, "time": 0.9542671918869019, "epoch": 8, "memory": 15585, "step": 14078} +{"lr": 7.810749162081216e-05, "data_time": 0.0016051769256591798, "loss": 0.0009449648205190897, "time": 0.9690500736236572, "epoch": 8, "memory": 15585, "step": 14178} +{"lr": 7.574343946758993e-05, "data_time": 0.00210416316986084, "loss": 0.0039619202725589275, "time": 1.0130390882492066, "epoch": 8, "memory": 15585, "step": 14278} +{"lr": 7.34127986522793e-05, "data_time": 0.002979564666748047, "loss": 0.008022097731009125, "time": 0.9794174909591675, "epoch": 8, "memory": 15585, "step": 14378} +{"lr": 7.111617161879136e-05, "data_time": 0.002597999572753906, "loss": 0.0021727455779910086, "time": 0.9529072046279907, "epoch": 8, "memory": 15585, "step": 14478} +{"lr": 6.885415201886485e-05, "data_time": 0.22874774932861328, "loss": 0.002567023318260908, "time": 0.9731489658355713, "epoch": 8, "memory": 15585, "step": 14578} +{"lr": 6.662732455861337e-05, "data_time": 0.40278670787811277, "loss": 0.002550022955983877, "time": 0.9697470426559448, "epoch": 8, "memory": 15585, "step": 14678} +{"lr": 6.443626484738661e-05, "data_time": 0.42852249145507815, "loss": 0.0007593533955514431, "time": 0.948227334022522, "epoch": 8, "memory": 15585, "step": 14778} +{"lr": 6.22815392489815e-05, "data_time": 0.4181420087814331, "loss": 0.0022932265419512987, "time": 0.9465721845626831, "epoch": 8, "memory": 15585, "step": 14878} +{"lr": 6.016370473524328e-05, "data_time": 0.42673871517181394, "loss": 0.003924101451411843, "time": 0.9452044010162354, "epoch": 8, "memory": 15585, "step": 14978} +{"lr": 5.80833087420956e-05, "data_time": 0.4155222177505493, "loss": 0.002293054014444351, "time": 0.9513051986694336, "epoch": 8, "memory": 15585, "step": 15078} +{"lr": 5.604088902803397e-05, "data_time": 0.4702103853225708, "loss": 0.001514874119311571, "time": 0.9896433115005493, "epoch": 8, "memory": 15585, "step": 15178} +{"lr": 5.403697353512148e-05, "data_time": 0.44001338481903074, "loss": 0.004679047269746661, "time": 0.9795157670974731, "epoch": 8, "memory": 15585, "step": 15278} +{"lr": 5.207208025252174e-05, "data_time": 0.45709240436553955, "loss": 0.0035715472418814897, "time": 0.9803754568099976, "epoch": 8, "memory": 15585, "step": 15378} +{"lr": 5.014671708260458e-05, "data_time": 0.4312600612640381, "loss": 0.0014895839150995017, "time": 0.9742679119110107, "epoch": 8, "memory": 15585, "step": 15478} +{"lr": 4.826138170965952e-05, "data_time": 0.460954475402832, "loss": 0.0054690070450305935, "time": 0.9832231998443604, "epoch": 8, "memory": 15585, "step": 15578} +{"accuracy/top1": 99.7773208618164, "data_time": 0.27094720204671224, "time": 0.6685602982838948, "step": 8} +{"lr": 4.5437373040402475e-05, "data_time": 0.0024146318435668947, "loss": 0.00034796479158103464, "time": 0.9479226350784302, "epoch": 9, "memory": 15585, "step": 15732} +{"lr": 4.3655875036148156e-05, "data_time": 0.002630448341369629, "loss": 0.0055925148539245125, "time": 0.970571231842041, "epoch": 9, "memory": 15585, "step": 15832} +{"lr": 4.19160826370641e-05, "data_time": 0.0027977943420410155, "loss": 0.0009660322219133377, "time": 0.9527229547500611, "epoch": 9, "memory": 15585, "step": 15932} +{"lr": 4.021844555952455e-05, "data_time": 0.0020281314849853516, "loss": 0.003684419672936201, "time": 0.9322951793670654, "epoch": 9, "memory": 15585, "step": 16032} +{"lr": 3.8563402623236864e-05, "data_time": 0.0024892330169677735, "loss": 0.002301778132095933, "time": 0.9459209203720093, "epoch": 9, "memory": 15585, "step": 16132} +{"lr": 3.695138163781135e-05, "data_time": 0.002077054977416992, "loss": 0.0024881566409021617, "time": 0.9121654272079468, "epoch": 9, "memory": 15585, "step": 16232} +{"lr": 3.538279929217756e-05, "data_time": 0.00218348503112793, "loss": 0.00040242779068648814, "time": 0.9328038215637207, "epoch": 9, "memory": 15585, "step": 16332} +{"lr": 3.385806104687531e-05, "data_time": 0.0023269414901733398, "loss": 0.0030994636472314596, "time": 0.969316053390503, "epoch": 9, "memory": 15585, "step": 16432} +{"lr": 3.237756102924734e-05, "data_time": 0.002310299873352051, "loss": 0.0015743127558380365, "time": 0.9078565835952759, "epoch": 9, "memory": 15585, "step": 16532} +{"lr": 3.094168193156213e-05, "data_time": 0.062359070777893065, "loss": 0.0036647174041718245, "time": 0.9517327070236206, "epoch": 9, "memory": 15585, "step": 16632} +{"lr": 2.95507949120924e-05, "data_time": 0.04834351539611816, "loss": 0.002974098129197955, "time": 0.9905012607574463, "epoch": 9, "memory": 15585, "step": 16732} +{"lr": 2.8205259499174602e-05, "data_time": 0.001982450485229492, "loss": 0.0016547942534089089, "time": 0.9796242475509643, "epoch": 9, "memory": 15585, "step": 16832} +{"lr": 2.690542349827549e-05, "data_time": 0.02330169677734375, "loss": 0.0021105783060193064, "time": 0.9274474620819092, "epoch": 9, "memory": 15585, "step": 16932} +{"lr": 2.565162290208795e-05, "data_time": 0.0024168968200683595, "loss": 0.0033429506700485946, "time": 0.9993813991546631, "epoch": 9, "memory": 15585, "step": 17032} +{"lr": 2.4444181803680975e-05, "data_time": 0.20685017108917236, "loss": 0.0005877256859093905, "time": 1.05042884349823, "epoch": 9, "memory": 15585, "step": 17132} +{"lr": 2.32834123127253e-05, "data_time": 0.43677310943603515, "loss": 0.0036239826120436193, "time": 0.9767820596694946, "epoch": 9, "memory": 15585, "step": 17232} +{"lr": 2.2169614474816763e-05, "data_time": 0.42950997352600095, "loss": 0.0007100783754140138, "time": 0.9541327714920044, "epoch": 9, "memory": 15585, "step": 17332} +{"lr": 2.110307619391749e-05, "data_time": 0.38581721782684325, "loss": 0.001673663267865777, "time": 0.9118571996688842, "epoch": 9, "memory": 15585, "step": 17432} +{"lr": 2.008407315793661e-05, "data_time": 0.4806783437728882, "loss": 0.00017982046119868755, "time": 0.9980745792388916, "epoch": 9, "memory": 15585, "step": 17532} +{"accuracy/top1": 99.83995056152344, "data_time": 0.2823968331019084, "time": 0.6790300051371256, "step": 9} +{"lr": 1.8608381954463387e-05, "data_time": 0.36521923542022705, "loss": 0.00047856238670647145, "time": 0.9121445417404175, "epoch": 10, "memory": 15585, "step": 17686} +{"lr": 1.771127410231484e-05, "data_time": 0.37261722087860105, "loss": 0.0006369306705892086, "time": 0.9656543016433716, "epoch": 10, "memory": 15585, "step": 17786} +{"lr": 1.6862578237010542e-05, "data_time": 0.24154651165008545, "loss": 0.003397755604237318, "time": 0.9507620811462403, "epoch": 10, "memory": 15585, "step": 17886} +{"lr": 1.6062513736694383e-05, "data_time": 0.18485028743743898, "loss": 0.0015289057977497579, "time": 0.9999956846237182, "epoch": 10, "memory": 15585, "step": 17986} +{"lr": 1.531128740886036e-05, "data_time": 0.38703265190124514, "loss": 0.0021364753134548663, "time": 1.0022279024124146, "epoch": 10, "memory": 15585, "step": 18086} +{"lr": 1.4609093436895328e-05, "data_time": 0.4147106885910034, "loss": 0.000597511325031519, "time": 0.9470914363861084, "epoch": 10, "memory": 15585, "step": 18186} +{"lr": 1.3956113329884566e-05, "data_time": 0.3815218210220337, "loss": 0.0022642647847533224, "time": 0.9054733037948608, "epoch": 10, "memory": 15585, "step": 18286} +{"lr": 1.335251587569401e-05, "data_time": 0.40713484287261964, "loss": 0.00237970226444304, "time": 0.9654510974884033, "epoch": 10, "memory": 15585, "step": 18386} +{"lr": 1.2798457097340337e-05, "data_time": 0.4161717176437378, "loss": 0.0007964265532791615, "time": 0.9454386711120606, "epoch": 10, "memory": 15585, "step": 18486} +{"lr": 1.229408021266094e-05, "data_time": 0.36980977058410647, "loss": 0.0018288569524884224, "time": 0.9056573629379272, "epoch": 10, "memory": 15585, "step": 18586} +{"lr": 1.1839515597293723e-05, "data_time": 0.3753602743148804, "loss": 0.002611189801245928, "time": 0.9475775003433228, "epoch": 10, "memory": 15585, "step": 18686} +{"lr": 1.1434880750976406e-05, "data_time": 0.40407772064208985, "loss": 0.0027824061922729014, "time": 0.9871024131774903, "epoch": 10, "memory": 15585, "step": 18786} +{"lr": 1.108028026717431e-05, "data_time": 0.40224418640136717, "loss": 0.0009329441469162703, "time": 0.9314643859863281, "epoch": 10, "memory": 15585, "step": 18886} +{"lr": 1.0775805806044138e-05, "data_time": 0.4564747095108032, "loss": 0.00026018787175416944, "time": 0.9844456911087036, "epoch": 10, "memory": 15585, "step": 18986} +{"lr": 1.0521536070740776e-05, "data_time": 0.42733771800994874, "loss": 0.001569018978625536, "time": 0.954958152770996, "epoch": 10, "memory": 15585, "step": 19086} +{"lr": 1.03175367870735e-05, "data_time": 0.3768294095993042, "loss": 0.0011358255986124277, "time": 0.8935638189315795, "epoch": 10, "memory": 15585, "step": 19186} +{"lr": 1.0163860686516628e-05, "data_time": 0.39269509315490725, "loss": 0.000978016061708331, "time": 0.9295082092285156, "epoch": 10, "memory": 15585, "step": 19286} +{"lr": 1.0060547492578997e-05, "data_time": 0.47974233627319335, "loss": 0.000620917696505785, "time": 1.0009328842163085, "epoch": 10, "memory": 15585, "step": 19386} +{"lr": 1.0007623910535892e-05, "data_time": 0.44156646728515625, "loss": 0.006867260346189142, "time": 0.9594136714935303, "epoch": 10, "memory": 15585, "step": 19486} +{"accuracy/top1": 99.88866424560547, "data_time": 0.2708349068959554, "time": 0.6577602783838908, "step": 10} diff --git a/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_5e-1/20230604_164906/vis_data/config.py b/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_5e-1/20230604_164906/vis_data/config.py new file mode 100644 index 0000000000000000000000000000000000000000..c6e47050fb7e76afd481119b9be206927b1f3b2d --- /dev/null +++ b/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_5e-1/20230604_164906/vis_data/config.py @@ -0,0 +1,247 @@ +optim_wrapper = dict( + optimizer=dict( + type='AdamW', + lr=0.0004, + weight_decay=0.05, + eps=1e-08, + betas=(0.9, 0.999), + _scope_='mmpretrain'), + paramwise_cfg=dict( + norm_decay_mult=0.0, + bias_decay_mult=0.0, + flat_decay_mult=0.0, + custom_keys=dict({ + '.absolute_pos_embed': dict(decay_mult=0.0), + '.relative_position_bias_table': dict(decay_mult=0.0) + })), + type='AmpOptimWrapper', + dtype='bfloat16', + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=False, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=4096) +model = dict( + type='ImageClassifier', + backbone=dict( + frozen_stages=2, + type='ConvNeXt', + arch='small', + drop_path_rate=0.4, + init_cfg=dict( + type='Pretrained', + checkpoint= + 'https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth', + prefix='backbone')), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=768, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + init_cfg=None), + init_cfg=dict( + type='TruncNormal', layer=['Conv2d', 'Linear'], std=0.02, bias=0.0), + train_cfg=None) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stablediffusionV1-5R2-dpmsolver-25-1m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/cc1m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_5e-1' diff --git a/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_5e-1/20230604_164906/vis_data/events.out.tfevents.1685868562.SH-IDC1-10-140-24-114.167863.0 b/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_5e-1/20230604_164906/vis_data/events.out.tfevents.1685868562.SH-IDC1-10-140-24-114.167863.0 new file mode 100644 index 0000000000000000000000000000000000000000..e2a8b8c70286394c858506e5d68a6003db31160f --- /dev/null +++ b/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_5e-1/20230604_164906/vis_data/events.out.tfevents.1685868562.SH-IDC1-10-140-24-114.167863.0 @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:dcbeb9bac51292903b4b7fecc492cca5da42eb93c8fb46b972160b7add754b60 +size 1300572 diff --git a/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_5e-1/20230604_164906/vis_data/scalars.json b/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_5e-1/20230604_164906/vis_data/scalars.json new file mode 100644 index 0000000000000000000000000000000000000000..1efad0b7b9cd2247948c01f5d24821f53f5ecb62 --- /dev/null +++ b/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_5e-1/20230604_164906/vis_data/scalars.json @@ -0,0 +1,200 @@ +{"lr": 0.00039997529889041205, "data_time": 0.27488863468170166, "loss": 0.044811129383742807, "time": 1.3829522132873535, "epoch": 1, "memory": 15587, "step": 100} +{"lr": 0.0003999002014233684, "data_time": 0.0025069713592529297, "loss": 0.03910655211657286, "time": 1.3254430055618287, "epoch": 1, "memory": 15587, "step": 200} +{"lr": 0.0003997747244903853, "data_time": 0.0017630100250244141, "loss": 0.030557064851745964, "time": 1.2637983798980712, "epoch": 1, "memory": 15587, "step": 300} +{"lr": 0.00039959890052580966, "data_time": 0.0018989086151123048, "loss": 0.022639472782611848, "time": 1.4661691665649415, "epoch": 1, "memory": 15587, "step": 400} +{"lr": 0.0003993727749781197, "data_time": 0.0017950057983398438, "loss": 0.016433806577697396, "time": 1.3433789491653443, "epoch": 1, "memory": 15587, "step": 500} +{"lr": 0.000399096406298175, "data_time": 0.0018087387084960937, "loss": 0.01657052696682513, "time": 1.3008520364761353, "epoch": 1, "memory": 15587, "step": 600} +{"lr": 0.0003987698659241082, "data_time": 0.0015877723693847657, "loss": 0.01950347237288952, "time": 1.2781208276748657, "epoch": 1, "memory": 15587, "step": 700} +{"lr": 0.0003983932382628598, "data_time": 0.001738572120666504, "loss": 0.0162202721927315, "time": 1.1779246568679809, "epoch": 1, "memory": 15587, "step": 800} +{"lr": 0.00039796662066835946, "data_time": 0.001804184913635254, "loss": 0.01106295855715871, "time": 1.1196962594985962, "epoch": 1, "memory": 15587, "step": 900} +{"lr": 0.00039749012341636, "data_time": 0.0017992734909057617, "loss": 0.014533220883458852, "time": 1.1227311372756958, "epoch": 1, "memory": 15587, "step": 1000} +{"lr": 0.00039696386967593436, "data_time": 0.0019058942794799804, "loss": 0.009314351435750722, "time": 1.113140082359314, "epoch": 1, "memory": 15587, "step": 1100} +{"lr": 0.00039638799547763706, "data_time": 0.030929255485534667, "loss": 0.016852918826043606, "time": 1.0852095127105712, "epoch": 1, "memory": 15587, "step": 1200} +{"lr": 0.0003957626496783407, "data_time": 0.03611502647399902, "loss": 0.02064823657274246, "time": 1.09158034324646, "epoch": 1, "memory": 15587, "step": 1300} +{"lr": 0.0003950879939227595, "data_time": 0.001755666732788086, "loss": 0.020769246947020294, "time": 1.0547627449035644, "epoch": 1, "memory": 15587, "step": 1400} +{"lr": 0.0003943642026016663, "data_time": 0.0024544477462768556, "loss": 0.015398759581148624, "time": 1.0406849384307861, "epoch": 1, "memory": 15587, "step": 1500} +{"lr": 0.0003935914628068133, "data_time": 0.0016436100006103516, "loss": 0.010116389906033873, "time": 1.02152578830719, "epoch": 1, "memory": 15587, "step": 1600} +{"lr": 0.00039276997428257205, "data_time": 0.002260255813598633, "loss": 0.022971110278740527, "time": 1.0216881513595581, "epoch": 1, "memory": 15587, "step": 1700} +{"lr": 0.0003918999493743001, "data_time": 0.0017485857009887696, "loss": 0.02214900734834373, "time": 0.9366251945495605, "epoch": 1, "memory": 15587, "step": 1800} +{"lr": 0.00039098161297345475, "data_time": 0.0018000125885009766, "loss": 0.013530656974762679, "time": 0.9412976026535034, "epoch": 1, "memory": 15587, "step": 1900} +{"accuracy/top1": 88.17369079589844, "data_time": 0.3702001982721789, "time": 0.839593139188043, "step": 1} +{"lr": 0.0003894734390071047, "data_time": 0.01877894401550293, "loss": 0.015119136963039636, "time": 0.9577568769454956, "epoch": 2, "memory": 15587, "step": 2054} +{"lr": 0.0003884334872841784, "data_time": 0.17825229167938234, "loss": 0.009091089013963938, "time": 0.9782303810119629, "epoch": 2, "memory": 15585, "step": 2154} +{"lr": 0.00038734612010935703, "data_time": 0.17850701808929442, "loss": 0.017891387827694415, "time": 0.945639705657959, "epoch": 2, "memory": 15585, "step": 2254} +{"lr": 0.00038621161855458035, "data_time": 0.22520716190338136, "loss": 0.01324383676983416, "time": 0.9511027812957764, "epoch": 2, "memory": 15585, "step": 2354} +{"lr": 0.0003850302758754841, "data_time": 0.17745506763458252, "loss": 0.008567770197987556, "time": 0.9681565523147583, "epoch": 2, "memory": 15585, "step": 2454} +{"lr": 0.00038380239743559646, "data_time": 0.24744791984558107, "loss": 0.014836873067542911, "time": 0.9981346607208252, "epoch": 2, "memory": 15585, "step": 2554} +{"lr": 0.00038252830062740647, "data_time": 0.4144535779953003, "loss": 0.014212223328649997, "time": 0.9331673145294189, "epoch": 2, "memory": 15585, "step": 2654} +{"lr": 0.00038120831479032175, "data_time": 0.32656862735748293, "loss": 0.012860441068187355, "time": 0.9716259717941285, "epoch": 2, "memory": 15585, "step": 2754} +{"lr": 0.0003798427811255361, "data_time": 0.3410447835922241, "loss": 0.009475871594622732, "time": 0.9721494436264038, "epoch": 2, "memory": 15585, "step": 2854} +{"lr": 0.0003784320526078357, "data_time": 0.4131464004516602, "loss": 0.010852901544421912, "time": 0.9363669872283935, "epoch": 2, "memory": 15585, "step": 2954} +{"lr": 0.0003769764938943556, "data_time": 0.41784465312957764, "loss": 0.010025016637519002, "time": 0.9517346858978272, "epoch": 2, "memory": 15585, "step": 3054} +{"lr": 0.0003754764812303242, "data_time": 0.4260796308517456, "loss": 0.010024727461859584, "time": 0.9602128028869629, "epoch": 2, "memory": 15585, "step": 3154} +{"lr": 0.0003739324023518048, "data_time": 0.48997960090637205, "loss": 0.010866807214915753, "time": 1.0086214065551757, "epoch": 2, "memory": 15585, "step": 3254} +{"lr": 0.0003723446563854724, "data_time": 0.4330077886581421, "loss": 0.015846089040860534, "time": 0.9536540269851684, "epoch": 2, "memory": 15585, "step": 3354} +{"lr": 0.0003707136537454424, "data_time": 0.4909473657608032, "loss": 0.015231353789567947, "time": 1.0086801052093506, "epoch": 2, "memory": 15585, "step": 3454} +{"lr": 0.0003690398160271841, "data_time": 0.42433533668518064, "loss": 0.014469049172475934, "time": 0.9541850328445435, "epoch": 2, "memory": 15585, "step": 3554} +{"lr": 0.00036732357589854426, "data_time": 0.4355283260345459, "loss": 0.007716640597209334, "time": 0.9569052457809448, "epoch": 2, "memory": 15585, "step": 3654} +{"lr": 0.00036556537698790475, "data_time": 0.4240563869476318, "loss": 0.013134542619809508, "time": 0.9421380519866943, "epoch": 2, "memory": 15585, "step": 3754} +{"lr": 0.0003637656737695118, "data_time": 0.4751794099807739, "loss": 0.021589742554351688, "time": 0.9941707134246827, "epoch": 2, "memory": 15585, "step": 3854} +{"accuracy/top1": 94.76705932617188, "data_time": 0.3208686908086141, "time": 0.7103821833928426, "step": 2} +{"lr": 0.00036091403361175644, "data_time": 0.0025679349899291994, "loss": 0.009165691211819649, "time": 1.0242275953292848, "epoch": 3, "memory": 15585, "step": 4008} +{"lr": 0.00035901102535335625, "data_time": 0.006280827522277832, "loss": 0.004569197678938508, "time": 0.9417719841003418, "epoch": 3, "memory": 15585, "step": 4108} +{"lr": 0.00035706820701190153, "data_time": 0.0019710779190063475, "loss": 0.012781585427001118, "time": 0.9085953235626221, "epoch": 3, "memory": 15585, "step": 4208} +{"lr": 0.00035508608078364215, "data_time": 0.0034751653671264648, "loss": 0.013173878146335482, "time": 1.0007141351699829, "epoch": 3, "memory": 15585, "step": 4308} +{"lr": 0.00035306515902546755, "data_time": 0.002584242820739746, "loss": 0.0109096250962466, "time": 0.96571843624115, "epoch": 3, "memory": 15585, "step": 4408} +{"lr": 0.0003510059641224655, "data_time": 0.0023252487182617186, "loss": 0.0054708153009414675, "time": 0.9249018907546998, "epoch": 3, "memory": 15585, "step": 4508} +{"lr": 0.00034890902835289306, "data_time": 0.0017351150512695313, "loss": 0.005799282202497124, "time": 0.9141477346420288, "epoch": 3, "memory": 15585, "step": 4608} +{"lr": 0.00034677489375058753, "data_time": 0.001969265937805176, "loss": 0.010284630768001079, "time": 0.9485032558441162, "epoch": 3, "memory": 15585, "step": 4708} +{"lr": 0.00034460411196485875, "data_time": 0.0019747734069824217, "loss": 0.01762869395315647, "time": 0.9205304384231567, "epoch": 3, "memory": 15585, "step": 4808} +{"lr": 0.0003423972441178926, "data_time": 0.010460853576660156, "loss": 0.011661151982843876, "time": 0.9256266117095947, "epoch": 3, "memory": 15585, "step": 4908} +{"lr": 0.0003401548606597075, "data_time": 0.14806385040283204, "loss": 0.009646823955699801, "time": 0.9667104721069336, "epoch": 3, "memory": 15585, "step": 5008} +{"lr": 0.0003378775412207003, "data_time": 0.05621922016143799, "loss": 0.010691530350595713, "time": 0.9566173791885376, "epoch": 3, "memory": 15585, "step": 5108} +{"lr": 0.0003355658744618178, "data_time": 0.1323199987411499, "loss": 0.009256822476163506, "time": 0.9530836343765259, "epoch": 3, "memory": 15585, "step": 5208} +{"lr": 0.0003332204579223944, "data_time": 0.0348684549331665, "loss": 0.009746877616271377, "time": 0.9325660705566406, "epoch": 3, "memory": 15585, "step": 5308} +{"lr": 0.0003308418978656968, "data_time": 0.04287197589874268, "loss": 0.015595878195017576, "time": 0.996539568901062, "epoch": 3, "memory": 15585, "step": 5408} +{"lr": 0.0003284308091222099, "data_time": 0.11746449470520019, "loss": 0.009972090134397149, "time": 0.9147012948989868, "epoch": 3, "memory": 15585, "step": 5508} +{"lr": 0.0003259878149307108, "data_time": 0.19142618179321289, "loss": 0.0053694368805736305, "time": 0.9154141187667847, "epoch": 3, "memory": 15585, "step": 5608} +{"lr": 0.0003235135467771688, "data_time": 0.29831852912902834, "loss": 0.007030118349939584, "time": 0.9449497938156128, "epoch": 3, "memory": 15585, "step": 5708} +{"lr": 0.0003210086442315141, "data_time": 0.12527620792388916, "loss": 0.009467535465955735, "time": 0.9195688724517822, "epoch": 3, "memory": 15585, "step": 5808} +{"accuracy/top1": 97.40092468261719, "data_time": 0.28207352956136067, "time": 0.6702582120895386, "step": 3} +{"lr": 0.00031709267628953966, "data_time": 0.11172959804534913, "loss": 0.00649794265627861, "time": 0.9505893230438233, "epoch": 4, "memory": 15585, "step": 5962} +{"lr": 0.00031451289120303685, "data_time": 0.0017980337142944336, "loss": 0.007613994600251317, "time": 0.9671101331710815, "epoch": 4, "memory": 15585, "step": 6062} +{"lr": 0.000311904798290621, "data_time": 0.0014580249786376952, "loss": 0.011619580676779151, "time": 1.003280806541443, "epoch": 4, "memory": 15585, "step": 6162} +{"lr": 0.00030926907171438685, "data_time": 0.0017405271530151368, "loss": 0.00666549876332283, "time": 0.9539995670318604, "epoch": 4, "memory": 15585, "step": 6262} +{"lr": 0.000306606392779414, "data_time": 0.002644658088684082, "loss": 0.011141629936173559, "time": 0.968813157081604, "epoch": 4, "memory": 15585, "step": 6362} +{"lr": 0.0003039174497576579, "data_time": 0.0018522262573242188, "loss": 0.008385690348222851, "time": 0.9816806793212891, "epoch": 4, "memory": 15585, "step": 6462} +{"lr": 0.0003012029377100387, "data_time": 0.0018495559692382813, "loss": 0.009659701539203525, "time": 0.9766746759414673, "epoch": 4, "memory": 15585, "step": 6562} +{"lr": 0.00029846355830677674, "data_time": 0.03559055328369141, "loss": 0.0036875926423817875, "time": 0.9105247259140015, "epoch": 4, "memory": 15585, "step": 6662} +{"lr": 0.0002957000196460183, "data_time": 0.13139536380767822, "loss": 0.007654532138258219, "time": 0.9805450916290284, "epoch": 4, "memory": 15585, "step": 6762} +{"lr": 0.0002929130360708003, "data_time": 0.5423965692520142, "loss": 0.0054247718770056965, "time": 1.0859703302383423, "epoch": 4, "memory": 15585, "step": 6862} +{"lr": 0.0002901033279844016, "data_time": 0.45169754028320314, "loss": 0.007872724533081054, "time": 0.9707653522491455, "epoch": 4, "memory": 15585, "step": 6962} +{"lr": 0.00028727162166412623, "data_time": 0.41893854141235354, "loss": 0.005585976736620068, "time": 0.9506667137145997, "epoch": 4, "memory": 15585, "step": 7062} +{"lr": 0.000284418649073569, "data_time": 0.41764733791351316, "loss": 0.0061735509429126974, "time": 0.9365954399108887, "epoch": 4, "memory": 15585, "step": 7162} +{"lr": 0.00028154514767341203, "data_time": 0.49697954654693605, "loss": 0.0049189492128789425, "time": 1.0262855052948, "epoch": 4, "memory": 15585, "step": 7262} +{"lr": 0.0002786518602307988, "data_time": 0.441841721534729, "loss": 0.0049474481493234634, "time": 0.9627981424331665, "epoch": 4, "memory": 15585, "step": 7362} +{"lr": 0.0002757395346273384, "data_time": 0.5250383377075195, "loss": 0.007277003536000848, "time": 1.0449779272079467, "epoch": 4, "memory": 15585, "step": 7462} +{"lr": 0.0002728089236657853, "data_time": 0.4268195390701294, "loss": 0.01237349109724164, "time": 0.9489672899246215, "epoch": 4, "memory": 15585, "step": 7562} +{"lr": 0.0002698607848754497, "data_time": 0.4164048433303833, "loss": 0.006725517986342311, "time": 0.9444144248962403, "epoch": 4, "memory": 15585, "step": 7662} +{"lr": 0.0002668958803163847, "data_time": 0.4745892286300659, "loss": 0.007470806688070297, "time": 0.9946723699569702, "epoch": 4, "memory": 15585, "step": 7762} +{"accuracy/top1": 98.79267120361328, "data_time": 0.3012315114339193, "time": 0.6895922422409058, "step": 4} +{"lr": 0.00026229890681650394, "data_time": 0.0768885612487793, "loss": 0.00528542841784656, "time": 0.9828031301498413, "epoch": 5, "memory": 15585, "step": 7916} +{"lr": 0.00025929487224058475, "data_time": 0.1465217113494873, "loss": 0.0072232729755342005, "time": 0.9394273281097412, "epoch": 5, "memory": 15585, "step": 8016} +{"lr": 0.0002562768030631272, "data_time": 0.001854109764099121, "loss": 0.004652506532147526, "time": 0.9927838325500489, "epoch": 5, "memory": 15585, "step": 8116} +{"lr": 0.0002532454794203871, "data_time": 0.02627689838409424, "loss": 0.0052061258815228936, "time": 0.9198421239852905, "epoch": 5, "memory": 15585, "step": 8216} +{"lr": 0.0002502016848747476, "data_time": 0.056214523315429685, "loss": 0.004920996772125363, "time": 0.969770073890686, "epoch": 5, "memory": 15585, "step": 8316} +{"lr": 0.0002471462062121758, "data_time": 0.16587018966674805, "loss": 0.0043186724185943605, "time": 0.9476407766342163, "epoch": 5, "memory": 15585, "step": 8416} +{"lr": 0.00024407983323884952, "data_time": 0.2657217502593994, "loss": 0.005234378017485141, "time": 0.9712835073471069, "epoch": 5, "memory": 15585, "step": 8516} +{"lr": 0.0002410033585770006, "data_time": 0.3135234355926514, "loss": 0.0069424071814864876, "time": 0.9429662942886352, "epoch": 5, "memory": 15585, "step": 8616} +{"lr": 0.00023791757746003292, "data_time": 0.448268723487854, "loss": 0.00430152346380055, "time": 0.976430082321167, "epoch": 5, "memory": 15585, "step": 8716} +{"lr": 0.00023482328752696182, "data_time": 0.4581224679946899, "loss": 0.006737135397270322, "time": 0.9894840717315674, "epoch": 5, "memory": 15585, "step": 8816} +{"lr": 0.00023172128861623345, "data_time": 0.44569220542907717, "loss": 0.0058538383338600395, "time": 0.9701451778411865, "epoch": 5, "memory": 15585, "step": 8916} +{"lr": 0.000228612382558976, "data_time": 0.459169602394104, "loss": 0.005569694563746452, "time": 1.0004676580429077, "epoch": 5, "memory": 15585, "step": 9016} +{"lr": 0.00022549737297173692, "data_time": 0.40810461044311525, "loss": 0.005624811025336384, "time": 0.932976484298706, "epoch": 5, "memory": 15585, "step": 9116} +{"lr": 0.00022237706504875477, "data_time": 0.4501085996627808, "loss": 0.003529950091615319, "time": 0.9704623937606811, "epoch": 5, "memory": 15585, "step": 9216} +{"lr": 0.00021925226535382818, "data_time": 0.4203763246536255, "loss": 0.006921368557959795, "time": 0.947316837310791, "epoch": 5, "memory": 15585, "step": 9316} +{"lr": 0.0002161237816118266, "data_time": 0.38816003799438475, "loss": 0.006047765770927071, "time": 0.9536850214004516, "epoch": 5, "memory": 15585, "step": 9416} +{"lr": 0.00021299242249990383, "data_time": 0.4218069314956665, "loss": 0.0039038790855556726, "time": 0.9412720441818238, "epoch": 5, "memory": 15585, "step": 9516} +{"lr": 0.00020985899743846363, "data_time": 0.41132094860076907, "loss": 0.005127867497503757, "time": 0.9767580270767212, "epoch": 5, "memory": 15585, "step": 9616} +{"lr": 0.00020672431638193363, "data_time": 0.3408376693725586, "loss": 0.005221713194623589, "time": 0.9531792163848877, "epoch": 5, "memory": 15585, "step": 9716} +{"accuracy/top1": 99.405029296875, "data_time": 0.2662598927815755, "time": 0.6586158990859985, "step": 5} +{"lr": 0.00020189632111931318, "data_time": 0.001739192008972168, "loss": 0.003452608222141862, "time": 0.9475012302398682, "epoch": 6, "memory": 15585, "step": 9870} +{"lr": 0.00019876209288860824, "data_time": 0.0024026155471801756, "loss": 0.005942829232662916, "time": 0.9519016027450562, "epoch": 6, "memory": 15585, "step": 9970} +{"lr": 0.00019562947708532286, "data_time": 0.0029773950576782227, "loss": 0.0071106948889791965, "time": 0.951548433303833, "epoch": 6, "memory": 15585, "step": 10070} +{"lr": 0.00019249928345470104, "data_time": 0.0027435302734375, "loss": 0.004475785465911031, "time": 0.9630252361297608, "epoch": 6, "memory": 15585, "step": 10170} +{"lr": 0.00018937232111588298, "data_time": 0.0018990755081176758, "loss": 0.003966807201504707, "time": 0.9069278717041016, "epoch": 6, "memory": 15585, "step": 10270} +{"lr": 0.000186249398352757, "data_time": 0.0015522956848144532, "loss": 0.008469361485913396, "time": 0.9342559576034546, "epoch": 6, "memory": 15585, "step": 10370} +{"lr": 0.0001831313224050277, "data_time": 0.40353171825408934, "loss": 0.004387742327526212, "time": 0.9232488870620728, "epoch": 6, "memory": 15585, "step": 10470} +{"lr": 0.00018001889925955291, "data_time": 0.39816813468933104, "loss": 0.0028657491318881513, "time": 0.9168015956878662, "epoch": 6, "memory": 15585, "step": 10570} +{"lr": 0.00017691293344200646, "data_time": 0.4150885820388794, "loss": 0.004340462340041995, "time": 0.9349114656448364, "epoch": 6, "memory": 15585, "step": 10670} +{"lr": 0.00017381422780891689, "data_time": 0.42396697998046873, "loss": 0.005025573493912816, "time": 0.9522564172744751, "epoch": 6, "memory": 15585, "step": 10770} +{"lr": 0.00017072358334013806, "data_time": 0.434910249710083, "loss": 0.0037409603130072354, "time": 0.9580995798110962, "epoch": 6, "memory": 15585, "step": 10870} +{"lr": 0.00016764179893180575, "data_time": 0.41643226146698, "loss": 0.00355774681083858, "time": 0.9355035066604614, "epoch": 6, "memory": 15585, "step": 10970} +{"lr": 0.00016456967118983176, "data_time": 0.3971834659576416, "loss": 0.0021645229775458575, "time": 0.916974139213562, "epoch": 6, "memory": 15585, "step": 11070} +{"lr": 0.00016150799422398984, "data_time": 0.44319169521331786, "loss": 0.0022221977822482588, "time": 0.9627624034881592, "epoch": 6, "memory": 15585, "step": 11170} +{"lr": 0.0001584575594426489, "data_time": 0.40769870281219484, "loss": 0.004099268745630979, "time": 0.9255609512329102, "epoch": 6, "memory": 15585, "step": 11270} +{"lr": 0.00015541915534820135, "data_time": 0.44083538055419924, "loss": 0.002200006926432252, "time": 0.9688750028610229, "epoch": 6, "memory": 15585, "step": 11370} +{"lr": 0.00015239356733324542, "data_time": 0.4261681795120239, "loss": 0.006635595066472888, "time": 0.9478034973144531, "epoch": 6, "memory": 15585, "step": 11470} +{"lr": 0.00014938157747756997, "data_time": 0.4163465738296509, "loss": 0.003441391373053193, "time": 0.9390012264251709, "epoch": 6, "memory": 15585, "step": 11570} +{"lr": 0.00014638396434599565, "data_time": 0.4195083141326904, "loss": 0.000967968488112092, "time": 0.9382271766662598, "epoch": 6, "memory": 15585, "step": 11670} +{"accuracy/top1": 99.6207504272461, "data_time": 0.26066848436991374, "time": 0.6483653624852498, "step": 6} +{"lr": 0.00014179754513556176, "data_time": 0.0016521453857421876, "loss": 0.005475955689325929, "time": 0.9159265518188476, "epoch": 7, "memory": 15585, "step": 11824} +{"lr": 0.00013883992334674444, "data_time": 0.0019083738327026368, "loss": 0.006218972522765398, "time": 0.9376081705093384, "epoch": 7, "memory": 15585, "step": 11924} +{"lr": 0.00013589940317867467, "data_time": 0.002809476852416992, "loss": 0.002176971593871713, "time": 1.4418408155441285, "epoch": 7, "memory": 15585, "step": 12024} +{"lr": 0.0001329767447220783, "data_time": 0.0026750564575195312, "loss": 0.002354126889258623, "time": 0.9647247791290283, "epoch": 7, "memory": 15585, "step": 12124} +{"lr": 0.00013007270345063488, "data_time": 0.002393651008605957, "loss": 0.0024485097732394935, "time": 0.9378865480422973, "epoch": 7, "memory": 15585, "step": 12224} +{"lr": 0.00012718803002569412, "data_time": 0.0017752647399902344, "loss": 0.0034284644760191442, "time": 0.9883497476577758, "epoch": 7, "memory": 15585, "step": 12324} +{"lr": 0.0001243234701022397, "data_time": 0.0025546073913574217, "loss": 0.005110947135835886, "time": 0.9463302373886109, "epoch": 7, "memory": 15585, "step": 12424} +{"lr": 0.00012147976413614597, "data_time": 0.0019263744354248047, "loss": 0.0031996916048228742, "time": 0.9885228395462036, "epoch": 7, "memory": 15585, "step": 12524} +{"lr": 0.0001186576471927784, "data_time": 0.001650834083557129, "loss": 0.002845912938937545, "time": 0.9857007265090942, "epoch": 7, "memory": 15585, "step": 12624} +{"lr": 0.00011585784875698867, "data_time": 0.020549511909484862, "loss": 0.0007590523455291986, "time": 0.9341190338134766, "epoch": 7, "memory": 15585, "step": 12724} +{"lr": 0.00011308109254454993, "data_time": 0.13373348712921143, "loss": 0.005037216423079372, "time": 0.9964096069335937, "epoch": 7, "memory": 15585, "step": 12824} +{"lr": 0.00011032809631508466, "data_time": 0.009465742111206054, "loss": 0.0036153506953269243, "time": 0.9257693767547608, "epoch": 7, "memory": 15585, "step": 12924} +{"lr": 0.00010759957168653253, "data_time": 0.07855479717254639, "loss": 0.001611655903980136, "time": 0.9202292680740356, "epoch": 7, "memory": 15585, "step": 13024} +{"lr": 0.0001048962239512054, "data_time": 0.24015531539916993, "loss": 0.006936500268056989, "time": 1.0145259618759155, "epoch": 7, "memory": 15585, "step": 13124} +{"lr": 0.00010221875189347644, "data_time": 0.46049938201904295, "loss": 0.005218543158844113, "time": 0.9803165197372437, "epoch": 7, "memory": 15585, "step": 13224} +{"lr": 9.95678476091537e-05, "data_time": 0.3974884510040283, "loss": 0.002453421102836728, "time": 0.9266520977020264, "epoch": 7, "memory": 15585, "step": 13324} +{"lr": 9.694419632657999e-05, "data_time": 0.4210886716842651, "loss": 0.0021715907845646144, "time": 0.9405019521713257, "epoch": 7, "memory": 15585, "step": 13424} +{"lr": 9.434847622951059e-05, "data_time": 0.3621471405029297, "loss": 0.004599348129704595, "time": 0.8858633041381836, "epoch": 7, "memory": 15585, "step": 13524} +{"lr": 9.178135828180905e-05, "data_time": 0.5840172529220581, "loss": 0.001961578754708171, "time": 1.105517578125, "epoch": 7, "memory": 15585, "step": 13624} +{"accuracy/top1": 99.71817779541016, "data_time": 0.4316521247227987, "time": 0.8269522666931153, "step": 7} +{"lr": 8.788546579134055e-05, "data_time": 0.0017841815948486327, "loss": 0.0019957524724304675, "time": 0.931977367401123, "epoch": 8, "memory": 15585, "step": 13778} +{"lr": 8.539396307206318e-05, "data_time": 0.0018756628036499024, "loss": 0.0039055104367434978, "time": 0.901304030418396, "epoch": 8, "memory": 15585, "step": 13878} +{"lr": 8.293337714106816e-05, "data_time": 0.002025413513183594, "loss": 0.0034841256681829693, "time": 0.9605952978134156, "epoch": 8, "memory": 15585, "step": 13978} +{"lr": 8.05043440315878e-05, "data_time": 0.0017398834228515626, "loss": 0.0014356670435518026, "time": 0.9542671918869019, "epoch": 8, "memory": 15585, "step": 14078} +{"lr": 7.810749162081216e-05, "data_time": 0.0016051769256591798, "loss": 0.0009449648205190897, "time": 0.9690500736236572, "epoch": 8, "memory": 15585, "step": 14178} +{"lr": 7.574343946758993e-05, "data_time": 0.00210416316986084, "loss": 0.0039619202725589275, "time": 1.0130390882492066, "epoch": 8, "memory": 15585, "step": 14278} +{"lr": 7.34127986522793e-05, "data_time": 0.002979564666748047, "loss": 0.008022097731009125, "time": 0.9794174909591675, "epoch": 8, "memory": 15585, "step": 14378} +{"lr": 7.111617161879136e-05, "data_time": 0.002597999572753906, "loss": 0.0021727455779910086, "time": 0.9529072046279907, "epoch": 8, "memory": 15585, "step": 14478} +{"lr": 6.885415201886485e-05, "data_time": 0.22874774932861328, "loss": 0.002567023318260908, "time": 0.9731489658355713, "epoch": 8, "memory": 15585, "step": 14578} +{"lr": 6.662732455861337e-05, "data_time": 0.40278670787811277, "loss": 0.002550022955983877, "time": 0.9697470426559448, "epoch": 8, "memory": 15585, "step": 14678} +{"lr": 6.443626484738661e-05, "data_time": 0.42852249145507815, "loss": 0.0007593533955514431, "time": 0.948227334022522, "epoch": 8, "memory": 15585, "step": 14778} +{"lr": 6.22815392489815e-05, "data_time": 0.4181420087814331, "loss": 0.0022932265419512987, "time": 0.9465721845626831, "epoch": 8, "memory": 15585, "step": 14878} +{"lr": 6.016370473524328e-05, "data_time": 0.42673871517181394, "loss": 0.003924101451411843, "time": 0.9452044010162354, "epoch": 8, "memory": 15585, "step": 14978} +{"lr": 5.80833087420956e-05, "data_time": 0.4155222177505493, "loss": 0.002293054014444351, "time": 0.9513051986694336, "epoch": 8, "memory": 15585, "step": 15078} +{"lr": 5.604088902803397e-05, "data_time": 0.4702103853225708, "loss": 0.001514874119311571, "time": 0.9896433115005493, "epoch": 8, "memory": 15585, "step": 15178} +{"lr": 5.403697353512148e-05, "data_time": 0.44001338481903074, "loss": 0.004679047269746661, "time": 0.9795157670974731, "epoch": 8, "memory": 15585, "step": 15278} +{"lr": 5.207208025252174e-05, "data_time": 0.45709240436553955, "loss": 0.0035715472418814897, "time": 0.9803754568099976, "epoch": 8, "memory": 15585, "step": 15378} +{"lr": 5.014671708260458e-05, "data_time": 0.4312600612640381, "loss": 0.0014895839150995017, "time": 0.9742679119110107, "epoch": 8, "memory": 15585, "step": 15478} +{"lr": 4.826138170965952e-05, "data_time": 0.460954475402832, "loss": 0.0054690070450305935, "time": 0.9832231998443604, "epoch": 8, "memory": 15585, "step": 15578} +{"accuracy/top1": 99.7773208618164, "data_time": 0.27094720204671224, "time": 0.6685602982838948, "step": 8} +{"lr": 4.5437373040402475e-05, "data_time": 0.0024146318435668947, "loss": 0.00034796479158103464, "time": 0.9479226350784302, "epoch": 9, "memory": 15585, "step": 15732} +{"lr": 4.3655875036148156e-05, "data_time": 0.002630448341369629, "loss": 0.0055925148539245125, "time": 0.970571231842041, "epoch": 9, "memory": 15585, "step": 15832} +{"lr": 4.19160826370641e-05, "data_time": 0.0027977943420410155, "loss": 0.0009660322219133377, "time": 0.9527229547500611, "epoch": 9, "memory": 15585, "step": 15932} +{"lr": 4.021844555952455e-05, "data_time": 0.0020281314849853516, "loss": 0.003684419672936201, "time": 0.9322951793670654, "epoch": 9, "memory": 15585, "step": 16032} +{"lr": 3.8563402623236864e-05, "data_time": 0.0024892330169677735, "loss": 0.002301778132095933, "time": 0.9459209203720093, "epoch": 9, "memory": 15585, "step": 16132} +{"lr": 3.695138163781135e-05, "data_time": 0.002077054977416992, "loss": 0.0024881566409021617, "time": 0.9121654272079468, "epoch": 9, "memory": 15585, "step": 16232} +{"lr": 3.538279929217756e-05, "data_time": 0.00218348503112793, "loss": 0.00040242779068648814, "time": 0.9328038215637207, "epoch": 9, "memory": 15585, "step": 16332} +{"lr": 3.385806104687531e-05, "data_time": 0.0023269414901733398, "loss": 0.0030994636472314596, "time": 0.969316053390503, "epoch": 9, "memory": 15585, "step": 16432} +{"lr": 3.237756102924734e-05, "data_time": 0.002310299873352051, "loss": 0.0015743127558380365, "time": 0.9078565835952759, "epoch": 9, "memory": 15585, "step": 16532} +{"lr": 3.094168193156213e-05, "data_time": 0.062359070777893065, "loss": 0.0036647174041718245, "time": 0.9517327070236206, "epoch": 9, "memory": 15585, "step": 16632} +{"lr": 2.95507949120924e-05, "data_time": 0.04834351539611816, "loss": 0.002974098129197955, "time": 0.9905012607574463, "epoch": 9, "memory": 15585, "step": 16732} +{"lr": 2.8205259499174602e-05, "data_time": 0.001982450485229492, "loss": 0.0016547942534089089, "time": 0.9796242475509643, "epoch": 9, "memory": 15585, "step": 16832} +{"lr": 2.690542349827549e-05, "data_time": 0.02330169677734375, "loss": 0.0021105783060193064, "time": 0.9274474620819092, "epoch": 9, "memory": 15585, "step": 16932} +{"lr": 2.565162290208795e-05, "data_time": 0.0024168968200683595, "loss": 0.0033429506700485946, "time": 0.9993813991546631, "epoch": 9, "memory": 15585, "step": 17032} +{"lr": 2.4444181803680975e-05, "data_time": 0.20685017108917236, "loss": 0.0005877256859093905, "time": 1.05042884349823, "epoch": 9, "memory": 15585, "step": 17132} +{"lr": 2.32834123127253e-05, "data_time": 0.43677310943603515, "loss": 0.0036239826120436193, "time": 0.9767820596694946, "epoch": 9, "memory": 15585, "step": 17232} +{"lr": 2.2169614474816763e-05, "data_time": 0.42950997352600095, "loss": 0.0007100783754140138, "time": 0.9541327714920044, "epoch": 9, "memory": 15585, "step": 17332} +{"lr": 2.110307619391749e-05, "data_time": 0.38581721782684325, "loss": 0.001673663267865777, "time": 0.9118571996688842, "epoch": 9, "memory": 15585, "step": 17432} +{"lr": 2.008407315793661e-05, "data_time": 0.4806783437728882, "loss": 0.00017982046119868755, "time": 0.9980745792388916, "epoch": 9, "memory": 15585, "step": 17532} +{"accuracy/top1": 99.83995056152344, "data_time": 0.2823968331019084, "time": 0.6790300051371256, "step": 9} +{"lr": 1.8608381954463387e-05, "data_time": 0.36521923542022705, "loss": 0.00047856238670647145, "time": 0.9121445417404175, "epoch": 10, "memory": 15585, "step": 17686} +{"lr": 1.771127410231484e-05, "data_time": 0.37261722087860105, "loss": 0.0006369306705892086, "time": 0.9656543016433716, "epoch": 10, "memory": 15585, "step": 17786} +{"lr": 1.6862578237010542e-05, "data_time": 0.24154651165008545, "loss": 0.003397755604237318, "time": 0.9507620811462403, "epoch": 10, "memory": 15585, "step": 17886} +{"lr": 1.6062513736694383e-05, "data_time": 0.18485028743743898, "loss": 0.0015289057977497579, "time": 0.9999956846237182, "epoch": 10, "memory": 15585, "step": 17986} +{"lr": 1.531128740886036e-05, "data_time": 0.38703265190124514, "loss": 0.0021364753134548663, "time": 1.0022279024124146, "epoch": 10, "memory": 15585, "step": 18086} +{"lr": 1.4609093436895328e-05, "data_time": 0.4147106885910034, "loss": 0.000597511325031519, "time": 0.9470914363861084, "epoch": 10, "memory": 15585, "step": 18186} +{"lr": 1.3956113329884566e-05, "data_time": 0.3815218210220337, "loss": 0.0022642647847533224, "time": 0.9054733037948608, "epoch": 10, "memory": 15585, "step": 18286} +{"lr": 1.335251587569401e-05, "data_time": 0.40713484287261964, "loss": 0.00237970226444304, "time": 0.9654510974884033, "epoch": 10, "memory": 15585, "step": 18386} +{"lr": 1.2798457097340337e-05, "data_time": 0.4161717176437378, "loss": 0.0007964265532791615, "time": 0.9454386711120606, "epoch": 10, "memory": 15585, "step": 18486} +{"lr": 1.229408021266094e-05, "data_time": 0.36980977058410647, "loss": 0.0018288569524884224, "time": 0.9056573629379272, "epoch": 10, "memory": 15585, "step": 18586} +{"lr": 1.1839515597293723e-05, "data_time": 0.3753602743148804, "loss": 0.002611189801245928, "time": 0.9475775003433228, "epoch": 10, "memory": 15585, "step": 18686} +{"lr": 1.1434880750976406e-05, "data_time": 0.40407772064208985, "loss": 0.0027824061922729014, "time": 0.9871024131774903, "epoch": 10, "memory": 15585, "step": 18786} +{"lr": 1.108028026717431e-05, "data_time": 0.40224418640136717, "loss": 0.0009329441469162703, "time": 0.9314643859863281, "epoch": 10, "memory": 15585, "step": 18886} +{"lr": 1.0775805806044138e-05, "data_time": 0.4564747095108032, "loss": 0.00026018787175416944, "time": 0.9844456911087036, "epoch": 10, "memory": 15585, "step": 18986} +{"lr": 1.0521536070740776e-05, "data_time": 0.42733771800994874, "loss": 0.001569018978625536, "time": 0.954958152770996, "epoch": 10, "memory": 15585, "step": 19086} +{"lr": 1.03175367870735e-05, "data_time": 0.3768294095993042, "loss": 0.0011358255986124277, "time": 0.8935638189315795, "epoch": 10, "memory": 15585, "step": 19186} +{"lr": 1.0163860686516628e-05, "data_time": 0.39269509315490725, "loss": 0.000978016061708331, "time": 0.9295082092285156, "epoch": 10, "memory": 15585, "step": 19286} +{"lr": 1.0060547492578997e-05, "data_time": 0.47974233627319335, "loss": 0.000620917696505785, "time": 1.0009328842163085, "epoch": 10, "memory": 15585, "step": 19386} +{"lr": 1.0007623910535892e-05, "data_time": 0.44156646728515625, "loss": 0.006867260346189142, "time": 0.9594136714935303, "epoch": 10, "memory": 15585, "step": 19486} +{"accuracy/top1": 99.88866424560547, "data_time": 0.2708349068959554, "time": 0.6577602783838908, "step": 10} diff --git a/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_5e-1/20230604_164906/vis_data/vis_image/0000000.png_1.png b/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_5e-1/20230604_164906/vis_data/vis_image/0000000.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..404799d1a523db2e6c1fed7012b0b04e9d79edfc Binary files /dev/null and b/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_5e-1/20230604_164906/vis_data/vis_image/0000000.png_1.png differ diff --git a/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_5e-1/20230604_164906/vis_data/vis_image/0000000.png_10.png b/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_5e-1/20230604_164906/vis_data/vis_image/0000000.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..0ff126a059b76f0995161cf3a69751dd5cb614cd Binary files /dev/null and b/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_5e-1/20230604_164906/vis_data/vis_image/0000000.png_10.png differ diff --git a/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_5e-1/20230604_164906/vis_data/vis_image/0000000.png_2.png b/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_5e-1/20230604_164906/vis_data/vis_image/0000000.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..0dab068e5ef81adad5a085bec4535bb7696b1bb4 Binary files /dev/null and b/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_5e-1/20230604_164906/vis_data/vis_image/0000000.png_2.png differ diff --git a/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_5e-1/20230604_164906/vis_data/vis_image/0000000.png_3.png b/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_5e-1/20230604_164906/vis_data/vis_image/0000000.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..7374347bcb369e101aa90dcd6dcd58f277a3b5c8 Binary files /dev/null and b/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_5e-1/20230604_164906/vis_data/vis_image/0000000.png_3.png differ diff --git a/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_5e-1/20230604_164906/vis_data/vis_image/0000000.png_4.png b/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_5e-1/20230604_164906/vis_data/vis_image/0000000.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..8479c74200bef559b36938e8a8a554c8e9e257e5 Binary files /dev/null and b/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_5e-1/20230604_164906/vis_data/vis_image/0000000.png_4.png differ diff --git a/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_5e-1/20230604_164906/vis_data/vis_image/0000000.png_5.png b/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_5e-1/20230604_164906/vis_data/vis_image/0000000.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..3fba49c305ae529766882f18731a3b438c72d0e3 Binary files /dev/null and b/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_5e-1/20230604_164906/vis_data/vis_image/0000000.png_5.png differ diff --git a/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_5e-1/20230604_164906/vis_data/vis_image/0000000.png_6.png b/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_5e-1/20230604_164906/vis_data/vis_image/0000000.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..89889bee86df930c2b4936768fbd07459ee006f3 Binary files /dev/null and b/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_5e-1/20230604_164906/vis_data/vis_image/0000000.png_6.png differ diff --git a/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_5e-1/20230604_164906/vis_data/vis_image/0000000.png_7.png b/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_5e-1/20230604_164906/vis_data/vis_image/0000000.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..d2bbc326605a567e0fa60f5954a57a0776d5ead0 Binary files /dev/null and b/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_5e-1/20230604_164906/vis_data/vis_image/0000000.png_7.png differ diff --git a/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_5e-1/20230604_164906/vis_data/vis_image/0000000.png_8.png b/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_5e-1/20230604_164906/vis_data/vis_image/0000000.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..4638c1f6682fd6c4a3c9e03054491142c8b12071 Binary files /dev/null and b/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_5e-1/20230604_164906/vis_data/vis_image/0000000.png_8.png differ diff --git a/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_5e-1/20230604_164906/vis_data/vis_image/0000000.png_9.png b/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_5e-1/20230604_164906/vis_data/vis_image/0000000.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..f9b7957f99e64f9034574d20e703d3c5b0210b33 Binary files /dev/null and b/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_5e-1/20230604_164906/vis_data/vis_image/0000000.png_9.png differ diff --git a/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_5e-1/20230604_164906/vis_data/vis_image/000005174.jpg_1.png b/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_5e-1/20230604_164906/vis_data/vis_image/000005174.jpg_1.png new file mode 100644 index 0000000000000000000000000000000000000000..2ec6e1472e792be326260c59bdbe308f879ea675 Binary files /dev/null and b/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_5e-1/20230604_164906/vis_data/vis_image/000005174.jpg_1.png differ diff --git a/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_5e-1/20230604_164906/vis_data/vis_image/000005174.jpg_10.png b/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_5e-1/20230604_164906/vis_data/vis_image/000005174.jpg_10.png new file mode 100644 index 0000000000000000000000000000000000000000..0d228ec4c1e80b00652f744534e75c0b843b80f6 Binary files /dev/null and b/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_5e-1/20230604_164906/vis_data/vis_image/000005174.jpg_10.png differ diff --git a/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_5e-1/20230604_164906/vis_data/vis_image/000005174.jpg_2.png b/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_5e-1/20230604_164906/vis_data/vis_image/000005174.jpg_2.png new file mode 100644 index 0000000000000000000000000000000000000000..bc3448bb1bf48f049516008d2b51ca516f0de0b6 Binary files /dev/null and b/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_5e-1/20230604_164906/vis_data/vis_image/000005174.jpg_2.png differ diff --git a/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_5e-1/20230604_164906/vis_data/vis_image/000005174.jpg_3.png b/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_5e-1/20230604_164906/vis_data/vis_image/000005174.jpg_3.png new file mode 100644 index 0000000000000000000000000000000000000000..d15109f4c45bbc0686f32008985ee0cdc945b6c8 Binary files /dev/null and b/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_5e-1/20230604_164906/vis_data/vis_image/000005174.jpg_3.png differ diff --git a/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_5e-1/20230604_164906/vis_data/vis_image/000005174.jpg_4.png b/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_5e-1/20230604_164906/vis_data/vis_image/000005174.jpg_4.png new file mode 100644 index 0000000000000000000000000000000000000000..e5f8ab085cb02867fd21c5772298a100efa3c06a Binary files /dev/null and b/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_5e-1/20230604_164906/vis_data/vis_image/000005174.jpg_4.png differ diff --git a/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_5e-1/20230604_164906/vis_data/vis_image/000005174.jpg_5.png b/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_5e-1/20230604_164906/vis_data/vis_image/000005174.jpg_5.png new file mode 100644 index 0000000000000000000000000000000000000000..3acb90dacdd853c5904e8a30118ee03c5b70e8da Binary files /dev/null and b/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_5e-1/20230604_164906/vis_data/vis_image/000005174.jpg_5.png differ diff --git a/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_5e-1/20230604_164906/vis_data/vis_image/000005174.jpg_6.png b/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_5e-1/20230604_164906/vis_data/vis_image/000005174.jpg_6.png new file mode 100644 index 0000000000000000000000000000000000000000..4c836b711cc22e560e1a324f9705e13f5cad7166 Binary files /dev/null and b/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_5e-1/20230604_164906/vis_data/vis_image/000005174.jpg_6.png differ diff --git a/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_5e-1/20230604_164906/vis_data/vis_image/000005174.jpg_7.png b/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_5e-1/20230604_164906/vis_data/vis_image/000005174.jpg_7.png new file mode 100644 index 0000000000000000000000000000000000000000..53d1c7b5a4dae2c5f25a9186379268051b13349e Binary files /dev/null and b/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_5e-1/20230604_164906/vis_data/vis_image/000005174.jpg_7.png differ diff --git a/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_5e-1/20230604_164906/vis_data/vis_image/000005174.jpg_8.png b/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_5e-1/20230604_164906/vis_data/vis_image/000005174.jpg_8.png new file mode 100644 index 0000000000000000000000000000000000000000..a0f7e3e456d5f51312384d5be0cafb7dbfa55113 Binary files /dev/null and b/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_5e-1/20230604_164906/vis_data/vis_image/000005174.jpg_8.png differ diff --git a/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_5e-1/20230604_164906/vis_data/vis_image/000005174.jpg_9.png b/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_5e-1/20230604_164906/vis_data/vis_image/000005174.jpg_9.png new file mode 100644 index 0000000000000000000000000000000000000000..5162b03fef7abde8c8301204c82ae2873f7acffa Binary files /dev/null and b/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_5e-1/20230604_164906/vis_data/vis_image/000005174.jpg_9.png differ diff --git a/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_5e-1/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_5e-1.py b/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_5e-1/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_5e-1.py new file mode 100644 index 0000000000000000000000000000000000000000..c6e47050fb7e76afd481119b9be206927b1f3b2d --- /dev/null +++ b/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_5e-1/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_5e-1.py @@ -0,0 +1,247 @@ +optim_wrapper = dict( + optimizer=dict( + type='AdamW', + lr=0.0004, + weight_decay=0.05, + eps=1e-08, + betas=(0.9, 0.999), + _scope_='mmpretrain'), + paramwise_cfg=dict( + norm_decay_mult=0.0, + bias_decay_mult=0.0, + flat_decay_mult=0.0, + custom_keys=dict({ + '.absolute_pos_embed': dict(decay_mult=0.0), + '.relative_position_bias_table': dict(decay_mult=0.0) + })), + type='AmpOptimWrapper', + dtype='bfloat16', + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=False, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=4096) +model = dict( + type='ImageClassifier', + backbone=dict( + frozen_stages=2, + type='ConvNeXt', + arch='small', + drop_path_rate=0.4, + init_cfg=dict( + type='Pretrained', + checkpoint= + 'https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth', + prefix='backbone')), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=768, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + init_cfg=None), + init_cfg=dict( + type='TruncNormal', layer=['Conv2d', 'Linear'], std=0.02, bias=0.0), + train_cfg=None) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stablediffusionV1-5R2-dpmsolver-25-1m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/cc1m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_5e-1' diff --git a/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_5e-1/epoch_1.pth b/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_5e-1/epoch_1.pth new file mode 100644 index 0000000000000000000000000000000000000000..3edf92a8b6b5048a3a890b6aebc740a1e7f3d10f --- /dev/null +++ b/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_5e-1/epoch_1.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:5c47b33b34ea28258d4ceea5257e17680e1305d9e1a9725e7504cf802e8da735 +size 782110490 diff --git a/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_5e-1/epoch_10.pth b/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_5e-1/epoch_10.pth new file mode 100644 index 0000000000000000000000000000000000000000..9ec9f4d6821c7d8551cec3f4670189b6104f8a55 --- /dev/null +++ b/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_5e-1/epoch_10.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:e91e8c5f0c0992c46a101a5e0b76aec8fece57c519a1e110e72f0000db3e8b72 +size 783392794 diff --git a/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_5e-1/epoch_2.pth b/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_5e-1/epoch_2.pth new file mode 100644 index 0000000000000000000000000000000000000000..9e27984a297cf6bdc07135928bf032940973f405 --- /dev/null +++ b/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_5e-1/epoch_2.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:b14c6c50ca9bc33369040de6b17f6feda5bf7fc6de276e921477c74135a264b1 +size 782254426 diff --git a/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_5e-1/epoch_3.pth b/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_5e-1/epoch_3.pth new file mode 100644 index 0000000000000000000000000000000000000000..c18e0e5bd5055284bd5ae37b2b6889f809380907 --- /dev/null +++ b/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_5e-1/epoch_3.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:dd073a7880cc0a9d7d60975ee3fd39b80b74ff3816341b656bbfb9b848319ea4 +size 782396954 diff --git a/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_5e-1/epoch_4.pth b/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_5e-1/epoch_4.pth new file mode 100644 index 0000000000000000000000000000000000000000..483f93f479d1e6ac28abdaafa107c9ca7da6c5e3 --- /dev/null +++ b/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_5e-1/epoch_4.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:f548ff4e6de5ffa2895717532fd0d42d79bd8a544c8103931e3cb7f826b85831 +size 782539354 diff --git a/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_5e-1/epoch_5.pth b/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_5e-1/epoch_5.pth new file mode 100644 index 0000000000000000000000000000000000000000..85ca401c9e8f76d15ca75caecef8e52bfda269da --- /dev/null +++ b/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_5e-1/epoch_5.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:476d4cfa7000cdaaff60b6f77734ffe014e67bd852112b8967ddbc0062185bfa +size 782681562 diff --git a/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_5e-1/epoch_6.pth b/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_5e-1/epoch_6.pth new file mode 100644 index 0000000000000000000000000000000000000000..9a8c2dd5831a89f422372803b797b1d1d91db3c6 --- /dev/null +++ b/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_5e-1/epoch_6.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:b90d423ef170cc9ae3cb11257605dd2fb10f2f54befed0f6d81364b160d54845 +size 782823386 diff --git a/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_5e-1/epoch_7.pth b/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_5e-1/epoch_7.pth new file mode 100644 index 0000000000000000000000000000000000000000..bfd7c33cde86ae143964892fd43c59428124d7ec --- /dev/null +++ b/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_5e-1/epoch_7.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:bfac35642815579314e43af885faa3132708f43a17c4e3dc7aa20e69a3db29b9 +size 782965274 diff --git a/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_5e-1/epoch_8.pth b/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_5e-1/epoch_8.pth new file mode 100644 index 0000000000000000000000000000000000000000..3a6d4d59dbb113fe4c5cdcc3a7d9801e965ff914 --- /dev/null +++ b/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_5e-1/epoch_8.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:b2bca36d25a893392b48c8d2802a9ae7cb1ca185860618fec2229598684ef2d8 +size 783106842 diff --git a/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_5e-1/epoch_9.pth b/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_5e-1/epoch_9.pth new file mode 100644 index 0000000000000000000000000000000000000000..cdf45dbfb0778de42041b55133c6fae7503b669f --- /dev/null +++ b/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_5e-1/epoch_9.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:1d79808abf15c64607d6db4b04c74900e4647733a9709733c094ddf49d628824 +size 783249370 diff --git a/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_5e-1/last_checkpoint b/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_5e-1/last_checkpoint new file mode 100644 index 0000000000000000000000000000000000000000..5d7d9e37d4ef6b25ef4cdafae952f0a20fd009cf --- /dev/null +++ b/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_5e-1/last_checkpoint @@ -0,0 +1 @@ +/mnt/petrelfs/luzeyu/workspace/fakebench/mmpretrain/workdir/convnext_small_4xb256_sdv1_1m_lr4e-4_aug_5e-1/epoch_10.pth \ No newline at end of file diff --git a/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_1e-1/20230603_172035/20230603_172035.log b/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_1e-1/20230603_172035/20230603_172035.log new file mode 100644 index 0000000000000000000000000000000000000000..eaad7fd24364dc1a03b30dbe317c8700a1cf2ee7 --- /dev/null +++ b/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_1e-1/20230603_172035/20230603_172035.log @@ -0,0 +1,1840 @@ +2023/06/03 17:20:40 - mmengine - INFO - +------------------------------------------------------------ +System environment: + sys.platform: linux + Python: 3.10.9 (main, Mar 8 2023, 10:47:38) [GCC 11.2.0] + CUDA available: True + numpy_random_seed: 2043574849 + GPU 0,1,2,3: NVIDIA A100-SXM4-80GB + CUDA_HOME: /mnt/petrelfs/share/cuda-11.6 + NVCC: Cuda compilation tools, release 11.6, V11.6.124 + GCC: gcc (GCC) 7.5.0 + PyTorch: 1.13.1 + PyTorch compiling details: PyTorch built with: + - GCC 9.3 + - C++ Version: 201402 + - Intel(R) oneAPI Math Kernel Library Version 2021.4-Product Build 20210904 for Intel(R) 64 architecture applications + - Intel(R) MKL-DNN v2.6.0 (Git Hash 52b5f107dd9cf10910aaa19cb47f3abf9b349815) + - OpenMP 201511 (a.k.a. OpenMP 4.5) + - LAPACK is enabled (usually provided by MKL) + - NNPACK is enabled + - CPU capability usage: AVX2 + - CUDA Runtime 11.6 + - NVCC architecture flags: -gencode;arch=compute_37,code=sm_37;-gencode;arch=compute_50,code=sm_50;-gencode;arch=compute_60,code=sm_60;-gencode;arch=compute_61,code=sm_61;-gencode;arch=compute_70,code=sm_70;-gencode;arch=compute_75,code=sm_75;-gencode;arch=compute_80,code=sm_80;-gencode;arch=compute_86,code=sm_86;-gencode;arch=compute_37,code=compute_37 + - CuDNN 8.3.2 (built against CUDA 11.5) + - Magma 2.6.1 + - Build settings: BLAS_INFO=mkl, BUILD_TYPE=Release, CUDA_VERSION=11.6, CUDNN_VERSION=8.3.2, CXX_COMPILER=/opt/rh/devtoolset-9/root/usr/bin/c++, CXX_FLAGS= -fabi-version=11 -Wno-deprecated -fvisibility-inlines-hidden -DUSE_PTHREADPOOL -fopenmp -DNDEBUG -DUSE_KINETO -DUSE_FBGEMM -DUSE_QNNPACK -DUSE_PYTORCH_QNNPACK -DUSE_XNNPACK -DSYMBOLICATE_MOBILE_DEBUG_HANDLE -DEDGE_PROFILER_USE_KINETO -O2 -fPIC -Wno-narrowing -Wall -Wextra -Werror=return-type -Werror=non-virtual-dtor -Wno-missing-field-initializers -Wno-type-limits -Wno-array-bounds -Wno-unknown-pragmas -Wunused-local-typedefs -Wno-unused-parameter -Wno-unused-function -Wno-unused-result -Wno-strict-overflow -Wno-strict-aliasing -Wno-error=deprecated-declarations -Wno-stringop-overflow -Wno-psabi -Wno-error=pedantic -Wno-error=redundant-decls -Wno-error=old-style-cast -fdiagnostics-color=always -faligned-new -Wno-unused-but-set-variable -Wno-maybe-uninitialized -fno-math-errno -fno-trapping-math -Werror=format -Werror=cast-function-type -Wno-stringop-overflow, LAPACK_INFO=mkl, PERF_WITH_AVX=1, PERF_WITH_AVX2=1, PERF_WITH_AVX512=1, TORCH_VERSION=1.13.1, USE_CUDA=ON, USE_CUDNN=ON, USE_EXCEPTION_PTR=1, USE_GFLAGS=OFF, USE_GLOG=OFF, USE_MKL=ON, USE_MKLDNN=ON, USE_MPI=OFF, USE_NCCL=ON, USE_NNPACK=ON, USE_OPENMP=ON, USE_ROCM=OFF, + + TorchVision: 0.14.1 + OpenCV: 4.7.0 + MMEngine: 0.7.3 + +Runtime environment: + cudnn_benchmark: True + mp_cfg: {'mp_start_method': 'fork', 'opencv_num_threads': 0} + dist_cfg: {'backend': 'nccl'} + seed: None + deterministic: False + Distributed launcher: slurm + Distributed training: True + GPU number: 4 +------------------------------------------------------------ + +2023/06/03 17:20:45 - mmengine - INFO - Config: +optim_wrapper = dict( + optimizer=dict( + type='AdamW', + lr=0.0004, + weight_decay=0.05, + eps=1e-08, + betas=(0.9, 0.999), + _scope_='mmpretrain'), + paramwise_cfg=dict( + norm_decay_mult=0.0, + bias_decay_mult=0.0, + flat_decay_mult=0.0, + custom_keys=dict({ + '.absolute_pos_embed': dict(decay_mult=0.0), + '.relative_position_bias_table': dict(decay_mult=0.0) + })), + type='AmpOptimWrapper', + dtype='bfloat16', + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=False, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=4096) +model = dict( + type='ImageClassifier', + backbone=dict( + frozen_stages=2, + type='ConvNeXt', + arch='small', + drop_path_rate=0.4, + init_cfg=dict( + type='Pretrained', + checkpoint= + 'https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth', + prefix='backbone')), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=768, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + init_cfg=None), + init_cfg=dict( + type='TruncNormal', layer=['Conv2d', 'Linear'], std=0.02, bias=0.0), + train_cfg=None) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stablediffusionV2-1-dpmsolver-25-1m.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/cc1m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV2-1-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV2-1-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_1e-1' + +2023/06/03 17:21:01 - mmengine - INFO - Hooks will be executed in the following order: +before_run: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_load_checkpoint: +(ABOVE_NORMAL) EMAHook + -------------------- +before_train: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_train_epoch: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(NORMAL ) DistSamplerSeedHook + -------------------- +before_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook + -------------------- +after_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +after_train_epoch: +(NORMAL ) IterTimerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_val_epoch: +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook + -------------------- +before_val_iter: +(NORMAL ) IterTimerHook + -------------------- +after_val_iter: +(NORMAL ) IterTimerHook +(NORMAL ) VisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_val_epoch: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_save_checkpoint: +(ABOVE_NORMAL) EMAHook + -------------------- +after_train: +(VERY_LOW ) CheckpointHook + -------------------- +before_test_epoch: +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook + -------------------- +before_test_iter: +(NORMAL ) IterTimerHook + -------------------- +after_test_iter: +(NORMAL ) IterTimerHook +(NORMAL ) VisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_test_epoch: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_run: +(BELOW_NORMAL) LoggerHook + -------------------- +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.2.0.weight:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.2.0.bias:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.2.1.bias:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.3.0.weight:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.3.0.bias:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.3.1.bias:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.0.gamma:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.0.depthwise_conv.bias:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.0.norm.weight:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.0.norm.bias:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.0.pointwise_conv1.bias:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.0.pointwise_conv2.bias:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.1.gamma:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.1.depthwise_conv.bias:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.1.norm.weight:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.1.norm.bias:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.1.pointwise_conv1.bias:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.1.pointwise_conv2.bias:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.2.gamma:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.2.depthwise_conv.bias:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.2.norm.weight:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.2.norm.bias:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.2.pointwise_conv1.bias:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.2.pointwise_conv2.bias:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.3.gamma:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.3.depthwise_conv.bias:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.3.norm.weight:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.3.norm.bias:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.3.pointwise_conv1.bias:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.3.pointwise_conv2.bias:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.4.gamma:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.4.depthwise_conv.bias:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.4.norm.weight:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.4.norm.bias:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.4.pointwise_conv1.bias:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.4.pointwise_conv2.bias:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.5.gamma:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.5.depthwise_conv.bias:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.5.norm.weight:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.5.norm.bias:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.5.pointwise_conv1.bias:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.5.pointwise_conv2.bias:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.6.gamma:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.6.depthwise_conv.bias:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.6.norm.weight:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.6.norm.bias:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.6.pointwise_conv1.bias:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.6.pointwise_conv2.bias:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.7.gamma:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.7.depthwise_conv.bias:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.7.norm.weight:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.7.norm.bias:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.7.pointwise_conv1.bias:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.7.pointwise_conv2.bias:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.8.gamma:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.8.depthwise_conv.bias:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.8.norm.weight:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.8.norm.bias:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.8.pointwise_conv1.bias:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.8.pointwise_conv2.bias:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.9.gamma:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.9.depthwise_conv.bias:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.9.norm.weight:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.9.norm.bias:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.9.pointwise_conv1.bias:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.9.pointwise_conv2.bias:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.10.gamma:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.10.depthwise_conv.bias:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.10.norm.weight:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.10.norm.bias:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.10.pointwise_conv1.bias:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.10.pointwise_conv2.bias:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.11.gamma:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.11.depthwise_conv.bias:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.11.norm.weight:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.11.norm.bias:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.11.pointwise_conv1.bias:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.11.pointwise_conv2.bias:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.12.gamma:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.12.depthwise_conv.bias:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.12.norm.weight:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.12.norm.bias:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.12.pointwise_conv1.bias:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.12.pointwise_conv2.bias:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.13.gamma:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.13.depthwise_conv.bias:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.13.norm.weight:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.13.norm.bias:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.13.pointwise_conv1.bias:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.13.pointwise_conv2.bias:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.14.gamma:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.14.depthwise_conv.bias:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.14.norm.weight:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.14.norm.bias:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.14.pointwise_conv1.bias:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.14.pointwise_conv2.bias:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.15.gamma:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.15.depthwise_conv.bias:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.15.norm.weight:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.15.norm.bias:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.15.pointwise_conv1.bias:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.15.pointwise_conv2.bias:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.16.gamma:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.16.depthwise_conv.bias:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.16.norm.weight:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.16.norm.bias:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.16.pointwise_conv1.bias:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.16.pointwise_conv2.bias:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.17.gamma:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.17.depthwise_conv.bias:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.17.norm.weight:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.17.norm.bias:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.17.pointwise_conv1.bias:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.17.pointwise_conv2.bias:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.18.gamma:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.18.depthwise_conv.bias:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.18.norm.weight:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.18.norm.bias:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.18.pointwise_conv1.bias:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.18.pointwise_conv2.bias:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.19.gamma:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.19.depthwise_conv.bias:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.19.norm.weight:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.19.norm.bias:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.19.pointwise_conv1.bias:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.19.pointwise_conv2.bias:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.20.gamma:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.20.depthwise_conv.bias:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.20.norm.weight:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.20.norm.bias:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.20.pointwise_conv1.bias:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.20.pointwise_conv2.bias:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.21.gamma:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.21.depthwise_conv.bias:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.21.norm.weight:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.21.norm.bias:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.21.pointwise_conv1.bias:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.21.pointwise_conv2.bias:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.22.gamma:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.22.depthwise_conv.bias:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.22.norm.weight:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.22.norm.bias:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.22.pointwise_conv1.bias:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.22.pointwise_conv2.bias:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.23.gamma:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.23.depthwise_conv.bias:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.23.norm.weight:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.23.norm.bias:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.23.pointwise_conv1.bias:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.23.pointwise_conv2.bias:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.24.gamma:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.24.depthwise_conv.bias:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.24.norm.weight:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.24.norm.bias:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.24.pointwise_conv1.bias:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.24.pointwise_conv2.bias:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.25.gamma:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.25.depthwise_conv.bias:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.25.norm.weight:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.25.norm.bias:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.25.pointwise_conv1.bias:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.25.pointwise_conv2.bias:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.26.gamma:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.26.depthwise_conv.bias:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.26.norm.weight:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.26.norm.bias:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.26.pointwise_conv1.bias:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.26.pointwise_conv2.bias:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.3.0.gamma:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.3.0.depthwise_conv.bias:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.3.0.norm.weight:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.3.0.norm.bias:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.3.0.pointwise_conv1.bias:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.3.0.pointwise_conv2.bias:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.3.1.gamma:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.3.1.depthwise_conv.bias:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.3.1.norm.weight:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.3.1.norm.bias:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.3.1.pointwise_conv1.bias:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.3.1.pointwise_conv2.bias:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.3.2.gamma:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.3.2.depthwise_conv.bias:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.3.2.norm.weight:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.3.2.norm.bias:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.3.2.pointwise_conv1.bias:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.stages.3.2.pointwise_conv2.bias:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.norm3.weight:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- backbone.norm3.bias:weight_decay=0.0 +2023/06/03 17:21:21 - mmengine - INFO - paramwise_options -- head.fc.bias:weight_decay=0.0 +2023/06/03 17:21:26 - mmengine - INFO - load backbone in model from: https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth +Name of parameter - Initialization information + +backbone.downsample_layers.0.0.weight - torch.Size([96, 3, 4, 4]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.0.0.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.0.1.weight - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.0.1.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.1.0.weight - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.1.0.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.1.1.weight - torch.Size([192, 96, 2, 2]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.1.1.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.2.0.weight - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.2.0.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.2.1.weight - torch.Size([384, 192, 2, 2]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.2.1.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.3.0.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.3.0.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.3.1.weight - torch.Size([768, 384, 2, 2]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.3.1.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.0.gamma - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.0.depthwise_conv.weight - torch.Size([96, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.0.depthwise_conv.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.0.norm.weight - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.0.norm.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.0.pointwise_conv1.weight - torch.Size([384, 96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.0.pointwise_conv1.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.0.pointwise_conv2.weight - torch.Size([96, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.0.pointwise_conv2.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.1.gamma - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.1.depthwise_conv.weight - torch.Size([96, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.1.depthwise_conv.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.1.norm.weight - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.1.norm.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.1.pointwise_conv1.weight - torch.Size([384, 96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.1.pointwise_conv1.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.1.pointwise_conv2.weight - torch.Size([96, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.1.pointwise_conv2.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.2.gamma - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.2.depthwise_conv.weight - torch.Size([96, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.2.depthwise_conv.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.2.norm.weight - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.2.norm.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.2.pointwise_conv1.weight - torch.Size([384, 96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.2.pointwise_conv1.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.2.pointwise_conv2.weight - torch.Size([96, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.2.pointwise_conv2.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.0.gamma - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.0.depthwise_conv.weight - torch.Size([192, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.0.depthwise_conv.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.0.norm.weight - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.0.norm.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.0.pointwise_conv1.weight - torch.Size([768, 192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.0.pointwise_conv1.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.0.pointwise_conv2.weight - torch.Size([192, 768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.0.pointwise_conv2.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.1.gamma - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.1.depthwise_conv.weight - torch.Size([192, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.1.depthwise_conv.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.1.norm.weight - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.1.norm.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.1.pointwise_conv1.weight - torch.Size([768, 192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.1.pointwise_conv1.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.1.pointwise_conv2.weight - torch.Size([192, 768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.1.pointwise_conv2.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.2.gamma - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.2.depthwise_conv.weight - torch.Size([192, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.2.depthwise_conv.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.2.norm.weight - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.2.norm.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.2.pointwise_conv1.weight - torch.Size([768, 192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.2.pointwise_conv1.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.2.pointwise_conv2.weight - torch.Size([192, 768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.2.pointwise_conv2.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.0.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.0.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.0.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.0.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.0.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.0.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.0.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.0.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.0.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.1.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.1.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.1.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.1.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.1.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.1.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.1.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.1.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.1.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.2.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.2.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.2.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.2.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.2.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.2.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.2.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.2.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.2.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.3.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.3.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.3.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.3.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.3.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.3.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.3.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.3.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.3.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.4.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.4.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.4.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.4.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.4.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.4.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.4.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.4.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.4.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.5.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.5.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.5.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.5.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.5.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.5.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.5.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.5.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.5.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.6.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.6.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.6.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.6.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.6.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.6.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.6.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.6.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.6.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.7.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.7.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.7.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.7.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.7.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.7.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.7.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.7.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.7.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.8.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.8.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.8.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.8.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.8.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.8.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.8.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.8.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.8.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.9.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.9.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.9.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.9.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.9.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.9.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.9.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.9.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.9.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.10.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.10.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.10.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.10.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.10.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.10.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.10.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.10.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.10.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.11.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.11.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.11.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.11.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.11.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.11.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.11.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.11.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.11.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.12.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.12.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.12.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.12.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.12.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.12.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.12.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.12.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.12.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.13.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.13.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.13.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.13.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.13.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.13.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.13.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.13.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.13.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.14.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.14.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.14.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.14.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.14.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.14.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.14.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.14.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.14.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.15.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.15.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.15.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.15.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.15.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.15.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.15.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.15.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.15.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.16.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.16.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.16.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.16.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.16.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.16.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.16.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.16.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.16.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.17.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.17.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.17.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.17.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.17.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.17.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.17.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.17.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.17.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.18.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.18.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.18.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.18.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.18.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.18.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.18.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.18.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.18.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.19.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.19.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.19.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.19.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.19.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.19.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.19.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.19.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.19.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.20.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.20.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.20.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.20.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.20.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.20.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.20.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.20.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.20.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.21.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.21.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.21.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.21.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.21.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.21.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.21.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.21.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.21.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.22.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.22.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.22.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.22.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.22.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.22.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.22.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.22.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.22.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.23.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.23.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.23.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.23.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.23.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.23.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.23.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.23.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.23.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.24.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.24.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.24.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.24.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.24.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.24.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.24.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.24.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.24.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.25.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.25.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.25.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.25.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.25.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.25.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.25.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.25.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.25.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.26.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.26.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.26.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.26.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.26.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.26.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.26.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.26.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.26.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.0.gamma - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.0.depthwise_conv.weight - torch.Size([768, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.0.depthwise_conv.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.0.norm.weight - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.0.norm.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.0.pointwise_conv1.weight - torch.Size([3072, 768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.0.pointwise_conv1.bias - torch.Size([3072]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.0.pointwise_conv2.weight - torch.Size([768, 3072]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.0.pointwise_conv2.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.1.gamma - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.1.depthwise_conv.weight - torch.Size([768, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.1.depthwise_conv.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.1.norm.weight - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.1.norm.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.1.pointwise_conv1.weight - torch.Size([3072, 768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.1.pointwise_conv1.bias - torch.Size([3072]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.1.pointwise_conv2.weight - torch.Size([768, 3072]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.1.pointwise_conv2.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.2.gamma - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.2.depthwise_conv.weight - torch.Size([768, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.2.depthwise_conv.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.2.norm.weight - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.2.norm.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.2.pointwise_conv1.weight - torch.Size([3072, 768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.2.pointwise_conv1.bias - torch.Size([3072]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.2.pointwise_conv2.weight - torch.Size([768, 3072]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.2.pointwise_conv2.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.norm3.weight - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.norm3.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +head.fc.weight - torch.Size([2, 768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +head.fc.bias - torch.Size([2]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 +2023/06/03 17:21:27 - mmengine - WARNING - "FileClient" will be deprecated in future. Please use io functions in https://mmengine.readthedocs.io/en/latest/api/fileio.html#file-io +2023/06/03 17:21:27 - mmengine - WARNING - "HardDiskBackend" is the alias of "LocalBackend" and the former will be deprecated in future. +2023/06/03 17:21:27 - mmengine - INFO - Checkpoints will be saved to /mnt/petrelfs/luzeyu/workspace/fakebench/mmpretrain/workdir/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_1e-1. +2023/06/03 17:25:15 - mmengine - INFO - Epoch(train) [1][ 100/1954] lr: 3.9998e-04 eta: 12:17:40 time: 2.2405 data_time: 0.8876 memory: 15587 loss: 0.0282 +2023/06/03 17:28:43 - mmengine - INFO - Epoch(train) [1][ 200/1954] lr: 3.9990e-04 eta: 11:42:31 time: 2.0559 data_time: 0.0119 memory: 15587 loss: 0.0199 +2023/06/03 17:32:03 - mmengine - INFO - Epoch(train) [1][ 300/1954] lr: 3.9977e-04 eta: 11:19:57 time: 2.2705 data_time: 0.0018 memory: 15587 loss: 0.0261 +2023/06/03 17:35:38 - mmengine - INFO - Epoch(train) [1][ 400/1954] lr: 3.9960e-04 eta: 11:18:27 time: 2.2109 data_time: 0.0026 memory: 15587 loss: 0.0129 +2023/06/03 17:39:05 - mmengine - INFO - Epoch(train) [1][ 500/1954] lr: 3.9937e-04 eta: 11:11:50 time: 2.1815 data_time: 0.0021 memory: 15587 loss: 0.0187 +2023/06/03 17:42:23 - mmengine - INFO - Epoch(train) [1][ 600/1954] lr: 3.9910e-04 eta: 11:00:41 time: 2.0332 data_time: 0.0020 memory: 15587 loss: 0.0112 +2023/06/03 17:45:54 - mmengine - INFO - Epoch(train) [1][ 700/1954] lr: 3.9877e-04 eta: 10:58:20 time: 2.2654 data_time: 0.0018 memory: 15587 loss: 0.0108 +2023/06/03 17:49:27 - mmengine - INFO - Epoch(train) [1][ 800/1954] lr: 3.9839e-04 eta: 10:56:08 time: 2.1147 data_time: 0.0016 memory: 15587 loss: 0.0154 +2023/06/03 17:52:52 - mmengine - INFO - Epoch(train) [1][ 900/1954] lr: 3.9797e-04 eta: 10:50:46 time: 1.3882 data_time: 0.0020 memory: 15587 loss: 0.0066 +2023/06/03 17:56:04 - mmengine - INFO - Exp name: convnext_small_4xb256_sdv2_1m_lr4e-4_aug_1e-1_20230603_172035 +2023/06/03 17:56:04 - mmengine - INFO - Epoch(train) [1][1000/1954] lr: 3.9749e-04 eta: 10:41:49 time: 2.1194 data_time: 0.0025 memory: 15587 loss: 0.0147 +2023/06/03 17:59:36 - mmengine - INFO - Epoch(train) [1][1100/1954] lr: 3.9696e-04 eta: 10:39:28 time: 2.2453 data_time: 0.0017 memory: 15587 loss: 0.0107 +2023/06/03 18:03:00 - mmengine - INFO - Epoch(train) [1][1200/1954] lr: 3.9639e-04 eta: 10:35:02 time: 2.0185 data_time: 0.0018 memory: 15587 loss: 0.0089 +2023/06/03 18:05:28 - mmengine - INFO - Epoch(train) [1][1300/1954] lr: 3.9576e-04 eta: 10:17:33 time: 1.4339 data_time: 0.0016 memory: 15587 loss: 0.0085 +2023/06/03 18:07:54 - mmengine - INFO - Epoch(train) [1][1400/1954] lr: 3.9509e-04 eta: 10:01:51 time: 1.4645 data_time: 0.0015 memory: 15587 loss: 0.0111 +2023/06/03 18:10:25 - mmengine - INFO - Epoch(train) [1][1500/1954] lr: 3.9436e-04 eta: 9:48:50 time: 1.5326 data_time: 0.0478 memory: 15587 loss: 0.0086 +2023/06/03 18:13:43 - mmengine - INFO - Epoch(train) [1][1600/1954] lr: 3.9359e-04 eta: 9:45:59 time: 0.7563 data_time: 0.0016 memory: 15587 loss: 0.0057 +2023/06/03 18:15:00 - mmengine - INFO - Epoch(train) [1][1700/1954] lr: 3.9277e-04 eta: 9:21:55 time: 0.8251 data_time: 0.0029 memory: 15587 loss: 0.0123 +2023/06/03 18:16:17 - mmengine - INFO - Epoch(train) [1][1800/1954] lr: 3.9190e-04 eta: 9:00:22 time: 0.7426 data_time: 0.0016 memory: 15587 loss: 0.0085 +2023/06/03 18:17:36 - mmengine - INFO - Epoch(train) [1][1900/1954] lr: 3.9098e-04 eta: 8:41:17 time: 0.7422 data_time: 0.0017 memory: 15587 loss: 0.0129 +2023/06/03 18:18:19 - mmengine - INFO - Exp name: convnext_small_4xb256_sdv2_1m_lr4e-4_aug_1e-1_20230603_172035 +2023/06/03 18:18:19 - mmengine - INFO - Saving checkpoint at 1 epochs +2023/06/03 18:19:34 - mmengine - INFO - Epoch(val) [1][29/29] accuracy/top1: 74.6808 single-label/precision_classwise: [98.57078552246094, 64.11041259765625] single-label/recall_classwise: [54.85764694213867, 99.02326202392578] single-label/f1-score_classwise: [70.4870834350586, 77.83092498779297] data_time: 1.4562 time: 1.9215 +2023/06/03 18:20:10 - mmengine - INFO - Exp name: convnext_small_4xb256_sdv2_1m_lr4e-4_aug_1e-1_20230603_172035 +2023/06/03 18:20:47 - mmengine - INFO - Epoch(train) [2][ 100/1954] lr: 3.8947e-04 eta: 8:14:26 time: 0.6186 data_time: 0.0021 memory: 15587 loss: 0.0067 +2023/06/03 18:21:53 - mmengine - INFO - Epoch(train) [2][ 200/1954] lr: 3.8843e-04 eta: 7:57:40 time: 0.6807 data_time: 0.0018 memory: 15585 loss: 0.0100 +2023/06/03 18:23:03 - mmengine - INFO - Epoch(train) [2][ 300/1954] lr: 3.8735e-04 eta: 7:42:52 time: 0.8138 data_time: 0.0025 memory: 15585 loss: 0.0101 +2023/06/03 18:24:08 - mmengine - INFO - Epoch(train) [2][ 400/1954] lr: 3.8621e-04 eta: 7:28:29 time: 0.6611 data_time: 0.0021 memory: 15585 loss: 0.0113 +2023/06/03 18:25:14 - mmengine - INFO - Epoch(train) [2][ 500/1954] lr: 3.8503e-04 eta: 7:15:22 time: 0.6108 data_time: 0.0018 memory: 15585 loss: 0.0077 +2023/06/03 18:26:20 - mmengine - INFO - Epoch(train) [2][ 600/1954] lr: 3.8380e-04 eta: 7:03:14 time: 0.7922 data_time: 0.0028 memory: 15585 loss: 0.0067 +2023/06/03 18:27:24 - mmengine - INFO - Epoch(train) [2][ 700/1954] lr: 3.8253e-04 eta: 6:51:40 time: 0.5612 data_time: 0.0027 memory: 15585 loss: 0.0120 +2023/06/03 18:28:30 - mmengine - INFO - Epoch(train) [2][ 800/1954] lr: 3.8121e-04 eta: 6:41:06 time: 0.6864 data_time: 0.0018 memory: 15585 loss: 0.0046 +2023/06/03 18:29:37 - mmengine - INFO - Epoch(train) [2][ 900/1954] lr: 3.7984e-04 eta: 6:31:16 time: 0.6648 data_time: 0.0020 memory: 15585 loss: 0.0057 +2023/06/03 18:30:42 - mmengine - INFO - Epoch(train) [2][1000/1954] lr: 3.7843e-04 eta: 6:21:50 time: 0.5865 data_time: 0.0017 memory: 15585 loss: 0.0097 +2023/06/03 18:31:12 - mmengine - INFO - Exp name: convnext_small_4xb256_sdv2_1m_lr4e-4_aug_1e-1_20230603_172035 +2023/06/03 18:31:46 - mmengine - INFO - Epoch(train) [2][1100/1954] lr: 3.7698e-04 eta: 6:12:50 time: 0.6202 data_time: 0.0016 memory: 15585 loss: 0.0105 +2023/06/03 18:32:52 - mmengine - INFO - Epoch(train) [2][1200/1954] lr: 3.7548e-04 eta: 6:04:31 time: 0.7500 data_time: 0.0033 memory: 15585 loss: 0.0086 +2023/06/03 18:33:59 - mmengine - INFO - Epoch(train) [2][1300/1954] lr: 3.7393e-04 eta: 5:56:47 time: 0.6538 data_time: 0.0017 memory: 15585 loss: 0.0074 +2023/06/03 18:35:05 - mmengine - INFO - Epoch(train) [2][1400/1954] lr: 3.7234e-04 eta: 5:49:18 time: 0.6083 data_time: 0.0019 memory: 15585 loss: 0.0139 +2023/06/03 18:36:10 - mmengine - INFO - Epoch(train) [2][1500/1954] lr: 3.7071e-04 eta: 5:42:10 time: 0.6342 data_time: 0.0019 memory: 15585 loss: 0.0051 +2023/06/03 18:37:17 - mmengine - INFO - Epoch(train) [2][1600/1954] lr: 3.6904e-04 eta: 5:35:29 time: 0.7065 data_time: 0.0024 memory: 15585 loss: 0.0075 +2023/06/03 18:38:24 - mmengine - INFO - Epoch(train) [2][1700/1954] lr: 3.6732e-04 eta: 5:29:07 time: 0.6910 data_time: 0.0015 memory: 15585 loss: 0.0028 +2023/06/03 18:39:31 - mmengine - INFO - Epoch(train) [2][1800/1954] lr: 3.6557e-04 eta: 5:23:02 time: 0.6923 data_time: 0.0016 memory: 15585 loss: 0.0046 +2023/06/03 18:40:37 - mmengine - INFO - Epoch(train) [2][1900/1954] lr: 3.6377e-04 eta: 5:17:08 time: 0.6549 data_time: 0.0031 memory: 15585 loss: 0.0069 +2023/06/03 18:41:13 - mmengine - INFO - Exp name: convnext_small_4xb256_sdv2_1m_lr4e-4_aug_1e-1_20230603_172035 +2023/06/03 18:41:13 - mmengine - INFO - Saving checkpoint at 2 epochs +2023/06/03 18:41:54 - mmengine - INFO - Epoch(val) [2][29/29] accuracy/top1: 96.3850 single-label/precision_classwise: [99.00025939941406, 93.4883041381836] single-label/recall_classwise: [94.394287109375, 98.82946014404297] single-label/f1-score_classwise: [96.64241790771484, 96.08470916748047] data_time: 0.4057 time: 0.7952 +2023/06/03 18:43:01 - mmengine - INFO - Exp name: convnext_small_4xb256_sdv2_1m_lr4e-4_aug_1e-1_20230603_172035 +2023/06/03 18:43:06 - mmengine - INFO - Epoch(train) [3][ 100/1954] lr: 3.6091e-04 eta: 5:08:54 time: 0.7163 data_time: 0.0020 memory: 15585 loss: 0.0066 +2023/06/03 18:44:09 - mmengine - INFO - Epoch(train) [3][ 200/1954] lr: 3.5901e-04 eta: 5:03:24 time: 0.6727 data_time: 0.0020 memory: 15585 loss: 0.0117 +2023/06/03 18:45:14 - mmengine - INFO - Epoch(train) [3][ 300/1954] lr: 3.5707e-04 eta: 4:58:14 time: 0.6574 data_time: 0.0021 memory: 15585 loss: 0.0073 +2023/06/03 18:46:21 - mmengine - INFO - Epoch(train) [3][ 400/1954] lr: 3.5509e-04 eta: 4:53:19 time: 0.6189 data_time: 0.0017 memory: 15585 loss: 0.0127 +2023/06/03 18:47:26 - mmengine - INFO - Epoch(train) [3][ 500/1954] lr: 3.5307e-04 eta: 4:48:30 time: 0.7311 data_time: 0.0020 memory: 15585 loss: 0.0057 +2023/06/03 18:48:30 - mmengine - INFO - Epoch(train) [3][ 600/1954] lr: 3.5101e-04 eta: 4:43:48 time: 0.6581 data_time: 0.0021 memory: 15585 loss: 0.0084 +2023/06/03 18:49:34 - mmengine - INFO - Epoch(train) [3][ 700/1954] lr: 3.4891e-04 eta: 4:39:15 time: 0.6895 data_time: 0.0021 memory: 15585 loss: 0.0063 +2023/06/03 18:50:39 - mmengine - INFO - Epoch(train) [3][ 800/1954] lr: 3.4677e-04 eta: 4:34:56 time: 0.5919 data_time: 0.0017 memory: 15585 loss: 0.0043 +2023/06/03 18:51:46 - mmengine - INFO - Epoch(train) [3][ 900/1954] lr: 3.4460e-04 eta: 4:30:48 time: 0.5827 data_time: 0.0019 memory: 15585 loss: 0.0085 +2023/06/03 18:52:50 - mmengine - INFO - Epoch(train) [3][1000/1954] lr: 3.4240e-04 eta: 4:26:39 time: 0.6836 data_time: 0.0024 memory: 15585 loss: 0.0058 +2023/06/03 18:53:53 - mmengine - INFO - Exp name: convnext_small_4xb256_sdv2_1m_lr4e-4_aug_1e-1_20230603_172035 +2023/06/03 18:53:58 - mmengine - INFO - Epoch(train) [3][1100/1954] lr: 3.4015e-04 eta: 4:22:50 time: 0.6142 data_time: 0.0019 memory: 15585 loss: 0.0044 +2023/06/03 18:55:01 - mmengine - INFO - Epoch(train) [3][1200/1954] lr: 3.3788e-04 eta: 4:18:52 time: 0.6562 data_time: 0.0030 memory: 15585 loss: 0.0099 +2023/06/03 18:56:07 - mmengine - INFO - Epoch(train) [3][1300/1954] lr: 3.3557e-04 eta: 4:15:12 time: 0.6658 data_time: 0.0022 memory: 15585 loss: 0.0028 +2023/06/03 18:57:11 - mmengine - INFO - Epoch(train) [3][1400/1954] lr: 3.3322e-04 eta: 4:11:30 time: 0.7218 data_time: 0.0017 memory: 15585 loss: 0.0048 +2023/06/03 18:58:15 - mmengine - INFO - Epoch(train) [3][1500/1954] lr: 3.3084e-04 eta: 4:07:54 time: 0.5950 data_time: 0.0020 memory: 15585 loss: 0.0063 +2023/06/03 18:59:21 - mmengine - INFO - Epoch(train) [3][1600/1954] lr: 3.2843e-04 eta: 4:04:29 time: 0.6757 data_time: 0.0022 memory: 15585 loss: 0.0076 +2023/06/03 19:00:27 - mmengine - INFO - Epoch(train) [3][1700/1954] lr: 3.2599e-04 eta: 4:01:08 time: 0.7135 data_time: 0.0029 memory: 15585 loss: 0.0048 +2023/06/03 19:01:34 - mmengine - INFO - Epoch(train) [3][1800/1954] lr: 3.2351e-04 eta: 3:57:54 time: 0.7037 data_time: 0.0018 memory: 15585 loss: 0.0055 +2023/06/03 19:02:43 - mmengine - INFO - Epoch(train) [3][1900/1954] lr: 3.2101e-04 eta: 3:54:50 time: 0.6197 data_time: 0.0024 memory: 15585 loss: 0.0050 +2023/06/03 19:03:19 - mmengine - INFO - Exp name: convnext_small_4xb256_sdv2_1m_lr4e-4_aug_1e-1_20230603_172035 +2023/06/03 19:03:19 - mmengine - INFO - Saving checkpoint at 3 epochs +2023/06/03 19:04:00 - mmengine - INFO - Epoch(val) [3][29/29] accuracy/top1: 99.2589 single-label/precision_classwise: [99.3370361328125, 99.1629867553711] single-label/recall_classwise: [99.31822204589844, 99.18604278564453] single-label/f1-score_classwise: [99.32762908935547, 99.17450714111328] data_time: 0.4040 time: 0.7969 +2023/06/03 19:05:12 - mmengine - INFO - Epoch(train) [4][ 100/1954] lr: 3.1709e-04 eta: 3:50:20 time: 0.6292 data_time: 0.0020 memory: 15585 loss: 0.0079 +2023/06/03 19:05:39 - mmengine - INFO - Exp name: convnext_small_4xb256_sdv2_1m_lr4e-4_aug_1e-1_20230603_172035 +2023/06/03 19:06:19 - mmengine - INFO - Epoch(train) [4][ 200/1954] lr: 3.1451e-04 eta: 3:47:22 time: 0.6463 data_time: 0.0024 memory: 15585 loss: 0.0070 +2023/06/03 19:07:25 - mmengine - INFO - Epoch(train) [4][ 300/1954] lr: 3.1190e-04 eta: 3:44:24 time: 0.7094 data_time: 0.0022 memory: 15585 loss: 0.0102 +2023/06/03 19:08:33 - mmengine - INFO - Epoch(train) [4][ 400/1954] lr: 3.0927e-04 eta: 3:41:34 time: 0.7614 data_time: 0.0024 memory: 15585 loss: 0.0035 +2023/06/03 19:09:40 - mmengine - INFO - Epoch(train) [4][ 500/1954] lr: 3.0661e-04 eta: 3:38:46 time: 0.6546 data_time: 0.0022 memory: 15585 loss: 0.0059 +2023/06/03 19:10:45 - mmengine - INFO - Epoch(train) [4][ 600/1954] lr: 3.0392e-04 eta: 3:35:55 time: 0.6833 data_time: 0.0019 memory: 15585 loss: 0.0054 +2023/06/03 19:11:50 - mmengine - INFO - Epoch(train) [4][ 700/1954] lr: 3.0120e-04 eta: 3:33:09 time: 0.7282 data_time: 0.0025 memory: 15585 loss: 0.0025 +2023/06/03 19:12:56 - mmengine - INFO - Epoch(train) [4][ 800/1954] lr: 2.9846e-04 eta: 3:30:27 time: 0.6195 data_time: 0.0021 memory: 15585 loss: 0.0051 +2023/06/03 19:14:01 - mmengine - INFO - Epoch(train) [4][ 900/1954] lr: 2.9570e-04 eta: 3:27:47 time: 0.6074 data_time: 0.0019 memory: 15585 loss: 0.0043 +2023/06/03 19:15:08 - mmengine - INFO - Epoch(train) [4][1000/1954] lr: 2.9291e-04 eta: 3:25:12 time: 0.6531 data_time: 0.0027 memory: 15585 loss: 0.0041 +2023/06/03 19:16:17 - mmengine - INFO - Epoch(train) [4][1100/1954] lr: 2.9010e-04 eta: 3:22:45 time: 0.6163 data_time: 0.0018 memory: 15585 loss: 0.0087 +2023/06/03 19:16:42 - mmengine - INFO - Exp name: convnext_small_4xb256_sdv2_1m_lr4e-4_aug_1e-1_20230603_172035 +2023/06/03 19:17:22 - mmengine - INFO - Epoch(train) [4][1200/1954] lr: 2.8727e-04 eta: 3:20:11 time: 0.6028 data_time: 0.0020 memory: 15585 loss: 0.0017 +2023/06/03 19:18:24 - mmengine - INFO - Epoch(train) [4][1300/1954] lr: 2.8442e-04 eta: 3:17:36 time: 0.5562 data_time: 0.0018 memory: 15585 loss: 0.0053 +2023/06/03 19:19:27 - mmengine - INFO - Epoch(train) [4][1400/1954] lr: 2.8155e-04 eta: 3:15:04 time: 0.5894 data_time: 0.0017 memory: 15585 loss: 0.0058 +2023/06/03 19:20:30 - mmengine - INFO - Epoch(train) [4][1500/1954] lr: 2.7865e-04 eta: 3:12:36 time: 0.6400 data_time: 0.0023 memory: 15585 loss: 0.0017 +2023/06/03 19:21:34 - mmengine - INFO - Epoch(train) [4][1600/1954] lr: 2.7574e-04 eta: 3:10:12 time: 0.6399 data_time: 0.0016 memory: 15585 loss: 0.0044 +2023/06/03 19:22:39 - mmengine - INFO - Epoch(train) [4][1700/1954] lr: 2.7281e-04 eta: 3:07:50 time: 0.6062 data_time: 0.0019 memory: 15585 loss: 0.0053 +2023/06/03 19:23:42 - mmengine - INFO - Epoch(train) [4][1800/1954] lr: 2.6986e-04 eta: 3:05:28 time: 0.6009 data_time: 0.0020 memory: 15585 loss: 0.0060 +2023/06/03 19:24:45 - mmengine - INFO - Epoch(train) [4][1900/1954] lr: 2.6690e-04 eta: 3:03:08 time: 0.6716 data_time: 0.0024 memory: 15585 loss: 0.0010 +2023/06/03 19:25:18 - mmengine - INFO - Exp name: convnext_small_4xb256_sdv2_1m_lr4e-4_aug_1e-1_20230603_172035 +2023/06/03 19:25:18 - mmengine - INFO - Saving checkpoint at 4 epochs +2023/06/03 19:25:55 - mmengine - INFO - Epoch(val) [4][29/29] accuracy/top1: 99.6416 single-label/precision_classwise: [99.46568298339844, 99.8597412109375] single-label/recall_classwise: [99.88636779785156, 99.34107971191406] single-label/f1-score_classwise: [99.67558288574219, 99.59973907470703] data_time: 0.2835 time: 0.6706 +2023/06/03 19:27:10 - mmengine - INFO - Epoch(train) [5][ 100/1954] lr: 2.6230e-04 eta: 2:59:51 time: 0.7761 data_time: 0.0018 memory: 15585 loss: 0.0034 +2023/06/03 19:28:04 - mmengine - INFO - Exp name: convnext_small_4xb256_sdv2_1m_lr4e-4_aug_1e-1_20230603_172035 +2023/06/03 19:28:15 - mmengine - INFO - Epoch(train) [5][ 200/1954] lr: 2.5929e-04 eta: 2:57:38 time: 0.7114 data_time: 0.0022 memory: 15585 loss: 0.0032 +2023/06/03 19:29:19 - mmengine - INFO - Epoch(train) [5][ 300/1954] lr: 2.5628e-04 eta: 2:55:26 time: 0.6504 data_time: 0.0018 memory: 15585 loss: 0.0058 +2023/06/03 19:30:24 - mmengine - INFO - Epoch(train) [5][ 400/1954] lr: 2.5325e-04 eta: 2:53:17 time: 0.6094 data_time: 0.0017 memory: 15585 loss: 0.0038 +2023/06/03 19:31:30 - mmengine - INFO - Epoch(train) [5][ 500/1954] lr: 2.5020e-04 eta: 2:51:09 time: 0.5812 data_time: 0.0031 memory: 15585 loss: 0.0018 +2023/06/03 19:32:36 - mmengine - INFO - Epoch(train) [5][ 600/1954] lr: 2.4715e-04 eta: 2:49:04 time: 0.7137 data_time: 0.0031 memory: 15585 loss: 0.0084 +2023/06/03 19:33:41 - mmengine - INFO - Epoch(train) [5][ 700/1954] lr: 2.4408e-04 eta: 2:46:59 time: 0.6228 data_time: 0.0015 memory: 15585 loss: 0.0034 +2023/06/03 19:34:46 - mmengine - INFO - Epoch(train) [5][ 800/1954] lr: 2.4100e-04 eta: 2:44:56 time: 0.6564 data_time: 0.0018 memory: 15585 loss: 0.0027 +2023/06/03 19:35:52 - mmengine - INFO - Epoch(train) [5][ 900/1954] lr: 2.3792e-04 eta: 2:42:55 time: 0.6469 data_time: 0.0016 memory: 15585 loss: 0.0008 +2023/06/03 19:36:59 - mmengine - INFO - Epoch(train) [5][1000/1954] lr: 2.3482e-04 eta: 2:40:55 time: 0.6467 data_time: 0.0016 memory: 15585 loss: 0.0027 +2023/06/03 19:38:06 - mmengine - INFO - Epoch(train) [5][1100/1954] lr: 2.3172e-04 eta: 2:38:59 time: 0.8062 data_time: 0.0026 memory: 15585 loss: 0.0044 +2023/06/03 19:39:03 - mmengine - INFO - Exp name: convnext_small_4xb256_sdv2_1m_lr4e-4_aug_1e-1_20230603_172035 +2023/06/03 19:39:13 - mmengine - INFO - Epoch(train) [5][1200/1954] lr: 2.2861e-04 eta: 2:37:02 time: 0.6702 data_time: 0.0021 memory: 15585 loss: 0.0089 +2023/06/03 19:40:20 - mmengine - INFO - Epoch(train) [5][1300/1954] lr: 2.2550e-04 eta: 2:35:07 time: 0.6787 data_time: 0.0019 memory: 15585 loss: 0.0052 +2023/06/03 19:41:28 - mmengine - INFO - Epoch(train) [5][1400/1954] lr: 2.2238e-04 eta: 2:33:14 time: 0.6485 data_time: 0.0017 memory: 15585 loss: 0.0020 +2023/06/03 19:42:34 - mmengine - INFO - Epoch(train) [5][1500/1954] lr: 2.1925e-04 eta: 2:31:19 time: 0.7062 data_time: 0.0021 memory: 15585 loss: 0.0025 +2023/06/03 19:43:40 - mmengine - INFO - Epoch(train) [5][1600/1954] lr: 2.1612e-04 eta: 2:29:26 time: 0.6657 data_time: 0.0023 memory: 15585 loss: 0.0042 +2023/06/03 19:44:45 - mmengine - INFO - Epoch(train) [5][1700/1954] lr: 2.1299e-04 eta: 2:27:32 time: 0.7266 data_time: 0.0019 memory: 15585 loss: 0.0017 +2023/06/03 19:45:50 - mmengine - INFO - Epoch(train) [5][1800/1954] lr: 2.0986e-04 eta: 2:25:40 time: 0.6141 data_time: 0.0015 memory: 15585 loss: 0.0026 +2023/06/03 19:46:57 - mmengine - INFO - Epoch(train) [5][1900/1954] lr: 2.0672e-04 eta: 2:23:51 time: 0.6750 data_time: 0.0016 memory: 15585 loss: 0.0023 +2023/06/03 19:47:34 - mmengine - INFO - Exp name: convnext_small_4xb256_sdv2_1m_lr4e-4_aug_1e-1_20230603_172035 +2023/06/03 19:47:34 - mmengine - INFO - Saving checkpoint at 5 epochs +2023/06/03 19:48:13 - mmengine - INFO - Epoch(val) [5][29/29] accuracy/top1: 99.7634 single-label/precision_classwise: [99.6412124633789, 99.9144287109375] single-label/recall_classwise: [99.93055725097656, 99.55813598632812] single-label/f1-score_classwise: [99.78567504882812, 99.7359619140625] data_time: 0.2945 time: 0.6848 +2023/06/03 19:49:26 - mmengine - INFO - Epoch(train) [6][ 100/1954] lr: 2.0190e-04 eta: 2:21:11 time: 0.6092 data_time: 0.0018 memory: 15585 loss: 0.0035 +2023/06/03 19:50:33 - mmengine - INFO - Epoch(train) [6][ 200/1954] lr: 1.9876e-04 eta: 2:19:24 time: 0.7729 data_time: 0.0028 memory: 15585 loss: 0.0031 +2023/06/03 19:50:54 - mmengine - INFO - Exp name: convnext_small_4xb256_sdv2_1m_lr4e-4_aug_1e-1_20230603_172035 +2023/06/03 19:51:42 - mmengine - INFO - Epoch(train) [6][ 300/1954] lr: 1.9563e-04 eta: 2:17:39 time: 0.7919 data_time: 0.0032 memory: 15585 loss: 0.0019 +2023/06/03 19:52:48 - mmengine - INFO - Epoch(train) [6][ 400/1954] lr: 1.9250e-04 eta: 2:15:52 time: 0.7147 data_time: 0.0025 memory: 15585 loss: 0.0025 +2023/06/03 19:53:54 - mmengine - INFO - Epoch(train) [6][ 500/1954] lr: 1.8937e-04 eta: 2:14:06 time: 0.7631 data_time: 0.0031 memory: 15585 loss: 0.0070 +2023/06/03 19:55:00 - mmengine - INFO - Epoch(train) [6][ 600/1954] lr: 1.8625e-04 eta: 2:12:21 time: 0.9600 data_time: 0.0018 memory: 15585 loss: 0.0021 +2023/06/03 19:56:10 - mmengine - INFO - Epoch(train) [6][ 700/1954] lr: 1.8313e-04 eta: 2:10:40 time: 0.5959 data_time: 0.0019 memory: 15585 loss: 0.0017 +2023/06/03 19:57:18 - mmengine - INFO - Epoch(train) [6][ 800/1954] lr: 1.8002e-04 eta: 2:08:58 time: 0.8121 data_time: 0.0027 memory: 15585 loss: 0.0058 +2023/06/03 19:58:25 - mmengine - INFO - Epoch(train) [6][ 900/1954] lr: 1.7691e-04 eta: 2:07:15 time: 0.6570 data_time: 0.0018 memory: 15585 loss: 0.0051 +2023/06/03 19:59:33 - mmengine - INFO - Epoch(train) [6][1000/1954] lr: 1.7381e-04 eta: 2:05:35 time: 0.6585 data_time: 0.0018 memory: 15585 loss: 0.0054 +2023/06/03 20:00:41 - mmengine - INFO - Epoch(train) [6][1100/1954] lr: 1.7072e-04 eta: 2:03:55 time: 0.7018 data_time: 0.0027 memory: 15585 loss: 0.0046 +2023/06/03 20:01:46 - mmengine - INFO - Epoch(train) [6][1200/1954] lr: 1.6764e-04 eta: 2:02:13 time: 0.6413 data_time: 0.0021 memory: 15585 loss: 0.0049 +2023/06/03 20:02:06 - mmengine - INFO - Exp name: convnext_small_4xb256_sdv2_1m_lr4e-4_aug_1e-1_20230603_172035 +2023/06/03 20:02:54 - mmengine - INFO - Epoch(train) [6][1300/1954] lr: 1.6457e-04 eta: 2:00:33 time: 0.6252 data_time: 0.0018 memory: 15585 loss: 0.0037 +2023/06/03 20:03:57 - mmengine - INFO - Epoch(train) [6][1400/1954] lr: 1.6151e-04 eta: 1:58:51 time: 0.7114 data_time: 0.0026 memory: 15585 loss: 0.0011 +2023/06/03 20:05:05 - mmengine - INFO - Epoch(train) [6][1500/1954] lr: 1.5846e-04 eta: 1:57:13 time: 0.6332 data_time: 0.0017 memory: 15585 loss: 0.0026 +2023/06/03 20:06:09 - mmengine - INFO - Epoch(train) [6][1600/1954] lr: 1.5542e-04 eta: 1:55:33 time: 0.6279 data_time: 0.0022 memory: 15585 loss: 0.0009 +2023/06/03 20:07:18 - mmengine - INFO - Epoch(train) [6][1700/1954] lr: 1.5239e-04 eta: 1:53:57 time: 0.7953 data_time: 0.0031 memory: 15585 loss: 0.0012 +2023/06/03 20:08:22 - mmengine - INFO - Epoch(train) [6][1800/1954] lr: 1.4938e-04 eta: 1:52:18 time: 0.6602 data_time: 0.0020 memory: 15585 loss: 0.0057 +2023/06/03 20:09:30 - mmengine - INFO - Epoch(train) [6][1900/1954] lr: 1.4638e-04 eta: 1:50:42 time: 0.7659 data_time: 0.0018 memory: 15585 loss: 0.0032 +2023/06/03 20:10:04 - mmengine - INFO - Exp name: convnext_small_4xb256_sdv2_1m_lr4e-4_aug_1e-1_20230603_172035 +2023/06/03 20:10:04 - mmengine - INFO - Saving checkpoint at 6 epochs +2023/06/03 20:10:45 - mmengine - INFO - Epoch(val) [6][29/29] accuracy/top1: 99.8365 single-label/precision_classwise: [99.76683044433594, 99.92231750488281] single-label/recall_classwise: [99.93687438964844, 99.71317291259766] single-label/f1-score_classwise: [99.85177612304688, 99.81763458251953] data_time: 0.4006 time: 0.7876 +2023/06/03 20:11:59 - mmengine - INFO - Epoch(train) [7][ 100/1954] lr: 1.4180e-04 eta: 1:48:19 time: 0.6309 data_time: 0.0019 memory: 15585 loss: 0.0070 +2023/06/03 20:13:06 - mmengine - INFO - Epoch(train) [7][ 200/1954] lr: 1.3884e-04 eta: 1:46:43 time: 0.6967 data_time: 0.0015 memory: 15585 loss: 0.0042 +2023/06/03 20:13:59 - mmengine - INFO - Exp name: convnext_small_4xb256_sdv2_1m_lr4e-4_aug_1e-1_20230603_172035 +2023/06/03 20:14:16 - mmengine - INFO - Epoch(train) [7][ 300/1954] lr: 1.3590e-04 eta: 1:45:10 time: 0.8140 data_time: 0.0021 memory: 15585 loss: 0.0035 +2023/06/03 20:15:24 - mmengine - INFO - Epoch(train) [7][ 400/1954] lr: 1.3298e-04 eta: 1:43:36 time: 0.6611 data_time: 0.0017 memory: 15585 loss: 0.0010 +2023/06/03 20:16:34 - mmengine - INFO - Epoch(train) [7][ 500/1954] lr: 1.3007e-04 eta: 1:42:05 time: 0.8553 data_time: 0.0023 memory: 15585 loss: 0.0060 +2023/06/03 20:17:45 - mmengine - INFO - Epoch(train) [7][ 600/1954] lr: 1.2719e-04 eta: 1:40:33 time: 0.6490 data_time: 0.0017 memory: 15585 loss: 0.0026 +2023/06/03 20:18:50 - mmengine - INFO - Epoch(train) [7][ 700/1954] lr: 1.2432e-04 eta: 1:38:59 time: 0.6495 data_time: 0.0016 memory: 15585 loss: 0.0009 +2023/06/03 20:19:59 - mmengine - INFO - Epoch(train) [7][ 800/1954] lr: 1.2148e-04 eta: 1:37:28 time: 0.6392 data_time: 0.0015 memory: 15585 loss: 0.0026 +2023/06/03 20:21:05 - mmengine - INFO - Epoch(train) [7][ 900/1954] lr: 1.1866e-04 eta: 1:35:55 time: 0.6212 data_time: 0.0018 memory: 15585 loss: 0.0050 +2023/06/03 20:22:11 - mmengine - INFO - Epoch(train) [7][1000/1954] lr: 1.1586e-04 eta: 1:34:22 time: 0.6220 data_time: 0.0029 memory: 15585 loss: 0.0045 +2023/06/03 20:23:18 - mmengine - INFO - Epoch(train) [7][1100/1954] lr: 1.1308e-04 eta: 1:32:50 time: 0.6195 data_time: 0.0016 memory: 15585 loss: 0.0030 +2023/06/03 20:24:24 - mmengine - INFO - Epoch(train) [7][1200/1954] lr: 1.1033e-04 eta: 1:31:19 time: 0.6924 data_time: 0.0020 memory: 15585 loss: 0.0032 +2023/06/03 20:25:18 - mmengine - INFO - Exp name: convnext_small_4xb256_sdv2_1m_lr4e-4_aug_1e-1_20230603_172035 +2023/06/03 20:25:34 - mmengine - INFO - Epoch(train) [7][1300/1954] lr: 1.0760e-04 eta: 1:29:50 time: 0.7162 data_time: 0.0021 memory: 15585 loss: 0.0043 +2023/06/03 20:26:39 - mmengine - INFO - Epoch(train) [7][1400/1954] lr: 1.0490e-04 eta: 1:28:18 time: 0.6079 data_time: 0.0017 memory: 15585 loss: 0.0020 +2023/06/03 20:27:43 - mmengine - INFO - Epoch(train) [7][1500/1954] lr: 1.0222e-04 eta: 1:26:47 time: 0.6890 data_time: 0.0035 memory: 15585 loss: 0.0033 +2023/06/03 20:28:47 - mmengine - INFO - Epoch(train) [7][1600/1954] lr: 9.9568e-05 eta: 1:25:16 time: 0.5878 data_time: 0.0016 memory: 15585 loss: 0.0010 +2023/06/03 20:29:49 - mmengine - INFO - Epoch(train) [7][1700/1954] lr: 9.6944e-05 eta: 1:23:44 time: 0.6265 data_time: 0.0016 memory: 15585 loss: 0.0026 +2023/06/03 20:30:53 - mmengine - INFO - Epoch(train) [7][1800/1954] lr: 9.4348e-05 eta: 1:22:14 time: 0.6686 data_time: 0.0018 memory: 15585 loss: 0.0026 +2023/06/03 20:31:59 - mmengine - INFO - Epoch(train) [7][1900/1954] lr: 9.1781e-05 eta: 1:20:45 time: 0.7247 data_time: 0.0021 memory: 15585 loss: 0.0021 +2023/06/03 20:32:32 - mmengine - INFO - Exp name: convnext_small_4xb256_sdv2_1m_lr4e-4_aug_1e-1_20230603_172035 +2023/06/03 20:32:32 - mmengine - INFO - Saving checkpoint at 7 epochs +2023/06/03 20:33:11 - mmengine - INFO - Epoch(val) [7][29/29] accuracy/top1: 99.8921 single-label/precision_classwise: [99.83609008789062, 99.961181640625] single-label/recall_classwise: [99.96842956542969, 99.79844665527344] single-label/f1-score_classwise: [99.90221405029297, 99.87974548339844] data_time: 0.3015 time: 0.6877 +2023/06/03 20:34:23 - mmengine - INFO - Epoch(train) [8][ 100/1954] lr: 8.7885e-05 eta: 1:18:30 time: 0.6268 data_time: 0.0021 memory: 15585 loss: 0.0019 +2023/06/03 20:35:28 - mmengine - INFO - Epoch(train) [8][ 200/1954] lr: 8.5394e-05 eta: 1:17:02 time: 0.6613 data_time: 0.0024 memory: 15585 loss: 0.0037 +2023/06/03 20:36:33 - mmengine - INFO - Epoch(train) [8][ 300/1954] lr: 8.2933e-05 eta: 1:15:33 time: 0.6687 data_time: 0.0017 memory: 15585 loss: 0.0025 +2023/06/03 20:36:48 - mmengine - INFO - Exp name: convnext_small_4xb256_sdv2_1m_lr4e-4_aug_1e-1_20230603_172035 +2023/06/03 20:37:38 - mmengine - INFO - Epoch(train) [8][ 400/1954] lr: 8.0504e-05 eta: 1:14:05 time: 0.5937 data_time: 0.0019 memory: 15585 loss: 0.0018 +2023/06/03 20:38:42 - mmengine - INFO - Epoch(train) [8][ 500/1954] lr: 7.8107e-05 eta: 1:12:37 time: 0.5466 data_time: 0.0016 memory: 15585 loss: 0.0017 +2023/06/03 20:39:46 - mmengine - INFO - Epoch(train) [8][ 600/1954] lr: 7.5743e-05 eta: 1:11:10 time: 0.7312 data_time: 0.0028 memory: 15585 loss: 0.0018 +2023/06/03 20:40:54 - mmengine - INFO - Epoch(train) [8][ 700/1954] lr: 7.3413e-05 eta: 1:09:44 time: 0.6396 data_time: 0.0018 memory: 15585 loss: 0.0037 +2023/06/03 20:42:02 - mmengine - INFO - Epoch(train) [8][ 800/1954] lr: 7.1116e-05 eta: 1:08:18 time: 0.6824 data_time: 0.0030 memory: 15585 loss: 0.0013 +2023/06/03 20:43:14 - mmengine - INFO - Epoch(train) [8][ 900/1954] lr: 6.8854e-05 eta: 1:06:54 time: 0.7072 data_time: 0.0022 memory: 15585 loss: 0.0008 +2023/06/03 20:44:21 - mmengine - INFO - Epoch(train) [8][1000/1954] lr: 6.6627e-05 eta: 1:05:29 time: 0.6273 data_time: 0.0022 memory: 15585 loss: 0.0007 +2023/06/03 20:45:32 - mmengine - INFO - Epoch(train) [8][1100/1954] lr: 6.4436e-05 eta: 1:04:05 time: 0.8115 data_time: 0.0027 memory: 15585 loss: 0.0009 +2023/06/03 20:46:36 - mmengine - INFO - Epoch(train) [8][1200/1954] lr: 6.2282e-05 eta: 1:02:39 time: 0.6569 data_time: 0.0021 memory: 15585 loss: 0.0020 +2023/06/03 20:47:43 - mmengine - INFO - Epoch(train) [8][1300/1954] lr: 6.0164e-05 eta: 1:01:14 time: 0.6631 data_time: 0.0018 memory: 15585 loss: 0.0015 +2023/06/03 20:47:56 - mmengine - INFO - Exp name: convnext_small_4xb256_sdv2_1m_lr4e-4_aug_1e-1_20230603_172035 +2023/06/03 20:48:50 - mmengine - INFO - Epoch(train) [8][1400/1954] lr: 5.8083e-05 eta: 0:59:49 time: 0.7107 data_time: 0.0024 memory: 15585 loss: 0.0046 +2023/06/03 20:49:56 - mmengine - INFO - Epoch(train) [8][1500/1954] lr: 5.6041e-05 eta: 0:58:25 time: 0.6852 data_time: 0.0021 memory: 15585 loss: 0.0010 +2023/06/03 20:51:00 - mmengine - INFO - Epoch(train) [8][1600/1954] lr: 5.4037e-05 eta: 0:57:00 time: 0.7780 data_time: 0.0016 memory: 15585 loss: 0.0015 +2023/06/03 20:52:06 - mmengine - INFO - Epoch(train) [8][1700/1954] lr: 5.2072e-05 eta: 0:55:36 time: 0.7209 data_time: 0.0032 memory: 15585 loss: 0.0008 +2023/06/03 20:53:13 - mmengine - INFO - Epoch(train) [8][1800/1954] lr: 5.0147e-05 eta: 0:54:12 time: 0.5953 data_time: 0.0016 memory: 15585 loss: 0.0024 +2023/06/03 20:54:19 - mmengine - INFO - Epoch(train) [8][1900/1954] lr: 4.8261e-05 eta: 0:52:48 time: 0.6689 data_time: 0.0021 memory: 15585 loss: 0.0010 +2023/06/03 20:54:54 - mmengine - INFO - Exp name: convnext_small_4xb256_sdv2_1m_lr4e-4_aug_1e-1_20230603_172035 +2023/06/03 20:54:54 - mmengine - INFO - Saving checkpoint at 8 epochs +2023/06/03 20:55:35 - mmengine - INFO - Epoch(val) [8][29/29] accuracy/top1: 99.9130 single-label/precision_classwise: [99.88015747070312, 99.95344543457031] single-label/recall_classwise: [99.96212005615234, 99.85271453857422] single-label/f1-score_classwise: [99.9211196899414, 99.90306091308594] data_time: 0.3978 time: 0.7914 +2023/06/03 20:56:48 - mmengine - INFO - Epoch(train) [9][ 100/1954] lr: 4.5437e-05 eta: 0:50:42 time: 0.6685 data_time: 0.0031 memory: 15585 loss: 0.0029 +2023/06/03 20:57:50 - mmengine - INFO - Epoch(train) [9][ 200/1954] lr: 4.3656e-05 eta: 0:49:18 time: 0.6349 data_time: 0.0020 memory: 15585 loss: 0.0011 +2023/06/03 20:59:00 - mmengine - INFO - Epoch(train) [9][ 300/1954] lr: 4.1916e-05 eta: 0:47:56 time: 0.8406 data_time: 0.0027 memory: 15585 loss: 0.0008 +2023/06/03 20:59:43 - mmengine - INFO - Exp name: convnext_small_4xb256_sdv2_1m_lr4e-4_aug_1e-1_20230603_172035 +2023/06/03 21:00:03 - mmengine - INFO - Epoch(train) [9][ 400/1954] lr: 4.0218e-05 eta: 0:46:32 time: 0.5914 data_time: 0.0019 memory: 15585 loss: 0.0018 +2023/06/03 21:01:07 - mmengine - INFO - Epoch(train) [9][ 500/1954] lr: 3.8563e-05 eta: 0:45:09 time: 0.6191 data_time: 0.0016 memory: 15585 loss: 0.0018 +2023/06/03 21:02:14 - mmengine - INFO - Epoch(train) [9][ 600/1954] lr: 3.6951e-05 eta: 0:43:47 time: 0.6164 data_time: 0.0023 memory: 15585 loss: 0.0005 +2023/06/03 21:03:23 - mmengine - INFO - Epoch(train) [9][ 700/1954] lr: 3.5383e-05 eta: 0:42:26 time: 0.6978 data_time: 0.0019 memory: 15585 loss: 0.0013 +2023/06/03 21:04:27 - mmengine - INFO - Epoch(train) [9][ 800/1954] lr: 3.3858e-05 eta: 0:41:03 time: 0.6358 data_time: 0.0016 memory: 15585 loss: 0.0006 +2023/06/03 21:05:33 - mmengine - INFO - Epoch(train) [9][ 900/1954] lr: 3.2378e-05 eta: 0:39:42 time: 0.6661 data_time: 0.0026 memory: 15585 loss: 0.0020 +2023/06/03 21:06:38 - mmengine - INFO - Epoch(train) [9][1000/1954] lr: 3.0942e-05 eta: 0:38:20 time: 0.6813 data_time: 0.0023 memory: 15585 loss: 0.0053 +2023/06/03 21:07:45 - mmengine - INFO - Epoch(train) [9][1100/1954] lr: 2.9551e-05 eta: 0:36:59 time: 0.6249 data_time: 0.0017 memory: 15585 loss: 0.0013 +2023/06/03 21:08:49 - mmengine - INFO - Epoch(train) [9][1200/1954] lr: 2.8205e-05 eta: 0:35:37 time: 0.5966 data_time: 0.0018 memory: 15585 loss: 0.0015 +2023/06/03 21:09:54 - mmengine - INFO - Epoch(train) [9][1300/1954] lr: 2.6905e-05 eta: 0:34:16 time: 0.6019 data_time: 0.0016 memory: 15585 loss: 0.0025 +2023/06/03 21:10:41 - mmengine - INFO - Exp name: convnext_small_4xb256_sdv2_1m_lr4e-4_aug_1e-1_20230603_172035 +2023/06/03 21:11:01 - mmengine - INFO - Epoch(train) [9][1400/1954] lr: 2.5652e-05 eta: 0:32:56 time: 0.6006 data_time: 0.0023 memory: 15585 loss: 0.0025 +2023/06/03 21:12:10 - mmengine - INFO - Epoch(train) [9][1500/1954] lr: 2.4444e-05 eta: 0:31:36 time: 0.5341 data_time: 0.0135 memory: 15585 loss: 0.0017 +2023/06/03 21:13:18 - mmengine - INFO - Epoch(train) [9][1600/1954] lr: 2.3283e-05 eta: 0:30:15 time: 0.9574 data_time: 0.0019 memory: 15585 loss: 0.0010 +2023/06/03 21:14:20 - mmengine - INFO - Epoch(train) [9][1700/1954] lr: 2.2170e-05 eta: 0:28:55 time: 0.5962 data_time: 0.0018 memory: 15585 loss: 0.0008 +2023/06/03 21:15:22 - mmengine - INFO - Epoch(train) [9][1800/1954] lr: 2.1103e-05 eta: 0:27:34 time: 0.7766 data_time: 0.0034 memory: 15585 loss: 0.0040 +2023/06/03 21:16:22 - mmengine - INFO - Epoch(train) [9][1900/1954] lr: 2.0084e-05 eta: 0:26:13 time: 0.6828 data_time: 0.0031 memory: 15585 loss: 0.0003 +2023/06/03 21:16:53 - mmengine - INFO - Exp name: convnext_small_4xb256_sdv2_1m_lr4e-4_aug_1e-1_20230603_172035 +2023/06/03 21:16:53 - mmengine - INFO - Saving checkpoint at 9 epochs +2023/06/03 21:17:30 - mmengine - INFO - Epoch(val) [9][29/29] accuracy/top1: 99.8852 single-label/precision_classwise: [99.81721496582031, 99.96893310546875] single-label/recall_classwise: [99.9747543334961, 99.77519226074219] single-label/f1-score_classwise: [99.89591979980469, 99.8719711303711] data_time: 0.2383 time: 0.6254 +2023/06/03 21:18:36 - mmengine - INFO - Epoch(train) [10][ 100/1954] lr: 1.8608e-05 eta: 0:24:10 time: 0.7067 data_time: 0.0029 memory: 15585 loss: 0.0010 +2023/06/03 21:19:38 - mmengine - INFO - Epoch(train) [10][ 200/1954] lr: 1.7711e-05 eta: 0:22:50 time: 0.5895 data_time: 0.0023 memory: 15585 loss: 0.0024 +2023/06/03 21:20:43 - mmengine - INFO - Epoch(train) [10][ 300/1954] lr: 1.6863e-05 eta: 0:21:31 time: 0.6707 data_time: 0.0025 memory: 15585 loss: 0.0012 +2023/06/03 21:21:47 - mmengine - INFO - Epoch(train) [10][ 400/1954] lr: 1.6063e-05 eta: 0:20:12 time: 0.5967 data_time: 0.0020 memory: 15585 loss: 0.0015 +2023/06/03 21:21:56 - mmengine - INFO - Exp name: convnext_small_4xb256_sdv2_1m_lr4e-4_aug_1e-1_20230603_172035 +2023/06/03 21:22:49 - mmengine - INFO - Epoch(train) [10][ 500/1954] lr: 1.5311e-05 eta: 0:18:52 time: 0.5839 data_time: 0.0020 memory: 15585 loss: 0.0017 +2023/06/03 21:23:50 - mmengine - INFO - Epoch(train) [10][ 600/1954] lr: 1.4609e-05 eta: 0:17:33 time: 0.6500 data_time: 0.0027 memory: 15585 loss: 0.0006 +2023/06/03 21:24:54 - mmengine - INFO - Epoch(train) [10][ 700/1954] lr: 1.3956e-05 eta: 0:16:14 time: 0.7252 data_time: 0.0023 memory: 15585 loss: 0.0010 +2023/06/03 21:25:56 - mmengine - INFO - Epoch(train) [10][ 800/1954] lr: 1.3353e-05 eta: 0:14:56 time: 0.5662 data_time: 0.0019 memory: 15585 loss: 0.0028 +2023/06/03 21:26:57 - mmengine - INFO - Epoch(train) [10][ 900/1954] lr: 1.2798e-05 eta: 0:13:37 time: 0.6044 data_time: 0.0019 memory: 15585 loss: 0.0003 +2023/06/03 21:28:01 - mmengine - INFO - Epoch(train) [10][1000/1954] lr: 1.2294e-05 eta: 0:12:19 time: 0.6125 data_time: 0.0022 memory: 15585 loss: 0.0004 +2023/06/03 21:29:04 - mmengine - INFO - Epoch(train) [10][1100/1954] lr: 1.1840e-05 eta: 0:11:01 time: 0.6679 data_time: 0.0028 memory: 15585 loss: 0.0004 +2023/06/03 21:30:05 - mmengine - INFO - Epoch(train) [10][1200/1954] lr: 1.1435e-05 eta: 0:09:43 time: 0.6810 data_time: 0.0017 memory: 15585 loss: 0.0034 +2023/06/03 21:31:08 - mmengine - INFO - Epoch(train) [10][1300/1954] lr: 1.1080e-05 eta: 0:08:25 time: 0.6506 data_time: 0.0017 memory: 15585 loss: 0.0003 +2023/06/03 21:32:10 - mmengine - INFO - Epoch(train) [10][1400/1954] lr: 1.0776e-05 eta: 0:07:07 time: 0.5854 data_time: 0.0018 memory: 15585 loss: 0.0019 +2023/06/03 21:32:20 - mmengine - INFO - Exp name: convnext_small_4xb256_sdv2_1m_lr4e-4_aug_1e-1_20230603_172035 +2023/06/03 21:33:15 - mmengine - INFO - Epoch(train) [10][1500/1954] lr: 1.0522e-05 eta: 0:05:50 time: 0.6268 data_time: 0.0046 memory: 15585 loss: 0.0007 +2023/06/03 21:34:18 - mmengine - INFO - Epoch(train) [10][1600/1954] lr: 1.0318e-05 eta: 0:04:32 time: 0.6911 data_time: 0.0027 memory: 15585 loss: 0.0023 +2023/06/03 21:35:19 - mmengine - INFO - Epoch(train) [10][1700/1954] lr: 1.0164e-05 eta: 0:03:15 time: 0.5970 data_time: 0.0028 memory: 15585 loss: 0.0014 +2023/06/03 21:36:23 - mmengine - INFO - Epoch(train) [10][1800/1954] lr: 1.0061e-05 eta: 0:01:58 time: 0.6389 data_time: 0.0020 memory: 15585 loss: 0.0009 +2023/06/03 21:37:23 - mmengine - INFO - Epoch(train) [10][1900/1954] lr: 1.0008e-05 eta: 0:00:41 time: 0.6003 data_time: 0.0019 memory: 15585 loss: 0.0004 +2023/06/03 21:37:55 - mmengine - INFO - Exp name: convnext_small_4xb256_sdv2_1m_lr4e-4_aug_1e-1_20230603_172035 +2023/06/03 21:37:55 - mmengine - INFO - Saving checkpoint at 10 epochs +2023/06/03 21:38:31 - mmengine - INFO - Epoch(val) [10][29/29] accuracy/top1: 99.9339 single-label/precision_classwise: [99.91167449951172, 99.96121215820312] single-label/recall_classwise: [99.96842956542969, 99.89147186279297] single-label/f1-score_classwise: [99.94005584716797, 99.92632293701172] data_time: 0.1915 time: 0.5792 diff --git a/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_1e-1/20230603_172035/vis_data/20230603_172035.json b/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_1e-1/20230603_172035/vis_data/20230603_172035.json new file mode 100644 index 0000000000000000000000000000000000000000..35d8484da0cea731e1374b660d952bdcf003f989 --- /dev/null +++ b/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_1e-1/20230603_172035/vis_data/20230603_172035.json @@ -0,0 +1,200 @@ +{"lr": 0.00039997529889041205, "data_time": 0.8876386165618897, "loss": 0.028243570402264596, "time": 2.2404876470565798, "epoch": 1, "memory": 15587, "step": 100} +{"lr": 0.0003999002014233684, "data_time": 0.011860418319702148, "loss": 0.019886883813887833, "time": 2.0559006690979005, "epoch": 1, "memory": 15587, "step": 200} +{"lr": 0.0003997747244903853, "data_time": 0.0017872333526611328, "loss": 0.026142413867637514, "time": 2.270457720756531, "epoch": 1, "memory": 15587, "step": 300} +{"lr": 0.00039959890052580966, "data_time": 0.002608990669250488, "loss": 0.012897801818326116, "time": 2.2108946800231934, "epoch": 1, "memory": 15587, "step": 400} +{"lr": 0.0003993727749781197, "data_time": 0.002134227752685547, "loss": 0.01874726158566773, "time": 2.18149094581604, "epoch": 1, "memory": 15587, "step": 500} +{"lr": 0.000399096406298175, "data_time": 0.002037715911865234, "loss": 0.011248944280669092, "time": 2.033232235908508, "epoch": 1, "memory": 15587, "step": 600} +{"lr": 0.0003987698659241082, "data_time": 0.0017608880996704101, "loss": 0.010819037957116962, "time": 2.265449380874634, "epoch": 1, "memory": 15587, "step": 700} +{"lr": 0.0003983932382628598, "data_time": 0.001622152328491211, "loss": 0.015379989007487892, "time": 2.1147350072860718, "epoch": 1, "memory": 15587, "step": 800} +{"lr": 0.00039796662066835946, "data_time": 0.0020137786865234374, "loss": 0.006578811584040523, "time": 1.3881823301315308, "epoch": 1, "memory": 15587, "step": 900} +{"lr": 0.00039749012341636, "data_time": 0.002531886100769043, "loss": 0.014720092481002211, "time": 2.1194176197052004, "epoch": 1, "memory": 15587, "step": 1000} +{"lr": 0.00039696386967593436, "data_time": 0.0016518354415893555, "loss": 0.0107122294139117, "time": 2.2452640771865844, "epoch": 1, "memory": 15587, "step": 1100} +{"lr": 0.00039638799547763706, "data_time": 0.0018185138702392577, "loss": 0.008884851774200796, "time": 2.018517756462097, "epoch": 1, "memory": 15587, "step": 1200} +{"lr": 0.0003957626496783407, "data_time": 0.001597142219543457, "loss": 0.008482847176492215, "time": 1.4338923454284669, "epoch": 1, "memory": 15587, "step": 1300} +{"lr": 0.0003950879939227595, "data_time": 0.0014984846115112306, "loss": 0.011113102734088897, "time": 1.4645151376724244, "epoch": 1, "memory": 15587, "step": 1400} +{"lr": 0.0003943642026016663, "data_time": 0.04784624576568604, "loss": 0.008591280272230505, "time": 1.5325900077819825, "epoch": 1, "memory": 15587, "step": 1500} +{"lr": 0.0003935914628068133, "data_time": 0.0015847206115722657, "loss": 0.0057054765988141295, "time": 0.7563295125961303, "epoch": 1, "memory": 15587, "step": 1600} +{"lr": 0.00039276997428257205, "data_time": 0.0028943300247192385, "loss": 0.012270110519602894, "time": 0.8250779867172241, "epoch": 1, "memory": 15587, "step": 1700} +{"lr": 0.0003918999493743001, "data_time": 0.0016125917434692382, "loss": 0.008454771200194954, "time": 0.7425617933273315, "epoch": 1, "memory": 15587, "step": 1800} +{"lr": 0.00039098161297345475, "data_time": 0.001694631576538086, "loss": 0.012858521239832044, "time": 0.7422439813613891, "epoch": 1, "memory": 15587, "step": 1900} +{"accuracy/top1": 74.68077087402344, "data_time": 1.456217494504205, "time": 1.9215284462632805, "step": 1} +{"lr": 0.0003894734390071047, "data_time": 0.0020853281021118164, "loss": 0.006674895528703928, "time": 0.6185879945755005, "epoch": 2, "memory": 15587, "step": 2054} +{"lr": 0.0003884334872841784, "data_time": 0.0017771005630493164, "loss": 0.009980583051219582, "time": 0.6806638956069946, "epoch": 2, "memory": 15585, "step": 2154} +{"lr": 0.00038734612010935703, "data_time": 0.0025436878204345703, "loss": 0.010095262108370662, "time": 0.8138118028640747, "epoch": 2, "memory": 15585, "step": 2254} +{"lr": 0.00038621161855458035, "data_time": 0.0020783185958862306, "loss": 0.011340158013626932, "time": 0.6611358404159546, "epoch": 2, "memory": 15585, "step": 2354} +{"lr": 0.0003850302758754841, "data_time": 0.0018435478210449218, "loss": 0.007705595251172781, "time": 0.6108261346817017, "epoch": 2, "memory": 15585, "step": 2454} +{"lr": 0.00038380239743559646, "data_time": 0.0028277873992919923, "loss": 0.006682888045907021, "time": 0.7921959161758423, "epoch": 2, "memory": 15585, "step": 2554} +{"lr": 0.00038252830062740647, "data_time": 0.00267488956451416, "loss": 0.012029745010659098, "time": 0.5611567974090577, "epoch": 2, "memory": 15585, "step": 2654} +{"lr": 0.00038120831479032175, "data_time": 0.0017798185348510743, "loss": 0.004605051083490253, "time": 0.6864491939544678, "epoch": 2, "memory": 15585, "step": 2754} +{"lr": 0.0003798427811255361, "data_time": 0.001991939544677734, "loss": 0.005698884697631001, "time": 0.6648210287094116, "epoch": 2, "memory": 15585, "step": 2854} +{"lr": 0.0003784320526078357, "data_time": 0.001694631576538086, "loss": 0.00974665260873735, "time": 0.5864532709121704, "epoch": 2, "memory": 15585, "step": 2954} +{"lr": 0.0003769764938943556, "data_time": 0.00160369873046875, "loss": 0.010500320605933667, "time": 0.6202181816101074, "epoch": 2, "memory": 15585, "step": 3054} +{"lr": 0.0003754764812303242, "data_time": 0.003342127799987793, "loss": 0.00864911307580769, "time": 0.7500198602676391, "epoch": 2, "memory": 15585, "step": 3154} +{"lr": 0.0003739324023518048, "data_time": 0.0016830921173095702, "loss": 0.00743028549477458, "time": 0.6538217544555665, "epoch": 2, "memory": 15585, "step": 3254} +{"lr": 0.0003723446563854724, "data_time": 0.0019375801086425782, "loss": 0.013865859108045697, "time": 0.6082741975784302, "epoch": 2, "memory": 15585, "step": 3354} +{"lr": 0.0003707136537454424, "data_time": 0.0018967151641845702, "loss": 0.005103431269526481, "time": 0.6341658353805542, "epoch": 2, "memory": 15585, "step": 3454} +{"lr": 0.0003690398160271841, "data_time": 0.002397584915161133, "loss": 0.007460079155862331, "time": 0.7064965248107911, "epoch": 2, "memory": 15585, "step": 3554} +{"lr": 0.00036732357589854426, "data_time": 0.001503133773803711, "loss": 0.002773802587762475, "time": 0.6909519910812378, "epoch": 2, "memory": 15585, "step": 3654} +{"lr": 0.00036556537698790475, "data_time": 0.0016093492507934571, "loss": 0.004564368352293968, "time": 0.6923271179199219, "epoch": 2, "memory": 15585, "step": 3754} +{"lr": 0.0003637656737695118, "data_time": 0.003058767318725586, "loss": 0.006924367323517799, "time": 0.6548510074615479, "epoch": 2, "memory": 15585, "step": 3854} +{"accuracy/top1": 96.38495635986328, "data_time": 0.405672820409139, "time": 0.7952233950297037, "step": 2} +{"lr": 0.00036091403361175644, "data_time": 0.0019866466522216798, "loss": 0.006628051167353988, "time": 0.7163207054138183, "epoch": 3, "memory": 15585, "step": 4008} +{"lr": 0.00035901102535335625, "data_time": 0.0019848108291625976, "loss": 0.011745957238599659, "time": 0.6726689815521241, "epoch": 3, "memory": 15585, "step": 4108} +{"lr": 0.00035706820701190153, "data_time": 0.0020529985427856444, "loss": 0.007274257158860565, "time": 0.6573845863342285, "epoch": 3, "memory": 15585, "step": 4208} +{"lr": 0.00035508608078364215, "data_time": 0.0017499208450317382, "loss": 0.012712002824991941, "time": 0.6188857316970825, "epoch": 3, "memory": 15585, "step": 4308} +{"lr": 0.00035306515902546755, "data_time": 0.0019903659820556642, "loss": 0.005651786830276251, "time": 0.7310869455337524, "epoch": 3, "memory": 15585, "step": 4408} +{"lr": 0.0003510059641224655, "data_time": 0.0020628929138183593, "loss": 0.008439291268587112, "time": 0.6581233501434326, "epoch": 3, "memory": 15585, "step": 4508} +{"lr": 0.00034890902835289306, "data_time": 0.0020843505859375, "loss": 0.006331758527085185, "time": 0.6894623756408691, "epoch": 3, "memory": 15585, "step": 4608} +{"lr": 0.00034677489375058753, "data_time": 0.0017110586166381836, "loss": 0.004314387775957584, "time": 0.5919305324554444, "epoch": 3, "memory": 15585, "step": 4708} +{"lr": 0.00034460411196485875, "data_time": 0.0019282817840576172, "loss": 0.008494613505899907, "time": 0.5827285051345825, "epoch": 3, "memory": 15585, "step": 4808} +{"lr": 0.0003423972441178926, "data_time": 0.0023634672164916993, "loss": 0.005785406613722443, "time": 0.6835513114929199, "epoch": 3, "memory": 15585, "step": 4908} +{"lr": 0.0003401548606597075, "data_time": 0.001933741569519043, "loss": 0.0043906884267926214, "time": 0.6141704320907593, "epoch": 3, "memory": 15585, "step": 5008} +{"lr": 0.0003378775412207003, "data_time": 0.002954411506652832, "loss": 0.009896530024707317, "time": 0.6562421321868896, "epoch": 3, "memory": 15585, "step": 5108} +{"lr": 0.0003355658744618178, "data_time": 0.00216059684753418, "loss": 0.0027915054000914098, "time": 0.6658342838287353, "epoch": 3, "memory": 15585, "step": 5208} +{"lr": 0.0003332204579223944, "data_time": 0.0017322301864624023, "loss": 0.004801007127389312, "time": 0.7218419790267945, "epoch": 3, "memory": 15585, "step": 5308} +{"lr": 0.0003308418978656968, "data_time": 0.001978492736816406, "loss": 0.00633810474537313, "time": 0.595003867149353, "epoch": 3, "memory": 15585, "step": 5408} +{"lr": 0.0003284308091222099, "data_time": 0.002231335639953613, "loss": 0.007586751598864794, "time": 0.6756606817245483, "epoch": 3, "memory": 15585, "step": 5508} +{"lr": 0.0003259878149307108, "data_time": 0.002854156494140625, "loss": 0.004832551255822182, "time": 0.7135452508926392, "epoch": 3, "memory": 15585, "step": 5608} +{"lr": 0.0003235135467771688, "data_time": 0.0018404006958007812, "loss": 0.005467744776979089, "time": 0.7036562442779541, "epoch": 3, "memory": 15585, "step": 5708} +{"lr": 0.0003210086442315141, "data_time": 0.0024457216262817384, "loss": 0.004971952084451914, "time": 0.6197478532791137, "epoch": 3, "memory": 15585, "step": 5808} +{"accuracy/top1": 99.25890350341797, "data_time": 0.40395546754201256, "time": 0.7969277540842692, "step": 3} +{"lr": 0.00031709267628953966, "data_time": 0.002018284797668457, "loss": 0.007866184692829847, "time": 0.6292434215545655, "epoch": 4, "memory": 15585, "step": 5962} +{"lr": 0.00031451289120303685, "data_time": 0.002409625053405762, "loss": 0.006956744473427534, "time": 0.6462630987167358, "epoch": 4, "memory": 15585, "step": 6062} +{"lr": 0.000311904798290621, "data_time": 0.0021924972534179688, "loss": 0.010229665692895651, "time": 0.7094257116317749, "epoch": 4, "memory": 15585, "step": 6162} +{"lr": 0.00030926907171438685, "data_time": 0.0024345874786376952, "loss": 0.003526915470138192, "time": 0.7614258527755737, "epoch": 4, "memory": 15585, "step": 6262} +{"lr": 0.000306606392779414, "data_time": 0.0021973371505737303, "loss": 0.00594695471227169, "time": 0.6545991182327271, "epoch": 4, "memory": 15585, "step": 6362} +{"lr": 0.0003039174497576579, "data_time": 0.0018962860107421876, "loss": 0.005417559342458844, "time": 0.6833284139633179, "epoch": 4, "memory": 15585, "step": 6462} +{"lr": 0.0003012029377100387, "data_time": 0.0024603605270385742, "loss": 0.002504068287089467, "time": 0.7282400608062745, "epoch": 4, "memory": 15585, "step": 6562} +{"lr": 0.00029846355830677674, "data_time": 0.002100968360900879, "loss": 0.005125401075929403, "time": 0.6195366144180298, "epoch": 4, "memory": 15585, "step": 6662} +{"lr": 0.0002957000196460183, "data_time": 0.0019130229949951172, "loss": 0.004303906485438347, "time": 0.6073935031890869, "epoch": 4, "memory": 15585, "step": 6762} +{"lr": 0.0002929130360708003, "data_time": 0.002662348747253418, "loss": 0.004120248276740313, "time": 0.6531259536743164, "epoch": 4, "memory": 15585, "step": 6862} +{"lr": 0.0002901033279844016, "data_time": 0.0018069744110107422, "loss": 0.008747064601629972, "time": 0.6163111209869385, "epoch": 4, "memory": 15585, "step": 6962} +{"lr": 0.00028727162166412623, "data_time": 0.0020123958587646485, "loss": 0.001685183309018612, "time": 0.6028306007385253, "epoch": 4, "memory": 15585, "step": 7062} +{"lr": 0.000284418649073569, "data_time": 0.0018334388732910156, "loss": 0.0052682790905237194, "time": 0.5562018871307373, "epoch": 4, "memory": 15585, "step": 7162} +{"lr": 0.00028154514767341203, "data_time": 0.001651453971862793, "loss": 0.005815492710098624, "time": 0.5893539190292358, "epoch": 4, "memory": 15585, "step": 7262} +{"lr": 0.0002786518602307988, "data_time": 0.002273893356323242, "loss": 0.0016718230675905944, "time": 0.6399655103683471, "epoch": 4, "memory": 15585, "step": 7362} +{"lr": 0.0002757395346273384, "data_time": 0.0016340494155883789, "loss": 0.0044249090366065506, "time": 0.6398932933807373, "epoch": 4, "memory": 15585, "step": 7462} +{"lr": 0.0002728089236657853, "data_time": 0.0018706560134887696, "loss": 0.00525554371997714, "time": 0.6062353849411011, "epoch": 4, "memory": 15585, "step": 7562} +{"lr": 0.0002698607848754497, "data_time": 0.0019966840744018556, "loss": 0.00600892910733819, "time": 0.6009192943572998, "epoch": 4, "memory": 15585, "step": 7662} +{"lr": 0.0002668958803163847, "data_time": 0.0023994922637939455, "loss": 0.0010058487765491008, "time": 0.6716243028640747, "epoch": 4, "memory": 15585, "step": 7762} +{"accuracy/top1": 99.64163208007812, "data_time": 0.2835105021794637, "time": 0.6705821593602498, "step": 4} +{"lr": 0.00026229890681650394, "data_time": 0.0017737150192260742, "loss": 0.0033932299818843605, "time": 0.7760605573654175, "epoch": 5, "memory": 15585, "step": 7916} +{"lr": 0.00025929487224058475, "data_time": 0.002240920066833496, "loss": 0.0032494002487510444, "time": 0.7113546133041382, "epoch": 5, "memory": 15585, "step": 8016} +{"lr": 0.0002562768030631272, "data_time": 0.0018189668655395508, "loss": 0.005762854823842645, "time": 0.6504134893417358, "epoch": 5, "memory": 15585, "step": 8116} +{"lr": 0.0002532454794203871, "data_time": 0.0017086029052734374, "loss": 0.003760502813383937, "time": 0.6094439506530762, "epoch": 5, "memory": 15585, "step": 8216} +{"lr": 0.0002502016848747476, "data_time": 0.0030717372894287108, "loss": 0.0017563470639288426, "time": 0.5812451362609863, "epoch": 5, "memory": 15585, "step": 8316} +{"lr": 0.0002471462062121758, "data_time": 0.003067421913146973, "loss": 0.008362926309928298, "time": 0.7136571645736695, "epoch": 5, "memory": 15585, "step": 8416} +{"lr": 0.00024407983323884952, "data_time": 0.0015196561813354491, "loss": 0.003412273805588484, "time": 0.6228043079376221, "epoch": 5, "memory": 15585, "step": 8516} +{"lr": 0.0002410033585770006, "data_time": 0.001826167106628418, "loss": 0.0027474768459796904, "time": 0.656356930732727, "epoch": 5, "memory": 15585, "step": 8616} +{"lr": 0.00023791757746003292, "data_time": 0.0016218900680541992, "loss": 0.0007744169328361749, "time": 0.646912670135498, "epoch": 5, "memory": 15585, "step": 8716} +{"lr": 0.00023482328752696182, "data_time": 0.0016498088836669922, "loss": 0.002677060244604945, "time": 0.6467357397079467, "epoch": 5, "memory": 15585, "step": 8816} +{"lr": 0.00023172128861623345, "data_time": 0.002625274658203125, "loss": 0.004404339054599404, "time": 0.806192684173584, "epoch": 5, "memory": 15585, "step": 8916} +{"lr": 0.000228612382558976, "data_time": 0.0021098852157592773, "loss": 0.008860325813293457, "time": 0.670192551612854, "epoch": 5, "memory": 15585, "step": 9016} +{"lr": 0.00022549737297173692, "data_time": 0.0018606185913085938, "loss": 0.005228723445907235, "time": 0.67873113155365, "epoch": 5, "memory": 15585, "step": 9116} +{"lr": 0.00022237706504875477, "data_time": 0.001738286018371582, "loss": 0.0019822140224277974, "time": 0.648528790473938, "epoch": 5, "memory": 15585, "step": 9216} +{"lr": 0.00021925226535382818, "data_time": 0.0020878791809082033, "loss": 0.0024604568723589184, "time": 0.7061655282974243, "epoch": 5, "memory": 15585, "step": 9316} +{"lr": 0.0002161237816118266, "data_time": 0.002283334732055664, "loss": 0.004188699461519718, "time": 0.6657184839248658, "epoch": 5, "memory": 15585, "step": 9416} +{"lr": 0.00021299242249990383, "data_time": 0.0018518686294555664, "loss": 0.0017229572404175997, "time": 0.726628589630127, "epoch": 5, "memory": 15585, "step": 9516} +{"lr": 0.00020985899743846363, "data_time": 0.001510763168334961, "loss": 0.002604235848411918, "time": 0.6141276121139526, "epoch": 5, "memory": 15585, "step": 9616} +{"lr": 0.00020672431638193363, "data_time": 0.0016469955444335938, "loss": 0.0023076528683304786, "time": 0.6750414371490479, "epoch": 5, "memory": 15585, "step": 9716} +{"accuracy/top1": 99.7634048461914, "data_time": 0.29454609553019206, "time": 0.6848266045252482, "step": 5} +{"lr": 0.00020189632111931318, "data_time": 0.0018271207809448242, "loss": 0.003468033019453287, "time": 0.6092110872268677, "epoch": 6, "memory": 15585, "step": 9870} +{"lr": 0.00019876209288860824, "data_time": 0.002828526496887207, "loss": 0.0030729669611901044, "time": 0.7728718042373657, "epoch": 6, "memory": 15585, "step": 9970} +{"lr": 0.00019562947708532286, "data_time": 0.0031618118286132813, "loss": 0.001884378120303154, "time": 0.7918673515319824, "epoch": 6, "memory": 15585, "step": 10070} +{"lr": 0.00019249928345470104, "data_time": 0.002530360221862793, "loss": 0.0024694283027201893, "time": 0.7146633625030517, "epoch": 6, "memory": 15585, "step": 10170} +{"lr": 0.00018937232111588298, "data_time": 0.0030922651290893554, "loss": 0.007032108958810568, "time": 0.7631094932556153, "epoch": 6, "memory": 15585, "step": 10270} +{"lr": 0.000186249398352757, "data_time": 0.0018081903457641602, "loss": 0.002067164471372962, "time": 0.9599995136260986, "epoch": 6, "memory": 15585, "step": 10370} +{"lr": 0.0001831313224050277, "data_time": 0.0018959522247314453, "loss": 0.0016676651313900948, "time": 0.5958833456039428, "epoch": 6, "memory": 15585, "step": 10470} +{"lr": 0.00018001889925955291, "data_time": 0.0026535987854003906, "loss": 0.005793391447514296, "time": 0.8120676755905152, "epoch": 6, "memory": 15585, "step": 10570} +{"lr": 0.00017691293344200646, "data_time": 0.0017524480819702149, "loss": 0.005128879193216562, "time": 0.6570345640182496, "epoch": 6, "memory": 15585, "step": 10670} +{"lr": 0.00017381422780891689, "data_time": 0.0018006086349487305, "loss": 0.0054186578840017315, "time": 0.6585187196731568, "epoch": 6, "memory": 15585, "step": 10770} +{"lr": 0.00017072358334013806, "data_time": 0.002730393409729004, "loss": 0.004597321106120944, "time": 0.7017984628677368, "epoch": 6, "memory": 15585, "step": 10870} +{"lr": 0.00016764179893180575, "data_time": 0.0021402835845947266, "loss": 0.004850745666772127, "time": 0.6412996053695679, "epoch": 6, "memory": 15585, "step": 10970} +{"lr": 0.00016456967118983176, "data_time": 0.0017818689346313476, "loss": 0.003687735600396991, "time": 0.6251964330673218, "epoch": 6, "memory": 15585, "step": 11070} +{"lr": 0.00016150799422398984, "data_time": 0.002596926689147949, "loss": 0.00107732517644763, "time": 0.7114237546920776, "epoch": 6, "memory": 15585, "step": 11170} +{"lr": 0.0001584575594426489, "data_time": 0.0016971349716186524, "loss": 0.002558692218735814, "time": 0.6332179069519043, "epoch": 6, "memory": 15585, "step": 11270} +{"lr": 0.00015541915534820135, "data_time": 0.002237415313720703, "loss": 0.000920321699231863, "time": 0.6279433727264404, "epoch": 6, "memory": 15585, "step": 11370} +{"lr": 0.00015239356733324542, "data_time": 0.0030861139297485352, "loss": 0.001232651760801673, "time": 0.7953397035598755, "epoch": 6, "memory": 15585, "step": 11470} +{"lr": 0.00014938157747756997, "data_time": 0.0020141363143920898, "loss": 0.005741068674251437, "time": 0.6601707935333252, "epoch": 6, "memory": 15585, "step": 11570} +{"lr": 0.00014638396434599565, "data_time": 0.0017972707748413086, "loss": 0.003222867427393794, "time": 0.7659487247467041, "epoch": 6, "memory": 15585, "step": 11670} +{"accuracy/top1": 99.83647155761719, "data_time": 0.4005567630132039, "time": 0.787556258837382, "step": 6} +{"lr": 0.00014179754513556176, "data_time": 0.0019487857818603516, "loss": 0.007016163924708963, "time": 0.6308538436889648, "epoch": 7, "memory": 15585, "step": 11824} +{"lr": 0.00013883992334674444, "data_time": 0.0015415191650390626, "loss": 0.004243675014004111, "time": 0.6966867923736573, "epoch": 7, "memory": 15585, "step": 11924} +{"lr": 0.00013589940317867467, "data_time": 0.002080273628234863, "loss": 0.003451137337833643, "time": 0.8140087842941284, "epoch": 7, "memory": 15585, "step": 12024} +{"lr": 0.0001329767447220783, "data_time": 0.001691150665283203, "loss": 0.0009962485171854496, "time": 0.6610537767410278, "epoch": 7, "memory": 15585, "step": 12124} +{"lr": 0.00013007270345063488, "data_time": 0.0023084163665771486, "loss": 0.006037602573633194, "time": 0.8553447246551513, "epoch": 7, "memory": 15585, "step": 12224} +{"lr": 0.00012718803002569412, "data_time": 0.0016849517822265625, "loss": 0.0025534901302307843, "time": 0.6490463018417358, "epoch": 7, "memory": 15585, "step": 12324} +{"lr": 0.0001243234701022397, "data_time": 0.0015696287155151367, "loss": 0.00087009915150702, "time": 0.6494510889053344, "epoch": 7, "memory": 15585, "step": 12424} +{"lr": 0.00012147976413614597, "data_time": 0.0015357732772827148, "loss": 0.002598012378439307, "time": 0.6391769170761108, "epoch": 7, "memory": 15585, "step": 12524} +{"lr": 0.0001186576471927784, "data_time": 0.00181124210357666, "loss": 0.005000874120742083, "time": 0.6212236166000367, "epoch": 7, "memory": 15585, "step": 12624} +{"lr": 0.00011585784875698867, "data_time": 0.0029226303100585937, "loss": 0.004526560194790363, "time": 0.6219683170318604, "epoch": 7, "memory": 15585, "step": 12724} +{"lr": 0.00011308109254454993, "data_time": 0.0016169786453247071, "loss": 0.0030195570085197686, "time": 0.6195187330245971, "epoch": 7, "memory": 15585, "step": 12824} +{"lr": 0.00011032809631508466, "data_time": 0.0020107746124267576, "loss": 0.003165309550240636, "time": 0.6924211740493774, "epoch": 7, "memory": 15585, "step": 12924} +{"lr": 0.00010759957168653253, "data_time": 0.0021471261978149416, "loss": 0.004263857891783118, "time": 0.7161961793899536, "epoch": 7, "memory": 15585, "step": 13024} +{"lr": 0.0001048962239512054, "data_time": 0.0017017841339111329, "loss": 0.0020327826496213675, "time": 0.6079156398773193, "epoch": 7, "memory": 15585, "step": 13124} +{"lr": 0.00010221875189347644, "data_time": 0.0034569501876831055, "loss": 0.0033277326729148626, "time": 0.6889806032180786, "epoch": 7, "memory": 15585, "step": 13224} +{"lr": 9.95678476091537e-05, "data_time": 0.0016147613525390625, "loss": 0.0009845915716141461, "time": 0.5877719163894654, "epoch": 7, "memory": 15585, "step": 13324} +{"lr": 9.694419632657999e-05, "data_time": 0.0015826940536499024, "loss": 0.0025791836436837912, "time": 0.626505184173584, "epoch": 7, "memory": 15585, "step": 13424} +{"lr": 9.434847622951059e-05, "data_time": 0.0017975568771362305, "loss": 0.002552804397419095, "time": 0.6686267852783203, "epoch": 7, "memory": 15585, "step": 13524} +{"lr": 9.178135828180905e-05, "data_time": 0.002054429054260254, "loss": 0.002057436853647232, "time": 0.7246514320373535, "epoch": 7, "memory": 15585, "step": 13624} +{"accuracy/top1": 99.89214324951172, "data_time": 0.3015367825826009, "time": 0.6877463261286417, "step": 7} +{"lr": 8.788546579134055e-05, "data_time": 0.0020665407180786135, "loss": 0.0019316677935421467, "time": 0.6268484354019165, "epoch": 8, "memory": 15585, "step": 13778} +{"lr": 8.539396307206318e-05, "data_time": 0.0024080991744995115, "loss": 0.0037416914012283085, "time": 0.6612948894500732, "epoch": 8, "memory": 15585, "step": 13878} +{"lr": 8.293337714106816e-05, "data_time": 0.0017055273056030273, "loss": 0.0025136511772871017, "time": 0.6687227487564087, "epoch": 8, "memory": 15585, "step": 13978} +{"lr": 8.05043440315878e-05, "data_time": 0.0018981695175170898, "loss": 0.0017824933398514985, "time": 0.5936944007873535, "epoch": 8, "memory": 15585, "step": 14078} +{"lr": 7.810749162081216e-05, "data_time": 0.0015995264053344726, "loss": 0.0017375380732119084, "time": 0.5466150045394897, "epoch": 8, "memory": 15585, "step": 14178} +{"lr": 7.574343946758993e-05, "data_time": 0.0028189897537231447, "loss": 0.0018452064599841834, "time": 0.7312486648559571, "epoch": 8, "memory": 15585, "step": 14278} +{"lr": 7.34127986522793e-05, "data_time": 0.0018247842788696289, "loss": 0.0037062933202832936, "time": 0.6396161079406738, "epoch": 8, "memory": 15585, "step": 14378} +{"lr": 7.111617161879136e-05, "data_time": 0.0030232667922973633, "loss": 0.001308464864268899, "time": 0.6824496030807495, "epoch": 8, "memory": 15585, "step": 14478} +{"lr": 6.885415201886485e-05, "data_time": 0.0022236108779907227, "loss": 0.0008456234354525805, "time": 0.7071934700012207, "epoch": 8, "memory": 15585, "step": 14578} +{"lr": 6.662732455861337e-05, "data_time": 0.002220606803894043, "loss": 0.0007111166138201952, "time": 0.6272669553756713, "epoch": 8, "memory": 15585, "step": 14678} +{"lr": 6.443626484738661e-05, "data_time": 0.002686810493469238, "loss": 0.0008525562472641468, "time": 0.8114670276641845, "epoch": 8, "memory": 15585, "step": 14778} +{"lr": 6.22815392489815e-05, "data_time": 0.0021439313888549803, "loss": 0.002012418955564499, "time": 0.6568866491317749, "epoch": 8, "memory": 15585, "step": 14878} +{"lr": 6.016370473524328e-05, "data_time": 0.0018251895904541015, "loss": 0.0014770927838981152, "time": 0.6630625486373901, "epoch": 8, "memory": 15585, "step": 14978} +{"lr": 5.80833087420956e-05, "data_time": 0.0023761987686157227, "loss": 0.004584140982478857, "time": 0.7106636047363282, "epoch": 8, "memory": 15585, "step": 15078} +{"lr": 5.604088902803397e-05, "data_time": 0.002094554901123047, "loss": 0.000984257971867919, "time": 0.6852351188659668, "epoch": 8, "memory": 15585, "step": 15178} +{"lr": 5.403697353512148e-05, "data_time": 0.0016436576843261719, "loss": 0.0015397735871374608, "time": 0.7779737949371338, "epoch": 8, "memory": 15585, "step": 15278} +{"lr": 5.207208025252174e-05, "data_time": 0.0031928300857543947, "loss": 0.0008444434963166714, "time": 0.720936918258667, "epoch": 8, "memory": 15585, "step": 15378} +{"lr": 5.014671708260458e-05, "data_time": 0.0016473293304443359, "loss": 0.002385187242180109, "time": 0.5953363895416259, "epoch": 8, "memory": 15585, "step": 15478} +{"lr": 4.826138170965952e-05, "data_time": 0.0020528078079223634, "loss": 0.0010014158207923174, "time": 0.6689431428909302, "epoch": 8, "memory": 15585, "step": 15578} +{"accuracy/top1": 99.91301727294922, "data_time": 0.3978493531545003, "time": 0.7914119005203247, "step": 8} +{"lr": 4.5437373040402475e-05, "data_time": 0.0031154394149780274, "loss": 0.002939263405278325, "time": 0.6684917211532593, "epoch": 9, "memory": 15585, "step": 15732} +{"lr": 4.3655875036148156e-05, "data_time": 0.0019965887069702147, "loss": 0.0011175610590726138, "time": 0.6349424123764038, "epoch": 9, "memory": 15585, "step": 15832} +{"lr": 4.19160826370641e-05, "data_time": 0.0027050495147705076, "loss": 0.0008449246641248464, "time": 0.8405579805374146, "epoch": 9, "memory": 15585, "step": 15932} +{"lr": 4.021844555952455e-05, "data_time": 0.0019388198852539062, "loss": 0.0017927538603544236, "time": 0.5914006471633911, "epoch": 9, "memory": 15585, "step": 16032} +{"lr": 3.8563402623236864e-05, "data_time": 0.0016057014465332032, "loss": 0.0018114028964191676, "time": 0.6190768957138062, "epoch": 9, "memory": 15585, "step": 16132} +{"lr": 3.695138163781135e-05, "data_time": 0.002338695526123047, "loss": 0.0004661856684833765, "time": 0.616426658630371, "epoch": 9, "memory": 15585, "step": 16232} +{"lr": 3.538279929217756e-05, "data_time": 0.001933455467224121, "loss": 0.0013150629587471486, "time": 0.6978412628173828, "epoch": 9, "memory": 15585, "step": 16332} +{"lr": 3.385806104687531e-05, "data_time": 0.0015532732009887694, "loss": 0.0005993003025650979, "time": 0.6358365535736084, "epoch": 9, "memory": 15585, "step": 16432} +{"lr": 3.237756102924734e-05, "data_time": 0.0025865793228149413, "loss": 0.0020424792543053626, "time": 0.6661193609237671, "epoch": 9, "memory": 15585, "step": 16532} +{"lr": 3.094168193156213e-05, "data_time": 0.0022933483123779297, "loss": 0.0053255957085639235, "time": 0.6813425064086914, "epoch": 9, "memory": 15585, "step": 16632} +{"lr": 2.95507949120924e-05, "data_time": 0.0017201662063598632, "loss": 0.0013453535735607146, "time": 0.6248709440231324, "epoch": 9, "memory": 15585, "step": 16732} +{"lr": 2.8205259499174602e-05, "data_time": 0.001804494857788086, "loss": 0.0014839986339211463, "time": 0.596595048904419, "epoch": 9, "memory": 15585, "step": 16832} +{"lr": 2.690542349827549e-05, "data_time": 0.0016358375549316406, "loss": 0.002500947145745158, "time": 0.6019083499908447, "epoch": 9, "memory": 15585, "step": 16932} +{"lr": 2.565162290208795e-05, "data_time": 0.002254962921142578, "loss": 0.002461536694318056, "time": 0.6005879163742065, "epoch": 9, "memory": 15585, "step": 17032} +{"lr": 2.4444181803680975e-05, "data_time": 0.013503074645996094, "loss": 0.001660887710750103, "time": 0.5341065168380738, "epoch": 9, "memory": 15585, "step": 17132} +{"lr": 2.32834123127253e-05, "data_time": 0.0018811702728271484, "loss": 0.0010136137250810861, "time": 0.957408356666565, "epoch": 9, "memory": 15585, "step": 17232} +{"lr": 2.2169614474816763e-05, "data_time": 0.001784205436706543, "loss": 0.0008078602608293295, "time": 0.5961580276489258, "epoch": 9, "memory": 15585, "step": 17332} +{"lr": 2.110307619391749e-05, "data_time": 0.003448748588562012, "loss": 0.003968959068879485, "time": 0.7766430377960205, "epoch": 9, "memory": 15585, "step": 17432} +{"lr": 2.008407315793661e-05, "data_time": 0.0031492948532104493, "loss": 0.00026420974172651767, "time": 0.6827759742736816, "epoch": 9, "memory": 15585, "step": 17532} +{"accuracy/top1": 99.88518524169922, "data_time": 0.23833333651224772, "time": 0.6253702561060588, "step": 9} +{"lr": 1.8608381954463387e-05, "data_time": 0.002872872352600098, "loss": 0.000977274589240551, "time": 0.7066831350326538, "epoch": 10, "memory": 15585, "step": 17686} +{"lr": 1.771127410231484e-05, "data_time": 0.0022667884826660157, "loss": 0.0023614218458533286, "time": 0.5895065546035767, "epoch": 10, "memory": 15585, "step": 17786} +{"lr": 1.6862578237010542e-05, "data_time": 0.002544879913330078, "loss": 0.0011892936658114196, "time": 0.6707264423370362, "epoch": 10, "memory": 15585, "step": 17886} +{"lr": 1.6062513736694383e-05, "data_time": 0.0019954442977905273, "loss": 0.0014919942710548638, "time": 0.5967118501663208, "epoch": 10, "memory": 15585, "step": 17986} +{"lr": 1.531128740886036e-05, "data_time": 0.002044820785522461, "loss": 0.001691504241898656, "time": 0.5839269876480102, "epoch": 10, "memory": 15585, "step": 18086} +{"lr": 1.4609093436895328e-05, "data_time": 0.0026756048202514647, "loss": 0.00056116315536201, "time": 0.650002384185791, "epoch": 10, "memory": 15585, "step": 18186} +{"lr": 1.3956113329884566e-05, "data_time": 0.0022931337356567384, "loss": 0.0010490994900465012, "time": 0.7251847743988037, "epoch": 10, "memory": 15585, "step": 18286} +{"lr": 1.335251587569401e-05, "data_time": 0.0018516778945922852, "loss": 0.002795368479564786, "time": 0.566193962097168, "epoch": 10, "memory": 15585, "step": 18386} +{"lr": 1.2798457097340337e-05, "data_time": 0.0019077301025390626, "loss": 0.00033678459003567697, "time": 0.6044378995895385, "epoch": 10, "memory": 15585, "step": 18486} +{"lr": 1.229408021266094e-05, "data_time": 0.002157902717590332, "loss": 0.000383464340120554, "time": 0.6124505996704102, "epoch": 10, "memory": 15585, "step": 18586} +{"lr": 1.1839515597293723e-05, "data_time": 0.002776336669921875, "loss": 0.00039589018560945986, "time": 0.6678505659103393, "epoch": 10, "memory": 15585, "step": 18686} +{"lr": 1.1434880750976406e-05, "data_time": 0.001705479621887207, "loss": 0.0033804717008024455, "time": 0.6809921503067017, "epoch": 10, "memory": 15585, "step": 18786} +{"lr": 1.108028026717431e-05, "data_time": 0.0017388582229614258, "loss": 0.0003105533309280872, "time": 0.650597357749939, "epoch": 10, "memory": 15585, "step": 18886} +{"lr": 1.0775805806044138e-05, "data_time": 0.001784682273864746, "loss": 0.0019477454479783773, "time": 0.5854311943054199, "epoch": 10, "memory": 15585, "step": 18986} +{"lr": 1.0521536070740776e-05, "data_time": 0.004630064964294434, "loss": 0.0007190603762865066, "time": 0.6267573833465576, "epoch": 10, "memory": 15585, "step": 19086} +{"lr": 1.03175367870735e-05, "data_time": 0.0026628255844116213, "loss": 0.002255278080701828, "time": 0.6910584926605224, "epoch": 10, "memory": 15585, "step": 19186} +{"lr": 1.0163860686516628e-05, "data_time": 0.002797985076904297, "loss": 0.0014115124009549617, "time": 0.5970310688018798, "epoch": 10, "memory": 15585, "step": 19286} +{"lr": 1.0060547492578997e-05, "data_time": 0.0020235061645507814, "loss": 0.0008689676877111196, "time": 0.638939642906189, "epoch": 10, "memory": 15585, "step": 19386} +{"lr": 1.0007623910535892e-05, "data_time": 0.0019460678100585937, "loss": 0.0004255445208400488, "time": 0.6003096103668213, "epoch": 10, "memory": 15585, "step": 19486} +{"accuracy/top1": 99.93389129638672, "data_time": 0.19146766662597656, "time": 0.5791983922322591, "step": 10} diff --git a/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_1e-1/20230603_172035/vis_data/config.py b/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_1e-1/20230603_172035/vis_data/config.py new file mode 100644 index 0000000000000000000000000000000000000000..470c9006a677f36d1d5911b9299657969c1a2ae4 --- /dev/null +++ b/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_1e-1/20230603_172035/vis_data/config.py @@ -0,0 +1,247 @@ +optim_wrapper = dict( + optimizer=dict( + type='AdamW', + lr=0.0004, + weight_decay=0.05, + eps=1e-08, + betas=(0.9, 0.999), + _scope_='mmpretrain'), + paramwise_cfg=dict( + norm_decay_mult=0.0, + bias_decay_mult=0.0, + flat_decay_mult=0.0, + custom_keys=dict({ + '.absolute_pos_embed': dict(decay_mult=0.0), + '.relative_position_bias_table': dict(decay_mult=0.0) + })), + type='AmpOptimWrapper', + dtype='bfloat16', + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=False, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=4096) +model = dict( + type='ImageClassifier', + backbone=dict( + frozen_stages=2, + type='ConvNeXt', + arch='small', + drop_path_rate=0.4, + init_cfg=dict( + type='Pretrained', + checkpoint= + 'https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth', + prefix='backbone')), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=768, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + init_cfg=None), + init_cfg=dict( + type='TruncNormal', layer=['Conv2d', 'Linear'], std=0.02, bias=0.0), + train_cfg=None) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stablediffusionV2-1-dpmsolver-25-1m.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/cc1m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV2-1-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV2-1-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_1e-1' diff --git a/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_1e-1/20230603_172035/vis_data/events.out.tfevents.1685784054.SH-IDC1-10-140-24-133.66739.0 b/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_1e-1/20230603_172035/vis_data/events.out.tfevents.1685784054.SH-IDC1-10-140-24-133.66739.0 new file mode 100644 index 0000000000000000000000000000000000000000..1ec4720d610de369403785878632a55a96f7db7a --- /dev/null +++ b/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_1e-1/20230603_172035/vis_data/events.out.tfevents.1685784054.SH-IDC1-10-140-24-133.66739.0 @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:29a4737e04ef79b7678a6d9e06c09fbe175ff8776f3a396223ca1664756fc21c +size 1612015 diff --git a/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_1e-1/20230603_172035/vis_data/scalars.json b/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_1e-1/20230603_172035/vis_data/scalars.json new file mode 100644 index 0000000000000000000000000000000000000000..35d8484da0cea731e1374b660d952bdcf003f989 --- /dev/null +++ b/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_1e-1/20230603_172035/vis_data/scalars.json @@ -0,0 +1,200 @@ +{"lr": 0.00039997529889041205, "data_time": 0.8876386165618897, "loss": 0.028243570402264596, "time": 2.2404876470565798, "epoch": 1, "memory": 15587, "step": 100} +{"lr": 0.0003999002014233684, "data_time": 0.011860418319702148, "loss": 0.019886883813887833, "time": 2.0559006690979005, "epoch": 1, "memory": 15587, "step": 200} +{"lr": 0.0003997747244903853, "data_time": 0.0017872333526611328, "loss": 0.026142413867637514, "time": 2.270457720756531, "epoch": 1, "memory": 15587, "step": 300} +{"lr": 0.00039959890052580966, "data_time": 0.002608990669250488, "loss": 0.012897801818326116, "time": 2.2108946800231934, "epoch": 1, "memory": 15587, "step": 400} +{"lr": 0.0003993727749781197, "data_time": 0.002134227752685547, "loss": 0.01874726158566773, "time": 2.18149094581604, "epoch": 1, "memory": 15587, "step": 500} +{"lr": 0.000399096406298175, "data_time": 0.002037715911865234, "loss": 0.011248944280669092, "time": 2.033232235908508, "epoch": 1, "memory": 15587, "step": 600} +{"lr": 0.0003987698659241082, "data_time": 0.0017608880996704101, "loss": 0.010819037957116962, "time": 2.265449380874634, "epoch": 1, "memory": 15587, "step": 700} +{"lr": 0.0003983932382628598, "data_time": 0.001622152328491211, "loss": 0.015379989007487892, "time": 2.1147350072860718, "epoch": 1, "memory": 15587, "step": 800} +{"lr": 0.00039796662066835946, "data_time": 0.0020137786865234374, "loss": 0.006578811584040523, "time": 1.3881823301315308, "epoch": 1, "memory": 15587, "step": 900} +{"lr": 0.00039749012341636, "data_time": 0.002531886100769043, "loss": 0.014720092481002211, "time": 2.1194176197052004, "epoch": 1, "memory": 15587, "step": 1000} +{"lr": 0.00039696386967593436, "data_time": 0.0016518354415893555, "loss": 0.0107122294139117, "time": 2.2452640771865844, "epoch": 1, "memory": 15587, "step": 1100} +{"lr": 0.00039638799547763706, "data_time": 0.0018185138702392577, "loss": 0.008884851774200796, "time": 2.018517756462097, "epoch": 1, "memory": 15587, "step": 1200} +{"lr": 0.0003957626496783407, "data_time": 0.001597142219543457, "loss": 0.008482847176492215, "time": 1.4338923454284669, "epoch": 1, "memory": 15587, "step": 1300} +{"lr": 0.0003950879939227595, "data_time": 0.0014984846115112306, "loss": 0.011113102734088897, "time": 1.4645151376724244, "epoch": 1, "memory": 15587, "step": 1400} +{"lr": 0.0003943642026016663, "data_time": 0.04784624576568604, "loss": 0.008591280272230505, "time": 1.5325900077819825, "epoch": 1, "memory": 15587, "step": 1500} +{"lr": 0.0003935914628068133, "data_time": 0.0015847206115722657, "loss": 0.0057054765988141295, "time": 0.7563295125961303, "epoch": 1, "memory": 15587, "step": 1600} +{"lr": 0.00039276997428257205, "data_time": 0.0028943300247192385, "loss": 0.012270110519602894, "time": 0.8250779867172241, "epoch": 1, "memory": 15587, "step": 1700} +{"lr": 0.0003918999493743001, "data_time": 0.0016125917434692382, "loss": 0.008454771200194954, "time": 0.7425617933273315, "epoch": 1, "memory": 15587, "step": 1800} +{"lr": 0.00039098161297345475, "data_time": 0.001694631576538086, "loss": 0.012858521239832044, "time": 0.7422439813613891, "epoch": 1, "memory": 15587, "step": 1900} +{"accuracy/top1": 74.68077087402344, "data_time": 1.456217494504205, "time": 1.9215284462632805, "step": 1} +{"lr": 0.0003894734390071047, "data_time": 0.0020853281021118164, "loss": 0.006674895528703928, "time": 0.6185879945755005, "epoch": 2, "memory": 15587, "step": 2054} +{"lr": 0.0003884334872841784, "data_time": 0.0017771005630493164, "loss": 0.009980583051219582, "time": 0.6806638956069946, "epoch": 2, "memory": 15585, "step": 2154} +{"lr": 0.00038734612010935703, "data_time": 0.0025436878204345703, "loss": 0.010095262108370662, "time": 0.8138118028640747, "epoch": 2, "memory": 15585, "step": 2254} +{"lr": 0.00038621161855458035, "data_time": 0.0020783185958862306, "loss": 0.011340158013626932, "time": 0.6611358404159546, "epoch": 2, "memory": 15585, "step": 2354} +{"lr": 0.0003850302758754841, "data_time": 0.0018435478210449218, "loss": 0.007705595251172781, "time": 0.6108261346817017, "epoch": 2, "memory": 15585, "step": 2454} +{"lr": 0.00038380239743559646, "data_time": 0.0028277873992919923, "loss": 0.006682888045907021, "time": 0.7921959161758423, "epoch": 2, "memory": 15585, "step": 2554} +{"lr": 0.00038252830062740647, "data_time": 0.00267488956451416, "loss": 0.012029745010659098, "time": 0.5611567974090577, "epoch": 2, "memory": 15585, "step": 2654} +{"lr": 0.00038120831479032175, "data_time": 0.0017798185348510743, "loss": 0.004605051083490253, "time": 0.6864491939544678, "epoch": 2, "memory": 15585, "step": 2754} +{"lr": 0.0003798427811255361, "data_time": 0.001991939544677734, "loss": 0.005698884697631001, "time": 0.6648210287094116, "epoch": 2, "memory": 15585, "step": 2854} +{"lr": 0.0003784320526078357, "data_time": 0.001694631576538086, "loss": 0.00974665260873735, "time": 0.5864532709121704, "epoch": 2, "memory": 15585, "step": 2954} +{"lr": 0.0003769764938943556, "data_time": 0.00160369873046875, "loss": 0.010500320605933667, "time": 0.6202181816101074, "epoch": 2, "memory": 15585, "step": 3054} +{"lr": 0.0003754764812303242, "data_time": 0.003342127799987793, "loss": 0.00864911307580769, "time": 0.7500198602676391, "epoch": 2, "memory": 15585, "step": 3154} +{"lr": 0.0003739324023518048, "data_time": 0.0016830921173095702, "loss": 0.00743028549477458, "time": 0.6538217544555665, "epoch": 2, "memory": 15585, "step": 3254} +{"lr": 0.0003723446563854724, "data_time": 0.0019375801086425782, "loss": 0.013865859108045697, "time": 0.6082741975784302, "epoch": 2, "memory": 15585, "step": 3354} +{"lr": 0.0003707136537454424, "data_time": 0.0018967151641845702, "loss": 0.005103431269526481, "time": 0.6341658353805542, "epoch": 2, "memory": 15585, "step": 3454} +{"lr": 0.0003690398160271841, "data_time": 0.002397584915161133, "loss": 0.007460079155862331, "time": 0.7064965248107911, "epoch": 2, "memory": 15585, "step": 3554} +{"lr": 0.00036732357589854426, "data_time": 0.001503133773803711, "loss": 0.002773802587762475, "time": 0.6909519910812378, "epoch": 2, "memory": 15585, "step": 3654} +{"lr": 0.00036556537698790475, "data_time": 0.0016093492507934571, "loss": 0.004564368352293968, "time": 0.6923271179199219, "epoch": 2, "memory": 15585, "step": 3754} +{"lr": 0.0003637656737695118, "data_time": 0.003058767318725586, "loss": 0.006924367323517799, "time": 0.6548510074615479, "epoch": 2, "memory": 15585, "step": 3854} +{"accuracy/top1": 96.38495635986328, "data_time": 0.405672820409139, "time": 0.7952233950297037, "step": 2} +{"lr": 0.00036091403361175644, "data_time": 0.0019866466522216798, "loss": 0.006628051167353988, "time": 0.7163207054138183, "epoch": 3, "memory": 15585, "step": 4008} +{"lr": 0.00035901102535335625, "data_time": 0.0019848108291625976, "loss": 0.011745957238599659, "time": 0.6726689815521241, "epoch": 3, "memory": 15585, "step": 4108} +{"lr": 0.00035706820701190153, "data_time": 0.0020529985427856444, "loss": 0.007274257158860565, "time": 0.6573845863342285, "epoch": 3, "memory": 15585, "step": 4208} +{"lr": 0.00035508608078364215, "data_time": 0.0017499208450317382, "loss": 0.012712002824991941, "time": 0.6188857316970825, "epoch": 3, "memory": 15585, "step": 4308} +{"lr": 0.00035306515902546755, "data_time": 0.0019903659820556642, "loss": 0.005651786830276251, "time": 0.7310869455337524, "epoch": 3, "memory": 15585, "step": 4408} +{"lr": 0.0003510059641224655, "data_time": 0.0020628929138183593, "loss": 0.008439291268587112, "time": 0.6581233501434326, "epoch": 3, "memory": 15585, "step": 4508} +{"lr": 0.00034890902835289306, "data_time": 0.0020843505859375, "loss": 0.006331758527085185, "time": 0.6894623756408691, "epoch": 3, "memory": 15585, "step": 4608} +{"lr": 0.00034677489375058753, "data_time": 0.0017110586166381836, "loss": 0.004314387775957584, "time": 0.5919305324554444, "epoch": 3, "memory": 15585, "step": 4708} +{"lr": 0.00034460411196485875, "data_time": 0.0019282817840576172, "loss": 0.008494613505899907, "time": 0.5827285051345825, "epoch": 3, "memory": 15585, "step": 4808} +{"lr": 0.0003423972441178926, "data_time": 0.0023634672164916993, "loss": 0.005785406613722443, "time": 0.6835513114929199, "epoch": 3, "memory": 15585, "step": 4908} +{"lr": 0.0003401548606597075, "data_time": 0.001933741569519043, "loss": 0.0043906884267926214, "time": 0.6141704320907593, "epoch": 3, "memory": 15585, "step": 5008} +{"lr": 0.0003378775412207003, "data_time": 0.002954411506652832, "loss": 0.009896530024707317, "time": 0.6562421321868896, "epoch": 3, "memory": 15585, "step": 5108} +{"lr": 0.0003355658744618178, "data_time": 0.00216059684753418, "loss": 0.0027915054000914098, "time": 0.6658342838287353, "epoch": 3, "memory": 15585, "step": 5208} +{"lr": 0.0003332204579223944, "data_time": 0.0017322301864624023, "loss": 0.004801007127389312, "time": 0.7218419790267945, "epoch": 3, "memory": 15585, "step": 5308} +{"lr": 0.0003308418978656968, "data_time": 0.001978492736816406, "loss": 0.00633810474537313, "time": 0.595003867149353, "epoch": 3, "memory": 15585, "step": 5408} +{"lr": 0.0003284308091222099, "data_time": 0.002231335639953613, "loss": 0.007586751598864794, "time": 0.6756606817245483, "epoch": 3, "memory": 15585, "step": 5508} +{"lr": 0.0003259878149307108, "data_time": 0.002854156494140625, "loss": 0.004832551255822182, "time": 0.7135452508926392, "epoch": 3, "memory": 15585, "step": 5608} +{"lr": 0.0003235135467771688, "data_time": 0.0018404006958007812, "loss": 0.005467744776979089, "time": 0.7036562442779541, "epoch": 3, "memory": 15585, "step": 5708} +{"lr": 0.0003210086442315141, "data_time": 0.0024457216262817384, "loss": 0.004971952084451914, "time": 0.6197478532791137, "epoch": 3, "memory": 15585, "step": 5808} +{"accuracy/top1": 99.25890350341797, "data_time": 0.40395546754201256, "time": 0.7969277540842692, "step": 3} +{"lr": 0.00031709267628953966, "data_time": 0.002018284797668457, "loss": 0.007866184692829847, "time": 0.6292434215545655, "epoch": 4, "memory": 15585, "step": 5962} +{"lr": 0.00031451289120303685, "data_time": 0.002409625053405762, "loss": 0.006956744473427534, "time": 0.6462630987167358, "epoch": 4, "memory": 15585, "step": 6062} +{"lr": 0.000311904798290621, "data_time": 0.0021924972534179688, "loss": 0.010229665692895651, "time": 0.7094257116317749, "epoch": 4, "memory": 15585, "step": 6162} +{"lr": 0.00030926907171438685, "data_time": 0.0024345874786376952, "loss": 0.003526915470138192, "time": 0.7614258527755737, "epoch": 4, "memory": 15585, "step": 6262} +{"lr": 0.000306606392779414, "data_time": 0.0021973371505737303, "loss": 0.00594695471227169, "time": 0.6545991182327271, "epoch": 4, "memory": 15585, "step": 6362} +{"lr": 0.0003039174497576579, "data_time": 0.0018962860107421876, "loss": 0.005417559342458844, "time": 0.6833284139633179, "epoch": 4, "memory": 15585, "step": 6462} +{"lr": 0.0003012029377100387, "data_time": 0.0024603605270385742, "loss": 0.002504068287089467, "time": 0.7282400608062745, "epoch": 4, "memory": 15585, "step": 6562} +{"lr": 0.00029846355830677674, "data_time": 0.002100968360900879, "loss": 0.005125401075929403, "time": 0.6195366144180298, "epoch": 4, "memory": 15585, "step": 6662} +{"lr": 0.0002957000196460183, "data_time": 0.0019130229949951172, "loss": 0.004303906485438347, "time": 0.6073935031890869, "epoch": 4, "memory": 15585, "step": 6762} +{"lr": 0.0002929130360708003, "data_time": 0.002662348747253418, "loss": 0.004120248276740313, "time": 0.6531259536743164, "epoch": 4, "memory": 15585, "step": 6862} +{"lr": 0.0002901033279844016, "data_time": 0.0018069744110107422, "loss": 0.008747064601629972, "time": 0.6163111209869385, "epoch": 4, "memory": 15585, "step": 6962} +{"lr": 0.00028727162166412623, "data_time": 0.0020123958587646485, "loss": 0.001685183309018612, "time": 0.6028306007385253, "epoch": 4, "memory": 15585, "step": 7062} +{"lr": 0.000284418649073569, "data_time": 0.0018334388732910156, "loss": 0.0052682790905237194, "time": 0.5562018871307373, "epoch": 4, "memory": 15585, "step": 7162} +{"lr": 0.00028154514767341203, "data_time": 0.001651453971862793, "loss": 0.005815492710098624, "time": 0.5893539190292358, "epoch": 4, "memory": 15585, "step": 7262} +{"lr": 0.0002786518602307988, "data_time": 0.002273893356323242, "loss": 0.0016718230675905944, "time": 0.6399655103683471, "epoch": 4, "memory": 15585, "step": 7362} +{"lr": 0.0002757395346273384, "data_time": 0.0016340494155883789, "loss": 0.0044249090366065506, "time": 0.6398932933807373, "epoch": 4, "memory": 15585, "step": 7462} +{"lr": 0.0002728089236657853, "data_time": 0.0018706560134887696, "loss": 0.00525554371997714, "time": 0.6062353849411011, "epoch": 4, "memory": 15585, "step": 7562} +{"lr": 0.0002698607848754497, "data_time": 0.0019966840744018556, "loss": 0.00600892910733819, "time": 0.6009192943572998, "epoch": 4, "memory": 15585, "step": 7662} +{"lr": 0.0002668958803163847, "data_time": 0.0023994922637939455, "loss": 0.0010058487765491008, "time": 0.6716243028640747, "epoch": 4, "memory": 15585, "step": 7762} +{"accuracy/top1": 99.64163208007812, "data_time": 0.2835105021794637, "time": 0.6705821593602498, "step": 4} +{"lr": 0.00026229890681650394, "data_time": 0.0017737150192260742, "loss": 0.0033932299818843605, "time": 0.7760605573654175, "epoch": 5, "memory": 15585, "step": 7916} +{"lr": 0.00025929487224058475, "data_time": 0.002240920066833496, "loss": 0.0032494002487510444, "time": 0.7113546133041382, "epoch": 5, "memory": 15585, "step": 8016} +{"lr": 0.0002562768030631272, "data_time": 0.0018189668655395508, "loss": 0.005762854823842645, "time": 0.6504134893417358, "epoch": 5, "memory": 15585, "step": 8116} +{"lr": 0.0002532454794203871, "data_time": 0.0017086029052734374, "loss": 0.003760502813383937, "time": 0.6094439506530762, "epoch": 5, "memory": 15585, "step": 8216} +{"lr": 0.0002502016848747476, "data_time": 0.0030717372894287108, "loss": 0.0017563470639288426, "time": 0.5812451362609863, "epoch": 5, "memory": 15585, "step": 8316} +{"lr": 0.0002471462062121758, "data_time": 0.003067421913146973, "loss": 0.008362926309928298, "time": 0.7136571645736695, "epoch": 5, "memory": 15585, "step": 8416} +{"lr": 0.00024407983323884952, "data_time": 0.0015196561813354491, "loss": 0.003412273805588484, "time": 0.6228043079376221, "epoch": 5, "memory": 15585, "step": 8516} +{"lr": 0.0002410033585770006, "data_time": 0.001826167106628418, "loss": 0.0027474768459796904, "time": 0.656356930732727, "epoch": 5, "memory": 15585, "step": 8616} +{"lr": 0.00023791757746003292, "data_time": 0.0016218900680541992, "loss": 0.0007744169328361749, "time": 0.646912670135498, "epoch": 5, "memory": 15585, "step": 8716} +{"lr": 0.00023482328752696182, "data_time": 0.0016498088836669922, "loss": 0.002677060244604945, "time": 0.6467357397079467, "epoch": 5, "memory": 15585, "step": 8816} +{"lr": 0.00023172128861623345, "data_time": 0.002625274658203125, "loss": 0.004404339054599404, "time": 0.806192684173584, "epoch": 5, "memory": 15585, "step": 8916} +{"lr": 0.000228612382558976, "data_time": 0.0021098852157592773, "loss": 0.008860325813293457, "time": 0.670192551612854, "epoch": 5, "memory": 15585, "step": 9016} +{"lr": 0.00022549737297173692, "data_time": 0.0018606185913085938, "loss": 0.005228723445907235, "time": 0.67873113155365, "epoch": 5, "memory": 15585, "step": 9116} +{"lr": 0.00022237706504875477, "data_time": 0.001738286018371582, "loss": 0.0019822140224277974, "time": 0.648528790473938, "epoch": 5, "memory": 15585, "step": 9216} +{"lr": 0.00021925226535382818, "data_time": 0.0020878791809082033, "loss": 0.0024604568723589184, "time": 0.7061655282974243, "epoch": 5, "memory": 15585, "step": 9316} +{"lr": 0.0002161237816118266, "data_time": 0.002283334732055664, "loss": 0.004188699461519718, "time": 0.6657184839248658, "epoch": 5, "memory": 15585, "step": 9416} +{"lr": 0.00021299242249990383, "data_time": 0.0018518686294555664, "loss": 0.0017229572404175997, "time": 0.726628589630127, "epoch": 5, "memory": 15585, "step": 9516} +{"lr": 0.00020985899743846363, "data_time": 0.001510763168334961, "loss": 0.002604235848411918, "time": 0.6141276121139526, "epoch": 5, "memory": 15585, "step": 9616} +{"lr": 0.00020672431638193363, "data_time": 0.0016469955444335938, "loss": 0.0023076528683304786, "time": 0.6750414371490479, "epoch": 5, "memory": 15585, "step": 9716} +{"accuracy/top1": 99.7634048461914, "data_time": 0.29454609553019206, "time": 0.6848266045252482, "step": 5} +{"lr": 0.00020189632111931318, "data_time": 0.0018271207809448242, "loss": 0.003468033019453287, "time": 0.6092110872268677, "epoch": 6, "memory": 15585, "step": 9870} +{"lr": 0.00019876209288860824, "data_time": 0.002828526496887207, "loss": 0.0030729669611901044, "time": 0.7728718042373657, "epoch": 6, "memory": 15585, "step": 9970} +{"lr": 0.00019562947708532286, "data_time": 0.0031618118286132813, "loss": 0.001884378120303154, "time": 0.7918673515319824, "epoch": 6, "memory": 15585, "step": 10070} +{"lr": 0.00019249928345470104, "data_time": 0.002530360221862793, "loss": 0.0024694283027201893, "time": 0.7146633625030517, "epoch": 6, "memory": 15585, "step": 10170} +{"lr": 0.00018937232111588298, "data_time": 0.0030922651290893554, "loss": 0.007032108958810568, "time": 0.7631094932556153, "epoch": 6, "memory": 15585, "step": 10270} +{"lr": 0.000186249398352757, "data_time": 0.0018081903457641602, "loss": 0.002067164471372962, "time": 0.9599995136260986, "epoch": 6, "memory": 15585, "step": 10370} +{"lr": 0.0001831313224050277, "data_time": 0.0018959522247314453, "loss": 0.0016676651313900948, "time": 0.5958833456039428, "epoch": 6, "memory": 15585, "step": 10470} +{"lr": 0.00018001889925955291, "data_time": 0.0026535987854003906, "loss": 0.005793391447514296, "time": 0.8120676755905152, "epoch": 6, "memory": 15585, "step": 10570} +{"lr": 0.00017691293344200646, "data_time": 0.0017524480819702149, "loss": 0.005128879193216562, "time": 0.6570345640182496, "epoch": 6, "memory": 15585, "step": 10670} +{"lr": 0.00017381422780891689, "data_time": 0.0018006086349487305, "loss": 0.0054186578840017315, "time": 0.6585187196731568, "epoch": 6, "memory": 15585, "step": 10770} +{"lr": 0.00017072358334013806, "data_time": 0.002730393409729004, "loss": 0.004597321106120944, "time": 0.7017984628677368, "epoch": 6, "memory": 15585, "step": 10870} +{"lr": 0.00016764179893180575, "data_time": 0.0021402835845947266, "loss": 0.004850745666772127, "time": 0.6412996053695679, "epoch": 6, "memory": 15585, "step": 10970} +{"lr": 0.00016456967118983176, "data_time": 0.0017818689346313476, "loss": 0.003687735600396991, "time": 0.6251964330673218, "epoch": 6, "memory": 15585, "step": 11070} +{"lr": 0.00016150799422398984, "data_time": 0.002596926689147949, "loss": 0.00107732517644763, "time": 0.7114237546920776, "epoch": 6, "memory": 15585, "step": 11170} +{"lr": 0.0001584575594426489, "data_time": 0.0016971349716186524, "loss": 0.002558692218735814, "time": 0.6332179069519043, "epoch": 6, "memory": 15585, "step": 11270} +{"lr": 0.00015541915534820135, "data_time": 0.002237415313720703, "loss": 0.000920321699231863, "time": 0.6279433727264404, "epoch": 6, "memory": 15585, "step": 11370} +{"lr": 0.00015239356733324542, "data_time": 0.0030861139297485352, "loss": 0.001232651760801673, "time": 0.7953397035598755, "epoch": 6, "memory": 15585, "step": 11470} +{"lr": 0.00014938157747756997, "data_time": 0.0020141363143920898, "loss": 0.005741068674251437, "time": 0.6601707935333252, "epoch": 6, "memory": 15585, "step": 11570} +{"lr": 0.00014638396434599565, "data_time": 0.0017972707748413086, "loss": 0.003222867427393794, "time": 0.7659487247467041, "epoch": 6, "memory": 15585, "step": 11670} +{"accuracy/top1": 99.83647155761719, "data_time": 0.4005567630132039, "time": 0.787556258837382, "step": 6} +{"lr": 0.00014179754513556176, "data_time": 0.0019487857818603516, "loss": 0.007016163924708963, "time": 0.6308538436889648, "epoch": 7, "memory": 15585, "step": 11824} +{"lr": 0.00013883992334674444, "data_time": 0.0015415191650390626, "loss": 0.004243675014004111, "time": 0.6966867923736573, "epoch": 7, "memory": 15585, "step": 11924} +{"lr": 0.00013589940317867467, "data_time": 0.002080273628234863, "loss": 0.003451137337833643, "time": 0.8140087842941284, "epoch": 7, "memory": 15585, "step": 12024} +{"lr": 0.0001329767447220783, "data_time": 0.001691150665283203, "loss": 0.0009962485171854496, "time": 0.6610537767410278, "epoch": 7, "memory": 15585, "step": 12124} +{"lr": 0.00013007270345063488, "data_time": 0.0023084163665771486, "loss": 0.006037602573633194, "time": 0.8553447246551513, "epoch": 7, "memory": 15585, "step": 12224} +{"lr": 0.00012718803002569412, "data_time": 0.0016849517822265625, "loss": 0.0025534901302307843, "time": 0.6490463018417358, "epoch": 7, "memory": 15585, "step": 12324} +{"lr": 0.0001243234701022397, "data_time": 0.0015696287155151367, "loss": 0.00087009915150702, "time": 0.6494510889053344, "epoch": 7, "memory": 15585, "step": 12424} +{"lr": 0.00012147976413614597, "data_time": 0.0015357732772827148, "loss": 0.002598012378439307, "time": 0.6391769170761108, "epoch": 7, "memory": 15585, "step": 12524} +{"lr": 0.0001186576471927784, "data_time": 0.00181124210357666, "loss": 0.005000874120742083, "time": 0.6212236166000367, "epoch": 7, "memory": 15585, "step": 12624} +{"lr": 0.00011585784875698867, "data_time": 0.0029226303100585937, "loss": 0.004526560194790363, "time": 0.6219683170318604, "epoch": 7, "memory": 15585, "step": 12724} +{"lr": 0.00011308109254454993, "data_time": 0.0016169786453247071, "loss": 0.0030195570085197686, "time": 0.6195187330245971, "epoch": 7, "memory": 15585, "step": 12824} +{"lr": 0.00011032809631508466, "data_time": 0.0020107746124267576, "loss": 0.003165309550240636, "time": 0.6924211740493774, "epoch": 7, "memory": 15585, "step": 12924} +{"lr": 0.00010759957168653253, "data_time": 0.0021471261978149416, "loss": 0.004263857891783118, "time": 0.7161961793899536, "epoch": 7, "memory": 15585, "step": 13024} +{"lr": 0.0001048962239512054, "data_time": 0.0017017841339111329, "loss": 0.0020327826496213675, "time": 0.6079156398773193, "epoch": 7, "memory": 15585, "step": 13124} +{"lr": 0.00010221875189347644, "data_time": 0.0034569501876831055, "loss": 0.0033277326729148626, "time": 0.6889806032180786, "epoch": 7, "memory": 15585, "step": 13224} +{"lr": 9.95678476091537e-05, "data_time": 0.0016147613525390625, "loss": 0.0009845915716141461, "time": 0.5877719163894654, "epoch": 7, "memory": 15585, "step": 13324} +{"lr": 9.694419632657999e-05, "data_time": 0.0015826940536499024, "loss": 0.0025791836436837912, "time": 0.626505184173584, "epoch": 7, "memory": 15585, "step": 13424} +{"lr": 9.434847622951059e-05, "data_time": 0.0017975568771362305, "loss": 0.002552804397419095, "time": 0.6686267852783203, "epoch": 7, "memory": 15585, "step": 13524} +{"lr": 9.178135828180905e-05, "data_time": 0.002054429054260254, "loss": 0.002057436853647232, "time": 0.7246514320373535, "epoch": 7, "memory": 15585, "step": 13624} +{"accuracy/top1": 99.89214324951172, "data_time": 0.3015367825826009, "time": 0.6877463261286417, "step": 7} +{"lr": 8.788546579134055e-05, "data_time": 0.0020665407180786135, "loss": 0.0019316677935421467, "time": 0.6268484354019165, "epoch": 8, "memory": 15585, "step": 13778} +{"lr": 8.539396307206318e-05, "data_time": 0.0024080991744995115, "loss": 0.0037416914012283085, "time": 0.6612948894500732, "epoch": 8, "memory": 15585, "step": 13878} +{"lr": 8.293337714106816e-05, "data_time": 0.0017055273056030273, "loss": 0.0025136511772871017, "time": 0.6687227487564087, "epoch": 8, "memory": 15585, "step": 13978} +{"lr": 8.05043440315878e-05, "data_time": 0.0018981695175170898, "loss": 0.0017824933398514985, "time": 0.5936944007873535, "epoch": 8, "memory": 15585, "step": 14078} +{"lr": 7.810749162081216e-05, "data_time": 0.0015995264053344726, "loss": 0.0017375380732119084, "time": 0.5466150045394897, "epoch": 8, "memory": 15585, "step": 14178} +{"lr": 7.574343946758993e-05, "data_time": 0.0028189897537231447, "loss": 0.0018452064599841834, "time": 0.7312486648559571, "epoch": 8, "memory": 15585, "step": 14278} +{"lr": 7.34127986522793e-05, "data_time": 0.0018247842788696289, "loss": 0.0037062933202832936, "time": 0.6396161079406738, "epoch": 8, "memory": 15585, "step": 14378} +{"lr": 7.111617161879136e-05, "data_time": 0.0030232667922973633, "loss": 0.001308464864268899, "time": 0.6824496030807495, "epoch": 8, "memory": 15585, "step": 14478} +{"lr": 6.885415201886485e-05, "data_time": 0.0022236108779907227, "loss": 0.0008456234354525805, "time": 0.7071934700012207, "epoch": 8, "memory": 15585, "step": 14578} +{"lr": 6.662732455861337e-05, "data_time": 0.002220606803894043, "loss": 0.0007111166138201952, "time": 0.6272669553756713, "epoch": 8, "memory": 15585, "step": 14678} +{"lr": 6.443626484738661e-05, "data_time": 0.002686810493469238, "loss": 0.0008525562472641468, "time": 0.8114670276641845, "epoch": 8, "memory": 15585, "step": 14778} +{"lr": 6.22815392489815e-05, "data_time": 0.0021439313888549803, "loss": 0.002012418955564499, "time": 0.6568866491317749, "epoch": 8, "memory": 15585, "step": 14878} +{"lr": 6.016370473524328e-05, "data_time": 0.0018251895904541015, "loss": 0.0014770927838981152, "time": 0.6630625486373901, "epoch": 8, "memory": 15585, "step": 14978} +{"lr": 5.80833087420956e-05, "data_time": 0.0023761987686157227, "loss": 0.004584140982478857, "time": 0.7106636047363282, "epoch": 8, "memory": 15585, "step": 15078} +{"lr": 5.604088902803397e-05, "data_time": 0.002094554901123047, "loss": 0.000984257971867919, "time": 0.6852351188659668, "epoch": 8, "memory": 15585, "step": 15178} +{"lr": 5.403697353512148e-05, "data_time": 0.0016436576843261719, "loss": 0.0015397735871374608, "time": 0.7779737949371338, "epoch": 8, "memory": 15585, "step": 15278} +{"lr": 5.207208025252174e-05, "data_time": 0.0031928300857543947, "loss": 0.0008444434963166714, "time": 0.720936918258667, "epoch": 8, "memory": 15585, "step": 15378} +{"lr": 5.014671708260458e-05, "data_time": 0.0016473293304443359, "loss": 0.002385187242180109, "time": 0.5953363895416259, "epoch": 8, "memory": 15585, "step": 15478} +{"lr": 4.826138170965952e-05, "data_time": 0.0020528078079223634, "loss": 0.0010014158207923174, "time": 0.6689431428909302, "epoch": 8, "memory": 15585, "step": 15578} +{"accuracy/top1": 99.91301727294922, "data_time": 0.3978493531545003, "time": 0.7914119005203247, "step": 8} +{"lr": 4.5437373040402475e-05, "data_time": 0.0031154394149780274, "loss": 0.002939263405278325, "time": 0.6684917211532593, "epoch": 9, "memory": 15585, "step": 15732} +{"lr": 4.3655875036148156e-05, "data_time": 0.0019965887069702147, "loss": 0.0011175610590726138, "time": 0.6349424123764038, "epoch": 9, "memory": 15585, "step": 15832} +{"lr": 4.19160826370641e-05, "data_time": 0.0027050495147705076, "loss": 0.0008449246641248464, "time": 0.8405579805374146, "epoch": 9, "memory": 15585, "step": 15932} +{"lr": 4.021844555952455e-05, "data_time": 0.0019388198852539062, "loss": 0.0017927538603544236, "time": 0.5914006471633911, "epoch": 9, "memory": 15585, "step": 16032} +{"lr": 3.8563402623236864e-05, "data_time": 0.0016057014465332032, "loss": 0.0018114028964191676, "time": 0.6190768957138062, "epoch": 9, "memory": 15585, "step": 16132} +{"lr": 3.695138163781135e-05, "data_time": 0.002338695526123047, "loss": 0.0004661856684833765, "time": 0.616426658630371, "epoch": 9, "memory": 15585, "step": 16232} +{"lr": 3.538279929217756e-05, "data_time": 0.001933455467224121, "loss": 0.0013150629587471486, "time": 0.6978412628173828, "epoch": 9, "memory": 15585, "step": 16332} +{"lr": 3.385806104687531e-05, "data_time": 0.0015532732009887694, "loss": 0.0005993003025650979, "time": 0.6358365535736084, "epoch": 9, "memory": 15585, "step": 16432} +{"lr": 3.237756102924734e-05, "data_time": 0.0025865793228149413, "loss": 0.0020424792543053626, "time": 0.6661193609237671, "epoch": 9, "memory": 15585, "step": 16532} +{"lr": 3.094168193156213e-05, "data_time": 0.0022933483123779297, "loss": 0.0053255957085639235, "time": 0.6813425064086914, "epoch": 9, "memory": 15585, "step": 16632} +{"lr": 2.95507949120924e-05, "data_time": 0.0017201662063598632, "loss": 0.0013453535735607146, "time": 0.6248709440231324, "epoch": 9, "memory": 15585, "step": 16732} +{"lr": 2.8205259499174602e-05, "data_time": 0.001804494857788086, "loss": 0.0014839986339211463, "time": 0.596595048904419, "epoch": 9, "memory": 15585, "step": 16832} +{"lr": 2.690542349827549e-05, "data_time": 0.0016358375549316406, "loss": 0.002500947145745158, "time": 0.6019083499908447, "epoch": 9, "memory": 15585, "step": 16932} +{"lr": 2.565162290208795e-05, "data_time": 0.002254962921142578, "loss": 0.002461536694318056, "time": 0.6005879163742065, "epoch": 9, "memory": 15585, "step": 17032} +{"lr": 2.4444181803680975e-05, "data_time": 0.013503074645996094, "loss": 0.001660887710750103, "time": 0.5341065168380738, "epoch": 9, "memory": 15585, "step": 17132} +{"lr": 2.32834123127253e-05, "data_time": 0.0018811702728271484, "loss": 0.0010136137250810861, "time": 0.957408356666565, "epoch": 9, "memory": 15585, "step": 17232} +{"lr": 2.2169614474816763e-05, "data_time": 0.001784205436706543, "loss": 0.0008078602608293295, "time": 0.5961580276489258, "epoch": 9, "memory": 15585, "step": 17332} +{"lr": 2.110307619391749e-05, "data_time": 0.003448748588562012, "loss": 0.003968959068879485, "time": 0.7766430377960205, "epoch": 9, "memory": 15585, "step": 17432} +{"lr": 2.008407315793661e-05, "data_time": 0.0031492948532104493, "loss": 0.00026420974172651767, "time": 0.6827759742736816, "epoch": 9, "memory": 15585, "step": 17532} +{"accuracy/top1": 99.88518524169922, "data_time": 0.23833333651224772, "time": 0.6253702561060588, "step": 9} +{"lr": 1.8608381954463387e-05, "data_time": 0.002872872352600098, "loss": 0.000977274589240551, "time": 0.7066831350326538, "epoch": 10, "memory": 15585, "step": 17686} +{"lr": 1.771127410231484e-05, "data_time": 0.0022667884826660157, "loss": 0.0023614218458533286, "time": 0.5895065546035767, "epoch": 10, "memory": 15585, "step": 17786} +{"lr": 1.6862578237010542e-05, "data_time": 0.002544879913330078, "loss": 0.0011892936658114196, "time": 0.6707264423370362, "epoch": 10, "memory": 15585, "step": 17886} +{"lr": 1.6062513736694383e-05, "data_time": 0.0019954442977905273, "loss": 0.0014919942710548638, "time": 0.5967118501663208, "epoch": 10, "memory": 15585, "step": 17986} +{"lr": 1.531128740886036e-05, "data_time": 0.002044820785522461, "loss": 0.001691504241898656, "time": 0.5839269876480102, "epoch": 10, "memory": 15585, "step": 18086} +{"lr": 1.4609093436895328e-05, "data_time": 0.0026756048202514647, "loss": 0.00056116315536201, "time": 0.650002384185791, "epoch": 10, "memory": 15585, "step": 18186} +{"lr": 1.3956113329884566e-05, "data_time": 0.0022931337356567384, "loss": 0.0010490994900465012, "time": 0.7251847743988037, "epoch": 10, "memory": 15585, "step": 18286} +{"lr": 1.335251587569401e-05, "data_time": 0.0018516778945922852, "loss": 0.002795368479564786, "time": 0.566193962097168, "epoch": 10, "memory": 15585, "step": 18386} +{"lr": 1.2798457097340337e-05, "data_time": 0.0019077301025390626, "loss": 0.00033678459003567697, "time": 0.6044378995895385, "epoch": 10, "memory": 15585, "step": 18486} +{"lr": 1.229408021266094e-05, "data_time": 0.002157902717590332, "loss": 0.000383464340120554, "time": 0.6124505996704102, "epoch": 10, "memory": 15585, "step": 18586} +{"lr": 1.1839515597293723e-05, "data_time": 0.002776336669921875, "loss": 0.00039589018560945986, "time": 0.6678505659103393, "epoch": 10, "memory": 15585, "step": 18686} +{"lr": 1.1434880750976406e-05, "data_time": 0.001705479621887207, "loss": 0.0033804717008024455, "time": 0.6809921503067017, "epoch": 10, "memory": 15585, "step": 18786} +{"lr": 1.108028026717431e-05, "data_time": 0.0017388582229614258, "loss": 0.0003105533309280872, "time": 0.650597357749939, "epoch": 10, "memory": 15585, "step": 18886} +{"lr": 1.0775805806044138e-05, "data_time": 0.001784682273864746, "loss": 0.0019477454479783773, "time": 0.5854311943054199, "epoch": 10, "memory": 15585, "step": 18986} +{"lr": 1.0521536070740776e-05, "data_time": 0.004630064964294434, "loss": 0.0007190603762865066, "time": 0.6267573833465576, "epoch": 10, "memory": 15585, "step": 19086} +{"lr": 1.03175367870735e-05, "data_time": 0.0026628255844116213, "loss": 0.002255278080701828, "time": 0.6910584926605224, "epoch": 10, "memory": 15585, "step": 19186} +{"lr": 1.0163860686516628e-05, "data_time": 0.002797985076904297, "loss": 0.0014115124009549617, "time": 0.5970310688018798, "epoch": 10, "memory": 15585, "step": 19286} +{"lr": 1.0060547492578997e-05, "data_time": 0.0020235061645507814, "loss": 0.0008689676877111196, "time": 0.638939642906189, "epoch": 10, "memory": 15585, "step": 19386} +{"lr": 1.0007623910535892e-05, "data_time": 0.0019460678100585937, "loss": 0.0004255445208400488, "time": 0.6003096103668213, "epoch": 10, "memory": 15585, "step": 19486} +{"accuracy/top1": 99.93389129638672, "data_time": 0.19146766662597656, "time": 0.5791983922322591, "step": 10} diff --git a/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_1e-1/20230603_172035/vis_data/vis_image/0000000.png_1.png b/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_1e-1/20230603_172035/vis_data/vis_image/0000000.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..a1c6c6127740e58e6931fc3dfb57c2325baed5e2 Binary files /dev/null and b/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_1e-1/20230603_172035/vis_data/vis_image/0000000.png_1.png differ diff --git a/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_1e-1/20230603_172035/vis_data/vis_image/0000000.png_10.png b/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_1e-1/20230603_172035/vis_data/vis_image/0000000.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..d53717b275cb99a8c54927283de9da415cb59664 Binary files /dev/null and b/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_1e-1/20230603_172035/vis_data/vis_image/0000000.png_10.png differ diff --git a/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_1e-1/20230603_172035/vis_data/vis_image/0000000.png_2.png b/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_1e-1/20230603_172035/vis_data/vis_image/0000000.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..1e7dbaa497b715e3d30768b793d0bcccbc825619 Binary files /dev/null and b/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_1e-1/20230603_172035/vis_data/vis_image/0000000.png_2.png differ diff --git a/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_1e-1/20230603_172035/vis_data/vis_image/0000000.png_3.png b/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_1e-1/20230603_172035/vis_data/vis_image/0000000.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..35d4e130f412a7416d4f5e191c85a944e58de2a6 Binary files /dev/null and b/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_1e-1/20230603_172035/vis_data/vis_image/0000000.png_3.png differ diff --git a/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_1e-1/20230603_172035/vis_data/vis_image/0000000.png_4.png b/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_1e-1/20230603_172035/vis_data/vis_image/0000000.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..40134fbc9e2be6a9d46531f2622e1f77683d0a8c Binary files /dev/null and b/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_1e-1/20230603_172035/vis_data/vis_image/0000000.png_4.png differ diff --git a/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_1e-1/20230603_172035/vis_data/vis_image/0000000.png_5.png b/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_1e-1/20230603_172035/vis_data/vis_image/0000000.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..695defc31de2f2fd8ae0d65ad37b2122d143e6b9 Binary files /dev/null and b/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_1e-1/20230603_172035/vis_data/vis_image/0000000.png_5.png differ diff --git a/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_1e-1/20230603_172035/vis_data/vis_image/0000000.png_6.png b/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_1e-1/20230603_172035/vis_data/vis_image/0000000.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..1ee94547b6a9ac2de768c8f0943da1c6bdff344b Binary files /dev/null and b/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_1e-1/20230603_172035/vis_data/vis_image/0000000.png_6.png differ diff --git a/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_1e-1/20230603_172035/vis_data/vis_image/0000000.png_7.png b/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_1e-1/20230603_172035/vis_data/vis_image/0000000.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..05d91623a0fdce7a1b9ae5b176f4cc7aef5c4272 Binary files /dev/null and b/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_1e-1/20230603_172035/vis_data/vis_image/0000000.png_7.png differ diff --git a/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_1e-1/20230603_172035/vis_data/vis_image/0000000.png_8.png b/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_1e-1/20230603_172035/vis_data/vis_image/0000000.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..3ea0e84ae46d33006056823db01374b5e1d0705f Binary files /dev/null and b/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_1e-1/20230603_172035/vis_data/vis_image/0000000.png_8.png differ diff --git a/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_1e-1/20230603_172035/vis_data/vis_image/0000000.png_9.png b/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_1e-1/20230603_172035/vis_data/vis_image/0000000.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..99579b9c5d979e8a4f3e5bfa70503277a11a1fea Binary files /dev/null and b/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_1e-1/20230603_172035/vis_data/vis_image/0000000.png_9.png differ diff --git a/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_1e-1/20230603_172035/vis_data/vis_image/000005174.jpg_1.png b/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_1e-1/20230603_172035/vis_data/vis_image/000005174.jpg_1.png new file mode 100644 index 0000000000000000000000000000000000000000..3122e0e692bf9ef8989225c047e3475d3f0f754a Binary files /dev/null and b/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_1e-1/20230603_172035/vis_data/vis_image/000005174.jpg_1.png differ diff --git a/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_1e-1/20230603_172035/vis_data/vis_image/000005174.jpg_10.png b/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_1e-1/20230603_172035/vis_data/vis_image/000005174.jpg_10.png new file mode 100644 index 0000000000000000000000000000000000000000..76f6e4cdcd6acdf6003a44c1447d85c94aa5e7c8 Binary files /dev/null and b/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_1e-1/20230603_172035/vis_data/vis_image/000005174.jpg_10.png differ diff --git a/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_1e-1/20230603_172035/vis_data/vis_image/000005174.jpg_2.png b/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_1e-1/20230603_172035/vis_data/vis_image/000005174.jpg_2.png new file mode 100644 index 0000000000000000000000000000000000000000..414bdb3c9e3b2134eee54d6da463450b3dbe701c Binary files /dev/null and b/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_1e-1/20230603_172035/vis_data/vis_image/000005174.jpg_2.png differ diff --git a/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_1e-1/20230603_172035/vis_data/vis_image/000005174.jpg_3.png b/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_1e-1/20230603_172035/vis_data/vis_image/000005174.jpg_3.png new file mode 100644 index 0000000000000000000000000000000000000000..f15af0263c498155ff5e27a55b9a8aa480910518 Binary files /dev/null and b/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_1e-1/20230603_172035/vis_data/vis_image/000005174.jpg_3.png differ diff --git a/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_1e-1/20230603_172035/vis_data/vis_image/000005174.jpg_4.png b/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_1e-1/20230603_172035/vis_data/vis_image/000005174.jpg_4.png new file mode 100644 index 0000000000000000000000000000000000000000..b23809ba5f4b07473a53ece4444ef9a47e16e51d Binary files /dev/null and b/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_1e-1/20230603_172035/vis_data/vis_image/000005174.jpg_4.png differ diff --git a/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_1e-1/20230603_172035/vis_data/vis_image/000005174.jpg_5.png b/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_1e-1/20230603_172035/vis_data/vis_image/000005174.jpg_5.png new file mode 100644 index 0000000000000000000000000000000000000000..ec9e2fb036ef05e42367ec3a688c3ed87a7c4989 Binary files /dev/null and b/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_1e-1/20230603_172035/vis_data/vis_image/000005174.jpg_5.png differ diff --git a/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_1e-1/20230603_172035/vis_data/vis_image/000005174.jpg_6.png b/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_1e-1/20230603_172035/vis_data/vis_image/000005174.jpg_6.png new file mode 100644 index 0000000000000000000000000000000000000000..cce18840ecf6342909683f104d89a1866c9bba43 Binary files /dev/null and b/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_1e-1/20230603_172035/vis_data/vis_image/000005174.jpg_6.png differ diff --git a/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_1e-1/20230603_172035/vis_data/vis_image/000005174.jpg_7.png b/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_1e-1/20230603_172035/vis_data/vis_image/000005174.jpg_7.png new file mode 100644 index 0000000000000000000000000000000000000000..dc334038163c4bc810af1b282e8d75503199a34c Binary files /dev/null and b/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_1e-1/20230603_172035/vis_data/vis_image/000005174.jpg_7.png differ diff --git a/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_1e-1/20230603_172035/vis_data/vis_image/000005174.jpg_8.png b/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_1e-1/20230603_172035/vis_data/vis_image/000005174.jpg_8.png new file mode 100644 index 0000000000000000000000000000000000000000..2177f510a6c0fcbab40fd3829040e1f850ec76da Binary files /dev/null and b/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_1e-1/20230603_172035/vis_data/vis_image/000005174.jpg_8.png differ diff --git a/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_1e-1/20230603_172035/vis_data/vis_image/000005174.jpg_9.png b/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_1e-1/20230603_172035/vis_data/vis_image/000005174.jpg_9.png new file mode 100644 index 0000000000000000000000000000000000000000..ec84015984cb9c8489b7f8fd0de69d19e76abab5 Binary files /dev/null and b/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_1e-1/20230603_172035/vis_data/vis_image/000005174.jpg_9.png differ diff --git a/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_1e-1/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_1e-1.py b/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_1e-1/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_1e-1.py new file mode 100644 index 0000000000000000000000000000000000000000..470c9006a677f36d1d5911b9299657969c1a2ae4 --- /dev/null +++ b/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_1e-1/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_1e-1.py @@ -0,0 +1,247 @@ +optim_wrapper = dict( + optimizer=dict( + type='AdamW', + lr=0.0004, + weight_decay=0.05, + eps=1e-08, + betas=(0.9, 0.999), + _scope_='mmpretrain'), + paramwise_cfg=dict( + norm_decay_mult=0.0, + bias_decay_mult=0.0, + flat_decay_mult=0.0, + custom_keys=dict({ + '.absolute_pos_embed': dict(decay_mult=0.0), + '.relative_position_bias_table': dict(decay_mult=0.0) + })), + type='AmpOptimWrapper', + dtype='bfloat16', + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=False, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=4096) +model = dict( + type='ImageClassifier', + backbone=dict( + frozen_stages=2, + type='ConvNeXt', + arch='small', + drop_path_rate=0.4, + init_cfg=dict( + type='Pretrained', + checkpoint= + 'https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth', + prefix='backbone')), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=768, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + init_cfg=None), + init_cfg=dict( + type='TruncNormal', layer=['Conv2d', 'Linear'], std=0.02, bias=0.0), + train_cfg=None) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stablediffusionV2-1-dpmsolver-25-1m.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/cc1m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV2-1-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV2-1-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_1e-1' diff --git a/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_1e-1/epoch_1.pth b/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_1e-1/epoch_1.pth new file mode 100644 index 0000000000000000000000000000000000000000..293f72bcbc035864dbaa70120aa4ded81fc42eee --- /dev/null +++ b/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_1e-1/epoch_1.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:12082b00bd67e08da192fb89f61ea666ea59126eda2f7f570335823dae367839 +size 782109786 diff --git a/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_1e-1/epoch_10.pth b/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_1e-1/epoch_10.pth new file mode 100644 index 0000000000000000000000000000000000000000..0d0395f11195999bd9f64f9f97b25aac7dbfe549 --- /dev/null +++ b/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_1e-1/epoch_10.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:4b98f816e1a5ee32bb7af64809ddd320f49b77e18c0052c5b59806367488b99f +size 783383962 diff --git a/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_1e-1/epoch_2.pth b/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_1e-1/epoch_2.pth new file mode 100644 index 0000000000000000000000000000000000000000..8ab7047034e80504ccb3c3c4d18d0f24224e4588 --- /dev/null +++ b/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_1e-1/epoch_2.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:b04064fb8c27398183756794a5c77221bcb45bfad6f5a8b14cd59e9eee3ea7ef +size 782252634 diff --git a/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_1e-1/epoch_3.pth b/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_1e-1/epoch_3.pth new file mode 100644 index 0000000000000000000000000000000000000000..ce3b0593a8b73134097c55c09bca124e902669b8 --- /dev/null +++ b/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_1e-1/epoch_3.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:7bcec0f652a6f0cd78177264b10bde4b069641982e692b8cb4073072d0bc1862 +size 782394138 diff --git a/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_1e-1/epoch_4.pth b/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_1e-1/epoch_4.pth new file mode 100644 index 0000000000000000000000000000000000000000..c965979531fa710642c19ea16ee96cc23f9238eb --- /dev/null +++ b/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_1e-1/epoch_4.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:1293c1de9975cfb2aadf10d926dee519eb3c82515b945c2cb3433128bbd95eb0 +size 782535642 diff --git a/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_1e-1/epoch_5.pth b/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_1e-1/epoch_5.pth new file mode 100644 index 0000000000000000000000000000000000000000..6f1636e3fc506d0ed95312bfb260f01735c347f0 --- /dev/null +++ b/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_1e-1/epoch_5.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:360eabc591457513b8e4c9799bd241e0b7fb7c4931e0fa4a89cc0b138e66526b +size 782676698 diff --git a/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_1e-1/epoch_6.pth b/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_1e-1/epoch_6.pth new file mode 100644 index 0000000000000000000000000000000000000000..78e6a1c793c99414e314066a3373725232f12054 --- /dev/null +++ b/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_1e-1/epoch_6.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:37e1b554d0faa5577e39ec15cf84d2b0dc4b07159be4658d8f17deaf40554384 +size 782817562 diff --git a/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_1e-1/epoch_7.pth b/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_1e-1/epoch_7.pth new file mode 100644 index 0000000000000000000000000000000000000000..1684a0b939a6bafb20fe12eb434a9e9cb7e4c248 --- /dev/null +++ b/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_1e-1/epoch_7.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:5279e666edffd6843ab83404baca690fc1aa3d3ab66d57930abb9ac4e3b16fe7 +size 782958490 diff --git a/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_1e-1/epoch_8.pth b/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_1e-1/epoch_8.pth new file mode 100644 index 0000000000000000000000000000000000000000..2f7f03c7201eef24917b545f424d41e0231bdef3 --- /dev/null +++ b/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_1e-1/epoch_8.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:491f8fe44d018d23febbf4d8c7873d4e190d397dbb6c7df9cc082296b0fe5f5f +size 783099418 diff --git a/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_1e-1/epoch_9.pth b/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_1e-1/epoch_9.pth new file mode 100644 index 0000000000000000000000000000000000000000..6807c8a0e8a596c9c0e23a26184374be0419423b --- /dev/null +++ b/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_1e-1/epoch_9.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c7335e9c20879342c6a830c1d37fc854d6097112d234d688a3d50405890e04b0 +size 783241114 diff --git a/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_1e-1/last_checkpoint b/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_1e-1/last_checkpoint new file mode 100644 index 0000000000000000000000000000000000000000..648c2500c1b729801fecaef5fcef709b88a3c812 --- /dev/null +++ b/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_1e-1/last_checkpoint @@ -0,0 +1 @@ +/mnt/petrelfs/luzeyu/workspace/fakebench/mmpretrain/workdir/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_1e-1/epoch_10.pth \ No newline at end of file diff --git a/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_5e-1/20230603_194605/20230603_194605.log b/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_5e-1/20230603_194605/20230603_194605.log new file mode 100644 index 0000000000000000000000000000000000000000..a627c535579900c64739299a6f76a00cb0094b4c --- /dev/null +++ b/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_5e-1/20230603_194605/20230603_194605.log @@ -0,0 +1,1840 @@ +2023/06/03 19:46:10 - mmengine - INFO - +------------------------------------------------------------ +System environment: + sys.platform: linux + Python: 3.10.9 (main, Mar 8 2023, 10:47:38) [GCC 11.2.0] + CUDA available: True + numpy_random_seed: 138418324 + GPU 0,1,2,3: NVIDIA A100-SXM4-80GB + CUDA_HOME: /mnt/petrelfs/share/cuda-11.6 + NVCC: Cuda compilation tools, release 11.6, V11.6.124 + GCC: gcc (GCC) 7.5.0 + PyTorch: 1.13.1 + PyTorch compiling details: PyTorch built with: + - GCC 9.3 + - C++ Version: 201402 + - Intel(R) oneAPI Math Kernel Library Version 2021.4-Product Build 20210904 for Intel(R) 64 architecture applications + - Intel(R) MKL-DNN v2.6.0 (Git Hash 52b5f107dd9cf10910aaa19cb47f3abf9b349815) + - OpenMP 201511 (a.k.a. OpenMP 4.5) + - LAPACK is enabled (usually provided by MKL) + - NNPACK is enabled + - CPU capability usage: AVX2 + - CUDA Runtime 11.6 + - NVCC architecture flags: -gencode;arch=compute_37,code=sm_37;-gencode;arch=compute_50,code=sm_50;-gencode;arch=compute_60,code=sm_60;-gencode;arch=compute_61,code=sm_61;-gencode;arch=compute_70,code=sm_70;-gencode;arch=compute_75,code=sm_75;-gencode;arch=compute_80,code=sm_80;-gencode;arch=compute_86,code=sm_86;-gencode;arch=compute_37,code=compute_37 + - CuDNN 8.3.2 (built against CUDA 11.5) + - Magma 2.6.1 + - Build settings: BLAS_INFO=mkl, BUILD_TYPE=Release, CUDA_VERSION=11.6, CUDNN_VERSION=8.3.2, CXX_COMPILER=/opt/rh/devtoolset-9/root/usr/bin/c++, CXX_FLAGS= -fabi-version=11 -Wno-deprecated -fvisibility-inlines-hidden -DUSE_PTHREADPOOL -fopenmp -DNDEBUG -DUSE_KINETO -DUSE_FBGEMM -DUSE_QNNPACK -DUSE_PYTORCH_QNNPACK -DUSE_XNNPACK -DSYMBOLICATE_MOBILE_DEBUG_HANDLE -DEDGE_PROFILER_USE_KINETO -O2 -fPIC -Wno-narrowing -Wall -Wextra -Werror=return-type -Werror=non-virtual-dtor -Wno-missing-field-initializers -Wno-type-limits -Wno-array-bounds -Wno-unknown-pragmas -Wunused-local-typedefs -Wno-unused-parameter -Wno-unused-function -Wno-unused-result -Wno-strict-overflow -Wno-strict-aliasing -Wno-error=deprecated-declarations -Wno-stringop-overflow -Wno-psabi -Wno-error=pedantic -Wno-error=redundant-decls -Wno-error=old-style-cast -fdiagnostics-color=always -faligned-new -Wno-unused-but-set-variable -Wno-maybe-uninitialized -fno-math-errno -fno-trapping-math -Werror=format -Werror=cast-function-type -Wno-stringop-overflow, LAPACK_INFO=mkl, PERF_WITH_AVX=1, PERF_WITH_AVX2=1, PERF_WITH_AVX512=1, TORCH_VERSION=1.13.1, USE_CUDA=ON, USE_CUDNN=ON, USE_EXCEPTION_PTR=1, USE_GFLAGS=OFF, USE_GLOG=OFF, USE_MKL=ON, USE_MKLDNN=ON, USE_MPI=OFF, USE_NCCL=ON, USE_NNPACK=ON, USE_OPENMP=ON, USE_ROCM=OFF, + + TorchVision: 0.14.1 + OpenCV: 4.7.0 + MMEngine: 0.7.3 + +Runtime environment: + cudnn_benchmark: True + mp_cfg: {'mp_start_method': 'fork', 'opencv_num_threads': 0} + dist_cfg: {'backend': 'nccl'} + seed: None + deterministic: False + Distributed launcher: slurm + Distributed training: True + GPU number: 4 +------------------------------------------------------------ + +2023/06/03 19:46:14 - mmengine - INFO - Config: +optim_wrapper = dict( + optimizer=dict( + type='AdamW', + lr=0.0004, + weight_decay=0.05, + eps=1e-08, + betas=(0.9, 0.999), + _scope_='mmpretrain'), + paramwise_cfg=dict( + norm_decay_mult=0.0, + bias_decay_mult=0.0, + flat_decay_mult=0.0, + custom_keys=dict({ + '.absolute_pos_embed': dict(decay_mult=0.0), + '.relative_position_bias_table': dict(decay_mult=0.0) + })), + type='AmpOptimWrapper', + dtype='bfloat16', + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=False, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=4096) +model = dict( + type='ImageClassifier', + backbone=dict( + frozen_stages=2, + type='ConvNeXt', + arch='small', + drop_path_rate=0.4, + init_cfg=dict( + type='Pretrained', + checkpoint= + 'https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth', + prefix='backbone')), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=768, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + init_cfg=None), + init_cfg=dict( + type='TruncNormal', layer=['Conv2d', 'Linear'], std=0.02, bias=0.0), + train_cfg=None) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stablediffusionV2-1-dpmsolver-25-1m.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/cc1m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV2-1-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV2-1-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_5e-1' + +2023/06/03 19:46:26 - mmengine - INFO - Hooks will be executed in the following order: +before_run: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_load_checkpoint: +(ABOVE_NORMAL) EMAHook + -------------------- +before_train: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_train_epoch: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(NORMAL ) DistSamplerSeedHook + -------------------- +before_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook + -------------------- +after_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +after_train_epoch: +(NORMAL ) IterTimerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_val_epoch: +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook + -------------------- +before_val_iter: +(NORMAL ) IterTimerHook + -------------------- +after_val_iter: +(NORMAL ) IterTimerHook +(NORMAL ) VisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_val_epoch: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_save_checkpoint: +(ABOVE_NORMAL) EMAHook + -------------------- +after_train: +(VERY_LOW ) CheckpointHook + -------------------- +before_test_epoch: +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook + -------------------- +before_test_iter: +(NORMAL ) IterTimerHook + -------------------- +after_test_iter: +(NORMAL ) IterTimerHook +(NORMAL ) VisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_test_epoch: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_run: +(BELOW_NORMAL) LoggerHook + -------------------- +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.2.0.weight:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.2.0.bias:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.2.1.bias:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.3.0.weight:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.3.0.bias:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.3.1.bias:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.2.0.gamma:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.2.0.depthwise_conv.bias:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.2.0.norm.weight:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.2.0.norm.bias:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.2.0.pointwise_conv1.bias:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.2.0.pointwise_conv2.bias:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.2.1.gamma:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.2.1.depthwise_conv.bias:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.2.1.norm.weight:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.2.1.norm.bias:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.2.1.pointwise_conv1.bias:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.2.1.pointwise_conv2.bias:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.2.2.gamma:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.2.2.depthwise_conv.bias:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.2.2.norm.weight:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.2.2.norm.bias:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.2.2.pointwise_conv1.bias:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.2.2.pointwise_conv2.bias:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.2.3.gamma:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.2.3.depthwise_conv.bias:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.2.3.norm.weight:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.2.3.norm.bias:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.2.3.pointwise_conv1.bias:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.2.3.pointwise_conv2.bias:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.2.4.gamma:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.2.4.depthwise_conv.bias:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.2.4.norm.weight:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.2.4.norm.bias:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.2.4.pointwise_conv1.bias:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.2.4.pointwise_conv2.bias:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.2.5.gamma:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.2.5.depthwise_conv.bias:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.2.5.norm.weight:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.2.5.norm.bias:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.2.5.pointwise_conv1.bias:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.2.5.pointwise_conv2.bias:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.2.6.gamma:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.2.6.depthwise_conv.bias:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.2.6.norm.weight:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.2.6.norm.bias:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.2.6.pointwise_conv1.bias:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.2.6.pointwise_conv2.bias:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.2.7.gamma:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.2.7.depthwise_conv.bias:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.2.7.norm.weight:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.2.7.norm.bias:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.2.7.pointwise_conv1.bias:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.2.7.pointwise_conv2.bias:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.2.8.gamma:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.2.8.depthwise_conv.bias:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.2.8.norm.weight:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.2.8.norm.bias:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.2.8.pointwise_conv1.bias:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.2.8.pointwise_conv2.bias:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.2.9.gamma:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.2.9.depthwise_conv.bias:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.2.9.norm.weight:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.2.9.norm.bias:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.2.9.pointwise_conv1.bias:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.2.9.pointwise_conv2.bias:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.2.10.gamma:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.2.10.depthwise_conv.bias:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.2.10.norm.weight:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.2.10.norm.bias:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.2.10.pointwise_conv1.bias:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.2.10.pointwise_conv2.bias:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.2.11.gamma:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.2.11.depthwise_conv.bias:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.2.11.norm.weight:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.2.11.norm.bias:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.2.11.pointwise_conv1.bias:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.2.11.pointwise_conv2.bias:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.2.12.gamma:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.2.12.depthwise_conv.bias:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.2.12.norm.weight:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.2.12.norm.bias:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.2.12.pointwise_conv1.bias:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.2.12.pointwise_conv2.bias:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.2.13.gamma:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.2.13.depthwise_conv.bias:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.2.13.norm.weight:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.2.13.norm.bias:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.2.13.pointwise_conv1.bias:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.2.13.pointwise_conv2.bias:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.2.14.gamma:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.2.14.depthwise_conv.bias:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.2.14.norm.weight:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.2.14.norm.bias:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.2.14.pointwise_conv1.bias:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.2.14.pointwise_conv2.bias:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.2.15.gamma:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.2.15.depthwise_conv.bias:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.2.15.norm.weight:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.2.15.norm.bias:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.2.15.pointwise_conv1.bias:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.2.15.pointwise_conv2.bias:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.2.16.gamma:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.2.16.depthwise_conv.bias:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.2.16.norm.weight:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.2.16.norm.bias:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.2.16.pointwise_conv1.bias:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.2.16.pointwise_conv2.bias:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.2.17.gamma:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.2.17.depthwise_conv.bias:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.2.17.norm.weight:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.2.17.norm.bias:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.2.17.pointwise_conv1.bias:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.2.17.pointwise_conv2.bias:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.2.18.gamma:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.2.18.depthwise_conv.bias:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.2.18.norm.weight:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.2.18.norm.bias:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.2.18.pointwise_conv1.bias:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.2.18.pointwise_conv2.bias:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.2.19.gamma:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.2.19.depthwise_conv.bias:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.2.19.norm.weight:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.2.19.norm.bias:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.2.19.pointwise_conv1.bias:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.2.19.pointwise_conv2.bias:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.2.20.gamma:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.2.20.depthwise_conv.bias:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.2.20.norm.weight:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.2.20.norm.bias:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.2.20.pointwise_conv1.bias:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.2.20.pointwise_conv2.bias:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.2.21.gamma:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.2.21.depthwise_conv.bias:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.2.21.norm.weight:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.2.21.norm.bias:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.2.21.pointwise_conv1.bias:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.2.21.pointwise_conv2.bias:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.2.22.gamma:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.2.22.depthwise_conv.bias:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.2.22.norm.weight:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.2.22.norm.bias:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.2.22.pointwise_conv1.bias:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.2.22.pointwise_conv2.bias:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.2.23.gamma:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.2.23.depthwise_conv.bias:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.2.23.norm.weight:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.2.23.norm.bias:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.2.23.pointwise_conv1.bias:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.2.23.pointwise_conv2.bias:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.2.24.gamma:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.2.24.depthwise_conv.bias:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.2.24.norm.weight:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.2.24.norm.bias:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.2.24.pointwise_conv1.bias:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.2.24.pointwise_conv2.bias:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.2.25.gamma:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.2.25.depthwise_conv.bias:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.2.25.norm.weight:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.2.25.norm.bias:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.2.25.pointwise_conv1.bias:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.2.25.pointwise_conv2.bias:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.2.26.gamma:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.2.26.depthwise_conv.bias:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.2.26.norm.weight:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.2.26.norm.bias:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.2.26.pointwise_conv1.bias:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.2.26.pointwise_conv2.bias:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.3.0.gamma:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.3.0.depthwise_conv.bias:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.3.0.norm.weight:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.3.0.norm.bias:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.3.0.pointwise_conv1.bias:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.3.0.pointwise_conv2.bias:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.3.1.gamma:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.3.1.depthwise_conv.bias:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.3.1.norm.weight:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.3.1.norm.bias:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.3.1.pointwise_conv1.bias:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.3.1.pointwise_conv2.bias:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.3.2.gamma:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.3.2.depthwise_conv.bias:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.3.2.norm.weight:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.3.2.norm.bias:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.3.2.pointwise_conv1.bias:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.stages.3.2.pointwise_conv2.bias:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.norm3.weight:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- backbone.norm3.bias:weight_decay=0.0 +2023/06/03 19:46:40 - mmengine - INFO - paramwise_options -- head.fc.bias:weight_decay=0.0 +2023/06/03 19:46:45 - mmengine - INFO - load backbone in model from: https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth +Name of parameter - Initialization information + +backbone.downsample_layers.0.0.weight - torch.Size([96, 3, 4, 4]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.0.0.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.0.1.weight - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.0.1.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.1.0.weight - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.1.0.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.1.1.weight - torch.Size([192, 96, 2, 2]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.1.1.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.2.0.weight - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.2.0.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.2.1.weight - torch.Size([384, 192, 2, 2]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.2.1.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.3.0.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.3.0.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.3.1.weight - torch.Size([768, 384, 2, 2]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.3.1.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.0.gamma - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.0.depthwise_conv.weight - torch.Size([96, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.0.depthwise_conv.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.0.norm.weight - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.0.norm.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.0.pointwise_conv1.weight - torch.Size([384, 96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.0.pointwise_conv1.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.0.pointwise_conv2.weight - torch.Size([96, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.0.pointwise_conv2.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.1.gamma - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.1.depthwise_conv.weight - torch.Size([96, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.1.depthwise_conv.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.1.norm.weight - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.1.norm.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.1.pointwise_conv1.weight - torch.Size([384, 96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.1.pointwise_conv1.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.1.pointwise_conv2.weight - torch.Size([96, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.1.pointwise_conv2.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.2.gamma - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.2.depthwise_conv.weight - torch.Size([96, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.2.depthwise_conv.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.2.norm.weight - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.2.norm.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.2.pointwise_conv1.weight - torch.Size([384, 96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.2.pointwise_conv1.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.2.pointwise_conv2.weight - torch.Size([96, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.2.pointwise_conv2.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.0.gamma - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.0.depthwise_conv.weight - torch.Size([192, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.0.depthwise_conv.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.0.norm.weight - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.0.norm.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.0.pointwise_conv1.weight - torch.Size([768, 192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.0.pointwise_conv1.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.0.pointwise_conv2.weight - torch.Size([192, 768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.0.pointwise_conv2.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.1.gamma - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.1.depthwise_conv.weight - torch.Size([192, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.1.depthwise_conv.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.1.norm.weight - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.1.norm.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.1.pointwise_conv1.weight - torch.Size([768, 192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.1.pointwise_conv1.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.1.pointwise_conv2.weight - torch.Size([192, 768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.1.pointwise_conv2.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.2.gamma - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.2.depthwise_conv.weight - torch.Size([192, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.2.depthwise_conv.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.2.norm.weight - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.2.norm.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.2.pointwise_conv1.weight - torch.Size([768, 192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.2.pointwise_conv1.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.2.pointwise_conv2.weight - torch.Size([192, 768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.2.pointwise_conv2.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.0.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.0.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.0.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.0.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.0.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.0.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.0.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.0.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.0.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.1.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.1.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.1.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.1.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.1.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.1.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.1.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.1.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.1.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.2.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.2.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.2.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.2.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.2.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.2.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.2.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.2.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.2.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.3.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.3.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.3.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.3.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.3.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.3.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.3.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.3.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.3.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.4.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.4.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.4.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.4.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.4.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.4.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.4.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.4.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.4.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.5.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.5.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.5.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.5.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.5.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.5.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.5.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.5.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.5.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.6.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.6.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.6.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.6.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.6.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.6.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.6.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.6.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.6.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.7.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.7.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.7.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.7.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.7.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.7.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.7.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.7.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.7.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.8.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.8.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.8.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.8.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.8.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.8.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.8.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.8.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.8.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.9.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.9.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.9.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.9.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.9.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.9.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.9.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.9.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.9.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.10.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.10.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.10.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.10.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.10.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.10.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.10.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.10.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.10.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.11.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.11.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.11.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.11.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.11.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.11.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.11.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.11.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.11.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.12.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.12.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.12.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.12.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.12.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.12.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.12.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.12.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.12.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.13.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.13.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.13.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.13.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.13.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.13.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.13.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.13.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.13.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.14.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.14.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.14.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.14.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.14.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.14.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.14.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.14.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.14.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.15.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.15.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.15.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.15.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.15.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.15.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.15.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.15.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.15.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.16.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.16.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.16.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.16.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.16.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.16.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.16.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.16.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.16.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.17.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.17.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.17.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.17.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.17.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.17.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.17.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.17.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.17.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.18.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.18.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.18.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.18.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.18.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.18.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.18.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.18.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.18.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.19.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.19.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.19.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.19.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.19.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.19.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.19.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.19.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.19.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.20.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.20.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.20.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.20.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.20.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.20.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.20.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.20.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.20.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.21.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.21.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.21.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.21.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.21.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.21.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.21.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.21.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.21.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.22.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.22.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.22.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.22.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.22.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.22.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.22.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.22.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.22.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.23.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.23.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.23.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.23.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.23.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.23.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.23.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.23.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.23.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.24.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.24.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.24.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.24.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.24.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.24.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.24.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.24.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.24.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.25.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.25.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.25.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.25.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.25.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.25.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.25.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.25.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.25.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.26.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.26.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.26.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.26.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.26.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.26.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.26.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.26.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.26.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.0.gamma - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.0.depthwise_conv.weight - torch.Size([768, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.0.depthwise_conv.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.0.norm.weight - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.0.norm.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.0.pointwise_conv1.weight - torch.Size([3072, 768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.0.pointwise_conv1.bias - torch.Size([3072]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.0.pointwise_conv2.weight - torch.Size([768, 3072]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.0.pointwise_conv2.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.1.gamma - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.1.depthwise_conv.weight - torch.Size([768, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.1.depthwise_conv.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.1.norm.weight - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.1.norm.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.1.pointwise_conv1.weight - torch.Size([3072, 768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.1.pointwise_conv1.bias - torch.Size([3072]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.1.pointwise_conv2.weight - torch.Size([768, 3072]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.1.pointwise_conv2.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.2.gamma - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.2.depthwise_conv.weight - torch.Size([768, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.2.depthwise_conv.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.2.norm.weight - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.2.norm.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.2.pointwise_conv1.weight - torch.Size([3072, 768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.2.pointwise_conv1.bias - torch.Size([3072]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.2.pointwise_conv2.weight - torch.Size([768, 3072]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.2.pointwise_conv2.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.norm3.weight - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.norm3.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +head.fc.weight - torch.Size([2, 768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +head.fc.bias - torch.Size([2]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 +2023/06/03 19:46:46 - mmengine - WARNING - "FileClient" will be deprecated in future. Please use io functions in https://mmengine.readthedocs.io/en/latest/api/fileio.html#file-io +2023/06/03 19:46:46 - mmengine - WARNING - "HardDiskBackend" is the alias of "LocalBackend" and the former will be deprecated in future. +2023/06/03 19:46:46 - mmengine - INFO - Checkpoints will be saved to /mnt/petrelfs/luzeyu/workspace/fakebench/mmpretrain/workdir/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_5e-1. +2023/06/03 19:48:07 - mmengine - INFO - Epoch(train) [1][ 100/1954] lr: 3.9998e-04 eta: 4:23:10 time: 0.8002 data_time: 0.0017 memory: 15587 loss: 0.0512 +2023/06/03 19:49:15 - mmengine - INFO - Epoch(train) [1][ 200/1954] lr: 3.9990e-04 eta: 3:59:48 time: 0.6619 data_time: 0.0017 memory: 15587 loss: 0.0415 +2023/06/03 19:50:23 - mmengine - INFO - Epoch(train) [1][ 300/1954] lr: 3.9977e-04 eta: 3:51:44 time: 0.6557 data_time: 0.0018 memory: 15587 loss: 0.0320 +2023/06/03 19:51:33 - mmengine - INFO - Epoch(train) [1][ 400/1954] lr: 3.9960e-04 eta: 3:48:49 time: 0.7310 data_time: 0.0020 memory: 15587 loss: 0.0243 +2023/06/03 19:52:44 - mmengine - INFO - Epoch(train) [1][ 500/1954] lr: 3.9937e-04 eta: 3:47:07 time: 0.7297 data_time: 0.0020 memory: 15587 loss: 0.0398 +2023/06/03 19:53:50 - mmengine - INFO - Epoch(train) [1][ 600/1954] lr: 3.9910e-04 eta: 3:43:14 time: 0.6628 data_time: 0.0018 memory: 15587 loss: 0.0253 +2023/06/03 19:55:01 - mmengine - INFO - Epoch(train) [1][ 700/1954] lr: 3.9877e-04 eta: 3:42:03 time: 0.9731 data_time: 0.0018 memory: 15587 loss: 0.0245 +2023/06/03 19:56:16 - mmengine - INFO - Epoch(train) [1][ 800/1954] lr: 3.9839e-04 eta: 3:42:42 time: 0.7067 data_time: 0.0025 memory: 15587 loss: 0.0287 +2023/06/03 19:57:26 - mmengine - INFO - Epoch(train) [1][ 900/1954] lr: 3.9797e-04 eta: 3:41:02 time: 0.6873 data_time: 0.0018 memory: 15587 loss: 0.0256 +2023/06/03 19:58:36 - mmengine - INFO - Exp name: convnext_small_4xb256_sdv2_1m_lr4e-4_aug_5e-1_20230603_194605 +2023/06/03 19:58:36 - mmengine - INFO - Epoch(train) [1][1000/1954] lr: 3.9749e-04 eta: 3:39:26 time: 0.6787 data_time: 0.0017 memory: 15587 loss: 0.0163 +2023/06/03 19:59:48 - mmengine - INFO - Epoch(train) [1][1100/1954] lr: 3.9696e-04 eta: 3:38:25 time: 0.7679 data_time: 0.0017 memory: 15587 loss: 0.0216 +2023/06/03 20:00:55 - mmengine - INFO - Epoch(train) [1][1200/1954] lr: 3.9639e-04 eta: 3:36:13 time: 0.6652 data_time: 0.0016 memory: 15587 loss: 0.0237 +2023/06/03 20:02:01 - mmengine - INFO - Epoch(train) [1][1300/1954] lr: 3.9576e-04 eta: 3:34:03 time: 0.7070 data_time: 0.0017 memory: 15587 loss: 0.0228 +2023/06/03 20:03:11 - mmengine - INFO - Epoch(train) [1][1400/1954] lr: 3.9509e-04 eta: 3:32:47 time: 0.6563 data_time: 0.0017 memory: 15587 loss: 0.0219 +2023/06/03 20:04:18 - mmengine - INFO - Epoch(train) [1][1500/1954] lr: 3.9436e-04 eta: 3:30:59 time: 0.6750 data_time: 0.0017 memory: 15587 loss: 0.0195 +2023/06/03 20:05:25 - mmengine - INFO - Epoch(train) [1][1600/1954] lr: 3.9359e-04 eta: 3:29:09 time: 0.6075 data_time: 0.0018 memory: 15587 loss: 0.0215 +2023/06/03 20:06:36 - mmengine - INFO - Epoch(train) [1][1700/1954] lr: 3.9277e-04 eta: 3:28:11 time: 0.6816 data_time: 0.0017 memory: 15587 loss: 0.0171 +2023/06/03 20:07:44 - mmengine - INFO - Epoch(train) [1][1800/1954] lr: 3.9190e-04 eta: 3:26:35 time: 0.6630 data_time: 0.0016 memory: 15587 loss: 0.0199 +2023/06/03 20:08:53 - mmengine - INFO - Epoch(train) [1][1900/1954] lr: 3.9098e-04 eta: 3:25:17 time: 0.6733 data_time: 0.0018 memory: 15587 loss: 0.0212 +2023/06/03 20:09:29 - mmengine - INFO - Exp name: convnext_small_4xb256_sdv2_1m_lr4e-4_aug_5e-1_20230603_194605 +2023/06/03 20:09:29 - mmengine - INFO - Saving checkpoint at 1 epochs +2023/06/03 20:10:09 - mmengine - INFO - Epoch(val) [1][29/29] accuracy/top1: 82.0709 single-label/precision_classwise: [76.77891540527344, 94.09220123291016] single-label/recall_classwise: [96.72369384765625, 64.07752227783203] single-label/f1-score_classwise: [85.60494232177734, 76.2370376586914] data_time: 0.3283 time: 0.7907 +2023/06/03 20:10:47 - mmengine - INFO - Exp name: convnext_small_4xb256_sdv2_1m_lr4e-4_aug_5e-1_20230603_194605 +2023/06/03 20:11:25 - mmengine - INFO - Epoch(train) [2][ 100/1954] lr: 3.8947e-04 eta: 3:24:09 time: 0.7513 data_time: 0.0017 memory: 15587 loss: 0.0120 +2023/06/03 20:12:36 - mmengine - INFO - Epoch(train) [2][ 200/1954] lr: 3.8843e-04 eta: 3:23:08 time: 0.6740 data_time: 0.0016 memory: 15585 loss: 0.0182 +2023/06/03 20:13:45 - mmengine - INFO - Epoch(train) [2][ 300/1954] lr: 3.8735e-04 eta: 3:21:48 time: 0.6747 data_time: 0.0019 memory: 15585 loss: 0.0229 +2023/06/03 20:14:58 - mmengine - INFO - Epoch(train) [2][ 400/1954] lr: 3.8621e-04 eta: 3:21:00 time: 0.6789 data_time: 0.0015 memory: 15585 loss: 0.0099 +2023/06/03 20:16:09 - mmengine - INFO - Epoch(train) [2][ 500/1954] lr: 3.8503e-04 eta: 3:19:55 time: 0.6590 data_time: 0.0017 memory: 15585 loss: 0.0129 +2023/06/03 20:17:20 - mmengine - INFO - Epoch(train) [2][ 600/1954] lr: 3.8380e-04 eta: 3:18:48 time: 0.6659 data_time: 0.0016 memory: 15585 loss: 0.0187 +2023/06/03 20:18:26 - mmengine - INFO - Epoch(train) [2][ 700/1954] lr: 3.8253e-04 eta: 3:17:12 time: 0.6611 data_time: 0.0014 memory: 15585 loss: 0.0124 +2023/06/03 20:19:38 - mmengine - INFO - Epoch(train) [2][ 800/1954] lr: 3.8121e-04 eta: 3:16:18 time: 0.7754 data_time: 0.0016 memory: 15585 loss: 0.0253 +2023/06/03 20:20:48 - mmengine - INFO - Epoch(train) [2][ 900/1954] lr: 3.7984e-04 eta: 3:15:02 time: 0.6869 data_time: 0.0014 memory: 15585 loss: 0.0143 +2023/06/03 20:21:55 - mmengine - INFO - Epoch(train) [2][1000/1954] lr: 3.7843e-04 eta: 3:13:34 time: 0.6961 data_time: 0.0015 memory: 15585 loss: 0.0129 +2023/06/03 20:22:27 - mmengine - INFO - Exp name: convnext_small_4xb256_sdv2_1m_lr4e-4_aug_5e-1_20230603_194605 +2023/06/03 20:23:05 - mmengine - INFO - Epoch(train) [2][1100/1954] lr: 3.7698e-04 eta: 3:12:26 time: 0.7613 data_time: 0.0018 memory: 15585 loss: 0.0165 +2023/06/03 20:24:13 - mmengine - INFO - Epoch(train) [2][1200/1954] lr: 3.7548e-04 eta: 3:11:07 time: 0.6289 data_time: 0.0018 memory: 15585 loss: 0.0078 +2023/06/03 20:25:22 - mmengine - INFO - Epoch(train) [2][1300/1954] lr: 3.7393e-04 eta: 3:09:50 time: 0.6919 data_time: 0.0016 memory: 15585 loss: 0.0143 +2023/06/03 20:26:31 - mmengine - INFO - Epoch(train) [2][1400/1954] lr: 3.7234e-04 eta: 3:08:34 time: 0.7064 data_time: 0.0015 memory: 15585 loss: 0.0172 +2023/06/03 20:27:37 - mmengine - INFO - Epoch(train) [2][1500/1954] lr: 3.7071e-04 eta: 3:07:08 time: 0.6244 data_time: 0.0019 memory: 15585 loss: 0.0117 +2023/06/03 20:28:41 - mmengine - INFO - Epoch(train) [2][1600/1954] lr: 3.6904e-04 eta: 3:05:32 time: 0.6623 data_time: 0.0018 memory: 15585 loss: 0.0113 +2023/06/03 20:29:48 - mmengine - INFO - Epoch(train) [2][1700/1954] lr: 3.6732e-04 eta: 3:04:11 time: 0.6861 data_time: 0.0015 memory: 15585 loss: 0.0141 +2023/06/03 20:30:55 - mmengine - INFO - Epoch(train) [2][1800/1954] lr: 3.6557e-04 eta: 3:02:49 time: 0.6885 data_time: 0.0016 memory: 15585 loss: 0.0199 +2023/06/03 20:32:02 - mmengine - INFO - Epoch(train) [2][1900/1954] lr: 3.6377e-04 eta: 3:01:31 time: 0.6978 data_time: 0.0016 memory: 15585 loss: 0.0159 +2023/06/03 20:32:38 - mmengine - INFO - Exp name: convnext_small_4xb256_sdv2_1m_lr4e-4_aug_5e-1_20230603_194605 +2023/06/03 20:32:38 - mmengine - INFO - Saving checkpoint at 2 epochs +2023/06/03 20:33:15 - mmengine - INFO - Epoch(val) [2][29/29] accuracy/top1: 92.2828 single-label/precision_classwise: [88.17463684082031, 99.00899505615234] single-label/recall_classwise: [99.31822204589844, 83.6434097290039] single-label/f1-score_classwise: [93.41527557373047, 90.67989349365234] data_time: 0.2989 time: 0.6843 +2023/06/03 20:34:25 - mmengine - INFO - Exp name: convnext_small_4xb256_sdv2_1m_lr4e-4_aug_5e-1_20230603_194605 +2023/06/03 20:34:29 - mmengine - INFO - Epoch(train) [3][ 100/1954] lr: 3.6091e-04 eta: 2:59:59 time: 0.6664 data_time: 0.0019 memory: 15585 loss: 0.0174 +2023/06/03 20:35:35 - mmengine - INFO - Epoch(train) [3][ 200/1954] lr: 3.5901e-04 eta: 2:58:35 time: 0.6456 data_time: 0.0016 memory: 15585 loss: 0.0210 +2023/06/03 20:36:44 - mmengine - INFO - Epoch(train) [3][ 300/1954] lr: 3.5707e-04 eta: 2:57:24 time: 0.6946 data_time: 0.0016 memory: 15585 loss: 0.0089 +2023/06/03 20:37:54 - mmengine - INFO - Epoch(train) [3][ 400/1954] lr: 3.5509e-04 eta: 2:56:16 time: 0.7003 data_time: 0.0016 memory: 15585 loss: 0.0138 +2023/06/03 20:39:01 - mmengine - INFO - Epoch(train) [3][ 500/1954] lr: 3.5307e-04 eta: 2:55:00 time: 0.6527 data_time: 0.0016 memory: 15585 loss: 0.0121 +2023/06/03 20:40:07 - mmengine - INFO - Epoch(train) [3][ 600/1954] lr: 3.5101e-04 eta: 2:53:39 time: 0.6449 data_time: 0.0017 memory: 15585 loss: 0.0089 +2023/06/03 20:41:16 - mmengine - INFO - Epoch(train) [3][ 700/1954] lr: 3.4891e-04 eta: 2:52:26 time: 0.6566 data_time: 0.0016 memory: 15585 loss: 0.0153 +2023/06/03 20:42:24 - mmengine - INFO - Epoch(train) [3][ 800/1954] lr: 3.4677e-04 eta: 2:51:15 time: 0.7227 data_time: 0.0015 memory: 15585 loss: 0.0116 +2023/06/03 20:43:31 - mmengine - INFO - Epoch(train) [3][ 900/1954] lr: 3.4460e-04 eta: 2:49:57 time: 0.7148 data_time: 0.0016 memory: 15585 loss: 0.0138 +2023/06/03 20:44:40 - mmengine - INFO - Epoch(train) [3][1000/1954] lr: 3.4240e-04 eta: 2:48:49 time: 0.6751 data_time: 0.0016 memory: 15585 loss: 0.0128 +2023/06/03 20:45:46 - mmengine - INFO - Exp name: convnext_small_4xb256_sdv2_1m_lr4e-4_aug_5e-1_20230603_194605 +2023/06/03 20:45:50 - mmengine - INFO - Epoch(train) [3][1100/1954] lr: 3.4015e-04 eta: 2:47:40 time: 0.6598 data_time: 0.0018 memory: 15585 loss: 0.0103 +2023/06/03 20:46:58 - mmengine - INFO - Epoch(train) [3][1200/1954] lr: 3.3788e-04 eta: 2:46:27 time: 0.6759 data_time: 0.0017 memory: 15585 loss: 0.0133 +2023/06/03 20:48:07 - mmengine - INFO - Epoch(train) [3][1300/1954] lr: 3.3557e-04 eta: 2:45:19 time: 0.6970 data_time: 0.0017 memory: 15585 loss: 0.0135 +2023/06/03 20:49:16 - mmengine - INFO - Epoch(train) [3][1400/1954] lr: 3.3322e-04 eta: 2:44:08 time: 0.6978 data_time: 0.0019 memory: 15585 loss: 0.0077 +2023/06/03 20:50:22 - mmengine - INFO - Epoch(train) [3][1500/1954] lr: 3.3084e-04 eta: 2:42:51 time: 0.6721 data_time: 0.0018 memory: 15585 loss: 0.0154 +2023/06/03 20:51:30 - mmengine - INFO - Epoch(train) [3][1600/1954] lr: 3.2843e-04 eta: 2:41:40 time: 0.6914 data_time: 0.0019 memory: 15585 loss: 0.0139 +2023/06/03 20:52:39 - mmengine - INFO - Epoch(train) [3][1700/1954] lr: 3.2599e-04 eta: 2:40:30 time: 0.6558 data_time: 0.0018 memory: 15585 loss: 0.0192 +2023/06/03 20:53:49 - mmengine - INFO - Epoch(train) [3][1800/1954] lr: 3.2351e-04 eta: 2:39:22 time: 0.6828 data_time: 0.0021 memory: 15585 loss: 0.0123 +2023/06/03 20:54:55 - mmengine - INFO - Epoch(train) [3][1900/1954] lr: 3.2101e-04 eta: 2:38:07 time: 0.6518 data_time: 0.0016 memory: 15585 loss: 0.0145 +2023/06/03 20:55:33 - mmengine - INFO - Exp name: convnext_small_4xb256_sdv2_1m_lr4e-4_aug_5e-1_20230603_194605 +2023/06/03 20:55:33 - mmengine - INFO - Saving checkpoint at 3 epochs +2023/06/03 20:56:10 - mmengine - INFO - Epoch(val) [3][29/29] accuracy/top1: 96.8477 single-label/precision_classwise: [94.88489532470703, 99.54560089111328] single-label/recall_classwise: [99.65279388427734, 93.4030990600586] single-label/f1-score_classwise: [97.21041870117188, 96.37657165527344] data_time: 0.3335 time: 0.7235 +2023/06/03 20:57:24 - mmengine - INFO - Epoch(train) [4][ 100/1954] lr: 3.1709e-04 eta: 2:36:31 time: 0.7409 data_time: 0.0020 memory: 15585 loss: 0.0115 +2023/06/03 20:57:51 - mmengine - INFO - Exp name: convnext_small_4xb256_sdv2_1m_lr4e-4_aug_5e-1_20230603_194605 +2023/06/03 20:58:34 - mmengine - INFO - Epoch(train) [4][ 200/1954] lr: 3.1451e-04 eta: 2:35:24 time: 0.6702 data_time: 0.0015 memory: 15585 loss: 0.0134 +2023/06/03 20:59:41 - mmengine - INFO - Epoch(train) [4][ 300/1954] lr: 3.1190e-04 eta: 2:34:08 time: 0.6307 data_time: 0.0017 memory: 15585 loss: 0.0073 +2023/06/03 21:00:48 - mmengine - INFO - Epoch(train) [4][ 400/1954] lr: 3.0927e-04 eta: 2:32:55 time: 0.7274 data_time: 0.0016 memory: 15585 loss: 0.0098 +2023/06/03 21:02:06 - mmengine - INFO - Epoch(train) [4][ 500/1954] lr: 3.0661e-04 eta: 2:32:05 time: 0.6560 data_time: 0.0017 memory: 15585 loss: 0.0148 +2023/06/03 21:03:13 - mmengine - INFO - Epoch(train) [4][ 600/1954] lr: 3.0392e-04 eta: 2:30:50 time: 0.6473 data_time: 0.0017 memory: 15585 loss: 0.0172 +2023/06/03 21:04:21 - mmengine - INFO - Epoch(train) [4][ 700/1954] lr: 3.0120e-04 eta: 2:29:40 time: 0.6559 data_time: 0.0016 memory: 15585 loss: 0.0108 +2023/06/03 21:05:28 - mmengine - INFO - Epoch(train) [4][ 800/1954] lr: 2.9846e-04 eta: 2:28:26 time: 0.6518 data_time: 0.0014 memory: 15585 loss: 0.0150 +2023/06/03 21:06:35 - mmengine - INFO - Epoch(train) [4][ 900/1954] lr: 2.9570e-04 eta: 2:27:13 time: 0.6484 data_time: 0.0014 memory: 15585 loss: 0.0092 +2023/06/03 21:07:45 - mmengine - INFO - Epoch(train) [4][1000/1954] lr: 2.9291e-04 eta: 2:26:06 time: 0.7806 data_time: 0.0014 memory: 15585 loss: 0.0106 +2023/06/03 21:08:55 - mmengine - INFO - Epoch(train) [4][1100/1954] lr: 2.9010e-04 eta: 2:24:57 time: 0.6899 data_time: 0.0014 memory: 15585 loss: 0.0114 +2023/06/03 21:09:20 - mmengine - INFO - Exp name: convnext_small_4xb256_sdv2_1m_lr4e-4_aug_5e-1_20230603_194605 +2023/06/03 21:10:01 - mmengine - INFO - Epoch(train) [4][1200/1954] lr: 2.8727e-04 eta: 2:23:43 time: 0.6519 data_time: 0.0014 memory: 15585 loss: 0.0132 +2023/06/03 21:11:16 - mmengine - INFO - Epoch(train) [4][1300/1954] lr: 2.8442e-04 eta: 2:22:44 time: 0.6678 data_time: 0.0016 memory: 15585 loss: 0.0117 +2023/06/03 21:12:21 - mmengine - INFO - Epoch(train) [4][1400/1954] lr: 2.8155e-04 eta: 2:21:27 time: 0.6377 data_time: 0.0015 memory: 15585 loss: 0.0084 +2023/06/03 21:13:31 - mmengine - INFO - Epoch(train) [4][1500/1954] lr: 2.7865e-04 eta: 2:20:20 time: 0.5352 data_time: 0.0149 memory: 15585 loss: 0.0084 +2023/06/03 21:14:48 - mmengine - INFO - Epoch(train) [4][1600/1954] lr: 2.7574e-04 eta: 2:19:24 time: 0.6818 data_time: 0.0018 memory: 15585 loss: 0.0090 +2023/06/03 21:15:51 - mmengine - INFO - Epoch(train) [4][1700/1954] lr: 2.7281e-04 eta: 2:18:04 time: 0.6514 data_time: 0.0017 memory: 15585 loss: 0.0055 +2023/06/03 21:16:54 - mmengine - INFO - Epoch(train) [4][1800/1954] lr: 2.6986e-04 eta: 2:16:46 time: 0.6486 data_time: 0.0015 memory: 15585 loss: 0.0176 +2023/06/03 21:17:57 - mmengine - INFO - Epoch(train) [4][1900/1954] lr: 2.6690e-04 eta: 2:15:27 time: 0.6373 data_time: 0.0016 memory: 15585 loss: 0.0096 +2023/06/03 21:18:30 - mmengine - INFO - Exp name: convnext_small_4xb256_sdv2_1m_lr4e-4_aug_5e-1_20230603_194605 +2023/06/03 21:18:30 - mmengine - INFO - Saving checkpoint at 4 epochs +2023/06/03 21:19:06 - mmengine - INFO - Epoch(val) [4][29/29] accuracy/top1: 98.5282 single-label/precision_classwise: [97.58641815185547, 99.74483489990234] single-label/recall_classwise: [99.7979965209961, 96.968994140625] single-label/f1-score_classwise: [98.67981719970703, 98.33732604980469] data_time: 0.2748 time: 0.6601 +2023/06/03 21:20:16 - mmengine - INFO - Epoch(train) [5][ 100/1954] lr: 2.6230e-04 eta: 2:13:35 time: 0.6195 data_time: 0.0018 memory: 15585 loss: 0.0048 +2023/06/03 21:21:11 - mmengine - INFO - Exp name: convnext_small_4xb256_sdv2_1m_lr4e-4_aug_5e-1_20230603_194605 +2023/06/03 21:21:21 - mmengine - INFO - Epoch(train) [5][ 200/1954] lr: 2.5929e-04 eta: 2:12:21 time: 0.6085 data_time: 0.0014 memory: 15585 loss: 0.0134 +2023/06/03 21:22:25 - mmengine - INFO - Epoch(train) [5][ 300/1954] lr: 2.5628e-04 eta: 2:11:06 time: 0.5957 data_time: 0.0017 memory: 15585 loss: 0.0121 +2023/06/03 21:23:31 - mmengine - INFO - Epoch(train) [5][ 400/1954] lr: 2.5325e-04 eta: 2:09:53 time: 0.6398 data_time: 0.0017 memory: 15585 loss: 0.0091 +2023/06/03 21:24:35 - mmengine - INFO - Epoch(train) [5][ 500/1954] lr: 2.5020e-04 eta: 2:08:38 time: 0.6455 data_time: 0.0020 memory: 15585 loss: 0.0056 +2023/06/03 21:25:39 - mmengine - INFO - Epoch(train) [5][ 600/1954] lr: 2.4715e-04 eta: 2:07:22 time: 0.5959 data_time: 0.0016 memory: 15585 loss: 0.0099 +2023/06/03 21:26:43 - mmengine - INFO - Epoch(train) [5][ 700/1954] lr: 2.4408e-04 eta: 2:06:07 time: 0.7091 data_time: 0.0016 memory: 15585 loss: 0.0091 +2023/06/03 21:27:48 - mmengine - INFO - Epoch(train) [5][ 800/1954] lr: 2.4100e-04 eta: 2:04:54 time: 0.6585 data_time: 0.0016 memory: 15585 loss: 0.0132 +2023/06/03 21:28:51 - mmengine - INFO - Epoch(train) [5][ 900/1954] lr: 2.3792e-04 eta: 2:03:38 time: 0.6162 data_time: 0.0015 memory: 15585 loss: 0.0050 +2023/06/03 21:29:55 - mmengine - INFO - Epoch(train) [5][1000/1954] lr: 2.3482e-04 eta: 2:02:25 time: 0.6582 data_time: 0.0016 memory: 15585 loss: 0.0112 +2023/06/03 21:31:00 - mmengine - INFO - Epoch(train) [5][1100/1954] lr: 2.3172e-04 eta: 2:01:12 time: 0.6510 data_time: 0.0015 memory: 15585 loss: 0.0067 +2023/06/03 21:31:56 - mmengine - INFO - Exp name: convnext_small_4xb256_sdv2_1m_lr4e-4_aug_5e-1_20230603_194605 +2023/06/03 21:32:06 - mmengine - INFO - Epoch(train) [5][1200/1954] lr: 2.2861e-04 eta: 2:00:01 time: 0.5934 data_time: 0.0015 memory: 15585 loss: 0.0081 +2023/06/03 21:33:10 - mmengine - INFO - Epoch(train) [5][1300/1954] lr: 2.2550e-04 eta: 1:58:47 time: 0.6141 data_time: 0.0015 memory: 15585 loss: 0.0089 +2023/06/03 21:34:16 - mmengine - INFO - Epoch(train) [5][1400/1954] lr: 2.2238e-04 eta: 1:57:36 time: 0.6153 data_time: 0.0016 memory: 15585 loss: 0.0088 +2023/06/03 21:35:19 - mmengine - INFO - Epoch(train) [5][1500/1954] lr: 2.1925e-04 eta: 1:56:22 time: 0.6370 data_time: 0.0014 memory: 15585 loss: 0.0056 +2023/06/03 21:36:23 - mmengine - INFO - Epoch(train) [5][1600/1954] lr: 2.1612e-04 eta: 1:55:09 time: 0.6627 data_time: 0.0015 memory: 15585 loss: 0.0062 +2023/06/03 21:37:25 - mmengine - INFO - Epoch(train) [5][1700/1954] lr: 2.1299e-04 eta: 1:53:55 time: 0.6244 data_time: 0.0014 memory: 15585 loss: 0.0123 +2023/06/03 21:38:31 - mmengine - INFO - Epoch(train) [5][1800/1954] lr: 2.0986e-04 eta: 1:52:43 time: 0.6566 data_time: 0.0015 memory: 15585 loss: 0.0047 +2023/06/03 21:39:33 - mmengine - INFO - Epoch(train) [5][1900/1954] lr: 2.0672e-04 eta: 1:51:29 time: 0.6219 data_time: 0.0014 memory: 15585 loss: 0.0097 +2023/06/03 21:40:06 - mmengine - INFO - Exp name: convnext_small_4xb256_sdv2_1m_lr4e-4_aug_5e-1_20230603_194605 +2023/06/03 21:40:06 - mmengine - INFO - Saving checkpoint at 5 epochs +2023/06/03 21:40:40 - mmengine - INFO - Epoch(val) [5][29/29] accuracy/top1: 99.2972 single-label/precision_classwise: [98.88104248046875, 99.81952667236328] single-label/recall_classwise: [99.85481262207031, 98.6124038696289] single-label/f1-score_classwise: [99.36553955078125, 99.21229553222656] data_time: 0.2399 time: 0.6237 +2023/06/03 21:41:52 - mmengine - INFO - Epoch(train) [6][ 100/1954] lr: 2.0190e-04 eta: 1:49:44 time: 0.6779 data_time: 0.0017 memory: 15585 loss: 0.0167 +2023/06/03 21:42:58 - mmengine - INFO - Epoch(train) [6][ 200/1954] lr: 1.9876e-04 eta: 1:48:34 time: 0.6569 data_time: 0.0018 memory: 15585 loss: 0.0076 +2023/06/03 21:43:16 - mmengine - INFO - Exp name: convnext_small_4xb256_sdv2_1m_lr4e-4_aug_5e-1_20230603_194605 +2023/06/03 21:44:01 - mmengine - INFO - Epoch(train) [6][ 300/1954] lr: 1.9563e-04 eta: 1:47:21 time: 0.6225 data_time: 0.0017 memory: 15585 loss: 0.0062 +2023/06/03 21:45:06 - mmengine - INFO - Epoch(train) [6][ 400/1954] lr: 1.9250e-04 eta: 1:46:11 time: 0.6652 data_time: 0.0018 memory: 15585 loss: 0.0066 +2023/06/03 21:46:11 - mmengine - INFO - Epoch(train) [6][ 500/1954] lr: 1.8937e-04 eta: 1:45:00 time: 0.6641 data_time: 0.0019 memory: 15585 loss: 0.0096 +2023/06/03 21:47:16 - mmengine - INFO - Epoch(train) [6][ 600/1954] lr: 1.8625e-04 eta: 1:43:49 time: 0.6893 data_time: 0.0017 memory: 15585 loss: 0.0071 +2023/06/03 21:48:21 - mmengine - INFO - Epoch(train) [6][ 700/1954] lr: 1.8313e-04 eta: 1:42:38 time: 0.6900 data_time: 0.0019 memory: 15585 loss: 0.0127 +2023/06/03 21:49:24 - mmengine - INFO - Epoch(train) [6][ 800/1954] lr: 1.8002e-04 eta: 1:41:27 time: 0.6615 data_time: 0.0017 memory: 15585 loss: 0.0095 +2023/06/03 21:50:27 - mmengine - INFO - Epoch(train) [6][ 900/1954] lr: 1.7691e-04 eta: 1:40:14 time: 0.5998 data_time: 0.0016 memory: 15585 loss: 0.0058 +2023/06/03 21:51:29 - mmengine - INFO - Epoch(train) [6][1000/1954] lr: 1.7381e-04 eta: 1:39:02 time: 0.5973 data_time: 0.0017 memory: 15585 loss: 0.0129 +2023/06/03 21:52:32 - mmengine - INFO - Epoch(train) [6][1100/1954] lr: 1.7072e-04 eta: 1:37:51 time: 0.5993 data_time: 0.0019 memory: 15585 loss: 0.0064 +2023/06/03 21:53:35 - mmengine - INFO - Epoch(train) [6][1200/1954] lr: 1.6764e-04 eta: 1:36:39 time: 0.6283 data_time: 0.0021 memory: 15585 loss: 0.0073 +2023/06/03 21:53:53 - mmengine - INFO - Exp name: convnext_small_4xb256_sdv2_1m_lr4e-4_aug_5e-1_20230603_194605 +2023/06/03 21:54:38 - mmengine - INFO - Epoch(train) [6][1300/1954] lr: 1.6457e-04 eta: 1:35:28 time: 0.6454 data_time: 0.0018 memory: 15585 loss: 0.0035 +2023/06/03 21:55:42 - mmengine - INFO - Epoch(train) [6][1400/1954] lr: 1.6151e-04 eta: 1:34:17 time: 0.6361 data_time: 0.0016 memory: 15585 loss: 0.0102 +2023/06/03 21:56:45 - mmengine - INFO - Epoch(train) [6][1500/1954] lr: 1.5846e-04 eta: 1:33:07 time: 0.6031 data_time: 0.0015 memory: 15585 loss: 0.0104 +2023/06/03 21:57:51 - mmengine - INFO - Epoch(train) [6][1600/1954] lr: 1.5542e-04 eta: 1:31:58 time: 0.6476 data_time: 0.0021 memory: 15585 loss: 0.0088 +2023/06/03 21:58:56 - mmengine - INFO - Epoch(train) [6][1700/1954] lr: 1.5239e-04 eta: 1:30:48 time: 0.6305 data_time: 0.0018 memory: 15585 loss: 0.0064 +2023/06/03 21:59:59 - mmengine - INFO - Epoch(train) [6][1800/1954] lr: 1.4938e-04 eta: 1:29:38 time: 0.6221 data_time: 0.0021 memory: 15585 loss: 0.0067 +2023/06/03 22:01:04 - mmengine - INFO - Epoch(train) [6][1900/1954] lr: 1.4638e-04 eta: 1:28:28 time: 0.6618 data_time: 0.0022 memory: 15585 loss: 0.0113 +2023/06/03 22:01:38 - mmengine - INFO - Exp name: convnext_small_4xb256_sdv2_1m_lr4e-4_aug_5e-1_20230603_194605 +2023/06/03 22:01:38 - mmengine - INFO - Saving checkpoint at 6 epochs +2023/06/03 22:02:14 - mmengine - INFO - Epoch(val) [6][29/29] accuracy/top1: 99.6138 single-label/precision_classwise: [99.44674682617188, 99.82080078125] single-label/recall_classwise: [99.85481262207031, 99.31783294677734] single-label/f1-score_classwise: [99.65036010742188, 99.56867218017578] data_time: 0.2569 time: 0.6421 +2023/06/03 22:03:24 - mmengine - INFO - Epoch(train) [7][ 100/1954] lr: 1.4180e-04 eta: 1:26:45 time: 0.6674 data_time: 0.0015 memory: 15585 loss: 0.0092 +2023/06/03 22:04:30 - mmengine - INFO - Epoch(train) [7][ 200/1954] lr: 1.3884e-04 eta: 1:25:36 time: 0.6143 data_time: 0.0015 memory: 15585 loss: 0.0071 +2023/06/03 22:05:21 - mmengine - INFO - Exp name: convnext_small_4xb256_sdv2_1m_lr4e-4_aug_5e-1_20230603_194605 +2023/06/03 22:05:36 - mmengine - INFO - Epoch(train) [7][ 300/1954] lr: 1.3590e-04 eta: 1:24:28 time: 0.6410 data_time: 0.0014 memory: 15585 loss: 0.0055 +2023/06/03 22:06:42 - mmengine - INFO - Epoch(train) [7][ 400/1954] lr: 1.3298e-04 eta: 1:23:19 time: 0.6931 data_time: 0.0017 memory: 15585 loss: 0.0066 +2023/06/03 22:07:46 - mmengine - INFO - Epoch(train) [7][ 500/1954] lr: 1.3007e-04 eta: 1:22:10 time: 0.6540 data_time: 0.0017 memory: 15585 loss: 0.0054 +2023/06/03 22:08:50 - mmengine - INFO - Epoch(train) [7][ 600/1954] lr: 1.2719e-04 eta: 1:21:01 time: 0.6361 data_time: 0.0017 memory: 15585 loss: 0.0053 +2023/06/03 22:09:55 - mmengine - INFO - Epoch(train) [7][ 700/1954] lr: 1.2432e-04 eta: 1:19:52 time: 0.6362 data_time: 0.0018 memory: 15585 loss: 0.0068 +2023/06/03 22:11:00 - mmengine - INFO - Epoch(train) [7][ 800/1954] lr: 1.2148e-04 eta: 1:18:43 time: 0.6669 data_time: 0.0019 memory: 15585 loss: 0.0028 +2023/06/03 22:12:05 - mmengine - INFO - Epoch(train) [7][ 900/1954] lr: 1.1866e-04 eta: 1:17:35 time: 0.6647 data_time: 0.0017 memory: 15585 loss: 0.0045 +2023/06/03 22:13:11 - mmengine - INFO - Epoch(train) [7][1000/1954] lr: 1.1586e-04 eta: 1:16:27 time: 0.6910 data_time: 0.0018 memory: 15585 loss: 0.0084 +2023/06/03 22:14:16 - mmengine - INFO - Epoch(train) [7][1100/1954] lr: 1.1308e-04 eta: 1:15:18 time: 0.6773 data_time: 0.0016 memory: 15585 loss: 0.0048 +2023/06/03 22:15:21 - mmengine - INFO - Epoch(train) [7][1200/1954] lr: 1.1033e-04 eta: 1:14:10 time: 0.6646 data_time: 0.0017 memory: 15585 loss: 0.0036 +2023/06/03 22:16:12 - mmengine - INFO - Exp name: convnext_small_4xb256_sdv2_1m_lr4e-4_aug_5e-1_20230603_194605 +2023/06/03 22:16:27 - mmengine - INFO - Epoch(train) [7][1300/1954] lr: 1.0760e-04 eta: 1:13:02 time: 0.6454 data_time: 0.0017 memory: 15585 loss: 0.0028 +2023/06/03 22:17:32 - mmengine - INFO - Epoch(train) [7][1400/1954] lr: 1.0490e-04 eta: 1:11:53 time: 0.6318 data_time: 0.0018 memory: 15585 loss: 0.0059 +2023/06/03 22:18:37 - mmengine - INFO - Epoch(train) [7][1500/1954] lr: 1.0222e-04 eta: 1:10:45 time: 0.6303 data_time: 0.0017 memory: 15585 loss: 0.0053 +2023/06/03 22:19:42 - mmengine - INFO - Epoch(train) [7][1600/1954] lr: 9.9568e-05 eta: 1:09:37 time: 0.6425 data_time: 0.0017 memory: 15585 loss: 0.0018 +2023/06/03 22:20:47 - mmengine - INFO - Epoch(train) [7][1700/1954] lr: 9.6944e-05 eta: 1:08:29 time: 0.6663 data_time: 0.0015 memory: 15585 loss: 0.0030 +2023/06/03 22:21:54 - mmengine - INFO - Epoch(train) [7][1800/1954] lr: 9.4348e-05 eta: 1:07:21 time: 0.7162 data_time: 0.0015 memory: 15585 loss: 0.0051 +2023/06/03 22:22:59 - mmengine - INFO - Epoch(train) [7][1900/1954] lr: 9.1781e-05 eta: 1:06:13 time: 0.6930 data_time: 0.0017 memory: 15585 loss: 0.0050 +2023/06/03 22:23:33 - mmengine - INFO - Exp name: convnext_small_4xb256_sdv2_1m_lr4e-4_aug_5e-1_20230603_194605 +2023/06/03 22:23:33 - mmengine - INFO - Saving checkpoint at 7 epochs +2023/06/03 22:24:08 - mmengine - INFO - Epoch(val) [7][29/29] accuracy/top1: 99.6695 single-label/precision_classwise: [99.47216796875, 99.91423797607422] single-label/recall_classwise: [99.93055725097656, 99.34883880615234] single-label/f1-score_classwise: [99.70083618164062, 99.6307373046875] data_time: 0.2506 time: 0.6394 +2023/06/03 22:25:19 - mmengine - INFO - Epoch(train) [8][ 100/1954] lr: 8.7885e-05 eta: 1:04:30 time: 0.6380 data_time: 0.0019 memory: 15585 loss: 0.0037 +2023/06/03 22:26:23 - mmengine - INFO - Epoch(train) [8][ 200/1954] lr: 8.5394e-05 eta: 1:03:22 time: 0.6349 data_time: 0.0018 memory: 15585 loss: 0.0079 +2023/06/03 22:27:28 - mmengine - INFO - Epoch(train) [8][ 300/1954] lr: 8.2933e-05 eta: 1:02:14 time: 0.6736 data_time: 0.0019 memory: 15585 loss: 0.0053 +2023/06/03 22:27:43 - mmengine - INFO - Exp name: convnext_small_4xb256_sdv2_1m_lr4e-4_aug_5e-1_20230603_194605 +2023/06/03 22:28:34 - mmengine - INFO - Epoch(train) [8][ 400/1954] lr: 8.0504e-05 eta: 1:01:06 time: 0.6391 data_time: 0.0018 memory: 15585 loss: 0.0046 +2023/06/03 22:29:39 - mmengine - INFO - Epoch(train) [8][ 500/1954] lr: 7.8107e-05 eta: 0:59:58 time: 0.6688 data_time: 0.0016 memory: 15585 loss: 0.0063 +2023/06/03 22:30:45 - mmengine - INFO - Epoch(train) [8][ 600/1954] lr: 7.5743e-05 eta: 0:58:51 time: 0.6456 data_time: 0.0016 memory: 15585 loss: 0.0038 +2023/06/03 22:31:52 - mmengine - INFO - Epoch(train) [8][ 700/1954] lr: 7.3413e-05 eta: 0:57:44 time: 0.6585 data_time: 0.0016 memory: 15585 loss: 0.0042 +2023/06/03 22:32:57 - mmengine - INFO - Epoch(train) [8][ 800/1954] lr: 7.1116e-05 eta: 0:56:36 time: 0.6674 data_time: 0.0015 memory: 15585 loss: 0.0030 +2023/06/03 22:34:02 - mmengine - INFO - Epoch(train) [8][ 900/1954] lr: 6.8854e-05 eta: 0:55:28 time: 0.6316 data_time: 0.0016 memory: 15585 loss: 0.0032 +2023/06/03 22:35:08 - mmengine - INFO - Epoch(train) [8][1000/1954] lr: 6.6627e-05 eta: 0:54:21 time: 0.6623 data_time: 0.0023 memory: 15585 loss: 0.0051 +2023/06/03 22:36:14 - mmengine - INFO - Epoch(train) [8][1100/1954] lr: 6.4436e-05 eta: 0:53:13 time: 0.6347 data_time: 0.0017 memory: 15585 loss: 0.0033 +2023/06/03 22:37:20 - mmengine - INFO - Epoch(train) [8][1200/1954] lr: 6.2282e-05 eta: 0:52:06 time: 0.6475 data_time: 0.0021 memory: 15585 loss: 0.0053 +2023/06/03 22:38:26 - mmengine - INFO - Epoch(train) [8][1300/1954] lr: 6.0164e-05 eta: 0:50:59 time: 0.6367 data_time: 0.0024 memory: 15585 loss: 0.0028 +2023/06/03 22:38:41 - mmengine - INFO - Exp name: convnext_small_4xb256_sdv2_1m_lr4e-4_aug_5e-1_20230603_194605 +2023/06/03 22:39:33 - mmengine - INFO - Epoch(train) [8][1400/1954] lr: 5.8083e-05 eta: 0:49:51 time: 0.6971 data_time: 0.0018 memory: 15585 loss: 0.0039 +2023/06/03 22:40:40 - mmengine - INFO - Epoch(train) [8][1500/1954] lr: 5.6041e-05 eta: 0:48:44 time: 0.6897 data_time: 0.0022 memory: 15585 loss: 0.0049 +2023/06/03 22:41:46 - mmengine - INFO - Epoch(train) [8][1600/1954] lr: 5.4037e-05 eta: 0:47:37 time: 0.6601 data_time: 0.0019 memory: 15585 loss: 0.0019 +2023/06/03 22:42:52 - mmengine - INFO - Epoch(train) [8][1700/1954] lr: 5.2072e-05 eta: 0:46:30 time: 0.6410 data_time: 0.0015 memory: 15585 loss: 0.0034 +2023/06/03 22:43:56 - mmengine - INFO - Epoch(train) [8][1800/1954] lr: 5.0147e-05 eta: 0:45:22 time: 0.6290 data_time: 0.0018 memory: 15585 loss: 0.0023 +2023/06/03 22:45:03 - mmengine - INFO - Epoch(train) [8][1900/1954] lr: 4.8261e-05 eta: 0:44:15 time: 0.6851 data_time: 0.0016 memory: 15585 loss: 0.0062 +2023/06/03 22:45:38 - mmengine - INFO - Exp name: convnext_small_4xb256_sdv2_1m_lr4e-4_aug_5e-1_20230603_194605 +2023/06/03 22:45:38 - mmengine - INFO - Saving checkpoint at 8 epochs +2023/06/03 22:46:13 - mmengine - INFO - Epoch(val) [8][29/29] accuracy/top1: 99.7530 single-label/precision_classwise: [99.67241668701172, 99.85233306884766] single-label/recall_classwise: [99.88005828857422, 99.5969009399414] single-label/f1-score_classwise: [99.77613067626953, 99.72445678710938] data_time: 0.2805 time: 0.6655 +2023/06/03 22:47:24 - mmengine - INFO - Epoch(train) [9][ 100/1954] lr: 4.5437e-05 eta: 0:42:32 time: 0.6607 data_time: 0.0017 memory: 15585 loss: 0.0021 +2023/06/03 22:48:30 - mmengine - INFO - Epoch(train) [9][ 200/1954] lr: 4.3656e-05 eta: 0:41:25 time: 0.6441 data_time: 0.0016 memory: 15585 loss: 0.0042 +2023/06/03 22:49:35 - mmengine - INFO - Epoch(train) [9][ 300/1954] lr: 4.1916e-05 eta: 0:40:17 time: 0.6652 data_time: 0.0016 memory: 15585 loss: 0.0050 +2023/06/03 22:50:21 - mmengine - INFO - Exp name: convnext_small_4xb256_sdv2_1m_lr4e-4_aug_5e-1_20230603_194605 +2023/06/03 22:50:41 - mmengine - INFO - Epoch(train) [9][ 400/1954] lr: 4.0218e-05 eta: 0:39:10 time: 0.6334 data_time: 0.0016 memory: 15585 loss: 0.0053 +2023/06/03 22:51:47 - mmengine - INFO - Epoch(train) [9][ 500/1954] lr: 3.8563e-05 eta: 0:38:03 time: 0.6555 data_time: 0.0017 memory: 15585 loss: 0.0046 +2023/06/03 22:52:54 - mmengine - INFO - Epoch(train) [9][ 600/1954] lr: 3.6951e-05 eta: 0:36:56 time: 0.6383 data_time: 0.0016 memory: 15585 loss: 0.0033 +2023/06/03 22:54:00 - mmengine - INFO - Epoch(train) [9][ 700/1954] lr: 3.5383e-05 eta: 0:35:49 time: 0.6229 data_time: 0.0018 memory: 15585 loss: 0.0040 +2023/06/03 22:55:07 - mmengine - INFO - Epoch(train) [9][ 800/1954] lr: 3.3858e-05 eta: 0:34:42 time: 0.7091 data_time: 0.0016 memory: 15585 loss: 0.0012 +2023/06/03 22:56:13 - mmengine - INFO - Epoch(train) [9][ 900/1954] lr: 3.2378e-05 eta: 0:33:34 time: 0.6770 data_time: 0.0016 memory: 15585 loss: 0.0068 +2023/06/03 22:57:19 - mmengine - INFO - Epoch(train) [9][1000/1954] lr: 3.0942e-05 eta: 0:32:27 time: 0.6279 data_time: 0.0017 memory: 15585 loss: 0.0076 +2023/06/03 22:58:25 - mmengine - INFO - Epoch(train) [9][1100/1954] lr: 2.9551e-05 eta: 0:31:20 time: 0.6590 data_time: 0.0019 memory: 15585 loss: 0.0019 +2023/06/03 22:59:30 - mmengine - INFO - Epoch(train) [9][1200/1954] lr: 2.8205e-05 eta: 0:30:13 time: 0.6033 data_time: 0.0017 memory: 15585 loss: 0.0029 +2023/06/03 23:00:38 - mmengine - INFO - Epoch(train) [9][1300/1954] lr: 2.6905e-05 eta: 0:29:06 time: 0.6482 data_time: 0.0016 memory: 15585 loss: 0.0028 +2023/06/03 23:01:23 - mmengine - INFO - Exp name: convnext_small_4xb256_sdv2_1m_lr4e-4_aug_5e-1_20230603_194605 +2023/06/03 23:01:45 - mmengine - INFO - Epoch(train) [9][1400/1954] lr: 2.5652e-05 eta: 0:27:59 time: 0.6410 data_time: 0.0016 memory: 15585 loss: 0.0032 +2023/06/03 23:02:52 - mmengine - INFO - Epoch(train) [9][1500/1954] lr: 2.4444e-05 eta: 0:26:52 time: 0.6821 data_time: 0.0016 memory: 15585 loss: 0.0015 +2023/06/03 23:03:59 - mmengine - INFO - Epoch(train) [9][1600/1954] lr: 2.3283e-05 eta: 0:25:45 time: 0.6349 data_time: 0.0014 memory: 15585 loss: 0.0025 +2023/06/03 23:05:07 - mmengine - INFO - Epoch(train) [9][1700/1954] lr: 2.2170e-05 eta: 0:24:38 time: 0.6543 data_time: 0.0016 memory: 15585 loss: 0.0034 +2023/06/03 23:06:14 - mmengine - INFO - Epoch(train) [9][1800/1954] lr: 2.1103e-05 eta: 0:23:31 time: 0.6806 data_time: 0.0015 memory: 15585 loss: 0.0020 +2023/06/03 23:07:21 - mmengine - INFO - Epoch(train) [9][1900/1954] lr: 2.0084e-05 eta: 0:22:24 time: 0.6655 data_time: 0.0016 memory: 15585 loss: 0.0013 +2023/06/03 23:07:56 - mmengine - INFO - Exp name: convnext_small_4xb256_sdv2_1m_lr4e-4_aug_5e-1_20230603_194605 +2023/06/03 23:07:56 - mmengine - INFO - Saving checkpoint at 9 epochs +2023/06/03 23:08:32 - mmengine - INFO - Epoch(val) [9][29/29] accuracy/top1: 99.8260 single-label/precision_classwise: [99.722900390625, 99.95335388183594] single-label/recall_classwise: [99.96212005615234, 99.6589126586914] single-label/f1-score_classwise: [99.84236145019531, 99.805908203125] data_time: 0.2769 time: 0.6611 +2023/06/03 23:09:45 - mmengine - INFO - Epoch(train) [10][ 100/1954] lr: 1.8608e-05 eta: 0:20:42 time: 0.6364 data_time: 0.0018 memory: 15585 loss: 0.0056 +2023/06/03 23:10:52 - mmengine - INFO - Epoch(train) [10][ 200/1954] lr: 1.7711e-05 eta: 0:19:35 time: 0.6549 data_time: 0.0017 memory: 15585 loss: 0.0044 +2023/06/03 23:11:58 - mmengine - INFO - Epoch(train) [10][ 300/1954] lr: 1.6863e-05 eta: 0:18:28 time: 0.6513 data_time: 0.0020 memory: 15585 loss: 0.0035 +2023/06/03 23:13:04 - mmengine - INFO - Epoch(train) [10][ 400/1954] lr: 1.6063e-05 eta: 0:17:21 time: 0.6334 data_time: 0.0018 memory: 15585 loss: 0.0023 +2023/06/03 23:13:14 - mmengine - INFO - Exp name: convnext_small_4xb256_sdv2_1m_lr4e-4_aug_5e-1_20230603_194605 +2023/06/03 23:14:10 - mmengine - INFO - Epoch(train) [10][ 500/1954] lr: 1.5311e-05 eta: 0:16:13 time: 0.6517 data_time: 0.0017 memory: 15585 loss: 0.0019 +2023/06/03 23:15:16 - mmengine - INFO - Epoch(train) [10][ 600/1954] lr: 1.4609e-05 eta: 0:15:06 time: 0.6436 data_time: 0.0016 memory: 15585 loss: 0.0017 +2023/06/03 23:16:23 - mmengine - INFO - Epoch(train) [10][ 700/1954] lr: 1.3956e-05 eta: 0:13:59 time: 0.6807 data_time: 0.0016 memory: 15585 loss: 0.0042 +2023/06/03 23:17:31 - mmengine - INFO - Epoch(train) [10][ 800/1954] lr: 1.3353e-05 eta: 0:12:52 time: 0.6580 data_time: 0.0017 memory: 15585 loss: 0.0063 +2023/06/03 23:18:36 - mmengine - INFO - Epoch(train) [10][ 900/1954] lr: 1.2798e-05 eta: 0:11:45 time: 0.6592 data_time: 0.0017 memory: 15585 loss: 0.0013 +2023/06/03 23:19:42 - mmengine - INFO - Epoch(train) [10][1000/1954] lr: 1.2294e-05 eta: 0:10:38 time: 0.6329 data_time: 0.0016 memory: 15585 loss: 0.0023 +2023/06/03 23:20:50 - mmengine - INFO - Epoch(train) [10][1100/1954] lr: 1.1840e-05 eta: 0:09:31 time: 0.6506 data_time: 0.0015 memory: 15585 loss: 0.0070 +2023/06/03 23:21:55 - mmengine - INFO - Epoch(train) [10][1200/1954] lr: 1.1435e-05 eta: 0:08:24 time: 0.6760 data_time: 0.0015 memory: 15585 loss: 0.0047 +2023/06/03 23:23:03 - mmengine - INFO - Epoch(train) [10][1300/1954] lr: 1.1080e-05 eta: 0:07:17 time: 0.6562 data_time: 0.0014 memory: 15585 loss: 0.0031 +2023/06/03 23:24:10 - mmengine - INFO - Epoch(train) [10][1400/1954] lr: 1.0776e-05 eta: 0:06:11 time: 0.6839 data_time: 0.0015 memory: 15585 loss: 0.0043 +2023/06/03 23:24:20 - mmengine - INFO - Exp name: convnext_small_4xb256_sdv2_1m_lr4e-4_aug_5e-1_20230603_194605 +2023/06/03 23:25:16 - mmengine - INFO - Epoch(train) [10][1500/1954] lr: 1.0522e-05 eta: 0:05:04 time: 0.6396 data_time: 0.0015 memory: 15585 loss: 0.0040 +2023/06/03 23:26:23 - mmengine - INFO - Epoch(train) [10][1600/1954] lr: 1.0318e-05 eta: 0:03:57 time: 0.6407 data_time: 0.0015 memory: 15585 loss: 0.0047 +2023/06/03 23:27:28 - mmengine - INFO - Epoch(train) [10][1700/1954] lr: 1.0164e-05 eta: 0:02:50 time: 0.6325 data_time: 0.0014 memory: 15585 loss: 0.0026 +2023/06/03 23:28:34 - mmengine - INFO - Epoch(train) [10][1800/1954] lr: 1.0061e-05 eta: 0:01:43 time: 0.6033 data_time: 0.0015 memory: 15585 loss: 0.0031 +2023/06/03 23:29:41 - mmengine - INFO - Epoch(train) [10][1900/1954] lr: 1.0008e-05 eta: 0:00:36 time: 0.7015 data_time: 0.0015 memory: 15585 loss: 0.0017 +2023/06/03 23:30:17 - mmengine - INFO - Exp name: convnext_small_4xb256_sdv2_1m_lr4e-4_aug_5e-1_20230603_194605 +2023/06/03 23:30:17 - mmengine - INFO - Saving checkpoint at 10 epochs +2023/06/03 23:30:53 - mmengine - INFO - Epoch(val) [10][29/29] accuracy/top1: 99.7808 single-label/precision_classwise: [99.68509674072266, 99.89893341064453] single-label/recall_classwise: [99.91793823242188, 99.6124038696289] single-label/f1-score_classwise: [99.80138397216797, 99.75546264648438] data_time: 0.2586 time: 0.6472 diff --git a/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_5e-1/20230603_194605/vis_data/20230603_194605.json b/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_5e-1/20230603_194605/vis_data/20230603_194605.json new file mode 100644 index 0000000000000000000000000000000000000000..71ae044effb85dda89e836d40de0c2773ebaa641 --- /dev/null +++ b/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_5e-1/20230603_194605/vis_data/20230603_194605.json @@ -0,0 +1,200 @@ +{"lr": 0.00039997529889041205, "data_time": 0.0016591310501098632, "loss": 0.051201630383729935, "time": 0.8002232551574707, "epoch": 1, "memory": 15587, "step": 100} +{"lr": 0.0003999002014233684, "data_time": 0.0016923189163208009, "loss": 0.04154322072863579, "time": 0.6618563413619996, "epoch": 1, "memory": 15587, "step": 200} +{"lr": 0.0003997747244903853, "data_time": 0.0017637252807617188, "loss": 0.03195568625815213, "time": 0.6556795835494995, "epoch": 1, "memory": 15587, "step": 300} +{"lr": 0.00039959890052580966, "data_time": 0.002024054527282715, "loss": 0.02430737824179232, "time": 0.7310150623321533, "epoch": 1, "memory": 15587, "step": 400} +{"lr": 0.0003993727749781197, "data_time": 0.002002573013305664, "loss": 0.039762046094983815, "time": 0.7297355175018311, "epoch": 1, "memory": 15587, "step": 500} +{"lr": 0.000399096406298175, "data_time": 0.0018381595611572266, "loss": 0.025331319961696862, "time": 0.66280677318573, "epoch": 1, "memory": 15587, "step": 600} +{"lr": 0.0003987698659241082, "data_time": 0.0017859697341918944, "loss": 0.02447650199756026, "time": 0.9731130838394165, "epoch": 1, "memory": 15587, "step": 700} +{"lr": 0.0003983932382628598, "data_time": 0.0025223731994628907, "loss": 0.02874429989606142, "time": 0.7066556215286255, "epoch": 1, "memory": 15587, "step": 800} +{"lr": 0.00039796662066835946, "data_time": 0.0018059015274047852, "loss": 0.025640539452433588, "time": 0.6873111486434936, "epoch": 1, "memory": 15587, "step": 900} +{"lr": 0.00039749012341636, "data_time": 0.0016995668411254883, "loss": 0.016293613985180855, "time": 0.6786811351776123, "epoch": 1, "memory": 15587, "step": 1000} +{"lr": 0.00039696386967593436, "data_time": 0.0017138481140136718, "loss": 0.0216406867839396, "time": 0.7678688764572144, "epoch": 1, "memory": 15587, "step": 1100} +{"lr": 0.00039638799547763706, "data_time": 0.0016362190246582032, "loss": 0.023681106604635717, "time": 0.665216326713562, "epoch": 1, "memory": 15587, "step": 1200} +{"lr": 0.0003957626496783407, "data_time": 0.0017006874084472656, "loss": 0.022824741108343004, "time": 0.7069957256317139, "epoch": 1, "memory": 15587, "step": 1300} +{"lr": 0.0003950879939227595, "data_time": 0.0016504287719726562, "loss": 0.02190188770182431, "time": 0.6563145399093628, "epoch": 1, "memory": 15587, "step": 1400} +{"lr": 0.0003943642026016663, "data_time": 0.001715540885925293, "loss": 0.019457390438765286, "time": 0.6750092029571533, "epoch": 1, "memory": 15587, "step": 1500} +{"lr": 0.0003935914628068133, "data_time": 0.001839423179626465, "loss": 0.021533434046432376, "time": 0.6075412511825562, "epoch": 1, "memory": 15587, "step": 1600} +{"lr": 0.00039276997428257205, "data_time": 0.0016817808151245116, "loss": 0.017051718244329096, "time": 0.6816421985626221, "epoch": 1, "memory": 15587, "step": 1700} +{"lr": 0.0003918999493743001, "data_time": 0.0016488313674926757, "loss": 0.0199392921756953, "time": 0.6629706621170044, "epoch": 1, "memory": 15587, "step": 1800} +{"lr": 0.00039098161297345475, "data_time": 0.0017903566360473633, "loss": 0.02119651101529598, "time": 0.6733246088027954, "epoch": 1, "memory": 15587, "step": 1900} +{"accuracy/top1": 82.07090759277344, "data_time": 0.32828064622550174, "time": 0.7906504499501196, "step": 1} +{"lr": 0.0003894734390071047, "data_time": 0.0017095327377319336, "loss": 0.012041058111935853, "time": 0.7513424634933472, "epoch": 2, "memory": 15587, "step": 2054} +{"lr": 0.0003884334872841784, "data_time": 0.0015699148178100586, "loss": 0.01816981448791921, "time": 0.6740405559539795, "epoch": 2, "memory": 15585, "step": 2154} +{"lr": 0.00038734612010935703, "data_time": 0.0018655776977539063, "loss": 0.022913835756480695, "time": 0.6747206449508667, "epoch": 2, "memory": 15585, "step": 2254} +{"lr": 0.00038621161855458035, "data_time": 0.0014965057373046875, "loss": 0.009866238804534078, "time": 0.6789412260055542, "epoch": 2, "memory": 15585, "step": 2354} +{"lr": 0.0003850302758754841, "data_time": 0.0017122983932495116, "loss": 0.012863134825602173, "time": 0.6590167045593261, "epoch": 2, "memory": 15585, "step": 2454} +{"lr": 0.00038380239743559646, "data_time": 0.001634502410888672, "loss": 0.018738905992358924, "time": 0.6659116744995117, "epoch": 2, "memory": 15585, "step": 2554} +{"lr": 0.00038252830062740647, "data_time": 0.0014419078826904297, "loss": 0.012373350374400616, "time": 0.6611159086227417, "epoch": 2, "memory": 15585, "step": 2654} +{"lr": 0.00038120831479032175, "data_time": 0.0015735149383544922, "loss": 0.0252923724707216, "time": 0.7753906726837159, "epoch": 2, "memory": 15585, "step": 2754} +{"lr": 0.0003798427811255361, "data_time": 0.001440119743347168, "loss": 0.014280840009450912, "time": 0.6868777751922608, "epoch": 2, "memory": 15585, "step": 2854} +{"lr": 0.0003784320526078357, "data_time": 0.0015321731567382812, "loss": 0.012855517165735363, "time": 0.6961020469665528, "epoch": 2, "memory": 15585, "step": 2954} +{"lr": 0.0003769764938943556, "data_time": 0.0018317222595214844, "loss": 0.01653548371978104, "time": 0.761281180381775, "epoch": 2, "memory": 15585, "step": 3054} +{"lr": 0.0003754764812303242, "data_time": 0.0017846584320068359, "loss": 0.007771245110780001, "time": 0.6288684606552124, "epoch": 2, "memory": 15585, "step": 3154} +{"lr": 0.0003739324023518048, "data_time": 0.0016299247741699218, "loss": 0.014337855158373713, "time": 0.6918784141540527, "epoch": 2, "memory": 15585, "step": 3254} +{"lr": 0.0003723446563854724, "data_time": 0.0015499114990234375, "loss": 0.01722486559301615, "time": 0.706419825553894, "epoch": 2, "memory": 15585, "step": 3354} +{"lr": 0.0003707136537454424, "data_time": 0.0019171476364135743, "loss": 0.011701448215171694, "time": 0.6243953466415405, "epoch": 2, "memory": 15585, "step": 3454} +{"lr": 0.0003690398160271841, "data_time": 0.0018057107925415039, "loss": 0.011284441966563464, "time": 0.6623225927352905, "epoch": 2, "memory": 15585, "step": 3554} +{"lr": 0.00036732357589854426, "data_time": 0.0015108346939086913, "loss": 0.014126776717603207, "time": 0.6861158609390259, "epoch": 2, "memory": 15585, "step": 3654} +{"lr": 0.00036556537698790475, "data_time": 0.0016100406646728516, "loss": 0.019877943117171525, "time": 0.6884613275527954, "epoch": 2, "memory": 15585, "step": 3754} +{"lr": 0.0003637656737695118, "data_time": 0.001631951332092285, "loss": 0.015939497854560614, "time": 0.6977732419967652, "epoch": 2, "memory": 15585, "step": 3854} +{"accuracy/top1": 92.28280639648438, "data_time": 0.2989017645517985, "time": 0.6843257665634155, "step": 2} +{"lr": 0.00036091403361175644, "data_time": 0.0018541336059570313, "loss": 0.017410039575770496, "time": 0.6663965940475464, "epoch": 3, "memory": 15585, "step": 4008} +{"lr": 0.00035901102535335625, "data_time": 0.001563119888305664, "loss": 0.02098084785975516, "time": 0.6456099271774292, "epoch": 3, "memory": 15585, "step": 4108} +{"lr": 0.00035706820701190153, "data_time": 0.0016266822814941407, "loss": 0.00888769249431789, "time": 0.6946443080902099, "epoch": 3, "memory": 15585, "step": 4208} +{"lr": 0.00035508608078364215, "data_time": 0.001626133918762207, "loss": 0.013776288786903023, "time": 0.7003012657165527, "epoch": 3, "memory": 15585, "step": 4308} +{"lr": 0.00035306515902546755, "data_time": 0.0015978336334228516, "loss": 0.01207076939754188, "time": 0.6527318716049194, "epoch": 3, "memory": 15585, "step": 4408} +{"lr": 0.0003510059641224655, "data_time": 0.0016560792922973634, "loss": 0.008928650245070457, "time": 0.6449427604675293, "epoch": 3, "memory": 15585, "step": 4508} +{"lr": 0.00034890902835289306, "data_time": 0.0015845537185668946, "loss": 0.015347963199019432, "time": 0.6566363096237182, "epoch": 3, "memory": 15585, "step": 4608} +{"lr": 0.00034677489375058753, "data_time": 0.0014883756637573242, "loss": 0.01160858809016645, "time": 0.7226941347122192, "epoch": 3, "memory": 15585, "step": 4708} +{"lr": 0.00034460411196485875, "data_time": 0.001604771614074707, "loss": 0.013774989219382406, "time": 0.714819860458374, "epoch": 3, "memory": 15585, "step": 4808} +{"lr": 0.0003423972441178926, "data_time": 0.001563715934753418, "loss": 0.012805009679868817, "time": 0.6751408100128173, "epoch": 3, "memory": 15585, "step": 4908} +{"lr": 0.0003401548606597075, "data_time": 0.001758885383605957, "loss": 0.010259871510788799, "time": 0.6597633600234986, "epoch": 3, "memory": 15585, "step": 5008} +{"lr": 0.0003378775412207003, "data_time": 0.0016581058502197266, "loss": 0.01333591747097671, "time": 0.6758508443832397, "epoch": 3, "memory": 15585, "step": 5108} +{"lr": 0.0003355658744618178, "data_time": 0.0017284870147705079, "loss": 0.013544637709856033, "time": 0.6969796180725097, "epoch": 3, "memory": 15585, "step": 5208} +{"lr": 0.0003332204579223944, "data_time": 0.0018582582473754884, "loss": 0.007732429029420018, "time": 0.6977839469909668, "epoch": 3, "memory": 15585, "step": 5308} +{"lr": 0.0003308418978656968, "data_time": 0.0017543077468872071, "loss": 0.015374886943027377, "time": 0.6721315860748291, "epoch": 3, "memory": 15585, "step": 5408} +{"lr": 0.0003284308091222099, "data_time": 0.0019192934036254884, "loss": 0.013944644061848522, "time": 0.6913657426834107, "epoch": 3, "memory": 15585, "step": 5508} +{"lr": 0.0003259878149307108, "data_time": 0.0017811775207519532, "loss": 0.019178403494879603, "time": 0.6557977437973023, "epoch": 3, "memory": 15585, "step": 5608} +{"lr": 0.0003235135467771688, "data_time": 0.002123856544494629, "loss": 0.012260586582124233, "time": 0.6827628135681152, "epoch": 3, "memory": 15585, "step": 5708} +{"lr": 0.0003210086442315141, "data_time": 0.0016324281692504882, "loss": 0.014499997021630406, "time": 0.651843810081482, "epoch": 3, "memory": 15585, "step": 5808} +{"accuracy/top1": 96.84770965576172, "data_time": 0.33346123695373536, "time": 0.7234560966491699, "step": 3} +{"lr": 0.00031709267628953966, "data_time": 0.0019713878631591798, "loss": 0.011463824100792408, "time": 0.7409474134445191, "epoch": 4, "memory": 15585, "step": 5962} +{"lr": 0.00031451289120303685, "data_time": 0.0015260696411132813, "loss": 0.013445825176313519, "time": 0.6702078342437744, "epoch": 4, "memory": 15585, "step": 6062} +{"lr": 0.000311904798290621, "data_time": 0.0017065763473510741, "loss": 0.007262719282880426, "time": 0.6307323694229126, "epoch": 4, "memory": 15585, "step": 6162} +{"lr": 0.00030926907171438685, "data_time": 0.0015701055526733398, "loss": 0.009844586765393615, "time": 0.727361798286438, "epoch": 4, "memory": 15585, "step": 6262} +{"lr": 0.000306606392779414, "data_time": 0.0017295122146606446, "loss": 0.014835256198421121, "time": 0.6559880495071411, "epoch": 4, "memory": 15585, "step": 6362} +{"lr": 0.0003039174497576579, "data_time": 0.0016888856887817382, "loss": 0.017229304742068054, "time": 0.6472847700119019, "epoch": 4, "memory": 15585, "step": 6462} +{"lr": 0.0003012029377100387, "data_time": 0.0016455411911010741, "loss": 0.010780877992510795, "time": 0.6559002161026001, "epoch": 4, "memory": 15585, "step": 6562} +{"lr": 0.00029846355830677674, "data_time": 0.00140380859375, "loss": 0.014993917709216476, "time": 0.6517593622207641, "epoch": 4, "memory": 15585, "step": 6662} +{"lr": 0.0002957000196460183, "data_time": 0.001419973373413086, "loss": 0.00921645383350551, "time": 0.6484264135360718, "epoch": 4, "memory": 15585, "step": 6762} +{"lr": 0.0002929130360708003, "data_time": 0.0013887405395507813, "loss": 0.010620530089363455, "time": 0.7805695772171021, "epoch": 4, "memory": 15585, "step": 6862} +{"lr": 0.0002901033279844016, "data_time": 0.0013953208923339843, "loss": 0.011406289180740714, "time": 0.6898817777633667, "epoch": 4, "memory": 15585, "step": 6962} +{"lr": 0.00028727162166412623, "data_time": 0.001413869857788086, "loss": 0.013239658391103148, "time": 0.651923656463623, "epoch": 4, "memory": 15585, "step": 7062} +{"lr": 0.000284418649073569, "data_time": 0.001578044891357422, "loss": 0.011714322678744792, "time": 0.6677604436874389, "epoch": 4, "memory": 15585, "step": 7162} +{"lr": 0.00028154514767341203, "data_time": 0.0015402555465698243, "loss": 0.008353339042514563, "time": 0.6376981735229492, "epoch": 4, "memory": 15585, "step": 7262} +{"lr": 0.0002786518602307988, "data_time": 0.014878654479980468, "loss": 0.00838736267760396, "time": 0.5351914405822754, "epoch": 4, "memory": 15585, "step": 7362} +{"lr": 0.0002757395346273384, "data_time": 0.0018158674240112305, "loss": 0.009013206278905273, "time": 0.6818314313888549, "epoch": 4, "memory": 15585, "step": 7462} +{"lr": 0.0002728089236657853, "data_time": 0.0016567707061767578, "loss": 0.005488904332742095, "time": 0.651411771774292, "epoch": 4, "memory": 15585, "step": 7562} +{"lr": 0.0002698607848754497, "data_time": 0.001488065719604492, "loss": 0.01764683909714222, "time": 0.6485535860061645, "epoch": 4, "memory": 15585, "step": 7662} +{"lr": 0.0002668958803163847, "data_time": 0.001552414894104004, "loss": 0.009556830395013094, "time": 0.6372662782669067, "epoch": 4, "memory": 15585, "step": 7762} +{"accuracy/top1": 98.52823638916016, "data_time": 0.27475206057230633, "time": 0.6601353089014689, "step": 4} +{"lr": 0.00026229890681650394, "data_time": 0.0017865657806396484, "loss": 0.004831953858956695, "time": 0.6194799661636352, "epoch": 5, "memory": 15585, "step": 7916} +{"lr": 0.00025929487224058475, "data_time": 0.001386880874633789, "loss": 0.013389514992013574, "time": 0.6084630250930786, "epoch": 5, "memory": 15585, "step": 8016} +{"lr": 0.0002562768030631272, "data_time": 0.0016942501068115234, "loss": 0.012080180924385786, "time": 0.5956696271896362, "epoch": 5, "memory": 15585, "step": 8116} +{"lr": 0.0002532454794203871, "data_time": 0.0017185688018798828, "loss": 0.009078635927289724, "time": 0.6397946357727051, "epoch": 5, "memory": 15585, "step": 8216} +{"lr": 0.0002502016848747476, "data_time": 0.0019526481628417969, "loss": 0.005615663947537541, "time": 0.6454926252365112, "epoch": 5, "memory": 15585, "step": 8316} +{"lr": 0.0002471462062121758, "data_time": 0.001575136184692383, "loss": 0.009888977743685246, "time": 0.5959043264389038, "epoch": 5, "memory": 15585, "step": 8416} +{"lr": 0.00024407983323884952, "data_time": 0.0016025781631469726, "loss": 0.00912858285009861, "time": 0.7090702056884766, "epoch": 5, "memory": 15585, "step": 8516} +{"lr": 0.0002410033585770006, "data_time": 0.0015970706939697266, "loss": 0.013221773970872163, "time": 0.6584749937057495, "epoch": 5, "memory": 15585, "step": 8616} +{"lr": 0.00023791757746003292, "data_time": 0.0015449762344360352, "loss": 0.005049741407856345, "time": 0.6161842584609986, "epoch": 5, "memory": 15585, "step": 8716} +{"lr": 0.00023482328752696182, "data_time": 0.0015813827514648438, "loss": 0.011168416449800133, "time": 0.6581512212753295, "epoch": 5, "memory": 15585, "step": 8816} +{"lr": 0.00023172128861623345, "data_time": 0.0015157938003540039, "loss": 0.0067132618278265, "time": 0.65101478099823, "epoch": 5, "memory": 15585, "step": 8916} +{"lr": 0.000228612382558976, "data_time": 0.0014892816543579102, "loss": 0.00811783829703927, "time": 0.5933526754379272, "epoch": 5, "memory": 15585, "step": 9016} +{"lr": 0.00022549737297173692, "data_time": 0.0015125274658203125, "loss": 0.008906740183010697, "time": 0.6140785932540893, "epoch": 5, "memory": 15585, "step": 9116} +{"lr": 0.00022237706504875477, "data_time": 0.0016047954559326172, "loss": 0.00877317083068192, "time": 0.6153039693832397, "epoch": 5, "memory": 15585, "step": 9216} +{"lr": 0.00021925226535382818, "data_time": 0.0014318227767944336, "loss": 0.005616085045039654, "time": 0.6370318174362183, "epoch": 5, "memory": 15585, "step": 9316} +{"lr": 0.0002161237816118266, "data_time": 0.001545548439025879, "loss": 0.006157375546172262, "time": 0.6626938343048095, "epoch": 5, "memory": 15585, "step": 9416} +{"lr": 0.00021299242249990383, "data_time": 0.0014414072036743164, "loss": 0.012326659588143229, "time": 0.6244205236434937, "epoch": 5, "memory": 15585, "step": 9516} +{"lr": 0.00020985899743846363, "data_time": 0.0014847517013549805, "loss": 0.004703784687444567, "time": 0.656559681892395, "epoch": 5, "memory": 15585, "step": 9616} +{"lr": 0.00020672431638193363, "data_time": 0.0014258623123168945, "loss": 0.009702467499300838, "time": 0.6219302415847778, "epoch": 5, "memory": 15585, "step": 9716} +{"accuracy/top1": 99.29717254638672, "data_time": 0.23988401095072429, "time": 0.6236536820729573, "step": 5} +{"lr": 0.00020189632111931318, "data_time": 0.00169219970703125, "loss": 0.01668799272738397, "time": 0.6779077529907227, "epoch": 6, "memory": 15585, "step": 9870} +{"lr": 0.00019876209288860824, "data_time": 0.0018379926681518556, "loss": 0.00760868419893086, "time": 0.6568680286407471, "epoch": 6, "memory": 15585, "step": 9970} +{"lr": 0.00019562947708532286, "data_time": 0.001674056053161621, "loss": 0.0062294354196637865, "time": 0.6225027561187744, "epoch": 6, "memory": 15585, "step": 10070} +{"lr": 0.00019249928345470104, "data_time": 0.001777315139770508, "loss": 0.006567248143255711, "time": 0.6651949882507324, "epoch": 6, "memory": 15585, "step": 10170} +{"lr": 0.00018937232111588298, "data_time": 0.0018759965896606445, "loss": 0.009640007140114903, "time": 0.6640912532806397, "epoch": 6, "memory": 15585, "step": 10270} +{"lr": 0.000186249398352757, "data_time": 0.0016901254653930663, "loss": 0.007108497293666005, "time": 0.6892532587051392, "epoch": 6, "memory": 15585, "step": 10370} +{"lr": 0.0001831313224050277, "data_time": 0.0019468069076538086, "loss": 0.01266483818180859, "time": 0.6900137186050415, "epoch": 6, "memory": 15585, "step": 10470} +{"lr": 0.00018001889925955291, "data_time": 0.0017454862594604493, "loss": 0.00954903052188456, "time": 0.6614962577819824, "epoch": 6, "memory": 15585, "step": 10570} +{"lr": 0.00017691293344200646, "data_time": 0.0015995025634765625, "loss": 0.005812606262043119, "time": 0.5997912645339966, "epoch": 6, "memory": 15585, "step": 10670} +{"lr": 0.00017381422780891689, "data_time": 0.0016822576522827148, "loss": 0.012861727038398385, "time": 0.5972685813903809, "epoch": 6, "memory": 15585, "step": 10770} +{"lr": 0.00017072358334013806, "data_time": 0.001943826675415039, "loss": 0.006443824712187051, "time": 0.5993451356887818, "epoch": 6, "memory": 15585, "step": 10870} +{"lr": 0.00016764179893180575, "data_time": 0.002106022834777832, "loss": 0.007273007370531559, "time": 0.6282766342163086, "epoch": 6, "memory": 15585, "step": 10970} +{"lr": 0.00016456967118983176, "data_time": 0.0017577886581420898, "loss": 0.003482454409822822, "time": 0.6453614711761475, "epoch": 6, "memory": 15585, "step": 11070} +{"lr": 0.00016150799422398984, "data_time": 0.0016405582427978516, "loss": 0.010242156451568007, "time": 0.6361419200897217, "epoch": 6, "memory": 15585, "step": 11170} +{"lr": 0.0001584575594426489, "data_time": 0.00153658390045166, "loss": 0.010424668900668621, "time": 0.6031476736068726, "epoch": 6, "memory": 15585, "step": 11270} +{"lr": 0.00015541915534820135, "data_time": 0.0021350860595703127, "loss": 0.008785891905426979, "time": 0.6475701570510864, "epoch": 6, "memory": 15585, "step": 11370} +{"lr": 0.00015239356733324542, "data_time": 0.001819014549255371, "loss": 0.006363527057692409, "time": 0.630529260635376, "epoch": 6, "memory": 15585, "step": 11470} +{"lr": 0.00014938157747756997, "data_time": 0.002148723602294922, "loss": 0.0067242414224892855, "time": 0.6221449851989747, "epoch": 6, "memory": 15585, "step": 11570} +{"lr": 0.00014638396434599565, "data_time": 0.0022005558013916014, "loss": 0.01133630545809865, "time": 0.6618329763412476, "epoch": 6, "memory": 15585, "step": 11670} +{"accuracy/top1": 99.6137924194336, "data_time": 0.25692954858144124, "time": 0.6420832316080729, "step": 6} +{"lr": 0.00014179754513556176, "data_time": 0.0015227079391479492, "loss": 0.00917971059679985, "time": 0.6673883676528931, "epoch": 7, "memory": 15585, "step": 11824} +{"lr": 0.00013883992334674444, "data_time": 0.001522660255432129, "loss": 0.007058839499950409, "time": 0.614302659034729, "epoch": 7, "memory": 15585, "step": 11924} +{"lr": 0.00013589940317867467, "data_time": 0.0014207124710083007, "loss": 0.00554033750668168, "time": 0.6409608125686646, "epoch": 7, "memory": 15585, "step": 12024} +{"lr": 0.0001329767447220783, "data_time": 0.0016649723052978515, "loss": 0.0065949358977377415, "time": 0.6931212425231934, "epoch": 7, "memory": 15585, "step": 12124} +{"lr": 0.00013007270345063488, "data_time": 0.0017387151718139648, "loss": 0.005416828859597445, "time": 0.6540053844451904, "epoch": 7, "memory": 15585, "step": 12224} +{"lr": 0.00012718803002569412, "data_time": 0.001673436164855957, "loss": 0.005337166460230946, "time": 0.636088514328003, "epoch": 7, "memory": 15585, "step": 12324} +{"lr": 0.0001243234701022397, "data_time": 0.0017577886581420898, "loss": 0.006841531582176685, "time": 0.6361660242080689, "epoch": 7, "memory": 15585, "step": 12424} +{"lr": 0.00012147976413614597, "data_time": 0.001942896842956543, "loss": 0.00278859231621027, "time": 0.6669306993484497, "epoch": 7, "memory": 15585, "step": 12524} +{"lr": 0.0001186576471927784, "data_time": 0.001684856414794922, "loss": 0.004544019978493452, "time": 0.6646654605865479, "epoch": 7, "memory": 15585, "step": 12624} +{"lr": 0.00011585784875698867, "data_time": 0.0018298625946044922, "loss": 0.00843923557549715, "time": 0.6909766435623169, "epoch": 7, "memory": 15585, "step": 12724} +{"lr": 0.00011308109254454993, "data_time": 0.001630091667175293, "loss": 0.0047781896777451035, "time": 0.6773380756378173, "epoch": 7, "memory": 15585, "step": 12824} +{"lr": 0.00011032809631508466, "data_time": 0.0017049312591552734, "loss": 0.0035765465814620256, "time": 0.6646187543869019, "epoch": 7, "memory": 15585, "step": 12924} +{"lr": 0.00010759957168653253, "data_time": 0.0016542196273803711, "loss": 0.0027839860413223505, "time": 0.64542236328125, "epoch": 7, "memory": 15585, "step": 13024} +{"lr": 0.0001048962239512054, "data_time": 0.0017742633819580079, "loss": 0.005897253239527345, "time": 0.6318056583404541, "epoch": 7, "memory": 15585, "step": 13124} +{"lr": 0.00010221875189347644, "data_time": 0.0016850948333740235, "loss": 0.005267477780580521, "time": 0.6302698135375977, "epoch": 7, "memory": 15585, "step": 13224} +{"lr": 9.95678476091537e-05, "data_time": 0.001740097999572754, "loss": 0.0017812042031437158, "time": 0.6425365924835205, "epoch": 7, "memory": 15585, "step": 13324} +{"lr": 9.694419632657999e-05, "data_time": 0.0014635324478149414, "loss": 0.003032888425514102, "time": 0.6663414478302002, "epoch": 7, "memory": 15585, "step": 13424} +{"lr": 9.434847622951059e-05, "data_time": 0.0015420913696289062, "loss": 0.005051023419946432, "time": 0.7161820411682129, "epoch": 7, "memory": 15585, "step": 13524} +{"lr": 9.178135828180905e-05, "data_time": 0.001733851432800293, "loss": 0.004972005914896727, "time": 0.6929910898208618, "epoch": 7, "memory": 15585, "step": 13624} +{"accuracy/top1": 99.66946411132812, "data_time": 0.25059472719828285, "time": 0.6394131183624268, "step": 7} +{"lr": 8.788546579134055e-05, "data_time": 0.0018557071685791015, "loss": 0.0036553694866597653, "time": 0.6380128860473633, "epoch": 8, "memory": 15585, "step": 13778} +{"lr": 8.539396307206318e-05, "data_time": 0.0018084049224853516, "loss": 0.007920496724545955, "time": 0.6349304914474487, "epoch": 8, "memory": 15585, "step": 13878} +{"lr": 8.293337714106816e-05, "data_time": 0.0019393682479858399, "loss": 0.005296564102172852, "time": 0.6735837936401368, "epoch": 8, "memory": 15585, "step": 13978} +{"lr": 8.05043440315878e-05, "data_time": 0.0017738580703735352, "loss": 0.004610680788755417, "time": 0.6390577793121338, "epoch": 8, "memory": 15585, "step": 14078} +{"lr": 7.810749162081216e-05, "data_time": 0.001647043228149414, "loss": 0.006256423704326153, "time": 0.668769383430481, "epoch": 8, "memory": 15585, "step": 14178} +{"lr": 7.574343946758993e-05, "data_time": 0.0015592575073242188, "loss": 0.003835038235411048, "time": 0.6455594539642334, "epoch": 8, "memory": 15585, "step": 14278} +{"lr": 7.34127986522793e-05, "data_time": 0.0015936613082885743, "loss": 0.004211691673845053, "time": 0.6585395574569702, "epoch": 8, "memory": 15585, "step": 14378} +{"lr": 7.111617161879136e-05, "data_time": 0.0015388250350952149, "loss": 0.002990959584712982, "time": 0.6674365520477294, "epoch": 8, "memory": 15585, "step": 14478} +{"lr": 6.885415201886485e-05, "data_time": 0.0016308307647705078, "loss": 0.003187550464645028, "time": 0.6316400289535522, "epoch": 8, "memory": 15585, "step": 14578} +{"lr": 6.662732455861337e-05, "data_time": 0.002294158935546875, "loss": 0.005082053178921342, "time": 0.6623456716537476, "epoch": 8, "memory": 15585, "step": 14678} +{"lr": 6.443626484738661e-05, "data_time": 0.0016690969467163086, "loss": 0.003276269603520632, "time": 0.6347261667251587, "epoch": 8, "memory": 15585, "step": 14778} +{"lr": 6.22815392489815e-05, "data_time": 0.002061271667480469, "loss": 0.0053091333247721195, "time": 0.6474648952484131, "epoch": 8, "memory": 15585, "step": 14878} +{"lr": 6.016370473524328e-05, "data_time": 0.002399301528930664, "loss": 0.0028497880790382623, "time": 0.6367097616195678, "epoch": 8, "memory": 15585, "step": 14978} +{"lr": 5.80833087420956e-05, "data_time": 0.0017988920211791993, "loss": 0.003937024902552366, "time": 0.6971336126327514, "epoch": 8, "memory": 15585, "step": 15078} +{"lr": 5.604088902803397e-05, "data_time": 0.0022139787673950196, "loss": 0.004871996259316802, "time": 0.6896669864654541, "epoch": 8, "memory": 15585, "step": 15178} +{"lr": 5.403697353512148e-05, "data_time": 0.0018764257431030274, "loss": 0.0019157560076564551, "time": 0.6600864171981812, "epoch": 8, "memory": 15585, "step": 15278} +{"lr": 5.207208025252174e-05, "data_time": 0.00151369571685791, "loss": 0.003417341411113739, "time": 0.6410274744033814, "epoch": 8, "memory": 15585, "step": 15378} +{"lr": 5.014671708260458e-05, "data_time": 0.0017699718475341797, "loss": 0.0023052831646054985, "time": 0.6289991617202759, "epoch": 8, "memory": 15585, "step": 15478} +{"lr": 4.826138170965952e-05, "data_time": 0.0016251802444458008, "loss": 0.006209521880373358, "time": 0.6850731134414673, "epoch": 8, "memory": 15585, "step": 15578} +{"accuracy/top1": 99.75296783447266, "data_time": 0.28049888610839846, "time": 0.6655471642812093, "step": 8} +{"lr": 4.5437373040402475e-05, "data_time": 0.0017120361328125, "loss": 0.002108980808407068, "time": 0.6607366085052491, "epoch": 9, "memory": 15585, "step": 15732} +{"lr": 4.3655875036148156e-05, "data_time": 0.0015893220901489259, "loss": 0.004160146508365869, "time": 0.6440839290618896, "epoch": 9, "memory": 15585, "step": 15832} +{"lr": 4.19160826370641e-05, "data_time": 0.001586151123046875, "loss": 0.0049891692586243154, "time": 0.6651614427566528, "epoch": 9, "memory": 15585, "step": 15932} +{"lr": 4.021844555952455e-05, "data_time": 0.0015690326690673828, "loss": 0.0053192850202322004, "time": 0.6334263563156128, "epoch": 9, "memory": 15585, "step": 16032} +{"lr": 3.8563402623236864e-05, "data_time": 0.0016942739486694336, "loss": 0.004595744470134377, "time": 0.6555308818817138, "epoch": 9, "memory": 15585, "step": 16132} +{"lr": 3.695138163781135e-05, "data_time": 0.0015754938125610352, "loss": 0.003329415759071708, "time": 0.6383215665817261, "epoch": 9, "memory": 15585, "step": 16232} +{"lr": 3.538279929217756e-05, "data_time": 0.0018115520477294921, "loss": 0.003979670349508524, "time": 0.6229106187820435, "epoch": 9, "memory": 15585, "step": 16332} +{"lr": 3.385806104687531e-05, "data_time": 0.0016431093215942382, "loss": 0.0011984257958829402, "time": 0.709062147140503, "epoch": 9, "memory": 15585, "step": 16432} +{"lr": 3.237756102924734e-05, "data_time": 0.001637887954711914, "loss": 0.006779159884899855, "time": 0.6769911527633667, "epoch": 9, "memory": 15585, "step": 16532} +{"lr": 3.094168193156213e-05, "data_time": 0.0017174959182739257, "loss": 0.007550827087834477, "time": 0.6279330492019654, "epoch": 9, "memory": 15585, "step": 16632} +{"lr": 2.95507949120924e-05, "data_time": 0.001858997344970703, "loss": 0.001896716421470046, "time": 0.6590387105941773, "epoch": 9, "memory": 15585, "step": 16732} +{"lr": 2.8205259499174602e-05, "data_time": 0.001654672622680664, "loss": 0.0028586310334503652, "time": 0.6033028602600098, "epoch": 9, "memory": 15585, "step": 16832} +{"lr": 2.690542349827549e-05, "data_time": 0.001623821258544922, "loss": 0.0027841584756970405, "time": 0.6482492923736572, "epoch": 9, "memory": 15585, "step": 16932} +{"lr": 2.565162290208795e-05, "data_time": 0.0016066789627075194, "loss": 0.003248716751113534, "time": 0.6410270214080811, "epoch": 9, "memory": 15585, "step": 17032} +{"lr": 2.4444181803680975e-05, "data_time": 0.0016053199768066406, "loss": 0.0015258102677762509, "time": 0.6820625543594361, "epoch": 9, "memory": 15585, "step": 17132} +{"lr": 2.32834123127253e-05, "data_time": 0.001449418067932129, "loss": 0.0025136827025562525, "time": 0.6349185228347778, "epoch": 9, "memory": 15585, "step": 17232} +{"lr": 2.2169614474816763e-05, "data_time": 0.0015571117401123047, "loss": 0.0033571912441402673, "time": 0.65428626537323, "epoch": 9, "memory": 15585, "step": 17332} +{"lr": 2.110307619391749e-05, "data_time": 0.0014767646789550781, "loss": 0.002041183551773429, "time": 0.6806483507156372, "epoch": 9, "memory": 15585, "step": 17432} +{"lr": 2.008407315793661e-05, "data_time": 0.0015740394592285156, "loss": 0.001289188675582409, "time": 0.6655269861221313, "epoch": 9, "memory": 15585, "step": 17532} +{"accuracy/top1": 99.82603454589844, "data_time": 0.2769359906514486, "time": 0.6611462116241456, "step": 9} +{"lr": 1.8608381954463387e-05, "data_time": 0.0017873525619506836, "loss": 0.0056225445587188005, "time": 0.6363729476928711, "epoch": 10, "memory": 15585, "step": 17686} +{"lr": 1.771127410231484e-05, "data_time": 0.0017302751541137696, "loss": 0.004442027769982815, "time": 0.6548747301101685, "epoch": 10, "memory": 15585, "step": 17786} +{"lr": 1.6862578237010542e-05, "data_time": 0.0019591093063354493, "loss": 0.003528768289834261, "time": 0.651309585571289, "epoch": 10, "memory": 15585, "step": 17886} +{"lr": 1.6062513736694383e-05, "data_time": 0.0017529010772705078, "loss": 0.002310699410736561, "time": 0.6333913087844849, "epoch": 10, "memory": 15585, "step": 17986} +{"lr": 1.531128740886036e-05, "data_time": 0.0016885757446289062, "loss": 0.0018918605986982585, "time": 0.6517352104187012, "epoch": 10, "memory": 15585, "step": 18086} +{"lr": 1.4609093436895328e-05, "data_time": 0.0016204595565795898, "loss": 0.0016631382051855326, "time": 0.6436481475830078, "epoch": 10, "memory": 15585, "step": 18186} +{"lr": 1.3956113329884566e-05, "data_time": 0.0016251564025878905, "loss": 0.004168464476242662, "time": 0.6807407140731812, "epoch": 10, "memory": 15585, "step": 18286} +{"lr": 1.335251587569401e-05, "data_time": 0.001718425750732422, "loss": 0.006327252555638552, "time": 0.6579589128494263, "epoch": 10, "memory": 15585, "step": 18386} +{"lr": 1.2798457097340337e-05, "data_time": 0.001715397834777832, "loss": 0.0012945509050041437, "time": 0.6591803550720214, "epoch": 10, "memory": 15585, "step": 18486} +{"lr": 1.229408021266094e-05, "data_time": 0.001591801643371582, "loss": 0.002268029190599918, "time": 0.6328505992889404, "epoch": 10, "memory": 15585, "step": 18586} +{"lr": 1.1839515597293723e-05, "data_time": 0.001513195037841797, "loss": 0.007035813201218844, "time": 0.6506047964096069, "epoch": 10, "memory": 15585, "step": 18686} +{"lr": 1.1434880750976406e-05, "data_time": 0.0015070199966430663, "loss": 0.004653706587851048, "time": 0.6760143995285034, "epoch": 10, "memory": 15585, "step": 18786} +{"lr": 1.108028026717431e-05, "data_time": 0.0014372587203979493, "loss": 0.0030619388911873103, "time": 0.656205677986145, "epoch": 10, "memory": 15585, "step": 18886} +{"lr": 1.0775805806044138e-05, "data_time": 0.001524639129638672, "loss": 0.004282281268388033, "time": 0.6839312791824341, "epoch": 10, "memory": 15585, "step": 18986} +{"lr": 1.0521536070740776e-05, "data_time": 0.0014963388442993165, "loss": 0.004044410912320018, "time": 0.6395697832107544, "epoch": 10, "memory": 15585, "step": 19086} +{"lr": 1.03175367870735e-05, "data_time": 0.001510286331176758, "loss": 0.004684901703149081, "time": 0.6406810522079468, "epoch": 10, "memory": 15585, "step": 19186} +{"lr": 1.0163860686516628e-05, "data_time": 0.0014453887939453124, "loss": 0.002646680362522602, "time": 0.6325014114379883, "epoch": 10, "memory": 15585, "step": 19286} +{"lr": 1.0060547492578997e-05, "data_time": 0.0014615774154663086, "loss": 0.003101595398038626, "time": 0.60328049659729, "epoch": 10, "memory": 15585, "step": 19386} +{"lr": 1.0007623910535892e-05, "data_time": 0.0015476703643798827, "loss": 0.001687302952632308, "time": 0.7015183448791504, "epoch": 10, "memory": 15585, "step": 19486} +{"accuracy/top1": 99.78079986572266, "data_time": 0.2585676829020182, "time": 0.6472060362497966, "step": 10} diff --git a/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_5e-1/20230603_194605/vis_data/config.py b/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_5e-1/20230603_194605/vis_data/config.py new file mode 100644 index 0000000000000000000000000000000000000000..58e70090387bd6ad28c5168ec1be12f28a9f050d --- /dev/null +++ b/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_5e-1/20230603_194605/vis_data/config.py @@ -0,0 +1,247 @@ +optim_wrapper = dict( + optimizer=dict( + type='AdamW', + lr=0.0004, + weight_decay=0.05, + eps=1e-08, + betas=(0.9, 0.999), + _scope_='mmpretrain'), + paramwise_cfg=dict( + norm_decay_mult=0.0, + bias_decay_mult=0.0, + flat_decay_mult=0.0, + custom_keys=dict({ + '.absolute_pos_embed': dict(decay_mult=0.0), + '.relative_position_bias_table': dict(decay_mult=0.0) + })), + type='AmpOptimWrapper', + dtype='bfloat16', + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=False, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=4096) +model = dict( + type='ImageClassifier', + backbone=dict( + frozen_stages=2, + type='ConvNeXt', + arch='small', + drop_path_rate=0.4, + init_cfg=dict( + type='Pretrained', + checkpoint= + 'https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth', + prefix='backbone')), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=768, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + init_cfg=None), + init_cfg=dict( + type='TruncNormal', layer=['Conv2d', 'Linear'], std=0.02, bias=0.0), + train_cfg=None) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stablediffusionV2-1-dpmsolver-25-1m.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/cc1m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV2-1-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV2-1-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_5e-1' diff --git a/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_5e-1/20230603_194605/vis_data/events.out.tfevents.1685792780.SH-IDC1-10-140-24-100.158150.0 b/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_5e-1/20230603_194605/vis_data/events.out.tfevents.1685792780.SH-IDC1-10-140-24-100.158150.0 new file mode 100644 index 0000000000000000000000000000000000000000..5efee2cf6624adcae2716f6ee31346234114c818 --- /dev/null +++ b/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_5e-1/20230603_194605/vis_data/events.out.tfevents.1685792780.SH-IDC1-10-140-24-100.158150.0 @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:5cfc182fe819d6a3ce5321eb8d788bc847c9b05419a83c44d30a139ec3dac100 +size 1381485 diff --git a/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_5e-1/20230603_194605/vis_data/scalars.json b/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_5e-1/20230603_194605/vis_data/scalars.json new file mode 100644 index 0000000000000000000000000000000000000000..71ae044effb85dda89e836d40de0c2773ebaa641 --- /dev/null +++ b/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_5e-1/20230603_194605/vis_data/scalars.json @@ -0,0 +1,200 @@ +{"lr": 0.00039997529889041205, "data_time": 0.0016591310501098632, "loss": 0.051201630383729935, "time": 0.8002232551574707, "epoch": 1, "memory": 15587, "step": 100} +{"lr": 0.0003999002014233684, "data_time": 0.0016923189163208009, "loss": 0.04154322072863579, "time": 0.6618563413619996, "epoch": 1, "memory": 15587, "step": 200} +{"lr": 0.0003997747244903853, "data_time": 0.0017637252807617188, "loss": 0.03195568625815213, "time": 0.6556795835494995, "epoch": 1, "memory": 15587, "step": 300} +{"lr": 0.00039959890052580966, "data_time": 0.002024054527282715, "loss": 0.02430737824179232, "time": 0.7310150623321533, "epoch": 1, "memory": 15587, "step": 400} +{"lr": 0.0003993727749781197, "data_time": 0.002002573013305664, "loss": 0.039762046094983815, "time": 0.7297355175018311, "epoch": 1, "memory": 15587, "step": 500} +{"lr": 0.000399096406298175, "data_time": 0.0018381595611572266, "loss": 0.025331319961696862, "time": 0.66280677318573, "epoch": 1, "memory": 15587, "step": 600} +{"lr": 0.0003987698659241082, "data_time": 0.0017859697341918944, "loss": 0.02447650199756026, "time": 0.9731130838394165, "epoch": 1, "memory": 15587, "step": 700} +{"lr": 0.0003983932382628598, "data_time": 0.0025223731994628907, "loss": 0.02874429989606142, "time": 0.7066556215286255, "epoch": 1, "memory": 15587, "step": 800} +{"lr": 0.00039796662066835946, "data_time": 0.0018059015274047852, "loss": 0.025640539452433588, "time": 0.6873111486434936, "epoch": 1, "memory": 15587, "step": 900} +{"lr": 0.00039749012341636, "data_time": 0.0016995668411254883, "loss": 0.016293613985180855, "time": 0.6786811351776123, "epoch": 1, "memory": 15587, "step": 1000} +{"lr": 0.00039696386967593436, "data_time": 0.0017138481140136718, "loss": 0.0216406867839396, "time": 0.7678688764572144, "epoch": 1, "memory": 15587, "step": 1100} +{"lr": 0.00039638799547763706, "data_time": 0.0016362190246582032, "loss": 0.023681106604635717, "time": 0.665216326713562, "epoch": 1, "memory": 15587, "step": 1200} +{"lr": 0.0003957626496783407, "data_time": 0.0017006874084472656, "loss": 0.022824741108343004, "time": 0.7069957256317139, "epoch": 1, "memory": 15587, "step": 1300} +{"lr": 0.0003950879939227595, "data_time": 0.0016504287719726562, "loss": 0.02190188770182431, "time": 0.6563145399093628, "epoch": 1, "memory": 15587, "step": 1400} +{"lr": 0.0003943642026016663, "data_time": 0.001715540885925293, "loss": 0.019457390438765286, "time": 0.6750092029571533, "epoch": 1, "memory": 15587, "step": 1500} +{"lr": 0.0003935914628068133, "data_time": 0.001839423179626465, "loss": 0.021533434046432376, "time": 0.6075412511825562, "epoch": 1, "memory": 15587, "step": 1600} +{"lr": 0.00039276997428257205, "data_time": 0.0016817808151245116, "loss": 0.017051718244329096, "time": 0.6816421985626221, "epoch": 1, "memory": 15587, "step": 1700} +{"lr": 0.0003918999493743001, "data_time": 0.0016488313674926757, "loss": 0.0199392921756953, "time": 0.6629706621170044, "epoch": 1, "memory": 15587, "step": 1800} +{"lr": 0.00039098161297345475, "data_time": 0.0017903566360473633, "loss": 0.02119651101529598, "time": 0.6733246088027954, "epoch": 1, "memory": 15587, "step": 1900} +{"accuracy/top1": 82.07090759277344, "data_time": 0.32828064622550174, "time": 0.7906504499501196, "step": 1} +{"lr": 0.0003894734390071047, "data_time": 0.0017095327377319336, "loss": 0.012041058111935853, "time": 0.7513424634933472, "epoch": 2, "memory": 15587, "step": 2054} +{"lr": 0.0003884334872841784, "data_time": 0.0015699148178100586, "loss": 0.01816981448791921, "time": 0.6740405559539795, "epoch": 2, "memory": 15585, "step": 2154} +{"lr": 0.00038734612010935703, "data_time": 0.0018655776977539063, "loss": 0.022913835756480695, "time": 0.6747206449508667, "epoch": 2, "memory": 15585, "step": 2254} +{"lr": 0.00038621161855458035, "data_time": 0.0014965057373046875, "loss": 0.009866238804534078, "time": 0.6789412260055542, "epoch": 2, "memory": 15585, "step": 2354} +{"lr": 0.0003850302758754841, "data_time": 0.0017122983932495116, "loss": 0.012863134825602173, "time": 0.6590167045593261, "epoch": 2, "memory": 15585, "step": 2454} +{"lr": 0.00038380239743559646, "data_time": 0.001634502410888672, "loss": 0.018738905992358924, "time": 0.6659116744995117, "epoch": 2, "memory": 15585, "step": 2554} +{"lr": 0.00038252830062740647, "data_time": 0.0014419078826904297, "loss": 0.012373350374400616, "time": 0.6611159086227417, "epoch": 2, "memory": 15585, "step": 2654} +{"lr": 0.00038120831479032175, "data_time": 0.0015735149383544922, "loss": 0.0252923724707216, "time": 0.7753906726837159, "epoch": 2, "memory": 15585, "step": 2754} +{"lr": 0.0003798427811255361, "data_time": 0.001440119743347168, "loss": 0.014280840009450912, "time": 0.6868777751922608, "epoch": 2, "memory": 15585, "step": 2854} +{"lr": 0.0003784320526078357, "data_time": 0.0015321731567382812, "loss": 0.012855517165735363, "time": 0.6961020469665528, "epoch": 2, "memory": 15585, "step": 2954} +{"lr": 0.0003769764938943556, "data_time": 0.0018317222595214844, "loss": 0.01653548371978104, "time": 0.761281180381775, "epoch": 2, "memory": 15585, "step": 3054} +{"lr": 0.0003754764812303242, "data_time": 0.0017846584320068359, "loss": 0.007771245110780001, "time": 0.6288684606552124, "epoch": 2, "memory": 15585, "step": 3154} +{"lr": 0.0003739324023518048, "data_time": 0.0016299247741699218, "loss": 0.014337855158373713, "time": 0.6918784141540527, "epoch": 2, "memory": 15585, "step": 3254} +{"lr": 0.0003723446563854724, "data_time": 0.0015499114990234375, "loss": 0.01722486559301615, "time": 0.706419825553894, "epoch": 2, "memory": 15585, "step": 3354} +{"lr": 0.0003707136537454424, "data_time": 0.0019171476364135743, "loss": 0.011701448215171694, "time": 0.6243953466415405, "epoch": 2, "memory": 15585, "step": 3454} +{"lr": 0.0003690398160271841, "data_time": 0.0018057107925415039, "loss": 0.011284441966563464, "time": 0.6623225927352905, "epoch": 2, "memory": 15585, "step": 3554} +{"lr": 0.00036732357589854426, "data_time": 0.0015108346939086913, "loss": 0.014126776717603207, "time": 0.6861158609390259, "epoch": 2, "memory": 15585, "step": 3654} +{"lr": 0.00036556537698790475, "data_time": 0.0016100406646728516, "loss": 0.019877943117171525, "time": 0.6884613275527954, "epoch": 2, "memory": 15585, "step": 3754} +{"lr": 0.0003637656737695118, "data_time": 0.001631951332092285, "loss": 0.015939497854560614, "time": 0.6977732419967652, "epoch": 2, "memory": 15585, "step": 3854} +{"accuracy/top1": 92.28280639648438, "data_time": 0.2989017645517985, "time": 0.6843257665634155, "step": 2} +{"lr": 0.00036091403361175644, "data_time": 0.0018541336059570313, "loss": 0.017410039575770496, "time": 0.6663965940475464, "epoch": 3, "memory": 15585, "step": 4008} +{"lr": 0.00035901102535335625, "data_time": 0.001563119888305664, "loss": 0.02098084785975516, "time": 0.6456099271774292, "epoch": 3, "memory": 15585, "step": 4108} +{"lr": 0.00035706820701190153, "data_time": 0.0016266822814941407, "loss": 0.00888769249431789, "time": 0.6946443080902099, "epoch": 3, "memory": 15585, "step": 4208} +{"lr": 0.00035508608078364215, "data_time": 0.001626133918762207, "loss": 0.013776288786903023, "time": 0.7003012657165527, "epoch": 3, "memory": 15585, "step": 4308} +{"lr": 0.00035306515902546755, "data_time": 0.0015978336334228516, "loss": 0.01207076939754188, "time": 0.6527318716049194, "epoch": 3, "memory": 15585, "step": 4408} +{"lr": 0.0003510059641224655, "data_time": 0.0016560792922973634, "loss": 0.008928650245070457, "time": 0.6449427604675293, "epoch": 3, "memory": 15585, "step": 4508} +{"lr": 0.00034890902835289306, "data_time": 0.0015845537185668946, "loss": 0.015347963199019432, "time": 0.6566363096237182, "epoch": 3, "memory": 15585, "step": 4608} +{"lr": 0.00034677489375058753, "data_time": 0.0014883756637573242, "loss": 0.01160858809016645, "time": 0.7226941347122192, "epoch": 3, "memory": 15585, "step": 4708} +{"lr": 0.00034460411196485875, "data_time": 0.001604771614074707, "loss": 0.013774989219382406, "time": 0.714819860458374, "epoch": 3, "memory": 15585, "step": 4808} +{"lr": 0.0003423972441178926, "data_time": 0.001563715934753418, "loss": 0.012805009679868817, "time": 0.6751408100128173, "epoch": 3, "memory": 15585, "step": 4908} +{"lr": 0.0003401548606597075, "data_time": 0.001758885383605957, "loss": 0.010259871510788799, "time": 0.6597633600234986, "epoch": 3, "memory": 15585, "step": 5008} +{"lr": 0.0003378775412207003, "data_time": 0.0016581058502197266, "loss": 0.01333591747097671, "time": 0.6758508443832397, "epoch": 3, "memory": 15585, "step": 5108} +{"lr": 0.0003355658744618178, "data_time": 0.0017284870147705079, "loss": 0.013544637709856033, "time": 0.6969796180725097, "epoch": 3, "memory": 15585, "step": 5208} +{"lr": 0.0003332204579223944, "data_time": 0.0018582582473754884, "loss": 0.007732429029420018, "time": 0.6977839469909668, "epoch": 3, "memory": 15585, "step": 5308} +{"lr": 0.0003308418978656968, "data_time": 0.0017543077468872071, "loss": 0.015374886943027377, "time": 0.6721315860748291, "epoch": 3, "memory": 15585, "step": 5408} +{"lr": 0.0003284308091222099, "data_time": 0.0019192934036254884, "loss": 0.013944644061848522, "time": 0.6913657426834107, "epoch": 3, "memory": 15585, "step": 5508} +{"lr": 0.0003259878149307108, "data_time": 0.0017811775207519532, "loss": 0.019178403494879603, "time": 0.6557977437973023, "epoch": 3, "memory": 15585, "step": 5608} +{"lr": 0.0003235135467771688, "data_time": 0.002123856544494629, "loss": 0.012260586582124233, "time": 0.6827628135681152, "epoch": 3, "memory": 15585, "step": 5708} +{"lr": 0.0003210086442315141, "data_time": 0.0016324281692504882, "loss": 0.014499997021630406, "time": 0.651843810081482, "epoch": 3, "memory": 15585, "step": 5808} +{"accuracy/top1": 96.84770965576172, "data_time": 0.33346123695373536, "time": 0.7234560966491699, "step": 3} +{"lr": 0.00031709267628953966, "data_time": 0.0019713878631591798, "loss": 0.011463824100792408, "time": 0.7409474134445191, "epoch": 4, "memory": 15585, "step": 5962} +{"lr": 0.00031451289120303685, "data_time": 0.0015260696411132813, "loss": 0.013445825176313519, "time": 0.6702078342437744, "epoch": 4, "memory": 15585, "step": 6062} +{"lr": 0.000311904798290621, "data_time": 0.0017065763473510741, "loss": 0.007262719282880426, "time": 0.6307323694229126, "epoch": 4, "memory": 15585, "step": 6162} +{"lr": 0.00030926907171438685, "data_time": 0.0015701055526733398, "loss": 0.009844586765393615, "time": 0.727361798286438, "epoch": 4, "memory": 15585, "step": 6262} +{"lr": 0.000306606392779414, "data_time": 0.0017295122146606446, "loss": 0.014835256198421121, "time": 0.6559880495071411, "epoch": 4, "memory": 15585, "step": 6362} +{"lr": 0.0003039174497576579, "data_time": 0.0016888856887817382, "loss": 0.017229304742068054, "time": 0.6472847700119019, "epoch": 4, "memory": 15585, "step": 6462} +{"lr": 0.0003012029377100387, "data_time": 0.0016455411911010741, "loss": 0.010780877992510795, "time": 0.6559002161026001, "epoch": 4, "memory": 15585, "step": 6562} +{"lr": 0.00029846355830677674, "data_time": 0.00140380859375, "loss": 0.014993917709216476, "time": 0.6517593622207641, "epoch": 4, "memory": 15585, "step": 6662} +{"lr": 0.0002957000196460183, "data_time": 0.001419973373413086, "loss": 0.00921645383350551, "time": 0.6484264135360718, "epoch": 4, "memory": 15585, "step": 6762} +{"lr": 0.0002929130360708003, "data_time": 0.0013887405395507813, "loss": 0.010620530089363455, "time": 0.7805695772171021, "epoch": 4, "memory": 15585, "step": 6862} +{"lr": 0.0002901033279844016, "data_time": 0.0013953208923339843, "loss": 0.011406289180740714, "time": 0.6898817777633667, "epoch": 4, "memory": 15585, "step": 6962} +{"lr": 0.00028727162166412623, "data_time": 0.001413869857788086, "loss": 0.013239658391103148, "time": 0.651923656463623, "epoch": 4, "memory": 15585, "step": 7062} +{"lr": 0.000284418649073569, "data_time": 0.001578044891357422, "loss": 0.011714322678744792, "time": 0.6677604436874389, "epoch": 4, "memory": 15585, "step": 7162} +{"lr": 0.00028154514767341203, "data_time": 0.0015402555465698243, "loss": 0.008353339042514563, "time": 0.6376981735229492, "epoch": 4, "memory": 15585, "step": 7262} +{"lr": 0.0002786518602307988, "data_time": 0.014878654479980468, "loss": 0.00838736267760396, "time": 0.5351914405822754, "epoch": 4, "memory": 15585, "step": 7362} +{"lr": 0.0002757395346273384, "data_time": 0.0018158674240112305, "loss": 0.009013206278905273, "time": 0.6818314313888549, "epoch": 4, "memory": 15585, "step": 7462} +{"lr": 0.0002728089236657853, "data_time": 0.0016567707061767578, "loss": 0.005488904332742095, "time": 0.651411771774292, "epoch": 4, "memory": 15585, "step": 7562} +{"lr": 0.0002698607848754497, "data_time": 0.001488065719604492, "loss": 0.01764683909714222, "time": 0.6485535860061645, "epoch": 4, "memory": 15585, "step": 7662} +{"lr": 0.0002668958803163847, "data_time": 0.001552414894104004, "loss": 0.009556830395013094, "time": 0.6372662782669067, "epoch": 4, "memory": 15585, "step": 7762} +{"accuracy/top1": 98.52823638916016, "data_time": 0.27475206057230633, "time": 0.6601353089014689, "step": 4} +{"lr": 0.00026229890681650394, "data_time": 0.0017865657806396484, "loss": 0.004831953858956695, "time": 0.6194799661636352, "epoch": 5, "memory": 15585, "step": 7916} +{"lr": 0.00025929487224058475, "data_time": 0.001386880874633789, "loss": 0.013389514992013574, "time": 0.6084630250930786, "epoch": 5, "memory": 15585, "step": 8016} +{"lr": 0.0002562768030631272, "data_time": 0.0016942501068115234, "loss": 0.012080180924385786, "time": 0.5956696271896362, "epoch": 5, "memory": 15585, "step": 8116} +{"lr": 0.0002532454794203871, "data_time": 0.0017185688018798828, "loss": 0.009078635927289724, "time": 0.6397946357727051, "epoch": 5, "memory": 15585, "step": 8216} +{"lr": 0.0002502016848747476, "data_time": 0.0019526481628417969, "loss": 0.005615663947537541, "time": 0.6454926252365112, "epoch": 5, "memory": 15585, "step": 8316} +{"lr": 0.0002471462062121758, "data_time": 0.001575136184692383, "loss": 0.009888977743685246, "time": 0.5959043264389038, "epoch": 5, "memory": 15585, "step": 8416} +{"lr": 0.00024407983323884952, "data_time": 0.0016025781631469726, "loss": 0.00912858285009861, "time": 0.7090702056884766, "epoch": 5, "memory": 15585, "step": 8516} +{"lr": 0.0002410033585770006, "data_time": 0.0015970706939697266, "loss": 0.013221773970872163, "time": 0.6584749937057495, "epoch": 5, "memory": 15585, "step": 8616} +{"lr": 0.00023791757746003292, "data_time": 0.0015449762344360352, "loss": 0.005049741407856345, "time": 0.6161842584609986, "epoch": 5, "memory": 15585, "step": 8716} +{"lr": 0.00023482328752696182, "data_time": 0.0015813827514648438, "loss": 0.011168416449800133, "time": 0.6581512212753295, "epoch": 5, "memory": 15585, "step": 8816} +{"lr": 0.00023172128861623345, "data_time": 0.0015157938003540039, "loss": 0.0067132618278265, "time": 0.65101478099823, "epoch": 5, "memory": 15585, "step": 8916} +{"lr": 0.000228612382558976, "data_time": 0.0014892816543579102, "loss": 0.00811783829703927, "time": 0.5933526754379272, "epoch": 5, "memory": 15585, "step": 9016} +{"lr": 0.00022549737297173692, "data_time": 0.0015125274658203125, "loss": 0.008906740183010697, "time": 0.6140785932540893, "epoch": 5, "memory": 15585, "step": 9116} +{"lr": 0.00022237706504875477, "data_time": 0.0016047954559326172, "loss": 0.00877317083068192, "time": 0.6153039693832397, "epoch": 5, "memory": 15585, "step": 9216} +{"lr": 0.00021925226535382818, "data_time": 0.0014318227767944336, "loss": 0.005616085045039654, "time": 0.6370318174362183, "epoch": 5, "memory": 15585, "step": 9316} +{"lr": 0.0002161237816118266, "data_time": 0.001545548439025879, "loss": 0.006157375546172262, "time": 0.6626938343048095, "epoch": 5, "memory": 15585, "step": 9416} +{"lr": 0.00021299242249990383, "data_time": 0.0014414072036743164, "loss": 0.012326659588143229, "time": 0.6244205236434937, "epoch": 5, "memory": 15585, "step": 9516} +{"lr": 0.00020985899743846363, "data_time": 0.0014847517013549805, "loss": 0.004703784687444567, "time": 0.656559681892395, "epoch": 5, "memory": 15585, "step": 9616} +{"lr": 0.00020672431638193363, "data_time": 0.0014258623123168945, "loss": 0.009702467499300838, "time": 0.6219302415847778, "epoch": 5, "memory": 15585, "step": 9716} +{"accuracy/top1": 99.29717254638672, "data_time": 0.23988401095072429, "time": 0.6236536820729573, "step": 5} +{"lr": 0.00020189632111931318, "data_time": 0.00169219970703125, "loss": 0.01668799272738397, "time": 0.6779077529907227, "epoch": 6, "memory": 15585, "step": 9870} +{"lr": 0.00019876209288860824, "data_time": 0.0018379926681518556, "loss": 0.00760868419893086, "time": 0.6568680286407471, "epoch": 6, "memory": 15585, "step": 9970} +{"lr": 0.00019562947708532286, "data_time": 0.001674056053161621, "loss": 0.0062294354196637865, "time": 0.6225027561187744, "epoch": 6, "memory": 15585, "step": 10070} +{"lr": 0.00019249928345470104, "data_time": 0.001777315139770508, "loss": 0.006567248143255711, "time": 0.6651949882507324, "epoch": 6, "memory": 15585, "step": 10170} +{"lr": 0.00018937232111588298, "data_time": 0.0018759965896606445, "loss": 0.009640007140114903, "time": 0.6640912532806397, "epoch": 6, "memory": 15585, "step": 10270} +{"lr": 0.000186249398352757, "data_time": 0.0016901254653930663, "loss": 0.007108497293666005, "time": 0.6892532587051392, "epoch": 6, "memory": 15585, "step": 10370} +{"lr": 0.0001831313224050277, "data_time": 0.0019468069076538086, "loss": 0.01266483818180859, "time": 0.6900137186050415, "epoch": 6, "memory": 15585, "step": 10470} +{"lr": 0.00018001889925955291, "data_time": 0.0017454862594604493, "loss": 0.00954903052188456, "time": 0.6614962577819824, "epoch": 6, "memory": 15585, "step": 10570} +{"lr": 0.00017691293344200646, "data_time": 0.0015995025634765625, "loss": 0.005812606262043119, "time": 0.5997912645339966, "epoch": 6, "memory": 15585, "step": 10670} +{"lr": 0.00017381422780891689, "data_time": 0.0016822576522827148, "loss": 0.012861727038398385, "time": 0.5972685813903809, "epoch": 6, "memory": 15585, "step": 10770} +{"lr": 0.00017072358334013806, "data_time": 0.001943826675415039, "loss": 0.006443824712187051, "time": 0.5993451356887818, "epoch": 6, "memory": 15585, "step": 10870} +{"lr": 0.00016764179893180575, "data_time": 0.002106022834777832, "loss": 0.007273007370531559, "time": 0.6282766342163086, "epoch": 6, "memory": 15585, "step": 10970} +{"lr": 0.00016456967118983176, "data_time": 0.0017577886581420898, "loss": 0.003482454409822822, "time": 0.6453614711761475, "epoch": 6, "memory": 15585, "step": 11070} +{"lr": 0.00016150799422398984, "data_time": 0.0016405582427978516, "loss": 0.010242156451568007, "time": 0.6361419200897217, "epoch": 6, "memory": 15585, "step": 11170} +{"lr": 0.0001584575594426489, "data_time": 0.00153658390045166, "loss": 0.010424668900668621, "time": 0.6031476736068726, "epoch": 6, "memory": 15585, "step": 11270} +{"lr": 0.00015541915534820135, "data_time": 0.0021350860595703127, "loss": 0.008785891905426979, "time": 0.6475701570510864, "epoch": 6, "memory": 15585, "step": 11370} +{"lr": 0.00015239356733324542, "data_time": 0.001819014549255371, "loss": 0.006363527057692409, "time": 0.630529260635376, "epoch": 6, "memory": 15585, "step": 11470} +{"lr": 0.00014938157747756997, "data_time": 0.002148723602294922, "loss": 0.0067242414224892855, "time": 0.6221449851989747, "epoch": 6, "memory": 15585, "step": 11570} +{"lr": 0.00014638396434599565, "data_time": 0.0022005558013916014, "loss": 0.01133630545809865, "time": 0.6618329763412476, "epoch": 6, "memory": 15585, "step": 11670} +{"accuracy/top1": 99.6137924194336, "data_time": 0.25692954858144124, "time": 0.6420832316080729, "step": 6} +{"lr": 0.00014179754513556176, "data_time": 0.0015227079391479492, "loss": 0.00917971059679985, "time": 0.6673883676528931, "epoch": 7, "memory": 15585, "step": 11824} +{"lr": 0.00013883992334674444, "data_time": 0.001522660255432129, "loss": 0.007058839499950409, "time": 0.614302659034729, "epoch": 7, "memory": 15585, "step": 11924} +{"lr": 0.00013589940317867467, "data_time": 0.0014207124710083007, "loss": 0.00554033750668168, "time": 0.6409608125686646, "epoch": 7, "memory": 15585, "step": 12024} +{"lr": 0.0001329767447220783, "data_time": 0.0016649723052978515, "loss": 0.0065949358977377415, "time": 0.6931212425231934, "epoch": 7, "memory": 15585, "step": 12124} +{"lr": 0.00013007270345063488, "data_time": 0.0017387151718139648, "loss": 0.005416828859597445, "time": 0.6540053844451904, "epoch": 7, "memory": 15585, "step": 12224} +{"lr": 0.00012718803002569412, "data_time": 0.001673436164855957, "loss": 0.005337166460230946, "time": 0.636088514328003, "epoch": 7, "memory": 15585, "step": 12324} +{"lr": 0.0001243234701022397, "data_time": 0.0017577886581420898, "loss": 0.006841531582176685, "time": 0.6361660242080689, "epoch": 7, "memory": 15585, "step": 12424} +{"lr": 0.00012147976413614597, "data_time": 0.001942896842956543, "loss": 0.00278859231621027, "time": 0.6669306993484497, "epoch": 7, "memory": 15585, "step": 12524} +{"lr": 0.0001186576471927784, "data_time": 0.001684856414794922, "loss": 0.004544019978493452, "time": 0.6646654605865479, "epoch": 7, "memory": 15585, "step": 12624} +{"lr": 0.00011585784875698867, "data_time": 0.0018298625946044922, "loss": 0.00843923557549715, "time": 0.6909766435623169, "epoch": 7, "memory": 15585, "step": 12724} +{"lr": 0.00011308109254454993, "data_time": 0.001630091667175293, "loss": 0.0047781896777451035, "time": 0.6773380756378173, "epoch": 7, "memory": 15585, "step": 12824} +{"lr": 0.00011032809631508466, "data_time": 0.0017049312591552734, "loss": 0.0035765465814620256, "time": 0.6646187543869019, "epoch": 7, "memory": 15585, "step": 12924} +{"lr": 0.00010759957168653253, "data_time": 0.0016542196273803711, "loss": 0.0027839860413223505, "time": 0.64542236328125, "epoch": 7, "memory": 15585, "step": 13024} +{"lr": 0.0001048962239512054, "data_time": 0.0017742633819580079, "loss": 0.005897253239527345, "time": 0.6318056583404541, "epoch": 7, "memory": 15585, "step": 13124} +{"lr": 0.00010221875189347644, "data_time": 0.0016850948333740235, "loss": 0.005267477780580521, "time": 0.6302698135375977, "epoch": 7, "memory": 15585, "step": 13224} +{"lr": 9.95678476091537e-05, "data_time": 0.001740097999572754, "loss": 0.0017812042031437158, "time": 0.6425365924835205, "epoch": 7, "memory": 15585, "step": 13324} +{"lr": 9.694419632657999e-05, "data_time": 0.0014635324478149414, "loss": 0.003032888425514102, "time": 0.6663414478302002, "epoch": 7, "memory": 15585, "step": 13424} +{"lr": 9.434847622951059e-05, "data_time": 0.0015420913696289062, "loss": 0.005051023419946432, "time": 0.7161820411682129, "epoch": 7, "memory": 15585, "step": 13524} +{"lr": 9.178135828180905e-05, "data_time": 0.001733851432800293, "loss": 0.004972005914896727, "time": 0.6929910898208618, "epoch": 7, "memory": 15585, "step": 13624} +{"accuracy/top1": 99.66946411132812, "data_time": 0.25059472719828285, "time": 0.6394131183624268, "step": 7} +{"lr": 8.788546579134055e-05, "data_time": 0.0018557071685791015, "loss": 0.0036553694866597653, "time": 0.6380128860473633, "epoch": 8, "memory": 15585, "step": 13778} +{"lr": 8.539396307206318e-05, "data_time": 0.0018084049224853516, "loss": 0.007920496724545955, "time": 0.6349304914474487, "epoch": 8, "memory": 15585, "step": 13878} +{"lr": 8.293337714106816e-05, "data_time": 0.0019393682479858399, "loss": 0.005296564102172852, "time": 0.6735837936401368, "epoch": 8, "memory": 15585, "step": 13978} +{"lr": 8.05043440315878e-05, "data_time": 0.0017738580703735352, "loss": 0.004610680788755417, "time": 0.6390577793121338, "epoch": 8, "memory": 15585, "step": 14078} +{"lr": 7.810749162081216e-05, "data_time": 0.001647043228149414, "loss": 0.006256423704326153, "time": 0.668769383430481, "epoch": 8, "memory": 15585, "step": 14178} +{"lr": 7.574343946758993e-05, "data_time": 0.0015592575073242188, "loss": 0.003835038235411048, "time": 0.6455594539642334, "epoch": 8, "memory": 15585, "step": 14278} +{"lr": 7.34127986522793e-05, "data_time": 0.0015936613082885743, "loss": 0.004211691673845053, "time": 0.6585395574569702, "epoch": 8, "memory": 15585, "step": 14378} +{"lr": 7.111617161879136e-05, "data_time": 0.0015388250350952149, "loss": 0.002990959584712982, "time": 0.6674365520477294, "epoch": 8, "memory": 15585, "step": 14478} +{"lr": 6.885415201886485e-05, "data_time": 0.0016308307647705078, "loss": 0.003187550464645028, "time": 0.6316400289535522, "epoch": 8, "memory": 15585, "step": 14578} +{"lr": 6.662732455861337e-05, "data_time": 0.002294158935546875, "loss": 0.005082053178921342, "time": 0.6623456716537476, "epoch": 8, "memory": 15585, "step": 14678} +{"lr": 6.443626484738661e-05, "data_time": 0.0016690969467163086, "loss": 0.003276269603520632, "time": 0.6347261667251587, "epoch": 8, "memory": 15585, "step": 14778} +{"lr": 6.22815392489815e-05, "data_time": 0.002061271667480469, "loss": 0.0053091333247721195, "time": 0.6474648952484131, "epoch": 8, "memory": 15585, "step": 14878} +{"lr": 6.016370473524328e-05, "data_time": 0.002399301528930664, "loss": 0.0028497880790382623, "time": 0.6367097616195678, "epoch": 8, "memory": 15585, "step": 14978} +{"lr": 5.80833087420956e-05, "data_time": 0.0017988920211791993, "loss": 0.003937024902552366, "time": 0.6971336126327514, "epoch": 8, "memory": 15585, "step": 15078} +{"lr": 5.604088902803397e-05, "data_time": 0.0022139787673950196, "loss": 0.004871996259316802, "time": 0.6896669864654541, "epoch": 8, "memory": 15585, "step": 15178} +{"lr": 5.403697353512148e-05, "data_time": 0.0018764257431030274, "loss": 0.0019157560076564551, "time": 0.6600864171981812, "epoch": 8, "memory": 15585, "step": 15278} +{"lr": 5.207208025252174e-05, "data_time": 0.00151369571685791, "loss": 0.003417341411113739, "time": 0.6410274744033814, "epoch": 8, "memory": 15585, "step": 15378} +{"lr": 5.014671708260458e-05, "data_time": 0.0017699718475341797, "loss": 0.0023052831646054985, "time": 0.6289991617202759, "epoch": 8, "memory": 15585, "step": 15478} +{"lr": 4.826138170965952e-05, "data_time": 0.0016251802444458008, "loss": 0.006209521880373358, "time": 0.6850731134414673, "epoch": 8, "memory": 15585, "step": 15578} +{"accuracy/top1": 99.75296783447266, "data_time": 0.28049888610839846, "time": 0.6655471642812093, "step": 8} +{"lr": 4.5437373040402475e-05, "data_time": 0.0017120361328125, "loss": 0.002108980808407068, "time": 0.6607366085052491, "epoch": 9, "memory": 15585, "step": 15732} +{"lr": 4.3655875036148156e-05, "data_time": 0.0015893220901489259, "loss": 0.004160146508365869, "time": 0.6440839290618896, "epoch": 9, "memory": 15585, "step": 15832} +{"lr": 4.19160826370641e-05, "data_time": 0.001586151123046875, "loss": 0.0049891692586243154, "time": 0.6651614427566528, "epoch": 9, "memory": 15585, "step": 15932} +{"lr": 4.021844555952455e-05, "data_time": 0.0015690326690673828, "loss": 0.0053192850202322004, "time": 0.6334263563156128, "epoch": 9, "memory": 15585, "step": 16032} +{"lr": 3.8563402623236864e-05, "data_time": 0.0016942739486694336, "loss": 0.004595744470134377, "time": 0.6555308818817138, "epoch": 9, "memory": 15585, "step": 16132} +{"lr": 3.695138163781135e-05, "data_time": 0.0015754938125610352, "loss": 0.003329415759071708, "time": 0.6383215665817261, "epoch": 9, "memory": 15585, "step": 16232} +{"lr": 3.538279929217756e-05, "data_time": 0.0018115520477294921, "loss": 0.003979670349508524, "time": 0.6229106187820435, "epoch": 9, "memory": 15585, "step": 16332} +{"lr": 3.385806104687531e-05, "data_time": 0.0016431093215942382, "loss": 0.0011984257958829402, "time": 0.709062147140503, "epoch": 9, "memory": 15585, "step": 16432} +{"lr": 3.237756102924734e-05, "data_time": 0.001637887954711914, "loss": 0.006779159884899855, "time": 0.6769911527633667, "epoch": 9, "memory": 15585, "step": 16532} +{"lr": 3.094168193156213e-05, "data_time": 0.0017174959182739257, "loss": 0.007550827087834477, "time": 0.6279330492019654, "epoch": 9, "memory": 15585, "step": 16632} +{"lr": 2.95507949120924e-05, "data_time": 0.001858997344970703, "loss": 0.001896716421470046, "time": 0.6590387105941773, "epoch": 9, "memory": 15585, "step": 16732} +{"lr": 2.8205259499174602e-05, "data_time": 0.001654672622680664, "loss": 0.0028586310334503652, "time": 0.6033028602600098, "epoch": 9, "memory": 15585, "step": 16832} +{"lr": 2.690542349827549e-05, "data_time": 0.001623821258544922, "loss": 0.0027841584756970405, "time": 0.6482492923736572, "epoch": 9, "memory": 15585, "step": 16932} +{"lr": 2.565162290208795e-05, "data_time": 0.0016066789627075194, "loss": 0.003248716751113534, "time": 0.6410270214080811, "epoch": 9, "memory": 15585, "step": 17032} +{"lr": 2.4444181803680975e-05, "data_time": 0.0016053199768066406, "loss": 0.0015258102677762509, "time": 0.6820625543594361, "epoch": 9, "memory": 15585, "step": 17132} +{"lr": 2.32834123127253e-05, "data_time": 0.001449418067932129, "loss": 0.0025136827025562525, "time": 0.6349185228347778, "epoch": 9, "memory": 15585, "step": 17232} +{"lr": 2.2169614474816763e-05, "data_time": 0.0015571117401123047, "loss": 0.0033571912441402673, "time": 0.65428626537323, "epoch": 9, "memory": 15585, "step": 17332} +{"lr": 2.110307619391749e-05, "data_time": 0.0014767646789550781, "loss": 0.002041183551773429, "time": 0.6806483507156372, "epoch": 9, "memory": 15585, "step": 17432} +{"lr": 2.008407315793661e-05, "data_time": 0.0015740394592285156, "loss": 0.001289188675582409, "time": 0.6655269861221313, "epoch": 9, "memory": 15585, "step": 17532} +{"accuracy/top1": 99.82603454589844, "data_time": 0.2769359906514486, "time": 0.6611462116241456, "step": 9} +{"lr": 1.8608381954463387e-05, "data_time": 0.0017873525619506836, "loss": 0.0056225445587188005, "time": 0.6363729476928711, "epoch": 10, "memory": 15585, "step": 17686} +{"lr": 1.771127410231484e-05, "data_time": 0.0017302751541137696, "loss": 0.004442027769982815, "time": 0.6548747301101685, "epoch": 10, "memory": 15585, "step": 17786} +{"lr": 1.6862578237010542e-05, "data_time": 0.0019591093063354493, "loss": 0.003528768289834261, "time": 0.651309585571289, "epoch": 10, "memory": 15585, "step": 17886} +{"lr": 1.6062513736694383e-05, "data_time": 0.0017529010772705078, "loss": 0.002310699410736561, "time": 0.6333913087844849, "epoch": 10, "memory": 15585, "step": 17986} +{"lr": 1.531128740886036e-05, "data_time": 0.0016885757446289062, "loss": 0.0018918605986982585, "time": 0.6517352104187012, "epoch": 10, "memory": 15585, "step": 18086} +{"lr": 1.4609093436895328e-05, "data_time": 0.0016204595565795898, "loss": 0.0016631382051855326, "time": 0.6436481475830078, "epoch": 10, "memory": 15585, "step": 18186} +{"lr": 1.3956113329884566e-05, "data_time": 0.0016251564025878905, "loss": 0.004168464476242662, "time": 0.6807407140731812, "epoch": 10, "memory": 15585, "step": 18286} +{"lr": 1.335251587569401e-05, "data_time": 0.001718425750732422, "loss": 0.006327252555638552, "time": 0.6579589128494263, "epoch": 10, "memory": 15585, "step": 18386} +{"lr": 1.2798457097340337e-05, "data_time": 0.001715397834777832, "loss": 0.0012945509050041437, "time": 0.6591803550720214, "epoch": 10, "memory": 15585, "step": 18486} +{"lr": 1.229408021266094e-05, "data_time": 0.001591801643371582, "loss": 0.002268029190599918, "time": 0.6328505992889404, "epoch": 10, "memory": 15585, "step": 18586} +{"lr": 1.1839515597293723e-05, "data_time": 0.001513195037841797, "loss": 0.007035813201218844, "time": 0.6506047964096069, "epoch": 10, "memory": 15585, "step": 18686} +{"lr": 1.1434880750976406e-05, "data_time": 0.0015070199966430663, "loss": 0.004653706587851048, "time": 0.6760143995285034, "epoch": 10, "memory": 15585, "step": 18786} +{"lr": 1.108028026717431e-05, "data_time": 0.0014372587203979493, "loss": 0.0030619388911873103, "time": 0.656205677986145, "epoch": 10, "memory": 15585, "step": 18886} +{"lr": 1.0775805806044138e-05, "data_time": 0.001524639129638672, "loss": 0.004282281268388033, "time": 0.6839312791824341, "epoch": 10, "memory": 15585, "step": 18986} +{"lr": 1.0521536070740776e-05, "data_time": 0.0014963388442993165, "loss": 0.004044410912320018, "time": 0.6395697832107544, "epoch": 10, "memory": 15585, "step": 19086} +{"lr": 1.03175367870735e-05, "data_time": 0.001510286331176758, "loss": 0.004684901703149081, "time": 0.6406810522079468, "epoch": 10, "memory": 15585, "step": 19186} +{"lr": 1.0163860686516628e-05, "data_time": 0.0014453887939453124, "loss": 0.002646680362522602, "time": 0.6325014114379883, "epoch": 10, "memory": 15585, "step": 19286} +{"lr": 1.0060547492578997e-05, "data_time": 0.0014615774154663086, "loss": 0.003101595398038626, "time": 0.60328049659729, "epoch": 10, "memory": 15585, "step": 19386} +{"lr": 1.0007623910535892e-05, "data_time": 0.0015476703643798827, "loss": 0.001687302952632308, "time": 0.7015183448791504, "epoch": 10, "memory": 15585, "step": 19486} +{"accuracy/top1": 99.78079986572266, "data_time": 0.2585676829020182, "time": 0.6472060362497966, "step": 10} diff --git a/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_5e-1/20230603_194605/vis_data/vis_image/0000000.png_1.png b/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_5e-1/20230603_194605/vis_data/vis_image/0000000.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..e8a3e067c83a1e195d15aea284df1a8be1e26ab6 Binary files /dev/null and b/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_5e-1/20230603_194605/vis_data/vis_image/0000000.png_1.png differ diff --git a/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_5e-1/20230603_194605/vis_data/vis_image/0000000.png_10.png b/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_5e-1/20230603_194605/vis_data/vis_image/0000000.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..fa34a215ce9150a571447e3abe3b5d2f66b94ffa Binary files /dev/null and b/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_5e-1/20230603_194605/vis_data/vis_image/0000000.png_10.png differ diff --git a/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_5e-1/20230603_194605/vis_data/vis_image/0000000.png_2.png b/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_5e-1/20230603_194605/vis_data/vis_image/0000000.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..35638cdd4c7eeb2220de2420be2d7f6b66ef92e4 Binary files /dev/null and b/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_5e-1/20230603_194605/vis_data/vis_image/0000000.png_2.png differ diff --git a/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_5e-1/20230603_194605/vis_data/vis_image/0000000.png_3.png b/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_5e-1/20230603_194605/vis_data/vis_image/0000000.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..6d29e7da37d2eec708bed0b2898a1786e41963f3 Binary files /dev/null and b/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_5e-1/20230603_194605/vis_data/vis_image/0000000.png_3.png differ diff --git a/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_5e-1/20230603_194605/vis_data/vis_image/0000000.png_4.png b/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_5e-1/20230603_194605/vis_data/vis_image/0000000.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..343d8895138775cb96c05524788a5aa224fec9f9 Binary files /dev/null and b/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_5e-1/20230603_194605/vis_data/vis_image/0000000.png_4.png differ diff --git a/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_5e-1/20230603_194605/vis_data/vis_image/0000000.png_5.png b/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_5e-1/20230603_194605/vis_data/vis_image/0000000.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..40c3325aeeb766de62c9ed62d08849282373a8a6 Binary files /dev/null and b/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_5e-1/20230603_194605/vis_data/vis_image/0000000.png_5.png differ diff --git a/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_5e-1/20230603_194605/vis_data/vis_image/0000000.png_6.png b/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_5e-1/20230603_194605/vis_data/vis_image/0000000.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..270cb1c072922a077bb698e08ac03b94a0f1ae8a Binary files /dev/null and b/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_5e-1/20230603_194605/vis_data/vis_image/0000000.png_6.png differ diff --git a/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_5e-1/20230603_194605/vis_data/vis_image/0000000.png_7.png b/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_5e-1/20230603_194605/vis_data/vis_image/0000000.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..e49d6d66f3b3b1f4137ba76a8c8d9028b5233627 Binary files /dev/null and b/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_5e-1/20230603_194605/vis_data/vis_image/0000000.png_7.png differ diff --git a/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_5e-1/20230603_194605/vis_data/vis_image/0000000.png_8.png b/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_5e-1/20230603_194605/vis_data/vis_image/0000000.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..78e68522f385892e8f841756f5a4cc2fb268fa4d Binary files /dev/null and b/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_5e-1/20230603_194605/vis_data/vis_image/0000000.png_8.png differ diff --git a/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_5e-1/20230603_194605/vis_data/vis_image/0000000.png_9.png b/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_5e-1/20230603_194605/vis_data/vis_image/0000000.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..989b39b33c9da491bcecf6899d0fbd31cc3164f2 Binary files /dev/null and b/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_5e-1/20230603_194605/vis_data/vis_image/0000000.png_9.png differ diff --git a/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_5e-1/20230603_194605/vis_data/vis_image/000005174.jpg_1.png b/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_5e-1/20230603_194605/vis_data/vis_image/000005174.jpg_1.png new file mode 100644 index 0000000000000000000000000000000000000000..4760c347a8792b4c64c60c73c5076f633b8aae08 Binary files /dev/null and b/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_5e-1/20230603_194605/vis_data/vis_image/000005174.jpg_1.png differ diff --git a/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_5e-1/20230603_194605/vis_data/vis_image/000005174.jpg_10.png b/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_5e-1/20230603_194605/vis_data/vis_image/000005174.jpg_10.png new file mode 100644 index 0000000000000000000000000000000000000000..6bae13f827674db9fa2c8156b5e2f9364ffbad8b Binary files /dev/null and b/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_5e-1/20230603_194605/vis_data/vis_image/000005174.jpg_10.png differ diff --git a/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_5e-1/20230603_194605/vis_data/vis_image/000005174.jpg_2.png b/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_5e-1/20230603_194605/vis_data/vis_image/000005174.jpg_2.png new file mode 100644 index 0000000000000000000000000000000000000000..b4805a34418abf6b2a242e6ed312babf18b4ca5e Binary files /dev/null and b/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_5e-1/20230603_194605/vis_data/vis_image/000005174.jpg_2.png differ diff --git a/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_5e-1/20230603_194605/vis_data/vis_image/000005174.jpg_3.png b/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_5e-1/20230603_194605/vis_data/vis_image/000005174.jpg_3.png new file mode 100644 index 0000000000000000000000000000000000000000..0f1d3167aae3f48685a92ef978ada617f39f67b3 Binary files /dev/null and b/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_5e-1/20230603_194605/vis_data/vis_image/000005174.jpg_3.png differ diff --git a/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_5e-1/20230603_194605/vis_data/vis_image/000005174.jpg_4.png b/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_5e-1/20230603_194605/vis_data/vis_image/000005174.jpg_4.png new file mode 100644 index 0000000000000000000000000000000000000000..ce559ee6f8b29321fc81d5667484ea0b10cc5e14 Binary files /dev/null and b/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_5e-1/20230603_194605/vis_data/vis_image/000005174.jpg_4.png differ diff --git a/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_5e-1/20230603_194605/vis_data/vis_image/000005174.jpg_5.png b/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_5e-1/20230603_194605/vis_data/vis_image/000005174.jpg_5.png new file mode 100644 index 0000000000000000000000000000000000000000..d1845e387c9533580299787248500c78287bc7a5 Binary files /dev/null and b/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_5e-1/20230603_194605/vis_data/vis_image/000005174.jpg_5.png differ diff --git a/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_5e-1/20230603_194605/vis_data/vis_image/000005174.jpg_6.png b/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_5e-1/20230603_194605/vis_data/vis_image/000005174.jpg_6.png new file mode 100644 index 0000000000000000000000000000000000000000..64d2518da93303bf967248bff32c8286c0dcad6d Binary files /dev/null and b/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_5e-1/20230603_194605/vis_data/vis_image/000005174.jpg_6.png differ diff --git a/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_5e-1/20230603_194605/vis_data/vis_image/000005174.jpg_7.png b/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_5e-1/20230603_194605/vis_data/vis_image/000005174.jpg_7.png new file mode 100644 index 0000000000000000000000000000000000000000..6b406bd24288bf2e82d9a3d813443f5fa371fd16 Binary files /dev/null and b/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_5e-1/20230603_194605/vis_data/vis_image/000005174.jpg_7.png differ diff --git a/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_5e-1/20230603_194605/vis_data/vis_image/000005174.jpg_8.png b/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_5e-1/20230603_194605/vis_data/vis_image/000005174.jpg_8.png new file mode 100644 index 0000000000000000000000000000000000000000..3b71406b06ba09006152bdd95944948252d944a8 Binary files /dev/null and b/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_5e-1/20230603_194605/vis_data/vis_image/000005174.jpg_8.png differ diff --git a/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_5e-1/20230603_194605/vis_data/vis_image/000005174.jpg_9.png b/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_5e-1/20230603_194605/vis_data/vis_image/000005174.jpg_9.png new file mode 100644 index 0000000000000000000000000000000000000000..279a25c5f37a33dd164f00908975b189991defa8 Binary files /dev/null and b/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_5e-1/20230603_194605/vis_data/vis_image/000005174.jpg_9.png differ diff --git a/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_5e-1/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_5e-1.py b/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_5e-1/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_5e-1.py new file mode 100644 index 0000000000000000000000000000000000000000..58e70090387bd6ad28c5168ec1be12f28a9f050d --- /dev/null +++ b/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_5e-1/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_5e-1.py @@ -0,0 +1,247 @@ +optim_wrapper = dict( + optimizer=dict( + type='AdamW', + lr=0.0004, + weight_decay=0.05, + eps=1e-08, + betas=(0.9, 0.999), + _scope_='mmpretrain'), + paramwise_cfg=dict( + norm_decay_mult=0.0, + bias_decay_mult=0.0, + flat_decay_mult=0.0, + custom_keys=dict({ + '.absolute_pos_embed': dict(decay_mult=0.0), + '.relative_position_bias_table': dict(decay_mult=0.0) + })), + type='AmpOptimWrapper', + dtype='bfloat16', + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=False, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=4096) +model = dict( + type='ImageClassifier', + backbone=dict( + frozen_stages=2, + type='ConvNeXt', + arch='small', + drop_path_rate=0.4, + init_cfg=dict( + type='Pretrained', + checkpoint= + 'https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth', + prefix='backbone')), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=768, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + init_cfg=None), + init_cfg=dict( + type='TruncNormal', layer=['Conv2d', 'Linear'], std=0.02, bias=0.0), + train_cfg=None) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stablediffusionV2-1-dpmsolver-25-1m.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/cc1m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV2-1-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV2-1-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_5e-1' diff --git a/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_5e-1/epoch_1.pth b/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_5e-1/epoch_1.pth new file mode 100644 index 0000000000000000000000000000000000000000..5b60db502a40a12ab610506e702495766a7df8e6 --- /dev/null +++ b/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_5e-1/epoch_1.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:6c418b3fe8671c79ea74049a9c1c044235b144b0c9714abe79dfbcde2b1a92f4 +size 782110490 diff --git a/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_5e-1/epoch_10.pth b/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_5e-1/epoch_10.pth new file mode 100644 index 0000000000000000000000000000000000000000..54570b5223b0d1d5cc637a9b8a98910274c94cb3 --- /dev/null +++ b/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_5e-1/epoch_10.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:f6375014a7e0fab53ee41759f00e7ad339dee37acbcd6fcc05f9c30b5fa731d2 +size 783390554 diff --git a/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_5e-1/epoch_2.pth b/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_5e-1/epoch_2.pth new file mode 100644 index 0000000000000000000000000000000000000000..7afffb6ff10e5032d8f6a681a889c439daa066fe --- /dev/null +++ b/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_5e-1/epoch_2.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:fd344c9baab2dbb89283c1a76d80f8f7b21ae599416d1fec450a6e13e433b6e3 +size 782253914 diff --git a/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_5e-1/epoch_3.pth b/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_5e-1/epoch_3.pth new file mode 100644 index 0000000000000000000000000000000000000000..6e4ed4d30512ebae9ff884077400d01611086f71 --- /dev/null +++ b/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_5e-1/epoch_3.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:b69f4f3cd450d0119670ac12926e0bb341989e853e269feb2a790dec50832c29 +size 782396122 diff --git a/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_5e-1/epoch_4.pth b/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_5e-1/epoch_4.pth new file mode 100644 index 0000000000000000000000000000000000000000..5bd2489278573ad8df18ee379482576c9ff0dfd0 --- /dev/null +++ b/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_5e-1/epoch_4.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:fb9d3ebca3e0e5f03720dd55b924c8cf11b2b23215de3627c26d9482611c124a +size 782538458 diff --git a/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_5e-1/epoch_5.pth b/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_5e-1/epoch_5.pth new file mode 100644 index 0000000000000000000000000000000000000000..96b33cf12e78b5017c4f206b8f0ab775170cb34e --- /dev/null +++ b/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_5e-1/epoch_5.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:8bf8dcf6ca44cd517e648b64ef8016ac3e0d3ad5bf240b2f2a51f904763fdba9 +size 782680410 diff --git a/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_5e-1/epoch_6.pth b/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_5e-1/epoch_6.pth new file mode 100644 index 0000000000000000000000000000000000000000..c586359d488b7cb7ddb6bc9d4e89c5c3c40415dc --- /dev/null +++ b/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_5e-1/epoch_6.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:4f2c84ad90394ab0d9a3c76ece48732a1ee872535c84fea9d69bdb4d6005b3c4 +size 782822490 diff --git a/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_5e-1/epoch_7.pth b/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_5e-1/epoch_7.pth new file mode 100644 index 0000000000000000000000000000000000000000..04c245650b9851c7c0be3400d12de92a485bb6bb --- /dev/null +++ b/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_5e-1/epoch_7.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:2d81e221bdd24828adb727b3826e836fdf6753a9b8b94dfab831c03ce7ea1779 +size 782963994 diff --git a/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_5e-1/epoch_8.pth b/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_5e-1/epoch_8.pth new file mode 100644 index 0000000000000000000000000000000000000000..3526f391576a7ead9f3d14411effdc4ab98d9493 --- /dev/null +++ b/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_5e-1/epoch_8.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:bfbbcf30946ce0d56a9e09a3064d71615a50bd8e5bdc21f6bbd4b4fbb4b22758 +size 783105434 diff --git a/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_5e-1/epoch_9.pth b/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_5e-1/epoch_9.pth new file mode 100644 index 0000000000000000000000000000000000000000..5c9ab586fb1593babd7dbd705dd89be56c1180e7 --- /dev/null +++ b/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_5e-1/epoch_9.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:8010265c4844de53a2cdb8755b4f75587ca692541408886b45693fc8d590bfec +size 783247386 diff --git a/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_5e-1/last_checkpoint b/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_5e-1/last_checkpoint new file mode 100644 index 0000000000000000000000000000000000000000..04f107236b436002ccc72dd93d42f1e443d0efec --- /dev/null +++ b/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_5e-1/last_checkpoint @@ -0,0 +1 @@ +/mnt/petrelfs/luzeyu/workspace/fakebench/mmpretrain/workdir/convnext_small_4xb256_sdv2_1m_lr4e-4_aug_5e-1/epoch_10.pth \ No newline at end of file diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/20230605_130757.log b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/20230605_130757.log new file mode 100644 index 0000000000000000000000000000000000000000..8d0ab251c1970f139d269b0dfd8d504bf7497953 --- /dev/null +++ b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/20230605_130757.log @@ -0,0 +1,1652 @@ +2023/06/05 13:08:01 - mmengine - INFO - +------------------------------------------------------------ +System environment: + sys.platform: linux + Python: 3.10.9 (main, Mar 8 2023, 10:47:38) [GCC 11.2.0] + CUDA available: True + numpy_random_seed: 1607240931 + GPU 0,1,2,3: NVIDIA A100-SXM4-80GB + CUDA_HOME: /mnt/petrelfs/share/cuda-11.6 + NVCC: Cuda compilation tools, release 11.6, V11.6.124 + GCC: gcc (GCC) 7.5.0 + PyTorch: 1.13.1 + PyTorch compiling details: PyTorch built with: + - GCC 9.3 + - C++ Version: 201402 + - Intel(R) oneAPI Math Kernel Library Version 2021.4-Product Build 20210904 for Intel(R) 64 architecture applications + - Intel(R) MKL-DNN v2.6.0 (Git Hash 52b5f107dd9cf10910aaa19cb47f3abf9b349815) + - OpenMP 201511 (a.k.a. OpenMP 4.5) + - LAPACK is enabled (usually provided by MKL) + - NNPACK is enabled + - CPU capability usage: AVX2 + - CUDA Runtime 11.6 + - NVCC architecture flags: -gencode;arch=compute_37,code=sm_37;-gencode;arch=compute_50,code=sm_50;-gencode;arch=compute_60,code=sm_60;-gencode;arch=compute_61,code=sm_61;-gencode;arch=compute_70,code=sm_70;-gencode;arch=compute_75,code=sm_75;-gencode;arch=compute_80,code=sm_80;-gencode;arch=compute_86,code=sm_86;-gencode;arch=compute_37,code=compute_37 + - CuDNN 8.3.2 (built against CUDA 11.5) + - Magma 2.6.1 + - Build settings: BLAS_INFO=mkl, BUILD_TYPE=Release, CUDA_VERSION=11.6, CUDNN_VERSION=8.3.2, CXX_COMPILER=/opt/rh/devtoolset-9/root/usr/bin/c++, CXX_FLAGS= -fabi-version=11 -Wno-deprecated -fvisibility-inlines-hidden -DUSE_PTHREADPOOL -fopenmp -DNDEBUG -DUSE_KINETO -DUSE_FBGEMM -DUSE_QNNPACK -DUSE_PYTORCH_QNNPACK -DUSE_XNNPACK -DSYMBOLICATE_MOBILE_DEBUG_HANDLE -DEDGE_PROFILER_USE_KINETO -O2 -fPIC -Wno-narrowing -Wall -Wextra -Werror=return-type -Werror=non-virtual-dtor -Wno-missing-field-initializers -Wno-type-limits -Wno-array-bounds -Wno-unknown-pragmas -Wunused-local-typedefs -Wno-unused-parameter -Wno-unused-function -Wno-unused-result -Wno-strict-overflow -Wno-strict-aliasing -Wno-error=deprecated-declarations -Wno-stringop-overflow -Wno-psabi -Wno-error=pedantic -Wno-error=redundant-decls -Wno-error=old-style-cast -fdiagnostics-color=always -faligned-new -Wno-unused-but-set-variable -Wno-maybe-uninitialized -fno-math-errno -fno-trapping-math -Werror=format -Werror=cast-function-type -Wno-stringop-overflow, LAPACK_INFO=mkl, PERF_WITH_AVX=1, PERF_WITH_AVX2=1, PERF_WITH_AVX512=1, TORCH_VERSION=1.13.1, USE_CUDA=ON, USE_CUDNN=ON, USE_EXCEPTION_PTR=1, USE_GFLAGS=OFF, USE_GLOG=OFF, USE_MKL=ON, USE_MKLDNN=ON, USE_MPI=OFF, USE_NCCL=ON, USE_NNPACK=ON, USE_OPENMP=ON, USE_ROCM=OFF, + + TorchVision: 0.14.1 + OpenCV: 4.7.0 + MMEngine: 0.7.3 + +Runtime environment: + cudnn_benchmark: True + mp_cfg: {'mp_start_method': 'fork', 'opencv_num_threads': 0} + dist_cfg: {'backend': 'nccl'} + seed: None + deterministic: False + Distributed launcher: slurm + Distributed training: True + GPU number: 4 +------------------------------------------------------------ + +2023/06/05 13:08:05 - mmengine - INFO - Config: +optim_wrapper = dict( + optimizer=dict( + type='AdamW', + lr=0.0001, + weight_decay=0.05, + eps=1e-08, + betas=(0.9, 0.999), + _scope_='mmpretrain'), + paramwise_cfg=dict( + norm_decay_mult=0.0, + bias_decay_mult=0.0, + flat_decay_mult=0.0, + custom_keys=dict({ + '.absolute_pos_embed': dict(decay_mult=0.0), + '.relative_position_bias_table': dict(decay_mult=0.0) + })), + type='AmpOptimWrapper', + dtype='bfloat16', + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=False, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=4096) +model = dict( + type='ImageClassifier', + backbone=dict( + frozen_stages=2, + type='ConvNeXt', + arch='small', + drop_path_rate=0.4, + init_cfg=dict( + type='Pretrained', + checkpoint= + 'https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth', + prefix='backbone')), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=768, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + init_cfg=None), + init_cfg=dict( + type='TruncNormal', layer=['Conv2d', 'Linear'], std=0.02, bias=0.0), + train_cfg=None) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3fake8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3real8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3fake8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3real8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3fake8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3real8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1' + +2023/06/05 13:08:16 - mmengine - INFO - Hooks will be executed in the following order: +before_run: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_load_checkpoint: +(ABOVE_NORMAL) EMAHook + -------------------- +before_train: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_train_epoch: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(NORMAL ) DistSamplerSeedHook + -------------------- +before_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook + -------------------- +after_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +after_train_epoch: +(NORMAL ) IterTimerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_val_epoch: +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook + -------------------- +before_val_iter: +(NORMAL ) IterTimerHook + -------------------- +after_val_iter: +(NORMAL ) IterTimerHook +(NORMAL ) VisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_val_epoch: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_save_checkpoint: +(ABOVE_NORMAL) EMAHook + -------------------- +after_train: +(VERY_LOW ) CheckpointHook + -------------------- +before_test_epoch: +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook + -------------------- +before_test_iter: +(NORMAL ) IterTimerHook + -------------------- +after_test_iter: +(NORMAL ) IterTimerHook +(NORMAL ) VisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_test_epoch: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_run: +(BELOW_NORMAL) LoggerHook + -------------------- +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.2.0.weight:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.2.0.bias:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.2.1.bias:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.3.0.weight:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.3.0.bias:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.3.1.bias:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.0.gamma:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.0.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.0.norm.weight:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.0.norm.bias:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.0.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.0.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.1.gamma:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.1.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.1.norm.weight:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.1.norm.bias:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.1.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.1.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.2.gamma:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.2.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.2.norm.weight:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.2.norm.bias:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.2.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.2.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.3.gamma:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.3.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.3.norm.weight:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.3.norm.bias:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.3.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.3.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.4.gamma:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.4.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.4.norm.weight:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.4.norm.bias:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.4.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.4.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.5.gamma:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.5.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.5.norm.weight:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.5.norm.bias:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.5.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.5.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.6.gamma:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.6.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.6.norm.weight:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.6.norm.bias:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.6.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.6.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.7.gamma:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.7.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.7.norm.weight:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.7.norm.bias:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.7.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.7.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.8.gamma:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.8.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.8.norm.weight:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.8.norm.bias:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.8.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.8.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.9.gamma:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.9.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.9.norm.weight:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.9.norm.bias:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.9.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.9.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.10.gamma:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.10.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.10.norm.weight:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.10.norm.bias:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.10.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.10.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.11.gamma:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.11.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.11.norm.weight:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.11.norm.bias:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.11.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.11.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.12.gamma:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.12.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.12.norm.weight:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.12.norm.bias:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.12.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.12.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.13.gamma:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.13.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.13.norm.weight:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.13.norm.bias:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.13.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.13.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.14.gamma:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.14.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.14.norm.weight:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.14.norm.bias:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.14.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.14.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.15.gamma:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.15.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.15.norm.weight:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.15.norm.bias:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.15.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.15.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.16.gamma:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.16.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.16.norm.weight:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.16.norm.bias:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.16.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.16.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.17.gamma:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.17.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.17.norm.weight:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.17.norm.bias:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.17.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.17.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.18.gamma:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.18.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.18.norm.weight:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.18.norm.bias:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.18.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.18.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.19.gamma:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.19.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.19.norm.weight:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.19.norm.bias:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.19.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.19.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.20.gamma:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.20.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.20.norm.weight:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.20.norm.bias:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.20.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.20.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.21.gamma:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.21.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.21.norm.weight:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.21.norm.bias:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.21.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.21.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.22.gamma:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.22.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.22.norm.weight:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.22.norm.bias:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.22.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.22.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.23.gamma:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.23.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.23.norm.weight:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.23.norm.bias:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.23.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.23.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.24.gamma:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.24.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.24.norm.weight:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.24.norm.bias:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.24.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.24.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.25.gamma:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.25.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.25.norm.weight:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.25.norm.bias:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.25.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.25.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.26.gamma:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.26.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.26.norm.weight:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.26.norm.bias:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.26.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.26.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.3.0.gamma:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.3.0.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.3.0.norm.weight:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.3.0.norm.bias:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.3.0.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.3.0.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.3.1.gamma:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.3.1.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.3.1.norm.weight:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.3.1.norm.bias:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.3.1.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.3.1.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.3.2.gamma:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.3.2.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.3.2.norm.weight:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.3.2.norm.bias:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.3.2.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.stages.3.2.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.norm3.weight:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- backbone.norm3.bias:weight_decay=0.0 +2023/06/05 13:08:21 - mmengine - INFO - paramwise_options -- head.fc.bias:weight_decay=0.0 +2023/06/05 13:08:26 - mmengine - INFO - load backbone in model from: https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth +Name of parameter - Initialization information + +backbone.downsample_layers.0.0.weight - torch.Size([96, 3, 4, 4]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.0.0.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.0.1.weight - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.0.1.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.1.0.weight - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.1.0.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.1.1.weight - torch.Size([192, 96, 2, 2]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.1.1.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.2.0.weight - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.2.0.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.2.1.weight - torch.Size([384, 192, 2, 2]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.2.1.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.3.0.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.3.0.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.3.1.weight - torch.Size([768, 384, 2, 2]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.3.1.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.0.gamma - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.0.depthwise_conv.weight - torch.Size([96, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.0.depthwise_conv.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.0.norm.weight - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.0.norm.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.0.pointwise_conv1.weight - torch.Size([384, 96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.0.pointwise_conv1.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.0.pointwise_conv2.weight - torch.Size([96, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.0.pointwise_conv2.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.1.gamma - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.1.depthwise_conv.weight - torch.Size([96, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.1.depthwise_conv.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.1.norm.weight - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.1.norm.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.1.pointwise_conv1.weight - torch.Size([384, 96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.1.pointwise_conv1.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.1.pointwise_conv2.weight - torch.Size([96, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.1.pointwise_conv2.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.2.gamma - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.2.depthwise_conv.weight - torch.Size([96, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.2.depthwise_conv.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.2.norm.weight - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.2.norm.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.2.pointwise_conv1.weight - torch.Size([384, 96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.2.pointwise_conv1.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.2.pointwise_conv2.weight - torch.Size([96, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.2.pointwise_conv2.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.0.gamma - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.0.depthwise_conv.weight - torch.Size([192, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.0.depthwise_conv.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.0.norm.weight - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.0.norm.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.0.pointwise_conv1.weight - torch.Size([768, 192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.0.pointwise_conv1.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.0.pointwise_conv2.weight - torch.Size([192, 768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.0.pointwise_conv2.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.1.gamma - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.1.depthwise_conv.weight - torch.Size([192, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.1.depthwise_conv.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.1.norm.weight - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.1.norm.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.1.pointwise_conv1.weight - torch.Size([768, 192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.1.pointwise_conv1.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.1.pointwise_conv2.weight - torch.Size([192, 768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.1.pointwise_conv2.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.2.gamma - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.2.depthwise_conv.weight - torch.Size([192, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.2.depthwise_conv.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.2.norm.weight - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.2.norm.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.2.pointwise_conv1.weight - torch.Size([768, 192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.2.pointwise_conv1.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.2.pointwise_conv2.weight - torch.Size([192, 768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.2.pointwise_conv2.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.0.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.0.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.0.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.0.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.0.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.0.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.0.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.0.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.0.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.1.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.1.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.1.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.1.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.1.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.1.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.1.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.1.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.1.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.2.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.2.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.2.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.2.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.2.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.2.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.2.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.2.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.2.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.3.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.3.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.3.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.3.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.3.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.3.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.3.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.3.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.3.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.4.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.4.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.4.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.4.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.4.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.4.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.4.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.4.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.4.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.5.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.5.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.5.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.5.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.5.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.5.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.5.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.5.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.5.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.6.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.6.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.6.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.6.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.6.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.6.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.6.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.6.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.6.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.7.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.7.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.7.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.7.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.7.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.7.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.7.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.7.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.7.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.8.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.8.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.8.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.8.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.8.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.8.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.8.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.8.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.8.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.9.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.9.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.9.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.9.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.9.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.9.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.9.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.9.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.9.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.10.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.10.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.10.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.10.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.10.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.10.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.10.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.10.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.10.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.11.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.11.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.11.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.11.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.11.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.11.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.11.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.11.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.11.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.12.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.12.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.12.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.12.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.12.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.12.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.12.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.12.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.12.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.13.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.13.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.13.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.13.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.13.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.13.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.13.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.13.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.13.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.14.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.14.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.14.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.14.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.14.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.14.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.14.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.14.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.14.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.15.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.15.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.15.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.15.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.15.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.15.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.15.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.15.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.15.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.16.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.16.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.16.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.16.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.16.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.16.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.16.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.16.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.16.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.17.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.17.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.17.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.17.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.17.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.17.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.17.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.17.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.17.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.18.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.18.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.18.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.18.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.18.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.18.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.18.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.18.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.18.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.19.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.19.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.19.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.19.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.19.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.19.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.19.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.19.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.19.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.20.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.20.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.20.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.20.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.20.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.20.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.20.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.20.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.20.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.21.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.21.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.21.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.21.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.21.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.21.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.21.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.21.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.21.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.22.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.22.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.22.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.22.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.22.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.22.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.22.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.22.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.22.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.23.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.23.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.23.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.23.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.23.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.23.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.23.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.23.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.23.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.24.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.24.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.24.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.24.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.24.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.24.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.24.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.24.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.24.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.25.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.25.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.25.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.25.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.25.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.25.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.25.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.25.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.25.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.26.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.26.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.26.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.26.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.26.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.26.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.26.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.26.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.26.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.0.gamma - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.0.depthwise_conv.weight - torch.Size([768, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.0.depthwise_conv.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.0.norm.weight - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.0.norm.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.0.pointwise_conv1.weight - torch.Size([3072, 768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.0.pointwise_conv1.bias - torch.Size([3072]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.0.pointwise_conv2.weight - torch.Size([768, 3072]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.0.pointwise_conv2.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.1.gamma - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.1.depthwise_conv.weight - torch.Size([768, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.1.depthwise_conv.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.1.norm.weight - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.1.norm.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.1.pointwise_conv1.weight - torch.Size([3072, 768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.1.pointwise_conv1.bias - torch.Size([3072]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.1.pointwise_conv2.weight - torch.Size([768, 3072]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.1.pointwise_conv2.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.2.gamma - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.2.depthwise_conv.weight - torch.Size([768, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.2.depthwise_conv.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.2.norm.weight - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.2.norm.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.2.pointwise_conv1.weight - torch.Size([3072, 768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.2.pointwise_conv1.bias - torch.Size([3072]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.2.pointwise_conv2.weight - torch.Size([768, 3072]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.2.pointwise_conv2.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.norm3.weight - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.norm3.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +head.fc.weight - torch.Size([2, 768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +head.fc.bias - torch.Size([2]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 +2023/06/05 13:08:26 - mmengine - WARNING - "FileClient" will be deprecated in future. Please use io functions in https://mmengine.readthedocs.io/en/latest/api/fileio.html#file-io +2023/06/05 13:08:26 - mmengine - WARNING - "HardDiskBackend" is the alias of "LocalBackend" and the former will be deprecated in future. +2023/06/05 13:08:26 - mmengine - INFO - Checkpoints will be saved to /mnt/petrelfs/luzeyu/workspace/fakebench/mmpretrain/workdir/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1. +2023/06/05 13:10:38 - mmengine - INFO - Epoch(train) [1][100/171] lr: 9.9258e-05 eta: 0:35:15 time: 1.1906 data_time: 0.0016 memory: 15587 loss: 0.0588 +2023/06/05 13:12:03 - mmengine - INFO - Exp name: convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1_20230605_130757 +2023/06/05 13:12:03 - mmengine - INFO - Saving checkpoint at 1 epochs +2023/06/05 13:13:59 - mmengine - INFO - Epoch(val) [1][100/171] eta: 0:01:17 time: 0.4144 data_time: 0.0070 memory: 15587 +2023/06/05 13:17:26 - mmengine - INFO - Epoch(val) [1][171/171] accuracy/top1: 52.3769 single-label/precision_classwise: [51.544517517089844, 55.00535202026367] single-label/recall_classwise: [78.34169006347656, 26.444276809692383] single-label/f1-score_classwise: [62.178810119628906, 35.71720886230469] data_time: 1.0213 time: 1.4468 +2023/06/05 13:19:39 - mmengine - INFO - Epoch(train) [2][100/171] lr: 9.4576e-05 eta: 0:30:51 time: 1.2289 data_time: 0.0017 memory: 15582 loss: 0.0347 +2023/06/05 13:21:05 - mmengine - INFO - Exp name: convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1_20230605_130757 +2023/06/05 13:21:05 - mmengine - INFO - Saving checkpoint at 2 epochs +2023/06/05 13:22:59 - mmengine - INFO - Epoch(val) [2][100/171] eta: 0:01:16 time: 0.4150 data_time: 0.0074 memory: 15582 +2023/06/05 13:26:22 - mmengine - INFO - Epoch(val) [2][171/171] accuracy/top1: 53.6470 single-label/precision_classwise: [52.47761154174805, 56.80461120605469] single-label/recall_classwise: [76.63848876953125, 30.68387222290039] single-label/f1-score_classwise: [62.2974967956543, 39.844913482666016] data_time: 0.9905 time: 1.4025 +2023/06/05 13:28:30 - mmengine - INFO - Epoch(train) [3][100/171] lr: 8.6021e-05 eta: 0:26:56 time: 1.2553 data_time: 0.0018 memory: 15582 loss: 0.0275 +2023/06/05 13:29:59 - mmengine - INFO - Exp name: convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1_20230605_130757 +2023/06/05 13:29:59 - mmengine - INFO - Saving checkpoint at 3 epochs +2023/06/05 13:31:54 - mmengine - INFO - Epoch(val) [3][100/171] eta: 0:01:17 time: 0.4162 data_time: 0.0082 memory: 15582 +2023/06/05 13:35:19 - mmengine - INFO - Epoch(val) [3][171/171] accuracy/top1: 55.3915 single-label/precision_classwise: [53.84211730957031, 58.9658203125] single-label/recall_classwise: [75.16780090332031, 35.63959884643555] single-label/f1-score_classwise: [62.74236297607422, 44.42700958251953] data_time: 1.0106 time: 1.4233 +2023/06/05 13:37:28 - mmengine - INFO - Epoch(train) [4][100/171] lr: 7.4429e-05 eta: 0:23:17 time: 1.2697 data_time: 0.0434 memory: 15582 loss: 0.0158 +2023/06/05 13:38:56 - mmengine - INFO - Exp name: convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1_20230605_130757 +2023/06/05 13:38:56 - mmengine - INFO - Saving checkpoint at 4 epochs +2023/06/05 13:40:50 - mmengine - INFO - Epoch(val) [4][100/171] eta: 0:01:16 time: 0.4150 data_time: 0.0072 memory: 15582 +2023/06/05 13:44:22 - mmengine - INFO - Epoch(val) [4][171/171] accuracy/top1: 57.1824 single-label/precision_classwise: [55.405487060546875, 60.65468215942383] single-label/recall_classwise: [73.34661865234375, 41.038047790527344] single-label/f1-score_classwise: [63.12603759765625, 48.95432662963867] data_time: 1.0070 time: 1.4198 +2023/06/05 13:46:33 - mmengine - INFO - Epoch(train) [5][100/171] lr: 6.0935e-05 eta: 0:19:39 time: 1.2135 data_time: 0.0027 memory: 15582 loss: 0.0188 +2023/06/05 13:48:01 - mmengine - INFO - Exp name: convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1_20230605_130757 +2023/06/05 13:48:01 - mmengine - INFO - Saving checkpoint at 5 epochs +2023/06/05 13:49:56 - mmengine - INFO - Epoch(val) [5][100/171] eta: 0:01:16 time: 0.4154 data_time: 0.0075 memory: 15582 +2023/06/05 13:53:27 - mmengine - INFO - Epoch(val) [5][171/171] accuracy/top1: 59.4763 single-label/precision_classwise: [57.5817985534668, 62.605674743652344] single-label/recall_classwise: [71.77971649169922, 47.18809509277344] single-label/f1-score_classwise: [63.90161895751953, 53.81439208984375] data_time: 1.0025 time: 1.4152 +2023/06/05 13:55:38 - mmengine - INFO - Epoch(train) [6][100/171] lr: 4.6860e-05 eta: 0:16:03 time: 1.2294 data_time: 0.0020 memory: 15582 loss: 0.0178 +2023/06/05 13:56:38 - mmengine - INFO - Exp name: convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1_20230605_130757 +2023/06/05 13:57:08 - mmengine - INFO - Exp name: convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1_20230605_130757 +2023/06/05 13:57:08 - mmengine - INFO - Saving checkpoint at 6 epochs +2023/06/05 13:59:02 - mmengine - INFO - Epoch(val) [6][100/171] eta: 0:01:17 time: 0.4261 data_time: 0.0068 memory: 15582 +2023/06/05 14:02:35 - mmengine - INFO - Epoch(val) [6][171/171] accuracy/top1: 61.4967 single-label/precision_classwise: [59.810001373291016, 63.868343353271484] single-label/recall_classwise: [69.94822692871094, 53.055572509765625] single-label/f1-score_classwise: [64.4830551147461, 57.96199035644531] data_time: 1.0000 time: 1.4121 +2023/06/05 14:04:44 - mmengine - INFO - Epoch(train) [7][100/171] lr: 3.3582e-05 eta: 0:12:25 time: 1.2830 data_time: 0.0023 memory: 15582 loss: 0.0153 +2023/06/05 14:06:11 - mmengine - INFO - Exp name: convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1_20230605_130757 +2023/06/05 14:06:11 - mmengine - INFO - Saving checkpoint at 7 epochs +2023/06/05 14:08:05 - mmengine - INFO - Epoch(val) [7][100/171] eta: 0:01:16 time: 0.4192 data_time: 0.0092 memory: 15582 +2023/06/05 14:11:37 - mmengine - INFO - Epoch(val) [7][171/171] accuracy/top1: 64.1884 single-label/precision_classwise: [63.04036331176758, 65.55160522460938] single-label/recall_classwise: [68.48326110839844, 59.89887237548828] single-label/f1-score_classwise: [65.64918518066406, 62.59788513183594] data_time: 0.9856 time: 1.4026 +2023/06/05 14:13:50 - mmengine - INFO - Epoch(train) [8][100/171] lr: 2.2401e-05 eta: 0:08:47 time: 1.2424 data_time: 0.0023 memory: 15582 loss: 0.0137 +2023/06/05 14:15:17 - mmengine - INFO - Exp name: convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1_20230605_130757 +2023/06/05 14:15:17 - mmengine - INFO - Saving checkpoint at 8 epochs +2023/06/05 14:17:13 - mmengine - INFO - Epoch(val) [8][100/171] eta: 0:01:17 time: 0.4328 data_time: 0.0090 memory: 15582 +2023/06/05 14:20:45 - mmengine - INFO - Epoch(val) [8][171/171] accuracy/top1: 66.3032 single-label/precision_classwise: [66.15746307373047, 66.45108032226562] single-label/recall_classwise: [66.66896057128906, 65.93795013427734] single-label/f1-score_classwise: [66.4122314453125, 66.19351959228516] data_time: 1.0086 time: 1.4223 +2023/06/05 14:22:58 - mmengine - INFO - Epoch(train) [9][100/171] lr: 1.4410e-05 eta: 0:05:09 time: 1.1147 data_time: 0.0016 memory: 15582 loss: 0.0147 +2023/06/05 14:24:26 - mmengine - INFO - Exp name: convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1_20230605_130757 +2023/06/05 14:24:26 - mmengine - INFO - Saving checkpoint at 9 epochs +2023/06/05 14:26:23 - mmengine - INFO - Epoch(val) [9][100/171] eta: 0:01:17 time: 0.4162 data_time: 0.0079 memory: 15582 +2023/06/05 14:29:53 - mmengine - INFO - Epoch(val) [9][171/171] accuracy/top1: 68.4318 single-label/precision_classwise: [69.77731323242188, 67.26134490966797] single-label/recall_classwise: [64.96117401123047, 71.8980941772461] single-label/f1-score_classwise: [67.28316497802734, 69.50247192382812] data_time: 1.0155 time: 1.4296 +2023/06/05 14:32:03 - mmengine - INFO - Epoch(train) [10][100/171] lr: 1.0393e-05 eta: 0:01:30 time: 1.2152 data_time: 0.0023 memory: 15582 loss: 0.0132 +2023/06/05 14:33:29 - mmengine - INFO - Exp name: convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1_20230605_130757 +2023/06/05 14:33:29 - mmengine - INFO - Saving checkpoint at 10 epochs +2023/06/05 14:35:24 - mmengine - INFO - Epoch(val) [10][100/171] eta: 0:01:16 time: 0.4331 data_time: 0.0083 memory: 15582 +2023/06/05 14:38:54 - mmengine - INFO - Epoch(val) [10][171/171] accuracy/top1: 70.4739 single-label/precision_classwise: [73.69762420654297, 68.02873992919922] single-label/recall_classwise: [63.61532974243164, 77.3239974975586] single-label/f1-score_classwise: [68.28633117675781, 72.379150390625] data_time: 0.9868 time: 1.4014 diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/20230605_130757.json b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/20230605_130757.json new file mode 100644 index 0000000000000000000000000000000000000000..7bc4d351bffc725528cc947c17dc92313ad0b852 --- /dev/null +++ b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/20230605_130757.json @@ -0,0 +1,20 @@ +{"lr": 9.925772836925723e-05, "data_time": 0.0015854358673095704, "loss": 0.05876042116433382, "time": 1.190629005432129, "epoch": 1, "memory": 15587, "step": 100} +{"accuracy/top1": 52.376949310302734, "data_time": 1.0213420349254942, "time": 1.446752931639465, "step": 1} +{"lr": 9.457631880429199e-05, "data_time": 0.0017116069793701172, "loss": 0.03468327410519123, "time": 1.228912591934204, "epoch": 2, "memory": 15582, "step": 271} +{"accuracy/top1": 53.64698028564453, "data_time": 0.9905425836873609, "time": 1.4025463267814282, "step": 2} +{"lr": 8.602090341033552e-05, "data_time": 0.001788759231567383, "loss": 0.027548110019415616, "time": 1.2553491830825805, "epoch": 3, "memory": 15582, "step": 442} +{"accuracy/top1": 55.39148712158203, "data_time": 1.0105970418730448, "time": 1.4233206787774728, "step": 3} +{"lr": 7.442894585523233e-05, "data_time": 0.043445587158203125, "loss": 0.015781699679791927, "time": 1.2696798324584961, "epoch": 4, "memory": 15582, "step": 613} +{"accuracy/top1": 57.182350158691406, "data_time": 1.007041709367619, "time": 1.419775573320167, "step": 4} +{"lr": 6.0935147710393706e-05, "data_time": 0.0027019023895263673, "loss": 0.01876516002230346, "time": 1.2135411739349364, "epoch": 5, "memory": 15582, "step": 784} +{"accuracy/top1": 59.4763069152832, "data_time": 1.0025147573892461, "time": 1.4151686166608057, "step": 5} +{"lr": 4.6860375955055494e-05, "data_time": 0.0020409107208251955, "loss": 0.017837236821651458, "time": 1.2294014930725097, "epoch": 6, "memory": 15582, "step": 955} +{"accuracy/top1": 61.496681213378906, "data_time": 0.9999547767084699, "time": 1.4121168366698331, "step": 6} +{"lr": 3.3582367313332126e-05, "data_time": 0.0022835016250610353, "loss": 0.015252875862643123, "time": 1.2829726696014405, "epoch": 7, "memory": 15582, "step": 1126} +{"accuracy/top1": 64.18841552734375, "data_time": 0.9856473692627841, "time": 1.4026432744292325, "step": 7} +{"lr": 2.2400865784401354e-05, "data_time": 0.002253079414367676, "loss": 0.013688745116814971, "time": 1.2424066305160522, "epoch": 8, "memory": 15582, "step": 1297} +{"accuracy/top1": 66.30323028564453, "data_time": 1.0086486880169359, "time": 1.4222926497459412, "step": 8} +{"lr": 1.4410394644016972e-05, "data_time": 0.0016428709030151367, "loss": 0.01465936629101634, "time": 1.1146580696105957, "epoch": 9, "memory": 15582, "step": 1468} +{"accuracy/top1": 68.43177795410156, "data_time": 1.0154968361521877, "time": 1.4295822908711988, "step": 9} +{"lr": 1.0393116880286122e-05, "data_time": 0.002296924591064453, "loss": 0.013170435465872288, "time": 1.2152070999145508, "epoch": 10, "memory": 15582, "step": 1639} +{"accuracy/top1": 70.47390747070312, "data_time": 0.9867724041606105, "time": 1.4014408020086067, "step": 10} diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/config.py b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/config.py new file mode 100644 index 0000000000000000000000000000000000000000..68ef5c25438d567b1d1e89c1b328e8593cb4b96a --- /dev/null +++ b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/config.py @@ -0,0 +1,247 @@ +optim_wrapper = dict( + optimizer=dict( + type='AdamW', + lr=0.0001, + weight_decay=0.05, + eps=1e-08, + betas=(0.9, 0.999), + _scope_='mmpretrain'), + paramwise_cfg=dict( + norm_decay_mult=0.0, + bias_decay_mult=0.0, + flat_decay_mult=0.0, + custom_keys=dict({ + '.absolute_pos_embed': dict(decay_mult=0.0), + '.relative_position_bias_table': dict(decay_mult=0.0) + })), + type='AmpOptimWrapper', + dtype='bfloat16', + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=False, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=4096) +model = dict( + type='ImageClassifier', + backbone=dict( + frozen_stages=2, + type='ConvNeXt', + arch='small', + drop_path_rate=0.4, + init_cfg=dict( + type='Pretrained', + checkpoint= + 'https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth', + prefix='backbone')), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=768, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + init_cfg=None), + init_cfg=dict( + type='TruncNormal', layer=['Conv2d', 'Linear'], std=0.02, bias=0.0), + train_cfg=None) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3fake8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3real8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3fake8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3real8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3fake8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3real8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1' diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/events.out.tfevents.1685941691.SH-IDC1-10-140-24-119.213595.0 b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/events.out.tfevents.1685941691.SH-IDC1-10-140-24-119.213595.0 new file mode 100644 index 0000000000000000000000000000000000000000..e89bb12211c7fbba6d10e5d28643fd8ad186bb56 --- /dev/null +++ b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/events.out.tfevents.1685941691.SH-IDC1-10-140-24-119.213595.0 @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:bcd1dab8668d14c85b46756d3c8d4ab15164f8dbd0affd17bc5812481221c1d1 +size 7075966 diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/scalars.json b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/scalars.json new file mode 100644 index 0000000000000000000000000000000000000000..7bc4d351bffc725528cc947c17dc92313ad0b852 --- /dev/null +++ b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/scalars.json @@ -0,0 +1,20 @@ +{"lr": 9.925772836925723e-05, "data_time": 0.0015854358673095704, "loss": 0.05876042116433382, "time": 1.190629005432129, "epoch": 1, "memory": 15587, "step": 100} +{"accuracy/top1": 52.376949310302734, "data_time": 1.0213420349254942, "time": 1.446752931639465, "step": 1} +{"lr": 9.457631880429199e-05, "data_time": 0.0017116069793701172, "loss": 0.03468327410519123, "time": 1.228912591934204, "epoch": 2, "memory": 15582, "step": 271} +{"accuracy/top1": 53.64698028564453, "data_time": 0.9905425836873609, "time": 1.4025463267814282, "step": 2} +{"lr": 8.602090341033552e-05, "data_time": 0.001788759231567383, "loss": 0.027548110019415616, "time": 1.2553491830825805, "epoch": 3, "memory": 15582, "step": 442} +{"accuracy/top1": 55.39148712158203, "data_time": 1.0105970418730448, "time": 1.4233206787774728, "step": 3} +{"lr": 7.442894585523233e-05, "data_time": 0.043445587158203125, "loss": 0.015781699679791927, "time": 1.2696798324584961, "epoch": 4, "memory": 15582, "step": 613} +{"accuracy/top1": 57.182350158691406, "data_time": 1.007041709367619, "time": 1.419775573320167, "step": 4} +{"lr": 6.0935147710393706e-05, "data_time": 0.0027019023895263673, "loss": 0.01876516002230346, "time": 1.2135411739349364, "epoch": 5, "memory": 15582, "step": 784} +{"accuracy/top1": 59.4763069152832, "data_time": 1.0025147573892461, "time": 1.4151686166608057, "step": 5} +{"lr": 4.6860375955055494e-05, "data_time": 0.0020409107208251955, "loss": 0.017837236821651458, "time": 1.2294014930725097, "epoch": 6, "memory": 15582, "step": 955} +{"accuracy/top1": 61.496681213378906, "data_time": 0.9999547767084699, "time": 1.4121168366698331, "step": 6} +{"lr": 3.3582367313332126e-05, "data_time": 0.0022835016250610353, "loss": 0.015252875862643123, "time": 1.2829726696014405, "epoch": 7, "memory": 15582, "step": 1126} +{"accuracy/top1": 64.18841552734375, "data_time": 0.9856473692627841, "time": 1.4026432744292325, "step": 7} +{"lr": 2.2400865784401354e-05, "data_time": 0.002253079414367676, "loss": 0.013688745116814971, "time": 1.2424066305160522, "epoch": 8, "memory": 15582, "step": 1297} +{"accuracy/top1": 66.30323028564453, "data_time": 1.0086486880169359, "time": 1.4222926497459412, "step": 8} +{"lr": 1.4410394644016972e-05, "data_time": 0.0016428709030151367, "loss": 0.01465936629101634, "time": 1.1146580696105957, "epoch": 9, "memory": 15582, "step": 1468} +{"accuracy/top1": 68.43177795410156, "data_time": 1.0154968361521877, "time": 1.4295822908711988, "step": 9} +{"lr": 1.0393116880286122e-05, "data_time": 0.002296924591064453, "loss": 0.013170435465872288, "time": 1.2152070999145508, "epoch": 10, "memory": 15582, "step": 1639} +{"accuracy/top1": 70.47390747070312, "data_time": 0.9867724041606105, "time": 1.4014408020086067, "step": 10} diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/35624.png_1.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/35624.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..d44a42dbf135cd5a56a2d4487febd0d47a87047d Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/35624.png_1.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/35624.png_10.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/35624.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..5890582be1abfd263eb7a92eff9ea4660891ffba Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/35624.png_10.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/35624.png_2.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/35624.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..8bf2ed8a0f884653d57821c997d0a6e54837f6cf Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/35624.png_2.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/35624.png_3.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/35624.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..49314e8ce3fb90faf011713c637f9d76faa388c7 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/35624.png_3.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/35624.png_4.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/35624.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..a9b266d0c17259469a0680fa8c83c3d3d6c47db0 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/35624.png_4.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/35624.png_5.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/35624.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..5a3dd80476fca5a4fc7f6e4deee84b976769bc04 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/35624.png_5.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/35624.png_6.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/35624.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..3e729dcfd23ccdc945b553abd1b89d3bbf375075 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/35624.png_6.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/35624.png_7.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/35624.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..8cab10b404d25df77fc86f202a500d336266e62c Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/35624.png_7.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/35624.png_8.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/35624.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..f907f0c94e921d6ee822f07e4f8c803b70732ac3 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/35624.png_8.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/35624.png_9.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/35624.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..87e5fca3112e6c578b3729338483f7957b9bdc6c Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/35624.png_9.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/61024.png_1.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/61024.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..60faefa06b9e8cd9272312e40d6e872cb6cee09a Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/61024.png_1.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/61024.png_10.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/61024.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..0211c8fe064b2799072bd1d4a88b86b65dc4ad68 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/61024.png_10.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/61024.png_2.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/61024.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..6b081528d2e2da6fe65c3adcf7b6836e44ffbf7a Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/61024.png_2.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/61024.png_3.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/61024.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..12d6a23beaab15cf58e06f8e1e85c724f698e4e8 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/61024.png_3.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/61024.png_4.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/61024.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..d28417a1def2a2cd9627293d023e6ec4766392dc Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/61024.png_4.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/61024.png_5.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/61024.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..6e743c5146890ce234ee1b64431de8f5072e8fb9 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/61024.png_5.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/61024.png_6.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/61024.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..3be324714bda4f6834254726364bed2dd0ed575c Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/61024.png_6.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/61024.png_7.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/61024.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..1d36ddd4fe2a7f48a391873533fc8b0ea2cbcd92 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/61024.png_7.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/61024.png_8.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/61024.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..f8781b28e4dd442c344738ab565b9017e32a651d Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/61024.png_8.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/61024.png_9.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/61024.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..b66b8647fbf402ade55dd77b5addd19e7d37adbb Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/61024.png_9.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/66390.png_1.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/66390.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..ebc1cd13f3bdae8b69b192af73104fb64e908e92 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/66390.png_1.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/66390.png_10.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/66390.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..aa548013f56ec90e9aa27a18fa3515187736f1a2 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/66390.png_10.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/66390.png_2.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/66390.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..80b1adb5c42141edf92d3e82be38fb4a0188f388 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/66390.png_2.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/66390.png_3.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/66390.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..69dbd051fc9e7bb91271683bb23ab01bf521a883 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/66390.png_3.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/66390.png_4.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/66390.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..7959fbd99d33247935c9c2cd9b092b7b9b9b7aa8 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/66390.png_4.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/66390.png_5.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/66390.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..68392954feddef749833a065d11650b02b4b6623 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/66390.png_5.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/66390.png_6.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/66390.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..6c01b82d185fb33e1f7f40e6431a158307a2196f Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/66390.png_6.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/66390.png_7.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/66390.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..7f946693d7377f169d8ad0ed9864faf13b3a129b Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/66390.png_7.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/66390.png_8.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/66390.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..d4ca145b37d97141ba0eeda525ca4387a745194d Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/66390.png_8.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/66390.png_9.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/66390.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..41aa408948034e63a8229ed0a1a8cfa3f58b5a7b Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/66390.png_9.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/pixabay_dog_002352.jpg_1.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/pixabay_dog_002352.jpg_1.png new file mode 100644 index 0000000000000000000000000000000000000000..309f0b953af85fc87712b23320abf9bc1bfd4681 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/pixabay_dog_002352.jpg_1.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/pixabay_dog_002352.jpg_10.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/pixabay_dog_002352.jpg_10.png new file mode 100644 index 0000000000000000000000000000000000000000..8973db85ba206163c7970d845c96f986f357a0ed Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/pixabay_dog_002352.jpg_10.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/pixabay_dog_002352.jpg_2.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/pixabay_dog_002352.jpg_2.png new file mode 100644 index 0000000000000000000000000000000000000000..3656ec56edeb5996a9b8529f709d2440af01c013 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/pixabay_dog_002352.jpg_2.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/pixabay_dog_002352.jpg_3.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/pixabay_dog_002352.jpg_3.png new file mode 100644 index 0000000000000000000000000000000000000000..d53339d4f0e4255472e8756a02fc2c2ba1c985cf Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/pixabay_dog_002352.jpg_3.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/pixabay_dog_002352.jpg_4.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/pixabay_dog_002352.jpg_4.png new file mode 100644 index 0000000000000000000000000000000000000000..616060af033863034c56b05da6910fdc52cb62f2 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/pixabay_dog_002352.jpg_4.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/pixabay_dog_002352.jpg_5.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/pixabay_dog_002352.jpg_5.png new file mode 100644 index 0000000000000000000000000000000000000000..0068a4559ed52cc904c42bf2efc6ada4692e7a55 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/pixabay_dog_002352.jpg_5.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/pixabay_dog_002352.jpg_6.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/pixabay_dog_002352.jpg_6.png new file mode 100644 index 0000000000000000000000000000000000000000..b69f7148f41e55848e1b5dae0d94d9dcac710fb9 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/pixabay_dog_002352.jpg_6.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/pixabay_dog_002352.jpg_7.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/pixabay_dog_002352.jpg_7.png new file mode 100644 index 0000000000000000000000000000000000000000..c0a218b26fba9d6f1fe3493d34cff6609ca0d81a Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/pixabay_dog_002352.jpg_7.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/pixabay_dog_002352.jpg_8.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/pixabay_dog_002352.jpg_8.png new file mode 100644 index 0000000000000000000000000000000000000000..0d3835327842b6df04f37068f3390242cacbc889 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/pixabay_dog_002352.jpg_8.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/pixabay_dog_002352.jpg_9.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/pixabay_dog_002352.jpg_9.png new file mode 100644 index 0000000000000000000000000000000000000000..bfe21f06e9e3018452805bfda644cfc916760dfd Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/pixabay_dog_002352.jpg_9.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/seed14088.png_1.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/seed14088.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..8bf90986b701d36ffc3897e3485fec99efcaa240 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/seed14088.png_1.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/seed14088.png_10.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/seed14088.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..6f9ab7bb41241b1286cfd338f6d65931c0620bf5 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/seed14088.png_10.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/seed14088.png_2.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/seed14088.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..4d94e9751248d0c7dfc05a253bd80c1d5741f722 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/seed14088.png_2.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/seed14088.png_3.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/seed14088.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..d8070d5fa5affc5887a0cd065a4ad32dd5deef16 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/seed14088.png_3.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/seed14088.png_4.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/seed14088.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..3ca5baee0589a208524f5bdcc0e89e2b5a516d46 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/seed14088.png_4.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/seed14088.png_5.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/seed14088.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..30d487e189dae12742751181291bdfd25e5e2829 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/seed14088.png_5.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/seed14088.png_6.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/seed14088.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..055f782cc5a7a9c76b7be85efa35b5fc7baae663 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/seed14088.png_6.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/seed14088.png_7.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/seed14088.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..643795118dbdcacb62d6ca1e1ac572f111f12197 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/seed14088.png_7.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/seed14088.png_8.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/seed14088.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..85ea8cff6de7b113a89c317805654ea6c46ddc42 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/seed14088.png_8.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/seed14088.png_9.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/seed14088.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..09e035d2e28f69b1460cc0248292a199f8dfc433 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/seed14088.png_9.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/seed17282.png_1.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/seed17282.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..be96db344f8e59065d7e46631b52a052a7ab507d Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/seed17282.png_1.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/seed17282.png_10.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/seed17282.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..2df15f30e712d9e6569ad959f87d62ee5283ad19 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/seed17282.png_10.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/seed17282.png_2.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/seed17282.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..3b8084d5fa0cf35e63690f12f0e2eaad2ad22bba Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/seed17282.png_2.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/seed17282.png_3.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/seed17282.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..cefdccc1f79f7e9b504f7e4fb719cd7d8e9fe78d Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/seed17282.png_3.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/seed17282.png_4.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/seed17282.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..078aac3f91bff0f9ee3b0022de5339e508b0483d Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/seed17282.png_4.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/seed17282.png_5.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/seed17282.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..c9db0209f7856dffb46d578fe9003e4770602d7d Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/seed17282.png_5.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/seed17282.png_6.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/seed17282.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..9231d61c8b1d0b72bf083ceb0430511a78ca9941 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/seed17282.png_6.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/seed17282.png_7.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/seed17282.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..64095bde48a43dcba5eb95fe5e1a0d88b1275ae2 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/seed17282.png_7.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/seed17282.png_8.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/seed17282.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..f2ba5a397a050ec398643ce260af81639eca852d Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/seed17282.png_8.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/seed17282.png_9.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/seed17282.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..5ce940149c3ced3693f7f06d9057bdad5c80c1be Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/seed17282.png_9.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/seed23602.png_1.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/seed23602.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..2e886f424412e59c2d610c337a967100cfd432c5 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/seed23602.png_1.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/seed23602.png_10.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/seed23602.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..c1be92153002a61ea7f7c988744f42d31c2a0298 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/seed23602.png_10.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/seed23602.png_2.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/seed23602.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..064fcf62a2f63fde0096c7672eafd47544ee57e8 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/seed23602.png_2.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/seed23602.png_3.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/seed23602.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..db4b9e4735ddbce18fc938464492b32243550cae Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/seed23602.png_3.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/seed23602.png_4.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/seed23602.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..8eb0bf9fc6d1242b8ee7d32eebc56bf296b68982 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/seed23602.png_4.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/seed23602.png_5.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/seed23602.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..233bc5b435e59c65262b7938e48bfb59c78ff5f5 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/seed23602.png_5.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/seed23602.png_6.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/seed23602.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..4bd3852185b24ee934a7d8b7d52175ecd3e922e4 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/seed23602.png_6.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/seed23602.png_7.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/seed23602.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..0f2d74badebe817fe9e896a664bb7195a58b6820 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/seed23602.png_7.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/seed23602.png_8.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/seed23602.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..517d5a579fa9e318f7923a77783d1fd510abe698 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/seed23602.png_8.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/seed23602.png_9.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/seed23602.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..7588689311f066fea74cc9391ff4051a6c39559f Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/seed23602.png_9.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/seed24366.png_1.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/seed24366.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..4595c0df0f39e9ff64add71fa7807c6dae8410ff Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/seed24366.png_1.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/seed24366.png_10.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/seed24366.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..fae38d40af952be61c5458098029a730270b6087 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/seed24366.png_10.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/seed24366.png_2.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/seed24366.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..cce25301c4e88502744c4aa29b03a344cd806e3a Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/seed24366.png_2.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/seed24366.png_3.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/seed24366.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..b2873899cc565425c34c88df22e3434150c32098 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/seed24366.png_3.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/seed24366.png_4.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/seed24366.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..66a28d4cf980a6abcf2cb1dee9650ab6232bc450 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/seed24366.png_4.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/seed24366.png_5.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/seed24366.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..9e99ce4311ed589ea71e99d675cdff5543f3a750 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/seed24366.png_5.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/seed24366.png_6.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/seed24366.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..ab546fe661121d98d34b679010c196e1412ffd2d Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/seed24366.png_6.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/seed24366.png_7.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/seed24366.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..885ae64f165edfc7223cc073f1eb9e045c862ccf Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/seed24366.png_7.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/seed24366.png_8.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/seed24366.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..f254d0e187d397f3083c4bf73914590059231f89 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/seed24366.png_8.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/seed24366.png_9.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/seed24366.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..4c82b0057f9812734a68e3dc4273d001c89f51c2 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/seed24366.png_9.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/seed44086.png_1.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/seed44086.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..a4a774f43c820ade90f2aceaab1123c8cd1ce053 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/seed44086.png_1.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/seed44086.png_10.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/seed44086.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..b9c2d4ad2b35b74d1d9b18b4eb326b44f175ec43 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/seed44086.png_10.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/seed44086.png_2.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/seed44086.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..3ee53d2d7284ebae53ae9dd29a28d4ae0e22c38d Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/seed44086.png_2.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/seed44086.png_3.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/seed44086.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..9658ad60517397a9d8e661f104e7eeb8cf8a92f3 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/seed44086.png_3.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/seed44086.png_4.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/seed44086.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..21b7418fe8846e13bb02bc09dd47414c7023e387 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/seed44086.png_4.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/seed44086.png_5.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/seed44086.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..c0a7225b8f5b110fce2bc77de0c975707bf068fa Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/seed44086.png_5.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/seed44086.png_6.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/seed44086.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..6bc82fbcce7932da450ec1ef6087df5890e24f41 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/seed44086.png_6.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/seed44086.png_7.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/seed44086.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..fc71fd9bf533e71c004b1a8e07709dfa725a6f92 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/seed44086.png_7.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/seed44086.png_8.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/seed44086.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..887e8c49fedb6b162ec0488be6e83941ce6c55fa Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/seed44086.png_8.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/seed44086.png_9.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/seed44086.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..d975358ebb65d30daaa5c8612059667d877b9188 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_130757/vis_data/vis_image/seed44086.png_9.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1.py b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1.py new file mode 100644 index 0000000000000000000000000000000000000000..68ef5c25438d567b1d1e89c1b328e8593cb4b96a --- /dev/null +++ b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1.py @@ -0,0 +1,247 @@ +optim_wrapper = dict( + optimizer=dict( + type='AdamW', + lr=0.0001, + weight_decay=0.05, + eps=1e-08, + betas=(0.9, 0.999), + _scope_='mmpretrain'), + paramwise_cfg=dict( + norm_decay_mult=0.0, + bias_decay_mult=0.0, + flat_decay_mult=0.0, + custom_keys=dict({ + '.absolute_pos_embed': dict(decay_mult=0.0), + '.relative_position_bias_table': dict(decay_mult=0.0) + })), + type='AmpOptimWrapper', + dtype='bfloat16', + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=False, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=4096) +model = dict( + type='ImageClassifier', + backbone=dict( + frozen_stages=2, + type='ConvNeXt', + arch='small', + drop_path_rate=0.4, + init_cfg=dict( + type='Pretrained', + checkpoint= + 'https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth', + prefix='backbone')), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=768, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + init_cfg=None), + init_cfg=dict( + type='TruncNormal', layer=['Conv2d', 'Linear'], std=0.02, bias=0.0), + train_cfg=None) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3fake8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3real8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3fake8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3real8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3fake8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3real8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1' diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/epoch_1.pth b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/epoch_1.pth new file mode 100644 index 0000000000000000000000000000000000000000..21b1e73c51fe8f6cf8d9230488217ed9e229502f --- /dev/null +++ b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/epoch_1.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:99385beceb9eeb5ce1905e7c7043314d892dbc712f95f82a747430a406522964 +size 781981338 diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/epoch_10.pth b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/epoch_10.pth new file mode 100644 index 0000000000000000000000000000000000000000..b7af38ec76120042afd9d54b1b0ed0f376eca544 --- /dev/null +++ b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/epoch_10.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:1d0a56938a87c4be84d40abcf83907e9e8d25ec122dff6e13322b87807b9d0d6 +size 782147674 diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/epoch_2.pth b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/epoch_2.pth new file mode 100644 index 0000000000000000000000000000000000000000..132ec54343348524593944be909bf028dfadbe47 --- /dev/null +++ b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/epoch_2.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:0dfab7b0d4f925cbb734613dee64bb970fc4c0b970f2ba9ebf9d67fe3477b982 +size 782000794 diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/epoch_3.pth b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/epoch_3.pth new file mode 100644 index 0000000000000000000000000000000000000000..cb2e08e8c5493931bd203da23eac9ff1c9352c05 --- /dev/null +++ b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/epoch_3.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:d3347ce6d3f66a1e3aaedff2e67c633f663cb612658a4030c90fadc11a1be054 +size 782019098 diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/epoch_4.pth b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/epoch_4.pth new file mode 100644 index 0000000000000000000000000000000000000000..88f9923817342acc354659db4620d6d5d41cc0ff --- /dev/null +++ b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/epoch_4.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:90857721fa9ff9c617e04078c34d4a3ed9637dc399e5003deca3c26dba011d83 +size 782037402 diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/epoch_5.pth b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/epoch_5.pth new file mode 100644 index 0000000000000000000000000000000000000000..cace0fe24fec5ca2d28bda90f17c33b37568d876 --- /dev/null +++ b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/epoch_5.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:fbe7ffe6732486299912a4068ebe4e0e10c2efb0250b9aed4c6e4c6b18814467 +size 782055706 diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/epoch_6.pth b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/epoch_6.pth new file mode 100644 index 0000000000000000000000000000000000000000..b2326aff607bf5da098c1d524bd2cade0affe419 --- /dev/null +++ b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/epoch_6.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:9644403a3132405857f5accc7c8db1aeb071aa47c72952466b944a96f3d5b0b2 +size 782074074 diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/epoch_7.pth b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/epoch_7.pth new file mode 100644 index 0000000000000000000000000000000000000000..6fa4538e7ded111b1c7cf7e6cc4f8fcea896ca05 --- /dev/null +++ b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/epoch_7.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:85e23084acaa882557e809d3f9d7ccd846dfb8956ebe4472dc8c5d943fa51d24 +size 782092442 diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/epoch_8.pth b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/epoch_8.pth new file mode 100644 index 0000000000000000000000000000000000000000..30e072245e8b828dff1d59e06a5be3cb691eb50a --- /dev/null +++ b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/epoch_8.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ecbd06e553f85215b50038bc77386d245d0aaac2e662951faf9c45966fc09cc3 +size 782110874 diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/epoch_9.pth b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/epoch_9.pth new file mode 100644 index 0000000000000000000000000000000000000000..218b217299d117f9304229b39579ea47d31f90ff --- /dev/null +++ b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/epoch_9.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:bd29ee9d02e2c2f1c92ea406fd4e141e418607b60aca97e0b0ee6e97f4444dc0 +size 782129306 diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/last_checkpoint b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/last_checkpoint new file mode 100644 index 0000000000000000000000000000000000000000..5813c040c40100ae65f78fcd45d4326fb71cd952 --- /dev/null +++ b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/last_checkpoint @@ -0,0 +1 @@ +/mnt/petrelfs/luzeyu/workspace/fakebench/mmpretrain/workdir/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_1e-1/epoch_10.pth \ No newline at end of file diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/20230605_124607.log b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/20230605_124607.log new file mode 100644 index 0000000000000000000000000000000000000000..53b9832423d9abc3e0496a0b099382fce3067083 --- /dev/null +++ b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/20230605_124607.log @@ -0,0 +1,1652 @@ +2023/06/05 12:46:14 - mmengine - INFO - +------------------------------------------------------------ +System environment: + sys.platform: linux + Python: 3.10.9 (main, Mar 8 2023, 10:47:38) [GCC 11.2.0] + CUDA available: True + numpy_random_seed: 98458905 + GPU 0,1,2,3: NVIDIA A100-SXM4-80GB + CUDA_HOME: /mnt/petrelfs/share/cuda-11.6 + NVCC: Cuda compilation tools, release 11.6, V11.6.124 + GCC: gcc (GCC) 7.5.0 + PyTorch: 1.13.1 + PyTorch compiling details: PyTorch built with: + - GCC 9.3 + - C++ Version: 201402 + - Intel(R) oneAPI Math Kernel Library Version 2021.4-Product Build 20210904 for Intel(R) 64 architecture applications + - Intel(R) MKL-DNN v2.6.0 (Git Hash 52b5f107dd9cf10910aaa19cb47f3abf9b349815) + - OpenMP 201511 (a.k.a. OpenMP 4.5) + - LAPACK is enabled (usually provided by MKL) + - NNPACK is enabled + - CPU capability usage: AVX2 + - CUDA Runtime 11.6 + - NVCC architecture flags: -gencode;arch=compute_37,code=sm_37;-gencode;arch=compute_50,code=sm_50;-gencode;arch=compute_60,code=sm_60;-gencode;arch=compute_61,code=sm_61;-gencode;arch=compute_70,code=sm_70;-gencode;arch=compute_75,code=sm_75;-gencode;arch=compute_80,code=sm_80;-gencode;arch=compute_86,code=sm_86;-gencode;arch=compute_37,code=compute_37 + - CuDNN 8.3.2 (built against CUDA 11.5) + - Magma 2.6.1 + - Build settings: BLAS_INFO=mkl, BUILD_TYPE=Release, CUDA_VERSION=11.6, CUDNN_VERSION=8.3.2, CXX_COMPILER=/opt/rh/devtoolset-9/root/usr/bin/c++, CXX_FLAGS= -fabi-version=11 -Wno-deprecated -fvisibility-inlines-hidden -DUSE_PTHREADPOOL -fopenmp -DNDEBUG -DUSE_KINETO -DUSE_FBGEMM -DUSE_QNNPACK -DUSE_PYTORCH_QNNPACK -DUSE_XNNPACK -DSYMBOLICATE_MOBILE_DEBUG_HANDLE -DEDGE_PROFILER_USE_KINETO -O2 -fPIC -Wno-narrowing -Wall -Wextra -Werror=return-type -Werror=non-virtual-dtor -Wno-missing-field-initializers -Wno-type-limits -Wno-array-bounds -Wno-unknown-pragmas -Wunused-local-typedefs -Wno-unused-parameter -Wno-unused-function -Wno-unused-result -Wno-strict-overflow -Wno-strict-aliasing -Wno-error=deprecated-declarations -Wno-stringop-overflow -Wno-psabi -Wno-error=pedantic -Wno-error=redundant-decls -Wno-error=old-style-cast -fdiagnostics-color=always -faligned-new -Wno-unused-but-set-variable -Wno-maybe-uninitialized -fno-math-errno -fno-trapping-math -Werror=format -Werror=cast-function-type -Wno-stringop-overflow, LAPACK_INFO=mkl, PERF_WITH_AVX=1, PERF_WITH_AVX2=1, PERF_WITH_AVX512=1, TORCH_VERSION=1.13.1, USE_CUDA=ON, USE_CUDNN=ON, USE_EXCEPTION_PTR=1, USE_GFLAGS=OFF, USE_GLOG=OFF, USE_MKL=ON, USE_MKLDNN=ON, USE_MPI=OFF, USE_NCCL=ON, USE_NNPACK=ON, USE_OPENMP=ON, USE_ROCM=OFF, + + TorchVision: 0.14.1 + OpenCV: 4.7.0 + MMEngine: 0.7.3 + +Runtime environment: + cudnn_benchmark: True + mp_cfg: {'mp_start_method': 'fork', 'opencv_num_threads': 0} + dist_cfg: {'backend': 'nccl'} + seed: None + deterministic: False + Distributed launcher: slurm + Distributed training: True + GPU number: 4 +------------------------------------------------------------ + +2023/06/05 12:46:18 - mmengine - INFO - Config: +optim_wrapper = dict( + optimizer=dict( + type='AdamW', + lr=0.0001, + weight_decay=0.05, + eps=1e-08, + betas=(0.9, 0.999), + _scope_='mmpretrain'), + paramwise_cfg=dict( + norm_decay_mult=0.0, + bias_decay_mult=0.0, + flat_decay_mult=0.0, + custom_keys=dict({ + '.absolute_pos_embed': dict(decay_mult=0.0), + '.relative_position_bias_table': dict(decay_mult=0.0) + })), + type='AmpOptimWrapper', + dtype='bfloat16', + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=False, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=4096) +model = dict( + type='ImageClassifier', + backbone=dict( + frozen_stages=2, + type='ConvNeXt', + arch='small', + drop_path_rate=0.4, + init_cfg=dict( + type='Pretrained', + checkpoint= + 'https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth', + prefix='backbone')), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=768, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + init_cfg=None), + init_cfg=dict( + type='TruncNormal', layer=['Conv2d', 'Linear'], std=0.02, bias=0.0), + train_cfg=None) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3fake8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3real8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3fake8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3real8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3fake8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3real8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1' + +2023/06/05 12:46:29 - mmengine - INFO - Hooks will be executed in the following order: +before_run: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_load_checkpoint: +(ABOVE_NORMAL) EMAHook + -------------------- +before_train: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_train_epoch: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(NORMAL ) DistSamplerSeedHook + -------------------- +before_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook + -------------------- +after_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +after_train_epoch: +(NORMAL ) IterTimerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_val_epoch: +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook + -------------------- +before_val_iter: +(NORMAL ) IterTimerHook + -------------------- +after_val_iter: +(NORMAL ) IterTimerHook +(NORMAL ) VisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_val_epoch: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_save_checkpoint: +(ABOVE_NORMAL) EMAHook + -------------------- +after_train: +(VERY_LOW ) CheckpointHook + -------------------- +before_test_epoch: +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook + -------------------- +before_test_iter: +(NORMAL ) IterTimerHook + -------------------- +after_test_iter: +(NORMAL ) IterTimerHook +(NORMAL ) VisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_test_epoch: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_run: +(BELOW_NORMAL) LoggerHook + -------------------- +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.2.0.weight:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.2.0.bias:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.2.1.bias:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.3.0.weight:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.3.0.bias:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.3.1.bias:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.0.gamma:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.0.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.0.norm.weight:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.0.norm.bias:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.0.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.0.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.1.gamma:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.1.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.1.norm.weight:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.1.norm.bias:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.1.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.1.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.2.gamma:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.2.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.2.norm.weight:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.2.norm.bias:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.2.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.2.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.3.gamma:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.3.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.3.norm.weight:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.3.norm.bias:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.3.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.3.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.4.gamma:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.4.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.4.norm.weight:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.4.norm.bias:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.4.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.4.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.5.gamma:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.5.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.5.norm.weight:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.5.norm.bias:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.5.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.5.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.6.gamma:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.6.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.6.norm.weight:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.6.norm.bias:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.6.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.6.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.7.gamma:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.7.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.7.norm.weight:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.7.norm.bias:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.7.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.7.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.8.gamma:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.8.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.8.norm.weight:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.8.norm.bias:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.8.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.8.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.9.gamma:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.9.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.9.norm.weight:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.9.norm.bias:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.9.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.9.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.10.gamma:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.10.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.10.norm.weight:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.10.norm.bias:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.10.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.10.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.11.gamma:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.11.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.11.norm.weight:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.11.norm.bias:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.11.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.11.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.12.gamma:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.12.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.12.norm.weight:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.12.norm.bias:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.12.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.12.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.13.gamma:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.13.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.13.norm.weight:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.13.norm.bias:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.13.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.13.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.14.gamma:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.14.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.14.norm.weight:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.14.norm.bias:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.14.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.14.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.15.gamma:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.15.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.15.norm.weight:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.15.norm.bias:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.15.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.15.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.16.gamma:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.16.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.16.norm.weight:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.16.norm.bias:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.16.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.16.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.17.gamma:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.17.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.17.norm.weight:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.17.norm.bias:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.17.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.17.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.18.gamma:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.18.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.18.norm.weight:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.18.norm.bias:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.18.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.18.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.19.gamma:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.19.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.19.norm.weight:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.19.norm.bias:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.19.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.19.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.20.gamma:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.20.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.20.norm.weight:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.20.norm.bias:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.20.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.20.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.21.gamma:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.21.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.21.norm.weight:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.21.norm.bias:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.21.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.21.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.22.gamma:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.22.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.22.norm.weight:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.22.norm.bias:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.22.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.22.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.23.gamma:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.23.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.23.norm.weight:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.23.norm.bias:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.23.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.23.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.24.gamma:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.24.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.24.norm.weight:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.24.norm.bias:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.24.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.24.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.25.gamma:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.25.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.25.norm.weight:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.25.norm.bias:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.25.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.25.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.26.gamma:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.26.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.26.norm.weight:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.26.norm.bias:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.26.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.26.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.3.0.gamma:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.3.0.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.3.0.norm.weight:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.3.0.norm.bias:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.3.0.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.3.0.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.3.1.gamma:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.3.1.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.3.1.norm.weight:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.3.1.norm.bias:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.3.1.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.3.1.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.3.2.gamma:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.3.2.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.3.2.norm.weight:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.3.2.norm.bias:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.3.2.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.stages.3.2.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.norm3.weight:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- backbone.norm3.bias:weight_decay=0.0 +2023/06/05 12:46:34 - mmengine - INFO - paramwise_options -- head.fc.bias:weight_decay=0.0 +2023/06/05 12:46:40 - mmengine - INFO - load backbone in model from: https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth +Name of parameter - Initialization information + +backbone.downsample_layers.0.0.weight - torch.Size([96, 3, 4, 4]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.0.0.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.0.1.weight - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.0.1.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.1.0.weight - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.1.0.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.1.1.weight - torch.Size([192, 96, 2, 2]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.1.1.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.2.0.weight - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.2.0.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.2.1.weight - torch.Size([384, 192, 2, 2]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.2.1.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.3.0.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.3.0.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.3.1.weight - torch.Size([768, 384, 2, 2]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.3.1.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.0.gamma - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.0.depthwise_conv.weight - torch.Size([96, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.0.depthwise_conv.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.0.norm.weight - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.0.norm.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.0.pointwise_conv1.weight - torch.Size([384, 96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.0.pointwise_conv1.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.0.pointwise_conv2.weight - torch.Size([96, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.0.pointwise_conv2.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.1.gamma - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.1.depthwise_conv.weight - torch.Size([96, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.1.depthwise_conv.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.1.norm.weight - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.1.norm.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.1.pointwise_conv1.weight - torch.Size([384, 96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.1.pointwise_conv1.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.1.pointwise_conv2.weight - torch.Size([96, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.1.pointwise_conv2.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.2.gamma - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.2.depthwise_conv.weight - torch.Size([96, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.2.depthwise_conv.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.2.norm.weight - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.2.norm.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.2.pointwise_conv1.weight - torch.Size([384, 96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.2.pointwise_conv1.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.2.pointwise_conv2.weight - torch.Size([96, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.2.pointwise_conv2.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.0.gamma - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.0.depthwise_conv.weight - torch.Size([192, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.0.depthwise_conv.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.0.norm.weight - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.0.norm.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.0.pointwise_conv1.weight - torch.Size([768, 192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.0.pointwise_conv1.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.0.pointwise_conv2.weight - torch.Size([192, 768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.0.pointwise_conv2.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.1.gamma - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.1.depthwise_conv.weight - torch.Size([192, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.1.depthwise_conv.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.1.norm.weight - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.1.norm.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.1.pointwise_conv1.weight - torch.Size([768, 192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.1.pointwise_conv1.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.1.pointwise_conv2.weight - torch.Size([192, 768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.1.pointwise_conv2.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.2.gamma - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.2.depthwise_conv.weight - torch.Size([192, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.2.depthwise_conv.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.2.norm.weight - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.2.norm.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.2.pointwise_conv1.weight - torch.Size([768, 192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.2.pointwise_conv1.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.2.pointwise_conv2.weight - torch.Size([192, 768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.2.pointwise_conv2.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.0.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.0.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.0.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.0.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.0.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.0.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.0.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.0.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.0.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.1.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.1.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.1.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.1.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.1.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.1.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.1.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.1.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.1.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.2.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.2.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.2.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.2.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.2.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.2.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.2.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.2.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.2.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.3.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.3.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.3.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.3.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.3.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.3.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.3.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.3.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.3.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.4.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.4.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.4.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.4.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.4.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.4.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.4.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.4.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.4.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.5.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.5.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.5.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.5.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.5.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.5.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.5.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.5.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.5.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.6.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.6.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.6.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.6.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.6.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.6.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.6.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.6.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.6.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.7.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.7.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.7.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.7.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.7.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.7.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.7.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.7.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.7.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.8.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.8.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.8.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.8.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.8.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.8.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.8.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.8.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.8.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.9.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.9.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.9.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.9.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.9.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.9.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.9.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.9.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.9.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.10.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.10.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.10.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.10.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.10.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.10.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.10.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.10.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.10.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.11.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.11.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.11.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.11.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.11.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.11.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.11.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.11.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.11.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.12.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.12.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.12.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.12.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.12.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.12.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.12.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.12.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.12.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.13.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.13.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.13.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.13.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.13.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.13.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.13.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.13.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.13.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.14.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.14.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.14.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.14.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.14.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.14.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.14.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.14.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.14.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.15.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.15.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.15.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.15.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.15.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.15.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.15.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.15.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.15.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.16.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.16.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.16.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.16.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.16.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.16.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.16.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.16.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.16.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.17.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.17.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.17.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.17.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.17.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.17.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.17.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.17.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.17.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.18.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.18.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.18.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.18.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.18.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.18.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.18.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.18.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.18.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.19.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.19.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.19.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.19.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.19.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.19.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.19.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.19.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.19.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.20.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.20.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.20.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.20.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.20.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.20.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.20.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.20.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.20.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.21.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.21.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.21.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.21.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.21.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.21.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.21.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.21.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.21.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.22.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.22.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.22.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.22.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.22.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.22.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.22.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.22.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.22.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.23.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.23.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.23.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.23.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.23.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.23.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.23.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.23.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.23.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.24.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.24.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.24.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.24.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.24.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.24.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.24.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.24.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.24.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.25.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.25.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.25.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.25.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.25.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.25.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.25.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.25.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.25.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.26.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.26.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.26.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.26.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.26.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.26.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.26.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.26.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.26.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.0.gamma - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.0.depthwise_conv.weight - torch.Size([768, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.0.depthwise_conv.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.0.norm.weight - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.0.norm.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.0.pointwise_conv1.weight - torch.Size([3072, 768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.0.pointwise_conv1.bias - torch.Size([3072]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.0.pointwise_conv2.weight - torch.Size([768, 3072]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.0.pointwise_conv2.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.1.gamma - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.1.depthwise_conv.weight - torch.Size([768, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.1.depthwise_conv.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.1.norm.weight - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.1.norm.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.1.pointwise_conv1.weight - torch.Size([3072, 768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.1.pointwise_conv1.bias - torch.Size([3072]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.1.pointwise_conv2.weight - torch.Size([768, 3072]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.1.pointwise_conv2.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.2.gamma - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.2.depthwise_conv.weight - torch.Size([768, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.2.depthwise_conv.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.2.norm.weight - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.2.norm.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.2.pointwise_conv1.weight - torch.Size([3072, 768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.2.pointwise_conv1.bias - torch.Size([3072]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.2.pointwise_conv2.weight - torch.Size([768, 3072]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.2.pointwise_conv2.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.norm3.weight - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.norm3.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +head.fc.weight - torch.Size([2, 768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +head.fc.bias - torch.Size([2]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 +2023/06/05 12:46:40 - mmengine - WARNING - "FileClient" will be deprecated in future. Please use io functions in https://mmengine.readthedocs.io/en/latest/api/fileio.html#file-io +2023/06/05 12:46:40 - mmengine - WARNING - "HardDiskBackend" is the alias of "LocalBackend" and the former will be deprecated in future. +2023/06/05 12:46:40 - mmengine - INFO - Checkpoints will be saved to /mnt/petrelfs/luzeyu/workspace/fakebench/mmpretrain/workdir/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1. +2023/06/05 12:48:55 - mmengine - INFO - Epoch(train) [1][100/171] lr: 9.9258e-05 eta: 0:36:19 time: 1.3130 data_time: 0.7898 memory: 15587 loss: 0.0787 +2023/06/05 12:50:28 - mmengine - INFO - Exp name: convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1_20230605_124607 +2023/06/05 12:50:28 - mmengine - INFO - Saving checkpoint at 1 epochs +2023/06/05 12:52:39 - mmengine - INFO - Epoch(val) [1][100/171] eta: 0:01:28 time: 0.4193 data_time: 0.0114 memory: 15587 +2023/06/05 12:55:58 - mmengine - INFO - Epoch(val) [1][171/171] accuracy/top1: 55.3354 single-label/precision_classwise: [53.94961166381836, 58.16838836669922] single-label/recall_classwise: [72.5013198852539, 38.190677642822266] single-label/f1-score_classwise: [61.86459732055664, 46.108585357666016] data_time: 1.1301 time: 1.5548 +2023/06/05 12:58:10 - mmengine - INFO - Epoch(train) [2][100/171] lr: 9.4576e-05 eta: 0:31:51 time: 1.3027 data_time: 0.7853 memory: 15582 loss: 0.0410 +2023/06/05 12:59:39 - mmengine - INFO - Exp name: convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1_20230605_124607 +2023/06/05 12:59:39 - mmengine - INFO - Saving checkpoint at 2 epochs +2023/06/05 13:01:51 - mmengine - INFO - Epoch(val) [2][100/171] eta: 0:01:28 time: 0.4147 data_time: 0.0073 memory: 15582 +2023/06/05 13:05:03 - mmengine - INFO - Epoch(val) [2][171/171] accuracy/top1: 57.0644 single-label/precision_classwise: [55.60972595214844, 59.5088005065918] single-label/recall_classwise: [69.76725769042969, 44.3773307800293] single-label/f1-score_classwise: [61.88916015625, 50.84108352661133] data_time: 1.1049 time: 1.5173 +2023/06/05 13:07:17 - mmengine - INFO - Epoch(train) [3][100/171] lr: 8.6021e-05 eta: 0:27:52 time: 1.2462 data_time: 0.7296 memory: 15582 loss: 0.0267 +2023/06/05 13:08:47 - mmengine - INFO - Exp name: convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1_20230605_124607 +2023/06/05 13:08:47 - mmengine - INFO - Saving checkpoint at 3 epochs +2023/06/05 13:10:56 - mmengine - INFO - Epoch(val) [3][100/171] eta: 0:01:26 time: 0.4151 data_time: 0.0074 memory: 15582 +2023/06/05 13:14:12 - mmengine - INFO - Epoch(val) [3][171/171] accuracy/top1: 59.1678 single-label/precision_classwise: [57.864349365234375, 60.974544525146484] single-label/recall_classwise: [67.26914978027344, 51.07648468017578] single-label/f1-score_classwise: [62.21332550048828, 55.58834457397461] data_time: 1.0999 time: 1.5125 +2023/06/05 13:16:24 - mmengine - INFO - Epoch(train) [4][100/171] lr: 7.4429e-05 eta: 0:24:00 time: 1.1503 data_time: 0.5665 memory: 15582 loss: 0.0227 +2023/06/05 13:17:55 - mmengine - INFO - Exp name: convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1_20230605_124607 +2023/06/05 13:17:55 - mmengine - INFO - Saving checkpoint at 4 epochs +2023/06/05 13:20:06 - mmengine - INFO - Epoch(val) [4][100/171] eta: 0:01:28 time: 0.4146 data_time: 0.0073 memory: 15582 +2023/06/05 13:23:24 - mmengine - INFO - Epoch(val) [4][171/171] accuracy/top1: 61.0840 single-label/precision_classwise: [60.46040725708008, 61.78255081176758] single-label/recall_classwise: [63.92573165893555, 58.245819091796875] single-label/f1-score_classwise: [62.144798278808594, 59.96207809448242] data_time: 1.1216 time: 1.5337 +2023/06/05 13:25:37 - mmengine - INFO - Epoch(train) [5][100/171] lr: 6.0935e-05 eta: 0:20:15 time: 1.3597 data_time: 0.8400 memory: 15582 loss: 0.0196 +2023/06/05 13:27:07 - mmengine - INFO - Exp name: convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1_20230605_124607 +2023/06/05 13:27:07 - mmengine - INFO - Saving checkpoint at 5 epochs +2023/06/05 13:29:17 - mmengine - INFO - Epoch(val) [5][100/171] eta: 0:01:27 time: 0.4153 data_time: 0.0076 memory: 15582 +2023/06/05 13:32:31 - mmengine - INFO - Epoch(val) [5][171/171] accuracy/top1: 63.2538 single-label/precision_classwise: [63.82039260864258, 62.734127044677734] single-label/recall_classwise: [61.09889221191406, 65.40599822998047] single-label/f1-score_classwise: [62.429996490478516, 64.04220581054688] data_time: 1.1051 time: 1.5179 +2023/06/05 13:34:45 - mmengine - INFO - Epoch(train) [6][100/171] lr: 4.6860e-05 eta: 0:16:30 time: 1.2642 data_time: 0.7471 memory: 15582 loss: 0.0249 +2023/06/05 13:35:48 - mmengine - INFO - Exp name: convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1_20230605_124607 +2023/06/05 13:36:16 - mmengine - INFO - Exp name: convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1_20230605_124607 +2023/06/05 13:36:16 - mmengine - INFO - Saving checkpoint at 6 epochs +2023/06/05 13:38:26 - mmengine - INFO - Epoch(val) [6][100/171] eta: 0:01:27 time: 0.4141 data_time: 0.0068 memory: 15582 +2023/06/05 13:41:43 - mmengine - INFO - Epoch(val) [6][171/171] accuracy/top1: 65.1213 single-label/precision_classwise: [67.60926055908203, 63.2560920715332] single-label/recall_classwise: [57.97425079345703, 72.25959014892578] single-label/f1-score_classwise: [62.42214584350586, 67.45874786376953] data_time: 1.1152 time: 1.5266 +2023/06/05 13:43:54 - mmengine - INFO - Epoch(train) [7][100/171] lr: 3.3582e-05 eta: 0:12:45 time: 1.2457 data_time: 0.6590 memory: 15582 loss: 0.0133 +2023/06/05 13:45:22 - mmengine - INFO - Exp name: convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1_20230605_124607 +2023/06/05 13:45:22 - mmengine - INFO - Saving checkpoint at 7 epochs +2023/06/05 13:47:31 - mmengine - INFO - Epoch(val) [7][100/171] eta: 0:01:27 time: 0.4144 data_time: 0.0072 memory: 15582 +2023/06/05 13:50:45 - mmengine - INFO - Epoch(val) [7][171/171] accuracy/top1: 66.8487 single-label/precision_classwise: [72.02302551269531, 63.65269470214844] single-label/recall_classwise: [55.034019470214844, 78.64872741699219] single-label/f1-score_classwise: [62.39270782470703, 70.36055755615234] data_time: 1.0949 time: 1.5062 +2023/06/05 13:53:00 - mmengine - INFO - Epoch(train) [8][100/171] lr: 2.2401e-05 eta: 0:09:00 time: 1.1702 data_time: 0.2989 memory: 15582 loss: 0.0178 +2023/06/05 13:54:27 - mmengine - INFO - Exp name: convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1_20230605_124607 +2023/06/05 13:54:27 - mmengine - INFO - Saving checkpoint at 8 epochs +2023/06/05 13:56:37 - mmengine - INFO - Epoch(val) [8][100/171] eta: 0:01:28 time: 0.4149 data_time: 0.0075 memory: 15582 +2023/06/05 13:59:52 - mmengine - INFO - Epoch(val) [8][171/171] accuracy/top1: 68.0002 single-label/precision_classwise: [76.03317260742188, 63.76682662963867] single-label/recall_classwise: [52.5141487121582, 83.46717834472656] single-label/f1-score_classwise: [62.12214279174805, 72.29901885986328] data_time: 1.1078 time: 1.5206 +2023/06/05 14:02:06 - mmengine - INFO - Epoch(train) [9][100/171] lr: 1.4410e-05 eta: 0:05:16 time: 1.1970 data_time: 0.6773 memory: 15582 loss: 0.0133 +2023/06/05 14:03:38 - mmengine - INFO - Exp name: convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1_20230605_124607 +2023/06/05 14:03:38 - mmengine - INFO - Saving checkpoint at 9 epochs +2023/06/05 14:05:49 - mmengine - INFO - Epoch(val) [9][100/171] eta: 0:01:27 time: 0.4148 data_time: 0.0075 memory: 15582 +2023/06/05 14:09:05 - mmengine - INFO - Epoch(val) [9][171/171] accuracy/top1: 68.9658 single-label/precision_classwise: [80.15715789794922, 63.84469985961914] single-label/recall_classwise: [50.35965347290039, 87.54890441894531] single-label/f1-score_classwise: [61.85696029663086, 73.84107971191406] data_time: 1.1153 time: 1.5259 +2023/06/05 14:11:17 - mmengine - INFO - Epoch(train) [10][100/171] lr: 1.0393e-05 eta: 0:01:32 time: 1.2602 data_time: 0.7340 memory: 15582 loss: 0.0136 +2023/06/05 14:12:46 - mmengine - INFO - Exp name: convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1_20230605_124607 +2023/06/05 14:12:46 - mmengine - INFO - Saving checkpoint at 10 epochs +2023/06/05 14:14:57 - mmengine - INFO - Epoch(val) [10][100/171] eta: 0:01:27 time: 0.4147 data_time: 0.0077 memory: 15582 +2023/06/05 14:18:14 - mmengine - INFO - Epoch(val) [10][171/171] accuracy/top1: 69.6766 single-label/precision_classwise: [84.07218170166016, 63.845306396484375] single-label/recall_classwise: [48.50525665283203, 90.82183837890625] single-label/f1-score_classwise: [61.51790237426758, 74.98099517822266] data_time: 1.1203 time: 1.5335 diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/20230605_124607.json b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/20230605_124607.json new file mode 100644 index 0000000000000000000000000000000000000000..11995b2523c5f1bceeed557b54a113bbdff79489 --- /dev/null +++ b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/20230605_124607.json @@ -0,0 +1,20 @@ +{"lr": 9.925772836925723e-05, "data_time": 0.7898063898086548, "loss": 0.07874756306409836, "time": 1.3130151748657226, "epoch": 1, "memory": 15587, "step": 100} +{"accuracy/top1": 55.33539581298828, "data_time": 1.130134353860777, "time": 1.554774628745185, "step": 1} +{"lr": 9.457631880429199e-05, "data_time": 0.7852777719497681, "loss": 0.04100975766777992, "time": 1.3027306079864502, "epoch": 2, "memory": 15582, "step": 271} +{"accuracy/top1": 57.064449310302734, "data_time": 1.1048849086428798, "time": 1.5172647806101067, "step": 2} +{"lr": 8.602090341033552e-05, "data_time": 0.7295974969863892, "loss": 0.026700264029204845, "time": 1.2462075471878051, "epoch": 3, "memory": 15582, "step": 442} +{"accuracy/top1": 59.16781234741211, "data_time": 1.099920044111651, "time": 1.5124684641527575, "step": 3} +{"lr": 7.442894585523233e-05, "data_time": 0.5664731740951539, "loss": 0.022700723446905612, "time": 1.150273609161377, "epoch": 4, "memory": 15582, "step": 613} +{"accuracy/top1": 61.084022521972656, "data_time": 1.1215637866840806, "time": 1.533672597519187, "step": 4} +{"lr": 6.0935147710393706e-05, "data_time": 0.8399810314178466, "loss": 0.01962084136903286, "time": 1.3596950769424438, "epoch": 5, "memory": 15582, "step": 784} +{"accuracy/top1": 63.253780364990234, "data_time": 1.1050903700118841, "time": 1.517908796321514, "step": 5} +{"lr": 4.6860375955055494e-05, "data_time": 0.7471021175384521, "loss": 0.024900213303044438, "time": 1.2641799926757813, "epoch": 6, "memory": 15582, "step": 955} +{"accuracy/top1": 65.121337890625, "data_time": 1.1151568459910015, "time": 1.5266117564467496, "step": 6} +{"lr": 3.3582367313332126e-05, "data_time": 0.6589953422546386, "loss": 0.01332604866474867, "time": 1.2457128047943116, "epoch": 7, "memory": 15582, "step": 1126} +{"accuracy/top1": 66.84867858886719, "data_time": 1.0949452976847804, "time": 1.5061987624611965, "step": 7} +{"lr": 2.2400865784401354e-05, "data_time": 0.29891166687011717, "loss": 0.017822381388396025, "time": 1.170210075378418, "epoch": 8, "memory": 15582, "step": 1297} +{"accuracy/top1": 68.00022888183594, "data_time": 1.107756872509801, "time": 1.5205856769583945, "step": 8} +{"lr": 1.4410394644016972e-05, "data_time": 0.6773425340652466, "loss": 0.013319657370448112, "time": 1.1969820499420165, "epoch": 9, "memory": 15582, "step": 1468} +{"accuracy/top1": 68.96577453613281, "data_time": 1.1152612553086392, "time": 1.5259180526400722, "step": 9} +{"lr": 1.0393116880286122e-05, "data_time": 0.7339940071105957, "loss": 0.013555871369317175, "time": 1.260174298286438, "epoch": 10, "memory": 15582, "step": 1639} +{"accuracy/top1": 69.67662811279297, "data_time": 1.1202679789343546, "time": 1.5334596841834311, "step": 10} diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/config.py b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/config.py new file mode 100644 index 0000000000000000000000000000000000000000..24166bc7056cbd8e3464ea8fbf43ad9b6b45a47d --- /dev/null +++ b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/config.py @@ -0,0 +1,247 @@ +optim_wrapper = dict( + optimizer=dict( + type='AdamW', + lr=0.0001, + weight_decay=0.05, + eps=1e-08, + betas=(0.9, 0.999), + _scope_='mmpretrain'), + paramwise_cfg=dict( + norm_decay_mult=0.0, + bias_decay_mult=0.0, + flat_decay_mult=0.0, + custom_keys=dict({ + '.absolute_pos_embed': dict(decay_mult=0.0), + '.relative_position_bias_table': dict(decay_mult=0.0) + })), + type='AmpOptimWrapper', + dtype='bfloat16', + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=False, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=4096) +model = dict( + type='ImageClassifier', + backbone=dict( + frozen_stages=2, + type='ConvNeXt', + arch='small', + drop_path_rate=0.4, + init_cfg=dict( + type='Pretrained', + checkpoint= + 'https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth', + prefix='backbone')), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=768, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + init_cfg=None), + init_cfg=dict( + type='TruncNormal', layer=['Conv2d', 'Linear'], std=0.02, bias=0.0), + train_cfg=None) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3fake8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3real8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3fake8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3real8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3fake8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3real8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1' diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/events.out.tfevents.1685940384.SH-IDC1-10-140-24-73.133582.0 b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/events.out.tfevents.1685940384.SH-IDC1-10-140-24-73.133582.0 new file mode 100644 index 0000000000000000000000000000000000000000..9748717a1c0f0e163296251bdbd63d21c01e48f8 --- /dev/null +++ b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/events.out.tfevents.1685940384.SH-IDC1-10-140-24-73.133582.0 @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:d3128135f4dd6fc312b9127756488bffc449062ce276537e420fcdcc26332b9f +size 7236762 diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/scalars.json b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/scalars.json new file mode 100644 index 0000000000000000000000000000000000000000..11995b2523c5f1bceeed557b54a113bbdff79489 --- /dev/null +++ b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/scalars.json @@ -0,0 +1,20 @@ +{"lr": 9.925772836925723e-05, "data_time": 0.7898063898086548, "loss": 0.07874756306409836, "time": 1.3130151748657226, "epoch": 1, "memory": 15587, "step": 100} +{"accuracy/top1": 55.33539581298828, "data_time": 1.130134353860777, "time": 1.554774628745185, "step": 1} +{"lr": 9.457631880429199e-05, "data_time": 0.7852777719497681, "loss": 0.04100975766777992, "time": 1.3027306079864502, "epoch": 2, "memory": 15582, "step": 271} +{"accuracy/top1": 57.064449310302734, "data_time": 1.1048849086428798, "time": 1.5172647806101067, "step": 2} +{"lr": 8.602090341033552e-05, "data_time": 0.7295974969863892, "loss": 0.026700264029204845, "time": 1.2462075471878051, "epoch": 3, "memory": 15582, "step": 442} +{"accuracy/top1": 59.16781234741211, "data_time": 1.099920044111651, "time": 1.5124684641527575, "step": 3} +{"lr": 7.442894585523233e-05, "data_time": 0.5664731740951539, "loss": 0.022700723446905612, "time": 1.150273609161377, "epoch": 4, "memory": 15582, "step": 613} +{"accuracy/top1": 61.084022521972656, "data_time": 1.1215637866840806, "time": 1.533672597519187, "step": 4} +{"lr": 6.0935147710393706e-05, "data_time": 0.8399810314178466, "loss": 0.01962084136903286, "time": 1.3596950769424438, "epoch": 5, "memory": 15582, "step": 784} +{"accuracy/top1": 63.253780364990234, "data_time": 1.1050903700118841, "time": 1.517908796321514, "step": 5} +{"lr": 4.6860375955055494e-05, "data_time": 0.7471021175384521, "loss": 0.024900213303044438, "time": 1.2641799926757813, "epoch": 6, "memory": 15582, "step": 955} +{"accuracy/top1": 65.121337890625, "data_time": 1.1151568459910015, "time": 1.5266117564467496, "step": 6} +{"lr": 3.3582367313332126e-05, "data_time": 0.6589953422546386, "loss": 0.01332604866474867, "time": 1.2457128047943116, "epoch": 7, "memory": 15582, "step": 1126} +{"accuracy/top1": 66.84867858886719, "data_time": 1.0949452976847804, "time": 1.5061987624611965, "step": 7} +{"lr": 2.2400865784401354e-05, "data_time": 0.29891166687011717, "loss": 0.017822381388396025, "time": 1.170210075378418, "epoch": 8, "memory": 15582, "step": 1297} +{"accuracy/top1": 68.00022888183594, "data_time": 1.107756872509801, "time": 1.5205856769583945, "step": 8} +{"lr": 1.4410394644016972e-05, "data_time": 0.6773425340652466, "loss": 0.013319657370448112, "time": 1.1969820499420165, "epoch": 9, "memory": 15582, "step": 1468} +{"accuracy/top1": 68.96577453613281, "data_time": 1.1152612553086392, "time": 1.5259180526400722, "step": 9} +{"lr": 1.0393116880286122e-05, "data_time": 0.7339940071105957, "loss": 0.013555871369317175, "time": 1.260174298286438, "epoch": 10, "memory": 15582, "step": 1639} +{"accuracy/top1": 69.67662811279297, "data_time": 1.1202679789343546, "time": 1.5334596841834311, "step": 10} diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/35624.png_1.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/35624.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..0b67fe685117fc79dc060e80f490b85d66b0c864 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/35624.png_1.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/35624.png_10.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/35624.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..afc5f53caf161ef01d06c9d46a795a3e369730c3 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/35624.png_10.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/35624.png_2.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/35624.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..a354d3b43610d39c711d0a6290f90015dedec57f Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/35624.png_2.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/35624.png_3.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/35624.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..a0b22bda8bdcc4d6d1f332ad859930078ecdb184 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/35624.png_3.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/35624.png_4.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/35624.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..32ff20758b1aae2c1fd933808197dca897d4302c Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/35624.png_4.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/35624.png_5.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/35624.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..0bcb39326bb5c7c57dc797226829a20fc7ab5ad9 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/35624.png_5.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/35624.png_6.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/35624.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..1c94ee732315b8b6cc168f563d34bce2764f9cf8 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/35624.png_6.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/35624.png_7.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/35624.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..e5e5aa626a3107ff4e22f165aa6dd1d61fd83188 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/35624.png_7.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/35624.png_8.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/35624.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..58b16e85b249f0ee2971205319d787bf1ef69904 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/35624.png_8.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/35624.png_9.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/35624.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..ef110ece7cf44f939bdaf9cd6c840085882a0640 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/35624.png_9.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/61024.png_1.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/61024.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..a9228a101470a1554f8066e791a6e43c0f725de2 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/61024.png_1.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/61024.png_10.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/61024.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..632d16f6be9c4222354d8c0f826d1761d884e0c5 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/61024.png_10.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/61024.png_2.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/61024.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..f696a26ee06711dba23c31c9c664def2a7f750df Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/61024.png_2.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/61024.png_3.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/61024.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..95f03a3ca2243d77f3337f4a50bbbb03ccd8ae9c Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/61024.png_3.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/61024.png_4.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/61024.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..72e92d0cec2185444dae27770cfaed8d2efae3c5 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/61024.png_4.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/61024.png_5.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/61024.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..189bf6d553070cd48c6e41988197b29d31605f1b Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/61024.png_5.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/61024.png_6.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/61024.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..b85775f1843c7c3fba79182e682b4b53cf461086 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/61024.png_6.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/61024.png_7.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/61024.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..5a90d5e37bf0459a6d361cbd50f412d2991f64ef Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/61024.png_7.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/61024.png_8.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/61024.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..1d56f0aa82d65128bf040c338e0e0776af1a7222 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/61024.png_8.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/61024.png_9.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/61024.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..4f271c7f3a04319776a4d272385d352f0ef77636 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/61024.png_9.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/66390.png_1.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/66390.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..69d5a799bcd01b2f98a1fa8446ee41be849f570a Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/66390.png_1.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/66390.png_10.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/66390.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..6aa90f29713f51a2f14de19deff1b8dec5596c21 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/66390.png_10.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/66390.png_2.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/66390.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..4a494a14860730e6119171b34312fff861d90dba Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/66390.png_2.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/66390.png_3.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/66390.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..fe5ce36b10cf2715132771f529fe1fd46a83bb69 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/66390.png_3.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/66390.png_4.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/66390.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..3bdb4bebca6d4bcad2980cf2e38f37fc46fccadc Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/66390.png_4.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/66390.png_5.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/66390.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..60cb3107f9ca5c0be36b6e27078f74d8c2b47737 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/66390.png_5.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/66390.png_6.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/66390.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..3cc2f20168b4e3a11af9370ae01fbd1b7c9a3031 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/66390.png_6.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/66390.png_7.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/66390.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..2603000c17fd2fc203399008f1acfa36e5a33119 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/66390.png_7.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/66390.png_8.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/66390.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..1949748d2b5e6ff84dfcc3e8827c61b5d3544aab Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/66390.png_8.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/66390.png_9.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/66390.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..66a5ceee9e8f258903fe176a520ca1f265475b76 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/66390.png_9.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/pixabay_dog_002352.jpg_1.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/pixabay_dog_002352.jpg_1.png new file mode 100644 index 0000000000000000000000000000000000000000..6a14cfe60cbb3e10802b6774db4c3a7150d41aee Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/pixabay_dog_002352.jpg_1.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/pixabay_dog_002352.jpg_10.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/pixabay_dog_002352.jpg_10.png new file mode 100644 index 0000000000000000000000000000000000000000..db4ffe62de7c2b04f7a1e4a554898fe5c8b0aa15 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/pixabay_dog_002352.jpg_10.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/pixabay_dog_002352.jpg_2.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/pixabay_dog_002352.jpg_2.png new file mode 100644 index 0000000000000000000000000000000000000000..3a0656f99291041b3d0dfef8c061c49a5e79aec0 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/pixabay_dog_002352.jpg_2.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/pixabay_dog_002352.jpg_3.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/pixabay_dog_002352.jpg_3.png new file mode 100644 index 0000000000000000000000000000000000000000..42f6f422a17a69b858d0dcf588638429814113ed Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/pixabay_dog_002352.jpg_3.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/pixabay_dog_002352.jpg_4.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/pixabay_dog_002352.jpg_4.png new file mode 100644 index 0000000000000000000000000000000000000000..2691514ab2ff67f74c224e738b01a4b8cb363072 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/pixabay_dog_002352.jpg_4.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/pixabay_dog_002352.jpg_5.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/pixabay_dog_002352.jpg_5.png new file mode 100644 index 0000000000000000000000000000000000000000..60ab80f6fce3576e0b1277a391428282e6646ee0 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/pixabay_dog_002352.jpg_5.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/pixabay_dog_002352.jpg_6.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/pixabay_dog_002352.jpg_6.png new file mode 100644 index 0000000000000000000000000000000000000000..170454ca29b42bfc358aad68b05dea08cc438e85 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/pixabay_dog_002352.jpg_6.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/pixabay_dog_002352.jpg_7.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/pixabay_dog_002352.jpg_7.png new file mode 100644 index 0000000000000000000000000000000000000000..1770b9d811e08d5893c24b084c90d29d98a3e49f Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/pixabay_dog_002352.jpg_7.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/pixabay_dog_002352.jpg_8.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/pixabay_dog_002352.jpg_8.png new file mode 100644 index 0000000000000000000000000000000000000000..ec09b02c4fdba3e06f7298dab896c959f49359cf Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/pixabay_dog_002352.jpg_8.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/pixabay_dog_002352.jpg_9.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/pixabay_dog_002352.jpg_9.png new file mode 100644 index 0000000000000000000000000000000000000000..0d0db0c04bc474a9977dd67435a5649446fd3806 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/pixabay_dog_002352.jpg_9.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/seed14088.png_1.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/seed14088.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..b26d396c81c056da60af170b4d7f558ba339aa78 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/seed14088.png_1.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/seed14088.png_10.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/seed14088.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..33f98bf2def2e343f6dd508b9c5b40033c09f215 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/seed14088.png_10.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/seed14088.png_2.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/seed14088.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..b7999fbc1dfc33a5b307b0e6790a4870724ffdb9 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/seed14088.png_2.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/seed14088.png_3.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/seed14088.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..f5b9a7e881164f524ba9bf603380a1d7c3092d95 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/seed14088.png_3.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/seed14088.png_4.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/seed14088.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..b7c66d65d6e10307d3b1fc125d326f107f92d76a Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/seed14088.png_4.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/seed14088.png_5.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/seed14088.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..562004744aa8fcb42bf24dbe94fd1402c7031ab8 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/seed14088.png_5.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/seed14088.png_6.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/seed14088.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..3c045782ebcbcc4a94656e457618f69c44f15c2e Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/seed14088.png_6.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/seed14088.png_7.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/seed14088.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..4ca9542f975a2efdd88a20db22877105853ebf60 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/seed14088.png_7.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/seed14088.png_8.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/seed14088.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..208e0cbe3115122cd5c9e970ef57ab511acb7f4c Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/seed14088.png_8.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/seed14088.png_9.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/seed14088.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..5a72881cd0dae555a626c2b2177b8d2656b87705 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/seed14088.png_9.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/seed17282.png_1.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/seed17282.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..7ebf6db0b966ce6bdda3be1c27c27a60b91bbbf7 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/seed17282.png_1.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/seed17282.png_10.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/seed17282.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..0943921377424f4a8c722426d73efae8d4661763 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/seed17282.png_10.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/seed17282.png_2.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/seed17282.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..248c9547d85297fa4727c66db2375856afbc5a50 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/seed17282.png_2.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/seed17282.png_3.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/seed17282.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..532e7864f9baf2143687bfa9d70e8d4cb5f386b3 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/seed17282.png_3.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/seed17282.png_4.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/seed17282.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..1f471c55cc29de74f3c50f30d5b7e18aef9da3b9 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/seed17282.png_4.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/seed17282.png_5.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/seed17282.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..33cee8c602d71b0a8193f3a5f3d2ac9eb50755ae Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/seed17282.png_5.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/seed17282.png_6.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/seed17282.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..09581f4dd55b4fd6cfcff3d763d1ba0b71e0fa4c Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/seed17282.png_6.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/seed17282.png_7.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/seed17282.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..6d18524ea2d81a886e2bcfd7d4b157bfb286830a Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/seed17282.png_7.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/seed17282.png_8.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/seed17282.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..7d73f685b484a3552ab7fc6d16c4709d1b443e9f Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/seed17282.png_8.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/seed17282.png_9.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/seed17282.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..40a7f98c5ead3338351f3be1cdc824676f1ea5fc Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/seed17282.png_9.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/seed23602.png_1.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/seed23602.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..7d73fd8bd2694e2fe716a14d431d81cd9a6c3990 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/seed23602.png_1.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/seed23602.png_10.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/seed23602.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..ada3f09b6b83af1efad5920a46fb3d12e756c86d Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/seed23602.png_10.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/seed23602.png_2.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/seed23602.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..fc12aff78d37279e025f06bfec21924c6d6a2789 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/seed23602.png_2.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/seed23602.png_3.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/seed23602.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..a5a543dde3d70cf74673144399a6697e3bb25449 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/seed23602.png_3.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/seed23602.png_4.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/seed23602.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..068189bc605acec1cb41a7e5a3ebf872d37a644a Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/seed23602.png_4.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/seed23602.png_5.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/seed23602.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..5eb8681014f113b9ec44ce673b7a55508f060786 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/seed23602.png_5.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/seed23602.png_6.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/seed23602.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..b44a1ba3e19dbb4712bcdaf2bc947def2726acb1 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/seed23602.png_6.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/seed23602.png_7.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/seed23602.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..595855e82446134913e19d53e82b864339d710ac Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/seed23602.png_7.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/seed23602.png_8.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/seed23602.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..921be3e0ede4fc64729165f25deb97dce93df5fd Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/seed23602.png_8.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/seed23602.png_9.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/seed23602.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..1e2e9b8cec0ee5f6ec0533d994f474542f81e2e3 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/seed23602.png_9.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/seed24366.png_1.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/seed24366.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..528f388145c6f6d2ae129ed9f2ad984a8fa4b5f1 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/seed24366.png_1.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/seed24366.png_10.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/seed24366.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..c6bfd100b2ac0e3f10662f75bfd5e1d2f82d6509 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/seed24366.png_10.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/seed24366.png_2.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/seed24366.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..513f25978ca463419422718f6a35176838e19c65 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/seed24366.png_2.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/seed24366.png_3.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/seed24366.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..9cecb7551532c2bfb33f16d630a4ebcaa55f60ac Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/seed24366.png_3.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/seed24366.png_4.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/seed24366.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..98e057e5de2400aca5ae7a33027fcff979576941 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/seed24366.png_4.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/seed24366.png_5.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/seed24366.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..8516a0a9c0c46346a0d470218d81592b9b4aedd7 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/seed24366.png_5.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/seed24366.png_6.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/seed24366.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..a3a907867debfc2c569f9372adcdd46d8771578a Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/seed24366.png_6.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/seed24366.png_7.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/seed24366.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..627577292abd467a6530699129a75cc34a103631 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/seed24366.png_7.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/seed24366.png_8.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/seed24366.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..36c79ee472d0f12e0c225a31b15aeeedd2316f53 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/seed24366.png_8.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/seed24366.png_9.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/seed24366.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..6f6d19b54a5112eb5cbe9eb306d7faa633a7ca60 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/seed24366.png_9.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/seed44086.png_1.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/seed44086.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..c46bd636c115a61b8c9e377e062d889d0a77b550 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/seed44086.png_1.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/seed44086.png_10.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/seed44086.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..8fa89fb90be8c925394b47c1f76cec17abb45ce1 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/seed44086.png_10.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/seed44086.png_2.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/seed44086.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..f0da3a6cc624249b1804600541543addb99ff373 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/seed44086.png_2.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/seed44086.png_3.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/seed44086.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..f7c4c98bdb972817cc6373f7206479c67d135291 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/seed44086.png_3.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/seed44086.png_4.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/seed44086.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..78e3d3f1dee1f15a3666facf19dc6bd0191ea725 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/seed44086.png_4.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/seed44086.png_5.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/seed44086.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..c71898b4501fed66f53d323d8cbf47f168c47c98 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/seed44086.png_5.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/seed44086.png_6.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/seed44086.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..e53c169bd94d3107f04d6a1e8ca922b5e687c6d4 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/seed44086.png_6.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/seed44086.png_7.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/seed44086.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..fedb1843b380c06ef9f7ea078adaaccf390e7c63 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/seed44086.png_7.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/seed44086.png_8.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/seed44086.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..0b178261cfecc0a49c137b16db57f0158f8315d4 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/seed44086.png_8.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/seed44086.png_9.png b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/seed44086.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..b4b5ca04e5141b95e30a7bf446213713a04a757d Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230605_124607/vis_data/vis_image/seed44086.png_9.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1.py b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1.py new file mode 100644 index 0000000000000000000000000000000000000000..24166bc7056cbd8e3464ea8fbf43ad9b6b45a47d --- /dev/null +++ b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1.py @@ -0,0 +1,247 @@ +optim_wrapper = dict( + optimizer=dict( + type='AdamW', + lr=0.0001, + weight_decay=0.05, + eps=1e-08, + betas=(0.9, 0.999), + _scope_='mmpretrain'), + paramwise_cfg=dict( + norm_decay_mult=0.0, + bias_decay_mult=0.0, + flat_decay_mult=0.0, + custom_keys=dict({ + '.absolute_pos_embed': dict(decay_mult=0.0), + '.relative_position_bias_table': dict(decay_mult=0.0) + })), + type='AmpOptimWrapper', + dtype='bfloat16', + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=False, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=4096) +model = dict( + type='ImageClassifier', + backbone=dict( + frozen_stages=2, + type='ConvNeXt', + arch='small', + drop_path_rate=0.4, + init_cfg=dict( + type='Pretrained', + checkpoint= + 'https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth', + prefix='backbone')), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=768, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + init_cfg=None), + init_cfg=dict( + type='TruncNormal', layer=['Conv2d', 'Linear'], std=0.02, bias=0.0), + train_cfg=None) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3fake8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3real8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3fake8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3real8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3fake8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3real8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1' diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/epoch_1.pth b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/epoch_1.pth new file mode 100644 index 0000000000000000000000000000000000000000..6cb242818cffe95b79e45d5a7cd853fbe38c17bc --- /dev/null +++ b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/epoch_1.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:aa7cf56016a599f514463c53cf937885571b1614c9c5e5fde4ad484efe8caaed +size 781981338 diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/epoch_10.pth b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/epoch_10.pth new file mode 100644 index 0000000000000000000000000000000000000000..ad6b10be6aadb1c38307000b59d4189ec7c945b4 --- /dev/null +++ b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/epoch_10.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:f6ffc6bfdee81d857816384965b0ccd290685d7b1edba0f8b056bdace9217fea +size 782147930 diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/epoch_2.pth b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/epoch_2.pth new file mode 100644 index 0000000000000000000000000000000000000000..452b32a0a71be659dc0da0f93b0f23e6ba98fa2f --- /dev/null +++ b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/epoch_2.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:98ddfa07ece952bf3a9482946eeff5be73743cbe3cdea872292a234d23943bce +size 782000858 diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/epoch_3.pth b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/epoch_3.pth new file mode 100644 index 0000000000000000000000000000000000000000..c7b9032027dea18e4850dca0634cbbaea6d5458a --- /dev/null +++ b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/epoch_3.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:4d49c910d24f71faf305cd555d42427f979824a24d5bd98a109f2c53fa12ce32 +size 782019162 diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/epoch_4.pth b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/epoch_4.pth new file mode 100644 index 0000000000000000000000000000000000000000..2c6ba599a462bb30013ca3f4c6509b23f0d20c39 --- /dev/null +++ b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/epoch_4.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ca4d1a3cac783cc0d39eabfc90acd124fff6f643586b4d1a1686721456c86d15 +size 782037530 diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/epoch_5.pth b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/epoch_5.pth new file mode 100644 index 0000000000000000000000000000000000000000..0cd42b0214b5568c542050796f47cc7b4c937971 --- /dev/null +++ b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/epoch_5.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:4c11e9e77a27a5ca8bb7717504b12e2372cab4725c0ccffcb28bdb1d54d5c28d +size 782055834 diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/epoch_6.pth b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/epoch_6.pth new file mode 100644 index 0000000000000000000000000000000000000000..e3e18f5a52ccdd8bb804b3ada59f373dd074054c --- /dev/null +++ b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/epoch_6.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ebf875355f08ea67201d495d2cfa47aacd6851a69e2e69b2e4ad3af3aa15130e +size 782074138 diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/epoch_7.pth b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/epoch_7.pth new file mode 100644 index 0000000000000000000000000000000000000000..37a757921b45fa2df781e6bdbc6b5f0b12d90434 --- /dev/null +++ b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/epoch_7.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:a4a331da9afa1eb4ba36e066a6f384c5de7cf7486154c64fe2bb77313fd2ba8f +size 782092506 diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/epoch_8.pth b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/epoch_8.pth new file mode 100644 index 0000000000000000000000000000000000000000..0e8aa88078fd650039eb8b880c373e8f27746f3a --- /dev/null +++ b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/epoch_8.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:118548172313113afee31809908d9513ed80fb01ca8171bc2b788d1056441f41 +size 782110938 diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/epoch_9.pth b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/epoch_9.pth new file mode 100644 index 0000000000000000000000000000000000000000..405b8e58842f5800a218743bf46ef3fd63d16626 --- /dev/null +++ b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/epoch_9.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:214d29f1d9eae853880664c70ee8081d0264a24371c75fb561f763914bf4c0d3 +size 782129434 diff --git a/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/last_checkpoint b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/last_checkpoint new file mode 100644 index 0000000000000000000000000000000000000000..b9bb5b0f3af27325b91330be7afcf9380ee01be5 --- /dev/null +++ b/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/last_checkpoint @@ -0,0 +1 @@ +/mnt/petrelfs/luzeyu/workspace/fakebench/mmpretrain/workdir/convnext_small_4xb256_stylegan3_1m_lr1e-4_aug_5e-1/epoch_10.pth \ No newline at end of file diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/20230605_010733.log b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/20230605_010733.log new file mode 100644 index 0000000000000000000000000000000000000000..8742352a47cedea3a3b2da23d172063121fca71e --- /dev/null +++ b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/20230605_010733.log @@ -0,0 +1,1652 @@ +2023/06/05 01:07:37 - mmengine - INFO - +------------------------------------------------------------ +System environment: + sys.platform: linux + Python: 3.10.9 (main, Mar 8 2023, 10:47:38) [GCC 11.2.0] + CUDA available: True + numpy_random_seed: 682731865 + GPU 0,1,2,3: NVIDIA A100-SXM4-80GB + CUDA_HOME: /mnt/petrelfs/share/cuda-11.6 + NVCC: Cuda compilation tools, release 11.6, V11.6.124 + GCC: gcc (GCC) 7.5.0 + PyTorch: 1.13.1 + PyTorch compiling details: PyTorch built with: + - GCC 9.3 + - C++ Version: 201402 + - Intel(R) oneAPI Math Kernel Library Version 2021.4-Product Build 20210904 for Intel(R) 64 architecture applications + - Intel(R) MKL-DNN v2.6.0 (Git Hash 52b5f107dd9cf10910aaa19cb47f3abf9b349815) + - OpenMP 201511 (a.k.a. OpenMP 4.5) + - LAPACK is enabled (usually provided by MKL) + - NNPACK is enabled + - CPU capability usage: AVX2 + - CUDA Runtime 11.6 + - NVCC architecture flags: -gencode;arch=compute_37,code=sm_37;-gencode;arch=compute_50,code=sm_50;-gencode;arch=compute_60,code=sm_60;-gencode;arch=compute_61,code=sm_61;-gencode;arch=compute_70,code=sm_70;-gencode;arch=compute_75,code=sm_75;-gencode;arch=compute_80,code=sm_80;-gencode;arch=compute_86,code=sm_86;-gencode;arch=compute_37,code=compute_37 + - CuDNN 8.3.2 (built against CUDA 11.5) + - Magma 2.6.1 + - Build settings: BLAS_INFO=mkl, BUILD_TYPE=Release, CUDA_VERSION=11.6, CUDNN_VERSION=8.3.2, CXX_COMPILER=/opt/rh/devtoolset-9/root/usr/bin/c++, CXX_FLAGS= -fabi-version=11 -Wno-deprecated -fvisibility-inlines-hidden -DUSE_PTHREADPOOL -fopenmp -DNDEBUG -DUSE_KINETO -DUSE_FBGEMM -DUSE_QNNPACK -DUSE_PYTORCH_QNNPACK -DUSE_XNNPACK -DSYMBOLICATE_MOBILE_DEBUG_HANDLE -DEDGE_PROFILER_USE_KINETO -O2 -fPIC -Wno-narrowing -Wall -Wextra -Werror=return-type -Werror=non-virtual-dtor -Wno-missing-field-initializers -Wno-type-limits -Wno-array-bounds -Wno-unknown-pragmas -Wunused-local-typedefs -Wno-unused-parameter -Wno-unused-function -Wno-unused-result -Wno-strict-overflow -Wno-strict-aliasing -Wno-error=deprecated-declarations -Wno-stringop-overflow -Wno-psabi -Wno-error=pedantic -Wno-error=redundant-decls -Wno-error=old-style-cast -fdiagnostics-color=always -faligned-new -Wno-unused-but-set-variable -Wno-maybe-uninitialized -fno-math-errno -fno-trapping-math -Werror=format -Werror=cast-function-type -Wno-stringop-overflow, LAPACK_INFO=mkl, PERF_WITH_AVX=1, PERF_WITH_AVX2=1, PERF_WITH_AVX512=1, TORCH_VERSION=1.13.1, USE_CUDA=ON, USE_CUDNN=ON, USE_EXCEPTION_PTR=1, USE_GFLAGS=OFF, USE_GLOG=OFF, USE_MKL=ON, USE_MKLDNN=ON, USE_MPI=OFF, USE_NCCL=ON, USE_NNPACK=ON, USE_OPENMP=ON, USE_ROCM=OFF, + + TorchVision: 0.14.1 + OpenCV: 4.7.0 + MMEngine: 0.7.3 + +Runtime environment: + cudnn_benchmark: True + mp_cfg: {'mp_start_method': 'fork', 'opencv_num_threads': 0} + dist_cfg: {'backend': 'nccl'} + seed: None + deterministic: False + Distributed launcher: slurm + Distributed training: True + GPU number: 4 +------------------------------------------------------------ + +2023/06/05 01:07:41 - mmengine - INFO - Config: +optim_wrapper = dict( + optimizer=dict( + type='AdamW', + lr=0.0004, + weight_decay=0.05, + eps=1e-08, + betas=(0.9, 0.999), + _scope_='mmpretrain'), + paramwise_cfg=dict( + norm_decay_mult=0.0, + bias_decay_mult=0.0, + flat_decay_mult=0.0, + custom_keys=dict({ + '.absolute_pos_embed': dict(decay_mult=0.0), + '.relative_position_bias_table': dict(decay_mult=0.0) + })), + type='AmpOptimWrapper', + dtype='bfloat16', + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=False, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=4096) +model = dict( + type='ImageClassifier', + backbone=dict( + frozen_stages=2, + type='ConvNeXt', + arch='small', + drop_path_rate=0.4, + init_cfg=dict( + type='Pretrained', + checkpoint= + 'https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth', + prefix='backbone')), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=768, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + init_cfg=None), + init_cfg=dict( + type='TruncNormal', layer=['Conv2d', 'Linear'], std=0.02, bias=0.0), + train_cfg=None) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3fake8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3real8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3fake8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3real8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3fake8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3real8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1' + +2023/06/05 01:07:53 - mmengine - INFO - Hooks will be executed in the following order: +before_run: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_load_checkpoint: +(ABOVE_NORMAL) EMAHook + -------------------- +before_train: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_train_epoch: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(NORMAL ) DistSamplerSeedHook + -------------------- +before_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook + -------------------- +after_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +after_train_epoch: +(NORMAL ) IterTimerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_val_epoch: +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook + -------------------- +before_val_iter: +(NORMAL ) IterTimerHook + -------------------- +after_val_iter: +(NORMAL ) IterTimerHook +(NORMAL ) VisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_val_epoch: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_save_checkpoint: +(ABOVE_NORMAL) EMAHook + -------------------- +after_train: +(VERY_LOW ) CheckpointHook + -------------------- +before_test_epoch: +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook + -------------------- +before_test_iter: +(NORMAL ) IterTimerHook + -------------------- +after_test_iter: +(NORMAL ) IterTimerHook +(NORMAL ) VisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_test_epoch: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_run: +(BELOW_NORMAL) LoggerHook + -------------------- +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.2.0.weight:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.2.0.bias:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.2.1.bias:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.3.0.weight:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.3.0.bias:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.3.1.bias:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.2.0.gamma:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.2.0.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.2.0.norm.weight:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.2.0.norm.bias:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.2.0.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.2.0.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.2.1.gamma:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.2.1.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.2.1.norm.weight:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.2.1.norm.bias:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.2.1.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.2.1.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.2.2.gamma:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.2.2.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.2.2.norm.weight:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.2.2.norm.bias:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.2.2.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.2.2.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.2.3.gamma:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.2.3.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.2.3.norm.weight:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.2.3.norm.bias:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.2.3.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.2.3.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.2.4.gamma:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.2.4.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.2.4.norm.weight:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.2.4.norm.bias:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.2.4.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.2.4.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.2.5.gamma:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.2.5.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.2.5.norm.weight:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.2.5.norm.bias:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.2.5.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.2.5.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.2.6.gamma:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.2.6.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.2.6.norm.weight:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.2.6.norm.bias:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.2.6.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.2.6.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.2.7.gamma:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.2.7.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.2.7.norm.weight:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.2.7.norm.bias:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.2.7.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.2.7.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.2.8.gamma:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.2.8.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.2.8.norm.weight:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.2.8.norm.bias:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.2.8.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.2.8.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.2.9.gamma:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.2.9.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.2.9.norm.weight:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.2.9.norm.bias:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.2.9.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.2.9.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.2.10.gamma:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.2.10.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.2.10.norm.weight:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.2.10.norm.bias:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.2.10.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.2.10.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.2.11.gamma:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.2.11.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.2.11.norm.weight:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.2.11.norm.bias:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.2.11.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.2.11.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.2.12.gamma:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.2.12.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.2.12.norm.weight:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.2.12.norm.bias:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.2.12.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.2.12.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.2.13.gamma:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.2.13.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.2.13.norm.weight:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.2.13.norm.bias:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.2.13.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.2.13.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.2.14.gamma:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.2.14.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.2.14.norm.weight:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.2.14.norm.bias:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.2.14.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.2.14.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.2.15.gamma:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.2.15.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.2.15.norm.weight:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.2.15.norm.bias:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.2.15.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.2.15.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.2.16.gamma:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.2.16.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.2.16.norm.weight:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.2.16.norm.bias:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.2.16.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.2.16.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.2.17.gamma:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.2.17.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.2.17.norm.weight:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.2.17.norm.bias:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.2.17.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.2.17.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.2.18.gamma:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.2.18.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.2.18.norm.weight:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.2.18.norm.bias:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.2.18.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.2.18.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.2.19.gamma:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.2.19.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.2.19.norm.weight:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.2.19.norm.bias:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.2.19.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.2.19.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.2.20.gamma:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.2.20.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.2.20.norm.weight:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.2.20.norm.bias:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.2.20.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.2.20.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.2.21.gamma:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.2.21.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.2.21.norm.weight:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.2.21.norm.bias:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.2.21.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.2.21.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.2.22.gamma:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.2.22.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.2.22.norm.weight:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.2.22.norm.bias:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.2.22.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.2.22.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.2.23.gamma:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.2.23.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.2.23.norm.weight:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.2.23.norm.bias:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.2.23.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.2.23.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.2.24.gamma:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.2.24.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.2.24.norm.weight:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.2.24.norm.bias:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.2.24.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.2.24.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.2.25.gamma:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.2.25.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.2.25.norm.weight:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.2.25.norm.bias:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.2.25.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.2.25.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.2.26.gamma:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.2.26.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.2.26.norm.weight:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.2.26.norm.bias:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.2.26.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.2.26.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.3.0.gamma:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.3.0.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.3.0.norm.weight:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.3.0.norm.bias:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.3.0.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.3.0.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.3.1.gamma:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.3.1.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.3.1.norm.weight:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.3.1.norm.bias:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.3.1.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.3.1.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.3.2.gamma:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.3.2.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.3.2.norm.weight:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.3.2.norm.bias:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.3.2.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.stages.3.2.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.norm3.weight:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- backbone.norm3.bias:weight_decay=0.0 +2023/06/05 01:07:59 - mmengine - INFO - paramwise_options -- head.fc.bias:weight_decay=0.0 +2023/06/05 01:08:05 - mmengine - INFO - load backbone in model from: https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth +Name of parameter - Initialization information + +backbone.downsample_layers.0.0.weight - torch.Size([96, 3, 4, 4]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.0.0.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.0.1.weight - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.0.1.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.1.0.weight - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.1.0.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.1.1.weight - torch.Size([192, 96, 2, 2]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.1.1.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.2.0.weight - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.2.0.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.2.1.weight - torch.Size([384, 192, 2, 2]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.2.1.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.3.0.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.3.0.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.3.1.weight - torch.Size([768, 384, 2, 2]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.3.1.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.0.gamma - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.0.depthwise_conv.weight - torch.Size([96, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.0.depthwise_conv.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.0.norm.weight - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.0.norm.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.0.pointwise_conv1.weight - torch.Size([384, 96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.0.pointwise_conv1.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.0.pointwise_conv2.weight - torch.Size([96, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.0.pointwise_conv2.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.1.gamma - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.1.depthwise_conv.weight - torch.Size([96, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.1.depthwise_conv.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.1.norm.weight - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.1.norm.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.1.pointwise_conv1.weight - torch.Size([384, 96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.1.pointwise_conv1.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.1.pointwise_conv2.weight - torch.Size([96, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.1.pointwise_conv2.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.2.gamma - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.2.depthwise_conv.weight - torch.Size([96, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.2.depthwise_conv.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.2.norm.weight - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.2.norm.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.2.pointwise_conv1.weight - torch.Size([384, 96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.2.pointwise_conv1.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.2.pointwise_conv2.weight - torch.Size([96, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.2.pointwise_conv2.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.0.gamma - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.0.depthwise_conv.weight - torch.Size([192, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.0.depthwise_conv.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.0.norm.weight - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.0.norm.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.0.pointwise_conv1.weight - torch.Size([768, 192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.0.pointwise_conv1.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.0.pointwise_conv2.weight - torch.Size([192, 768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.0.pointwise_conv2.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.1.gamma - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.1.depthwise_conv.weight - torch.Size([192, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.1.depthwise_conv.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.1.norm.weight - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.1.norm.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.1.pointwise_conv1.weight - torch.Size([768, 192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.1.pointwise_conv1.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.1.pointwise_conv2.weight - torch.Size([192, 768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.1.pointwise_conv2.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.2.gamma - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.2.depthwise_conv.weight - torch.Size([192, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.2.depthwise_conv.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.2.norm.weight - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.2.norm.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.2.pointwise_conv1.weight - torch.Size([768, 192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.2.pointwise_conv1.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.2.pointwise_conv2.weight - torch.Size([192, 768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.2.pointwise_conv2.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.0.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.0.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.0.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.0.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.0.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.0.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.0.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.0.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.0.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.1.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.1.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.1.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.1.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.1.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.1.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.1.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.1.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.1.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.2.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.2.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.2.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.2.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.2.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.2.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.2.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.2.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.2.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.3.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.3.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.3.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.3.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.3.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.3.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.3.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.3.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.3.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.4.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.4.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.4.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.4.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.4.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.4.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.4.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.4.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.4.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.5.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.5.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.5.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.5.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.5.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.5.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.5.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.5.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.5.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.6.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.6.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.6.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.6.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.6.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.6.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.6.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.6.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.6.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.7.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.7.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.7.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.7.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.7.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.7.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.7.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.7.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.7.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.8.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.8.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.8.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.8.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.8.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.8.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.8.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.8.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.8.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.9.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.9.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.9.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.9.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.9.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.9.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.9.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.9.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.9.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.10.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.10.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.10.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.10.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.10.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.10.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.10.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.10.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.10.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.11.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.11.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.11.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.11.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.11.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.11.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.11.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.11.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.11.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.12.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.12.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.12.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.12.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.12.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.12.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.12.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.12.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.12.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.13.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.13.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.13.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.13.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.13.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.13.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.13.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.13.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.13.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.14.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.14.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.14.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.14.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.14.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.14.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.14.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.14.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.14.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.15.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.15.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.15.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.15.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.15.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.15.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.15.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.15.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.15.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.16.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.16.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.16.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.16.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.16.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.16.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.16.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.16.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.16.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.17.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.17.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.17.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.17.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.17.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.17.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.17.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.17.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.17.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.18.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.18.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.18.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.18.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.18.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.18.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.18.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.18.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.18.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.19.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.19.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.19.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.19.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.19.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.19.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.19.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.19.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.19.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.20.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.20.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.20.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.20.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.20.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.20.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.20.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.20.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.20.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.21.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.21.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.21.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.21.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.21.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.21.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.21.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.21.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.21.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.22.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.22.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.22.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.22.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.22.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.22.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.22.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.22.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.22.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.23.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.23.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.23.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.23.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.23.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.23.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.23.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.23.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.23.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.24.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.24.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.24.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.24.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.24.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.24.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.24.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.24.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.24.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.25.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.25.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.25.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.25.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.25.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.25.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.25.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.25.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.25.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.26.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.26.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.26.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.26.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.26.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.26.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.26.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.26.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.26.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.0.gamma - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.0.depthwise_conv.weight - torch.Size([768, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.0.depthwise_conv.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.0.norm.weight - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.0.norm.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.0.pointwise_conv1.weight - torch.Size([3072, 768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.0.pointwise_conv1.bias - torch.Size([3072]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.0.pointwise_conv2.weight - torch.Size([768, 3072]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.0.pointwise_conv2.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.1.gamma - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.1.depthwise_conv.weight - torch.Size([768, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.1.depthwise_conv.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.1.norm.weight - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.1.norm.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.1.pointwise_conv1.weight - torch.Size([3072, 768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.1.pointwise_conv1.bias - torch.Size([3072]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.1.pointwise_conv2.weight - torch.Size([768, 3072]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.1.pointwise_conv2.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.2.gamma - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.2.depthwise_conv.weight - torch.Size([768, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.2.depthwise_conv.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.2.norm.weight - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.2.norm.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.2.pointwise_conv1.weight - torch.Size([3072, 768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.2.pointwise_conv1.bias - torch.Size([3072]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.2.pointwise_conv2.weight - torch.Size([768, 3072]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.2.pointwise_conv2.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.norm3.weight - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.norm3.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +head.fc.weight - torch.Size([2, 768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +head.fc.bias - torch.Size([2]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 +2023/06/05 01:08:05 - mmengine - WARNING - "FileClient" will be deprecated in future. Please use io functions in https://mmengine.readthedocs.io/en/latest/api/fileio.html#file-io +2023/06/05 01:08:05 - mmengine - WARNING - "HardDiskBackend" is the alias of "LocalBackend" and the former will be deprecated in future. +2023/06/05 01:08:05 - mmengine - INFO - Checkpoints will be saved to /mnt/petrelfs/luzeyu/workspace/fakebench/mmpretrain/workdir/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1. +2023/06/05 01:10:33 - mmengine - INFO - Epoch(train) [1][100/171] lr: 3.9678e-04 eta: 0:39:35 time: 1.4661 data_time: 0.5541 memory: 15587 loss: 0.0548 +2023/06/05 01:12:06 - mmengine - INFO - Exp name: convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1_20230605_010733 +2023/06/05 01:12:06 - mmengine - INFO - Saving checkpoint at 1 epochs +2023/06/05 01:14:26 - mmengine - INFO - Epoch(val) [1][100/171] eta: 0:01:33 time: 0.4189 data_time: 0.0080 memory: 15587 +2023/06/05 01:17:53 - mmengine - INFO - Epoch(val) [1][171/171] accuracy/top1: 53.1834 single-label/precision_classwise: [51.83223342895508, 61.502540588378906] single-label/recall_classwise: [89.23556518554688, 17.175739288330078] single-label/f1-score_classwise: [65.57526397705078, 26.85243034362793] data_time: 1.1971 time: 1.6268 +2023/06/05 01:20:12 - mmengine - INFO - Epoch(train) [2][100/171] lr: 3.7650e-04 eta: 0:33:37 time: 1.3117 data_time: 0.3619 memory: 15582 loss: 0.0314 +2023/06/05 01:21:42 - mmengine - INFO - Exp name: convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1_20230605_010733 +2023/06/05 01:21:42 - mmengine - INFO - Saving checkpoint at 2 epochs +2023/06/05 01:24:00 - mmengine - INFO - Epoch(val) [2][100/171] eta: 0:01:32 time: 0.4221 data_time: 0.0115 memory: 15582 +2023/06/05 01:27:27 - mmengine - INFO - Epoch(val) [2][171/171] accuracy/top1: 53.6756 single-label/precision_classwise: [52.119476318359375, 63.219566345214844] single-label/recall_classwise: [89.68112182617188, 17.714553833007812] single-label/f1-score_classwise: [65.9254379272461, 27.674518585205078] data_time: 1.2081 time: 1.6207 +2023/06/05 01:29:46 - mmengine - INFO - Epoch(train) [3][100/171] lr: 3.3942e-04 eta: 0:29:07 time: 1.2893 data_time: 0.6879 memory: 15582 loss: 0.0181 +2023/06/05 01:31:22 - mmengine - INFO - Exp name: convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1_20230605_010733 +2023/06/05 01:31:22 - mmengine - INFO - Saving checkpoint at 3 epochs +2023/06/05 01:33:41 - mmengine - INFO - Epoch(val) [3][100/171] eta: 0:01:33 time: 0.4164 data_time: 0.0076 memory: 15582 +2023/06/05 01:37:10 - mmengine - INFO - Epoch(val) [3][171/171] accuracy/top1: 54.2388 single-label/precision_classwise: [52.45213317871094, 65.03425598144531] single-label/recall_classwise: [90.06368255615234, 18.458141326904297] single-label/f1-score_classwise: [66.29485321044922, 28.754987716674805] data_time: 1.2197 time: 1.6318 +2023/06/05 01:39:33 - mmengine - INFO - Epoch(train) [4][100/171] lr: 2.8919e-04 eta: 0:25:15 time: 1.3836 data_time: 0.2745 memory: 15582 loss: 0.0161 +2023/06/05 01:41:07 - mmengine - INFO - Exp name: convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1_20230605_010733 +2023/06/05 01:41:07 - mmengine - INFO - Saving checkpoint at 4 epochs +2023/06/05 01:43:25 - mmengine - INFO - Epoch(val) [4][100/171] eta: 0:01:33 time: 0.4170 data_time: 0.0087 memory: 15582 +2023/06/05 01:46:56 - mmengine - INFO - Epoch(val) [4][171/171] accuracy/top1: 54.8014 single-label/precision_classwise: [52.783512115478516, 66.88667297363281] single-label/recall_classwise: [90.5184097290039, 19.128515243530273] single-label/f1-score_classwise: [66.68269348144531, 29.74922752380371] data_time: 1.2254 time: 1.6392 +2023/06/05 01:49:18 - mmengine - INFO - Epoch(train) [5][100/171] lr: 2.3072e-04 eta: 0:21:18 time: 1.2235 data_time: 0.0019 memory: 15582 loss: 0.0169 +2023/06/05 01:50:56 - mmengine - INFO - Exp name: convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1_20230605_010733 +2023/06/05 01:50:56 - mmengine - INFO - Saving checkpoint at 5 epochs +2023/06/05 01:53:13 - mmengine - INFO - Epoch(val) [5][100/171] eta: 0:01:32 time: 0.4193 data_time: 0.0101 memory: 15582 +2023/06/05 01:56:39 - mmengine - INFO - Epoch(val) [5][171/171] accuracy/top1: 55.5260 single-label/precision_classwise: [53.223388671875, 68.82099151611328] single-label/recall_classwise: [90.7887191772461, 20.30681610107422] single-label/f1-score_classwise: [67.10662841796875, 31.360252380371094] data_time: 1.1877 time: 1.6021 +2023/06/05 01:59:06 - mmengine - INFO - Epoch(train) [6][100/171] lr: 1.6973e-04 eta: 0:17:28 time: 1.5369 data_time: 0.3604 memory: 15582 loss: 0.0098 +2023/06/05 02:00:08 - mmengine - INFO - Exp name: convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1_20230605_010733 +2023/06/05 02:00:39 - mmengine - INFO - Exp name: convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1_20230605_010733 +2023/06/05 02:00:39 - mmengine - INFO - Saving checkpoint at 6 epochs +2023/06/05 02:02:59 - mmengine - INFO - Epoch(val) [6][100/171] eta: 0:01:34 time: 0.4175 data_time: 0.0075 memory: 15582 +2023/06/05 02:06:30 - mmengine - INFO - Epoch(val) [6][171/171] accuracy/top1: 56.3771 single-label/precision_classwise: [53.74812316894531, 70.88338470458984] single-label/recall_classwise: [91.0601806640625, 21.736793518066406] single-label/f1-score_classwise: [67.5971450805664, 33.27088165283203] data_time: 1.2262 time: 1.6402 +2023/06/05 02:08:57 - mmengine - INFO - Epoch(train) [7][100/171] lr: 1.1219e-04 eta: 0:13:32 time: 1.3888 data_time: 0.0020 memory: 15582 loss: 0.0072 +2023/06/05 02:10:33 - mmengine - INFO - Exp name: convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1_20230605_010733 +2023/06/05 02:10:33 - mmengine - INFO - Saving checkpoint at 7 epochs +2023/06/05 02:12:53 - mmengine - INFO - Epoch(val) [7][100/171] eta: 0:01:33 time: 0.4182 data_time: 0.0080 memory: 15582 +2023/06/05 02:16:25 - mmengine - INFO - Epoch(val) [7][171/171] accuracy/top1: 57.3037 single-label/precision_classwise: [54.32814407348633, 72.94538879394531] single-label/recall_classwise: [91.34652709960938, 23.30290412902832] single-label/f1-score_classwise: [68.13384246826172, 35.321964263916016] data_time: 1.2242 time: 1.6425 +2023/06/05 02:18:47 - mmengine - INFO - Epoch(train) [8][100/171] lr: 6.3737e-05 eta: 0:09:34 time: 1.2699 data_time: 0.1410 memory: 15582 loss: 0.0097 +2023/06/05 02:20:19 - mmengine - INFO - Exp name: convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1_20230605_010733 +2023/06/05 02:20:19 - mmengine - INFO - Saving checkpoint at 8 epochs +2023/06/05 02:22:39 - mmengine - INFO - Epoch(val) [8][100/171] eta: 0:01:33 time: 0.4160 data_time: 0.0079 memory: 15582 +2023/06/05 02:26:05 - mmengine - INFO - Epoch(val) [8][171/171] accuracy/top1: 58.5159 single-label/precision_classwise: [55.106117248535156, 75.26648712158203] single-label/recall_classwise: [91.6282958984375, 25.444435119628906] single-label/f1-score_classwise: [68.82202911376953, 38.03188705444336] data_time: 1.1954 time: 1.6103 +2023/06/05 02:28:27 - mmengine - INFO - Epoch(train) [9][100/171] lr: 2.9112e-05 eta: 0:05:36 time: 1.5239 data_time: 0.0018 memory: 15582 loss: 0.0051 +2023/06/05 02:30:12 - mmengine - INFO - Exp name: convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1_20230605_010733 +2023/06/05 02:30:12 - mmengine - INFO - Saving checkpoint at 9 epochs +2023/06/05 02:32:31 - mmengine - INFO - Epoch(val) [9][100/171] eta: 0:01:33 time: 0.4195 data_time: 0.0089 memory: 15582 +2023/06/05 02:35:59 - mmengine - INFO - Epoch(val) [9][171/171] accuracy/top1: 60.1397 single-label/precision_classwise: [56.179588317871094, 77.9279556274414] single-label/recall_classwise: [91.95702362060547, 28.361589431762695] single-label/f1-score_classwise: [69.74788665771484, 41.587547302246094] data_time: 1.1925 time: 1.6055 +2023/06/05 02:38:18 - mmengine - INFO - Epoch(train) [10][100/171] lr: 1.1704e-05 eta: 0:01:38 time: 1.2706 data_time: 0.0779 memory: 15582 loss: 0.0090 +2023/06/05 02:39:53 - mmengine - INFO - Exp name: convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1_20230605_010733 +2023/06/05 02:39:53 - mmengine - INFO - Saving checkpoint at 10 epochs +2023/06/05 02:42:12 - mmengine - INFO - Epoch(val) [10][100/171] eta: 0:01:32 time: 0.4177 data_time: 0.0081 memory: 15582 +2023/06/05 02:45:43 - mmengine - INFO - Epoch(val) [10][171/171] accuracy/top1: 62.0713 single-label/precision_classwise: [57.52552032470703, 80.251220703125] single-label/recall_classwise: [92.09447479248047, 32.08524703979492] single-label/f1-score_classwise: [70.8165054321289, 45.842281341552734] data_time: 1.1848 time: 1.5980 diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/20230605_010733.json b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/20230605_010733.json new file mode 100644 index 0000000000000000000000000000000000000000..642177873a7dafbf9df3619575bfdf2815d51828 --- /dev/null +++ b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/20230605_010733.json @@ -0,0 +1,20 @@ +{"lr": 0.0003967834896001145, "data_time": 0.554125165939331, "loss": 0.05475399363785982, "time": 1.4660505294799804, "epoch": 1, "memory": 15587, "step": 100} +{"accuracy/top1": 53.183380126953125, "data_time": 1.197097284752026, "time": 1.6268110888743261, "step": 1} +{"lr": 0.000376497381485265, "data_time": 0.36186254024505615, "loss": 0.03136100564152002, "time": 1.311652660369873, "epoch": 2, "memory": 15582, "step": 271} +{"accuracy/top1": 53.67559814453125, "data_time": 1.2080852791320447, "time": 1.6206542336663534, "step": 2} +{"lr": 0.00033942391477812043, "data_time": 0.6879224300384521, "loss": 0.018135064095258713, "time": 1.2893144130706786, "epoch": 3, "memory": 15582, "step": 442} +{"accuracy/top1": 54.23878479003906, "data_time": 1.219717217046161, "time": 1.631797710130381, "step": 3} +{"lr": 0.0002891920987060062, "data_time": 0.2744947910308838, "loss": 0.01606695996597409, "time": 1.3836426496505738, "epoch": 4, "memory": 15582, "step": 613} +{"accuracy/top1": 54.80139923095703, "data_time": 1.2254282568776331, "time": 1.6391759833624198, "step": 4} +{"lr": 0.000230718973411704, "data_time": 0.001909160614013672, "loss": 0.016870868671685458, "time": 1.223513698577881, "epoch": 5, "memory": 15582, "step": 784} +{"accuracy/top1": 55.52598571777344, "data_time": 1.187709550524867, "time": 1.602063507534737, "step": 5} +{"lr": 0.00016972829580523899, "data_time": 0.3604011058807373, "loss": 0.00983569915406406, "time": 1.5368563890457154, "epoch": 6, "memory": 15582, "step": 955} +{"accuracy/top1": 56.3770637512207, "data_time": 1.2261716085810994, "time": 1.640196746171907, "step": 6} +{"lr": 0.00011219025835777107, "data_time": 0.0019702672958374023, "loss": 0.0071703892201185225, "time": 1.3887849569320678, "epoch": 7, "memory": 15582, "step": 1126} +{"accuracy/top1": 57.303688049316406, "data_time": 1.2241768601328828, "time": 1.6424788929695306, "step": 7} +{"lr": 6.373708506573799e-05, "data_time": 0.14099416732788086, "loss": 0.00969743775203824, "time": 1.2699143648147584, "epoch": 8, "memory": 15582, "step": 1297} +{"accuracy/top1": 58.51591491699219, "data_time": 1.1953882103742555, "time": 1.6103371229282646, "step": 8} +{"lr": 2.9111710124073225e-05, "data_time": 0.0018050193786621094, "loss": 0.005062832310795784, "time": 1.5239089965820312, "epoch": 9, "memory": 15582, "step": 1468} +{"accuracy/top1": 60.13965606689453, "data_time": 1.192514610844989, "time": 1.6054761686990426, "step": 9} +{"lr": 1.1703506481239862e-05, "data_time": 0.07794067859649659, "loss": 0.00901974057778716, "time": 1.2705679178237914, "epoch": 10, "memory": 15582, "step": 1639} +{"accuracy/top1": 62.07131576538086, "data_time": 1.1848429247390393, "time": 1.5980328196703002, "step": 10} diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/config.py b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/config.py new file mode 100644 index 0000000000000000000000000000000000000000..e452889e6187289e7e8fade71cf0f2603ccc9eef --- /dev/null +++ b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/config.py @@ -0,0 +1,247 @@ +optim_wrapper = dict( + optimizer=dict( + type='AdamW', + lr=0.0004, + weight_decay=0.05, + eps=1e-08, + betas=(0.9, 0.999), + _scope_='mmpretrain'), + paramwise_cfg=dict( + norm_decay_mult=0.0, + bias_decay_mult=0.0, + flat_decay_mult=0.0, + custom_keys=dict({ + '.absolute_pos_embed': dict(decay_mult=0.0), + '.relative_position_bias_table': dict(decay_mult=0.0) + })), + type='AmpOptimWrapper', + dtype='bfloat16', + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=False, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=4096) +model = dict( + type='ImageClassifier', + backbone=dict( + frozen_stages=2, + type='ConvNeXt', + arch='small', + drop_path_rate=0.4, + init_cfg=dict( + type='Pretrained', + checkpoint= + 'https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth', + prefix='backbone')), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=768, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + init_cfg=None), + init_cfg=dict( + type='TruncNormal', layer=['Conv2d', 'Linear'], std=0.02, bias=0.0), + train_cfg=None) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3fake8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3real8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3fake8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3real8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3fake8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3real8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1' diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/events.out.tfevents.1685898468.SH-IDC1-10-140-24-130.106936.0 b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/events.out.tfevents.1685898468.SH-IDC1-10-140-24-130.106936.0 new file mode 100644 index 0000000000000000000000000000000000000000..ec227750f01716c43b555132cf8706e4a06e21aa --- /dev/null +++ b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/events.out.tfevents.1685898468.SH-IDC1-10-140-24-130.106936.0 @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:92c80fcd4a333710c05dab7c3aeb08abd5e9218e15820e10d573b81ccf8e83d9 +size 7133722 diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/scalars.json b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/scalars.json new file mode 100644 index 0000000000000000000000000000000000000000..642177873a7dafbf9df3619575bfdf2815d51828 --- /dev/null +++ b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/scalars.json @@ -0,0 +1,20 @@ +{"lr": 0.0003967834896001145, "data_time": 0.554125165939331, "loss": 0.05475399363785982, "time": 1.4660505294799804, "epoch": 1, "memory": 15587, "step": 100} +{"accuracy/top1": 53.183380126953125, "data_time": 1.197097284752026, "time": 1.6268110888743261, "step": 1} +{"lr": 0.000376497381485265, "data_time": 0.36186254024505615, "loss": 0.03136100564152002, "time": 1.311652660369873, "epoch": 2, "memory": 15582, "step": 271} +{"accuracy/top1": 53.67559814453125, "data_time": 1.2080852791320447, "time": 1.6206542336663534, "step": 2} +{"lr": 0.00033942391477812043, "data_time": 0.6879224300384521, "loss": 0.018135064095258713, "time": 1.2893144130706786, "epoch": 3, "memory": 15582, "step": 442} +{"accuracy/top1": 54.23878479003906, "data_time": 1.219717217046161, "time": 1.631797710130381, "step": 3} +{"lr": 0.0002891920987060062, "data_time": 0.2744947910308838, "loss": 0.01606695996597409, "time": 1.3836426496505738, "epoch": 4, "memory": 15582, "step": 613} +{"accuracy/top1": 54.80139923095703, "data_time": 1.2254282568776331, "time": 1.6391759833624198, "step": 4} +{"lr": 0.000230718973411704, "data_time": 0.001909160614013672, "loss": 0.016870868671685458, "time": 1.223513698577881, "epoch": 5, "memory": 15582, "step": 784} +{"accuracy/top1": 55.52598571777344, "data_time": 1.187709550524867, "time": 1.602063507534737, "step": 5} +{"lr": 0.00016972829580523899, "data_time": 0.3604011058807373, "loss": 0.00983569915406406, "time": 1.5368563890457154, "epoch": 6, "memory": 15582, "step": 955} +{"accuracy/top1": 56.3770637512207, "data_time": 1.2261716085810994, "time": 1.640196746171907, "step": 6} +{"lr": 0.00011219025835777107, "data_time": 0.0019702672958374023, "loss": 0.0071703892201185225, "time": 1.3887849569320678, "epoch": 7, "memory": 15582, "step": 1126} +{"accuracy/top1": 57.303688049316406, "data_time": 1.2241768601328828, "time": 1.6424788929695306, "step": 7} +{"lr": 6.373708506573799e-05, "data_time": 0.14099416732788086, "loss": 0.00969743775203824, "time": 1.2699143648147584, "epoch": 8, "memory": 15582, "step": 1297} +{"accuracy/top1": 58.51591491699219, "data_time": 1.1953882103742555, "time": 1.6103371229282646, "step": 8} +{"lr": 2.9111710124073225e-05, "data_time": 0.0018050193786621094, "loss": 0.005062832310795784, "time": 1.5239089965820312, "epoch": 9, "memory": 15582, "step": 1468} +{"accuracy/top1": 60.13965606689453, "data_time": 1.192514610844989, "time": 1.6054761686990426, "step": 9} +{"lr": 1.1703506481239862e-05, "data_time": 0.07794067859649659, "loss": 0.00901974057778716, "time": 1.2705679178237914, "epoch": 10, "memory": 15582, "step": 1639} +{"accuracy/top1": 62.07131576538086, "data_time": 1.1848429247390393, "time": 1.5980328196703002, "step": 10} diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/35624.png_1.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/35624.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..c695b3ce53bf1b09566b8555301bbb9f9225840f Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/35624.png_1.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/35624.png_10.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/35624.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..7be60d34c251b6ec4703a347be89ec6cd9fdf613 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/35624.png_10.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/35624.png_2.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/35624.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..a5cc96e23dc0d8724d803d6b70e59aae8fd6d46f Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/35624.png_2.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/35624.png_3.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/35624.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..dda010ae266f63554d245f26af29bb7809077990 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/35624.png_3.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/35624.png_4.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/35624.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..6971484f600d3f8659c0aad5aa0f88038c32c3f5 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/35624.png_4.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/35624.png_5.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/35624.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..756574c5ea4830c9860120c70397c5fd2f175523 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/35624.png_5.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/35624.png_6.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/35624.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..83823636058a30510822320f7718ce305c697d98 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/35624.png_6.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/35624.png_7.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/35624.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..e568e4a7a994a6cc333d875b9a0006107e6c1254 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/35624.png_7.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/35624.png_8.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/35624.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..857f5131e4e05048bb7478058e1e43c1e261bfa5 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/35624.png_8.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/35624.png_9.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/35624.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..ec23db497343073ddb9ea1a313365b70abf7e324 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/35624.png_9.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/61024.png_1.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/61024.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..1a8ff9048b9bbb5961bf72976ebd514da5864ab2 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/61024.png_1.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/61024.png_10.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/61024.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..776cb1d4cf07433750264527451b1de0c092eb17 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/61024.png_10.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/61024.png_2.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/61024.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..b41e9cfdd58a56ae06529e45cfb605b59cc12da1 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/61024.png_2.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/61024.png_3.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/61024.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..8cab6a9c94bdda635075a14b701fd0c387b24be7 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/61024.png_3.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/61024.png_4.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/61024.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..5c50cdd57c2aba9c1df20af1faca23a7ac438b3b Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/61024.png_4.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/61024.png_5.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/61024.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..deb806bb5a6321f81a89fe5683acb0579ba5c0ef Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/61024.png_5.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/61024.png_6.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/61024.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..55a3c8e96efa2611cb7ab39b501a0f85d2320480 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/61024.png_6.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/61024.png_7.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/61024.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..1357ac2bd1b3e8d4ab4a8de98dd2c04bd6129441 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/61024.png_7.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/61024.png_8.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/61024.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..de765f8fa239b73a1717e9cc13f899cf35a435ad Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/61024.png_8.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/61024.png_9.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/61024.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..9f0c1662349a9b6732ccd53167a213b1acbd5dfb Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/61024.png_9.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/66390.png_1.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/66390.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..e0b9fb77ae786c5651f7f630ddd0d84c4ffae28c Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/66390.png_1.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/66390.png_10.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/66390.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..d894a1249914b5a553979a9eb46b42b0e9260348 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/66390.png_10.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/66390.png_2.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/66390.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..5e8feff23fd9d80b5d6e486f2fdb70ec2ed3923f Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/66390.png_2.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/66390.png_3.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/66390.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..8bc99b39043af092aa45646303be1de9ae12b262 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/66390.png_3.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/66390.png_4.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/66390.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..7ecd3712cf9481c0a83bc3bc1b97821992d41a68 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/66390.png_4.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/66390.png_5.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/66390.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..68c42ab513d266a3930eda85ca427bcf6bd9aaea Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/66390.png_5.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/66390.png_6.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/66390.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..e7abb0d75ab717717ad2024f60df4820397ed258 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/66390.png_6.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/66390.png_7.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/66390.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..aec7960b7d0489e9e1da1526991ed4d8ca7d3800 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/66390.png_7.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/66390.png_8.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/66390.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..b0ad01e13d28aa9b310c7b09224756711a269775 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/66390.png_8.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/66390.png_9.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/66390.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..474d80c6e85d889e8c32da4d1b8b9d81faabd534 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/66390.png_9.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/pixabay_dog_002352.jpg_1.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/pixabay_dog_002352.jpg_1.png new file mode 100644 index 0000000000000000000000000000000000000000..2008bb81082015f8370c79503eea230994d0440b Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/pixabay_dog_002352.jpg_1.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/pixabay_dog_002352.jpg_10.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/pixabay_dog_002352.jpg_10.png new file mode 100644 index 0000000000000000000000000000000000000000..ced2f770e09996d7dc325f5976872e76dd7d8d7b Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/pixabay_dog_002352.jpg_10.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/pixabay_dog_002352.jpg_2.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/pixabay_dog_002352.jpg_2.png new file mode 100644 index 0000000000000000000000000000000000000000..3e925ef298dd7ff88b574b8c11caab671b8a90f7 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/pixabay_dog_002352.jpg_2.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/pixabay_dog_002352.jpg_3.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/pixabay_dog_002352.jpg_3.png new file mode 100644 index 0000000000000000000000000000000000000000..8865465a90d733404053546206dd9515fba06a57 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/pixabay_dog_002352.jpg_3.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/pixabay_dog_002352.jpg_4.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/pixabay_dog_002352.jpg_4.png new file mode 100644 index 0000000000000000000000000000000000000000..a04cc04733e68052d7a5ceea2b1ead16135aeadc Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/pixabay_dog_002352.jpg_4.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/pixabay_dog_002352.jpg_5.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/pixabay_dog_002352.jpg_5.png new file mode 100644 index 0000000000000000000000000000000000000000..518f56079569df8cbc9a8e7ad905b2ed9a772615 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/pixabay_dog_002352.jpg_5.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/pixabay_dog_002352.jpg_6.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/pixabay_dog_002352.jpg_6.png new file mode 100644 index 0000000000000000000000000000000000000000..6d6193c4b84f1a64d781628e4887a8bf91b7cd08 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/pixabay_dog_002352.jpg_6.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/pixabay_dog_002352.jpg_7.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/pixabay_dog_002352.jpg_7.png new file mode 100644 index 0000000000000000000000000000000000000000..cc5b119d838501739a1690b828b91b410e21a55a Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/pixabay_dog_002352.jpg_7.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/pixabay_dog_002352.jpg_8.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/pixabay_dog_002352.jpg_8.png new file mode 100644 index 0000000000000000000000000000000000000000..3348de331d955e736d6a5ab9cab5af069ea87001 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/pixabay_dog_002352.jpg_8.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/pixabay_dog_002352.jpg_9.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/pixabay_dog_002352.jpg_9.png new file mode 100644 index 0000000000000000000000000000000000000000..0821a6310f78d0bd35a82acbc0fccbea52631ec7 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/pixabay_dog_002352.jpg_9.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/seed14088.png_1.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/seed14088.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..f1e68cfa152baf21e7e4ae1114b6be8956699c90 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/seed14088.png_1.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/seed14088.png_10.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/seed14088.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..f570102c7656be873415f985da900a635e81935d Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/seed14088.png_10.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/seed14088.png_2.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/seed14088.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..7132f1cbee547f271b017e45d9f5d548a3a8d53c Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/seed14088.png_2.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/seed14088.png_3.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/seed14088.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..6db6bb857d402ab664fa408e030700ee89ca87df Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/seed14088.png_3.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/seed14088.png_4.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/seed14088.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..0bf2a32f0d20a7c1be5bae9787fd8bf50f85c075 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/seed14088.png_4.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/seed14088.png_5.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/seed14088.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..a8d12ae3096a3804ad3d0ab15edbf8afcd0087c8 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/seed14088.png_5.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/seed14088.png_6.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/seed14088.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..db6bc73b00778284198139f2f2a1d29ee070025d Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/seed14088.png_6.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/seed14088.png_7.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/seed14088.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..b80fed790139f739257829767d04b601f3f4f31c Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/seed14088.png_7.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/seed14088.png_8.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/seed14088.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..667f58e2505584f85a6d59ce50b78b25225f4e21 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/seed14088.png_8.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/seed14088.png_9.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/seed14088.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..1ec854a5ae37116aec45dcc078a28cfba93cf6fd Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/seed14088.png_9.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/seed17282.png_1.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/seed17282.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..fceeb73cf7c97fdbf3b1de1d1ae9beac149ae392 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/seed17282.png_1.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/seed17282.png_10.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/seed17282.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..c2daa8e595601443f73a26e9328d2b4f0be93bcf Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/seed17282.png_10.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/seed17282.png_2.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/seed17282.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..abfd35bb0136b434bf9e0ea84afaa9c49a3e2967 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/seed17282.png_2.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/seed17282.png_3.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/seed17282.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..c3f925ee55bfbdca9d3af689bba9387c3beb502d Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/seed17282.png_3.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/seed17282.png_4.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/seed17282.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..54509651d8dd14a61717e93d2333dd928c85e863 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/seed17282.png_4.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/seed17282.png_5.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/seed17282.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..74c1a216cec099453db581fb80d331178de58728 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/seed17282.png_5.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/seed17282.png_6.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/seed17282.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..cc5a520822e98dd2e5792fe5c7d5fe45e5fd49c7 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/seed17282.png_6.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/seed17282.png_7.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/seed17282.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..b2939fb04d2feb057e5000297d1a7809538bf2cb Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/seed17282.png_7.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/seed17282.png_8.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/seed17282.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..fae9987a02e2d388fa10a5b0af3f5e4dd21ab110 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/seed17282.png_8.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/seed17282.png_9.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/seed17282.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..72d9448c648e794fd0c98f245e3f1f4971d09076 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/seed17282.png_9.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/seed23602.png_1.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/seed23602.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..0b444ba44b070f95138ffc13bef645797026ee42 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/seed23602.png_1.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/seed23602.png_10.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/seed23602.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..0472d16e2df91de1a18695b4303e99bef8f1068a Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/seed23602.png_10.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/seed23602.png_2.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/seed23602.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..fd8bbfa4be45c43f70fe762b8ede00e58e2de302 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/seed23602.png_2.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/seed23602.png_3.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/seed23602.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..a6fa9dce991dd9af166e829d417186d4585a51c8 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/seed23602.png_3.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/seed23602.png_4.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/seed23602.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..7c086a2df0624ed800aaf33f4b4cafb8eca95ccd Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/seed23602.png_4.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/seed23602.png_5.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/seed23602.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..36c76dff0abb602f631ec296532ee7296b448b03 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/seed23602.png_5.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/seed23602.png_6.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/seed23602.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..b38a48920825b23d5592f4e6d624ef34a5bff5f5 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/seed23602.png_6.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/seed23602.png_7.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/seed23602.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..9d4ae51bafa08396e46a23d254c65f6d7323e418 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/seed23602.png_7.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/seed23602.png_8.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/seed23602.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..73624cb8a21038baac355a2346ef7b2e473240b0 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/seed23602.png_8.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/seed23602.png_9.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/seed23602.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..b3881270c25c5802c784d29a269ed2427aec3033 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/seed23602.png_9.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/seed24366.png_1.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/seed24366.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..d5a732296661130b39374c20d66c402de3bfb88a Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/seed24366.png_1.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/seed24366.png_10.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/seed24366.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..f1e8b149513495dc940e9134c944a1f7fbe74417 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/seed24366.png_10.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/seed24366.png_2.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/seed24366.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..d0591751cc38950e8f6df9a4e7d3cd20b48c7f8a Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/seed24366.png_2.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/seed24366.png_3.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/seed24366.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..72aed0133d3167ce87381890f70c5831ec475d27 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/seed24366.png_3.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/seed24366.png_4.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/seed24366.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..d667bf71a7d2f394ff896b068b64e3ddaca9c93d Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/seed24366.png_4.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/seed24366.png_5.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/seed24366.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..62447d44ce5ae8d32d9c03ec44afb8d2cb016a55 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/seed24366.png_5.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/seed24366.png_6.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/seed24366.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..634f4d8c842c8794529bfcb84b64ff4cd2f19009 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/seed24366.png_6.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/seed24366.png_7.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/seed24366.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..ef462b5e5720e3d37a1c3248999ba19ca6d4b7e5 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/seed24366.png_7.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/seed24366.png_8.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/seed24366.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..50f3e1ded4b4f9a72c3a4b54c7922c7caec27cce Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/seed24366.png_8.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/seed24366.png_9.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/seed24366.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..fb5c47a94aace0a341eb9da761d5ce0a0172e16a Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/seed24366.png_9.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/seed44086.png_1.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/seed44086.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..43565968d7a76751964c2b8f5e1d1ff64618f554 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/seed44086.png_1.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/seed44086.png_10.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/seed44086.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..e8df9f9ff910d3e7230cf9bc97f5ca02ade6acef Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/seed44086.png_10.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/seed44086.png_2.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/seed44086.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..4c48b47cb22d90b03aeadcbbea6d3f5dd8a82edc Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/seed44086.png_2.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/seed44086.png_3.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/seed44086.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..2d7be349039f67eabc6f640126f3dd5d2c573ad1 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/seed44086.png_3.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/seed44086.png_4.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/seed44086.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..92a72bf2ce504c13355d8fb93f9a4d7bf8bd5385 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/seed44086.png_4.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/seed44086.png_5.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/seed44086.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..6fecae25f21f471f4c0915bffdb8d08e6839b7ae Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/seed44086.png_5.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/seed44086.png_6.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/seed44086.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..3bb254df2148633b55d297c43cb8f2c99547d900 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/seed44086.png_6.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/seed44086.png_7.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/seed44086.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..1eff7102ab44ce4f2452e26f2c678fcd8a79f216 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/seed44086.png_7.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/seed44086.png_8.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/seed44086.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..e47979d967a6dd7ff2e317bbf626e4019c02b56d Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/seed44086.png_8.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/seed44086.png_9.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/seed44086.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..38eef5ecb4b9ee2a4fcaf3ca9b617ad1d569e5e8 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/20230605_010733/vis_data/vis_image/seed44086.png_9.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1.py b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1.py new file mode 100644 index 0000000000000000000000000000000000000000..e452889e6187289e7e8fade71cf0f2603ccc9eef --- /dev/null +++ b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1.py @@ -0,0 +1,247 @@ +optim_wrapper = dict( + optimizer=dict( + type='AdamW', + lr=0.0004, + weight_decay=0.05, + eps=1e-08, + betas=(0.9, 0.999), + _scope_='mmpretrain'), + paramwise_cfg=dict( + norm_decay_mult=0.0, + bias_decay_mult=0.0, + flat_decay_mult=0.0, + custom_keys=dict({ + '.absolute_pos_embed': dict(decay_mult=0.0), + '.relative_position_bias_table': dict(decay_mult=0.0) + })), + type='AmpOptimWrapper', + dtype='bfloat16', + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=False, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=4096) +model = dict( + type='ImageClassifier', + backbone=dict( + frozen_stages=2, + type='ConvNeXt', + arch='small', + drop_path_rate=0.4, + init_cfg=dict( + type='Pretrained', + checkpoint= + 'https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth', + prefix='backbone')), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=768, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + init_cfg=None), + init_cfg=dict( + type='TruncNormal', layer=['Conv2d', 'Linear'], std=0.02, bias=0.0), + train_cfg=None) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3fake8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3real8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3fake8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3real8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3fake8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3real8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1' diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/epoch_1.pth b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/epoch_1.pth new file mode 100644 index 0000000000000000000000000000000000000000..f320d049c2609f6392fd2bb9a14d2bd304069001 --- /dev/null +++ b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/epoch_1.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:2ca65c6867683f2225158a5f41a341cc4202e4214c23d58db47effa9150adc88 +size 781981274 diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/epoch_10.pth b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/epoch_10.pth new file mode 100644 index 0000000000000000000000000000000000000000..21ffdb8cdae1001e82d1d2ad0c739848167633f7 --- /dev/null +++ b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/epoch_10.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:883d8c3a498b7076454d48ac5927eb534c5e208d7ba6f9396eecfe38557bb309 +size 782147162 diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/epoch_2.pth b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/epoch_2.pth new file mode 100644 index 0000000000000000000000000000000000000000..4379d2c8ce5a61c63538ec617f917fee586504f1 --- /dev/null +++ b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/epoch_2.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:48799cfffe83550011e56a426c9b1ad19947e7b01688954720575957b985373f +size 782000794 diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/epoch_3.pth b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/epoch_3.pth new file mode 100644 index 0000000000000000000000000000000000000000..5f5edcd23d85838c071c8b677c0094139f20a462 --- /dev/null +++ b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/epoch_3.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c7b0d4e007ccc021473393c23a448c3bf4bebcf22099be72b758fccd64098c87 +size 782019034 diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/epoch_4.pth b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/epoch_4.pth new file mode 100644 index 0000000000000000000000000000000000000000..3c7f0af35f0f55f0d8fae7350c0e0d3dded55c6f --- /dev/null +++ b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/epoch_4.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:32dd8f155e8d44801382c5d2dc1c2b8540fe28f9cc4415cb7f63aac211806d33 +size 782037402 diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/epoch_5.pth b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/epoch_5.pth new file mode 100644 index 0000000000000000000000000000000000000000..60637c04fe4de7c19d56082ddf97dc7867e46da9 --- /dev/null +++ b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/epoch_5.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:6e02199d02afb7c1b4fc2e1c88803765ccd9dae9673d127a82b82e7da900282b +size 782055706 diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/epoch_6.pth b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/epoch_6.pth new file mode 100644 index 0000000000000000000000000000000000000000..78ac22b320918109014fe7eedb263b555235826b --- /dev/null +++ b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/epoch_6.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:69508cbf33eddc67056dc6fdffe72d59df06edefee7fbfd6c81be079766f15bd +size 782074010 diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/epoch_7.pth b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/epoch_7.pth new file mode 100644 index 0000000000000000000000000000000000000000..52ad6c0a8091ea25a2aa6e76b347ba50f8e5b7b1 --- /dev/null +++ b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/epoch_7.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:98345b5e0c0f3f737e955e70ad47e3b7f68d9893c0f912483054e3433a33b681 +size 782092250 diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/epoch_8.pth b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/epoch_8.pth new file mode 100644 index 0000000000000000000000000000000000000000..429c4242f05d633f0ef7be1ea16d12d06593e0f3 --- /dev/null +++ b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/epoch_8.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:9ad05c8e916b4753ae9a2fa2b48ed4c77c435fb62659bc1e2b06c13739998023 +size 782110554 diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/epoch_9.pth b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/epoch_9.pth new file mode 100644 index 0000000000000000000000000000000000000000..90beb3f4b2371c01ad65ca7f60bb1bbe0c0fefae --- /dev/null +++ b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/epoch_9.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ffc8282681b28b8dd42bd749d3b8725c79b41c0e8e1b10cb586e86cbe0303dc0 +size 782128794 diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/last_checkpoint b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/last_checkpoint new file mode 100644 index 0000000000000000000000000000000000000000..dead29107971d988d0217131c6740bf4fb6e5f0d --- /dev/null +++ b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/last_checkpoint @@ -0,0 +1 @@ +/mnt/petrelfs/luzeyu/workspace/fakebench/mmpretrain/workdir/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_1e-1/epoch_10.pth \ No newline at end of file diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/20230605_010924.log b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/20230605_010924.log new file mode 100644 index 0000000000000000000000000000000000000000..14eb269e7fcd29ee9c1cb4d6c91c64f7757d3375 --- /dev/null +++ b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/20230605_010924.log @@ -0,0 +1,1652 @@ +2023/06/05 01:09:29 - mmengine - INFO - +------------------------------------------------------------ +System environment: + sys.platform: linux + Python: 3.10.9 (main, Mar 8 2023, 10:47:38) [GCC 11.2.0] + CUDA available: True + numpy_random_seed: 752651729 + GPU 0,1,2,3: NVIDIA A100-SXM4-80GB + CUDA_HOME: /mnt/petrelfs/share/cuda-11.6 + NVCC: Cuda compilation tools, release 11.6, V11.6.124 + GCC: gcc (GCC) 7.5.0 + PyTorch: 1.13.1 + PyTorch compiling details: PyTorch built with: + - GCC 9.3 + - C++ Version: 201402 + - Intel(R) oneAPI Math Kernel Library Version 2021.4-Product Build 20210904 for Intel(R) 64 architecture applications + - Intel(R) MKL-DNN v2.6.0 (Git Hash 52b5f107dd9cf10910aaa19cb47f3abf9b349815) + - OpenMP 201511 (a.k.a. OpenMP 4.5) + - LAPACK is enabled (usually provided by MKL) + - NNPACK is enabled + - CPU capability usage: AVX2 + - CUDA Runtime 11.6 + - NVCC architecture flags: -gencode;arch=compute_37,code=sm_37;-gencode;arch=compute_50,code=sm_50;-gencode;arch=compute_60,code=sm_60;-gencode;arch=compute_61,code=sm_61;-gencode;arch=compute_70,code=sm_70;-gencode;arch=compute_75,code=sm_75;-gencode;arch=compute_80,code=sm_80;-gencode;arch=compute_86,code=sm_86;-gencode;arch=compute_37,code=compute_37 + - CuDNN 8.3.2 (built against CUDA 11.5) + - Magma 2.6.1 + - Build settings: BLAS_INFO=mkl, BUILD_TYPE=Release, CUDA_VERSION=11.6, CUDNN_VERSION=8.3.2, CXX_COMPILER=/opt/rh/devtoolset-9/root/usr/bin/c++, CXX_FLAGS= -fabi-version=11 -Wno-deprecated -fvisibility-inlines-hidden -DUSE_PTHREADPOOL -fopenmp -DNDEBUG -DUSE_KINETO -DUSE_FBGEMM -DUSE_QNNPACK -DUSE_PYTORCH_QNNPACK -DUSE_XNNPACK -DSYMBOLICATE_MOBILE_DEBUG_HANDLE -DEDGE_PROFILER_USE_KINETO -O2 -fPIC -Wno-narrowing -Wall -Wextra -Werror=return-type -Werror=non-virtual-dtor -Wno-missing-field-initializers -Wno-type-limits -Wno-array-bounds -Wno-unknown-pragmas -Wunused-local-typedefs -Wno-unused-parameter -Wno-unused-function -Wno-unused-result -Wno-strict-overflow -Wno-strict-aliasing -Wno-error=deprecated-declarations -Wno-stringop-overflow -Wno-psabi -Wno-error=pedantic -Wno-error=redundant-decls -Wno-error=old-style-cast -fdiagnostics-color=always -faligned-new -Wno-unused-but-set-variable -Wno-maybe-uninitialized -fno-math-errno -fno-trapping-math -Werror=format -Werror=cast-function-type -Wno-stringop-overflow, LAPACK_INFO=mkl, PERF_WITH_AVX=1, PERF_WITH_AVX2=1, PERF_WITH_AVX512=1, TORCH_VERSION=1.13.1, USE_CUDA=ON, USE_CUDNN=ON, USE_EXCEPTION_PTR=1, USE_GFLAGS=OFF, USE_GLOG=OFF, USE_MKL=ON, USE_MKLDNN=ON, USE_MPI=OFF, USE_NCCL=ON, USE_NNPACK=ON, USE_OPENMP=ON, USE_ROCM=OFF, + + TorchVision: 0.14.1 + OpenCV: 4.7.0 + MMEngine: 0.7.3 + +Runtime environment: + cudnn_benchmark: True + mp_cfg: {'mp_start_method': 'fork', 'opencv_num_threads': 0} + dist_cfg: {'backend': 'nccl'} + seed: None + deterministic: False + Distributed launcher: slurm + Distributed training: True + GPU number: 4 +------------------------------------------------------------ + +2023/06/05 01:09:33 - mmengine - INFO - Config: +optim_wrapper = dict( + optimizer=dict( + type='AdamW', + lr=0.0004, + weight_decay=0.05, + eps=1e-08, + betas=(0.9, 0.999), + _scope_='mmpretrain'), + paramwise_cfg=dict( + norm_decay_mult=0.0, + bias_decay_mult=0.0, + flat_decay_mult=0.0, + custom_keys=dict({ + '.absolute_pos_embed': dict(decay_mult=0.0), + '.relative_position_bias_table': dict(decay_mult=0.0) + })), + type='AmpOptimWrapper', + dtype='bfloat16', + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=False, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=4096) +model = dict( + type='ImageClassifier', + backbone=dict( + frozen_stages=2, + type='ConvNeXt', + arch='small', + drop_path_rate=0.4, + init_cfg=dict( + type='Pretrained', + checkpoint= + 'https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth', + prefix='backbone')), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=768, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + init_cfg=None), + init_cfg=dict( + type='TruncNormal', layer=['Conv2d', 'Linear'], std=0.02, bias=0.0), + train_cfg=None) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3fake8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3real8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3fake8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3real8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3fake8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3real8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1' + +2023/06/05 01:09:44 - mmengine - INFO - Hooks will be executed in the following order: +before_run: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_load_checkpoint: +(ABOVE_NORMAL) EMAHook + -------------------- +before_train: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_train_epoch: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(NORMAL ) DistSamplerSeedHook + -------------------- +before_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook + -------------------- +after_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +after_train_epoch: +(NORMAL ) IterTimerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_val_epoch: +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook + -------------------- +before_val_iter: +(NORMAL ) IterTimerHook + -------------------- +after_val_iter: +(NORMAL ) IterTimerHook +(NORMAL ) VisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_val_epoch: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_save_checkpoint: +(ABOVE_NORMAL) EMAHook + -------------------- +after_train: +(VERY_LOW ) CheckpointHook + -------------------- +before_test_epoch: +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook + -------------------- +before_test_iter: +(NORMAL ) IterTimerHook + -------------------- +after_test_iter: +(NORMAL ) IterTimerHook +(NORMAL ) VisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_test_epoch: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_run: +(BELOW_NORMAL) LoggerHook + -------------------- +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.2.0.weight:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.2.0.bias:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.2.1.bias:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.3.0.weight:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.3.0.bias:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.3.1.bias:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.2.0.gamma:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.2.0.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.2.0.norm.weight:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.2.0.norm.bias:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.2.0.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.2.0.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.2.1.gamma:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.2.1.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.2.1.norm.weight:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.2.1.norm.bias:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.2.1.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.2.1.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.2.2.gamma:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.2.2.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.2.2.norm.weight:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.2.2.norm.bias:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.2.2.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.2.2.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.2.3.gamma:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.2.3.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.2.3.norm.weight:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.2.3.norm.bias:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.2.3.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.2.3.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.2.4.gamma:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.2.4.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.2.4.norm.weight:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.2.4.norm.bias:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.2.4.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.2.4.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.2.5.gamma:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.2.5.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.2.5.norm.weight:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.2.5.norm.bias:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.2.5.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.2.5.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.2.6.gamma:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.2.6.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.2.6.norm.weight:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.2.6.norm.bias:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.2.6.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.2.6.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.2.7.gamma:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.2.7.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.2.7.norm.weight:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.2.7.norm.bias:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.2.7.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.2.7.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.2.8.gamma:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.2.8.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.2.8.norm.weight:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.2.8.norm.bias:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.2.8.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.2.8.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.2.9.gamma:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.2.9.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.2.9.norm.weight:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.2.9.norm.bias:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.2.9.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.2.9.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.2.10.gamma:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.2.10.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.2.10.norm.weight:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.2.10.norm.bias:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.2.10.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.2.10.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.2.11.gamma:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.2.11.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.2.11.norm.weight:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.2.11.norm.bias:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.2.11.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.2.11.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.2.12.gamma:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.2.12.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.2.12.norm.weight:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.2.12.norm.bias:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.2.12.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.2.12.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.2.13.gamma:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.2.13.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.2.13.norm.weight:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.2.13.norm.bias:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.2.13.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.2.13.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.2.14.gamma:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.2.14.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.2.14.norm.weight:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.2.14.norm.bias:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.2.14.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.2.14.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.2.15.gamma:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.2.15.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.2.15.norm.weight:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.2.15.norm.bias:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.2.15.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.2.15.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.2.16.gamma:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.2.16.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.2.16.norm.weight:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.2.16.norm.bias:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.2.16.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.2.16.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.2.17.gamma:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.2.17.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.2.17.norm.weight:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.2.17.norm.bias:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.2.17.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.2.17.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.2.18.gamma:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.2.18.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.2.18.norm.weight:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.2.18.norm.bias:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.2.18.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.2.18.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.2.19.gamma:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.2.19.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.2.19.norm.weight:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.2.19.norm.bias:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.2.19.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.2.19.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.2.20.gamma:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.2.20.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.2.20.norm.weight:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.2.20.norm.bias:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.2.20.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.2.20.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.2.21.gamma:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.2.21.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.2.21.norm.weight:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.2.21.norm.bias:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.2.21.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.2.21.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.2.22.gamma:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.2.22.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.2.22.norm.weight:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.2.22.norm.bias:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.2.22.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.2.22.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.2.23.gamma:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.2.23.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.2.23.norm.weight:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.2.23.norm.bias:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.2.23.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.2.23.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.2.24.gamma:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.2.24.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.2.24.norm.weight:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.2.24.norm.bias:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.2.24.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.2.24.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.2.25.gamma:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.2.25.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.2.25.norm.weight:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.2.25.norm.bias:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.2.25.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.2.25.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.2.26.gamma:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.2.26.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.2.26.norm.weight:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.2.26.norm.bias:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.2.26.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.2.26.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.3.0.gamma:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.3.0.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.3.0.norm.weight:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.3.0.norm.bias:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.3.0.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.3.0.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.3.1.gamma:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.3.1.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.3.1.norm.weight:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.3.1.norm.bias:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.3.1.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.3.1.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.3.2.gamma:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.3.2.depthwise_conv.bias:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.3.2.norm.weight:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.3.2.norm.bias:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.3.2.pointwise_conv1.bias:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.stages.3.2.pointwise_conv2.bias:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.norm3.weight:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- backbone.norm3.bias:weight_decay=0.0 +2023/06/05 01:09:49 - mmengine - INFO - paramwise_options -- head.fc.bias:weight_decay=0.0 +2023/06/05 01:09:54 - mmengine - INFO - load backbone in model from: https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth +Name of parameter - Initialization information + +backbone.downsample_layers.0.0.weight - torch.Size([96, 3, 4, 4]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.0.0.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.0.1.weight - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.0.1.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.1.0.weight - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.1.0.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.1.1.weight - torch.Size([192, 96, 2, 2]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.1.1.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.2.0.weight - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.2.0.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.2.1.weight - torch.Size([384, 192, 2, 2]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.2.1.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.3.0.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.3.0.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.3.1.weight - torch.Size([768, 384, 2, 2]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.downsample_layers.3.1.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.0.gamma - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.0.depthwise_conv.weight - torch.Size([96, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.0.depthwise_conv.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.0.norm.weight - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.0.norm.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.0.pointwise_conv1.weight - torch.Size([384, 96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.0.pointwise_conv1.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.0.pointwise_conv2.weight - torch.Size([96, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.0.pointwise_conv2.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.1.gamma - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.1.depthwise_conv.weight - torch.Size([96, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.1.depthwise_conv.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.1.norm.weight - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.1.norm.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.1.pointwise_conv1.weight - torch.Size([384, 96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.1.pointwise_conv1.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.1.pointwise_conv2.weight - torch.Size([96, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.1.pointwise_conv2.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.2.gamma - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.2.depthwise_conv.weight - torch.Size([96, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.2.depthwise_conv.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.2.norm.weight - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.2.norm.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.2.pointwise_conv1.weight - torch.Size([384, 96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.2.pointwise_conv1.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.2.pointwise_conv2.weight - torch.Size([96, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.0.2.pointwise_conv2.bias - torch.Size([96]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.0.gamma - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.0.depthwise_conv.weight - torch.Size([192, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.0.depthwise_conv.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.0.norm.weight - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.0.norm.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.0.pointwise_conv1.weight - torch.Size([768, 192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.0.pointwise_conv1.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.0.pointwise_conv2.weight - torch.Size([192, 768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.0.pointwise_conv2.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.1.gamma - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.1.depthwise_conv.weight - torch.Size([192, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.1.depthwise_conv.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.1.norm.weight - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.1.norm.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.1.pointwise_conv1.weight - torch.Size([768, 192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.1.pointwise_conv1.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.1.pointwise_conv2.weight - torch.Size([192, 768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.1.pointwise_conv2.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.2.gamma - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.2.depthwise_conv.weight - torch.Size([192, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.2.depthwise_conv.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.2.norm.weight - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.2.norm.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.2.pointwise_conv1.weight - torch.Size([768, 192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.2.pointwise_conv1.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.2.pointwise_conv2.weight - torch.Size([192, 768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.1.2.pointwise_conv2.bias - torch.Size([192]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.0.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.0.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.0.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.0.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.0.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.0.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.0.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.0.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.0.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.1.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.1.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.1.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.1.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.1.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.1.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.1.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.1.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.1.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.2.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.2.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.2.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.2.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.2.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.2.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.2.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.2.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.2.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.3.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.3.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.3.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.3.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.3.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.3.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.3.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.3.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.3.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.4.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.4.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.4.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.4.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.4.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.4.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.4.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.4.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.4.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.5.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.5.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.5.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.5.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.5.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.5.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.5.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.5.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.5.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.6.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.6.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.6.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.6.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.6.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.6.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.6.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.6.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.6.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.7.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.7.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.7.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.7.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.7.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.7.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.7.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.7.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.7.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.8.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.8.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.8.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.8.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.8.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.8.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.8.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.8.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.8.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.9.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.9.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.9.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.9.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.9.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.9.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.9.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.9.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.9.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.10.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.10.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.10.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.10.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.10.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.10.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.10.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.10.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.10.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.11.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.11.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.11.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.11.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.11.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.11.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.11.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.11.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.11.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.12.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.12.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.12.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.12.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.12.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.12.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.12.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.12.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.12.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.13.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.13.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.13.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.13.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.13.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.13.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.13.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.13.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.13.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.14.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.14.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.14.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.14.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.14.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.14.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.14.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.14.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.14.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.15.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.15.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.15.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.15.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.15.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.15.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.15.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.15.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.15.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.16.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.16.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.16.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.16.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.16.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.16.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.16.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.16.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.16.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.17.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.17.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.17.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.17.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.17.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.17.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.17.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.17.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.17.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.18.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.18.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.18.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.18.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.18.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.18.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.18.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.18.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.18.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.19.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.19.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.19.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.19.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.19.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.19.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.19.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.19.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.19.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.20.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.20.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.20.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.20.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.20.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.20.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.20.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.20.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.20.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.21.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.21.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.21.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.21.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.21.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.21.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.21.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.21.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.21.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.22.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.22.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.22.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.22.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.22.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.22.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.22.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.22.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.22.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.23.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.23.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.23.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.23.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.23.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.23.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.23.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.23.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.23.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.24.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.24.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.24.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.24.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.24.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.24.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.24.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.24.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.24.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.25.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.25.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.25.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.25.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.25.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.25.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.25.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.25.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.25.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.26.gamma - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.26.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.26.depthwise_conv.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.26.norm.weight - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.26.norm.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.26.pointwise_conv1.weight - torch.Size([1536, 384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.26.pointwise_conv1.bias - torch.Size([1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.26.pointwise_conv2.weight - torch.Size([384, 1536]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.2.26.pointwise_conv2.bias - torch.Size([384]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.0.gamma - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.0.depthwise_conv.weight - torch.Size([768, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.0.depthwise_conv.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.0.norm.weight - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.0.norm.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.0.pointwise_conv1.weight - torch.Size([3072, 768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.0.pointwise_conv1.bias - torch.Size([3072]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.0.pointwise_conv2.weight - torch.Size([768, 3072]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.0.pointwise_conv2.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.1.gamma - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.1.depthwise_conv.weight - torch.Size([768, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.1.depthwise_conv.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.1.norm.weight - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.1.norm.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.1.pointwise_conv1.weight - torch.Size([3072, 768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.1.pointwise_conv1.bias - torch.Size([3072]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.1.pointwise_conv2.weight - torch.Size([768, 3072]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.1.pointwise_conv2.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.2.gamma - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.2.depthwise_conv.weight - torch.Size([768, 1, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.2.depthwise_conv.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.2.norm.weight - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.2.norm.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.2.pointwise_conv1.weight - torch.Size([3072, 768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.2.pointwise_conv1.bias - torch.Size([3072]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.2.pointwise_conv2.weight - torch.Size([768, 3072]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.stages.3.2.pointwise_conv2.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.norm3.weight - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +backbone.norm3.bias - torch.Size([768]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth + +head.fc.weight - torch.Size([2, 768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +head.fc.bias - torch.Size([2]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 +2023/06/05 01:09:55 - mmengine - WARNING - "FileClient" will be deprecated in future. Please use io functions in https://mmengine.readthedocs.io/en/latest/api/fileio.html#file-io +2023/06/05 01:09:55 - mmengine - WARNING - "HardDiskBackend" is the alias of "LocalBackend" and the former will be deprecated in future. +2023/06/05 01:09:55 - mmengine - INFO - Checkpoints will be saved to /mnt/petrelfs/luzeyu/workspace/fakebench/mmpretrain/workdir/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1. +2023/06/05 01:12:06 - mmengine - INFO - Epoch(train) [1][100/171] lr: 3.9678e-04 eta: 0:35:23 time: 1.2840 data_time: 0.0020 memory: 15587 loss: 0.0609 +2023/06/05 01:13:37 - mmengine - INFO - Exp name: convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1_20230605_010924 +2023/06/05 01:13:37 - mmengine - INFO - Saving checkpoint at 1 epochs +2023/06/05 01:15:48 - mmengine - INFO - Epoch(val) [1][100/171] eta: 0:01:28 time: 0.4135 data_time: 0.0070 memory: 15587 +2023/06/05 01:19:02 - mmengine - INFO - Epoch(val) [1][171/171] accuracy/top1: 49.9691 single-label/precision_classwise: [49.969093322753906, 0.0] single-label/recall_classwise: [100.0, 0.0] single-label/f1-score_classwise: [66.63919067382812, 0.0] data_time: 1.1121 time: 1.5349 +2023/06/05 01:21:12 - mmengine - INFO - Epoch(train) [2][100/171] lr: 3.7650e-04 eta: 0:31:10 time: 1.3321 data_time: 0.0017 memory: 15582 loss: 0.0376 +2023/06/05 01:22:43 - mmengine - INFO - Exp name: convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1_20230605_010924 +2023/06/05 01:22:43 - mmengine - INFO - Saving checkpoint at 2 epochs +2023/06/05 01:24:52 - mmengine - INFO - Epoch(val) [2][100/171] eta: 0:01:27 time: 0.4149 data_time: 0.0070 memory: 15582 +2023/06/05 01:28:09 - mmengine - INFO - Epoch(val) [2][171/171] accuracy/top1: 49.9691 single-label/precision_classwise: [49.969093322753906, 0.0] single-label/recall_classwise: [100.0, 0.0] single-label/f1-score_classwise: [66.63919067382812, 0.0] data_time: 1.1107 time: 1.5214 +2023/06/05 01:30:19 - mmengine - INFO - Epoch(train) [3][100/171] lr: 3.3942e-04 eta: 0:27:25 time: 1.2439 data_time: 0.0020 memory: 15582 loss: 0.0287 +2023/06/05 01:31:46 - mmengine - INFO - Exp name: convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1_20230605_010924 +2023/06/05 01:31:46 - mmengine - INFO - Saving checkpoint at 3 epochs +2023/06/05 01:33:55 - mmengine - INFO - Epoch(val) [3][100/171] eta: 0:01:27 time: 0.4142 data_time: 0.0072 memory: 15582 +2023/06/05 01:37:09 - mmengine - INFO - Epoch(val) [3][171/171] accuracy/top1: 49.9697 single-label/precision_classwise: [49.96937942504883, 100.0] single-label/recall_classwise: [100.0, 0.0011439815862104297] single-label/f1-score_classwise: [66.63944244384766, 0.00228793709538877] data_time: 1.1135 time: 1.5247 +2023/06/05 01:39:23 - mmengine - INFO - Epoch(train) [4][100/171] lr: 2.8919e-04 eta: 0:23:42 time: 1.2642 data_time: 0.0280 memory: 15582 loss: 0.0196 +2023/06/05 01:40:54 - mmengine - INFO - Exp name: convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1_20230605_010924 +2023/06/05 01:40:54 - mmengine - INFO - Saving checkpoint at 4 epochs +2023/06/05 01:43:05 - mmengine - INFO - Epoch(val) [4][100/171] eta: 0:01:28 time: 0.4137 data_time: 0.0070 memory: 15582 +2023/06/05 01:46:21 - mmengine - INFO - Epoch(val) [4][171/171] accuracy/top1: 49.9737 single-label/precision_classwise: [49.97138214111328, 100.0] single-label/recall_classwise: [100.0, 0.009151852689683437] single-label/f1-score_classwise: [66.64122772216797, 0.018302028998732567] data_time: 1.1249 time: 1.5353 +2023/06/05 01:48:30 - mmengine - INFO - Epoch(train) [5][100/171] lr: 2.3072e-04 eta: 0:19:59 time: 1.1882 data_time: 0.0024 memory: 15582 loss: 0.0105 +2023/06/05 01:50:01 - mmengine - INFO - Exp name: convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1_20230605_010924 +2023/06/05 01:50:01 - mmengine - INFO - Saving checkpoint at 5 epochs +2023/06/05 01:52:11 - mmengine - INFO - Epoch(val) [5][100/171] eta: 0:01:27 time: 0.4152 data_time: 0.0079 memory: 15582 +2023/06/05 01:55:28 - mmengine - INFO - Epoch(val) [5][171/171] accuracy/top1: 49.9857 single-label/precision_classwise: [49.97739028930664, 100.0] single-label/recall_classwise: [100.0, 0.03317546471953392] single-label/f1-score_classwise: [66.64656829833984, 0.06632892787456512] data_time: 1.1309 time: 1.5400 +2023/06/05 01:57:43 - mmengine - INFO - Epoch(train) [6][100/171] lr: 1.6973e-04 eta: 0:16:20 time: 1.2903 data_time: 0.0017 memory: 15582 loss: 0.0130 +2023/06/05 01:58:43 - mmengine - INFO - Exp name: convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1_20230605_010924 +2023/06/05 01:59:15 - mmengine - INFO - Exp name: convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1_20230605_010924 +2023/06/05 01:59:15 - mmengine - INFO - Saving checkpoint at 6 epochs +2023/06/05 02:01:25 - mmengine - INFO - Epoch(val) [6][100/171] eta: 0:01:27 time: 0.4138 data_time: 0.0065 memory: 15582 +2023/06/05 02:04:42 - mmengine - INFO - Epoch(val) [6][171/171] accuracy/top1: 50.0177 single-label/precision_classwise: [49.993412017822266, 96.70330047607422] single-label/recall_classwise: [99.99656677246094, 0.10067036747932434] single-label/f1-score_classwise: [66.66004943847656, 0.20113135874271393] data_time: 1.1584 time: 1.5693 +2023/06/05 02:07:00 - mmengine - INFO - Epoch(train) [7][100/171] lr: 1.1219e-04 eta: 0:12:42 time: 1.3285 data_time: 0.0017 memory: 15582 loss: 0.0124 +2023/06/05 02:08:31 - mmengine - INFO - Exp name: convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1_20230605_010924 +2023/06/05 02:08:31 - mmengine - INFO - Saving checkpoint at 7 epochs +2023/06/05 02:10:41 - mmengine - INFO - Epoch(val) [7][100/171] eta: 0:01:27 time: 0.4140 data_time: 0.0069 memory: 15582 +2023/06/05 02:13:56 - mmengine - INFO - Epoch(val) [7][171/171] accuracy/top1: 50.0767 single-label/precision_classwise: [50.022918701171875, 97.47474670410156] single-label/recall_classwise: [99.99427032470703, 0.22078843414783478] single-label/f1-score_classwise: [66.6857681274414, 0.4405789375305176] data_time: 1.1618 time: 1.5730 +2023/06/05 02:16:07 - mmengine - INFO - Epoch(train) [8][100/171] lr: 6.3737e-05 eta: 0:08:59 time: 1.2254 data_time: 0.0020 memory: 15582 loss: 0.0079 +2023/06/05 02:17:34 - mmengine - INFO - Exp name: convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1_20230605_010924 +2023/06/05 02:17:34 - mmengine - INFO - Saving checkpoint at 8 epochs +2023/06/05 02:19:44 - mmengine - INFO - Epoch(val) [8][100/171] eta: 0:01:27 time: 0.4135 data_time: 0.0068 memory: 15582 +2023/06/05 02:22:57 - mmengine - INFO - Epoch(val) [8][171/171] accuracy/top1: 50.1688 single-label/precision_classwise: [50.069122314453125, 95.09043884277344] single-label/recall_classwise: [99.97824096679688, 0.4209851920604706] single-label/f1-score_classwise: [66.72323608398438, 0.838259220123291] data_time: 1.1212 time: 1.5301 +2023/06/05 02:25:11 - mmengine - INFO - Epoch(train) [9][100/171] lr: 2.9112e-05 eta: 0:05:15 time: 1.2187 data_time: 0.0017 memory: 15582 loss: 0.0089 +2023/06/05 02:26:40 - mmengine - INFO - Exp name: convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1_20230605_010924 +2023/06/05 02:26:40 - mmengine - INFO - Saving checkpoint at 9 epochs +2023/06/05 02:28:48 - mmengine - INFO - Epoch(val) [9][100/171] eta: 0:01:26 time: 0.4139 data_time: 0.0070 memory: 15582 +2023/06/05 02:32:06 - mmengine - INFO - Epoch(val) [9][171/171] accuracy/top1: 50.3142 single-label/precision_classwise: [50.142215728759766, 93.75907135009766] single-label/recall_classwise: [99.95075225830078, 0.7390120625495911] single-label/f1-score_classwise: [66.78196716308594, 1.4664653539657593] data_time: 1.1056 time: 1.5157 +2023/06/05 02:34:21 - mmengine - INFO - Epoch(train) [10][100/171] lr: 1.1704e-05 eta: 0:01:32 time: 1.2197 data_time: 0.0015 memory: 15582 loss: 0.0083 +2023/06/05 02:35:49 - mmengine - INFO - Exp name: convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1_20230605_010924 +2023/06/05 02:35:49 - mmengine - INFO - Saving checkpoint at 10 epochs +2023/06/05 02:37:57 - mmengine - INFO - Epoch(val) [10][100/171] eta: 0:01:26 time: 0.4165 data_time: 0.0081 memory: 15582 +2023/06/05 02:41:11 - mmengine - INFO - Epoch(val) [10][171/171] accuracy/top1: 50.4894 single-label/precision_classwise: [50.23061752319336, 93.16239166259766] single-label/recall_classwise: [99.91753387451172, 1.1222457885742188] single-label/f1-score_classwise: [66.85289764404297, 2.21777606010437] data_time: 1.0866 time: 1.4986 diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/20230605_010924.json b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/20230605_010924.json new file mode 100644 index 0000000000000000000000000000000000000000..c88b7d7a607b80f6ddffc898cceb89b9e5ca7d68 --- /dev/null +++ b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/20230605_010924.json @@ -0,0 +1,20 @@ +{"lr": 0.0003967834896001145, "data_time": 0.002004241943359375, "loss": 0.060886699333786966, "time": 1.283958101272583, "epoch": 1, "memory": 15587, "step": 100} +{"accuracy/top1": 49.96909713745117, "data_time": 1.1120571992550676, "time": 1.5349291109899332, "step": 1} +{"lr": 0.000376497381485265, "data_time": 0.001748061180114746, "loss": 0.037596475379541514, "time": 1.3321064472198487, "epoch": 2, "memory": 15582, "step": 271} +{"accuracy/top1": 49.96909713745117, "data_time": 1.1107065372688825, "time": 1.5214041734850683, "step": 2} +{"lr": 0.00033942391477812043, "data_time": 0.0020072221755981444, "loss": 0.02871889676898718, "time": 1.243894839286804, "epoch": 3, "memory": 15582, "step": 442} +{"accuracy/top1": 49.969669342041016, "data_time": 1.1134790681129279, "time": 1.524714401988096, "step": 3} +{"lr": 0.0002891920987060062, "data_time": 0.027992820739746092, "loss": 0.019611307745799424, "time": 1.2641853094100952, "epoch": 4, "memory": 15582, "step": 613} +{"accuracy/top1": 49.97367477416992, "data_time": 1.1248791536619498, "time": 1.535257224426713, "step": 4} +{"lr": 0.000230718973411704, "data_time": 0.0023558378219604493, "loss": 0.010542389191687106, "time": 1.1881954193115234, "epoch": 5, "memory": 15582, "step": 784} +{"accuracy/top1": 49.985694885253906, "data_time": 1.1308857063914455, "time": 1.5399510125781215, "step": 5} +{"lr": 0.00016972829580523899, "data_time": 0.0016524553298950194, "loss": 0.012980625219643115, "time": 1.2902986526489257, "epoch": 6, "memory": 15582, "step": 955} +{"accuracy/top1": 50.01774597167969, "data_time": 1.1584049934564635, "time": 1.5692680367203646, "step": 6} +{"lr": 0.00011219025835777107, "data_time": 0.0016577005386352538, "loss": 0.012409109668806196, "time": 1.3285138845443725, "epoch": 7, "memory": 15582, "step": 1126} +{"accuracy/top1": 50.07669448852539, "data_time": 1.161834533824477, "time": 1.5730493900387785, "step": 7} +{"lr": 6.373708506573799e-05, "data_time": 0.002012157440185547, "loss": 0.007878543762490153, "time": 1.22543044090271, "epoch": 8, "memory": 15582, "step": 1297} +{"accuracy/top1": 50.16884231567383, "data_time": 1.1211500722308492, "time": 1.5300801077554391, "step": 8} +{"lr": 2.9111710124073225e-05, "data_time": 0.0017465591430664063, "loss": 0.00894083846360445, "time": 1.2186597824096679, "epoch": 9, "memory": 15582, "step": 1468} +{"accuracy/top1": 50.3142204284668, "data_time": 1.1055657600247584, "time": 1.515679121017456, "step": 9} +{"lr": 1.1703506481239862e-05, "data_time": 0.0015125274658203125, "loss": 0.008270366303622722, "time": 1.2197386503219605, "epoch": 10, "memory": 15582, "step": 1639} +{"accuracy/top1": 50.489356994628906, "data_time": 1.0865808545156967, "time": 1.4986090840295303, "step": 10} diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/config.py b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/config.py new file mode 100644 index 0000000000000000000000000000000000000000..d73748fea7730e205b16a806e1406b9f773cf3eb --- /dev/null +++ b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/config.py @@ -0,0 +1,247 @@ +optim_wrapper = dict( + optimizer=dict( + type='AdamW', + lr=0.0004, + weight_decay=0.05, + eps=1e-08, + betas=(0.9, 0.999), + _scope_='mmpretrain'), + paramwise_cfg=dict( + norm_decay_mult=0.0, + bias_decay_mult=0.0, + flat_decay_mult=0.0, + custom_keys=dict({ + '.absolute_pos_embed': dict(decay_mult=0.0), + '.relative_position_bias_table': dict(decay_mult=0.0) + })), + type='AmpOptimWrapper', + dtype='bfloat16', + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=False, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=4096) +model = dict( + type='ImageClassifier', + backbone=dict( + frozen_stages=2, + type='ConvNeXt', + arch='small', + drop_path_rate=0.4, + init_cfg=dict( + type='Pretrained', + checkpoint= + 'https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth', + prefix='backbone')), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=768, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + init_cfg=None), + init_cfg=dict( + type='TruncNormal', layer=['Conv2d', 'Linear'], std=0.02, bias=0.0), + train_cfg=None) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3fake8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3real8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3fake8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3real8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3fake8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3real8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1' diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/events.out.tfevents.1685898579.SH-IDC1-10-140-24-15.33681.0 b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/events.out.tfevents.1685898579.SH-IDC1-10-140-24-15.33681.0 new file mode 100644 index 0000000000000000000000000000000000000000..8fa431026006e0ee9f82f16fdd22dc764ceb35a2 --- /dev/null +++ b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/events.out.tfevents.1685898579.SH-IDC1-10-140-24-15.33681.0 @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:e73589b892ae35c85ae96f74979931410cc09fed330c413ae2986802346eefaf +size 6920202 diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/scalars.json b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/scalars.json new file mode 100644 index 0000000000000000000000000000000000000000..c88b7d7a607b80f6ddffc898cceb89b9e5ca7d68 --- /dev/null +++ b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/scalars.json @@ -0,0 +1,20 @@ +{"lr": 0.0003967834896001145, "data_time": 0.002004241943359375, "loss": 0.060886699333786966, "time": 1.283958101272583, "epoch": 1, "memory": 15587, "step": 100} +{"accuracy/top1": 49.96909713745117, "data_time": 1.1120571992550676, "time": 1.5349291109899332, "step": 1} +{"lr": 0.000376497381485265, "data_time": 0.001748061180114746, "loss": 0.037596475379541514, "time": 1.3321064472198487, "epoch": 2, "memory": 15582, "step": 271} +{"accuracy/top1": 49.96909713745117, "data_time": 1.1107065372688825, "time": 1.5214041734850683, "step": 2} +{"lr": 0.00033942391477812043, "data_time": 0.0020072221755981444, "loss": 0.02871889676898718, "time": 1.243894839286804, "epoch": 3, "memory": 15582, "step": 442} +{"accuracy/top1": 49.969669342041016, "data_time": 1.1134790681129279, "time": 1.524714401988096, "step": 3} +{"lr": 0.0002891920987060062, "data_time": 0.027992820739746092, "loss": 0.019611307745799424, "time": 1.2641853094100952, "epoch": 4, "memory": 15582, "step": 613} +{"accuracy/top1": 49.97367477416992, "data_time": 1.1248791536619498, "time": 1.535257224426713, "step": 4} +{"lr": 0.000230718973411704, "data_time": 0.0023558378219604493, "loss": 0.010542389191687106, "time": 1.1881954193115234, "epoch": 5, "memory": 15582, "step": 784} +{"accuracy/top1": 49.985694885253906, "data_time": 1.1308857063914455, "time": 1.5399510125781215, "step": 5} +{"lr": 0.00016972829580523899, "data_time": 0.0016524553298950194, "loss": 0.012980625219643115, "time": 1.2902986526489257, "epoch": 6, "memory": 15582, "step": 955} +{"accuracy/top1": 50.01774597167969, "data_time": 1.1584049934564635, "time": 1.5692680367203646, "step": 6} +{"lr": 0.00011219025835777107, "data_time": 0.0016577005386352538, "loss": 0.012409109668806196, "time": 1.3285138845443725, "epoch": 7, "memory": 15582, "step": 1126} +{"accuracy/top1": 50.07669448852539, "data_time": 1.161834533824477, "time": 1.5730493900387785, "step": 7} +{"lr": 6.373708506573799e-05, "data_time": 0.002012157440185547, "loss": 0.007878543762490153, "time": 1.22543044090271, "epoch": 8, "memory": 15582, "step": 1297} +{"accuracy/top1": 50.16884231567383, "data_time": 1.1211500722308492, "time": 1.5300801077554391, "step": 8} +{"lr": 2.9111710124073225e-05, "data_time": 0.0017465591430664063, "loss": 0.00894083846360445, "time": 1.2186597824096679, "epoch": 9, "memory": 15582, "step": 1468} +{"accuracy/top1": 50.3142204284668, "data_time": 1.1055657600247584, "time": 1.515679121017456, "step": 9} +{"lr": 1.1703506481239862e-05, "data_time": 0.0015125274658203125, "loss": 0.008270366303622722, "time": 1.2197386503219605, "epoch": 10, "memory": 15582, "step": 1639} +{"accuracy/top1": 50.489356994628906, "data_time": 1.0865808545156967, "time": 1.4986090840295303, "step": 10} diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/35624.png_1.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/35624.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..bbe68bed4ba6bc1f5f3542f79b64e655d8e36e95 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/35624.png_1.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/35624.png_10.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/35624.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..9642fb3b50fb92851b1509f814f0972ba20ef6f2 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/35624.png_10.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/35624.png_2.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/35624.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..cb5311250355dbb36d5783fc3f21f51dda9dc30c Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/35624.png_2.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/35624.png_3.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/35624.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..9d93c6ac1b66002cb40cf805d75da507e11d2e7f Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/35624.png_3.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/35624.png_4.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/35624.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..1c5710742c06f29b65e6c9ee8b08e8066127bf4f Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/35624.png_4.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/35624.png_5.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/35624.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..a89d8327e2a9bb8d67459cc50f4216976fef8803 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/35624.png_5.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/35624.png_6.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/35624.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..703d3da69296e5fbc2f595f21ea06bc57ff704f4 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/35624.png_6.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/35624.png_7.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/35624.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..66e9a8bb5c3e892690a1740035b0d772b4f5e062 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/35624.png_7.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/35624.png_8.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/35624.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..e487de7819ff53e36d6659c9b7eb8071fcbac779 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/35624.png_8.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/35624.png_9.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/35624.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..ac9bc0b6e42d67436e3f775f00beea747f7bcd16 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/35624.png_9.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/61024.png_1.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/61024.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..6b4cd0cadcbe4a6bf867e8065c01eed98595cc6f Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/61024.png_1.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/61024.png_10.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/61024.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..24c563bc38ce303c0ad77b4055ec28efc40701e0 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/61024.png_10.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/61024.png_2.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/61024.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..9000ef6fa7c11595892b40e9c724532673972caf Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/61024.png_2.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/61024.png_3.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/61024.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..43052d9c556e34d94d49c1515e68bf27e84dc424 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/61024.png_3.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/61024.png_4.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/61024.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..804b80ad715abb0290ad11ab140e4d3c5086797c Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/61024.png_4.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/61024.png_5.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/61024.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..fc6f8bbeb208283192661abe00e1a3a0c3e52593 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/61024.png_5.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/61024.png_6.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/61024.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..87a32a65e3fcccebe2d29bc7dded4e1ec1b6907d Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/61024.png_6.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/61024.png_7.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/61024.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..3a36bf784f86ddc157c98fe22fb5075b79d2f5dd Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/61024.png_7.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/61024.png_8.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/61024.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..c7acc76f98f36bd935a62b709660b62799cb2978 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/61024.png_8.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/61024.png_9.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/61024.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..8da0217465b2def239ea40a6817f439837738548 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/61024.png_9.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/66390.png_1.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/66390.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..25832a3d109805a10ed41eef1c21f98cf0860f9e Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/66390.png_1.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/66390.png_10.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/66390.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..0e75206dda46460859515590fbabf6b19fea5c04 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/66390.png_10.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/66390.png_2.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/66390.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..c3979e578b3ce1465a910aaa69ad7345714df028 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/66390.png_2.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/66390.png_3.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/66390.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..00ebd81d4586d97861c90d9343ab964fe3f4f283 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/66390.png_3.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/66390.png_4.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/66390.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..b2ce7761c941b13fdf9e921ff736b6da6a4a5df6 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/66390.png_4.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/66390.png_5.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/66390.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..0e779b29cba2d08f787851e82681eaa29967b8ac Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/66390.png_5.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/66390.png_6.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/66390.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..7905b104d1f596a8c0cbaab44311c7796f075609 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/66390.png_6.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/66390.png_7.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/66390.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..3f5efec3bb33f7cb97c15f46e05be946ce1710da Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/66390.png_7.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/66390.png_8.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/66390.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..95f4d6d52f93fb51773092d1358e39ac82d3898e Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/66390.png_8.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/66390.png_9.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/66390.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..4c1130ca7335497bd7d6d277b86ca716f521d8e3 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/66390.png_9.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/pixabay_dog_002352.jpg_1.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/pixabay_dog_002352.jpg_1.png new file mode 100644 index 0000000000000000000000000000000000000000..43c973dbb4ce4adde38a46a27c917ff51473f138 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/pixabay_dog_002352.jpg_1.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/pixabay_dog_002352.jpg_10.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/pixabay_dog_002352.jpg_10.png new file mode 100644 index 0000000000000000000000000000000000000000..c834fdb86b4f25fedf3a88fcf0dc8440743c9a81 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/pixabay_dog_002352.jpg_10.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/pixabay_dog_002352.jpg_2.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/pixabay_dog_002352.jpg_2.png new file mode 100644 index 0000000000000000000000000000000000000000..66e8f3d04af710bcc5867e65362c9a547743aa38 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/pixabay_dog_002352.jpg_2.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/pixabay_dog_002352.jpg_3.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/pixabay_dog_002352.jpg_3.png new file mode 100644 index 0000000000000000000000000000000000000000..23bbbd02bee00dbbc0194cd01713c0482ebd443f Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/pixabay_dog_002352.jpg_3.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/pixabay_dog_002352.jpg_4.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/pixabay_dog_002352.jpg_4.png new file mode 100644 index 0000000000000000000000000000000000000000..dd062699f0cd6ba2d059427dc399fdd69f27ddb5 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/pixabay_dog_002352.jpg_4.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/pixabay_dog_002352.jpg_5.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/pixabay_dog_002352.jpg_5.png new file mode 100644 index 0000000000000000000000000000000000000000..16d0239f3d38afcefadf108aa0ba38c750711ff8 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/pixabay_dog_002352.jpg_5.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/pixabay_dog_002352.jpg_6.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/pixabay_dog_002352.jpg_6.png new file mode 100644 index 0000000000000000000000000000000000000000..e65677099c1398e1573da554ad32180286ebc3b2 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/pixabay_dog_002352.jpg_6.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/pixabay_dog_002352.jpg_7.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/pixabay_dog_002352.jpg_7.png new file mode 100644 index 0000000000000000000000000000000000000000..f9712e7ff31867de0e96e92fc9b37c5a8a053931 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/pixabay_dog_002352.jpg_7.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/pixabay_dog_002352.jpg_8.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/pixabay_dog_002352.jpg_8.png new file mode 100644 index 0000000000000000000000000000000000000000..5e2d48b86d0c237392c2b274b16b5256b898f4f0 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/pixabay_dog_002352.jpg_8.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/pixabay_dog_002352.jpg_9.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/pixabay_dog_002352.jpg_9.png new file mode 100644 index 0000000000000000000000000000000000000000..9c76d1d71269d8cdaa3313abfe967fb731994ad8 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/pixabay_dog_002352.jpg_9.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/seed14088.png_1.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/seed14088.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..6b422c444bc841f088d1a39c6e79d32addb92718 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/seed14088.png_1.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/seed14088.png_10.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/seed14088.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..942dea678b3085bcde55dd186bdce2bc5aa97be9 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/seed14088.png_10.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/seed14088.png_2.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/seed14088.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..f5f1a10766a1ed01c4993dd24b1722c7e078b068 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/seed14088.png_2.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/seed14088.png_3.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/seed14088.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..9192a6ee7b7bc29c59e8f27014cbbe5b894c6216 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/seed14088.png_3.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/seed14088.png_4.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/seed14088.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..f1437d72efb1f700190597062618b86403e11c10 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/seed14088.png_4.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/seed14088.png_5.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/seed14088.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..25728794518eb1deeee15f285efa5355dbf2c072 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/seed14088.png_5.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/seed14088.png_6.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/seed14088.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..6cc22023422770c97c87853c520a8fb162e616a1 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/seed14088.png_6.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/seed14088.png_7.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/seed14088.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..e0b7825cdbc1dd2281d55dfd7aeaaf99d4483364 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/seed14088.png_7.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/seed14088.png_8.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/seed14088.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..ce5c8487b225c8cbff5a59a69792312f1454a2d8 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/seed14088.png_8.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/seed14088.png_9.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/seed14088.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..a10c7fd21c7176a1a7ca233085ba92205ada24d4 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/seed14088.png_9.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/seed17282.png_1.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/seed17282.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..751e0cdd908071753378d83d0731d567584273c9 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/seed17282.png_1.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/seed17282.png_10.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/seed17282.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..2e02e3e52511ed5f626c572a329cdd63f8ff2098 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/seed17282.png_10.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/seed17282.png_2.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/seed17282.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..4d4e01ea0568135ba7279df56212ba016eb8238a Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/seed17282.png_2.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/seed17282.png_3.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/seed17282.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..ca3520d531d4e08fc94263b97c78cdcfb12bf2a9 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/seed17282.png_3.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/seed17282.png_4.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/seed17282.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..002c17b92c65c26e9670722656e3eee96d8e4ee9 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/seed17282.png_4.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/seed17282.png_5.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/seed17282.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..d4fa37dc9ab4fe0687604303b9ed3dc8b5ba3b08 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/seed17282.png_5.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/seed17282.png_6.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/seed17282.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..da57b612de3824cb874f56515e17405895ae2f74 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/seed17282.png_6.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/seed17282.png_7.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/seed17282.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..cbc659757b9a38f016b48a4484588fbd3fabee9b Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/seed17282.png_7.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/seed17282.png_8.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/seed17282.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..6749bdf7cc3a08d6b1b8b7296ce3588dd5fe7cfb Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/seed17282.png_8.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/seed17282.png_9.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/seed17282.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..df603d8e34251c834340b36941d9847b6ccdd1f0 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/seed17282.png_9.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/seed23602.png_1.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/seed23602.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..0a86888f7353df3b9df21bd2a67b07acfcd188d5 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/seed23602.png_1.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/seed23602.png_10.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/seed23602.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..69ed8f6b4d6e7c4a24154bcb7b42cef1bd4a4894 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/seed23602.png_10.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/seed23602.png_2.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/seed23602.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..7d0c8d2d7fed0af5eb4a2b1a7263856d780d1e8c Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/seed23602.png_2.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/seed23602.png_3.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/seed23602.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..09a7e1305ee85f15f1700e04635940c7e4a20b08 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/seed23602.png_3.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/seed23602.png_4.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/seed23602.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..04391422690d6772006229579d26837ba51073eb Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/seed23602.png_4.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/seed23602.png_5.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/seed23602.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..30662d8627dac944b8096c0ce03c4db20b9e148a Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/seed23602.png_5.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/seed23602.png_6.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/seed23602.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..0269513a02d94b9836e7e487d37083dc84aa50aa Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/seed23602.png_6.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/seed23602.png_7.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/seed23602.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..5fca2d99721411632df6cbb8f8f760a6f6157728 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/seed23602.png_7.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/seed23602.png_8.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/seed23602.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..5ba3f3c1db5c70a5a473e8b8eaeea6fb860dac38 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/seed23602.png_8.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/seed23602.png_9.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/seed23602.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..35d22a04849579afebe16053c14164ca17f0ce14 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/seed23602.png_9.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/seed24366.png_1.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/seed24366.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..50e24dcc8dc8b53c88424307dc7ec360c8089818 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/seed24366.png_1.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/seed24366.png_10.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/seed24366.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..b86be6c4947679d1fdbcd0a0e746a76982ec411e Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/seed24366.png_10.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/seed24366.png_2.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/seed24366.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..5d26aac479d43daae26e9aaa91afa934fce80188 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/seed24366.png_2.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/seed24366.png_3.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/seed24366.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..734672ad5ab0cff217260d217f1bc5f808e85411 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/seed24366.png_3.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/seed24366.png_4.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/seed24366.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..d60cc20d485a71494ee7f0858a8d7015d4de1ef1 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/seed24366.png_4.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/seed24366.png_5.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/seed24366.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..178c69c3cc2c3eafdf1d0af639aad7043cb81236 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/seed24366.png_5.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/seed24366.png_6.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/seed24366.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..911faabf57bc60f983d5353ee1f56722c57da420 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/seed24366.png_6.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/seed24366.png_7.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/seed24366.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..e237688ba4390ec45818f5d5f000fcd4de72c716 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/seed24366.png_7.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/seed24366.png_8.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/seed24366.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..957eccca88f3752a90307df44ef524d28cfe0b87 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/seed24366.png_8.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/seed24366.png_9.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/seed24366.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..24b829635ccf5480665ca457b67b44122bb298ca Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/seed24366.png_9.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/seed44086.png_1.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/seed44086.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..abe072d0bd9c479674f93e8d21c58614f2c54336 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/seed44086.png_1.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/seed44086.png_10.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/seed44086.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..51663329886c503871b92d94a5cfe581102921ea Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/seed44086.png_10.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/seed44086.png_2.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/seed44086.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..d64e11ad9cd96f06bc65ccb0291db953e0705a34 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/seed44086.png_2.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/seed44086.png_3.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/seed44086.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..5a275503c3c9d4089c46c6cbc50ec14a902a5ac2 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/seed44086.png_3.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/seed44086.png_4.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/seed44086.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..0c4e696d7e06dccb93455940ae28181ab5bc7e41 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/seed44086.png_4.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/seed44086.png_5.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/seed44086.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..78231ddac24f6f7939dddee8d831ce0c99865d38 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/seed44086.png_5.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/seed44086.png_6.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/seed44086.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..f25eff467da064a492856c595380bf16bda39597 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/seed44086.png_6.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/seed44086.png_7.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/seed44086.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..619b8b8d8ff93492fe8df8daa7144af56a3bdaa0 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/seed44086.png_7.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/seed44086.png_8.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/seed44086.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..952d9cc3fd5bc616a1c6fcddba2ddbdad3eaa860 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/seed44086.png_8.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/seed44086.png_9.png b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/seed44086.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..5de26b05f50237b42ceed794a062ad8e2c1bd5d6 Binary files /dev/null and b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/20230605_010924/vis_data/vis_image/seed44086.png_9.png differ diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1.py b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1.py new file mode 100644 index 0000000000000000000000000000000000000000..d73748fea7730e205b16a806e1406b9f773cf3eb --- /dev/null +++ b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1.py @@ -0,0 +1,247 @@ +optim_wrapper = dict( + optimizer=dict( + type='AdamW', + lr=0.0004, + weight_decay=0.05, + eps=1e-08, + betas=(0.9, 0.999), + _scope_='mmpretrain'), + paramwise_cfg=dict( + norm_decay_mult=0.0, + bias_decay_mult=0.0, + flat_decay_mult=0.0, + custom_keys=dict({ + '.absolute_pos_embed': dict(decay_mult=0.0), + '.relative_position_bias_table': dict(decay_mult=0.0) + })), + type='AmpOptimWrapper', + dtype='bfloat16', + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=False, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=4096) +model = dict( + type='ImageClassifier', + backbone=dict( + frozen_stages=2, + type='ConvNeXt', + arch='small', + drop_path_rate=0.4, + init_cfg=dict( + type='Pretrained', + checkpoint= + 'https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_in21k-pre_3rdparty_in1k_20221219-aeca4c93.pth', + prefix='backbone')), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=768, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + init_cfg=None), + init_cfg=dict( + type='TruncNormal', layer=['Conv2d', 'Linear'], std=0.02, bias=0.0), + train_cfg=None) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3fake8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3real8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3fake8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3real8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3fake8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3real8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1' diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/epoch_1.pth b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/epoch_1.pth new file mode 100644 index 0000000000000000000000000000000000000000..7adb3bacb16c08353c7be05590110317eace68a7 --- /dev/null +++ b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/epoch_1.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:1c097068b43a953bd2bca761d0193bbadaf04dd8ba9785d58fd0baf894fb155d +size 781981274 diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/epoch_10.pth b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/epoch_10.pth new file mode 100644 index 0000000000000000000000000000000000000000..5756c8682173c4eef25a27b1cb77d0e7edb61480 --- /dev/null +++ b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/epoch_10.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:bd874970045059e4fa8033604ddce30f7bbb8a29c3d7e0ab406a0fb84c70cfca +size 782146906 diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/epoch_2.pth b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/epoch_2.pth new file mode 100644 index 0000000000000000000000000000000000000000..914c4185d09b904c5c4637ad73dbc0866a7d6f79 --- /dev/null +++ b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/epoch_2.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:41c2f19bba73eb0829135c2b718e18f1bff26a283ec230c30f63ab858dce1a0e +size 782000730 diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/epoch_3.pth b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/epoch_3.pth new file mode 100644 index 0000000000000000000000000000000000000000..df928bbc665ff095954ce572578a2bb79bba692b --- /dev/null +++ b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/epoch_3.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:84c3725d0e9955769f10fc0a6a427f5057218f2a6c7f7b9aec4432692cc5bbca +size 782019098 diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/epoch_4.pth b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/epoch_4.pth new file mode 100644 index 0000000000000000000000000000000000000000..9f7457cda2c30d7e7cfe0ca06f8c457c94c12a72 --- /dev/null +++ b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/epoch_4.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:41f22a297640d2f969ce7460f5f6507d3a63ddd28f66607b1bfaa8e29fee6954 +size 782037338 diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/epoch_5.pth b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/epoch_5.pth new file mode 100644 index 0000000000000000000000000000000000000000..0965cc8006b78b4c709f7d4b716197694a5234e6 --- /dev/null +++ b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/epoch_5.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:6fd3774804eae2d5bc1188749b277c08ca227c851b1a1f3f48606cc23c4c0479 +size 782055578 diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/epoch_6.pth b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/epoch_6.pth new file mode 100644 index 0000000000000000000000000000000000000000..73e5315c477d388c901ebd704bf68c8f2fea4897 --- /dev/null +++ b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/epoch_6.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ecaba9ccc2d5d3e625ee8837f9425b8e6fa1543b9f72e3742358476c85861d1b +size 782073818 diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/epoch_7.pth b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/epoch_7.pth new file mode 100644 index 0000000000000000000000000000000000000000..b3484fa7849cfb1a9acb06e0bab0375f6e2c23bc --- /dev/null +++ b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/epoch_7.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:1cda558dd725e1f275854f932d0c0bddcfbf9d072557953c85437f10a27b9bce +size 782092058 diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/epoch_8.pth b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/epoch_8.pth new file mode 100644 index 0000000000000000000000000000000000000000..950557d78945517e389c04553ce31cad797b61a2 --- /dev/null +++ b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/epoch_8.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:4778c7a9ab94438ebc88c7ab182260c09e3820ee3e364ab6012d5c0d2701190a +size 782110298 diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/epoch_9.pth b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/epoch_9.pth new file mode 100644 index 0000000000000000000000000000000000000000..2f817decd3d57cc48619720916733a556ffd7acb --- /dev/null +++ b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/epoch_9.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:24b176c44d177da21ff1538ffe7151c9c80b87ecac3540c52e56c603eaa71ab4 +size 782128538 diff --git a/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/last_checkpoint b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/last_checkpoint new file mode 100644 index 0000000000000000000000000000000000000000..b786724073bc1f598a63b0cf869bcabd59c147d3 --- /dev/null +++ b/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/last_checkpoint @@ -0,0 +1 @@ +/mnt/petrelfs/luzeyu/workspace/fakebench/mmpretrain/workdir/convnext_small_4xb256_stylegan3_1m_lr4e-4_aug_5e-1/epoch_10.pth \ No newline at end of file diff --git a/convnext_tiny_4xb1024_4e-3lr_5m/20230531_172740/20230531_172740.log b/convnext_tiny_4xb1024_4e-3lr_5m/20230531_172740/20230531_172740.log new file mode 100644 index 0000000000000000000000000000000000000000..65e075f7ec1dc6cea23a1f9321f5038143c48aca --- /dev/null +++ b/convnext_tiny_4xb1024_4e-3lr_5m/20230531_172740/20230531_172740.log @@ -0,0 +1,1275 @@ +2023/05/31 17:27:44 - mmengine - INFO - +------------------------------------------------------------ +System environment: + sys.platform: linux + Python: 3.10.9 (main, Mar 8 2023, 10:47:38) [GCC 11.2.0] + CUDA available: True + numpy_random_seed: 622910272 + GPU 0,1,2,3: NVIDIA A100-SXM4-80GB + CUDA_HOME: /mnt/petrelfs/share/cuda-11.6 + NVCC: Cuda compilation tools, release 11.6, V11.6.124 + GCC: gcc (GCC) 4.8.5 20150623 (Red Hat 4.8.5-44) + PyTorch: 1.13.1 + PyTorch compiling details: PyTorch built with: + - GCC 9.3 + - C++ Version: 201402 + - Intel(R) oneAPI Math Kernel Library Version 2021.4-Product Build 20210904 for Intel(R) 64 architecture applications + - Intel(R) MKL-DNN v2.6.0 (Git Hash 52b5f107dd9cf10910aaa19cb47f3abf9b349815) + - OpenMP 201511 (a.k.a. OpenMP 4.5) + - LAPACK is enabled (usually provided by MKL) + - NNPACK is enabled + - CPU capability usage: AVX2 + - CUDA Runtime 11.6 + - NVCC architecture flags: -gencode;arch=compute_37,code=sm_37;-gencode;arch=compute_50,code=sm_50;-gencode;arch=compute_60,code=sm_60;-gencode;arch=compute_61,code=sm_61;-gencode;arch=compute_70,code=sm_70;-gencode;arch=compute_75,code=sm_75;-gencode;arch=compute_80,code=sm_80;-gencode;arch=compute_86,code=sm_86;-gencode;arch=compute_37,code=compute_37 + - CuDNN 8.3.2 (built against CUDA 11.5) + - Magma 2.6.1 + - Build settings: BLAS_INFO=mkl, BUILD_TYPE=Release, CUDA_VERSION=11.6, CUDNN_VERSION=8.3.2, CXX_COMPILER=/opt/rh/devtoolset-9/root/usr/bin/c++, CXX_FLAGS= -fabi-version=11 -Wno-deprecated -fvisibility-inlines-hidden -DUSE_PTHREADPOOL -fopenmp -DNDEBUG -DUSE_KINETO -DUSE_FBGEMM -DUSE_QNNPACK -DUSE_PYTORCH_QNNPACK -DUSE_XNNPACK -DSYMBOLICATE_MOBILE_DEBUG_HANDLE -DEDGE_PROFILER_USE_KINETO -O2 -fPIC -Wno-narrowing -Wall -Wextra -Werror=return-type -Werror=non-virtual-dtor -Wno-missing-field-initializers -Wno-type-limits -Wno-array-bounds -Wno-unknown-pragmas -Wunused-local-typedefs -Wno-unused-parameter -Wno-unused-function -Wno-unused-result -Wno-strict-overflow -Wno-strict-aliasing -Wno-error=deprecated-declarations -Wno-stringop-overflow -Wno-psabi -Wno-error=pedantic -Wno-error=redundant-decls -Wno-error=old-style-cast -fdiagnostics-color=always -faligned-new -Wno-unused-but-set-variable -Wno-maybe-uninitialized -fno-math-errno -fno-trapping-math -Werror=format -Werror=cast-function-type -Wno-stringop-overflow, LAPACK_INFO=mkl, PERF_WITH_AVX=1, PERF_WITH_AVX2=1, PERF_WITH_AVX512=1, TORCH_VERSION=1.13.1, USE_CUDA=ON, USE_CUDNN=ON, USE_EXCEPTION_PTR=1, USE_GFLAGS=OFF, USE_GLOG=OFF, USE_MKL=ON, USE_MKLDNN=ON, USE_MPI=OFF, USE_NCCL=ON, USE_NNPACK=ON, USE_OPENMP=ON, USE_ROCM=OFF, + + TorchVision: 0.14.1 + OpenCV: 4.7.0 + MMEngine: 0.7.3 + +Runtime environment: + cudnn_benchmark: True + mp_cfg: {'mp_start_method': 'fork', 'opencv_num_threads': 0} + dist_cfg: {'backend': 'nccl'} + seed: None + deterministic: False + Distributed launcher: slurm + Distributed training: True + GPU number: 4 +------------------------------------------------------------ + +2023/05/31 17:27:44 - mmengine - INFO - Config: +optim_wrapper = dict( + optimizer=dict( + type='AdamW', + lr=0.004, + weight_decay=0.05, + eps=1e-08, + betas=(0.9, 0.999), + _scope_='mmpretrain'), + paramwise_cfg=dict( + norm_decay_mult=0.0, + bias_decay_mult=0.0, + flat_decay_mult=0.0, + custom_keys=dict({ + '.absolute_pos_embed': dict(decay_mult=0.0), + '.relative_position_bias_table': dict(decay_mult=0.0) + })), + type='AmpOptimWrapper', + dtype='bfloat16', + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=True, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=20, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=4096) +model = dict( + type='ImageClassifier', + backbone=dict(type='ConvNeXt', arch='tiny', drop_path_rate=0.1), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=768, + loss=dict( + type='LabelSmoothLoss', label_smooth_val=0.1, mode='original'), + init_cfg=None), + init_cfg=dict( + type='TruncNormal', layer=['Conv2d', 'Linear'], std=0.02, bias=0.0), + train_cfg=None) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=1024, + num_workers=10, + dataset=dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stablediffusionV1-5R2-dpmsolver-25-5m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/convnext_tiny_4xb1024_4e-3lr_5m' + +2023/05/31 17:27:48 - mmengine - INFO - Hooks will be executed in the following order: +before_run: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_load_checkpoint: +(ABOVE_NORMAL) EMAHook + -------------------- +before_train: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_train_epoch: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(NORMAL ) DistSamplerSeedHook + -------------------- +before_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook + -------------------- +after_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +after_train_epoch: +(NORMAL ) IterTimerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_val_epoch: +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook + -------------------- +before_val_iter: +(NORMAL ) IterTimerHook + -------------------- +after_val_iter: +(NORMAL ) IterTimerHook +(NORMAL ) VisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_val_epoch: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_save_checkpoint: +(ABOVE_NORMAL) EMAHook + -------------------- +after_train: +(VERY_LOW ) CheckpointHook + -------------------- +before_test_epoch: +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook + -------------------- +before_test_iter: +(NORMAL ) IterTimerHook + -------------------- +after_test_iter: +(NORMAL ) IterTimerHook +(NORMAL ) VisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_test_epoch: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_run: +(BELOW_NORMAL) LoggerHook + -------------------- +2023/05/31 17:28:15 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.0.0.bias:weight_decay=0.0 +2023/05/31 17:28:15 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.0.1.weight:weight_decay=0.0 +2023/05/31 17:28:15 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.0.1.bias:weight_decay=0.0 +2023/05/31 17:28:15 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.1.0.weight:weight_decay=0.0 +2023/05/31 17:28:15 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.1.0.bias:weight_decay=0.0 +2023/05/31 17:28:15 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.1.1.bias:weight_decay=0.0 +2023/05/31 17:28:15 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.2.0.weight:weight_decay=0.0 +2023/05/31 17:28:15 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.2.0.bias:weight_decay=0.0 +2023/05/31 17:28:15 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.2.1.bias:weight_decay=0.0 +2023/05/31 17:28:15 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.3.0.weight:weight_decay=0.0 +2023/05/31 17:28:15 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.3.0.bias:weight_decay=0.0 +2023/05/31 17:28:15 - mmengine - INFO - paramwise_options -- backbone.downsample_layers.3.1.bias:weight_decay=0.0 +2023/05/31 17:28:15 - mmengine - INFO - paramwise_options -- backbone.stages.0.0.gamma:weight_decay=0.0 +2023/05/31 17:28:15 - mmengine - INFO - paramwise_options -- backbone.stages.0.0.depthwise_conv.bias:weight_decay=0.0 +2023/05/31 17:28:15 - mmengine - INFO - paramwise_options -- backbone.stages.0.0.norm.weight:weight_decay=0.0 +2023/05/31 17:28:15 - mmengine - INFO - paramwise_options -- backbone.stages.0.0.norm.bias:weight_decay=0.0 +2023/05/31 17:28:15 - mmengine - INFO - paramwise_options -- backbone.stages.0.0.pointwise_conv1.bias:weight_decay=0.0 +2023/05/31 17:28:15 - mmengine - INFO - paramwise_options -- backbone.stages.0.0.pointwise_conv2.bias:weight_decay=0.0 +2023/05/31 17:28:15 - mmengine - INFO - paramwise_options -- backbone.stages.0.1.gamma:weight_decay=0.0 +2023/05/31 17:28:15 - mmengine - INFO - paramwise_options -- backbone.stages.0.1.depthwise_conv.bias:weight_decay=0.0 +2023/05/31 17:28:15 - mmengine - INFO - paramwise_options -- backbone.stages.0.1.norm.weight:weight_decay=0.0 +2023/05/31 17:28:15 - mmengine - INFO - paramwise_options -- backbone.stages.0.1.norm.bias:weight_decay=0.0 +2023/05/31 17:28:15 - mmengine - INFO - paramwise_options -- backbone.stages.0.1.pointwise_conv1.bias:weight_decay=0.0 +2023/05/31 17:28:15 - mmengine - INFO - paramwise_options -- backbone.stages.0.1.pointwise_conv2.bias:weight_decay=0.0 +2023/05/31 17:28:15 - mmengine - INFO - paramwise_options -- backbone.stages.0.2.gamma:weight_decay=0.0 +2023/05/31 17:28:15 - mmengine - INFO - paramwise_options -- backbone.stages.0.2.depthwise_conv.bias:weight_decay=0.0 +2023/05/31 17:28:15 - mmengine - INFO - paramwise_options -- backbone.stages.0.2.norm.weight:weight_decay=0.0 +2023/05/31 17:28:15 - mmengine - INFO - paramwise_options -- backbone.stages.0.2.norm.bias:weight_decay=0.0 +2023/05/31 17:28:15 - mmengine - INFO - paramwise_options -- backbone.stages.0.2.pointwise_conv1.bias:weight_decay=0.0 +2023/05/31 17:28:15 - mmengine - INFO - paramwise_options -- backbone.stages.0.2.pointwise_conv2.bias:weight_decay=0.0 +2023/05/31 17:28:15 - mmengine - INFO - paramwise_options -- backbone.stages.1.0.gamma:weight_decay=0.0 +2023/05/31 17:28:15 - mmengine - INFO - paramwise_options -- backbone.stages.1.0.depthwise_conv.bias:weight_decay=0.0 +2023/05/31 17:28:15 - mmengine - INFO - paramwise_options -- backbone.stages.1.0.norm.weight:weight_decay=0.0 +2023/05/31 17:28:15 - mmengine - INFO - paramwise_options -- backbone.stages.1.0.norm.bias:weight_decay=0.0 +2023/05/31 17:28:15 - mmengine - INFO - paramwise_options -- backbone.stages.1.0.pointwise_conv1.bias:weight_decay=0.0 +2023/05/31 17:28:15 - mmengine - INFO - paramwise_options -- backbone.stages.1.0.pointwise_conv2.bias:weight_decay=0.0 +2023/05/31 17:28:15 - mmengine - INFO - paramwise_options -- backbone.stages.1.1.gamma:weight_decay=0.0 +2023/05/31 17:28:15 - mmengine - INFO - paramwise_options -- backbone.stages.1.1.depthwise_conv.bias:weight_decay=0.0 +2023/05/31 17:28:15 - mmengine - INFO - paramwise_options -- backbone.stages.1.1.norm.weight:weight_decay=0.0 +2023/05/31 17:28:15 - mmengine - INFO - paramwise_options -- backbone.stages.1.1.norm.bias:weight_decay=0.0 +2023/05/31 17:28:15 - mmengine - INFO - paramwise_options -- backbone.stages.1.1.pointwise_conv1.bias:weight_decay=0.0 +2023/05/31 17:28:15 - mmengine - INFO - paramwise_options -- backbone.stages.1.1.pointwise_conv2.bias:weight_decay=0.0 +2023/05/31 17:28:15 - mmengine - INFO - paramwise_options -- backbone.stages.1.2.gamma:weight_decay=0.0 +2023/05/31 17:28:15 - mmengine - INFO - paramwise_options -- backbone.stages.1.2.depthwise_conv.bias:weight_decay=0.0 +2023/05/31 17:28:15 - mmengine - INFO - paramwise_options -- backbone.stages.1.2.norm.weight:weight_decay=0.0 +2023/05/31 17:28:15 - mmengine - INFO - paramwise_options -- backbone.stages.1.2.norm.bias:weight_decay=0.0 +2023/05/31 17:28:15 - mmengine - INFO - paramwise_options -- backbone.stages.1.2.pointwise_conv1.bias:weight_decay=0.0 +2023/05/31 17:28:15 - mmengine - INFO - paramwise_options -- backbone.stages.1.2.pointwise_conv2.bias:weight_decay=0.0 +2023/05/31 17:28:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.0.gamma:weight_decay=0.0 +2023/05/31 17:28:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.0.depthwise_conv.bias:weight_decay=0.0 +2023/05/31 17:28:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.0.norm.weight:weight_decay=0.0 +2023/05/31 17:28:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.0.norm.bias:weight_decay=0.0 +2023/05/31 17:28:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.0.pointwise_conv1.bias:weight_decay=0.0 +2023/05/31 17:28:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.0.pointwise_conv2.bias:weight_decay=0.0 +2023/05/31 17:28:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.1.gamma:weight_decay=0.0 +2023/05/31 17:28:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.1.depthwise_conv.bias:weight_decay=0.0 +2023/05/31 17:28:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.1.norm.weight:weight_decay=0.0 +2023/05/31 17:28:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.1.norm.bias:weight_decay=0.0 +2023/05/31 17:28:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.1.pointwise_conv1.bias:weight_decay=0.0 +2023/05/31 17:28:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.1.pointwise_conv2.bias:weight_decay=0.0 +2023/05/31 17:28:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.2.gamma:weight_decay=0.0 +2023/05/31 17:28:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.2.depthwise_conv.bias:weight_decay=0.0 +2023/05/31 17:28:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.2.norm.weight:weight_decay=0.0 +2023/05/31 17:28:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.2.norm.bias:weight_decay=0.0 +2023/05/31 17:28:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.2.pointwise_conv1.bias:weight_decay=0.0 +2023/05/31 17:28:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.2.pointwise_conv2.bias:weight_decay=0.0 +2023/05/31 17:28:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.3.gamma:weight_decay=0.0 +2023/05/31 17:28:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.3.depthwise_conv.bias:weight_decay=0.0 +2023/05/31 17:28:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.3.norm.weight:weight_decay=0.0 +2023/05/31 17:28:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.3.norm.bias:weight_decay=0.0 +2023/05/31 17:28:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.3.pointwise_conv1.bias:weight_decay=0.0 +2023/05/31 17:28:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.3.pointwise_conv2.bias:weight_decay=0.0 +2023/05/31 17:28:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.4.gamma:weight_decay=0.0 +2023/05/31 17:28:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.4.depthwise_conv.bias:weight_decay=0.0 +2023/05/31 17:28:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.4.norm.weight:weight_decay=0.0 +2023/05/31 17:28:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.4.norm.bias:weight_decay=0.0 +2023/05/31 17:28:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.4.pointwise_conv1.bias:weight_decay=0.0 +2023/05/31 17:28:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.4.pointwise_conv2.bias:weight_decay=0.0 +2023/05/31 17:28:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.5.gamma:weight_decay=0.0 +2023/05/31 17:28:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.5.depthwise_conv.bias:weight_decay=0.0 +2023/05/31 17:28:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.5.norm.weight:weight_decay=0.0 +2023/05/31 17:28:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.5.norm.bias:weight_decay=0.0 +2023/05/31 17:28:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.5.pointwise_conv1.bias:weight_decay=0.0 +2023/05/31 17:28:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.5.pointwise_conv2.bias:weight_decay=0.0 +2023/05/31 17:28:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.6.gamma:weight_decay=0.0 +2023/05/31 17:28:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.6.depthwise_conv.bias:weight_decay=0.0 +2023/05/31 17:28:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.6.norm.weight:weight_decay=0.0 +2023/05/31 17:28:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.6.norm.bias:weight_decay=0.0 +2023/05/31 17:28:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.6.pointwise_conv1.bias:weight_decay=0.0 +2023/05/31 17:28:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.6.pointwise_conv2.bias:weight_decay=0.0 +2023/05/31 17:28:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.7.gamma:weight_decay=0.0 +2023/05/31 17:28:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.7.depthwise_conv.bias:weight_decay=0.0 +2023/05/31 17:28:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.7.norm.weight:weight_decay=0.0 +2023/05/31 17:28:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.7.norm.bias:weight_decay=0.0 +2023/05/31 17:28:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.7.pointwise_conv1.bias:weight_decay=0.0 +2023/05/31 17:28:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.7.pointwise_conv2.bias:weight_decay=0.0 +2023/05/31 17:28:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.8.gamma:weight_decay=0.0 +2023/05/31 17:28:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.8.depthwise_conv.bias:weight_decay=0.0 +2023/05/31 17:28:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.8.norm.weight:weight_decay=0.0 +2023/05/31 17:28:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.8.norm.bias:weight_decay=0.0 +2023/05/31 17:28:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.8.pointwise_conv1.bias:weight_decay=0.0 +2023/05/31 17:28:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.8.pointwise_conv2.bias:weight_decay=0.0 +2023/05/31 17:28:15 - mmengine - INFO - paramwise_options -- backbone.stages.3.0.gamma:weight_decay=0.0 +2023/05/31 17:28:15 - mmengine - INFO - paramwise_options -- backbone.stages.3.0.depthwise_conv.bias:weight_decay=0.0 +2023/05/31 17:28:15 - mmengine - INFO - paramwise_options -- backbone.stages.3.0.norm.weight:weight_decay=0.0 +2023/05/31 17:28:15 - mmengine - INFO - paramwise_options -- backbone.stages.3.0.norm.bias:weight_decay=0.0 +2023/05/31 17:28:15 - mmengine - INFO - paramwise_options -- backbone.stages.3.0.pointwise_conv1.bias:weight_decay=0.0 +2023/05/31 17:28:15 - mmengine - INFO - paramwise_options -- backbone.stages.3.0.pointwise_conv2.bias:weight_decay=0.0 +2023/05/31 17:28:15 - mmengine - INFO - paramwise_options -- backbone.stages.3.1.gamma:weight_decay=0.0 +2023/05/31 17:28:15 - mmengine - INFO - paramwise_options -- backbone.stages.3.1.depthwise_conv.bias:weight_decay=0.0 +2023/05/31 17:28:15 - mmengine - INFO - paramwise_options -- backbone.stages.3.1.norm.weight:weight_decay=0.0 +2023/05/31 17:28:15 - mmengine - INFO - paramwise_options -- backbone.stages.3.1.norm.bias:weight_decay=0.0 +2023/05/31 17:28:15 - mmengine - INFO - paramwise_options -- backbone.stages.3.1.pointwise_conv1.bias:weight_decay=0.0 +2023/05/31 17:28:15 - mmengine - INFO - paramwise_options -- backbone.stages.3.1.pointwise_conv2.bias:weight_decay=0.0 +2023/05/31 17:28:15 - mmengine - INFO - paramwise_options -- backbone.stages.3.2.gamma:weight_decay=0.0 +2023/05/31 17:28:15 - mmengine - INFO - paramwise_options -- backbone.stages.3.2.depthwise_conv.bias:weight_decay=0.0 +2023/05/31 17:28:15 - mmengine - INFO - paramwise_options -- backbone.stages.3.2.norm.weight:weight_decay=0.0 +2023/05/31 17:28:15 - mmengine - INFO - paramwise_options -- backbone.stages.3.2.norm.bias:weight_decay=0.0 +2023/05/31 17:28:15 - mmengine - INFO - paramwise_options -- backbone.stages.3.2.pointwise_conv1.bias:weight_decay=0.0 +2023/05/31 17:28:15 - mmengine - INFO - paramwise_options -- backbone.stages.3.2.pointwise_conv2.bias:weight_decay=0.0 +2023/05/31 17:28:15 - mmengine - INFO - paramwise_options -- backbone.norm3.weight:weight_decay=0.0 +2023/05/31 17:28:15 - mmengine - INFO - paramwise_options -- backbone.norm3.bias:weight_decay=0.0 +2023/05/31 17:28:15 - mmengine - INFO - paramwise_options -- head.fc.bias:weight_decay=0.0 +Name of parameter - Initialization information + +backbone.downsample_layers.0.0.weight - torch.Size([96, 3, 4, 4]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.downsample_layers.0.0.bias - torch.Size([96]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.downsample_layers.0.1.weight - torch.Size([96]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.downsample_layers.0.1.bias - torch.Size([96]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.downsample_layers.1.0.weight - torch.Size([96]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.downsample_layers.1.0.bias - torch.Size([96]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.downsample_layers.1.1.weight - torch.Size([192, 96, 2, 2]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.downsample_layers.1.1.bias - torch.Size([192]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.downsample_layers.2.0.weight - torch.Size([192]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.downsample_layers.2.0.bias - torch.Size([192]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.downsample_layers.2.1.weight - torch.Size([384, 192, 2, 2]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.downsample_layers.2.1.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.downsample_layers.3.0.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.downsample_layers.3.0.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.downsample_layers.3.1.weight - torch.Size([768, 384, 2, 2]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.downsample_layers.3.1.bias - torch.Size([768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.0.gamma - torch.Size([96]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.0.0.depthwise_conv.weight - torch.Size([96, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.0.depthwise_conv.bias - torch.Size([96]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.0.norm.weight - torch.Size([96]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.0.0.norm.bias - torch.Size([96]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.0.0.pointwise_conv1.weight - torch.Size([384, 96]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.0.pointwise_conv1.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.0.pointwise_conv2.weight - torch.Size([96, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.0.pointwise_conv2.bias - torch.Size([96]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.1.gamma - torch.Size([96]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.0.1.depthwise_conv.weight - torch.Size([96, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.1.depthwise_conv.bias - torch.Size([96]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.1.norm.weight - torch.Size([96]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.0.1.norm.bias - torch.Size([96]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.0.1.pointwise_conv1.weight - torch.Size([384, 96]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.1.pointwise_conv1.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.1.pointwise_conv2.weight - torch.Size([96, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.1.pointwise_conv2.bias - torch.Size([96]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.2.gamma - torch.Size([96]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.0.2.depthwise_conv.weight - torch.Size([96, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.2.depthwise_conv.bias - torch.Size([96]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.2.norm.weight - torch.Size([96]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.0.2.norm.bias - torch.Size([96]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.0.2.pointwise_conv1.weight - torch.Size([384, 96]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.2.pointwise_conv1.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.2.pointwise_conv2.weight - torch.Size([96, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.2.pointwise_conv2.bias - torch.Size([96]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.0.gamma - torch.Size([192]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.1.0.depthwise_conv.weight - torch.Size([192, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.0.depthwise_conv.bias - torch.Size([192]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.0.norm.weight - torch.Size([192]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.1.0.norm.bias - torch.Size([192]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.1.0.pointwise_conv1.weight - torch.Size([768, 192]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.0.pointwise_conv1.bias - torch.Size([768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.0.pointwise_conv2.weight - torch.Size([192, 768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.0.pointwise_conv2.bias - torch.Size([192]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.1.gamma - torch.Size([192]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.1.1.depthwise_conv.weight - torch.Size([192, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.1.depthwise_conv.bias - torch.Size([192]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.1.norm.weight - torch.Size([192]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.1.1.norm.bias - torch.Size([192]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.1.1.pointwise_conv1.weight - torch.Size([768, 192]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.1.pointwise_conv1.bias - torch.Size([768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.1.pointwise_conv2.weight - torch.Size([192, 768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.1.pointwise_conv2.bias - torch.Size([192]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.2.gamma - torch.Size([192]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.1.2.depthwise_conv.weight - torch.Size([192, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.2.depthwise_conv.bias - torch.Size([192]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.2.norm.weight - torch.Size([192]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.1.2.norm.bias - torch.Size([192]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.1.2.pointwise_conv1.weight - torch.Size([768, 192]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.2.pointwise_conv1.bias - torch.Size([768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.2.pointwise_conv2.weight - torch.Size([192, 768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.2.pointwise_conv2.bias - torch.Size([192]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.0.gamma - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.0.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.0.depthwise_conv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.0.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.0.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.0.pointwise_conv1.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.0.pointwise_conv1.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.0.pointwise_conv2.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.0.pointwise_conv2.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.1.gamma - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.1.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.1.depthwise_conv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.1.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.1.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.1.pointwise_conv1.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.1.pointwise_conv1.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.1.pointwise_conv2.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.1.pointwise_conv2.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.2.gamma - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.2.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.2.depthwise_conv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.2.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.2.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.2.pointwise_conv1.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.2.pointwise_conv1.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.2.pointwise_conv2.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.2.pointwise_conv2.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.3.gamma - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.3.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.3.depthwise_conv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.3.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.3.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.3.pointwise_conv1.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.3.pointwise_conv1.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.3.pointwise_conv2.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.3.pointwise_conv2.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.4.gamma - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.4.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.4.depthwise_conv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.4.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.4.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.4.pointwise_conv1.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.4.pointwise_conv1.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.4.pointwise_conv2.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.4.pointwise_conv2.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.5.gamma - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.5.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.5.depthwise_conv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.5.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.5.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.5.pointwise_conv1.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.5.pointwise_conv1.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.5.pointwise_conv2.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.5.pointwise_conv2.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.6.gamma - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.6.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.6.depthwise_conv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.6.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.6.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.6.pointwise_conv1.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.6.pointwise_conv1.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.6.pointwise_conv2.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.6.pointwise_conv2.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.7.gamma - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.7.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.7.depthwise_conv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.7.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.7.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.7.pointwise_conv1.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.7.pointwise_conv1.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.7.pointwise_conv2.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.7.pointwise_conv2.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.8.gamma - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.8.depthwise_conv.weight - torch.Size([384, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.8.depthwise_conv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.8.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.8.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.8.pointwise_conv1.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.8.pointwise_conv1.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.8.pointwise_conv2.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.8.pointwise_conv2.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.0.gamma - torch.Size([768]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.3.0.depthwise_conv.weight - torch.Size([768, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.0.depthwise_conv.bias - torch.Size([768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.0.norm.weight - torch.Size([768]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.3.0.norm.bias - torch.Size([768]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.3.0.pointwise_conv1.weight - torch.Size([3072, 768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.0.pointwise_conv1.bias - torch.Size([3072]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.0.pointwise_conv2.weight - torch.Size([768, 3072]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.0.pointwise_conv2.bias - torch.Size([768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.1.gamma - torch.Size([768]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.3.1.depthwise_conv.weight - torch.Size([768, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.1.depthwise_conv.bias - torch.Size([768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.1.norm.weight - torch.Size([768]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.3.1.norm.bias - torch.Size([768]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.3.1.pointwise_conv1.weight - torch.Size([3072, 768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.1.pointwise_conv1.bias - torch.Size([3072]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.1.pointwise_conv2.weight - torch.Size([768, 3072]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.1.pointwise_conv2.bias - torch.Size([768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.2.gamma - torch.Size([768]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.3.2.depthwise_conv.weight - torch.Size([768, 1, 7, 7]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.2.depthwise_conv.bias - torch.Size([768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.2.norm.weight - torch.Size([768]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.3.2.norm.bias - torch.Size([768]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.3.2.pointwise_conv1.weight - torch.Size([3072, 768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.2.pointwise_conv1.bias - torch.Size([3072]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.2.pointwise_conv2.weight - torch.Size([768, 3072]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.2.pointwise_conv2.bias - torch.Size([768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.norm3.weight - torch.Size([768]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.norm3.bias - torch.Size([768]): +The value is the same before and after calling `init_weights` of ImageClassifier + +head.fc.weight - torch.Size([2, 768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +head.fc.bias - torch.Size([2]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 +2023/05/31 17:28:15 - mmengine - WARNING - "FileClient" will be deprecated in future. Please use io functions in https://mmengine.readthedocs.io/en/latest/api/fileio.html#file-io +2023/05/31 17:28:15 - mmengine - WARNING - "HardDiskBackend" is the alias of "LocalBackend" and the former will be deprecated in future. +2023/05/31 17:28:15 - mmengine - INFO - Checkpoints will be saved to /mnt/petrelfs/luzeyu/workspace/fakebench/mmpretrain/workdir/convnext_tiny_4xb1024_4e-3lr_5m. +2023/05/31 17:38:14 - mmengine - INFO - Epoch(train) [1][ 100/1440] lr: 4.0000e-03 eta: 1 day, 23:42:27 time: 4.3614 data_time: 1.7131 memory: 61143 loss: 0.6281 +2023/05/31 17:45:17 - mmengine - INFO - Epoch(train) [1][ 200/1440] lr: 4.0000e-03 eta: 1 day, 16:36:00 time: 4.2238 data_time: 1.3604 memory: 61143 loss: 0.6048 +2023/05/31 17:51:44 - mmengine - INFO - Epoch(train) [1][ 300/1440] lr: 4.0000e-03 eta: 1 day, 13:10:09 time: 3.8278 data_time: 0.8225 memory: 61143 loss: 0.5771 +2023/05/31 17:58:18 - mmengine - INFO - Epoch(train) [1][ 400/1440] lr: 4.0000e-03 eta: 1 day, 11:33:08 time: 4.0650 data_time: 1.9558 memory: 61143 loss: 0.5526 +2023/05/31 18:04:48 - mmengine - INFO - Epoch(train) [1][ 500/1440] lr: 4.0000e-03 eta: 1 day, 10:28:34 time: 3.9281 data_time: 2.1526 memory: 61143 loss: 0.5663 +2023/05/31 18:10:47 - mmengine - INFO - Epoch(train) [1][ 600/1440] lr: 4.0000e-03 eta: 1 day, 9:18:32 time: 3.4568 data_time: 1.6816 memory: 61143 loss: 0.5426 +2023/05/31 18:16:16 - mmengine - INFO - Epoch(train) [1][ 700/1440] lr: 4.0000e-03 eta: 1 day, 8:07:34 time: 3.1568 data_time: 1.3889 memory: 61143 loss: 0.5292 +2023/05/31 18:22:13 - mmengine - INFO - Epoch(train) [1][ 800/1440] lr: 4.0000e-03 eta: 1 day, 7:28:23 time: 3.6402 data_time: 1.8797 memory: 61143 loss: 0.5095 +2023/05/31 18:28:22 - mmengine - INFO - Epoch(train) [1][ 900/1440] lr: 4.0000e-03 eta: 1 day, 7:03:20 time: 3.6294 data_time: 1.8447 memory: 61143 loss: 0.4994 +2023/05/31 18:34:00 - mmengine - INFO - Exp name: convnext_tiny_4xb1024_fake5m_20230531_172740 +2023/05/31 18:34:00 - mmengine - INFO - Epoch(train) [1][1000/1440] lr: 4.0000e-03 eta: 1 day, 6:27:45 time: 3.4610 data_time: 1.6932 memory: 61143 loss: 0.4899 +2023/05/31 18:39:54 - mmengine - INFO - Epoch(train) [1][1100/1440] lr: 4.0000e-03 eta: 1 day, 6:04:05 time: 3.3571 data_time: 1.6030 memory: 61143 loss: 0.4797 +2023/05/31 18:45:49 - mmengine - INFO - Epoch(train) [1][1200/1440] lr: 4.0000e-03 eta: 1 day, 5:43:43 time: 3.6645 data_time: 1.8829 memory: 61143 loss: 0.4687 +2023/05/31 18:52:21 - mmengine - INFO - Epoch(train) [1][1300/1440] lr: 4.0000e-03 eta: 1 day, 5:38:58 time: 3.8143 data_time: 2.0433 memory: 61143 loss: 0.4437 +2023/05/31 18:58:50 - mmengine - INFO - Epoch(train) [1][1400/1440] lr: 4.0000e-03 eta: 1 day, 5:32:39 time: 3.8834 data_time: 2.1126 memory: 61143 loss: 0.4518 +2023/05/31 19:01:19 - mmengine - INFO - Exp name: convnext_tiny_4xb1024_fake5m_20230531_172740 +2023/05/31 19:01:19 - mmengine - INFO - Saving checkpoint at 1 epochs +2023/05/31 19:01:37 - mmengine - INFO - Epoch(val) [1][16/16] accuracy/top1: 54.3337 single-label/precision_classwise: [100.0, 0.0] single-label/recall_classwise: [54.33369445800781, 0.0] single-label/f1-score_classwise: [70.41067504882812, 0.0] data_time: 0.2721 time: 0.6429 +2023/05/31 19:08:07 - mmengine - INFO - Epoch(train) [2][ 100/1440] lr: 3.9754e-03 eta: 1 day, 5:22:19 time: 4.2048 data_time: 2.4369 memory: 61146 loss: 0.4514 +2023/05/31 19:14:33 - mmengine - INFO - Epoch(train) [2][ 200/1440] lr: 3.9754e-03 eta: 1 day, 5:15:25 time: 3.5233 data_time: 1.7614 memory: 61145 loss: 0.4308 +2023/05/31 19:20:27 - mmengine - INFO - Epoch(train) [2][ 300/1440] lr: 3.9754e-03 eta: 1 day, 5:00:07 time: 3.4567 data_time: 1.6979 memory: 61145 loss: 0.4170 +2023/05/31 19:26:28 - mmengine - INFO - Epoch(train) [2][ 400/1440] lr: 3.9754e-03 eta: 1 day, 4:47:31 time: 3.7817 data_time: 1.9897 memory: 61145 loss: 0.4038 +2023/05/31 19:32:26 - mmengine - INFO - Epoch(train) [2][ 500/1440] lr: 3.9754e-03 eta: 1 day, 4:35:04 time: 3.5252 data_time: 1.7586 memory: 61145 loss: 0.3877 +2023/05/31 19:35:57 - mmengine - INFO - Exp name: convnext_tiny_4xb1024_fake5m_20230531_172740 +2023/05/31 19:38:06 - mmengine - INFO - Epoch(train) [2][ 600/1440] lr: 3.9754e-03 eta: 1 day, 4:19:21 time: 3.1745 data_time: 1.4087 memory: 61145 loss: 0.3851 +2023/05/31 19:43:33 - mmengine - INFO - Epoch(train) [2][ 700/1440] lr: 3.9754e-03 eta: 1 day, 4:01:49 time: 3.2064 data_time: 1.4148 memory: 61145 loss: 0.4002 +2023/05/31 19:48:46 - mmengine - INFO - Epoch(train) [2][ 800/1440] lr: 3.9754e-03 eta: 1 day, 3:42:32 time: 2.9946 data_time: 1.2440 memory: 61145 loss: 0.3552 +2023/05/31 19:53:54 - mmengine - INFO - Epoch(train) [2][ 900/1440] lr: 3.9754e-03 eta: 1 day, 3:23:29 time: 3.0410 data_time: 1.2612 memory: 61145 loss: 0.3580 +2023/05/31 19:59:05 - mmengine - INFO - Epoch(train) [2][1000/1440] lr: 3.9754e-03 eta: 1 day, 3:06:07 time: 3.3049 data_time: 1.5412 memory: 61145 loss: 0.3450 +2023/05/31 20:04:25 - mmengine - INFO - Epoch(train) [2][1100/1440] lr: 3.9754e-03 eta: 1 day, 2:51:25 time: 3.1558 data_time: 1.3897 memory: 61145 loss: 0.3331 +2023/05/31 20:10:05 - mmengine - INFO - Epoch(train) [2][1200/1440] lr: 3.9754e-03 eta: 1 day, 2:40:30 time: 3.2056 data_time: 1.4384 memory: 61145 loss: 0.3146 +2023/05/31 20:15:44 - mmengine - INFO - Epoch(train) [2][1300/1440] lr: 3.9754e-03 eta: 1 day, 2:29:56 time: 3.3135 data_time: 1.5567 memory: 61145 loss: 0.3398 +2023/05/31 20:21:24 - mmengine - INFO - Epoch(train) [2][1400/1440] lr: 3.9754e-03 eta: 1 day, 2:19:58 time: 3.4440 data_time: 1.6529 memory: 61145 loss: 0.3053 +2023/05/31 20:23:38 - mmengine - INFO - Exp name: convnext_tiny_4xb1024_fake5m_20230531_172740 +2023/05/31 20:23:38 - mmengine - INFO - Saving checkpoint at 2 epochs +2023/05/31 20:23:53 - mmengine - INFO - Epoch(val) [2][16/16] accuracy/top1: 73.4486 single-label/precision_classwise: [100.0, 0.0] single-label/recall_classwise: [73.44864654541016, 0.0] single-label/f1-score_classwise: [84.69210052490234, 0.0] data_time: 0.2554 time: 0.4853 +2023/05/31 20:29:47 - mmengine - INFO - Epoch(train) [3][ 100/1440] lr: 3.9024e-03 eta: 1 day, 2:07:58 time: 3.5298 data_time: 1.6939 memory: 61145 loss: 0.2797 +2023/05/31 20:30:58 - mmengine - INFO - Exp name: convnext_tiny_4xb1024_fake5m_20230531_172740 +2023/05/31 20:35:32 - mmengine - INFO - Epoch(train) [3][ 200/1440] lr: 3.9024e-03 eta: 1 day, 1:59:13 time: 3.3690 data_time: 1.2517 memory: 61145 loss: 0.2745 +2023/05/31 20:41:09 - mmengine - INFO - Epoch(train) [3][ 300/1440] lr: 3.9024e-03 eta: 1 day, 1:49:37 time: 3.4114 data_time: 1.6347 memory: 61145 loss: 0.2831 +2023/05/31 20:46:45 - mmengine - INFO - Epoch(train) [3][ 400/1440] lr: 3.9024e-03 eta: 1 day, 1:40:02 time: 3.3154 data_time: 1.5658 memory: 61145 loss: 0.2657 +2023/05/31 20:52:20 - mmengine - INFO - Epoch(train) [3][ 500/1440] lr: 3.9024e-03 eta: 1 day, 1:30:40 time: 3.2786 data_time: 1.5092 memory: 61145 loss: 0.2652 +2023/05/31 20:57:55 - mmengine - INFO - Epoch(train) [3][ 600/1440] lr: 3.9024e-03 eta: 1 day, 1:21:24 time: 3.4873 data_time: 1.7051 memory: 61145 loss: 0.2564 +2023/05/31 21:03:24 - mmengine - INFO - Epoch(train) [3][ 700/1440] lr: 3.9024e-03 eta: 1 day, 1:11:43 time: 3.1659 data_time: 1.3825 memory: 61145 loss: 0.2584 +2023/05/31 21:08:54 - mmengine - INFO - Epoch(train) [3][ 800/1440] lr: 3.9024e-03 eta: 1 day, 1:02:22 time: 3.3623 data_time: 1.5842 memory: 61145 loss: 0.2746 +2023/05/31 21:14:22 - mmengine - INFO - Epoch(train) [3][ 900/1440] lr: 3.9024e-03 eta: 1 day, 0:52:57 time: 3.2797 data_time: 1.4913 memory: 61145 loss: 0.4432 +2023/05/31 21:19:51 - mmengine - INFO - Epoch(train) [3][1000/1440] lr: 3.9024e-03 eta: 1 day, 0:43:51 time: 3.3689 data_time: 1.6024 memory: 61145 loss: 0.3017 +2023/05/31 21:25:26 - mmengine - INFO - Epoch(train) [3][1100/1440] lr: 3.9024e-03 eta: 1 day, 0:35:36 time: 3.7910 data_time: 2.0036 memory: 61145 loss: 0.2460 +2023/05/31 21:26:34 - mmengine - INFO - Exp name: convnext_tiny_4xb1024_fake5m_20230531_172740 +2023/05/31 21:31:02 - mmengine - INFO - Epoch(train) [3][1200/1440] lr: 3.9024e-03 eta: 1 day, 0:27:34 time: 3.4165 data_time: 1.6337 memory: 61145 loss: 0.2390 +2023/05/31 21:36:41 - mmengine - INFO - Epoch(train) [3][1300/1440] lr: 3.9024e-03 eta: 1 day, 0:19:59 time: 3.4488 data_time: 1.6870 memory: 61145 loss: 0.2319 +2023/05/31 21:42:17 - mmengine - INFO - Epoch(train) [3][1400/1440] lr: 3.9024e-03 eta: 1 day, 0:12:07 time: 3.4106 data_time: 1.6247 memory: 61145 loss: 0.2377 +2023/05/31 21:44:31 - mmengine - INFO - Exp name: convnext_tiny_4xb1024_fake5m_20230531_172740 +2023/05/31 21:44:31 - mmengine - INFO - Saving checkpoint at 3 epochs +2023/05/31 21:44:46 - mmengine - INFO - Epoch(val) [3][16/16] accuracy/top1: 80.7588 single-label/precision_classwise: [100.0, 0.0] single-label/recall_classwise: [80.7587890625, 0.0] single-label/f1-score_classwise: [89.35530853271484, 0.0] data_time: 0.2534 time: 0.4867 +2023/05/31 21:50:32 - mmengine - INFO - Epoch(train) [4][ 100/1440] lr: 3.7826e-03 eta: 1 day, 0:02:14 time: 3.2469 data_time: 1.4799 memory: 61145 loss: 0.2305 +2023/05/31 21:56:01 - mmengine - INFO - Epoch(train) [4][ 200/1440] lr: 3.7826e-03 eta: 23:53:56 time: 3.2973 data_time: 1.5249 memory: 61145 loss: 0.2314 +2023/05/31 22:01:36 - mmengine - INFO - Epoch(train) [4][ 300/1440] lr: 3.7826e-03 eta: 23:46:24 time: 3.3672 data_time: 1.5883 memory: 61145 loss: 0.2256 +2023/05/31 22:07:14 - mmengine - INFO - Epoch(train) [4][ 400/1440] lr: 3.7826e-03 eta: 23:39:07 time: 3.2423 data_time: 1.4663 memory: 61145 loss: 0.2337 +2023/05/31 22:12:32 - mmengine - INFO - Epoch(train) [4][ 500/1440] lr: 3.7826e-03 eta: 23:30:14 time: 3.1382 data_time: 1.3721 memory: 61145 loss: 0.2306 +2023/05/31 22:17:50 - mmengine - INFO - Epoch(train) [4][ 600/1440] lr: 3.7826e-03 eta: 23:21:36 time: 3.1335 data_time: 1.3680 memory: 61145 loss: 0.2205 +2023/05/31 22:22:07 - mmengine - INFO - Exp name: convnext_tiny_4xb1024_fake5m_20230531_172740 +2023/05/31 22:23:09 - mmengine - INFO - Epoch(train) [4][ 700/1440] lr: 3.7826e-03 eta: 23:13:07 time: 3.1160 data_time: 1.3327 memory: 61145 loss: 0.2213 +2023/05/31 22:28:26 - mmengine - INFO - Epoch(train) [4][ 800/1440] lr: 3.7826e-03 eta: 23:04:34 time: 3.2489 data_time: 1.4970 memory: 61145 loss: 0.2181 +2023/05/31 22:33:44 - mmengine - INFO - Epoch(train) [4][ 900/1440] lr: 3.7826e-03 eta: 22:56:16 time: 3.1870 data_time: 1.4183 memory: 61145 loss: 0.4881 +2023/05/31 22:38:53 - mmengine - INFO - Epoch(train) [4][1000/1440] lr: 3.7826e-03 eta: 22:47:21 time: 3.0847 data_time: 1.3343 memory: 61145 loss: 0.3468 +2023/05/31 22:44:05 - mmengine - INFO - Epoch(train) [4][1100/1440] lr: 3.7826e-03 eta: 22:38:51 time: 3.2474 data_time: 1.4968 memory: 61145 loss: 0.2656 +2023/05/31 22:49:26 - mmengine - INFO - Epoch(train) [4][1200/1440] lr: 3.7826e-03 eta: 22:31:06 time: 3.2647 data_time: 1.4720 memory: 61145 loss: 0.2267 +2023/05/31 22:54:45 - mmengine - INFO - Epoch(train) [4][1300/1440] lr: 3.7826e-03 eta: 22:23:18 time: 3.0902 data_time: 0.0023 memory: 61145 loss: 0.2243 +2023/05/31 22:59:49 - mmengine - INFO - Epoch(train) [4][1400/1440] lr: 3.7826e-03 eta: 22:14:34 time: 3.0753 data_time: 0.7537 memory: 61145 loss: 0.2172 +2023/05/31 23:01:54 - mmengine - INFO - Exp name: convnext_tiny_4xb1024_fake5m_20230531_172740 +2023/05/31 23:01:54 - mmengine - INFO - Saving checkpoint at 4 epochs +2023/05/31 23:02:09 - mmengine - INFO - Epoch(val) [4][16/16] accuracy/top1: 80.5000 single-label/precision_classwise: [100.0, 0.0] single-label/recall_classwise: [80.49996948242188, 0.0] single-label/f1-score_classwise: [89.1966552734375, 0.0] data_time: 0.2560 time: 0.4827 +2023/05/31 23:07:34 - mmengine - INFO - Epoch(train) [5][ 100/1440] lr: 3.6190e-03 eta: 22:04:08 time: 3.0755 data_time: 1.2481 memory: 61145 loss: 0.5187 +2023/05/31 23:12:43 - mmengine - INFO - Epoch(train) [5][ 200/1440] lr: 3.6190e-03 eta: 21:55:57 time: 3.0854 data_time: 0.6788 memory: 61145 loss: 0.4294 +2023/05/31 23:14:50 - mmengine - INFO - Exp name: convnext_tiny_4xb1024_fake5m_20230531_172740 +2023/05/31 23:17:52 - mmengine - INFO - Epoch(train) [5][ 300/1440] lr: 3.6190e-03 eta: 21:47:52 time: 3.0641 data_time: 1.1823 memory: 61145 loss: 0.3695 +2023/05/31 23:23:05 - mmengine - INFO - Epoch(train) [5][ 400/1440] lr: 3.6190e-03 eta: 21:40:09 time: 3.0508 data_time: 1.2857 memory: 61145 loss: 0.2451 +2023/05/31 23:28:21 - mmengine - INFO - Epoch(train) [5][ 500/1440] lr: 3.6190e-03 eta: 21:32:43 time: 3.0749 data_time: 1.2892 memory: 61145 loss: 0.2295 +2023/05/31 23:33:38 - mmengine - INFO - Epoch(train) [5][ 600/1440] lr: 3.6190e-03 eta: 21:25:22 time: 3.0173 data_time: 1.2538 memory: 61145 loss: 0.2202 +2023/05/31 23:38:49 - mmengine - INFO - Epoch(train) [5][ 700/1440] lr: 3.6190e-03 eta: 21:17:48 time: 3.0805 data_time: 1.2854 memory: 61145 loss: 0.6604 +2023/05/31 23:44:02 - mmengine - INFO - Epoch(train) [5][ 800/1440] lr: 3.6190e-03 eta: 21:10:20 time: 3.1783 data_time: 1.4264 memory: 61145 loss: 0.4415 +2023/05/31 23:49:16 - mmengine - INFO - Epoch(train) [5][ 900/1440] lr: 3.6190e-03 eta: 21:03:03 time: 3.1799 data_time: 1.4117 memory: 61145 loss: 0.2370 +2023/05/31 23:54:20 - mmengine - INFO - Epoch(train) [5][1000/1440] lr: 3.6190e-03 eta: 20:55:18 time: 3.0851 data_time: 1.3251 memory: 61145 loss: 0.2167 +2023/05/31 23:59:27 - mmengine - INFO - Epoch(train) [5][1100/1440] lr: 3.6190e-03 eta: 20:47:43 time: 3.1024 data_time: 1.3410 memory: 61145 loss: 0.2216 +2023/06/01 00:04:36 - mmengine - INFO - Epoch(train) [5][1200/1440] lr: 3.6190e-03 eta: 20:40:20 time: 3.0994 data_time: 1.3417 memory: 61145 loss: 0.2179 +2023/06/01 00:06:38 - mmengine - INFO - Exp name: convnext_tiny_4xb1024_fake5m_20230531_172740 +2023/06/01 00:09:41 - mmengine - INFO - Epoch(train) [5][1300/1440] lr: 3.6190e-03 eta: 20:32:51 time: 3.0856 data_time: 1.3213 memory: 61145 loss: 0.2147 +2023/06/01 00:14:47 - mmengine - INFO - Epoch(train) [5][1400/1440] lr: 3.6190e-03 eta: 20:25:28 time: 3.0591 data_time: 1.2700 memory: 61145 loss: 0.2130 +2023/06/01 00:16:47 - mmengine - INFO - Exp name: convnext_tiny_4xb1024_fake5m_20230531_172740 +2023/06/01 00:16:47 - mmengine - INFO - Saving checkpoint at 5 epochs +2023/06/01 00:17:02 - mmengine - INFO - Epoch(val) [5][16/16] accuracy/top1: 73.1961 single-label/precision_classwise: [100.0, 0.0] single-label/recall_classwise: [73.19613647460938, 0.0] single-label/f1-score_classwise: [84.52398681640625, 0.0] data_time: 0.2506 time: 0.4771 +2023/06/01 00:22:23 - mmengine - INFO - Epoch(train) [6][ 100/1440] lr: 3.4157e-03 eta: 20:15:47 time: 3.0956 data_time: 1.3272 memory: 61145 loss: 0.2109 +2023/06/01 00:27:24 - mmengine - INFO - Epoch(train) [6][ 200/1440] lr: 3.4157e-03 eta: 20:08:18 time: 3.0313 data_time: 1.2757 memory: 61145 loss: 0.2103 +2023/06/01 00:32:30 - mmengine - INFO - Epoch(train) [6][ 300/1440] lr: 3.4157e-03 eta: 20:01:06 time: 3.0531 data_time: 1.2686 memory: 61145 loss: 0.2119 +2023/06/01 00:37:33 - mmengine - INFO - Epoch(train) [6][ 400/1440] lr: 3.4157e-03 eta: 19:53:49 time: 3.1303 data_time: 1.3811 memory: 61145 loss: 0.2154 +2023/06/01 00:42:36 - mmengine - INFO - Epoch(train) [6][ 500/1440] lr: 3.4157e-03 eta: 19:46:36 time: 2.9587 data_time: 1.1833 memory: 61145 loss: 0.2110 +2023/06/01 00:47:46 - mmengine - INFO - Epoch(train) [6][ 600/1440] lr: 3.4157e-03 eta: 19:39:43 time: 3.1617 data_time: 1.4080 memory: 61145 loss: 0.2335 +2023/06/01 00:52:56 - mmengine - INFO - Epoch(train) [6][ 700/1440] lr: 3.4157e-03 eta: 19:32:56 time: 3.1356 data_time: 1.3428 memory: 61145 loss: 0.2092 +2023/06/01 00:58:02 - mmengine - INFO - Exp name: convnext_tiny_4xb1024_fake5m_20230531_172740 +2023/06/01 00:58:02 - mmengine - INFO - Epoch(train) [6][ 800/1440] lr: 3.4157e-03 eta: 19:26:00 time: 3.1164 data_time: 1.3544 memory: 61145 loss: 0.2123 +2023/06/01 01:03:04 - mmengine - INFO - Epoch(train) [6][ 900/1440] lr: 3.4157e-03 eta: 19:18:56 time: 3.0944 data_time: 1.3150 memory: 61145 loss: 0.2122 +2023/06/01 01:08:42 - mmengine - INFO - Epoch(train) [6][1000/1440] lr: 3.4157e-03 eta: 19:13:25 time: 6.3566 data_time: 1.2379 memory: 61145 loss: 0.2294 +2023/06/01 01:13:37 - mmengine - INFO - Epoch(train) [6][1100/1440] lr: 3.4157e-03 eta: 19:06:07 time: 2.9337 data_time: 0.0022 memory: 61145 loss: 0.5894 +2023/06/01 01:18:28 - mmengine - INFO - Epoch(train) [6][1200/1440] lr: 3.4157e-03 eta: 18:58:45 time: 3.0414 data_time: 0.4232 memory: 61145 loss: 0.5871 +2023/06/01 01:23:33 - mmengine - INFO - Epoch(train) [6][1300/1440] lr: 3.4157e-03 eta: 18:51:57 time: 2.9096 data_time: 1.1399 memory: 61145 loss: 0.5092 +2023/06/01 01:28:35 - mmengine - INFO - Epoch(train) [6][1400/1440] lr: 3.4157e-03 eta: 18:45:07 time: 3.0509 data_time: 1.2904 memory: 61145 loss: 0.4725 +2023/06/01 01:30:35 - mmengine - INFO - Exp name: convnext_tiny_4xb1024_fake5m_20230531_172740 +2023/06/01 01:30:35 - mmengine - INFO - Saving checkpoint at 6 epochs +2023/06/01 01:30:51 - mmengine - INFO - Epoch(val) [6][16/16] accuracy/top1: 78.4168 single-label/precision_classwise: [100.0, 0.0] single-label/recall_classwise: [78.41676330566406, 0.0] single-label/f1-score_classwise: [87.90290832519531, 0.0] data_time: 0.2397 time: 0.4670 +2023/06/01 01:36:14 - mmengine - INFO - Epoch(train) [7][ 100/1440] lr: 3.1776e-03 eta: 18:36:22 time: 3.0124 data_time: 1.2088 memory: 61145 loss: 0.4403 +2023/06/01 01:41:16 - mmengine - INFO - Epoch(train) [7][ 200/1440] lr: 3.1776e-03 eta: 18:29:38 time: 3.1734 data_time: 1.3573 memory: 61145 loss: 0.3954 +2023/06/01 01:46:21 - mmengine - INFO - Epoch(train) [7][ 300/1440] lr: 3.1776e-03 eta: 18:23:00 time: 2.9848 data_time: 1.1978 memory: 61145 loss: 0.2557 +2023/06/01 01:49:28 - mmengine - INFO - Exp name: convnext_tiny_4xb1024_fake5m_20230531_172740 +2023/06/01 01:51:33 - mmengine - INFO - Epoch(train) [7][ 400/1440] lr: 3.1776e-03 eta: 18:16:40 time: 3.1612 data_time: 1.3735 memory: 61145 loss: 0.2293 +2023/06/01 01:56:41 - mmengine - INFO - Epoch(train) [7][ 500/1440] lr: 3.1776e-03 eta: 18:10:13 time: 3.1020 data_time: 1.3456 memory: 61145 loss: 0.2181 +2023/06/01 02:01:53 - mmengine - INFO - Epoch(train) [7][ 600/1440] lr: 3.1776e-03 eta: 18:03:56 time: 3.0814 data_time: 1.2816 memory: 61145 loss: 0.2119 +2023/06/01 02:07:02 - mmengine - INFO - Epoch(train) [7][ 700/1440] lr: 3.1776e-03 eta: 17:57:34 time: 3.1496 data_time: 1.3925 memory: 61145 loss: 0.2117 +2023/06/01 02:12:13 - mmengine - INFO - Epoch(train) [7][ 800/1440] lr: 3.1776e-03 eta: 17:51:20 time: 3.0924 data_time: 1.2872 memory: 61145 loss: 0.2100 +2023/06/01 02:17:23 - mmengine - INFO - Epoch(train) [7][ 900/1440] lr: 3.1776e-03 eta: 17:45:02 time: 3.0860 data_time: 1.3338 memory: 61145 loss: 0.2105 +2023/06/01 02:22:36 - mmengine - INFO - Epoch(train) [7][1000/1440] lr: 3.1776e-03 eta: 17:38:53 time: 3.0671 data_time: 1.2958 memory: 61145 loss: 0.2075 +2023/06/01 02:27:49 - mmengine - INFO - Epoch(train) [7][1100/1440] lr: 3.1776e-03 eta: 17:32:46 time: 3.0555 data_time: 1.2907 memory: 61145 loss: 0.2066 +2023/06/01 02:33:05 - mmengine - INFO - Epoch(train) [7][1200/1440] lr: 3.1776e-03 eta: 17:26:44 time: 3.1656 data_time: 1.3843 memory: 61145 loss: 0.2134 +2023/06/01 02:38:18 - mmengine - INFO - Epoch(train) [7][1300/1440] lr: 3.1776e-03 eta: 17:20:38 time: 3.1757 data_time: 1.4102 memory: 61145 loss: 0.6153 +2023/06/01 02:41:28 - mmengine - INFO - Exp name: convnext_tiny_4xb1024_fake5m_20230531_172740 +2023/06/01 02:43:34 - mmengine - INFO - Epoch(train) [7][1400/1440] lr: 3.1776e-03 eta: 17:14:39 time: 3.1776 data_time: 1.4122 memory: 61145 loss: 0.5245 +2023/06/01 02:45:40 - mmengine - INFO - Exp name: convnext_tiny_4xb1024_fake5m_20230531_172740 +2023/06/01 02:45:40 - mmengine - INFO - Saving checkpoint at 7 epochs +2023/06/01 02:45:55 - mmengine - INFO - Epoch(val) [7][16/16] accuracy/top1: 93.3337 single-label/precision_classwise: [100.0, 0.0] single-label/recall_classwise: [93.33375549316406, 0.0] single-label/f1-score_classwise: [96.55195617675781, 0.0] data_time: 0.2456 time: 0.4723 +2023/06/01 02:51:22 - mmengine - INFO - Epoch(train) [8][ 100/1440] lr: 2.9107e-03 eta: 17:06:36 time: 3.0902 data_time: 0.6951 memory: 61145 loss: 0.4658 +2023/06/01 02:56:35 - mmengine - INFO - Epoch(train) [8][ 200/1440] lr: 2.9107e-03 eta: 17:00:33 time: 3.2131 data_time: 0.8308 memory: 61145 loss: 0.4396 +2023/06/01 03:01:49 - mmengine - INFO - Epoch(train) [8][ 300/1440] lr: 2.9107e-03 eta: 16:54:32 time: 3.1430 data_time: 1.0770 memory: 61145 loss: 0.4115 +2023/06/01 03:06:58 - mmengine - INFO - Epoch(train) [8][ 400/1440] lr: 2.9107e-03 eta: 16:48:25 time: 3.0551 data_time: 1.2908 memory: 61145 loss: 0.4118 +2023/06/01 03:12:10 - mmengine - INFO - Epoch(train) [8][ 500/1440] lr: 2.9107e-03 eta: 16:42:23 time: 3.0869 data_time: 1.2974 memory: 61145 loss: 0.3912 +2023/06/01 03:17:24 - mmengine - INFO - Epoch(train) [8][ 600/1440] lr: 2.9107e-03 eta: 16:36:25 time: 3.0950 data_time: 1.2923 memory: 61145 loss: 0.3675 +2023/06/01 03:22:38 - mmengine - INFO - Epoch(train) [8][ 700/1440] lr: 2.9107e-03 eta: 16:30:29 time: 3.0924 data_time: 1.3209 memory: 61145 loss: 0.3533 +2023/06/01 03:27:51 - mmengine - INFO - Epoch(train) [8][ 800/1440] lr: 2.9107e-03 eta: 16:24:32 time: 3.0916 data_time: 1.3250 memory: 61145 loss: 0.3378 +2023/06/01 03:33:05 - mmengine - INFO - Epoch(train) [8][ 900/1440] lr: 2.9107e-03 eta: 16:18:36 time: 3.1861 data_time: 1.3871 memory: 61145 loss: 0.3249 +2023/06/01 03:34:07 - mmengine - INFO - Exp name: convnext_tiny_4xb1024_fake5m_20230531_172740 +2023/06/01 03:38:20 - mmengine - INFO - Epoch(train) [8][1000/1440] lr: 2.9107e-03 eta: 16:12:44 time: 3.1566 data_time: 1.3787 memory: 61145 loss: 0.3034 +2023/06/01 03:43:32 - mmengine - INFO - Epoch(train) [8][1100/1440] lr: 2.9107e-03 eta: 16:06:48 time: 3.1391 data_time: 1.3681 memory: 61145 loss: 0.3182 +2023/06/01 03:48:42 - mmengine - INFO - Epoch(train) [8][1200/1440] lr: 2.9107e-03 eta: 16:00:49 time: 3.1060 data_time: 1.3256 memory: 61145 loss: 0.2937 +2023/06/01 03:53:57 - mmengine - INFO - Epoch(train) [8][1300/1440] lr: 2.9107e-03 eta: 15:54:57 time: 3.0547 data_time: 1.2850 memory: 61145 loss: 0.2927 +2023/06/01 03:59:09 - mmengine - INFO - Epoch(train) [8][1400/1440] lr: 2.9107e-03 eta: 15:49:03 time: 3.2542 data_time: 1.4537 memory: 61145 loss: 0.2789 +2023/06/01 04:01:01 - mmengine - INFO - Exp name: convnext_tiny_4xb1024_fake5m_20230531_172740 +2023/06/01 04:01:01 - mmengine - INFO - Saving checkpoint at 8 epochs +2023/06/01 04:01:16 - mmengine - INFO - Epoch(val) [8][16/16] accuracy/top1: 97.1340 single-label/precision_classwise: [100.0, 0.0] single-label/recall_classwise: [97.13401794433594, 0.0] single-label/f1-score_classwise: [98.54617309570312, 0.0] data_time: 0.2475 time: 0.4747 +2023/06/01 04:06:45 - mmengine - INFO - Epoch(train) [9][ 100/1440] lr: 2.6215e-03 eta: 15:40:55 time: 3.1850 data_time: 0.8897 memory: 61145 loss: 0.2795 +2023/06/01 04:11:53 - mmengine - INFO - Epoch(train) [9][ 200/1440] lr: 2.6215e-03 eta: 15:34:56 time: 3.1250 data_time: 1.1256 memory: 61145 loss: 0.2640 +2023/06/01 04:17:03 - mmengine - INFO - Epoch(train) [9][ 300/1440] lr: 2.6215e-03 eta: 15:29:00 time: 3.1158 data_time: 1.3541 memory: 61145 loss: 0.2597 +2023/06/01 04:22:16 - mmengine - INFO - Epoch(train) [9][ 400/1440] lr: 2.6215e-03 eta: 15:23:10 time: 3.1486 data_time: 1.3617 memory: 61145 loss: 0.2476 +2023/06/01 04:26:25 - mmengine - INFO - Exp name: convnext_tiny_4xb1024_fake5m_20230531_172740 +2023/06/01 04:27:27 - mmengine - INFO - Epoch(train) [9][ 500/1440] lr: 2.6215e-03 eta: 15:17:19 time: 3.0533 data_time: 1.2851 memory: 61145 loss: 0.2372 +2023/06/01 04:32:40 - mmengine - INFO - Epoch(train) [9][ 600/1440] lr: 2.6215e-03 eta: 15:11:30 time: 3.1301 data_time: 1.3581 memory: 61145 loss: 0.2365 +2023/06/01 04:37:50 - mmengine - INFO - Epoch(train) [9][ 700/1440] lr: 2.6215e-03 eta: 15:05:38 time: 3.0736 data_time: 1.2984 memory: 61145 loss: 0.2316 +2023/06/01 04:43:05 - mmengine - INFO - Epoch(train) [9][ 800/1440] lr: 2.6215e-03 eta: 14:59:54 time: 3.0875 data_time: 1.3187 memory: 61145 loss: 0.2319 +2023/06/01 04:48:16 - mmengine - INFO - Epoch(train) [9][ 900/1440] lr: 2.6215e-03 eta: 14:54:03 time: 3.1726 data_time: 1.3965 memory: 61145 loss: 0.2304 +2023/06/01 04:53:29 - mmengine - INFO - Epoch(train) [9][1000/1440] lr: 2.6215e-03 eta: 14:48:17 time: 3.0857 data_time: 1.3124 memory: 61145 loss: 0.2226 +2023/06/01 04:58:40 - mmengine - INFO - Epoch(train) [9][1100/1440] lr: 2.6215e-03 eta: 14:42:29 time: 3.0722 data_time: 1.2785 memory: 61145 loss: 0.2280 +2023/06/01 05:03:53 - mmengine - INFO - Epoch(train) [9][1200/1440] lr: 2.6215e-03 eta: 14:36:44 time: 3.0199 data_time: 1.2362 memory: 61145 loss: 0.2230 +2023/06/01 05:09:07 - mmengine - INFO - Epoch(train) [9][1300/1440] lr: 2.6215e-03 eta: 14:31:00 time: 3.1816 data_time: 1.4119 memory: 61145 loss: 0.2218 +2023/06/01 05:14:23 - mmengine - INFO - Epoch(train) [9][1400/1440] lr: 2.6215e-03 eta: 14:25:20 time: 3.1760 data_time: 1.3612 memory: 61145 loss: 0.2184 +2023/06/01 05:16:24 - mmengine - INFO - Exp name: convnext_tiny_4xb1024_fake5m_20230531_172740 +2023/06/01 05:16:24 - mmengine - INFO - Saving checkpoint at 9 epochs +2023/06/01 05:16:39 - mmengine - INFO - Epoch(val) [9][16/16] accuracy/top1: 98.0178 single-label/precision_classwise: [100.0, 0.0] single-label/recall_classwise: [98.0177993774414, 0.0] single-label/f1-score_classwise: [98.99898529052734, 0.0] data_time: 0.2387 time: 0.4705 +2023/06/01 05:19:00 - mmengine - INFO - Exp name: convnext_tiny_4xb1024_fake5m_20230531_172740 +2023/06/01 05:22:04 - mmengine - INFO - Epoch(train) [10][ 100/1440] lr: 2.3171e-03 eta: 14:17:28 time: 3.1092 data_time: 1.1508 memory: 61145 loss: 0.2145 +2023/06/01 05:27:14 - mmengine - INFO - Epoch(train) [10][ 200/1440] lr: 2.3171e-03 eta: 14:11:41 time: 3.0942 data_time: 1.3323 memory: 61145 loss: 0.2162 +2023/06/01 05:32:24 - mmengine - INFO - Epoch(train) [10][ 300/1440] lr: 2.3171e-03 eta: 14:05:55 time: 3.0775 data_time: 1.2937 memory: 61145 loss: 0.2160 +2023/06/01 05:37:36 - mmengine - INFO - Epoch(train) [10][ 400/1440] lr: 2.3171e-03 eta: 14:00:11 time: 3.1045 data_time: 1.3234 memory: 61145 loss: 0.2188 +2023/06/01 05:42:45 - mmengine - INFO - Epoch(train) [10][ 500/1440] lr: 2.3171e-03 eta: 13:54:24 time: 3.1302 data_time: 1.3379 memory: 61145 loss: 0.2421 +2023/06/01 05:47:54 - mmengine - INFO - Epoch(train) [10][ 600/1440] lr: 2.3171e-03 eta: 13:48:39 time: 3.0870 data_time: 1.3331 memory: 61145 loss: 0.2091 +2023/06/01 05:53:07 - mmengine - INFO - Epoch(train) [10][ 700/1440] lr: 2.3171e-03 eta: 13:42:58 time: 3.1577 data_time: 1.3870 memory: 61145 loss: 0.2137 +2023/06/01 05:58:22 - mmengine - INFO - Epoch(train) [10][ 800/1440] lr: 2.3171e-03 eta: 13:37:19 time: 3.2528 data_time: 1.4971 memory: 61145 loss: 0.2121 +2023/06/01 06:03:34 - mmengine - INFO - Epoch(train) [10][ 900/1440] lr: 2.3171e-03 eta: 13:31:38 time: 3.1807 data_time: 1.4304 memory: 61145 loss: 0.2086 +2023/06/01 06:08:49 - mmengine - INFO - Epoch(train) [10][1000/1440] lr: 2.3171e-03 eta: 13:26:00 time: 3.1125 data_time: 1.3377 memory: 61145 loss: 0.2134 +2023/06/01 06:10:54 - mmengine - INFO - Exp name: convnext_tiny_4xb1024_fake5m_20230531_172740 +2023/06/01 06:14:02 - mmengine - INFO - Epoch(train) [10][1100/1440] lr: 2.3171e-03 eta: 13:20:21 time: 3.2460 data_time: 1.4528 memory: 61145 loss: 0.2106 +2023/06/01 06:19:13 - mmengine - INFO - Epoch(train) [10][1200/1440] lr: 2.3171e-03 eta: 13:14:40 time: 3.1990 data_time: 1.4445 memory: 61145 loss: 0.2082 +2023/06/01 06:24:28 - mmengine - INFO - Epoch(train) [10][1300/1440] lr: 2.3171e-03 eta: 13:09:04 time: 3.1214 data_time: 1.3376 memory: 61145 loss: 0.4386 +2023/06/01 06:29:47 - mmengine - INFO - Epoch(train) [10][1400/1440] lr: 2.3171e-03 eta: 13:03:31 time: 3.1819 data_time: 1.3947 memory: 61145 loss: 0.2292 +2023/06/01 06:31:52 - mmengine - INFO - Exp name: convnext_tiny_4xb1024_fake5m_20230531_172740 +2023/06/01 06:31:52 - mmengine - INFO - Saving checkpoint at 10 epochs +2023/06/01 06:32:07 - mmengine - INFO - Epoch(val) [10][16/16] accuracy/top1: 96.9383 single-label/precision_classwise: [100.0, 0.0] single-label/recall_classwise: [96.93832397460938, 0.0] single-label/f1-score_classwise: [98.44535827636719, 0.0] data_time: 0.2335 time: 0.4679 +2023/06/01 06:37:32 - mmengine - INFO - Epoch(train) [11][ 100/1440] lr: 2.0050e-03 eta: 12:55:50 time: 3.0897 data_time: 1.0506 memory: 61145 loss: 0.2140 +2023/06/01 06:42:44 - mmengine - INFO - Epoch(train) [11][ 200/1440] lr: 2.0050e-03 eta: 12:50:11 time: 3.1347 data_time: 1.0325 memory: 61145 loss: 0.2098 +2023/06/01 06:48:00 - mmengine - INFO - Epoch(train) [11][ 300/1440] lr: 2.0050e-03 eta: 12:44:36 time: 3.1370 data_time: 1.3851 memory: 61145 loss: 0.2154 +2023/06/01 06:53:17 - mmengine - INFO - Epoch(train) [11][ 400/1440] lr: 2.0050e-03 eta: 12:39:03 time: 3.0547 data_time: 1.2575 memory: 61145 loss: 0.2089 +2023/06/01 06:58:31 - mmengine - INFO - Epoch(train) [11][ 500/1440] lr: 2.0050e-03 eta: 12:33:27 time: 3.0892 data_time: 1.3201 memory: 61145 loss: 0.2102 +2023/06/01 07:03:43 - mmengine - INFO - Exp name: convnext_tiny_4xb1024_fake5m_20230531_172740 +2023/06/01 07:03:43 - mmengine - INFO - Epoch(train) [11][ 600/1440] lr: 2.0050e-03 eta: 12:27:50 time: 3.1587 data_time: 1.3789 memory: 61145 loss: 0.2074 +2023/06/01 07:08:53 - mmengine - INFO - Epoch(train) [11][ 700/1440] lr: 2.0050e-03 eta: 12:22:11 time: 3.0238 data_time: 1.2731 memory: 61145 loss: 0.2058 +2023/06/01 07:14:09 - mmengine - INFO - Epoch(train) [11][ 800/1440] lr: 2.0050e-03 eta: 12:16:37 time: 3.2337 data_time: 1.4580 memory: 61145 loss: 0.2100 +2023/06/01 07:19:19 - mmengine - INFO - Epoch(train) [11][ 900/1440] lr: 2.0050e-03 eta: 12:10:59 time: 3.0961 data_time: 1.3105 memory: 61145 loss: 0.2043 +2023/06/01 07:24:33 - mmengine - INFO - Epoch(train) [11][1000/1440] lr: 2.0050e-03 eta: 12:05:25 time: 3.1072 data_time: 1.3269 memory: 61145 loss: 0.2049 +2023/06/01 07:29:46 - mmengine - INFO - Epoch(train) [11][1100/1440] lr: 2.0050e-03 eta: 11:59:50 time: 3.1478 data_time: 1.3773 memory: 61145 loss: 0.2150 +2023/06/01 07:35:00 - mmengine - INFO - Epoch(train) [11][1200/1440] lr: 2.0050e-03 eta: 11:54:16 time: 3.2485 data_time: 1.4597 memory: 61145 loss: 0.2066 +2023/06/01 07:40:12 - mmengine - INFO - Epoch(train) [11][1300/1440] lr: 2.0050e-03 eta: 11:48:41 time: 3.1004 data_time: 1.3403 memory: 61145 loss: 0.2052 +2023/06/01 07:45:29 - mmengine - INFO - Epoch(train) [11][1400/1440] lr: 2.0050e-03 eta: 11:43:10 time: 3.1783 data_time: 1.3930 memory: 61145 loss: 0.3890 +2023/06/01 07:47:37 - mmengine - INFO - Exp name: convnext_tiny_4xb1024_fake5m_20230531_172740 +2023/06/01 07:47:37 - mmengine - INFO - Saving checkpoint at 11 epochs +2023/06/01 07:47:52 - mmengine - INFO - Epoch(val) [11][16/16] accuracy/top1: 92.3932 single-label/precision_classwise: [100.0, 0.0] single-label/recall_classwise: [92.39315795898438, 0.0] single-label/f1-score_classwise: [96.04619598388672, 0.0] data_time: 0.2547 time: 0.4821 +2023/06/01 07:53:17 - mmengine - INFO - Epoch(train) [12][ 100/1440] lr: 1.6929e-03 eta: 11:35:34 time: 3.1339 data_time: 1.1678 memory: 61145 loss: 0.2115 +2023/06/01 07:56:21 - mmengine - INFO - Exp name: convnext_tiny_4xb1024_fake5m_20230531_172740 +2023/06/01 07:58:27 - mmengine - INFO - Epoch(train) [12][ 200/1440] lr: 1.6929e-03 eta: 11:29:58 time: 3.1980 data_time: 1.4426 memory: 61145 loss: 0.2072 +2023/06/01 08:03:48 - mmengine - INFO - Epoch(train) [12][ 300/1440] lr: 1.6929e-03 eta: 11:24:31 time: 3.2064 data_time: 1.4528 memory: 61145 loss: 0.2081 +2023/06/01 08:09:09 - mmengine - INFO - Epoch(train) [12][ 400/1440] lr: 1.6929e-03 eta: 11:19:04 time: 3.2709 data_time: 1.4935 memory: 61145 loss: 0.2063 +2023/06/01 08:14:31 - mmengine - INFO - Epoch(train) [12][ 500/1440] lr: 1.6929e-03 eta: 11:13:38 time: 3.2489 data_time: 1.4901 memory: 61145 loss: 0.2068 +2023/06/01 08:19:54 - mmengine - INFO - Epoch(train) [12][ 600/1440] lr: 1.6929e-03 eta: 11:08:13 time: 3.0888 data_time: 1.3309 memory: 61145 loss: 0.2061 +2023/06/01 08:25:19 - mmengine - INFO - Epoch(train) [12][ 700/1440] lr: 1.6929e-03 eta: 11:02:49 time: 3.2565 data_time: 1.4864 memory: 61145 loss: 0.2067 +2023/06/01 08:30:38 - mmengine - INFO - Epoch(train) [12][ 800/1440] lr: 1.6929e-03 eta: 10:57:20 time: 3.1482 data_time: 1.3911 memory: 61145 loss: 0.2045 +2023/06/01 08:36:03 - mmengine - INFO - Epoch(train) [12][ 900/1440] lr: 1.6929e-03 eta: 10:51:56 time: 3.3144 data_time: 1.5198 memory: 61145 loss: 0.2034 +2023/06/01 08:41:28 - mmengine - INFO - Epoch(train) [12][1000/1440] lr: 1.6929e-03 eta: 10:46:33 time: 3.2255 data_time: 1.4723 memory: 61145 loss: 0.2042 +2023/06/01 08:46:53 - mmengine - INFO - Epoch(train) [12][1100/1440] lr: 1.6929e-03 eta: 10:41:09 time: 3.2166 data_time: 1.4404 memory: 61145 loss: 0.2078 +2023/06/01 08:50:08 - mmengine - INFO - Exp name: convnext_tiny_4xb1024_fake5m_20230531_172740 +2023/06/01 08:52:18 - mmengine - INFO - Epoch(train) [12][1200/1440] lr: 1.6929e-03 eta: 10:35:44 time: 3.2394 data_time: 1.4543 memory: 61145 loss: 0.2035 +2023/06/01 08:57:43 - mmengine - INFO - Epoch(train) [12][1300/1440] lr: 1.6929e-03 eta: 10:30:20 time: 3.2678 data_time: 1.4798 memory: 61145 loss: 0.2025 +2023/06/01 09:03:08 - mmengine - INFO - Epoch(train) [12][1400/1440] lr: 1.6929e-03 eta: 10:24:57 time: 3.2344 data_time: 1.4094 memory: 61145 loss: 0.2038 +2023/06/01 09:05:15 - mmengine - INFO - Exp name: convnext_tiny_4xb1024_fake5m_20230531_172740 +2023/06/01 09:05:15 - mmengine - INFO - Saving checkpoint at 12 epochs +2023/06/01 09:05:31 - mmengine - INFO - Epoch(val) [12][16/16] accuracy/top1: 75.1720 single-label/precision_classwise: [100.0, 0.0] single-label/recall_classwise: [75.1720199584961, 0.0] single-label/f1-score_classwise: [85.8265151977539, 0.0] data_time: 0.2597 time: 0.4906 +2023/06/01 09:11:14 - mmengine - INFO - Epoch(train) [13][ 100/1440] lr: 1.3885e-03 eta: 10:17:33 time: 3.0909 data_time: 0.8689 memory: 61145 loss: 0.2030 +2023/06/01 09:17:09 - mmengine - INFO - Epoch(train) [13][ 200/1440] lr: 1.3885e-03 eta: 10:12:29 time: 3.2317 data_time: 0.5883 memory: 61145 loss: 0.2036 +2023/06/01 09:22:35 - mmengine - INFO - Epoch(train) [13][ 300/1440] lr: 1.3885e-03 eta: 10:07:05 time: 3.1851 data_time: 0.5509 memory: 61145 loss: 0.2052 +2023/06/01 09:27:58 - mmengine - INFO - Epoch(train) [13][ 400/1440] lr: 1.3885e-03 eta: 10:01:39 time: 3.2718 data_time: 1.3912 memory: 61145 loss: 0.2074 +2023/06/01 09:33:37 - mmengine - INFO - Epoch(train) [13][ 500/1440] lr: 1.3885e-03 eta: 9:56:23 time: 4.0069 data_time: 2.2406 memory: 61145 loss: 0.2060 +2023/06/01 09:39:01 - mmengine - INFO - Epoch(train) [13][ 600/1440] lr: 1.3885e-03 eta: 9:50:58 time: 3.2260 data_time: 1.4577 memory: 61145 loss: 0.2040 +2023/06/01 09:44:23 - mmengine - INFO - Epoch(train) [13][ 700/1440] lr: 1.3885e-03 eta: 9:45:32 time: 3.1958 data_time: 1.4013 memory: 61145 loss: 0.2033 +2023/06/01 09:45:27 - mmengine - INFO - Exp name: convnext_tiny_4xb1024_fake5m_20230531_172740 +2023/06/01 09:49:46 - mmengine - INFO - Epoch(train) [13][ 800/1440] lr: 1.3885e-03 eta: 9:40:06 time: 3.2272 data_time: 1.4415 memory: 61145 loss: 0.2040 +2023/06/01 09:55:08 - mmengine - INFO - Epoch(train) [13][ 900/1440] lr: 1.3885e-03 eta: 9:34:40 time: 3.1846 data_time: 1.4219 memory: 61145 loss: 0.2033 +2023/06/01 10:00:31 - mmengine - INFO - Epoch(train) [13][1000/1440] lr: 1.3885e-03 eta: 9:29:14 time: 3.2232 data_time: 1.4633 memory: 61145 loss: 0.2035 +2023/06/01 10:06:01 - mmengine - INFO - Epoch(train) [13][1100/1440] lr: 1.3885e-03 eta: 9:23:53 time: 3.2273 data_time: 1.4749 memory: 61145 loss: 0.2013 +2023/06/01 10:11:27 - mmengine - INFO - Epoch(train) [13][1200/1440] lr: 1.3885e-03 eta: 9:18:29 time: 3.2653 data_time: 1.4675 memory: 61145 loss: 0.2028 +2023/06/01 10:16:54 - mmengine - INFO - Epoch(train) [13][1300/1440] lr: 1.3885e-03 eta: 9:13:05 time: 3.2732 data_time: 1.5046 memory: 61145 loss: 0.2019 +2023/06/01 10:22:21 - mmengine - INFO - Epoch(train) [13][1400/1440] lr: 1.3885e-03 eta: 9:07:42 time: 3.1675 data_time: 1.3992 memory: 61145 loss: 0.2039 +2023/06/01 10:24:31 - mmengine - INFO - Exp name: convnext_tiny_4xb1024_fake5m_20230531_172740 +2023/06/01 10:24:31 - mmengine - INFO - Saving checkpoint at 13 epochs +2023/06/01 10:24:45 - mmengine - INFO - Epoch(val) [13][16/16] accuracy/top1: 61.7701 single-label/precision_classwise: [100.0, 0.0] single-label/recall_classwise: [61.77008819580078, 0.0] single-label/f1-score_classwise: [76.36775207519531, 0.0] data_time: 0.2610 time: 0.4888 +2023/06/01 10:30:24 - mmengine - INFO - Epoch(train) [14][ 100/1440] lr: 1.0993e-03 eta: 9:00:14 time: 3.1806 data_time: 1.1026 memory: 61145 loss: 0.2025 +2023/06/01 10:35:49 - mmengine - INFO - Epoch(train) [14][ 200/1440] lr: 1.0993e-03 eta: 8:54:50 time: 3.2537 data_time: 1.4801 memory: 61145 loss: 0.2027 +2023/06/01 10:40:12 - mmengine - INFO - Exp name: convnext_tiny_4xb1024_fake5m_20230531_172740 +2023/06/01 10:41:18 - mmengine - INFO - Epoch(train) [14][ 300/1440] lr: 1.0993e-03 eta: 8:49:27 time: 3.3251 data_time: 1.5635 memory: 61145 loss: 0.2014 +2023/06/01 10:46:43 - mmengine - INFO - Epoch(train) [14][ 400/1440] lr: 1.0993e-03 eta: 8:44:02 time: 3.2223 data_time: 1.4504 memory: 61145 loss: 0.2016 +2023/06/01 10:52:14 - mmengine - INFO - Epoch(train) [14][ 500/1440] lr: 1.0993e-03 eta: 8:38:41 time: 3.3354 data_time: 1.5506 memory: 61145 loss: 0.2016 +2023/06/01 10:57:46 - mmengine - INFO - Epoch(train) [14][ 600/1440] lr: 1.0993e-03 eta: 8:33:19 time: 3.3589 data_time: 1.5880 memory: 61145 loss: 0.2029 +2023/06/01 11:03:16 - mmengine - INFO - Epoch(train) [14][ 700/1440] lr: 1.0993e-03 eta: 8:27:57 time: 3.2776 data_time: 1.4978 memory: 61145 loss: 0.2031 +2023/06/01 11:08:45 - mmengine - INFO - Epoch(train) [14][ 800/1440] lr: 1.0993e-03 eta: 8:22:34 time: 3.3003 data_time: 1.5182 memory: 61145 loss: 0.2038 +2023/06/01 11:14:23 - mmengine - INFO - Epoch(train) [14][ 900/1440] lr: 1.0993e-03 eta: 8:17:15 time: 3.4478 data_time: 1.6688 memory: 61145 loss: 0.2018 +2023/06/01 11:20:01 - mmengine - INFO - Epoch(train) [14][1000/1440] lr: 1.0993e-03 eta: 8:11:56 time: 3.2508 data_time: 1.4506 memory: 61145 loss: 0.2023 +2023/06/01 11:25:40 - mmengine - INFO - Epoch(train) [14][1100/1440] lr: 1.0993e-03 eta: 8:06:37 time: 3.3529 data_time: 1.5441 memory: 61145 loss: 0.2021 +2023/06/01 11:31:10 - mmengine - INFO - Epoch(train) [14][1200/1440] lr: 1.0993e-03 eta: 8:01:14 time: 3.4700 data_time: 1.6898 memory: 61145 loss: 0.2015 +2023/06/01 11:35:40 - mmengine - INFO - Exp name: convnext_tiny_4xb1024_fake5m_20230531_172740 +2023/06/01 11:36:48 - mmengine - INFO - Epoch(train) [14][1300/1440] lr: 1.0993e-03 eta: 7:55:55 time: 3.4183 data_time: 1.6409 memory: 61145 loss: 0.5079 +2023/06/01 11:42:26 - mmengine - INFO - Epoch(train) [14][1400/1440] lr: 1.0993e-03 eta: 7:50:35 time: 3.3625 data_time: 1.5968 memory: 61145 loss: 0.2150 +2023/06/01 11:44:34 - mmengine - INFO - Exp name: convnext_tiny_4xb1024_fake5m_20230531_172740 +2023/06/01 11:44:34 - mmengine - INFO - Saving checkpoint at 14 epochs +2023/06/01 11:44:49 - mmengine - INFO - Epoch(val) [14][16/16] accuracy/top1: 73.1014 single-label/precision_classwise: [100.0, 0.0] single-label/recall_classwise: [73.1014404296875, 0.0] single-label/f1-score_classwise: [84.46080780029297, 0.0] data_time: 0.2505 time: 0.4782 +2023/06/01 11:50:20 - mmengine - INFO - Epoch(train) [15][ 100/1440] lr: 8.3237e-04 eta: 7:43:01 time: 3.2008 data_time: 1.4396 memory: 61145 loss: 0.2063 +2023/06/01 11:55:35 - mmengine - INFO - Epoch(train) [15][ 200/1440] lr: 8.3237e-04 eta: 7:37:32 time: 3.1488 data_time: 1.3790 memory: 61145 loss: 0.2038 +2023/06/01 12:00:51 - mmengine - INFO - Epoch(train) [15][ 300/1440] lr: 8.3237e-04 eta: 7:32:02 time: 3.1446 data_time: 1.3874 memory: 61145 loss: 0.2030 +2023/06/01 12:06:05 - mmengine - INFO - Epoch(train) [15][ 400/1440] lr: 8.3237e-04 eta: 7:26:32 time: 3.0224 data_time: 1.2620 memory: 61145 loss: 0.2024 +2023/06/01 12:11:15 - mmengine - INFO - Epoch(train) [15][ 500/1440] lr: 8.3237e-04 eta: 7:21:02 time: 3.0775 data_time: 1.2974 memory: 61145 loss: 0.2039 +2023/06/01 12:16:33 - mmengine - INFO - Epoch(train) [15][ 600/1440] lr: 8.3237e-04 eta: 7:15:34 time: 3.1038 data_time: 1.3352 memory: 61145 loss: 0.2050 +2023/06/01 12:21:55 - mmengine - INFO - Epoch(train) [15][ 700/1440] lr: 8.3237e-04 eta: 7:10:07 time: 3.2604 data_time: 1.4990 memory: 61145 loss: 0.2004 +2023/06/01 12:27:11 - mmengine - INFO - Epoch(train) [15][ 800/1440] lr: 8.3237e-04 eta: 7:04:39 time: 3.1540 data_time: 1.3388 memory: 61145 loss: 0.2008 +2023/06/01 12:29:20 - mmengine - INFO - Exp name: convnext_tiny_4xb1024_fake5m_20230531_172740 +2023/06/01 12:32:28 - mmengine - INFO - Epoch(train) [15][ 900/1440] lr: 8.3237e-04 eta: 6:59:11 time: 3.1246 data_time: 1.3477 memory: 61145 loss: 0.2016 +2023/06/01 12:37:41 - mmengine - INFO - Epoch(train) [15][1000/1440] lr: 8.3237e-04 eta: 6:53:42 time: 3.1804 data_time: 1.4056 memory: 61145 loss: 0.2012 +2023/06/01 12:42:54 - mmengine - INFO - Epoch(train) [15][1100/1440] lr: 8.3237e-04 eta: 6:48:13 time: 3.1857 data_time: 1.3900 memory: 61145 loss: 0.2018 +2023/06/01 12:48:08 - mmengine - INFO - Epoch(train) [15][1200/1440] lr: 8.3237e-04 eta: 6:42:44 time: 3.1122 data_time: 1.3087 memory: 61145 loss: 0.2034 +2023/06/01 12:53:20 - mmengine - INFO - Epoch(train) [15][1300/1440] lr: 8.3237e-04 eta: 6:37:15 time: 3.0883 data_time: 1.3111 memory: 61145 loss: 0.2014 +2023/06/01 12:58:30 - mmengine - INFO - Epoch(train) [15][1400/1440] lr: 8.3237e-04 eta: 6:31:45 time: 3.0851 data_time: 1.3131 memory: 61145 loss: 0.2014 +2023/06/01 13:00:25 - mmengine - INFO - Exp name: convnext_tiny_4xb1024_fake5m_20230531_172740 +2023/06/01 13:00:25 - mmengine - INFO - Saving checkpoint at 15 epochs +2023/06/01 13:00:40 - mmengine - INFO - Epoch(val) [15][16/16] accuracy/top1: 94.0155 single-label/precision_classwise: [100.0, 0.0] single-label/recall_classwise: [94.0155258178711, 0.0] single-label/f1-score_classwise: [96.91546630859375, 0.0] data_time: 0.2400 time: 0.4684 +2023/06/01 13:06:09 - mmengine - INFO - Epoch(train) [16][ 100/1440] lr: 5.9432e-04 eta: 6:24:07 time: 3.0457 data_time: 1.0311 memory: 61145 loss: 0.2003 +2023/06/01 13:11:27 - mmengine - INFO - Epoch(train) [16][ 200/1440] lr: 5.9432e-04 eta: 6:18:41 time: 3.1365 data_time: 0.2847 memory: 61145 loss: 0.2004 +2023/06/01 13:16:46 - mmengine - INFO - Epoch(train) [16][ 300/1440] lr: 5.9432e-04 eta: 6:13:14 time: 3.6608 data_time: 1.3309 memory: 61145 loss: 0.2009 +2023/06/01 13:22:16 - mmengine - INFO - Exp name: convnext_tiny_4xb1024_fake5m_20230531_172740 +2023/06/01 13:22:16 - mmengine - INFO - Epoch(train) [16][ 400/1440] lr: 5.9432e-04 eta: 6:07:51 time: 3.1618 data_time: 1.4055 memory: 61145 loss: 0.2004 +2023/06/01 13:27:46 - mmengine - INFO - Epoch(train) [16][ 500/1440] lr: 5.9432e-04 eta: 6:02:28 time: 3.4606 data_time: 1.7095 memory: 61145 loss: 0.2003 +2023/06/01 13:33:13 - mmengine - INFO - Epoch(train) [16][ 600/1440] lr: 5.9432e-04 eta: 5:57:05 time: 3.2758 data_time: 1.5108 memory: 61145 loss: 0.2016 +2023/06/01 13:38:35 - mmengine - INFO - Epoch(train) [16][ 700/1440] lr: 5.9432e-04 eta: 5:51:39 time: 3.1949 data_time: 1.4147 memory: 61145 loss: 0.2009 +2023/06/01 13:43:53 - mmengine - INFO - Epoch(train) [16][ 800/1440] lr: 5.9432e-04 eta: 5:46:13 time: 3.1362 data_time: 1.3603 memory: 61145 loss: 0.2001 +2023/06/01 13:49:15 - mmengine - INFO - Epoch(train) [16][ 900/1440] lr: 5.9432e-04 eta: 5:40:47 time: 3.2374 data_time: 1.4581 memory: 61145 loss: 0.1996 +2023/06/01 13:54:32 - mmengine - INFO - Epoch(train) [16][1000/1440] lr: 5.9432e-04 eta: 5:35:21 time: 3.1935 data_time: 1.4217 memory: 61145 loss: 0.2007 +2023/06/01 13:59:53 - mmengine - INFO - Epoch(train) [16][1100/1440] lr: 5.9432e-04 eta: 5:29:55 time: 3.3168 data_time: 1.5334 memory: 61145 loss: 0.2016 +2023/06/01 14:05:09 - mmengine - INFO - Epoch(train) [16][1200/1440] lr: 5.9432e-04 eta: 5:24:28 time: 3.1746 data_time: 1.3878 memory: 61145 loss: 0.2012 +2023/06/01 14:10:30 - mmengine - INFO - Epoch(train) [16][1300/1440] lr: 5.9432e-04 eta: 5:19:03 time: 3.1452 data_time: 1.3828 memory: 61145 loss: 0.2011 +2023/06/01 14:15:52 - mmengine - INFO - Exp name: convnext_tiny_4xb1024_fake5m_20230531_172740 +2023/06/01 14:15:52 - mmengine - INFO - Epoch(train) [16][1400/1440] lr: 5.9432e-04 eta: 5:13:38 time: 3.2832 data_time: 1.5014 memory: 61145 loss: 0.2013 +2023/06/01 14:18:02 - mmengine - INFO - Exp name: convnext_tiny_4xb1024_fake5m_20230531_172740 +2023/06/01 14:18:02 - mmengine - INFO - Saving checkpoint at 16 epochs +2023/06/01 14:18:17 - mmengine - INFO - Epoch(val) [16][16/16] accuracy/top1: 98.2261 single-label/precision_classwise: [100.0, 0.0] single-label/recall_classwise: [98.22612762451172, 0.0] single-label/f1-score_classwise: [99.10511779785156, 0.0] data_time: 0.2630 time: 0.4892 +2023/06/01 14:23:55 - mmengine - INFO - Epoch(train) [17][ 100/1440] lr: 3.9101e-04 eta: 5:06:07 time: 3.1291 data_time: 1.3749 memory: 61145 loss: 0.2003 +2023/06/01 14:29:22 - mmengine - INFO - Epoch(train) [17][ 200/1440] lr: 3.9101e-04 eta: 5:00:43 time: 3.4322 data_time: 1.6813 memory: 61145 loss: 0.1996 +2023/06/01 14:35:02 - mmengine - INFO - Epoch(train) [17][ 300/1440] lr: 3.9101e-04 eta: 4:55:22 time: 3.4481 data_time: 1.2939 memory: 61145 loss: 0.2001 +2023/06/01 14:40:30 - mmengine - INFO - Epoch(train) [17][ 400/1440] lr: 3.9101e-04 eta: 4:49:58 time: 3.2638 data_time: 1.5064 memory: 61145 loss: 0.2002 +2023/06/01 14:45:53 - mmengine - INFO - Epoch(train) [17][ 500/1440] lr: 3.9101e-04 eta: 4:44:33 time: 3.1671 data_time: 1.3796 memory: 61145 loss: 0.2016 +2023/06/01 14:51:12 - mmengine - INFO - Epoch(train) [17][ 600/1440] lr: 3.9101e-04 eta: 4:39:08 time: 3.0302 data_time: 1.2577 memory: 61145 loss: 0.1997 +2023/06/01 14:56:17 - mmengine - INFO - Epoch(train) [17][ 700/1440] lr: 3.9101e-04 eta: 4:33:39 time: 3.1236 data_time: 1.3265 memory: 61145 loss: 0.1996 +2023/06/01 15:01:27 - mmengine - INFO - Epoch(train) [17][ 800/1440] lr: 3.9101e-04 eta: 4:28:11 time: 2.9701 data_time: 1.1868 memory: 61145 loss: 0.1999 +2023/06/01 15:06:28 - mmengine - INFO - Epoch(train) [17][ 900/1440] lr: 3.9101e-04 eta: 4:22:42 time: 3.0264 data_time: 1.2686 memory: 61145 loss: 0.2004 diff --git a/convnext_tiny_4xb1024_4e-3lr_5m/20230531_172740/vis_data/20230531_172740.json b/convnext_tiny_4xb1024_4e-3lr_5m/20230531_172740/vis_data/20230531_172740.json new file mode 100644 index 0000000000000000000000000000000000000000..631d6e1b9b30662998a124b4d8433a8b6225320d --- /dev/null +++ b/convnext_tiny_4xb1024_4e-3lr_5m/20230531_172740/vis_data/20230531_172740.json @@ -0,0 +1,249 @@ +{"lr": 0.004, "data_time": 1.7130630493164063, "loss": 0.6280613899230957, "time": 4.36142795085907, "epoch": 1, "memory": 61143, "step": 100} +{"lr": 0.004, "data_time": 1.3604084968566894, "loss": 0.604778128862381, "time": 4.2237999677658085, "epoch": 1, "memory": 61143, "step": 200} +{"lr": 0.004, "data_time": 0.8225233078002929, "loss": 0.5770811200141907, "time": 3.827829909324646, "epoch": 1, "memory": 61143, "step": 300} +{"lr": 0.004, "data_time": 1.9557789325714112, "loss": 0.552632063627243, "time": 4.0649792671203615, "epoch": 1, "memory": 61143, "step": 400} +{"lr": 0.004, "data_time": 2.152598524093628, "loss": 0.5663142740726471, "time": 3.9281469106674196, "epoch": 1, "memory": 61143, "step": 500} +{"lr": 0.004, "data_time": 1.68164381980896, "loss": 0.5425537526607513, "time": 3.4568049907684326, "epoch": 1, "memory": 61143, "step": 600} +{"lr": 0.004, "data_time": 1.3889002323150634, "loss": 0.5292092591524125, "time": 3.1568449258804323, "epoch": 1, "memory": 61143, "step": 700} +{"lr": 0.004, "data_time": 1.8797118663787842, "loss": 0.5095146715641021, "time": 3.640153408050537, "epoch": 1, "memory": 61143, "step": 800} +{"lr": 0.004, "data_time": 1.844734215736389, "loss": 0.4993878424167633, "time": 3.6293958187103272, "epoch": 1, "memory": 61143, "step": 900} +{"lr": 0.004, "data_time": 1.6931977033615113, "loss": 0.4898508757352829, "time": 3.461032843589783, "epoch": 1, "memory": 61143, "step": 1000} +{"lr": 0.004, "data_time": 1.602983283996582, "loss": 0.4797444552183151, "time": 3.357076120376587, "epoch": 1, "memory": 61143, "step": 1100} +{"lr": 0.004, "data_time": 1.8828502655029298, "loss": 0.4687023848295212, "time": 3.6645478010177612, "epoch": 1, "memory": 61143, "step": 1200} +{"lr": 0.004, "data_time": 2.0432738780975344, "loss": 0.4436558812856674, "time": 3.814341735839844, "epoch": 1, "memory": 61143, "step": 1300} +{"lr": 0.004, "data_time": 2.112619733810425, "loss": 0.45178292095661166, "time": 3.8833593130111694, "epoch": 1, "memory": 61143, "step": 1400} +{"accuracy/top1": 54.33369064331055, "data_time": 0.27214300632476807, "time": 0.6429488658905029, "step": 1} +{"lr": 0.0039754382394873, "data_time": 2.4369296550750734, "loss": 0.45136943757534026, "time": 4.204821467399597, "epoch": 2, "memory": 61146, "step": 1540} +{"lr": 0.0039754382394873, "data_time": 1.7613685846328735, "loss": 0.43077549040317537, "time": 3.5233321666717528, "epoch": 2, "memory": 61145, "step": 1640} +{"lr": 0.0039754382394873, "data_time": 1.6979450225830077, "loss": 0.41698336601257324, "time": 3.4566946268081664, "epoch": 2, "memory": 61145, "step": 1740} +{"lr": 0.0039754382394873, "data_time": 1.9897242546081544, "loss": 0.4037817716598511, "time": 3.781732964515686, "epoch": 2, "memory": 61145, "step": 1840} +{"lr": 0.0039754382394873, "data_time": 1.7586026430130004, "loss": 0.3876725047826767, "time": 3.5252096176147463, "epoch": 2, "memory": 61145, "step": 1940} +{"lr": 0.0039754382394873, "data_time": 1.4087422132492065, "loss": 0.3851060479879379, "time": 3.174519109725952, "epoch": 2, "memory": 61145, "step": 2040} +{"lr": 0.0039754382394873, "data_time": 1.414848518371582, "loss": 0.4002061516046524, "time": 3.206371474266052, "epoch": 2, "memory": 61145, "step": 2140} +{"lr": 0.0039754382394873, "data_time": 1.243985915184021, "loss": 0.35516210496425626, "time": 2.9946089029312133, "epoch": 2, "memory": 61145, "step": 2240} +{"lr": 0.0039754382394873, "data_time": 1.2611984968185426, "loss": 0.35801794528961184, "time": 3.040983867645264, "epoch": 2, "memory": 61145, "step": 2340} +{"lr": 0.0039754382394873, "data_time": 1.5412312030792237, "loss": 0.34500221312046053, "time": 3.3048946380615236, "epoch": 2, "memory": 61145, "step": 2440} +{"lr": 0.0039754382394873, "data_time": 1.3897242307662965, "loss": 0.33305284976959226, "time": 3.1557993650436402, "epoch": 2, "memory": 61145, "step": 2540} +{"lr": 0.0039754382394873, "data_time": 1.438400959968567, "loss": 0.3145768165588379, "time": 3.2056346416473387, "epoch": 2, "memory": 61145, "step": 2640} +{"lr": 0.0039754382394873, "data_time": 1.5566883325576781, "loss": 0.3397660493850708, "time": 3.313472104072571, "epoch": 2, "memory": 61145, "step": 2740} +{"lr": 0.0039754382394873, "data_time": 1.6529077529907226, "loss": 0.30533234775066376, "time": 3.4440282821655273, "epoch": 2, "memory": 61145, "step": 2840} +{"accuracy/top1": 73.44864654541016, "data_time": 0.2554305020500632, "time": 0.4852958426756017, "step": 2} +{"lr": 0.0039023577500088323, "data_time": 1.6938936471939088, "loss": 0.27967030107975005, "time": 3.5297956705093383, "epoch": 3, "memory": 61145, "step": 2980} +{"lr": 0.0039023577500088323, "data_time": 1.2517368078231812, "loss": 0.2744703829288483, "time": 3.3689878225326537, "epoch": 3, "memory": 61145, "step": 3080} +{"lr": 0.0039023577500088323, "data_time": 1.634689211845398, "loss": 0.2830836445093155, "time": 3.411437654495239, "epoch": 3, "memory": 61145, "step": 3180} +{"lr": 0.0039023577500088323, "data_time": 1.5658483505249023, "loss": 0.2656904995441437, "time": 3.315410280227661, "epoch": 3, "memory": 61145, "step": 3280} +{"lr": 0.0039023577500088323, "data_time": 1.5091981887817383, "loss": 0.26522556096315386, "time": 3.27861430644989, "epoch": 3, "memory": 61145, "step": 3380} +{"lr": 0.0039023577500088323, "data_time": 1.7050632953643798, "loss": 0.25637071281671525, "time": 3.4873088359832765, "epoch": 3, "memory": 61145, "step": 3480} +{"lr": 0.0039023577500088323, "data_time": 1.382461929321289, "loss": 0.25835521817207335, "time": 3.165927195549011, "epoch": 3, "memory": 61145, "step": 3580} +{"lr": 0.0039023577500088323, "data_time": 1.5841979265213013, "loss": 0.2746317759156227, "time": 3.362286019325256, "epoch": 3, "memory": 61145, "step": 3680} +{"lr": 0.0039023577500088323, "data_time": 1.4913410186767577, "loss": 0.44324982464313506, "time": 3.279749298095703, "epoch": 3, "memory": 61145, "step": 3780} +{"lr": 0.0039023577500088323, "data_time": 1.6023799180984497, "loss": 0.30170021057128904, "time": 3.368915319442749, "epoch": 3, "memory": 61145, "step": 3880} +{"lr": 0.0039023577500088323, "data_time": 2.003620076179504, "loss": 0.24604385048151017, "time": 3.7910067558288576, "epoch": 3, "memory": 61145, "step": 3980} +{"lr": 0.0039023577500088323, "data_time": 1.633652687072754, "loss": 0.2390464022755623, "time": 3.4165264129638673, "epoch": 3, "memory": 61145, "step": 4080} +{"lr": 0.0039023577500088323, "data_time": 1.686966323852539, "loss": 0.23189062476158143, "time": 3.448843550682068, "epoch": 3, "memory": 61145, "step": 4180} +{"lr": 0.0039023577500088323, "data_time": 1.6247499704360961, "loss": 0.23772178888320922, "time": 3.410560417175293, "epoch": 3, "memory": 61145, "step": 4280} +{"accuracy/top1": 80.7587890625, "data_time": 0.25339679157032685, "time": 0.486673256930183, "step": 3} +{"lr": 0.0037825580157557948, "data_time": 1.4799125432968139, "loss": 0.23052257299423218, "time": 3.2469041109085084, "epoch": 4, "memory": 61145, "step": 4420} +{"lr": 0.0037825580157557948, "data_time": 1.524852418899536, "loss": 0.23137079030275345, "time": 3.2973489284515383, "epoch": 4, "memory": 61145, "step": 4520} +{"lr": 0.0037825580157557948, "data_time": 1.588316512107849, "loss": 0.22561066150665282, "time": 3.3671595811843873, "epoch": 4, "memory": 61145, "step": 4620} +{"lr": 0.0037825580157557948, "data_time": 1.4662907600402832, "loss": 0.23371999561786652, "time": 3.2422826528549193, "epoch": 4, "memory": 61145, "step": 4720} +{"lr": 0.0037825580157557948, "data_time": 1.3720803022384644, "loss": 0.23059415072202682, "time": 3.138239860534668, "epoch": 4, "memory": 61145, "step": 4820} +{"lr": 0.0037825580157557948, "data_time": 1.367960023880005, "loss": 0.22049878537654877, "time": 3.133543682098389, "epoch": 4, "memory": 61145, "step": 4920} +{"lr": 0.0037825580157557948, "data_time": 1.3326706886291504, "loss": 0.22130948007106782, "time": 3.1160457134246826, "epoch": 4, "memory": 61145, "step": 5020} +{"lr": 0.0037825580157557948, "data_time": 1.496960711479187, "loss": 0.21808547526597977, "time": 3.248895525932312, "epoch": 4, "memory": 61145, "step": 5120} +{"lr": 0.0037825580157557948, "data_time": 1.4183381795883179, "loss": 0.4881200700998306, "time": 3.1870269775390625, "epoch": 4, "memory": 61145, "step": 5220} +{"lr": 0.0037825580157557948, "data_time": 1.3342695951461792, "loss": 0.3468328148126602, "time": 3.084699034690857, "epoch": 4, "memory": 61145, "step": 5320} +{"lr": 0.0037825580157557948, "data_time": 1.4967683553695679, "loss": 0.26561583280563356, "time": 3.247421407699585, "epoch": 4, "memory": 61145, "step": 5420} +{"lr": 0.0037825580157557948, "data_time": 1.4720463752746582, "loss": 0.22668404877185822, "time": 3.26472749710083, "epoch": 4, "memory": 61145, "step": 5520} +{"lr": 0.0037825580157557948, "data_time": 0.0023345708847045898, "loss": 0.224278225004673, "time": 3.0901735067367553, "epoch": 4, "memory": 61145, "step": 5620} +{"lr": 0.0037825580157557948, "data_time": 0.7537124872207641, "loss": 0.21717673391103745, "time": 3.075292778015137, "epoch": 4, "memory": 61145, "step": 5720} +{"accuracy/top1": 80.49996948242188, "data_time": 0.25596802374895883, "time": 0.4827101651360007, "step": 4} +{"lr": 0.0036189889037780212, "data_time": 1.2480610609054565, "loss": 0.5187392711639405, "time": 3.075471377372742, "epoch": 5, "memory": 61145, "step": 5860} +{"lr": 0.0036189889037780212, "data_time": 0.6788094997406006, "loss": 0.42944357395172117, "time": 3.0854479551315306, "epoch": 5, "memory": 61145, "step": 5960} +{"lr": 0.0036189889037780212, "data_time": 1.1822561740875244, "loss": 0.3694631189107895, "time": 3.0640578508377074, "epoch": 5, "memory": 61145, "step": 6060} +{"lr": 0.0036189889037780212, "data_time": 1.2857311964035034, "loss": 0.24512279480695726, "time": 3.0508487462997436, "epoch": 5, "memory": 61145, "step": 6160} +{"lr": 0.0036189889037780212, "data_time": 1.289154601097107, "loss": 0.22952439188957213, "time": 3.0749281644821167, "epoch": 5, "memory": 61145, "step": 6260} +{"lr": 0.0036189889037780212, "data_time": 1.2537958383560182, "loss": 0.2201613202691078, "time": 3.0172586917877195, "epoch": 5, "memory": 61145, "step": 6360} +{"lr": 0.0036189889037780212, "data_time": 1.285356378555298, "loss": 0.6604279816150666, "time": 3.080531668663025, "epoch": 5, "memory": 61145, "step": 6460} +{"lr": 0.0036189889037780212, "data_time": 1.426414966583252, "loss": 0.4415350526571274, "time": 3.17825186252594, "epoch": 5, "memory": 61145, "step": 6560} +{"lr": 0.0036189889037780212, "data_time": 1.4116779804229735, "loss": 0.23701888024806977, "time": 3.1798554420471192, "epoch": 5, "memory": 61145, "step": 6660} +{"lr": 0.0036189889037780212, "data_time": 1.3251469373703002, "loss": 0.21668200343847274, "time": 3.0851128101348877, "epoch": 5, "memory": 61145, "step": 6760} +{"lr": 0.0036189889037780212, "data_time": 1.3410115718841553, "loss": 0.22156630903482438, "time": 3.1023576736450194, "epoch": 5, "memory": 61145, "step": 6860} +{"lr": 0.0036189889037780212, "data_time": 1.341738796234131, "loss": 0.21785434931516648, "time": 3.099390721321106, "epoch": 5, "memory": 61145, "step": 6960} +{"lr": 0.0036189889037780212, "data_time": 1.3212546110153198, "loss": 0.2146911159157753, "time": 3.08558611869812, "epoch": 5, "memory": 61145, "step": 7060} +{"lr": 0.0036189889037780212, "data_time": 1.2699919939041138, "loss": 0.2130149260163307, "time": 3.059149074554443, "epoch": 5, "memory": 61145, "step": 7160} +{"accuracy/top1": 73.19613647460938, "data_time": 0.2506285695468678, "time": 0.47710183087517233, "step": 5} +{"lr": 0.003415678028467163, "data_time": 1.327199935913086, "loss": 0.21091615557670593, "time": 3.0956477403640745, "epoch": 6, "memory": 61145, "step": 7300} +{"lr": 0.003415678028467163, "data_time": 1.2757323503494262, "loss": 0.21031467467546464, "time": 3.031269073486328, "epoch": 6, "memory": 61145, "step": 7400} +{"lr": 0.003415678028467163, "data_time": 1.2686207294464111, "loss": 0.21194825917482377, "time": 3.0530646562576296, "epoch": 6, "memory": 61145, "step": 7500} +{"lr": 0.003415678028467163, "data_time": 1.3811330556869508, "loss": 0.21540516316890718, "time": 3.1302960395812987, "epoch": 6, "memory": 61145, "step": 7600} +{"lr": 0.003415678028467163, "data_time": 1.1832636594772339, "loss": 0.2109585776925087, "time": 2.958653521537781, "epoch": 6, "memory": 61145, "step": 7700} +{"lr": 0.003415678028467163, "data_time": 1.407991075515747, "loss": 0.23347438722848893, "time": 3.161695456504822, "epoch": 6, "memory": 61145, "step": 7800} +{"lr": 0.003415678028467163, "data_time": 1.342753291130066, "loss": 0.209233358502388, "time": 3.1355624437332152, "epoch": 6, "memory": 61145, "step": 7900} +{"lr": 0.003415678028467163, "data_time": 1.354392170906067, "loss": 0.2123303681612015, "time": 3.116430330276489, "epoch": 6, "memory": 61145, "step": 8000} +{"lr": 0.003415678028467163, "data_time": 1.3150188207626343, "loss": 0.21217477917671204, "time": 3.0944385051727297, "epoch": 6, "memory": 61145, "step": 8100} +{"lr": 0.003415678028467163, "data_time": 1.2378691673278808, "loss": 0.22938902974128722, "time": 6.356593608856201, "epoch": 6, "memory": 61145, "step": 8200} +{"lr": 0.003415678028467163, "data_time": 0.0022025108337402344, "loss": 0.5893660843372345, "time": 2.933739995956421, "epoch": 6, "memory": 61145, "step": 8300} +{"lr": 0.003415678028467163, "data_time": 0.42315075397491453, "loss": 0.5871405005455017, "time": 3.0413651943206785, "epoch": 6, "memory": 61145, "step": 8400} +{"lr": 0.003415678028467163, "data_time": 1.139907145500183, "loss": 0.5091971963644027, "time": 2.9095616579055785, "epoch": 6, "memory": 61145, "step": 8500} +{"lr": 0.003415678028467163, "data_time": 1.2904117345809936, "loss": 0.4725072294473648, "time": 3.0508846282958983, "epoch": 6, "memory": 61145, "step": 8600} +{"accuracy/top1": 78.41676330566406, "data_time": 0.23968013595132268, "time": 0.46704961271846995, "step": 6} +{"lr": 0.003177631578323485, "data_time": 1.2088043689727783, "loss": 0.4402738243341446, "time": 3.012364149093628, "epoch": 7, "memory": 61145, "step": 8740} +{"lr": 0.003177631578323485, "data_time": 1.3572895765304565, "loss": 0.3953871577978134, "time": 3.1733875274658203, "epoch": 7, "memory": 61145, "step": 8840} +{"lr": 0.003177631578323485, "data_time": 1.1977622270584107, "loss": 0.2557254731655121, "time": 2.9848200559616087, "epoch": 7, "memory": 61145, "step": 8940} +{"lr": 0.003177631578323485, "data_time": 1.3734773635864257, "loss": 0.22934531271457673, "time": 3.161170172691345, "epoch": 7, "memory": 61145, "step": 9040} +{"lr": 0.003177631578323485, "data_time": 1.3455553293228149, "loss": 0.21808532625436783, "time": 3.101999068260193, "epoch": 7, "memory": 61145, "step": 9140} +{"lr": 0.003177631578323485, "data_time": 1.2815997838973998, "loss": 0.21185320168733596, "time": 3.081390380859375, "epoch": 7, "memory": 61145, "step": 9240} +{"lr": 0.003177631578323485, "data_time": 1.3924853563308717, "loss": 0.2116745427250862, "time": 3.149598789215088, "epoch": 7, "memory": 61145, "step": 9340} +{"lr": 0.003177631578323485, "data_time": 1.2871636152267456, "loss": 0.2099997103214264, "time": 3.092425560951233, "epoch": 7, "memory": 61145, "step": 9440} +{"lr": 0.003177631578323485, "data_time": 1.3338433980941773, "loss": 0.21052040457725524, "time": 3.0859866380691527, "epoch": 7, "memory": 61145, "step": 9540} +{"lr": 0.003177631578323485, "data_time": 1.2957664251327514, "loss": 0.20748849660158158, "time": 3.067079782485962, "epoch": 7, "memory": 61145, "step": 9640} +{"lr": 0.003177631578323485, "data_time": 1.2907065153121948, "loss": 0.2065814405679703, "time": 3.0555028676986695, "epoch": 7, "memory": 61145, "step": 9740} +{"lr": 0.003177631578323485, "data_time": 1.3843186855316163, "loss": 0.2133854016661644, "time": 3.1655680894851685, "epoch": 7, "memory": 61145, "step": 9840} +{"lr": 0.003177631578323485, "data_time": 1.410230803489685, "loss": 0.6152868151664734, "time": 3.1757494449615478, "epoch": 7, "memory": 61145, "step": 9940} +{"lr": 0.003177631578323485, "data_time": 1.4121645212173461, "loss": 0.5245011568069458, "time": 3.177628183364868, "epoch": 7, "memory": 61145, "step": 10040} +{"accuracy/top1": 93.33374786376953, "data_time": 0.24559831619262695, "time": 0.4722733637865852, "step": 7} +{"lr": 0.0029107110469803964, "data_time": 0.6950544357299805, "loss": 0.4657981514930725, "time": 3.0901869773864745, "epoch": 8, "memory": 61145, "step": 10180} +{"lr": 0.0029107110469803964, "data_time": 0.8307860851287842, "loss": 0.43963272869586945, "time": 3.2131316661834717, "epoch": 8, "memory": 61145, "step": 10280} +{"lr": 0.0029107110469803964, "data_time": 1.0770029544830322, "loss": 0.41147109866142273, "time": 3.1430297613143923, "epoch": 8, "memory": 61145, "step": 10380} +{"lr": 0.0029107110469803964, "data_time": 1.290766429901123, "loss": 0.41178798079490664, "time": 3.055069160461426, "epoch": 8, "memory": 61145, "step": 10480} +{"lr": 0.0029107110469803964, "data_time": 1.297412061691284, "loss": 0.3912085950374603, "time": 3.0869417667388914, "epoch": 8, "memory": 61145, "step": 10580} +{"lr": 0.0029107110469803964, "data_time": 1.2923460483551026, "loss": 0.3674752235412598, "time": 3.0950441122055055, "epoch": 8, "memory": 61145, "step": 10680} +{"lr": 0.0029107110469803964, "data_time": 1.3209062814712524, "loss": 0.353262397646904, "time": 3.092371368408203, "epoch": 8, "memory": 61145, "step": 10780} +{"lr": 0.0029107110469803964, "data_time": 1.3249934196472168, "loss": 0.3378433495759964, "time": 3.0916277408599853, "epoch": 8, "memory": 61145, "step": 10880} +{"lr": 0.0029107110469803964, "data_time": 1.3870600700378417, "loss": 0.3248699098825455, "time": 3.1861149311065673, "epoch": 8, "memory": 61145, "step": 10980} +{"lr": 0.0029107110469803964, "data_time": 1.3786678552627563, "loss": 0.3034124970436096, "time": 3.1566009759902953, "epoch": 8, "memory": 61145, "step": 11080} +{"lr": 0.0029107110469803964, "data_time": 1.3681430101394654, "loss": 0.31824258863925936, "time": 3.139137887954712, "epoch": 8, "memory": 61145, "step": 11180} +{"lr": 0.0029107110469803964, "data_time": 1.3256127834320068, "loss": 0.2937474995851517, "time": 3.105956268310547, "epoch": 8, "memory": 61145, "step": 11280} +{"lr": 0.0029107110469803964, "data_time": 1.2849518060684204, "loss": 0.2927419424057007, "time": 3.0546663761138917, "epoch": 8, "memory": 61145, "step": 11380} +{"lr": 0.0029107110469803964, "data_time": 1.453737735748291, "loss": 0.2789072453975677, "time": 3.2541993856430054, "epoch": 8, "memory": 61145, "step": 11480} +{"accuracy/top1": 97.13401794433594, "data_time": 0.24754604171304143, "time": 0.4746855567483341, "step": 8} +{"lr": 0.0026214889037780207, "data_time": 0.8896720170974731, "loss": 0.27948315143585206, "time": 3.1850438594818113, "epoch": 9, "memory": 61145, "step": 11620} +{"lr": 0.0026214889037780207, "data_time": 1.125617504119873, "loss": 0.2639809250831604, "time": 3.1249682664871217, "epoch": 9, "memory": 61145, "step": 11720} +{"lr": 0.0026214889037780207, "data_time": 1.3541236162185668, "loss": 0.25968233346939085, "time": 3.1157520055770873, "epoch": 9, "memory": 61145, "step": 11820} +{"lr": 0.0026214889037780207, "data_time": 1.3616967439651488, "loss": 0.247555910050869, "time": 3.1485676050186155, "epoch": 9, "memory": 61145, "step": 11920} +{"lr": 0.0026214889037780207, "data_time": 1.2850595474243165, "loss": 0.23718794882297517, "time": 3.053327775001526, "epoch": 9, "memory": 61145, "step": 12020} +{"lr": 0.0026214889037780207, "data_time": 1.3581131219863891, "loss": 0.23645456433296203, "time": 3.1300790071487428, "epoch": 9, "memory": 61145, "step": 12120} +{"lr": 0.0026214889037780207, "data_time": 1.2983941078186034, "loss": 0.23156867027282715, "time": 3.0735569477081297, "epoch": 9, "memory": 61145, "step": 12220} +{"lr": 0.0026214889037780207, "data_time": 1.3186868906021119, "loss": 0.23194513469934464, "time": 3.0875450134277345, "epoch": 9, "memory": 61145, "step": 12320} +{"lr": 0.0026214889037780207, "data_time": 1.3964525938034058, "loss": 0.2303655579686165, "time": 3.172581696510315, "epoch": 9, "memory": 61145, "step": 12420} +{"lr": 0.0026214889037780207, "data_time": 1.3123528718948365, "loss": 0.2225619450211525, "time": 3.0856523752212524, "epoch": 9, "memory": 61145, "step": 12520} +{"lr": 0.0026214889037780207, "data_time": 1.2784934282302856, "loss": 0.22804639786481856, "time": 3.0721927165985106, "epoch": 9, "memory": 61145, "step": 12620} +{"lr": 0.0026214889037780207, "data_time": 1.2362180471420288, "loss": 0.22302616387605667, "time": 3.0199347734451294, "epoch": 9, "memory": 61145, "step": 12720} +{"lr": 0.0026214889037780207, "data_time": 1.411906933784485, "loss": 0.22183109670877457, "time": 3.1815821886062623, "epoch": 9, "memory": 61145, "step": 12820} +{"lr": 0.0026214889037780207, "data_time": 1.3612128019332885, "loss": 0.21841464340686798, "time": 3.1760150671005247, "epoch": 9, "memory": 61145, "step": 12920} +{"accuracy/top1": 98.0177993774414, "data_time": 0.23865142990561092, "time": 0.47052425496718464, "step": 9} +{"lr": 0.002317086757755261, "data_time": 1.1507697582244873, "loss": 0.214485327899456, "time": 3.1092417240142822, "epoch": 10, "memory": 61145, "step": 13060} +{"lr": 0.002317086757755261, "data_time": 1.3323131322860717, "loss": 0.2162133902311325, "time": 3.0941875696182253, "epoch": 10, "memory": 61145, "step": 13160} +{"lr": 0.002317086757755261, "data_time": 1.2937342405319214, "loss": 0.21600172072649002, "time": 3.077526497840881, "epoch": 10, "memory": 61145, "step": 13260} +{"lr": 0.002317086757755261, "data_time": 1.3234427452087403, "loss": 0.21877665519714357, "time": 3.1045006036758425, "epoch": 10, "memory": 61145, "step": 13360} +{"lr": 0.002317086757755261, "data_time": 1.3379269123077393, "loss": 0.2420973151922226, "time": 3.1301849603652956, "epoch": 10, "memory": 61145, "step": 13460} +{"lr": 0.002317086757755261, "data_time": 1.3330594539642333, "loss": 0.2090570256114006, "time": 3.0870211601257322, "epoch": 10, "memory": 61145, "step": 13560} +{"lr": 0.002317086757755261, "data_time": 1.386960792541504, "loss": 0.2136906072497368, "time": 3.1576788902282713, "epoch": 10, "memory": 61145, "step": 13660} +{"lr": 0.002317086757755261, "data_time": 1.4971266746520997, "loss": 0.2120979145169258, "time": 3.252758240699768, "epoch": 10, "memory": 61145, "step": 13760} +{"lr": 0.002317086757755261, "data_time": 1.4303601264953614, "loss": 0.20855435580015183, "time": 3.180695915222168, "epoch": 10, "memory": 61145, "step": 13860} +{"lr": 0.002317086757755261, "data_time": 1.3377168893814086, "loss": 0.21341949701309204, "time": 3.112483334541321, "epoch": 10, "memory": 61145, "step": 13960} +{"lr": 0.002317086757755261, "data_time": 1.4528234243392943, "loss": 0.2106423109769821, "time": 3.2459860324859617, "epoch": 10, "memory": 61145, "step": 14060} +{"lr": 0.002317086757755261, "data_time": 1.4445148944854735, "loss": 0.20821882039308548, "time": 3.198951506614685, "epoch": 10, "memory": 61145, "step": 14160} +{"lr": 0.002317086757755261, "data_time": 1.3375514984130858, "loss": 0.43864227533340455, "time": 3.1213868618011475, "epoch": 10, "memory": 61145, "step": 14260} +{"lr": 0.002317086757755261, "data_time": 1.3947025537490845, "loss": 0.22921686321496965, "time": 3.181926941871643, "epoch": 10, "memory": 61145, "step": 14360} +{"accuracy/top1": 96.93832397460938, "data_time": 0.23350057882421157, "time": 0.4679492782143986, "step": 10} +{"lr": 0.0020050000000000003, "data_time": 1.0506031274795533, "loss": 0.21396497637033463, "time": 3.0896746635437013, "epoch": 11, "memory": 61145, "step": 14500} +{"lr": 0.0020050000000000003, "data_time": 1.0325023412704468, "loss": 0.20982069671154022, "time": 3.13472900390625, "epoch": 11, "memory": 61145, "step": 14600} +{"lr": 0.0020050000000000003, "data_time": 1.3851130247116088, "loss": 0.2154417008161545, "time": 3.136973476409912, "epoch": 11, "memory": 61145, "step": 14700} +{"lr": 0.0020050000000000003, "data_time": 1.2574612140655517, "loss": 0.20886597335338591, "time": 3.0546548843383787, "epoch": 11, "memory": 61145, "step": 14800} +{"lr": 0.0020050000000000003, "data_time": 1.3200727462768556, "loss": 0.2102435976266861, "time": 3.0892250776290893, "epoch": 11, "memory": 61145, "step": 14900} +{"lr": 0.0020050000000000003, "data_time": 1.3789181470870973, "loss": 0.20736741572618483, "time": 3.158732056617737, "epoch": 11, "memory": 61145, "step": 15000} +{"lr": 0.0020050000000000003, "data_time": 1.273102331161499, "loss": 0.20581120550632476, "time": 3.023810052871704, "epoch": 11, "memory": 61145, "step": 15100} +{"lr": 0.0020050000000000003, "data_time": 1.4580342531204225, "loss": 0.21001167744398117, "time": 3.2336506605148316, "epoch": 11, "memory": 61145, "step": 15200} +{"lr": 0.0020050000000000003, "data_time": 1.310532021522522, "loss": 0.20431315153837204, "time": 3.0961413383483887, "epoch": 11, "memory": 61145, "step": 15300} +{"lr": 0.0020050000000000003, "data_time": 1.3269094944000244, "loss": 0.2048868253827095, "time": 3.1072128295898436, "epoch": 11, "memory": 61145, "step": 15400} +{"lr": 0.0020050000000000003, "data_time": 1.377335000038147, "loss": 0.21495335549116135, "time": 3.147787880897522, "epoch": 11, "memory": 61145, "step": 15500} +{"lr": 0.0020050000000000003, "data_time": 1.4597194194793701, "loss": 0.2065896660089493, "time": 3.248529052734375, "epoch": 11, "memory": 61145, "step": 15600} +{"lr": 0.0020050000000000003, "data_time": 1.3403338193893433, "loss": 0.20524612218141555, "time": 3.1004098415374757, "epoch": 11, "memory": 61145, "step": 15700} +{"lr": 0.0020050000000000003, "data_time": 1.3929861068725586, "loss": 0.3890265688300133, "time": 3.178326463699341, "epoch": 11, "memory": 61145, "step": 15800} +{"accuracy/top1": 92.39315032958984, "data_time": 0.25474998530219584, "time": 0.48205336402444277, "step": 11} +{"lr": 0.0016929132422447402, "data_time": 1.1677767753601074, "loss": 0.2115398958325386, "time": 3.1338894605636596, "epoch": 12, "memory": 61145, "step": 15940} +{"lr": 0.0016929132422447402, "data_time": 1.4426311731338501, "loss": 0.20715700685977936, "time": 3.198032522201538, "epoch": 12, "memory": 61145, "step": 16040} +{"lr": 0.0016929132422447402, "data_time": 1.4528038024902343, "loss": 0.20810561776161193, "time": 3.2064133167266844, "epoch": 12, "memory": 61145, "step": 16140} +{"lr": 0.0016929132422447402, "data_time": 1.493489408493042, "loss": 0.2062857300043106, "time": 3.2709250688552856, "epoch": 12, "memory": 61145, "step": 16240} +{"lr": 0.0016929132422447402, "data_time": 1.4901140689849854, "loss": 0.20680598616600038, "time": 3.2488963127136232, "epoch": 12, "memory": 61145, "step": 16340} +{"lr": 0.0016929132422447402, "data_time": 1.330900454521179, "loss": 0.20609278678894044, "time": 3.088787579536438, "epoch": 12, "memory": 61145, "step": 16440} +{"lr": 0.0016929132422447402, "data_time": 1.4864380836486817, "loss": 0.20673569440841674, "time": 3.256498193740845, "epoch": 12, "memory": 61145, "step": 16540} +{"lr": 0.0016929132422447402, "data_time": 1.3910550355911255, "loss": 0.20453892946243285, "time": 3.1482223987579347, "epoch": 12, "memory": 61145, "step": 16640} +{"lr": 0.0016929132422447402, "data_time": 1.5198049783706664, "loss": 0.20335696935653685, "time": 3.314421033859253, "epoch": 12, "memory": 61145, "step": 16740} +{"lr": 0.0016929132422447402, "data_time": 1.4722851753234862, "loss": 0.20423238277435302, "time": 3.2255250453948974, "epoch": 12, "memory": 61145, "step": 16840} +{"lr": 0.0016929132422447402, "data_time": 1.4404170513153076, "loss": 0.2077995851635933, "time": 3.2166297674179076, "epoch": 12, "memory": 61145, "step": 16940} +{"lr": 0.0016929132422447402, "data_time": 1.4542959451675415, "loss": 0.20349851250648499, "time": 3.2394116163253783, "epoch": 12, "memory": 61145, "step": 17040} +{"lr": 0.0016929132422447402, "data_time": 1.479795551300049, "loss": 0.20247942060232163, "time": 3.2678136348724367, "epoch": 12, "memory": 61145, "step": 17140} +{"lr": 0.0016929132422447402, "data_time": 1.4093853950500488, "loss": 0.20381221920251846, "time": 3.2344154357910155, "epoch": 12, "memory": 61145, "step": 17240} +{"accuracy/top1": 75.1720199584961, "data_time": 0.2596929774564855, "time": 0.49056057369007783, "step": 12} +{"lr": 0.0013885110962219803, "data_time": 0.8688613176345825, "loss": 0.2030133917927742, "time": 3.0908605575561525, "epoch": 13, "memory": 61145, "step": 17380} +{"lr": 0.0013885110962219803, "data_time": 0.5883035898208618, "loss": 0.20363679230213166, "time": 3.231678676605225, "epoch": 13, "memory": 61145, "step": 17480} +{"lr": 0.0013885110962219803, "data_time": 0.5508976221084595, "loss": 0.20518180578947068, "time": 3.185136890411377, "epoch": 13, "memory": 61145, "step": 17580} +{"lr": 0.0013885110962219803, "data_time": 1.391202735900879, "loss": 0.20737201869487762, "time": 3.2718473434448243, "epoch": 13, "memory": 61145, "step": 17680} +{"lr": 0.0013885110962219803, "data_time": 2.2405836820602416, "loss": 0.20596887171268463, "time": 4.006868648529053, "epoch": 13, "memory": 61145, "step": 17780} +{"lr": 0.0013885110962219803, "data_time": 1.4577032804489136, "loss": 0.20395630449056626, "time": 3.2259729623794557, "epoch": 13, "memory": 61145, "step": 17880} +{"lr": 0.0013885110962219803, "data_time": 1.4013015270233153, "loss": 0.20326561331748963, "time": 3.195839858055115, "epoch": 13, "memory": 61145, "step": 17980} +{"lr": 0.0013885110962219803, "data_time": 1.4415432929992675, "loss": 0.204030404984951, "time": 3.2272146463394167, "epoch": 13, "memory": 61145, "step": 18080} +{"lr": 0.0013885110962219803, "data_time": 1.4219282388687133, "loss": 0.20333558320999146, "time": 3.1845631837844848, "epoch": 13, "memory": 61145, "step": 18180} +{"lr": 0.0013885110962219803, "data_time": 1.4633343458175658, "loss": 0.20350033938884735, "time": 3.2231768131256104, "epoch": 13, "memory": 61145, "step": 18280} +{"lr": 0.0013885110962219803, "data_time": 1.474947166442871, "loss": 0.2013031557202339, "time": 3.2273441791534423, "epoch": 13, "memory": 61145, "step": 18380} +{"lr": 0.0013885110962219803, "data_time": 1.4674747467041016, "loss": 0.2027623936533928, "time": 3.2653159618377687, "epoch": 13, "memory": 61145, "step": 18480} +{"lr": 0.0013885110962219803, "data_time": 1.5045531749725343, "loss": 0.20190396159887314, "time": 3.2731655836105347, "epoch": 13, "memory": 61145, "step": 18580} +{"lr": 0.0013885110962219803, "data_time": 1.3992159605026244, "loss": 0.20390971451997758, "time": 3.1675029754638673, "epoch": 13, "memory": 61145, "step": 18680} +{"accuracy/top1": 61.77008819580078, "data_time": 0.2610341240377987, "time": 0.48876219637253704, "step": 13} +{"lr": 0.0010992889530196043, "data_time": 1.102649712562561, "loss": 0.20250055342912673, "time": 3.1805857181549073, "epoch": 14, "memory": 61145, "step": 18820} +{"lr": 0.0010992889530196043, "data_time": 1.4801113843917846, "loss": 0.20265392512083052, "time": 3.2537014961242674, "epoch": 14, "memory": 61145, "step": 18920} +{"lr": 0.0010992889530196043, "data_time": 1.5634637832641602, "loss": 0.20136702358722686, "time": 3.3251278400421143, "epoch": 14, "memory": 61145, "step": 19020} +{"lr": 0.0010992889530196043, "data_time": 1.4503780841827392, "loss": 0.20164737701416016, "time": 3.2222634077072145, "epoch": 14, "memory": 61145, "step": 19120} +{"lr": 0.0010992889530196043, "data_time": 1.5506370067596436, "loss": 0.20156373530626298, "time": 3.3353641271591186, "epoch": 14, "memory": 61145, "step": 19220} +{"lr": 0.0010992889530196043, "data_time": 1.5880133152008056, "loss": 0.20293943583965302, "time": 3.3588680982589723, "epoch": 14, "memory": 61145, "step": 19320} +{"lr": 0.0010992889530196043, "data_time": 1.4978216409683227, "loss": 0.2030561536550522, "time": 3.2776420593261717, "epoch": 14, "memory": 61145, "step": 19420} +{"lr": 0.0010992889530196043, "data_time": 1.5181540727615357, "loss": 0.20381435453891755, "time": 3.30031898021698, "epoch": 14, "memory": 61145, "step": 19520} +{"lr": 0.0010992889530196043, "data_time": 1.6687667608261108, "loss": 0.20175562500953675, "time": 3.447770690917969, "epoch": 14, "memory": 61145, "step": 19620} +{"lr": 0.0010992889530196043, "data_time": 1.4506142616271973, "loss": 0.20231660157442094, "time": 3.250755023956299, "epoch": 14, "memory": 61145, "step": 19720} +{"lr": 0.0010992889530196043, "data_time": 1.5440868377685546, "loss": 0.20214795619249343, "time": 3.3528568744659424, "epoch": 14, "memory": 61145, "step": 19820} +{"lr": 0.0010992889530196043, "data_time": 1.6898084878921509, "loss": 0.20152749568223954, "time": 3.469971776008606, "epoch": 14, "memory": 61145, "step": 19920} +{"lr": 0.0010992889530196043, "data_time": 1.6408870935440063, "loss": 0.5079215049743653, "time": 3.4183467388153077, "epoch": 14, "memory": 61145, "step": 20020} +{"lr": 0.0010992889530196043, "data_time": 1.5968194961547852, "loss": 0.21502864360809326, "time": 3.3625492572784426, "epoch": 14, "memory": 61145, "step": 20120} +{"accuracy/top1": 73.1014404296875, "data_time": 0.2505163445192225, "time": 0.47824956389034495, "step": 14} +{"lr": 0.0008323684216765164, "data_time": 1.439625096321106, "loss": 0.20632852613925934, "time": 3.2007597208023073, "epoch": 15, "memory": 61145, "step": 20260} +{"lr": 0.0008323684216765164, "data_time": 1.3790383338928223, "loss": 0.20381699800491332, "time": 3.1487918138504027, "epoch": 15, "memory": 61145, "step": 20360} +{"lr": 0.0008323684216765164, "data_time": 1.3874056100845338, "loss": 0.20298607647418976, "time": 3.1445539474487303, "epoch": 15, "memory": 61145, "step": 20460} +{"lr": 0.0008323684216765164, "data_time": 1.2619595527648926, "loss": 0.2023540496826172, "time": 3.0224238634109497, "epoch": 15, "memory": 61145, "step": 20560} +{"lr": 0.0008323684216765164, "data_time": 1.2974002838134766, "loss": 0.20385382920503617, "time": 3.0774792432785034, "epoch": 15, "memory": 61145, "step": 20660} +{"lr": 0.0008323684216765164, "data_time": 1.3352447986602782, "loss": 0.2050051286816597, "time": 3.1037906646728515, "epoch": 15, "memory": 61145, "step": 20760} +{"lr": 0.0008323684216765164, "data_time": 1.4990469217300415, "loss": 0.20038449317216872, "time": 3.2603516817092895, "epoch": 15, "memory": 61145, "step": 20860} +{"lr": 0.0008323684216765164, "data_time": 1.338820719718933, "loss": 0.20078101605176926, "time": 3.153957962989807, "epoch": 15, "memory": 61145, "step": 20960} +{"lr": 0.0008323684216765164, "data_time": 1.3476698637008666, "loss": 0.2015976220369339, "time": 3.1245984077453612, "epoch": 15, "memory": 61145, "step": 21060} +{"lr": 0.0008323684216765164, "data_time": 1.4055803537368774, "loss": 0.2011843204498291, "time": 3.1803860425949098, "epoch": 15, "memory": 61145, "step": 21160} +{"lr": 0.0008323684216765164, "data_time": 1.3900239706039428, "loss": 0.20184639394283294, "time": 3.1857438325881957, "epoch": 15, "memory": 61145, "step": 21260} +{"lr": 0.0008323684216765164, "data_time": 1.3086981058120728, "loss": 0.2033592864871025, "time": 3.1122119426727295, "epoch": 15, "memory": 61145, "step": 21360} +{"lr": 0.0008323684216765164, "data_time": 1.3110873699188232, "loss": 0.20138298124074935, "time": 3.0882763385772707, "epoch": 15, "memory": 61145, "step": 21460} +{"lr": 0.0008323684216765164, "data_time": 1.3130765438079834, "loss": 0.20135413855314255, "time": 3.0850628852844237, "epoch": 15, "memory": 61145, "step": 21560} +{"accuracy/top1": 94.0155258178711, "data_time": 0.23996667300953584, "time": 0.4683537342969109, "step": 15} +{"lr": 0.0005943219715328379, "data_time": 1.0311064720153809, "loss": 0.20026372075080873, "time": 3.045671558380127, "epoch": 16, "memory": 61145, "step": 21700} +{"lr": 0.0005943219715328379, "data_time": 0.2846883773803711, "loss": 0.20036861002445222, "time": 3.1365254402160643, "epoch": 16, "memory": 61145, "step": 21800} +{"lr": 0.0005943219715328379, "data_time": 1.3309248208999633, "loss": 0.20091276466846467, "time": 3.660755920410156, "epoch": 16, "memory": 61145, "step": 21900} +{"lr": 0.0005943219715328379, "data_time": 1.4055038690567017, "loss": 0.20040241926908492, "time": 3.161810564994812, "epoch": 16, "memory": 61145, "step": 22000} +{"lr": 0.0005943219715328379, "data_time": 1.7095362663269043, "loss": 0.20026686489582063, "time": 3.460629391670227, "epoch": 16, "memory": 61145, "step": 22100} +{"lr": 0.0005943219715328379, "data_time": 1.5108324766159058, "loss": 0.20159268528223037, "time": 3.275766682624817, "epoch": 16, "memory": 61145, "step": 22200} +{"lr": 0.0005943219715328379, "data_time": 1.4146517992019654, "loss": 0.2009176418185234, "time": 3.194920015335083, "epoch": 16, "memory": 61145, "step": 22300} +{"lr": 0.0005943219715328379, "data_time": 1.3603282928466798, "loss": 0.20006120651960374, "time": 3.1361812114715577, "epoch": 16, "memory": 61145, "step": 22400} +{"lr": 0.0005943219715328379, "data_time": 1.4580895900726318, "loss": 0.19959874004125594, "time": 3.2373690605163574, "epoch": 16, "memory": 61145, "step": 22500} +{"lr": 0.0005943219715328379, "data_time": 1.4217192649841308, "loss": 0.2007407784461975, "time": 3.1934733152389527, "epoch": 16, "memory": 61145, "step": 22600} +{"lr": 0.0005943219715328379, "data_time": 1.5334298849105834, "loss": 0.2016201823949814, "time": 3.316812753677368, "epoch": 16, "memory": 61145, "step": 22700} +{"lr": 0.0005943219715328379, "data_time": 1.387758493423462, "loss": 0.20124318450689316, "time": 3.174627113342285, "epoch": 16, "memory": 61145, "step": 22800} +{"lr": 0.0005943219715328379, "data_time": 1.382840871810913, "loss": 0.20111599266529084, "time": 3.1452425956726073, "epoch": 16, "memory": 61145, "step": 22900} +{"lr": 0.0005943219715328379, "data_time": 1.5014278173446656, "loss": 0.20131610035896302, "time": 3.28324134349823, "epoch": 16, "memory": 61145, "step": 23000} +{"accuracy/top1": 98.22611999511719, "data_time": 0.2630082579220043, "time": 0.4892232137567857, "step": 16} +{"lr": 0.0003910110962219801, "data_time": 1.3749398231506347, "loss": 0.20030642449855804, "time": 3.1291100025177, "epoch": 17, "memory": 61145, "step": 23140} +{"lr": 0.0003910110962219801, "data_time": 1.6813330888748168, "loss": 0.1995615243911743, "time": 3.4321943283081056, "epoch": 17, "memory": 61145, "step": 23240} +{"lr": 0.0003910110962219801, "data_time": 1.2939404010772706, "loss": 0.20007950067520142, "time": 3.448142981529236, "epoch": 17, "memory": 61145, "step": 23340} +{"lr": 0.0003910110962219801, "data_time": 1.5063703536987305, "loss": 0.20023534446954727, "time": 3.2638447761535643, "epoch": 17, "memory": 61145, "step": 23440} +{"lr": 0.0003910110962219801, "data_time": 1.379599666595459, "loss": 0.20158216059207917, "time": 3.1671363353729247, "epoch": 17, "memory": 61145, "step": 23540} +{"lr": 0.0003910110962219801, "data_time": 1.2577183485031127, "loss": 0.19966453313827515, "time": 3.0302154541015627, "epoch": 17, "memory": 61145, "step": 23640} +{"lr": 0.0003910110962219801, "data_time": 1.3265140771865844, "loss": 0.19958391636610032, "time": 3.1236394166946413, "epoch": 17, "memory": 61145, "step": 23740} +{"lr": 0.0003910110962219801, "data_time": 1.1868301391601563, "loss": 0.19993222057819365, "time": 2.970109748840332, "epoch": 17, "memory": 61145, "step": 23840} +{"lr": 0.0003910110962219801, "data_time": 1.2685500621795653, "loss": 0.2004125952720642, "time": 3.0263702869415283, "epoch": 17, "memory": 61145, "step": 23940} diff --git a/convnext_tiny_4xb1024_4e-3lr_5m/20230531_172740/vis_data/config.py b/convnext_tiny_4xb1024_4e-3lr_5m/20230531_172740/vis_data/config.py new file mode 100644 index 0000000000000000000000000000000000000000..662338ce17f9d9fd29063fd5b315ce1d8866f4f0 --- /dev/null +++ b/convnext_tiny_4xb1024_4e-3lr_5m/20230531_172740/vis_data/config.py @@ -0,0 +1,170 @@ +optim_wrapper = dict( + optimizer=dict( + type='AdamW', + lr=0.004, + weight_decay=0.05, + eps=1e-08, + betas=(0.9, 0.999), + _scope_='mmpretrain'), + paramwise_cfg=dict( + norm_decay_mult=0.0, + bias_decay_mult=0.0, + flat_decay_mult=0.0, + custom_keys=dict({ + '.absolute_pos_embed': dict(decay_mult=0.0), + '.relative_position_bias_table': dict(decay_mult=0.0) + })), + type='AmpOptimWrapper', + dtype='bfloat16', + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=True, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=20, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=4096) +model = dict( + type='ImageClassifier', + backbone=dict(type='ConvNeXt', arch='tiny', drop_path_rate=0.1), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=768, + loss=dict( + type='LabelSmoothLoss', label_smooth_val=0.1, mode='original'), + init_cfg=None), + init_cfg=dict( + type='TruncNormal', layer=['Conv2d', 'Linear'], std=0.02, bias=0.0), + train_cfg=None) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=1024, + num_workers=10, + dataset=dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stablediffusionV1-5R2-dpmsolver-25-5m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/convnext_tiny_4xb1024_4e-3lr_5m' diff --git a/convnext_tiny_4xb1024_4e-3lr_5m/20230531_172740/vis_data/events.out.tfevents.1685525267.SH-IDC1-10-140-24-119.217723.0 b/convnext_tiny_4xb1024_4e-3lr_5m/20230531_172740/vis_data/events.out.tfevents.1685525267.SH-IDC1-10-140-24-119.217723.0 new file mode 100644 index 0000000000000000000000000000000000000000..6d27021d7d86705862d66454ba31c5703ad761f6 --- /dev/null +++ b/convnext_tiny_4xb1024_4e-3lr_5m/20230531_172740/vis_data/events.out.tfevents.1685525267.SH-IDC1-10-140-24-119.217723.0 @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:8b31b43ea9106e0e0938a50003bb2436ea3789faf4fe4cea37248ea1c1e5b9ec +size 1204020 diff --git a/convnext_tiny_4xb1024_4e-3lr_5m/20230531_172740/vis_data/scalars.json b/convnext_tiny_4xb1024_4e-3lr_5m/20230531_172740/vis_data/scalars.json new file mode 100644 index 0000000000000000000000000000000000000000..631d6e1b9b30662998a124b4d8433a8b6225320d --- /dev/null +++ b/convnext_tiny_4xb1024_4e-3lr_5m/20230531_172740/vis_data/scalars.json @@ -0,0 +1,249 @@ +{"lr": 0.004, "data_time": 1.7130630493164063, "loss": 0.6280613899230957, "time": 4.36142795085907, "epoch": 1, "memory": 61143, "step": 100} +{"lr": 0.004, "data_time": 1.3604084968566894, "loss": 0.604778128862381, "time": 4.2237999677658085, "epoch": 1, "memory": 61143, "step": 200} +{"lr": 0.004, "data_time": 0.8225233078002929, "loss": 0.5770811200141907, "time": 3.827829909324646, "epoch": 1, "memory": 61143, "step": 300} +{"lr": 0.004, "data_time": 1.9557789325714112, "loss": 0.552632063627243, "time": 4.0649792671203615, "epoch": 1, "memory": 61143, "step": 400} +{"lr": 0.004, "data_time": 2.152598524093628, "loss": 0.5663142740726471, "time": 3.9281469106674196, "epoch": 1, "memory": 61143, "step": 500} +{"lr": 0.004, "data_time": 1.68164381980896, "loss": 0.5425537526607513, "time": 3.4568049907684326, "epoch": 1, "memory": 61143, "step": 600} +{"lr": 0.004, "data_time": 1.3889002323150634, "loss": 0.5292092591524125, "time": 3.1568449258804323, "epoch": 1, "memory": 61143, "step": 700} +{"lr": 0.004, "data_time": 1.8797118663787842, "loss": 0.5095146715641021, "time": 3.640153408050537, "epoch": 1, "memory": 61143, "step": 800} +{"lr": 0.004, "data_time": 1.844734215736389, "loss": 0.4993878424167633, "time": 3.6293958187103272, "epoch": 1, "memory": 61143, "step": 900} +{"lr": 0.004, "data_time": 1.6931977033615113, "loss": 0.4898508757352829, "time": 3.461032843589783, "epoch": 1, "memory": 61143, "step": 1000} +{"lr": 0.004, "data_time": 1.602983283996582, "loss": 0.4797444552183151, "time": 3.357076120376587, "epoch": 1, "memory": 61143, "step": 1100} +{"lr": 0.004, "data_time": 1.8828502655029298, "loss": 0.4687023848295212, "time": 3.6645478010177612, "epoch": 1, "memory": 61143, "step": 1200} +{"lr": 0.004, "data_time": 2.0432738780975344, "loss": 0.4436558812856674, "time": 3.814341735839844, "epoch": 1, "memory": 61143, "step": 1300} +{"lr": 0.004, "data_time": 2.112619733810425, "loss": 0.45178292095661166, "time": 3.8833593130111694, "epoch": 1, "memory": 61143, "step": 1400} +{"accuracy/top1": 54.33369064331055, "data_time": 0.27214300632476807, "time": 0.6429488658905029, "step": 1} +{"lr": 0.0039754382394873, "data_time": 2.4369296550750734, "loss": 0.45136943757534026, "time": 4.204821467399597, "epoch": 2, "memory": 61146, "step": 1540} +{"lr": 0.0039754382394873, "data_time": 1.7613685846328735, "loss": 0.43077549040317537, "time": 3.5233321666717528, "epoch": 2, "memory": 61145, "step": 1640} +{"lr": 0.0039754382394873, "data_time": 1.6979450225830077, "loss": 0.41698336601257324, "time": 3.4566946268081664, "epoch": 2, "memory": 61145, "step": 1740} +{"lr": 0.0039754382394873, "data_time": 1.9897242546081544, "loss": 0.4037817716598511, "time": 3.781732964515686, "epoch": 2, "memory": 61145, "step": 1840} +{"lr": 0.0039754382394873, "data_time": 1.7586026430130004, "loss": 0.3876725047826767, "time": 3.5252096176147463, "epoch": 2, "memory": 61145, "step": 1940} +{"lr": 0.0039754382394873, "data_time": 1.4087422132492065, "loss": 0.3851060479879379, "time": 3.174519109725952, "epoch": 2, "memory": 61145, "step": 2040} +{"lr": 0.0039754382394873, "data_time": 1.414848518371582, "loss": 0.4002061516046524, "time": 3.206371474266052, "epoch": 2, "memory": 61145, "step": 2140} +{"lr": 0.0039754382394873, "data_time": 1.243985915184021, "loss": 0.35516210496425626, "time": 2.9946089029312133, "epoch": 2, "memory": 61145, "step": 2240} +{"lr": 0.0039754382394873, "data_time": 1.2611984968185426, "loss": 0.35801794528961184, "time": 3.040983867645264, "epoch": 2, "memory": 61145, "step": 2340} +{"lr": 0.0039754382394873, "data_time": 1.5412312030792237, "loss": 0.34500221312046053, "time": 3.3048946380615236, "epoch": 2, "memory": 61145, "step": 2440} +{"lr": 0.0039754382394873, "data_time": 1.3897242307662965, "loss": 0.33305284976959226, "time": 3.1557993650436402, "epoch": 2, "memory": 61145, "step": 2540} +{"lr": 0.0039754382394873, "data_time": 1.438400959968567, "loss": 0.3145768165588379, "time": 3.2056346416473387, "epoch": 2, "memory": 61145, "step": 2640} +{"lr": 0.0039754382394873, "data_time": 1.5566883325576781, "loss": 0.3397660493850708, "time": 3.313472104072571, "epoch": 2, "memory": 61145, "step": 2740} +{"lr": 0.0039754382394873, "data_time": 1.6529077529907226, "loss": 0.30533234775066376, "time": 3.4440282821655273, "epoch": 2, "memory": 61145, "step": 2840} +{"accuracy/top1": 73.44864654541016, "data_time": 0.2554305020500632, "time": 0.4852958426756017, "step": 2} +{"lr": 0.0039023577500088323, "data_time": 1.6938936471939088, "loss": 0.27967030107975005, "time": 3.5297956705093383, "epoch": 3, "memory": 61145, "step": 2980} +{"lr": 0.0039023577500088323, "data_time": 1.2517368078231812, "loss": 0.2744703829288483, "time": 3.3689878225326537, "epoch": 3, "memory": 61145, "step": 3080} +{"lr": 0.0039023577500088323, "data_time": 1.634689211845398, "loss": 0.2830836445093155, "time": 3.411437654495239, "epoch": 3, "memory": 61145, "step": 3180} +{"lr": 0.0039023577500088323, "data_time": 1.5658483505249023, "loss": 0.2656904995441437, "time": 3.315410280227661, "epoch": 3, "memory": 61145, "step": 3280} +{"lr": 0.0039023577500088323, "data_time": 1.5091981887817383, "loss": 0.26522556096315386, "time": 3.27861430644989, "epoch": 3, "memory": 61145, "step": 3380} +{"lr": 0.0039023577500088323, "data_time": 1.7050632953643798, "loss": 0.25637071281671525, "time": 3.4873088359832765, "epoch": 3, "memory": 61145, "step": 3480} +{"lr": 0.0039023577500088323, "data_time": 1.382461929321289, "loss": 0.25835521817207335, "time": 3.165927195549011, "epoch": 3, "memory": 61145, "step": 3580} +{"lr": 0.0039023577500088323, "data_time": 1.5841979265213013, "loss": 0.2746317759156227, "time": 3.362286019325256, "epoch": 3, "memory": 61145, "step": 3680} +{"lr": 0.0039023577500088323, "data_time": 1.4913410186767577, "loss": 0.44324982464313506, "time": 3.279749298095703, "epoch": 3, "memory": 61145, "step": 3780} +{"lr": 0.0039023577500088323, "data_time": 1.6023799180984497, "loss": 0.30170021057128904, "time": 3.368915319442749, "epoch": 3, "memory": 61145, "step": 3880} +{"lr": 0.0039023577500088323, "data_time": 2.003620076179504, "loss": 0.24604385048151017, "time": 3.7910067558288576, "epoch": 3, "memory": 61145, "step": 3980} +{"lr": 0.0039023577500088323, "data_time": 1.633652687072754, "loss": 0.2390464022755623, "time": 3.4165264129638673, "epoch": 3, "memory": 61145, "step": 4080} +{"lr": 0.0039023577500088323, "data_time": 1.686966323852539, "loss": 0.23189062476158143, "time": 3.448843550682068, "epoch": 3, "memory": 61145, "step": 4180} +{"lr": 0.0039023577500088323, "data_time": 1.6247499704360961, "loss": 0.23772178888320922, "time": 3.410560417175293, "epoch": 3, "memory": 61145, "step": 4280} +{"accuracy/top1": 80.7587890625, "data_time": 0.25339679157032685, "time": 0.486673256930183, "step": 3} +{"lr": 0.0037825580157557948, "data_time": 1.4799125432968139, "loss": 0.23052257299423218, "time": 3.2469041109085084, "epoch": 4, "memory": 61145, "step": 4420} +{"lr": 0.0037825580157557948, "data_time": 1.524852418899536, "loss": 0.23137079030275345, "time": 3.2973489284515383, "epoch": 4, "memory": 61145, "step": 4520} +{"lr": 0.0037825580157557948, "data_time": 1.588316512107849, "loss": 0.22561066150665282, "time": 3.3671595811843873, "epoch": 4, "memory": 61145, "step": 4620} +{"lr": 0.0037825580157557948, "data_time": 1.4662907600402832, "loss": 0.23371999561786652, "time": 3.2422826528549193, "epoch": 4, "memory": 61145, "step": 4720} +{"lr": 0.0037825580157557948, "data_time": 1.3720803022384644, "loss": 0.23059415072202682, "time": 3.138239860534668, "epoch": 4, "memory": 61145, "step": 4820} +{"lr": 0.0037825580157557948, "data_time": 1.367960023880005, "loss": 0.22049878537654877, "time": 3.133543682098389, "epoch": 4, "memory": 61145, "step": 4920} +{"lr": 0.0037825580157557948, "data_time": 1.3326706886291504, "loss": 0.22130948007106782, "time": 3.1160457134246826, "epoch": 4, "memory": 61145, "step": 5020} +{"lr": 0.0037825580157557948, "data_time": 1.496960711479187, "loss": 0.21808547526597977, "time": 3.248895525932312, "epoch": 4, "memory": 61145, "step": 5120} +{"lr": 0.0037825580157557948, "data_time": 1.4183381795883179, "loss": 0.4881200700998306, "time": 3.1870269775390625, "epoch": 4, "memory": 61145, "step": 5220} +{"lr": 0.0037825580157557948, "data_time": 1.3342695951461792, "loss": 0.3468328148126602, "time": 3.084699034690857, "epoch": 4, "memory": 61145, "step": 5320} +{"lr": 0.0037825580157557948, "data_time": 1.4967683553695679, "loss": 0.26561583280563356, "time": 3.247421407699585, "epoch": 4, "memory": 61145, "step": 5420} +{"lr": 0.0037825580157557948, "data_time": 1.4720463752746582, "loss": 0.22668404877185822, "time": 3.26472749710083, "epoch": 4, "memory": 61145, "step": 5520} +{"lr": 0.0037825580157557948, "data_time": 0.0023345708847045898, "loss": 0.224278225004673, "time": 3.0901735067367553, "epoch": 4, "memory": 61145, "step": 5620} +{"lr": 0.0037825580157557948, "data_time": 0.7537124872207641, "loss": 0.21717673391103745, "time": 3.075292778015137, "epoch": 4, "memory": 61145, "step": 5720} +{"accuracy/top1": 80.49996948242188, "data_time": 0.25596802374895883, "time": 0.4827101651360007, "step": 4} +{"lr": 0.0036189889037780212, "data_time": 1.2480610609054565, "loss": 0.5187392711639405, "time": 3.075471377372742, "epoch": 5, "memory": 61145, "step": 5860} +{"lr": 0.0036189889037780212, "data_time": 0.6788094997406006, "loss": 0.42944357395172117, "time": 3.0854479551315306, "epoch": 5, "memory": 61145, "step": 5960} +{"lr": 0.0036189889037780212, "data_time": 1.1822561740875244, "loss": 0.3694631189107895, "time": 3.0640578508377074, "epoch": 5, "memory": 61145, "step": 6060} +{"lr": 0.0036189889037780212, "data_time": 1.2857311964035034, "loss": 0.24512279480695726, "time": 3.0508487462997436, "epoch": 5, "memory": 61145, "step": 6160} +{"lr": 0.0036189889037780212, "data_time": 1.289154601097107, "loss": 0.22952439188957213, "time": 3.0749281644821167, "epoch": 5, "memory": 61145, "step": 6260} +{"lr": 0.0036189889037780212, "data_time": 1.2537958383560182, "loss": 0.2201613202691078, "time": 3.0172586917877195, "epoch": 5, "memory": 61145, "step": 6360} +{"lr": 0.0036189889037780212, "data_time": 1.285356378555298, "loss": 0.6604279816150666, "time": 3.080531668663025, "epoch": 5, "memory": 61145, "step": 6460} +{"lr": 0.0036189889037780212, "data_time": 1.426414966583252, "loss": 0.4415350526571274, "time": 3.17825186252594, "epoch": 5, "memory": 61145, "step": 6560} +{"lr": 0.0036189889037780212, "data_time": 1.4116779804229735, "loss": 0.23701888024806977, "time": 3.1798554420471192, "epoch": 5, "memory": 61145, "step": 6660} +{"lr": 0.0036189889037780212, "data_time": 1.3251469373703002, "loss": 0.21668200343847274, "time": 3.0851128101348877, "epoch": 5, "memory": 61145, "step": 6760} +{"lr": 0.0036189889037780212, "data_time": 1.3410115718841553, "loss": 0.22156630903482438, "time": 3.1023576736450194, "epoch": 5, "memory": 61145, "step": 6860} +{"lr": 0.0036189889037780212, "data_time": 1.341738796234131, "loss": 0.21785434931516648, "time": 3.099390721321106, "epoch": 5, "memory": 61145, "step": 6960} +{"lr": 0.0036189889037780212, "data_time": 1.3212546110153198, "loss": 0.2146911159157753, "time": 3.08558611869812, "epoch": 5, "memory": 61145, "step": 7060} +{"lr": 0.0036189889037780212, "data_time": 1.2699919939041138, "loss": 0.2130149260163307, "time": 3.059149074554443, "epoch": 5, "memory": 61145, "step": 7160} +{"accuracy/top1": 73.19613647460938, "data_time": 0.2506285695468678, "time": 0.47710183087517233, "step": 5} +{"lr": 0.003415678028467163, "data_time": 1.327199935913086, "loss": 0.21091615557670593, "time": 3.0956477403640745, "epoch": 6, "memory": 61145, "step": 7300} +{"lr": 0.003415678028467163, "data_time": 1.2757323503494262, "loss": 0.21031467467546464, "time": 3.031269073486328, "epoch": 6, "memory": 61145, "step": 7400} +{"lr": 0.003415678028467163, "data_time": 1.2686207294464111, "loss": 0.21194825917482377, "time": 3.0530646562576296, "epoch": 6, "memory": 61145, "step": 7500} +{"lr": 0.003415678028467163, "data_time": 1.3811330556869508, "loss": 0.21540516316890718, "time": 3.1302960395812987, "epoch": 6, "memory": 61145, "step": 7600} +{"lr": 0.003415678028467163, "data_time": 1.1832636594772339, "loss": 0.2109585776925087, "time": 2.958653521537781, "epoch": 6, "memory": 61145, "step": 7700} +{"lr": 0.003415678028467163, "data_time": 1.407991075515747, "loss": 0.23347438722848893, "time": 3.161695456504822, "epoch": 6, "memory": 61145, "step": 7800} +{"lr": 0.003415678028467163, "data_time": 1.342753291130066, "loss": 0.209233358502388, "time": 3.1355624437332152, "epoch": 6, "memory": 61145, "step": 7900} +{"lr": 0.003415678028467163, "data_time": 1.354392170906067, "loss": 0.2123303681612015, "time": 3.116430330276489, "epoch": 6, "memory": 61145, "step": 8000} +{"lr": 0.003415678028467163, "data_time": 1.3150188207626343, "loss": 0.21217477917671204, "time": 3.0944385051727297, "epoch": 6, "memory": 61145, "step": 8100} +{"lr": 0.003415678028467163, "data_time": 1.2378691673278808, "loss": 0.22938902974128722, "time": 6.356593608856201, "epoch": 6, "memory": 61145, "step": 8200} +{"lr": 0.003415678028467163, "data_time": 0.0022025108337402344, "loss": 0.5893660843372345, "time": 2.933739995956421, "epoch": 6, "memory": 61145, "step": 8300} +{"lr": 0.003415678028467163, "data_time": 0.42315075397491453, "loss": 0.5871405005455017, "time": 3.0413651943206785, "epoch": 6, "memory": 61145, "step": 8400} +{"lr": 0.003415678028467163, "data_time": 1.139907145500183, "loss": 0.5091971963644027, "time": 2.9095616579055785, "epoch": 6, "memory": 61145, "step": 8500} +{"lr": 0.003415678028467163, "data_time": 1.2904117345809936, "loss": 0.4725072294473648, "time": 3.0508846282958983, "epoch": 6, "memory": 61145, "step": 8600} +{"accuracy/top1": 78.41676330566406, "data_time": 0.23968013595132268, "time": 0.46704961271846995, "step": 6} +{"lr": 0.003177631578323485, "data_time": 1.2088043689727783, "loss": 0.4402738243341446, "time": 3.012364149093628, "epoch": 7, "memory": 61145, "step": 8740} +{"lr": 0.003177631578323485, "data_time": 1.3572895765304565, "loss": 0.3953871577978134, "time": 3.1733875274658203, "epoch": 7, "memory": 61145, "step": 8840} +{"lr": 0.003177631578323485, "data_time": 1.1977622270584107, "loss": 0.2557254731655121, "time": 2.9848200559616087, "epoch": 7, "memory": 61145, "step": 8940} +{"lr": 0.003177631578323485, "data_time": 1.3734773635864257, "loss": 0.22934531271457673, "time": 3.161170172691345, "epoch": 7, "memory": 61145, "step": 9040} +{"lr": 0.003177631578323485, "data_time": 1.3455553293228149, "loss": 0.21808532625436783, "time": 3.101999068260193, "epoch": 7, "memory": 61145, "step": 9140} +{"lr": 0.003177631578323485, "data_time": 1.2815997838973998, "loss": 0.21185320168733596, "time": 3.081390380859375, "epoch": 7, "memory": 61145, "step": 9240} +{"lr": 0.003177631578323485, "data_time": 1.3924853563308717, "loss": 0.2116745427250862, "time": 3.149598789215088, "epoch": 7, "memory": 61145, "step": 9340} +{"lr": 0.003177631578323485, "data_time": 1.2871636152267456, "loss": 0.2099997103214264, "time": 3.092425560951233, "epoch": 7, "memory": 61145, "step": 9440} +{"lr": 0.003177631578323485, "data_time": 1.3338433980941773, "loss": 0.21052040457725524, "time": 3.0859866380691527, "epoch": 7, "memory": 61145, "step": 9540} +{"lr": 0.003177631578323485, "data_time": 1.2957664251327514, "loss": 0.20748849660158158, "time": 3.067079782485962, "epoch": 7, "memory": 61145, "step": 9640} +{"lr": 0.003177631578323485, "data_time": 1.2907065153121948, "loss": 0.2065814405679703, "time": 3.0555028676986695, "epoch": 7, "memory": 61145, "step": 9740} +{"lr": 0.003177631578323485, "data_time": 1.3843186855316163, "loss": 0.2133854016661644, "time": 3.1655680894851685, "epoch": 7, "memory": 61145, "step": 9840} +{"lr": 0.003177631578323485, "data_time": 1.410230803489685, "loss": 0.6152868151664734, "time": 3.1757494449615478, "epoch": 7, "memory": 61145, "step": 9940} +{"lr": 0.003177631578323485, "data_time": 1.4121645212173461, "loss": 0.5245011568069458, "time": 3.177628183364868, "epoch": 7, "memory": 61145, "step": 10040} +{"accuracy/top1": 93.33374786376953, "data_time": 0.24559831619262695, "time": 0.4722733637865852, "step": 7} +{"lr": 0.0029107110469803964, "data_time": 0.6950544357299805, "loss": 0.4657981514930725, "time": 3.0901869773864745, "epoch": 8, "memory": 61145, "step": 10180} +{"lr": 0.0029107110469803964, "data_time": 0.8307860851287842, "loss": 0.43963272869586945, "time": 3.2131316661834717, "epoch": 8, "memory": 61145, "step": 10280} +{"lr": 0.0029107110469803964, "data_time": 1.0770029544830322, "loss": 0.41147109866142273, "time": 3.1430297613143923, "epoch": 8, "memory": 61145, "step": 10380} +{"lr": 0.0029107110469803964, "data_time": 1.290766429901123, "loss": 0.41178798079490664, "time": 3.055069160461426, "epoch": 8, "memory": 61145, "step": 10480} +{"lr": 0.0029107110469803964, "data_time": 1.297412061691284, "loss": 0.3912085950374603, "time": 3.0869417667388914, "epoch": 8, "memory": 61145, "step": 10580} +{"lr": 0.0029107110469803964, "data_time": 1.2923460483551026, "loss": 0.3674752235412598, "time": 3.0950441122055055, "epoch": 8, "memory": 61145, "step": 10680} +{"lr": 0.0029107110469803964, "data_time": 1.3209062814712524, "loss": 0.353262397646904, "time": 3.092371368408203, "epoch": 8, "memory": 61145, "step": 10780} +{"lr": 0.0029107110469803964, "data_time": 1.3249934196472168, "loss": 0.3378433495759964, "time": 3.0916277408599853, "epoch": 8, "memory": 61145, "step": 10880} +{"lr": 0.0029107110469803964, "data_time": 1.3870600700378417, "loss": 0.3248699098825455, "time": 3.1861149311065673, "epoch": 8, "memory": 61145, "step": 10980} +{"lr": 0.0029107110469803964, "data_time": 1.3786678552627563, "loss": 0.3034124970436096, "time": 3.1566009759902953, "epoch": 8, "memory": 61145, "step": 11080} +{"lr": 0.0029107110469803964, "data_time": 1.3681430101394654, "loss": 0.31824258863925936, "time": 3.139137887954712, "epoch": 8, "memory": 61145, "step": 11180} +{"lr": 0.0029107110469803964, "data_time": 1.3256127834320068, "loss": 0.2937474995851517, "time": 3.105956268310547, "epoch": 8, "memory": 61145, "step": 11280} +{"lr": 0.0029107110469803964, "data_time": 1.2849518060684204, "loss": 0.2927419424057007, "time": 3.0546663761138917, "epoch": 8, "memory": 61145, "step": 11380} +{"lr": 0.0029107110469803964, "data_time": 1.453737735748291, "loss": 0.2789072453975677, "time": 3.2541993856430054, "epoch": 8, "memory": 61145, "step": 11480} +{"accuracy/top1": 97.13401794433594, "data_time": 0.24754604171304143, "time": 0.4746855567483341, "step": 8} +{"lr": 0.0026214889037780207, "data_time": 0.8896720170974731, "loss": 0.27948315143585206, "time": 3.1850438594818113, "epoch": 9, "memory": 61145, "step": 11620} +{"lr": 0.0026214889037780207, "data_time": 1.125617504119873, "loss": 0.2639809250831604, "time": 3.1249682664871217, "epoch": 9, "memory": 61145, "step": 11720} +{"lr": 0.0026214889037780207, "data_time": 1.3541236162185668, "loss": 0.25968233346939085, "time": 3.1157520055770873, "epoch": 9, "memory": 61145, "step": 11820} +{"lr": 0.0026214889037780207, "data_time": 1.3616967439651488, "loss": 0.247555910050869, "time": 3.1485676050186155, "epoch": 9, "memory": 61145, "step": 11920} +{"lr": 0.0026214889037780207, "data_time": 1.2850595474243165, "loss": 0.23718794882297517, "time": 3.053327775001526, "epoch": 9, "memory": 61145, "step": 12020} +{"lr": 0.0026214889037780207, "data_time": 1.3581131219863891, "loss": 0.23645456433296203, "time": 3.1300790071487428, "epoch": 9, "memory": 61145, "step": 12120} +{"lr": 0.0026214889037780207, "data_time": 1.2983941078186034, "loss": 0.23156867027282715, "time": 3.0735569477081297, "epoch": 9, "memory": 61145, "step": 12220} +{"lr": 0.0026214889037780207, "data_time": 1.3186868906021119, "loss": 0.23194513469934464, "time": 3.0875450134277345, "epoch": 9, "memory": 61145, "step": 12320} +{"lr": 0.0026214889037780207, "data_time": 1.3964525938034058, "loss": 0.2303655579686165, "time": 3.172581696510315, "epoch": 9, "memory": 61145, "step": 12420} +{"lr": 0.0026214889037780207, "data_time": 1.3123528718948365, "loss": 0.2225619450211525, "time": 3.0856523752212524, "epoch": 9, "memory": 61145, "step": 12520} +{"lr": 0.0026214889037780207, "data_time": 1.2784934282302856, "loss": 0.22804639786481856, "time": 3.0721927165985106, "epoch": 9, "memory": 61145, "step": 12620} +{"lr": 0.0026214889037780207, "data_time": 1.2362180471420288, "loss": 0.22302616387605667, "time": 3.0199347734451294, "epoch": 9, "memory": 61145, "step": 12720} +{"lr": 0.0026214889037780207, "data_time": 1.411906933784485, "loss": 0.22183109670877457, "time": 3.1815821886062623, "epoch": 9, "memory": 61145, "step": 12820} +{"lr": 0.0026214889037780207, "data_time": 1.3612128019332885, "loss": 0.21841464340686798, "time": 3.1760150671005247, "epoch": 9, "memory": 61145, "step": 12920} +{"accuracy/top1": 98.0177993774414, "data_time": 0.23865142990561092, "time": 0.47052425496718464, "step": 9} +{"lr": 0.002317086757755261, "data_time": 1.1507697582244873, "loss": 0.214485327899456, "time": 3.1092417240142822, "epoch": 10, "memory": 61145, "step": 13060} +{"lr": 0.002317086757755261, "data_time": 1.3323131322860717, "loss": 0.2162133902311325, "time": 3.0941875696182253, "epoch": 10, "memory": 61145, "step": 13160} +{"lr": 0.002317086757755261, "data_time": 1.2937342405319214, "loss": 0.21600172072649002, "time": 3.077526497840881, "epoch": 10, "memory": 61145, "step": 13260} +{"lr": 0.002317086757755261, "data_time": 1.3234427452087403, "loss": 0.21877665519714357, "time": 3.1045006036758425, "epoch": 10, "memory": 61145, "step": 13360} +{"lr": 0.002317086757755261, "data_time": 1.3379269123077393, "loss": 0.2420973151922226, "time": 3.1301849603652956, "epoch": 10, "memory": 61145, "step": 13460} +{"lr": 0.002317086757755261, "data_time": 1.3330594539642333, "loss": 0.2090570256114006, "time": 3.0870211601257322, "epoch": 10, "memory": 61145, "step": 13560} +{"lr": 0.002317086757755261, "data_time": 1.386960792541504, "loss": 0.2136906072497368, "time": 3.1576788902282713, "epoch": 10, "memory": 61145, "step": 13660} +{"lr": 0.002317086757755261, "data_time": 1.4971266746520997, "loss": 0.2120979145169258, "time": 3.252758240699768, "epoch": 10, "memory": 61145, "step": 13760} +{"lr": 0.002317086757755261, "data_time": 1.4303601264953614, "loss": 0.20855435580015183, "time": 3.180695915222168, "epoch": 10, "memory": 61145, "step": 13860} +{"lr": 0.002317086757755261, "data_time": 1.3377168893814086, "loss": 0.21341949701309204, "time": 3.112483334541321, "epoch": 10, "memory": 61145, "step": 13960} +{"lr": 0.002317086757755261, "data_time": 1.4528234243392943, "loss": 0.2106423109769821, "time": 3.2459860324859617, "epoch": 10, "memory": 61145, "step": 14060} +{"lr": 0.002317086757755261, "data_time": 1.4445148944854735, "loss": 0.20821882039308548, "time": 3.198951506614685, "epoch": 10, "memory": 61145, "step": 14160} +{"lr": 0.002317086757755261, "data_time": 1.3375514984130858, "loss": 0.43864227533340455, "time": 3.1213868618011475, "epoch": 10, "memory": 61145, "step": 14260} +{"lr": 0.002317086757755261, "data_time": 1.3947025537490845, "loss": 0.22921686321496965, "time": 3.181926941871643, "epoch": 10, "memory": 61145, "step": 14360} +{"accuracy/top1": 96.93832397460938, "data_time": 0.23350057882421157, "time": 0.4679492782143986, "step": 10} +{"lr": 0.0020050000000000003, "data_time": 1.0506031274795533, "loss": 0.21396497637033463, "time": 3.0896746635437013, "epoch": 11, "memory": 61145, "step": 14500} +{"lr": 0.0020050000000000003, "data_time": 1.0325023412704468, "loss": 0.20982069671154022, "time": 3.13472900390625, "epoch": 11, "memory": 61145, "step": 14600} +{"lr": 0.0020050000000000003, "data_time": 1.3851130247116088, "loss": 0.2154417008161545, "time": 3.136973476409912, "epoch": 11, "memory": 61145, "step": 14700} +{"lr": 0.0020050000000000003, "data_time": 1.2574612140655517, "loss": 0.20886597335338591, "time": 3.0546548843383787, "epoch": 11, "memory": 61145, "step": 14800} +{"lr": 0.0020050000000000003, "data_time": 1.3200727462768556, "loss": 0.2102435976266861, "time": 3.0892250776290893, "epoch": 11, "memory": 61145, "step": 14900} +{"lr": 0.0020050000000000003, "data_time": 1.3789181470870973, "loss": 0.20736741572618483, "time": 3.158732056617737, "epoch": 11, "memory": 61145, "step": 15000} +{"lr": 0.0020050000000000003, "data_time": 1.273102331161499, "loss": 0.20581120550632476, "time": 3.023810052871704, "epoch": 11, "memory": 61145, "step": 15100} +{"lr": 0.0020050000000000003, "data_time": 1.4580342531204225, "loss": 0.21001167744398117, "time": 3.2336506605148316, "epoch": 11, "memory": 61145, "step": 15200} +{"lr": 0.0020050000000000003, "data_time": 1.310532021522522, "loss": 0.20431315153837204, "time": 3.0961413383483887, "epoch": 11, "memory": 61145, "step": 15300} +{"lr": 0.0020050000000000003, "data_time": 1.3269094944000244, "loss": 0.2048868253827095, "time": 3.1072128295898436, "epoch": 11, "memory": 61145, "step": 15400} +{"lr": 0.0020050000000000003, "data_time": 1.377335000038147, "loss": 0.21495335549116135, "time": 3.147787880897522, "epoch": 11, "memory": 61145, "step": 15500} +{"lr": 0.0020050000000000003, "data_time": 1.4597194194793701, "loss": 0.2065896660089493, "time": 3.248529052734375, "epoch": 11, "memory": 61145, "step": 15600} +{"lr": 0.0020050000000000003, "data_time": 1.3403338193893433, "loss": 0.20524612218141555, "time": 3.1004098415374757, "epoch": 11, "memory": 61145, "step": 15700} +{"lr": 0.0020050000000000003, "data_time": 1.3929861068725586, "loss": 0.3890265688300133, "time": 3.178326463699341, "epoch": 11, "memory": 61145, "step": 15800} +{"accuracy/top1": 92.39315032958984, "data_time": 0.25474998530219584, "time": 0.48205336402444277, "step": 11} +{"lr": 0.0016929132422447402, "data_time": 1.1677767753601074, "loss": 0.2115398958325386, "time": 3.1338894605636596, "epoch": 12, "memory": 61145, "step": 15940} +{"lr": 0.0016929132422447402, "data_time": 1.4426311731338501, "loss": 0.20715700685977936, "time": 3.198032522201538, "epoch": 12, "memory": 61145, "step": 16040} +{"lr": 0.0016929132422447402, "data_time": 1.4528038024902343, "loss": 0.20810561776161193, "time": 3.2064133167266844, "epoch": 12, "memory": 61145, "step": 16140} +{"lr": 0.0016929132422447402, "data_time": 1.493489408493042, "loss": 0.2062857300043106, "time": 3.2709250688552856, "epoch": 12, "memory": 61145, "step": 16240} +{"lr": 0.0016929132422447402, "data_time": 1.4901140689849854, "loss": 0.20680598616600038, "time": 3.2488963127136232, "epoch": 12, "memory": 61145, "step": 16340} +{"lr": 0.0016929132422447402, "data_time": 1.330900454521179, "loss": 0.20609278678894044, "time": 3.088787579536438, "epoch": 12, "memory": 61145, "step": 16440} +{"lr": 0.0016929132422447402, "data_time": 1.4864380836486817, "loss": 0.20673569440841674, "time": 3.256498193740845, "epoch": 12, "memory": 61145, "step": 16540} +{"lr": 0.0016929132422447402, "data_time": 1.3910550355911255, "loss": 0.20453892946243285, "time": 3.1482223987579347, "epoch": 12, "memory": 61145, "step": 16640} +{"lr": 0.0016929132422447402, "data_time": 1.5198049783706664, "loss": 0.20335696935653685, "time": 3.314421033859253, "epoch": 12, "memory": 61145, "step": 16740} +{"lr": 0.0016929132422447402, "data_time": 1.4722851753234862, "loss": 0.20423238277435302, "time": 3.2255250453948974, "epoch": 12, "memory": 61145, "step": 16840} +{"lr": 0.0016929132422447402, "data_time": 1.4404170513153076, "loss": 0.2077995851635933, "time": 3.2166297674179076, "epoch": 12, "memory": 61145, "step": 16940} +{"lr": 0.0016929132422447402, "data_time": 1.4542959451675415, "loss": 0.20349851250648499, "time": 3.2394116163253783, "epoch": 12, "memory": 61145, "step": 17040} +{"lr": 0.0016929132422447402, "data_time": 1.479795551300049, "loss": 0.20247942060232163, "time": 3.2678136348724367, "epoch": 12, "memory": 61145, "step": 17140} +{"lr": 0.0016929132422447402, "data_time": 1.4093853950500488, "loss": 0.20381221920251846, "time": 3.2344154357910155, "epoch": 12, "memory": 61145, "step": 17240} +{"accuracy/top1": 75.1720199584961, "data_time": 0.2596929774564855, "time": 0.49056057369007783, "step": 12} +{"lr": 0.0013885110962219803, "data_time": 0.8688613176345825, "loss": 0.2030133917927742, "time": 3.0908605575561525, "epoch": 13, "memory": 61145, "step": 17380} +{"lr": 0.0013885110962219803, "data_time": 0.5883035898208618, "loss": 0.20363679230213166, "time": 3.231678676605225, "epoch": 13, "memory": 61145, "step": 17480} +{"lr": 0.0013885110962219803, "data_time": 0.5508976221084595, "loss": 0.20518180578947068, "time": 3.185136890411377, "epoch": 13, "memory": 61145, "step": 17580} +{"lr": 0.0013885110962219803, "data_time": 1.391202735900879, "loss": 0.20737201869487762, "time": 3.2718473434448243, "epoch": 13, "memory": 61145, "step": 17680} +{"lr": 0.0013885110962219803, "data_time": 2.2405836820602416, "loss": 0.20596887171268463, "time": 4.006868648529053, "epoch": 13, "memory": 61145, "step": 17780} +{"lr": 0.0013885110962219803, "data_time": 1.4577032804489136, "loss": 0.20395630449056626, "time": 3.2259729623794557, "epoch": 13, "memory": 61145, "step": 17880} +{"lr": 0.0013885110962219803, "data_time": 1.4013015270233153, "loss": 0.20326561331748963, "time": 3.195839858055115, "epoch": 13, "memory": 61145, "step": 17980} +{"lr": 0.0013885110962219803, "data_time": 1.4415432929992675, "loss": 0.204030404984951, "time": 3.2272146463394167, "epoch": 13, "memory": 61145, "step": 18080} +{"lr": 0.0013885110962219803, "data_time": 1.4219282388687133, "loss": 0.20333558320999146, "time": 3.1845631837844848, "epoch": 13, "memory": 61145, "step": 18180} +{"lr": 0.0013885110962219803, "data_time": 1.4633343458175658, "loss": 0.20350033938884735, "time": 3.2231768131256104, "epoch": 13, "memory": 61145, "step": 18280} +{"lr": 0.0013885110962219803, "data_time": 1.474947166442871, "loss": 0.2013031557202339, "time": 3.2273441791534423, "epoch": 13, "memory": 61145, "step": 18380} +{"lr": 0.0013885110962219803, "data_time": 1.4674747467041016, "loss": 0.2027623936533928, "time": 3.2653159618377687, "epoch": 13, "memory": 61145, "step": 18480} +{"lr": 0.0013885110962219803, "data_time": 1.5045531749725343, "loss": 0.20190396159887314, "time": 3.2731655836105347, "epoch": 13, "memory": 61145, "step": 18580} +{"lr": 0.0013885110962219803, "data_time": 1.3992159605026244, "loss": 0.20390971451997758, "time": 3.1675029754638673, "epoch": 13, "memory": 61145, "step": 18680} +{"accuracy/top1": 61.77008819580078, "data_time": 0.2610341240377987, "time": 0.48876219637253704, "step": 13} +{"lr": 0.0010992889530196043, "data_time": 1.102649712562561, "loss": 0.20250055342912673, "time": 3.1805857181549073, "epoch": 14, "memory": 61145, "step": 18820} +{"lr": 0.0010992889530196043, "data_time": 1.4801113843917846, "loss": 0.20265392512083052, "time": 3.2537014961242674, "epoch": 14, "memory": 61145, "step": 18920} +{"lr": 0.0010992889530196043, "data_time": 1.5634637832641602, "loss": 0.20136702358722686, "time": 3.3251278400421143, "epoch": 14, "memory": 61145, "step": 19020} +{"lr": 0.0010992889530196043, "data_time": 1.4503780841827392, "loss": 0.20164737701416016, "time": 3.2222634077072145, "epoch": 14, "memory": 61145, "step": 19120} +{"lr": 0.0010992889530196043, "data_time": 1.5506370067596436, "loss": 0.20156373530626298, "time": 3.3353641271591186, "epoch": 14, "memory": 61145, "step": 19220} +{"lr": 0.0010992889530196043, "data_time": 1.5880133152008056, "loss": 0.20293943583965302, "time": 3.3588680982589723, "epoch": 14, "memory": 61145, "step": 19320} +{"lr": 0.0010992889530196043, "data_time": 1.4978216409683227, "loss": 0.2030561536550522, "time": 3.2776420593261717, "epoch": 14, "memory": 61145, "step": 19420} +{"lr": 0.0010992889530196043, "data_time": 1.5181540727615357, "loss": 0.20381435453891755, "time": 3.30031898021698, "epoch": 14, "memory": 61145, "step": 19520} +{"lr": 0.0010992889530196043, "data_time": 1.6687667608261108, "loss": 0.20175562500953675, "time": 3.447770690917969, "epoch": 14, "memory": 61145, "step": 19620} +{"lr": 0.0010992889530196043, "data_time": 1.4506142616271973, "loss": 0.20231660157442094, "time": 3.250755023956299, "epoch": 14, "memory": 61145, "step": 19720} +{"lr": 0.0010992889530196043, "data_time": 1.5440868377685546, "loss": 0.20214795619249343, "time": 3.3528568744659424, "epoch": 14, "memory": 61145, "step": 19820} +{"lr": 0.0010992889530196043, "data_time": 1.6898084878921509, "loss": 0.20152749568223954, "time": 3.469971776008606, "epoch": 14, "memory": 61145, "step": 19920} +{"lr": 0.0010992889530196043, "data_time": 1.6408870935440063, "loss": 0.5079215049743653, "time": 3.4183467388153077, "epoch": 14, "memory": 61145, "step": 20020} +{"lr": 0.0010992889530196043, "data_time": 1.5968194961547852, "loss": 0.21502864360809326, "time": 3.3625492572784426, "epoch": 14, "memory": 61145, "step": 20120} +{"accuracy/top1": 73.1014404296875, "data_time": 0.2505163445192225, "time": 0.47824956389034495, "step": 14} +{"lr": 0.0008323684216765164, "data_time": 1.439625096321106, "loss": 0.20632852613925934, "time": 3.2007597208023073, "epoch": 15, "memory": 61145, "step": 20260} +{"lr": 0.0008323684216765164, "data_time": 1.3790383338928223, "loss": 0.20381699800491332, "time": 3.1487918138504027, "epoch": 15, "memory": 61145, "step": 20360} +{"lr": 0.0008323684216765164, "data_time": 1.3874056100845338, "loss": 0.20298607647418976, "time": 3.1445539474487303, "epoch": 15, "memory": 61145, "step": 20460} +{"lr": 0.0008323684216765164, "data_time": 1.2619595527648926, "loss": 0.2023540496826172, "time": 3.0224238634109497, "epoch": 15, "memory": 61145, "step": 20560} +{"lr": 0.0008323684216765164, "data_time": 1.2974002838134766, "loss": 0.20385382920503617, "time": 3.0774792432785034, "epoch": 15, "memory": 61145, "step": 20660} +{"lr": 0.0008323684216765164, "data_time": 1.3352447986602782, "loss": 0.2050051286816597, "time": 3.1037906646728515, "epoch": 15, "memory": 61145, "step": 20760} +{"lr": 0.0008323684216765164, "data_time": 1.4990469217300415, "loss": 0.20038449317216872, "time": 3.2603516817092895, "epoch": 15, "memory": 61145, "step": 20860} +{"lr": 0.0008323684216765164, "data_time": 1.338820719718933, "loss": 0.20078101605176926, "time": 3.153957962989807, "epoch": 15, "memory": 61145, "step": 20960} +{"lr": 0.0008323684216765164, "data_time": 1.3476698637008666, "loss": 0.2015976220369339, "time": 3.1245984077453612, "epoch": 15, "memory": 61145, "step": 21060} +{"lr": 0.0008323684216765164, "data_time": 1.4055803537368774, "loss": 0.2011843204498291, "time": 3.1803860425949098, "epoch": 15, "memory": 61145, "step": 21160} +{"lr": 0.0008323684216765164, "data_time": 1.3900239706039428, "loss": 0.20184639394283294, "time": 3.1857438325881957, "epoch": 15, "memory": 61145, "step": 21260} +{"lr": 0.0008323684216765164, "data_time": 1.3086981058120728, "loss": 0.2033592864871025, "time": 3.1122119426727295, "epoch": 15, "memory": 61145, "step": 21360} +{"lr": 0.0008323684216765164, "data_time": 1.3110873699188232, "loss": 0.20138298124074935, "time": 3.0882763385772707, "epoch": 15, "memory": 61145, "step": 21460} +{"lr": 0.0008323684216765164, "data_time": 1.3130765438079834, "loss": 0.20135413855314255, "time": 3.0850628852844237, "epoch": 15, "memory": 61145, "step": 21560} +{"accuracy/top1": 94.0155258178711, "data_time": 0.23996667300953584, "time": 0.4683537342969109, "step": 15} +{"lr": 0.0005943219715328379, "data_time": 1.0311064720153809, "loss": 0.20026372075080873, "time": 3.045671558380127, "epoch": 16, "memory": 61145, "step": 21700} +{"lr": 0.0005943219715328379, "data_time": 0.2846883773803711, "loss": 0.20036861002445222, "time": 3.1365254402160643, "epoch": 16, "memory": 61145, "step": 21800} +{"lr": 0.0005943219715328379, "data_time": 1.3309248208999633, "loss": 0.20091276466846467, "time": 3.660755920410156, "epoch": 16, "memory": 61145, "step": 21900} +{"lr": 0.0005943219715328379, "data_time": 1.4055038690567017, "loss": 0.20040241926908492, "time": 3.161810564994812, "epoch": 16, "memory": 61145, "step": 22000} +{"lr": 0.0005943219715328379, "data_time": 1.7095362663269043, "loss": 0.20026686489582063, "time": 3.460629391670227, "epoch": 16, "memory": 61145, "step": 22100} +{"lr": 0.0005943219715328379, "data_time": 1.5108324766159058, "loss": 0.20159268528223037, "time": 3.275766682624817, "epoch": 16, "memory": 61145, "step": 22200} +{"lr": 0.0005943219715328379, "data_time": 1.4146517992019654, "loss": 0.2009176418185234, "time": 3.194920015335083, "epoch": 16, "memory": 61145, "step": 22300} +{"lr": 0.0005943219715328379, "data_time": 1.3603282928466798, "loss": 0.20006120651960374, "time": 3.1361812114715577, "epoch": 16, "memory": 61145, "step": 22400} +{"lr": 0.0005943219715328379, "data_time": 1.4580895900726318, "loss": 0.19959874004125594, "time": 3.2373690605163574, "epoch": 16, "memory": 61145, "step": 22500} +{"lr": 0.0005943219715328379, "data_time": 1.4217192649841308, "loss": 0.2007407784461975, "time": 3.1934733152389527, "epoch": 16, "memory": 61145, "step": 22600} +{"lr": 0.0005943219715328379, "data_time": 1.5334298849105834, "loss": 0.2016201823949814, "time": 3.316812753677368, "epoch": 16, "memory": 61145, "step": 22700} +{"lr": 0.0005943219715328379, "data_time": 1.387758493423462, "loss": 0.20124318450689316, "time": 3.174627113342285, "epoch": 16, "memory": 61145, "step": 22800} +{"lr": 0.0005943219715328379, "data_time": 1.382840871810913, "loss": 0.20111599266529084, "time": 3.1452425956726073, "epoch": 16, "memory": 61145, "step": 22900} +{"lr": 0.0005943219715328379, "data_time": 1.5014278173446656, "loss": 0.20131610035896302, "time": 3.28324134349823, "epoch": 16, "memory": 61145, "step": 23000} +{"accuracy/top1": 98.22611999511719, "data_time": 0.2630082579220043, "time": 0.4892232137567857, "step": 16} +{"lr": 0.0003910110962219801, "data_time": 1.3749398231506347, "loss": 0.20030642449855804, "time": 3.1291100025177, "epoch": 17, "memory": 61145, "step": 23140} +{"lr": 0.0003910110962219801, "data_time": 1.6813330888748168, "loss": 0.1995615243911743, "time": 3.4321943283081056, "epoch": 17, "memory": 61145, "step": 23240} +{"lr": 0.0003910110962219801, "data_time": 1.2939404010772706, "loss": 0.20007950067520142, "time": 3.448142981529236, "epoch": 17, "memory": 61145, "step": 23340} +{"lr": 0.0003910110962219801, "data_time": 1.5063703536987305, "loss": 0.20023534446954727, "time": 3.2638447761535643, "epoch": 17, "memory": 61145, "step": 23440} +{"lr": 0.0003910110962219801, "data_time": 1.379599666595459, "loss": 0.20158216059207917, "time": 3.1671363353729247, "epoch": 17, "memory": 61145, "step": 23540} +{"lr": 0.0003910110962219801, "data_time": 1.2577183485031127, "loss": 0.19966453313827515, "time": 3.0302154541015627, "epoch": 17, "memory": 61145, "step": 23640} +{"lr": 0.0003910110962219801, "data_time": 1.3265140771865844, "loss": 0.19958391636610032, "time": 3.1236394166946413, "epoch": 17, "memory": 61145, "step": 23740} +{"lr": 0.0003910110962219801, "data_time": 1.1868301391601563, "loss": 0.19993222057819365, "time": 2.970109748840332, "epoch": 17, "memory": 61145, "step": 23840} +{"lr": 0.0003910110962219801, "data_time": 1.2685500621795653, "loss": 0.2004125952720642, "time": 3.0263702869415283, "epoch": 17, "memory": 61145, "step": 23940} diff --git a/convnext_tiny_4xb1024_4e-3lr_5m/20230531_172740/vis_data/vis_image/0000000.png_1.png b/convnext_tiny_4xb1024_4e-3lr_5m/20230531_172740/vis_data/vis_image/0000000.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..9f499cda61efb33d102a61aee957b611dc504144 Binary files /dev/null and b/convnext_tiny_4xb1024_4e-3lr_5m/20230531_172740/vis_data/vis_image/0000000.png_1.png differ diff --git a/convnext_tiny_4xb1024_4e-3lr_5m/20230531_172740/vis_data/vis_image/0000000.png_10.png b/convnext_tiny_4xb1024_4e-3lr_5m/20230531_172740/vis_data/vis_image/0000000.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..dbd00a21bdda4f44b04908f3673b5c490c33ef18 Binary files /dev/null and b/convnext_tiny_4xb1024_4e-3lr_5m/20230531_172740/vis_data/vis_image/0000000.png_10.png differ diff --git a/convnext_tiny_4xb1024_4e-3lr_5m/20230531_172740/vis_data/vis_image/0000000.png_11.png b/convnext_tiny_4xb1024_4e-3lr_5m/20230531_172740/vis_data/vis_image/0000000.png_11.png new file mode 100644 index 0000000000000000000000000000000000000000..1be5dd4998bb6baa8c3ce36c1a4d861cdaf6b2c7 Binary files /dev/null and b/convnext_tiny_4xb1024_4e-3lr_5m/20230531_172740/vis_data/vis_image/0000000.png_11.png differ diff --git a/convnext_tiny_4xb1024_4e-3lr_5m/20230531_172740/vis_data/vis_image/0000000.png_12.png b/convnext_tiny_4xb1024_4e-3lr_5m/20230531_172740/vis_data/vis_image/0000000.png_12.png new file mode 100644 index 0000000000000000000000000000000000000000..ae323dfb2f99afa256eed926b072e5d438f42b32 Binary files /dev/null and b/convnext_tiny_4xb1024_4e-3lr_5m/20230531_172740/vis_data/vis_image/0000000.png_12.png differ diff --git a/convnext_tiny_4xb1024_4e-3lr_5m/20230531_172740/vis_data/vis_image/0000000.png_13.png b/convnext_tiny_4xb1024_4e-3lr_5m/20230531_172740/vis_data/vis_image/0000000.png_13.png new file mode 100644 index 0000000000000000000000000000000000000000..359ab8b499d0c0d2b132b04283ae399f01124a8b Binary files /dev/null and b/convnext_tiny_4xb1024_4e-3lr_5m/20230531_172740/vis_data/vis_image/0000000.png_13.png differ diff --git a/convnext_tiny_4xb1024_4e-3lr_5m/20230531_172740/vis_data/vis_image/0000000.png_14.png b/convnext_tiny_4xb1024_4e-3lr_5m/20230531_172740/vis_data/vis_image/0000000.png_14.png new file mode 100644 index 0000000000000000000000000000000000000000..642fd7a9f39412c9d046f0e4ed7caf12ec84c305 Binary files /dev/null and b/convnext_tiny_4xb1024_4e-3lr_5m/20230531_172740/vis_data/vis_image/0000000.png_14.png differ diff --git a/convnext_tiny_4xb1024_4e-3lr_5m/20230531_172740/vis_data/vis_image/0000000.png_15.png b/convnext_tiny_4xb1024_4e-3lr_5m/20230531_172740/vis_data/vis_image/0000000.png_15.png new file mode 100644 index 0000000000000000000000000000000000000000..33117125e77c9142d11af1dae31710a85f7e591a Binary files /dev/null and b/convnext_tiny_4xb1024_4e-3lr_5m/20230531_172740/vis_data/vis_image/0000000.png_15.png differ diff --git a/convnext_tiny_4xb1024_4e-3lr_5m/20230531_172740/vis_data/vis_image/0000000.png_16.png b/convnext_tiny_4xb1024_4e-3lr_5m/20230531_172740/vis_data/vis_image/0000000.png_16.png new file mode 100644 index 0000000000000000000000000000000000000000..618fd617b542a468231948eb6d7afb6266c85540 Binary files /dev/null and b/convnext_tiny_4xb1024_4e-3lr_5m/20230531_172740/vis_data/vis_image/0000000.png_16.png differ diff --git a/convnext_tiny_4xb1024_4e-3lr_5m/20230531_172740/vis_data/vis_image/0000000.png_2.png b/convnext_tiny_4xb1024_4e-3lr_5m/20230531_172740/vis_data/vis_image/0000000.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..a8ea7e04408b929559153ff7ebf279bda0e86429 Binary files /dev/null and b/convnext_tiny_4xb1024_4e-3lr_5m/20230531_172740/vis_data/vis_image/0000000.png_2.png differ diff --git a/convnext_tiny_4xb1024_4e-3lr_5m/20230531_172740/vis_data/vis_image/0000000.png_3.png b/convnext_tiny_4xb1024_4e-3lr_5m/20230531_172740/vis_data/vis_image/0000000.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..1ac34d4a4ac3fa4def08506155df955df0db01ab Binary files /dev/null and b/convnext_tiny_4xb1024_4e-3lr_5m/20230531_172740/vis_data/vis_image/0000000.png_3.png differ diff --git a/convnext_tiny_4xb1024_4e-3lr_5m/20230531_172740/vis_data/vis_image/0000000.png_4.png b/convnext_tiny_4xb1024_4e-3lr_5m/20230531_172740/vis_data/vis_image/0000000.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..e3aaa512873642c40167ca56ce1e7a49ac713f90 Binary files /dev/null and b/convnext_tiny_4xb1024_4e-3lr_5m/20230531_172740/vis_data/vis_image/0000000.png_4.png differ diff --git a/convnext_tiny_4xb1024_4e-3lr_5m/20230531_172740/vis_data/vis_image/0000000.png_5.png b/convnext_tiny_4xb1024_4e-3lr_5m/20230531_172740/vis_data/vis_image/0000000.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..fa721e057afb3c14016b7443e65231b49ceb7d0b Binary files /dev/null and b/convnext_tiny_4xb1024_4e-3lr_5m/20230531_172740/vis_data/vis_image/0000000.png_5.png differ diff --git a/convnext_tiny_4xb1024_4e-3lr_5m/20230531_172740/vis_data/vis_image/0000000.png_6.png b/convnext_tiny_4xb1024_4e-3lr_5m/20230531_172740/vis_data/vis_image/0000000.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..ca22324d2315160190fa6d22fd98a2d9a36626fa Binary files /dev/null and b/convnext_tiny_4xb1024_4e-3lr_5m/20230531_172740/vis_data/vis_image/0000000.png_6.png differ diff --git a/convnext_tiny_4xb1024_4e-3lr_5m/20230531_172740/vis_data/vis_image/0000000.png_7.png b/convnext_tiny_4xb1024_4e-3lr_5m/20230531_172740/vis_data/vis_image/0000000.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..a9e9e345b5f3bfd64830209c3175c487d039c749 Binary files /dev/null and b/convnext_tiny_4xb1024_4e-3lr_5m/20230531_172740/vis_data/vis_image/0000000.png_7.png differ diff --git a/convnext_tiny_4xb1024_4e-3lr_5m/20230531_172740/vis_data/vis_image/0000000.png_8.png b/convnext_tiny_4xb1024_4e-3lr_5m/20230531_172740/vis_data/vis_image/0000000.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..a9e9e345b5f3bfd64830209c3175c487d039c749 Binary files /dev/null and b/convnext_tiny_4xb1024_4e-3lr_5m/20230531_172740/vis_data/vis_image/0000000.png_8.png differ diff --git a/convnext_tiny_4xb1024_4e-3lr_5m/20230531_172740/vis_data/vis_image/0000000.png_9.png b/convnext_tiny_4xb1024_4e-3lr_5m/20230531_172740/vis_data/vis_image/0000000.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..1be5dd4998bb6baa8c3ce36c1a4d861cdaf6b2c7 Binary files /dev/null and b/convnext_tiny_4xb1024_4e-3lr_5m/20230531_172740/vis_data/vis_image/0000000.png_9.png differ diff --git a/convnext_tiny_4xb1024_4e-3lr_5m/convnext_tiny_4xb1024_fake5m.py b/convnext_tiny_4xb1024_4e-3lr_5m/convnext_tiny_4xb1024_fake5m.py new file mode 100644 index 0000000000000000000000000000000000000000..662338ce17f9d9fd29063fd5b315ce1d8866f4f0 --- /dev/null +++ b/convnext_tiny_4xb1024_4e-3lr_5m/convnext_tiny_4xb1024_fake5m.py @@ -0,0 +1,170 @@ +optim_wrapper = dict( + optimizer=dict( + type='AdamW', + lr=0.004, + weight_decay=0.05, + eps=1e-08, + betas=(0.9, 0.999), + _scope_='mmpretrain'), + paramwise_cfg=dict( + norm_decay_mult=0.0, + bias_decay_mult=0.0, + flat_decay_mult=0.0, + custom_keys=dict({ + '.absolute_pos_embed': dict(decay_mult=0.0), + '.relative_position_bias_table': dict(decay_mult=0.0) + })), + type='AmpOptimWrapper', + dtype='bfloat16', + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=True, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=20, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=4096) +model = dict( + type='ImageClassifier', + backbone=dict(type='ConvNeXt', arch='tiny', drop_path_rate=0.1), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=768, + loss=dict( + type='LabelSmoothLoss', label_smooth_val=0.1, mode='original'), + init_cfg=None), + init_cfg=dict( + type='TruncNormal', layer=['Conv2d', 'Linear'], std=0.02, bias=0.0), + train_cfg=None) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=1024, + num_workers=10, + dataset=dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stablediffusionV1-5R2-dpmsolver-25-5m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/convnext_tiny_4xb1024_4e-3lr_5m' diff --git a/convnext_tiny_4xb1024_4e-3lr_5m/epoch_1.pth b/convnext_tiny_4xb1024_4e-3lr_5m/epoch_1.pth new file mode 100644 index 0000000000000000000000000000000000000000..78389d7b9eac423162cb862164790ab63984c207 --- /dev/null +++ b/convnext_tiny_4xb1024_4e-3lr_5m/epoch_1.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:f8388c515f975ba06c52fc827fa9b139e9284f61596b9ee46fc57d4357d3fdaf +size 445571747 diff --git a/convnext_tiny_4xb1024_4e-3lr_5m/epoch_10.pth b/convnext_tiny_4xb1024_4e-3lr_5m/epoch_10.pth new file mode 100644 index 0000000000000000000000000000000000000000..c24d072ed34d5982fcbe67820fae4383e7de1f4c --- /dev/null +++ b/convnext_tiny_4xb1024_4e-3lr_5m/epoch_10.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:b464a4cf02a290dfe5629a3ab73b380e437c51a66f38ca0d9ec4ae5a596227cf +size 446519139 diff --git a/convnext_tiny_4xb1024_4e-3lr_5m/epoch_11.pth b/convnext_tiny_4xb1024_4e-3lr_5m/epoch_11.pth new file mode 100644 index 0000000000000000000000000000000000000000..7548ac1ac4f8cbe979a2c71da9dd01696db93d7e --- /dev/null +++ b/convnext_tiny_4xb1024_4e-3lr_5m/epoch_11.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ecac8a8e8dab8219e4fd7a2020d9835afb8dfc6597a3016ef7f5db54bc7b7539 +size 446624675 diff --git a/convnext_tiny_4xb1024_4e-3lr_5m/epoch_12.pth b/convnext_tiny_4xb1024_4e-3lr_5m/epoch_12.pth new file mode 100644 index 0000000000000000000000000000000000000000..65e944ac0d49c72207240bfe7baaafd8b0aff065 --- /dev/null +++ b/convnext_tiny_4xb1024_4e-3lr_5m/epoch_12.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:88610358fcf5973a2a950a6fc7e519c390b355b585c693f6d87c189fc68a620d +size 446732963 diff --git a/convnext_tiny_4xb1024_4e-3lr_5m/epoch_13.pth b/convnext_tiny_4xb1024_4e-3lr_5m/epoch_13.pth new file mode 100644 index 0000000000000000000000000000000000000000..6943fbbea3ebaaeff7bb06584eb3075ad7105f58 --- /dev/null +++ b/convnext_tiny_4xb1024_4e-3lr_5m/epoch_13.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:751d3500d36ecc33b2b39f1e74d764c621c0b0539d1ec13c6423ed08ef93d065 +size 446839971 diff --git a/convnext_tiny_4xb1024_4e-3lr_5m/epoch_14.pth b/convnext_tiny_4xb1024_4e-3lr_5m/epoch_14.pth new file mode 100644 index 0000000000000000000000000000000000000000..b47c0cbc32e7e9b47db5790c8cc349abd38c6bae --- /dev/null +++ b/convnext_tiny_4xb1024_4e-3lr_5m/epoch_14.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:656cce9fb059bb2ed8c5dfac524361ef8586df10e93763156b173a03b22c988d +size 446945699 diff --git a/convnext_tiny_4xb1024_4e-3lr_5m/epoch_15.pth b/convnext_tiny_4xb1024_4e-3lr_5m/epoch_15.pth new file mode 100644 index 0000000000000000000000000000000000000000..3332b245de85a7cd2ad2205f8c0c8497a18fec72 --- /dev/null +++ b/convnext_tiny_4xb1024_4e-3lr_5m/epoch_15.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ab11b68bda6d7bb3d5ba1c96e4bf2489362f9cd8f2de3040f3761dda68e76121 +size 447049827 diff --git a/convnext_tiny_4xb1024_4e-3lr_5m/epoch_16.pth b/convnext_tiny_4xb1024_4e-3lr_5m/epoch_16.pth new file mode 100644 index 0000000000000000000000000000000000000000..94eda72fec3f6ac0076c5cf764add8d66bead146 --- /dev/null +++ b/convnext_tiny_4xb1024_4e-3lr_5m/epoch_16.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:acd05c8458f8345ad94d38bf2102f38a6786ae78e03fcc11dac114a5a5ce0694 +size 447155619 diff --git a/convnext_tiny_4xb1024_4e-3lr_5m/epoch_2.pth b/convnext_tiny_4xb1024_4e-3lr_5m/epoch_2.pth new file mode 100644 index 0000000000000000000000000000000000000000..f79749e3ce3b471ac6da3fc582d6e887a13da2e1 --- /dev/null +++ b/convnext_tiny_4xb1024_4e-3lr_5m/epoch_2.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:4d188c759762afacd425af1c27dada30d3172b91787e6af4675ce7b658a09d7d +size 445678307 diff --git a/convnext_tiny_4xb1024_4e-3lr_5m/epoch_3.pth b/convnext_tiny_4xb1024_4e-3lr_5m/epoch_3.pth new file mode 100644 index 0000000000000000000000000000000000000000..f1ddaa3904b3a482b9154a6a90e430975f97e749 --- /dev/null +++ b/convnext_tiny_4xb1024_4e-3lr_5m/epoch_3.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ff95ae95939e16caa3b1e8582b51fb73c8dfa5590c2edf1179492f447158c8f2 +size 445782179 diff --git a/convnext_tiny_4xb1024_4e-3lr_5m/epoch_4.pth b/convnext_tiny_4xb1024_4e-3lr_5m/epoch_4.pth new file mode 100644 index 0000000000000000000000000000000000000000..32643011f69c5a6d5a59bd88a61738b98f3bb1d6 --- /dev/null +++ b/convnext_tiny_4xb1024_4e-3lr_5m/epoch_4.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:40e4015a76756e95c5aa4bf2a54fab7a463b618b1abb569465a5ca6f94348ee9 +size 445887651 diff --git a/convnext_tiny_4xb1024_4e-3lr_5m/epoch_5.pth b/convnext_tiny_4xb1024_4e-3lr_5m/epoch_5.pth new file mode 100644 index 0000000000000000000000000000000000000000..d1c60100a6e94b258fafb4d4146cf14806959ed0 --- /dev/null +++ b/convnext_tiny_4xb1024_4e-3lr_5m/epoch_5.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:7aba23498ef6c3a50a6ed12eb078cca3b9c9288c521fa4bcbc68a9ae17f1b7da +size 445996067 diff --git a/convnext_tiny_4xb1024_4e-3lr_5m/epoch_6.pth b/convnext_tiny_4xb1024_4e-3lr_5m/epoch_6.pth new file mode 100644 index 0000000000000000000000000000000000000000..27ec4a3d58465b161e0736b4415441b0e048cd0a --- /dev/null +++ b/convnext_tiny_4xb1024_4e-3lr_5m/epoch_6.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:a50513015cb2df39dc85159db9c9c525611cdfe2f837a13fe0aefd56a4c1b153 +size 446101539 diff --git a/convnext_tiny_4xb1024_4e-3lr_5m/epoch_7.pth b/convnext_tiny_4xb1024_4e-3lr_5m/epoch_7.pth new file mode 100644 index 0000000000000000000000000000000000000000..e8f10c3fa9efb9d6a8d449c2c3f269e3a73f6693 --- /dev/null +++ b/convnext_tiny_4xb1024_4e-3lr_5m/epoch_7.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:cb17710ad964c15358045f36b76dfbdc75090af2b1d2e6138cac370e3ec93550 +size 446209955 diff --git a/convnext_tiny_4xb1024_4e-3lr_5m/epoch_8.pth b/convnext_tiny_4xb1024_4e-3lr_5m/epoch_8.pth new file mode 100644 index 0000000000000000000000000000000000000000..e807c2387fcf35aa1533b5906dc46a06e26b4dd4 --- /dev/null +++ b/convnext_tiny_4xb1024_4e-3lr_5m/epoch_8.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:654d3f4c6b62dbcd4b2c34a20858e7612d97a526d2d85463e95ed5871484a191 +size 446312419 diff --git a/convnext_tiny_4xb1024_4e-3lr_5m/epoch_9.pth b/convnext_tiny_4xb1024_4e-3lr_5m/epoch_9.pth new file mode 100644 index 0000000000000000000000000000000000000000..d759f8ddd5ef5fbf00988fc24f4d162da54d0cc9 --- /dev/null +++ b/convnext_tiny_4xb1024_4e-3lr_5m/epoch_9.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c95ebccf4b70931f8bcdcfca2b3605de1e5f2527e46241c0cad5f213f6607577 +size 446415075 diff --git a/convnext_tiny_4xb1024_4e-3lr_5m/last_checkpoint b/convnext_tiny_4xb1024_4e-3lr_5m/last_checkpoint new file mode 100644 index 0000000000000000000000000000000000000000..4b70671e9db2ea893a4d0f6082a1c4a35d5b4600 --- /dev/null +++ b/convnext_tiny_4xb1024_4e-3lr_5m/last_checkpoint @@ -0,0 +1 @@ +/mnt/petrelfs/luzeyu/workspace/fakebench/mmpretrain/workdir/convnext_tiny_4xb1024_4e-3lr_5m/epoch_16.pth \ No newline at end of file diff --git a/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1/20230606_050038/20230606_050038.log b/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1/20230606_050038/20230606_050038.log new file mode 100644 index 0000000000000000000000000000000000000000..88c6fec483bef63973d15d343ae6bf229f5b6f2b --- /dev/null +++ b/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1/20230606_050038/20230606_050038.log @@ -0,0 +1,863 @@ +2023/06/06 05:00:45 - mmengine - INFO - +------------------------------------------------------------ +System environment: + sys.platform: linux + Python: 3.10.9 (main, Mar 8 2023, 10:47:38) [GCC 11.2.0] + CUDA available: True + numpy_random_seed: 2008498557 + GPU 0,1: NVIDIA A100-SXM4-80GB + CUDA_HOME: /mnt/petrelfs/share/cuda-11.6 + NVCC: Cuda compilation tools, release 11.6, V11.6.124 + GCC: gcc (GCC) 7.5.0 + PyTorch: 1.13.1 + PyTorch compiling details: PyTorch built with: + - GCC 9.3 + - C++ Version: 201402 + - Intel(R) oneAPI Math Kernel Library Version 2021.4-Product Build 20210904 for Intel(R) 64 architecture applications + - Intel(R) MKL-DNN v2.6.0 (Git Hash 52b5f107dd9cf10910aaa19cb47f3abf9b349815) + - OpenMP 201511 (a.k.a. OpenMP 4.5) + - LAPACK is enabled (usually provided by MKL) + - NNPACK is enabled + - CPU capability usage: AVX2 + - CUDA Runtime 11.6 + - NVCC architecture flags: -gencode;arch=compute_37,code=sm_37;-gencode;arch=compute_50,code=sm_50;-gencode;arch=compute_60,code=sm_60;-gencode;arch=compute_61,code=sm_61;-gencode;arch=compute_70,code=sm_70;-gencode;arch=compute_75,code=sm_75;-gencode;arch=compute_80,code=sm_80;-gencode;arch=compute_86,code=sm_86;-gencode;arch=compute_37,code=compute_37 + - CuDNN 8.3.2 (built against CUDA 11.5) + - Magma 2.6.1 + - Build settings: BLAS_INFO=mkl, BUILD_TYPE=Release, CUDA_VERSION=11.6, CUDNN_VERSION=8.3.2, CXX_COMPILER=/opt/rh/devtoolset-9/root/usr/bin/c++, CXX_FLAGS= -fabi-version=11 -Wno-deprecated -fvisibility-inlines-hidden -DUSE_PTHREADPOOL -fopenmp -DNDEBUG -DUSE_KINETO -DUSE_FBGEMM -DUSE_QNNPACK -DUSE_PYTORCH_QNNPACK -DUSE_XNNPACK -DSYMBOLICATE_MOBILE_DEBUG_HANDLE -DEDGE_PROFILER_USE_KINETO -O2 -fPIC -Wno-narrowing -Wall -Wextra -Werror=return-type -Werror=non-virtual-dtor -Wno-missing-field-initializers -Wno-type-limits -Wno-array-bounds -Wno-unknown-pragmas -Wunused-local-typedefs -Wno-unused-parameter -Wno-unused-function -Wno-unused-result -Wno-strict-overflow -Wno-strict-aliasing -Wno-error=deprecated-declarations -Wno-stringop-overflow -Wno-psabi -Wno-error=pedantic -Wno-error=redundant-decls -Wno-error=old-style-cast -fdiagnostics-color=always -faligned-new -Wno-unused-but-set-variable -Wno-maybe-uninitialized -fno-math-errno -fno-trapping-math -Werror=format -Werror=cast-function-type -Wno-stringop-overflow, LAPACK_INFO=mkl, PERF_WITH_AVX=1, PERF_WITH_AVX2=1, PERF_WITH_AVX512=1, TORCH_VERSION=1.13.1, USE_CUDA=ON, USE_CUDNN=ON, USE_EXCEPTION_PTR=1, USE_GFLAGS=OFF, USE_GLOG=OFF, USE_MKL=ON, USE_MKLDNN=ON, USE_MPI=OFF, USE_NCCL=ON, USE_NNPACK=ON, USE_OPENMP=ON, USE_ROCM=OFF, + + TorchVision: 0.14.1 + OpenCV: 4.7.0 + MMEngine: 0.7.3 + +Runtime environment: + cudnn_benchmark: True + mp_cfg: {'mp_start_method': 'fork', 'opencv_num_threads': 0} + dist_cfg: {'backend': 'nccl'} + seed: None + deterministic: False + Distributed launcher: slurm + Distributed training: True + GPU number: 2 +------------------------------------------------------------ + +2023/06/06 05:00:49 - mmengine - INFO - Config: +optim_wrapper = dict( + optimizer=dict( + type='SGD', + lr=0.0001, + momentum=0.9, + weight_decay=0.0001, + _scope_='mmpretrain'), + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=False, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=512) +model = dict( + type='ImageClassifier', + backbone=dict( + frozen_stages=2, + type='ResNet', + depth=50, + num_stages=4, + out_indices=(3, ), + style='pytorch', + init_cfg=dict( + type='Pretrained', + checkpoint= + 'https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth', + prefix='backbone')), + neck=dict(type='GlobalAveragePooling'), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=2048, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + topk=1)) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/IF80w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/if-dpmsolver++-50-20w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/cc1m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/if-dpmsolver++-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = dict(type='Accuracy', topk=1) +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/if-dpmsolver++-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = dict(type='Accuracy', topk=1) +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1' + +2023/06/06 05:00:59 - mmengine - INFO - Hooks will be executed in the following order: +before_run: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_load_checkpoint: +(ABOVE_NORMAL) EMAHook + -------------------- +before_train: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_train_epoch: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(NORMAL ) DistSamplerSeedHook + -------------------- +before_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook + -------------------- +after_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +after_train_epoch: +(NORMAL ) IterTimerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_val_epoch: +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook + -------------------- +before_val_iter: +(NORMAL ) IterTimerHook + -------------------- +after_val_iter: +(NORMAL ) IterTimerHook +(NORMAL ) VisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_val_epoch: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_save_checkpoint: +(ABOVE_NORMAL) EMAHook + -------------------- +after_train: +(VERY_LOW ) CheckpointHook + -------------------- +before_test_epoch: +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook + -------------------- +before_test_iter: +(NORMAL ) IterTimerHook + -------------------- +after_test_iter: +(NORMAL ) IterTimerHook +(NORMAL ) VisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_test_epoch: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_run: +(BELOW_NORMAL) LoggerHook + -------------------- +2023/06/06 05:01:17 - mmengine - INFO - load backbone in model from: https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth +Name of parameter - Initialization information + +backbone.conv1.weight - torch.Size([64, 3, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.bn1.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.bn1.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.conv1.weight - torch.Size([64, 64, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.bn1.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.bn1.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.conv2.weight - torch.Size([64, 64, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.bn2.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.bn2.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.conv3.weight - torch.Size([256, 64, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.bn3.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.bn3.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.downsample.0.weight - torch.Size([256, 64, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.downsample.1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.downsample.1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.conv1.weight - torch.Size([64, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.bn1.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.bn1.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.conv2.weight - torch.Size([64, 64, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.bn2.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.bn2.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.conv3.weight - torch.Size([256, 64, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.bn3.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.bn3.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.conv1.weight - torch.Size([64, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.bn1.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.bn1.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.conv2.weight - torch.Size([64, 64, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.bn2.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.bn2.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.conv3.weight - torch.Size([256, 64, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.bn3.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.bn3.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.conv1.weight - torch.Size([128, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.bn1.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.bn1.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.conv2.weight - torch.Size([128, 128, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.bn2.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.bn2.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.conv3.weight - torch.Size([512, 128, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.bn3.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.bn3.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.downsample.0.weight - torch.Size([512, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.downsample.1.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.downsample.1.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.conv1.weight - torch.Size([128, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.bn1.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.bn1.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.conv2.weight - torch.Size([128, 128, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.bn2.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.bn2.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.conv3.weight - torch.Size([512, 128, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.bn3.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.bn3.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.conv1.weight - torch.Size([128, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.bn1.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.bn1.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.conv2.weight - torch.Size([128, 128, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.bn2.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.bn2.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.conv3.weight - torch.Size([512, 128, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.bn3.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.bn3.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.conv1.weight - torch.Size([128, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.bn1.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.bn1.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.conv2.weight - torch.Size([128, 128, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.bn2.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.bn2.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.conv3.weight - torch.Size([512, 128, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.bn3.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.bn3.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.conv1.weight - torch.Size([256, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.bn1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.bn1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.conv2.weight - torch.Size([256, 256, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.bn2.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.bn2.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.conv3.weight - torch.Size([1024, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.bn3.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.bn3.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.downsample.0.weight - torch.Size([1024, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.downsample.1.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.downsample.1.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.conv1.weight - torch.Size([256, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.bn1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.bn1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.conv2.weight - torch.Size([256, 256, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.bn2.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.bn2.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.conv3.weight - torch.Size([1024, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.bn3.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.bn3.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.conv1.weight - torch.Size([256, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.bn1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.bn1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.conv2.weight - torch.Size([256, 256, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.bn2.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.bn2.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.conv3.weight - torch.Size([1024, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.bn3.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.bn3.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.conv1.weight - torch.Size([256, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.bn1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.bn1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.conv2.weight - torch.Size([256, 256, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.bn2.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.bn2.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.conv3.weight - torch.Size([1024, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.bn3.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.bn3.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.conv1.weight - torch.Size([256, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.bn1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.bn1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.conv2.weight - torch.Size([256, 256, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.bn2.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.bn2.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.conv3.weight - torch.Size([1024, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.bn3.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.bn3.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.conv1.weight - torch.Size([256, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.bn1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.bn1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.conv2.weight - torch.Size([256, 256, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.bn2.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.bn2.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.conv3.weight - torch.Size([1024, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.bn3.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.bn3.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.conv1.weight - torch.Size([512, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.bn1.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.bn1.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.conv2.weight - torch.Size([512, 512, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.bn2.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.bn2.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.conv3.weight - torch.Size([2048, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.bn3.weight - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.bn3.bias - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.downsample.0.weight - torch.Size([2048, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.downsample.1.weight - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.downsample.1.bias - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.conv1.weight - torch.Size([512, 2048, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.bn1.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.bn1.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.conv2.weight - torch.Size([512, 512, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.bn2.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.bn2.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.conv3.weight - torch.Size([2048, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.bn3.weight - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.bn3.bias - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.conv1.weight - torch.Size([512, 2048, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.bn1.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.bn1.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.conv2.weight - torch.Size([512, 512, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.bn2.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.bn2.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.conv3.weight - torch.Size([2048, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.bn3.weight - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.bn3.bias - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +head.fc.weight - torch.Size([2, 2048]): +NormalInit: mean=0, std=0.01, bias=0 + +head.fc.bias - torch.Size([2]): +NormalInit: mean=0, std=0.01, bias=0 +2023/06/06 05:01:17 - mmengine - WARNING - "FileClient" will be deprecated in future. Please use io functions in https://mmengine.readthedocs.io/en/latest/api/fileio.html#file-io +2023/06/06 05:01:17 - mmengine - WARNING - "HardDiskBackend" is the alias of "LocalBackend" and the former will be deprecated in future. +2023/06/06 05:01:17 - mmengine - INFO - Checkpoints will be saved to /mnt/petrelfs/luzeyu/workspace/fakebench/mmpretrain/workdir/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1. diff --git a/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1/20230606_050038/vis_data/config.py b/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1/20230606_050038/vis_data/config.py new file mode 100644 index 0000000000000000000000000000000000000000..9fa82a87719e4caa4a661d2572ccc1b988bdb43d --- /dev/null +++ b/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1/20230606_050038/vis_data/config.py @@ -0,0 +1,247 @@ +optim_wrapper = dict( + optimizer=dict( + type='SGD', + lr=0.0001, + momentum=0.9, + weight_decay=0.0001, + _scope_='mmpretrain'), + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=False, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=512) +model = dict( + type='ImageClassifier', + backbone=dict( + frozen_stages=2, + type='ResNet', + depth=50, + num_stages=4, + out_indices=(3, ), + style='pytorch', + init_cfg=dict( + type='Pretrained', + checkpoint= + 'https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth', + prefix='backbone')), + neck=dict(type='GlobalAveragePooling'), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=2048, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + topk=1)) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/IF80w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/if-dpmsolver++-50-20w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/cc1m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/if-dpmsolver++-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = dict(type='Accuracy', topk=1) +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/if-dpmsolver++-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = dict(type='Accuracy', topk=1) +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1' diff --git a/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1/20230606_050038/vis_data/events.out.tfevents.1685998855.SH-IDC1-10-140-24-124.191231.0 b/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1/20230606_050038/vis_data/events.out.tfevents.1685998855.SH-IDC1-10-140-24-124.191231.0 new file mode 100644 index 0000000000000000000000000000000000000000..583642bda36f2da8ed52dbdd0080bce4c8b2cdaf --- /dev/null +++ b/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1/20230606_050038/vis_data/events.out.tfevents.1685998855.SH-IDC1-10-140-24-124.191231.0 @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:64401208c2ac465e433f8538eb556f32fdf9abdf00eed15afd6734b543c59127 +size 9590 diff --git a/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1/20230606_050639/20230606_050639.log b/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1/20230606_050639/20230606_050639.log new file mode 100644 index 0000000000000000000000000000000000000000..fc5667be8acb3e2210b062722fadea52bbfa1b46 --- /dev/null +++ b/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1/20230606_050639/20230606_050639.log @@ -0,0 +1,863 @@ +2023/06/06 05:06:44 - mmengine - INFO - +------------------------------------------------------------ +System environment: + sys.platform: linux + Python: 3.10.9 (main, Mar 8 2023, 10:47:38) [GCC 11.2.0] + CUDA available: True + numpy_random_seed: 912363288 + GPU 0,1: NVIDIA A100-SXM4-80GB + CUDA_HOME: /mnt/petrelfs/share/cuda-11.6 + NVCC: Cuda compilation tools, release 11.6, V11.6.124 + GCC: gcc (GCC) 7.5.0 + PyTorch: 1.13.1 + PyTorch compiling details: PyTorch built with: + - GCC 9.3 + - C++ Version: 201402 + - Intel(R) oneAPI Math Kernel Library Version 2021.4-Product Build 20210904 for Intel(R) 64 architecture applications + - Intel(R) MKL-DNN v2.6.0 (Git Hash 52b5f107dd9cf10910aaa19cb47f3abf9b349815) + - OpenMP 201511 (a.k.a. OpenMP 4.5) + - LAPACK is enabled (usually provided by MKL) + - NNPACK is enabled + - CPU capability usage: AVX2 + - CUDA Runtime 11.6 + - NVCC architecture flags: -gencode;arch=compute_37,code=sm_37;-gencode;arch=compute_50,code=sm_50;-gencode;arch=compute_60,code=sm_60;-gencode;arch=compute_61,code=sm_61;-gencode;arch=compute_70,code=sm_70;-gencode;arch=compute_75,code=sm_75;-gencode;arch=compute_80,code=sm_80;-gencode;arch=compute_86,code=sm_86;-gencode;arch=compute_37,code=compute_37 + - CuDNN 8.3.2 (built against CUDA 11.5) + - Magma 2.6.1 + - Build settings: BLAS_INFO=mkl, BUILD_TYPE=Release, CUDA_VERSION=11.6, CUDNN_VERSION=8.3.2, CXX_COMPILER=/opt/rh/devtoolset-9/root/usr/bin/c++, CXX_FLAGS= -fabi-version=11 -Wno-deprecated -fvisibility-inlines-hidden -DUSE_PTHREADPOOL -fopenmp -DNDEBUG -DUSE_KINETO -DUSE_FBGEMM -DUSE_QNNPACK -DUSE_PYTORCH_QNNPACK -DUSE_XNNPACK -DSYMBOLICATE_MOBILE_DEBUG_HANDLE -DEDGE_PROFILER_USE_KINETO -O2 -fPIC -Wno-narrowing -Wall -Wextra -Werror=return-type -Werror=non-virtual-dtor -Wno-missing-field-initializers -Wno-type-limits -Wno-array-bounds -Wno-unknown-pragmas -Wunused-local-typedefs -Wno-unused-parameter -Wno-unused-function -Wno-unused-result -Wno-strict-overflow -Wno-strict-aliasing -Wno-error=deprecated-declarations -Wno-stringop-overflow -Wno-psabi -Wno-error=pedantic -Wno-error=redundant-decls -Wno-error=old-style-cast -fdiagnostics-color=always -faligned-new -Wno-unused-but-set-variable -Wno-maybe-uninitialized -fno-math-errno -fno-trapping-math -Werror=format -Werror=cast-function-type -Wno-stringop-overflow, LAPACK_INFO=mkl, PERF_WITH_AVX=1, PERF_WITH_AVX2=1, PERF_WITH_AVX512=1, TORCH_VERSION=1.13.1, USE_CUDA=ON, USE_CUDNN=ON, USE_EXCEPTION_PTR=1, USE_GFLAGS=OFF, USE_GLOG=OFF, USE_MKL=ON, USE_MKLDNN=ON, USE_MPI=OFF, USE_NCCL=ON, USE_NNPACK=ON, USE_OPENMP=ON, USE_ROCM=OFF, + + TorchVision: 0.14.1 + OpenCV: 4.7.0 + MMEngine: 0.7.3 + +Runtime environment: + cudnn_benchmark: True + mp_cfg: {'mp_start_method': 'fork', 'opencv_num_threads': 0} + dist_cfg: {'backend': 'nccl'} + seed: None + deterministic: False + Distributed launcher: slurm + Distributed training: True + GPU number: 2 +------------------------------------------------------------ + +2023/06/06 05:06:48 - mmengine - INFO - Config: +optim_wrapper = dict( + optimizer=dict( + type='SGD', + lr=0.0001, + momentum=0.9, + weight_decay=0.0001, + _scope_='mmpretrain'), + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=False, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=512) +model = dict( + type='ImageClassifier', + backbone=dict( + frozen_stages=2, + type='ResNet', + depth=50, + num_stages=4, + out_indices=(3, ), + style='pytorch', + init_cfg=dict( + type='Pretrained', + checkpoint= + 'https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth', + prefix='backbone')), + neck=dict(type='GlobalAveragePooling'), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=2048, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + topk=1)) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/IF80w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/if-dpmsolver++-50-20w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/cc1m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/if-dpmsolver++-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = dict(type='Accuracy', topk=1) +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/if-dpmsolver++-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = dict(type='Accuracy', topk=1) +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1' + +2023/06/06 05:06:59 - mmengine - INFO - Hooks will be executed in the following order: +before_run: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_load_checkpoint: +(ABOVE_NORMAL) EMAHook + -------------------- +before_train: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_train_epoch: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(NORMAL ) DistSamplerSeedHook + -------------------- +before_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook + -------------------- +after_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +after_train_epoch: +(NORMAL ) IterTimerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_val_epoch: +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook + -------------------- +before_val_iter: +(NORMAL ) IterTimerHook + -------------------- +after_val_iter: +(NORMAL ) IterTimerHook +(NORMAL ) VisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_val_epoch: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_save_checkpoint: +(ABOVE_NORMAL) EMAHook + -------------------- +after_train: +(VERY_LOW ) CheckpointHook + -------------------- +before_test_epoch: +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook + -------------------- +before_test_iter: +(NORMAL ) IterTimerHook + -------------------- +after_test_iter: +(NORMAL ) IterTimerHook +(NORMAL ) VisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_test_epoch: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_run: +(BELOW_NORMAL) LoggerHook + -------------------- +2023/06/06 05:07:16 - mmengine - INFO - load backbone in model from: https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth +Name of parameter - Initialization information + +backbone.conv1.weight - torch.Size([64, 3, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.bn1.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.bn1.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.conv1.weight - torch.Size([64, 64, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.bn1.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.bn1.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.conv2.weight - torch.Size([64, 64, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.bn2.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.bn2.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.conv3.weight - torch.Size([256, 64, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.bn3.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.bn3.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.downsample.0.weight - torch.Size([256, 64, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.downsample.1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.downsample.1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.conv1.weight - torch.Size([64, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.bn1.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.bn1.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.conv2.weight - torch.Size([64, 64, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.bn2.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.bn2.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.conv3.weight - torch.Size([256, 64, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.bn3.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.bn3.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.conv1.weight - torch.Size([64, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.bn1.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.bn1.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.conv2.weight - torch.Size([64, 64, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.bn2.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.bn2.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.conv3.weight - torch.Size([256, 64, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.bn3.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.bn3.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.conv1.weight - torch.Size([128, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.bn1.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.bn1.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.conv2.weight - torch.Size([128, 128, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.bn2.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.bn2.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.conv3.weight - torch.Size([512, 128, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.bn3.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.bn3.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.downsample.0.weight - torch.Size([512, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.downsample.1.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.downsample.1.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.conv1.weight - torch.Size([128, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.bn1.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.bn1.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.conv2.weight - torch.Size([128, 128, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.bn2.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.bn2.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.conv3.weight - torch.Size([512, 128, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.bn3.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.bn3.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.conv1.weight - torch.Size([128, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.bn1.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.bn1.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.conv2.weight - torch.Size([128, 128, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.bn2.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.bn2.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.conv3.weight - torch.Size([512, 128, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.bn3.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.bn3.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.conv1.weight - torch.Size([128, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.bn1.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.bn1.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.conv2.weight - torch.Size([128, 128, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.bn2.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.bn2.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.conv3.weight - torch.Size([512, 128, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.bn3.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.bn3.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.conv1.weight - torch.Size([256, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.bn1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.bn1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.conv2.weight - torch.Size([256, 256, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.bn2.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.bn2.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.conv3.weight - torch.Size([1024, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.bn3.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.bn3.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.downsample.0.weight - torch.Size([1024, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.downsample.1.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.downsample.1.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.conv1.weight - torch.Size([256, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.bn1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.bn1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.conv2.weight - torch.Size([256, 256, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.bn2.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.bn2.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.conv3.weight - torch.Size([1024, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.bn3.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.bn3.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.conv1.weight - torch.Size([256, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.bn1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.bn1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.conv2.weight - torch.Size([256, 256, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.bn2.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.bn2.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.conv3.weight - torch.Size([1024, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.bn3.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.bn3.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.conv1.weight - torch.Size([256, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.bn1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.bn1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.conv2.weight - torch.Size([256, 256, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.bn2.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.bn2.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.conv3.weight - torch.Size([1024, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.bn3.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.bn3.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.conv1.weight - torch.Size([256, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.bn1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.bn1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.conv2.weight - torch.Size([256, 256, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.bn2.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.bn2.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.conv3.weight - torch.Size([1024, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.bn3.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.bn3.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.conv1.weight - torch.Size([256, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.bn1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.bn1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.conv2.weight - torch.Size([256, 256, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.bn2.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.bn2.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.conv3.weight - torch.Size([1024, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.bn3.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.bn3.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.conv1.weight - torch.Size([512, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.bn1.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.bn1.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.conv2.weight - torch.Size([512, 512, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.bn2.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.bn2.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.conv3.weight - torch.Size([2048, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.bn3.weight - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.bn3.bias - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.downsample.0.weight - torch.Size([2048, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.downsample.1.weight - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.downsample.1.bias - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.conv1.weight - torch.Size([512, 2048, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.bn1.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.bn1.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.conv2.weight - torch.Size([512, 512, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.bn2.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.bn2.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.conv3.weight - torch.Size([2048, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.bn3.weight - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.bn3.bias - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.conv1.weight - torch.Size([512, 2048, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.bn1.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.bn1.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.conv2.weight - torch.Size([512, 512, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.bn2.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.bn2.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.conv3.weight - torch.Size([2048, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.bn3.weight - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.bn3.bias - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +head.fc.weight - torch.Size([2, 2048]): +NormalInit: mean=0, std=0.01, bias=0 + +head.fc.bias - torch.Size([2]): +NormalInit: mean=0, std=0.01, bias=0 +2023/06/06 05:07:17 - mmengine - WARNING - "FileClient" will be deprecated in future. Please use io functions in https://mmengine.readthedocs.io/en/latest/api/fileio.html#file-io +2023/06/06 05:07:17 - mmengine - WARNING - "HardDiskBackend" is the alias of "LocalBackend" and the former will be deprecated in future. +2023/06/06 05:07:17 - mmengine - INFO - Checkpoints will be saved to /mnt/petrelfs/luzeyu/workspace/fakebench/mmpretrain/workdir/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1. diff --git a/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1/20230606_050639/vis_data/config.py b/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1/20230606_050639/vis_data/config.py new file mode 100644 index 0000000000000000000000000000000000000000..9fa82a87719e4caa4a661d2572ccc1b988bdb43d --- /dev/null +++ b/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1/20230606_050639/vis_data/config.py @@ -0,0 +1,247 @@ +optim_wrapper = dict( + optimizer=dict( + type='SGD', + lr=0.0001, + momentum=0.9, + weight_decay=0.0001, + _scope_='mmpretrain'), + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=False, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=512) +model = dict( + type='ImageClassifier', + backbone=dict( + frozen_stages=2, + type='ResNet', + depth=50, + num_stages=4, + out_indices=(3, ), + style='pytorch', + init_cfg=dict( + type='Pretrained', + checkpoint= + 'https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth', + prefix='backbone')), + neck=dict(type='GlobalAveragePooling'), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=2048, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + topk=1)) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/IF80w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/if-dpmsolver++-50-20w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/cc1m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/if-dpmsolver++-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = dict(type='Accuracy', topk=1) +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/if-dpmsolver++-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = dict(type='Accuracy', topk=1) +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1' diff --git a/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1/20230606_050639/vis_data/events.out.tfevents.1685999214.SH-IDC1-10-140-24-124.212397.0 b/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1/20230606_050639/vis_data/events.out.tfevents.1685999214.SH-IDC1-10-140-24-124.212397.0 new file mode 100644 index 0000000000000000000000000000000000000000..acfe70acab36bed2e372e49952595dd9b7f70d5d --- /dev/null +++ b/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1/20230606_050639/vis_data/events.out.tfevents.1685999214.SH-IDC1-10-140-24-124.212397.0 @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:f47a052177054f947a75d06215ae5e31eb1b3180dbc8c437cc76aae169ee3923 +size 9590 diff --git a/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1/20230606_051539/20230606_051539.log b/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1/20230606_051539/20230606_051539.log new file mode 100644 index 0000000000000000000000000000000000000000..1a6d9fd209a9a561f50f5ab4916226bbbc11ee74 --- /dev/null +++ b/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1/20230606_051539/20230606_051539.log @@ -0,0 +1,1322 @@ +2023/06/06 05:15:42 - mmengine - INFO - +------------------------------------------------------------ +System environment: + sys.platform: linux + Python: 3.10.9 (main, Mar 8 2023, 10:47:38) [GCC 11.2.0] + CUDA available: True + numpy_random_seed: 1434538273 + GPU 0,1: NVIDIA A100-SXM4-80GB + CUDA_HOME: /mnt/petrelfs/share/cuda-11.6 + NVCC: Cuda compilation tools, release 11.6, V11.6.124 + GCC: gcc (GCC) 7.5.0 + PyTorch: 1.13.1 + PyTorch compiling details: PyTorch built with: + - GCC 9.3 + - C++ Version: 201402 + - Intel(R) oneAPI Math Kernel Library Version 2021.4-Product Build 20210904 for Intel(R) 64 architecture applications + - Intel(R) MKL-DNN v2.6.0 (Git Hash 52b5f107dd9cf10910aaa19cb47f3abf9b349815) + - OpenMP 201511 (a.k.a. OpenMP 4.5) + - LAPACK is enabled (usually provided by MKL) + - NNPACK is enabled + - CPU capability usage: AVX2 + - CUDA Runtime 11.6 + - NVCC architecture flags: -gencode;arch=compute_37,code=sm_37;-gencode;arch=compute_50,code=sm_50;-gencode;arch=compute_60,code=sm_60;-gencode;arch=compute_61,code=sm_61;-gencode;arch=compute_70,code=sm_70;-gencode;arch=compute_75,code=sm_75;-gencode;arch=compute_80,code=sm_80;-gencode;arch=compute_86,code=sm_86;-gencode;arch=compute_37,code=compute_37 + - CuDNN 8.3.2 (built against CUDA 11.5) + - Magma 2.6.1 + - Build settings: BLAS_INFO=mkl, BUILD_TYPE=Release, CUDA_VERSION=11.6, CUDNN_VERSION=8.3.2, CXX_COMPILER=/opt/rh/devtoolset-9/root/usr/bin/c++, CXX_FLAGS= -fabi-version=11 -Wno-deprecated -fvisibility-inlines-hidden -DUSE_PTHREADPOOL -fopenmp -DNDEBUG -DUSE_KINETO -DUSE_FBGEMM -DUSE_QNNPACK -DUSE_PYTORCH_QNNPACK -DUSE_XNNPACK -DSYMBOLICATE_MOBILE_DEBUG_HANDLE -DEDGE_PROFILER_USE_KINETO -O2 -fPIC -Wno-narrowing -Wall -Wextra -Werror=return-type -Werror=non-virtual-dtor -Wno-missing-field-initializers -Wno-type-limits -Wno-array-bounds -Wno-unknown-pragmas -Wunused-local-typedefs -Wno-unused-parameter -Wno-unused-function -Wno-unused-result -Wno-strict-overflow -Wno-strict-aliasing -Wno-error=deprecated-declarations -Wno-stringop-overflow -Wno-psabi -Wno-error=pedantic -Wno-error=redundant-decls -Wno-error=old-style-cast -fdiagnostics-color=always -faligned-new -Wno-unused-but-set-variable -Wno-maybe-uninitialized -fno-math-errno -fno-trapping-math -Werror=format -Werror=cast-function-type -Wno-stringop-overflow, LAPACK_INFO=mkl, PERF_WITH_AVX=1, PERF_WITH_AVX2=1, PERF_WITH_AVX512=1, TORCH_VERSION=1.13.1, USE_CUDA=ON, USE_CUDNN=ON, USE_EXCEPTION_PTR=1, USE_GFLAGS=OFF, USE_GLOG=OFF, USE_MKL=ON, USE_MKLDNN=ON, USE_MPI=OFF, USE_NCCL=ON, USE_NNPACK=ON, USE_OPENMP=ON, USE_ROCM=OFF, + + TorchVision: 0.14.1 + OpenCV: 4.7.0 + MMEngine: 0.7.3 + +Runtime environment: + cudnn_benchmark: True + mp_cfg: {'mp_start_method': 'fork', 'opencv_num_threads': 0} + dist_cfg: {'backend': 'nccl'} + seed: None + deterministic: False + Distributed launcher: slurm + Distributed training: True + GPU number: 2 +------------------------------------------------------------ + +2023/06/06 05:15:46 - mmengine - INFO - Config: +optim_wrapper = dict( + optimizer=dict( + type='SGD', + lr=0.0001, + momentum=0.9, + weight_decay=0.0001, + _scope_='mmpretrain'), + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=False, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=512) +model = dict( + type='ImageClassifier', + backbone=dict( + frozen_stages=2, + type='ResNet', + depth=50, + num_stages=4, + out_indices=(3, ), + style='pytorch', + init_cfg=dict( + type='Pretrained', + checkpoint= + 'https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth', + prefix='backbone')), + neck=dict(type='GlobalAveragePooling'), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=2048, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + topk=1)) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/IF80w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/if-dpmsolver++-50-20w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/cc1m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/if-dpmsolver++-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = dict(type='Accuracy', topk=1) +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/if-dpmsolver++-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = dict(type='Accuracy', topk=1) +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1' + +2023/06/06 05:15:57 - mmengine - INFO - Hooks will be executed in the following order: +before_run: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_load_checkpoint: +(ABOVE_NORMAL) EMAHook + -------------------- +before_train: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_train_epoch: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(NORMAL ) DistSamplerSeedHook + -------------------- +before_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook + -------------------- +after_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +after_train_epoch: +(NORMAL ) IterTimerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_val_epoch: +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook + -------------------- +before_val_iter: +(NORMAL ) IterTimerHook + -------------------- +after_val_iter: +(NORMAL ) IterTimerHook +(NORMAL ) VisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_val_epoch: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_save_checkpoint: +(ABOVE_NORMAL) EMAHook + -------------------- +after_train: +(VERY_LOW ) CheckpointHook + -------------------- +before_test_epoch: +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook + -------------------- +before_test_iter: +(NORMAL ) IterTimerHook + -------------------- +after_test_iter: +(NORMAL ) IterTimerHook +(NORMAL ) VisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_test_epoch: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_run: +(BELOW_NORMAL) LoggerHook + -------------------- +2023/06/06 05:16:15 - mmengine - INFO - load backbone in model from: https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth +Name of parameter - Initialization information + +backbone.conv1.weight - torch.Size([64, 3, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.bn1.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.bn1.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.conv1.weight - torch.Size([64, 64, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.bn1.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.bn1.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.conv2.weight - torch.Size([64, 64, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.bn2.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.bn2.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.conv3.weight - torch.Size([256, 64, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.bn3.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.bn3.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.downsample.0.weight - torch.Size([256, 64, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.downsample.1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.downsample.1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.conv1.weight - torch.Size([64, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.bn1.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.bn1.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.conv2.weight - torch.Size([64, 64, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.bn2.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.bn2.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.conv3.weight - torch.Size([256, 64, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.bn3.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.bn3.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.conv1.weight - torch.Size([64, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.bn1.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.bn1.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.conv2.weight - torch.Size([64, 64, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.bn2.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.bn2.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.conv3.weight - torch.Size([256, 64, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.bn3.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.bn3.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.conv1.weight - torch.Size([128, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.bn1.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.bn1.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.conv2.weight - torch.Size([128, 128, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.bn2.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.bn2.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.conv3.weight - torch.Size([512, 128, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.bn3.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.bn3.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.downsample.0.weight - torch.Size([512, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.downsample.1.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.downsample.1.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.conv1.weight - torch.Size([128, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.bn1.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.bn1.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.conv2.weight - torch.Size([128, 128, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.bn2.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.bn2.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.conv3.weight - torch.Size([512, 128, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.bn3.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.bn3.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.conv1.weight - torch.Size([128, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.bn1.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.bn1.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.conv2.weight - torch.Size([128, 128, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.bn2.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.bn2.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.conv3.weight - torch.Size([512, 128, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.bn3.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.bn3.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.conv1.weight - torch.Size([128, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.bn1.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.bn1.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.conv2.weight - torch.Size([128, 128, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.bn2.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.bn2.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.conv3.weight - torch.Size([512, 128, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.bn3.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.bn3.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.conv1.weight - torch.Size([256, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.bn1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.bn1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.conv2.weight - torch.Size([256, 256, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.bn2.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.bn2.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.conv3.weight - torch.Size([1024, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.bn3.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.bn3.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.downsample.0.weight - torch.Size([1024, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.downsample.1.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.downsample.1.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.conv1.weight - torch.Size([256, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.bn1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.bn1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.conv2.weight - torch.Size([256, 256, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.bn2.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.bn2.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.conv3.weight - torch.Size([1024, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.bn3.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.bn3.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.conv1.weight - torch.Size([256, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.bn1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.bn1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.conv2.weight - torch.Size([256, 256, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.bn2.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.bn2.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.conv3.weight - torch.Size([1024, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.bn3.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.bn3.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.conv1.weight - torch.Size([256, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.bn1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.bn1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.conv2.weight - torch.Size([256, 256, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.bn2.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.bn2.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.conv3.weight - torch.Size([1024, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.bn3.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.bn3.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.conv1.weight - torch.Size([256, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.bn1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.bn1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.conv2.weight - torch.Size([256, 256, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.bn2.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.bn2.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.conv3.weight - torch.Size([1024, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.bn3.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.bn3.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.conv1.weight - torch.Size([256, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.bn1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.bn1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.conv2.weight - torch.Size([256, 256, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.bn2.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.bn2.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.conv3.weight - torch.Size([1024, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.bn3.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.bn3.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.conv1.weight - torch.Size([512, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.bn1.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.bn1.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.conv2.weight - torch.Size([512, 512, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.bn2.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.bn2.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.conv3.weight - torch.Size([2048, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.bn3.weight - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.bn3.bias - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.downsample.0.weight - torch.Size([2048, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.downsample.1.weight - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.downsample.1.bias - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.conv1.weight - torch.Size([512, 2048, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.bn1.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.bn1.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.conv2.weight - torch.Size([512, 512, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.bn2.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.bn2.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.conv3.weight - torch.Size([2048, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.bn3.weight - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.bn3.bias - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.conv1.weight - torch.Size([512, 2048, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.bn1.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.bn1.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.conv2.weight - torch.Size([512, 512, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.bn2.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.bn2.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.conv3.weight - torch.Size([2048, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.bn3.weight - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.bn3.bias - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +head.fc.weight - torch.Size([2, 2048]): +NormalInit: mean=0, std=0.01, bias=0 + +head.fc.bias - torch.Size([2]): +NormalInit: mean=0, std=0.01, bias=0 +2023/06/06 05:16:15 - mmengine - WARNING - "FileClient" will be deprecated in future. Please use io functions in https://mmengine.readthedocs.io/en/latest/api/fileio.html#file-io +2023/06/06 05:16:15 - mmengine - WARNING - "HardDiskBackend" is the alias of "LocalBackend" and the former will be deprecated in future. +2023/06/06 05:16:15 - mmengine - INFO - Checkpoints will be saved to /mnt/petrelfs/luzeyu/workspace/fakebench/mmpretrain/workdir/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1. +2023/06/06 05:17:09 - mmengine - INFO - Epoch(train) [1][ 100/3937] lr: 9.9999e-05 eta: 5:49:41 time: 0.5047 data_time: 0.3514 memory: 9436 loss: 0.6373 +2023/06/06 05:18:07 - mmengine - INFO - Epoch(train) [1][ 200/3937] lr: 9.9994e-05 eta: 6:05:34 time: 0.5414 data_time: 0.4017 memory: 6319 loss: 0.5732 +2023/06/06 05:19:13 - mmengine - INFO - Epoch(train) [1][ 300/3937] lr: 9.9987e-05 eta: 6:25:38 time: 0.6411 data_time: 0.4869 memory: 6319 loss: 0.5214 +2023/06/06 05:20:14 - mmengine - INFO - Epoch(train) [1][ 400/3937] lr: 9.9977e-05 eta: 6:28:07 time: 0.6815 data_time: 0.5418 memory: 6319 loss: 0.4485 +2023/06/06 05:21:20 - mmengine - INFO - Epoch(train) [1][ 500/3937] lr: 9.9964e-05 eta: 6:34:51 time: 0.5462 data_time: 0.4061 memory: 6319 loss: 0.4105 +2023/06/06 05:22:21 - mmengine - INFO - Epoch(train) [1][ 600/3937] lr: 9.9949e-05 eta: 6:33:26 time: 0.6187 data_time: 0.4788 memory: 6319 loss: 0.3614 +2023/06/06 05:23:22 - mmengine - INFO - Epoch(train) [1][ 700/3937] lr: 9.9930e-05 eta: 6:32:17 time: 0.6050 data_time: 0.4635 memory: 6319 loss: 0.3202 +2023/06/06 05:24:24 - mmengine - INFO - Epoch(train) [1][ 800/3937] lr: 9.9909e-05 eta: 6:32:24 time: 0.6451 data_time: 0.5031 memory: 6319 loss: 0.3191 +2023/06/06 05:25:33 - mmengine - INFO - Epoch(train) [1][ 900/3937] lr: 9.9884e-05 eta: 6:36:55 time: 0.5433 data_time: 0.4020 memory: 6319 loss: 0.3023 +2023/06/06 05:26:30 - mmengine - INFO - Exp name: resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1_20230606_051539 +2023/06/06 05:26:30 - mmengine - INFO - Epoch(train) [1][1000/3937] lr: 9.9857e-05 eta: 6:33:17 time: 0.6019 data_time: 0.4612 memory: 6319 loss: 0.2838 +2023/06/06 05:27:27 - mmengine - INFO - Epoch(train) [1][1100/3937] lr: 9.9827e-05 eta: 6:29:27 time: 0.5616 data_time: 0.4169 memory: 6319 loss: 0.2604 +2023/06/06 05:28:28 - mmengine - INFO - Epoch(train) [1][1200/3937] lr: 9.9794e-05 eta: 6:28:07 time: 0.5829 data_time: 0.4422 memory: 6319 loss: 0.2618 +2023/06/06 05:29:25 - mmengine - INFO - Epoch(train) [1][1300/3937] lr: 9.9758e-05 eta: 6:25:32 time: 0.5849 data_time: 0.4418 memory: 6319 loss: 0.2529 +2023/06/06 05:30:23 - mmengine - INFO - Epoch(train) [1][1400/3937] lr: 9.9720e-05 eta: 6:23:09 time: 0.5753 data_time: 0.4348 memory: 6319 loss: 0.2381 +2023/06/06 05:31:21 - mmengine - INFO - Epoch(train) [1][1500/3937] lr: 9.9678e-05 eta: 6:21:05 time: 0.5428 data_time: 0.3910 memory: 6319 loss: 0.2377 +2023/06/06 05:32:19 - mmengine - INFO - Epoch(train) [1][1600/3937] lr: 9.9634e-05 eta: 6:19:15 time: 0.5666 data_time: 0.4258 memory: 6319 loss: 0.2392 +2023/06/06 05:33:16 - mmengine - INFO - Epoch(train) [1][1700/3937] lr: 9.9587e-05 eta: 6:17:01 time: 0.5770 data_time: 0.4286 memory: 6319 loss: 0.2350 +2023/06/06 05:34:13 - mmengine - INFO - Epoch(train) [1][1800/3937] lr: 9.9537e-05 eta: 6:14:44 time: 0.5512 data_time: 0.4116 memory: 6319 loss: 0.2359 +2023/06/06 05:36:42 - mmengine - INFO - Epoch(train) [1][1900/3937] lr: 9.9484e-05 eta: 6:43:14 time: 0.6120 data_time: 0.4719 memory: 6319 loss: 0.2265 +2023/06/06 05:37:39 - mmengine - INFO - Exp name: resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1_20230606_051539 +2023/06/06 05:37:39 - mmengine - INFO - Epoch(train) [1][2000/3937] lr: 9.9429e-05 eta: 6:39:48 time: 0.6012 data_time: 0.4618 memory: 6319 loss: 0.2139 +2023/06/06 05:38:44 - mmengine - INFO - Epoch(train) [1][2100/3937] lr: 9.9370e-05 eta: 6:38:52 time: 0.5164 data_time: 0.3744 memory: 6319 loss: 0.2222 +2023/06/06 05:39:47 - mmengine - INFO - Epoch(train) [1][2200/3937] lr: 9.9309e-05 eta: 6:37:26 time: 0.5578 data_time: 0.4180 memory: 6319 loss: 0.2177 +2023/06/06 05:40:44 - mmengine - INFO - Epoch(train) [1][2300/3937] lr: 9.9245e-05 eta: 6:34:27 time: 0.5166 data_time: 0.3758 memory: 6319 loss: 0.2220 +2023/06/06 05:41:38 - mmengine - INFO - Epoch(train) [1][2400/3937] lr: 9.9178e-05 eta: 6:31:00 time: 0.5589 data_time: 0.4190 memory: 6319 loss: 0.2072 +2023/06/06 05:42:36 - mmengine - INFO - Epoch(train) [1][2500/3937] lr: 9.9108e-05 eta: 6:28:24 time: 0.5872 data_time: 0.4474 memory: 6319 loss: 0.2102 +2023/06/06 05:43:35 - mmengine - INFO - Epoch(train) [1][2600/3937] lr: 9.9036e-05 eta: 6:26:22 time: 0.5721 data_time: 0.4291 memory: 6319 loss: 0.2182 +2023/06/06 05:44:30 - mmengine - INFO - Epoch(train) [1][2700/3937] lr: 9.8960e-05 eta: 6:23:36 time: 0.5731 data_time: 0.4330 memory: 6319 loss: 0.2115 +2023/06/06 05:45:26 - mmengine - INFO - Epoch(train) [1][2800/3937] lr: 9.8882e-05 eta: 6:21:08 time: 0.5482 data_time: 0.4070 memory: 6319 loss: 0.1990 +2023/06/06 05:46:23 - mmengine - INFO - Epoch(train) [1][2900/3937] lr: 9.8801e-05 eta: 6:18:56 time: 0.5595 data_time: 0.4196 memory: 6319 loss: 0.1944 +2023/06/06 05:47:21 - mmengine - INFO - Exp name: resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1_20230606_051539 +2023/06/06 05:47:21 - mmengine - INFO - Epoch(train) [1][3000/3937] lr: 9.8718e-05 eta: 6:16:56 time: 0.5798 data_time: 0.4379 memory: 6319 loss: 0.1924 +2023/06/06 05:48:19 - mmengine - INFO - Epoch(train) [1][3100/3937] lr: 9.8631e-05 eta: 6:15:03 time: 0.5713 data_time: 0.4307 memory: 6319 loss: 0.1904 +2023/06/06 05:49:15 - mmengine - INFO - Epoch(train) [1][3200/3937] lr: 9.8542e-05 eta: 6:12:58 time: 0.5764 data_time: 0.4360 memory: 6319 loss: 0.1930 +2023/06/06 05:50:13 - mmengine - INFO - Epoch(train) [1][3300/3937] lr: 9.8450e-05 eta: 6:11:07 time: 0.5986 data_time: 0.4568 memory: 6319 loss: 0.1836 +2023/06/06 05:51:10 - mmengine - INFO - Epoch(train) [1][3400/3937] lr: 9.8355e-05 eta: 6:09:22 time: 0.5555 data_time: 0.4158 memory: 6319 loss: 0.1891 +2023/06/06 05:52:07 - mmengine - INFO - Epoch(train) [1][3500/3937] lr: 9.8257e-05 eta: 6:07:28 time: 0.5643 data_time: 0.4242 memory: 6319 loss: 0.2057 +2023/06/06 05:53:04 - mmengine - INFO - Epoch(train) [1][3600/3937] lr: 9.8157e-05 eta: 6:05:46 time: 0.5689 data_time: 0.4279 memory: 6319 loss: 0.1987 +2023/06/06 05:54:01 - mmengine - INFO - Epoch(train) [1][3700/3937] lr: 9.8054e-05 eta: 6:03:58 time: 0.5294 data_time: 0.3884 memory: 6319 loss: 0.1900 +2023/06/06 05:55:07 - mmengine - INFO - Epoch(train) [1][3800/3937] lr: 9.7948e-05 eta: 6:03:46 time: 0.5100 data_time: 0.3698 memory: 6319 loss: 0.1846 +2023/06/06 05:56:00 - mmengine - INFO - Epoch(train) [1][3900/3937] lr: 9.7840e-05 eta: 6:01:28 time: 0.5300 data_time: 0.3885 memory: 6319 loss: 0.1877 +2023/06/06 05:56:22 - mmengine - INFO - Exp name: resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1_20230606_051539 +2023/06/06 05:56:22 - mmengine - INFO - Saving checkpoint at 1 epochs +2023/06/06 05:56:58 - mmengine - INFO - Epoch(val) [1][57/57] accuracy/top1: 65.2830 data_time: 0.3456 time: 0.4378 +2023/06/06 05:57:37 - mmengine - INFO - Exp name: resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1_20230606_051539 +2023/06/06 05:57:54 - mmengine - INFO - Epoch(train) [2][ 100/3937] lr: 9.7686e-05 eta: 5:59:09 time: 0.5510 data_time: 0.3870 memory: 8776 loss: 0.1776 +2023/06/06 05:58:47 - mmengine - INFO - Epoch(train) [2][ 200/3937] lr: 9.7571e-05 eta: 5:57:03 time: 0.4744 data_time: 0.1604 memory: 6318 loss: 0.1698 +2023/06/06 05:59:38 - mmengine - INFO - Epoch(train) [2][ 300/3937] lr: 9.7454e-05 eta: 5:54:41 time: 0.5031 data_time: 0.0548 memory: 6318 loss: 0.1841 +2023/06/06 06:00:33 - mmengine - INFO - Epoch(train) [2][ 400/3937] lr: 9.7333e-05 eta: 5:52:49 time: 0.6180 data_time: 0.0845 memory: 6318 loss: 0.1695 +2023/06/06 06:01:23 - mmengine - INFO - Epoch(train) [2][ 500/3937] lr: 9.7210e-05 eta: 5:50:26 time: 0.4621 data_time: 0.0924 memory: 6318 loss: 0.1613 +2023/06/06 06:02:10 - mmengine - INFO - Epoch(train) [2][ 600/3937] lr: 9.7084e-05 eta: 5:47:46 time: 0.5158 data_time: 0.1673 memory: 6318 loss: 0.1570 +2023/06/06 06:03:02 - mmengine - INFO - Epoch(train) [2][ 700/3937] lr: 9.6956e-05 eta: 5:45:48 time: 0.4993 data_time: 0.3292 memory: 6318 loss: 0.1773 +2023/06/06 06:03:55 - mmengine - INFO - Epoch(train) [2][ 800/3937] lr: 9.6825e-05 eta: 5:43:58 time: 0.5105 data_time: 0.3632 memory: 6318 loss: 0.1792 +2023/06/06 06:04:48 - mmengine - INFO - Epoch(train) [2][ 900/3937] lr: 9.6691e-05 eta: 5:42:12 time: 0.5395 data_time: 0.3273 memory: 6318 loss: 0.1745 +2023/06/06 06:05:42 - mmengine - INFO - Epoch(train) [2][1000/3937] lr: 9.6554e-05 eta: 5:40:34 time: 0.5209 data_time: 0.2691 memory: 6318 loss: 0.1860 +2023/06/06 06:06:18 - mmengine - INFO - Exp name: resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1_20230606_051539 +2023/06/06 06:06:35 - mmengine - INFO - Epoch(train) [2][1100/3937] lr: 9.6415e-05 eta: 5:38:50 time: 0.5503 data_time: 0.3627 memory: 6318 loss: 0.1756 +2023/06/06 06:07:28 - mmengine - INFO - Epoch(train) [2][1200/3937] lr: 9.6273e-05 eta: 5:37:12 time: 0.5431 data_time: 0.2301 memory: 6318 loss: 0.1674 +2023/06/06 06:08:25 - mmengine - INFO - Epoch(train) [2][1300/3937] lr: 9.6129e-05 eta: 5:35:56 time: 0.5604 data_time: 0.1979 memory: 6318 loss: 0.1689 +2023/06/06 06:09:19 - mmengine - INFO - Epoch(train) [2][1400/3937] lr: 9.5982e-05 eta: 5:34:25 time: 0.5552 data_time: 0.2883 memory: 6318 loss: 0.1605 +2023/06/06 06:10:11 - mmengine - INFO - Epoch(train) [2][1500/3937] lr: 9.5832e-05 eta: 5:32:45 time: 0.4835 data_time: 0.0954 memory: 6318 loss: 0.1733 +2023/06/06 06:11:04 - mmengine - INFO - Epoch(train) [2][1600/3937] lr: 9.5680e-05 eta: 5:31:12 time: 0.5141 data_time: 0.1621 memory: 6318 loss: 0.1646 +2023/06/06 06:11:57 - mmengine - INFO - Epoch(train) [2][1700/3937] lr: 9.5525e-05 eta: 5:29:34 time: 0.5090 data_time: 0.2315 memory: 6318 loss: 0.1661 +2023/06/06 06:12:50 - mmengine - INFO - Epoch(train) [2][1800/3937] lr: 9.5368e-05 eta: 5:28:05 time: 0.5375 data_time: 0.3950 memory: 6318 loss: 0.1593 +2023/06/06 06:13:44 - mmengine - INFO - Epoch(train) [2][1900/3937] lr: 9.5208e-05 eta: 5:26:40 time: 0.5412 data_time: 0.4003 memory: 6318 loss: 0.1620 +2023/06/06 06:14:35 - mmengine - INFO - Epoch(train) [2][2000/3937] lr: 9.5045e-05 eta: 5:25:01 time: 0.4904 data_time: 0.3501 memory: 6318 loss: 0.1636 +2023/06/06 06:15:07 - mmengine - INFO - Exp name: resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1_20230606_051539 +2023/06/06 06:15:27 - mmengine - INFO - Epoch(train) [2][2100/3937] lr: 9.4880e-05 eta: 5:23:29 time: 0.5136 data_time: 0.3732 memory: 6318 loss: 0.1564 +2023/06/06 06:16:21 - mmengine - INFO - Epoch(train) [2][2200/3937] lr: 9.4713e-05 eta: 5:22:08 time: 0.5361 data_time: 0.3961 memory: 6318 loss: 0.1632 +2023/06/06 06:17:14 - mmengine - INFO - Epoch(train) [2][2300/3937] lr: 9.4543e-05 eta: 5:20:40 time: 0.5368 data_time: 0.3964 memory: 6318 loss: 0.1699 +2023/06/06 06:18:06 - mmengine - INFO - Epoch(train) [2][2400/3937] lr: 9.4370e-05 eta: 5:19:13 time: 0.5168 data_time: 0.3755 memory: 6318 loss: 0.1533 +2023/06/06 06:18:59 - mmengine - INFO - Epoch(train) [2][2500/3937] lr: 9.4195e-05 eta: 5:17:46 time: 0.5169 data_time: 0.3764 memory: 6318 loss: 0.1366 +2023/06/06 06:19:53 - mmengine - INFO - Epoch(train) [2][2600/3937] lr: 9.4017e-05 eta: 5:16:32 time: 0.7384 data_time: 0.5957 memory: 6318 loss: 0.1449 +2023/06/06 06:20:47 - mmengine - INFO - Epoch(train) [2][2700/3937] lr: 9.3837e-05 eta: 5:15:14 time: 0.5579 data_time: 0.4174 memory: 6318 loss: 0.1521 +2023/06/06 06:21:36 - mmengine - INFO - Epoch(train) [2][2800/3937] lr: 9.3654e-05 eta: 5:13:35 time: 0.4788 data_time: 0.3381 memory: 6318 loss: 0.1664 +2023/06/06 06:22:28 - mmengine - INFO - Epoch(train) [2][2900/3937] lr: 9.3469e-05 eta: 5:12:10 time: 0.5536 data_time: 0.4131 memory: 6318 loss: 0.1545 +2023/06/06 06:23:22 - mmengine - INFO - Epoch(train) [2][3000/3937] lr: 9.3282e-05 eta: 5:10:53 time: 0.5363 data_time: 0.3947 memory: 6318 loss: 0.1534 +2023/06/06 06:23:54 - mmengine - INFO - Exp name: resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1_20230606_051539 +2023/06/06 06:24:18 - mmengine - INFO - Epoch(train) [2][3100/3937] lr: 9.3092e-05 eta: 5:09:48 time: 0.5498 data_time: 0.3988 memory: 6318 loss: 0.1516 +2023/06/06 06:25:11 - mmengine - INFO - Epoch(train) [2][3200/3937] lr: 9.2899e-05 eta: 5:08:32 time: 0.5706 data_time: 0.4294 memory: 6318 loss: 0.1571 +2023/06/06 06:26:05 - mmengine - INFO - Epoch(train) [2][3300/3937] lr: 9.2705e-05 eta: 5:07:17 time: 0.5533 data_time: 0.4121 memory: 6318 loss: 0.1340 +2023/06/06 06:26:57 - mmengine - INFO - Epoch(train) [2][3400/3937] lr: 9.2507e-05 eta: 5:05:59 time: 0.5596 data_time: 0.4186 memory: 6318 loss: 0.1446 +2023/06/06 06:27:51 - mmengine - INFO - Epoch(train) [2][3500/3937] lr: 9.2308e-05 eta: 5:04:46 time: 0.5405 data_time: 0.3979 memory: 6318 loss: 0.1461 +2023/06/06 06:28:43 - mmengine - INFO - Epoch(train) [2][3600/3937] lr: 9.2106e-05 eta: 5:03:26 time: 0.5186 data_time: 0.3776 memory: 6318 loss: 0.1462 +2023/06/06 06:29:35 - mmengine - INFO - Epoch(train) [2][3700/3937] lr: 9.1902e-05 eta: 5:02:09 time: 0.5137 data_time: 0.3727 memory: 6318 loss: 0.1502 +2023/06/06 06:30:28 - mmengine - INFO - Epoch(train) [2][3800/3937] lr: 9.1695e-05 eta: 5:00:55 time: 0.5301 data_time: 0.3903 memory: 6318 loss: 0.1527 +2023/06/06 06:31:20 - mmengine - INFO - Epoch(train) [2][3900/3937] lr: 9.1486e-05 eta: 4:59:38 time: 0.5244 data_time: 0.3838 memory: 6318 loss: 0.1423 +2023/06/06 06:31:37 - mmengine - INFO - Exp name: resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1_20230606_051539 +2023/06/06 06:31:37 - mmengine - INFO - Saving checkpoint at 2 epochs +2023/06/06 06:32:08 - mmengine - INFO - Epoch(val) [2][57/57] accuracy/top1: 86.7993 data_time: 0.2487 time: 0.3374 +2023/06/06 06:33:03 - mmengine - INFO - Epoch(train) [3][ 100/3937] lr: 9.1196e-05 eta: 4:57:55 time: 0.5260 data_time: 0.3863 memory: 6318 loss: 0.1453 +2023/06/06 06:33:18 - mmengine - INFO - Exp name: resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1_20230606_051539 +2023/06/06 06:33:55 - mmengine - INFO - Epoch(train) [3][ 200/3937] lr: 9.0981e-05 eta: 4:56:40 time: 0.5138 data_time: 0.3732 memory: 6318 loss: 0.1224 +2023/06/06 06:34:48 - mmengine - INFO - Epoch(train) [3][ 300/3937] lr: 9.0764e-05 eta: 4:55:30 time: 0.5260 data_time: 0.3859 memory: 6318 loss: 0.1420 +2023/06/06 06:35:41 - mmengine - INFO - Epoch(train) [3][ 400/3937] lr: 9.0545e-05 eta: 4:54:16 time: 0.5375 data_time: 0.3987 memory: 6318 loss: 0.1296 +2023/06/06 06:36:44 - mmengine - INFO - Epoch(train) [3][ 500/3937] lr: 9.0324e-05 eta: 4:53:42 time: 0.5341 data_time: 0.3945 memory: 6318 loss: 0.1484 +2023/06/06 06:37:36 - mmengine - INFO - Epoch(train) [3][ 600/3937] lr: 9.0100e-05 eta: 4:52:29 time: 0.4678 data_time: 0.3276 memory: 6318 loss: 0.1336 +2023/06/06 06:38:26 - mmengine - INFO - Epoch(train) [3][ 700/3937] lr: 8.9875e-05 eta: 4:51:09 time: 0.5399 data_time: 0.3987 memory: 6318 loss: 0.1381 +2023/06/06 06:39:19 - mmengine - INFO - Epoch(train) [3][ 800/3937] lr: 8.9647e-05 eta: 4:49:56 time: 0.5374 data_time: 0.3964 memory: 6318 loss: 0.1518 +2023/06/06 06:40:11 - mmengine - INFO - Epoch(train) [3][ 900/3937] lr: 8.9416e-05 eta: 4:48:45 time: 0.5084 data_time: 0.3686 memory: 6318 loss: 0.1569 +2023/06/06 06:41:03 - mmengine - INFO - Epoch(train) [3][1000/3937] lr: 8.9184e-05 eta: 4:47:32 time: 0.4690 data_time: 0.3288 memory: 6318 loss: 0.1136 +2023/06/06 06:41:53 - mmengine - INFO - Epoch(train) [3][1100/3937] lr: 8.8949e-05 eta: 4:46:12 time: 0.5258 data_time: 0.3855 memory: 6318 loss: 0.1454 +2023/06/06 06:42:08 - mmengine - INFO - Exp name: resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1_20230606_051539 +2023/06/06 06:42:46 - mmengine - INFO - Epoch(train) [3][1200/3937] lr: 8.8712e-05 eta: 4:45:06 time: 0.5376 data_time: 0.3976 memory: 6318 loss: 0.1523 +2023/06/06 06:43:39 - mmengine - INFO - Epoch(train) [3][1300/3937] lr: 8.8474e-05 eta: 4:43:58 time: 0.5290 data_time: 0.3884 memory: 6318 loss: 0.1266 +2023/06/06 06:44:35 - mmengine - INFO - Epoch(train) [3][1400/3937] lr: 8.8232e-05 eta: 4:43:00 time: 0.5105 data_time: 0.3702 memory: 6318 loss: 0.1287 +2023/06/06 06:45:28 - mmengine - INFO - Epoch(train) [3][1500/3937] lr: 8.7989e-05 eta: 4:41:52 time: 0.5381 data_time: 0.3972 memory: 6318 loss: 0.1386 +2023/06/06 06:46:22 - mmengine - INFO - Epoch(train) [3][1600/3937] lr: 8.7744e-05 eta: 4:40:48 time: 0.5485 data_time: 0.4087 memory: 6318 loss: 0.1346 +2023/06/06 06:47:15 - mmengine - INFO - Epoch(train) [3][1700/3937] lr: 8.7497e-05 eta: 4:39:40 time: 0.5403 data_time: 0.4002 memory: 6318 loss: 0.1432 +2023/06/06 06:48:08 - mmengine - INFO - Epoch(train) [3][1800/3937] lr: 8.7247e-05 eta: 4:38:36 time: 0.5743 data_time: 0.4343 memory: 6318 loss: 0.1401 +2023/06/06 06:49:01 - mmengine - INFO - Epoch(train) [3][1900/3937] lr: 8.6996e-05 eta: 4:37:29 time: 0.5147 data_time: 0.3745 memory: 6318 loss: 0.1292 +2023/06/06 06:49:53 - mmengine - INFO - Epoch(train) [3][2000/3937] lr: 8.6742e-05 eta: 4:36:20 time: 0.5084 data_time: 0.3644 memory: 6318 loss: 0.1330 +2023/06/06 06:50:47 - mmengine - INFO - Epoch(train) [3][2100/3937] lr: 8.6487e-05 eta: 4:35:16 time: 0.5392 data_time: 0.3924 memory: 6318 loss: 0.1320 +2023/06/06 06:51:01 - mmengine - INFO - Exp name: resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1_20230606_051539 +2023/06/06 06:51:40 - mmengine - INFO - Epoch(train) [3][2200/3937] lr: 8.6229e-05 eta: 4:34:11 time: 0.7131 data_time: 0.5733 memory: 6318 loss: 0.1407 +2023/06/06 06:52:31 - mmengine - INFO - Epoch(train) [3][2300/3937] lr: 8.5970e-05 eta: 4:33:01 time: 0.5335 data_time: 0.3915 memory: 6318 loss: 0.1336 +2023/06/06 06:53:23 - mmengine - INFO - Epoch(train) [3][2400/3937] lr: 8.5708e-05 eta: 4:31:53 time: 0.5171 data_time: 0.3767 memory: 6318 loss: 0.1252 +2023/06/06 06:54:17 - mmengine - INFO - Epoch(train) [3][2500/3937] lr: 8.5445e-05 eta: 4:30:51 time: 0.5400 data_time: 0.3926 memory: 6318 loss: 0.1189 +2023/06/06 06:55:10 - mmengine - INFO - Epoch(train) [3][2600/3937] lr: 8.5179e-05 eta: 4:29:46 time: 0.5304 data_time: 0.3909 memory: 6318 loss: 0.1370 +2023/06/06 06:56:02 - mmengine - INFO - Epoch(train) [3][2700/3937] lr: 8.4912e-05 eta: 4:28:38 time: 0.5066 data_time: 0.3662 memory: 6318 loss: 0.1270 +2023/06/06 06:56:53 - mmengine - INFO - Epoch(train) [3][2800/3937] lr: 8.4643e-05 eta: 4:27:31 time: 0.4913 data_time: 0.3500 memory: 6318 loss: 0.1258 +2023/06/06 06:57:47 - mmengine - INFO - Epoch(train) [3][2900/3937] lr: 8.4372e-05 eta: 4:26:28 time: 0.5241 data_time: 0.3839 memory: 6318 loss: 0.1231 +2023/06/06 06:58:40 - mmengine - INFO - Epoch(train) [3][3000/3937] lr: 8.4099e-05 eta: 4:25:26 time: 0.5343 data_time: 0.3930 memory: 6318 loss: 0.1207 +2023/06/06 06:59:34 - mmengine - INFO - Epoch(train) [3][3100/3937] lr: 8.3824e-05 eta: 4:24:24 time: 0.5280 data_time: 0.3880 memory: 6318 loss: 0.1270 +2023/06/06 06:59:49 - mmengine - INFO - Exp name: resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1_20230606_051539 +2023/06/06 07:00:26 - mmengine - INFO - Epoch(train) [3][3200/3937] lr: 8.3547e-05 eta: 4:23:19 time: 0.5151 data_time: 0.3738 memory: 6318 loss: 0.1255 +2023/06/06 07:01:14 - mmengine - INFO - Epoch(train) [3][3300/3937] lr: 8.3269e-05 eta: 4:22:03 time: 0.4971 data_time: 0.3565 memory: 6318 loss: 0.1128 +2023/06/06 07:02:06 - mmengine - INFO - Epoch(train) [3][3400/3937] lr: 8.2988e-05 eta: 4:20:59 time: 0.5572 data_time: 0.4172 memory: 6318 loss: 0.1208 +2023/06/06 07:03:00 - mmengine - INFO - Epoch(train) [3][3500/3937] lr: 8.2706e-05 eta: 4:19:57 time: 0.5464 data_time: 0.4059 memory: 6318 loss: 0.1126 +2023/06/06 07:03:55 - mmengine - INFO - Epoch(train) [3][3600/3937] lr: 8.2423e-05 eta: 4:19:00 time: 0.5714 data_time: 0.4318 memory: 6318 loss: 0.1323 +2023/06/06 07:04:47 - mmengine - INFO - Epoch(train) [3][3700/3937] lr: 8.2137e-05 eta: 4:17:55 time: 0.5590 data_time: 0.4175 memory: 6318 loss: 0.1208 +2023/06/06 07:05:38 - mmengine - INFO - Epoch(train) [3][3800/3937] lr: 8.1850e-05 eta: 4:16:49 time: 0.5164 data_time: 0.3455 memory: 6318 loss: 0.1324 +2023/06/06 07:06:31 - mmengine - INFO - Epoch(train) [3][3900/3937] lr: 8.1561e-05 eta: 4:15:47 time: 0.5163 data_time: 0.3638 memory: 6318 loss: 0.1253 +2023/06/06 07:06:51 - mmengine - INFO - Exp name: resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1_20230606_051539 +2023/06/06 07:06:51 - mmengine - INFO - Saving checkpoint at 3 epochs +2023/06/06 07:07:24 - mmengine - INFO - Epoch(val) [3][57/57] accuracy/top1: 91.9036 data_time: 0.3081 time: 0.3949 +2023/06/06 07:08:18 - mmengine - INFO - Epoch(train) [4][ 100/3937] lr: 8.1162e-05 eta: 4:14:26 time: 0.5493 data_time: 0.4092 memory: 6318 loss: 0.1251 +2023/06/06 07:09:06 - mmengine - INFO - Exp name: resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1_20230606_051539 +2023/06/06 07:09:12 - mmengine - INFO - Epoch(train) [4][ 200/3937] lr: 8.0869e-05 eta: 4:13:25 time: 0.5319 data_time: 0.3922 memory: 6318 loss: 0.1404 +2023/06/06 07:10:05 - mmengine - INFO - Epoch(train) [4][ 300/3937] lr: 8.0574e-05 eta: 4:12:25 time: 0.5273 data_time: 0.3872 memory: 6318 loss: 0.1384 +2023/06/06 07:10:58 - mmengine - INFO - Epoch(train) [4][ 400/3937] lr: 8.0278e-05 eta: 4:11:23 time: 0.5216 data_time: 0.3695 memory: 6318 loss: 0.1159 +2023/06/06 07:11:50 - mmengine - INFO - Epoch(train) [4][ 500/3937] lr: 7.9980e-05 eta: 4:10:20 time: 0.5605 data_time: 0.4206 memory: 6318 loss: 0.1300 +2023/06/06 07:12:42 - mmengine - INFO - Epoch(train) [4][ 600/3937] lr: 7.9681e-05 eta: 4:09:17 time: 0.5050 data_time: 0.3648 memory: 6318 loss: 0.1427 +2023/06/06 07:13:35 - mmengine - INFO - Epoch(train) [4][ 700/3937] lr: 7.9380e-05 eta: 4:08:16 time: 0.5670 data_time: 0.4261 memory: 6318 loss: 0.1226 +2023/06/06 07:14:28 - mmengine - INFO - Epoch(train) [4][ 800/3937] lr: 7.9077e-05 eta: 4:07:16 time: 0.5410 data_time: 0.4006 memory: 6318 loss: 0.1108 +2023/06/06 07:15:21 - mmengine - INFO - Epoch(train) [4][ 900/3937] lr: 7.8773e-05 eta: 4:06:15 time: 0.5204 data_time: 0.3793 memory: 6318 loss: 0.1147 +2023/06/06 07:16:14 - mmengine - INFO - Epoch(train) [4][1000/3937] lr: 7.8467e-05 eta: 4:05:15 time: 0.4980 data_time: 0.3544 memory: 6318 loss: 0.1201 +2023/06/06 07:17:08 - mmengine - INFO - Epoch(train) [4][1100/3937] lr: 7.8160e-05 eta: 4:04:17 time: 0.5739 data_time: 0.4329 memory: 6318 loss: 0.1147 +2023/06/06 07:17:57 - mmengine - INFO - Exp name: resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1_20230606_051539 +2023/06/06 07:18:01 - mmengine - INFO - Epoch(train) [4][1200/3937] lr: 7.7852e-05 eta: 4:03:16 time: 0.4707 data_time: 0.3300 memory: 6318 loss: 0.1175 +2023/06/06 07:18:56 - mmengine - INFO - Epoch(train) [4][1300/3937] lr: 7.7541e-05 eta: 4:02:19 time: 0.5458 data_time: 0.4063 memory: 6318 loss: 0.1247 +2023/06/06 07:19:49 - mmengine - INFO - Epoch(train) [4][1400/3937] lr: 7.7230e-05 eta: 4:01:19 time: 0.5519 data_time: 0.4119 memory: 6318 loss: 0.1383 +2023/06/06 07:20:42 - mmengine - INFO - Epoch(train) [4][1500/3937] lr: 7.6917e-05 eta: 4:00:20 time: 0.4628 data_time: 0.3220 memory: 6318 loss: 0.1111 +2023/06/06 07:21:37 - mmengine - INFO - Epoch(train) [4][1600/3937] lr: 7.6603e-05 eta: 3:59:23 time: 0.5125 data_time: 0.3714 memory: 6318 loss: 0.1200 +2023/06/06 07:22:30 - mmengine - INFO - Epoch(train) [4][1700/3937] lr: 7.6287e-05 eta: 3:58:23 time: 0.5060 data_time: 0.3652 memory: 6318 loss: 0.1127 +2023/06/06 07:23:23 - mmengine - INFO - Epoch(train) [4][1800/3937] lr: 7.5970e-05 eta: 3:57:24 time: 0.5295 data_time: 0.3892 memory: 6318 loss: 0.1079 +2023/06/06 07:24:17 - mmengine - INFO - Epoch(train) [4][1900/3937] lr: 7.5652e-05 eta: 3:56:26 time: 0.5295 data_time: 0.3884 memory: 6318 loss: 0.1077 +2023/06/06 07:25:08 - mmengine - INFO - Epoch(train) [4][2000/3937] lr: 7.5332e-05 eta: 3:55:23 time: 0.4793 data_time: 0.3397 memory: 6318 loss: 0.1228 +2023/06/06 07:26:03 - mmengine - INFO - Epoch(train) [4][2100/3937] lr: 7.5011e-05 eta: 3:54:28 time: 0.5257 data_time: 0.3724 memory: 6318 loss: 0.1336 +2023/06/06 07:26:51 - mmengine - INFO - Exp name: resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1_20230606_051539 +2023/06/06 07:26:56 - mmengine - INFO - Epoch(train) [4][2200/3937] lr: 7.4689e-05 eta: 3:53:28 time: 0.5282 data_time: 0.3885 memory: 6318 loss: 0.1253 +2023/06/06 07:27:49 - mmengine - INFO - Epoch(train) [4][2300/3937] lr: 7.4365e-05 eta: 3:52:29 time: 0.5276 data_time: 0.3871 memory: 6318 loss: 0.1141 +2023/06/06 07:28:42 - mmengine - INFO - Epoch(train) [4][2400/3937] lr: 7.4040e-05 eta: 3:51:29 time: 0.5059 data_time: 0.3657 memory: 6318 loss: 0.1167 +2023/06/06 07:29:34 - mmengine - INFO - Epoch(train) [4][2500/3937] lr: 7.3714e-05 eta: 3:50:29 time: 0.5242 data_time: 0.3831 memory: 6318 loss: 0.1068 +2023/06/06 07:30:28 - mmengine - INFO - Epoch(train) [4][2600/3937] lr: 7.3387e-05 eta: 3:49:31 time: 0.5403 data_time: 0.3998 memory: 6318 loss: 0.1111 +2023/06/06 07:31:22 - mmengine - INFO - Epoch(train) [4][2700/3937] lr: 7.3059e-05 eta: 3:48:35 time: 0.5112 data_time: 0.3698 memory: 6318 loss: 0.1154 +2023/06/06 07:32:16 - mmengine - INFO - Epoch(train) [4][2800/3937] lr: 7.2730e-05 eta: 3:47:38 time: 0.5330 data_time: 0.3927 memory: 6318 loss: 0.1196 +2023/06/06 07:33:12 - mmengine - INFO - Epoch(train) [4][2900/3937] lr: 7.2399e-05 eta: 3:46:43 time: 0.4828 data_time: 0.3426 memory: 6318 loss: 0.1188 +2023/06/06 07:34:06 - mmengine - INFO - Epoch(train) [4][3000/3937] lr: 7.2067e-05 eta: 3:45:46 time: 0.5592 data_time: 0.4177 memory: 6318 loss: 0.1068 +2023/06/06 07:34:59 - mmengine - INFO - Epoch(train) [4][3100/3937] lr: 7.1734e-05 eta: 3:44:48 time: 0.5557 data_time: 0.4155 memory: 6318 loss: 0.1081 +2023/06/06 07:35:47 - mmengine - INFO - Exp name: resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1_20230606_051539 +2023/06/06 07:35:53 - mmengine - INFO - Epoch(train) [4][3200/3937] lr: 7.1401e-05 eta: 3:43:50 time: 0.5145 data_time: 0.3741 memory: 6318 loss: 0.1165 +2023/06/06 07:36:46 - mmengine - INFO - Epoch(train) [4][3300/3937] lr: 7.1066e-05 eta: 3:42:51 time: 0.5350 data_time: 0.3953 memory: 6318 loss: 0.1208 +2023/06/06 07:37:45 - mmengine - INFO - Epoch(train) [4][3400/3937] lr: 7.0730e-05 eta: 3:42:02 time: 0.5341 data_time: 0.3927 memory: 6318 loss: 0.1159 +2023/06/06 07:38:38 - mmengine - INFO - Epoch(train) [4][3500/3937] lr: 7.0393e-05 eta: 3:41:04 time: 0.5602 data_time: 0.4198 memory: 6318 loss: 0.1173 +2023/06/06 07:39:32 - mmengine - INFO - Epoch(train) [4][3600/3937] lr: 7.0055e-05 eta: 3:40:07 time: 0.5138 data_time: 0.3735 memory: 6318 loss: 0.1098 +2023/06/06 07:40:25 - mmengine - INFO - Epoch(train) [4][3700/3937] lr: 6.9716e-05 eta: 3:39:09 time: 0.5398 data_time: 0.3990 memory: 6318 loss: 0.1083 +2023/06/06 07:41:14 - mmengine - INFO - Epoch(train) [4][3800/3937] lr: 6.9376e-05 eta: 3:38:05 time: 0.4256 data_time: 0.2859 memory: 6318 loss: 0.0960 +2023/06/06 07:42:08 - mmengine - INFO - Epoch(train) [4][3900/3937] lr: 6.9035e-05 eta: 3:37:08 time: 0.5472 data_time: 0.4022 memory: 6318 loss: 0.1149 +2023/06/06 07:42:27 - mmengine - INFO - Exp name: resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1_20230606_051539 +2023/06/06 07:42:27 - mmengine - INFO - Saving checkpoint at 4 epochs +2023/06/06 07:43:06 - mmengine - INFO - Epoch(val) [4][57/57] accuracy/top1: 93.9946 data_time: 0.3987 time: 0.4851 +2023/06/06 07:44:02 - mmengine - INFO - Epoch(train) [5][ 100/3937] lr: 6.8567e-05 eta: 3:35:51 time: 0.5796 data_time: 0.4392 memory: 6318 loss: 0.1268 +2023/06/06 07:44:54 - mmengine - INFO - Epoch(train) [5][ 200/3937] lr: 6.8224e-05 eta: 3:34:53 time: 0.5533 data_time: 0.4073 memory: 6318 loss: 0.1138 +2023/06/06 07:45:25 - mmengine - INFO - Exp name: resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1_20230606_051539 +2023/06/06 07:45:48 - mmengine - INFO - Epoch(train) [5][ 300/3937] lr: 6.7881e-05 eta: 3:33:55 time: 0.5136 data_time: 0.3716 memory: 6318 loss: 0.1120 +2023/06/06 07:46:43 - mmengine - INFO - Epoch(train) [5][ 400/3937] lr: 6.7536e-05 eta: 3:33:00 time: 0.5638 data_time: 0.4237 memory: 6318 loss: 0.1069 +2023/06/06 07:47:35 - mmengine - INFO - Epoch(train) [5][ 500/3937] lr: 6.7191e-05 eta: 3:32:01 time: 0.5221 data_time: 0.3810 memory: 6318 loss: 0.1148 +2023/06/06 07:48:28 - mmengine - INFO - Epoch(train) [5][ 600/3937] lr: 6.6845e-05 eta: 3:31:03 time: 0.5582 data_time: 0.4182 memory: 6318 loss: 0.1209 +2023/06/06 07:49:20 - mmengine - INFO - Epoch(train) [5][ 700/3937] lr: 6.6498e-05 eta: 3:30:04 time: 0.5336 data_time: 0.3941 memory: 6318 loss: 0.1143 +2023/06/06 07:50:13 - mmengine - INFO - Epoch(train) [5][ 800/3937] lr: 6.6151e-05 eta: 3:29:07 time: 0.5414 data_time: 0.3909 memory: 6318 loss: 0.1151 +2023/06/06 07:51:07 - mmengine - INFO - Epoch(train) [5][ 900/3937] lr: 6.5802e-05 eta: 3:28:10 time: 0.5284 data_time: 0.3882 memory: 6318 loss: 0.1190 +2023/06/06 07:52:01 - mmengine - INFO - Epoch(train) [5][1000/3937] lr: 6.5454e-05 eta: 3:27:14 time: 0.5280 data_time: 0.3874 memory: 6318 loss: 0.0977 +2023/06/06 07:52:53 - mmengine - INFO - Epoch(train) [5][1100/3937] lr: 6.5104e-05 eta: 3:26:15 time: 0.5452 data_time: 0.4036 memory: 6318 loss: 0.1031 +2023/06/06 07:53:47 - mmengine - INFO - Epoch(train) [5][1200/3937] lr: 6.4754e-05 eta: 3:25:19 time: 0.5358 data_time: 0.3943 memory: 6318 loss: 0.1205 +2023/06/06 07:54:14 - mmengine - INFO - Exp name: resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1_20230606_051539 +2023/06/06 07:54:39 - mmengine - INFO - Epoch(train) [5][1300/3937] lr: 6.4403e-05 eta: 3:24:20 time: 0.5059 data_time: 0.3663 memory: 6318 loss: 0.1149 +2023/06/06 07:55:30 - mmengine - INFO - Epoch(train) [5][1400/3937] lr: 6.4051e-05 eta: 3:23:20 time: 0.5050 data_time: 0.3636 memory: 6318 loss: 0.1093 +2023/06/06 07:56:22 - mmengine - INFO - Epoch(train) [5][1500/3937] lr: 6.3699e-05 eta: 3:22:21 time: 0.5233 data_time: 0.3828 memory: 6318 loss: 0.1071 +2023/06/06 07:57:15 - mmengine - INFO - Epoch(train) [5][1600/3937] lr: 6.3347e-05 eta: 3:21:24 time: 0.5349 data_time: 0.3948 memory: 6318 loss: 0.1112 +2023/06/06 07:58:10 - mmengine - INFO - Epoch(train) [5][1700/3937] lr: 6.2994e-05 eta: 3:20:29 time: 0.5508 data_time: 0.4100 memory: 6318 loss: 0.1143 +2023/06/06 07:59:04 - mmengine - INFO - Epoch(train) [5][1800/3937] lr: 6.2640e-05 eta: 3:19:33 time: 0.5132 data_time: 0.3744 memory: 6318 loss: 0.1044 +2023/06/06 07:59:57 - mmengine - INFO - Epoch(train) [5][1900/3937] lr: 6.2286e-05 eta: 3:18:37 time: 0.4946 data_time: 0.3545 memory: 6318 loss: 0.1073 +2023/06/06 08:00:46 - mmengine - INFO - Epoch(train) [5][2000/3937] lr: 6.1931e-05 eta: 3:17:34 time: 0.4647 data_time: 0.3237 memory: 6318 loss: 0.0993 +2023/06/06 08:01:37 - mmengine - INFO - Epoch(train) [5][2100/3937] lr: 6.1576e-05 eta: 3:16:35 time: 0.4765 data_time: 0.3361 memory: 6318 loss: 0.1078 +2023/06/06 08:02:28 - mmengine - INFO - Epoch(train) [5][2200/3937] lr: 6.1221e-05 eta: 3:15:35 time: 0.4753 data_time: 0.3357 memory: 6318 loss: 0.1031 +2023/06/06 08:02:55 - mmengine - INFO - Exp name: resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1_20230606_051539 +2023/06/06 08:03:19 - mmengine - INFO - Epoch(train) [5][2300/3937] lr: 6.0865e-05 eta: 3:14:36 time: 0.4953 data_time: 0.3543 memory: 6318 loss: 0.1024 +2023/06/06 08:04:09 - mmengine - INFO - Epoch(train) [5][2400/3937] lr: 6.0509e-05 eta: 3:13:35 time: 0.5209 data_time: 0.3811 memory: 6318 loss: 0.1015 +2023/06/06 08:05:00 - mmengine - INFO - Epoch(train) [5][2500/3937] lr: 6.0152e-05 eta: 3:12:36 time: 0.5036 data_time: 0.3631 memory: 6318 loss: 0.0985 +2023/06/06 08:05:51 - mmengine - INFO - Epoch(train) [5][2600/3937] lr: 5.9795e-05 eta: 3:11:38 time: 0.4994 data_time: 0.3588 memory: 6318 loss: 0.1022 +2023/06/06 08:06:43 - mmengine - INFO - Epoch(train) [5][2700/3937] lr: 5.9438e-05 eta: 3:10:39 time: 0.5059 data_time: 0.3660 memory: 6318 loss: 0.0967 +2023/06/06 08:07:34 - mmengine - INFO - Epoch(train) [5][2800/3937] lr: 5.9081e-05 eta: 3:09:41 time: 0.5142 data_time: 0.3740 memory: 6318 loss: 0.0941 +2023/06/06 08:08:24 - mmengine - INFO - Epoch(train) [5][2900/3937] lr: 5.8723e-05 eta: 3:08:41 time: 0.5547 data_time: 0.4150 memory: 6318 loss: 0.1097 +2023/06/06 08:09:15 - mmengine - INFO - Epoch(train) [5][3000/3937] lr: 5.8365e-05 eta: 3:07:43 time: 0.5186 data_time: 0.3781 memory: 6318 loss: 0.1051 +2023/06/06 08:10:07 - mmengine - INFO - Epoch(train) [5][3100/3937] lr: 5.8007e-05 eta: 3:06:45 time: 0.5408 data_time: 0.4018 memory: 6318 loss: 0.1064 +2023/06/06 08:10:59 - mmengine - INFO - Epoch(train) [5][3200/3937] lr: 5.7649e-05 eta: 3:05:47 time: 0.5261 data_time: 0.3858 memory: 6318 loss: 0.1119 +2023/06/06 08:11:25 - mmengine - INFO - Exp name: resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1_20230606_051539 +2023/06/06 08:11:50 - mmengine - INFO - Epoch(train) [5][3300/3937] lr: 5.7290e-05 eta: 3:04:49 time: 0.4934 data_time: 0.3537 memory: 6318 loss: 0.1185 +2023/06/06 08:12:41 - mmengine - INFO - Epoch(train) [5][3400/3937] lr: 5.6931e-05 eta: 3:03:51 time: 0.5185 data_time: 0.3781 memory: 6318 loss: 0.1136 +2023/06/06 08:13:32 - mmengine - INFO - Epoch(train) [5][3500/3937] lr: 5.6572e-05 eta: 3:02:53 time: 0.5417 data_time: 0.3987 memory: 6318 loss: 0.1119 +2023/06/06 08:14:24 - mmengine - INFO - Epoch(train) [5][3600/3937] lr: 5.6214e-05 eta: 3:01:55 time: 0.5275 data_time: 0.3866 memory: 6318 loss: 0.1118 +2023/06/06 08:15:15 - mmengine - INFO - Epoch(train) [5][3700/3937] lr: 5.5855e-05 eta: 3:00:57 time: 0.5254 data_time: 0.3854 memory: 6318 loss: 0.0928 +2023/06/06 08:16:05 - mmengine - INFO - Epoch(train) [5][3800/3937] lr: 5.5496e-05 eta: 2:59:59 time: 0.5167 data_time: 0.3780 memory: 6318 loss: 0.1013 +2023/06/06 08:16:57 - mmengine - INFO - Epoch(train) [5][3900/3937] lr: 5.5136e-05 eta: 2:59:01 time: 0.5152 data_time: 0.3741 memory: 6318 loss: 0.1064 +2023/06/06 08:17:15 - mmengine - INFO - Exp name: resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1_20230606_051539 +2023/06/06 08:17:15 - mmengine - INFO - Saving checkpoint at 5 epochs +2023/06/06 08:17:54 - mmengine - INFO - Epoch(val) [5][57/57] accuracy/top1: 94.8575 data_time: 0.3959 time: 0.4819 +2023/06/06 08:18:47 - mmengine - INFO - Epoch(train) [6][ 100/3937] lr: 5.4645e-05 eta: 2:57:43 time: 0.4448 data_time: 0.3041 memory: 6318 loss: 0.1017 +2023/06/06 08:19:38 - mmengine - INFO - Epoch(train) [6][ 200/3937] lr: 5.4285e-05 eta: 2:56:45 time: 0.5063 data_time: 0.3651 memory: 6318 loss: 0.1123 +2023/06/06 08:20:27 - mmengine - INFO - Epoch(train) [6][ 300/3937] lr: 5.3926e-05 eta: 2:55:46 time: 0.4770 data_time: 0.3365 memory: 6318 loss: 0.0997 +2023/06/06 08:20:36 - mmengine - INFO - Exp name: resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1_20230606_051539 +2023/06/06 08:21:18 - mmengine - INFO - Epoch(train) [6][ 400/3937] lr: 5.3567e-05 eta: 2:54:49 time: 0.4970 data_time: 0.3476 memory: 6318 loss: 0.1072 +2023/06/06 08:22:11 - mmengine - INFO - Epoch(train) [6][ 500/3937] lr: 5.3209e-05 eta: 2:53:52 time: 0.4878 data_time: 0.3484 memory: 6318 loss: 0.0981 +2023/06/06 08:23:02 - mmengine - INFO - Epoch(train) [6][ 600/3937] lr: 5.2850e-05 eta: 2:52:55 time: 0.4863 data_time: 0.3465 memory: 6318 loss: 0.1041 +2023/06/06 08:23:54 - mmengine - INFO - Epoch(train) [6][ 700/3937] lr: 5.2491e-05 eta: 2:51:59 time: 0.6157 data_time: 0.4766 memory: 6318 loss: 0.1127 +2023/06/06 08:24:47 - mmengine - INFO - Epoch(train) [6][ 800/3937] lr: 5.2133e-05 eta: 2:51:03 time: 0.5636 data_time: 0.4233 memory: 6318 loss: 0.0979 +2023/06/06 08:25:38 - mmengine - INFO - Epoch(train) [6][ 900/3937] lr: 5.1775e-05 eta: 2:50:06 time: 0.5405 data_time: 0.4006 memory: 6318 loss: 0.1037 +2023/06/06 08:26:30 - mmengine - INFO - Epoch(train) [6][1000/3937] lr: 5.1417e-05 eta: 2:49:09 time: 0.5186 data_time: 0.3783 memory: 6318 loss: 0.0983 +2023/06/06 08:27:22 - mmengine - INFO - Epoch(train) [6][1100/3937] lr: 5.1059e-05 eta: 2:48:13 time: 0.5142 data_time: 0.3742 memory: 6318 loss: 0.1162 +2023/06/06 08:28:15 - mmengine - INFO - Epoch(train) [6][1200/3937] lr: 5.0701e-05 eta: 2:47:17 time: 0.5352 data_time: 0.3945 memory: 6318 loss: 0.0945 +2023/06/06 08:29:06 - mmengine - INFO - Epoch(train) [6][1300/3937] lr: 5.0344e-05 eta: 2:46:20 time: 0.5427 data_time: 0.4024 memory: 6318 loss: 0.1034 +2023/06/06 08:29:12 - mmengine - INFO - Exp name: resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1_20230606_051539 +2023/06/06 08:29:58 - mmengine - INFO - Epoch(train) [6][1400/3937] lr: 4.9987e-05 eta: 2:45:24 time: 0.5366 data_time: 0.3955 memory: 6318 loss: 0.1004 +2023/06/06 08:30:53 - mmengine - INFO - Epoch(train) [6][1500/3937] lr: 4.9630e-05 eta: 2:44:30 time: 0.7973 data_time: 0.6567 memory: 6318 loss: 0.1043 +2023/06/06 08:31:46 - mmengine - INFO - Epoch(train) [6][1600/3937] lr: 4.9274e-05 eta: 2:43:34 time: 0.5309 data_time: 0.3899 memory: 6318 loss: 0.1053 +2023/06/06 08:32:36 - mmengine - INFO - Epoch(train) [6][1700/3937] lr: 4.8918e-05 eta: 2:42:37 time: 0.5113 data_time: 0.3708 memory: 6318 loss: 0.1106 +2023/06/06 08:33:26 - mmengine - INFO - Epoch(train) [6][1800/3937] lr: 4.8562e-05 eta: 2:41:39 time: 0.5040 data_time: 0.3638 memory: 6318 loss: 0.1211 +2023/06/06 08:34:19 - mmengine - INFO - Epoch(train) [6][1900/3937] lr: 4.8207e-05 eta: 2:40:43 time: 0.5424 data_time: 0.4025 memory: 6318 loss: 0.1014 +2023/06/06 08:35:11 - mmengine - INFO - Epoch(train) [6][2000/3937] lr: 4.7852e-05 eta: 2:39:47 time: 0.4983 data_time: 0.3580 memory: 6318 loss: 0.1176 +2023/06/06 08:36:02 - mmengine - INFO - Epoch(train) [6][2100/3937] lr: 4.7498e-05 eta: 2:38:51 time: 0.5116 data_time: 0.3715 memory: 6318 loss: 0.0972 +2023/06/06 08:36:54 - mmengine - INFO - Epoch(train) [6][2200/3937] lr: 4.7144e-05 eta: 2:37:55 time: 0.5229 data_time: 0.3824 memory: 6318 loss: 0.1076 +2023/06/06 08:37:44 - mmengine - INFO - Epoch(train) [6][2300/3937] lr: 4.6791e-05 eta: 2:36:58 time: 0.5383 data_time: 0.3979 memory: 6318 loss: 0.1107 +2023/06/06 08:37:51 - mmengine - INFO - Exp name: resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1_20230606_051539 +2023/06/06 08:38:36 - mmengine - INFO - Epoch(train) [6][2400/3937] lr: 4.6438e-05 eta: 2:36:01 time: 0.5162 data_time: 0.3760 memory: 6318 loss: 0.1006 +2023/06/06 08:39:32 - mmengine - INFO - Epoch(train) [6][2500/3937] lr: 4.6086e-05 eta: 2:35:09 time: 0.5263 data_time: 0.3849 memory: 6318 loss: 0.1085 +2023/06/06 08:40:20 - mmengine - INFO - Epoch(train) [6][2600/3937] lr: 4.5734e-05 eta: 2:34:09 time: 0.4587 data_time: 0.3120 memory: 6318 loss: 0.1082 +2023/06/06 08:41:08 - mmengine - INFO - Epoch(train) [6][2700/3937] lr: 4.5383e-05 eta: 2:33:11 time: 0.4705 data_time: 0.3301 memory: 6318 loss: 0.1032 +2023/06/06 08:41:56 - mmengine - INFO - Epoch(train) [6][2800/3937] lr: 4.5033e-05 eta: 2:32:12 time: 0.5014 data_time: 0.3610 memory: 6318 loss: 0.1047 +2023/06/06 08:42:43 - mmengine - INFO - Epoch(train) [6][2900/3937] lr: 4.4683e-05 eta: 2:31:13 time: 0.4705 data_time: 0.3295 memory: 6318 loss: 0.1073 +2023/06/06 08:43:32 - mmengine - INFO - Epoch(train) [6][3000/3937] lr: 4.4334e-05 eta: 2:30:15 time: 0.4635 data_time: 0.3223 memory: 6318 loss: 0.1008 +2023/06/06 08:44:20 - mmengine - INFO - Epoch(train) [6][3100/3937] lr: 4.3985e-05 eta: 2:29:16 time: 0.4707 data_time: 0.3296 memory: 6318 loss: 0.1005 +2023/06/06 08:45:08 - mmengine - INFO - Epoch(train) [6][3200/3937] lr: 4.3637e-05 eta: 2:28:18 time: 0.4723 data_time: 0.3312 memory: 6318 loss: 0.0981 +2023/06/06 08:45:57 - mmengine - INFO - Epoch(train) [6][3300/3937] lr: 4.3290e-05 eta: 2:27:21 time: 0.4898 data_time: 0.3498 memory: 6318 loss: 0.0869 +2023/06/06 08:46:03 - mmengine - INFO - Exp name: resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1_20230606_051539 +2023/06/06 08:46:45 - mmengine - INFO - Epoch(train) [6][3400/3937] lr: 4.2944e-05 eta: 2:26:23 time: 0.4689 data_time: 0.3280 memory: 6318 loss: 0.1017 +2023/06/06 08:47:34 - mmengine - INFO - Epoch(train) [6][3500/3937] lr: 4.2598e-05 eta: 2:25:25 time: 0.4516 data_time: 0.3117 memory: 6318 loss: 0.1040 +2023/06/06 08:48:23 - mmengine - INFO - Epoch(train) [6][3600/3937] lr: 4.2253e-05 eta: 2:24:28 time: 0.4694 data_time: 0.3300 memory: 6318 loss: 0.1043 +2023/06/06 08:49:12 - mmengine - INFO - Epoch(train) [6][3700/3937] lr: 4.1909e-05 eta: 2:23:30 time: 0.4968 data_time: 0.2668 memory: 6318 loss: 0.1057 +2023/06/06 08:49:59 - mmengine - INFO - Epoch(train) [6][3800/3937] lr: 4.1566e-05 eta: 2:22:32 time: 0.4473 data_time: 0.2770 memory: 6318 loss: 0.1044 +2023/06/06 08:50:47 - mmengine - INFO - Epoch(train) [6][3900/3937] lr: 4.1224e-05 eta: 2:21:34 time: 0.4809 data_time: 0.2977 memory: 6318 loss: 0.1120 +2023/06/06 08:51:06 - mmengine - INFO - Exp name: resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1_20230606_051539 +2023/06/06 08:51:06 - mmengine - INFO - Saving checkpoint at 6 epochs +2023/06/06 08:51:43 - mmengine - INFO - Epoch(val) [6][57/57] accuracy/top1: 95.6578 data_time: 0.3769 time: 0.4641 +2023/06/06 08:52:34 - mmengine - INFO - Epoch(train) [7][ 100/3937] lr: 4.0757e-05 eta: 2:20:18 time: 0.4974 data_time: 0.3529 memory: 6318 loss: 0.0987 +2023/06/06 08:53:23 - mmengine - INFO - Epoch(train) [7][ 200/3937] lr: 4.0416e-05 eta: 2:19:21 time: 0.4690 data_time: 0.3285 memory: 6318 loss: 0.1043 +2023/06/06 08:54:11 - mmengine - INFO - Epoch(train) [7][ 300/3937] lr: 4.0077e-05 eta: 2:18:24 time: 0.4892 data_time: 0.3483 memory: 6318 loss: 0.1056 +2023/06/06 08:54:52 - mmengine - INFO - Exp name: resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1_20230606_051539 +2023/06/06 08:55:02 - mmengine - INFO - Epoch(train) [7][ 400/3937] lr: 3.9739e-05 eta: 2:17:28 time: 0.4886 data_time: 0.3486 memory: 6318 loss: 0.0884 +2023/06/06 08:55:51 - mmengine - INFO - Epoch(train) [7][ 500/3937] lr: 3.9402e-05 eta: 2:16:31 time: 0.4728 data_time: 0.3242 memory: 6318 loss: 0.0926 +2023/06/06 08:56:39 - mmengine - INFO - Epoch(train) [7][ 600/3937] lr: 3.9065e-05 eta: 2:15:34 time: 0.5062 data_time: 0.1736 memory: 6318 loss: 0.1043 +2023/06/06 08:57:28 - mmengine - INFO - Epoch(train) [7][ 700/3937] lr: 3.8730e-05 eta: 2:14:37 time: 0.4877 data_time: 0.1162 memory: 6318 loss: 0.1075 +2023/06/06 08:58:17 - mmengine - INFO - Epoch(train) [7][ 800/3937] lr: 3.8396e-05 eta: 2:13:41 time: 0.4622 data_time: 0.0617 memory: 6318 loss: 0.1010 +2023/06/06 08:59:06 - mmengine - INFO - Epoch(train) [7][ 900/3937] lr: 3.8062e-05 eta: 2:12:44 time: 0.5068 data_time: 0.1352 memory: 6318 loss: 0.0981 +2023/06/06 08:59:55 - mmengine - INFO - Epoch(train) [7][1000/3937] lr: 3.7730e-05 eta: 2:11:48 time: 0.4476 data_time: 0.1468 memory: 6318 loss: 0.0984 +2023/06/06 09:00:44 - mmengine - INFO - Epoch(train) [7][1100/3937] lr: 3.7399e-05 eta: 2:10:51 time: 0.4717 data_time: 0.3320 memory: 6318 loss: 0.0954 +2023/06/06 09:01:33 - mmengine - INFO - Epoch(train) [7][1200/3937] lr: 3.7069e-05 eta: 2:09:55 time: 0.4521 data_time: 0.3119 memory: 6318 loss: 0.0964 +2023/06/06 09:02:22 - mmengine - INFO - Epoch(train) [7][1300/3937] lr: 3.6741e-05 eta: 2:08:59 time: 0.5170 data_time: 0.3764 memory: 6318 loss: 0.0928 +2023/06/06 09:03:01 - mmengine - INFO - Exp name: resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1_20230606_051539 +2023/06/06 09:03:10 - mmengine - INFO - Epoch(train) [7][1400/3937] lr: 3.6413e-05 eta: 2:08:02 time: 0.4686 data_time: 0.3064 memory: 6318 loss: 0.0961 +2023/06/06 09:03:58 - mmengine - INFO - Epoch(train) [7][1500/3937] lr: 3.6087e-05 eta: 2:07:06 time: 0.4651 data_time: 0.2846 memory: 6318 loss: 0.0884 +2023/06/06 09:04:46 - mmengine - INFO - Epoch(train) [7][1600/3937] lr: 3.5761e-05 eta: 2:06:09 time: 0.5051 data_time: 0.3652 memory: 6318 loss: 0.0990 +2023/06/06 09:05:35 - mmengine - INFO - Epoch(train) [7][1700/3937] lr: 3.5437e-05 eta: 2:05:13 time: 0.4787 data_time: 0.3387 memory: 6318 loss: 0.1021 +2023/06/06 09:06:23 - mmengine - INFO - Epoch(train) [7][1800/3937] lr: 3.5115e-05 eta: 2:04:17 time: 0.4712 data_time: 0.3305 memory: 6318 loss: 0.0942 +2023/06/06 09:07:13 - mmengine - INFO - Epoch(train) [7][1900/3937] lr: 3.4793e-05 eta: 2:03:21 time: 0.5552 data_time: 0.2782 memory: 6318 loss: 0.1006 +2023/06/06 09:08:01 - mmengine - INFO - Epoch(train) [7][2000/3937] lr: 3.4473e-05 eta: 2:02:25 time: 0.4948 data_time: 0.1854 memory: 6318 loss: 0.1051 +2023/06/06 09:08:49 - mmengine - INFO - Epoch(train) [7][2100/3937] lr: 3.4154e-05 eta: 2:01:28 time: 0.4275 data_time: 0.0934 memory: 6318 loss: 0.0947 +2023/06/06 09:09:37 - mmengine - INFO - Epoch(train) [7][2200/3937] lr: 3.3836e-05 eta: 2:00:32 time: 0.4643 data_time: 0.3228 memory: 6318 loss: 0.1072 +2023/06/06 09:10:25 - mmengine - INFO - Epoch(train) [7][2300/3937] lr: 3.3520e-05 eta: 1:59:36 time: 0.4775 data_time: 0.2345 memory: 6318 loss: 0.1121 +2023/06/06 09:11:03 - mmengine - INFO - Exp name: resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1_20230606_051539 +2023/06/06 09:11:13 - mmengine - INFO - Epoch(train) [7][2400/3937] lr: 3.3205e-05 eta: 1:58:40 time: 0.4920 data_time: 0.2082 memory: 6318 loss: 0.1025 +2023/06/06 09:12:05 - mmengine - INFO - Epoch(train) [7][2500/3937] lr: 3.2892e-05 eta: 1:57:45 time: 0.4967 data_time: 0.0349 memory: 6318 loss: 0.1035 +2023/06/06 09:12:53 - mmengine - INFO - Epoch(train) [7][2600/3937] lr: 3.2580e-05 eta: 1:56:50 time: 0.4440 data_time: 0.0009 memory: 6318 loss: 0.0892 +2023/06/06 09:13:43 - mmengine - INFO - Epoch(train) [7][2700/3937] lr: 3.2269e-05 eta: 1:55:55 time: 0.4798 data_time: 0.0118 memory: 6318 loss: 0.1024 +2023/06/06 09:14:31 - mmengine - INFO - Epoch(train) [7][2800/3937] lr: 3.1960e-05 eta: 1:54:59 time: 0.5114 data_time: 0.0010 memory: 6318 loss: 0.0986 +2023/06/06 09:15:21 - mmengine - INFO - Epoch(train) [7][2900/3937] lr: 3.1652e-05 eta: 1:54:04 time: 0.5245 data_time: 0.1300 memory: 6318 loss: 0.1037 +2023/06/06 09:16:08 - mmengine - INFO - Epoch(train) [7][3000/3937] lr: 3.1346e-05 eta: 1:53:08 time: 0.4480 data_time: 0.1224 memory: 6318 loss: 0.1240 +2023/06/06 09:16:56 - mmengine - INFO - Epoch(train) [7][3100/3937] lr: 3.1041e-05 eta: 1:52:12 time: 0.4890 data_time: 0.1909 memory: 6318 loss: 0.0977 +2023/06/06 09:17:46 - mmengine - INFO - Epoch(train) [7][3200/3937] lr: 3.0738e-05 eta: 1:51:17 time: 0.5318 data_time: 0.3903 memory: 6318 loss: 0.0990 +2023/06/06 09:18:34 - mmengine - INFO - Epoch(train) [7][3300/3937] lr: 3.0437e-05 eta: 1:50:22 time: 0.4417 data_time: 0.2983 memory: 6318 loss: 0.0883 +2023/06/06 09:19:16 - mmengine - INFO - Exp name: resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1_20230606_051539 +2023/06/06 09:19:25 - mmengine - INFO - Epoch(train) [7][3400/3937] lr: 3.0136e-05 eta: 1:49:27 time: 0.4484 data_time: 0.3087 memory: 6318 loss: 0.0871 +2023/06/06 09:20:14 - mmengine - INFO - Epoch(train) [7][3500/3937] lr: 2.9838e-05 eta: 1:48:32 time: 0.4788 data_time: 0.3389 memory: 6318 loss: 0.1077 +2023/06/06 09:21:02 - mmengine - INFO - Epoch(train) [7][3600/3937] lr: 2.9541e-05 eta: 1:47:37 time: 0.4860 data_time: 0.3454 memory: 6318 loss: 0.1038 +2023/06/06 09:21:50 - mmengine - INFO - Epoch(train) [7][3700/3937] lr: 2.9246e-05 eta: 1:46:42 time: 0.4617 data_time: 0.3214 memory: 6318 loss: 0.1015 +2023/06/06 09:22:39 - mmengine - INFO - Epoch(train) [7][3800/3937] lr: 2.8952e-05 eta: 1:45:46 time: 0.4959 data_time: 0.3556 memory: 6318 loss: 0.1085 +2023/06/06 09:23:33 - mmengine - INFO - Epoch(train) [7][3900/3937] lr: 2.8660e-05 eta: 1:44:54 time: 0.9999 data_time: 0.8572 memory: 6318 loss: 0.0842 +2023/06/06 09:23:51 - mmengine - INFO - Exp name: resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1_20230606_051539 +2023/06/06 09:23:51 - mmengine - INFO - Saving checkpoint at 7 epochs +2023/06/06 09:24:29 - mmengine - INFO - Epoch(val) [7][57/57] accuracy/top1: 95.9779 data_time: 0.3753 time: 0.4617 +2023/06/06 09:25:27 - mmengine - INFO - Epoch(train) [8][ 100/3937] lr: 2.8263e-05 eta: 1:43:42 time: 0.4893 data_time: 0.3492 memory: 6318 loss: 0.1015 +2023/06/06 09:26:17 - mmengine - INFO - Epoch(train) [8][ 200/3937] lr: 2.7975e-05 eta: 1:42:48 time: 0.4971 data_time: 0.3564 memory: 6318 loss: 0.0957 +2023/06/06 09:27:06 - mmengine - INFO - Epoch(train) [8][ 300/3937] lr: 2.7689e-05 eta: 1:41:53 time: 0.4953 data_time: 0.3549 memory: 6318 loss: 0.0953 +2023/06/06 09:27:54 - mmengine - INFO - Epoch(train) [8][ 400/3937] lr: 2.7404e-05 eta: 1:40:58 time: 0.4518 data_time: 0.3116 memory: 6318 loss: 0.0931 +2023/06/06 09:28:16 - mmengine - INFO - Exp name: resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1_20230606_051539 +2023/06/06 09:28:43 - mmengine - INFO - Epoch(train) [8][ 500/3937] lr: 2.7121e-05 eta: 1:40:03 time: 0.4707 data_time: 0.3314 memory: 6318 loss: 0.1123 +2023/06/06 09:29:30 - mmengine - INFO - Epoch(train) [8][ 600/3937] lr: 2.6840e-05 eta: 1:39:08 time: 0.4655 data_time: 0.3272 memory: 6318 loss: 0.0829 +2023/06/06 09:30:18 - mmengine - INFO - Epoch(train) [8][ 700/3937] lr: 2.6561e-05 eta: 1:38:13 time: 0.4717 data_time: 0.3334 memory: 6318 loss: 0.0996 +2023/06/06 09:31:06 - mmengine - INFO - Epoch(train) [8][ 800/3937] lr: 2.6284e-05 eta: 1:37:18 time: 0.4751 data_time: 0.3352 memory: 6318 loss: 0.0895 +2023/06/06 09:31:55 - mmengine - INFO - Epoch(train) [8][ 900/3937] lr: 2.6008e-05 eta: 1:36:23 time: 0.4863 data_time: 0.3455 memory: 6318 loss: 0.0949 +2023/06/06 09:32:43 - mmengine - INFO - Epoch(train) [8][1000/3937] lr: 2.5735e-05 eta: 1:35:28 time: 0.4779 data_time: 0.3382 memory: 6318 loss: 0.0884 +2023/06/06 09:33:32 - mmengine - INFO - Epoch(train) [8][1100/3937] lr: 2.5463e-05 eta: 1:34:34 time: 0.4916 data_time: 0.3517 memory: 6318 loss: 0.0989 +2023/06/06 09:34:20 - mmengine - INFO - Epoch(train) [8][1200/3937] lr: 2.5193e-05 eta: 1:33:39 time: 0.4995 data_time: 0.3603 memory: 6318 loss: 0.1010 +2023/06/06 09:35:08 - mmengine - INFO - Epoch(train) [8][1300/3937] lr: 2.4925e-05 eta: 1:32:44 time: 0.4772 data_time: 0.3378 memory: 6318 loss: 0.1026 +2023/06/06 09:35:57 - mmengine - INFO - Epoch(train) [8][1400/3937] lr: 2.4659e-05 eta: 1:31:50 time: 0.4885 data_time: 0.3492 memory: 6318 loss: 0.1079 +2023/06/06 09:36:17 - mmengine - INFO - Exp name: resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1_20230606_051539 +2023/06/06 09:36:45 - mmengine - INFO - Epoch(train) [8][1500/3937] lr: 2.4394e-05 eta: 1:30:55 time: 0.5131 data_time: 0.3725 memory: 6318 loss: 0.0903 +2023/06/06 09:37:33 - mmengine - INFO - Epoch(train) [8][1600/3937] lr: 2.4132e-05 eta: 1:30:01 time: 0.4856 data_time: 0.3452 memory: 6318 loss: 0.0912 +2023/06/06 09:38:22 - mmengine - INFO - Epoch(train) [8][1700/3937] lr: 2.3872e-05 eta: 1:29:06 time: 0.4990 data_time: 0.3584 memory: 6318 loss: 0.1099 +2023/06/06 09:39:10 - mmengine - INFO - Epoch(train) [8][1800/3937] lr: 2.3613e-05 eta: 1:28:12 time: 0.4877 data_time: 0.3469 memory: 6318 loss: 0.1033 +2023/06/06 09:39:58 - mmengine - INFO - Epoch(train) [8][1900/3937] lr: 2.3357e-05 eta: 1:27:17 time: 0.4446 data_time: 0.2944 memory: 6318 loss: 0.0841 +2023/06/06 09:40:46 - mmengine - INFO - Epoch(train) [8][2000/3937] lr: 2.3103e-05 eta: 1:26:23 time: 0.4660 data_time: 0.3251 memory: 6318 loss: 0.0903 +2023/06/06 09:41:36 - mmengine - INFO - Epoch(train) [8][2100/3937] lr: 2.2851e-05 eta: 1:25:29 time: 0.5002 data_time: 0.3601 memory: 6318 loss: 0.0948 +2023/06/06 09:42:24 - mmengine - INFO - Epoch(train) [8][2200/3937] lr: 2.2600e-05 eta: 1:24:35 time: 0.4388 data_time: 0.2989 memory: 6318 loss: 0.0793 +2023/06/06 09:43:12 - mmengine - INFO - Epoch(train) [8][2300/3937] lr: 2.2352e-05 eta: 1:23:40 time: 0.5249 data_time: 0.3846 memory: 6318 loss: 0.1010 +2023/06/06 09:44:00 - mmengine - INFO - Epoch(train) [8][2400/3937] lr: 2.2106e-05 eta: 1:22:46 time: 0.4711 data_time: 0.3309 memory: 6318 loss: 0.0983 +2023/06/06 09:44:19 - mmengine - INFO - Exp name: resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1_20230606_051539 +2023/06/06 09:44:48 - mmengine - INFO - Epoch(train) [8][2500/3937] lr: 2.1862e-05 eta: 1:21:52 time: 0.5036 data_time: 0.3639 memory: 6318 loss: 0.0915 +2023/06/06 09:45:35 - mmengine - INFO - Epoch(train) [8][2600/3937] lr: 2.1620e-05 eta: 1:20:57 time: 0.4820 data_time: 0.3417 memory: 6318 loss: 0.0944 +2023/06/06 09:46:22 - mmengine - INFO - Epoch(train) [8][2700/3937] lr: 2.1380e-05 eta: 1:20:03 time: 0.5053 data_time: 0.3642 memory: 6318 loss: 0.0903 +2023/06/06 09:47:10 - mmengine - INFO - Epoch(train) [8][2800/3937] lr: 2.1143e-05 eta: 1:19:09 time: 0.4527 data_time: 0.3121 memory: 6318 loss: 0.1001 +2023/06/06 09:48:08 - mmengine - INFO - Epoch(train) [8][2900/3937] lr: 2.0907e-05 eta: 1:18:18 time: 0.5498 data_time: 0.4081 memory: 6318 loss: 0.0997 +2023/06/06 09:48:58 - mmengine - INFO - Epoch(train) [8][3000/3937] lr: 2.0674e-05 eta: 1:17:24 time: 0.5058 data_time: 0.3659 memory: 6318 loss: 0.1028 +2023/06/06 09:49:45 - mmengine - INFO - Epoch(train) [8][3100/3937] lr: 2.0443e-05 eta: 1:16:30 time: 0.4466 data_time: 0.3062 memory: 6318 loss: 0.1055 +2023/06/06 09:50:33 - mmengine - INFO - Epoch(train) [8][3200/3937] lr: 2.0214e-05 eta: 1:15:36 time: 0.4221 data_time: 0.2819 memory: 6318 loss: 0.1006 +2023/06/06 09:51:21 - mmengine - INFO - Epoch(train) [8][3300/3937] lr: 1.9987e-05 eta: 1:14:42 time: 0.5357 data_time: 0.3933 memory: 6318 loss: 0.1032 +2023/06/06 09:52:10 - mmengine - INFO - Epoch(train) [8][3400/3937] lr: 1.9763e-05 eta: 1:13:48 time: 0.5375 data_time: 0.3967 memory: 6318 loss: 0.0990 +2023/06/06 09:52:30 - mmengine - INFO - Exp name: resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1_20230606_051539 +2023/06/06 09:52:58 - mmengine - INFO - Epoch(train) [8][3500/3937] lr: 1.9541e-05 eta: 1:12:54 time: 0.4961 data_time: 0.3552 memory: 6318 loss: 0.0946 +2023/06/06 09:53:47 - mmengine - INFO - Epoch(train) [8][3600/3937] lr: 1.9321e-05 eta: 1:12:00 time: 0.4753 data_time: 0.3366 memory: 6318 loss: 0.0846 +2023/06/06 09:54:35 - mmengine - INFO - Epoch(train) [8][3700/3937] lr: 1.9103e-05 eta: 1:11:07 time: 0.4752 data_time: 0.3363 memory: 6318 loss: 0.1105 +2023/06/06 09:55:23 - mmengine - INFO - Epoch(train) [8][3800/3937] lr: 1.8888e-05 eta: 1:10:13 time: 0.5039 data_time: 0.3650 memory: 6318 loss: 0.0986 +2023/06/06 09:56:11 - mmengine - INFO - Epoch(train) [8][3900/3937] lr: 1.8675e-05 eta: 1:09:19 time: 0.4995 data_time: 0.3594 memory: 6318 loss: 0.1031 +2023/06/06 09:56:30 - mmengine - INFO - Exp name: resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1_20230606_051539 +2023/06/06 09:56:30 - mmengine - INFO - Saving checkpoint at 8 epochs +2023/06/06 09:57:08 - mmengine - INFO - Epoch(val) [8][57/57] accuracy/top1: 96.0996 data_time: 0.3809 time: 0.4695 +2023/06/06 09:57:58 - mmengine - INFO - Epoch(train) [9][ 100/3937] lr: 1.8386e-05 eta: 1:08:06 time: 0.4895 data_time: 0.2757 memory: 6318 loss: 0.0985 +2023/06/06 09:58:46 - mmengine - INFO - Epoch(train) [9][ 200/3937] lr: 1.8179e-05 eta: 1:07:13 time: 0.4643 data_time: 0.3203 memory: 6318 loss: 0.0889 +2023/06/06 09:59:35 - mmengine - INFO - Epoch(train) [9][ 300/3937] lr: 1.7974e-05 eta: 1:06:19 time: 0.5105 data_time: 0.3589 memory: 6318 loss: 0.0767 +2023/06/06 10:00:24 - mmengine - INFO - Epoch(train) [9][ 400/3937] lr: 1.7771e-05 eta: 1:05:26 time: 0.4834 data_time: 0.3182 memory: 6318 loss: 0.0912 +2023/06/06 10:01:12 - mmengine - INFO - Epoch(train) [9][ 500/3937] lr: 1.7570e-05 eta: 1:04:32 time: 0.4951 data_time: 0.3545 memory: 6318 loss: 0.0979 +2023/06/06 10:01:16 - mmengine - INFO - Exp name: resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1_20230606_051539 +2023/06/06 10:02:03 - mmengine - INFO - Epoch(train) [9][ 600/3937] lr: 1.7372e-05 eta: 1:03:39 time: 0.5042 data_time: 0.3643 memory: 6318 loss: 0.1015 +2023/06/06 10:02:52 - mmengine - INFO - Epoch(train) [9][ 700/3937] lr: 1.7176e-05 eta: 1:02:46 time: 0.4697 data_time: 0.3254 memory: 6318 loss: 0.1068 +2023/06/06 10:03:39 - mmengine - INFO - Epoch(train) [9][ 800/3937] lr: 1.6983e-05 eta: 1:01:53 time: 0.4708 data_time: 0.3297 memory: 6318 loss: 0.0952 +2023/06/06 10:04:28 - mmengine - INFO - Epoch(train) [9][ 900/3937] lr: 1.6792e-05 eta: 1:00:59 time: 0.4917 data_time: 0.1874 memory: 6318 loss: 0.1064 +2023/06/06 10:05:15 - mmengine - INFO - Epoch(train) [9][1000/3937] lr: 1.6604e-05 eta: 1:00:06 time: 0.4402 data_time: 0.2063 memory: 6318 loss: 0.0954 +2023/06/06 10:06:03 - mmengine - INFO - Epoch(train) [9][1100/3937] lr: 1.6418e-05 eta: 0:59:12 time: 0.4709 data_time: 0.3185 memory: 6318 loss: 0.0950 +2023/06/06 10:06:52 - mmengine - INFO - Epoch(train) [9][1200/3937] lr: 1.6234e-05 eta: 0:58:19 time: 0.4505 data_time: 0.1621 memory: 6318 loss: 0.1014 +2023/06/06 10:07:42 - mmengine - INFO - Epoch(train) [9][1300/3937] lr: 1.6053e-05 eta: 0:57:26 time: 0.4600 data_time: 0.0015 memory: 6318 loss: 0.0885 +2023/06/06 10:08:31 - mmengine - INFO - Epoch(train) [9][1400/3937] lr: 1.5874e-05 eta: 0:56:33 time: 0.4757 data_time: 0.0012 memory: 6318 loss: 0.0928 +2023/06/06 10:09:20 - mmengine - INFO - Epoch(train) [9][1500/3937] lr: 1.5698e-05 eta: 0:55:40 time: 0.4927 data_time: 0.0009 memory: 6318 loss: 0.0994 +2023/06/06 10:09:21 - mmengine - INFO - Exp name: resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1_20230606_051539 +2023/06/06 10:10:09 - mmengine - INFO - Epoch(train) [9][1600/3937] lr: 1.5524e-05 eta: 0:54:47 time: 0.4405 data_time: 0.0009 memory: 6318 loss: 0.0883 +2023/06/06 10:10:57 - mmengine - INFO - Epoch(train) [9][1700/3937] lr: 1.5353e-05 eta: 0:53:54 time: 0.4558 data_time: 0.0009 memory: 6318 loss: 0.0957 +2023/06/06 10:11:46 - mmengine - INFO - Epoch(train) [9][1800/3937] lr: 1.5185e-05 eta: 0:53:01 time: 0.4894 data_time: 0.0008 memory: 6318 loss: 0.0951 +2023/06/06 10:12:36 - mmengine - INFO - Epoch(train) [9][1900/3937] lr: 1.5019e-05 eta: 0:52:08 time: 0.5259 data_time: 0.0009 memory: 6318 loss: 0.0923 +2023/06/06 10:13:24 - mmengine - INFO - Epoch(train) [9][2000/3937] lr: 1.4855e-05 eta: 0:51:15 time: 0.4856 data_time: 0.0009 memory: 6318 loss: 0.1047 +2023/06/06 10:14:12 - mmengine - INFO - Epoch(train) [9][2100/3937] lr: 1.4694e-05 eta: 0:50:22 time: 0.4717 data_time: 0.0011 memory: 6318 loss: 0.0939 +2023/06/06 10:15:00 - mmengine - INFO - Epoch(train) [9][2200/3937] lr: 1.4536e-05 eta: 0:49:29 time: 0.4394 data_time: 0.0012 memory: 6318 loss: 0.0965 +2023/06/06 10:15:48 - mmengine - INFO - Epoch(train) [9][2300/3937] lr: 1.4380e-05 eta: 0:48:36 time: 0.5093 data_time: 0.0013 memory: 6318 loss: 0.0903 +2023/06/06 10:16:36 - mmengine - INFO - Epoch(train) [9][2400/3937] lr: 1.4227e-05 eta: 0:47:43 time: 0.4522 data_time: 0.0011 memory: 6318 loss: 0.0929 +2023/06/06 10:17:25 - mmengine - INFO - Epoch(train) [9][2500/3937] lr: 1.4076e-05 eta: 0:46:50 time: 0.5187 data_time: 0.0015 memory: 6318 loss: 0.0882 +2023/06/06 10:17:25 - mmengine - INFO - Exp name: resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1_20230606_051539 +2023/06/06 10:18:12 - mmengine - INFO - Epoch(train) [9][2600/3937] lr: 1.3928e-05 eta: 0:45:57 time: 0.4956 data_time: 0.0011 memory: 6318 loss: 0.0949 +2023/06/06 10:18:59 - mmengine - INFO - Epoch(train) [9][2700/3937] lr: 1.3783e-05 eta: 0:45:04 time: 0.4372 data_time: 0.0010 memory: 6318 loss: 0.0738 +2023/06/06 10:19:47 - mmengine - INFO - Epoch(train) [9][2800/3937] lr: 1.3640e-05 eta: 0:44:11 time: 0.4920 data_time: 0.0564 memory: 6318 loss: 0.1008 +2023/06/06 10:20:35 - mmengine - INFO - Epoch(train) [9][2900/3937] lr: 1.3500e-05 eta: 0:43:18 time: 0.4997 data_time: 0.2096 memory: 6318 loss: 0.0955 +2023/06/06 10:21:24 - mmengine - INFO - Epoch(train) [9][3000/3937] lr: 1.3362e-05 eta: 0:42:25 time: 0.4949 data_time: 0.0522 memory: 6318 loss: 0.1057 +2023/06/06 10:22:12 - mmengine - INFO - Epoch(train) [9][3100/3937] lr: 1.3227e-05 eta: 0:41:32 time: 0.5162 data_time: 0.1237 memory: 6318 loss: 0.0888 +2023/06/06 10:23:00 - mmengine - INFO - Epoch(train) [9][3200/3937] lr: 1.3095e-05 eta: 0:40:39 time: 0.4896 data_time: 0.1300 memory: 6318 loss: 0.0921 +2023/06/06 10:23:53 - mmengine - INFO - Epoch(train) [9][3300/3937] lr: 1.2966e-05 eta: 0:39:47 time: 0.4965 data_time: 0.1501 memory: 6318 loss: 0.0902 +2023/06/06 10:24:42 - mmengine - INFO - Epoch(train) [9][3400/3937] lr: 1.2839e-05 eta: 0:38:55 time: 0.5437 data_time: 0.1679 memory: 6318 loss: 0.1042 +2023/06/06 10:25:37 - mmengine - INFO - Epoch(train) [9][3500/3937] lr: 1.2715e-05 eta: 0:38:03 time: 0.4753 data_time: 0.3287 memory: 6318 loss: 0.0917 +2023/06/06 10:25:42 - mmengine - INFO - Exp name: resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1_20230606_051539 +2023/06/06 10:26:29 - mmengine - INFO - Epoch(train) [9][3600/3937] lr: 1.2593e-05 eta: 0:37:11 time: 0.5048 data_time: 0.3644 memory: 6318 loss: 0.0963 +2023/06/06 10:27:24 - mmengine - INFO - Epoch(train) [9][3700/3937] lr: 1.2474e-05 eta: 0:36:19 time: 0.5743 data_time: 0.4333 memory: 6318 loss: 0.0851 +2023/06/06 10:28:19 - mmengine - INFO - Epoch(train) [9][3800/3937] lr: 1.2358e-05 eta: 0:35:27 time: 0.5269 data_time: 0.2728 memory: 6318 loss: 0.0974 +2023/06/06 10:29:10 - mmengine - INFO - Epoch(train) [9][3900/3937] lr: 1.2245e-05 eta: 0:34:35 time: 0.4858 data_time: 0.3203 memory: 6318 loss: 0.0945 +2023/06/06 10:29:28 - mmengine - INFO - Exp name: resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1_20230606_051539 +2023/06/06 10:29:28 - mmengine - INFO - Saving checkpoint at 9 epochs +2023/06/06 10:30:06 - mmengine - INFO - Epoch(val) [9][57/57] accuracy/top1: 96.6181 data_time: 0.3733 time: 0.4594 +2023/06/06 10:31:00 - mmengine - INFO - Epoch(train) [10][ 100/3937] lr: 1.2094e-05 eta: 0:33:23 time: 0.4406 data_time: 0.3008 memory: 6318 loss: 0.1128 +2023/06/06 10:31:50 - mmengine - INFO - Epoch(train) [10][ 200/3937] lr: 1.1987e-05 eta: 0:32:31 time: 0.5170 data_time: 0.3775 memory: 6318 loss: 0.0882 +2023/06/06 10:32:38 - mmengine - INFO - Epoch(train) [10][ 300/3937] lr: 1.1883e-05 eta: 0:31:38 time: 0.4734 data_time: 0.3339 memory: 6318 loss: 0.0974 +2023/06/06 10:33:27 - mmengine - INFO - Epoch(train) [10][ 400/3937] lr: 1.1781e-05 eta: 0:30:46 time: 0.4880 data_time: 0.3488 memory: 6318 loss: 0.0869 +2023/06/06 10:34:16 - mmengine - INFO - Epoch(train) [10][ 500/3937] lr: 1.1683e-05 eta: 0:29:53 time: 0.4829 data_time: 0.3382 memory: 6318 loss: 0.0848 +2023/06/06 10:34:50 - mmengine - INFO - Exp name: resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1_20230606_051539 +2023/06/06 10:35:05 - mmengine - INFO - Epoch(train) [10][ 600/3937] lr: 1.1587e-05 eta: 0:29:01 time: 0.4530 data_time: 0.3130 memory: 6318 loss: 0.0926 +2023/06/06 10:35:53 - mmengine - INFO - Epoch(train) [10][ 700/3937] lr: 1.1494e-05 eta: 0:28:08 time: 0.4570 data_time: 0.3006 memory: 6318 loss: 0.0889 +2023/06/06 10:36:43 - mmengine - INFO - Epoch(train) [10][ 800/3937] lr: 1.1403e-05 eta: 0:27:16 time: 0.4851 data_time: 0.3454 memory: 6318 loss: 0.0857 +2023/06/06 10:37:31 - mmengine - INFO - Epoch(train) [10][ 900/3937] lr: 1.1316e-05 eta: 0:26:23 time: 0.4892 data_time: 0.3492 memory: 6318 loss: 0.0870 +2023/06/06 10:38:20 - mmengine - INFO - Epoch(train) [10][1000/3937] lr: 1.1231e-05 eta: 0:25:31 time: 0.4914 data_time: 0.3515 memory: 6318 loss: 0.0925 +2023/06/06 10:39:11 - mmengine - INFO - Epoch(train) [10][1100/3937] lr: 1.1149e-05 eta: 0:24:39 time: 0.7087 data_time: 0.5686 memory: 6318 loss: 0.0996 +2023/06/06 10:40:06 - mmengine - INFO - Epoch(train) [10][1200/3937] lr: 1.1070e-05 eta: 0:23:47 time: 0.5362 data_time: 0.3947 memory: 6318 loss: 0.0914 +2023/06/06 10:40:58 - mmengine - INFO - Epoch(train) [10][1300/3937] lr: 1.0993e-05 eta: 0:22:54 time: 0.4398 data_time: 0.2985 memory: 6318 loss: 0.0791 +2023/06/06 10:41:49 - mmengine - INFO - Epoch(train) [10][1400/3937] lr: 1.0920e-05 eta: 0:22:02 time: 0.5093 data_time: 0.3679 memory: 6318 loss: 0.0915 +2023/06/06 10:42:47 - mmengine - INFO - Epoch(train) [10][1500/3937] lr: 1.0849e-05 eta: 0:21:10 time: 1.0797 data_time: 0.7514 memory: 6318 loss: 0.0949 +2023/06/06 10:43:23 - mmengine - INFO - Exp name: resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1_20230606_051539 +2023/06/06 10:43:36 - mmengine - INFO - Epoch(train) [10][1600/3937] lr: 1.0781e-05 eta: 0:20:18 time: 0.4554 data_time: 0.3007 memory: 6318 loss: 0.0859 +2023/06/06 10:44:27 - mmengine - INFO - Epoch(train) [10][1700/3937] lr: 1.0716e-05 eta: 0:19:26 time: 0.4756 data_time: 0.1418 memory: 6318 loss: 0.1055 +2023/06/06 10:45:15 - mmengine - INFO - Epoch(train) [10][1800/3937] lr: 1.0653e-05 eta: 0:18:34 time: 0.4653 data_time: 0.1102 memory: 6318 loss: 0.0865 +2023/06/06 10:46:03 - mmengine - INFO - Epoch(train) [10][1900/3937] lr: 1.0594e-05 eta: 0:17:41 time: 0.4915 data_time: 0.2065 memory: 6318 loss: 0.0876 +2023/06/06 10:46:52 - mmengine - INFO - Epoch(train) [10][2000/3937] lr: 1.0537e-05 eta: 0:16:49 time: 0.4768 data_time: 0.2327 memory: 6318 loss: 0.0855 +2023/06/06 10:47:40 - mmengine - INFO - Epoch(train) [10][2100/3937] lr: 1.0483e-05 eta: 0:15:57 time: 0.5312 data_time: 0.3084 memory: 6318 loss: 0.0829 +2023/06/06 10:48:30 - mmengine - INFO - Epoch(train) [10][2200/3937] lr: 1.0432e-05 eta: 0:15:04 time: 0.5082 data_time: 0.3671 memory: 6318 loss: 0.0817 +2023/06/06 10:49:21 - mmengine - INFO - Epoch(train) [10][2300/3937] lr: 1.0384e-05 eta: 0:14:12 time: 0.4603 data_time: 0.3196 memory: 6318 loss: 0.0890 +2023/06/06 10:50:09 - mmengine - INFO - Epoch(train) [10][2400/3937] lr: 1.0338e-05 eta: 0:13:20 time: 0.4919 data_time: 0.3505 memory: 6318 loss: 0.0840 +2023/06/06 10:50:58 - mmengine - INFO - Epoch(train) [10][2500/3937] lr: 1.0296e-05 eta: 0:12:28 time: 0.4809 data_time: 0.3404 memory: 6318 loss: 0.0957 +2023/06/06 10:51:31 - mmengine - INFO - Exp name: resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1_20230606_051539 +2023/06/06 10:51:47 - mmengine - INFO - Epoch(train) [10][2600/3937] lr: 1.0256e-05 eta: 0:11:36 time: 0.4657 data_time: 0.3240 memory: 6318 loss: 0.0906 +2023/06/06 10:52:36 - mmengine - INFO - Epoch(train) [10][2700/3937] lr: 1.0219e-05 eta: 0:10:43 time: 0.4838 data_time: 0.3422 memory: 6318 loss: 0.0938 +2023/06/06 10:53:27 - mmengine - INFO - Epoch(train) [10][2800/3937] lr: 1.0185e-05 eta: 0:09:51 time: 0.4433 data_time: 0.3036 memory: 6318 loss: 0.0948 +2023/06/06 10:54:17 - mmengine - INFO - Epoch(train) [10][2900/3937] lr: 1.0154e-05 eta: 0:08:59 time: 0.5015 data_time: 0.3606 memory: 6318 loss: 0.0872 +2023/06/06 10:55:07 - mmengine - INFO - Epoch(train) [10][3000/3937] lr: 1.0126e-05 eta: 0:08:07 time: 0.4892 data_time: 0.3489 memory: 6318 loss: 0.0881 +2023/06/06 10:55:57 - mmengine - INFO - Epoch(train) [10][3100/3937] lr: 1.0101e-05 eta: 0:07:15 time: 0.4903 data_time: 0.3498 memory: 6318 loss: 0.1071 +2023/06/06 10:56:45 - mmengine - INFO - Epoch(train) [10][3200/3937] lr: 1.0078e-05 eta: 0:06:23 time: 0.4542 data_time: 0.3144 memory: 6318 loss: 0.1080 +2023/06/06 10:57:34 - mmengine - INFO - Epoch(train) [10][3300/3937] lr: 1.0058e-05 eta: 0:05:31 time: 0.4817 data_time: 0.3398 memory: 6318 loss: 0.1010 +2023/06/06 10:58:24 - mmengine - INFO - Epoch(train) [10][3400/3937] lr: 1.0041e-05 eta: 0:04:39 time: 0.5132 data_time: 0.3725 memory: 6318 loss: 0.1021 +2023/06/06 10:59:14 - mmengine - INFO - Epoch(train) [10][3500/3937] lr: 1.0027e-05 eta: 0:03:47 time: 0.4768 data_time: 0.3359 memory: 6318 loss: 0.0997 +2023/06/06 10:59:48 - mmengine - INFO - Exp name: resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1_20230606_051539 +2023/06/06 11:00:02 - mmengine - INFO - Epoch(train) [10][3600/3937] lr: 1.0016e-05 eta: 0:02:55 time: 0.4510 data_time: 0.3103 memory: 6318 loss: 0.1002 +2023/06/06 11:00:48 - mmengine - INFO - Epoch(train) [10][3700/3937] lr: 1.0008e-05 eta: 0:02:03 time: 0.4924 data_time: 0.3524 memory: 6318 loss: 0.0881 +2023/06/06 11:01:37 - mmengine - INFO - Epoch(train) [10][3800/3937] lr: 1.0003e-05 eta: 0:01:11 time: 0.5142 data_time: 0.3727 memory: 6318 loss: 0.0865 +2023/06/06 11:02:26 - mmengine - INFO - Epoch(train) [10][3900/3937] lr: 1.0000e-05 eta: 0:00:19 time: 0.5273 data_time: 0.3860 memory: 6318 loss: 0.0944 +2023/06/06 11:02:44 - mmengine - INFO - Exp name: resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1_20230606_051539 +2023/06/06 11:02:44 - mmengine - INFO - Saving checkpoint at 10 epochs +2023/06/06 11:03:23 - mmengine - INFO - Epoch(val) [10][57/57] accuracy/top1: 96.4859 data_time: 0.3758 time: 0.4634 diff --git a/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1/20230606_051539/vis_data/20230606_051539.json b/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1/20230606_051539/vis_data/20230606_051539.json new file mode 100644 index 0000000000000000000000000000000000000000..74dcd1eb050eed777b0c77735158f43f6f45d36d --- /dev/null +++ b/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1/20230606_051539/vis_data/20230606_051539.json @@ -0,0 +1,400 @@ +{"lr": 9.999859583040624e-05, "data_time": 0.35135469436645506, "loss": 0.6373136699199676, "time": 0.5047009229660034, "epoch": 1, "memory": 9436, "step": 100} +{"lr": 9.999432653394512e-05, "data_time": 0.40174314975738523, "loss": 0.573208886384964, "time": 0.5414288520812989, "epoch": 1, "memory": 6319, "step": 200} +{"lr": 9.998719223919393e-05, "data_time": 0.48686139583587645, "loss": 0.5214441359043122, "time": 0.6410719394683838, "epoch": 1, "memory": 6319, "step": 300} +{"lr": 9.997719340042649e-05, "data_time": 0.5417542457580566, "loss": 0.44845099151134493, "time": 0.6814687013626098, "epoch": 1, "memory": 6319, "step": 400} +{"lr": 9.996433065431542e-05, "data_time": 0.4060805082321167, "loss": 0.410491606593132, "time": 0.5461702108383178, "epoch": 1, "memory": 6319, "step": 500} +{"lr": 9.994860481989157e-05, "data_time": 0.4788053512573242, "loss": 0.361409854888916, "time": 0.6187043905258178, "epoch": 1, "memory": 6319, "step": 600} +{"lr": 9.993001689849214e-05, "data_time": 0.4635407209396362, "loss": 0.32016042768955233, "time": 0.6049500703811646, "epoch": 1, "memory": 6319, "step": 700} +{"lr": 9.990856807369658e-05, "data_time": 0.5030936002731323, "loss": 0.3190604686737061, "time": 0.6450572490692139, "epoch": 1, "memory": 6319, "step": 800} +{"lr": 9.98842597112514e-05, "data_time": 0.40195062160491946, "loss": 0.3022842317819595, "time": 0.5432976722717285, "epoch": 1, "memory": 6319, "step": 900} +{"lr": 9.985709335898336e-05, "data_time": 0.4612396717071533, "loss": 0.2837853774428368, "time": 0.6018701076507569, "epoch": 1, "memory": 6319, "step": 1000} +{"lr": 9.982707074670047e-05, "data_time": 0.4168861389160156, "loss": 0.2603603795170784, "time": 0.5616236209869385, "epoch": 1, "memory": 6319, "step": 1100} +{"lr": 9.979419378608245e-05, "data_time": 0.44224417209625244, "loss": 0.26175386607646944, "time": 0.582902193069458, "epoch": 1, "memory": 6319, "step": 1200} +{"lr": 9.975846457055832e-05, "data_time": 0.4418195247650146, "loss": 0.25294673144817353, "time": 0.5848827123641968, "epoch": 1, "memory": 6319, "step": 1300} +{"lr": 9.97198853751737e-05, "data_time": 0.43476912975311277, "loss": 0.23805642426013945, "time": 0.5752826452255249, "epoch": 1, "memory": 6319, "step": 1400} +{"lr": 9.967845865644554e-05, "data_time": 0.3910046100616455, "loss": 0.23767492026090623, "time": 0.5428294658660888, "epoch": 1, "memory": 6319, "step": 1500} +{"lr": 9.963418705220597e-05, "data_time": 0.42578513622283937, "loss": 0.23924392610788345, "time": 0.5666382074356079, "epoch": 1, "memory": 6319, "step": 1600} +{"lr": 9.958707338143413e-05, "data_time": 0.428590202331543, "loss": 0.23500001281499863, "time": 0.5770107507705688, "epoch": 1, "memory": 6319, "step": 1700} +{"lr": 9.953712064407689e-05, "data_time": 0.41157917976379393, "loss": 0.23590199053287506, "time": 0.5512240409851075, "epoch": 1, "memory": 6319, "step": 1800} +{"lr": 9.948433202085752e-05, "data_time": 0.47192094326019285, "loss": 0.22650814056396484, "time": 0.6119584083557129, "epoch": 1, "memory": 6319, "step": 1900} +{"lr": 9.942871087307363e-05, "data_time": 0.46183364391326903, "loss": 0.21390928328037262, "time": 0.6011797189712524, "epoch": 1, "memory": 6319, "step": 2000} +{"lr": 9.937026074238262e-05, "data_time": 0.37435388565063477, "loss": 0.22222377955913544, "time": 0.5163755178451538, "epoch": 1, "memory": 6319, "step": 2100} +{"lr": 9.930898535057657e-05, "data_time": 0.41797561645507814, "loss": 0.2177472949028015, "time": 0.5578161716461182, "epoch": 1, "memory": 6319, "step": 2200} +{"lr": 9.924488859934492e-05, "data_time": 0.37576615810394287, "loss": 0.2220224842429161, "time": 0.5166475057601929, "epoch": 1, "memory": 6319, "step": 2300} +{"lr": 9.917797457002633e-05, "data_time": 0.41900794506072997, "loss": 0.20719822645187377, "time": 0.5588636875152588, "epoch": 1, "memory": 6319, "step": 2400} +{"lr": 9.910824752334866e-05, "data_time": 0.4473739147186279, "loss": 0.21015606373548507, "time": 0.5872008323669433, "epoch": 1, "memory": 6319, "step": 2500} +{"lr": 9.903571189915766e-05, "data_time": 0.42905564308166505, "loss": 0.2182312563061714, "time": 0.5721040725708008, "epoch": 1, "memory": 6319, "step": 2600} +{"lr": 9.896037231613424e-05, "data_time": 0.4330267429351807, "loss": 0.21145364344120027, "time": 0.5730567455291748, "epoch": 1, "memory": 6319, "step": 2700} +{"lr": 9.888223357150049e-05, "data_time": 0.40704665184020994, "loss": 0.1990158200263977, "time": 0.5482313394546509, "epoch": 1, "memory": 6319, "step": 2800} +{"lr": 9.880130064071408e-05, "data_time": 0.41959223747253416, "loss": 0.19440122246742247, "time": 0.5595078945159913, "epoch": 1, "memory": 6319, "step": 2900} +{"lr": 9.87175786771517e-05, "data_time": 0.4379056453704834, "loss": 0.19239158928394318, "time": 0.5797731399536132, "epoch": 1, "memory": 6319, "step": 3000} +{"lr": 9.863107301178059e-05, "data_time": 0.4306967735290527, "loss": 0.1904392957687378, "time": 0.5712720870971679, "epoch": 1, "memory": 6319, "step": 3100} +{"lr": 9.854178915281925e-05, "data_time": 0.43600437641143797, "loss": 0.19300480633974076, "time": 0.5763952970504761, "epoch": 1, "memory": 6319, "step": 3200} +{"lr": 9.84497327853868e-05, "data_time": 0.45678741931915284, "loss": 0.18359561264514923, "time": 0.598597502708435, "epoch": 1, "memory": 6319, "step": 3300} +{"lr": 9.83549097711407e-05, "data_time": 0.41577074527740476, "loss": 0.18914546370506286, "time": 0.5554603338241577, "epoch": 1, "memory": 6319, "step": 3400} +{"lr": 9.825732614790392e-05, "data_time": 0.42415382862091067, "loss": 0.20574819892644883, "time": 0.5642972946166992, "epoch": 1, "memory": 6319, "step": 3500} +{"lr": 9.815698812928018e-05, "data_time": 0.42788641452789306, "loss": 0.1987052395939827, "time": 0.5689198732376098, "epoch": 1, "memory": 6319, "step": 3600} +{"lr": 9.805390210425831e-05, "data_time": 0.3883518695831299, "loss": 0.19001712948083876, "time": 0.5294172286987304, "epoch": 1, "memory": 6319, "step": 3700} +{"lr": 9.794807463680567e-05, "data_time": 0.36976687908172606, "loss": 0.18455516248941423, "time": 0.5100427389144897, "epoch": 1, "memory": 6319, "step": 3800} +{"lr": 9.783951246544974e-05, "data_time": 0.38853485584259034, "loss": 0.1877005636692047, "time": 0.5299545288085937, "epoch": 1, "memory": 6319, "step": 3900} +{"accuracy/top1": 65.28304290771484, "data_time": 0.34556789983782854, "time": 0.43776849696510717, "step": 1} +{"lr": 9.768635527139677e-05, "data_time": 0.38695855140686036, "loss": 0.17761119082570076, "time": 0.5509963750839233, "epoch": 2, "memory": 8776, "step": 4037} +{"lr": 9.75713397738286e-05, "data_time": 0.1603771448135376, "loss": 0.16979795843362808, "time": 0.47443265914916993, "epoch": 2, "memory": 6318, "step": 4137} +{"lr": 9.745361356082025e-05, "data_time": 0.0548288106918335, "loss": 0.18406075686216355, "time": 0.5031360864639283, "epoch": 2, "memory": 6318, "step": 4237} +{"lr": 9.733318412854784e-05, "data_time": 0.08445332050323487, "loss": 0.16945173293352128, "time": 0.6180483579635621, "epoch": 2, "memory": 6318, "step": 4337} +{"lr": 9.721005914531412e-05, "data_time": 0.09240703582763672, "loss": 0.16133614033460617, "time": 0.462139368057251, "epoch": 2, "memory": 6318, "step": 4437} +{"lr": 9.708424645106017e-05, "data_time": 0.1672619342803955, "loss": 0.1570103347301483, "time": 0.5158196687698364, "epoch": 2, "memory": 6318, "step": 4537} +{"lr": 9.695575405686603e-05, "data_time": 0.3291741609573364, "loss": 0.17726304531097412, "time": 0.4993450403213501, "epoch": 2, "memory": 6318, "step": 4637} +{"lr": 9.682459014444089e-05, "data_time": 0.36315526962280276, "loss": 0.17917748242616655, "time": 0.5105035066604614, "epoch": 2, "memory": 6318, "step": 4737} +{"lr": 9.669076306560176e-05, "data_time": 0.32734673023223876, "loss": 0.17449757009744643, "time": 0.5395374298095703, "epoch": 2, "memory": 6318, "step": 4837} +{"lr": 9.655428134174183e-05, "data_time": 0.2691281795501709, "loss": 0.18600485920906068, "time": 0.5208764553070069, "epoch": 2, "memory": 6318, "step": 4937} +{"lr": 9.641515366328807e-05, "data_time": 0.3626711845397949, "loss": 0.1755761295557022, "time": 0.5503090143203735, "epoch": 2, "memory": 6318, "step": 5037} +{"lr": 9.627338888914757e-05, "data_time": 0.2300825834274292, "loss": 0.16736908555030822, "time": 0.5431391477584839, "epoch": 2, "memory": 6318, "step": 5137} +{"lr": 9.612899604614378e-05, "data_time": 0.19789693355560303, "loss": 0.16888393759727477, "time": 0.5604245901107788, "epoch": 2, "memory": 6318, "step": 5237} +{"lr": 9.598198432844131e-05, "data_time": 0.288263726234436, "loss": 0.16049372628331185, "time": 0.5551909685134888, "epoch": 2, "memory": 6318, "step": 5337} +{"lr": 9.583236309696081e-05, "data_time": 0.09539401531219482, "loss": 0.17333296090364456, "time": 0.4834871292114258, "epoch": 2, "memory": 6318, "step": 5437} +{"lr": 9.568014187878279e-05, "data_time": 0.162140417098999, "loss": 0.1645927131175995, "time": 0.5141091108322143, "epoch": 2, "memory": 6318, "step": 5537} +{"lr": 9.552533036654103e-05, "data_time": 0.23149263858795166, "loss": 0.16614100635051726, "time": 0.5089569568634034, "epoch": 2, "memory": 6318, "step": 5637} +{"lr": 9.536793841780541e-05, "data_time": 0.3949670076370239, "loss": 0.15926927030086518, "time": 0.537546968460083, "epoch": 2, "memory": 6318, "step": 5737} +{"lr": 9.52079760544543e-05, "data_time": 0.40032660961151123, "loss": 0.16198524683713914, "time": 0.5412147283554077, "epoch": 2, "memory": 6318, "step": 5837} +{"lr": 9.504545346203619e-05, "data_time": 0.35010015964508057, "loss": 0.1635586552321911, "time": 0.4903889179229736, "epoch": 2, "memory": 6318, "step": 5937} +{"lr": 9.488038098912122e-05, "data_time": 0.373197603225708, "loss": 0.15642861276865005, "time": 0.5136023759841919, "epoch": 2, "memory": 6318, "step": 6037} +{"lr": 9.471276914664257e-05, "data_time": 0.39609856605529786, "loss": 0.16319777071475983, "time": 0.536125922203064, "epoch": 2, "memory": 6318, "step": 6137} +{"lr": 9.454262860722667e-05, "data_time": 0.39641575813293456, "loss": 0.1698547422885895, "time": 0.5367683172225952, "epoch": 2, "memory": 6318, "step": 6237} +{"lr": 9.436997020451398e-05, "data_time": 0.3755208969116211, "loss": 0.15329988598823546, "time": 0.5168079376220703, "epoch": 2, "memory": 6318, "step": 6337} +{"lr": 9.419480493246903e-05, "data_time": 0.37643425464630126, "loss": 0.13655854389071465, "time": 0.5169485330581665, "epoch": 2, "memory": 6318, "step": 6437} +{"lr": 9.401714394468036e-05, "data_time": 0.5956509828567504, "loss": 0.14493875205516815, "time": 0.7383939027786255, "epoch": 2, "memory": 6318, "step": 6537} +{"lr": 9.383699855365022e-05, "data_time": 0.41739416122436523, "loss": 0.15211182758212088, "time": 0.55793297290802, "epoch": 2, "memory": 6318, "step": 6637} +{"lr": 9.365438023007469e-05, "data_time": 0.33812565803527833, "loss": 0.16641210317611693, "time": 0.4787991762161255, "epoch": 2, "memory": 6318, "step": 6737} +{"lr": 9.346930060211267e-05, "data_time": 0.41308434009552003, "loss": 0.15445083230733872, "time": 0.5535926342010498, "epoch": 2, "memory": 6318, "step": 6837} +{"lr": 9.328177145464599e-05, "data_time": 0.394673490524292, "loss": 0.15339895337820053, "time": 0.5363324165344239, "epoch": 2, "memory": 6318, "step": 6937} +{"lr": 9.309180472852871e-05, "data_time": 0.39875857830047606, "loss": 0.15161959379911422, "time": 0.5497647285461426, "epoch": 2, "memory": 6318, "step": 7037} +{"lr": 9.289941251982683e-05, "data_time": 0.4293885469436646, "loss": 0.15708278119564056, "time": 0.5705724716186523, "epoch": 2, "memory": 6318, "step": 7137} +{"lr": 9.270460707904827e-05, "data_time": 0.4121288537979126, "loss": 0.13404012173414231, "time": 0.5533035278320313, "epoch": 2, "memory": 6318, "step": 7237} +{"lr": 9.250740081036243e-05, "data_time": 0.418600869178772, "loss": 0.14458792954683303, "time": 0.5596281528472901, "epoch": 2, "memory": 6318, "step": 7337} +{"lr": 9.230780627081088e-05, "data_time": 0.39788527488708497, "loss": 0.1460558496415615, "time": 0.540451192855835, "epoch": 2, "memory": 6318, "step": 7437} +{"lr": 9.210583616950722e-05, "data_time": 0.3776459932327271, "loss": 0.14620409682393073, "time": 0.5185774564743042, "epoch": 2, "memory": 6318, "step": 7537} +{"lr": 9.190150336682842e-05, "data_time": 0.3727187395095825, "loss": 0.1502346560359001, "time": 0.5137011289596558, "epoch": 2, "memory": 6318, "step": 7637} +{"lr": 9.16948208735955e-05, "data_time": 0.39030919075012205, "loss": 0.15273726880550384, "time": 0.5300938367843628, "epoch": 2, "memory": 6318, "step": 7737} +{"lr": 9.148580185024521e-05, "data_time": 0.3838090419769287, "loss": 0.1422838144004345, "time": 0.5244406938552857, "epoch": 2, "memory": 6318, "step": 7837} +{"accuracy/top1": 86.79934692382812, "data_time": 0.24867882399723448, "time": 0.3373623223140322, "step": 2} +{"lr": 9.119567684509738e-05, "data_time": 0.3863436222076416, "loss": 0.14530004188418388, "time": 0.5260138034820556, "epoch": 3, "memory": 6318, "step": 7974} +{"lr": 9.098117366468421e-05, "data_time": 0.3732325553894043, "loss": 0.1224223829805851, "time": 0.5137952089309692, "epoch": 3, "memory": 6318, "step": 8074} +{"lr": 9.076437939539471e-05, "data_time": 0.3858577013015747, "loss": 0.14201032146811485, "time": 0.5259620904922485, "epoch": 3, "memory": 6318, "step": 8174} +{"lr": 9.05453078415294e-05, "data_time": 0.3986901998519897, "loss": 0.12961676418781282, "time": 0.5374616861343384, "epoch": 3, "memory": 6318, "step": 8274} +{"lr": 9.032397295239427e-05, "data_time": 0.3945038080215454, "loss": 0.14835804626345633, "time": 0.534127163887024, "epoch": 3, "memory": 6318, "step": 8374} +{"lr": 9.010038882141241e-05, "data_time": 0.3276252031326294, "loss": 0.13363704234361648, "time": 0.4678341388702393, "epoch": 3, "memory": 6318, "step": 8474} +{"lr": 8.987456968522655e-05, "data_time": 0.3987130880355835, "loss": 0.1381424091756344, "time": 0.539873480796814, "epoch": 3, "memory": 6318, "step": 8574} +{"lr": 8.964652992279262e-05, "data_time": 0.3964304685592651, "loss": 0.15181969180703164, "time": 0.5373668432235718, "epoch": 3, "memory": 6318, "step": 8674} +{"lr": 8.941628405446428e-05, "data_time": 0.368574333190918, "loss": 0.15693323239684104, "time": 0.5083968877792359, "epoch": 3, "memory": 6318, "step": 8774} +{"lr": 8.918384674106797e-05, "data_time": 0.32884399890899657, "loss": 0.1135924369096756, "time": 0.4690085411071777, "epoch": 3, "memory": 6318, "step": 8874} +{"lr": 8.894923278296992e-05, "data_time": 0.38549854755401614, "loss": 0.14542728960514067, "time": 0.5258296251296997, "epoch": 3, "memory": 6318, "step": 8974} +{"lr": 8.871245711913324e-05, "data_time": 0.3975813388824463, "loss": 0.15234951972961425, "time": 0.5375964164733886, "epoch": 3, "memory": 6318, "step": 9074} +{"lr": 8.847353482616703e-05, "data_time": 0.38844666481018064, "loss": 0.12656183168292046, "time": 0.5289930582046509, "epoch": 3, "memory": 6318, "step": 9174} +{"lr": 8.823248111736638e-05, "data_time": 0.3702134132385254, "loss": 0.12871245965361594, "time": 0.510457181930542, "epoch": 3, "memory": 6318, "step": 9274} +{"lr": 8.798931134174327e-05, "data_time": 0.3971710681915283, "loss": 0.1386072114109993, "time": 0.5380873918533325, "epoch": 3, "memory": 6318, "step": 9374} +{"lr": 8.774404098304961e-05, "data_time": 0.40869979858398436, "loss": 0.13455229476094246, "time": 0.5485035419464112, "epoch": 3, "memory": 6318, "step": 9474} +{"lr": 8.749668565879136e-05, "data_time": 0.40019822120666504, "loss": 0.14316971302032472, "time": 0.5402791738510132, "epoch": 3, "memory": 6318, "step": 9574} +{"lr": 8.724726111923386e-05, "data_time": 0.43427319526672364, "loss": 0.14011319503188133, "time": 0.5742615938186646, "epoch": 3, "memory": 6318, "step": 9674} +{"lr": 8.699578324639898e-05, "data_time": 0.3744964122772217, "loss": 0.1291911445558071, "time": 0.514688229560852, "epoch": 3, "memory": 6318, "step": 9774} +{"lr": 8.67422680530539e-05, "data_time": 0.36439154148101804, "loss": 0.13297815918922423, "time": 0.5083678007125855, "epoch": 3, "memory": 6318, "step": 9874} +{"lr": 8.648673168169147e-05, "data_time": 0.39237961769104, "loss": 0.13199843987822532, "time": 0.5391510725021362, "epoch": 3, "memory": 6318, "step": 9974} +{"lr": 8.622919040350244e-05, "data_time": 0.5732649326324463, "loss": 0.1406612977385521, "time": 0.7130593299865723, "epoch": 3, "memory": 6318, "step": 10074} +{"lr": 8.596966061733922e-05, "data_time": 0.3915122032165527, "loss": 0.1336354061961174, "time": 0.5335070371627808, "epoch": 3, "memory": 6318, "step": 10174} +{"lr": 8.570815884867185e-05, "data_time": 0.3767182111740112, "loss": 0.1251736029982567, "time": 0.5171158790588379, "epoch": 3, "memory": 6318, "step": 10274} +{"lr": 8.544470174853563e-05, "data_time": 0.3925620079040527, "loss": 0.11885619238018989, "time": 0.5400431394577027, "epoch": 3, "memory": 6318, "step": 10374} +{"lr": 8.517930609247077e-05, "data_time": 0.39089391231536863, "loss": 0.1369865544140339, "time": 0.5303629159927368, "epoch": 3, "memory": 6318, "step": 10474} +{"lr": 8.491198877945477e-05, "data_time": 0.3661817073822021, "loss": 0.12701885998249055, "time": 0.5065898656845093, "epoch": 3, "memory": 6318, "step": 10574} +{"lr": 8.464276683082565e-05, "data_time": 0.349953818321228, "loss": 0.12582506984472275, "time": 0.49130213260650635, "epoch": 3, "memory": 6318, "step": 10674} +{"lr": 8.437165738919852e-05, "data_time": 0.38394405841827395, "loss": 0.1230903796851635, "time": 0.524141001701355, "epoch": 3, "memory": 6318, "step": 10774} +{"lr": 8.409867771737398e-05, "data_time": 0.3929825067520142, "loss": 0.12072777599096299, "time": 0.5342830181121826, "epoch": 3, "memory": 6318, "step": 10874} +{"lr": 8.382384519723895e-05, "data_time": 0.3879563331604004, "loss": 0.12704971060156822, "time": 0.5280010938644409, "epoch": 3, "memory": 6318, "step": 10974} +{"lr": 8.35471773286596e-05, "data_time": 0.3738083839416504, "loss": 0.12546228989958763, "time": 0.5150954246520996, "epoch": 3, "memory": 6318, "step": 11074} +{"lr": 8.326869172836765e-05, "data_time": 0.35648431777954104, "loss": 0.11281569972634316, "time": 0.4971273422241211, "epoch": 3, "memory": 6318, "step": 11174} +{"lr": 8.298840612883789e-05, "data_time": 0.4172009229660034, "loss": 0.12077075466513634, "time": 0.5572011232376098, "epoch": 3, "memory": 6318, "step": 11274} +{"lr": 8.270633837715962e-05, "data_time": 0.40588066577911375, "loss": 0.11257560476660729, "time": 0.54643874168396, "epoch": 3, "memory": 6318, "step": 11374} +{"lr": 8.242250643390022e-05, "data_time": 0.43177099227905275, "loss": 0.1323128454387188, "time": 0.5713624238967896, "epoch": 3, "memory": 6318, "step": 11474} +{"lr": 8.2136928371961e-05, "data_time": 0.4174847364425659, "loss": 0.12084009796380997, "time": 0.5590145587921143, "epoch": 3, "memory": 6318, "step": 11574} +{"lr": 8.184962237542712e-05, "data_time": 0.3454891681671143, "loss": 0.1324061408638954, "time": 0.5164076566696167, "epoch": 3, "memory": 6318, "step": 11674} +{"lr": 8.156060673840912e-05, "data_time": 0.36376371383666994, "loss": 0.1252976857125759, "time": 0.516312837600708, "epoch": 3, "memory": 6318, "step": 11774} +{"accuracy/top1": 91.90355682373047, "data_time": 0.30805200543896905, "time": 0.39490813222424737, "step": 3} +{"lr": 8.116191337909002e-05, "data_time": 0.40924673080444335, "loss": 0.12510494589805604, "time": 0.549347186088562, "epoch": 4, "memory": 6318, "step": 11911} +{"lr": 8.086891958166228e-05, "data_time": 0.392189621925354, "loss": 0.1404468335211277, "time": 0.5318895101547241, "epoch": 4, "memory": 6318, "step": 12011} +{"lr": 8.057427858966412e-05, "data_time": 0.38721375465393065, "loss": 0.13838243559002877, "time": 0.5272845983505249, "epoch": 4, "memory": 6318, "step": 12111} +{"lr": 8.027800916425949e-05, "data_time": 0.3695343494415283, "loss": 0.11593601852655411, "time": 0.5215901374816895, "epoch": 4, "memory": 6318, "step": 12211} +{"lr": 7.998013017030234e-05, "data_time": 0.4206414699554443, "loss": 0.1300233282148838, "time": 0.5605233192443848, "epoch": 4, "memory": 6318, "step": 12311} +{"lr": 7.968066057513528e-05, "data_time": 0.3648210048675537, "loss": 0.14271003305912017, "time": 0.5050093650817871, "epoch": 4, "memory": 6318, "step": 12411} +{"lr": 7.937961944738199e-05, "data_time": 0.4261133670806885, "loss": 0.12264132052659989, "time": 0.566995644569397, "epoch": 4, "memory": 6318, "step": 12511} +{"lr": 7.907702595573289e-05, "data_time": 0.40063564777374266, "loss": 0.1107692077755928, "time": 0.540999436378479, "epoch": 4, "memory": 6318, "step": 12611} +{"lr": 7.877289936772454e-05, "data_time": 0.3793018341064453, "loss": 0.11471677720546722, "time": 0.5203649997711182, "epoch": 4, "memory": 6318, "step": 12711} +{"lr": 7.8467259048513e-05, "data_time": 0.35439746379852294, "loss": 0.12011936008930206, "time": 0.49799973964691163, "epoch": 4, "memory": 6318, "step": 12811} +{"lr": 7.816012445964079e-05, "data_time": 0.4329079627990723, "loss": 0.11465063616633415, "time": 0.5739084720611572, "epoch": 4, "memory": 6318, "step": 12911} +{"lr": 7.785151515779718e-05, "data_time": 0.3299607038497925, "loss": 0.11752335876226425, "time": 0.470651650428772, "epoch": 4, "memory": 6318, "step": 13011} +{"lr": 7.754145079357376e-05, "data_time": 0.40634267330169677, "loss": 0.12472412064671516, "time": 0.5458418369293213, "epoch": 4, "memory": 6318, "step": 13111} +{"lr": 7.722995111021218e-05, "data_time": 0.41192777156829835, "loss": 0.13826683908700943, "time": 0.5518895626068115, "epoch": 4, "memory": 6318, "step": 13211} +{"lr": 7.691703594234788e-05, "data_time": 0.32203140258789065, "loss": 0.1110983096063137, "time": 0.4627741098403931, "epoch": 4, "memory": 6318, "step": 13311} +{"lr": 7.660272521474673e-05, "data_time": 0.3714022159576416, "loss": 0.12002493664622307, "time": 0.5125049591064453, "epoch": 4, "memory": 6318, "step": 13411} +{"lr": 7.628703894103628e-05, "data_time": 0.3652146100997925, "loss": 0.11274023503065109, "time": 0.506047534942627, "epoch": 4, "memory": 6318, "step": 13511} +{"lr": 7.596999722243168e-05, "data_time": 0.3891819953918457, "loss": 0.10794650241732598, "time": 0.5294886350631713, "epoch": 4, "memory": 6318, "step": 13611} +{"lr": 7.565162024645551e-05, "data_time": 0.3883901834487915, "loss": 0.10773054882884026, "time": 0.5295039176940918, "epoch": 4, "memory": 6318, "step": 13711} +{"lr": 7.533192828565237e-05, "data_time": 0.33973870277404783, "loss": 0.12283506020903587, "time": 0.47932968139648435, "epoch": 4, "memory": 6318, "step": 13811} +{"lr": 7.501094169629802e-05, "data_time": 0.37243316173553465, "loss": 0.13361043035984038, "time": 0.5256778955459595, "epoch": 4, "memory": 6318, "step": 13911} +{"lr": 7.468868091710333e-05, "data_time": 0.38852458000183104, "loss": 0.125302542001009, "time": 0.5282164096832276, "epoch": 4, "memory": 6318, "step": 14011} +{"lr": 7.436516646791276e-05, "data_time": 0.3871271848678589, "loss": 0.11411226466298104, "time": 0.527640962600708, "epoch": 4, "memory": 6318, "step": 14111} +{"lr": 7.404041894839768e-05, "data_time": 0.36572928428649903, "loss": 0.11670994758605957, "time": 0.5059444427490234, "epoch": 4, "memory": 6318, "step": 14211} +{"lr": 7.371445903674494e-05, "data_time": 0.38312344551086425, "loss": 0.10678449794650077, "time": 0.5241638898849488, "epoch": 4, "memory": 6318, "step": 14311} +{"lr": 7.338730748834014e-05, "data_time": 0.3997997522354126, "loss": 0.11111574321985244, "time": 0.5402564287185669, "epoch": 4, "memory": 6318, "step": 14411} +{"lr": 7.305898513444536e-05, "data_time": 0.36980490684509276, "loss": 0.11540628373622894, "time": 0.5111876964569092, "epoch": 4, "memory": 6318, "step": 14511} +{"lr": 7.27295128808741e-05, "data_time": 0.3927323579788208, "loss": 0.11961081027984619, "time": 0.5329862356185913, "epoch": 4, "memory": 6318, "step": 14611} +{"lr": 7.239891170665884e-05, "data_time": 0.3426256895065308, "loss": 0.11875466629862785, "time": 0.4827617645263672, "epoch": 4, "memory": 6318, "step": 14711} +{"lr": 7.206720266271582e-05, "data_time": 0.4176710367202759, "loss": 0.10677655637264252, "time": 0.5592361211776733, "epoch": 4, "memory": 6318, "step": 14811} +{"lr": 7.173440687050449e-05, "data_time": 0.41552095413208007, "loss": 0.10806555673480034, "time": 0.5557098865509034, "epoch": 4, "memory": 6318, "step": 14911} +{"lr": 7.140054552068265e-05, "data_time": 0.37405297756195066, "loss": 0.1164758674800396, "time": 0.5145105838775634, "epoch": 4, "memory": 6318, "step": 15011} +{"lr": 7.106563987175705e-05, "data_time": 0.39530482292175295, "loss": 0.12077103331685066, "time": 0.5350132465362549, "epoch": 4, "memory": 6318, "step": 15111} +{"lr": 7.072971124873013e-05, "data_time": 0.39270269870758057, "loss": 0.11589161828160285, "time": 0.5341485500335693, "epoch": 4, "memory": 6318, "step": 15211} +{"lr": 7.039278104174143e-05, "data_time": 0.4198456764221191, "loss": 0.11728076860308648, "time": 0.5601847410202027, "epoch": 4, "memory": 6318, "step": 15311} +{"lr": 7.005487070470615e-05, "data_time": 0.373486065864563, "loss": 0.10975094363093377, "time": 0.513847827911377, "epoch": 4, "memory": 6318, "step": 15411} +{"lr": 6.971600175394891e-05, "data_time": 0.39898760318756105, "loss": 0.10830138102173806, "time": 0.5398354530334473, "epoch": 4, "memory": 6318, "step": 15511} +{"lr": 6.937619576683373e-05, "data_time": 0.28586156368255616, "loss": 0.09604218527674675, "time": 0.4256209135055542, "epoch": 4, "memory": 6318, "step": 15611} +{"lr": 6.903547438039021e-05, "data_time": 0.4022240161895752, "loss": 0.11488519236445427, "time": 0.547239112854004, "epoch": 4, "memory": 6318, "step": 15711} +{"accuracy/top1": 93.99464416503906, "data_time": 0.3987178555850325, "time": 0.4850656081890238, "step": 4} +{"lr": 6.856723955077045e-05, "data_time": 0.43915011882781985, "loss": 0.12675577998161316, "time": 0.5796188592910767, "epoch": 5, "memory": 6318, "step": 15848} +{"lr": 6.822443541062844e-05, "data_time": 0.40730764865875246, "loss": 0.11382345259189605, "time": 0.553311538696289, "epoch": 5, "memory": 6318, "step": 15948} +{"lr": 6.788078920910325e-05, "data_time": 0.37160983085632326, "loss": 0.11200935617089272, "time": 0.5135897397994995, "epoch": 5, "memory": 6318, "step": 16048} +{"lr": 6.753632282774874e-05, "data_time": 0.4236746311187744, "loss": 0.10685943216085433, "time": 0.5638228416442871, "epoch": 5, "memory": 6318, "step": 16148} +{"lr": 6.719105820034336e-05, "data_time": 0.38102164268493655, "loss": 0.11480921283364295, "time": 0.522061538696289, "epoch": 5, "memory": 6318, "step": 16248} +{"lr": 6.68450173114937e-05, "data_time": 0.418174147605896, "loss": 0.12086585462093354, "time": 0.5582307815551758, "epoch": 5, "memory": 6318, "step": 16348} +{"lr": 6.64982221952347e-05, "data_time": 0.39405083656311035, "loss": 0.11434520930051803, "time": 0.5336458444595337, "epoch": 5, "memory": 6318, "step": 16448} +{"lr": 6.615069493362599e-05, "data_time": 0.390876030921936, "loss": 0.11514672338962555, "time": 0.5414099216461181, "epoch": 5, "memory": 6318, "step": 16548} +{"lr": 6.580245765534682e-05, "data_time": 0.3882064580917358, "loss": 0.11900919079780578, "time": 0.5284212589263916, "epoch": 5, "memory": 6318, "step": 16648} +{"lr": 6.54535325342862e-05, "data_time": 0.38741657733917234, "loss": 0.09771207123994827, "time": 0.5280201435089111, "epoch": 5, "memory": 6318, "step": 16748} +{"lr": 6.510394178813155e-05, "data_time": 0.4036487817764282, "loss": 0.10312446132302285, "time": 0.545156741142273, "epoch": 5, "memory": 6318, "step": 16848} +{"lr": 6.475370767695331e-05, "data_time": 0.3942956686019897, "loss": 0.1205051064491272, "time": 0.5358115196228027, "epoch": 5, "memory": 6318, "step": 16948} +{"lr": 6.440285250178845e-05, "data_time": 0.3663409948348999, "loss": 0.11486078724265099, "time": 0.5059177160263062, "epoch": 5, "memory": 6318, "step": 17048} +{"lr": 6.405139860321955e-05, "data_time": 0.36364991664886476, "loss": 0.10932580530643463, "time": 0.5050055027008057, "epoch": 5, "memory": 6318, "step": 17148} +{"lr": 6.369936835995306e-05, "data_time": 0.38280601501464845, "loss": 0.10711326822638512, "time": 0.5232627868652344, "epoch": 5, "memory": 6318, "step": 17248} +{"lr": 6.334678418739388e-05, "data_time": 0.3948099374771118, "loss": 0.11118068918585777, "time": 0.5349216938018799, "epoch": 5, "memory": 6318, "step": 17348} +{"lr": 6.299366853621796e-05, "data_time": 0.4100250005722046, "loss": 0.11425347253680229, "time": 0.5508455038070679, "epoch": 5, "memory": 6318, "step": 17448} +{"lr": 6.264004389094315e-05, "data_time": 0.37438867092132566, "loss": 0.10442727133631706, "time": 0.5132425785064697, "epoch": 5, "memory": 6318, "step": 17548} +{"lr": 6.22859327684974e-05, "data_time": 0.3545452833175659, "loss": 0.10730602070689202, "time": 0.4946286678314209, "epoch": 5, "memory": 6318, "step": 17648} +{"lr": 6.193135771678478e-05, "data_time": 0.3236617803573608, "loss": 0.09932406432926655, "time": 0.4646850347518921, "epoch": 5, "memory": 6318, "step": 17748} +{"lr": 6.157634131324983e-05, "data_time": 0.33607330322265627, "loss": 0.10781325846910476, "time": 0.4764808177947998, "epoch": 5, "memory": 6318, "step": 17848} +{"lr": 6.12209061634402e-05, "data_time": 0.3357051372528076, "loss": 0.10310060307383537, "time": 0.47526533603668214, "epoch": 5, "memory": 6318, "step": 17948} +{"lr": 6.086507489956686e-05, "data_time": 0.3543382167816162, "loss": 0.10237922705709934, "time": 0.49532651901245117, "epoch": 5, "memory": 6318, "step": 18048} +{"lr": 6.0508870179063345e-05, "data_time": 0.3810558795928955, "loss": 0.10153815671801567, "time": 0.5208540916442871, "epoch": 5, "memory": 6318, "step": 18148} +{"lr": 6.015231468314303e-05, "data_time": 0.3631471633911133, "loss": 0.09847992844879627, "time": 0.5036010980606079, "epoch": 5, "memory": 6318, "step": 18248} +{"lr": 5.979543111535473e-05, "data_time": 0.35880897045135496, "loss": 0.10222495719790459, "time": 0.49943222999572756, "epoch": 5, "memory": 6318, "step": 18348} +{"lr": 5.943824220013694e-05, "data_time": 0.3660011053085327, "loss": 0.09672838598489761, "time": 0.505923056602478, "epoch": 5, "memory": 6318, "step": 18448} +{"lr": 5.908077068137128e-05, "data_time": 0.374019455909729, "loss": 0.09413647279143333, "time": 0.5142240047454834, "epoch": 5, "memory": 6318, "step": 18548} +{"lr": 5.872303932093392e-05, "data_time": 0.4149693727493286, "loss": 0.10969468727707862, "time": 0.5547202825546265, "epoch": 5, "memory": 6318, "step": 18648} +{"lr": 5.836507089724633e-05, "data_time": 0.3780791759490967, "loss": 0.10508418828248978, "time": 0.5185910940170289, "epoch": 5, "memory": 6318, "step": 18748} +{"lr": 5.800688820382498e-05, "data_time": 0.4017800331115723, "loss": 0.10637912973761558, "time": 0.5407953977584838, "epoch": 5, "memory": 6318, "step": 18848} +{"lr": 5.7648514047829945e-05, "data_time": 0.385778021812439, "loss": 0.11188452839851379, "time": 0.5261282920837402, "epoch": 5, "memory": 6318, "step": 18948} +{"lr": 5.7289971248612474e-05, "data_time": 0.3536604642868042, "loss": 0.11847224235534667, "time": 0.4933535099029541, "epoch": 5, "memory": 6318, "step": 19048} +{"lr": 5.693128263626215e-05, "data_time": 0.3780566692352295, "loss": 0.11355301141738891, "time": 0.5185465812683105, "epoch": 5, "memory": 6318, "step": 19148} +{"lr": 5.6572471050153236e-05, "data_time": 0.3987316370010376, "loss": 0.11193219497799874, "time": 0.5417137861251831, "epoch": 5, "memory": 6318, "step": 19248} +{"lr": 5.621355933749018e-05, "data_time": 0.3865906000137329, "loss": 0.11184052303433419, "time": 0.5274548768997193, "epoch": 5, "memory": 6318, "step": 19348} +{"lr": 5.5854570351853106e-05, "data_time": 0.38541455268859864, "loss": 0.0927666824311018, "time": 0.5253933429718017, "epoch": 5, "memory": 6318, "step": 19448} +{"lr": 5.5495526951742315e-05, "data_time": 0.3779834508895874, "loss": 0.10130834355950355, "time": 0.5167313337326049, "epoch": 5, "memory": 6318, "step": 19548} +{"lr": 5.513645199912304e-05, "data_time": 0.37412319183349607, "loss": 0.10644027814269066, "time": 0.5151644706726074, "epoch": 5, "memory": 6318, "step": 19648} +{"accuracy/top1": 94.8575210571289, "data_time": 0.39589020301555766, "time": 0.4818695619188506, "step": 5} +{"lr": 5.464450978670876e-05, "data_time": 0.30408563613891604, "loss": 0.10174725353717803, "time": 0.44475257396698, "epoch": 6, "memory": 6318, "step": 19785} +{"lr": 5.4285451361545876e-05, "data_time": 0.3650784730911255, "loss": 0.11228409856557846, "time": 0.5063369989395141, "epoch": 6, "memory": 6318, "step": 19885} +{"lr": 5.3926438435022166e-05, "data_time": 0.3365126609802246, "loss": 0.09968096762895584, "time": 0.47697179317474364, "epoch": 6, "memory": 6318, "step": 19985} +{"lr": 5.356749386716241e-05, "data_time": 0.3476013422012329, "loss": 0.10715125873684883, "time": 0.49696924686431887, "epoch": 6, "memory": 6318, "step": 20085} +{"lr": 5.320864051363873e-05, "data_time": 0.3483818292617798, "loss": 0.09811747744679451, "time": 0.4877533674240112, "epoch": 6, "memory": 6318, "step": 20185} +{"lr": 5.2849901224315175e-05, "data_time": 0.34649107456207273, "loss": 0.10409680530428886, "time": 0.4863214254379272, "epoch": 6, "memory": 6318, "step": 20285} +{"lr": 5.249129884179276e-05, "data_time": 0.47657930850982666, "loss": 0.11266562156379223, "time": 0.6157445907592773, "epoch": 6, "memory": 6318, "step": 20385} +{"lr": 5.2132856199954985e-05, "data_time": 0.4233354091644287, "loss": 0.09786221906542777, "time": 0.5636491537094116, "epoch": 6, "memory": 6318, "step": 20485} +{"lr": 5.177459612251411e-05, "data_time": 0.40064258575439454, "loss": 0.10372036993503571, "time": 0.5405067682266236, "epoch": 6, "memory": 6318, "step": 20585} +{"lr": 5.141654142155744e-05, "data_time": 0.3782562971115112, "loss": 0.09827804118394852, "time": 0.5185530424118042, "epoch": 6, "memory": 6318, "step": 20685} +{"lr": 5.105871489609512e-05, "data_time": 0.37419517040252687, "loss": 0.11622708737850189, "time": 0.5142244100570679, "epoch": 6, "memory": 6318, "step": 20785} +{"lr": 5.070113933060828e-05, "data_time": 0.3944650411605835, "loss": 0.0945027407258749, "time": 0.5351663827896118, "epoch": 6, "memory": 6318, "step": 20885} +{"lr": 5.0343837493598264e-05, "data_time": 0.4023817777633667, "loss": 0.10340002775192261, "time": 0.5427498102188111, "epoch": 6, "memory": 6318, "step": 20985} +{"lr": 4.998683213613675e-05, "data_time": 0.3955320119857788, "loss": 0.10043730363249778, "time": 0.5366127252578735, "epoch": 6, "memory": 6318, "step": 21085} +{"lr": 4.963014599041739e-05, "data_time": 0.6567476511001586, "loss": 0.10432487577199936, "time": 0.7972772121429443, "epoch": 6, "memory": 6318, "step": 21185} +{"lr": 4.927380176830793e-05, "data_time": 0.3898554563522339, "loss": 0.10532468929886818, "time": 0.5309202671051025, "epoch": 6, "memory": 6318, "step": 21285} +{"lr": 4.89178221599044e-05, "data_time": 0.37075784206390383, "loss": 0.11055532693862916, "time": 0.5113343000411987, "epoch": 6, "memory": 6318, "step": 21385} +{"lr": 4.85622298320861e-05, "data_time": 0.36381983757019043, "loss": 0.1211197942495346, "time": 0.5040167570114136, "epoch": 6, "memory": 6318, "step": 21485} +{"lr": 4.8207047427072355e-05, "data_time": 0.4025398254394531, "loss": 0.10142585933208466, "time": 0.5424248218536377, "epoch": 6, "memory": 6318, "step": 21585} +{"lr": 4.7852297560980876e-05, "data_time": 0.35801057815551757, "loss": 0.11761986836791039, "time": 0.49829421043395994, "epoch": 6, "memory": 6318, "step": 21685} +{"lr": 4.749800282238749e-05, "data_time": 0.37150614261627196, "loss": 0.09721977785229682, "time": 0.5115808725357056, "epoch": 6, "memory": 6318, "step": 21785} +{"lr": 4.714418577088805e-05, "data_time": 0.3824015617370605, "loss": 0.107598315179348, "time": 0.5228906393051147, "epoch": 6, "memory": 6318, "step": 21885} +{"lr": 4.679086893566184e-05, "data_time": 0.39792261123657224, "loss": 0.11070032864809036, "time": 0.5383080005645752, "epoch": 6, "memory": 6318, "step": 21985} +{"lr": 4.643807481403694e-05, "data_time": 0.37602369785308837, "loss": 0.10062609389424323, "time": 0.5161623954772949, "epoch": 6, "memory": 6318, "step": 22085} +{"lr": 4.6085825870057946e-05, "data_time": 0.3849221706390381, "loss": 0.10851487964391708, "time": 0.5263155937194824, "epoch": 6, "memory": 6318, "step": 22185} +{"lr": 4.573414453305537e-05, "data_time": 0.311960506439209, "loss": 0.10821853280067444, "time": 0.458675479888916, "epoch": 6, "memory": 6318, "step": 22285} +{"lr": 4.5383053196217514e-05, "data_time": 0.3300865650177002, "loss": 0.10318279191851616, "time": 0.470539927482605, "epoch": 6, "memory": 6318, "step": 22385} +{"lr": 4.503257421516477e-05, "data_time": 0.360955810546875, "loss": 0.10468805134296418, "time": 0.5014185190200806, "epoch": 6, "memory": 6318, "step": 22485} +{"lr": 4.468272990652586e-05, "data_time": 0.32948927879333495, "loss": 0.10729858800768852, "time": 0.47054457664489746, "epoch": 6, "memory": 6318, "step": 22585} +{"lr": 4.433354254651697e-05, "data_time": 0.32231760025024414, "loss": 0.10081584528088569, "time": 0.463468337059021, "epoch": 6, "memory": 6318, "step": 22685} +{"lr": 4.398503436952338e-05, "data_time": 0.3296447992324829, "loss": 0.10046930015087127, "time": 0.47072160243988037, "epoch": 6, "memory": 6318, "step": 22785} +{"lr": 4.363722756668351e-05, "data_time": 0.33118667602539065, "loss": 0.0980569876730442, "time": 0.4722886085510254, "epoch": 6, "memory": 6318, "step": 22885} +{"lr": 4.329014428447611e-05, "data_time": 0.34980001449584963, "loss": 0.08689046539366245, "time": 0.4898162126541138, "epoch": 6, "memory": 6318, "step": 22985} +{"lr": 4.294380662331e-05, "data_time": 0.3280327558517456, "loss": 0.10173123776912689, "time": 0.4689242601394653, "epoch": 6, "memory": 6318, "step": 23085} +{"lr": 4.259823663611678e-05, "data_time": 0.3117323398590088, "loss": 0.10396727472543717, "time": 0.4515509605407715, "epoch": 6, "memory": 6318, "step": 23185} +{"lr": 4.22534563269468e-05, "data_time": 0.32997186183929444, "loss": 0.10429118499159813, "time": 0.4694223403930664, "epoch": 6, "memory": 6318, "step": 23285} +{"lr": 4.1909487649567694e-05, "data_time": 0.26676576137542723, "loss": 0.1056912124156952, "time": 0.4968456506729126, "epoch": 6, "memory": 6318, "step": 23385} +{"lr": 4.156635250606698e-05, "data_time": 0.2769592761993408, "loss": 0.10443811044096947, "time": 0.44726712703704835, "epoch": 6, "memory": 6318, "step": 23485} +{"lr": 4.122407274545703e-05, "data_time": 0.2976927518844604, "loss": 0.11196320131421089, "time": 0.4809021711349487, "epoch": 6, "memory": 6318, "step": 23585} +{"accuracy/top1": 95.65777587890625, "data_time": 0.3768713720913591, "time": 0.4641124462259227, "step": 6} +{"lr": 4.075657788156285e-05, "data_time": 0.35293667316436766, "loss": 0.0987037293612957, "time": 0.4974281072616577, "epoch": 7, "memory": 6318, "step": 23722} +{"lr": 4.0416412311741566e-05, "data_time": 0.3285355567932129, "loss": 0.10428239926695823, "time": 0.4689664363861084, "epoch": 7, "memory": 6318, "step": 23822} +{"lr": 4.007717534685821e-05, "data_time": 0.34826011657714845, "loss": 0.1055700808763504, "time": 0.489159893989563, "epoch": 7, "memory": 6318, "step": 23922} +{"lr": 3.973888858770982e-05, "data_time": 0.3486127138137817, "loss": 0.08842075392603874, "time": 0.48857219219207765, "epoch": 7, "memory": 6318, "step": 24022} +{"lr": 3.9401573574589535e-05, "data_time": 0.3241807699203491, "loss": 0.09261823780834674, "time": 0.47283239364624025, "epoch": 7, "memory": 6318, "step": 24122} +{"lr": 3.906525178591511e-05, "data_time": 0.1736433982849121, "loss": 0.10432448759675025, "time": 0.5062353610992432, "epoch": 7, "memory": 6318, "step": 24222} +{"lr": 3.87299446368609e-05, "data_time": 0.11616425514221192, "loss": 0.10748223438858986, "time": 0.48773789405822754, "epoch": 7, "memory": 6318, "step": 24322} +{"lr": 3.839567347799442e-05, "data_time": 0.06165766716003418, "loss": 0.10103201270103454, "time": 0.4622232437133789, "epoch": 7, "memory": 6318, "step": 24422} +{"lr": 3.806245959391694e-05, "data_time": 0.13522274494171144, "loss": 0.09810488671064377, "time": 0.506796407699585, "epoch": 7, "memory": 6318, "step": 24522} +{"lr": 3.773032420190816e-05, "data_time": 0.14675233364105225, "loss": 0.0984394371509552, "time": 0.44760453701019287, "epoch": 7, "memory": 6318, "step": 24622} +{"lr": 3.739928845057507e-05, "data_time": 0.33199052810668944, "loss": 0.09539510831236839, "time": 0.47165915966033933, "epoch": 7, "memory": 6318, "step": 24722} +{"lr": 3.706937341850541e-05, "data_time": 0.3119044780731201, "loss": 0.09639893397688866, "time": 0.45209035873413084, "epoch": 7, "memory": 6318, "step": 24822} +{"lr": 3.674060011292556e-05, "data_time": 0.3763544797897339, "loss": 0.09282100200653076, "time": 0.517037034034729, "epoch": 7, "memory": 6318, "step": 24922} +{"lr": 3.641298946836287e-05, "data_time": 0.30643692016601565, "loss": 0.09610395208001137, "time": 0.4686431884765625, "epoch": 7, "memory": 6318, "step": 25022} +{"lr": 3.6086562345312595e-05, "data_time": 0.28463425636291506, "loss": 0.08842328265309334, "time": 0.465103006362915, "epoch": 7, "memory": 6318, "step": 25122} +{"lr": 3.576133952890974e-05, "data_time": 0.3651818513870239, "loss": 0.09897352643311023, "time": 0.5050853967666626, "epoch": 7, "memory": 6318, "step": 25222} +{"lr": 3.543734172760541e-05, "data_time": 0.33870925903320315, "loss": 0.10209830403327942, "time": 0.47871851921081543, "epoch": 7, "memory": 6318, "step": 25322} +{"lr": 3.511458957184831e-05, "data_time": 0.3305222749710083, "loss": 0.09417184218764305, "time": 0.4711876630783081, "epoch": 7, "memory": 6318, "step": 25422} +{"lr": 3.479310361277118e-05, "data_time": 0.2781820774078369, "loss": 0.10061065405607224, "time": 0.5552470207214355, "epoch": 7, "memory": 6318, "step": 25522} +{"lr": 3.4472904320882065e-05, "data_time": 0.1854102849960327, "loss": 0.10505718365311623, "time": 0.4947544813156128, "epoch": 7, "memory": 6318, "step": 25622} +{"lr": 3.4154012084760864e-05, "data_time": 0.09341263771057129, "loss": 0.09473652243614197, "time": 0.4275254011154175, "epoch": 7, "memory": 6318, "step": 25722} +{"lr": 3.38364472097612e-05, "data_time": 0.3228015661239624, "loss": 0.10724164620041847, "time": 0.46433560848236083, "epoch": 7, "memory": 6318, "step": 25822} +{"lr": 3.3520229916717485e-05, "data_time": 0.2344813585281372, "loss": 0.11206526979804039, "time": 0.47754745483398436, "epoch": 7, "memory": 6318, "step": 25922} +{"lr": 3.320538034065722e-05, "data_time": 0.20822627544403077, "loss": 0.10250148177146912, "time": 0.49203219413757326, "epoch": 7, "memory": 6318, "step": 26022} +{"lr": 3.289191852951901e-05, "data_time": 0.03485782146453857, "loss": 0.1034554086625576, "time": 0.4966739177703857, "epoch": 7, "memory": 6318, "step": 26122} +{"lr": 3.257986444287605e-05, "data_time": 0.0009075164794921875, "loss": 0.08920665830373764, "time": 0.44400362968444823, "epoch": 7, "memory": 6318, "step": 26222} +{"lr": 3.226923795066519e-05, "data_time": 0.011821389198303223, "loss": 0.10243339091539383, "time": 0.479843807220459, "epoch": 7, "memory": 6318, "step": 26322} +{"lr": 3.196005883192158e-05, "data_time": 0.001040482521057129, "loss": 0.09856786578893661, "time": 0.5114281177520752, "epoch": 7, "memory": 6318, "step": 26422} +{"lr": 3.165234677351947e-05, "data_time": 0.1300366163253784, "loss": 0.1036761175841093, "time": 0.5245226383209228, "epoch": 7, "memory": 6318, "step": 26522} +{"lr": 3.134612136891845e-05, "data_time": 0.12235636711120605, "loss": 0.1240170106291771, "time": 0.4479642391204834, "epoch": 7, "memory": 6318, "step": 26622} +{"lr": 3.104140211691602e-05, "data_time": 0.19092833995819092, "loss": 0.09769907668232918, "time": 0.4890336275100708, "epoch": 7, "memory": 6318, "step": 26722} +{"lr": 3.073820842040592e-05, "data_time": 0.3902705192565918, "loss": 0.0990387573838234, "time": 0.5317658424377442, "epoch": 7, "memory": 6318, "step": 26822} +{"lr": 3.043655958514262e-05, "data_time": 0.298284649848938, "loss": 0.08826142475008965, "time": 0.4417326211929321, "epoch": 7, "memory": 6318, "step": 26922} +{"lr": 3.013647481851217e-05, "data_time": 0.3087068796157837, "loss": 0.08713078796863556, "time": 0.4484199047088623, "epoch": 7, "memory": 6318, "step": 27022} +{"lr": 2.983797322830899e-05, "data_time": 0.33893604278564454, "loss": 0.10766812637448311, "time": 0.4787930727005005, "epoch": 7, "memory": 6318, "step": 27122} +{"lr": 2.9541073821519346e-05, "data_time": 0.345421576499939, "loss": 0.10380721762776375, "time": 0.4860095024108887, "epoch": 7, "memory": 6318, "step": 27222} +{"lr": 2.924579550311095e-05, "data_time": 0.3214087009429932, "loss": 0.10154414102435112, "time": 0.46173095703125, "epoch": 7, "memory": 6318, "step": 27322} +{"lr": 2.8952157074829418e-05, "data_time": 0.3555649518966675, "loss": 0.10850416794419289, "time": 0.49585533142089844, "epoch": 7, "memory": 6318, "step": 27422} +{"lr": 2.866017723400071e-05, "data_time": 0.8572046279907226, "loss": 0.0841876830905676, "time": 0.9998951911926269, "epoch": 7, "memory": 6318, "step": 27522} +{"accuracy/top1": 95.97787475585938, "data_time": 0.3753005225082924, "time": 0.46172171625597724, "step": 7} +{"lr": 2.8262891373132927e-05, "data_time": 0.3492342233657837, "loss": 0.10151149034500122, "time": 0.48925559520721434, "epoch": 8, "memory": 6318, "step": 27659} +{"lr": 2.797491643470029e-05, "data_time": 0.35636391639709475, "loss": 0.09571295753121375, "time": 0.4970815181732178, "epoch": 8, "memory": 6318, "step": 27759} +{"lr": 2.768866230918266e-05, "data_time": 0.35486526489257814, "loss": 0.095349533110857, "time": 0.49532508850097656, "epoch": 8, "memory": 6318, "step": 27859} +{"lr": 2.7404147223713185e-05, "data_time": 0.31156766414642334, "loss": 0.09307517856359482, "time": 0.451794958114624, "epoch": 8, "memory": 6318, "step": 27959} +{"lr": 2.712138929469222e-05, "data_time": 0.33141589164733887, "loss": 0.11230526641011238, "time": 0.47066478729248046, "epoch": 8, "memory": 6318, "step": 28059} +{"lr": 2.6840406526633896e-05, "data_time": 0.3271724462509155, "loss": 0.08288054689764976, "time": 0.46551098823547366, "epoch": 8, "memory": 6318, "step": 28159} +{"lr": 2.6561216811019345e-05, "data_time": 0.3334389925003052, "loss": 0.0996477335691452, "time": 0.4717499494552612, "epoch": 8, "memory": 6318, "step": 28259} +{"lr": 2.6283837925157908e-05, "data_time": 0.3351886749267578, "loss": 0.08946491815149785, "time": 0.475070858001709, "epoch": 8, "memory": 6318, "step": 28359} +{"lr": 2.6008287531054805e-05, "data_time": 0.34549751281738283, "loss": 0.09487205892801284, "time": 0.4862659931182861, "epoch": 8, "memory": 6318, "step": 28459} +{"lr": 2.5734583174286796e-05, "data_time": 0.3381718635559082, "loss": 0.0884143128991127, "time": 0.4779308319091797, "epoch": 8, "memory": 6318, "step": 28559} +{"lr": 2.546274228288488e-05, "data_time": 0.35170581340789797, "loss": 0.098910591006279, "time": 0.491578197479248, "epoch": 8, "memory": 6318, "step": 28659} +{"lr": 2.5192782166224308e-05, "data_time": 0.36029911041259766, "loss": 0.10099406689405441, "time": 0.4994908571243286, "epoch": 8, "memory": 6318, "step": 28759} +{"lr": 2.492472001392276e-05, "data_time": 0.3378411054611206, "loss": 0.10261294022202491, "time": 0.477240777015686, "epoch": 8, "memory": 6318, "step": 28859} +{"lr": 2.4658572894745857e-05, "data_time": 0.3492377758026123, "loss": 0.10785484537482262, "time": 0.4885077476501465, "epoch": 8, "memory": 6318, "step": 28959} +{"lr": 2.439435775551997e-05, "data_time": 0.3725048780441284, "loss": 0.09029463529586793, "time": 0.5131029605865478, "epoch": 8, "memory": 6318, "step": 29059} +{"lr": 2.4132091420053357e-05, "data_time": 0.3452200412750244, "loss": 0.09120306968688965, "time": 0.48561174869537355, "epoch": 8, "memory": 6318, "step": 29159} +{"lr": 2.387179058806494e-05, "data_time": 0.35843653678894044, "loss": 0.10985691845417023, "time": 0.49903619289398193, "epoch": 8, "memory": 6318, "step": 29259} +{"lr": 2.361347183412073e-05, "data_time": 0.3468772411346436, "loss": 0.10330413170158863, "time": 0.487683629989624, "epoch": 8, "memory": 6318, "step": 29359} +{"lr": 2.3357151606578728e-05, "data_time": 0.2944147825241089, "loss": 0.08409390300512314, "time": 0.44464750289916993, "epoch": 8, "memory": 6318, "step": 29459} +{"lr": 2.3102846226541355e-05, "data_time": 0.32512702941894533, "loss": 0.0902996003627777, "time": 0.4659524440765381, "epoch": 8, "memory": 6318, "step": 29559} +{"lr": 2.2850571886816365e-05, "data_time": 0.3600844144821167, "loss": 0.09483547136187553, "time": 0.5001818418502808, "epoch": 8, "memory": 6318, "step": 29659} +{"lr": 2.260034465088568e-05, "data_time": 0.2988943099975586, "loss": 0.079250842705369, "time": 0.43875305652618407, "epoch": 8, "memory": 6318, "step": 29759} +{"lr": 2.2352180451882643e-05, "data_time": 0.38458776473999023, "loss": 0.10099021680653095, "time": 0.524933123588562, "epoch": 8, "memory": 6318, "step": 29859} +{"lr": 2.210609509157741e-05, "data_time": 0.33091471195220945, "loss": 0.09827216044068336, "time": 0.4710878849029541, "epoch": 8, "memory": 6318, "step": 29959} +{"lr": 2.1862104239370772e-05, "data_time": 0.3638679265975952, "loss": 0.09145785272121429, "time": 0.5036179780960083, "epoch": 8, "memory": 6318, "step": 30059} +{"lr": 2.1620223431296407e-05, "data_time": 0.3417232513427734, "loss": 0.09443091079592705, "time": 0.4819511651992798, "epoch": 8, "memory": 6318, "step": 30159} +{"lr": 2.1380468069031744e-05, "data_time": 0.36420743465423583, "loss": 0.09029760658740997, "time": 0.505265760421753, "epoch": 8, "memory": 6318, "step": 30259} +{"lr": 2.1142853418917185e-05, "data_time": 0.312066650390625, "loss": 0.10014737099409103, "time": 0.4526773691177368, "epoch": 8, "memory": 6318, "step": 30359} +{"lr": 2.0907394610983892e-05, "data_time": 0.40809338092803954, "loss": 0.09974495694041252, "time": 0.5498457670211792, "epoch": 8, "memory": 6318, "step": 30459} +{"lr": 2.0674106637990643e-05, "data_time": 0.3659006118774414, "loss": 0.10275045484304428, "time": 0.5057721853256225, "epoch": 8, "memory": 6318, "step": 30559} +{"lr": 2.0443004354468975e-05, "data_time": 0.306179404258728, "loss": 0.1055156260728836, "time": 0.4466289281845093, "epoch": 8, "memory": 6318, "step": 30659} +{"lr": 2.0214102475777453e-05, "data_time": 0.2818570852279663, "loss": 0.10063871070742607, "time": 0.42209014892578123, "epoch": 8, "memory": 6318, "step": 30759} +{"lr": 1.9987415577164583e-05, "data_time": 0.39331493377685545, "loss": 0.10316462069749832, "time": 0.5357325077056885, "epoch": 8, "memory": 6318, "step": 30859} +{"lr": 1.9762958092840814e-05, "data_time": 0.39673705101013185, "loss": 0.09898230955004692, "time": 0.5374812602996826, "epoch": 8, "memory": 6318, "step": 30959} +{"lr": 1.9540744315059334e-05, "data_time": 0.35516085624694826, "loss": 0.09462017826735973, "time": 0.4960510730743408, "epoch": 8, "memory": 6318, "step": 31059} +{"lr": 1.9320788393206132e-05, "data_time": 0.33659427165985106, "loss": 0.08460660092532635, "time": 0.475318455696106, "epoch": 8, "memory": 6318, "step": 31159} +{"lr": 1.9103104332899027e-05, "data_time": 0.3362543821334839, "loss": 0.11053954884409904, "time": 0.475168514251709, "epoch": 8, "memory": 6318, "step": 31259} +{"lr": 1.888770599509573e-05, "data_time": 0.36495540142059324, "loss": 0.09862048029899598, "time": 0.5039122104644775, "epoch": 8, "memory": 6318, "step": 31359} +{"lr": 1.8674607095211516e-05, "data_time": 0.35937793254852296, "loss": 0.10310119204223156, "time": 0.49950060844421384, "epoch": 8, "memory": 6318, "step": 31459} +{"accuracy/top1": 96.09964752197266, "data_time": 0.3809464265560282, "time": 0.46951562371747246, "step": 8} +{"lr": 1.838641934548488e-05, "data_time": 0.2756743192672729, "loss": 0.0984993189573288, "time": 0.4895286560058594, "epoch": 9, "memory": 6318, "step": 31596} +{"lr": 1.817882403185378e-05, "data_time": 0.3203425884246826, "loss": 0.0889295682311058, "time": 0.4642958164215088, "epoch": 9, "memory": 6318, "step": 31696} +{"lr": 1.797357329395729e-05, "data_time": 0.3588765382766724, "loss": 0.07670582234859466, "time": 0.5104834556579589, "epoch": 9, "memory": 6318, "step": 31796} +{"lr": 1.7770680201065646e-05, "data_time": 0.3181695222854614, "loss": 0.09116340428590775, "time": 0.4833985328674316, "epoch": 9, "memory": 6318, "step": 31896} +{"lr": 1.7570157672326885e-05, "data_time": 0.3544581651687622, "loss": 0.09792966432869435, "time": 0.49512252807617185, "epoch": 9, "memory": 6318, "step": 31996} +{"lr": 1.7372018475944145e-05, "data_time": 0.3643302917480469, "loss": 0.101457005366683, "time": 0.5042122602462769, "epoch": 9, "memory": 6318, "step": 32096} +{"lr": 1.717627522836271e-05, "data_time": 0.3254039764404297, "loss": 0.10682158172130585, "time": 0.4696770668029785, "epoch": 9, "memory": 6318, "step": 32196} +{"lr": 1.6982940393466685e-05, "data_time": 0.3297384738922119, "loss": 0.09523039907217026, "time": 0.4708113670349121, "epoch": 9, "memory": 6318, "step": 32296} +{"lr": 1.6792026281785218e-05, "data_time": 0.18743543624877929, "loss": 0.10642971992492675, "time": 0.49167566299438475, "epoch": 9, "memory": 6318, "step": 32396} +{"lr": 1.660354504970888e-05, "data_time": 0.20626299381256102, "loss": 0.09539871588349343, "time": 0.4402266263961792, "epoch": 9, "memory": 6318, "step": 32496} +{"lr": 1.641750869871534e-05, "data_time": 0.3184908390045166, "loss": 0.09498243555426597, "time": 0.47086918354034424, "epoch": 9, "memory": 6318, "step": 32596} +{"lr": 1.6233929074605374e-05, "data_time": 0.16208815574645996, "loss": 0.1013569600880146, "time": 0.45047998428344727, "epoch": 9, "memory": 6318, "step": 32696} +{"lr": 1.6052817866748513e-05, "data_time": 0.0014832019805908203, "loss": 0.08854821175336838, "time": 0.45998849868774416, "epoch": 9, "memory": 6318, "step": 32796} +{"lr": 1.587418660733871e-05, "data_time": 0.0011659860610961914, "loss": 0.09278179705142975, "time": 0.47566611766815187, "epoch": 9, "memory": 6318, "step": 32896} +{"lr": 1.5698046670660044e-05, "data_time": 0.0008747339248657227, "loss": 0.09943257197737694, "time": 0.4926849603652954, "epoch": 9, "memory": 6318, "step": 32996} +{"lr": 1.5524409272362524e-05, "data_time": 0.0009041309356689453, "loss": 0.0882600724697113, "time": 0.440548300743103, "epoch": 9, "memory": 6318, "step": 33096} +{"lr": 1.5353285468747853e-05, "data_time": 0.0009421586990356445, "loss": 0.09573952667415142, "time": 0.45577971935272216, "epoch": 9, "memory": 6318, "step": 33196} +{"lr": 1.5184686156065442e-05, "data_time": 0.0008428096771240234, "loss": 0.0951104611158371, "time": 0.4894315958023071, "epoch": 9, "memory": 6318, "step": 33296} +{"lr": 1.5018622069818636e-05, "data_time": 0.0009381055831909179, "loss": 0.09233948960900307, "time": 0.5258806467056274, "epoch": 9, "memory": 6318, "step": 33396} +{"lr": 1.4855103784081099e-05, "data_time": 0.0008928298950195313, "loss": 0.10468098372220994, "time": 0.48557496070861816, "epoch": 9, "memory": 6318, "step": 33496} +{"lr": 1.4694141710823508e-05, "data_time": 0.001069927215576172, "loss": 0.09393770471215249, "time": 0.4717130422592163, "epoch": 9, "memory": 6318, "step": 33596} +{"lr": 1.4535746099250614e-05, "data_time": 0.0011947154998779297, "loss": 0.09650558307766914, "time": 0.4393578052520752, "epoch": 9, "memory": 6318, "step": 33696} +{"lr": 1.4379927035148576e-05, "data_time": 0.0013447284698486328, "loss": 0.09026336446404457, "time": 0.5092681884765625, "epoch": 9, "memory": 6318, "step": 33796} +{"lr": 1.42266944402428e-05, "data_time": 0.0010889530181884765, "loss": 0.0928823359310627, "time": 0.4522377967834473, "epoch": 9, "memory": 6318, "step": 33896} +{"lr": 1.4076058071566133e-05, "data_time": 0.0014900922775268554, "loss": 0.08821561448276043, "time": 0.5187438488006592, "epoch": 9, "memory": 6318, "step": 33996} +{"lr": 1.3928027520837635e-05, "data_time": 0.001142573356628418, "loss": 0.09485570564866067, "time": 0.4955988883972168, "epoch": 9, "memory": 6318, "step": 34096} +{"lr": 1.3782612213851759e-05, "data_time": 0.0010311126708984375, "loss": 0.07379315569996833, "time": 0.4372359275817871, "epoch": 9, "memory": 6318, "step": 34196} +{"lr": 1.3639821409878263e-05, "data_time": 0.056370210647583005, "loss": 0.10082559026777745, "time": 0.4919793128967285, "epoch": 9, "memory": 6318, "step": 34296} +{"lr": 1.3499664201072546e-05, "data_time": 0.20958466529846193, "loss": 0.09546709209680557, "time": 0.4997179269790649, "epoch": 9, "memory": 6318, "step": 34396} +{"lr": 1.336214951189674e-05, "data_time": 0.05220351219177246, "loss": 0.10569841042160988, "time": 0.49487416744232177, "epoch": 9, "memory": 6318, "step": 34496} +{"lr": 1.3227286098551456e-05, "data_time": 0.12370331287384033, "loss": 0.08879945501685142, "time": 0.5162428140640258, "epoch": 9, "memory": 6318, "step": 34596} +{"lr": 1.3095082548418224e-05, "data_time": 0.13003745079040527, "loss": 0.09212391600012779, "time": 0.48962717056274413, "epoch": 9, "memory": 6318, "step": 34696} +{"lr": 1.2965547279512709e-05, "data_time": 0.15012824535369873, "loss": 0.09019293710589409, "time": 0.49649505615234374, "epoch": 9, "memory": 6318, "step": 34796} +{"lr": 1.2838688539948659e-05, "data_time": 0.1678551197052002, "loss": 0.10418915376067162, "time": 0.5437381029129028, "epoch": 9, "memory": 6318, "step": 34896} +{"lr": 1.2714514407412763e-05, "data_time": 0.328688907623291, "loss": 0.09173640720546246, "time": 0.4752793788909912, "epoch": 9, "memory": 6318, "step": 34996} +{"lr": 1.2593032788650256e-05, "data_time": 0.36444292068481443, "loss": 0.09632524996995925, "time": 0.5047950267791748, "epoch": 9, "memory": 6318, "step": 35096} +{"lr": 1.2474251418961484e-05, "data_time": 0.433282732963562, "loss": 0.08514822125434876, "time": 0.5742971658706665, "epoch": 9, "memory": 6318, "step": 35196} +{"lr": 1.2358177861709387e-05, "data_time": 0.2728005647659302, "loss": 0.09738615900278091, "time": 0.5269148349761963, "epoch": 9, "memory": 6318, "step": 35296} +{"lr": 1.2244819507837855e-05, "data_time": 0.32030434608459474, "loss": 0.0945310577750206, "time": 0.48576650619506834, "epoch": 9, "memory": 6318, "step": 35396} +{"accuracy/top1": 96.61807250976562, "data_time": 0.3732914760194976, "time": 0.4594194683535346, "step": 9} +{"lr": 1.2093939694118031e-05, "data_time": 0.30083954334259033, "loss": 0.1128064289689064, "time": 0.44062635898590086, "epoch": 10, "memory": 6318, "step": 35533} +{"lr": 1.1987044880988498e-05, "data_time": 0.3775103330612183, "loss": 0.08815532177686691, "time": 0.516980791091919, "epoch": 10, "memory": 6318, "step": 35633} +{"lr": 1.1882888903004625e-05, "data_time": 0.33387250900268556, "loss": 0.09743470251560211, "time": 0.4734462022781372, "epoch": 10, "memory": 6318, "step": 35733} +{"lr": 1.178147839226253e-05, "data_time": 0.34879398345947266, "loss": 0.08691700845956803, "time": 0.4880156278610229, "epoch": 10, "memory": 6318, "step": 35833} +{"lr": 1.1682819806041682e-05, "data_time": 0.3381638765335083, "loss": 0.08479001820087433, "time": 0.4828916311264038, "epoch": 10, "memory": 6318, "step": 35933} +{"lr": 1.1586919426393653e-05, "data_time": 0.31300647258758546, "loss": 0.09261269718408585, "time": 0.45303988456726074, "epoch": 10, "memory": 6318, "step": 36033} +{"lr": 1.1493783359742196e-05, "data_time": 0.30055234432220457, "loss": 0.08887546919286252, "time": 0.4570359706878662, "epoch": 10, "memory": 6318, "step": 36133} +{"lr": 1.1403417536494367e-05, "data_time": 0.3454045295715332, "loss": 0.08568267896771431, "time": 0.48510792255401614, "epoch": 10, "memory": 6318, "step": 36233} +{"lr": 1.1315827710662908e-05, "data_time": 0.3492480993270874, "loss": 0.08701312318444251, "time": 0.48921523094177244, "epoch": 10, "memory": 6318, "step": 36333} +{"lr": 1.1231019459499938e-05, "data_time": 0.3514674425125122, "loss": 0.09252516627311706, "time": 0.49141180515289307, "epoch": 10, "memory": 6318, "step": 36433} +{"lr": 1.1148998183141688e-05, "data_time": 0.5685892581939698, "loss": 0.09964828714728355, "time": 0.708743953704834, "epoch": 10, "memory": 6318, "step": 36533} +{"lr": 1.1069769104264779e-05, "data_time": 0.3947058439254761, "loss": 0.09140113741159439, "time": 0.5361949443817139, "epoch": 10, "memory": 6318, "step": 36633} +{"lr": 1.0993337267753594e-05, "data_time": 0.298458456993103, "loss": 0.07913832552731037, "time": 0.439842963218689, "epoch": 10, "memory": 6318, "step": 36733} +{"lr": 1.0919707540379088e-05, "data_time": 0.3678726673126221, "loss": 0.09146838709712028, "time": 0.5092720270156861, "epoch": 10, "memory": 6318, "step": 36833} +{"lr": 1.0848884610488858e-05, "data_time": 0.7514483213424683, "loss": 0.0948871422559023, "time": 1.0796780109405517, "epoch": 10, "memory": 6318, "step": 36933} +{"lr": 1.0780872987708645e-05, "data_time": 0.3007082462310791, "loss": 0.08585912808775902, "time": 0.45542125701904296, "epoch": 10, "memory": 6318, "step": 37033} +{"lr": 1.0715677002655166e-05, "data_time": 0.14177203178405762, "loss": 0.10550803169608117, "time": 0.4755758047103882, "epoch": 10, "memory": 6318, "step": 37133} +{"lr": 1.0653300806660372e-05, "data_time": 0.11019306182861328, "loss": 0.08652376122772694, "time": 0.4652707099914551, "epoch": 10, "memory": 6318, "step": 37233} +{"lr": 1.0593748371507151e-05, "data_time": 0.2064640998840332, "loss": 0.0875510510057211, "time": 0.4914530038833618, "epoch": 10, "memory": 6318, "step": 37333} +{"lr": 1.0537023489176325e-05, "data_time": 0.23270184993743898, "loss": 0.08551195412874221, "time": 0.4768069744110107, "epoch": 10, "memory": 6318, "step": 37433} +{"lr": 1.0483129771605328e-05, "data_time": 0.3083606958389282, "loss": 0.08292020373046398, "time": 0.5312201499938964, "epoch": 10, "memory": 6318, "step": 37533} +{"lr": 1.0432070650458095e-05, "data_time": 0.36708760261535645, "loss": 0.08170890659093857, "time": 0.5082449913024902, "epoch": 10, "memory": 6318, "step": 37633} +{"lr": 1.0383849376906634e-05, "data_time": 0.31961779594421386, "loss": 0.08897578455507756, "time": 0.46034243106842043, "epoch": 10, "memory": 6318, "step": 37733} +{"lr": 1.0338469021423977e-05, "data_time": 0.3505417823791504, "loss": 0.08403019309043884, "time": 0.4918610334396362, "epoch": 10, "memory": 6318, "step": 37833} +{"lr": 1.029593247358868e-05, "data_time": 0.34041595458984375, "loss": 0.09566556289792061, "time": 0.480924129486084, "epoch": 10, "memory": 6318, "step": 37933} +{"lr": 1.0256242441900827e-05, "data_time": 0.3240325927734375, "loss": 0.09055039659142494, "time": 0.4656890869140625, "epoch": 10, "memory": 6318, "step": 38033} +{"lr": 1.0219401453609533e-05, "data_time": 0.34224896430969237, "loss": 0.09382965825498105, "time": 0.48375849723815917, "epoch": 10, "memory": 6318, "step": 38133} +{"lr": 1.018541185455209e-05, "data_time": 0.30359065532684326, "loss": 0.09477643221616745, "time": 0.4432615041732788, "epoch": 10, "memory": 6318, "step": 38233} +{"lr": 1.015427580900455e-05, "data_time": 0.3605957508087158, "loss": 0.08722190409898758, "time": 0.5014996290206909, "epoch": 10, "memory": 6318, "step": 38333} +{"lr": 1.0125995299543914e-05, "data_time": 0.3488884925842285, "loss": 0.08811897188425064, "time": 0.4892071962356567, "epoch": 10, "memory": 6318, "step": 38433} +{"lr": 1.0100572126921902e-05, "data_time": 0.34983506202697756, "loss": 0.1071148082613945, "time": 0.4902963638305664, "epoch": 10, "memory": 6318, "step": 38533} +{"lr": 1.007800790995031e-05, "data_time": 0.31435959339141845, "loss": 0.10802561715245247, "time": 0.4542102575302124, "epoch": 10, "memory": 6318, "step": 38633} +{"lr": 1.0058304085397905e-05, "data_time": 0.33975374698638916, "loss": 0.10101355090737343, "time": 0.481682276725769, "epoch": 10, "memory": 6318, "step": 38733} +{"lr": 1.0041461907898941e-05, "data_time": 0.37254621982574465, "loss": 0.10207248032093048, "time": 0.513184380531311, "epoch": 10, "memory": 6318, "step": 38833} +{"lr": 1.0027482449873297e-05, "data_time": 0.33594613075256347, "loss": 0.09972277358174324, "time": 0.47675719261169436, "epoch": 10, "memory": 6318, "step": 38933} +{"lr": 1.0016366601458162e-05, "data_time": 0.31034648418426514, "loss": 0.10015354640781879, "time": 0.4509953260421753, "epoch": 10, "memory": 6318, "step": 39033} +{"lr": 1.0008115070451358e-05, "data_time": 0.35239882469177247, "loss": 0.0880768671631813, "time": 0.49243271350860596, "epoch": 10, "memory": 6318, "step": 39133} +{"lr": 1.0002728382266284e-05, "data_time": 0.37269542217254636, "loss": 0.08653453812003135, "time": 0.5141926765441894, "epoch": 10, "memory": 6318, "step": 39233} +{"lr": 1.0000206879898465e-05, "data_time": 0.38600831031799315, "loss": 0.09437817335128784, "time": 0.5272750616073608, "epoch": 10, "memory": 6318, "step": 39333} +{"accuracy/top1": 96.48585510253906, "data_time": 0.3758496046066284, "time": 0.46335516304805363, "step": 10} diff --git a/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1/20230606_051539/vis_data/config.py b/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1/20230606_051539/vis_data/config.py new file mode 100644 index 0000000000000000000000000000000000000000..9fa82a87719e4caa4a661d2572ccc1b988bdb43d --- /dev/null +++ b/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1/20230606_051539/vis_data/config.py @@ -0,0 +1,247 @@ +optim_wrapper = dict( + optimizer=dict( + type='SGD', + lr=0.0001, + momentum=0.9, + weight_decay=0.0001, + _scope_='mmpretrain'), + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=False, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=512) +model = dict( + type='ImageClassifier', + backbone=dict( + frozen_stages=2, + type='ResNet', + depth=50, + num_stages=4, + out_indices=(3, ), + style='pytorch', + init_cfg=dict( + type='Pretrained', + checkpoint= + 'https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth', + prefix='backbone')), + neck=dict(type='GlobalAveragePooling'), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=2048, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + topk=1)) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/IF80w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/if-dpmsolver++-50-20w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/cc1m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/if-dpmsolver++-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = dict(type='Accuracy', topk=1) +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/if-dpmsolver++-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = dict(type='Accuracy', topk=1) +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1' diff --git a/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1/20230606_051539/vis_data/events.out.tfevents.1685999753.SH-IDC1-10-140-24-124.246723.0 b/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1/20230606_051539/vis_data/events.out.tfevents.1685999753.SH-IDC1-10-140-24-124.246723.0 new file mode 100644 index 0000000000000000000000000000000000000000..a6d2a43e6b12f17a0348dfca4d66e47cc833260a --- /dev/null +++ b/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1/20230606_051539/vis_data/events.out.tfevents.1685999753.SH-IDC1-10-140-24-124.246723.0 @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:babefd38212d2c64afff58207d7feb15db8a4280d356c66a12b30f61c1cdce1c +size 3099216 diff --git a/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1/20230606_051539/vis_data/scalars.json b/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1/20230606_051539/vis_data/scalars.json new file mode 100644 index 0000000000000000000000000000000000000000..74dcd1eb050eed777b0c77735158f43f6f45d36d --- /dev/null +++ b/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1/20230606_051539/vis_data/scalars.json @@ -0,0 +1,400 @@ +{"lr": 9.999859583040624e-05, "data_time": 0.35135469436645506, "loss": 0.6373136699199676, "time": 0.5047009229660034, "epoch": 1, "memory": 9436, "step": 100} +{"lr": 9.999432653394512e-05, "data_time": 0.40174314975738523, "loss": 0.573208886384964, "time": 0.5414288520812989, "epoch": 1, "memory": 6319, "step": 200} +{"lr": 9.998719223919393e-05, "data_time": 0.48686139583587645, "loss": 0.5214441359043122, "time": 0.6410719394683838, "epoch": 1, "memory": 6319, "step": 300} +{"lr": 9.997719340042649e-05, "data_time": 0.5417542457580566, "loss": 0.44845099151134493, "time": 0.6814687013626098, "epoch": 1, "memory": 6319, "step": 400} +{"lr": 9.996433065431542e-05, "data_time": 0.4060805082321167, "loss": 0.410491606593132, "time": 0.5461702108383178, "epoch": 1, "memory": 6319, "step": 500} +{"lr": 9.994860481989157e-05, "data_time": 0.4788053512573242, "loss": 0.361409854888916, "time": 0.6187043905258178, "epoch": 1, "memory": 6319, "step": 600} +{"lr": 9.993001689849214e-05, "data_time": 0.4635407209396362, "loss": 0.32016042768955233, "time": 0.6049500703811646, "epoch": 1, "memory": 6319, "step": 700} +{"lr": 9.990856807369658e-05, "data_time": 0.5030936002731323, "loss": 0.3190604686737061, "time": 0.6450572490692139, "epoch": 1, "memory": 6319, "step": 800} +{"lr": 9.98842597112514e-05, "data_time": 0.40195062160491946, "loss": 0.3022842317819595, "time": 0.5432976722717285, "epoch": 1, "memory": 6319, "step": 900} +{"lr": 9.985709335898336e-05, "data_time": 0.4612396717071533, "loss": 0.2837853774428368, "time": 0.6018701076507569, "epoch": 1, "memory": 6319, "step": 1000} +{"lr": 9.982707074670047e-05, "data_time": 0.4168861389160156, "loss": 0.2603603795170784, "time": 0.5616236209869385, "epoch": 1, "memory": 6319, "step": 1100} +{"lr": 9.979419378608245e-05, "data_time": 0.44224417209625244, "loss": 0.26175386607646944, "time": 0.582902193069458, "epoch": 1, "memory": 6319, "step": 1200} +{"lr": 9.975846457055832e-05, "data_time": 0.4418195247650146, "loss": 0.25294673144817353, "time": 0.5848827123641968, "epoch": 1, "memory": 6319, "step": 1300} +{"lr": 9.97198853751737e-05, "data_time": 0.43476912975311277, "loss": 0.23805642426013945, "time": 0.5752826452255249, "epoch": 1, "memory": 6319, "step": 1400} +{"lr": 9.967845865644554e-05, "data_time": 0.3910046100616455, "loss": 0.23767492026090623, "time": 0.5428294658660888, "epoch": 1, "memory": 6319, "step": 1500} +{"lr": 9.963418705220597e-05, "data_time": 0.42578513622283937, "loss": 0.23924392610788345, "time": 0.5666382074356079, "epoch": 1, "memory": 6319, "step": 1600} +{"lr": 9.958707338143413e-05, "data_time": 0.428590202331543, "loss": 0.23500001281499863, "time": 0.5770107507705688, "epoch": 1, "memory": 6319, "step": 1700} +{"lr": 9.953712064407689e-05, "data_time": 0.41157917976379393, "loss": 0.23590199053287506, "time": 0.5512240409851075, "epoch": 1, "memory": 6319, "step": 1800} +{"lr": 9.948433202085752e-05, "data_time": 0.47192094326019285, "loss": 0.22650814056396484, "time": 0.6119584083557129, "epoch": 1, "memory": 6319, "step": 1900} +{"lr": 9.942871087307363e-05, "data_time": 0.46183364391326903, "loss": 0.21390928328037262, "time": 0.6011797189712524, "epoch": 1, "memory": 6319, "step": 2000} +{"lr": 9.937026074238262e-05, "data_time": 0.37435388565063477, "loss": 0.22222377955913544, "time": 0.5163755178451538, "epoch": 1, "memory": 6319, "step": 2100} +{"lr": 9.930898535057657e-05, "data_time": 0.41797561645507814, "loss": 0.2177472949028015, "time": 0.5578161716461182, "epoch": 1, "memory": 6319, "step": 2200} +{"lr": 9.924488859934492e-05, "data_time": 0.37576615810394287, "loss": 0.2220224842429161, "time": 0.5166475057601929, "epoch": 1, "memory": 6319, "step": 2300} +{"lr": 9.917797457002633e-05, "data_time": 0.41900794506072997, "loss": 0.20719822645187377, "time": 0.5588636875152588, "epoch": 1, "memory": 6319, "step": 2400} +{"lr": 9.910824752334866e-05, "data_time": 0.4473739147186279, "loss": 0.21015606373548507, "time": 0.5872008323669433, "epoch": 1, "memory": 6319, "step": 2500} +{"lr": 9.903571189915766e-05, "data_time": 0.42905564308166505, "loss": 0.2182312563061714, "time": 0.5721040725708008, "epoch": 1, "memory": 6319, "step": 2600} +{"lr": 9.896037231613424e-05, "data_time": 0.4330267429351807, "loss": 0.21145364344120027, "time": 0.5730567455291748, "epoch": 1, "memory": 6319, "step": 2700} +{"lr": 9.888223357150049e-05, "data_time": 0.40704665184020994, "loss": 0.1990158200263977, "time": 0.5482313394546509, "epoch": 1, "memory": 6319, "step": 2800} +{"lr": 9.880130064071408e-05, "data_time": 0.41959223747253416, "loss": 0.19440122246742247, "time": 0.5595078945159913, "epoch": 1, "memory": 6319, "step": 2900} +{"lr": 9.87175786771517e-05, "data_time": 0.4379056453704834, "loss": 0.19239158928394318, "time": 0.5797731399536132, "epoch": 1, "memory": 6319, "step": 3000} +{"lr": 9.863107301178059e-05, "data_time": 0.4306967735290527, "loss": 0.1904392957687378, "time": 0.5712720870971679, "epoch": 1, "memory": 6319, "step": 3100} +{"lr": 9.854178915281925e-05, "data_time": 0.43600437641143797, "loss": 0.19300480633974076, "time": 0.5763952970504761, "epoch": 1, "memory": 6319, "step": 3200} +{"lr": 9.84497327853868e-05, "data_time": 0.45678741931915284, "loss": 0.18359561264514923, "time": 0.598597502708435, "epoch": 1, "memory": 6319, "step": 3300} +{"lr": 9.83549097711407e-05, "data_time": 0.41577074527740476, "loss": 0.18914546370506286, "time": 0.5554603338241577, "epoch": 1, "memory": 6319, "step": 3400} +{"lr": 9.825732614790392e-05, "data_time": 0.42415382862091067, "loss": 0.20574819892644883, "time": 0.5642972946166992, "epoch": 1, "memory": 6319, "step": 3500} +{"lr": 9.815698812928018e-05, "data_time": 0.42788641452789306, "loss": 0.1987052395939827, "time": 0.5689198732376098, "epoch": 1, "memory": 6319, "step": 3600} +{"lr": 9.805390210425831e-05, "data_time": 0.3883518695831299, "loss": 0.19001712948083876, "time": 0.5294172286987304, "epoch": 1, "memory": 6319, "step": 3700} +{"lr": 9.794807463680567e-05, "data_time": 0.36976687908172606, "loss": 0.18455516248941423, "time": 0.5100427389144897, "epoch": 1, "memory": 6319, "step": 3800} +{"lr": 9.783951246544974e-05, "data_time": 0.38853485584259034, "loss": 0.1877005636692047, "time": 0.5299545288085937, "epoch": 1, "memory": 6319, "step": 3900} +{"accuracy/top1": 65.28304290771484, "data_time": 0.34556789983782854, "time": 0.43776849696510717, "step": 1} +{"lr": 9.768635527139677e-05, "data_time": 0.38695855140686036, "loss": 0.17761119082570076, "time": 0.5509963750839233, "epoch": 2, "memory": 8776, "step": 4037} +{"lr": 9.75713397738286e-05, "data_time": 0.1603771448135376, "loss": 0.16979795843362808, "time": 0.47443265914916993, "epoch": 2, "memory": 6318, "step": 4137} +{"lr": 9.745361356082025e-05, "data_time": 0.0548288106918335, "loss": 0.18406075686216355, "time": 0.5031360864639283, "epoch": 2, "memory": 6318, "step": 4237} +{"lr": 9.733318412854784e-05, "data_time": 0.08445332050323487, "loss": 0.16945173293352128, "time": 0.6180483579635621, "epoch": 2, "memory": 6318, "step": 4337} +{"lr": 9.721005914531412e-05, "data_time": 0.09240703582763672, "loss": 0.16133614033460617, "time": 0.462139368057251, "epoch": 2, "memory": 6318, "step": 4437} +{"lr": 9.708424645106017e-05, "data_time": 0.1672619342803955, "loss": 0.1570103347301483, "time": 0.5158196687698364, "epoch": 2, "memory": 6318, "step": 4537} +{"lr": 9.695575405686603e-05, "data_time": 0.3291741609573364, "loss": 0.17726304531097412, "time": 0.4993450403213501, "epoch": 2, "memory": 6318, "step": 4637} +{"lr": 9.682459014444089e-05, "data_time": 0.36315526962280276, "loss": 0.17917748242616655, "time": 0.5105035066604614, "epoch": 2, "memory": 6318, "step": 4737} +{"lr": 9.669076306560176e-05, "data_time": 0.32734673023223876, "loss": 0.17449757009744643, "time": 0.5395374298095703, "epoch": 2, "memory": 6318, "step": 4837} +{"lr": 9.655428134174183e-05, "data_time": 0.2691281795501709, "loss": 0.18600485920906068, "time": 0.5208764553070069, "epoch": 2, "memory": 6318, "step": 4937} +{"lr": 9.641515366328807e-05, "data_time": 0.3626711845397949, "loss": 0.1755761295557022, "time": 0.5503090143203735, "epoch": 2, "memory": 6318, "step": 5037} +{"lr": 9.627338888914757e-05, "data_time": 0.2300825834274292, "loss": 0.16736908555030822, "time": 0.5431391477584839, "epoch": 2, "memory": 6318, "step": 5137} +{"lr": 9.612899604614378e-05, "data_time": 0.19789693355560303, "loss": 0.16888393759727477, "time": 0.5604245901107788, "epoch": 2, "memory": 6318, "step": 5237} +{"lr": 9.598198432844131e-05, "data_time": 0.288263726234436, "loss": 0.16049372628331185, "time": 0.5551909685134888, "epoch": 2, "memory": 6318, "step": 5337} +{"lr": 9.583236309696081e-05, "data_time": 0.09539401531219482, "loss": 0.17333296090364456, "time": 0.4834871292114258, "epoch": 2, "memory": 6318, "step": 5437} +{"lr": 9.568014187878279e-05, "data_time": 0.162140417098999, "loss": 0.1645927131175995, "time": 0.5141091108322143, "epoch": 2, "memory": 6318, "step": 5537} +{"lr": 9.552533036654103e-05, "data_time": 0.23149263858795166, "loss": 0.16614100635051726, "time": 0.5089569568634034, "epoch": 2, "memory": 6318, "step": 5637} +{"lr": 9.536793841780541e-05, "data_time": 0.3949670076370239, "loss": 0.15926927030086518, "time": 0.537546968460083, "epoch": 2, "memory": 6318, "step": 5737} +{"lr": 9.52079760544543e-05, "data_time": 0.40032660961151123, "loss": 0.16198524683713914, "time": 0.5412147283554077, "epoch": 2, "memory": 6318, "step": 5837} +{"lr": 9.504545346203619e-05, "data_time": 0.35010015964508057, "loss": 0.1635586552321911, "time": 0.4903889179229736, "epoch": 2, "memory": 6318, "step": 5937} +{"lr": 9.488038098912122e-05, "data_time": 0.373197603225708, "loss": 0.15642861276865005, "time": 0.5136023759841919, "epoch": 2, "memory": 6318, "step": 6037} +{"lr": 9.471276914664257e-05, "data_time": 0.39609856605529786, "loss": 0.16319777071475983, "time": 0.536125922203064, "epoch": 2, "memory": 6318, "step": 6137} +{"lr": 9.454262860722667e-05, "data_time": 0.39641575813293456, "loss": 0.1698547422885895, "time": 0.5367683172225952, "epoch": 2, "memory": 6318, "step": 6237} +{"lr": 9.436997020451398e-05, "data_time": 0.3755208969116211, "loss": 0.15329988598823546, "time": 0.5168079376220703, "epoch": 2, "memory": 6318, "step": 6337} +{"lr": 9.419480493246903e-05, "data_time": 0.37643425464630126, "loss": 0.13655854389071465, "time": 0.5169485330581665, "epoch": 2, "memory": 6318, "step": 6437} +{"lr": 9.401714394468036e-05, "data_time": 0.5956509828567504, "loss": 0.14493875205516815, "time": 0.7383939027786255, "epoch": 2, "memory": 6318, "step": 6537} +{"lr": 9.383699855365022e-05, "data_time": 0.41739416122436523, "loss": 0.15211182758212088, "time": 0.55793297290802, "epoch": 2, "memory": 6318, "step": 6637} +{"lr": 9.365438023007469e-05, "data_time": 0.33812565803527833, "loss": 0.16641210317611693, "time": 0.4787991762161255, "epoch": 2, "memory": 6318, "step": 6737} +{"lr": 9.346930060211267e-05, "data_time": 0.41308434009552003, "loss": 0.15445083230733872, "time": 0.5535926342010498, "epoch": 2, "memory": 6318, "step": 6837} +{"lr": 9.328177145464599e-05, "data_time": 0.394673490524292, "loss": 0.15339895337820053, "time": 0.5363324165344239, "epoch": 2, "memory": 6318, "step": 6937} +{"lr": 9.309180472852871e-05, "data_time": 0.39875857830047606, "loss": 0.15161959379911422, "time": 0.5497647285461426, "epoch": 2, "memory": 6318, "step": 7037} +{"lr": 9.289941251982683e-05, "data_time": 0.4293885469436646, "loss": 0.15708278119564056, "time": 0.5705724716186523, "epoch": 2, "memory": 6318, "step": 7137} +{"lr": 9.270460707904827e-05, "data_time": 0.4121288537979126, "loss": 0.13404012173414231, "time": 0.5533035278320313, "epoch": 2, "memory": 6318, "step": 7237} +{"lr": 9.250740081036243e-05, "data_time": 0.418600869178772, "loss": 0.14458792954683303, "time": 0.5596281528472901, "epoch": 2, "memory": 6318, "step": 7337} +{"lr": 9.230780627081088e-05, "data_time": 0.39788527488708497, "loss": 0.1460558496415615, "time": 0.540451192855835, "epoch": 2, "memory": 6318, "step": 7437} +{"lr": 9.210583616950722e-05, "data_time": 0.3776459932327271, "loss": 0.14620409682393073, "time": 0.5185774564743042, "epoch": 2, "memory": 6318, "step": 7537} +{"lr": 9.190150336682842e-05, "data_time": 0.3727187395095825, "loss": 0.1502346560359001, "time": 0.5137011289596558, "epoch": 2, "memory": 6318, "step": 7637} +{"lr": 9.16948208735955e-05, "data_time": 0.39030919075012205, "loss": 0.15273726880550384, "time": 0.5300938367843628, "epoch": 2, "memory": 6318, "step": 7737} +{"lr": 9.148580185024521e-05, "data_time": 0.3838090419769287, "loss": 0.1422838144004345, "time": 0.5244406938552857, "epoch": 2, "memory": 6318, "step": 7837} +{"accuracy/top1": 86.79934692382812, "data_time": 0.24867882399723448, "time": 0.3373623223140322, "step": 2} +{"lr": 9.119567684509738e-05, "data_time": 0.3863436222076416, "loss": 0.14530004188418388, "time": 0.5260138034820556, "epoch": 3, "memory": 6318, "step": 7974} +{"lr": 9.098117366468421e-05, "data_time": 0.3732325553894043, "loss": 0.1224223829805851, "time": 0.5137952089309692, "epoch": 3, "memory": 6318, "step": 8074} +{"lr": 9.076437939539471e-05, "data_time": 0.3858577013015747, "loss": 0.14201032146811485, "time": 0.5259620904922485, "epoch": 3, "memory": 6318, "step": 8174} +{"lr": 9.05453078415294e-05, "data_time": 0.3986901998519897, "loss": 0.12961676418781282, "time": 0.5374616861343384, "epoch": 3, "memory": 6318, "step": 8274} +{"lr": 9.032397295239427e-05, "data_time": 0.3945038080215454, "loss": 0.14835804626345633, "time": 0.534127163887024, "epoch": 3, "memory": 6318, "step": 8374} +{"lr": 9.010038882141241e-05, "data_time": 0.3276252031326294, "loss": 0.13363704234361648, "time": 0.4678341388702393, "epoch": 3, "memory": 6318, "step": 8474} +{"lr": 8.987456968522655e-05, "data_time": 0.3987130880355835, "loss": 0.1381424091756344, "time": 0.539873480796814, "epoch": 3, "memory": 6318, "step": 8574} +{"lr": 8.964652992279262e-05, "data_time": 0.3964304685592651, "loss": 0.15181969180703164, "time": 0.5373668432235718, "epoch": 3, "memory": 6318, "step": 8674} +{"lr": 8.941628405446428e-05, "data_time": 0.368574333190918, "loss": 0.15693323239684104, "time": 0.5083968877792359, "epoch": 3, "memory": 6318, "step": 8774} +{"lr": 8.918384674106797e-05, "data_time": 0.32884399890899657, "loss": 0.1135924369096756, "time": 0.4690085411071777, "epoch": 3, "memory": 6318, "step": 8874} +{"lr": 8.894923278296992e-05, "data_time": 0.38549854755401614, "loss": 0.14542728960514067, "time": 0.5258296251296997, "epoch": 3, "memory": 6318, "step": 8974} +{"lr": 8.871245711913324e-05, "data_time": 0.3975813388824463, "loss": 0.15234951972961425, "time": 0.5375964164733886, "epoch": 3, "memory": 6318, "step": 9074} +{"lr": 8.847353482616703e-05, "data_time": 0.38844666481018064, "loss": 0.12656183168292046, "time": 0.5289930582046509, "epoch": 3, "memory": 6318, "step": 9174} +{"lr": 8.823248111736638e-05, "data_time": 0.3702134132385254, "loss": 0.12871245965361594, "time": 0.510457181930542, "epoch": 3, "memory": 6318, "step": 9274} +{"lr": 8.798931134174327e-05, "data_time": 0.3971710681915283, "loss": 0.1386072114109993, "time": 0.5380873918533325, "epoch": 3, "memory": 6318, "step": 9374} +{"lr": 8.774404098304961e-05, "data_time": 0.40869979858398436, "loss": 0.13455229476094246, "time": 0.5485035419464112, "epoch": 3, "memory": 6318, "step": 9474} +{"lr": 8.749668565879136e-05, "data_time": 0.40019822120666504, "loss": 0.14316971302032472, "time": 0.5402791738510132, "epoch": 3, "memory": 6318, "step": 9574} +{"lr": 8.724726111923386e-05, "data_time": 0.43427319526672364, "loss": 0.14011319503188133, "time": 0.5742615938186646, "epoch": 3, "memory": 6318, "step": 9674} +{"lr": 8.699578324639898e-05, "data_time": 0.3744964122772217, "loss": 0.1291911445558071, "time": 0.514688229560852, "epoch": 3, "memory": 6318, "step": 9774} +{"lr": 8.67422680530539e-05, "data_time": 0.36439154148101804, "loss": 0.13297815918922423, "time": 0.5083678007125855, "epoch": 3, "memory": 6318, "step": 9874} +{"lr": 8.648673168169147e-05, "data_time": 0.39237961769104, "loss": 0.13199843987822532, "time": 0.5391510725021362, "epoch": 3, "memory": 6318, "step": 9974} +{"lr": 8.622919040350244e-05, "data_time": 0.5732649326324463, "loss": 0.1406612977385521, "time": 0.7130593299865723, "epoch": 3, "memory": 6318, "step": 10074} +{"lr": 8.596966061733922e-05, "data_time": 0.3915122032165527, "loss": 0.1336354061961174, "time": 0.5335070371627808, "epoch": 3, "memory": 6318, "step": 10174} +{"lr": 8.570815884867185e-05, "data_time": 0.3767182111740112, "loss": 0.1251736029982567, "time": 0.5171158790588379, "epoch": 3, "memory": 6318, "step": 10274} +{"lr": 8.544470174853563e-05, "data_time": 0.3925620079040527, "loss": 0.11885619238018989, "time": 0.5400431394577027, "epoch": 3, "memory": 6318, "step": 10374} +{"lr": 8.517930609247077e-05, "data_time": 0.39089391231536863, "loss": 0.1369865544140339, "time": 0.5303629159927368, "epoch": 3, "memory": 6318, "step": 10474} +{"lr": 8.491198877945477e-05, "data_time": 0.3661817073822021, "loss": 0.12701885998249055, "time": 0.5065898656845093, "epoch": 3, "memory": 6318, "step": 10574} +{"lr": 8.464276683082565e-05, "data_time": 0.349953818321228, "loss": 0.12582506984472275, "time": 0.49130213260650635, "epoch": 3, "memory": 6318, "step": 10674} +{"lr": 8.437165738919852e-05, "data_time": 0.38394405841827395, "loss": 0.1230903796851635, "time": 0.524141001701355, "epoch": 3, "memory": 6318, "step": 10774} +{"lr": 8.409867771737398e-05, "data_time": 0.3929825067520142, "loss": 0.12072777599096299, "time": 0.5342830181121826, "epoch": 3, "memory": 6318, "step": 10874} +{"lr": 8.382384519723895e-05, "data_time": 0.3879563331604004, "loss": 0.12704971060156822, "time": 0.5280010938644409, "epoch": 3, "memory": 6318, "step": 10974} +{"lr": 8.35471773286596e-05, "data_time": 0.3738083839416504, "loss": 0.12546228989958763, "time": 0.5150954246520996, "epoch": 3, "memory": 6318, "step": 11074} +{"lr": 8.326869172836765e-05, "data_time": 0.35648431777954104, "loss": 0.11281569972634316, "time": 0.4971273422241211, "epoch": 3, "memory": 6318, "step": 11174} +{"lr": 8.298840612883789e-05, "data_time": 0.4172009229660034, "loss": 0.12077075466513634, "time": 0.5572011232376098, "epoch": 3, "memory": 6318, "step": 11274} +{"lr": 8.270633837715962e-05, "data_time": 0.40588066577911375, "loss": 0.11257560476660729, "time": 0.54643874168396, "epoch": 3, "memory": 6318, "step": 11374} +{"lr": 8.242250643390022e-05, "data_time": 0.43177099227905275, "loss": 0.1323128454387188, "time": 0.5713624238967896, "epoch": 3, "memory": 6318, "step": 11474} +{"lr": 8.2136928371961e-05, "data_time": 0.4174847364425659, "loss": 0.12084009796380997, "time": 0.5590145587921143, "epoch": 3, "memory": 6318, "step": 11574} +{"lr": 8.184962237542712e-05, "data_time": 0.3454891681671143, "loss": 0.1324061408638954, "time": 0.5164076566696167, "epoch": 3, "memory": 6318, "step": 11674} +{"lr": 8.156060673840912e-05, "data_time": 0.36376371383666994, "loss": 0.1252976857125759, "time": 0.516312837600708, "epoch": 3, "memory": 6318, "step": 11774} +{"accuracy/top1": 91.90355682373047, "data_time": 0.30805200543896905, "time": 0.39490813222424737, "step": 3} +{"lr": 8.116191337909002e-05, "data_time": 0.40924673080444335, "loss": 0.12510494589805604, "time": 0.549347186088562, "epoch": 4, "memory": 6318, "step": 11911} +{"lr": 8.086891958166228e-05, "data_time": 0.392189621925354, "loss": 0.1404468335211277, "time": 0.5318895101547241, "epoch": 4, "memory": 6318, "step": 12011} +{"lr": 8.057427858966412e-05, "data_time": 0.38721375465393065, "loss": 0.13838243559002877, "time": 0.5272845983505249, "epoch": 4, "memory": 6318, "step": 12111} +{"lr": 8.027800916425949e-05, "data_time": 0.3695343494415283, "loss": 0.11593601852655411, "time": 0.5215901374816895, "epoch": 4, "memory": 6318, "step": 12211} +{"lr": 7.998013017030234e-05, "data_time": 0.4206414699554443, "loss": 0.1300233282148838, "time": 0.5605233192443848, "epoch": 4, "memory": 6318, "step": 12311} +{"lr": 7.968066057513528e-05, "data_time": 0.3648210048675537, "loss": 0.14271003305912017, "time": 0.5050093650817871, "epoch": 4, "memory": 6318, "step": 12411} +{"lr": 7.937961944738199e-05, "data_time": 0.4261133670806885, "loss": 0.12264132052659989, "time": 0.566995644569397, "epoch": 4, "memory": 6318, "step": 12511} +{"lr": 7.907702595573289e-05, "data_time": 0.40063564777374266, "loss": 0.1107692077755928, "time": 0.540999436378479, "epoch": 4, "memory": 6318, "step": 12611} +{"lr": 7.877289936772454e-05, "data_time": 0.3793018341064453, "loss": 0.11471677720546722, "time": 0.5203649997711182, "epoch": 4, "memory": 6318, "step": 12711} +{"lr": 7.8467259048513e-05, "data_time": 0.35439746379852294, "loss": 0.12011936008930206, "time": 0.49799973964691163, "epoch": 4, "memory": 6318, "step": 12811} +{"lr": 7.816012445964079e-05, "data_time": 0.4329079627990723, "loss": 0.11465063616633415, "time": 0.5739084720611572, "epoch": 4, "memory": 6318, "step": 12911} +{"lr": 7.785151515779718e-05, "data_time": 0.3299607038497925, "loss": 0.11752335876226425, "time": 0.470651650428772, "epoch": 4, "memory": 6318, "step": 13011} +{"lr": 7.754145079357376e-05, "data_time": 0.40634267330169677, "loss": 0.12472412064671516, "time": 0.5458418369293213, "epoch": 4, "memory": 6318, "step": 13111} +{"lr": 7.722995111021218e-05, "data_time": 0.41192777156829835, "loss": 0.13826683908700943, "time": 0.5518895626068115, "epoch": 4, "memory": 6318, "step": 13211} +{"lr": 7.691703594234788e-05, "data_time": 0.32203140258789065, "loss": 0.1110983096063137, "time": 0.4627741098403931, "epoch": 4, "memory": 6318, "step": 13311} +{"lr": 7.660272521474673e-05, "data_time": 0.3714022159576416, "loss": 0.12002493664622307, "time": 0.5125049591064453, "epoch": 4, "memory": 6318, "step": 13411} +{"lr": 7.628703894103628e-05, "data_time": 0.3652146100997925, "loss": 0.11274023503065109, "time": 0.506047534942627, "epoch": 4, "memory": 6318, "step": 13511} +{"lr": 7.596999722243168e-05, "data_time": 0.3891819953918457, "loss": 0.10794650241732598, "time": 0.5294886350631713, "epoch": 4, "memory": 6318, "step": 13611} +{"lr": 7.565162024645551e-05, "data_time": 0.3883901834487915, "loss": 0.10773054882884026, "time": 0.5295039176940918, "epoch": 4, "memory": 6318, "step": 13711} +{"lr": 7.533192828565237e-05, "data_time": 0.33973870277404783, "loss": 0.12283506020903587, "time": 0.47932968139648435, "epoch": 4, "memory": 6318, "step": 13811} +{"lr": 7.501094169629802e-05, "data_time": 0.37243316173553465, "loss": 0.13361043035984038, "time": 0.5256778955459595, "epoch": 4, "memory": 6318, "step": 13911} +{"lr": 7.468868091710333e-05, "data_time": 0.38852458000183104, "loss": 0.125302542001009, "time": 0.5282164096832276, "epoch": 4, "memory": 6318, "step": 14011} +{"lr": 7.436516646791276e-05, "data_time": 0.3871271848678589, "loss": 0.11411226466298104, "time": 0.527640962600708, "epoch": 4, "memory": 6318, "step": 14111} +{"lr": 7.404041894839768e-05, "data_time": 0.36572928428649903, "loss": 0.11670994758605957, "time": 0.5059444427490234, "epoch": 4, "memory": 6318, "step": 14211} +{"lr": 7.371445903674494e-05, "data_time": 0.38312344551086425, "loss": 0.10678449794650077, "time": 0.5241638898849488, "epoch": 4, "memory": 6318, "step": 14311} +{"lr": 7.338730748834014e-05, "data_time": 0.3997997522354126, "loss": 0.11111574321985244, "time": 0.5402564287185669, "epoch": 4, "memory": 6318, "step": 14411} +{"lr": 7.305898513444536e-05, "data_time": 0.36980490684509276, "loss": 0.11540628373622894, "time": 0.5111876964569092, "epoch": 4, "memory": 6318, "step": 14511} +{"lr": 7.27295128808741e-05, "data_time": 0.3927323579788208, "loss": 0.11961081027984619, "time": 0.5329862356185913, "epoch": 4, "memory": 6318, "step": 14611} +{"lr": 7.239891170665884e-05, "data_time": 0.3426256895065308, "loss": 0.11875466629862785, "time": 0.4827617645263672, "epoch": 4, "memory": 6318, "step": 14711} +{"lr": 7.206720266271582e-05, "data_time": 0.4176710367202759, "loss": 0.10677655637264252, "time": 0.5592361211776733, "epoch": 4, "memory": 6318, "step": 14811} +{"lr": 7.173440687050449e-05, "data_time": 0.41552095413208007, "loss": 0.10806555673480034, "time": 0.5557098865509034, "epoch": 4, "memory": 6318, "step": 14911} +{"lr": 7.140054552068265e-05, "data_time": 0.37405297756195066, "loss": 0.1164758674800396, "time": 0.5145105838775634, "epoch": 4, "memory": 6318, "step": 15011} +{"lr": 7.106563987175705e-05, "data_time": 0.39530482292175295, "loss": 0.12077103331685066, "time": 0.5350132465362549, "epoch": 4, "memory": 6318, "step": 15111} +{"lr": 7.072971124873013e-05, "data_time": 0.39270269870758057, "loss": 0.11589161828160285, "time": 0.5341485500335693, "epoch": 4, "memory": 6318, "step": 15211} +{"lr": 7.039278104174143e-05, "data_time": 0.4198456764221191, "loss": 0.11728076860308648, "time": 0.5601847410202027, "epoch": 4, "memory": 6318, "step": 15311} +{"lr": 7.005487070470615e-05, "data_time": 0.373486065864563, "loss": 0.10975094363093377, "time": 0.513847827911377, "epoch": 4, "memory": 6318, "step": 15411} +{"lr": 6.971600175394891e-05, "data_time": 0.39898760318756105, "loss": 0.10830138102173806, "time": 0.5398354530334473, "epoch": 4, "memory": 6318, "step": 15511} +{"lr": 6.937619576683373e-05, "data_time": 0.28586156368255616, "loss": 0.09604218527674675, "time": 0.4256209135055542, "epoch": 4, "memory": 6318, "step": 15611} +{"lr": 6.903547438039021e-05, "data_time": 0.4022240161895752, "loss": 0.11488519236445427, "time": 0.547239112854004, "epoch": 4, "memory": 6318, "step": 15711} +{"accuracy/top1": 93.99464416503906, "data_time": 0.3987178555850325, "time": 0.4850656081890238, "step": 4} +{"lr": 6.856723955077045e-05, "data_time": 0.43915011882781985, "loss": 0.12675577998161316, "time": 0.5796188592910767, "epoch": 5, "memory": 6318, "step": 15848} +{"lr": 6.822443541062844e-05, "data_time": 0.40730764865875246, "loss": 0.11382345259189605, "time": 0.553311538696289, "epoch": 5, "memory": 6318, "step": 15948} +{"lr": 6.788078920910325e-05, "data_time": 0.37160983085632326, "loss": 0.11200935617089272, "time": 0.5135897397994995, "epoch": 5, "memory": 6318, "step": 16048} +{"lr": 6.753632282774874e-05, "data_time": 0.4236746311187744, "loss": 0.10685943216085433, "time": 0.5638228416442871, "epoch": 5, "memory": 6318, "step": 16148} +{"lr": 6.719105820034336e-05, "data_time": 0.38102164268493655, "loss": 0.11480921283364295, "time": 0.522061538696289, "epoch": 5, "memory": 6318, "step": 16248} +{"lr": 6.68450173114937e-05, "data_time": 0.418174147605896, "loss": 0.12086585462093354, "time": 0.5582307815551758, "epoch": 5, "memory": 6318, "step": 16348} +{"lr": 6.64982221952347e-05, "data_time": 0.39405083656311035, "loss": 0.11434520930051803, "time": 0.5336458444595337, "epoch": 5, "memory": 6318, "step": 16448} +{"lr": 6.615069493362599e-05, "data_time": 0.390876030921936, "loss": 0.11514672338962555, "time": 0.5414099216461181, "epoch": 5, "memory": 6318, "step": 16548} +{"lr": 6.580245765534682e-05, "data_time": 0.3882064580917358, "loss": 0.11900919079780578, "time": 0.5284212589263916, "epoch": 5, "memory": 6318, "step": 16648} +{"lr": 6.54535325342862e-05, "data_time": 0.38741657733917234, "loss": 0.09771207123994827, "time": 0.5280201435089111, "epoch": 5, "memory": 6318, "step": 16748} +{"lr": 6.510394178813155e-05, "data_time": 0.4036487817764282, "loss": 0.10312446132302285, "time": 0.545156741142273, "epoch": 5, "memory": 6318, "step": 16848} +{"lr": 6.475370767695331e-05, "data_time": 0.3942956686019897, "loss": 0.1205051064491272, "time": 0.5358115196228027, "epoch": 5, "memory": 6318, "step": 16948} +{"lr": 6.440285250178845e-05, "data_time": 0.3663409948348999, "loss": 0.11486078724265099, "time": 0.5059177160263062, "epoch": 5, "memory": 6318, "step": 17048} +{"lr": 6.405139860321955e-05, "data_time": 0.36364991664886476, "loss": 0.10932580530643463, "time": 0.5050055027008057, "epoch": 5, "memory": 6318, "step": 17148} +{"lr": 6.369936835995306e-05, "data_time": 0.38280601501464845, "loss": 0.10711326822638512, "time": 0.5232627868652344, "epoch": 5, "memory": 6318, "step": 17248} +{"lr": 6.334678418739388e-05, "data_time": 0.3948099374771118, "loss": 0.11118068918585777, "time": 0.5349216938018799, "epoch": 5, "memory": 6318, "step": 17348} +{"lr": 6.299366853621796e-05, "data_time": 0.4100250005722046, "loss": 0.11425347253680229, "time": 0.5508455038070679, "epoch": 5, "memory": 6318, "step": 17448} +{"lr": 6.264004389094315e-05, "data_time": 0.37438867092132566, "loss": 0.10442727133631706, "time": 0.5132425785064697, "epoch": 5, "memory": 6318, "step": 17548} +{"lr": 6.22859327684974e-05, "data_time": 0.3545452833175659, "loss": 0.10730602070689202, "time": 0.4946286678314209, "epoch": 5, "memory": 6318, "step": 17648} +{"lr": 6.193135771678478e-05, "data_time": 0.3236617803573608, "loss": 0.09932406432926655, "time": 0.4646850347518921, "epoch": 5, "memory": 6318, "step": 17748} +{"lr": 6.157634131324983e-05, "data_time": 0.33607330322265627, "loss": 0.10781325846910476, "time": 0.4764808177947998, "epoch": 5, "memory": 6318, "step": 17848} +{"lr": 6.12209061634402e-05, "data_time": 0.3357051372528076, "loss": 0.10310060307383537, "time": 0.47526533603668214, "epoch": 5, "memory": 6318, "step": 17948} +{"lr": 6.086507489956686e-05, "data_time": 0.3543382167816162, "loss": 0.10237922705709934, "time": 0.49532651901245117, "epoch": 5, "memory": 6318, "step": 18048} +{"lr": 6.0508870179063345e-05, "data_time": 0.3810558795928955, "loss": 0.10153815671801567, "time": 0.5208540916442871, "epoch": 5, "memory": 6318, "step": 18148} +{"lr": 6.015231468314303e-05, "data_time": 0.3631471633911133, "loss": 0.09847992844879627, "time": 0.5036010980606079, "epoch": 5, "memory": 6318, "step": 18248} +{"lr": 5.979543111535473e-05, "data_time": 0.35880897045135496, "loss": 0.10222495719790459, "time": 0.49943222999572756, "epoch": 5, "memory": 6318, "step": 18348} +{"lr": 5.943824220013694e-05, "data_time": 0.3660011053085327, "loss": 0.09672838598489761, "time": 0.505923056602478, "epoch": 5, "memory": 6318, "step": 18448} +{"lr": 5.908077068137128e-05, "data_time": 0.374019455909729, "loss": 0.09413647279143333, "time": 0.5142240047454834, "epoch": 5, "memory": 6318, "step": 18548} +{"lr": 5.872303932093392e-05, "data_time": 0.4149693727493286, "loss": 0.10969468727707862, "time": 0.5547202825546265, "epoch": 5, "memory": 6318, "step": 18648} +{"lr": 5.836507089724633e-05, "data_time": 0.3780791759490967, "loss": 0.10508418828248978, "time": 0.5185910940170289, "epoch": 5, "memory": 6318, "step": 18748} +{"lr": 5.800688820382498e-05, "data_time": 0.4017800331115723, "loss": 0.10637912973761558, "time": 0.5407953977584838, "epoch": 5, "memory": 6318, "step": 18848} +{"lr": 5.7648514047829945e-05, "data_time": 0.385778021812439, "loss": 0.11188452839851379, "time": 0.5261282920837402, "epoch": 5, "memory": 6318, "step": 18948} +{"lr": 5.7289971248612474e-05, "data_time": 0.3536604642868042, "loss": 0.11847224235534667, "time": 0.4933535099029541, "epoch": 5, "memory": 6318, "step": 19048} +{"lr": 5.693128263626215e-05, "data_time": 0.3780566692352295, "loss": 0.11355301141738891, "time": 0.5185465812683105, "epoch": 5, "memory": 6318, "step": 19148} +{"lr": 5.6572471050153236e-05, "data_time": 0.3987316370010376, "loss": 0.11193219497799874, "time": 0.5417137861251831, "epoch": 5, "memory": 6318, "step": 19248} +{"lr": 5.621355933749018e-05, "data_time": 0.3865906000137329, "loss": 0.11184052303433419, "time": 0.5274548768997193, "epoch": 5, "memory": 6318, "step": 19348} +{"lr": 5.5854570351853106e-05, "data_time": 0.38541455268859864, "loss": 0.0927666824311018, "time": 0.5253933429718017, "epoch": 5, "memory": 6318, "step": 19448} +{"lr": 5.5495526951742315e-05, "data_time": 0.3779834508895874, "loss": 0.10130834355950355, "time": 0.5167313337326049, "epoch": 5, "memory": 6318, "step": 19548} +{"lr": 5.513645199912304e-05, "data_time": 0.37412319183349607, "loss": 0.10644027814269066, "time": 0.5151644706726074, "epoch": 5, "memory": 6318, "step": 19648} +{"accuracy/top1": 94.8575210571289, "data_time": 0.39589020301555766, "time": 0.4818695619188506, "step": 5} +{"lr": 5.464450978670876e-05, "data_time": 0.30408563613891604, "loss": 0.10174725353717803, "time": 0.44475257396698, "epoch": 6, "memory": 6318, "step": 19785} +{"lr": 5.4285451361545876e-05, "data_time": 0.3650784730911255, "loss": 0.11228409856557846, "time": 0.5063369989395141, "epoch": 6, "memory": 6318, "step": 19885} +{"lr": 5.3926438435022166e-05, "data_time": 0.3365126609802246, "loss": 0.09968096762895584, "time": 0.47697179317474364, "epoch": 6, "memory": 6318, "step": 19985} +{"lr": 5.356749386716241e-05, "data_time": 0.3476013422012329, "loss": 0.10715125873684883, "time": 0.49696924686431887, "epoch": 6, "memory": 6318, "step": 20085} +{"lr": 5.320864051363873e-05, "data_time": 0.3483818292617798, "loss": 0.09811747744679451, "time": 0.4877533674240112, "epoch": 6, "memory": 6318, "step": 20185} +{"lr": 5.2849901224315175e-05, "data_time": 0.34649107456207273, "loss": 0.10409680530428886, "time": 0.4863214254379272, "epoch": 6, "memory": 6318, "step": 20285} +{"lr": 5.249129884179276e-05, "data_time": 0.47657930850982666, "loss": 0.11266562156379223, "time": 0.6157445907592773, "epoch": 6, "memory": 6318, "step": 20385} +{"lr": 5.2132856199954985e-05, "data_time": 0.4233354091644287, "loss": 0.09786221906542777, "time": 0.5636491537094116, "epoch": 6, "memory": 6318, "step": 20485} +{"lr": 5.177459612251411e-05, "data_time": 0.40064258575439454, "loss": 0.10372036993503571, "time": 0.5405067682266236, "epoch": 6, "memory": 6318, "step": 20585} +{"lr": 5.141654142155744e-05, "data_time": 0.3782562971115112, "loss": 0.09827804118394852, "time": 0.5185530424118042, "epoch": 6, "memory": 6318, "step": 20685} +{"lr": 5.105871489609512e-05, "data_time": 0.37419517040252687, "loss": 0.11622708737850189, "time": 0.5142244100570679, "epoch": 6, "memory": 6318, "step": 20785} +{"lr": 5.070113933060828e-05, "data_time": 0.3944650411605835, "loss": 0.0945027407258749, "time": 0.5351663827896118, "epoch": 6, "memory": 6318, "step": 20885} +{"lr": 5.0343837493598264e-05, "data_time": 0.4023817777633667, "loss": 0.10340002775192261, "time": 0.5427498102188111, "epoch": 6, "memory": 6318, "step": 20985} +{"lr": 4.998683213613675e-05, "data_time": 0.3955320119857788, "loss": 0.10043730363249778, "time": 0.5366127252578735, "epoch": 6, "memory": 6318, "step": 21085} +{"lr": 4.963014599041739e-05, "data_time": 0.6567476511001586, "loss": 0.10432487577199936, "time": 0.7972772121429443, "epoch": 6, "memory": 6318, "step": 21185} +{"lr": 4.927380176830793e-05, "data_time": 0.3898554563522339, "loss": 0.10532468929886818, "time": 0.5309202671051025, "epoch": 6, "memory": 6318, "step": 21285} +{"lr": 4.89178221599044e-05, "data_time": 0.37075784206390383, "loss": 0.11055532693862916, "time": 0.5113343000411987, "epoch": 6, "memory": 6318, "step": 21385} +{"lr": 4.85622298320861e-05, "data_time": 0.36381983757019043, "loss": 0.1211197942495346, "time": 0.5040167570114136, "epoch": 6, "memory": 6318, "step": 21485} +{"lr": 4.8207047427072355e-05, "data_time": 0.4025398254394531, "loss": 0.10142585933208466, "time": 0.5424248218536377, "epoch": 6, "memory": 6318, "step": 21585} +{"lr": 4.7852297560980876e-05, "data_time": 0.35801057815551757, "loss": 0.11761986836791039, "time": 0.49829421043395994, "epoch": 6, "memory": 6318, "step": 21685} +{"lr": 4.749800282238749e-05, "data_time": 0.37150614261627196, "loss": 0.09721977785229682, "time": 0.5115808725357056, "epoch": 6, "memory": 6318, "step": 21785} +{"lr": 4.714418577088805e-05, "data_time": 0.3824015617370605, "loss": 0.107598315179348, "time": 0.5228906393051147, "epoch": 6, "memory": 6318, "step": 21885} +{"lr": 4.679086893566184e-05, "data_time": 0.39792261123657224, "loss": 0.11070032864809036, "time": 0.5383080005645752, "epoch": 6, "memory": 6318, "step": 21985} +{"lr": 4.643807481403694e-05, "data_time": 0.37602369785308837, "loss": 0.10062609389424323, "time": 0.5161623954772949, "epoch": 6, "memory": 6318, "step": 22085} +{"lr": 4.6085825870057946e-05, "data_time": 0.3849221706390381, "loss": 0.10851487964391708, "time": 0.5263155937194824, "epoch": 6, "memory": 6318, "step": 22185} +{"lr": 4.573414453305537e-05, "data_time": 0.311960506439209, "loss": 0.10821853280067444, "time": 0.458675479888916, "epoch": 6, "memory": 6318, "step": 22285} +{"lr": 4.5383053196217514e-05, "data_time": 0.3300865650177002, "loss": 0.10318279191851616, "time": 0.470539927482605, "epoch": 6, "memory": 6318, "step": 22385} +{"lr": 4.503257421516477e-05, "data_time": 0.360955810546875, "loss": 0.10468805134296418, "time": 0.5014185190200806, "epoch": 6, "memory": 6318, "step": 22485} +{"lr": 4.468272990652586e-05, "data_time": 0.32948927879333495, "loss": 0.10729858800768852, "time": 0.47054457664489746, "epoch": 6, "memory": 6318, "step": 22585} +{"lr": 4.433354254651697e-05, "data_time": 0.32231760025024414, "loss": 0.10081584528088569, "time": 0.463468337059021, "epoch": 6, "memory": 6318, "step": 22685} +{"lr": 4.398503436952338e-05, "data_time": 0.3296447992324829, "loss": 0.10046930015087127, "time": 0.47072160243988037, "epoch": 6, "memory": 6318, "step": 22785} +{"lr": 4.363722756668351e-05, "data_time": 0.33118667602539065, "loss": 0.0980569876730442, "time": 0.4722886085510254, "epoch": 6, "memory": 6318, "step": 22885} +{"lr": 4.329014428447611e-05, "data_time": 0.34980001449584963, "loss": 0.08689046539366245, "time": 0.4898162126541138, "epoch": 6, "memory": 6318, "step": 22985} +{"lr": 4.294380662331e-05, "data_time": 0.3280327558517456, "loss": 0.10173123776912689, "time": 0.4689242601394653, "epoch": 6, "memory": 6318, "step": 23085} +{"lr": 4.259823663611678e-05, "data_time": 0.3117323398590088, "loss": 0.10396727472543717, "time": 0.4515509605407715, "epoch": 6, "memory": 6318, "step": 23185} +{"lr": 4.22534563269468e-05, "data_time": 0.32997186183929444, "loss": 0.10429118499159813, "time": 0.4694223403930664, "epoch": 6, "memory": 6318, "step": 23285} +{"lr": 4.1909487649567694e-05, "data_time": 0.26676576137542723, "loss": 0.1056912124156952, "time": 0.4968456506729126, "epoch": 6, "memory": 6318, "step": 23385} +{"lr": 4.156635250606698e-05, "data_time": 0.2769592761993408, "loss": 0.10443811044096947, "time": 0.44726712703704835, "epoch": 6, "memory": 6318, "step": 23485} +{"lr": 4.122407274545703e-05, "data_time": 0.2976927518844604, "loss": 0.11196320131421089, "time": 0.4809021711349487, "epoch": 6, "memory": 6318, "step": 23585} +{"accuracy/top1": 95.65777587890625, "data_time": 0.3768713720913591, "time": 0.4641124462259227, "step": 6} +{"lr": 4.075657788156285e-05, "data_time": 0.35293667316436766, "loss": 0.0987037293612957, "time": 0.4974281072616577, "epoch": 7, "memory": 6318, "step": 23722} +{"lr": 4.0416412311741566e-05, "data_time": 0.3285355567932129, "loss": 0.10428239926695823, "time": 0.4689664363861084, "epoch": 7, "memory": 6318, "step": 23822} +{"lr": 4.007717534685821e-05, "data_time": 0.34826011657714845, "loss": 0.1055700808763504, "time": 0.489159893989563, "epoch": 7, "memory": 6318, "step": 23922} +{"lr": 3.973888858770982e-05, "data_time": 0.3486127138137817, "loss": 0.08842075392603874, "time": 0.48857219219207765, "epoch": 7, "memory": 6318, "step": 24022} +{"lr": 3.9401573574589535e-05, "data_time": 0.3241807699203491, "loss": 0.09261823780834674, "time": 0.47283239364624025, "epoch": 7, "memory": 6318, "step": 24122} +{"lr": 3.906525178591511e-05, "data_time": 0.1736433982849121, "loss": 0.10432448759675025, "time": 0.5062353610992432, "epoch": 7, "memory": 6318, "step": 24222} +{"lr": 3.87299446368609e-05, "data_time": 0.11616425514221192, "loss": 0.10748223438858986, "time": 0.48773789405822754, "epoch": 7, "memory": 6318, "step": 24322} +{"lr": 3.839567347799442e-05, "data_time": 0.06165766716003418, "loss": 0.10103201270103454, "time": 0.4622232437133789, "epoch": 7, "memory": 6318, "step": 24422} +{"lr": 3.806245959391694e-05, "data_time": 0.13522274494171144, "loss": 0.09810488671064377, "time": 0.506796407699585, "epoch": 7, "memory": 6318, "step": 24522} +{"lr": 3.773032420190816e-05, "data_time": 0.14675233364105225, "loss": 0.0984394371509552, "time": 0.44760453701019287, "epoch": 7, "memory": 6318, "step": 24622} +{"lr": 3.739928845057507e-05, "data_time": 0.33199052810668944, "loss": 0.09539510831236839, "time": 0.47165915966033933, "epoch": 7, "memory": 6318, "step": 24722} +{"lr": 3.706937341850541e-05, "data_time": 0.3119044780731201, "loss": 0.09639893397688866, "time": 0.45209035873413084, "epoch": 7, "memory": 6318, "step": 24822} +{"lr": 3.674060011292556e-05, "data_time": 0.3763544797897339, "loss": 0.09282100200653076, "time": 0.517037034034729, "epoch": 7, "memory": 6318, "step": 24922} +{"lr": 3.641298946836287e-05, "data_time": 0.30643692016601565, "loss": 0.09610395208001137, "time": 0.4686431884765625, "epoch": 7, "memory": 6318, "step": 25022} +{"lr": 3.6086562345312595e-05, "data_time": 0.28463425636291506, "loss": 0.08842328265309334, "time": 0.465103006362915, "epoch": 7, "memory": 6318, "step": 25122} +{"lr": 3.576133952890974e-05, "data_time": 0.3651818513870239, "loss": 0.09897352643311023, "time": 0.5050853967666626, "epoch": 7, "memory": 6318, "step": 25222} +{"lr": 3.543734172760541e-05, "data_time": 0.33870925903320315, "loss": 0.10209830403327942, "time": 0.47871851921081543, "epoch": 7, "memory": 6318, "step": 25322} +{"lr": 3.511458957184831e-05, "data_time": 0.3305222749710083, "loss": 0.09417184218764305, "time": 0.4711876630783081, "epoch": 7, "memory": 6318, "step": 25422} +{"lr": 3.479310361277118e-05, "data_time": 0.2781820774078369, "loss": 0.10061065405607224, "time": 0.5552470207214355, "epoch": 7, "memory": 6318, "step": 25522} +{"lr": 3.4472904320882065e-05, "data_time": 0.1854102849960327, "loss": 0.10505718365311623, "time": 0.4947544813156128, "epoch": 7, "memory": 6318, "step": 25622} +{"lr": 3.4154012084760864e-05, "data_time": 0.09341263771057129, "loss": 0.09473652243614197, "time": 0.4275254011154175, "epoch": 7, "memory": 6318, "step": 25722} +{"lr": 3.38364472097612e-05, "data_time": 0.3228015661239624, "loss": 0.10724164620041847, "time": 0.46433560848236083, "epoch": 7, "memory": 6318, "step": 25822} +{"lr": 3.3520229916717485e-05, "data_time": 0.2344813585281372, "loss": 0.11206526979804039, "time": 0.47754745483398436, "epoch": 7, "memory": 6318, "step": 25922} +{"lr": 3.320538034065722e-05, "data_time": 0.20822627544403077, "loss": 0.10250148177146912, "time": 0.49203219413757326, "epoch": 7, "memory": 6318, "step": 26022} +{"lr": 3.289191852951901e-05, "data_time": 0.03485782146453857, "loss": 0.1034554086625576, "time": 0.4966739177703857, "epoch": 7, "memory": 6318, "step": 26122} +{"lr": 3.257986444287605e-05, "data_time": 0.0009075164794921875, "loss": 0.08920665830373764, "time": 0.44400362968444823, "epoch": 7, "memory": 6318, "step": 26222} +{"lr": 3.226923795066519e-05, "data_time": 0.011821389198303223, "loss": 0.10243339091539383, "time": 0.479843807220459, "epoch": 7, "memory": 6318, "step": 26322} +{"lr": 3.196005883192158e-05, "data_time": 0.001040482521057129, "loss": 0.09856786578893661, "time": 0.5114281177520752, "epoch": 7, "memory": 6318, "step": 26422} +{"lr": 3.165234677351947e-05, "data_time": 0.1300366163253784, "loss": 0.1036761175841093, "time": 0.5245226383209228, "epoch": 7, "memory": 6318, "step": 26522} +{"lr": 3.134612136891845e-05, "data_time": 0.12235636711120605, "loss": 0.1240170106291771, "time": 0.4479642391204834, "epoch": 7, "memory": 6318, "step": 26622} +{"lr": 3.104140211691602e-05, "data_time": 0.19092833995819092, "loss": 0.09769907668232918, "time": 0.4890336275100708, "epoch": 7, "memory": 6318, "step": 26722} +{"lr": 3.073820842040592e-05, "data_time": 0.3902705192565918, "loss": 0.0990387573838234, "time": 0.5317658424377442, "epoch": 7, "memory": 6318, "step": 26822} +{"lr": 3.043655958514262e-05, "data_time": 0.298284649848938, "loss": 0.08826142475008965, "time": 0.4417326211929321, "epoch": 7, "memory": 6318, "step": 26922} +{"lr": 3.013647481851217e-05, "data_time": 0.3087068796157837, "loss": 0.08713078796863556, "time": 0.4484199047088623, "epoch": 7, "memory": 6318, "step": 27022} +{"lr": 2.983797322830899e-05, "data_time": 0.33893604278564454, "loss": 0.10766812637448311, "time": 0.4787930727005005, "epoch": 7, "memory": 6318, "step": 27122} +{"lr": 2.9541073821519346e-05, "data_time": 0.345421576499939, "loss": 0.10380721762776375, "time": 0.4860095024108887, "epoch": 7, "memory": 6318, "step": 27222} +{"lr": 2.924579550311095e-05, "data_time": 0.3214087009429932, "loss": 0.10154414102435112, "time": 0.46173095703125, "epoch": 7, "memory": 6318, "step": 27322} +{"lr": 2.8952157074829418e-05, "data_time": 0.3555649518966675, "loss": 0.10850416794419289, "time": 0.49585533142089844, "epoch": 7, "memory": 6318, "step": 27422} +{"lr": 2.866017723400071e-05, "data_time": 0.8572046279907226, "loss": 0.0841876830905676, "time": 0.9998951911926269, "epoch": 7, "memory": 6318, "step": 27522} +{"accuracy/top1": 95.97787475585938, "data_time": 0.3753005225082924, "time": 0.46172171625597724, "step": 7} +{"lr": 2.8262891373132927e-05, "data_time": 0.3492342233657837, "loss": 0.10151149034500122, "time": 0.48925559520721434, "epoch": 8, "memory": 6318, "step": 27659} +{"lr": 2.797491643470029e-05, "data_time": 0.35636391639709475, "loss": 0.09571295753121375, "time": 0.4970815181732178, "epoch": 8, "memory": 6318, "step": 27759} +{"lr": 2.768866230918266e-05, "data_time": 0.35486526489257814, "loss": 0.095349533110857, "time": 0.49532508850097656, "epoch": 8, "memory": 6318, "step": 27859} +{"lr": 2.7404147223713185e-05, "data_time": 0.31156766414642334, "loss": 0.09307517856359482, "time": 0.451794958114624, "epoch": 8, "memory": 6318, "step": 27959} +{"lr": 2.712138929469222e-05, "data_time": 0.33141589164733887, "loss": 0.11230526641011238, "time": 0.47066478729248046, "epoch": 8, "memory": 6318, "step": 28059} +{"lr": 2.6840406526633896e-05, "data_time": 0.3271724462509155, "loss": 0.08288054689764976, "time": 0.46551098823547366, "epoch": 8, "memory": 6318, "step": 28159} +{"lr": 2.6561216811019345e-05, "data_time": 0.3334389925003052, "loss": 0.0996477335691452, "time": 0.4717499494552612, "epoch": 8, "memory": 6318, "step": 28259} +{"lr": 2.6283837925157908e-05, "data_time": 0.3351886749267578, "loss": 0.08946491815149785, "time": 0.475070858001709, "epoch": 8, "memory": 6318, "step": 28359} +{"lr": 2.6008287531054805e-05, "data_time": 0.34549751281738283, "loss": 0.09487205892801284, "time": 0.4862659931182861, "epoch": 8, "memory": 6318, "step": 28459} +{"lr": 2.5734583174286796e-05, "data_time": 0.3381718635559082, "loss": 0.0884143128991127, "time": 0.4779308319091797, "epoch": 8, "memory": 6318, "step": 28559} +{"lr": 2.546274228288488e-05, "data_time": 0.35170581340789797, "loss": 0.098910591006279, "time": 0.491578197479248, "epoch": 8, "memory": 6318, "step": 28659} +{"lr": 2.5192782166224308e-05, "data_time": 0.36029911041259766, "loss": 0.10099406689405441, "time": 0.4994908571243286, "epoch": 8, "memory": 6318, "step": 28759} +{"lr": 2.492472001392276e-05, "data_time": 0.3378411054611206, "loss": 0.10261294022202491, "time": 0.477240777015686, "epoch": 8, "memory": 6318, "step": 28859} +{"lr": 2.4658572894745857e-05, "data_time": 0.3492377758026123, "loss": 0.10785484537482262, "time": 0.4885077476501465, "epoch": 8, "memory": 6318, "step": 28959} +{"lr": 2.439435775551997e-05, "data_time": 0.3725048780441284, "loss": 0.09029463529586793, "time": 0.5131029605865478, "epoch": 8, "memory": 6318, "step": 29059} +{"lr": 2.4132091420053357e-05, "data_time": 0.3452200412750244, "loss": 0.09120306968688965, "time": 0.48561174869537355, "epoch": 8, "memory": 6318, "step": 29159} +{"lr": 2.387179058806494e-05, "data_time": 0.35843653678894044, "loss": 0.10985691845417023, "time": 0.49903619289398193, "epoch": 8, "memory": 6318, "step": 29259} +{"lr": 2.361347183412073e-05, "data_time": 0.3468772411346436, "loss": 0.10330413170158863, "time": 0.487683629989624, "epoch": 8, "memory": 6318, "step": 29359} +{"lr": 2.3357151606578728e-05, "data_time": 0.2944147825241089, "loss": 0.08409390300512314, "time": 0.44464750289916993, "epoch": 8, "memory": 6318, "step": 29459} +{"lr": 2.3102846226541355e-05, "data_time": 0.32512702941894533, "loss": 0.0902996003627777, "time": 0.4659524440765381, "epoch": 8, "memory": 6318, "step": 29559} +{"lr": 2.2850571886816365e-05, "data_time": 0.3600844144821167, "loss": 0.09483547136187553, "time": 0.5001818418502808, "epoch": 8, "memory": 6318, "step": 29659} +{"lr": 2.260034465088568e-05, "data_time": 0.2988943099975586, "loss": 0.079250842705369, "time": 0.43875305652618407, "epoch": 8, "memory": 6318, "step": 29759} +{"lr": 2.2352180451882643e-05, "data_time": 0.38458776473999023, "loss": 0.10099021680653095, "time": 0.524933123588562, "epoch": 8, "memory": 6318, "step": 29859} +{"lr": 2.210609509157741e-05, "data_time": 0.33091471195220945, "loss": 0.09827216044068336, "time": 0.4710878849029541, "epoch": 8, "memory": 6318, "step": 29959} +{"lr": 2.1862104239370772e-05, "data_time": 0.3638679265975952, "loss": 0.09145785272121429, "time": 0.5036179780960083, "epoch": 8, "memory": 6318, "step": 30059} +{"lr": 2.1620223431296407e-05, "data_time": 0.3417232513427734, "loss": 0.09443091079592705, "time": 0.4819511651992798, "epoch": 8, "memory": 6318, "step": 30159} +{"lr": 2.1380468069031744e-05, "data_time": 0.36420743465423583, "loss": 0.09029760658740997, "time": 0.505265760421753, "epoch": 8, "memory": 6318, "step": 30259} +{"lr": 2.1142853418917185e-05, "data_time": 0.312066650390625, "loss": 0.10014737099409103, "time": 0.4526773691177368, "epoch": 8, "memory": 6318, "step": 30359} +{"lr": 2.0907394610983892e-05, "data_time": 0.40809338092803954, "loss": 0.09974495694041252, "time": 0.5498457670211792, "epoch": 8, "memory": 6318, "step": 30459} +{"lr": 2.0674106637990643e-05, "data_time": 0.3659006118774414, "loss": 0.10275045484304428, "time": 0.5057721853256225, "epoch": 8, "memory": 6318, "step": 30559} +{"lr": 2.0443004354468975e-05, "data_time": 0.306179404258728, "loss": 0.1055156260728836, "time": 0.4466289281845093, "epoch": 8, "memory": 6318, "step": 30659} +{"lr": 2.0214102475777453e-05, "data_time": 0.2818570852279663, "loss": 0.10063871070742607, "time": 0.42209014892578123, "epoch": 8, "memory": 6318, "step": 30759} +{"lr": 1.9987415577164583e-05, "data_time": 0.39331493377685545, "loss": 0.10316462069749832, "time": 0.5357325077056885, "epoch": 8, "memory": 6318, "step": 30859} +{"lr": 1.9762958092840814e-05, "data_time": 0.39673705101013185, "loss": 0.09898230955004692, "time": 0.5374812602996826, "epoch": 8, "memory": 6318, "step": 30959} +{"lr": 1.9540744315059334e-05, "data_time": 0.35516085624694826, "loss": 0.09462017826735973, "time": 0.4960510730743408, "epoch": 8, "memory": 6318, "step": 31059} +{"lr": 1.9320788393206132e-05, "data_time": 0.33659427165985106, "loss": 0.08460660092532635, "time": 0.475318455696106, "epoch": 8, "memory": 6318, "step": 31159} +{"lr": 1.9103104332899027e-05, "data_time": 0.3362543821334839, "loss": 0.11053954884409904, "time": 0.475168514251709, "epoch": 8, "memory": 6318, "step": 31259} +{"lr": 1.888770599509573e-05, "data_time": 0.36495540142059324, "loss": 0.09862048029899598, "time": 0.5039122104644775, "epoch": 8, "memory": 6318, "step": 31359} +{"lr": 1.8674607095211516e-05, "data_time": 0.35937793254852296, "loss": 0.10310119204223156, "time": 0.49950060844421384, "epoch": 8, "memory": 6318, "step": 31459} +{"accuracy/top1": 96.09964752197266, "data_time": 0.3809464265560282, "time": 0.46951562371747246, "step": 8} +{"lr": 1.838641934548488e-05, "data_time": 0.2756743192672729, "loss": 0.0984993189573288, "time": 0.4895286560058594, "epoch": 9, "memory": 6318, "step": 31596} +{"lr": 1.817882403185378e-05, "data_time": 0.3203425884246826, "loss": 0.0889295682311058, "time": 0.4642958164215088, "epoch": 9, "memory": 6318, "step": 31696} +{"lr": 1.797357329395729e-05, "data_time": 0.3588765382766724, "loss": 0.07670582234859466, "time": 0.5104834556579589, "epoch": 9, "memory": 6318, "step": 31796} +{"lr": 1.7770680201065646e-05, "data_time": 0.3181695222854614, "loss": 0.09116340428590775, "time": 0.4833985328674316, "epoch": 9, "memory": 6318, "step": 31896} +{"lr": 1.7570157672326885e-05, "data_time": 0.3544581651687622, "loss": 0.09792966432869435, "time": 0.49512252807617185, "epoch": 9, "memory": 6318, "step": 31996} +{"lr": 1.7372018475944145e-05, "data_time": 0.3643302917480469, "loss": 0.101457005366683, "time": 0.5042122602462769, "epoch": 9, "memory": 6318, "step": 32096} +{"lr": 1.717627522836271e-05, "data_time": 0.3254039764404297, "loss": 0.10682158172130585, "time": 0.4696770668029785, "epoch": 9, "memory": 6318, "step": 32196} +{"lr": 1.6982940393466685e-05, "data_time": 0.3297384738922119, "loss": 0.09523039907217026, "time": 0.4708113670349121, "epoch": 9, "memory": 6318, "step": 32296} +{"lr": 1.6792026281785218e-05, "data_time": 0.18743543624877929, "loss": 0.10642971992492675, "time": 0.49167566299438475, "epoch": 9, "memory": 6318, "step": 32396} +{"lr": 1.660354504970888e-05, "data_time": 0.20626299381256102, "loss": 0.09539871588349343, "time": 0.4402266263961792, "epoch": 9, "memory": 6318, "step": 32496} +{"lr": 1.641750869871534e-05, "data_time": 0.3184908390045166, "loss": 0.09498243555426597, "time": 0.47086918354034424, "epoch": 9, "memory": 6318, "step": 32596} +{"lr": 1.6233929074605374e-05, "data_time": 0.16208815574645996, "loss": 0.1013569600880146, "time": 0.45047998428344727, "epoch": 9, "memory": 6318, "step": 32696} +{"lr": 1.6052817866748513e-05, "data_time": 0.0014832019805908203, "loss": 0.08854821175336838, "time": 0.45998849868774416, "epoch": 9, "memory": 6318, "step": 32796} +{"lr": 1.587418660733871e-05, "data_time": 0.0011659860610961914, "loss": 0.09278179705142975, "time": 0.47566611766815187, "epoch": 9, "memory": 6318, "step": 32896} +{"lr": 1.5698046670660044e-05, "data_time": 0.0008747339248657227, "loss": 0.09943257197737694, "time": 0.4926849603652954, "epoch": 9, "memory": 6318, "step": 32996} +{"lr": 1.5524409272362524e-05, "data_time": 0.0009041309356689453, "loss": 0.0882600724697113, "time": 0.440548300743103, "epoch": 9, "memory": 6318, "step": 33096} +{"lr": 1.5353285468747853e-05, "data_time": 0.0009421586990356445, "loss": 0.09573952667415142, "time": 0.45577971935272216, "epoch": 9, "memory": 6318, "step": 33196} +{"lr": 1.5184686156065442e-05, "data_time": 0.0008428096771240234, "loss": 0.0951104611158371, "time": 0.4894315958023071, "epoch": 9, "memory": 6318, "step": 33296} +{"lr": 1.5018622069818636e-05, "data_time": 0.0009381055831909179, "loss": 0.09233948960900307, "time": 0.5258806467056274, "epoch": 9, "memory": 6318, "step": 33396} +{"lr": 1.4855103784081099e-05, "data_time": 0.0008928298950195313, "loss": 0.10468098372220994, "time": 0.48557496070861816, "epoch": 9, "memory": 6318, "step": 33496} +{"lr": 1.4694141710823508e-05, "data_time": 0.001069927215576172, "loss": 0.09393770471215249, "time": 0.4717130422592163, "epoch": 9, "memory": 6318, "step": 33596} +{"lr": 1.4535746099250614e-05, "data_time": 0.0011947154998779297, "loss": 0.09650558307766914, "time": 0.4393578052520752, "epoch": 9, "memory": 6318, "step": 33696} +{"lr": 1.4379927035148576e-05, "data_time": 0.0013447284698486328, "loss": 0.09026336446404457, "time": 0.5092681884765625, "epoch": 9, "memory": 6318, "step": 33796} +{"lr": 1.42266944402428e-05, "data_time": 0.0010889530181884765, "loss": 0.0928823359310627, "time": 0.4522377967834473, "epoch": 9, "memory": 6318, "step": 33896} +{"lr": 1.4076058071566133e-05, "data_time": 0.0014900922775268554, "loss": 0.08821561448276043, "time": 0.5187438488006592, "epoch": 9, "memory": 6318, "step": 33996} +{"lr": 1.3928027520837635e-05, "data_time": 0.001142573356628418, "loss": 0.09485570564866067, "time": 0.4955988883972168, "epoch": 9, "memory": 6318, "step": 34096} +{"lr": 1.3782612213851759e-05, "data_time": 0.0010311126708984375, "loss": 0.07379315569996833, "time": 0.4372359275817871, "epoch": 9, "memory": 6318, "step": 34196} +{"lr": 1.3639821409878263e-05, "data_time": 0.056370210647583005, "loss": 0.10082559026777745, "time": 0.4919793128967285, "epoch": 9, "memory": 6318, "step": 34296} +{"lr": 1.3499664201072546e-05, "data_time": 0.20958466529846193, "loss": 0.09546709209680557, "time": 0.4997179269790649, "epoch": 9, "memory": 6318, "step": 34396} +{"lr": 1.336214951189674e-05, "data_time": 0.05220351219177246, "loss": 0.10569841042160988, "time": 0.49487416744232177, "epoch": 9, "memory": 6318, "step": 34496} +{"lr": 1.3227286098551456e-05, "data_time": 0.12370331287384033, "loss": 0.08879945501685142, "time": 0.5162428140640258, "epoch": 9, "memory": 6318, "step": 34596} +{"lr": 1.3095082548418224e-05, "data_time": 0.13003745079040527, "loss": 0.09212391600012779, "time": 0.48962717056274413, "epoch": 9, "memory": 6318, "step": 34696} +{"lr": 1.2965547279512709e-05, "data_time": 0.15012824535369873, "loss": 0.09019293710589409, "time": 0.49649505615234374, "epoch": 9, "memory": 6318, "step": 34796} +{"lr": 1.2838688539948659e-05, "data_time": 0.1678551197052002, "loss": 0.10418915376067162, "time": 0.5437381029129028, "epoch": 9, "memory": 6318, "step": 34896} +{"lr": 1.2714514407412763e-05, "data_time": 0.328688907623291, "loss": 0.09173640720546246, "time": 0.4752793788909912, "epoch": 9, "memory": 6318, "step": 34996} +{"lr": 1.2593032788650256e-05, "data_time": 0.36444292068481443, "loss": 0.09632524996995925, "time": 0.5047950267791748, "epoch": 9, "memory": 6318, "step": 35096} +{"lr": 1.2474251418961484e-05, "data_time": 0.433282732963562, "loss": 0.08514822125434876, "time": 0.5742971658706665, "epoch": 9, "memory": 6318, "step": 35196} +{"lr": 1.2358177861709387e-05, "data_time": 0.2728005647659302, "loss": 0.09738615900278091, "time": 0.5269148349761963, "epoch": 9, "memory": 6318, "step": 35296} +{"lr": 1.2244819507837855e-05, "data_time": 0.32030434608459474, "loss": 0.0945310577750206, "time": 0.48576650619506834, "epoch": 9, "memory": 6318, "step": 35396} +{"accuracy/top1": 96.61807250976562, "data_time": 0.3732914760194976, "time": 0.4594194683535346, "step": 9} +{"lr": 1.2093939694118031e-05, "data_time": 0.30083954334259033, "loss": 0.1128064289689064, "time": 0.44062635898590086, "epoch": 10, "memory": 6318, "step": 35533} +{"lr": 1.1987044880988498e-05, "data_time": 0.3775103330612183, "loss": 0.08815532177686691, "time": 0.516980791091919, "epoch": 10, "memory": 6318, "step": 35633} +{"lr": 1.1882888903004625e-05, "data_time": 0.33387250900268556, "loss": 0.09743470251560211, "time": 0.4734462022781372, "epoch": 10, "memory": 6318, "step": 35733} +{"lr": 1.178147839226253e-05, "data_time": 0.34879398345947266, "loss": 0.08691700845956803, "time": 0.4880156278610229, "epoch": 10, "memory": 6318, "step": 35833} +{"lr": 1.1682819806041682e-05, "data_time": 0.3381638765335083, "loss": 0.08479001820087433, "time": 0.4828916311264038, "epoch": 10, "memory": 6318, "step": 35933} +{"lr": 1.1586919426393653e-05, "data_time": 0.31300647258758546, "loss": 0.09261269718408585, "time": 0.45303988456726074, "epoch": 10, "memory": 6318, "step": 36033} +{"lr": 1.1493783359742196e-05, "data_time": 0.30055234432220457, "loss": 0.08887546919286252, "time": 0.4570359706878662, "epoch": 10, "memory": 6318, "step": 36133} +{"lr": 1.1403417536494367e-05, "data_time": 0.3454045295715332, "loss": 0.08568267896771431, "time": 0.48510792255401614, "epoch": 10, "memory": 6318, "step": 36233} +{"lr": 1.1315827710662908e-05, "data_time": 0.3492480993270874, "loss": 0.08701312318444251, "time": 0.48921523094177244, "epoch": 10, "memory": 6318, "step": 36333} +{"lr": 1.1231019459499938e-05, "data_time": 0.3514674425125122, "loss": 0.09252516627311706, "time": 0.49141180515289307, "epoch": 10, "memory": 6318, "step": 36433} +{"lr": 1.1148998183141688e-05, "data_time": 0.5685892581939698, "loss": 0.09964828714728355, "time": 0.708743953704834, "epoch": 10, "memory": 6318, "step": 36533} +{"lr": 1.1069769104264779e-05, "data_time": 0.3947058439254761, "loss": 0.09140113741159439, "time": 0.5361949443817139, "epoch": 10, "memory": 6318, "step": 36633} +{"lr": 1.0993337267753594e-05, "data_time": 0.298458456993103, "loss": 0.07913832552731037, "time": 0.439842963218689, "epoch": 10, "memory": 6318, "step": 36733} +{"lr": 1.0919707540379088e-05, "data_time": 0.3678726673126221, "loss": 0.09146838709712028, "time": 0.5092720270156861, "epoch": 10, "memory": 6318, "step": 36833} +{"lr": 1.0848884610488858e-05, "data_time": 0.7514483213424683, "loss": 0.0948871422559023, "time": 1.0796780109405517, "epoch": 10, "memory": 6318, "step": 36933} +{"lr": 1.0780872987708645e-05, "data_time": 0.3007082462310791, "loss": 0.08585912808775902, "time": 0.45542125701904296, "epoch": 10, "memory": 6318, "step": 37033} +{"lr": 1.0715677002655166e-05, "data_time": 0.14177203178405762, "loss": 0.10550803169608117, "time": 0.4755758047103882, "epoch": 10, "memory": 6318, "step": 37133} +{"lr": 1.0653300806660372e-05, "data_time": 0.11019306182861328, "loss": 0.08652376122772694, "time": 0.4652707099914551, "epoch": 10, "memory": 6318, "step": 37233} +{"lr": 1.0593748371507151e-05, "data_time": 0.2064640998840332, "loss": 0.0875510510057211, "time": 0.4914530038833618, "epoch": 10, "memory": 6318, "step": 37333} +{"lr": 1.0537023489176325e-05, "data_time": 0.23270184993743898, "loss": 0.08551195412874221, "time": 0.4768069744110107, "epoch": 10, "memory": 6318, "step": 37433} +{"lr": 1.0483129771605328e-05, "data_time": 0.3083606958389282, "loss": 0.08292020373046398, "time": 0.5312201499938964, "epoch": 10, "memory": 6318, "step": 37533} +{"lr": 1.0432070650458095e-05, "data_time": 0.36708760261535645, "loss": 0.08170890659093857, "time": 0.5082449913024902, "epoch": 10, "memory": 6318, "step": 37633} +{"lr": 1.0383849376906634e-05, "data_time": 0.31961779594421386, "loss": 0.08897578455507756, "time": 0.46034243106842043, "epoch": 10, "memory": 6318, "step": 37733} +{"lr": 1.0338469021423977e-05, "data_time": 0.3505417823791504, "loss": 0.08403019309043884, "time": 0.4918610334396362, "epoch": 10, "memory": 6318, "step": 37833} +{"lr": 1.029593247358868e-05, "data_time": 0.34041595458984375, "loss": 0.09566556289792061, "time": 0.480924129486084, "epoch": 10, "memory": 6318, "step": 37933} +{"lr": 1.0256242441900827e-05, "data_time": 0.3240325927734375, "loss": 0.09055039659142494, "time": 0.4656890869140625, "epoch": 10, "memory": 6318, "step": 38033} +{"lr": 1.0219401453609533e-05, "data_time": 0.34224896430969237, "loss": 0.09382965825498105, "time": 0.48375849723815917, "epoch": 10, "memory": 6318, "step": 38133} +{"lr": 1.018541185455209e-05, "data_time": 0.30359065532684326, "loss": 0.09477643221616745, "time": 0.4432615041732788, "epoch": 10, "memory": 6318, "step": 38233} +{"lr": 1.015427580900455e-05, "data_time": 0.3605957508087158, "loss": 0.08722190409898758, "time": 0.5014996290206909, "epoch": 10, "memory": 6318, "step": 38333} +{"lr": 1.0125995299543914e-05, "data_time": 0.3488884925842285, "loss": 0.08811897188425064, "time": 0.4892071962356567, "epoch": 10, "memory": 6318, "step": 38433} +{"lr": 1.0100572126921902e-05, "data_time": 0.34983506202697756, "loss": 0.1071148082613945, "time": 0.4902963638305664, "epoch": 10, "memory": 6318, "step": 38533} +{"lr": 1.007800790995031e-05, "data_time": 0.31435959339141845, "loss": 0.10802561715245247, "time": 0.4542102575302124, "epoch": 10, "memory": 6318, "step": 38633} +{"lr": 1.0058304085397905e-05, "data_time": 0.33975374698638916, "loss": 0.10101355090737343, "time": 0.481682276725769, "epoch": 10, "memory": 6318, "step": 38733} +{"lr": 1.0041461907898941e-05, "data_time": 0.37254621982574465, "loss": 0.10207248032093048, "time": 0.513184380531311, "epoch": 10, "memory": 6318, "step": 38833} +{"lr": 1.0027482449873297e-05, "data_time": 0.33594613075256347, "loss": 0.09972277358174324, "time": 0.47675719261169436, "epoch": 10, "memory": 6318, "step": 38933} +{"lr": 1.0016366601458162e-05, "data_time": 0.31034648418426514, "loss": 0.10015354640781879, "time": 0.4509953260421753, "epoch": 10, "memory": 6318, "step": 39033} +{"lr": 1.0008115070451358e-05, "data_time": 0.35239882469177247, "loss": 0.0880768671631813, "time": 0.49243271350860596, "epoch": 10, "memory": 6318, "step": 39133} +{"lr": 1.0002728382266284e-05, "data_time": 0.37269542217254636, "loss": 0.08653453812003135, "time": 0.5141926765441894, "epoch": 10, "memory": 6318, "step": 39233} +{"lr": 1.0000206879898465e-05, "data_time": 0.38600831031799315, "loss": 0.09437817335128784, "time": 0.5272750616073608, "epoch": 10, "memory": 6318, "step": 39333} +{"accuracy/top1": 96.48585510253906, "data_time": 0.3758496046066284, "time": 0.46335516304805363, "step": 10} diff --git a/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1/20230606_051539/vis_data/vis_image/0000000.png_1.png b/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1/20230606_051539/vis_data/vis_image/0000000.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..8c9d9d70d8c50caff84e819ba3e20c20126c2700 Binary files /dev/null and b/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1/20230606_051539/vis_data/vis_image/0000000.png_1.png differ diff --git a/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1/20230606_051539/vis_data/vis_image/0000000.png_10.png b/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1/20230606_051539/vis_data/vis_image/0000000.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..db32fe66286e22eb8ee96c5800175753ad974c99 Binary files /dev/null and b/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1/20230606_051539/vis_data/vis_image/0000000.png_10.png differ diff --git a/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1/20230606_051539/vis_data/vis_image/0000000.png_2.png b/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1/20230606_051539/vis_data/vis_image/0000000.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..c9f7678f1a062c709157775ac2ff8ecc43b7bd97 Binary files /dev/null and b/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1/20230606_051539/vis_data/vis_image/0000000.png_2.png differ diff --git a/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1/20230606_051539/vis_data/vis_image/0000000.png_3.png b/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1/20230606_051539/vis_data/vis_image/0000000.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..bb37abea093f08e237273988019c4e7bae14db9c Binary files /dev/null and b/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1/20230606_051539/vis_data/vis_image/0000000.png_3.png differ diff --git a/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1/20230606_051539/vis_data/vis_image/0000000.png_4.png b/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1/20230606_051539/vis_data/vis_image/0000000.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..1157a9714047959ff1238079d8a8e7ac997626a2 Binary files /dev/null and b/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1/20230606_051539/vis_data/vis_image/0000000.png_4.png differ diff --git a/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1/20230606_051539/vis_data/vis_image/0000000.png_5.png b/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1/20230606_051539/vis_data/vis_image/0000000.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..7bbc6eea1be6df3dbee61e66bef696ae2e501c74 Binary files /dev/null and b/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1/20230606_051539/vis_data/vis_image/0000000.png_5.png differ diff --git a/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1/20230606_051539/vis_data/vis_image/0000000.png_6.png b/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1/20230606_051539/vis_data/vis_image/0000000.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..6803afc81867f1d5093c352d42c4a8d66693f3d6 Binary files /dev/null and b/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1/20230606_051539/vis_data/vis_image/0000000.png_6.png differ diff --git a/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1/20230606_051539/vis_data/vis_image/0000000.png_7.png b/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1/20230606_051539/vis_data/vis_image/0000000.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..ec53badc6c83c3b0ce778b0b806f86a8b3420d60 Binary files /dev/null and b/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1/20230606_051539/vis_data/vis_image/0000000.png_7.png differ diff --git a/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1/20230606_051539/vis_data/vis_image/0000000.png_8.png b/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1/20230606_051539/vis_data/vis_image/0000000.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..67e5b7dd8d6ea667162ba8cfe07ec808723494fe Binary files /dev/null and b/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1/20230606_051539/vis_data/vis_image/0000000.png_8.png differ diff --git a/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1/20230606_051539/vis_data/vis_image/0000000.png_9.png b/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1/20230606_051539/vis_data/vis_image/0000000.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..1d6efd6f61d28f4d6a9b1d3d8daa55269d8e86e8 Binary files /dev/null and b/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1/20230606_051539/vis_data/vis_image/0000000.png_9.png differ diff --git a/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1/20230606_051539/vis_data/vis_image/000005174.jpg_1.png b/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1/20230606_051539/vis_data/vis_image/000005174.jpg_1.png new file mode 100644 index 0000000000000000000000000000000000000000..af8f95cba2d1fd117ec1e914e087936047b5ca33 Binary files /dev/null and b/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1/20230606_051539/vis_data/vis_image/000005174.jpg_1.png differ diff --git a/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1/20230606_051539/vis_data/vis_image/000005174.jpg_10.png b/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1/20230606_051539/vis_data/vis_image/000005174.jpg_10.png new file mode 100644 index 0000000000000000000000000000000000000000..93c76146ebd67bbbce74b28412a99b69e98247da Binary files /dev/null and b/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1/20230606_051539/vis_data/vis_image/000005174.jpg_10.png differ diff --git a/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1/20230606_051539/vis_data/vis_image/000005174.jpg_2.png b/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1/20230606_051539/vis_data/vis_image/000005174.jpg_2.png new file mode 100644 index 0000000000000000000000000000000000000000..9f7eccf96dcf96a2f8b8583f9d85ae23fde7c6c3 Binary files /dev/null and b/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1/20230606_051539/vis_data/vis_image/000005174.jpg_2.png differ diff --git a/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1/20230606_051539/vis_data/vis_image/000005174.jpg_3.png b/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1/20230606_051539/vis_data/vis_image/000005174.jpg_3.png new file mode 100644 index 0000000000000000000000000000000000000000..45192e973336bb045084c5889f49ea69c10ebb0c Binary files /dev/null and b/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1/20230606_051539/vis_data/vis_image/000005174.jpg_3.png differ diff --git a/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1/20230606_051539/vis_data/vis_image/000005174.jpg_4.png b/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1/20230606_051539/vis_data/vis_image/000005174.jpg_4.png new file mode 100644 index 0000000000000000000000000000000000000000..794b851c56d042a20ad143c196f3e0814f193a40 Binary files /dev/null and b/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1/20230606_051539/vis_data/vis_image/000005174.jpg_4.png differ diff --git a/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1/20230606_051539/vis_data/vis_image/000005174.jpg_5.png b/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1/20230606_051539/vis_data/vis_image/000005174.jpg_5.png new file mode 100644 index 0000000000000000000000000000000000000000..91df9f4b2b4b8eceb2f3debb4bbe3152107779d8 Binary files /dev/null and b/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1/20230606_051539/vis_data/vis_image/000005174.jpg_5.png differ diff --git a/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1/20230606_051539/vis_data/vis_image/000005174.jpg_6.png b/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1/20230606_051539/vis_data/vis_image/000005174.jpg_6.png new file mode 100644 index 0000000000000000000000000000000000000000..ed7b489951a9062ff123d62fd1d6b6116db4a099 Binary files /dev/null and b/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1/20230606_051539/vis_data/vis_image/000005174.jpg_6.png differ diff --git a/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1/20230606_051539/vis_data/vis_image/000005174.jpg_7.png b/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1/20230606_051539/vis_data/vis_image/000005174.jpg_7.png new file mode 100644 index 0000000000000000000000000000000000000000..d9902fa217c44684208ab5839bceec8d271f6eea Binary files /dev/null and b/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1/20230606_051539/vis_data/vis_image/000005174.jpg_7.png differ diff --git a/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1/20230606_051539/vis_data/vis_image/000005174.jpg_8.png b/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1/20230606_051539/vis_data/vis_image/000005174.jpg_8.png new file mode 100644 index 0000000000000000000000000000000000000000..357f49a0e8003fe07d1a15a6027574b7d6159cc9 Binary files /dev/null and b/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1/20230606_051539/vis_data/vis_image/000005174.jpg_8.png differ diff --git a/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1/20230606_051539/vis_data/vis_image/000005174.jpg_9.png b/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1/20230606_051539/vis_data/vis_image/000005174.jpg_9.png new file mode 100644 index 0000000000000000000000000000000000000000..2521e63dc765b5a9412245f6816459d4e266eac1 Binary files /dev/null and b/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1/20230606_051539/vis_data/vis_image/000005174.jpg_9.png differ diff --git a/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1/20230606_051539/vis_data/vis_image/0010000.png_1.png b/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1/20230606_051539/vis_data/vis_image/0010000.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..ac28240fa1e2009bf84c0436a2e38cc9b0d47816 Binary files /dev/null and b/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1/20230606_051539/vis_data/vis_image/0010000.png_1.png differ diff --git a/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1/20230606_051539/vis_data/vis_image/0010000.png_10.png b/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1/20230606_051539/vis_data/vis_image/0010000.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..e4784ff3123efef0d77c93848e1edc00527db0a1 Binary files /dev/null and b/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1/20230606_051539/vis_data/vis_image/0010000.png_10.png differ diff --git a/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1/20230606_051539/vis_data/vis_image/0010000.png_2.png b/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1/20230606_051539/vis_data/vis_image/0010000.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..0dba34625079da0bf5f081d63e500231c006d3be Binary files /dev/null and b/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1/20230606_051539/vis_data/vis_image/0010000.png_2.png differ diff --git a/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1/20230606_051539/vis_data/vis_image/0010000.png_3.png b/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1/20230606_051539/vis_data/vis_image/0010000.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..151389467befed09e293bba6931c5f6ca67f07c6 Binary files /dev/null and b/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1/20230606_051539/vis_data/vis_image/0010000.png_3.png differ diff --git a/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1/20230606_051539/vis_data/vis_image/0010000.png_4.png b/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1/20230606_051539/vis_data/vis_image/0010000.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..c6d9cdfab94a8d545da03cf4273fb6b0a51dc40e Binary files /dev/null and b/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1/20230606_051539/vis_data/vis_image/0010000.png_4.png differ diff --git a/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1/20230606_051539/vis_data/vis_image/0010000.png_5.png b/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1/20230606_051539/vis_data/vis_image/0010000.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..d31776537450b8d147a8bb1f10f5356c386ce043 Binary files /dev/null and b/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1/20230606_051539/vis_data/vis_image/0010000.png_5.png differ diff --git a/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1/20230606_051539/vis_data/vis_image/0010000.png_6.png b/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1/20230606_051539/vis_data/vis_image/0010000.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..13d693329d469adeaff2bcb43f8afc9f1a2c72e7 Binary files /dev/null and b/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1/20230606_051539/vis_data/vis_image/0010000.png_6.png differ diff --git a/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1/20230606_051539/vis_data/vis_image/0010000.png_7.png b/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1/20230606_051539/vis_data/vis_image/0010000.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..c3287af1f0575ee8754037639722bf0803abbdaf Binary files /dev/null and b/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1/20230606_051539/vis_data/vis_image/0010000.png_7.png differ diff --git a/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1/20230606_051539/vis_data/vis_image/0010000.png_8.png b/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1/20230606_051539/vis_data/vis_image/0010000.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..ed70829db8dd2fbec5cb208aac459a3183428a69 Binary files /dev/null and b/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1/20230606_051539/vis_data/vis_image/0010000.png_8.png differ diff --git a/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1/20230606_051539/vis_data/vis_image/0010000.png_9.png b/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1/20230606_051539/vis_data/vis_image/0010000.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..fce89ec1f391544798e4200e91096398910bf002 Binary files /dev/null and b/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1/20230606_051539/vis_data/vis_image/0010000.png_9.png differ diff --git a/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1/epoch_1.pth b/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1/epoch_1.pth new file mode 100644 index 0000000000000000000000000000000000000000..50d69c5d944e875b43a2b12b275ab71da663ba91 --- /dev/null +++ b/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1/epoch_1.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:6a705af71606f3679ac68ede0bf106d4d6046ef566f42f81430933976616fba2 +size 277335181 diff --git a/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1/epoch_10.pth b/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1/epoch_10.pth new file mode 100644 index 0000000000000000000000000000000000000000..b88f288da1d058df73e3cfc11e5aeca1c772db76 --- /dev/null +++ b/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1/epoch_10.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:cb216c0b8ec37e0a69c46e3c0d91ce6a1cb1776bc59b0b03359f4a036fc0a222 +size 279934477 diff --git a/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1/epoch_2.pth b/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1/epoch_2.pth new file mode 100644 index 0000000000000000000000000000000000000000..c38467ec413acff81c59e86f9289ed72b050ad26 --- /dev/null +++ b/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1/epoch_2.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c3e9c216cc3c396a06bbff8b583c83ab9ad0d60401999e609b75774629da9e39 +size 277622733 diff --git a/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1/epoch_3.pth b/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1/epoch_3.pth new file mode 100644 index 0000000000000000000000000000000000000000..0159fe72ae79a55b2a57d3ec566a16b4b1e12613 --- /dev/null +++ b/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1/epoch_3.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:bac3148a42298c91d70a496fe6a8b4ee1325a5cd1abe7716e75b9e8b7593dc26 +size 277910605 diff --git a/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1/epoch_4.pth b/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1/epoch_4.pth new file mode 100644 index 0000000000000000000000000000000000000000..1e3db93a6a7207364e95abf3f6f8643bce372eb1 --- /dev/null +++ b/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1/epoch_4.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:15b7102b66da4ed167d83f909aa56b5b9a1d29c84bc34828f3c2f5c5153a9dfc +size 278197453 diff --git a/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1/epoch_5.pth b/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1/epoch_5.pth new file mode 100644 index 0000000000000000000000000000000000000000..c8d03ab5c21c887fa382d1f9e5ea5ebd898eb8f9 --- /dev/null +++ b/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1/epoch_5.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:0f7a8e87fbb7b07a063b6c667aec15bc2da1a5036e2afe662061431c34aa82fd +size 278485453 diff --git a/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1/epoch_6.pth b/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1/epoch_6.pth new file mode 100644 index 0000000000000000000000000000000000000000..01cfac5ec2ebbaa23c3b526646e119996d30c83a --- /dev/null +++ b/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1/epoch_6.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:a60728ade303c3394853d097749e59489331e8277987066d510b8b6643597d66 +size 278773133 diff --git a/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1/epoch_7.pth b/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1/epoch_7.pth new file mode 100644 index 0000000000000000000000000000000000000000..9ef76da53e9a81cc2a48bffaad1115352d191eb9 --- /dev/null +++ b/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1/epoch_7.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:a992bf5d32c582f993cccf75d0efeff172593380a066218cd150b025f1fd7c89 +size 279061261 diff --git a/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1/epoch_8.pth b/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1/epoch_8.pth new file mode 100644 index 0000000000000000000000000000000000000000..9ec72d077b4dc70a6c5c948e270adf249cdbc2db --- /dev/null +++ b/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1/epoch_8.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:2bea03c69caac1f435045a99be2bb68a2b418678c5207083c7fe8352aa267772 +size 279353293 diff --git a/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1/epoch_9.pth b/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1/epoch_9.pth new file mode 100644 index 0000000000000000000000000000000000000000..d207346b344445f007ceb4535f778bc0d5d368b9 --- /dev/null +++ b/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1/epoch_9.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:50b7245d61d927b23977e345abb12dfc024b7a838711cc56835a66424c3ef59f +size 279643853 diff --git a/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1/last_checkpoint b/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1/last_checkpoint new file mode 100644 index 0000000000000000000000000000000000000000..ab0deaccf65a1fda459994ec25de03ad25b81663 --- /dev/null +++ b/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1/last_checkpoint @@ -0,0 +1 @@ +/mnt/petrelfs/luzeyu/workspace/fakebench/mmpretrain/workdir/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1/epoch_10.pth \ No newline at end of file diff --git a/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1.py b/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1.py new file mode 100644 index 0000000000000000000000000000000000000000..9fa82a87719e4caa4a661d2572ccc1b988bdb43d --- /dev/null +++ b/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1.py @@ -0,0 +1,247 @@ +optim_wrapper = dict( + optimizer=dict( + type='SGD', + lr=0.0001, + momentum=0.9, + weight_decay=0.0001, + _scope_='mmpretrain'), + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=False, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=512) +model = dict( + type='ImageClassifier', + backbone=dict( + frozen_stages=2, + type='ResNet', + depth=50, + num_stages=4, + out_indices=(3, ), + style='pytorch', + init_cfg=dict( + type='Pretrained', + checkpoint= + 'https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth', + prefix='backbone')), + neck=dict(type='GlobalAveragePooling'), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=2048, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + topk=1)) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/IF80w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/if-dpmsolver++-50-20w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/cc1m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/if-dpmsolver++-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = dict(type='Accuracy', topk=1) +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/if-dpmsolver++-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = dict(type='Accuracy', topk=1) +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/resnet50_2xb256_IF_1m_lr1e-4_aug_1e-1' diff --git a/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1/20230606_050038/20230606_050038.log b/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1/20230606_050038/20230606_050038.log new file mode 100644 index 0000000000000000000000000000000000000000..1d659eff5adc1456b9f59a58ecafe5919186f76b --- /dev/null +++ b/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1/20230606_050038/20230606_050038.log @@ -0,0 +1,863 @@ +2023/06/06 05:00:46 - mmengine - INFO - +------------------------------------------------------------ +System environment: + sys.platform: linux + Python: 3.10.9 (main, Mar 8 2023, 10:47:38) [GCC 11.2.0] + CUDA available: True + numpy_random_seed: 2141664782 + GPU 0,1: NVIDIA A100-SXM4-80GB + CUDA_HOME: /mnt/petrelfs/share/cuda-11.6 + NVCC: Cuda compilation tools, release 11.6, V11.6.124 + GCC: gcc (GCC) 7.5.0 + PyTorch: 1.13.1 + PyTorch compiling details: PyTorch built with: + - GCC 9.3 + - C++ Version: 201402 + - Intel(R) oneAPI Math Kernel Library Version 2021.4-Product Build 20210904 for Intel(R) 64 architecture applications + - Intel(R) MKL-DNN v2.6.0 (Git Hash 52b5f107dd9cf10910aaa19cb47f3abf9b349815) + - OpenMP 201511 (a.k.a. OpenMP 4.5) + - LAPACK is enabled (usually provided by MKL) + - NNPACK is enabled + - CPU capability usage: AVX2 + - CUDA Runtime 11.6 + - NVCC architecture flags: -gencode;arch=compute_37,code=sm_37;-gencode;arch=compute_50,code=sm_50;-gencode;arch=compute_60,code=sm_60;-gencode;arch=compute_61,code=sm_61;-gencode;arch=compute_70,code=sm_70;-gencode;arch=compute_75,code=sm_75;-gencode;arch=compute_80,code=sm_80;-gencode;arch=compute_86,code=sm_86;-gencode;arch=compute_37,code=compute_37 + - CuDNN 8.3.2 (built against CUDA 11.5) + - Magma 2.6.1 + - Build settings: BLAS_INFO=mkl, BUILD_TYPE=Release, CUDA_VERSION=11.6, CUDNN_VERSION=8.3.2, CXX_COMPILER=/opt/rh/devtoolset-9/root/usr/bin/c++, CXX_FLAGS= -fabi-version=11 -Wno-deprecated -fvisibility-inlines-hidden -DUSE_PTHREADPOOL -fopenmp -DNDEBUG -DUSE_KINETO -DUSE_FBGEMM -DUSE_QNNPACK -DUSE_PYTORCH_QNNPACK -DUSE_XNNPACK -DSYMBOLICATE_MOBILE_DEBUG_HANDLE -DEDGE_PROFILER_USE_KINETO -O2 -fPIC -Wno-narrowing -Wall -Wextra -Werror=return-type -Werror=non-virtual-dtor -Wno-missing-field-initializers -Wno-type-limits -Wno-array-bounds -Wno-unknown-pragmas -Wunused-local-typedefs -Wno-unused-parameter -Wno-unused-function -Wno-unused-result -Wno-strict-overflow -Wno-strict-aliasing -Wno-error=deprecated-declarations -Wno-stringop-overflow -Wno-psabi -Wno-error=pedantic -Wno-error=redundant-decls -Wno-error=old-style-cast -fdiagnostics-color=always -faligned-new -Wno-unused-but-set-variable -Wno-maybe-uninitialized -fno-math-errno -fno-trapping-math -Werror=format -Werror=cast-function-type -Wno-stringop-overflow, LAPACK_INFO=mkl, PERF_WITH_AVX=1, PERF_WITH_AVX2=1, PERF_WITH_AVX512=1, TORCH_VERSION=1.13.1, USE_CUDA=ON, USE_CUDNN=ON, USE_EXCEPTION_PTR=1, USE_GFLAGS=OFF, USE_GLOG=OFF, USE_MKL=ON, USE_MKLDNN=ON, USE_MPI=OFF, USE_NCCL=ON, USE_NNPACK=ON, USE_OPENMP=ON, USE_ROCM=OFF, + + TorchVision: 0.14.1 + OpenCV: 4.7.0 + MMEngine: 0.7.3 + +Runtime environment: + cudnn_benchmark: True + mp_cfg: {'mp_start_method': 'fork', 'opencv_num_threads': 0} + dist_cfg: {'backend': 'nccl'} + seed: None + deterministic: False + Distributed launcher: slurm + Distributed training: True + GPU number: 2 +------------------------------------------------------------ + +2023/06/06 05:00:50 - mmengine - INFO - Config: +optim_wrapper = dict( + optimizer=dict( + type='SGD', + lr=0.0001, + momentum=0.9, + weight_decay=0.0001, + _scope_='mmpretrain'), + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=False, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=512) +model = dict( + type='ImageClassifier', + backbone=dict( + frozen_stages=2, + type='ResNet', + depth=50, + num_stages=4, + out_indices=(3, ), + style='pytorch', + init_cfg=dict( + type='Pretrained', + checkpoint= + 'https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth', + prefix='backbone')), + neck=dict(type='GlobalAveragePooling'), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=2048, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + topk=1)) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/IF80w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/if-dpmsolver++-50-20w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/cc1m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/if-dpmsolver++-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = dict(type='Accuracy', topk=1) +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/if-dpmsolver++-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = dict(type='Accuracy', topk=1) +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1' + +2023/06/06 05:01:01 - mmengine - INFO - Hooks will be executed in the following order: +before_run: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_load_checkpoint: +(ABOVE_NORMAL) EMAHook + -------------------- +before_train: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_train_epoch: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(NORMAL ) DistSamplerSeedHook + -------------------- +before_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook + -------------------- +after_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +after_train_epoch: +(NORMAL ) IterTimerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_val_epoch: +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook + -------------------- +before_val_iter: +(NORMAL ) IterTimerHook + -------------------- +after_val_iter: +(NORMAL ) IterTimerHook +(NORMAL ) VisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_val_epoch: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_save_checkpoint: +(ABOVE_NORMAL) EMAHook + -------------------- +after_train: +(VERY_LOW ) CheckpointHook + -------------------- +before_test_epoch: +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook + -------------------- +before_test_iter: +(NORMAL ) IterTimerHook + -------------------- +after_test_iter: +(NORMAL ) IterTimerHook +(NORMAL ) VisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_test_epoch: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_run: +(BELOW_NORMAL) LoggerHook + -------------------- +2023/06/06 05:01:18 - mmengine - INFO - load backbone in model from: https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth +Name of parameter - Initialization information + +backbone.conv1.weight - torch.Size([64, 3, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.bn1.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.bn1.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.conv1.weight - torch.Size([64, 64, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.bn1.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.bn1.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.conv2.weight - torch.Size([64, 64, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.bn2.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.bn2.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.conv3.weight - torch.Size([256, 64, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.bn3.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.bn3.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.downsample.0.weight - torch.Size([256, 64, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.downsample.1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.downsample.1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.conv1.weight - torch.Size([64, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.bn1.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.bn1.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.conv2.weight - torch.Size([64, 64, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.bn2.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.bn2.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.conv3.weight - torch.Size([256, 64, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.bn3.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.bn3.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.conv1.weight - torch.Size([64, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.bn1.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.bn1.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.conv2.weight - torch.Size([64, 64, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.bn2.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.bn2.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.conv3.weight - torch.Size([256, 64, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.bn3.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.bn3.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.conv1.weight - torch.Size([128, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.bn1.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.bn1.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.conv2.weight - torch.Size([128, 128, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.bn2.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.bn2.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.conv3.weight - torch.Size([512, 128, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.bn3.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.bn3.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.downsample.0.weight - torch.Size([512, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.downsample.1.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.downsample.1.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.conv1.weight - torch.Size([128, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.bn1.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.bn1.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.conv2.weight - torch.Size([128, 128, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.bn2.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.bn2.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.conv3.weight - torch.Size([512, 128, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.bn3.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.bn3.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.conv1.weight - torch.Size([128, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.bn1.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.bn1.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.conv2.weight - torch.Size([128, 128, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.bn2.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.bn2.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.conv3.weight - torch.Size([512, 128, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.bn3.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.bn3.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.conv1.weight - torch.Size([128, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.bn1.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.bn1.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.conv2.weight - torch.Size([128, 128, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.bn2.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.bn2.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.conv3.weight - torch.Size([512, 128, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.bn3.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.bn3.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.conv1.weight - torch.Size([256, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.bn1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.bn1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.conv2.weight - torch.Size([256, 256, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.bn2.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.bn2.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.conv3.weight - torch.Size([1024, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.bn3.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.bn3.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.downsample.0.weight - torch.Size([1024, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.downsample.1.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.downsample.1.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.conv1.weight - torch.Size([256, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.bn1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.bn1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.conv2.weight - torch.Size([256, 256, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.bn2.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.bn2.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.conv3.weight - torch.Size([1024, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.bn3.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.bn3.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.conv1.weight - torch.Size([256, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.bn1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.bn1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.conv2.weight - torch.Size([256, 256, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.bn2.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.bn2.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.conv3.weight - torch.Size([1024, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.bn3.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.bn3.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.conv1.weight - torch.Size([256, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.bn1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.bn1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.conv2.weight - torch.Size([256, 256, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.bn2.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.bn2.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.conv3.weight - torch.Size([1024, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.bn3.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.bn3.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.conv1.weight - torch.Size([256, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.bn1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.bn1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.conv2.weight - torch.Size([256, 256, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.bn2.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.bn2.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.conv3.weight - torch.Size([1024, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.bn3.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.bn3.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.conv1.weight - torch.Size([256, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.bn1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.bn1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.conv2.weight - torch.Size([256, 256, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.bn2.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.bn2.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.conv3.weight - torch.Size([1024, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.bn3.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.bn3.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.conv1.weight - torch.Size([512, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.bn1.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.bn1.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.conv2.weight - torch.Size([512, 512, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.bn2.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.bn2.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.conv3.weight - torch.Size([2048, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.bn3.weight - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.bn3.bias - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.downsample.0.weight - torch.Size([2048, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.downsample.1.weight - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.downsample.1.bias - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.conv1.weight - torch.Size([512, 2048, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.bn1.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.bn1.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.conv2.weight - torch.Size([512, 512, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.bn2.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.bn2.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.conv3.weight - torch.Size([2048, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.bn3.weight - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.bn3.bias - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.conv1.weight - torch.Size([512, 2048, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.bn1.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.bn1.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.conv2.weight - torch.Size([512, 512, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.bn2.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.bn2.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.conv3.weight - torch.Size([2048, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.bn3.weight - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.bn3.bias - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +head.fc.weight - torch.Size([2, 2048]): +NormalInit: mean=0, std=0.01, bias=0 + +head.fc.bias - torch.Size([2]): +NormalInit: mean=0, std=0.01, bias=0 +2023/06/06 05:01:19 - mmengine - WARNING - "FileClient" will be deprecated in future. Please use io functions in https://mmengine.readthedocs.io/en/latest/api/fileio.html#file-io +2023/06/06 05:01:19 - mmengine - WARNING - "HardDiskBackend" is the alias of "LocalBackend" and the former will be deprecated in future. +2023/06/06 05:01:19 - mmengine - INFO - Checkpoints will be saved to /mnt/petrelfs/luzeyu/workspace/fakebench/mmpretrain/workdir/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1. diff --git a/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1/20230606_050038/vis_data/config.py b/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1/20230606_050038/vis_data/config.py new file mode 100644 index 0000000000000000000000000000000000000000..07f49b3c87537f52b4f76320195407e52d9fcb36 --- /dev/null +++ b/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1/20230606_050038/vis_data/config.py @@ -0,0 +1,247 @@ +optim_wrapper = dict( + optimizer=dict( + type='SGD', + lr=0.0001, + momentum=0.9, + weight_decay=0.0001, + _scope_='mmpretrain'), + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=False, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=512) +model = dict( + type='ImageClassifier', + backbone=dict( + frozen_stages=2, + type='ResNet', + depth=50, + num_stages=4, + out_indices=(3, ), + style='pytorch', + init_cfg=dict( + type='Pretrained', + checkpoint= + 'https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth', + prefix='backbone')), + neck=dict(type='GlobalAveragePooling'), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=2048, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + topk=1)) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/IF80w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/if-dpmsolver++-50-20w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/cc1m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/if-dpmsolver++-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = dict(type='Accuracy', topk=1) +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/if-dpmsolver++-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = dict(type='Accuracy', topk=1) +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1' diff --git a/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1/20230606_050038/vis_data/events.out.tfevents.1685998856.SH-IDC1-10-140-24-124.191224.0 b/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1/20230606_050038/vis_data/events.out.tfevents.1685998856.SH-IDC1-10-140-24-124.191224.0 new file mode 100644 index 0000000000000000000000000000000000000000..b082b7cce12db4fa75697eae3adf72db5dafcb56 --- /dev/null +++ b/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1/20230606_050038/vis_data/events.out.tfevents.1685998856.SH-IDC1-10-140-24-124.191224.0 @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:0e2cba458c21e9290bc0d54acccad9a5e1cc032d0da3afe426ce3b15aaca2cac +size 9590 diff --git a/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1/20230606_050638/20230606_050638.log b/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1/20230606_050638/20230606_050638.log new file mode 100644 index 0000000000000000000000000000000000000000..e2f514cd49e9689ea11a19a29ce8d58df913126f --- /dev/null +++ b/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1/20230606_050638/20230606_050638.log @@ -0,0 +1,863 @@ +2023/06/06 05:06:43 - mmengine - INFO - +------------------------------------------------------------ +System environment: + sys.platform: linux + Python: 3.10.9 (main, Mar 8 2023, 10:47:38) [GCC 11.2.0] + CUDA available: True + numpy_random_seed: 1798664996 + GPU 0,1: NVIDIA A100-SXM4-80GB + CUDA_HOME: /mnt/petrelfs/share/cuda-11.6 + NVCC: Cuda compilation tools, release 11.6, V11.6.124 + GCC: gcc (GCC) 7.5.0 + PyTorch: 1.13.1 + PyTorch compiling details: PyTorch built with: + - GCC 9.3 + - C++ Version: 201402 + - Intel(R) oneAPI Math Kernel Library Version 2021.4-Product Build 20210904 for Intel(R) 64 architecture applications + - Intel(R) MKL-DNN v2.6.0 (Git Hash 52b5f107dd9cf10910aaa19cb47f3abf9b349815) + - OpenMP 201511 (a.k.a. OpenMP 4.5) + - LAPACK is enabled (usually provided by MKL) + - NNPACK is enabled + - CPU capability usage: AVX2 + - CUDA Runtime 11.6 + - NVCC architecture flags: -gencode;arch=compute_37,code=sm_37;-gencode;arch=compute_50,code=sm_50;-gencode;arch=compute_60,code=sm_60;-gencode;arch=compute_61,code=sm_61;-gencode;arch=compute_70,code=sm_70;-gencode;arch=compute_75,code=sm_75;-gencode;arch=compute_80,code=sm_80;-gencode;arch=compute_86,code=sm_86;-gencode;arch=compute_37,code=compute_37 + - CuDNN 8.3.2 (built against CUDA 11.5) + - Magma 2.6.1 + - Build settings: BLAS_INFO=mkl, BUILD_TYPE=Release, CUDA_VERSION=11.6, CUDNN_VERSION=8.3.2, CXX_COMPILER=/opt/rh/devtoolset-9/root/usr/bin/c++, CXX_FLAGS= -fabi-version=11 -Wno-deprecated -fvisibility-inlines-hidden -DUSE_PTHREADPOOL -fopenmp -DNDEBUG -DUSE_KINETO -DUSE_FBGEMM -DUSE_QNNPACK -DUSE_PYTORCH_QNNPACK -DUSE_XNNPACK -DSYMBOLICATE_MOBILE_DEBUG_HANDLE -DEDGE_PROFILER_USE_KINETO -O2 -fPIC -Wno-narrowing -Wall -Wextra -Werror=return-type -Werror=non-virtual-dtor -Wno-missing-field-initializers -Wno-type-limits -Wno-array-bounds -Wno-unknown-pragmas -Wunused-local-typedefs -Wno-unused-parameter -Wno-unused-function -Wno-unused-result -Wno-strict-overflow -Wno-strict-aliasing -Wno-error=deprecated-declarations -Wno-stringop-overflow -Wno-psabi -Wno-error=pedantic -Wno-error=redundant-decls -Wno-error=old-style-cast -fdiagnostics-color=always -faligned-new -Wno-unused-but-set-variable -Wno-maybe-uninitialized -fno-math-errno -fno-trapping-math -Werror=format -Werror=cast-function-type -Wno-stringop-overflow, LAPACK_INFO=mkl, PERF_WITH_AVX=1, PERF_WITH_AVX2=1, PERF_WITH_AVX512=1, TORCH_VERSION=1.13.1, USE_CUDA=ON, USE_CUDNN=ON, USE_EXCEPTION_PTR=1, USE_GFLAGS=OFF, USE_GLOG=OFF, USE_MKL=ON, USE_MKLDNN=ON, USE_MPI=OFF, USE_NCCL=ON, USE_NNPACK=ON, USE_OPENMP=ON, USE_ROCM=OFF, + + TorchVision: 0.14.1 + OpenCV: 4.7.0 + MMEngine: 0.7.3 + +Runtime environment: + cudnn_benchmark: True + mp_cfg: {'mp_start_method': 'fork', 'opencv_num_threads': 0} + dist_cfg: {'backend': 'nccl'} + seed: None + deterministic: False + Distributed launcher: slurm + Distributed training: True + GPU number: 2 +------------------------------------------------------------ + +2023/06/06 05:06:47 - mmengine - INFO - Config: +optim_wrapper = dict( + optimizer=dict( + type='SGD', + lr=0.0001, + momentum=0.9, + weight_decay=0.0001, + _scope_='mmpretrain'), + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=False, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=512) +model = dict( + type='ImageClassifier', + backbone=dict( + frozen_stages=2, + type='ResNet', + depth=50, + num_stages=4, + out_indices=(3, ), + style='pytorch', + init_cfg=dict( + type='Pretrained', + checkpoint= + 'https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth', + prefix='backbone')), + neck=dict(type='GlobalAveragePooling'), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=2048, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + topk=1)) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/IF80w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/if-dpmsolver++-50-20w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/cc1m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/if-dpmsolver++-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = dict(type='Accuracy', topk=1) +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/if-dpmsolver++-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = dict(type='Accuracy', topk=1) +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1' + +2023/06/06 05:06:57 - mmengine - INFO - Hooks will be executed in the following order: +before_run: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_load_checkpoint: +(ABOVE_NORMAL) EMAHook + -------------------- +before_train: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_train_epoch: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(NORMAL ) DistSamplerSeedHook + -------------------- +before_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook + -------------------- +after_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +after_train_epoch: +(NORMAL ) IterTimerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_val_epoch: +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook + -------------------- +before_val_iter: +(NORMAL ) IterTimerHook + -------------------- +after_val_iter: +(NORMAL ) IterTimerHook +(NORMAL ) VisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_val_epoch: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_save_checkpoint: +(ABOVE_NORMAL) EMAHook + -------------------- +after_train: +(VERY_LOW ) CheckpointHook + -------------------- +before_test_epoch: +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook + -------------------- +before_test_iter: +(NORMAL ) IterTimerHook + -------------------- +after_test_iter: +(NORMAL ) IterTimerHook +(NORMAL ) VisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_test_epoch: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_run: +(BELOW_NORMAL) LoggerHook + -------------------- +2023/06/06 05:07:15 - mmengine - INFO - load backbone in model from: https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth +Name of parameter - Initialization information + +backbone.conv1.weight - torch.Size([64, 3, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.bn1.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.bn1.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.conv1.weight - torch.Size([64, 64, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.bn1.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.bn1.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.conv2.weight - torch.Size([64, 64, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.bn2.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.bn2.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.conv3.weight - torch.Size([256, 64, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.bn3.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.bn3.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.downsample.0.weight - torch.Size([256, 64, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.downsample.1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.downsample.1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.conv1.weight - torch.Size([64, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.bn1.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.bn1.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.conv2.weight - torch.Size([64, 64, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.bn2.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.bn2.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.conv3.weight - torch.Size([256, 64, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.bn3.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.bn3.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.conv1.weight - torch.Size([64, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.bn1.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.bn1.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.conv2.weight - torch.Size([64, 64, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.bn2.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.bn2.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.conv3.weight - torch.Size([256, 64, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.bn3.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.bn3.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.conv1.weight - torch.Size([128, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.bn1.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.bn1.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.conv2.weight - torch.Size([128, 128, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.bn2.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.bn2.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.conv3.weight - torch.Size([512, 128, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.bn3.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.bn3.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.downsample.0.weight - torch.Size([512, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.downsample.1.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.downsample.1.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.conv1.weight - torch.Size([128, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.bn1.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.bn1.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.conv2.weight - torch.Size([128, 128, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.bn2.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.bn2.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.conv3.weight - torch.Size([512, 128, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.bn3.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.bn3.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.conv1.weight - torch.Size([128, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.bn1.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.bn1.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.conv2.weight - torch.Size([128, 128, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.bn2.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.bn2.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.conv3.weight - torch.Size([512, 128, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.bn3.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.bn3.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.conv1.weight - torch.Size([128, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.bn1.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.bn1.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.conv2.weight - torch.Size([128, 128, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.bn2.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.bn2.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.conv3.weight - torch.Size([512, 128, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.bn3.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.bn3.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.conv1.weight - torch.Size([256, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.bn1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.bn1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.conv2.weight - torch.Size([256, 256, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.bn2.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.bn2.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.conv3.weight - torch.Size([1024, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.bn3.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.bn3.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.downsample.0.weight - torch.Size([1024, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.downsample.1.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.downsample.1.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.conv1.weight - torch.Size([256, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.bn1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.bn1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.conv2.weight - torch.Size([256, 256, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.bn2.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.bn2.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.conv3.weight - torch.Size([1024, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.bn3.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.bn3.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.conv1.weight - torch.Size([256, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.bn1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.bn1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.conv2.weight - torch.Size([256, 256, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.bn2.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.bn2.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.conv3.weight - torch.Size([1024, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.bn3.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.bn3.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.conv1.weight - torch.Size([256, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.bn1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.bn1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.conv2.weight - torch.Size([256, 256, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.bn2.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.bn2.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.conv3.weight - torch.Size([1024, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.bn3.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.bn3.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.conv1.weight - torch.Size([256, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.bn1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.bn1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.conv2.weight - torch.Size([256, 256, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.bn2.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.bn2.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.conv3.weight - torch.Size([1024, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.bn3.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.bn3.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.conv1.weight - torch.Size([256, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.bn1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.bn1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.conv2.weight - torch.Size([256, 256, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.bn2.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.bn2.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.conv3.weight - torch.Size([1024, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.bn3.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.bn3.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.conv1.weight - torch.Size([512, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.bn1.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.bn1.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.conv2.weight - torch.Size([512, 512, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.bn2.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.bn2.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.conv3.weight - torch.Size([2048, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.bn3.weight - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.bn3.bias - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.downsample.0.weight - torch.Size([2048, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.downsample.1.weight - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.downsample.1.bias - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.conv1.weight - torch.Size([512, 2048, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.bn1.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.bn1.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.conv2.weight - torch.Size([512, 512, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.bn2.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.bn2.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.conv3.weight - torch.Size([2048, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.bn3.weight - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.bn3.bias - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.conv1.weight - torch.Size([512, 2048, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.bn1.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.bn1.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.conv2.weight - torch.Size([512, 512, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.bn2.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.bn2.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.conv3.weight - torch.Size([2048, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.bn3.weight - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.bn3.bias - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +head.fc.weight - torch.Size([2, 2048]): +NormalInit: mean=0, std=0.01, bias=0 + +head.fc.bias - torch.Size([2]): +NormalInit: mean=0, std=0.01, bias=0 +2023/06/06 05:07:15 - mmengine - WARNING - "FileClient" will be deprecated in future. Please use io functions in https://mmengine.readthedocs.io/en/latest/api/fileio.html#file-io +2023/06/06 05:07:15 - mmengine - WARNING - "HardDiskBackend" is the alias of "LocalBackend" and the former will be deprecated in future. +2023/06/06 05:07:15 - mmengine - INFO - Checkpoints will be saved to /mnt/petrelfs/luzeyu/workspace/fakebench/mmpretrain/workdir/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1. diff --git a/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1/20230606_050638/vis_data/config.py b/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1/20230606_050638/vis_data/config.py new file mode 100644 index 0000000000000000000000000000000000000000..07f49b3c87537f52b4f76320195407e52d9fcb36 --- /dev/null +++ b/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1/20230606_050638/vis_data/config.py @@ -0,0 +1,247 @@ +optim_wrapper = dict( + optimizer=dict( + type='SGD', + lr=0.0001, + momentum=0.9, + weight_decay=0.0001, + _scope_='mmpretrain'), + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=False, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=512) +model = dict( + type='ImageClassifier', + backbone=dict( + frozen_stages=2, + type='ResNet', + depth=50, + num_stages=4, + out_indices=(3, ), + style='pytorch', + init_cfg=dict( + type='Pretrained', + checkpoint= + 'https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth', + prefix='backbone')), + neck=dict(type='GlobalAveragePooling'), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=2048, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + topk=1)) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/IF80w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/if-dpmsolver++-50-20w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/cc1m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/if-dpmsolver++-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = dict(type='Accuracy', topk=1) +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/if-dpmsolver++-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = dict(type='Accuracy', topk=1) +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1' diff --git a/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1/20230606_050638/vis_data/events.out.tfevents.1685999213.SH-IDC1-10-140-24-124.212405.0 b/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1/20230606_050638/vis_data/events.out.tfevents.1685999213.SH-IDC1-10-140-24-124.212405.0 new file mode 100644 index 0000000000000000000000000000000000000000..a9db47411faa8ea659d0373ef397c4a83afa90cd --- /dev/null +++ b/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1/20230606_050638/vis_data/events.out.tfevents.1685999213.SH-IDC1-10-140-24-124.212405.0 @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:4d1395ffc90e27a0ec54f880c42b169a24b76d29d8d0fe87858441d356123472 +size 9590 diff --git a/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1/20230606_051509/20230606_051509.log b/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1/20230606_051509/20230606_051509.log new file mode 100644 index 0000000000000000000000000000000000000000..cc18955feb1a9abeb9de354e3d5acb507cc00494 --- /dev/null +++ b/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1/20230606_051509/20230606_051509.log @@ -0,0 +1,1322 @@ +2023/06/06 05:15:13 - mmengine - INFO - +------------------------------------------------------------ +System environment: + sys.platform: linux + Python: 3.10.9 (main, Mar 8 2023, 10:47:38) [GCC 11.2.0] + CUDA available: True + numpy_random_seed: 1184274118 + GPU 0,1: NVIDIA A100-SXM4-80GB + CUDA_HOME: /mnt/petrelfs/share/cuda-11.6 + NVCC: Cuda compilation tools, release 11.6, V11.6.124 + GCC: gcc (GCC) 7.5.0 + PyTorch: 1.13.1 + PyTorch compiling details: PyTorch built with: + - GCC 9.3 + - C++ Version: 201402 + - Intel(R) oneAPI Math Kernel Library Version 2021.4-Product Build 20210904 for Intel(R) 64 architecture applications + - Intel(R) MKL-DNN v2.6.0 (Git Hash 52b5f107dd9cf10910aaa19cb47f3abf9b349815) + - OpenMP 201511 (a.k.a. OpenMP 4.5) + - LAPACK is enabled (usually provided by MKL) + - NNPACK is enabled + - CPU capability usage: AVX2 + - CUDA Runtime 11.6 + - NVCC architecture flags: -gencode;arch=compute_37,code=sm_37;-gencode;arch=compute_50,code=sm_50;-gencode;arch=compute_60,code=sm_60;-gencode;arch=compute_61,code=sm_61;-gencode;arch=compute_70,code=sm_70;-gencode;arch=compute_75,code=sm_75;-gencode;arch=compute_80,code=sm_80;-gencode;arch=compute_86,code=sm_86;-gencode;arch=compute_37,code=compute_37 + - CuDNN 8.3.2 (built against CUDA 11.5) + - Magma 2.6.1 + - Build settings: BLAS_INFO=mkl, BUILD_TYPE=Release, CUDA_VERSION=11.6, CUDNN_VERSION=8.3.2, CXX_COMPILER=/opt/rh/devtoolset-9/root/usr/bin/c++, CXX_FLAGS= -fabi-version=11 -Wno-deprecated -fvisibility-inlines-hidden -DUSE_PTHREADPOOL -fopenmp -DNDEBUG -DUSE_KINETO -DUSE_FBGEMM -DUSE_QNNPACK -DUSE_PYTORCH_QNNPACK -DUSE_XNNPACK -DSYMBOLICATE_MOBILE_DEBUG_HANDLE -DEDGE_PROFILER_USE_KINETO -O2 -fPIC -Wno-narrowing -Wall -Wextra -Werror=return-type -Werror=non-virtual-dtor -Wno-missing-field-initializers -Wno-type-limits -Wno-array-bounds -Wno-unknown-pragmas -Wunused-local-typedefs -Wno-unused-parameter -Wno-unused-function -Wno-unused-result -Wno-strict-overflow -Wno-strict-aliasing -Wno-error=deprecated-declarations -Wno-stringop-overflow -Wno-psabi -Wno-error=pedantic -Wno-error=redundant-decls -Wno-error=old-style-cast -fdiagnostics-color=always -faligned-new -Wno-unused-but-set-variable -Wno-maybe-uninitialized -fno-math-errno -fno-trapping-math -Werror=format -Werror=cast-function-type -Wno-stringop-overflow, LAPACK_INFO=mkl, PERF_WITH_AVX=1, PERF_WITH_AVX2=1, PERF_WITH_AVX512=1, TORCH_VERSION=1.13.1, USE_CUDA=ON, USE_CUDNN=ON, USE_EXCEPTION_PTR=1, USE_GFLAGS=OFF, USE_GLOG=OFF, USE_MKL=ON, USE_MKLDNN=ON, USE_MPI=OFF, USE_NCCL=ON, USE_NNPACK=ON, USE_OPENMP=ON, USE_ROCM=OFF, + + TorchVision: 0.14.1 + OpenCV: 4.7.0 + MMEngine: 0.7.3 + +Runtime environment: + cudnn_benchmark: True + mp_cfg: {'mp_start_method': 'fork', 'opencv_num_threads': 0} + dist_cfg: {'backend': 'nccl'} + seed: None + deterministic: False + Distributed launcher: slurm + Distributed training: True + GPU number: 2 +------------------------------------------------------------ + +2023/06/06 05:15:17 - mmengine - INFO - Config: +optim_wrapper = dict( + optimizer=dict( + type='SGD', + lr=0.0001, + momentum=0.9, + weight_decay=0.0001, + _scope_='mmpretrain'), + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=False, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=512) +model = dict( + type='ImageClassifier', + backbone=dict( + frozen_stages=2, + type='ResNet', + depth=50, + num_stages=4, + out_indices=(3, ), + style='pytorch', + init_cfg=dict( + type='Pretrained', + checkpoint= + 'https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth', + prefix='backbone')), + neck=dict(type='GlobalAveragePooling'), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=2048, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + topk=1)) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/IF80w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/if-dpmsolver++-50-20w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/cc1m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/if-dpmsolver++-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = dict(type='Accuracy', topk=1) +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/if-dpmsolver++-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = dict(type='Accuracy', topk=1) +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1' + +2023/06/06 05:15:27 - mmengine - INFO - Hooks will be executed in the following order: +before_run: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_load_checkpoint: +(ABOVE_NORMAL) EMAHook + -------------------- +before_train: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_train_epoch: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(NORMAL ) DistSamplerSeedHook + -------------------- +before_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook + -------------------- +after_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +after_train_epoch: +(NORMAL ) IterTimerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_val_epoch: +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook + -------------------- +before_val_iter: +(NORMAL ) IterTimerHook + -------------------- +after_val_iter: +(NORMAL ) IterTimerHook +(NORMAL ) VisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_val_epoch: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_save_checkpoint: +(ABOVE_NORMAL) EMAHook + -------------------- +after_train: +(VERY_LOW ) CheckpointHook + -------------------- +before_test_epoch: +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook + -------------------- +before_test_iter: +(NORMAL ) IterTimerHook + -------------------- +after_test_iter: +(NORMAL ) IterTimerHook +(NORMAL ) VisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_test_epoch: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_run: +(BELOW_NORMAL) LoggerHook + -------------------- +2023/06/06 05:15:45 - mmengine - INFO - load backbone in model from: https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth +Name of parameter - Initialization information + +backbone.conv1.weight - torch.Size([64, 3, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.bn1.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.bn1.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.conv1.weight - torch.Size([64, 64, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.bn1.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.bn1.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.conv2.weight - torch.Size([64, 64, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.bn2.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.bn2.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.conv3.weight - torch.Size([256, 64, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.bn3.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.bn3.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.downsample.0.weight - torch.Size([256, 64, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.downsample.1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.downsample.1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.conv1.weight - torch.Size([64, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.bn1.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.bn1.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.conv2.weight - torch.Size([64, 64, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.bn2.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.bn2.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.conv3.weight - torch.Size([256, 64, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.bn3.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.bn3.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.conv1.weight - torch.Size([64, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.bn1.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.bn1.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.conv2.weight - torch.Size([64, 64, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.bn2.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.bn2.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.conv3.weight - torch.Size([256, 64, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.bn3.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.bn3.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.conv1.weight - torch.Size([128, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.bn1.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.bn1.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.conv2.weight - torch.Size([128, 128, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.bn2.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.bn2.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.conv3.weight - torch.Size([512, 128, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.bn3.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.bn3.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.downsample.0.weight - torch.Size([512, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.downsample.1.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.downsample.1.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.conv1.weight - torch.Size([128, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.bn1.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.bn1.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.conv2.weight - torch.Size([128, 128, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.bn2.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.bn2.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.conv3.weight - torch.Size([512, 128, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.bn3.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.bn3.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.conv1.weight - torch.Size([128, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.bn1.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.bn1.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.conv2.weight - torch.Size([128, 128, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.bn2.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.bn2.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.conv3.weight - torch.Size([512, 128, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.bn3.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.bn3.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.conv1.weight - torch.Size([128, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.bn1.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.bn1.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.conv2.weight - torch.Size([128, 128, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.bn2.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.bn2.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.conv3.weight - torch.Size([512, 128, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.bn3.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.bn3.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.conv1.weight - torch.Size([256, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.bn1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.bn1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.conv2.weight - torch.Size([256, 256, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.bn2.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.bn2.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.conv3.weight - torch.Size([1024, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.bn3.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.bn3.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.downsample.0.weight - torch.Size([1024, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.downsample.1.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.downsample.1.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.conv1.weight - torch.Size([256, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.bn1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.bn1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.conv2.weight - torch.Size([256, 256, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.bn2.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.bn2.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.conv3.weight - torch.Size([1024, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.bn3.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.bn3.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.conv1.weight - torch.Size([256, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.bn1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.bn1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.conv2.weight - torch.Size([256, 256, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.bn2.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.bn2.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.conv3.weight - torch.Size([1024, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.bn3.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.bn3.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.conv1.weight - torch.Size([256, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.bn1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.bn1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.conv2.weight - torch.Size([256, 256, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.bn2.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.bn2.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.conv3.weight - torch.Size([1024, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.bn3.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.bn3.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.conv1.weight - torch.Size([256, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.bn1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.bn1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.conv2.weight - torch.Size([256, 256, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.bn2.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.bn2.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.conv3.weight - torch.Size([1024, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.bn3.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.bn3.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.conv1.weight - torch.Size([256, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.bn1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.bn1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.conv2.weight - torch.Size([256, 256, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.bn2.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.bn2.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.conv3.weight - torch.Size([1024, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.bn3.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.bn3.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.conv1.weight - torch.Size([512, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.bn1.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.bn1.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.conv2.weight - torch.Size([512, 512, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.bn2.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.bn2.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.conv3.weight - torch.Size([2048, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.bn3.weight - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.bn3.bias - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.downsample.0.weight - torch.Size([2048, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.downsample.1.weight - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.downsample.1.bias - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.conv1.weight - torch.Size([512, 2048, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.bn1.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.bn1.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.conv2.weight - torch.Size([512, 512, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.bn2.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.bn2.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.conv3.weight - torch.Size([2048, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.bn3.weight - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.bn3.bias - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.conv1.weight - torch.Size([512, 2048, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.bn1.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.bn1.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.conv2.weight - torch.Size([512, 512, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.bn2.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.bn2.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.conv3.weight - torch.Size([2048, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.bn3.weight - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.bn3.bias - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +head.fc.weight - torch.Size([2, 2048]): +NormalInit: mean=0, std=0.01, bias=0 + +head.fc.bias - torch.Size([2]): +NormalInit: mean=0, std=0.01, bias=0 +2023/06/06 05:15:45 - mmengine - WARNING - "FileClient" will be deprecated in future. Please use io functions in https://mmengine.readthedocs.io/en/latest/api/fileio.html#file-io +2023/06/06 05:15:45 - mmengine - WARNING - "HardDiskBackend" is the alias of "LocalBackend" and the former will be deprecated in future. +2023/06/06 05:15:45 - mmengine - INFO - Checkpoints will be saved to /mnt/petrelfs/luzeyu/workspace/fakebench/mmpretrain/workdir/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1. +2023/06/06 05:16:44 - mmengine - INFO - Epoch(train) [1][ 100/3937] lr: 9.9999e-05 eta: 6:23:04 time: 0.5295 data_time: 0.2599 memory: 9436 loss: 0.6402 +2023/06/06 05:17:43 - mmengine - INFO - Epoch(train) [1][ 200/3937] lr: 9.9994e-05 eta: 6:24:07 time: 0.6861 data_time: 0.5440 memory: 6319 loss: 0.5861 +2023/06/06 05:18:55 - mmengine - INFO - Epoch(train) [1][ 300/3937] lr: 9.9987e-05 eta: 6:52:38 time: 0.6989 data_time: 0.5587 memory: 6319 loss: 0.5367 +2023/06/06 05:20:00 - mmengine - INFO - Epoch(train) [1][ 400/3937] lr: 9.9977e-05 eta: 6:54:28 time: 0.6627 data_time: 0.5214 memory: 6319 loss: 0.4854 +2023/06/06 05:21:12 - mmengine - INFO - Epoch(train) [1][ 500/3937] lr: 9.9964e-05 eta: 7:04:07 time: 0.6675 data_time: 0.5269 memory: 6319 loss: 0.4375 +2023/06/06 05:22:19 - mmengine - INFO - Epoch(train) [1][ 600/3937] lr: 9.9949e-05 eta: 7:03:56 time: 0.6597 data_time: 0.5153 memory: 6319 loss: 0.3924 +2023/06/06 05:23:30 - mmengine - INFO - Epoch(train) [1][ 700/3937] lr: 9.9930e-05 eta: 7:08:13 time: 1.1468 data_time: 1.0034 memory: 6319 loss: 0.3822 +2023/06/06 05:24:37 - mmengine - INFO - Epoch(train) [1][ 800/3937] lr: 9.9909e-05 eta: 7:07:28 time: 0.5727 data_time: 0.4322 memory: 6319 loss: 0.3580 +2023/06/06 05:25:47 - mmengine - INFO - Epoch(train) [1][ 900/3937] lr: 9.9884e-05 eta: 7:09:09 time: 0.6277 data_time: 0.4866 memory: 6319 loss: 0.3337 +2023/06/06 05:26:50 - mmengine - INFO - Exp name: resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1_20230606_051509 +2023/06/06 05:26:50 - mmengine - INFO - Epoch(train) [1][1000/3937] lr: 9.9857e-05 eta: 7:05:05 time: 0.6407 data_time: 0.4991 memory: 6319 loss: 0.3300 +2023/06/06 05:27:52 - mmengine - INFO - Epoch(train) [1][1100/3937] lr: 9.9827e-05 eta: 7:01:15 time: 0.6052 data_time: 0.4647 memory: 6319 loss: 0.3195 +2023/06/06 05:28:56 - mmengine - INFO - Epoch(train) [1][1200/3937] lr: 9.9794e-05 eta: 6:59:33 time: 0.6464 data_time: 0.5044 memory: 6319 loss: 0.2946 +2023/06/06 05:30:00 - mmengine - INFO - Epoch(train) [1][1300/3937] lr: 9.9758e-05 eta: 6:57:03 time: 0.6669 data_time: 0.5261 memory: 6319 loss: 0.2901 +2023/06/06 05:31:03 - mmengine - INFO - Epoch(train) [1][1400/3937] lr: 9.9720e-05 eta: 6:55:00 time: 0.6472 data_time: 0.5073 memory: 6319 loss: 0.3030 +2023/06/06 05:32:07 - mmengine - INFO - Epoch(train) [1][1500/3937] lr: 9.9678e-05 eta: 6:53:18 time: 0.6278 data_time: 0.4881 memory: 6319 loss: 0.2953 +2023/06/06 05:33:10 - mmengine - INFO - Epoch(train) [1][1600/3937] lr: 9.9634e-05 eta: 6:51:04 time: 0.6350 data_time: 0.4939 memory: 6319 loss: 0.2869 +2023/06/06 05:34:11 - mmengine - INFO - Epoch(train) [1][1700/3937] lr: 9.9587e-05 eta: 6:48:37 time: 0.6069 data_time: 0.4661 memory: 6319 loss: 0.2917 +2023/06/06 05:36:46 - mmengine - INFO - Epoch(train) [1][1800/3937] lr: 9.9537e-05 eta: 7:18:42 time: 0.5937 data_time: 0.4530 memory: 6319 loss: 0.2901 +2023/06/06 05:37:50 - mmengine - INFO - Epoch(train) [1][1900/3937] lr: 9.9484e-05 eta: 7:15:21 time: 0.6577 data_time: 0.5169 memory: 6319 loss: 0.2740 +2023/06/06 05:39:02 - mmengine - INFO - Exp name: resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1_20230606_051509 +2023/06/06 05:39:02 - mmengine - INFO - Epoch(train) [1][2000/3937] lr: 9.9429e-05 eta: 7:15:01 time: 0.6394 data_time: 0.4975 memory: 6319 loss: 0.2567 +2023/06/06 05:40:09 - mmengine - INFO - Epoch(train) [1][2100/3937] lr: 9.9370e-05 eta: 7:12:53 time: 0.6727 data_time: 0.5305 memory: 6319 loss: 0.2537 +2023/06/06 05:41:03 - mmengine - INFO - Epoch(train) [1][2200/3937] lr: 9.9309e-05 eta: 7:07:31 time: 0.5333 data_time: 0.3924 memory: 6319 loss: 0.2436 +2023/06/06 05:42:05 - mmengine - INFO - Epoch(train) [1][2300/3937] lr: 9.9245e-05 eta: 7:04:16 time: 0.5985 data_time: 0.4582 memory: 6319 loss: 0.2623 +2023/06/06 05:43:04 - mmengine - INFO - Epoch(train) [1][2400/3937] lr: 9.9178e-05 eta: 7:00:53 time: 0.5757 data_time: 0.4275 memory: 6319 loss: 0.2483 +2023/06/06 05:44:06 - mmengine - INFO - Epoch(train) [1][2500/3937] lr: 9.9108e-05 eta: 6:58:09 time: 0.5948 data_time: 0.1390 memory: 6319 loss: 0.2680 +2023/06/06 05:45:09 - mmengine - INFO - Epoch(train) [1][2600/3937] lr: 9.9036e-05 eta: 6:55:39 time: 0.5902 data_time: 0.0009 memory: 6319 loss: 0.2405 +2023/06/06 05:46:10 - mmengine - INFO - Epoch(train) [1][2700/3937] lr: 9.8960e-05 eta: 6:53:02 time: 0.6121 data_time: 0.0009 memory: 6319 loss: 0.2365 +2023/06/06 05:47:12 - mmengine - INFO - Epoch(train) [1][2800/3937] lr: 9.8882e-05 eta: 6:50:40 time: 0.6036 data_time: 0.0008 memory: 6319 loss: 0.2466 +2023/06/06 05:48:13 - mmengine - INFO - Epoch(train) [1][2900/3937] lr: 9.8801e-05 eta: 6:48:22 time: 0.6612 data_time: 0.0009 memory: 6319 loss: 0.2303 +2023/06/06 05:49:15 - mmengine - INFO - Exp name: resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1_20230606_051509 +2023/06/06 05:49:15 - mmengine - INFO - Epoch(train) [1][3000/3937] lr: 9.8718e-05 eta: 6:46:09 time: 0.6020 data_time: 0.0009 memory: 6319 loss: 0.2394 +2023/06/06 05:50:19 - mmengine - INFO - Epoch(train) [1][3100/3937] lr: 9.8631e-05 eta: 6:44:23 time: 0.6488 data_time: 0.0010 memory: 6319 loss: 0.2439 +2023/06/06 05:51:21 - mmengine - INFO - Epoch(train) [1][3200/3937] lr: 9.8542e-05 eta: 6:42:24 time: 0.6052 data_time: 0.0010 memory: 6319 loss: 0.2438 +2023/06/06 05:52:25 - mmengine - INFO - Epoch(train) [1][3300/3937] lr: 9.8450e-05 eta: 6:40:41 time: 0.6643 data_time: 0.0010 memory: 6319 loss: 0.2250 +2023/06/06 05:53:27 - mmengine - INFO - Epoch(train) [1][3400/3937] lr: 9.8355e-05 eta: 6:38:54 time: 0.6440 data_time: 0.0010 memory: 6319 loss: 0.2478 +2023/06/06 05:54:40 - mmengine - INFO - Epoch(train) [1][3500/3937] lr: 9.8257e-05 eta: 6:38:50 time: 0.5712 data_time: 0.0009 memory: 6319 loss: 0.2369 +2023/06/06 05:55:38 - mmengine - INFO - Epoch(train) [1][3600/3937] lr: 9.8157e-05 eta: 6:36:13 time: 0.5564 data_time: 0.0009 memory: 6319 loss: 0.2181 +2023/06/06 05:56:35 - mmengine - INFO - Epoch(train) [1][3700/3937] lr: 9.8054e-05 eta: 6:33:35 time: 0.5984 data_time: 0.0010 memory: 6319 loss: 0.2178 +2023/06/06 05:57:30 - mmengine - INFO - Epoch(train) [1][3800/3937] lr: 9.7948e-05 eta: 6:30:47 time: 0.5614 data_time: 0.0010 memory: 6319 loss: 0.2163 +2023/06/06 05:58:29 - mmengine - INFO - Epoch(train) [1][3900/3937] lr: 9.7840e-05 eta: 6:28:41 time: 0.5717 data_time: 0.0009 memory: 6319 loss: 0.2113 +2023/06/06 05:58:49 - mmengine - INFO - Exp name: resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1_20230606_051509 +2023/06/06 05:58:49 - mmengine - INFO - Saving checkpoint at 1 epochs +2023/06/06 05:59:32 - mmengine - INFO - Epoch(val) [1][57/57] accuracy/top1: 60.0257 data_time: 0.4937 time: 0.5865 +2023/06/06 06:00:15 - mmengine - INFO - Exp name: resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1_20230606_051509 +2023/06/06 06:00:33 - mmengine - INFO - Epoch(train) [2][ 100/3937] lr: 9.7686e-05 eta: 6:25:39 time: 0.5931 data_time: 0.4532 memory: 8776 loss: 0.2233 +2023/06/06 06:01:27 - mmengine - INFO - Epoch(train) [2][ 200/3937] lr: 9.7571e-05 eta: 6:22:54 time: 0.5014 data_time: 0.2470 memory: 6318 loss: 0.2178 +2023/06/06 06:02:20 - mmengine - INFO - Epoch(train) [2][ 300/3937] lr: 9.7454e-05 eta: 6:20:13 time: 0.5809 data_time: 0.0009 memory: 6318 loss: 0.2273 +2023/06/06 06:03:18 - mmengine - INFO - Epoch(train) [2][ 400/3937] lr: 9.7333e-05 eta: 6:18:13 time: 0.5686 data_time: 0.0010 memory: 6318 loss: 0.2305 +2023/06/06 06:04:16 - mmengine - INFO - Epoch(train) [2][ 500/3937] lr: 9.7210e-05 eta: 6:16:15 time: 0.5727 data_time: 0.0008 memory: 6318 loss: 0.2172 +2023/06/06 06:05:15 - mmengine - INFO - Epoch(train) [2][ 600/3937] lr: 9.7084e-05 eta: 6:14:20 time: 0.5349 data_time: 0.0010 memory: 6318 loss: 0.2231 +2023/06/06 06:06:12 - mmengine - INFO - Epoch(train) [2][ 700/3937] lr: 9.6956e-05 eta: 6:12:27 time: 0.5726 data_time: 0.0009 memory: 6318 loss: 0.2080 +2023/06/06 06:07:11 - mmengine - INFO - Epoch(train) [2][ 800/3937] lr: 9.6825e-05 eta: 6:10:41 time: 0.5798 data_time: 0.0009 memory: 6318 loss: 0.2255 +2023/06/06 06:08:12 - mmengine - INFO - Epoch(train) [2][ 900/3937] lr: 9.6691e-05 eta: 6:09:14 time: 0.5881 data_time: 0.0009 memory: 6318 loss: 0.2255 +2023/06/06 06:09:09 - mmengine - INFO - Epoch(train) [2][1000/3937] lr: 9.6554e-05 eta: 6:07:17 time: 0.5565 data_time: 0.0009 memory: 6318 loss: 0.2115 +2023/06/06 06:09:45 - mmengine - INFO - Exp name: resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1_20230606_051509 +2023/06/06 06:10:07 - mmengine - INFO - Epoch(train) [2][1100/3937] lr: 9.6415e-05 eta: 6:05:36 time: 0.5930 data_time: 0.0008 memory: 6318 loss: 0.2049 +2023/06/06 06:11:06 - mmengine - INFO - Epoch(train) [2][1200/3937] lr: 9.6273e-05 eta: 6:03:55 time: 0.5737 data_time: 0.0009 memory: 6318 loss: 0.2254 +2023/06/06 06:12:05 - mmengine - INFO - Epoch(train) [2][1300/3937] lr: 9.6129e-05 eta: 6:02:20 time: 0.5922 data_time: 0.0009 memory: 6318 loss: 0.2058 +2023/06/06 06:13:03 - mmengine - INFO - Epoch(train) [2][1400/3937] lr: 9.5982e-05 eta: 6:00:40 time: 0.5570 data_time: 0.0009 memory: 6318 loss: 0.2013 +2023/06/06 06:14:02 - mmengine - INFO - Epoch(train) [2][1500/3937] lr: 9.5832e-05 eta: 5:59:07 time: 0.5517 data_time: 0.0009 memory: 6318 loss: 0.1963 +2023/06/06 06:14:57 - mmengine - INFO - Epoch(train) [2][1600/3937] lr: 9.5680e-05 eta: 5:57:16 time: 0.5703 data_time: 0.0008 memory: 6318 loss: 0.2013 +2023/06/06 06:15:56 - mmengine - INFO - Epoch(train) [2][1700/3937] lr: 9.5525e-05 eta: 5:55:45 time: 0.6077 data_time: 0.0008 memory: 6318 loss: 0.2135 +2023/06/06 06:16:54 - mmengine - INFO - Epoch(train) [2][1800/3937] lr: 9.5368e-05 eta: 5:54:11 time: 0.5826 data_time: 0.0010 memory: 6318 loss: 0.2057 +2023/06/06 06:17:53 - mmengine - INFO - Epoch(train) [2][1900/3937] lr: 9.5208e-05 eta: 5:52:44 time: 0.5742 data_time: 0.0009 memory: 6318 loss: 0.2116 +2023/06/06 06:18:51 - mmengine - INFO - Epoch(train) [2][2000/3937] lr: 9.5045e-05 eta: 5:51:12 time: 0.5615 data_time: 0.0009 memory: 6318 loss: 0.1854 +2023/06/06 06:19:26 - mmengine - INFO - Exp name: resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1_20230606_051509 +2023/06/06 06:19:50 - mmengine - INFO - Epoch(train) [2][2100/3937] lr: 9.4880e-05 eta: 5:49:48 time: 0.7452 data_time: 0.0009 memory: 6318 loss: 0.2020 +2023/06/06 06:20:49 - mmengine - INFO - Epoch(train) [2][2200/3937] lr: 9.4713e-05 eta: 5:48:23 time: 0.5835 data_time: 0.0009 memory: 6318 loss: 0.2122 +2023/06/06 06:21:42 - mmengine - INFO - Epoch(train) [2][2300/3937] lr: 9.4543e-05 eta: 5:46:26 time: 0.5199 data_time: 0.0012 memory: 6318 loss: 0.2048 +2023/06/06 06:22:39 - mmengine - INFO - Epoch(train) [2][2400/3937] lr: 9.4370e-05 eta: 5:44:55 time: 0.5328 data_time: 0.0009 memory: 6318 loss: 0.2028 +2023/06/06 06:23:40 - mmengine - INFO - Epoch(train) [2][2500/3937] lr: 9.4195e-05 eta: 5:43:41 time: 0.5861 data_time: 0.0009 memory: 6318 loss: 0.1882 +2023/06/06 06:24:40 - mmengine - INFO - Epoch(train) [2][2600/3937] lr: 9.4017e-05 eta: 5:42:28 time: 0.5689 data_time: 0.0008 memory: 6318 loss: 0.1942 +2023/06/06 06:25:37 - mmengine - INFO - Epoch(train) [2][2700/3937] lr: 9.3837e-05 eta: 5:40:59 time: 0.5423 data_time: 0.0009 memory: 6318 loss: 0.2137 +2023/06/06 06:26:35 - mmengine - INFO - Epoch(train) [2][2800/3937] lr: 9.3654e-05 eta: 5:39:34 time: 0.5685 data_time: 0.0010 memory: 6318 loss: 0.2045 +2023/06/06 06:27:34 - mmengine - INFO - Epoch(train) [2][2900/3937] lr: 9.3469e-05 eta: 5:38:15 time: 0.5431 data_time: 0.0009 memory: 6318 loss: 0.2244 +2023/06/06 06:28:31 - mmengine - INFO - Epoch(train) [2][3000/3937] lr: 9.3282e-05 eta: 5:36:47 time: 0.5705 data_time: 0.0009 memory: 6318 loss: 0.1909 +2023/06/06 06:29:05 - mmengine - INFO - Exp name: resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1_20230606_051509 +2023/06/06 06:29:28 - mmengine - INFO - Epoch(train) [2][3100/3937] lr: 9.3092e-05 eta: 5:35:18 time: 0.5858 data_time: 0.0009 memory: 6318 loss: 0.2038 +2023/06/06 06:30:26 - mmengine - INFO - Epoch(train) [2][3200/3937] lr: 9.2899e-05 eta: 5:33:56 time: 0.5599 data_time: 0.0009 memory: 6318 loss: 0.1844 +2023/06/06 06:31:23 - mmengine - INFO - Epoch(train) [2][3300/3937] lr: 9.2705e-05 eta: 5:32:34 time: 0.5775 data_time: 0.0009 memory: 6318 loss: 0.1814 +2023/06/06 06:32:18 - mmengine - INFO - Epoch(train) [2][3400/3937] lr: 9.2507e-05 eta: 5:31:01 time: 0.5599 data_time: 0.0010 memory: 6318 loss: 0.1833 +2023/06/06 06:33:16 - mmengine - INFO - Epoch(train) [2][3500/3937] lr: 9.2308e-05 eta: 5:29:39 time: 0.5388 data_time: 0.0009 memory: 6318 loss: 0.1710 +2023/06/06 06:34:14 - mmengine - INFO - Epoch(train) [2][3600/3937] lr: 9.2106e-05 eta: 5:28:24 time: 0.5797 data_time: 0.0008 memory: 6318 loss: 0.2101 +2023/06/06 06:35:13 - mmengine - INFO - Epoch(train) [2][3700/3937] lr: 9.1902e-05 eta: 5:27:07 time: 0.5979 data_time: 0.0009 memory: 6318 loss: 0.1647 +2023/06/06 06:36:20 - mmengine - INFO - Epoch(train) [2][3800/3937] lr: 9.1695e-05 eta: 5:26:25 time: 0.5740 data_time: 0.0009 memory: 6318 loss: 0.1844 +2023/06/06 06:37:18 - mmengine - INFO - Epoch(train) [2][3900/3937] lr: 9.1486e-05 eta: 5:25:11 time: 0.6056 data_time: 0.0010 memory: 6318 loss: 0.1855 +2023/06/06 06:37:41 - mmengine - INFO - Exp name: resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1_20230606_051509 +2023/06/06 06:37:41 - mmengine - INFO - Saving checkpoint at 2 epochs +2023/06/06 06:38:22 - mmengine - INFO - Epoch(val) [2][57/57] accuracy/top1: 84.0089 data_time: 0.4414 time: 0.5313 +2023/06/06 06:39:22 - mmengine - INFO - Epoch(train) [3][ 100/3937] lr: 9.1196e-05 eta: 5:23:37 time: 0.5919 data_time: 0.3175 memory: 6318 loss: 0.1898 +2023/06/06 06:39:39 - mmengine - INFO - Exp name: resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1_20230606_051509 +2023/06/06 06:40:27 - mmengine - INFO - Epoch(train) [3][ 200/3937] lr: 9.0981e-05 eta: 5:22:46 time: 1.2847 data_time: 0.1723 memory: 6318 loss: 0.1745 +2023/06/06 06:41:24 - mmengine - INFO - Epoch(train) [3][ 300/3937] lr: 9.0764e-05 eta: 5:21:25 time: 0.5463 data_time: 0.0008 memory: 6318 loss: 0.1745 +2023/06/06 06:42:23 - mmengine - INFO - Epoch(train) [3][ 400/3937] lr: 9.0545e-05 eta: 5:20:13 time: 0.5926 data_time: 0.0009 memory: 6318 loss: 0.1823 +2023/06/06 06:43:22 - mmengine - INFO - Epoch(train) [3][ 500/3937] lr: 9.0324e-05 eta: 5:19:01 time: 0.5694 data_time: 0.0012 memory: 6318 loss: 0.1747 +2023/06/06 06:44:20 - mmengine - INFO - Epoch(train) [3][ 600/3937] lr: 9.0100e-05 eta: 5:17:46 time: 0.5914 data_time: 0.0008 memory: 6318 loss: 0.1855 +2023/06/06 06:45:19 - mmengine - INFO - Epoch(train) [3][ 700/3937] lr: 8.9875e-05 eta: 5:16:36 time: 0.5474 data_time: 0.0010 memory: 6318 loss: 0.1901 +2023/06/06 06:46:17 - mmengine - INFO - Epoch(train) [3][ 800/3937] lr: 8.9647e-05 eta: 5:15:21 time: 0.5780 data_time: 0.0009 memory: 6318 loss: 0.1888 +2023/06/06 06:47:14 - mmengine - INFO - Epoch(train) [3][ 900/3937] lr: 8.9416e-05 eta: 5:14:03 time: 0.5762 data_time: 0.0010 memory: 6318 loss: 0.1732 +2023/06/06 06:48:11 - mmengine - INFO - Epoch(train) [3][1000/3937] lr: 8.9184e-05 eta: 5:12:45 time: 0.5683 data_time: 0.0008 memory: 6318 loss: 0.1767 +2023/06/06 06:49:11 - mmengine - INFO - Epoch(train) [3][1100/3937] lr: 8.8949e-05 eta: 5:11:39 time: 0.5964 data_time: 0.0009 memory: 6318 loss: 0.2008 +2023/06/06 06:49:24 - mmengine - INFO - Exp name: resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1_20230606_051509 +2023/06/06 06:50:07 - mmengine - INFO - Epoch(train) [3][1200/3937] lr: 8.8712e-05 eta: 5:10:20 time: 0.5894 data_time: 0.0012 memory: 6318 loss: 0.1724 +2023/06/06 06:51:04 - mmengine - INFO - Epoch(train) [3][1300/3937] lr: 8.8474e-05 eta: 5:09:02 time: 0.5614 data_time: 0.0009 memory: 6318 loss: 0.1705 +2023/06/06 06:52:03 - mmengine - INFO - Epoch(train) [3][1400/3937] lr: 8.8232e-05 eta: 5:07:53 time: 0.5741 data_time: 0.0009 memory: 6318 loss: 0.1763 +2023/06/06 06:52:59 - mmengine - INFO - Epoch(train) [3][1500/3937] lr: 8.7989e-05 eta: 5:06:34 time: 0.5469 data_time: 0.0009 memory: 6318 loss: 0.1659 +2023/06/06 06:53:57 - mmengine - INFO - Epoch(train) [3][1600/3937] lr: 8.7744e-05 eta: 5:05:22 time: 0.5935 data_time: 0.0011 memory: 6318 loss: 0.1798 +2023/06/06 06:54:54 - mmengine - INFO - Epoch(train) [3][1700/3937] lr: 8.7497e-05 eta: 5:04:10 time: 0.5515 data_time: 0.0009 memory: 6318 loss: 0.1756 +2023/06/06 06:55:52 - mmengine - INFO - Epoch(train) [3][1800/3937] lr: 8.7247e-05 eta: 5:02:57 time: 0.5799 data_time: 0.0009 memory: 6318 loss: 0.1722 +2023/06/06 06:56:51 - mmengine - INFO - Epoch(train) [3][1900/3937] lr: 8.6996e-05 eta: 5:01:48 time: 0.5886 data_time: 0.0009 memory: 6318 loss: 0.1667 +2023/06/06 06:57:48 - mmengine - INFO - Epoch(train) [3][2000/3937] lr: 8.6742e-05 eta: 5:00:37 time: 0.5694 data_time: 0.0009 memory: 6318 loss: 0.1593 +2023/06/06 06:58:47 - mmengine - INFO - Epoch(train) [3][2100/3937] lr: 8.6487e-05 eta: 4:59:27 time: 0.5852 data_time: 0.0009 memory: 6318 loss: 0.1758 +2023/06/06 06:59:01 - mmengine - INFO - Exp name: resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1_20230606_051509 +2023/06/06 06:59:44 - mmengine - INFO - Epoch(train) [3][2200/3937] lr: 8.6229e-05 eta: 4:58:15 time: 0.6135 data_time: 0.0009 memory: 6318 loss: 0.1730 +2023/06/06 07:00:40 - mmengine - INFO - Epoch(train) [3][2300/3937] lr: 8.5970e-05 eta: 4:57:00 time: 0.5560 data_time: 0.0011 memory: 6318 loss: 0.1819 +2023/06/06 07:01:33 - mmengine - INFO - Epoch(train) [3][2400/3937] lr: 8.5708e-05 eta: 4:55:36 time: 0.5502 data_time: 0.0008 memory: 6318 loss: 0.1792 +2023/06/06 07:02:32 - mmengine - INFO - Epoch(train) [3][2500/3937] lr: 8.5445e-05 eta: 4:54:30 time: 0.6083 data_time: 0.0011 memory: 6318 loss: 0.1626 +2023/06/06 07:03:30 - mmengine - INFO - Epoch(train) [3][2600/3937] lr: 8.5179e-05 eta: 4:53:19 time: 0.5666 data_time: 0.0008 memory: 6318 loss: 0.1758 +2023/06/06 07:04:28 - mmengine - INFO - Epoch(train) [3][2700/3937] lr: 8.4912e-05 eta: 4:52:11 time: 0.6049 data_time: 0.0010 memory: 6318 loss: 0.1712 +2023/06/06 07:05:25 - mmengine - INFO - Epoch(train) [3][2800/3937] lr: 8.4643e-05 eta: 4:51:01 time: 0.5280 data_time: 0.0010 memory: 6318 loss: 0.1630 +2023/06/06 07:06:23 - mmengine - INFO - Epoch(train) [3][2900/3937] lr: 8.4372e-05 eta: 4:49:52 time: 0.5584 data_time: 0.0011 memory: 6318 loss: 0.1777 +2023/06/06 07:07:18 - mmengine - INFO - Epoch(train) [3][3000/3937] lr: 8.4099e-05 eta: 4:48:35 time: 0.5353 data_time: 0.0009 memory: 6318 loss: 0.1592 +2023/06/06 07:08:15 - mmengine - INFO - Epoch(train) [3][3100/3937] lr: 8.3824e-05 eta: 4:47:24 time: 0.5936 data_time: 0.0009 memory: 6318 loss: 0.1803 +2023/06/06 07:08:30 - mmengine - INFO - Exp name: resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1_20230606_051509 +2023/06/06 07:09:12 - mmengine - INFO - Epoch(train) [3][3200/3937] lr: 8.3547e-05 eta: 4:46:16 time: 0.5961 data_time: 0.0010 memory: 6318 loss: 0.1638 +2023/06/06 07:10:09 - mmengine - INFO - Epoch(train) [3][3300/3937] lr: 8.3269e-05 eta: 4:45:05 time: 0.5805 data_time: 0.0009 memory: 6318 loss: 0.1925 +2023/06/06 07:11:06 - mmengine - INFO - Epoch(train) [3][3400/3937] lr: 8.2988e-05 eta: 4:43:55 time: 0.5538 data_time: 0.0010 memory: 6318 loss: 0.1608 +2023/06/06 07:12:04 - mmengine - INFO - Epoch(train) [3][3500/3937] lr: 8.2706e-05 eta: 4:42:46 time: 0.5589 data_time: 0.0010 memory: 6318 loss: 0.1523 +2023/06/06 07:13:01 - mmengine - INFO - Epoch(train) [3][3600/3937] lr: 8.2423e-05 eta: 4:41:38 time: 0.5637 data_time: 0.0008 memory: 6318 loss: 0.1623 +2023/06/06 07:14:00 - mmengine - INFO - Epoch(train) [3][3700/3937] lr: 8.2137e-05 eta: 4:40:33 time: 0.5647 data_time: 0.0010 memory: 6318 loss: 0.1633 +2023/06/06 07:14:58 - mmengine - INFO - Epoch(train) [3][3800/3937] lr: 8.1850e-05 eta: 4:39:26 time: 0.5698 data_time: 0.0009 memory: 6318 loss: 0.1582 +2023/06/06 07:15:56 - mmengine - INFO - Epoch(train) [3][3900/3937] lr: 8.1561e-05 eta: 4:38:20 time: 0.5732 data_time: 0.0009 memory: 6318 loss: 0.1608 +2023/06/06 07:16:17 - mmengine - INFO - Exp name: resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1_20230606_051509 +2023/06/06 07:16:17 - mmengine - INFO - Saving checkpoint at 3 epochs +2023/06/06 07:16:59 - mmengine - INFO - Epoch(val) [3][57/57] accuracy/top1: 90.7241 data_time: 0.4454 time: 0.5321 +2023/06/06 07:17:59 - mmengine - INFO - Epoch(train) [4][ 100/3937] lr: 8.1162e-05 eta: 4:36:55 time: 0.5837 data_time: 0.2446 memory: 6318 loss: 0.1625 +2023/06/06 07:18:52 - mmengine - INFO - Exp name: resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1_20230606_051509 +2023/06/06 07:18:58 - mmengine - INFO - Epoch(train) [4][ 200/3937] lr: 8.0869e-05 eta: 4:35:50 time: 0.5979 data_time: 0.0009 memory: 6318 loss: 0.1551 +2023/06/06 07:19:57 - mmengine - INFO - Epoch(train) [4][ 300/3937] lr: 8.0574e-05 eta: 4:34:46 time: 0.5647 data_time: 0.0009 memory: 6318 loss: 0.1561 +2023/06/06 07:20:54 - mmengine - INFO - Epoch(train) [4][ 400/3937] lr: 8.0278e-05 eta: 4:33:38 time: 0.5088 data_time: 0.0009 memory: 6318 loss: 0.1554 +2023/06/06 07:21:54 - mmengine - INFO - Epoch(train) [4][ 500/3937] lr: 7.9980e-05 eta: 4:32:35 time: 0.5707 data_time: 0.0010 memory: 6318 loss: 0.1611 +2023/06/06 07:22:51 - mmengine - INFO - Epoch(train) [4][ 600/3937] lr: 7.9681e-05 eta: 4:31:29 time: 0.5646 data_time: 0.0008 memory: 6318 loss: 0.1700 +2023/06/06 07:23:51 - mmengine - INFO - Epoch(train) [4][ 700/3937] lr: 7.9380e-05 eta: 4:30:26 time: 0.6077 data_time: 0.0011 memory: 6318 loss: 0.1626 +2023/06/06 07:24:48 - mmengine - INFO - Epoch(train) [4][ 800/3937] lr: 7.9077e-05 eta: 4:29:19 time: 0.5592 data_time: 0.0009 memory: 6318 loss: 0.1516 +2023/06/06 07:25:46 - mmengine - INFO - Epoch(train) [4][ 900/3937] lr: 7.8773e-05 eta: 4:28:13 time: 0.6144 data_time: 0.0010 memory: 6318 loss: 0.1790 +2023/06/06 07:26:46 - mmengine - INFO - Epoch(train) [4][1000/3937] lr: 7.8467e-05 eta: 4:27:13 time: 0.5863 data_time: 0.0009 memory: 6318 loss: 0.1618 +2023/06/06 07:27:44 - mmengine - INFO - Epoch(train) [4][1100/3937] lr: 7.8160e-05 eta: 4:26:08 time: 0.6009 data_time: 0.0009 memory: 6318 loss: 0.1544 +2023/06/06 07:28:35 - mmengine - INFO - Exp name: resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1_20230606_051509 +2023/06/06 07:28:41 - mmengine - INFO - Epoch(train) [4][1200/3937] lr: 7.7852e-05 eta: 4:25:00 time: 0.5613 data_time: 0.0009 memory: 6318 loss: 0.1451 +2023/06/06 07:29:38 - mmengine - INFO - Epoch(train) [4][1300/3937] lr: 7.7541e-05 eta: 4:23:54 time: 0.5536 data_time: 0.0009 memory: 6318 loss: 0.1528 +2023/06/06 07:30:37 - mmengine - INFO - Epoch(train) [4][1400/3937] lr: 7.7230e-05 eta: 4:22:50 time: 0.6178 data_time: 0.0009 memory: 6318 loss: 0.1458 +2023/06/06 07:31:37 - mmengine - INFO - Epoch(train) [4][1500/3937] lr: 7.6917e-05 eta: 4:21:49 time: 0.5422 data_time: 0.0012 memory: 6318 loss: 0.1695 +2023/06/06 07:32:34 - mmengine - INFO - Epoch(train) [4][1600/3937] lr: 7.6603e-05 eta: 4:20:44 time: 0.5812 data_time: 0.0009 memory: 6318 loss: 0.1484 +2023/06/06 07:33:36 - mmengine - INFO - Epoch(train) [4][1700/3937] lr: 7.6287e-05 eta: 4:19:47 time: 0.5679 data_time: 0.0010 memory: 6318 loss: 0.1634 +2023/06/06 07:34:35 - mmengine - INFO - Epoch(train) [4][1800/3937] lr: 7.5970e-05 eta: 4:18:44 time: 0.5532 data_time: 0.0009 memory: 6318 loss: 0.1580 +2023/06/06 07:35:34 - mmengine - INFO - Epoch(train) [4][1900/3937] lr: 7.5652e-05 eta: 4:17:41 time: 0.5672 data_time: 0.0009 memory: 6318 loss: 0.1517 +2023/06/06 07:36:32 - mmengine - INFO - Epoch(train) [4][2000/3937] lr: 7.5332e-05 eta: 4:16:36 time: 0.5388 data_time: 0.0009 memory: 6318 loss: 0.1487 +2023/06/06 07:37:34 - mmengine - INFO - Epoch(train) [4][2100/3937] lr: 7.5011e-05 eta: 4:15:39 time: 0.9346 data_time: 0.0008 memory: 6318 loss: 0.1681 +2023/06/06 07:38:25 - mmengine - INFO - Exp name: resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1_20230606_051509 +2023/06/06 07:38:31 - mmengine - INFO - Epoch(train) [4][2200/3937] lr: 7.4689e-05 eta: 4:14:34 time: 0.5879 data_time: 0.0009 memory: 6318 loss: 0.1622 +2023/06/06 07:39:29 - mmengine - INFO - Epoch(train) [4][2300/3937] lr: 7.4365e-05 eta: 4:13:28 time: 0.5491 data_time: 0.0010 memory: 6318 loss: 0.1659 +2023/06/06 07:40:27 - mmengine - INFO - Epoch(train) [4][2400/3937] lr: 7.4040e-05 eta: 4:12:25 time: 0.6071 data_time: 0.0010 memory: 6318 loss: 0.1606 +2023/06/06 07:41:21 - mmengine - INFO - Epoch(train) [4][2500/3937] lr: 7.3714e-05 eta: 4:11:14 time: 0.4698 data_time: 0.0009 memory: 6318 loss: 0.1526 +2023/06/06 07:42:18 - mmengine - INFO - Epoch(train) [4][2600/3937] lr: 7.3387e-05 eta: 4:10:08 time: 0.5767 data_time: 0.0011 memory: 6318 loss: 0.1433 +2023/06/06 07:43:12 - mmengine - INFO - Epoch(train) [4][2700/3937] lr: 7.3059e-05 eta: 4:08:58 time: 0.6133 data_time: 0.0011 memory: 6318 loss: 0.1601 +2023/06/06 07:44:12 - mmengine - INFO - Epoch(train) [4][2800/3937] lr: 7.2730e-05 eta: 4:07:57 time: 0.6078 data_time: 0.0009 memory: 6318 loss: 0.1602 +2023/06/06 07:45:10 - mmengine - INFO - Epoch(train) [4][2900/3937] lr: 7.2399e-05 eta: 4:06:54 time: 0.5470 data_time: 0.0010 memory: 6318 loss: 0.1578 +2023/06/06 07:46:08 - mmengine - INFO - Epoch(train) [4][3000/3937] lr: 7.2067e-05 eta: 4:05:50 time: 0.6376 data_time: 0.0009 memory: 6318 loss: 0.1417 +2023/06/06 07:47:06 - mmengine - INFO - Epoch(train) [4][3100/3937] lr: 7.1734e-05 eta: 4:04:47 time: 0.5643 data_time: 0.0009 memory: 6318 loss: 0.1480 +2023/06/06 07:47:58 - mmengine - INFO - Exp name: resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1_20230606_051509 +2023/06/06 07:48:04 - mmengine - INFO - Epoch(train) [4][3200/3937] lr: 7.1401e-05 eta: 4:03:43 time: 0.5604 data_time: 0.0009 memory: 6318 loss: 0.1504 +2023/06/06 07:49:02 - mmengine - INFO - Epoch(train) [4][3300/3937] lr: 7.1066e-05 eta: 4:02:40 time: 0.6343 data_time: 0.0009 memory: 6318 loss: 0.1690 +2023/06/06 07:49:59 - mmengine - INFO - Epoch(train) [4][3400/3937] lr: 7.0730e-05 eta: 4:01:36 time: 0.5949 data_time: 0.0010 memory: 6318 loss: 0.1599 +2023/06/06 07:50:57 - mmengine - INFO - Epoch(train) [4][3500/3937] lr: 7.0393e-05 eta: 4:00:33 time: 0.5798 data_time: 0.0009 memory: 6318 loss: 0.1438 +2023/06/06 07:51:55 - mmengine - INFO - Epoch(train) [4][3600/3937] lr: 7.0055e-05 eta: 3:59:29 time: 0.5852 data_time: 0.0010 memory: 6318 loss: 0.1646 +2023/06/06 07:52:54 - mmengine - INFO - Epoch(train) [4][3700/3937] lr: 6.9716e-05 eta: 3:58:27 time: 0.6001 data_time: 0.0009 memory: 6318 loss: 0.1458 +2023/06/06 07:53:53 - mmengine - INFO - Epoch(train) [4][3800/3937] lr: 6.9376e-05 eta: 3:57:25 time: 0.5926 data_time: 0.0009 memory: 6318 loss: 0.1710 +2023/06/06 07:54:50 - mmengine - INFO - Epoch(train) [4][3900/3937] lr: 6.9035e-05 eta: 3:56:21 time: 0.5645 data_time: 0.0010 memory: 6318 loss: 0.1476 +2023/06/06 07:55:11 - mmengine - INFO - Exp name: resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1_20230606_051509 +2023/06/06 07:55:11 - mmengine - INFO - Saving checkpoint at 4 epochs +2023/06/06 07:55:53 - mmengine - INFO - Epoch(val) [4][57/57] accuracy/top1: 92.7421 data_time: 0.4560 time: 0.5418 +2023/06/06 07:56:54 - mmengine - INFO - Epoch(train) [5][ 100/3937] lr: 6.8567e-05 eta: 3:54:59 time: 0.5487 data_time: 0.0008 memory: 6318 loss: 0.1450 +2023/06/06 07:57:52 - mmengine - INFO - Epoch(train) [5][ 200/3937] lr: 6.8224e-05 eta: 3:53:58 time: 0.6196 data_time: 0.0008 memory: 6318 loss: 0.1527 +2023/06/06 07:58:27 - mmengine - INFO - Exp name: resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1_20230606_051509 +2023/06/06 07:58:53 - mmengine - INFO - Epoch(train) [5][ 300/3937] lr: 6.7881e-05 eta: 3:52:59 time: 0.5848 data_time: 0.0008 memory: 6318 loss: 0.1566 +2023/06/06 07:59:52 - mmengine - INFO - Epoch(train) [5][ 400/3937] lr: 6.7536e-05 eta: 3:51:57 time: 0.5540 data_time: 0.0010 memory: 6318 loss: 0.1581 +2023/06/06 08:00:47 - mmengine - INFO - Epoch(train) [5][ 500/3937] lr: 6.7191e-05 eta: 3:50:51 time: 0.5832 data_time: 0.0012 memory: 6318 loss: 0.1559 +2023/06/06 08:01:42 - mmengine - INFO - Epoch(train) [5][ 600/3937] lr: 6.6845e-05 eta: 3:49:43 time: 0.5620 data_time: 0.0010 memory: 6318 loss: 0.1627 +2023/06/06 08:02:39 - mmengine - INFO - Epoch(train) [5][ 700/3937] lr: 6.6498e-05 eta: 3:48:40 time: 0.5661 data_time: 0.0010 memory: 6318 loss: 0.1359 +2023/06/06 08:03:36 - mmengine - INFO - Epoch(train) [5][ 800/3937] lr: 6.6151e-05 eta: 3:47:35 time: 0.5730 data_time: 0.0009 memory: 6318 loss: 0.1652 +2023/06/06 08:04:32 - mmengine - INFO - Epoch(train) [5][ 900/3937] lr: 6.5802e-05 eta: 3:46:30 time: 0.6023 data_time: 0.0010 memory: 6318 loss: 0.1497 +2023/06/06 08:05:28 - mmengine - INFO - Epoch(train) [5][1000/3937] lr: 6.5454e-05 eta: 3:45:26 time: 0.5546 data_time: 0.0010 memory: 6318 loss: 0.1397 +2023/06/06 08:06:23 - mmengine - INFO - Epoch(train) [5][1100/3937] lr: 6.5104e-05 eta: 3:44:20 time: 0.5315 data_time: 0.0009 memory: 6318 loss: 0.1610 +2023/06/06 08:07:19 - mmengine - INFO - Epoch(train) [5][1200/3937] lr: 6.4754e-05 eta: 3:43:15 time: 0.5730 data_time: 0.0009 memory: 6318 loss: 0.1423 +2023/06/06 08:07:52 - mmengine - INFO - Exp name: resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1_20230606_051509 +2023/06/06 08:08:16 - mmengine - INFO - Epoch(train) [5][1300/3937] lr: 6.4403e-05 eta: 3:42:11 time: 0.5521 data_time: 0.0010 memory: 6318 loss: 0.1626 +2023/06/06 08:09:12 - mmengine - INFO - Epoch(train) [5][1400/3937] lr: 6.4051e-05 eta: 3:41:07 time: 0.5918 data_time: 0.0010 memory: 6318 loss: 0.1473 +2023/06/06 08:10:09 - mmengine - INFO - Epoch(train) [5][1500/3937] lr: 6.3699e-05 eta: 3:40:04 time: 0.6313 data_time: 0.0010 memory: 6318 loss: 0.1543 +2023/06/06 08:11:06 - mmengine - INFO - Epoch(train) [5][1600/3937] lr: 6.3347e-05 eta: 3:39:01 time: 0.5824 data_time: 0.0009 memory: 6318 loss: 0.1421 +2023/06/06 08:12:01 - mmengine - INFO - Epoch(train) [5][1700/3937] lr: 6.2994e-05 eta: 3:37:56 time: 0.5913 data_time: 0.0010 memory: 6318 loss: 0.1583 +2023/06/06 08:12:58 - mmengine - INFO - Epoch(train) [5][1800/3937] lr: 6.2640e-05 eta: 3:36:52 time: 0.5646 data_time: 0.0011 memory: 6318 loss: 0.1450 +2023/06/06 08:13:54 - mmengine - INFO - Epoch(train) [5][1900/3937] lr: 6.2286e-05 eta: 3:35:47 time: 0.5523 data_time: 0.0010 memory: 6318 loss: 0.1400 +2023/06/06 08:14:51 - mmengine - INFO - Epoch(train) [5][2000/3937] lr: 6.1931e-05 eta: 3:34:46 time: 0.5800 data_time: 0.0012 memory: 6318 loss: 0.1443 +2023/06/06 08:15:48 - mmengine - INFO - Epoch(train) [5][2100/3937] lr: 6.1576e-05 eta: 3:33:43 time: 0.5655 data_time: 0.0009 memory: 6318 loss: 0.1535 +2023/06/06 08:16:46 - mmengine - INFO - Epoch(train) [5][2200/3937] lr: 6.1221e-05 eta: 3:32:41 time: 0.5932 data_time: 0.0009 memory: 6318 loss: 0.1531 +2023/06/06 08:17:14 - mmengine - INFO - Exp name: resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1_20230606_051509 +2023/06/06 08:17:40 - mmengine - INFO - Epoch(train) [5][2300/3937] lr: 6.0865e-05 eta: 3:31:35 time: 0.4993 data_time: 0.0012 memory: 6318 loss: 0.1525 +2023/06/06 08:18:36 - mmengine - INFO - Epoch(train) [5][2400/3937] lr: 6.0509e-05 eta: 3:30:32 time: 0.5357 data_time: 0.0009 memory: 6318 loss: 0.1452 +2023/06/06 08:19:33 - mmengine - INFO - Epoch(train) [5][2500/3937] lr: 6.0152e-05 eta: 3:29:28 time: 0.5359 data_time: 0.0010 memory: 6318 loss: 0.1547 +2023/06/06 08:20:28 - mmengine - INFO - Epoch(train) [5][2600/3937] lr: 5.9795e-05 eta: 3:28:24 time: 0.5863 data_time: 0.0008 memory: 6318 loss: 0.1354 +2023/06/06 08:21:23 - mmengine - INFO - Epoch(train) [5][2700/3937] lr: 5.9438e-05 eta: 3:27:19 time: 0.5796 data_time: 0.0009 memory: 6318 loss: 0.1521 +2023/06/06 08:22:20 - mmengine - INFO - Epoch(train) [5][2800/3937] lr: 5.9081e-05 eta: 3:26:17 time: 0.5718 data_time: 0.0009 memory: 6318 loss: 0.1568 +2023/06/06 08:23:16 - mmengine - INFO - Epoch(train) [5][2900/3937] lr: 5.8723e-05 eta: 3:25:14 time: 0.5511 data_time: 0.0009 memory: 6318 loss: 0.1609 +2023/06/06 08:24:14 - mmengine - INFO - Epoch(train) [5][3000/3937] lr: 5.8365e-05 eta: 3:24:13 time: 0.5489 data_time: 0.0010 memory: 6318 loss: 0.1484 +2023/06/06 08:25:11 - mmengine - INFO - Epoch(train) [5][3100/3937] lr: 5.8007e-05 eta: 3:23:11 time: 0.5856 data_time: 0.0008 memory: 6318 loss: 0.1374 +2023/06/06 08:26:08 - mmengine - INFO - Epoch(train) [5][3200/3937] lr: 5.7649e-05 eta: 3:22:08 time: 0.5738 data_time: 0.0009 memory: 6318 loss: 0.1389 +2023/06/06 08:26:37 - mmengine - INFO - Exp name: resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1_20230606_051509 +2023/06/06 08:27:03 - mmengine - INFO - Epoch(train) [5][3300/3937] lr: 5.7290e-05 eta: 3:21:05 time: 0.5383 data_time: 0.0009 memory: 6318 loss: 0.1546 +2023/06/06 08:27:59 - mmengine - INFO - Epoch(train) [5][3400/3937] lr: 5.6931e-05 eta: 3:20:02 time: 0.5499 data_time: 0.0011 memory: 6318 loss: 0.1322 +2023/06/06 08:28:54 - mmengine - INFO - Epoch(train) [5][3500/3937] lr: 5.6572e-05 eta: 3:18:58 time: 0.5480 data_time: 0.0009 memory: 6318 loss: 0.1386 +2023/06/06 08:29:51 - mmengine - INFO - Epoch(train) [5][3600/3937] lr: 5.6214e-05 eta: 3:17:56 time: 0.5767 data_time: 0.0009 memory: 6318 loss: 0.1499 +2023/06/06 08:30:47 - mmengine - INFO - Epoch(train) [5][3700/3937] lr: 5.5855e-05 eta: 3:16:54 time: 0.5520 data_time: 0.0009 memory: 6318 loss: 0.1284 +2023/06/06 08:31:49 - mmengine - INFO - Epoch(train) [5][3800/3937] lr: 5.5496e-05 eta: 3:15:57 time: 0.5452 data_time: 0.0009 memory: 6318 loss: 0.1420 +2023/06/06 08:32:45 - mmengine - INFO - Epoch(train) [5][3900/3937] lr: 5.5136e-05 eta: 3:14:54 time: 0.5302 data_time: 0.0011 memory: 6318 loss: 0.1391 +2023/06/06 08:33:03 - mmengine - INFO - Exp name: resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1_20230606_051509 +2023/06/06 08:33:03 - mmengine - INFO - Saving checkpoint at 5 epochs +2023/06/06 08:33:45 - mmengine - INFO - Epoch(val) [5][57/57] accuracy/top1: 93.6502 data_time: 0.4446 time: 0.5333 +2023/06/06 08:34:44 - mmengine - INFO - Epoch(train) [6][ 100/3937] lr: 5.4645e-05 eta: 3:13:29 time: 0.5866 data_time: 0.0584 memory: 6318 loss: 0.1412 +2023/06/06 08:35:42 - mmengine - INFO - Epoch(train) [6][ 200/3937] lr: 5.4285e-05 eta: 3:12:28 time: 0.5428 data_time: 0.0009 memory: 6318 loss: 0.1390 +2023/06/06 08:36:38 - mmengine - INFO - Epoch(train) [6][ 300/3937] lr: 5.3926e-05 eta: 3:11:26 time: 0.5615 data_time: 0.0010 memory: 6318 loss: 0.1340 +2023/06/06 08:36:48 - mmengine - INFO - Exp name: resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1_20230606_051509 +2023/06/06 08:37:34 - mmengine - INFO - Epoch(train) [6][ 400/3937] lr: 5.3567e-05 eta: 3:10:23 time: 0.5408 data_time: 0.0009 memory: 6318 loss: 0.1313 +2023/06/06 08:38:29 - mmengine - INFO - Epoch(train) [6][ 500/3937] lr: 5.3209e-05 eta: 3:09:20 time: 0.5378 data_time: 0.0010 memory: 6318 loss: 0.1389 +2023/06/06 08:39:30 - mmengine - INFO - Epoch(train) [6][ 600/3937] lr: 5.2850e-05 eta: 3:08:23 time: 0.5119 data_time: 0.0010 memory: 6318 loss: 0.1533 +2023/06/06 08:40:25 - mmengine - INFO - Epoch(train) [6][ 700/3937] lr: 5.2491e-05 eta: 3:07:19 time: 0.5120 data_time: 0.0009 memory: 6318 loss: 0.1379 +2023/06/06 08:41:18 - mmengine - INFO - Epoch(train) [6][ 800/3937] lr: 5.2133e-05 eta: 3:06:14 time: 0.5415 data_time: 0.0009 memory: 6318 loss: 0.1381 +2023/06/06 08:42:09 - mmengine - INFO - Epoch(train) [6][ 900/3937] lr: 5.1775e-05 eta: 3:05:08 time: 0.5622 data_time: 0.0009 memory: 6318 loss: 0.1505 +2023/06/06 08:43:02 - mmengine - INFO - Epoch(train) [6][1000/3937] lr: 5.1417e-05 eta: 3:04:04 time: 0.5471 data_time: 0.0009 memory: 6318 loss: 0.1377 +2023/06/06 08:43:55 - mmengine - INFO - Epoch(train) [6][1100/3937] lr: 5.1059e-05 eta: 3:02:59 time: 0.4803 data_time: 0.0009 memory: 6318 loss: 0.1422 +2023/06/06 08:44:50 - mmengine - INFO - Epoch(train) [6][1200/3937] lr: 5.0701e-05 eta: 3:01:57 time: 0.5403 data_time: 0.0008 memory: 6318 loss: 0.1364 +2023/06/06 08:45:43 - mmengine - INFO - Epoch(train) [6][1300/3937] lr: 5.0344e-05 eta: 3:00:52 time: 0.5780 data_time: 0.0008 memory: 6318 loss: 0.1449 +2023/06/06 08:45:53 - mmengine - INFO - Exp name: resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1_20230606_051509 +2023/06/06 08:46:37 - mmengine - INFO - Epoch(train) [6][1400/3937] lr: 4.9987e-05 eta: 2:59:48 time: 0.5561 data_time: 0.0009 memory: 6318 loss: 0.1285 +2023/06/06 08:47:31 - mmengine - INFO - Epoch(train) [6][1500/3937] lr: 4.9630e-05 eta: 2:58:45 time: 0.5164 data_time: 0.0008 memory: 6318 loss: 0.1519 +2023/06/06 08:48:24 - mmengine - INFO - Epoch(train) [6][1600/3937] lr: 4.9274e-05 eta: 2:57:41 time: 0.5363 data_time: 0.0008 memory: 6318 loss: 0.1337 +2023/06/06 08:49:17 - mmengine - INFO - Epoch(train) [6][1700/3937] lr: 4.8918e-05 eta: 2:56:37 time: 0.5094 data_time: 0.0008 memory: 6318 loss: 0.1456 +2023/06/06 08:50:11 - mmengine - INFO - Epoch(train) [6][1800/3937] lr: 4.8562e-05 eta: 2:55:34 time: 0.5598 data_time: 0.0009 memory: 6318 loss: 0.1324 +2023/06/06 08:51:04 - mmengine - INFO - Epoch(train) [6][1900/3937] lr: 4.8207e-05 eta: 2:54:30 time: 0.5867 data_time: 0.0008 memory: 6318 loss: 0.1638 +2023/06/06 08:51:58 - mmengine - INFO - Epoch(train) [6][2000/3937] lr: 4.7852e-05 eta: 2:53:27 time: 0.5411 data_time: 0.0010 memory: 6318 loss: 0.1349 +2023/06/06 08:52:51 - mmengine - INFO - Epoch(train) [6][2100/3937] lr: 4.7498e-05 eta: 2:52:24 time: 0.5153 data_time: 0.0009 memory: 6318 loss: 0.1360 +2023/06/06 08:53:43 - mmengine - INFO - Epoch(train) [6][2200/3937] lr: 4.7144e-05 eta: 2:51:20 time: 0.5270 data_time: 0.0009 memory: 6318 loss: 0.1340 +2023/06/06 08:54:36 - mmengine - INFO - Epoch(train) [6][2300/3937] lr: 4.6791e-05 eta: 2:50:16 time: 0.5041 data_time: 0.0009 memory: 6318 loss: 0.1345 +2023/06/06 08:54:46 - mmengine - INFO - Exp name: resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1_20230606_051509 +2023/06/06 08:55:29 - mmengine - INFO - Epoch(train) [6][2400/3937] lr: 4.6438e-05 eta: 2:49:13 time: 0.5262 data_time: 0.0009 memory: 6318 loss: 0.1462 +2023/06/06 08:56:22 - mmengine - INFO - Epoch(train) [6][2500/3937] lr: 4.6086e-05 eta: 2:48:10 time: 0.4997 data_time: 0.0009 memory: 6318 loss: 0.1326 +2023/06/06 08:57:15 - mmengine - INFO - Epoch(train) [6][2600/3937] lr: 4.5734e-05 eta: 2:47:07 time: 0.5210 data_time: 0.0008 memory: 6318 loss: 0.1429 +2023/06/06 08:58:09 - mmengine - INFO - Epoch(train) [6][2700/3937] lr: 4.5383e-05 eta: 2:46:05 time: 0.5254 data_time: 0.0009 memory: 6318 loss: 0.1457 +2023/06/06 08:59:02 - mmengine - INFO - Epoch(train) [6][2800/3937] lr: 4.5033e-05 eta: 2:45:01 time: 0.4974 data_time: 0.0009 memory: 6318 loss: 0.1358 +2023/06/06 08:59:58 - mmengine - INFO - Epoch(train) [6][2900/3937] lr: 4.4683e-05 eta: 2:44:01 time: 0.5563 data_time: 0.0009 memory: 6318 loss: 0.1353 +2023/06/06 09:00:52 - mmengine - INFO - Epoch(train) [6][3000/3937] lr: 4.4334e-05 eta: 2:42:58 time: 0.5399 data_time: 0.0010 memory: 6318 loss: 0.1328 +2023/06/06 09:01:45 - mmengine - INFO - Epoch(train) [6][3100/3937] lr: 4.3985e-05 eta: 2:41:56 time: 0.5462 data_time: 0.0010 memory: 6318 loss: 0.1218 +2023/06/06 09:02:38 - mmengine - INFO - Epoch(train) [6][3200/3937] lr: 4.3637e-05 eta: 2:40:53 time: 0.5394 data_time: 0.0009 memory: 6318 loss: 0.1382 +2023/06/06 09:03:32 - mmengine - INFO - Epoch(train) [6][3300/3937] lr: 4.3290e-05 eta: 2:39:52 time: 0.5260 data_time: 0.0009 memory: 6318 loss: 0.1435 +2023/06/06 09:03:42 - mmengine - INFO - Exp name: resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1_20230606_051509 +2023/06/06 09:04:25 - mmengine - INFO - Epoch(train) [6][3400/3937] lr: 4.2944e-05 eta: 2:38:49 time: 0.5178 data_time: 0.0009 memory: 6318 loss: 0.1359 +2023/06/06 09:05:19 - mmengine - INFO - Epoch(train) [6][3500/3937] lr: 4.2598e-05 eta: 2:37:47 time: 0.5816 data_time: 0.0009 memory: 6318 loss: 0.1336 +2023/06/06 09:06:14 - mmengine - INFO - Epoch(train) [6][3600/3937] lr: 4.2253e-05 eta: 2:36:47 time: 0.5552 data_time: 0.0009 memory: 6318 loss: 0.1295 +2023/06/06 09:07:07 - mmengine - INFO - Epoch(train) [6][3700/3937] lr: 4.1909e-05 eta: 2:35:44 time: 0.5511 data_time: 0.0009 memory: 6318 loss: 0.1377 +2023/06/06 09:07:59 - mmengine - INFO - Epoch(train) [6][3800/3937] lr: 4.1566e-05 eta: 2:34:42 time: 0.5127 data_time: 0.0010 memory: 6318 loss: 0.1415 +2023/06/06 09:08:53 - mmengine - INFO - Epoch(train) [6][3900/3937] lr: 4.1224e-05 eta: 2:33:40 time: 0.5430 data_time: 0.0009 memory: 6318 loss: 0.1289 +2023/06/06 09:09:13 - mmengine - INFO - Exp name: resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1_20230606_051509 +2023/06/06 09:09:13 - mmengine - INFO - Saving checkpoint at 6 epochs +2023/06/06 09:09:53 - mmengine - INFO - Epoch(val) [6][57/57] accuracy/top1: 94.4191 data_time: 0.4195 time: 0.5064 +2023/06/06 09:10:49 - mmengine - INFO - Epoch(train) [7][ 100/3937] lr: 4.0757e-05 eta: 2:32:17 time: 0.5188 data_time: 0.0112 memory: 6318 loss: 0.1292 +2023/06/06 09:11:44 - mmengine - INFO - Epoch(train) [7][ 200/3937] lr: 4.0416e-05 eta: 2:31:16 time: 0.5394 data_time: 0.0008 memory: 6318 loss: 0.1331 +2023/06/06 09:12:36 - mmengine - INFO - Epoch(train) [7][ 300/3937] lr: 4.0077e-05 eta: 2:30:14 time: 0.5402 data_time: 0.0009 memory: 6318 loss: 0.1257 +2023/06/06 09:13:18 - mmengine - INFO - Exp name: resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1_20230606_051509 +2023/06/06 09:13:29 - mmengine - INFO - Epoch(train) [7][ 400/3937] lr: 3.9739e-05 eta: 2:29:12 time: 0.5293 data_time: 0.0010 memory: 6318 loss: 0.1323 +2023/06/06 09:14:24 - mmengine - INFO - Epoch(train) [7][ 500/3937] lr: 3.9402e-05 eta: 2:28:12 time: 0.5393 data_time: 0.0010 memory: 6318 loss: 0.1349 +2023/06/06 09:15:19 - mmengine - INFO - Epoch(train) [7][ 600/3937] lr: 3.9065e-05 eta: 2:27:12 time: 0.5323 data_time: 0.0009 memory: 6318 loss: 0.1317 +2023/06/06 09:16:13 - mmengine - INFO - Epoch(train) [7][ 700/3937] lr: 3.8730e-05 eta: 2:26:11 time: 0.5518 data_time: 0.0008 memory: 6318 loss: 0.1390 +2023/06/06 09:17:06 - mmengine - INFO - Epoch(train) [7][ 800/3937] lr: 3.8396e-05 eta: 2:25:09 time: 0.5597 data_time: 0.0009 memory: 6318 loss: 0.1329 +2023/06/06 09:18:00 - mmengine - INFO - Epoch(train) [7][ 900/3937] lr: 3.8062e-05 eta: 2:24:08 time: 0.5361 data_time: 0.0008 memory: 6318 loss: 0.1310 +2023/06/06 09:18:52 - mmengine - INFO - Epoch(train) [7][1000/3937] lr: 3.7730e-05 eta: 2:23:06 time: 0.5403 data_time: 0.0009 memory: 6318 loss: 0.1411 +2023/06/06 09:19:49 - mmengine - INFO - Epoch(train) [7][1100/3937] lr: 3.7399e-05 eta: 2:22:07 time: 0.5777 data_time: 0.0009 memory: 6318 loss: 0.1303 +2023/06/06 09:20:42 - mmengine - INFO - Epoch(train) [7][1200/3937] lr: 3.7069e-05 eta: 2:21:06 time: 0.5675 data_time: 0.0012 memory: 6318 loss: 0.1454 +2023/06/06 09:21:35 - mmengine - INFO - Epoch(train) [7][1300/3937] lr: 3.6741e-05 eta: 2:20:05 time: 0.5143 data_time: 0.0009 memory: 6318 loss: 0.1293 +2023/06/06 09:22:16 - mmengine - INFO - Exp name: resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1_20230606_051509 +2023/06/06 09:22:27 - mmengine - INFO - Epoch(train) [7][1400/3937] lr: 3.6413e-05 eta: 2:19:03 time: 0.5373 data_time: 0.0009 memory: 6318 loss: 0.1435 +2023/06/06 09:23:21 - mmengine - INFO - Epoch(train) [7][1500/3937] lr: 3.6087e-05 eta: 2:18:02 time: 0.4947 data_time: 0.0008 memory: 6318 loss: 0.1324 +2023/06/06 09:24:21 - mmengine - INFO - Epoch(train) [7][1600/3937] lr: 3.5761e-05 eta: 2:17:05 time: 0.5269 data_time: 0.0010 memory: 6318 loss: 0.1382 +2023/06/06 09:25:23 - mmengine - INFO - Epoch(train) [7][1700/3937] lr: 3.5437e-05 eta: 2:16:10 time: 0.4827 data_time: 0.0009 memory: 6318 loss: 0.1507 +2023/06/06 09:26:20 - mmengine - INFO - Epoch(train) [7][1800/3937] lr: 3.5115e-05 eta: 2:15:10 time: 0.5382 data_time: 0.0010 memory: 6318 loss: 0.1317 +2023/06/06 09:27:12 - mmengine - INFO - Epoch(train) [7][1900/3937] lr: 3.4793e-05 eta: 2:14:09 time: 0.5494 data_time: 0.0009 memory: 6318 loss: 0.1275 +2023/06/06 09:28:05 - mmengine - INFO - Epoch(train) [7][2000/3937] lr: 3.4473e-05 eta: 2:13:08 time: 0.5356 data_time: 0.0008 memory: 6318 loss: 0.1354 +2023/06/06 09:28:57 - mmengine - INFO - Epoch(train) [7][2100/3937] lr: 3.4154e-05 eta: 2:12:07 time: 0.4970 data_time: 0.0009 memory: 6318 loss: 0.1335 +2023/06/06 09:29:50 - mmengine - INFO - Epoch(train) [7][2200/3937] lr: 3.3836e-05 eta: 2:11:06 time: 0.5666 data_time: 0.0009 memory: 6318 loss: 0.1418 +2023/06/06 09:30:45 - mmengine - INFO - Epoch(train) [7][2300/3937] lr: 3.3520e-05 eta: 2:10:07 time: 0.5835 data_time: 0.0012 memory: 6318 loss: 0.1448 +2023/06/06 09:31:28 - mmengine - INFO - Exp name: resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1_20230606_051509 +2023/06/06 09:31:38 - mmengine - INFO - Epoch(train) [7][2400/3937] lr: 3.3205e-05 eta: 2:09:06 time: 0.5074 data_time: 0.0008 memory: 6318 loss: 0.1481 +2023/06/06 09:32:32 - mmengine - INFO - Epoch(train) [7][2500/3937] lr: 3.2892e-05 eta: 2:08:06 time: 0.5074 data_time: 0.0008 memory: 6318 loss: 0.1255 +2023/06/06 09:33:24 - mmengine - INFO - Epoch(train) [7][2600/3937] lr: 3.2580e-05 eta: 2:07:05 time: 0.5433 data_time: 0.0009 memory: 6318 loss: 0.1125 +2023/06/06 09:34:19 - mmengine - INFO - Epoch(train) [7][2700/3937] lr: 3.2269e-05 eta: 2:06:05 time: 0.5444 data_time: 0.0009 memory: 6318 loss: 0.1457 +2023/06/06 09:35:11 - mmengine - INFO - Epoch(train) [7][2800/3937] lr: 3.1960e-05 eta: 2:05:04 time: 0.4820 data_time: 0.0009 memory: 6318 loss: 0.1288 +2023/06/06 09:36:05 - mmengine - INFO - Epoch(train) [7][2900/3937] lr: 3.1652e-05 eta: 2:04:04 time: 0.5225 data_time: 0.0009 memory: 6318 loss: 0.1380 +2023/06/06 09:36:56 - mmengine - INFO - Epoch(train) [7][3000/3937] lr: 3.1346e-05 eta: 2:03:03 time: 0.5135 data_time: 0.0009 memory: 6318 loss: 0.1177 +2023/06/06 09:37:48 - mmengine - INFO - Epoch(train) [7][3100/3937] lr: 3.1041e-05 eta: 2:02:02 time: 0.5220 data_time: 0.0009 memory: 6318 loss: 0.1315 +2023/06/06 09:38:41 - mmengine - INFO - Epoch(train) [7][3200/3937] lr: 3.0738e-05 eta: 2:01:02 time: 0.5505 data_time: 0.0009 memory: 6318 loss: 0.1470 +2023/06/06 09:39:35 - mmengine - INFO - Epoch(train) [7][3300/3937] lr: 3.0437e-05 eta: 2:00:03 time: 0.4891 data_time: 0.0009 memory: 6318 loss: 0.1254 +2023/06/06 09:40:16 - mmengine - INFO - Exp name: resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1_20230606_051509 +2023/06/06 09:40:27 - mmengine - INFO - Epoch(train) [7][3400/3937] lr: 3.0136e-05 eta: 1:59:02 time: 0.4903 data_time: 0.0009 memory: 6318 loss: 0.1375 +2023/06/06 09:41:23 - mmengine - INFO - Epoch(train) [7][3500/3937] lr: 2.9838e-05 eta: 1:58:03 time: 0.5489 data_time: 0.0008 memory: 6318 loss: 0.1491 +2023/06/06 09:42:15 - mmengine - INFO - Epoch(train) [7][3600/3937] lr: 2.9541e-05 eta: 1:57:03 time: 0.4885 data_time: 0.0008 memory: 6318 loss: 0.1285 +2023/06/06 09:43:08 - mmengine - INFO - Epoch(train) [7][3700/3937] lr: 2.9246e-05 eta: 1:56:03 time: 0.5033 data_time: 0.0009 memory: 6318 loss: 0.1362 +2023/06/06 09:44:01 - mmengine - INFO - Epoch(train) [7][3800/3937] lr: 2.8952e-05 eta: 1:55:03 time: 0.5200 data_time: 0.0009 memory: 6318 loss: 0.1217 +2023/06/06 09:44:53 - mmengine - INFO - Epoch(train) [7][3900/3937] lr: 2.8660e-05 eta: 1:54:03 time: 0.5441 data_time: 0.0009 memory: 6318 loss: 0.1232 +2023/06/06 09:45:12 - mmengine - INFO - Exp name: resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1_20230606_051509 +2023/06/06 09:45:12 - mmengine - INFO - Saving checkpoint at 7 epochs +2023/06/06 09:45:52 - mmengine - INFO - Epoch(val) [7][57/57] accuracy/top1: 94.5653 data_time: 0.4192 time: 0.5061 +2023/06/06 09:46:47 - mmengine - INFO - Epoch(train) [8][ 100/3937] lr: 2.8263e-05 eta: 1:52:42 time: 0.5189 data_time: 0.1424 memory: 6318 loss: 0.1267 +2023/06/06 09:47:50 - mmengine - INFO - Epoch(train) [8][ 200/3937] lr: 2.7975e-05 eta: 1:51:46 time: 0.5797 data_time: 0.0983 memory: 6318 loss: 0.1330 +2023/06/06 09:48:43 - mmengine - INFO - Epoch(train) [8][ 300/3937] lr: 2.7689e-05 eta: 1:50:46 time: 0.5567 data_time: 0.0008 memory: 6318 loss: 0.1328 +2023/06/06 09:49:34 - mmengine - INFO - Epoch(train) [8][ 400/3937] lr: 2.7404e-05 eta: 1:49:46 time: 0.4748 data_time: 0.0009 memory: 6318 loss: 0.1181 +2023/06/06 09:49:56 - mmengine - INFO - Exp name: resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1_20230606_051509 +2023/06/06 09:50:27 - mmengine - INFO - Epoch(train) [8][ 500/3937] lr: 2.7121e-05 eta: 1:48:46 time: 0.5220 data_time: 0.0009 memory: 6318 loss: 0.1305 +2023/06/06 09:51:20 - mmengine - INFO - Epoch(train) [8][ 600/3937] lr: 2.6840e-05 eta: 1:47:47 time: 0.5292 data_time: 0.0010 memory: 6318 loss: 0.1483 +2023/06/06 09:52:14 - mmengine - INFO - Epoch(train) [8][ 700/3937] lr: 2.6561e-05 eta: 1:46:48 time: 0.5099 data_time: 0.0010 memory: 6318 loss: 0.1250 +2023/06/06 09:53:07 - mmengine - INFO - Epoch(train) [8][ 800/3937] lr: 2.6284e-05 eta: 1:45:48 time: 0.5096 data_time: 0.0009 memory: 6318 loss: 0.1485 +2023/06/06 09:54:01 - mmengine - INFO - Epoch(train) [8][ 900/3937] lr: 2.6008e-05 eta: 1:44:49 time: 0.5155 data_time: 0.0009 memory: 6318 loss: 0.1244 +2023/06/06 09:54:54 - mmengine - INFO - Epoch(train) [8][1000/3937] lr: 2.5735e-05 eta: 1:43:49 time: 0.5150 data_time: 0.0009 memory: 6318 loss: 0.1286 +2023/06/06 09:55:48 - mmengine - INFO - Epoch(train) [8][1100/3937] lr: 2.5463e-05 eta: 1:42:51 time: 0.6032 data_time: 0.0008 memory: 6318 loss: 0.1421 +2023/06/06 09:56:41 - mmengine - INFO - Epoch(train) [8][1200/3937] lr: 2.5193e-05 eta: 1:41:51 time: 0.5605 data_time: 0.0008 memory: 6318 loss: 0.1340 +2023/06/06 09:57:35 - mmengine - INFO - Epoch(train) [8][1300/3937] lr: 2.4925e-05 eta: 1:40:52 time: 0.5194 data_time: 0.0009 memory: 6318 loss: 0.1375 +2023/06/06 09:58:29 - mmengine - INFO - Epoch(train) [8][1400/3937] lr: 2.4659e-05 eta: 1:39:53 time: 0.5091 data_time: 0.0012 memory: 6318 loss: 0.1490 +2023/06/06 09:58:50 - mmengine - INFO - Exp name: resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1_20230606_051509 +2023/06/06 09:59:21 - mmengine - INFO - Epoch(train) [8][1500/3937] lr: 2.4394e-05 eta: 1:38:54 time: 0.5405 data_time: 0.0010 memory: 6318 loss: 0.1342 +2023/06/06 10:00:14 - mmengine - INFO - Epoch(train) [8][1600/3937] lr: 2.4132e-05 eta: 1:37:55 time: 0.5230 data_time: 0.0009 memory: 6318 loss: 0.1352 +2023/06/06 10:01:06 - mmengine - INFO - Epoch(train) [8][1700/3937] lr: 2.3872e-05 eta: 1:36:55 time: 0.5531 data_time: 0.0009 memory: 6318 loss: 0.1209 +2023/06/06 10:02:00 - mmengine - INFO - Epoch(train) [8][1800/3937] lr: 2.3613e-05 eta: 1:35:56 time: 0.5433 data_time: 0.0009 memory: 6318 loss: 0.1342 +2023/06/06 10:02:53 - mmengine - INFO - Epoch(train) [8][1900/3937] lr: 2.3357e-05 eta: 1:34:58 time: 0.5657 data_time: 0.0009 memory: 6318 loss: 0.1345 +2023/06/06 10:03:47 - mmengine - INFO - Epoch(train) [8][2000/3937] lr: 2.3103e-05 eta: 1:33:59 time: 0.5370 data_time: 0.0010 memory: 6318 loss: 0.1330 +2023/06/06 10:04:42 - mmengine - INFO - Epoch(train) [8][2100/3937] lr: 2.2851e-05 eta: 1:33:00 time: 0.5332 data_time: 0.0011 memory: 6318 loss: 0.1571 +2023/06/06 10:05:35 - mmengine - INFO - Epoch(train) [8][2200/3937] lr: 2.2600e-05 eta: 1:32:02 time: 0.5693 data_time: 0.0009 memory: 6318 loss: 0.1248 +2023/06/06 10:06:29 - mmengine - INFO - Epoch(train) [8][2300/3937] lr: 2.2352e-05 eta: 1:31:03 time: 0.5215 data_time: 0.0009 memory: 6318 loss: 0.1259 +2023/06/06 10:07:20 - mmengine - INFO - Epoch(train) [8][2400/3937] lr: 2.2106e-05 eta: 1:30:04 time: 0.4952 data_time: 0.0008 memory: 6318 loss: 0.1369 +2023/06/06 10:07:41 - mmengine - INFO - Exp name: resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1_20230606_051509 +2023/06/06 10:08:13 - mmengine - INFO - Epoch(train) [8][2500/3937] lr: 2.1862e-05 eta: 1:29:05 time: 0.5486 data_time: 0.0008 memory: 6318 loss: 0.1288 +2023/06/06 10:09:05 - mmengine - INFO - Epoch(train) [8][2600/3937] lr: 2.1620e-05 eta: 1:28:06 time: 0.5337 data_time: 0.0008 memory: 6318 loss: 0.1295 +2023/06/06 10:09:59 - mmengine - INFO - Epoch(train) [8][2700/3937] lr: 2.1380e-05 eta: 1:27:07 time: 0.5489 data_time: 0.0008 memory: 6318 loss: 0.1199 +2023/06/06 10:10:52 - mmengine - INFO - Epoch(train) [8][2800/3937] lr: 2.1143e-05 eta: 1:26:09 time: 0.4954 data_time: 0.0008 memory: 6318 loss: 0.1314 +2023/06/06 10:11:44 - mmengine - INFO - Epoch(train) [8][2900/3937] lr: 2.0907e-05 eta: 1:25:10 time: 0.5195 data_time: 0.0009 memory: 6318 loss: 0.1098 +2023/06/06 10:12:37 - mmengine - INFO - Epoch(train) [8][3000/3937] lr: 2.0674e-05 eta: 1:24:11 time: 0.5493 data_time: 0.0009 memory: 6318 loss: 0.1238 +2023/06/06 10:13:30 - mmengine - INFO - Epoch(train) [8][3100/3937] lr: 2.0443e-05 eta: 1:23:13 time: 0.5381 data_time: 0.0010 memory: 6318 loss: 0.1322 +2023/06/06 10:14:22 - mmengine - INFO - Epoch(train) [8][3200/3937] lr: 2.0214e-05 eta: 1:22:14 time: 0.5168 data_time: 0.0009 memory: 6318 loss: 0.1291 +2023/06/06 10:15:15 - mmengine - INFO - Epoch(train) [8][3300/3937] lr: 1.9987e-05 eta: 1:21:15 time: 0.5395 data_time: 0.0010 memory: 6318 loss: 0.1437 +2023/06/06 10:16:08 - mmengine - INFO - Epoch(train) [8][3400/3937] lr: 1.9763e-05 eta: 1:20:17 time: 0.5408 data_time: 0.0008 memory: 6318 loss: 0.1322 +2023/06/06 10:16:29 - mmengine - INFO - Exp name: resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1_20230606_051509 +2023/06/06 10:17:01 - mmengine - INFO - Epoch(train) [8][3500/3937] lr: 1.9541e-05 eta: 1:19:18 time: 0.5330 data_time: 0.0009 memory: 6318 loss: 0.1139 +2023/06/06 10:17:55 - mmengine - INFO - Epoch(train) [8][3600/3937] lr: 1.9321e-05 eta: 1:18:20 time: 0.5538 data_time: 0.0008 memory: 6318 loss: 0.1422 +2023/06/06 10:18:48 - mmengine - INFO - Epoch(train) [8][3700/3937] lr: 1.9103e-05 eta: 1:17:22 time: 0.5306 data_time: 0.0009 memory: 6318 loss: 0.1313 +2023/06/06 10:19:41 - mmengine - INFO - Epoch(train) [8][3800/3937] lr: 1.8888e-05 eta: 1:16:23 time: 0.5093 data_time: 0.0009 memory: 6318 loss: 0.1176 +2023/06/06 10:20:34 - mmengine - INFO - Epoch(train) [8][3900/3937] lr: 1.8675e-05 eta: 1:15:25 time: 0.5236 data_time: 0.0013 memory: 6318 loss: 0.1347 +2023/06/06 10:20:54 - mmengine - INFO - Exp name: resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1_20230606_051509 +2023/06/06 10:20:54 - mmengine - INFO - Saving checkpoint at 8 epochs +2023/06/06 10:21:34 - mmengine - INFO - Epoch(val) [8][57/57] accuracy/top1: 94.8819 data_time: 0.4241 time: 0.5100 +2023/06/06 10:22:29 - mmengine - INFO - Epoch(train) [9][ 100/3937] lr: 1.8386e-05 eta: 1:14:06 time: 0.5421 data_time: 0.2025 memory: 6318 loss: 0.1304 +2023/06/06 10:23:26 - mmengine - INFO - Epoch(train) [9][ 200/3937] lr: 1.8179e-05 eta: 1:13:09 time: 0.5205 data_time: 0.0011 memory: 6318 loss: 0.1482 +2023/06/06 10:24:19 - mmengine - INFO - Epoch(train) [9][ 300/3937] lr: 1.7974e-05 eta: 1:12:11 time: 0.5717 data_time: 0.0007 memory: 6318 loss: 0.1405 +2023/06/06 10:25:20 - mmengine - INFO - Epoch(train) [9][ 400/3937] lr: 1.7771e-05 eta: 1:11:14 time: 0.6035 data_time: 0.0008 memory: 6318 loss: 0.1273 +2023/06/06 10:26:16 - mmengine - INFO - Epoch(train) [9][ 500/3937] lr: 1.7570e-05 eta: 1:10:17 time: 0.5485 data_time: 0.0008 memory: 6318 loss: 0.1331 +2023/06/06 10:26:21 - mmengine - INFO - Exp name: resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1_20230606_051509 +2023/06/06 10:27:16 - mmengine - INFO - Epoch(train) [9][ 600/3937] lr: 1.7372e-05 eta: 1:09:20 time: 0.6123 data_time: 0.0009 memory: 6318 loss: 0.1296 +2023/06/06 10:28:15 - mmengine - INFO - Epoch(train) [9][ 700/3937] lr: 1.7176e-05 eta: 1:08:23 time: 1.0484 data_time: 0.0008 memory: 6318 loss: 0.1255 +2023/06/06 10:29:09 - mmengine - INFO - Epoch(train) [9][ 800/3937] lr: 1.6983e-05 eta: 1:07:26 time: 0.5582 data_time: 0.0008 memory: 6318 loss: 0.1302 +2023/06/06 10:30:03 - mmengine - INFO - Epoch(train) [9][ 900/3937] lr: 1.6792e-05 eta: 1:06:28 time: 0.5453 data_time: 0.0008 memory: 6318 loss: 0.1210 +2023/06/06 10:30:59 - mmengine - INFO - Epoch(train) [9][1000/3937] lr: 1.6604e-05 eta: 1:05:30 time: 0.5354 data_time: 0.0008 memory: 6318 loss: 0.1339 +2023/06/06 10:31:52 - mmengine - INFO - Epoch(train) [9][1100/3937] lr: 1.6418e-05 eta: 1:04:32 time: 0.5070 data_time: 0.0008 memory: 6318 loss: 0.1308 +2023/06/06 10:32:46 - mmengine - INFO - Epoch(train) [9][1200/3937] lr: 1.6234e-05 eta: 1:03:34 time: 0.5148 data_time: 0.0008 memory: 6318 loss: 0.1205 +2023/06/06 10:33:40 - mmengine - INFO - Epoch(train) [9][1300/3937] lr: 1.6053e-05 eta: 1:02:37 time: 0.5323 data_time: 0.0008 memory: 6318 loss: 0.1348 +2023/06/06 10:34:32 - mmengine - INFO - Epoch(train) [9][1400/3937] lr: 1.5874e-05 eta: 1:01:38 time: 0.5126 data_time: 0.0008 memory: 6318 loss: 0.1238 +2023/06/06 10:35:25 - mmengine - INFO - Epoch(train) [9][1500/3937] lr: 1.5698e-05 eta: 1:00:41 time: 0.5361 data_time: 0.0009 memory: 6318 loss: 0.1377 +2023/06/06 10:35:29 - mmengine - INFO - Exp name: resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1_20230606_051509 +2023/06/06 10:36:18 - mmengine - INFO - Epoch(train) [9][1600/3937] lr: 1.5524e-05 eta: 0:59:43 time: 0.5586 data_time: 0.0007 memory: 6318 loss: 0.1396 +2023/06/06 10:37:12 - mmengine - INFO - Epoch(train) [9][1700/3937] lr: 1.5353e-05 eta: 0:58:45 time: 0.5245 data_time: 0.0008 memory: 6318 loss: 0.1172 +2023/06/06 10:38:05 - mmengine - INFO - Epoch(train) [9][1800/3937] lr: 1.5185e-05 eta: 0:57:47 time: 0.5342 data_time: 0.0009 memory: 6318 loss: 0.1274 +2023/06/06 10:38:57 - mmengine - INFO - Epoch(train) [9][1900/3937] lr: 1.5019e-05 eta: 0:56:49 time: 0.5742 data_time: 0.0007 memory: 6318 loss: 0.1315 +2023/06/06 10:39:59 - mmengine - INFO - Epoch(train) [9][2000/3937] lr: 1.4855e-05 eta: 0:55:53 time: 0.5847 data_time: 0.0010 memory: 6318 loss: 0.1492 +2023/06/06 10:40:54 - mmengine - INFO - Epoch(train) [9][2100/3937] lr: 1.4694e-05 eta: 0:54:55 time: 0.5479 data_time: 0.0009 memory: 6318 loss: 0.1379 +2023/06/06 10:41:49 - mmengine - INFO - Epoch(train) [9][2200/3937] lr: 1.4536e-05 eta: 0:53:58 time: 0.5629 data_time: 0.0009 memory: 6318 loss: 0.1211 +2023/06/06 10:42:52 - mmengine - INFO - Epoch(train) [9][2300/3937] lr: 1.4380e-05 eta: 0:53:02 time: 0.5173 data_time: 0.0009 memory: 6318 loss: 0.1101 +2023/06/06 10:43:46 - mmengine - INFO - Epoch(train) [9][2400/3937] lr: 1.4227e-05 eta: 0:52:04 time: 0.5427 data_time: 0.0008 memory: 6318 loss: 0.1227 +2023/06/06 10:44:39 - mmengine - INFO - Epoch(train) [9][2500/3937] lr: 1.4076e-05 eta: 0:51:07 time: 0.5356 data_time: 0.0008 memory: 6318 loss: 0.1317 +2023/06/06 10:44:40 - mmengine - INFO - Exp name: resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1_20230606_051509 +2023/06/06 10:45:32 - mmengine - INFO - Epoch(train) [9][2600/3937] lr: 1.3928e-05 eta: 0:50:09 time: 0.5027 data_time: 0.0009 memory: 6318 loss: 0.1292 +2023/06/06 10:46:25 - mmengine - INFO - Epoch(train) [9][2700/3937] lr: 1.3783e-05 eta: 0:49:11 time: 0.5145 data_time: 0.0008 memory: 6318 loss: 0.1310 +2023/06/06 10:47:19 - mmengine - INFO - Epoch(train) [9][2800/3937] lr: 1.3640e-05 eta: 0:48:14 time: 0.5618 data_time: 0.0008 memory: 6318 loss: 0.1216 +2023/06/06 10:48:12 - mmengine - INFO - Epoch(train) [9][2900/3937] lr: 1.3500e-05 eta: 0:47:16 time: 0.5470 data_time: 0.0009 memory: 6318 loss: 0.1460 +2023/06/06 10:49:08 - mmengine - INFO - Epoch(train) [9][3000/3937] lr: 1.3362e-05 eta: 0:46:19 time: 0.4852 data_time: 0.0008 memory: 6318 loss: 0.1223 +2023/06/06 10:50:02 - mmengine - INFO - Epoch(train) [9][3100/3937] lr: 1.3227e-05 eta: 0:45:21 time: 0.5468 data_time: 0.0009 memory: 6318 loss: 0.1295 +2023/06/06 10:50:56 - mmengine - INFO - Epoch(train) [9][3200/3937] lr: 1.3095e-05 eta: 0:44:24 time: 0.5216 data_time: 0.0008 memory: 6318 loss: 0.1247 +2023/06/06 10:51:50 - mmengine - INFO - Epoch(train) [9][3300/3937] lr: 1.2966e-05 eta: 0:43:27 time: 0.5606 data_time: 0.0009 memory: 6318 loss: 0.1334 +2023/06/06 10:52:43 - mmengine - INFO - Epoch(train) [9][3400/3937] lr: 1.2839e-05 eta: 0:42:29 time: 0.5482 data_time: 0.0009 memory: 6318 loss: 0.1220 +2023/06/06 10:53:36 - mmengine - INFO - Epoch(train) [9][3500/3937] lr: 1.2715e-05 eta: 0:41:32 time: 0.5413 data_time: 0.0008 memory: 6318 loss: 0.1313 +2023/06/06 10:53:37 - mmengine - INFO - Exp name: resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1_20230606_051509 +2023/06/06 10:54:28 - mmengine - INFO - Epoch(train) [9][3600/3937] lr: 1.2593e-05 eta: 0:40:34 time: 0.5148 data_time: 0.0009 memory: 6318 loss: 0.1219 +2023/06/06 10:55:21 - mmengine - INFO - Epoch(train) [9][3700/3937] lr: 1.2474e-05 eta: 0:39:37 time: 0.5154 data_time: 0.0009 memory: 6318 loss: 0.1419 +2023/06/06 10:56:14 - mmengine - INFO - Epoch(train) [9][3800/3937] lr: 1.2358e-05 eta: 0:38:39 time: 0.5051 data_time: 0.0012 memory: 6318 loss: 0.1183 +2023/06/06 10:57:08 - mmengine - INFO - Epoch(train) [9][3900/3937] lr: 1.2245e-05 eta: 0:37:42 time: 0.5408 data_time: 0.0009 memory: 6318 loss: 0.1339 +2023/06/06 10:57:29 - mmengine - INFO - Exp name: resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1_20230606_051509 +2023/06/06 10:57:29 - mmengine - INFO - Saving checkpoint at 9 epochs +2023/06/06 10:58:08 - mmengine - INFO - Epoch(val) [9][57/57] accuracy/top1: 95.1985 data_time: 0.4250 time: 0.5118 +2023/06/06 10:59:04 - mmengine - INFO - Epoch(train) [10][ 100/3937] lr: 1.2094e-05 eta: 0:36:24 time: 0.5395 data_time: 0.1659 memory: 6318 loss: 0.1428 +2023/06/06 10:59:57 - mmengine - INFO - Epoch(train) [10][ 200/3937] lr: 1.1987e-05 eta: 0:35:26 time: 0.5352 data_time: 0.0011 memory: 6318 loss: 0.1319 +2023/06/06 11:00:47 - mmengine - INFO - Epoch(train) [10][ 300/3937] lr: 1.1883e-05 eta: 0:34:29 time: 0.4939 data_time: 0.0009 memory: 6318 loss: 0.1195 +2023/06/06 11:01:40 - mmengine - INFO - Epoch(train) [10][ 400/3937] lr: 1.1781e-05 eta: 0:33:32 time: 0.5419 data_time: 0.0009 memory: 6318 loss: 0.1323 +2023/06/06 11:02:34 - mmengine - INFO - Epoch(train) [10][ 500/3937] lr: 1.1683e-05 eta: 0:32:34 time: 0.5449 data_time: 0.0009 memory: 6318 loss: 0.1271 +2023/06/06 11:03:12 - mmengine - INFO - Exp name: resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1_20230606_051509 +2023/06/06 11:03:28 - mmengine - INFO - Epoch(train) [10][ 600/3937] lr: 1.1587e-05 eta: 0:31:37 time: 0.5249 data_time: 0.0008 memory: 6318 loss: 0.1158 +2023/06/06 11:04:21 - mmengine - INFO - Epoch(train) [10][ 700/3937] lr: 1.1494e-05 eta: 0:30:40 time: 0.5559 data_time: 0.0008 memory: 6318 loss: 0.1198 +2023/06/06 11:05:13 - mmengine - INFO - Epoch(train) [10][ 800/3937] lr: 1.1403e-05 eta: 0:29:43 time: 0.5160 data_time: 0.0009 memory: 6318 loss: 0.1254 +2023/06/06 11:06:10 - mmengine - INFO - Epoch(train) [10][ 900/3937] lr: 1.1316e-05 eta: 0:28:46 time: 0.9006 data_time: 0.0009 memory: 6318 loss: 0.1325 +2023/06/06 11:07:02 - mmengine - INFO - Epoch(train) [10][1000/3937] lr: 1.1231e-05 eta: 0:27:49 time: 0.4761 data_time: 0.0009 memory: 6318 loss: 0.1279 +2023/06/06 11:07:54 - mmengine - INFO - Epoch(train) [10][1100/3937] lr: 1.1149e-05 eta: 0:26:51 time: 0.4900 data_time: 0.0008 memory: 6318 loss: 0.1210 +2023/06/06 11:08:44 - mmengine - INFO - Epoch(train) [10][1200/3937] lr: 1.1070e-05 eta: 0:25:54 time: 0.4979 data_time: 0.0009 memory: 6318 loss: 0.1333 +2023/06/06 11:09:36 - mmengine - INFO - Epoch(train) [10][1300/3937] lr: 1.0993e-05 eta: 0:24:57 time: 0.5278 data_time: 0.0008 memory: 6318 loss: 0.1285 +2023/06/06 11:10:29 - mmengine - INFO - Epoch(train) [10][1400/3937] lr: 1.0920e-05 eta: 0:24:00 time: 0.5615 data_time: 0.0007 memory: 6318 loss: 0.1329 +2023/06/06 11:11:21 - mmengine - INFO - Epoch(train) [10][1500/3937] lr: 1.0849e-05 eta: 0:23:03 time: 0.5142 data_time: 0.0010 memory: 6318 loss: 0.1231 +2023/06/06 11:11:57 - mmengine - INFO - Exp name: resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1_20230606_051509 +2023/06/06 11:12:13 - mmengine - INFO - Epoch(train) [10][1600/3937] lr: 1.0781e-05 eta: 0:22:06 time: 0.4944 data_time: 0.0009 memory: 6318 loss: 0.1282 +2023/06/06 11:13:04 - mmengine - INFO - Epoch(train) [10][1700/3937] lr: 1.0716e-05 eta: 0:21:09 time: 0.5297 data_time: 0.0010 memory: 6318 loss: 0.1526 +2023/06/06 11:13:57 - mmengine - INFO - Epoch(train) [10][1800/3937] lr: 1.0653e-05 eta: 0:20:12 time: 0.5135 data_time: 0.0009 memory: 6318 loss: 0.1091 +2023/06/06 11:14:49 - mmengine - INFO - Epoch(train) [10][1900/3937] lr: 1.0594e-05 eta: 0:19:15 time: 0.5665 data_time: 0.0010 memory: 6318 loss: 0.1348 +2023/06/06 11:15:44 - mmengine - INFO - Epoch(train) [10][2000/3937] lr: 1.0537e-05 eta: 0:18:18 time: 0.5381 data_time: 0.0008 memory: 6318 loss: 0.1401 +2023/06/06 11:16:36 - mmengine - INFO - Epoch(train) [10][2100/3937] lr: 1.0483e-05 eta: 0:17:21 time: 0.5410 data_time: 0.0008 memory: 6318 loss: 0.1243 +2023/06/06 11:17:30 - mmengine - INFO - Epoch(train) [10][2200/3937] lr: 1.0432e-05 eta: 0:16:24 time: 0.5008 data_time: 0.0010 memory: 6318 loss: 0.1417 +2023/06/06 11:18:23 - mmengine - INFO - Epoch(train) [10][2300/3937] lr: 1.0384e-05 eta: 0:15:27 time: 0.5252 data_time: 0.0009 memory: 6318 loss: 0.1196 +2023/06/06 11:19:16 - mmengine - INFO - Epoch(train) [10][2400/3937] lr: 1.0338e-05 eta: 0:14:30 time: 0.5254 data_time: 0.0008 memory: 6318 loss: 0.1327 +2023/06/06 11:20:09 - mmengine - INFO - Epoch(train) [10][2500/3937] lr: 1.0296e-05 eta: 0:13:34 time: 0.5297 data_time: 0.0009 memory: 6318 loss: 0.1197 +2023/06/06 11:20:48 - mmengine - INFO - Exp name: resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1_20230606_051509 +2023/06/06 11:21:04 - mmengine - INFO - Epoch(train) [10][2600/3937] lr: 1.0256e-05 eta: 0:12:37 time: 0.5383 data_time: 0.0008 memory: 6318 loss: 0.1361 +2023/06/06 11:21:57 - mmengine - INFO - Epoch(train) [10][2700/3937] lr: 1.0219e-05 eta: 0:11:40 time: 0.5563 data_time: 0.0007 memory: 6318 loss: 0.1322 +2023/06/06 11:22:49 - mmengine - INFO - Epoch(train) [10][2800/3937] lr: 1.0185e-05 eta: 0:10:43 time: 0.5264 data_time: 0.0008 memory: 6318 loss: 0.1303 +2023/06/06 11:23:43 - mmengine - INFO - Epoch(train) [10][2900/3937] lr: 1.0154e-05 eta: 0:09:47 time: 0.5644 data_time: 0.0007 memory: 6318 loss: 0.1294 +2023/06/06 11:24:35 - mmengine - INFO - Epoch(train) [10][3000/3937] lr: 1.0126e-05 eta: 0:08:50 time: 0.5225 data_time: 0.0008 memory: 6318 loss: 0.1229 +2023/06/06 11:25:28 - mmengine - INFO - Epoch(train) [10][3100/3937] lr: 1.0101e-05 eta: 0:07:53 time: 0.5481 data_time: 0.0007 memory: 6318 loss: 0.1237 +2023/06/06 11:26:20 - mmengine - INFO - Epoch(train) [10][3200/3937] lr: 1.0078e-05 eta: 0:06:57 time: 0.5012 data_time: 0.0010 memory: 6318 loss: 0.1179 +2023/06/06 11:27:12 - mmengine - INFO - Epoch(train) [10][3300/3937] lr: 1.0058e-05 eta: 0:06:00 time: 0.5307 data_time: 0.0009 memory: 6318 loss: 0.1240 +2023/06/06 11:28:14 - mmengine - INFO - Epoch(train) [10][3400/3937] lr: 1.0041e-05 eta: 0:05:03 time: 0.8462 data_time: 0.0010 memory: 6318 loss: 0.1239 +2023/06/06 11:29:08 - mmengine - INFO - Epoch(train) [10][3500/3937] lr: 1.0027e-05 eta: 0:04:07 time: 0.5163 data_time: 0.0009 memory: 6318 loss: 0.1245 +2023/06/06 11:29:49 - mmengine - INFO - Exp name: resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1_20230606_051509 +2023/06/06 11:30:06 - mmengine - INFO - Epoch(train) [10][3600/3937] lr: 1.0016e-05 eta: 0:03:10 time: 0.5723 data_time: 0.0009 memory: 6318 loss: 0.1358 +2023/06/06 11:31:06 - mmengine - INFO - Epoch(train) [10][3700/3937] lr: 1.0008e-05 eta: 0:02:14 time: 0.5714 data_time: 0.0008 memory: 6318 loss: 0.1320 +2023/06/06 11:32:04 - mmengine - INFO - Epoch(train) [10][3800/3937] lr: 1.0003e-05 eta: 0:01:17 time: 0.5763 data_time: 0.0008 memory: 6318 loss: 0.1347 +2023/06/06 11:33:04 - mmengine - INFO - Epoch(train) [10][3900/3937] lr: 1.0000e-05 eta: 0:00:20 time: 0.5147 data_time: 0.0007 memory: 6318 loss: 0.1506 +2023/06/06 11:33:24 - mmengine - INFO - Exp name: resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1_20230606_051509 +2023/06/06 11:33:24 - mmengine - INFO - Saving checkpoint at 10 epochs +2023/06/06 11:34:04 - mmengine - INFO - Epoch(val) [10][57/57] accuracy/top1: 95.4734 data_time: 0.4221 time: 0.5078 diff --git a/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1/20230606_051509/vis_data/20230606_051509.json b/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1/20230606_051509/vis_data/20230606_051509.json new file mode 100644 index 0000000000000000000000000000000000000000..9d49377fbb2d936b6395471c65d52f190cab1533 --- /dev/null +++ b/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1/20230606_051509/vis_data/20230606_051509.json @@ -0,0 +1,400 @@ +{"lr": 9.999859583040624e-05, "data_time": 0.25990498065948486, "loss": 0.6402046918869019, "time": 0.5295006036758423, "epoch": 1, "memory": 9436, "step": 100} +{"lr": 9.999432653394512e-05, "data_time": 0.5439581155776978, "loss": 0.5861357271671295, "time": 0.6861336708068848, "epoch": 1, "memory": 6319, "step": 200} +{"lr": 9.998719223919393e-05, "data_time": 0.5587227582931519, "loss": 0.5367321491241455, "time": 0.6989289999008179, "epoch": 1, "memory": 6319, "step": 300} +{"lr": 9.997719340042649e-05, "data_time": 0.5213950872421265, "loss": 0.48539854884147643, "time": 0.6626511812210083, "epoch": 1, "memory": 6319, "step": 400} +{"lr": 9.996433065431542e-05, "data_time": 0.5269391059875488, "loss": 0.43747425079345703, "time": 0.6674677848815918, "epoch": 1, "memory": 6319, "step": 500} +{"lr": 9.994860481989157e-05, "data_time": 0.5152857303619385, "loss": 0.39237564206123354, "time": 0.6596978187561036, "epoch": 1, "memory": 6319, "step": 600} +{"lr": 9.993001689849214e-05, "data_time": 1.0034260749816895, "loss": 0.3821966916322708, "time": 1.1468387842178345, "epoch": 1, "memory": 6319, "step": 700} +{"lr": 9.990856807369658e-05, "data_time": 0.4322436571121216, "loss": 0.35795452296733854, "time": 0.5726524353027344, "epoch": 1, "memory": 6319, "step": 800} +{"lr": 9.98842597112514e-05, "data_time": 0.48664350509643556, "loss": 0.33374722898006437, "time": 0.6277008056640625, "epoch": 1, "memory": 6319, "step": 900} +{"lr": 9.985709335898336e-05, "data_time": 0.4990594148635864, "loss": 0.32999047040939333, "time": 0.6407262325286865, "epoch": 1, "memory": 6319, "step": 1000} +{"lr": 9.982707074670047e-05, "data_time": 0.4646873712539673, "loss": 0.31953787207603457, "time": 0.6051527976989746, "epoch": 1, "memory": 6319, "step": 1100} +{"lr": 9.979419378608245e-05, "data_time": 0.5044169664382935, "loss": 0.2946313828229904, "time": 0.6464436292648316, "epoch": 1, "memory": 6319, "step": 1200} +{"lr": 9.975846457055832e-05, "data_time": 0.5260628938674927, "loss": 0.29008248448371887, "time": 0.6668855428695679, "epoch": 1, "memory": 6319, "step": 1300} +{"lr": 9.97198853751737e-05, "data_time": 0.5073223114013672, "loss": 0.3029852151870728, "time": 0.6472033977508544, "epoch": 1, "memory": 6319, "step": 1400} +{"lr": 9.967845865644554e-05, "data_time": 0.48812687397003174, "loss": 0.29525449872016907, "time": 0.6277704000473022, "epoch": 1, "memory": 6319, "step": 1500} +{"lr": 9.963418705220597e-05, "data_time": 0.493870210647583, "loss": 0.28693257868289945, "time": 0.6350355625152588, "epoch": 1, "memory": 6319, "step": 1600} +{"lr": 9.958707338143413e-05, "data_time": 0.4661304235458374, "loss": 0.29168210327625277, "time": 0.6068583488464355, "epoch": 1, "memory": 6319, "step": 1700} +{"lr": 9.953712064407689e-05, "data_time": 0.45297794342041015, "loss": 0.29011504650115966, "time": 0.5937061309814453, "epoch": 1, "memory": 6319, "step": 1800} +{"lr": 9.948433202085752e-05, "data_time": 0.5169339418411255, "loss": 0.27398703396320345, "time": 0.6577034235000611, "epoch": 1, "memory": 6319, "step": 1900} +{"lr": 9.942871087307363e-05, "data_time": 0.4975031137466431, "loss": 0.2566560238599777, "time": 0.6394488096237183, "epoch": 1, "memory": 6319, "step": 2000} +{"lr": 9.937026074238262e-05, "data_time": 0.5305297374725342, "loss": 0.2536715030670166, "time": 0.6727217197418213, "epoch": 1, "memory": 6319, "step": 2100} +{"lr": 9.930898535057657e-05, "data_time": 0.3923900365829468, "loss": 0.2435808628797531, "time": 0.5332682132720947, "epoch": 1, "memory": 6319, "step": 2200} +{"lr": 9.924488859934492e-05, "data_time": 0.45818946361541746, "loss": 0.262268927693367, "time": 0.5985460996627807, "epoch": 1, "memory": 6319, "step": 2300} +{"lr": 9.917797457002633e-05, "data_time": 0.42747619152069094, "loss": 0.24827308654785157, "time": 0.5756779193878174, "epoch": 1, "memory": 6319, "step": 2400} +{"lr": 9.910824752334866e-05, "data_time": 0.13895781040191652, "loss": 0.26804320961236955, "time": 0.5948331832885743, "epoch": 1, "memory": 6319, "step": 2500} +{"lr": 9.903571189915766e-05, "data_time": 0.0008750200271606445, "loss": 0.2405380979180336, "time": 0.5901623010635376, "epoch": 1, "memory": 6319, "step": 2600} +{"lr": 9.896037231613424e-05, "data_time": 0.0009298086166381836, "loss": 0.2365088403224945, "time": 0.6120794773101806, "epoch": 1, "memory": 6319, "step": 2700} +{"lr": 9.888223357150049e-05, "data_time": 0.0008474588394165039, "loss": 0.2465743228793144, "time": 0.6035980701446533, "epoch": 1, "memory": 6319, "step": 2800} +{"lr": 9.880130064071408e-05, "data_time": 0.000853586196899414, "loss": 0.23031350076198578, "time": 0.6611725330352783, "epoch": 1, "memory": 6319, "step": 2900} +{"lr": 9.87175786771517e-05, "data_time": 0.000911259651184082, "loss": 0.23938852399587632, "time": 0.6019812822341919, "epoch": 1, "memory": 6319, "step": 3000} +{"lr": 9.863107301178059e-05, "data_time": 0.001007819175720215, "loss": 0.24390990138053895, "time": 0.6488357782363892, "epoch": 1, "memory": 6319, "step": 3100} +{"lr": 9.854178915281925e-05, "data_time": 0.0009762287139892578, "loss": 0.24378345012664795, "time": 0.6051812410354614, "epoch": 1, "memory": 6319, "step": 3200} +{"lr": 9.84497327853868e-05, "data_time": 0.0009861469268798828, "loss": 0.22497162222862244, "time": 0.6643328189849853, "epoch": 1, "memory": 6319, "step": 3300} +{"lr": 9.83549097711407e-05, "data_time": 0.0009825944900512696, "loss": 0.24779136180877687, "time": 0.6440266609191895, "epoch": 1, "memory": 6319, "step": 3400} +{"lr": 9.825732614790392e-05, "data_time": 0.0009270668029785156, "loss": 0.2368963912129402, "time": 0.57118718624115, "epoch": 1, "memory": 6319, "step": 3500} +{"lr": 9.815698812928018e-05, "data_time": 0.0009441852569580078, "loss": 0.21808661967515947, "time": 0.5564011573791504, "epoch": 1, "memory": 6319, "step": 3600} +{"lr": 9.805390210425831e-05, "data_time": 0.0009526252746582031, "loss": 0.21776784658432008, "time": 0.5983808755874633, "epoch": 1, "memory": 6319, "step": 3700} +{"lr": 9.794807463680567e-05, "data_time": 0.0009661197662353515, "loss": 0.21626628786325455, "time": 0.5614094018936158, "epoch": 1, "memory": 6319, "step": 3800} +{"lr": 9.783951246544974e-05, "data_time": 0.000871586799621582, "loss": 0.2112686291337013, "time": 0.5717248439788818, "epoch": 1, "memory": 6319, "step": 3900} +{"accuracy/top1": 60.02574920654297, "data_time": 0.4936740942168654, "time": 0.5865166772875869, "step": 1} +{"lr": 9.768635527139677e-05, "data_time": 0.4531930685043335, "loss": 0.22331490516662597, "time": 0.5931108713150024, "epoch": 2, "memory": 8776, "step": 4037} +{"lr": 9.75713397738286e-05, "data_time": 0.24699089527130128, "loss": 0.21777475923299788, "time": 0.5014456748962403, "epoch": 2, "memory": 6318, "step": 4137} +{"lr": 9.745361356082025e-05, "data_time": 0.0008819341659545898, "loss": 0.22727794647216798, "time": 0.580863094329834, "epoch": 2, "memory": 6318, "step": 4237} +{"lr": 9.733318412854784e-05, "data_time": 0.000957942008972168, "loss": 0.23054793179035188, "time": 0.5685622930526734, "epoch": 2, "memory": 6318, "step": 4337} +{"lr": 9.721005914531412e-05, "data_time": 0.000832819938659668, "loss": 0.21719527691602708, "time": 0.5727081298828125, "epoch": 2, "memory": 6318, "step": 4437} +{"lr": 9.708424645106017e-05, "data_time": 0.0009802818298339845, "loss": 0.2230851247906685, "time": 0.5348587512969971, "epoch": 2, "memory": 6318, "step": 4537} +{"lr": 9.695575405686603e-05, "data_time": 0.0009038209915161133, "loss": 0.2079911008477211, "time": 0.5726252317428588, "epoch": 2, "memory": 6318, "step": 4637} +{"lr": 9.682459014444089e-05, "data_time": 0.0009235620498657226, "loss": 0.22554663568735123, "time": 0.5798112392425537, "epoch": 2, "memory": 6318, "step": 4737} +{"lr": 9.669076306560176e-05, "data_time": 0.0008735418319702149, "loss": 0.22549958974123002, "time": 0.5881304025650025, "epoch": 2, "memory": 6318, "step": 4837} +{"lr": 9.655428134174183e-05, "data_time": 0.0009006261825561523, "loss": 0.21149692237377166, "time": 0.556542444229126, "epoch": 2, "memory": 6318, "step": 4937} +{"lr": 9.641515366328807e-05, "data_time": 0.0008460760116577148, "loss": 0.20487848669290543, "time": 0.5929847478866577, "epoch": 2, "memory": 6318, "step": 5037} +{"lr": 9.627338888914757e-05, "data_time": 0.0009048700332641602, "loss": 0.22542578727006912, "time": 0.5736958742141723, "epoch": 2, "memory": 6318, "step": 5137} +{"lr": 9.612899604614378e-05, "data_time": 0.0009108781814575195, "loss": 0.20576413869857788, "time": 0.5921751260757446, "epoch": 2, "memory": 6318, "step": 5237} +{"lr": 9.598198432844131e-05, "data_time": 0.0009085416793823242, "loss": 0.20128974169492722, "time": 0.557023549079895, "epoch": 2, "memory": 6318, "step": 5337} +{"lr": 9.583236309696081e-05, "data_time": 0.0008863210678100586, "loss": 0.19630965441465378, "time": 0.551726245880127, "epoch": 2, "memory": 6318, "step": 5437} +{"lr": 9.568014187878279e-05, "data_time": 0.0008133649826049805, "loss": 0.20126755982637407, "time": 0.5702857255935669, "epoch": 2, "memory": 6318, "step": 5537} +{"lr": 9.552533036654103e-05, "data_time": 0.0008449077606201172, "loss": 0.21352028101682663, "time": 0.60773766040802, "epoch": 2, "memory": 6318, "step": 5637} +{"lr": 9.536793841780541e-05, "data_time": 0.000992894172668457, "loss": 0.20573764741420747, "time": 0.5825971126556396, "epoch": 2, "memory": 6318, "step": 5737} +{"lr": 9.52079760544543e-05, "data_time": 0.0008639097213745117, "loss": 0.21162491738796235, "time": 0.5742067098617554, "epoch": 2, "memory": 6318, "step": 5837} +{"lr": 9.504545346203619e-05, "data_time": 0.0008871793746948242, "loss": 0.18542599380016328, "time": 0.5614900112152099, "epoch": 2, "memory": 6318, "step": 5937} +{"lr": 9.488038098912122e-05, "data_time": 0.000901341438293457, "loss": 0.2020304471254349, "time": 0.74519784450531, "epoch": 2, "memory": 6318, "step": 6037} +{"lr": 9.471276914664257e-05, "data_time": 0.0009241104125976562, "loss": 0.2121737465262413, "time": 0.5834649085998536, "epoch": 2, "memory": 6318, "step": 6137} +{"lr": 9.454262860722667e-05, "data_time": 0.001154470443725586, "loss": 0.20481036901473998, "time": 0.5199038028717041, "epoch": 2, "memory": 6318, "step": 6237} +{"lr": 9.436997020451398e-05, "data_time": 0.0009418964385986328, "loss": 0.2028192847967148, "time": 0.5328365802764893, "epoch": 2, "memory": 6318, "step": 6337} +{"lr": 9.419480493246903e-05, "data_time": 0.0009444713592529297, "loss": 0.18822189718484877, "time": 0.5861220359802246, "epoch": 2, "memory": 6318, "step": 6437} +{"lr": 9.401714394468036e-05, "data_time": 0.0008361577987670898, "loss": 0.19415918588638306, "time": 0.5689403057098389, "epoch": 2, "memory": 6318, "step": 6537} +{"lr": 9.383699855365022e-05, "data_time": 0.000895547866821289, "loss": 0.21370866596698762, "time": 0.542294192314148, "epoch": 2, "memory": 6318, "step": 6637} +{"lr": 9.365438023007469e-05, "data_time": 0.0010315656661987304, "loss": 0.20446225702762605, "time": 0.5685359239578247, "epoch": 2, "memory": 6318, "step": 6737} +{"lr": 9.346930060211267e-05, "data_time": 0.0009064674377441406, "loss": 0.2243996813893318, "time": 0.54306640625, "epoch": 2, "memory": 6318, "step": 6837} +{"lr": 9.328177145464599e-05, "data_time": 0.0009489059448242188, "loss": 0.19088304787874222, "time": 0.5704697847366333, "epoch": 2, "memory": 6318, "step": 6937} +{"lr": 9.309180472852871e-05, "data_time": 0.0008680343627929688, "loss": 0.20378941744565965, "time": 0.585848617553711, "epoch": 2, "memory": 6318, "step": 7037} +{"lr": 9.289941251982683e-05, "data_time": 0.0009253025054931641, "loss": 0.1844475522637367, "time": 0.5599412202835083, "epoch": 2, "memory": 6318, "step": 7137} +{"lr": 9.270460707904827e-05, "data_time": 0.0009270906448364258, "loss": 0.18135594725608825, "time": 0.5775460243225098, "epoch": 2, "memory": 6318, "step": 7237} +{"lr": 9.250740081036243e-05, "data_time": 0.000993657112121582, "loss": 0.18333010673522948, "time": 0.559870719909668, "epoch": 2, "memory": 6318, "step": 7337} +{"lr": 9.230780627081088e-05, "data_time": 0.0009291172027587891, "loss": 0.17100572139024733, "time": 0.5388106346130371, "epoch": 2, "memory": 6318, "step": 7437} +{"lr": 9.210583616950722e-05, "data_time": 0.0008291482925415039, "loss": 0.21008660644292831, "time": 0.5797322034835816, "epoch": 2, "memory": 6318, "step": 7537} +{"lr": 9.190150336682842e-05, "data_time": 0.0008919954299926758, "loss": 0.1646827444434166, "time": 0.5979239702224731, "epoch": 2, "memory": 6318, "step": 7637} +{"lr": 9.16948208735955e-05, "data_time": 0.000911879539489746, "loss": 0.18441036492586135, "time": 0.5740241050720215, "epoch": 2, "memory": 6318, "step": 7737} +{"lr": 9.148580185024521e-05, "data_time": 0.0010335922241210937, "loss": 0.18545248359441757, "time": 0.605552339553833, "epoch": 2, "memory": 6318, "step": 7837} +{"accuracy/top1": 84.0089111328125, "data_time": 0.4414039809128334, "time": 0.5312507851370449, "step": 2} +{"lr": 9.119567684509738e-05, "data_time": 0.31747140884399416, "loss": 0.18982364982366562, "time": 0.5918580055236816, "epoch": 3, "memory": 6318, "step": 7974} +{"lr": 9.098117366468421e-05, "data_time": 0.17231969833374022, "loss": 0.17452817857265474, "time": 1.284656810760498, "epoch": 3, "memory": 6318, "step": 8074} +{"lr": 9.076437939539471e-05, "data_time": 0.0007827043533325196, "loss": 0.17449144423007965, "time": 0.546287226676941, "epoch": 3, "memory": 6318, "step": 8174} +{"lr": 9.05453078415294e-05, "data_time": 0.0009041309356689453, "loss": 0.18233867287635802, "time": 0.592623257637024, "epoch": 3, "memory": 6318, "step": 8274} +{"lr": 9.032397295239427e-05, "data_time": 0.0011610031127929688, "loss": 0.1747233659029007, "time": 0.5693724393844605, "epoch": 3, "memory": 6318, "step": 8374} +{"lr": 9.010038882141241e-05, "data_time": 0.0008029460906982422, "loss": 0.1854666903614998, "time": 0.5913819551467896, "epoch": 3, "memory": 6318, "step": 8474} +{"lr": 8.987456968522655e-05, "data_time": 0.0010127544403076172, "loss": 0.19007351696491243, "time": 0.5473955869674683, "epoch": 3, "memory": 6318, "step": 8574} +{"lr": 8.964652992279262e-05, "data_time": 0.0008792638778686524, "loss": 0.18883423656225204, "time": 0.578046727180481, "epoch": 3, "memory": 6318, "step": 8674} +{"lr": 8.941628405446428e-05, "data_time": 0.0009624004364013672, "loss": 0.173228719830513, "time": 0.5761780977249146, "epoch": 3, "memory": 6318, "step": 8774} +{"lr": 8.918384674106797e-05, "data_time": 0.0008363723754882812, "loss": 0.17672196328639983, "time": 0.5682754516601562, "epoch": 3, "memory": 6318, "step": 8874} +{"lr": 8.894923278296992e-05, "data_time": 0.0009339570999145508, "loss": 0.2008403167128563, "time": 0.5964117050170898, "epoch": 3, "memory": 6318, "step": 8974} +{"lr": 8.871245711913324e-05, "data_time": 0.0011966705322265625, "loss": 0.17237788140773774, "time": 0.5893830299377442, "epoch": 3, "memory": 6318, "step": 9074} +{"lr": 8.847353482616703e-05, "data_time": 0.0009474277496337891, "loss": 0.17052201330661773, "time": 0.5614013195037841, "epoch": 3, "memory": 6318, "step": 9174} +{"lr": 8.823248111736638e-05, "data_time": 0.0008767843246459961, "loss": 0.17628194242715836, "time": 0.5740657806396484, "epoch": 3, "memory": 6318, "step": 9274} +{"lr": 8.798931134174327e-05, "data_time": 0.0008599519729614258, "loss": 0.16591138392686844, "time": 0.5469395637512207, "epoch": 3, "memory": 6318, "step": 9374} +{"lr": 8.774404098304961e-05, "data_time": 0.001050257682800293, "loss": 0.1798093795776367, "time": 0.5934871435165405, "epoch": 3, "memory": 6318, "step": 9474} +{"lr": 8.749668565879136e-05, "data_time": 0.000908970832824707, "loss": 0.17558502554893493, "time": 0.551546573638916, "epoch": 3, "memory": 6318, "step": 9574} +{"lr": 8.724726111923386e-05, "data_time": 0.0009049654006958007, "loss": 0.17216941863298416, "time": 0.5799022197723389, "epoch": 3, "memory": 6318, "step": 9674} +{"lr": 8.699578324639898e-05, "data_time": 0.0008556842803955078, "loss": 0.16674857661128045, "time": 0.5886175870895386, "epoch": 3, "memory": 6318, "step": 9774} +{"lr": 8.67422680530539e-05, "data_time": 0.0009267807006835937, "loss": 0.1592809021472931, "time": 0.569447112083435, "epoch": 3, "memory": 6318, "step": 9874} +{"lr": 8.648673168169147e-05, "data_time": 0.0008530139923095704, "loss": 0.17575456351041793, "time": 0.5851914882659912, "epoch": 3, "memory": 6318, "step": 9974} +{"lr": 8.622919040350244e-05, "data_time": 0.0008959293365478516, "loss": 0.17296219244599342, "time": 0.6134592533111572, "epoch": 3, "memory": 6318, "step": 10074} +{"lr": 8.596966061733922e-05, "data_time": 0.0010569334030151368, "loss": 0.18186536133289338, "time": 0.5560044288635254, "epoch": 3, "memory": 6318, "step": 10174} +{"lr": 8.570815884867185e-05, "data_time": 0.0008459329605102539, "loss": 0.17924930304288864, "time": 0.5501826047897339, "epoch": 3, "memory": 6318, "step": 10274} +{"lr": 8.544470174853563e-05, "data_time": 0.0011171579360961914, "loss": 0.16258630454540252, "time": 0.6082868337631225, "epoch": 3, "memory": 6318, "step": 10374} +{"lr": 8.517930609247077e-05, "data_time": 0.0008495330810546875, "loss": 0.17577800005674363, "time": 0.5666185617446899, "epoch": 3, "memory": 6318, "step": 10474} +{"lr": 8.491198877945477e-05, "data_time": 0.0009595394134521484, "loss": 0.17118353992700577, "time": 0.6048997640609741, "epoch": 3, "memory": 6318, "step": 10574} +{"lr": 8.464276683082565e-05, "data_time": 0.0010280370712280273, "loss": 0.1630304455757141, "time": 0.5279735803604126, "epoch": 3, "memory": 6318, "step": 10674} +{"lr": 8.437165738919852e-05, "data_time": 0.0011461019515991212, "loss": 0.17769434154033661, "time": 0.5584010124206543, "epoch": 3, "memory": 6318, "step": 10774} +{"lr": 8.409867771737398e-05, "data_time": 0.0008713245391845703, "loss": 0.15921278446912765, "time": 0.5352803945541382, "epoch": 3, "memory": 6318, "step": 10874} +{"lr": 8.382384519723895e-05, "data_time": 0.0009061336517333984, "loss": 0.18028637617826462, "time": 0.5936290502548218, "epoch": 3, "memory": 6318, "step": 10974} +{"lr": 8.35471773286596e-05, "data_time": 0.0009564638137817382, "loss": 0.16379580199718474, "time": 0.596064281463623, "epoch": 3, "memory": 6318, "step": 11074} +{"lr": 8.326869172836765e-05, "data_time": 0.0009354591369628906, "loss": 0.19253600984811783, "time": 0.5805278539657592, "epoch": 3, "memory": 6318, "step": 11174} +{"lr": 8.298840612883789e-05, "data_time": 0.0010148048400878905, "loss": 0.16079765260219575, "time": 0.5537952184677124, "epoch": 3, "memory": 6318, "step": 11274} +{"lr": 8.270633837715962e-05, "data_time": 0.0009637832641601563, "loss": 0.1523275852203369, "time": 0.5589352369308471, "epoch": 3, "memory": 6318, "step": 11374} +{"lr": 8.242250643390022e-05, "data_time": 0.0008406639099121094, "loss": 0.16228190660476685, "time": 0.5637318372726441, "epoch": 3, "memory": 6318, "step": 11474} +{"lr": 8.2136928371961e-05, "data_time": 0.0009596824645996093, "loss": 0.16325031816959382, "time": 0.5646752834320068, "epoch": 3, "memory": 6318, "step": 11574} +{"lr": 8.184962237542712e-05, "data_time": 0.0009074211120605469, "loss": 0.15824640095233916, "time": 0.569823431968689, "epoch": 3, "memory": 6318, "step": 11674} +{"lr": 8.156060673840912e-05, "data_time": 0.0008990049362182617, "loss": 0.16078418865799904, "time": 0.5731776714324951, "epoch": 3, "memory": 6318, "step": 11774} +{"accuracy/top1": 90.72405242919922, "data_time": 0.44536342291996395, "time": 0.5320818670864763, "step": 3} +{"lr": 8.116191337909002e-05, "data_time": 0.2445814847946167, "loss": 0.16252445727586745, "time": 0.5837427139282226, "epoch": 4, "memory": 6318, "step": 11911} +{"lr": 8.086891958166228e-05, "data_time": 0.0008585929870605468, "loss": 0.1551465429365635, "time": 0.5978672504425049, "epoch": 4, "memory": 6318, "step": 12011} +{"lr": 8.057427858966412e-05, "data_time": 0.0008853912353515625, "loss": 0.1561238296329975, "time": 0.5647099733352661, "epoch": 4, "memory": 6318, "step": 12111} +{"lr": 8.027800916425949e-05, "data_time": 0.000876927375793457, "loss": 0.15540698170661926, "time": 0.5088225126266479, "epoch": 4, "memory": 6318, "step": 12211} +{"lr": 7.998013017030234e-05, "data_time": 0.0009682178497314453, "loss": 0.16109662726521493, "time": 0.5707410335540771, "epoch": 4, "memory": 6318, "step": 12311} +{"lr": 7.968066057513528e-05, "data_time": 0.0008210897445678711, "loss": 0.17002131566405296, "time": 0.564606761932373, "epoch": 4, "memory": 6318, "step": 12411} +{"lr": 7.937961944738199e-05, "data_time": 0.0010721206665039063, "loss": 0.1626214936375618, "time": 0.6076979398727417, "epoch": 4, "memory": 6318, "step": 12511} +{"lr": 7.907702595573289e-05, "data_time": 0.0009151220321655274, "loss": 0.15160149782896043, "time": 0.5591695070266723, "epoch": 4, "memory": 6318, "step": 12611} +{"lr": 7.877289936772454e-05, "data_time": 0.0009728193283081055, "loss": 0.1789689317345619, "time": 0.6144480466842651, "epoch": 4, "memory": 6318, "step": 12711} +{"lr": 7.8467259048513e-05, "data_time": 0.0008974075317382812, "loss": 0.16176287233829498, "time": 0.5863381147384643, "epoch": 4, "memory": 6318, "step": 12811} +{"lr": 7.816012445964079e-05, "data_time": 0.000920557975769043, "loss": 0.1543870359659195, "time": 0.6008711576461792, "epoch": 4, "memory": 6318, "step": 12911} +{"lr": 7.785151515779718e-05, "data_time": 0.0009092092514038086, "loss": 0.14510656893253326, "time": 0.5613490343093872, "epoch": 4, "memory": 6318, "step": 13011} +{"lr": 7.754145079357376e-05, "data_time": 0.0009084224700927734, "loss": 0.152776862680912, "time": 0.553636622428894, "epoch": 4, "memory": 6318, "step": 13111} +{"lr": 7.722995111021218e-05, "data_time": 0.0009069442749023438, "loss": 0.1458202764391899, "time": 0.6177518606185913, "epoch": 4, "memory": 6318, "step": 13211} +{"lr": 7.691703594234788e-05, "data_time": 0.0012085914611816406, "loss": 0.16952790766954423, "time": 0.5421665430068969, "epoch": 4, "memory": 6318, "step": 13311} +{"lr": 7.660272521474673e-05, "data_time": 0.0008891105651855468, "loss": 0.14836854189634324, "time": 0.5812160015106201, "epoch": 4, "memory": 6318, "step": 13411} +{"lr": 7.628703894103628e-05, "data_time": 0.0009837865829467774, "loss": 0.1634357288479805, "time": 0.5678518056869507, "epoch": 4, "memory": 6318, "step": 13511} +{"lr": 7.596999722243168e-05, "data_time": 0.0009011030197143555, "loss": 0.15797903537750244, "time": 0.5531563997268677, "epoch": 4, "memory": 6318, "step": 13611} +{"lr": 7.565162024645551e-05, "data_time": 0.0009408712387084961, "loss": 0.15172194316983223, "time": 0.5672452926635743, "epoch": 4, "memory": 6318, "step": 13711} +{"lr": 7.533192828565237e-05, "data_time": 0.000873565673828125, "loss": 0.14865531995892525, "time": 0.5387712001800538, "epoch": 4, "memory": 6318, "step": 13811} +{"lr": 7.501094169629802e-05, "data_time": 0.0008395195007324218, "loss": 0.16808363944292068, "time": 0.9345919370651246, "epoch": 4, "memory": 6318, "step": 13911} +{"lr": 7.468868091710333e-05, "data_time": 0.0008895635604858398, "loss": 0.16219151765108109, "time": 0.5878986120223999, "epoch": 4, "memory": 6318, "step": 14011} +{"lr": 7.436516646791276e-05, "data_time": 0.0009893655776977539, "loss": 0.16593725457787514, "time": 0.5491437911987305, "epoch": 4, "memory": 6318, "step": 14111} +{"lr": 7.404041894839768e-05, "data_time": 0.001008296012878418, "loss": 0.1606047883629799, "time": 0.607050108909607, "epoch": 4, "memory": 6318, "step": 14211} +{"lr": 7.371445903674494e-05, "data_time": 0.0009079217910766601, "loss": 0.15264861583709716, "time": 0.4697632074356079, "epoch": 4, "memory": 6318, "step": 14311} +{"lr": 7.338730748834014e-05, "data_time": 0.001061415672302246, "loss": 0.14325622469186783, "time": 0.5767011404037475, "epoch": 4, "memory": 6318, "step": 14411} +{"lr": 7.305898513444536e-05, "data_time": 0.0011385202407836914, "loss": 0.16008485481142998, "time": 0.6132961511611938, "epoch": 4, "memory": 6318, "step": 14511} +{"lr": 7.27295128808741e-05, "data_time": 0.0008663415908813476, "loss": 0.16022886633872985, "time": 0.6077814817428588, "epoch": 4, "memory": 6318, "step": 14611} +{"lr": 7.239891170665884e-05, "data_time": 0.0009595870971679688, "loss": 0.15783049985766412, "time": 0.5469722509384155, "epoch": 4, "memory": 6318, "step": 14711} +{"lr": 7.206720266271582e-05, "data_time": 0.0009441375732421875, "loss": 0.14166781231760978, "time": 0.6376216411590576, "epoch": 4, "memory": 6318, "step": 14811} +{"lr": 7.173440687050449e-05, "data_time": 0.0008619308471679687, "loss": 0.14800475165247917, "time": 0.5642582654953003, "epoch": 4, "memory": 6318, "step": 14911} +{"lr": 7.140054552068265e-05, "data_time": 0.0009479284286499024, "loss": 0.15040959268808365, "time": 0.5604031562805176, "epoch": 4, "memory": 6318, "step": 15011} +{"lr": 7.106563987175705e-05, "data_time": 0.0009463548660278321, "loss": 0.16903550773859025, "time": 0.6342783689498901, "epoch": 4, "memory": 6318, "step": 15111} +{"lr": 7.072971124873013e-05, "data_time": 0.0009508371353149414, "loss": 0.15986801832914352, "time": 0.59492027759552, "epoch": 4, "memory": 6318, "step": 15211} +{"lr": 7.039278104174143e-05, "data_time": 0.0009119987487792968, "loss": 0.14383111372590066, "time": 0.579813814163208, "epoch": 4, "memory": 6318, "step": 15311} +{"lr": 7.005487070470615e-05, "data_time": 0.0009538650512695313, "loss": 0.16456588879227638, "time": 0.5852491617202759, "epoch": 4, "memory": 6318, "step": 15411} +{"lr": 6.971600175394891e-05, "data_time": 0.000942063331604004, "loss": 0.14582718461751937, "time": 0.6001242160797119, "epoch": 4, "memory": 6318, "step": 15511} +{"lr": 6.937619576683373e-05, "data_time": 0.0008599042892456054, "loss": 0.1710476577281952, "time": 0.5925651788711548, "epoch": 4, "memory": 6318, "step": 15611} +{"lr": 6.903547438039021e-05, "data_time": 0.000958395004272461, "loss": 0.14759338274598122, "time": 0.5645423650741577, "epoch": 4, "memory": 6318, "step": 15711} +{"accuracy/top1": 92.74208068847656, "data_time": 0.4560331969425596, "time": 0.5418031955587452, "step": 4} +{"lr": 6.856723955077045e-05, "data_time": 0.0007763147354125976, "loss": 0.14500877037644386, "time": 0.5487354040145874, "epoch": 5, "memory": 6318, "step": 15848} +{"lr": 6.822443541062844e-05, "data_time": 0.000841522216796875, "loss": 0.15268248543143273, "time": 0.6195852518081665, "epoch": 5, "memory": 6318, "step": 15948} +{"lr": 6.788078920910325e-05, "data_time": 0.0008341312408447265, "loss": 0.15663531944155692, "time": 0.5848025798797607, "epoch": 5, "memory": 6318, "step": 16048} +{"lr": 6.753632282774874e-05, "data_time": 0.0009988784790039063, "loss": 0.15814314037561417, "time": 0.5539665937423706, "epoch": 5, "memory": 6318, "step": 16148} +{"lr": 6.719105820034336e-05, "data_time": 0.0011941194534301758, "loss": 0.1559486523270607, "time": 0.5831831455230713, "epoch": 5, "memory": 6318, "step": 16248} +{"lr": 6.68450173114937e-05, "data_time": 0.00100252628326416, "loss": 0.16272845044732093, "time": 0.5619689702987671, "epoch": 5, "memory": 6318, "step": 16348} +{"lr": 6.64982221952347e-05, "data_time": 0.0009701013565063477, "loss": 0.13593565300107002, "time": 0.5660976648330689, "epoch": 5, "memory": 6318, "step": 16448} +{"lr": 6.615069493362599e-05, "data_time": 0.0009341716766357421, "loss": 0.16524188295006753, "time": 0.5730363368988037, "epoch": 5, "memory": 6318, "step": 16548} +{"lr": 6.580245765534682e-05, "data_time": 0.0010108709335327148, "loss": 0.14973906874656678, "time": 0.6023453950881958, "epoch": 5, "memory": 6318, "step": 16648} +{"lr": 6.54535325342862e-05, "data_time": 0.0009762048721313477, "loss": 0.1397172324359417, "time": 0.5546247005462647, "epoch": 5, "memory": 6318, "step": 16748} +{"lr": 6.510394178813155e-05, "data_time": 0.0009490489959716797, "loss": 0.16103667989373208, "time": 0.5314825773239136, "epoch": 5, "memory": 6318, "step": 16848} +{"lr": 6.475370767695331e-05, "data_time": 0.00085601806640625, "loss": 0.14232759326696395, "time": 0.5730009555816651, "epoch": 5, "memory": 6318, "step": 16948} +{"lr": 6.440285250178845e-05, "data_time": 0.0009927749633789062, "loss": 0.16256664842367172, "time": 0.5520715236663818, "epoch": 5, "memory": 6318, "step": 17048} +{"lr": 6.405139860321955e-05, "data_time": 0.0010034799575805663, "loss": 0.14732514321804047, "time": 0.5918293237686157, "epoch": 5, "memory": 6318, "step": 17148} +{"lr": 6.369936835995306e-05, "data_time": 0.0009931564331054688, "loss": 0.15432692989706992, "time": 0.6313480377197266, "epoch": 5, "memory": 6318, "step": 17248} +{"lr": 6.334678418739388e-05, "data_time": 0.0009126901626586914, "loss": 0.14209817796945573, "time": 0.5824315547943115, "epoch": 5, "memory": 6318, "step": 17348} +{"lr": 6.299366853621796e-05, "data_time": 0.0009621381759643555, "loss": 0.15826880782842637, "time": 0.5913075923919677, "epoch": 5, "memory": 6318, "step": 17448} +{"lr": 6.264004389094315e-05, "data_time": 0.0010961532592773438, "loss": 0.14500583633780478, "time": 0.5645835876464844, "epoch": 5, "memory": 6318, "step": 17548} +{"lr": 6.22859327684974e-05, "data_time": 0.0009605646133422852, "loss": 0.13999637290835382, "time": 0.5523180484771728, "epoch": 5, "memory": 6318, "step": 17648} +{"lr": 6.193135771678478e-05, "data_time": 0.0011732101440429688, "loss": 0.14428636059165, "time": 0.5800431489944458, "epoch": 5, "memory": 6318, "step": 17748} +{"lr": 6.157634131324983e-05, "data_time": 0.000865316390991211, "loss": 0.15345237776637077, "time": 0.5654687404632568, "epoch": 5, "memory": 6318, "step": 17848} +{"lr": 6.12209061634402e-05, "data_time": 0.0009256839752197265, "loss": 0.1531187653541565, "time": 0.5931895494461059, "epoch": 5, "memory": 6318, "step": 17948} +{"lr": 6.086507489956686e-05, "data_time": 0.0012007951736450195, "loss": 0.1524558886885643, "time": 0.4993189811706543, "epoch": 5, "memory": 6318, "step": 18048} +{"lr": 6.0508870179063345e-05, "data_time": 0.0009344816207885742, "loss": 0.1451650455594063, "time": 0.5356508255004883, "epoch": 5, "memory": 6318, "step": 18148} +{"lr": 6.015231468314303e-05, "data_time": 0.001004505157470703, "loss": 0.15471096336841583, "time": 0.5358565807342529, "epoch": 5, "memory": 6318, "step": 18248} +{"lr": 5.979543111535473e-05, "data_time": 0.0008063554763793945, "loss": 0.13539359495043754, "time": 0.5862555503845215, "epoch": 5, "memory": 6318, "step": 18348} +{"lr": 5.943824220013694e-05, "data_time": 0.0009228467941284179, "loss": 0.15212441831827164, "time": 0.5795838117599488, "epoch": 5, "memory": 6318, "step": 18448} +{"lr": 5.908077068137128e-05, "data_time": 0.0008740901947021484, "loss": 0.15683663859963418, "time": 0.5718268394470215, "epoch": 5, "memory": 6318, "step": 18548} +{"lr": 5.872303932093392e-05, "data_time": 0.0009355545043945312, "loss": 0.1608651399612427, "time": 0.5510890245437622, "epoch": 5, "memory": 6318, "step": 18648} +{"lr": 5.836507089724633e-05, "data_time": 0.000998210906982422, "loss": 0.14838469922542571, "time": 0.5488568305969238, "epoch": 5, "memory": 6318, "step": 18748} +{"lr": 5.800688820382498e-05, "data_time": 0.000847625732421875, "loss": 0.1373772993683815, "time": 0.5856449365615845, "epoch": 5, "memory": 6318, "step": 18848} +{"lr": 5.7648514047829945e-05, "data_time": 0.0008908271789550781, "loss": 0.13893911689519883, "time": 0.573755931854248, "epoch": 5, "memory": 6318, "step": 18948} +{"lr": 5.7289971248612474e-05, "data_time": 0.0008938312530517578, "loss": 0.15458082184195518, "time": 0.5382607221603394, "epoch": 5, "memory": 6318, "step": 19048} +{"lr": 5.693128263626215e-05, "data_time": 0.0010591745376586914, "loss": 0.13220496401190757, "time": 0.5498858451843261, "epoch": 5, "memory": 6318, "step": 19148} +{"lr": 5.6572471050153236e-05, "data_time": 0.0009000539779663086, "loss": 0.13859168887138368, "time": 0.5480077266693115, "epoch": 5, "memory": 6318, "step": 19248} +{"lr": 5.621355933749018e-05, "data_time": 0.0009001970291137695, "loss": 0.14990852773189545, "time": 0.5767415761947632, "epoch": 5, "memory": 6318, "step": 19348} +{"lr": 5.5854570351853106e-05, "data_time": 0.0009374856948852539, "loss": 0.12843604907393455, "time": 0.5519895553588867, "epoch": 5, "memory": 6318, "step": 19448} +{"lr": 5.5495526951742315e-05, "data_time": 0.0009349584579467773, "loss": 0.14197040423750878, "time": 0.5452033519744873, "epoch": 5, "memory": 6318, "step": 19548} +{"lr": 5.513645199912304e-05, "data_time": 0.0011122703552246093, "loss": 0.13910222202539443, "time": 0.5302355766296387, "epoch": 5, "memory": 6318, "step": 19648} +{"accuracy/top1": 93.65018463134766, "data_time": 0.44459915983265846, "time": 0.5333290798910733, "step": 5} +{"lr": 5.464450978670876e-05, "data_time": 0.05841796398162842, "loss": 0.14123359322547913, "time": 0.5865522861480713, "epoch": 6, "memory": 6318, "step": 19785} +{"lr": 5.4285451361545876e-05, "data_time": 0.0008589029312133789, "loss": 0.13904216289520263, "time": 0.5427911281585693, "epoch": 6, "memory": 6318, "step": 19885} +{"lr": 5.3926438435022166e-05, "data_time": 0.0009965896606445312, "loss": 0.13401658236980438, "time": 0.5614965677261352, "epoch": 6, "memory": 6318, "step": 19985} +{"lr": 5.356749386716241e-05, "data_time": 0.0008932113647460938, "loss": 0.13127534538507463, "time": 0.5408131122589112, "epoch": 6, "memory": 6318, "step": 20085} +{"lr": 5.320864051363873e-05, "data_time": 0.0009876251220703124, "loss": 0.13893340304493904, "time": 0.5377917051315307, "epoch": 6, "memory": 6318, "step": 20185} +{"lr": 5.2849901224315175e-05, "data_time": 0.0009930133819580078, "loss": 0.1533495657145977, "time": 0.5118759155273438, "epoch": 6, "memory": 6318, "step": 20285} +{"lr": 5.249129884179276e-05, "data_time": 0.0008838176727294922, "loss": 0.1379082404077053, "time": 0.511989974975586, "epoch": 6, "memory": 6318, "step": 20385} +{"lr": 5.2132856199954985e-05, "data_time": 0.0008897066116333007, "loss": 0.13805076330900193, "time": 0.541465449333191, "epoch": 6, "memory": 6318, "step": 20485} +{"lr": 5.177459612251411e-05, "data_time": 0.000903010368347168, "loss": 0.15050484389066696, "time": 0.562174940109253, "epoch": 6, "memory": 6318, "step": 20585} +{"lr": 5.141654142155744e-05, "data_time": 0.0009062528610229492, "loss": 0.13774992823600768, "time": 0.5470547914505005, "epoch": 6, "memory": 6318, "step": 20685} +{"lr": 5.105871489609512e-05, "data_time": 0.0009209632873535157, "loss": 0.14223484694957733, "time": 0.4802698135375977, "epoch": 6, "memory": 6318, "step": 20785} +{"lr": 5.070113933060828e-05, "data_time": 0.0008120059967041016, "loss": 0.13637217432260512, "time": 0.5403118371963501, "epoch": 6, "memory": 6318, "step": 20885} +{"lr": 5.0343837493598264e-05, "data_time": 0.0008483648300170899, "loss": 0.14491448029875756, "time": 0.5779665946960449, "epoch": 6, "memory": 6318, "step": 20985} +{"lr": 4.998683213613675e-05, "data_time": 0.0008725404739379882, "loss": 0.12851423695683478, "time": 0.5560952425003052, "epoch": 6, "memory": 6318, "step": 21085} +{"lr": 4.963014599041739e-05, "data_time": 0.0008286714553833007, "loss": 0.15191066414117813, "time": 0.5163877964019775, "epoch": 6, "memory": 6318, "step": 21185} +{"lr": 4.927380176830793e-05, "data_time": 0.0008488178253173828, "loss": 0.13374930918216704, "time": 0.5363269567489624, "epoch": 6, "memory": 6318, "step": 21285} +{"lr": 4.89178221599044e-05, "data_time": 0.000800013542175293, "loss": 0.14563221335411072, "time": 0.5094030380249024, "epoch": 6, "memory": 6318, "step": 21385} +{"lr": 4.85622298320861e-05, "data_time": 0.0009118318557739258, "loss": 0.1324270635843277, "time": 0.559794545173645, "epoch": 6, "memory": 6318, "step": 21485} +{"lr": 4.8207047427072355e-05, "data_time": 0.0007805347442626954, "loss": 0.16379940062761306, "time": 0.5867275953292846, "epoch": 6, "memory": 6318, "step": 21585} +{"lr": 4.7852297560980876e-05, "data_time": 0.0010346651077270507, "loss": 0.1348823480308056, "time": 0.5411486625671387, "epoch": 6, "memory": 6318, "step": 21685} +{"lr": 4.749800282238749e-05, "data_time": 0.0009227991104125977, "loss": 0.13596554175019265, "time": 0.5152548789978028, "epoch": 6, "memory": 6318, "step": 21785} +{"lr": 4.714418577088805e-05, "data_time": 0.0009283304214477539, "loss": 0.1340097300708294, "time": 0.5270004987716674, "epoch": 6, "memory": 6318, "step": 21885} +{"lr": 4.679086893566184e-05, "data_time": 0.000896286964416504, "loss": 0.13446456789970399, "time": 0.5040998458862305, "epoch": 6, "memory": 6318, "step": 21985} +{"lr": 4.643807481403694e-05, "data_time": 0.0008671760559082031, "loss": 0.1462169148027897, "time": 0.5261975049972534, "epoch": 6, "memory": 6318, "step": 22085} +{"lr": 4.6085825870057946e-05, "data_time": 0.0008907794952392578, "loss": 0.13258469328284264, "time": 0.4997079849243164, "epoch": 6, "memory": 6318, "step": 22185} +{"lr": 4.573414453305537e-05, "data_time": 0.0008089303970336914, "loss": 0.14285623654723167, "time": 0.5209615468978882, "epoch": 6, "memory": 6318, "step": 22285} +{"lr": 4.5383053196217514e-05, "data_time": 0.0008890390396118164, "loss": 0.14570192396640777, "time": 0.5254142999649047, "epoch": 6, "memory": 6318, "step": 22385} +{"lr": 4.503257421516477e-05, "data_time": 0.0008598804473876953, "loss": 0.1358324870467186, "time": 0.4974297761917114, "epoch": 6, "memory": 6318, "step": 22485} +{"lr": 4.468272990652586e-05, "data_time": 0.0008741617202758789, "loss": 0.13534300327301024, "time": 0.5563331604003906, "epoch": 6, "memory": 6318, "step": 22585} +{"lr": 4.433354254651697e-05, "data_time": 0.0009692907333374023, "loss": 0.13278524056077004, "time": 0.5399020671844482, "epoch": 6, "memory": 6318, "step": 22685} +{"lr": 4.398503436952338e-05, "data_time": 0.0009827136993408204, "loss": 0.12180397585034371, "time": 0.5461867809295654, "epoch": 6, "memory": 6318, "step": 22785} +{"lr": 4.363722756668351e-05, "data_time": 0.0009311199188232422, "loss": 0.13818999454379083, "time": 0.5394057035446167, "epoch": 6, "memory": 6318, "step": 22885} +{"lr": 4.329014428447611e-05, "data_time": 0.0008503198623657227, "loss": 0.1434588126838207, "time": 0.5260370969772339, "epoch": 6, "memory": 6318, "step": 22985} +{"lr": 4.294380662331e-05, "data_time": 0.0008500099182128906, "loss": 0.13585112169384955, "time": 0.5178012132644654, "epoch": 6, "memory": 6318, "step": 23085} +{"lr": 4.259823663611678e-05, "data_time": 0.0009212255477905274, "loss": 0.13363206684589385, "time": 0.5816402912139893, "epoch": 6, "memory": 6318, "step": 23185} +{"lr": 4.22534563269468e-05, "data_time": 0.0009413003921508789, "loss": 0.12947214618325234, "time": 0.5552320003509521, "epoch": 6, "memory": 6318, "step": 23285} +{"lr": 4.1909487649567694e-05, "data_time": 0.0008799076080322265, "loss": 0.1377067133784294, "time": 0.5511159181594849, "epoch": 6, "memory": 6318, "step": 23385} +{"lr": 4.156635250606698e-05, "data_time": 0.0009684085845947266, "loss": 0.14151004552841187, "time": 0.5126610040664673, "epoch": 6, "memory": 6318, "step": 23485} +{"lr": 4.122407274545703e-05, "data_time": 0.00086212158203125, "loss": 0.12892805486917497, "time": 0.5429720878601074, "epoch": 6, "memory": 6318, "step": 23585} +{"accuracy/top1": 94.41912078857422, "data_time": 0.4195119882452077, "time": 0.5064442938771742, "step": 6} +{"lr": 4.075657788156285e-05, "data_time": 0.0111541748046875, "loss": 0.12921048030257226, "time": 0.5187873840332031, "epoch": 7, "memory": 6318, "step": 23722} +{"lr": 4.0416412311741566e-05, "data_time": 0.0008292913436889649, "loss": 0.13311309665441512, "time": 0.5393574237823486, "epoch": 7, "memory": 6318, "step": 23822} +{"lr": 4.007717534685821e-05, "data_time": 0.0008679866790771484, "loss": 0.1257340505719185, "time": 0.5401736259460449, "epoch": 7, "memory": 6318, "step": 23922} +{"lr": 3.973888858770982e-05, "data_time": 0.0009850502014160157, "loss": 0.13229683339595794, "time": 0.5292590856552124, "epoch": 7, "memory": 6318, "step": 24022} +{"lr": 3.9401573574589535e-05, "data_time": 0.0010236740112304688, "loss": 0.13488859087228774, "time": 0.5393239259719849, "epoch": 7, "memory": 6318, "step": 24122} +{"lr": 3.906525178591511e-05, "data_time": 0.0008519411087036133, "loss": 0.13168733865022658, "time": 0.5322802066802979, "epoch": 7, "memory": 6318, "step": 24222} +{"lr": 3.87299446368609e-05, "data_time": 0.0008388996124267578, "loss": 0.13895001113414765, "time": 0.5517948865890503, "epoch": 7, "memory": 6318, "step": 24322} +{"lr": 3.839567347799442e-05, "data_time": 0.0008920669555664062, "loss": 0.1328663319349289, "time": 0.5596678018569946, "epoch": 7, "memory": 6318, "step": 24422} +{"lr": 3.806245959391694e-05, "data_time": 0.0008179903030395508, "loss": 0.13098523691296576, "time": 0.5361457586288452, "epoch": 7, "memory": 6318, "step": 24522} +{"lr": 3.773032420190816e-05, "data_time": 0.0008769989013671875, "loss": 0.1411207750439644, "time": 0.5402635097503662, "epoch": 7, "memory": 6318, "step": 24622} +{"lr": 3.739928845057507e-05, "data_time": 0.0009146928787231445, "loss": 0.13031796142458915, "time": 0.577673864364624, "epoch": 7, "memory": 6318, "step": 24722} +{"lr": 3.706937341850541e-05, "data_time": 0.0011984825134277344, "loss": 0.14541516155004502, "time": 0.567496943473816, "epoch": 7, "memory": 6318, "step": 24822} +{"lr": 3.674060011292556e-05, "data_time": 0.0008799314498901367, "loss": 0.12934816256165504, "time": 0.5143082141876221, "epoch": 7, "memory": 6318, "step": 24922} +{"lr": 3.641298946836287e-05, "data_time": 0.0009241819381713868, "loss": 0.14347289130091667, "time": 0.5372864007949829, "epoch": 7, "memory": 6318, "step": 25022} +{"lr": 3.6086562345312595e-05, "data_time": 0.0008380651473999023, "loss": 0.1324164941906929, "time": 0.49466605186462403, "epoch": 7, "memory": 6318, "step": 25122} +{"lr": 3.576133952890974e-05, "data_time": 0.0009708881378173828, "loss": 0.13816252574324608, "time": 0.5268621921539307, "epoch": 7, "memory": 6318, "step": 25222} +{"lr": 3.543734172760541e-05, "data_time": 0.0009027242660522461, "loss": 0.15072961449623107, "time": 0.4827312469482422, "epoch": 7, "memory": 6318, "step": 25322} +{"lr": 3.511458957184831e-05, "data_time": 0.0009831905364990235, "loss": 0.13165954500436783, "time": 0.5381730318069458, "epoch": 7, "memory": 6318, "step": 25422} +{"lr": 3.479310361277118e-05, "data_time": 0.0008530378341674805, "loss": 0.12745990008115768, "time": 0.5493924856185913, "epoch": 7, "memory": 6318, "step": 25522} +{"lr": 3.4472904320882065e-05, "data_time": 0.0008347034454345703, "loss": 0.1353730984032154, "time": 0.5355998754501343, "epoch": 7, "memory": 6318, "step": 25622} +{"lr": 3.4154012084760864e-05, "data_time": 0.000901484489440918, "loss": 0.13354340568184853, "time": 0.49695773124694825, "epoch": 7, "memory": 6318, "step": 25722} +{"lr": 3.38364472097612e-05, "data_time": 0.000924372673034668, "loss": 0.14177048951387405, "time": 0.5665776014328003, "epoch": 7, "memory": 6318, "step": 25822} +{"lr": 3.3520229916717485e-05, "data_time": 0.0011937856674194337, "loss": 0.14481553956866264, "time": 0.5834773540496826, "epoch": 7, "memory": 6318, "step": 25922} +{"lr": 3.320538034065722e-05, "data_time": 0.000817108154296875, "loss": 0.14810217991471292, "time": 0.5073909044265748, "epoch": 7, "memory": 6318, "step": 26022} +{"lr": 3.289191852951901e-05, "data_time": 0.0008272647857666016, "loss": 0.125517837703228, "time": 0.5073810815811157, "epoch": 7, "memory": 6318, "step": 26122} +{"lr": 3.257986444287605e-05, "data_time": 0.0009132862091064453, "loss": 0.11254951730370522, "time": 0.5433212995529175, "epoch": 7, "memory": 6318, "step": 26222} +{"lr": 3.226923795066519e-05, "data_time": 0.0008596181869506836, "loss": 0.14573151916265487, "time": 0.5444326877593995, "epoch": 7, "memory": 6318, "step": 26322} +{"lr": 3.196005883192158e-05, "data_time": 0.0008810043334960937, "loss": 0.12877408564090728, "time": 0.48203999996185304, "epoch": 7, "memory": 6318, "step": 26422} +{"lr": 3.165234677351947e-05, "data_time": 0.0009072542190551758, "loss": 0.1379546657204628, "time": 0.5224812746047973, "epoch": 7, "memory": 6318, "step": 26522} +{"lr": 3.134612136891845e-05, "data_time": 0.0009146451950073243, "loss": 0.11773192062973976, "time": 0.5135346412658691, "epoch": 7, "memory": 6318, "step": 26622} +{"lr": 3.104140211691602e-05, "data_time": 0.0008940696716308594, "loss": 0.13146670460700988, "time": 0.5220028162002563, "epoch": 7, "memory": 6318, "step": 26722} +{"lr": 3.073820842040592e-05, "data_time": 0.0008766651153564453, "loss": 0.14702612683176994, "time": 0.5504681587219238, "epoch": 7, "memory": 6318, "step": 26822} +{"lr": 3.043655958514262e-05, "data_time": 0.0008579254150390625, "loss": 0.1253778614103794, "time": 0.4890860080718994, "epoch": 7, "memory": 6318, "step": 26922} +{"lr": 3.013647481851217e-05, "data_time": 0.0008542299270629883, "loss": 0.13747862502932548, "time": 0.49033732414245607, "epoch": 7, "memory": 6318, "step": 27022} +{"lr": 2.983797322830899e-05, "data_time": 0.0008123874664306641, "loss": 0.1490783706307411, "time": 0.5488513946533203, "epoch": 7, "memory": 6318, "step": 27122} +{"lr": 2.9541073821519346e-05, "data_time": 0.0008112668991088867, "loss": 0.1284991666674614, "time": 0.4884809494018555, "epoch": 7, "memory": 6318, "step": 27222} +{"lr": 2.924579550311095e-05, "data_time": 0.0008758544921875, "loss": 0.1362491451203823, "time": 0.5032778024673462, "epoch": 7, "memory": 6318, "step": 27322} +{"lr": 2.8952157074829418e-05, "data_time": 0.0009056806564331054, "loss": 0.1216659426689148, "time": 0.5200213670730591, "epoch": 7, "memory": 6318, "step": 27422} +{"lr": 2.866017723400071e-05, "data_time": 0.000866103172302246, "loss": 0.12323641031980515, "time": 0.5441327571868897, "epoch": 7, "memory": 6318, "step": 27522} +{"accuracy/top1": 94.56525421142578, "data_time": 0.41920522163654195, "time": 0.5060606454980785, "step": 7} +{"lr": 2.8262891373132927e-05, "data_time": 0.14242446422576904, "loss": 0.1266934886574745, "time": 0.5188592672348022, "epoch": 8, "memory": 6318, "step": 27659} +{"lr": 2.797491643470029e-05, "data_time": 0.09833438396453857, "loss": 0.13300009965896606, "time": 0.579714035987854, "epoch": 8, "memory": 6318, "step": 27759} +{"lr": 2.768866230918266e-05, "data_time": 0.0008493423461914063, "loss": 0.13277756944298744, "time": 0.5567246675491333, "epoch": 8, "memory": 6318, "step": 27859} +{"lr": 2.7404147223713185e-05, "data_time": 0.0009485721588134765, "loss": 0.11810171157121659, "time": 0.4748399257659912, "epoch": 8, "memory": 6318, "step": 27959} +{"lr": 2.712138929469222e-05, "data_time": 0.0009093046188354492, "loss": 0.13050267472863197, "time": 0.5220201253890991, "epoch": 8, "memory": 6318, "step": 28059} +{"lr": 2.6840406526633896e-05, "data_time": 0.000984954833984375, "loss": 0.148331019282341, "time": 0.5292039632797241, "epoch": 8, "memory": 6318, "step": 28159} +{"lr": 2.6561216811019345e-05, "data_time": 0.000967097282409668, "loss": 0.12500903531908988, "time": 0.509897780418396, "epoch": 8, "memory": 6318, "step": 28259} +{"lr": 2.6283837925157908e-05, "data_time": 0.0008904695510864258, "loss": 0.14851940795779228, "time": 0.5096237421035766, "epoch": 8, "memory": 6318, "step": 28359} +{"lr": 2.6008287531054805e-05, "data_time": 0.0008517742156982422, "loss": 0.12435762360692024, "time": 0.5155454397201538, "epoch": 8, "memory": 6318, "step": 28459} +{"lr": 2.5734583174286796e-05, "data_time": 0.0008705377578735352, "loss": 0.12857967764139175, "time": 0.5149909257888794, "epoch": 8, "memory": 6318, "step": 28559} +{"lr": 2.546274228288488e-05, "data_time": 0.0008172988891601562, "loss": 0.1420954018831253, "time": 0.603237509727478, "epoch": 8, "memory": 6318, "step": 28659} +{"lr": 2.5192782166224308e-05, "data_time": 0.0008053779602050781, "loss": 0.1340021774172783, "time": 0.5605469465255737, "epoch": 8, "memory": 6318, "step": 28759} +{"lr": 2.492472001392276e-05, "data_time": 0.000878143310546875, "loss": 0.13753439858555794, "time": 0.5194282293319702, "epoch": 8, "memory": 6318, "step": 28859} +{"lr": 2.4658572894745857e-05, "data_time": 0.0012406587600708007, "loss": 0.14896167889237405, "time": 0.5091075897216797, "epoch": 8, "memory": 6318, "step": 28959} +{"lr": 2.439435775551997e-05, "data_time": 0.0009567975997924804, "loss": 0.13417726755142212, "time": 0.5405267000198364, "epoch": 8, "memory": 6318, "step": 29059} +{"lr": 2.4132091420053357e-05, "data_time": 0.0009279251098632812, "loss": 0.13520755246281624, "time": 0.5230050325393677, "epoch": 8, "memory": 6318, "step": 29159} +{"lr": 2.387179058806494e-05, "data_time": 0.0009148120880126953, "loss": 0.1209425762295723, "time": 0.5530800580978393, "epoch": 8, "memory": 6318, "step": 29259} +{"lr": 2.361347183412073e-05, "data_time": 0.0009204626083374023, "loss": 0.13417719900608063, "time": 0.543345856666565, "epoch": 8, "memory": 6318, "step": 29359} +{"lr": 2.3357151606578728e-05, "data_time": 0.0008663415908813476, "loss": 0.13448622971773147, "time": 0.5656534433364868, "epoch": 8, "memory": 6318, "step": 29459} +{"lr": 2.3102846226541355e-05, "data_time": 0.0009607553482055664, "loss": 0.13304293677210807, "time": 0.5370008945465088, "epoch": 8, "memory": 6318, "step": 29559} +{"lr": 2.2850571886816365e-05, "data_time": 0.001056528091430664, "loss": 0.15710695087909698, "time": 0.5332360029220581, "epoch": 8, "memory": 6318, "step": 29659} +{"lr": 2.260034465088568e-05, "data_time": 0.000875401496887207, "loss": 0.12475125044584275, "time": 0.5693212985992432, "epoch": 8, "memory": 6318, "step": 29759} +{"lr": 2.2352180451882643e-05, "data_time": 0.0008660316467285156, "loss": 0.12586882412433625, "time": 0.5214612483978271, "epoch": 8, "memory": 6318, "step": 29859} +{"lr": 2.210609509157741e-05, "data_time": 0.0008114814758300781, "loss": 0.13693549633026122, "time": 0.49515542984008787, "epoch": 8, "memory": 6318, "step": 29959} +{"lr": 2.1862104239370772e-05, "data_time": 0.0008178472518920899, "loss": 0.12876120433211327, "time": 0.548601508140564, "epoch": 8, "memory": 6318, "step": 30059} +{"lr": 2.1620223431296407e-05, "data_time": 0.0008268117904663086, "loss": 0.12953976020216942, "time": 0.5336732864379883, "epoch": 8, "memory": 6318, "step": 30159} +{"lr": 2.1380468069031744e-05, "data_time": 0.0008353233337402344, "loss": 0.11989609152078629, "time": 0.5489434003829956, "epoch": 8, "memory": 6318, "step": 30259} +{"lr": 2.1142853418917185e-05, "data_time": 0.0008357048034667968, "loss": 0.131401364505291, "time": 0.4954005479812622, "epoch": 8, "memory": 6318, "step": 30359} +{"lr": 2.0907394610983892e-05, "data_time": 0.0009056329727172852, "loss": 0.10983985811471939, "time": 0.5195320129394532, "epoch": 8, "memory": 6318, "step": 30459} +{"lr": 2.0674106637990643e-05, "data_time": 0.000891733169555664, "loss": 0.12375066280364991, "time": 0.5492771863937378, "epoch": 8, "memory": 6318, "step": 30559} +{"lr": 2.0443004354468975e-05, "data_time": 0.000986194610595703, "loss": 0.13219113796949386, "time": 0.5381386518478394, "epoch": 8, "memory": 6318, "step": 30659} +{"lr": 2.0214102475777453e-05, "data_time": 0.0009032011032104493, "loss": 0.12914354279637336, "time": 0.5168377161026001, "epoch": 8, "memory": 6318, "step": 30759} +{"lr": 1.9987415577164583e-05, "data_time": 0.0010039567947387694, "loss": 0.14365241378545762, "time": 0.539490008354187, "epoch": 8, "memory": 6318, "step": 30859} +{"lr": 1.9762958092840814e-05, "data_time": 0.0008420467376708985, "loss": 0.132231717556715, "time": 0.5407920598983764, "epoch": 8, "memory": 6318, "step": 30959} +{"lr": 1.9540744315059334e-05, "data_time": 0.000885176658630371, "loss": 0.1139188252389431, "time": 0.533031678199768, "epoch": 8, "memory": 6318, "step": 31059} +{"lr": 1.9320788393206132e-05, "data_time": 0.0008019447326660156, "loss": 0.14223519787192346, "time": 0.553769302368164, "epoch": 8, "memory": 6318, "step": 31159} +{"lr": 1.9103104332899027e-05, "data_time": 0.0008715629577636719, "loss": 0.1313345156610012, "time": 0.5306265115737915, "epoch": 8, "memory": 6318, "step": 31259} +{"lr": 1.888770599509573e-05, "data_time": 0.000914311408996582, "loss": 0.11755105927586555, "time": 0.5092584609985351, "epoch": 8, "memory": 6318, "step": 31359} +{"lr": 1.8674607095211516e-05, "data_time": 0.0012692451477050782, "loss": 0.13470162004232406, "time": 0.5235853672027588, "epoch": 8, "memory": 6318, "step": 31459} +{"accuracy/top1": 94.88188171386719, "data_time": 0.42409575396570665, "time": 0.5100471644565977, "step": 8} +{"lr": 1.838641934548488e-05, "data_time": 0.20252509117126466, "loss": 0.13036711141467094, "time": 0.5421467781066894, "epoch": 9, "memory": 6318, "step": 31596} +{"lr": 1.817882403185378e-05, "data_time": 0.0011459112167358399, "loss": 0.14821603745222092, "time": 0.5204514741897583, "epoch": 9, "memory": 6318, "step": 31696} +{"lr": 1.797357329395729e-05, "data_time": 0.0007365226745605468, "loss": 0.14046533778309822, "time": 0.5716930627822876, "epoch": 9, "memory": 6318, "step": 31796} +{"lr": 1.7770680201065646e-05, "data_time": 0.0007657051086425781, "loss": 0.1272752359509468, "time": 0.6035057783126831, "epoch": 9, "memory": 6318, "step": 31896} +{"lr": 1.7570157672326885e-05, "data_time": 0.0007543802261352539, "loss": 0.13307861760258674, "time": 0.5485161542892456, "epoch": 9, "memory": 6318, "step": 31996} +{"lr": 1.7372018475944145e-05, "data_time": 0.0008781671524047852, "loss": 0.12959629371762277, "time": 0.6123078107833863, "epoch": 9, "memory": 6318, "step": 32096} +{"lr": 1.717627522836271e-05, "data_time": 0.0007833957672119141, "loss": 0.1255411498248577, "time": 1.0483634710311889, "epoch": 9, "memory": 6318, "step": 32196} +{"lr": 1.6982940393466685e-05, "data_time": 0.0007716655731201172, "loss": 0.13023898154497146, "time": 0.5581950426101685, "epoch": 9, "memory": 6318, "step": 32296} +{"lr": 1.6792026281785218e-05, "data_time": 0.0008478164672851562, "loss": 0.12103265821933747, "time": 0.5452906370162964, "epoch": 9, "memory": 6318, "step": 32396} +{"lr": 1.660354504970888e-05, "data_time": 0.0007548332214355469, "loss": 0.1338508330285549, "time": 0.5353586435317993, "epoch": 9, "memory": 6318, "step": 32496} +{"lr": 1.641750869871534e-05, "data_time": 0.0008444547653198242, "loss": 0.13075316399335862, "time": 0.5070287704467773, "epoch": 9, "memory": 6318, "step": 32596} +{"lr": 1.6233929074605374e-05, "data_time": 0.0007917881011962891, "loss": 0.12046018317341804, "time": 0.5147933959960938, "epoch": 9, "memory": 6318, "step": 32696} +{"lr": 1.6052817866748513e-05, "data_time": 0.0008064746856689453, "loss": 0.13475532457232475, "time": 0.5322628974914551, "epoch": 9, "memory": 6318, "step": 32796} +{"lr": 1.587418660733871e-05, "data_time": 0.0007519721984863281, "loss": 0.12381435409188271, "time": 0.5126032352447509, "epoch": 9, "memory": 6318, "step": 32896} +{"lr": 1.5698046670660044e-05, "data_time": 0.0008713722229003907, "loss": 0.13767931684851648, "time": 0.5360569953918457, "epoch": 9, "memory": 6318, "step": 32996} +{"lr": 1.5524409272362524e-05, "data_time": 0.0007413148880004883, "loss": 0.13964034393429756, "time": 0.5586385250091552, "epoch": 9, "memory": 6318, "step": 33096} +{"lr": 1.5353285468747853e-05, "data_time": 0.0007857799530029297, "loss": 0.11721767336130143, "time": 0.5244704246520996, "epoch": 9, "memory": 6318, "step": 33196} +{"lr": 1.5184686156065442e-05, "data_time": 0.0008677482604980469, "loss": 0.12741190642118455, "time": 0.5342095613479614, "epoch": 9, "memory": 6318, "step": 33296} +{"lr": 1.5018622069818636e-05, "data_time": 0.0007383108139038086, "loss": 0.13146117106080055, "time": 0.5741929531097412, "epoch": 9, "memory": 6318, "step": 33396} +{"lr": 1.4855103784081099e-05, "data_time": 0.0009666204452514649, "loss": 0.14917017370462418, "time": 0.5846997261047363, "epoch": 9, "memory": 6318, "step": 33496} +{"lr": 1.4694141710823508e-05, "data_time": 0.0009002208709716797, "loss": 0.137880901992321, "time": 0.5479081630706787, "epoch": 9, "memory": 6318, "step": 33596} +{"lr": 1.4535746099250614e-05, "data_time": 0.000886225700378418, "loss": 0.12109714597463608, "time": 0.562928295135498, "epoch": 9, "memory": 6318, "step": 33696} +{"lr": 1.4379927035148576e-05, "data_time": 0.0008934974670410156, "loss": 0.11005139574408532, "time": 0.5172569274902343, "epoch": 9, "memory": 6318, "step": 33796} +{"lr": 1.42266944402428e-05, "data_time": 0.0008431434631347656, "loss": 0.12268985509872436, "time": 0.542741322517395, "epoch": 9, "memory": 6318, "step": 33896} +{"lr": 1.4076058071566133e-05, "data_time": 0.0008301496505737305, "loss": 0.13168108239769935, "time": 0.535594129562378, "epoch": 9, "memory": 6318, "step": 33996} +{"lr": 1.3928027520837635e-05, "data_time": 0.0008788585662841796, "loss": 0.12917825505137442, "time": 0.5027451276779175, "epoch": 9, "memory": 6318, "step": 34096} +{"lr": 1.3782612213851759e-05, "data_time": 0.0008229255676269531, "loss": 0.1309836857020855, "time": 0.5145463705062866, "epoch": 9, "memory": 6318, "step": 34196} +{"lr": 1.3639821409878263e-05, "data_time": 0.000844264030456543, "loss": 0.1216252364218235, "time": 0.5617667198181152, "epoch": 9, "memory": 6318, "step": 34296} +{"lr": 1.3499664201072546e-05, "data_time": 0.0008713245391845703, "loss": 0.14603504464030265, "time": 0.5469597101211547, "epoch": 9, "memory": 6318, "step": 34396} +{"lr": 1.336214951189674e-05, "data_time": 0.0008215427398681641, "loss": 0.12230904772877693, "time": 0.48522827625274656, "epoch": 9, "memory": 6318, "step": 34496} +{"lr": 1.3227286098551456e-05, "data_time": 0.0008590936660766602, "loss": 0.12949518114328384, "time": 0.5468153953552246, "epoch": 9, "memory": 6318, "step": 34596} +{"lr": 1.3095082548418224e-05, "data_time": 0.0008005857467651367, "loss": 0.12470810413360596, "time": 0.5216195106506347, "epoch": 9, "memory": 6318, "step": 34696} +{"lr": 1.2965547279512709e-05, "data_time": 0.0009238481521606446, "loss": 0.1334141582250595, "time": 0.5606493711471557, "epoch": 9, "memory": 6318, "step": 34796} +{"lr": 1.2838688539948659e-05, "data_time": 0.0009300947189331054, "loss": 0.12197697162628174, "time": 0.5481856822967529, "epoch": 9, "memory": 6318, "step": 34896} +{"lr": 1.2714514407412763e-05, "data_time": 0.0008265495300292969, "loss": 0.13125229775905609, "time": 0.5412568569183349, "epoch": 9, "memory": 6318, "step": 34996} +{"lr": 1.2593032788650256e-05, "data_time": 0.0008857488632202148, "loss": 0.12186857983469963, "time": 0.5148493766784668, "epoch": 9, "memory": 6318, "step": 35096} +{"lr": 1.2474251418961484e-05, "data_time": 0.0008896350860595703, "loss": 0.1418699562549591, "time": 0.5153765201568603, "epoch": 9, "memory": 6318, "step": 35196} +{"lr": 1.2358177861709387e-05, "data_time": 0.0011574029922485352, "loss": 0.11833802014589309, "time": 0.5051170825958252, "epoch": 9, "memory": 6318, "step": 35296} +{"lr": 1.2244819507837855e-05, "data_time": 0.0009081602096557617, "loss": 0.13394040688872338, "time": 0.5407806634902954, "epoch": 9, "memory": 6318, "step": 35396} +{"accuracy/top1": 95.19850158691406, "data_time": 0.4250262235773021, "time": 0.5117717488058682, "step": 9} +{"lr": 1.2093939694118031e-05, "data_time": 0.1658930778503418, "loss": 0.1427712269127369, "time": 0.539451289176941, "epoch": 10, "memory": 6318, "step": 35533} +{"lr": 1.1987044880988498e-05, "data_time": 0.001073932647705078, "loss": 0.1318757027387619, "time": 0.5351577043533325, "epoch": 10, "memory": 6318, "step": 35633} +{"lr": 1.1882888903004625e-05, "data_time": 0.0009023427963256836, "loss": 0.11952565461397172, "time": 0.49394822120666504, "epoch": 10, "memory": 6318, "step": 35733} +{"lr": 1.178147839226253e-05, "data_time": 0.0008791446685791016, "loss": 0.1322527289390564, "time": 0.5419137001037597, "epoch": 10, "memory": 6318, "step": 35833} +{"lr": 1.1682819806041682e-05, "data_time": 0.0008666276931762696, "loss": 0.12705398648977279, "time": 0.5449024200439453, "epoch": 10, "memory": 6318, "step": 35933} +{"lr": 1.1586919426393653e-05, "data_time": 0.0008065938949584961, "loss": 0.11582861244678497, "time": 0.5248615503311157, "epoch": 10, "memory": 6318, "step": 36033} +{"lr": 1.1493783359742196e-05, "data_time": 0.0007873058319091796, "loss": 0.1197835549712181, "time": 0.5558968544006347, "epoch": 10, "memory": 6318, "step": 36133} +{"lr": 1.1403417536494367e-05, "data_time": 0.0008743524551391602, "loss": 0.12535134106874465, "time": 0.5160489082336426, "epoch": 10, "memory": 6318, "step": 36233} +{"lr": 1.1315827710662908e-05, "data_time": 0.000906991958618164, "loss": 0.13251739665865897, "time": 0.900600814819336, "epoch": 10, "memory": 6318, "step": 36333} +{"lr": 1.1231019459499938e-05, "data_time": 0.0009277582168579102, "loss": 0.12785017788410186, "time": 0.47608993053436277, "epoch": 10, "memory": 6318, "step": 36433} +{"lr": 1.1148998183141688e-05, "data_time": 0.000840449333190918, "loss": 0.12100049778819084, "time": 0.4899609565734863, "epoch": 10, "memory": 6318, "step": 36533} +{"lr": 1.1069769104264779e-05, "data_time": 0.0008569478988647461, "loss": 0.1333050012588501, "time": 0.4979268550872803, "epoch": 10, "memory": 6318, "step": 36633} +{"lr": 1.0993337267753594e-05, "data_time": 0.0008299589157104492, "loss": 0.12846382558345795, "time": 0.5277811527252197, "epoch": 10, "memory": 6318, "step": 36733} +{"lr": 1.0919707540379088e-05, "data_time": 0.000736236572265625, "loss": 0.1328561447560787, "time": 0.5615181684494018, "epoch": 10, "memory": 6318, "step": 36833} +{"lr": 1.0848884610488858e-05, "data_time": 0.00095062255859375, "loss": 0.12305434346199036, "time": 0.5141811370849609, "epoch": 10, "memory": 6318, "step": 36933} +{"lr": 1.0780872987708645e-05, "data_time": 0.0008921623229980469, "loss": 0.1282338745892048, "time": 0.4944150447845459, "epoch": 10, "memory": 6318, "step": 37033} +{"lr": 1.0715677002655166e-05, "data_time": 0.0009976625442504883, "loss": 0.15263360738754272, "time": 0.5296923637390136, "epoch": 10, "memory": 6318, "step": 37133} +{"lr": 1.0653300806660372e-05, "data_time": 0.0009070873260498047, "loss": 0.10912934839725494, "time": 0.5134855508804321, "epoch": 10, "memory": 6318, "step": 37233} +{"lr": 1.0593748371507151e-05, "data_time": 0.0010263919830322266, "loss": 0.13477461710572242, "time": 0.5665380477905273, "epoch": 10, "memory": 6318, "step": 37333} +{"lr": 1.0537023489176325e-05, "data_time": 0.0008136749267578125, "loss": 0.14008670598268508, "time": 0.538142204284668, "epoch": 10, "memory": 6318, "step": 37433} +{"lr": 1.0483129771605328e-05, "data_time": 0.0007574558258056641, "loss": 0.1242856040596962, "time": 0.5409637689590454, "epoch": 10, "memory": 6318, "step": 37533} +{"lr": 1.0432070650458095e-05, "data_time": 0.000991082191467285, "loss": 0.14170307591557502, "time": 0.5007960319519043, "epoch": 10, "memory": 6318, "step": 37633} +{"lr": 1.0383849376906634e-05, "data_time": 0.0008839607238769531, "loss": 0.11963862031698227, "time": 0.525249719619751, "epoch": 10, "memory": 6318, "step": 37733} +{"lr": 1.0338469021423977e-05, "data_time": 0.000841832160949707, "loss": 0.13267244696617125, "time": 0.5253978252410889, "epoch": 10, "memory": 6318, "step": 37833} +{"lr": 1.029593247358868e-05, "data_time": 0.0009142875671386719, "loss": 0.11969713270664215, "time": 0.5296550035476685, "epoch": 10, "memory": 6318, "step": 37933} +{"lr": 1.0256242441900827e-05, "data_time": 0.0008339643478393555, "loss": 0.13607119396328926, "time": 0.5382554054260253, "epoch": 10, "memory": 6318, "step": 38033} +{"lr": 1.0219401453609533e-05, "data_time": 0.000739884376525879, "loss": 0.13219364285469054, "time": 0.5563067197799683, "epoch": 10, "memory": 6318, "step": 38133} +{"lr": 1.018541185455209e-05, "data_time": 0.0008130550384521484, "loss": 0.1302885577082634, "time": 0.5263590097427369, "epoch": 10, "memory": 6318, "step": 38233} +{"lr": 1.015427580900455e-05, "data_time": 0.0007379531860351562, "loss": 0.12939881533384323, "time": 0.5643665075302124, "epoch": 10, "memory": 6318, "step": 38333} +{"lr": 1.0125995299543914e-05, "data_time": 0.0007515192031860351, "loss": 0.12289058044552803, "time": 0.5224899768829345, "epoch": 10, "memory": 6318, "step": 38433} +{"lr": 1.0100572126921902e-05, "data_time": 0.0007193803787231446, "loss": 0.12374185994267464, "time": 0.5481310606002807, "epoch": 10, "memory": 6318, "step": 38533} +{"lr": 1.007800790995031e-05, "data_time": 0.0009577274322509766, "loss": 0.11785700395703316, "time": 0.5011749267578125, "epoch": 10, "memory": 6318, "step": 38633} +{"lr": 1.0058304085397905e-05, "data_time": 0.0008551836013793946, "loss": 0.12400827035307885, "time": 0.5307268381118775, "epoch": 10, "memory": 6318, "step": 38733} +{"lr": 1.0041461907898941e-05, "data_time": 0.0010039091110229492, "loss": 0.12385998293757439, "time": 0.846207070350647, "epoch": 10, "memory": 6318, "step": 38833} +{"lr": 1.0027482449873297e-05, "data_time": 0.0008578777313232421, "loss": 0.1244684875011444, "time": 0.5163285493850708, "epoch": 10, "memory": 6318, "step": 38933} +{"lr": 1.0016366601458162e-05, "data_time": 0.0009111881256103516, "loss": 0.13576598688960076, "time": 0.5722871065139771, "epoch": 10, "memory": 6318, "step": 39033} +{"lr": 1.0008115070451358e-05, "data_time": 0.0007571220397949219, "loss": 0.13201383501291275, "time": 0.5714087009429931, "epoch": 10, "memory": 6318, "step": 39133} +{"lr": 1.0002728382266284e-05, "data_time": 0.0007715702056884765, "loss": 0.13470007590949534, "time": 0.5762731313705445, "epoch": 10, "memory": 6318, "step": 39233} +{"lr": 1.0000206879898465e-05, "data_time": 0.0007406234741210938, "loss": 0.15056282579898833, "time": 0.5147181272506713, "epoch": 10, "memory": 6318, "step": 39333} +{"accuracy/top1": 95.4733657836914, "data_time": 0.4220685424475834, "time": 0.5078481312455803, "step": 10} diff --git a/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1/20230606_051509/vis_data/config.py b/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1/20230606_051509/vis_data/config.py new file mode 100644 index 0000000000000000000000000000000000000000..07f49b3c87537f52b4f76320195407e52d9fcb36 --- /dev/null +++ b/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1/20230606_051509/vis_data/config.py @@ -0,0 +1,247 @@ +optim_wrapper = dict( + optimizer=dict( + type='SGD', + lr=0.0001, + momentum=0.9, + weight_decay=0.0001, + _scope_='mmpretrain'), + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=False, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=512) +model = dict( + type='ImageClassifier', + backbone=dict( + frozen_stages=2, + type='ResNet', + depth=50, + num_stages=4, + out_indices=(3, ), + style='pytorch', + init_cfg=dict( + type='Pretrained', + checkpoint= + 'https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth', + prefix='backbone')), + neck=dict(type='GlobalAveragePooling'), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=2048, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + topk=1)) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/IF80w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/if-dpmsolver++-50-20w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/cc1m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/if-dpmsolver++-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = dict(type='Accuracy', topk=1) +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/if-dpmsolver++-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = dict(type='Accuracy', topk=1) +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1' diff --git a/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1/20230606_051509/vis_data/events.out.tfevents.1685999722.SH-IDC1-10-140-24-124.245811.0 b/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1/20230606_051509/vis_data/events.out.tfevents.1685999722.SH-IDC1-10-140-24-124.245811.0 new file mode 100644 index 0000000000000000000000000000000000000000..ca233501cc83ee8d8ae5728aabce75013ecf7700 --- /dev/null +++ b/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1/20230606_051509/vis_data/events.out.tfevents.1685999722.SH-IDC1-10-140-24-124.245811.0 @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:174b8a4a2e8b77418a0ca2f0bb216533234b647a1417b7e1141332b70ecd2f8b +size 2940382 diff --git a/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1/20230606_051509/vis_data/scalars.json b/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1/20230606_051509/vis_data/scalars.json new file mode 100644 index 0000000000000000000000000000000000000000..9d49377fbb2d936b6395471c65d52f190cab1533 --- /dev/null +++ b/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1/20230606_051509/vis_data/scalars.json @@ -0,0 +1,400 @@ +{"lr": 9.999859583040624e-05, "data_time": 0.25990498065948486, "loss": 0.6402046918869019, "time": 0.5295006036758423, "epoch": 1, "memory": 9436, "step": 100} +{"lr": 9.999432653394512e-05, "data_time": 0.5439581155776978, "loss": 0.5861357271671295, "time": 0.6861336708068848, "epoch": 1, "memory": 6319, "step": 200} +{"lr": 9.998719223919393e-05, "data_time": 0.5587227582931519, "loss": 0.5367321491241455, "time": 0.6989289999008179, "epoch": 1, "memory": 6319, "step": 300} +{"lr": 9.997719340042649e-05, "data_time": 0.5213950872421265, "loss": 0.48539854884147643, "time": 0.6626511812210083, "epoch": 1, "memory": 6319, "step": 400} +{"lr": 9.996433065431542e-05, "data_time": 0.5269391059875488, "loss": 0.43747425079345703, "time": 0.6674677848815918, "epoch": 1, "memory": 6319, "step": 500} +{"lr": 9.994860481989157e-05, "data_time": 0.5152857303619385, "loss": 0.39237564206123354, "time": 0.6596978187561036, "epoch": 1, "memory": 6319, "step": 600} +{"lr": 9.993001689849214e-05, "data_time": 1.0034260749816895, "loss": 0.3821966916322708, "time": 1.1468387842178345, "epoch": 1, "memory": 6319, "step": 700} +{"lr": 9.990856807369658e-05, "data_time": 0.4322436571121216, "loss": 0.35795452296733854, "time": 0.5726524353027344, "epoch": 1, "memory": 6319, "step": 800} +{"lr": 9.98842597112514e-05, "data_time": 0.48664350509643556, "loss": 0.33374722898006437, "time": 0.6277008056640625, "epoch": 1, "memory": 6319, "step": 900} +{"lr": 9.985709335898336e-05, "data_time": 0.4990594148635864, "loss": 0.32999047040939333, "time": 0.6407262325286865, "epoch": 1, "memory": 6319, "step": 1000} +{"lr": 9.982707074670047e-05, "data_time": 0.4646873712539673, "loss": 0.31953787207603457, "time": 0.6051527976989746, "epoch": 1, "memory": 6319, "step": 1100} +{"lr": 9.979419378608245e-05, "data_time": 0.5044169664382935, "loss": 0.2946313828229904, "time": 0.6464436292648316, "epoch": 1, "memory": 6319, "step": 1200} +{"lr": 9.975846457055832e-05, "data_time": 0.5260628938674927, "loss": 0.29008248448371887, "time": 0.6668855428695679, "epoch": 1, "memory": 6319, "step": 1300} +{"lr": 9.97198853751737e-05, "data_time": 0.5073223114013672, "loss": 0.3029852151870728, "time": 0.6472033977508544, "epoch": 1, "memory": 6319, "step": 1400} +{"lr": 9.967845865644554e-05, "data_time": 0.48812687397003174, "loss": 0.29525449872016907, "time": 0.6277704000473022, "epoch": 1, "memory": 6319, "step": 1500} +{"lr": 9.963418705220597e-05, "data_time": 0.493870210647583, "loss": 0.28693257868289945, "time": 0.6350355625152588, "epoch": 1, "memory": 6319, "step": 1600} +{"lr": 9.958707338143413e-05, "data_time": 0.4661304235458374, "loss": 0.29168210327625277, "time": 0.6068583488464355, "epoch": 1, "memory": 6319, "step": 1700} +{"lr": 9.953712064407689e-05, "data_time": 0.45297794342041015, "loss": 0.29011504650115966, "time": 0.5937061309814453, "epoch": 1, "memory": 6319, "step": 1800} +{"lr": 9.948433202085752e-05, "data_time": 0.5169339418411255, "loss": 0.27398703396320345, "time": 0.6577034235000611, "epoch": 1, "memory": 6319, "step": 1900} +{"lr": 9.942871087307363e-05, "data_time": 0.4975031137466431, "loss": 0.2566560238599777, "time": 0.6394488096237183, "epoch": 1, "memory": 6319, "step": 2000} +{"lr": 9.937026074238262e-05, "data_time": 0.5305297374725342, "loss": 0.2536715030670166, "time": 0.6727217197418213, "epoch": 1, "memory": 6319, "step": 2100} +{"lr": 9.930898535057657e-05, "data_time": 0.3923900365829468, "loss": 0.2435808628797531, "time": 0.5332682132720947, "epoch": 1, "memory": 6319, "step": 2200} +{"lr": 9.924488859934492e-05, "data_time": 0.45818946361541746, "loss": 0.262268927693367, "time": 0.5985460996627807, "epoch": 1, "memory": 6319, "step": 2300} +{"lr": 9.917797457002633e-05, "data_time": 0.42747619152069094, "loss": 0.24827308654785157, "time": 0.5756779193878174, "epoch": 1, "memory": 6319, "step": 2400} +{"lr": 9.910824752334866e-05, "data_time": 0.13895781040191652, "loss": 0.26804320961236955, "time": 0.5948331832885743, "epoch": 1, "memory": 6319, "step": 2500} +{"lr": 9.903571189915766e-05, "data_time": 0.0008750200271606445, "loss": 0.2405380979180336, "time": 0.5901623010635376, "epoch": 1, "memory": 6319, "step": 2600} +{"lr": 9.896037231613424e-05, "data_time": 0.0009298086166381836, "loss": 0.2365088403224945, "time": 0.6120794773101806, "epoch": 1, "memory": 6319, "step": 2700} +{"lr": 9.888223357150049e-05, "data_time": 0.0008474588394165039, "loss": 0.2465743228793144, "time": 0.6035980701446533, "epoch": 1, "memory": 6319, "step": 2800} +{"lr": 9.880130064071408e-05, "data_time": 0.000853586196899414, "loss": 0.23031350076198578, "time": 0.6611725330352783, "epoch": 1, "memory": 6319, "step": 2900} +{"lr": 9.87175786771517e-05, "data_time": 0.000911259651184082, "loss": 0.23938852399587632, "time": 0.6019812822341919, "epoch": 1, "memory": 6319, "step": 3000} +{"lr": 9.863107301178059e-05, "data_time": 0.001007819175720215, "loss": 0.24390990138053895, "time": 0.6488357782363892, "epoch": 1, "memory": 6319, "step": 3100} +{"lr": 9.854178915281925e-05, "data_time": 0.0009762287139892578, "loss": 0.24378345012664795, "time": 0.6051812410354614, "epoch": 1, "memory": 6319, "step": 3200} +{"lr": 9.84497327853868e-05, "data_time": 0.0009861469268798828, "loss": 0.22497162222862244, "time": 0.6643328189849853, "epoch": 1, "memory": 6319, "step": 3300} +{"lr": 9.83549097711407e-05, "data_time": 0.0009825944900512696, "loss": 0.24779136180877687, "time": 0.6440266609191895, "epoch": 1, "memory": 6319, "step": 3400} +{"lr": 9.825732614790392e-05, "data_time": 0.0009270668029785156, "loss": 0.2368963912129402, "time": 0.57118718624115, "epoch": 1, "memory": 6319, "step": 3500} +{"lr": 9.815698812928018e-05, "data_time": 0.0009441852569580078, "loss": 0.21808661967515947, "time": 0.5564011573791504, "epoch": 1, "memory": 6319, "step": 3600} +{"lr": 9.805390210425831e-05, "data_time": 0.0009526252746582031, "loss": 0.21776784658432008, "time": 0.5983808755874633, "epoch": 1, "memory": 6319, "step": 3700} +{"lr": 9.794807463680567e-05, "data_time": 0.0009661197662353515, "loss": 0.21626628786325455, "time": 0.5614094018936158, "epoch": 1, "memory": 6319, "step": 3800} +{"lr": 9.783951246544974e-05, "data_time": 0.000871586799621582, "loss": 0.2112686291337013, "time": 0.5717248439788818, "epoch": 1, "memory": 6319, "step": 3900} +{"accuracy/top1": 60.02574920654297, "data_time": 0.4936740942168654, "time": 0.5865166772875869, "step": 1} +{"lr": 9.768635527139677e-05, "data_time": 0.4531930685043335, "loss": 0.22331490516662597, "time": 0.5931108713150024, "epoch": 2, "memory": 8776, "step": 4037} +{"lr": 9.75713397738286e-05, "data_time": 0.24699089527130128, "loss": 0.21777475923299788, "time": 0.5014456748962403, "epoch": 2, "memory": 6318, "step": 4137} +{"lr": 9.745361356082025e-05, "data_time": 0.0008819341659545898, "loss": 0.22727794647216798, "time": 0.580863094329834, "epoch": 2, "memory": 6318, "step": 4237} +{"lr": 9.733318412854784e-05, "data_time": 0.000957942008972168, "loss": 0.23054793179035188, "time": 0.5685622930526734, "epoch": 2, "memory": 6318, "step": 4337} +{"lr": 9.721005914531412e-05, "data_time": 0.000832819938659668, "loss": 0.21719527691602708, "time": 0.5727081298828125, "epoch": 2, "memory": 6318, "step": 4437} +{"lr": 9.708424645106017e-05, "data_time": 0.0009802818298339845, "loss": 0.2230851247906685, "time": 0.5348587512969971, "epoch": 2, "memory": 6318, "step": 4537} +{"lr": 9.695575405686603e-05, "data_time": 0.0009038209915161133, "loss": 0.2079911008477211, "time": 0.5726252317428588, "epoch": 2, "memory": 6318, "step": 4637} +{"lr": 9.682459014444089e-05, "data_time": 0.0009235620498657226, "loss": 0.22554663568735123, "time": 0.5798112392425537, "epoch": 2, "memory": 6318, "step": 4737} +{"lr": 9.669076306560176e-05, "data_time": 0.0008735418319702149, "loss": 0.22549958974123002, "time": 0.5881304025650025, "epoch": 2, "memory": 6318, "step": 4837} +{"lr": 9.655428134174183e-05, "data_time": 0.0009006261825561523, "loss": 0.21149692237377166, "time": 0.556542444229126, "epoch": 2, "memory": 6318, "step": 4937} +{"lr": 9.641515366328807e-05, "data_time": 0.0008460760116577148, "loss": 0.20487848669290543, "time": 0.5929847478866577, "epoch": 2, "memory": 6318, "step": 5037} +{"lr": 9.627338888914757e-05, "data_time": 0.0009048700332641602, "loss": 0.22542578727006912, "time": 0.5736958742141723, "epoch": 2, "memory": 6318, "step": 5137} +{"lr": 9.612899604614378e-05, "data_time": 0.0009108781814575195, "loss": 0.20576413869857788, "time": 0.5921751260757446, "epoch": 2, "memory": 6318, "step": 5237} +{"lr": 9.598198432844131e-05, "data_time": 0.0009085416793823242, "loss": 0.20128974169492722, "time": 0.557023549079895, "epoch": 2, "memory": 6318, "step": 5337} +{"lr": 9.583236309696081e-05, "data_time": 0.0008863210678100586, "loss": 0.19630965441465378, "time": 0.551726245880127, "epoch": 2, "memory": 6318, "step": 5437} +{"lr": 9.568014187878279e-05, "data_time": 0.0008133649826049805, "loss": 0.20126755982637407, "time": 0.5702857255935669, "epoch": 2, "memory": 6318, "step": 5537} +{"lr": 9.552533036654103e-05, "data_time": 0.0008449077606201172, "loss": 0.21352028101682663, "time": 0.60773766040802, "epoch": 2, "memory": 6318, "step": 5637} +{"lr": 9.536793841780541e-05, "data_time": 0.000992894172668457, "loss": 0.20573764741420747, "time": 0.5825971126556396, "epoch": 2, "memory": 6318, "step": 5737} +{"lr": 9.52079760544543e-05, "data_time": 0.0008639097213745117, "loss": 0.21162491738796235, "time": 0.5742067098617554, "epoch": 2, "memory": 6318, "step": 5837} +{"lr": 9.504545346203619e-05, "data_time": 0.0008871793746948242, "loss": 0.18542599380016328, "time": 0.5614900112152099, "epoch": 2, "memory": 6318, "step": 5937} +{"lr": 9.488038098912122e-05, "data_time": 0.000901341438293457, "loss": 0.2020304471254349, "time": 0.74519784450531, "epoch": 2, "memory": 6318, "step": 6037} +{"lr": 9.471276914664257e-05, "data_time": 0.0009241104125976562, "loss": 0.2121737465262413, "time": 0.5834649085998536, "epoch": 2, "memory": 6318, "step": 6137} +{"lr": 9.454262860722667e-05, "data_time": 0.001154470443725586, "loss": 0.20481036901473998, "time": 0.5199038028717041, "epoch": 2, "memory": 6318, "step": 6237} +{"lr": 9.436997020451398e-05, "data_time": 0.0009418964385986328, "loss": 0.2028192847967148, "time": 0.5328365802764893, "epoch": 2, "memory": 6318, "step": 6337} +{"lr": 9.419480493246903e-05, "data_time": 0.0009444713592529297, "loss": 0.18822189718484877, "time": 0.5861220359802246, "epoch": 2, "memory": 6318, "step": 6437} +{"lr": 9.401714394468036e-05, "data_time": 0.0008361577987670898, "loss": 0.19415918588638306, "time": 0.5689403057098389, "epoch": 2, "memory": 6318, "step": 6537} +{"lr": 9.383699855365022e-05, "data_time": 0.000895547866821289, "loss": 0.21370866596698762, "time": 0.542294192314148, "epoch": 2, "memory": 6318, "step": 6637} +{"lr": 9.365438023007469e-05, "data_time": 0.0010315656661987304, "loss": 0.20446225702762605, "time": 0.5685359239578247, "epoch": 2, "memory": 6318, "step": 6737} +{"lr": 9.346930060211267e-05, "data_time": 0.0009064674377441406, "loss": 0.2243996813893318, "time": 0.54306640625, "epoch": 2, "memory": 6318, "step": 6837} +{"lr": 9.328177145464599e-05, "data_time": 0.0009489059448242188, "loss": 0.19088304787874222, "time": 0.5704697847366333, "epoch": 2, "memory": 6318, "step": 6937} +{"lr": 9.309180472852871e-05, "data_time": 0.0008680343627929688, "loss": 0.20378941744565965, "time": 0.585848617553711, "epoch": 2, "memory": 6318, "step": 7037} +{"lr": 9.289941251982683e-05, "data_time": 0.0009253025054931641, "loss": 0.1844475522637367, "time": 0.5599412202835083, "epoch": 2, "memory": 6318, "step": 7137} +{"lr": 9.270460707904827e-05, "data_time": 0.0009270906448364258, "loss": 0.18135594725608825, "time": 0.5775460243225098, "epoch": 2, "memory": 6318, "step": 7237} +{"lr": 9.250740081036243e-05, "data_time": 0.000993657112121582, "loss": 0.18333010673522948, "time": 0.559870719909668, "epoch": 2, "memory": 6318, "step": 7337} +{"lr": 9.230780627081088e-05, "data_time": 0.0009291172027587891, "loss": 0.17100572139024733, "time": 0.5388106346130371, "epoch": 2, "memory": 6318, "step": 7437} +{"lr": 9.210583616950722e-05, "data_time": 0.0008291482925415039, "loss": 0.21008660644292831, "time": 0.5797322034835816, "epoch": 2, "memory": 6318, "step": 7537} +{"lr": 9.190150336682842e-05, "data_time": 0.0008919954299926758, "loss": 0.1646827444434166, "time": 0.5979239702224731, "epoch": 2, "memory": 6318, "step": 7637} +{"lr": 9.16948208735955e-05, "data_time": 0.000911879539489746, "loss": 0.18441036492586135, "time": 0.5740241050720215, "epoch": 2, "memory": 6318, "step": 7737} +{"lr": 9.148580185024521e-05, "data_time": 0.0010335922241210937, "loss": 0.18545248359441757, "time": 0.605552339553833, "epoch": 2, "memory": 6318, "step": 7837} +{"accuracy/top1": 84.0089111328125, "data_time": 0.4414039809128334, "time": 0.5312507851370449, "step": 2} +{"lr": 9.119567684509738e-05, "data_time": 0.31747140884399416, "loss": 0.18982364982366562, "time": 0.5918580055236816, "epoch": 3, "memory": 6318, "step": 7974} +{"lr": 9.098117366468421e-05, "data_time": 0.17231969833374022, "loss": 0.17452817857265474, "time": 1.284656810760498, "epoch": 3, "memory": 6318, "step": 8074} +{"lr": 9.076437939539471e-05, "data_time": 0.0007827043533325196, "loss": 0.17449144423007965, "time": 0.546287226676941, "epoch": 3, "memory": 6318, "step": 8174} +{"lr": 9.05453078415294e-05, "data_time": 0.0009041309356689453, "loss": 0.18233867287635802, "time": 0.592623257637024, "epoch": 3, "memory": 6318, "step": 8274} +{"lr": 9.032397295239427e-05, "data_time": 0.0011610031127929688, "loss": 0.1747233659029007, "time": 0.5693724393844605, "epoch": 3, "memory": 6318, "step": 8374} +{"lr": 9.010038882141241e-05, "data_time": 0.0008029460906982422, "loss": 0.1854666903614998, "time": 0.5913819551467896, "epoch": 3, "memory": 6318, "step": 8474} +{"lr": 8.987456968522655e-05, "data_time": 0.0010127544403076172, "loss": 0.19007351696491243, "time": 0.5473955869674683, "epoch": 3, "memory": 6318, "step": 8574} +{"lr": 8.964652992279262e-05, "data_time": 0.0008792638778686524, "loss": 0.18883423656225204, "time": 0.578046727180481, "epoch": 3, "memory": 6318, "step": 8674} +{"lr": 8.941628405446428e-05, "data_time": 0.0009624004364013672, "loss": 0.173228719830513, "time": 0.5761780977249146, "epoch": 3, "memory": 6318, "step": 8774} +{"lr": 8.918384674106797e-05, "data_time": 0.0008363723754882812, "loss": 0.17672196328639983, "time": 0.5682754516601562, "epoch": 3, "memory": 6318, "step": 8874} +{"lr": 8.894923278296992e-05, "data_time": 0.0009339570999145508, "loss": 0.2008403167128563, "time": 0.5964117050170898, "epoch": 3, "memory": 6318, "step": 8974} +{"lr": 8.871245711913324e-05, "data_time": 0.0011966705322265625, "loss": 0.17237788140773774, "time": 0.5893830299377442, "epoch": 3, "memory": 6318, "step": 9074} +{"lr": 8.847353482616703e-05, "data_time": 0.0009474277496337891, "loss": 0.17052201330661773, "time": 0.5614013195037841, "epoch": 3, "memory": 6318, "step": 9174} +{"lr": 8.823248111736638e-05, "data_time": 0.0008767843246459961, "loss": 0.17628194242715836, "time": 0.5740657806396484, "epoch": 3, "memory": 6318, "step": 9274} +{"lr": 8.798931134174327e-05, "data_time": 0.0008599519729614258, "loss": 0.16591138392686844, "time": 0.5469395637512207, "epoch": 3, "memory": 6318, "step": 9374} +{"lr": 8.774404098304961e-05, "data_time": 0.001050257682800293, "loss": 0.1798093795776367, "time": 0.5934871435165405, "epoch": 3, "memory": 6318, "step": 9474} +{"lr": 8.749668565879136e-05, "data_time": 0.000908970832824707, "loss": 0.17558502554893493, "time": 0.551546573638916, "epoch": 3, "memory": 6318, "step": 9574} +{"lr": 8.724726111923386e-05, "data_time": 0.0009049654006958007, "loss": 0.17216941863298416, "time": 0.5799022197723389, "epoch": 3, "memory": 6318, "step": 9674} +{"lr": 8.699578324639898e-05, "data_time": 0.0008556842803955078, "loss": 0.16674857661128045, "time": 0.5886175870895386, "epoch": 3, "memory": 6318, "step": 9774} +{"lr": 8.67422680530539e-05, "data_time": 0.0009267807006835937, "loss": 0.1592809021472931, "time": 0.569447112083435, "epoch": 3, "memory": 6318, "step": 9874} +{"lr": 8.648673168169147e-05, "data_time": 0.0008530139923095704, "loss": 0.17575456351041793, "time": 0.5851914882659912, "epoch": 3, "memory": 6318, "step": 9974} +{"lr": 8.622919040350244e-05, "data_time": 0.0008959293365478516, "loss": 0.17296219244599342, "time": 0.6134592533111572, "epoch": 3, "memory": 6318, "step": 10074} +{"lr": 8.596966061733922e-05, "data_time": 0.0010569334030151368, "loss": 0.18186536133289338, "time": 0.5560044288635254, "epoch": 3, "memory": 6318, "step": 10174} +{"lr": 8.570815884867185e-05, "data_time": 0.0008459329605102539, "loss": 0.17924930304288864, "time": 0.5501826047897339, "epoch": 3, "memory": 6318, "step": 10274} +{"lr": 8.544470174853563e-05, "data_time": 0.0011171579360961914, "loss": 0.16258630454540252, "time": 0.6082868337631225, "epoch": 3, "memory": 6318, "step": 10374} +{"lr": 8.517930609247077e-05, "data_time": 0.0008495330810546875, "loss": 0.17577800005674363, "time": 0.5666185617446899, "epoch": 3, "memory": 6318, "step": 10474} +{"lr": 8.491198877945477e-05, "data_time": 0.0009595394134521484, "loss": 0.17118353992700577, "time": 0.6048997640609741, "epoch": 3, "memory": 6318, "step": 10574} +{"lr": 8.464276683082565e-05, "data_time": 0.0010280370712280273, "loss": 0.1630304455757141, "time": 0.5279735803604126, "epoch": 3, "memory": 6318, "step": 10674} +{"lr": 8.437165738919852e-05, "data_time": 0.0011461019515991212, "loss": 0.17769434154033661, "time": 0.5584010124206543, "epoch": 3, "memory": 6318, "step": 10774} +{"lr": 8.409867771737398e-05, "data_time": 0.0008713245391845703, "loss": 0.15921278446912765, "time": 0.5352803945541382, "epoch": 3, "memory": 6318, "step": 10874} +{"lr": 8.382384519723895e-05, "data_time": 0.0009061336517333984, "loss": 0.18028637617826462, "time": 0.5936290502548218, "epoch": 3, "memory": 6318, "step": 10974} +{"lr": 8.35471773286596e-05, "data_time": 0.0009564638137817382, "loss": 0.16379580199718474, "time": 0.596064281463623, "epoch": 3, "memory": 6318, "step": 11074} +{"lr": 8.326869172836765e-05, "data_time": 0.0009354591369628906, "loss": 0.19253600984811783, "time": 0.5805278539657592, "epoch": 3, "memory": 6318, "step": 11174} +{"lr": 8.298840612883789e-05, "data_time": 0.0010148048400878905, "loss": 0.16079765260219575, "time": 0.5537952184677124, "epoch": 3, "memory": 6318, "step": 11274} +{"lr": 8.270633837715962e-05, "data_time": 0.0009637832641601563, "loss": 0.1523275852203369, "time": 0.5589352369308471, "epoch": 3, "memory": 6318, "step": 11374} +{"lr": 8.242250643390022e-05, "data_time": 0.0008406639099121094, "loss": 0.16228190660476685, "time": 0.5637318372726441, "epoch": 3, "memory": 6318, "step": 11474} +{"lr": 8.2136928371961e-05, "data_time": 0.0009596824645996093, "loss": 0.16325031816959382, "time": 0.5646752834320068, "epoch": 3, "memory": 6318, "step": 11574} +{"lr": 8.184962237542712e-05, "data_time": 0.0009074211120605469, "loss": 0.15824640095233916, "time": 0.569823431968689, "epoch": 3, "memory": 6318, "step": 11674} +{"lr": 8.156060673840912e-05, "data_time": 0.0008990049362182617, "loss": 0.16078418865799904, "time": 0.5731776714324951, "epoch": 3, "memory": 6318, "step": 11774} +{"accuracy/top1": 90.72405242919922, "data_time": 0.44536342291996395, "time": 0.5320818670864763, "step": 3} +{"lr": 8.116191337909002e-05, "data_time": 0.2445814847946167, "loss": 0.16252445727586745, "time": 0.5837427139282226, "epoch": 4, "memory": 6318, "step": 11911} +{"lr": 8.086891958166228e-05, "data_time": 0.0008585929870605468, "loss": 0.1551465429365635, "time": 0.5978672504425049, "epoch": 4, "memory": 6318, "step": 12011} +{"lr": 8.057427858966412e-05, "data_time": 0.0008853912353515625, "loss": 0.1561238296329975, "time": 0.5647099733352661, "epoch": 4, "memory": 6318, "step": 12111} +{"lr": 8.027800916425949e-05, "data_time": 0.000876927375793457, "loss": 0.15540698170661926, "time": 0.5088225126266479, "epoch": 4, "memory": 6318, "step": 12211} +{"lr": 7.998013017030234e-05, "data_time": 0.0009682178497314453, "loss": 0.16109662726521493, "time": 0.5707410335540771, "epoch": 4, "memory": 6318, "step": 12311} +{"lr": 7.968066057513528e-05, "data_time": 0.0008210897445678711, "loss": 0.17002131566405296, "time": 0.564606761932373, "epoch": 4, "memory": 6318, "step": 12411} +{"lr": 7.937961944738199e-05, "data_time": 0.0010721206665039063, "loss": 0.1626214936375618, "time": 0.6076979398727417, "epoch": 4, "memory": 6318, "step": 12511} +{"lr": 7.907702595573289e-05, "data_time": 0.0009151220321655274, "loss": 0.15160149782896043, "time": 0.5591695070266723, "epoch": 4, "memory": 6318, "step": 12611} +{"lr": 7.877289936772454e-05, "data_time": 0.0009728193283081055, "loss": 0.1789689317345619, "time": 0.6144480466842651, "epoch": 4, "memory": 6318, "step": 12711} +{"lr": 7.8467259048513e-05, "data_time": 0.0008974075317382812, "loss": 0.16176287233829498, "time": 0.5863381147384643, "epoch": 4, "memory": 6318, "step": 12811} +{"lr": 7.816012445964079e-05, "data_time": 0.000920557975769043, "loss": 0.1543870359659195, "time": 0.6008711576461792, "epoch": 4, "memory": 6318, "step": 12911} +{"lr": 7.785151515779718e-05, "data_time": 0.0009092092514038086, "loss": 0.14510656893253326, "time": 0.5613490343093872, "epoch": 4, "memory": 6318, "step": 13011} +{"lr": 7.754145079357376e-05, "data_time": 0.0009084224700927734, "loss": 0.152776862680912, "time": 0.553636622428894, "epoch": 4, "memory": 6318, "step": 13111} +{"lr": 7.722995111021218e-05, "data_time": 0.0009069442749023438, "loss": 0.1458202764391899, "time": 0.6177518606185913, "epoch": 4, "memory": 6318, "step": 13211} +{"lr": 7.691703594234788e-05, "data_time": 0.0012085914611816406, "loss": 0.16952790766954423, "time": 0.5421665430068969, "epoch": 4, "memory": 6318, "step": 13311} +{"lr": 7.660272521474673e-05, "data_time": 0.0008891105651855468, "loss": 0.14836854189634324, "time": 0.5812160015106201, "epoch": 4, "memory": 6318, "step": 13411} +{"lr": 7.628703894103628e-05, "data_time": 0.0009837865829467774, "loss": 0.1634357288479805, "time": 0.5678518056869507, "epoch": 4, "memory": 6318, "step": 13511} +{"lr": 7.596999722243168e-05, "data_time": 0.0009011030197143555, "loss": 0.15797903537750244, "time": 0.5531563997268677, "epoch": 4, "memory": 6318, "step": 13611} +{"lr": 7.565162024645551e-05, "data_time": 0.0009408712387084961, "loss": 0.15172194316983223, "time": 0.5672452926635743, "epoch": 4, "memory": 6318, "step": 13711} +{"lr": 7.533192828565237e-05, "data_time": 0.000873565673828125, "loss": 0.14865531995892525, "time": 0.5387712001800538, "epoch": 4, "memory": 6318, "step": 13811} +{"lr": 7.501094169629802e-05, "data_time": 0.0008395195007324218, "loss": 0.16808363944292068, "time": 0.9345919370651246, "epoch": 4, "memory": 6318, "step": 13911} +{"lr": 7.468868091710333e-05, "data_time": 0.0008895635604858398, "loss": 0.16219151765108109, "time": 0.5878986120223999, "epoch": 4, "memory": 6318, "step": 14011} +{"lr": 7.436516646791276e-05, "data_time": 0.0009893655776977539, "loss": 0.16593725457787514, "time": 0.5491437911987305, "epoch": 4, "memory": 6318, "step": 14111} +{"lr": 7.404041894839768e-05, "data_time": 0.001008296012878418, "loss": 0.1606047883629799, "time": 0.607050108909607, "epoch": 4, "memory": 6318, "step": 14211} +{"lr": 7.371445903674494e-05, "data_time": 0.0009079217910766601, "loss": 0.15264861583709716, "time": 0.4697632074356079, "epoch": 4, "memory": 6318, "step": 14311} +{"lr": 7.338730748834014e-05, "data_time": 0.001061415672302246, "loss": 0.14325622469186783, "time": 0.5767011404037475, "epoch": 4, "memory": 6318, "step": 14411} +{"lr": 7.305898513444536e-05, "data_time": 0.0011385202407836914, "loss": 0.16008485481142998, "time": 0.6132961511611938, "epoch": 4, "memory": 6318, "step": 14511} +{"lr": 7.27295128808741e-05, "data_time": 0.0008663415908813476, "loss": 0.16022886633872985, "time": 0.6077814817428588, "epoch": 4, "memory": 6318, "step": 14611} +{"lr": 7.239891170665884e-05, "data_time": 0.0009595870971679688, "loss": 0.15783049985766412, "time": 0.5469722509384155, "epoch": 4, "memory": 6318, "step": 14711} +{"lr": 7.206720266271582e-05, "data_time": 0.0009441375732421875, "loss": 0.14166781231760978, "time": 0.6376216411590576, "epoch": 4, "memory": 6318, "step": 14811} +{"lr": 7.173440687050449e-05, "data_time": 0.0008619308471679687, "loss": 0.14800475165247917, "time": 0.5642582654953003, "epoch": 4, "memory": 6318, "step": 14911} +{"lr": 7.140054552068265e-05, "data_time": 0.0009479284286499024, "loss": 0.15040959268808365, "time": 0.5604031562805176, "epoch": 4, "memory": 6318, "step": 15011} +{"lr": 7.106563987175705e-05, "data_time": 0.0009463548660278321, "loss": 0.16903550773859025, "time": 0.6342783689498901, "epoch": 4, "memory": 6318, "step": 15111} +{"lr": 7.072971124873013e-05, "data_time": 0.0009508371353149414, "loss": 0.15986801832914352, "time": 0.59492027759552, "epoch": 4, "memory": 6318, "step": 15211} +{"lr": 7.039278104174143e-05, "data_time": 0.0009119987487792968, "loss": 0.14383111372590066, "time": 0.579813814163208, "epoch": 4, "memory": 6318, "step": 15311} +{"lr": 7.005487070470615e-05, "data_time": 0.0009538650512695313, "loss": 0.16456588879227638, "time": 0.5852491617202759, "epoch": 4, "memory": 6318, "step": 15411} +{"lr": 6.971600175394891e-05, "data_time": 0.000942063331604004, "loss": 0.14582718461751937, "time": 0.6001242160797119, "epoch": 4, "memory": 6318, "step": 15511} +{"lr": 6.937619576683373e-05, "data_time": 0.0008599042892456054, "loss": 0.1710476577281952, "time": 0.5925651788711548, "epoch": 4, "memory": 6318, "step": 15611} +{"lr": 6.903547438039021e-05, "data_time": 0.000958395004272461, "loss": 0.14759338274598122, "time": 0.5645423650741577, "epoch": 4, "memory": 6318, "step": 15711} +{"accuracy/top1": 92.74208068847656, "data_time": 0.4560331969425596, "time": 0.5418031955587452, "step": 4} +{"lr": 6.856723955077045e-05, "data_time": 0.0007763147354125976, "loss": 0.14500877037644386, "time": 0.5487354040145874, "epoch": 5, "memory": 6318, "step": 15848} +{"lr": 6.822443541062844e-05, "data_time": 0.000841522216796875, "loss": 0.15268248543143273, "time": 0.6195852518081665, "epoch": 5, "memory": 6318, "step": 15948} +{"lr": 6.788078920910325e-05, "data_time": 0.0008341312408447265, "loss": 0.15663531944155692, "time": 0.5848025798797607, "epoch": 5, "memory": 6318, "step": 16048} +{"lr": 6.753632282774874e-05, "data_time": 0.0009988784790039063, "loss": 0.15814314037561417, "time": 0.5539665937423706, "epoch": 5, "memory": 6318, "step": 16148} +{"lr": 6.719105820034336e-05, "data_time": 0.0011941194534301758, "loss": 0.1559486523270607, "time": 0.5831831455230713, "epoch": 5, "memory": 6318, "step": 16248} +{"lr": 6.68450173114937e-05, "data_time": 0.00100252628326416, "loss": 0.16272845044732093, "time": 0.5619689702987671, "epoch": 5, "memory": 6318, "step": 16348} +{"lr": 6.64982221952347e-05, "data_time": 0.0009701013565063477, "loss": 0.13593565300107002, "time": 0.5660976648330689, "epoch": 5, "memory": 6318, "step": 16448} +{"lr": 6.615069493362599e-05, "data_time": 0.0009341716766357421, "loss": 0.16524188295006753, "time": 0.5730363368988037, "epoch": 5, "memory": 6318, "step": 16548} +{"lr": 6.580245765534682e-05, "data_time": 0.0010108709335327148, "loss": 0.14973906874656678, "time": 0.6023453950881958, "epoch": 5, "memory": 6318, "step": 16648} +{"lr": 6.54535325342862e-05, "data_time": 0.0009762048721313477, "loss": 0.1397172324359417, "time": 0.5546247005462647, "epoch": 5, "memory": 6318, "step": 16748} +{"lr": 6.510394178813155e-05, "data_time": 0.0009490489959716797, "loss": 0.16103667989373208, "time": 0.5314825773239136, "epoch": 5, "memory": 6318, "step": 16848} +{"lr": 6.475370767695331e-05, "data_time": 0.00085601806640625, "loss": 0.14232759326696395, "time": 0.5730009555816651, "epoch": 5, "memory": 6318, "step": 16948} +{"lr": 6.440285250178845e-05, "data_time": 0.0009927749633789062, "loss": 0.16256664842367172, "time": 0.5520715236663818, "epoch": 5, "memory": 6318, "step": 17048} +{"lr": 6.405139860321955e-05, "data_time": 0.0010034799575805663, "loss": 0.14732514321804047, "time": 0.5918293237686157, "epoch": 5, "memory": 6318, "step": 17148} +{"lr": 6.369936835995306e-05, "data_time": 0.0009931564331054688, "loss": 0.15432692989706992, "time": 0.6313480377197266, "epoch": 5, "memory": 6318, "step": 17248} +{"lr": 6.334678418739388e-05, "data_time": 0.0009126901626586914, "loss": 0.14209817796945573, "time": 0.5824315547943115, "epoch": 5, "memory": 6318, "step": 17348} +{"lr": 6.299366853621796e-05, "data_time": 0.0009621381759643555, "loss": 0.15826880782842637, "time": 0.5913075923919677, "epoch": 5, "memory": 6318, "step": 17448} +{"lr": 6.264004389094315e-05, "data_time": 0.0010961532592773438, "loss": 0.14500583633780478, "time": 0.5645835876464844, "epoch": 5, "memory": 6318, "step": 17548} +{"lr": 6.22859327684974e-05, "data_time": 0.0009605646133422852, "loss": 0.13999637290835382, "time": 0.5523180484771728, "epoch": 5, "memory": 6318, "step": 17648} +{"lr": 6.193135771678478e-05, "data_time": 0.0011732101440429688, "loss": 0.14428636059165, "time": 0.5800431489944458, "epoch": 5, "memory": 6318, "step": 17748} +{"lr": 6.157634131324983e-05, "data_time": 0.000865316390991211, "loss": 0.15345237776637077, "time": 0.5654687404632568, "epoch": 5, "memory": 6318, "step": 17848} +{"lr": 6.12209061634402e-05, "data_time": 0.0009256839752197265, "loss": 0.1531187653541565, "time": 0.5931895494461059, "epoch": 5, "memory": 6318, "step": 17948} +{"lr": 6.086507489956686e-05, "data_time": 0.0012007951736450195, "loss": 0.1524558886885643, "time": 0.4993189811706543, "epoch": 5, "memory": 6318, "step": 18048} +{"lr": 6.0508870179063345e-05, "data_time": 0.0009344816207885742, "loss": 0.1451650455594063, "time": 0.5356508255004883, "epoch": 5, "memory": 6318, "step": 18148} +{"lr": 6.015231468314303e-05, "data_time": 0.001004505157470703, "loss": 0.15471096336841583, "time": 0.5358565807342529, "epoch": 5, "memory": 6318, "step": 18248} +{"lr": 5.979543111535473e-05, "data_time": 0.0008063554763793945, "loss": 0.13539359495043754, "time": 0.5862555503845215, "epoch": 5, "memory": 6318, "step": 18348} +{"lr": 5.943824220013694e-05, "data_time": 0.0009228467941284179, "loss": 0.15212441831827164, "time": 0.5795838117599488, "epoch": 5, "memory": 6318, "step": 18448} +{"lr": 5.908077068137128e-05, "data_time": 0.0008740901947021484, "loss": 0.15683663859963418, "time": 0.5718268394470215, "epoch": 5, "memory": 6318, "step": 18548} +{"lr": 5.872303932093392e-05, "data_time": 0.0009355545043945312, "loss": 0.1608651399612427, "time": 0.5510890245437622, "epoch": 5, "memory": 6318, "step": 18648} +{"lr": 5.836507089724633e-05, "data_time": 0.000998210906982422, "loss": 0.14838469922542571, "time": 0.5488568305969238, "epoch": 5, "memory": 6318, "step": 18748} +{"lr": 5.800688820382498e-05, "data_time": 0.000847625732421875, "loss": 0.1373772993683815, "time": 0.5856449365615845, "epoch": 5, "memory": 6318, "step": 18848} +{"lr": 5.7648514047829945e-05, "data_time": 0.0008908271789550781, "loss": 0.13893911689519883, "time": 0.573755931854248, "epoch": 5, "memory": 6318, "step": 18948} +{"lr": 5.7289971248612474e-05, "data_time": 0.0008938312530517578, "loss": 0.15458082184195518, "time": 0.5382607221603394, "epoch": 5, "memory": 6318, "step": 19048} +{"lr": 5.693128263626215e-05, "data_time": 0.0010591745376586914, "loss": 0.13220496401190757, "time": 0.5498858451843261, "epoch": 5, "memory": 6318, "step": 19148} +{"lr": 5.6572471050153236e-05, "data_time": 0.0009000539779663086, "loss": 0.13859168887138368, "time": 0.5480077266693115, "epoch": 5, "memory": 6318, "step": 19248} +{"lr": 5.621355933749018e-05, "data_time": 0.0009001970291137695, "loss": 0.14990852773189545, "time": 0.5767415761947632, "epoch": 5, "memory": 6318, "step": 19348} +{"lr": 5.5854570351853106e-05, "data_time": 0.0009374856948852539, "loss": 0.12843604907393455, "time": 0.5519895553588867, "epoch": 5, "memory": 6318, "step": 19448} +{"lr": 5.5495526951742315e-05, "data_time": 0.0009349584579467773, "loss": 0.14197040423750878, "time": 0.5452033519744873, "epoch": 5, "memory": 6318, "step": 19548} +{"lr": 5.513645199912304e-05, "data_time": 0.0011122703552246093, "loss": 0.13910222202539443, "time": 0.5302355766296387, "epoch": 5, "memory": 6318, "step": 19648} +{"accuracy/top1": 93.65018463134766, "data_time": 0.44459915983265846, "time": 0.5333290798910733, "step": 5} +{"lr": 5.464450978670876e-05, "data_time": 0.05841796398162842, "loss": 0.14123359322547913, "time": 0.5865522861480713, "epoch": 6, "memory": 6318, "step": 19785} +{"lr": 5.4285451361545876e-05, "data_time": 0.0008589029312133789, "loss": 0.13904216289520263, "time": 0.5427911281585693, "epoch": 6, "memory": 6318, "step": 19885} +{"lr": 5.3926438435022166e-05, "data_time": 0.0009965896606445312, "loss": 0.13401658236980438, "time": 0.5614965677261352, "epoch": 6, "memory": 6318, "step": 19985} +{"lr": 5.356749386716241e-05, "data_time": 0.0008932113647460938, "loss": 0.13127534538507463, "time": 0.5408131122589112, "epoch": 6, "memory": 6318, "step": 20085} +{"lr": 5.320864051363873e-05, "data_time": 0.0009876251220703124, "loss": 0.13893340304493904, "time": 0.5377917051315307, "epoch": 6, "memory": 6318, "step": 20185} +{"lr": 5.2849901224315175e-05, "data_time": 0.0009930133819580078, "loss": 0.1533495657145977, "time": 0.5118759155273438, "epoch": 6, "memory": 6318, "step": 20285} +{"lr": 5.249129884179276e-05, "data_time": 0.0008838176727294922, "loss": 0.1379082404077053, "time": 0.511989974975586, "epoch": 6, "memory": 6318, "step": 20385} +{"lr": 5.2132856199954985e-05, "data_time": 0.0008897066116333007, "loss": 0.13805076330900193, "time": 0.541465449333191, "epoch": 6, "memory": 6318, "step": 20485} +{"lr": 5.177459612251411e-05, "data_time": 0.000903010368347168, "loss": 0.15050484389066696, "time": 0.562174940109253, "epoch": 6, "memory": 6318, "step": 20585} +{"lr": 5.141654142155744e-05, "data_time": 0.0009062528610229492, "loss": 0.13774992823600768, "time": 0.5470547914505005, "epoch": 6, "memory": 6318, "step": 20685} +{"lr": 5.105871489609512e-05, "data_time": 0.0009209632873535157, "loss": 0.14223484694957733, "time": 0.4802698135375977, "epoch": 6, "memory": 6318, "step": 20785} +{"lr": 5.070113933060828e-05, "data_time": 0.0008120059967041016, "loss": 0.13637217432260512, "time": 0.5403118371963501, "epoch": 6, "memory": 6318, "step": 20885} +{"lr": 5.0343837493598264e-05, "data_time": 0.0008483648300170899, "loss": 0.14491448029875756, "time": 0.5779665946960449, "epoch": 6, "memory": 6318, "step": 20985} +{"lr": 4.998683213613675e-05, "data_time": 0.0008725404739379882, "loss": 0.12851423695683478, "time": 0.5560952425003052, "epoch": 6, "memory": 6318, "step": 21085} +{"lr": 4.963014599041739e-05, "data_time": 0.0008286714553833007, "loss": 0.15191066414117813, "time": 0.5163877964019775, "epoch": 6, "memory": 6318, "step": 21185} +{"lr": 4.927380176830793e-05, "data_time": 0.0008488178253173828, "loss": 0.13374930918216704, "time": 0.5363269567489624, "epoch": 6, "memory": 6318, "step": 21285} +{"lr": 4.89178221599044e-05, "data_time": 0.000800013542175293, "loss": 0.14563221335411072, "time": 0.5094030380249024, "epoch": 6, "memory": 6318, "step": 21385} +{"lr": 4.85622298320861e-05, "data_time": 0.0009118318557739258, "loss": 0.1324270635843277, "time": 0.559794545173645, "epoch": 6, "memory": 6318, "step": 21485} +{"lr": 4.8207047427072355e-05, "data_time": 0.0007805347442626954, "loss": 0.16379940062761306, "time": 0.5867275953292846, "epoch": 6, "memory": 6318, "step": 21585} +{"lr": 4.7852297560980876e-05, "data_time": 0.0010346651077270507, "loss": 0.1348823480308056, "time": 0.5411486625671387, "epoch": 6, "memory": 6318, "step": 21685} +{"lr": 4.749800282238749e-05, "data_time": 0.0009227991104125977, "loss": 0.13596554175019265, "time": 0.5152548789978028, "epoch": 6, "memory": 6318, "step": 21785} +{"lr": 4.714418577088805e-05, "data_time": 0.0009283304214477539, "loss": 0.1340097300708294, "time": 0.5270004987716674, "epoch": 6, "memory": 6318, "step": 21885} +{"lr": 4.679086893566184e-05, "data_time": 0.000896286964416504, "loss": 0.13446456789970399, "time": 0.5040998458862305, "epoch": 6, "memory": 6318, "step": 21985} +{"lr": 4.643807481403694e-05, "data_time": 0.0008671760559082031, "loss": 0.1462169148027897, "time": 0.5261975049972534, "epoch": 6, "memory": 6318, "step": 22085} +{"lr": 4.6085825870057946e-05, "data_time": 0.0008907794952392578, "loss": 0.13258469328284264, "time": 0.4997079849243164, "epoch": 6, "memory": 6318, "step": 22185} +{"lr": 4.573414453305537e-05, "data_time": 0.0008089303970336914, "loss": 0.14285623654723167, "time": 0.5209615468978882, "epoch": 6, "memory": 6318, "step": 22285} +{"lr": 4.5383053196217514e-05, "data_time": 0.0008890390396118164, "loss": 0.14570192396640777, "time": 0.5254142999649047, "epoch": 6, "memory": 6318, "step": 22385} +{"lr": 4.503257421516477e-05, "data_time": 0.0008598804473876953, "loss": 0.1358324870467186, "time": 0.4974297761917114, "epoch": 6, "memory": 6318, "step": 22485} +{"lr": 4.468272990652586e-05, "data_time": 0.0008741617202758789, "loss": 0.13534300327301024, "time": 0.5563331604003906, "epoch": 6, "memory": 6318, "step": 22585} +{"lr": 4.433354254651697e-05, "data_time": 0.0009692907333374023, "loss": 0.13278524056077004, "time": 0.5399020671844482, "epoch": 6, "memory": 6318, "step": 22685} +{"lr": 4.398503436952338e-05, "data_time": 0.0009827136993408204, "loss": 0.12180397585034371, "time": 0.5461867809295654, "epoch": 6, "memory": 6318, "step": 22785} +{"lr": 4.363722756668351e-05, "data_time": 0.0009311199188232422, "loss": 0.13818999454379083, "time": 0.5394057035446167, "epoch": 6, "memory": 6318, "step": 22885} +{"lr": 4.329014428447611e-05, "data_time": 0.0008503198623657227, "loss": 0.1434588126838207, "time": 0.5260370969772339, "epoch": 6, "memory": 6318, "step": 22985} +{"lr": 4.294380662331e-05, "data_time": 0.0008500099182128906, "loss": 0.13585112169384955, "time": 0.5178012132644654, "epoch": 6, "memory": 6318, "step": 23085} +{"lr": 4.259823663611678e-05, "data_time": 0.0009212255477905274, "loss": 0.13363206684589385, "time": 0.5816402912139893, "epoch": 6, "memory": 6318, "step": 23185} +{"lr": 4.22534563269468e-05, "data_time": 0.0009413003921508789, "loss": 0.12947214618325234, "time": 0.5552320003509521, "epoch": 6, "memory": 6318, "step": 23285} +{"lr": 4.1909487649567694e-05, "data_time": 0.0008799076080322265, "loss": 0.1377067133784294, "time": 0.5511159181594849, "epoch": 6, "memory": 6318, "step": 23385} +{"lr": 4.156635250606698e-05, "data_time": 0.0009684085845947266, "loss": 0.14151004552841187, "time": 0.5126610040664673, "epoch": 6, "memory": 6318, "step": 23485} +{"lr": 4.122407274545703e-05, "data_time": 0.00086212158203125, "loss": 0.12892805486917497, "time": 0.5429720878601074, "epoch": 6, "memory": 6318, "step": 23585} +{"accuracy/top1": 94.41912078857422, "data_time": 0.4195119882452077, "time": 0.5064442938771742, "step": 6} +{"lr": 4.075657788156285e-05, "data_time": 0.0111541748046875, "loss": 0.12921048030257226, "time": 0.5187873840332031, "epoch": 7, "memory": 6318, "step": 23722} +{"lr": 4.0416412311741566e-05, "data_time": 0.0008292913436889649, "loss": 0.13311309665441512, "time": 0.5393574237823486, "epoch": 7, "memory": 6318, "step": 23822} +{"lr": 4.007717534685821e-05, "data_time": 0.0008679866790771484, "loss": 0.1257340505719185, "time": 0.5401736259460449, "epoch": 7, "memory": 6318, "step": 23922} +{"lr": 3.973888858770982e-05, "data_time": 0.0009850502014160157, "loss": 0.13229683339595794, "time": 0.5292590856552124, "epoch": 7, "memory": 6318, "step": 24022} +{"lr": 3.9401573574589535e-05, "data_time": 0.0010236740112304688, "loss": 0.13488859087228774, "time": 0.5393239259719849, "epoch": 7, "memory": 6318, "step": 24122} +{"lr": 3.906525178591511e-05, "data_time": 0.0008519411087036133, "loss": 0.13168733865022658, "time": 0.5322802066802979, "epoch": 7, "memory": 6318, "step": 24222} +{"lr": 3.87299446368609e-05, "data_time": 0.0008388996124267578, "loss": 0.13895001113414765, "time": 0.5517948865890503, "epoch": 7, "memory": 6318, "step": 24322} +{"lr": 3.839567347799442e-05, "data_time": 0.0008920669555664062, "loss": 0.1328663319349289, "time": 0.5596678018569946, "epoch": 7, "memory": 6318, "step": 24422} +{"lr": 3.806245959391694e-05, "data_time": 0.0008179903030395508, "loss": 0.13098523691296576, "time": 0.5361457586288452, "epoch": 7, "memory": 6318, "step": 24522} +{"lr": 3.773032420190816e-05, "data_time": 0.0008769989013671875, "loss": 0.1411207750439644, "time": 0.5402635097503662, "epoch": 7, "memory": 6318, "step": 24622} +{"lr": 3.739928845057507e-05, "data_time": 0.0009146928787231445, "loss": 0.13031796142458915, "time": 0.577673864364624, "epoch": 7, "memory": 6318, "step": 24722} +{"lr": 3.706937341850541e-05, "data_time": 0.0011984825134277344, "loss": 0.14541516155004502, "time": 0.567496943473816, "epoch": 7, "memory": 6318, "step": 24822} +{"lr": 3.674060011292556e-05, "data_time": 0.0008799314498901367, "loss": 0.12934816256165504, "time": 0.5143082141876221, "epoch": 7, "memory": 6318, "step": 24922} +{"lr": 3.641298946836287e-05, "data_time": 0.0009241819381713868, "loss": 0.14347289130091667, "time": 0.5372864007949829, "epoch": 7, "memory": 6318, "step": 25022} +{"lr": 3.6086562345312595e-05, "data_time": 0.0008380651473999023, "loss": 0.1324164941906929, "time": 0.49466605186462403, "epoch": 7, "memory": 6318, "step": 25122} +{"lr": 3.576133952890974e-05, "data_time": 0.0009708881378173828, "loss": 0.13816252574324608, "time": 0.5268621921539307, "epoch": 7, "memory": 6318, "step": 25222} +{"lr": 3.543734172760541e-05, "data_time": 0.0009027242660522461, "loss": 0.15072961449623107, "time": 0.4827312469482422, "epoch": 7, "memory": 6318, "step": 25322} +{"lr": 3.511458957184831e-05, "data_time": 0.0009831905364990235, "loss": 0.13165954500436783, "time": 0.5381730318069458, "epoch": 7, "memory": 6318, "step": 25422} +{"lr": 3.479310361277118e-05, "data_time": 0.0008530378341674805, "loss": 0.12745990008115768, "time": 0.5493924856185913, "epoch": 7, "memory": 6318, "step": 25522} +{"lr": 3.4472904320882065e-05, "data_time": 0.0008347034454345703, "loss": 0.1353730984032154, "time": 0.5355998754501343, "epoch": 7, "memory": 6318, "step": 25622} +{"lr": 3.4154012084760864e-05, "data_time": 0.000901484489440918, "loss": 0.13354340568184853, "time": 0.49695773124694825, "epoch": 7, "memory": 6318, "step": 25722} +{"lr": 3.38364472097612e-05, "data_time": 0.000924372673034668, "loss": 0.14177048951387405, "time": 0.5665776014328003, "epoch": 7, "memory": 6318, "step": 25822} +{"lr": 3.3520229916717485e-05, "data_time": 0.0011937856674194337, "loss": 0.14481553956866264, "time": 0.5834773540496826, "epoch": 7, "memory": 6318, "step": 25922} +{"lr": 3.320538034065722e-05, "data_time": 0.000817108154296875, "loss": 0.14810217991471292, "time": 0.5073909044265748, "epoch": 7, "memory": 6318, "step": 26022} +{"lr": 3.289191852951901e-05, "data_time": 0.0008272647857666016, "loss": 0.125517837703228, "time": 0.5073810815811157, "epoch": 7, "memory": 6318, "step": 26122} +{"lr": 3.257986444287605e-05, "data_time": 0.0009132862091064453, "loss": 0.11254951730370522, "time": 0.5433212995529175, "epoch": 7, "memory": 6318, "step": 26222} +{"lr": 3.226923795066519e-05, "data_time": 0.0008596181869506836, "loss": 0.14573151916265487, "time": 0.5444326877593995, "epoch": 7, "memory": 6318, "step": 26322} +{"lr": 3.196005883192158e-05, "data_time": 0.0008810043334960937, "loss": 0.12877408564090728, "time": 0.48203999996185304, "epoch": 7, "memory": 6318, "step": 26422} +{"lr": 3.165234677351947e-05, "data_time": 0.0009072542190551758, "loss": 0.1379546657204628, "time": 0.5224812746047973, "epoch": 7, "memory": 6318, "step": 26522} +{"lr": 3.134612136891845e-05, "data_time": 0.0009146451950073243, "loss": 0.11773192062973976, "time": 0.5135346412658691, "epoch": 7, "memory": 6318, "step": 26622} +{"lr": 3.104140211691602e-05, "data_time": 0.0008940696716308594, "loss": 0.13146670460700988, "time": 0.5220028162002563, "epoch": 7, "memory": 6318, "step": 26722} +{"lr": 3.073820842040592e-05, "data_time": 0.0008766651153564453, "loss": 0.14702612683176994, "time": 0.5504681587219238, "epoch": 7, "memory": 6318, "step": 26822} +{"lr": 3.043655958514262e-05, "data_time": 0.0008579254150390625, "loss": 0.1253778614103794, "time": 0.4890860080718994, "epoch": 7, "memory": 6318, "step": 26922} +{"lr": 3.013647481851217e-05, "data_time": 0.0008542299270629883, "loss": 0.13747862502932548, "time": 0.49033732414245607, "epoch": 7, "memory": 6318, "step": 27022} +{"lr": 2.983797322830899e-05, "data_time": 0.0008123874664306641, "loss": 0.1490783706307411, "time": 0.5488513946533203, "epoch": 7, "memory": 6318, "step": 27122} +{"lr": 2.9541073821519346e-05, "data_time": 0.0008112668991088867, "loss": 0.1284991666674614, "time": 0.4884809494018555, "epoch": 7, "memory": 6318, "step": 27222} +{"lr": 2.924579550311095e-05, "data_time": 0.0008758544921875, "loss": 0.1362491451203823, "time": 0.5032778024673462, "epoch": 7, "memory": 6318, "step": 27322} +{"lr": 2.8952157074829418e-05, "data_time": 0.0009056806564331054, "loss": 0.1216659426689148, "time": 0.5200213670730591, "epoch": 7, "memory": 6318, "step": 27422} +{"lr": 2.866017723400071e-05, "data_time": 0.000866103172302246, "loss": 0.12323641031980515, "time": 0.5441327571868897, "epoch": 7, "memory": 6318, "step": 27522} +{"accuracy/top1": 94.56525421142578, "data_time": 0.41920522163654195, "time": 0.5060606454980785, "step": 7} +{"lr": 2.8262891373132927e-05, "data_time": 0.14242446422576904, "loss": 0.1266934886574745, "time": 0.5188592672348022, "epoch": 8, "memory": 6318, "step": 27659} +{"lr": 2.797491643470029e-05, "data_time": 0.09833438396453857, "loss": 0.13300009965896606, "time": 0.579714035987854, "epoch": 8, "memory": 6318, "step": 27759} +{"lr": 2.768866230918266e-05, "data_time": 0.0008493423461914063, "loss": 0.13277756944298744, "time": 0.5567246675491333, "epoch": 8, "memory": 6318, "step": 27859} +{"lr": 2.7404147223713185e-05, "data_time": 0.0009485721588134765, "loss": 0.11810171157121659, "time": 0.4748399257659912, "epoch": 8, "memory": 6318, "step": 27959} +{"lr": 2.712138929469222e-05, "data_time": 0.0009093046188354492, "loss": 0.13050267472863197, "time": 0.5220201253890991, "epoch": 8, "memory": 6318, "step": 28059} +{"lr": 2.6840406526633896e-05, "data_time": 0.000984954833984375, "loss": 0.148331019282341, "time": 0.5292039632797241, "epoch": 8, "memory": 6318, "step": 28159} +{"lr": 2.6561216811019345e-05, "data_time": 0.000967097282409668, "loss": 0.12500903531908988, "time": 0.509897780418396, "epoch": 8, "memory": 6318, "step": 28259} +{"lr": 2.6283837925157908e-05, "data_time": 0.0008904695510864258, "loss": 0.14851940795779228, "time": 0.5096237421035766, "epoch": 8, "memory": 6318, "step": 28359} +{"lr": 2.6008287531054805e-05, "data_time": 0.0008517742156982422, "loss": 0.12435762360692024, "time": 0.5155454397201538, "epoch": 8, "memory": 6318, "step": 28459} +{"lr": 2.5734583174286796e-05, "data_time": 0.0008705377578735352, "loss": 0.12857967764139175, "time": 0.5149909257888794, "epoch": 8, "memory": 6318, "step": 28559} +{"lr": 2.546274228288488e-05, "data_time": 0.0008172988891601562, "loss": 0.1420954018831253, "time": 0.603237509727478, "epoch": 8, "memory": 6318, "step": 28659} +{"lr": 2.5192782166224308e-05, "data_time": 0.0008053779602050781, "loss": 0.1340021774172783, "time": 0.5605469465255737, "epoch": 8, "memory": 6318, "step": 28759} +{"lr": 2.492472001392276e-05, "data_time": 0.000878143310546875, "loss": 0.13753439858555794, "time": 0.5194282293319702, "epoch": 8, "memory": 6318, "step": 28859} +{"lr": 2.4658572894745857e-05, "data_time": 0.0012406587600708007, "loss": 0.14896167889237405, "time": 0.5091075897216797, "epoch": 8, "memory": 6318, "step": 28959} +{"lr": 2.439435775551997e-05, "data_time": 0.0009567975997924804, "loss": 0.13417726755142212, "time": 0.5405267000198364, "epoch": 8, "memory": 6318, "step": 29059} +{"lr": 2.4132091420053357e-05, "data_time": 0.0009279251098632812, "loss": 0.13520755246281624, "time": 0.5230050325393677, "epoch": 8, "memory": 6318, "step": 29159} +{"lr": 2.387179058806494e-05, "data_time": 0.0009148120880126953, "loss": 0.1209425762295723, "time": 0.5530800580978393, "epoch": 8, "memory": 6318, "step": 29259} +{"lr": 2.361347183412073e-05, "data_time": 0.0009204626083374023, "loss": 0.13417719900608063, "time": 0.543345856666565, "epoch": 8, "memory": 6318, "step": 29359} +{"lr": 2.3357151606578728e-05, "data_time": 0.0008663415908813476, "loss": 0.13448622971773147, "time": 0.5656534433364868, "epoch": 8, "memory": 6318, "step": 29459} +{"lr": 2.3102846226541355e-05, "data_time": 0.0009607553482055664, "loss": 0.13304293677210807, "time": 0.5370008945465088, "epoch": 8, "memory": 6318, "step": 29559} +{"lr": 2.2850571886816365e-05, "data_time": 0.001056528091430664, "loss": 0.15710695087909698, "time": 0.5332360029220581, "epoch": 8, "memory": 6318, "step": 29659} +{"lr": 2.260034465088568e-05, "data_time": 0.000875401496887207, "loss": 0.12475125044584275, "time": 0.5693212985992432, "epoch": 8, "memory": 6318, "step": 29759} +{"lr": 2.2352180451882643e-05, "data_time": 0.0008660316467285156, "loss": 0.12586882412433625, "time": 0.5214612483978271, "epoch": 8, "memory": 6318, "step": 29859} +{"lr": 2.210609509157741e-05, "data_time": 0.0008114814758300781, "loss": 0.13693549633026122, "time": 0.49515542984008787, "epoch": 8, "memory": 6318, "step": 29959} +{"lr": 2.1862104239370772e-05, "data_time": 0.0008178472518920899, "loss": 0.12876120433211327, "time": 0.548601508140564, "epoch": 8, "memory": 6318, "step": 30059} +{"lr": 2.1620223431296407e-05, "data_time": 0.0008268117904663086, "loss": 0.12953976020216942, "time": 0.5336732864379883, "epoch": 8, "memory": 6318, "step": 30159} +{"lr": 2.1380468069031744e-05, "data_time": 0.0008353233337402344, "loss": 0.11989609152078629, "time": 0.5489434003829956, "epoch": 8, "memory": 6318, "step": 30259} +{"lr": 2.1142853418917185e-05, "data_time": 0.0008357048034667968, "loss": 0.131401364505291, "time": 0.4954005479812622, "epoch": 8, "memory": 6318, "step": 30359} +{"lr": 2.0907394610983892e-05, "data_time": 0.0009056329727172852, "loss": 0.10983985811471939, "time": 0.5195320129394532, "epoch": 8, "memory": 6318, "step": 30459} +{"lr": 2.0674106637990643e-05, "data_time": 0.000891733169555664, "loss": 0.12375066280364991, "time": 0.5492771863937378, "epoch": 8, "memory": 6318, "step": 30559} +{"lr": 2.0443004354468975e-05, "data_time": 0.000986194610595703, "loss": 0.13219113796949386, "time": 0.5381386518478394, "epoch": 8, "memory": 6318, "step": 30659} +{"lr": 2.0214102475777453e-05, "data_time": 0.0009032011032104493, "loss": 0.12914354279637336, "time": 0.5168377161026001, "epoch": 8, "memory": 6318, "step": 30759} +{"lr": 1.9987415577164583e-05, "data_time": 0.0010039567947387694, "loss": 0.14365241378545762, "time": 0.539490008354187, "epoch": 8, "memory": 6318, "step": 30859} +{"lr": 1.9762958092840814e-05, "data_time": 0.0008420467376708985, "loss": 0.132231717556715, "time": 0.5407920598983764, "epoch": 8, "memory": 6318, "step": 30959} +{"lr": 1.9540744315059334e-05, "data_time": 0.000885176658630371, "loss": 0.1139188252389431, "time": 0.533031678199768, "epoch": 8, "memory": 6318, "step": 31059} +{"lr": 1.9320788393206132e-05, "data_time": 0.0008019447326660156, "loss": 0.14223519787192346, "time": 0.553769302368164, "epoch": 8, "memory": 6318, "step": 31159} +{"lr": 1.9103104332899027e-05, "data_time": 0.0008715629577636719, "loss": 0.1313345156610012, "time": 0.5306265115737915, "epoch": 8, "memory": 6318, "step": 31259} +{"lr": 1.888770599509573e-05, "data_time": 0.000914311408996582, "loss": 0.11755105927586555, "time": 0.5092584609985351, "epoch": 8, "memory": 6318, "step": 31359} +{"lr": 1.8674607095211516e-05, "data_time": 0.0012692451477050782, "loss": 0.13470162004232406, "time": 0.5235853672027588, "epoch": 8, "memory": 6318, "step": 31459} +{"accuracy/top1": 94.88188171386719, "data_time": 0.42409575396570665, "time": 0.5100471644565977, "step": 8} +{"lr": 1.838641934548488e-05, "data_time": 0.20252509117126466, "loss": 0.13036711141467094, "time": 0.5421467781066894, "epoch": 9, "memory": 6318, "step": 31596} +{"lr": 1.817882403185378e-05, "data_time": 0.0011459112167358399, "loss": 0.14821603745222092, "time": 0.5204514741897583, "epoch": 9, "memory": 6318, "step": 31696} +{"lr": 1.797357329395729e-05, "data_time": 0.0007365226745605468, "loss": 0.14046533778309822, "time": 0.5716930627822876, "epoch": 9, "memory": 6318, "step": 31796} +{"lr": 1.7770680201065646e-05, "data_time": 0.0007657051086425781, "loss": 0.1272752359509468, "time": 0.6035057783126831, "epoch": 9, "memory": 6318, "step": 31896} +{"lr": 1.7570157672326885e-05, "data_time": 0.0007543802261352539, "loss": 0.13307861760258674, "time": 0.5485161542892456, "epoch": 9, "memory": 6318, "step": 31996} +{"lr": 1.7372018475944145e-05, "data_time": 0.0008781671524047852, "loss": 0.12959629371762277, "time": 0.6123078107833863, "epoch": 9, "memory": 6318, "step": 32096} +{"lr": 1.717627522836271e-05, "data_time": 0.0007833957672119141, "loss": 0.1255411498248577, "time": 1.0483634710311889, "epoch": 9, "memory": 6318, "step": 32196} +{"lr": 1.6982940393466685e-05, "data_time": 0.0007716655731201172, "loss": 0.13023898154497146, "time": 0.5581950426101685, "epoch": 9, "memory": 6318, "step": 32296} +{"lr": 1.6792026281785218e-05, "data_time": 0.0008478164672851562, "loss": 0.12103265821933747, "time": 0.5452906370162964, "epoch": 9, "memory": 6318, "step": 32396} +{"lr": 1.660354504970888e-05, "data_time": 0.0007548332214355469, "loss": 0.1338508330285549, "time": 0.5353586435317993, "epoch": 9, "memory": 6318, "step": 32496} +{"lr": 1.641750869871534e-05, "data_time": 0.0008444547653198242, "loss": 0.13075316399335862, "time": 0.5070287704467773, "epoch": 9, "memory": 6318, "step": 32596} +{"lr": 1.6233929074605374e-05, "data_time": 0.0007917881011962891, "loss": 0.12046018317341804, "time": 0.5147933959960938, "epoch": 9, "memory": 6318, "step": 32696} +{"lr": 1.6052817866748513e-05, "data_time": 0.0008064746856689453, "loss": 0.13475532457232475, "time": 0.5322628974914551, "epoch": 9, "memory": 6318, "step": 32796} +{"lr": 1.587418660733871e-05, "data_time": 0.0007519721984863281, "loss": 0.12381435409188271, "time": 0.5126032352447509, "epoch": 9, "memory": 6318, "step": 32896} +{"lr": 1.5698046670660044e-05, "data_time": 0.0008713722229003907, "loss": 0.13767931684851648, "time": 0.5360569953918457, "epoch": 9, "memory": 6318, "step": 32996} +{"lr": 1.5524409272362524e-05, "data_time": 0.0007413148880004883, "loss": 0.13964034393429756, "time": 0.5586385250091552, "epoch": 9, "memory": 6318, "step": 33096} +{"lr": 1.5353285468747853e-05, "data_time": 0.0007857799530029297, "loss": 0.11721767336130143, "time": 0.5244704246520996, "epoch": 9, "memory": 6318, "step": 33196} +{"lr": 1.5184686156065442e-05, "data_time": 0.0008677482604980469, "loss": 0.12741190642118455, "time": 0.5342095613479614, "epoch": 9, "memory": 6318, "step": 33296} +{"lr": 1.5018622069818636e-05, "data_time": 0.0007383108139038086, "loss": 0.13146117106080055, "time": 0.5741929531097412, "epoch": 9, "memory": 6318, "step": 33396} +{"lr": 1.4855103784081099e-05, "data_time": 0.0009666204452514649, "loss": 0.14917017370462418, "time": 0.5846997261047363, "epoch": 9, "memory": 6318, "step": 33496} +{"lr": 1.4694141710823508e-05, "data_time": 0.0009002208709716797, "loss": 0.137880901992321, "time": 0.5479081630706787, "epoch": 9, "memory": 6318, "step": 33596} +{"lr": 1.4535746099250614e-05, "data_time": 0.000886225700378418, "loss": 0.12109714597463608, "time": 0.562928295135498, "epoch": 9, "memory": 6318, "step": 33696} +{"lr": 1.4379927035148576e-05, "data_time": 0.0008934974670410156, "loss": 0.11005139574408532, "time": 0.5172569274902343, "epoch": 9, "memory": 6318, "step": 33796} +{"lr": 1.42266944402428e-05, "data_time": 0.0008431434631347656, "loss": 0.12268985509872436, "time": 0.542741322517395, "epoch": 9, "memory": 6318, "step": 33896} +{"lr": 1.4076058071566133e-05, "data_time": 0.0008301496505737305, "loss": 0.13168108239769935, "time": 0.535594129562378, "epoch": 9, "memory": 6318, "step": 33996} +{"lr": 1.3928027520837635e-05, "data_time": 0.0008788585662841796, "loss": 0.12917825505137442, "time": 0.5027451276779175, "epoch": 9, "memory": 6318, "step": 34096} +{"lr": 1.3782612213851759e-05, "data_time": 0.0008229255676269531, "loss": 0.1309836857020855, "time": 0.5145463705062866, "epoch": 9, "memory": 6318, "step": 34196} +{"lr": 1.3639821409878263e-05, "data_time": 0.000844264030456543, "loss": 0.1216252364218235, "time": 0.5617667198181152, "epoch": 9, "memory": 6318, "step": 34296} +{"lr": 1.3499664201072546e-05, "data_time": 0.0008713245391845703, "loss": 0.14603504464030265, "time": 0.5469597101211547, "epoch": 9, "memory": 6318, "step": 34396} +{"lr": 1.336214951189674e-05, "data_time": 0.0008215427398681641, "loss": 0.12230904772877693, "time": 0.48522827625274656, "epoch": 9, "memory": 6318, "step": 34496} +{"lr": 1.3227286098551456e-05, "data_time": 0.0008590936660766602, "loss": 0.12949518114328384, "time": 0.5468153953552246, "epoch": 9, "memory": 6318, "step": 34596} +{"lr": 1.3095082548418224e-05, "data_time": 0.0008005857467651367, "loss": 0.12470810413360596, "time": 0.5216195106506347, "epoch": 9, "memory": 6318, "step": 34696} +{"lr": 1.2965547279512709e-05, "data_time": 0.0009238481521606446, "loss": 0.1334141582250595, "time": 0.5606493711471557, "epoch": 9, "memory": 6318, "step": 34796} +{"lr": 1.2838688539948659e-05, "data_time": 0.0009300947189331054, "loss": 0.12197697162628174, "time": 0.5481856822967529, "epoch": 9, "memory": 6318, "step": 34896} +{"lr": 1.2714514407412763e-05, "data_time": 0.0008265495300292969, "loss": 0.13125229775905609, "time": 0.5412568569183349, "epoch": 9, "memory": 6318, "step": 34996} +{"lr": 1.2593032788650256e-05, "data_time": 0.0008857488632202148, "loss": 0.12186857983469963, "time": 0.5148493766784668, "epoch": 9, "memory": 6318, "step": 35096} +{"lr": 1.2474251418961484e-05, "data_time": 0.0008896350860595703, "loss": 0.1418699562549591, "time": 0.5153765201568603, "epoch": 9, "memory": 6318, "step": 35196} +{"lr": 1.2358177861709387e-05, "data_time": 0.0011574029922485352, "loss": 0.11833802014589309, "time": 0.5051170825958252, "epoch": 9, "memory": 6318, "step": 35296} +{"lr": 1.2244819507837855e-05, "data_time": 0.0009081602096557617, "loss": 0.13394040688872338, "time": 0.5407806634902954, "epoch": 9, "memory": 6318, "step": 35396} +{"accuracy/top1": 95.19850158691406, "data_time": 0.4250262235773021, "time": 0.5117717488058682, "step": 9} +{"lr": 1.2093939694118031e-05, "data_time": 0.1658930778503418, "loss": 0.1427712269127369, "time": 0.539451289176941, "epoch": 10, "memory": 6318, "step": 35533} +{"lr": 1.1987044880988498e-05, "data_time": 0.001073932647705078, "loss": 0.1318757027387619, "time": 0.5351577043533325, "epoch": 10, "memory": 6318, "step": 35633} +{"lr": 1.1882888903004625e-05, "data_time": 0.0009023427963256836, "loss": 0.11952565461397172, "time": 0.49394822120666504, "epoch": 10, "memory": 6318, "step": 35733} +{"lr": 1.178147839226253e-05, "data_time": 0.0008791446685791016, "loss": 0.1322527289390564, "time": 0.5419137001037597, "epoch": 10, "memory": 6318, "step": 35833} +{"lr": 1.1682819806041682e-05, "data_time": 0.0008666276931762696, "loss": 0.12705398648977279, "time": 0.5449024200439453, "epoch": 10, "memory": 6318, "step": 35933} +{"lr": 1.1586919426393653e-05, "data_time": 0.0008065938949584961, "loss": 0.11582861244678497, "time": 0.5248615503311157, "epoch": 10, "memory": 6318, "step": 36033} +{"lr": 1.1493783359742196e-05, "data_time": 0.0007873058319091796, "loss": 0.1197835549712181, "time": 0.5558968544006347, "epoch": 10, "memory": 6318, "step": 36133} +{"lr": 1.1403417536494367e-05, "data_time": 0.0008743524551391602, "loss": 0.12535134106874465, "time": 0.5160489082336426, "epoch": 10, "memory": 6318, "step": 36233} +{"lr": 1.1315827710662908e-05, "data_time": 0.000906991958618164, "loss": 0.13251739665865897, "time": 0.900600814819336, "epoch": 10, "memory": 6318, "step": 36333} +{"lr": 1.1231019459499938e-05, "data_time": 0.0009277582168579102, "loss": 0.12785017788410186, "time": 0.47608993053436277, "epoch": 10, "memory": 6318, "step": 36433} +{"lr": 1.1148998183141688e-05, "data_time": 0.000840449333190918, "loss": 0.12100049778819084, "time": 0.4899609565734863, "epoch": 10, "memory": 6318, "step": 36533} +{"lr": 1.1069769104264779e-05, "data_time": 0.0008569478988647461, "loss": 0.1333050012588501, "time": 0.4979268550872803, "epoch": 10, "memory": 6318, "step": 36633} +{"lr": 1.0993337267753594e-05, "data_time": 0.0008299589157104492, "loss": 0.12846382558345795, "time": 0.5277811527252197, "epoch": 10, "memory": 6318, "step": 36733} +{"lr": 1.0919707540379088e-05, "data_time": 0.000736236572265625, "loss": 0.1328561447560787, "time": 0.5615181684494018, "epoch": 10, "memory": 6318, "step": 36833} +{"lr": 1.0848884610488858e-05, "data_time": 0.00095062255859375, "loss": 0.12305434346199036, "time": 0.5141811370849609, "epoch": 10, "memory": 6318, "step": 36933} +{"lr": 1.0780872987708645e-05, "data_time": 0.0008921623229980469, "loss": 0.1282338745892048, "time": 0.4944150447845459, "epoch": 10, "memory": 6318, "step": 37033} +{"lr": 1.0715677002655166e-05, "data_time": 0.0009976625442504883, "loss": 0.15263360738754272, "time": 0.5296923637390136, "epoch": 10, "memory": 6318, "step": 37133} +{"lr": 1.0653300806660372e-05, "data_time": 0.0009070873260498047, "loss": 0.10912934839725494, "time": 0.5134855508804321, "epoch": 10, "memory": 6318, "step": 37233} +{"lr": 1.0593748371507151e-05, "data_time": 0.0010263919830322266, "loss": 0.13477461710572242, "time": 0.5665380477905273, "epoch": 10, "memory": 6318, "step": 37333} +{"lr": 1.0537023489176325e-05, "data_time": 0.0008136749267578125, "loss": 0.14008670598268508, "time": 0.538142204284668, "epoch": 10, "memory": 6318, "step": 37433} +{"lr": 1.0483129771605328e-05, "data_time": 0.0007574558258056641, "loss": 0.1242856040596962, "time": 0.5409637689590454, "epoch": 10, "memory": 6318, "step": 37533} +{"lr": 1.0432070650458095e-05, "data_time": 0.000991082191467285, "loss": 0.14170307591557502, "time": 0.5007960319519043, "epoch": 10, "memory": 6318, "step": 37633} +{"lr": 1.0383849376906634e-05, "data_time": 0.0008839607238769531, "loss": 0.11963862031698227, "time": 0.525249719619751, "epoch": 10, "memory": 6318, "step": 37733} +{"lr": 1.0338469021423977e-05, "data_time": 0.000841832160949707, "loss": 0.13267244696617125, "time": 0.5253978252410889, "epoch": 10, "memory": 6318, "step": 37833} +{"lr": 1.029593247358868e-05, "data_time": 0.0009142875671386719, "loss": 0.11969713270664215, "time": 0.5296550035476685, "epoch": 10, "memory": 6318, "step": 37933} +{"lr": 1.0256242441900827e-05, "data_time": 0.0008339643478393555, "loss": 0.13607119396328926, "time": 0.5382554054260253, "epoch": 10, "memory": 6318, "step": 38033} +{"lr": 1.0219401453609533e-05, "data_time": 0.000739884376525879, "loss": 0.13219364285469054, "time": 0.5563067197799683, "epoch": 10, "memory": 6318, "step": 38133} +{"lr": 1.018541185455209e-05, "data_time": 0.0008130550384521484, "loss": 0.1302885577082634, "time": 0.5263590097427369, "epoch": 10, "memory": 6318, "step": 38233} +{"lr": 1.015427580900455e-05, "data_time": 0.0007379531860351562, "loss": 0.12939881533384323, "time": 0.5643665075302124, "epoch": 10, "memory": 6318, "step": 38333} +{"lr": 1.0125995299543914e-05, "data_time": 0.0007515192031860351, "loss": 0.12289058044552803, "time": 0.5224899768829345, "epoch": 10, "memory": 6318, "step": 38433} +{"lr": 1.0100572126921902e-05, "data_time": 0.0007193803787231446, "loss": 0.12374185994267464, "time": 0.5481310606002807, "epoch": 10, "memory": 6318, "step": 38533} +{"lr": 1.007800790995031e-05, "data_time": 0.0009577274322509766, "loss": 0.11785700395703316, "time": 0.5011749267578125, "epoch": 10, "memory": 6318, "step": 38633} +{"lr": 1.0058304085397905e-05, "data_time": 0.0008551836013793946, "loss": 0.12400827035307885, "time": 0.5307268381118775, "epoch": 10, "memory": 6318, "step": 38733} +{"lr": 1.0041461907898941e-05, "data_time": 0.0010039091110229492, "loss": 0.12385998293757439, "time": 0.846207070350647, "epoch": 10, "memory": 6318, "step": 38833} +{"lr": 1.0027482449873297e-05, "data_time": 0.0008578777313232421, "loss": 0.1244684875011444, "time": 0.5163285493850708, "epoch": 10, "memory": 6318, "step": 38933} +{"lr": 1.0016366601458162e-05, "data_time": 0.0009111881256103516, "loss": 0.13576598688960076, "time": 0.5722871065139771, "epoch": 10, "memory": 6318, "step": 39033} +{"lr": 1.0008115070451358e-05, "data_time": 0.0007571220397949219, "loss": 0.13201383501291275, "time": 0.5714087009429931, "epoch": 10, "memory": 6318, "step": 39133} +{"lr": 1.0002728382266284e-05, "data_time": 0.0007715702056884765, "loss": 0.13470007590949534, "time": 0.5762731313705445, "epoch": 10, "memory": 6318, "step": 39233} +{"lr": 1.0000206879898465e-05, "data_time": 0.0007406234741210938, "loss": 0.15056282579898833, "time": 0.5147181272506713, "epoch": 10, "memory": 6318, "step": 39333} +{"accuracy/top1": 95.4733657836914, "data_time": 0.4220685424475834, "time": 0.5078481312455803, "step": 10} diff --git a/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1/20230606_051509/vis_data/vis_image/0000000.png_1.png b/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1/20230606_051509/vis_data/vis_image/0000000.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..e4922def59c604163c2cd36ed30fa0fa1365fac6 Binary files /dev/null and b/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1/20230606_051509/vis_data/vis_image/0000000.png_1.png differ diff --git a/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1/20230606_051509/vis_data/vis_image/0000000.png_10.png b/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1/20230606_051509/vis_data/vis_image/0000000.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..b6e22c99bc5bc58381bf612a7a83e1aa1cbf12fd Binary files /dev/null and b/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1/20230606_051509/vis_data/vis_image/0000000.png_10.png differ diff --git a/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1/20230606_051509/vis_data/vis_image/0000000.png_2.png b/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1/20230606_051509/vis_data/vis_image/0000000.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..a6a97ced21eabc96131bbc5d9cdbcd05c549278c Binary files /dev/null and b/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1/20230606_051509/vis_data/vis_image/0000000.png_2.png differ diff --git a/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1/20230606_051509/vis_data/vis_image/0000000.png_3.png b/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1/20230606_051509/vis_data/vis_image/0000000.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..0a9e7046f4a3a29246eeecea59aa2741d4b39980 Binary files /dev/null and b/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1/20230606_051509/vis_data/vis_image/0000000.png_3.png differ diff --git a/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1/20230606_051509/vis_data/vis_image/0000000.png_4.png b/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1/20230606_051509/vis_data/vis_image/0000000.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..70e3de7d84560d2c651a156db79d405f74db5e3a Binary files /dev/null and b/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1/20230606_051509/vis_data/vis_image/0000000.png_4.png differ diff --git a/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1/20230606_051509/vis_data/vis_image/0000000.png_5.png b/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1/20230606_051509/vis_data/vis_image/0000000.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..434bbb2016a547be981a0f1b7b7efc9d402ff762 Binary files /dev/null and b/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1/20230606_051509/vis_data/vis_image/0000000.png_5.png differ diff --git a/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1/20230606_051509/vis_data/vis_image/0000000.png_6.png b/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1/20230606_051509/vis_data/vis_image/0000000.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..d268e992dfaa9cef7a9ec30bb43cfb81dd4eed55 Binary files /dev/null and b/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1/20230606_051509/vis_data/vis_image/0000000.png_6.png differ diff --git a/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1/20230606_051509/vis_data/vis_image/0000000.png_7.png b/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1/20230606_051509/vis_data/vis_image/0000000.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..1073c1246969b8aae6b30595e2242e3b5f8795bd Binary files /dev/null and b/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1/20230606_051509/vis_data/vis_image/0000000.png_7.png differ diff --git a/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1/20230606_051509/vis_data/vis_image/0000000.png_8.png b/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1/20230606_051509/vis_data/vis_image/0000000.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..3eba0c72de5cea577f416378cae4f642420946d7 Binary files /dev/null and b/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1/20230606_051509/vis_data/vis_image/0000000.png_8.png differ diff --git a/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1/20230606_051509/vis_data/vis_image/0000000.png_9.png b/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1/20230606_051509/vis_data/vis_image/0000000.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..11bfaa1b53f7789e15481e6076d8e5f84bf23028 Binary files /dev/null and b/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1/20230606_051509/vis_data/vis_image/0000000.png_9.png differ diff --git a/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1/20230606_051509/vis_data/vis_image/000005174.jpg_1.png b/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1/20230606_051509/vis_data/vis_image/000005174.jpg_1.png new file mode 100644 index 0000000000000000000000000000000000000000..6e63add25231bd8f39850dd73f4b23f2c3530a0a Binary files /dev/null and b/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1/20230606_051509/vis_data/vis_image/000005174.jpg_1.png differ diff --git a/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1/20230606_051509/vis_data/vis_image/000005174.jpg_10.png b/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1/20230606_051509/vis_data/vis_image/000005174.jpg_10.png new file mode 100644 index 0000000000000000000000000000000000000000..55fa296fbf2af7d07c6a0432d5654ddb789c83d0 Binary files /dev/null and b/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1/20230606_051509/vis_data/vis_image/000005174.jpg_10.png differ diff --git a/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1/20230606_051509/vis_data/vis_image/000005174.jpg_2.png b/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1/20230606_051509/vis_data/vis_image/000005174.jpg_2.png new file mode 100644 index 0000000000000000000000000000000000000000..eabb0b12f4d243f3c9d61f790d875bec2e7e2a4e Binary files /dev/null and b/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1/20230606_051509/vis_data/vis_image/000005174.jpg_2.png differ diff --git a/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1/20230606_051509/vis_data/vis_image/000005174.jpg_3.png b/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1/20230606_051509/vis_data/vis_image/000005174.jpg_3.png new file mode 100644 index 0000000000000000000000000000000000000000..035e024ab0171bc6bd2ea98468e3fdc72066bfe9 Binary files /dev/null and b/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1/20230606_051509/vis_data/vis_image/000005174.jpg_3.png differ diff --git a/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1/20230606_051509/vis_data/vis_image/000005174.jpg_4.png b/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1/20230606_051509/vis_data/vis_image/000005174.jpg_4.png new file mode 100644 index 0000000000000000000000000000000000000000..5ac313170fcf1618c8d9f7f45af23bbf8b9850c3 Binary files /dev/null and b/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1/20230606_051509/vis_data/vis_image/000005174.jpg_4.png differ diff --git a/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1/20230606_051509/vis_data/vis_image/000005174.jpg_5.png b/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1/20230606_051509/vis_data/vis_image/000005174.jpg_5.png new file mode 100644 index 0000000000000000000000000000000000000000..5443688db72371bd6f43c35039579d653914c16a Binary files /dev/null and b/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1/20230606_051509/vis_data/vis_image/000005174.jpg_5.png differ diff --git a/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1/20230606_051509/vis_data/vis_image/000005174.jpg_6.png b/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1/20230606_051509/vis_data/vis_image/000005174.jpg_6.png new file mode 100644 index 0000000000000000000000000000000000000000..917396c6c3a8a66ca0b6a517f4f1e2d36b0b91b5 Binary files /dev/null and b/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1/20230606_051509/vis_data/vis_image/000005174.jpg_6.png differ diff --git a/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1/20230606_051509/vis_data/vis_image/000005174.jpg_7.png b/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1/20230606_051509/vis_data/vis_image/000005174.jpg_7.png new file mode 100644 index 0000000000000000000000000000000000000000..b4dc5f98e8e599c5916faecae447289d31457084 Binary files /dev/null and b/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1/20230606_051509/vis_data/vis_image/000005174.jpg_7.png differ diff --git a/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1/20230606_051509/vis_data/vis_image/000005174.jpg_8.png b/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1/20230606_051509/vis_data/vis_image/000005174.jpg_8.png new file mode 100644 index 0000000000000000000000000000000000000000..5e6d1ff0108391e6e322d71888de0ebc3f1e2457 Binary files /dev/null and b/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1/20230606_051509/vis_data/vis_image/000005174.jpg_8.png differ diff --git a/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1/20230606_051509/vis_data/vis_image/000005174.jpg_9.png b/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1/20230606_051509/vis_data/vis_image/000005174.jpg_9.png new file mode 100644 index 0000000000000000000000000000000000000000..864120da11a5c16e17d8a02292afa8a9533fc810 Binary files /dev/null and b/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1/20230606_051509/vis_data/vis_image/000005174.jpg_9.png differ diff --git a/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1/20230606_051509/vis_data/vis_image/0010000.png_1.png b/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1/20230606_051509/vis_data/vis_image/0010000.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..c09721536689bbc2e3f6749202f57ad770d88c72 Binary files /dev/null and b/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1/20230606_051509/vis_data/vis_image/0010000.png_1.png differ diff --git a/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1/20230606_051509/vis_data/vis_image/0010000.png_10.png b/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1/20230606_051509/vis_data/vis_image/0010000.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..07481dc4ac59d2f0f2cc9c365a29c69578685670 Binary files /dev/null and b/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1/20230606_051509/vis_data/vis_image/0010000.png_10.png differ diff --git a/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1/20230606_051509/vis_data/vis_image/0010000.png_2.png b/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1/20230606_051509/vis_data/vis_image/0010000.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..a4335c00c6e894e268ae8c6a731fc763d2a924c0 Binary files /dev/null and b/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1/20230606_051509/vis_data/vis_image/0010000.png_2.png differ diff --git a/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1/20230606_051509/vis_data/vis_image/0010000.png_3.png b/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1/20230606_051509/vis_data/vis_image/0010000.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..f1fcd9431476fb0856e00411ac2568d474e9a397 Binary files /dev/null and b/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1/20230606_051509/vis_data/vis_image/0010000.png_3.png differ diff --git a/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1/20230606_051509/vis_data/vis_image/0010000.png_4.png b/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1/20230606_051509/vis_data/vis_image/0010000.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..cd6b39bcf3e725c5b66c52de8930b76ddba000dd Binary files /dev/null and b/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1/20230606_051509/vis_data/vis_image/0010000.png_4.png differ diff --git a/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1/20230606_051509/vis_data/vis_image/0010000.png_5.png b/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1/20230606_051509/vis_data/vis_image/0010000.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..1aadf5fac97cf16dff27510cf80a17fc14ba9128 Binary files /dev/null and b/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1/20230606_051509/vis_data/vis_image/0010000.png_5.png differ diff --git a/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1/20230606_051509/vis_data/vis_image/0010000.png_6.png b/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1/20230606_051509/vis_data/vis_image/0010000.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..f1e293d37931200643ef46b229815fc1168dfacd Binary files /dev/null and b/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1/20230606_051509/vis_data/vis_image/0010000.png_6.png differ diff --git a/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1/20230606_051509/vis_data/vis_image/0010000.png_7.png b/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1/20230606_051509/vis_data/vis_image/0010000.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..a85b724d18ed1fc0bb1da82d2f1a5fbcdc2a6d78 Binary files /dev/null and b/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1/20230606_051509/vis_data/vis_image/0010000.png_7.png differ diff --git a/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1/20230606_051509/vis_data/vis_image/0010000.png_8.png b/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1/20230606_051509/vis_data/vis_image/0010000.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..34f306099c9fa820d1ee63cb2fabebc80a3e8769 Binary files /dev/null and b/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1/20230606_051509/vis_data/vis_image/0010000.png_8.png differ diff --git a/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1/20230606_051509/vis_data/vis_image/0010000.png_9.png b/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1/20230606_051509/vis_data/vis_image/0010000.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..1f72771c331c318851d14dc2628a59514dcc4b82 Binary files /dev/null and b/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1/20230606_051509/vis_data/vis_image/0010000.png_9.png differ diff --git a/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1/epoch_1.pth b/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1/epoch_1.pth new file mode 100644 index 0000000000000000000000000000000000000000..159468f7838f83ec80c5e2c091b209af70f5cdc8 --- /dev/null +++ b/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1/epoch_1.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:12dcccfaedc1691b0ddc4beb09d09725dcb21f5e7c138133de44582e15829f44 +size 277335053 diff --git a/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1/epoch_10.pth b/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1/epoch_10.pth new file mode 100644 index 0000000000000000000000000000000000000000..18d7479a126cb11d815bfe4fa4c219d118347da3 --- /dev/null +++ b/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1/epoch_10.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:d9315385337c49aa4138cecc0465cb141e8e8df3c6dee3a82d55ea4e7db245e4 +size 279924045 diff --git a/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1/epoch_2.pth b/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1/epoch_2.pth new file mode 100644 index 0000000000000000000000000000000000000000..5716356877a93f529f9ffe7facffc605152a401e --- /dev/null +++ b/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1/epoch_2.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c073dc626d048291c39aed8e4a733060fa4362bdc703ad2a3e7f5c60ce83fec5 +size 277621965 diff --git a/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1/epoch_3.pth b/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1/epoch_3.pth new file mode 100644 index 0000000000000000000000000000000000000000..d511bdfe8360c72866402f356e3bf1767c9d6760 --- /dev/null +++ b/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1/epoch_3.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:2fd3b4e9f57f4d7506bf4bec88c11493f09a01fc7b38e41337d41b270693bbdf +size 277908301 diff --git a/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1/epoch_4.pth b/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1/epoch_4.pth new file mode 100644 index 0000000000000000000000000000000000000000..e185c4c2ba2606d162d5c4908e5114bfb14ec688 --- /dev/null +++ b/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1/epoch_4.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:e662d0919fe2ef4b124708895205db45c5416398fbf011a1b2e89cbdd67ba301 +size 278194445 diff --git a/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1/epoch_5.pth b/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1/epoch_5.pth new file mode 100644 index 0000000000000000000000000000000000000000..8df0a6577e9d7e0f804b8ba359cfbcb58d5a0a8d --- /dev/null +++ b/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1/epoch_5.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:a30a501df55bafe854a9a1c430d2c8b2bb1590bd35b8e1ee9a7a13f31007a091 +size 278480781 diff --git a/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1/epoch_6.pth b/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1/epoch_6.pth new file mode 100644 index 0000000000000000000000000000000000000000..df1eaad412e58e056eb64a479797df9af68ff300 --- /dev/null +++ b/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1/epoch_6.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:bb3c485d64b7d944f2a7ed340f260d61c1766d575d260c380ffb1fa69fe098fe +size 278767117 diff --git a/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1/epoch_7.pth b/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1/epoch_7.pth new file mode 100644 index 0000000000000000000000000000000000000000..98d2d9054ff69e43025b22cb191ad16e42f375ab --- /dev/null +++ b/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1/epoch_7.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:874dc1745fe8a880c3d0e6d54529bf816110e7604450506cc2b2fc5dfe4948d6 +size 279053773 diff --git a/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1/epoch_8.pth b/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1/epoch_8.pth new file mode 100644 index 0000000000000000000000000000000000000000..c73c29652a3a53adbf21c119bad061783d5a8160 --- /dev/null +++ b/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1/epoch_8.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:a3a347fb2e269dc8fbd0ab5867884df56f57595f8a8fa001e95a6dabe05ad093 +size 279344013 diff --git a/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1/epoch_9.pth b/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1/epoch_9.pth new file mode 100644 index 0000000000000000000000000000000000000000..a5353cf9dfdc79b0d10eba6919ed00ad2eae53d4 --- /dev/null +++ b/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1/epoch_9.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:0b1d67a1691df9fc3ea88c1ae3d4d66b6f70727f6cc23ae23c297e53d6b0ecfc +size 279634253 diff --git a/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1/last_checkpoint b/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1/last_checkpoint new file mode 100644 index 0000000000000000000000000000000000000000..e9191c1543b8e9d023f1e940fb83c27d1186b9d7 --- /dev/null +++ b/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1/last_checkpoint @@ -0,0 +1 @@ +/mnt/petrelfs/luzeyu/workspace/fakebench/mmpretrain/workdir/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1/epoch_10.pth \ No newline at end of file diff --git a/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1.py b/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1.py new file mode 100644 index 0000000000000000000000000000000000000000..07f49b3c87537f52b4f76320195407e52d9fcb36 --- /dev/null +++ b/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1.py @@ -0,0 +1,247 @@ +optim_wrapper = dict( + optimizer=dict( + type='SGD', + lr=0.0001, + momentum=0.9, + weight_decay=0.0001, + _scope_='mmpretrain'), + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=False, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=512) +model = dict( + type='ImageClassifier', + backbone=dict( + frozen_stages=2, + type='ResNet', + depth=50, + num_stages=4, + out_indices=(3, ), + style='pytorch', + init_cfg=dict( + type='Pretrained', + checkpoint= + 'https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth', + prefix='backbone')), + neck=dict(type='GlobalAveragePooling'), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=2048, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + topk=1)) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/IF80w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/if-dpmsolver++-50-20w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/cc1m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/if-dpmsolver++-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = dict(type='Accuracy', topk=1) +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/if-dpmsolver++-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = dict(type='Accuracy', topk=1) +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/resnet50_2xb256_IF_1m_lr1e-4_aug_5e-1' diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/20230606_005743.log b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/20230606_005743.log new file mode 100644 index 0000000000000000000000000000000000000000..33de14ace0e0a80bc2fba097b68760984b3bf7c8 --- /dev/null +++ b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/20230606_005743.log @@ -0,0 +1,1445 @@ +2023/06/06 00:57:46 - mmengine - INFO - +------------------------------------------------------------ +System environment: + sys.platform: linux + Python: 3.10.9 (main, Mar 8 2023, 10:47:38) [GCC 11.2.0] + CUDA available: True + numpy_random_seed: 1427001271 + GPU 0,1: NVIDIA A100-SXM4-80GB + CUDA_HOME: /mnt/petrelfs/share/cuda-11.6 + NVCC: Cuda compilation tools, release 11.6, V11.6.124 + GCC: gcc (GCC) 7.5.0 + PyTorch: 1.13.1 + PyTorch compiling details: PyTorch built with: + - GCC 9.3 + - C++ Version: 201402 + - Intel(R) oneAPI Math Kernel Library Version 2021.4-Product Build 20210904 for Intel(R) 64 architecture applications + - Intel(R) MKL-DNN v2.6.0 (Git Hash 52b5f107dd9cf10910aaa19cb47f3abf9b349815) + - OpenMP 201511 (a.k.a. OpenMP 4.5) + - LAPACK is enabled (usually provided by MKL) + - NNPACK is enabled + - CPU capability usage: AVX2 + - CUDA Runtime 11.6 + - NVCC architecture flags: -gencode;arch=compute_37,code=sm_37;-gencode;arch=compute_50,code=sm_50;-gencode;arch=compute_60,code=sm_60;-gencode;arch=compute_61,code=sm_61;-gencode;arch=compute_70,code=sm_70;-gencode;arch=compute_75,code=sm_75;-gencode;arch=compute_80,code=sm_80;-gencode;arch=compute_86,code=sm_86;-gencode;arch=compute_37,code=compute_37 + - CuDNN 8.3.2 (built against CUDA 11.5) + - Magma 2.6.1 + - Build settings: BLAS_INFO=mkl, BUILD_TYPE=Release, CUDA_VERSION=11.6, CUDNN_VERSION=8.3.2, CXX_COMPILER=/opt/rh/devtoolset-9/root/usr/bin/c++, CXX_FLAGS= -fabi-version=11 -Wno-deprecated -fvisibility-inlines-hidden -DUSE_PTHREADPOOL -fopenmp -DNDEBUG -DUSE_KINETO -DUSE_FBGEMM -DUSE_QNNPACK -DUSE_PYTORCH_QNNPACK -DUSE_XNNPACK -DSYMBOLICATE_MOBILE_DEBUG_HANDLE -DEDGE_PROFILER_USE_KINETO -O2 -fPIC -Wno-narrowing -Wall -Wextra -Werror=return-type -Werror=non-virtual-dtor -Wno-missing-field-initializers -Wno-type-limits -Wno-array-bounds -Wno-unknown-pragmas -Wunused-local-typedefs -Wno-unused-parameter -Wno-unused-function -Wno-unused-result -Wno-strict-overflow -Wno-strict-aliasing -Wno-error=deprecated-declarations -Wno-stringop-overflow -Wno-psabi -Wno-error=pedantic -Wno-error=redundant-decls -Wno-error=old-style-cast -fdiagnostics-color=always -faligned-new -Wno-unused-but-set-variable -Wno-maybe-uninitialized -fno-math-errno -fno-trapping-math -Werror=format -Werror=cast-function-type -Wno-stringop-overflow, LAPACK_INFO=mkl, PERF_WITH_AVX=1, PERF_WITH_AVX2=1, PERF_WITH_AVX512=1, TORCH_VERSION=1.13.1, USE_CUDA=ON, USE_CUDNN=ON, USE_EXCEPTION_PTR=1, USE_GFLAGS=OFF, USE_GLOG=OFF, USE_MKL=ON, USE_MKLDNN=ON, USE_MPI=OFF, USE_NCCL=ON, USE_NNPACK=ON, USE_OPENMP=ON, USE_ROCM=OFF, + + TorchVision: 0.14.1 + OpenCV: 4.7.0 + MMEngine: 0.7.3 + +Runtime environment: + cudnn_benchmark: True + mp_cfg: {'mp_start_method': 'fork', 'opencv_num_threads': 0} + dist_cfg: {'backend': 'nccl'} + seed: None + deterministic: False + Distributed launcher: slurm + Distributed training: True + GPU number: 2 +------------------------------------------------------------ + +2023/06/06 00:57:50 - mmengine - INFO - Config: +optim_wrapper = dict( + optimizer=dict( + type='SGD', + lr=0.0001, + momentum=0.9, + weight_decay=0.0001, + _scope_='mmpretrain'), + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=False, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=512) +model = dict( + type='ImageClassifier', + backbone=dict( + frozen_stages=2, + type='ResNet', + depth=50, + num_stages=4, + out_indices=(3, ), + style='pytorch', + init_cfg=dict( + type='Pretrained', + checkpoint= + 'https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth', + prefix='backbone')), + neck=dict(type='GlobalAveragePooling'), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=2048, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + topk=1)) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/all_0.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/all_1.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3fake8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/cc1m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3real8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV2-1-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/if-dpmsolver++-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = dict(type='Accuracy', topk=1) +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV2-1-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/if-dpmsolver++-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = dict(type='Accuracy', topk=1) +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1' + +2023/06/06 00:58:02 - mmengine - INFO - Hooks will be executed in the following order: +before_run: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_load_checkpoint: +(ABOVE_NORMAL) EMAHook + -------------------- +before_train: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_train_epoch: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(NORMAL ) DistSamplerSeedHook + -------------------- +before_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook + -------------------- +after_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +after_train_epoch: +(NORMAL ) IterTimerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_val_epoch: +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook + -------------------- +before_val_iter: +(NORMAL ) IterTimerHook + -------------------- +after_val_iter: +(NORMAL ) IterTimerHook +(NORMAL ) VisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_val_epoch: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_save_checkpoint: +(ABOVE_NORMAL) EMAHook + -------------------- +after_train: +(VERY_LOW ) CheckpointHook + -------------------- +before_test_epoch: +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook + -------------------- +before_test_iter: +(NORMAL ) IterTimerHook + -------------------- +after_test_iter: +(NORMAL ) IterTimerHook +(NORMAL ) VisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_test_epoch: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_run: +(BELOW_NORMAL) LoggerHook + -------------------- +2023/06/06 00:58:23 - mmengine - INFO - load backbone in model from: https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth +Name of parameter - Initialization information + +backbone.conv1.weight - torch.Size([64, 3, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.bn1.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.bn1.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.conv1.weight - torch.Size([64, 64, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.bn1.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.bn1.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.conv2.weight - torch.Size([64, 64, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.bn2.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.bn2.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.conv3.weight - torch.Size([256, 64, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.bn3.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.bn3.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.downsample.0.weight - torch.Size([256, 64, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.downsample.1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.downsample.1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.conv1.weight - torch.Size([64, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.bn1.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.bn1.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.conv2.weight - torch.Size([64, 64, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.bn2.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.bn2.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.conv3.weight - torch.Size([256, 64, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.bn3.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.bn3.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.conv1.weight - torch.Size([64, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.bn1.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.bn1.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.conv2.weight - torch.Size([64, 64, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.bn2.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.bn2.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.conv3.weight - torch.Size([256, 64, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.bn3.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.bn3.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.conv1.weight - torch.Size([128, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.bn1.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.bn1.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.conv2.weight - torch.Size([128, 128, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.bn2.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.bn2.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.conv3.weight - torch.Size([512, 128, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.bn3.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.bn3.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.downsample.0.weight - torch.Size([512, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.downsample.1.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.downsample.1.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.conv1.weight - torch.Size([128, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.bn1.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.bn1.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.conv2.weight - torch.Size([128, 128, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.bn2.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.bn2.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.conv3.weight - torch.Size([512, 128, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.bn3.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.bn3.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.conv1.weight - torch.Size([128, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.bn1.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.bn1.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.conv2.weight - torch.Size([128, 128, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.bn2.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.bn2.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.conv3.weight - torch.Size([512, 128, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.bn3.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.bn3.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.conv1.weight - torch.Size([128, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.bn1.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.bn1.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.conv2.weight - torch.Size([128, 128, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.bn2.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.bn2.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.conv3.weight - torch.Size([512, 128, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.bn3.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.bn3.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.conv1.weight - torch.Size([256, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.bn1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.bn1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.conv2.weight - torch.Size([256, 256, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.bn2.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.bn2.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.conv3.weight - torch.Size([1024, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.bn3.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.bn3.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.downsample.0.weight - torch.Size([1024, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.downsample.1.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.downsample.1.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.conv1.weight - torch.Size([256, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.bn1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.bn1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.conv2.weight - torch.Size([256, 256, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.bn2.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.bn2.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.conv3.weight - torch.Size([1024, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.bn3.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.bn3.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.conv1.weight - torch.Size([256, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.bn1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.bn1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.conv2.weight - torch.Size([256, 256, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.bn2.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.bn2.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.conv3.weight - torch.Size([1024, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.bn3.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.bn3.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.conv1.weight - torch.Size([256, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.bn1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.bn1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.conv2.weight - torch.Size([256, 256, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.bn2.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.bn2.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.conv3.weight - torch.Size([1024, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.bn3.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.bn3.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.conv1.weight - torch.Size([256, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.bn1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.bn1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.conv2.weight - torch.Size([256, 256, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.bn2.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.bn2.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.conv3.weight - torch.Size([1024, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.bn3.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.bn3.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.conv1.weight - torch.Size([256, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.bn1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.bn1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.conv2.weight - torch.Size([256, 256, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.bn2.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.bn2.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.conv3.weight - torch.Size([1024, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.bn3.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.bn3.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.conv1.weight - torch.Size([512, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.bn1.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.bn1.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.conv2.weight - torch.Size([512, 512, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.bn2.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.bn2.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.conv3.weight - torch.Size([2048, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.bn3.weight - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.bn3.bias - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.downsample.0.weight - torch.Size([2048, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.downsample.1.weight - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.downsample.1.bias - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.conv1.weight - torch.Size([512, 2048, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.bn1.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.bn1.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.conv2.weight - torch.Size([512, 512, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.bn2.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.bn2.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.conv3.weight - torch.Size([2048, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.bn3.weight - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.bn3.bias - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.conv1.weight - torch.Size([512, 2048, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.bn1.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.bn1.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.conv2.weight - torch.Size([512, 512, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.bn2.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.bn2.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.conv3.weight - torch.Size([2048, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.bn3.weight - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.bn3.bias - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +head.fc.weight - torch.Size([2, 2048]): +NormalInit: mean=0, std=0.01, bias=0 + +head.fc.bias - torch.Size([2]): +NormalInit: mean=0, std=0.01, bias=0 +2023/06/06 00:58:24 - mmengine - WARNING - "FileClient" will be deprecated in future. Please use io functions in https://mmengine.readthedocs.io/en/latest/api/fileio.html#file-io +2023/06/06 00:58:24 - mmengine - WARNING - "HardDiskBackend" is the alias of "LocalBackend" and the former will be deprecated in future. +2023/06/06 00:58:24 - mmengine - INFO - Checkpoints will be saved to /mnt/petrelfs/luzeyu/workspace/fakebench/mmpretrain/workdir/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1. +2023/06/06 00:59:44 - mmengine - INFO - Epoch(train) [1][ 100/4092] lr: 9.9999e-05 eta: 9:04:18 time: 0.8090 data_time: 0.2358 memory: 9436 loss: 0.6417 +2023/06/06 01:01:02 - mmengine - INFO - Epoch(train) [1][ 200/4092] lr: 9.9995e-05 eta: 8:56:17 time: 0.7791 data_time: 0.2187 memory: 6319 loss: 0.5971 +2023/06/06 01:02:15 - mmengine - INFO - Epoch(train) [1][ 300/4092] lr: 9.9988e-05 eta: 8:43:14 time: 0.7500 data_time: 0.0009 memory: 6319 loss: 0.5584 +2023/06/06 01:03:30 - mmengine - INFO - Epoch(train) [1][ 400/4092] lr: 9.9979e-05 eta: 8:37:10 time: 0.7361 data_time: 0.0008 memory: 6319 loss: 0.5302 +2023/06/06 01:04:47 - mmengine - INFO - Epoch(train) [1][ 500/4092] lr: 9.9967e-05 eta: 8:36:24 time: 0.7582 data_time: 0.2011 memory: 6319 loss: 0.5015 +2023/06/06 01:06:01 - mmengine - INFO - Epoch(train) [1][ 600/4092] lr: 9.9952e-05 eta: 8:32:47 time: 0.7184 data_time: 0.3321 memory: 6319 loss: 0.4688 +2023/06/06 01:07:14 - mmengine - INFO - Epoch(train) [1][ 700/4092] lr: 9.9935e-05 eta: 8:28:14 time: 0.7379 data_time: 0.0201 memory: 6319 loss: 0.4382 +2023/06/06 01:08:30 - mmengine - INFO - Epoch(train) [1][ 800/4092] lr: 9.9915e-05 eta: 8:26:42 time: 0.7372 data_time: 0.0008 memory: 6319 loss: 0.4320 +2023/06/06 01:09:43 - mmengine - INFO - Epoch(train) [1][ 900/4092] lr: 9.9893e-05 eta: 8:23:25 time: 0.7632 data_time: 0.1161 memory: 6319 loss: 0.4201 +2023/06/06 01:10:55 - mmengine - INFO - Exp name: resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1_20230606_005743 +2023/06/06 01:10:55 - mmengine - INFO - Epoch(train) [1][1000/4092] lr: 9.9868e-05 eta: 8:20:02 time: 0.7913 data_time: 0.0008 memory: 6319 loss: 0.4004 +2023/06/06 01:12:09 - mmengine - INFO - Epoch(train) [1][1100/4092] lr: 9.9840e-05 eta: 8:17:54 time: 0.8244 data_time: 0.0008 memory: 6319 loss: 0.3950 +2023/06/06 01:13:23 - mmengine - INFO - Epoch(train) [1][1200/4092] lr: 9.9809e-05 eta: 8:16:10 time: 0.7227 data_time: 0.0007 memory: 6319 loss: 0.3863 +2023/06/06 01:14:34 - mmengine - INFO - Epoch(train) [1][1300/4092] lr: 9.9776e-05 eta: 8:12:44 time: 0.7242 data_time: 0.0008 memory: 6319 loss: 0.3782 +2023/06/06 01:15:45 - mmengine - INFO - Epoch(train) [1][1400/4092] lr: 9.9741e-05 eta: 8:10:11 time: 0.7568 data_time: 0.0008 memory: 6319 loss: 0.3711 +2023/06/06 01:16:57 - mmengine - INFO - Epoch(train) [1][1500/4092] lr: 9.9702e-05 eta: 8:07:33 time: 0.7378 data_time: 0.0007 memory: 6319 loss: 0.3743 +2023/06/06 01:18:09 - mmengine - INFO - Epoch(train) [1][1600/4092] lr: 9.9661e-05 eta: 8:05:26 time: 0.6979 data_time: 0.0009 memory: 6319 loss: 0.3582 +2023/06/06 01:19:21 - mmengine - INFO - Epoch(train) [1][1700/4092] lr: 9.9618e-05 eta: 8:03:33 time: 0.7231 data_time: 0.0011 memory: 6319 loss: 0.3595 +2023/06/06 01:20:35 - mmengine - INFO - Epoch(train) [1][1800/4092] lr: 9.9571e-05 eta: 8:02:08 time: 0.7151 data_time: 0.0008 memory: 6319 loss: 0.3502 +2023/06/06 01:21:48 - mmengine - INFO - Epoch(train) [1][1900/4092] lr: 9.9523e-05 eta: 8:00:33 time: 0.7786 data_time: 0.0007 memory: 6319 loss: 0.3434 +2023/06/06 01:23:01 - mmengine - INFO - Exp name: resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1_20230606_005743 +2023/06/06 01:23:01 - mmengine - INFO - Epoch(train) [1][2000/4092] lr: 9.9471e-05 eta: 7:59:10 time: 0.7474 data_time: 0.0008 memory: 6319 loss: 0.3303 +2023/06/06 01:24:14 - mmengine - INFO - Epoch(train) [1][2100/4092] lr: 9.9417e-05 eta: 7:57:46 time: 0.7527 data_time: 0.0009 memory: 6319 loss: 0.3392 +2023/06/06 01:25:26 - mmengine - INFO - Epoch(train) [1][2200/4092] lr: 9.9360e-05 eta: 7:55:56 time: 0.7442 data_time: 0.0008 memory: 6319 loss: 0.3544 +2023/06/06 01:28:12 - mmengine - INFO - Epoch(train) [1][2300/4092] lr: 9.9301e-05 eta: 8:20:31 time: 0.7476 data_time: 0.0010 memory: 6319 loss: 0.3193 +2023/06/06 01:29:23 - mmengine - INFO - Epoch(train) [1][2400/4092] lr: 9.9239e-05 eta: 8:17:24 time: 0.7450 data_time: 0.0011 memory: 6319 loss: 0.3394 +2023/06/06 01:30:35 - mmengine - INFO - Epoch(train) [1][2500/4092] lr: 9.9174e-05 eta: 8:14:44 time: 0.7152 data_time: 0.0007 memory: 6319 loss: 0.3217 +2023/06/06 01:31:46 - mmengine - INFO - Epoch(train) [1][2600/4092] lr: 9.9107e-05 eta: 8:11:45 time: 0.6926 data_time: 0.0009 memory: 6319 loss: 0.3210 +2023/06/06 01:32:58 - mmengine - INFO - Epoch(train) [1][2700/4092] lr: 9.9037e-05 eta: 8:09:19 time: 0.6957 data_time: 0.0008 memory: 6319 loss: 0.3038 +2023/06/06 01:34:10 - mmengine - INFO - Epoch(train) [1][2800/4092] lr: 9.8965e-05 eta: 8:06:57 time: 0.7176 data_time: 0.0009 memory: 6319 loss: 0.3177 +2023/06/06 01:35:22 - mmengine - INFO - Epoch(train) [1][2900/4092] lr: 9.8890e-05 eta: 8:04:44 time: 0.7144 data_time: 0.0009 memory: 6319 loss: 0.3155 +2023/06/06 01:36:35 - mmengine - INFO - Exp name: resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1_20230606_005743 +2023/06/06 01:36:35 - mmengine - INFO - Epoch(train) [1][3000/4092] lr: 9.8812e-05 eta: 8:02:41 time: 0.6949 data_time: 0.0008 memory: 6319 loss: 0.3076 +2023/06/06 01:37:46 - mmengine - INFO - Epoch(train) [1][3100/4092] lr: 9.8732e-05 eta: 8:00:18 time: 0.7028 data_time: 0.0008 memory: 6319 loss: 0.3098 +2023/06/06 01:38:56 - mmengine - INFO - Epoch(train) [1][3200/4092] lr: 9.8650e-05 eta: 7:57:53 time: 0.7453 data_time: 0.0008 memory: 6319 loss: 0.3065 +2023/06/06 01:40:06 - mmengine - INFO - Epoch(train) [1][3300/4092] lr: 9.8564e-05 eta: 7:55:31 time: 0.6711 data_time: 0.0007 memory: 6319 loss: 0.2986 +2023/06/06 01:41:16 - mmengine - INFO - Epoch(train) [1][3400/4092] lr: 9.8476e-05 eta: 7:53:09 time: 0.7275 data_time: 0.0009 memory: 6319 loss: 0.2963 +2023/06/06 01:42:28 - mmengine - INFO - Epoch(train) [1][3500/4092] lr: 9.8386e-05 eta: 7:51:08 time: 0.6771 data_time: 0.0009 memory: 6319 loss: 0.2964 +2023/06/06 01:43:42 - mmengine - INFO - Epoch(train) [1][3600/4092] lr: 9.8293e-05 eta: 7:49:44 time: 0.7794 data_time: 0.0008 memory: 6319 loss: 0.3021 +2023/06/06 01:44:56 - mmengine - INFO - Epoch(train) [1][3700/4092] lr: 9.8198e-05 eta: 7:48:06 time: 0.8267 data_time: 0.0008 memory: 6319 loss: 0.3013 +2023/06/06 01:46:10 - mmengine - INFO - Epoch(train) [1][3800/4092] lr: 9.8099e-05 eta: 7:46:39 time: 0.7225 data_time: 0.0009 memory: 6319 loss: 0.3013 +2023/06/06 01:47:25 - mmengine - INFO - Epoch(train) [1][3900/4092] lr: 9.7999e-05 eta: 7:45:20 time: 0.7070 data_time: 0.0008 memory: 6319 loss: 0.2915 +2023/06/06 01:48:37 - mmengine - INFO - Exp name: resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1_20230606_005743 +2023/06/06 01:48:37 - mmengine - INFO - Epoch(train) [1][4000/4092] lr: 9.7896e-05 eta: 7:43:37 time: 0.7585 data_time: 0.0009 memory: 6319 loss: 0.3216 +2023/06/06 01:49:46 - mmengine - INFO - Exp name: resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1_20230606_005743 +2023/06/06 01:49:46 - mmengine - INFO - Saving checkpoint at 1 epochs +2023/06/06 01:50:30 - mmengine - INFO - Epoch(val) [1][100/119] eta: 0:00:07 time: 0.6816 data_time: 0.5936 memory: 6319 +2023/06/06 01:50:56 - mmengine - INFO - Epoch(val) [1][119/119] accuracy/top1: 83.0131 data_time: 0.3644 time: 0.4524 +2023/06/06 01:52:10 - mmengine - INFO - Epoch(train) [2][ 100/4092] lr: 9.7691e-05 eta: 7:40:49 time: 0.6733 data_time: 0.3978 memory: 6319 loss: 0.3066 +2023/06/06 01:53:20 - mmengine - INFO - Epoch(train) [2][ 200/4092] lr: 9.7580e-05 eta: 7:38:49 time: 0.6840 data_time: 0.1511 memory: 6319 loss: 0.2823 +2023/06/06 01:54:32 - mmengine - INFO - Epoch(train) [2][ 300/4092] lr: 9.7467e-05 eta: 7:37:01 time: 0.7212 data_time: 0.0024 memory: 6319 loss: 0.2794 +2023/06/06 01:55:43 - mmengine - INFO - Epoch(train) [2][ 400/4092] lr: 9.7352e-05 eta: 7:35:17 time: 0.7382 data_time: 0.0008 memory: 6319 loss: 0.2703 +2023/06/06 01:56:53 - mmengine - INFO - Epoch(train) [2][ 500/4092] lr: 9.7234e-05 eta: 7:33:23 time: 0.6471 data_time: 0.0008 memory: 6319 loss: 0.2680 +2023/06/06 01:58:06 - mmengine - INFO - Epoch(train) [2][ 600/4092] lr: 9.7113e-05 eta: 7:31:55 time: 0.8386 data_time: 0.0008 memory: 6319 loss: 0.2711 +2023/06/06 01:59:21 - mmengine - INFO - Epoch(train) [2][ 700/4092] lr: 9.6990e-05 eta: 7:30:36 time: 0.8524 data_time: 0.0009 memory: 6319 loss: 0.2783 +2023/06/06 02:00:31 - mmengine - INFO - Epoch(train) [2][ 800/4092] lr: 9.6865e-05 eta: 7:28:45 time: 0.6748 data_time: 0.0009 memory: 6319 loss: 0.2656 +2023/06/06 02:01:38 - mmengine - INFO - Epoch(train) [2][ 900/4092] lr: 9.6737e-05 eta: 7:26:40 time: 0.6461 data_time: 0.0010 memory: 6319 loss: 0.2584 +2023/06/06 02:01:45 - mmengine - INFO - Exp name: resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1_20230606_005743 +2023/06/06 02:02:48 - mmengine - INFO - Epoch(train) [2][1000/4092] lr: 9.6606e-05 eta: 7:24:52 time: 0.7332 data_time: 0.0012 memory: 6319 loss: 0.2699 +2023/06/06 02:04:00 - mmengine - INFO - Epoch(train) [2][1100/4092] lr: 9.6473e-05 eta: 7:23:19 time: 0.7268 data_time: 0.0009 memory: 6319 loss: 0.2681 +2023/06/06 02:05:12 - mmengine - INFO - Epoch(train) [2][1200/4092] lr: 9.6338e-05 eta: 7:21:45 time: 0.7184 data_time: 0.0010 memory: 6319 loss: 0.2734 +2023/06/06 02:06:22 - mmengine - INFO - Epoch(train) [2][1300/4092] lr: 9.6200e-05 eta: 7:20:05 time: 0.7383 data_time: 0.0007 memory: 6319 loss: 0.2741 +2023/06/06 02:07:32 - mmengine - INFO - Epoch(train) [2][1400/4092] lr: 9.6060e-05 eta: 7:18:19 time: 0.7033 data_time: 0.0007 memory: 6319 loss: 0.2644 +2023/06/06 02:08:39 - mmengine - INFO - Epoch(train) [2][1500/4092] lr: 9.5918e-05 eta: 7:16:24 time: 0.6655 data_time: 0.0010 memory: 6319 loss: 0.2552 +2023/06/06 02:09:47 - mmengine - INFO - Epoch(train) [2][1600/4092] lr: 9.5773e-05 eta: 7:14:28 time: 0.6554 data_time: 0.0010 memory: 6319 loss: 0.2630 +2023/06/06 02:10:56 - mmengine - INFO - Epoch(train) [2][1700/4092] lr: 9.5625e-05 eta: 7:12:43 time: 0.7198 data_time: 0.0009 memory: 6319 loss: 0.2622 +2023/06/06 02:12:01 - mmengine - INFO - Epoch(train) [2][1800/4092] lr: 9.5475e-05 eta: 7:10:37 time: 0.6761 data_time: 0.0008 memory: 6319 loss: 0.2603 +2023/06/06 02:13:14 - mmengine - INFO - Epoch(train) [2][1900/4092] lr: 9.5323e-05 eta: 7:09:20 time: 0.9657 data_time: 0.0009 memory: 6319 loss: 0.2766 +2023/06/06 02:13:21 - mmengine - INFO - Exp name: resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1_20230606_005743 +2023/06/06 02:14:23 - mmengine - INFO - Epoch(train) [2][2000/4092] lr: 9.5169e-05 eta: 7:07:37 time: 0.7123 data_time: 0.0009 memory: 6319 loss: 0.2436 +2023/06/06 02:15:30 - mmengine - INFO - Epoch(train) [2][2100/4092] lr: 9.5012e-05 eta: 7:05:49 time: 0.6767 data_time: 0.0008 memory: 6319 loss: 0.2770 +2023/06/06 02:16:41 - mmengine - INFO - Epoch(train) [2][2200/4092] lr: 9.4853e-05 eta: 7:04:18 time: 0.6891 data_time: 0.0011 memory: 6319 loss: 0.2426 +2023/06/06 02:17:51 - mmengine - INFO - Epoch(train) [2][2300/4092] lr: 9.4691e-05 eta: 7:02:49 time: 0.7354 data_time: 0.0008 memory: 6319 loss: 0.2462 +2023/06/06 02:18:59 - mmengine - INFO - Epoch(train) [2][2400/4092] lr: 9.4527e-05 eta: 7:01:07 time: 0.6790 data_time: 0.0008 memory: 6319 loss: 0.2553 +2023/06/06 02:20:08 - mmengine - INFO - Epoch(train) [2][2500/4092] lr: 9.4361e-05 eta: 6:59:31 time: 0.6513 data_time: 0.0008 memory: 6319 loss: 0.2425 +2023/06/06 02:21:18 - mmengine - INFO - Epoch(train) [2][2600/4092] lr: 9.4192e-05 eta: 6:58:00 time: 0.6749 data_time: 0.0008 memory: 6319 loss: 0.2689 +2023/06/06 02:22:27 - mmengine - INFO - Epoch(train) [2][2700/4092] lr: 9.4021e-05 eta: 6:56:25 time: 0.6558 data_time: 0.0007 memory: 6319 loss: 0.2409 +2023/06/06 02:23:36 - mmengine - INFO - Epoch(train) [2][2800/4092] lr: 9.3848e-05 eta: 6:54:52 time: 0.7019 data_time: 0.0009 memory: 6319 loss: 0.2770 +2023/06/06 02:24:45 - mmengine - INFO - Epoch(train) [2][2900/4092] lr: 9.3672e-05 eta: 6:53:19 time: 0.7396 data_time: 0.0008 memory: 6319 loss: 0.2310 +2023/06/06 02:24:52 - mmengine - INFO - Exp name: resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1_20230606_005743 +2023/06/06 02:25:55 - mmengine - INFO - Epoch(train) [2][3000/4092] lr: 9.3495e-05 eta: 6:51:50 time: 0.6743 data_time: 0.0012 memory: 6319 loss: 0.2553 +2023/06/06 02:27:04 - mmengine - INFO - Epoch(train) [2][3100/4092] lr: 9.3315e-05 eta: 6:50:17 time: 0.6631 data_time: 0.0009 memory: 6319 loss: 0.2469 +2023/06/06 02:28:14 - mmengine - INFO - Epoch(train) [2][3200/4092] lr: 9.3132e-05 eta: 6:48:48 time: 0.7029 data_time: 0.0016 memory: 6319 loss: 0.2429 +2023/06/06 02:29:24 - mmengine - INFO - Epoch(train) [2][3300/4092] lr: 9.2948e-05 eta: 6:47:26 time: 0.6996 data_time: 0.0008 memory: 6319 loss: 0.2546 +2023/06/06 02:30:35 - mmengine - INFO - Epoch(train) [2][3400/4092] lr: 9.2761e-05 eta: 6:46:04 time: 0.6885 data_time: 0.0007 memory: 6319 loss: 0.2544 +2023/06/06 02:31:46 - mmengine - INFO - Epoch(train) [2][3500/4092] lr: 9.2572e-05 eta: 6:44:40 time: 0.7154 data_time: 0.0009 memory: 6319 loss: 0.2529 +2023/06/06 02:32:57 - mmengine - INFO - Epoch(train) [2][3600/4092] lr: 9.2381e-05 eta: 6:43:20 time: 0.7095 data_time: 0.0008 memory: 6319 loss: 0.2443 +2023/06/06 02:34:14 - mmengine - INFO - Epoch(train) [2][3700/4092] lr: 9.2187e-05 eta: 6:42:24 time: 0.7587 data_time: 0.0011 memory: 6319 loss: 0.2309 +2023/06/06 02:35:26 - mmengine - INFO - Epoch(train) [2][3800/4092] lr: 9.1991e-05 eta: 6:41:08 time: 0.6877 data_time: 0.0011 memory: 6319 loss: 0.2499 +2023/06/06 02:36:39 - mmengine - INFO - Epoch(train) [2][3900/4092] lr: 9.1794e-05 eta: 6:39:55 time: 0.6867 data_time: 0.0007 memory: 6319 loss: 0.2335 +2023/06/06 02:36:46 - mmengine - INFO - Exp name: resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1_20230606_005743 +2023/06/06 02:37:57 - mmengine - INFO - Epoch(train) [2][4000/4092] lr: 9.1594e-05 eta: 6:39:04 time: 0.7114 data_time: 0.0007 memory: 6319 loss: 0.2362 +2023/06/06 02:39:02 - mmengine - INFO - Exp name: resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1_20230606_005743 +2023/06/06 02:39:02 - mmengine - INFO - Saving checkpoint at 2 epochs +2023/06/06 02:39:43 - mmengine - INFO - Epoch(val) [2][100/119] eta: 0:00:06 time: 0.6301 data_time: 0.5429 memory: 6319 +2023/06/06 02:40:10 - mmengine - INFO - Epoch(val) [2][119/119] accuracy/top1: 81.0155 data_time: 0.3427 time: 0.4298 +2023/06/06 02:41:23 - mmengine - INFO - Epoch(train) [3][ 100/4092] lr: 9.1204e-05 eta: 6:36:35 time: 0.6739 data_time: 0.4689 memory: 6319 loss: 0.2375 +2023/06/06 02:42:34 - mmengine - INFO - Epoch(train) [3][ 200/4092] lr: 9.0997e-05 eta: 6:35:13 time: 0.7003 data_time: 0.3339 memory: 6319 loss: 0.2527 +2023/06/06 02:43:43 - mmengine - INFO - Epoch(train) [3][ 300/4092] lr: 9.0789e-05 eta: 6:33:47 time: 0.6683 data_time: 0.2901 memory: 6319 loss: 0.2347 +2023/06/06 02:44:58 - mmengine - INFO - Epoch(train) [3][ 400/4092] lr: 9.0579e-05 eta: 6:32:40 time: 0.6562 data_time: 0.3168 memory: 6319 loss: 0.2416 +2023/06/06 02:46:08 - mmengine - INFO - Epoch(train) [3][ 500/4092] lr: 9.0366e-05 eta: 6:31:17 time: 0.7558 data_time: 0.5717 memory: 6319 loss: 0.2165 +2023/06/06 02:47:19 - mmengine - INFO - Epoch(train) [3][ 600/4092] lr: 9.0151e-05 eta: 6:29:59 time: 0.6772 data_time: 0.5375 memory: 6319 loss: 0.2230 +2023/06/06 02:48:29 - mmengine - INFO - Epoch(train) [3][ 700/4092] lr: 8.9935e-05 eta: 6:28:35 time: 0.6904 data_time: 0.4909 memory: 6319 loss: 0.2312 +2023/06/06 02:49:42 - mmengine - INFO - Epoch(train) [3][ 800/4092] lr: 8.9716e-05 eta: 6:27:22 time: 0.6909 data_time: 0.5512 memory: 6319 loss: 0.2282 +2023/06/06 02:49:55 - mmengine - INFO - Exp name: resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1_20230606_005743 +2023/06/06 02:50:53 - mmengine - INFO - Epoch(train) [3][ 900/4092] lr: 8.9495e-05 eta: 6:26:04 time: 0.7130 data_time: 0.5730 memory: 6319 loss: 0.2252 +2023/06/06 02:52:05 - mmengine - INFO - Epoch(train) [3][1000/4092] lr: 8.9272e-05 eta: 6:24:49 time: 0.7019 data_time: 0.5608 memory: 6319 loss: 0.2453 +2023/06/06 02:53:15 - mmengine - INFO - Epoch(train) [3][1100/4092] lr: 8.9047e-05 eta: 6:23:25 time: 0.7277 data_time: 0.5887 memory: 6319 loss: 0.2265 +2023/06/06 02:54:26 - mmengine - INFO - Epoch(train) [3][1200/4092] lr: 8.8820e-05 eta: 6:22:06 time: 0.6716 data_time: 0.5279 memory: 6319 loss: 0.2352 +2023/06/06 02:55:35 - mmengine - INFO - Epoch(train) [3][1300/4092] lr: 8.8591e-05 eta: 6:20:43 time: 0.7038 data_time: 0.5630 memory: 6319 loss: 0.2377 +2023/06/06 02:56:46 - mmengine - INFO - Epoch(train) [3][1400/4092] lr: 8.8360e-05 eta: 6:19:23 time: 0.6769 data_time: 0.5362 memory: 6319 loss: 0.2327 +2023/06/06 02:57:55 - mmengine - INFO - Epoch(train) [3][1500/4092] lr: 8.8128e-05 eta: 6:17:59 time: 0.7121 data_time: 0.5612 memory: 6319 loss: 0.2297 +2023/06/06 02:59:10 - mmengine - INFO - Epoch(train) [3][1600/4092] lr: 8.7893e-05 eta: 6:16:54 time: 0.6463 data_time: 0.5060 memory: 6319 loss: 0.2340 +2023/06/06 03:00:20 - mmengine - INFO - Epoch(train) [3][1700/4092] lr: 8.7656e-05 eta: 6:15:34 time: 0.6686 data_time: 0.5291 memory: 6319 loss: 0.2326 +2023/06/06 03:01:30 - mmengine - INFO - Epoch(train) [3][1800/4092] lr: 8.7417e-05 eta: 6:14:14 time: 0.6653 data_time: 0.5214 memory: 6319 loss: 0.2310 +2023/06/06 03:01:42 - mmengine - INFO - Exp name: resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1_20230606_005743 +2023/06/06 03:02:37 - mmengine - INFO - Epoch(train) [3][1900/4092] lr: 8.7177e-05 eta: 6:12:44 time: 0.7430 data_time: 0.4484 memory: 6319 loss: 0.2426 +2023/06/06 03:03:47 - mmengine - INFO - Epoch(train) [3][2000/4092] lr: 8.6934e-05 eta: 6:11:24 time: 0.6700 data_time: 0.3338 memory: 6319 loss: 0.2122 +2023/06/06 03:04:57 - mmengine - INFO - Epoch(train) [3][2100/4092] lr: 8.6690e-05 eta: 6:10:05 time: 0.7080 data_time: 0.1465 memory: 6319 loss: 0.2160 +2023/06/06 03:06:07 - mmengine - INFO - Epoch(train) [3][2200/4092] lr: 8.6444e-05 eta: 6:08:43 time: 0.6683 data_time: 0.0652 memory: 6319 loss: 0.2329 +2023/06/06 03:07:21 - mmengine - INFO - Epoch(train) [3][2300/4092] lr: 8.6196e-05 eta: 6:07:36 time: 0.7149 data_time: 0.1617 memory: 6319 loss: 0.2339 +2023/06/06 03:08:34 - mmengine - INFO - Epoch(train) [3][2400/4092] lr: 8.5946e-05 eta: 6:06:26 time: 0.7036 data_time: 0.1618 memory: 6319 loss: 0.2193 +2023/06/06 03:09:46 - mmengine - INFO - Epoch(train) [3][2500/4092] lr: 8.5694e-05 eta: 6:05:12 time: 0.6862 data_time: 0.0010 memory: 6319 loss: 0.2251 +2023/06/06 03:10:57 - mmengine - INFO - Epoch(train) [3][2600/4092] lr: 8.5441e-05 eta: 6:03:55 time: 0.7535 data_time: 0.0008 memory: 6319 loss: 0.2095 +2023/06/06 03:12:09 - mmengine - INFO - Epoch(train) [3][2700/4092] lr: 8.5185e-05 eta: 6:02:41 time: 0.7049 data_time: 0.0008 memory: 6319 loss: 0.2110 +2023/06/06 03:13:23 - mmengine - INFO - Epoch(train) [3][2800/4092] lr: 8.4928e-05 eta: 6:01:32 time: 0.7746 data_time: 0.0009 memory: 6319 loss: 0.2166 +2023/06/06 03:13:31 - mmengine - INFO - Exp name: resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1_20230606_005743 +2023/06/06 03:14:38 - mmengine - INFO - Epoch(train) [3][2900/4092] lr: 8.4669e-05 eta: 6:00:27 time: 0.7173 data_time: 0.0008 memory: 6319 loss: 0.2160 +2023/06/06 03:15:49 - mmengine - INFO - Epoch(train) [3][3000/4092] lr: 8.4409e-05 eta: 5:59:11 time: 0.6983 data_time: 0.0008 memory: 6319 loss: 0.2236 +2023/06/06 03:17:02 - mmengine - INFO - Epoch(train) [3][3100/4092] lr: 8.4146e-05 eta: 5:57:59 time: 0.7384 data_time: 0.0009 memory: 6319 loss: 0.2074 +2023/06/06 03:18:15 - mmengine - INFO - Epoch(train) [3][3200/4092] lr: 8.3882e-05 eta: 5:56:49 time: 0.7332 data_time: 0.0009 memory: 6319 loss: 0.2073 +2023/06/06 03:19:27 - mmengine - INFO - Epoch(train) [3][3300/4092] lr: 8.3616e-05 eta: 5:55:34 time: 0.7300 data_time: 0.0009 memory: 6319 loss: 0.2079 +2023/06/06 03:20:40 - mmengine - INFO - Epoch(train) [3][3400/4092] lr: 8.3349e-05 eta: 5:54:24 time: 0.8191 data_time: 0.0009 memory: 6319 loss: 0.1952 +2023/06/06 03:21:52 - mmengine - INFO - Epoch(train) [3][3500/4092] lr: 8.3080e-05 eta: 5:53:08 time: 0.7053 data_time: 0.0009 memory: 6319 loss: 0.2118 +2023/06/06 03:23:10 - mmengine - INFO - Epoch(train) [3][3600/4092] lr: 8.2809e-05 eta: 5:52:11 time: 1.2415 data_time: 0.0008 memory: 6319 loss: 0.2278 +2023/06/06 03:24:25 - mmengine - INFO - Epoch(train) [3][3700/4092] lr: 8.2537e-05 eta: 5:51:04 time: 0.6985 data_time: 0.0008 memory: 6319 loss: 0.2131 +2023/06/06 03:25:38 - mmengine - INFO - Epoch(train) [3][3800/4092] lr: 8.2263e-05 eta: 5:49:53 time: 0.7291 data_time: 0.0007 memory: 6319 loss: 0.2232 +2023/06/06 03:25:52 - mmengine - INFO - Exp name: resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1_20230606_005743 +2023/06/06 03:26:51 - mmengine - INFO - Epoch(train) [3][3900/4092] lr: 8.1987e-05 eta: 5:48:42 time: 0.7389 data_time: 0.0008 memory: 6319 loss: 0.2098 +2023/06/06 03:28:04 - mmengine - INFO - Epoch(train) [3][4000/4092] lr: 8.1710e-05 eta: 5:47:31 time: 0.7250 data_time: 0.0007 memory: 6319 loss: 0.2235 +2023/06/06 03:29:14 - mmengine - INFO - Exp name: resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1_20230606_005743 +2023/06/06 03:29:14 - mmengine - INFO - Saving checkpoint at 3 epochs +2023/06/06 03:29:55 - mmengine - INFO - Epoch(val) [3][100/119] eta: 0:00:06 time: 0.6817 data_time: 0.5920 memory: 6319 +2023/06/06 03:30:22 - mmengine - INFO - Epoch(val) [3][119/119] accuracy/top1: 82.5017 data_time: 0.3352 time: 0.4236 +2023/06/06 03:31:35 - mmengine - INFO - Epoch(train) [4][ 100/4092] lr: 8.1173e-05 eta: 5:45:20 time: 0.7653 data_time: 0.3452 memory: 6319 loss: 0.2124 +2023/06/06 03:32:48 - mmengine - INFO - Epoch(train) [4][ 200/4092] lr: 8.0891e-05 eta: 5:44:06 time: 0.7129 data_time: 0.1697 memory: 6319 loss: 0.2119 +2023/06/06 03:33:59 - mmengine - INFO - Epoch(train) [4][ 300/4092] lr: 8.0608e-05 eta: 5:42:51 time: 0.6693 data_time: 0.2002 memory: 6319 loss: 0.2057 +2023/06/06 03:35:14 - mmengine - INFO - Epoch(train) [4][ 400/4092] lr: 8.0323e-05 eta: 5:41:44 time: 0.7680 data_time: 0.0008 memory: 6319 loss: 0.2183 +2023/06/06 03:36:27 - mmengine - INFO - Epoch(train) [4][ 500/4092] lr: 8.0037e-05 eta: 5:40:32 time: 0.7451 data_time: 0.0007 memory: 6319 loss: 0.2009 +2023/06/06 03:37:41 - mmengine - INFO - Epoch(train) [4][ 600/4092] lr: 7.9749e-05 eta: 5:39:22 time: 0.7786 data_time: 0.0137 memory: 6319 loss: 0.2086 +2023/06/06 03:38:52 - mmengine - INFO - Epoch(train) [4][ 700/4092] lr: 7.9459e-05 eta: 5:38:07 time: 0.7002 data_time: 0.0008 memory: 6319 loss: 0.2154 +2023/06/06 03:39:07 - mmengine - INFO - Exp name: resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1_20230606_005743 +2023/06/06 03:40:09 - mmengine - INFO - Epoch(train) [4][ 800/4092] lr: 7.9169e-05 eta: 5:37:03 time: 0.7613 data_time: 0.0008 memory: 6319 loss: 0.2074 +2023/06/06 03:41:21 - mmengine - INFO - Epoch(train) [4][ 900/4092] lr: 7.8877e-05 eta: 5:35:49 time: 0.7719 data_time: 0.0010 memory: 6319 loss: 0.2295 +2023/06/06 03:42:34 - mmengine - INFO - Epoch(train) [4][1000/4092] lr: 7.8583e-05 eta: 5:34:37 time: 0.6818 data_time: 0.0007 memory: 6319 loss: 0.2154 +2023/06/06 03:43:48 - mmengine - INFO - Epoch(train) [4][1100/4092] lr: 7.8288e-05 eta: 5:33:28 time: 0.7383 data_time: 0.0008 memory: 6319 loss: 0.2053 +2023/06/06 03:45:00 - mmengine - INFO - Epoch(train) [4][1200/4092] lr: 7.7992e-05 eta: 5:32:14 time: 0.7470 data_time: 0.0008 memory: 6319 loss: 0.2038 +2023/06/06 03:46:14 - mmengine - INFO - Epoch(train) [4][1300/4092] lr: 7.7694e-05 eta: 5:31:03 time: 0.7960 data_time: 0.0008 memory: 6319 loss: 0.1935 +2023/06/06 03:47:26 - mmengine - INFO - Epoch(train) [4][1400/4092] lr: 7.7395e-05 eta: 5:29:50 time: 0.7113 data_time: 0.0009 memory: 6319 loss: 0.1868 +2023/06/06 03:48:41 - mmengine - INFO - Epoch(train) [4][1500/4092] lr: 7.7095e-05 eta: 5:28:41 time: 0.7173 data_time: 0.0008 memory: 6319 loss: 0.2074 +2023/06/06 03:49:57 - mmengine - INFO - Epoch(train) [4][1600/4092] lr: 7.6793e-05 eta: 5:27:36 time: 0.7048 data_time: 0.0008 memory: 6319 loss: 0.2050 +2023/06/06 03:51:11 - mmengine - INFO - Epoch(train) [4][1700/4092] lr: 7.6490e-05 eta: 5:26:26 time: 0.7432 data_time: 0.0011 memory: 6319 loss: 0.2100 +2023/06/06 03:51:26 - mmengine - INFO - Exp name: resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1_20230606_005743 +2023/06/06 03:52:25 - mmengine - INFO - Epoch(train) [4][1800/4092] lr: 7.6186e-05 eta: 5:25:16 time: 0.7548 data_time: 0.0008 memory: 6319 loss: 0.2036 +2023/06/06 03:53:40 - mmengine - INFO - Epoch(train) [4][1900/4092] lr: 7.5881e-05 eta: 5:24:07 time: 0.7574 data_time: 0.0009 memory: 6319 loss: 0.2252 +2023/06/06 03:54:53 - mmengine - INFO - Epoch(train) [4][2000/4092] lr: 7.5574e-05 eta: 5:22:55 time: 0.6884 data_time: 0.0010 memory: 6319 loss: 0.1938 +2023/06/06 03:56:07 - mmengine - INFO - Epoch(train) [4][2100/4092] lr: 7.5266e-05 eta: 5:21:43 time: 0.7812 data_time: 0.0008 memory: 6319 loss: 0.1937 +2023/06/06 03:57:20 - mmengine - INFO - Epoch(train) [4][2200/4092] lr: 7.4957e-05 eta: 5:20:31 time: 0.6806 data_time: 0.0008 memory: 6319 loss: 0.2127 +2023/06/06 03:58:33 - mmengine - INFO - Epoch(train) [4][2300/4092] lr: 7.4647e-05 eta: 5:19:20 time: 0.7339 data_time: 0.0008 memory: 6319 loss: 0.2008 +2023/06/06 03:59:42 - mmengine - INFO - Epoch(train) [4][2400/4092] lr: 7.4336e-05 eta: 5:18:00 time: 0.8199 data_time: 0.0009 memory: 6319 loss: 0.2070 +2023/06/06 04:00:49 - mmengine - INFO - Epoch(train) [4][2500/4092] lr: 7.4023e-05 eta: 5:16:38 time: 0.6465 data_time: 0.0008 memory: 6319 loss: 0.2020 +2023/06/06 04:01:59 - mmengine - INFO - Epoch(train) [4][2600/4092] lr: 7.3709e-05 eta: 5:15:20 time: 0.6653 data_time: 0.0007 memory: 6319 loss: 0.1918 +2023/06/06 04:03:12 - mmengine - INFO - Epoch(train) [4][2700/4092] lr: 7.3395e-05 eta: 5:14:08 time: 0.7855 data_time: 0.0009 memory: 6319 loss: 0.2002 +2023/06/06 04:03:26 - mmengine - INFO - Exp name: resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1_20230606_005743 +2023/06/06 04:04:23 - mmengine - INFO - Epoch(train) [4][2800/4092] lr: 7.3079e-05 eta: 5:12:52 time: 0.7003 data_time: 0.0008 memory: 6319 loss: 0.1922 +2023/06/06 04:05:33 - mmengine - INFO - Epoch(train) [4][2900/4092] lr: 7.2762e-05 eta: 5:11:35 time: 0.6478 data_time: 0.0011 memory: 6319 loss: 0.1829 +2023/06/06 04:06:43 - mmengine - INFO - Epoch(train) [4][3000/4092] lr: 7.2444e-05 eta: 5:10:19 time: 0.6514 data_time: 0.0009 memory: 6319 loss: 0.2042 +2023/06/06 04:07:52 - mmengine - INFO - Epoch(train) [4][3100/4092] lr: 7.2125e-05 eta: 5:09:00 time: 0.6860 data_time: 0.0008 memory: 6319 loss: 0.2087 +2023/06/06 04:09:02 - mmengine - INFO - Epoch(train) [4][3200/4092] lr: 7.1805e-05 eta: 5:07:42 time: 0.6912 data_time: 0.0009 memory: 6319 loss: 0.1959 +2023/06/06 04:10:14 - mmengine - INFO - Epoch(train) [4][3300/4092] lr: 7.1484e-05 eta: 5:06:29 time: 0.7713 data_time: 0.0008 memory: 6319 loss: 0.1932 +2023/06/06 04:11:24 - mmengine - INFO - Epoch(train) [4][3400/4092] lr: 7.1162e-05 eta: 5:05:13 time: 0.6500 data_time: 0.0008 memory: 6319 loss: 0.2022 +2023/06/06 04:12:35 - mmengine - INFO - Epoch(train) [4][3500/4092] lr: 7.0839e-05 eta: 5:03:58 time: 0.6702 data_time: 0.0008 memory: 6319 loss: 0.2044 +2023/06/06 04:13:51 - mmengine - INFO - Epoch(train) [4][3600/4092] lr: 7.0515e-05 eta: 5:02:50 time: 0.7106 data_time: 0.0008 memory: 6319 loss: 0.1811 +2023/06/06 04:15:08 - mmengine - INFO - Epoch(train) [4][3700/4092] lr: 7.0191e-05 eta: 5:01:45 time: 0.6673 data_time: 0.0008 memory: 6319 loss: 0.1854 +2023/06/06 04:15:23 - mmengine - INFO - Exp name: resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1_20230606_005743 +2023/06/06 04:16:18 - mmengine - INFO - Epoch(train) [4][3800/4092] lr: 6.9865e-05 eta: 5:00:29 time: 0.6873 data_time: 0.0008 memory: 6319 loss: 0.1951 +2023/06/06 04:17:27 - mmengine - INFO - Epoch(train) [4][3900/4092] lr: 6.9538e-05 eta: 4:59:11 time: 0.6796 data_time: 0.0008 memory: 6319 loss: 0.1844 +2023/06/06 04:18:39 - mmengine - INFO - Epoch(train) [4][4000/4092] lr: 6.9211e-05 eta: 4:57:56 time: 0.7152 data_time: 0.0008 memory: 6319 loss: 0.1930 +2023/06/06 04:19:41 - mmengine - INFO - Exp name: resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1_20230606_005743 +2023/06/06 04:19:41 - mmengine - INFO - Saving checkpoint at 4 epochs +2023/06/06 04:20:23 - mmengine - INFO - Epoch(val) [4][100/119] eta: 0:00:06 time: 0.6212 data_time: 0.5338 memory: 6319 +2023/06/06 04:20:49 - mmengine - INFO - Epoch(val) [4][119/119] accuracy/top1: 87.8523 data_time: 0.3359 time: 0.4238 +2023/06/06 04:22:03 - mmengine - INFO - Epoch(train) [5][ 100/4092] lr: 6.8580e-05 eta: 4:55:33 time: 0.7527 data_time: 0.4232 memory: 6319 loss: 0.1879 +2023/06/06 04:23:16 - mmengine - INFO - Epoch(train) [5][ 200/4092] lr: 6.8250e-05 eta: 4:54:21 time: 0.7094 data_time: 0.0855 memory: 6319 loss: 0.1848 +2023/06/06 04:24:29 - mmengine - INFO - Epoch(train) [5][ 300/4092] lr: 6.7920e-05 eta: 4:53:09 time: 0.7260 data_time: 0.0009 memory: 6319 loss: 0.1809 +2023/06/06 04:25:43 - mmengine - INFO - Epoch(train) [5][ 400/4092] lr: 6.7588e-05 eta: 4:51:58 time: 0.7914 data_time: 0.0009 memory: 6319 loss: 0.1817 +2023/06/06 04:26:55 - mmengine - INFO - Epoch(train) [5][ 500/4092] lr: 6.7256e-05 eta: 4:50:45 time: 0.7367 data_time: 0.0010 memory: 6319 loss: 0.1991 +2023/06/06 04:28:09 - mmengine - INFO - Epoch(train) [5][ 600/4092] lr: 6.6924e-05 eta: 4:49:35 time: 0.7934 data_time: 0.0008 memory: 6319 loss: 0.1778 +2023/06/06 04:28:31 - mmengine - INFO - Exp name: resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1_20230606_005743 +2023/06/06 04:29:22 - mmengine - INFO - Epoch(train) [5][ 700/4092] lr: 6.6590e-05 eta: 4:48:23 time: 0.7456 data_time: 0.0008 memory: 6319 loss: 0.1894 +2023/06/06 04:30:34 - mmengine - INFO - Epoch(train) [5][ 800/4092] lr: 6.6256e-05 eta: 4:47:10 time: 0.7498 data_time: 0.0008 memory: 6319 loss: 0.1915 +2023/06/06 04:31:47 - mmengine - INFO - Epoch(train) [5][ 900/4092] lr: 6.5921e-05 eta: 4:45:58 time: 0.7293 data_time: 0.0008 memory: 6319 loss: 0.1988 +2023/06/06 04:32:59 - mmengine - INFO - Epoch(train) [5][1000/4092] lr: 6.5586e-05 eta: 4:44:44 time: 0.7052 data_time: 0.0089 memory: 6319 loss: 0.1941 +2023/06/06 04:34:12 - mmengine - INFO - Epoch(train) [5][1100/4092] lr: 6.5250e-05 eta: 4:43:33 time: 0.7157 data_time: 0.3020 memory: 6319 loss: 0.2048 +2023/06/06 04:35:24 - mmengine - INFO - Epoch(train) [5][1200/4092] lr: 6.4913e-05 eta: 4:42:19 time: 0.7213 data_time: 0.1752 memory: 6319 loss: 0.1896 +2023/06/06 04:36:46 - mmengine - INFO - Epoch(train) [5][1300/4092] lr: 6.4576e-05 eta: 4:41:20 time: 1.1040 data_time: 0.1012 memory: 6319 loss: 0.1927 +2023/06/06 04:38:00 - mmengine - INFO - Epoch(train) [5][1400/4092] lr: 6.4238e-05 eta: 4:40:09 time: 0.7296 data_time: 0.0009 memory: 6319 loss: 0.2042 +2023/06/06 04:39:13 - mmengine - INFO - Epoch(train) [5][1500/4092] lr: 6.3899e-05 eta: 4:38:57 time: 0.7837 data_time: 0.1670 memory: 6319 loss: 0.1982 +2023/06/06 04:40:27 - mmengine - INFO - Epoch(train) [5][1600/4092] lr: 6.3560e-05 eta: 4:37:46 time: 0.7807 data_time: 0.3579 memory: 6319 loss: 0.1835 +2023/06/06 04:40:49 - mmengine - INFO - Exp name: resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1_20230606_005743 +2023/06/06 04:41:38 - mmengine - INFO - Epoch(train) [5][1700/4092] lr: 6.3221e-05 eta: 4:36:31 time: 0.7191 data_time: 0.2286 memory: 6319 loss: 0.1985 +2023/06/06 04:42:52 - mmengine - INFO - Epoch(train) [5][1800/4092] lr: 6.2881e-05 eta: 4:35:20 time: 0.7471 data_time: 0.2171 memory: 6319 loss: 0.1943 +2023/06/06 04:44:04 - mmengine - INFO - Epoch(train) [5][1900/4092] lr: 6.2541e-05 eta: 4:34:07 time: 0.6766 data_time: 0.3013 memory: 6319 loss: 0.1803 +2023/06/06 04:45:16 - mmengine - INFO - Epoch(train) [5][2000/4092] lr: 6.2200e-05 eta: 4:32:54 time: 0.8018 data_time: 0.0008 memory: 6319 loss: 0.1824 +2023/06/06 04:46:29 - mmengine - INFO - Epoch(train) [5][2100/4092] lr: 6.1859e-05 eta: 4:31:42 time: 0.7042 data_time: 0.0009 memory: 6319 loss: 0.1862 +2023/06/06 04:47:43 - mmengine - INFO - Epoch(train) [5][2200/4092] lr: 6.1517e-05 eta: 4:30:30 time: 0.7313 data_time: 0.0009 memory: 6319 loss: 0.1891 +2023/06/06 04:48:54 - mmengine - INFO - Epoch(train) [5][2300/4092] lr: 6.1175e-05 eta: 4:29:16 time: 0.6894 data_time: 0.0008 memory: 6319 loss: 0.2005 +2023/06/06 04:50:07 - mmengine - INFO - Epoch(train) [5][2400/4092] lr: 6.0833e-05 eta: 4:28:04 time: 0.7661 data_time: 0.0008 memory: 6319 loss: 0.1876 +2023/06/06 04:51:21 - mmengine - INFO - Epoch(train) [5][2500/4092] lr: 6.0490e-05 eta: 4:26:53 time: 0.7509 data_time: 0.0009 memory: 6319 loss: 0.1776 +2023/06/06 04:52:34 - mmengine - INFO - Epoch(train) [5][2600/4092] lr: 6.0147e-05 eta: 4:25:41 time: 0.7639 data_time: 0.0009 memory: 6319 loss: 0.1656 +2023/06/06 04:52:58 - mmengine - INFO - Exp name: resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1_20230606_005743 +2023/06/06 04:53:51 - mmengine - INFO - Epoch(train) [5][2700/4092] lr: 5.9803e-05 eta: 4:24:33 time: 1.1508 data_time: 0.0010 memory: 6319 loss: 0.1880 +2023/06/06 04:55:04 - mmengine - INFO - Epoch(train) [5][2800/4092] lr: 5.9460e-05 eta: 4:23:20 time: 0.6970 data_time: 0.0010 memory: 6319 loss: 0.1880 +2023/06/06 04:56:17 - mmengine - INFO - Epoch(train) [5][2900/4092] lr: 5.9116e-05 eta: 4:22:09 time: 0.6837 data_time: 0.0009 memory: 6319 loss: 0.1804 +2023/06/06 04:57:30 - mmengine - INFO - Epoch(train) [5][3000/4092] lr: 5.8772e-05 eta: 4:20:56 time: 0.7440 data_time: 0.0008 memory: 6319 loss: 0.1687 +2023/06/06 04:58:41 - mmengine - INFO - Epoch(train) [5][3100/4092] lr: 5.8427e-05 eta: 4:19:42 time: 0.6986 data_time: 0.0010 memory: 6319 loss: 0.1765 +2023/06/06 04:59:53 - mmengine - INFO - Epoch(train) [5][3200/4092] lr: 5.8083e-05 eta: 4:18:28 time: 0.7481 data_time: 0.0008 memory: 6319 loss: 0.1798 +2023/06/06 05:01:05 - mmengine - INFO - Epoch(train) [5][3300/4092] lr: 5.7738e-05 eta: 4:17:15 time: 0.7215 data_time: 0.0008 memory: 6319 loss: 0.1844 +2023/06/06 05:02:21 - mmengine - INFO - Epoch(train) [5][3400/4092] lr: 5.7393e-05 eta: 4:16:06 time: 0.7331 data_time: 0.0008 memory: 6319 loss: 0.1776 +2023/06/06 05:03:34 - mmengine - INFO - Epoch(train) [5][3500/4092] lr: 5.7048e-05 eta: 4:14:54 time: 0.7413 data_time: 0.0008 memory: 6319 loss: 0.1897 +2023/06/06 05:04:47 - mmengine - INFO - Epoch(train) [5][3600/4092] lr: 5.6703e-05 eta: 4:13:41 time: 0.7372 data_time: 0.0008 memory: 6319 loss: 0.1966 +2023/06/06 05:05:10 - mmengine - INFO - Exp name: resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1_20230606_005743 +2023/06/06 05:06:00 - mmengine - INFO - Epoch(train) [5][3700/4092] lr: 5.6358e-05 eta: 4:12:29 time: 0.7181 data_time: 0.0008 memory: 6319 loss: 0.1727 +2023/06/06 05:07:14 - mmengine - INFO - Epoch(train) [5][3800/4092] lr: 5.6012e-05 eta: 4:11:18 time: 0.7618 data_time: 0.0008 memory: 6319 loss: 0.1865 +2023/06/06 05:08:27 - mmengine - INFO - Epoch(train) [5][3900/4092] lr: 5.5667e-05 eta: 4:10:06 time: 0.7369 data_time: 0.0009 memory: 6319 loss: 0.1806 +2023/06/06 05:09:42 - mmengine - INFO - Epoch(train) [5][4000/4092] lr: 5.5321e-05 eta: 4:08:55 time: 0.7057 data_time: 0.0008 memory: 6319 loss: 0.1744 +2023/06/06 05:10:47 - mmengine - INFO - Exp name: resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1_20230606_005743 +2023/06/06 05:10:47 - mmengine - INFO - Saving checkpoint at 5 epochs +2023/06/06 05:11:28 - mmengine - INFO - Epoch(val) [5][100/119] eta: 0:00:06 time: 0.6983 data_time: 0.6109 memory: 6319 +2023/06/06 05:11:55 - mmengine - INFO - Epoch(val) [5][119/119] accuracy/top1: 90.0419 data_time: 0.3324 time: 0.4194 +2023/06/06 05:13:11 - mmengine - INFO - Epoch(train) [6][ 100/4092] lr: 5.4658e-05 eta: 4:06:37 time: 0.7909 data_time: 0.4620 memory: 6319 loss: 0.1931 +2023/06/06 05:14:25 - mmengine - INFO - Epoch(train) [6][ 200/4092] lr: 5.4313e-05 eta: 4:05:26 time: 0.7966 data_time: 0.2947 memory: 6319 loss: 0.1745 +2023/06/06 05:15:39 - mmengine - INFO - Epoch(train) [6][ 300/4092] lr: 5.3967e-05 eta: 4:04:14 time: 0.7237 data_time: 0.0064 memory: 6319 loss: 0.1741 +2023/06/06 05:16:51 - mmengine - INFO - Epoch(train) [6][ 400/4092] lr: 5.3622e-05 eta: 4:03:01 time: 0.7044 data_time: 0.0008 memory: 6319 loss: 0.1696 +2023/06/06 05:18:03 - mmengine - INFO - Epoch(train) [6][ 500/4092] lr: 5.3276e-05 eta: 4:01:49 time: 0.7209 data_time: 0.0009 memory: 6319 loss: 0.1818 +2023/06/06 05:18:37 - mmengine - INFO - Exp name: resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1_20230606_005743 +2023/06/06 05:19:24 - mmengine - INFO - Epoch(train) [6][ 600/4092] lr: 5.2931e-05 eta: 4:00:43 time: 0.6940 data_time: 0.0009 memory: 6319 loss: 0.1800 +2023/06/06 05:20:45 - mmengine - INFO - Epoch(train) [6][ 700/4092] lr: 5.2586e-05 eta: 3:59:38 time: 0.7270 data_time: 0.0012 memory: 6319 loss: 0.1700 +2023/06/06 05:21:57 - mmengine - INFO - Epoch(train) [6][ 800/4092] lr: 5.2241e-05 eta: 3:58:25 time: 0.7370 data_time: 0.0009 memory: 6319 loss: 0.1735 +2023/06/06 05:23:10 - mmengine - INFO - Epoch(train) [6][ 900/4092] lr: 5.1897e-05 eta: 3:57:12 time: 0.7386 data_time: 0.0009 memory: 6319 loss: 0.1841 +2023/06/06 05:24:30 - mmengine - INFO - Epoch(train) [6][1000/4092] lr: 5.1552e-05 eta: 3:56:06 time: 0.9761 data_time: 0.0009 memory: 6319 loss: 0.1671 +2023/06/06 05:25:50 - mmengine - INFO - Epoch(train) [6][1100/4092] lr: 5.1208e-05 eta: 3:55:00 time: 0.7023 data_time: 0.0009 memory: 6319 loss: 0.1853 +2023/06/06 05:27:00 - mmengine - INFO - Epoch(train) [6][1200/4092] lr: 5.0864e-05 eta: 3:53:45 time: 0.7260 data_time: 0.0008 memory: 6319 loss: 0.1834 +2023/06/06 05:28:12 - mmengine - INFO - Epoch(train) [6][1300/4092] lr: 5.0520e-05 eta: 3:52:31 time: 0.6930 data_time: 0.0008 memory: 6319 loss: 0.1743 +2023/06/06 05:29:19 - mmengine - INFO - Epoch(train) [6][1400/4092] lr: 5.0176e-05 eta: 3:51:13 time: 0.6507 data_time: 0.0008 memory: 6319 loss: 0.1836 +2023/06/06 05:30:34 - mmengine - INFO - Epoch(train) [6][1500/4092] lr: 4.9833e-05 eta: 3:50:02 time: 0.7906 data_time: 0.0009 memory: 6319 loss: 0.1709 +2023/06/06 05:31:02 - mmengine - INFO - Exp name: resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1_20230606_005743 +2023/06/06 05:31:45 - mmengine - INFO - Epoch(train) [6][1600/4092] lr: 4.9490e-05 eta: 3:48:48 time: 0.6766 data_time: 0.0010 memory: 6319 loss: 0.1594 +2023/06/06 05:32:56 - mmengine - INFO - Epoch(train) [6][1700/4092] lr: 4.9147e-05 eta: 3:47:33 time: 0.7720 data_time: 0.0009 memory: 6319 loss: 0.1905 +2023/06/06 05:34:07 - mmengine - INFO - Epoch(train) [6][1800/4092] lr: 4.8805e-05 eta: 3:46:19 time: 0.7351 data_time: 0.0009 memory: 6319 loss: 0.1803 +2023/06/06 05:36:49 - mmengine - INFO - Epoch(train) [6][1900/4092] lr: 4.8462e-05 eta: 3:46:21 time: 0.6649 data_time: 0.0009 memory: 6319 loss: 0.1862 +2023/06/06 05:38:02 - mmengine - INFO - Epoch(train) [6][2000/4092] lr: 4.8121e-05 eta: 3:45:07 time: 0.7321 data_time: 0.0009 memory: 6319 loss: 0.1694 +2023/06/06 05:39:24 - mmengine - INFO - Epoch(train) [6][2100/4092] lr: 4.7780e-05 eta: 3:44:01 time: 0.6986 data_time: 0.0007 memory: 6319 loss: 0.1774 +2023/06/06 05:40:36 - mmengine - INFO - Epoch(train) [6][2200/4092] lr: 4.7439e-05 eta: 3:42:47 time: 0.6732 data_time: 0.0008 memory: 6319 loss: 0.1733 +2023/06/06 05:41:47 - mmengine - INFO - Epoch(train) [6][2300/4092] lr: 4.7099e-05 eta: 3:41:32 time: 0.6546 data_time: 0.0010 memory: 6319 loss: 0.1711 +2023/06/06 05:42:58 - mmengine - INFO - Epoch(train) [6][2400/4092] lr: 4.6759e-05 eta: 3:40:17 time: 0.6798 data_time: 0.0009 memory: 6319 loss: 0.1711 +2023/06/06 05:44:08 - mmengine - INFO - Epoch(train) [6][2500/4092] lr: 4.6419e-05 eta: 3:39:02 time: 0.7296 data_time: 0.0009 memory: 6319 loss: 0.1920 +2023/06/06 05:44:37 - mmengine - INFO - Exp name: resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1_20230606_005743 +2023/06/06 05:45:20 - mmengine - INFO - Epoch(train) [6][2600/4092] lr: 4.6080e-05 eta: 3:37:47 time: 0.7309 data_time: 0.0009 memory: 6319 loss: 0.1615 +2023/06/06 05:46:29 - mmengine - INFO - Epoch(train) [6][2700/4092] lr: 4.5742e-05 eta: 3:36:31 time: 0.7305 data_time: 0.0011 memory: 6319 loss: 0.1842 +2023/06/06 05:47:40 - mmengine - INFO - Epoch(train) [6][2800/4092] lr: 4.5404e-05 eta: 3:35:16 time: 0.7066 data_time: 0.0009 memory: 6319 loss: 0.1651 +2023/06/06 05:48:50 - mmengine - INFO - Epoch(train) [6][2900/4092] lr: 4.5067e-05 eta: 3:34:00 time: 0.6536 data_time: 0.0009 memory: 6319 loss: 0.1729 +2023/06/06 05:49:59 - mmengine - INFO - Epoch(train) [6][3000/4092] lr: 4.4730e-05 eta: 3:32:45 time: 0.7351 data_time: 0.0008 memory: 6319 loss: 0.1614 +2023/06/06 05:51:09 - mmengine - INFO - Epoch(train) [6][3100/4092] lr: 4.4394e-05 eta: 3:31:29 time: 0.6639 data_time: 0.0009 memory: 6319 loss: 0.1661 +2023/06/06 05:52:21 - mmengine - INFO - Epoch(train) [6][3200/4092] lr: 4.4059e-05 eta: 3:30:15 time: 0.7033 data_time: 0.0014 memory: 6319 loss: 0.1747 +2023/06/06 05:53:31 - mmengine - INFO - Epoch(train) [6][3300/4092] lr: 4.3724e-05 eta: 3:28:59 time: 0.6926 data_time: 0.0012 memory: 6319 loss: 0.1679 +2023/06/06 05:54:52 - mmengine - INFO - Epoch(train) [6][3400/4092] lr: 4.3390e-05 eta: 3:27:52 time: 0.6704 data_time: 0.0008 memory: 6319 loss: 0.1794 +2023/06/06 05:56:03 - mmengine - INFO - Epoch(train) [6][3500/4092] lr: 4.3056e-05 eta: 3:26:38 time: 0.7032 data_time: 0.1426 memory: 6319 loss: 0.2007 +2023/06/06 05:56:32 - mmengine - INFO - Exp name: resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1_20230606_005743 +2023/06/06 05:57:14 - mmengine - INFO - Epoch(train) [6][3600/4092] lr: 4.2724e-05 eta: 3:25:23 time: 0.6976 data_time: 0.1217 memory: 6319 loss: 0.1539 +2023/06/06 05:58:23 - mmengine - INFO - Epoch(train) [6][3700/4092] lr: 4.2392e-05 eta: 3:24:07 time: 0.7112 data_time: 0.2130 memory: 6319 loss: 0.1662 +2023/06/06 05:59:34 - mmengine - INFO - Epoch(train) [6][3800/4092] lr: 4.2060e-05 eta: 3:22:53 time: 0.7096 data_time: 0.2057 memory: 6319 loss: 0.1629 +2023/06/06 06:00:40 - mmengine - INFO - Epoch(train) [6][3900/4092] lr: 4.1730e-05 eta: 3:21:35 time: 0.7004 data_time: 0.3036 memory: 6319 loss: 0.1888 +2023/06/06 06:01:46 - mmengine - INFO - Epoch(train) [6][4000/4092] lr: 4.1400e-05 eta: 3:20:17 time: 0.6331 data_time: 0.2031 memory: 6319 loss: 0.1910 +2023/06/06 06:02:45 - mmengine - INFO - Exp name: resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1_20230606_005743 +2023/06/06 06:02:45 - mmengine - INFO - Saving checkpoint at 6 epochs +2023/06/06 06:03:25 - mmengine - INFO - Epoch(val) [6][100/119] eta: 0:00:06 time: 0.6358 data_time: 0.5457 memory: 6319 +2023/06/06 06:03:51 - mmengine - INFO - Epoch(val) [6][119/119] accuracy/top1: 91.2931 data_time: 0.3151 time: 0.4053 +2023/06/06 06:05:04 - mmengine - INFO - Epoch(train) [7][ 100/4092] lr: 4.0769e-05 eta: 3:17:51 time: 0.7606 data_time: 0.6124 memory: 6319 loss: 0.1614 +2023/06/06 06:06:14 - mmengine - INFO - Epoch(train) [7][ 200/4092] lr: 4.0442e-05 eta: 3:16:36 time: 0.7394 data_time: 0.5981 memory: 6319 loss: 0.1814 +2023/06/06 06:07:24 - mmengine - INFO - Epoch(train) [7][ 300/4092] lr: 4.0116e-05 eta: 3:15:21 time: 0.7033 data_time: 0.5624 memory: 6319 loss: 0.1696 +2023/06/06 06:08:36 - mmengine - INFO - Epoch(train) [7][ 400/4092] lr: 3.9790e-05 eta: 3:14:08 time: 0.6878 data_time: 0.5472 memory: 6319 loss: 0.1757 +2023/06/06 06:09:11 - mmengine - INFO - Exp name: resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1_20230606_005743 +2023/06/06 06:09:45 - mmengine - INFO - Epoch(train) [7][ 500/4092] lr: 3.9465e-05 eta: 3:12:52 time: 0.7204 data_time: 0.5762 memory: 6319 loss: 0.1739 +2023/06/06 06:10:57 - mmengine - INFO - Epoch(train) [7][ 600/4092] lr: 3.9141e-05 eta: 3:11:39 time: 0.6916 data_time: 0.2864 memory: 6319 loss: 0.1654 +2023/06/06 06:12:06 - mmengine - INFO - Epoch(train) [7][ 700/4092] lr: 3.8819e-05 eta: 3:10:23 time: 0.6537 data_time: 0.2910 memory: 6319 loss: 0.1748 +2023/06/06 06:13:17 - mmengine - INFO - Epoch(train) [7][ 800/4092] lr: 3.8497e-05 eta: 3:09:09 time: 0.7169 data_time: 0.2875 memory: 6319 loss: 0.1674 +2023/06/06 06:14:25 - mmengine - INFO - Epoch(train) [7][ 900/4092] lr: 3.8176e-05 eta: 3:07:53 time: 0.6882 data_time: 0.4835 memory: 6319 loss: 0.1732 +2023/06/06 06:15:34 - mmengine - INFO - Epoch(train) [7][1000/4092] lr: 3.7856e-05 eta: 3:06:38 time: 0.7223 data_time: 0.4904 memory: 6319 loss: 0.1564 +2023/06/06 06:16:43 - mmengine - INFO - Epoch(train) [7][1100/4092] lr: 3.7537e-05 eta: 3:05:23 time: 0.7047 data_time: 0.5655 memory: 6319 loss: 0.1568 +2023/06/06 06:17:55 - mmengine - INFO - Epoch(train) [7][1200/4092] lr: 3.7219e-05 eta: 3:04:10 time: 0.7078 data_time: 0.5667 memory: 6319 loss: 0.1883 +2023/06/06 06:19:06 - mmengine - INFO - Epoch(train) [7][1300/4092] lr: 3.6902e-05 eta: 3:02:56 time: 0.7599 data_time: 0.6204 memory: 6319 loss: 0.1799 +2023/06/06 06:20:18 - mmengine - INFO - Epoch(train) [7][1400/4092] lr: 3.6586e-05 eta: 3:01:42 time: 0.7283 data_time: 0.5881 memory: 6319 loss: 0.1736 +2023/06/06 06:20:52 - mmengine - INFO - Exp name: resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1_20230606_005743 +2023/06/06 06:21:27 - mmengine - INFO - Epoch(train) [7][1500/4092] lr: 3.6272e-05 eta: 3:00:27 time: 0.6873 data_time: 0.5465 memory: 6319 loss: 0.1626 +2023/06/06 06:22:36 - mmengine - INFO - Epoch(train) [7][1600/4092] lr: 3.5958e-05 eta: 2:59:12 time: 0.7093 data_time: 0.5690 memory: 6319 loss: 0.1856 +2023/06/06 06:23:46 - mmengine - INFO - Epoch(train) [7][1700/4092] lr: 3.5646e-05 eta: 2:57:58 time: 0.6989 data_time: 0.5577 memory: 6319 loss: 0.1672 +2023/06/06 06:24:59 - mmengine - INFO - Epoch(train) [7][1800/4092] lr: 3.5334e-05 eta: 2:56:45 time: 0.6936 data_time: 0.5526 memory: 6319 loss: 0.1687 +2023/06/06 06:26:08 - mmengine - INFO - Epoch(train) [7][1900/4092] lr: 3.5024e-05 eta: 2:55:31 time: 0.6866 data_time: 0.5436 memory: 6319 loss: 0.1585 +2023/06/06 06:27:18 - mmengine - INFO - Epoch(train) [7][2000/4092] lr: 3.4715e-05 eta: 2:54:16 time: 0.6927 data_time: 0.3834 memory: 6319 loss: 0.1630 +2023/06/06 06:28:29 - mmengine - INFO - Epoch(train) [7][2100/4092] lr: 3.4407e-05 eta: 2:53:02 time: 0.6746 data_time: 0.2664 memory: 6319 loss: 0.1672 +2023/06/06 06:29:37 - mmengine - INFO - Epoch(train) [7][2200/4092] lr: 3.4101e-05 eta: 2:51:47 time: 0.7114 data_time: 0.4306 memory: 6319 loss: 0.1680 +2023/06/06 06:30:46 - mmengine - INFO - Epoch(train) [7][2300/4092] lr: 3.3796e-05 eta: 2:50:32 time: 0.6894 data_time: 0.4674 memory: 6319 loss: 0.1741 +2023/06/06 06:31:57 - mmengine - INFO - Epoch(train) [7][2400/4092] lr: 3.3491e-05 eta: 2:49:19 time: 0.7370 data_time: 0.5961 memory: 6319 loss: 0.1940 +2023/06/06 06:32:31 - mmengine - INFO - Exp name: resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1_20230606_005743 +2023/06/06 06:33:07 - mmengine - INFO - Epoch(train) [7][2500/4092] lr: 3.3189e-05 eta: 2:48:05 time: 0.6764 data_time: 0.5340 memory: 6319 loss: 0.1768 +2023/06/06 06:34:16 - mmengine - INFO - Epoch(train) [7][2600/4092] lr: 3.2887e-05 eta: 2:46:50 time: 0.6853 data_time: 0.5447 memory: 6319 loss: 0.1754 +2023/06/06 06:35:26 - mmengine - INFO - Epoch(train) [7][2700/4092] lr: 3.2587e-05 eta: 2:45:36 time: 0.6966 data_time: 0.5490 memory: 6319 loss: 0.1641 +2023/06/06 06:36:46 - mmengine - INFO - Epoch(train) [7][2800/4092] lr: 3.2288e-05 eta: 2:44:27 time: 0.6792 data_time: 0.5382 memory: 6319 loss: 0.1656 +2023/06/06 06:37:57 - mmengine - INFO - Epoch(train) [7][2900/4092] lr: 3.1990e-05 eta: 2:43:13 time: 0.6708 data_time: 0.5310 memory: 6319 loss: 0.1735 +2023/06/06 06:39:06 - mmengine - INFO - Epoch(train) [7][3000/4092] lr: 3.1694e-05 eta: 2:41:59 time: 0.6380 data_time: 0.4976 memory: 6319 loss: 0.1528 +2023/06/06 06:40:16 - mmengine - INFO - Epoch(train) [7][3100/4092] lr: 3.1399e-05 eta: 2:40:45 time: 0.6814 data_time: 0.5413 memory: 6319 loss: 0.1775 +2023/06/06 06:41:26 - mmengine - INFO - Epoch(train) [7][3200/4092] lr: 3.1106e-05 eta: 2:39:31 time: 0.7420 data_time: 0.5962 memory: 6319 loss: 0.1535 +2023/06/06 06:42:36 - mmengine - INFO - Epoch(train) [7][3300/4092] lr: 3.0814e-05 eta: 2:38:17 time: 0.7288 data_time: 0.5891 memory: 6319 loss: 0.1650 +2023/06/06 06:43:45 - mmengine - INFO - Epoch(train) [7][3400/4092] lr: 3.0523e-05 eta: 2:37:02 time: 0.6746 data_time: 0.5338 memory: 6319 loss: 0.1656 +2023/06/06 06:44:19 - mmengine - INFO - Exp name: resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1_20230606_005743 +2023/06/06 06:44:55 - mmengine - INFO - Epoch(train) [7][3500/4092] lr: 3.0234e-05 eta: 2:35:49 time: 0.6423 data_time: 0.5015 memory: 6319 loss: 0.1782 +2023/06/06 06:46:05 - mmengine - INFO - Epoch(train) [7][3600/4092] lr: 2.9946e-05 eta: 2:34:34 time: 0.7051 data_time: 0.5658 memory: 6319 loss: 0.1733 +2023/06/06 06:47:15 - mmengine - INFO - Epoch(train) [7][3700/4092] lr: 2.9660e-05 eta: 2:33:21 time: 0.6682 data_time: 0.5270 memory: 6319 loss: 0.1612 +2023/06/06 06:48:24 - mmengine - INFO - Epoch(train) [7][3800/4092] lr: 2.9375e-05 eta: 2:32:07 time: 0.6775 data_time: 0.5378 memory: 6319 loss: 0.1648 +2023/06/06 06:49:34 - mmengine - INFO - Epoch(train) [7][3900/4092] lr: 2.9092e-05 eta: 2:30:53 time: 0.6970 data_time: 0.5563 memory: 6319 loss: 0.1648 +2023/06/06 06:50:42 - mmengine - INFO - Epoch(train) [7][4000/4092] lr: 2.8810e-05 eta: 2:29:38 time: 0.6848 data_time: 0.5439 memory: 6319 loss: 0.1697 +2023/06/06 06:51:47 - mmengine - INFO - Exp name: resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1_20230606_005743 +2023/06/06 06:51:47 - mmengine - INFO - Saving checkpoint at 7 epochs +2023/06/06 06:52:28 - mmengine - INFO - Epoch(val) [7][100/119] eta: 0:00:06 time: 0.6404 data_time: 0.5521 memory: 6319 +2023/06/06 06:52:55 - mmengine - INFO - Epoch(val) [7][119/119] accuracy/top1: 91.7349 data_time: 0.3212 time: 0.4097 +2023/06/06 06:54:07 - mmengine - INFO - Epoch(train) [8][ 100/4092] lr: 2.8274e-05 eta: 2:27:18 time: 0.6692 data_time: 0.4350 memory: 6319 loss: 0.1593 +2023/06/06 06:55:17 - mmengine - INFO - Epoch(train) [8][ 200/4092] lr: 2.7997e-05 eta: 2:26:04 time: 0.6596 data_time: 0.4416 memory: 6319 loss: 0.1724 +2023/06/06 06:56:27 - mmengine - INFO - Epoch(train) [8][ 300/4092] lr: 2.7721e-05 eta: 2:24:50 time: 0.6673 data_time: 0.4922 memory: 6319 loss: 0.1620 +2023/06/06 06:57:08 - mmengine - INFO - Exp name: resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1_20230606_005743 +2023/06/06 06:57:38 - mmengine - INFO - Epoch(train) [8][ 400/4092] lr: 2.7447e-05 eta: 2:23:37 time: 0.7153 data_time: 0.3490 memory: 6319 loss: 0.1694 +2023/06/06 06:58:47 - mmengine - INFO - Epoch(train) [8][ 500/4092] lr: 2.7175e-05 eta: 2:22:24 time: 0.7138 data_time: 0.4309 memory: 6319 loss: 0.1526 +2023/06/06 06:59:56 - mmengine - INFO - Epoch(train) [8][ 600/4092] lr: 2.6904e-05 eta: 2:21:10 time: 0.6754 data_time: 0.1599 memory: 6319 loss: 0.1712 +2023/06/06 07:01:03 - mmengine - INFO - Epoch(train) [8][ 700/4092] lr: 2.6635e-05 eta: 2:19:55 time: 0.6670 data_time: 0.0714 memory: 6319 loss: 0.1931 +2023/06/06 07:02:13 - mmengine - INFO - Epoch(train) [8][ 800/4092] lr: 2.6368e-05 eta: 2:18:41 time: 0.7235 data_time: 0.1936 memory: 6319 loss: 0.1605 +2023/06/06 07:03:24 - mmengine - INFO - Epoch(train) [8][ 900/4092] lr: 2.6102e-05 eta: 2:17:28 time: 0.7862 data_time: 0.3424 memory: 6319 loss: 0.1638 +2023/06/06 07:04:36 - mmengine - INFO - Epoch(train) [8][1000/4092] lr: 2.5838e-05 eta: 2:16:16 time: 0.7583 data_time: 0.0010 memory: 6319 loss: 0.1731 +2023/06/06 07:05:47 - mmengine - INFO - Epoch(train) [8][1100/4092] lr: 2.5576e-05 eta: 2:15:02 time: 0.7509 data_time: 0.0009 memory: 6319 loss: 0.1605 +2023/06/06 07:07:00 - mmengine - INFO - Epoch(train) [8][1200/4092] lr: 2.5315e-05 eta: 2:13:50 time: 0.7036 data_time: 0.0010 memory: 6319 loss: 0.1636 +2023/06/06 07:08:10 - mmengine - INFO - Epoch(train) [8][1300/4092] lr: 2.5056e-05 eta: 2:12:37 time: 0.6537 data_time: 0.0011 memory: 6319 loss: 0.1692 +2023/06/06 07:08:51 - mmengine - INFO - Exp name: resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1_20230606_005743 +2023/06/06 07:09:21 - mmengine - INFO - Epoch(train) [8][1400/4092] lr: 2.4799e-05 eta: 2:11:23 time: 0.7158 data_time: 0.0011 memory: 6319 loss: 0.1660 +2023/06/06 07:10:32 - mmengine - INFO - Epoch(train) [8][1500/4092] lr: 2.4544e-05 eta: 2:10:10 time: 0.6812 data_time: 0.0008 memory: 6319 loss: 0.1564 +2023/06/06 07:11:42 - mmengine - INFO - Epoch(train) [8][1600/4092] lr: 2.4291e-05 eta: 2:08:57 time: 0.7033 data_time: 0.0010 memory: 6319 loss: 0.1701 +2023/06/06 07:12:51 - mmengine - INFO - Epoch(train) [8][1700/4092] lr: 2.4039e-05 eta: 2:07:44 time: 0.7390 data_time: 0.0010 memory: 6319 loss: 0.1743 +2023/06/06 07:14:01 - mmengine - INFO - Epoch(train) [8][1800/4092] lr: 2.3789e-05 eta: 2:06:30 time: 0.6613 data_time: 0.0008 memory: 6319 loss: 0.1634 +2023/06/06 07:15:10 - mmengine - INFO - Epoch(train) [8][1900/4092] lr: 2.3541e-05 eta: 2:05:17 time: 0.7069 data_time: 0.0012 memory: 6319 loss: 0.1791 +2023/06/06 07:16:19 - mmengine - INFO - Epoch(train) [8][2000/4092] lr: 2.3295e-05 eta: 2:04:03 time: 0.7016 data_time: 0.0008 memory: 6319 loss: 0.1675 +2023/06/06 07:17:32 - mmengine - INFO - Epoch(train) [8][2100/4092] lr: 2.3051e-05 eta: 2:02:50 time: 0.7067 data_time: 0.0009 memory: 6319 loss: 0.1815 +2023/06/06 07:18:41 - mmengine - INFO - Epoch(train) [8][2200/4092] lr: 2.2809e-05 eta: 2:01:37 time: 0.7133 data_time: 0.0010 memory: 6319 loss: 0.1514 +2023/06/06 07:19:53 - mmengine - INFO - Epoch(train) [8][2300/4092] lr: 2.2568e-05 eta: 2:00:24 time: 0.6976 data_time: 0.0009 memory: 6319 loss: 0.1597 +2023/06/06 07:20:35 - mmengine - INFO - Exp name: resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1_20230606_005743 +2023/06/06 07:21:06 - mmengine - INFO - Epoch(train) [8][2400/4092] lr: 2.2330e-05 eta: 1:59:12 time: 0.9631 data_time: 0.0009 memory: 6319 loss: 0.1851 +2023/06/06 07:22:17 - mmengine - INFO - Epoch(train) [8][2500/4092] lr: 2.2093e-05 eta: 1:57:59 time: 0.7878 data_time: 0.0009 memory: 6319 loss: 0.1673 +2023/06/06 07:23:27 - mmengine - INFO - Epoch(train) [8][2600/4092] lr: 2.1858e-05 eta: 1:56:46 time: 0.7213 data_time: 0.0009 memory: 6319 loss: 0.1604 +2023/06/06 07:24:36 - mmengine - INFO - Epoch(train) [8][2700/4092] lr: 2.1626e-05 eta: 1:55:33 time: 0.6758 data_time: 0.0011 memory: 6319 loss: 0.1625 +2023/06/06 07:25:46 - mmengine - INFO - Epoch(train) [8][2800/4092] lr: 2.1395e-05 eta: 1:54:20 time: 0.7037 data_time: 0.0010 memory: 6319 loss: 0.1778 +2023/06/06 07:26:59 - mmengine - INFO - Epoch(train) [8][2900/4092] lr: 2.1166e-05 eta: 1:53:07 time: 0.6738 data_time: 0.0010 memory: 6319 loss: 0.1790 +2023/06/06 07:28:08 - mmengine - INFO - Epoch(train) [8][3000/4092] lr: 2.0939e-05 eta: 1:51:54 time: 0.6913 data_time: 0.0009 memory: 6319 loss: 0.1610 +2023/06/06 07:29:19 - mmengine - INFO - Epoch(train) [8][3100/4092] lr: 2.0715e-05 eta: 1:50:41 time: 0.7478 data_time: 0.0007 memory: 6319 loss: 0.1729 +2023/06/06 07:30:29 - mmengine - INFO - Epoch(train) [8][3200/4092] lr: 2.0492e-05 eta: 1:49:28 time: 0.7185 data_time: 0.0010 memory: 6319 loss: 0.1998 +2023/06/06 07:31:40 - mmengine - INFO - Epoch(train) [8][3300/4092] lr: 2.0271e-05 eta: 1:48:15 time: 0.7237 data_time: 0.0009 memory: 6319 loss: 0.1761 +2023/06/06 07:32:22 - mmengine - INFO - Exp name: resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1_20230606_005743 +2023/06/06 07:32:50 - mmengine - INFO - Epoch(train) [8][3400/4092] lr: 2.0052e-05 eta: 1:47:03 time: 0.7074 data_time: 0.0008 memory: 6319 loss: 0.1756 +2023/06/06 07:34:02 - mmengine - INFO - Epoch(train) [8][3500/4092] lr: 1.9836e-05 eta: 1:45:50 time: 0.7198 data_time: 0.0008 memory: 6319 loss: 0.1502 +2023/06/06 07:35:11 - mmengine - INFO - Epoch(train) [8][3600/4092] lr: 1.9621e-05 eta: 1:44:37 time: 0.7024 data_time: 0.0009 memory: 6319 loss: 0.1591 +2023/06/06 07:36:22 - mmengine - INFO - Epoch(train) [8][3700/4092] lr: 1.9409e-05 eta: 1:43:24 time: 0.7456 data_time: 0.0009 memory: 6319 loss: 0.1811 +2023/06/06 07:37:35 - mmengine - INFO - Epoch(train) [8][3800/4092] lr: 1.9198e-05 eta: 1:42:12 time: 1.0399 data_time: 0.0009 memory: 6319 loss: 0.1697 +2023/06/06 07:38:45 - mmengine - INFO - Epoch(train) [8][3900/4092] lr: 1.8990e-05 eta: 1:40:59 time: 0.7251 data_time: 0.0009 memory: 6319 loss: 0.1785 +2023/06/06 07:39:54 - mmengine - INFO - Epoch(train) [8][4000/4092] lr: 1.8784e-05 eta: 1:39:46 time: 0.7142 data_time: 0.0008 memory: 6319 loss: 0.1625 +2023/06/06 07:40:57 - mmengine - INFO - Exp name: resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1_20230606_005743 +2023/06/06 07:40:57 - mmengine - INFO - Saving checkpoint at 8 epochs +2023/06/06 07:41:37 - mmengine - INFO - Epoch(val) [8][100/119] eta: 0:00:06 time: 0.6134 data_time: 0.5254 memory: 6319 +2023/06/06 07:42:03 - mmengine - INFO - Epoch(val) [8][119/119] accuracy/top1: 92.9067 data_time: 0.3212 time: 0.4082 +2023/06/06 07:43:14 - mmengine - INFO - Epoch(train) [9][ 100/4092] lr: 1.8394e-05 eta: 1:37:26 time: 0.7207 data_time: 0.1873 memory: 6319 loss: 0.1546 +2023/06/06 07:44:25 - mmengine - INFO - Epoch(train) [9][ 200/4092] lr: 1.8194e-05 eta: 1:36:13 time: 0.6816 data_time: 0.1899 memory: 6319 loss: 0.1688 +2023/06/06 07:45:15 - mmengine - INFO - Exp name: resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1_20230606_005743 +2023/06/06 07:45:37 - mmengine - INFO - Epoch(train) [9][ 300/4092] lr: 1.7997e-05 eta: 1:35:00 time: 0.6806 data_time: 0.3172 memory: 6319 loss: 0.1572 +2023/06/06 07:46:47 - mmengine - INFO - Epoch(train) [9][ 400/4092] lr: 1.7801e-05 eta: 1:33:48 time: 0.7137 data_time: 0.3877 memory: 6319 loss: 0.1750 +2023/06/06 07:47:57 - mmengine - INFO - Epoch(train) [9][ 500/4092] lr: 1.7608e-05 eta: 1:32:35 time: 0.6825 data_time: 0.2948 memory: 6319 loss: 0.1449 +2023/06/06 07:49:07 - mmengine - INFO - Epoch(train) [9][ 600/4092] lr: 1.7417e-05 eta: 1:31:22 time: 0.7056 data_time: 0.1645 memory: 6319 loss: 0.1794 +2023/06/06 07:50:16 - mmengine - INFO - Epoch(train) [9][ 700/4092] lr: 1.7228e-05 eta: 1:30:09 time: 0.7026 data_time: 0.0008 memory: 6319 loss: 0.1795 +2023/06/06 07:51:27 - mmengine - INFO - Epoch(train) [9][ 800/4092] lr: 1.7041e-05 eta: 1:28:57 time: 0.7037 data_time: 0.0009 memory: 6319 loss: 0.1639 +2023/06/06 07:52:38 - mmengine - INFO - Epoch(train) [9][ 900/4092] lr: 1.6857e-05 eta: 1:27:44 time: 0.7312 data_time: 0.2464 memory: 6319 loss: 0.1575 +2023/06/06 07:53:48 - mmengine - INFO - Epoch(train) [9][1000/4092] lr: 1.6675e-05 eta: 1:26:31 time: 0.6530 data_time: 0.1508 memory: 6319 loss: 0.1495 +2023/06/06 07:54:58 - mmengine - INFO - Epoch(train) [9][1100/4092] lr: 1.6495e-05 eta: 1:25:18 time: 0.6659 data_time: 0.2675 memory: 6319 loss: 0.1704 +2023/06/06 07:56:09 - mmengine - INFO - Epoch(train) [9][1200/4092] lr: 1.6317e-05 eta: 1:24:06 time: 0.7438 data_time: 0.2079 memory: 6319 loss: 0.1612 +2023/06/06 07:56:57 - mmengine - INFO - Exp name: resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1_20230606_005743 +2023/06/06 07:57:20 - mmengine - INFO - Epoch(train) [9][1300/4092] lr: 1.6142e-05 eta: 1:22:53 time: 0.7149 data_time: 0.0339 memory: 6319 loss: 0.1829 +2023/06/06 07:58:32 - mmengine - INFO - Epoch(train) [9][1400/4092] lr: 1.5969e-05 eta: 1:21:41 time: 0.7060 data_time: 0.0010 memory: 6319 loss: 0.1559 +2023/06/06 07:59:42 - mmengine - INFO - Epoch(train) [9][1500/4092] lr: 1.5798e-05 eta: 1:20:28 time: 0.6935 data_time: 0.0008 memory: 6319 loss: 0.1688 +2023/06/06 08:00:52 - mmengine - INFO - Epoch(train) [9][1600/4092] lr: 1.5629e-05 eta: 1:19:16 time: 0.6837 data_time: 0.0010 memory: 6319 loss: 0.1697 +2023/06/06 08:02:04 - mmengine - INFO - Epoch(train) [9][1700/4092] lr: 1.5463e-05 eta: 1:18:03 time: 0.7375 data_time: 0.0009 memory: 6319 loss: 0.1715 +2023/06/06 08:03:15 - mmengine - INFO - Epoch(train) [9][1800/4092] lr: 1.5299e-05 eta: 1:16:51 time: 0.7000 data_time: 0.0011 memory: 6319 loss: 0.1785 +2023/06/06 08:04:25 - mmengine - INFO - Epoch(train) [9][1900/4092] lr: 1.5138e-05 eta: 1:15:38 time: 0.7424 data_time: 0.0008 memory: 6319 loss: 0.1638 +2023/06/06 08:05:35 - mmengine - INFO - Epoch(train) [9][2000/4092] lr: 1.4979e-05 eta: 1:14:26 time: 0.6368 data_time: 0.0007 memory: 6319 loss: 0.1598 +2023/06/06 08:06:45 - mmengine - INFO - Epoch(train) [9][2100/4092] lr: 1.4822e-05 eta: 1:13:13 time: 0.6890 data_time: 0.0009 memory: 6319 loss: 0.1589 +2023/06/06 08:07:54 - mmengine - INFO - Epoch(train) [9][2200/4092] lr: 1.4668e-05 eta: 1:12:00 time: 0.7143 data_time: 0.0009 memory: 6319 loss: 0.1602 +2023/06/06 08:08:42 - mmengine - INFO - Exp name: resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1_20230606_005743 +2023/06/06 08:09:03 - mmengine - INFO - Epoch(train) [9][2300/4092] lr: 1.4515e-05 eta: 1:10:48 time: 0.7422 data_time: 0.0009 memory: 6319 loss: 0.1585 +2023/06/06 08:10:12 - mmengine - INFO - Epoch(train) [9][2400/4092] lr: 1.4366e-05 eta: 1:09:35 time: 0.7104 data_time: 0.0008 memory: 6319 loss: 0.1651 +2023/06/06 08:11:23 - mmengine - INFO - Epoch(train) [9][2500/4092] lr: 1.4219e-05 eta: 1:08:22 time: 0.7485 data_time: 0.0009 memory: 6319 loss: 0.1632 +2023/06/06 08:12:34 - mmengine - INFO - Epoch(train) [9][2600/4092] lr: 1.4074e-05 eta: 1:07:10 time: 0.6742 data_time: 0.0010 memory: 6319 loss: 0.1598 +2023/06/06 08:13:44 - mmengine - INFO - Epoch(train) [9][2700/4092] lr: 1.3931e-05 eta: 1:05:58 time: 0.6640 data_time: 0.0009 memory: 6319 loss: 0.1691 +2023/06/06 08:14:53 - mmengine - INFO - Epoch(train) [9][2800/4092] lr: 1.3791e-05 eta: 1:04:45 time: 0.6555 data_time: 0.0011 memory: 6319 loss: 0.1560 +2023/06/06 08:16:04 - mmengine - INFO - Epoch(train) [9][2900/4092] lr: 1.3654e-05 eta: 1:03:33 time: 0.6874 data_time: 0.0011 memory: 6319 loss: 0.1809 +2023/06/06 08:17:14 - mmengine - INFO - Epoch(train) [9][3000/4092] lr: 1.3519e-05 eta: 1:02:20 time: 0.7585 data_time: 0.0009 memory: 6319 loss: 0.1703 +2023/06/06 08:18:23 - mmengine - INFO - Epoch(train) [9][3100/4092] lr: 1.3386e-05 eta: 1:01:08 time: 0.7467 data_time: 0.2073 memory: 6319 loss: 0.1638 +2023/06/06 08:19:35 - mmengine - INFO - Epoch(train) [9][3200/4092] lr: 1.3256e-05 eta: 0:59:55 time: 0.7195 data_time: 0.1893 memory: 6319 loss: 0.1659 +2023/06/06 08:20:22 - mmengine - INFO - Exp name: resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1_20230606_005743 +2023/06/06 08:20:48 - mmengine - INFO - Epoch(train) [9][3300/4092] lr: 1.3128e-05 eta: 0:58:43 time: 0.7629 data_time: 0.0010 memory: 6319 loss: 0.1620 +2023/06/06 08:21:57 - mmengine - INFO - Epoch(train) [9][3400/4092] lr: 1.3003e-05 eta: 0:57:31 time: 0.7198 data_time: 0.0009 memory: 6319 loss: 0.1666 +2023/06/06 08:23:08 - mmengine - INFO - Epoch(train) [9][3500/4092] lr: 1.2880e-05 eta: 0:56:18 time: 0.7428 data_time: 0.0009 memory: 6319 loss: 0.1598 +2023/06/06 08:24:19 - mmengine - INFO - Epoch(train) [9][3600/4092] lr: 1.2759e-05 eta: 0:55:06 time: 0.7897 data_time: 0.0009 memory: 6319 loss: 0.1594 +2023/06/06 08:25:30 - mmengine - INFO - Epoch(train) [9][3700/4092] lr: 1.2641e-05 eta: 0:53:54 time: 0.6903 data_time: 0.0008 memory: 6319 loss: 0.1692 +2023/06/06 08:26:40 - mmengine - INFO - Epoch(train) [9][3800/4092] lr: 1.2526e-05 eta: 0:52:41 time: 0.6901 data_time: 0.0008 memory: 6319 loss: 0.1662 +2023/06/06 08:27:49 - mmengine - INFO - Epoch(train) [9][3900/4092] lr: 1.2413e-05 eta: 0:51:29 time: 0.6892 data_time: 0.0009 memory: 6319 loss: 0.1717 +2023/06/06 08:29:02 - mmengine - INFO - Epoch(train) [9][4000/4092] lr: 1.2303e-05 eta: 0:50:17 time: 0.7204 data_time: 0.0008 memory: 6319 loss: 0.1668 +2023/06/06 08:30:05 - mmengine - INFO - Exp name: resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1_20230606_005743 +2023/06/06 08:30:05 - mmengine - INFO - Saving checkpoint at 9 epochs +2023/06/06 08:30:45 - mmengine - INFO - Epoch(val) [9][100/119] eta: 0:00:06 time: 0.6766 data_time: 0.5872 memory: 6319 +2023/06/06 08:31:10 - mmengine - INFO - Epoch(val) [9][119/119] accuracy/top1: 93.0159 data_time: 0.3134 time: 0.4029 +2023/06/06 08:32:23 - mmengine - INFO - Epoch(train) [10][ 100/4092] lr: 1.2098e-05 eta: 0:47:58 time: 0.6929 data_time: 0.0987 memory: 6319 loss: 0.1676 +2023/06/06 08:33:14 - mmengine - INFO - Exp name: resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1_20230606_005743 +2023/06/06 08:33:33 - mmengine - INFO - Epoch(train) [10][ 200/4092] lr: 1.1995e-05 eta: 0:46:46 time: 0.7161 data_time: 0.2022 memory: 6319 loss: 0.1678 +2023/06/06 08:34:42 - mmengine - INFO - Epoch(train) [10][ 300/4092] lr: 1.1895e-05 eta: 0:45:33 time: 0.7055 data_time: 0.0009 memory: 6319 loss: 0.1426 +2023/06/06 08:35:54 - mmengine - INFO - Epoch(train) [10][ 400/4092] lr: 1.1797e-05 eta: 0:44:21 time: 0.7239 data_time: 0.0009 memory: 6319 loss: 0.1644 +2023/06/06 08:37:04 - mmengine - INFO - Epoch(train) [10][ 500/4092] lr: 1.1701e-05 eta: 0:43:09 time: 0.6705 data_time: 0.0009 memory: 6319 loss: 0.1632 +2023/06/06 08:38:15 - mmengine - INFO - Epoch(train) [10][ 600/4092] lr: 1.1608e-05 eta: 0:41:57 time: 0.6627 data_time: 0.0010 memory: 6319 loss: 0.1848 +2023/06/06 08:39:31 - mmengine - INFO - Epoch(train) [10][ 700/4092] lr: 1.1518e-05 eta: 0:40:45 time: 0.6897 data_time: 0.0009 memory: 6319 loss: 0.1533 +2023/06/06 08:40:41 - mmengine - INFO - Epoch(train) [10][ 800/4092] lr: 1.1430e-05 eta: 0:39:33 time: 0.7070 data_time: 0.0009 memory: 6319 loss: 0.1482 +2023/06/06 08:41:50 - mmengine - INFO - Epoch(train) [10][ 900/4092] lr: 1.1345e-05 eta: 0:38:20 time: 0.7723 data_time: 0.0010 memory: 6319 loss: 0.1497 +2023/06/06 08:42:59 - mmengine - INFO - Epoch(train) [10][1000/4092] lr: 1.1263e-05 eta: 0:37:08 time: 0.6892 data_time: 0.0009 memory: 6319 loss: 0.1525 +2023/06/06 08:44:09 - mmengine - INFO - Epoch(train) [10][1100/4092] lr: 1.1183e-05 eta: 0:35:56 time: 0.6578 data_time: 0.0009 memory: 6319 loss: 0.1548 +2023/06/06 08:44:58 - mmengine - INFO - Exp name: resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1_20230606_005743 +2023/06/06 08:45:19 - mmengine - INFO - Epoch(train) [10][1200/4092] lr: 1.1105e-05 eta: 0:34:44 time: 0.6491 data_time: 0.0010 memory: 6319 loss: 0.1501 +2023/06/06 08:46:29 - mmengine - INFO - Epoch(train) [10][1300/4092] lr: 1.1031e-05 eta: 0:33:31 time: 0.7208 data_time: 0.0009 memory: 6319 loss: 0.1507 +2023/06/06 08:47:42 - mmengine - INFO - Epoch(train) [10][1400/4092] lr: 1.0958e-05 eta: 0:32:19 time: 0.6935 data_time: 0.0011 memory: 6319 loss: 0.1692 +2023/06/06 08:48:52 - mmengine - INFO - Epoch(train) [10][1500/4092] lr: 1.0889e-05 eta: 0:31:07 time: 0.6928 data_time: 0.0010 memory: 6319 loss: 0.1632 +2023/06/06 08:50:04 - mmengine - INFO - Epoch(train) [10][1600/4092] lr: 1.0822e-05 eta: 0:29:55 time: 0.7008 data_time: 0.0010 memory: 6319 loss: 0.1710 +2023/06/06 08:51:15 - mmengine - INFO - Epoch(train) [10][1700/4092] lr: 1.0757e-05 eta: 0:28:43 time: 0.7213 data_time: 0.0010 memory: 6319 loss: 0.1795 +2023/06/06 08:52:25 - mmengine - INFO - Epoch(train) [10][1800/4092] lr: 1.0696e-05 eta: 0:27:31 time: 0.7137 data_time: 0.0009 memory: 6319 loss: 0.1598 +2023/06/06 08:53:37 - mmengine - INFO - Epoch(train) [10][1900/4092] lr: 1.0636e-05 eta: 0:26:19 time: 0.7228 data_time: 0.0010 memory: 6319 loss: 0.1737 +2023/06/06 08:54:49 - mmengine - INFO - Epoch(train) [10][2000/4092] lr: 1.0580e-05 eta: 0:25:07 time: 0.6470 data_time: 0.0009 memory: 6319 loss: 0.1519 +2023/06/06 08:55:58 - mmengine - INFO - Epoch(train) [10][2100/4092] lr: 1.0526e-05 eta: 0:23:55 time: 0.7224 data_time: 0.0009 memory: 6319 loss: 0.1799 +2023/06/06 08:56:49 - mmengine - INFO - Exp name: resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1_20230606_005743 +2023/06/06 08:57:11 - mmengine - INFO - Epoch(train) [10][2200/4092] lr: 1.0474e-05 eta: 0:22:43 time: 0.7347 data_time: 0.0009 memory: 6319 loss: 0.1714 +2023/06/06 08:58:23 - mmengine - INFO - Epoch(train) [10][2300/4092] lr: 1.0426e-05 eta: 0:21:31 time: 0.6977 data_time: 0.0010 memory: 6319 loss: 0.1523 +2023/06/06 08:59:36 - mmengine - INFO - Epoch(train) [10][2400/4092] lr: 1.0380e-05 eta: 0:20:19 time: 0.7016 data_time: 0.0009 memory: 6319 loss: 0.1691 +2023/06/06 09:00:45 - mmengine - INFO - Epoch(train) [10][2500/4092] lr: 1.0336e-05 eta: 0:19:06 time: 0.6834 data_time: 0.0009 memory: 6319 loss: 0.1661 +2023/06/06 09:01:54 - mmengine - INFO - Epoch(train) [10][2600/4092] lr: 1.0295e-05 eta: 0:17:54 time: 0.6715 data_time: 0.0008 memory: 6319 loss: 0.1630 +2023/06/06 09:03:06 - mmengine - INFO - Epoch(train) [10][2700/4092] lr: 1.0257e-05 eta: 0:16:42 time: 0.6889 data_time: 0.0008 memory: 6319 loss: 0.1664 +2023/06/06 09:04:17 - mmengine - INFO - Epoch(train) [10][2800/4092] lr: 1.0222e-05 eta: 0:15:30 time: 0.7031 data_time: 0.0009 memory: 6319 loss: 0.1494 +2023/06/06 09:05:28 - mmengine - INFO - Epoch(train) [10][2900/4092] lr: 1.0189e-05 eta: 0:14:18 time: 0.7960 data_time: 0.0008 memory: 6319 loss: 0.1693 +2023/06/06 09:06:37 - mmengine - INFO - Epoch(train) [10][3000/4092] lr: 1.0158e-05 eta: 0:13:06 time: 0.6756 data_time: 0.0008 memory: 6319 loss: 0.1615 +2023/06/06 09:07:47 - mmengine - INFO - Epoch(train) [10][3100/4092] lr: 1.0131e-05 eta: 0:11:54 time: 0.6999 data_time: 0.0010 memory: 6319 loss: 0.1593 +2023/06/06 09:08:36 - mmengine - INFO - Exp name: resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1_20230606_005743 +2023/06/06 09:08:56 - mmengine - INFO - Epoch(train) [10][3200/4092] lr: 1.0106e-05 eta: 0:10:42 time: 0.7180 data_time: 0.0009 memory: 6319 loss: 0.1709 +2023/06/06 09:10:09 - mmengine - INFO - Epoch(train) [10][3300/4092] lr: 1.0083e-05 eta: 0:09:30 time: 0.7259 data_time: 0.0009 memory: 6319 loss: 0.1506 +2023/06/06 09:11:18 - mmengine - INFO - Epoch(train) [10][3400/4092] lr: 1.0064e-05 eta: 0:08:18 time: 0.6662 data_time: 0.0012 memory: 6319 loss: 0.1570 +2023/06/06 09:12:31 - mmengine - INFO - Epoch(train) [10][3500/4092] lr: 1.0047e-05 eta: 0:07:06 time: 0.6967 data_time: 0.0012 memory: 6319 loss: 0.1607 +2023/06/06 09:13:44 - mmengine - INFO - Epoch(train) [10][3600/4092] lr: 1.0032e-05 eta: 0:05:54 time: 0.9085 data_time: 0.0008 memory: 6319 loss: 0.1496 +2023/06/06 09:14:55 - mmengine - INFO - Epoch(train) [10][3700/4092] lr: 1.0020e-05 eta: 0:04:42 time: 0.7271 data_time: 0.0010 memory: 6319 loss: 0.1523 +2023/06/06 09:16:09 - mmengine - INFO - Epoch(train) [10][3800/4092] lr: 1.0011e-05 eta: 0:03:30 time: 0.7483 data_time: 0.0010 memory: 6319 loss: 0.1580 +2023/06/06 09:17:19 - mmengine - INFO - Epoch(train) [10][3900/4092] lr: 1.0005e-05 eta: 0:02:18 time: 0.7160 data_time: 0.0009 memory: 6319 loss: 0.1690 +2023/06/06 09:18:28 - mmengine - INFO - Epoch(train) [10][4000/4092] lr: 1.0001e-05 eta: 0:01:06 time: 0.6836 data_time: 0.0009 memory: 6319 loss: 0.1600 +2023/06/06 09:19:36 - mmengine - INFO - Exp name: resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1_20230606_005743 +2023/06/06 09:19:36 - mmengine - INFO - Saving checkpoint at 10 epochs +2023/06/06 09:20:17 - mmengine - INFO - Epoch(val) [10][100/119] eta: 0:00:06 time: 0.7365 data_time: 0.6372 memory: 6319 +2023/06/06 09:20:43 - mmengine - INFO - Epoch(val) [10][119/119] accuracy/top1: 92.8636 data_time: 0.3246 time: 0.4125 diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/20230606_005743.json b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/20230606_005743.json new file mode 100644 index 0000000000000000000000000000000000000000..c44a590b61948449a56eed401d7d3473b7008063 --- /dev/null +++ b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/20230606_005743.json @@ -0,0 +1,410 @@ +{"lr": 9.999870019168206e-05, "data_time": 0.23580620288848878, "loss": 0.6417255282402039, "time": 0.8090263366699219, "epoch": 1, "memory": 9436, "step": 100} +{"lr": 9.99947481934738e-05, "data_time": 0.21867220401763915, "loss": 0.5970972716808319, "time": 0.7791444778442382, "epoch": 1, "memory": 6319, "step": 200} +{"lr": 9.99881441056942e-05, "data_time": 0.0008538007736206054, "loss": 0.5583989799022675, "time": 0.7499716997146606, "epoch": 1, "memory": 6319, "step": 300} +{"lr": 9.997888831760252e-05, "data_time": 0.0008380651473999023, "loss": 0.5301752865314484, "time": 0.7361132621765136, "epoch": 1, "memory": 6319, "step": 400} +{"lr": 9.996698137475549e-05, "data_time": 0.2010805606842041, "loss": 0.5015071302652359, "time": 0.7582030534744263, "epoch": 1, "memory": 6319, "step": 500} +{"lr": 9.99524239789746e-05, "data_time": 0.3320797920227051, "loss": 0.46877075731754303, "time": 0.7183988094329834, "epoch": 1, "memory": 6319, "step": 600} +{"lr": 9.993521698830466e-05, "data_time": 0.02013735771179199, "loss": 0.43822180926799775, "time": 0.7378515005111694, "epoch": 1, "memory": 6319, "step": 700} +{"lr": 9.991536141696371e-05, "data_time": 0.0007852315902709961, "loss": 0.4319634079933167, "time": 0.7372121334075927, "epoch": 1, "memory": 6319, "step": 800} +{"lr": 9.989285843528274e-05, "data_time": 0.11607332229614258, "loss": 0.42011411786079406, "time": 0.7631598949432373, "epoch": 1, "memory": 6319, "step": 900} +{"lr": 9.986770936963713e-05, "data_time": 0.0008153676986694336, "loss": 0.4003512740135193, "time": 0.7913449287414551, "epoch": 1, "memory": 6319, "step": 1000} +{"lr": 9.983991570236818e-05, "data_time": 0.0007603883743286133, "loss": 0.39497354030609133, "time": 0.8244112253189086, "epoch": 1, "memory": 6319, "step": 1100} +{"lr": 9.980947907169598e-05, "data_time": 0.0007162094116210938, "loss": 0.3863217025995255, "time": 0.722666573524475, "epoch": 1, "memory": 6319, "step": 1200} +{"lr": 9.977640127162247e-05, "data_time": 0.0008350849151611328, "loss": 0.37821408808231355, "time": 0.7241732358932496, "epoch": 1, "memory": 6319, "step": 1300} +{"lr": 9.974068425182603e-05, "data_time": 0.0007997035980224609, "loss": 0.37111017405986785, "time": 0.7567805051803589, "epoch": 1, "memory": 6319, "step": 1400} +{"lr": 9.970233011754657e-05, "data_time": 0.0007455587387084961, "loss": 0.3742601066827774, "time": 0.7378098011016846, "epoch": 1, "memory": 6319, "step": 1500} +{"lr": 9.966134112946122e-05, "data_time": 0.0008747339248657227, "loss": 0.3582422524690628, "time": 0.6978653907775879, "epoch": 1, "memory": 6319, "step": 1600} +{"lr": 9.961771970355126e-05, "data_time": 0.0010759592056274413, "loss": 0.3595370054244995, "time": 0.7230527639389038, "epoch": 1, "memory": 6319, "step": 1700} +{"lr": 9.957146841095961e-05, "data_time": 0.0007981300354003906, "loss": 0.3502417117357254, "time": 0.7151001453399658, "epoch": 1, "memory": 6319, "step": 1800} +{"lr": 9.952258997783933e-05, "data_time": 0.0007390499114990235, "loss": 0.34335593283176424, "time": 0.7786037206649781, "epoch": 1, "memory": 6319, "step": 1900} +{"lr": 9.947108728519288e-05, "data_time": 0.0007749319076538086, "loss": 0.33032276928424836, "time": 0.7474444389343262, "epoch": 1, "memory": 6319, "step": 2000} +{"lr": 9.941696336870245e-05, "data_time": 0.0009086132049560547, "loss": 0.33920717239379883, "time": 0.7526807308197021, "epoch": 1, "memory": 6319, "step": 2100} +{"lr": 9.936022141855094e-05, "data_time": 0.0007847785949707031, "loss": 0.35441104471683504, "time": 0.7442209720611572, "epoch": 1, "memory": 6319, "step": 2200} +{"lr": 9.93008647792338e-05, "data_time": 0.000978565216064453, "loss": 0.3193003237247467, "time": 0.7475589513778687, "epoch": 1, "memory": 6319, "step": 2300} +{"lr": 9.923889694936212e-05, "data_time": 0.001101851463317871, "loss": 0.3393714040517807, "time": 0.7450142621994018, "epoch": 1, "memory": 6319, "step": 2400} +{"lr": 9.917432158145639e-05, "data_time": 0.0007400751113891601, "loss": 0.3217262476682663, "time": 0.7152126550674438, "epoch": 1, "memory": 6319, "step": 2500} +{"lr": 9.91071424817309e-05, "data_time": 0.0009122371673583984, "loss": 0.32098470330238343, "time": 0.6925920009613037, "epoch": 1, "memory": 6319, "step": 2600} +{"lr": 9.903736360986979e-05, "data_time": 0.0007892131805419921, "loss": 0.3037663847208023, "time": 0.6956831455230713, "epoch": 1, "memory": 6319, "step": 2700} +{"lr": 9.89649890787935e-05, "data_time": 0.0008501768112182617, "loss": 0.317654749751091, "time": 0.7175958633422852, "epoch": 1, "memory": 6319, "step": 2800} +{"lr": 9.889002315441642e-05, "data_time": 0.0008883237838745117, "loss": 0.3155059784650803, "time": 0.7144453048706054, "epoch": 1, "memory": 6319, "step": 2900} +{"lr": 9.881247025539515e-05, "data_time": 0.0007520437240600586, "loss": 0.3076498419046402, "time": 0.6948991060256958, "epoch": 1, "memory": 6319, "step": 3000} +{"lr": 9.873233495286844e-05, "data_time": 0.0007988214492797852, "loss": 0.30978504568338394, "time": 0.7028228044509888, "epoch": 1, "memory": 6319, "step": 3100} +{"lr": 9.864962197018756e-05, "data_time": 0.0007736444473266602, "loss": 0.3064905434846878, "time": 0.7453366994857789, "epoch": 1, "memory": 6319, "step": 3200} +{"lr": 9.856433618263779e-05, "data_time": 0.0007336616516113281, "loss": 0.29859074354171755, "time": 0.6710611343383789, "epoch": 1, "memory": 6319, "step": 3300} +{"lr": 9.847648261715132e-05, "data_time": 0.0009308576583862305, "loss": 0.2963121831417084, "time": 0.7274848937988281, "epoch": 1, "memory": 6319, "step": 3400} +{"lr": 9.838606645201093e-05, "data_time": 0.0008805036544799804, "loss": 0.29638334810733796, "time": 0.6771203517913819, "epoch": 1, "memory": 6319, "step": 3500} +{"lr": 9.829309301654456e-05, "data_time": 0.0007841110229492188, "loss": 0.30206618905067445, "time": 0.7793800830841064, "epoch": 1, "memory": 6319, "step": 3600} +{"lr": 9.819756779081127e-05, "data_time": 0.0007678985595703125, "loss": 0.30126443058252333, "time": 0.8266619920730591, "epoch": 1, "memory": 6319, "step": 3700} +{"lr": 9.809949640527814e-05, "data_time": 0.000860285758972168, "loss": 0.3012932062149048, "time": 0.7224830627441406, "epoch": 1, "memory": 6319, "step": 3800} +{"lr": 9.799888464048879e-05, "data_time": 0.0008294820785522461, "loss": 0.2915215402841568, "time": 0.7069743633270263, "epoch": 1, "memory": 6319, "step": 3900} +{"lr": 9.789573842672242e-05, "data_time": 0.0009275674819946289, "loss": 0.3215578556060791, "time": 0.7584826707839966, "epoch": 1, "memory": 6319, "step": 4000} +{"accuracy/top1": 83.01309204101562, "data_time": 0.36436522708219643, "time": 0.4523949502896862, "step": 1} +{"lr": 9.7690615523361e-05, "data_time": 0.39775900840759276, "loss": 0.306628492474556, "time": 0.6732662677764892, "epoch": 2, "memory": 6319, "step": 4192} +{"lr": 9.758010405927656e-05, "data_time": 0.15111191272735597, "loss": 0.2822544798254967, "time": 0.683975625038147, "epoch": 2, "memory": 6319, "step": 4292} +{"lr": 9.746708283006492e-05, "data_time": 0.002385663986206055, "loss": 0.2794072195887566, "time": 0.7211886405944824, "epoch": 2, "memory": 6319, "step": 4392} +{"lr": 9.735155849744634e-05, "data_time": 0.0008119344711303711, "loss": 0.2702935427427292, "time": 0.7382085084915161, "epoch": 2, "memory": 6319, "step": 4492} +{"lr": 9.723353787067928e-05, "data_time": 0.0007747411727905273, "loss": 0.26802886426448824, "time": 0.6471272230148315, "epoch": 2, "memory": 6319, "step": 4592} +{"lr": 9.711302790615951e-05, "data_time": 0.0008364439010620118, "loss": 0.2711342617869377, "time": 0.8386395931243896, "epoch": 2, "memory": 6319, "step": 4692} +{"lr": 9.699003570700986e-05, "data_time": 0.0009061574935913086, "loss": 0.27828952819108965, "time": 0.8523646354675293, "epoch": 2, "memory": 6319, "step": 4792} +{"lr": 9.686456852266142e-05, "data_time": 0.0009198904037475586, "loss": 0.2655982568860054, "time": 0.6747840881347656, "epoch": 2, "memory": 6319, "step": 4892} +{"lr": 9.673663374842644e-05, "data_time": 0.0010221004486083984, "loss": 0.25839774757623674, "time": 0.6460691213607788, "epoch": 2, "memory": 6319, "step": 4992} +{"lr": 9.660623892506218e-05, "data_time": 0.0012143611907958984, "loss": 0.2699123755097389, "time": 0.7331794500350952, "epoch": 2, "memory": 6319, "step": 5092} +{"lr": 9.647339173832693e-05, "data_time": 0.0008815288543701172, "loss": 0.26809143722057344, "time": 0.7267658948898316, "epoch": 2, "memory": 6319, "step": 5192} +{"lr": 9.633810001852653e-05, "data_time": 0.000978231430053711, "loss": 0.2733738929033279, "time": 0.7184287071228027, "epoch": 2, "memory": 6319, "step": 5292} +{"lr": 9.620037174005274e-05, "data_time": 0.0007425546646118164, "loss": 0.27412050366401675, "time": 0.7382806062698364, "epoch": 2, "memory": 6319, "step": 5392} +{"lr": 9.606021502091389e-05, "data_time": 0.0006863832473754883, "loss": 0.2644442543387413, "time": 0.7033350706100464, "epoch": 2, "memory": 6319, "step": 5492} +{"lr": 9.591763812225569e-05, "data_time": 0.00098726749420166, "loss": 0.2552147641777992, "time": 0.6655296802520752, "epoch": 2, "memory": 6319, "step": 5592} +{"lr": 9.577264944787459e-05, "data_time": 0.0009675979614257813, "loss": 0.26296195983886717, "time": 0.6554296016693115, "epoch": 2, "memory": 6319, "step": 5692} +{"lr": 9.562525754372252e-05, "data_time": 0.0009081125259399414, "loss": 0.26218927949666976, "time": 0.7197994709014892, "epoch": 2, "memory": 6319, "step": 5792} +{"lr": 9.547547109740282e-05, "data_time": 0.000800633430480957, "loss": 0.26032392531633375, "time": 0.6760532855987549, "epoch": 2, "memory": 6319, "step": 5892} +{"lr": 9.532329893765858e-05, "data_time": 0.0009043455123901367, "loss": 0.27663236260414126, "time": 0.9657118082046509, "epoch": 2, "memory": 6319, "step": 5992} +{"lr": 9.516875003385223e-05, "data_time": 0.0008503675460815429, "loss": 0.24357614368200303, "time": 0.7123346090316772, "epoch": 2, "memory": 6319, "step": 6092} +{"lr": 9.501183349543653e-05, "data_time": 0.0007915019989013672, "loss": 0.27699775248765945, "time": 0.676742434501648, "epoch": 2, "memory": 6319, "step": 6192} +{"lr": 9.48525585714179e-05, "data_time": 0.0010963678359985352, "loss": 0.2426074355840683, "time": 0.6890865564346313, "epoch": 2, "memory": 6319, "step": 6292} +{"lr": 9.469093464981119e-05, "data_time": 0.0007992982864379883, "loss": 0.2461544841527939, "time": 0.7354315996170044, "epoch": 2, "memory": 6319, "step": 6392} +{"lr": 9.452697125708636e-05, "data_time": 0.000832056999206543, "loss": 0.25532765686511993, "time": 0.6789705991744995, "epoch": 2, "memory": 6319, "step": 6492} +{"lr": 9.436067805760705e-05, "data_time": 0.0008382558822631836, "loss": 0.2425134912133217, "time": 0.6513427019119262, "epoch": 2, "memory": 6319, "step": 6592} +{"lr": 9.419206485306072e-05, "data_time": 0.0008160352706909179, "loss": 0.2688654959201813, "time": 0.6749155998229981, "epoch": 2, "memory": 6319, "step": 6692} +{"lr": 9.4021141581881e-05, "data_time": 0.0007049083709716797, "loss": 0.24094131886959075, "time": 0.6558093547821044, "epoch": 2, "memory": 6319, "step": 6792} +{"lr": 9.384791831866193e-05, "data_time": 0.0009281158447265625, "loss": 0.2769829869270325, "time": 0.7018612146377563, "epoch": 2, "memory": 6319, "step": 6892} +{"lr": 9.367240527356428e-05, "data_time": 0.0007861375808715821, "loss": 0.23097819536924363, "time": 0.7395848512649537, "epoch": 2, "memory": 6319, "step": 6992} +{"lr": 9.349461279171362e-05, "data_time": 0.0011840343475341796, "loss": 0.2553142011165619, "time": 0.6742855787277222, "epoch": 2, "memory": 6319, "step": 7092} +{"lr": 9.33145513525904e-05, "data_time": 0.0009016990661621094, "loss": 0.24693509191274643, "time": 0.6630861043930054, "epoch": 2, "memory": 6319, "step": 7192} +{"lr": 9.313223156941242e-05, "data_time": 0.001598024368286133, "loss": 0.2428608626127243, "time": 0.7029174327850342, "epoch": 2, "memory": 6319, "step": 7292} +{"lr": 9.294766418850942e-05, "data_time": 0.0007671117782592773, "loss": 0.2545601323246956, "time": 0.699622368812561, "epoch": 2, "memory": 6319, "step": 7392} +{"lr": 9.276086008868929e-05, "data_time": 0.0007414340972900391, "loss": 0.2544305339455605, "time": 0.6884544610977172, "epoch": 2, "memory": 6319, "step": 7492} +{"lr": 9.257183028059726e-05, "data_time": 0.0009234428405761718, "loss": 0.25289510786533353, "time": 0.7154263496398926, "epoch": 2, "memory": 6319, "step": 7592} +{"lr": 9.238058590606654e-05, "data_time": 0.0007828950881958007, "loss": 0.24429425746202468, "time": 0.7094822406768799, "epoch": 2, "memory": 6319, "step": 7692} +{"lr": 9.218713823746189e-05, "data_time": 0.0011223077774047852, "loss": 0.23091641813516617, "time": 0.7586504697799683, "epoch": 2, "memory": 6319, "step": 7792} +{"lr": 9.199149867701492e-05, "data_time": 0.0011123180389404296, "loss": 0.24989114403724672, "time": 0.6876688003540039, "epoch": 2, "memory": 6319, "step": 7892} +{"lr": 9.179367875615232e-05, "data_time": 0.0007442951202392579, "loss": 0.23347796648740768, "time": 0.6866568803787232, "epoch": 2, "memory": 6319, "step": 7992} +{"lr": 9.159369013481585e-05, "data_time": 0.0007095098495483399, "loss": 0.23617833852767944, "time": 0.711391544342041, "epoch": 2, "memory": 6319, "step": 8092} +{"accuracy/top1": 81.01551055908203, "data_time": 0.3427091976006826, "time": 0.42977015376091005, "step": 2} +{"lr": 9.120367596460475e-05, "data_time": 0.4689095258712769, "loss": 0.23746268898248674, "time": 0.6739122152328492, "epoch": 3, "memory": 6319, "step": 8284} +{"lr": 9.099742266672751e-05, "data_time": 0.33391757011413575, "loss": 0.2527444750070572, "time": 0.700292706489563, "epoch": 3, "memory": 6319, "step": 8384} +{"lr": 9.078904760146445e-05, "data_time": 0.2901160478591919, "loss": 0.23466432243585586, "time": 0.668321967124939, "epoch": 3, "memory": 6319, "step": 8484} +{"lr": 9.057856305090082e-05, "data_time": 0.3167825937271118, "loss": 0.24158217310905455, "time": 0.6561912059783935, "epoch": 3, "memory": 6319, "step": 8584} +{"lr": 9.036598142145979e-05, "data_time": 0.5717406272888184, "loss": 0.21645256280899047, "time": 0.7558250188827514, "epoch": 3, "memory": 6319, "step": 8684} +{"lr": 9.015131524317079e-05, "data_time": 0.537515926361084, "loss": 0.22299962341785431, "time": 0.6771813154220581, "epoch": 3, "memory": 6319, "step": 8784} +{"lr": 8.993457716893124e-05, "data_time": 0.4909390449523926, "loss": 0.2311646595597267, "time": 0.6904112815856933, "epoch": 3, "memory": 6319, "step": 8884} +{"lr": 8.971577997376067e-05, "data_time": 0.5511940240859985, "loss": 0.22824979573488235, "time": 0.6908522605895996, "epoch": 3, "memory": 6319, "step": 8984} +{"lr": 8.949493655404777e-05, "data_time": 0.5729543924331665, "loss": 0.22518849074840547, "time": 0.7130033493041992, "epoch": 3, "memory": 6319, "step": 9084} +{"lr": 8.927205992679013e-05, "data_time": 0.560827088356018, "loss": 0.2453401893377304, "time": 0.7018682479858398, "epoch": 3, "memory": 6319, "step": 9184} +{"lr": 8.9047163228827e-05, "data_time": 0.5887284994125366, "loss": 0.22647609263658525, "time": 0.7276627063751221, "epoch": 3, "memory": 6319, "step": 9284} +{"lr": 8.88202597160652e-05, "data_time": 0.5279091358184814, "loss": 0.23519654124975203, "time": 0.6715782403945922, "epoch": 3, "memory": 6319, "step": 9384} +{"lr": 8.859136276269775e-05, "data_time": 0.5629609584808349, "loss": 0.23769256919622422, "time": 0.7038479328155518, "epoch": 3, "memory": 6319, "step": 9484} +{"lr": 8.836048586041507e-05, "data_time": 0.5362492084503174, "loss": 0.2327328860759735, "time": 0.676891565322876, "epoch": 3, "memory": 6319, "step": 9584} +{"lr": 8.812764261761037e-05, "data_time": 0.561246395111084, "loss": 0.22969091087579727, "time": 0.7121063232421875, "epoch": 3, "memory": 6319, "step": 9684} +{"lr": 8.789284675857743e-05, "data_time": 0.5059587478637695, "loss": 0.23402496427297592, "time": 0.6462924480438232, "epoch": 3, "memory": 6319, "step": 9784} +{"lr": 8.76561121227013e-05, "data_time": 0.5290754556655883, "loss": 0.23261655420064925, "time": 0.6685675859451294, "epoch": 3, "memory": 6319, "step": 9884} +{"lr": 8.741745266364282e-05, "data_time": 0.5213753700256347, "loss": 0.231000679731369, "time": 0.6652604818344117, "epoch": 3, "memory": 6319, "step": 9984} +{"lr": 8.717688244851635e-05, "data_time": 0.4483925819396973, "loss": 0.24258920401334763, "time": 0.7430192470550537, "epoch": 3, "memory": 6319, "step": 10084} +{"lr": 8.693441565706024e-05, "data_time": 0.3338231801986694, "loss": 0.21221618801355363, "time": 0.6699972867965698, "epoch": 3, "memory": 6319, "step": 10184} +{"lr": 8.669006658080132e-05, "data_time": 0.14651296138763428, "loss": 0.21599060893058777, "time": 0.7080436229705811, "epoch": 3, "memory": 6319, "step": 10284} +{"lr": 8.644384962221226e-05, "data_time": 0.06523208618164063, "loss": 0.23294903635978698, "time": 0.6683426618576049, "epoch": 3, "memory": 6319, "step": 10384} +{"lr": 8.619577929386303e-05, "data_time": 0.16167752742767333, "loss": 0.23386239856481553, "time": 0.7149275064468383, "epoch": 3, "memory": 6319, "step": 10484} +{"lr": 8.594587021756517e-05, "data_time": 0.16182565689086914, "loss": 0.21927913427352905, "time": 0.7035971879959106, "epoch": 3, "memory": 6319, "step": 10584} +{"lr": 8.569413712351023e-05, "data_time": 0.0009621858596801757, "loss": 0.22514743208885193, "time": 0.6861767292022705, "epoch": 3, "memory": 6319, "step": 10684} +{"lr": 8.544059484940119e-05, "data_time": 0.0008236169815063477, "loss": 0.20948270410299302, "time": 0.7535413503646851, "epoch": 3, "memory": 6319, "step": 10784} +{"lr": 8.518525833957818e-05, "data_time": 0.0008296966552734375, "loss": 0.2109537348151207, "time": 0.7048580408096313, "epoch": 3, "memory": 6319, "step": 10884} +{"lr": 8.49281426441377e-05, "data_time": 0.0009437799453735352, "loss": 0.21660107374191284, "time": 0.7745984554290771, "epoch": 3, "memory": 6319, "step": 10984} +{"lr": 8.466926291804501e-05, "data_time": 0.0007861852645874023, "loss": 0.21598206758499144, "time": 0.7172518014907837, "epoch": 3, "memory": 6319, "step": 11084} +{"lr": 8.440863442024147e-05, "data_time": 0.000759124755859375, "loss": 0.2235696107149124, "time": 0.6983484745025634, "epoch": 3, "memory": 6319, "step": 11184} +{"lr": 8.414627251274488e-05, "data_time": 0.000874471664428711, "loss": 0.20743883401155472, "time": 0.738426685333252, "epoch": 3, "memory": 6319, "step": 11284} +{"lr": 8.388219265974407e-05, "data_time": 0.0008832931518554688, "loss": 0.2073444053530693, "time": 0.7331681489944458, "epoch": 3, "memory": 6319, "step": 11384} +{"lr": 8.361641042668715e-05, "data_time": 0.0008800029754638672, "loss": 0.20792239159345627, "time": 0.7300339937210083, "epoch": 3, "memory": 6319, "step": 11484} +{"lr": 8.33489414793644e-05, "data_time": 0.0009262561798095703, "loss": 0.19522866904735564, "time": 0.8190644979476929, "epoch": 3, "memory": 6319, "step": 11584} +{"lr": 8.30798015829848e-05, "data_time": 0.0009125709533691406, "loss": 0.21183450520038605, "time": 0.7052595138549804, "epoch": 3, "memory": 6319, "step": 11684} +{"lr": 8.280900660124665e-05, "data_time": 0.0007672309875488281, "loss": 0.2278122529387474, "time": 1.241529941558838, "epoch": 3, "memory": 6319, "step": 11784} +{"lr": 8.253657249540274e-05, "data_time": 0.0008351564407348633, "loss": 0.21307228058576583, "time": 0.6984730958938599, "epoch": 3, "memory": 6319, "step": 11884} +{"lr": 8.226251532331926e-05, "data_time": 0.000736403465270996, "loss": 0.22321422547101974, "time": 0.7290765523910523, "epoch": 3, "memory": 6319, "step": 11984} +{"lr": 8.198685123852965e-05, "data_time": 0.0008404731750488281, "loss": 0.20979103744029998, "time": 0.738861083984375, "epoch": 3, "memory": 6319, "step": 12084} +{"lr": 8.17095964892824e-05, "data_time": 0.0007280588150024414, "loss": 0.2234665036201477, "time": 0.7249564170837403, "epoch": 3, "memory": 6319, "step": 12184} +{"accuracy/top1": 82.50170135498047, "data_time": 0.3352497100830078, "time": 0.42358460823694866, "step": 3} +{"lr": 8.11728683565904e-05, "data_time": 0.3452092170715332, "loss": 0.2123613715171814, "time": 0.7652813911437988, "epoch": 4, "memory": 6319, "step": 12376} +{"lr": 8.089106273476776e-05, "data_time": 0.16974735260009766, "loss": 0.2119458019733429, "time": 0.712868070602417, "epoch": 4, "memory": 6319, "step": 12476} +{"lr": 8.060773103667566e-05, "data_time": 0.2002340078353882, "loss": 0.20567522943019867, "time": 0.6693098306655884, "epoch": 4, "memory": 6319, "step": 12576} +{"lr": 8.03228899625083e-05, "data_time": 0.0007893562316894532, "loss": 0.21832259595394135, "time": 0.7679970741271973, "epoch": 4, "memory": 6319, "step": 12676} +{"lr": 8.003655630142627e-05, "data_time": 0.0007112979888916016, "loss": 0.20089872777462006, "time": 0.7451218128204345, "epoch": 4, "memory": 6319, "step": 12776} +{"lr": 7.974874693056617e-05, "data_time": 0.013694691658020019, "loss": 0.20858096480369567, "time": 0.7786278486251831, "epoch": 4, "memory": 6319, "step": 12876} +{"lr": 7.945947881404618e-05, "data_time": 0.00084991455078125, "loss": 0.2153647467494011, "time": 0.7001819372177124, "epoch": 4, "memory": 6319, "step": 12976} +{"lr": 7.916876900196646e-05, "data_time": 0.0008081674575805664, "loss": 0.20742841958999633, "time": 0.7613331794738769, "epoch": 4, "memory": 6319, "step": 13076} +{"lr": 7.887663462940359e-05, "data_time": 0.0009556055068969726, "loss": 0.2295478105545044, "time": 0.7719241619110108, "epoch": 4, "memory": 6319, "step": 13176} +{"lr": 7.858309291540111e-05, "data_time": 0.0007396697998046875, "loss": 0.21536858528852462, "time": 0.6818386554718018, "epoch": 4, "memory": 6319, "step": 13276} +{"lr": 7.828816116195411e-05, "data_time": 0.0007997512817382813, "loss": 0.20525012463331221, "time": 0.7383459091186524, "epoch": 4, "memory": 6319, "step": 13376} +{"lr": 7.799185675298994e-05, "data_time": 0.000770401954650879, "loss": 0.20376997888088227, "time": 0.747047758102417, "epoch": 4, "memory": 6319, "step": 13476} +{"lr": 7.769419715334297e-05, "data_time": 0.0008090019226074218, "loss": 0.19351460486650468, "time": 0.7959878444671631, "epoch": 4, "memory": 6319, "step": 13576} +{"lr": 7.739519990772583e-05, "data_time": 0.0008514881134033203, "loss": 0.1868076041340828, "time": 0.7113405942916871, "epoch": 4, "memory": 6319, "step": 13676} +{"lr": 7.709488263969475e-05, "data_time": 0.0007879972457885742, "loss": 0.20743744373321532, "time": 0.717255973815918, "epoch": 4, "memory": 6319, "step": 13776} +{"lr": 7.67932630506109e-05, "data_time": 0.0008163213729858399, "loss": 0.20504215657711028, "time": 0.7047784566879273, "epoch": 4, "memory": 6319, "step": 13876} +{"lr": 7.64903589185975e-05, "data_time": 0.0010724544525146484, "loss": 0.21003948599100114, "time": 0.7431734800338745, "epoch": 4, "memory": 6319, "step": 13976} +{"lr": 7.618618809749134e-05, "data_time": 0.0007904291152954102, "loss": 0.20357482731342316, "time": 0.7548447132110596, "epoch": 4, "memory": 6319, "step": 14076} +{"lr": 7.588076851579028e-05, "data_time": 0.0009006023406982422, "loss": 0.22520921230316163, "time": 0.7574375152587891, "epoch": 4, "memory": 6319, "step": 14176} +{"lr": 7.557411817559755e-05, "data_time": 0.0009590625762939453, "loss": 0.1938050150871277, "time": 0.6884126663208008, "epoch": 4, "memory": 6319, "step": 14276} +{"lr": 7.526625515155957e-05, "data_time": 0.0008490800857543946, "loss": 0.19368332177400588, "time": 0.7812108516693115, "epoch": 4, "memory": 6319, "step": 14376} +{"lr": 7.495719758980122e-05, "data_time": 0.0007518291473388672, "loss": 0.21273910254240036, "time": 0.6806473970413208, "epoch": 4, "memory": 6319, "step": 14476} +{"lr": 7.464696370685586e-05, "data_time": 0.0007924318313598632, "loss": 0.20084236264228822, "time": 0.7339329242706298, "epoch": 4, "memory": 6319, "step": 14576} +{"lr": 7.433557178859198e-05, "data_time": 0.0008692502975463867, "loss": 0.20700835138559343, "time": 0.819854998588562, "epoch": 4, "memory": 6319, "step": 14676} +{"lr": 7.402304018913512e-05, "data_time": 0.0007573366165161133, "loss": 0.20201543271541594, "time": 0.6464768886566162, "epoch": 4, "memory": 6319, "step": 14776} +{"lr": 7.37093873297861e-05, "data_time": 0.000745391845703125, "loss": 0.19180578589439393, "time": 0.6652537822723389, "epoch": 4, "memory": 6319, "step": 14876} +{"lr": 7.339463169793563e-05, "data_time": 0.0008716106414794922, "loss": 0.2002347320318222, "time": 0.7854808330535888, "epoch": 4, "memory": 6319, "step": 14976} +{"lr": 7.307879184597369e-05, "data_time": 0.0008253097534179687, "loss": 0.19215106219053268, "time": 0.7002780437469482, "epoch": 4, "memory": 6319, "step": 15076} +{"lr": 7.276188639019704e-05, "data_time": 0.0011428594589233398, "loss": 0.1828625664114952, "time": 0.6477735996246338, "epoch": 4, "memory": 6319, "step": 15176} +{"lr": 7.244393400971127e-05, "data_time": 0.000862884521484375, "loss": 0.2041605144739151, "time": 0.6513532400131226, "epoch": 4, "memory": 6319, "step": 15276} +{"lr": 7.212495344533006e-05, "data_time": 0.000845789909362793, "loss": 0.20874846130609512, "time": 0.6859585523605347, "epoch": 4, "memory": 6319, "step": 15376} +{"lr": 7.180496349847055e-05, "data_time": 0.0008579492568969727, "loss": 0.19593610018491744, "time": 0.6912116527557373, "epoch": 4, "memory": 6319, "step": 15476} +{"lr": 7.148398303004499e-05, "data_time": 0.0008430242538452149, "loss": 0.19316211193799973, "time": 0.7713447332382202, "epoch": 4, "memory": 6319, "step": 15576} +{"lr": 7.11620309593493e-05, "data_time": 0.00077972412109375, "loss": 0.2021595597267151, "time": 0.6499639034271241, "epoch": 4, "memory": 6319, "step": 15676} +{"lr": 7.083912626294795e-05, "data_time": 0.0008289575576782227, "loss": 0.20438037663698197, "time": 0.6702100753784179, "epoch": 4, "memory": 6319, "step": 15776} +{"lr": 7.051528797355494e-05, "data_time": 0.0008092164993286133, "loss": 0.1811400607228279, "time": 0.7106074810028076, "epoch": 4, "memory": 6319, "step": 15876} +{"lr": 7.019053517891214e-05, "data_time": 0.0007821321487426758, "loss": 0.18542954176664353, "time": 0.6673015594482422, "epoch": 4, "memory": 6319, "step": 15976} +{"lr": 6.986488702066473e-05, "data_time": 0.0007546663284301758, "loss": 0.19507016092538834, "time": 0.6872706174850464, "epoch": 4, "memory": 6319, "step": 16076} +{"lr": 6.953836269323245e-05, "data_time": 0.0008318901062011718, "loss": 0.18438905328512192, "time": 0.6796255588531495, "epoch": 4, "memory": 6319, "step": 16176} +{"lr": 6.921098144267838e-05, "data_time": 0.0008318424224853516, "loss": 0.19302212595939636, "time": 0.7152482986450195, "epoch": 4, "memory": 6319, "step": 16276} +{"accuracy/top1": 87.85231018066406, "data_time": 0.3359352926413218, "time": 0.4237587094306946, "step": 4} +{"lr": 6.858007803673889e-05, "data_time": 0.4231771469116211, "loss": 0.18794462084770203, "time": 0.7526944875717163, "epoch": 5, "memory": 6319, "step": 16468} +{"lr": 6.825030518869046e-05, "data_time": 0.08548085689544678, "loss": 0.18475014269351958, "time": 0.7094476461410523, "epoch": 5, "memory": 6319, "step": 16568} +{"lr": 6.791975133846686e-05, "data_time": 0.0008550167083740234, "loss": 0.18087070882320405, "time": 0.7259903907775879, "epoch": 5, "memory": 6319, "step": 16668} +{"lr": 6.758843596964013e-05, "data_time": 0.0008980512619018554, "loss": 0.18168546110391617, "time": 0.7913861989974975, "epoch": 5, "memory": 6319, "step": 16768} +{"lr": 6.725637861066804e-05, "data_time": 0.0010109424591064453, "loss": 0.19905896335840226, "time": 0.7366887092590332, "epoch": 5, "memory": 6319, "step": 16868} +{"lr": 6.692359883374264e-05, "data_time": 0.0007719278335571289, "loss": 0.17777239829301833, "time": 0.7934297800064087, "epoch": 5, "memory": 6319, "step": 16968} +{"lr": 6.659011625363718e-05, "data_time": 0.0008303165435791015, "loss": 0.18943231999874116, "time": 0.7456103086471557, "epoch": 5, "memory": 6319, "step": 17068} +{"lr": 6.625595052654927e-05, "data_time": 0.0008040904998779297, "loss": 0.19153810441493987, "time": 0.7497786760330201, "epoch": 5, "memory": 6319, "step": 17168} +{"lr": 6.592112134894315e-05, "data_time": 0.0007994651794433593, "loss": 0.19877799451351166, "time": 0.7292736053466797, "epoch": 5, "memory": 6319, "step": 17268} +{"lr": 6.558564845638802e-05, "data_time": 0.008930206298828125, "loss": 0.19411127716302873, "time": 0.7052309274673462, "epoch": 5, "memory": 6319, "step": 17368} +{"lr": 6.524955162239518e-05, "data_time": 0.30195157527923583, "loss": 0.2047702983021736, "time": 0.7157135963439941, "epoch": 5, "memory": 6319, "step": 17468} +{"lr": 6.491285065725238e-05, "data_time": 0.17519099712371827, "loss": 0.18956020325422288, "time": 0.7212751388549805, "epoch": 5, "memory": 6319, "step": 17568} +{"lr": 6.457556540685618e-05, "data_time": 0.10123610496520996, "loss": 0.1927041381597519, "time": 1.1040440559387208, "epoch": 5, "memory": 6319, "step": 17668} +{"lr": 6.423771575154208e-05, "data_time": 0.000852203369140625, "loss": 0.2042357802391052, "time": 0.7295528888702393, "epoch": 5, "memory": 6319, "step": 17768} +{"lr": 6.389932160491309e-05, "data_time": 0.166975998878479, "loss": 0.19824224710464478, "time": 0.7837016582489014, "epoch": 5, "memory": 6319, "step": 17868} +{"lr": 6.356040291266556e-05, "data_time": 0.35793533325195315, "loss": 0.18350497186183928, "time": 0.780746603012085, "epoch": 5, "memory": 6319, "step": 17968} +{"lr": 6.322097965141391e-05, "data_time": 0.22860174179077147, "loss": 0.19847580641508103, "time": 0.7191197633743286, "epoch": 5, "memory": 6319, "step": 18068} +{"lr": 6.288107182751267e-05, "data_time": 0.21714768409729004, "loss": 0.1943136215209961, "time": 0.7471305131912231, "epoch": 5, "memory": 6319, "step": 18168} +{"lr": 6.254069947587782e-05, "data_time": 0.30128490924835205, "loss": 0.18034227788448334, "time": 0.6765732526779175, "epoch": 5, "memory": 6319, "step": 18268} +{"lr": 6.219988265880537e-05, "data_time": 0.0007728815078735351, "loss": 0.18240047991275787, "time": 0.8017641067504883, "epoch": 5, "memory": 6319, "step": 18368} +{"lr": 6.185864146478941e-05, "data_time": 0.0009111881256103516, "loss": 0.1861817166209221, "time": 0.7041740894317627, "epoch": 5, "memory": 6319, "step": 18468} +{"lr": 6.151699600733752e-05, "data_time": 0.0009175777435302734, "loss": 0.18908145129680634, "time": 0.731346845626831, "epoch": 5, "memory": 6319, "step": 18568} +{"lr": 6.117496642378554e-05, "data_time": 0.00077667236328125, "loss": 0.20048560649156572, "time": 0.6893887281417846, "epoch": 5, "memory": 6319, "step": 18668} +{"lr": 6.0832572874110315e-05, "data_time": 0.0008406162261962891, "loss": 0.18763316720724105, "time": 0.7661089420318603, "epoch": 5, "memory": 6319, "step": 18768} +{"lr": 6.0489835539741924e-05, "data_time": 0.0009041786193847656, "loss": 0.17761170342564583, "time": 0.7509448289871216, "epoch": 5, "memory": 6319, "step": 18868} +{"lr": 6.0146774622374006e-05, "data_time": 0.0008766889572143554, "loss": 0.1656336508691311, "time": 0.7639059066772461, "epoch": 5, "memory": 6319, "step": 18968} +{"lr": 5.980341034277274e-05, "data_time": 0.000999140739440918, "loss": 0.18799278885126114, "time": 1.1507911920547484, "epoch": 5, "memory": 6319, "step": 19068} +{"lr": 5.945976293958505e-05, "data_time": 0.0009579658508300781, "loss": 0.18799471259117126, "time": 0.6969701290130615, "epoch": 5, "memory": 6319, "step": 19168} +{"lr": 5.911585266814587e-05, "data_time": 0.0009127140045166015, "loss": 0.18035973310470582, "time": 0.6836655616760254, "epoch": 5, "memory": 6319, "step": 19268} +{"lr": 5.877169979928426e-05, "data_time": 0.0008069038391113281, "loss": 0.16871726214885713, "time": 0.7440181016921997, "epoch": 5, "memory": 6319, "step": 19368} +{"lr": 5.842732461812828e-05, "data_time": 0.0010048627853393554, "loss": 0.17651473730802536, "time": 0.6986016988754272, "epoch": 5, "memory": 6319, "step": 19468} +{"lr": 5.8082747422909646e-05, "data_time": 0.0008463859558105469, "loss": 0.17984080761671067, "time": 0.7480592727661133, "epoch": 5, "memory": 6319, "step": 19568} +{"lr": 5.7737988523767406e-05, "data_time": 0.000822901725769043, "loss": 0.1844061106443405, "time": 0.7214935779571533, "epoch": 5, "memory": 6319, "step": 19668} +{"lr": 5.739306824155042e-05, "data_time": 0.0007722377777099609, "loss": 0.17757899910211564, "time": 0.7331432342529297, "epoch": 5, "memory": 6319, "step": 19768} +{"lr": 5.704800690661993e-05, "data_time": 0.0008005380630493164, "loss": 0.1896573856472969, "time": 0.7413455009460449, "epoch": 5, "memory": 6319, "step": 19868} +{"lr": 5.6702824857650994e-05, "data_time": 0.0008304595947265625, "loss": 0.19663527756929397, "time": 0.7371852397918701, "epoch": 5, "memory": 6319, "step": 19968} +{"lr": 5.635754244043411e-05, "data_time": 0.000824594497680664, "loss": 0.17274728864431382, "time": 0.7180647373199462, "epoch": 5, "memory": 6319, "step": 20068} +{"lr": 5.60121800066754e-05, "data_time": 0.0007601022720336914, "loss": 0.186477167904377, "time": 0.7617677450180054, "epoch": 5, "memory": 6319, "step": 20168} +{"lr": 5.5666757912797516e-05, "data_time": 0.0008510351181030273, "loss": 0.18061977177858352, "time": 0.7368639707565308, "epoch": 5, "memory": 6319, "step": 20268} +{"lr": 5.532129651873955e-05, "data_time": 0.0008313417434692383, "loss": 0.17435094714164734, "time": 0.7056990385055542, "epoch": 5, "memory": 6319, "step": 20368} +{"accuracy/top1": 90.04187774658203, "data_time": 0.33237088123957315, "time": 0.41941052079200747, "step": 5} +{"lr": 5.465797506068693e-05, "data_time": 0.46198620796203616, "loss": 0.19311447143554689, "time": 0.7909291505813598, "epoch": 6, "memory": 6319, "step": 20560} +{"lr": 5.431251545044144e-05, "data_time": 0.29471142292022706, "loss": 0.1744712382555008, "time": 0.7965962886810303, "epoch": 6, "memory": 6319, "step": 20660} +{"lr": 5.396709636204968e-05, "data_time": 0.006396985054016114, "loss": 0.17412697225809098, "time": 0.7236768484115601, "epoch": 6, "memory": 6319, "step": 20760} +{"lr": 5.3621738155273566e-05, "data_time": 0.0007611274719238281, "loss": 0.1695677876472473, "time": 0.7043702840805054, "epoch": 6, "memory": 6319, "step": 20860} +{"lr": 5.3276461186286584e-05, "data_time": 0.0008725643157958985, "loss": 0.18179067671298982, "time": 0.7208564043045044, "epoch": 6, "memory": 6319, "step": 20960} +{"lr": 5.293128580647384e-05, "data_time": 0.0009331226348876954, "loss": 0.18003389686346055, "time": 0.6939608097076416, "epoch": 6, "memory": 6319, "step": 21060} +{"lr": 5.258623236123256e-05, "data_time": 0.001152968406677246, "loss": 0.17001144737005233, "time": 0.7269580364227295, "epoch": 6, "memory": 6319, "step": 21160} +{"lr": 5.224132118877281e-05, "data_time": 0.0008795738220214843, "loss": 0.1735054597258568, "time": 0.7369820356369019, "epoch": 6, "memory": 6319, "step": 21260} +{"lr": 5.189657261891887e-05, "data_time": 0.0008654594421386719, "loss": 0.18412974625825881, "time": 0.7385596990585327, "epoch": 6, "memory": 6319, "step": 21360} +{"lr": 5.1552006971910974e-05, "data_time": 0.000860285758972168, "loss": 0.16710021048784257, "time": 0.9760957717895508, "epoch": 6, "memory": 6319, "step": 21460} +{"lr": 5.120764455720728e-05, "data_time": 0.0009122371673583984, "loss": 0.18533318638801574, "time": 0.7022658824920655, "epoch": 6, "memory": 6319, "step": 21560} +{"lr": 5.0863505672287034e-05, "data_time": 0.000771188735961914, "loss": 0.1834379807114601, "time": 0.7259957075119019, "epoch": 6, "memory": 6319, "step": 21660} +{"lr": 5.051961060145421e-05, "data_time": 0.0008374452590942383, "loss": 0.17431159615516661, "time": 0.6929921865463257, "epoch": 6, "memory": 6319, "step": 21760} +{"lr": 5.017597961464177e-05, "data_time": 0.000815272331237793, "loss": 0.18364601582288742, "time": 0.6506641387939454, "epoch": 6, "memory": 6319, "step": 21860} +{"lr": 4.983263296621707e-05, "data_time": 0.0009006023406982422, "loss": 0.17091528028249742, "time": 0.7906309366226196, "epoch": 6, "memory": 6319, "step": 21960} +{"lr": 4.948959089378785e-05, "data_time": 0.0009644746780395508, "loss": 0.15937474891543388, "time": 0.6765654802322387, "epoch": 6, "memory": 6319, "step": 22060} +{"lr": 4.91468736170094e-05, "data_time": 0.0008697509765625, "loss": 0.19052128940820695, "time": 0.7719676494598389, "epoch": 6, "memory": 6319, "step": 22160} +{"lr": 4.880450133639314e-05, "data_time": 0.0009163141250610352, "loss": 0.1802923262119293, "time": 0.7350878715515137, "epoch": 6, "memory": 6319, "step": 22260} +{"lr": 4.8462494232115335e-05, "data_time": 0.0009006500244140625, "loss": 0.18618869334459304, "time": 0.6649278163909912, "epoch": 6, "memory": 6319, "step": 22360} +{"lr": 4.812087246282806e-05, "data_time": 0.0008510828018188476, "loss": 0.16936117336153983, "time": 0.7321164846420288, "epoch": 6, "memory": 6319, "step": 22460} +{"lr": 4.7779656164470954e-05, "data_time": 0.0007488727569580078, "loss": 0.17743580937385559, "time": 0.6986096143722534, "epoch": 6, "memory": 6319, "step": 22560} +{"lr": 4.7438865449084246e-05, "data_time": 0.0007977962493896484, "loss": 0.17333071529865265, "time": 0.6732075214385986, "epoch": 6, "memory": 6319, "step": 22660} +{"lr": 4.709852040362338e-05, "data_time": 0.0010052919387817383, "loss": 0.1711229793727398, "time": 0.6546087265014648, "epoch": 6, "memory": 6319, "step": 22760} +{"lr": 4.675864108877499e-05, "data_time": 0.0009387969970703125, "loss": 0.17113893181085588, "time": 0.6797942399978638, "epoch": 6, "memory": 6319, "step": 22860} +{"lr": 4.641924753777464e-05, "data_time": 0.0009348154067993164, "loss": 0.19200259745121, "time": 0.7295567035675049, "epoch": 6, "memory": 6319, "step": 22960} +{"lr": 4.608035975522578e-05, "data_time": 0.0009101390838623047, "loss": 0.16151309609413148, "time": 0.7308872699737549, "epoch": 6, "memory": 6319, "step": 23060} +{"lr": 4.574199771592085e-05, "data_time": 0.0011149406433105468, "loss": 0.18423908054828644, "time": 0.7304690837860107, "epoch": 6, "memory": 6319, "step": 23160} +{"lr": 4.540418136366372e-05, "data_time": 0.0008737087249755859, "loss": 0.16510149985551834, "time": 0.7066318035125733, "epoch": 6, "memory": 6319, "step": 23260} +{"lr": 4.506693061009437e-05, "data_time": 0.0009009361267089844, "loss": 0.17288915663957596, "time": 0.6536128520965576, "epoch": 6, "memory": 6319, "step": 23360} +{"lr": 4.473026533351511e-05, "data_time": 0.0008421897888183594, "loss": 0.16139207482337953, "time": 0.7351397514343262, "epoch": 6, "memory": 6319, "step": 23460} +{"lr": 4.439420537771888e-05, "data_time": 0.0009276628494262696, "loss": 0.16605090945959092, "time": 0.6639213800430298, "epoch": 6, "memory": 6319, "step": 23560} +{"lr": 4.405877055081978e-05, "data_time": 0.0014353513717651368, "loss": 0.1747357040643692, "time": 0.7033367872238159, "epoch": 6, "memory": 6319, "step": 23660} +{"lr": 4.3723980624085373e-05, "data_time": 0.0012249469757080079, "loss": 0.16790425330400466, "time": 0.6926178693771362, "epoch": 6, "memory": 6319, "step": 23760} +{"lr": 4.3389855330771394e-05, "data_time": 0.0008136272430419922, "loss": 0.17936118245124816, "time": 0.6703760147094726, "epoch": 6, "memory": 6319, "step": 23860} +{"lr": 4.305641436495869e-05, "data_time": 0.1425553560256958, "loss": 0.20069527477025986, "time": 0.7031758308410645, "epoch": 6, "memory": 6319, "step": 23960} +{"lr": 4.272367738039222e-05, "data_time": 0.12170403003692627, "loss": 0.1538866363465786, "time": 0.6975866794586182, "epoch": 6, "memory": 6319, "step": 24060} +{"lr": 4.239166398932278e-05, "data_time": 0.2130486249923706, "loss": 0.16621210426092148, "time": 0.7111610889434814, "epoch": 6, "memory": 6319, "step": 24160} +{"lr": 4.206039376135108e-05, "data_time": 0.2057497262954712, "loss": 0.16290929019451142, "time": 0.7096094846725464, "epoch": 6, "memory": 6319, "step": 24260} +{"lr": 4.1729886222273985e-05, "data_time": 0.3035950899124146, "loss": 0.1887648656964302, "time": 0.7003968715667724, "epoch": 6, "memory": 6319, "step": 24360} +{"lr": 4.140016085293389e-05, "data_time": 0.20308949947357177, "loss": 0.19095413088798524, "time": 0.6330536127090454, "epoch": 6, "memory": 6319, "step": 24460} +{"accuracy/top1": 91.29305267333984, "data_time": 0.31512669920921327, "time": 0.4052566925684611, "step": 6} +{"lr": 4.0769351868715556e-05, "data_time": 0.6124338865280151, "loss": 0.16141433119773865, "time": 0.7606054782867432, "epoch": 7, "memory": 6319, "step": 24652} +{"lr": 4.044202148988293e-05, "data_time": 0.5981223106384277, "loss": 0.1813565731048584, "time": 0.7393783807754517, "epoch": 7, "memory": 6319, "step": 24752} +{"lr": 4.0115549190371644e-05, "data_time": 0.56242835521698, "loss": 0.16964461356401445, "time": 0.7032805919647217, "epoch": 7, "memory": 6319, "step": 24852} +{"lr": 3.978995421317808e-05, "data_time": 0.5471855640411377, "loss": 0.1756830707192421, "time": 0.6877955675125123, "epoch": 7, "memory": 6319, "step": 24952} +{"lr": 3.94652557495873e-05, "data_time": 0.5761992692947387, "loss": 0.17393409907817842, "time": 0.7203640699386596, "epoch": 7, "memory": 6319, "step": 25052} +{"lr": 3.914147293804207e-05, "data_time": 0.28637559413909913, "loss": 0.16538164466619493, "time": 0.6916460037231446, "epoch": 7, "memory": 6319, "step": 25152} +{"lr": 3.881862486301448e-05, "data_time": 0.29096641540527346, "loss": 0.17479893416166306, "time": 0.6537070989608764, "epoch": 7, "memory": 6319, "step": 25252} +{"lr": 3.849673055388122e-05, "data_time": 0.28748221397399903, "loss": 0.1674222767353058, "time": 0.7168793678283691, "epoch": 7, "memory": 6319, "step": 25352} +{"lr": 3.817580898380197e-05, "data_time": 0.48348379135131836, "loss": 0.17319290190935135, "time": 0.6882036209106446, "epoch": 7, "memory": 6319, "step": 25452} +{"lr": 3.7855879068600975e-05, "data_time": 0.49041578769683836, "loss": 0.15639640837907792, "time": 0.7223220109939575, "epoch": 7, "memory": 6319, "step": 25552} +{"lr": 3.753695966565216e-05, "data_time": 0.565480089187622, "loss": 0.156770521402359, "time": 0.7047290086746216, "epoch": 7, "memory": 6319, "step": 25652} +{"lr": 3.7219069572767636e-05, "data_time": 0.5666501522064209, "loss": 0.18828204572200774, "time": 0.7078155994415283, "epoch": 7, "memory": 6319, "step": 25752} +{"lr": 3.6902227527089725e-05, "data_time": 0.6203948736190796, "loss": 0.1799191102385521, "time": 0.7598810434341431, "epoch": 7, "memory": 6319, "step": 25852} +{"lr": 3.65864522039865e-05, "data_time": 0.5880690336227417, "loss": 0.1735661044716835, "time": 0.7283035278320312, "epoch": 7, "memory": 6319, "step": 25952} +{"lr": 3.627176221595109e-05, "data_time": 0.5464682817459107, "loss": 0.1626366138458252, "time": 0.6872701644897461, "epoch": 7, "memory": 6319, "step": 26052} +{"lr": 3.595817611150461e-05, "data_time": 0.5690017938613892, "loss": 0.1855938032269478, "time": 0.7093132019042969, "epoch": 7, "memory": 6319, "step": 26152} +{"lr": 3.5645712374102824e-05, "data_time": 0.5577040910720825, "loss": 0.16719404011964797, "time": 0.6989094734191894, "epoch": 7, "memory": 6319, "step": 26252} +{"lr": 3.533438942104663e-05, "data_time": 0.5526257276535034, "loss": 0.16869386434555053, "time": 0.6935862541198731, "epoch": 7, "memory": 6319, "step": 26352} +{"lr": 3.5024225602396704e-05, "data_time": 0.5435844421386719, "loss": 0.15850681364536284, "time": 0.6865845918655396, "epoch": 7, "memory": 6319, "step": 26452} +{"lr": 3.4715239199891735e-05, "data_time": 0.3833840131759644, "loss": 0.1629615105688572, "time": 0.6926536321640014, "epoch": 7, "memory": 6319, "step": 26552} +{"lr": 3.44074484258709e-05, "data_time": 0.26637959480285645, "loss": 0.16717423796653746, "time": 0.6746260643005371, "epoch": 7, "memory": 6319, "step": 26652} +{"lr": 3.4100871422200426e-05, "data_time": 0.43064260482788086, "loss": 0.16795639991760253, "time": 0.7114032506942749, "epoch": 7, "memory": 6319, "step": 26752} +{"lr": 3.379552625920419e-05, "data_time": 0.46738812923431394, "loss": 0.17410181760787963, "time": 0.689372181892395, "epoch": 7, "memory": 6319, "step": 26852} +{"lr": 3.349143093459868e-05, "data_time": 0.5961261510848999, "loss": 0.19399193972349166, "time": 0.7370005130767823, "epoch": 7, "memory": 6319, "step": 26952} +{"lr": 3.318860337243226e-05, "data_time": 0.5339762687683105, "loss": 0.1768278032541275, "time": 0.676361870765686, "epoch": 7, "memory": 6319, "step": 27052} +{"lr": 3.288706142202845e-05, "data_time": 0.5447427034378052, "loss": 0.17543355897068977, "time": 0.6852931976318359, "epoch": 7, "memory": 6319, "step": 27152} +{"lr": 3.25868228569341e-05, "data_time": 0.5489830493927002, "loss": 0.16413650959730147, "time": 0.6966290950775147, "epoch": 7, "memory": 6319, "step": 27252} +{"lr": 3.228790537387148e-05, "data_time": 0.5382426500320434, "loss": 0.16559230983257295, "time": 0.6792003154754639, "epoch": 7, "memory": 6319, "step": 27352} +{"lr": 3.199032659169556e-05, "data_time": 0.5310146093368531, "loss": 0.17347155064344405, "time": 0.6708183288574219, "epoch": 7, "memory": 6319, "step": 27452} +{"lr": 3.169410405035527e-05, "data_time": 0.4976022958755493, "loss": 0.1528003215789795, "time": 0.6380467891693116, "epoch": 7, "memory": 6319, "step": 27552} +{"lr": 3.139925520985965e-05, "data_time": 0.5412569046020508, "loss": 0.1775051087141037, "time": 0.6814433097839355, "epoch": 7, "memory": 6319, "step": 27652} +{"lr": 3.1105797449248916e-05, "data_time": 0.5962021589279175, "loss": 0.15354120805859567, "time": 0.7419838666915893, "epoch": 7, "memory": 6319, "step": 27752} +{"lr": 3.081374806556986e-05, "data_time": 0.5891196489334106, "loss": 0.16502247899770736, "time": 0.7288280725479126, "epoch": 7, "memory": 6319, "step": 27852} +{"lr": 3.052312427285649e-05, "data_time": 0.5337841033935546, "loss": 0.1655918374657631, "time": 0.6746386766433716, "epoch": 7, "memory": 6319, "step": 27952} +{"lr": 3.0233943201115213e-05, "data_time": 0.5015422105789185, "loss": 0.1782305881381035, "time": 0.6422603607177735, "epoch": 7, "memory": 6319, "step": 28052} +{"lr": 2.9946221895315525e-05, "data_time": 0.5657840013504029, "loss": 0.17332247346639634, "time": 0.7050846576690674, "epoch": 7, "memory": 6319, "step": 28152} +{"lr": 2.9659977314384953e-05, "data_time": 0.5270125150680542, "loss": 0.1612229660153389, "time": 0.6682371377944947, "epoch": 7, "memory": 6319, "step": 28252} +{"lr": 2.93752263302096e-05, "data_time": 0.5378301620483399, "loss": 0.1648441880941391, "time": 0.6775393009185791, "epoch": 7, "memory": 6319, "step": 28352} +{"lr": 2.9091985726639714e-05, "data_time": 0.5562945365905761, "loss": 0.16484554409980773, "time": 0.697039794921875, "epoch": 7, "memory": 6319, "step": 28452} +{"lr": 2.881027219850035e-05, "data_time": 0.5438944339752197, "loss": 0.16967946738004686, "time": 0.684760332107544, "epoch": 7, "memory": 6319, "step": 28552} +{"accuracy/top1": 91.73493957519531, "data_time": 0.3212118367354075, "time": 0.4096764226754506, "step": 7} +{"lr": 2.827372366659638e-05, "data_time": 0.43496387004852294, "loss": 0.15929679274559022, "time": 0.6692456007003784, "epoch": 8, "memory": 6319, "step": 28744} +{"lr": 2.7996563897301616e-05, "data_time": 0.4415527582168579, "loss": 0.17239178866147994, "time": 0.6596245527267456, "epoch": 8, "memory": 6319, "step": 28844} +{"lr": 2.7720995770013028e-05, "data_time": 0.49218761920928955, "loss": 0.1620272070169449, "time": 0.6672914028167725, "epoch": 8, "memory": 6319, "step": 28944} +{"lr": 2.744703552732299e-05, "data_time": 0.34904003143310547, "loss": 0.16939831748604775, "time": 0.7152575969696044, "epoch": 8, "memory": 6319, "step": 29044} +{"lr": 2.717469931705171e-05, "data_time": 0.4309261322021484, "loss": 0.1526232421398163, "time": 0.7138391971588135, "epoch": 8, "memory": 6319, "step": 29144} +{"lr": 2.690400319129529e-05, "data_time": 0.15989320278167723, "loss": 0.17122574746608735, "time": 0.6754186630249024, "epoch": 8, "memory": 6319, "step": 29244} +{"lr": 2.6634963105479716e-05, "data_time": 0.07138471603393555, "loss": 0.1931123211979866, "time": 0.6669699907302856, "epoch": 8, "memory": 6319, "step": 29344} +{"lr": 2.6367594917420177e-05, "data_time": 0.1935683012008667, "loss": 0.16052377671003343, "time": 0.7234967708587646, "epoch": 8, "memory": 6319, "step": 29444} +{"lr": 2.610191438638673e-05, "data_time": 0.34243483543395997, "loss": 0.16379099935293198, "time": 0.7862113237380981, "epoch": 8, "memory": 6319, "step": 29544} +{"lr": 2.5837937172175032e-05, "data_time": 0.0009986400604248048, "loss": 0.17309513613581656, "time": 0.7582850933074952, "epoch": 8, "memory": 6319, "step": 29644} +{"lr": 2.557567883418364e-05, "data_time": 0.0009107112884521485, "loss": 0.16048759445548058, "time": 0.7508659601211548, "epoch": 8, "memory": 6319, "step": 29744} +{"lr": 2.531515483049656e-05, "data_time": 0.0010088682174682617, "loss": 0.16359711214900016, "time": 0.7036072015762329, "epoch": 8, "memory": 6319, "step": 29844} +{"lr": 2.5056380516972453e-05, "data_time": 0.0010594844818115235, "loss": 0.1691916584968567, "time": 0.6536807298660279, "epoch": 8, "memory": 6319, "step": 29944} +{"lr": 2.4799371146339435e-05, "data_time": 0.0011473894119262695, "loss": 0.1659874677658081, "time": 0.7157558679580689, "epoch": 8, "memory": 6319, "step": 30044} +{"lr": 2.4544141867295954e-05, "data_time": 0.0008456945419311523, "loss": 0.1564167097210884, "time": 0.6811789035797119, "epoch": 8, "memory": 6319, "step": 30144} +{"lr": 2.429070772361796e-05, "data_time": 0.0009846210479736329, "loss": 0.17010714411735534, "time": 0.7032801389694214, "epoch": 8, "memory": 6319, "step": 30244} +{"lr": 2.4039083653272075e-05, "data_time": 0.0010410070419311524, "loss": 0.1742545932531357, "time": 0.7389954328536987, "epoch": 8, "memory": 6319, "step": 30344} +{"lr": 2.378928448753535e-05, "data_time": 0.0008489370346069336, "loss": 0.1634024515748024, "time": 0.6612615823745728, "epoch": 8, "memory": 6319, "step": 30444} +{"lr": 2.354132495012082e-05, "data_time": 0.0011972665786743164, "loss": 0.17912216931581498, "time": 0.7069468021392822, "epoch": 8, "memory": 6319, "step": 30544} +{"lr": 2.329521965630983e-05, "data_time": 0.0008386611938476562, "loss": 0.16748422980308533, "time": 0.7016327142715454, "epoch": 8, "memory": 6319, "step": 30644} +{"lr": 2.3050983112090472e-05, "data_time": 0.0009271621704101563, "loss": 0.18147973343729973, "time": 0.7067485332489014, "epoch": 8, "memory": 6319, "step": 30744} +{"lr": 2.2808629713302708e-05, "data_time": 0.0009601831436157227, "loss": 0.15142384618520738, "time": 0.7132920503616333, "epoch": 8, "memory": 6319, "step": 30844} +{"lr": 2.2568173744789706e-05, "data_time": 0.0009085655212402344, "loss": 0.15970299243927003, "time": 0.6975794076919556, "epoch": 8, "memory": 6319, "step": 30944} +{"lr": 2.2329629379555897e-05, "data_time": 0.0009399890899658203, "loss": 0.18512292504310607, "time": 0.9630717515945435, "epoch": 8, "memory": 6319, "step": 31044} +{"lr": 2.2093010677931667e-05, "data_time": 0.0009394168853759765, "loss": 0.16732430309057236, "time": 0.7877877712249756, "epoch": 8, "memory": 6319, "step": 31144} +{"lr": 2.1858331586744434e-05, "data_time": 0.0009273529052734375, "loss": 0.16037182062864302, "time": 0.7212547779083252, "epoch": 8, "memory": 6319, "step": 31244} +{"lr": 2.1625605938496814e-05, "data_time": 0.0010665416717529296, "loss": 0.16246872991323472, "time": 0.6757916927337646, "epoch": 8, "memory": 6319, "step": 31344} +{"lr": 2.1394847450551186e-05, "data_time": 0.0010218381881713866, "loss": 0.17783166021108626, "time": 0.7036640882492066, "epoch": 8, "memory": 6319, "step": 31444} +{"lr": 2.1166069724321078e-05, "data_time": 0.0009870052337646485, "loss": 0.17901988476514816, "time": 0.6738017797470093, "epoch": 8, "memory": 6319, "step": 31544} +{"lr": 2.0939286244469643e-05, "data_time": 0.0009018898010253907, "loss": 0.1610145889222622, "time": 0.6912663459777832, "epoch": 8, "memory": 6319, "step": 31644} +{"lr": 2.071451037811473e-05, "data_time": 0.0007386684417724609, "loss": 0.17289062291383744, "time": 0.7477646350860596, "epoch": 8, "memory": 6319, "step": 31744} +{"lr": 2.0491755374041045e-05, "data_time": 0.000998210906982422, "loss": 0.19977452903985976, "time": 0.7184840440750122, "epoch": 8, "memory": 6319, "step": 31844} +{"lr": 2.0271034361919193e-05, "data_time": 0.0008982658386230469, "loss": 0.17608511745929717, "time": 0.7237038612365723, "epoch": 8, "memory": 6319, "step": 31944} +{"lr": 2.0052360351531814e-05, "data_time": 0.0008471965789794922, "loss": 0.17564684599637986, "time": 0.7074095010757446, "epoch": 8, "memory": 6319, "step": 32044} +{"lr": 1.983574623200682e-05, "data_time": 0.0008153915405273438, "loss": 0.15017342045903206, "time": 0.7198257923126221, "epoch": 8, "memory": 6319, "step": 32144} +{"lr": 1.9621204771057533e-05, "data_time": 0.0008871078491210938, "loss": 0.1591176211833954, "time": 0.7024089813232421, "epoch": 8, "memory": 6319, "step": 32244} +{"lr": 1.940874861423025e-05, "data_time": 0.0008514404296875, "loss": 0.18107160776853562, "time": 0.7456432104110717, "epoch": 8, "memory": 6319, "step": 32344} +{"lr": 1.9198390284158783e-05, "data_time": 0.0008984088897705078, "loss": 0.16974645629525184, "time": 1.039924430847168, "epoch": 8, "memory": 6319, "step": 32444} +{"lr": 1.8990142179826573e-05, "data_time": 0.0008800745010375977, "loss": 0.17854541093111037, "time": 0.7250747203826904, "epoch": 8, "memory": 6319, "step": 32544} +{"lr": 1.8784016575835568e-05, "data_time": 0.0008195638656616211, "loss": 0.16246017515659333, "time": 0.7141840696334839, "epoch": 8, "memory": 6319, "step": 32644} +{"accuracy/top1": 92.90667724609375, "data_time": 0.3211768865585327, "time": 0.40820035139719646, "step": 8} +{"lr": 1.8394249604901642e-05, "data_time": 0.18729810714721679, "loss": 0.15460986346006395, "time": 0.7207031011581421, "epoch": 9, "memory": 6319, "step": 32836} +{"lr": 1.81943907744488e-05, "data_time": 0.18993062973022462, "loss": 0.16875767558813096, "time": 0.6815762996673584, "epoch": 9, "memory": 6319, "step": 32936} +{"lr": 1.7996701347677866e-05, "data_time": 0.31720824241638185, "loss": 0.1572006583213806, "time": 0.6806392431259155, "epoch": 9, "memory": 6319, "step": 33036} +{"lr": 1.780119297683906e-05, "data_time": 0.3877443313598633, "loss": 0.1749775841832161, "time": 0.7137496709823609, "epoch": 9, "memory": 6319, "step": 33136} +{"lr": 1.7607877185626483e-05, "data_time": 0.294804310798645, "loss": 0.1448626011610031, "time": 0.6825323104858398, "epoch": 9, "memory": 6319, "step": 33236} +{"lr": 1.7416765368498675e-05, "data_time": 0.16448497772216797, "loss": 0.17938916236162186, "time": 0.705608582496643, "epoch": 9, "memory": 6319, "step": 33336} +{"lr": 1.7227868790007105e-05, "data_time": 0.0008262157440185547, "loss": 0.17950266003608703, "time": 0.7025579690933228, "epoch": 9, "memory": 6319, "step": 33436} +{"lr": 1.7041198584132233e-05, "data_time": 0.0009088516235351562, "loss": 0.16385626047849655, "time": 0.7037282705307006, "epoch": 9, "memory": 6319, "step": 33536} +{"lr": 1.6856765753627146e-05, "data_time": 0.24636623859405518, "loss": 0.1574540063738823, "time": 0.7312356233596802, "epoch": 9, "memory": 6319, "step": 33636} +{"lr": 1.6674581169369163e-05, "data_time": 0.15081188678741456, "loss": 0.14947786033153534, "time": 0.6530324459075928, "epoch": 9, "memory": 6319, "step": 33736} +{"lr": 1.6494655569718982e-05, "data_time": 0.2675036430358887, "loss": 0.17042155265808107, "time": 0.6659437894821167, "epoch": 9, "memory": 6319, "step": 33836} +{"lr": 1.6316999559887786e-05, "data_time": 0.2079176664352417, "loss": 0.1611701712012291, "time": 0.7437841415405273, "epoch": 9, "memory": 6319, "step": 33936} +{"lr": 1.6141623611312105e-05, "data_time": 0.03393511772155762, "loss": 0.1829184889793396, "time": 0.714853572845459, "epoch": 9, "memory": 6319, "step": 34036} +{"lr": 1.5968538061036705e-05, "data_time": 0.0009735345840454102, "loss": 0.15594484135508538, "time": 0.7060147047042846, "epoch": 9, "memory": 6319, "step": 34136} +{"lr": 1.5797753111105157e-05, "data_time": 0.0008349418640136719, "loss": 0.16882977783679962, "time": 0.6934847831726074, "epoch": 9, "memory": 6319, "step": 34236} +{"lr": 1.5629278827958574e-05, "data_time": 0.0009967327117919923, "loss": 0.16971203088760375, "time": 0.6836693525314331, "epoch": 9, "memory": 6319, "step": 34336} +{"lr": 1.5463125141842392e-05, "data_time": 0.0009215116500854492, "loss": 0.1715298682451248, "time": 0.7374971151351929, "epoch": 9, "memory": 6319, "step": 34436} +{"lr": 1.5299301846220796e-05, "data_time": 0.00105438232421875, "loss": 0.17847218066453935, "time": 0.7000346899032592, "epoch": 9, "memory": 6319, "step": 34536} +{"lr": 1.5137818597199751e-05, "data_time": 0.0008035182952880859, "loss": 0.1638461098074913, "time": 0.7424283027648926, "epoch": 9, "memory": 6319, "step": 34636} +{"lr": 1.4978684912957622e-05, "data_time": 0.0007372379302978515, "loss": 0.15982463583350182, "time": 0.6368038415908813, "epoch": 9, "memory": 6319, "step": 34736} +{"lr": 1.4821910173184321e-05, "data_time": 0.0008972406387329102, "loss": 0.1588974714279175, "time": 0.6889965295791626, "epoch": 9, "memory": 6319, "step": 34836} +{"lr": 1.4667503618528337e-05, "data_time": 0.0008920431137084961, "loss": 0.16023080199956893, "time": 0.7143422365188599, "epoch": 9, "memory": 6319, "step": 34936} +{"lr": 1.4515474350052097e-05, "data_time": 0.0008902549743652344, "loss": 0.15848468691110612, "time": 0.7422490358352661, "epoch": 9, "memory": 6319, "step": 35036} +{"lr": 1.4365831328695616e-05, "data_time": 0.0008068323135375976, "loss": 0.1650614470243454, "time": 0.7103931188583374, "epoch": 9, "memory": 6319, "step": 35136} +{"lr": 1.4218583374748169e-05, "data_time": 0.0008558034896850586, "loss": 0.1632140651345253, "time": 0.7484535932540893, "epoch": 9, "memory": 6319, "step": 35236} +{"lr": 1.4073739167328542e-05, "data_time": 0.0010104894638061524, "loss": 0.15982672795653344, "time": 0.6741731882095336, "epoch": 9, "memory": 6319, "step": 35336} +{"lr": 1.3931307243873344e-05, "data_time": 0.0009338140487670898, "loss": 0.16906670555472375, "time": 0.6640015125274659, "epoch": 9, "memory": 6319, "step": 35436} +{"lr": 1.3791295999633911e-05, "data_time": 0.0010547876358032227, "loss": 0.15604328289628028, "time": 0.6555068731307984, "epoch": 9, "memory": 6319, "step": 35536} +{"lr": 1.3653713687181394e-05, "data_time": 0.0010912418365478516, "loss": 0.1809113949537277, "time": 0.6874321699142456, "epoch": 9, "memory": 6319, "step": 35636} +{"lr": 1.3518568415920337e-05, "data_time": 0.0009035348892211914, "loss": 0.17026009708642958, "time": 0.7585430383682251, "epoch": 9, "memory": 6319, "step": 35736} +{"lr": 1.3385868151610723e-05, "data_time": 0.20730414390563964, "loss": 0.1638484090566635, "time": 0.7466943740844727, "epoch": 9, "memory": 6319, "step": 35836} +{"lr": 1.3255620715898442e-05, "data_time": 0.18925862312316893, "loss": 0.16586991995573044, "time": 0.7194545030593872, "epoch": 9, "memory": 6319, "step": 35936} +{"lr": 1.3127833785854261e-05, "data_time": 0.0010066032409667969, "loss": 0.1620241478085518, "time": 0.7628704071044922, "epoch": 9, "memory": 6319, "step": 36036} +{"lr": 1.3002514893521313e-05, "data_time": 0.0009039878845214844, "loss": 0.16661669462919235, "time": 0.7198026180267334, "epoch": 9, "memory": 6319, "step": 36136} +{"lr": 1.287967142547117e-05, "data_time": 0.000866556167602539, "loss": 0.15979148373007773, "time": 0.7427562475204468, "epoch": 9, "memory": 6319, "step": 36236} +{"lr": 1.2759310622368435e-05, "data_time": 0.000870656967163086, "loss": 0.15942969620227815, "time": 0.7897002935409546, "epoch": 9, "memory": 6319, "step": 36336} +{"lr": 1.2641439578543968e-05, "data_time": 0.0008477210998535156, "loss": 0.16918978691101075, "time": 0.6903006792068481, "epoch": 9, "memory": 6319, "step": 36436} +{"lr": 1.252606524157674e-05, "data_time": 0.0008438825607299805, "loss": 0.16622457951307296, "time": 0.6901096343994141, "epoch": 9, "memory": 6319, "step": 36536} +{"lr": 1.2413194411884297e-05, "data_time": 0.0008791923522949219, "loss": 0.17171796560287475, "time": 0.68916335105896, "epoch": 9, "memory": 6319, "step": 36636} +{"lr": 1.2302833742321981e-05, "data_time": 0.0008104562759399414, "loss": 0.1667586825788021, "time": 0.7203540563583374, "epoch": 9, "memory": 6319, "step": 36736} +{"accuracy/top1": 93.01590728759766, "data_time": 0.31340978542963666, "time": 0.40285993019739785, "step": 9} +{"lr": 1.2098001439139718e-05, "data_time": 0.09869544506072998, "loss": 0.16758147701621057, "time": 0.6929279088973999, "epoch": 10, "memory": 6319, "step": 36928} +{"lr": 1.1995007122352056e-05, "data_time": 0.20224294662475586, "loss": 0.1677936092019081, "time": 0.7161273002624512, "epoch": 10, "memory": 6319, "step": 37028} +{"lr": 1.1894547614574522e-05, "data_time": 0.000911259651184082, "loss": 0.1425803378224373, "time": 0.7055025577545166, "epoch": 10, "memory": 6319, "step": 37128} +{"lr": 1.1796628837111806e-05, "data_time": 0.0008507013320922851, "loss": 0.1644194945693016, "time": 0.7238788604736328, "epoch": 10, "memory": 6319, "step": 37228} +{"lr": 1.1701256561512343e-05, "data_time": 0.0008981466293334961, "loss": 0.16315096914768218, "time": 0.6704913854599, "epoch": 10, "memory": 6319, "step": 37328} +{"lr": 1.1608436409228136e-05, "data_time": 0.0009629487991333008, "loss": 0.18481188416481018, "time": 0.6627022743225097, "epoch": 10, "memory": 6319, "step": 37428} +{"lr": 1.1518173851283401e-05, "data_time": 0.0008953094482421875, "loss": 0.1533253327012062, "time": 0.6897203207015992, "epoch": 10, "memory": 6319, "step": 37528} +{"lr": 1.1430474207952176e-05, "data_time": 0.0009173393249511719, "loss": 0.14821891188621522, "time": 0.7070162296295166, "epoch": 10, "memory": 6319, "step": 37628} +{"lr": 1.1345342648444593e-05, "data_time": 0.001033473014831543, "loss": 0.14969507306814195, "time": 0.7723404169082642, "epoch": 10, "memory": 6319, "step": 37728} +{"lr": 1.12627841906023e-05, "data_time": 0.0008768081665039062, "loss": 0.15251275449991225, "time": 0.6891923427581788, "epoch": 10, "memory": 6319, "step": 37828} +{"lr": 1.1182803700602659e-05, "data_time": 0.0009001731872558594, "loss": 0.15483831539750098, "time": 0.6577965259552002, "epoch": 10, "memory": 6319, "step": 37928} +{"lr": 1.110540589267192e-05, "data_time": 0.0010291576385498048, "loss": 0.150139519572258, "time": 0.6490852355957031, "epoch": 10, "memory": 6319, "step": 38028} +{"lr": 1.1030595328807396e-05, "data_time": 0.0008701562881469726, "loss": 0.15066432654857637, "time": 0.7208055257797241, "epoch": 10, "memory": 6319, "step": 38128} +{"lr": 1.0958376418508494e-05, "data_time": 0.0010786056518554688, "loss": 0.16919270902872086, "time": 0.6935332775115967, "epoch": 10, "memory": 6319, "step": 38228} +{"lr": 1.088875341851686e-05, "data_time": 0.001039886474609375, "loss": 0.1632241539657116, "time": 0.6928199768066406, "epoch": 10, "memory": 6319, "step": 38328} +{"lr": 1.0821730432565486e-05, "data_time": 0.0010054588317871093, "loss": 0.17095913887023925, "time": 0.7007759571075439, "epoch": 10, "memory": 6319, "step": 38428} +{"lr": 1.0757311411136797e-05, "data_time": 0.0009536981582641601, "loss": 0.17949067652225495, "time": 0.7212576150894165, "epoch": 10, "memory": 6319, "step": 38528} +{"lr": 1.0695500151229833e-05, "data_time": 0.0008974075317382812, "loss": 0.15975531786680222, "time": 0.7136516571044922, "epoch": 10, "memory": 6319, "step": 38628} +{"lr": 1.0636300296136373e-05, "data_time": 0.0009805679321289063, "loss": 0.17370918840169908, "time": 0.7227862358093262, "epoch": 10, "memory": 6319, "step": 38728} +{"lr": 1.0579715335226322e-05, "data_time": 0.0008877754211425781, "loss": 0.15194374546408654, "time": 0.6469579935073853, "epoch": 10, "memory": 6319, "step": 38828} +{"lr": 1.0525748603741896e-05, "data_time": 0.0008641958236694336, "loss": 0.1799135535955429, "time": 0.7223961591720581, "epoch": 10, "memory": 6319, "step": 38928} +{"lr": 1.0474403282601119e-05, "data_time": 0.0009482860565185547, "loss": 0.17140405997633934, "time": 0.7346735954284668, "epoch": 10, "memory": 6319, "step": 39028} +{"lr": 1.0425682398210336e-05, "data_time": 0.0009703397750854492, "loss": 0.15233578756451607, "time": 0.697683310508728, "epoch": 10, "memory": 6319, "step": 39128} +{"lr": 1.0379588822285784e-05, "data_time": 0.0009176492691040039, "loss": 0.16907952725887299, "time": 0.7015511274337769, "epoch": 10, "memory": 6319, "step": 39228} +{"lr": 1.0336125271684364e-05, "data_time": 0.0008747339248657227, "loss": 0.16613706797361374, "time": 0.683406400680542, "epoch": 10, "memory": 6319, "step": 39328} +{"lr": 1.0295294308243494e-05, "data_time": 0.0008446693420410157, "loss": 0.16299415975809098, "time": 0.6715137004852295, "epoch": 10, "memory": 6319, "step": 39428} +{"lr": 1.0257098338630086e-05, "data_time": 0.0007773160934448243, "loss": 0.16641470342874526, "time": 0.688904595375061, "epoch": 10, "memory": 6319, "step": 39528} +{"lr": 1.0221539614198729e-05, "data_time": 0.0008801937103271485, "loss": 0.14935189485549927, "time": 0.7031291723251343, "epoch": 10, "memory": 6319, "step": 39628} +{"lr": 1.0188620230858978e-05, "data_time": 0.0008347511291503906, "loss": 0.1692873314023018, "time": 0.7959968566894531, "epoch": 10, "memory": 6319, "step": 39728} +{"lr": 1.01583421289518e-05, "data_time": 0.0007643461227416992, "loss": 0.1614592969417572, "time": 0.6756434440612793, "epoch": 10, "memory": 6319, "step": 39828} +{"lr": 1.0130707093135206e-05, "data_time": 0.0010109663009643555, "loss": 0.15934257507324218, "time": 0.6999173879623413, "epoch": 10, "memory": 6319, "step": 39928} +{"lr": 1.01057167522791e-05, "data_time": 0.0008799314498901367, "loss": 0.17094288468360902, "time": 0.7179607629776001, "epoch": 10, "memory": 6319, "step": 40028} +{"lr": 1.008337257936921e-05, "data_time": 0.0009057283401489258, "loss": 0.1505805179476738, "time": 0.7259350538253784, "epoch": 10, "memory": 6319, "step": 40128} +{"lr": 1.0063675891420294e-05, "data_time": 0.001208662986755371, "loss": 0.15701159834861755, "time": 0.6661712646484375, "epoch": 10, "memory": 6319, "step": 40228} +{"lr": 1.0046627849398548e-05, "data_time": 0.0011827707290649413, "loss": 0.16071770191192628, "time": 0.6966933250427246, "epoch": 10, "memory": 6319, "step": 40328} +{"lr": 1.0032229458153106e-05, "data_time": 0.0008065462112426758, "loss": 0.14961961209774016, "time": 0.9084820985794068, "epoch": 10, "memory": 6319, "step": 40428} +{"lr": 1.0020481566356862e-05, "data_time": 0.0009717702865600586, "loss": 0.15234164893627167, "time": 0.7270920753479004, "epoch": 10, "memory": 6319, "step": 40528} +{"lr": 1.0011384866456456e-05, "data_time": 0.00099637508392334, "loss": 0.15804436951875686, "time": 0.7482522487640381, "epoch": 10, "memory": 6319, "step": 40628} +{"lr": 1.00049398946314e-05, "data_time": 0.0008790493011474609, "loss": 0.16904067248106003, "time": 0.7159698009490967, "epoch": 10, "memory": 6319, "step": 40728} +{"lr": 1.0001147030762536e-05, "data_time": 0.000933384895324707, "loss": 0.15995430648326875, "time": 0.6835578918457031, "epoch": 10, "memory": 6319, "step": 40828} +{"accuracy/top1": 92.8636474609375, "data_time": 0.3246283312638601, "time": 0.4124751389026642, "step": 10} diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/config.py b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/config.py new file mode 100644 index 0000000000000000000000000000000000000000..0577dd2008d2b155b4427d6d87779bcabd4cbeec --- /dev/null +++ b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/config.py @@ -0,0 +1,349 @@ +optim_wrapper = dict( + optimizer=dict( + type='SGD', + lr=0.0001, + momentum=0.9, + weight_decay=0.0001, + _scope_='mmpretrain'), + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=False, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=512) +model = dict( + type='ImageClassifier', + backbone=dict( + frozen_stages=2, + type='ResNet', + depth=50, + num_stages=4, + out_indices=(3, ), + style='pytorch', + init_cfg=dict( + type='Pretrained', + checkpoint= + 'https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth', + prefix='backbone')), + neck=dict(type='GlobalAveragePooling'), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=2048, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + topk=1)) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/all_0.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/all_1.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3fake8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/cc1m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3real8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV2-1-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/if-dpmsolver++-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = dict(type='Accuracy', topk=1) +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV2-1-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/if-dpmsolver++-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = dict(type='Accuracy', topk=1) +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1' diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/events.out.tfevents.1685984276.SH-IDC1-10-140-24-103.212173.0 b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/events.out.tfevents.1685984276.SH-IDC1-10-140-24-103.212173.0 new file mode 100644 index 0000000000000000000000000000000000000000..28230c36ff00017e45bfbadd20568ac013bb336e --- /dev/null +++ b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/events.out.tfevents.1685984276.SH-IDC1-10-140-24-103.212173.0 @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:a73ab0ae60673e7323ee6172ccfa58de8c75ec4b740538790b67606009157028 +size 5587606 diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/scalars.json b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/scalars.json new file mode 100644 index 0000000000000000000000000000000000000000..c44a590b61948449a56eed401d7d3473b7008063 --- /dev/null +++ b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/scalars.json @@ -0,0 +1,410 @@ +{"lr": 9.999870019168206e-05, "data_time": 0.23580620288848878, "loss": 0.6417255282402039, "time": 0.8090263366699219, "epoch": 1, "memory": 9436, "step": 100} +{"lr": 9.99947481934738e-05, "data_time": 0.21867220401763915, "loss": 0.5970972716808319, "time": 0.7791444778442382, "epoch": 1, "memory": 6319, "step": 200} +{"lr": 9.99881441056942e-05, "data_time": 0.0008538007736206054, "loss": 0.5583989799022675, "time": 0.7499716997146606, "epoch": 1, "memory": 6319, "step": 300} +{"lr": 9.997888831760252e-05, "data_time": 0.0008380651473999023, "loss": 0.5301752865314484, "time": 0.7361132621765136, "epoch": 1, "memory": 6319, "step": 400} +{"lr": 9.996698137475549e-05, "data_time": 0.2010805606842041, "loss": 0.5015071302652359, "time": 0.7582030534744263, "epoch": 1, "memory": 6319, "step": 500} +{"lr": 9.99524239789746e-05, "data_time": 0.3320797920227051, "loss": 0.46877075731754303, "time": 0.7183988094329834, "epoch": 1, "memory": 6319, "step": 600} +{"lr": 9.993521698830466e-05, "data_time": 0.02013735771179199, "loss": 0.43822180926799775, "time": 0.7378515005111694, "epoch": 1, "memory": 6319, "step": 700} +{"lr": 9.991536141696371e-05, "data_time": 0.0007852315902709961, "loss": 0.4319634079933167, "time": 0.7372121334075927, "epoch": 1, "memory": 6319, "step": 800} +{"lr": 9.989285843528274e-05, "data_time": 0.11607332229614258, "loss": 0.42011411786079406, "time": 0.7631598949432373, "epoch": 1, "memory": 6319, "step": 900} +{"lr": 9.986770936963713e-05, "data_time": 0.0008153676986694336, "loss": 0.4003512740135193, "time": 0.7913449287414551, "epoch": 1, "memory": 6319, "step": 1000} +{"lr": 9.983991570236818e-05, "data_time": 0.0007603883743286133, "loss": 0.39497354030609133, "time": 0.8244112253189086, "epoch": 1, "memory": 6319, "step": 1100} +{"lr": 9.980947907169598e-05, "data_time": 0.0007162094116210938, "loss": 0.3863217025995255, "time": 0.722666573524475, "epoch": 1, "memory": 6319, "step": 1200} +{"lr": 9.977640127162247e-05, "data_time": 0.0008350849151611328, "loss": 0.37821408808231355, "time": 0.7241732358932496, "epoch": 1, "memory": 6319, "step": 1300} +{"lr": 9.974068425182603e-05, "data_time": 0.0007997035980224609, "loss": 0.37111017405986785, "time": 0.7567805051803589, "epoch": 1, "memory": 6319, "step": 1400} +{"lr": 9.970233011754657e-05, "data_time": 0.0007455587387084961, "loss": 0.3742601066827774, "time": 0.7378098011016846, "epoch": 1, "memory": 6319, "step": 1500} +{"lr": 9.966134112946122e-05, "data_time": 0.0008747339248657227, "loss": 0.3582422524690628, "time": 0.6978653907775879, "epoch": 1, "memory": 6319, "step": 1600} +{"lr": 9.961771970355126e-05, "data_time": 0.0010759592056274413, "loss": 0.3595370054244995, "time": 0.7230527639389038, "epoch": 1, "memory": 6319, "step": 1700} +{"lr": 9.957146841095961e-05, "data_time": 0.0007981300354003906, "loss": 0.3502417117357254, "time": 0.7151001453399658, "epoch": 1, "memory": 6319, "step": 1800} +{"lr": 9.952258997783933e-05, "data_time": 0.0007390499114990235, "loss": 0.34335593283176424, "time": 0.7786037206649781, "epoch": 1, "memory": 6319, "step": 1900} +{"lr": 9.947108728519288e-05, "data_time": 0.0007749319076538086, "loss": 0.33032276928424836, "time": 0.7474444389343262, "epoch": 1, "memory": 6319, "step": 2000} +{"lr": 9.941696336870245e-05, "data_time": 0.0009086132049560547, "loss": 0.33920717239379883, "time": 0.7526807308197021, "epoch": 1, "memory": 6319, "step": 2100} +{"lr": 9.936022141855094e-05, "data_time": 0.0007847785949707031, "loss": 0.35441104471683504, "time": 0.7442209720611572, "epoch": 1, "memory": 6319, "step": 2200} +{"lr": 9.93008647792338e-05, "data_time": 0.000978565216064453, "loss": 0.3193003237247467, "time": 0.7475589513778687, "epoch": 1, "memory": 6319, "step": 2300} +{"lr": 9.923889694936212e-05, "data_time": 0.001101851463317871, "loss": 0.3393714040517807, "time": 0.7450142621994018, "epoch": 1, "memory": 6319, "step": 2400} +{"lr": 9.917432158145639e-05, "data_time": 0.0007400751113891601, "loss": 0.3217262476682663, "time": 0.7152126550674438, "epoch": 1, "memory": 6319, "step": 2500} +{"lr": 9.91071424817309e-05, "data_time": 0.0009122371673583984, "loss": 0.32098470330238343, "time": 0.6925920009613037, "epoch": 1, "memory": 6319, "step": 2600} +{"lr": 9.903736360986979e-05, "data_time": 0.0007892131805419921, "loss": 0.3037663847208023, "time": 0.6956831455230713, "epoch": 1, "memory": 6319, "step": 2700} +{"lr": 9.89649890787935e-05, "data_time": 0.0008501768112182617, "loss": 0.317654749751091, "time": 0.7175958633422852, "epoch": 1, "memory": 6319, "step": 2800} +{"lr": 9.889002315441642e-05, "data_time": 0.0008883237838745117, "loss": 0.3155059784650803, "time": 0.7144453048706054, "epoch": 1, "memory": 6319, "step": 2900} +{"lr": 9.881247025539515e-05, "data_time": 0.0007520437240600586, "loss": 0.3076498419046402, "time": 0.6948991060256958, "epoch": 1, "memory": 6319, "step": 3000} +{"lr": 9.873233495286844e-05, "data_time": 0.0007988214492797852, "loss": 0.30978504568338394, "time": 0.7028228044509888, "epoch": 1, "memory": 6319, "step": 3100} +{"lr": 9.864962197018756e-05, "data_time": 0.0007736444473266602, "loss": 0.3064905434846878, "time": 0.7453366994857789, "epoch": 1, "memory": 6319, "step": 3200} +{"lr": 9.856433618263779e-05, "data_time": 0.0007336616516113281, "loss": 0.29859074354171755, "time": 0.6710611343383789, "epoch": 1, "memory": 6319, "step": 3300} +{"lr": 9.847648261715132e-05, "data_time": 0.0009308576583862305, "loss": 0.2963121831417084, "time": 0.7274848937988281, "epoch": 1, "memory": 6319, "step": 3400} +{"lr": 9.838606645201093e-05, "data_time": 0.0008805036544799804, "loss": 0.29638334810733796, "time": 0.6771203517913819, "epoch": 1, "memory": 6319, "step": 3500} +{"lr": 9.829309301654456e-05, "data_time": 0.0007841110229492188, "loss": 0.30206618905067445, "time": 0.7793800830841064, "epoch": 1, "memory": 6319, "step": 3600} +{"lr": 9.819756779081127e-05, "data_time": 0.0007678985595703125, "loss": 0.30126443058252333, "time": 0.8266619920730591, "epoch": 1, "memory": 6319, "step": 3700} +{"lr": 9.809949640527814e-05, "data_time": 0.000860285758972168, "loss": 0.3012932062149048, "time": 0.7224830627441406, "epoch": 1, "memory": 6319, "step": 3800} +{"lr": 9.799888464048879e-05, "data_time": 0.0008294820785522461, "loss": 0.2915215402841568, "time": 0.7069743633270263, "epoch": 1, "memory": 6319, "step": 3900} +{"lr": 9.789573842672242e-05, "data_time": 0.0009275674819946289, "loss": 0.3215578556060791, "time": 0.7584826707839966, "epoch": 1, "memory": 6319, "step": 4000} +{"accuracy/top1": 83.01309204101562, "data_time": 0.36436522708219643, "time": 0.4523949502896862, "step": 1} +{"lr": 9.7690615523361e-05, "data_time": 0.39775900840759276, "loss": 0.306628492474556, "time": 0.6732662677764892, "epoch": 2, "memory": 6319, "step": 4192} +{"lr": 9.758010405927656e-05, "data_time": 0.15111191272735597, "loss": 0.2822544798254967, "time": 0.683975625038147, "epoch": 2, "memory": 6319, "step": 4292} +{"lr": 9.746708283006492e-05, "data_time": 0.002385663986206055, "loss": 0.2794072195887566, "time": 0.7211886405944824, "epoch": 2, "memory": 6319, "step": 4392} +{"lr": 9.735155849744634e-05, "data_time": 0.0008119344711303711, "loss": 0.2702935427427292, "time": 0.7382085084915161, "epoch": 2, "memory": 6319, "step": 4492} +{"lr": 9.723353787067928e-05, "data_time": 0.0007747411727905273, "loss": 0.26802886426448824, "time": 0.6471272230148315, "epoch": 2, "memory": 6319, "step": 4592} +{"lr": 9.711302790615951e-05, "data_time": 0.0008364439010620118, "loss": 0.2711342617869377, "time": 0.8386395931243896, "epoch": 2, "memory": 6319, "step": 4692} +{"lr": 9.699003570700986e-05, "data_time": 0.0009061574935913086, "loss": 0.27828952819108965, "time": 0.8523646354675293, "epoch": 2, "memory": 6319, "step": 4792} +{"lr": 9.686456852266142e-05, "data_time": 0.0009198904037475586, "loss": 0.2655982568860054, "time": 0.6747840881347656, "epoch": 2, "memory": 6319, "step": 4892} +{"lr": 9.673663374842644e-05, "data_time": 0.0010221004486083984, "loss": 0.25839774757623674, "time": 0.6460691213607788, "epoch": 2, "memory": 6319, "step": 4992} +{"lr": 9.660623892506218e-05, "data_time": 0.0012143611907958984, "loss": 0.2699123755097389, "time": 0.7331794500350952, "epoch": 2, "memory": 6319, "step": 5092} +{"lr": 9.647339173832693e-05, "data_time": 0.0008815288543701172, "loss": 0.26809143722057344, "time": 0.7267658948898316, "epoch": 2, "memory": 6319, "step": 5192} +{"lr": 9.633810001852653e-05, "data_time": 0.000978231430053711, "loss": 0.2733738929033279, "time": 0.7184287071228027, "epoch": 2, "memory": 6319, "step": 5292} +{"lr": 9.620037174005274e-05, "data_time": 0.0007425546646118164, "loss": 0.27412050366401675, "time": 0.7382806062698364, "epoch": 2, "memory": 6319, "step": 5392} +{"lr": 9.606021502091389e-05, "data_time": 0.0006863832473754883, "loss": 0.2644442543387413, "time": 0.7033350706100464, "epoch": 2, "memory": 6319, "step": 5492} +{"lr": 9.591763812225569e-05, "data_time": 0.00098726749420166, "loss": 0.2552147641777992, "time": 0.6655296802520752, "epoch": 2, "memory": 6319, "step": 5592} +{"lr": 9.577264944787459e-05, "data_time": 0.0009675979614257813, "loss": 0.26296195983886717, "time": 0.6554296016693115, "epoch": 2, "memory": 6319, "step": 5692} +{"lr": 9.562525754372252e-05, "data_time": 0.0009081125259399414, "loss": 0.26218927949666976, "time": 0.7197994709014892, "epoch": 2, "memory": 6319, "step": 5792} +{"lr": 9.547547109740282e-05, "data_time": 0.000800633430480957, "loss": 0.26032392531633375, "time": 0.6760532855987549, "epoch": 2, "memory": 6319, "step": 5892} +{"lr": 9.532329893765858e-05, "data_time": 0.0009043455123901367, "loss": 0.27663236260414126, "time": 0.9657118082046509, "epoch": 2, "memory": 6319, "step": 5992} +{"lr": 9.516875003385223e-05, "data_time": 0.0008503675460815429, "loss": 0.24357614368200303, "time": 0.7123346090316772, "epoch": 2, "memory": 6319, "step": 6092} +{"lr": 9.501183349543653e-05, "data_time": 0.0007915019989013672, "loss": 0.27699775248765945, "time": 0.676742434501648, "epoch": 2, "memory": 6319, "step": 6192} +{"lr": 9.48525585714179e-05, "data_time": 0.0010963678359985352, "loss": 0.2426074355840683, "time": 0.6890865564346313, "epoch": 2, "memory": 6319, "step": 6292} +{"lr": 9.469093464981119e-05, "data_time": 0.0007992982864379883, "loss": 0.2461544841527939, "time": 0.7354315996170044, "epoch": 2, "memory": 6319, "step": 6392} +{"lr": 9.452697125708636e-05, "data_time": 0.000832056999206543, "loss": 0.25532765686511993, "time": 0.6789705991744995, "epoch": 2, "memory": 6319, "step": 6492} +{"lr": 9.436067805760705e-05, "data_time": 0.0008382558822631836, "loss": 0.2425134912133217, "time": 0.6513427019119262, "epoch": 2, "memory": 6319, "step": 6592} +{"lr": 9.419206485306072e-05, "data_time": 0.0008160352706909179, "loss": 0.2688654959201813, "time": 0.6749155998229981, "epoch": 2, "memory": 6319, "step": 6692} +{"lr": 9.4021141581881e-05, "data_time": 0.0007049083709716797, "loss": 0.24094131886959075, "time": 0.6558093547821044, "epoch": 2, "memory": 6319, "step": 6792} +{"lr": 9.384791831866193e-05, "data_time": 0.0009281158447265625, "loss": 0.2769829869270325, "time": 0.7018612146377563, "epoch": 2, "memory": 6319, "step": 6892} +{"lr": 9.367240527356428e-05, "data_time": 0.0007861375808715821, "loss": 0.23097819536924363, "time": 0.7395848512649537, "epoch": 2, "memory": 6319, "step": 6992} +{"lr": 9.349461279171362e-05, "data_time": 0.0011840343475341796, "loss": 0.2553142011165619, "time": 0.6742855787277222, "epoch": 2, "memory": 6319, "step": 7092} +{"lr": 9.33145513525904e-05, "data_time": 0.0009016990661621094, "loss": 0.24693509191274643, "time": 0.6630861043930054, "epoch": 2, "memory": 6319, "step": 7192} +{"lr": 9.313223156941242e-05, "data_time": 0.001598024368286133, "loss": 0.2428608626127243, "time": 0.7029174327850342, "epoch": 2, "memory": 6319, "step": 7292} +{"lr": 9.294766418850942e-05, "data_time": 0.0007671117782592773, "loss": 0.2545601323246956, "time": 0.699622368812561, "epoch": 2, "memory": 6319, "step": 7392} +{"lr": 9.276086008868929e-05, "data_time": 0.0007414340972900391, "loss": 0.2544305339455605, "time": 0.6884544610977172, "epoch": 2, "memory": 6319, "step": 7492} +{"lr": 9.257183028059726e-05, "data_time": 0.0009234428405761718, "loss": 0.25289510786533353, "time": 0.7154263496398926, "epoch": 2, "memory": 6319, "step": 7592} +{"lr": 9.238058590606654e-05, "data_time": 0.0007828950881958007, "loss": 0.24429425746202468, "time": 0.7094822406768799, "epoch": 2, "memory": 6319, "step": 7692} +{"lr": 9.218713823746189e-05, "data_time": 0.0011223077774047852, "loss": 0.23091641813516617, "time": 0.7586504697799683, "epoch": 2, "memory": 6319, "step": 7792} +{"lr": 9.199149867701492e-05, "data_time": 0.0011123180389404296, "loss": 0.24989114403724672, "time": 0.6876688003540039, "epoch": 2, "memory": 6319, "step": 7892} +{"lr": 9.179367875615232e-05, "data_time": 0.0007442951202392579, "loss": 0.23347796648740768, "time": 0.6866568803787232, "epoch": 2, "memory": 6319, "step": 7992} +{"lr": 9.159369013481585e-05, "data_time": 0.0007095098495483399, "loss": 0.23617833852767944, "time": 0.711391544342041, "epoch": 2, "memory": 6319, "step": 8092} +{"accuracy/top1": 81.01551055908203, "data_time": 0.3427091976006826, "time": 0.42977015376091005, "step": 2} +{"lr": 9.120367596460475e-05, "data_time": 0.4689095258712769, "loss": 0.23746268898248674, "time": 0.6739122152328492, "epoch": 3, "memory": 6319, "step": 8284} +{"lr": 9.099742266672751e-05, "data_time": 0.33391757011413575, "loss": 0.2527444750070572, "time": 0.700292706489563, "epoch": 3, "memory": 6319, "step": 8384} +{"lr": 9.078904760146445e-05, "data_time": 0.2901160478591919, "loss": 0.23466432243585586, "time": 0.668321967124939, "epoch": 3, "memory": 6319, "step": 8484} +{"lr": 9.057856305090082e-05, "data_time": 0.3167825937271118, "loss": 0.24158217310905455, "time": 0.6561912059783935, "epoch": 3, "memory": 6319, "step": 8584} +{"lr": 9.036598142145979e-05, "data_time": 0.5717406272888184, "loss": 0.21645256280899047, "time": 0.7558250188827514, "epoch": 3, "memory": 6319, "step": 8684} +{"lr": 9.015131524317079e-05, "data_time": 0.537515926361084, "loss": 0.22299962341785431, "time": 0.6771813154220581, "epoch": 3, "memory": 6319, "step": 8784} +{"lr": 8.993457716893124e-05, "data_time": 0.4909390449523926, "loss": 0.2311646595597267, "time": 0.6904112815856933, "epoch": 3, "memory": 6319, "step": 8884} +{"lr": 8.971577997376067e-05, "data_time": 0.5511940240859985, "loss": 0.22824979573488235, "time": 0.6908522605895996, "epoch": 3, "memory": 6319, "step": 8984} +{"lr": 8.949493655404777e-05, "data_time": 0.5729543924331665, "loss": 0.22518849074840547, "time": 0.7130033493041992, "epoch": 3, "memory": 6319, "step": 9084} +{"lr": 8.927205992679013e-05, "data_time": 0.560827088356018, "loss": 0.2453401893377304, "time": 0.7018682479858398, "epoch": 3, "memory": 6319, "step": 9184} +{"lr": 8.9047163228827e-05, "data_time": 0.5887284994125366, "loss": 0.22647609263658525, "time": 0.7276627063751221, "epoch": 3, "memory": 6319, "step": 9284} +{"lr": 8.88202597160652e-05, "data_time": 0.5279091358184814, "loss": 0.23519654124975203, "time": 0.6715782403945922, "epoch": 3, "memory": 6319, "step": 9384} +{"lr": 8.859136276269775e-05, "data_time": 0.5629609584808349, "loss": 0.23769256919622422, "time": 0.7038479328155518, "epoch": 3, "memory": 6319, "step": 9484} +{"lr": 8.836048586041507e-05, "data_time": 0.5362492084503174, "loss": 0.2327328860759735, "time": 0.676891565322876, "epoch": 3, "memory": 6319, "step": 9584} +{"lr": 8.812764261761037e-05, "data_time": 0.561246395111084, "loss": 0.22969091087579727, "time": 0.7121063232421875, "epoch": 3, "memory": 6319, "step": 9684} +{"lr": 8.789284675857743e-05, "data_time": 0.5059587478637695, "loss": 0.23402496427297592, "time": 0.6462924480438232, "epoch": 3, "memory": 6319, "step": 9784} +{"lr": 8.76561121227013e-05, "data_time": 0.5290754556655883, "loss": 0.23261655420064925, "time": 0.6685675859451294, "epoch": 3, "memory": 6319, "step": 9884} +{"lr": 8.741745266364282e-05, "data_time": 0.5213753700256347, "loss": 0.231000679731369, "time": 0.6652604818344117, "epoch": 3, "memory": 6319, "step": 9984} +{"lr": 8.717688244851635e-05, "data_time": 0.4483925819396973, "loss": 0.24258920401334763, "time": 0.7430192470550537, "epoch": 3, "memory": 6319, "step": 10084} +{"lr": 8.693441565706024e-05, "data_time": 0.3338231801986694, "loss": 0.21221618801355363, "time": 0.6699972867965698, "epoch": 3, "memory": 6319, "step": 10184} +{"lr": 8.669006658080132e-05, "data_time": 0.14651296138763428, "loss": 0.21599060893058777, "time": 0.7080436229705811, "epoch": 3, "memory": 6319, "step": 10284} +{"lr": 8.644384962221226e-05, "data_time": 0.06523208618164063, "loss": 0.23294903635978698, "time": 0.6683426618576049, "epoch": 3, "memory": 6319, "step": 10384} +{"lr": 8.619577929386303e-05, "data_time": 0.16167752742767333, "loss": 0.23386239856481553, "time": 0.7149275064468383, "epoch": 3, "memory": 6319, "step": 10484} +{"lr": 8.594587021756517e-05, "data_time": 0.16182565689086914, "loss": 0.21927913427352905, "time": 0.7035971879959106, "epoch": 3, "memory": 6319, "step": 10584} +{"lr": 8.569413712351023e-05, "data_time": 0.0009621858596801757, "loss": 0.22514743208885193, "time": 0.6861767292022705, "epoch": 3, "memory": 6319, "step": 10684} +{"lr": 8.544059484940119e-05, "data_time": 0.0008236169815063477, "loss": 0.20948270410299302, "time": 0.7535413503646851, "epoch": 3, "memory": 6319, "step": 10784} +{"lr": 8.518525833957818e-05, "data_time": 0.0008296966552734375, "loss": 0.2109537348151207, "time": 0.7048580408096313, "epoch": 3, "memory": 6319, "step": 10884} +{"lr": 8.49281426441377e-05, "data_time": 0.0009437799453735352, "loss": 0.21660107374191284, "time": 0.7745984554290771, "epoch": 3, "memory": 6319, "step": 10984} +{"lr": 8.466926291804501e-05, "data_time": 0.0007861852645874023, "loss": 0.21598206758499144, "time": 0.7172518014907837, "epoch": 3, "memory": 6319, "step": 11084} +{"lr": 8.440863442024147e-05, "data_time": 0.000759124755859375, "loss": 0.2235696107149124, "time": 0.6983484745025634, "epoch": 3, "memory": 6319, "step": 11184} +{"lr": 8.414627251274488e-05, "data_time": 0.000874471664428711, "loss": 0.20743883401155472, "time": 0.738426685333252, "epoch": 3, "memory": 6319, "step": 11284} +{"lr": 8.388219265974407e-05, "data_time": 0.0008832931518554688, "loss": 0.2073444053530693, "time": 0.7331681489944458, "epoch": 3, "memory": 6319, "step": 11384} +{"lr": 8.361641042668715e-05, "data_time": 0.0008800029754638672, "loss": 0.20792239159345627, "time": 0.7300339937210083, "epoch": 3, "memory": 6319, "step": 11484} +{"lr": 8.33489414793644e-05, "data_time": 0.0009262561798095703, "loss": 0.19522866904735564, "time": 0.8190644979476929, "epoch": 3, "memory": 6319, "step": 11584} +{"lr": 8.30798015829848e-05, "data_time": 0.0009125709533691406, "loss": 0.21183450520038605, "time": 0.7052595138549804, "epoch": 3, "memory": 6319, "step": 11684} +{"lr": 8.280900660124665e-05, "data_time": 0.0007672309875488281, "loss": 0.2278122529387474, "time": 1.241529941558838, "epoch": 3, "memory": 6319, "step": 11784} +{"lr": 8.253657249540274e-05, "data_time": 0.0008351564407348633, "loss": 0.21307228058576583, "time": 0.6984730958938599, "epoch": 3, "memory": 6319, "step": 11884} +{"lr": 8.226251532331926e-05, "data_time": 0.000736403465270996, "loss": 0.22321422547101974, "time": 0.7290765523910523, "epoch": 3, "memory": 6319, "step": 11984} +{"lr": 8.198685123852965e-05, "data_time": 0.0008404731750488281, "loss": 0.20979103744029998, "time": 0.738861083984375, "epoch": 3, "memory": 6319, "step": 12084} +{"lr": 8.17095964892824e-05, "data_time": 0.0007280588150024414, "loss": 0.2234665036201477, "time": 0.7249564170837403, "epoch": 3, "memory": 6319, "step": 12184} +{"accuracy/top1": 82.50170135498047, "data_time": 0.3352497100830078, "time": 0.42358460823694866, "step": 3} +{"lr": 8.11728683565904e-05, "data_time": 0.3452092170715332, "loss": 0.2123613715171814, "time": 0.7652813911437988, "epoch": 4, "memory": 6319, "step": 12376} +{"lr": 8.089106273476776e-05, "data_time": 0.16974735260009766, "loss": 0.2119458019733429, "time": 0.712868070602417, "epoch": 4, "memory": 6319, "step": 12476} +{"lr": 8.060773103667566e-05, "data_time": 0.2002340078353882, "loss": 0.20567522943019867, "time": 0.6693098306655884, "epoch": 4, "memory": 6319, "step": 12576} +{"lr": 8.03228899625083e-05, "data_time": 0.0007893562316894532, "loss": 0.21832259595394135, "time": 0.7679970741271973, "epoch": 4, "memory": 6319, "step": 12676} +{"lr": 8.003655630142627e-05, "data_time": 0.0007112979888916016, "loss": 0.20089872777462006, "time": 0.7451218128204345, "epoch": 4, "memory": 6319, "step": 12776} +{"lr": 7.974874693056617e-05, "data_time": 0.013694691658020019, "loss": 0.20858096480369567, "time": 0.7786278486251831, "epoch": 4, "memory": 6319, "step": 12876} +{"lr": 7.945947881404618e-05, "data_time": 0.00084991455078125, "loss": 0.2153647467494011, "time": 0.7001819372177124, "epoch": 4, "memory": 6319, "step": 12976} +{"lr": 7.916876900196646e-05, "data_time": 0.0008081674575805664, "loss": 0.20742841958999633, "time": 0.7613331794738769, "epoch": 4, "memory": 6319, "step": 13076} +{"lr": 7.887663462940359e-05, "data_time": 0.0009556055068969726, "loss": 0.2295478105545044, "time": 0.7719241619110108, "epoch": 4, "memory": 6319, "step": 13176} +{"lr": 7.858309291540111e-05, "data_time": 0.0007396697998046875, "loss": 0.21536858528852462, "time": 0.6818386554718018, "epoch": 4, "memory": 6319, "step": 13276} +{"lr": 7.828816116195411e-05, "data_time": 0.0007997512817382813, "loss": 0.20525012463331221, "time": 0.7383459091186524, "epoch": 4, "memory": 6319, "step": 13376} +{"lr": 7.799185675298994e-05, "data_time": 0.000770401954650879, "loss": 0.20376997888088227, "time": 0.747047758102417, "epoch": 4, "memory": 6319, "step": 13476} +{"lr": 7.769419715334297e-05, "data_time": 0.0008090019226074218, "loss": 0.19351460486650468, "time": 0.7959878444671631, "epoch": 4, "memory": 6319, "step": 13576} +{"lr": 7.739519990772583e-05, "data_time": 0.0008514881134033203, "loss": 0.1868076041340828, "time": 0.7113405942916871, "epoch": 4, "memory": 6319, "step": 13676} +{"lr": 7.709488263969475e-05, "data_time": 0.0007879972457885742, "loss": 0.20743744373321532, "time": 0.717255973815918, "epoch": 4, "memory": 6319, "step": 13776} +{"lr": 7.67932630506109e-05, "data_time": 0.0008163213729858399, "loss": 0.20504215657711028, "time": 0.7047784566879273, "epoch": 4, "memory": 6319, "step": 13876} +{"lr": 7.64903589185975e-05, "data_time": 0.0010724544525146484, "loss": 0.21003948599100114, "time": 0.7431734800338745, "epoch": 4, "memory": 6319, "step": 13976} +{"lr": 7.618618809749134e-05, "data_time": 0.0007904291152954102, "loss": 0.20357482731342316, "time": 0.7548447132110596, "epoch": 4, "memory": 6319, "step": 14076} +{"lr": 7.588076851579028e-05, "data_time": 0.0009006023406982422, "loss": 0.22520921230316163, "time": 0.7574375152587891, "epoch": 4, "memory": 6319, "step": 14176} +{"lr": 7.557411817559755e-05, "data_time": 0.0009590625762939453, "loss": 0.1938050150871277, "time": 0.6884126663208008, "epoch": 4, "memory": 6319, "step": 14276} +{"lr": 7.526625515155957e-05, "data_time": 0.0008490800857543946, "loss": 0.19368332177400588, "time": 0.7812108516693115, "epoch": 4, "memory": 6319, "step": 14376} +{"lr": 7.495719758980122e-05, "data_time": 0.0007518291473388672, "loss": 0.21273910254240036, "time": 0.6806473970413208, "epoch": 4, "memory": 6319, "step": 14476} +{"lr": 7.464696370685586e-05, "data_time": 0.0007924318313598632, "loss": 0.20084236264228822, "time": 0.7339329242706298, "epoch": 4, "memory": 6319, "step": 14576} +{"lr": 7.433557178859198e-05, "data_time": 0.0008692502975463867, "loss": 0.20700835138559343, "time": 0.819854998588562, "epoch": 4, "memory": 6319, "step": 14676} +{"lr": 7.402304018913512e-05, "data_time": 0.0007573366165161133, "loss": 0.20201543271541594, "time": 0.6464768886566162, "epoch": 4, "memory": 6319, "step": 14776} +{"lr": 7.37093873297861e-05, "data_time": 0.000745391845703125, "loss": 0.19180578589439393, "time": 0.6652537822723389, "epoch": 4, "memory": 6319, "step": 14876} +{"lr": 7.339463169793563e-05, "data_time": 0.0008716106414794922, "loss": 0.2002347320318222, "time": 0.7854808330535888, "epoch": 4, "memory": 6319, "step": 14976} +{"lr": 7.307879184597369e-05, "data_time": 0.0008253097534179687, "loss": 0.19215106219053268, "time": 0.7002780437469482, "epoch": 4, "memory": 6319, "step": 15076} +{"lr": 7.276188639019704e-05, "data_time": 0.0011428594589233398, "loss": 0.1828625664114952, "time": 0.6477735996246338, "epoch": 4, "memory": 6319, "step": 15176} +{"lr": 7.244393400971127e-05, "data_time": 0.000862884521484375, "loss": 0.2041605144739151, "time": 0.6513532400131226, "epoch": 4, "memory": 6319, "step": 15276} +{"lr": 7.212495344533006e-05, "data_time": 0.000845789909362793, "loss": 0.20874846130609512, "time": 0.6859585523605347, "epoch": 4, "memory": 6319, "step": 15376} +{"lr": 7.180496349847055e-05, "data_time": 0.0008579492568969727, "loss": 0.19593610018491744, "time": 0.6912116527557373, "epoch": 4, "memory": 6319, "step": 15476} +{"lr": 7.148398303004499e-05, "data_time": 0.0008430242538452149, "loss": 0.19316211193799973, "time": 0.7713447332382202, "epoch": 4, "memory": 6319, "step": 15576} +{"lr": 7.11620309593493e-05, "data_time": 0.00077972412109375, "loss": 0.2021595597267151, "time": 0.6499639034271241, "epoch": 4, "memory": 6319, "step": 15676} +{"lr": 7.083912626294795e-05, "data_time": 0.0008289575576782227, "loss": 0.20438037663698197, "time": 0.6702100753784179, "epoch": 4, "memory": 6319, "step": 15776} +{"lr": 7.051528797355494e-05, "data_time": 0.0008092164993286133, "loss": 0.1811400607228279, "time": 0.7106074810028076, "epoch": 4, "memory": 6319, "step": 15876} +{"lr": 7.019053517891214e-05, "data_time": 0.0007821321487426758, "loss": 0.18542954176664353, "time": 0.6673015594482422, "epoch": 4, "memory": 6319, "step": 15976} +{"lr": 6.986488702066473e-05, "data_time": 0.0007546663284301758, "loss": 0.19507016092538834, "time": 0.6872706174850464, "epoch": 4, "memory": 6319, "step": 16076} +{"lr": 6.953836269323245e-05, "data_time": 0.0008318901062011718, "loss": 0.18438905328512192, "time": 0.6796255588531495, "epoch": 4, "memory": 6319, "step": 16176} +{"lr": 6.921098144267838e-05, "data_time": 0.0008318424224853516, "loss": 0.19302212595939636, "time": 0.7152482986450195, "epoch": 4, "memory": 6319, "step": 16276} +{"accuracy/top1": 87.85231018066406, "data_time": 0.3359352926413218, "time": 0.4237587094306946, "step": 4} +{"lr": 6.858007803673889e-05, "data_time": 0.4231771469116211, "loss": 0.18794462084770203, "time": 0.7526944875717163, "epoch": 5, "memory": 6319, "step": 16468} +{"lr": 6.825030518869046e-05, "data_time": 0.08548085689544678, "loss": 0.18475014269351958, "time": 0.7094476461410523, "epoch": 5, "memory": 6319, "step": 16568} +{"lr": 6.791975133846686e-05, "data_time": 0.0008550167083740234, "loss": 0.18087070882320405, "time": 0.7259903907775879, "epoch": 5, "memory": 6319, "step": 16668} +{"lr": 6.758843596964013e-05, "data_time": 0.0008980512619018554, "loss": 0.18168546110391617, "time": 0.7913861989974975, "epoch": 5, "memory": 6319, "step": 16768} +{"lr": 6.725637861066804e-05, "data_time": 0.0010109424591064453, "loss": 0.19905896335840226, "time": 0.7366887092590332, "epoch": 5, "memory": 6319, "step": 16868} +{"lr": 6.692359883374264e-05, "data_time": 0.0007719278335571289, "loss": 0.17777239829301833, "time": 0.7934297800064087, "epoch": 5, "memory": 6319, "step": 16968} +{"lr": 6.659011625363718e-05, "data_time": 0.0008303165435791015, "loss": 0.18943231999874116, "time": 0.7456103086471557, "epoch": 5, "memory": 6319, "step": 17068} +{"lr": 6.625595052654927e-05, "data_time": 0.0008040904998779297, "loss": 0.19153810441493987, "time": 0.7497786760330201, "epoch": 5, "memory": 6319, "step": 17168} +{"lr": 6.592112134894315e-05, "data_time": 0.0007994651794433593, "loss": 0.19877799451351166, "time": 0.7292736053466797, "epoch": 5, "memory": 6319, "step": 17268} +{"lr": 6.558564845638802e-05, "data_time": 0.008930206298828125, "loss": 0.19411127716302873, "time": 0.7052309274673462, "epoch": 5, "memory": 6319, "step": 17368} +{"lr": 6.524955162239518e-05, "data_time": 0.30195157527923583, "loss": 0.2047702983021736, "time": 0.7157135963439941, "epoch": 5, "memory": 6319, "step": 17468} +{"lr": 6.491285065725238e-05, "data_time": 0.17519099712371827, "loss": 0.18956020325422288, "time": 0.7212751388549805, "epoch": 5, "memory": 6319, "step": 17568} +{"lr": 6.457556540685618e-05, "data_time": 0.10123610496520996, "loss": 0.1927041381597519, "time": 1.1040440559387208, "epoch": 5, "memory": 6319, "step": 17668} +{"lr": 6.423771575154208e-05, "data_time": 0.000852203369140625, "loss": 0.2042357802391052, "time": 0.7295528888702393, "epoch": 5, "memory": 6319, "step": 17768} +{"lr": 6.389932160491309e-05, "data_time": 0.166975998878479, "loss": 0.19824224710464478, "time": 0.7837016582489014, "epoch": 5, "memory": 6319, "step": 17868} +{"lr": 6.356040291266556e-05, "data_time": 0.35793533325195315, "loss": 0.18350497186183928, "time": 0.780746603012085, "epoch": 5, "memory": 6319, "step": 17968} +{"lr": 6.322097965141391e-05, "data_time": 0.22860174179077147, "loss": 0.19847580641508103, "time": 0.7191197633743286, "epoch": 5, "memory": 6319, "step": 18068} +{"lr": 6.288107182751267e-05, "data_time": 0.21714768409729004, "loss": 0.1943136215209961, "time": 0.7471305131912231, "epoch": 5, "memory": 6319, "step": 18168} +{"lr": 6.254069947587782e-05, "data_time": 0.30128490924835205, "loss": 0.18034227788448334, "time": 0.6765732526779175, "epoch": 5, "memory": 6319, "step": 18268} +{"lr": 6.219988265880537e-05, "data_time": 0.0007728815078735351, "loss": 0.18240047991275787, "time": 0.8017641067504883, "epoch": 5, "memory": 6319, "step": 18368} +{"lr": 6.185864146478941e-05, "data_time": 0.0009111881256103516, "loss": 0.1861817166209221, "time": 0.7041740894317627, "epoch": 5, "memory": 6319, "step": 18468} +{"lr": 6.151699600733752e-05, "data_time": 0.0009175777435302734, "loss": 0.18908145129680634, "time": 0.731346845626831, "epoch": 5, "memory": 6319, "step": 18568} +{"lr": 6.117496642378554e-05, "data_time": 0.00077667236328125, "loss": 0.20048560649156572, "time": 0.6893887281417846, "epoch": 5, "memory": 6319, "step": 18668} +{"lr": 6.0832572874110315e-05, "data_time": 0.0008406162261962891, "loss": 0.18763316720724105, "time": 0.7661089420318603, "epoch": 5, "memory": 6319, "step": 18768} +{"lr": 6.0489835539741924e-05, "data_time": 0.0009041786193847656, "loss": 0.17761170342564583, "time": 0.7509448289871216, "epoch": 5, "memory": 6319, "step": 18868} +{"lr": 6.0146774622374006e-05, "data_time": 0.0008766889572143554, "loss": 0.1656336508691311, "time": 0.7639059066772461, "epoch": 5, "memory": 6319, "step": 18968} +{"lr": 5.980341034277274e-05, "data_time": 0.000999140739440918, "loss": 0.18799278885126114, "time": 1.1507911920547484, "epoch": 5, "memory": 6319, "step": 19068} +{"lr": 5.945976293958505e-05, "data_time": 0.0009579658508300781, "loss": 0.18799471259117126, "time": 0.6969701290130615, "epoch": 5, "memory": 6319, "step": 19168} +{"lr": 5.911585266814587e-05, "data_time": 0.0009127140045166015, "loss": 0.18035973310470582, "time": 0.6836655616760254, "epoch": 5, "memory": 6319, "step": 19268} +{"lr": 5.877169979928426e-05, "data_time": 0.0008069038391113281, "loss": 0.16871726214885713, "time": 0.7440181016921997, "epoch": 5, "memory": 6319, "step": 19368} +{"lr": 5.842732461812828e-05, "data_time": 0.0010048627853393554, "loss": 0.17651473730802536, "time": 0.6986016988754272, "epoch": 5, "memory": 6319, "step": 19468} +{"lr": 5.8082747422909646e-05, "data_time": 0.0008463859558105469, "loss": 0.17984080761671067, "time": 0.7480592727661133, "epoch": 5, "memory": 6319, "step": 19568} +{"lr": 5.7737988523767406e-05, "data_time": 0.000822901725769043, "loss": 0.1844061106443405, "time": 0.7214935779571533, "epoch": 5, "memory": 6319, "step": 19668} +{"lr": 5.739306824155042e-05, "data_time": 0.0007722377777099609, "loss": 0.17757899910211564, "time": 0.7331432342529297, "epoch": 5, "memory": 6319, "step": 19768} +{"lr": 5.704800690661993e-05, "data_time": 0.0008005380630493164, "loss": 0.1896573856472969, "time": 0.7413455009460449, "epoch": 5, "memory": 6319, "step": 19868} +{"lr": 5.6702824857650994e-05, "data_time": 0.0008304595947265625, "loss": 0.19663527756929397, "time": 0.7371852397918701, "epoch": 5, "memory": 6319, "step": 19968} +{"lr": 5.635754244043411e-05, "data_time": 0.000824594497680664, "loss": 0.17274728864431382, "time": 0.7180647373199462, "epoch": 5, "memory": 6319, "step": 20068} +{"lr": 5.60121800066754e-05, "data_time": 0.0007601022720336914, "loss": 0.186477167904377, "time": 0.7617677450180054, "epoch": 5, "memory": 6319, "step": 20168} +{"lr": 5.5666757912797516e-05, "data_time": 0.0008510351181030273, "loss": 0.18061977177858352, "time": 0.7368639707565308, "epoch": 5, "memory": 6319, "step": 20268} +{"lr": 5.532129651873955e-05, "data_time": 0.0008313417434692383, "loss": 0.17435094714164734, "time": 0.7056990385055542, "epoch": 5, "memory": 6319, "step": 20368} +{"accuracy/top1": 90.04187774658203, "data_time": 0.33237088123957315, "time": 0.41941052079200747, "step": 5} +{"lr": 5.465797506068693e-05, "data_time": 0.46198620796203616, "loss": 0.19311447143554689, "time": 0.7909291505813598, "epoch": 6, "memory": 6319, "step": 20560} +{"lr": 5.431251545044144e-05, "data_time": 0.29471142292022706, "loss": 0.1744712382555008, "time": 0.7965962886810303, "epoch": 6, "memory": 6319, "step": 20660} +{"lr": 5.396709636204968e-05, "data_time": 0.006396985054016114, "loss": 0.17412697225809098, "time": 0.7236768484115601, "epoch": 6, "memory": 6319, "step": 20760} +{"lr": 5.3621738155273566e-05, "data_time": 0.0007611274719238281, "loss": 0.1695677876472473, "time": 0.7043702840805054, "epoch": 6, "memory": 6319, "step": 20860} +{"lr": 5.3276461186286584e-05, "data_time": 0.0008725643157958985, "loss": 0.18179067671298982, "time": 0.7208564043045044, "epoch": 6, "memory": 6319, "step": 20960} +{"lr": 5.293128580647384e-05, "data_time": 0.0009331226348876954, "loss": 0.18003389686346055, "time": 0.6939608097076416, "epoch": 6, "memory": 6319, "step": 21060} +{"lr": 5.258623236123256e-05, "data_time": 0.001152968406677246, "loss": 0.17001144737005233, "time": 0.7269580364227295, "epoch": 6, "memory": 6319, "step": 21160} +{"lr": 5.224132118877281e-05, "data_time": 0.0008795738220214843, "loss": 0.1735054597258568, "time": 0.7369820356369019, "epoch": 6, "memory": 6319, "step": 21260} +{"lr": 5.189657261891887e-05, "data_time": 0.0008654594421386719, "loss": 0.18412974625825881, "time": 0.7385596990585327, "epoch": 6, "memory": 6319, "step": 21360} +{"lr": 5.1552006971910974e-05, "data_time": 0.000860285758972168, "loss": 0.16710021048784257, "time": 0.9760957717895508, "epoch": 6, "memory": 6319, "step": 21460} +{"lr": 5.120764455720728e-05, "data_time": 0.0009122371673583984, "loss": 0.18533318638801574, "time": 0.7022658824920655, "epoch": 6, "memory": 6319, "step": 21560} +{"lr": 5.0863505672287034e-05, "data_time": 0.000771188735961914, "loss": 0.1834379807114601, "time": 0.7259957075119019, "epoch": 6, "memory": 6319, "step": 21660} +{"lr": 5.051961060145421e-05, "data_time": 0.0008374452590942383, "loss": 0.17431159615516661, "time": 0.6929921865463257, "epoch": 6, "memory": 6319, "step": 21760} +{"lr": 5.017597961464177e-05, "data_time": 0.000815272331237793, "loss": 0.18364601582288742, "time": 0.6506641387939454, "epoch": 6, "memory": 6319, "step": 21860} +{"lr": 4.983263296621707e-05, "data_time": 0.0009006023406982422, "loss": 0.17091528028249742, "time": 0.7906309366226196, "epoch": 6, "memory": 6319, "step": 21960} +{"lr": 4.948959089378785e-05, "data_time": 0.0009644746780395508, "loss": 0.15937474891543388, "time": 0.6765654802322387, "epoch": 6, "memory": 6319, "step": 22060} +{"lr": 4.91468736170094e-05, "data_time": 0.0008697509765625, "loss": 0.19052128940820695, "time": 0.7719676494598389, "epoch": 6, "memory": 6319, "step": 22160} +{"lr": 4.880450133639314e-05, "data_time": 0.0009163141250610352, "loss": 0.1802923262119293, "time": 0.7350878715515137, "epoch": 6, "memory": 6319, "step": 22260} +{"lr": 4.8462494232115335e-05, "data_time": 0.0009006500244140625, "loss": 0.18618869334459304, "time": 0.6649278163909912, "epoch": 6, "memory": 6319, "step": 22360} +{"lr": 4.812087246282806e-05, "data_time": 0.0008510828018188476, "loss": 0.16936117336153983, "time": 0.7321164846420288, "epoch": 6, "memory": 6319, "step": 22460} +{"lr": 4.7779656164470954e-05, "data_time": 0.0007488727569580078, "loss": 0.17743580937385559, "time": 0.6986096143722534, "epoch": 6, "memory": 6319, "step": 22560} +{"lr": 4.7438865449084246e-05, "data_time": 0.0007977962493896484, "loss": 0.17333071529865265, "time": 0.6732075214385986, "epoch": 6, "memory": 6319, "step": 22660} +{"lr": 4.709852040362338e-05, "data_time": 0.0010052919387817383, "loss": 0.1711229793727398, "time": 0.6546087265014648, "epoch": 6, "memory": 6319, "step": 22760} +{"lr": 4.675864108877499e-05, "data_time": 0.0009387969970703125, "loss": 0.17113893181085588, "time": 0.6797942399978638, "epoch": 6, "memory": 6319, "step": 22860} +{"lr": 4.641924753777464e-05, "data_time": 0.0009348154067993164, "loss": 0.19200259745121, "time": 0.7295567035675049, "epoch": 6, "memory": 6319, "step": 22960} +{"lr": 4.608035975522578e-05, "data_time": 0.0009101390838623047, "loss": 0.16151309609413148, "time": 0.7308872699737549, "epoch": 6, "memory": 6319, "step": 23060} +{"lr": 4.574199771592085e-05, "data_time": 0.0011149406433105468, "loss": 0.18423908054828644, "time": 0.7304690837860107, "epoch": 6, "memory": 6319, "step": 23160} +{"lr": 4.540418136366372e-05, "data_time": 0.0008737087249755859, "loss": 0.16510149985551834, "time": 0.7066318035125733, "epoch": 6, "memory": 6319, "step": 23260} +{"lr": 4.506693061009437e-05, "data_time": 0.0009009361267089844, "loss": 0.17288915663957596, "time": 0.6536128520965576, "epoch": 6, "memory": 6319, "step": 23360} +{"lr": 4.473026533351511e-05, "data_time": 0.0008421897888183594, "loss": 0.16139207482337953, "time": 0.7351397514343262, "epoch": 6, "memory": 6319, "step": 23460} +{"lr": 4.439420537771888e-05, "data_time": 0.0009276628494262696, "loss": 0.16605090945959092, "time": 0.6639213800430298, "epoch": 6, "memory": 6319, "step": 23560} +{"lr": 4.405877055081978e-05, "data_time": 0.0014353513717651368, "loss": 0.1747357040643692, "time": 0.7033367872238159, "epoch": 6, "memory": 6319, "step": 23660} +{"lr": 4.3723980624085373e-05, "data_time": 0.0012249469757080079, "loss": 0.16790425330400466, "time": 0.6926178693771362, "epoch": 6, "memory": 6319, "step": 23760} +{"lr": 4.3389855330771394e-05, "data_time": 0.0008136272430419922, "loss": 0.17936118245124816, "time": 0.6703760147094726, "epoch": 6, "memory": 6319, "step": 23860} +{"lr": 4.305641436495869e-05, "data_time": 0.1425553560256958, "loss": 0.20069527477025986, "time": 0.7031758308410645, "epoch": 6, "memory": 6319, "step": 23960} +{"lr": 4.272367738039222e-05, "data_time": 0.12170403003692627, "loss": 0.1538866363465786, "time": 0.6975866794586182, "epoch": 6, "memory": 6319, "step": 24060} +{"lr": 4.239166398932278e-05, "data_time": 0.2130486249923706, "loss": 0.16621210426092148, "time": 0.7111610889434814, "epoch": 6, "memory": 6319, "step": 24160} +{"lr": 4.206039376135108e-05, "data_time": 0.2057497262954712, "loss": 0.16290929019451142, "time": 0.7096094846725464, "epoch": 6, "memory": 6319, "step": 24260} +{"lr": 4.1729886222273985e-05, "data_time": 0.3035950899124146, "loss": 0.1887648656964302, "time": 0.7003968715667724, "epoch": 6, "memory": 6319, "step": 24360} +{"lr": 4.140016085293389e-05, "data_time": 0.20308949947357177, "loss": 0.19095413088798524, "time": 0.6330536127090454, "epoch": 6, "memory": 6319, "step": 24460} +{"accuracy/top1": 91.29305267333984, "data_time": 0.31512669920921327, "time": 0.4052566925684611, "step": 6} +{"lr": 4.0769351868715556e-05, "data_time": 0.6124338865280151, "loss": 0.16141433119773865, "time": 0.7606054782867432, "epoch": 7, "memory": 6319, "step": 24652} +{"lr": 4.044202148988293e-05, "data_time": 0.5981223106384277, "loss": 0.1813565731048584, "time": 0.7393783807754517, "epoch": 7, "memory": 6319, "step": 24752} +{"lr": 4.0115549190371644e-05, "data_time": 0.56242835521698, "loss": 0.16964461356401445, "time": 0.7032805919647217, "epoch": 7, "memory": 6319, "step": 24852} +{"lr": 3.978995421317808e-05, "data_time": 0.5471855640411377, "loss": 0.1756830707192421, "time": 0.6877955675125123, "epoch": 7, "memory": 6319, "step": 24952} +{"lr": 3.94652557495873e-05, "data_time": 0.5761992692947387, "loss": 0.17393409907817842, "time": 0.7203640699386596, "epoch": 7, "memory": 6319, "step": 25052} +{"lr": 3.914147293804207e-05, "data_time": 0.28637559413909913, "loss": 0.16538164466619493, "time": 0.6916460037231446, "epoch": 7, "memory": 6319, "step": 25152} +{"lr": 3.881862486301448e-05, "data_time": 0.29096641540527346, "loss": 0.17479893416166306, "time": 0.6537070989608764, "epoch": 7, "memory": 6319, "step": 25252} +{"lr": 3.849673055388122e-05, "data_time": 0.28748221397399903, "loss": 0.1674222767353058, "time": 0.7168793678283691, "epoch": 7, "memory": 6319, "step": 25352} +{"lr": 3.817580898380197e-05, "data_time": 0.48348379135131836, "loss": 0.17319290190935135, "time": 0.6882036209106446, "epoch": 7, "memory": 6319, "step": 25452} +{"lr": 3.7855879068600975e-05, "data_time": 0.49041578769683836, "loss": 0.15639640837907792, "time": 0.7223220109939575, "epoch": 7, "memory": 6319, "step": 25552} +{"lr": 3.753695966565216e-05, "data_time": 0.565480089187622, "loss": 0.156770521402359, "time": 0.7047290086746216, "epoch": 7, "memory": 6319, "step": 25652} +{"lr": 3.7219069572767636e-05, "data_time": 0.5666501522064209, "loss": 0.18828204572200774, "time": 0.7078155994415283, "epoch": 7, "memory": 6319, "step": 25752} +{"lr": 3.6902227527089725e-05, "data_time": 0.6203948736190796, "loss": 0.1799191102385521, "time": 0.7598810434341431, "epoch": 7, "memory": 6319, "step": 25852} +{"lr": 3.65864522039865e-05, "data_time": 0.5880690336227417, "loss": 0.1735661044716835, "time": 0.7283035278320312, "epoch": 7, "memory": 6319, "step": 25952} +{"lr": 3.627176221595109e-05, "data_time": 0.5464682817459107, "loss": 0.1626366138458252, "time": 0.6872701644897461, "epoch": 7, "memory": 6319, "step": 26052} +{"lr": 3.595817611150461e-05, "data_time": 0.5690017938613892, "loss": 0.1855938032269478, "time": 0.7093132019042969, "epoch": 7, "memory": 6319, "step": 26152} +{"lr": 3.5645712374102824e-05, "data_time": 0.5577040910720825, "loss": 0.16719404011964797, "time": 0.6989094734191894, "epoch": 7, "memory": 6319, "step": 26252} +{"lr": 3.533438942104663e-05, "data_time": 0.5526257276535034, "loss": 0.16869386434555053, "time": 0.6935862541198731, "epoch": 7, "memory": 6319, "step": 26352} +{"lr": 3.5024225602396704e-05, "data_time": 0.5435844421386719, "loss": 0.15850681364536284, "time": 0.6865845918655396, "epoch": 7, "memory": 6319, "step": 26452} +{"lr": 3.4715239199891735e-05, "data_time": 0.3833840131759644, "loss": 0.1629615105688572, "time": 0.6926536321640014, "epoch": 7, "memory": 6319, "step": 26552} +{"lr": 3.44074484258709e-05, "data_time": 0.26637959480285645, "loss": 0.16717423796653746, "time": 0.6746260643005371, "epoch": 7, "memory": 6319, "step": 26652} +{"lr": 3.4100871422200426e-05, "data_time": 0.43064260482788086, "loss": 0.16795639991760253, "time": 0.7114032506942749, "epoch": 7, "memory": 6319, "step": 26752} +{"lr": 3.379552625920419e-05, "data_time": 0.46738812923431394, "loss": 0.17410181760787963, "time": 0.689372181892395, "epoch": 7, "memory": 6319, "step": 26852} +{"lr": 3.349143093459868e-05, "data_time": 0.5961261510848999, "loss": 0.19399193972349166, "time": 0.7370005130767823, "epoch": 7, "memory": 6319, "step": 26952} +{"lr": 3.318860337243226e-05, "data_time": 0.5339762687683105, "loss": 0.1768278032541275, "time": 0.676361870765686, "epoch": 7, "memory": 6319, "step": 27052} +{"lr": 3.288706142202845e-05, "data_time": 0.5447427034378052, "loss": 0.17543355897068977, "time": 0.6852931976318359, "epoch": 7, "memory": 6319, "step": 27152} +{"lr": 3.25868228569341e-05, "data_time": 0.5489830493927002, "loss": 0.16413650959730147, "time": 0.6966290950775147, "epoch": 7, "memory": 6319, "step": 27252} +{"lr": 3.228790537387148e-05, "data_time": 0.5382426500320434, "loss": 0.16559230983257295, "time": 0.6792003154754639, "epoch": 7, "memory": 6319, "step": 27352} +{"lr": 3.199032659169556e-05, "data_time": 0.5310146093368531, "loss": 0.17347155064344405, "time": 0.6708183288574219, "epoch": 7, "memory": 6319, "step": 27452} +{"lr": 3.169410405035527e-05, "data_time": 0.4976022958755493, "loss": 0.1528003215789795, "time": 0.6380467891693116, "epoch": 7, "memory": 6319, "step": 27552} +{"lr": 3.139925520985965e-05, "data_time": 0.5412569046020508, "loss": 0.1775051087141037, "time": 0.6814433097839355, "epoch": 7, "memory": 6319, "step": 27652} +{"lr": 3.1105797449248916e-05, "data_time": 0.5962021589279175, "loss": 0.15354120805859567, "time": 0.7419838666915893, "epoch": 7, "memory": 6319, "step": 27752} +{"lr": 3.081374806556986e-05, "data_time": 0.5891196489334106, "loss": 0.16502247899770736, "time": 0.7288280725479126, "epoch": 7, "memory": 6319, "step": 27852} +{"lr": 3.052312427285649e-05, "data_time": 0.5337841033935546, "loss": 0.1655918374657631, "time": 0.6746386766433716, "epoch": 7, "memory": 6319, "step": 27952} +{"lr": 3.0233943201115213e-05, "data_time": 0.5015422105789185, "loss": 0.1782305881381035, "time": 0.6422603607177735, "epoch": 7, "memory": 6319, "step": 28052} +{"lr": 2.9946221895315525e-05, "data_time": 0.5657840013504029, "loss": 0.17332247346639634, "time": 0.7050846576690674, "epoch": 7, "memory": 6319, "step": 28152} +{"lr": 2.9659977314384953e-05, "data_time": 0.5270125150680542, "loss": 0.1612229660153389, "time": 0.6682371377944947, "epoch": 7, "memory": 6319, "step": 28252} +{"lr": 2.93752263302096e-05, "data_time": 0.5378301620483399, "loss": 0.1648441880941391, "time": 0.6775393009185791, "epoch": 7, "memory": 6319, "step": 28352} +{"lr": 2.9091985726639714e-05, "data_time": 0.5562945365905761, "loss": 0.16484554409980773, "time": 0.697039794921875, "epoch": 7, "memory": 6319, "step": 28452} +{"lr": 2.881027219850035e-05, "data_time": 0.5438944339752197, "loss": 0.16967946738004686, "time": 0.684760332107544, "epoch": 7, "memory": 6319, "step": 28552} +{"accuracy/top1": 91.73493957519531, "data_time": 0.3212118367354075, "time": 0.4096764226754506, "step": 7} +{"lr": 2.827372366659638e-05, "data_time": 0.43496387004852294, "loss": 0.15929679274559022, "time": 0.6692456007003784, "epoch": 8, "memory": 6319, "step": 28744} +{"lr": 2.7996563897301616e-05, "data_time": 0.4415527582168579, "loss": 0.17239178866147994, "time": 0.6596245527267456, "epoch": 8, "memory": 6319, "step": 28844} +{"lr": 2.7720995770013028e-05, "data_time": 0.49218761920928955, "loss": 0.1620272070169449, "time": 0.6672914028167725, "epoch": 8, "memory": 6319, "step": 28944} +{"lr": 2.744703552732299e-05, "data_time": 0.34904003143310547, "loss": 0.16939831748604775, "time": 0.7152575969696044, "epoch": 8, "memory": 6319, "step": 29044} +{"lr": 2.717469931705171e-05, "data_time": 0.4309261322021484, "loss": 0.1526232421398163, "time": 0.7138391971588135, "epoch": 8, "memory": 6319, "step": 29144} +{"lr": 2.690400319129529e-05, "data_time": 0.15989320278167723, "loss": 0.17122574746608735, "time": 0.6754186630249024, "epoch": 8, "memory": 6319, "step": 29244} +{"lr": 2.6634963105479716e-05, "data_time": 0.07138471603393555, "loss": 0.1931123211979866, "time": 0.6669699907302856, "epoch": 8, "memory": 6319, "step": 29344} +{"lr": 2.6367594917420177e-05, "data_time": 0.1935683012008667, "loss": 0.16052377671003343, "time": 0.7234967708587646, "epoch": 8, "memory": 6319, "step": 29444} +{"lr": 2.610191438638673e-05, "data_time": 0.34243483543395997, "loss": 0.16379099935293198, "time": 0.7862113237380981, "epoch": 8, "memory": 6319, "step": 29544} +{"lr": 2.5837937172175032e-05, "data_time": 0.0009986400604248048, "loss": 0.17309513613581656, "time": 0.7582850933074952, "epoch": 8, "memory": 6319, "step": 29644} +{"lr": 2.557567883418364e-05, "data_time": 0.0009107112884521485, "loss": 0.16048759445548058, "time": 0.7508659601211548, "epoch": 8, "memory": 6319, "step": 29744} +{"lr": 2.531515483049656e-05, "data_time": 0.0010088682174682617, "loss": 0.16359711214900016, "time": 0.7036072015762329, "epoch": 8, "memory": 6319, "step": 29844} +{"lr": 2.5056380516972453e-05, "data_time": 0.0010594844818115235, "loss": 0.1691916584968567, "time": 0.6536807298660279, "epoch": 8, "memory": 6319, "step": 29944} +{"lr": 2.4799371146339435e-05, "data_time": 0.0011473894119262695, "loss": 0.1659874677658081, "time": 0.7157558679580689, "epoch": 8, "memory": 6319, "step": 30044} +{"lr": 2.4544141867295954e-05, "data_time": 0.0008456945419311523, "loss": 0.1564167097210884, "time": 0.6811789035797119, "epoch": 8, "memory": 6319, "step": 30144} +{"lr": 2.429070772361796e-05, "data_time": 0.0009846210479736329, "loss": 0.17010714411735534, "time": 0.7032801389694214, "epoch": 8, "memory": 6319, "step": 30244} +{"lr": 2.4039083653272075e-05, "data_time": 0.0010410070419311524, "loss": 0.1742545932531357, "time": 0.7389954328536987, "epoch": 8, "memory": 6319, "step": 30344} +{"lr": 2.378928448753535e-05, "data_time": 0.0008489370346069336, "loss": 0.1634024515748024, "time": 0.6612615823745728, "epoch": 8, "memory": 6319, "step": 30444} +{"lr": 2.354132495012082e-05, "data_time": 0.0011972665786743164, "loss": 0.17912216931581498, "time": 0.7069468021392822, "epoch": 8, "memory": 6319, "step": 30544} +{"lr": 2.329521965630983e-05, "data_time": 0.0008386611938476562, "loss": 0.16748422980308533, "time": 0.7016327142715454, "epoch": 8, "memory": 6319, "step": 30644} +{"lr": 2.3050983112090472e-05, "data_time": 0.0009271621704101563, "loss": 0.18147973343729973, "time": 0.7067485332489014, "epoch": 8, "memory": 6319, "step": 30744} +{"lr": 2.2808629713302708e-05, "data_time": 0.0009601831436157227, "loss": 0.15142384618520738, "time": 0.7132920503616333, "epoch": 8, "memory": 6319, "step": 30844} +{"lr": 2.2568173744789706e-05, "data_time": 0.0009085655212402344, "loss": 0.15970299243927003, "time": 0.6975794076919556, "epoch": 8, "memory": 6319, "step": 30944} +{"lr": 2.2329629379555897e-05, "data_time": 0.0009399890899658203, "loss": 0.18512292504310607, "time": 0.9630717515945435, "epoch": 8, "memory": 6319, "step": 31044} +{"lr": 2.2093010677931667e-05, "data_time": 0.0009394168853759765, "loss": 0.16732430309057236, "time": 0.7877877712249756, "epoch": 8, "memory": 6319, "step": 31144} +{"lr": 2.1858331586744434e-05, "data_time": 0.0009273529052734375, "loss": 0.16037182062864302, "time": 0.7212547779083252, "epoch": 8, "memory": 6319, "step": 31244} +{"lr": 2.1625605938496814e-05, "data_time": 0.0010665416717529296, "loss": 0.16246872991323472, "time": 0.6757916927337646, "epoch": 8, "memory": 6319, "step": 31344} +{"lr": 2.1394847450551186e-05, "data_time": 0.0010218381881713866, "loss": 0.17783166021108626, "time": 0.7036640882492066, "epoch": 8, "memory": 6319, "step": 31444} +{"lr": 2.1166069724321078e-05, "data_time": 0.0009870052337646485, "loss": 0.17901988476514816, "time": 0.6738017797470093, "epoch": 8, "memory": 6319, "step": 31544} +{"lr": 2.0939286244469643e-05, "data_time": 0.0009018898010253907, "loss": 0.1610145889222622, "time": 0.6912663459777832, "epoch": 8, "memory": 6319, "step": 31644} +{"lr": 2.071451037811473e-05, "data_time": 0.0007386684417724609, "loss": 0.17289062291383744, "time": 0.7477646350860596, "epoch": 8, "memory": 6319, "step": 31744} +{"lr": 2.0491755374041045e-05, "data_time": 0.000998210906982422, "loss": 0.19977452903985976, "time": 0.7184840440750122, "epoch": 8, "memory": 6319, "step": 31844} +{"lr": 2.0271034361919193e-05, "data_time": 0.0008982658386230469, "loss": 0.17608511745929717, "time": 0.7237038612365723, "epoch": 8, "memory": 6319, "step": 31944} +{"lr": 2.0052360351531814e-05, "data_time": 0.0008471965789794922, "loss": 0.17564684599637986, "time": 0.7074095010757446, "epoch": 8, "memory": 6319, "step": 32044} +{"lr": 1.983574623200682e-05, "data_time": 0.0008153915405273438, "loss": 0.15017342045903206, "time": 0.7198257923126221, "epoch": 8, "memory": 6319, "step": 32144} +{"lr": 1.9621204771057533e-05, "data_time": 0.0008871078491210938, "loss": 0.1591176211833954, "time": 0.7024089813232421, "epoch": 8, "memory": 6319, "step": 32244} +{"lr": 1.940874861423025e-05, "data_time": 0.0008514404296875, "loss": 0.18107160776853562, "time": 0.7456432104110717, "epoch": 8, "memory": 6319, "step": 32344} +{"lr": 1.9198390284158783e-05, "data_time": 0.0008984088897705078, "loss": 0.16974645629525184, "time": 1.039924430847168, "epoch": 8, "memory": 6319, "step": 32444} +{"lr": 1.8990142179826573e-05, "data_time": 0.0008800745010375977, "loss": 0.17854541093111037, "time": 0.7250747203826904, "epoch": 8, "memory": 6319, "step": 32544} +{"lr": 1.8784016575835568e-05, "data_time": 0.0008195638656616211, "loss": 0.16246017515659333, "time": 0.7141840696334839, "epoch": 8, "memory": 6319, "step": 32644} +{"accuracy/top1": 92.90667724609375, "data_time": 0.3211768865585327, "time": 0.40820035139719646, "step": 8} +{"lr": 1.8394249604901642e-05, "data_time": 0.18729810714721679, "loss": 0.15460986346006395, "time": 0.7207031011581421, "epoch": 9, "memory": 6319, "step": 32836} +{"lr": 1.81943907744488e-05, "data_time": 0.18993062973022462, "loss": 0.16875767558813096, "time": 0.6815762996673584, "epoch": 9, "memory": 6319, "step": 32936} +{"lr": 1.7996701347677866e-05, "data_time": 0.31720824241638185, "loss": 0.1572006583213806, "time": 0.6806392431259155, "epoch": 9, "memory": 6319, "step": 33036} +{"lr": 1.780119297683906e-05, "data_time": 0.3877443313598633, "loss": 0.1749775841832161, "time": 0.7137496709823609, "epoch": 9, "memory": 6319, "step": 33136} +{"lr": 1.7607877185626483e-05, "data_time": 0.294804310798645, "loss": 0.1448626011610031, "time": 0.6825323104858398, "epoch": 9, "memory": 6319, "step": 33236} +{"lr": 1.7416765368498675e-05, "data_time": 0.16448497772216797, "loss": 0.17938916236162186, "time": 0.705608582496643, "epoch": 9, "memory": 6319, "step": 33336} +{"lr": 1.7227868790007105e-05, "data_time": 0.0008262157440185547, "loss": 0.17950266003608703, "time": 0.7025579690933228, "epoch": 9, "memory": 6319, "step": 33436} +{"lr": 1.7041198584132233e-05, "data_time": 0.0009088516235351562, "loss": 0.16385626047849655, "time": 0.7037282705307006, "epoch": 9, "memory": 6319, "step": 33536} +{"lr": 1.6856765753627146e-05, "data_time": 0.24636623859405518, "loss": 0.1574540063738823, "time": 0.7312356233596802, "epoch": 9, "memory": 6319, "step": 33636} +{"lr": 1.6674581169369163e-05, "data_time": 0.15081188678741456, "loss": 0.14947786033153534, "time": 0.6530324459075928, "epoch": 9, "memory": 6319, "step": 33736} +{"lr": 1.6494655569718982e-05, "data_time": 0.2675036430358887, "loss": 0.17042155265808107, "time": 0.6659437894821167, "epoch": 9, "memory": 6319, "step": 33836} +{"lr": 1.6316999559887786e-05, "data_time": 0.2079176664352417, "loss": 0.1611701712012291, "time": 0.7437841415405273, "epoch": 9, "memory": 6319, "step": 33936} +{"lr": 1.6141623611312105e-05, "data_time": 0.03393511772155762, "loss": 0.1829184889793396, "time": 0.714853572845459, "epoch": 9, "memory": 6319, "step": 34036} +{"lr": 1.5968538061036705e-05, "data_time": 0.0009735345840454102, "loss": 0.15594484135508538, "time": 0.7060147047042846, "epoch": 9, "memory": 6319, "step": 34136} +{"lr": 1.5797753111105157e-05, "data_time": 0.0008349418640136719, "loss": 0.16882977783679962, "time": 0.6934847831726074, "epoch": 9, "memory": 6319, "step": 34236} +{"lr": 1.5629278827958574e-05, "data_time": 0.0009967327117919923, "loss": 0.16971203088760375, "time": 0.6836693525314331, "epoch": 9, "memory": 6319, "step": 34336} +{"lr": 1.5463125141842392e-05, "data_time": 0.0009215116500854492, "loss": 0.1715298682451248, "time": 0.7374971151351929, "epoch": 9, "memory": 6319, "step": 34436} +{"lr": 1.5299301846220796e-05, "data_time": 0.00105438232421875, "loss": 0.17847218066453935, "time": 0.7000346899032592, "epoch": 9, "memory": 6319, "step": 34536} +{"lr": 1.5137818597199751e-05, "data_time": 0.0008035182952880859, "loss": 0.1638461098074913, "time": 0.7424283027648926, "epoch": 9, "memory": 6319, "step": 34636} +{"lr": 1.4978684912957622e-05, "data_time": 0.0007372379302978515, "loss": 0.15982463583350182, "time": 0.6368038415908813, "epoch": 9, "memory": 6319, "step": 34736} +{"lr": 1.4821910173184321e-05, "data_time": 0.0008972406387329102, "loss": 0.1588974714279175, "time": 0.6889965295791626, "epoch": 9, "memory": 6319, "step": 34836} +{"lr": 1.4667503618528337e-05, "data_time": 0.0008920431137084961, "loss": 0.16023080199956893, "time": 0.7143422365188599, "epoch": 9, "memory": 6319, "step": 34936} +{"lr": 1.4515474350052097e-05, "data_time": 0.0008902549743652344, "loss": 0.15848468691110612, "time": 0.7422490358352661, "epoch": 9, "memory": 6319, "step": 35036} +{"lr": 1.4365831328695616e-05, "data_time": 0.0008068323135375976, "loss": 0.1650614470243454, "time": 0.7103931188583374, "epoch": 9, "memory": 6319, "step": 35136} +{"lr": 1.4218583374748169e-05, "data_time": 0.0008558034896850586, "loss": 0.1632140651345253, "time": 0.7484535932540893, "epoch": 9, "memory": 6319, "step": 35236} +{"lr": 1.4073739167328542e-05, "data_time": 0.0010104894638061524, "loss": 0.15982672795653344, "time": 0.6741731882095336, "epoch": 9, "memory": 6319, "step": 35336} +{"lr": 1.3931307243873344e-05, "data_time": 0.0009338140487670898, "loss": 0.16906670555472375, "time": 0.6640015125274659, "epoch": 9, "memory": 6319, "step": 35436} +{"lr": 1.3791295999633911e-05, "data_time": 0.0010547876358032227, "loss": 0.15604328289628028, "time": 0.6555068731307984, "epoch": 9, "memory": 6319, "step": 35536} +{"lr": 1.3653713687181394e-05, "data_time": 0.0010912418365478516, "loss": 0.1809113949537277, "time": 0.6874321699142456, "epoch": 9, "memory": 6319, "step": 35636} +{"lr": 1.3518568415920337e-05, "data_time": 0.0009035348892211914, "loss": 0.17026009708642958, "time": 0.7585430383682251, "epoch": 9, "memory": 6319, "step": 35736} +{"lr": 1.3385868151610723e-05, "data_time": 0.20730414390563964, "loss": 0.1638484090566635, "time": 0.7466943740844727, "epoch": 9, "memory": 6319, "step": 35836} +{"lr": 1.3255620715898442e-05, "data_time": 0.18925862312316893, "loss": 0.16586991995573044, "time": 0.7194545030593872, "epoch": 9, "memory": 6319, "step": 35936} +{"lr": 1.3127833785854261e-05, "data_time": 0.0010066032409667969, "loss": 0.1620241478085518, "time": 0.7628704071044922, "epoch": 9, "memory": 6319, "step": 36036} +{"lr": 1.3002514893521313e-05, "data_time": 0.0009039878845214844, "loss": 0.16661669462919235, "time": 0.7198026180267334, "epoch": 9, "memory": 6319, "step": 36136} +{"lr": 1.287967142547117e-05, "data_time": 0.000866556167602539, "loss": 0.15979148373007773, "time": 0.7427562475204468, "epoch": 9, "memory": 6319, "step": 36236} +{"lr": 1.2759310622368435e-05, "data_time": 0.000870656967163086, "loss": 0.15942969620227815, "time": 0.7897002935409546, "epoch": 9, "memory": 6319, "step": 36336} +{"lr": 1.2641439578543968e-05, "data_time": 0.0008477210998535156, "loss": 0.16918978691101075, "time": 0.6903006792068481, "epoch": 9, "memory": 6319, "step": 36436} +{"lr": 1.252606524157674e-05, "data_time": 0.0008438825607299805, "loss": 0.16622457951307296, "time": 0.6901096343994141, "epoch": 9, "memory": 6319, "step": 36536} +{"lr": 1.2413194411884297e-05, "data_time": 0.0008791923522949219, "loss": 0.17171796560287475, "time": 0.68916335105896, "epoch": 9, "memory": 6319, "step": 36636} +{"lr": 1.2302833742321981e-05, "data_time": 0.0008104562759399414, "loss": 0.1667586825788021, "time": 0.7203540563583374, "epoch": 9, "memory": 6319, "step": 36736} +{"accuracy/top1": 93.01590728759766, "data_time": 0.31340978542963666, "time": 0.40285993019739785, "step": 9} +{"lr": 1.2098001439139718e-05, "data_time": 0.09869544506072998, "loss": 0.16758147701621057, "time": 0.6929279088973999, "epoch": 10, "memory": 6319, "step": 36928} +{"lr": 1.1995007122352056e-05, "data_time": 0.20224294662475586, "loss": 0.1677936092019081, "time": 0.7161273002624512, "epoch": 10, "memory": 6319, "step": 37028} +{"lr": 1.1894547614574522e-05, "data_time": 0.000911259651184082, "loss": 0.1425803378224373, "time": 0.7055025577545166, "epoch": 10, "memory": 6319, "step": 37128} +{"lr": 1.1796628837111806e-05, "data_time": 0.0008507013320922851, "loss": 0.1644194945693016, "time": 0.7238788604736328, "epoch": 10, "memory": 6319, "step": 37228} +{"lr": 1.1701256561512343e-05, "data_time": 0.0008981466293334961, "loss": 0.16315096914768218, "time": 0.6704913854599, "epoch": 10, "memory": 6319, "step": 37328} +{"lr": 1.1608436409228136e-05, "data_time": 0.0009629487991333008, "loss": 0.18481188416481018, "time": 0.6627022743225097, "epoch": 10, "memory": 6319, "step": 37428} +{"lr": 1.1518173851283401e-05, "data_time": 0.0008953094482421875, "loss": 0.1533253327012062, "time": 0.6897203207015992, "epoch": 10, "memory": 6319, "step": 37528} +{"lr": 1.1430474207952176e-05, "data_time": 0.0009173393249511719, "loss": 0.14821891188621522, "time": 0.7070162296295166, "epoch": 10, "memory": 6319, "step": 37628} +{"lr": 1.1345342648444593e-05, "data_time": 0.001033473014831543, "loss": 0.14969507306814195, "time": 0.7723404169082642, "epoch": 10, "memory": 6319, "step": 37728} +{"lr": 1.12627841906023e-05, "data_time": 0.0008768081665039062, "loss": 0.15251275449991225, "time": 0.6891923427581788, "epoch": 10, "memory": 6319, "step": 37828} +{"lr": 1.1182803700602659e-05, "data_time": 0.0009001731872558594, "loss": 0.15483831539750098, "time": 0.6577965259552002, "epoch": 10, "memory": 6319, "step": 37928} +{"lr": 1.110540589267192e-05, "data_time": 0.0010291576385498048, "loss": 0.150139519572258, "time": 0.6490852355957031, "epoch": 10, "memory": 6319, "step": 38028} +{"lr": 1.1030595328807396e-05, "data_time": 0.0008701562881469726, "loss": 0.15066432654857637, "time": 0.7208055257797241, "epoch": 10, "memory": 6319, "step": 38128} +{"lr": 1.0958376418508494e-05, "data_time": 0.0010786056518554688, "loss": 0.16919270902872086, "time": 0.6935332775115967, "epoch": 10, "memory": 6319, "step": 38228} +{"lr": 1.088875341851686e-05, "data_time": 0.001039886474609375, "loss": 0.1632241539657116, "time": 0.6928199768066406, "epoch": 10, "memory": 6319, "step": 38328} +{"lr": 1.0821730432565486e-05, "data_time": 0.0010054588317871093, "loss": 0.17095913887023925, "time": 0.7007759571075439, "epoch": 10, "memory": 6319, "step": 38428} +{"lr": 1.0757311411136797e-05, "data_time": 0.0009536981582641601, "loss": 0.17949067652225495, "time": 0.7212576150894165, "epoch": 10, "memory": 6319, "step": 38528} +{"lr": 1.0695500151229833e-05, "data_time": 0.0008974075317382812, "loss": 0.15975531786680222, "time": 0.7136516571044922, "epoch": 10, "memory": 6319, "step": 38628} +{"lr": 1.0636300296136373e-05, "data_time": 0.0009805679321289063, "loss": 0.17370918840169908, "time": 0.7227862358093262, "epoch": 10, "memory": 6319, "step": 38728} +{"lr": 1.0579715335226322e-05, "data_time": 0.0008877754211425781, "loss": 0.15194374546408654, "time": 0.6469579935073853, "epoch": 10, "memory": 6319, "step": 38828} +{"lr": 1.0525748603741896e-05, "data_time": 0.0008641958236694336, "loss": 0.1799135535955429, "time": 0.7223961591720581, "epoch": 10, "memory": 6319, "step": 38928} +{"lr": 1.0474403282601119e-05, "data_time": 0.0009482860565185547, "loss": 0.17140405997633934, "time": 0.7346735954284668, "epoch": 10, "memory": 6319, "step": 39028} +{"lr": 1.0425682398210336e-05, "data_time": 0.0009703397750854492, "loss": 0.15233578756451607, "time": 0.697683310508728, "epoch": 10, "memory": 6319, "step": 39128} +{"lr": 1.0379588822285784e-05, "data_time": 0.0009176492691040039, "loss": 0.16907952725887299, "time": 0.7015511274337769, "epoch": 10, "memory": 6319, "step": 39228} +{"lr": 1.0336125271684364e-05, "data_time": 0.0008747339248657227, "loss": 0.16613706797361374, "time": 0.683406400680542, "epoch": 10, "memory": 6319, "step": 39328} +{"lr": 1.0295294308243494e-05, "data_time": 0.0008446693420410157, "loss": 0.16299415975809098, "time": 0.6715137004852295, "epoch": 10, "memory": 6319, "step": 39428} +{"lr": 1.0257098338630086e-05, "data_time": 0.0007773160934448243, "loss": 0.16641470342874526, "time": 0.688904595375061, "epoch": 10, "memory": 6319, "step": 39528} +{"lr": 1.0221539614198729e-05, "data_time": 0.0008801937103271485, "loss": 0.14935189485549927, "time": 0.7031291723251343, "epoch": 10, "memory": 6319, "step": 39628} +{"lr": 1.0188620230858978e-05, "data_time": 0.0008347511291503906, "loss": 0.1692873314023018, "time": 0.7959968566894531, "epoch": 10, "memory": 6319, "step": 39728} +{"lr": 1.01583421289518e-05, "data_time": 0.0007643461227416992, "loss": 0.1614592969417572, "time": 0.6756434440612793, "epoch": 10, "memory": 6319, "step": 39828} +{"lr": 1.0130707093135206e-05, "data_time": 0.0010109663009643555, "loss": 0.15934257507324218, "time": 0.6999173879623413, "epoch": 10, "memory": 6319, "step": 39928} +{"lr": 1.01057167522791e-05, "data_time": 0.0008799314498901367, "loss": 0.17094288468360902, "time": 0.7179607629776001, "epoch": 10, "memory": 6319, "step": 40028} +{"lr": 1.008337257936921e-05, "data_time": 0.0009057283401489258, "loss": 0.1505805179476738, "time": 0.7259350538253784, "epoch": 10, "memory": 6319, "step": 40128} +{"lr": 1.0063675891420294e-05, "data_time": 0.001208662986755371, "loss": 0.15701159834861755, "time": 0.6661712646484375, "epoch": 10, "memory": 6319, "step": 40228} +{"lr": 1.0046627849398548e-05, "data_time": 0.0011827707290649413, "loss": 0.16071770191192628, "time": 0.6966933250427246, "epoch": 10, "memory": 6319, "step": 40328} +{"lr": 1.0032229458153106e-05, "data_time": 0.0008065462112426758, "loss": 0.14961961209774016, "time": 0.9084820985794068, "epoch": 10, "memory": 6319, "step": 40428} +{"lr": 1.0020481566356862e-05, "data_time": 0.0009717702865600586, "loss": 0.15234164893627167, "time": 0.7270920753479004, "epoch": 10, "memory": 6319, "step": 40528} +{"lr": 1.0011384866456456e-05, "data_time": 0.00099637508392334, "loss": 0.15804436951875686, "time": 0.7482522487640381, "epoch": 10, "memory": 6319, "step": 40628} +{"lr": 1.00049398946314e-05, "data_time": 0.0008790493011474609, "loss": 0.16904067248106003, "time": 0.7159698009490967, "epoch": 10, "memory": 6319, "step": 40728} +{"lr": 1.0001147030762536e-05, "data_time": 0.000933384895324707, "loss": 0.15995430648326875, "time": 0.6835578918457031, "epoch": 10, "memory": 6319, "step": 40828} +{"accuracy/top1": 92.8636474609375, "data_time": 0.3246283312638601, "time": 0.4124751389026642, "step": 10} diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/0000000.png_1.png b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/0000000.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..670fc15cc3173587a48a8f5a980913ca84720c9e Binary files /dev/null and b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/0000000.png_1.png differ diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/0000000.png_10.png b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/0000000.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..5013b7c5b8f23777ddcd7f5188f79bbd93bce6b3 Binary files /dev/null and b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/0000000.png_10.png differ diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/0000000.png_2.png b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/0000000.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..6d5b36a7c973b0146ffd29ae6bfafbe9b3962ead Binary files /dev/null and b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/0000000.png_2.png differ diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/0000000.png_3.png b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/0000000.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..8f8d9d9fd1c26fc954157a9a2b05752da03988f3 Binary files /dev/null and b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/0000000.png_3.png differ diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/0000000.png_4.png b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/0000000.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..9fca3aeccf92cd6defef1a7a7b6c9c4359d06b5d Binary files /dev/null and b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/0000000.png_4.png differ diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/0000000.png_5.png b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/0000000.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..998c2a688c02228b9d79f6d3fab4eeded59660fc Binary files /dev/null and b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/0000000.png_5.png differ diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/0000000.png_6.png b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/0000000.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..7563f7246cb7c79d694aabdd1052c0e01e76c894 Binary files /dev/null and b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/0000000.png_6.png differ diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/0000000.png_7.png b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/0000000.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..a4e8bdafa4dda835d71f2aa4d808169b20304d75 Binary files /dev/null and b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/0000000.png_7.png differ diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/0000000.png_8.png b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/0000000.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..bbe1c6302b9c598aa8fa19ea2af7df8f5ac2fc1c Binary files /dev/null and b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/0000000.png_8.png differ diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/0000000.png_9.png b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/0000000.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..5cf98d7b45f3fae9af05d653f423863aa691d1c0 Binary files /dev/null and b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/0000000.png_9.png differ diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/000003093.jpg_1.png b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/000003093.jpg_1.png new file mode 100644 index 0000000000000000000000000000000000000000..dead13c821cb85d97b31841eefa51d4c45dacaa3 Binary files /dev/null and b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/000003093.jpg_1.png differ diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/000003093.jpg_10.png b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/000003093.jpg_10.png new file mode 100644 index 0000000000000000000000000000000000000000..1d4580f296e50ab8339c8eb264bac3654f84d3c4 Binary files /dev/null and b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/000003093.jpg_10.png differ diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/000003093.jpg_2.png b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/000003093.jpg_2.png new file mode 100644 index 0000000000000000000000000000000000000000..aaeb1d76b5194ee661cd16768a35d1e1c66c2d30 Binary files /dev/null and b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/000003093.jpg_2.png differ diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/000003093.jpg_3.png b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/000003093.jpg_3.png new file mode 100644 index 0000000000000000000000000000000000000000..2c2a0d7ade4a79b997ca727f728b3ea2a50b7d94 Binary files /dev/null and b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/000003093.jpg_3.png differ diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/000003093.jpg_4.png b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/000003093.jpg_4.png new file mode 100644 index 0000000000000000000000000000000000000000..f7374369f988b8e904dfa03d8b45f70e5cdd96f3 Binary files /dev/null and b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/000003093.jpg_4.png differ diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/000003093.jpg_5.png b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/000003093.jpg_5.png new file mode 100644 index 0000000000000000000000000000000000000000..c4eadc87c65bad14dbad18400f8742de6431285c Binary files /dev/null and b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/000003093.jpg_5.png differ diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/000003093.jpg_6.png b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/000003093.jpg_6.png new file mode 100644 index 0000000000000000000000000000000000000000..84d36a116137efee0bfb59b819616f219c542661 Binary files /dev/null and b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/000003093.jpg_6.png differ diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/000003093.jpg_7.png b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/000003093.jpg_7.png new file mode 100644 index 0000000000000000000000000000000000000000..7d0e2f69cdbebb1ad7bc97c1b57b274c1ab2c9a1 Binary files /dev/null and b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/000003093.jpg_7.png differ diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/000003093.jpg_8.png b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/000003093.jpg_8.png new file mode 100644 index 0000000000000000000000000000000000000000..fadbd48c4a5364a37912291a70a85416cde1f4fc Binary files /dev/null and b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/000003093.jpg_8.png differ diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/000003093.jpg_9.png b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/000003093.jpg_9.png new file mode 100644 index 0000000000000000000000000000000000000000..4009e46419bfd9d40d4d82ae0332c8ba9c2691fe Binary files /dev/null and b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/000003093.jpg_9.png differ diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/000015337.jpg_1.png b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/000015337.jpg_1.png new file mode 100644 index 0000000000000000000000000000000000000000..7ff22a047a61492da05c84e78338c4459453448f Binary files /dev/null and b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/000015337.jpg_1.png differ diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/000015337.jpg_10.png b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/000015337.jpg_10.png new file mode 100644 index 0000000000000000000000000000000000000000..e70d7809aa099b4c545b659730ab4b10bb7e4c80 Binary files /dev/null and b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/000015337.jpg_10.png differ diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/000015337.jpg_2.png b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/000015337.jpg_2.png new file mode 100644 index 0000000000000000000000000000000000000000..70a53ed9245b5ebd48a8d00d91288ed304de990c Binary files /dev/null and b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/000015337.jpg_2.png differ diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/000015337.jpg_3.png b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/000015337.jpg_3.png new file mode 100644 index 0000000000000000000000000000000000000000..f4eee09ddc7ffefb703edd9c9ae1968b8bdf9b40 Binary files /dev/null and b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/000015337.jpg_3.png differ diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/000015337.jpg_4.png b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/000015337.jpg_4.png new file mode 100644 index 0000000000000000000000000000000000000000..afac7a11170100af6269862f1469b06f1a52757f Binary files /dev/null and b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/000015337.jpg_4.png differ diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/000015337.jpg_5.png b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/000015337.jpg_5.png new file mode 100644 index 0000000000000000000000000000000000000000..b5cda8f6ce3c0a5fa506d1994725b073fb77813a Binary files /dev/null and b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/000015337.jpg_5.png differ diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/000015337.jpg_6.png b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/000015337.jpg_6.png new file mode 100644 index 0000000000000000000000000000000000000000..ab7eeab74d7795e4b54b44c9c23bb65f7ef3b9e2 Binary files /dev/null and b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/000015337.jpg_6.png differ diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/000015337.jpg_7.png b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/000015337.jpg_7.png new file mode 100644 index 0000000000000000000000000000000000000000..6b159fab57f69c1407b3fa77f0bd7ce0be521516 Binary files /dev/null and b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/000015337.jpg_7.png differ diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/000015337.jpg_8.png b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/000015337.jpg_8.png new file mode 100644 index 0000000000000000000000000000000000000000..33c4085440adda95143297de31e8b7266cde2c3d Binary files /dev/null and b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/000015337.jpg_8.png differ diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/000015337.jpg_9.png b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/000015337.jpg_9.png new file mode 100644 index 0000000000000000000000000000000000000000..c126eed325882f503c95534d9e1a054281301385 Binary files /dev/null and b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/000015337.jpg_9.png differ diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/0004159.png_1.png b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/0004159.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..a2543c2f59d17f992dc3911840b70ae091fc1378 Binary files /dev/null and b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/0004159.png_1.png differ diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/0004159.png_10.png b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/0004159.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..e055c951ced42c95d8f07351986d6ee71dbbef86 Binary files /dev/null and b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/0004159.png_10.png differ diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/0004159.png_2.png b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/0004159.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..52c4ee4b0630ce57e336cd60a3b8c6395841e24a Binary files /dev/null and b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/0004159.png_2.png differ diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/0004159.png_3.png b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/0004159.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..c6a49b8c46f45c4064396f655fa4c405a9d854d9 Binary files /dev/null and b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/0004159.png_3.png differ diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/0004159.png_4.png b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/0004159.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..1e58d5dfb581784804ffb7d2b86ad3318e1e0e4a Binary files /dev/null and b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/0004159.png_4.png differ diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/0004159.png_5.png b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/0004159.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..0e1722226deea939494fb73ea1eac001bcad7622 Binary files /dev/null and b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/0004159.png_5.png differ diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/0004159.png_6.png b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/0004159.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..eec811dcf2e04db503713ccdce07baaaed822052 Binary files /dev/null and b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/0004159.png_6.png differ diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/0004159.png_7.png b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/0004159.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..193aab92e592a81f21d8305ad3e96c6c684e35b1 Binary files /dev/null and b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/0004159.png_7.png differ diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/0004159.png_8.png b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/0004159.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..65a6f4aafe8bfe59564879affd3bad75ffd69bab Binary files /dev/null and b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/0004159.png_8.png differ diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/0004159.png_9.png b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/0004159.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..5567663b180e48b279542dc22bf88d8cf0be75a1 Binary files /dev/null and b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/0004159.png_9.png differ diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/0008318.png_1.png b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/0008318.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..dc01ebea4c9db10c4335382972ed7765b09d909f Binary files /dev/null and b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/0008318.png_1.png differ diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/0008318.png_10.png b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/0008318.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..b126a5241379642c191adf1d5c47f742bf4c32a6 Binary files /dev/null and b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/0008318.png_10.png differ diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/0008318.png_2.png b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/0008318.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..8a99b89e9ce4e62fecdbbdf31d1b9eb0913fbd4c Binary files /dev/null and b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/0008318.png_2.png differ diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/0008318.png_3.png b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/0008318.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..d44fa9d77105a106379a72b8bdaecea8a7fa321b Binary files /dev/null and b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/0008318.png_3.png differ diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/0008318.png_4.png b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/0008318.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..79ec12d372fac75e2b476f345fafae90b03cac80 Binary files /dev/null and b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/0008318.png_4.png differ diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/0008318.png_5.png b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/0008318.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..cdbadc20ebcc262c89301310769f000d76c370b6 Binary files /dev/null and b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/0008318.png_5.png differ diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/0008318.png_6.png b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/0008318.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..4ad19a45dd4f91f2c0bf053900ede5ee55c28144 Binary files /dev/null and b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/0008318.png_6.png differ diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/0008318.png_7.png b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/0008318.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..dc3ce3117c56afbc1d776120370ed373195e8f89 Binary files /dev/null and b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/0008318.png_7.png differ diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/0008318.png_8.png b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/0008318.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..ee398ecb445b7184a2d45d97b379967b29c8dcbe Binary files /dev/null and b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/0008318.png_8.png differ diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/0008318.png_9.png b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/0008318.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..edd0adef0b13da3319fda057b460869d7f0616e9 Binary files /dev/null and b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/0008318.png_9.png differ diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/0010000.png_1.png b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/0010000.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..1568da2af2f95925dffbb4e43acad73eab1bb6f1 Binary files /dev/null and b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/0010000.png_1.png differ diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/0010000.png_10.png b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/0010000.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..7250bdf06147c9ea0fa8f263fa8513ac52ddb128 Binary files /dev/null and b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/0010000.png_10.png differ diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/0010000.png_2.png b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/0010000.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..8439dd5673071dcd8bcfdc4d224be54709b89a4c Binary files /dev/null and b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/0010000.png_2.png differ diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/0010000.png_3.png b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/0010000.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..cc9d3d1cf94ac5496b03dac04e6b4a012211dd99 Binary files /dev/null and b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/0010000.png_3.png differ diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/0010000.png_4.png b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/0010000.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..cfd006e4df9e19dc39edc44ce2cfb8559a5d5055 Binary files /dev/null and b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/0010000.png_4.png differ diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/0010000.png_5.png b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/0010000.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..03227b0ff5ac61ad95fdfb13cbe55ab79c0f0063 Binary files /dev/null and b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/0010000.png_5.png differ diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/0010000.png_6.png b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/0010000.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..66b4bac37585e775ee1305bec5d5541dc96797f6 Binary files /dev/null and b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/0010000.png_6.png differ diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/0010000.png_7.png b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/0010000.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..889d88e7d43f207775dd536a9f6e396bad32c9e5 Binary files /dev/null and b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/0010000.png_7.png differ diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/0010000.png_8.png b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/0010000.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..22d1808b0e42dfe949104d6d018eb11af2d9e283 Binary files /dev/null and b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/0010000.png_8.png differ diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/0010000.png_9.png b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/0010000.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..21c27171d557579c505bc19b2a0669680a59c9a6 Binary files /dev/null and b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/0010000.png_9.png differ diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/0014159.png_1.png b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/0014159.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..f92dddbf9d391e5d3aae15cd0b22207ece61a381 Binary files /dev/null and b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/0014159.png_1.png differ diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/0014159.png_10.png b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/0014159.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..508f82c9c0c4631215871c34cb2420955ff005bd Binary files /dev/null and b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/0014159.png_10.png differ diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/0014159.png_2.png b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/0014159.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..3e2135106b1bafcc70e6da45d113d93f7b5845cc Binary files /dev/null and b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/0014159.png_2.png differ diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/0014159.png_3.png b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/0014159.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..5af195823c51eb1dbb20a41da80fdad59f72bb57 Binary files /dev/null and b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/0014159.png_3.png differ diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/0014159.png_4.png b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/0014159.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..7f3c4249bbe3ab71d7105fe3fed2b5581d2aa083 Binary files /dev/null and b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/0014159.png_4.png differ diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/0014159.png_5.png b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/0014159.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..4aac68ec45bb97d77ba0d056e2d82720336e05b5 Binary files /dev/null and b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/0014159.png_5.png differ diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/0014159.png_6.png b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/0014159.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..92029754a846f349ed5a164b2e594ecf1c9def86 Binary files /dev/null and b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/0014159.png_6.png differ diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/0014159.png_7.png b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/0014159.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..6e66f82562bfbb55f115b64504b1824c6c0c6770 Binary files /dev/null and b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/0014159.png_7.png differ diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/0014159.png_8.png b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/0014159.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..58ea0364bbb9b7e30a4da2a30a332df8b705f1d2 Binary files /dev/null and b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/0014159.png_8.png differ diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/0014159.png_9.png b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/0014159.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..6858d35756e84231afdffb254b7e2c49953a9d1f Binary files /dev/null and b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/20230606_005743/vis_data/vis_image/0014159.png_9.png differ diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/epoch_1.pth b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/epoch_1.pth new file mode 100644 index 0000000000000000000000000000000000000000..0c7af387e857b86d79fb4bf936e0dc80e27340df --- /dev/null +++ b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/epoch_1.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:bdd54daae37f9e1442e48dedab0623a4403e67800f9bdd3a7334bd1408d65390 +size 277355917 diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/epoch_10.pth b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/epoch_10.pth new file mode 100644 index 0000000000000000000000000000000000000000..47581095592a471fa380209aa8bcbb856f6f7bab --- /dev/null +++ b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/epoch_10.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:562991a25b0241b4514071c531132ffa4f622a42efc244dbcdf0fcb379ec27a8 +size 280068173 diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/epoch_2.pth b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/epoch_2.pth new file mode 100644 index 0000000000000000000000000000000000000000..f8b0ffabe996302134d4d765d5bb25255ea173ca --- /dev/null +++ b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/epoch_2.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:9cd8dc3dd698d9f5395824b3afb8a5ad3d28089898ebc305566d60fe96b07438 +size 277656333 diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/epoch_3.pth b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/epoch_3.pth new file mode 100644 index 0000000000000000000000000000000000000000..08b52ec0ff8be65338dda5758493969b25a733ce --- /dev/null +++ b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/epoch_3.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:4a56bb615f55d4f41f148d842bb1caabae929f5d275ce6e50e44489d6f1be2ac +size 277956749 diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/epoch_4.pth b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/epoch_4.pth new file mode 100644 index 0000000000000000000000000000000000000000..1b74a3b0a6fb305540d990b08bc9f4b5c2e8fd1f --- /dev/null +++ b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/epoch_4.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:895efa8e21899fbb6a94c478abfbc6cbe854b29f84dcc7bc7d8634b39aa7f88f +size 278256525 diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/epoch_5.pth b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/epoch_5.pth new file mode 100644 index 0000000000000000000000000000000000000000..78788b0f82b11714dbeba4482c1468f72bea354d --- /dev/null +++ b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/epoch_5.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:483e494d89e18734e92f58b657d829b863ddda5304ddca98c8c8c12e20cf10b9 +size 278556365 diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/epoch_6.pth b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/epoch_6.pth new file mode 100644 index 0000000000000000000000000000000000000000..a87d5e1c59f9faa780c186bb96e1cd2b87c536de --- /dev/null +++ b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/epoch_6.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:fe6803a9bc191ee1cd9763c44b24a2af328a1da3edac332c980d3252bcfe0dc2 +size 278856141 diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/epoch_7.pth b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/epoch_7.pth new file mode 100644 index 0000000000000000000000000000000000000000..02dfd2a0ddef3770182e58c73096bfe2d75ea384 --- /dev/null +++ b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/epoch_7.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:f229ad177c32d892eb70185beaa4dd590f3f0739982b891e965aa96c740cc449 +size 279156877 diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/epoch_8.pth b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/epoch_8.pth new file mode 100644 index 0000000000000000000000000000000000000000..67eb1e20a549fabe6612306cf08232057de53cdf --- /dev/null +++ b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/epoch_8.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:edaf9c3fad95773ba05a3e3d7042a553c3e96bd446f0261b38c295b95fce7ee7 +size 279461069 diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/epoch_9.pth b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/epoch_9.pth new file mode 100644 index 0000000000000000000000000000000000000000..41d929b9a81cc49908e4f5caae29a644719cafa6 --- /dev/null +++ b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/epoch_9.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:d591a250340d0c25b6d4a2f5cee8c141b29823c2f7e40b98e3675ff55650ddfc +size 279764749 diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/last_checkpoint b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/last_checkpoint new file mode 100644 index 0000000000000000000000000000000000000000..7e46b6a6b37479d22a9f528a9714c675b46abf03 --- /dev/null +++ b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/last_checkpoint @@ -0,0 +1 @@ +/mnt/petrelfs/luzeyu/workspace/fakebench/mmpretrain/workdir/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/epoch_10.pth \ No newline at end of file diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1.py b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1.py new file mode 100644 index 0000000000000000000000000000000000000000..0577dd2008d2b155b4427d6d87779bcabd4cbeec --- /dev/null +++ b/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1.py @@ -0,0 +1,349 @@ +optim_wrapper = dict( + optimizer=dict( + type='SGD', + lr=0.0001, + momentum=0.9, + weight_decay=0.0001, + _scope_='mmpretrain'), + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=False, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=512) +model = dict( + type='ImageClassifier', + backbone=dict( + frozen_stages=2, + type='ResNet', + depth=50, + num_stages=4, + out_indices=(3, ), + style='pytorch', + init_cfg=dict( + type='Pretrained', + checkpoint= + 'https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth', + prefix='backbone')), + neck=dict(type='GlobalAveragePooling'), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=2048, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + topk=1)) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/all_0.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/all_1.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3fake8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/cc1m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3real8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV2-1-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/if-dpmsolver++-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = dict(type='Accuracy', topk=1) +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV2-1-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/if-dpmsolver++-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = dict(type='Accuracy', topk=1) +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/resnet50_2xb256_all2_1m_lr1e-4_aug_1e-1' diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/20230606_005813.log b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/20230606_005813.log new file mode 100644 index 0000000000000000000000000000000000000000..eea62e886465f968c2beeca4288fbfedd8edbf6c --- /dev/null +++ b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/20230606_005813.log @@ -0,0 +1,1445 @@ +2023/06/06 00:58:16 - mmengine - INFO - +------------------------------------------------------------ +System environment: + sys.platform: linux + Python: 3.10.9 (main, Mar 8 2023, 10:47:38) [GCC 11.2.0] + CUDA available: True + numpy_random_seed: 1389536065 + GPU 0,1: NVIDIA A100-SXM4-80GB + CUDA_HOME: /mnt/petrelfs/share/cuda-11.6 + NVCC: Cuda compilation tools, release 11.6, V11.6.124 + GCC: gcc (GCC) 7.5.0 + PyTorch: 1.13.1 + PyTorch compiling details: PyTorch built with: + - GCC 9.3 + - C++ Version: 201402 + - Intel(R) oneAPI Math Kernel Library Version 2021.4-Product Build 20210904 for Intel(R) 64 architecture applications + - Intel(R) MKL-DNN v2.6.0 (Git Hash 52b5f107dd9cf10910aaa19cb47f3abf9b349815) + - OpenMP 201511 (a.k.a. OpenMP 4.5) + - LAPACK is enabled (usually provided by MKL) + - NNPACK is enabled + - CPU capability usage: AVX2 + - CUDA Runtime 11.6 + - NVCC architecture flags: -gencode;arch=compute_37,code=sm_37;-gencode;arch=compute_50,code=sm_50;-gencode;arch=compute_60,code=sm_60;-gencode;arch=compute_61,code=sm_61;-gencode;arch=compute_70,code=sm_70;-gencode;arch=compute_75,code=sm_75;-gencode;arch=compute_80,code=sm_80;-gencode;arch=compute_86,code=sm_86;-gencode;arch=compute_37,code=compute_37 + - CuDNN 8.3.2 (built against CUDA 11.5) + - Magma 2.6.1 + - Build settings: BLAS_INFO=mkl, BUILD_TYPE=Release, CUDA_VERSION=11.6, CUDNN_VERSION=8.3.2, CXX_COMPILER=/opt/rh/devtoolset-9/root/usr/bin/c++, CXX_FLAGS= -fabi-version=11 -Wno-deprecated -fvisibility-inlines-hidden -DUSE_PTHREADPOOL -fopenmp -DNDEBUG -DUSE_KINETO -DUSE_FBGEMM -DUSE_QNNPACK -DUSE_PYTORCH_QNNPACK -DUSE_XNNPACK -DSYMBOLICATE_MOBILE_DEBUG_HANDLE -DEDGE_PROFILER_USE_KINETO -O2 -fPIC -Wno-narrowing -Wall -Wextra -Werror=return-type -Werror=non-virtual-dtor -Wno-missing-field-initializers -Wno-type-limits -Wno-array-bounds -Wno-unknown-pragmas -Wunused-local-typedefs -Wno-unused-parameter -Wno-unused-function -Wno-unused-result -Wno-strict-overflow -Wno-strict-aliasing -Wno-error=deprecated-declarations -Wno-stringop-overflow -Wno-psabi -Wno-error=pedantic -Wno-error=redundant-decls -Wno-error=old-style-cast -fdiagnostics-color=always -faligned-new -Wno-unused-but-set-variable -Wno-maybe-uninitialized -fno-math-errno -fno-trapping-math -Werror=format -Werror=cast-function-type -Wno-stringop-overflow, LAPACK_INFO=mkl, PERF_WITH_AVX=1, PERF_WITH_AVX2=1, PERF_WITH_AVX512=1, TORCH_VERSION=1.13.1, USE_CUDA=ON, USE_CUDNN=ON, USE_EXCEPTION_PTR=1, USE_GFLAGS=OFF, USE_GLOG=OFF, USE_MKL=ON, USE_MKLDNN=ON, USE_MPI=OFF, USE_NCCL=ON, USE_NNPACK=ON, USE_OPENMP=ON, USE_ROCM=OFF, + + TorchVision: 0.14.1 + OpenCV: 4.7.0 + MMEngine: 0.7.3 + +Runtime environment: + cudnn_benchmark: True + mp_cfg: {'mp_start_method': 'fork', 'opencv_num_threads': 0} + dist_cfg: {'backend': 'nccl'} + seed: None + deterministic: False + Distributed launcher: slurm + Distributed training: True + GPU number: 2 +------------------------------------------------------------ + +2023/06/06 00:58:21 - mmengine - INFO - Config: +optim_wrapper = dict( + optimizer=dict( + type='SGD', + lr=0.0001, + momentum=0.9, + weight_decay=0.0001, + _scope_='mmpretrain'), + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=False, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=512) +model = dict( + type='ImageClassifier', + backbone=dict( + frozen_stages=2, + type='ResNet', + depth=50, + num_stages=4, + out_indices=(3, ), + style='pytorch', + init_cfg=dict( + type='Pretrained', + checkpoint= + 'https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth', + prefix='backbone')), + neck=dict(type='GlobalAveragePooling'), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=2048, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + topk=1)) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/all_0.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/all_1.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3fake8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/cc1m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3real8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV2-1-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/if-dpmsolver++-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = dict(type='Accuracy', topk=1) +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV2-1-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/if-dpmsolver++-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = dict(type='Accuracy', topk=1) +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1' + +2023/06/06 00:58:32 - mmengine - INFO - Hooks will be executed in the following order: +before_run: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_load_checkpoint: +(ABOVE_NORMAL) EMAHook + -------------------- +before_train: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_train_epoch: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(NORMAL ) DistSamplerSeedHook + -------------------- +before_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook + -------------------- +after_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +after_train_epoch: +(NORMAL ) IterTimerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_val_epoch: +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook + -------------------- +before_val_iter: +(NORMAL ) IterTimerHook + -------------------- +after_val_iter: +(NORMAL ) IterTimerHook +(NORMAL ) VisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_val_epoch: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_save_checkpoint: +(ABOVE_NORMAL) EMAHook + -------------------- +after_train: +(VERY_LOW ) CheckpointHook + -------------------- +before_test_epoch: +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook + -------------------- +before_test_iter: +(NORMAL ) IterTimerHook + -------------------- +after_test_iter: +(NORMAL ) IterTimerHook +(NORMAL ) VisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_test_epoch: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_run: +(BELOW_NORMAL) LoggerHook + -------------------- +2023/06/06 00:58:53 - mmengine - INFO - load backbone in model from: https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth +Name of parameter - Initialization information + +backbone.conv1.weight - torch.Size([64, 3, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.bn1.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.bn1.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.conv1.weight - torch.Size([64, 64, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.bn1.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.bn1.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.conv2.weight - torch.Size([64, 64, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.bn2.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.bn2.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.conv3.weight - torch.Size([256, 64, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.bn3.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.bn3.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.downsample.0.weight - torch.Size([256, 64, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.downsample.1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.downsample.1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.conv1.weight - torch.Size([64, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.bn1.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.bn1.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.conv2.weight - torch.Size([64, 64, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.bn2.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.bn2.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.conv3.weight - torch.Size([256, 64, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.bn3.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.bn3.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.conv1.weight - torch.Size([64, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.bn1.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.bn1.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.conv2.weight - torch.Size([64, 64, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.bn2.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.bn2.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.conv3.weight - torch.Size([256, 64, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.bn3.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.bn3.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.conv1.weight - torch.Size([128, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.bn1.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.bn1.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.conv2.weight - torch.Size([128, 128, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.bn2.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.bn2.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.conv3.weight - torch.Size([512, 128, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.bn3.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.bn3.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.downsample.0.weight - torch.Size([512, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.downsample.1.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.downsample.1.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.conv1.weight - torch.Size([128, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.bn1.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.bn1.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.conv2.weight - torch.Size([128, 128, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.bn2.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.bn2.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.conv3.weight - torch.Size([512, 128, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.bn3.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.bn3.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.conv1.weight - torch.Size([128, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.bn1.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.bn1.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.conv2.weight - torch.Size([128, 128, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.bn2.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.bn2.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.conv3.weight - torch.Size([512, 128, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.bn3.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.bn3.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.conv1.weight - torch.Size([128, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.bn1.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.bn1.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.conv2.weight - torch.Size([128, 128, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.bn2.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.bn2.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.conv3.weight - torch.Size([512, 128, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.bn3.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.bn3.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.conv1.weight - torch.Size([256, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.bn1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.bn1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.conv2.weight - torch.Size([256, 256, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.bn2.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.bn2.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.conv3.weight - torch.Size([1024, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.bn3.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.bn3.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.downsample.0.weight - torch.Size([1024, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.downsample.1.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.downsample.1.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.conv1.weight - torch.Size([256, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.bn1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.bn1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.conv2.weight - torch.Size([256, 256, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.bn2.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.bn2.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.conv3.weight - torch.Size([1024, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.bn3.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.bn3.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.conv1.weight - torch.Size([256, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.bn1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.bn1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.conv2.weight - torch.Size([256, 256, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.bn2.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.bn2.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.conv3.weight - torch.Size([1024, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.bn3.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.bn3.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.conv1.weight - torch.Size([256, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.bn1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.bn1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.conv2.weight - torch.Size([256, 256, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.bn2.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.bn2.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.conv3.weight - torch.Size([1024, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.bn3.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.bn3.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.conv1.weight - torch.Size([256, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.bn1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.bn1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.conv2.weight - torch.Size([256, 256, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.bn2.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.bn2.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.conv3.weight - torch.Size([1024, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.bn3.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.bn3.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.conv1.weight - torch.Size([256, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.bn1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.bn1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.conv2.weight - torch.Size([256, 256, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.bn2.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.bn2.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.conv3.weight - torch.Size([1024, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.bn3.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.bn3.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.conv1.weight - torch.Size([512, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.bn1.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.bn1.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.conv2.weight - torch.Size([512, 512, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.bn2.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.bn2.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.conv3.weight - torch.Size([2048, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.bn3.weight - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.bn3.bias - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.downsample.0.weight - torch.Size([2048, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.downsample.1.weight - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.downsample.1.bias - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.conv1.weight - torch.Size([512, 2048, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.bn1.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.bn1.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.conv2.weight - torch.Size([512, 512, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.bn2.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.bn2.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.conv3.weight - torch.Size([2048, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.bn3.weight - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.bn3.bias - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.conv1.weight - torch.Size([512, 2048, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.bn1.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.bn1.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.conv2.weight - torch.Size([512, 512, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.bn2.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.bn2.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.conv3.weight - torch.Size([2048, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.bn3.weight - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.bn3.bias - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +head.fc.weight - torch.Size([2, 2048]): +NormalInit: mean=0, std=0.01, bias=0 + +head.fc.bias - torch.Size([2]): +NormalInit: mean=0, std=0.01, bias=0 +2023/06/06 00:58:54 - mmengine - WARNING - "FileClient" will be deprecated in future. Please use io functions in https://mmengine.readthedocs.io/en/latest/api/fileio.html#file-io +2023/06/06 00:58:54 - mmengine - WARNING - "HardDiskBackend" is the alias of "LocalBackend" and the former will be deprecated in future. +2023/06/06 00:58:54 - mmengine - INFO - Checkpoints will be saved to /mnt/petrelfs/luzeyu/workspace/fakebench/mmpretrain/workdir/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1. +2023/06/06 01:00:18 - mmengine - INFO - Epoch(train) [1][ 100/4092] lr: 9.9999e-05 eta: 9:32:51 time: 0.7276 data_time: 0.0567 memory: 9436 loss: 0.6467 +2023/06/06 01:01:37 - mmengine - INFO - Epoch(train) [1][ 200/4092] lr: 9.9995e-05 eta: 9:13:48 time: 0.7685 data_time: 0.0007 memory: 6319 loss: 0.6184 +2023/06/06 01:02:56 - mmengine - INFO - Epoch(train) [1][ 300/4092] lr: 9.9988e-05 eta: 9:07:10 time: 0.7909 data_time: 0.0009 memory: 6319 loss: 0.5818 +2023/06/06 01:04:15 - mmengine - INFO - Epoch(train) [1][ 400/4092] lr: 9.9979e-05 eta: 9:02:34 time: 0.8350 data_time: 0.0008 memory: 6319 loss: 0.5466 +2023/06/06 01:05:35 - mmengine - INFO - Epoch(train) [1][ 500/4092] lr: 9.9967e-05 eta: 9:01:06 time: 0.8036 data_time: 0.0007 memory: 6319 loss: 0.5174 +2023/06/06 01:06:52 - mmengine - INFO - Epoch(train) [1][ 600/4092] lr: 9.9952e-05 eta: 8:55:32 time: 0.7820 data_time: 0.0009 memory: 6319 loss: 0.5113 +2023/06/06 01:08:10 - mmengine - INFO - Epoch(train) [1][ 700/4092] lr: 9.9935e-05 eta: 8:52:31 time: 0.7548 data_time: 0.0009 memory: 6319 loss: 0.4770 +2023/06/06 01:09:26 - mmengine - INFO - Epoch(train) [1][ 800/4092] lr: 9.9915e-05 eta: 8:48:34 time: 0.7602 data_time: 0.0011 memory: 6319 loss: 0.4670 +2023/06/06 01:10:43 - mmengine - INFO - Epoch(train) [1][ 900/4092] lr: 9.9893e-05 eta: 8:45:45 time: 0.7635 data_time: 0.0009 memory: 6319 loss: 0.4512 +2023/06/06 01:11:59 - mmengine - INFO - Exp name: resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1_20230606_005813 +2023/06/06 01:11:59 - mmengine - INFO - Epoch(train) [1][1000/4092] lr: 9.9868e-05 eta: 8:42:20 time: 0.7467 data_time: 0.0009 memory: 6319 loss: 0.4458 +2023/06/06 01:13:13 - mmengine - INFO - Epoch(train) [1][1100/4092] lr: 9.9840e-05 eta: 8:38:23 time: 0.7451 data_time: 0.0015 memory: 6319 loss: 0.4359 +2023/06/06 01:14:29 - mmengine - INFO - Epoch(train) [1][1200/4092] lr: 9.9809e-05 eta: 8:36:17 time: 0.8331 data_time: 0.0008 memory: 6319 loss: 0.4352 +2023/06/06 01:15:46 - mmengine - INFO - Epoch(train) [1][1300/4092] lr: 9.9776e-05 eta: 8:34:02 time: 0.7362 data_time: 0.0009 memory: 6319 loss: 0.4102 +2023/06/06 01:17:02 - mmengine - INFO - Epoch(train) [1][1400/4092] lr: 9.9741e-05 eta: 8:32:05 time: 0.7338 data_time: 0.0009 memory: 6319 loss: 0.4059 +2023/06/06 01:18:17 - mmengine - INFO - Epoch(train) [1][1500/4092] lr: 9.9702e-05 eta: 8:29:38 time: 0.7408 data_time: 0.0008 memory: 6319 loss: 0.3949 +2023/06/06 01:19:32 - mmengine - INFO - Epoch(train) [1][1600/4092] lr: 9.9661e-05 eta: 8:27:16 time: 0.7590 data_time: 0.0009 memory: 6319 loss: 0.3928 +2023/06/06 01:20:48 - mmengine - INFO - Epoch(train) [1][1700/4092] lr: 9.9618e-05 eta: 8:25:13 time: 0.6609 data_time: 0.0009 memory: 6319 loss: 0.3838 +2023/06/06 01:22:02 - mmengine - INFO - Epoch(train) [1][1800/4092] lr: 9.9571e-05 eta: 8:22:50 time: 0.7454 data_time: 0.0008 memory: 6319 loss: 0.3867 +2023/06/06 01:23:17 - mmengine - INFO - Epoch(train) [1][1900/4092] lr: 9.9523e-05 eta: 8:20:55 time: 0.7970 data_time: 0.0008 memory: 6319 loss: 0.3863 +2023/06/06 01:24:30 - mmengine - INFO - Exp name: resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1_20230606_005813 +2023/06/06 01:24:30 - mmengine - INFO - Epoch(train) [1][2000/4092] lr: 9.9471e-05 eta: 8:18:22 time: 0.7101 data_time: 0.0008 memory: 6319 loss: 0.3688 +2023/06/06 01:25:45 - mmengine - INFO - Epoch(train) [1][2100/4092] lr: 9.9417e-05 eta: 8:16:19 time: 0.7135 data_time: 0.0009 memory: 6319 loss: 0.3723 +2023/06/06 01:28:34 - mmengine - INFO - Epoch(train) [1][2200/4092] lr: 9.9360e-05 eta: 8:42:12 time: 0.8075 data_time: 0.0008 memory: 6319 loss: 0.3746 +2023/06/06 01:29:52 - mmengine - INFO - Epoch(train) [1][2300/4092] lr: 9.9301e-05 eta: 8:39:57 time: 0.7297 data_time: 0.0010 memory: 6319 loss: 0.3827 +2023/06/06 01:31:05 - mmengine - INFO - Epoch(train) [1][2400/4092] lr: 9.9239e-05 eta: 8:36:35 time: 0.7086 data_time: 0.0008 memory: 6319 loss: 0.3737 +2023/06/06 01:32:17 - mmengine - INFO - Epoch(train) [1][2500/4092] lr: 9.9174e-05 eta: 8:33:14 time: 0.6826 data_time: 0.0008 memory: 6319 loss: 0.3717 +2023/06/06 01:33:33 - mmengine - INFO - Epoch(train) [1][2600/4092] lr: 9.9107e-05 eta: 8:30:39 time: 0.7525 data_time: 0.0008 memory: 6319 loss: 0.3504 +2023/06/06 01:34:48 - mmengine - INFO - Epoch(train) [1][2700/4092] lr: 9.9037e-05 eta: 8:28:09 time: 0.7436 data_time: 0.0009 memory: 6319 loss: 0.3462 +2023/06/06 01:36:01 - mmengine - INFO - Epoch(train) [1][2800/4092] lr: 9.8965e-05 eta: 8:25:27 time: 0.6897 data_time: 0.0010 memory: 6319 loss: 0.3519 +2023/06/06 01:37:16 - mmengine - INFO - Epoch(train) [1][2900/4092] lr: 9.8890e-05 eta: 8:22:58 time: 0.7586 data_time: 0.0010 memory: 6319 loss: 0.3407 +2023/06/06 01:38:30 - mmengine - INFO - Exp name: resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1_20230606_005813 +2023/06/06 01:38:30 - mmengine - INFO - Epoch(train) [1][3000/4092] lr: 9.8812e-05 eta: 8:20:34 time: 0.7910 data_time: 0.0009 memory: 6319 loss: 0.3508 +2023/06/06 01:39:43 - mmengine - INFO - Epoch(train) [1][3100/4092] lr: 9.8732e-05 eta: 8:18:02 time: 0.7930 data_time: 0.0010 memory: 6319 loss: 0.3420 +2023/06/06 01:40:56 - mmengine - INFO - Epoch(train) [1][3200/4092] lr: 9.8650e-05 eta: 8:15:32 time: 0.6975 data_time: 0.0012 memory: 6319 loss: 0.3493 +2023/06/06 01:42:08 - mmengine - INFO - Epoch(train) [1][3300/4092] lr: 9.8564e-05 eta: 8:12:59 time: 0.7505 data_time: 0.0009 memory: 6319 loss: 0.3305 +2023/06/06 01:43:23 - mmengine - INFO - Epoch(train) [1][3400/4092] lr: 9.8476e-05 eta: 8:11:00 time: 0.7414 data_time: 0.0008 memory: 6319 loss: 0.3468 +2023/06/06 01:44:39 - mmengine - INFO - Epoch(train) [1][3500/4092] lr: 9.8386e-05 eta: 8:09:08 time: 0.7580 data_time: 0.0010 memory: 6319 loss: 0.3498 +2023/06/06 01:45:55 - mmengine - INFO - Epoch(train) [1][3600/4092] lr: 9.8293e-05 eta: 8:07:23 time: 0.7445 data_time: 0.0008 memory: 6319 loss: 0.3350 +2023/06/06 01:47:14 - mmengine - INFO - Epoch(train) [1][3700/4092] lr: 9.8198e-05 eta: 8:06:17 time: 0.7830 data_time: 0.0009 memory: 6319 loss: 0.3254 +2023/06/06 01:48:30 - mmengine - INFO - Epoch(train) [1][3800/4092] lr: 9.8099e-05 eta: 8:04:35 time: 0.7673 data_time: 0.0010 memory: 6319 loss: 0.3400 +2023/06/06 01:49:46 - mmengine - INFO - Epoch(train) [1][3900/4092] lr: 9.7999e-05 eta: 8:02:50 time: 0.7695 data_time: 0.0009 memory: 6319 loss: 0.3286 +2023/06/06 01:51:00 - mmengine - INFO - Exp name: resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1_20230606_005813 +2023/06/06 01:51:00 - mmengine - INFO - Epoch(train) [1][4000/4092] lr: 9.7896e-05 eta: 8:00:57 time: 0.7636 data_time: 0.0009 memory: 6319 loss: 0.3149 +2023/06/06 01:52:10 - mmengine - INFO - Exp name: resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1_20230606_005813 +2023/06/06 01:52:10 - mmengine - INFO - Saving checkpoint at 1 epochs +2023/06/06 01:52:57 - mmengine - INFO - Epoch(val) [1][100/119] eta: 0:00:08 time: 0.8238 data_time: 0.7352 memory: 6319 +2023/06/06 01:53:25 - mmengine - INFO - Epoch(val) [1][119/119] accuracy/top1: 82.7698 data_time: 0.3998 time: 0.4893 +2023/06/06 01:54:40 - mmengine - INFO - Epoch(train) [2][ 100/4092] lr: 9.7691e-05 eta: 7:57:40 time: 0.7403 data_time: 0.4555 memory: 6319 loss: 0.3017 +2023/06/06 01:55:55 - mmengine - INFO - Epoch(train) [2][ 200/4092] lr: 9.7580e-05 eta: 7:55:57 time: 0.7495 data_time: 0.2718 memory: 6319 loss: 0.3268 +2023/06/06 01:57:08 - mmengine - INFO - Epoch(train) [2][ 300/4092] lr: 9.7467e-05 eta: 7:53:56 time: 0.7373 data_time: 0.1156 memory: 6319 loss: 0.3033 +2023/06/06 01:58:21 - mmengine - INFO - Epoch(train) [2][ 400/4092] lr: 9.7352e-05 eta: 7:52:05 time: 0.8112 data_time: 0.2385 memory: 6319 loss: 0.3228 +2023/06/06 01:59:35 - mmengine - INFO - Epoch(train) [2][ 500/4092] lr: 9.7234e-05 eta: 7:50:15 time: 0.7327 data_time: 0.0173 memory: 6319 loss: 0.3332 +2023/06/06 02:00:52 - mmengine - INFO - Epoch(train) [2][ 600/4092] lr: 9.7113e-05 eta: 7:48:49 time: 0.7490 data_time: 0.0010 memory: 6319 loss: 0.3017 +2023/06/06 02:02:02 - mmengine - INFO - Epoch(train) [2][ 700/4092] lr: 9.6990e-05 eta: 7:46:35 time: 0.7078 data_time: 0.0009 memory: 6319 loss: 0.3062 +2023/06/06 02:03:16 - mmengine - INFO - Epoch(train) [2][ 800/4092] lr: 9.6865e-05 eta: 7:44:55 time: 0.7500 data_time: 0.0011 memory: 6319 loss: 0.3307 +2023/06/06 02:04:31 - mmengine - INFO - Epoch(train) [2][ 900/4092] lr: 9.6737e-05 eta: 7:43:18 time: 0.7641 data_time: 0.0009 memory: 6319 loss: 0.3279 +2023/06/06 02:04:39 - mmengine - INFO - Exp name: resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1_20230606_005813 +2023/06/06 02:05:45 - mmengine - INFO - Epoch(train) [2][1000/4092] lr: 9.6606e-05 eta: 7:41:36 time: 0.7436 data_time: 0.0009 memory: 6319 loss: 0.3087 +2023/06/06 02:06:59 - mmengine - INFO - Epoch(train) [2][1100/4092] lr: 9.6473e-05 eta: 7:40:00 time: 0.8010 data_time: 0.0010 memory: 6319 loss: 0.2861 +2023/06/06 02:08:14 - mmengine - INFO - Epoch(train) [2][1200/4092] lr: 9.6338e-05 eta: 7:38:28 time: 0.7218 data_time: 0.0010 memory: 6319 loss: 0.3208 +2023/06/06 02:09:27 - mmengine - INFO - Epoch(train) [2][1300/4092] lr: 9.6200e-05 eta: 7:36:40 time: 0.7188 data_time: 0.0009 memory: 6319 loss: 0.3038 +2023/06/06 02:10:39 - mmengine - INFO - Epoch(train) [2][1400/4092] lr: 9.6060e-05 eta: 7:34:49 time: 0.6924 data_time: 0.0008 memory: 6319 loss: 0.2916 +2023/06/06 02:11:51 - mmengine - INFO - Epoch(train) [2][1500/4092] lr: 9.5918e-05 eta: 7:33:00 time: 0.6918 data_time: 0.0009 memory: 6319 loss: 0.3138 +2023/06/06 02:13:06 - mmengine - INFO - Epoch(train) [2][1600/4092] lr: 9.5773e-05 eta: 7:31:35 time: 0.7407 data_time: 0.0009 memory: 6319 loss: 0.2937 +2023/06/06 02:14:26 - mmengine - INFO - Epoch(train) [2][1700/4092] lr: 9.5625e-05 eta: 7:30:32 time: 0.8106 data_time: 0.0010 memory: 6319 loss: 0.3085 +2023/06/06 02:15:41 - mmengine - INFO - Epoch(train) [2][1800/4092] lr: 9.5475e-05 eta: 7:29:07 time: 0.7120 data_time: 0.0009 memory: 6319 loss: 0.3005 +2023/06/06 02:16:55 - mmengine - INFO - Epoch(train) [2][1900/4092] lr: 9.5323e-05 eta: 7:27:32 time: 0.7047 data_time: 0.0009 memory: 6319 loss: 0.3085 +2023/06/06 02:17:03 - mmengine - INFO - Exp name: resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1_20230606_005813 +2023/06/06 02:18:08 - mmengine - INFO - Epoch(train) [2][2000/4092] lr: 9.5169e-05 eta: 7:25:53 time: 0.7012 data_time: 0.0010 memory: 6319 loss: 0.2957 +2023/06/06 02:19:24 - mmengine - INFO - Epoch(train) [2][2100/4092] lr: 9.5012e-05 eta: 7:24:30 time: 0.7679 data_time: 0.0008 memory: 6319 loss: 0.2921 +2023/06/06 02:20:37 - mmengine - INFO - Epoch(train) [2][2200/4092] lr: 9.4853e-05 eta: 7:22:54 time: 0.7100 data_time: 0.0010 memory: 6319 loss: 0.2963 +2023/06/06 02:21:51 - mmengine - INFO - Epoch(train) [2][2300/4092] lr: 9.4691e-05 eta: 7:21:20 time: 0.7452 data_time: 0.0011 memory: 6319 loss: 0.2916 +2023/06/06 02:23:05 - mmengine - INFO - Epoch(train) [2][2400/4092] lr: 9.4527e-05 eta: 7:19:53 time: 0.7754 data_time: 0.0009 memory: 6319 loss: 0.2812 +2023/06/06 02:24:20 - mmengine - INFO - Epoch(train) [2][2500/4092] lr: 9.4361e-05 eta: 7:18:24 time: 0.8683 data_time: 0.0010 memory: 6319 loss: 0.3007 +2023/06/06 02:25:34 - mmengine - INFO - Epoch(train) [2][2600/4092] lr: 9.4192e-05 eta: 7:16:54 time: 0.7002 data_time: 0.0008 memory: 6319 loss: 0.2915 +2023/06/06 02:26:50 - mmengine - INFO - Epoch(train) [2][2700/4092] lr: 9.4021e-05 eta: 7:15:35 time: 0.7760 data_time: 0.0009 memory: 6319 loss: 0.3125 +2023/06/06 02:28:06 - mmengine - INFO - Epoch(train) [2][2800/4092] lr: 9.3848e-05 eta: 7:14:14 time: 0.7401 data_time: 0.0009 memory: 6319 loss: 0.2726 +2023/06/06 02:29:22 - mmengine - INFO - Epoch(train) [2][2900/4092] lr: 9.3672e-05 eta: 7:12:57 time: 0.7292 data_time: 0.0009 memory: 6319 loss: 0.3004 +2023/06/06 02:29:30 - mmengine - INFO - Exp name: resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1_20230606_005813 +2023/06/06 02:30:39 - mmengine - INFO - Epoch(train) [2][3000/4092] lr: 9.3495e-05 eta: 7:11:41 time: 0.7777 data_time: 0.0011 memory: 6319 loss: 0.2778 +2023/06/06 02:31:55 - mmengine - INFO - Epoch(train) [2][3100/4092] lr: 9.3315e-05 eta: 7:10:25 time: 0.7420 data_time: 0.0010 memory: 6319 loss: 0.3007 +2023/06/06 02:33:09 - mmengine - INFO - Epoch(train) [2][3200/4092] lr: 9.3132e-05 eta: 7:08:56 time: 0.7074 data_time: 0.0008 memory: 6319 loss: 0.2931 +2023/06/06 02:34:30 - mmengine - INFO - Epoch(train) [2][3300/4092] lr: 9.2948e-05 eta: 7:07:57 time: 0.8083 data_time: 0.0013 memory: 6319 loss: 0.2762 +2023/06/06 02:35:45 - mmengine - INFO - Epoch(train) [2][3400/4092] lr: 9.2761e-05 eta: 7:06:36 time: 0.8479 data_time: 0.0008 memory: 6319 loss: 0.2856 +2023/06/06 02:37:00 - mmengine - INFO - Epoch(train) [2][3500/4092] lr: 9.2572e-05 eta: 7:05:13 time: 0.6747 data_time: 0.0009 memory: 6319 loss: 0.2842 +2023/06/06 02:38:24 - mmengine - INFO - Epoch(train) [2][3600/4092] lr: 9.2381e-05 eta: 7:04:28 time: 0.7300 data_time: 0.0009 memory: 6319 loss: 0.2812 +2023/06/06 02:39:38 - mmengine - INFO - Epoch(train) [2][3700/4092] lr: 9.2187e-05 eta: 7:02:57 time: 0.7394 data_time: 0.0009 memory: 6319 loss: 0.2923 +2023/06/06 02:40:56 - mmengine - INFO - Epoch(train) [2][3800/4092] lr: 9.1991e-05 eta: 7:01:47 time: 0.7227 data_time: 0.0011 memory: 6319 loss: 0.3005 +2023/06/06 02:42:09 - mmengine - INFO - Epoch(train) [2][3900/4092] lr: 9.1794e-05 eta: 7:00:17 time: 0.7644 data_time: 0.0009 memory: 6319 loss: 0.2758 +2023/06/06 02:42:16 - mmengine - INFO - Exp name: resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1_20230606_005813 +2023/06/06 02:43:23 - mmengine - INFO - Epoch(train) [2][4000/4092] lr: 9.1594e-05 eta: 6:58:51 time: 0.7619 data_time: 0.0010 memory: 6319 loss: 0.2709 +2023/06/06 02:44:35 - mmengine - INFO - Exp name: resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1_20230606_005813 +2023/06/06 02:44:35 - mmengine - INFO - Saving checkpoint at 2 epochs +2023/06/06 02:45:20 - mmengine - INFO - Epoch(val) [2][100/119] eta: 0:00:07 time: 0.6763 data_time: 0.5850 memory: 6319 +2023/06/06 02:45:47 - mmengine - INFO - Epoch(val) [2][119/119] accuracy/top1: 81.9224 data_time: 0.3713 time: 0.4613 +2023/06/06 02:47:05 - mmengine - INFO - Epoch(train) [3][ 100/4092] lr: 9.1204e-05 eta: 6:56:33 time: 0.7784 data_time: 0.3623 memory: 6319 loss: 0.2805 +2023/06/06 02:48:19 - mmengine - INFO - Epoch(train) [3][ 200/4092] lr: 9.0997e-05 eta: 6:55:09 time: 0.7314 data_time: 0.2476 memory: 6319 loss: 0.2813 +2023/06/06 02:49:33 - mmengine - INFO - Epoch(train) [3][ 300/4092] lr: 9.0789e-05 eta: 6:53:43 time: 0.7502 data_time: 0.0113 memory: 6319 loss: 0.2848 +2023/06/06 02:50:49 - mmengine - INFO - Epoch(train) [3][ 400/4092] lr: 9.0579e-05 eta: 6:52:22 time: 0.7843 data_time: 0.0010 memory: 6319 loss: 0.2834 +2023/06/06 02:52:03 - mmengine - INFO - Epoch(train) [3][ 500/4092] lr: 9.0366e-05 eta: 6:50:58 time: 0.7473 data_time: 0.0008 memory: 6319 loss: 0.2887 +2023/06/06 02:53:17 - mmengine - INFO - Epoch(train) [3][ 600/4092] lr: 9.0151e-05 eta: 6:49:33 time: 0.7299 data_time: 0.0008 memory: 6319 loss: 0.2831 +2023/06/06 02:54:32 - mmengine - INFO - Epoch(train) [3][ 700/4092] lr: 8.9935e-05 eta: 6:48:10 time: 0.7469 data_time: 0.0009 memory: 6319 loss: 0.2758 +2023/06/06 02:55:47 - mmengine - INFO - Epoch(train) [3][ 800/4092] lr: 8.9716e-05 eta: 6:46:48 time: 0.7335 data_time: 0.0008 memory: 6319 loss: 0.2611 +2023/06/06 02:56:02 - mmengine - INFO - Exp name: resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1_20230606_005813 +2023/06/06 02:57:02 - mmengine - INFO - Epoch(train) [3][ 900/4092] lr: 8.9495e-05 eta: 6:45:26 time: 0.7905 data_time: 0.0007 memory: 6319 loss: 0.2864 +2023/06/06 02:58:21 - mmengine - INFO - Epoch(train) [3][1000/4092] lr: 8.9272e-05 eta: 6:44:21 time: 0.8574 data_time: 0.0008 memory: 6319 loss: 0.2859 +2023/06/06 02:59:37 - mmengine - INFO - Epoch(train) [3][1100/4092] lr: 8.9047e-05 eta: 6:43:01 time: 0.7644 data_time: 0.0010 memory: 6319 loss: 0.2825 +2023/06/06 03:00:54 - mmengine - INFO - Epoch(train) [3][1200/4092] lr: 8.8820e-05 eta: 6:41:46 time: 0.7886 data_time: 0.0009 memory: 6319 loss: 0.2827 +2023/06/06 03:02:05 - mmengine - INFO - Epoch(train) [3][1300/4092] lr: 8.8591e-05 eta: 6:40:13 time: 0.7629 data_time: 0.0009 memory: 6319 loss: 0.2766 +2023/06/06 03:03:20 - mmengine - INFO - Epoch(train) [3][1400/4092] lr: 8.8360e-05 eta: 6:38:52 time: 0.7176 data_time: 0.0009 memory: 6319 loss: 0.2816 +2023/06/06 03:04:36 - mmengine - INFO - Epoch(train) [3][1500/4092] lr: 8.8128e-05 eta: 6:37:33 time: 0.8579 data_time: 0.0009 memory: 6319 loss: 0.2614 +2023/06/06 03:05:49 - mmengine - INFO - Epoch(train) [3][1600/4092] lr: 8.7893e-05 eta: 6:36:07 time: 0.7500 data_time: 0.0008 memory: 6319 loss: 0.2704 +2023/06/06 03:07:03 - mmengine - INFO - Epoch(train) [3][1700/4092] lr: 8.7656e-05 eta: 6:34:44 time: 0.6923 data_time: 0.0008 memory: 6319 loss: 0.2542 +2023/06/06 03:08:18 - mmengine - INFO - Epoch(train) [3][1800/4092] lr: 8.7417e-05 eta: 6:33:21 time: 0.7331 data_time: 0.0007 memory: 6319 loss: 0.2760 +2023/06/06 03:08:32 - mmengine - INFO - Exp name: resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1_20230606_005813 +2023/06/06 03:09:33 - mmengine - INFO - Epoch(train) [3][1900/4092] lr: 8.7177e-05 eta: 6:32:04 time: 0.7779 data_time: 0.0008 memory: 6319 loss: 0.2653 +2023/06/06 03:10:47 - mmengine - INFO - Epoch(train) [3][2000/4092] lr: 8.6934e-05 eta: 6:30:39 time: 0.6819 data_time: 0.0010 memory: 6319 loss: 0.2725 +2023/06/06 03:12:02 - mmengine - INFO - Epoch(train) [3][2100/4092] lr: 8.6690e-05 eta: 6:29:20 time: 0.7411 data_time: 0.0010 memory: 6319 loss: 0.2684 +2023/06/06 03:13:19 - mmengine - INFO - Epoch(train) [3][2200/4092] lr: 8.6444e-05 eta: 6:28:06 time: 0.7439 data_time: 0.0009 memory: 6319 loss: 0.2677 +2023/06/06 03:14:38 - mmengine - INFO - Epoch(train) [3][2300/4092] lr: 8.6196e-05 eta: 6:26:57 time: 0.7392 data_time: 0.0009 memory: 6319 loss: 0.2721 +2023/06/06 03:15:54 - mmengine - INFO - Epoch(train) [3][2400/4092] lr: 8.5946e-05 eta: 6:25:40 time: 0.7655 data_time: 0.0008 memory: 6319 loss: 0.2583 +2023/06/06 03:17:10 - mmengine - INFO - Epoch(train) [3][2500/4092] lr: 8.5694e-05 eta: 6:24:22 time: 0.7567 data_time: 0.0009 memory: 6319 loss: 0.2638 +2023/06/06 03:18:26 - mmengine - INFO - Epoch(train) [3][2600/4092] lr: 8.5441e-05 eta: 6:23:05 time: 0.7387 data_time: 0.0009 memory: 6319 loss: 0.2830 +2023/06/06 03:19:40 - mmengine - INFO - Epoch(train) [3][2700/4092] lr: 8.5185e-05 eta: 6:21:44 time: 0.7503 data_time: 0.0008 memory: 6319 loss: 0.2826 +2023/06/06 03:20:55 - mmengine - INFO - Epoch(train) [3][2800/4092] lr: 8.4928e-05 eta: 6:20:24 time: 0.7174 data_time: 0.0010 memory: 6319 loss: 0.2440 +2023/06/06 03:21:10 - mmengine - INFO - Exp name: resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1_20230606_005813 +2023/06/06 03:22:13 - mmengine - INFO - Epoch(train) [3][2900/4092] lr: 8.4669e-05 eta: 6:19:12 time: 0.6999 data_time: 0.0011 memory: 6319 loss: 0.2680 +2023/06/06 03:23:34 - mmengine - INFO - Epoch(train) [3][3000/4092] lr: 8.4409e-05 eta: 6:18:09 time: 0.7636 data_time: 0.0009 memory: 6319 loss: 0.2751 +2023/06/06 03:24:50 - mmengine - INFO - Epoch(train) [3][3100/4092] lr: 8.4146e-05 eta: 6:16:51 time: 0.7459 data_time: 0.0010 memory: 6319 loss: 0.2576 +2023/06/06 03:26:06 - mmengine - INFO - Epoch(train) [3][3200/4092] lr: 8.3882e-05 eta: 6:15:34 time: 0.7709 data_time: 0.0007 memory: 6319 loss: 0.2576 +2023/06/06 03:27:22 - mmengine - INFO - Epoch(train) [3][3300/4092] lr: 8.3616e-05 eta: 6:14:17 time: 0.7078 data_time: 0.0008 memory: 6319 loss: 0.2428 +2023/06/06 03:28:40 - mmengine - INFO - Epoch(train) [3][3400/4092] lr: 8.3349e-05 eta: 6:13:06 time: 0.7724 data_time: 0.0009 memory: 6319 loss: 0.2670 +2023/06/06 03:29:57 - mmengine - INFO - Epoch(train) [3][3500/4092] lr: 8.3080e-05 eta: 6:11:50 time: 0.7585 data_time: 0.0010 memory: 6319 loss: 0.2671 +2023/06/06 03:31:12 - mmengine - INFO - Epoch(train) [3][3600/4092] lr: 8.2809e-05 eta: 6:10:31 time: 0.7335 data_time: 0.0008 memory: 6319 loss: 0.2550 +2023/06/06 03:32:28 - mmengine - INFO - Epoch(train) [3][3700/4092] lr: 8.2537e-05 eta: 6:09:13 time: 0.7529 data_time: 0.0010 memory: 6319 loss: 0.2653 +2023/06/06 03:33:45 - mmengine - INFO - Epoch(train) [3][3800/4092] lr: 8.2263e-05 eta: 6:08:00 time: 0.7682 data_time: 0.0011 memory: 6319 loss: 0.2576 +2023/06/06 03:33:59 - mmengine - INFO - Exp name: resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1_20230606_005813 +2023/06/06 03:35:00 - mmengine - INFO - Epoch(train) [3][3900/4092] lr: 8.1987e-05 eta: 6:06:39 time: 0.7986 data_time: 0.0010 memory: 6319 loss: 0.2617 +2023/06/06 03:36:14 - mmengine - INFO - Epoch(train) [3][4000/4092] lr: 8.1710e-05 eta: 6:05:18 time: 0.7045 data_time: 0.0011 memory: 6319 loss: 0.2548 +2023/06/06 03:37:27 - mmengine - INFO - Exp name: resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1_20230606_005813 +2023/06/06 03:37:27 - mmengine - INFO - Saving checkpoint at 3 epochs +2023/06/06 03:38:12 - mmengine - INFO - Epoch(val) [3][100/119] eta: 0:00:07 time: 0.6277 data_time: 0.5376 memory: 6319 +2023/06/06 03:38:39 - mmengine - INFO - Epoch(val) [3][119/119] accuracy/top1: 81.7387 data_time: 0.3711 time: 0.4602 +2023/06/06 03:40:00 - mmengine - INFO - Epoch(train) [4][ 100/4092] lr: 8.1173e-05 eta: 6:03:09 time: 0.7065 data_time: 0.4401 memory: 6319 loss: 0.2581 +2023/06/06 03:41:15 - mmengine - INFO - Epoch(train) [4][ 200/4092] lr: 8.0891e-05 eta: 6:01:50 time: 0.7446 data_time: 0.0747 memory: 6319 loss: 0.2434 +2023/06/06 03:42:31 - mmengine - INFO - Epoch(train) [4][ 300/4092] lr: 8.0608e-05 eta: 6:00:32 time: 0.7453 data_time: 0.0009 memory: 6319 loss: 0.2466 +2023/06/06 03:43:46 - mmengine - INFO - Epoch(train) [4][ 400/4092] lr: 8.0323e-05 eta: 5:59:14 time: 0.7624 data_time: 0.0010 memory: 6319 loss: 0.2469 +2023/06/06 03:45:03 - mmengine - INFO - Epoch(train) [4][ 500/4092] lr: 8.0037e-05 eta: 5:57:59 time: 0.7806 data_time: 0.0009 memory: 6319 loss: 0.2450 +2023/06/06 03:46:19 - mmengine - INFO - Epoch(train) [4][ 600/4092] lr: 7.9749e-05 eta: 5:56:42 time: 0.7400 data_time: 0.0009 memory: 6319 loss: 0.2516 +2023/06/06 03:47:37 - mmengine - INFO - Epoch(train) [4][ 700/4092] lr: 7.9459e-05 eta: 5:55:29 time: 0.7678 data_time: 0.0009 memory: 6319 loss: 0.2599 +2023/06/06 03:47:57 - mmengine - INFO - Exp name: resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1_20230606_005813 +2023/06/06 03:48:54 - mmengine - INFO - Epoch(train) [4][ 800/4092] lr: 7.9169e-05 eta: 5:54:13 time: 0.7357 data_time: 0.0010 memory: 6319 loss: 0.2558 +2023/06/06 03:50:11 - mmengine - INFO - Epoch(train) [4][ 900/4092] lr: 7.8877e-05 eta: 5:52:59 time: 0.7497 data_time: 0.0009 memory: 6319 loss: 0.2480 +2023/06/06 03:51:27 - mmengine - INFO - Epoch(train) [4][1000/4092] lr: 7.8583e-05 eta: 5:51:41 time: 0.7437 data_time: 0.0010 memory: 6319 loss: 0.2589 +2023/06/06 03:52:41 - mmengine - INFO - Epoch(train) [4][1100/4092] lr: 7.8288e-05 eta: 5:50:21 time: 0.7458 data_time: 0.0009 memory: 6319 loss: 0.2309 +2023/06/06 03:53:58 - mmengine - INFO - Epoch(train) [4][1200/4092] lr: 7.7992e-05 eta: 5:49:06 time: 0.7534 data_time: 0.0010 memory: 6319 loss: 0.2401 +2023/06/06 03:55:13 - mmengine - INFO - Epoch(train) [4][1300/4092] lr: 7.7694e-05 eta: 5:47:47 time: 0.7990 data_time: 0.0010 memory: 6319 loss: 0.2420 +2023/06/06 03:56:28 - mmengine - INFO - Epoch(train) [4][1400/4092] lr: 7.7395e-05 eta: 5:46:28 time: 0.7417 data_time: 0.0012 memory: 6319 loss: 0.2452 +2023/06/06 03:57:42 - mmengine - INFO - Epoch(train) [4][1500/4092] lr: 7.7095e-05 eta: 5:45:06 time: 0.7450 data_time: 0.0010 memory: 6319 loss: 0.2393 +2023/06/06 03:58:58 - mmengine - INFO - Epoch(train) [4][1600/4092] lr: 7.6793e-05 eta: 5:43:49 time: 0.7494 data_time: 0.0008 memory: 6319 loss: 0.2357 +2023/06/06 04:00:12 - mmengine - INFO - Epoch(train) [4][1700/4092] lr: 7.6490e-05 eta: 5:42:30 time: 0.7087 data_time: 0.0009 memory: 6319 loss: 0.2427 +2023/06/06 04:00:33 - mmengine - INFO - Exp name: resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1_20230606_005813 +2023/06/06 04:01:30 - mmengine - INFO - Epoch(train) [4][1800/4092] lr: 7.6186e-05 eta: 5:41:16 time: 0.7755 data_time: 0.0008 memory: 6319 loss: 0.2352 +2023/06/06 04:02:43 - mmengine - INFO - Epoch(train) [4][1900/4092] lr: 7.5881e-05 eta: 5:39:54 time: 0.7532 data_time: 0.0009 memory: 6319 loss: 0.2549 +2023/06/06 04:03:59 - mmengine - INFO - Epoch(train) [4][2000/4092] lr: 7.5574e-05 eta: 5:38:36 time: 0.7648 data_time: 0.0009 memory: 6319 loss: 0.2572 +2023/06/06 04:05:14 - mmengine - INFO - Epoch(train) [4][2100/4092] lr: 7.5266e-05 eta: 5:37:19 time: 0.6891 data_time: 0.0009 memory: 6319 loss: 0.2320 +2023/06/06 04:06:31 - mmengine - INFO - Epoch(train) [4][2200/4092] lr: 7.4957e-05 eta: 5:36:03 time: 0.7295 data_time: 0.0009 memory: 6319 loss: 0.2601 +2023/06/06 04:07:45 - mmengine - INFO - Epoch(train) [4][2300/4092] lr: 7.4647e-05 eta: 5:34:44 time: 0.8304 data_time: 0.0008 memory: 6319 loss: 0.2433 +2023/06/06 04:09:01 - mmengine - INFO - Epoch(train) [4][2400/4092] lr: 7.4336e-05 eta: 5:33:26 time: 0.7421 data_time: 0.0009 memory: 6319 loss: 0.2347 +2023/06/06 04:10:16 - mmengine - INFO - Epoch(train) [4][2500/4092] lr: 7.4023e-05 eta: 5:32:08 time: 0.7426 data_time: 0.0011 memory: 6319 loss: 0.2413 +2023/06/06 04:11:30 - mmengine - INFO - Epoch(train) [4][2600/4092] lr: 7.3709e-05 eta: 5:30:49 time: 0.7354 data_time: 0.0010 memory: 6319 loss: 0.2345 +2023/06/06 04:12:44 - mmengine - INFO - Epoch(train) [4][2700/4092] lr: 7.3395e-05 eta: 5:29:28 time: 0.8386 data_time: 0.0010 memory: 6319 loss: 0.2358 +2023/06/06 04:13:01 - mmengine - INFO - Exp name: resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1_20230606_005813 +2023/06/06 04:14:07 - mmengine - INFO - Epoch(train) [4][2800/4092] lr: 7.3079e-05 eta: 5:28:24 time: 1.4188 data_time: 0.0009 memory: 6319 loss: 0.2593 +2023/06/06 04:15:25 - mmengine - INFO - Epoch(train) [4][2900/4092] lr: 7.2762e-05 eta: 5:27:10 time: 0.7593 data_time: 0.0009 memory: 6319 loss: 0.2396 +2023/06/06 04:16:43 - mmengine - INFO - Epoch(train) [4][3000/4092] lr: 7.2444e-05 eta: 5:25:56 time: 0.7510 data_time: 0.0009 memory: 6319 loss: 0.2313 +2023/06/06 04:17:58 - mmengine - INFO - Epoch(train) [4][3100/4092] lr: 7.2125e-05 eta: 5:24:38 time: 0.7700 data_time: 0.0008 memory: 6319 loss: 0.2398 +2023/06/06 04:19:15 - mmengine - INFO - Epoch(train) [4][3200/4092] lr: 7.1805e-05 eta: 5:23:23 time: 0.8533 data_time: 0.0011 memory: 6319 loss: 0.2264 +2023/06/06 04:20:32 - mmengine - INFO - Epoch(train) [4][3300/4092] lr: 7.1484e-05 eta: 5:22:09 time: 0.7617 data_time: 0.0010 memory: 6319 loss: 0.2567 +2023/06/06 04:21:48 - mmengine - INFO - Epoch(train) [4][3400/4092] lr: 7.1162e-05 eta: 5:20:52 time: 0.8312 data_time: 0.0008 memory: 6319 loss: 0.2266 +2023/06/06 04:23:05 - mmengine - INFO - Epoch(train) [4][3500/4092] lr: 7.0839e-05 eta: 5:19:37 time: 0.7463 data_time: 0.0010 memory: 6319 loss: 0.2386 +2023/06/06 04:24:21 - mmengine - INFO - Epoch(train) [4][3600/4092] lr: 7.0515e-05 eta: 5:18:21 time: 0.7131 data_time: 0.0008 memory: 6319 loss: 0.2441 +2023/06/06 04:25:34 - mmengine - INFO - Epoch(train) [4][3700/4092] lr: 7.0191e-05 eta: 5:16:59 time: 0.7135 data_time: 0.0009 memory: 6319 loss: 0.2493 +2023/06/06 04:25:49 - mmengine - INFO - Exp name: resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1_20230606_005813 +2023/06/06 04:26:46 - mmengine - INFO - Epoch(train) [4][3800/4092] lr: 6.9865e-05 eta: 5:15:36 time: 0.7190 data_time: 0.0008 memory: 6319 loss: 0.2373 +2023/06/06 04:28:00 - mmengine - INFO - Epoch(train) [4][3900/4092] lr: 6.9538e-05 eta: 5:14:16 time: 0.8029 data_time: 0.0009 memory: 6319 loss: 0.2515 +2023/06/06 04:29:16 - mmengine - INFO - Epoch(train) [4][4000/4092] lr: 6.9211e-05 eta: 5:12:59 time: 0.7749 data_time: 0.0009 memory: 6319 loss: 0.2331 +2023/06/06 04:30:24 - mmengine - INFO - Exp name: resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1_20230606_005813 +2023/06/06 04:30:24 - mmengine - INFO - Saving checkpoint at 4 epochs +2023/06/06 04:31:08 - mmengine - INFO - Epoch(val) [4][100/119] eta: 0:00:07 time: 0.4918 data_time: 0.4018 memory: 6319 +2023/06/06 04:31:32 - mmengine - INFO - Epoch(val) [4][119/119] accuracy/top1: 85.1083 data_time: 0.3380 time: 0.4257 +2023/06/06 04:32:48 - mmengine - INFO - Epoch(train) [5][ 100/4092] lr: 6.8580e-05 eta: 5:10:31 time: 0.7358 data_time: 0.3022 memory: 6319 loss: 0.2545 +2023/06/06 04:34:04 - mmengine - INFO - Epoch(train) [5][ 200/4092] lr: 6.8250e-05 eta: 5:09:14 time: 0.7363 data_time: 0.0010 memory: 6319 loss: 0.2320 +2023/06/06 04:35:19 - mmengine - INFO - Epoch(train) [5][ 300/4092] lr: 6.7920e-05 eta: 5:07:56 time: 0.7698 data_time: 0.0010 memory: 6319 loss: 0.2312 +2023/06/06 04:36:39 - mmengine - INFO - Epoch(train) [5][ 400/4092] lr: 6.7588e-05 eta: 5:06:45 time: 1.3156 data_time: 0.0007 memory: 6319 loss: 0.2396 +2023/06/06 04:37:58 - mmengine - INFO - Epoch(train) [5][ 500/4092] lr: 6.7256e-05 eta: 5:05:34 time: 0.7898 data_time: 0.0009 memory: 6319 loss: 0.2282 +2023/06/06 04:39:14 - mmengine - INFO - Epoch(train) [5][ 600/4092] lr: 6.6924e-05 eta: 5:04:17 time: 0.8250 data_time: 0.0009 memory: 6319 loss: 0.2323 +2023/06/06 04:39:42 - mmengine - INFO - Exp name: resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1_20230606_005813 +2023/06/06 04:40:27 - mmengine - INFO - Epoch(train) [5][ 700/4092] lr: 6.6590e-05 eta: 5:02:56 time: 0.7066 data_time: 0.0008 memory: 6319 loss: 0.2311 +2023/06/06 04:41:42 - mmengine - INFO - Epoch(train) [5][ 800/4092] lr: 6.6256e-05 eta: 5:01:38 time: 0.7500 data_time: 0.0008 memory: 6319 loss: 0.2330 +2023/06/06 04:42:58 - mmengine - INFO - Epoch(train) [5][ 900/4092] lr: 6.5921e-05 eta: 5:00:22 time: 0.7828 data_time: 0.0009 memory: 6319 loss: 0.2508 +2023/06/06 04:44:14 - mmengine - INFO - Epoch(train) [5][1000/4092] lr: 6.5586e-05 eta: 4:59:05 time: 0.7278 data_time: 0.0009 memory: 6319 loss: 0.2467 +2023/06/06 04:45:29 - mmengine - INFO - Epoch(train) [5][1100/4092] lr: 6.5250e-05 eta: 4:57:48 time: 0.7504 data_time: 0.0010 memory: 6319 loss: 0.2559 +2023/06/06 04:46:45 - mmengine - INFO - Epoch(train) [5][1200/4092] lr: 6.4913e-05 eta: 4:56:30 time: 0.7106 data_time: 0.0010 memory: 6319 loss: 0.2270 +2023/06/06 04:48:01 - mmengine - INFO - Epoch(train) [5][1300/4092] lr: 6.4576e-05 eta: 4:55:14 time: 0.7481 data_time: 0.0008 memory: 6319 loss: 0.2416 +2023/06/06 04:49:17 - mmengine - INFO - Epoch(train) [5][1400/4092] lr: 6.4238e-05 eta: 4:53:57 time: 0.7690 data_time: 0.0009 memory: 6319 loss: 0.2311 +2023/06/06 04:50:33 - mmengine - INFO - Epoch(train) [5][1500/4092] lr: 6.3899e-05 eta: 4:52:41 time: 0.7814 data_time: 0.0009 memory: 6319 loss: 0.2490 +2023/06/06 04:51:48 - mmengine - INFO - Epoch(train) [5][1600/4092] lr: 6.3560e-05 eta: 4:51:24 time: 0.7281 data_time: 0.0009 memory: 6319 loss: 0.2329 +2023/06/06 04:52:17 - mmengine - INFO - Exp name: resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1_20230606_005813 +2023/06/06 04:53:04 - mmengine - INFO - Epoch(train) [5][1700/4092] lr: 6.3221e-05 eta: 4:50:08 time: 0.7271 data_time: 0.0008 memory: 6319 loss: 0.2362 +2023/06/06 04:54:24 - mmengine - INFO - Epoch(train) [5][1800/4092] lr: 6.2881e-05 eta: 4:48:56 time: 0.7365 data_time: 0.0009 memory: 6319 loss: 0.2285 +2023/06/06 04:55:38 - mmengine - INFO - Epoch(train) [5][1900/4092] lr: 6.2541e-05 eta: 4:47:38 time: 0.7225 data_time: 0.0008 memory: 6319 loss: 0.2323 +2023/06/06 04:56:54 - mmengine - INFO - Epoch(train) [5][2000/4092] lr: 6.2200e-05 eta: 4:46:21 time: 0.8011 data_time: 0.0010 memory: 6319 loss: 0.2440 +2023/06/06 04:58:08 - mmengine - INFO - Epoch(train) [5][2100/4092] lr: 6.1859e-05 eta: 4:45:02 time: 0.7546 data_time: 0.0011 memory: 6319 loss: 0.2435 +2023/06/06 04:59:23 - mmengine - INFO - Epoch(train) [5][2200/4092] lr: 6.1517e-05 eta: 4:43:44 time: 0.7521 data_time: 0.0012 memory: 6319 loss: 0.2370 +2023/06/06 05:00:36 - mmengine - INFO - Epoch(train) [5][2300/4092] lr: 6.1175e-05 eta: 4:42:25 time: 0.7016 data_time: 0.0009 memory: 6319 loss: 0.2205 +2023/06/06 05:01:49 - mmengine - INFO - Epoch(train) [5][2400/4092] lr: 6.0833e-05 eta: 4:41:05 time: 0.7140 data_time: 0.0008 memory: 6319 loss: 0.2285 +2023/06/06 05:03:10 - mmengine - INFO - Epoch(train) [5][2500/4092] lr: 6.0490e-05 eta: 4:39:54 time: 0.6932 data_time: 0.0011 memory: 6319 loss: 0.2296 +2023/06/06 05:04:25 - mmengine - INFO - Epoch(train) [5][2600/4092] lr: 6.0147e-05 eta: 4:38:37 time: 0.7555 data_time: 0.0009 memory: 6319 loss: 0.2294 +2023/06/06 05:04:54 - mmengine - INFO - Exp name: resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1_20230606_005813 +2023/06/06 05:05:43 - mmengine - INFO - Epoch(train) [5][2700/4092] lr: 5.9803e-05 eta: 4:37:22 time: 0.7669 data_time: 0.0010 memory: 6319 loss: 0.2333 +2023/06/06 05:06:57 - mmengine - INFO - Epoch(train) [5][2800/4092] lr: 5.9460e-05 eta: 4:36:04 time: 0.8021 data_time: 0.0009 memory: 6319 loss: 0.2551 +2023/06/06 05:08:13 - mmengine - INFO - Epoch(train) [5][2900/4092] lr: 5.9116e-05 eta: 4:34:47 time: 0.7087 data_time: 0.0009 memory: 6319 loss: 0.2471 +2023/06/06 05:09:27 - mmengine - INFO - Epoch(train) [5][3000/4092] lr: 5.8772e-05 eta: 4:33:29 time: 0.7371 data_time: 0.0010 memory: 6319 loss: 0.2278 +2023/06/06 05:10:44 - mmengine - INFO - Epoch(train) [5][3100/4092] lr: 5.8427e-05 eta: 4:32:13 time: 0.7397 data_time: 0.0011 memory: 6319 loss: 0.2247 +2023/06/06 05:12:01 - mmengine - INFO - Epoch(train) [5][3200/4092] lr: 5.8083e-05 eta: 4:30:58 time: 0.7367 data_time: 0.0009 memory: 6319 loss: 0.2403 +2023/06/06 05:13:17 - mmengine - INFO - Epoch(train) [5][3300/4092] lr: 5.7738e-05 eta: 4:29:43 time: 0.7156 data_time: 0.0009 memory: 6319 loss: 0.2145 +2023/06/06 05:14:34 - mmengine - INFO - Epoch(train) [5][3400/4092] lr: 5.7393e-05 eta: 4:28:27 time: 0.7647 data_time: 0.0011 memory: 6319 loss: 0.2224 +2023/06/06 05:15:50 - mmengine - INFO - Epoch(train) [5][3500/4092] lr: 5.7048e-05 eta: 4:27:11 time: 0.8112 data_time: 0.0011 memory: 6319 loss: 0.2200 +2023/06/06 05:17:06 - mmengine - INFO - Epoch(train) [5][3600/4092] lr: 5.6703e-05 eta: 4:25:54 time: 0.7341 data_time: 0.0009 memory: 6319 loss: 0.2186 +2023/06/06 05:17:29 - mmengine - INFO - Exp name: resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1_20230606_005813 +2023/06/06 05:18:21 - mmengine - INFO - Epoch(train) [5][3700/4092] lr: 5.6358e-05 eta: 4:24:37 time: 0.7233 data_time: 0.0009 memory: 6319 loss: 0.2396 +2023/06/06 05:19:41 - mmengine - INFO - Epoch(train) [5][3800/4092] lr: 5.6012e-05 eta: 4:23:25 time: 0.7516 data_time: 0.0011 memory: 6319 loss: 0.2451 +2023/06/06 05:21:02 - mmengine - INFO - Epoch(train) [5][3900/4092] lr: 5.5667e-05 eta: 4:22:13 time: 0.7594 data_time: 0.0011 memory: 6319 loss: 0.2264 +2023/06/06 05:22:17 - mmengine - INFO - Epoch(train) [5][4000/4092] lr: 5.5321e-05 eta: 4:20:56 time: 0.7460 data_time: 0.0010 memory: 6319 loss: 0.2093 +2023/06/06 05:23:31 - mmengine - INFO - Exp name: resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1_20230606_005813 +2023/06/06 05:23:31 - mmengine - INFO - Saving checkpoint at 5 epochs +2023/06/06 05:24:16 - mmengine - INFO - Epoch(val) [5][100/119] eta: 0:00:07 time: 0.6853 data_time: 0.5955 memory: 6319 +2023/06/06 05:24:43 - mmengine - INFO - Epoch(val) [5][119/119] accuracy/top1: 88.7295 data_time: 0.3710 time: 0.4601 +2023/06/06 05:26:06 - mmengine - INFO - Epoch(train) [6][ 100/4092] lr: 5.4658e-05 eta: 4:18:40 time: 0.7422 data_time: 0.3618 memory: 6319 loss: 0.2210 +2023/06/06 05:27:21 - mmengine - INFO - Epoch(train) [6][ 200/4092] lr: 5.4313e-05 eta: 4:17:24 time: 0.7930 data_time: 0.0892 memory: 6319 loss: 0.2218 +2023/06/06 05:28:39 - mmengine - INFO - Epoch(train) [6][ 300/4092] lr: 5.3967e-05 eta: 4:16:09 time: 0.8125 data_time: 0.0008 memory: 6319 loss: 0.2273 +2023/06/06 05:29:55 - mmengine - INFO - Epoch(train) [6][ 400/4092] lr: 5.3622e-05 eta: 4:14:52 time: 0.7553 data_time: 0.0008 memory: 6319 loss: 0.2373 +2023/06/06 05:31:13 - mmengine - INFO - Epoch(train) [6][ 500/4092] lr: 5.3276e-05 eta: 4:13:38 time: 0.7640 data_time: 0.0010 memory: 6319 loss: 0.2372 +2023/06/06 05:31:44 - mmengine - INFO - Exp name: resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1_20230606_005813 +2023/06/06 05:32:30 - mmengine - INFO - Epoch(train) [6][ 600/4092] lr: 5.2931e-05 eta: 4:12:22 time: 0.7425 data_time: 0.0011 memory: 6319 loss: 0.2217 +2023/06/06 05:33:44 - mmengine - INFO - Epoch(train) [6][ 700/4092] lr: 5.2586e-05 eta: 4:11:04 time: 0.7136 data_time: 0.0010 memory: 6319 loss: 0.2179 +2023/06/06 05:36:30 - mmengine - INFO - Epoch(train) [6][ 800/4092] lr: 5.2241e-05 eta: 4:11:11 time: 0.7487 data_time: 0.0016 memory: 6319 loss: 0.2180 +2023/06/06 05:37:45 - mmengine - INFO - Epoch(train) [6][ 900/4092] lr: 5.1897e-05 eta: 4:09:52 time: 0.7653 data_time: 0.0009 memory: 6319 loss: 0.2441 +2023/06/06 05:39:12 - mmengine - INFO - Epoch(train) [6][1000/4092] lr: 5.1552e-05 eta: 4:08:45 time: 0.8699 data_time: 0.0007 memory: 6319 loss: 0.2157 +2023/06/06 05:40:31 - mmengine - INFO - Epoch(train) [6][1100/4092] lr: 5.1208e-05 eta: 4:07:30 time: 0.7318 data_time: 0.0012 memory: 6319 loss: 0.2166 +2023/06/06 05:41:47 - mmengine - INFO - Epoch(train) [6][1200/4092] lr: 5.0864e-05 eta: 4:06:13 time: 0.8226 data_time: 0.0010 memory: 6319 loss: 0.2266 +2023/06/06 05:43:01 - mmengine - INFO - Epoch(train) [6][1300/4092] lr: 5.0520e-05 eta: 4:04:54 time: 0.7309 data_time: 0.0009 memory: 6319 loss: 0.2154 +2023/06/06 05:44:16 - mmengine - INFO - Epoch(train) [6][1400/4092] lr: 5.0176e-05 eta: 4:03:36 time: 0.7729 data_time: 0.0010 memory: 6319 loss: 0.2330 +2023/06/06 05:45:32 - mmengine - INFO - Epoch(train) [6][1500/4092] lr: 4.9833e-05 eta: 4:02:18 time: 0.7289 data_time: 0.0009 memory: 6319 loss: 0.2136 +2023/06/06 05:46:03 - mmengine - INFO - Exp name: resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1_20230606_005813 +2023/06/06 05:46:47 - mmengine - INFO - Epoch(train) [6][1600/4092] lr: 4.9490e-05 eta: 4:01:01 time: 0.7354 data_time: 0.0009 memory: 6319 loss: 0.2199 +2023/06/06 05:48:03 - mmengine - INFO - Epoch(train) [6][1700/4092] lr: 4.9147e-05 eta: 3:59:43 time: 0.7417 data_time: 0.0008 memory: 6319 loss: 0.2292 +2023/06/06 05:49:18 - mmengine - INFO - Epoch(train) [6][1800/4092] lr: 4.8805e-05 eta: 3:58:25 time: 0.7459 data_time: 0.0011 memory: 6319 loss: 0.2248 +2023/06/06 05:50:33 - mmengine - INFO - Epoch(train) [6][1900/4092] lr: 4.8462e-05 eta: 3:57:07 time: 0.7584 data_time: 0.0009 memory: 6319 loss: 0.2131 +2023/06/06 05:51:48 - mmengine - INFO - Epoch(train) [6][2000/4092] lr: 4.8121e-05 eta: 3:55:49 time: 0.7301 data_time: 0.0010 memory: 6319 loss: 0.2117 +2023/06/06 05:53:04 - mmengine - INFO - Epoch(train) [6][2100/4092] lr: 4.7780e-05 eta: 3:54:32 time: 0.7756 data_time: 0.0014 memory: 6319 loss: 0.2200 +2023/06/06 05:54:31 - mmengine - INFO - Epoch(train) [6][2200/4092] lr: 4.7439e-05 eta: 3:53:24 time: 0.7684 data_time: 0.0009 memory: 6319 loss: 0.2077 +2023/06/06 05:55:46 - mmengine - INFO - Epoch(train) [6][2300/4092] lr: 4.7099e-05 eta: 3:52:05 time: 0.7135 data_time: 0.0009 memory: 6319 loss: 0.2112 +2023/06/06 05:57:01 - mmengine - INFO - Epoch(train) [6][2400/4092] lr: 4.6759e-05 eta: 3:50:47 time: 0.7669 data_time: 0.0009 memory: 6319 loss: 0.2421 +2023/06/06 05:58:16 - mmengine - INFO - Epoch(train) [6][2500/4092] lr: 4.6419e-05 eta: 3:49:30 time: 0.7091 data_time: 0.0011 memory: 6319 loss: 0.2262 +2023/06/06 05:58:45 - mmengine - INFO - Exp name: resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1_20230606_005813 +2023/06/06 05:59:30 - mmengine - INFO - Epoch(train) [6][2600/4092] lr: 4.6080e-05 eta: 3:48:11 time: 0.7463 data_time: 0.0008 memory: 6319 loss: 0.2426 +2023/06/06 06:00:41 - mmengine - INFO - Epoch(train) [6][2700/4092] lr: 4.5742e-05 eta: 3:46:50 time: 0.7600 data_time: 0.0009 memory: 6319 loss: 0.2342 +2023/06/06 06:01:51 - mmengine - INFO - Epoch(train) [6][2800/4092] lr: 4.5404e-05 eta: 3:45:28 time: 0.7358 data_time: 0.0010 memory: 6319 loss: 0.2428 +2023/06/06 06:03:04 - mmengine - INFO - Epoch(train) [6][2900/4092] lr: 4.5067e-05 eta: 3:44:09 time: 0.7264 data_time: 0.0011 memory: 6319 loss: 0.2073 +2023/06/06 06:04:17 - mmengine - INFO - Epoch(train) [6][3000/4092] lr: 4.4730e-05 eta: 3:42:50 time: 0.7639 data_time: 0.0010 memory: 6319 loss: 0.2154 +2023/06/06 06:05:32 - mmengine - INFO - Epoch(train) [6][3100/4092] lr: 4.4394e-05 eta: 3:41:32 time: 0.7721 data_time: 0.0010 memory: 6319 loss: 0.2066 +2023/06/06 06:06:47 - mmengine - INFO - Epoch(train) [6][3200/4092] lr: 4.4059e-05 eta: 3:40:14 time: 0.7492 data_time: 0.0012 memory: 6319 loss: 0.2201 +2023/06/06 06:08:03 - mmengine - INFO - Epoch(train) [6][3300/4092] lr: 4.3724e-05 eta: 3:38:57 time: 0.6997 data_time: 0.0010 memory: 6319 loss: 0.2286 +2023/06/06 06:09:17 - mmengine - INFO - Epoch(train) [6][3400/4092] lr: 4.3390e-05 eta: 3:37:39 time: 0.7599 data_time: 0.0009 memory: 6319 loss: 0.2251 +2023/06/06 06:10:32 - mmengine - INFO - Epoch(train) [6][3500/4092] lr: 4.3056e-05 eta: 3:36:21 time: 0.7302 data_time: 0.0008 memory: 6319 loss: 0.2147 +2023/06/06 06:11:02 - mmengine - INFO - Exp name: resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1_20230606_005813 +2023/06/06 06:11:47 - mmengine - INFO - Epoch(train) [6][3600/4092] lr: 4.2724e-05 eta: 3:35:04 time: 0.7878 data_time: 0.0011 memory: 6319 loss: 0.2023 +2023/06/06 06:13:03 - mmengine - INFO - Epoch(train) [6][3700/4092] lr: 4.2392e-05 eta: 3:33:46 time: 0.7331 data_time: 0.0008 memory: 6319 loss: 0.2047 +2023/06/06 06:14:18 - mmengine - INFO - Epoch(train) [6][3800/4092] lr: 4.2060e-05 eta: 3:32:29 time: 0.7452 data_time: 0.0010 memory: 6319 loss: 0.2449 +2023/06/06 06:15:35 - mmengine - INFO - Epoch(train) [6][3900/4092] lr: 4.1730e-05 eta: 3:31:12 time: 0.7946 data_time: 0.0010 memory: 6319 loss: 0.2128 +2023/06/06 06:16:49 - mmengine - INFO - Epoch(train) [6][4000/4092] lr: 4.1400e-05 eta: 3:29:55 time: 0.7635 data_time: 0.0008 memory: 6319 loss: 0.2154 +2023/06/06 06:17:56 - mmengine - INFO - Exp name: resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1_20230606_005813 +2023/06/06 06:17:56 - mmengine - INFO - Saving checkpoint at 6 epochs +2023/06/06 06:18:41 - mmengine - INFO - Epoch(val) [6][100/119] eta: 0:00:07 time: 0.6478 data_time: 0.5579 memory: 6319 +2023/06/06 06:19:08 - mmengine - INFO - Epoch(val) [6][119/119] accuracy/top1: 89.2938 data_time: 0.3655 time: 0.4556 +2023/06/06 06:20:27 - mmengine - INFO - Epoch(train) [7][ 100/4092] lr: 4.0769e-05 eta: 3:27:27 time: 0.7437 data_time: 0.0676 memory: 6319 loss: 0.2146 +2023/06/06 06:21:42 - mmengine - INFO - Epoch(train) [7][ 200/4092] lr: 4.0442e-05 eta: 3:26:10 time: 0.7826 data_time: 0.0009 memory: 6319 loss: 0.2203 +2023/06/06 06:22:59 - mmengine - INFO - Epoch(train) [7][ 300/4092] lr: 4.0116e-05 eta: 3:24:53 time: 0.8765 data_time: 0.0008 memory: 6319 loss: 0.2284 +2023/06/06 06:24:15 - mmengine - INFO - Epoch(train) [7][ 400/4092] lr: 3.9790e-05 eta: 3:23:36 time: 0.7793 data_time: 0.0011 memory: 6319 loss: 0.2285 +2023/06/06 06:24:52 - mmengine - INFO - Exp name: resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1_20230606_005813 +2023/06/06 06:25:31 - mmengine - INFO - Epoch(train) [7][ 500/4092] lr: 3.9465e-05 eta: 3:22:20 time: 0.6905 data_time: 0.0010 memory: 6319 loss: 0.2083 +2023/06/06 06:26:45 - mmengine - INFO - Epoch(train) [7][ 600/4092] lr: 3.9141e-05 eta: 3:21:01 time: 0.7519 data_time: 0.0012 memory: 6319 loss: 0.2073 +2023/06/06 06:28:00 - mmengine - INFO - Epoch(train) [7][ 700/4092] lr: 3.8819e-05 eta: 3:19:44 time: 0.8323 data_time: 0.0011 memory: 6319 loss: 0.2279 +2023/06/06 06:29:15 - mmengine - INFO - Epoch(train) [7][ 800/4092] lr: 3.8497e-05 eta: 3:18:27 time: 0.7189 data_time: 0.0010 memory: 6319 loss: 0.2198 +2023/06/06 06:30:31 - mmengine - INFO - Epoch(train) [7][ 900/4092] lr: 3.8176e-05 eta: 3:17:10 time: 0.7632 data_time: 0.0010 memory: 6319 loss: 0.2288 +2023/06/06 06:31:44 - mmengine - INFO - Epoch(train) [7][1000/4092] lr: 3.7856e-05 eta: 3:15:51 time: 0.7142 data_time: 0.0009 memory: 6319 loss: 0.2212 +2023/06/06 06:32:59 - mmengine - INFO - Epoch(train) [7][1100/4092] lr: 3.7537e-05 eta: 3:14:34 time: 0.7266 data_time: 0.0012 memory: 6319 loss: 0.2185 +2023/06/06 06:34:13 - mmengine - INFO - Epoch(train) [7][1200/4092] lr: 3.7219e-05 eta: 3:13:16 time: 0.7236 data_time: 0.0010 memory: 6319 loss: 0.2188 +2023/06/06 06:35:28 - mmengine - INFO - Epoch(train) [7][1300/4092] lr: 3.6902e-05 eta: 3:11:59 time: 0.7414 data_time: 0.0011 memory: 6319 loss: 0.2190 +2023/06/06 06:36:49 - mmengine - INFO - Epoch(train) [7][1400/4092] lr: 3.6586e-05 eta: 3:10:45 time: 0.7303 data_time: 0.0009 memory: 6319 loss: 0.2032 +2023/06/06 06:37:26 - mmengine - INFO - Exp name: resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1_20230606_005813 +2023/06/06 06:38:03 - mmengine - INFO - Epoch(train) [7][1500/4092] lr: 3.6272e-05 eta: 3:09:27 time: 0.7634 data_time: 0.0009 memory: 6319 loss: 0.2278 +2023/06/06 06:39:18 - mmengine - INFO - Epoch(train) [7][1600/4092] lr: 3.5958e-05 eta: 3:08:10 time: 0.7389 data_time: 0.0010 memory: 6319 loss: 0.2195 +2023/06/06 06:40:34 - mmengine - INFO - Epoch(train) [7][1700/4092] lr: 3.5646e-05 eta: 3:06:53 time: 0.7473 data_time: 0.0012 memory: 6319 loss: 0.2245 +2023/06/06 06:41:49 - mmengine - INFO - Epoch(train) [7][1800/4092] lr: 3.5334e-05 eta: 3:05:36 time: 0.6744 data_time: 0.0011 memory: 6319 loss: 0.2088 +2023/06/06 06:43:03 - mmengine - INFO - Epoch(train) [7][1900/4092] lr: 3.5024e-05 eta: 3:04:18 time: 0.7298 data_time: 0.0011 memory: 6319 loss: 0.2193 +2023/06/06 06:44:17 - mmengine - INFO - Epoch(train) [7][2000/4092] lr: 3.4715e-05 eta: 3:03:01 time: 0.7075 data_time: 0.0010 memory: 6319 loss: 0.2291 +2023/06/06 06:45:36 - mmengine - INFO - Epoch(train) [7][2100/4092] lr: 3.4407e-05 eta: 3:01:45 time: 0.7866 data_time: 0.0011 memory: 6319 loss: 0.2196 +2023/06/06 06:46:52 - mmengine - INFO - Epoch(train) [7][2200/4092] lr: 3.4101e-05 eta: 3:00:28 time: 0.7293 data_time: 0.0010 memory: 6319 loss: 0.2215 +2023/06/06 06:48:06 - mmengine - INFO - Epoch(train) [7][2300/4092] lr: 3.3796e-05 eta: 2:59:11 time: 0.7372 data_time: 0.0010 memory: 6319 loss: 0.2180 +2023/06/06 06:49:21 - mmengine - INFO - Epoch(train) [7][2400/4092] lr: 3.3491e-05 eta: 2:57:54 time: 0.7334 data_time: 0.0009 memory: 6319 loss: 0.2225 +2023/06/06 06:49:53 - mmengine - INFO - Exp name: resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1_20230606_005813 +2023/06/06 06:50:35 - mmengine - INFO - Epoch(train) [7][2500/4092] lr: 3.3189e-05 eta: 2:56:36 time: 0.7129 data_time: 0.0013 memory: 6319 loss: 0.2247 +2023/06/06 06:51:52 - mmengine - INFO - Epoch(train) [7][2600/4092] lr: 3.2887e-05 eta: 2:55:20 time: 0.7457 data_time: 0.0011 memory: 6319 loss: 0.2133 +2023/06/06 06:53:07 - mmengine - INFO - Epoch(train) [7][2700/4092] lr: 3.2587e-05 eta: 2:54:03 time: 0.7748 data_time: 0.0010 memory: 6319 loss: 0.2101 +2023/06/06 06:54:25 - mmengine - INFO - Epoch(train) [7][2800/4092] lr: 3.2288e-05 eta: 2:52:47 time: 0.7487 data_time: 0.0008 memory: 6319 loss: 0.2170 +2023/06/06 06:55:39 - mmengine - INFO - Epoch(train) [7][2900/4092] lr: 3.1990e-05 eta: 2:51:30 time: 0.7505 data_time: 0.0008 memory: 6319 loss: 0.2392 +2023/06/06 06:56:54 - mmengine - INFO - Epoch(train) [7][3000/4092] lr: 3.1694e-05 eta: 2:50:12 time: 0.7179 data_time: 0.0009 memory: 6319 loss: 0.2333 +2023/06/06 06:58:10 - mmengine - INFO - Epoch(train) [7][3100/4092] lr: 3.1399e-05 eta: 2:48:56 time: 0.7442 data_time: 0.0008 memory: 6319 loss: 0.2169 +2023/06/06 06:59:23 - mmengine - INFO - Epoch(train) [7][3200/4092] lr: 3.1106e-05 eta: 2:47:38 time: 0.7484 data_time: 0.0010 memory: 6319 loss: 0.2172 +2023/06/06 07:00:35 - mmengine - INFO - Epoch(train) [7][3300/4092] lr: 3.0814e-05 eta: 2:46:19 time: 0.6734 data_time: 0.0008 memory: 6319 loss: 0.2253 +2023/06/06 07:01:47 - mmengine - INFO - Epoch(train) [7][3400/4092] lr: 3.0523e-05 eta: 2:45:01 time: 0.7776 data_time: 0.0008 memory: 6319 loss: 0.2241 +2023/06/06 07:02:19 - mmengine - INFO - Exp name: resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1_20230606_005813 +2023/06/06 07:03:02 - mmengine - INFO - Epoch(train) [7][3500/4092] lr: 3.0234e-05 eta: 2:43:44 time: 0.7468 data_time: 0.0008 memory: 6319 loss: 0.2086 +2023/06/06 07:04:17 - mmengine - INFO - Epoch(train) [7][3600/4092] lr: 2.9946e-05 eta: 2:42:27 time: 0.7438 data_time: 0.0011 memory: 6319 loss: 0.2213 +2023/06/06 07:05:32 - mmengine - INFO - Epoch(train) [7][3700/4092] lr: 2.9660e-05 eta: 2:41:10 time: 0.7352 data_time: 0.0008 memory: 6319 loss: 0.2156 +2023/06/06 07:06:47 - mmengine - INFO - Epoch(train) [7][3800/4092] lr: 2.9375e-05 eta: 2:39:53 time: 0.7198 data_time: 0.0010 memory: 6319 loss: 0.2027 +2023/06/06 07:08:02 - mmengine - INFO - Epoch(train) [7][3900/4092] lr: 2.9092e-05 eta: 2:38:36 time: 0.7665 data_time: 0.0010 memory: 6319 loss: 0.2086 +2023/06/06 07:09:15 - mmengine - INFO - Epoch(train) [7][4000/4092] lr: 2.8810e-05 eta: 2:37:18 time: 0.7162 data_time: 0.0010 memory: 6319 loss: 0.2280 +2023/06/06 07:10:22 - mmengine - INFO - Exp name: resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1_20230606_005813 +2023/06/06 07:10:22 - mmengine - INFO - Saving checkpoint at 7 epochs +2023/06/06 07:11:08 - mmengine - INFO - Epoch(val) [7][100/119] eta: 0:00:07 time: 0.7160 data_time: 0.6154 memory: 6319 +2023/06/06 07:11:34 - mmengine - INFO - Epoch(val) [7][119/119] accuracy/top1: 90.8644 data_time: 0.3731 time: 0.4618 +2023/06/06 07:12:51 - mmengine - INFO - Epoch(train) [8][ 100/4092] lr: 2.8274e-05 eta: 2:34:51 time: 0.7781 data_time: 0.2035 memory: 6319 loss: 0.2022 +2023/06/06 07:14:06 - mmengine - INFO - Epoch(train) [8][ 200/4092] lr: 2.7997e-05 eta: 2:33:34 time: 0.7858 data_time: 0.0012 memory: 6319 loss: 0.2192 +2023/06/06 07:15:21 - mmengine - INFO - Epoch(train) [8][ 300/4092] lr: 2.7721e-05 eta: 2:32:17 time: 0.7734 data_time: 0.0010 memory: 6319 loss: 0.2223 +2023/06/06 07:16:04 - mmengine - INFO - Exp name: resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1_20230606_005813 +2023/06/06 07:16:35 - mmengine - INFO - Epoch(train) [8][ 400/4092] lr: 2.7447e-05 eta: 2:31:00 time: 0.7600 data_time: 0.0010 memory: 6319 loss: 0.2042 +2023/06/06 07:17:53 - mmengine - INFO - Epoch(train) [8][ 500/4092] lr: 2.7175e-05 eta: 2:29:44 time: 0.7366 data_time: 0.0008 memory: 6319 loss: 0.2056 +2023/06/06 07:19:07 - mmengine - INFO - Epoch(train) [8][ 600/4092] lr: 2.6904e-05 eta: 2:28:27 time: 0.7605 data_time: 0.0008 memory: 6319 loss: 0.2115 +2023/06/06 07:20:24 - mmengine - INFO - Epoch(train) [8][ 700/4092] lr: 2.6635e-05 eta: 2:27:11 time: 0.8980 data_time: 0.0009 memory: 6319 loss: 0.2169 +2023/06/06 07:21:41 - mmengine - INFO - Epoch(train) [8][ 800/4092] lr: 2.6368e-05 eta: 2:25:55 time: 0.7452 data_time: 0.0008 memory: 6319 loss: 0.2182 +2023/06/06 07:22:56 - mmengine - INFO - Epoch(train) [8][ 900/4092] lr: 2.6102e-05 eta: 2:24:38 time: 0.7357 data_time: 0.0008 memory: 6319 loss: 0.2177 +2023/06/06 07:24:10 - mmengine - INFO - Epoch(train) [8][1000/4092] lr: 2.5838e-05 eta: 2:23:21 time: 0.7183 data_time: 0.0008 memory: 6319 loss: 0.2219 +2023/06/06 07:25:25 - mmengine - INFO - Epoch(train) [8][1100/4092] lr: 2.5576e-05 eta: 2:22:04 time: 0.7524 data_time: 0.0008 memory: 6319 loss: 0.2043 +2023/06/06 07:26:43 - mmengine - INFO - Epoch(train) [8][1200/4092] lr: 2.5315e-05 eta: 2:20:49 time: 0.7501 data_time: 0.0009 memory: 6319 loss: 0.2100 +2023/06/06 07:27:59 - mmengine - INFO - Epoch(train) [8][1300/4092] lr: 2.5056e-05 eta: 2:19:32 time: 0.7853 data_time: 0.0010 memory: 6319 loss: 0.2195 +2023/06/06 07:28:43 - mmengine - INFO - Exp name: resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1_20230606_005813 +2023/06/06 07:29:17 - mmengine - INFO - Epoch(train) [8][1400/4092] lr: 2.4799e-05 eta: 2:18:16 time: 0.8273 data_time: 0.0010 memory: 6319 loss: 0.2153 +2023/06/06 07:30:31 - mmengine - INFO - Epoch(train) [8][1500/4092] lr: 2.4544e-05 eta: 2:16:59 time: 0.7533 data_time: 0.0011 memory: 6319 loss: 0.2166 +2023/06/06 07:31:46 - mmengine - INFO - Epoch(train) [8][1600/4092] lr: 2.4291e-05 eta: 2:15:43 time: 0.7239 data_time: 0.0009 memory: 6319 loss: 0.2020 +2023/06/06 07:33:05 - mmengine - INFO - Epoch(train) [8][1700/4092] lr: 2.4039e-05 eta: 2:14:27 time: 0.7509 data_time: 0.0010 memory: 6319 loss: 0.2096 +2023/06/06 07:34:21 - mmengine - INFO - Epoch(train) [8][1800/4092] lr: 2.3789e-05 eta: 2:13:11 time: 0.7533 data_time: 0.0011 memory: 6319 loss: 0.2192 +2023/06/06 07:35:36 - mmengine - INFO - Epoch(train) [8][1900/4092] lr: 2.3541e-05 eta: 2:11:54 time: 0.7710 data_time: 0.0011 memory: 6319 loss: 0.2098 +2023/06/06 07:36:51 - mmengine - INFO - Epoch(train) [8][2000/4092] lr: 2.3295e-05 eta: 2:10:37 time: 0.7488 data_time: 0.0009 memory: 6319 loss: 0.2031 +2023/06/06 07:38:10 - mmengine - INFO - Epoch(train) [8][2100/4092] lr: 2.3051e-05 eta: 2:09:22 time: 0.7827 data_time: 0.0010 memory: 6319 loss: 0.2172 +2023/06/06 07:39:24 - mmengine - INFO - Epoch(train) [8][2200/4092] lr: 2.2809e-05 eta: 2:08:05 time: 0.7382 data_time: 0.0010 memory: 6319 loss: 0.2142 +2023/06/06 07:40:39 - mmengine - INFO - Epoch(train) [8][2300/4092] lr: 2.2568e-05 eta: 2:06:49 time: 0.7396 data_time: 0.0009 memory: 6319 loss: 0.2159 +2023/06/06 07:41:18 - mmengine - INFO - Exp name: resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1_20230606_005813 +2023/06/06 07:41:55 - mmengine - INFO - Epoch(train) [8][2400/4092] lr: 2.2330e-05 eta: 2:05:32 time: 0.8223 data_time: 0.0009 memory: 6319 loss: 0.2172 +2023/06/06 07:43:09 - mmengine - INFO - Epoch(train) [8][2500/4092] lr: 2.2093e-05 eta: 2:04:15 time: 0.7885 data_time: 0.0010 memory: 6319 loss: 0.1949 +2023/06/06 07:44:24 - mmengine - INFO - Epoch(train) [8][2600/4092] lr: 2.1858e-05 eta: 2:02:58 time: 0.7555 data_time: 0.0010 memory: 6319 loss: 0.2133 +2023/06/06 07:45:38 - mmengine - INFO - Epoch(train) [8][2700/4092] lr: 2.1626e-05 eta: 2:01:41 time: 0.7514 data_time: 0.0010 memory: 6319 loss: 0.2072 +2023/06/06 07:46:55 - mmengine - INFO - Epoch(train) [8][2800/4092] lr: 2.1395e-05 eta: 2:00:25 time: 0.8021 data_time: 0.0012 memory: 6319 loss: 0.2115 +2023/06/06 07:48:08 - mmengine - INFO - Epoch(train) [8][2900/4092] lr: 2.1166e-05 eta: 1:59:08 time: 0.7590 data_time: 0.0011 memory: 6319 loss: 0.2106 +2023/06/06 07:49:24 - mmengine - INFO - Epoch(train) [8][3000/4092] lr: 2.0939e-05 eta: 1:57:52 time: 0.7446 data_time: 0.0009 memory: 6319 loss: 0.2148 +2023/06/06 07:50:42 - mmengine - INFO - Epoch(train) [8][3100/4092] lr: 2.0715e-05 eta: 1:56:36 time: 0.7885 data_time: 0.0009 memory: 6319 loss: 0.2003 +2023/06/06 07:51:57 - mmengine - INFO - Epoch(train) [8][3200/4092] lr: 2.0492e-05 eta: 1:55:19 time: 0.7916 data_time: 0.0012 memory: 6319 loss: 0.2210 +2023/06/06 07:53:12 - mmengine - INFO - Epoch(train) [8][3300/4092] lr: 2.0271e-05 eta: 1:54:03 time: 0.7740 data_time: 0.0011 memory: 6319 loss: 0.2124 +2023/06/06 07:53:50 - mmengine - INFO - Exp name: resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1_20230606_005813 +2023/06/06 07:54:27 - mmengine - INFO - Epoch(train) [8][3400/4092] lr: 2.0052e-05 eta: 1:52:46 time: 0.7203 data_time: 0.0011 memory: 6319 loss: 0.2249 +2023/06/06 07:55:43 - mmengine - INFO - Epoch(train) [8][3500/4092] lr: 1.9836e-05 eta: 1:51:30 time: 0.7695 data_time: 0.0009 memory: 6319 loss: 0.2137 +2023/06/06 07:56:58 - mmengine - INFO - Epoch(train) [8][3600/4092] lr: 1.9621e-05 eta: 1:50:14 time: 0.7751 data_time: 0.0010 memory: 6319 loss: 0.2158 +2023/06/06 07:58:15 - mmengine - INFO - Epoch(train) [8][3700/4092] lr: 1.9409e-05 eta: 1:48:57 time: 0.7498 data_time: 0.0011 memory: 6319 loss: 0.2102 +2023/06/06 07:59:30 - mmengine - INFO - Epoch(train) [8][3800/4092] lr: 1.9198e-05 eta: 1:47:41 time: 0.7471 data_time: 0.0011 memory: 6319 loss: 0.2317 +2023/06/06 08:00:45 - mmengine - INFO - Epoch(train) [8][3900/4092] lr: 1.8990e-05 eta: 1:46:24 time: 0.7907 data_time: 0.0009 memory: 6319 loss: 0.2089 +2023/06/06 08:02:02 - mmengine - INFO - Epoch(train) [8][4000/4092] lr: 1.8784e-05 eta: 1:45:08 time: 0.7369 data_time: 0.0008 memory: 6319 loss: 0.2050 +2023/06/06 08:03:08 - mmengine - INFO - Exp name: resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1_20230606_005813 +2023/06/06 08:03:08 - mmengine - INFO - Saving checkpoint at 8 epochs +2023/06/06 08:03:53 - mmengine - INFO - Epoch(val) [8][100/119] eta: 0:00:07 time: 0.7422 data_time: 0.6513 memory: 6319 +2023/06/06 08:04:20 - mmengine - INFO - Epoch(val) [8][119/119] accuracy/top1: 91.2815 data_time: 0.3766 time: 0.4667 +2023/06/06 08:05:37 - mmengine - INFO - Epoch(train) [9][ 100/4092] lr: 1.8394e-05 eta: 1:42:41 time: 0.7444 data_time: 0.1491 memory: 6319 loss: 0.2149 +2023/06/06 08:06:52 - mmengine - INFO - Epoch(train) [9][ 200/4092] lr: 1.8194e-05 eta: 1:41:24 time: 0.7720 data_time: 0.0010 memory: 6319 loss: 0.2059 +2023/06/06 08:07:43 - mmengine - INFO - Exp name: resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1_20230606_005813 +2023/06/06 08:08:07 - mmengine - INFO - Epoch(train) [9][ 300/4092] lr: 1.7997e-05 eta: 1:40:08 time: 0.7507 data_time: 0.0010 memory: 6319 loss: 0.2010 +2023/06/06 08:09:21 - mmengine - INFO - Epoch(train) [9][ 400/4092] lr: 1.7801e-05 eta: 1:38:51 time: 0.7334 data_time: 0.0011 memory: 6319 loss: 0.2070 +2023/06/06 08:10:36 - mmengine - INFO - Epoch(train) [9][ 500/4092] lr: 1.7608e-05 eta: 1:37:35 time: 0.7890 data_time: 0.0011 memory: 6319 loss: 0.2013 +2023/06/06 08:11:52 - mmengine - INFO - Epoch(train) [9][ 600/4092] lr: 1.7417e-05 eta: 1:36:18 time: 0.7321 data_time: 0.0009 memory: 6319 loss: 0.1978 +2023/06/06 08:13:06 - mmengine - INFO - Epoch(train) [9][ 700/4092] lr: 1.7228e-05 eta: 1:35:02 time: 0.7511 data_time: 0.0009 memory: 6319 loss: 0.2191 +2023/06/06 08:14:20 - mmengine - INFO - Epoch(train) [9][ 800/4092] lr: 1.7041e-05 eta: 1:33:45 time: 0.7092 data_time: 0.0011 memory: 6319 loss: 0.2122 +2023/06/06 08:15:36 - mmengine - INFO - Epoch(train) [9][ 900/4092] lr: 1.6857e-05 eta: 1:32:29 time: 0.7433 data_time: 0.0010 memory: 6319 loss: 0.2143 +2023/06/06 08:16:51 - mmengine - INFO - Epoch(train) [9][1000/4092] lr: 1.6675e-05 eta: 1:31:13 time: 0.7043 data_time: 0.0011 memory: 6319 loss: 0.2056 +2023/06/06 08:18:06 - mmengine - INFO - Epoch(train) [9][1100/4092] lr: 1.6495e-05 eta: 1:29:56 time: 0.6958 data_time: 0.0009 memory: 6319 loss: 0.2154 +2023/06/06 08:19:21 - mmengine - INFO - Epoch(train) [9][1200/4092] lr: 1.6317e-05 eta: 1:28:40 time: 0.7735 data_time: 0.0010 memory: 6319 loss: 0.2105 +2023/06/06 08:20:11 - mmengine - INFO - Exp name: resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1_20230606_005813 +2023/06/06 08:20:38 - mmengine - INFO - Epoch(train) [9][1300/4092] lr: 1.6142e-05 eta: 1:27:24 time: 0.7008 data_time: 0.0011 memory: 6319 loss: 0.2146 +2023/06/06 08:21:53 - mmengine - INFO - Epoch(train) [9][1400/4092] lr: 1.5969e-05 eta: 1:26:07 time: 0.7708 data_time: 0.0009 memory: 6319 loss: 0.2127 +2023/06/06 08:23:09 - mmengine - INFO - Epoch(train) [9][1500/4092] lr: 1.5798e-05 eta: 1:24:51 time: 0.7452 data_time: 0.0010 memory: 6319 loss: 0.2144 +2023/06/06 08:24:27 - mmengine - INFO - Epoch(train) [9][1600/4092] lr: 1.5629e-05 eta: 1:23:35 time: 0.7562 data_time: 0.0010 memory: 6319 loss: 0.2263 +2023/06/06 08:25:43 - mmengine - INFO - Epoch(train) [9][1700/4092] lr: 1.5463e-05 eta: 1:22:19 time: 0.7675 data_time: 0.0010 memory: 6319 loss: 0.2165 +2023/06/06 08:26:59 - mmengine - INFO - Epoch(train) [9][1800/4092] lr: 1.5299e-05 eta: 1:21:03 time: 0.7118 data_time: 0.0011 memory: 6319 loss: 0.2004 +2023/06/06 08:28:16 - mmengine - INFO - Epoch(train) [9][1900/4092] lr: 1.5138e-05 eta: 1:19:47 time: 0.7517 data_time: 0.0010 memory: 6319 loss: 0.2170 +2023/06/06 08:29:32 - mmengine - INFO - Epoch(train) [9][2000/4092] lr: 1.4979e-05 eta: 1:18:30 time: 0.7448 data_time: 0.0010 memory: 6319 loss: 0.2017 +2023/06/06 08:30:45 - mmengine - INFO - Epoch(train) [9][2100/4092] lr: 1.4822e-05 eta: 1:17:14 time: 0.7401 data_time: 0.0010 memory: 6319 loss: 0.1936 +2023/06/06 08:31:59 - mmengine - INFO - Epoch(train) [9][2200/4092] lr: 1.4668e-05 eta: 1:15:57 time: 0.7106 data_time: 0.0010 memory: 6319 loss: 0.2115 +2023/06/06 08:32:46 - mmengine - INFO - Exp name: resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1_20230606_005813 +2023/06/06 08:33:15 - mmengine - INFO - Epoch(train) [9][2300/4092] lr: 1.4515e-05 eta: 1:14:41 time: 0.7421 data_time: 0.0010 memory: 6319 loss: 0.1955 +2023/06/06 08:34:31 - mmengine - INFO - Epoch(train) [9][2400/4092] lr: 1.4366e-05 eta: 1:13:25 time: 0.7416 data_time: 0.0009 memory: 6319 loss: 0.2198 +2023/06/06 08:35:45 - mmengine - INFO - Epoch(train) [9][2500/4092] lr: 1.4219e-05 eta: 1:12:08 time: 0.8412 data_time: 0.0012 memory: 6319 loss: 0.2145 +2023/06/06 08:37:02 - mmengine - INFO - Epoch(train) [9][2600/4092] lr: 1.4074e-05 eta: 1:10:52 time: 0.7456 data_time: 0.0010 memory: 6319 loss: 0.2063 +2023/06/06 08:38:17 - mmengine - INFO - Epoch(train) [9][2700/4092] lr: 1.3931e-05 eta: 1:09:36 time: 0.8046 data_time: 0.0009 memory: 6319 loss: 0.1999 +2023/06/06 08:39:40 - mmengine - INFO - Epoch(train) [9][2800/4092] lr: 1.3791e-05 eta: 1:08:21 time: 0.7733 data_time: 0.0008 memory: 6319 loss: 0.2130 +2023/06/06 08:40:56 - mmengine - INFO - Epoch(train) [9][2900/4092] lr: 1.3654e-05 eta: 1:07:05 time: 0.7395 data_time: 0.0008 memory: 6319 loss: 0.2052 +2023/06/06 08:42:11 - mmengine - INFO - Epoch(train) [9][3000/4092] lr: 1.3519e-05 eta: 1:05:48 time: 0.7833 data_time: 0.0010 memory: 6319 loss: 0.2027 +2023/06/06 08:43:27 - mmengine - INFO - Epoch(train) [9][3100/4092] lr: 1.3386e-05 eta: 1:04:32 time: 0.8056 data_time: 0.0010 memory: 6319 loss: 0.2251 +2023/06/06 08:44:42 - mmengine - INFO - Epoch(train) [9][3200/4092] lr: 1.3256e-05 eta: 1:03:16 time: 0.7167 data_time: 0.0010 memory: 6319 loss: 0.2165 +2023/06/06 08:45:29 - mmengine - INFO - Exp name: resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1_20230606_005813 +2023/06/06 08:46:00 - mmengine - INFO - Epoch(train) [9][3300/4092] lr: 1.3128e-05 eta: 1:02:00 time: 0.7907 data_time: 0.0010 memory: 6319 loss: 0.2154 +2023/06/06 08:47:15 - mmengine - INFO - Epoch(train) [9][3400/4092] lr: 1.3003e-05 eta: 1:00:44 time: 0.7522 data_time: 0.0014 memory: 6319 loss: 0.2204 +2023/06/06 08:48:33 - mmengine - INFO - Epoch(train) [9][3500/4092] lr: 1.2880e-05 eta: 0:59:28 time: 0.7013 data_time: 0.0009 memory: 6319 loss: 0.2208 +2023/06/06 08:49:48 - mmengine - INFO - Epoch(train) [9][3600/4092] lr: 1.2759e-05 eta: 0:58:11 time: 0.7600 data_time: 0.0010 memory: 6319 loss: 0.2284 +2023/06/06 08:51:03 - mmengine - INFO - Epoch(train) [9][3700/4092] lr: 1.2641e-05 eta: 0:56:55 time: 0.7758 data_time: 0.0008 memory: 6319 loss: 0.2244 +2023/06/06 08:52:20 - mmengine - INFO - Epoch(train) [9][3800/4092] lr: 1.2526e-05 eta: 0:55:39 time: 0.8104 data_time: 0.0010 memory: 6319 loss: 0.2115 +2023/06/06 08:53:34 - mmengine - INFO - Epoch(train) [9][3900/4092] lr: 1.2413e-05 eta: 0:54:22 time: 0.7465 data_time: 0.0010 memory: 6319 loss: 0.2135 +2023/06/06 08:54:51 - mmengine - INFO - Epoch(train) [9][4000/4092] lr: 1.2303e-05 eta: 0:53:06 time: 0.7471 data_time: 0.0010 memory: 6319 loss: 0.2111 +2023/06/06 08:55:59 - mmengine - INFO - Exp name: resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1_20230606_005813 +2023/06/06 08:55:59 - mmengine - INFO - Saving checkpoint at 9 epochs +2023/06/06 08:56:45 - mmengine - INFO - Epoch(val) [9][100/119] eta: 0:00:07 time: 0.7141 data_time: 0.6247 memory: 6319 +2023/06/06 08:57:11 - mmengine - INFO - Epoch(val) [9][119/119] accuracy/top1: 91.4784 data_time: 0.3669 time: 0.4550 +2023/06/06 08:58:31 - mmengine - INFO - Epoch(train) [10][ 100/4092] lr: 1.2098e-05 eta: 0:50:40 time: 0.7676 data_time: 0.1648 memory: 6319 loss: 0.2166 +2023/06/06 08:59:30 - mmengine - INFO - Exp name: resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1_20230606_005813 +2023/06/06 08:59:50 - mmengine - INFO - Epoch(train) [10][ 200/4092] lr: 1.1995e-05 eta: 0:49:24 time: 0.7874 data_time: 0.0011 memory: 6319 loss: 0.2215 +2023/06/06 09:01:04 - mmengine - INFO - Epoch(train) [10][ 300/4092] lr: 1.1895e-05 eta: 0:48:08 time: 0.7119 data_time: 0.0010 memory: 6319 loss: 0.2005 +2023/06/06 09:02:19 - mmengine - INFO - Epoch(train) [10][ 400/4092] lr: 1.1797e-05 eta: 0:46:52 time: 0.7177 data_time: 0.0009 memory: 6319 loss: 0.2159 +2023/06/06 09:03:35 - mmengine - INFO - Epoch(train) [10][ 500/4092] lr: 1.1701e-05 eta: 0:45:36 time: 0.7312 data_time: 0.0011 memory: 6319 loss: 0.2015 +2023/06/06 09:04:49 - mmengine - INFO - Epoch(train) [10][ 600/4092] lr: 1.1608e-05 eta: 0:44:19 time: 0.7109 data_time: 0.0010 memory: 6319 loss: 0.2013 +2023/06/06 09:06:03 - mmengine - INFO - Epoch(train) [10][ 700/4092] lr: 1.1518e-05 eta: 0:43:03 time: 0.7603 data_time: 0.0009 memory: 6319 loss: 0.2202 +2023/06/06 09:07:19 - mmengine - INFO - Epoch(train) [10][ 800/4092] lr: 1.1430e-05 eta: 0:41:47 time: 0.7416 data_time: 0.0010 memory: 6319 loss: 0.2060 +2023/06/06 09:08:34 - mmengine - INFO - Epoch(train) [10][ 900/4092] lr: 1.1345e-05 eta: 0:40:30 time: 0.7928 data_time: 0.0010 memory: 6319 loss: 0.2178 +2023/06/06 09:09:51 - mmengine - INFO - Epoch(train) [10][1000/4092] lr: 1.1263e-05 eta: 0:39:14 time: 0.7332 data_time: 0.0010 memory: 6319 loss: 0.2236 +2023/06/06 09:11:10 - mmengine - INFO - Epoch(train) [10][1100/4092] lr: 1.1183e-05 eta: 0:37:58 time: 0.7501 data_time: 0.0013 memory: 6319 loss: 0.1999 +2023/06/06 09:12:04 - mmengine - INFO - Exp name: resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1_20230606_005813 +2023/06/06 09:12:27 - mmengine - INFO - Epoch(train) [10][1200/4092] lr: 1.1105e-05 eta: 0:36:42 time: 0.7648 data_time: 0.0010 memory: 6319 loss: 0.2071 +2023/06/06 09:13:43 - mmengine - INFO - Epoch(train) [10][1300/4092] lr: 1.1031e-05 eta: 0:35:26 time: 0.9441 data_time: 0.0011 memory: 6319 loss: 0.2190 +2023/06/06 09:14:57 - mmengine - INFO - Epoch(train) [10][1400/4092] lr: 1.0958e-05 eta: 0:34:10 time: 0.7064 data_time: 0.0011 memory: 6319 loss: 0.2044 +2023/06/06 09:16:14 - mmengine - INFO - Epoch(train) [10][1500/4092] lr: 1.0889e-05 eta: 0:32:54 time: 0.7076 data_time: 0.0011 memory: 6319 loss: 0.2198 +2023/06/06 09:17:29 - mmengine - INFO - Epoch(train) [10][1600/4092] lr: 1.0822e-05 eta: 0:31:37 time: 0.7296 data_time: 0.0010 memory: 6319 loss: 0.2096 +2023/06/06 09:18:45 - mmengine - INFO - Epoch(train) [10][1700/4092] lr: 1.0757e-05 eta: 0:30:21 time: 0.7686 data_time: 0.0009 memory: 6319 loss: 0.2070 +2023/06/06 09:20:04 - mmengine - INFO - Epoch(train) [10][1800/4092] lr: 1.0696e-05 eta: 0:29:05 time: 0.7310 data_time: 0.0010 memory: 6319 loss: 0.2095 +2023/06/06 09:21:18 - mmengine - INFO - Epoch(train) [10][1900/4092] lr: 1.0636e-05 eta: 0:27:49 time: 0.7421 data_time: 0.0016 memory: 6319 loss: 0.2009 +2023/06/06 09:22:32 - mmengine - INFO - Epoch(train) [10][2000/4092] lr: 1.0580e-05 eta: 0:26:33 time: 0.7543 data_time: 0.0015 memory: 6319 loss: 0.2024 +2023/06/06 09:23:52 - mmengine - INFO - Epoch(train) [10][2100/4092] lr: 1.0526e-05 eta: 0:25:17 time: 0.8326 data_time: 0.0013 memory: 6319 loss: 0.1982 +2023/06/06 09:24:47 - mmengine - INFO - Exp name: resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1_20230606_005813 +2023/06/06 09:25:14 - mmengine - INFO - Epoch(train) [10][2200/4092] lr: 1.0474e-05 eta: 0:24:01 time: 1.1544 data_time: 0.0011 memory: 6319 loss: 0.2192 +2023/06/06 09:26:31 - mmengine - INFO - Epoch(train) [10][2300/4092] lr: 1.0426e-05 eta: 0:22:45 time: 0.7191 data_time: 0.0009 memory: 6319 loss: 0.1905 +2023/06/06 09:27:46 - mmengine - INFO - Epoch(train) [10][2400/4092] lr: 1.0380e-05 eta: 0:21:28 time: 0.7957 data_time: 0.0013 memory: 6319 loss: 0.2189 +2023/06/06 09:29:02 - mmengine - INFO - Epoch(train) [10][2500/4092] lr: 1.0336e-05 eta: 0:20:12 time: 0.7244 data_time: 0.0012 memory: 6319 loss: 0.2137 +2023/06/06 09:30:16 - mmengine - INFO - Epoch(train) [10][2600/4092] lr: 1.0295e-05 eta: 0:18:56 time: 0.7591 data_time: 0.0010 memory: 6319 loss: 0.1928 +2023/06/06 09:31:32 - mmengine - INFO - Epoch(train) [10][2700/4092] lr: 1.0257e-05 eta: 0:17:40 time: 0.7740 data_time: 0.0010 memory: 6319 loss: 0.1990 +2023/06/06 09:32:48 - mmengine - INFO - Epoch(train) [10][2800/4092] lr: 1.0222e-05 eta: 0:16:24 time: 0.7677 data_time: 0.0008 memory: 6319 loss: 0.2072 +2023/06/06 09:34:02 - mmengine - INFO - Epoch(train) [10][2900/4092] lr: 1.0189e-05 eta: 0:15:07 time: 0.7200 data_time: 0.0013 memory: 6319 loss: 0.2193 +2023/06/06 09:35:17 - mmengine - INFO - Epoch(train) [10][3000/4092] lr: 1.0158e-05 eta: 0:13:51 time: 0.7599 data_time: 0.0011 memory: 6319 loss: 0.2092 +2023/06/06 09:36:33 - mmengine - INFO - Epoch(train) [10][3100/4092] lr: 1.0131e-05 eta: 0:12:35 time: 0.7480 data_time: 0.0011 memory: 6319 loss: 0.2111 +2023/06/06 09:37:30 - mmengine - INFO - Exp name: resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1_20230606_005813 +2023/06/06 09:37:48 - mmengine - INFO - Epoch(train) [10][3200/4092] lr: 1.0106e-05 eta: 0:11:19 time: 0.7729 data_time: 0.0012 memory: 6319 loss: 0.2009 +2023/06/06 09:39:03 - mmengine - INFO - Epoch(train) [10][3300/4092] lr: 1.0083e-05 eta: 0:10:03 time: 0.7603 data_time: 0.0010 memory: 6319 loss: 0.1973 +2023/06/06 09:40:20 - mmengine - INFO - Epoch(train) [10][3400/4092] lr: 1.0064e-05 eta: 0:08:47 time: 0.7466 data_time: 0.0010 memory: 6319 loss: 0.2091 +2023/06/06 09:41:36 - mmengine - INFO - Epoch(train) [10][3500/4092] lr: 1.0047e-05 eta: 0:07:30 time: 0.8362 data_time: 0.0009 memory: 6319 loss: 0.2217 +2023/06/06 09:42:52 - mmengine - INFO - Epoch(train) [10][3600/4092] lr: 1.0032e-05 eta: 0:06:14 time: 0.7650 data_time: 0.0010 memory: 6319 loss: 0.2146 +2023/06/06 09:44:07 - mmengine - INFO - Epoch(train) [10][3700/4092] lr: 1.0020e-05 eta: 0:04:58 time: 0.7336 data_time: 0.0018 memory: 6319 loss: 0.2005 +2023/06/06 09:45:24 - mmengine - INFO - Epoch(train) [10][3800/4092] lr: 1.0011e-05 eta: 0:03:42 time: 0.7361 data_time: 0.0012 memory: 6319 loss: 0.1984 +2023/06/06 09:46:39 - mmengine - INFO - Epoch(train) [10][3900/4092] lr: 1.0005e-05 eta: 0:02:26 time: 0.7410 data_time: 0.0008 memory: 6319 loss: 0.2112 +2023/06/06 09:48:02 - mmengine - INFO - Epoch(train) [10][4000/4092] lr: 1.0001e-05 eta: 0:01:10 time: 0.7572 data_time: 0.0011 memory: 6319 loss: 0.2112 +2023/06/06 09:49:10 - mmengine - INFO - Exp name: resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1_20230606_005813 +2023/06/06 09:49:10 - mmengine - INFO - Saving checkpoint at 10 epochs +2023/06/06 09:49:56 - mmengine - INFO - Epoch(val) [10][100/119] eta: 0:00:07 time: 0.7298 data_time: 0.6389 memory: 6319 +2023/06/06 09:50:23 - mmengine - INFO - Epoch(val) [10][119/119] accuracy/top1: 91.9004 data_time: 0.3828 time: 0.4717 diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/20230606_005813.json b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/20230606_005813.json new file mode 100644 index 0000000000000000000000000000000000000000..6529b6690b2940fc50c96e3d0898101d50592287 --- /dev/null +++ b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/20230606_005813.json @@ -0,0 +1,410 @@ +{"lr": 9.999870019168206e-05, "data_time": 0.056679391860961915, "loss": 0.6467106103897095, "time": 0.7275593042373657, "epoch": 1, "memory": 9436, "step": 100} +{"lr": 9.99947481934738e-05, "data_time": 0.000693058967590332, "loss": 0.6183649361133575, "time": 0.7685099601745605, "epoch": 1, "memory": 6319, "step": 200} +{"lr": 9.99881441056942e-05, "data_time": 0.000945425033569336, "loss": 0.5818252265453339, "time": 0.7908675193786621, "epoch": 1, "memory": 6319, "step": 300} +{"lr": 9.997888831760252e-05, "data_time": 0.0008037805557250977, "loss": 0.5466198921203613, "time": 0.8349925994873046, "epoch": 1, "memory": 6319, "step": 400} +{"lr": 9.996698137475549e-05, "data_time": 0.0007397413253784179, "loss": 0.5174447357654571, "time": 0.8036102294921875, "epoch": 1, "memory": 6319, "step": 500} +{"lr": 9.99524239789746e-05, "data_time": 0.0009145259857177735, "loss": 0.5112586677074432, "time": 0.7819676399230957, "epoch": 1, "memory": 6319, "step": 600} +{"lr": 9.993521698830466e-05, "data_time": 0.0008785486221313477, "loss": 0.47699070870876314, "time": 0.7547931909561157, "epoch": 1, "memory": 6319, "step": 700} +{"lr": 9.991536141696371e-05, "data_time": 0.0011017799377441406, "loss": 0.4670168846845627, "time": 0.7601818799972534, "epoch": 1, "memory": 6319, "step": 800} +{"lr": 9.989285843528274e-05, "data_time": 0.0009255647659301758, "loss": 0.451236367225647, "time": 0.7634871482849122, "epoch": 1, "memory": 6319, "step": 900} +{"lr": 9.986770936963713e-05, "data_time": 0.0009256362915039063, "loss": 0.4457569599151611, "time": 0.74671790599823, "epoch": 1, "memory": 6319, "step": 1000} +{"lr": 9.983991570236818e-05, "data_time": 0.001474618911743164, "loss": 0.4359045296907425, "time": 0.745058250427246, "epoch": 1, "memory": 6319, "step": 1100} +{"lr": 9.980947907169598e-05, "data_time": 0.000811624526977539, "loss": 0.4352470725774765, "time": 0.8330950736999512, "epoch": 1, "memory": 6319, "step": 1200} +{"lr": 9.977640127162247e-05, "data_time": 0.0009439945220947266, "loss": 0.41018485128879545, "time": 0.7362479925155639, "epoch": 1, "memory": 6319, "step": 1300} +{"lr": 9.974068425182603e-05, "data_time": 0.0008623600006103516, "loss": 0.4059330612421036, "time": 0.7337836503982544, "epoch": 1, "memory": 6319, "step": 1400} +{"lr": 9.970233011754657e-05, "data_time": 0.0008291244506835937, "loss": 0.3948523014783859, "time": 0.7408311128616333, "epoch": 1, "memory": 6319, "step": 1500} +{"lr": 9.966134112946122e-05, "data_time": 0.0008630990982055664, "loss": 0.39278132617473605, "time": 0.759043526649475, "epoch": 1, "memory": 6319, "step": 1600} +{"lr": 9.961771970355126e-05, "data_time": 0.0009460926055908203, "loss": 0.383815535902977, "time": 0.66086847782135, "epoch": 1, "memory": 6319, "step": 1700} +{"lr": 9.957146841095961e-05, "data_time": 0.0007583856582641602, "loss": 0.3866573393344879, "time": 0.7454272747039795, "epoch": 1, "memory": 6319, "step": 1800} +{"lr": 9.952258997783933e-05, "data_time": 0.0007628202438354492, "loss": 0.3863100469112396, "time": 0.7970134735107421, "epoch": 1, "memory": 6319, "step": 1900} +{"lr": 9.947108728519288e-05, "data_time": 0.0008378028869628906, "loss": 0.36880527436733246, "time": 0.7101362943649292, "epoch": 1, "memory": 6319, "step": 2000} +{"lr": 9.941696336870245e-05, "data_time": 0.0008944988250732422, "loss": 0.3723113119602203, "time": 0.7135246753692627, "epoch": 1, "memory": 6319, "step": 2100} +{"lr": 9.936022141855094e-05, "data_time": 0.0008395910263061523, "loss": 0.374591463804245, "time": 0.8074831962585449, "epoch": 1, "memory": 6319, "step": 2200} +{"lr": 9.93008647792338e-05, "data_time": 0.000999617576599121, "loss": 0.38273478150367735, "time": 0.729729413986206, "epoch": 1, "memory": 6319, "step": 2300} +{"lr": 9.923889694936212e-05, "data_time": 0.0007698535919189453, "loss": 0.3737144827842712, "time": 0.7085587024688721, "epoch": 1, "memory": 6319, "step": 2400} +{"lr": 9.917432158145639e-05, "data_time": 0.0008183479309082031, "loss": 0.3717155665159225, "time": 0.6826485633850098, "epoch": 1, "memory": 6319, "step": 2500} +{"lr": 9.91071424817309e-05, "data_time": 0.0008365631103515625, "loss": 0.3504417181015015, "time": 0.7524774312973023, "epoch": 1, "memory": 6319, "step": 2600} +{"lr": 9.903736360986979e-05, "data_time": 0.0008868932723999024, "loss": 0.34623405933380125, "time": 0.74358229637146, "epoch": 1, "memory": 6319, "step": 2700} +{"lr": 9.89649890787935e-05, "data_time": 0.001009058952331543, "loss": 0.3519168525934219, "time": 0.6897189617156982, "epoch": 1, "memory": 6319, "step": 2800} +{"lr": 9.889002315441642e-05, "data_time": 0.0010261774063110352, "loss": 0.34067114591598513, "time": 0.7585605144500732, "epoch": 1, "memory": 6319, "step": 2900} +{"lr": 9.881247025539515e-05, "data_time": 0.0008624076843261718, "loss": 0.3508040338754654, "time": 0.7910493373870849, "epoch": 1, "memory": 6319, "step": 3000} +{"lr": 9.873233495286844e-05, "data_time": 0.001009964942932129, "loss": 0.3420281380414963, "time": 0.7930436372756958, "epoch": 1, "memory": 6319, "step": 3100} +{"lr": 9.864962197018756e-05, "data_time": 0.001180124282836914, "loss": 0.34925495088100433, "time": 0.6975256919860839, "epoch": 1, "memory": 6319, "step": 3200} +{"lr": 9.856433618263779e-05, "data_time": 0.0009097814559936524, "loss": 0.3305091917514801, "time": 0.7505454301834107, "epoch": 1, "memory": 6319, "step": 3300} +{"lr": 9.847648261715132e-05, "data_time": 0.0008378505706787109, "loss": 0.3468285262584686, "time": 0.741396713256836, "epoch": 1, "memory": 6319, "step": 3400} +{"lr": 9.838606645201093e-05, "data_time": 0.0009875297546386719, "loss": 0.3498389571905136, "time": 0.7579817295074462, "epoch": 1, "memory": 6319, "step": 3500} +{"lr": 9.829309301654456e-05, "data_time": 0.0008234739303588867, "loss": 0.33495093882083893, "time": 0.7445247888565063, "epoch": 1, "memory": 6319, "step": 3600} +{"lr": 9.819756779081127e-05, "data_time": 0.0008818864822387696, "loss": 0.32544230222702025, "time": 0.7830398797988891, "epoch": 1, "memory": 6319, "step": 3700} +{"lr": 9.809949640527814e-05, "data_time": 0.0010114431381225586, "loss": 0.33996055722236634, "time": 0.7672888517379761, "epoch": 1, "memory": 6319, "step": 3800} +{"lr": 9.799888464048879e-05, "data_time": 0.0008689165115356445, "loss": 0.32864971160888673, "time": 0.769545316696167, "epoch": 1, "memory": 6319, "step": 3900} +{"lr": 9.789573842672242e-05, "data_time": 0.0008574485778808594, "loss": 0.3149187624454498, "time": 0.7635641098022461, "epoch": 1, "memory": 6319, "step": 4000} +{"accuracy/top1": 82.76980590820312, "data_time": 0.3998277928648877, "time": 0.4892594173174946, "step": 1} +{"lr": 9.7690615523361e-05, "data_time": 0.45548300743103026, "loss": 0.30172773599624636, "time": 0.740324330329895, "epoch": 2, "memory": 6319, "step": 4192} +{"lr": 9.758010405927656e-05, "data_time": 0.27181134223937986, "loss": 0.3268072783946991, "time": 0.7494885444641113, "epoch": 2, "memory": 6319, "step": 4292} +{"lr": 9.746708283006492e-05, "data_time": 0.11563518047332763, "loss": 0.3033058255910873, "time": 0.7373050928115845, "epoch": 2, "memory": 6319, "step": 4392} +{"lr": 9.735155849744634e-05, "data_time": 0.23847999572753906, "loss": 0.32278717756271363, "time": 0.8111628293991089, "epoch": 2, "memory": 6319, "step": 4492} +{"lr": 9.723353787067928e-05, "data_time": 0.017300224304199217, "loss": 0.3331650346517563, "time": 0.7327423095703125, "epoch": 2, "memory": 6319, "step": 4592} +{"lr": 9.711302790615951e-05, "data_time": 0.001006484031677246, "loss": 0.30172336399555205, "time": 0.7489974498748779, "epoch": 2, "memory": 6319, "step": 4692} +{"lr": 9.699003570700986e-05, "data_time": 0.0008834362030029297, "loss": 0.30616395473480223, "time": 0.7078217029571533, "epoch": 2, "memory": 6319, "step": 4792} +{"lr": 9.686456852266142e-05, "data_time": 0.0010566473007202148, "loss": 0.3306979715824127, "time": 0.7499691247940063, "epoch": 2, "memory": 6319, "step": 4892} +{"lr": 9.673663374842644e-05, "data_time": 0.0009475231170654296, "loss": 0.3279380649328232, "time": 0.7641098976135254, "epoch": 2, "memory": 6319, "step": 4992} +{"lr": 9.660623892506218e-05, "data_time": 0.0008621692657470703, "loss": 0.30873491168022155, "time": 0.743563199043274, "epoch": 2, "memory": 6319, "step": 5092} +{"lr": 9.647339173832693e-05, "data_time": 0.0010243654251098633, "loss": 0.2860911160707474, "time": 0.8009820461273194, "epoch": 2, "memory": 6319, "step": 5192} +{"lr": 9.633810001852653e-05, "data_time": 0.0009762525558471679, "loss": 0.32079139053821565, "time": 0.7218252658843994, "epoch": 2, "memory": 6319, "step": 5292} +{"lr": 9.620037174005274e-05, "data_time": 0.0009346485137939453, "loss": 0.3037598103284836, "time": 0.7187831401824951, "epoch": 2, "memory": 6319, "step": 5392} +{"lr": 9.606021502091389e-05, "data_time": 0.0008486747741699219, "loss": 0.2915645271539688, "time": 0.6923610687255859, "epoch": 2, "memory": 6319, "step": 5492} +{"lr": 9.591763812225569e-05, "data_time": 0.0008571147918701172, "loss": 0.313795217871666, "time": 0.6917939424514771, "epoch": 2, "memory": 6319, "step": 5592} +{"lr": 9.577264944787459e-05, "data_time": 0.0009221792221069336, "loss": 0.2936663508415222, "time": 0.7407344818115235, "epoch": 2, "memory": 6319, "step": 5692} +{"lr": 9.562525754372252e-05, "data_time": 0.0009955644607543945, "loss": 0.30850992500782015, "time": 0.8105541229248047, "epoch": 2, "memory": 6319, "step": 5792} +{"lr": 9.547547109740282e-05, "data_time": 0.0009494543075561524, "loss": 0.3004725694656372, "time": 0.7119806051254273, "epoch": 2, "memory": 6319, "step": 5892} +{"lr": 9.532329893765858e-05, "data_time": 0.0009148836135864258, "loss": 0.30849842727184296, "time": 0.7046910762786865, "epoch": 2, "memory": 6319, "step": 5992} +{"lr": 9.516875003385223e-05, "data_time": 0.0009780168533325196, "loss": 0.2957474052906036, "time": 0.7011518239974975, "epoch": 2, "memory": 6319, "step": 6092} +{"lr": 9.501183349543653e-05, "data_time": 0.0007869243621826172, "loss": 0.29207600057125094, "time": 0.7678962469100952, "epoch": 2, "memory": 6319, "step": 6192} +{"lr": 9.48525585714179e-05, "data_time": 0.0009556293487548828, "loss": 0.2962953090667725, "time": 0.7099686384201049, "epoch": 2, "memory": 6319, "step": 6292} +{"lr": 9.469093464981119e-05, "data_time": 0.0010651111602783202, "loss": 0.29158331602811816, "time": 0.7452439308166504, "epoch": 2, "memory": 6319, "step": 6392} +{"lr": 9.452697125708636e-05, "data_time": 0.0008752107620239258, "loss": 0.2812229782342911, "time": 0.7753681421279908, "epoch": 2, "memory": 6319, "step": 6492} +{"lr": 9.436067805760705e-05, "data_time": 0.0009801149368286132, "loss": 0.300742506980896, "time": 0.868291974067688, "epoch": 2, "memory": 6319, "step": 6592} +{"lr": 9.419206485306072e-05, "data_time": 0.0008069753646850586, "loss": 0.2915066510438919, "time": 0.7001688003540039, "epoch": 2, "memory": 6319, "step": 6692} +{"lr": 9.4021141581881e-05, "data_time": 0.0008558273315429687, "loss": 0.31252866685390474, "time": 0.7760474205017089, "epoch": 2, "memory": 6319, "step": 6792} +{"lr": 9.384791831866193e-05, "data_time": 0.0009261846542358398, "loss": 0.27259269952774046, "time": 0.7401493549346924, "epoch": 2, "memory": 6319, "step": 6892} +{"lr": 9.367240527356428e-05, "data_time": 0.0009130477905273437, "loss": 0.30035998821258547, "time": 0.7291918277740479, "epoch": 2, "memory": 6319, "step": 6992} +{"lr": 9.349461279171362e-05, "data_time": 0.0010501861572265625, "loss": 0.277800253033638, "time": 0.7776790142059327, "epoch": 2, "memory": 6319, "step": 7092} +{"lr": 9.33145513525904e-05, "data_time": 0.0010162353515625, "loss": 0.3006815016269684, "time": 0.7420114040374756, "epoch": 2, "memory": 6319, "step": 7192} +{"lr": 9.313223156941242e-05, "data_time": 0.0007892608642578125, "loss": 0.2931386008858681, "time": 0.7074287891387939, "epoch": 2, "memory": 6319, "step": 7292} +{"lr": 9.294766418850942e-05, "data_time": 0.0012914657592773438, "loss": 0.27622135281562804, "time": 0.8082712650299072, "epoch": 2, "memory": 6319, "step": 7392} +{"lr": 9.276086008868929e-05, "data_time": 0.0007749557495117188, "loss": 0.2855895534157753, "time": 0.8478997468948364, "epoch": 2, "memory": 6319, "step": 7492} +{"lr": 9.257183028059726e-05, "data_time": 0.0009172439575195312, "loss": 0.28419963717460633, "time": 0.6746926069259643, "epoch": 2, "memory": 6319, "step": 7592} +{"lr": 9.238058590606654e-05, "data_time": 0.000856328010559082, "loss": 0.2811689838767052, "time": 0.7300053596496582, "epoch": 2, "memory": 6319, "step": 7692} +{"lr": 9.218713823746189e-05, "data_time": 0.0009179115295410156, "loss": 0.2922960877418518, "time": 0.7394140243530274, "epoch": 2, "memory": 6319, "step": 7792} +{"lr": 9.199149867701492e-05, "data_time": 0.0010849475860595704, "loss": 0.3005239307880402, "time": 0.722748064994812, "epoch": 2, "memory": 6319, "step": 7892} +{"lr": 9.179367875615232e-05, "data_time": 0.0009314298629760742, "loss": 0.27577583491802216, "time": 0.7644084453582763, "epoch": 2, "memory": 6319, "step": 7992} +{"lr": 9.159369013481585e-05, "data_time": 0.000997447967529297, "loss": 0.2708585262298584, "time": 0.7618850231170654, "epoch": 2, "memory": 6319, "step": 8092} +{"accuracy/top1": 81.92244720458984, "data_time": 0.3712924559911092, "time": 0.4612708806991577, "step": 2} +{"lr": 9.120367596460475e-05, "data_time": 0.36234614849090574, "loss": 0.2805113524198532, "time": 0.7783551454544068, "epoch": 3, "memory": 6319, "step": 8284} +{"lr": 9.099742266672751e-05, "data_time": 0.24762063026428222, "loss": 0.2812726765871048, "time": 0.7314421653747558, "epoch": 3, "memory": 6319, "step": 8384} +{"lr": 9.078904760146445e-05, "data_time": 0.011304831504821778, "loss": 0.2848355293273926, "time": 0.7501899003982544, "epoch": 3, "memory": 6319, "step": 8484} +{"lr": 9.057856305090082e-05, "data_time": 0.0009741306304931641, "loss": 0.28337272703647615, "time": 0.7842912435531616, "epoch": 3, "memory": 6319, "step": 8584} +{"lr": 9.036598142145979e-05, "data_time": 0.0008115053176879882, "loss": 0.2887425020337105, "time": 0.7473387718200684, "epoch": 3, "memory": 6319, "step": 8684} +{"lr": 9.015131524317079e-05, "data_time": 0.0007852315902709961, "loss": 0.28306626081466674, "time": 0.7298887252807618, "epoch": 3, "memory": 6319, "step": 8784} +{"lr": 8.993457716893124e-05, "data_time": 0.0009052038192749023, "loss": 0.2758233904838562, "time": 0.7469028949737548, "epoch": 3, "memory": 6319, "step": 8884} +{"lr": 8.971577997376067e-05, "data_time": 0.0007829427719116211, "loss": 0.26109735667705536, "time": 0.7335215091705323, "epoch": 3, "memory": 6319, "step": 8984} +{"lr": 8.949493655404777e-05, "data_time": 0.0007196664810180664, "loss": 0.28642739951610563, "time": 0.7904765367507934, "epoch": 3, "memory": 6319, "step": 9084} +{"lr": 8.927205992679013e-05, "data_time": 0.0008393764495849609, "loss": 0.28590462505817416, "time": 0.8574235677719116, "epoch": 3, "memory": 6319, "step": 9184} +{"lr": 8.9047163228827e-05, "data_time": 0.0010226964950561523, "loss": 0.2825495645403862, "time": 0.7643748760223389, "epoch": 3, "memory": 6319, "step": 9284} +{"lr": 8.88202597160652e-05, "data_time": 0.0009260177612304688, "loss": 0.28266668170690534, "time": 0.7886485815048218, "epoch": 3, "memory": 6319, "step": 9384} +{"lr": 8.859136276269775e-05, "data_time": 0.000884103775024414, "loss": 0.27662210166454315, "time": 0.7629488229751586, "epoch": 3, "memory": 6319, "step": 9484} +{"lr": 8.836048586041507e-05, "data_time": 0.0008762359619140625, "loss": 0.28159307688474655, "time": 0.7175558567047119, "epoch": 3, "memory": 6319, "step": 9584} +{"lr": 8.812764261761037e-05, "data_time": 0.0009073734283447266, "loss": 0.26138431280851365, "time": 0.8578940629959106, "epoch": 3, "memory": 6319, "step": 9684} +{"lr": 8.789284675857743e-05, "data_time": 0.0007905006408691406, "loss": 0.2703880339860916, "time": 0.7499826908111572, "epoch": 3, "memory": 6319, "step": 9784} +{"lr": 8.76561121227013e-05, "data_time": 0.0007951259613037109, "loss": 0.25421659350395204, "time": 0.6923384189605712, "epoch": 3, "memory": 6319, "step": 9884} +{"lr": 8.741745266364282e-05, "data_time": 0.0007498264312744141, "loss": 0.2759989321231842, "time": 0.7331349611282348, "epoch": 3, "memory": 6319, "step": 9984} +{"lr": 8.717688244851635e-05, "data_time": 0.0008265495300292969, "loss": 0.2653114333748817, "time": 0.7779464721679688, "epoch": 3, "memory": 6319, "step": 10084} +{"lr": 8.693441565706024e-05, "data_time": 0.0010085105895996094, "loss": 0.27250864207744596, "time": 0.6818900108337402, "epoch": 3, "memory": 6319, "step": 10184} +{"lr": 8.669006658080132e-05, "data_time": 0.0010039567947387694, "loss": 0.2683863386511803, "time": 0.7410855054855346, "epoch": 3, "memory": 6319, "step": 10284} +{"lr": 8.644384962221226e-05, "data_time": 0.000935816764831543, "loss": 0.26774340569972993, "time": 0.7439370393753052, "epoch": 3, "memory": 6319, "step": 10384} +{"lr": 8.619577929386303e-05, "data_time": 0.000908660888671875, "loss": 0.27209125012159346, "time": 0.7391719818115234, "epoch": 3, "memory": 6319, "step": 10484} +{"lr": 8.594587021756517e-05, "data_time": 0.0008325338363647461, "loss": 0.2583186015486717, "time": 0.7654966831207275, "epoch": 3, "memory": 6319, "step": 10584} +{"lr": 8.569413712351023e-05, "data_time": 0.0009442806243896484, "loss": 0.26382465809583666, "time": 0.7567195892333984, "epoch": 3, "memory": 6319, "step": 10684} +{"lr": 8.544059484940119e-05, "data_time": 0.0008954048156738281, "loss": 0.28303846418857576, "time": 0.7387443542480469, "epoch": 3, "memory": 6319, "step": 10784} +{"lr": 8.518525833957818e-05, "data_time": 0.0007791996002197265, "loss": 0.2826122909784317, "time": 0.7503203868865966, "epoch": 3, "memory": 6319, "step": 10884} +{"lr": 8.49281426441377e-05, "data_time": 0.0009593486785888672, "loss": 0.24396598637104033, "time": 0.7173983812332153, "epoch": 3, "memory": 6319, "step": 10984} +{"lr": 8.466926291804501e-05, "data_time": 0.0010905742645263671, "loss": 0.26798778772354126, "time": 0.6998972415924072, "epoch": 3, "memory": 6319, "step": 11084} +{"lr": 8.440863442024147e-05, "data_time": 0.0008717060089111328, "loss": 0.275073966383934, "time": 0.7636283874511719, "epoch": 3, "memory": 6319, "step": 11184} +{"lr": 8.414627251274488e-05, "data_time": 0.0009631156921386719, "loss": 0.257562592625618, "time": 0.7458836793899536, "epoch": 3, "memory": 6319, "step": 11284} +{"lr": 8.388219265974407e-05, "data_time": 0.0007251024246215821, "loss": 0.2575678452849388, "time": 0.7709132432937622, "epoch": 3, "memory": 6319, "step": 11384} +{"lr": 8.361641042668715e-05, "data_time": 0.0007683038711547852, "loss": 0.2427933245897293, "time": 0.7077937364578247, "epoch": 3, "memory": 6319, "step": 11484} +{"lr": 8.33489414793644e-05, "data_time": 0.0009012937545776368, "loss": 0.2670065313577652, "time": 0.7723667621612549, "epoch": 3, "memory": 6319, "step": 11584} +{"lr": 8.30798015829848e-05, "data_time": 0.0009708881378173828, "loss": 0.26711356043815615, "time": 0.7584527015686036, "epoch": 3, "memory": 6319, "step": 11684} +{"lr": 8.280900660124665e-05, "data_time": 0.0008171796798706055, "loss": 0.25504983514547347, "time": 0.7334778308868408, "epoch": 3, "memory": 6319, "step": 11784} +{"lr": 8.253657249540274e-05, "data_time": 0.0009721040725708008, "loss": 0.2652930110692978, "time": 0.7529042959213257, "epoch": 3, "memory": 6319, "step": 11884} +{"lr": 8.226251532331926e-05, "data_time": 0.001077413558959961, "loss": 0.25756279528141024, "time": 0.7682068109512329, "epoch": 3, "memory": 6319, "step": 11984} +{"lr": 8.198685123852965e-05, "data_time": 0.0009561300277709961, "loss": 0.26171095967292785, "time": 0.7986436128616333, "epoch": 3, "memory": 6319, "step": 12084} +{"lr": 8.17095964892824e-05, "data_time": 0.00106048583984375, "loss": 0.25484343320131303, "time": 0.704461932182312, "epoch": 3, "memory": 6319, "step": 12184} +{"accuracy/top1": 81.7387466430664, "data_time": 0.3710541745026906, "time": 0.46016467014948526, "step": 3} +{"lr": 8.11728683565904e-05, "data_time": 0.44006059169769285, "loss": 0.25812426656484605, "time": 0.7065037012100219, "epoch": 4, "memory": 6319, "step": 12376} +{"lr": 8.089106273476776e-05, "data_time": 0.07465133666992188, "loss": 0.24335235059261323, "time": 0.7446303129196167, "epoch": 4, "memory": 6319, "step": 12476} +{"lr": 8.060773103667566e-05, "data_time": 0.0009151697158813477, "loss": 0.24658887535333635, "time": 0.745331597328186, "epoch": 4, "memory": 6319, "step": 12576} +{"lr": 8.03228899625083e-05, "data_time": 0.0010189294815063476, "loss": 0.24688779711723327, "time": 0.7623863458633423, "epoch": 4, "memory": 6319, "step": 12676} +{"lr": 8.003655630142627e-05, "data_time": 0.0008771657943725586, "loss": 0.2450454831123352, "time": 0.7806264162063599, "epoch": 4, "memory": 6319, "step": 12776} +{"lr": 7.974874693056617e-05, "data_time": 0.0009460210800170899, "loss": 0.2516448900103569, "time": 0.7400216102600098, "epoch": 4, "memory": 6319, "step": 12876} +{"lr": 7.945947881404618e-05, "data_time": 0.0009044170379638672, "loss": 0.25994026511907575, "time": 0.7677648067474365, "epoch": 4, "memory": 6319, "step": 12976} +{"lr": 7.916876900196646e-05, "data_time": 0.0009540081024169922, "loss": 0.2557961091399193, "time": 0.7357230663299561, "epoch": 4, "memory": 6319, "step": 13076} +{"lr": 7.887663462940359e-05, "data_time": 0.0009281635284423828, "loss": 0.24798357039690017, "time": 0.7497295379638672, "epoch": 4, "memory": 6319, "step": 13176} +{"lr": 7.858309291540111e-05, "data_time": 0.0010016679763793946, "loss": 0.258859620988369, "time": 0.7436985015869141, "epoch": 4, "memory": 6319, "step": 13276} +{"lr": 7.828816116195411e-05, "data_time": 0.0008557796478271485, "loss": 0.2308677166700363, "time": 0.7458075761795044, "epoch": 4, "memory": 6319, "step": 13376} +{"lr": 7.799185675298994e-05, "data_time": 0.0010064363479614258, "loss": 0.24006345719099045, "time": 0.7534310102462769, "epoch": 4, "memory": 6319, "step": 13476} +{"lr": 7.769419715334297e-05, "data_time": 0.0010347843170166015, "loss": 0.24197655767202378, "time": 0.7990076303482055, "epoch": 4, "memory": 6319, "step": 13576} +{"lr": 7.739519990772583e-05, "data_time": 0.0012361526489257813, "loss": 0.24522503465414047, "time": 0.7416842699050903, "epoch": 4, "memory": 6319, "step": 13676} +{"lr": 7.709488263969475e-05, "data_time": 0.0010061025619506835, "loss": 0.2393003895878792, "time": 0.7449990272521972, "epoch": 4, "memory": 6319, "step": 13776} +{"lr": 7.67932630506109e-05, "data_time": 0.0008320093154907226, "loss": 0.2356681138277054, "time": 0.7493997812271118, "epoch": 4, "memory": 6319, "step": 13876} +{"lr": 7.64903589185975e-05, "data_time": 0.0008714199066162109, "loss": 0.24267417639493943, "time": 0.7086737155914307, "epoch": 4, "memory": 6319, "step": 13976} +{"lr": 7.618618809749134e-05, "data_time": 0.0008412599563598633, "loss": 0.23524534702301025, "time": 0.7754576921463012, "epoch": 4, "memory": 6319, "step": 14076} +{"lr": 7.588076851579028e-05, "data_time": 0.0008939027786254883, "loss": 0.25485320687294005, "time": 0.7532485961914063, "epoch": 4, "memory": 6319, "step": 14176} +{"lr": 7.557411817559755e-05, "data_time": 0.0009070396423339844, "loss": 0.2572333440184593, "time": 0.764794921875, "epoch": 4, "memory": 6319, "step": 14276} +{"lr": 7.526625515155957e-05, "data_time": 0.0009389400482177734, "loss": 0.23196712732315064, "time": 0.6891292572021485, "epoch": 4, "memory": 6319, "step": 14376} +{"lr": 7.495719758980122e-05, "data_time": 0.0008939743041992187, "loss": 0.26005916446447375, "time": 0.7295319318771363, "epoch": 4, "memory": 6319, "step": 14476} +{"lr": 7.464696370685586e-05, "data_time": 0.0008395671844482422, "loss": 0.24326847940683366, "time": 0.8304219484329224, "epoch": 4, "memory": 6319, "step": 14576} +{"lr": 7.433557178859198e-05, "data_time": 0.0009120941162109375, "loss": 0.23468608111143113, "time": 0.7421232461929321, "epoch": 4, "memory": 6319, "step": 14676} +{"lr": 7.402304018913512e-05, "data_time": 0.001082634925842285, "loss": 0.24131477624177933, "time": 0.7425987243652343, "epoch": 4, "memory": 6319, "step": 14776} +{"lr": 7.37093873297861e-05, "data_time": 0.0009795427322387695, "loss": 0.2344892367720604, "time": 0.7353773593902588, "epoch": 4, "memory": 6319, "step": 14876} +{"lr": 7.339463169793563e-05, "data_time": 0.0009601593017578125, "loss": 0.2357519179582596, "time": 0.8386085748672485, "epoch": 4, "memory": 6319, "step": 14976} +{"lr": 7.307879184597369e-05, "data_time": 0.0008764982223510743, "loss": 0.25934259295463563, "time": 1.4188462972640992, "epoch": 4, "memory": 6319, "step": 15076} +{"lr": 7.276188639019704e-05, "data_time": 0.0009303569793701171, "loss": 0.23962348699569702, "time": 0.7592996120452881, "epoch": 4, "memory": 6319, "step": 15176} +{"lr": 7.244393400971127e-05, "data_time": 0.0008882284164428711, "loss": 0.23132585138082504, "time": 0.7509804487228393, "epoch": 4, "memory": 6319, "step": 15276} +{"lr": 7.212495344533006e-05, "data_time": 0.0007544994354248047, "loss": 0.2398171529173851, "time": 0.7699744462966919, "epoch": 4, "memory": 6319, "step": 15376} +{"lr": 7.180496349847055e-05, "data_time": 0.0010762214660644531, "loss": 0.22637886852025985, "time": 0.8533263444900513, "epoch": 4, "memory": 6319, "step": 15476} +{"lr": 7.148398303004499e-05, "data_time": 0.0010487556457519532, "loss": 0.25670819282531737, "time": 0.7617285251617432, "epoch": 4, "memory": 6319, "step": 15576} +{"lr": 7.11620309593493e-05, "data_time": 0.0008179903030395508, "loss": 0.22662961184978486, "time": 0.8312148809432983, "epoch": 4, "memory": 6319, "step": 15676} +{"lr": 7.083912626294795e-05, "data_time": 0.0010239362716674804, "loss": 0.23864293545484544, "time": 0.7462512969970703, "epoch": 4, "memory": 6319, "step": 15776} +{"lr": 7.051528797355494e-05, "data_time": 0.0007915973663330078, "loss": 0.24410039633512498, "time": 0.7130609750747681, "epoch": 4, "memory": 6319, "step": 15876} +{"lr": 7.019053517891214e-05, "data_time": 0.0008965730667114258, "loss": 0.2492641806602478, "time": 0.7135108947753906, "epoch": 4, "memory": 6319, "step": 15976} +{"lr": 6.986488702066473e-05, "data_time": 0.0008026123046875, "loss": 0.23730763494968415, "time": 0.7190106868743896, "epoch": 4, "memory": 6319, "step": 16076} +{"lr": 6.953836269323245e-05, "data_time": 0.0009332895278930664, "loss": 0.2515053525567055, "time": 0.8029151678085327, "epoch": 4, "memory": 6319, "step": 16176} +{"lr": 6.921098144267838e-05, "data_time": 0.0009343624114990234, "loss": 0.2330927789211273, "time": 0.7749029159545898, "epoch": 4, "memory": 6319, "step": 16276} +{"accuracy/top1": 85.10832214355469, "data_time": 0.3379850685596466, "time": 0.42568763891855876, "step": 4} +{"lr": 6.858007803673889e-05, "data_time": 0.30223779678344725, "loss": 0.2544656455516815, "time": 0.7357942581176757, "epoch": 5, "memory": 6319, "step": 16468} +{"lr": 6.825030518869046e-05, "data_time": 0.0009554862976074219, "loss": 0.23199640065431595, "time": 0.7362879276275635, "epoch": 5, "memory": 6319, "step": 16568} +{"lr": 6.791975133846686e-05, "data_time": 0.0009989261627197266, "loss": 0.23119949102401732, "time": 0.7697957277297973, "epoch": 5, "memory": 6319, "step": 16668} +{"lr": 6.758843596964013e-05, "data_time": 0.0007458925247192383, "loss": 0.239640474319458, "time": 1.3155960321426392, "epoch": 5, "memory": 6319, "step": 16768} +{"lr": 6.725637861066804e-05, "data_time": 0.0008596420288085937, "loss": 0.22820000052452089, "time": 0.7898166656494141, "epoch": 5, "memory": 6319, "step": 16868} +{"lr": 6.692359883374264e-05, "data_time": 0.0008529901504516601, "loss": 0.23233971148729324, "time": 0.8249887466430664, "epoch": 5, "memory": 6319, "step": 16968} +{"lr": 6.659011625363718e-05, "data_time": 0.0008084297180175782, "loss": 0.23109327405691146, "time": 0.70664381980896, "epoch": 5, "memory": 6319, "step": 17068} +{"lr": 6.625595052654927e-05, "data_time": 0.0008114814758300781, "loss": 0.23301598131656648, "time": 0.7499979972839356, "epoch": 5, "memory": 6319, "step": 17168} +{"lr": 6.592112134894315e-05, "data_time": 0.000906825065612793, "loss": 0.25075470060110094, "time": 0.7827856063842773, "epoch": 5, "memory": 6319, "step": 17268} +{"lr": 6.558564845638802e-05, "data_time": 0.0009220361709594727, "loss": 0.2466512367129326, "time": 0.7277509450912476, "epoch": 5, "memory": 6319, "step": 17368} +{"lr": 6.524955162239518e-05, "data_time": 0.0009633302688598633, "loss": 0.255857852101326, "time": 0.7504189014434814, "epoch": 5, "memory": 6319, "step": 17468} +{"lr": 6.491285065725238e-05, "data_time": 0.0010142326354980469, "loss": 0.22700437754392624, "time": 0.7106493949890137, "epoch": 5, "memory": 6319, "step": 17568} +{"lr": 6.457556540685618e-05, "data_time": 0.0007700920104980469, "loss": 0.24164402335882187, "time": 0.7481193304061889, "epoch": 5, "memory": 6319, "step": 17668} +{"lr": 6.423771575154208e-05, "data_time": 0.0009311914443969726, "loss": 0.23107862025499343, "time": 0.7690333604812623, "epoch": 5, "memory": 6319, "step": 17768} +{"lr": 6.389932160491309e-05, "data_time": 0.0008931636810302734, "loss": 0.24904999434947966, "time": 0.7814180135726929, "epoch": 5, "memory": 6319, "step": 17868} +{"lr": 6.356040291266556e-05, "data_time": 0.000942683219909668, "loss": 0.23286448419094086, "time": 0.7281208276748657, "epoch": 5, "memory": 6319, "step": 17968} +{"lr": 6.322097965141391e-05, "data_time": 0.000815272331237793, "loss": 0.23617454320192338, "time": 0.7271085500717163, "epoch": 5, "memory": 6319, "step": 18068} +{"lr": 6.288107182751267e-05, "data_time": 0.0009077787399291992, "loss": 0.22853530049324036, "time": 0.7365193843841553, "epoch": 5, "memory": 6319, "step": 18168} +{"lr": 6.254069947587782e-05, "data_time": 0.000838780403137207, "loss": 0.23225885778665542, "time": 0.7225019454956054, "epoch": 5, "memory": 6319, "step": 18268} +{"lr": 6.219988265880537e-05, "data_time": 0.0009633541107177734, "loss": 0.24403035044670104, "time": 0.8011118650436402, "epoch": 5, "memory": 6319, "step": 18368} +{"lr": 6.185864146478941e-05, "data_time": 0.0011283397674560548, "loss": 0.24354525804519653, "time": 0.7545641660690308, "epoch": 5, "memory": 6319, "step": 18468} +{"lr": 6.151699600733752e-05, "data_time": 0.001166391372680664, "loss": 0.2370014175772667, "time": 0.7521046161651611, "epoch": 5, "memory": 6319, "step": 18568} +{"lr": 6.117496642378554e-05, "data_time": 0.0008833169937133789, "loss": 0.22047528475522996, "time": 0.7016199350357055, "epoch": 5, "memory": 6319, "step": 18668} +{"lr": 6.0832572874110315e-05, "data_time": 0.0008026123046875, "loss": 0.22848183959722518, "time": 0.7139702796936035, "epoch": 5, "memory": 6319, "step": 18768} +{"lr": 6.0489835539741924e-05, "data_time": 0.0010669708251953125, "loss": 0.22957628965377808, "time": 0.6931902408599854, "epoch": 5, "memory": 6319, "step": 18868} +{"lr": 6.0146774622374006e-05, "data_time": 0.0009476900100708008, "loss": 0.22941379994153976, "time": 0.7554872751235961, "epoch": 5, "memory": 6319, "step": 18968} +{"lr": 5.980341034277274e-05, "data_time": 0.0010091781616210938, "loss": 0.23331795632839203, "time": 0.7668699026107788, "epoch": 5, "memory": 6319, "step": 19068} +{"lr": 5.945976293958505e-05, "data_time": 0.0009149551391601562, "loss": 0.2550873875617981, "time": 0.8021178722381592, "epoch": 5, "memory": 6319, "step": 19168} +{"lr": 5.911585266814587e-05, "data_time": 0.0009304285049438477, "loss": 0.2470840498805046, "time": 0.7086685419082641, "epoch": 5, "memory": 6319, "step": 19268} +{"lr": 5.877169979928426e-05, "data_time": 0.0010169267654418946, "loss": 0.22779435217380523, "time": 0.7370935678482056, "epoch": 5, "memory": 6319, "step": 19368} +{"lr": 5.842732461812828e-05, "data_time": 0.0010720252990722655, "loss": 0.22472428679466247, "time": 0.7397331714630127, "epoch": 5, "memory": 6319, "step": 19468} +{"lr": 5.8082747422909646e-05, "data_time": 0.0009012699127197265, "loss": 0.24025715589523317, "time": 0.7366528749465943, "epoch": 5, "memory": 6319, "step": 19568} +{"lr": 5.7737988523767406e-05, "data_time": 0.000866103172302246, "loss": 0.21452924907207488, "time": 0.7156328678131103, "epoch": 5, "memory": 6319, "step": 19668} +{"lr": 5.739306824155042e-05, "data_time": 0.00106503963470459, "loss": 0.22239121943712234, "time": 0.7647297143936157, "epoch": 5, "memory": 6319, "step": 19768} +{"lr": 5.704800690661993e-05, "data_time": 0.0010779857635498046, "loss": 0.22004520297050476, "time": 0.8111758708953858, "epoch": 5, "memory": 6319, "step": 19868} +{"lr": 5.6702824857650994e-05, "data_time": 0.0008751630783081054, "loss": 0.218601992726326, "time": 0.734070086479187, "epoch": 5, "memory": 6319, "step": 19968} +{"lr": 5.635754244043411e-05, "data_time": 0.0008763790130615235, "loss": 0.2396036222577095, "time": 0.723269009590149, "epoch": 5, "memory": 6319, "step": 20068} +{"lr": 5.60121800066754e-05, "data_time": 0.0011265993118286133, "loss": 0.24507984519004822, "time": 0.7516209125518799, "epoch": 5, "memory": 6319, "step": 20168} +{"lr": 5.5666757912797516e-05, "data_time": 0.0010689020156860352, "loss": 0.2263699859380722, "time": 0.7594061374664307, "epoch": 5, "memory": 6319, "step": 20268} +{"lr": 5.532129651873955e-05, "data_time": 0.0009542226791381836, "loss": 0.2092998206615448, "time": 0.7460134744644165, "epoch": 5, "memory": 6319, "step": 20368} +{"accuracy/top1": 88.72946166992188, "data_time": 0.37101765473683673, "time": 0.46014336347579954, "step": 5} +{"lr": 5.465797506068693e-05, "data_time": 0.3617511749267578, "loss": 0.22097585052251817, "time": 0.7422191143035889, "epoch": 6, "memory": 6319, "step": 20560} +{"lr": 5.431251545044144e-05, "data_time": 0.08920300006866455, "loss": 0.22178382724523543, "time": 0.7930021047592163, "epoch": 6, "memory": 6319, "step": 20660} +{"lr": 5.396709636204968e-05, "data_time": 0.000785374641418457, "loss": 0.22732941210269927, "time": 0.812529730796814, "epoch": 6, "memory": 6319, "step": 20760} +{"lr": 5.3621738155273566e-05, "data_time": 0.0008415937423706055, "loss": 0.2372802272439003, "time": 0.7552832365036011, "epoch": 6, "memory": 6319, "step": 20860} +{"lr": 5.3276461186286584e-05, "data_time": 0.0010387659072875976, "loss": 0.2372077465057373, "time": 0.7639619827270507, "epoch": 6, "memory": 6319, "step": 20960} +{"lr": 5.293128580647384e-05, "data_time": 0.0010754823684692382, "loss": 0.22168179750442504, "time": 0.7425286769866943, "epoch": 6, "memory": 6319, "step": 21060} +{"lr": 5.258623236123256e-05, "data_time": 0.0009563446044921875, "loss": 0.21794785112142562, "time": 0.7136249780654907, "epoch": 6, "memory": 6319, "step": 21160} +{"lr": 5.224132118877281e-05, "data_time": 0.0016475200653076171, "loss": 0.21800965070724487, "time": 0.7487157821655274, "epoch": 6, "memory": 6319, "step": 21260} +{"lr": 5.189657261891887e-05, "data_time": 0.0009242534637451172, "loss": 0.24410653710365296, "time": 0.7653291463851929, "epoch": 6, "memory": 6319, "step": 21360} +{"lr": 5.1552006971910974e-05, "data_time": 0.0007371187210083007, "loss": 0.2157185912132263, "time": 0.8699023962020874, "epoch": 6, "memory": 6319, "step": 21460} +{"lr": 5.120764455720728e-05, "data_time": 0.0012006282806396485, "loss": 0.21663552671670913, "time": 0.7318055391311645, "epoch": 6, "memory": 6319, "step": 21560} +{"lr": 5.0863505672287034e-05, "data_time": 0.0009980440139770509, "loss": 0.22659144848585128, "time": 0.8226141214370728, "epoch": 6, "memory": 6319, "step": 21660} +{"lr": 5.051961060145421e-05, "data_time": 0.0009065389633178711, "loss": 0.21539592742919922, "time": 0.7309055328369141, "epoch": 6, "memory": 6319, "step": 21760} +{"lr": 5.017597961464177e-05, "data_time": 0.0009898900985717773, "loss": 0.23304681926965715, "time": 0.772872018814087, "epoch": 6, "memory": 6319, "step": 21860} +{"lr": 4.983263296621707e-05, "data_time": 0.0008554697036743164, "loss": 0.21356911659240724, "time": 0.72887282371521, "epoch": 6, "memory": 6319, "step": 21960} +{"lr": 4.948959089378785e-05, "data_time": 0.0008954286575317382, "loss": 0.21990423947572707, "time": 0.7353631019592285, "epoch": 6, "memory": 6319, "step": 22060} +{"lr": 4.91468736170094e-05, "data_time": 0.0008244991302490234, "loss": 0.22922151684761047, "time": 0.7417398691177368, "epoch": 6, "memory": 6319, "step": 22160} +{"lr": 4.880450133639314e-05, "data_time": 0.0011096000671386719, "loss": 0.22476275265216827, "time": 0.7459015130996705, "epoch": 6, "memory": 6319, "step": 22260} +{"lr": 4.8462494232115335e-05, "data_time": 0.000895380973815918, "loss": 0.21308909356594086, "time": 0.7584392309188843, "epoch": 6, "memory": 6319, "step": 22360} +{"lr": 4.812087246282806e-05, "data_time": 0.0009870529174804688, "loss": 0.21166886687278746, "time": 0.7301443099975586, "epoch": 6, "memory": 6319, "step": 22460} +{"lr": 4.7779656164470954e-05, "data_time": 0.0013822317123413086, "loss": 0.2199772208929062, "time": 0.7756128787994385, "epoch": 6, "memory": 6319, "step": 22560} +{"lr": 4.7438865449084246e-05, "data_time": 0.0008816242218017579, "loss": 0.2076578766107559, "time": 0.768445348739624, "epoch": 6, "memory": 6319, "step": 22660} +{"lr": 4.709852040362338e-05, "data_time": 0.0009389638900756836, "loss": 0.21123078167438508, "time": 0.7135071277618408, "epoch": 6, "memory": 6319, "step": 22760} +{"lr": 4.675864108877499e-05, "data_time": 0.0008680343627929688, "loss": 0.24213262051343917, "time": 0.7669236183166503, "epoch": 6, "memory": 6319, "step": 22860} +{"lr": 4.641924753777464e-05, "data_time": 0.00105743408203125, "loss": 0.22617121487855912, "time": 0.7091005086898804, "epoch": 6, "memory": 6319, "step": 22960} +{"lr": 4.608035975522578e-05, "data_time": 0.0007989645004272461, "loss": 0.24261579364538194, "time": 0.7463184356689453, "epoch": 6, "memory": 6319, "step": 23060} +{"lr": 4.574199771592085e-05, "data_time": 0.0009153127670288086, "loss": 0.23417214751243592, "time": 0.7600003242492676, "epoch": 6, "memory": 6319, "step": 23160} +{"lr": 4.540418136366372e-05, "data_time": 0.000960850715637207, "loss": 0.24277285784482955, "time": 0.7358004093170166, "epoch": 6, "memory": 6319, "step": 23260} +{"lr": 4.506693061009437e-05, "data_time": 0.0011164188385009766, "loss": 0.20732830762863158, "time": 0.7263560771942139, "epoch": 6, "memory": 6319, "step": 23360} +{"lr": 4.473026533351511e-05, "data_time": 0.001043844223022461, "loss": 0.2153569743037224, "time": 0.7638959407806396, "epoch": 6, "memory": 6319, "step": 23460} +{"lr": 4.439420537771888e-05, "data_time": 0.0010238170623779296, "loss": 0.2066410392522812, "time": 0.7720999002456665, "epoch": 6, "memory": 6319, "step": 23560} +{"lr": 4.405877055081978e-05, "data_time": 0.0012173891067504884, "loss": 0.22008976042270662, "time": 0.7492174625396728, "epoch": 6, "memory": 6319, "step": 23660} +{"lr": 4.3723980624085373e-05, "data_time": 0.0009804725646972655, "loss": 0.2285812348127365, "time": 0.6996923446655273, "epoch": 6, "memory": 6319, "step": 23760} +{"lr": 4.3389855330771394e-05, "data_time": 0.000852513313293457, "loss": 0.22510926574468612, "time": 0.7598581790924073, "epoch": 6, "memory": 6319, "step": 23860} +{"lr": 4.305641436495869e-05, "data_time": 0.0008151769638061523, "loss": 0.21471958756446838, "time": 0.7302027702331543, "epoch": 6, "memory": 6319, "step": 23960} +{"lr": 4.272367738039222e-05, "data_time": 0.001053190231323242, "loss": 0.20233013182878495, "time": 0.7877671957015991, "epoch": 6, "memory": 6319, "step": 24060} +{"lr": 4.239166398932278e-05, "data_time": 0.0008234500885009766, "loss": 0.20474919229745864, "time": 0.7330990076065064, "epoch": 6, "memory": 6319, "step": 24160} +{"lr": 4.206039376135108e-05, "data_time": 0.0009666681289672852, "loss": 0.24494716972112657, "time": 0.7451794385910034, "epoch": 6, "memory": 6319, "step": 24260} +{"lr": 4.1729886222273985e-05, "data_time": 0.0010160684585571289, "loss": 0.21276360154151916, "time": 0.7946240425109863, "epoch": 6, "memory": 6319, "step": 24360} +{"lr": 4.140016085293389e-05, "data_time": 0.0008048057556152344, "loss": 0.21543313711881637, "time": 0.7634910106658935, "epoch": 6, "memory": 6319, "step": 24460} +{"accuracy/top1": 89.29381561279297, "data_time": 0.3655286471048991, "time": 0.45561564366022744, "step": 6} +{"lr": 4.0769351868715556e-05, "data_time": 0.06758735179901124, "loss": 0.21464929431676866, "time": 0.7436929702758789, "epoch": 7, "memory": 6319, "step": 24652} +{"lr": 4.044202148988293e-05, "data_time": 0.0009186506271362304, "loss": 0.22030574530363084, "time": 0.7825503826141358, "epoch": 7, "memory": 6319, "step": 24752} +{"lr": 4.0115549190371644e-05, "data_time": 0.0007718324661254883, "loss": 0.22843018025159836, "time": 0.8764800548553466, "epoch": 7, "memory": 6319, "step": 24852} +{"lr": 3.978995421317808e-05, "data_time": 0.0010705947875976562, "loss": 0.2284991666674614, "time": 0.7792644262313843, "epoch": 7, "memory": 6319, "step": 24952} +{"lr": 3.94652557495873e-05, "data_time": 0.0010233163833618165, "loss": 0.20828754156827928, "time": 0.6905143976211547, "epoch": 7, "memory": 6319, "step": 25052} +{"lr": 3.914147293804207e-05, "data_time": 0.0012282133102416992, "loss": 0.20730749368667603, "time": 0.7519061326980591, "epoch": 7, "memory": 6319, "step": 25152} +{"lr": 3.881862486301448e-05, "data_time": 0.0010767459869384765, "loss": 0.22793783396482467, "time": 0.8323409795761109, "epoch": 7, "memory": 6319, "step": 25252} +{"lr": 3.849673055388122e-05, "data_time": 0.0010067224502563477, "loss": 0.2198184385895729, "time": 0.7189236164093018, "epoch": 7, "memory": 6319, "step": 25352} +{"lr": 3.817580898380197e-05, "data_time": 0.0009664535522460937, "loss": 0.22881572246551513, "time": 0.7632386207580566, "epoch": 7, "memory": 6319, "step": 25452} +{"lr": 3.7855879068600975e-05, "data_time": 0.000894618034362793, "loss": 0.22118446081876755, "time": 0.7141993284225464, "epoch": 7, "memory": 6319, "step": 25552} +{"lr": 3.753695966565216e-05, "data_time": 0.0012279272079467774, "loss": 0.21850900799036027, "time": 0.726638388633728, "epoch": 7, "memory": 6319, "step": 25652} +{"lr": 3.7219069572767636e-05, "data_time": 0.0010407924652099609, "loss": 0.21882450729608535, "time": 0.7235573530197144, "epoch": 7, "memory": 6319, "step": 25752} +{"lr": 3.6902227527089725e-05, "data_time": 0.001052522659301758, "loss": 0.21903080642223358, "time": 0.7414105415344239, "epoch": 7, "memory": 6319, "step": 25852} +{"lr": 3.65864522039865e-05, "data_time": 0.0009268522262573242, "loss": 0.20320204198360442, "time": 0.7302920103073121, "epoch": 7, "memory": 6319, "step": 25952} +{"lr": 3.627176221595109e-05, "data_time": 0.0008532524108886719, "loss": 0.22782006859779358, "time": 0.7634247541427612, "epoch": 7, "memory": 6319, "step": 26052} +{"lr": 3.595817611150461e-05, "data_time": 0.0010223388671875, "loss": 0.2195037305355072, "time": 0.7388573169708252, "epoch": 7, "memory": 6319, "step": 26152} +{"lr": 3.5645712374102824e-05, "data_time": 0.0011571884155273438, "loss": 0.22454675883054734, "time": 0.7472745180130005, "epoch": 7, "memory": 6319, "step": 26252} +{"lr": 3.533438942104663e-05, "data_time": 0.0010906219482421874, "loss": 0.2088132008910179, "time": 0.6744309663772583, "epoch": 7, "memory": 6319, "step": 26352} +{"lr": 3.5024225602396704e-05, "data_time": 0.0010611772537231444, "loss": 0.21927298307418824, "time": 0.7297801017761231, "epoch": 7, "memory": 6319, "step": 26452} +{"lr": 3.4715239199891735e-05, "data_time": 0.001035928726196289, "loss": 0.22908590584993363, "time": 0.7075364589691162, "epoch": 7, "memory": 6319, "step": 26552} +{"lr": 3.44074484258709e-05, "data_time": 0.0010905265808105469, "loss": 0.21959491521120073, "time": 0.786644434928894, "epoch": 7, "memory": 6319, "step": 26652} +{"lr": 3.4100871422200426e-05, "data_time": 0.0009969472885131836, "loss": 0.2215401530265808, "time": 0.7292780160903931, "epoch": 7, "memory": 6319, "step": 26752} +{"lr": 3.379552625920419e-05, "data_time": 0.0010289192199707032, "loss": 0.21800469607114792, "time": 0.7371968507766724, "epoch": 7, "memory": 6319, "step": 26852} +{"lr": 3.349143093459868e-05, "data_time": 0.0009022235870361328, "loss": 0.2225462332367897, "time": 0.7333777666091919, "epoch": 7, "memory": 6319, "step": 26952} +{"lr": 3.318860337243226e-05, "data_time": 0.0012842893600463867, "loss": 0.22466444969177246, "time": 0.7128596305847168, "epoch": 7, "memory": 6319, "step": 27052} +{"lr": 3.288706142202845e-05, "data_time": 0.001050424575805664, "loss": 0.21334576606750488, "time": 0.7456519603729248, "epoch": 7, "memory": 6319, "step": 27152} +{"lr": 3.25868228569341e-05, "data_time": 0.0009516716003417969, "loss": 0.21006509959697722, "time": 0.7747654914855957, "epoch": 7, "memory": 6319, "step": 27252} +{"lr": 3.228790537387148e-05, "data_time": 0.0008413076400756836, "loss": 0.21703979820013047, "time": 0.7487269163131713, "epoch": 7, "memory": 6319, "step": 27352} +{"lr": 3.199032659169556e-05, "data_time": 0.000826716423034668, "loss": 0.23915895074605942, "time": 0.7505497217178345, "epoch": 7, "memory": 6319, "step": 27452} +{"lr": 3.169410405035527e-05, "data_time": 0.0008703947067260742, "loss": 0.23334365487098693, "time": 0.7179245948791504, "epoch": 7, "memory": 6319, "step": 27552} +{"lr": 3.139925520985965e-05, "data_time": 0.000834512710571289, "loss": 0.2168520823121071, "time": 0.7441962242126465, "epoch": 7, "memory": 6319, "step": 27652} +{"lr": 3.1105797449248916e-05, "data_time": 0.0009701013565063477, "loss": 0.21715566962957383, "time": 0.7484423637390136, "epoch": 7, "memory": 6319, "step": 27752} +{"lr": 3.081374806556986e-05, "data_time": 0.0008368730545043945, "loss": 0.2253271907567978, "time": 0.6734090328216553, "epoch": 7, "memory": 6319, "step": 27852} +{"lr": 3.052312427285649e-05, "data_time": 0.0008078336715698243, "loss": 0.22412145286798477, "time": 0.7776374101638794, "epoch": 7, "memory": 6319, "step": 27952} +{"lr": 3.0233943201115213e-05, "data_time": 0.000777745246887207, "loss": 0.20857783108949662, "time": 0.7468486785888672, "epoch": 7, "memory": 6319, "step": 28052} +{"lr": 2.9946221895315525e-05, "data_time": 0.0010638713836669921, "loss": 0.22127020359039307, "time": 0.7438156366348266, "epoch": 7, "memory": 6319, "step": 28152} +{"lr": 2.9659977314384953e-05, "data_time": 0.0008136749267578125, "loss": 0.2155949130654335, "time": 0.7352388143539429, "epoch": 7, "memory": 6319, "step": 28252} +{"lr": 2.93752263302096e-05, "data_time": 0.0009730815887451172, "loss": 0.20265680700540542, "time": 0.7197938203811646, "epoch": 7, "memory": 6319, "step": 28352} +{"lr": 2.9091985726639714e-05, "data_time": 0.0009970664978027344, "loss": 0.20858417302370072, "time": 0.766517949104309, "epoch": 7, "memory": 6319, "step": 28452} +{"lr": 2.881027219850035e-05, "data_time": 0.0010104656219482421, "loss": 0.22796138674020766, "time": 0.7162450551986694, "epoch": 7, "memory": 6319, "step": 28552} +{"accuracy/top1": 90.86441040039062, "data_time": 0.37313953836758934, "time": 0.46175772746404015, "step": 7} +{"lr": 2.827372366659638e-05, "data_time": 0.2035313367843628, "loss": 0.20221212059259414, "time": 0.778073787689209, "epoch": 8, "memory": 6319, "step": 28744} +{"lr": 2.7996563897301616e-05, "data_time": 0.001152348518371582, "loss": 0.2191552385687828, "time": 0.7858430624008179, "epoch": 8, "memory": 6319, "step": 28844} +{"lr": 2.7720995770013028e-05, "data_time": 0.0009917259216308595, "loss": 0.222273613512516, "time": 0.773386287689209, "epoch": 8, "memory": 6319, "step": 28944} +{"lr": 2.744703552732299e-05, "data_time": 0.0010296106338500977, "loss": 0.20417798608541488, "time": 0.7599788904190063, "epoch": 8, "memory": 6319, "step": 29044} +{"lr": 2.717469931705171e-05, "data_time": 0.0008471965789794922, "loss": 0.2055888667702675, "time": 0.7366097450256348, "epoch": 8, "memory": 6319, "step": 29144} +{"lr": 2.690400319129529e-05, "data_time": 0.0008450031280517578, "loss": 0.21146828234195708, "time": 0.7604643344879151, "epoch": 8, "memory": 6319, "step": 29244} +{"lr": 2.6634963105479716e-05, "data_time": 0.0009392976760864257, "loss": 0.21690429002046585, "time": 0.8979794979095459, "epoch": 8, "memory": 6319, "step": 29344} +{"lr": 2.6367594917420177e-05, "data_time": 0.0008240222930908203, "loss": 0.21820570677518844, "time": 0.7451939105987548, "epoch": 8, "memory": 6319, "step": 29444} +{"lr": 2.610191438638673e-05, "data_time": 0.000759577751159668, "loss": 0.2177011102437973, "time": 0.7357152700424194, "epoch": 8, "memory": 6319, "step": 29544} +{"lr": 2.5837937172175032e-05, "data_time": 0.0008301019668579101, "loss": 0.22187730818986892, "time": 0.7183035373687744, "epoch": 8, "memory": 6319, "step": 29644} +{"lr": 2.557567883418364e-05, "data_time": 0.0008044242858886719, "loss": 0.20425197333097458, "time": 0.7523620367050171, "epoch": 8, "memory": 6319, "step": 29744} +{"lr": 2.531515483049656e-05, "data_time": 0.000857853889465332, "loss": 0.21000173687934875, "time": 0.7500866413116455, "epoch": 8, "memory": 6319, "step": 29844} +{"lr": 2.5056380516972453e-05, "data_time": 0.0009514093399047852, "loss": 0.21952392905950546, "time": 0.7852553129196167, "epoch": 8, "memory": 6319, "step": 29944} +{"lr": 2.4799371146339435e-05, "data_time": 0.0009852886199951173, "loss": 0.21530640572309495, "time": 0.8273231029510498, "epoch": 8, "memory": 6319, "step": 30044} +{"lr": 2.4544141867295954e-05, "data_time": 0.0010799884796142579, "loss": 0.2165604904294014, "time": 0.7533233642578125, "epoch": 8, "memory": 6319, "step": 30144} +{"lr": 2.429070772361796e-05, "data_time": 0.0009200096130371094, "loss": 0.20199739933013916, "time": 0.7238510608673095, "epoch": 8, "memory": 6319, "step": 30244} +{"lr": 2.4039083653272075e-05, "data_time": 0.0010300636291503905, "loss": 0.20960500985383987, "time": 0.7509066104888916, "epoch": 8, "memory": 6319, "step": 30344} +{"lr": 2.378928448753535e-05, "data_time": 0.0011282920837402343, "loss": 0.21917520612478256, "time": 0.7533273696899414, "epoch": 8, "memory": 6319, "step": 30444} +{"lr": 2.354132495012082e-05, "data_time": 0.001062321662902832, "loss": 0.20982220470905305, "time": 0.7709638595581054, "epoch": 8, "memory": 6319, "step": 30544} +{"lr": 2.329521965630983e-05, "data_time": 0.0008909225463867188, "loss": 0.20313867926597595, "time": 0.7487683773040772, "epoch": 8, "memory": 6319, "step": 30644} +{"lr": 2.3050983112090472e-05, "data_time": 0.0009563684463500977, "loss": 0.21723445504903793, "time": 0.7826807975769043, "epoch": 8, "memory": 6319, "step": 30744} +{"lr": 2.2808629713302708e-05, "data_time": 0.000956416130065918, "loss": 0.2141541585326195, "time": 0.7382060050964355, "epoch": 8, "memory": 6319, "step": 30844} +{"lr": 2.2568173744789706e-05, "data_time": 0.0009347200393676758, "loss": 0.2159138426184654, "time": 0.7396485090255738, "epoch": 8, "memory": 6319, "step": 30944} +{"lr": 2.2329629379555897e-05, "data_time": 0.000901484489440918, "loss": 0.2172321707010269, "time": 0.8223274230957032, "epoch": 8, "memory": 6319, "step": 31044} +{"lr": 2.2093010677931667e-05, "data_time": 0.0010088205337524414, "loss": 0.19494126737117767, "time": 0.7884708642959595, "epoch": 8, "memory": 6319, "step": 31144} +{"lr": 2.1858331586744434e-05, "data_time": 0.0010422468185424805, "loss": 0.21327109932899474, "time": 0.755548620223999, "epoch": 8, "memory": 6319, "step": 31244} +{"lr": 2.1625605938496814e-05, "data_time": 0.0009706735610961914, "loss": 0.20719381868839265, "time": 0.7514196395874023, "epoch": 8, "memory": 6319, "step": 31344} +{"lr": 2.1394847450551186e-05, "data_time": 0.0011536598205566407, "loss": 0.21146022379398347, "time": 0.802072286605835, "epoch": 8, "memory": 6319, "step": 31444} +{"lr": 2.1166069724321078e-05, "data_time": 0.0010943651199340821, "loss": 0.2106250748038292, "time": 0.758974552154541, "epoch": 8, "memory": 6319, "step": 31544} +{"lr": 2.0939286244469643e-05, "data_time": 0.0009427785873413086, "loss": 0.21476707756519317, "time": 0.7446159362792969, "epoch": 8, "memory": 6319, "step": 31644} +{"lr": 2.071451037811473e-05, "data_time": 0.0009332656860351563, "loss": 0.20031267106533052, "time": 0.7885329723358154, "epoch": 8, "memory": 6319, "step": 31744} +{"lr": 2.0491755374041045e-05, "data_time": 0.0012426137924194335, "loss": 0.2209850773215294, "time": 0.7915916919708252, "epoch": 8, "memory": 6319, "step": 31844} +{"lr": 2.0271034361919193e-05, "data_time": 0.001116776466369629, "loss": 0.21243432015180588, "time": 0.7739881753921509, "epoch": 8, "memory": 6319, "step": 31944} +{"lr": 2.0052360351531814e-05, "data_time": 0.0011266708374023438, "loss": 0.22488388866186143, "time": 0.7203091621398926, "epoch": 8, "memory": 6319, "step": 32044} +{"lr": 1.983574623200682e-05, "data_time": 0.0009102106094360351, "loss": 0.21372572481632232, "time": 0.7694578886032104, "epoch": 8, "memory": 6319, "step": 32144} +{"lr": 1.9621204771057533e-05, "data_time": 0.001020669937133789, "loss": 0.2158012017607689, "time": 0.7751044750213623, "epoch": 8, "memory": 6319, "step": 32244} +{"lr": 1.940874861423025e-05, "data_time": 0.0011316776275634766, "loss": 0.21017730981111526, "time": 0.7497865676879882, "epoch": 8, "memory": 6319, "step": 32344} +{"lr": 1.9198390284158783e-05, "data_time": 0.001055145263671875, "loss": 0.2317439556121826, "time": 0.7470786571502686, "epoch": 8, "memory": 6319, "step": 32444} +{"lr": 1.8990142179826573e-05, "data_time": 0.0009215354919433593, "loss": 0.20885702073574067, "time": 0.7906820774078369, "epoch": 8, "memory": 6319, "step": 32544} +{"lr": 1.8784016575835568e-05, "data_time": 0.0008343219757080078, "loss": 0.20501026064157485, "time": 0.7369314432144165, "epoch": 8, "memory": 6319, "step": 32644} +{"accuracy/top1": 91.2814712524414, "data_time": 0.37664963205655416, "time": 0.4666530847549438, "step": 8} +{"lr": 1.8394249604901642e-05, "data_time": 0.14910335540771485, "loss": 0.2148888349533081, "time": 0.7443775653839111, "epoch": 9, "memory": 6319, "step": 32836} +{"lr": 1.81943907744488e-05, "data_time": 0.0009862422943115235, "loss": 0.20585200190544128, "time": 0.7719999790191651, "epoch": 9, "memory": 6319, "step": 32936} +{"lr": 1.7996701347677866e-05, "data_time": 0.0009792089462280274, "loss": 0.20101186335086824, "time": 0.7506574392318726, "epoch": 9, "memory": 6319, "step": 33036} +{"lr": 1.780119297683906e-05, "data_time": 0.0010838508605957031, "loss": 0.20698247998952865, "time": 0.7333967685699463, "epoch": 9, "memory": 6319, "step": 33136} +{"lr": 1.7607877185626483e-05, "data_time": 0.0010756492614746095, "loss": 0.2013411834836006, "time": 0.7890335321426392, "epoch": 9, "memory": 6319, "step": 33236} +{"lr": 1.7416765368498675e-05, "data_time": 0.000881028175354004, "loss": 0.19780399799346923, "time": 0.73207688331604, "epoch": 9, "memory": 6319, "step": 33336} +{"lr": 1.7227868790007105e-05, "data_time": 0.0009068489074707031, "loss": 0.21908696591854096, "time": 0.751064682006836, "epoch": 9, "memory": 6319, "step": 33436} +{"lr": 1.7041198584132233e-05, "data_time": 0.0011165142059326172, "loss": 0.21218812465667725, "time": 0.7092317104339599, "epoch": 9, "memory": 6319, "step": 33536} +{"lr": 1.6856765753627146e-05, "data_time": 0.0009967327117919923, "loss": 0.21433231979608536, "time": 0.743303632736206, "epoch": 9, "memory": 6319, "step": 33636} +{"lr": 1.6674581169369163e-05, "data_time": 0.0010551691055297851, "loss": 0.2055973917245865, "time": 0.7042733192443847, "epoch": 9, "memory": 6319, "step": 33736} +{"lr": 1.6494655569718982e-05, "data_time": 0.000939631462097168, "loss": 0.21540264040231705, "time": 0.6957808017730713, "epoch": 9, "memory": 6319, "step": 33836} +{"lr": 1.6316999559887786e-05, "data_time": 0.0010408878326416016, "loss": 0.2105497658252716, "time": 0.7735482454299927, "epoch": 9, "memory": 6319, "step": 33936} +{"lr": 1.6141623611312105e-05, "data_time": 0.0010665655136108398, "loss": 0.21456100344657897, "time": 0.7007798433303833, "epoch": 9, "memory": 6319, "step": 34036} +{"lr": 1.5968538061036705e-05, "data_time": 0.000921940803527832, "loss": 0.21272813975811006, "time": 0.7707955837249756, "epoch": 9, "memory": 6319, "step": 34136} +{"lr": 1.5797753111105157e-05, "data_time": 0.0009696722030639648, "loss": 0.2143809661269188, "time": 0.745158338546753, "epoch": 9, "memory": 6319, "step": 34236} +{"lr": 1.5629278827958574e-05, "data_time": 0.001006746292114258, "loss": 0.22629391998052598, "time": 0.7561637163162231, "epoch": 9, "memory": 6319, "step": 34336} +{"lr": 1.5463125141842392e-05, "data_time": 0.0009626388549804687, "loss": 0.21649148613214492, "time": 0.7674808502197266, "epoch": 9, "memory": 6319, "step": 34436} +{"lr": 1.5299301846220796e-05, "data_time": 0.0010663509368896485, "loss": 0.20040907859802246, "time": 0.7117836952209473, "epoch": 9, "memory": 6319, "step": 34536} +{"lr": 1.5137818597199751e-05, "data_time": 0.0009500741958618164, "loss": 0.2169900044798851, "time": 0.7516935110092163, "epoch": 9, "memory": 6319, "step": 34636} +{"lr": 1.4978684912957622e-05, "data_time": 0.0009848356246948242, "loss": 0.20170099139213563, "time": 0.7448330879211426, "epoch": 9, "memory": 6319, "step": 34736} +{"lr": 1.4821910173184321e-05, "data_time": 0.0010041236877441407, "loss": 0.19362652748823167, "time": 0.7400836944580078, "epoch": 9, "memory": 6319, "step": 34836} +{"lr": 1.4667503618528337e-05, "data_time": 0.0009563922882080078, "loss": 0.21154155135154723, "time": 0.7105942964553833, "epoch": 9, "memory": 6319, "step": 34936} +{"lr": 1.4515474350052097e-05, "data_time": 0.0010356903076171875, "loss": 0.19550230354070663, "time": 0.7421216487884521, "epoch": 9, "memory": 6319, "step": 35036} +{"lr": 1.4365831328695616e-05, "data_time": 0.0009246349334716796, "loss": 0.21979222744703292, "time": 0.7416331529617309, "epoch": 9, "memory": 6319, "step": 35136} +{"lr": 1.4218583374748169e-05, "data_time": 0.0012225627899169922, "loss": 0.21451364159584047, "time": 0.8412116050720215, "epoch": 9, "memory": 6319, "step": 35236} +{"lr": 1.4073739167328542e-05, "data_time": 0.0009931325912475586, "loss": 0.20629584938287734, "time": 0.7455545425415039, "epoch": 9, "memory": 6319, "step": 35336} +{"lr": 1.3931307243873344e-05, "data_time": 0.0008781909942626953, "loss": 0.19987259954214096, "time": 0.8046121835708618, "epoch": 9, "memory": 6319, "step": 35436} +{"lr": 1.3791295999633911e-05, "data_time": 0.0007792472839355469, "loss": 0.2130444049835205, "time": 0.7733380556106567, "epoch": 9, "memory": 6319, "step": 35536} +{"lr": 1.3653713687181394e-05, "data_time": 0.0007951498031616211, "loss": 0.2052200198173523, "time": 0.7395313262939454, "epoch": 9, "memory": 6319, "step": 35636} +{"lr": 1.3518568415920337e-05, "data_time": 0.0009598016738891601, "loss": 0.20269481539726258, "time": 0.7832774877548218, "epoch": 9, "memory": 6319, "step": 35736} +{"lr": 1.3385868151610723e-05, "data_time": 0.0010293245315551758, "loss": 0.2250720202922821, "time": 0.8056441307067871, "epoch": 9, "memory": 6319, "step": 35836} +{"lr": 1.3255620715898442e-05, "data_time": 0.000977301597595215, "loss": 0.2165330559015274, "time": 0.7166641235351563, "epoch": 9, "memory": 6319, "step": 35936} +{"lr": 1.3127833785854261e-05, "data_time": 0.0010109424591064453, "loss": 0.21536608934402465, "time": 0.7907366752624512, "epoch": 9, "memory": 6319, "step": 36036} +{"lr": 1.3002514893521313e-05, "data_time": 0.0014429092407226562, "loss": 0.2203567698597908, "time": 0.7522102117538452, "epoch": 9, "memory": 6319, "step": 36136} +{"lr": 1.287967142547117e-05, "data_time": 0.0008806228637695312, "loss": 0.22083668708801268, "time": 0.7012571096420288, "epoch": 9, "memory": 6319, "step": 36236} +{"lr": 1.2759310622368435e-05, "data_time": 0.0009766817092895508, "loss": 0.22841697484254836, "time": 0.7599836826324463, "epoch": 9, "memory": 6319, "step": 36336} +{"lr": 1.2641439578543968e-05, "data_time": 0.0008215427398681641, "loss": 0.2243849068880081, "time": 0.7758415699005127, "epoch": 9, "memory": 6319, "step": 36436} +{"lr": 1.252606524157674e-05, "data_time": 0.000990581512451172, "loss": 0.2114819511771202, "time": 0.810365104675293, "epoch": 9, "memory": 6319, "step": 36536} +{"lr": 1.2413194411884297e-05, "data_time": 0.001035022735595703, "loss": 0.21351986080408097, "time": 0.7465147733688354, "epoch": 9, "memory": 6319, "step": 36636} +{"lr": 1.2302833742321981e-05, "data_time": 0.0010212898254394532, "loss": 0.21108139753341676, "time": 0.7470567941665649, "epoch": 9, "memory": 6319, "step": 36736} +{"accuracy/top1": 91.4784164428711, "data_time": 0.36692734162012736, "time": 0.45497892101605736, "step": 9} +{"lr": 1.2098001439139718e-05, "data_time": 0.16482934951782227, "loss": 0.21655374020338058, "time": 0.7675949096679687, "epoch": 10, "memory": 6319, "step": 36928} +{"lr": 1.1995007122352056e-05, "data_time": 0.0011049270629882812, "loss": 0.2215420961380005, "time": 0.7873828172683716, "epoch": 10, "memory": 6319, "step": 37028} +{"lr": 1.1894547614574522e-05, "data_time": 0.0009546041488647461, "loss": 0.20045895576477052, "time": 0.7119117736816406, "epoch": 10, "memory": 6319, "step": 37128} +{"lr": 1.1796628837111806e-05, "data_time": 0.000859379768371582, "loss": 0.21591064631938933, "time": 0.7177269220352173, "epoch": 10, "memory": 6319, "step": 37228} +{"lr": 1.1701256561512343e-05, "data_time": 0.001059722900390625, "loss": 0.20153772830963135, "time": 0.7311688899993897, "epoch": 10, "memory": 6319, "step": 37328} +{"lr": 1.1608436409228136e-05, "data_time": 0.001040172576904297, "loss": 0.2012565940618515, "time": 0.7108525991439819, "epoch": 10, "memory": 6319, "step": 37428} +{"lr": 1.1518173851283401e-05, "data_time": 0.0009455680847167969, "loss": 0.22018181681632995, "time": 0.7602965593338012, "epoch": 10, "memory": 6319, "step": 37528} +{"lr": 1.1430474207952176e-05, "data_time": 0.0010172605514526367, "loss": 0.2060492068529129, "time": 0.7416475772857666, "epoch": 10, "memory": 6319, "step": 37628} +{"lr": 1.1345342648444593e-05, "data_time": 0.0010094642639160156, "loss": 0.2178183004260063, "time": 0.792791199684143, "epoch": 10, "memory": 6319, "step": 37728} +{"lr": 1.12627841906023e-05, "data_time": 0.0010157346725463868, "loss": 0.2236344709992409, "time": 0.733181357383728, "epoch": 10, "memory": 6319, "step": 37828} +{"lr": 1.1182803700602659e-05, "data_time": 0.0012716054916381836, "loss": 0.1999393418431282, "time": 0.7501315832138061, "epoch": 10, "memory": 6319, "step": 37928} +{"lr": 1.110540589267192e-05, "data_time": 0.0010054349899291993, "loss": 0.20714259147644043, "time": 0.7648464918136597, "epoch": 10, "memory": 6319, "step": 38028} +{"lr": 1.1030595328807396e-05, "data_time": 0.001136612892150879, "loss": 0.21901369988918304, "time": 0.9440932035446167, "epoch": 10, "memory": 6319, "step": 38128} +{"lr": 1.0958376418508494e-05, "data_time": 0.0011468648910522461, "loss": 0.2044278934597969, "time": 0.7064155578613281, "epoch": 10, "memory": 6319, "step": 38228} +{"lr": 1.088875341851686e-05, "data_time": 0.0010543346405029296, "loss": 0.2198479488492012, "time": 0.7076034784317017, "epoch": 10, "memory": 6319, "step": 38328} +{"lr": 1.0821730432565486e-05, "data_time": 0.000953841209411621, "loss": 0.20958588868379593, "time": 0.7295926094055176, "epoch": 10, "memory": 6319, "step": 38428} +{"lr": 1.0757311411136797e-05, "data_time": 0.0008830070495605469, "loss": 0.20698675513267517, "time": 0.7685664653778076, "epoch": 10, "memory": 6319, "step": 38528} +{"lr": 1.0695500151229833e-05, "data_time": 0.0010093212127685546, "loss": 0.2094940572977066, "time": 0.7310352802276612, "epoch": 10, "memory": 6319, "step": 38628} +{"lr": 1.0636300296136373e-05, "data_time": 0.001578521728515625, "loss": 0.2008810892701149, "time": 0.7421096086502075, "epoch": 10, "memory": 6319, "step": 38728} +{"lr": 1.0579715335226322e-05, "data_time": 0.001484513282775879, "loss": 0.2024385690689087, "time": 0.7543192863464355, "epoch": 10, "memory": 6319, "step": 38828} +{"lr": 1.0525748603741896e-05, "data_time": 0.0013012170791625976, "loss": 0.19819287359714508, "time": 0.8326394319534302, "epoch": 10, "memory": 6319, "step": 38928} +{"lr": 1.0474403282601119e-05, "data_time": 0.0010962724685668946, "loss": 0.21920082122087478, "time": 1.1544252157211303, "epoch": 10, "memory": 6319, "step": 39028} +{"lr": 1.0425682398210336e-05, "data_time": 0.0009479522705078125, "loss": 0.1904679462313652, "time": 0.7190548181533813, "epoch": 10, "memory": 6319, "step": 39128} +{"lr": 1.0379588822285784e-05, "data_time": 0.0012860536575317384, "loss": 0.21891588419675828, "time": 0.795698881149292, "epoch": 10, "memory": 6319, "step": 39228} +{"lr": 1.0336125271684364e-05, "data_time": 0.0011662960052490235, "loss": 0.21366954743862152, "time": 0.7244150161743164, "epoch": 10, "memory": 6319, "step": 39328} +{"lr": 1.0295294308243494e-05, "data_time": 0.0010058164596557617, "loss": 0.19279357194900512, "time": 0.7590964555740356, "epoch": 10, "memory": 6319, "step": 39428} +{"lr": 1.0257098338630086e-05, "data_time": 0.0009763240814208984, "loss": 0.19902547746896743, "time": 0.773990249633789, "epoch": 10, "memory": 6319, "step": 39528} +{"lr": 1.0221539614198729e-05, "data_time": 0.0008453369140625, "loss": 0.20715045630931855, "time": 0.7676884889602661, "epoch": 10, "memory": 6319, "step": 39628} +{"lr": 1.0188620230858978e-05, "data_time": 0.0013334035873413086, "loss": 0.21932879239320754, "time": 0.7200077772140503, "epoch": 10, "memory": 6319, "step": 39728} +{"lr": 1.01583421289518e-05, "data_time": 0.0010752916336059571, "loss": 0.20916725993156432, "time": 0.7598609685897827, "epoch": 10, "memory": 6319, "step": 39828} +{"lr": 1.0130707093135206e-05, "data_time": 0.0011291265487670898, "loss": 0.21107010394334794, "time": 0.7480459928512573, "epoch": 10, "memory": 6319, "step": 39928} +{"lr": 1.01057167522791e-05, "data_time": 0.001167893409729004, "loss": 0.20092849880456926, "time": 0.772868013381958, "epoch": 10, "memory": 6319, "step": 40028} +{"lr": 1.008337257936921e-05, "data_time": 0.0010306835174560547, "loss": 0.1973315417766571, "time": 0.7603323221206665, "epoch": 10, "memory": 6319, "step": 40128} +{"lr": 1.0063675891420294e-05, "data_time": 0.0009876489639282227, "loss": 0.2090567946434021, "time": 0.7465966224670411, "epoch": 10, "memory": 6319, "step": 40228} +{"lr": 1.0046627849398548e-05, "data_time": 0.0008569002151489258, "loss": 0.22173888087272645, "time": 0.8362118482589722, "epoch": 10, "memory": 6319, "step": 40328} +{"lr": 1.0032229458153106e-05, "data_time": 0.0009721994400024414, "loss": 0.21461016237735747, "time": 0.7650289535522461, "epoch": 10, "memory": 6319, "step": 40428} +{"lr": 1.0020481566356862e-05, "data_time": 0.0018167734146118165, "loss": 0.2005124107003212, "time": 0.7336384534835816, "epoch": 10, "memory": 6319, "step": 40528} +{"lr": 1.0011384866456456e-05, "data_time": 0.001233386993408203, "loss": 0.1984155297279358, "time": 0.7361355304718018, "epoch": 10, "memory": 6319, "step": 40628} +{"lr": 1.00049398946314e-05, "data_time": 0.0008237361907958984, "loss": 0.21123445630073548, "time": 0.7409549236297608, "epoch": 10, "memory": 6319, "step": 40728} +{"lr": 1.0001147030762536e-05, "data_time": 0.00107574462890625, "loss": 0.21124005317687988, "time": 0.7571876049041748, "epoch": 10, "memory": 6319, "step": 40828} +{"accuracy/top1": 91.90043640136719, "data_time": 0.3827948351701101, "time": 0.4716785252094269, "step": 10} diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/config.py b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/config.py new file mode 100644 index 0000000000000000000000000000000000000000..7a0f9f2c3060ac00c4c97a418905c91572dba3b1 --- /dev/null +++ b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/config.py @@ -0,0 +1,349 @@ +optim_wrapper = dict( + optimizer=dict( + type='SGD', + lr=0.0001, + momentum=0.9, + weight_decay=0.0001, + _scope_='mmpretrain'), + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=False, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=512) +model = dict( + type='ImageClassifier', + backbone=dict( + frozen_stages=2, + type='ResNet', + depth=50, + num_stages=4, + out_indices=(3, ), + style='pytorch', + init_cfg=dict( + type='Pretrained', + checkpoint= + 'https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth', + prefix='backbone')), + neck=dict(type='GlobalAveragePooling'), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=2048, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + topk=1)) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/all_0.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/all_1.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3fake8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/cc1m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3real8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV2-1-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/if-dpmsolver++-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = dict(type='Accuracy', topk=1) +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV2-1-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/if-dpmsolver++-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = dict(type='Accuracy', topk=1) +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1' diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/events.out.tfevents.1685984307.SH-IDC1-10-140-24-118.71685.0 b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/events.out.tfevents.1685984307.SH-IDC1-10-140-24-118.71685.0 new file mode 100644 index 0000000000000000000000000000000000000000..02cd54ad19fbb088173b7fd4b774a22d08a29697 --- /dev/null +++ b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/events.out.tfevents.1685984307.SH-IDC1-10-140-24-118.71685.0 @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:f709fc2aa9f71cb6f9503de0d1e8c3e07c1c82e4a0141b5c35099089441aeac8 +size 5424630 diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/scalars.json b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/scalars.json new file mode 100644 index 0000000000000000000000000000000000000000..6529b6690b2940fc50c96e3d0898101d50592287 --- /dev/null +++ b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/scalars.json @@ -0,0 +1,410 @@ +{"lr": 9.999870019168206e-05, "data_time": 0.056679391860961915, "loss": 0.6467106103897095, "time": 0.7275593042373657, "epoch": 1, "memory": 9436, "step": 100} +{"lr": 9.99947481934738e-05, "data_time": 0.000693058967590332, "loss": 0.6183649361133575, "time": 0.7685099601745605, "epoch": 1, "memory": 6319, "step": 200} +{"lr": 9.99881441056942e-05, "data_time": 0.000945425033569336, "loss": 0.5818252265453339, "time": 0.7908675193786621, "epoch": 1, "memory": 6319, "step": 300} +{"lr": 9.997888831760252e-05, "data_time": 0.0008037805557250977, "loss": 0.5466198921203613, "time": 0.8349925994873046, "epoch": 1, "memory": 6319, "step": 400} +{"lr": 9.996698137475549e-05, "data_time": 0.0007397413253784179, "loss": 0.5174447357654571, "time": 0.8036102294921875, "epoch": 1, "memory": 6319, "step": 500} +{"lr": 9.99524239789746e-05, "data_time": 0.0009145259857177735, "loss": 0.5112586677074432, "time": 0.7819676399230957, "epoch": 1, "memory": 6319, "step": 600} +{"lr": 9.993521698830466e-05, "data_time": 0.0008785486221313477, "loss": 0.47699070870876314, "time": 0.7547931909561157, "epoch": 1, "memory": 6319, "step": 700} +{"lr": 9.991536141696371e-05, "data_time": 0.0011017799377441406, "loss": 0.4670168846845627, "time": 0.7601818799972534, "epoch": 1, "memory": 6319, "step": 800} +{"lr": 9.989285843528274e-05, "data_time": 0.0009255647659301758, "loss": 0.451236367225647, "time": 0.7634871482849122, "epoch": 1, "memory": 6319, "step": 900} +{"lr": 9.986770936963713e-05, "data_time": 0.0009256362915039063, "loss": 0.4457569599151611, "time": 0.74671790599823, "epoch": 1, "memory": 6319, "step": 1000} +{"lr": 9.983991570236818e-05, "data_time": 0.001474618911743164, "loss": 0.4359045296907425, "time": 0.745058250427246, "epoch": 1, "memory": 6319, "step": 1100} +{"lr": 9.980947907169598e-05, "data_time": 0.000811624526977539, "loss": 0.4352470725774765, "time": 0.8330950736999512, "epoch": 1, "memory": 6319, "step": 1200} +{"lr": 9.977640127162247e-05, "data_time": 0.0009439945220947266, "loss": 0.41018485128879545, "time": 0.7362479925155639, "epoch": 1, "memory": 6319, "step": 1300} +{"lr": 9.974068425182603e-05, "data_time": 0.0008623600006103516, "loss": 0.4059330612421036, "time": 0.7337836503982544, "epoch": 1, "memory": 6319, "step": 1400} +{"lr": 9.970233011754657e-05, "data_time": 0.0008291244506835937, "loss": 0.3948523014783859, "time": 0.7408311128616333, "epoch": 1, "memory": 6319, "step": 1500} +{"lr": 9.966134112946122e-05, "data_time": 0.0008630990982055664, "loss": 0.39278132617473605, "time": 0.759043526649475, "epoch": 1, "memory": 6319, "step": 1600} +{"lr": 9.961771970355126e-05, "data_time": 0.0009460926055908203, "loss": 0.383815535902977, "time": 0.66086847782135, "epoch": 1, "memory": 6319, "step": 1700} +{"lr": 9.957146841095961e-05, "data_time": 0.0007583856582641602, "loss": 0.3866573393344879, "time": 0.7454272747039795, "epoch": 1, "memory": 6319, "step": 1800} +{"lr": 9.952258997783933e-05, "data_time": 0.0007628202438354492, "loss": 0.3863100469112396, "time": 0.7970134735107421, "epoch": 1, "memory": 6319, "step": 1900} +{"lr": 9.947108728519288e-05, "data_time": 0.0008378028869628906, "loss": 0.36880527436733246, "time": 0.7101362943649292, "epoch": 1, "memory": 6319, "step": 2000} +{"lr": 9.941696336870245e-05, "data_time": 0.0008944988250732422, "loss": 0.3723113119602203, "time": 0.7135246753692627, "epoch": 1, "memory": 6319, "step": 2100} +{"lr": 9.936022141855094e-05, "data_time": 0.0008395910263061523, "loss": 0.374591463804245, "time": 0.8074831962585449, "epoch": 1, "memory": 6319, "step": 2200} +{"lr": 9.93008647792338e-05, "data_time": 0.000999617576599121, "loss": 0.38273478150367735, "time": 0.729729413986206, "epoch": 1, "memory": 6319, "step": 2300} +{"lr": 9.923889694936212e-05, "data_time": 0.0007698535919189453, "loss": 0.3737144827842712, "time": 0.7085587024688721, "epoch": 1, "memory": 6319, "step": 2400} +{"lr": 9.917432158145639e-05, "data_time": 0.0008183479309082031, "loss": 0.3717155665159225, "time": 0.6826485633850098, "epoch": 1, "memory": 6319, "step": 2500} +{"lr": 9.91071424817309e-05, "data_time": 0.0008365631103515625, "loss": 0.3504417181015015, "time": 0.7524774312973023, "epoch": 1, "memory": 6319, "step": 2600} +{"lr": 9.903736360986979e-05, "data_time": 0.0008868932723999024, "loss": 0.34623405933380125, "time": 0.74358229637146, "epoch": 1, "memory": 6319, "step": 2700} +{"lr": 9.89649890787935e-05, "data_time": 0.001009058952331543, "loss": 0.3519168525934219, "time": 0.6897189617156982, "epoch": 1, "memory": 6319, "step": 2800} +{"lr": 9.889002315441642e-05, "data_time": 0.0010261774063110352, "loss": 0.34067114591598513, "time": 0.7585605144500732, "epoch": 1, "memory": 6319, "step": 2900} +{"lr": 9.881247025539515e-05, "data_time": 0.0008624076843261718, "loss": 0.3508040338754654, "time": 0.7910493373870849, "epoch": 1, "memory": 6319, "step": 3000} +{"lr": 9.873233495286844e-05, "data_time": 0.001009964942932129, "loss": 0.3420281380414963, "time": 0.7930436372756958, "epoch": 1, "memory": 6319, "step": 3100} +{"lr": 9.864962197018756e-05, "data_time": 0.001180124282836914, "loss": 0.34925495088100433, "time": 0.6975256919860839, "epoch": 1, "memory": 6319, "step": 3200} +{"lr": 9.856433618263779e-05, "data_time": 0.0009097814559936524, "loss": 0.3305091917514801, "time": 0.7505454301834107, "epoch": 1, "memory": 6319, "step": 3300} +{"lr": 9.847648261715132e-05, "data_time": 0.0008378505706787109, "loss": 0.3468285262584686, "time": 0.741396713256836, "epoch": 1, "memory": 6319, "step": 3400} +{"lr": 9.838606645201093e-05, "data_time": 0.0009875297546386719, "loss": 0.3498389571905136, "time": 0.7579817295074462, "epoch": 1, "memory": 6319, "step": 3500} +{"lr": 9.829309301654456e-05, "data_time": 0.0008234739303588867, "loss": 0.33495093882083893, "time": 0.7445247888565063, "epoch": 1, "memory": 6319, "step": 3600} +{"lr": 9.819756779081127e-05, "data_time": 0.0008818864822387696, "loss": 0.32544230222702025, "time": 0.7830398797988891, "epoch": 1, "memory": 6319, "step": 3700} +{"lr": 9.809949640527814e-05, "data_time": 0.0010114431381225586, "loss": 0.33996055722236634, "time": 0.7672888517379761, "epoch": 1, "memory": 6319, "step": 3800} +{"lr": 9.799888464048879e-05, "data_time": 0.0008689165115356445, "loss": 0.32864971160888673, "time": 0.769545316696167, "epoch": 1, "memory": 6319, "step": 3900} +{"lr": 9.789573842672242e-05, "data_time": 0.0008574485778808594, "loss": 0.3149187624454498, "time": 0.7635641098022461, "epoch": 1, "memory": 6319, "step": 4000} +{"accuracy/top1": 82.76980590820312, "data_time": 0.3998277928648877, "time": 0.4892594173174946, "step": 1} +{"lr": 9.7690615523361e-05, "data_time": 0.45548300743103026, "loss": 0.30172773599624636, "time": 0.740324330329895, "epoch": 2, "memory": 6319, "step": 4192} +{"lr": 9.758010405927656e-05, "data_time": 0.27181134223937986, "loss": 0.3268072783946991, "time": 0.7494885444641113, "epoch": 2, "memory": 6319, "step": 4292} +{"lr": 9.746708283006492e-05, "data_time": 0.11563518047332763, "loss": 0.3033058255910873, "time": 0.7373050928115845, "epoch": 2, "memory": 6319, "step": 4392} +{"lr": 9.735155849744634e-05, "data_time": 0.23847999572753906, "loss": 0.32278717756271363, "time": 0.8111628293991089, "epoch": 2, "memory": 6319, "step": 4492} +{"lr": 9.723353787067928e-05, "data_time": 0.017300224304199217, "loss": 0.3331650346517563, "time": 0.7327423095703125, "epoch": 2, "memory": 6319, "step": 4592} +{"lr": 9.711302790615951e-05, "data_time": 0.001006484031677246, "loss": 0.30172336399555205, "time": 0.7489974498748779, "epoch": 2, "memory": 6319, "step": 4692} +{"lr": 9.699003570700986e-05, "data_time": 0.0008834362030029297, "loss": 0.30616395473480223, "time": 0.7078217029571533, "epoch": 2, "memory": 6319, "step": 4792} +{"lr": 9.686456852266142e-05, "data_time": 0.0010566473007202148, "loss": 0.3306979715824127, "time": 0.7499691247940063, "epoch": 2, "memory": 6319, "step": 4892} +{"lr": 9.673663374842644e-05, "data_time": 0.0009475231170654296, "loss": 0.3279380649328232, "time": 0.7641098976135254, "epoch": 2, "memory": 6319, "step": 4992} +{"lr": 9.660623892506218e-05, "data_time": 0.0008621692657470703, "loss": 0.30873491168022155, "time": 0.743563199043274, "epoch": 2, "memory": 6319, "step": 5092} +{"lr": 9.647339173832693e-05, "data_time": 0.0010243654251098633, "loss": 0.2860911160707474, "time": 0.8009820461273194, "epoch": 2, "memory": 6319, "step": 5192} +{"lr": 9.633810001852653e-05, "data_time": 0.0009762525558471679, "loss": 0.32079139053821565, "time": 0.7218252658843994, "epoch": 2, "memory": 6319, "step": 5292} +{"lr": 9.620037174005274e-05, "data_time": 0.0009346485137939453, "loss": 0.3037598103284836, "time": 0.7187831401824951, "epoch": 2, "memory": 6319, "step": 5392} +{"lr": 9.606021502091389e-05, "data_time": 0.0008486747741699219, "loss": 0.2915645271539688, "time": 0.6923610687255859, "epoch": 2, "memory": 6319, "step": 5492} +{"lr": 9.591763812225569e-05, "data_time": 0.0008571147918701172, "loss": 0.313795217871666, "time": 0.6917939424514771, "epoch": 2, "memory": 6319, "step": 5592} +{"lr": 9.577264944787459e-05, "data_time": 0.0009221792221069336, "loss": 0.2936663508415222, "time": 0.7407344818115235, "epoch": 2, "memory": 6319, "step": 5692} +{"lr": 9.562525754372252e-05, "data_time": 0.0009955644607543945, "loss": 0.30850992500782015, "time": 0.8105541229248047, "epoch": 2, "memory": 6319, "step": 5792} +{"lr": 9.547547109740282e-05, "data_time": 0.0009494543075561524, "loss": 0.3004725694656372, "time": 0.7119806051254273, "epoch": 2, "memory": 6319, "step": 5892} +{"lr": 9.532329893765858e-05, "data_time": 0.0009148836135864258, "loss": 0.30849842727184296, "time": 0.7046910762786865, "epoch": 2, "memory": 6319, "step": 5992} +{"lr": 9.516875003385223e-05, "data_time": 0.0009780168533325196, "loss": 0.2957474052906036, "time": 0.7011518239974975, "epoch": 2, "memory": 6319, "step": 6092} +{"lr": 9.501183349543653e-05, "data_time": 0.0007869243621826172, "loss": 0.29207600057125094, "time": 0.7678962469100952, "epoch": 2, "memory": 6319, "step": 6192} +{"lr": 9.48525585714179e-05, "data_time": 0.0009556293487548828, "loss": 0.2962953090667725, "time": 0.7099686384201049, "epoch": 2, "memory": 6319, "step": 6292} +{"lr": 9.469093464981119e-05, "data_time": 0.0010651111602783202, "loss": 0.29158331602811816, "time": 0.7452439308166504, "epoch": 2, "memory": 6319, "step": 6392} +{"lr": 9.452697125708636e-05, "data_time": 0.0008752107620239258, "loss": 0.2812229782342911, "time": 0.7753681421279908, "epoch": 2, "memory": 6319, "step": 6492} +{"lr": 9.436067805760705e-05, "data_time": 0.0009801149368286132, "loss": 0.300742506980896, "time": 0.868291974067688, "epoch": 2, "memory": 6319, "step": 6592} +{"lr": 9.419206485306072e-05, "data_time": 0.0008069753646850586, "loss": 0.2915066510438919, "time": 0.7001688003540039, "epoch": 2, "memory": 6319, "step": 6692} +{"lr": 9.4021141581881e-05, "data_time": 0.0008558273315429687, "loss": 0.31252866685390474, "time": 0.7760474205017089, "epoch": 2, "memory": 6319, "step": 6792} +{"lr": 9.384791831866193e-05, "data_time": 0.0009261846542358398, "loss": 0.27259269952774046, "time": 0.7401493549346924, "epoch": 2, "memory": 6319, "step": 6892} +{"lr": 9.367240527356428e-05, "data_time": 0.0009130477905273437, "loss": 0.30035998821258547, "time": 0.7291918277740479, "epoch": 2, "memory": 6319, "step": 6992} +{"lr": 9.349461279171362e-05, "data_time": 0.0010501861572265625, "loss": 0.277800253033638, "time": 0.7776790142059327, "epoch": 2, "memory": 6319, "step": 7092} +{"lr": 9.33145513525904e-05, "data_time": 0.0010162353515625, "loss": 0.3006815016269684, "time": 0.7420114040374756, "epoch": 2, "memory": 6319, "step": 7192} +{"lr": 9.313223156941242e-05, "data_time": 0.0007892608642578125, "loss": 0.2931386008858681, "time": 0.7074287891387939, "epoch": 2, "memory": 6319, "step": 7292} +{"lr": 9.294766418850942e-05, "data_time": 0.0012914657592773438, "loss": 0.27622135281562804, "time": 0.8082712650299072, "epoch": 2, "memory": 6319, "step": 7392} +{"lr": 9.276086008868929e-05, "data_time": 0.0007749557495117188, "loss": 0.2855895534157753, "time": 0.8478997468948364, "epoch": 2, "memory": 6319, "step": 7492} +{"lr": 9.257183028059726e-05, "data_time": 0.0009172439575195312, "loss": 0.28419963717460633, "time": 0.6746926069259643, "epoch": 2, "memory": 6319, "step": 7592} +{"lr": 9.238058590606654e-05, "data_time": 0.000856328010559082, "loss": 0.2811689838767052, "time": 0.7300053596496582, "epoch": 2, "memory": 6319, "step": 7692} +{"lr": 9.218713823746189e-05, "data_time": 0.0009179115295410156, "loss": 0.2922960877418518, "time": 0.7394140243530274, "epoch": 2, "memory": 6319, "step": 7792} +{"lr": 9.199149867701492e-05, "data_time": 0.0010849475860595704, "loss": 0.3005239307880402, "time": 0.722748064994812, "epoch": 2, "memory": 6319, "step": 7892} +{"lr": 9.179367875615232e-05, "data_time": 0.0009314298629760742, "loss": 0.27577583491802216, "time": 0.7644084453582763, "epoch": 2, "memory": 6319, "step": 7992} +{"lr": 9.159369013481585e-05, "data_time": 0.000997447967529297, "loss": 0.2708585262298584, "time": 0.7618850231170654, "epoch": 2, "memory": 6319, "step": 8092} +{"accuracy/top1": 81.92244720458984, "data_time": 0.3712924559911092, "time": 0.4612708806991577, "step": 2} +{"lr": 9.120367596460475e-05, "data_time": 0.36234614849090574, "loss": 0.2805113524198532, "time": 0.7783551454544068, "epoch": 3, "memory": 6319, "step": 8284} +{"lr": 9.099742266672751e-05, "data_time": 0.24762063026428222, "loss": 0.2812726765871048, "time": 0.7314421653747558, "epoch": 3, "memory": 6319, "step": 8384} +{"lr": 9.078904760146445e-05, "data_time": 0.011304831504821778, "loss": 0.2848355293273926, "time": 0.7501899003982544, "epoch": 3, "memory": 6319, "step": 8484} +{"lr": 9.057856305090082e-05, "data_time": 0.0009741306304931641, "loss": 0.28337272703647615, "time": 0.7842912435531616, "epoch": 3, "memory": 6319, "step": 8584} +{"lr": 9.036598142145979e-05, "data_time": 0.0008115053176879882, "loss": 0.2887425020337105, "time": 0.7473387718200684, "epoch": 3, "memory": 6319, "step": 8684} +{"lr": 9.015131524317079e-05, "data_time": 0.0007852315902709961, "loss": 0.28306626081466674, "time": 0.7298887252807618, "epoch": 3, "memory": 6319, "step": 8784} +{"lr": 8.993457716893124e-05, "data_time": 0.0009052038192749023, "loss": 0.2758233904838562, "time": 0.7469028949737548, "epoch": 3, "memory": 6319, "step": 8884} +{"lr": 8.971577997376067e-05, "data_time": 0.0007829427719116211, "loss": 0.26109735667705536, "time": 0.7335215091705323, "epoch": 3, "memory": 6319, "step": 8984} +{"lr": 8.949493655404777e-05, "data_time": 0.0007196664810180664, "loss": 0.28642739951610563, "time": 0.7904765367507934, "epoch": 3, "memory": 6319, "step": 9084} +{"lr": 8.927205992679013e-05, "data_time": 0.0008393764495849609, "loss": 0.28590462505817416, "time": 0.8574235677719116, "epoch": 3, "memory": 6319, "step": 9184} +{"lr": 8.9047163228827e-05, "data_time": 0.0010226964950561523, "loss": 0.2825495645403862, "time": 0.7643748760223389, "epoch": 3, "memory": 6319, "step": 9284} +{"lr": 8.88202597160652e-05, "data_time": 0.0009260177612304688, "loss": 0.28266668170690534, "time": 0.7886485815048218, "epoch": 3, "memory": 6319, "step": 9384} +{"lr": 8.859136276269775e-05, "data_time": 0.000884103775024414, "loss": 0.27662210166454315, "time": 0.7629488229751586, "epoch": 3, "memory": 6319, "step": 9484} +{"lr": 8.836048586041507e-05, "data_time": 0.0008762359619140625, "loss": 0.28159307688474655, "time": 0.7175558567047119, "epoch": 3, "memory": 6319, "step": 9584} +{"lr": 8.812764261761037e-05, "data_time": 0.0009073734283447266, "loss": 0.26138431280851365, "time": 0.8578940629959106, "epoch": 3, "memory": 6319, "step": 9684} +{"lr": 8.789284675857743e-05, "data_time": 0.0007905006408691406, "loss": 0.2703880339860916, "time": 0.7499826908111572, "epoch": 3, "memory": 6319, "step": 9784} +{"lr": 8.76561121227013e-05, "data_time": 0.0007951259613037109, "loss": 0.25421659350395204, "time": 0.6923384189605712, "epoch": 3, "memory": 6319, "step": 9884} +{"lr": 8.741745266364282e-05, "data_time": 0.0007498264312744141, "loss": 0.2759989321231842, "time": 0.7331349611282348, "epoch": 3, "memory": 6319, "step": 9984} +{"lr": 8.717688244851635e-05, "data_time": 0.0008265495300292969, "loss": 0.2653114333748817, "time": 0.7779464721679688, "epoch": 3, "memory": 6319, "step": 10084} +{"lr": 8.693441565706024e-05, "data_time": 0.0010085105895996094, "loss": 0.27250864207744596, "time": 0.6818900108337402, "epoch": 3, "memory": 6319, "step": 10184} +{"lr": 8.669006658080132e-05, "data_time": 0.0010039567947387694, "loss": 0.2683863386511803, "time": 0.7410855054855346, "epoch": 3, "memory": 6319, "step": 10284} +{"lr": 8.644384962221226e-05, "data_time": 0.000935816764831543, "loss": 0.26774340569972993, "time": 0.7439370393753052, "epoch": 3, "memory": 6319, "step": 10384} +{"lr": 8.619577929386303e-05, "data_time": 0.000908660888671875, "loss": 0.27209125012159346, "time": 0.7391719818115234, "epoch": 3, "memory": 6319, "step": 10484} +{"lr": 8.594587021756517e-05, "data_time": 0.0008325338363647461, "loss": 0.2583186015486717, "time": 0.7654966831207275, "epoch": 3, "memory": 6319, "step": 10584} +{"lr": 8.569413712351023e-05, "data_time": 0.0009442806243896484, "loss": 0.26382465809583666, "time": 0.7567195892333984, "epoch": 3, "memory": 6319, "step": 10684} +{"lr": 8.544059484940119e-05, "data_time": 0.0008954048156738281, "loss": 0.28303846418857576, "time": 0.7387443542480469, "epoch": 3, "memory": 6319, "step": 10784} +{"lr": 8.518525833957818e-05, "data_time": 0.0007791996002197265, "loss": 0.2826122909784317, "time": 0.7503203868865966, "epoch": 3, "memory": 6319, "step": 10884} +{"lr": 8.49281426441377e-05, "data_time": 0.0009593486785888672, "loss": 0.24396598637104033, "time": 0.7173983812332153, "epoch": 3, "memory": 6319, "step": 10984} +{"lr": 8.466926291804501e-05, "data_time": 0.0010905742645263671, "loss": 0.26798778772354126, "time": 0.6998972415924072, "epoch": 3, "memory": 6319, "step": 11084} +{"lr": 8.440863442024147e-05, "data_time": 0.0008717060089111328, "loss": 0.275073966383934, "time": 0.7636283874511719, "epoch": 3, "memory": 6319, "step": 11184} +{"lr": 8.414627251274488e-05, "data_time": 0.0009631156921386719, "loss": 0.257562592625618, "time": 0.7458836793899536, "epoch": 3, "memory": 6319, "step": 11284} +{"lr": 8.388219265974407e-05, "data_time": 0.0007251024246215821, "loss": 0.2575678452849388, "time": 0.7709132432937622, "epoch": 3, "memory": 6319, "step": 11384} +{"lr": 8.361641042668715e-05, "data_time": 0.0007683038711547852, "loss": 0.2427933245897293, "time": 0.7077937364578247, "epoch": 3, "memory": 6319, "step": 11484} +{"lr": 8.33489414793644e-05, "data_time": 0.0009012937545776368, "loss": 0.2670065313577652, "time": 0.7723667621612549, "epoch": 3, "memory": 6319, "step": 11584} +{"lr": 8.30798015829848e-05, "data_time": 0.0009708881378173828, "loss": 0.26711356043815615, "time": 0.7584527015686036, "epoch": 3, "memory": 6319, "step": 11684} +{"lr": 8.280900660124665e-05, "data_time": 0.0008171796798706055, "loss": 0.25504983514547347, "time": 0.7334778308868408, "epoch": 3, "memory": 6319, "step": 11784} +{"lr": 8.253657249540274e-05, "data_time": 0.0009721040725708008, "loss": 0.2652930110692978, "time": 0.7529042959213257, "epoch": 3, "memory": 6319, "step": 11884} +{"lr": 8.226251532331926e-05, "data_time": 0.001077413558959961, "loss": 0.25756279528141024, "time": 0.7682068109512329, "epoch": 3, "memory": 6319, "step": 11984} +{"lr": 8.198685123852965e-05, "data_time": 0.0009561300277709961, "loss": 0.26171095967292785, "time": 0.7986436128616333, "epoch": 3, "memory": 6319, "step": 12084} +{"lr": 8.17095964892824e-05, "data_time": 0.00106048583984375, "loss": 0.25484343320131303, "time": 0.704461932182312, "epoch": 3, "memory": 6319, "step": 12184} +{"accuracy/top1": 81.7387466430664, "data_time": 0.3710541745026906, "time": 0.46016467014948526, "step": 3} +{"lr": 8.11728683565904e-05, "data_time": 0.44006059169769285, "loss": 0.25812426656484605, "time": 0.7065037012100219, "epoch": 4, "memory": 6319, "step": 12376} +{"lr": 8.089106273476776e-05, "data_time": 0.07465133666992188, "loss": 0.24335235059261323, "time": 0.7446303129196167, "epoch": 4, "memory": 6319, "step": 12476} +{"lr": 8.060773103667566e-05, "data_time": 0.0009151697158813477, "loss": 0.24658887535333635, "time": 0.745331597328186, "epoch": 4, "memory": 6319, "step": 12576} +{"lr": 8.03228899625083e-05, "data_time": 0.0010189294815063476, "loss": 0.24688779711723327, "time": 0.7623863458633423, "epoch": 4, "memory": 6319, "step": 12676} +{"lr": 8.003655630142627e-05, "data_time": 0.0008771657943725586, "loss": 0.2450454831123352, "time": 0.7806264162063599, "epoch": 4, "memory": 6319, "step": 12776} +{"lr": 7.974874693056617e-05, "data_time": 0.0009460210800170899, "loss": 0.2516448900103569, "time": 0.7400216102600098, "epoch": 4, "memory": 6319, "step": 12876} +{"lr": 7.945947881404618e-05, "data_time": 0.0009044170379638672, "loss": 0.25994026511907575, "time": 0.7677648067474365, "epoch": 4, "memory": 6319, "step": 12976} +{"lr": 7.916876900196646e-05, "data_time": 0.0009540081024169922, "loss": 0.2557961091399193, "time": 0.7357230663299561, "epoch": 4, "memory": 6319, "step": 13076} +{"lr": 7.887663462940359e-05, "data_time": 0.0009281635284423828, "loss": 0.24798357039690017, "time": 0.7497295379638672, "epoch": 4, "memory": 6319, "step": 13176} +{"lr": 7.858309291540111e-05, "data_time": 0.0010016679763793946, "loss": 0.258859620988369, "time": 0.7436985015869141, "epoch": 4, "memory": 6319, "step": 13276} +{"lr": 7.828816116195411e-05, "data_time": 0.0008557796478271485, "loss": 0.2308677166700363, "time": 0.7458075761795044, "epoch": 4, "memory": 6319, "step": 13376} +{"lr": 7.799185675298994e-05, "data_time": 0.0010064363479614258, "loss": 0.24006345719099045, "time": 0.7534310102462769, "epoch": 4, "memory": 6319, "step": 13476} +{"lr": 7.769419715334297e-05, "data_time": 0.0010347843170166015, "loss": 0.24197655767202378, "time": 0.7990076303482055, "epoch": 4, "memory": 6319, "step": 13576} +{"lr": 7.739519990772583e-05, "data_time": 0.0012361526489257813, "loss": 0.24522503465414047, "time": 0.7416842699050903, "epoch": 4, "memory": 6319, "step": 13676} +{"lr": 7.709488263969475e-05, "data_time": 0.0010061025619506835, "loss": 0.2393003895878792, "time": 0.7449990272521972, "epoch": 4, "memory": 6319, "step": 13776} +{"lr": 7.67932630506109e-05, "data_time": 0.0008320093154907226, "loss": 0.2356681138277054, "time": 0.7493997812271118, "epoch": 4, "memory": 6319, "step": 13876} +{"lr": 7.64903589185975e-05, "data_time": 0.0008714199066162109, "loss": 0.24267417639493943, "time": 0.7086737155914307, "epoch": 4, "memory": 6319, "step": 13976} +{"lr": 7.618618809749134e-05, "data_time": 0.0008412599563598633, "loss": 0.23524534702301025, "time": 0.7754576921463012, "epoch": 4, "memory": 6319, "step": 14076} +{"lr": 7.588076851579028e-05, "data_time": 0.0008939027786254883, "loss": 0.25485320687294005, "time": 0.7532485961914063, "epoch": 4, "memory": 6319, "step": 14176} +{"lr": 7.557411817559755e-05, "data_time": 0.0009070396423339844, "loss": 0.2572333440184593, "time": 0.764794921875, "epoch": 4, "memory": 6319, "step": 14276} +{"lr": 7.526625515155957e-05, "data_time": 0.0009389400482177734, "loss": 0.23196712732315064, "time": 0.6891292572021485, "epoch": 4, "memory": 6319, "step": 14376} +{"lr": 7.495719758980122e-05, "data_time": 0.0008939743041992187, "loss": 0.26005916446447375, "time": 0.7295319318771363, "epoch": 4, "memory": 6319, "step": 14476} +{"lr": 7.464696370685586e-05, "data_time": 0.0008395671844482422, "loss": 0.24326847940683366, "time": 0.8304219484329224, "epoch": 4, "memory": 6319, "step": 14576} +{"lr": 7.433557178859198e-05, "data_time": 0.0009120941162109375, "loss": 0.23468608111143113, "time": 0.7421232461929321, "epoch": 4, "memory": 6319, "step": 14676} +{"lr": 7.402304018913512e-05, "data_time": 0.001082634925842285, "loss": 0.24131477624177933, "time": 0.7425987243652343, "epoch": 4, "memory": 6319, "step": 14776} +{"lr": 7.37093873297861e-05, "data_time": 0.0009795427322387695, "loss": 0.2344892367720604, "time": 0.7353773593902588, "epoch": 4, "memory": 6319, "step": 14876} +{"lr": 7.339463169793563e-05, "data_time": 0.0009601593017578125, "loss": 0.2357519179582596, "time": 0.8386085748672485, "epoch": 4, "memory": 6319, "step": 14976} +{"lr": 7.307879184597369e-05, "data_time": 0.0008764982223510743, "loss": 0.25934259295463563, "time": 1.4188462972640992, "epoch": 4, "memory": 6319, "step": 15076} +{"lr": 7.276188639019704e-05, "data_time": 0.0009303569793701171, "loss": 0.23962348699569702, "time": 0.7592996120452881, "epoch": 4, "memory": 6319, "step": 15176} +{"lr": 7.244393400971127e-05, "data_time": 0.0008882284164428711, "loss": 0.23132585138082504, "time": 0.7509804487228393, "epoch": 4, "memory": 6319, "step": 15276} +{"lr": 7.212495344533006e-05, "data_time": 0.0007544994354248047, "loss": 0.2398171529173851, "time": 0.7699744462966919, "epoch": 4, "memory": 6319, "step": 15376} +{"lr": 7.180496349847055e-05, "data_time": 0.0010762214660644531, "loss": 0.22637886852025985, "time": 0.8533263444900513, "epoch": 4, "memory": 6319, "step": 15476} +{"lr": 7.148398303004499e-05, "data_time": 0.0010487556457519532, "loss": 0.25670819282531737, "time": 0.7617285251617432, "epoch": 4, "memory": 6319, "step": 15576} +{"lr": 7.11620309593493e-05, "data_time": 0.0008179903030395508, "loss": 0.22662961184978486, "time": 0.8312148809432983, "epoch": 4, "memory": 6319, "step": 15676} +{"lr": 7.083912626294795e-05, "data_time": 0.0010239362716674804, "loss": 0.23864293545484544, "time": 0.7462512969970703, "epoch": 4, "memory": 6319, "step": 15776} +{"lr": 7.051528797355494e-05, "data_time": 0.0007915973663330078, "loss": 0.24410039633512498, "time": 0.7130609750747681, "epoch": 4, "memory": 6319, "step": 15876} +{"lr": 7.019053517891214e-05, "data_time": 0.0008965730667114258, "loss": 0.2492641806602478, "time": 0.7135108947753906, "epoch": 4, "memory": 6319, "step": 15976} +{"lr": 6.986488702066473e-05, "data_time": 0.0008026123046875, "loss": 0.23730763494968415, "time": 0.7190106868743896, "epoch": 4, "memory": 6319, "step": 16076} +{"lr": 6.953836269323245e-05, "data_time": 0.0009332895278930664, "loss": 0.2515053525567055, "time": 0.8029151678085327, "epoch": 4, "memory": 6319, "step": 16176} +{"lr": 6.921098144267838e-05, "data_time": 0.0009343624114990234, "loss": 0.2330927789211273, "time": 0.7749029159545898, "epoch": 4, "memory": 6319, "step": 16276} +{"accuracy/top1": 85.10832214355469, "data_time": 0.3379850685596466, "time": 0.42568763891855876, "step": 4} +{"lr": 6.858007803673889e-05, "data_time": 0.30223779678344725, "loss": 0.2544656455516815, "time": 0.7357942581176757, "epoch": 5, "memory": 6319, "step": 16468} +{"lr": 6.825030518869046e-05, "data_time": 0.0009554862976074219, "loss": 0.23199640065431595, "time": 0.7362879276275635, "epoch": 5, "memory": 6319, "step": 16568} +{"lr": 6.791975133846686e-05, "data_time": 0.0009989261627197266, "loss": 0.23119949102401732, "time": 0.7697957277297973, "epoch": 5, "memory": 6319, "step": 16668} +{"lr": 6.758843596964013e-05, "data_time": 0.0007458925247192383, "loss": 0.239640474319458, "time": 1.3155960321426392, "epoch": 5, "memory": 6319, "step": 16768} +{"lr": 6.725637861066804e-05, "data_time": 0.0008596420288085937, "loss": 0.22820000052452089, "time": 0.7898166656494141, "epoch": 5, "memory": 6319, "step": 16868} +{"lr": 6.692359883374264e-05, "data_time": 0.0008529901504516601, "loss": 0.23233971148729324, "time": 0.8249887466430664, "epoch": 5, "memory": 6319, "step": 16968} +{"lr": 6.659011625363718e-05, "data_time": 0.0008084297180175782, "loss": 0.23109327405691146, "time": 0.70664381980896, "epoch": 5, "memory": 6319, "step": 17068} +{"lr": 6.625595052654927e-05, "data_time": 0.0008114814758300781, "loss": 0.23301598131656648, "time": 0.7499979972839356, "epoch": 5, "memory": 6319, "step": 17168} +{"lr": 6.592112134894315e-05, "data_time": 0.000906825065612793, "loss": 0.25075470060110094, "time": 0.7827856063842773, "epoch": 5, "memory": 6319, "step": 17268} +{"lr": 6.558564845638802e-05, "data_time": 0.0009220361709594727, "loss": 0.2466512367129326, "time": 0.7277509450912476, "epoch": 5, "memory": 6319, "step": 17368} +{"lr": 6.524955162239518e-05, "data_time": 0.0009633302688598633, "loss": 0.255857852101326, "time": 0.7504189014434814, "epoch": 5, "memory": 6319, "step": 17468} +{"lr": 6.491285065725238e-05, "data_time": 0.0010142326354980469, "loss": 0.22700437754392624, "time": 0.7106493949890137, "epoch": 5, "memory": 6319, "step": 17568} +{"lr": 6.457556540685618e-05, "data_time": 0.0007700920104980469, "loss": 0.24164402335882187, "time": 0.7481193304061889, "epoch": 5, "memory": 6319, "step": 17668} +{"lr": 6.423771575154208e-05, "data_time": 0.0009311914443969726, "loss": 0.23107862025499343, "time": 0.7690333604812623, "epoch": 5, "memory": 6319, "step": 17768} +{"lr": 6.389932160491309e-05, "data_time": 0.0008931636810302734, "loss": 0.24904999434947966, "time": 0.7814180135726929, "epoch": 5, "memory": 6319, "step": 17868} +{"lr": 6.356040291266556e-05, "data_time": 0.000942683219909668, "loss": 0.23286448419094086, "time": 0.7281208276748657, "epoch": 5, "memory": 6319, "step": 17968} +{"lr": 6.322097965141391e-05, "data_time": 0.000815272331237793, "loss": 0.23617454320192338, "time": 0.7271085500717163, "epoch": 5, "memory": 6319, "step": 18068} +{"lr": 6.288107182751267e-05, "data_time": 0.0009077787399291992, "loss": 0.22853530049324036, "time": 0.7365193843841553, "epoch": 5, "memory": 6319, "step": 18168} +{"lr": 6.254069947587782e-05, "data_time": 0.000838780403137207, "loss": 0.23225885778665542, "time": 0.7225019454956054, "epoch": 5, "memory": 6319, "step": 18268} +{"lr": 6.219988265880537e-05, "data_time": 0.0009633541107177734, "loss": 0.24403035044670104, "time": 0.8011118650436402, "epoch": 5, "memory": 6319, "step": 18368} +{"lr": 6.185864146478941e-05, "data_time": 0.0011283397674560548, "loss": 0.24354525804519653, "time": 0.7545641660690308, "epoch": 5, "memory": 6319, "step": 18468} +{"lr": 6.151699600733752e-05, "data_time": 0.001166391372680664, "loss": 0.2370014175772667, "time": 0.7521046161651611, "epoch": 5, "memory": 6319, "step": 18568} +{"lr": 6.117496642378554e-05, "data_time": 0.0008833169937133789, "loss": 0.22047528475522996, "time": 0.7016199350357055, "epoch": 5, "memory": 6319, "step": 18668} +{"lr": 6.0832572874110315e-05, "data_time": 0.0008026123046875, "loss": 0.22848183959722518, "time": 0.7139702796936035, "epoch": 5, "memory": 6319, "step": 18768} +{"lr": 6.0489835539741924e-05, "data_time": 0.0010669708251953125, "loss": 0.22957628965377808, "time": 0.6931902408599854, "epoch": 5, "memory": 6319, "step": 18868} +{"lr": 6.0146774622374006e-05, "data_time": 0.0009476900100708008, "loss": 0.22941379994153976, "time": 0.7554872751235961, "epoch": 5, "memory": 6319, "step": 18968} +{"lr": 5.980341034277274e-05, "data_time": 0.0010091781616210938, "loss": 0.23331795632839203, "time": 0.7668699026107788, "epoch": 5, "memory": 6319, "step": 19068} +{"lr": 5.945976293958505e-05, "data_time": 0.0009149551391601562, "loss": 0.2550873875617981, "time": 0.8021178722381592, "epoch": 5, "memory": 6319, "step": 19168} +{"lr": 5.911585266814587e-05, "data_time": 0.0009304285049438477, "loss": 0.2470840498805046, "time": 0.7086685419082641, "epoch": 5, "memory": 6319, "step": 19268} +{"lr": 5.877169979928426e-05, "data_time": 0.0010169267654418946, "loss": 0.22779435217380523, "time": 0.7370935678482056, "epoch": 5, "memory": 6319, "step": 19368} +{"lr": 5.842732461812828e-05, "data_time": 0.0010720252990722655, "loss": 0.22472428679466247, "time": 0.7397331714630127, "epoch": 5, "memory": 6319, "step": 19468} +{"lr": 5.8082747422909646e-05, "data_time": 0.0009012699127197265, "loss": 0.24025715589523317, "time": 0.7366528749465943, "epoch": 5, "memory": 6319, "step": 19568} +{"lr": 5.7737988523767406e-05, "data_time": 0.000866103172302246, "loss": 0.21452924907207488, "time": 0.7156328678131103, "epoch": 5, "memory": 6319, "step": 19668} +{"lr": 5.739306824155042e-05, "data_time": 0.00106503963470459, "loss": 0.22239121943712234, "time": 0.7647297143936157, "epoch": 5, "memory": 6319, "step": 19768} +{"lr": 5.704800690661993e-05, "data_time": 0.0010779857635498046, "loss": 0.22004520297050476, "time": 0.8111758708953858, "epoch": 5, "memory": 6319, "step": 19868} +{"lr": 5.6702824857650994e-05, "data_time": 0.0008751630783081054, "loss": 0.218601992726326, "time": 0.734070086479187, "epoch": 5, "memory": 6319, "step": 19968} +{"lr": 5.635754244043411e-05, "data_time": 0.0008763790130615235, "loss": 0.2396036222577095, "time": 0.723269009590149, "epoch": 5, "memory": 6319, "step": 20068} +{"lr": 5.60121800066754e-05, "data_time": 0.0011265993118286133, "loss": 0.24507984519004822, "time": 0.7516209125518799, "epoch": 5, "memory": 6319, "step": 20168} +{"lr": 5.5666757912797516e-05, "data_time": 0.0010689020156860352, "loss": 0.2263699859380722, "time": 0.7594061374664307, "epoch": 5, "memory": 6319, "step": 20268} +{"lr": 5.532129651873955e-05, "data_time": 0.0009542226791381836, "loss": 0.2092998206615448, "time": 0.7460134744644165, "epoch": 5, "memory": 6319, "step": 20368} +{"accuracy/top1": 88.72946166992188, "data_time": 0.37101765473683673, "time": 0.46014336347579954, "step": 5} +{"lr": 5.465797506068693e-05, "data_time": 0.3617511749267578, "loss": 0.22097585052251817, "time": 0.7422191143035889, "epoch": 6, "memory": 6319, "step": 20560} +{"lr": 5.431251545044144e-05, "data_time": 0.08920300006866455, "loss": 0.22178382724523543, "time": 0.7930021047592163, "epoch": 6, "memory": 6319, "step": 20660} +{"lr": 5.396709636204968e-05, "data_time": 0.000785374641418457, "loss": 0.22732941210269927, "time": 0.812529730796814, "epoch": 6, "memory": 6319, "step": 20760} +{"lr": 5.3621738155273566e-05, "data_time": 0.0008415937423706055, "loss": 0.2372802272439003, "time": 0.7552832365036011, "epoch": 6, "memory": 6319, "step": 20860} +{"lr": 5.3276461186286584e-05, "data_time": 0.0010387659072875976, "loss": 0.2372077465057373, "time": 0.7639619827270507, "epoch": 6, "memory": 6319, "step": 20960} +{"lr": 5.293128580647384e-05, "data_time": 0.0010754823684692382, "loss": 0.22168179750442504, "time": 0.7425286769866943, "epoch": 6, "memory": 6319, "step": 21060} +{"lr": 5.258623236123256e-05, "data_time": 0.0009563446044921875, "loss": 0.21794785112142562, "time": 0.7136249780654907, "epoch": 6, "memory": 6319, "step": 21160} +{"lr": 5.224132118877281e-05, "data_time": 0.0016475200653076171, "loss": 0.21800965070724487, "time": 0.7487157821655274, "epoch": 6, "memory": 6319, "step": 21260} +{"lr": 5.189657261891887e-05, "data_time": 0.0009242534637451172, "loss": 0.24410653710365296, "time": 0.7653291463851929, "epoch": 6, "memory": 6319, "step": 21360} +{"lr": 5.1552006971910974e-05, "data_time": 0.0007371187210083007, "loss": 0.2157185912132263, "time": 0.8699023962020874, "epoch": 6, "memory": 6319, "step": 21460} +{"lr": 5.120764455720728e-05, "data_time": 0.0012006282806396485, "loss": 0.21663552671670913, "time": 0.7318055391311645, "epoch": 6, "memory": 6319, "step": 21560} +{"lr": 5.0863505672287034e-05, "data_time": 0.0009980440139770509, "loss": 0.22659144848585128, "time": 0.8226141214370728, "epoch": 6, "memory": 6319, "step": 21660} +{"lr": 5.051961060145421e-05, "data_time": 0.0009065389633178711, "loss": 0.21539592742919922, "time": 0.7309055328369141, "epoch": 6, "memory": 6319, "step": 21760} +{"lr": 5.017597961464177e-05, "data_time": 0.0009898900985717773, "loss": 0.23304681926965715, "time": 0.772872018814087, "epoch": 6, "memory": 6319, "step": 21860} +{"lr": 4.983263296621707e-05, "data_time": 0.0008554697036743164, "loss": 0.21356911659240724, "time": 0.72887282371521, "epoch": 6, "memory": 6319, "step": 21960} +{"lr": 4.948959089378785e-05, "data_time": 0.0008954286575317382, "loss": 0.21990423947572707, "time": 0.7353631019592285, "epoch": 6, "memory": 6319, "step": 22060} +{"lr": 4.91468736170094e-05, "data_time": 0.0008244991302490234, "loss": 0.22922151684761047, "time": 0.7417398691177368, "epoch": 6, "memory": 6319, "step": 22160} +{"lr": 4.880450133639314e-05, "data_time": 0.0011096000671386719, "loss": 0.22476275265216827, "time": 0.7459015130996705, "epoch": 6, "memory": 6319, "step": 22260} +{"lr": 4.8462494232115335e-05, "data_time": 0.000895380973815918, "loss": 0.21308909356594086, "time": 0.7584392309188843, "epoch": 6, "memory": 6319, "step": 22360} +{"lr": 4.812087246282806e-05, "data_time": 0.0009870529174804688, "loss": 0.21166886687278746, "time": 0.7301443099975586, "epoch": 6, "memory": 6319, "step": 22460} +{"lr": 4.7779656164470954e-05, "data_time": 0.0013822317123413086, "loss": 0.2199772208929062, "time": 0.7756128787994385, "epoch": 6, "memory": 6319, "step": 22560} +{"lr": 4.7438865449084246e-05, "data_time": 0.0008816242218017579, "loss": 0.2076578766107559, "time": 0.768445348739624, "epoch": 6, "memory": 6319, "step": 22660} +{"lr": 4.709852040362338e-05, "data_time": 0.0009389638900756836, "loss": 0.21123078167438508, "time": 0.7135071277618408, "epoch": 6, "memory": 6319, "step": 22760} +{"lr": 4.675864108877499e-05, "data_time": 0.0008680343627929688, "loss": 0.24213262051343917, "time": 0.7669236183166503, "epoch": 6, "memory": 6319, "step": 22860} +{"lr": 4.641924753777464e-05, "data_time": 0.00105743408203125, "loss": 0.22617121487855912, "time": 0.7091005086898804, "epoch": 6, "memory": 6319, "step": 22960} +{"lr": 4.608035975522578e-05, "data_time": 0.0007989645004272461, "loss": 0.24261579364538194, "time": 0.7463184356689453, "epoch": 6, "memory": 6319, "step": 23060} +{"lr": 4.574199771592085e-05, "data_time": 0.0009153127670288086, "loss": 0.23417214751243592, "time": 0.7600003242492676, "epoch": 6, "memory": 6319, "step": 23160} +{"lr": 4.540418136366372e-05, "data_time": 0.000960850715637207, "loss": 0.24277285784482955, "time": 0.7358004093170166, "epoch": 6, "memory": 6319, "step": 23260} +{"lr": 4.506693061009437e-05, "data_time": 0.0011164188385009766, "loss": 0.20732830762863158, "time": 0.7263560771942139, "epoch": 6, "memory": 6319, "step": 23360} +{"lr": 4.473026533351511e-05, "data_time": 0.001043844223022461, "loss": 0.2153569743037224, "time": 0.7638959407806396, "epoch": 6, "memory": 6319, "step": 23460} +{"lr": 4.439420537771888e-05, "data_time": 0.0010238170623779296, "loss": 0.2066410392522812, "time": 0.7720999002456665, "epoch": 6, "memory": 6319, "step": 23560} +{"lr": 4.405877055081978e-05, "data_time": 0.0012173891067504884, "loss": 0.22008976042270662, "time": 0.7492174625396728, "epoch": 6, "memory": 6319, "step": 23660} +{"lr": 4.3723980624085373e-05, "data_time": 0.0009804725646972655, "loss": 0.2285812348127365, "time": 0.6996923446655273, "epoch": 6, "memory": 6319, "step": 23760} +{"lr": 4.3389855330771394e-05, "data_time": 0.000852513313293457, "loss": 0.22510926574468612, "time": 0.7598581790924073, "epoch": 6, "memory": 6319, "step": 23860} +{"lr": 4.305641436495869e-05, "data_time": 0.0008151769638061523, "loss": 0.21471958756446838, "time": 0.7302027702331543, "epoch": 6, "memory": 6319, "step": 23960} +{"lr": 4.272367738039222e-05, "data_time": 0.001053190231323242, "loss": 0.20233013182878495, "time": 0.7877671957015991, "epoch": 6, "memory": 6319, "step": 24060} +{"lr": 4.239166398932278e-05, "data_time": 0.0008234500885009766, "loss": 0.20474919229745864, "time": 0.7330990076065064, "epoch": 6, "memory": 6319, "step": 24160} +{"lr": 4.206039376135108e-05, "data_time": 0.0009666681289672852, "loss": 0.24494716972112657, "time": 0.7451794385910034, "epoch": 6, "memory": 6319, "step": 24260} +{"lr": 4.1729886222273985e-05, "data_time": 0.0010160684585571289, "loss": 0.21276360154151916, "time": 0.7946240425109863, "epoch": 6, "memory": 6319, "step": 24360} +{"lr": 4.140016085293389e-05, "data_time": 0.0008048057556152344, "loss": 0.21543313711881637, "time": 0.7634910106658935, "epoch": 6, "memory": 6319, "step": 24460} +{"accuracy/top1": 89.29381561279297, "data_time": 0.3655286471048991, "time": 0.45561564366022744, "step": 6} +{"lr": 4.0769351868715556e-05, "data_time": 0.06758735179901124, "loss": 0.21464929431676866, "time": 0.7436929702758789, "epoch": 7, "memory": 6319, "step": 24652} +{"lr": 4.044202148988293e-05, "data_time": 0.0009186506271362304, "loss": 0.22030574530363084, "time": 0.7825503826141358, "epoch": 7, "memory": 6319, "step": 24752} +{"lr": 4.0115549190371644e-05, "data_time": 0.0007718324661254883, "loss": 0.22843018025159836, "time": 0.8764800548553466, "epoch": 7, "memory": 6319, "step": 24852} +{"lr": 3.978995421317808e-05, "data_time": 0.0010705947875976562, "loss": 0.2284991666674614, "time": 0.7792644262313843, "epoch": 7, "memory": 6319, "step": 24952} +{"lr": 3.94652557495873e-05, "data_time": 0.0010233163833618165, "loss": 0.20828754156827928, "time": 0.6905143976211547, "epoch": 7, "memory": 6319, "step": 25052} +{"lr": 3.914147293804207e-05, "data_time": 0.0012282133102416992, "loss": 0.20730749368667603, "time": 0.7519061326980591, "epoch": 7, "memory": 6319, "step": 25152} +{"lr": 3.881862486301448e-05, "data_time": 0.0010767459869384765, "loss": 0.22793783396482467, "time": 0.8323409795761109, "epoch": 7, "memory": 6319, "step": 25252} +{"lr": 3.849673055388122e-05, "data_time": 0.0010067224502563477, "loss": 0.2198184385895729, "time": 0.7189236164093018, "epoch": 7, "memory": 6319, "step": 25352} +{"lr": 3.817580898380197e-05, "data_time": 0.0009664535522460937, "loss": 0.22881572246551513, "time": 0.7632386207580566, "epoch": 7, "memory": 6319, "step": 25452} +{"lr": 3.7855879068600975e-05, "data_time": 0.000894618034362793, "loss": 0.22118446081876755, "time": 0.7141993284225464, "epoch": 7, "memory": 6319, "step": 25552} +{"lr": 3.753695966565216e-05, "data_time": 0.0012279272079467774, "loss": 0.21850900799036027, "time": 0.726638388633728, "epoch": 7, "memory": 6319, "step": 25652} +{"lr": 3.7219069572767636e-05, "data_time": 0.0010407924652099609, "loss": 0.21882450729608535, "time": 0.7235573530197144, "epoch": 7, "memory": 6319, "step": 25752} +{"lr": 3.6902227527089725e-05, "data_time": 0.001052522659301758, "loss": 0.21903080642223358, "time": 0.7414105415344239, "epoch": 7, "memory": 6319, "step": 25852} +{"lr": 3.65864522039865e-05, "data_time": 0.0009268522262573242, "loss": 0.20320204198360442, "time": 0.7302920103073121, "epoch": 7, "memory": 6319, "step": 25952} +{"lr": 3.627176221595109e-05, "data_time": 0.0008532524108886719, "loss": 0.22782006859779358, "time": 0.7634247541427612, "epoch": 7, "memory": 6319, "step": 26052} +{"lr": 3.595817611150461e-05, "data_time": 0.0010223388671875, "loss": 0.2195037305355072, "time": 0.7388573169708252, "epoch": 7, "memory": 6319, "step": 26152} +{"lr": 3.5645712374102824e-05, "data_time": 0.0011571884155273438, "loss": 0.22454675883054734, "time": 0.7472745180130005, "epoch": 7, "memory": 6319, "step": 26252} +{"lr": 3.533438942104663e-05, "data_time": 0.0010906219482421874, "loss": 0.2088132008910179, "time": 0.6744309663772583, "epoch": 7, "memory": 6319, "step": 26352} +{"lr": 3.5024225602396704e-05, "data_time": 0.0010611772537231444, "loss": 0.21927298307418824, "time": 0.7297801017761231, "epoch": 7, "memory": 6319, "step": 26452} +{"lr": 3.4715239199891735e-05, "data_time": 0.001035928726196289, "loss": 0.22908590584993363, "time": 0.7075364589691162, "epoch": 7, "memory": 6319, "step": 26552} +{"lr": 3.44074484258709e-05, "data_time": 0.0010905265808105469, "loss": 0.21959491521120073, "time": 0.786644434928894, "epoch": 7, "memory": 6319, "step": 26652} +{"lr": 3.4100871422200426e-05, "data_time": 0.0009969472885131836, "loss": 0.2215401530265808, "time": 0.7292780160903931, "epoch": 7, "memory": 6319, "step": 26752} +{"lr": 3.379552625920419e-05, "data_time": 0.0010289192199707032, "loss": 0.21800469607114792, "time": 0.7371968507766724, "epoch": 7, "memory": 6319, "step": 26852} +{"lr": 3.349143093459868e-05, "data_time": 0.0009022235870361328, "loss": 0.2225462332367897, "time": 0.7333777666091919, "epoch": 7, "memory": 6319, "step": 26952} +{"lr": 3.318860337243226e-05, "data_time": 0.0012842893600463867, "loss": 0.22466444969177246, "time": 0.7128596305847168, "epoch": 7, "memory": 6319, "step": 27052} +{"lr": 3.288706142202845e-05, "data_time": 0.001050424575805664, "loss": 0.21334576606750488, "time": 0.7456519603729248, "epoch": 7, "memory": 6319, "step": 27152} +{"lr": 3.25868228569341e-05, "data_time": 0.0009516716003417969, "loss": 0.21006509959697722, "time": 0.7747654914855957, "epoch": 7, "memory": 6319, "step": 27252} +{"lr": 3.228790537387148e-05, "data_time": 0.0008413076400756836, "loss": 0.21703979820013047, "time": 0.7487269163131713, "epoch": 7, "memory": 6319, "step": 27352} +{"lr": 3.199032659169556e-05, "data_time": 0.000826716423034668, "loss": 0.23915895074605942, "time": 0.7505497217178345, "epoch": 7, "memory": 6319, "step": 27452} +{"lr": 3.169410405035527e-05, "data_time": 0.0008703947067260742, "loss": 0.23334365487098693, "time": 0.7179245948791504, "epoch": 7, "memory": 6319, "step": 27552} +{"lr": 3.139925520985965e-05, "data_time": 0.000834512710571289, "loss": 0.2168520823121071, "time": 0.7441962242126465, "epoch": 7, "memory": 6319, "step": 27652} +{"lr": 3.1105797449248916e-05, "data_time": 0.0009701013565063477, "loss": 0.21715566962957383, "time": 0.7484423637390136, "epoch": 7, "memory": 6319, "step": 27752} +{"lr": 3.081374806556986e-05, "data_time": 0.0008368730545043945, "loss": 0.2253271907567978, "time": 0.6734090328216553, "epoch": 7, "memory": 6319, "step": 27852} +{"lr": 3.052312427285649e-05, "data_time": 0.0008078336715698243, "loss": 0.22412145286798477, "time": 0.7776374101638794, "epoch": 7, "memory": 6319, "step": 27952} +{"lr": 3.0233943201115213e-05, "data_time": 0.000777745246887207, "loss": 0.20857783108949662, "time": 0.7468486785888672, "epoch": 7, "memory": 6319, "step": 28052} +{"lr": 2.9946221895315525e-05, "data_time": 0.0010638713836669921, "loss": 0.22127020359039307, "time": 0.7438156366348266, "epoch": 7, "memory": 6319, "step": 28152} +{"lr": 2.9659977314384953e-05, "data_time": 0.0008136749267578125, "loss": 0.2155949130654335, "time": 0.7352388143539429, "epoch": 7, "memory": 6319, "step": 28252} +{"lr": 2.93752263302096e-05, "data_time": 0.0009730815887451172, "loss": 0.20265680700540542, "time": 0.7197938203811646, "epoch": 7, "memory": 6319, "step": 28352} +{"lr": 2.9091985726639714e-05, "data_time": 0.0009970664978027344, "loss": 0.20858417302370072, "time": 0.766517949104309, "epoch": 7, "memory": 6319, "step": 28452} +{"lr": 2.881027219850035e-05, "data_time": 0.0010104656219482421, "loss": 0.22796138674020766, "time": 0.7162450551986694, "epoch": 7, "memory": 6319, "step": 28552} +{"accuracy/top1": 90.86441040039062, "data_time": 0.37313953836758934, "time": 0.46175772746404015, "step": 7} +{"lr": 2.827372366659638e-05, "data_time": 0.2035313367843628, "loss": 0.20221212059259414, "time": 0.778073787689209, "epoch": 8, "memory": 6319, "step": 28744} +{"lr": 2.7996563897301616e-05, "data_time": 0.001152348518371582, "loss": 0.2191552385687828, "time": 0.7858430624008179, "epoch": 8, "memory": 6319, "step": 28844} +{"lr": 2.7720995770013028e-05, "data_time": 0.0009917259216308595, "loss": 0.222273613512516, "time": 0.773386287689209, "epoch": 8, "memory": 6319, "step": 28944} +{"lr": 2.744703552732299e-05, "data_time": 0.0010296106338500977, "loss": 0.20417798608541488, "time": 0.7599788904190063, "epoch": 8, "memory": 6319, "step": 29044} +{"lr": 2.717469931705171e-05, "data_time": 0.0008471965789794922, "loss": 0.2055888667702675, "time": 0.7366097450256348, "epoch": 8, "memory": 6319, "step": 29144} +{"lr": 2.690400319129529e-05, "data_time": 0.0008450031280517578, "loss": 0.21146828234195708, "time": 0.7604643344879151, "epoch": 8, "memory": 6319, "step": 29244} +{"lr": 2.6634963105479716e-05, "data_time": 0.0009392976760864257, "loss": 0.21690429002046585, "time": 0.8979794979095459, "epoch": 8, "memory": 6319, "step": 29344} +{"lr": 2.6367594917420177e-05, "data_time": 0.0008240222930908203, "loss": 0.21820570677518844, "time": 0.7451939105987548, "epoch": 8, "memory": 6319, "step": 29444} +{"lr": 2.610191438638673e-05, "data_time": 0.000759577751159668, "loss": 0.2177011102437973, "time": 0.7357152700424194, "epoch": 8, "memory": 6319, "step": 29544} +{"lr": 2.5837937172175032e-05, "data_time": 0.0008301019668579101, "loss": 0.22187730818986892, "time": 0.7183035373687744, "epoch": 8, "memory": 6319, "step": 29644} +{"lr": 2.557567883418364e-05, "data_time": 0.0008044242858886719, "loss": 0.20425197333097458, "time": 0.7523620367050171, "epoch": 8, "memory": 6319, "step": 29744} +{"lr": 2.531515483049656e-05, "data_time": 0.000857853889465332, "loss": 0.21000173687934875, "time": 0.7500866413116455, "epoch": 8, "memory": 6319, "step": 29844} +{"lr": 2.5056380516972453e-05, "data_time": 0.0009514093399047852, "loss": 0.21952392905950546, "time": 0.7852553129196167, "epoch": 8, "memory": 6319, "step": 29944} +{"lr": 2.4799371146339435e-05, "data_time": 0.0009852886199951173, "loss": 0.21530640572309495, "time": 0.8273231029510498, "epoch": 8, "memory": 6319, "step": 30044} +{"lr": 2.4544141867295954e-05, "data_time": 0.0010799884796142579, "loss": 0.2165604904294014, "time": 0.7533233642578125, "epoch": 8, "memory": 6319, "step": 30144} +{"lr": 2.429070772361796e-05, "data_time": 0.0009200096130371094, "loss": 0.20199739933013916, "time": 0.7238510608673095, "epoch": 8, "memory": 6319, "step": 30244} +{"lr": 2.4039083653272075e-05, "data_time": 0.0010300636291503905, "loss": 0.20960500985383987, "time": 0.7509066104888916, "epoch": 8, "memory": 6319, "step": 30344} +{"lr": 2.378928448753535e-05, "data_time": 0.0011282920837402343, "loss": 0.21917520612478256, "time": 0.7533273696899414, "epoch": 8, "memory": 6319, "step": 30444} +{"lr": 2.354132495012082e-05, "data_time": 0.001062321662902832, "loss": 0.20982220470905305, "time": 0.7709638595581054, "epoch": 8, "memory": 6319, "step": 30544} +{"lr": 2.329521965630983e-05, "data_time": 0.0008909225463867188, "loss": 0.20313867926597595, "time": 0.7487683773040772, "epoch": 8, "memory": 6319, "step": 30644} +{"lr": 2.3050983112090472e-05, "data_time": 0.0009563684463500977, "loss": 0.21723445504903793, "time": 0.7826807975769043, "epoch": 8, "memory": 6319, "step": 30744} +{"lr": 2.2808629713302708e-05, "data_time": 0.000956416130065918, "loss": 0.2141541585326195, "time": 0.7382060050964355, "epoch": 8, "memory": 6319, "step": 30844} +{"lr": 2.2568173744789706e-05, "data_time": 0.0009347200393676758, "loss": 0.2159138426184654, "time": 0.7396485090255738, "epoch": 8, "memory": 6319, "step": 30944} +{"lr": 2.2329629379555897e-05, "data_time": 0.000901484489440918, "loss": 0.2172321707010269, "time": 0.8223274230957032, "epoch": 8, "memory": 6319, "step": 31044} +{"lr": 2.2093010677931667e-05, "data_time": 0.0010088205337524414, "loss": 0.19494126737117767, "time": 0.7884708642959595, "epoch": 8, "memory": 6319, "step": 31144} +{"lr": 2.1858331586744434e-05, "data_time": 0.0010422468185424805, "loss": 0.21327109932899474, "time": 0.755548620223999, "epoch": 8, "memory": 6319, "step": 31244} +{"lr": 2.1625605938496814e-05, "data_time": 0.0009706735610961914, "loss": 0.20719381868839265, "time": 0.7514196395874023, "epoch": 8, "memory": 6319, "step": 31344} +{"lr": 2.1394847450551186e-05, "data_time": 0.0011536598205566407, "loss": 0.21146022379398347, "time": 0.802072286605835, "epoch": 8, "memory": 6319, "step": 31444} +{"lr": 2.1166069724321078e-05, "data_time": 0.0010943651199340821, "loss": 0.2106250748038292, "time": 0.758974552154541, "epoch": 8, "memory": 6319, "step": 31544} +{"lr": 2.0939286244469643e-05, "data_time": 0.0009427785873413086, "loss": 0.21476707756519317, "time": 0.7446159362792969, "epoch": 8, "memory": 6319, "step": 31644} +{"lr": 2.071451037811473e-05, "data_time": 0.0009332656860351563, "loss": 0.20031267106533052, "time": 0.7885329723358154, "epoch": 8, "memory": 6319, "step": 31744} +{"lr": 2.0491755374041045e-05, "data_time": 0.0012426137924194335, "loss": 0.2209850773215294, "time": 0.7915916919708252, "epoch": 8, "memory": 6319, "step": 31844} +{"lr": 2.0271034361919193e-05, "data_time": 0.001116776466369629, "loss": 0.21243432015180588, "time": 0.7739881753921509, "epoch": 8, "memory": 6319, "step": 31944} +{"lr": 2.0052360351531814e-05, "data_time": 0.0011266708374023438, "loss": 0.22488388866186143, "time": 0.7203091621398926, "epoch": 8, "memory": 6319, "step": 32044} +{"lr": 1.983574623200682e-05, "data_time": 0.0009102106094360351, "loss": 0.21372572481632232, "time": 0.7694578886032104, "epoch": 8, "memory": 6319, "step": 32144} +{"lr": 1.9621204771057533e-05, "data_time": 0.001020669937133789, "loss": 0.2158012017607689, "time": 0.7751044750213623, "epoch": 8, "memory": 6319, "step": 32244} +{"lr": 1.940874861423025e-05, "data_time": 0.0011316776275634766, "loss": 0.21017730981111526, "time": 0.7497865676879882, "epoch": 8, "memory": 6319, "step": 32344} +{"lr": 1.9198390284158783e-05, "data_time": 0.001055145263671875, "loss": 0.2317439556121826, "time": 0.7470786571502686, "epoch": 8, "memory": 6319, "step": 32444} +{"lr": 1.8990142179826573e-05, "data_time": 0.0009215354919433593, "loss": 0.20885702073574067, "time": 0.7906820774078369, "epoch": 8, "memory": 6319, "step": 32544} +{"lr": 1.8784016575835568e-05, "data_time": 0.0008343219757080078, "loss": 0.20501026064157485, "time": 0.7369314432144165, "epoch": 8, "memory": 6319, "step": 32644} +{"accuracy/top1": 91.2814712524414, "data_time": 0.37664963205655416, "time": 0.4666530847549438, "step": 8} +{"lr": 1.8394249604901642e-05, "data_time": 0.14910335540771485, "loss": 0.2148888349533081, "time": 0.7443775653839111, "epoch": 9, "memory": 6319, "step": 32836} +{"lr": 1.81943907744488e-05, "data_time": 0.0009862422943115235, "loss": 0.20585200190544128, "time": 0.7719999790191651, "epoch": 9, "memory": 6319, "step": 32936} +{"lr": 1.7996701347677866e-05, "data_time": 0.0009792089462280274, "loss": 0.20101186335086824, "time": 0.7506574392318726, "epoch": 9, "memory": 6319, "step": 33036} +{"lr": 1.780119297683906e-05, "data_time": 0.0010838508605957031, "loss": 0.20698247998952865, "time": 0.7333967685699463, "epoch": 9, "memory": 6319, "step": 33136} +{"lr": 1.7607877185626483e-05, "data_time": 0.0010756492614746095, "loss": 0.2013411834836006, "time": 0.7890335321426392, "epoch": 9, "memory": 6319, "step": 33236} +{"lr": 1.7416765368498675e-05, "data_time": 0.000881028175354004, "loss": 0.19780399799346923, "time": 0.73207688331604, "epoch": 9, "memory": 6319, "step": 33336} +{"lr": 1.7227868790007105e-05, "data_time": 0.0009068489074707031, "loss": 0.21908696591854096, "time": 0.751064682006836, "epoch": 9, "memory": 6319, "step": 33436} +{"lr": 1.7041198584132233e-05, "data_time": 0.0011165142059326172, "loss": 0.21218812465667725, "time": 0.7092317104339599, "epoch": 9, "memory": 6319, "step": 33536} +{"lr": 1.6856765753627146e-05, "data_time": 0.0009967327117919923, "loss": 0.21433231979608536, "time": 0.743303632736206, "epoch": 9, "memory": 6319, "step": 33636} +{"lr": 1.6674581169369163e-05, "data_time": 0.0010551691055297851, "loss": 0.2055973917245865, "time": 0.7042733192443847, "epoch": 9, "memory": 6319, "step": 33736} +{"lr": 1.6494655569718982e-05, "data_time": 0.000939631462097168, "loss": 0.21540264040231705, "time": 0.6957808017730713, "epoch": 9, "memory": 6319, "step": 33836} +{"lr": 1.6316999559887786e-05, "data_time": 0.0010408878326416016, "loss": 0.2105497658252716, "time": 0.7735482454299927, "epoch": 9, "memory": 6319, "step": 33936} +{"lr": 1.6141623611312105e-05, "data_time": 0.0010665655136108398, "loss": 0.21456100344657897, "time": 0.7007798433303833, "epoch": 9, "memory": 6319, "step": 34036} +{"lr": 1.5968538061036705e-05, "data_time": 0.000921940803527832, "loss": 0.21272813975811006, "time": 0.7707955837249756, "epoch": 9, "memory": 6319, "step": 34136} +{"lr": 1.5797753111105157e-05, "data_time": 0.0009696722030639648, "loss": 0.2143809661269188, "time": 0.745158338546753, "epoch": 9, "memory": 6319, "step": 34236} +{"lr": 1.5629278827958574e-05, "data_time": 0.001006746292114258, "loss": 0.22629391998052598, "time": 0.7561637163162231, "epoch": 9, "memory": 6319, "step": 34336} +{"lr": 1.5463125141842392e-05, "data_time": 0.0009626388549804687, "loss": 0.21649148613214492, "time": 0.7674808502197266, "epoch": 9, "memory": 6319, "step": 34436} +{"lr": 1.5299301846220796e-05, "data_time": 0.0010663509368896485, "loss": 0.20040907859802246, "time": 0.7117836952209473, "epoch": 9, "memory": 6319, "step": 34536} +{"lr": 1.5137818597199751e-05, "data_time": 0.0009500741958618164, "loss": 0.2169900044798851, "time": 0.7516935110092163, "epoch": 9, "memory": 6319, "step": 34636} +{"lr": 1.4978684912957622e-05, "data_time": 0.0009848356246948242, "loss": 0.20170099139213563, "time": 0.7448330879211426, "epoch": 9, "memory": 6319, "step": 34736} +{"lr": 1.4821910173184321e-05, "data_time": 0.0010041236877441407, "loss": 0.19362652748823167, "time": 0.7400836944580078, "epoch": 9, "memory": 6319, "step": 34836} +{"lr": 1.4667503618528337e-05, "data_time": 0.0009563922882080078, "loss": 0.21154155135154723, "time": 0.7105942964553833, "epoch": 9, "memory": 6319, "step": 34936} +{"lr": 1.4515474350052097e-05, "data_time": 0.0010356903076171875, "loss": 0.19550230354070663, "time": 0.7421216487884521, "epoch": 9, "memory": 6319, "step": 35036} +{"lr": 1.4365831328695616e-05, "data_time": 0.0009246349334716796, "loss": 0.21979222744703292, "time": 0.7416331529617309, "epoch": 9, "memory": 6319, "step": 35136} +{"lr": 1.4218583374748169e-05, "data_time": 0.0012225627899169922, "loss": 0.21451364159584047, "time": 0.8412116050720215, "epoch": 9, "memory": 6319, "step": 35236} +{"lr": 1.4073739167328542e-05, "data_time": 0.0009931325912475586, "loss": 0.20629584938287734, "time": 0.7455545425415039, "epoch": 9, "memory": 6319, "step": 35336} +{"lr": 1.3931307243873344e-05, "data_time": 0.0008781909942626953, "loss": 0.19987259954214096, "time": 0.8046121835708618, "epoch": 9, "memory": 6319, "step": 35436} +{"lr": 1.3791295999633911e-05, "data_time": 0.0007792472839355469, "loss": 0.2130444049835205, "time": 0.7733380556106567, "epoch": 9, "memory": 6319, "step": 35536} +{"lr": 1.3653713687181394e-05, "data_time": 0.0007951498031616211, "loss": 0.2052200198173523, "time": 0.7395313262939454, "epoch": 9, "memory": 6319, "step": 35636} +{"lr": 1.3518568415920337e-05, "data_time": 0.0009598016738891601, "loss": 0.20269481539726258, "time": 0.7832774877548218, "epoch": 9, "memory": 6319, "step": 35736} +{"lr": 1.3385868151610723e-05, "data_time": 0.0010293245315551758, "loss": 0.2250720202922821, "time": 0.8056441307067871, "epoch": 9, "memory": 6319, "step": 35836} +{"lr": 1.3255620715898442e-05, "data_time": 0.000977301597595215, "loss": 0.2165330559015274, "time": 0.7166641235351563, "epoch": 9, "memory": 6319, "step": 35936} +{"lr": 1.3127833785854261e-05, "data_time": 0.0010109424591064453, "loss": 0.21536608934402465, "time": 0.7907366752624512, "epoch": 9, "memory": 6319, "step": 36036} +{"lr": 1.3002514893521313e-05, "data_time": 0.0014429092407226562, "loss": 0.2203567698597908, "time": 0.7522102117538452, "epoch": 9, "memory": 6319, "step": 36136} +{"lr": 1.287967142547117e-05, "data_time": 0.0008806228637695312, "loss": 0.22083668708801268, "time": 0.7012571096420288, "epoch": 9, "memory": 6319, "step": 36236} +{"lr": 1.2759310622368435e-05, "data_time": 0.0009766817092895508, "loss": 0.22841697484254836, "time": 0.7599836826324463, "epoch": 9, "memory": 6319, "step": 36336} +{"lr": 1.2641439578543968e-05, "data_time": 0.0008215427398681641, "loss": 0.2243849068880081, "time": 0.7758415699005127, "epoch": 9, "memory": 6319, "step": 36436} +{"lr": 1.252606524157674e-05, "data_time": 0.000990581512451172, "loss": 0.2114819511771202, "time": 0.810365104675293, "epoch": 9, "memory": 6319, "step": 36536} +{"lr": 1.2413194411884297e-05, "data_time": 0.001035022735595703, "loss": 0.21351986080408097, "time": 0.7465147733688354, "epoch": 9, "memory": 6319, "step": 36636} +{"lr": 1.2302833742321981e-05, "data_time": 0.0010212898254394532, "loss": 0.21108139753341676, "time": 0.7470567941665649, "epoch": 9, "memory": 6319, "step": 36736} +{"accuracy/top1": 91.4784164428711, "data_time": 0.36692734162012736, "time": 0.45497892101605736, "step": 9} +{"lr": 1.2098001439139718e-05, "data_time": 0.16482934951782227, "loss": 0.21655374020338058, "time": 0.7675949096679687, "epoch": 10, "memory": 6319, "step": 36928} +{"lr": 1.1995007122352056e-05, "data_time": 0.0011049270629882812, "loss": 0.2215420961380005, "time": 0.7873828172683716, "epoch": 10, "memory": 6319, "step": 37028} +{"lr": 1.1894547614574522e-05, "data_time": 0.0009546041488647461, "loss": 0.20045895576477052, "time": 0.7119117736816406, "epoch": 10, "memory": 6319, "step": 37128} +{"lr": 1.1796628837111806e-05, "data_time": 0.000859379768371582, "loss": 0.21591064631938933, "time": 0.7177269220352173, "epoch": 10, "memory": 6319, "step": 37228} +{"lr": 1.1701256561512343e-05, "data_time": 0.001059722900390625, "loss": 0.20153772830963135, "time": 0.7311688899993897, "epoch": 10, "memory": 6319, "step": 37328} +{"lr": 1.1608436409228136e-05, "data_time": 0.001040172576904297, "loss": 0.2012565940618515, "time": 0.7108525991439819, "epoch": 10, "memory": 6319, "step": 37428} +{"lr": 1.1518173851283401e-05, "data_time": 0.0009455680847167969, "loss": 0.22018181681632995, "time": 0.7602965593338012, "epoch": 10, "memory": 6319, "step": 37528} +{"lr": 1.1430474207952176e-05, "data_time": 0.0010172605514526367, "loss": 0.2060492068529129, "time": 0.7416475772857666, "epoch": 10, "memory": 6319, "step": 37628} +{"lr": 1.1345342648444593e-05, "data_time": 0.0010094642639160156, "loss": 0.2178183004260063, "time": 0.792791199684143, "epoch": 10, "memory": 6319, "step": 37728} +{"lr": 1.12627841906023e-05, "data_time": 0.0010157346725463868, "loss": 0.2236344709992409, "time": 0.733181357383728, "epoch": 10, "memory": 6319, "step": 37828} +{"lr": 1.1182803700602659e-05, "data_time": 0.0012716054916381836, "loss": 0.1999393418431282, "time": 0.7501315832138061, "epoch": 10, "memory": 6319, "step": 37928} +{"lr": 1.110540589267192e-05, "data_time": 0.0010054349899291993, "loss": 0.20714259147644043, "time": 0.7648464918136597, "epoch": 10, "memory": 6319, "step": 38028} +{"lr": 1.1030595328807396e-05, "data_time": 0.001136612892150879, "loss": 0.21901369988918304, "time": 0.9440932035446167, "epoch": 10, "memory": 6319, "step": 38128} +{"lr": 1.0958376418508494e-05, "data_time": 0.0011468648910522461, "loss": 0.2044278934597969, "time": 0.7064155578613281, "epoch": 10, "memory": 6319, "step": 38228} +{"lr": 1.088875341851686e-05, "data_time": 0.0010543346405029296, "loss": 0.2198479488492012, "time": 0.7076034784317017, "epoch": 10, "memory": 6319, "step": 38328} +{"lr": 1.0821730432565486e-05, "data_time": 0.000953841209411621, "loss": 0.20958588868379593, "time": 0.7295926094055176, "epoch": 10, "memory": 6319, "step": 38428} +{"lr": 1.0757311411136797e-05, "data_time": 0.0008830070495605469, "loss": 0.20698675513267517, "time": 0.7685664653778076, "epoch": 10, "memory": 6319, "step": 38528} +{"lr": 1.0695500151229833e-05, "data_time": 0.0010093212127685546, "loss": 0.2094940572977066, "time": 0.7310352802276612, "epoch": 10, "memory": 6319, "step": 38628} +{"lr": 1.0636300296136373e-05, "data_time": 0.001578521728515625, "loss": 0.2008810892701149, "time": 0.7421096086502075, "epoch": 10, "memory": 6319, "step": 38728} +{"lr": 1.0579715335226322e-05, "data_time": 0.001484513282775879, "loss": 0.2024385690689087, "time": 0.7543192863464355, "epoch": 10, "memory": 6319, "step": 38828} +{"lr": 1.0525748603741896e-05, "data_time": 0.0013012170791625976, "loss": 0.19819287359714508, "time": 0.8326394319534302, "epoch": 10, "memory": 6319, "step": 38928} +{"lr": 1.0474403282601119e-05, "data_time": 0.0010962724685668946, "loss": 0.21920082122087478, "time": 1.1544252157211303, "epoch": 10, "memory": 6319, "step": 39028} +{"lr": 1.0425682398210336e-05, "data_time": 0.0009479522705078125, "loss": 0.1904679462313652, "time": 0.7190548181533813, "epoch": 10, "memory": 6319, "step": 39128} +{"lr": 1.0379588822285784e-05, "data_time": 0.0012860536575317384, "loss": 0.21891588419675828, "time": 0.795698881149292, "epoch": 10, "memory": 6319, "step": 39228} +{"lr": 1.0336125271684364e-05, "data_time": 0.0011662960052490235, "loss": 0.21366954743862152, "time": 0.7244150161743164, "epoch": 10, "memory": 6319, "step": 39328} +{"lr": 1.0295294308243494e-05, "data_time": 0.0010058164596557617, "loss": 0.19279357194900512, "time": 0.7590964555740356, "epoch": 10, "memory": 6319, "step": 39428} +{"lr": 1.0257098338630086e-05, "data_time": 0.0009763240814208984, "loss": 0.19902547746896743, "time": 0.773990249633789, "epoch": 10, "memory": 6319, "step": 39528} +{"lr": 1.0221539614198729e-05, "data_time": 0.0008453369140625, "loss": 0.20715045630931855, "time": 0.7676884889602661, "epoch": 10, "memory": 6319, "step": 39628} +{"lr": 1.0188620230858978e-05, "data_time": 0.0013334035873413086, "loss": 0.21932879239320754, "time": 0.7200077772140503, "epoch": 10, "memory": 6319, "step": 39728} +{"lr": 1.01583421289518e-05, "data_time": 0.0010752916336059571, "loss": 0.20916725993156432, "time": 0.7598609685897827, "epoch": 10, "memory": 6319, "step": 39828} +{"lr": 1.0130707093135206e-05, "data_time": 0.0011291265487670898, "loss": 0.21107010394334794, "time": 0.7480459928512573, "epoch": 10, "memory": 6319, "step": 39928} +{"lr": 1.01057167522791e-05, "data_time": 0.001167893409729004, "loss": 0.20092849880456926, "time": 0.772868013381958, "epoch": 10, "memory": 6319, "step": 40028} +{"lr": 1.008337257936921e-05, "data_time": 0.0010306835174560547, "loss": 0.1973315417766571, "time": 0.7603323221206665, "epoch": 10, "memory": 6319, "step": 40128} +{"lr": 1.0063675891420294e-05, "data_time": 0.0009876489639282227, "loss": 0.2090567946434021, "time": 0.7465966224670411, "epoch": 10, "memory": 6319, "step": 40228} +{"lr": 1.0046627849398548e-05, "data_time": 0.0008569002151489258, "loss": 0.22173888087272645, "time": 0.8362118482589722, "epoch": 10, "memory": 6319, "step": 40328} +{"lr": 1.0032229458153106e-05, "data_time": 0.0009721994400024414, "loss": 0.21461016237735747, "time": 0.7650289535522461, "epoch": 10, "memory": 6319, "step": 40428} +{"lr": 1.0020481566356862e-05, "data_time": 0.0018167734146118165, "loss": 0.2005124107003212, "time": 0.7336384534835816, "epoch": 10, "memory": 6319, "step": 40528} +{"lr": 1.0011384866456456e-05, "data_time": 0.001233386993408203, "loss": 0.1984155297279358, "time": 0.7361355304718018, "epoch": 10, "memory": 6319, "step": 40628} +{"lr": 1.00049398946314e-05, "data_time": 0.0008237361907958984, "loss": 0.21123445630073548, "time": 0.7409549236297608, "epoch": 10, "memory": 6319, "step": 40728} +{"lr": 1.0001147030762536e-05, "data_time": 0.00107574462890625, "loss": 0.21124005317687988, "time": 0.7571876049041748, "epoch": 10, "memory": 6319, "step": 40828} +{"accuracy/top1": 91.90043640136719, "data_time": 0.3827948351701101, "time": 0.4716785252094269, "step": 10} diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/0000000.png_1.png b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/0000000.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..b81d94a9485f6865bc76e2d9597a60b7113e5c7b Binary files /dev/null and b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/0000000.png_1.png differ diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/0000000.png_10.png b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/0000000.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..8565b9411f6db5381dab8e5bf91b4f76c4cb3da6 Binary files /dev/null and b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/0000000.png_10.png differ diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/0000000.png_2.png b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/0000000.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..2c19930899e10873e0a114a88972b1c969316ade Binary files /dev/null and b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/0000000.png_2.png differ diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/0000000.png_3.png b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/0000000.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..c2ba0d47b3af74c7d3b925b646084201bbe1059b Binary files /dev/null and b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/0000000.png_3.png differ diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/0000000.png_4.png b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/0000000.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..83c06ba5ce517e2d9af8c18cd8df703b18dec5e7 Binary files /dev/null and b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/0000000.png_4.png differ diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/0000000.png_5.png b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/0000000.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..23290a1fa20ab59c6ec3491e5e105936e5f28d97 Binary files /dev/null and b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/0000000.png_5.png differ diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/0000000.png_6.png b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/0000000.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..00a51363ba46c9e07f5d57801c6f8f419f56a22c Binary files /dev/null and b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/0000000.png_6.png differ diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/0000000.png_7.png b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/0000000.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..d0cddfe62957448c2fc40b8650540a1c7e396e9b Binary files /dev/null and b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/0000000.png_7.png differ diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/0000000.png_8.png b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/0000000.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..ff67b937b1fa50d9096286bde39f1041e315bd18 Binary files /dev/null and b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/0000000.png_8.png differ diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/0000000.png_9.png b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/0000000.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..aef4093b4978113c3a991f4ae8b6ecfe6a238fa3 Binary files /dev/null and b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/0000000.png_9.png differ diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/000003093.jpg_1.png b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/000003093.jpg_1.png new file mode 100644 index 0000000000000000000000000000000000000000..5d61a3ff615bd35bcfad7dd81aeeeda1a7b75684 Binary files /dev/null and b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/000003093.jpg_1.png differ diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/000003093.jpg_10.png b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/000003093.jpg_10.png new file mode 100644 index 0000000000000000000000000000000000000000..9b96b33f6573a7e4d5cef80359dbb9868df23469 Binary files /dev/null and b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/000003093.jpg_10.png differ diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/000003093.jpg_2.png b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/000003093.jpg_2.png new file mode 100644 index 0000000000000000000000000000000000000000..beefe92a5e35b895b6baf44abca51f6a53be7952 Binary files /dev/null and b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/000003093.jpg_2.png differ diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/000003093.jpg_3.png b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/000003093.jpg_3.png new file mode 100644 index 0000000000000000000000000000000000000000..48dcd1d66e8e6ae9c53e4770dae8afe852abc83c Binary files /dev/null and b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/000003093.jpg_3.png differ diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/000003093.jpg_4.png b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/000003093.jpg_4.png new file mode 100644 index 0000000000000000000000000000000000000000..dd291f0a8c395c1e1c69e26053cd16e534527a12 Binary files /dev/null and b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/000003093.jpg_4.png differ diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/000003093.jpg_5.png b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/000003093.jpg_5.png new file mode 100644 index 0000000000000000000000000000000000000000..166ad9cb82d29a6989b04bd5094ed7a2d0690608 Binary files /dev/null and b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/000003093.jpg_5.png differ diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/000003093.jpg_6.png b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/000003093.jpg_6.png new file mode 100644 index 0000000000000000000000000000000000000000..7cc4dac674f9b6d2c2a0677e0b4c9ef0747c88c7 Binary files /dev/null and b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/000003093.jpg_6.png differ diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/000003093.jpg_7.png b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/000003093.jpg_7.png new file mode 100644 index 0000000000000000000000000000000000000000..94eed1e9a27368902f1099f08e163df44df6f59a Binary files /dev/null and b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/000003093.jpg_7.png differ diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/000003093.jpg_8.png b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/000003093.jpg_8.png new file mode 100644 index 0000000000000000000000000000000000000000..6e4eaf9014def76a53232c549053e4f27395c092 Binary files /dev/null and b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/000003093.jpg_8.png differ diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/000003093.jpg_9.png b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/000003093.jpg_9.png new file mode 100644 index 0000000000000000000000000000000000000000..4a001c428dd65a16232d2b6dedd78160cc915eec Binary files /dev/null and b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/000003093.jpg_9.png differ diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/000015337.jpg_1.png b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/000015337.jpg_1.png new file mode 100644 index 0000000000000000000000000000000000000000..3d33287ba363f2be17cdbfb627c80602184ae478 Binary files /dev/null and b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/000015337.jpg_1.png differ diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/000015337.jpg_10.png b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/000015337.jpg_10.png new file mode 100644 index 0000000000000000000000000000000000000000..8018c77e3c34caf7448a230021a3f669967c44cb Binary files /dev/null and b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/000015337.jpg_10.png differ diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/000015337.jpg_2.png b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/000015337.jpg_2.png new file mode 100644 index 0000000000000000000000000000000000000000..7db5be22c25bedd5a32115e502cae666a884f571 Binary files /dev/null and b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/000015337.jpg_2.png differ diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/000015337.jpg_3.png b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/000015337.jpg_3.png new file mode 100644 index 0000000000000000000000000000000000000000..30052ac3265d5b7be7e2022d1120b943b3aa3dd3 Binary files /dev/null and b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/000015337.jpg_3.png differ diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/000015337.jpg_4.png b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/000015337.jpg_4.png new file mode 100644 index 0000000000000000000000000000000000000000..a5f909ebb3d98bbc38acba152db30946580a0b56 Binary files /dev/null and b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/000015337.jpg_4.png differ diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/000015337.jpg_5.png b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/000015337.jpg_5.png new file mode 100644 index 0000000000000000000000000000000000000000..58ae1f82cea275f6983ccc3542fd82d7a355d354 Binary files /dev/null and b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/000015337.jpg_5.png differ diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/000015337.jpg_6.png b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/000015337.jpg_6.png new file mode 100644 index 0000000000000000000000000000000000000000..463e40374c90ba7c45d1b01f9a48c9722a4c5244 Binary files /dev/null and b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/000015337.jpg_6.png differ diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/000015337.jpg_7.png b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/000015337.jpg_7.png new file mode 100644 index 0000000000000000000000000000000000000000..55ecaab80fda1ee9dd41369dfc19ffd278d39847 Binary files /dev/null and b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/000015337.jpg_7.png differ diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/000015337.jpg_8.png b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/000015337.jpg_8.png new file mode 100644 index 0000000000000000000000000000000000000000..8e88b8d5987479b8c1fcaf91389dc3ece62f1e16 Binary files /dev/null and b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/000015337.jpg_8.png differ diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/000015337.jpg_9.png b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/000015337.jpg_9.png new file mode 100644 index 0000000000000000000000000000000000000000..c67244ccb093ba9d2618ac97fadb08887db1f130 Binary files /dev/null and b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/000015337.jpg_9.png differ diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/0004159.png_1.png b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/0004159.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..bc30bf1c06ff707511901792073ae466af4368bb Binary files /dev/null and b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/0004159.png_1.png differ diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/0004159.png_10.png b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/0004159.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..4b217e116b891ec9e225c0ff0a2ba00f040fb64a Binary files /dev/null and b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/0004159.png_10.png differ diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/0004159.png_2.png b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/0004159.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..5371e7e8c2e17bf156f41d65b93c9a574c1b97ba Binary files /dev/null and b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/0004159.png_2.png differ diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/0004159.png_3.png b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/0004159.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..7ac09cd98c5bc1f8db0df7b3d02fc632c6d3680b Binary files /dev/null and b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/0004159.png_3.png differ diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/0004159.png_4.png b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/0004159.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..34a1f64acafcf75aadf7189c6ebeea58a4aca1f1 Binary files /dev/null and b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/0004159.png_4.png differ diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/0004159.png_5.png b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/0004159.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..1f7632d4772df0b0c20e991dd8c8df9f3bdc6702 Binary files /dev/null and b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/0004159.png_5.png differ diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/0004159.png_6.png b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/0004159.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..3bd83a04ce4cca2276cdd9391b5d6e67cef00fa2 Binary files /dev/null and b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/0004159.png_6.png differ diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/0004159.png_7.png b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/0004159.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..841d35895d1138ea245f6f486942d79c33700f36 Binary files /dev/null and b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/0004159.png_7.png differ diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/0004159.png_8.png b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/0004159.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..f045f351cf184e81e3b655bb7c3bbe017c3cbbe5 Binary files /dev/null and b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/0004159.png_8.png differ diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/0004159.png_9.png b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/0004159.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..36df7201271c40054e7e6a84174c835b09f306c3 Binary files /dev/null and b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/0004159.png_9.png differ diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/0008318.png_1.png b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/0008318.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..ad6d823468104fbac83f60826117f9a7ef892675 Binary files /dev/null and b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/0008318.png_1.png differ diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/0008318.png_10.png b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/0008318.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..36464c39cc404b231c49b4bd2ef7a1e3b0d18d73 Binary files /dev/null and b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/0008318.png_10.png differ diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/0008318.png_2.png b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/0008318.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..a6042d769ae701159d2ba674cad9d1d5dfd3d2e5 Binary files /dev/null and b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/0008318.png_2.png differ diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/0008318.png_3.png b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/0008318.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..0b3421826cb035183fd4a5e13627300077da042b Binary files /dev/null and b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/0008318.png_3.png differ diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/0008318.png_4.png b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/0008318.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..c1ee2ed2f460392ed3146fa07fdd496c3c32c20e Binary files /dev/null and b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/0008318.png_4.png differ diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/0008318.png_5.png b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/0008318.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..fe279fe5234d99c3b814a975f5490917ed0a6581 Binary files /dev/null and b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/0008318.png_5.png differ diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/0008318.png_6.png b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/0008318.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..b467d2e3d01e55197dfae74c83f5954f0d09661b Binary files /dev/null and b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/0008318.png_6.png differ diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/0008318.png_7.png b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/0008318.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..99444b937f69b4cfded63edac1ce1c36a3dd1556 Binary files /dev/null and b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/0008318.png_7.png differ diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/0008318.png_8.png b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/0008318.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..25bb1a0a4a545e3d71b5292fc7b7d1a8fc1ea287 Binary files /dev/null and b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/0008318.png_8.png differ diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/0008318.png_9.png b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/0008318.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..6b67cf373d24df81babcafcb17446695bb44265a Binary files /dev/null and b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/0008318.png_9.png differ diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/0010000.png_1.png b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/0010000.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..2e159f266f1540def49df9a4cc8932372dcc6537 Binary files /dev/null and b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/0010000.png_1.png differ diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/0010000.png_10.png b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/0010000.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..aa978eb0d526c956642511de86cc0f4580412504 Binary files /dev/null and b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/0010000.png_10.png differ diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/0010000.png_2.png b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/0010000.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..ca8c9902c72cf3c9a997c492f6cec1babf29b7b4 Binary files /dev/null and b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/0010000.png_2.png differ diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/0010000.png_3.png b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/0010000.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..81e146f4c4b346ed0056567c3bed87120e6389e9 Binary files /dev/null and b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/0010000.png_3.png differ diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/0010000.png_4.png b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/0010000.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..9eb8ee45ee8f45b7f162671249275626d8a9a7d8 Binary files /dev/null and b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/0010000.png_4.png differ diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/0010000.png_5.png b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/0010000.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..d16d095fe82a6869743765856b632cc0944bf7cd Binary files /dev/null and b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/0010000.png_5.png differ diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/0010000.png_6.png b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/0010000.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..119ca739d12b7c78e4b8b86ce6f0479d914ec222 Binary files /dev/null and b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/0010000.png_6.png differ diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/0010000.png_7.png b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/0010000.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..f93a3bf80678d9d66e9436eb56abadb39b1ac981 Binary files /dev/null and b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/0010000.png_7.png differ diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/0010000.png_8.png b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/0010000.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..800539d152622d7d626dfa5b81b3adf3ab7cfa69 Binary files /dev/null and b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/0010000.png_8.png differ diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/0010000.png_9.png b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/0010000.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..dca15b847deafe514a637b2c8d6b092482f7fc05 Binary files /dev/null and b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/0010000.png_9.png differ diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/0014159.png_1.png b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/0014159.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..3bf41f41cbe6be10ceff451978f39d138518eeb7 Binary files /dev/null and b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/0014159.png_1.png differ diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/0014159.png_10.png b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/0014159.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..ea808c681f0146d4349398799d501026bcc3eca6 Binary files /dev/null and b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/0014159.png_10.png differ diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/0014159.png_2.png b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/0014159.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..c889bd68b1e422404582d2ccca286ae00fef5d00 Binary files /dev/null and b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/0014159.png_2.png differ diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/0014159.png_3.png b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/0014159.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..feb4423cfb9584ce09e9d6b1b815f901f911164f Binary files /dev/null and b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/0014159.png_3.png differ diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/0014159.png_4.png b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/0014159.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..6279b62b6d7e470c3d6c6e0db50b57e33137ccd5 Binary files /dev/null and b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/0014159.png_4.png differ diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/0014159.png_5.png b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/0014159.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..4332d4a7e805f250b598f05156af81df7a30cac5 Binary files /dev/null and b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/0014159.png_5.png differ diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/0014159.png_6.png b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/0014159.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..00c3f20e3237bd211a27ea177a7ec4713ad488d7 Binary files /dev/null and b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/0014159.png_6.png differ diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/0014159.png_7.png b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/0014159.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..e26c58f242bb7a5ad515b74966c1c913c43ea540 Binary files /dev/null and b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/0014159.png_7.png differ diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/0014159.png_8.png b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/0014159.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..93792c06f7b591efba088794df76539eb257f01e Binary files /dev/null and b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/0014159.png_8.png differ diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/0014159.png_9.png b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/0014159.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..cb5547c2486825bab55ad0e38597555054806f94 Binary files /dev/null and b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/20230606_005813/vis_data/vis_image/0014159.png_9.png differ diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/epoch_1.pth b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/epoch_1.pth new file mode 100644 index 0000000000000000000000000000000000000000..812949b1103d3079aed25bc200b150974f7983fd --- /dev/null +++ b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/epoch_1.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:e48db027eb713bf1a9f2c324a03b541c419036829d31bdb3524df18e036ebf66 +size 277355661 diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/epoch_10.pth b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/epoch_10.pth new file mode 100644 index 0000000000000000000000000000000000000000..09366d06b3e53a0d69b2c6d594b10ea917cc40ef --- /dev/null +++ b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/epoch_10.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:25ebd90720131db828268e3bda2c0582ed75f4f2083cbf4330e32a92e3f6dc6e +size 280066701 diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/epoch_2.pth b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/epoch_2.pth new file mode 100644 index 0000000000000000000000000000000000000000..630edd8b2aa5d35dba62d57de2a9e77bff9a6975 --- /dev/null +++ b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/epoch_2.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c0f562aad4a839072c656a8980de75c430faafcfdff72785a451f4e12e221aec +size 277656269 diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/epoch_3.pth b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/epoch_3.pth new file mode 100644 index 0000000000000000000000000000000000000000..9720d5fde26eb23bc57bbf1889d291f152ecbf16 --- /dev/null +++ b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/epoch_3.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:16e9a11c670c27ce07e2813992eb07a8868656f3df4bf562e585297a31e1dc7f +size 277956301 diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/epoch_4.pth b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/epoch_4.pth new file mode 100644 index 0000000000000000000000000000000000000000..168d35274d7844094bac677f4dfb6dfaec32c403 --- /dev/null +++ b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/epoch_4.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:cd4f960704338e161a69a5b93af1e3a585712ad73f927b2b64aff1474796c059 +size 278255821 diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/epoch_5.pth b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/epoch_5.pth new file mode 100644 index 0000000000000000000000000000000000000000..4387934df2afaa8c9b8bd57c9b3fd1f5c1fd291e --- /dev/null +++ b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/epoch_5.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:923215cd5f829e18ab305a2e8da82ed7b66dfa130423efd5d48d8d5df9a39545 +size 278555533 diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/epoch_6.pth b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/epoch_6.pth new file mode 100644 index 0000000000000000000000000000000000000000..f9bea9d91f56aa7ac69ae2aeee83a17d035262d1 --- /dev/null +++ b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/epoch_6.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:b80692d24c063f4172d766f17dd890aa99429e1f6421b0480565bf2ae26efa9f +size 278855245 diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/epoch_7.pth b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/epoch_7.pth new file mode 100644 index 0000000000000000000000000000000000000000..4382bed17cba16f82f6b6ad65403fc14737aac94 --- /dev/null +++ b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/epoch_7.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:47707df2ac5c9861eddf8a21dca8154f517ffaa9fd28db08aaa365bb75a94b0d +size 279155469 diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/epoch_8.pth b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/epoch_8.pth new file mode 100644 index 0000000000000000000000000000000000000000..7416a240040261f1cc95e4e2f32b258958b33866 --- /dev/null +++ b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/epoch_8.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:259665aea62bf49838dba6256673046ce015a20ecc06c9141e65ad79437345db +size 279459469 diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/epoch_9.pth b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/epoch_9.pth new file mode 100644 index 0000000000000000000000000000000000000000..f475af26751b9ea0e5ff23547fb2bb9df70020d0 --- /dev/null +++ b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/epoch_9.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:2bcc9c7b5eac5dc989f2a710a7d0a50c9c0e8bb10042df6f606eb0fafc0797d1 +size 279763341 diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/last_checkpoint b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/last_checkpoint new file mode 100644 index 0000000000000000000000000000000000000000..54536066bf2fbfbff01673798e62d3f2adb6f1b3 --- /dev/null +++ b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/last_checkpoint @@ -0,0 +1 @@ +/mnt/petrelfs/luzeyu/workspace/fakebench/mmpretrain/workdir/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/epoch_10.pth \ No newline at end of file diff --git a/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1.py b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1.py new file mode 100644 index 0000000000000000000000000000000000000000..7a0f9f2c3060ac00c4c97a418905c91572dba3b1 --- /dev/null +++ b/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1.py @@ -0,0 +1,349 @@ +optim_wrapper = dict( + optimizer=dict( + type='SGD', + lr=0.0001, + momentum=0.9, + weight_decay=0.0001, + _scope_='mmpretrain'), + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=False, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=512) +model = dict( + type='ImageClassifier', + backbone=dict( + frozen_stages=2, + type='ResNet', + depth=50, + num_stages=4, + out_indices=(3, ), + style='pytorch', + init_cfg=dict( + type='Pretrained', + checkpoint= + 'https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth', + prefix='backbone')), + neck=dict(type='GlobalAveragePooling'), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=2048, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + topk=1)) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/all_0.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/all_1.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3fake8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/cc1m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3real8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV2-1-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/if-dpmsolver++-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = dict(type='Accuracy', topk=1) +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV2-1-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/if-dpmsolver++-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = dict(type='Accuracy', topk=1) +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/resnet50_2xb256_all2_1m_lr1e-4_aug_5e-1' diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/20230605_042304.log b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/20230605_042304.log new file mode 100644 index 0000000000000000000000000000000000000000..6d67c65247d853188f592e57801cc59d869f1dde --- /dev/null +++ b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/20230605_042304.log @@ -0,0 +1,1473 @@ +2023/06/05 04:23:08 - mmengine - INFO - +------------------------------------------------------------ +System environment: + sys.platform: linux + Python: 3.10.9 (main, Mar 8 2023, 10:47:38) [GCC 11.2.0] + CUDA available: True + numpy_random_seed: 836701732 + GPU 0,1: NVIDIA A100-SXM4-80GB + CUDA_HOME: /mnt/petrelfs/share/cuda-11.6 + NVCC: Cuda compilation tools, release 11.6, V11.6.124 + GCC: gcc (GCC) 7.5.0 + PyTorch: 1.13.1 + PyTorch compiling details: PyTorch built with: + - GCC 9.3 + - C++ Version: 201402 + - Intel(R) oneAPI Math Kernel Library Version 2021.4-Product Build 20210904 for Intel(R) 64 architecture applications + - Intel(R) MKL-DNN v2.6.0 (Git Hash 52b5f107dd9cf10910aaa19cb47f3abf9b349815) + - OpenMP 201511 (a.k.a. OpenMP 4.5) + - LAPACK is enabled (usually provided by MKL) + - NNPACK is enabled + - CPU capability usage: AVX2 + - CUDA Runtime 11.6 + - NVCC architecture flags: -gencode;arch=compute_37,code=sm_37;-gencode;arch=compute_50,code=sm_50;-gencode;arch=compute_60,code=sm_60;-gencode;arch=compute_61,code=sm_61;-gencode;arch=compute_70,code=sm_70;-gencode;arch=compute_75,code=sm_75;-gencode;arch=compute_80,code=sm_80;-gencode;arch=compute_86,code=sm_86;-gencode;arch=compute_37,code=compute_37 + - CuDNN 8.3.2 (built against CUDA 11.5) + - Magma 2.6.1 + - Build settings: BLAS_INFO=mkl, BUILD_TYPE=Release, CUDA_VERSION=11.6, CUDNN_VERSION=8.3.2, CXX_COMPILER=/opt/rh/devtoolset-9/root/usr/bin/c++, CXX_FLAGS= -fabi-version=11 -Wno-deprecated -fvisibility-inlines-hidden -DUSE_PTHREADPOOL -fopenmp -DNDEBUG -DUSE_KINETO -DUSE_FBGEMM -DUSE_QNNPACK -DUSE_PYTORCH_QNNPACK -DUSE_XNNPACK -DSYMBOLICATE_MOBILE_DEBUG_HANDLE -DEDGE_PROFILER_USE_KINETO -O2 -fPIC -Wno-narrowing -Wall -Wextra -Werror=return-type -Werror=non-virtual-dtor -Wno-missing-field-initializers -Wno-type-limits -Wno-array-bounds -Wno-unknown-pragmas -Wunused-local-typedefs -Wno-unused-parameter -Wno-unused-function -Wno-unused-result -Wno-strict-overflow -Wno-strict-aliasing -Wno-error=deprecated-declarations -Wno-stringop-overflow -Wno-psabi -Wno-error=pedantic -Wno-error=redundant-decls -Wno-error=old-style-cast -fdiagnostics-color=always -faligned-new -Wno-unused-but-set-variable -Wno-maybe-uninitialized -fno-math-errno -fno-trapping-math -Werror=format -Werror=cast-function-type -Wno-stringop-overflow, LAPACK_INFO=mkl, PERF_WITH_AVX=1, PERF_WITH_AVX2=1, PERF_WITH_AVX512=1, TORCH_VERSION=1.13.1, USE_CUDA=ON, USE_CUDNN=ON, USE_EXCEPTION_PTR=1, USE_GFLAGS=OFF, USE_GLOG=OFF, USE_MKL=ON, USE_MKLDNN=ON, USE_MPI=OFF, USE_NCCL=ON, USE_NNPACK=ON, USE_OPENMP=ON, USE_ROCM=OFF, + + TorchVision: 0.14.1 + OpenCV: 4.7.0 + MMEngine: 0.7.3 + +Runtime environment: + cudnn_benchmark: True + mp_cfg: {'mp_start_method': 'fork', 'opencv_num_threads': 0} + dist_cfg: {'backend': 'nccl'} + seed: None + deterministic: False + Distributed launcher: slurm + Distributed training: True + GPU number: 2 +------------------------------------------------------------ + +2023/06/05 04:23:12 - mmengine - INFO - Config: +optim_wrapper = dict( + optimizer=dict( + type='SGD', + lr=0.0001, + momentum=0.9, + weight_decay=0.0001, + _scope_='mmpretrain'), + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=False, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=512) +model = dict( + type='ImageClassifier', + backbone=dict( + frozen_stages=2, + type='ResNet', + depth=50, + num_stages=4, + out_indices=(3, ), + style='pytorch', + init_cfg=dict( + type='Pretrained', + checkpoint= + 'https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth', + prefix='backbone')), + neck=dict(type='GlobalAveragePooling'), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=2048, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + topk=1)) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/all1.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/all2.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/all3.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3fake8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/cc1m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3real8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV2-1-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/if-dpmsolver++-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = dict(type='Accuracy', topk=1) +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV2-1-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/if-dpmsolver++-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = dict(type='Accuracy', topk=1) +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1' + +2023/06/05 04:23:25 - mmengine - INFO - Hooks will be executed in the following order: +before_run: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_load_checkpoint: +(ABOVE_NORMAL) EMAHook + -------------------- +before_train: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_train_epoch: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(NORMAL ) DistSamplerSeedHook + -------------------- +before_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook + -------------------- +after_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +after_train_epoch: +(NORMAL ) IterTimerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_val_epoch: +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook + -------------------- +before_val_iter: +(NORMAL ) IterTimerHook + -------------------- +after_val_iter: +(NORMAL ) IterTimerHook +(NORMAL ) VisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_val_epoch: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_save_checkpoint: +(ABOVE_NORMAL) EMAHook + -------------------- +after_train: +(VERY_LOW ) CheckpointHook + -------------------- +before_test_epoch: +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook + -------------------- +before_test_iter: +(NORMAL ) IterTimerHook + -------------------- +after_test_iter: +(NORMAL ) IterTimerHook +(NORMAL ) VisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_test_epoch: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_run: +(BELOW_NORMAL) LoggerHook + -------------------- +2023/06/05 04:23:47 - mmengine - INFO - load backbone in model from: https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth +Name of parameter - Initialization information + +backbone.conv1.weight - torch.Size([64, 3, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.bn1.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.bn1.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.conv1.weight - torch.Size([64, 64, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.bn1.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.bn1.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.conv2.weight - torch.Size([64, 64, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.bn2.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.bn2.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.conv3.weight - torch.Size([256, 64, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.bn3.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.bn3.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.downsample.0.weight - torch.Size([256, 64, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.downsample.1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.downsample.1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.conv1.weight - torch.Size([64, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.bn1.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.bn1.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.conv2.weight - torch.Size([64, 64, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.bn2.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.bn2.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.conv3.weight - torch.Size([256, 64, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.bn3.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.bn3.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.conv1.weight - torch.Size([64, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.bn1.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.bn1.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.conv2.weight - torch.Size([64, 64, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.bn2.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.bn2.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.conv3.weight - torch.Size([256, 64, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.bn3.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.bn3.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.conv1.weight - torch.Size([128, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.bn1.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.bn1.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.conv2.weight - torch.Size([128, 128, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.bn2.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.bn2.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.conv3.weight - torch.Size([512, 128, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.bn3.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.bn3.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.downsample.0.weight - torch.Size([512, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.downsample.1.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.downsample.1.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.conv1.weight - torch.Size([128, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.bn1.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.bn1.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.conv2.weight - torch.Size([128, 128, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.bn2.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.bn2.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.conv3.weight - torch.Size([512, 128, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.bn3.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.bn3.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.conv1.weight - torch.Size([128, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.bn1.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.bn1.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.conv2.weight - torch.Size([128, 128, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.bn2.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.bn2.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.conv3.weight - torch.Size([512, 128, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.bn3.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.bn3.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.conv1.weight - torch.Size([128, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.bn1.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.bn1.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.conv2.weight - torch.Size([128, 128, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.bn2.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.bn2.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.conv3.weight - torch.Size([512, 128, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.bn3.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.bn3.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.conv1.weight - torch.Size([256, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.bn1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.bn1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.conv2.weight - torch.Size([256, 256, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.bn2.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.bn2.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.conv3.weight - torch.Size([1024, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.bn3.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.bn3.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.downsample.0.weight - torch.Size([1024, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.downsample.1.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.downsample.1.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.conv1.weight - torch.Size([256, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.bn1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.bn1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.conv2.weight - torch.Size([256, 256, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.bn2.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.bn2.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.conv3.weight - torch.Size([1024, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.bn3.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.bn3.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.conv1.weight - torch.Size([256, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.bn1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.bn1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.conv2.weight - torch.Size([256, 256, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.bn2.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.bn2.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.conv3.weight - torch.Size([1024, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.bn3.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.bn3.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.conv1.weight - torch.Size([256, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.bn1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.bn1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.conv2.weight - torch.Size([256, 256, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.bn2.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.bn2.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.conv3.weight - torch.Size([1024, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.bn3.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.bn3.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.conv1.weight - torch.Size([256, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.bn1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.bn1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.conv2.weight - torch.Size([256, 256, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.bn2.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.bn2.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.conv3.weight - torch.Size([1024, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.bn3.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.bn3.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.conv1.weight - torch.Size([256, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.bn1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.bn1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.conv2.weight - torch.Size([256, 256, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.bn2.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.bn2.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.conv3.weight - torch.Size([1024, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.bn3.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.bn3.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.conv1.weight - torch.Size([512, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.bn1.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.bn1.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.conv2.weight - torch.Size([512, 512, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.bn2.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.bn2.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.conv3.weight - torch.Size([2048, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.bn3.weight - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.bn3.bias - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.downsample.0.weight - torch.Size([2048, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.downsample.1.weight - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.downsample.1.bias - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.conv1.weight - torch.Size([512, 2048, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.bn1.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.bn1.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.conv2.weight - torch.Size([512, 512, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.bn2.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.bn2.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.conv3.weight - torch.Size([2048, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.bn3.weight - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.bn3.bias - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.conv1.weight - torch.Size([512, 2048, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.bn1.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.bn1.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.conv2.weight - torch.Size([512, 512, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.bn2.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.bn2.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.conv3.weight - torch.Size([2048, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.bn3.weight - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.bn3.bias - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +head.fc.weight - torch.Size([2, 2048]): +NormalInit: mean=0, std=0.01, bias=0 + +head.fc.bias - torch.Size([2]): +NormalInit: mean=0, std=0.01, bias=0 +2023/06/05 04:23:47 - mmengine - WARNING - "FileClient" will be deprecated in future. Please use io functions in https://mmengine.readthedocs.io/en/latest/api/fileio.html#file-io +2023/06/05 04:23:47 - mmengine - WARNING - "HardDiskBackend" is the alias of "LocalBackend" and the former will be deprecated in future. +2023/06/05 04:23:47 - mmengine - INFO - Checkpoints will be saved to /mnt/petrelfs/luzeyu/workspace/fakebench/mmpretrain/workdir/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1. +2023/06/05 04:25:00 - mmengine - INFO - Epoch(train) [1][ 100/4111] lr: 9.9999e-05 eta: 8:13:35 time: 0.7016 data_time: 0.5597 memory: 9436 loss: 0.6434 +2023/06/05 04:27:47 - mmengine - INFO - Epoch(train) [1][ 200/4111] lr: 9.9995e-05 eta: 13:37:18 time: 1.2976 data_time: 1.1454 memory: 6319 loss: 0.5886 +2023/06/05 04:28:54 - mmengine - INFO - Epoch(train) [1][ 300/4111] lr: 9.9988e-05 eta: 11:35:03 time: 0.6639 data_time: 0.5219 memory: 6319 loss: 0.5549 +2023/06/05 04:30:02 - mmengine - INFO - Epoch(train) [1][ 400/4111] lr: 9.9979e-05 eta: 10:35:08 time: 0.6636 data_time: 0.5151 memory: 6319 loss: 0.5157 +2023/06/05 04:31:08 - mmengine - INFO - Epoch(train) [1][ 500/4111] lr: 9.9967e-05 eta: 9:57:03 time: 0.6272 data_time: 0.4860 memory: 6319 loss: 0.4935 +2023/06/05 04:32:16 - mmengine - INFO - Epoch(train) [1][ 600/4111] lr: 9.9953e-05 eta: 9:32:17 time: 0.6772 data_time: 0.5247 memory: 6319 loss: 0.4670 +2023/06/05 04:33:26 - mmengine - INFO - Epoch(train) [1][ 700/4111] lr: 9.9936e-05 eta: 9:16:30 time: 0.6669 data_time: 0.5264 memory: 6319 loss: 0.4350 +2023/06/05 04:34:36 - mmengine - INFO - Epoch(train) [1][ 800/4111] lr: 9.9916e-05 eta: 9:04:34 time: 0.7353 data_time: 0.5882 memory: 6319 loss: 0.4327 +2023/06/05 04:35:42 - mmengine - INFO - Epoch(train) [1][ 900/4111] lr: 9.9894e-05 eta: 8:52:04 time: 0.6301 data_time: 0.4891 memory: 6319 loss: 0.4018 +2023/06/05 04:36:57 - mmengine - INFO - Exp name: resnet50_2xb256_all_1m_lr1e-4_aug_1e-1_20230605_042304 +2023/06/05 04:36:57 - mmengine - INFO - Epoch(train) [1][1000/4111] lr: 9.9869e-05 eta: 8:47:34 time: 0.6469 data_time: 0.5072 memory: 6319 loss: 0.4035 +2023/06/05 04:38:02 - mmengine - INFO - Epoch(train) [1][1100/4111] lr: 9.9841e-05 eta: 8:38:21 time: 0.6600 data_time: 0.5214 memory: 6319 loss: 0.3875 +2023/06/05 04:39:13 - mmengine - INFO - Epoch(train) [1][1200/4111] lr: 9.9811e-05 eta: 8:32:51 time: 0.6956 data_time: 0.5416 memory: 6319 loss: 0.3899 +2023/06/05 04:40:20 - mmengine - INFO - Epoch(train) [1][1300/4111] lr: 9.9778e-05 eta: 8:26:39 time: 0.6730 data_time: 0.5328 memory: 6319 loss: 0.3806 +2023/06/05 04:41:39 - mmengine - INFO - Epoch(train) [1][1400/4111] lr: 9.9743e-05 eta: 8:26:45 time: 0.6127 data_time: 0.4738 memory: 6319 loss: 0.3726 +2023/06/05 04:42:50 - mmengine - INFO - Epoch(train) [1][1500/4111] lr: 9.9705e-05 eta: 8:22:46 time: 0.6967 data_time: 0.5568 memory: 6319 loss: 0.3778 +2023/06/05 04:44:05 - mmengine - INFO - Epoch(train) [1][1600/4111] lr: 9.9664e-05 eta: 8:21:04 time: 0.6558 data_time: 0.5161 memory: 6319 loss: 0.3588 +2023/06/05 04:45:14 - mmengine - INFO - Epoch(train) [1][1700/4111] lr: 9.9621e-05 eta: 8:17:06 time: 0.7394 data_time: 0.5993 memory: 6319 loss: 0.3490 +2023/06/05 04:46:33 - mmengine - INFO - Epoch(train) [1][1800/4111] lr: 9.9575e-05 eta: 8:17:07 time: 0.6608 data_time: 0.5212 memory: 6319 loss: 0.3332 +2023/06/05 04:47:43 - mmengine - INFO - Epoch(train) [1][1900/4111] lr: 9.9527e-05 eta: 8:13:51 time: 0.6402 data_time: 0.5000 memory: 6319 loss: 0.3416 +2023/06/05 04:48:52 - mmengine - INFO - Exp name: resnet50_2xb256_all_1m_lr1e-4_aug_1e-1_20230605_042304 +2023/06/05 04:48:52 - mmengine - INFO - Epoch(train) [1][2000/4111] lr: 9.9476e-05 eta: 8:10:16 time: 0.6580 data_time: 0.5169 memory: 6319 loss: 0.3345 +2023/06/05 04:49:59 - mmengine - INFO - Epoch(train) [1][2100/4111] lr: 9.9422e-05 eta: 8:06:32 time: 0.6466 data_time: 0.5055 memory: 6319 loss: 0.3519 +2023/06/05 04:51:07 - mmengine - INFO - Epoch(train) [1][2200/4111] lr: 9.9366e-05 eta: 8:03:23 time: 0.6427 data_time: 0.5036 memory: 6319 loss: 0.3288 +2023/06/05 04:52:16 - mmengine - INFO - Epoch(train) [1][2300/4111] lr: 9.9307e-05 eta: 8:00:27 time: 0.7374 data_time: 0.5937 memory: 6319 loss: 0.3231 +2023/06/05 04:53:25 - mmengine - INFO - Epoch(train) [1][2400/4111] lr: 9.9246e-05 eta: 7:57:42 time: 0.6693 data_time: 0.5299 memory: 6319 loss: 0.3177 +2023/06/05 04:54:34 - mmengine - INFO - Epoch(train) [1][2500/4111] lr: 9.9182e-05 eta: 7:55:23 time: 0.7474 data_time: 0.6086 memory: 6319 loss: 0.3123 +2023/06/05 04:55:44 - mmengine - INFO - Epoch(train) [1][2600/4111] lr: 9.9115e-05 eta: 7:53:13 time: 0.6542 data_time: 0.5145 memory: 6319 loss: 0.3173 +2023/06/05 04:56:52 - mmengine - INFO - Epoch(train) [1][2700/4111] lr: 9.9046e-05 eta: 7:50:38 time: 0.7132 data_time: 0.5705 memory: 6319 loss: 0.3289 +2023/06/05 04:58:01 - mmengine - INFO - Epoch(train) [1][2800/4111] lr: 9.8974e-05 eta: 7:48:14 time: 0.6319 data_time: 0.4922 memory: 6319 loss: 0.3010 +2023/06/05 04:59:08 - mmengine - INFO - Epoch(train) [1][2900/4111] lr: 9.8900e-05 eta: 7:45:43 time: 0.6673 data_time: 0.5272 memory: 6319 loss: 0.3105 +2023/06/05 05:00:16 - mmengine - INFO - Exp name: resnet50_2xb256_all_1m_lr1e-4_aug_1e-1_20230605_042304 +2023/06/05 05:00:16 - mmengine - INFO - Epoch(train) [1][3000/4111] lr: 9.8823e-05 eta: 7:43:23 time: 0.6718 data_time: 0.5320 memory: 6319 loss: 0.3096 +2023/06/05 05:01:28 - mmengine - INFO - Epoch(train) [1][3100/4111] lr: 9.8744e-05 eta: 7:42:00 time: 0.8636 data_time: 0.7241 memory: 6319 loss: 0.3106 +2023/06/05 05:02:44 - mmengine - INFO - Epoch(train) [1][3200/4111] lr: 9.8662e-05 eta: 7:41:20 time: 0.6952 data_time: 0.5562 memory: 6319 loss: 0.3068 +2023/06/05 05:03:53 - mmengine - INFO - Epoch(train) [1][3300/4111] lr: 9.8578e-05 eta: 7:39:19 time: 0.6560 data_time: 0.5106 memory: 6319 loss: 0.3132 +2023/06/05 05:05:03 - mmengine - INFO - Epoch(train) [1][3400/4111] lr: 9.8490e-05 eta: 7:37:31 time: 0.6594 data_time: 0.5197 memory: 6319 loss: 0.2942 +2023/06/05 05:06:09 - mmengine - INFO - Epoch(train) [1][3500/4111] lr: 9.8401e-05 eta: 7:35:15 time: 0.7079 data_time: 0.5684 memory: 6319 loss: 0.3102 +2023/06/05 05:07:17 - mmengine - INFO - Epoch(train) [1][3600/4111] lr: 9.8309e-05 eta: 7:33:14 time: 0.6479 data_time: 0.5084 memory: 6319 loss: 0.2963 +2023/06/05 05:08:24 - mmengine - INFO - Epoch(train) [1][3700/4111] lr: 9.8214e-05 eta: 7:31:00 time: 0.7166 data_time: 0.5766 memory: 6319 loss: 0.3107 +2023/06/05 05:09:30 - mmengine - INFO - Epoch(train) [1][3800/4111] lr: 9.8117e-05 eta: 7:28:48 time: 0.6658 data_time: 0.5262 memory: 6319 loss: 0.2898 +2023/06/05 05:10:40 - mmengine - INFO - Epoch(train) [1][3900/4111] lr: 9.8017e-05 eta: 7:27:17 time: 0.6627 data_time: 0.5220 memory: 6319 loss: 0.2886 +2023/06/05 05:11:48 - mmengine - INFO - Exp name: resnet50_2xb256_all_1m_lr1e-4_aug_1e-1_20230605_042304 +2023/06/05 05:11:48 - mmengine - INFO - Epoch(train) [1][4000/4111] lr: 9.7915e-05 eta: 7:25:23 time: 0.8035 data_time: 0.6571 memory: 6319 loss: 0.2782 +2023/06/05 05:12:54 - mmengine - INFO - Epoch(train) [1][4100/4111] lr: 9.7810e-05 eta: 7:23:18 time: 0.6754 data_time: 0.5351 memory: 6319 loss: 0.2950 +2023/06/05 05:13:02 - mmengine - INFO - Exp name: resnet50_2xb256_all_1m_lr1e-4_aug_1e-1_20230605_042304 +2023/06/05 05:13:02 - mmengine - INFO - Saving checkpoint at 1 epochs +2023/06/05 05:13:45 - mmengine - INFO - Epoch(val) [1][100/119] eta: 0:00:07 time: 0.6731 data_time: 0.5845 memory: 8106 +2023/06/05 05:14:10 - mmengine - INFO - Epoch(val) [1][119/119] accuracy/top1: 80.6084 data_time: 0.3415 time: 0.4320 +2023/06/05 05:15:40 - mmengine - INFO - Epoch(train) [2][ 100/4111] lr: 9.7691e-05 eta: 7:24:33 time: 0.7329 data_time: 0.5789 memory: 6318 loss: 0.2884 +2023/06/05 05:16:53 - mmengine - INFO - Epoch(train) [2][ 200/4111] lr: 9.7581e-05 eta: 7:23:22 time: 0.7605 data_time: 0.6203 memory: 6318 loss: 0.2871 +2023/06/05 05:18:02 - mmengine - INFO - Epoch(train) [2][ 300/4111] lr: 9.7469e-05 eta: 7:21:42 time: 0.7376 data_time: 0.5982 memory: 6318 loss: 0.2807 +2023/06/05 05:19:10 - mmengine - INFO - Epoch(train) [2][ 400/4111] lr: 9.7354e-05 eta: 7:19:59 time: 0.6557 data_time: 0.5157 memory: 6318 loss: 0.2862 +2023/06/05 05:20:18 - mmengine - INFO - Epoch(train) [2][ 500/4111] lr: 9.7236e-05 eta: 7:18:12 time: 0.6786 data_time: 0.5338 memory: 6318 loss: 0.2887 +2023/06/05 05:21:23 - mmengine - INFO - Epoch(train) [2][ 600/4111] lr: 9.7116e-05 eta: 7:16:09 time: 0.6572 data_time: 0.5173 memory: 6318 loss: 0.2789 +2023/06/05 05:22:30 - mmengine - INFO - Epoch(train) [2][ 700/4111] lr: 9.6994e-05 eta: 7:14:18 time: 0.6556 data_time: 0.5163 memory: 6318 loss: 0.2543 +2023/06/05 05:23:39 - mmengine - INFO - Epoch(train) [2][ 800/4111] lr: 9.6869e-05 eta: 7:12:49 time: 0.6456 data_time: 0.5058 memory: 6318 loss: 0.2657 +2023/06/05 05:24:40 - mmengine - INFO - Exp name: resnet50_2xb256_all_1m_lr1e-4_aug_1e-1_20230605_042304 +2023/06/05 05:24:46 - mmengine - INFO - Epoch(train) [2][ 900/4111] lr: 9.6742e-05 eta: 7:11:03 time: 0.6640 data_time: 0.5169 memory: 6318 loss: 0.2808 +2023/06/05 05:25:55 - mmengine - INFO - Epoch(train) [2][1000/4111] lr: 9.6612e-05 eta: 7:09:30 time: 0.7165 data_time: 0.5766 memory: 6318 loss: 0.2722 +2023/06/05 05:27:05 - mmengine - INFO - Epoch(train) [2][1100/4111] lr: 9.6480e-05 eta: 7:08:10 time: 0.6846 data_time: 0.5435 memory: 6318 loss: 0.2801 +2023/06/05 05:28:14 - mmengine - INFO - Epoch(train) [2][1200/4111] lr: 9.6346e-05 eta: 7:06:41 time: 0.7721 data_time: 0.6312 memory: 6318 loss: 0.2790 +2023/06/05 05:29:26 - mmengine - INFO - Epoch(train) [2][1300/4111] lr: 9.6209e-05 eta: 7:05:30 time: 0.7352 data_time: 0.5963 memory: 6318 loss: 0.2939 +2023/06/05 05:30:38 - mmengine - INFO - Epoch(train) [2][1400/4111] lr: 9.6069e-05 eta: 7:04:24 time: 0.7055 data_time: 0.5664 memory: 6318 loss: 0.2679 +2023/06/05 05:31:46 - mmengine - INFO - Epoch(train) [2][1500/4111] lr: 9.5928e-05 eta: 7:02:49 time: 0.6749 data_time: 0.5336 memory: 6318 loss: 0.2764 +2023/06/05 05:32:53 - mmengine - INFO - Epoch(train) [2][1600/4111] lr: 9.5783e-05 eta: 7:01:08 time: 0.6455 data_time: 0.5049 memory: 6318 loss: 0.2670 +2023/06/05 05:34:00 - mmengine - INFO - Epoch(train) [2][1700/4111] lr: 9.5637e-05 eta: 6:59:31 time: 0.6840 data_time: 0.5443 memory: 6318 loss: 0.2562 +2023/06/05 05:35:07 - mmengine - INFO - Epoch(train) [2][1800/4111] lr: 9.5488e-05 eta: 6:57:56 time: 0.7537 data_time: 0.6135 memory: 6318 loss: 0.2651 +2023/06/05 05:36:08 - mmengine - INFO - Exp name: resnet50_2xb256_all_1m_lr1e-4_aug_1e-1_20230605_042304 +2023/06/05 05:36:14 - mmengine - INFO - Epoch(train) [2][1900/4111] lr: 9.5337e-05 eta: 6:56:20 time: 0.6432 data_time: 0.5038 memory: 6318 loss: 0.2727 +2023/06/05 05:37:21 - mmengine - INFO - Epoch(train) [2][2000/4111] lr: 9.5183e-05 eta: 6:54:41 time: 0.6921 data_time: 0.5514 memory: 6318 loss: 0.2482 +2023/06/05 05:38:29 - mmengine - INFO - Epoch(train) [2][2100/4111] lr: 9.5027e-05 eta: 6:53:14 time: 0.6437 data_time: 0.5037 memory: 6318 loss: 0.2737 +2023/06/05 05:39:35 - mmengine - INFO - Epoch(train) [2][2200/4111] lr: 9.4869e-05 eta: 6:51:36 time: 0.6386 data_time: 0.4852 memory: 6318 loss: 0.2672 +2023/06/05 05:40:45 - mmengine - INFO - Epoch(train) [2][2300/4111] lr: 9.4708e-05 eta: 6:50:18 time: 0.6691 data_time: 0.5302 memory: 6318 loss: 0.2820 +2023/06/05 05:41:57 - mmengine - INFO - Epoch(train) [2][2400/4111] lr: 9.4545e-05 eta: 6:49:15 time: 0.6558 data_time: 0.5160 memory: 6318 loss: 0.2813 +2023/06/05 05:43:05 - mmengine - INFO - Epoch(train) [2][2500/4111] lr: 9.4380e-05 eta: 6:47:48 time: 0.7393 data_time: 0.5992 memory: 6318 loss: 0.2600 +2023/06/05 05:44:17 - mmengine - INFO - Epoch(train) [2][2600/4111] lr: 9.4212e-05 eta: 6:46:43 time: 0.7186 data_time: 0.5790 memory: 6318 loss: 0.2557 +2023/06/05 05:45:26 - mmengine - INFO - Epoch(train) [2][2700/4111] lr: 9.4043e-05 eta: 6:45:22 time: 0.6528 data_time: 0.5132 memory: 6318 loss: 0.2733 +2023/06/05 05:46:35 - mmengine - INFO - Epoch(train) [2][2800/4111] lr: 9.3870e-05 eta: 6:44:02 time: 0.6648 data_time: 0.5245 memory: 6318 loss: 0.2612 +2023/06/05 05:47:36 - mmengine - INFO - Exp name: resnet50_2xb256_all_1m_lr1e-4_aug_1e-1_20230605_042304 +2023/06/05 05:47:44 - mmengine - INFO - Epoch(train) [2][2900/4111] lr: 9.3696e-05 eta: 6:42:42 time: 0.8142 data_time: 0.6749 memory: 6318 loss: 0.2527 +2023/06/05 05:48:59 - mmengine - INFO - Epoch(train) [2][3000/4111] lr: 9.3519e-05 eta: 6:41:52 time: 0.6556 data_time: 0.5158 memory: 6318 loss: 0.2518 +2023/06/05 05:50:10 - mmengine - INFO - Epoch(train) [2][3100/4111] lr: 9.3340e-05 eta: 6:40:40 time: 0.6868 data_time: 0.5484 memory: 6318 loss: 0.2574 +2023/06/05 05:51:16 - mmengine - INFO - Epoch(train) [2][3200/4111] lr: 9.3159e-05 eta: 6:39:07 time: 0.6466 data_time: 0.5062 memory: 6318 loss: 0.2500 +2023/06/05 05:52:22 - mmengine - INFO - Epoch(train) [2][3300/4111] lr: 9.2976e-05 eta: 6:37:31 time: 0.7152 data_time: 0.5768 memory: 6318 loss: 0.2577 +2023/06/05 05:53:33 - mmengine - INFO - Epoch(train) [2][3400/4111] lr: 9.2790e-05 eta: 6:36:22 time: 0.6677 data_time: 0.5277 memory: 6318 loss: 0.2390 +2023/06/05 05:54:43 - mmengine - INFO - Epoch(train) [2][3500/4111] lr: 9.2603e-05 eta: 6:35:10 time: 0.7328 data_time: 0.5924 memory: 6318 loss: 0.2464 +2023/06/05 05:55:54 - mmengine - INFO - Epoch(train) [2][3600/4111] lr: 9.2413e-05 eta: 6:33:59 time: 0.7079 data_time: 0.5584 memory: 6318 loss: 0.2338 +2023/06/05 05:57:02 - mmengine - INFO - Epoch(train) [2][3700/4111] lr: 9.2220e-05 eta: 6:32:37 time: 0.7019 data_time: 0.5611 memory: 6318 loss: 0.2387 +2023/06/05 05:58:13 - mmengine - INFO - Epoch(train) [2][3800/4111] lr: 9.2026e-05 eta: 6:31:25 time: 0.6808 data_time: 0.5362 memory: 6318 loss: 0.2550 +2023/06/05 05:59:15 - mmengine - INFO - Exp name: resnet50_2xb256_all_1m_lr1e-4_aug_1e-1_20230605_042304 +2023/06/05 05:59:22 - mmengine - INFO - Epoch(train) [2][3900/4111] lr: 9.1829e-05 eta: 6:30:08 time: 0.7194 data_time: 0.5798 memory: 6318 loss: 0.2351 +2023/06/05 06:00:29 - mmengine - INFO - Epoch(train) [2][4000/4111] lr: 9.1631e-05 eta: 6:28:41 time: 0.6537 data_time: 0.5132 memory: 6318 loss: 0.2341 +2023/06/05 06:01:35 - mmengine - INFO - Epoch(train) [2][4100/4111] lr: 9.1430e-05 eta: 6:27:13 time: 0.6221 data_time: 0.4816 memory: 6318 loss: 0.2507 +2023/06/05 06:01:41 - mmengine - INFO - Exp name: resnet50_2xb256_all_1m_lr1e-4_aug_1e-1_20230605_042304 +2023/06/05 06:01:41 - mmengine - INFO - Saving checkpoint at 2 epochs +2023/06/05 06:02:23 - mmengine - INFO - Epoch(val) [2][100/119] eta: 0:00:06 time: 0.6119 data_time: 0.5233 memory: 6318 +2023/06/05 06:02:48 - mmengine - INFO - Epoch(val) [2][119/119] accuracy/top1: 76.6182 data_time: 0.3264 time: 0.4149 +2023/06/05 06:03:57 - mmengine - INFO - Epoch(train) [3][ 100/4111] lr: 9.1205e-05 eta: 6:25:40 time: 0.6529 data_time: 0.5121 memory: 6318 loss: 0.2351 +2023/06/05 06:05:06 - mmengine - INFO - Epoch(train) [3][ 200/4111] lr: 9.0999e-05 eta: 6:24:22 time: 0.6927 data_time: 0.5536 memory: 6318 loss: 0.2398 +2023/06/05 06:06:12 - mmengine - INFO - Epoch(train) [3][ 300/4111] lr: 9.0792e-05 eta: 6:22:54 time: 0.6558 data_time: 0.5149 memory: 6318 loss: 0.2404 +2023/06/05 06:07:19 - mmengine - INFO - Epoch(train) [3][ 400/4111] lr: 9.0582e-05 eta: 6:21:32 time: 0.6465 data_time: 0.4992 memory: 6318 loss: 0.2312 +2023/06/05 06:08:30 - mmengine - INFO - Epoch(train) [3][ 500/4111] lr: 9.0371e-05 eta: 6:20:22 time: 0.6673 data_time: 0.5261 memory: 6318 loss: 0.2261 +2023/06/05 06:09:41 - mmengine - INFO - Epoch(train) [3][ 600/4111] lr: 9.0157e-05 eta: 6:19:13 time: 0.7101 data_time: 0.5704 memory: 6318 loss: 0.2586 +2023/06/05 06:10:51 - mmengine - INFO - Epoch(train) [3][ 700/4111] lr: 8.9942e-05 eta: 6:18:02 time: 0.6768 data_time: 0.5367 memory: 6318 loss: 0.2211 +2023/06/05 06:11:46 - mmengine - INFO - Exp name: resnet50_2xb256_all_1m_lr1e-4_aug_1e-1_20230605_042304 +2023/06/05 06:11:59 - mmengine - INFO - Epoch(train) [3][ 800/4111] lr: 8.9724e-05 eta: 6:16:43 time: 0.6304 data_time: 0.4898 memory: 6318 loss: 0.2293 +2023/06/05 06:13:07 - mmengine - INFO - Epoch(train) [3][ 900/4111] lr: 8.9504e-05 eta: 6:15:23 time: 0.6777 data_time: 0.5377 memory: 6318 loss: 0.2454 +2023/06/05 06:14:23 - mmengine - INFO - Epoch(train) [3][1000/4111] lr: 8.9282e-05 eta: 6:14:34 time: 0.6233 data_time: 0.4842 memory: 6318 loss: 0.2248 +2023/06/05 06:15:32 - mmengine - INFO - Epoch(train) [3][1100/4111] lr: 8.9059e-05 eta: 6:13:16 time: 0.6809 data_time: 0.5413 memory: 6318 loss: 0.2316 +2023/06/05 06:16:40 - mmengine - INFO - Epoch(train) [3][1200/4111] lr: 8.8833e-05 eta: 6:11:58 time: 0.6518 data_time: 0.5114 memory: 6318 loss: 0.2279 +2023/06/05 06:17:51 - mmengine - INFO - Epoch(train) [3][1300/4111] lr: 8.8605e-05 eta: 6:10:51 time: 0.6892 data_time: 0.5494 memory: 6318 loss: 0.2144 +2023/06/05 06:19:05 - mmengine - INFO - Epoch(train) [3][1400/4111] lr: 8.8375e-05 eta: 6:09:50 time: 0.7070 data_time: 0.5674 memory: 6318 loss: 0.2563 +2023/06/05 06:20:12 - mmengine - INFO - Epoch(train) [3][1500/4111] lr: 8.8144e-05 eta: 6:08:29 time: 0.6540 data_time: 0.5146 memory: 6318 loss: 0.2295 +2023/06/05 06:21:18 - mmengine - INFO - Epoch(train) [3][1600/4111] lr: 8.7910e-05 eta: 6:07:04 time: 0.6279 data_time: 0.4882 memory: 6318 loss: 0.2419 +2023/06/05 06:22:25 - mmengine - INFO - Epoch(train) [3][1700/4111] lr: 8.7675e-05 eta: 6:05:42 time: 0.6413 data_time: 0.5018 memory: 6318 loss: 0.2370 +2023/06/05 06:23:18 - mmengine - INFO - Exp name: resnet50_2xb256_all_1m_lr1e-4_aug_1e-1_20230605_042304 +2023/06/05 06:23:32 - mmengine - INFO - Epoch(train) [3][1800/4111] lr: 8.7437e-05 eta: 6:04:24 time: 0.7454 data_time: 0.6060 memory: 6318 loss: 0.2405 +2023/06/05 06:24:40 - mmengine - INFO - Epoch(train) [3][1900/4111] lr: 8.7198e-05 eta: 6:03:07 time: 0.6996 data_time: 0.5599 memory: 6318 loss: 0.2388 +2023/06/05 06:25:50 - mmengine - INFO - Epoch(train) [3][2000/4111] lr: 8.6957e-05 eta: 6:01:54 time: 0.6953 data_time: 0.5556 memory: 6318 loss: 0.2218 +2023/06/05 06:26:58 - mmengine - INFO - Epoch(train) [3][2100/4111] lr: 8.6714e-05 eta: 6:00:39 time: 0.6900 data_time: 0.5482 memory: 6318 loss: 0.2269 +2023/06/05 06:28:06 - mmengine - INFO - Epoch(train) [3][2200/4111] lr: 8.6469e-05 eta: 5:59:20 time: 0.6623 data_time: 0.5227 memory: 6318 loss: 0.2424 +2023/06/05 06:29:13 - mmengine - INFO - Epoch(train) [3][2300/4111] lr: 8.6222e-05 eta: 5:58:00 time: 0.7008 data_time: 0.5609 memory: 6318 loss: 0.2247 +2023/06/05 06:30:21 - mmengine - INFO - Epoch(train) [3][2400/4111] lr: 8.5974e-05 eta: 5:56:43 time: 0.6677 data_time: 0.5278 memory: 6318 loss: 0.2358 +2023/06/05 06:31:27 - mmengine - INFO - Epoch(train) [3][2500/4111] lr: 8.5723e-05 eta: 5:55:22 time: 0.7358 data_time: 0.5971 memory: 6318 loss: 0.2281 +2023/06/05 06:32:37 - mmengine - INFO - Epoch(train) [3][2600/4111] lr: 8.5471e-05 eta: 5:54:12 time: 0.6892 data_time: 0.5493 memory: 6318 loss: 0.2266 +2023/06/05 06:33:44 - mmengine - INFO - Epoch(train) [3][2700/4111] lr: 8.5217e-05 eta: 5:52:52 time: 0.6604 data_time: 0.5140 memory: 6318 loss: 0.2166 +2023/06/05 06:34:34 - mmengine - INFO - Exp name: resnet50_2xb256_all_1m_lr1e-4_aug_1e-1_20230605_042304 +2023/06/05 06:34:53 - mmengine - INFO - Epoch(train) [3][2800/4111] lr: 8.4962e-05 eta: 5:51:39 time: 0.6740 data_time: 0.5343 memory: 6318 loss: 0.2256 +2023/06/05 06:36:00 - mmengine - INFO - Epoch(train) [3][2900/4111] lr: 8.4704e-05 eta: 5:50:20 time: 0.6729 data_time: 0.5242 memory: 6318 loss: 0.2222 +2023/06/05 06:37:07 - mmengine - INFO - Epoch(train) [3][3000/4111] lr: 8.4445e-05 eta: 5:49:03 time: 0.6918 data_time: 0.5515 memory: 6318 loss: 0.2274 +2023/06/05 06:38:23 - mmengine - INFO - Epoch(train) [3][3100/4111] lr: 8.4184e-05 eta: 5:48:08 time: 0.6403 data_time: 0.4957 memory: 6318 loss: 0.2424 +2023/06/05 06:39:33 - mmengine - INFO - Epoch(train) [3][3200/4111] lr: 8.3921e-05 eta: 5:46:58 time: 0.6715 data_time: 0.5314 memory: 6318 loss: 0.2269 +2023/06/05 06:40:40 - mmengine - INFO - Epoch(train) [3][3300/4111] lr: 8.3657e-05 eta: 5:45:38 time: 0.6542 data_time: 0.5149 memory: 6318 loss: 0.2328 +2023/06/05 06:41:48 - mmengine - INFO - Epoch(train) [3][3400/4111] lr: 8.3391e-05 eta: 5:44:23 time: 0.7934 data_time: 0.6533 memory: 6318 loss: 0.2138 +2023/06/05 06:42:56 - mmengine - INFO - Epoch(train) [3][3500/4111] lr: 8.3123e-05 eta: 5:43:07 time: 0.7261 data_time: 0.5869 memory: 6318 loss: 0.2085 +2023/06/05 06:44:04 - mmengine - INFO - Epoch(train) [3][3600/4111] lr: 8.2854e-05 eta: 5:41:53 time: 0.7015 data_time: 0.5618 memory: 6318 loss: 0.2099 +2023/06/05 06:45:13 - mmengine - INFO - Epoch(train) [3][3700/4111] lr: 8.2583e-05 eta: 5:40:40 time: 0.7207 data_time: 0.5813 memory: 6318 loss: 0.2214 +2023/06/05 06:46:00 - mmengine - INFO - Exp name: resnet50_2xb256_all_1m_lr1e-4_aug_1e-1_20230605_042304 +2023/06/05 06:46:19 - mmengine - INFO - Epoch(train) [3][3800/4111] lr: 8.2311e-05 eta: 5:39:21 time: 0.6813 data_time: 0.5420 memory: 6318 loss: 0.2224 +2023/06/05 06:47:26 - mmengine - INFO - Epoch(train) [3][3900/4111] lr: 8.2037e-05 eta: 5:38:03 time: 0.6851 data_time: 0.5457 memory: 6318 loss: 0.2083 +2023/06/05 06:48:34 - mmengine - INFO - Epoch(train) [3][4000/4111] lr: 8.1761e-05 eta: 5:36:48 time: 0.6498 data_time: 0.5097 memory: 6318 loss: 0.2100 +2023/06/05 06:49:40 - mmengine - INFO - Epoch(train) [3][4100/4111] lr: 8.1484e-05 eta: 5:35:30 time: 0.6188 data_time: 0.4794 memory: 6318 loss: 0.2088 +2023/06/05 06:49:47 - mmengine - INFO - Exp name: resnet50_2xb256_all_1m_lr1e-4_aug_1e-1_20230605_042304 +2023/06/05 06:49:47 - mmengine - INFO - Saving checkpoint at 3 epochs +2023/06/05 06:50:29 - mmengine - INFO - Epoch(val) [3][100/119] eta: 0:00:06 time: 0.6094 data_time: 0.5207 memory: 6318 +2023/06/05 06:50:55 - mmengine - INFO - Epoch(val) [3][119/119] accuracy/top1: 82.1740 data_time: 0.3334 time: 0.4222 +2023/06/05 06:52:06 - mmengine - INFO - Epoch(train) [4][ 100/4111] lr: 8.1174e-05 eta: 5:34:14 time: 0.6891 data_time: 0.5500 memory: 6318 loss: 0.2066 +2023/06/05 06:53:13 - mmengine - INFO - Epoch(train) [4][ 200/4111] lr: 8.0894e-05 eta: 5:32:57 time: 0.6719 data_time: 0.5330 memory: 6318 loss: 0.2162 +2023/06/05 06:54:21 - mmengine - INFO - Epoch(train) [4][ 300/4111] lr: 8.0612e-05 eta: 5:31:43 time: 0.7122 data_time: 0.5733 memory: 6318 loss: 0.2147 +2023/06/05 06:55:53 - mmengine - INFO - Epoch(train) [4][ 400/4111] lr: 8.0328e-05 eta: 5:31:23 time: 0.9807 data_time: 0.8413 memory: 6318 loss: 0.2095 +2023/06/05 06:57:08 - mmengine - INFO - Epoch(train) [4][ 500/4111] lr: 8.0043e-05 eta: 5:30:24 time: 0.7969 data_time: 0.6574 memory: 6318 loss: 0.2101 +2023/06/05 06:59:07 - mmengine - INFO - Epoch(train) [4][ 600/4111] lr: 7.9757e-05 eta: 5:30:59 time: 3.7587 data_time: 3.6199 memory: 6318 loss: 0.1999 +2023/06/05 07:00:04 - mmengine - INFO - Exp name: resnet50_2xb256_all_1m_lr1e-4_aug_1e-1_20230605_042304 +2023/06/05 07:00:28 - mmengine - INFO - Epoch(train) [4][ 700/4111] lr: 7.9469e-05 eta: 5:30:12 time: 0.9326 data_time: 0.7939 memory: 6318 loss: 0.2053 +2023/06/05 07:01:41 - mmengine - INFO - Epoch(train) [4][ 800/4111] lr: 7.9180e-05 eta: 5:29:07 time: 0.6549 data_time: 0.5148 memory: 6318 loss: 0.2334 +2023/06/05 07:02:54 - mmengine - INFO - Epoch(train) [4][ 900/4111] lr: 7.8889e-05 eta: 5:28:01 time: 0.6627 data_time: 0.5238 memory: 6318 loss: 0.2003 +2023/06/05 07:04:05 - mmengine - INFO - Epoch(train) [4][1000/4111] lr: 7.8597e-05 eta: 5:26:52 time: 0.6817 data_time: 0.5417 memory: 6318 loss: 0.2139 +2023/06/05 07:05:31 - mmengine - INFO - Epoch(train) [4][1100/4111] lr: 7.8303e-05 eta: 5:26:12 time: 0.7733 data_time: 0.6320 memory: 6318 loss: 0.2207 +2023/06/05 07:07:16 - mmengine - INFO - Epoch(train) [4][1200/4111] lr: 7.8008e-05 eta: 5:26:11 time: 0.6919 data_time: 0.5527 memory: 6318 loss: 0.2070 +2023/06/05 07:08:26 - mmengine - INFO - Epoch(train) [4][1300/4111] lr: 7.7712e-05 eta: 5:24:58 time: 0.6996 data_time: 0.5588 memory: 6318 loss: 0.2056 +2023/06/05 07:09:52 - mmengine - INFO - Epoch(train) [4][1400/4111] lr: 7.7415e-05 eta: 5:24:18 time: 0.7376 data_time: 0.5980 memory: 6318 loss: 0.2001 +2023/06/05 07:11:07 - mmengine - INFO - Epoch(train) [4][1500/4111] lr: 7.7116e-05 eta: 5:23:15 time: 0.8817 data_time: 0.7419 memory: 6318 loss: 0.2053 +2023/06/05 07:12:18 - mmengine - INFO - Epoch(train) [4][1600/4111] lr: 7.6816e-05 eta: 5:22:04 time: 0.7149 data_time: 0.5750 memory: 6318 loss: 0.1991 +2023/06/05 07:13:06 - mmengine - INFO - Exp name: resnet50_2xb256_all_1m_lr1e-4_aug_1e-1_20230605_042304 +2023/06/05 07:13:26 - mmengine - INFO - Epoch(train) [4][1700/4111] lr: 7.6514e-05 eta: 5:20:47 time: 0.6677 data_time: 0.5282 memory: 6318 loss: 0.2221 +2023/06/05 07:14:33 - mmengine - INFO - Epoch(train) [4][1800/4111] lr: 7.6212e-05 eta: 5:19:27 time: 0.6753 data_time: 0.5357 memory: 6318 loss: 0.2149 +2023/06/05 07:15:43 - mmengine - INFO - Epoch(train) [4][1900/4111] lr: 7.5908e-05 eta: 5:18:14 time: 0.6480 data_time: 0.5065 memory: 6318 loss: 0.2118 +2023/06/05 07:16:51 - mmengine - INFO - Epoch(train) [4][2000/4111] lr: 7.5603e-05 eta: 5:16:58 time: 0.6751 data_time: 0.5268 memory: 6318 loss: 0.2203 +2023/06/05 07:18:00 - mmengine - INFO - Epoch(train) [4][2100/4111] lr: 7.5296e-05 eta: 5:15:42 time: 0.7193 data_time: 0.5785 memory: 6318 loss: 0.2021 +2023/06/05 07:19:09 - mmengine - INFO - Epoch(train) [4][2200/4111] lr: 7.4989e-05 eta: 5:14:29 time: 0.7157 data_time: 0.5727 memory: 6318 loss: 0.2078 +2023/06/05 07:20:22 - mmengine - INFO - Epoch(train) [4][2300/4111] lr: 7.4680e-05 eta: 5:13:20 time: 0.6712 data_time: 0.5318 memory: 6318 loss: 0.1939 +2023/06/05 07:21:30 - mmengine - INFO - Epoch(train) [4][2400/4111] lr: 7.4370e-05 eta: 5:12:04 time: 0.7483 data_time: 0.6084 memory: 6318 loss: 0.1979 +2023/06/05 07:22:42 - mmengine - INFO - Epoch(train) [4][2500/4111] lr: 7.4059e-05 eta: 5:10:54 time: 0.6741 data_time: 0.5344 memory: 6318 loss: 0.2179 +2023/06/05 07:23:51 - mmengine - INFO - Epoch(train) [4][2600/4111] lr: 7.3747e-05 eta: 5:09:40 time: 0.6585 data_time: 0.5175 memory: 6318 loss: 0.2071 +2023/06/05 07:24:37 - mmengine - INFO - Exp name: resnet50_2xb256_all_1m_lr1e-4_aug_1e-1_20230605_042304 +2023/06/05 07:25:01 - mmengine - INFO - Epoch(train) [4][2700/4111] lr: 7.3434e-05 eta: 5:08:28 time: 0.7038 data_time: 0.5623 memory: 6318 loss: 0.2106 +2023/06/05 07:26:11 - mmengine - INFO - Epoch(train) [4][2800/4111] lr: 7.3120e-05 eta: 5:07:16 time: 0.6760 data_time: 0.5360 memory: 6318 loss: 0.1990 +2023/06/05 07:27:24 - mmengine - INFO - Epoch(train) [4][2900/4111] lr: 7.2804e-05 eta: 5:06:07 time: 1.0544 data_time: 0.9147 memory: 6318 loss: 0.2062 +2023/06/05 07:28:32 - mmengine - INFO - Epoch(train) [4][3000/4111] lr: 7.2488e-05 eta: 5:04:52 time: 0.6835 data_time: 0.5441 memory: 6318 loss: 0.1991 +2023/06/05 07:29:41 - mmengine - INFO - Epoch(train) [4][3100/4111] lr: 7.2171e-05 eta: 5:03:37 time: 0.6813 data_time: 0.5419 memory: 6318 loss: 0.1903 +2023/06/05 07:30:49 - mmengine - INFO - Epoch(train) [4][3200/4111] lr: 7.1852e-05 eta: 5:02:22 time: 0.6837 data_time: 0.5426 memory: 6318 loss: 0.2156 +2023/06/05 07:31:58 - mmengine - INFO - Epoch(train) [4][3300/4111] lr: 7.1533e-05 eta: 5:01:07 time: 0.6905 data_time: 0.5513 memory: 6318 loss: 0.2176 +2023/06/05 07:33:07 - mmengine - INFO - Epoch(train) [4][3400/4111] lr: 7.1213e-05 eta: 4:59:53 time: 0.6949 data_time: 0.5554 memory: 6318 loss: 0.2083 +2023/06/05 07:34:23 - mmengine - INFO - Epoch(train) [4][3500/4111] lr: 7.0891e-05 eta: 4:58:50 time: 0.9142 data_time: 0.7726 memory: 6318 loss: 0.2121 +2023/06/05 07:35:32 - mmengine - INFO - Epoch(train) [4][3600/4111] lr: 7.0569e-05 eta: 4:57:36 time: 0.7232 data_time: 0.5817 memory: 6318 loss: 0.2016 +2023/06/05 07:36:16 - mmengine - INFO - Exp name: resnet50_2xb256_all_1m_lr1e-4_aug_1e-1_20230605_042304 +2023/06/05 07:36:40 - mmengine - INFO - Epoch(train) [4][3700/4111] lr: 7.0246e-05 eta: 4:56:21 time: 0.6925 data_time: 0.5392 memory: 6318 loss: 0.2005 +2023/06/05 07:37:48 - mmengine - INFO - Epoch(train) [4][3800/4111] lr: 6.9922e-05 eta: 4:55:05 time: 0.6424 data_time: 0.5014 memory: 6318 loss: 0.2120 +2023/06/05 07:38:57 - mmengine - INFO - Epoch(train) [4][3900/4111] lr: 6.9597e-05 eta: 4:53:52 time: 0.6864 data_time: 0.5450 memory: 6318 loss: 0.1946 +2023/06/05 07:40:11 - mmengine - INFO - Epoch(train) [4][4000/4111] lr: 6.9272e-05 eta: 4:52:45 time: 0.6641 data_time: 0.5248 memory: 6318 loss: 0.2059 +2023/06/05 07:41:21 - mmengine - INFO - Epoch(train) [4][4100/4111] lr: 6.8945e-05 eta: 4:51:33 time: 0.6902 data_time: 0.5502 memory: 6318 loss: 0.2112 +2023/06/05 07:41:28 - mmengine - INFO - Exp name: resnet50_2xb256_all_1m_lr1e-4_aug_1e-1_20230605_042304 +2023/06/05 07:41:28 - mmengine - INFO - Saving checkpoint at 4 epochs +2023/06/05 07:42:10 - mmengine - INFO - Epoch(val) [4][100/119] eta: 0:00:06 time: 0.5669 data_time: 0.4781 memory: 6318 +2023/06/05 07:42:36 - mmengine - INFO - Epoch(val) [4][119/119] accuracy/top1: 87.5693 data_time: 0.3290 time: 0.4164 +2023/06/05 07:43:50 - mmengine - INFO - Epoch(train) [5][ 100/4111] lr: 6.8582e-05 eta: 4:50:20 time: 0.7525 data_time: 0.6104 memory: 6318 loss: 0.1965 +2023/06/05 07:45:02 - mmengine - INFO - Epoch(train) [5][ 200/4111] lr: 6.8253e-05 eta: 4:49:11 time: 0.6520 data_time: 0.5069 memory: 6318 loss: 0.1992 +2023/06/05 07:46:14 - mmengine - INFO - Epoch(train) [5][ 300/4111] lr: 6.7924e-05 eta: 4:48:00 time: 0.7026 data_time: 0.5620 memory: 6318 loss: 0.2052 +2023/06/05 07:47:25 - mmengine - INFO - Epoch(train) [5][ 400/4111] lr: 6.7595e-05 eta: 4:46:50 time: 0.6679 data_time: 0.5227 memory: 6318 loss: 0.1997 +2023/06/05 07:48:45 - mmengine - INFO - Epoch(train) [5][ 500/4111] lr: 6.7264e-05 eta: 4:45:51 time: 0.8475 data_time: 0.7072 memory: 6318 loss: 0.1740 +2023/06/05 07:49:27 - mmengine - INFO - Exp name: resnet50_2xb256_all_1m_lr1e-4_aug_1e-1_20230605_042304 +2023/06/05 07:49:56 - mmengine - INFO - Epoch(train) [5][ 600/4111] lr: 6.6933e-05 eta: 4:44:41 time: 0.6908 data_time: 0.5499 memory: 6318 loss: 0.1807 +2023/06/05 07:51:05 - mmengine - INFO - Epoch(train) [5][ 700/4111] lr: 6.6601e-05 eta: 4:43:27 time: 0.6313 data_time: 0.4910 memory: 6318 loss: 0.1913 +2023/06/05 07:52:10 - mmengine - INFO - Epoch(train) [5][ 800/4111] lr: 6.6268e-05 eta: 4:42:08 time: 0.6615 data_time: 0.5211 memory: 6318 loss: 0.2028 +2023/06/05 07:53:17 - mmengine - INFO - Epoch(train) [5][ 900/4111] lr: 6.5935e-05 eta: 4:40:51 time: 0.6667 data_time: 0.5267 memory: 6318 loss: 0.1880 +2023/06/05 07:54:24 - mmengine - INFO - Epoch(train) [5][1000/4111] lr: 6.5601e-05 eta: 4:39:36 time: 0.6608 data_time: 0.4699 memory: 6318 loss: 0.1942 +2023/06/05 07:55:34 - mmengine - INFO - Epoch(train) [5][1100/4111] lr: 6.5267e-05 eta: 4:38:23 time: 0.7252 data_time: 0.5861 memory: 6318 loss: 0.1862 +2023/06/05 07:56:42 - mmengine - INFO - Epoch(train) [5][1200/4111] lr: 6.4932e-05 eta: 4:37:09 time: 0.6648 data_time: 0.5248 memory: 6318 loss: 0.1850 +2023/06/05 07:57:50 - mmengine - INFO - Epoch(train) [5][1300/4111] lr: 6.4596e-05 eta: 4:35:53 time: 0.6707 data_time: 0.5299 memory: 6318 loss: 0.1880 +2023/06/05 07:58:59 - mmengine - INFO - Epoch(train) [5][1400/4111] lr: 6.4260e-05 eta: 4:34:40 time: 0.6519 data_time: 0.5015 memory: 6318 loss: 0.1826 +2023/06/05 08:00:08 - mmengine - INFO - Epoch(train) [5][1500/4111] lr: 6.3923e-05 eta: 4:33:28 time: 0.7340 data_time: 0.5950 memory: 6318 loss: 0.1818 +2023/06/05 08:00:48 - mmengine - INFO - Exp name: resnet50_2xb256_all_1m_lr1e-4_aug_1e-1_20230605_042304 +2023/06/05 08:01:17 - mmengine - INFO - Epoch(train) [5][1600/4111] lr: 6.3585e-05 eta: 4:32:14 time: 0.6649 data_time: 0.5260 memory: 6318 loss: 0.2020 +2023/06/05 08:02:26 - mmengine - INFO - Epoch(train) [5][1700/4111] lr: 6.3248e-05 eta: 4:31:01 time: 0.6833 data_time: 0.5348 memory: 6318 loss: 0.2267 +2023/06/05 08:03:35 - mmengine - INFO - Epoch(train) [5][1800/4111] lr: 6.2909e-05 eta: 4:29:48 time: 0.6969 data_time: 0.5557 memory: 6318 loss: 0.1825 +2023/06/05 08:04:44 - mmengine - INFO - Epoch(train) [5][1900/4111] lr: 6.2571e-05 eta: 4:28:35 time: 0.6793 data_time: 0.5397 memory: 6318 loss: 0.2001 +2023/06/05 08:05:52 - mmengine - INFO - Epoch(train) [5][2000/4111] lr: 6.2231e-05 eta: 4:27:21 time: 0.7036 data_time: 0.5628 memory: 6318 loss: 0.2003 +2023/06/05 08:07:00 - mmengine - INFO - Epoch(train) [5][2100/4111] lr: 6.1892e-05 eta: 4:26:06 time: 0.6916 data_time: 0.5520 memory: 6318 loss: 0.1820 +2023/06/05 08:08:08 - mmengine - INFO - Epoch(train) [5][2200/4111] lr: 6.1552e-05 eta: 4:24:52 time: 0.6888 data_time: 0.5490 memory: 6318 loss: 0.2004 +2023/06/05 08:09:17 - mmengine - INFO - Epoch(train) [5][2300/4111] lr: 6.1211e-05 eta: 4:23:40 time: 0.7217 data_time: 0.5800 memory: 6318 loss: 0.1835 +2023/06/05 08:10:27 - mmengine - INFO - Epoch(train) [5][2400/4111] lr: 6.0871e-05 eta: 4:22:28 time: 0.6632 data_time: 0.5233 memory: 6318 loss: 0.2027 +2023/06/05 08:11:35 - mmengine - INFO - Epoch(train) [5][2500/4111] lr: 6.0529e-05 eta: 4:21:14 time: 0.6963 data_time: 0.5548 memory: 6318 loss: 0.2032 +2023/06/05 08:12:12 - mmengine - INFO - Exp name: resnet50_2xb256_all_1m_lr1e-4_aug_1e-1_20230605_042304 +2023/06/05 08:12:44 - mmengine - INFO - Epoch(train) [5][2600/4111] lr: 6.0188e-05 eta: 4:20:01 time: 0.6788 data_time: 0.5398 memory: 6318 loss: 0.1846 +2023/06/05 08:13:52 - mmengine - INFO - Epoch(train) [5][2700/4111] lr: 5.9846e-05 eta: 4:18:47 time: 0.6945 data_time: 0.5467 memory: 6318 loss: 0.2016 +2023/06/05 08:15:02 - mmengine - INFO - Epoch(train) [5][2800/4111] lr: 5.9504e-05 eta: 4:17:36 time: 0.6891 data_time: 0.5480 memory: 6318 loss: 0.1944 +2023/06/05 08:16:10 - mmengine - INFO - Epoch(train) [5][2900/4111] lr: 5.9162e-05 eta: 4:16:22 time: 0.6932 data_time: 0.5531 memory: 6318 loss: 0.1817 +2023/06/05 08:17:17 - mmengine - INFO - Epoch(train) [5][3000/4111] lr: 5.8819e-05 eta: 4:15:08 time: 0.6642 data_time: 0.5245 memory: 6318 loss: 0.1841 +2023/06/05 08:18:26 - mmengine - INFO - Epoch(train) [5][3100/4111] lr: 5.8477e-05 eta: 4:13:55 time: 0.6804 data_time: 0.5391 memory: 6318 loss: 0.1948 +2023/06/05 08:19:36 - mmengine - INFO - Epoch(train) [5][3200/4111] lr: 5.8134e-05 eta: 4:12:43 time: 0.7249 data_time: 0.5829 memory: 6318 loss: 0.2007 +2023/06/05 08:20:45 - mmengine - INFO - Epoch(train) [5][3300/4111] lr: 5.7791e-05 eta: 4:11:32 time: 0.6943 data_time: 0.5544 memory: 6318 loss: 0.1966 +2023/06/05 08:21:57 - mmengine - INFO - Epoch(train) [5][3400/4111] lr: 5.7447e-05 eta: 4:10:22 time: 0.6463 data_time: 0.5061 memory: 6318 loss: 0.2002 +2023/06/05 08:23:07 - mmengine - INFO - Epoch(train) [5][3500/4111] lr: 5.7104e-05 eta: 4:09:11 time: 0.6664 data_time: 0.5270 memory: 6318 loss: 0.1879 +2023/06/05 08:23:45 - mmengine - INFO - Exp name: resnet50_2xb256_all_1m_lr1e-4_aug_1e-1_20230605_042304 +2023/06/05 08:24:19 - mmengine - INFO - Epoch(train) [5][3600/4111] lr: 5.6760e-05 eta: 4:08:01 time: 0.6527 data_time: 0.5110 memory: 6318 loss: 0.1934 +2023/06/05 08:25:28 - mmengine - INFO - Epoch(train) [5][3700/4111] lr: 5.6417e-05 eta: 4:06:49 time: 0.7501 data_time: 0.6092 memory: 6318 loss: 0.2002 +2023/06/05 08:26:37 - mmengine - INFO - Epoch(train) [5][3800/4111] lr: 5.6073e-05 eta: 4:05:37 time: 0.6989 data_time: 0.5596 memory: 6318 loss: 0.1893 +2023/06/05 08:27:45 - mmengine - INFO - Epoch(train) [5][3900/4111] lr: 5.5729e-05 eta: 4:04:23 time: 0.6485 data_time: 0.5081 memory: 6318 loss: 0.1763 +2023/06/05 08:28:52 - mmengine - INFO - Epoch(train) [5][4000/4111] lr: 5.5385e-05 eta: 4:03:09 time: 0.7505 data_time: 0.6104 memory: 6318 loss: 0.1867 +2023/06/05 08:30:04 - mmengine - INFO - Epoch(train) [5][4100/4111] lr: 5.5041e-05 eta: 4:02:00 time: 0.6874 data_time: 0.5456 memory: 6318 loss: 0.1997 +2023/06/05 08:30:11 - mmengine - INFO - Exp name: resnet50_2xb256_all_1m_lr1e-4_aug_1e-1_20230605_042304 +2023/06/05 08:30:11 - mmengine - INFO - Saving checkpoint at 5 epochs +2023/06/05 08:30:53 - mmengine - INFO - Epoch(val) [5][100/119] eta: 0:00:06 time: 0.6590 data_time: 0.5708 memory: 6318 +2023/06/05 08:31:22 - mmengine - INFO - Epoch(val) [5][119/119] accuracy/top1: 90.7916 data_time: 0.3614 time: 0.4499 +2023/06/05 08:32:40 - mmengine - INFO - Epoch(train) [6][ 100/4111] lr: 5.4660e-05 eta: 4:00:48 time: 0.7297 data_time: 0.5882 memory: 6318 loss: 0.1936 +2023/06/05 08:33:49 - mmengine - INFO - Epoch(train) [6][ 200/4111] lr: 5.4316e-05 eta: 3:59:36 time: 0.7079 data_time: 0.5668 memory: 6318 loss: 0.1863 +2023/06/05 08:35:05 - mmengine - INFO - Epoch(train) [6][ 300/4111] lr: 5.3972e-05 eta: 3:58:30 time: 0.6163 data_time: 0.4769 memory: 6318 loss: 0.2100 +2023/06/05 08:36:17 - mmengine - INFO - Epoch(train) [6][ 400/4111] lr: 5.3628e-05 eta: 3:57:21 time: 0.9684 data_time: 0.8281 memory: 6318 loss: 0.1943 +2023/06/05 08:36:49 - mmengine - INFO - Exp name: resnet50_2xb256_all_1m_lr1e-4_aug_1e-1_20230605_042304 +2023/06/05 08:37:30 - mmengine - INFO - Epoch(train) [6][ 500/4111] lr: 5.3284e-05 eta: 3:56:13 time: 0.7643 data_time: 0.6247 memory: 6318 loss: 0.1848 +2023/06/05 08:38:41 - mmengine - INFO - Epoch(train) [6][ 600/4111] lr: 5.2941e-05 eta: 3:55:03 time: 0.6981 data_time: 0.5587 memory: 6318 loss: 0.1864 +2023/06/05 08:39:48 - mmengine - INFO - Epoch(train) [6][ 700/4111] lr: 5.2597e-05 eta: 3:53:48 time: 0.6411 data_time: 0.5001 memory: 6318 loss: 0.1788 +2023/06/05 08:40:57 - mmengine - INFO - Epoch(train) [6][ 800/4111] lr: 5.2254e-05 eta: 3:52:36 time: 0.6403 data_time: 0.4992 memory: 6318 loss: 0.1884 +2023/06/05 08:42:05 - mmengine - INFO - Epoch(train) [6][ 900/4111] lr: 5.1911e-05 eta: 3:51:23 time: 0.7014 data_time: 0.5623 memory: 6318 loss: 0.1881 +2023/06/05 08:43:14 - mmengine - INFO - Epoch(train) [6][1000/4111] lr: 5.1568e-05 eta: 3:50:11 time: 0.7145 data_time: 0.5729 memory: 6318 loss: 0.1664 +2023/06/05 08:44:24 - mmengine - INFO - Epoch(train) [6][1100/4111] lr: 5.1225e-05 eta: 3:49:00 time: 0.7451 data_time: 0.6028 memory: 6318 loss: 0.1892 +2023/06/05 08:45:34 - mmengine - INFO - Epoch(train) [6][1200/4111] lr: 5.0883e-05 eta: 3:47:49 time: 0.8238 data_time: 0.6845 memory: 6318 loss: 0.1976 +2023/06/05 08:46:43 - mmengine - INFO - Epoch(train) [6][1300/4111] lr: 5.0540e-05 eta: 3:46:37 time: 0.6799 data_time: 0.5400 memory: 6318 loss: 0.2103 +2023/06/05 08:47:52 - mmengine - INFO - Epoch(train) [6][1400/4111] lr: 5.0198e-05 eta: 3:45:25 time: 0.6935 data_time: 0.5545 memory: 6318 loss: 0.1891 +2023/06/05 08:48:20 - mmengine - INFO - Exp name: resnet50_2xb256_all_1m_lr1e-4_aug_1e-1_20230605_042304 +2023/06/05 08:48:59 - mmengine - INFO - Epoch(train) [6][1500/4111] lr: 4.9856e-05 eta: 3:44:11 time: 0.6637 data_time: 0.5248 memory: 6318 loss: 0.1875 +2023/06/05 08:50:09 - mmengine - INFO - Epoch(train) [6][1600/4111] lr: 4.9515e-05 eta: 3:43:00 time: 0.6708 data_time: 0.5312 memory: 6318 loss: 0.1783 +2023/06/05 08:51:16 - mmengine - INFO - Epoch(train) [6][1700/4111] lr: 4.9174e-05 eta: 3:41:46 time: 0.6817 data_time: 0.5416 memory: 6318 loss: 0.1846 +2023/06/05 08:52:23 - mmengine - INFO - Epoch(train) [6][1800/4111] lr: 4.8833e-05 eta: 3:40:32 time: 0.6841 data_time: 0.5440 memory: 6318 loss: 0.1957 +2023/06/05 08:53:30 - mmengine - INFO - Epoch(train) [6][1900/4111] lr: 4.8492e-05 eta: 3:39:19 time: 0.6858 data_time: 0.5453 memory: 6318 loss: 0.1802 +2023/06/05 08:54:38 - mmengine - INFO - Epoch(train) [6][2000/4111] lr: 4.8152e-05 eta: 3:38:06 time: 0.6768 data_time: 0.5378 memory: 6318 loss: 0.1841 +2023/06/05 08:55:46 - mmengine - INFO - Epoch(train) [6][2100/4111] lr: 4.7813e-05 eta: 3:36:54 time: 0.7972 data_time: 0.6555 memory: 6318 loss: 0.1690 +2023/06/05 08:56:55 - mmengine - INFO - Epoch(train) [6][2200/4111] lr: 4.7473e-05 eta: 3:35:42 time: 0.6761 data_time: 0.5349 memory: 6318 loss: 0.1722 +2023/06/05 08:58:05 - mmengine - INFO - Epoch(train) [6][2300/4111] lr: 4.7135e-05 eta: 3:34:31 time: 0.6606 data_time: 0.5202 memory: 6318 loss: 0.1744 +2023/06/05 08:59:13 - mmengine - INFO - Epoch(train) [6][2400/4111] lr: 4.6796e-05 eta: 3:33:19 time: 0.7129 data_time: 0.5723 memory: 6318 loss: 0.1782 +2023/06/05 08:59:41 - mmengine - INFO - Exp name: resnet50_2xb256_all_1m_lr1e-4_aug_1e-1_20230605_042304 +2023/06/05 09:00:22 - mmengine - INFO - Epoch(train) [6][2500/4111] lr: 4.6458e-05 eta: 3:32:07 time: 0.6966 data_time: 0.5570 memory: 6318 loss: 0.1884 +2023/06/05 09:01:33 - mmengine - INFO - Epoch(train) [6][2600/4111] lr: 4.6121e-05 eta: 3:30:56 time: 0.7554 data_time: 0.6069 memory: 6318 loss: 0.1869 +2023/06/05 09:02:48 - mmengine - INFO - Epoch(train) [6][2700/4111] lr: 4.5784e-05 eta: 3:29:50 time: 0.6648 data_time: 0.5252 memory: 6318 loss: 0.1785 +2023/06/05 09:04:07 - mmengine - INFO - Epoch(train) [6][2800/4111] lr: 4.5448e-05 eta: 3:28:46 time: 0.6645 data_time: 0.5238 memory: 6318 loss: 0.1825 +2023/06/05 09:05:13 - mmengine - INFO - Epoch(train) [6][2900/4111] lr: 4.5112e-05 eta: 3:27:31 time: 0.6811 data_time: 0.5413 memory: 6318 loss: 0.1810 +2023/06/05 09:06:21 - mmengine - INFO - Epoch(train) [6][3000/4111] lr: 4.4777e-05 eta: 3:26:19 time: 0.6667 data_time: 0.5271 memory: 6318 loss: 0.1888 +2023/06/05 09:07:29 - mmengine - INFO - Epoch(train) [6][3100/4111] lr: 4.4442e-05 eta: 3:25:07 time: 0.7421 data_time: 0.6008 memory: 6318 loss: 0.1906 +2023/06/05 09:08:39 - mmengine - INFO - Epoch(train) [6][3200/4111] lr: 4.4108e-05 eta: 3:23:56 time: 0.6488 data_time: 0.5090 memory: 6318 loss: 0.1860 +2023/06/05 09:09:52 - mmengine - INFO - Epoch(train) [6][3300/4111] lr: 4.3775e-05 eta: 3:22:47 time: 0.7039 data_time: 0.5628 memory: 6318 loss: 0.1724 +2023/06/05 09:10:59 - mmengine - INFO - Epoch(train) [6][3400/4111] lr: 4.3442e-05 eta: 3:21:34 time: 0.6503 data_time: 0.5091 memory: 6318 loss: 0.1883 +2023/06/05 09:11:27 - mmengine - INFO - Exp name: resnet50_2xb256_all_1m_lr1e-4_aug_1e-1_20230605_042304 +2023/06/05 09:12:07 - mmengine - INFO - Epoch(train) [6][3500/4111] lr: 4.3110e-05 eta: 3:20:22 time: 0.7181 data_time: 0.5778 memory: 6318 loss: 0.1944 +2023/06/05 09:13:17 - mmengine - INFO - Epoch(train) [6][3600/4111] lr: 4.2779e-05 eta: 3:19:11 time: 0.7065 data_time: 0.5667 memory: 6318 loss: 0.1760 +2023/06/05 09:14:24 - mmengine - INFO - Epoch(train) [6][3700/4111] lr: 4.2448e-05 eta: 3:17:58 time: 0.6673 data_time: 0.5272 memory: 6318 loss: 0.1834 +2023/06/05 09:15:35 - mmengine - INFO - Epoch(train) [6][3800/4111] lr: 4.2119e-05 eta: 3:16:48 time: 0.6924 data_time: 0.5520 memory: 6318 loss: 0.1931 +2023/06/05 09:16:44 - mmengine - INFO - Epoch(train) [6][3900/4111] lr: 4.1789e-05 eta: 3:15:36 time: 0.6757 data_time: 0.5355 memory: 6318 loss: 0.1664 +2023/06/05 09:17:53 - mmengine - INFO - Epoch(train) [6][4000/4111] lr: 4.1461e-05 eta: 3:14:25 time: 0.6254 data_time: 0.4852 memory: 6318 loss: 0.1958 +2023/06/05 09:19:02 - mmengine - INFO - Epoch(train) [6][4100/4111] lr: 4.1133e-05 eta: 3:13:13 time: 0.7272 data_time: 0.5881 memory: 6318 loss: 0.2020 +2023/06/05 09:19:08 - mmengine - INFO - Exp name: resnet50_2xb256_all_1m_lr1e-4_aug_1e-1_20230605_042304 +2023/06/05 09:19:08 - mmengine - INFO - Saving checkpoint at 6 epochs +2023/06/05 09:19:50 - mmengine - INFO - Epoch(val) [6][100/119] eta: 0:00:06 time: 0.5611 data_time: 0.4726 memory: 6318 +2023/06/05 09:20:16 - mmengine - INFO - Epoch(val) [6][119/119] accuracy/top1: 92.5244 data_time: 0.3328 time: 0.4211 +2023/06/05 09:21:27 - mmengine - INFO - Epoch(train) [7][ 100/4111] lr: 4.0771e-05 eta: 3:11:55 time: 0.7193 data_time: 0.4646 memory: 6318 loss: 0.1814 +2023/06/05 09:22:37 - mmengine - INFO - Epoch(train) [7][ 200/4111] lr: 4.0445e-05 eta: 3:10:44 time: 0.6639 data_time: 0.5248 memory: 6318 loss: 0.1694 +2023/06/05 09:23:57 - mmengine - INFO - Epoch(train) [7][ 300/4111] lr: 4.0120e-05 eta: 3:09:40 time: 0.7094 data_time: 0.5701 memory: 6318 loss: 0.1868 +2023/06/05 09:24:23 - mmengine - INFO - Exp name: resnet50_2xb256_all_1m_lr1e-4_aug_1e-1_20230605_042304 +2023/06/05 09:25:07 - mmengine - INFO - Epoch(train) [7][ 400/4111] lr: 3.9796e-05 eta: 3:08:29 time: 0.6476 data_time: 0.5069 memory: 6318 loss: 0.1741 +2023/06/05 09:26:15 - mmengine - INFO - Epoch(train) [7][ 500/4111] lr: 3.9473e-05 eta: 3:07:17 time: 0.6585 data_time: 0.5183 memory: 6318 loss: 0.1977 +2023/06/05 09:27:23 - mmengine - INFO - Epoch(train) [7][ 600/4111] lr: 3.9150e-05 eta: 3:06:04 time: 0.6441 data_time: 0.5041 memory: 6318 loss: 0.1667 +2023/06/05 09:28:40 - mmengine - INFO - Epoch(train) [7][ 700/4111] lr: 3.8829e-05 eta: 3:04:58 time: 0.7195 data_time: 0.5778 memory: 6318 loss: 0.1736 +2023/06/05 09:29:49 - mmengine - INFO - Epoch(train) [7][ 800/4111] lr: 3.8509e-05 eta: 3:03:47 time: 0.7399 data_time: 0.6008 memory: 6318 loss: 0.1912 +2023/06/05 09:30:59 - mmengine - INFO - Epoch(train) [7][ 900/4111] lr: 3.8189e-05 eta: 3:02:36 time: 0.6967 data_time: 0.5565 memory: 6318 loss: 0.1841 +2023/06/05 09:32:07 - mmengine - INFO - Epoch(train) [7][1000/4111] lr: 3.7871e-05 eta: 3:01:24 time: 0.7850 data_time: 0.6453 memory: 6318 loss: 0.1779 +2023/06/05 09:33:17 - mmengine - INFO - Epoch(train) [7][1100/4111] lr: 3.7553e-05 eta: 3:00:13 time: 0.6913 data_time: 0.5507 memory: 6318 loss: 0.1802 +2023/06/05 09:34:25 - mmengine - INFO - Epoch(train) [7][1200/4111] lr: 3.7237e-05 eta: 2:59:02 time: 0.7087 data_time: 0.5684 memory: 6318 loss: 0.1848 +2023/06/05 09:35:34 - mmengine - INFO - Epoch(train) [7][1300/4111] lr: 3.6921e-05 eta: 2:57:50 time: 0.6983 data_time: 0.5580 memory: 6318 loss: 0.1797 +2023/06/05 09:36:01 - mmengine - INFO - Exp name: resnet50_2xb256_all_1m_lr1e-4_aug_1e-1_20230605_042304 +2023/06/05 09:36:45 - mmengine - INFO - Epoch(train) [7][1400/4111] lr: 3.6607e-05 eta: 2:56:40 time: 0.7293 data_time: 0.5900 memory: 6318 loss: 0.1740 +2023/06/05 09:37:54 - mmengine - INFO - Epoch(train) [7][1500/4111] lr: 3.6294e-05 eta: 2:55:28 time: 0.6585 data_time: 0.5165 memory: 6318 loss: 0.1933 +2023/06/05 09:39:03 - mmengine - INFO - Epoch(train) [7][1600/4111] lr: 3.5981e-05 eta: 2:54:18 time: 0.7451 data_time: 0.6045 memory: 6318 loss: 0.2084 +2023/06/05 09:40:12 - mmengine - INFO - Epoch(train) [7][1700/4111] lr: 3.5670e-05 eta: 2:53:06 time: 0.6747 data_time: 0.5336 memory: 6318 loss: 0.1621 +2023/06/05 09:41:23 - mmengine - INFO - Epoch(train) [7][1800/4111] lr: 3.5360e-05 eta: 2:51:56 time: 0.7210 data_time: 0.5795 memory: 6318 loss: 0.1837 +2023/06/05 09:42:31 - mmengine - INFO - Epoch(train) [7][1900/4111] lr: 3.5051e-05 eta: 2:50:44 time: 0.6737 data_time: 0.5326 memory: 6318 loss: 0.2047 +2023/06/05 09:43:41 - mmengine - INFO - Epoch(train) [7][2000/4111] lr: 3.4744e-05 eta: 2:49:34 time: 0.6825 data_time: 0.5428 memory: 6318 loss: 0.1599 +2023/06/05 09:44:51 - mmengine - INFO - Epoch(train) [7][2100/4111] lr: 3.4437e-05 eta: 2:48:23 time: 0.6812 data_time: 0.5406 memory: 6318 loss: 0.1679 +2023/06/05 09:46:01 - mmengine - INFO - Epoch(train) [7][2200/4111] lr: 3.4132e-05 eta: 2:47:12 time: 0.6666 data_time: 0.5270 memory: 6318 loss: 0.1751 +2023/06/05 09:47:12 - mmengine - INFO - Epoch(train) [7][2300/4111] lr: 3.3828e-05 eta: 2:46:02 time: 0.6772 data_time: 0.5361 memory: 6318 loss: 0.1872 +2023/06/05 09:47:33 - mmengine - INFO - Exp name: resnet50_2xb256_all_1m_lr1e-4_aug_1e-1_20230605_042304 +2023/06/05 09:48:21 - mmengine - INFO - Epoch(train) [7][2400/4111] lr: 3.3525e-05 eta: 2:44:51 time: 0.6718 data_time: 0.5320 memory: 6318 loss: 0.1854 +2023/06/05 09:49:30 - mmengine - INFO - Epoch(train) [7][2500/4111] lr: 3.3224e-05 eta: 2:43:40 time: 0.6549 data_time: 0.5130 memory: 6318 loss: 0.1695 +2023/06/05 09:50:38 - mmengine - INFO - Epoch(train) [7][2600/4111] lr: 3.2923e-05 eta: 2:42:28 time: 0.6954 data_time: 0.5537 memory: 6318 loss: 0.1784 +2023/06/05 09:51:48 - mmengine - INFO - Epoch(train) [7][2700/4111] lr: 3.2624e-05 eta: 2:41:17 time: 0.7574 data_time: 0.6148 memory: 6318 loss: 0.1982 +2023/06/05 09:52:57 - mmengine - INFO - Epoch(train) [7][2800/4111] lr: 3.2326e-05 eta: 2:40:06 time: 0.7150 data_time: 0.5746 memory: 6318 loss: 0.1773 +2023/06/05 09:54:04 - mmengine - INFO - Epoch(train) [7][2900/4111] lr: 3.2030e-05 eta: 2:38:54 time: 0.6553 data_time: 0.5163 memory: 6318 loss: 0.1781 +2023/06/05 09:55:14 - mmengine - INFO - Epoch(train) [7][3000/4111] lr: 3.1735e-05 eta: 2:37:43 time: 0.6583 data_time: 0.5178 memory: 6318 loss: 0.1846 +2023/06/05 09:56:21 - mmengine - INFO - Epoch(train) [7][3100/4111] lr: 3.1441e-05 eta: 2:36:31 time: 0.6836 data_time: 0.5437 memory: 6318 loss: 0.1862 +2023/06/05 09:57:30 - mmengine - INFO - Epoch(train) [7][3200/4111] lr: 3.1149e-05 eta: 2:35:20 time: 0.6799 data_time: 0.5386 memory: 6318 loss: 0.1819 +2023/06/05 09:58:39 - mmengine - INFO - Epoch(train) [7][3300/4111] lr: 3.0858e-05 eta: 2:34:09 time: 0.6873 data_time: 0.5464 memory: 6318 loss: 0.1833 +2023/06/05 09:59:00 - mmengine - INFO - Exp name: resnet50_2xb256_all_1m_lr1e-4_aug_1e-1_20230605_042304 +2023/06/05 09:59:46 - mmengine - INFO - Epoch(train) [7][3400/4111] lr: 3.0569e-05 eta: 2:32:58 time: 0.6369 data_time: 0.4966 memory: 6318 loss: 0.1984 +2023/06/05 10:00:54 - mmengine - INFO - Epoch(train) [7][3500/4111] lr: 3.0281e-05 eta: 2:31:46 time: 0.6819 data_time: 0.5429 memory: 6318 loss: 0.1919 +2023/06/05 10:02:03 - mmengine - INFO - Epoch(train) [7][3600/4111] lr: 2.9994e-05 eta: 2:30:35 time: 0.6860 data_time: 0.5452 memory: 6318 loss: 0.1728 +2023/06/05 10:03:12 - mmengine - INFO - Epoch(train) [7][3700/4111] lr: 2.9709e-05 eta: 2:29:24 time: 0.6525 data_time: 0.5103 memory: 6318 loss: 0.2020 +2023/06/05 10:04:20 - mmengine - INFO - Epoch(train) [7][3800/4111] lr: 2.9425e-05 eta: 2:28:13 time: 0.6337 data_time: 0.4938 memory: 6318 loss: 0.1569 +2023/06/05 10:05:29 - mmengine - INFO - Epoch(train) [7][3900/4111] lr: 2.9143e-05 eta: 2:27:02 time: 0.6560 data_time: 0.5153 memory: 6318 loss: 0.1754 +2023/06/05 10:06:39 - mmengine - INFO - Epoch(train) [7][4000/4111] lr: 2.8862e-05 eta: 2:25:51 time: 0.6811 data_time: 0.5369 memory: 6318 loss: 0.1651 +2023/06/05 10:07:46 - mmengine - INFO - Epoch(train) [7][4100/4111] lr: 2.8583e-05 eta: 2:24:39 time: 0.6952 data_time: 0.5557 memory: 6318 loss: 0.1785 +2023/06/05 10:07:53 - mmengine - INFO - Exp name: resnet50_2xb256_all_1m_lr1e-4_aug_1e-1_20230605_042304 +2023/06/05 10:07:53 - mmengine - INFO - Saving checkpoint at 7 epochs +2023/06/05 10:08:35 - mmengine - INFO - Epoch(val) [7][100/119] eta: 0:00:06 time: 0.6411 data_time: 0.5524 memory: 6318 +2023/06/05 10:09:02 - mmengine - INFO - Epoch(val) [7][119/119] accuracy/top1: 93.6084 data_time: 0.3388 time: 0.4280 +2023/06/05 10:10:17 - mmengine - INFO - Epoch(train) [8][ 100/4111] lr: 2.8275e-05 eta: 2:23:23 time: 0.7000 data_time: 0.3466 memory: 6318 loss: 0.1809 +2023/06/05 10:11:26 - mmengine - INFO - Epoch(train) [8][ 200/4111] lr: 2.7999e-05 eta: 2:22:12 time: 0.7807 data_time: 0.3971 memory: 6318 loss: 0.1934 +2023/06/05 10:11:44 - mmengine - INFO - Exp name: resnet50_2xb256_all_1m_lr1e-4_aug_1e-1_20230605_042304 +2023/06/05 10:12:36 - mmengine - INFO - Epoch(train) [8][ 300/4111] lr: 2.7725e-05 eta: 2:21:01 time: 0.6687 data_time: 0.5286 memory: 6318 loss: 0.1711 +2023/06/05 10:13:44 - mmengine - INFO - Epoch(train) [8][ 400/4111] lr: 2.7452e-05 eta: 2:19:50 time: 0.6724 data_time: 0.5334 memory: 6318 loss: 0.1987 +2023/06/05 10:14:56 - mmengine - INFO - Epoch(train) [8][ 500/4111] lr: 2.7181e-05 eta: 2:18:41 time: 0.7010 data_time: 0.5628 memory: 6318 loss: 0.1737 +2023/06/05 10:16:04 - mmengine - INFO - Epoch(train) [8][ 600/4111] lr: 2.6911e-05 eta: 2:17:30 time: 0.7668 data_time: 0.6279 memory: 6318 loss: 0.1853 +2023/06/05 10:17:12 - mmengine - INFO - Epoch(train) [8][ 700/4111] lr: 2.6644e-05 eta: 2:16:18 time: 0.7733 data_time: 0.6329 memory: 6318 loss: 0.1672 +2023/06/05 10:18:22 - mmengine - INFO - Epoch(train) [8][ 800/4111] lr: 2.6377e-05 eta: 2:15:08 time: 0.6908 data_time: 0.5497 memory: 6318 loss: 0.1769 +2023/06/05 10:19:31 - mmengine - INFO - Epoch(train) [8][ 900/4111] lr: 2.6113e-05 eta: 2:13:57 time: 0.7253 data_time: 0.5836 memory: 6318 loss: 0.1720 +2023/06/05 10:20:40 - mmengine - INFO - Epoch(train) [8][1000/4111] lr: 2.5850e-05 eta: 2:12:46 time: 0.6694 data_time: 0.5291 memory: 6318 loss: 0.1731 +2023/06/05 10:21:49 - mmengine - INFO - Epoch(train) [8][1100/4111] lr: 2.5589e-05 eta: 2:11:35 time: 0.7433 data_time: 0.6032 memory: 6318 loss: 0.1501 +2023/06/05 10:23:00 - mmengine - INFO - Epoch(train) [8][1200/4111] lr: 2.5330e-05 eta: 2:10:25 time: 0.7386 data_time: 0.5976 memory: 6318 loss: 0.1694 +2023/06/05 10:23:14 - mmengine - INFO - Exp name: resnet50_2xb256_all_1m_lr1e-4_aug_1e-1_20230605_042304 +2023/06/05 10:24:09 - mmengine - INFO - Epoch(train) [8][1300/4111] lr: 2.5072e-05 eta: 2:09:15 time: 0.6841 data_time: 0.5418 memory: 6318 loss: 0.1813 +2023/06/05 10:25:17 - mmengine - INFO - Epoch(train) [8][1400/4111] lr: 2.4816e-05 eta: 2:08:04 time: 0.6441 data_time: 0.5030 memory: 6318 loss: 0.1685 +2023/06/05 10:26:26 - mmengine - INFO - Epoch(train) [8][1500/4111] lr: 2.4562e-05 eta: 2:06:53 time: 0.6538 data_time: 0.5140 memory: 6318 loss: 0.1859 +2023/06/05 10:27:38 - mmengine - INFO - Epoch(train) [8][1600/4111] lr: 2.4309e-05 eta: 2:05:43 time: 0.6847 data_time: 0.5446 memory: 6318 loss: 0.1759 +2023/06/05 10:28:47 - mmengine - INFO - Epoch(train) [8][1700/4111] lr: 2.4059e-05 eta: 2:04:32 time: 0.7024 data_time: 0.5620 memory: 6318 loss: 0.1682 +2023/06/05 10:30:01 - mmengine - INFO - Epoch(train) [8][1800/4111] lr: 2.3810e-05 eta: 2:03:24 time: 0.6897 data_time: 0.5496 memory: 6318 loss: 0.1661 +2023/06/05 10:31:11 - mmengine - INFO - Epoch(train) [8][1900/4111] lr: 2.3563e-05 eta: 2:02:13 time: 0.6882 data_time: 0.5490 memory: 6318 loss: 0.1699 +2023/06/05 10:32:20 - mmengine - INFO - Epoch(train) [8][2000/4111] lr: 2.3318e-05 eta: 2:01:02 time: 0.7059 data_time: 0.5658 memory: 6318 loss: 0.1782 +2023/06/05 10:33:30 - mmengine - INFO - Epoch(train) [8][2100/4111] lr: 2.3075e-05 eta: 1:59:52 time: 0.6718 data_time: 0.5316 memory: 6318 loss: 0.1860 +2023/06/05 10:34:40 - mmengine - INFO - Epoch(train) [8][2200/4111] lr: 2.2833e-05 eta: 1:58:42 time: 0.7300 data_time: 0.5765 memory: 6318 loss: 0.1732 +2023/06/05 10:34:55 - mmengine - INFO - Exp name: resnet50_2xb256_all_1m_lr1e-4_aug_1e-1_20230605_042304 +2023/06/05 10:35:50 - mmengine - INFO - Epoch(train) [8][2300/4111] lr: 2.2594e-05 eta: 1:57:31 time: 0.7166 data_time: 0.5759 memory: 6318 loss: 0.1827 +2023/06/05 10:37:11 - mmengine - INFO - Epoch(train) [8][2400/4111] lr: 2.2356e-05 eta: 1:56:24 time: 0.6596 data_time: 0.5200 memory: 6318 loss: 0.1858 +2023/06/05 10:38:20 - mmengine - INFO - Epoch(train) [8][2500/4111] lr: 2.2120e-05 eta: 1:55:14 time: 0.6646 data_time: 0.5255 memory: 6318 loss: 0.1778 +2023/06/05 10:39:30 - mmengine - INFO - Epoch(train) [8][2600/4111] lr: 2.1886e-05 eta: 1:54:03 time: 0.6579 data_time: 0.5154 memory: 6318 loss: 0.1764 +2023/06/05 10:40:40 - mmengine - INFO - Epoch(train) [8][2700/4111] lr: 2.1655e-05 eta: 1:52:53 time: 0.6683 data_time: 0.5291 memory: 6318 loss: 0.1867 +2023/06/05 10:41:47 - mmengine - INFO - Epoch(train) [8][2800/4111] lr: 2.1425e-05 eta: 1:51:41 time: 0.6617 data_time: 0.5164 memory: 6318 loss: 0.1630 +2023/06/05 10:44:28 - mmengine - INFO - Epoch(train) [8][2900/4111] lr: 2.1197e-05 eta: 1:50:58 time: 0.7043 data_time: 0.5644 memory: 6318 loss: 0.1776 +2023/06/05 10:45:35 - mmengine - INFO - Epoch(train) [8][3000/4111] lr: 2.0971e-05 eta: 1:49:46 time: 0.6814 data_time: 0.5326 memory: 6318 loss: 0.1851 +2023/06/05 10:46:42 - mmengine - INFO - Epoch(train) [8][3100/4111] lr: 2.0747e-05 eta: 1:48:35 time: 0.6248 data_time: 0.4842 memory: 6318 loss: 0.1678 +2023/06/05 10:47:48 - mmengine - INFO - Epoch(train) [8][3200/4111] lr: 2.0525e-05 eta: 1:47:23 time: 0.7102 data_time: 0.5699 memory: 6318 loss: 0.1741 +2023/06/05 10:48:01 - mmengine - INFO - Exp name: resnet50_2xb256_all_1m_lr1e-4_aug_1e-1_20230605_042304 +2023/06/05 10:48:54 - mmengine - INFO - Epoch(train) [8][3300/4111] lr: 2.0305e-05 eta: 1:46:11 time: 0.7096 data_time: 0.5702 memory: 6318 loss: 0.1513 +2023/06/05 10:50:00 - mmengine - INFO - Epoch(train) [8][3400/4111] lr: 2.0087e-05 eta: 1:44:59 time: 0.6107 data_time: 0.4709 memory: 6318 loss: 0.1840 +2023/06/05 10:51:05 - mmengine - INFO - Epoch(train) [8][3500/4111] lr: 1.9871e-05 eta: 1:43:47 time: 0.6468 data_time: 0.5076 memory: 6318 loss: 0.1777 +2023/06/05 10:52:10 - mmengine - INFO - Epoch(train) [8][3600/4111] lr: 1.9657e-05 eta: 1:42:35 time: 0.6388 data_time: 0.5000 memory: 6318 loss: 0.1928 +2023/06/05 10:53:15 - mmengine - INFO - Epoch(train) [8][3700/4111] lr: 1.9445e-05 eta: 1:41:23 time: 0.6394 data_time: 0.3368 memory: 6318 loss: 0.1737 +2023/06/05 10:54:20 - mmengine - INFO - Epoch(train) [8][3800/4111] lr: 1.9235e-05 eta: 1:40:11 time: 0.6786 data_time: 0.2336 memory: 6318 loss: 0.1746 +2023/06/05 10:55:25 - mmengine - INFO - Epoch(train) [8][3900/4111] lr: 1.9028e-05 eta: 1:39:00 time: 0.6521 data_time: 0.3715 memory: 6318 loss: 0.1874 +2023/06/05 10:56:31 - mmengine - INFO - Epoch(train) [8][4000/4111] lr: 1.8822e-05 eta: 1:37:48 time: 0.6261 data_time: 0.4321 memory: 6318 loss: 0.1712 +2023/06/05 10:57:37 - mmengine - INFO - Epoch(train) [8][4100/4111] lr: 1.8619e-05 eta: 1:36:36 time: 0.6536 data_time: 0.4312 memory: 6318 loss: 0.1740 +2023/06/05 10:57:44 - mmengine - INFO - Exp name: resnet50_2xb256_all_1m_lr1e-4_aug_1e-1_20230605_042304 +2023/06/05 10:57:44 - mmengine - INFO - Saving checkpoint at 8 epochs +2023/06/05 10:58:26 - mmengine - INFO - Epoch(val) [8][100/119] eta: 0:00:06 time: 0.5305 data_time: 0.4418 memory: 6318 +2023/06/05 10:58:51 - mmengine - INFO - Epoch(val) [8][119/119] accuracy/top1: 94.4293 data_time: 0.3191 time: 0.4082 +2023/06/05 11:00:00 - mmengine - INFO - Epoch(train) [9][ 100/4111] lr: 1.8395e-05 eta: 1:35:18 time: 0.6420 data_time: 0.3468 memory: 6318 loss: 0.1785 +2023/06/05 11:00:10 - mmengine - INFO - Exp name: resnet50_2xb256_all_1m_lr1e-4_aug_1e-1_20230605_042304 +2023/06/05 11:01:05 - mmengine - INFO - Epoch(train) [9][ 200/4111] lr: 1.8196e-05 eta: 1:34:06 time: 0.6008 data_time: 0.4608 memory: 6318 loss: 0.1632 +2023/06/05 11:02:13 - mmengine - INFO - Epoch(train) [9][ 300/4111] lr: 1.7999e-05 eta: 1:32:55 time: 0.8737 data_time: 0.7202 memory: 6318 loss: 0.1667 +2023/06/05 11:03:24 - mmengine - INFO - Epoch(train) [9][ 400/4111] lr: 1.7805e-05 eta: 1:31:45 time: 0.7355 data_time: 0.3189 memory: 6318 loss: 0.1734 +2023/06/05 11:04:38 - mmengine - INFO - Epoch(train) [9][ 500/4111] lr: 1.7612e-05 eta: 1:30:35 time: 0.7606 data_time: 0.3928 memory: 6318 loss: 0.1776 +2023/06/05 11:05:49 - mmengine - INFO - Epoch(train) [9][ 600/4111] lr: 1.7422e-05 eta: 1:29:25 time: 0.6478 data_time: 0.1640 memory: 6318 loss: 0.1790 +2023/06/05 11:07:00 - mmengine - INFO - Epoch(train) [9][ 700/4111] lr: 1.7234e-05 eta: 1:28:15 time: 0.7338 data_time: 0.0442 memory: 6318 loss: 0.1730 +2023/06/05 11:08:11 - mmengine - INFO - Epoch(train) [9][ 800/4111] lr: 1.7048e-05 eta: 1:27:04 time: 0.6869 data_time: 0.0009 memory: 6318 loss: 0.1847 +2023/06/05 11:09:24 - mmengine - INFO - Epoch(train) [9][ 900/4111] lr: 1.6864e-05 eta: 1:25:54 time: 0.8214 data_time: 0.0012 memory: 6318 loss: 0.1775 +2023/06/05 11:10:39 - mmengine - INFO - Epoch(train) [9][1000/4111] lr: 1.6683e-05 eta: 1:24:45 time: 0.6815 data_time: 0.5313 memory: 6318 loss: 0.1760 +2023/06/05 11:11:49 - mmengine - INFO - Epoch(train) [9][1100/4111] lr: 1.6504e-05 eta: 1:23:35 time: 0.6951 data_time: 0.2721 memory: 6318 loss: 0.1712 +2023/06/05 11:11:56 - mmengine - INFO - Exp name: resnet50_2xb256_all_1m_lr1e-4_aug_1e-1_20230605_042304 +2023/06/05 11:12:56 - mmengine - INFO - Epoch(train) [9][1200/4111] lr: 1.6327e-05 eta: 1:22:24 time: 0.6473 data_time: 0.0601 memory: 6318 loss: 0.1605 +2023/06/05 11:14:06 - mmengine - INFO - Epoch(train) [9][1300/4111] lr: 1.6152e-05 eta: 1:21:13 time: 0.7316 data_time: 0.1700 memory: 6318 loss: 0.1614 +2023/06/05 11:15:14 - mmengine - INFO - Epoch(train) [9][1400/4111] lr: 1.5980e-05 eta: 1:20:02 time: 0.6472 data_time: 0.0019 memory: 6318 loss: 0.1671 +2023/06/05 11:16:24 - mmengine - INFO - Epoch(train) [9][1500/4111] lr: 1.5810e-05 eta: 1:18:52 time: 0.6969 data_time: 0.1018 memory: 6318 loss: 0.1774 +2023/06/05 11:17:32 - mmengine - INFO - Epoch(train) [9][1600/4111] lr: 1.5642e-05 eta: 1:17:41 time: 0.6635 data_time: 0.1210 memory: 6318 loss: 0.2010 +2023/06/05 11:18:43 - mmengine - INFO - Epoch(train) [9][1700/4111] lr: 1.5476e-05 eta: 1:16:30 time: 0.7296 data_time: 0.0376 memory: 6318 loss: 0.1773 +2023/06/05 11:19:54 - mmengine - INFO - Epoch(train) [9][1800/4111] lr: 1.5313e-05 eta: 1:15:20 time: 0.7693 data_time: 0.0762 memory: 6318 loss: 0.1808 +2023/06/05 11:21:04 - mmengine - INFO - Epoch(train) [9][1900/4111] lr: 1.5152e-05 eta: 1:14:10 time: 0.7373 data_time: 0.2208 memory: 6318 loss: 0.1700 +2023/06/05 11:22:14 - mmengine - INFO - Epoch(train) [9][2000/4111] lr: 1.4993e-05 eta: 1:12:59 time: 0.6940 data_time: 0.5547 memory: 6318 loss: 0.1677 +2023/06/05 11:23:25 - mmengine - INFO - Epoch(train) [9][2100/4111] lr: 1.4837e-05 eta: 1:11:49 time: 0.7662 data_time: 0.6261 memory: 6318 loss: 0.1616 +2023/06/05 11:23:31 - mmengine - INFO - Exp name: resnet50_2xb256_all_1m_lr1e-4_aug_1e-1_20230605_042304 +2023/06/05 11:24:36 - mmengine - INFO - Epoch(train) [9][2200/4111] lr: 1.4683e-05 eta: 1:10:39 time: 0.6872 data_time: 0.5476 memory: 6318 loss: 0.1703 +2023/06/05 11:25:44 - mmengine - INFO - Epoch(train) [9][2300/4111] lr: 1.4532e-05 eta: 1:09:28 time: 0.7560 data_time: 0.6144 memory: 6318 loss: 0.1780 +2023/06/05 11:26:55 - mmengine - INFO - Epoch(train) [9][2400/4111] lr: 1.4382e-05 eta: 1:08:18 time: 0.7118 data_time: 0.5710 memory: 6318 loss: 0.1599 +2023/06/05 11:28:09 - mmengine - INFO - Epoch(train) [9][2500/4111] lr: 1.4235e-05 eta: 1:07:08 time: 0.6652 data_time: 0.5250 memory: 6318 loss: 0.1810 +2023/06/05 11:29:18 - mmengine - INFO - Epoch(train) [9][2600/4111] lr: 1.4091e-05 eta: 1:05:57 time: 0.6633 data_time: 0.5211 memory: 6318 loss: 0.2039 +2023/06/05 11:30:30 - mmengine - INFO - Epoch(train) [9][2700/4111] lr: 1.3949e-05 eta: 1:04:47 time: 0.6528 data_time: 0.5101 memory: 6318 loss: 0.1734 +2023/06/05 11:31:42 - mmengine - INFO - Epoch(train) [9][2800/4111] lr: 1.3809e-05 eta: 1:03:37 time: 0.7015 data_time: 0.5619 memory: 6318 loss: 0.1807 +2023/06/05 11:32:51 - mmengine - INFO - Epoch(train) [9][2900/4111] lr: 1.3672e-05 eta: 1:02:26 time: 0.6962 data_time: 0.5552 memory: 6318 loss: 0.1915 +2023/06/05 11:34:06 - mmengine - INFO - Epoch(train) [9][3000/4111] lr: 1.3537e-05 eta: 1:01:17 time: 0.6545 data_time: 0.5141 memory: 6318 loss: 0.1836 +2023/06/05 11:35:20 - mmengine - INFO - Epoch(train) [9][3100/4111] lr: 1.3405e-05 eta: 1:00:07 time: 0.7365 data_time: 0.5975 memory: 6318 loss: 0.1667 +2023/06/05 11:35:28 - mmengine - INFO - Exp name: resnet50_2xb256_all_1m_lr1e-4_aug_1e-1_20230605_042304 +2023/06/05 11:36:30 - mmengine - INFO - Epoch(train) [9][3200/4111] lr: 1.3275e-05 eta: 0:58:56 time: 0.6958 data_time: 0.5556 memory: 6318 loss: 0.1723 +2023/06/05 11:37:41 - mmengine - INFO - Epoch(train) [9][3300/4111] lr: 1.3147e-05 eta: 0:57:46 time: 0.6644 data_time: 0.5238 memory: 6318 loss: 0.1840 +2023/06/05 11:38:55 - mmengine - INFO - Epoch(train) [9][3400/4111] lr: 1.3022e-05 eta: 0:56:36 time: 0.7057 data_time: 0.5662 memory: 6318 loss: 0.1802 +2023/06/05 11:40:05 - mmengine - INFO - Epoch(train) [9][3500/4111] lr: 1.2899e-05 eta: 0:55:25 time: 0.6263 data_time: 0.4857 memory: 6318 loss: 0.1728 +2023/06/05 11:41:16 - mmengine - INFO - Epoch(train) [9][3600/4111] lr: 1.2779e-05 eta: 0:54:15 time: 0.8215 data_time: 0.6819 memory: 6318 loss: 0.1657 +2023/06/05 11:42:27 - mmengine - INFO - Epoch(train) [9][3700/4111] lr: 1.2661e-05 eta: 0:53:05 time: 0.6506 data_time: 0.5093 memory: 6318 loss: 0.1669 +2023/06/05 11:43:49 - mmengine - INFO - Epoch(train) [9][3800/4111] lr: 1.2546e-05 eta: 0:51:56 time: 1.2272 data_time: 1.0880 memory: 6318 loss: 0.1649 +2023/06/05 11:45:04 - mmengine - INFO - Epoch(train) [9][3900/4111] lr: 1.2433e-05 eta: 0:50:46 time: 0.8262 data_time: 0.6861 memory: 6318 loss: 0.1716 +2023/06/05 11:46:16 - mmengine - INFO - Epoch(train) [9][4000/4111] lr: 1.2323e-05 eta: 0:49:35 time: 0.7812 data_time: 0.6412 memory: 6318 loss: 0.1585 +2023/06/05 11:47:27 - mmengine - INFO - Epoch(train) [9][4100/4111] lr: 1.2215e-05 eta: 0:48:25 time: 0.6845 data_time: 0.5433 memory: 6318 loss: 0.1710 +2023/06/05 11:47:34 - mmengine - INFO - Exp name: resnet50_2xb256_all_1m_lr1e-4_aug_1e-1_20230605_042304 +2023/06/05 11:47:34 - mmengine - INFO - Saving checkpoint at 9 epochs +2023/06/05 11:48:18 - mmengine - INFO - Epoch(val) [9][100/119] eta: 0:00:07 time: 0.8372 data_time: 0.7481 memory: 6318 +2023/06/05 11:48:46 - mmengine - INFO - Epoch(val) [9][119/119] accuracy/top1: 94.5451 data_time: 0.3506 time: 0.4392 +2023/06/05 11:48:53 - mmengine - INFO - Exp name: resnet50_2xb256_all_1m_lr1e-4_aug_1e-1_20230605_042304 +2023/06/05 11:49:59 - mmengine - INFO - Epoch(train) [10][ 100/4111] lr: 1.2098e-05 eta: 0:47:07 time: 0.6648 data_time: 0.2433 memory: 6318 loss: 0.1597 +2023/06/05 11:51:09 - mmengine - INFO - Epoch(train) [10][ 200/4111] lr: 1.1996e-05 eta: 0:45:56 time: 0.7985 data_time: 0.6584 memory: 6318 loss: 0.1560 +2023/06/05 11:52:17 - mmengine - INFO - Epoch(train) [10][ 300/4111] lr: 1.1896e-05 eta: 0:44:46 time: 0.6541 data_time: 0.5123 memory: 6318 loss: 0.1761 +2023/06/05 11:53:26 - mmengine - INFO - Epoch(train) [10][ 400/4111] lr: 1.1798e-05 eta: 0:43:35 time: 0.6463 data_time: 0.5055 memory: 6318 loss: 0.1699 +2023/06/05 11:54:39 - mmengine - INFO - Epoch(train) [10][ 500/4111] lr: 1.1703e-05 eta: 0:42:25 time: 1.0077 data_time: 0.8644 memory: 6318 loss: 0.1651 +2023/06/05 11:56:57 - mmengine - INFO - Epoch(train) [10][ 600/4111] lr: 1.1611e-05 eta: 0:41:20 time: 0.7169 data_time: 0.5759 memory: 6318 loss: 0.1731 +2023/06/05 11:58:05 - mmengine - INFO - Epoch(train) [10][ 700/4111] lr: 1.1521e-05 eta: 0:40:10 time: 0.6585 data_time: 0.5187 memory: 6318 loss: 0.1673 +2023/06/05 11:59:14 - mmengine - INFO - Epoch(train) [10][ 800/4111] lr: 1.1434e-05 eta: 0:38:59 time: 0.6699 data_time: 0.5278 memory: 6318 loss: 0.1681 +2023/06/05 12:00:21 - mmengine - INFO - Epoch(train) [10][ 900/4111] lr: 1.1349e-05 eta: 0:37:48 time: 0.6334 data_time: 0.4925 memory: 6318 loss: 0.1536 +2023/06/05 12:01:32 - mmengine - INFO - Epoch(train) [10][1000/4111] lr: 1.1267e-05 eta: 0:36:37 time: 0.6854 data_time: 0.5401 memory: 6318 loss: 0.1602 +2023/06/05 12:01:32 - mmengine - INFO - Exp name: resnet50_2xb256_all_1m_lr1e-4_aug_1e-1_20230605_042304 +2023/06/05 12:02:41 - mmengine - INFO - Epoch(train) [10][1100/4111] lr: 1.1187e-05 eta: 0:35:26 time: 0.7267 data_time: 0.5870 memory: 6318 loss: 0.1692 +2023/06/05 12:03:52 - mmengine - INFO - Epoch(train) [10][1200/4111] lr: 1.1110e-05 eta: 0:34:16 time: 0.6680 data_time: 0.5291 memory: 6318 loss: 0.1670 +2023/06/05 12:05:01 - mmengine - INFO - Epoch(train) [10][1300/4111] lr: 1.1035e-05 eta: 0:33:05 time: 0.6356 data_time: 0.4959 memory: 6318 loss: 0.1667 +2023/06/05 12:06:11 - mmengine - INFO - Epoch(train) [10][1400/4111] lr: 1.0963e-05 eta: 0:31:54 time: 0.6594 data_time: 0.5132 memory: 6318 loss: 0.1806 +2023/06/05 12:07:32 - mmengine - INFO - Epoch(train) [10][1500/4111] lr: 1.0893e-05 eta: 0:30:44 time: 0.6971 data_time: 0.5574 memory: 6318 loss: 0.1712 +2023/06/05 12:08:41 - mmengine - INFO - Epoch(train) [10][1600/4111] lr: 1.0827e-05 eta: 0:29:34 time: 0.7271 data_time: 0.5870 memory: 6318 loss: 0.1878 +2023/06/05 12:09:51 - mmengine - INFO - Epoch(train) [10][1700/4111] lr: 1.0762e-05 eta: 0:28:23 time: 0.6739 data_time: 0.5329 memory: 6318 loss: 0.1806 +2023/06/05 12:11:01 - mmengine - INFO - Epoch(train) [10][1800/4111] lr: 1.0701e-05 eta: 0:27:12 time: 0.6983 data_time: 0.5583 memory: 6318 loss: 0.1936 +2023/06/05 12:12:13 - mmengine - INFO - Epoch(train) [10][1900/4111] lr: 1.0641e-05 eta: 0:26:02 time: 0.7575 data_time: 0.6159 memory: 6318 loss: 0.1804 +2023/06/05 12:13:22 - mmengine - INFO - Epoch(train) [10][2000/4111] lr: 1.0585e-05 eta: 0:24:51 time: 0.6712 data_time: 0.5298 memory: 6318 loss: 0.1770 +2023/06/05 12:13:22 - mmengine - INFO - Exp name: resnet50_2xb256_all_1m_lr1e-4_aug_1e-1_20230605_042304 +2023/06/05 12:14:31 - mmengine - INFO - Epoch(train) [10][2100/4111] lr: 1.0531e-05 eta: 0:23:40 time: 0.7146 data_time: 0.5743 memory: 6318 loss: 0.1720 +2023/06/05 12:15:42 - mmengine - INFO - Epoch(train) [10][2200/4111] lr: 1.0480e-05 eta: 0:22:30 time: 0.7426 data_time: 0.6036 memory: 6318 loss: 0.1623 +2023/06/05 12:16:50 - mmengine - INFO - Epoch(train) [10][2300/4111] lr: 1.0431e-05 eta: 0:21:19 time: 0.6865 data_time: 0.5477 memory: 6318 loss: 0.1617 +2023/06/05 12:18:00 - mmengine - INFO - Epoch(train) [10][2400/4111] lr: 1.0385e-05 eta: 0:20:08 time: 0.7210 data_time: 0.5811 memory: 6318 loss: 0.1715 +2023/06/05 12:19:09 - mmengine - INFO - Epoch(train) [10][2500/4111] lr: 1.0341e-05 eta: 0:18:57 time: 0.7228 data_time: 0.5829 memory: 6318 loss: 0.1631 +2023/06/05 12:20:17 - mmengine - INFO - Epoch(train) [10][2600/4111] lr: 1.0300e-05 eta: 0:17:47 time: 0.6030 data_time: 0.4642 memory: 6318 loss: 0.1728 +2023/06/05 12:21:25 - mmengine - INFO - Epoch(train) [10][2700/4111] lr: 1.0262e-05 eta: 0:16:36 time: 0.6022 data_time: 0.4624 memory: 6318 loss: 0.1872 +2023/06/05 12:22:32 - mmengine - INFO - Epoch(train) [10][2800/4111] lr: 1.0226e-05 eta: 0:15:25 time: 0.6113 data_time: 0.4722 memory: 6318 loss: 0.1667 +2023/06/05 12:23:42 - mmengine - INFO - Epoch(train) [10][2900/4111] lr: 1.0193e-05 eta: 0:14:15 time: 0.7533 data_time: 0.5411 memory: 6318 loss: 0.1810 +2023/06/05 12:24:50 - mmengine - INFO - Epoch(train) [10][3000/4111] lr: 1.0162e-05 eta: 0:13:04 time: 0.6399 data_time: 0.1822 memory: 6318 loss: 0.1852 +2023/06/05 12:24:50 - mmengine - INFO - Exp name: resnet50_2xb256_all_1m_lr1e-4_aug_1e-1_20230605_042304 +2023/06/05 12:26:00 - mmengine - INFO - Epoch(train) [10][3100/4111] lr: 1.0135e-05 eta: 0:11:53 time: 0.6463 data_time: 0.1517 memory: 6318 loss: 0.1541 +2023/06/05 12:27:09 - mmengine - INFO - Epoch(train) [10][3200/4111] lr: 1.0109e-05 eta: 0:10:43 time: 0.7540 data_time: 0.6154 memory: 6318 loss: 0.1657 +2023/06/05 12:28:20 - mmengine - INFO - Epoch(train) [10][3300/4111] lr: 1.0087e-05 eta: 0:09:32 time: 0.6389 data_time: 0.4992 memory: 6318 loss: 0.1606 +2023/06/05 12:29:35 - mmengine - INFO - Epoch(train) [10][3400/4111] lr: 1.0067e-05 eta: 0:08:22 time: 0.6840 data_time: 0.5434 memory: 6318 loss: 0.1712 +2023/06/05 12:30:46 - mmengine - INFO - Epoch(train) [10][3500/4111] lr: 1.0049e-05 eta: 0:07:11 time: 0.7084 data_time: 0.5598 memory: 6318 loss: 0.1637 +2023/06/05 12:32:00 - mmengine - INFO - Epoch(train) [10][3600/4111] lr: 1.0034e-05 eta: 0:06:00 time: 0.7190 data_time: 0.5804 memory: 6318 loss: 0.1933 +2023/06/05 12:33:10 - mmengine - INFO - Epoch(train) [10][3700/4111] lr: 1.0022e-05 eta: 0:04:50 time: 0.6487 data_time: 0.5082 memory: 6318 loss: 0.1668 +2023/06/05 12:34:21 - mmengine - INFO - Epoch(train) [10][3800/4111] lr: 1.0013e-05 eta: 0:03:39 time: 0.6828 data_time: 0.5437 memory: 6318 loss: 0.1766 +2023/06/05 12:35:32 - mmengine - INFO - Epoch(train) [10][3900/4111] lr: 1.0006e-05 eta: 0:02:29 time: 0.6787 data_time: 0.5390 memory: 6318 loss: 0.1523 +2023/06/05 12:36:41 - mmengine - INFO - Epoch(train) [10][4000/4111] lr: 1.0002e-05 eta: 0:01:18 time: 0.6802 data_time: 0.5397 memory: 6318 loss: 0.1713 +2023/06/05 12:36:41 - mmengine - INFO - Exp name: resnet50_2xb256_all_1m_lr1e-4_aug_1e-1_20230605_042304 +2023/06/05 12:37:53 - mmengine - INFO - Epoch(train) [10][4100/4111] lr: 1.0000e-05 eta: 0:00:07 time: 0.6985 data_time: 0.5588 memory: 6318 loss: 0.1691 +2023/06/05 12:38:00 - mmengine - INFO - Exp name: resnet50_2xb256_all_1m_lr1e-4_aug_1e-1_20230605_042304 +2023/06/05 12:38:00 - mmengine - INFO - Saving checkpoint at 10 epochs +2023/06/05 12:38:43 - mmengine - INFO - Epoch(val) [10][100/119] eta: 0:00:06 time: 0.5867 data_time: 0.4975 memory: 6318 +2023/06/05 12:39:09 - mmengine - INFO - Epoch(val) [10][119/119] accuracy/top1: 94.7007 data_time: 0.3315 time: 0.4218 diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/20230605_042304.json b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/20230605_042304.json new file mode 100644 index 0000000000000000000000000000000000000000..a6de6d26a15177c3856eb62ac93cc96c4eaebd61 --- /dev/null +++ b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/20230605_042304.json @@ -0,0 +1,420 @@ +{"lr": 9.999871217862936e-05, "data_time": 0.5597489595413208, "loss": 0.6434031903743744, "time": 0.7016031503677368, "epoch": 1, "memory": 9436, "step": 100} +{"lr": 9.999479662539605e-05, "data_time": 1.1453526020050049, "loss": 0.5886454939842224, "time": 1.2975964784622191, "epoch": 1, "memory": 6319, "step": 200} +{"lr": 9.99882534375654e-05, "data_time": 0.5219174861907959, "loss": 0.5548661112785339, "time": 0.6638822317123413, "epoch": 1, "memory": 6319, "step": 300} +{"lr": 9.997908299725075e-05, "data_time": 0.5151443719863892, "loss": 0.5156890898942947, "time": 0.6636269569396973, "epoch": 1, "memory": 6319, "step": 400} +{"lr": 9.996728583999318e-05, "data_time": 0.4860126256942749, "loss": 0.4935474991798401, "time": 0.6272287130355835, "epoch": 1, "memory": 6319, "step": 500} +{"lr": 9.995286265473058e-05, "data_time": 0.5246894598007202, "loss": 0.467021507024765, "time": 0.6771746873855591, "epoch": 1, "memory": 6319, "step": 600} +{"lr": 9.993581428375725e-05, "data_time": 0.5264055728912354, "loss": 0.4350478231906891, "time": 0.6668889284133911, "epoch": 1, "memory": 6319, "step": 700} +{"lr": 9.991614172267466e-05, "data_time": 0.588233494758606, "loss": 0.4326759189367294, "time": 0.7353369951248169, "epoch": 1, "memory": 6319, "step": 800} +{"lr": 9.989384612033355e-05, "data_time": 0.48908228874206544, "loss": 0.4017670780420303, "time": 0.6301378726959228, "epoch": 1, "memory": 6319, "step": 900} +{"lr": 9.986892877876658e-05, "data_time": 0.5072320222854614, "loss": 0.4035114675760269, "time": 0.6469413995742798, "epoch": 1, "memory": 6319, "step": 1000} +{"lr": 9.984139115311244e-05, "data_time": 0.5214408874511719, "loss": 0.38751137554645537, "time": 0.6600114822387695, "epoch": 1, "memory": 6319, "step": 1100} +{"lr": 9.981123485153081e-05, "data_time": 0.5415996551513672, "loss": 0.3898611694574356, "time": 0.6956294775009155, "epoch": 1, "memory": 6319, "step": 1200} +{"lr": 9.977846163510851e-05, "data_time": 0.5327706336975098, "loss": 0.3806368261575699, "time": 0.6729814052581787, "epoch": 1, "memory": 6319, "step": 1300} +{"lr": 9.974307341775653e-05, "data_time": 0.47383880615234375, "loss": 0.3725622326135635, "time": 0.6127111196517945, "epoch": 1, "memory": 6319, "step": 1400} +{"lr": 9.970507226609833e-05, "data_time": 0.5567919492721558, "loss": 0.3777932286262512, "time": 0.6967374324798584, "epoch": 1, "memory": 6319, "step": 1500} +{"lr": 9.966446039934934e-05, "data_time": 0.5160897493362426, "loss": 0.3588222533464432, "time": 0.655788516998291, "epoch": 1, "memory": 6319, "step": 1600} +{"lr": 9.962124018918683e-05, "data_time": 0.5992703199386596, "loss": 0.3490312427282333, "time": 0.7394388914108276, "epoch": 1, "memory": 6319, "step": 1700} +{"lr": 9.957541415961212e-05, "data_time": 0.5212093114852905, "loss": 0.33320796191692353, "time": 0.6608426332473755, "epoch": 1, "memory": 6319, "step": 1800} +{"lr": 9.952698498680267e-05, "data_time": 0.5000231504440308, "loss": 0.341569259762764, "time": 0.6402387380599975, "epoch": 1, "memory": 6319, "step": 1900} +{"lr": 9.947595549895583e-05, "data_time": 0.5169378280639648, "loss": 0.33452337682247163, "time": 0.6580249786376953, "epoch": 1, "memory": 6319, "step": 2000} +{"lr": 9.942232867612417e-05, "data_time": 0.5054732084274292, "loss": 0.35189352333545687, "time": 0.6466429233551025, "epoch": 1, "memory": 6319, "step": 2100} +{"lr": 9.93661076500406e-05, "data_time": 0.5036224365234375, "loss": 0.3288008451461792, "time": 0.6427456140518188, "epoch": 1, "memory": 6319, "step": 2200} +{"lr": 9.930729570393622e-05, "data_time": 0.5937491416931152, "loss": 0.32314945161342623, "time": 0.737432861328125, "epoch": 1, "memory": 6319, "step": 2300} +{"lr": 9.924589627234824e-05, "data_time": 0.5298749923706054, "loss": 0.317729064822197, "time": 0.6692728281021119, "epoch": 1, "memory": 6319, "step": 2400} +{"lr": 9.918191294091956e-05, "data_time": 0.608555817604065, "loss": 0.31226462721824644, "time": 0.7473854541778564, "epoch": 1, "memory": 6319, "step": 2500} +{"lr": 9.91153494461892e-05, "data_time": 0.5144747257232666, "loss": 0.3173118382692337, "time": 0.6541839361190795, "epoch": 1, "memory": 6319, "step": 2600} +{"lr": 9.904620967537422e-05, "data_time": 0.5705208063125611, "loss": 0.3288555830717087, "time": 0.7131868124008178, "epoch": 1, "memory": 6319, "step": 2700} +{"lr": 9.897449766614277e-05, "data_time": 0.49216086864471437, "loss": 0.3009959772229195, "time": 0.6318735599517822, "epoch": 1, "memory": 6319, "step": 2800} +{"lr": 9.890021760637804e-05, "data_time": 0.5271918296813964, "loss": 0.3104993671178818, "time": 0.667259693145752, "epoch": 1, "memory": 6319, "step": 2900} +{"lr": 9.882337383393409e-05, "data_time": 0.5319823980331421, "loss": 0.3096489369869232, "time": 0.6717957973480224, "epoch": 1, "memory": 6319, "step": 3000} +{"lr": 9.874397083638193e-05, "data_time": 0.7240642070770263, "loss": 0.3106270909309387, "time": 0.863601565361023, "epoch": 1, "memory": 6319, "step": 3100} +{"lr": 9.866201325074828e-05, "data_time": 0.5561637639999389, "loss": 0.30683611035346986, "time": 0.6951672554016113, "epoch": 1, "memory": 6319, "step": 3200} +{"lr": 9.857750586324393e-05, "data_time": 0.5105583190917968, "loss": 0.31323845982551574, "time": 0.6559603929519653, "epoch": 1, "memory": 6319, "step": 3300} +{"lr": 9.84904536089848e-05, "data_time": 0.5197272300720215, "loss": 0.29418307840824126, "time": 0.6593654870986938, "epoch": 1, "memory": 6319, "step": 3400} +{"lr": 9.840086157170351e-05, "data_time": 0.5684110879898071, "loss": 0.31018399596214297, "time": 0.7078615903854371, "epoch": 1, "memory": 6319, "step": 3500} +{"lr": 9.830873498345255e-05, "data_time": 0.5084229946136475, "loss": 0.29629366993904116, "time": 0.6479480504989624, "epoch": 1, "memory": 6319, "step": 3600} +{"lr": 9.821407922429864e-05, "data_time": 0.5765625, "loss": 0.31072622537612915, "time": 0.7166357278823853, "epoch": 1, "memory": 6319, "step": 3700} +{"lr": 9.811689982200877e-05, "data_time": 0.5262469053268433, "loss": 0.28984043300151824, "time": 0.6657505989074707, "epoch": 1, "memory": 6319, "step": 3800} +{"lr": 9.801720245172702e-05, "data_time": 0.521976089477539, "loss": 0.28857015669345853, "time": 0.6626749277114868, "epoch": 1, "memory": 6319, "step": 3900} +{"lr": 9.791499293564358e-05, "data_time": 0.6571306705474853, "loss": 0.27816821485757826, "time": 0.8035236597061157, "epoch": 1, "memory": 6319, "step": 4000} +{"lr": 9.781027724265439e-05, "data_time": 0.5350852727890014, "loss": 0.29500730335712433, "time": 0.6754218339920044, "epoch": 1, "memory": 6319, "step": 4100} +{"accuracy/top1": 80.60838317871094, "data_time": 0.341507623175613, "time": 0.432024621162094, "step": 1} +{"lr": 9.769111539369434e-05, "data_time": 0.5789459705352783, "loss": 0.2883513569831848, "time": 0.7328791379928589, "epoch": 2, "memory": 6318, "step": 4211} +{"lr": 9.758113190911623e-05, "data_time": 0.6202581644058227, "loss": 0.28709604442119596, "time": 0.7605458259582519, "epoch": 2, "memory": 6318, "step": 4311} +{"lr": 9.746866174465984e-05, "data_time": 0.5982344627380372, "loss": 0.28065086156129837, "time": 0.7375700950622559, "epoch": 2, "memory": 6318, "step": 4411} +{"lr": 9.735371146842894e-05, "data_time": 0.5156761884689331, "loss": 0.28615678250789645, "time": 0.6557076454162598, "epoch": 2, "memory": 6318, "step": 4511} +{"lr": 9.723628779336255e-05, "data_time": 0.5337749242782592, "loss": 0.2887206941843033, "time": 0.6786488056182861, "epoch": 2, "memory": 6318, "step": 4611} +{"lr": 9.711639757684265e-05, "data_time": 0.5172728538513184, "loss": 0.27893856167793274, "time": 0.657161021232605, "epoch": 2, "memory": 6318, "step": 4711} +{"lr": 9.699404782029395e-05, "data_time": 0.5163314342498779, "loss": 0.2543041467666626, "time": 0.6555690765380859, "epoch": 2, "memory": 6318, "step": 4811} +{"lr": 9.686924566877498e-05, "data_time": 0.5058374881744385, "loss": 0.2656920820474625, "time": 0.6456071615219117, "epoch": 2, "memory": 6318, "step": 4911} +{"lr": 9.674199841056078e-05, "data_time": 0.5168680429458619, "loss": 0.28078637421131136, "time": 0.6639743328094483, "epoch": 2, "memory": 6318, "step": 5011} +{"lr": 9.661231347671705e-05, "data_time": 0.5765825033187866, "loss": 0.2722261160612106, "time": 0.7164970636367798, "epoch": 2, "memory": 6318, "step": 5111} +{"lr": 9.64801984406668e-05, "data_time": 0.5434545040130615, "loss": 0.2801385968923569, "time": 0.6845673084259033, "epoch": 2, "memory": 6318, "step": 5211} +{"lr": 9.634566101774737e-05, "data_time": 0.6311676740646363, "loss": 0.2790187358856201, "time": 0.7720628023147583, "epoch": 2, "memory": 6318, "step": 5311} +{"lr": 9.62087090647604e-05, "data_time": 0.5963162422180176, "loss": 0.2938770279288292, "time": 0.735174012184143, "epoch": 2, "memory": 6318, "step": 5411} +{"lr": 9.606935057951247e-05, "data_time": 0.5663928508758544, "loss": 0.2679371178150177, "time": 0.7054675817489624, "epoch": 2, "memory": 6318, "step": 5511} +{"lr": 9.592759370034884e-05, "data_time": 0.5335516214370728, "loss": 0.2764046609401703, "time": 0.6748976945877075, "epoch": 2, "memory": 6318, "step": 5611} +{"lr": 9.578344670567721e-05, "data_time": 0.5049342632293701, "loss": 0.2669560417532921, "time": 0.6455360889434815, "epoch": 2, "memory": 6318, "step": 5711} +{"lr": 9.563691801348495e-05, "data_time": 0.544314193725586, "loss": 0.2562394917011261, "time": 0.6839758634567261, "epoch": 2, "memory": 6318, "step": 5811} +{"lr": 9.548801618084738e-05, "data_time": 0.6134768009185791, "loss": 0.2650818094611168, "time": 0.753722882270813, "epoch": 2, "memory": 6318, "step": 5911} +{"lr": 9.533674990342786e-05, "data_time": 0.5038453102111816, "loss": 0.2726507052779198, "time": 0.6431791067123414, "epoch": 2, "memory": 6318, "step": 6011} +{"lr": 9.518312801497029e-05, "data_time": 0.5513953924179077, "loss": 0.24824558347463607, "time": 0.6920597076416015, "epoch": 2, "memory": 6318, "step": 6111} +{"lr": 9.502715948678276e-05, "data_time": 0.5037060976028442, "loss": 0.2736586719751358, "time": 0.6437305688858033, "epoch": 2, "memory": 6318, "step": 6211} +{"lr": 9.486885342721427e-05, "data_time": 0.4852008819580078, "loss": 0.267157718539238, "time": 0.6385726690292358, "epoch": 2, "memory": 6318, "step": 6311} +{"lr": 9.470821908112202e-05, "data_time": 0.5302046298980713, "loss": 0.28201734721660615, "time": 0.66909339427948, "epoch": 2, "memory": 6318, "step": 6411} +{"lr": 9.454526582933214e-05, "data_time": 0.5160373687744141, "loss": 0.2813108071684837, "time": 0.655771017074585, "epoch": 2, "memory": 6318, "step": 6511} +{"lr": 9.438000318809176e-05, "data_time": 0.5991506576538086, "loss": 0.2600247323513031, "time": 0.7393239736557007, "epoch": 2, "memory": 6318, "step": 6611} +{"lr": 9.42124408085132e-05, "data_time": 0.5790022373199463, "loss": 0.2557353124022484, "time": 0.7185931444168091, "epoch": 2, "memory": 6318, "step": 6711} +{"lr": 9.404258847601036e-05, "data_time": 0.5132441282272339, "loss": 0.27334317564964294, "time": 0.6528227806091309, "epoch": 2, "memory": 6318, "step": 6811} +{"lr": 9.387045610972717e-05, "data_time": 0.5245428085327148, "loss": 0.2611568719148636, "time": 0.6647748231887818, "epoch": 2, "memory": 6318, "step": 6911} +{"lr": 9.369605376195838e-05, "data_time": 0.6749463558197022, "loss": 0.2527195826172829, "time": 0.814208197593689, "epoch": 2, "memory": 6318, "step": 7011} +{"lr": 9.351939161756269e-05, "data_time": 0.5157665491104126, "loss": 0.2518232688307762, "time": 0.655615234375, "epoch": 2, "memory": 6318, "step": 7111} +{"lr": 9.334047999336764e-05, "data_time": 0.5483863353729248, "loss": 0.2573984503746033, "time": 0.6868363142013549, "epoch": 2, "memory": 6318, "step": 7211} +{"lr": 9.315932933756752e-05, "data_time": 0.506188178062439, "loss": 0.24999005645513533, "time": 0.6465755224227905, "epoch": 2, "memory": 6318, "step": 7311} +{"lr": 9.297595022911287e-05, "data_time": 0.5768125534057618, "loss": 0.2576883599162102, "time": 0.7152302026748657, "epoch": 2, "memory": 6318, "step": 7411} +{"lr": 9.279035337709295e-05, "data_time": 0.5277339696884156, "loss": 0.23903103917837143, "time": 0.6676866769790649, "epoch": 2, "memory": 6318, "step": 7511} +{"lr": 9.260254962011016e-05, "data_time": 0.5923547744750977, "loss": 0.24638890773057937, "time": 0.7327769994735718, "epoch": 2, "memory": 6318, "step": 7611} +{"lr": 9.241254992564702e-05, "data_time": 0.5583524703979492, "loss": 0.23376426547765733, "time": 0.70789954662323, "epoch": 2, "memory": 6318, "step": 7711} +{"lr": 9.22203653894259e-05, "data_time": 0.5611158609390259, "loss": 0.23874622881412505, "time": 0.7019243240356445, "epoch": 2, "memory": 6318, "step": 7811} +{"lr": 9.202600723476098e-05, "data_time": 0.536197304725647, "loss": 0.25499413162469864, "time": 0.6808391094207764, "epoch": 2, "memory": 6318, "step": 7911} +{"lr": 9.182948681190268e-05, "data_time": 0.5797640323638916, "loss": 0.2351435124874115, "time": 0.7194280624389648, "epoch": 2, "memory": 6318, "step": 8011} +{"lr": 9.163081559737497e-05, "data_time": 0.513187575340271, "loss": 0.2341068908572197, "time": 0.6537068128585816, "epoch": 2, "memory": 6318, "step": 8111} +{"lr": 9.143000519330517e-05, "data_time": 0.481587553024292, "loss": 0.250732509791851, "time": 0.6221200466156006, "epoch": 2, "memory": 6318, "step": 8211} +{"accuracy/top1": 76.61817932128906, "data_time": 0.3263594190279643, "time": 0.41486122210820514, "step": 2} +{"lr": 9.120461478753376e-05, "data_time": 0.5121278762817383, "loss": 0.23507924228906632, "time": 0.652920937538147, "epoch": 3, "memory": 6318, "step": 8322} +{"lr": 9.099932932138134e-05, "data_time": 0.5536199331283569, "loss": 0.23975561708211898, "time": 0.692713737487793, "epoch": 3, "memory": 6318, "step": 8422} +{"lr": 9.079194154362181e-05, "data_time": 0.5148802995681763, "loss": 0.24038059413433074, "time": 0.6558120489120484, "epoch": 3, "memory": 6318, "step": 8522} +{"lr": 9.058246356541787e-05, "data_time": 0.4992228984832764, "loss": 0.23119065016508103, "time": 0.6465325593948364, "epoch": 3, "memory": 6318, "step": 8622} +{"lr": 9.037090761999699e-05, "data_time": 0.5260629892349243, "loss": 0.22607937157154084, "time": 0.6672863006591797, "epoch": 3, "memory": 6318, "step": 8722} +{"lr": 9.015728606193701e-05, "data_time": 0.5703656673431396, "loss": 0.2585846409201622, "time": 0.7101302862167358, "epoch": 3, "memory": 6318, "step": 8822} +{"lr": 8.994161136644488e-05, "data_time": 0.5367135763168335, "loss": 0.22107840478420257, "time": 0.6768431901931763, "epoch": 3, "memory": 6318, "step": 8922} +{"lr": 8.972389612862792e-05, "data_time": 0.4898061752319336, "loss": 0.22925304472446442, "time": 0.6304153680801392, "epoch": 3, "memory": 6318, "step": 9022} +{"lr": 8.950415306275833e-05, "data_time": 0.5377302885055542, "loss": 0.2453791007399559, "time": 0.6776947259902955, "epoch": 3, "memory": 6318, "step": 9122} +{"lr": 8.928239500153059e-05, "data_time": 0.4841548681259155, "loss": 0.22476376742124557, "time": 0.6232578039169312, "epoch": 3, "memory": 6318, "step": 9222} +{"lr": 8.905863489531232e-05, "data_time": 0.5413121223449707, "loss": 0.2316204234957695, "time": 0.6809134006500244, "epoch": 3, "memory": 6318, "step": 9322} +{"lr": 8.883288581138775e-05, "data_time": 0.5113520383834839, "loss": 0.22785344570875168, "time": 0.6517601490020752, "epoch": 3, "memory": 6318, "step": 9422} +{"lr": 8.86051609331947e-05, "data_time": 0.5493632078170776, "loss": 0.2144322007894516, "time": 0.6891978502273559, "epoch": 3, "memory": 6318, "step": 9522} +{"lr": 8.837547355955467e-05, "data_time": 0.5674475193023681, "loss": 0.25626387894153596, "time": 0.707038140296936, "epoch": 3, "memory": 6318, "step": 9622} +{"lr": 8.814383710389614e-05, "data_time": 0.5146476030349731, "loss": 0.22953106611967086, "time": 0.653994345664978, "epoch": 3, "memory": 6318, "step": 9722} +{"lr": 8.791026509347146e-05, "data_time": 0.4882441759109497, "loss": 0.2418530121445656, "time": 0.6278645753860473, "epoch": 3, "memory": 6318, "step": 9822} +{"lr": 8.767477116856659e-05, "data_time": 0.5018274307250976, "loss": 0.2370159775018692, "time": 0.6412617683410644, "epoch": 3, "memory": 6318, "step": 9922} +{"lr": 8.743736908170466e-05, "data_time": 0.6059984683990478, "loss": 0.24052680283784866, "time": 0.7454351186752319, "epoch": 3, "memory": 6318, "step": 10022} +{"lr": 8.719807269684305e-05, "data_time": 0.5599032878875733, "loss": 0.23884377032518386, "time": 0.6996105432510376, "epoch": 3, "memory": 6318, "step": 10122} +{"lr": 8.695689598856356e-05, "data_time": 0.5556319236755372, "loss": 0.22184799760580062, "time": 0.6952784776687622, "epoch": 3, "memory": 6318, "step": 10222} +{"lr": 8.67138530412559e-05, "data_time": 0.5482355833053589, "loss": 0.22685093879699708, "time": 0.6899597406387329, "epoch": 3, "memory": 6318, "step": 10322} +{"lr": 8.646895804829604e-05, "data_time": 0.5226839303970336, "loss": 0.24236670285463333, "time": 0.6623435974121094, "epoch": 3, "memory": 6318, "step": 10422} +{"lr": 8.622222531121664e-05, "data_time": 0.5609231948852539, "loss": 0.2246718943119049, "time": 0.700794267654419, "epoch": 3, "memory": 6318, "step": 10522} +{"lr": 8.597366923887211e-05, "data_time": 0.5277501344680786, "loss": 0.23577700108289718, "time": 0.6677085638046265, "epoch": 3, "memory": 6318, "step": 10622} +{"lr": 8.57233043465972e-05, "data_time": 0.5971243143081665, "loss": 0.22813340425491332, "time": 0.7357851505279541, "epoch": 3, "memory": 6318, "step": 10722} +{"lr": 8.547114525535937e-05, "data_time": 0.5492791891098022, "loss": 0.22657681703567506, "time": 0.6891911029815674, "epoch": 3, "memory": 6318, "step": 10822} +{"lr": 8.521720669090474e-05, "data_time": 0.5139815807342529, "loss": 0.2166411831974983, "time": 0.6603597402572632, "epoch": 3, "memory": 6318, "step": 10922} +{"lr": 8.496150348289821e-05, "data_time": 0.5343125820159912, "loss": 0.2255546674132347, "time": 0.6739606142044068, "epoch": 3, "memory": 6318, "step": 11022} +{"lr": 8.470405056405736e-05, "data_time": 0.5242126226425171, "loss": 0.2221732333302498, "time": 0.6729058504104615, "epoch": 3, "memory": 6318, "step": 11122} +{"lr": 8.44448629692807e-05, "data_time": 0.5514846563339233, "loss": 0.2273815467953682, "time": 0.6917582511901855, "epoch": 3, "memory": 6318, "step": 11222} +{"lr": 8.418395583476941e-05, "data_time": 0.49566290378570554, "loss": 0.24244913905858995, "time": 0.6403186798095704, "epoch": 3, "memory": 6318, "step": 11322} +{"lr": 8.392134439714336e-05, "data_time": 0.5313705205917358, "loss": 0.22687100023031234, "time": 0.6714829206466675, "epoch": 3, "memory": 6318, "step": 11422} +{"lr": 8.365704399255133e-05, "data_time": 0.5148557424545288, "loss": 0.23275596052408218, "time": 0.6542460441589355, "epoch": 3, "memory": 6318, "step": 11522} +{"lr": 8.339107005577562e-05, "data_time": 0.6533173322677612, "loss": 0.2137850731611252, "time": 0.7933687686920166, "epoch": 3, "memory": 6318, "step": 11622} +{"lr": 8.31234381193305e-05, "data_time": 0.5868905782699585, "loss": 0.20853554606437683, "time": 0.7260604619979858, "epoch": 3, "memory": 6318, "step": 11722} +{"lr": 8.285416381255504e-05, "data_time": 0.5618477106094361, "loss": 0.20987340956926345, "time": 0.7015027523040771, "epoch": 3, "memory": 6318, "step": 11822} +{"lr": 8.258326286070047e-05, "data_time": 0.58131103515625, "loss": 0.22143910080194473, "time": 0.7206623792648316, "epoch": 3, "memory": 6318, "step": 11922} +{"lr": 8.231075108401202e-05, "data_time": 0.5419651508331299, "loss": 0.22237882316112517, "time": 0.6813149929046631, "epoch": 3, "memory": 6318, "step": 12022} +{"lr": 8.203664439680493e-05, "data_time": 0.5456916570663453, "loss": 0.2083484187722206, "time": 0.685080885887146, "epoch": 3, "memory": 6318, "step": 12122} +{"lr": 8.176095880653475e-05, "data_time": 0.5096866130828858, "loss": 0.20996739864349365, "time": 0.6498157262802124, "epoch": 3, "memory": 6318, "step": 12222} +{"lr": 8.148371041286308e-05, "data_time": 0.4793927907943726, "loss": 0.208797886967659, "time": 0.618826937675476, "epoch": 3, "memory": 6318, "step": 12322} +{"accuracy/top1": 82.17401123046875, "data_time": 0.3333737671375275, "time": 0.42224480708440143, "step": 3} +{"lr": 8.117415423358443e-05, "data_time": 0.5500046253204346, "loss": 0.20664273351430892, "time": 0.6890757322311402, "epoch": 4, "memory": 6318, "step": 12433} +{"lr": 8.089366155759266e-05, "data_time": 0.5329584598541259, "loss": 0.2161662459373474, "time": 0.6718969583511353, "epoch": 4, "memory": 6318, "step": 12533} +{"lr": 8.061165672717273e-05, "data_time": 0.5733454704284668, "loss": 0.21466794312000276, "time": 0.7122225761413574, "epoch": 4, "memory": 6318, "step": 12633} +{"lr": 8.032815621102116e-05, "data_time": 0.8412500619888306, "loss": 0.20947342216968537, "time": 0.9806801557540894, "epoch": 4, "memory": 6318, "step": 12733} +{"lr": 8.004317656518042e-05, "data_time": 0.6573879241943359, "loss": 0.2101093128323555, "time": 0.7969398498535156, "epoch": 4, "memory": 6318, "step": 12833} +{"lr": 7.975673443207223e-05, "data_time": 3.6199007987976075, "loss": 0.19993355125188828, "time": 3.7586714744567873, "epoch": 4, "memory": 6318, "step": 12933} +{"lr": 7.946884653952537e-05, "data_time": 0.7938543319702148, "loss": 0.205340938270092, "time": 0.9326043128967285, "epoch": 4, "memory": 6318, "step": 13033} +{"lr": 7.91795296997992e-05, "data_time": 0.5147840738296509, "loss": 0.23342226445674896, "time": 0.6549011707305908, "epoch": 4, "memory": 6318, "step": 13133} +{"lr": 7.888880080860151e-05, "data_time": 0.5238473176956177, "loss": 0.20029883682727814, "time": 0.6626713752746582, "epoch": 4, "memory": 6318, "step": 13233} +{"lr": 7.859667684410205e-05, "data_time": 0.5417109489440918, "loss": 0.21394192576408386, "time": 0.6817146301269531, "epoch": 4, "memory": 6318, "step": 13333} +{"lr": 7.830317486594082e-05, "data_time": 0.6320316553115845, "loss": 0.220724980533123, "time": 0.7733464956283569, "epoch": 4, "memory": 6318, "step": 13433} +{"lr": 7.800831201423207e-05, "data_time": 0.5526538848876953, "loss": 0.20695157945156098, "time": 0.6918729066848754, "epoch": 4, "memory": 6318, "step": 13533} +{"lr": 7.771210550856309e-05, "data_time": 0.5588021278381348, "loss": 0.20560887157917024, "time": 0.6995813131332398, "epoch": 4, "memory": 6318, "step": 13633} +{"lr": 7.741457264698884e-05, "data_time": 0.5980195760726928, "loss": 0.20007935613393785, "time": 0.7375678539276123, "epoch": 4, "memory": 6318, "step": 13733} +{"lr": 7.711573080502147e-05, "data_time": 0.7418617486953736, "loss": 0.205252206325531, "time": 0.8816857099533081, "epoch": 4, "memory": 6318, "step": 13833} +{"lr": 7.681559743461609e-05, "data_time": 0.5750361204147338, "loss": 0.199148528277874, "time": 0.7148786306381225, "epoch": 4, "memory": 6318, "step": 13933} +{"lr": 7.651419006315099e-05, "data_time": 0.5282118320465088, "loss": 0.22205454707145691, "time": 0.6676740169525146, "epoch": 4, "memory": 6318, "step": 14033} +{"lr": 7.621152629240453e-05, "data_time": 0.5356722831726074, "loss": 0.21486999541521073, "time": 0.6752888441085816, "epoch": 4, "memory": 6318, "step": 14133} +{"lr": 7.590762379752719e-05, "data_time": 0.5064670085906983, "loss": 0.21178832054138183, "time": 0.6480307102203369, "epoch": 4, "memory": 6318, "step": 14233} +{"lr": 7.560250032600911e-05, "data_time": 0.5267542362213135, "loss": 0.22033770978450776, "time": 0.6750738859176636, "epoch": 4, "memory": 6318, "step": 14333} +{"lr": 7.52961736966437e-05, "data_time": 0.5785313367843627, "loss": 0.20205722451210023, "time": 0.719299030303955, "epoch": 4, "memory": 6318, "step": 14433} +{"lr": 7.498866179848732e-05, "data_time": 0.5726711750030518, "loss": 0.2077656090259552, "time": 0.715662670135498, "epoch": 4, "memory": 6318, "step": 14533} +{"lr": 7.467998258981415e-05, "data_time": 0.5317749738693237, "loss": 0.19386776238679887, "time": 0.671155309677124, "epoch": 4, "memory": 6318, "step": 14633} +{"lr": 7.437015409706804e-05, "data_time": 0.6084056377410889, "loss": 0.1978773444890976, "time": 0.7483431100845337, "epoch": 4, "memory": 6318, "step": 14733} +{"lr": 7.405919441380929e-05, "data_time": 0.5344118595123291, "loss": 0.21792005449533464, "time": 0.6741207599639892, "epoch": 4, "memory": 6318, "step": 14833} +{"lr": 7.374712169965821e-05, "data_time": 0.5174609422683716, "loss": 0.20713909566402436, "time": 0.6584647178649903, "epoch": 4, "memory": 6318, "step": 14933} +{"lr": 7.343395417923461e-05, "data_time": 0.5622954845428467, "loss": 0.2105598047375679, "time": 0.7038269519805909, "epoch": 4, "memory": 6318, "step": 15033} +{"lr": 7.311971014109342e-05, "data_time": 0.5359824895858765, "loss": 0.19901027977466584, "time": 0.6760349750518799, "epoch": 4, "memory": 6318, "step": 15133} +{"lr": 7.280440793665682e-05, "data_time": 0.914686918258667, "loss": 0.2062069669365883, "time": 1.0543843507766724, "epoch": 4, "memory": 6318, "step": 15233} +{"lr": 7.248806597914234e-05, "data_time": 0.5441090822219848, "loss": 0.19911188036203384, "time": 0.683473777770996, "epoch": 4, "memory": 6318, "step": 15333} +{"lr": 7.217070274248765e-05, "data_time": 0.5418741464614868, "loss": 0.1903040498495102, "time": 0.6813448667526245, "epoch": 4, "memory": 6318, "step": 15433} +{"lr": 7.185233676027203e-05, "data_time": 0.5426470994949341, "loss": 0.21562436372041702, "time": 0.6836632251739502, "epoch": 4, "memory": 6318, "step": 15533} +{"lr": 7.153298662463338e-05, "data_time": 0.5512782335281372, "loss": 0.2175891324877739, "time": 0.6904650449752807, "epoch": 4, "memory": 6318, "step": 15633} +{"lr": 7.121267098518291e-05, "data_time": 0.555422043800354, "loss": 0.20827249586582183, "time": 0.6948675394058228, "epoch": 4, "memory": 6318, "step": 15733} +{"lr": 7.089140854791618e-05, "data_time": 0.7726138591766357, "loss": 0.21206145882606506, "time": 0.914221215248108, "epoch": 4, "memory": 6318, "step": 15833} +{"lr": 7.056921807412025e-05, "data_time": 0.581702995300293, "loss": 0.20164531767368316, "time": 0.7231520652770996, "epoch": 4, "memory": 6318, "step": 15933} +{"lr": 7.024611837927802e-05, "data_time": 0.5392264127731323, "loss": 0.2005120888352394, "time": 0.6925257682800293, "epoch": 4, "memory": 6318, "step": 16033} +{"lr": 6.992212833196998e-05, "data_time": 0.5013979196548461, "loss": 0.21203254908323288, "time": 0.6423643589019775, "epoch": 4, "memory": 6318, "step": 16133} +{"lr": 6.959726685277174e-05, "data_time": 0.5449551343917847, "loss": 0.1945815697312355, "time": 0.6863574028015137, "epoch": 4, "memory": 6318, "step": 16233} +{"lr": 6.927155291314949e-05, "data_time": 0.5248435258865356, "loss": 0.20592499673366546, "time": 0.6640942096710205, "epoch": 4, "memory": 6318, "step": 16333} +{"lr": 6.894500553435204e-05, "data_time": 0.5502176761627198, "loss": 0.21122395545244216, "time": 0.6902376413345337, "epoch": 4, "memory": 6318, "step": 16433} +{"accuracy/top1": 87.56930541992188, "data_time": 0.3289952198664347, "time": 0.41643391648928324, "step": 4} +{"lr": 6.858158509719903e-05, "data_time": 0.6103754281997681, "loss": 0.19654458463191987, "time": 0.7525153160095215, "epoch": 5, "memory": 6318, "step": 16544} +{"lr": 6.825334178880171e-05, "data_time": 0.5068995714187622, "loss": 0.19921637028455735, "time": 0.6520355224609375, "epoch": 5, "memory": 6318, "step": 16644} +{"lr": 6.792432450336645e-05, "data_time": 0.5620324850082398, "loss": 0.20516614019870758, "time": 0.7025948286056518, "epoch": 5, "memory": 6318, "step": 16744} +{"lr": 6.759455245505286e-05, "data_time": 0.5227182865142822, "loss": 0.1997493878006935, "time": 0.6678563356399536, "epoch": 5, "memory": 6318, "step": 16844} +{"lr": 6.726404490209753e-05, "data_time": 0.7072115182876587, "loss": 0.174032524228096, "time": 0.8474693059921264, "epoch": 5, "memory": 6318, "step": 16944} +{"lr": 6.693282114568972e-05, "data_time": 0.5498621463775635, "loss": 0.18071813136339188, "time": 0.6907564640045166, "epoch": 5, "memory": 6318, "step": 17044} +{"lr": 6.660090052884388e-05, "data_time": 0.49100098609924314, "loss": 0.19130819737911225, "time": 0.6312556743621827, "epoch": 5, "memory": 6318, "step": 17144} +{"lr": 6.626830243527016e-05, "data_time": 0.5211453914642334, "loss": 0.20275127589702607, "time": 0.66148362159729, "epoch": 5, "memory": 6318, "step": 17244} +{"lr": 6.593504628824236e-05, "data_time": 0.5267242193222046, "loss": 0.18796692788600922, "time": 0.6666906118392945, "epoch": 5, "memory": 6318, "step": 17344} +{"lr": 6.560115154946387e-05, "data_time": 0.4699209451675415, "loss": 0.19421417266130447, "time": 0.660755729675293, "epoch": 5, "memory": 6318, "step": 17444} +{"lr": 6.526663771793081e-05, "data_time": 0.5860954999923706, "loss": 0.1862255036830902, "time": 0.7251555681228637, "epoch": 5, "memory": 6318, "step": 17544} +{"lr": 6.493152432879363e-05, "data_time": 0.5247564077377319, "loss": 0.18504683375358583, "time": 0.6648433446884155, "epoch": 5, "memory": 6318, "step": 17644} +{"lr": 6.459583095221596e-05, "data_time": 0.5299063920974731, "loss": 0.18796184808015823, "time": 0.6706902265548706, "epoch": 5, "memory": 6318, "step": 17744} +{"lr": 6.425957719223197e-05, "data_time": 0.5014869689941406, "loss": 0.1826095849275589, "time": 0.6518988609313965, "epoch": 5, "memory": 6318, "step": 17844} +{"lr": 6.392278268560153e-05, "data_time": 0.594998049736023, "loss": 0.18176030963659287, "time": 0.7340277671813965, "epoch": 5, "memory": 6318, "step": 17944} +{"lr": 6.358546710066316e-05, "data_time": 0.5260184049606323, "loss": 0.2019784778356552, "time": 0.6648510217666626, "epoch": 5, "memory": 6318, "step": 18044} +{"lr": 6.324765013618573e-05, "data_time": 0.5347894430160522, "loss": 0.2267347291111946, "time": 0.683300518989563, "epoch": 5, "memory": 6318, "step": 18144} +{"lr": 6.290935152021818e-05, "data_time": 0.5557122230529785, "loss": 0.18254578113555908, "time": 0.6968997716903687, "epoch": 5, "memory": 6318, "step": 18244} +{"lr": 6.2570591008937e-05, "data_time": 0.5397093296051025, "loss": 0.20010170340538025, "time": 0.6792788982391358, "epoch": 5, "memory": 6318, "step": 18344} +{"lr": 6.223138838549294e-05, "data_time": 0.5627518415451049, "loss": 0.20028055757284163, "time": 0.7036221742630004, "epoch": 5, "memory": 6318, "step": 18444} +{"lr": 6.18917634588553e-05, "data_time": 0.5519994258880615, "loss": 0.1819985553622246, "time": 0.6915643215179443, "epoch": 5, "memory": 6318, "step": 18544} +{"lr": 6.155173606265555e-05, "data_time": 0.549025583267212, "loss": 0.20035212188959123, "time": 0.6887681245803833, "epoch": 5, "memory": 6318, "step": 18644} +{"lr": 6.121132605402843e-05, "data_time": 0.579982852935791, "loss": 0.1834666386246681, "time": 0.7216654777526855, "epoch": 5, "memory": 6318, "step": 18744} +{"lr": 6.087055331245287e-05, "data_time": 0.5233453035354614, "loss": 0.2027045249938965, "time": 0.6631776094436646, "epoch": 5, "memory": 6318, "step": 18844} +{"lr": 6.05294377385911e-05, "data_time": 0.5548267841339112, "loss": 0.20322794914245607, "time": 0.6963297367095947, "epoch": 5, "memory": 6318, "step": 18944} +{"lr": 6.0187999253126355e-05, "data_time": 0.5397990226745606, "loss": 0.18461829870939256, "time": 0.6788439750671387, "epoch": 5, "memory": 6318, "step": 19044} +{"lr": 5.9846257795599195e-05, "data_time": 0.5467065095901489, "loss": 0.20155278593301773, "time": 0.6944790124893189, "epoch": 5, "memory": 6318, "step": 19144} +{"lr": 5.950423332324355e-05, "data_time": 0.5479861259460449, "loss": 0.19435943961143493, "time": 0.689105224609375, "epoch": 5, "memory": 6318, "step": 19244} +{"lr": 5.9161945809820855e-05, "data_time": 0.5530983209609985, "loss": 0.18167036026716232, "time": 0.6932096958160401, "epoch": 5, "memory": 6318, "step": 19344} +{"lr": 5.8819415244453975e-05, "data_time": 0.5245201110839843, "loss": 0.18412984013557435, "time": 0.6642180442810058, "epoch": 5, "memory": 6318, "step": 19444} +{"lr": 5.8476661630459405e-05, "data_time": 0.5391457796096801, "loss": 0.19481215327978135, "time": 0.6804267168045044, "epoch": 5, "memory": 6318, "step": 19544} +{"lr": 5.81337049841796e-05, "data_time": 0.582946491241455, "loss": 0.2007267326116562, "time": 0.7248592138290405, "epoch": 5, "memory": 6318, "step": 19644} +{"lr": 5.779056533381373e-05, "data_time": 0.5544168949127197, "loss": 0.19656536728143692, "time": 0.6942979574203492, "epoch": 5, "memory": 6318, "step": 19744} +{"lr": 5.744726271824817e-05, "data_time": 0.5061422109603881, "loss": 0.2001613974571228, "time": 0.6463444471359253, "epoch": 5, "memory": 6318, "step": 19844} +{"lr": 5.710381718588622e-05, "data_time": 0.5269783496856689, "loss": 0.18792343586683274, "time": 0.6664208889007568, "epoch": 5, "memory": 6318, "step": 19944} +{"lr": 5.676024879347741e-05, "data_time": 0.5109683752059937, "loss": 0.19335417598485946, "time": 0.6527363538742066, "epoch": 5, "memory": 6318, "step": 20044} +{"lr": 5.6416577604945966e-05, "data_time": 0.6091557502746582, "loss": 0.2001790001988411, "time": 0.7500738382339478, "epoch": 5, "memory": 6318, "step": 20144} +{"lr": 5.6072823690219385e-05, "data_time": 0.5595529556274415, "loss": 0.18933072239160537, "time": 0.6989166736602783, "epoch": 5, "memory": 6318, "step": 20244} +{"lr": 5.572900712405624e-05, "data_time": 0.508144211769104, "loss": 0.1762568861246109, "time": 0.6485060453414917, "epoch": 5, "memory": 6318, "step": 20344} +{"lr": 5.538514798487397e-05, "data_time": 0.6103987693786621, "loss": 0.18673183917999267, "time": 0.7505025148391724, "epoch": 5, "memory": 6318, "step": 20444} +{"lr": 5.5041266353576e-05, "data_time": 0.5456227779388427, "loss": 0.19972552955150605, "time": 0.6874054193496704, "epoch": 5, "memory": 6318, "step": 20544} +{"accuracy/top1": 90.79158782958984, "data_time": 0.361405078570048, "time": 0.44991439779599507, "step": 5} +{"lr": 5.4659555783035384e-05, "data_time": 0.5881729364395142, "loss": 0.19360779970884323, "time": 0.7296860694885254, "epoch": 6, "memory": 6318, "step": 20655} +{"lr": 5.431569258426336e-05, "data_time": 0.5668072938919068, "loss": 0.18631560802459718, "time": 0.7078919649124146, "epoch": 6, "memory": 6318, "step": 20755} +{"lr": 5.397186934810864e-05, "data_time": 0.47689697742462156, "loss": 0.20995801091194152, "time": 0.6162860155105591, "epoch": 6, "memory": 6318, "step": 20855} +{"lr": 5.362810615337815e-05, "data_time": 0.8281100511550903, "loss": 0.1943479910492897, "time": 0.968401575088501, "epoch": 6, "memory": 6318, "step": 20955} +{"lr": 5.328442307537233e-05, "data_time": 0.6247039556503295, "loss": 0.18478670567274094, "time": 0.7642682790756226, "epoch": 6, "memory": 6318, "step": 21055} +{"lr": 5.294084018471303e-05, "data_time": 0.5587382793426514, "loss": 0.1863896906375885, "time": 0.6981104850769043, "epoch": 6, "memory": 6318, "step": 21155} +{"lr": 5.259737754617132e-05, "data_time": 0.500095009803772, "loss": 0.17880396991968156, "time": 0.6411489725112915, "epoch": 6, "memory": 6318, "step": 21255} +{"lr": 5.2254055217495525e-05, "data_time": 0.49918527603149415, "loss": 0.18837087452411652, "time": 0.6402879238128663, "epoch": 6, "memory": 6318, "step": 21355} +{"lr": 5.1910893248240283e-05, "data_time": 0.5623347997665405, "loss": 0.18812087029218674, "time": 0.7013975620269776, "epoch": 6, "memory": 6318, "step": 21455} +{"lr": 5.156791167859532e-05, "data_time": 0.5729370594024659, "loss": 0.16644491553306578, "time": 0.7145461082458496, "epoch": 6, "memory": 6318, "step": 21555} +{"lr": 5.122513053821533e-05, "data_time": 0.6027938842773437, "loss": 0.18924373239278794, "time": 0.7450995206832886, "epoch": 6, "memory": 6318, "step": 21655} +{"lr": 5.0882569845050173e-05, "data_time": 0.6844998359680176, "loss": 0.19755019694566728, "time": 0.8237540006637574, "epoch": 6, "memory": 6318, "step": 21755} +{"lr": 5.054024960417592e-05, "data_time": 0.5399816513061524, "loss": 0.2102901577949524, "time": 0.6798722505569458, "epoch": 6, "memory": 6318, "step": 21855} +{"lr": 5.0198189806626494e-05, "data_time": 0.5544996738433838, "loss": 0.18910381197929382, "time": 0.6935258388519288, "epoch": 6, "memory": 6318, "step": 21955} +{"lr": 4.985641042822647e-05, "data_time": 0.5247689962387085, "loss": 0.187516987323761, "time": 0.6636535167694092, "epoch": 6, "memory": 6318, "step": 22055} +{"lr": 4.951493142842412e-05, "data_time": 0.5312407493591309, "loss": 0.1782834127545357, "time": 0.6708173036575318, "epoch": 6, "memory": 6318, "step": 22155} +{"lr": 4.917377274912603e-05, "data_time": 0.5415554761886596, "loss": 0.18463821411132814, "time": 0.6817080736160278, "epoch": 6, "memory": 6318, "step": 22255} +{"lr": 4.88329543135327e-05, "data_time": 0.543966817855835, "loss": 0.19569617211818696, "time": 0.684067964553833, "epoch": 6, "memory": 6318, "step": 22355} +{"lr": 4.849249602497467e-05, "data_time": 0.5453457593917846, "loss": 0.18015597462654115, "time": 0.6858213424682618, "epoch": 6, "memory": 6318, "step": 22455} +{"lr": 4.8152417765750456e-05, "data_time": 0.5377744436264038, "loss": 0.18413397967815398, "time": 0.676759910583496, "epoch": 6, "memory": 6318, "step": 22555} +{"lr": 4.7812739395965355e-05, "data_time": 0.6555386304855346, "loss": 0.16897769868373871, "time": 0.7971973657608032, "epoch": 6, "memory": 6318, "step": 22655} +{"lr": 4.74734807523717e-05, "data_time": 0.5349490165710449, "loss": 0.172246565669775, "time": 0.6760582685470581, "epoch": 6, "memory": 6318, "step": 22755} +{"lr": 4.7134661647210416e-05, "data_time": 0.5201940536499023, "loss": 0.1744161680340767, "time": 0.6606065034866333, "epoch": 6, "memory": 6318, "step": 22855} +{"lr": 4.679630186705388e-05, "data_time": 0.572269868850708, "loss": 0.17823325768113135, "time": 0.7128920793533325, "epoch": 6, "memory": 6318, "step": 22955} +{"lr": 4.6458421171650644e-05, "data_time": 0.556977105140686, "loss": 0.18836547285318375, "time": 0.6965638399124146, "epoch": 6, "memory": 6318, "step": 23055} +{"lr": 4.6121039292771295e-05, "data_time": 0.6068763971328736, "loss": 0.18693819493055344, "time": 0.7554102897644043, "epoch": 6, "memory": 6318, "step": 23155} +{"lr": 4.5784175933056197e-05, "data_time": 0.5251615762710571, "loss": 0.17846366316080092, "time": 0.6648310422897339, "epoch": 6, "memory": 6318, "step": 23255} +{"lr": 4.544785076486494e-05, "data_time": 0.5238264322280883, "loss": 0.1825462445616722, "time": 0.6644751787185669, "epoch": 6, "memory": 6318, "step": 23355} +{"lr": 4.5112083429127475e-05, "data_time": 0.5412717580795288, "loss": 0.1809942215681076, "time": 0.6811113119125366, "epoch": 6, "memory": 6318, "step": 23455} +{"lr": 4.477689353419706e-05, "data_time": 0.5270626068115234, "loss": 0.18884348273277282, "time": 0.6666532516479492, "epoch": 6, "memory": 6318, "step": 23555} +{"lr": 4.4442300654705266e-05, "data_time": 0.6008017778396606, "loss": 0.190585695207119, "time": 0.7420772552490235, "epoch": 6, "memory": 6318, "step": 23655} +{"lr": 4.410832433041866e-05, "data_time": 0.5090028524398804, "loss": 0.18598672747612, "time": 0.6488039493560791, "epoch": 6, "memory": 6318, "step": 23755} +{"lr": 4.377498406509792e-05, "data_time": 0.5628119945526123, "loss": 0.17237576544284822, "time": 0.7039343833923339, "epoch": 6, "memory": 6318, "step": 23855} +{"lr": 4.344229932535875e-05, "data_time": 0.5091280460357666, "loss": 0.18829593807458878, "time": 0.650333023071289, "epoch": 6, "memory": 6318, "step": 23955} +{"lr": 4.3110289539535115e-05, "data_time": 0.5777626276016236, "loss": 0.19435689896345137, "time": 0.7181173324584961, "epoch": 6, "memory": 6318, "step": 24055} +{"lr": 4.2778974096544436e-05, "data_time": 0.5667340755462646, "loss": 0.17597020864486695, "time": 0.7064597845077515, "epoch": 6, "memory": 6318, "step": 24155} +{"lr": 4.2448372344755544e-05, "data_time": 0.5271841049194336, "loss": 0.18336962014436722, "time": 0.6673299551010132, "epoch": 6, "memory": 6318, "step": 24255} +{"lr": 4.211850359085871e-05, "data_time": 0.5520426988601684, "loss": 0.19305049180984496, "time": 0.6924070835113525, "epoch": 6, "memory": 6318, "step": 24355} +{"lr": 4.17893870987381e-05, "data_time": 0.5354948520660401, "loss": 0.16639301627874375, "time": 0.6757314682006836, "epoch": 6, "memory": 6318, "step": 24455} +{"lr": 4.146104208834679e-05, "data_time": 0.4852054834365845, "loss": 0.19581633508205415, "time": 0.6253786563873291, "epoch": 6, "memory": 6318, "step": 24555} +{"lr": 4.113348773458442e-05, "data_time": 0.5880526065826416, "loss": 0.2019881710410118, "time": 0.7272488355636597, "epoch": 6, "memory": 6318, "step": 24655} +{"accuracy/top1": 92.52437591552734, "data_time": 0.3327964166800181, "time": 0.42110156019528705, "step": 6} +{"lr": 4.0770851520835164e-05, "data_time": 0.46455338001251223, "loss": 0.1814323514699936, "time": 0.7192839860916138, "epoch": 7, "memory": 6318, "step": 24766} +{"lr": 4.044502815742039e-05, "data_time": 0.5248425960540771, "loss": 0.16935558319091798, "time": 0.6638788223266602, "epoch": 7, "memory": 6318, "step": 24866} +{"lr": 4.012005478445896e-05, "data_time": 0.5700514793395997, "loss": 0.18679607212543486, "time": 0.7094446659088135, "epoch": 7, "memory": 6318, "step": 24966} +{"lr": 3.9795950379951394e-05, "data_time": 0.5068717002868652, "loss": 0.1741093322634697, "time": 0.6475993394851685, "epoch": 7, "memory": 6318, "step": 25066} +{"lr": 3.947273387115169e-05, "data_time": 0.51826331615448, "loss": 0.19768147319555282, "time": 0.6584887266159057, "epoch": 7, "memory": 6318, "step": 25166} +{"lr": 3.9150424133462044e-05, "data_time": 0.504107666015625, "loss": 0.16667136996984483, "time": 0.64411301612854, "epoch": 7, "memory": 6318, "step": 25266} +{"lr": 3.882903998933037e-05, "data_time": 0.5777571201324463, "loss": 0.1736258640885353, "time": 0.7195043325424194, "epoch": 7, "memory": 6318, "step": 25366} +{"lr": 3.8508600207151264e-05, "data_time": 0.6008114576339721, "loss": 0.1911705791950226, "time": 0.7398993730545044, "epoch": 7, "memory": 6318, "step": 25466} +{"lr": 3.818912350016978e-05, "data_time": 0.5565430641174316, "loss": 0.1841244339942932, "time": 0.6966655015945434, "epoch": 7, "memory": 6318, "step": 25566} +{"lr": 3.787062852538883e-05, "data_time": 0.6452668190002442, "loss": 0.17789336740970613, "time": 0.7850041866302491, "epoch": 7, "memory": 6318, "step": 25666} +{"lr": 3.7553133882479346e-05, "data_time": 0.5507402420043945, "loss": 0.18023606538772582, "time": 0.6913224220275879, "epoch": 7, "memory": 6318, "step": 25766} +{"lr": 3.723665811269437e-05, "data_time": 0.5683861255645752, "loss": 0.18480758666992186, "time": 0.7086949110031128, "epoch": 7, "memory": 6318, "step": 25866} +{"lr": 3.6921219697786176e-05, "data_time": 0.5580015182495117, "loss": 0.17970621585845947, "time": 0.69829843044281, "epoch": 7, "memory": 6318, "step": 25966} +{"lr": 3.660683705892677e-05, "data_time": 0.5900080442428589, "loss": 0.1739823713898659, "time": 0.7292535543441773, "epoch": 7, "memory": 6318, "step": 26066} +{"lr": 3.6293528555632435e-05, "data_time": 0.5164843320846557, "loss": 0.19332556873559953, "time": 0.6585107326507569, "epoch": 7, "memory": 6318, "step": 26166} +{"lr": 3.598131248469134e-05, "data_time": 0.6044678211212158, "loss": 0.20842364579439163, "time": 0.7450649738311768, "epoch": 7, "memory": 6318, "step": 26266} +{"lr": 3.567020707909507e-05, "data_time": 0.5335680961608886, "loss": 0.16210654824972154, "time": 0.6747160434722901, "epoch": 7, "memory": 6318, "step": 26366} +{"lr": 3.5360230506973935e-05, "data_time": 0.579534363746643, "loss": 0.18372683674097062, "time": 0.720951509475708, "epoch": 7, "memory": 6318, "step": 26466} +{"lr": 3.5051400870535877e-05, "data_time": 0.5325656175613404, "loss": 0.20466922968626022, "time": 0.673722505569458, "epoch": 7, "memory": 6318, "step": 26566} +{"lr": 3.47437362050094e-05, "data_time": 0.5428436040878296, "loss": 0.15994197130203247, "time": 0.6824786424636841, "epoch": 7, "memory": 6318, "step": 26666} +{"lr": 3.443725447759023e-05, "data_time": 0.540584135055542, "loss": 0.16788210272789, "time": 0.6811728954315186, "epoch": 7, "memory": 6318, "step": 26766} +{"lr": 3.413197358639224e-05, "data_time": 0.5269869565963745, "loss": 0.17511895298957825, "time": 0.6666075468063355, "epoch": 7, "memory": 6318, "step": 26866} +{"lr": 3.382791135940194e-05, "data_time": 0.5360666751861572, "loss": 0.18723165094852448, "time": 0.6772321462631226, "epoch": 7, "memory": 6318, "step": 26966} +{"lr": 3.35250855534377e-05, "data_time": 0.53197340965271, "loss": 0.18540700823068618, "time": 0.6717989683151245, "epoch": 7, "memory": 6318, "step": 27066} +{"lr": 3.3223513853112476e-05, "data_time": 0.5130138397216797, "loss": 0.16946106925606727, "time": 0.6549042701721192, "epoch": 7, "memory": 6318, "step": 27166} +{"lr": 3.292321386980115e-05, "data_time": 0.5537219524383545, "loss": 0.17841366678476334, "time": 0.6954458236694336, "epoch": 7, "memory": 6318, "step": 27266} +{"lr": 3.262420314061217e-05, "data_time": 0.6147526741027832, "loss": 0.1982018008828163, "time": 0.7574407577514648, "epoch": 7, "memory": 6318, "step": 27366} +{"lr": 3.2326499127363195e-05, "data_time": 0.5746092319488525, "loss": 0.1772906720638275, "time": 0.7149666786193848, "epoch": 7, "memory": 6318, "step": 27466} +{"lr": 3.2030119215561613e-05, "data_time": 0.5163069009780884, "loss": 0.1781323418021202, "time": 0.6553100347518921, "epoch": 7, "memory": 6318, "step": 27566} +{"lr": 3.173508071338896e-05, "data_time": 0.517768669128418, "loss": 0.18461388498544692, "time": 0.6582580804824829, "epoch": 7, "memory": 6318, "step": 27666} +{"lr": 3.144140085069037e-05, "data_time": 0.5436537742614747, "loss": 0.18624600172042846, "time": 0.6835618734359741, "epoch": 7, "memory": 6318, "step": 27766} +{"lr": 3.114909677796814e-05, "data_time": 0.5386141777038574, "loss": 0.1818668231368065, "time": 0.6798846960067749, "epoch": 7, "memory": 6318, "step": 27866} +{"lr": 3.085818556538059e-05, "data_time": 0.5464069366455078, "loss": 0.18328240066766738, "time": 0.6873406410217285, "epoch": 7, "memory": 6318, "step": 27966} +{"lr": 3.056868420174455e-05, "data_time": 0.496645450592041, "loss": 0.19840313047170638, "time": 0.6369399547576904, "epoch": 7, "memory": 6318, "step": 28066} +{"lr": 3.028060959354392e-05, "data_time": 0.5428974866867066, "loss": 0.1919218197464943, "time": 0.6818622589111328, "epoch": 7, "memory": 6318, "step": 28166} +{"lr": 2.9993978563941876e-05, "data_time": 0.5451676368713378, "loss": 0.17275484502315522, "time": 0.6860358953475952, "epoch": 7, "memory": 6318, "step": 28266} +{"lr": 2.9708807851798587e-05, "data_time": 0.510317587852478, "loss": 0.2020210310816765, "time": 0.6525311946868897, "epoch": 7, "memory": 6318, "step": 28366} +{"lr": 2.9425114110693717e-05, "data_time": 0.49378926753997804, "loss": 0.1569451205432415, "time": 0.6336827039718628, "epoch": 7, "memory": 6318, "step": 28466} +{"lr": 2.9142913907953844e-05, "data_time": 0.5152788639068604, "loss": 0.175429967045784, "time": 0.6560280799865723, "epoch": 7, "memory": 6318, "step": 28566} +{"lr": 2.8862223723684912e-05, "data_time": 0.5369388818740845, "loss": 0.16513666808605193, "time": 0.681144118309021, "epoch": 7, "memory": 6318, "step": 28666} +{"lr": 2.8583059949809854e-05, "data_time": 0.5557304859161377, "loss": 0.17852320224046708, "time": 0.6952420234680176, "epoch": 7, "memory": 6318, "step": 28766} +{"accuracy/top1": 93.6083984375, "data_time": 0.33878278930981953, "time": 0.4279541730880737, "step": 7} +{"lr": 2.827499545208893e-05, "data_time": 0.3465777158737183, "loss": 0.18094997555017472, "time": 0.7000437259674073, "epoch": 8, "memory": 6318, "step": 28877} +{"lr": 2.7999105784987406e-05, "data_time": 0.39706437587738036, "loss": 0.19337422549724578, "time": 0.7807217597961426, "epoch": 8, "memory": 6318, "step": 28977} +{"lr": 2.772479293318711e-05, "data_time": 0.5286478996276855, "loss": 0.17107584774494172, "time": 0.6686571836471558, "epoch": 8, "memory": 6318, "step": 29077} +{"lr": 2.7452072916183603e-05, "data_time": 0.5333987474441528, "loss": 0.19867638498544693, "time": 0.6723517179489136, "epoch": 8, "memory": 6318, "step": 29177} +{"lr": 2.7180961660452883e-05, "data_time": 0.5628446102142334, "loss": 0.17373458296060562, "time": 0.701029372215271, "epoch": 8, "memory": 6318, "step": 29277} +{"lr": 2.6911474998521622e-05, "data_time": 0.627887487411499, "loss": 0.18528251200914383, "time": 0.7667783260345459, "epoch": 8, "memory": 6318, "step": 29377} +{"lr": 2.664362866804239e-05, "data_time": 0.6329079866409302, "loss": 0.1671963721513748, "time": 0.7732800960540771, "epoch": 8, "memory": 6318, "step": 29477} +{"lr": 2.637743831087459e-05, "data_time": 0.5496542453765869, "loss": 0.1769385442137718, "time": 0.6907904624938965, "epoch": 8, "memory": 6318, "step": 29577} +{"lr": 2.6112919472171114e-05, "data_time": 0.5835949420928955, "loss": 0.17195185422897338, "time": 0.725251579284668, "epoch": 8, "memory": 6318, "step": 29677} +{"lr": 2.5850087599470434e-05, "data_time": 0.5291482210159302, "loss": 0.17310436815023422, "time": 0.6693939447402955, "epoch": 8, "memory": 6318, "step": 29777} +{"lr": 2.5588958041794503e-05, "data_time": 0.6031557559967041, "loss": 0.15010228902101516, "time": 0.7433319091796875, "epoch": 8, "memory": 6318, "step": 29877} +{"lr": 2.5329546048752453e-05, "data_time": 0.5976364612579346, "loss": 0.1693695940077305, "time": 0.7385983228683471, "epoch": 8, "memory": 6318, "step": 29977} +{"lr": 2.5071866769649806e-05, "data_time": 0.5417542219161987, "loss": 0.18125804811716079, "time": 0.6841305017471313, "epoch": 8, "memory": 6318, "step": 30077} +{"lr": 2.4815935252604222e-05, "data_time": 0.5030296564102172, "loss": 0.1685178168118, "time": 0.6441235780715943, "epoch": 8, "memory": 6318, "step": 30177} +{"lr": 2.4561766443666384e-05, "data_time": 0.5139962434768677, "loss": 0.18587067425251008, "time": 0.6538240194320679, "epoch": 8, "memory": 6318, "step": 30277} +{"lr": 2.4309375185947164e-05, "data_time": 0.5446279764175415, "loss": 0.17585383355617523, "time": 0.6846682071685791, "epoch": 8, "memory": 6318, "step": 30377} +{"lr": 2.40587762187509e-05, "data_time": 0.5620025634765625, "loss": 0.16819631159305573, "time": 0.7023976087570191, "epoch": 8, "memory": 6318, "step": 30477} +{"lr": 2.3809984176714634e-05, "data_time": 0.5496367454528809, "loss": 0.1661178320646286, "time": 0.6897157430648804, "epoch": 8, "memory": 6318, "step": 30577} +{"lr": 2.3563013588953406e-05, "data_time": 0.5490324258804321, "loss": 0.16991938948631286, "time": 0.6881859302520752, "epoch": 8, "memory": 6318, "step": 30677} +{"lr": 2.331787887821182e-05, "data_time": 0.5657930374145508, "loss": 0.1782354325056076, "time": 0.7059160947799683, "epoch": 8, "memory": 6318, "step": 30777} +{"lr": 2.307459436002179e-05, "data_time": 0.5315551042556763, "loss": 0.18601717352867125, "time": 0.671811318397522, "epoch": 8, "memory": 6318, "step": 30877} +{"lr": 2.2833174241866527e-05, "data_time": 0.5765299320220947, "loss": 0.1732058271765709, "time": 0.7299833536148072, "epoch": 8, "memory": 6318, "step": 30977} +{"lr": 2.259363262235086e-05, "data_time": 0.5758986711502075, "loss": 0.18273567110300065, "time": 0.7165857076644897, "epoch": 8, "memory": 6318, "step": 31077} +{"lr": 2.2355983490377735e-05, "data_time": 0.5200314521789551, "loss": 0.1857576221227646, "time": 0.6596033096313476, "epoch": 8, "memory": 6318, "step": 31177} +{"lr": 2.2120240724331546e-05, "data_time": 0.5255465984344483, "loss": 0.17776864022016525, "time": 0.6645726680755615, "epoch": 8, "memory": 6318, "step": 31277} +{"lr": 2.1886418091267445e-05, "data_time": 0.5153661966323853, "loss": 0.1763664811849594, "time": 0.6579375743865967, "epoch": 8, "memory": 6318, "step": 31377} +{"lr": 2.1654529246107496e-05, "data_time": 0.5290921926498413, "loss": 0.18671398758888244, "time": 0.6683281660079956, "epoch": 8, "memory": 6318, "step": 31477} +{"lr": 2.1424587730843184e-05, "data_time": 0.5164127588272095, "loss": 0.16295654326677322, "time": 0.6617313623428345, "epoch": 8, "memory": 6318, "step": 31577} +{"lr": 2.1196606973744593e-05, "data_time": 0.5644389152526855, "loss": 0.1775700107216835, "time": 0.7042914390563965, "epoch": 8, "memory": 6318, "step": 31677} +{"lr": 2.0970600288576148e-05, "data_time": 0.5326156377792358, "loss": 0.18511711657047272, "time": 0.6814176082611084, "epoch": 8, "memory": 6318, "step": 31777} +{"lr": 2.0746580873819274e-05, "data_time": 0.4841681957244873, "loss": 0.16780608594417573, "time": 0.6248322486877441, "epoch": 8, "memory": 6318, "step": 31877} +{"lr": 2.0524561811901426e-05, "data_time": 0.5698666095733642, "loss": 0.1741199880838394, "time": 0.7101840496063232, "epoch": 8, "memory": 6318, "step": 31977} +{"lr": 2.0304556068432268e-05, "data_time": 0.570238733291626, "loss": 0.15130266547203064, "time": 0.7095555305480957, "epoch": 8, "memory": 6318, "step": 32077} +{"lr": 2.0086576491446333e-05, "data_time": 0.47088942527770994, "loss": 0.18402259349822997, "time": 0.6107290029525757, "epoch": 8, "memory": 6318, "step": 32177} +{"lr": 1.987063581065291e-05, "data_time": 0.5076226949691772, "loss": 0.17769983410835266, "time": 0.6468212366104126, "epoch": 8, "memory": 6318, "step": 32277} +{"lr": 1.965674663669245e-05, "data_time": 0.4999563455581665, "loss": 0.19281096756458282, "time": 0.6388433694839477, "epoch": 8, "memory": 6318, "step": 32377} +{"lr": 1.9444921460400247e-05, "data_time": 0.3367817163467407, "loss": 0.1737343743443489, "time": 0.6393883228302002, "epoch": 8, "memory": 6318, "step": 32477} +{"lr": 1.923517265207697e-05, "data_time": 0.23358845710754395, "loss": 0.174641951918602, "time": 0.6786292552947998, "epoch": 8, "memory": 6318, "step": 32577} +{"lr": 1.9027512460766204e-05, "data_time": 0.3714548110961914, "loss": 0.1874246269464493, "time": 0.6521084308624268, "epoch": 8, "memory": 6318, "step": 32677} +{"lr": 1.8821953013539136e-05, "data_time": 0.432128381729126, "loss": 0.17116330415010453, "time": 0.6260940313339234, "epoch": 8, "memory": 6318, "step": 32777} +{"lr": 1.8618506314786433e-05, "data_time": 0.43123228549957277, "loss": 0.1739860475063324, "time": 0.653626537322998, "epoch": 8, "memory": 6318, "step": 32877} +{"accuracy/top1": 94.42927551269531, "data_time": 0.31908140977223715, "time": 0.40818392038345336, "step": 8} +{"lr": 1.8395169032542028e-05, "data_time": 0.3467705249786377, "loss": 0.17853446155786515, "time": 0.6419506788253784, "epoch": 9, "memory": 6318, "step": 32988} +{"lr": 1.819621906460573e-05, "data_time": 0.46077885627746584, "loss": 0.16322045177221298, "time": 0.6007961988449096, "epoch": 9, "memory": 6318, "step": 33088} +{"lr": 1.799941838715077e-05, "data_time": 0.7201919794082642, "loss": 0.16671231687068938, "time": 0.8737469434738159, "epoch": 9, "memory": 6318, "step": 33188} +{"lr": 1.7804778493067625e-05, "data_time": 0.31894209384918215, "loss": 0.17341336756944656, "time": 0.7355125188827515, "epoch": 9, "memory": 6318, "step": 33288} +{"lr": 1.761231074905996e-05, "data_time": 0.3927642583847046, "loss": 0.17762706875801088, "time": 0.7605992317199707, "epoch": 9, "memory": 6318, "step": 33388} +{"lr": 1.742202639498083e-05, "data_time": 0.16396536827087402, "loss": 0.17904257625341416, "time": 0.6477619647979737, "epoch": 9, "memory": 6318, "step": 33488} +{"lr": 1.7233936543176347e-05, "data_time": 0.04417450428009033, "loss": 0.17295984625816346, "time": 0.7338368415832519, "epoch": 9, "memory": 6318, "step": 33588} +{"lr": 1.704805217783665e-05, "data_time": 0.0008935689926147461, "loss": 0.18471588641405107, "time": 0.6868875741958618, "epoch": 9, "memory": 6318, "step": 33688} +{"lr": 1.68643841543545e-05, "data_time": 0.0012085437774658203, "loss": 0.17745250910520555, "time": 0.8213760614395141, "epoch": 9, "memory": 6318, "step": 33788} +{"lr": 1.6682943198691327e-05, "data_time": 0.5312901496887207, "loss": 0.1759701833128929, "time": 0.6814525127410889, "epoch": 9, "memory": 6318, "step": 33888} +{"lr": 1.6503739906750822e-05, "data_time": 0.27212936878204347, "loss": 0.17120040208101273, "time": 0.6951136350631714, "epoch": 9, "memory": 6318, "step": 33988} +{"lr": 1.6326784743760187e-05, "data_time": 0.0600679874420166, "loss": 0.16045070067048073, "time": 0.6472667694091797, "epoch": 9, "memory": 6318, "step": 34088} +{"lr": 1.6152088043659013e-05, "data_time": 0.16998941898345948, "loss": 0.16141234263777732, "time": 0.7316301584243774, "epoch": 9, "memory": 6318, "step": 34188} +{"lr": 1.59796600084957e-05, "data_time": 0.0019454479217529297, "loss": 0.16706182956695556, "time": 0.6471722602844239, "epoch": 9, "memory": 6318, "step": 34288} +{"lr": 1.5809510707831687e-05, "data_time": 0.10176880359649658, "loss": 0.17744836509227752, "time": 0.6969247579574585, "epoch": 9, "memory": 6318, "step": 34388} +{"lr": 1.5641650078153473e-05, "data_time": 0.12099258899688721, "loss": 0.2010004073381424, "time": 0.6634851217269897, "epoch": 9, "memory": 6318, "step": 34488} +{"lr": 1.547608792229235e-05, "data_time": 0.03756430149078369, "loss": 0.17726728916168213, "time": 0.7295563220977783, "epoch": 9, "memory": 6318, "step": 34588} +{"lr": 1.5312833908851794e-05, "data_time": 0.07615690231323242, "loss": 0.18076025247573851, "time": 0.7693311929702759, "epoch": 9, "memory": 6318, "step": 34688} +{"lr": 1.5151897571642958e-05, "data_time": 0.22081241607666016, "loss": 0.17004195898771285, "time": 0.7372802734375, "epoch": 9, "memory": 6318, "step": 34788} +{"lr": 1.4993288309127877e-05, "data_time": 0.5547281265258789, "loss": 0.16769114434719085, "time": 0.6940059185028076, "epoch": 9, "memory": 6318, "step": 34888} +{"lr": 1.4837015383870593e-05, "data_time": 0.6260907649993896, "loss": 0.1616027683019638, "time": 0.7661632299423218, "epoch": 9, "memory": 6318, "step": 34988} +{"lr": 1.4683087921996264e-05, "data_time": 0.5476421594619751, "loss": 0.17030781656503677, "time": 0.687248420715332, "epoch": 9, "memory": 6318, "step": 35088} +{"lr": 1.453151491265819e-05, "data_time": 0.6143709182739258, "loss": 0.1779827207326889, "time": 0.7559512615203857, "epoch": 9, "memory": 6318, "step": 35188} +{"lr": 1.4382305207512847e-05, "data_time": 0.5709915161132812, "loss": 0.15990345180034637, "time": 0.7118298292160035, "epoch": 9, "memory": 6318, "step": 35288} +{"lr": 1.423546752020298e-05, "data_time": 0.5250293016433716, "loss": 0.18098425567150117, "time": 0.6652384757995605, "epoch": 9, "memory": 6318, "step": 35388} +{"lr": 1.4091010425848762e-05, "data_time": 0.5211315393447876, "loss": 0.20388921201229096, "time": 0.6633057832717896, "epoch": 9, "memory": 6318, "step": 35488} +{"lr": 1.3948942360546966e-05, "data_time": 0.5101306676864624, "loss": 0.17343863919377328, "time": 0.6527750015258789, "epoch": 9, "memory": 6318, "step": 35588} +{"lr": 1.3809271620878346e-05, "data_time": 0.5618909597396851, "loss": 0.18072735220193864, "time": 0.7015467405319213, "epoch": 9, "memory": 6318, "step": 35688} +{"lr": 1.3672006363423091e-05, "data_time": 0.5551931619644165, "loss": 0.19147171825170517, "time": 0.6961599826812744, "epoch": 9, "memory": 6318, "step": 35788} +{"lr": 1.3537154604284566e-05, "data_time": 0.514067006111145, "loss": 0.18362155109643935, "time": 0.6545395135879517, "epoch": 9, "memory": 6318, "step": 35888} +{"lr": 1.3404724218621087e-05, "data_time": 0.5974668979644775, "loss": 0.16670264750719072, "time": 0.7365384340286255, "epoch": 9, "memory": 6318, "step": 35988} +{"lr": 1.3274722940186121e-05, "data_time": 0.5556386709213257, "loss": 0.17231397479772567, "time": 0.6957824230194092, "epoch": 9, "memory": 6318, "step": 36088} +{"lr": 1.3147158360876541e-05, "data_time": 0.5237772464752197, "loss": 0.18398341089487075, "time": 0.6643592357635498, "epoch": 9, "memory": 6318, "step": 36188} +{"lr": 1.3022037930289352e-05, "data_time": 0.5662123441696167, "loss": 0.18018409311771394, "time": 0.7056707620620728, "epoch": 9, "memory": 6318, "step": 36288} +{"lr": 1.2899368955286627e-05, "data_time": 0.4856699228286743, "loss": 0.17277174741029738, "time": 0.6262874364852905, "epoch": 9, "memory": 6318, "step": 36388} +{"lr": 1.2779158599568751e-05, "data_time": 0.6819154977798462, "loss": 0.16570268720388412, "time": 0.8214737892150878, "epoch": 9, "memory": 6318, "step": 36488} +{"lr": 1.2661413883256134e-05, "data_time": 0.5092743158340454, "loss": 0.16690452620387078, "time": 0.6505688428878784, "epoch": 9, "memory": 6318, "step": 36588} +{"lr": 1.2546141682479203e-05, "data_time": 1.088001561164856, "loss": 0.1649048961699009, "time": 1.2272021532058717, "epoch": 9, "memory": 6318, "step": 36688} +{"lr": 1.2433348728976846e-05, "data_time": 0.6861448287963867, "loss": 0.171642704308033, "time": 0.8261954545974731, "epoch": 9, "memory": 6318, "step": 36788} +{"lr": 1.2323041609703344e-05, "data_time": 0.6412292718887329, "loss": 0.15848518610000611, "time": 0.7812103271484375, "epoch": 9, "memory": 6318, "step": 36888} +{"lr": 1.2215226766443614e-05, "data_time": 0.5432866334915161, "loss": 0.17100584656000137, "time": 0.6845231771469116, "epoch": 9, "memory": 6318, "step": 36988} +{"accuracy/top1": 94.54512786865234, "data_time": 0.3506173054377238, "time": 0.4392131467660268, "step": 9} +{"lr": 1.209847850894437e-05, "data_time": 0.24326579570770263, "loss": 0.15971334725618364, "time": 0.6647909879684448, "epoch": 10, "memory": 6318, "step": 37099} +{"lr": 1.1995942849201055e-05, "data_time": 0.6583766460418701, "loss": 0.15596782565116882, "time": 0.7985041856765747, "epoch": 10, "memory": 6318, "step": 37199} +{"lr": 1.189591856759183e-05, "data_time": 0.5123385429382324, "loss": 0.1760866865515709, "time": 0.6541135787963868, "epoch": 10, "memory": 6318, "step": 37299} +{"lr": 1.1798411505398016e-05, "data_time": 0.5055289983749389, "loss": 0.1699337735772133, "time": 0.6463497638702392, "epoch": 10, "memory": 6318, "step": 37399} +{"lr": 1.1703427356898678e-05, "data_time": 0.864405345916748, "loss": 0.16508603394031524, "time": 1.0076769590377808, "epoch": 10, "memory": 6318, "step": 37499} +{"lr": 1.1610971669038223e-05, "data_time": 0.5758638143539428, "loss": 0.17306023538112641, "time": 0.7169496297836304, "epoch": 10, "memory": 6318, "step": 37599} +{"lr": 1.1521049841102396e-05, "data_time": 0.5187180995941162, "loss": 0.16726333349943162, "time": 0.6584716796875, "epoch": 10, "memory": 6318, "step": 37699} +{"lr": 1.1433667124402997e-05, "data_time": 0.5278132200241089, "loss": 0.1680594876408577, "time": 0.6698819637298584, "epoch": 10, "memory": 6318, "step": 37799} +{"lr": 1.1348828621971181e-05, "data_time": 0.49249792098999023, "loss": 0.15358172878623008, "time": 0.6334294319152832, "epoch": 10, "memory": 6318, "step": 37899} +{"lr": 1.1266539288259486e-05, "data_time": 0.5400561094284058, "loss": 0.16024808287620546, "time": 0.6853986024856568, "epoch": 10, "memory": 6318, "step": 37999} +{"lr": 1.1186803928852477e-05, "data_time": 0.5870069026947021, "loss": 0.16915135011076926, "time": 0.726749324798584, "epoch": 10, "memory": 6318, "step": 38099} +{"lr": 1.1109627200186125e-05, "data_time": 0.5290631532669068, "loss": 0.16700850576162338, "time": 0.6680032253265381, "epoch": 10, "memory": 6318, "step": 38199} +{"lr": 1.1035013609275835e-05, "data_time": 0.4958626747131348, "loss": 0.16667198538780212, "time": 0.6355530023574829, "epoch": 10, "memory": 6318, "step": 38299} +{"lr": 1.0962967513453298e-05, "data_time": 0.513212776184082, "loss": 0.18055977672338486, "time": 0.659446382522583, "epoch": 10, "memory": 6318, "step": 38399} +{"lr": 1.0893493120111999e-05, "data_time": 0.5574291229248047, "loss": 0.17115170061588286, "time": 0.6971151828765869, "epoch": 10, "memory": 6318, "step": 38499} +{"lr": 1.0826594486461509e-05, "data_time": 0.5870451927185059, "loss": 0.18780031204223632, "time": 0.7271421909332275, "epoch": 10, "memory": 6318, "step": 38599} +{"lr": 1.0762275519290557e-05, "data_time": 0.53290696144104, "loss": 0.1806088149547577, "time": 0.673856544494629, "epoch": 10, "memory": 6318, "step": 38699} +{"lr": 1.0700539974738862e-05, "data_time": 0.5583453893661499, "loss": 0.19364455938339234, "time": 0.6982730865478516, "epoch": 10, "memory": 6318, "step": 38799} +{"lr": 1.064139145807783e-05, "data_time": 0.6159309387207031, "loss": 0.18038648068904878, "time": 0.7574919700622559, "epoch": 10, "memory": 6318, "step": 38899} +{"lr": 1.0584833423499947e-05, "data_time": 0.5298036336898804, "loss": 0.1769822955131531, "time": 0.6711828708648682, "epoch": 10, "memory": 6318, "step": 38999} +{"lr": 1.0530869173917099e-05, "data_time": 0.5743390798568726, "loss": 0.17203499972820283, "time": 0.7146025419235229, "epoch": 10, "memory": 6318, "step": 39099} +{"lr": 1.0479501860767654e-05, "data_time": 0.603572940826416, "loss": 0.16231856718659401, "time": 0.7425681591033936, "epoch": 10, "memory": 6318, "step": 39199} +{"lr": 1.043073448383247e-05, "data_time": 0.547653317451477, "loss": 0.1617319703102112, "time": 0.6864744186401367, "epoch": 10, "memory": 6318, "step": 39299} +{"lr": 1.038456989105968e-05, "data_time": 0.5811430931091308, "loss": 0.1714554473757744, "time": 0.7210165739059449, "epoch": 10, "memory": 6318, "step": 39399} +{"lr": 1.034101077839838e-05, "data_time": 0.5828921079635621, "loss": 0.16311801671981813, "time": 0.7227906942367553, "epoch": 10, "memory": 6318, "step": 39499} +{"lr": 1.0300059689641194e-05, "data_time": 0.46419026851654055, "loss": 0.17278907746076583, "time": 0.6029969453811646, "epoch": 10, "memory": 6318, "step": 39599} +{"lr": 1.0261719016275683e-05, "data_time": 0.4624385118484497, "loss": 0.18718645274639129, "time": 0.6022111177444458, "epoch": 10, "memory": 6318, "step": 39699} +{"lr": 1.0225990997344757e-05, "data_time": 0.47218492031097414, "loss": 0.16667576581239701, "time": 0.6113409042358399, "epoch": 10, "memory": 6318, "step": 39799} +{"lr": 1.019287771931587e-05, "data_time": 0.5410808324813843, "loss": 0.18095844238996506, "time": 0.7533207893371582, "epoch": 10, "memory": 6318, "step": 39899} +{"lr": 1.0162381115959187e-05, "data_time": 0.1821967840194702, "loss": 0.18518355935811998, "time": 0.6399319410324097, "epoch": 10, "memory": 6318, "step": 39999} +{"lr": 1.0134502968234636e-05, "data_time": 0.15167815685272218, "loss": 0.15409796461462974, "time": 0.6462615013122559, "epoch": 10, "memory": 6318, "step": 40099} +{"lr": 1.0109244904187918e-05, "data_time": 0.6154161691665649, "loss": 0.16570380181074143, "time": 0.7540058612823486, "epoch": 10, "memory": 6318, "step": 40199} +{"lr": 1.0086608398855446e-05, "data_time": 0.4992237567901611, "loss": 0.16055154353380202, "time": 0.6389468669891357, "epoch": 10, "memory": 6318, "step": 40299} +{"lr": 1.0066594774178176e-05, "data_time": 0.5433523178100585, "loss": 0.17121364176273346, "time": 0.6839509725570678, "epoch": 10, "memory": 6318, "step": 40399} +{"lr": 1.0049205198924413e-05, "data_time": 0.5597763061523438, "loss": 0.16367753297090532, "time": 0.7084070920944214, "epoch": 10, "memory": 6318, "step": 40499} +{"lr": 1.0034440688621577e-05, "data_time": 0.5803892612457275, "loss": 0.1932917758822441, "time": 0.7189661026000976, "epoch": 10, "memory": 6318, "step": 40599} +{"lr": 1.0022302105496896e-05, "data_time": 0.5081663131713867, "loss": 0.16676972433924675, "time": 0.6487033843994141, "epoch": 10, "memory": 6318, "step": 40699} +{"lr": 1.0012790158427017e-05, "data_time": 0.5436817169189453, "loss": 0.1766459122300148, "time": 0.6827889680862427, "epoch": 10, "memory": 6318, "step": 40799} +{"lr": 1.000590540289662e-05, "data_time": 0.5389787673950195, "loss": 0.1522808313369751, "time": 0.6787208557128906, "epoch": 10, "memory": 6318, "step": 40899} +{"lr": 1.0001648240966046e-05, "data_time": 0.5397365808486938, "loss": 0.17131981998682022, "time": 0.680224347114563, "epoch": 10, "memory": 6318, "step": 40999} +{"lr": 1.0000018921247718e-05, "data_time": 0.558778715133667, "loss": 0.1690695308148861, "time": 0.6985341310501099, "epoch": 10, "memory": 6318, "step": 41099} +{"accuracy/top1": 94.70069885253906, "data_time": 0.3315279563268026, "time": 0.42175288597742716, "step": 10} diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/config.py b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/config.py new file mode 100644 index 0000000000000000000000000000000000000000..e1db61d967e52e39f7a4971468760d3ff016bef0 --- /dev/null +++ b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/config.py @@ -0,0 +1,366 @@ +optim_wrapper = dict( + optimizer=dict( + type='SGD', + lr=0.0001, + momentum=0.9, + weight_decay=0.0001, + _scope_='mmpretrain'), + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=False, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=512) +model = dict( + type='ImageClassifier', + backbone=dict( + frozen_stages=2, + type='ResNet', + depth=50, + num_stages=4, + out_indices=(3, ), + style='pytorch', + init_cfg=dict( + type='Pretrained', + checkpoint= + 'https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth', + prefix='backbone')), + neck=dict(type='GlobalAveragePooling'), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=2048, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + topk=1)) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/all1.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/all2.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/all3.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3fake8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/cc1m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3real8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV2-1-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/if-dpmsolver++-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = dict(type='Accuracy', topk=1) +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV2-1-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/if-dpmsolver++-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = dict(type='Accuracy', topk=1) +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1' diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/events.out.tfevents.1685910199.SH-IDC1-10-140-24-106.139935.0 b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/events.out.tfevents.1685910199.SH-IDC1-10-140-24-106.139935.0 new file mode 100644 index 0000000000000000000000000000000000000000..b169e03c5023953bf4d63c69cbf2ed71249d5036 --- /dev/null +++ b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/events.out.tfevents.1685910199.SH-IDC1-10-140-24-106.139935.0 @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:17a5c2c4235e3d451a8da0484314a5aeec475c90b6ddd059cb9d0a74f8d97a96 +size 5788703 diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/scalars.json b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/scalars.json new file mode 100644 index 0000000000000000000000000000000000000000..a6de6d26a15177c3856eb62ac93cc96c4eaebd61 --- /dev/null +++ b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/scalars.json @@ -0,0 +1,420 @@ +{"lr": 9.999871217862936e-05, "data_time": 0.5597489595413208, "loss": 0.6434031903743744, "time": 0.7016031503677368, "epoch": 1, "memory": 9436, "step": 100} +{"lr": 9.999479662539605e-05, "data_time": 1.1453526020050049, "loss": 0.5886454939842224, "time": 1.2975964784622191, "epoch": 1, "memory": 6319, "step": 200} +{"lr": 9.99882534375654e-05, "data_time": 0.5219174861907959, "loss": 0.5548661112785339, "time": 0.6638822317123413, "epoch": 1, "memory": 6319, "step": 300} +{"lr": 9.997908299725075e-05, "data_time": 0.5151443719863892, "loss": 0.5156890898942947, "time": 0.6636269569396973, "epoch": 1, "memory": 6319, "step": 400} +{"lr": 9.996728583999318e-05, "data_time": 0.4860126256942749, "loss": 0.4935474991798401, "time": 0.6272287130355835, "epoch": 1, "memory": 6319, "step": 500} +{"lr": 9.995286265473058e-05, "data_time": 0.5246894598007202, "loss": 0.467021507024765, "time": 0.6771746873855591, "epoch": 1, "memory": 6319, "step": 600} +{"lr": 9.993581428375725e-05, "data_time": 0.5264055728912354, "loss": 0.4350478231906891, "time": 0.6668889284133911, "epoch": 1, "memory": 6319, "step": 700} +{"lr": 9.991614172267466e-05, "data_time": 0.588233494758606, "loss": 0.4326759189367294, "time": 0.7353369951248169, "epoch": 1, "memory": 6319, "step": 800} +{"lr": 9.989384612033355e-05, "data_time": 0.48908228874206544, "loss": 0.4017670780420303, "time": 0.6301378726959228, "epoch": 1, "memory": 6319, "step": 900} +{"lr": 9.986892877876658e-05, "data_time": 0.5072320222854614, "loss": 0.4035114675760269, "time": 0.6469413995742798, "epoch": 1, "memory": 6319, "step": 1000} +{"lr": 9.984139115311244e-05, "data_time": 0.5214408874511719, "loss": 0.38751137554645537, "time": 0.6600114822387695, "epoch": 1, "memory": 6319, "step": 1100} +{"lr": 9.981123485153081e-05, "data_time": 0.5415996551513672, "loss": 0.3898611694574356, "time": 0.6956294775009155, "epoch": 1, "memory": 6319, "step": 1200} +{"lr": 9.977846163510851e-05, "data_time": 0.5327706336975098, "loss": 0.3806368261575699, "time": 0.6729814052581787, "epoch": 1, "memory": 6319, "step": 1300} +{"lr": 9.974307341775653e-05, "data_time": 0.47383880615234375, "loss": 0.3725622326135635, "time": 0.6127111196517945, "epoch": 1, "memory": 6319, "step": 1400} +{"lr": 9.970507226609833e-05, "data_time": 0.5567919492721558, "loss": 0.3777932286262512, "time": 0.6967374324798584, "epoch": 1, "memory": 6319, "step": 1500} +{"lr": 9.966446039934934e-05, "data_time": 0.5160897493362426, "loss": 0.3588222533464432, "time": 0.655788516998291, "epoch": 1, "memory": 6319, "step": 1600} +{"lr": 9.962124018918683e-05, "data_time": 0.5992703199386596, "loss": 0.3490312427282333, "time": 0.7394388914108276, "epoch": 1, "memory": 6319, "step": 1700} +{"lr": 9.957541415961212e-05, "data_time": 0.5212093114852905, "loss": 0.33320796191692353, "time": 0.6608426332473755, "epoch": 1, "memory": 6319, "step": 1800} +{"lr": 9.952698498680267e-05, "data_time": 0.5000231504440308, "loss": 0.341569259762764, "time": 0.6402387380599975, "epoch": 1, "memory": 6319, "step": 1900} +{"lr": 9.947595549895583e-05, "data_time": 0.5169378280639648, "loss": 0.33452337682247163, "time": 0.6580249786376953, "epoch": 1, "memory": 6319, "step": 2000} +{"lr": 9.942232867612417e-05, "data_time": 0.5054732084274292, "loss": 0.35189352333545687, "time": 0.6466429233551025, "epoch": 1, "memory": 6319, "step": 2100} +{"lr": 9.93661076500406e-05, "data_time": 0.5036224365234375, "loss": 0.3288008451461792, "time": 0.6427456140518188, "epoch": 1, "memory": 6319, "step": 2200} +{"lr": 9.930729570393622e-05, "data_time": 0.5937491416931152, "loss": 0.32314945161342623, "time": 0.737432861328125, "epoch": 1, "memory": 6319, "step": 2300} +{"lr": 9.924589627234824e-05, "data_time": 0.5298749923706054, "loss": 0.317729064822197, "time": 0.6692728281021119, "epoch": 1, "memory": 6319, "step": 2400} +{"lr": 9.918191294091956e-05, "data_time": 0.608555817604065, "loss": 0.31226462721824644, "time": 0.7473854541778564, "epoch": 1, "memory": 6319, "step": 2500} +{"lr": 9.91153494461892e-05, "data_time": 0.5144747257232666, "loss": 0.3173118382692337, "time": 0.6541839361190795, "epoch": 1, "memory": 6319, "step": 2600} +{"lr": 9.904620967537422e-05, "data_time": 0.5705208063125611, "loss": 0.3288555830717087, "time": 0.7131868124008178, "epoch": 1, "memory": 6319, "step": 2700} +{"lr": 9.897449766614277e-05, "data_time": 0.49216086864471437, "loss": 0.3009959772229195, "time": 0.6318735599517822, "epoch": 1, "memory": 6319, "step": 2800} +{"lr": 9.890021760637804e-05, "data_time": 0.5271918296813964, "loss": 0.3104993671178818, "time": 0.667259693145752, "epoch": 1, "memory": 6319, "step": 2900} +{"lr": 9.882337383393409e-05, "data_time": 0.5319823980331421, "loss": 0.3096489369869232, "time": 0.6717957973480224, "epoch": 1, "memory": 6319, "step": 3000} +{"lr": 9.874397083638193e-05, "data_time": 0.7240642070770263, "loss": 0.3106270909309387, "time": 0.863601565361023, "epoch": 1, "memory": 6319, "step": 3100} +{"lr": 9.866201325074828e-05, "data_time": 0.5561637639999389, "loss": 0.30683611035346986, "time": 0.6951672554016113, "epoch": 1, "memory": 6319, "step": 3200} +{"lr": 9.857750586324393e-05, "data_time": 0.5105583190917968, "loss": 0.31323845982551574, "time": 0.6559603929519653, "epoch": 1, "memory": 6319, "step": 3300} +{"lr": 9.84904536089848e-05, "data_time": 0.5197272300720215, "loss": 0.29418307840824126, "time": 0.6593654870986938, "epoch": 1, "memory": 6319, "step": 3400} +{"lr": 9.840086157170351e-05, "data_time": 0.5684110879898071, "loss": 0.31018399596214297, "time": 0.7078615903854371, "epoch": 1, "memory": 6319, "step": 3500} +{"lr": 9.830873498345255e-05, "data_time": 0.5084229946136475, "loss": 0.29629366993904116, "time": 0.6479480504989624, "epoch": 1, "memory": 6319, "step": 3600} +{"lr": 9.821407922429864e-05, "data_time": 0.5765625, "loss": 0.31072622537612915, "time": 0.7166357278823853, "epoch": 1, "memory": 6319, "step": 3700} +{"lr": 9.811689982200877e-05, "data_time": 0.5262469053268433, "loss": 0.28984043300151824, "time": 0.6657505989074707, "epoch": 1, "memory": 6319, "step": 3800} +{"lr": 9.801720245172702e-05, "data_time": 0.521976089477539, "loss": 0.28857015669345853, "time": 0.6626749277114868, "epoch": 1, "memory": 6319, "step": 3900} +{"lr": 9.791499293564358e-05, "data_time": 0.6571306705474853, "loss": 0.27816821485757826, "time": 0.8035236597061157, "epoch": 1, "memory": 6319, "step": 4000} +{"lr": 9.781027724265439e-05, "data_time": 0.5350852727890014, "loss": 0.29500730335712433, "time": 0.6754218339920044, "epoch": 1, "memory": 6319, "step": 4100} +{"accuracy/top1": 80.60838317871094, "data_time": 0.341507623175613, "time": 0.432024621162094, "step": 1} +{"lr": 9.769111539369434e-05, "data_time": 0.5789459705352783, "loss": 0.2883513569831848, "time": 0.7328791379928589, "epoch": 2, "memory": 6318, "step": 4211} +{"lr": 9.758113190911623e-05, "data_time": 0.6202581644058227, "loss": 0.28709604442119596, "time": 0.7605458259582519, "epoch": 2, "memory": 6318, "step": 4311} +{"lr": 9.746866174465984e-05, "data_time": 0.5982344627380372, "loss": 0.28065086156129837, "time": 0.7375700950622559, "epoch": 2, "memory": 6318, "step": 4411} +{"lr": 9.735371146842894e-05, "data_time": 0.5156761884689331, "loss": 0.28615678250789645, "time": 0.6557076454162598, "epoch": 2, "memory": 6318, "step": 4511} +{"lr": 9.723628779336255e-05, "data_time": 0.5337749242782592, "loss": 0.2887206941843033, "time": 0.6786488056182861, "epoch": 2, "memory": 6318, "step": 4611} +{"lr": 9.711639757684265e-05, "data_time": 0.5172728538513184, "loss": 0.27893856167793274, "time": 0.657161021232605, "epoch": 2, "memory": 6318, "step": 4711} +{"lr": 9.699404782029395e-05, "data_time": 0.5163314342498779, "loss": 0.2543041467666626, "time": 0.6555690765380859, "epoch": 2, "memory": 6318, "step": 4811} +{"lr": 9.686924566877498e-05, "data_time": 0.5058374881744385, "loss": 0.2656920820474625, "time": 0.6456071615219117, "epoch": 2, "memory": 6318, "step": 4911} +{"lr": 9.674199841056078e-05, "data_time": 0.5168680429458619, "loss": 0.28078637421131136, "time": 0.6639743328094483, "epoch": 2, "memory": 6318, "step": 5011} +{"lr": 9.661231347671705e-05, "data_time": 0.5765825033187866, "loss": 0.2722261160612106, "time": 0.7164970636367798, "epoch": 2, "memory": 6318, "step": 5111} +{"lr": 9.64801984406668e-05, "data_time": 0.5434545040130615, "loss": 0.2801385968923569, "time": 0.6845673084259033, "epoch": 2, "memory": 6318, "step": 5211} +{"lr": 9.634566101774737e-05, "data_time": 0.6311676740646363, "loss": 0.2790187358856201, "time": 0.7720628023147583, "epoch": 2, "memory": 6318, "step": 5311} +{"lr": 9.62087090647604e-05, "data_time": 0.5963162422180176, "loss": 0.2938770279288292, "time": 0.735174012184143, "epoch": 2, "memory": 6318, "step": 5411} +{"lr": 9.606935057951247e-05, "data_time": 0.5663928508758544, "loss": 0.2679371178150177, "time": 0.7054675817489624, "epoch": 2, "memory": 6318, "step": 5511} +{"lr": 9.592759370034884e-05, "data_time": 0.5335516214370728, "loss": 0.2764046609401703, "time": 0.6748976945877075, "epoch": 2, "memory": 6318, "step": 5611} +{"lr": 9.578344670567721e-05, "data_time": 0.5049342632293701, "loss": 0.2669560417532921, "time": 0.6455360889434815, "epoch": 2, "memory": 6318, "step": 5711} +{"lr": 9.563691801348495e-05, "data_time": 0.544314193725586, "loss": 0.2562394917011261, "time": 0.6839758634567261, "epoch": 2, "memory": 6318, "step": 5811} +{"lr": 9.548801618084738e-05, "data_time": 0.6134768009185791, "loss": 0.2650818094611168, "time": 0.753722882270813, "epoch": 2, "memory": 6318, "step": 5911} +{"lr": 9.533674990342786e-05, "data_time": 0.5038453102111816, "loss": 0.2726507052779198, "time": 0.6431791067123414, "epoch": 2, "memory": 6318, "step": 6011} +{"lr": 9.518312801497029e-05, "data_time": 0.5513953924179077, "loss": 0.24824558347463607, "time": 0.6920597076416015, "epoch": 2, "memory": 6318, "step": 6111} +{"lr": 9.502715948678276e-05, "data_time": 0.5037060976028442, "loss": 0.2736586719751358, "time": 0.6437305688858033, "epoch": 2, "memory": 6318, "step": 6211} +{"lr": 9.486885342721427e-05, "data_time": 0.4852008819580078, "loss": 0.267157718539238, "time": 0.6385726690292358, "epoch": 2, "memory": 6318, "step": 6311} +{"lr": 9.470821908112202e-05, "data_time": 0.5302046298980713, "loss": 0.28201734721660615, "time": 0.66909339427948, "epoch": 2, "memory": 6318, "step": 6411} +{"lr": 9.454526582933214e-05, "data_time": 0.5160373687744141, "loss": 0.2813108071684837, "time": 0.655771017074585, "epoch": 2, "memory": 6318, "step": 6511} +{"lr": 9.438000318809176e-05, "data_time": 0.5991506576538086, "loss": 0.2600247323513031, "time": 0.7393239736557007, "epoch": 2, "memory": 6318, "step": 6611} +{"lr": 9.42124408085132e-05, "data_time": 0.5790022373199463, "loss": 0.2557353124022484, "time": 0.7185931444168091, "epoch": 2, "memory": 6318, "step": 6711} +{"lr": 9.404258847601036e-05, "data_time": 0.5132441282272339, "loss": 0.27334317564964294, "time": 0.6528227806091309, "epoch": 2, "memory": 6318, "step": 6811} +{"lr": 9.387045610972717e-05, "data_time": 0.5245428085327148, "loss": 0.2611568719148636, "time": 0.6647748231887818, "epoch": 2, "memory": 6318, "step": 6911} +{"lr": 9.369605376195838e-05, "data_time": 0.6749463558197022, "loss": 0.2527195826172829, "time": 0.814208197593689, "epoch": 2, "memory": 6318, "step": 7011} +{"lr": 9.351939161756269e-05, "data_time": 0.5157665491104126, "loss": 0.2518232688307762, "time": 0.655615234375, "epoch": 2, "memory": 6318, "step": 7111} +{"lr": 9.334047999336764e-05, "data_time": 0.5483863353729248, "loss": 0.2573984503746033, "time": 0.6868363142013549, "epoch": 2, "memory": 6318, "step": 7211} +{"lr": 9.315932933756752e-05, "data_time": 0.506188178062439, "loss": 0.24999005645513533, "time": 0.6465755224227905, "epoch": 2, "memory": 6318, "step": 7311} +{"lr": 9.297595022911287e-05, "data_time": 0.5768125534057618, "loss": 0.2576883599162102, "time": 0.7152302026748657, "epoch": 2, "memory": 6318, "step": 7411} +{"lr": 9.279035337709295e-05, "data_time": 0.5277339696884156, "loss": 0.23903103917837143, "time": 0.6676866769790649, "epoch": 2, "memory": 6318, "step": 7511} +{"lr": 9.260254962011016e-05, "data_time": 0.5923547744750977, "loss": 0.24638890773057937, "time": 0.7327769994735718, "epoch": 2, "memory": 6318, "step": 7611} +{"lr": 9.241254992564702e-05, "data_time": 0.5583524703979492, "loss": 0.23376426547765733, "time": 0.70789954662323, "epoch": 2, "memory": 6318, "step": 7711} +{"lr": 9.22203653894259e-05, "data_time": 0.5611158609390259, "loss": 0.23874622881412505, "time": 0.7019243240356445, "epoch": 2, "memory": 6318, "step": 7811} +{"lr": 9.202600723476098e-05, "data_time": 0.536197304725647, "loss": 0.25499413162469864, "time": 0.6808391094207764, "epoch": 2, "memory": 6318, "step": 7911} +{"lr": 9.182948681190268e-05, "data_time": 0.5797640323638916, "loss": 0.2351435124874115, "time": 0.7194280624389648, "epoch": 2, "memory": 6318, "step": 8011} +{"lr": 9.163081559737497e-05, "data_time": 0.513187575340271, "loss": 0.2341068908572197, "time": 0.6537068128585816, "epoch": 2, "memory": 6318, "step": 8111} +{"lr": 9.143000519330517e-05, "data_time": 0.481587553024292, "loss": 0.250732509791851, "time": 0.6221200466156006, "epoch": 2, "memory": 6318, "step": 8211} +{"accuracy/top1": 76.61817932128906, "data_time": 0.3263594190279643, "time": 0.41486122210820514, "step": 2} +{"lr": 9.120461478753376e-05, "data_time": 0.5121278762817383, "loss": 0.23507924228906632, "time": 0.652920937538147, "epoch": 3, "memory": 6318, "step": 8322} +{"lr": 9.099932932138134e-05, "data_time": 0.5536199331283569, "loss": 0.23975561708211898, "time": 0.692713737487793, "epoch": 3, "memory": 6318, "step": 8422} +{"lr": 9.079194154362181e-05, "data_time": 0.5148802995681763, "loss": 0.24038059413433074, "time": 0.6558120489120484, "epoch": 3, "memory": 6318, "step": 8522} +{"lr": 9.058246356541787e-05, "data_time": 0.4992228984832764, "loss": 0.23119065016508103, "time": 0.6465325593948364, "epoch": 3, "memory": 6318, "step": 8622} +{"lr": 9.037090761999699e-05, "data_time": 0.5260629892349243, "loss": 0.22607937157154084, "time": 0.6672863006591797, "epoch": 3, "memory": 6318, "step": 8722} +{"lr": 9.015728606193701e-05, "data_time": 0.5703656673431396, "loss": 0.2585846409201622, "time": 0.7101302862167358, "epoch": 3, "memory": 6318, "step": 8822} +{"lr": 8.994161136644488e-05, "data_time": 0.5367135763168335, "loss": 0.22107840478420257, "time": 0.6768431901931763, "epoch": 3, "memory": 6318, "step": 8922} +{"lr": 8.972389612862792e-05, "data_time": 0.4898061752319336, "loss": 0.22925304472446442, "time": 0.6304153680801392, "epoch": 3, "memory": 6318, "step": 9022} +{"lr": 8.950415306275833e-05, "data_time": 0.5377302885055542, "loss": 0.2453791007399559, "time": 0.6776947259902955, "epoch": 3, "memory": 6318, "step": 9122} +{"lr": 8.928239500153059e-05, "data_time": 0.4841548681259155, "loss": 0.22476376742124557, "time": 0.6232578039169312, "epoch": 3, "memory": 6318, "step": 9222} +{"lr": 8.905863489531232e-05, "data_time": 0.5413121223449707, "loss": 0.2316204234957695, "time": 0.6809134006500244, "epoch": 3, "memory": 6318, "step": 9322} +{"lr": 8.883288581138775e-05, "data_time": 0.5113520383834839, "loss": 0.22785344570875168, "time": 0.6517601490020752, "epoch": 3, "memory": 6318, "step": 9422} +{"lr": 8.86051609331947e-05, "data_time": 0.5493632078170776, "loss": 0.2144322007894516, "time": 0.6891978502273559, "epoch": 3, "memory": 6318, "step": 9522} +{"lr": 8.837547355955467e-05, "data_time": 0.5674475193023681, "loss": 0.25626387894153596, "time": 0.707038140296936, "epoch": 3, "memory": 6318, "step": 9622} +{"lr": 8.814383710389614e-05, "data_time": 0.5146476030349731, "loss": 0.22953106611967086, "time": 0.653994345664978, "epoch": 3, "memory": 6318, "step": 9722} +{"lr": 8.791026509347146e-05, "data_time": 0.4882441759109497, "loss": 0.2418530121445656, "time": 0.6278645753860473, "epoch": 3, "memory": 6318, "step": 9822} +{"lr": 8.767477116856659e-05, "data_time": 0.5018274307250976, "loss": 0.2370159775018692, "time": 0.6412617683410644, "epoch": 3, "memory": 6318, "step": 9922} +{"lr": 8.743736908170466e-05, "data_time": 0.6059984683990478, "loss": 0.24052680283784866, "time": 0.7454351186752319, "epoch": 3, "memory": 6318, "step": 10022} +{"lr": 8.719807269684305e-05, "data_time": 0.5599032878875733, "loss": 0.23884377032518386, "time": 0.6996105432510376, "epoch": 3, "memory": 6318, "step": 10122} +{"lr": 8.695689598856356e-05, "data_time": 0.5556319236755372, "loss": 0.22184799760580062, "time": 0.6952784776687622, "epoch": 3, "memory": 6318, "step": 10222} +{"lr": 8.67138530412559e-05, "data_time": 0.5482355833053589, "loss": 0.22685093879699708, "time": 0.6899597406387329, "epoch": 3, "memory": 6318, "step": 10322} +{"lr": 8.646895804829604e-05, "data_time": 0.5226839303970336, "loss": 0.24236670285463333, "time": 0.6623435974121094, "epoch": 3, "memory": 6318, "step": 10422} +{"lr": 8.622222531121664e-05, "data_time": 0.5609231948852539, "loss": 0.2246718943119049, "time": 0.700794267654419, "epoch": 3, "memory": 6318, "step": 10522} +{"lr": 8.597366923887211e-05, "data_time": 0.5277501344680786, "loss": 0.23577700108289718, "time": 0.6677085638046265, "epoch": 3, "memory": 6318, "step": 10622} +{"lr": 8.57233043465972e-05, "data_time": 0.5971243143081665, "loss": 0.22813340425491332, "time": 0.7357851505279541, "epoch": 3, "memory": 6318, "step": 10722} +{"lr": 8.547114525535937e-05, "data_time": 0.5492791891098022, "loss": 0.22657681703567506, "time": 0.6891911029815674, "epoch": 3, "memory": 6318, "step": 10822} +{"lr": 8.521720669090474e-05, "data_time": 0.5139815807342529, "loss": 0.2166411831974983, "time": 0.6603597402572632, "epoch": 3, "memory": 6318, "step": 10922} +{"lr": 8.496150348289821e-05, "data_time": 0.5343125820159912, "loss": 0.2255546674132347, "time": 0.6739606142044068, "epoch": 3, "memory": 6318, "step": 11022} +{"lr": 8.470405056405736e-05, "data_time": 0.5242126226425171, "loss": 0.2221732333302498, "time": 0.6729058504104615, "epoch": 3, "memory": 6318, "step": 11122} +{"lr": 8.44448629692807e-05, "data_time": 0.5514846563339233, "loss": 0.2273815467953682, "time": 0.6917582511901855, "epoch": 3, "memory": 6318, "step": 11222} +{"lr": 8.418395583476941e-05, "data_time": 0.49566290378570554, "loss": 0.24244913905858995, "time": 0.6403186798095704, "epoch": 3, "memory": 6318, "step": 11322} +{"lr": 8.392134439714336e-05, "data_time": 0.5313705205917358, "loss": 0.22687100023031234, "time": 0.6714829206466675, "epoch": 3, "memory": 6318, "step": 11422} +{"lr": 8.365704399255133e-05, "data_time": 0.5148557424545288, "loss": 0.23275596052408218, "time": 0.6542460441589355, "epoch": 3, "memory": 6318, "step": 11522} +{"lr": 8.339107005577562e-05, "data_time": 0.6533173322677612, "loss": 0.2137850731611252, "time": 0.7933687686920166, "epoch": 3, "memory": 6318, "step": 11622} +{"lr": 8.31234381193305e-05, "data_time": 0.5868905782699585, "loss": 0.20853554606437683, "time": 0.7260604619979858, "epoch": 3, "memory": 6318, "step": 11722} +{"lr": 8.285416381255504e-05, "data_time": 0.5618477106094361, "loss": 0.20987340956926345, "time": 0.7015027523040771, "epoch": 3, "memory": 6318, "step": 11822} +{"lr": 8.258326286070047e-05, "data_time": 0.58131103515625, "loss": 0.22143910080194473, "time": 0.7206623792648316, "epoch": 3, "memory": 6318, "step": 11922} +{"lr": 8.231075108401202e-05, "data_time": 0.5419651508331299, "loss": 0.22237882316112517, "time": 0.6813149929046631, "epoch": 3, "memory": 6318, "step": 12022} +{"lr": 8.203664439680493e-05, "data_time": 0.5456916570663453, "loss": 0.2083484187722206, "time": 0.685080885887146, "epoch": 3, "memory": 6318, "step": 12122} +{"lr": 8.176095880653475e-05, "data_time": 0.5096866130828858, "loss": 0.20996739864349365, "time": 0.6498157262802124, "epoch": 3, "memory": 6318, "step": 12222} +{"lr": 8.148371041286308e-05, "data_time": 0.4793927907943726, "loss": 0.208797886967659, "time": 0.618826937675476, "epoch": 3, "memory": 6318, "step": 12322} +{"accuracy/top1": 82.17401123046875, "data_time": 0.3333737671375275, "time": 0.42224480708440143, "step": 3} +{"lr": 8.117415423358443e-05, "data_time": 0.5500046253204346, "loss": 0.20664273351430892, "time": 0.6890757322311402, "epoch": 4, "memory": 6318, "step": 12433} +{"lr": 8.089366155759266e-05, "data_time": 0.5329584598541259, "loss": 0.2161662459373474, "time": 0.6718969583511353, "epoch": 4, "memory": 6318, "step": 12533} +{"lr": 8.061165672717273e-05, "data_time": 0.5733454704284668, "loss": 0.21466794312000276, "time": 0.7122225761413574, "epoch": 4, "memory": 6318, "step": 12633} +{"lr": 8.032815621102116e-05, "data_time": 0.8412500619888306, "loss": 0.20947342216968537, "time": 0.9806801557540894, "epoch": 4, "memory": 6318, "step": 12733} +{"lr": 8.004317656518042e-05, "data_time": 0.6573879241943359, "loss": 0.2101093128323555, "time": 0.7969398498535156, "epoch": 4, "memory": 6318, "step": 12833} +{"lr": 7.975673443207223e-05, "data_time": 3.6199007987976075, "loss": 0.19993355125188828, "time": 3.7586714744567873, "epoch": 4, "memory": 6318, "step": 12933} +{"lr": 7.946884653952537e-05, "data_time": 0.7938543319702148, "loss": 0.205340938270092, "time": 0.9326043128967285, "epoch": 4, "memory": 6318, "step": 13033} +{"lr": 7.91795296997992e-05, "data_time": 0.5147840738296509, "loss": 0.23342226445674896, "time": 0.6549011707305908, "epoch": 4, "memory": 6318, "step": 13133} +{"lr": 7.888880080860151e-05, "data_time": 0.5238473176956177, "loss": 0.20029883682727814, "time": 0.6626713752746582, "epoch": 4, "memory": 6318, "step": 13233} +{"lr": 7.859667684410205e-05, "data_time": 0.5417109489440918, "loss": 0.21394192576408386, "time": 0.6817146301269531, "epoch": 4, "memory": 6318, "step": 13333} +{"lr": 7.830317486594082e-05, "data_time": 0.6320316553115845, "loss": 0.220724980533123, "time": 0.7733464956283569, "epoch": 4, "memory": 6318, "step": 13433} +{"lr": 7.800831201423207e-05, "data_time": 0.5526538848876953, "loss": 0.20695157945156098, "time": 0.6918729066848754, "epoch": 4, "memory": 6318, "step": 13533} +{"lr": 7.771210550856309e-05, "data_time": 0.5588021278381348, "loss": 0.20560887157917024, "time": 0.6995813131332398, "epoch": 4, "memory": 6318, "step": 13633} +{"lr": 7.741457264698884e-05, "data_time": 0.5980195760726928, "loss": 0.20007935613393785, "time": 0.7375678539276123, "epoch": 4, "memory": 6318, "step": 13733} +{"lr": 7.711573080502147e-05, "data_time": 0.7418617486953736, "loss": 0.205252206325531, "time": 0.8816857099533081, "epoch": 4, "memory": 6318, "step": 13833} +{"lr": 7.681559743461609e-05, "data_time": 0.5750361204147338, "loss": 0.199148528277874, "time": 0.7148786306381225, "epoch": 4, "memory": 6318, "step": 13933} +{"lr": 7.651419006315099e-05, "data_time": 0.5282118320465088, "loss": 0.22205454707145691, "time": 0.6676740169525146, "epoch": 4, "memory": 6318, "step": 14033} +{"lr": 7.621152629240453e-05, "data_time": 0.5356722831726074, "loss": 0.21486999541521073, "time": 0.6752888441085816, "epoch": 4, "memory": 6318, "step": 14133} +{"lr": 7.590762379752719e-05, "data_time": 0.5064670085906983, "loss": 0.21178832054138183, "time": 0.6480307102203369, "epoch": 4, "memory": 6318, "step": 14233} +{"lr": 7.560250032600911e-05, "data_time": 0.5267542362213135, "loss": 0.22033770978450776, "time": 0.6750738859176636, "epoch": 4, "memory": 6318, "step": 14333} +{"lr": 7.52961736966437e-05, "data_time": 0.5785313367843627, "loss": 0.20205722451210023, "time": 0.719299030303955, "epoch": 4, "memory": 6318, "step": 14433} +{"lr": 7.498866179848732e-05, "data_time": 0.5726711750030518, "loss": 0.2077656090259552, "time": 0.715662670135498, "epoch": 4, "memory": 6318, "step": 14533} +{"lr": 7.467998258981415e-05, "data_time": 0.5317749738693237, "loss": 0.19386776238679887, "time": 0.671155309677124, "epoch": 4, "memory": 6318, "step": 14633} +{"lr": 7.437015409706804e-05, "data_time": 0.6084056377410889, "loss": 0.1978773444890976, "time": 0.7483431100845337, "epoch": 4, "memory": 6318, "step": 14733} +{"lr": 7.405919441380929e-05, "data_time": 0.5344118595123291, "loss": 0.21792005449533464, "time": 0.6741207599639892, "epoch": 4, "memory": 6318, "step": 14833} +{"lr": 7.374712169965821e-05, "data_time": 0.5174609422683716, "loss": 0.20713909566402436, "time": 0.6584647178649903, "epoch": 4, "memory": 6318, "step": 14933} +{"lr": 7.343395417923461e-05, "data_time": 0.5622954845428467, "loss": 0.2105598047375679, "time": 0.7038269519805909, "epoch": 4, "memory": 6318, "step": 15033} +{"lr": 7.311971014109342e-05, "data_time": 0.5359824895858765, "loss": 0.19901027977466584, "time": 0.6760349750518799, "epoch": 4, "memory": 6318, "step": 15133} +{"lr": 7.280440793665682e-05, "data_time": 0.914686918258667, "loss": 0.2062069669365883, "time": 1.0543843507766724, "epoch": 4, "memory": 6318, "step": 15233} +{"lr": 7.248806597914234e-05, "data_time": 0.5441090822219848, "loss": 0.19911188036203384, "time": 0.683473777770996, "epoch": 4, "memory": 6318, "step": 15333} +{"lr": 7.217070274248765e-05, "data_time": 0.5418741464614868, "loss": 0.1903040498495102, "time": 0.6813448667526245, "epoch": 4, "memory": 6318, "step": 15433} +{"lr": 7.185233676027203e-05, "data_time": 0.5426470994949341, "loss": 0.21562436372041702, "time": 0.6836632251739502, "epoch": 4, "memory": 6318, "step": 15533} +{"lr": 7.153298662463338e-05, "data_time": 0.5512782335281372, "loss": 0.2175891324877739, "time": 0.6904650449752807, "epoch": 4, "memory": 6318, "step": 15633} +{"lr": 7.121267098518291e-05, "data_time": 0.555422043800354, "loss": 0.20827249586582183, "time": 0.6948675394058228, "epoch": 4, "memory": 6318, "step": 15733} +{"lr": 7.089140854791618e-05, "data_time": 0.7726138591766357, "loss": 0.21206145882606506, "time": 0.914221215248108, "epoch": 4, "memory": 6318, "step": 15833} +{"lr": 7.056921807412025e-05, "data_time": 0.581702995300293, "loss": 0.20164531767368316, "time": 0.7231520652770996, "epoch": 4, "memory": 6318, "step": 15933} +{"lr": 7.024611837927802e-05, "data_time": 0.5392264127731323, "loss": 0.2005120888352394, "time": 0.6925257682800293, "epoch": 4, "memory": 6318, "step": 16033} +{"lr": 6.992212833196998e-05, "data_time": 0.5013979196548461, "loss": 0.21203254908323288, "time": 0.6423643589019775, "epoch": 4, "memory": 6318, "step": 16133} +{"lr": 6.959726685277174e-05, "data_time": 0.5449551343917847, "loss": 0.1945815697312355, "time": 0.6863574028015137, "epoch": 4, "memory": 6318, "step": 16233} +{"lr": 6.927155291314949e-05, "data_time": 0.5248435258865356, "loss": 0.20592499673366546, "time": 0.6640942096710205, "epoch": 4, "memory": 6318, "step": 16333} +{"lr": 6.894500553435204e-05, "data_time": 0.5502176761627198, "loss": 0.21122395545244216, "time": 0.6902376413345337, "epoch": 4, "memory": 6318, "step": 16433} +{"accuracy/top1": 87.56930541992188, "data_time": 0.3289952198664347, "time": 0.41643391648928324, "step": 4} +{"lr": 6.858158509719903e-05, "data_time": 0.6103754281997681, "loss": 0.19654458463191987, "time": 0.7525153160095215, "epoch": 5, "memory": 6318, "step": 16544} +{"lr": 6.825334178880171e-05, "data_time": 0.5068995714187622, "loss": 0.19921637028455735, "time": 0.6520355224609375, "epoch": 5, "memory": 6318, "step": 16644} +{"lr": 6.792432450336645e-05, "data_time": 0.5620324850082398, "loss": 0.20516614019870758, "time": 0.7025948286056518, "epoch": 5, "memory": 6318, "step": 16744} +{"lr": 6.759455245505286e-05, "data_time": 0.5227182865142822, "loss": 0.1997493878006935, "time": 0.6678563356399536, "epoch": 5, "memory": 6318, "step": 16844} +{"lr": 6.726404490209753e-05, "data_time": 0.7072115182876587, "loss": 0.174032524228096, "time": 0.8474693059921264, "epoch": 5, "memory": 6318, "step": 16944} +{"lr": 6.693282114568972e-05, "data_time": 0.5498621463775635, "loss": 0.18071813136339188, "time": 0.6907564640045166, "epoch": 5, "memory": 6318, "step": 17044} +{"lr": 6.660090052884388e-05, "data_time": 0.49100098609924314, "loss": 0.19130819737911225, "time": 0.6312556743621827, "epoch": 5, "memory": 6318, "step": 17144} +{"lr": 6.626830243527016e-05, "data_time": 0.5211453914642334, "loss": 0.20275127589702607, "time": 0.66148362159729, "epoch": 5, "memory": 6318, "step": 17244} +{"lr": 6.593504628824236e-05, "data_time": 0.5267242193222046, "loss": 0.18796692788600922, "time": 0.6666906118392945, "epoch": 5, "memory": 6318, "step": 17344} +{"lr": 6.560115154946387e-05, "data_time": 0.4699209451675415, "loss": 0.19421417266130447, "time": 0.660755729675293, "epoch": 5, "memory": 6318, "step": 17444} +{"lr": 6.526663771793081e-05, "data_time": 0.5860954999923706, "loss": 0.1862255036830902, "time": 0.7251555681228637, "epoch": 5, "memory": 6318, "step": 17544} +{"lr": 6.493152432879363e-05, "data_time": 0.5247564077377319, "loss": 0.18504683375358583, "time": 0.6648433446884155, "epoch": 5, "memory": 6318, "step": 17644} +{"lr": 6.459583095221596e-05, "data_time": 0.5299063920974731, "loss": 0.18796184808015823, "time": 0.6706902265548706, "epoch": 5, "memory": 6318, "step": 17744} +{"lr": 6.425957719223197e-05, "data_time": 0.5014869689941406, "loss": 0.1826095849275589, "time": 0.6518988609313965, "epoch": 5, "memory": 6318, "step": 17844} +{"lr": 6.392278268560153e-05, "data_time": 0.594998049736023, "loss": 0.18176030963659287, "time": 0.7340277671813965, "epoch": 5, "memory": 6318, "step": 17944} +{"lr": 6.358546710066316e-05, "data_time": 0.5260184049606323, "loss": 0.2019784778356552, "time": 0.6648510217666626, "epoch": 5, "memory": 6318, "step": 18044} +{"lr": 6.324765013618573e-05, "data_time": 0.5347894430160522, "loss": 0.2267347291111946, "time": 0.683300518989563, "epoch": 5, "memory": 6318, "step": 18144} +{"lr": 6.290935152021818e-05, "data_time": 0.5557122230529785, "loss": 0.18254578113555908, "time": 0.6968997716903687, "epoch": 5, "memory": 6318, "step": 18244} +{"lr": 6.2570591008937e-05, "data_time": 0.5397093296051025, "loss": 0.20010170340538025, "time": 0.6792788982391358, "epoch": 5, "memory": 6318, "step": 18344} +{"lr": 6.223138838549294e-05, "data_time": 0.5627518415451049, "loss": 0.20028055757284163, "time": 0.7036221742630004, "epoch": 5, "memory": 6318, "step": 18444} +{"lr": 6.18917634588553e-05, "data_time": 0.5519994258880615, "loss": 0.1819985553622246, "time": 0.6915643215179443, "epoch": 5, "memory": 6318, "step": 18544} +{"lr": 6.155173606265555e-05, "data_time": 0.549025583267212, "loss": 0.20035212188959123, "time": 0.6887681245803833, "epoch": 5, "memory": 6318, "step": 18644} +{"lr": 6.121132605402843e-05, "data_time": 0.579982852935791, "loss": 0.1834666386246681, "time": 0.7216654777526855, "epoch": 5, "memory": 6318, "step": 18744} +{"lr": 6.087055331245287e-05, "data_time": 0.5233453035354614, "loss": 0.2027045249938965, "time": 0.6631776094436646, "epoch": 5, "memory": 6318, "step": 18844} +{"lr": 6.05294377385911e-05, "data_time": 0.5548267841339112, "loss": 0.20322794914245607, "time": 0.6963297367095947, "epoch": 5, "memory": 6318, "step": 18944} +{"lr": 6.0187999253126355e-05, "data_time": 0.5397990226745606, "loss": 0.18461829870939256, "time": 0.6788439750671387, "epoch": 5, "memory": 6318, "step": 19044} +{"lr": 5.9846257795599195e-05, "data_time": 0.5467065095901489, "loss": 0.20155278593301773, "time": 0.6944790124893189, "epoch": 5, "memory": 6318, "step": 19144} +{"lr": 5.950423332324355e-05, "data_time": 0.5479861259460449, "loss": 0.19435943961143493, "time": 0.689105224609375, "epoch": 5, "memory": 6318, "step": 19244} +{"lr": 5.9161945809820855e-05, "data_time": 0.5530983209609985, "loss": 0.18167036026716232, "time": 0.6932096958160401, "epoch": 5, "memory": 6318, "step": 19344} +{"lr": 5.8819415244453975e-05, "data_time": 0.5245201110839843, "loss": 0.18412984013557435, "time": 0.6642180442810058, "epoch": 5, "memory": 6318, "step": 19444} +{"lr": 5.8476661630459405e-05, "data_time": 0.5391457796096801, "loss": 0.19481215327978135, "time": 0.6804267168045044, "epoch": 5, "memory": 6318, "step": 19544} +{"lr": 5.81337049841796e-05, "data_time": 0.582946491241455, "loss": 0.2007267326116562, "time": 0.7248592138290405, "epoch": 5, "memory": 6318, "step": 19644} +{"lr": 5.779056533381373e-05, "data_time": 0.5544168949127197, "loss": 0.19656536728143692, "time": 0.6942979574203492, "epoch": 5, "memory": 6318, "step": 19744} +{"lr": 5.744726271824817e-05, "data_time": 0.5061422109603881, "loss": 0.2001613974571228, "time": 0.6463444471359253, "epoch": 5, "memory": 6318, "step": 19844} +{"lr": 5.710381718588622e-05, "data_time": 0.5269783496856689, "loss": 0.18792343586683274, "time": 0.6664208889007568, "epoch": 5, "memory": 6318, "step": 19944} +{"lr": 5.676024879347741e-05, "data_time": 0.5109683752059937, "loss": 0.19335417598485946, "time": 0.6527363538742066, "epoch": 5, "memory": 6318, "step": 20044} +{"lr": 5.6416577604945966e-05, "data_time": 0.6091557502746582, "loss": 0.2001790001988411, "time": 0.7500738382339478, "epoch": 5, "memory": 6318, "step": 20144} +{"lr": 5.6072823690219385e-05, "data_time": 0.5595529556274415, "loss": 0.18933072239160537, "time": 0.6989166736602783, "epoch": 5, "memory": 6318, "step": 20244} +{"lr": 5.572900712405624e-05, "data_time": 0.508144211769104, "loss": 0.1762568861246109, "time": 0.6485060453414917, "epoch": 5, "memory": 6318, "step": 20344} +{"lr": 5.538514798487397e-05, "data_time": 0.6103987693786621, "loss": 0.18673183917999267, "time": 0.7505025148391724, "epoch": 5, "memory": 6318, "step": 20444} +{"lr": 5.5041266353576e-05, "data_time": 0.5456227779388427, "loss": 0.19972552955150605, "time": 0.6874054193496704, "epoch": 5, "memory": 6318, "step": 20544} +{"accuracy/top1": 90.79158782958984, "data_time": 0.361405078570048, "time": 0.44991439779599507, "step": 5} +{"lr": 5.4659555783035384e-05, "data_time": 0.5881729364395142, "loss": 0.19360779970884323, "time": 0.7296860694885254, "epoch": 6, "memory": 6318, "step": 20655} +{"lr": 5.431569258426336e-05, "data_time": 0.5668072938919068, "loss": 0.18631560802459718, "time": 0.7078919649124146, "epoch": 6, "memory": 6318, "step": 20755} +{"lr": 5.397186934810864e-05, "data_time": 0.47689697742462156, "loss": 0.20995801091194152, "time": 0.6162860155105591, "epoch": 6, "memory": 6318, "step": 20855} +{"lr": 5.362810615337815e-05, "data_time": 0.8281100511550903, "loss": 0.1943479910492897, "time": 0.968401575088501, "epoch": 6, "memory": 6318, "step": 20955} +{"lr": 5.328442307537233e-05, "data_time": 0.6247039556503295, "loss": 0.18478670567274094, "time": 0.7642682790756226, "epoch": 6, "memory": 6318, "step": 21055} +{"lr": 5.294084018471303e-05, "data_time": 0.5587382793426514, "loss": 0.1863896906375885, "time": 0.6981104850769043, "epoch": 6, "memory": 6318, "step": 21155} +{"lr": 5.259737754617132e-05, "data_time": 0.500095009803772, "loss": 0.17880396991968156, "time": 0.6411489725112915, "epoch": 6, "memory": 6318, "step": 21255} +{"lr": 5.2254055217495525e-05, "data_time": 0.49918527603149415, "loss": 0.18837087452411652, "time": 0.6402879238128663, "epoch": 6, "memory": 6318, "step": 21355} +{"lr": 5.1910893248240283e-05, "data_time": 0.5623347997665405, "loss": 0.18812087029218674, "time": 0.7013975620269776, "epoch": 6, "memory": 6318, "step": 21455} +{"lr": 5.156791167859532e-05, "data_time": 0.5729370594024659, "loss": 0.16644491553306578, "time": 0.7145461082458496, "epoch": 6, "memory": 6318, "step": 21555} +{"lr": 5.122513053821533e-05, "data_time": 0.6027938842773437, "loss": 0.18924373239278794, "time": 0.7450995206832886, "epoch": 6, "memory": 6318, "step": 21655} +{"lr": 5.0882569845050173e-05, "data_time": 0.6844998359680176, "loss": 0.19755019694566728, "time": 0.8237540006637574, "epoch": 6, "memory": 6318, "step": 21755} +{"lr": 5.054024960417592e-05, "data_time": 0.5399816513061524, "loss": 0.2102901577949524, "time": 0.6798722505569458, "epoch": 6, "memory": 6318, "step": 21855} +{"lr": 5.0198189806626494e-05, "data_time": 0.5544996738433838, "loss": 0.18910381197929382, "time": 0.6935258388519288, "epoch": 6, "memory": 6318, "step": 21955} +{"lr": 4.985641042822647e-05, "data_time": 0.5247689962387085, "loss": 0.187516987323761, "time": 0.6636535167694092, "epoch": 6, "memory": 6318, "step": 22055} +{"lr": 4.951493142842412e-05, "data_time": 0.5312407493591309, "loss": 0.1782834127545357, "time": 0.6708173036575318, "epoch": 6, "memory": 6318, "step": 22155} +{"lr": 4.917377274912603e-05, "data_time": 0.5415554761886596, "loss": 0.18463821411132814, "time": 0.6817080736160278, "epoch": 6, "memory": 6318, "step": 22255} +{"lr": 4.88329543135327e-05, "data_time": 0.543966817855835, "loss": 0.19569617211818696, "time": 0.684067964553833, "epoch": 6, "memory": 6318, "step": 22355} +{"lr": 4.849249602497467e-05, "data_time": 0.5453457593917846, "loss": 0.18015597462654115, "time": 0.6858213424682618, "epoch": 6, "memory": 6318, "step": 22455} +{"lr": 4.8152417765750456e-05, "data_time": 0.5377744436264038, "loss": 0.18413397967815398, "time": 0.676759910583496, "epoch": 6, "memory": 6318, "step": 22555} +{"lr": 4.7812739395965355e-05, "data_time": 0.6555386304855346, "loss": 0.16897769868373871, "time": 0.7971973657608032, "epoch": 6, "memory": 6318, "step": 22655} +{"lr": 4.74734807523717e-05, "data_time": 0.5349490165710449, "loss": 0.172246565669775, "time": 0.6760582685470581, "epoch": 6, "memory": 6318, "step": 22755} +{"lr": 4.7134661647210416e-05, "data_time": 0.5201940536499023, "loss": 0.1744161680340767, "time": 0.6606065034866333, "epoch": 6, "memory": 6318, "step": 22855} +{"lr": 4.679630186705388e-05, "data_time": 0.572269868850708, "loss": 0.17823325768113135, "time": 0.7128920793533325, "epoch": 6, "memory": 6318, "step": 22955} +{"lr": 4.6458421171650644e-05, "data_time": 0.556977105140686, "loss": 0.18836547285318375, "time": 0.6965638399124146, "epoch": 6, "memory": 6318, "step": 23055} +{"lr": 4.6121039292771295e-05, "data_time": 0.6068763971328736, "loss": 0.18693819493055344, "time": 0.7554102897644043, "epoch": 6, "memory": 6318, "step": 23155} +{"lr": 4.5784175933056197e-05, "data_time": 0.5251615762710571, "loss": 0.17846366316080092, "time": 0.6648310422897339, "epoch": 6, "memory": 6318, "step": 23255} +{"lr": 4.544785076486494e-05, "data_time": 0.5238264322280883, "loss": 0.1825462445616722, "time": 0.6644751787185669, "epoch": 6, "memory": 6318, "step": 23355} +{"lr": 4.5112083429127475e-05, "data_time": 0.5412717580795288, "loss": 0.1809942215681076, "time": 0.6811113119125366, "epoch": 6, "memory": 6318, "step": 23455} +{"lr": 4.477689353419706e-05, "data_time": 0.5270626068115234, "loss": 0.18884348273277282, "time": 0.6666532516479492, "epoch": 6, "memory": 6318, "step": 23555} +{"lr": 4.4442300654705266e-05, "data_time": 0.6008017778396606, "loss": 0.190585695207119, "time": 0.7420772552490235, "epoch": 6, "memory": 6318, "step": 23655} +{"lr": 4.410832433041866e-05, "data_time": 0.5090028524398804, "loss": 0.18598672747612, "time": 0.6488039493560791, "epoch": 6, "memory": 6318, "step": 23755} +{"lr": 4.377498406509792e-05, "data_time": 0.5628119945526123, "loss": 0.17237576544284822, "time": 0.7039343833923339, "epoch": 6, "memory": 6318, "step": 23855} +{"lr": 4.344229932535875e-05, "data_time": 0.5091280460357666, "loss": 0.18829593807458878, "time": 0.650333023071289, "epoch": 6, "memory": 6318, "step": 23955} +{"lr": 4.3110289539535115e-05, "data_time": 0.5777626276016236, "loss": 0.19435689896345137, "time": 0.7181173324584961, "epoch": 6, "memory": 6318, "step": 24055} +{"lr": 4.2778974096544436e-05, "data_time": 0.5667340755462646, "loss": 0.17597020864486695, "time": 0.7064597845077515, "epoch": 6, "memory": 6318, "step": 24155} +{"lr": 4.2448372344755544e-05, "data_time": 0.5271841049194336, "loss": 0.18336962014436722, "time": 0.6673299551010132, "epoch": 6, "memory": 6318, "step": 24255} +{"lr": 4.211850359085871e-05, "data_time": 0.5520426988601684, "loss": 0.19305049180984496, "time": 0.6924070835113525, "epoch": 6, "memory": 6318, "step": 24355} +{"lr": 4.17893870987381e-05, "data_time": 0.5354948520660401, "loss": 0.16639301627874375, "time": 0.6757314682006836, "epoch": 6, "memory": 6318, "step": 24455} +{"lr": 4.146104208834679e-05, "data_time": 0.4852054834365845, "loss": 0.19581633508205415, "time": 0.6253786563873291, "epoch": 6, "memory": 6318, "step": 24555} +{"lr": 4.113348773458442e-05, "data_time": 0.5880526065826416, "loss": 0.2019881710410118, "time": 0.7272488355636597, "epoch": 6, "memory": 6318, "step": 24655} +{"accuracy/top1": 92.52437591552734, "data_time": 0.3327964166800181, "time": 0.42110156019528705, "step": 6} +{"lr": 4.0770851520835164e-05, "data_time": 0.46455338001251223, "loss": 0.1814323514699936, "time": 0.7192839860916138, "epoch": 7, "memory": 6318, "step": 24766} +{"lr": 4.044502815742039e-05, "data_time": 0.5248425960540771, "loss": 0.16935558319091798, "time": 0.6638788223266602, "epoch": 7, "memory": 6318, "step": 24866} +{"lr": 4.012005478445896e-05, "data_time": 0.5700514793395997, "loss": 0.18679607212543486, "time": 0.7094446659088135, "epoch": 7, "memory": 6318, "step": 24966} +{"lr": 3.9795950379951394e-05, "data_time": 0.5068717002868652, "loss": 0.1741093322634697, "time": 0.6475993394851685, "epoch": 7, "memory": 6318, "step": 25066} +{"lr": 3.947273387115169e-05, "data_time": 0.51826331615448, "loss": 0.19768147319555282, "time": 0.6584887266159057, "epoch": 7, "memory": 6318, "step": 25166} +{"lr": 3.9150424133462044e-05, "data_time": 0.504107666015625, "loss": 0.16667136996984483, "time": 0.64411301612854, "epoch": 7, "memory": 6318, "step": 25266} +{"lr": 3.882903998933037e-05, "data_time": 0.5777571201324463, "loss": 0.1736258640885353, "time": 0.7195043325424194, "epoch": 7, "memory": 6318, "step": 25366} +{"lr": 3.8508600207151264e-05, "data_time": 0.6008114576339721, "loss": 0.1911705791950226, "time": 0.7398993730545044, "epoch": 7, "memory": 6318, "step": 25466} +{"lr": 3.818912350016978e-05, "data_time": 0.5565430641174316, "loss": 0.1841244339942932, "time": 0.6966655015945434, "epoch": 7, "memory": 6318, "step": 25566} +{"lr": 3.787062852538883e-05, "data_time": 0.6452668190002442, "loss": 0.17789336740970613, "time": 0.7850041866302491, "epoch": 7, "memory": 6318, "step": 25666} +{"lr": 3.7553133882479346e-05, "data_time": 0.5507402420043945, "loss": 0.18023606538772582, "time": 0.6913224220275879, "epoch": 7, "memory": 6318, "step": 25766} +{"lr": 3.723665811269437e-05, "data_time": 0.5683861255645752, "loss": 0.18480758666992186, "time": 0.7086949110031128, "epoch": 7, "memory": 6318, "step": 25866} +{"lr": 3.6921219697786176e-05, "data_time": 0.5580015182495117, "loss": 0.17970621585845947, "time": 0.69829843044281, "epoch": 7, "memory": 6318, "step": 25966} +{"lr": 3.660683705892677e-05, "data_time": 0.5900080442428589, "loss": 0.1739823713898659, "time": 0.7292535543441773, "epoch": 7, "memory": 6318, "step": 26066} +{"lr": 3.6293528555632435e-05, "data_time": 0.5164843320846557, "loss": 0.19332556873559953, "time": 0.6585107326507569, "epoch": 7, "memory": 6318, "step": 26166} +{"lr": 3.598131248469134e-05, "data_time": 0.6044678211212158, "loss": 0.20842364579439163, "time": 0.7450649738311768, "epoch": 7, "memory": 6318, "step": 26266} +{"lr": 3.567020707909507e-05, "data_time": 0.5335680961608886, "loss": 0.16210654824972154, "time": 0.6747160434722901, "epoch": 7, "memory": 6318, "step": 26366} +{"lr": 3.5360230506973935e-05, "data_time": 0.579534363746643, "loss": 0.18372683674097062, "time": 0.720951509475708, "epoch": 7, "memory": 6318, "step": 26466} +{"lr": 3.5051400870535877e-05, "data_time": 0.5325656175613404, "loss": 0.20466922968626022, "time": 0.673722505569458, "epoch": 7, "memory": 6318, "step": 26566} +{"lr": 3.47437362050094e-05, "data_time": 0.5428436040878296, "loss": 0.15994197130203247, "time": 0.6824786424636841, "epoch": 7, "memory": 6318, "step": 26666} +{"lr": 3.443725447759023e-05, "data_time": 0.540584135055542, "loss": 0.16788210272789, "time": 0.6811728954315186, "epoch": 7, "memory": 6318, "step": 26766} +{"lr": 3.413197358639224e-05, "data_time": 0.5269869565963745, "loss": 0.17511895298957825, "time": 0.6666075468063355, "epoch": 7, "memory": 6318, "step": 26866} +{"lr": 3.382791135940194e-05, "data_time": 0.5360666751861572, "loss": 0.18723165094852448, "time": 0.6772321462631226, "epoch": 7, "memory": 6318, "step": 26966} +{"lr": 3.35250855534377e-05, "data_time": 0.53197340965271, "loss": 0.18540700823068618, "time": 0.6717989683151245, "epoch": 7, "memory": 6318, "step": 27066} +{"lr": 3.3223513853112476e-05, "data_time": 0.5130138397216797, "loss": 0.16946106925606727, "time": 0.6549042701721192, "epoch": 7, "memory": 6318, "step": 27166} +{"lr": 3.292321386980115e-05, "data_time": 0.5537219524383545, "loss": 0.17841366678476334, "time": 0.6954458236694336, "epoch": 7, "memory": 6318, "step": 27266} +{"lr": 3.262420314061217e-05, "data_time": 0.6147526741027832, "loss": 0.1982018008828163, "time": 0.7574407577514648, "epoch": 7, "memory": 6318, "step": 27366} +{"lr": 3.2326499127363195e-05, "data_time": 0.5746092319488525, "loss": 0.1772906720638275, "time": 0.7149666786193848, "epoch": 7, "memory": 6318, "step": 27466} +{"lr": 3.2030119215561613e-05, "data_time": 0.5163069009780884, "loss": 0.1781323418021202, "time": 0.6553100347518921, "epoch": 7, "memory": 6318, "step": 27566} +{"lr": 3.173508071338896e-05, "data_time": 0.517768669128418, "loss": 0.18461388498544692, "time": 0.6582580804824829, "epoch": 7, "memory": 6318, "step": 27666} +{"lr": 3.144140085069037e-05, "data_time": 0.5436537742614747, "loss": 0.18624600172042846, "time": 0.6835618734359741, "epoch": 7, "memory": 6318, "step": 27766} +{"lr": 3.114909677796814e-05, "data_time": 0.5386141777038574, "loss": 0.1818668231368065, "time": 0.6798846960067749, "epoch": 7, "memory": 6318, "step": 27866} +{"lr": 3.085818556538059e-05, "data_time": 0.5464069366455078, "loss": 0.18328240066766738, "time": 0.6873406410217285, "epoch": 7, "memory": 6318, "step": 27966} +{"lr": 3.056868420174455e-05, "data_time": 0.496645450592041, "loss": 0.19840313047170638, "time": 0.6369399547576904, "epoch": 7, "memory": 6318, "step": 28066} +{"lr": 3.028060959354392e-05, "data_time": 0.5428974866867066, "loss": 0.1919218197464943, "time": 0.6818622589111328, "epoch": 7, "memory": 6318, "step": 28166} +{"lr": 2.9993978563941876e-05, "data_time": 0.5451676368713378, "loss": 0.17275484502315522, "time": 0.6860358953475952, "epoch": 7, "memory": 6318, "step": 28266} +{"lr": 2.9708807851798587e-05, "data_time": 0.510317587852478, "loss": 0.2020210310816765, "time": 0.6525311946868897, "epoch": 7, "memory": 6318, "step": 28366} +{"lr": 2.9425114110693717e-05, "data_time": 0.49378926753997804, "loss": 0.1569451205432415, "time": 0.6336827039718628, "epoch": 7, "memory": 6318, "step": 28466} +{"lr": 2.9142913907953844e-05, "data_time": 0.5152788639068604, "loss": 0.175429967045784, "time": 0.6560280799865723, "epoch": 7, "memory": 6318, "step": 28566} +{"lr": 2.8862223723684912e-05, "data_time": 0.5369388818740845, "loss": 0.16513666808605193, "time": 0.681144118309021, "epoch": 7, "memory": 6318, "step": 28666} +{"lr": 2.8583059949809854e-05, "data_time": 0.5557304859161377, "loss": 0.17852320224046708, "time": 0.6952420234680176, "epoch": 7, "memory": 6318, "step": 28766} +{"accuracy/top1": 93.6083984375, "data_time": 0.33878278930981953, "time": 0.4279541730880737, "step": 7} +{"lr": 2.827499545208893e-05, "data_time": 0.3465777158737183, "loss": 0.18094997555017472, "time": 0.7000437259674073, "epoch": 8, "memory": 6318, "step": 28877} +{"lr": 2.7999105784987406e-05, "data_time": 0.39706437587738036, "loss": 0.19337422549724578, "time": 0.7807217597961426, "epoch": 8, "memory": 6318, "step": 28977} +{"lr": 2.772479293318711e-05, "data_time": 0.5286478996276855, "loss": 0.17107584774494172, "time": 0.6686571836471558, "epoch": 8, "memory": 6318, "step": 29077} +{"lr": 2.7452072916183603e-05, "data_time": 0.5333987474441528, "loss": 0.19867638498544693, "time": 0.6723517179489136, "epoch": 8, "memory": 6318, "step": 29177} +{"lr": 2.7180961660452883e-05, "data_time": 0.5628446102142334, "loss": 0.17373458296060562, "time": 0.701029372215271, "epoch": 8, "memory": 6318, "step": 29277} +{"lr": 2.6911474998521622e-05, "data_time": 0.627887487411499, "loss": 0.18528251200914383, "time": 0.7667783260345459, "epoch": 8, "memory": 6318, "step": 29377} +{"lr": 2.664362866804239e-05, "data_time": 0.6329079866409302, "loss": 0.1671963721513748, "time": 0.7732800960540771, "epoch": 8, "memory": 6318, "step": 29477} +{"lr": 2.637743831087459e-05, "data_time": 0.5496542453765869, "loss": 0.1769385442137718, "time": 0.6907904624938965, "epoch": 8, "memory": 6318, "step": 29577} +{"lr": 2.6112919472171114e-05, "data_time": 0.5835949420928955, "loss": 0.17195185422897338, "time": 0.725251579284668, "epoch": 8, "memory": 6318, "step": 29677} +{"lr": 2.5850087599470434e-05, "data_time": 0.5291482210159302, "loss": 0.17310436815023422, "time": 0.6693939447402955, "epoch": 8, "memory": 6318, "step": 29777} +{"lr": 2.5588958041794503e-05, "data_time": 0.6031557559967041, "loss": 0.15010228902101516, "time": 0.7433319091796875, "epoch": 8, "memory": 6318, "step": 29877} +{"lr": 2.5329546048752453e-05, "data_time": 0.5976364612579346, "loss": 0.1693695940077305, "time": 0.7385983228683471, "epoch": 8, "memory": 6318, "step": 29977} +{"lr": 2.5071866769649806e-05, "data_time": 0.5417542219161987, "loss": 0.18125804811716079, "time": 0.6841305017471313, "epoch": 8, "memory": 6318, "step": 30077} +{"lr": 2.4815935252604222e-05, "data_time": 0.5030296564102172, "loss": 0.1685178168118, "time": 0.6441235780715943, "epoch": 8, "memory": 6318, "step": 30177} +{"lr": 2.4561766443666384e-05, "data_time": 0.5139962434768677, "loss": 0.18587067425251008, "time": 0.6538240194320679, "epoch": 8, "memory": 6318, "step": 30277} +{"lr": 2.4309375185947164e-05, "data_time": 0.5446279764175415, "loss": 0.17585383355617523, "time": 0.6846682071685791, "epoch": 8, "memory": 6318, "step": 30377} +{"lr": 2.40587762187509e-05, "data_time": 0.5620025634765625, "loss": 0.16819631159305573, "time": 0.7023976087570191, "epoch": 8, "memory": 6318, "step": 30477} +{"lr": 2.3809984176714634e-05, "data_time": 0.5496367454528809, "loss": 0.1661178320646286, "time": 0.6897157430648804, "epoch": 8, "memory": 6318, "step": 30577} +{"lr": 2.3563013588953406e-05, "data_time": 0.5490324258804321, "loss": 0.16991938948631286, "time": 0.6881859302520752, "epoch": 8, "memory": 6318, "step": 30677} +{"lr": 2.331787887821182e-05, "data_time": 0.5657930374145508, "loss": 0.1782354325056076, "time": 0.7059160947799683, "epoch": 8, "memory": 6318, "step": 30777} +{"lr": 2.307459436002179e-05, "data_time": 0.5315551042556763, "loss": 0.18601717352867125, "time": 0.671811318397522, "epoch": 8, "memory": 6318, "step": 30877} +{"lr": 2.2833174241866527e-05, "data_time": 0.5765299320220947, "loss": 0.1732058271765709, "time": 0.7299833536148072, "epoch": 8, "memory": 6318, "step": 30977} +{"lr": 2.259363262235086e-05, "data_time": 0.5758986711502075, "loss": 0.18273567110300065, "time": 0.7165857076644897, "epoch": 8, "memory": 6318, "step": 31077} +{"lr": 2.2355983490377735e-05, "data_time": 0.5200314521789551, "loss": 0.1857576221227646, "time": 0.6596033096313476, "epoch": 8, "memory": 6318, "step": 31177} +{"lr": 2.2120240724331546e-05, "data_time": 0.5255465984344483, "loss": 0.17776864022016525, "time": 0.6645726680755615, "epoch": 8, "memory": 6318, "step": 31277} +{"lr": 2.1886418091267445e-05, "data_time": 0.5153661966323853, "loss": 0.1763664811849594, "time": 0.6579375743865967, "epoch": 8, "memory": 6318, "step": 31377} +{"lr": 2.1654529246107496e-05, "data_time": 0.5290921926498413, "loss": 0.18671398758888244, "time": 0.6683281660079956, "epoch": 8, "memory": 6318, "step": 31477} +{"lr": 2.1424587730843184e-05, "data_time": 0.5164127588272095, "loss": 0.16295654326677322, "time": 0.6617313623428345, "epoch": 8, "memory": 6318, "step": 31577} +{"lr": 2.1196606973744593e-05, "data_time": 0.5644389152526855, "loss": 0.1775700107216835, "time": 0.7042914390563965, "epoch": 8, "memory": 6318, "step": 31677} +{"lr": 2.0970600288576148e-05, "data_time": 0.5326156377792358, "loss": 0.18511711657047272, "time": 0.6814176082611084, "epoch": 8, "memory": 6318, "step": 31777} +{"lr": 2.0746580873819274e-05, "data_time": 0.4841681957244873, "loss": 0.16780608594417573, "time": 0.6248322486877441, "epoch": 8, "memory": 6318, "step": 31877} +{"lr": 2.0524561811901426e-05, "data_time": 0.5698666095733642, "loss": 0.1741199880838394, "time": 0.7101840496063232, "epoch": 8, "memory": 6318, "step": 31977} +{"lr": 2.0304556068432268e-05, "data_time": 0.570238733291626, "loss": 0.15130266547203064, "time": 0.7095555305480957, "epoch": 8, "memory": 6318, "step": 32077} +{"lr": 2.0086576491446333e-05, "data_time": 0.47088942527770994, "loss": 0.18402259349822997, "time": 0.6107290029525757, "epoch": 8, "memory": 6318, "step": 32177} +{"lr": 1.987063581065291e-05, "data_time": 0.5076226949691772, "loss": 0.17769983410835266, "time": 0.6468212366104126, "epoch": 8, "memory": 6318, "step": 32277} +{"lr": 1.965674663669245e-05, "data_time": 0.4999563455581665, "loss": 0.19281096756458282, "time": 0.6388433694839477, "epoch": 8, "memory": 6318, "step": 32377} +{"lr": 1.9444921460400247e-05, "data_time": 0.3367817163467407, "loss": 0.1737343743443489, "time": 0.6393883228302002, "epoch": 8, "memory": 6318, "step": 32477} +{"lr": 1.923517265207697e-05, "data_time": 0.23358845710754395, "loss": 0.174641951918602, "time": 0.6786292552947998, "epoch": 8, "memory": 6318, "step": 32577} +{"lr": 1.9027512460766204e-05, "data_time": 0.3714548110961914, "loss": 0.1874246269464493, "time": 0.6521084308624268, "epoch": 8, "memory": 6318, "step": 32677} +{"lr": 1.8821953013539136e-05, "data_time": 0.432128381729126, "loss": 0.17116330415010453, "time": 0.6260940313339234, "epoch": 8, "memory": 6318, "step": 32777} +{"lr": 1.8618506314786433e-05, "data_time": 0.43123228549957277, "loss": 0.1739860475063324, "time": 0.653626537322998, "epoch": 8, "memory": 6318, "step": 32877} +{"accuracy/top1": 94.42927551269531, "data_time": 0.31908140977223715, "time": 0.40818392038345336, "step": 8} +{"lr": 1.8395169032542028e-05, "data_time": 0.3467705249786377, "loss": 0.17853446155786515, "time": 0.6419506788253784, "epoch": 9, "memory": 6318, "step": 32988} +{"lr": 1.819621906460573e-05, "data_time": 0.46077885627746584, "loss": 0.16322045177221298, "time": 0.6007961988449096, "epoch": 9, "memory": 6318, "step": 33088} +{"lr": 1.799941838715077e-05, "data_time": 0.7201919794082642, "loss": 0.16671231687068938, "time": 0.8737469434738159, "epoch": 9, "memory": 6318, "step": 33188} +{"lr": 1.7804778493067625e-05, "data_time": 0.31894209384918215, "loss": 0.17341336756944656, "time": 0.7355125188827515, "epoch": 9, "memory": 6318, "step": 33288} +{"lr": 1.761231074905996e-05, "data_time": 0.3927642583847046, "loss": 0.17762706875801088, "time": 0.7605992317199707, "epoch": 9, "memory": 6318, "step": 33388} +{"lr": 1.742202639498083e-05, "data_time": 0.16396536827087402, "loss": 0.17904257625341416, "time": 0.6477619647979737, "epoch": 9, "memory": 6318, "step": 33488} +{"lr": 1.7233936543176347e-05, "data_time": 0.04417450428009033, "loss": 0.17295984625816346, "time": 0.7338368415832519, "epoch": 9, "memory": 6318, "step": 33588} +{"lr": 1.704805217783665e-05, "data_time": 0.0008935689926147461, "loss": 0.18471588641405107, "time": 0.6868875741958618, "epoch": 9, "memory": 6318, "step": 33688} +{"lr": 1.68643841543545e-05, "data_time": 0.0012085437774658203, "loss": 0.17745250910520555, "time": 0.8213760614395141, "epoch": 9, "memory": 6318, "step": 33788} +{"lr": 1.6682943198691327e-05, "data_time": 0.5312901496887207, "loss": 0.1759701833128929, "time": 0.6814525127410889, "epoch": 9, "memory": 6318, "step": 33888} +{"lr": 1.6503739906750822e-05, "data_time": 0.27212936878204347, "loss": 0.17120040208101273, "time": 0.6951136350631714, "epoch": 9, "memory": 6318, "step": 33988} +{"lr": 1.6326784743760187e-05, "data_time": 0.0600679874420166, "loss": 0.16045070067048073, "time": 0.6472667694091797, "epoch": 9, "memory": 6318, "step": 34088} +{"lr": 1.6152088043659013e-05, "data_time": 0.16998941898345948, "loss": 0.16141234263777732, "time": 0.7316301584243774, "epoch": 9, "memory": 6318, "step": 34188} +{"lr": 1.59796600084957e-05, "data_time": 0.0019454479217529297, "loss": 0.16706182956695556, "time": 0.6471722602844239, "epoch": 9, "memory": 6318, "step": 34288} +{"lr": 1.5809510707831687e-05, "data_time": 0.10176880359649658, "loss": 0.17744836509227752, "time": 0.6969247579574585, "epoch": 9, "memory": 6318, "step": 34388} +{"lr": 1.5641650078153473e-05, "data_time": 0.12099258899688721, "loss": 0.2010004073381424, "time": 0.6634851217269897, "epoch": 9, "memory": 6318, "step": 34488} +{"lr": 1.547608792229235e-05, "data_time": 0.03756430149078369, "loss": 0.17726728916168213, "time": 0.7295563220977783, "epoch": 9, "memory": 6318, "step": 34588} +{"lr": 1.5312833908851794e-05, "data_time": 0.07615690231323242, "loss": 0.18076025247573851, "time": 0.7693311929702759, "epoch": 9, "memory": 6318, "step": 34688} +{"lr": 1.5151897571642958e-05, "data_time": 0.22081241607666016, "loss": 0.17004195898771285, "time": 0.7372802734375, "epoch": 9, "memory": 6318, "step": 34788} +{"lr": 1.4993288309127877e-05, "data_time": 0.5547281265258789, "loss": 0.16769114434719085, "time": 0.6940059185028076, "epoch": 9, "memory": 6318, "step": 34888} +{"lr": 1.4837015383870593e-05, "data_time": 0.6260907649993896, "loss": 0.1616027683019638, "time": 0.7661632299423218, "epoch": 9, "memory": 6318, "step": 34988} +{"lr": 1.4683087921996264e-05, "data_time": 0.5476421594619751, "loss": 0.17030781656503677, "time": 0.687248420715332, "epoch": 9, "memory": 6318, "step": 35088} +{"lr": 1.453151491265819e-05, "data_time": 0.6143709182739258, "loss": 0.1779827207326889, "time": 0.7559512615203857, "epoch": 9, "memory": 6318, "step": 35188} +{"lr": 1.4382305207512847e-05, "data_time": 0.5709915161132812, "loss": 0.15990345180034637, "time": 0.7118298292160035, "epoch": 9, "memory": 6318, "step": 35288} +{"lr": 1.423546752020298e-05, "data_time": 0.5250293016433716, "loss": 0.18098425567150117, "time": 0.6652384757995605, "epoch": 9, "memory": 6318, "step": 35388} +{"lr": 1.4091010425848762e-05, "data_time": 0.5211315393447876, "loss": 0.20388921201229096, "time": 0.6633057832717896, "epoch": 9, "memory": 6318, "step": 35488} +{"lr": 1.3948942360546966e-05, "data_time": 0.5101306676864624, "loss": 0.17343863919377328, "time": 0.6527750015258789, "epoch": 9, "memory": 6318, "step": 35588} +{"lr": 1.3809271620878346e-05, "data_time": 0.5618909597396851, "loss": 0.18072735220193864, "time": 0.7015467405319213, "epoch": 9, "memory": 6318, "step": 35688} +{"lr": 1.3672006363423091e-05, "data_time": 0.5551931619644165, "loss": 0.19147171825170517, "time": 0.6961599826812744, "epoch": 9, "memory": 6318, "step": 35788} +{"lr": 1.3537154604284566e-05, "data_time": 0.514067006111145, "loss": 0.18362155109643935, "time": 0.6545395135879517, "epoch": 9, "memory": 6318, "step": 35888} +{"lr": 1.3404724218621087e-05, "data_time": 0.5974668979644775, "loss": 0.16670264750719072, "time": 0.7365384340286255, "epoch": 9, "memory": 6318, "step": 35988} +{"lr": 1.3274722940186121e-05, "data_time": 0.5556386709213257, "loss": 0.17231397479772567, "time": 0.6957824230194092, "epoch": 9, "memory": 6318, "step": 36088} +{"lr": 1.3147158360876541e-05, "data_time": 0.5237772464752197, "loss": 0.18398341089487075, "time": 0.6643592357635498, "epoch": 9, "memory": 6318, "step": 36188} +{"lr": 1.3022037930289352e-05, "data_time": 0.5662123441696167, "loss": 0.18018409311771394, "time": 0.7056707620620728, "epoch": 9, "memory": 6318, "step": 36288} +{"lr": 1.2899368955286627e-05, "data_time": 0.4856699228286743, "loss": 0.17277174741029738, "time": 0.6262874364852905, "epoch": 9, "memory": 6318, "step": 36388} +{"lr": 1.2779158599568751e-05, "data_time": 0.6819154977798462, "loss": 0.16570268720388412, "time": 0.8214737892150878, "epoch": 9, "memory": 6318, "step": 36488} +{"lr": 1.2661413883256134e-05, "data_time": 0.5092743158340454, "loss": 0.16690452620387078, "time": 0.6505688428878784, "epoch": 9, "memory": 6318, "step": 36588} +{"lr": 1.2546141682479203e-05, "data_time": 1.088001561164856, "loss": 0.1649048961699009, "time": 1.2272021532058717, "epoch": 9, "memory": 6318, "step": 36688} +{"lr": 1.2433348728976846e-05, "data_time": 0.6861448287963867, "loss": 0.171642704308033, "time": 0.8261954545974731, "epoch": 9, "memory": 6318, "step": 36788} +{"lr": 1.2323041609703344e-05, "data_time": 0.6412292718887329, "loss": 0.15848518610000611, "time": 0.7812103271484375, "epoch": 9, "memory": 6318, "step": 36888} +{"lr": 1.2215226766443614e-05, "data_time": 0.5432866334915161, "loss": 0.17100584656000137, "time": 0.6845231771469116, "epoch": 9, "memory": 6318, "step": 36988} +{"accuracy/top1": 94.54512786865234, "data_time": 0.3506173054377238, "time": 0.4392131467660268, "step": 9} +{"lr": 1.209847850894437e-05, "data_time": 0.24326579570770263, "loss": 0.15971334725618364, "time": 0.6647909879684448, "epoch": 10, "memory": 6318, "step": 37099} +{"lr": 1.1995942849201055e-05, "data_time": 0.6583766460418701, "loss": 0.15596782565116882, "time": 0.7985041856765747, "epoch": 10, "memory": 6318, "step": 37199} +{"lr": 1.189591856759183e-05, "data_time": 0.5123385429382324, "loss": 0.1760866865515709, "time": 0.6541135787963868, "epoch": 10, "memory": 6318, "step": 37299} +{"lr": 1.1798411505398016e-05, "data_time": 0.5055289983749389, "loss": 0.1699337735772133, "time": 0.6463497638702392, "epoch": 10, "memory": 6318, "step": 37399} +{"lr": 1.1703427356898678e-05, "data_time": 0.864405345916748, "loss": 0.16508603394031524, "time": 1.0076769590377808, "epoch": 10, "memory": 6318, "step": 37499} +{"lr": 1.1610971669038223e-05, "data_time": 0.5758638143539428, "loss": 0.17306023538112641, "time": 0.7169496297836304, "epoch": 10, "memory": 6318, "step": 37599} +{"lr": 1.1521049841102396e-05, "data_time": 0.5187180995941162, "loss": 0.16726333349943162, "time": 0.6584716796875, "epoch": 10, "memory": 6318, "step": 37699} +{"lr": 1.1433667124402997e-05, "data_time": 0.5278132200241089, "loss": 0.1680594876408577, "time": 0.6698819637298584, "epoch": 10, "memory": 6318, "step": 37799} +{"lr": 1.1348828621971181e-05, "data_time": 0.49249792098999023, "loss": 0.15358172878623008, "time": 0.6334294319152832, "epoch": 10, "memory": 6318, "step": 37899} +{"lr": 1.1266539288259486e-05, "data_time": 0.5400561094284058, "loss": 0.16024808287620546, "time": 0.6853986024856568, "epoch": 10, "memory": 6318, "step": 37999} +{"lr": 1.1186803928852477e-05, "data_time": 0.5870069026947021, "loss": 0.16915135011076926, "time": 0.726749324798584, "epoch": 10, "memory": 6318, "step": 38099} +{"lr": 1.1109627200186125e-05, "data_time": 0.5290631532669068, "loss": 0.16700850576162338, "time": 0.6680032253265381, "epoch": 10, "memory": 6318, "step": 38199} +{"lr": 1.1035013609275835e-05, "data_time": 0.4958626747131348, "loss": 0.16667198538780212, "time": 0.6355530023574829, "epoch": 10, "memory": 6318, "step": 38299} +{"lr": 1.0962967513453298e-05, "data_time": 0.513212776184082, "loss": 0.18055977672338486, "time": 0.659446382522583, "epoch": 10, "memory": 6318, "step": 38399} +{"lr": 1.0893493120111999e-05, "data_time": 0.5574291229248047, "loss": 0.17115170061588286, "time": 0.6971151828765869, "epoch": 10, "memory": 6318, "step": 38499} +{"lr": 1.0826594486461509e-05, "data_time": 0.5870451927185059, "loss": 0.18780031204223632, "time": 0.7271421909332275, "epoch": 10, "memory": 6318, "step": 38599} +{"lr": 1.0762275519290557e-05, "data_time": 0.53290696144104, "loss": 0.1806088149547577, "time": 0.673856544494629, "epoch": 10, "memory": 6318, "step": 38699} +{"lr": 1.0700539974738862e-05, "data_time": 0.5583453893661499, "loss": 0.19364455938339234, "time": 0.6982730865478516, "epoch": 10, "memory": 6318, "step": 38799} +{"lr": 1.064139145807783e-05, "data_time": 0.6159309387207031, "loss": 0.18038648068904878, "time": 0.7574919700622559, "epoch": 10, "memory": 6318, "step": 38899} +{"lr": 1.0584833423499947e-05, "data_time": 0.5298036336898804, "loss": 0.1769822955131531, "time": 0.6711828708648682, "epoch": 10, "memory": 6318, "step": 38999} +{"lr": 1.0530869173917099e-05, "data_time": 0.5743390798568726, "loss": 0.17203499972820283, "time": 0.7146025419235229, "epoch": 10, "memory": 6318, "step": 39099} +{"lr": 1.0479501860767654e-05, "data_time": 0.603572940826416, "loss": 0.16231856718659401, "time": 0.7425681591033936, "epoch": 10, "memory": 6318, "step": 39199} +{"lr": 1.043073448383247e-05, "data_time": 0.547653317451477, "loss": 0.1617319703102112, "time": 0.6864744186401367, "epoch": 10, "memory": 6318, "step": 39299} +{"lr": 1.038456989105968e-05, "data_time": 0.5811430931091308, "loss": 0.1714554473757744, "time": 0.7210165739059449, "epoch": 10, "memory": 6318, "step": 39399} +{"lr": 1.034101077839838e-05, "data_time": 0.5828921079635621, "loss": 0.16311801671981813, "time": 0.7227906942367553, "epoch": 10, "memory": 6318, "step": 39499} +{"lr": 1.0300059689641194e-05, "data_time": 0.46419026851654055, "loss": 0.17278907746076583, "time": 0.6029969453811646, "epoch": 10, "memory": 6318, "step": 39599} +{"lr": 1.0261719016275683e-05, "data_time": 0.4624385118484497, "loss": 0.18718645274639129, "time": 0.6022111177444458, "epoch": 10, "memory": 6318, "step": 39699} +{"lr": 1.0225990997344757e-05, "data_time": 0.47218492031097414, "loss": 0.16667576581239701, "time": 0.6113409042358399, "epoch": 10, "memory": 6318, "step": 39799} +{"lr": 1.019287771931587e-05, "data_time": 0.5410808324813843, "loss": 0.18095844238996506, "time": 0.7533207893371582, "epoch": 10, "memory": 6318, "step": 39899} +{"lr": 1.0162381115959187e-05, "data_time": 0.1821967840194702, "loss": 0.18518355935811998, "time": 0.6399319410324097, "epoch": 10, "memory": 6318, "step": 39999} +{"lr": 1.0134502968234636e-05, "data_time": 0.15167815685272218, "loss": 0.15409796461462974, "time": 0.6462615013122559, "epoch": 10, "memory": 6318, "step": 40099} +{"lr": 1.0109244904187918e-05, "data_time": 0.6154161691665649, "loss": 0.16570380181074143, "time": 0.7540058612823486, "epoch": 10, "memory": 6318, "step": 40199} +{"lr": 1.0086608398855446e-05, "data_time": 0.4992237567901611, "loss": 0.16055154353380202, "time": 0.6389468669891357, "epoch": 10, "memory": 6318, "step": 40299} +{"lr": 1.0066594774178176e-05, "data_time": 0.5433523178100585, "loss": 0.17121364176273346, "time": 0.6839509725570678, "epoch": 10, "memory": 6318, "step": 40399} +{"lr": 1.0049205198924413e-05, "data_time": 0.5597763061523438, "loss": 0.16367753297090532, "time": 0.7084070920944214, "epoch": 10, "memory": 6318, "step": 40499} +{"lr": 1.0034440688621577e-05, "data_time": 0.5803892612457275, "loss": 0.1932917758822441, "time": 0.7189661026000976, "epoch": 10, "memory": 6318, "step": 40599} +{"lr": 1.0022302105496896e-05, "data_time": 0.5081663131713867, "loss": 0.16676972433924675, "time": 0.6487033843994141, "epoch": 10, "memory": 6318, "step": 40699} +{"lr": 1.0012790158427017e-05, "data_time": 0.5436817169189453, "loss": 0.1766459122300148, "time": 0.6827889680862427, "epoch": 10, "memory": 6318, "step": 40799} +{"lr": 1.000590540289662e-05, "data_time": 0.5389787673950195, "loss": 0.1522808313369751, "time": 0.6787208557128906, "epoch": 10, "memory": 6318, "step": 40899} +{"lr": 1.0001648240966046e-05, "data_time": 0.5397365808486938, "loss": 0.17131981998682022, "time": 0.680224347114563, "epoch": 10, "memory": 6318, "step": 40999} +{"lr": 1.0000018921247718e-05, "data_time": 0.558778715133667, "loss": 0.1690695308148861, "time": 0.6985341310501099, "epoch": 10, "memory": 6318, "step": 41099} +{"accuracy/top1": 94.70069885253906, "data_time": 0.3315279563268026, "time": 0.42175288597742716, "step": 10} diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/0000000.png_1.png b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/0000000.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..04a4b38fc60c262c15e2420dc49544a61a232fa8 Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/0000000.png_1.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/0000000.png_10.png b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/0000000.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..7efdea6cb4c59f63a166ea313fcd6d93a9e5ea81 Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/0000000.png_10.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/0000000.png_2.png b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/0000000.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..dab0e40b264ee25adcd6df21548ddb2ad3ee3c1a Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/0000000.png_2.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/0000000.png_3.png b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/0000000.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..7174dbd6e5506947e3737fb2ffc757a95e249541 Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/0000000.png_3.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/0000000.png_4.png b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/0000000.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..081282ee00914ec7a93d8a9fc0e913bc961bd0a8 Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/0000000.png_4.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/0000000.png_5.png b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/0000000.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..07e370580cd89e060c380268baff18ad77564fac Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/0000000.png_5.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/0000000.png_6.png b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/0000000.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..a4f43d871ec4380f248a4c2131b02685ea0ba0eb Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/0000000.png_6.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/0000000.png_7.png b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/0000000.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..e204d7f918c8a09e12c393d98f95beae9c21fe59 Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/0000000.png_7.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/0000000.png_8.png b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/0000000.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..f229679c6eed931848978615b87f9579a7083653 Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/0000000.png_8.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/0000000.png_9.png b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/0000000.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..72eece39d1aa0eaa8b6f8bc71e70e27ef29f2319 Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/0000000.png_9.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/000003093.jpg_1.png b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/000003093.jpg_1.png new file mode 100644 index 0000000000000000000000000000000000000000..91abb99766d7877909b4ec6567b33256d0141ad4 Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/000003093.jpg_1.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/000003093.jpg_10.png b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/000003093.jpg_10.png new file mode 100644 index 0000000000000000000000000000000000000000..4d1556b514e3a657efd66aadba68a7269eee065d Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/000003093.jpg_10.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/000003093.jpg_2.png b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/000003093.jpg_2.png new file mode 100644 index 0000000000000000000000000000000000000000..54db0053928ccf798e734864287b19a389afabdc Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/000003093.jpg_2.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/000003093.jpg_3.png b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/000003093.jpg_3.png new file mode 100644 index 0000000000000000000000000000000000000000..16dff298e446c0b34587782dc7e98486cef3e3fa Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/000003093.jpg_3.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/000003093.jpg_4.png b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/000003093.jpg_4.png new file mode 100644 index 0000000000000000000000000000000000000000..48bd2598774759afec744da3fc630700a787b18a Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/000003093.jpg_4.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/000003093.jpg_5.png b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/000003093.jpg_5.png new file mode 100644 index 0000000000000000000000000000000000000000..2620a3cad15922f1f7dc0902952a052c76a9aeaf Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/000003093.jpg_5.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/000003093.jpg_6.png b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/000003093.jpg_6.png new file mode 100644 index 0000000000000000000000000000000000000000..76f8dd15bde10058ced8f8d4dd678b8961474dc8 Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/000003093.jpg_6.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/000003093.jpg_7.png b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/000003093.jpg_7.png new file mode 100644 index 0000000000000000000000000000000000000000..159e443b5fa550dcec849d98d4b4fe3dfef5792c Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/000003093.jpg_7.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/000003093.jpg_8.png b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/000003093.jpg_8.png new file mode 100644 index 0000000000000000000000000000000000000000..5eff715736806a8a79c17f8e5b9ac7f277056d03 Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/000003093.jpg_8.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/000003093.jpg_9.png b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/000003093.jpg_9.png new file mode 100644 index 0000000000000000000000000000000000000000..5bd1ad14b9f1be37f85063dc013cfc7c87d628f9 Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/000003093.jpg_9.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/000015337.jpg_1.png b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/000015337.jpg_1.png new file mode 100644 index 0000000000000000000000000000000000000000..5c09ec1ceed87366a821731e56667d20a88e0cf9 Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/000015337.jpg_1.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/000015337.jpg_10.png b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/000015337.jpg_10.png new file mode 100644 index 0000000000000000000000000000000000000000..803089c4c3b9e47aebeb716df061b45a002c4be2 Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/000015337.jpg_10.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/000015337.jpg_2.png b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/000015337.jpg_2.png new file mode 100644 index 0000000000000000000000000000000000000000..4fb5c32dba4fc72dabea2e8dc438915a8d17e53c Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/000015337.jpg_2.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/000015337.jpg_3.png b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/000015337.jpg_3.png new file mode 100644 index 0000000000000000000000000000000000000000..1fe1ac7437a0ae7d7872e8f28486ce8cbdbf557c Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/000015337.jpg_3.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/000015337.jpg_4.png b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/000015337.jpg_4.png new file mode 100644 index 0000000000000000000000000000000000000000..2ca6956fe783f6a8c1f614d6e00f7bd679016971 Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/000015337.jpg_4.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/000015337.jpg_5.png b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/000015337.jpg_5.png new file mode 100644 index 0000000000000000000000000000000000000000..5d1d9c1d0603dbf153f98f9d0a303589c83b161d Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/000015337.jpg_5.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/000015337.jpg_6.png b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/000015337.jpg_6.png new file mode 100644 index 0000000000000000000000000000000000000000..790a1d09fdaa3de119e477a35903b11dd890f4bd Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/000015337.jpg_6.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/000015337.jpg_7.png b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/000015337.jpg_7.png new file mode 100644 index 0000000000000000000000000000000000000000..4616eb1e7ce1a298ec979d6f1cb6f3951747fd94 Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/000015337.jpg_7.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/000015337.jpg_8.png b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/000015337.jpg_8.png new file mode 100644 index 0000000000000000000000000000000000000000..fa5dee42441317c729b5731be3f6e5ddddbbdd97 Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/000015337.jpg_8.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/000015337.jpg_9.png b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/000015337.jpg_9.png new file mode 100644 index 0000000000000000000000000000000000000000..5119e5014940799ea72e3b82362f2cd15241a73c Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/000015337.jpg_9.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/0004159.png_1.png b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/0004159.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..cc995b53b6df1a648995f78ac130eb006a587ebf Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/0004159.png_1.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/0004159.png_10.png b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/0004159.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..a66a5823fd2b0650079bfb3cd8d24fb762764331 Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/0004159.png_10.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/0004159.png_2.png b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/0004159.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..797042197dfcad67369441f8a3e77988e99d365d Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/0004159.png_2.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/0004159.png_3.png b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/0004159.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..6c358f0b4d219a21041b9f22a5eec71d8ed94fcb Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/0004159.png_3.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/0004159.png_4.png b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/0004159.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..158f3cc43bb882aaed78665bb9b30eb4d173f794 Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/0004159.png_4.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/0004159.png_5.png b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/0004159.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..faa4fe1c9bcdcdacfae35a805946f4ff1032ddf2 Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/0004159.png_5.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/0004159.png_6.png b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/0004159.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..4ac6cb3112875a83dfa45b3e0cf5e052410faaaf Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/0004159.png_6.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/0004159.png_7.png b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/0004159.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..0576cbc51d8093621c8a8891c6297c00d465f0b8 Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/0004159.png_7.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/0004159.png_8.png b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/0004159.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..473cc51dcc503004bb472d1fb0516712bc5b2e8b Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/0004159.png_8.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/0004159.png_9.png b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/0004159.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..1483890a8f6e4038aee142d9c1fb90bb773e202f Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/0004159.png_9.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/0008318.png_1.png b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/0008318.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..4e1aa7e4f7e5d7005b6f7c5cf28817e302fb676a Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/0008318.png_1.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/0008318.png_10.png b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/0008318.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..fabe4d347b52ce3e652545916d43f1b0e63d2125 Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/0008318.png_10.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/0008318.png_2.png b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/0008318.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..6ed43d3ee1acedf01a42604522f392aef711e995 Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/0008318.png_2.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/0008318.png_3.png b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/0008318.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..cf14b7407f0c44b3bf96afc59bd31d630fc1c772 Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/0008318.png_3.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/0008318.png_4.png b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/0008318.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..ebf55a96102d45ba1eeed8f8b8f870ad510b2891 Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/0008318.png_4.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/0008318.png_5.png b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/0008318.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..fa57fd66b621f971fd2b71dd1bf621f98c52ca16 Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/0008318.png_5.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/0008318.png_6.png b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/0008318.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..16d9dba93a0f065258f241f4002b0bc473898c89 Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/0008318.png_6.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/0008318.png_7.png b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/0008318.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..e2b1f88bd3f03aba84be0bd3086f146f59329a9b Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/0008318.png_7.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/0008318.png_8.png b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/0008318.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..b1aa138c9361663e42981907437595cd1855494c Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/0008318.png_8.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/0008318.png_9.png b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/0008318.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..c97e85fe3840dfd7a16f2e04ec9152efaf3c0e15 Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/0008318.png_9.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/0010000.png_1.png b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/0010000.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..4692d52bf340d2a92ee7a8c863f70f0331a81702 Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/0010000.png_1.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/0010000.png_10.png b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/0010000.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..c0007f2c6896bfae8b567f328bbe1e49b187f17e Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/0010000.png_10.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/0010000.png_2.png b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/0010000.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..66c8a45d49385a6d05538046d3a11c8f3bfadb83 Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/0010000.png_2.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/0010000.png_3.png b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/0010000.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..6ce784948610ab3800b736b14eda71dd2017af46 Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/0010000.png_3.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/0010000.png_4.png b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/0010000.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..fb59756a3b465fbbc98b84ed64cf2822d90ab457 Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/0010000.png_4.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/0010000.png_5.png b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/0010000.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..65f5c63d6d79c7c0e0a19441b852bc0611c85d00 Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/0010000.png_5.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/0010000.png_6.png b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/0010000.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..1ea3cbad7c94d174df4db09ce9baa64da00eea2d Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/0010000.png_6.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/0010000.png_7.png b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/0010000.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..766347cd0a8b1396d82d520fe104fb037a463c4e Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/0010000.png_7.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/0010000.png_8.png b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/0010000.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..f579a2a965243367fccddc9e088b9b8e4bdc7085 Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/0010000.png_8.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/0010000.png_9.png b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/0010000.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..608642d92ab68de0e259123a8a9394a9fe5f584c Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/0010000.png_9.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/0014159.png_1.png b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/0014159.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..892e8a9acd7aa8c14ea6b8b49aff7b6f1eb7af12 Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/0014159.png_1.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/0014159.png_10.png b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/0014159.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..478e3faf706a8372260de2014a11b66e35d5f6ad Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/0014159.png_10.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/0014159.png_2.png b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/0014159.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..a29ab62deefbd686853716ed9c5ff7275cec664c Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/0014159.png_2.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/0014159.png_3.png b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/0014159.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..2e80328524ff31438af3f54547bf52ead952966c Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/0014159.png_3.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/0014159.png_4.png b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/0014159.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..d57745fa7ae01bb20518ec91a22484ac040e7a35 Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/0014159.png_4.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/0014159.png_5.png b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/0014159.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..982eccfc77125c91e8110dc490e736e266bbc8fa Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/0014159.png_5.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/0014159.png_6.png b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/0014159.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..f474619a550f29c7f29bde4601ea5b7e77585719 Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/0014159.png_6.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/0014159.png_7.png b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/0014159.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..7ccecf0f98d249f54f6bcff6a252fb553a542a9a Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/0014159.png_7.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/0014159.png_8.png b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/0014159.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..e8af7958fbc49ef8b16e2449f1cb37c3a32234f2 Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/0014159.png_8.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/0014159.png_9.png b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/0014159.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..5e7b66e7fffbdf5c8b4998a6861af688da7545a4 Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/20230605_042304/vis_data/vis_image/0014159.png_9.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/epoch_1.pth b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/epoch_1.pth new file mode 100644 index 0000000000000000000000000000000000000000..4fe724c649c54e65748fdc9d711c0dd41dece57c --- /dev/null +++ b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/epoch_1.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:381e7b2bbbd9f344f1fa48f8f3bf8611d095af66b937fd0529fbddd1ae824b17 +size 277359501 diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/epoch_10.pth b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/epoch_10.pth new file mode 100644 index 0000000000000000000000000000000000000000..b8a47c8c8cd664b681390c78871567c91d74fcd2 --- /dev/null +++ b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/epoch_10.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:9dea8c2970749cbe6e20d6d2294f29e4af143d4bfd5fc5734ca806dd3b0b5832 +size 280090957 diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/epoch_2.pth b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/epoch_2.pth new file mode 100644 index 0000000000000000000000000000000000000000..6dfae2565d2bee4c77781d64bc8cf2d40c10ffd4 --- /dev/null +++ b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/epoch_2.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:b25917b6d1c7d747c29ce3729e15fdec26d086293ccc8cf04b30cd3acca6af8e +size 277661581 diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/epoch_3.pth b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/epoch_3.pth new file mode 100644 index 0000000000000000000000000000000000000000..1ec2aab0dabae1b315456b2471322c92d91f3723 --- /dev/null +++ b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/epoch_3.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:04f6a6a394284fcb8b132c8d115aeb891419e9634b23e9cf9cacb4c63a2ff645 +size 277963853 diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/epoch_4.pth b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/epoch_4.pth new file mode 100644 index 0000000000000000000000000000000000000000..0383b5f2b45e48f5ab413fb32a54dda575fe385a --- /dev/null +++ b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/epoch_4.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:fa8ea846c8caf7cd94c9436e9ddfdd7d2da542f763c3f3441c8fa2acd53a5476 +size 278265933 diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/epoch_5.pth b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/epoch_5.pth new file mode 100644 index 0000000000000000000000000000000000000000..4fd5ce6801ba17da2fdcddb05348ce24a135fa62 --- /dev/null +++ b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/epoch_5.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:817449dd78343764cff4b5c348f067e8ee26d6575c44fc41ac212a879cefef46 +size 278568525 diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/epoch_6.pth b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/epoch_6.pth new file mode 100644 index 0000000000000000000000000000000000000000..7f5ac11a35eabcbc74337feb1b39776924595912 --- /dev/null +++ b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/epoch_6.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:75e7d67bf9cc7e28d70af0d9efb61e49b393c6cc8daf7be441ff21c62ed24f31 +size 278870797 diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/epoch_7.pth b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/epoch_7.pth new file mode 100644 index 0000000000000000000000000000000000000000..85d84da8683ae96ca158e16ee378f22782d7bce6 --- /dev/null +++ b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/epoch_7.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:53f8b38bdaa459d209eccb60442675a0bc14e6e845e7d10a585a97c45ca5926a +size 279173645 diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/epoch_8.pth b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/epoch_8.pth new file mode 100644 index 0000000000000000000000000000000000000000..305b388487d1f3bfbb827f7957b49dd706651247 --- /dev/null +++ b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/epoch_8.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:1cb6311d181176fe57c0cfbc853ad5c6918f9e3699d07af5d0359690ebc9bff0 +size 279479821 diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/epoch_9.pth b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/epoch_9.pth new file mode 100644 index 0000000000000000000000000000000000000000..720e86ad8b842b61bebdf6d7351bbb6ce1cd488e --- /dev/null +++ b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/epoch_9.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:a7111f3783ee15a57f90ec98afa870a72c6b10772d0d4ba46e9b800d357abe28 +size 279785549 diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/last_checkpoint b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/last_checkpoint new file mode 100644 index 0000000000000000000000000000000000000000..7c06f6155edc9e9ba963f3538105fe2a9ce2c5cc --- /dev/null +++ b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/last_checkpoint @@ -0,0 +1 @@ +/mnt/petrelfs/luzeyu/workspace/fakebench/mmpretrain/workdir/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/epoch_10.pth \ No newline at end of file diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1.py b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1.py new file mode 100644 index 0000000000000000000000000000000000000000..e1db61d967e52e39f7a4971468760d3ff016bef0 --- /dev/null +++ b/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1.py @@ -0,0 +1,366 @@ +optim_wrapper = dict( + optimizer=dict( + type='SGD', + lr=0.0001, + momentum=0.9, + weight_decay=0.0001, + _scope_='mmpretrain'), + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=False, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=512) +model = dict( + type='ImageClassifier', + backbone=dict( + frozen_stages=2, + type='ResNet', + depth=50, + num_stages=4, + out_indices=(3, ), + style='pytorch', + init_cfg=dict( + type='Pretrained', + checkpoint= + 'https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth', + prefix='backbone')), + neck=dict(type='GlobalAveragePooling'), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=2048, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + topk=1)) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/all1.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/all2.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/all3.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3fake8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/cc1m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3real8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV2-1-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/if-dpmsolver++-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = dict(type='Accuracy', topk=1) +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV2-1-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/if-dpmsolver++-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = dict(type='Accuracy', topk=1) +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/resnet50_2xb256_all_1m_lr1e-4_aug_1e-1' diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_043015/20230605_043015.log b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_043015/20230605_043015.log new file mode 100644 index 0000000000000000000000000000000000000000..0d7d4bb0e5932a099d996bba9e714211c3793620 --- /dev/null +++ b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_043015/20230605_043015.log @@ -0,0 +1,1363 @@ +2023/06/05 04:30:19 - mmengine - INFO - +------------------------------------------------------------ +System environment: + sys.platform: linux + Python: 3.10.9 (main, Mar 8 2023, 10:47:38) [GCC 11.2.0] + CUDA available: True + numpy_random_seed: 2061627647 + GPU 0,1: NVIDIA A100-SXM4-80GB + CUDA_HOME: /mnt/petrelfs/share/cuda-11.6 + NVCC: Cuda compilation tools, release 11.6, V11.6.124 + GCC: gcc (GCC) 7.5.0 + PyTorch: 1.13.1 + PyTorch compiling details: PyTorch built with: + - GCC 9.3 + - C++ Version: 201402 + - Intel(R) oneAPI Math Kernel Library Version 2021.4-Product Build 20210904 for Intel(R) 64 architecture applications + - Intel(R) MKL-DNN v2.6.0 (Git Hash 52b5f107dd9cf10910aaa19cb47f3abf9b349815) + - OpenMP 201511 (a.k.a. OpenMP 4.5) + - LAPACK is enabled (usually provided by MKL) + - NNPACK is enabled + - CPU capability usage: AVX2 + - CUDA Runtime 11.6 + - NVCC architecture flags: -gencode;arch=compute_37,code=sm_37;-gencode;arch=compute_50,code=sm_50;-gencode;arch=compute_60,code=sm_60;-gencode;arch=compute_61,code=sm_61;-gencode;arch=compute_70,code=sm_70;-gencode;arch=compute_75,code=sm_75;-gencode;arch=compute_80,code=sm_80;-gencode;arch=compute_86,code=sm_86;-gencode;arch=compute_37,code=compute_37 + - CuDNN 8.3.2 (built against CUDA 11.5) + - Magma 2.6.1 + - Build settings: BLAS_INFO=mkl, BUILD_TYPE=Release, CUDA_VERSION=11.6, CUDNN_VERSION=8.3.2, CXX_COMPILER=/opt/rh/devtoolset-9/root/usr/bin/c++, CXX_FLAGS= -fabi-version=11 -Wno-deprecated -fvisibility-inlines-hidden -DUSE_PTHREADPOOL -fopenmp -DNDEBUG -DUSE_KINETO -DUSE_FBGEMM -DUSE_QNNPACK -DUSE_PYTORCH_QNNPACK -DUSE_XNNPACK -DSYMBOLICATE_MOBILE_DEBUG_HANDLE -DEDGE_PROFILER_USE_KINETO -O2 -fPIC -Wno-narrowing -Wall -Wextra -Werror=return-type -Werror=non-virtual-dtor -Wno-missing-field-initializers -Wno-type-limits -Wno-array-bounds -Wno-unknown-pragmas -Wunused-local-typedefs -Wno-unused-parameter -Wno-unused-function -Wno-unused-result -Wno-strict-overflow -Wno-strict-aliasing -Wno-error=deprecated-declarations -Wno-stringop-overflow -Wno-psabi -Wno-error=pedantic -Wno-error=redundant-decls -Wno-error=old-style-cast -fdiagnostics-color=always -faligned-new -Wno-unused-but-set-variable -Wno-maybe-uninitialized -fno-math-errno -fno-trapping-math -Werror=format -Werror=cast-function-type -Wno-stringop-overflow, LAPACK_INFO=mkl, PERF_WITH_AVX=1, PERF_WITH_AVX2=1, PERF_WITH_AVX512=1, TORCH_VERSION=1.13.1, USE_CUDA=ON, USE_CUDNN=ON, USE_EXCEPTION_PTR=1, USE_GFLAGS=OFF, USE_GLOG=OFF, USE_MKL=ON, USE_MKLDNN=ON, USE_MPI=OFF, USE_NCCL=ON, USE_NNPACK=ON, USE_OPENMP=ON, USE_ROCM=OFF, + + TorchVision: 0.14.1 + OpenCV: 4.7.0 + MMEngine: 0.7.3 + +Runtime environment: + cudnn_benchmark: True + mp_cfg: {'mp_start_method': 'fork', 'opencv_num_threads': 0} + dist_cfg: {'backend': 'nccl'} + seed: None + deterministic: False + Distributed launcher: slurm + Distributed training: True + GPU number: 2 +------------------------------------------------------------ + +2023/06/05 04:30:23 - mmengine - INFO - Config: +optim_wrapper = dict( + optimizer=dict( + type='SGD', + lr=0.0001, + momentum=0.9, + weight_decay=0.0001, + _scope_='mmpretrain'), + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=False, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=512) +model = dict( + type='ImageClassifier', + backbone=dict( + frozen_stages=2, + type='ResNet', + depth=50, + num_stages=4, + out_indices=(3, ), + style='pytorch', + init_cfg=dict( + type='Pretrained', + checkpoint= + 'https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth', + prefix='backbone')), + neck=dict(type='GlobalAveragePooling'), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=2048, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + topk=1)) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/all1.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/all2.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/all3.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3fake8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/cc1m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3real8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV2-1-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/if-dpmsolver++-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = dict(type='Accuracy', topk=1) +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV2-1-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/if-dpmsolver++-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = dict(type='Accuracy', topk=1) +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1' + +2023/06/05 04:30:37 - mmengine - INFO - Hooks will be executed in the following order: +before_run: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_load_checkpoint: +(ABOVE_NORMAL) EMAHook + -------------------- +before_train: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_train_epoch: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(NORMAL ) DistSamplerSeedHook + -------------------- +before_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook + -------------------- +after_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +after_train_epoch: +(NORMAL ) IterTimerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_val_epoch: +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook + -------------------- +before_val_iter: +(NORMAL ) IterTimerHook + -------------------- +after_val_iter: +(NORMAL ) IterTimerHook +(NORMAL ) VisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_val_epoch: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_save_checkpoint: +(ABOVE_NORMAL) EMAHook + -------------------- +after_train: +(VERY_LOW ) CheckpointHook + -------------------- +before_test_epoch: +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook + -------------------- +before_test_iter: +(NORMAL ) IterTimerHook + -------------------- +after_test_iter: +(NORMAL ) IterTimerHook +(NORMAL ) VisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_test_epoch: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_run: +(BELOW_NORMAL) LoggerHook + -------------------- +2023/06/05 04:30:59 - mmengine - INFO - load backbone in model from: https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth +Name of parameter - Initialization information + +backbone.conv1.weight - torch.Size([64, 3, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.bn1.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.bn1.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.conv1.weight - torch.Size([64, 64, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.bn1.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.bn1.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.conv2.weight - torch.Size([64, 64, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.bn2.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.bn2.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.conv3.weight - torch.Size([256, 64, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.bn3.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.bn3.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.downsample.0.weight - torch.Size([256, 64, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.downsample.1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.downsample.1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.conv1.weight - torch.Size([64, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.bn1.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.bn1.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.conv2.weight - torch.Size([64, 64, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.bn2.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.bn2.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.conv3.weight - torch.Size([256, 64, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.bn3.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.bn3.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.conv1.weight - torch.Size([64, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.bn1.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.bn1.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.conv2.weight - torch.Size([64, 64, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.bn2.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.bn2.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.conv3.weight - torch.Size([256, 64, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.bn3.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.bn3.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.conv1.weight - torch.Size([128, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.bn1.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.bn1.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.conv2.weight - torch.Size([128, 128, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.bn2.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.bn2.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.conv3.weight - torch.Size([512, 128, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.bn3.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.bn3.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.downsample.0.weight - torch.Size([512, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.downsample.1.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.downsample.1.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.conv1.weight - torch.Size([128, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.bn1.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.bn1.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.conv2.weight - torch.Size([128, 128, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.bn2.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.bn2.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.conv3.weight - torch.Size([512, 128, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.bn3.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.bn3.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.conv1.weight - torch.Size([128, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.bn1.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.bn1.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.conv2.weight - torch.Size([128, 128, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.bn2.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.bn2.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.conv3.weight - torch.Size([512, 128, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.bn3.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.bn3.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.conv1.weight - torch.Size([128, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.bn1.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.bn1.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.conv2.weight - torch.Size([128, 128, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.bn2.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.bn2.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.conv3.weight - torch.Size([512, 128, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.bn3.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.bn3.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.conv1.weight - torch.Size([256, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.bn1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.bn1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.conv2.weight - torch.Size([256, 256, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.bn2.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.bn2.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.conv3.weight - torch.Size([1024, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.bn3.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.bn3.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.downsample.0.weight - torch.Size([1024, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.downsample.1.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.downsample.1.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.conv1.weight - torch.Size([256, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.bn1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.bn1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.conv2.weight - torch.Size([256, 256, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.bn2.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.bn2.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.conv3.weight - torch.Size([1024, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.bn3.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.bn3.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.conv1.weight - torch.Size([256, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.bn1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.bn1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.conv2.weight - torch.Size([256, 256, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.bn2.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.bn2.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.conv3.weight - torch.Size([1024, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.bn3.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.bn3.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.conv1.weight - torch.Size([256, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.bn1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.bn1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.conv2.weight - torch.Size([256, 256, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.bn2.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.bn2.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.conv3.weight - torch.Size([1024, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.bn3.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.bn3.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.conv1.weight - torch.Size([256, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.bn1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.bn1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.conv2.weight - torch.Size([256, 256, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.bn2.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.bn2.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.conv3.weight - torch.Size([1024, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.bn3.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.bn3.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.conv1.weight - torch.Size([256, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.bn1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.bn1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.conv2.weight - torch.Size([256, 256, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.bn2.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.bn2.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.conv3.weight - torch.Size([1024, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.bn3.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.bn3.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.conv1.weight - torch.Size([512, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.bn1.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.bn1.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.conv2.weight - torch.Size([512, 512, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.bn2.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.bn2.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.conv3.weight - torch.Size([2048, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.bn3.weight - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.bn3.bias - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.downsample.0.weight - torch.Size([2048, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.downsample.1.weight - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.downsample.1.bias - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.conv1.weight - torch.Size([512, 2048, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.bn1.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.bn1.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.conv2.weight - torch.Size([512, 512, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.bn2.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.bn2.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.conv3.weight - torch.Size([2048, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.bn3.weight - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.bn3.bias - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.conv1.weight - torch.Size([512, 2048, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.bn1.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.bn1.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.conv2.weight - torch.Size([512, 512, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.bn2.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.bn2.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.conv3.weight - torch.Size([2048, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.bn3.weight - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.bn3.bias - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +head.fc.weight - torch.Size([2, 2048]): +NormalInit: mean=0, std=0.01, bias=0 + +head.fc.bias - torch.Size([2]): +NormalInit: mean=0, std=0.01, bias=0 +2023/06/05 04:30:59 - mmengine - WARNING - "FileClient" will be deprecated in future. Please use io functions in https://mmengine.readthedocs.io/en/latest/api/fileio.html#file-io +2023/06/05 04:30:59 - mmengine - WARNING - "HardDiskBackend" is the alias of "LocalBackend" and the former will be deprecated in future. +2023/06/05 04:30:59 - mmengine - INFO - Checkpoints will be saved to /mnt/petrelfs/luzeyu/workspace/fakebench/mmpretrain/workdir/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1. +2023/06/05 04:32:16 - mmengine - INFO - Epoch(train) [1][ 100/4111] lr: 9.9999e-05 eta: 8:45:25 time: 0.7237 data_time: 0.5834 memory: 9436 loss: 0.6458 +2023/06/05 04:33:34 - mmengine - INFO - Epoch(train) [1][ 200/4111] lr: 9.9995e-05 eta: 8:49:36 time: 0.9103 data_time: 0.7669 memory: 6319 loss: 0.6040 +2023/06/05 04:34:45 - mmengine - INFO - Epoch(train) [1][ 300/4111] lr: 9.9988e-05 eta: 8:33:42 time: 0.7237 data_time: 0.5825 memory: 6319 loss: 0.5770 +2023/06/05 04:35:59 - mmengine - INFO - Epoch(train) [1][ 400/4111] lr: 9.9979e-05 eta: 8:28:24 time: 0.7018 data_time: 0.5610 memory: 6319 loss: 0.5383 +2023/06/05 04:37:12 - mmengine - INFO - Epoch(train) [1][ 500/4111] lr: 9.9967e-05 eta: 8:24:25 time: 0.7145 data_time: 0.5732 memory: 6319 loss: 0.5250 +2023/06/05 04:38:27 - mmengine - INFO - Epoch(train) [1][ 600/4111] lr: 9.9953e-05 eta: 8:23:43 time: 0.8062 data_time: 0.6640 memory: 6319 loss: 0.5100 +2023/06/05 04:39:42 - mmengine - INFO - Epoch(train) [1][ 700/4111] lr: 9.9936e-05 eta: 8:23:22 time: 0.9325 data_time: 0.7929 memory: 6319 loss: 0.4634 +2023/06/05 04:40:56 - mmengine - INFO - Epoch(train) [1][ 800/4111] lr: 9.9916e-05 eta: 8:21:30 time: 0.7230 data_time: 0.5825 memory: 6319 loss: 0.4538 +2023/06/05 04:42:10 - mmengine - INFO - Epoch(train) [1][ 900/4111] lr: 9.9894e-05 eta: 8:19:59 time: 0.7644 data_time: 0.6241 memory: 6319 loss: 0.4483 +2023/06/05 04:43:24 - mmengine - INFO - Exp name: resnet50_2xb256_all_1m_lr1e-4_aug_5e-1_20230605_043015 +2023/06/05 04:43:24 - mmengine - INFO - Epoch(train) [1][1000/4111] lr: 9.9869e-05 eta: 8:18:19 time: 0.7761 data_time: 0.6361 memory: 6319 loss: 0.4394 +2023/06/05 04:44:37 - mmengine - INFO - Epoch(train) [1][1100/4111] lr: 9.9841e-05 eta: 8:15:45 time: 0.7667 data_time: 0.6265 memory: 6319 loss: 0.4465 +2023/06/05 04:45:49 - mmengine - INFO - Epoch(train) [1][1200/4111] lr: 9.9811e-05 eta: 8:13:07 time: 0.7231 data_time: 0.5830 memory: 6319 loss: 0.4156 +2023/06/05 04:47:02 - mmengine - INFO - Epoch(train) [1][1300/4111] lr: 9.9778e-05 eta: 8:11:25 time: 0.7630 data_time: 0.6212 memory: 6319 loss: 0.4051 +2023/06/05 04:48:16 - mmengine - INFO - Epoch(train) [1][1400/4111] lr: 9.9743e-05 eta: 8:10:18 time: 0.6811 data_time: 0.5405 memory: 6319 loss: 0.3932 +2023/06/05 04:49:29 - mmengine - INFO - Epoch(train) [1][1500/4111] lr: 9.9705e-05 eta: 8:08:38 time: 0.7653 data_time: 0.6264 memory: 6319 loss: 0.3778 +2023/06/05 04:50:42 - mmengine - INFO - Epoch(train) [1][1600/4111] lr: 9.9664e-05 eta: 8:07:02 time: 0.7213 data_time: 0.5815 memory: 6319 loss: 0.3830 +2023/06/05 04:51:56 - mmengine - INFO - Epoch(train) [1][1700/4111] lr: 9.9621e-05 eta: 8:05:50 time: 0.7056 data_time: 0.5648 memory: 6319 loss: 0.3915 +2023/06/05 04:53:09 - mmengine - INFO - Epoch(train) [1][1800/4111] lr: 9.9575e-05 eta: 8:04:15 time: 0.7551 data_time: 0.6145 memory: 6319 loss: 0.3696 +2023/06/05 04:54:22 - mmengine - INFO - Epoch(train) [1][1900/4111] lr: 9.9527e-05 eta: 8:02:45 time: 0.7103 data_time: 0.5710 memory: 6319 loss: 0.3797 +2023/06/05 04:55:35 - mmengine - INFO - Exp name: resnet50_2xb256_all_1m_lr1e-4_aug_5e-1_20230605_043015 +2023/06/05 04:55:35 - mmengine - INFO - Epoch(train) [1][2000/4111] lr: 9.9476e-05 eta: 8:01:03 time: 0.7544 data_time: 0.6139 memory: 6319 loss: 0.3580 +2023/06/05 04:56:50 - mmengine - INFO - Epoch(train) [1][2100/4111] lr: 9.9422e-05 eta: 8:00:08 time: 0.7179 data_time: 0.5774 memory: 6319 loss: 0.3667 +2023/06/05 04:58:04 - mmengine - INFO - Epoch(train) [1][2200/4111] lr: 9.9366e-05 eta: 7:59:01 time: 0.7219 data_time: 0.5804 memory: 6319 loss: 0.3506 +2023/06/05 04:59:16 - mmengine - INFO - Epoch(train) [1][2300/4111] lr: 9.9307e-05 eta: 7:57:13 time: 0.6951 data_time: 0.5541 memory: 6319 loss: 0.3724 +2023/06/05 05:00:30 - mmengine - INFO - Epoch(train) [1][2400/4111] lr: 9.9246e-05 eta: 7:56:13 time: 0.7629 data_time: 0.6224 memory: 6319 loss: 0.3514 +2023/06/05 05:01:42 - mmengine - INFO - Epoch(train) [1][2500/4111] lr: 9.9182e-05 eta: 7:54:19 time: 0.7212 data_time: 0.5803 memory: 6319 loss: 0.3630 +2023/06/05 05:02:55 - mmengine - INFO - Epoch(train) [1][2600/4111] lr: 9.9115e-05 eta: 7:53:03 time: 0.7224 data_time: 0.5827 memory: 6319 loss: 0.3590 +2023/06/05 05:04:10 - mmengine - INFO - Epoch(train) [1][2700/4111] lr: 9.9046e-05 eta: 7:52:02 time: 0.7428 data_time: 0.6027 memory: 6319 loss: 0.3396 +2023/06/05 05:05:21 - mmengine - INFO - Epoch(train) [1][2800/4111] lr: 9.8974e-05 eta: 7:50:16 time: 0.7444 data_time: 0.6042 memory: 6319 loss: 0.3580 +2023/06/05 05:06:34 - mmengine - INFO - Epoch(train) [1][2900/4111] lr: 9.8900e-05 eta: 7:48:55 time: 0.6727 data_time: 0.5304 memory: 6319 loss: 0.3416 +2023/06/05 05:07:48 - mmengine - INFO - Exp name: resnet50_2xb256_all_1m_lr1e-4_aug_5e-1_20230605_043015 +2023/06/05 05:07:48 - mmengine - INFO - Epoch(train) [1][3000/4111] lr: 9.8823e-05 eta: 7:47:41 time: 0.6958 data_time: 0.5570 memory: 6319 loss: 0.3409 +2023/06/05 05:09:01 - mmengine - INFO - Epoch(train) [1][3100/4111] lr: 9.8744e-05 eta: 7:46:23 time: 0.7817 data_time: 0.6411 memory: 6319 loss: 0.3344 +2023/06/05 05:10:13 - mmengine - INFO - Epoch(train) [1][3200/4111] lr: 9.8662e-05 eta: 7:44:45 time: 0.7727 data_time: 0.6330 memory: 6319 loss: 0.3328 +2023/06/05 05:11:27 - mmengine - INFO - Epoch(train) [1][3300/4111] lr: 9.8578e-05 eta: 7:43:39 time: 0.7490 data_time: 0.6098 memory: 6319 loss: 0.3460 +2023/06/05 05:12:39 - mmengine - INFO - Epoch(train) [1][3400/4111] lr: 9.8490e-05 eta: 7:42:02 time: 0.7145 data_time: 0.5731 memory: 6319 loss: 0.3187 +2023/06/05 05:13:53 - mmengine - INFO - Epoch(train) [1][3500/4111] lr: 9.8401e-05 eta: 7:41:03 time: 0.7148 data_time: 0.5743 memory: 6319 loss: 0.3388 +2023/06/05 05:15:08 - mmengine - INFO - Epoch(train) [1][3600/4111] lr: 9.8309e-05 eta: 7:39:56 time: 0.6919 data_time: 0.5523 memory: 6319 loss: 0.3252 +2023/06/05 05:16:23 - mmengine - INFO - Epoch(train) [1][3700/4111] lr: 9.8214e-05 eta: 7:39:05 time: 0.7662 data_time: 0.6264 memory: 6319 loss: 0.3327 +2023/06/05 05:17:39 - mmengine - INFO - Epoch(train) [1][3800/4111] lr: 9.8117e-05 eta: 7:38:14 time: 0.7660 data_time: 0.6109 memory: 6319 loss: 0.3214 +2023/06/05 05:18:52 - mmengine - INFO - Epoch(train) [1][3900/4111] lr: 9.8017e-05 eta: 7:36:47 time: 0.7375 data_time: 0.5972 memory: 6319 loss: 0.3192 +2023/06/05 05:20:06 - mmengine - INFO - Exp name: resnet50_2xb256_all_1m_lr1e-4_aug_5e-1_20230605_043015 +2023/06/05 05:20:06 - mmengine - INFO - Epoch(train) [1][4000/4111] lr: 9.7915e-05 eta: 7:35:43 time: 0.7446 data_time: 0.6052 memory: 6319 loss: 0.3149 +2023/06/05 05:21:18 - mmengine - INFO - Epoch(train) [1][4100/4111] lr: 9.7810e-05 eta: 7:34:09 time: 0.7587 data_time: 0.6189 memory: 6319 loss: 0.3222 +2023/06/05 05:21:26 - mmengine - INFO - Exp name: resnet50_2xb256_all_1m_lr1e-4_aug_5e-1_20230605_043015 +2023/06/05 05:21:26 - mmengine - INFO - Saving checkpoint at 1 epochs +2023/06/05 05:22:14 - mmengine - INFO - Epoch(val) [1][100/119] eta: 0:00:08 time: 0.7489 data_time: 0.6575 memory: 8106 +2023/06/05 05:22:43 - mmengine - INFO - Epoch(val) [1][119/119] accuracy/top1: 82.1376 data_time: 0.3993 time: 0.4887 +2023/06/05 05:23:58 - mmengine - INFO - Epoch(train) [2][ 100/4111] lr: 9.7691e-05 eta: 7:33:08 time: 0.7325 data_time: 0.5902 memory: 6318 loss: 0.3236 +2023/06/05 05:25:12 - mmengine - INFO - Epoch(train) [2][ 200/4111] lr: 9.7581e-05 eta: 7:31:54 time: 0.7627 data_time: 0.6221 memory: 6318 loss: 0.3212 +2023/06/05 05:26:26 - mmengine - INFO - Epoch(train) [2][ 300/4111] lr: 9.7469e-05 eta: 7:30:47 time: 0.7274 data_time: 0.5857 memory: 6318 loss: 0.3311 +2023/06/05 05:27:40 - mmengine - INFO - Epoch(train) [2][ 400/4111] lr: 9.7354e-05 eta: 7:29:35 time: 0.7569 data_time: 0.6168 memory: 6318 loss: 0.3035 +2023/06/05 05:28:56 - mmengine - INFO - Epoch(train) [2][ 500/4111] lr: 9.7236e-05 eta: 7:28:37 time: 0.7368 data_time: 0.5955 memory: 6318 loss: 0.3121 +2023/06/05 05:30:10 - mmengine - INFO - Epoch(train) [2][ 600/4111] lr: 9.7116e-05 eta: 7:27:25 time: 0.7184 data_time: 0.5795 memory: 6318 loss: 0.3228 +2023/06/05 05:31:25 - mmengine - INFO - Epoch(train) [2][ 700/4111] lr: 9.6994e-05 eta: 7:26:23 time: 0.7987 data_time: 0.6580 memory: 6318 loss: 0.3153 +2023/06/05 05:32:40 - mmengine - INFO - Epoch(train) [2][ 800/4111] lr: 9.6869e-05 eta: 7:25:18 time: 0.7385 data_time: 0.5990 memory: 6318 loss: 0.3126 +2023/06/05 05:33:41 - mmengine - INFO - Exp name: resnet50_2xb256_all_1m_lr1e-4_aug_5e-1_20230605_043015 +2023/06/05 05:33:54 - mmengine - INFO - Epoch(train) [2][ 900/4111] lr: 9.6742e-05 eta: 7:24:05 time: 0.7581 data_time: 0.6177 memory: 6318 loss: 0.3055 +2023/06/05 05:35:06 - mmengine - INFO - Epoch(train) [2][1000/4111] lr: 9.6612e-05 eta: 7:22:38 time: 0.7713 data_time: 0.6324 memory: 6318 loss: 0.3197 +2023/06/05 05:36:19 - mmengine - INFO - Epoch(train) [2][1100/4111] lr: 9.6480e-05 eta: 7:21:16 time: 0.6825 data_time: 0.5424 memory: 6318 loss: 0.3171 +2023/06/05 05:37:31 - mmengine - INFO - Epoch(train) [2][1200/4111] lr: 9.6346e-05 eta: 7:19:51 time: 0.7205 data_time: 0.5808 memory: 6318 loss: 0.3128 +2023/06/05 05:38:44 - mmengine - INFO - Epoch(train) [2][1300/4111] lr: 9.6209e-05 eta: 7:18:33 time: 0.7196 data_time: 0.5788 memory: 6318 loss: 0.3116 +2023/06/05 05:39:58 - mmengine - INFO - Epoch(train) [2][1400/4111] lr: 9.6069e-05 eta: 7:17:21 time: 0.7109 data_time: 0.5683 memory: 6318 loss: 0.3233 +2023/06/05 05:41:11 - mmengine - INFO - Epoch(train) [2][1500/4111] lr: 9.5928e-05 eta: 7:16:04 time: 0.7785 data_time: 0.6389 memory: 6318 loss: 0.3070 +2023/06/05 05:42:23 - mmengine - INFO - Epoch(train) [2][1600/4111] lr: 9.5783e-05 eta: 7:14:38 time: 0.7593 data_time: 0.6137 memory: 6318 loss: 0.2970 +2023/06/05 05:43:36 - mmengine - INFO - Epoch(train) [2][1700/4111] lr: 9.5637e-05 eta: 7:13:22 time: 0.7305 data_time: 0.5899 memory: 6318 loss: 0.2924 +2023/06/05 05:44:49 - mmengine - INFO - Epoch(train) [2][1800/4111] lr: 9.5488e-05 eta: 7:12:01 time: 0.7623 data_time: 0.6219 memory: 6318 loss: 0.3016 +2023/06/05 05:45:50 - mmengine - INFO - Exp name: resnet50_2xb256_all_1m_lr1e-4_aug_5e-1_20230605_043015 +2023/06/05 05:46:03 - mmengine - INFO - Epoch(train) [2][1900/4111] lr: 9.5337e-05 eta: 7:10:53 time: 0.7546 data_time: 0.6148 memory: 6318 loss: 0.3011 +2023/06/05 05:47:16 - mmengine - INFO - Epoch(train) [2][2000/4111] lr: 9.5183e-05 eta: 7:09:35 time: 0.7169 data_time: 0.5771 memory: 6318 loss: 0.2821 +2023/06/05 05:48:31 - mmengine - INFO - Epoch(train) [2][2100/4111] lr: 9.5027e-05 eta: 7:08:25 time: 0.7317 data_time: 0.5909 memory: 6318 loss: 0.2922 +2023/06/05 05:49:44 - mmengine - INFO - Epoch(train) [2][2200/4111] lr: 9.4869e-05 eta: 7:07:13 time: 0.7633 data_time: 0.6227 memory: 6318 loss: 0.3060 +2023/06/05 05:50:59 - mmengine - INFO - Epoch(train) [2][2300/4111] lr: 9.4708e-05 eta: 7:06:03 time: 0.6975 data_time: 0.5570 memory: 6318 loss: 0.2896 +2023/06/05 05:52:13 - mmengine - INFO - Epoch(train) [2][2400/4111] lr: 9.4545e-05 eta: 7:04:52 time: 0.7587 data_time: 0.6190 memory: 6318 loss: 0.2919 +2023/06/05 05:53:26 - mmengine - INFO - Epoch(train) [2][2500/4111] lr: 9.4380e-05 eta: 7:03:36 time: 0.6919 data_time: 0.5511 memory: 6318 loss: 0.2964 +2023/06/05 05:54:37 - mmengine - INFO - Epoch(train) [2][2600/4111] lr: 9.4212e-05 eta: 7:02:10 time: 0.7404 data_time: 0.5999 memory: 6318 loss: 0.3184 +2023/06/05 05:55:51 - mmengine - INFO - Epoch(train) [2][2700/4111] lr: 9.4043e-05 eta: 7:00:54 time: 0.7273 data_time: 0.5878 memory: 6318 loss: 0.2691 +2023/06/05 05:57:03 - mmengine - INFO - Epoch(train) [2][2800/4111] lr: 9.3870e-05 eta: 6:59:34 time: 0.7112 data_time: 0.5716 memory: 6318 loss: 0.2867 +2023/06/05 05:58:10 - mmengine - INFO - Exp name: resnet50_2xb256_all_1m_lr1e-4_aug_5e-1_20230605_043015 +2023/06/05 05:58:19 - mmengine - INFO - Epoch(train) [2][2900/4111] lr: 9.3696e-05 eta: 6:58:30 time: 0.8053 data_time: 0.6654 memory: 6318 loss: 0.2685 +2023/06/05 05:59:34 - mmengine - INFO - Epoch(train) [2][3000/4111] lr: 9.3519e-05 eta: 6:57:26 time: 0.6782 data_time: 0.5378 memory: 6318 loss: 0.2746 +2023/06/05 06:00:47 - mmengine - INFO - Epoch(train) [2][3100/4111] lr: 9.3340e-05 eta: 6:56:09 time: 0.7422 data_time: 0.6017 memory: 6318 loss: 0.2676 +2023/06/05 06:02:02 - mmengine - INFO - Epoch(train) [2][3200/4111] lr: 9.3159e-05 eta: 6:55:02 time: 0.7827 data_time: 0.6432 memory: 6318 loss: 0.2729 +2023/06/05 06:03:14 - mmengine - INFO - Epoch(train) [2][3300/4111] lr: 9.2976e-05 eta: 6:53:40 time: 0.7189 data_time: 0.5789 memory: 6318 loss: 0.2948 +2023/06/05 06:04:28 - mmengine - INFO - Epoch(train) [2][3400/4111] lr: 9.2790e-05 eta: 6:52:28 time: 0.8703 data_time: 0.7300 memory: 6318 loss: 0.2894 +2023/06/05 06:05:44 - mmengine - INFO - Epoch(train) [2][3500/4111] lr: 9.2603e-05 eta: 6:51:22 time: 0.7461 data_time: 0.5967 memory: 6318 loss: 0.2779 +2023/06/05 06:06:57 - mmengine - INFO - Epoch(train) [2][3600/4111] lr: 9.2413e-05 eta: 6:50:06 time: 0.7983 data_time: 0.6364 memory: 6318 loss: 0.2850 +2023/06/05 06:08:12 - mmengine - INFO - Epoch(train) [2][3700/4111] lr: 9.2220e-05 eta: 6:48:58 time: 0.7547 data_time: 0.6148 memory: 6318 loss: 0.2755 +2023/06/05 06:09:27 - mmengine - INFO - Epoch(train) [2][3800/4111] lr: 9.2026e-05 eta: 6:47:51 time: 0.7754 data_time: 0.6347 memory: 6318 loss: 0.2761 +2023/06/05 06:10:36 - mmengine - INFO - Exp name: resnet50_2xb256_all_1m_lr1e-4_aug_5e-1_20230605_043015 +2023/06/05 06:10:43 - mmengine - INFO - Epoch(train) [2][3900/4111] lr: 9.1829e-05 eta: 6:46:47 time: 0.7377 data_time: 0.5981 memory: 6318 loss: 0.2720 +2023/06/05 06:11:57 - mmengine - INFO - Epoch(train) [2][4000/4111] lr: 9.1631e-05 eta: 6:45:34 time: 0.7340 data_time: 0.5946 memory: 6318 loss: 0.2814 +2023/06/05 06:13:11 - mmengine - INFO - Epoch(train) [2][4100/4111] lr: 9.1430e-05 eta: 6:44:23 time: 0.6995 data_time: 0.5590 memory: 6318 loss: 0.2801 +2023/06/05 06:13:18 - mmengine - INFO - Exp name: resnet50_2xb256_all_1m_lr1e-4_aug_5e-1_20230605_043015 +2023/06/05 06:13:18 - mmengine - INFO - Saving checkpoint at 2 epochs +2023/06/05 06:14:05 - mmengine - INFO - Epoch(val) [2][100/119] eta: 0:00:07 time: 0.8455 data_time: 0.7543 memory: 6318 +2023/06/05 06:14:33 - mmengine - INFO - Epoch(val) [2][119/119] accuracy/top1: 85.1646 data_time: 0.3821 time: 0.4728 +2023/06/05 06:15:50 - mmengine - INFO - Epoch(train) [3][ 100/4111] lr: 9.1205e-05 eta: 6:43:09 time: 0.7175 data_time: 0.5781 memory: 6318 loss: 0.2836 +2023/06/05 06:17:07 - mmengine - INFO - Epoch(train) [3][ 200/4111] lr: 9.0999e-05 eta: 6:42:05 time: 0.6776 data_time: 0.5375 memory: 6318 loss: 0.2693 +2023/06/05 06:18:29 - mmengine - INFO - Epoch(train) [3][ 300/4111] lr: 9.0792e-05 eta: 6:41:26 time: 0.9346 data_time: 0.7944 memory: 6318 loss: 0.2957 +2023/06/05 06:19:46 - mmengine - INFO - Epoch(train) [3][ 400/4111] lr: 9.0582e-05 eta: 6:40:22 time: 0.7733 data_time: 0.6330 memory: 6318 loss: 0.2778 +2023/06/05 06:21:03 - mmengine - INFO - Epoch(train) [3][ 500/4111] lr: 9.0371e-05 eta: 6:39:21 time: 0.8288 data_time: 0.6890 memory: 6318 loss: 0.2761 +2023/06/05 06:22:21 - mmengine - INFO - Epoch(train) [3][ 600/4111] lr: 9.0157e-05 eta: 6:38:20 time: 0.8028 data_time: 0.6631 memory: 6318 loss: 0.2906 +2023/06/05 06:23:37 - mmengine - INFO - Epoch(train) [3][ 700/4111] lr: 8.9942e-05 eta: 6:37:13 time: 0.7149 data_time: 0.5741 memory: 6318 loss: 0.3003 +2023/06/05 06:24:38 - mmengine - INFO - Exp name: resnet50_2xb256_all_1m_lr1e-4_aug_5e-1_20230605_043015 +2023/06/05 06:24:54 - mmengine - INFO - Epoch(train) [3][ 800/4111] lr: 8.9724e-05 eta: 6:36:11 time: 0.7742 data_time: 0.6332 memory: 6318 loss: 0.2660 +2023/06/05 06:26:09 - mmengine - INFO - Epoch(train) [3][ 900/4111] lr: 8.9504e-05 eta: 6:34:58 time: 0.7672 data_time: 0.6184 memory: 6318 loss: 0.2671 +2023/06/05 06:27:20 - mmengine - INFO - Epoch(train) [3][1000/4111] lr: 8.9282e-05 eta: 6:33:35 time: 0.7322 data_time: 0.5923 memory: 6318 loss: 0.2828 +2023/06/05 06:28:34 - mmengine - INFO - Epoch(train) [3][1100/4111] lr: 8.9059e-05 eta: 6:32:21 time: 0.7666 data_time: 0.6252 memory: 6318 loss: 0.2615 +2023/06/05 06:29:48 - mmengine - INFO - Epoch(train) [3][1200/4111] lr: 8.8833e-05 eta: 6:31:07 time: 0.7139 data_time: 0.5713 memory: 6318 loss: 0.2783 +2023/06/05 06:30:58 - mmengine - INFO - Epoch(train) [3][1300/4111] lr: 8.8605e-05 eta: 6:29:40 time: 0.6606 data_time: 0.5197 memory: 6318 loss: 0.2669 +2023/06/05 06:32:15 - mmengine - INFO - Epoch(train) [3][1400/4111] lr: 8.8375e-05 eta: 6:28:36 time: 0.7681 data_time: 0.6248 memory: 6318 loss: 0.2652 +2023/06/05 06:33:30 - mmengine - INFO - Epoch(train) [3][1500/4111] lr: 8.8144e-05 eta: 6:27:24 time: 0.7508 data_time: 0.6111 memory: 6318 loss: 0.2883 +2023/06/05 06:34:44 - mmengine - INFO - Epoch(train) [3][1600/4111] lr: 8.7910e-05 eta: 6:26:10 time: 0.7789 data_time: 0.6371 memory: 6318 loss: 0.2538 +2023/06/05 06:35:58 - mmengine - INFO - Epoch(train) [3][1700/4111] lr: 8.7675e-05 eta: 6:24:54 time: 0.7568 data_time: 0.6164 memory: 6318 loss: 0.2568 +2023/06/05 06:36:52 - mmengine - INFO - Exp name: resnet50_2xb256_all_1m_lr1e-4_aug_5e-1_20230605_043015 +2023/06/05 06:37:11 - mmengine - INFO - Epoch(train) [3][1800/4111] lr: 8.7437e-05 eta: 6:23:39 time: 0.6756 data_time: 0.5362 memory: 6318 loss: 0.2945 +2023/06/05 06:38:32 - mmengine - INFO - Epoch(train) [3][1900/4111] lr: 8.7198e-05 eta: 6:22:44 time: 0.7427 data_time: 0.6020 memory: 6318 loss: 0.2514 +2023/06/05 06:39:48 - mmengine - INFO - Epoch(train) [3][2000/4111] lr: 8.6957e-05 eta: 6:21:35 time: 0.7496 data_time: 0.6082 memory: 6318 loss: 0.2642 +2023/06/05 06:41:03 - mmengine - INFO - Epoch(train) [3][2100/4111] lr: 8.6714e-05 eta: 6:20:24 time: 0.7483 data_time: 0.6089 memory: 6318 loss: 0.2693 +2023/06/05 06:42:17 - mmengine - INFO - Epoch(train) [3][2200/4111] lr: 8.6469e-05 eta: 6:19:09 time: 0.7361 data_time: 0.5897 memory: 6318 loss: 0.2740 +2023/06/05 06:43:31 - mmengine - INFO - Epoch(train) [3][2300/4111] lr: 8.6222e-05 eta: 6:17:55 time: 0.7437 data_time: 0.6015 memory: 6318 loss: 0.2598 +2023/06/05 06:44:45 - mmengine - INFO - Epoch(train) [3][2400/4111] lr: 8.5974e-05 eta: 6:16:42 time: 0.7424 data_time: 0.6019 memory: 6318 loss: 0.2573 +2023/06/05 06:45:58 - mmengine - INFO - Epoch(train) [3][2500/4111] lr: 8.5723e-05 eta: 6:15:25 time: 0.6946 data_time: 0.5541 memory: 6318 loss: 0.2633 +2023/06/05 06:47:12 - mmengine - INFO - Epoch(train) [3][2600/4111] lr: 8.5471e-05 eta: 6:14:08 time: 0.7374 data_time: 0.5862 memory: 6318 loss: 0.2482 +2023/06/05 06:48:23 - mmengine - INFO - Epoch(train) [3][2700/4111] lr: 8.5217e-05 eta: 6:12:46 time: 0.6802 data_time: 0.5396 memory: 6318 loss: 0.2803 +2023/06/05 06:49:14 - mmengine - INFO - Exp name: resnet50_2xb256_all_1m_lr1e-4_aug_5e-1_20230605_043015 +2023/06/05 06:49:34 - mmengine - INFO - Epoch(train) [3][2800/4111] lr: 8.4962e-05 eta: 6:11:25 time: 0.7176 data_time: 0.5594 memory: 6318 loss: 0.2621 +2023/06/05 06:50:45 - mmengine - INFO - Epoch(train) [3][2900/4111] lr: 8.4704e-05 eta: 6:10:03 time: 0.7138 data_time: 0.5735 memory: 6318 loss: 0.2568 +2023/06/05 06:51:59 - mmengine - INFO - Epoch(train) [3][3000/4111] lr: 8.4445e-05 eta: 6:08:48 time: 0.7514 data_time: 0.6116 memory: 6318 loss: 0.2490 +2023/06/05 06:53:12 - mmengine - INFO - Epoch(train) [3][3100/4111] lr: 8.4184e-05 eta: 6:07:31 time: 0.6965 data_time: 0.5562 memory: 6318 loss: 0.2693 +2023/06/05 06:54:24 - mmengine - INFO - Epoch(train) [3][3200/4111] lr: 8.3921e-05 eta: 6:06:11 time: 0.7037 data_time: 0.5575 memory: 6318 loss: 0.2406 +2023/06/05 06:55:38 - mmengine - INFO - Epoch(train) [3][3300/4111] lr: 8.3657e-05 eta: 6:04:57 time: 0.7339 data_time: 0.5934 memory: 6318 loss: 0.2416 +2023/06/05 06:56:48 - mmengine - INFO - Epoch(train) [3][3400/4111] lr: 8.3391e-05 eta: 6:03:34 time: 0.7273 data_time: 0.5866 memory: 6318 loss: 0.2674 +2023/06/05 06:58:03 - mmengine - INFO - Epoch(train) [3][3500/4111] lr: 8.3123e-05 eta: 6:02:22 time: 0.7548 data_time: 0.6135 memory: 6318 loss: 0.2591 +2023/06/05 06:59:18 - mmengine - INFO - Epoch(train) [3][3600/4111] lr: 8.2854e-05 eta: 6:01:09 time: 0.7723 data_time: 0.6322 memory: 6318 loss: 0.2432 +2023/06/05 07:00:39 - mmengine - INFO - Epoch(train) [3][3700/4111] lr: 8.2583e-05 eta: 6:00:12 time: 0.7632 data_time: 0.6222 memory: 6318 loss: 0.2620 +2023/06/05 07:01:30 - mmengine - INFO - Exp name: resnet50_2xb256_all_1m_lr1e-4_aug_5e-1_20230605_043015 +2023/06/05 07:01:52 - mmengine - INFO - Epoch(train) [3][3800/4111] lr: 8.2311e-05 eta: 5:58:56 time: 0.7383 data_time: 0.5979 memory: 6318 loss: 0.2561 +2023/06/05 07:03:09 - mmengine - INFO - Epoch(train) [3][3900/4111] lr: 8.2037e-05 eta: 5:57:51 time: 0.8333 data_time: 0.6930 memory: 6318 loss: 0.2536 +2023/06/05 07:04:24 - mmengine - INFO - Epoch(train) [3][4000/4111] lr: 8.1761e-05 eta: 5:56:37 time: 0.7675 data_time: 0.6269 memory: 6318 loss: 0.2563 +2023/06/05 07:05:37 - mmengine - INFO - Epoch(train) [3][4100/4111] lr: 8.1484e-05 eta: 5:55:22 time: 0.7358 data_time: 0.5956 memory: 6318 loss: 0.2519 +2023/06/05 07:05:44 - mmengine - INFO - Exp name: resnet50_2xb256_all_1m_lr1e-4_aug_5e-1_20230605_043015 +2023/06/05 07:05:44 - mmengine - INFO - Saving checkpoint at 3 epochs +2023/06/05 07:06:31 - mmengine - INFO - Epoch(val) [3][100/119] eta: 0:00:07 time: 0.7675 data_time: 0.6765 memory: 6318 +2023/06/05 07:07:00 - mmengine - INFO - Epoch(val) [3][119/119] accuracy/top1: 83.9581 data_time: 0.3894 time: 0.4792 +2023/06/05 07:08:17 - mmengine - INFO - Epoch(train) [4][ 100/4111] lr: 8.1174e-05 eta: 5:54:05 time: 0.7959 data_time: 0.6538 memory: 6318 loss: 0.2366 +2023/06/05 07:09:34 - mmengine - INFO - Epoch(train) [4][ 200/4111] lr: 8.0894e-05 eta: 5:52:58 time: 0.7718 data_time: 0.6240 memory: 6318 loss: 0.2417 +2023/06/05 07:10:50 - mmengine - INFO - Epoch(train) [4][ 300/4111] lr: 8.0612e-05 eta: 5:51:47 time: 0.7352 data_time: 0.5951 memory: 6318 loss: 0.2568 +2023/06/05 07:12:07 - mmengine - INFO - Epoch(train) [4][ 400/4111] lr: 8.0328e-05 eta: 5:50:39 time: 0.7188 data_time: 0.2905 memory: 6318 loss: 0.2458 +2023/06/05 07:13:22 - mmengine - INFO - Epoch(train) [4][ 500/4111] lr: 8.0043e-05 eta: 5:49:26 time: 0.7400 data_time: 0.4857 memory: 6318 loss: 0.2530 +2023/06/05 07:14:37 - mmengine - INFO - Epoch(train) [4][ 600/4111] lr: 7.9757e-05 eta: 5:48:14 time: 0.7419 data_time: 0.3014 memory: 6318 loss: 0.2484 +2023/06/05 07:15:28 - mmengine - INFO - Exp name: resnet50_2xb256_all_1m_lr1e-4_aug_5e-1_20230605_043015 +2023/06/05 07:15:52 - mmengine - INFO - Epoch(train) [4][ 700/4111] lr: 7.9469e-05 eta: 5:47:02 time: 0.8202 data_time: 0.6789 memory: 6318 loss: 0.2391 +2023/06/05 07:17:07 - mmengine - INFO - Epoch(train) [4][ 800/4111] lr: 7.9180e-05 eta: 5:45:49 time: 0.7638 data_time: 0.6203 memory: 6318 loss: 0.2586 +2023/06/05 07:18:21 - mmengine - INFO - Epoch(train) [4][ 900/4111] lr: 7.8889e-05 eta: 5:44:36 time: 0.7317 data_time: 0.5903 memory: 6318 loss: 0.2585 +2023/06/05 07:19:43 - mmengine - INFO - Epoch(train) [4][1000/4111] lr: 7.8597e-05 eta: 5:43:38 time: 0.7474 data_time: 0.6059 memory: 6318 loss: 0.2260 +2023/06/05 07:21:00 - mmengine - INFO - Epoch(train) [4][1100/4111] lr: 7.8303e-05 eta: 5:42:28 time: 0.7612 data_time: 0.6215 memory: 6318 loss: 0.2503 +2023/06/05 07:22:23 - mmengine - INFO - Epoch(train) [4][1200/4111] lr: 7.8008e-05 eta: 5:41:31 time: 0.7009 data_time: 0.5605 memory: 6318 loss: 0.2542 +2023/06/05 07:23:37 - mmengine - INFO - Epoch(train) [4][1300/4111] lr: 7.7712e-05 eta: 5:40:17 time: 0.7578 data_time: 0.6174 memory: 6318 loss: 0.2565 +2023/06/05 07:24:53 - mmengine - INFO - Epoch(train) [4][1400/4111] lr: 7.7415e-05 eta: 5:39:07 time: 0.8520 data_time: 0.7111 memory: 6318 loss: 0.2498 +2023/06/05 07:26:08 - mmengine - INFO - Epoch(train) [4][1500/4111] lr: 7.7116e-05 eta: 5:37:54 time: 0.6926 data_time: 0.5510 memory: 6318 loss: 0.2502 +2023/06/05 07:27:27 - mmengine - INFO - Epoch(train) [4][1600/4111] lr: 7.6816e-05 eta: 5:36:48 time: 0.7112 data_time: 0.5713 memory: 6318 loss: 0.2363 +2023/06/05 07:28:18 - mmengine - INFO - Exp name: resnet50_2xb256_all_1m_lr1e-4_aug_5e-1_20230605_043015 +2023/06/05 07:28:41 - mmengine - INFO - Epoch(train) [4][1700/4111] lr: 7.6514e-05 eta: 5:35:33 time: 0.7617 data_time: 0.6209 memory: 6318 loss: 0.2541 +2023/06/05 07:29:58 - mmengine - INFO - Epoch(train) [4][1800/4111] lr: 7.6212e-05 eta: 5:34:23 time: 0.7496 data_time: 0.6097 memory: 6318 loss: 0.2435 +2023/06/05 07:31:11 - mmengine - INFO - Epoch(train) [4][1900/4111] lr: 7.5908e-05 eta: 5:33:08 time: 0.7485 data_time: 0.6082 memory: 6318 loss: 0.2418 +2023/06/05 07:32:26 - mmengine - INFO - Epoch(train) [4][2000/4111] lr: 7.5603e-05 eta: 5:31:54 time: 0.7510 data_time: 0.6105 memory: 6318 loss: 0.2273 +2023/06/05 07:33:40 - mmengine - INFO - Epoch(train) [4][2100/4111] lr: 7.5296e-05 eta: 5:30:39 time: 0.7382 data_time: 0.5980 memory: 6318 loss: 0.2368 +2023/06/05 07:35:03 - mmengine - INFO - Epoch(train) [4][2200/4111] lr: 7.4989e-05 eta: 5:29:40 time: 0.7455 data_time: 0.6029 memory: 6318 loss: 0.2513 +2023/06/05 07:36:18 - mmengine - INFO - Epoch(train) [4][2300/4111] lr: 7.4680e-05 eta: 5:28:26 time: 0.7461 data_time: 0.6061 memory: 6318 loss: 0.2445 +2023/06/05 07:37:32 - mmengine - INFO - Epoch(train) [4][2400/4111] lr: 7.4370e-05 eta: 5:27:11 time: 0.7388 data_time: 0.5987 memory: 6318 loss: 0.2446 +2023/06/05 07:38:45 - mmengine - INFO - Epoch(train) [4][2500/4111] lr: 7.4059e-05 eta: 5:25:55 time: 0.7198 data_time: 0.5786 memory: 6318 loss: 0.2439 +2023/06/05 07:40:01 - mmengine - INFO - Epoch(train) [4][2600/4111] lr: 7.3747e-05 eta: 5:24:42 time: 0.7635 data_time: 0.6222 memory: 6318 loss: 0.2538 +2023/06/05 07:40:52 - mmengine - INFO - Exp name: resnet50_2xb256_all_1m_lr1e-4_aug_5e-1_20230605_043015 +2023/06/05 07:41:16 - mmengine - INFO - Epoch(train) [4][2700/4111] lr: 7.3434e-05 eta: 5:23:29 time: 0.8363 data_time: 0.6962 memory: 6318 loss: 0.2388 +2023/06/05 07:42:32 - mmengine - INFO - Epoch(train) [4][2800/4111] lr: 7.3120e-05 eta: 5:22:18 time: 0.7793 data_time: 0.6280 memory: 6318 loss: 0.2328 +2023/06/05 07:43:48 - mmengine - INFO - Epoch(train) [4][2900/4111] lr: 7.2804e-05 eta: 5:21:05 time: 0.7282 data_time: 0.5790 memory: 6318 loss: 0.2476 +2023/06/05 07:45:05 - mmengine - INFO - Epoch(train) [4][3000/4111] lr: 7.2488e-05 eta: 5:19:56 time: 0.8058 data_time: 0.6665 memory: 6318 loss: 0.2336 +2023/06/05 07:46:20 - mmengine - INFO - Epoch(train) [4][3100/4111] lr: 7.2171e-05 eta: 5:18:42 time: 0.7802 data_time: 0.6399 memory: 6318 loss: 0.2335 +2023/06/05 07:47:36 - mmengine - INFO - Epoch(train) [4][3200/4111] lr: 7.1852e-05 eta: 5:17:29 time: 0.7682 data_time: 0.6243 memory: 6318 loss: 0.2131 +2023/06/05 07:48:49 - mmengine - INFO - Epoch(train) [4][3300/4111] lr: 7.1533e-05 eta: 5:16:13 time: 0.7452 data_time: 0.6001 memory: 6318 loss: 0.2271 +2023/06/05 07:50:06 - mmengine - INFO - Epoch(train) [4][3400/4111] lr: 7.1213e-05 eta: 5:15:02 time: 0.7746 data_time: 0.6265 memory: 6318 loss: 0.2712 +2023/06/05 07:51:22 - mmengine - INFO - Epoch(train) [4][3500/4111] lr: 7.0891e-05 eta: 5:13:51 time: 0.7252 data_time: 0.5846 memory: 6318 loss: 0.2259 +2023/06/05 07:52:36 - mmengine - INFO - Epoch(train) [4][3600/4111] lr: 7.0569e-05 eta: 5:12:36 time: 0.7984 data_time: 0.6570 memory: 6318 loss: 0.2314 +2023/06/05 07:53:29 - mmengine - INFO - Exp name: resnet50_2xb256_all_1m_lr1e-4_aug_5e-1_20230605_043015 +2023/06/05 07:53:52 - mmengine - INFO - Epoch(train) [4][3700/4111] lr: 7.0246e-05 eta: 5:11:23 time: 0.7741 data_time: 0.6339 memory: 6318 loss: 0.2543 +2023/06/05 07:55:08 - mmengine - INFO - Epoch(train) [4][3800/4111] lr: 6.9922e-05 eta: 5:10:11 time: 0.7275 data_time: 0.5854 memory: 6318 loss: 0.2378 +2023/06/05 07:56:24 - mmengine - INFO - Epoch(train) [4][3900/4111] lr: 6.9597e-05 eta: 5:08:59 time: 0.7083 data_time: 0.5680 memory: 6318 loss: 0.2365 +2023/06/05 07:57:40 - mmengine - INFO - Epoch(train) [4][4000/4111] lr: 6.9272e-05 eta: 5:07:47 time: 0.7952 data_time: 0.6556 memory: 6318 loss: 0.2481 +2023/06/05 07:58:57 - mmengine - INFO - Epoch(train) [4][4100/4111] lr: 6.8945e-05 eta: 5:06:36 time: 0.7165 data_time: 0.5760 memory: 6318 loss: 0.2459 +2023/06/05 07:59:04 - mmengine - INFO - Exp name: resnet50_2xb256_all_1m_lr1e-4_aug_5e-1_20230605_043015 +2023/06/05 07:59:04 - mmengine - INFO - Saving checkpoint at 4 epochs +2023/06/05 07:59:50 - mmengine - INFO - Epoch(val) [4][100/119] eta: 0:00:07 time: 0.7620 data_time: 0.6713 memory: 6318 +2023/06/05 08:00:18 - mmengine - INFO - Epoch(val) [4][119/119] accuracy/top1: 88.2131 data_time: 0.3677 time: 0.4580 +2023/06/05 08:01:46 - mmengine - INFO - Epoch(train) [5][ 100/4111] lr: 6.8582e-05 eta: 5:05:31 time: 0.8502 data_time: 0.6343 memory: 6318 loss: 0.2429 +2023/06/05 08:03:07 - mmengine - INFO - Epoch(train) [5][ 200/4111] lr: 6.8253e-05 eta: 5:04:27 time: 0.7924 data_time: 0.5777 memory: 6318 loss: 0.2437 +2023/06/05 08:04:28 - mmengine - INFO - Epoch(train) [5][ 300/4111] lr: 6.7924e-05 eta: 5:03:20 time: 0.7397 data_time: 0.3348 memory: 6318 loss: 0.2249 +2023/06/05 08:05:50 - mmengine - INFO - Epoch(train) [5][ 400/4111] lr: 6.7595e-05 eta: 5:02:16 time: 0.7925 data_time: 0.2291 memory: 6318 loss: 0.2554 +2023/06/05 08:07:07 - mmengine - INFO - Epoch(train) [5][ 500/4111] lr: 6.7264e-05 eta: 5:01:05 time: 0.7338 data_time: 0.3372 memory: 6318 loss: 0.2268 +2023/06/05 08:07:48 - mmengine - INFO - Exp name: resnet50_2xb256_all_1m_lr1e-4_aug_5e-1_20230605_043015 +2023/06/05 08:08:24 - mmengine - INFO - Epoch(train) [5][ 600/4111] lr: 6.6933e-05 eta: 4:59:53 time: 0.7636 data_time: 0.4749 memory: 6318 loss: 0.2321 +2023/06/05 08:09:39 - mmengine - INFO - Epoch(train) [5][ 700/4111] lr: 6.6601e-05 eta: 4:58:39 time: 0.7585 data_time: 0.3626 memory: 6318 loss: 0.2265 +2023/06/05 08:10:55 - mmengine - INFO - Epoch(train) [5][ 800/4111] lr: 6.6268e-05 eta: 4:57:25 time: 0.7528 data_time: 0.4017 memory: 6318 loss: 0.2389 +2023/06/05 08:12:11 - mmengine - INFO - Epoch(train) [5][ 900/4111] lr: 6.5935e-05 eta: 4:56:13 time: 0.7246 data_time: 0.2533 memory: 6318 loss: 0.2270 +2023/06/05 08:13:28 - mmengine - INFO - Epoch(train) [5][1000/4111] lr: 6.5601e-05 eta: 4:55:00 time: 0.7384 data_time: 0.4445 memory: 6318 loss: 0.2537 +2023/06/05 08:14:45 - mmengine - INFO - Epoch(train) [5][1100/4111] lr: 6.5267e-05 eta: 4:53:49 time: 0.7676 data_time: 0.4799 memory: 6318 loss: 0.2218 +2023/06/05 08:16:04 - mmengine - INFO - Epoch(train) [5][1200/4111] lr: 6.4932e-05 eta: 4:52:39 time: 0.7836 data_time: 0.5886 memory: 6318 loss: 0.2378 +2023/06/05 08:17:21 - mmengine - INFO - Epoch(train) [5][1300/4111] lr: 6.4596e-05 eta: 4:51:28 time: 0.8033 data_time: 0.6632 memory: 6318 loss: 0.2267 +2023/06/05 08:18:36 - mmengine - INFO - Epoch(train) [5][1400/4111] lr: 6.4260e-05 eta: 4:50:13 time: 0.7110 data_time: 0.5715 memory: 6318 loss: 0.2149 +2023/06/05 08:19:52 - mmengine - INFO - Epoch(train) [5][1500/4111] lr: 6.3923e-05 eta: 4:48:59 time: 0.7345 data_time: 0.5947 memory: 6318 loss: 0.2365 +2023/06/05 08:20:31 - mmengine - INFO - Exp name: resnet50_2xb256_all_1m_lr1e-4_aug_5e-1_20230605_043015 +2023/06/05 08:21:09 - mmengine - INFO - Epoch(train) [5][1600/4111] lr: 6.3585e-05 eta: 4:47:47 time: 0.6995 data_time: 0.5587 memory: 6318 loss: 0.2586 +2023/06/05 08:22:26 - mmengine - INFO - Epoch(train) [5][1700/4111] lr: 6.3248e-05 eta: 4:46:35 time: 0.7580 data_time: 0.6117 memory: 6318 loss: 0.2160 +2023/06/05 08:23:46 - mmengine - INFO - Epoch(train) [5][1800/4111] lr: 6.2909e-05 eta: 4:45:27 time: 0.9177 data_time: 0.7779 memory: 6318 loss: 0.2267 +2023/06/05 08:25:03 - mmengine - INFO - Epoch(train) [5][1900/4111] lr: 6.2571e-05 eta: 4:44:15 time: 0.7560 data_time: 0.6114 memory: 6318 loss: 0.2213 +2023/06/05 08:26:20 - mmengine - INFO - Epoch(train) [5][2000/4111] lr: 6.2231e-05 eta: 4:43:02 time: 0.7217 data_time: 0.5817 memory: 6318 loss: 0.2266 +2023/06/05 08:27:37 - mmengine - INFO - Epoch(train) [5][2100/4111] lr: 6.1892e-05 eta: 4:41:50 time: 0.7467 data_time: 0.6068 memory: 6318 loss: 0.2212 +2023/06/05 08:28:53 - mmengine - INFO - Epoch(train) [5][2200/4111] lr: 6.1552e-05 eta: 4:40:36 time: 0.7319 data_time: 0.5907 memory: 6318 loss: 0.2312 +2023/06/05 08:30:10 - mmengine - INFO - Epoch(train) [5][2300/4111] lr: 6.1211e-05 eta: 4:39:24 time: 0.8046 data_time: 0.6644 memory: 6318 loss: 0.2282 +2023/06/05 08:31:28 - mmengine - INFO - Epoch(train) [5][2400/4111] lr: 6.0871e-05 eta: 4:38:12 time: 0.7199 data_time: 0.5794 memory: 6318 loss: 0.2259 +2023/06/05 08:32:49 - mmengine - INFO - Epoch(train) [5][2500/4111] lr: 6.0529e-05 eta: 4:37:04 time: 0.7875 data_time: 0.6479 memory: 6318 loss: 0.2182 +2023/06/05 08:33:28 - mmengine - INFO - Exp name: resnet50_2xb256_all_1m_lr1e-4_aug_5e-1_20230605_043015 +2023/06/05 08:34:06 - mmengine - INFO - Epoch(train) [5][2600/4111] lr: 6.0188e-05 eta: 4:35:51 time: 0.7740 data_time: 0.6332 memory: 6318 loss: 0.2096 +2023/06/05 08:35:22 - mmengine - INFO - Epoch(train) [5][2700/4111] lr: 5.9846e-05 eta: 4:34:38 time: 0.7155 data_time: 0.5757 memory: 6318 loss: 0.2194 +2023/06/05 08:36:39 - mmengine - INFO - Epoch(train) [5][2800/4111] lr: 5.9504e-05 eta: 4:33:25 time: 0.9219 data_time: 0.7819 memory: 6318 loss: 0.2381 +2023/06/05 08:37:55 - mmengine - INFO - Epoch(train) [5][2900/4111] lr: 5.9162e-05 eta: 4:32:11 time: 0.7291 data_time: 0.5880 memory: 6318 loss: 0.2236 +2023/06/05 08:39:13 - mmengine - INFO - Epoch(train) [5][3000/4111] lr: 5.8819e-05 eta: 4:31:00 time: 0.8175 data_time: 0.6765 memory: 6318 loss: 0.2237 +2023/06/05 08:40:29 - mmengine - INFO - Epoch(train) [5][3100/4111] lr: 5.8477e-05 eta: 4:29:46 time: 0.7782 data_time: 0.6368 memory: 6318 loss: 0.2358 +2023/06/05 08:41:46 - mmengine - INFO - Epoch(train) [5][3200/4111] lr: 5.8134e-05 eta: 4:28:32 time: 0.7205 data_time: 0.5800 memory: 6318 loss: 0.2395 +2023/06/05 08:43:02 - mmengine - INFO - Epoch(train) [5][3300/4111] lr: 5.7791e-05 eta: 4:27:19 time: 0.8234 data_time: 0.6822 memory: 6318 loss: 0.2373 +2023/06/05 08:44:19 - mmengine - INFO - Epoch(train) [5][3400/4111] lr: 5.7447e-05 eta: 4:26:05 time: 0.7395 data_time: 0.5984 memory: 6318 loss: 0.2118 +2023/06/05 08:45:38 - mmengine - INFO - Epoch(train) [5][3500/4111] lr: 5.7104e-05 eta: 4:24:55 time: 0.7593 data_time: 0.6185 memory: 6318 loss: 0.2370 +2023/06/05 08:46:18 - mmengine - INFO - Exp name: resnet50_2xb256_all_1m_lr1e-4_aug_5e-1_20230605_043015 +2023/06/05 08:46:57 - mmengine - INFO - Epoch(train) [5][3600/4111] lr: 5.6760e-05 eta: 4:23:43 time: 0.7787 data_time: 0.6385 memory: 6318 loss: 0.2290 +2023/06/05 08:48:13 - mmengine - INFO - Epoch(train) [5][3700/4111] lr: 5.6417e-05 eta: 4:22:30 time: 0.7277 data_time: 0.5880 memory: 6318 loss: 0.2421 +2023/06/05 08:49:30 - mmengine - INFO - Epoch(train) [5][3800/4111] lr: 5.6073e-05 eta: 4:21:16 time: 0.7863 data_time: 0.6431 memory: 6318 loss: 0.2238 +2023/06/05 08:50:47 - mmengine - INFO - Epoch(train) [5][3900/4111] lr: 5.5729e-05 eta: 4:20:03 time: 0.7448 data_time: 0.6051 memory: 6318 loss: 0.2060 +2023/06/05 08:52:03 - mmengine - INFO - Epoch(train) [5][4000/4111] lr: 5.5385e-05 eta: 4:18:49 time: 0.7857 data_time: 0.6456 memory: 6318 loss: 0.2495 +2023/06/05 08:53:20 - mmengine - INFO - Epoch(train) [5][4100/4111] lr: 5.5041e-05 eta: 4:17:35 time: 0.7250 data_time: 0.5834 memory: 6318 loss: 0.2377 +2023/06/05 08:53:27 - mmengine - INFO - Exp name: resnet50_2xb256_all_1m_lr1e-4_aug_5e-1_20230605_043015 +2023/06/05 08:53:27 - mmengine - INFO - Saving checkpoint at 5 epochs +2023/06/05 08:54:15 - mmengine - INFO - Epoch(val) [5][100/119] eta: 0:00:07 time: 0.7754 data_time: 0.6851 memory: 6318 +2023/06/05 08:54:42 - mmengine - INFO - Epoch(val) [5][119/119] accuracy/top1: 89.2740 data_time: 0.3910 time: 0.4800 +2023/06/05 08:56:02 - mmengine - INFO - Epoch(train) [6][ 100/4111] lr: 5.4660e-05 eta: 4:16:15 time: 0.7475 data_time: 0.0166 memory: 6318 loss: 0.2214 +2023/06/05 08:57:18 - mmengine - INFO - Epoch(train) [6][ 200/4111] lr: 5.4316e-05 eta: 4:15:01 time: 0.7511 data_time: 0.0785 memory: 6318 loss: 0.2265 +2023/06/05 08:58:33 - mmengine - INFO - Epoch(train) [6][ 300/4111] lr: 5.3972e-05 eta: 4:13:45 time: 0.7852 data_time: 0.1737 memory: 6318 loss: 0.2242 +2023/06/05 08:59:49 - mmengine - INFO - Epoch(train) [6][ 400/4111] lr: 5.3628e-05 eta: 4:12:31 time: 0.7674 data_time: 0.2014 memory: 6318 loss: 0.2276 +2023/06/05 09:00:27 - mmengine - INFO - Exp name: resnet50_2xb256_all_1m_lr1e-4_aug_5e-1_20230605_043015 +2023/06/05 09:01:06 - mmengine - INFO - Epoch(train) [6][ 500/4111] lr: 5.3284e-05 eta: 4:11:17 time: 0.6924 data_time: 0.3652 memory: 6318 loss: 0.2286 +2023/06/05 09:02:23 - mmengine - INFO - Epoch(train) [6][ 600/4111] lr: 5.2941e-05 eta: 4:10:04 time: 0.7842 data_time: 0.6162 memory: 6318 loss: 0.2208 +2023/06/05 09:03:41 - mmengine - INFO - Epoch(train) [6][ 700/4111] lr: 5.2597e-05 eta: 4:08:51 time: 0.7421 data_time: 0.6010 memory: 6318 loss: 0.2224 +2023/06/05 09:04:57 - mmengine - INFO - Epoch(train) [6][ 800/4111] lr: 5.2254e-05 eta: 4:07:37 time: 0.7267 data_time: 0.5866 memory: 6318 loss: 0.2223 +2023/06/05 09:06:15 - mmengine - INFO - Epoch(train) [6][ 900/4111] lr: 5.1911e-05 eta: 4:06:24 time: 0.7500 data_time: 0.6092 memory: 6318 loss: 0.2368 +2023/06/05 09:07:42 - mmengine - INFO - Epoch(train) [6][1000/4111] lr: 5.1568e-05 eta: 4:05:20 time: 0.8278 data_time: 0.6736 memory: 6318 loss: 0.2162 +2023/06/05 09:09:07 - mmengine - INFO - Epoch(train) [6][1100/4111] lr: 5.1225e-05 eta: 4:04:13 time: 1.0904 data_time: 0.9498 memory: 6318 loss: 0.2426 +2023/06/05 09:10:26 - mmengine - INFO - Epoch(train) [6][1200/4111] lr: 5.0883e-05 eta: 4:03:01 time: 0.8106 data_time: 0.6630 memory: 6318 loss: 0.2254 +2023/06/05 09:11:49 - mmengine - INFO - Epoch(train) [6][1300/4111] lr: 5.0540e-05 eta: 4:01:53 time: 0.7666 data_time: 0.6265 memory: 6318 loss: 0.2206 +2023/06/05 09:13:09 - mmengine - INFO - Epoch(train) [6][1400/4111] lr: 5.0198e-05 eta: 4:00:41 time: 0.7937 data_time: 0.6423 memory: 6318 loss: 0.2258 +2023/06/05 09:13:47 - mmengine - INFO - Exp name: resnet50_2xb256_all_1m_lr1e-4_aug_5e-1_20230605_043015 +2023/06/05 09:14:26 - mmengine - INFO - Epoch(train) [6][1500/4111] lr: 4.9856e-05 eta: 3:59:27 time: 0.7680 data_time: 0.6275 memory: 6318 loss: 0.2214 +2023/06/05 09:15:45 - mmengine - INFO - Epoch(train) [6][1600/4111] lr: 4.9515e-05 eta: 3:58:15 time: 0.7070 data_time: 0.5662 memory: 6318 loss: 0.2415 +2023/06/05 09:17:03 - mmengine - INFO - Epoch(train) [6][1700/4111] lr: 4.9174e-05 eta: 3:57:01 time: 0.7056 data_time: 0.5645 memory: 6318 loss: 0.2150 +2023/06/05 09:18:20 - mmengine - INFO - Epoch(train) [6][1800/4111] lr: 4.8833e-05 eta: 3:55:48 time: 0.7788 data_time: 0.6381 memory: 6318 loss: 0.2316 +2023/06/05 09:19:37 - mmengine - INFO - Epoch(train) [6][1900/4111] lr: 4.8492e-05 eta: 3:54:33 time: 0.8003 data_time: 0.6599 memory: 6318 loss: 0.2283 +2023/06/05 09:20:54 - mmengine - INFO - Epoch(train) [6][2000/4111] lr: 4.8152e-05 eta: 3:53:19 time: 0.7127 data_time: 0.5728 memory: 6318 loss: 0.2311 +2023/06/05 09:22:12 - mmengine - INFO - Epoch(train) [6][2100/4111] lr: 4.7813e-05 eta: 3:52:06 time: 0.7974 data_time: 0.6469 memory: 6318 loss: 0.2061 +2023/06/05 09:23:41 - mmengine - INFO - Epoch(train) [6][2200/4111] lr: 4.7473e-05 eta: 3:51:01 time: 0.7579 data_time: 0.6173 memory: 6318 loss: 0.2191 +2023/06/05 09:24:56 - mmengine - INFO - Epoch(train) [6][2300/4111] lr: 4.7135e-05 eta: 3:49:46 time: 0.7390 data_time: 0.5941 memory: 6318 loss: 0.2172 +2023/06/05 09:26:13 - mmengine - INFO - Epoch(train) [6][2400/4111] lr: 4.6796e-05 eta: 3:48:31 time: 0.7471 data_time: 0.6067 memory: 6318 loss: 0.2189 +2023/06/05 09:26:52 - mmengine - INFO - Exp name: resnet50_2xb256_all_1m_lr1e-4_aug_5e-1_20230605_043015 +2023/06/05 09:27:31 - mmengine - INFO - Epoch(train) [6][2500/4111] lr: 4.6458e-05 eta: 3:47:18 time: 0.7909 data_time: 0.6506 memory: 6318 loss: 0.2199 +2023/06/05 09:28:55 - mmengine - INFO - Epoch(train) [6][2600/4111] lr: 4.6121e-05 eta: 3:46:09 time: 0.7845 data_time: 0.6432 memory: 6318 loss: 0.2149 +2023/06/05 09:30:11 - mmengine - INFO - Epoch(train) [6][2700/4111] lr: 4.5784e-05 eta: 3:44:53 time: 0.7774 data_time: 0.6262 memory: 6318 loss: 0.2251 +2023/06/05 09:31:28 - mmengine - INFO - Epoch(train) [6][2800/4111] lr: 4.5448e-05 eta: 3:43:38 time: 0.7820 data_time: 0.6411 memory: 6318 loss: 0.2306 +2023/06/05 09:32:43 - mmengine - INFO - Epoch(train) [6][2900/4111] lr: 4.5112e-05 eta: 3:42:23 time: 0.7187 data_time: 0.5785 memory: 6318 loss: 0.2257 +2023/06/05 09:33:58 - mmengine - INFO - Epoch(train) [6][3000/4111] lr: 4.4777e-05 eta: 3:41:06 time: 0.7169 data_time: 0.5760 memory: 6318 loss: 0.2216 +2023/06/05 09:35:12 - mmengine - INFO - Epoch(train) [6][3100/4111] lr: 4.4442e-05 eta: 3:39:50 time: 0.7510 data_time: 0.6024 memory: 6318 loss: 0.2115 +2023/06/05 09:36:28 - mmengine - INFO - Epoch(train) [6][3200/4111] lr: 4.4108e-05 eta: 3:38:34 time: 0.7712 data_time: 0.6314 memory: 6318 loss: 0.2095 +2023/06/05 09:37:44 - mmengine - INFO - Epoch(train) [6][3300/4111] lr: 4.3775e-05 eta: 3:37:19 time: 0.7868 data_time: 0.6470 memory: 6318 loss: 0.2167 +2023/06/05 09:39:01 - mmengine - INFO - Epoch(train) [6][3400/4111] lr: 4.3442e-05 eta: 3:36:04 time: 0.7350 data_time: 0.5941 memory: 6318 loss: 0.2367 +2023/06/05 09:39:36 - mmengine - INFO - Exp name: resnet50_2xb256_all_1m_lr1e-4_aug_5e-1_20230605_043015 +2023/06/05 09:40:15 - mmengine - INFO - Epoch(train) [6][3500/4111] lr: 4.3110e-05 eta: 3:34:48 time: 0.7931 data_time: 0.6539 memory: 6318 loss: 0.2163 +2023/06/05 09:41:31 - mmengine - INFO - Epoch(train) [6][3600/4111] lr: 4.2779e-05 eta: 3:33:33 time: 0.7742 data_time: 0.6331 memory: 6318 loss: 0.2247 +2023/06/05 09:42:48 - mmengine - INFO - Epoch(train) [6][3700/4111] lr: 4.2448e-05 eta: 3:32:18 time: 0.7315 data_time: 0.5891 memory: 6318 loss: 0.2151 +2023/06/05 09:44:05 - mmengine - INFO - Epoch(train) [6][3800/4111] lr: 4.2119e-05 eta: 3:31:04 time: 0.8074 data_time: 0.6673 memory: 6318 loss: 0.2139 +2023/06/05 09:45:21 - mmengine - INFO - Epoch(train) [6][3900/4111] lr: 4.1789e-05 eta: 3:29:48 time: 0.7303 data_time: 0.5842 memory: 6318 loss: 0.2248 +2023/06/05 09:46:36 - mmengine - INFO - Epoch(train) [6][4000/4111] lr: 4.1461e-05 eta: 3:28:32 time: 0.7560 data_time: 0.6124 memory: 6318 loss: 0.2094 +2023/06/05 09:47:52 - mmengine - INFO - Epoch(train) [6][4100/4111] lr: 4.1133e-05 eta: 3:27:17 time: 0.7096 data_time: 0.5682 memory: 6318 loss: 0.2220 +2023/06/05 09:47:59 - mmengine - INFO - Exp name: resnet50_2xb256_all_1m_lr1e-4_aug_5e-1_20230605_043015 +2023/06/05 09:47:59 - mmengine - INFO - Saving checkpoint at 6 epochs +2023/06/05 09:48:47 - mmengine - INFO - Epoch(val) [6][100/119] eta: 0:00:07 time: 0.6989 data_time: 0.6093 memory: 6318 +2023/06/05 09:49:13 - mmengine - INFO - Epoch(val) [6][119/119] accuracy/top1: 91.0647 data_time: 0.3798 time: 0.4705 +2023/06/05 09:50:31 - mmengine - INFO - Epoch(train) [7][ 100/4111] lr: 4.0771e-05 eta: 3:25:54 time: 0.7361 data_time: 0.3209 memory: 6318 loss: 0.2120 +2023/06/05 09:51:47 - mmengine - INFO - Epoch(train) [7][ 200/4111] lr: 4.0445e-05 eta: 3:24:39 time: 0.7527 data_time: 0.2320 memory: 6318 loss: 0.2310 +2023/06/05 09:53:02 - mmengine - INFO - Epoch(train) [7][ 300/4111] lr: 4.0120e-05 eta: 3:23:23 time: 0.7051 data_time: 0.1945 memory: 6318 loss: 0.2183 +2023/06/05 09:53:32 - mmengine - INFO - Exp name: resnet50_2xb256_all_1m_lr1e-4_aug_5e-1_20230605_043015 +2023/06/05 09:54:18 - mmengine - INFO - Epoch(train) [7][ 400/4111] lr: 3.9796e-05 eta: 3:22:07 time: 0.7813 data_time: 0.4923 memory: 6318 loss: 0.2312 +2023/06/05 09:55:31 - mmengine - INFO - Epoch(train) [7][ 500/4111] lr: 3.9473e-05 eta: 3:20:50 time: 0.7242 data_time: 0.5172 memory: 6318 loss: 0.2258 +2023/06/05 09:56:44 - mmengine - INFO - Epoch(train) [7][ 600/4111] lr: 3.9150e-05 eta: 3:19:33 time: 0.7151 data_time: 0.5714 memory: 6318 loss: 0.2375 +2023/06/05 09:57:58 - mmengine - INFO - Epoch(train) [7][ 700/4111] lr: 3.8829e-05 eta: 3:18:16 time: 0.7413 data_time: 0.5933 memory: 6318 loss: 0.2357 +2023/06/05 09:59:12 - mmengine - INFO - Epoch(train) [7][ 800/4111] lr: 3.8509e-05 eta: 3:17:00 time: 0.7514 data_time: 0.6053 memory: 6318 loss: 0.2198 +2023/06/05 10:00:27 - mmengine - INFO - Epoch(train) [7][ 900/4111] lr: 3.8189e-05 eta: 3:15:44 time: 0.7382 data_time: 0.5937 memory: 6318 loss: 0.2058 +2023/06/05 10:01:44 - mmengine - INFO - Epoch(train) [7][1000/4111] lr: 3.7871e-05 eta: 3:14:29 time: 0.7230 data_time: 0.5775 memory: 6318 loss: 0.2120 +2023/06/05 10:03:00 - mmengine - INFO - Epoch(train) [7][1100/4111] lr: 3.7553e-05 eta: 3:13:14 time: 0.7402 data_time: 0.5991 memory: 6318 loss: 0.2268 +2023/06/05 10:04:16 - mmengine - INFO - Epoch(train) [7][1200/4111] lr: 3.7237e-05 eta: 3:11:58 time: 0.7691 data_time: 0.5241 memory: 6318 loss: 0.2126 +2023/06/05 10:05:31 - mmengine - INFO - Epoch(train) [7][1300/4111] lr: 3.6921e-05 eta: 3:10:43 time: 0.7883 data_time: 0.3424 memory: 6318 loss: 0.2062 +2023/06/05 10:06:00 - mmengine - INFO - Exp name: resnet50_2xb256_all_1m_lr1e-4_aug_5e-1_20230605_043015 +2023/06/05 10:06:48 - mmengine - INFO - Epoch(train) [7][1400/4111] lr: 3.6607e-05 eta: 3:09:28 time: 0.7385 data_time: 0.3458 memory: 6318 loss: 0.2030 +2023/06/05 10:08:05 - mmengine - INFO - Epoch(train) [7][1500/4111] lr: 3.6294e-05 eta: 3:08:13 time: 0.8354 data_time: 0.5796 memory: 6318 loss: 0.2325 +2023/06/05 10:09:23 - mmengine - INFO - Epoch(train) [7][1600/4111] lr: 3.5981e-05 eta: 3:06:59 time: 0.7662 data_time: 0.6268 memory: 6318 loss: 0.2258 +2023/06/05 10:10:37 - mmengine - INFO - Epoch(train) [7][1700/4111] lr: 3.5670e-05 eta: 3:05:42 time: 0.7412 data_time: 0.6018 memory: 6318 loss: 0.2239 +2023/06/05 10:11:52 - mmengine - INFO - Epoch(train) [7][1800/4111] lr: 3.5360e-05 eta: 3:04:27 time: 0.7617 data_time: 0.6216 memory: 6318 loss: 0.2138 +2023/06/05 10:13:16 - mmengine - INFO - Epoch(train) [7][1900/4111] lr: 3.5051e-05 eta: 3:03:15 time: 0.7515 data_time: 0.5958 memory: 6318 loss: 0.2314 +2023/06/05 10:14:39 - mmengine - INFO - Epoch(train) [7][2000/4111] lr: 3.4744e-05 eta: 3:02:04 time: 0.7904 data_time: 0.6483 memory: 6318 loss: 0.2143 +2023/06/05 10:15:59 - mmengine - INFO - Epoch(train) [7][2100/4111] lr: 3.4437e-05 eta: 3:00:51 time: 0.7967 data_time: 0.6556 memory: 6318 loss: 0.2142 +2023/06/05 10:17:18 - mmengine - INFO - Epoch(train) [7][2200/4111] lr: 3.4132e-05 eta: 2:59:37 time: 1.0781 data_time: 0.9353 memory: 6318 loss: 0.2335 +2023/06/05 10:18:36 - mmengine - INFO - Epoch(train) [7][2300/4111] lr: 3.3828e-05 eta: 2:58:22 time: 0.7348 data_time: 0.5899 memory: 6318 loss: 0.2184 +2023/06/05 10:18:58 - mmengine - INFO - Exp name: resnet50_2xb256_all_1m_lr1e-4_aug_5e-1_20230605_043015 +2023/06/05 10:19:51 - mmengine - INFO - Epoch(train) [7][2400/4111] lr: 3.3525e-05 eta: 2:57:06 time: 0.7644 data_time: 0.6249 memory: 6318 loss: 0.2104 +2023/06/05 10:21:07 - mmengine - INFO - Epoch(train) [7][2500/4111] lr: 3.3224e-05 eta: 2:55:51 time: 0.7665 data_time: 0.6146 memory: 6318 loss: 0.2241 +2023/06/05 10:22:24 - mmengine - INFO - Epoch(train) [7][2600/4111] lr: 3.2923e-05 eta: 2:54:36 time: 0.7305 data_time: 0.5899 memory: 6318 loss: 0.2180 +2023/06/05 10:23:42 - mmengine - INFO - Epoch(train) [7][2700/4111] lr: 3.2624e-05 eta: 2:53:21 time: 0.8325 data_time: 0.6916 memory: 6318 loss: 0.2028 +2023/06/05 10:24:59 - mmengine - INFO - Epoch(train) [7][2800/4111] lr: 3.2326e-05 eta: 2:52:06 time: 0.7610 data_time: 0.6181 memory: 6318 loss: 0.2209 +2023/06/05 10:26:18 - mmengine - INFO - Epoch(train) [7][2900/4111] lr: 3.2030e-05 eta: 2:50:52 time: 0.7727 data_time: 0.6322 memory: 6318 loss: 0.2266 +2023/06/05 10:27:37 - mmengine - INFO - Epoch(train) [7][3000/4111] lr: 3.1735e-05 eta: 2:49:38 time: 0.8124 data_time: 0.6718 memory: 6318 loss: 0.2239 +2023/06/05 10:28:54 - mmengine - INFO - Epoch(train) [7][3100/4111] lr: 3.1441e-05 eta: 2:48:23 time: 0.7880 data_time: 0.6478 memory: 6318 loss: 0.2045 +2023/06/05 10:30:15 - mmengine - INFO - Epoch(train) [7][3200/4111] lr: 3.1149e-05 eta: 2:47:10 time: 0.8066 data_time: 0.6670 memory: 6318 loss: 0.2192 +2023/06/05 10:31:33 - mmengine - INFO - Epoch(train) [7][3300/4111] lr: 3.0858e-05 eta: 2:45:55 time: 0.7689 data_time: 0.6286 memory: 6318 loss: 0.2159 +2023/06/05 10:31:58 - mmengine - INFO - Exp name: resnet50_2xb256_all_1m_lr1e-4_aug_5e-1_20230605_043015 +2023/06/05 10:32:51 - mmengine - INFO - Epoch(train) [7][3400/4111] lr: 3.0569e-05 eta: 2:44:41 time: 0.7191 data_time: 0.5792 memory: 6318 loss: 0.2176 +2023/06/05 10:34:08 - mmengine - INFO - Epoch(train) [7][3500/4111] lr: 3.0281e-05 eta: 2:43:25 time: 0.7587 data_time: 0.6122 memory: 6318 loss: 0.2165 +2023/06/05 10:35:25 - mmengine - INFO - Epoch(train) [7][3600/4111] lr: 2.9994e-05 eta: 2:42:10 time: 0.7239 data_time: 0.5830 memory: 6318 loss: 0.2313 +2023/06/05 10:36:53 - mmengine - INFO - Epoch(train) [7][3700/4111] lr: 2.9709e-05 eta: 2:41:00 time: 0.7592 data_time: 0.6110 memory: 6318 loss: 0.2153 +2023/06/05 10:38:10 - mmengine - INFO - Epoch(train) [7][3800/4111] lr: 2.9425e-05 eta: 2:39:45 time: 0.8211 data_time: 0.6804 memory: 6318 loss: 0.2069 +2023/06/05 10:39:27 - mmengine - INFO - Epoch(train) [7][3900/4111] lr: 2.9143e-05 eta: 2:38:29 time: 0.7671 data_time: 0.6137 memory: 6318 loss: 0.2201 +2023/06/05 10:40:44 - mmengine - INFO - Epoch(train) [7][4000/4111] lr: 2.8862e-05 eta: 2:37:14 time: 0.7612 data_time: 0.6210 memory: 6318 loss: 0.2262 +2023/06/05 10:43:30 - mmengine - INFO - Epoch(train) [7][4100/4111] lr: 2.8583e-05 eta: 2:36:37 time: 0.6743 data_time: 0.5121 memory: 6318 loss: 0.2341 +2023/06/05 10:43:36 - mmengine - INFO - Exp name: resnet50_2xb256_all_1m_lr1e-4_aug_5e-1_20230605_043015 +2023/06/05 10:43:36 - mmengine - INFO - Saving checkpoint at 7 epochs +2023/06/05 10:44:23 - mmengine - INFO - Epoch(val) [7][100/119] eta: 0:00:07 time: 0.6871 data_time: 0.5898 memory: 6318 +2023/06/05 10:44:50 - mmengine - INFO - Epoch(val) [7][119/119] accuracy/top1: 92.3605 data_time: 0.3819 time: 0.4709 +2023/06/05 10:46:07 - mmengine - INFO - Epoch(train) [8][ 100/4111] lr: 2.8275e-05 eta: 2:35:12 time: 0.7521 data_time: 0.6104 memory: 6318 loss: 0.2263 +2023/06/05 10:47:21 - mmengine - INFO - Epoch(train) [8][ 200/4111] lr: 2.7999e-05 eta: 2:33:55 time: 0.7882 data_time: 0.6448 memory: 6318 loss: 0.2168 +2023/06/05 10:47:36 - mmengine - INFO - Exp name: resnet50_2xb256_all_1m_lr1e-4_aug_5e-1_20230605_043015 +2023/06/05 10:48:35 - mmengine - INFO - Epoch(train) [8][ 300/4111] lr: 2.7725e-05 eta: 2:32:38 time: 0.7282 data_time: 0.5874 memory: 6318 loss: 0.2134 +2023/06/05 10:49:48 - mmengine - INFO - Epoch(train) [8][ 400/4111] lr: 2.7452e-05 eta: 2:31:20 time: 0.7312 data_time: 0.5237 memory: 6318 loss: 0.2160 +2023/06/05 10:51:02 - mmengine - INFO - Epoch(train) [8][ 500/4111] lr: 2.7181e-05 eta: 2:30:03 time: 0.7278 data_time: 0.5871 memory: 6318 loss: 0.2142 +2023/06/05 10:52:18 - mmengine - INFO - Epoch(train) [8][ 600/4111] lr: 2.6911e-05 eta: 2:28:47 time: 0.7979 data_time: 0.6580 memory: 6318 loss: 0.2258 +2023/06/05 10:53:30 - mmengine - INFO - Epoch(train) [8][ 700/4111] lr: 2.6644e-05 eta: 2:27:29 time: 0.6881 data_time: 0.5466 memory: 6318 loss: 0.2087 +2023/06/05 10:54:43 - mmengine - INFO - Epoch(train) [8][ 800/4111] lr: 2.6377e-05 eta: 2:26:12 time: 0.7980 data_time: 0.6518 memory: 6318 loss: 0.2018 +2023/06/05 10:55:57 - mmengine - INFO - Epoch(train) [8][ 900/4111] lr: 2.6113e-05 eta: 2:24:55 time: 0.7836 data_time: 0.6430 memory: 6318 loss: 0.2118 +2023/06/05 10:57:11 - mmengine - INFO - Epoch(train) [8][1000/4111] lr: 2.5850e-05 eta: 2:23:38 time: 0.7499 data_time: 0.5988 memory: 6318 loss: 0.2003 +2023/06/05 10:58:24 - mmengine - INFO - Epoch(train) [8][1100/4111] lr: 2.5589e-05 eta: 2:22:21 time: 0.7286 data_time: 0.5884 memory: 6318 loss: 0.1964 +2023/06/05 10:59:39 - mmengine - INFO - Epoch(train) [8][1200/4111] lr: 2.5330e-05 eta: 2:21:05 time: 0.8365 data_time: 0.6950 memory: 6318 loss: 0.2181 +2023/06/05 10:59:54 - mmengine - INFO - Exp name: resnet50_2xb256_all_1m_lr1e-4_aug_5e-1_20230605_043015 +2023/06/05 11:00:53 - mmengine - INFO - Epoch(train) [8][1300/4111] lr: 2.5072e-05 eta: 2:19:48 time: 0.7476 data_time: 0.6070 memory: 6318 loss: 0.2108 +2023/06/05 11:02:09 - mmengine - INFO - Epoch(train) [8][1400/4111] lr: 2.4816e-05 eta: 2:18:32 time: 0.8261 data_time: 0.6724 memory: 6318 loss: 0.2193 +2023/06/05 11:03:27 - mmengine - INFO - Epoch(train) [8][1500/4111] lr: 2.4562e-05 eta: 2:17:17 time: 0.8157 data_time: 0.6751 memory: 6318 loss: 0.2210 +2023/06/05 11:04:50 - mmengine - INFO - Epoch(train) [8][1600/4111] lr: 2.4309e-05 eta: 2:16:03 time: 0.7475 data_time: 0.6079 memory: 6318 loss: 0.2409 +2023/06/05 11:06:07 - mmengine - INFO - Epoch(train) [8][1700/4111] lr: 2.4059e-05 eta: 2:14:47 time: 0.7203 data_time: 0.4834 memory: 6318 loss: 0.1997 +2023/06/05 11:07:26 - mmengine - INFO - Epoch(train) [8][1800/4111] lr: 2.3810e-05 eta: 2:13:32 time: 0.7506 data_time: 0.2910 memory: 6318 loss: 0.2177 +2023/06/05 11:08:45 - mmengine - INFO - Epoch(train) [8][1900/4111] lr: 2.3563e-05 eta: 2:12:17 time: 0.7565 data_time: 0.3451 memory: 6318 loss: 0.2236 +2023/06/05 11:10:05 - mmengine - INFO - Epoch(train) [8][2000/4111] lr: 2.3318e-05 eta: 2:11:03 time: 0.6942 data_time: 0.3461 memory: 6318 loss: 0.2359 +2023/06/05 11:11:22 - mmengine - INFO - Epoch(train) [8][2100/4111] lr: 2.3075e-05 eta: 2:09:47 time: 0.7598 data_time: 0.3456 memory: 6318 loss: 0.2040 +2023/06/05 11:12:39 - mmengine - INFO - Epoch(train) [8][2200/4111] lr: 2.2833e-05 eta: 2:08:31 time: 0.8021 data_time: 0.3676 memory: 6318 loss: 0.2068 +2023/06/05 11:12:55 - mmengine - INFO - Exp name: resnet50_2xb256_all_1m_lr1e-4_aug_5e-1_20230605_043015 +2023/06/05 11:13:55 - mmengine - INFO - Epoch(train) [8][2300/4111] lr: 2.2594e-05 eta: 2:07:15 time: 0.7440 data_time: 0.1274 memory: 6318 loss: 0.2076 +2023/06/05 11:15:12 - mmengine - INFO - Epoch(train) [8][2400/4111] lr: 2.2356e-05 eta: 2:05:59 time: 0.7982 data_time: 0.2360 memory: 6318 loss: 0.2207 +2023/06/05 11:16:28 - mmengine - INFO - Epoch(train) [8][2500/4111] lr: 2.2120e-05 eta: 2:04:43 time: 0.6828 data_time: 0.2441 memory: 6318 loss: 0.2162 +2023/06/05 11:17:41 - mmengine - INFO - Epoch(train) [8][2600/4111] lr: 2.1886e-05 eta: 2:03:26 time: 0.7177 data_time: 0.0572 memory: 6318 loss: 0.2165 +2023/06/05 11:19:06 - mmengine - INFO - Epoch(train) [8][2700/4111] lr: 2.1655e-05 eta: 2:02:12 time: 0.8127 data_time: 0.4509 memory: 6318 loss: 0.2098 +2023/06/05 11:20:25 - mmengine - INFO - Epoch(train) [8][2800/4111] lr: 2.1425e-05 eta: 2:00:57 time: 0.7857 data_time: 0.3818 memory: 6318 loss: 0.2035 +2023/06/05 11:21:45 - mmengine - INFO - Epoch(train) [8][2900/4111] lr: 2.1197e-05 eta: 1:59:42 time: 0.7535 data_time: 0.6132 memory: 6318 loss: 0.2222 +2023/06/05 11:23:05 - mmengine - INFO - Epoch(train) [8][3000/4111] lr: 2.0971e-05 eta: 1:58:27 time: 0.7328 data_time: 0.5931 memory: 6318 loss: 0.2138 +2023/06/05 11:24:26 - mmengine - INFO - Epoch(train) [8][3100/4111] lr: 2.0747e-05 eta: 1:57:12 time: 0.7588 data_time: 0.6189 memory: 6318 loss: 0.2127 +2023/06/05 11:25:43 - mmengine - INFO - Epoch(train) [8][3200/4111] lr: 2.0525e-05 eta: 1:55:57 time: 0.8382 data_time: 0.0010 memory: 6318 loss: 0.2055 +2023/06/05 11:25:59 - mmengine - INFO - Exp name: resnet50_2xb256_all_1m_lr1e-4_aug_5e-1_20230605_043015 +2023/06/05 11:27:06 - mmengine - INFO - Epoch(train) [8][3300/4111] lr: 2.0305e-05 eta: 1:54:42 time: 0.7376 data_time: 0.0010 memory: 6318 loss: 0.2018 +2023/06/05 11:28:24 - mmengine - INFO - Epoch(train) [8][3400/4111] lr: 2.0087e-05 eta: 1:53:26 time: 0.7884 data_time: 0.0011 memory: 6318 loss: 0.2080 diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_043015/vis_data/20230605_043015.json b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_043015/vis_data/20230605_043015.json new file mode 100644 index 0000000000000000000000000000000000000000..81fcbfd8e34ca0334bac8e28d84a7610d675daeb --- /dev/null +++ b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_043015/vis_data/20230605_043015.json @@ -0,0 +1,328 @@ +{"lr": 9.999871217862936e-05, "data_time": 0.5833782434463501, "loss": 0.6457757472991943, "time": 0.7236799240112305, "epoch": 1, "memory": 9436, "step": 100} +{"lr": 9.999479662539605e-05, "data_time": 0.7668869018554687, "loss": 0.6040080726146698, "time": 0.9103182554244995, "epoch": 1, "memory": 6319, "step": 200} +{"lr": 9.99882534375654e-05, "data_time": 0.5824718236923218, "loss": 0.5769660413265228, "time": 0.7236875057220459, "epoch": 1, "memory": 6319, "step": 300} +{"lr": 9.997908299725075e-05, "data_time": 0.5609920263290405, "loss": 0.538285905122757, "time": 0.7017717838287354, "epoch": 1, "memory": 6319, "step": 400} +{"lr": 9.996728583999318e-05, "data_time": 0.5731720924377441, "loss": 0.5249818980693817, "time": 0.7144674062728882, "epoch": 1, "memory": 6319, "step": 500} +{"lr": 9.995286265473058e-05, "data_time": 0.6639786243438721, "loss": 0.5100289076566696, "time": 0.8061728477478027, "epoch": 1, "memory": 6319, "step": 600} +{"lr": 9.993581428375725e-05, "data_time": 0.7929261684417724, "loss": 0.46335644125938413, "time": 0.9325023651123047, "epoch": 1, "memory": 6319, "step": 700} +{"lr": 9.991614172267466e-05, "data_time": 0.582516360282898, "loss": 0.4537742078304291, "time": 0.7229749202728272, "epoch": 1, "memory": 6319, "step": 800} +{"lr": 9.989384612033355e-05, "data_time": 0.6241274118423462, "loss": 0.44828479588031767, "time": 0.7643932342529297, "epoch": 1, "memory": 6319, "step": 900} +{"lr": 9.986892877876658e-05, "data_time": 0.6361027956008911, "loss": 0.4394378662109375, "time": 0.7761248111724853, "epoch": 1, "memory": 6319, "step": 1000} +{"lr": 9.984139115311244e-05, "data_time": 0.626507306098938, "loss": 0.44652384519577026, "time": 0.7666589021682739, "epoch": 1, "memory": 6319, "step": 1100} +{"lr": 9.981123485153081e-05, "data_time": 0.5829660177230835, "loss": 0.41557493805885315, "time": 0.7230542421340942, "epoch": 1, "memory": 6319, "step": 1200} +{"lr": 9.977846163510851e-05, "data_time": 0.6212214946746826, "loss": 0.4050597667694092, "time": 0.7630456447601318, "epoch": 1, "memory": 6319, "step": 1300} +{"lr": 9.974307341775653e-05, "data_time": 0.5404500246047974, "loss": 0.39317560791969297, "time": 0.6810670614242553, "epoch": 1, "memory": 6319, "step": 1400} +{"lr": 9.970507226609833e-05, "data_time": 0.6263991117477417, "loss": 0.37779472172260287, "time": 0.7653239011764527, "epoch": 1, "memory": 6319, "step": 1500} +{"lr": 9.966446039934934e-05, "data_time": 0.5814517498016357, "loss": 0.3829806953668594, "time": 0.7212536334991455, "epoch": 1, "memory": 6319, "step": 1600} +{"lr": 9.962124018918683e-05, "data_time": 0.5648025751113892, "loss": 0.3914806544780731, "time": 0.7056179523468018, "epoch": 1, "memory": 6319, "step": 1700} +{"lr": 9.957541415961212e-05, "data_time": 0.6145240545272828, "loss": 0.3696164727210999, "time": 0.7550813674926757, "epoch": 1, "memory": 6319, "step": 1800} +{"lr": 9.952698498680267e-05, "data_time": 0.5710381269454956, "loss": 0.37971139550209043, "time": 0.7102778673171997, "epoch": 1, "memory": 6319, "step": 1900} +{"lr": 9.947595549895583e-05, "data_time": 0.6139296770095826, "loss": 0.3580131262540817, "time": 0.7543619871139526, "epoch": 1, "memory": 6319, "step": 2000} +{"lr": 9.942232867612417e-05, "data_time": 0.5774040699005127, "loss": 0.3667486310005188, "time": 0.7178636074066163, "epoch": 1, "memory": 6319, "step": 2100} +{"lr": 9.93661076500406e-05, "data_time": 0.5804235696792602, "loss": 0.35057703852653505, "time": 0.7219144105911255, "epoch": 1, "memory": 6319, "step": 2200} +{"lr": 9.930729570393622e-05, "data_time": 0.5540996074676514, "loss": 0.3724305748939514, "time": 0.6951204061508178, "epoch": 1, "memory": 6319, "step": 2300} +{"lr": 9.924589627234824e-05, "data_time": 0.6224174976348877, "loss": 0.35138237476348877, "time": 0.7629217624664306, "epoch": 1, "memory": 6319, "step": 2400} +{"lr": 9.918191294091956e-05, "data_time": 0.5802772521972657, "loss": 0.36304775178432463, "time": 0.7212441444396973, "epoch": 1, "memory": 6319, "step": 2500} +{"lr": 9.91153494461892e-05, "data_time": 0.5826991081237793, "loss": 0.3589605003595352, "time": 0.7224309206008911, "epoch": 1, "memory": 6319, "step": 2600} +{"lr": 9.904620967537422e-05, "data_time": 0.6027488946914673, "loss": 0.33957523107528687, "time": 0.7427972793579102, "epoch": 1, "memory": 6319, "step": 2700} +{"lr": 9.897449766614277e-05, "data_time": 0.6042295217514038, "loss": 0.3580081135034561, "time": 0.7444121360778808, "epoch": 1, "memory": 6319, "step": 2800} +{"lr": 9.890021760637804e-05, "data_time": 0.5304379224777221, "loss": 0.341555318236351, "time": 0.6727274417877197, "epoch": 1, "memory": 6319, "step": 2900} +{"lr": 9.882337383393409e-05, "data_time": 0.5570443153381348, "loss": 0.34094529151916503, "time": 0.6958114862442016, "epoch": 1, "memory": 6319, "step": 3000} +{"lr": 9.874397083638193e-05, "data_time": 0.6410630941390991, "loss": 0.3343875378370285, "time": 0.7816831827163696, "epoch": 1, "memory": 6319, "step": 3100} +{"lr": 9.866201325074828e-05, "data_time": 0.632978367805481, "loss": 0.332791805267334, "time": 0.772688364982605, "epoch": 1, "memory": 6319, "step": 3200} +{"lr": 9.857750586324393e-05, "data_time": 0.6097943782806396, "loss": 0.34595970809459686, "time": 0.7490038394927978, "epoch": 1, "memory": 6319, "step": 3300} +{"lr": 9.84904536089848e-05, "data_time": 0.5731130599975586, "loss": 0.31871800422668456, "time": 0.7145023107528686, "epoch": 1, "memory": 6319, "step": 3400} +{"lr": 9.840086157170351e-05, "data_time": 0.5743430137634278, "loss": 0.33876071572303773, "time": 0.7148260831832886, "epoch": 1, "memory": 6319, "step": 3500} +{"lr": 9.830873498345255e-05, "data_time": 0.5523132085800171, "loss": 0.3251511067152023, "time": 0.691866397857666, "epoch": 1, "memory": 6319, "step": 3600} +{"lr": 9.821407922429864e-05, "data_time": 0.6264314889907837, "loss": 0.332692089676857, "time": 0.7661503314971924, "epoch": 1, "memory": 6319, "step": 3700} +{"lr": 9.811689982200877e-05, "data_time": 0.6108566045761108, "loss": 0.3214162290096283, "time": 0.7660102844238281, "epoch": 1, "memory": 6319, "step": 3800} +{"lr": 9.801720245172702e-05, "data_time": 0.5971901655197144, "loss": 0.31916169822216034, "time": 0.7374595165252685, "epoch": 1, "memory": 6319, "step": 3900} +{"lr": 9.791499293564358e-05, "data_time": 0.6052390336990356, "loss": 0.3148622393608093, "time": 0.7446340322494507, "epoch": 1, "memory": 6319, "step": 4000} +{"lr": 9.781027724265439e-05, "data_time": 0.6188958644866943, "loss": 0.32222879827022555, "time": 0.7586504220962524, "epoch": 1, "memory": 6319, "step": 4100} +{"accuracy/top1": 82.1375961303711, "data_time": 0.3992989824599579, "time": 0.48873059088442505, "step": 1} +{"lr": 9.769111539369434e-05, "data_time": 0.5902113676071167, "loss": 0.32364955842494963, "time": 0.7324928283691406, "epoch": 2, "memory": 6318, "step": 4211} +{"lr": 9.758113190911623e-05, "data_time": 0.6220621347427369, "loss": 0.32121524810791013, "time": 0.762659215927124, "epoch": 2, "memory": 6318, "step": 4311} +{"lr": 9.746866174465984e-05, "data_time": 0.5857436656951904, "loss": 0.33113714158535, "time": 0.7273895263671875, "epoch": 2, "memory": 6318, "step": 4411} +{"lr": 9.735371146842894e-05, "data_time": 0.6168334245681762, "loss": 0.30353142619132994, "time": 0.7569495916366578, "epoch": 2, "memory": 6318, "step": 4511} +{"lr": 9.723628779336255e-05, "data_time": 0.5955076694488526, "loss": 0.3121369868516922, "time": 0.7367850542068481, "epoch": 2, "memory": 6318, "step": 4611} +{"lr": 9.711639757684265e-05, "data_time": 0.5795055389404297, "loss": 0.32282961905002594, "time": 0.7184475183486938, "epoch": 2, "memory": 6318, "step": 4711} +{"lr": 9.699404782029395e-05, "data_time": 0.6580286741256713, "loss": 0.31532838344573977, "time": 0.7986724853515625, "epoch": 2, "memory": 6318, "step": 4811} +{"lr": 9.686924566877498e-05, "data_time": 0.5989612817764283, "loss": 0.3125740885734558, "time": 0.7385095357894897, "epoch": 2, "memory": 6318, "step": 4911} +{"lr": 9.674199841056078e-05, "data_time": 0.6176668643951416, "loss": 0.30549986362457277, "time": 0.7581380844116211, "epoch": 2, "memory": 6318, "step": 5011} +{"lr": 9.661231347671705e-05, "data_time": 0.6323682785034179, "loss": 0.3196987360715866, "time": 0.7712761640548706, "epoch": 2, "memory": 6318, "step": 5111} +{"lr": 9.64801984406668e-05, "data_time": 0.5423715591430665, "loss": 0.31711987853050233, "time": 0.6824635982513427, "epoch": 2, "memory": 6318, "step": 5211} +{"lr": 9.634566101774737e-05, "data_time": 0.5807841539382934, "loss": 0.3128260254859924, "time": 0.7204586029052734, "epoch": 2, "memory": 6318, "step": 5311} +{"lr": 9.62087090647604e-05, "data_time": 0.5787700414657593, "loss": 0.31155178844928744, "time": 0.7196438789367676, "epoch": 2, "memory": 6318, "step": 5411} +{"lr": 9.606935057951247e-05, "data_time": 0.568332028388977, "loss": 0.3232632249593735, "time": 0.7109239339828491, "epoch": 2, "memory": 6318, "step": 5511} +{"lr": 9.592759370034884e-05, "data_time": 0.638883638381958, "loss": 0.3069745123386383, "time": 0.7785221099853515, "epoch": 2, "memory": 6318, "step": 5611} +{"lr": 9.578344670567721e-05, "data_time": 0.6136923789978027, "loss": 0.29695713222026826, "time": 0.7593366622924804, "epoch": 2, "memory": 6318, "step": 5711} +{"lr": 9.563691801348495e-05, "data_time": 0.5899014711380005, "loss": 0.2924453437328339, "time": 0.7304668188095093, "epoch": 2, "memory": 6318, "step": 5811} +{"lr": 9.548801618084738e-05, "data_time": 0.6219239234924316, "loss": 0.30155878365039823, "time": 0.7622530937194825, "epoch": 2, "memory": 6318, "step": 5911} +{"lr": 9.533674990342786e-05, "data_time": 0.6147721529006958, "loss": 0.30114450454711916, "time": 0.7545836210250855, "epoch": 2, "memory": 6318, "step": 6011} +{"lr": 9.518312801497029e-05, "data_time": 0.5771173715591431, "loss": 0.28212161362171173, "time": 0.7168561935424804, "epoch": 2, "memory": 6318, "step": 6111} +{"lr": 9.502715948678276e-05, "data_time": 0.5909449100494385, "loss": 0.2921979516744614, "time": 0.7317031383514404, "epoch": 2, "memory": 6318, "step": 6211} +{"lr": 9.486885342721427e-05, "data_time": 0.622680401802063, "loss": 0.3059606194496155, "time": 0.7632664203643799, "epoch": 2, "memory": 6318, "step": 6311} +{"lr": 9.470821908112202e-05, "data_time": 0.5569717168807984, "loss": 0.2895622730255127, "time": 0.6975443840026856, "epoch": 2, "memory": 6318, "step": 6411} +{"lr": 9.454526582933214e-05, "data_time": 0.6190478563308716, "loss": 0.2918767884373665, "time": 0.7586658477783204, "epoch": 2, "memory": 6318, "step": 6511} +{"lr": 9.438000318809176e-05, "data_time": 0.5510854959487915, "loss": 0.2964482635259628, "time": 0.6919265508651733, "epoch": 2, "memory": 6318, "step": 6611} +{"lr": 9.42124408085132e-05, "data_time": 0.599868655204773, "loss": 0.3183589786291122, "time": 0.7403755426406861, "epoch": 2, "memory": 6318, "step": 6711} +{"lr": 9.404258847601036e-05, "data_time": 0.5877859592437744, "loss": 0.2691179126501083, "time": 0.7272505521774292, "epoch": 2, "memory": 6318, "step": 6811} +{"lr": 9.387045610972717e-05, "data_time": 0.5716188192367554, "loss": 0.2867441624403, "time": 0.7112274646759034, "epoch": 2, "memory": 6318, "step": 6911} +{"lr": 9.369605376195838e-05, "data_time": 0.6654116392135621, "loss": 0.26854095458984373, "time": 0.8053451776504517, "epoch": 2, "memory": 6318, "step": 7011} +{"lr": 9.351939161756269e-05, "data_time": 0.53777174949646, "loss": 0.2746407687664032, "time": 0.6781685829162598, "epoch": 2, "memory": 6318, "step": 7111} +{"lr": 9.334047999336764e-05, "data_time": 0.6016851663589478, "loss": 0.26764721125364305, "time": 0.7421585321426392, "epoch": 2, "memory": 6318, "step": 7211} +{"lr": 9.315932933756752e-05, "data_time": 0.643176007270813, "loss": 0.27293182611465455, "time": 0.7826913833618164, "epoch": 2, "memory": 6318, "step": 7311} +{"lr": 9.297595022911287e-05, "data_time": 0.5789018630981445, "loss": 0.2947516679763794, "time": 0.718891716003418, "epoch": 2, "memory": 6318, "step": 7411} +{"lr": 9.279035337709295e-05, "data_time": 0.7300005674362182, "loss": 0.2894325017929077, "time": 0.8703007698059082, "epoch": 2, "memory": 6318, "step": 7511} +{"lr": 9.260254962011016e-05, "data_time": 0.5967065334320069, "loss": 0.2779200911521912, "time": 0.7461466550827026, "epoch": 2, "memory": 6318, "step": 7611} +{"lr": 9.241254992564702e-05, "data_time": 0.6363636493682862, "loss": 0.28503446877002714, "time": 0.7983319282531738, "epoch": 2, "memory": 6318, "step": 7711} +{"lr": 9.22203653894259e-05, "data_time": 0.614773178100586, "loss": 0.2754530429840088, "time": 0.7547361135482789, "epoch": 2, "memory": 6318, "step": 7811} +{"lr": 9.202600723476098e-05, "data_time": 0.6346525192260742, "loss": 0.27611604928970335, "time": 0.7754125833511353, "epoch": 2, "memory": 6318, "step": 7911} +{"lr": 9.182948681190268e-05, "data_time": 0.5980861186981201, "loss": 0.27201849818229673, "time": 0.7377133131027221, "epoch": 2, "memory": 6318, "step": 8011} +{"lr": 9.163081559737497e-05, "data_time": 0.5946156263351441, "loss": 0.28144574016332624, "time": 0.7340287446975708, "epoch": 2, "memory": 6318, "step": 8111} +{"lr": 9.143000519330517e-05, "data_time": 0.5590303421020508, "loss": 0.2801020860671997, "time": 0.6994771003723145, "epoch": 2, "memory": 6318, "step": 8211} +{"accuracy/top1": 85.16458892822266, "data_time": 0.38208451072374977, "time": 0.4727638363838196, "step": 2} +{"lr": 9.120461478753376e-05, "data_time": 0.5781090021133423, "loss": 0.28358772546052935, "time": 0.7174955606460571, "epoch": 3, "memory": 6318, "step": 8322} +{"lr": 9.099932932138134e-05, "data_time": 0.5374660730361939, "loss": 0.2692747175693512, "time": 0.6775541782379151, "epoch": 3, "memory": 6318, "step": 8422} +{"lr": 9.079194154362181e-05, "data_time": 0.7943992853164673, "loss": 0.29565121084451673, "time": 0.9346028327941894, "epoch": 3, "memory": 6318, "step": 8522} +{"lr": 9.058246356541787e-05, "data_time": 0.6330480098724365, "loss": 0.2778221368789673, "time": 0.7732642650604248, "epoch": 3, "memory": 6318, "step": 8622} +{"lr": 9.037090761999699e-05, "data_time": 0.6890178442001342, "loss": 0.27614607810974123, "time": 0.8287760257720947, "epoch": 3, "memory": 6318, "step": 8722} +{"lr": 9.015728606193701e-05, "data_time": 0.6631336450576782, "loss": 0.290597602725029, "time": 0.8027716159820557, "epoch": 3, "memory": 6318, "step": 8822} +{"lr": 8.994161136644488e-05, "data_time": 0.5741037845611572, "loss": 0.3002932250499725, "time": 0.7148855209350586, "epoch": 3, "memory": 6318, "step": 8922} +{"lr": 8.972389612862792e-05, "data_time": 0.6331704139709473, "loss": 0.26597594022750853, "time": 0.774188494682312, "epoch": 3, "memory": 6318, "step": 9022} +{"lr": 8.950415306275833e-05, "data_time": 0.6183555603027344, "loss": 0.26705462485551834, "time": 0.7672089099884033, "epoch": 3, "memory": 6318, "step": 9122} +{"lr": 8.928239500153059e-05, "data_time": 0.5922861576080323, "loss": 0.2828182190656662, "time": 0.7321900129318237, "epoch": 3, "memory": 6318, "step": 9222} +{"lr": 8.905863489531232e-05, "data_time": 0.6252435684204102, "loss": 0.26145609468221664, "time": 0.7666311740875245, "epoch": 3, "memory": 6318, "step": 9322} +{"lr": 8.883288581138775e-05, "data_time": 0.5713315486907959, "loss": 0.27831234633922575, "time": 0.7139128684997559, "epoch": 3, "memory": 6318, "step": 9422} +{"lr": 8.86051609331947e-05, "data_time": 0.5196505069732666, "loss": 0.2668722853064537, "time": 0.6605884552001953, "epoch": 3, "memory": 6318, "step": 9522} +{"lr": 8.837547355955467e-05, "data_time": 0.6247809886932373, "loss": 0.2651767820119858, "time": 0.7680954933166504, "epoch": 3, "memory": 6318, "step": 9622} +{"lr": 8.814383710389614e-05, "data_time": 0.6110742330551148, "loss": 0.2882920101284981, "time": 0.7508492708206177, "epoch": 3, "memory": 6318, "step": 9722} +{"lr": 8.791026509347146e-05, "data_time": 0.6371283054351806, "loss": 0.25376144647598264, "time": 0.7788569927215576, "epoch": 3, "memory": 6318, "step": 9822} +{"lr": 8.767477116856659e-05, "data_time": 0.6164314270019531, "loss": 0.2567665338516235, "time": 0.7568058252334595, "epoch": 3, "memory": 6318, "step": 9922} +{"lr": 8.743736908170466e-05, "data_time": 0.5361625671386718, "loss": 0.29451175779104233, "time": 0.6756283044815063, "epoch": 3, "memory": 6318, "step": 10022} +{"lr": 8.719807269684305e-05, "data_time": 0.6020381212234497, "loss": 0.25140334069728854, "time": 0.7426766395568848, "epoch": 3, "memory": 6318, "step": 10122} +{"lr": 8.695689598856356e-05, "data_time": 0.6081812381744385, "loss": 0.2641803830862045, "time": 0.749635910987854, "epoch": 3, "memory": 6318, "step": 10222} +{"lr": 8.67138530412559e-05, "data_time": 0.6089127063751221, "loss": 0.2692801162600517, "time": 0.7482808113098145, "epoch": 3, "memory": 6318, "step": 10322} +{"lr": 8.646895804829604e-05, "data_time": 0.5897334575653076, "loss": 0.27401638180017474, "time": 0.7361309289932251, "epoch": 3, "memory": 6318, "step": 10422} +{"lr": 8.622222531121664e-05, "data_time": 0.6014791488647461, "loss": 0.2598450928926468, "time": 0.743660569190979, "epoch": 3, "memory": 6318, "step": 10522} +{"lr": 8.597366923887211e-05, "data_time": 0.6019439935684204, "loss": 0.2573289662599564, "time": 0.742353081703186, "epoch": 3, "memory": 6318, "step": 10622} +{"lr": 8.57233043465972e-05, "data_time": 0.5541431665420532, "loss": 0.26331354677677155, "time": 0.6946407318115234, "epoch": 3, "memory": 6318, "step": 10722} +{"lr": 8.547114525535937e-05, "data_time": 0.5862284421920776, "loss": 0.248159721493721, "time": 0.7373787641525269, "epoch": 3, "memory": 6318, "step": 10822} +{"lr": 8.521720669090474e-05, "data_time": 0.5396308183670044, "loss": 0.28025137782096865, "time": 0.6801734447479248, "epoch": 3, "memory": 6318, "step": 10922} +{"lr": 8.496150348289821e-05, "data_time": 0.5594420671463013, "loss": 0.2621003046631813, "time": 0.7176491022109985, "epoch": 3, "memory": 6318, "step": 11022} +{"lr": 8.470405056405736e-05, "data_time": 0.5735392570495605, "loss": 0.256819885969162, "time": 0.7138025283813476, "epoch": 3, "memory": 6318, "step": 11122} +{"lr": 8.44448629692807e-05, "data_time": 0.6116055965423584, "loss": 0.24901504814624786, "time": 0.7513501167297363, "epoch": 3, "memory": 6318, "step": 11222} +{"lr": 8.418395583476941e-05, "data_time": 0.5562048435211182, "loss": 0.26933703422546384, "time": 0.6964932203292846, "epoch": 3, "memory": 6318, "step": 11322} +{"lr": 8.392134439714336e-05, "data_time": 0.5575484991073608, "loss": 0.24063725173473358, "time": 0.7036512851715088, "epoch": 3, "memory": 6318, "step": 11422} +{"lr": 8.365704399255133e-05, "data_time": 0.5933792114257812, "loss": 0.2415594533085823, "time": 0.73389732837677, "epoch": 3, "memory": 6318, "step": 11522} +{"lr": 8.339107005577562e-05, "data_time": 0.5866213798522949, "loss": 0.2673858031630516, "time": 0.727308201789856, "epoch": 3, "memory": 6318, "step": 11622} +{"lr": 8.31234381193305e-05, "data_time": 0.613468599319458, "loss": 0.25906232595443723, "time": 0.7547986507415771, "epoch": 3, "memory": 6318, "step": 11722} +{"lr": 8.285416381255504e-05, "data_time": 0.6322313070297241, "loss": 0.24322355687618255, "time": 0.7723243474960327, "epoch": 3, "memory": 6318, "step": 11822} +{"lr": 8.258326286070047e-05, "data_time": 0.6222434282302857, "loss": 0.2620311677455902, "time": 0.7631978988647461, "epoch": 3, "memory": 6318, "step": 11922} +{"lr": 8.231075108401202e-05, "data_time": 0.5978820323944092, "loss": 0.2560994476079941, "time": 0.7383124113082886, "epoch": 3, "memory": 6318, "step": 12022} +{"lr": 8.203664439680493e-05, "data_time": 0.6929920196533204, "loss": 0.2536120295524597, "time": 0.833314061164856, "epoch": 3, "memory": 6318, "step": 12122} +{"lr": 8.176095880653475e-05, "data_time": 0.6269451141357422, "loss": 0.2562736988067627, "time": 0.7674845933914185, "epoch": 3, "memory": 6318, "step": 12222} +{"lr": 8.148371041286308e-05, "data_time": 0.5955604076385498, "loss": 0.25193780809640887, "time": 0.7358064889907837, "epoch": 3, "memory": 6318, "step": 12322} +{"accuracy/top1": 83.95809936523438, "data_time": 0.38941847284634906, "time": 0.4791889111200968, "step": 3} +{"lr": 8.117415423358443e-05, "data_time": 0.6537652254104614, "loss": 0.23664241433143615, "time": 0.795924186706543, "epoch": 4, "memory": 6318, "step": 12433} +{"lr": 8.089366155759266e-05, "data_time": 0.6240051031112671, "loss": 0.24171775281429292, "time": 0.7717869997024536, "epoch": 4, "memory": 6318, "step": 12533} +{"lr": 8.061165672717273e-05, "data_time": 0.5950781583786011, "loss": 0.2567700624465942, "time": 0.7352271318435669, "epoch": 4, "memory": 6318, "step": 12633} +{"lr": 8.032815621102116e-05, "data_time": 0.29046666622161865, "loss": 0.2457823485136032, "time": 0.7187775373458862, "epoch": 4, "memory": 6318, "step": 12733} +{"lr": 8.004317656518042e-05, "data_time": 0.4857169151306152, "loss": 0.25299454629421236, "time": 0.7400350570678711, "epoch": 4, "memory": 6318, "step": 12833} +{"lr": 7.975673443207223e-05, "data_time": 0.3014135122299194, "loss": 0.248407806456089, "time": 0.7418803930282593, "epoch": 4, "memory": 6318, "step": 12933} +{"lr": 7.946884653952537e-05, "data_time": 0.6789196014404297, "loss": 0.23907822072505952, "time": 0.8201666355133057, "epoch": 4, "memory": 6318, "step": 13033} +{"lr": 7.91795296997992e-05, "data_time": 0.6203129768371582, "loss": 0.2586119264364243, "time": 0.763839316368103, "epoch": 4, "memory": 6318, "step": 13133} +{"lr": 7.888880080860151e-05, "data_time": 0.5903083562850953, "loss": 0.2584664598107338, "time": 0.7317310810089112, "epoch": 4, "memory": 6318, "step": 13233} +{"lr": 7.859667684410205e-05, "data_time": 0.6058751344680786, "loss": 0.22599205672740935, "time": 0.7473642826080322, "epoch": 4, "memory": 6318, "step": 13333} +{"lr": 7.830317486594082e-05, "data_time": 0.6215017795562744, "loss": 0.25034799575805666, "time": 0.761247205734253, "epoch": 4, "memory": 6318, "step": 13433} +{"lr": 7.800831201423207e-05, "data_time": 0.560490345954895, "loss": 0.25423461496829985, "time": 0.700891661643982, "epoch": 4, "memory": 6318, "step": 13533} +{"lr": 7.771210550856309e-05, "data_time": 0.6174039125442505, "loss": 0.2564885541796684, "time": 0.7578014135360718, "epoch": 4, "memory": 6318, "step": 13633} +{"lr": 7.741457264698884e-05, "data_time": 0.7111117839813232, "loss": 0.2498186483979225, "time": 0.852037763595581, "epoch": 4, "memory": 6318, "step": 13733} +{"lr": 7.711573080502147e-05, "data_time": 0.5510381937026978, "loss": 0.25023491531610487, "time": 0.6926404476165772, "epoch": 4, "memory": 6318, "step": 13833} +{"lr": 7.681559743461609e-05, "data_time": 0.5712744951248169, "loss": 0.2362610563635826, "time": 0.7112491607666016, "epoch": 4, "memory": 6318, "step": 13933} +{"lr": 7.651419006315099e-05, "data_time": 0.6209005355834961, "loss": 0.254113607108593, "time": 0.7617431163787842, "epoch": 4, "memory": 6318, "step": 14033} +{"lr": 7.621152629240453e-05, "data_time": 0.6097277164459228, "loss": 0.24347356259822844, "time": 0.7496002197265625, "epoch": 4, "memory": 6318, "step": 14133} +{"lr": 7.590762379752719e-05, "data_time": 0.6082467794418335, "loss": 0.2417900651693344, "time": 0.7484559774398803, "epoch": 4, "memory": 6318, "step": 14233} +{"lr": 7.560250032600911e-05, "data_time": 0.6105065822601319, "loss": 0.22725880444049834, "time": 0.751008152961731, "epoch": 4, "memory": 6318, "step": 14333} +{"lr": 7.52961736966437e-05, "data_time": 0.5979675054550171, "loss": 0.23678749948740005, "time": 0.7382069826126099, "epoch": 4, "memory": 6318, "step": 14433} +{"lr": 7.498866179848732e-05, "data_time": 0.602944278717041, "loss": 0.25134188383817674, "time": 0.745530652999878, "epoch": 4, "memory": 6318, "step": 14533} +{"lr": 7.467998258981415e-05, "data_time": 0.6060527324676513, "loss": 0.24453014135360718, "time": 0.7461367607116699, "epoch": 4, "memory": 6318, "step": 14633} +{"lr": 7.437015409706804e-05, "data_time": 0.5987406969070435, "loss": 0.244582162797451, "time": 0.7388062715530396, "epoch": 4, "memory": 6318, "step": 14733} +{"lr": 7.405919441380929e-05, "data_time": 0.5786459445953369, "loss": 0.24392749071121217, "time": 0.7198333501815796, "epoch": 4, "memory": 6318, "step": 14833} +{"lr": 7.374712169965821e-05, "data_time": 0.6221743106842041, "loss": 0.25384840965270994, "time": 0.7634946346282959, "epoch": 4, "memory": 6318, "step": 14933} +{"lr": 7.343395417923461e-05, "data_time": 0.6961986541748046, "loss": 0.23877278566360474, "time": 0.8362715721130372, "epoch": 4, "memory": 6318, "step": 15033} +{"lr": 7.311971014109342e-05, "data_time": 0.6279742479324341, "loss": 0.23283368051052095, "time": 0.7792951822280884, "epoch": 4, "memory": 6318, "step": 15133} +{"lr": 7.280440793665682e-05, "data_time": 0.5790355443954468, "loss": 0.2475575938820839, "time": 0.7281638860702515, "epoch": 4, "memory": 6318, "step": 15233} +{"lr": 7.248806597914234e-05, "data_time": 0.6665335655212402, "loss": 0.23362425565719605, "time": 0.8058000802993774, "epoch": 4, "memory": 6318, "step": 15333} +{"lr": 7.217070274248765e-05, "data_time": 0.6398946046829224, "loss": 0.23349884301424026, "time": 0.7801725387573242, "epoch": 4, "memory": 6318, "step": 15433} +{"lr": 7.185233676027203e-05, "data_time": 0.6243457555770874, "loss": 0.2130759537220001, "time": 0.7681803226470947, "epoch": 4, "memory": 6318, "step": 15533} +{"lr": 7.153298662463338e-05, "data_time": 0.6000623226165771, "loss": 0.2271370217204094, "time": 0.7452205419540405, "epoch": 4, "memory": 6318, "step": 15633} +{"lr": 7.121267098518291e-05, "data_time": 0.6265114545822144, "loss": 0.27117105275392533, "time": 0.7745836496353149, "epoch": 4, "memory": 6318, "step": 15733} +{"lr": 7.089140854791618e-05, "data_time": 0.5846446514129638, "loss": 0.22592310905456542, "time": 0.7251559495925903, "epoch": 4, "memory": 6318, "step": 15833} +{"lr": 7.056921807412025e-05, "data_time": 0.6570351123809814, "loss": 0.2314138889312744, "time": 0.7984226703643799, "epoch": 4, "memory": 6318, "step": 15933} +{"lr": 7.024611837927802e-05, "data_time": 0.6338525295257569, "loss": 0.2543148994445801, "time": 0.7740651845932007, "epoch": 4, "memory": 6318, "step": 16033} +{"lr": 6.992212833196998e-05, "data_time": 0.5853900909423828, "loss": 0.23783159255981445, "time": 0.7274734020233155, "epoch": 4, "memory": 6318, "step": 16133} +{"lr": 6.959726685277174e-05, "data_time": 0.5680332660675049, "loss": 0.23654840439558028, "time": 0.7082507610321045, "epoch": 4, "memory": 6318, "step": 16233} +{"lr": 6.927155291314949e-05, "data_time": 0.6556339263916016, "loss": 0.24805202186107636, "time": 0.7951558589935303, "epoch": 4, "memory": 6318, "step": 16333} +{"lr": 6.894500553435204e-05, "data_time": 0.5759670019149781, "loss": 0.2458961620926857, "time": 0.7164654731750488, "epoch": 4, "memory": 6318, "step": 16433} +{"accuracy/top1": 88.21310424804688, "data_time": 0.36774808367093403, "time": 0.45801969766616824, "step": 4} +{"lr": 6.858158509719903e-05, "data_time": 0.634275221824646, "loss": 0.24288712292909623, "time": 0.8501934289932251, "epoch": 5, "memory": 6318, "step": 16544} +{"lr": 6.825334178880171e-05, "data_time": 0.5777467966079712, "loss": 0.24366383105516434, "time": 0.7923753023147583, "epoch": 5, "memory": 6318, "step": 16644} +{"lr": 6.792432450336645e-05, "data_time": 0.33480277061462405, "loss": 0.22490052431821822, "time": 0.7397310972213745, "epoch": 5, "memory": 6318, "step": 16744} +{"lr": 6.759455245505286e-05, "data_time": 0.2291111707687378, "loss": 0.2554459750652313, "time": 0.7925480365753174, "epoch": 5, "memory": 6318, "step": 16844} +{"lr": 6.726404490209753e-05, "data_time": 0.33718101978302, "loss": 0.22675415724515915, "time": 0.7337573766708374, "epoch": 5, "memory": 6318, "step": 16944} +{"lr": 6.693282114568972e-05, "data_time": 0.4749349594116211, "loss": 0.2320703700184822, "time": 0.763561749458313, "epoch": 5, "memory": 6318, "step": 17044} +{"lr": 6.660090052884388e-05, "data_time": 0.3626111030578613, "loss": 0.22647695541381835, "time": 0.7585147380828857, "epoch": 5, "memory": 6318, "step": 17144} +{"lr": 6.626830243527016e-05, "data_time": 0.4016925096511841, "loss": 0.23891469538211824, "time": 0.7527997016906738, "epoch": 5, "memory": 6318, "step": 17244} +{"lr": 6.593504628824236e-05, "data_time": 0.2532955646514893, "loss": 0.2269635319709778, "time": 0.7245667219161988, "epoch": 5, "memory": 6318, "step": 17344} +{"lr": 6.560115154946387e-05, "data_time": 0.4445216178894043, "loss": 0.25372825264930726, "time": 0.7384096622467041, "epoch": 5, "memory": 6318, "step": 17444} +{"lr": 6.526663771793081e-05, "data_time": 0.4799137830734253, "loss": 0.22184183895587922, "time": 0.7676207065582276, "epoch": 5, "memory": 6318, "step": 17544} +{"lr": 6.493152432879363e-05, "data_time": 0.5886412620544433, "loss": 0.23784775137901307, "time": 0.7836074590682983, "epoch": 5, "memory": 6318, "step": 17644} +{"lr": 6.459583095221596e-05, "data_time": 0.6631577491760254, "loss": 0.22672287821769715, "time": 0.8032900094985962, "epoch": 5, "memory": 6318, "step": 17744} +{"lr": 6.425957719223197e-05, "data_time": 0.5715037822723389, "loss": 0.2148836672306061, "time": 0.7109566926956177, "epoch": 5, "memory": 6318, "step": 17844} +{"lr": 6.392278268560153e-05, "data_time": 0.5947471141815186, "loss": 0.2364882528781891, "time": 0.7344706058502197, "epoch": 5, "memory": 6318, "step": 17944} +{"lr": 6.358546710066316e-05, "data_time": 0.5587163925170898, "loss": 0.2586208418011665, "time": 0.6994696617126465, "epoch": 5, "memory": 6318, "step": 18044} +{"lr": 6.324765013618573e-05, "data_time": 0.6116652488708496, "loss": 0.21596358716487885, "time": 0.7579565048217773, "epoch": 5, "memory": 6318, "step": 18144} +{"lr": 6.290935152021818e-05, "data_time": 0.7778600454330444, "loss": 0.2266959324479103, "time": 0.9177164554595947, "epoch": 5, "memory": 6318, "step": 18244} +{"lr": 6.2570591008937e-05, "data_time": 0.6113502264022828, "loss": 0.22128949016332627, "time": 0.7559655189514161, "epoch": 5, "memory": 6318, "step": 18344} +{"lr": 6.223138838549294e-05, "data_time": 0.581654691696167, "loss": 0.22663779109716414, "time": 0.721722149848938, "epoch": 5, "memory": 6318, "step": 18444} +{"lr": 6.18917634588553e-05, "data_time": 0.6068169116973877, "loss": 0.22119206190109253, "time": 0.7467026948928833, "epoch": 5, "memory": 6318, "step": 18544} +{"lr": 6.155173606265555e-05, "data_time": 0.590725040435791, "loss": 0.23121103793382644, "time": 0.7319432258605957, "epoch": 5, "memory": 6318, "step": 18644} +{"lr": 6.121132605402843e-05, "data_time": 0.664388108253479, "loss": 0.22820659428834916, "time": 0.8045989513397217, "epoch": 5, "memory": 6318, "step": 18744} +{"lr": 6.087055331245287e-05, "data_time": 0.5794222354888916, "loss": 0.22593510150909424, "time": 0.7198932886123657, "epoch": 5, "memory": 6318, "step": 18844} +{"lr": 6.05294377385911e-05, "data_time": 0.6478742361068726, "loss": 0.21822001785039902, "time": 0.7875202655792236, "epoch": 5, "memory": 6318, "step": 18944} +{"lr": 6.0187999253126355e-05, "data_time": 0.6332026243209838, "loss": 0.20956631898880004, "time": 0.7739517688751221, "epoch": 5, "memory": 6318, "step": 19044} +{"lr": 5.9846257795599195e-05, "data_time": 0.5757324695587158, "loss": 0.21944597065448762, "time": 0.715477442741394, "epoch": 5, "memory": 6318, "step": 19144} +{"lr": 5.950423332324355e-05, "data_time": 0.7819278478622437, "loss": 0.23812942504882811, "time": 0.921880578994751, "epoch": 5, "memory": 6318, "step": 19244} +{"lr": 5.9161945809820855e-05, "data_time": 0.5880257368087769, "loss": 0.22361236661672593, "time": 0.7290972948074341, "epoch": 5, "memory": 6318, "step": 19344} +{"lr": 5.8819415244453975e-05, "data_time": 0.6764802217483521, "loss": 0.2236904874444008, "time": 0.8174845457077027, "epoch": 5, "memory": 6318, "step": 19444} +{"lr": 5.8476661630459405e-05, "data_time": 0.6367762804031372, "loss": 0.23582514822483064, "time": 0.7781792402267456, "epoch": 5, "memory": 6318, "step": 19544} +{"lr": 5.81337049841796e-05, "data_time": 0.5800023555755616, "loss": 0.23952553421258926, "time": 0.720527172088623, "epoch": 5, "memory": 6318, "step": 19644} +{"lr": 5.779056533381373e-05, "data_time": 0.682162880897522, "loss": 0.2372528463602066, "time": 0.8233830451965332, "epoch": 5, "memory": 6318, "step": 19744} +{"lr": 5.744726271824817e-05, "data_time": 0.5984102964401246, "loss": 0.21181246042251586, "time": 0.7395000219345093, "epoch": 5, "memory": 6318, "step": 19844} +{"lr": 5.710381718588622e-05, "data_time": 0.6184651851654053, "loss": 0.23702518194913863, "time": 0.759309196472168, "epoch": 5, "memory": 6318, "step": 19944} +{"lr": 5.676024879347741e-05, "data_time": 0.6384994506835937, "loss": 0.2289959654211998, "time": 0.7787276268005371, "epoch": 5, "memory": 6318, "step": 20044} +{"lr": 5.6416577604945966e-05, "data_time": 0.5880339860916137, "loss": 0.24212294965982437, "time": 0.7277263641357422, "epoch": 5, "memory": 6318, "step": 20144} +{"lr": 5.6072823690219385e-05, "data_time": 0.6430895805358887, "loss": 0.22380499690771102, "time": 0.7862739324569702, "epoch": 5, "memory": 6318, "step": 20244} +{"lr": 5.572900712405624e-05, "data_time": 0.6050524234771728, "loss": 0.2059819668531418, "time": 0.74480721950531, "epoch": 5, "memory": 6318, "step": 20344} +{"lr": 5.538514798487397e-05, "data_time": 0.6456482648849488, "loss": 0.2494832068681717, "time": 0.7857419013977051, "epoch": 5, "memory": 6318, "step": 20444} +{"lr": 5.5041266353576e-05, "data_time": 0.5834356307983398, "loss": 0.23772364109754562, "time": 0.7250214099884034, "epoch": 5, "memory": 6318, "step": 20544} +{"accuracy/top1": 89.27395629882812, "data_time": 0.39095328251520794, "time": 0.48003365397453307, "step": 5} +{"lr": 5.4659555783035384e-05, "data_time": 0.016553759574890137, "loss": 0.22144537419080734, "time": 0.7474731922149658, "epoch": 6, "memory": 6318, "step": 20655} +{"lr": 5.431569258426336e-05, "data_time": 0.07853200435638427, "loss": 0.2264638990163803, "time": 0.751076626777649, "epoch": 6, "memory": 6318, "step": 20755} +{"lr": 5.397186934810864e-05, "data_time": 0.1736987352371216, "loss": 0.22422888725996018, "time": 0.7852191209793091, "epoch": 6, "memory": 6318, "step": 20855} +{"lr": 5.362810615337815e-05, "data_time": 0.201401686668396, "loss": 0.2276495799422264, "time": 0.7673514366149903, "epoch": 6, "memory": 6318, "step": 20955} +{"lr": 5.328442307537233e-05, "data_time": 0.3652447462081909, "loss": 0.2285897895693779, "time": 0.6923784732818603, "epoch": 6, "memory": 6318, "step": 21055} +{"lr": 5.294084018471303e-05, "data_time": 0.6161708116531373, "loss": 0.22080680429935456, "time": 0.7842055320739746, "epoch": 6, "memory": 6318, "step": 21155} +{"lr": 5.259737754617132e-05, "data_time": 0.6010441780090332, "loss": 0.22242555171251296, "time": 0.7420645952224731, "epoch": 6, "memory": 6318, "step": 21255} +{"lr": 5.2254055217495525e-05, "data_time": 0.5866324186325074, "loss": 0.22232354879379274, "time": 0.726702070236206, "epoch": 6, "memory": 6318, "step": 21355} +{"lr": 5.1910893248240283e-05, "data_time": 0.6091682434082031, "loss": 0.2368236780166626, "time": 0.7500319242477417, "epoch": 6, "memory": 6318, "step": 21455} +{"lr": 5.156791167859532e-05, "data_time": 0.67364022731781, "loss": 0.2161878764629364, "time": 0.8277969121932983, "epoch": 6, "memory": 6318, "step": 21555} +{"lr": 5.122513053821533e-05, "data_time": 0.9497917175292969, "loss": 0.2425651580095291, "time": 1.090404224395752, "epoch": 6, "memory": 6318, "step": 21655} +{"lr": 5.0882569845050173e-05, "data_time": 0.6629886627197266, "loss": 0.22540825456380845, "time": 0.8105605125427247, "epoch": 6, "memory": 6318, "step": 21755} +{"lr": 5.054024960417592e-05, "data_time": 0.6264513731002808, "loss": 0.22059424519538878, "time": 0.7666175365447998, "epoch": 6, "memory": 6318, "step": 21855} +{"lr": 5.0198189806626494e-05, "data_time": 0.6422939300537109, "loss": 0.2257988378405571, "time": 0.7936925411224365, "epoch": 6, "memory": 6318, "step": 21955} +{"lr": 4.985641042822647e-05, "data_time": 0.6275058031082154, "loss": 0.22144672274589539, "time": 0.7679895877838134, "epoch": 6, "memory": 6318, "step": 22055} +{"lr": 4.951493142842412e-05, "data_time": 0.5662127256393432, "loss": 0.24146845936775208, "time": 0.706992769241333, "epoch": 6, "memory": 6318, "step": 22155} +{"lr": 4.917377274912603e-05, "data_time": 0.5645421504974365, "loss": 0.21499475240707397, "time": 0.705557894706726, "epoch": 6, "memory": 6318, "step": 22255} +{"lr": 4.88329543135327e-05, "data_time": 0.6381146907806396, "loss": 0.2316078558564186, "time": 0.7787853956222535, "epoch": 6, "memory": 6318, "step": 22355} +{"lr": 4.849249602497467e-05, "data_time": 0.6599221706390381, "loss": 0.22825948745012284, "time": 0.8002774000167847, "epoch": 6, "memory": 6318, "step": 22455} +{"lr": 4.8152417765750456e-05, "data_time": 0.5728270530700683, "loss": 0.2311478853225708, "time": 0.7126530408859253, "epoch": 6, "memory": 6318, "step": 22555} +{"lr": 4.7812739395965355e-05, "data_time": 0.6469001293182373, "loss": 0.20607706904411316, "time": 0.7973691463470459, "epoch": 6, "memory": 6318, "step": 22655} +{"lr": 4.74734807523717e-05, "data_time": 0.6172947645187378, "loss": 0.2191238969564438, "time": 0.7578845500946045, "epoch": 6, "memory": 6318, "step": 22755} +{"lr": 4.7134661647210416e-05, "data_time": 0.5940561294555664, "loss": 0.21722680926322938, "time": 0.7390312671661377, "epoch": 6, "memory": 6318, "step": 22855} +{"lr": 4.679630186705388e-05, "data_time": 0.6066904544830323, "loss": 0.2189228653907776, "time": 0.7470699548721313, "epoch": 6, "memory": 6318, "step": 22955} +{"lr": 4.6458421171650644e-05, "data_time": 0.6506342411041259, "loss": 0.21992774307727814, "time": 0.7908859491348267, "epoch": 6, "memory": 6318, "step": 23055} +{"lr": 4.6121039292771295e-05, "data_time": 0.6431920528411865, "loss": 0.21492919474840164, "time": 0.784492802619934, "epoch": 6, "memory": 6318, "step": 23155} +{"lr": 4.5784175933056197e-05, "data_time": 0.6262296438217163, "loss": 0.22508245557546616, "time": 0.7773866653442383, "epoch": 6, "memory": 6318, "step": 23255} +{"lr": 4.544785076486494e-05, "data_time": 0.6410591840744019, "loss": 0.23063731491565703, "time": 0.7819517135620118, "epoch": 6, "memory": 6318, "step": 23355} +{"lr": 4.5112083429127475e-05, "data_time": 0.5785451173782349, "loss": 0.22566608637571334, "time": 0.7186836004257202, "epoch": 6, "memory": 6318, "step": 23455} +{"lr": 4.477689353419706e-05, "data_time": 0.5760425090789795, "loss": 0.22161386758089066, "time": 0.7169084072113037, "epoch": 6, "memory": 6318, "step": 23555} +{"lr": 4.4442300654705266e-05, "data_time": 0.6023649215698242, "loss": 0.2114580288529396, "time": 0.7510221242904663, "epoch": 6, "memory": 6318, "step": 23655} +{"lr": 4.410832433041866e-05, "data_time": 0.6313949584960937, "loss": 0.2094666451215744, "time": 0.7711641550064087, "epoch": 6, "memory": 6318, "step": 23755} +{"lr": 4.377498406509792e-05, "data_time": 0.6469725847244263, "loss": 0.21666625291109085, "time": 0.786846113204956, "epoch": 6, "memory": 6318, "step": 23855} +{"lr": 4.344229932535875e-05, "data_time": 0.5941392660140992, "loss": 0.2367297574877739, "time": 0.7350420951843262, "epoch": 6, "memory": 6318, "step": 23955} +{"lr": 4.3110289539535115e-05, "data_time": 0.6539195775985718, "loss": 0.21627108454704286, "time": 0.793128776550293, "epoch": 6, "memory": 6318, "step": 24055} +{"lr": 4.2778974096544436e-05, "data_time": 0.6331358194351197, "loss": 0.22466585040092468, "time": 0.7742410659790039, "epoch": 6, "memory": 6318, "step": 24155} +{"lr": 4.2448372344755544e-05, "data_time": 0.5891356706619263, "loss": 0.21508681029081345, "time": 0.7315256357192993, "epoch": 6, "memory": 6318, "step": 24255} +{"lr": 4.211850359085871e-05, "data_time": 0.66728355884552, "loss": 0.21390046775341034, "time": 0.8074127912521363, "epoch": 6, "memory": 6318, "step": 24355} +{"lr": 4.17893870987381e-05, "data_time": 0.5841787815093994, "loss": 0.22484858185052872, "time": 0.7302618503570557, "epoch": 6, "memory": 6318, "step": 24455} +{"lr": 4.146104208834679e-05, "data_time": 0.6124205589294434, "loss": 0.20944125652313234, "time": 0.7560169219970703, "epoch": 6, "memory": 6318, "step": 24555} +{"lr": 4.113348773458442e-05, "data_time": 0.5681646823883056, "loss": 0.2219816580414772, "time": 0.7095581531524658, "epoch": 6, "memory": 6318, "step": 24655} +{"accuracy/top1": 91.06466674804688, "data_time": 0.3798326849937439, "time": 0.47045179406801857, "step": 6} +{"lr": 4.0770851520835164e-05, "data_time": 0.32087929248809816, "loss": 0.21204094886779784, "time": 0.7360997200012207, "epoch": 7, "memory": 6318, "step": 24766} +{"lr": 4.044502815742039e-05, "data_time": 0.2319826602935791, "loss": 0.23103538602590562, "time": 0.7526503324508667, "epoch": 7, "memory": 6318, "step": 24866} +{"lr": 4.012005478445896e-05, "data_time": 0.19448323249816896, "loss": 0.21832422018051148, "time": 0.7050837516784668, "epoch": 7, "memory": 6318, "step": 24966} +{"lr": 3.9795950379951394e-05, "data_time": 0.49231514930725095, "loss": 0.2312387242913246, "time": 0.7812903881072998, "epoch": 7, "memory": 6318, "step": 25066} +{"lr": 3.947273387115169e-05, "data_time": 0.5172301530838013, "loss": 0.22581694275140762, "time": 0.7241688013076782, "epoch": 7, "memory": 6318, "step": 25166} +{"lr": 3.9150424133462044e-05, "data_time": 0.5714112997055054, "loss": 0.23753017634153367, "time": 0.7150854587554931, "epoch": 7, "memory": 6318, "step": 25266} +{"lr": 3.882903998933037e-05, "data_time": 0.5933114528656006, "loss": 0.23574368059635162, "time": 0.7412816047668457, "epoch": 7, "memory": 6318, "step": 25366} +{"lr": 3.8508600207151264e-05, "data_time": 0.6053260087966919, "loss": 0.21976497322320937, "time": 0.7513857364654541, "epoch": 7, "memory": 6318, "step": 25466} +{"lr": 3.818912350016978e-05, "data_time": 0.5936861753463745, "loss": 0.2058437153697014, "time": 0.7381568193435669, "epoch": 7, "memory": 6318, "step": 25566} +{"lr": 3.787062852538883e-05, "data_time": 0.5775141239166259, "loss": 0.21195100247859955, "time": 0.7230489730834961, "epoch": 7, "memory": 6318, "step": 25666} +{"lr": 3.7553133882479346e-05, "data_time": 0.5990762710571289, "loss": 0.22675100564956666, "time": 0.7402037858963013, "epoch": 7, "memory": 6318, "step": 25766} +{"lr": 3.723665811269437e-05, "data_time": 0.5240511178970337, "loss": 0.21259500086307526, "time": 0.7690648794174194, "epoch": 7, "memory": 6318, "step": 25866} +{"lr": 3.6921219697786176e-05, "data_time": 0.34241557121276855, "loss": 0.2061529576778412, "time": 0.7883437871932983, "epoch": 7, "memory": 6318, "step": 25966} +{"lr": 3.660683705892677e-05, "data_time": 0.3457735300064087, "loss": 0.20303792506456375, "time": 0.7385384321212769, "epoch": 7, "memory": 6318, "step": 26066} +{"lr": 3.6293528555632435e-05, "data_time": 0.5796358823776245, "loss": 0.23252723664045333, "time": 0.8354166269302368, "epoch": 7, "memory": 6318, "step": 26166} +{"lr": 3.598131248469134e-05, "data_time": 0.6267612934112549, "loss": 0.22580407410860062, "time": 0.766175365447998, "epoch": 7, "memory": 6318, "step": 26266} +{"lr": 3.567020707909507e-05, "data_time": 0.6017528772354126, "loss": 0.22389102578163148, "time": 0.7411997079849243, "epoch": 7, "memory": 6318, "step": 26366} +{"lr": 3.5360230506973935e-05, "data_time": 0.6216040134429932, "loss": 0.21377645581960678, "time": 0.7616827011108398, "epoch": 7, "memory": 6318, "step": 26466} +{"lr": 3.5051400870535877e-05, "data_time": 0.5958281755447388, "loss": 0.2313540667295456, "time": 0.7514890909194947, "epoch": 7, "memory": 6318, "step": 26566} +{"lr": 3.47437362050094e-05, "data_time": 0.648256254196167, "loss": 0.2143481358885765, "time": 0.7904376268386841, "epoch": 7, "memory": 6318, "step": 26666} +{"lr": 3.443725447759023e-05, "data_time": 0.6556012868881226, "loss": 0.21419825702905654, "time": 0.7967238187789917, "epoch": 7, "memory": 6318, "step": 26766} +{"lr": 3.413197358639224e-05, "data_time": 0.9352653980255127, "loss": 0.2334531933069229, "time": 1.078092384338379, "epoch": 7, "memory": 6318, "step": 26866} +{"lr": 3.382791135940194e-05, "data_time": 0.5899015903472901, "loss": 0.21842274218797683, "time": 0.7348134994506836, "epoch": 7, "memory": 6318, "step": 26966} +{"lr": 3.35250855534377e-05, "data_time": 0.6248748064041137, "loss": 0.21042381525039672, "time": 0.764406704902649, "epoch": 7, "memory": 6318, "step": 27066} +{"lr": 3.3223513853112476e-05, "data_time": 0.6145540952682496, "loss": 0.22413529455661774, "time": 0.7665135860443115, "epoch": 7, "memory": 6318, "step": 27166} +{"lr": 3.292321386980115e-05, "data_time": 0.589860463142395, "loss": 0.21802660971879959, "time": 0.7305471181869507, "epoch": 7, "memory": 6318, "step": 27266} +{"lr": 3.262420314061217e-05, "data_time": 0.691631269454956, "loss": 0.20277985632419587, "time": 0.8325157880783081, "epoch": 7, "memory": 6318, "step": 27366} +{"lr": 3.2326499127363195e-05, "data_time": 0.6181166172027588, "loss": 0.2208991304039955, "time": 0.7609508037567139, "epoch": 7, "memory": 6318, "step": 27466} +{"lr": 3.2030119215561613e-05, "data_time": 0.6322191238403321, "loss": 0.22663544714450837, "time": 0.7726822376251221, "epoch": 7, "memory": 6318, "step": 27566} +{"lr": 3.173508071338896e-05, "data_time": 0.6718494653701782, "loss": 0.22389377951622008, "time": 0.8124493360519409, "epoch": 7, "memory": 6318, "step": 27666} +{"lr": 3.144140085069037e-05, "data_time": 0.6478348255157471, "loss": 0.20451682657003403, "time": 0.7879785776138306, "epoch": 7, "memory": 6318, "step": 27766} +{"lr": 3.114909677796814e-05, "data_time": 0.6669594526290894, "loss": 0.21915597915649415, "time": 0.8066153526306152, "epoch": 7, "memory": 6318, "step": 27866} +{"lr": 3.085818556538059e-05, "data_time": 0.6285897970199585, "loss": 0.21591716706752778, "time": 0.7689493894577026, "epoch": 7, "memory": 6318, "step": 27966} +{"lr": 3.056868420174455e-05, "data_time": 0.5792436838150025, "loss": 0.21763637363910676, "time": 0.7191099643707275, "epoch": 7, "memory": 6318, "step": 28066} +{"lr": 3.028060959354392e-05, "data_time": 0.612235164642334, "loss": 0.2165424942970276, "time": 0.758737850189209, "epoch": 7, "memory": 6318, "step": 28166} +{"lr": 2.9993978563941876e-05, "data_time": 0.582972240447998, "loss": 0.23131911009550093, "time": 0.7238704919815063, "epoch": 7, "memory": 6318, "step": 28266} +{"lr": 2.9708807851798587e-05, "data_time": 0.6110140085220337, "loss": 0.2152962863445282, "time": 0.7591787338256836, "epoch": 7, "memory": 6318, "step": 28366} +{"lr": 2.9425114110693717e-05, "data_time": 0.6803679704666138, "loss": 0.20685464441776275, "time": 0.8211320877075196, "epoch": 7, "memory": 6318, "step": 28466} +{"lr": 2.9142913907953844e-05, "data_time": 0.6136783838272095, "loss": 0.22012823671102524, "time": 0.7671475172042846, "epoch": 7, "memory": 6318, "step": 28566} +{"lr": 2.8862223723684912e-05, "data_time": 0.6209859132766724, "loss": 0.22623719722032548, "time": 0.7611841440200806, "epoch": 7, "memory": 6318, "step": 28666} +{"lr": 2.8583059949809854e-05, "data_time": 0.5121319532394409, "loss": 0.23409331738948821, "time": 0.6742612600326539, "epoch": 7, "memory": 6318, "step": 28766} +{"accuracy/top1": 92.36052703857422, "data_time": 0.38193190296490986, "time": 0.4709255814552307, "step": 7} +{"lr": 2.827499545208893e-05, "data_time": 0.610418176651001, "loss": 0.22626211494207382, "time": 0.7521275043487549, "epoch": 8, "memory": 6318, "step": 28877} +{"lr": 2.7999105784987406e-05, "data_time": 0.6447580814361572, "loss": 0.21675214320421218, "time": 0.7882266283035279, "epoch": 8, "memory": 6318, "step": 28977} +{"lr": 2.772479293318711e-05, "data_time": 0.5873958587646484, "loss": 0.2133768081665039, "time": 0.7282128810882569, "epoch": 8, "memory": 6318, "step": 29077} +{"lr": 2.7452072916183603e-05, "data_time": 0.5236689805984497, "loss": 0.21600501984357834, "time": 0.7311687231063843, "epoch": 8, "memory": 6318, "step": 29177} +{"lr": 2.7180961660452883e-05, "data_time": 0.5871314764022827, "loss": 0.21424666345119475, "time": 0.7278459548950196, "epoch": 8, "memory": 6318, "step": 29277} +{"lr": 2.6911474998521622e-05, "data_time": 0.6580465078353882, "loss": 0.22579137682914735, "time": 0.7979048013687133, "epoch": 8, "memory": 6318, "step": 29377} +{"lr": 2.664362866804239e-05, "data_time": 0.546606469154358, "loss": 0.2086592972278595, "time": 0.6880997657775879, "epoch": 8, "memory": 6318, "step": 29477} +{"lr": 2.637743831087459e-05, "data_time": 0.6517598152160644, "loss": 0.2018253117799759, "time": 0.7980185985565186, "epoch": 8, "memory": 6318, "step": 29577} +{"lr": 2.6112919472171114e-05, "data_time": 0.6430217742919921, "loss": 0.2117971658706665, "time": 0.7836272239685058, "epoch": 8, "memory": 6318, "step": 29677} +{"lr": 2.5850087599470434e-05, "data_time": 0.5987901210784912, "loss": 0.20031181424856187, "time": 0.7499037027359009, "epoch": 8, "memory": 6318, "step": 29777} +{"lr": 2.5588958041794503e-05, "data_time": 0.5884083032608032, "loss": 0.196441125869751, "time": 0.7286111354827881, "epoch": 8, "memory": 6318, "step": 29877} +{"lr": 2.5329546048752453e-05, "data_time": 0.6950069904327393, "loss": 0.21808973401784898, "time": 0.8365283012390137, "epoch": 8, "memory": 6318, "step": 29977} +{"lr": 2.5071866769649806e-05, "data_time": 0.6069842338562011, "loss": 0.2107990190386772, "time": 0.7475652456283569, "epoch": 8, "memory": 6318, "step": 30077} +{"lr": 2.4815935252604222e-05, "data_time": 0.6724363565444946, "loss": 0.21933402717113495, "time": 0.8261152029037475, "epoch": 8, "memory": 6318, "step": 30177} +{"lr": 2.4561766443666384e-05, "data_time": 0.6750922918319702, "loss": 0.22104105055332185, "time": 0.815656328201294, "epoch": 8, "memory": 6318, "step": 30277} +{"lr": 2.4309375185947164e-05, "data_time": 0.6079245805740356, "loss": 0.2408744901418686, "time": 0.7474660634994507, "epoch": 8, "memory": 6318, "step": 30377} +{"lr": 2.40587762187509e-05, "data_time": 0.48343064785003664, "loss": 0.19970002323389052, "time": 0.7202982425689697, "epoch": 8, "memory": 6318, "step": 30477} +{"lr": 2.3809984176714634e-05, "data_time": 0.2909902334213257, "loss": 0.21772560924291612, "time": 0.7505961656570435, "epoch": 8, "memory": 6318, "step": 30577} +{"lr": 2.3563013588953406e-05, "data_time": 0.34513111114501954, "loss": 0.22359851002693176, "time": 0.7564613580703735, "epoch": 8, "memory": 6318, "step": 30677} +{"lr": 2.331787887821182e-05, "data_time": 0.3460818290710449, "loss": 0.23591297566890718, "time": 0.6941813468933106, "epoch": 8, "memory": 6318, "step": 30777} +{"lr": 2.307459436002179e-05, "data_time": 0.3455662727355957, "loss": 0.20404092818498612, "time": 0.7597615718841553, "epoch": 8, "memory": 6318, "step": 30877} +{"lr": 2.2833174241866527e-05, "data_time": 0.36763978004455566, "loss": 0.20681032091379165, "time": 0.8021057367324829, "epoch": 8, "memory": 6318, "step": 30977} +{"lr": 2.259363262235086e-05, "data_time": 0.1274261713027954, "loss": 0.2076197549700737, "time": 0.743987250328064, "epoch": 8, "memory": 6318, "step": 31077} +{"lr": 2.2355983490377735e-05, "data_time": 0.23604495525360109, "loss": 0.22067641466856003, "time": 0.7981895446777344, "epoch": 8, "memory": 6318, "step": 31177} +{"lr": 2.2120240724331546e-05, "data_time": 0.2441396951675415, "loss": 0.21618660241365434, "time": 0.6827727079391479, "epoch": 8, "memory": 6318, "step": 31277} +{"lr": 2.1886418091267445e-05, "data_time": 0.05718188285827637, "loss": 0.21654512733221054, "time": 0.717699670791626, "epoch": 8, "memory": 6318, "step": 31377} +{"lr": 2.1654529246107496e-05, "data_time": 0.4508920192718506, "loss": 0.2097909063100815, "time": 0.8126692056655884, "epoch": 8, "memory": 6318, "step": 31477} +{"lr": 2.1424587730843184e-05, "data_time": 0.3818422794342041, "loss": 0.20350787043571472, "time": 0.785672378540039, "epoch": 8, "memory": 6318, "step": 31577} +{"lr": 2.1196606973744593e-05, "data_time": 0.6131561517715454, "loss": 0.22217413336038588, "time": 0.7535250425338745, "epoch": 8, "memory": 6318, "step": 31677} +{"lr": 2.0970600288576148e-05, "data_time": 0.5930640935897827, "loss": 0.21378509551286698, "time": 0.7328116655349731, "epoch": 8, "memory": 6318, "step": 31777} +{"lr": 2.0746580873819274e-05, "data_time": 0.6189353227615356, "loss": 0.21270658373832702, "time": 0.7587697982788086, "epoch": 8, "memory": 6318, "step": 31877} +{"lr": 2.0524561811901426e-05, "data_time": 0.0010198354721069336, "loss": 0.20548191368579866, "time": 0.8381837368011474, "epoch": 8, "memory": 6318, "step": 31977} +{"lr": 2.0304556068432268e-05, "data_time": 0.0009754180908203125, "loss": 0.20180732905864715, "time": 0.7376081228256226, "epoch": 8, "memory": 6318, "step": 32077} +{"lr": 2.0086576491446333e-05, "data_time": 0.0010756254196166992, "loss": 0.20802976489067077, "time": 0.7883803606033325, "epoch": 8, "memory": 6318, "step": 32177} diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_043015/vis_data/config.py b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_043015/vis_data/config.py new file mode 100644 index 0000000000000000000000000000000000000000..b35ee250106aa84d1065073f108965cfa202c34d --- /dev/null +++ b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_043015/vis_data/config.py @@ -0,0 +1,366 @@ +optim_wrapper = dict( + optimizer=dict( + type='SGD', + lr=0.0001, + momentum=0.9, + weight_decay=0.0001, + _scope_='mmpretrain'), + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=False, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=512) +model = dict( + type='ImageClassifier', + backbone=dict( + frozen_stages=2, + type='ResNet', + depth=50, + num_stages=4, + out_indices=(3, ), + style='pytorch', + init_cfg=dict( + type='Pretrained', + checkpoint= + 'https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth', + prefix='backbone')), + neck=dict(type='GlobalAveragePooling'), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=2048, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + topk=1)) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/all1.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/all2.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/all3.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3fake8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/cc1m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3real8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV2-1-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/if-dpmsolver++-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = dict(type='Accuracy', topk=1) +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV2-1-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/if-dpmsolver++-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = dict(type='Accuracy', topk=1) +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1' diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_043015/vis_data/events.out.tfevents.1685910631.SH-IDC1-10-140-24-125.193454.0 b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_043015/vis_data/events.out.tfevents.1685910631.SH-IDC1-10-140-24-125.193454.0 new file mode 100644 index 0000000000000000000000000000000000000000..6b5943f491943914d60af678dac8508c75e36055 --- /dev/null +++ b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_043015/vis_data/events.out.tfevents.1685910631.SH-IDC1-10-140-24-125.193454.0 @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:03a0836075eac23de528fe96e50f31bb0eaa69db85061a96005301f2c15d55b7 +size 3927908 diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_043015/vis_data/scalars.json b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_043015/vis_data/scalars.json new file mode 100644 index 0000000000000000000000000000000000000000..81fcbfd8e34ca0334bac8e28d84a7610d675daeb --- /dev/null +++ b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_043015/vis_data/scalars.json @@ -0,0 +1,328 @@ +{"lr": 9.999871217862936e-05, "data_time": 0.5833782434463501, "loss": 0.6457757472991943, "time": 0.7236799240112305, "epoch": 1, "memory": 9436, "step": 100} +{"lr": 9.999479662539605e-05, "data_time": 0.7668869018554687, "loss": 0.6040080726146698, "time": 0.9103182554244995, "epoch": 1, "memory": 6319, "step": 200} +{"lr": 9.99882534375654e-05, "data_time": 0.5824718236923218, "loss": 0.5769660413265228, "time": 0.7236875057220459, "epoch": 1, "memory": 6319, "step": 300} +{"lr": 9.997908299725075e-05, "data_time": 0.5609920263290405, "loss": 0.538285905122757, "time": 0.7017717838287354, "epoch": 1, "memory": 6319, "step": 400} +{"lr": 9.996728583999318e-05, "data_time": 0.5731720924377441, "loss": 0.5249818980693817, "time": 0.7144674062728882, "epoch": 1, "memory": 6319, "step": 500} +{"lr": 9.995286265473058e-05, "data_time": 0.6639786243438721, "loss": 0.5100289076566696, "time": 0.8061728477478027, "epoch": 1, "memory": 6319, "step": 600} +{"lr": 9.993581428375725e-05, "data_time": 0.7929261684417724, "loss": 0.46335644125938413, "time": 0.9325023651123047, "epoch": 1, "memory": 6319, "step": 700} +{"lr": 9.991614172267466e-05, "data_time": 0.582516360282898, "loss": 0.4537742078304291, "time": 0.7229749202728272, "epoch": 1, "memory": 6319, "step": 800} +{"lr": 9.989384612033355e-05, "data_time": 0.6241274118423462, "loss": 0.44828479588031767, "time": 0.7643932342529297, "epoch": 1, "memory": 6319, "step": 900} +{"lr": 9.986892877876658e-05, "data_time": 0.6361027956008911, "loss": 0.4394378662109375, "time": 0.7761248111724853, "epoch": 1, "memory": 6319, "step": 1000} +{"lr": 9.984139115311244e-05, "data_time": 0.626507306098938, "loss": 0.44652384519577026, "time": 0.7666589021682739, "epoch": 1, "memory": 6319, "step": 1100} +{"lr": 9.981123485153081e-05, "data_time": 0.5829660177230835, "loss": 0.41557493805885315, "time": 0.7230542421340942, "epoch": 1, "memory": 6319, "step": 1200} +{"lr": 9.977846163510851e-05, "data_time": 0.6212214946746826, "loss": 0.4050597667694092, "time": 0.7630456447601318, "epoch": 1, "memory": 6319, "step": 1300} +{"lr": 9.974307341775653e-05, "data_time": 0.5404500246047974, "loss": 0.39317560791969297, "time": 0.6810670614242553, "epoch": 1, "memory": 6319, "step": 1400} +{"lr": 9.970507226609833e-05, "data_time": 0.6263991117477417, "loss": 0.37779472172260287, "time": 0.7653239011764527, "epoch": 1, "memory": 6319, "step": 1500} +{"lr": 9.966446039934934e-05, "data_time": 0.5814517498016357, "loss": 0.3829806953668594, "time": 0.7212536334991455, "epoch": 1, "memory": 6319, "step": 1600} +{"lr": 9.962124018918683e-05, "data_time": 0.5648025751113892, "loss": 0.3914806544780731, "time": 0.7056179523468018, "epoch": 1, "memory": 6319, "step": 1700} +{"lr": 9.957541415961212e-05, "data_time": 0.6145240545272828, "loss": 0.3696164727210999, "time": 0.7550813674926757, "epoch": 1, "memory": 6319, "step": 1800} +{"lr": 9.952698498680267e-05, "data_time": 0.5710381269454956, "loss": 0.37971139550209043, "time": 0.7102778673171997, "epoch": 1, "memory": 6319, "step": 1900} +{"lr": 9.947595549895583e-05, "data_time": 0.6139296770095826, "loss": 0.3580131262540817, "time": 0.7543619871139526, "epoch": 1, "memory": 6319, "step": 2000} +{"lr": 9.942232867612417e-05, "data_time": 0.5774040699005127, "loss": 0.3667486310005188, "time": 0.7178636074066163, "epoch": 1, "memory": 6319, "step": 2100} +{"lr": 9.93661076500406e-05, "data_time": 0.5804235696792602, "loss": 0.35057703852653505, "time": 0.7219144105911255, "epoch": 1, "memory": 6319, "step": 2200} +{"lr": 9.930729570393622e-05, "data_time": 0.5540996074676514, "loss": 0.3724305748939514, "time": 0.6951204061508178, "epoch": 1, "memory": 6319, "step": 2300} +{"lr": 9.924589627234824e-05, "data_time": 0.6224174976348877, "loss": 0.35138237476348877, "time": 0.7629217624664306, "epoch": 1, "memory": 6319, "step": 2400} +{"lr": 9.918191294091956e-05, "data_time": 0.5802772521972657, "loss": 0.36304775178432463, "time": 0.7212441444396973, "epoch": 1, "memory": 6319, "step": 2500} +{"lr": 9.91153494461892e-05, "data_time": 0.5826991081237793, "loss": 0.3589605003595352, "time": 0.7224309206008911, "epoch": 1, "memory": 6319, "step": 2600} +{"lr": 9.904620967537422e-05, "data_time": 0.6027488946914673, "loss": 0.33957523107528687, "time": 0.7427972793579102, "epoch": 1, "memory": 6319, "step": 2700} +{"lr": 9.897449766614277e-05, "data_time": 0.6042295217514038, "loss": 0.3580081135034561, "time": 0.7444121360778808, "epoch": 1, "memory": 6319, "step": 2800} +{"lr": 9.890021760637804e-05, "data_time": 0.5304379224777221, "loss": 0.341555318236351, "time": 0.6727274417877197, "epoch": 1, "memory": 6319, "step": 2900} +{"lr": 9.882337383393409e-05, "data_time": 0.5570443153381348, "loss": 0.34094529151916503, "time": 0.6958114862442016, "epoch": 1, "memory": 6319, "step": 3000} +{"lr": 9.874397083638193e-05, "data_time": 0.6410630941390991, "loss": 0.3343875378370285, "time": 0.7816831827163696, "epoch": 1, "memory": 6319, "step": 3100} +{"lr": 9.866201325074828e-05, "data_time": 0.632978367805481, "loss": 0.332791805267334, "time": 0.772688364982605, "epoch": 1, "memory": 6319, "step": 3200} +{"lr": 9.857750586324393e-05, "data_time": 0.6097943782806396, "loss": 0.34595970809459686, "time": 0.7490038394927978, "epoch": 1, "memory": 6319, "step": 3300} +{"lr": 9.84904536089848e-05, "data_time": 0.5731130599975586, "loss": 0.31871800422668456, "time": 0.7145023107528686, "epoch": 1, "memory": 6319, "step": 3400} +{"lr": 9.840086157170351e-05, "data_time": 0.5743430137634278, "loss": 0.33876071572303773, "time": 0.7148260831832886, "epoch": 1, "memory": 6319, "step": 3500} +{"lr": 9.830873498345255e-05, "data_time": 0.5523132085800171, "loss": 0.3251511067152023, "time": 0.691866397857666, "epoch": 1, "memory": 6319, "step": 3600} +{"lr": 9.821407922429864e-05, "data_time": 0.6264314889907837, "loss": 0.332692089676857, "time": 0.7661503314971924, "epoch": 1, "memory": 6319, "step": 3700} +{"lr": 9.811689982200877e-05, "data_time": 0.6108566045761108, "loss": 0.3214162290096283, "time": 0.7660102844238281, "epoch": 1, "memory": 6319, "step": 3800} +{"lr": 9.801720245172702e-05, "data_time": 0.5971901655197144, "loss": 0.31916169822216034, "time": 0.7374595165252685, "epoch": 1, "memory": 6319, "step": 3900} +{"lr": 9.791499293564358e-05, "data_time": 0.6052390336990356, "loss": 0.3148622393608093, "time": 0.7446340322494507, "epoch": 1, "memory": 6319, "step": 4000} +{"lr": 9.781027724265439e-05, "data_time": 0.6188958644866943, "loss": 0.32222879827022555, "time": 0.7586504220962524, "epoch": 1, "memory": 6319, "step": 4100} +{"accuracy/top1": 82.1375961303711, "data_time": 0.3992989824599579, "time": 0.48873059088442505, "step": 1} +{"lr": 9.769111539369434e-05, "data_time": 0.5902113676071167, "loss": 0.32364955842494963, "time": 0.7324928283691406, "epoch": 2, "memory": 6318, "step": 4211} +{"lr": 9.758113190911623e-05, "data_time": 0.6220621347427369, "loss": 0.32121524810791013, "time": 0.762659215927124, "epoch": 2, "memory": 6318, "step": 4311} +{"lr": 9.746866174465984e-05, "data_time": 0.5857436656951904, "loss": 0.33113714158535, "time": 0.7273895263671875, "epoch": 2, "memory": 6318, "step": 4411} +{"lr": 9.735371146842894e-05, "data_time": 0.6168334245681762, "loss": 0.30353142619132994, "time": 0.7569495916366578, "epoch": 2, "memory": 6318, "step": 4511} +{"lr": 9.723628779336255e-05, "data_time": 0.5955076694488526, "loss": 0.3121369868516922, "time": 0.7367850542068481, "epoch": 2, "memory": 6318, "step": 4611} +{"lr": 9.711639757684265e-05, "data_time": 0.5795055389404297, "loss": 0.32282961905002594, "time": 0.7184475183486938, "epoch": 2, "memory": 6318, "step": 4711} +{"lr": 9.699404782029395e-05, "data_time": 0.6580286741256713, "loss": 0.31532838344573977, "time": 0.7986724853515625, "epoch": 2, "memory": 6318, "step": 4811} +{"lr": 9.686924566877498e-05, "data_time": 0.5989612817764283, "loss": 0.3125740885734558, "time": 0.7385095357894897, "epoch": 2, "memory": 6318, "step": 4911} +{"lr": 9.674199841056078e-05, "data_time": 0.6176668643951416, "loss": 0.30549986362457277, "time": 0.7581380844116211, "epoch": 2, "memory": 6318, "step": 5011} +{"lr": 9.661231347671705e-05, "data_time": 0.6323682785034179, "loss": 0.3196987360715866, "time": 0.7712761640548706, "epoch": 2, "memory": 6318, "step": 5111} +{"lr": 9.64801984406668e-05, "data_time": 0.5423715591430665, "loss": 0.31711987853050233, "time": 0.6824635982513427, "epoch": 2, "memory": 6318, "step": 5211} +{"lr": 9.634566101774737e-05, "data_time": 0.5807841539382934, "loss": 0.3128260254859924, "time": 0.7204586029052734, "epoch": 2, "memory": 6318, "step": 5311} +{"lr": 9.62087090647604e-05, "data_time": 0.5787700414657593, "loss": 0.31155178844928744, "time": 0.7196438789367676, "epoch": 2, "memory": 6318, "step": 5411} +{"lr": 9.606935057951247e-05, "data_time": 0.568332028388977, "loss": 0.3232632249593735, "time": 0.7109239339828491, "epoch": 2, "memory": 6318, "step": 5511} +{"lr": 9.592759370034884e-05, "data_time": 0.638883638381958, "loss": 0.3069745123386383, "time": 0.7785221099853515, "epoch": 2, "memory": 6318, "step": 5611} +{"lr": 9.578344670567721e-05, "data_time": 0.6136923789978027, "loss": 0.29695713222026826, "time": 0.7593366622924804, "epoch": 2, "memory": 6318, "step": 5711} +{"lr": 9.563691801348495e-05, "data_time": 0.5899014711380005, "loss": 0.2924453437328339, "time": 0.7304668188095093, "epoch": 2, "memory": 6318, "step": 5811} +{"lr": 9.548801618084738e-05, "data_time": 0.6219239234924316, "loss": 0.30155878365039823, "time": 0.7622530937194825, "epoch": 2, "memory": 6318, "step": 5911} +{"lr": 9.533674990342786e-05, "data_time": 0.6147721529006958, "loss": 0.30114450454711916, "time": 0.7545836210250855, "epoch": 2, "memory": 6318, "step": 6011} +{"lr": 9.518312801497029e-05, "data_time": 0.5771173715591431, "loss": 0.28212161362171173, "time": 0.7168561935424804, "epoch": 2, "memory": 6318, "step": 6111} +{"lr": 9.502715948678276e-05, "data_time": 0.5909449100494385, "loss": 0.2921979516744614, "time": 0.7317031383514404, "epoch": 2, "memory": 6318, "step": 6211} +{"lr": 9.486885342721427e-05, "data_time": 0.622680401802063, "loss": 0.3059606194496155, "time": 0.7632664203643799, "epoch": 2, "memory": 6318, "step": 6311} +{"lr": 9.470821908112202e-05, "data_time": 0.5569717168807984, "loss": 0.2895622730255127, "time": 0.6975443840026856, "epoch": 2, "memory": 6318, "step": 6411} +{"lr": 9.454526582933214e-05, "data_time": 0.6190478563308716, "loss": 0.2918767884373665, "time": 0.7586658477783204, "epoch": 2, "memory": 6318, "step": 6511} +{"lr": 9.438000318809176e-05, "data_time": 0.5510854959487915, "loss": 0.2964482635259628, "time": 0.6919265508651733, "epoch": 2, "memory": 6318, "step": 6611} +{"lr": 9.42124408085132e-05, "data_time": 0.599868655204773, "loss": 0.3183589786291122, "time": 0.7403755426406861, "epoch": 2, "memory": 6318, "step": 6711} +{"lr": 9.404258847601036e-05, "data_time": 0.5877859592437744, "loss": 0.2691179126501083, "time": 0.7272505521774292, "epoch": 2, "memory": 6318, "step": 6811} +{"lr": 9.387045610972717e-05, "data_time": 0.5716188192367554, "loss": 0.2867441624403, "time": 0.7112274646759034, "epoch": 2, "memory": 6318, "step": 6911} +{"lr": 9.369605376195838e-05, "data_time": 0.6654116392135621, "loss": 0.26854095458984373, "time": 0.8053451776504517, "epoch": 2, "memory": 6318, "step": 7011} +{"lr": 9.351939161756269e-05, "data_time": 0.53777174949646, "loss": 0.2746407687664032, "time": 0.6781685829162598, "epoch": 2, "memory": 6318, "step": 7111} +{"lr": 9.334047999336764e-05, "data_time": 0.6016851663589478, "loss": 0.26764721125364305, "time": 0.7421585321426392, "epoch": 2, "memory": 6318, "step": 7211} +{"lr": 9.315932933756752e-05, "data_time": 0.643176007270813, "loss": 0.27293182611465455, "time": 0.7826913833618164, "epoch": 2, "memory": 6318, "step": 7311} +{"lr": 9.297595022911287e-05, "data_time": 0.5789018630981445, "loss": 0.2947516679763794, "time": 0.718891716003418, "epoch": 2, "memory": 6318, "step": 7411} +{"lr": 9.279035337709295e-05, "data_time": 0.7300005674362182, "loss": 0.2894325017929077, "time": 0.8703007698059082, "epoch": 2, "memory": 6318, "step": 7511} +{"lr": 9.260254962011016e-05, "data_time": 0.5967065334320069, "loss": 0.2779200911521912, "time": 0.7461466550827026, "epoch": 2, "memory": 6318, "step": 7611} +{"lr": 9.241254992564702e-05, "data_time": 0.6363636493682862, "loss": 0.28503446877002714, "time": 0.7983319282531738, "epoch": 2, "memory": 6318, "step": 7711} +{"lr": 9.22203653894259e-05, "data_time": 0.614773178100586, "loss": 0.2754530429840088, "time": 0.7547361135482789, "epoch": 2, "memory": 6318, "step": 7811} +{"lr": 9.202600723476098e-05, "data_time": 0.6346525192260742, "loss": 0.27611604928970335, "time": 0.7754125833511353, "epoch": 2, "memory": 6318, "step": 7911} +{"lr": 9.182948681190268e-05, "data_time": 0.5980861186981201, "loss": 0.27201849818229673, "time": 0.7377133131027221, "epoch": 2, "memory": 6318, "step": 8011} +{"lr": 9.163081559737497e-05, "data_time": 0.5946156263351441, "loss": 0.28144574016332624, "time": 0.7340287446975708, "epoch": 2, "memory": 6318, "step": 8111} +{"lr": 9.143000519330517e-05, "data_time": 0.5590303421020508, "loss": 0.2801020860671997, "time": 0.6994771003723145, "epoch": 2, "memory": 6318, "step": 8211} +{"accuracy/top1": 85.16458892822266, "data_time": 0.38208451072374977, "time": 0.4727638363838196, "step": 2} +{"lr": 9.120461478753376e-05, "data_time": 0.5781090021133423, "loss": 0.28358772546052935, "time": 0.7174955606460571, "epoch": 3, "memory": 6318, "step": 8322} +{"lr": 9.099932932138134e-05, "data_time": 0.5374660730361939, "loss": 0.2692747175693512, "time": 0.6775541782379151, "epoch": 3, "memory": 6318, "step": 8422} +{"lr": 9.079194154362181e-05, "data_time": 0.7943992853164673, "loss": 0.29565121084451673, "time": 0.9346028327941894, "epoch": 3, "memory": 6318, "step": 8522} +{"lr": 9.058246356541787e-05, "data_time": 0.6330480098724365, "loss": 0.2778221368789673, "time": 0.7732642650604248, "epoch": 3, "memory": 6318, "step": 8622} +{"lr": 9.037090761999699e-05, "data_time": 0.6890178442001342, "loss": 0.27614607810974123, "time": 0.8287760257720947, "epoch": 3, "memory": 6318, "step": 8722} +{"lr": 9.015728606193701e-05, "data_time": 0.6631336450576782, "loss": 0.290597602725029, "time": 0.8027716159820557, "epoch": 3, "memory": 6318, "step": 8822} +{"lr": 8.994161136644488e-05, "data_time": 0.5741037845611572, "loss": 0.3002932250499725, "time": 0.7148855209350586, "epoch": 3, "memory": 6318, "step": 8922} +{"lr": 8.972389612862792e-05, "data_time": 0.6331704139709473, "loss": 0.26597594022750853, "time": 0.774188494682312, "epoch": 3, "memory": 6318, "step": 9022} +{"lr": 8.950415306275833e-05, "data_time": 0.6183555603027344, "loss": 0.26705462485551834, "time": 0.7672089099884033, "epoch": 3, "memory": 6318, "step": 9122} +{"lr": 8.928239500153059e-05, "data_time": 0.5922861576080323, "loss": 0.2828182190656662, "time": 0.7321900129318237, "epoch": 3, "memory": 6318, "step": 9222} +{"lr": 8.905863489531232e-05, "data_time": 0.6252435684204102, "loss": 0.26145609468221664, "time": 0.7666311740875245, "epoch": 3, "memory": 6318, "step": 9322} +{"lr": 8.883288581138775e-05, "data_time": 0.5713315486907959, "loss": 0.27831234633922575, "time": 0.7139128684997559, "epoch": 3, "memory": 6318, "step": 9422} +{"lr": 8.86051609331947e-05, "data_time": 0.5196505069732666, "loss": 0.2668722853064537, "time": 0.6605884552001953, "epoch": 3, "memory": 6318, "step": 9522} +{"lr": 8.837547355955467e-05, "data_time": 0.6247809886932373, "loss": 0.2651767820119858, "time": 0.7680954933166504, "epoch": 3, "memory": 6318, "step": 9622} +{"lr": 8.814383710389614e-05, "data_time": 0.6110742330551148, "loss": 0.2882920101284981, "time": 0.7508492708206177, "epoch": 3, "memory": 6318, "step": 9722} +{"lr": 8.791026509347146e-05, "data_time": 0.6371283054351806, "loss": 0.25376144647598264, "time": 0.7788569927215576, "epoch": 3, "memory": 6318, "step": 9822} +{"lr": 8.767477116856659e-05, "data_time": 0.6164314270019531, "loss": 0.2567665338516235, "time": 0.7568058252334595, "epoch": 3, "memory": 6318, "step": 9922} +{"lr": 8.743736908170466e-05, "data_time": 0.5361625671386718, "loss": 0.29451175779104233, "time": 0.6756283044815063, "epoch": 3, "memory": 6318, "step": 10022} +{"lr": 8.719807269684305e-05, "data_time": 0.6020381212234497, "loss": 0.25140334069728854, "time": 0.7426766395568848, "epoch": 3, "memory": 6318, "step": 10122} +{"lr": 8.695689598856356e-05, "data_time": 0.6081812381744385, "loss": 0.2641803830862045, "time": 0.749635910987854, "epoch": 3, "memory": 6318, "step": 10222} +{"lr": 8.67138530412559e-05, "data_time": 0.6089127063751221, "loss": 0.2692801162600517, "time": 0.7482808113098145, "epoch": 3, "memory": 6318, "step": 10322} +{"lr": 8.646895804829604e-05, "data_time": 0.5897334575653076, "loss": 0.27401638180017474, "time": 0.7361309289932251, "epoch": 3, "memory": 6318, "step": 10422} +{"lr": 8.622222531121664e-05, "data_time": 0.6014791488647461, "loss": 0.2598450928926468, "time": 0.743660569190979, "epoch": 3, "memory": 6318, "step": 10522} +{"lr": 8.597366923887211e-05, "data_time": 0.6019439935684204, "loss": 0.2573289662599564, "time": 0.742353081703186, "epoch": 3, "memory": 6318, "step": 10622} +{"lr": 8.57233043465972e-05, "data_time": 0.5541431665420532, "loss": 0.26331354677677155, "time": 0.6946407318115234, "epoch": 3, "memory": 6318, "step": 10722} +{"lr": 8.547114525535937e-05, "data_time": 0.5862284421920776, "loss": 0.248159721493721, "time": 0.7373787641525269, "epoch": 3, "memory": 6318, "step": 10822} +{"lr": 8.521720669090474e-05, "data_time": 0.5396308183670044, "loss": 0.28025137782096865, "time": 0.6801734447479248, "epoch": 3, "memory": 6318, "step": 10922} +{"lr": 8.496150348289821e-05, "data_time": 0.5594420671463013, "loss": 0.2621003046631813, "time": 0.7176491022109985, "epoch": 3, "memory": 6318, "step": 11022} +{"lr": 8.470405056405736e-05, "data_time": 0.5735392570495605, "loss": 0.256819885969162, "time": 0.7138025283813476, "epoch": 3, "memory": 6318, "step": 11122} +{"lr": 8.44448629692807e-05, "data_time": 0.6116055965423584, "loss": 0.24901504814624786, "time": 0.7513501167297363, "epoch": 3, "memory": 6318, "step": 11222} +{"lr": 8.418395583476941e-05, "data_time": 0.5562048435211182, "loss": 0.26933703422546384, "time": 0.6964932203292846, "epoch": 3, "memory": 6318, "step": 11322} +{"lr": 8.392134439714336e-05, "data_time": 0.5575484991073608, "loss": 0.24063725173473358, "time": 0.7036512851715088, "epoch": 3, "memory": 6318, "step": 11422} +{"lr": 8.365704399255133e-05, "data_time": 0.5933792114257812, "loss": 0.2415594533085823, "time": 0.73389732837677, "epoch": 3, "memory": 6318, "step": 11522} +{"lr": 8.339107005577562e-05, "data_time": 0.5866213798522949, "loss": 0.2673858031630516, "time": 0.727308201789856, "epoch": 3, "memory": 6318, "step": 11622} +{"lr": 8.31234381193305e-05, "data_time": 0.613468599319458, "loss": 0.25906232595443723, "time": 0.7547986507415771, "epoch": 3, "memory": 6318, "step": 11722} +{"lr": 8.285416381255504e-05, "data_time": 0.6322313070297241, "loss": 0.24322355687618255, "time": 0.7723243474960327, "epoch": 3, "memory": 6318, "step": 11822} +{"lr": 8.258326286070047e-05, "data_time": 0.6222434282302857, "loss": 0.2620311677455902, "time": 0.7631978988647461, "epoch": 3, "memory": 6318, "step": 11922} +{"lr": 8.231075108401202e-05, "data_time": 0.5978820323944092, "loss": 0.2560994476079941, "time": 0.7383124113082886, "epoch": 3, "memory": 6318, "step": 12022} +{"lr": 8.203664439680493e-05, "data_time": 0.6929920196533204, "loss": 0.2536120295524597, "time": 0.833314061164856, "epoch": 3, "memory": 6318, "step": 12122} +{"lr": 8.176095880653475e-05, "data_time": 0.6269451141357422, "loss": 0.2562736988067627, "time": 0.7674845933914185, "epoch": 3, "memory": 6318, "step": 12222} +{"lr": 8.148371041286308e-05, "data_time": 0.5955604076385498, "loss": 0.25193780809640887, "time": 0.7358064889907837, "epoch": 3, "memory": 6318, "step": 12322} +{"accuracy/top1": 83.95809936523438, "data_time": 0.38941847284634906, "time": 0.4791889111200968, "step": 3} +{"lr": 8.117415423358443e-05, "data_time": 0.6537652254104614, "loss": 0.23664241433143615, "time": 0.795924186706543, "epoch": 4, "memory": 6318, "step": 12433} +{"lr": 8.089366155759266e-05, "data_time": 0.6240051031112671, "loss": 0.24171775281429292, "time": 0.7717869997024536, "epoch": 4, "memory": 6318, "step": 12533} +{"lr": 8.061165672717273e-05, "data_time": 0.5950781583786011, "loss": 0.2567700624465942, "time": 0.7352271318435669, "epoch": 4, "memory": 6318, "step": 12633} +{"lr": 8.032815621102116e-05, "data_time": 0.29046666622161865, "loss": 0.2457823485136032, "time": 0.7187775373458862, "epoch": 4, "memory": 6318, "step": 12733} +{"lr": 8.004317656518042e-05, "data_time": 0.4857169151306152, "loss": 0.25299454629421236, "time": 0.7400350570678711, "epoch": 4, "memory": 6318, "step": 12833} +{"lr": 7.975673443207223e-05, "data_time": 0.3014135122299194, "loss": 0.248407806456089, "time": 0.7418803930282593, "epoch": 4, "memory": 6318, "step": 12933} +{"lr": 7.946884653952537e-05, "data_time": 0.6789196014404297, "loss": 0.23907822072505952, "time": 0.8201666355133057, "epoch": 4, "memory": 6318, "step": 13033} +{"lr": 7.91795296997992e-05, "data_time": 0.6203129768371582, "loss": 0.2586119264364243, "time": 0.763839316368103, "epoch": 4, "memory": 6318, "step": 13133} +{"lr": 7.888880080860151e-05, "data_time": 0.5903083562850953, "loss": 0.2584664598107338, "time": 0.7317310810089112, "epoch": 4, "memory": 6318, "step": 13233} +{"lr": 7.859667684410205e-05, "data_time": 0.6058751344680786, "loss": 0.22599205672740935, "time": 0.7473642826080322, "epoch": 4, "memory": 6318, "step": 13333} +{"lr": 7.830317486594082e-05, "data_time": 0.6215017795562744, "loss": 0.25034799575805666, "time": 0.761247205734253, "epoch": 4, "memory": 6318, "step": 13433} +{"lr": 7.800831201423207e-05, "data_time": 0.560490345954895, "loss": 0.25423461496829985, "time": 0.700891661643982, "epoch": 4, "memory": 6318, "step": 13533} +{"lr": 7.771210550856309e-05, "data_time": 0.6174039125442505, "loss": 0.2564885541796684, "time": 0.7578014135360718, "epoch": 4, "memory": 6318, "step": 13633} +{"lr": 7.741457264698884e-05, "data_time": 0.7111117839813232, "loss": 0.2498186483979225, "time": 0.852037763595581, "epoch": 4, "memory": 6318, "step": 13733} +{"lr": 7.711573080502147e-05, "data_time": 0.5510381937026978, "loss": 0.25023491531610487, "time": 0.6926404476165772, "epoch": 4, "memory": 6318, "step": 13833} +{"lr": 7.681559743461609e-05, "data_time": 0.5712744951248169, "loss": 0.2362610563635826, "time": 0.7112491607666016, "epoch": 4, "memory": 6318, "step": 13933} +{"lr": 7.651419006315099e-05, "data_time": 0.6209005355834961, "loss": 0.254113607108593, "time": 0.7617431163787842, "epoch": 4, "memory": 6318, "step": 14033} +{"lr": 7.621152629240453e-05, "data_time": 0.6097277164459228, "loss": 0.24347356259822844, "time": 0.7496002197265625, "epoch": 4, "memory": 6318, "step": 14133} +{"lr": 7.590762379752719e-05, "data_time": 0.6082467794418335, "loss": 0.2417900651693344, "time": 0.7484559774398803, "epoch": 4, "memory": 6318, "step": 14233} +{"lr": 7.560250032600911e-05, "data_time": 0.6105065822601319, "loss": 0.22725880444049834, "time": 0.751008152961731, "epoch": 4, "memory": 6318, "step": 14333} +{"lr": 7.52961736966437e-05, "data_time": 0.5979675054550171, "loss": 0.23678749948740005, "time": 0.7382069826126099, "epoch": 4, "memory": 6318, "step": 14433} +{"lr": 7.498866179848732e-05, "data_time": 0.602944278717041, "loss": 0.25134188383817674, "time": 0.745530652999878, "epoch": 4, "memory": 6318, "step": 14533} +{"lr": 7.467998258981415e-05, "data_time": 0.6060527324676513, "loss": 0.24453014135360718, "time": 0.7461367607116699, "epoch": 4, "memory": 6318, "step": 14633} +{"lr": 7.437015409706804e-05, "data_time": 0.5987406969070435, "loss": 0.244582162797451, "time": 0.7388062715530396, "epoch": 4, "memory": 6318, "step": 14733} +{"lr": 7.405919441380929e-05, "data_time": 0.5786459445953369, "loss": 0.24392749071121217, "time": 0.7198333501815796, "epoch": 4, "memory": 6318, "step": 14833} +{"lr": 7.374712169965821e-05, "data_time": 0.6221743106842041, "loss": 0.25384840965270994, "time": 0.7634946346282959, "epoch": 4, "memory": 6318, "step": 14933} +{"lr": 7.343395417923461e-05, "data_time": 0.6961986541748046, "loss": 0.23877278566360474, "time": 0.8362715721130372, "epoch": 4, "memory": 6318, "step": 15033} +{"lr": 7.311971014109342e-05, "data_time": 0.6279742479324341, "loss": 0.23283368051052095, "time": 0.7792951822280884, "epoch": 4, "memory": 6318, "step": 15133} +{"lr": 7.280440793665682e-05, "data_time": 0.5790355443954468, "loss": 0.2475575938820839, "time": 0.7281638860702515, "epoch": 4, "memory": 6318, "step": 15233} +{"lr": 7.248806597914234e-05, "data_time": 0.6665335655212402, "loss": 0.23362425565719605, "time": 0.8058000802993774, "epoch": 4, "memory": 6318, "step": 15333} +{"lr": 7.217070274248765e-05, "data_time": 0.6398946046829224, "loss": 0.23349884301424026, "time": 0.7801725387573242, "epoch": 4, "memory": 6318, "step": 15433} +{"lr": 7.185233676027203e-05, "data_time": 0.6243457555770874, "loss": 0.2130759537220001, "time": 0.7681803226470947, "epoch": 4, "memory": 6318, "step": 15533} +{"lr": 7.153298662463338e-05, "data_time": 0.6000623226165771, "loss": 0.2271370217204094, "time": 0.7452205419540405, "epoch": 4, "memory": 6318, "step": 15633} +{"lr": 7.121267098518291e-05, "data_time": 0.6265114545822144, "loss": 0.27117105275392533, "time": 0.7745836496353149, "epoch": 4, "memory": 6318, "step": 15733} +{"lr": 7.089140854791618e-05, "data_time": 0.5846446514129638, "loss": 0.22592310905456542, "time": 0.7251559495925903, "epoch": 4, "memory": 6318, "step": 15833} +{"lr": 7.056921807412025e-05, "data_time": 0.6570351123809814, "loss": 0.2314138889312744, "time": 0.7984226703643799, "epoch": 4, "memory": 6318, "step": 15933} +{"lr": 7.024611837927802e-05, "data_time": 0.6338525295257569, "loss": 0.2543148994445801, "time": 0.7740651845932007, "epoch": 4, "memory": 6318, "step": 16033} +{"lr": 6.992212833196998e-05, "data_time": 0.5853900909423828, "loss": 0.23783159255981445, "time": 0.7274734020233155, "epoch": 4, "memory": 6318, "step": 16133} +{"lr": 6.959726685277174e-05, "data_time": 0.5680332660675049, "loss": 0.23654840439558028, "time": 0.7082507610321045, "epoch": 4, "memory": 6318, "step": 16233} +{"lr": 6.927155291314949e-05, "data_time": 0.6556339263916016, "loss": 0.24805202186107636, "time": 0.7951558589935303, "epoch": 4, "memory": 6318, "step": 16333} +{"lr": 6.894500553435204e-05, "data_time": 0.5759670019149781, "loss": 0.2458961620926857, "time": 0.7164654731750488, "epoch": 4, "memory": 6318, "step": 16433} +{"accuracy/top1": 88.21310424804688, "data_time": 0.36774808367093403, "time": 0.45801969766616824, "step": 4} +{"lr": 6.858158509719903e-05, "data_time": 0.634275221824646, "loss": 0.24288712292909623, "time": 0.8501934289932251, "epoch": 5, "memory": 6318, "step": 16544} +{"lr": 6.825334178880171e-05, "data_time": 0.5777467966079712, "loss": 0.24366383105516434, "time": 0.7923753023147583, "epoch": 5, "memory": 6318, "step": 16644} +{"lr": 6.792432450336645e-05, "data_time": 0.33480277061462405, "loss": 0.22490052431821822, "time": 0.7397310972213745, "epoch": 5, "memory": 6318, "step": 16744} +{"lr": 6.759455245505286e-05, "data_time": 0.2291111707687378, "loss": 0.2554459750652313, "time": 0.7925480365753174, "epoch": 5, "memory": 6318, "step": 16844} +{"lr": 6.726404490209753e-05, "data_time": 0.33718101978302, "loss": 0.22675415724515915, "time": 0.7337573766708374, "epoch": 5, "memory": 6318, "step": 16944} +{"lr": 6.693282114568972e-05, "data_time": 0.4749349594116211, "loss": 0.2320703700184822, "time": 0.763561749458313, "epoch": 5, "memory": 6318, "step": 17044} +{"lr": 6.660090052884388e-05, "data_time": 0.3626111030578613, "loss": 0.22647695541381835, "time": 0.7585147380828857, "epoch": 5, "memory": 6318, "step": 17144} +{"lr": 6.626830243527016e-05, "data_time": 0.4016925096511841, "loss": 0.23891469538211824, "time": 0.7527997016906738, "epoch": 5, "memory": 6318, "step": 17244} +{"lr": 6.593504628824236e-05, "data_time": 0.2532955646514893, "loss": 0.2269635319709778, "time": 0.7245667219161988, "epoch": 5, "memory": 6318, "step": 17344} +{"lr": 6.560115154946387e-05, "data_time": 0.4445216178894043, "loss": 0.25372825264930726, "time": 0.7384096622467041, "epoch": 5, "memory": 6318, "step": 17444} +{"lr": 6.526663771793081e-05, "data_time": 0.4799137830734253, "loss": 0.22184183895587922, "time": 0.7676207065582276, "epoch": 5, "memory": 6318, "step": 17544} +{"lr": 6.493152432879363e-05, "data_time": 0.5886412620544433, "loss": 0.23784775137901307, "time": 0.7836074590682983, "epoch": 5, "memory": 6318, "step": 17644} +{"lr": 6.459583095221596e-05, "data_time": 0.6631577491760254, "loss": 0.22672287821769715, "time": 0.8032900094985962, "epoch": 5, "memory": 6318, "step": 17744} +{"lr": 6.425957719223197e-05, "data_time": 0.5715037822723389, "loss": 0.2148836672306061, "time": 0.7109566926956177, "epoch": 5, "memory": 6318, "step": 17844} +{"lr": 6.392278268560153e-05, "data_time": 0.5947471141815186, "loss": 0.2364882528781891, "time": 0.7344706058502197, "epoch": 5, "memory": 6318, "step": 17944} +{"lr": 6.358546710066316e-05, "data_time": 0.5587163925170898, "loss": 0.2586208418011665, "time": 0.6994696617126465, "epoch": 5, "memory": 6318, "step": 18044} +{"lr": 6.324765013618573e-05, "data_time": 0.6116652488708496, "loss": 0.21596358716487885, "time": 0.7579565048217773, "epoch": 5, "memory": 6318, "step": 18144} +{"lr": 6.290935152021818e-05, "data_time": 0.7778600454330444, "loss": 0.2266959324479103, "time": 0.9177164554595947, "epoch": 5, "memory": 6318, "step": 18244} +{"lr": 6.2570591008937e-05, "data_time": 0.6113502264022828, "loss": 0.22128949016332627, "time": 0.7559655189514161, "epoch": 5, "memory": 6318, "step": 18344} +{"lr": 6.223138838549294e-05, "data_time": 0.581654691696167, "loss": 0.22663779109716414, "time": 0.721722149848938, "epoch": 5, "memory": 6318, "step": 18444} +{"lr": 6.18917634588553e-05, "data_time": 0.6068169116973877, "loss": 0.22119206190109253, "time": 0.7467026948928833, "epoch": 5, "memory": 6318, "step": 18544} +{"lr": 6.155173606265555e-05, "data_time": 0.590725040435791, "loss": 0.23121103793382644, "time": 0.7319432258605957, "epoch": 5, "memory": 6318, "step": 18644} +{"lr": 6.121132605402843e-05, "data_time": 0.664388108253479, "loss": 0.22820659428834916, "time": 0.8045989513397217, "epoch": 5, "memory": 6318, "step": 18744} +{"lr": 6.087055331245287e-05, "data_time": 0.5794222354888916, "loss": 0.22593510150909424, "time": 0.7198932886123657, "epoch": 5, "memory": 6318, "step": 18844} +{"lr": 6.05294377385911e-05, "data_time": 0.6478742361068726, "loss": 0.21822001785039902, "time": 0.7875202655792236, "epoch": 5, "memory": 6318, "step": 18944} +{"lr": 6.0187999253126355e-05, "data_time": 0.6332026243209838, "loss": 0.20956631898880004, "time": 0.7739517688751221, "epoch": 5, "memory": 6318, "step": 19044} +{"lr": 5.9846257795599195e-05, "data_time": 0.5757324695587158, "loss": 0.21944597065448762, "time": 0.715477442741394, "epoch": 5, "memory": 6318, "step": 19144} +{"lr": 5.950423332324355e-05, "data_time": 0.7819278478622437, "loss": 0.23812942504882811, "time": 0.921880578994751, "epoch": 5, "memory": 6318, "step": 19244} +{"lr": 5.9161945809820855e-05, "data_time": 0.5880257368087769, "loss": 0.22361236661672593, "time": 0.7290972948074341, "epoch": 5, "memory": 6318, "step": 19344} +{"lr": 5.8819415244453975e-05, "data_time": 0.6764802217483521, "loss": 0.2236904874444008, "time": 0.8174845457077027, "epoch": 5, "memory": 6318, "step": 19444} +{"lr": 5.8476661630459405e-05, "data_time": 0.6367762804031372, "loss": 0.23582514822483064, "time": 0.7781792402267456, "epoch": 5, "memory": 6318, "step": 19544} +{"lr": 5.81337049841796e-05, "data_time": 0.5800023555755616, "loss": 0.23952553421258926, "time": 0.720527172088623, "epoch": 5, "memory": 6318, "step": 19644} +{"lr": 5.779056533381373e-05, "data_time": 0.682162880897522, "loss": 0.2372528463602066, "time": 0.8233830451965332, "epoch": 5, "memory": 6318, "step": 19744} +{"lr": 5.744726271824817e-05, "data_time": 0.5984102964401246, "loss": 0.21181246042251586, "time": 0.7395000219345093, "epoch": 5, "memory": 6318, "step": 19844} +{"lr": 5.710381718588622e-05, "data_time": 0.6184651851654053, "loss": 0.23702518194913863, "time": 0.759309196472168, "epoch": 5, "memory": 6318, "step": 19944} +{"lr": 5.676024879347741e-05, "data_time": 0.6384994506835937, "loss": 0.2289959654211998, "time": 0.7787276268005371, "epoch": 5, "memory": 6318, "step": 20044} +{"lr": 5.6416577604945966e-05, "data_time": 0.5880339860916137, "loss": 0.24212294965982437, "time": 0.7277263641357422, "epoch": 5, "memory": 6318, "step": 20144} +{"lr": 5.6072823690219385e-05, "data_time": 0.6430895805358887, "loss": 0.22380499690771102, "time": 0.7862739324569702, "epoch": 5, "memory": 6318, "step": 20244} +{"lr": 5.572900712405624e-05, "data_time": 0.6050524234771728, "loss": 0.2059819668531418, "time": 0.74480721950531, "epoch": 5, "memory": 6318, "step": 20344} +{"lr": 5.538514798487397e-05, "data_time": 0.6456482648849488, "loss": 0.2494832068681717, "time": 0.7857419013977051, "epoch": 5, "memory": 6318, "step": 20444} +{"lr": 5.5041266353576e-05, "data_time": 0.5834356307983398, "loss": 0.23772364109754562, "time": 0.7250214099884034, "epoch": 5, "memory": 6318, "step": 20544} +{"accuracy/top1": 89.27395629882812, "data_time": 0.39095328251520794, "time": 0.48003365397453307, "step": 5} +{"lr": 5.4659555783035384e-05, "data_time": 0.016553759574890137, "loss": 0.22144537419080734, "time": 0.7474731922149658, "epoch": 6, "memory": 6318, "step": 20655} +{"lr": 5.431569258426336e-05, "data_time": 0.07853200435638427, "loss": 0.2264638990163803, "time": 0.751076626777649, "epoch": 6, "memory": 6318, "step": 20755} +{"lr": 5.397186934810864e-05, "data_time": 0.1736987352371216, "loss": 0.22422888725996018, "time": 0.7852191209793091, "epoch": 6, "memory": 6318, "step": 20855} +{"lr": 5.362810615337815e-05, "data_time": 0.201401686668396, "loss": 0.2276495799422264, "time": 0.7673514366149903, "epoch": 6, "memory": 6318, "step": 20955} +{"lr": 5.328442307537233e-05, "data_time": 0.3652447462081909, "loss": 0.2285897895693779, "time": 0.6923784732818603, "epoch": 6, "memory": 6318, "step": 21055} +{"lr": 5.294084018471303e-05, "data_time": 0.6161708116531373, "loss": 0.22080680429935456, "time": 0.7842055320739746, "epoch": 6, "memory": 6318, "step": 21155} +{"lr": 5.259737754617132e-05, "data_time": 0.6010441780090332, "loss": 0.22242555171251296, "time": 0.7420645952224731, "epoch": 6, "memory": 6318, "step": 21255} +{"lr": 5.2254055217495525e-05, "data_time": 0.5866324186325074, "loss": 0.22232354879379274, "time": 0.726702070236206, "epoch": 6, "memory": 6318, "step": 21355} +{"lr": 5.1910893248240283e-05, "data_time": 0.6091682434082031, "loss": 0.2368236780166626, "time": 0.7500319242477417, "epoch": 6, "memory": 6318, "step": 21455} +{"lr": 5.156791167859532e-05, "data_time": 0.67364022731781, "loss": 0.2161878764629364, "time": 0.8277969121932983, "epoch": 6, "memory": 6318, "step": 21555} +{"lr": 5.122513053821533e-05, "data_time": 0.9497917175292969, "loss": 0.2425651580095291, "time": 1.090404224395752, "epoch": 6, "memory": 6318, "step": 21655} +{"lr": 5.0882569845050173e-05, "data_time": 0.6629886627197266, "loss": 0.22540825456380845, "time": 0.8105605125427247, "epoch": 6, "memory": 6318, "step": 21755} +{"lr": 5.054024960417592e-05, "data_time": 0.6264513731002808, "loss": 0.22059424519538878, "time": 0.7666175365447998, "epoch": 6, "memory": 6318, "step": 21855} +{"lr": 5.0198189806626494e-05, "data_time": 0.6422939300537109, "loss": 0.2257988378405571, "time": 0.7936925411224365, "epoch": 6, "memory": 6318, "step": 21955} +{"lr": 4.985641042822647e-05, "data_time": 0.6275058031082154, "loss": 0.22144672274589539, "time": 0.7679895877838134, "epoch": 6, "memory": 6318, "step": 22055} +{"lr": 4.951493142842412e-05, "data_time": 0.5662127256393432, "loss": 0.24146845936775208, "time": 0.706992769241333, "epoch": 6, "memory": 6318, "step": 22155} +{"lr": 4.917377274912603e-05, "data_time": 0.5645421504974365, "loss": 0.21499475240707397, "time": 0.705557894706726, "epoch": 6, "memory": 6318, "step": 22255} +{"lr": 4.88329543135327e-05, "data_time": 0.6381146907806396, "loss": 0.2316078558564186, "time": 0.7787853956222535, "epoch": 6, "memory": 6318, "step": 22355} +{"lr": 4.849249602497467e-05, "data_time": 0.6599221706390381, "loss": 0.22825948745012284, "time": 0.8002774000167847, "epoch": 6, "memory": 6318, "step": 22455} +{"lr": 4.8152417765750456e-05, "data_time": 0.5728270530700683, "loss": 0.2311478853225708, "time": 0.7126530408859253, "epoch": 6, "memory": 6318, "step": 22555} +{"lr": 4.7812739395965355e-05, "data_time": 0.6469001293182373, "loss": 0.20607706904411316, "time": 0.7973691463470459, "epoch": 6, "memory": 6318, "step": 22655} +{"lr": 4.74734807523717e-05, "data_time": 0.6172947645187378, "loss": 0.2191238969564438, "time": 0.7578845500946045, "epoch": 6, "memory": 6318, "step": 22755} +{"lr": 4.7134661647210416e-05, "data_time": 0.5940561294555664, "loss": 0.21722680926322938, "time": 0.7390312671661377, "epoch": 6, "memory": 6318, "step": 22855} +{"lr": 4.679630186705388e-05, "data_time": 0.6066904544830323, "loss": 0.2189228653907776, "time": 0.7470699548721313, "epoch": 6, "memory": 6318, "step": 22955} +{"lr": 4.6458421171650644e-05, "data_time": 0.6506342411041259, "loss": 0.21992774307727814, "time": 0.7908859491348267, "epoch": 6, "memory": 6318, "step": 23055} +{"lr": 4.6121039292771295e-05, "data_time": 0.6431920528411865, "loss": 0.21492919474840164, "time": 0.784492802619934, "epoch": 6, "memory": 6318, "step": 23155} +{"lr": 4.5784175933056197e-05, "data_time": 0.6262296438217163, "loss": 0.22508245557546616, "time": 0.7773866653442383, "epoch": 6, "memory": 6318, "step": 23255} +{"lr": 4.544785076486494e-05, "data_time": 0.6410591840744019, "loss": 0.23063731491565703, "time": 0.7819517135620118, "epoch": 6, "memory": 6318, "step": 23355} +{"lr": 4.5112083429127475e-05, "data_time": 0.5785451173782349, "loss": 0.22566608637571334, "time": 0.7186836004257202, "epoch": 6, "memory": 6318, "step": 23455} +{"lr": 4.477689353419706e-05, "data_time": 0.5760425090789795, "loss": 0.22161386758089066, "time": 0.7169084072113037, "epoch": 6, "memory": 6318, "step": 23555} +{"lr": 4.4442300654705266e-05, "data_time": 0.6023649215698242, "loss": 0.2114580288529396, "time": 0.7510221242904663, "epoch": 6, "memory": 6318, "step": 23655} +{"lr": 4.410832433041866e-05, "data_time": 0.6313949584960937, "loss": 0.2094666451215744, "time": 0.7711641550064087, "epoch": 6, "memory": 6318, "step": 23755} +{"lr": 4.377498406509792e-05, "data_time": 0.6469725847244263, "loss": 0.21666625291109085, "time": 0.786846113204956, "epoch": 6, "memory": 6318, "step": 23855} +{"lr": 4.344229932535875e-05, "data_time": 0.5941392660140992, "loss": 0.2367297574877739, "time": 0.7350420951843262, "epoch": 6, "memory": 6318, "step": 23955} +{"lr": 4.3110289539535115e-05, "data_time": 0.6539195775985718, "loss": 0.21627108454704286, "time": 0.793128776550293, "epoch": 6, "memory": 6318, "step": 24055} +{"lr": 4.2778974096544436e-05, "data_time": 0.6331358194351197, "loss": 0.22466585040092468, "time": 0.7742410659790039, "epoch": 6, "memory": 6318, "step": 24155} +{"lr": 4.2448372344755544e-05, "data_time": 0.5891356706619263, "loss": 0.21508681029081345, "time": 0.7315256357192993, "epoch": 6, "memory": 6318, "step": 24255} +{"lr": 4.211850359085871e-05, "data_time": 0.66728355884552, "loss": 0.21390046775341034, "time": 0.8074127912521363, "epoch": 6, "memory": 6318, "step": 24355} +{"lr": 4.17893870987381e-05, "data_time": 0.5841787815093994, "loss": 0.22484858185052872, "time": 0.7302618503570557, "epoch": 6, "memory": 6318, "step": 24455} +{"lr": 4.146104208834679e-05, "data_time": 0.6124205589294434, "loss": 0.20944125652313234, "time": 0.7560169219970703, "epoch": 6, "memory": 6318, "step": 24555} +{"lr": 4.113348773458442e-05, "data_time": 0.5681646823883056, "loss": 0.2219816580414772, "time": 0.7095581531524658, "epoch": 6, "memory": 6318, "step": 24655} +{"accuracy/top1": 91.06466674804688, "data_time": 0.3798326849937439, "time": 0.47045179406801857, "step": 6} +{"lr": 4.0770851520835164e-05, "data_time": 0.32087929248809816, "loss": 0.21204094886779784, "time": 0.7360997200012207, "epoch": 7, "memory": 6318, "step": 24766} +{"lr": 4.044502815742039e-05, "data_time": 0.2319826602935791, "loss": 0.23103538602590562, "time": 0.7526503324508667, "epoch": 7, "memory": 6318, "step": 24866} +{"lr": 4.012005478445896e-05, "data_time": 0.19448323249816896, "loss": 0.21832422018051148, "time": 0.7050837516784668, "epoch": 7, "memory": 6318, "step": 24966} +{"lr": 3.9795950379951394e-05, "data_time": 0.49231514930725095, "loss": 0.2312387242913246, "time": 0.7812903881072998, "epoch": 7, "memory": 6318, "step": 25066} +{"lr": 3.947273387115169e-05, "data_time": 0.5172301530838013, "loss": 0.22581694275140762, "time": 0.7241688013076782, "epoch": 7, "memory": 6318, "step": 25166} +{"lr": 3.9150424133462044e-05, "data_time": 0.5714112997055054, "loss": 0.23753017634153367, "time": 0.7150854587554931, "epoch": 7, "memory": 6318, "step": 25266} +{"lr": 3.882903998933037e-05, "data_time": 0.5933114528656006, "loss": 0.23574368059635162, "time": 0.7412816047668457, "epoch": 7, "memory": 6318, "step": 25366} +{"lr": 3.8508600207151264e-05, "data_time": 0.6053260087966919, "loss": 0.21976497322320937, "time": 0.7513857364654541, "epoch": 7, "memory": 6318, "step": 25466} +{"lr": 3.818912350016978e-05, "data_time": 0.5936861753463745, "loss": 0.2058437153697014, "time": 0.7381568193435669, "epoch": 7, "memory": 6318, "step": 25566} +{"lr": 3.787062852538883e-05, "data_time": 0.5775141239166259, "loss": 0.21195100247859955, "time": 0.7230489730834961, "epoch": 7, "memory": 6318, "step": 25666} +{"lr": 3.7553133882479346e-05, "data_time": 0.5990762710571289, "loss": 0.22675100564956666, "time": 0.7402037858963013, "epoch": 7, "memory": 6318, "step": 25766} +{"lr": 3.723665811269437e-05, "data_time": 0.5240511178970337, "loss": 0.21259500086307526, "time": 0.7690648794174194, "epoch": 7, "memory": 6318, "step": 25866} +{"lr": 3.6921219697786176e-05, "data_time": 0.34241557121276855, "loss": 0.2061529576778412, "time": 0.7883437871932983, "epoch": 7, "memory": 6318, "step": 25966} +{"lr": 3.660683705892677e-05, "data_time": 0.3457735300064087, "loss": 0.20303792506456375, "time": 0.7385384321212769, "epoch": 7, "memory": 6318, "step": 26066} +{"lr": 3.6293528555632435e-05, "data_time": 0.5796358823776245, "loss": 0.23252723664045333, "time": 0.8354166269302368, "epoch": 7, "memory": 6318, "step": 26166} +{"lr": 3.598131248469134e-05, "data_time": 0.6267612934112549, "loss": 0.22580407410860062, "time": 0.766175365447998, "epoch": 7, "memory": 6318, "step": 26266} +{"lr": 3.567020707909507e-05, "data_time": 0.6017528772354126, "loss": 0.22389102578163148, "time": 0.7411997079849243, "epoch": 7, "memory": 6318, "step": 26366} +{"lr": 3.5360230506973935e-05, "data_time": 0.6216040134429932, "loss": 0.21377645581960678, "time": 0.7616827011108398, "epoch": 7, "memory": 6318, "step": 26466} +{"lr": 3.5051400870535877e-05, "data_time": 0.5958281755447388, "loss": 0.2313540667295456, "time": 0.7514890909194947, "epoch": 7, "memory": 6318, "step": 26566} +{"lr": 3.47437362050094e-05, "data_time": 0.648256254196167, "loss": 0.2143481358885765, "time": 0.7904376268386841, "epoch": 7, "memory": 6318, "step": 26666} +{"lr": 3.443725447759023e-05, "data_time": 0.6556012868881226, "loss": 0.21419825702905654, "time": 0.7967238187789917, "epoch": 7, "memory": 6318, "step": 26766} +{"lr": 3.413197358639224e-05, "data_time": 0.9352653980255127, "loss": 0.2334531933069229, "time": 1.078092384338379, "epoch": 7, "memory": 6318, "step": 26866} +{"lr": 3.382791135940194e-05, "data_time": 0.5899015903472901, "loss": 0.21842274218797683, "time": 0.7348134994506836, "epoch": 7, "memory": 6318, "step": 26966} +{"lr": 3.35250855534377e-05, "data_time": 0.6248748064041137, "loss": 0.21042381525039672, "time": 0.764406704902649, "epoch": 7, "memory": 6318, "step": 27066} +{"lr": 3.3223513853112476e-05, "data_time": 0.6145540952682496, "loss": 0.22413529455661774, "time": 0.7665135860443115, "epoch": 7, "memory": 6318, "step": 27166} +{"lr": 3.292321386980115e-05, "data_time": 0.589860463142395, "loss": 0.21802660971879959, "time": 0.7305471181869507, "epoch": 7, "memory": 6318, "step": 27266} +{"lr": 3.262420314061217e-05, "data_time": 0.691631269454956, "loss": 0.20277985632419587, "time": 0.8325157880783081, "epoch": 7, "memory": 6318, "step": 27366} +{"lr": 3.2326499127363195e-05, "data_time": 0.6181166172027588, "loss": 0.2208991304039955, "time": 0.7609508037567139, "epoch": 7, "memory": 6318, "step": 27466} +{"lr": 3.2030119215561613e-05, "data_time": 0.6322191238403321, "loss": 0.22663544714450837, "time": 0.7726822376251221, "epoch": 7, "memory": 6318, "step": 27566} +{"lr": 3.173508071338896e-05, "data_time": 0.6718494653701782, "loss": 0.22389377951622008, "time": 0.8124493360519409, "epoch": 7, "memory": 6318, "step": 27666} +{"lr": 3.144140085069037e-05, "data_time": 0.6478348255157471, "loss": 0.20451682657003403, "time": 0.7879785776138306, "epoch": 7, "memory": 6318, "step": 27766} +{"lr": 3.114909677796814e-05, "data_time": 0.6669594526290894, "loss": 0.21915597915649415, "time": 0.8066153526306152, "epoch": 7, "memory": 6318, "step": 27866} +{"lr": 3.085818556538059e-05, "data_time": 0.6285897970199585, "loss": 0.21591716706752778, "time": 0.7689493894577026, "epoch": 7, "memory": 6318, "step": 27966} +{"lr": 3.056868420174455e-05, "data_time": 0.5792436838150025, "loss": 0.21763637363910676, "time": 0.7191099643707275, "epoch": 7, "memory": 6318, "step": 28066} +{"lr": 3.028060959354392e-05, "data_time": 0.612235164642334, "loss": 0.2165424942970276, "time": 0.758737850189209, "epoch": 7, "memory": 6318, "step": 28166} +{"lr": 2.9993978563941876e-05, "data_time": 0.582972240447998, "loss": 0.23131911009550093, "time": 0.7238704919815063, "epoch": 7, "memory": 6318, "step": 28266} +{"lr": 2.9708807851798587e-05, "data_time": 0.6110140085220337, "loss": 0.2152962863445282, "time": 0.7591787338256836, "epoch": 7, "memory": 6318, "step": 28366} +{"lr": 2.9425114110693717e-05, "data_time": 0.6803679704666138, "loss": 0.20685464441776275, "time": 0.8211320877075196, "epoch": 7, "memory": 6318, "step": 28466} +{"lr": 2.9142913907953844e-05, "data_time": 0.6136783838272095, "loss": 0.22012823671102524, "time": 0.7671475172042846, "epoch": 7, "memory": 6318, "step": 28566} +{"lr": 2.8862223723684912e-05, "data_time": 0.6209859132766724, "loss": 0.22623719722032548, "time": 0.7611841440200806, "epoch": 7, "memory": 6318, "step": 28666} +{"lr": 2.8583059949809854e-05, "data_time": 0.5121319532394409, "loss": 0.23409331738948821, "time": 0.6742612600326539, "epoch": 7, "memory": 6318, "step": 28766} +{"accuracy/top1": 92.36052703857422, "data_time": 0.38193190296490986, "time": 0.4709255814552307, "step": 7} +{"lr": 2.827499545208893e-05, "data_time": 0.610418176651001, "loss": 0.22626211494207382, "time": 0.7521275043487549, "epoch": 8, "memory": 6318, "step": 28877} +{"lr": 2.7999105784987406e-05, "data_time": 0.6447580814361572, "loss": 0.21675214320421218, "time": 0.7882266283035279, "epoch": 8, "memory": 6318, "step": 28977} +{"lr": 2.772479293318711e-05, "data_time": 0.5873958587646484, "loss": 0.2133768081665039, "time": 0.7282128810882569, "epoch": 8, "memory": 6318, "step": 29077} +{"lr": 2.7452072916183603e-05, "data_time": 0.5236689805984497, "loss": 0.21600501984357834, "time": 0.7311687231063843, "epoch": 8, "memory": 6318, "step": 29177} +{"lr": 2.7180961660452883e-05, "data_time": 0.5871314764022827, "loss": 0.21424666345119475, "time": 0.7278459548950196, "epoch": 8, "memory": 6318, "step": 29277} +{"lr": 2.6911474998521622e-05, "data_time": 0.6580465078353882, "loss": 0.22579137682914735, "time": 0.7979048013687133, "epoch": 8, "memory": 6318, "step": 29377} +{"lr": 2.664362866804239e-05, "data_time": 0.546606469154358, "loss": 0.2086592972278595, "time": 0.6880997657775879, "epoch": 8, "memory": 6318, "step": 29477} +{"lr": 2.637743831087459e-05, "data_time": 0.6517598152160644, "loss": 0.2018253117799759, "time": 0.7980185985565186, "epoch": 8, "memory": 6318, "step": 29577} +{"lr": 2.6112919472171114e-05, "data_time": 0.6430217742919921, "loss": 0.2117971658706665, "time": 0.7836272239685058, "epoch": 8, "memory": 6318, "step": 29677} +{"lr": 2.5850087599470434e-05, "data_time": 0.5987901210784912, "loss": 0.20031181424856187, "time": 0.7499037027359009, "epoch": 8, "memory": 6318, "step": 29777} +{"lr": 2.5588958041794503e-05, "data_time": 0.5884083032608032, "loss": 0.196441125869751, "time": 0.7286111354827881, "epoch": 8, "memory": 6318, "step": 29877} +{"lr": 2.5329546048752453e-05, "data_time": 0.6950069904327393, "loss": 0.21808973401784898, "time": 0.8365283012390137, "epoch": 8, "memory": 6318, "step": 29977} +{"lr": 2.5071866769649806e-05, "data_time": 0.6069842338562011, "loss": 0.2107990190386772, "time": 0.7475652456283569, "epoch": 8, "memory": 6318, "step": 30077} +{"lr": 2.4815935252604222e-05, "data_time": 0.6724363565444946, "loss": 0.21933402717113495, "time": 0.8261152029037475, "epoch": 8, "memory": 6318, "step": 30177} +{"lr": 2.4561766443666384e-05, "data_time": 0.6750922918319702, "loss": 0.22104105055332185, "time": 0.815656328201294, "epoch": 8, "memory": 6318, "step": 30277} +{"lr": 2.4309375185947164e-05, "data_time": 0.6079245805740356, "loss": 0.2408744901418686, "time": 0.7474660634994507, "epoch": 8, "memory": 6318, "step": 30377} +{"lr": 2.40587762187509e-05, "data_time": 0.48343064785003664, "loss": 0.19970002323389052, "time": 0.7202982425689697, "epoch": 8, "memory": 6318, "step": 30477} +{"lr": 2.3809984176714634e-05, "data_time": 0.2909902334213257, "loss": 0.21772560924291612, "time": 0.7505961656570435, "epoch": 8, "memory": 6318, "step": 30577} +{"lr": 2.3563013588953406e-05, "data_time": 0.34513111114501954, "loss": 0.22359851002693176, "time": 0.7564613580703735, "epoch": 8, "memory": 6318, "step": 30677} +{"lr": 2.331787887821182e-05, "data_time": 0.3460818290710449, "loss": 0.23591297566890718, "time": 0.6941813468933106, "epoch": 8, "memory": 6318, "step": 30777} +{"lr": 2.307459436002179e-05, "data_time": 0.3455662727355957, "loss": 0.20404092818498612, "time": 0.7597615718841553, "epoch": 8, "memory": 6318, "step": 30877} +{"lr": 2.2833174241866527e-05, "data_time": 0.36763978004455566, "loss": 0.20681032091379165, "time": 0.8021057367324829, "epoch": 8, "memory": 6318, "step": 30977} +{"lr": 2.259363262235086e-05, "data_time": 0.1274261713027954, "loss": 0.2076197549700737, "time": 0.743987250328064, "epoch": 8, "memory": 6318, "step": 31077} +{"lr": 2.2355983490377735e-05, "data_time": 0.23604495525360109, "loss": 0.22067641466856003, "time": 0.7981895446777344, "epoch": 8, "memory": 6318, "step": 31177} +{"lr": 2.2120240724331546e-05, "data_time": 0.2441396951675415, "loss": 0.21618660241365434, "time": 0.6827727079391479, "epoch": 8, "memory": 6318, "step": 31277} +{"lr": 2.1886418091267445e-05, "data_time": 0.05718188285827637, "loss": 0.21654512733221054, "time": 0.717699670791626, "epoch": 8, "memory": 6318, "step": 31377} +{"lr": 2.1654529246107496e-05, "data_time": 0.4508920192718506, "loss": 0.2097909063100815, "time": 0.8126692056655884, "epoch": 8, "memory": 6318, "step": 31477} +{"lr": 2.1424587730843184e-05, "data_time": 0.3818422794342041, "loss": 0.20350787043571472, "time": 0.785672378540039, "epoch": 8, "memory": 6318, "step": 31577} +{"lr": 2.1196606973744593e-05, "data_time": 0.6131561517715454, "loss": 0.22217413336038588, "time": 0.7535250425338745, "epoch": 8, "memory": 6318, "step": 31677} +{"lr": 2.0970600288576148e-05, "data_time": 0.5930640935897827, "loss": 0.21378509551286698, "time": 0.7328116655349731, "epoch": 8, "memory": 6318, "step": 31777} +{"lr": 2.0746580873819274e-05, "data_time": 0.6189353227615356, "loss": 0.21270658373832702, "time": 0.7587697982788086, "epoch": 8, "memory": 6318, "step": 31877} +{"lr": 2.0524561811901426e-05, "data_time": 0.0010198354721069336, "loss": 0.20548191368579866, "time": 0.8381837368011474, "epoch": 8, "memory": 6318, "step": 31977} +{"lr": 2.0304556068432268e-05, "data_time": 0.0009754180908203125, "loss": 0.20180732905864715, "time": 0.7376081228256226, "epoch": 8, "memory": 6318, "step": 32077} +{"lr": 2.0086576491446333e-05, "data_time": 0.0010756254196166992, "loss": 0.20802976489067077, "time": 0.7883803606033325, "epoch": 8, "memory": 6318, "step": 32177} diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_043015/vis_data/vis_image/0000000.png_1.png b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_043015/vis_data/vis_image/0000000.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..9e061ab77420673d62df48ebb0814895600acdf9 Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_043015/vis_data/vis_image/0000000.png_1.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_043015/vis_data/vis_image/0000000.png_2.png b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_043015/vis_data/vis_image/0000000.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..fcffc5bc89c310ecd7c0d1f0cf5bfcabefa2c444 Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_043015/vis_data/vis_image/0000000.png_2.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_043015/vis_data/vis_image/0000000.png_3.png b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_043015/vis_data/vis_image/0000000.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..e09ada4af0f84b97905b11a3900d16dd30425312 Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_043015/vis_data/vis_image/0000000.png_3.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_043015/vis_data/vis_image/0000000.png_4.png b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_043015/vis_data/vis_image/0000000.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..3e9df76bb71545ac2c80d84020ef00f59b4e003b Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_043015/vis_data/vis_image/0000000.png_4.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_043015/vis_data/vis_image/0000000.png_5.png b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_043015/vis_data/vis_image/0000000.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..22ea03a66944016ee9366b8fa68de6dff46deb84 Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_043015/vis_data/vis_image/0000000.png_5.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_043015/vis_data/vis_image/0000000.png_6.png b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_043015/vis_data/vis_image/0000000.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..d772fa465488737da47a7e3921d04fbdf394f266 Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_043015/vis_data/vis_image/0000000.png_6.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_043015/vis_data/vis_image/0000000.png_7.png b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_043015/vis_data/vis_image/0000000.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..f66f504ba6faec5466bd16d30afc7e90cc0d9080 Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_043015/vis_data/vis_image/0000000.png_7.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_043015/vis_data/vis_image/000003093.jpg_1.png b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_043015/vis_data/vis_image/000003093.jpg_1.png new file mode 100644 index 0000000000000000000000000000000000000000..f85d346ec23c035a8a47cd60ec73dd68e01bc7ec Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_043015/vis_data/vis_image/000003093.jpg_1.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_043015/vis_data/vis_image/000003093.jpg_2.png b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_043015/vis_data/vis_image/000003093.jpg_2.png new file mode 100644 index 0000000000000000000000000000000000000000..e7040407ccb99ebdf5d5b554d3f8373097eca978 Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_043015/vis_data/vis_image/000003093.jpg_2.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_043015/vis_data/vis_image/000003093.jpg_3.png b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_043015/vis_data/vis_image/000003093.jpg_3.png new file mode 100644 index 0000000000000000000000000000000000000000..4b4f6ceef4a85b25ef55e9744845f02e21031e84 Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_043015/vis_data/vis_image/000003093.jpg_3.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_043015/vis_data/vis_image/000003093.jpg_4.png b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_043015/vis_data/vis_image/000003093.jpg_4.png new file mode 100644 index 0000000000000000000000000000000000000000..67f00c4ebd1cf92976c2ed423c45f4873f9663ca Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_043015/vis_data/vis_image/000003093.jpg_4.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_043015/vis_data/vis_image/000003093.jpg_5.png b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_043015/vis_data/vis_image/000003093.jpg_5.png new file mode 100644 index 0000000000000000000000000000000000000000..b974e32f4105909821bef4efbd67967e8e7eb9f8 Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_043015/vis_data/vis_image/000003093.jpg_5.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_043015/vis_data/vis_image/000003093.jpg_6.png b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_043015/vis_data/vis_image/000003093.jpg_6.png new file mode 100644 index 0000000000000000000000000000000000000000..24d89b478700850a8b000d07802ac06a70482fb9 Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_043015/vis_data/vis_image/000003093.jpg_6.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_043015/vis_data/vis_image/000003093.jpg_7.png b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_043015/vis_data/vis_image/000003093.jpg_7.png new file mode 100644 index 0000000000000000000000000000000000000000..5401216f38f4b9d80f4bcdf4061bff542793e635 Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_043015/vis_data/vis_image/000003093.jpg_7.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_043015/vis_data/vis_image/000015337.jpg_1.png b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_043015/vis_data/vis_image/000015337.jpg_1.png new file mode 100644 index 0000000000000000000000000000000000000000..3bf2602127c0d1147e0daf6e825280c20e9e8ab3 Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_043015/vis_data/vis_image/000015337.jpg_1.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_043015/vis_data/vis_image/000015337.jpg_2.png b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_043015/vis_data/vis_image/000015337.jpg_2.png new file mode 100644 index 0000000000000000000000000000000000000000..e07cb0e81097c88c6301db342f0819bf097ad819 Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_043015/vis_data/vis_image/000015337.jpg_2.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_043015/vis_data/vis_image/000015337.jpg_3.png b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_043015/vis_data/vis_image/000015337.jpg_3.png new file mode 100644 index 0000000000000000000000000000000000000000..00853707e7ab6aac3a33c55e7a0000fbd5e176e8 Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_043015/vis_data/vis_image/000015337.jpg_3.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_043015/vis_data/vis_image/000015337.jpg_4.png b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_043015/vis_data/vis_image/000015337.jpg_4.png new file mode 100644 index 0000000000000000000000000000000000000000..f983b0cb23e0e82498ea3a280b4540cbe23d99a1 Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_043015/vis_data/vis_image/000015337.jpg_4.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_043015/vis_data/vis_image/000015337.jpg_5.png b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_043015/vis_data/vis_image/000015337.jpg_5.png new file mode 100644 index 0000000000000000000000000000000000000000..126be3c7d8fde76a2fe360b803d7b0304b5982fc Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_043015/vis_data/vis_image/000015337.jpg_5.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_043015/vis_data/vis_image/000015337.jpg_6.png b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_043015/vis_data/vis_image/000015337.jpg_6.png new file mode 100644 index 0000000000000000000000000000000000000000..d7104d13c14ab368ef36c78e3db642d52a104c07 Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_043015/vis_data/vis_image/000015337.jpg_6.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_043015/vis_data/vis_image/000015337.jpg_7.png b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_043015/vis_data/vis_image/000015337.jpg_7.png new file mode 100644 index 0000000000000000000000000000000000000000..ee118c0598940d3d48eaceadb69f368bd0f5539f Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_043015/vis_data/vis_image/000015337.jpg_7.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_043015/vis_data/vis_image/0004159.png_1.png b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_043015/vis_data/vis_image/0004159.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..232f5032a564a990976a24faf12208df5af9f6eb Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_043015/vis_data/vis_image/0004159.png_1.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_043015/vis_data/vis_image/0004159.png_2.png b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_043015/vis_data/vis_image/0004159.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..d5cf917feab2b8057382b569ca63d7b23e48dc58 Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_043015/vis_data/vis_image/0004159.png_2.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_043015/vis_data/vis_image/0004159.png_3.png b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_043015/vis_data/vis_image/0004159.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..5903755de818b6931823a9d8398f897d16d3d28d Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_043015/vis_data/vis_image/0004159.png_3.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_043015/vis_data/vis_image/0004159.png_4.png b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_043015/vis_data/vis_image/0004159.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..854530fabb36d3805ab22bea5273953d304df533 Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_043015/vis_data/vis_image/0004159.png_4.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_043015/vis_data/vis_image/0004159.png_5.png b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_043015/vis_data/vis_image/0004159.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..5958378c265b9fa0e86bf67e38a25e5f43296505 Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_043015/vis_data/vis_image/0004159.png_5.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_043015/vis_data/vis_image/0004159.png_6.png b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_043015/vis_data/vis_image/0004159.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..111eeb8526ba2cdf415bd89ecb8098fd96b8bc3e Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_043015/vis_data/vis_image/0004159.png_6.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_043015/vis_data/vis_image/0004159.png_7.png b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_043015/vis_data/vis_image/0004159.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..f2faf5e160015d33b8303c637b2fab25d857aebb Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_043015/vis_data/vis_image/0004159.png_7.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_043015/vis_data/vis_image/0008318.png_1.png b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_043015/vis_data/vis_image/0008318.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..8a9106c5d1b6ffa0e260ba6cf394f0022409df4f Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_043015/vis_data/vis_image/0008318.png_1.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_043015/vis_data/vis_image/0008318.png_2.png b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_043015/vis_data/vis_image/0008318.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..20abdadaf8cc4f673fdc66b4c18fa337f23ad398 Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_043015/vis_data/vis_image/0008318.png_2.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_043015/vis_data/vis_image/0008318.png_3.png b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_043015/vis_data/vis_image/0008318.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..25d127c0c998a0f375394654e6d1dbce904f8952 Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_043015/vis_data/vis_image/0008318.png_3.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_043015/vis_data/vis_image/0008318.png_4.png b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_043015/vis_data/vis_image/0008318.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..ccac23bb07ebfcc99630a93897261657f276df6b Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_043015/vis_data/vis_image/0008318.png_4.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_043015/vis_data/vis_image/0008318.png_5.png b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_043015/vis_data/vis_image/0008318.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..6311556382b90296f83389a9ececc3dd82c54cbc Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_043015/vis_data/vis_image/0008318.png_5.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_043015/vis_data/vis_image/0008318.png_6.png b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_043015/vis_data/vis_image/0008318.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..ae2a588ac3e7c5dc6464d65fb738aae382f0efc8 Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_043015/vis_data/vis_image/0008318.png_6.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_043015/vis_data/vis_image/0008318.png_7.png b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_043015/vis_data/vis_image/0008318.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..69c5e5476944e747d99896aceaecba4c07d056b0 Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_043015/vis_data/vis_image/0008318.png_7.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_043015/vis_data/vis_image/0010000.png_1.png b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_043015/vis_data/vis_image/0010000.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..5a327ba71c40da6c8e70f42fad729d79a20880c8 Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_043015/vis_data/vis_image/0010000.png_1.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_043015/vis_data/vis_image/0010000.png_2.png b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_043015/vis_data/vis_image/0010000.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..00173f7cac6100981a1c1051a87e24806a0c42c9 Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_043015/vis_data/vis_image/0010000.png_2.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_043015/vis_data/vis_image/0010000.png_3.png b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_043015/vis_data/vis_image/0010000.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..4d5eeb1f5dcf847ce2f36f59dc16e83b5626e0e4 Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_043015/vis_data/vis_image/0010000.png_3.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_043015/vis_data/vis_image/0010000.png_4.png b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_043015/vis_data/vis_image/0010000.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..e9b63fc7a4ac6a5e2560e1dc2f357887cefd086e Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_043015/vis_data/vis_image/0010000.png_4.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_043015/vis_data/vis_image/0010000.png_5.png b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_043015/vis_data/vis_image/0010000.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..47d3dec36a501790ee1be5793ab0185c6f6c4597 Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_043015/vis_data/vis_image/0010000.png_5.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_043015/vis_data/vis_image/0010000.png_6.png b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_043015/vis_data/vis_image/0010000.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..352a288fec42ca468ed412186e266309412a72bc Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_043015/vis_data/vis_image/0010000.png_6.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_043015/vis_data/vis_image/0010000.png_7.png b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_043015/vis_data/vis_image/0010000.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..3f4c1a531e7344bd59a7b3cda05e3065535dfa1a Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_043015/vis_data/vis_image/0010000.png_7.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_043015/vis_data/vis_image/0014159.png_1.png b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_043015/vis_data/vis_image/0014159.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..fb40574c39bece67e543c20b4dacc647a8e2fc07 Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_043015/vis_data/vis_image/0014159.png_1.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_043015/vis_data/vis_image/0014159.png_2.png b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_043015/vis_data/vis_image/0014159.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..28ed6fccf4c4e13f13b6c144c8ae1b7b2b1085a7 Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_043015/vis_data/vis_image/0014159.png_2.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_043015/vis_data/vis_image/0014159.png_3.png b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_043015/vis_data/vis_image/0014159.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..725728385a0e34d23ea88743a874e126f4a794d2 Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_043015/vis_data/vis_image/0014159.png_3.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_043015/vis_data/vis_image/0014159.png_4.png b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_043015/vis_data/vis_image/0014159.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..c1ef543bd120adb203ae52bc0b90c8c0a78439c4 Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_043015/vis_data/vis_image/0014159.png_4.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_043015/vis_data/vis_image/0014159.png_5.png b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_043015/vis_data/vis_image/0014159.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..ce9e223d904e22d5619aad2d2e040b96783e6f04 Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_043015/vis_data/vis_image/0014159.png_5.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_043015/vis_data/vis_image/0014159.png_6.png b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_043015/vis_data/vis_image/0014159.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..ee365ffb1ee9808912d73fbb41c01cf1cd3fdb65 Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_043015/vis_data/vis_image/0014159.png_6.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_043015/vis_data/vis_image/0014159.png_7.png b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_043015/vis_data/vis_image/0014159.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..94187e715ec252f260dd00212ff5f470558c4d5b Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_043015/vis_data/vis_image/0014159.png_7.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_113243/20230605_113243.log b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_113243/20230605_113243.log new file mode 100644 index 0000000000000000000000000000000000000000..6dcc3dc353b630cbd21bcb4927beed4ef8ee63c8 --- /dev/null +++ b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_113243/20230605_113243.log @@ -0,0 +1,983 @@ +2023/06/05 11:32:46 - mmengine - INFO - +------------------------------------------------------------ +System environment: + sys.platform: linux + Python: 3.10.9 (main, Mar 8 2023, 10:47:38) [GCC 11.2.0] + CUDA available: True + numpy_random_seed: 1531645525 + GPU 0,1: NVIDIA A100-SXM4-80GB + CUDA_HOME: /mnt/petrelfs/share/cuda-11.6 + NVCC: Cuda compilation tools, release 11.6, V11.6.124 + GCC: gcc (GCC) 7.5.0 + PyTorch: 1.13.1 + PyTorch compiling details: PyTorch built with: + - GCC 9.3 + - C++ Version: 201402 + - Intel(R) oneAPI Math Kernel Library Version 2021.4-Product Build 20210904 for Intel(R) 64 architecture applications + - Intel(R) MKL-DNN v2.6.0 (Git Hash 52b5f107dd9cf10910aaa19cb47f3abf9b349815) + - OpenMP 201511 (a.k.a. OpenMP 4.5) + - LAPACK is enabled (usually provided by MKL) + - NNPACK is enabled + - CPU capability usage: AVX2 + - CUDA Runtime 11.6 + - NVCC architecture flags: -gencode;arch=compute_37,code=sm_37;-gencode;arch=compute_50,code=sm_50;-gencode;arch=compute_60,code=sm_60;-gencode;arch=compute_61,code=sm_61;-gencode;arch=compute_70,code=sm_70;-gencode;arch=compute_75,code=sm_75;-gencode;arch=compute_80,code=sm_80;-gencode;arch=compute_86,code=sm_86;-gencode;arch=compute_37,code=compute_37 + - CuDNN 8.3.2 (built against CUDA 11.5) + - Magma 2.6.1 + - Build settings: BLAS_INFO=mkl, BUILD_TYPE=Release, CUDA_VERSION=11.6, CUDNN_VERSION=8.3.2, CXX_COMPILER=/opt/rh/devtoolset-9/root/usr/bin/c++, CXX_FLAGS= -fabi-version=11 -Wno-deprecated -fvisibility-inlines-hidden -DUSE_PTHREADPOOL -fopenmp -DNDEBUG -DUSE_KINETO -DUSE_FBGEMM -DUSE_QNNPACK -DUSE_PYTORCH_QNNPACK -DUSE_XNNPACK -DSYMBOLICATE_MOBILE_DEBUG_HANDLE -DEDGE_PROFILER_USE_KINETO -O2 -fPIC -Wno-narrowing -Wall -Wextra -Werror=return-type -Werror=non-virtual-dtor -Wno-missing-field-initializers -Wno-type-limits -Wno-array-bounds -Wno-unknown-pragmas -Wunused-local-typedefs -Wno-unused-parameter -Wno-unused-function -Wno-unused-result -Wno-strict-overflow -Wno-strict-aliasing -Wno-error=deprecated-declarations -Wno-stringop-overflow -Wno-psabi -Wno-error=pedantic -Wno-error=redundant-decls -Wno-error=old-style-cast -fdiagnostics-color=always -faligned-new -Wno-unused-but-set-variable -Wno-maybe-uninitialized -fno-math-errno -fno-trapping-math -Werror=format -Werror=cast-function-type -Wno-stringop-overflow, LAPACK_INFO=mkl, PERF_WITH_AVX=1, PERF_WITH_AVX2=1, PERF_WITH_AVX512=1, TORCH_VERSION=1.13.1, USE_CUDA=ON, USE_CUDNN=ON, USE_EXCEPTION_PTR=1, USE_GFLAGS=OFF, USE_GLOG=OFF, USE_MKL=ON, USE_MKLDNN=ON, USE_MPI=OFF, USE_NCCL=ON, USE_NNPACK=ON, USE_OPENMP=ON, USE_ROCM=OFF, + + TorchVision: 0.14.1 + OpenCV: 4.7.0 + MMEngine: 0.7.3 + +Runtime environment: + cudnn_benchmark: True + mp_cfg: {'mp_start_method': 'fork', 'opencv_num_threads': 0} + dist_cfg: {'backend': 'nccl'} + seed: None + deterministic: False + Distributed launcher: slurm + Distributed training: True + GPU number: 2 +------------------------------------------------------------ + +2023/06/05 11:32:51 - mmengine - INFO - Config: +optim_wrapper = dict( + optimizer=dict( + type='SGD', + lr=0.0001, + momentum=0.9, + weight_decay=0.0001, + _scope_='mmpretrain'), + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=False, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=512) +model = dict( + type='ImageClassifier', + backbone=dict( + frozen_stages=2, + type='ResNet', + depth=50, + num_stages=4, + out_indices=(3, ), + style='pytorch', + init_cfg=dict( + type='Pretrained', + checkpoint= + 'https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth', + prefix='backbone')), + neck=dict(type='GlobalAveragePooling'), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=2048, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + topk=1)) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/all1.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/all2.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/all3.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3fake8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/cc1m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3real8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV2-1-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/if-dpmsolver++-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = dict(type='Accuracy', topk=1) +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV2-1-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/if-dpmsolver++-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = dict(type='Accuracy', topk=1) +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1' + +2023/06/05 11:33:07 - mmengine - INFO - Hooks will be executed in the following order: +before_run: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_load_checkpoint: +(ABOVE_NORMAL) EMAHook + -------------------- +before_train: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_train_epoch: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(NORMAL ) DistSamplerSeedHook + -------------------- +before_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook + -------------------- +after_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +after_train_epoch: +(NORMAL ) IterTimerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_val_epoch: +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook + -------------------- +before_val_iter: +(NORMAL ) IterTimerHook + -------------------- +after_val_iter: +(NORMAL ) IterTimerHook +(NORMAL ) VisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_val_epoch: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_save_checkpoint: +(ABOVE_NORMAL) EMAHook + -------------------- +after_train: +(VERY_LOW ) CheckpointHook + -------------------- +before_test_epoch: +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook + -------------------- +before_test_iter: +(NORMAL ) IterTimerHook + -------------------- +after_test_iter: +(NORMAL ) IterTimerHook +(NORMAL ) VisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_test_epoch: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_run: +(BELOW_NORMAL) LoggerHook + -------------------- +2023/06/05 11:33:28 - mmengine - INFO - load backbone in model from: https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth +Name of parameter - Initialization information + +backbone.conv1.weight - torch.Size([64, 3, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.bn1.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.bn1.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.conv1.weight - torch.Size([64, 64, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.bn1.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.bn1.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.conv2.weight - torch.Size([64, 64, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.bn2.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.bn2.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.conv3.weight - torch.Size([256, 64, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.bn3.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.bn3.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.downsample.0.weight - torch.Size([256, 64, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.downsample.1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.downsample.1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.conv1.weight - torch.Size([64, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.bn1.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.bn1.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.conv2.weight - torch.Size([64, 64, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.bn2.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.bn2.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.conv3.weight - torch.Size([256, 64, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.bn3.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.bn3.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.conv1.weight - torch.Size([64, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.bn1.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.bn1.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.conv2.weight - torch.Size([64, 64, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.bn2.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.bn2.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.conv3.weight - torch.Size([256, 64, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.bn3.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.bn3.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.conv1.weight - torch.Size([128, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.bn1.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.bn1.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.conv2.weight - torch.Size([128, 128, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.bn2.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.bn2.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.conv3.weight - torch.Size([512, 128, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.bn3.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.bn3.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.downsample.0.weight - torch.Size([512, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.downsample.1.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.downsample.1.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.conv1.weight - torch.Size([128, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.bn1.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.bn1.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.conv2.weight - torch.Size([128, 128, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.bn2.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.bn2.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.conv3.weight - torch.Size([512, 128, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.bn3.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.bn3.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.conv1.weight - torch.Size([128, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.bn1.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.bn1.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.conv2.weight - torch.Size([128, 128, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.bn2.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.bn2.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.conv3.weight - torch.Size([512, 128, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.bn3.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.bn3.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.conv1.weight - torch.Size([128, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.bn1.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.bn1.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.conv2.weight - torch.Size([128, 128, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.bn2.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.bn2.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.conv3.weight - torch.Size([512, 128, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.bn3.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.bn3.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.conv1.weight - torch.Size([256, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.bn1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.bn1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.conv2.weight - torch.Size([256, 256, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.bn2.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.bn2.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.conv3.weight - torch.Size([1024, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.bn3.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.bn3.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.downsample.0.weight - torch.Size([1024, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.downsample.1.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.downsample.1.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.conv1.weight - torch.Size([256, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.bn1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.bn1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.conv2.weight - torch.Size([256, 256, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.bn2.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.bn2.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.conv3.weight - torch.Size([1024, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.bn3.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.bn3.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.conv1.weight - torch.Size([256, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.bn1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.bn1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.conv2.weight - torch.Size([256, 256, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.bn2.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.bn2.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.conv3.weight - torch.Size([1024, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.bn3.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.bn3.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.conv1.weight - torch.Size([256, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.bn1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.bn1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.conv2.weight - torch.Size([256, 256, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.bn2.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.bn2.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.conv3.weight - torch.Size([1024, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.bn3.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.bn3.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.conv1.weight - torch.Size([256, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.bn1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.bn1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.conv2.weight - torch.Size([256, 256, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.bn2.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.bn2.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.conv3.weight - torch.Size([1024, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.bn3.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.bn3.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.conv1.weight - torch.Size([256, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.bn1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.bn1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.conv2.weight - torch.Size([256, 256, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.bn2.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.bn2.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.conv3.weight - torch.Size([1024, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.bn3.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.bn3.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.conv1.weight - torch.Size([512, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.bn1.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.bn1.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.conv2.weight - torch.Size([512, 512, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.bn2.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.bn2.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.conv3.weight - torch.Size([2048, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.bn3.weight - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.bn3.bias - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.downsample.0.weight - torch.Size([2048, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.downsample.1.weight - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.downsample.1.bias - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.conv1.weight - torch.Size([512, 2048, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.bn1.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.bn1.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.conv2.weight - torch.Size([512, 512, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.bn2.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.bn2.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.conv3.weight - torch.Size([2048, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.bn3.weight - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.bn3.bias - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.conv1.weight - torch.Size([512, 2048, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.bn1.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.bn1.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.conv2.weight - torch.Size([512, 512, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.bn2.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.bn2.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.conv3.weight - torch.Size([2048, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.bn3.weight - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.bn3.bias - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +head.fc.weight - torch.Size([2, 2048]): +NormalInit: mean=0, std=0.01, bias=0 + +head.fc.bias - torch.Size([2]): +NormalInit: mean=0, std=0.01, bias=0 +2023/06/05 11:33:28 - mmengine - WARNING - "FileClient" will be deprecated in future. Please use io functions in https://mmengine.readthedocs.io/en/latest/api/fileio.html#file-io +2023/06/05 11:33:28 - mmengine - WARNING - "HardDiskBackend" is the alias of "LocalBackend" and the former will be deprecated in future. +2023/06/05 11:33:28 - mmengine - INFO - Checkpoints will be saved to /mnt/petrelfs/luzeyu/workspace/fakebench/mmpretrain/workdir/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1. +2023/06/05 11:34:54 - mmengine - INFO - Epoch(train) [1][ 100/4111] lr: 9.9999e-05 eta: 9:43:29 time: 0.7275 data_time: 0.5866 memory: 9436 loss: 0.6508 diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_113243/vis_data/20230605_113243.json b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_113243/vis_data/20230605_113243.json new file mode 100644 index 0000000000000000000000000000000000000000..60a19c3439e7b49063589b39bbe7ffedfaa3973a --- /dev/null +++ b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_113243/vis_data/20230605_113243.json @@ -0,0 +1 @@ +{"lr": 9.999871217862936e-05, "data_time": 0.5866034984588623, "loss": 0.6508402526378632, "time": 0.7275227308273315, "epoch": 1, "memory": 9436, "step": 100} diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_113243/vis_data/config.py b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_113243/vis_data/config.py new file mode 100644 index 0000000000000000000000000000000000000000..b35ee250106aa84d1065073f108965cfa202c34d --- /dev/null +++ b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_113243/vis_data/config.py @@ -0,0 +1,366 @@ +optim_wrapper = dict( + optimizer=dict( + type='SGD', + lr=0.0001, + momentum=0.9, + weight_decay=0.0001, + _scope_='mmpretrain'), + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=False, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=512) +model = dict( + type='ImageClassifier', + backbone=dict( + frozen_stages=2, + type='ResNet', + depth=50, + num_stages=4, + out_indices=(3, ), + style='pytorch', + init_cfg=dict( + type='Pretrained', + checkpoint= + 'https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth', + prefix='backbone')), + neck=dict(type='GlobalAveragePooling'), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=2048, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + topk=1)) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/all1.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/all2.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/all3.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3fake8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/cc1m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3real8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV2-1-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/if-dpmsolver++-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = dict(type='Accuracy', topk=1) +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV2-1-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/if-dpmsolver++-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = dict(type='Accuracy', topk=1) +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1' diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_113243/vis_data/events.out.tfevents.1685935981.SH-IDC1-10-140-24-133.66110.0 b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_113243/vis_data/events.out.tfevents.1685935981.SH-IDC1-10-140-24-133.66110.0 new file mode 100644 index 0000000000000000000000000000000000000000..d8c3b84c5dc31ee2ddb77134167adc3183a8bbc1 --- /dev/null +++ b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_113243/vis_data/events.out.tfevents.1685935981.SH-IDC1-10-140-24-133.66110.0 @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:da7b935bb8f830c7832a247858aa915e4c3eb564a7f1638a3a301e26d3d8bd1f +size 15581 diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_113243/vis_data/scalars.json b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_113243/vis_data/scalars.json new file mode 100644 index 0000000000000000000000000000000000000000..60a19c3439e7b49063589b39bbe7ffedfaa3973a --- /dev/null +++ b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_113243/vis_data/scalars.json @@ -0,0 +1 @@ +{"lr": 9.999871217862936e-05, "data_time": 0.5866034984588623, "loss": 0.6508402526378632, "time": 0.7275227308273315, "epoch": 1, "memory": 9436, "step": 100} diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_114723/20230605_114723.log b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_114723/20230605_114723.log new file mode 100644 index 0000000000000000000000000000000000000000..222bc24c0a8d944a282f8c36066798e304ae4291 --- /dev/null +++ b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_114723/20230605_114723.log @@ -0,0 +1,1011 @@ +2023/06/05 11:47:27 - mmengine - INFO - +------------------------------------------------------------ +System environment: + sys.platform: linux + Python: 3.10.9 (main, Mar 8 2023, 10:47:38) [GCC 11.2.0] + CUDA available: True + numpy_random_seed: 1002084629 + GPU 0,1: NVIDIA A100-SXM4-80GB + CUDA_HOME: /mnt/petrelfs/share/cuda-11.6 + NVCC: Cuda compilation tools, release 11.6, V11.6.124 + GCC: gcc (GCC) 7.5.0 + PyTorch: 1.13.1 + PyTorch compiling details: PyTorch built with: + - GCC 9.3 + - C++ Version: 201402 + - Intel(R) oneAPI Math Kernel Library Version 2021.4-Product Build 20210904 for Intel(R) 64 architecture applications + - Intel(R) MKL-DNN v2.6.0 (Git Hash 52b5f107dd9cf10910aaa19cb47f3abf9b349815) + - OpenMP 201511 (a.k.a. OpenMP 4.5) + - LAPACK is enabled (usually provided by MKL) + - NNPACK is enabled + - CPU capability usage: AVX2 + - CUDA Runtime 11.6 + - NVCC architecture flags: -gencode;arch=compute_37,code=sm_37;-gencode;arch=compute_50,code=sm_50;-gencode;arch=compute_60,code=sm_60;-gencode;arch=compute_61,code=sm_61;-gencode;arch=compute_70,code=sm_70;-gencode;arch=compute_75,code=sm_75;-gencode;arch=compute_80,code=sm_80;-gencode;arch=compute_86,code=sm_86;-gencode;arch=compute_37,code=compute_37 + - CuDNN 8.3.2 (built against CUDA 11.5) + - Magma 2.6.1 + - Build settings: BLAS_INFO=mkl, BUILD_TYPE=Release, CUDA_VERSION=11.6, CUDNN_VERSION=8.3.2, CXX_COMPILER=/opt/rh/devtoolset-9/root/usr/bin/c++, CXX_FLAGS= -fabi-version=11 -Wno-deprecated -fvisibility-inlines-hidden -DUSE_PTHREADPOOL -fopenmp -DNDEBUG -DUSE_KINETO -DUSE_FBGEMM -DUSE_QNNPACK -DUSE_PYTORCH_QNNPACK -DUSE_XNNPACK -DSYMBOLICATE_MOBILE_DEBUG_HANDLE -DEDGE_PROFILER_USE_KINETO -O2 -fPIC -Wno-narrowing -Wall -Wextra -Werror=return-type -Werror=non-virtual-dtor -Wno-missing-field-initializers -Wno-type-limits -Wno-array-bounds -Wno-unknown-pragmas -Wunused-local-typedefs -Wno-unused-parameter -Wno-unused-function -Wno-unused-result -Wno-strict-overflow -Wno-strict-aliasing -Wno-error=deprecated-declarations -Wno-stringop-overflow -Wno-psabi -Wno-error=pedantic -Wno-error=redundant-decls -Wno-error=old-style-cast -fdiagnostics-color=always -faligned-new -Wno-unused-but-set-variable -Wno-maybe-uninitialized -fno-math-errno -fno-trapping-math -Werror=format -Werror=cast-function-type -Wno-stringop-overflow, LAPACK_INFO=mkl, PERF_WITH_AVX=1, PERF_WITH_AVX2=1, PERF_WITH_AVX512=1, TORCH_VERSION=1.13.1, USE_CUDA=ON, USE_CUDNN=ON, USE_EXCEPTION_PTR=1, USE_GFLAGS=OFF, USE_GLOG=OFF, USE_MKL=ON, USE_MKLDNN=ON, USE_MPI=OFF, USE_NCCL=ON, USE_NNPACK=ON, USE_OPENMP=ON, USE_ROCM=OFF, + + TorchVision: 0.14.1 + OpenCV: 4.7.0 + MMEngine: 0.7.3 + +Runtime environment: + cudnn_benchmark: True + mp_cfg: {'mp_start_method': 'fork', 'opencv_num_threads': 0} + dist_cfg: {'backend': 'nccl'} + seed: None + deterministic: False + Distributed launcher: slurm + Distributed training: True + GPU number: 2 +------------------------------------------------------------ + +2023/06/05 11:47:31 - mmengine - INFO - Config: +optim_wrapper = dict( + optimizer=dict( + type='SGD', + lr=0.0001, + momentum=0.9, + weight_decay=0.0001, + _scope_='mmpretrain'), + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=False, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=512) +model = dict( + type='ImageClassifier', + backbone=dict( + frozen_stages=2, + type='ResNet', + depth=50, + num_stages=4, + out_indices=(3, ), + style='pytorch', + init_cfg=dict( + type='Pretrained', + checkpoint= + 'https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth', + prefix='backbone')), + neck=dict(type='GlobalAveragePooling'), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=2048, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + topk=1)) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/all1.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/all2.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/all3.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3fake8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/cc1m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3real8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV2-1-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/if-dpmsolver++-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = dict(type='Accuracy', topk=1) +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV2-1-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/if-dpmsolver++-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = dict(type='Accuracy', topk=1) +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1' + +2023/06/05 11:47:43 - mmengine - INFO - Hooks will be executed in the following order: +before_run: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_load_checkpoint: +(ABOVE_NORMAL) EMAHook + -------------------- +before_train: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_train_epoch: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(NORMAL ) DistSamplerSeedHook + -------------------- +before_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook + -------------------- +after_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +after_train_epoch: +(NORMAL ) IterTimerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_val_epoch: +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook + -------------------- +before_val_iter: +(NORMAL ) IterTimerHook + -------------------- +after_val_iter: +(NORMAL ) IterTimerHook +(NORMAL ) VisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_val_epoch: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_save_checkpoint: +(ABOVE_NORMAL) EMAHook + -------------------- +after_train: +(VERY_LOW ) CheckpointHook + -------------------- +before_test_epoch: +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook + -------------------- +before_test_iter: +(NORMAL ) IterTimerHook + -------------------- +after_test_iter: +(NORMAL ) IterTimerHook +(NORMAL ) VisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_test_epoch: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_run: +(BELOW_NORMAL) LoggerHook + -------------------- +2023/06/05 11:48:04 - mmengine - INFO - load backbone in model from: https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth +Name of parameter - Initialization information + +backbone.conv1.weight - torch.Size([64, 3, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.bn1.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.bn1.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.conv1.weight - torch.Size([64, 64, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.bn1.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.bn1.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.conv2.weight - torch.Size([64, 64, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.bn2.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.bn2.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.conv3.weight - torch.Size([256, 64, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.bn3.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.bn3.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.downsample.0.weight - torch.Size([256, 64, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.downsample.1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.downsample.1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.conv1.weight - torch.Size([64, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.bn1.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.bn1.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.conv2.weight - torch.Size([64, 64, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.bn2.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.bn2.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.conv3.weight - torch.Size([256, 64, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.bn3.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.bn3.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.conv1.weight - torch.Size([64, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.bn1.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.bn1.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.conv2.weight - torch.Size([64, 64, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.bn2.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.bn2.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.conv3.weight - torch.Size([256, 64, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.bn3.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.bn3.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.conv1.weight - torch.Size([128, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.bn1.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.bn1.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.conv2.weight - torch.Size([128, 128, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.bn2.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.bn2.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.conv3.weight - torch.Size([512, 128, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.bn3.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.bn3.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.downsample.0.weight - torch.Size([512, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.downsample.1.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.downsample.1.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.conv1.weight - torch.Size([128, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.bn1.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.bn1.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.conv2.weight - torch.Size([128, 128, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.bn2.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.bn2.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.conv3.weight - torch.Size([512, 128, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.bn3.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.bn3.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.conv1.weight - torch.Size([128, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.bn1.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.bn1.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.conv2.weight - torch.Size([128, 128, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.bn2.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.bn2.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.conv3.weight - torch.Size([512, 128, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.bn3.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.bn3.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.conv1.weight - torch.Size([128, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.bn1.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.bn1.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.conv2.weight - torch.Size([128, 128, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.bn2.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.bn2.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.conv3.weight - torch.Size([512, 128, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.bn3.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.bn3.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.conv1.weight - torch.Size([256, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.bn1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.bn1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.conv2.weight - torch.Size([256, 256, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.bn2.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.bn2.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.conv3.weight - torch.Size([1024, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.bn3.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.bn3.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.downsample.0.weight - torch.Size([1024, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.downsample.1.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.downsample.1.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.conv1.weight - torch.Size([256, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.bn1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.bn1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.conv2.weight - torch.Size([256, 256, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.bn2.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.bn2.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.conv3.weight - torch.Size([1024, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.bn3.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.bn3.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.conv1.weight - torch.Size([256, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.bn1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.bn1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.conv2.weight - torch.Size([256, 256, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.bn2.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.bn2.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.conv3.weight - torch.Size([1024, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.bn3.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.bn3.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.conv1.weight - torch.Size([256, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.bn1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.bn1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.conv2.weight - torch.Size([256, 256, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.bn2.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.bn2.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.conv3.weight - torch.Size([1024, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.bn3.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.bn3.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.conv1.weight - torch.Size([256, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.bn1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.bn1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.conv2.weight - torch.Size([256, 256, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.bn2.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.bn2.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.conv3.weight - torch.Size([1024, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.bn3.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.bn3.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.conv1.weight - torch.Size([256, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.bn1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.bn1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.conv2.weight - torch.Size([256, 256, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.bn2.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.bn2.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.conv3.weight - torch.Size([1024, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.bn3.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.bn3.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.conv1.weight - torch.Size([512, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.bn1.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.bn1.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.conv2.weight - torch.Size([512, 512, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.bn2.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.bn2.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.conv3.weight - torch.Size([2048, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.bn3.weight - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.bn3.bias - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.downsample.0.weight - torch.Size([2048, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.downsample.1.weight - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.downsample.1.bias - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.conv1.weight - torch.Size([512, 2048, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.bn1.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.bn1.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.conv2.weight - torch.Size([512, 512, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.bn2.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.bn2.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.conv3.weight - torch.Size([2048, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.bn3.weight - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.bn3.bias - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.conv1.weight - torch.Size([512, 2048, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.bn1.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.bn1.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.conv2.weight - torch.Size([512, 512, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.bn2.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.bn2.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.conv3.weight - torch.Size([2048, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.bn3.weight - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.bn3.bias - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +head.fc.weight - torch.Size([2, 2048]): +NormalInit: mean=0, std=0.01, bias=0 + +head.fc.bias - torch.Size([2]): +NormalInit: mean=0, std=0.01, bias=0 +2023/06/05 11:48:04 - mmengine - WARNING - "FileClient" will be deprecated in future. Please use io functions in https://mmengine.readthedocs.io/en/latest/api/fileio.html#file-io +2023/06/05 11:48:04 - mmengine - WARNING - "HardDiskBackend" is the alias of "LocalBackend" and the former will be deprecated in future. +2023/06/05 11:48:04 - mmengine - INFO - Checkpoints will be saved to /mnt/petrelfs/luzeyu/workspace/fakebench/mmpretrain/workdir/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1. +2023/06/05 11:49:23 - mmengine - INFO - Epoch(train) [1][ 100/4111] lr: 9.9999e-05 eta: 9:02:24 time: 0.7415 data_time: 0.6021 memory: 9436 loss: 0.6495 +2023/06/05 11:50:43 - mmengine - INFO - Epoch(train) [1][ 200/4111] lr: 9.9995e-05 eta: 9:03:33 time: 0.7898 data_time: 0.6385 memory: 6319 loss: 0.6031 +2023/06/05 11:51:57 - mmengine - INFO - Epoch(train) [1][ 300/4111] lr: 9.9988e-05 eta: 8:48:02 time: 0.7234 data_time: 0.5820 memory: 6319 loss: 0.5771 +2023/06/05 11:53:13 - mmengine - INFO - Epoch(train) [1][ 400/4111] lr: 9.9979e-05 eta: 8:44:17 time: 0.7362 data_time: 0.5954 memory: 6319 loss: 0.5439 +2023/06/05 11:54:29 - mmengine - INFO - Epoch(train) [1][ 500/4111] lr: 9.9967e-05 eta: 8:40:59 time: 0.7681 data_time: 0.6277 memory: 6319 loss: 0.5165 +2023/06/05 11:55:44 - mmengine - INFO - Epoch(train) [1][ 600/4111] lr: 9.9953e-05 eta: 8:38:13 time: 0.7868 data_time: 0.6472 memory: 6319 loss: 0.5025 +2023/06/05 11:56:59 - mmengine - INFO - Epoch(train) [1][ 700/4111] lr: 9.9936e-05 eta: 8:35:10 time: 0.7355 data_time: 0.5949 memory: 6319 loss: 0.4765 +2023/06/05 11:58:16 - mmengine - INFO - Epoch(train) [1][ 800/4111] lr: 9.9916e-05 eta: 8:34:15 time: 0.7956 data_time: 0.6555 memory: 6319 loss: 0.4560 +2023/06/05 11:59:31 - mmengine - INFO - Epoch(train) [1][ 900/4111] lr: 9.9894e-05 eta: 8:31:41 time: 0.7758 data_time: 0.6346 memory: 6319 loss: 0.4433 +2023/06/05 12:00:45 - mmengine - INFO - Exp name: resnet50_2xb256_all_1m_lr1e-4_aug_5e-1_20230605_114723 +2023/06/05 12:00:45 - mmengine - INFO - Epoch(train) [1][1000/4111] lr: 9.9869e-05 eta: 8:28:56 time: 0.7063 data_time: 0.5671 memory: 6319 loss: 0.4438 +2023/06/05 12:02:01 - mmengine - INFO - Epoch(train) [1][1100/4111] lr: 9.9841e-05 eta: 8:27:17 time: 0.7406 data_time: 0.5994 memory: 6319 loss: 0.4015 +2023/06/05 12:03:18 - mmengine - INFO - Epoch(train) [1][1200/4111] lr: 9.9811e-05 eta: 8:26:54 time: 0.7685 data_time: 0.6274 memory: 6319 loss: 0.4095 +2023/06/05 12:04:34 - mmengine - INFO - Epoch(train) [1][1300/4111] lr: 9.9778e-05 eta: 8:25:33 time: 0.7007 data_time: 0.5607 memory: 6319 loss: 0.4167 +2023/06/05 12:05:48 - mmengine - INFO - Epoch(train) [1][1400/4111] lr: 9.9743e-05 eta: 8:23:18 time: 0.6940 data_time: 0.5535 memory: 6319 loss: 0.3922 +2023/06/05 12:07:18 - mmengine - INFO - Epoch(train) [1][1500/4111] lr: 9.9705e-05 eta: 8:27:56 time: 0.7725 data_time: 0.6325 memory: 6319 loss: 0.3987 +2023/06/05 12:08:35 - mmengine - INFO - Epoch(train) [1][1600/4111] lr: 9.9664e-05 eta: 8:26:31 time: 0.8433 data_time: 0.7027 memory: 6319 loss: 0.3887 +2023/06/05 12:09:46 - mmengine - INFO - Epoch(train) [1][1700/4111] lr: 9.9621e-05 eta: 8:23:08 time: 0.7507 data_time: 0.6097 memory: 6319 loss: 0.3801 +2023/06/05 12:10:59 - mmengine - INFO - Epoch(train) [1][1800/4111] lr: 9.9575e-05 eta: 8:20:31 time: 0.8186 data_time: 0.6780 memory: 6319 loss: 0.3910 +2023/06/05 12:12:15 - mmengine - INFO - Epoch(train) [1][1900/4111] lr: 9.9527e-05 eta: 8:19:07 time: 0.7826 data_time: 0.6428 memory: 6319 loss: 0.3794 +2023/06/05 12:13:28 - mmengine - INFO - Exp name: resnet50_2xb256_all_1m_lr1e-4_aug_5e-1_20230605_114723 +2023/06/05 12:13:28 - mmengine - INFO - Epoch(train) [1][2000/4111] lr: 9.9476e-05 eta: 8:16:49 time: 0.7061 data_time: 0.5656 memory: 6319 loss: 0.3654 +2023/06/05 12:14:40 - mmengine - INFO - Epoch(train) [1][2100/4111] lr: 9.9422e-05 eta: 8:14:13 time: 0.7454 data_time: 0.6046 memory: 6319 loss: 0.3579 +2023/06/05 12:15:56 - mmengine - INFO - Epoch(train) [1][2200/4111] lr: 9.9366e-05 eta: 8:12:47 time: 0.8435 data_time: 0.7029 memory: 6319 loss: 0.3607 +2023/06/05 12:17:11 - mmengine - INFO - Epoch(train) [1][2300/4111] lr: 9.9307e-05 eta: 8:11:17 time: 0.7672 data_time: 0.6265 memory: 6319 loss: 0.3667 +2023/06/05 12:18:26 - mmengine - INFO - Epoch(train) [1][2400/4111] lr: 9.9246e-05 eta: 8:09:43 time: 0.6848 data_time: 0.5447 memory: 6319 loss: 0.3394 +2023/06/05 12:19:41 - mmengine - INFO - Epoch(train) [1][2500/4111] lr: 9.9182e-05 eta: 8:08:19 time: 0.7179 data_time: 0.5786 memory: 6319 loss: 0.3501 +2023/06/05 12:20:52 - mmengine - INFO - Epoch(train) [1][2600/4111] lr: 9.9115e-05 eta: 8:05:57 time: 0.7215 data_time: 0.5806 memory: 6319 loss: 0.3724 +2023/06/05 12:22:06 - mmengine - INFO - Epoch(train) [1][2700/4111] lr: 9.9046e-05 eta: 8:04:04 time: 0.6675 data_time: 0.5273 memory: 6319 loss: 0.3434 diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_114723/vis_data/20230605_114723.json b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_114723/vis_data/20230605_114723.json new file mode 100644 index 0000000000000000000000000000000000000000..f218303722266d44b90df6d500f4b2f89d3b32b3 --- /dev/null +++ b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_114723/vis_data/20230605_114723.json @@ -0,0 +1,27 @@ +{"lr": 9.999871217862936e-05, "data_time": 0.6020993947982788, "loss": 0.6494906187057495, "time": 0.741540503501892, "epoch": 1, "memory": 9436, "step": 100} +{"lr": 9.999479662539605e-05, "data_time": 0.6385319471359253, "loss": 0.6031487882137299, "time": 0.7898366451263428, "epoch": 1, "memory": 6319, "step": 200} +{"lr": 9.99882534375654e-05, "data_time": 0.5819762229919434, "loss": 0.5770677506923676, "time": 0.7234013557434082, "epoch": 1, "memory": 6319, "step": 300} +{"lr": 9.997908299725075e-05, "data_time": 0.5954420804977417, "loss": 0.5438629209995269, "time": 0.7362374544143677, "epoch": 1, "memory": 6319, "step": 400} +{"lr": 9.996728583999318e-05, "data_time": 0.627727723121643, "loss": 0.5164570271968841, "time": 0.7681180000305176, "epoch": 1, "memory": 6319, "step": 500} +{"lr": 9.995286265473058e-05, "data_time": 0.6471993923187256, "loss": 0.5024662017822266, "time": 0.7867578506469727, "epoch": 1, "memory": 6319, "step": 600} +{"lr": 9.993581428375725e-05, "data_time": 0.5948776245117188, "loss": 0.476534429192543, "time": 0.735540246963501, "epoch": 1, "memory": 6319, "step": 700} +{"lr": 9.991614172267466e-05, "data_time": 0.6555254459381104, "loss": 0.45595123767852785, "time": 0.7956423044204712, "epoch": 1, "memory": 6319, "step": 800} +{"lr": 9.989384612033355e-05, "data_time": 0.634600830078125, "loss": 0.4433297276496887, "time": 0.7757574558258057, "epoch": 1, "memory": 6319, "step": 900} +{"lr": 9.986892877876658e-05, "data_time": 0.567132544517517, "loss": 0.44384285509586335, "time": 0.7062573671340943, "epoch": 1, "memory": 6319, "step": 1000} +{"lr": 9.984139115311244e-05, "data_time": 0.599405312538147, "loss": 0.4014589101076126, "time": 0.7406479120254517, "epoch": 1, "memory": 6319, "step": 1100} +{"lr": 9.981123485153081e-05, "data_time": 0.6274431943893433, "loss": 0.4095339685678482, "time": 0.7685052156448364, "epoch": 1, "memory": 6319, "step": 1200} +{"lr": 9.977846163510851e-05, "data_time": 0.5607189416885376, "loss": 0.4167290240526199, "time": 0.7007149219512939, "epoch": 1, "memory": 6319, "step": 1300} +{"lr": 9.974307341775653e-05, "data_time": 0.5535310268402099, "loss": 0.3922219753265381, "time": 0.6939513444900512, "epoch": 1, "memory": 6319, "step": 1400} +{"lr": 9.970507226609833e-05, "data_time": 0.6324576377868653, "loss": 0.39867656528949735, "time": 0.7724632501602173, "epoch": 1, "memory": 6319, "step": 1500} +{"lr": 9.966446039934934e-05, "data_time": 0.7027170658111572, "loss": 0.38865671753883363, "time": 0.8433010816574097, "epoch": 1, "memory": 6319, "step": 1600} +{"lr": 9.962124018918683e-05, "data_time": 0.6097376108169555, "loss": 0.38005087077617644, "time": 0.7507309436798095, "epoch": 1, "memory": 6319, "step": 1700} +{"lr": 9.957541415961212e-05, "data_time": 0.6780000448226928, "loss": 0.3909699708223343, "time": 0.8186037063598632, "epoch": 1, "memory": 6319, "step": 1800} +{"lr": 9.952698498680267e-05, "data_time": 0.642765212059021, "loss": 0.37944130003452303, "time": 0.7826313734054565, "epoch": 1, "memory": 6319, "step": 1900} +{"lr": 9.947595549895583e-05, "data_time": 0.5656037092208862, "loss": 0.36535309851169584, "time": 0.7060959815979004, "epoch": 1, "memory": 6319, "step": 2000} +{"lr": 9.942232867612417e-05, "data_time": 0.6045804738998413, "loss": 0.35787458121776583, "time": 0.7453924417495728, "epoch": 1, "memory": 6319, "step": 2100} +{"lr": 9.93661076500406e-05, "data_time": 0.7028562545776367, "loss": 0.3606951117515564, "time": 0.8435119152069092, "epoch": 1, "memory": 6319, "step": 2200} +{"lr": 9.930729570393622e-05, "data_time": 0.6264518022537231, "loss": 0.3666966199874878, "time": 0.7672350406646729, "epoch": 1, "memory": 6319, "step": 2300} +{"lr": 9.924589627234824e-05, "data_time": 0.5447150945663453, "loss": 0.33944440484046934, "time": 0.6848401308059693, "epoch": 1, "memory": 6319, "step": 2400} +{"lr": 9.918191294091956e-05, "data_time": 0.5785748958587646, "loss": 0.35007711946964265, "time": 0.717919135093689, "epoch": 1, "memory": 6319, "step": 2500} +{"lr": 9.91153494461892e-05, "data_time": 0.5806029558181762, "loss": 0.37237260341644285, "time": 0.7214610338211059, "epoch": 1, "memory": 6319, "step": 2600} +{"lr": 9.904620967537422e-05, "data_time": 0.5273166179656983, "loss": 0.3433519512414932, "time": 0.667465353012085, "epoch": 1, "memory": 6319, "step": 2700} diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_114723/vis_data/config.py b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_114723/vis_data/config.py new file mode 100644 index 0000000000000000000000000000000000000000..b35ee250106aa84d1065073f108965cfa202c34d --- /dev/null +++ b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_114723/vis_data/config.py @@ -0,0 +1,366 @@ +optim_wrapper = dict( + optimizer=dict( + type='SGD', + lr=0.0001, + momentum=0.9, + weight_decay=0.0001, + _scope_='mmpretrain'), + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=False, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=512) +model = dict( + type='ImageClassifier', + backbone=dict( + frozen_stages=2, + type='ResNet', + depth=50, + num_stages=4, + out_indices=(3, ), + style='pytorch', + init_cfg=dict( + type='Pretrained', + checkpoint= + 'https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth', + prefix='backbone')), + neck=dict(type='GlobalAveragePooling'), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=2048, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + topk=1)) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/all1.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/all2.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/all3.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3fake8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/cc1m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3real8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV2-1-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/if-dpmsolver++-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = dict(type='Accuracy', topk=1) +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV2-1-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/if-dpmsolver++-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = dict(type='Accuracy', topk=1) +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1' diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_114723/vis_data/events.out.tfevents.1685936858.SH-IDC1-10-140-24-17.136269.0 b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_114723/vis_data/events.out.tfevents.1685936858.SH-IDC1-10-140-24-17.136269.0 new file mode 100644 index 0000000000000000000000000000000000000000..9d53e81b1af68c6bdd203aa3569698d22172a894 --- /dev/null +++ b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_114723/vis_data/events.out.tfevents.1685936858.SH-IDC1-10-140-24-17.136269.0 @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ef81ac89d67d7b9cfb1bf8dfd8f447fcedee57909c6aa9d2a0a34c3559d50f44 +size 22445 diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_114723/vis_data/scalars.json b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_114723/vis_data/scalars.json new file mode 100644 index 0000000000000000000000000000000000000000..f218303722266d44b90df6d500f4b2f89d3b32b3 --- /dev/null +++ b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_114723/vis_data/scalars.json @@ -0,0 +1,27 @@ +{"lr": 9.999871217862936e-05, "data_time": 0.6020993947982788, "loss": 0.6494906187057495, "time": 0.741540503501892, "epoch": 1, "memory": 9436, "step": 100} +{"lr": 9.999479662539605e-05, "data_time": 0.6385319471359253, "loss": 0.6031487882137299, "time": 0.7898366451263428, "epoch": 1, "memory": 6319, "step": 200} +{"lr": 9.99882534375654e-05, "data_time": 0.5819762229919434, "loss": 0.5770677506923676, "time": 0.7234013557434082, "epoch": 1, "memory": 6319, "step": 300} +{"lr": 9.997908299725075e-05, "data_time": 0.5954420804977417, "loss": 0.5438629209995269, "time": 0.7362374544143677, "epoch": 1, "memory": 6319, "step": 400} +{"lr": 9.996728583999318e-05, "data_time": 0.627727723121643, "loss": 0.5164570271968841, "time": 0.7681180000305176, "epoch": 1, "memory": 6319, "step": 500} +{"lr": 9.995286265473058e-05, "data_time": 0.6471993923187256, "loss": 0.5024662017822266, "time": 0.7867578506469727, "epoch": 1, "memory": 6319, "step": 600} +{"lr": 9.993581428375725e-05, "data_time": 0.5948776245117188, "loss": 0.476534429192543, "time": 0.735540246963501, "epoch": 1, "memory": 6319, "step": 700} +{"lr": 9.991614172267466e-05, "data_time": 0.6555254459381104, "loss": 0.45595123767852785, "time": 0.7956423044204712, "epoch": 1, "memory": 6319, "step": 800} +{"lr": 9.989384612033355e-05, "data_time": 0.634600830078125, "loss": 0.4433297276496887, "time": 0.7757574558258057, "epoch": 1, "memory": 6319, "step": 900} +{"lr": 9.986892877876658e-05, "data_time": 0.567132544517517, "loss": 0.44384285509586335, "time": 0.7062573671340943, "epoch": 1, "memory": 6319, "step": 1000} +{"lr": 9.984139115311244e-05, "data_time": 0.599405312538147, "loss": 0.4014589101076126, "time": 0.7406479120254517, "epoch": 1, "memory": 6319, "step": 1100} +{"lr": 9.981123485153081e-05, "data_time": 0.6274431943893433, "loss": 0.4095339685678482, "time": 0.7685052156448364, "epoch": 1, "memory": 6319, "step": 1200} +{"lr": 9.977846163510851e-05, "data_time": 0.5607189416885376, "loss": 0.4167290240526199, "time": 0.7007149219512939, "epoch": 1, "memory": 6319, "step": 1300} +{"lr": 9.974307341775653e-05, "data_time": 0.5535310268402099, "loss": 0.3922219753265381, "time": 0.6939513444900512, "epoch": 1, "memory": 6319, "step": 1400} +{"lr": 9.970507226609833e-05, "data_time": 0.6324576377868653, "loss": 0.39867656528949735, "time": 0.7724632501602173, "epoch": 1, "memory": 6319, "step": 1500} +{"lr": 9.966446039934934e-05, "data_time": 0.7027170658111572, "loss": 0.38865671753883363, "time": 0.8433010816574097, "epoch": 1, "memory": 6319, "step": 1600} +{"lr": 9.962124018918683e-05, "data_time": 0.6097376108169555, "loss": 0.38005087077617644, "time": 0.7507309436798095, "epoch": 1, "memory": 6319, "step": 1700} +{"lr": 9.957541415961212e-05, "data_time": 0.6780000448226928, "loss": 0.3909699708223343, "time": 0.8186037063598632, "epoch": 1, "memory": 6319, "step": 1800} +{"lr": 9.952698498680267e-05, "data_time": 0.642765212059021, "loss": 0.37944130003452303, "time": 0.7826313734054565, "epoch": 1, "memory": 6319, "step": 1900} +{"lr": 9.947595549895583e-05, "data_time": 0.5656037092208862, "loss": 0.36535309851169584, "time": 0.7060959815979004, "epoch": 1, "memory": 6319, "step": 2000} +{"lr": 9.942232867612417e-05, "data_time": 0.6045804738998413, "loss": 0.35787458121776583, "time": 0.7453924417495728, "epoch": 1, "memory": 6319, "step": 2100} +{"lr": 9.93661076500406e-05, "data_time": 0.7028562545776367, "loss": 0.3606951117515564, "time": 0.8435119152069092, "epoch": 1, "memory": 6319, "step": 2200} +{"lr": 9.930729570393622e-05, "data_time": 0.6264518022537231, "loss": 0.3666966199874878, "time": 0.7672350406646729, "epoch": 1, "memory": 6319, "step": 2300} +{"lr": 9.924589627234824e-05, "data_time": 0.5447150945663453, "loss": 0.33944440484046934, "time": 0.6848401308059693, "epoch": 1, "memory": 6319, "step": 2400} +{"lr": 9.918191294091956e-05, "data_time": 0.5785748958587646, "loss": 0.35007711946964265, "time": 0.717919135093689, "epoch": 1, "memory": 6319, "step": 2500} +{"lr": 9.91153494461892e-05, "data_time": 0.5806029558181762, "loss": 0.37237260341644285, "time": 0.7214610338211059, "epoch": 1, "memory": 6319, "step": 2600} +{"lr": 9.904620967537422e-05, "data_time": 0.5273166179656983, "loss": 0.3433519512414932, "time": 0.667465353012085, "epoch": 1, "memory": 6319, "step": 2700} diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_122552/20230605_122552.log b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_122552/20230605_122552.log new file mode 100644 index 0000000000000000000000000000000000000000..a00b6c727100c6e18ac625675d1f0f274916ad48 --- /dev/null +++ b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_122552/20230605_122552.log @@ -0,0 +1,999 @@ +2023/06/05 12:25:56 - mmengine - INFO - +------------------------------------------------------------ +System environment: + sys.platform: linux + Python: 3.10.9 (main, Mar 8 2023, 10:47:38) [GCC 11.2.0] + CUDA available: True + numpy_random_seed: 2082862108 + GPU 0,1: NVIDIA A100-SXM4-80GB + CUDA_HOME: /mnt/petrelfs/share/cuda-11.6 + NVCC: Cuda compilation tools, release 11.6, V11.6.124 + GCC: gcc (GCC) 7.5.0 + PyTorch: 1.13.1 + PyTorch compiling details: PyTorch built with: + - GCC 9.3 + - C++ Version: 201402 + - Intel(R) oneAPI Math Kernel Library Version 2021.4-Product Build 20210904 for Intel(R) 64 architecture applications + - Intel(R) MKL-DNN v2.6.0 (Git Hash 52b5f107dd9cf10910aaa19cb47f3abf9b349815) + - OpenMP 201511 (a.k.a. OpenMP 4.5) + - LAPACK is enabled (usually provided by MKL) + - NNPACK is enabled + - CPU capability usage: AVX2 + - CUDA Runtime 11.6 + - NVCC architecture flags: -gencode;arch=compute_37,code=sm_37;-gencode;arch=compute_50,code=sm_50;-gencode;arch=compute_60,code=sm_60;-gencode;arch=compute_61,code=sm_61;-gencode;arch=compute_70,code=sm_70;-gencode;arch=compute_75,code=sm_75;-gencode;arch=compute_80,code=sm_80;-gencode;arch=compute_86,code=sm_86;-gencode;arch=compute_37,code=compute_37 + - CuDNN 8.3.2 (built against CUDA 11.5) + - Magma 2.6.1 + - Build settings: BLAS_INFO=mkl, BUILD_TYPE=Release, CUDA_VERSION=11.6, CUDNN_VERSION=8.3.2, CXX_COMPILER=/opt/rh/devtoolset-9/root/usr/bin/c++, CXX_FLAGS= -fabi-version=11 -Wno-deprecated -fvisibility-inlines-hidden -DUSE_PTHREADPOOL -fopenmp -DNDEBUG -DUSE_KINETO -DUSE_FBGEMM -DUSE_QNNPACK -DUSE_PYTORCH_QNNPACK -DUSE_XNNPACK -DSYMBOLICATE_MOBILE_DEBUG_HANDLE -DEDGE_PROFILER_USE_KINETO -O2 -fPIC -Wno-narrowing -Wall -Wextra -Werror=return-type -Werror=non-virtual-dtor -Wno-missing-field-initializers -Wno-type-limits -Wno-array-bounds -Wno-unknown-pragmas -Wunused-local-typedefs -Wno-unused-parameter -Wno-unused-function -Wno-unused-result -Wno-strict-overflow -Wno-strict-aliasing -Wno-error=deprecated-declarations -Wno-stringop-overflow -Wno-psabi -Wno-error=pedantic -Wno-error=redundant-decls -Wno-error=old-style-cast -fdiagnostics-color=always -faligned-new -Wno-unused-but-set-variable -Wno-maybe-uninitialized -fno-math-errno -fno-trapping-math -Werror=format -Werror=cast-function-type -Wno-stringop-overflow, LAPACK_INFO=mkl, PERF_WITH_AVX=1, PERF_WITH_AVX2=1, PERF_WITH_AVX512=1, TORCH_VERSION=1.13.1, USE_CUDA=ON, USE_CUDNN=ON, USE_EXCEPTION_PTR=1, USE_GFLAGS=OFF, USE_GLOG=OFF, USE_MKL=ON, USE_MKLDNN=ON, USE_MPI=OFF, USE_NCCL=ON, USE_NNPACK=ON, USE_OPENMP=ON, USE_ROCM=OFF, + + TorchVision: 0.14.1 + OpenCV: 4.7.0 + MMEngine: 0.7.3 + +Runtime environment: + cudnn_benchmark: True + mp_cfg: {'mp_start_method': 'fork', 'opencv_num_threads': 0} + dist_cfg: {'backend': 'nccl'} + seed: None + deterministic: False + Distributed launcher: slurm + Distributed training: True + GPU number: 2 +------------------------------------------------------------ + +2023/06/05 12:26:01 - mmengine - INFO - Config: +optim_wrapper = dict( + optimizer=dict( + type='SGD', + lr=0.0001, + momentum=0.9, + weight_decay=0.0001, + _scope_='mmpretrain'), + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=False, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=512) +model = dict( + type='ImageClassifier', + backbone=dict( + frozen_stages=2, + type='ResNet', + depth=50, + num_stages=4, + out_indices=(3, ), + style='pytorch', + init_cfg=dict( + type='Pretrained', + checkpoint= + 'https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth', + prefix='backbone')), + neck=dict(type='GlobalAveragePooling'), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=2048, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + topk=1)) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/all1.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/all2.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/all3.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3fake8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/cc1m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3real8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV2-1-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/if-dpmsolver++-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = dict(type='Accuracy', topk=1) +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV2-1-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/if-dpmsolver++-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = dict(type='Accuracy', topk=1) +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1' + +2023/06/05 12:26:16 - mmengine - INFO - Hooks will be executed in the following order: +before_run: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_load_checkpoint: +(ABOVE_NORMAL) EMAHook + -------------------- +before_train: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_train_epoch: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(NORMAL ) DistSamplerSeedHook + -------------------- +before_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook + -------------------- +after_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +after_train_epoch: +(NORMAL ) IterTimerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_val_epoch: +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook + -------------------- +before_val_iter: +(NORMAL ) IterTimerHook + -------------------- +after_val_iter: +(NORMAL ) IterTimerHook +(NORMAL ) VisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_val_epoch: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_save_checkpoint: +(ABOVE_NORMAL) EMAHook + -------------------- +after_train: +(VERY_LOW ) CheckpointHook + -------------------- +before_test_epoch: +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook + -------------------- +before_test_iter: +(NORMAL ) IterTimerHook + -------------------- +after_test_iter: +(NORMAL ) IterTimerHook +(NORMAL ) VisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_test_epoch: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_run: +(BELOW_NORMAL) LoggerHook + -------------------- +2023/06/05 12:26:38 - mmengine - INFO - load backbone in model from: https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth +Name of parameter - Initialization information + +backbone.conv1.weight - torch.Size([64, 3, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.bn1.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.bn1.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.conv1.weight - torch.Size([64, 64, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.bn1.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.bn1.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.conv2.weight - torch.Size([64, 64, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.bn2.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.bn2.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.conv3.weight - torch.Size([256, 64, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.bn3.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.bn3.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.downsample.0.weight - torch.Size([256, 64, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.downsample.1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.downsample.1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.conv1.weight - torch.Size([64, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.bn1.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.bn1.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.conv2.weight - torch.Size([64, 64, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.bn2.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.bn2.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.conv3.weight - torch.Size([256, 64, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.bn3.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.bn3.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.conv1.weight - torch.Size([64, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.bn1.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.bn1.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.conv2.weight - torch.Size([64, 64, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.bn2.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.bn2.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.conv3.weight - torch.Size([256, 64, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.bn3.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.bn3.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.conv1.weight - torch.Size([128, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.bn1.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.bn1.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.conv2.weight - torch.Size([128, 128, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.bn2.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.bn2.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.conv3.weight - torch.Size([512, 128, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.bn3.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.bn3.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.downsample.0.weight - torch.Size([512, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.downsample.1.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.downsample.1.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.conv1.weight - torch.Size([128, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.bn1.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.bn1.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.conv2.weight - torch.Size([128, 128, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.bn2.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.bn2.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.conv3.weight - torch.Size([512, 128, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.bn3.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.bn3.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.conv1.weight - torch.Size([128, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.bn1.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.bn1.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.conv2.weight - torch.Size([128, 128, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.bn2.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.bn2.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.conv3.weight - torch.Size([512, 128, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.bn3.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.bn3.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.conv1.weight - torch.Size([128, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.bn1.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.bn1.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.conv2.weight - torch.Size([128, 128, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.bn2.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.bn2.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.conv3.weight - torch.Size([512, 128, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.bn3.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.bn3.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.conv1.weight - torch.Size([256, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.bn1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.bn1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.conv2.weight - torch.Size([256, 256, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.bn2.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.bn2.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.conv3.weight - torch.Size([1024, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.bn3.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.bn3.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.downsample.0.weight - torch.Size([1024, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.downsample.1.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.downsample.1.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.conv1.weight - torch.Size([256, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.bn1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.bn1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.conv2.weight - torch.Size([256, 256, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.bn2.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.bn2.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.conv3.weight - torch.Size([1024, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.bn3.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.bn3.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.conv1.weight - torch.Size([256, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.bn1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.bn1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.conv2.weight - torch.Size([256, 256, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.bn2.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.bn2.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.conv3.weight - torch.Size([1024, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.bn3.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.bn3.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.conv1.weight - torch.Size([256, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.bn1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.bn1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.conv2.weight - torch.Size([256, 256, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.bn2.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.bn2.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.conv3.weight - torch.Size([1024, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.bn3.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.bn3.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.conv1.weight - torch.Size([256, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.bn1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.bn1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.conv2.weight - torch.Size([256, 256, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.bn2.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.bn2.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.conv3.weight - torch.Size([1024, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.bn3.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.bn3.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.conv1.weight - torch.Size([256, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.bn1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.bn1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.conv2.weight - torch.Size([256, 256, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.bn2.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.bn2.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.conv3.weight - torch.Size([1024, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.bn3.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.bn3.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.conv1.weight - torch.Size([512, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.bn1.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.bn1.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.conv2.weight - torch.Size([512, 512, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.bn2.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.bn2.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.conv3.weight - torch.Size([2048, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.bn3.weight - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.bn3.bias - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.downsample.0.weight - torch.Size([2048, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.downsample.1.weight - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.downsample.1.bias - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.conv1.weight - torch.Size([512, 2048, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.bn1.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.bn1.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.conv2.weight - torch.Size([512, 512, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.bn2.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.bn2.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.conv3.weight - torch.Size([2048, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.bn3.weight - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.bn3.bias - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.conv1.weight - torch.Size([512, 2048, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.bn1.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.bn1.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.conv2.weight - torch.Size([512, 512, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.bn2.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.bn2.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.conv3.weight - torch.Size([2048, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.bn3.weight - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.bn3.bias - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +head.fc.weight - torch.Size([2, 2048]): +NormalInit: mean=0, std=0.01, bias=0 + +head.fc.bias - torch.Size([2]): +NormalInit: mean=0, std=0.01, bias=0 +2023/06/05 12:26:39 - mmengine - WARNING - "FileClient" will be deprecated in future. Please use io functions in https://mmengine.readthedocs.io/en/latest/api/fileio.html#file-io +2023/06/05 12:26:39 - mmengine - WARNING - "HardDiskBackend" is the alias of "LocalBackend" and the former will be deprecated in future. +2023/06/05 12:26:39 - mmengine - INFO - Checkpoints will be saved to /mnt/petrelfs/luzeyu/workspace/fakebench/mmpretrain/workdir/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1. +2023/06/05 12:28:02 - mmengine - INFO - Epoch(train) [1][ 100/4111] lr: 9.9999e-05 eta: 9:29:52 time: 0.7207 data_time: 0.1963 memory: 9436 loss: 0.6488 +2023/06/05 12:29:19 - mmengine - INFO - Epoch(train) [1][ 200/4111] lr: 9.9995e-05 eta: 9:06:54 time: 0.7048 data_time: 0.0009 memory: 6319 loss: 0.5979 +2023/06/05 12:30:35 - mmengine - INFO - Epoch(train) [1][ 300/4111] lr: 9.9988e-05 eta: 8:56:48 time: 0.7198 data_time: 0.0009 memory: 6319 loss: 0.5720 +2023/06/05 12:31:51 - mmengine - INFO - Epoch(train) [1][ 400/4111] lr: 9.9979e-05 eta: 8:49:14 time: 0.7277 data_time: 0.0009 memory: 6319 loss: 0.5428 +2023/06/05 12:33:13 - mmengine - INFO - Epoch(train) [1][ 500/4111] lr: 9.9967e-05 eta: 8:53:49 time: 0.8745 data_time: 0.0008 memory: 6319 loss: 0.5163 +2023/06/05 12:34:31 - mmengine - INFO - Epoch(train) [1][ 600/4111] lr: 9.9953e-05 eta: 8:51:39 time: 1.0020 data_time: 0.0008 memory: 6319 loss: 0.4937 +2023/06/05 12:35:48 - mmengine - INFO - Epoch(train) [1][ 700/4111] lr: 9.9936e-05 eta: 8:48:21 time: 0.7460 data_time: 0.0010 memory: 6319 loss: 0.4777 +2023/06/05 12:37:05 - mmengine - INFO - Epoch(train) [1][ 800/4111] lr: 9.9916e-05 eta: 8:46:19 time: 0.7673 data_time: 0.0008 memory: 6319 loss: 0.4659 +2023/06/05 12:38:26 - mmengine - INFO - Epoch(train) [1][ 900/4111] lr: 9.9894e-05 eta: 8:46:29 time: 0.7225 data_time: 0.0008 memory: 6319 loss: 0.4472 +2023/06/05 12:39:43 - mmengine - INFO - Exp name: resnet50_2xb256_all_1m_lr1e-4_aug_5e-1_20230605_122552 +2023/06/05 12:39:43 - mmengine - INFO - Epoch(train) [1][1000/4111] lr: 9.9869e-05 eta: 8:44:01 time: 0.7457 data_time: 0.0008 memory: 6319 loss: 0.4317 +2023/06/05 12:40:59 - mmengine - INFO - Epoch(train) [1][1100/4111] lr: 9.9841e-05 eta: 8:41:36 time: 0.7586 data_time: 0.0007 memory: 6319 loss: 0.4348 +2023/06/05 12:42:17 - mmengine - INFO - Epoch(train) [1][1200/4111] lr: 9.9811e-05 eta: 8:40:19 time: 0.8087 data_time: 0.0009 memory: 6319 loss: 0.4089 +2023/06/05 12:43:34 - mmengine - INFO - Epoch(train) [1][1300/4111] lr: 9.9778e-05 eta: 8:38:19 time: 0.8170 data_time: 0.0007 memory: 6319 loss: 0.4063 +2023/06/05 12:44:52 - mmengine - INFO - Epoch(train) [1][1400/4111] lr: 9.9743e-05 eta: 8:36:51 time: 0.8277 data_time: 0.0007 memory: 6319 loss: 0.3895 +2023/06/05 12:46:10 - mmengine - INFO - Epoch(train) [1][1500/4111] lr: 9.9705e-05 eta: 8:35:36 time: 0.7795 data_time: 0.0011 memory: 6319 loss: 0.4049 +2023/06/05 12:47:28 - mmengine - INFO - Epoch(train) [1][1600/4111] lr: 9.9664e-05 eta: 8:34:19 time: 0.8086 data_time: 0.0007 memory: 6319 loss: 0.3747 diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_122552/vis_data/20230605_122552.json b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_122552/vis_data/20230605_122552.json new file mode 100644 index 0000000000000000000000000000000000000000..782161767ed67c0931c9d2259113b26ade7e1f7b --- /dev/null +++ b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_122552/vis_data/20230605_122552.json @@ -0,0 +1,16 @@ +{"lr": 9.999871217862936e-05, "data_time": 0.19629251956939697, "loss": 0.6487768113613128, "time": 0.7206579446792603, "epoch": 1, "memory": 9436, "step": 100} +{"lr": 9.999479662539605e-05, "data_time": 0.0008574724197387695, "loss": 0.597885936498642, "time": 0.704778528213501, "epoch": 1, "memory": 6319, "step": 200} +{"lr": 9.99882534375654e-05, "data_time": 0.0008840084075927735, "loss": 0.5719808757305145, "time": 0.7198325157165527, "epoch": 1, "memory": 6319, "step": 300} +{"lr": 9.997908299725075e-05, "data_time": 0.0008769989013671875, "loss": 0.5428020298480988, "time": 0.7276598453521729, "epoch": 1, "memory": 6319, "step": 400} +{"lr": 9.996728583999318e-05, "data_time": 0.0008464336395263671, "loss": 0.5163365423679351, "time": 0.8745182514190674, "epoch": 1, "memory": 6319, "step": 500} +{"lr": 9.995286265473058e-05, "data_time": 0.0008020639419555664, "loss": 0.4937333583831787, "time": 1.00198655128479, "epoch": 1, "memory": 6319, "step": 600} +{"lr": 9.993581428375725e-05, "data_time": 0.0009988784790039063, "loss": 0.47774254381656645, "time": 0.7459773540496826, "epoch": 1, "memory": 6319, "step": 700} +{"lr": 9.991614172267466e-05, "data_time": 0.0007680654525756836, "loss": 0.46589950323104856, "time": 0.7673243045806885, "epoch": 1, "memory": 6319, "step": 800} +{"lr": 9.989384612033355e-05, "data_time": 0.0008303403854370117, "loss": 0.44720605909824374, "time": 0.7224833965301514, "epoch": 1, "memory": 6319, "step": 900} +{"lr": 9.986892877876658e-05, "data_time": 0.000790262222290039, "loss": 0.4317399799823761, "time": 0.7456749677658081, "epoch": 1, "memory": 6319, "step": 1000} +{"lr": 9.984139115311244e-05, "data_time": 0.0007363080978393555, "loss": 0.43475731909275056, "time": 0.7585577964782715, "epoch": 1, "memory": 6319, "step": 1100} +{"lr": 9.981123485153081e-05, "data_time": 0.0009210824966430664, "loss": 0.4089220702648163, "time": 0.8086841106414795, "epoch": 1, "memory": 6319, "step": 1200} +{"lr": 9.977846163510851e-05, "data_time": 0.0007385492324829101, "loss": 0.40625119507312774, "time": 0.8170331001281739, "epoch": 1, "memory": 6319, "step": 1300} +{"lr": 9.974307341775653e-05, "data_time": 0.0007062673568725586, "loss": 0.3894824951887131, "time": 0.8277269601821899, "epoch": 1, "memory": 6319, "step": 1400} +{"lr": 9.970507226609833e-05, "data_time": 0.0011449098587036134, "loss": 0.4049314379692078, "time": 0.7794937133789063, "epoch": 1, "memory": 6319, "step": 1500} +{"lr": 9.966446039934934e-05, "data_time": 0.0007276535034179688, "loss": 0.3747121334075928, "time": 0.8086159944534301, "epoch": 1, "memory": 6319, "step": 1600} diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_122552/vis_data/config.py b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_122552/vis_data/config.py new file mode 100644 index 0000000000000000000000000000000000000000..b35ee250106aa84d1065073f108965cfa202c34d --- /dev/null +++ b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_122552/vis_data/config.py @@ -0,0 +1,366 @@ +optim_wrapper = dict( + optimizer=dict( + type='SGD', + lr=0.0001, + momentum=0.9, + weight_decay=0.0001, + _scope_='mmpretrain'), + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=False, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=512) +model = dict( + type='ImageClassifier', + backbone=dict( + frozen_stages=2, + type='ResNet', + depth=50, + num_stages=4, + out_indices=(3, ), + style='pytorch', + init_cfg=dict( + type='Pretrained', + checkpoint= + 'https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth', + prefix='backbone')), + neck=dict(type='GlobalAveragePooling'), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=2048, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + topk=1)) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/all1.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/all2.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/all3.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3fake8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/cc1m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3real8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV2-1-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/if-dpmsolver++-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = dict(type='Accuracy', topk=1) +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV2-1-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/if-dpmsolver++-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = dict(type='Accuracy', topk=1) +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1' diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_122552/vis_data/events.out.tfevents.1685939169.SH-IDC1-10-140-24-105.190930.0 b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_122552/vis_data/events.out.tfevents.1685939169.SH-IDC1-10-140-24-105.190930.0 new file mode 100644 index 0000000000000000000000000000000000000000..fa7f0b5146562003b97a14af60f0d85639a93d06 --- /dev/null +++ b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_122552/vis_data/events.out.tfevents.1685939169.SH-IDC1-10-140-24-105.190930.0 @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:4db33a3ff53d6306c12113e1af254fb86f03a0a2ee6e8d7cf23bfe63a31b11a9 +size 19541 diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_122552/vis_data/scalars.json b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_122552/vis_data/scalars.json new file mode 100644 index 0000000000000000000000000000000000000000..782161767ed67c0931c9d2259113b26ade7e1f7b --- /dev/null +++ b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_122552/vis_data/scalars.json @@ -0,0 +1,16 @@ +{"lr": 9.999871217862936e-05, "data_time": 0.19629251956939697, "loss": 0.6487768113613128, "time": 0.7206579446792603, "epoch": 1, "memory": 9436, "step": 100} +{"lr": 9.999479662539605e-05, "data_time": 0.0008574724197387695, "loss": 0.597885936498642, "time": 0.704778528213501, "epoch": 1, "memory": 6319, "step": 200} +{"lr": 9.99882534375654e-05, "data_time": 0.0008840084075927735, "loss": 0.5719808757305145, "time": 0.7198325157165527, "epoch": 1, "memory": 6319, "step": 300} +{"lr": 9.997908299725075e-05, "data_time": 0.0008769989013671875, "loss": 0.5428020298480988, "time": 0.7276598453521729, "epoch": 1, "memory": 6319, "step": 400} +{"lr": 9.996728583999318e-05, "data_time": 0.0008464336395263671, "loss": 0.5163365423679351, "time": 0.8745182514190674, "epoch": 1, "memory": 6319, "step": 500} +{"lr": 9.995286265473058e-05, "data_time": 0.0008020639419555664, "loss": 0.4937333583831787, "time": 1.00198655128479, "epoch": 1, "memory": 6319, "step": 600} +{"lr": 9.993581428375725e-05, "data_time": 0.0009988784790039063, "loss": 0.47774254381656645, "time": 0.7459773540496826, "epoch": 1, "memory": 6319, "step": 700} +{"lr": 9.991614172267466e-05, "data_time": 0.0007680654525756836, "loss": 0.46589950323104856, "time": 0.7673243045806885, "epoch": 1, "memory": 6319, "step": 800} +{"lr": 9.989384612033355e-05, "data_time": 0.0008303403854370117, "loss": 0.44720605909824374, "time": 0.7224833965301514, "epoch": 1, "memory": 6319, "step": 900} +{"lr": 9.986892877876658e-05, "data_time": 0.000790262222290039, "loss": 0.4317399799823761, "time": 0.7456749677658081, "epoch": 1, "memory": 6319, "step": 1000} +{"lr": 9.984139115311244e-05, "data_time": 0.0007363080978393555, "loss": 0.43475731909275056, "time": 0.7585577964782715, "epoch": 1, "memory": 6319, "step": 1100} +{"lr": 9.981123485153081e-05, "data_time": 0.0009210824966430664, "loss": 0.4089220702648163, "time": 0.8086841106414795, "epoch": 1, "memory": 6319, "step": 1200} +{"lr": 9.977846163510851e-05, "data_time": 0.0007385492324829101, "loss": 0.40625119507312774, "time": 0.8170331001281739, "epoch": 1, "memory": 6319, "step": 1300} +{"lr": 9.974307341775653e-05, "data_time": 0.0007062673568725586, "loss": 0.3894824951887131, "time": 0.8277269601821899, "epoch": 1, "memory": 6319, "step": 1400} +{"lr": 9.970507226609833e-05, "data_time": 0.0011449098587036134, "loss": 0.4049314379692078, "time": 0.7794937133789063, "epoch": 1, "memory": 6319, "step": 1500} +{"lr": 9.966446039934934e-05, "data_time": 0.0007276535034179688, "loss": 0.3747121334075928, "time": 0.8086159944534301, "epoch": 1, "memory": 6319, "step": 1600} diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_125106/20230605_125106.log b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_125106/20230605_125106.log new file mode 100644 index 0000000000000000000000000000000000000000..15e8e064ec9bd7190493e06e308492cc504c562b --- /dev/null +++ b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_125106/20230605_125106.log @@ -0,0 +1,984 @@ +2023/06/05 12:51:10 - mmengine - INFO - +------------------------------------------------------------ +System environment: + sys.platform: linux + Python: 3.10.9 (main, Mar 8 2023, 10:47:38) [GCC 11.2.0] + CUDA available: True + numpy_random_seed: 481121810 + GPU 0,1: NVIDIA A100-SXM4-80GB + CUDA_HOME: /mnt/petrelfs/share/cuda-11.6 + NVCC: Cuda compilation tools, release 11.6, V11.6.124 + GCC: gcc (GCC) 7.5.0 + PyTorch: 1.13.1 + PyTorch compiling details: PyTorch built with: + - GCC 9.3 + - C++ Version: 201402 + - Intel(R) oneAPI Math Kernel Library Version 2021.4-Product Build 20210904 for Intel(R) 64 architecture applications + - Intel(R) MKL-DNN v2.6.0 (Git Hash 52b5f107dd9cf10910aaa19cb47f3abf9b349815) + - OpenMP 201511 (a.k.a. OpenMP 4.5) + - LAPACK is enabled (usually provided by MKL) + - NNPACK is enabled + - CPU capability usage: AVX2 + - CUDA Runtime 11.6 + - NVCC architecture flags: -gencode;arch=compute_37,code=sm_37;-gencode;arch=compute_50,code=sm_50;-gencode;arch=compute_60,code=sm_60;-gencode;arch=compute_61,code=sm_61;-gencode;arch=compute_70,code=sm_70;-gencode;arch=compute_75,code=sm_75;-gencode;arch=compute_80,code=sm_80;-gencode;arch=compute_86,code=sm_86;-gencode;arch=compute_37,code=compute_37 + - CuDNN 8.3.2 (built against CUDA 11.5) + - Magma 2.6.1 + - Build settings: BLAS_INFO=mkl, BUILD_TYPE=Release, CUDA_VERSION=11.6, CUDNN_VERSION=8.3.2, CXX_COMPILER=/opt/rh/devtoolset-9/root/usr/bin/c++, CXX_FLAGS= -fabi-version=11 -Wno-deprecated -fvisibility-inlines-hidden -DUSE_PTHREADPOOL -fopenmp -DNDEBUG -DUSE_KINETO -DUSE_FBGEMM -DUSE_QNNPACK -DUSE_PYTORCH_QNNPACK -DUSE_XNNPACK -DSYMBOLICATE_MOBILE_DEBUG_HANDLE -DEDGE_PROFILER_USE_KINETO -O2 -fPIC -Wno-narrowing -Wall -Wextra -Werror=return-type -Werror=non-virtual-dtor -Wno-missing-field-initializers -Wno-type-limits -Wno-array-bounds -Wno-unknown-pragmas -Wunused-local-typedefs -Wno-unused-parameter -Wno-unused-function -Wno-unused-result -Wno-strict-overflow -Wno-strict-aliasing -Wno-error=deprecated-declarations -Wno-stringop-overflow -Wno-psabi -Wno-error=pedantic -Wno-error=redundant-decls -Wno-error=old-style-cast -fdiagnostics-color=always -faligned-new -Wno-unused-but-set-variable -Wno-maybe-uninitialized -fno-math-errno -fno-trapping-math -Werror=format -Werror=cast-function-type -Wno-stringop-overflow, LAPACK_INFO=mkl, PERF_WITH_AVX=1, PERF_WITH_AVX2=1, PERF_WITH_AVX512=1, TORCH_VERSION=1.13.1, USE_CUDA=ON, USE_CUDNN=ON, USE_EXCEPTION_PTR=1, USE_GFLAGS=OFF, USE_GLOG=OFF, USE_MKL=ON, USE_MKLDNN=ON, USE_MPI=OFF, USE_NCCL=ON, USE_NNPACK=ON, USE_OPENMP=ON, USE_ROCM=OFF, + + TorchVision: 0.14.1 + OpenCV: 4.7.0 + MMEngine: 0.7.3 + +Runtime environment: + cudnn_benchmark: True + mp_cfg: {'mp_start_method': 'fork', 'opencv_num_threads': 0} + dist_cfg: {'backend': 'nccl'} + seed: None + deterministic: False + Distributed launcher: slurm + Distributed training: True + GPU number: 2 +------------------------------------------------------------ + +2023/06/05 12:51:15 - mmengine - INFO - Config: +optim_wrapper = dict( + optimizer=dict( + type='SGD', + lr=0.0001, + momentum=0.9, + weight_decay=0.0001, + _scope_='mmpretrain'), + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=False, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=512) +model = dict( + type='ImageClassifier', + backbone=dict( + frozen_stages=2, + type='ResNet', + depth=50, + num_stages=4, + out_indices=(3, ), + style='pytorch', + init_cfg=dict( + type='Pretrained', + checkpoint= + 'https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth', + prefix='backbone')), + neck=dict(type='GlobalAveragePooling'), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=2048, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + topk=1)) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/all1.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/all2.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/all3.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3fake8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/cc1m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3real8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV2-1-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/if-dpmsolver++-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = dict(type='Accuracy', topk=1) +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV2-1-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/if-dpmsolver++-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = dict(type='Accuracy', topk=1) +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1' + +2023/06/05 12:51:27 - mmengine - INFO - Hooks will be executed in the following order: +before_run: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_load_checkpoint: +(ABOVE_NORMAL) EMAHook + -------------------- +before_train: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_train_epoch: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(NORMAL ) DistSamplerSeedHook + -------------------- +before_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook + -------------------- +after_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +after_train_epoch: +(NORMAL ) IterTimerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_val_epoch: +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook + -------------------- +before_val_iter: +(NORMAL ) IterTimerHook + -------------------- +after_val_iter: +(NORMAL ) IterTimerHook +(NORMAL ) VisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_val_epoch: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_save_checkpoint: +(ABOVE_NORMAL) EMAHook + -------------------- +after_train: +(VERY_LOW ) CheckpointHook + -------------------- +before_test_epoch: +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook + -------------------- +before_test_iter: +(NORMAL ) IterTimerHook + -------------------- +after_test_iter: +(NORMAL ) IterTimerHook +(NORMAL ) VisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_test_epoch: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_run: +(BELOW_NORMAL) LoggerHook + -------------------- +2023/06/05 12:51:53 - mmengine - INFO - load backbone in model from: https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth +Name of parameter - Initialization information + +backbone.conv1.weight - torch.Size([64, 3, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.bn1.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.bn1.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.conv1.weight - torch.Size([64, 64, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.bn1.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.bn1.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.conv2.weight - torch.Size([64, 64, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.bn2.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.bn2.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.conv3.weight - torch.Size([256, 64, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.bn3.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.bn3.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.downsample.0.weight - torch.Size([256, 64, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.downsample.1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.downsample.1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.conv1.weight - torch.Size([64, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.bn1.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.bn1.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.conv2.weight - torch.Size([64, 64, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.bn2.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.bn2.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.conv3.weight - torch.Size([256, 64, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.bn3.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.bn3.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.conv1.weight - torch.Size([64, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.bn1.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.bn1.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.conv2.weight - torch.Size([64, 64, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.bn2.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.bn2.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.conv3.weight - torch.Size([256, 64, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.bn3.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.bn3.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.conv1.weight - torch.Size([128, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.bn1.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.bn1.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.conv2.weight - torch.Size([128, 128, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.bn2.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.bn2.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.conv3.weight - torch.Size([512, 128, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.bn3.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.bn3.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.downsample.0.weight - torch.Size([512, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.downsample.1.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.downsample.1.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.conv1.weight - torch.Size([128, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.bn1.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.bn1.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.conv2.weight - torch.Size([128, 128, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.bn2.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.bn2.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.conv3.weight - torch.Size([512, 128, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.bn3.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.bn3.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.conv1.weight - torch.Size([128, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.bn1.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.bn1.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.conv2.weight - torch.Size([128, 128, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.bn2.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.bn2.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.conv3.weight - torch.Size([512, 128, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.bn3.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.bn3.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.conv1.weight - torch.Size([128, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.bn1.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.bn1.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.conv2.weight - torch.Size([128, 128, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.bn2.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.bn2.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.conv3.weight - torch.Size([512, 128, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.bn3.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.bn3.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.conv1.weight - torch.Size([256, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.bn1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.bn1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.conv2.weight - torch.Size([256, 256, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.bn2.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.bn2.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.conv3.weight - torch.Size([1024, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.bn3.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.bn3.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.downsample.0.weight - torch.Size([1024, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.downsample.1.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.downsample.1.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.conv1.weight - torch.Size([256, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.bn1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.bn1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.conv2.weight - torch.Size([256, 256, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.bn2.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.bn2.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.conv3.weight - torch.Size([1024, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.bn3.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.bn3.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.conv1.weight - torch.Size([256, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.bn1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.bn1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.conv2.weight - torch.Size([256, 256, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.bn2.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.bn2.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.conv3.weight - torch.Size([1024, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.bn3.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.bn3.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.conv1.weight - torch.Size([256, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.bn1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.bn1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.conv2.weight - torch.Size([256, 256, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.bn2.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.bn2.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.conv3.weight - torch.Size([1024, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.bn3.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.bn3.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.conv1.weight - torch.Size([256, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.bn1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.bn1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.conv2.weight - torch.Size([256, 256, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.bn2.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.bn2.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.conv3.weight - torch.Size([1024, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.bn3.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.bn3.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.conv1.weight - torch.Size([256, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.bn1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.bn1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.conv2.weight - torch.Size([256, 256, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.bn2.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.bn2.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.conv3.weight - torch.Size([1024, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.bn3.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.bn3.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.conv1.weight - torch.Size([512, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.bn1.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.bn1.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.conv2.weight - torch.Size([512, 512, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.bn2.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.bn2.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.conv3.weight - torch.Size([2048, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.bn3.weight - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.bn3.bias - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.downsample.0.weight - torch.Size([2048, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.downsample.1.weight - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.downsample.1.bias - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.conv1.weight - torch.Size([512, 2048, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.bn1.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.bn1.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.conv2.weight - torch.Size([512, 512, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.bn2.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.bn2.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.conv3.weight - torch.Size([2048, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.bn3.weight - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.bn3.bias - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.conv1.weight - torch.Size([512, 2048, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.bn1.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.bn1.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.conv2.weight - torch.Size([512, 512, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.bn2.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.bn2.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.conv3.weight - torch.Size([2048, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.bn3.weight - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.bn3.bias - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +head.fc.weight - torch.Size([2, 2048]): +NormalInit: mean=0, std=0.01, bias=0 + +head.fc.bias - torch.Size([2]): +NormalInit: mean=0, std=0.01, bias=0 +2023/06/05 12:51:53 - mmengine - WARNING - "FileClient" will be deprecated in future. Please use io functions in https://mmengine.readthedocs.io/en/latest/api/fileio.html#file-io +2023/06/05 12:51:53 - mmengine - WARNING - "HardDiskBackend" is the alias of "LocalBackend" and the former will be deprecated in future. +2023/06/05 12:51:53 - mmengine - INFO - Checkpoints will be saved to /mnt/petrelfs/luzeyu/workspace/fakebench/mmpretrain/workdir/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1. +2023/06/05 12:53:22 - mmengine - INFO - Epoch(train) [1][ 100/4111] lr: 9.9999e-05 eta: 10:10:22 time: 0.6535 data_time: 0.1267 memory: 9436 loss: 0.6368 +2023/06/05 12:54:33 - mmengine - INFO - Epoch(train) [1][ 200/4111] lr: 9.9995e-05 eta: 9:05:25 time: 0.6773 data_time: 0.0828 memory: 6319 loss: 0.5959 diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_125106/vis_data/20230605_125106.json b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_125106/vis_data/20230605_125106.json new file mode 100644 index 0000000000000000000000000000000000000000..b981cf6c029350aa7de57bfe9615090264265616 --- /dev/null +++ b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_125106/vis_data/20230605_125106.json @@ -0,0 +1,2 @@ +{"lr": 9.999871217862936e-05, "data_time": 0.12667274475097656, "loss": 0.6367849469184875, "time": 0.6534832239151, "epoch": 1, "memory": 9436, "step": 100} +{"lr": 9.999479662539605e-05, "data_time": 0.08284163475036621, "loss": 0.5958549380302429, "time": 0.6773184061050415, "epoch": 1, "memory": 6319, "step": 200} diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_125106/vis_data/config.py b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_125106/vis_data/config.py new file mode 100644 index 0000000000000000000000000000000000000000..b35ee250106aa84d1065073f108965cfa202c34d --- /dev/null +++ b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_125106/vis_data/config.py @@ -0,0 +1,366 @@ +optim_wrapper = dict( + optimizer=dict( + type='SGD', + lr=0.0001, + momentum=0.9, + weight_decay=0.0001, + _scope_='mmpretrain'), + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=False, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=512) +model = dict( + type='ImageClassifier', + backbone=dict( + frozen_stages=2, + type='ResNet', + depth=50, + num_stages=4, + out_indices=(3, ), + style='pytorch', + init_cfg=dict( + type='Pretrained', + checkpoint= + 'https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth', + prefix='backbone')), + neck=dict(type='GlobalAveragePooling'), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=2048, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + topk=1)) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/all1.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/all2.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/all3.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3fake8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/cc1m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3real8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV2-1-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/if-dpmsolver++-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = dict(type='Accuracy', topk=1) +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV2-1-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/if-dpmsolver++-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = dict(type='Accuracy', topk=1) +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1' diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_125106/vis_data/events.out.tfevents.1685940682.SH-IDC1-10-140-24-108.67786.0 b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_125106/vis_data/events.out.tfevents.1685940682.SH-IDC1-10-140-24-108.67786.0 new file mode 100644 index 0000000000000000000000000000000000000000..a00fd8421294aa2b8f7913e77fe270fe3a8b5eee --- /dev/null +++ b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_125106/vis_data/events.out.tfevents.1685940682.SH-IDC1-10-140-24-108.67786.0 @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:27b5d0dfb06bd69199de8056ec00afae2505fa6131911242e6f29c11f4ebaf66 +size 15845 diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_125106/vis_data/scalars.json b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_125106/vis_data/scalars.json new file mode 100644 index 0000000000000000000000000000000000000000..b981cf6c029350aa7de57bfe9615090264265616 --- /dev/null +++ b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_125106/vis_data/scalars.json @@ -0,0 +1,2 @@ +{"lr": 9.999871217862936e-05, "data_time": 0.12667274475097656, "loss": 0.6367849469184875, "time": 0.6534832239151, "epoch": 1, "memory": 9436, "step": 100} +{"lr": 9.999479662539605e-05, "data_time": 0.08284163475036621, "loss": 0.5958549380302429, "time": 0.6773184061050415, "epoch": 1, "memory": 6319, "step": 200} diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_125740/20230605_125740.log b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_125740/20230605_125740.log new file mode 100644 index 0000000000000000000000000000000000000000..25cf301becfacfef702913d6017f0c8e28ae7309 --- /dev/null +++ b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_125740/20230605_125740.log @@ -0,0 +1,1108 @@ +2023/06/05 12:57:43 - mmengine - INFO - +------------------------------------------------------------ +System environment: + sys.platform: linux + Python: 3.10.9 (main, Mar 8 2023, 10:47:38) [GCC 11.2.0] + CUDA available: True + numpy_random_seed: 331852259 + GPU 0,1: NVIDIA A100-SXM4-80GB + CUDA_HOME: /mnt/petrelfs/share/cuda-11.6 + NVCC: Cuda compilation tools, release 11.6, V11.6.124 + GCC: gcc (GCC) 7.5.0 + PyTorch: 1.13.1 + PyTorch compiling details: PyTorch built with: + - GCC 9.3 + - C++ Version: 201402 + - Intel(R) oneAPI Math Kernel Library Version 2021.4-Product Build 20210904 for Intel(R) 64 architecture applications + - Intel(R) MKL-DNN v2.6.0 (Git Hash 52b5f107dd9cf10910aaa19cb47f3abf9b349815) + - OpenMP 201511 (a.k.a. OpenMP 4.5) + - LAPACK is enabled (usually provided by MKL) + - NNPACK is enabled + - CPU capability usage: AVX2 + - CUDA Runtime 11.6 + - NVCC architecture flags: -gencode;arch=compute_37,code=sm_37;-gencode;arch=compute_50,code=sm_50;-gencode;arch=compute_60,code=sm_60;-gencode;arch=compute_61,code=sm_61;-gencode;arch=compute_70,code=sm_70;-gencode;arch=compute_75,code=sm_75;-gencode;arch=compute_80,code=sm_80;-gencode;arch=compute_86,code=sm_86;-gencode;arch=compute_37,code=compute_37 + - CuDNN 8.3.2 (built against CUDA 11.5) + - Magma 2.6.1 + - Build settings: BLAS_INFO=mkl, BUILD_TYPE=Release, CUDA_VERSION=11.6, CUDNN_VERSION=8.3.2, CXX_COMPILER=/opt/rh/devtoolset-9/root/usr/bin/c++, CXX_FLAGS= -fabi-version=11 -Wno-deprecated -fvisibility-inlines-hidden -DUSE_PTHREADPOOL -fopenmp -DNDEBUG -DUSE_KINETO -DUSE_FBGEMM -DUSE_QNNPACK -DUSE_PYTORCH_QNNPACK -DUSE_XNNPACK -DSYMBOLICATE_MOBILE_DEBUG_HANDLE -DEDGE_PROFILER_USE_KINETO -O2 -fPIC -Wno-narrowing -Wall -Wextra -Werror=return-type -Werror=non-virtual-dtor -Wno-missing-field-initializers -Wno-type-limits -Wno-array-bounds -Wno-unknown-pragmas -Wunused-local-typedefs -Wno-unused-parameter -Wno-unused-function -Wno-unused-result -Wno-strict-overflow -Wno-strict-aliasing -Wno-error=deprecated-declarations -Wno-stringop-overflow -Wno-psabi -Wno-error=pedantic -Wno-error=redundant-decls -Wno-error=old-style-cast -fdiagnostics-color=always -faligned-new -Wno-unused-but-set-variable -Wno-maybe-uninitialized -fno-math-errno -fno-trapping-math -Werror=format -Werror=cast-function-type -Wno-stringop-overflow, LAPACK_INFO=mkl, PERF_WITH_AVX=1, PERF_WITH_AVX2=1, PERF_WITH_AVX512=1, TORCH_VERSION=1.13.1, USE_CUDA=ON, USE_CUDNN=ON, USE_EXCEPTION_PTR=1, USE_GFLAGS=OFF, USE_GLOG=OFF, USE_MKL=ON, USE_MKLDNN=ON, USE_MPI=OFF, USE_NCCL=ON, USE_NNPACK=ON, USE_OPENMP=ON, USE_ROCM=OFF, + + TorchVision: 0.14.1 + OpenCV: 4.7.0 + MMEngine: 0.7.3 + +Runtime environment: + cudnn_benchmark: True + mp_cfg: {'mp_start_method': 'fork', 'opencv_num_threads': 0} + dist_cfg: {'backend': 'nccl'} + seed: None + deterministic: False + Distributed launcher: slurm + Distributed training: True + GPU number: 2 +------------------------------------------------------------ + +2023/06/05 12:57:48 - mmengine - INFO - Config: +optim_wrapper = dict( + optimizer=dict( + type='SGD', + lr=0.0001, + momentum=0.9, + weight_decay=0.0001, + _scope_='mmpretrain'), + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=False, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=512) +model = dict( + type='ImageClassifier', + backbone=dict( + frozen_stages=2, + type='ResNet', + depth=50, + num_stages=4, + out_indices=(3, ), + style='pytorch', + init_cfg=dict( + type='Pretrained', + checkpoint= + 'https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth', + prefix='backbone')), + neck=dict(type='GlobalAveragePooling'), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=2048, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + topk=1)) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/all1.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/all2.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/all3.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3fake8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/cc1m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3real8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV2-1-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/if-dpmsolver++-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = dict(type='Accuracy', topk=1) +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV2-1-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/if-dpmsolver++-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = dict(type='Accuracy', topk=1) +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1' + +2023/06/05 12:58:00 - mmengine - INFO - Hooks will be executed in the following order: +before_run: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_load_checkpoint: +(ABOVE_NORMAL) EMAHook + -------------------- +before_train: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_train_epoch: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(NORMAL ) DistSamplerSeedHook + -------------------- +before_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook + -------------------- +after_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +after_train_epoch: +(NORMAL ) IterTimerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_val_epoch: +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook + -------------------- +before_val_iter: +(NORMAL ) IterTimerHook + -------------------- +after_val_iter: +(NORMAL ) IterTimerHook +(NORMAL ) VisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_val_epoch: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_save_checkpoint: +(ABOVE_NORMAL) EMAHook + -------------------- +after_train: +(VERY_LOW ) CheckpointHook + -------------------- +before_test_epoch: +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook + -------------------- +before_test_iter: +(NORMAL ) IterTimerHook + -------------------- +after_test_iter: +(NORMAL ) IterTimerHook +(NORMAL ) VisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_test_epoch: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_run: +(BELOW_NORMAL) LoggerHook + -------------------- +2023/06/05 12:58:21 - mmengine - INFO - load backbone in model from: https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth +Name of parameter - Initialization information + +backbone.conv1.weight - torch.Size([64, 3, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.bn1.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.bn1.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.conv1.weight - torch.Size([64, 64, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.bn1.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.bn1.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.conv2.weight - torch.Size([64, 64, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.bn2.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.bn2.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.conv3.weight - torch.Size([256, 64, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.bn3.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.bn3.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.downsample.0.weight - torch.Size([256, 64, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.downsample.1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.downsample.1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.conv1.weight - torch.Size([64, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.bn1.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.bn1.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.conv2.weight - torch.Size([64, 64, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.bn2.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.bn2.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.conv3.weight - torch.Size([256, 64, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.bn3.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.bn3.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.conv1.weight - torch.Size([64, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.bn1.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.bn1.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.conv2.weight - torch.Size([64, 64, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.bn2.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.bn2.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.conv3.weight - torch.Size([256, 64, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.bn3.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.bn3.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.conv1.weight - torch.Size([128, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.bn1.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.bn1.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.conv2.weight - torch.Size([128, 128, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.bn2.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.bn2.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.conv3.weight - torch.Size([512, 128, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.bn3.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.bn3.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.downsample.0.weight - torch.Size([512, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.downsample.1.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.downsample.1.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.conv1.weight - torch.Size([128, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.bn1.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.bn1.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.conv2.weight - torch.Size([128, 128, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.bn2.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.bn2.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.conv3.weight - torch.Size([512, 128, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.bn3.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.bn3.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.conv1.weight - torch.Size([128, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.bn1.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.bn1.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.conv2.weight - torch.Size([128, 128, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.bn2.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.bn2.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.conv3.weight - torch.Size([512, 128, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.bn3.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.bn3.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.conv1.weight - torch.Size([128, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.bn1.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.bn1.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.conv2.weight - torch.Size([128, 128, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.bn2.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.bn2.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.conv3.weight - torch.Size([512, 128, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.bn3.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.bn3.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.conv1.weight - torch.Size([256, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.bn1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.bn1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.conv2.weight - torch.Size([256, 256, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.bn2.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.bn2.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.conv3.weight - torch.Size([1024, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.bn3.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.bn3.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.downsample.0.weight - torch.Size([1024, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.downsample.1.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.downsample.1.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.conv1.weight - torch.Size([256, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.bn1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.bn1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.conv2.weight - torch.Size([256, 256, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.bn2.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.bn2.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.conv3.weight - torch.Size([1024, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.bn3.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.bn3.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.conv1.weight - torch.Size([256, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.bn1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.bn1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.conv2.weight - torch.Size([256, 256, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.bn2.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.bn2.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.conv3.weight - torch.Size([1024, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.bn3.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.bn3.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.conv1.weight - torch.Size([256, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.bn1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.bn1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.conv2.weight - torch.Size([256, 256, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.bn2.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.bn2.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.conv3.weight - torch.Size([1024, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.bn3.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.bn3.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.conv1.weight - torch.Size([256, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.bn1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.bn1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.conv2.weight - torch.Size([256, 256, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.bn2.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.bn2.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.conv3.weight - torch.Size([1024, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.bn3.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.bn3.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.conv1.weight - torch.Size([256, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.bn1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.bn1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.conv2.weight - torch.Size([256, 256, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.bn2.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.bn2.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.conv3.weight - torch.Size([1024, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.bn3.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.bn3.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.conv1.weight - torch.Size([512, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.bn1.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.bn1.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.conv2.weight - torch.Size([512, 512, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.bn2.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.bn2.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.conv3.weight - torch.Size([2048, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.bn3.weight - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.bn3.bias - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.downsample.0.weight - torch.Size([2048, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.downsample.1.weight - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.downsample.1.bias - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.conv1.weight - torch.Size([512, 2048, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.bn1.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.bn1.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.conv2.weight - torch.Size([512, 512, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.bn2.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.bn2.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.conv3.weight - torch.Size([2048, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.bn3.weight - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.bn3.bias - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.conv1.weight - torch.Size([512, 2048, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.bn1.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.bn1.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.conv2.weight - torch.Size([512, 512, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.bn2.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.bn2.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.conv3.weight - torch.Size([2048, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.bn3.weight - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.bn3.bias - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +head.fc.weight - torch.Size([2, 2048]): +NormalInit: mean=0, std=0.01, bias=0 + +head.fc.bias - torch.Size([2]): +NormalInit: mean=0, std=0.01, bias=0 +2023/06/05 12:58:21 - mmengine - WARNING - "FileClient" will be deprecated in future. Please use io functions in https://mmengine.readthedocs.io/en/latest/api/fileio.html#file-io +2023/06/05 12:58:21 - mmengine - WARNING - "HardDiskBackend" is the alias of "LocalBackend" and the former will be deprecated in future. +2023/06/05 12:58:21 - mmengine - INFO - Checkpoints will be saved to /mnt/petrelfs/luzeyu/workspace/fakebench/mmpretrain/workdir/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1. +2023/06/05 12:59:40 - mmengine - INFO - Epoch(train) [1][ 100/4111] lr: 9.9999e-05 eta: 9:03:55 time: 0.7742 data_time: 0.2041 memory: 9436 loss: 0.6428 +2023/06/05 13:01:00 - mmengine - INFO - Epoch(train) [1][ 200/4111] lr: 9.9995e-05 eta: 9:02:38 time: 0.7534 data_time: 0.3893 memory: 6319 loss: 0.5993 +2023/06/05 13:02:12 - mmengine - INFO - Epoch(train) [1][ 300/4111] lr: 9.9988e-05 eta: 8:44:07 time: 0.7856 data_time: 0.1730 memory: 6319 loss: 0.5667 +2023/06/05 13:03:27 - mmengine - INFO - Epoch(train) [1][ 400/4111] lr: 9.9979e-05 eta: 8:39:54 time: 0.8013 data_time: 0.3532 memory: 6319 loss: 0.5497 +2023/06/05 13:04:43 - mmengine - INFO - Epoch(train) [1][ 500/4111] lr: 9.9967e-05 eta: 8:37:03 time: 0.8361 data_time: 0.2420 memory: 6319 loss: 0.5153 +2023/06/05 13:05:59 - mmengine - INFO - Epoch(train) [1][ 600/4111] lr: 9.9953e-05 eta: 8:35:10 time: 0.6796 data_time: 0.0007 memory: 6319 loss: 0.5026 +2023/06/05 13:07:11 - mmengine - INFO - Epoch(train) [1][ 700/4111] lr: 9.9936e-05 eta: 8:29:41 time: 0.6836 data_time: 0.0008 memory: 6319 loss: 0.4585 +2023/06/05 13:08:26 - mmengine - INFO - Epoch(train) [1][ 800/4111] lr: 9.9916e-05 eta: 8:28:28 time: 0.8039 data_time: 0.0007 memory: 6319 loss: 0.4638 +2023/06/05 13:09:40 - mmengine - INFO - Epoch(train) [1][ 900/4111] lr: 9.9894e-05 eta: 8:25:34 time: 0.7464 data_time: 0.0011 memory: 6319 loss: 0.4339 +2023/06/05 13:10:55 - mmengine - INFO - Exp name: resnet50_2xb256_all_1m_lr1e-4_aug_5e-1_20230605_125740 +2023/06/05 13:10:55 - mmengine - INFO - Epoch(train) [1][1000/4111] lr: 9.9869e-05 eta: 8:23:57 time: 0.6744 data_time: 0.0009 memory: 6319 loss: 0.4469 +2023/06/05 13:12:06 - mmengine - INFO - Epoch(train) [1][1100/4111] lr: 9.9841e-05 eta: 8:20:31 time: 0.7514 data_time: 0.0009 memory: 6319 loss: 0.4297 +2023/06/05 13:13:20 - mmengine - INFO - Epoch(train) [1][1200/4111] lr: 9.9811e-05 eta: 8:18:20 time: 0.6752 data_time: 0.0008 memory: 6319 loss: 0.4130 +2023/06/05 13:14:35 - mmengine - INFO - Epoch(train) [1][1300/4111] lr: 9.9778e-05 eta: 8:17:23 time: 0.7549 data_time: 0.0010 memory: 6319 loss: 0.4018 +2023/06/05 13:15:52 - mmengine - INFO - Epoch(train) [1][1400/4111] lr: 9.9743e-05 eta: 8:17:00 time: 0.7881 data_time: 0.0010 memory: 6319 loss: 0.4014 +2023/06/05 13:17:11 - mmengine - INFO - Epoch(train) [1][1500/4111] lr: 9.9705e-05 eta: 8:17:34 time: 0.7475 data_time: 0.0009 memory: 6319 loss: 0.4052 +2023/06/05 13:18:27 - mmengine - INFO - Epoch(train) [1][1600/4111] lr: 9.9664e-05 eta: 8:16:29 time: 0.7959 data_time: 0.0007 memory: 6319 loss: 0.4016 +2023/06/05 13:19:42 - mmengine - INFO - Epoch(train) [1][1700/4111] lr: 9.9621e-05 eta: 8:15:07 time: 0.7259 data_time: 0.0010 memory: 6319 loss: 0.3793 +2023/06/05 13:20:59 - mmengine - INFO - Epoch(train) [1][1800/4111] lr: 9.9575e-05 eta: 8:14:19 time: 0.7215 data_time: 0.0007 memory: 6319 loss: 0.3755 +2023/06/05 13:22:14 - mmengine - INFO - Epoch(train) [1][1900/4111] lr: 9.9527e-05 eta: 8:13:06 time: 0.7830 data_time: 0.0010 memory: 6319 loss: 0.3783 +2023/06/05 13:23:33 - mmengine - INFO - Exp name: resnet50_2xb256_all_1m_lr1e-4_aug_5e-1_20230605_125740 +2023/06/05 13:23:33 - mmengine - INFO - Epoch(train) [1][2000/4111] lr: 9.9476e-05 eta: 8:12:51 time: 0.7923 data_time: 0.0008 memory: 6319 loss: 0.4020 +2023/06/05 13:24:53 - mmengine - INFO - Epoch(train) [1][2100/4111] lr: 9.9422e-05 eta: 8:13:02 time: 0.8824 data_time: 0.0014 memory: 6319 loss: 0.3705 +2023/06/05 13:26:11 - mmengine - INFO - Epoch(train) [1][2200/4111] lr: 9.9366e-05 eta: 8:12:27 time: 0.7648 data_time: 0.0008 memory: 6319 loss: 0.3587 +2023/06/05 13:27:30 - mmengine - INFO - Epoch(train) [1][2300/4111] lr: 9.9307e-05 eta: 8:11:47 time: 0.8399 data_time: 0.0009 memory: 6319 loss: 0.3483 +2023/06/05 13:28:46 - mmengine - INFO - Epoch(train) [1][2400/4111] lr: 9.9246e-05 eta: 8:10:36 time: 0.7457 data_time: 0.0009 memory: 6319 loss: 0.3611 +2023/06/05 13:30:05 - mmengine - INFO - Epoch(train) [1][2500/4111] lr: 9.9182e-05 eta: 8:10:01 time: 0.7930 data_time: 0.0010 memory: 6319 loss: 0.3527 +2023/06/05 13:31:23 - mmengine - INFO - Epoch(train) [1][2600/4111] lr: 9.9115e-05 eta: 8:09:24 time: 0.7478 data_time: 0.0009 memory: 6319 loss: 0.3477 +2023/06/05 13:33:00 - mmengine - INFO - Epoch(train) [1][2700/4111] lr: 9.9046e-05 eta: 8:12:54 time: 0.8190 data_time: 0.0008 memory: 6319 loss: 0.3634 +2023/06/05 13:34:19 - mmengine - INFO - Epoch(train) [1][2800/4111] lr: 9.8974e-05 eta: 8:12:03 time: 0.8434 data_time: 0.0009 memory: 6319 loss: 0.3451 +2023/06/05 13:35:38 - mmengine - INFO - Epoch(train) [1][2900/4111] lr: 9.8900e-05 eta: 8:11:19 time: 0.7545 data_time: 0.0010 memory: 6319 loss: 0.3416 +2023/06/05 13:36:59 - mmengine - INFO - Exp name: resnet50_2xb256_all_1m_lr1e-4_aug_5e-1_20230605_125740 +2023/06/05 13:36:59 - mmengine - INFO - Epoch(train) [1][3000/4111] lr: 9.8823e-05 eta: 8:10:45 time: 0.8981 data_time: 0.0010 memory: 6319 loss: 0.3353 +2023/06/05 13:38:18 - mmengine - INFO - Epoch(train) [1][3100/4111] lr: 9.8744e-05 eta: 8:09:52 time: 0.7958 data_time: 0.0013 memory: 6319 loss: 0.3577 +2023/06/05 13:39:40 - mmengine - INFO - Epoch(train) [1][3200/4111] lr: 9.8662e-05 eta: 8:09:31 time: 0.7664 data_time: 0.0009 memory: 6319 loss: 0.3388 +2023/06/05 13:41:02 - mmengine - INFO - Epoch(train) [1][3300/4111] lr: 9.8578e-05 eta: 8:09:02 time: 0.8132 data_time: 0.0011 memory: 6319 loss: 0.3491 +2023/06/05 13:42:26 - mmengine - INFO - Epoch(train) [1][3400/4111] lr: 9.8490e-05 eta: 8:08:55 time: 0.8212 data_time: 0.0010 memory: 6319 loss: 0.3376 +2023/06/05 13:43:47 - mmengine - INFO - Epoch(train) [1][3500/4111] lr: 9.8401e-05 eta: 8:08:18 time: 0.8311 data_time: 0.0010 memory: 6319 loss: 0.3234 +2023/06/05 13:45:11 - mmengine - INFO - Epoch(train) [1][3600/4111] lr: 9.8309e-05 eta: 8:07:56 time: 0.8697 data_time: 0.0008 memory: 6319 loss: 0.3167 +2023/06/05 13:46:33 - mmengine - INFO - Epoch(train) [1][3700/4111] lr: 9.8214e-05 eta: 8:07:16 time: 0.8765 data_time: 0.0008 memory: 6319 loss: 0.3220 +2023/06/05 13:48:00 - mmengine - INFO - Epoch(train) [1][3800/4111] lr: 9.8117e-05 eta: 8:07:25 time: 0.8392 data_time: 0.0009 memory: 6319 loss: 0.3052 +2023/06/05 13:49:22 - mmengine - INFO - Epoch(train) [1][3900/4111] lr: 9.8017e-05 eta: 8:06:45 time: 0.8501 data_time: 0.0009 memory: 6319 loss: 0.3324 +2023/06/05 13:50:48 - mmengine - INFO - Exp name: resnet50_2xb256_all_1m_lr1e-4_aug_5e-1_20230605_125740 +2023/06/05 13:50:48 - mmengine - INFO - Epoch(train) [1][4000/4111] lr: 9.7915e-05 eta: 8:06:35 time: 0.8163 data_time: 0.0009 memory: 6319 loss: 0.3272 +2023/06/05 13:52:14 - mmengine - INFO - Epoch(train) [1][4100/4111] lr: 9.7810e-05 eta: 8:06:24 time: 0.8786 data_time: 0.0020 memory: 6319 loss: 0.3239 +2023/06/05 13:52:24 - mmengine - INFO - Exp name: resnet50_2xb256_all_1m_lr1e-4_aug_5e-1_20230605_125740 +2023/06/05 13:52:24 - mmengine - INFO - Saving checkpoint at 1 epochs +2023/06/05 13:53:12 - mmengine - INFO - Epoch(val) [1][100/119] eta: 0:00:08 time: 0.6596 data_time: 0.5705 memory: 8106 +2023/06/05 13:53:42 - mmengine - INFO - Epoch(val) [1][119/119] accuracy/top1: 83.5626 data_time: 0.4138 time: 0.5049 +2023/06/05 13:55:03 - mmengine - INFO - Epoch(train) [2][ 100/4111] lr: 9.7691e-05 eta: 8:05:26 time: 0.7789 data_time: 0.1826 memory: 6318 loss: 0.3109 +2023/06/05 13:56:21 - mmengine - INFO - Epoch(train) [2][ 200/4111] lr: 9.7581e-05 eta: 8:04:00 time: 0.9014 data_time: 0.0009 memory: 6318 loss: 0.3000 +2023/06/05 13:57:44 - mmengine - INFO - Epoch(train) [2][ 300/4111] lr: 9.7469e-05 eta: 8:03:15 time: 0.6464 data_time: 0.2909 memory: 6318 loss: 0.3160 +2023/06/05 13:59:02 - mmengine - INFO - Epoch(train) [2][ 400/4111] lr: 9.7354e-05 eta: 8:01:51 time: 0.7634 data_time: 0.0867 memory: 6318 loss: 0.3169 +2023/06/05 14:00:18 - mmengine - INFO - Epoch(train) [2][ 500/4111] lr: 9.7236e-05 eta: 8:00:07 time: 0.7674 data_time: 0.4066 memory: 6318 loss: 0.3082 +2023/06/05 14:01:31 - mmengine - INFO - Epoch(train) [2][ 600/4111] lr: 9.7116e-05 eta: 7:58:02 time: 0.7093 data_time: 0.0009 memory: 6318 loss: 0.3261 +2023/06/05 14:02:46 - mmengine - INFO - Epoch(train) [2][ 700/4111] lr: 9.6994e-05 eta: 7:56:13 time: 0.7093 data_time: 0.0010 memory: 6318 loss: 0.2998 +2023/06/05 14:03:59 - mmengine - INFO - Epoch(train) [2][ 800/4111] lr: 9.6869e-05 eta: 7:54:14 time: 0.6999 data_time: 0.0008 memory: 6318 loss: 0.3144 +2023/06/05 14:05:04 - mmengine - INFO - Exp name: resnet50_2xb256_all_1m_lr1e-4_aug_5e-1_20230605_125740 +2023/06/05 14:05:11 - mmengine - INFO - Epoch(train) [2][ 900/4111] lr: 9.6742e-05 eta: 7:52:07 time: 0.6869 data_time: 0.0008 memory: 6318 loss: 0.2896 +2023/06/05 14:06:25 - mmengine - INFO - Epoch(train) [2][1000/4111] lr: 9.6612e-05 eta: 7:50:15 time: 0.7754 data_time: 0.0009 memory: 6318 loss: 0.2940 +2023/06/05 14:07:37 - mmengine - INFO - Epoch(train) [2][1100/4111] lr: 9.6480e-05 eta: 7:48:11 time: 0.6769 data_time: 0.0008 memory: 6318 loss: 0.2966 +2023/06/05 14:08:51 - mmengine - INFO - Epoch(train) [2][1200/4111] lr: 9.6346e-05 eta: 7:46:26 time: 0.6986 data_time: 0.0009 memory: 6318 loss: 0.3084 +2023/06/05 14:10:03 - mmengine - INFO - Epoch(train) [2][1300/4111] lr: 9.6209e-05 eta: 7:44:29 time: 0.8171 data_time: 0.0009 memory: 6318 loss: 0.2755 +2023/06/05 14:11:17 - mmengine - INFO - Epoch(train) [2][1400/4111] lr: 9.6069e-05 eta: 7:42:44 time: 0.6307 data_time: 0.0009 memory: 6318 loss: 0.2861 +2023/06/05 14:12:29 - mmengine - INFO - Epoch(train) [2][1500/4111] lr: 9.5928e-05 eta: 7:40:48 time: 0.7079 data_time: 0.0012 memory: 6318 loss: 0.2763 +2023/06/05 14:13:41 - mmengine - INFO - Epoch(train) [2][1600/4111] lr: 9.5783e-05 eta: 7:38:56 time: 0.7026 data_time: 0.0013 memory: 6318 loss: 0.2957 +2023/06/05 14:14:55 - mmengine - INFO - Epoch(train) [2][1700/4111] lr: 9.5637e-05 eta: 7:37:10 time: 0.6900 data_time: 0.0011 memory: 6318 loss: 0.3187 +2023/06/05 14:16:07 - mmengine - INFO - Epoch(train) [2][1800/4111] lr: 9.5488e-05 eta: 7:35:20 time: 0.7827 data_time: 0.0008 memory: 6318 loss: 0.2890 +2023/06/05 14:17:13 - mmengine - INFO - Exp name: resnet50_2xb256_all_1m_lr1e-4_aug_5e-1_20230605_125740 +2023/06/05 14:17:20 - mmengine - INFO - Epoch(train) [2][1900/4111] lr: 9.5337e-05 eta: 7:33:37 time: 0.6677 data_time: 0.0008 memory: 6318 loss: 0.3179 +2023/06/05 14:18:32 - mmengine - INFO - Epoch(train) [2][2000/4111] lr: 9.5183e-05 eta: 7:31:48 time: 0.7555 data_time: 0.0008 memory: 6318 loss: 0.2722 +2023/06/05 14:19:48 - mmengine - INFO - Epoch(train) [2][2100/4111] lr: 9.5027e-05 eta: 7:30:21 time: 0.7338 data_time: 0.0008 memory: 6318 loss: 0.2902 +2023/06/05 14:21:02 - mmengine - INFO - Epoch(train) [2][2200/4111] lr: 9.4869e-05 eta: 7:28:44 time: 0.7299 data_time: 0.0009 memory: 6318 loss: 0.2824 +2023/06/05 14:22:22 - mmengine - INFO - Epoch(train) [2][2300/4111] lr: 9.4708e-05 eta: 7:27:41 time: 1.1837 data_time: 0.0009 memory: 6318 loss: 0.2830 +2023/06/05 14:23:37 - mmengine - INFO - Epoch(train) [2][2400/4111] lr: 9.4545e-05 eta: 7:26:10 time: 0.8235 data_time: 0.0010 memory: 6318 loss: 0.3017 +2023/06/05 14:24:50 - mmengine - INFO - Epoch(train) [2][2500/4111] lr: 9.4380e-05 eta: 7:24:33 time: 0.7664 data_time: 0.0012 memory: 6318 loss: 0.2948 +2023/06/05 14:26:07 - mmengine - INFO - Epoch(train) [2][2600/4111] lr: 9.4212e-05 eta: 7:23:10 time: 0.7472 data_time: 0.0008 memory: 6318 loss: 0.2975 +2023/06/05 14:27:35 - mmengine - INFO - Epoch(train) [2][2700/4111] lr: 9.4043e-05 eta: 7:22:50 time: 1.9882 data_time: 0.0009 memory: 6318 loss: 0.2873 +2023/06/05 14:28:49 - mmengine - INFO - Epoch(train) [2][2800/4111] lr: 9.3870e-05 eta: 7:21:16 time: 0.7905 data_time: 0.0011 memory: 6318 loss: 0.2905 +2023/06/05 14:29:59 - mmengine - INFO - Exp name: resnet50_2xb256_all_1m_lr1e-4_aug_5e-1_20230605_125740 +2023/06/05 14:30:07 - mmengine - INFO - Epoch(train) [2][2900/4111] lr: 9.3696e-05 eta: 7:19:57 time: 0.7432 data_time: 0.0009 memory: 6318 loss: 0.2926 +2023/06/05 14:31:21 - mmengine - INFO - Epoch(train) [2][3000/4111] lr: 9.3519e-05 eta: 7:18:23 time: 0.6783 data_time: 0.0009 memory: 6318 loss: 0.3042 +2023/06/05 14:32:36 - mmengine - INFO - Epoch(train) [2][3100/4111] lr: 9.3340e-05 eta: 7:16:56 time: 0.8870 data_time: 0.0009 memory: 6318 loss: 0.2720 +2023/06/05 14:33:50 - mmengine - INFO - Epoch(train) [2][3200/4111] lr: 9.3159e-05 eta: 7:15:23 time: 0.7410 data_time: 0.0010 memory: 6318 loss: 0.2902 +2023/06/05 14:35:09 - mmengine - INFO - Epoch(train) [2][3300/4111] lr: 9.2976e-05 eta: 7:14:13 time: 0.7320 data_time: 0.0009 memory: 6318 loss: 0.2734 +2023/06/05 14:36:22 - mmengine - INFO - Epoch(train) [2][3400/4111] lr: 9.2790e-05 eta: 7:12:40 time: 0.7835 data_time: 0.0009 memory: 6318 loss: 0.2973 +2023/06/05 14:37:40 - mmengine - INFO - Epoch(train) [2][3500/4111] lr: 9.2603e-05 eta: 7:11:23 time: 0.7930 data_time: 0.0009 memory: 6318 loss: 0.2961 +2023/06/05 14:38:55 - mmengine - INFO - Epoch(train) [2][3600/4111] lr: 9.2413e-05 eta: 7:09:58 time: 0.7792 data_time: 0.0009 memory: 6318 loss: 0.2653 +2023/06/05 14:40:16 - mmengine - INFO - Epoch(train) [2][3700/4111] lr: 9.2220e-05 eta: 7:08:54 time: 0.7937 data_time: 0.0009 memory: 6318 loss: 0.2825 +2023/06/05 14:41:31 - mmengine - INFO - Epoch(train) [2][3800/4111] lr: 9.2026e-05 eta: 7:07:28 time: 0.7585 data_time: 0.0010 memory: 6318 loss: 0.2954 +2023/06/05 14:42:41 - mmengine - INFO - Exp name: resnet50_2xb256_all_1m_lr1e-4_aug_5e-1_20230605_125740 +2023/06/05 14:42:49 - mmengine - INFO - Epoch(train) [2][3900/4111] lr: 9.1829e-05 eta: 7:06:15 time: 0.7750 data_time: 0.0009 memory: 6318 loss: 0.2654 +2023/06/05 14:44:07 - mmengine - INFO - Epoch(train) [2][4000/4111] lr: 9.1631e-05 eta: 7:05:01 time: 0.7844 data_time: 0.0011 memory: 6318 loss: 0.2730 +2023/06/05 14:45:28 - mmengine - INFO - Epoch(train) [2][4100/4111] lr: 9.1430e-05 eta: 7:03:56 time: 0.8431 data_time: 0.0017 memory: 6318 loss: 0.2942 +2023/06/05 14:45:35 - mmengine - INFO - Exp name: resnet50_2xb256_all_1m_lr1e-4_aug_5e-1_20230605_125740 +2023/06/05 14:45:35 - mmengine - INFO - Saving checkpoint at 2 epochs +2023/06/05 14:46:21 - mmengine - INFO - Epoch(val) [2][100/119] eta: 0:00:07 time: 0.5283 data_time: 0.4396 memory: 6318 +2023/06/05 14:46:47 - mmengine - INFO - Epoch(val) [2][119/119] accuracy/top1: 80.3899 data_time: 0.3671 time: 0.4557 +2023/06/05 14:48:03 - mmengine - INFO - Epoch(train) [3][ 100/4111] lr: 9.1205e-05 eta: 7:02:22 time: 0.6853 data_time: 0.0007 memory: 6318 loss: 0.2650 +2023/06/05 14:49:16 - mmengine - INFO - Epoch(train) [3][ 200/4111] lr: 9.0999e-05 eta: 7:00:46 time: 0.7479 data_time: 0.0008 memory: 6318 loss: 0.2731 +2023/06/05 14:50:26 - mmengine - INFO - Epoch(train) [3][ 300/4111] lr: 9.0792e-05 eta: 6:59:04 time: 0.7287 data_time: 0.0007 memory: 6318 loss: 0.2826 +2023/06/05 14:51:35 - mmengine - INFO - Epoch(train) [3][ 400/4111] lr: 9.0582e-05 eta: 6:57:15 time: 0.7380 data_time: 0.0011 memory: 6318 loss: 0.2907 +2023/06/05 14:52:46 - mmengine - INFO - Epoch(train) [3][ 500/4111] lr: 9.0371e-05 eta: 6:55:36 time: 0.6743 data_time: 0.0009 memory: 6318 loss: 0.2685 +2023/06/05 14:53:58 - mmengine - INFO - Epoch(train) [3][ 600/4111] lr: 9.0157e-05 eta: 6:54:02 time: 0.7188 data_time: 0.1056 memory: 6318 loss: 0.2740 +2023/06/05 14:55:10 - mmengine - INFO - Epoch(train) [3][ 700/4111] lr: 8.9942e-05 eta: 6:52:27 time: 0.7447 data_time: 0.3643 memory: 6318 loss: 0.2815 +2023/06/05 14:56:09 - mmengine - INFO - Exp name: resnet50_2xb256_all_1m_lr1e-4_aug_5e-1_20230605_125740 +2023/06/05 14:56:24 - mmengine - INFO - Epoch(train) [3][ 800/4111] lr: 8.9724e-05 eta: 6:50:59 time: 0.7448 data_time: 0.6018 memory: 6318 loss: 0.2692 +2023/06/05 14:57:35 - mmengine - INFO - Epoch(train) [3][ 900/4111] lr: 8.9504e-05 eta: 6:49:20 time: 0.7695 data_time: 0.5832 memory: 6318 loss: 0.2782 +2023/06/05 14:58:45 - mmengine - INFO - Epoch(train) [3][1000/4111] lr: 8.9282e-05 eta: 6:47:40 time: 0.6863 data_time: 0.5471 memory: 6318 loss: 0.2763 +2023/06/05 14:59:55 - mmengine - INFO - Epoch(train) [3][1100/4111] lr: 8.9059e-05 eta: 6:45:59 time: 0.6888 data_time: 0.5433 memory: 6318 loss: 0.2706 +2023/06/05 15:01:05 - mmengine - INFO - Epoch(train) [3][1200/4111] lr: 8.8833e-05 eta: 6:44:23 time: 0.6886 data_time: 0.5491 memory: 6318 loss: 0.2769 +2023/06/05 15:02:16 - mmengine - INFO - Epoch(train) [3][1300/4111] lr: 8.8605e-05 eta: 6:42:45 time: 0.6792 data_time: 0.5289 memory: 6318 loss: 0.2669 +2023/06/05 15:03:24 - mmengine - INFO - Epoch(train) [3][1400/4111] lr: 8.8375e-05 eta: 6:41:03 time: 0.6331 data_time: 0.4923 memory: 6318 loss: 0.2619 +2023/06/05 15:04:33 - mmengine - INFO - Epoch(train) [3][1500/4111] lr: 8.8144e-05 eta: 6:39:20 time: 0.6950 data_time: 0.3788 memory: 6318 loss: 0.2719 +2023/06/05 15:05:45 - mmengine - INFO - Epoch(train) [3][1600/4111] lr: 8.7910e-05 eta: 6:37:52 time: 0.7249 data_time: 0.5856 memory: 6318 loss: 0.2664 +2023/06/05 15:06:52 - mmengine - INFO - Epoch(train) [3][1700/4111] lr: 8.7675e-05 eta: 6:36:05 time: 0.6689 data_time: 0.4381 memory: 6318 loss: 0.2720 +2023/06/05 15:07:50 - mmengine - INFO - Exp name: resnet50_2xb256_all_1m_lr1e-4_aug_5e-1_20230605_125740 +2023/06/05 15:08:04 - mmengine - INFO - Epoch(train) [3][1800/4111] lr: 8.7437e-05 eta: 6:34:36 time: 0.6393 data_time: 0.1606 memory: 6318 loss: 0.2647 +2023/06/05 15:09:17 - mmengine - INFO - Epoch(train) [3][1900/4111] lr: 8.7198e-05 eta: 6:33:11 time: 0.6986 data_time: 0.0009 memory: 6318 loss: 0.2876 +2023/06/05 15:10:27 - mmengine - INFO - Epoch(train) [3][2000/4111] lr: 8.6957e-05 eta: 6:31:36 time: 0.6364 data_time: 0.0010 memory: 6318 loss: 0.2736 +2023/06/05 15:11:38 - mmengine - INFO - Epoch(train) [3][2100/4111] lr: 8.6714e-05 eta: 6:30:05 time: 0.7715 data_time: 0.0009 memory: 6318 loss: 0.2636 +2023/06/05 15:12:50 - mmengine - INFO - Epoch(train) [3][2200/4111] lr: 8.6469e-05 eta: 6:28:39 time: 0.7740 data_time: 0.0010 memory: 6318 loss: 0.2497 +2023/06/05 15:13:59 - mmengine - INFO - Epoch(train) [3][2300/4111] lr: 8.6222e-05 eta: 6:27:02 time: 0.6491 data_time: 0.0013 memory: 6318 loss: 0.2686 +2023/06/05 15:15:08 - mmengine - INFO - Epoch(train) [3][2400/4111] lr: 8.5974e-05 eta: 6:25:25 time: 0.6311 data_time: 0.0007 memory: 6318 loss: 0.2690 +2023/06/05 15:16:20 - mmengine - INFO - Epoch(train) [3][2500/4111] lr: 8.5723e-05 eta: 6:23:58 time: 0.6957 data_time: 0.0009 memory: 6318 loss: 0.2824 +2023/06/05 15:17:33 - mmengine - INFO - Epoch(train) [3][2600/4111] lr: 8.5471e-05 eta: 6:22:35 time: 0.7715 data_time: 0.0012 memory: 6318 loss: 0.2520 diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_125740/vis_data/20230605_125740.json b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_125740/vis_data/20230605_125740.json new file mode 100644 index 0000000000000000000000000000000000000000..cdbeba31b0196004977e47409c7542c58955ab80 --- /dev/null +++ b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_125740/vis_data/20230605_125740.json @@ -0,0 +1,110 @@ +{"lr": 9.999871217862936e-05, "data_time": 0.20409960746765138, "loss": 0.6428110539913178, "time": 0.7742290496826172, "epoch": 1, "memory": 9436, "step": 100} +{"lr": 9.999479662539605e-05, "data_time": 0.3893429279327393, "loss": 0.5992599487304687, "time": 0.7533838510513305, "epoch": 1, "memory": 6319, "step": 200} +{"lr": 9.99882534375654e-05, "data_time": 0.17300114631652833, "loss": 0.5666563391685486, "time": 0.7856077432632447, "epoch": 1, "memory": 6319, "step": 300} +{"lr": 9.997908299725075e-05, "data_time": 0.3531585931777954, "loss": 0.5496869027614594, "time": 0.8012889385223388, "epoch": 1, "memory": 6319, "step": 400} +{"lr": 9.996728583999318e-05, "data_time": 0.24204785823822023, "loss": 0.5152768701314926, "time": 0.8361482620239258, "epoch": 1, "memory": 6319, "step": 500} +{"lr": 9.995286265473058e-05, "data_time": 0.0007352590560913086, "loss": 0.5025729775428772, "time": 0.6796221256256103, "epoch": 1, "memory": 6319, "step": 600} +{"lr": 9.993581428375725e-05, "data_time": 0.0007589817047119141, "loss": 0.4584820032119751, "time": 0.6835838794708252, "epoch": 1, "memory": 6319, "step": 700} +{"lr": 9.991614172267466e-05, "data_time": 0.0006952524185180664, "loss": 0.4638307332992554, "time": 0.8038980722427368, "epoch": 1, "memory": 6319, "step": 800} +{"lr": 9.989384612033355e-05, "data_time": 0.0010694503784179688, "loss": 0.4339237719774246, "time": 0.7463749170303344, "epoch": 1, "memory": 6319, "step": 900} +{"lr": 9.986892877876658e-05, "data_time": 0.0009260177612304688, "loss": 0.44686152040958405, "time": 0.6744317770004272, "epoch": 1, "memory": 6319, "step": 1000} +{"lr": 9.984139115311244e-05, "data_time": 0.0009229660034179687, "loss": 0.4297418177127838, "time": 0.7513756513595581, "epoch": 1, "memory": 6319, "step": 1100} +{"lr": 9.981123485153081e-05, "data_time": 0.0008283376693725586, "loss": 0.41301066875457765, "time": 0.6751856565475464, "epoch": 1, "memory": 6319, "step": 1200} +{"lr": 9.977846163510851e-05, "data_time": 0.0009643316268920898, "loss": 0.4018182665109634, "time": 0.7549192667007446, "epoch": 1, "memory": 6319, "step": 1300} +{"lr": 9.974307341775653e-05, "data_time": 0.0009546279907226562, "loss": 0.4013751924037933, "time": 0.7881002426147461, "epoch": 1, "memory": 6319, "step": 1400} +{"lr": 9.970507226609833e-05, "data_time": 0.0009330987930297851, "loss": 0.40516377389431, "time": 0.7475346088409424, "epoch": 1, "memory": 6319, "step": 1500} +{"lr": 9.966446039934934e-05, "data_time": 0.0006869554519653321, "loss": 0.4016081541776657, "time": 0.7958840370178223, "epoch": 1, "memory": 6319, "step": 1600} +{"lr": 9.962124018918683e-05, "data_time": 0.0009645938873291016, "loss": 0.3792893081903458, "time": 0.7258940458297729, "epoch": 1, "memory": 6319, "step": 1700} +{"lr": 9.957541415961212e-05, "data_time": 0.0007101774215698242, "loss": 0.3755451083183289, "time": 0.7215308427810669, "epoch": 1, "memory": 6319, "step": 1800} +{"lr": 9.952698498680267e-05, "data_time": 0.001020526885986328, "loss": 0.3782971054315567, "time": 0.7829531908035279, "epoch": 1, "memory": 6319, "step": 1900} +{"lr": 9.947595549895583e-05, "data_time": 0.0008440017700195312, "loss": 0.4020328789949417, "time": 0.7923438072204589, "epoch": 1, "memory": 6319, "step": 2000} +{"lr": 9.942232867612417e-05, "data_time": 0.0014200448989868165, "loss": 0.3704873651266098, "time": 0.882388973236084, "epoch": 1, "memory": 6319, "step": 2100} +{"lr": 9.93661076500406e-05, "data_time": 0.0008404970169067383, "loss": 0.3586942821741104, "time": 0.7647693872451782, "epoch": 1, "memory": 6319, "step": 2200} +{"lr": 9.930729570393622e-05, "data_time": 0.0009474754333496094, "loss": 0.3483234286308289, "time": 0.8398890256881714, "epoch": 1, "memory": 6319, "step": 2300} +{"lr": 9.924589627234824e-05, "data_time": 0.0008964061737060547, "loss": 0.3611366808414459, "time": 0.7457397222518921, "epoch": 1, "memory": 6319, "step": 2400} +{"lr": 9.918191294091956e-05, "data_time": 0.0009618759155273438, "loss": 0.3527453541755676, "time": 0.7929868221282959, "epoch": 1, "memory": 6319, "step": 2500} +{"lr": 9.91153494461892e-05, "data_time": 0.0009335756301879882, "loss": 0.3476579636335373, "time": 0.747836422920227, "epoch": 1, "memory": 6319, "step": 2600} +{"lr": 9.904620967537422e-05, "data_time": 0.000835585594177246, "loss": 0.36337176263332366, "time": 0.8190121650695801, "epoch": 1, "memory": 6319, "step": 2700} +{"lr": 9.897449766614277e-05, "data_time": 0.0008677005767822266, "loss": 0.34508435130119325, "time": 0.8433783531188965, "epoch": 1, "memory": 6319, "step": 2800} +{"lr": 9.890021760637804e-05, "data_time": 0.0009719133377075195, "loss": 0.34163249731063844, "time": 0.7544926643371582, "epoch": 1, "memory": 6319, "step": 2900} +{"lr": 9.882337383393409e-05, "data_time": 0.0009678125381469727, "loss": 0.33529525995254517, "time": 0.8980875492095948, "epoch": 1, "memory": 6319, "step": 3000} +{"lr": 9.874397083638193e-05, "data_time": 0.001253199577331543, "loss": 0.3577060431241989, "time": 0.7957884073257446, "epoch": 1, "memory": 6319, "step": 3100} +{"lr": 9.866201325074828e-05, "data_time": 0.0008949756622314454, "loss": 0.33875913321971896, "time": 0.7663751840591431, "epoch": 1, "memory": 6319, "step": 3200} +{"lr": 9.857750586324393e-05, "data_time": 0.0011025190353393555, "loss": 0.34913310110569, "time": 0.8131855010986329, "epoch": 1, "memory": 6319, "step": 3300} +{"lr": 9.84904536089848e-05, "data_time": 0.000963902473449707, "loss": 0.3376062482595444, "time": 0.8211594820022583, "epoch": 1, "memory": 6319, "step": 3400} +{"lr": 9.840086157170351e-05, "data_time": 0.0010434865951538086, "loss": 0.32343386113643646, "time": 0.8311334133148194, "epoch": 1, "memory": 6319, "step": 3500} +{"lr": 9.830873498345255e-05, "data_time": 0.0008426666259765625, "loss": 0.31666031777858733, "time": 0.8696626663208008, "epoch": 1, "memory": 6319, "step": 3600} +{"lr": 9.821407922429864e-05, "data_time": 0.0008366584777832031, "loss": 0.3220408380031586, "time": 0.8764559507369996, "epoch": 1, "memory": 6319, "step": 3700} +{"lr": 9.811689982200877e-05, "data_time": 0.0009337186813354493, "loss": 0.3052338659763336, "time": 0.8391748666763306, "epoch": 1, "memory": 6319, "step": 3800} +{"lr": 9.801720245172702e-05, "data_time": 0.0008714199066162109, "loss": 0.33241842687129974, "time": 0.8501276731491089, "epoch": 1, "memory": 6319, "step": 3900} +{"lr": 9.791499293564358e-05, "data_time": 0.0009468555450439453, "loss": 0.3272052824497223, "time": 0.8162740707397461, "epoch": 1, "memory": 6319, "step": 4000} +{"lr": 9.781027724265439e-05, "data_time": 0.001980876922607422, "loss": 0.3239219099283218, "time": 0.8785770654678344, "epoch": 1, "memory": 6319, "step": 4100} +{"accuracy/top1": 83.56255340576172, "data_time": 0.4138187961418088, "time": 0.504850471720976, "step": 1} +{"lr": 9.769111539369434e-05, "data_time": 0.18262119293212892, "loss": 0.3109478265047073, "time": 0.7789007425308228, "epoch": 2, "memory": 6318, "step": 4211} +{"lr": 9.758113190911623e-05, "data_time": 0.0008503198623657227, "loss": 0.3000048279762268, "time": 0.9014130115509034, "epoch": 2, "memory": 6318, "step": 4311} +{"lr": 9.746866174465984e-05, "data_time": 0.29087731838226316, "loss": 0.3160201162099838, "time": 0.6463502883911133, "epoch": 2, "memory": 6318, "step": 4411} +{"lr": 9.735371146842894e-05, "data_time": 0.08671903610229492, "loss": 0.3168675124645233, "time": 0.7634302377700806, "epoch": 2, "memory": 6318, "step": 4511} +{"lr": 9.723628779336255e-05, "data_time": 0.4065819501876831, "loss": 0.30820069313049314, "time": 0.7674283742904663, "epoch": 2, "memory": 6318, "step": 4611} +{"lr": 9.711639757684265e-05, "data_time": 0.0009097814559936524, "loss": 0.32611868977546693, "time": 0.7092680215835572, "epoch": 2, "memory": 6318, "step": 4711} +{"lr": 9.699404782029395e-05, "data_time": 0.0009825468063354493, "loss": 0.29978167116641996, "time": 0.7092856645584107, "epoch": 2, "memory": 6318, "step": 4811} +{"lr": 9.686924566877498e-05, "data_time": 0.0008087158203125, "loss": 0.3144329011440277, "time": 0.6998948574066162, "epoch": 2, "memory": 6318, "step": 4911} +{"lr": 9.674199841056078e-05, "data_time": 0.0008284330368041992, "loss": 0.2896211877465248, "time": 0.6869336605072022, "epoch": 2, "memory": 6318, "step": 5011} +{"lr": 9.661231347671705e-05, "data_time": 0.0008912563323974609, "loss": 0.2940391093492508, "time": 0.7754358291625977, "epoch": 2, "memory": 6318, "step": 5111} +{"lr": 9.64801984406668e-05, "data_time": 0.0008103132247924804, "loss": 0.29659841358661654, "time": 0.6768763780593872, "epoch": 2, "memory": 6318, "step": 5211} +{"lr": 9.634566101774737e-05, "data_time": 0.0009378433227539062, "loss": 0.30843727588653563, "time": 0.6985715389251709, "epoch": 2, "memory": 6318, "step": 5311} +{"lr": 9.62087090647604e-05, "data_time": 0.0009206056594848633, "loss": 0.2754927545785904, "time": 0.8171066284179688, "epoch": 2, "memory": 6318, "step": 5411} +{"lr": 9.606935057951247e-05, "data_time": 0.0009344816207885742, "loss": 0.28609300702810286, "time": 0.63073251247406, "epoch": 2, "memory": 6318, "step": 5511} +{"lr": 9.592759370034884e-05, "data_time": 0.0011746406555175782, "loss": 0.2763012573122978, "time": 0.70788414478302, "epoch": 2, "memory": 6318, "step": 5611} +{"lr": 9.578344670567721e-05, "data_time": 0.0013096332550048828, "loss": 0.2957438349723816, "time": 0.7025631189346313, "epoch": 2, "memory": 6318, "step": 5711} +{"lr": 9.563691801348495e-05, "data_time": 0.0011096715927124024, "loss": 0.3187422066926956, "time": 0.6899694919586181, "epoch": 2, "memory": 6318, "step": 5811} +{"lr": 9.548801618084738e-05, "data_time": 0.0007781744003295899, "loss": 0.2889773666858673, "time": 0.7827431440353394, "epoch": 2, "memory": 6318, "step": 5911} +{"lr": 9.533674990342786e-05, "data_time": 0.0007563591003417969, "loss": 0.3179330348968506, "time": 0.6677175045013428, "epoch": 2, "memory": 6318, "step": 6011} +{"lr": 9.518312801497029e-05, "data_time": 0.0008057355880737305, "loss": 0.2721507683396339, "time": 0.755538821220398, "epoch": 2, "memory": 6318, "step": 6111} +{"lr": 9.502715948678276e-05, "data_time": 0.0008047580718994141, "loss": 0.29019091129302976, "time": 0.7338272094726562, "epoch": 2, "memory": 6318, "step": 6211} +{"lr": 9.486885342721427e-05, "data_time": 0.0008751630783081054, "loss": 0.2824150174856186, "time": 0.729899549484253, "epoch": 2, "memory": 6318, "step": 6311} +{"lr": 9.470821908112202e-05, "data_time": 0.0009315967559814453, "loss": 0.28303675949573515, "time": 1.1837078332901, "epoch": 2, "memory": 6318, "step": 6411} +{"lr": 9.454526582933214e-05, "data_time": 0.0009738683700561523, "loss": 0.3016718864440918, "time": 0.823495602607727, "epoch": 2, "memory": 6318, "step": 6511} +{"lr": 9.438000318809176e-05, "data_time": 0.0012355089187622071, "loss": 0.29478231221437456, "time": 0.7664000988006592, "epoch": 2, "memory": 6318, "step": 6611} +{"lr": 9.42124408085132e-05, "data_time": 0.0008062124252319336, "loss": 0.29754903316497805, "time": 0.7471588134765625, "epoch": 2, "memory": 6318, "step": 6711} +{"lr": 9.404258847601036e-05, "data_time": 0.0009423017501831055, "loss": 0.2872541010379791, "time": 1.9881983518600463, "epoch": 2, "memory": 6318, "step": 6811} +{"lr": 9.387045610972717e-05, "data_time": 0.0010599851608276366, "loss": 0.29051762670278547, "time": 0.7904786348342896, "epoch": 2, "memory": 6318, "step": 6911} +{"lr": 9.369605376195838e-05, "data_time": 0.0009445667266845704, "loss": 0.29260163754224777, "time": 0.7432496309280395, "epoch": 2, "memory": 6318, "step": 7011} +{"lr": 9.351939161756269e-05, "data_time": 0.0009049415588378906, "loss": 0.30424838662147524, "time": 0.6783438205718995, "epoch": 2, "memory": 6318, "step": 7111} +{"lr": 9.334047999336764e-05, "data_time": 0.0009110450744628906, "loss": 0.2720350831747055, "time": 0.887003231048584, "epoch": 2, "memory": 6318, "step": 7211} +{"lr": 9.315932933756752e-05, "data_time": 0.0009765148162841797, "loss": 0.2902287930250168, "time": 0.7409857749938965, "epoch": 2, "memory": 6318, "step": 7311} +{"lr": 9.297595022911287e-05, "data_time": 0.0008769750595092774, "loss": 0.2734398484230042, "time": 0.7320471286773682, "epoch": 2, "memory": 6318, "step": 7411} +{"lr": 9.279035337709295e-05, "data_time": 0.0008745908737182618, "loss": 0.29733290821313857, "time": 0.7834531784057617, "epoch": 2, "memory": 6318, "step": 7511} +{"lr": 9.260254962011016e-05, "data_time": 0.0008535146713256836, "loss": 0.29611544907093046, "time": 0.793001675605774, "epoch": 2, "memory": 6318, "step": 7611} +{"lr": 9.241254992564702e-05, "data_time": 0.0009152412414550782, "loss": 0.2653490588068962, "time": 0.77921142578125, "epoch": 2, "memory": 6318, "step": 7711} +{"lr": 9.22203653894259e-05, "data_time": 0.0009097576141357422, "loss": 0.28250943422317504, "time": 0.7937284946441651, "epoch": 2, "memory": 6318, "step": 7811} +{"lr": 9.202600723476098e-05, "data_time": 0.000969099998474121, "loss": 0.29540292024612425, "time": 0.7584635734558105, "epoch": 2, "memory": 6318, "step": 7911} +{"lr": 9.182948681190268e-05, "data_time": 0.0009256124496459961, "loss": 0.2653553158044815, "time": 0.7749513387680054, "epoch": 2, "memory": 6318, "step": 8011} +{"lr": 9.163081559737497e-05, "data_time": 0.0011258125305175781, "loss": 0.2730012521147728, "time": 0.7843770265579224, "epoch": 2, "memory": 6318, "step": 8111} +{"lr": 9.143000519330517e-05, "data_time": 0.0016515731811523437, "loss": 0.29417825043201445, "time": 0.8431460380554199, "epoch": 2, "memory": 6318, "step": 8211} +{"accuracy/top1": 80.38992309570312, "data_time": 0.36706274350484214, "time": 0.4556770404179891, "step": 2} +{"lr": 9.120461478753376e-05, "data_time": 0.0007345914840698242, "loss": 0.2649999871850014, "time": 0.685315752029419, "epoch": 3, "memory": 6318, "step": 8322} +{"lr": 9.099932932138134e-05, "data_time": 0.000763702392578125, "loss": 0.2731035828590393, "time": 0.7478806495666503, "epoch": 3, "memory": 6318, "step": 8422} +{"lr": 9.079194154362181e-05, "data_time": 0.0007415533065795899, "loss": 0.2825735807418823, "time": 0.7287466764450073, "epoch": 3, "memory": 6318, "step": 8522} +{"lr": 9.058246356541787e-05, "data_time": 0.0011133670806884766, "loss": 0.2907098770141602, "time": 0.7379800319671631, "epoch": 3, "memory": 6318, "step": 8622} +{"lr": 9.037090761999699e-05, "data_time": 0.0009167671203613281, "loss": 0.2684899285435677, "time": 0.6742784976959229, "epoch": 3, "memory": 6318, "step": 8722} +{"lr": 9.015728606193701e-05, "data_time": 0.10558381080627441, "loss": 0.2739625841379166, "time": 0.7187713384628296, "epoch": 3, "memory": 6318, "step": 8822} +{"lr": 8.994161136644488e-05, "data_time": 0.3643131494522095, "loss": 0.28152443170547486, "time": 0.7447021484375, "epoch": 3, "memory": 6318, "step": 8922} +{"lr": 8.972389612862792e-05, "data_time": 0.6017848014831543, "loss": 0.2692335307598114, "time": 0.7448285102844239, "epoch": 3, "memory": 6318, "step": 9022} +{"lr": 8.950415306275833e-05, "data_time": 0.5831636428833008, "loss": 0.2781596094369888, "time": 0.769466757774353, "epoch": 3, "memory": 6318, "step": 9122} +{"lr": 8.928239500153059e-05, "data_time": 0.5470522165298461, "loss": 0.2762856870889664, "time": 0.6863395214080811, "epoch": 3, "memory": 6318, "step": 9222} +{"lr": 8.905863489531232e-05, "data_time": 0.5433248043060303, "loss": 0.2705572977662086, "time": 0.6887992858886719, "epoch": 3, "memory": 6318, "step": 9322} +{"lr": 8.883288581138775e-05, "data_time": 0.5490938901901246, "loss": 0.27692133486270903, "time": 0.6886429071426392, "epoch": 3, "memory": 6318, "step": 9422} +{"lr": 8.86051609331947e-05, "data_time": 0.5289304971694946, "loss": 0.2668500617146492, "time": 0.6792001485824585, "epoch": 3, "memory": 6318, "step": 9522} +{"lr": 8.837547355955467e-05, "data_time": 0.4923022508621216, "loss": 0.26194403767585756, "time": 0.633077073097229, "epoch": 3, "memory": 6318, "step": 9622} +{"lr": 8.814383710389614e-05, "data_time": 0.3787827730178833, "loss": 0.27189348489046095, "time": 0.6949618816375732, "epoch": 3, "memory": 6318, "step": 9722} +{"lr": 8.791026509347146e-05, "data_time": 0.5856317520141602, "loss": 0.2663997381925583, "time": 0.724916672706604, "epoch": 3, "memory": 6318, "step": 9822} +{"lr": 8.767477116856659e-05, "data_time": 0.43811945915222167, "loss": 0.2720366507768631, "time": 0.6689430236816406, "epoch": 3, "memory": 6318, "step": 9922} +{"lr": 8.743736908170466e-05, "data_time": 0.16061439514160156, "loss": 0.2646900206804276, "time": 0.6393389463424682, "epoch": 3, "memory": 6318, "step": 10022} +{"lr": 8.719807269684305e-05, "data_time": 0.0008947849273681641, "loss": 0.2876169040799141, "time": 0.698579216003418, "epoch": 3, "memory": 6318, "step": 10122} +{"lr": 8.695689598856356e-05, "data_time": 0.000968170166015625, "loss": 0.2736373618245125, "time": 0.63644380569458, "epoch": 3, "memory": 6318, "step": 10222} +{"lr": 8.67138530412559e-05, "data_time": 0.0009461641311645508, "loss": 0.263558404147625, "time": 0.7714944124221802, "epoch": 3, "memory": 6318, "step": 10322} +{"lr": 8.646895804829604e-05, "data_time": 0.0009682655334472656, "loss": 0.24974735379219054, "time": 0.773975419998169, "epoch": 3, "memory": 6318, "step": 10422} +{"lr": 8.622222531121664e-05, "data_time": 0.0013257265090942383, "loss": 0.2685848161578178, "time": 0.6491158723831176, "epoch": 3, "memory": 6318, "step": 10522} +{"lr": 8.597366923887211e-05, "data_time": 0.0007126808166503906, "loss": 0.26895226538181305, "time": 0.6310829401016236, "epoch": 3, "memory": 6318, "step": 10622} +{"lr": 8.57233043465972e-05, "data_time": 0.0009105682373046875, "loss": 0.2824095621705055, "time": 0.695707893371582, "epoch": 3, "memory": 6318, "step": 10722} +{"lr": 8.547114525535937e-05, "data_time": 0.0011605262756347657, "loss": 0.25198017954826357, "time": 0.7714751482009887, "epoch": 3, "memory": 6318, "step": 10822} diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_125740/vis_data/config.py b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_125740/vis_data/config.py new file mode 100644 index 0000000000000000000000000000000000000000..b35ee250106aa84d1065073f108965cfa202c34d --- /dev/null +++ b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_125740/vis_data/config.py @@ -0,0 +1,366 @@ +optim_wrapper = dict( + optimizer=dict( + type='SGD', + lr=0.0001, + momentum=0.9, + weight_decay=0.0001, + _scope_='mmpretrain'), + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=False, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=512) +model = dict( + type='ImageClassifier', + backbone=dict( + frozen_stages=2, + type='ResNet', + depth=50, + num_stages=4, + out_indices=(3, ), + style='pytorch', + init_cfg=dict( + type='Pretrained', + checkpoint= + 'https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth', + prefix='backbone')), + neck=dict(type='GlobalAveragePooling'), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=2048, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + topk=1)) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/all1.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/all2.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/all3.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3fake8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/cc1m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3real8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV2-1-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/if-dpmsolver++-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = dict(type='Accuracy', topk=1) +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV2-1-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/if-dpmsolver++-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = dict(type='Accuracy', topk=1) +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1' diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_125740/vis_data/events.out.tfevents.1685941074.SH-IDC1-10-140-24-132.170636.0 b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_125740/vis_data/events.out.tfevents.1685941074.SH-IDC1-10-140-24-132.170636.0 new file mode 100644 index 0000000000000000000000000000000000000000..f9c290131c1533c03e06995a55649443f2d1bfd3 --- /dev/null +++ b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_125740/vis_data/events.out.tfevents.1685941074.SH-IDC1-10-140-24-132.170636.0 @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:36fd67503838bfa5edf232c55b4c955b8201f34ae6048e1857c3139761938468 +size 1113828 diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_125740/vis_data/scalars.json b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_125740/vis_data/scalars.json new file mode 100644 index 0000000000000000000000000000000000000000..cdbeba31b0196004977e47409c7542c58955ab80 --- /dev/null +++ b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_125740/vis_data/scalars.json @@ -0,0 +1,110 @@ +{"lr": 9.999871217862936e-05, "data_time": 0.20409960746765138, "loss": 0.6428110539913178, "time": 0.7742290496826172, "epoch": 1, "memory": 9436, "step": 100} +{"lr": 9.999479662539605e-05, "data_time": 0.3893429279327393, "loss": 0.5992599487304687, "time": 0.7533838510513305, "epoch": 1, "memory": 6319, "step": 200} +{"lr": 9.99882534375654e-05, "data_time": 0.17300114631652833, "loss": 0.5666563391685486, "time": 0.7856077432632447, "epoch": 1, "memory": 6319, "step": 300} +{"lr": 9.997908299725075e-05, "data_time": 0.3531585931777954, "loss": 0.5496869027614594, "time": 0.8012889385223388, "epoch": 1, "memory": 6319, "step": 400} +{"lr": 9.996728583999318e-05, "data_time": 0.24204785823822023, "loss": 0.5152768701314926, "time": 0.8361482620239258, "epoch": 1, "memory": 6319, "step": 500} +{"lr": 9.995286265473058e-05, "data_time": 0.0007352590560913086, "loss": 0.5025729775428772, "time": 0.6796221256256103, "epoch": 1, "memory": 6319, "step": 600} +{"lr": 9.993581428375725e-05, "data_time": 0.0007589817047119141, "loss": 0.4584820032119751, "time": 0.6835838794708252, "epoch": 1, "memory": 6319, "step": 700} +{"lr": 9.991614172267466e-05, "data_time": 0.0006952524185180664, "loss": 0.4638307332992554, "time": 0.8038980722427368, "epoch": 1, "memory": 6319, "step": 800} +{"lr": 9.989384612033355e-05, "data_time": 0.0010694503784179688, "loss": 0.4339237719774246, "time": 0.7463749170303344, "epoch": 1, "memory": 6319, "step": 900} +{"lr": 9.986892877876658e-05, "data_time": 0.0009260177612304688, "loss": 0.44686152040958405, "time": 0.6744317770004272, "epoch": 1, "memory": 6319, "step": 1000} +{"lr": 9.984139115311244e-05, "data_time": 0.0009229660034179687, "loss": 0.4297418177127838, "time": 0.7513756513595581, "epoch": 1, "memory": 6319, "step": 1100} +{"lr": 9.981123485153081e-05, "data_time": 0.0008283376693725586, "loss": 0.41301066875457765, "time": 0.6751856565475464, "epoch": 1, "memory": 6319, "step": 1200} +{"lr": 9.977846163510851e-05, "data_time": 0.0009643316268920898, "loss": 0.4018182665109634, "time": 0.7549192667007446, "epoch": 1, "memory": 6319, "step": 1300} +{"lr": 9.974307341775653e-05, "data_time": 0.0009546279907226562, "loss": 0.4013751924037933, "time": 0.7881002426147461, "epoch": 1, "memory": 6319, "step": 1400} +{"lr": 9.970507226609833e-05, "data_time": 0.0009330987930297851, "loss": 0.40516377389431, "time": 0.7475346088409424, "epoch": 1, "memory": 6319, "step": 1500} +{"lr": 9.966446039934934e-05, "data_time": 0.0006869554519653321, "loss": 0.4016081541776657, "time": 0.7958840370178223, "epoch": 1, "memory": 6319, "step": 1600} +{"lr": 9.962124018918683e-05, "data_time": 0.0009645938873291016, "loss": 0.3792893081903458, "time": 0.7258940458297729, "epoch": 1, "memory": 6319, "step": 1700} +{"lr": 9.957541415961212e-05, "data_time": 0.0007101774215698242, "loss": 0.3755451083183289, "time": 0.7215308427810669, "epoch": 1, "memory": 6319, "step": 1800} +{"lr": 9.952698498680267e-05, "data_time": 0.001020526885986328, "loss": 0.3782971054315567, "time": 0.7829531908035279, "epoch": 1, "memory": 6319, "step": 1900} +{"lr": 9.947595549895583e-05, "data_time": 0.0008440017700195312, "loss": 0.4020328789949417, "time": 0.7923438072204589, "epoch": 1, "memory": 6319, "step": 2000} +{"lr": 9.942232867612417e-05, "data_time": 0.0014200448989868165, "loss": 0.3704873651266098, "time": 0.882388973236084, "epoch": 1, "memory": 6319, "step": 2100} +{"lr": 9.93661076500406e-05, "data_time": 0.0008404970169067383, "loss": 0.3586942821741104, "time": 0.7647693872451782, "epoch": 1, "memory": 6319, "step": 2200} +{"lr": 9.930729570393622e-05, "data_time": 0.0009474754333496094, "loss": 0.3483234286308289, "time": 0.8398890256881714, "epoch": 1, "memory": 6319, "step": 2300} +{"lr": 9.924589627234824e-05, "data_time": 0.0008964061737060547, "loss": 0.3611366808414459, "time": 0.7457397222518921, "epoch": 1, "memory": 6319, "step": 2400} +{"lr": 9.918191294091956e-05, "data_time": 0.0009618759155273438, "loss": 0.3527453541755676, "time": 0.7929868221282959, "epoch": 1, "memory": 6319, "step": 2500} +{"lr": 9.91153494461892e-05, "data_time": 0.0009335756301879882, "loss": 0.3476579636335373, "time": 0.747836422920227, "epoch": 1, "memory": 6319, "step": 2600} +{"lr": 9.904620967537422e-05, "data_time": 0.000835585594177246, "loss": 0.36337176263332366, "time": 0.8190121650695801, "epoch": 1, "memory": 6319, "step": 2700} +{"lr": 9.897449766614277e-05, "data_time": 0.0008677005767822266, "loss": 0.34508435130119325, "time": 0.8433783531188965, "epoch": 1, "memory": 6319, "step": 2800} +{"lr": 9.890021760637804e-05, "data_time": 0.0009719133377075195, "loss": 0.34163249731063844, "time": 0.7544926643371582, "epoch": 1, "memory": 6319, "step": 2900} +{"lr": 9.882337383393409e-05, "data_time": 0.0009678125381469727, "loss": 0.33529525995254517, "time": 0.8980875492095948, "epoch": 1, "memory": 6319, "step": 3000} +{"lr": 9.874397083638193e-05, "data_time": 0.001253199577331543, "loss": 0.3577060431241989, "time": 0.7957884073257446, "epoch": 1, "memory": 6319, "step": 3100} +{"lr": 9.866201325074828e-05, "data_time": 0.0008949756622314454, "loss": 0.33875913321971896, "time": 0.7663751840591431, "epoch": 1, "memory": 6319, "step": 3200} +{"lr": 9.857750586324393e-05, "data_time": 0.0011025190353393555, "loss": 0.34913310110569, "time": 0.8131855010986329, "epoch": 1, "memory": 6319, "step": 3300} +{"lr": 9.84904536089848e-05, "data_time": 0.000963902473449707, "loss": 0.3376062482595444, "time": 0.8211594820022583, "epoch": 1, "memory": 6319, "step": 3400} +{"lr": 9.840086157170351e-05, "data_time": 0.0010434865951538086, "loss": 0.32343386113643646, "time": 0.8311334133148194, "epoch": 1, "memory": 6319, "step": 3500} +{"lr": 9.830873498345255e-05, "data_time": 0.0008426666259765625, "loss": 0.31666031777858733, "time": 0.8696626663208008, "epoch": 1, "memory": 6319, "step": 3600} +{"lr": 9.821407922429864e-05, "data_time": 0.0008366584777832031, "loss": 0.3220408380031586, "time": 0.8764559507369996, "epoch": 1, "memory": 6319, "step": 3700} +{"lr": 9.811689982200877e-05, "data_time": 0.0009337186813354493, "loss": 0.3052338659763336, "time": 0.8391748666763306, "epoch": 1, "memory": 6319, "step": 3800} +{"lr": 9.801720245172702e-05, "data_time": 0.0008714199066162109, "loss": 0.33241842687129974, "time": 0.8501276731491089, "epoch": 1, "memory": 6319, "step": 3900} +{"lr": 9.791499293564358e-05, "data_time": 0.0009468555450439453, "loss": 0.3272052824497223, "time": 0.8162740707397461, "epoch": 1, "memory": 6319, "step": 4000} +{"lr": 9.781027724265439e-05, "data_time": 0.001980876922607422, "loss": 0.3239219099283218, "time": 0.8785770654678344, "epoch": 1, "memory": 6319, "step": 4100} +{"accuracy/top1": 83.56255340576172, "data_time": 0.4138187961418088, "time": 0.504850471720976, "step": 1} +{"lr": 9.769111539369434e-05, "data_time": 0.18262119293212892, "loss": 0.3109478265047073, "time": 0.7789007425308228, "epoch": 2, "memory": 6318, "step": 4211} +{"lr": 9.758113190911623e-05, "data_time": 0.0008503198623657227, "loss": 0.3000048279762268, "time": 0.9014130115509034, "epoch": 2, "memory": 6318, "step": 4311} +{"lr": 9.746866174465984e-05, "data_time": 0.29087731838226316, "loss": 0.3160201162099838, "time": 0.6463502883911133, "epoch": 2, "memory": 6318, "step": 4411} +{"lr": 9.735371146842894e-05, "data_time": 0.08671903610229492, "loss": 0.3168675124645233, "time": 0.7634302377700806, "epoch": 2, "memory": 6318, "step": 4511} +{"lr": 9.723628779336255e-05, "data_time": 0.4065819501876831, "loss": 0.30820069313049314, "time": 0.7674283742904663, "epoch": 2, "memory": 6318, "step": 4611} +{"lr": 9.711639757684265e-05, "data_time": 0.0009097814559936524, "loss": 0.32611868977546693, "time": 0.7092680215835572, "epoch": 2, "memory": 6318, "step": 4711} +{"lr": 9.699404782029395e-05, "data_time": 0.0009825468063354493, "loss": 0.29978167116641996, "time": 0.7092856645584107, "epoch": 2, "memory": 6318, "step": 4811} +{"lr": 9.686924566877498e-05, "data_time": 0.0008087158203125, "loss": 0.3144329011440277, "time": 0.6998948574066162, "epoch": 2, "memory": 6318, "step": 4911} +{"lr": 9.674199841056078e-05, "data_time": 0.0008284330368041992, "loss": 0.2896211877465248, "time": 0.6869336605072022, "epoch": 2, "memory": 6318, "step": 5011} +{"lr": 9.661231347671705e-05, "data_time": 0.0008912563323974609, "loss": 0.2940391093492508, "time": 0.7754358291625977, "epoch": 2, "memory": 6318, "step": 5111} +{"lr": 9.64801984406668e-05, "data_time": 0.0008103132247924804, "loss": 0.29659841358661654, "time": 0.6768763780593872, "epoch": 2, "memory": 6318, "step": 5211} +{"lr": 9.634566101774737e-05, "data_time": 0.0009378433227539062, "loss": 0.30843727588653563, "time": 0.6985715389251709, "epoch": 2, "memory": 6318, "step": 5311} +{"lr": 9.62087090647604e-05, "data_time": 0.0009206056594848633, "loss": 0.2754927545785904, "time": 0.8171066284179688, "epoch": 2, "memory": 6318, "step": 5411} +{"lr": 9.606935057951247e-05, "data_time": 0.0009344816207885742, "loss": 0.28609300702810286, "time": 0.63073251247406, "epoch": 2, "memory": 6318, "step": 5511} +{"lr": 9.592759370034884e-05, "data_time": 0.0011746406555175782, "loss": 0.2763012573122978, "time": 0.70788414478302, "epoch": 2, "memory": 6318, "step": 5611} +{"lr": 9.578344670567721e-05, "data_time": 0.0013096332550048828, "loss": 0.2957438349723816, "time": 0.7025631189346313, "epoch": 2, "memory": 6318, "step": 5711} +{"lr": 9.563691801348495e-05, "data_time": 0.0011096715927124024, "loss": 0.3187422066926956, "time": 0.6899694919586181, "epoch": 2, "memory": 6318, "step": 5811} +{"lr": 9.548801618084738e-05, "data_time": 0.0007781744003295899, "loss": 0.2889773666858673, "time": 0.7827431440353394, "epoch": 2, "memory": 6318, "step": 5911} +{"lr": 9.533674990342786e-05, "data_time": 0.0007563591003417969, "loss": 0.3179330348968506, "time": 0.6677175045013428, "epoch": 2, "memory": 6318, "step": 6011} +{"lr": 9.518312801497029e-05, "data_time": 0.0008057355880737305, "loss": 0.2721507683396339, "time": 0.755538821220398, "epoch": 2, "memory": 6318, "step": 6111} +{"lr": 9.502715948678276e-05, "data_time": 0.0008047580718994141, "loss": 0.29019091129302976, "time": 0.7338272094726562, "epoch": 2, "memory": 6318, "step": 6211} +{"lr": 9.486885342721427e-05, "data_time": 0.0008751630783081054, "loss": 0.2824150174856186, "time": 0.729899549484253, "epoch": 2, "memory": 6318, "step": 6311} +{"lr": 9.470821908112202e-05, "data_time": 0.0009315967559814453, "loss": 0.28303675949573515, "time": 1.1837078332901, "epoch": 2, "memory": 6318, "step": 6411} +{"lr": 9.454526582933214e-05, "data_time": 0.0009738683700561523, "loss": 0.3016718864440918, "time": 0.823495602607727, "epoch": 2, "memory": 6318, "step": 6511} +{"lr": 9.438000318809176e-05, "data_time": 0.0012355089187622071, "loss": 0.29478231221437456, "time": 0.7664000988006592, "epoch": 2, "memory": 6318, "step": 6611} +{"lr": 9.42124408085132e-05, "data_time": 0.0008062124252319336, "loss": 0.29754903316497805, "time": 0.7471588134765625, "epoch": 2, "memory": 6318, "step": 6711} +{"lr": 9.404258847601036e-05, "data_time": 0.0009423017501831055, "loss": 0.2872541010379791, "time": 1.9881983518600463, "epoch": 2, "memory": 6318, "step": 6811} +{"lr": 9.387045610972717e-05, "data_time": 0.0010599851608276366, "loss": 0.29051762670278547, "time": 0.7904786348342896, "epoch": 2, "memory": 6318, "step": 6911} +{"lr": 9.369605376195838e-05, "data_time": 0.0009445667266845704, "loss": 0.29260163754224777, "time": 0.7432496309280395, "epoch": 2, "memory": 6318, "step": 7011} +{"lr": 9.351939161756269e-05, "data_time": 0.0009049415588378906, "loss": 0.30424838662147524, "time": 0.6783438205718995, "epoch": 2, "memory": 6318, "step": 7111} +{"lr": 9.334047999336764e-05, "data_time": 0.0009110450744628906, "loss": 0.2720350831747055, "time": 0.887003231048584, "epoch": 2, "memory": 6318, "step": 7211} +{"lr": 9.315932933756752e-05, "data_time": 0.0009765148162841797, "loss": 0.2902287930250168, "time": 0.7409857749938965, "epoch": 2, "memory": 6318, "step": 7311} +{"lr": 9.297595022911287e-05, "data_time": 0.0008769750595092774, "loss": 0.2734398484230042, "time": 0.7320471286773682, "epoch": 2, "memory": 6318, "step": 7411} +{"lr": 9.279035337709295e-05, "data_time": 0.0008745908737182618, "loss": 0.29733290821313857, "time": 0.7834531784057617, "epoch": 2, "memory": 6318, "step": 7511} +{"lr": 9.260254962011016e-05, "data_time": 0.0008535146713256836, "loss": 0.29611544907093046, "time": 0.793001675605774, "epoch": 2, "memory": 6318, "step": 7611} +{"lr": 9.241254992564702e-05, "data_time": 0.0009152412414550782, "loss": 0.2653490588068962, "time": 0.77921142578125, "epoch": 2, "memory": 6318, "step": 7711} +{"lr": 9.22203653894259e-05, "data_time": 0.0009097576141357422, "loss": 0.28250943422317504, "time": 0.7937284946441651, "epoch": 2, "memory": 6318, "step": 7811} +{"lr": 9.202600723476098e-05, "data_time": 0.000969099998474121, "loss": 0.29540292024612425, "time": 0.7584635734558105, "epoch": 2, "memory": 6318, "step": 7911} +{"lr": 9.182948681190268e-05, "data_time": 0.0009256124496459961, "loss": 0.2653553158044815, "time": 0.7749513387680054, "epoch": 2, "memory": 6318, "step": 8011} +{"lr": 9.163081559737497e-05, "data_time": 0.0011258125305175781, "loss": 0.2730012521147728, "time": 0.7843770265579224, "epoch": 2, "memory": 6318, "step": 8111} +{"lr": 9.143000519330517e-05, "data_time": 0.0016515731811523437, "loss": 0.29417825043201445, "time": 0.8431460380554199, "epoch": 2, "memory": 6318, "step": 8211} +{"accuracy/top1": 80.38992309570312, "data_time": 0.36706274350484214, "time": 0.4556770404179891, "step": 2} +{"lr": 9.120461478753376e-05, "data_time": 0.0007345914840698242, "loss": 0.2649999871850014, "time": 0.685315752029419, "epoch": 3, "memory": 6318, "step": 8322} +{"lr": 9.099932932138134e-05, "data_time": 0.000763702392578125, "loss": 0.2731035828590393, "time": 0.7478806495666503, "epoch": 3, "memory": 6318, "step": 8422} +{"lr": 9.079194154362181e-05, "data_time": 0.0007415533065795899, "loss": 0.2825735807418823, "time": 0.7287466764450073, "epoch": 3, "memory": 6318, "step": 8522} +{"lr": 9.058246356541787e-05, "data_time": 0.0011133670806884766, "loss": 0.2907098770141602, "time": 0.7379800319671631, "epoch": 3, "memory": 6318, "step": 8622} +{"lr": 9.037090761999699e-05, "data_time": 0.0009167671203613281, "loss": 0.2684899285435677, "time": 0.6742784976959229, "epoch": 3, "memory": 6318, "step": 8722} +{"lr": 9.015728606193701e-05, "data_time": 0.10558381080627441, "loss": 0.2739625841379166, "time": 0.7187713384628296, "epoch": 3, "memory": 6318, "step": 8822} +{"lr": 8.994161136644488e-05, "data_time": 0.3643131494522095, "loss": 0.28152443170547486, "time": 0.7447021484375, "epoch": 3, "memory": 6318, "step": 8922} +{"lr": 8.972389612862792e-05, "data_time": 0.6017848014831543, "loss": 0.2692335307598114, "time": 0.7448285102844239, "epoch": 3, "memory": 6318, "step": 9022} +{"lr": 8.950415306275833e-05, "data_time": 0.5831636428833008, "loss": 0.2781596094369888, "time": 0.769466757774353, "epoch": 3, "memory": 6318, "step": 9122} +{"lr": 8.928239500153059e-05, "data_time": 0.5470522165298461, "loss": 0.2762856870889664, "time": 0.6863395214080811, "epoch": 3, "memory": 6318, "step": 9222} +{"lr": 8.905863489531232e-05, "data_time": 0.5433248043060303, "loss": 0.2705572977662086, "time": 0.6887992858886719, "epoch": 3, "memory": 6318, "step": 9322} +{"lr": 8.883288581138775e-05, "data_time": 0.5490938901901246, "loss": 0.27692133486270903, "time": 0.6886429071426392, "epoch": 3, "memory": 6318, "step": 9422} +{"lr": 8.86051609331947e-05, "data_time": 0.5289304971694946, "loss": 0.2668500617146492, "time": 0.6792001485824585, "epoch": 3, "memory": 6318, "step": 9522} +{"lr": 8.837547355955467e-05, "data_time": 0.4923022508621216, "loss": 0.26194403767585756, "time": 0.633077073097229, "epoch": 3, "memory": 6318, "step": 9622} +{"lr": 8.814383710389614e-05, "data_time": 0.3787827730178833, "loss": 0.27189348489046095, "time": 0.6949618816375732, "epoch": 3, "memory": 6318, "step": 9722} +{"lr": 8.791026509347146e-05, "data_time": 0.5856317520141602, "loss": 0.2663997381925583, "time": 0.724916672706604, "epoch": 3, "memory": 6318, "step": 9822} +{"lr": 8.767477116856659e-05, "data_time": 0.43811945915222167, "loss": 0.2720366507768631, "time": 0.6689430236816406, "epoch": 3, "memory": 6318, "step": 9922} +{"lr": 8.743736908170466e-05, "data_time": 0.16061439514160156, "loss": 0.2646900206804276, "time": 0.6393389463424682, "epoch": 3, "memory": 6318, "step": 10022} +{"lr": 8.719807269684305e-05, "data_time": 0.0008947849273681641, "loss": 0.2876169040799141, "time": 0.698579216003418, "epoch": 3, "memory": 6318, "step": 10122} +{"lr": 8.695689598856356e-05, "data_time": 0.000968170166015625, "loss": 0.2736373618245125, "time": 0.63644380569458, "epoch": 3, "memory": 6318, "step": 10222} +{"lr": 8.67138530412559e-05, "data_time": 0.0009461641311645508, "loss": 0.263558404147625, "time": 0.7714944124221802, "epoch": 3, "memory": 6318, "step": 10322} +{"lr": 8.646895804829604e-05, "data_time": 0.0009682655334472656, "loss": 0.24974735379219054, "time": 0.773975419998169, "epoch": 3, "memory": 6318, "step": 10422} +{"lr": 8.622222531121664e-05, "data_time": 0.0013257265090942383, "loss": 0.2685848161578178, "time": 0.6491158723831176, "epoch": 3, "memory": 6318, "step": 10522} +{"lr": 8.597366923887211e-05, "data_time": 0.0007126808166503906, "loss": 0.26895226538181305, "time": 0.6310829401016236, "epoch": 3, "memory": 6318, "step": 10622} +{"lr": 8.57233043465972e-05, "data_time": 0.0009105682373046875, "loss": 0.2824095621705055, "time": 0.695707893371582, "epoch": 3, "memory": 6318, "step": 10722} +{"lr": 8.547114525535937e-05, "data_time": 0.0011605262756347657, "loss": 0.25198017954826357, "time": 0.7714751482009887, "epoch": 3, "memory": 6318, "step": 10822} diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_125740/vis_data/vis_image/0000000.png_1.png b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_125740/vis_data/vis_image/0000000.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..dc5bf307fa0ed5b0887e2f188cdadddbf8d5863f Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_125740/vis_data/vis_image/0000000.png_1.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_125740/vis_data/vis_image/0000000.png_2.png b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_125740/vis_data/vis_image/0000000.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..1abd2f81b225afbe6b6b8f56e70920d9ce8da385 Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_125740/vis_data/vis_image/0000000.png_2.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_125740/vis_data/vis_image/000003093.jpg_1.png b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_125740/vis_data/vis_image/000003093.jpg_1.png new file mode 100644 index 0000000000000000000000000000000000000000..98fac0079b08f9ae57a809cfabb64096a0f0d042 Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_125740/vis_data/vis_image/000003093.jpg_1.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_125740/vis_data/vis_image/000003093.jpg_2.png b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_125740/vis_data/vis_image/000003093.jpg_2.png new file mode 100644 index 0000000000000000000000000000000000000000..9db89d333156235b99def79b58580fbe7dabf6a4 Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_125740/vis_data/vis_image/000003093.jpg_2.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_125740/vis_data/vis_image/000015337.jpg_1.png b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_125740/vis_data/vis_image/000015337.jpg_1.png new file mode 100644 index 0000000000000000000000000000000000000000..0bbe63bf1eb4b247408dca3327ff140d1ba9d156 Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_125740/vis_data/vis_image/000015337.jpg_1.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_125740/vis_data/vis_image/000015337.jpg_2.png b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_125740/vis_data/vis_image/000015337.jpg_2.png new file mode 100644 index 0000000000000000000000000000000000000000..26c40a68364696a08d5eda77c80acae05680a497 Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_125740/vis_data/vis_image/000015337.jpg_2.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_125740/vis_data/vis_image/0004159.png_1.png b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_125740/vis_data/vis_image/0004159.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..18c6d5adf418f2ff11ab3324c8c55b5efd48438c Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_125740/vis_data/vis_image/0004159.png_1.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_125740/vis_data/vis_image/0004159.png_2.png b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_125740/vis_data/vis_image/0004159.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..92af1283a7615452a069915a63393d2c6c2ca387 Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_125740/vis_data/vis_image/0004159.png_2.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_125740/vis_data/vis_image/0008318.png_1.png b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_125740/vis_data/vis_image/0008318.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..d0bad16a72f6f3ef5808749c48d3e50ab1e65936 Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_125740/vis_data/vis_image/0008318.png_1.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_125740/vis_data/vis_image/0008318.png_2.png b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_125740/vis_data/vis_image/0008318.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..d11662d720bb86fc37979d43303e1b9f8a1d92ff Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_125740/vis_data/vis_image/0008318.png_2.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_125740/vis_data/vis_image/0010000.png_1.png b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_125740/vis_data/vis_image/0010000.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..c3a2949189097b04a29efe5cb3f5c2046baa6867 Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_125740/vis_data/vis_image/0010000.png_1.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_125740/vis_data/vis_image/0010000.png_2.png b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_125740/vis_data/vis_image/0010000.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..a313ceb10152772041219181b8455a107e65a054 Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_125740/vis_data/vis_image/0010000.png_2.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_125740/vis_data/vis_image/0014159.png_1.png b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_125740/vis_data/vis_image/0014159.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..54a1647a1cbb94fdf221f6dedb2e8343d05c8f2a Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_125740/vis_data/vis_image/0014159.png_1.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_125740/vis_data/vis_image/0014159.png_2.png b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_125740/vis_data/vis_image/0014159.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..875cc2332dfc8c4aa2d3284e5fbec3f13018f786 Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_125740/vis_data/vis_image/0014159.png_2.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/20230605_152040.log b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/20230605_152040.log new file mode 100644 index 0000000000000000000000000000000000000000..b6858e7d22e46a762ffe76e1e4034d4e9963c2d1 --- /dev/null +++ b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/20230605_152040.log @@ -0,0 +1,1473 @@ +2023/06/05 15:20:43 - mmengine - INFO - +------------------------------------------------------------ +System environment: + sys.platform: linux + Python: 3.10.9 (main, Mar 8 2023, 10:47:38) [GCC 11.2.0] + CUDA available: True + numpy_random_seed: 2106773263 + GPU 0,1: NVIDIA A100-SXM4-80GB + CUDA_HOME: /mnt/petrelfs/share/cuda-11.6 + NVCC: Cuda compilation tools, release 11.6, V11.6.124 + GCC: gcc (GCC) 7.5.0 + PyTorch: 1.13.1 + PyTorch compiling details: PyTorch built with: + - GCC 9.3 + - C++ Version: 201402 + - Intel(R) oneAPI Math Kernel Library Version 2021.4-Product Build 20210904 for Intel(R) 64 architecture applications + - Intel(R) MKL-DNN v2.6.0 (Git Hash 52b5f107dd9cf10910aaa19cb47f3abf9b349815) + - OpenMP 201511 (a.k.a. OpenMP 4.5) + - LAPACK is enabled (usually provided by MKL) + - NNPACK is enabled + - CPU capability usage: AVX2 + - CUDA Runtime 11.6 + - NVCC architecture flags: -gencode;arch=compute_37,code=sm_37;-gencode;arch=compute_50,code=sm_50;-gencode;arch=compute_60,code=sm_60;-gencode;arch=compute_61,code=sm_61;-gencode;arch=compute_70,code=sm_70;-gencode;arch=compute_75,code=sm_75;-gencode;arch=compute_80,code=sm_80;-gencode;arch=compute_86,code=sm_86;-gencode;arch=compute_37,code=compute_37 + - CuDNN 8.3.2 (built against CUDA 11.5) + - Magma 2.6.1 + - Build settings: BLAS_INFO=mkl, BUILD_TYPE=Release, CUDA_VERSION=11.6, CUDNN_VERSION=8.3.2, CXX_COMPILER=/opt/rh/devtoolset-9/root/usr/bin/c++, CXX_FLAGS= -fabi-version=11 -Wno-deprecated -fvisibility-inlines-hidden -DUSE_PTHREADPOOL -fopenmp -DNDEBUG -DUSE_KINETO -DUSE_FBGEMM -DUSE_QNNPACK -DUSE_PYTORCH_QNNPACK -DUSE_XNNPACK -DSYMBOLICATE_MOBILE_DEBUG_HANDLE -DEDGE_PROFILER_USE_KINETO -O2 -fPIC -Wno-narrowing -Wall -Wextra -Werror=return-type -Werror=non-virtual-dtor -Wno-missing-field-initializers -Wno-type-limits -Wno-array-bounds -Wno-unknown-pragmas -Wunused-local-typedefs -Wno-unused-parameter -Wno-unused-function -Wno-unused-result -Wno-strict-overflow -Wno-strict-aliasing -Wno-error=deprecated-declarations -Wno-stringop-overflow -Wno-psabi -Wno-error=pedantic -Wno-error=redundant-decls -Wno-error=old-style-cast -fdiagnostics-color=always -faligned-new -Wno-unused-but-set-variable -Wno-maybe-uninitialized -fno-math-errno -fno-trapping-math -Werror=format -Werror=cast-function-type -Wno-stringop-overflow, LAPACK_INFO=mkl, PERF_WITH_AVX=1, PERF_WITH_AVX2=1, PERF_WITH_AVX512=1, TORCH_VERSION=1.13.1, USE_CUDA=ON, USE_CUDNN=ON, USE_EXCEPTION_PTR=1, USE_GFLAGS=OFF, USE_GLOG=OFF, USE_MKL=ON, USE_MKLDNN=ON, USE_MPI=OFF, USE_NCCL=ON, USE_NNPACK=ON, USE_OPENMP=ON, USE_ROCM=OFF, + + TorchVision: 0.14.1 + OpenCV: 4.7.0 + MMEngine: 0.7.3 + +Runtime environment: + cudnn_benchmark: True + mp_cfg: {'mp_start_method': 'fork', 'opencv_num_threads': 0} + dist_cfg: {'backend': 'nccl'} + seed: None + deterministic: False + Distributed launcher: slurm + Distributed training: True + GPU number: 2 +------------------------------------------------------------ + +2023/06/05 15:20:48 - mmengine - INFO - Config: +optim_wrapper = dict( + optimizer=dict( + type='SGD', + lr=0.0001, + momentum=0.9, + weight_decay=0.0001, + _scope_='mmpretrain'), + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=False, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=512) +model = dict( + type='ImageClassifier', + backbone=dict( + frozen_stages=2, + type='ResNet', + depth=50, + num_stages=4, + out_indices=(3, ), + style='pytorch', + init_cfg=dict( + type='Pretrained', + checkpoint= + 'https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth', + prefix='backbone')), + neck=dict(type='GlobalAveragePooling'), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=2048, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + topk=1)) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/all1.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/all2.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/all3.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3fake8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/cc1m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3real8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV2-1-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/if-dpmsolver++-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = dict(type='Accuracy', topk=1) +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV2-1-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/if-dpmsolver++-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = dict(type='Accuracy', topk=1) +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1' + +2023/06/05 15:20:59 - mmengine - INFO - Hooks will be executed in the following order: +before_run: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_load_checkpoint: +(ABOVE_NORMAL) EMAHook + -------------------- +before_train: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_train_epoch: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(NORMAL ) DistSamplerSeedHook + -------------------- +before_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook + -------------------- +after_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +after_train_epoch: +(NORMAL ) IterTimerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_val_epoch: +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook + -------------------- +before_val_iter: +(NORMAL ) IterTimerHook + -------------------- +after_val_iter: +(NORMAL ) IterTimerHook +(NORMAL ) VisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_val_epoch: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_save_checkpoint: +(ABOVE_NORMAL) EMAHook + -------------------- +after_train: +(VERY_LOW ) CheckpointHook + -------------------- +before_test_epoch: +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook + -------------------- +before_test_iter: +(NORMAL ) IterTimerHook + -------------------- +after_test_iter: +(NORMAL ) IterTimerHook +(NORMAL ) VisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_test_epoch: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_run: +(BELOW_NORMAL) LoggerHook + -------------------- +2023/06/05 15:21:19 - mmengine - INFO - load backbone in model from: https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth +Name of parameter - Initialization information + +backbone.conv1.weight - torch.Size([64, 3, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.bn1.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.bn1.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.conv1.weight - torch.Size([64, 64, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.bn1.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.bn1.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.conv2.weight - torch.Size([64, 64, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.bn2.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.bn2.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.conv3.weight - torch.Size([256, 64, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.bn3.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.bn3.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.downsample.0.weight - torch.Size([256, 64, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.downsample.1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.downsample.1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.conv1.weight - torch.Size([64, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.bn1.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.bn1.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.conv2.weight - torch.Size([64, 64, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.bn2.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.bn2.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.conv3.weight - torch.Size([256, 64, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.bn3.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.bn3.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.conv1.weight - torch.Size([64, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.bn1.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.bn1.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.conv2.weight - torch.Size([64, 64, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.bn2.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.bn2.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.conv3.weight - torch.Size([256, 64, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.bn3.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.bn3.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.conv1.weight - torch.Size([128, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.bn1.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.bn1.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.conv2.weight - torch.Size([128, 128, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.bn2.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.bn2.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.conv3.weight - torch.Size([512, 128, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.bn3.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.bn3.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.downsample.0.weight - torch.Size([512, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.downsample.1.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.downsample.1.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.conv1.weight - torch.Size([128, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.bn1.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.bn1.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.conv2.weight - torch.Size([128, 128, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.bn2.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.bn2.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.conv3.weight - torch.Size([512, 128, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.bn3.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.bn3.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.conv1.weight - torch.Size([128, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.bn1.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.bn1.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.conv2.weight - torch.Size([128, 128, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.bn2.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.bn2.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.conv3.weight - torch.Size([512, 128, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.bn3.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.bn3.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.conv1.weight - torch.Size([128, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.bn1.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.bn1.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.conv2.weight - torch.Size([128, 128, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.bn2.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.bn2.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.conv3.weight - torch.Size([512, 128, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.bn3.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.bn3.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.conv1.weight - torch.Size([256, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.bn1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.bn1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.conv2.weight - torch.Size([256, 256, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.bn2.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.bn2.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.conv3.weight - torch.Size([1024, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.bn3.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.bn3.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.downsample.0.weight - torch.Size([1024, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.downsample.1.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.downsample.1.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.conv1.weight - torch.Size([256, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.bn1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.bn1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.conv2.weight - torch.Size([256, 256, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.bn2.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.bn2.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.conv3.weight - torch.Size([1024, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.bn3.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.bn3.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.conv1.weight - torch.Size([256, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.bn1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.bn1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.conv2.weight - torch.Size([256, 256, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.bn2.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.bn2.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.conv3.weight - torch.Size([1024, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.bn3.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.bn3.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.conv1.weight - torch.Size([256, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.bn1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.bn1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.conv2.weight - torch.Size([256, 256, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.bn2.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.bn2.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.conv3.weight - torch.Size([1024, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.bn3.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.bn3.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.conv1.weight - torch.Size([256, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.bn1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.bn1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.conv2.weight - torch.Size([256, 256, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.bn2.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.bn2.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.conv3.weight - torch.Size([1024, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.bn3.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.bn3.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.conv1.weight - torch.Size([256, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.bn1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.bn1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.conv2.weight - torch.Size([256, 256, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.bn2.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.bn2.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.conv3.weight - torch.Size([1024, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.bn3.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.bn3.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.conv1.weight - torch.Size([512, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.bn1.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.bn1.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.conv2.weight - torch.Size([512, 512, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.bn2.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.bn2.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.conv3.weight - torch.Size([2048, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.bn3.weight - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.bn3.bias - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.downsample.0.weight - torch.Size([2048, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.downsample.1.weight - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.downsample.1.bias - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.conv1.weight - torch.Size([512, 2048, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.bn1.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.bn1.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.conv2.weight - torch.Size([512, 512, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.bn2.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.bn2.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.conv3.weight - torch.Size([2048, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.bn3.weight - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.bn3.bias - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.conv1.weight - torch.Size([512, 2048, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.bn1.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.bn1.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.conv2.weight - torch.Size([512, 512, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.bn2.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.bn2.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.conv3.weight - torch.Size([2048, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.bn3.weight - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.bn3.bias - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +head.fc.weight - torch.Size([2, 2048]): +NormalInit: mean=0, std=0.01, bias=0 + +head.fc.bias - torch.Size([2]): +NormalInit: mean=0, std=0.01, bias=0 +2023/06/05 15:21:19 - mmengine - WARNING - "FileClient" will be deprecated in future. Please use io functions in https://mmengine.readthedocs.io/en/latest/api/fileio.html#file-io +2023/06/05 15:21:19 - mmengine - WARNING - "HardDiskBackend" is the alias of "LocalBackend" and the former will be deprecated in future. +2023/06/05 15:21:19 - mmengine - INFO - Checkpoints will be saved to /mnt/petrelfs/luzeyu/workspace/fakebench/mmpretrain/workdir/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1. +2023/06/05 15:22:35 - mmengine - INFO - Epoch(train) [1][ 100/4111] lr: 9.9999e-05 eta: 8:38:26 time: 0.6935 data_time: 0.5506 memory: 9436 loss: 0.6506 +2023/06/05 15:23:45 - mmengine - INFO - Epoch(train) [1][ 200/4111] lr: 9.9995e-05 eta: 8:17:30 time: 0.6552 data_time: 0.5153 memory: 6319 loss: 0.5974 +2023/06/05 15:25:14 - mmengine - INFO - Epoch(train) [1][ 300/4111] lr: 9.9988e-05 eta: 8:52:42 time: 0.6513 data_time: 0.0010 memory: 6319 loss: 0.5777 +2023/06/05 15:26:19 - mmengine - INFO - Epoch(train) [1][ 400/4111] lr: 9.9979e-05 eta: 8:28:23 time: 0.6429 data_time: 0.0009 memory: 6319 loss: 0.5506 +2023/06/05 15:27:25 - mmengine - INFO - Epoch(train) [1][ 500/4111] lr: 9.9967e-05 eta: 8:15:50 time: 0.6031 data_time: 0.2106 memory: 6319 loss: 0.5142 +2023/06/05 15:28:36 - mmengine - INFO - Epoch(train) [1][ 600/4111] lr: 9.9953e-05 eta: 8:11:12 time: 0.7148 data_time: 0.3470 memory: 6319 loss: 0.5076 +2023/06/05 15:29:46 - mmengine - INFO - Epoch(train) [1][ 700/4111] lr: 9.9936e-05 eta: 8:08:10 time: 0.7294 data_time: 0.4602 memory: 6319 loss: 0.4696 +2023/06/05 15:30:57 - mmengine - INFO - Epoch(train) [1][ 800/4111] lr: 9.9916e-05 eta: 8:05:35 time: 0.6684 data_time: 0.4658 memory: 6319 loss: 0.4617 +2023/06/05 15:32:07 - mmengine - INFO - Epoch(train) [1][ 900/4111] lr: 9.9894e-05 eta: 8:02:40 time: 0.7080 data_time: 0.4330 memory: 6319 loss: 0.4414 +2023/06/05 15:33:16 - mmengine - INFO - Exp name: resnet50_2xb256_all_1m_lr1e-4_aug_5e-1_20230605_152040 +2023/06/05 15:33:16 - mmengine - INFO - Epoch(train) [1][1000/4111] lr: 9.9869e-05 eta: 7:59:10 time: 0.6917 data_time: 0.4350 memory: 6319 loss: 0.4226 +2023/06/05 15:34:26 - mmengine - INFO - Epoch(train) [1][1100/4111] lr: 9.9841e-05 eta: 7:57:16 time: 0.6448 data_time: 0.5045 memory: 6319 loss: 0.4144 +2023/06/05 15:35:37 - mmengine - INFO - Epoch(train) [1][1200/4111] lr: 9.9811e-05 eta: 7:55:26 time: 0.6686 data_time: 0.2969 memory: 6319 loss: 0.4110 +2023/06/05 15:36:47 - mmengine - INFO - Epoch(train) [1][1300/4111] lr: 9.9778e-05 eta: 7:53:26 time: 0.6431 data_time: 0.1998 memory: 6319 loss: 0.4085 +2023/06/05 15:38:00 - mmengine - INFO - Epoch(train) [1][1400/4111] lr: 9.9743e-05 eta: 7:53:24 time: 0.6990 data_time: 0.5204 memory: 6319 loss: 0.3970 +2023/06/05 15:39:16 - mmengine - INFO - Epoch(train) [1][1500/4111] lr: 9.9705e-05 eta: 7:53:50 time: 0.6899 data_time: 0.5476 memory: 6319 loss: 0.3997 +2023/06/05 15:40:33 - mmengine - INFO - Epoch(train) [1][1600/4111] lr: 9.9664e-05 eta: 7:54:50 time: 0.7155 data_time: 0.5750 memory: 6319 loss: 0.3951 +2023/06/05 15:41:42 - mmengine - INFO - Epoch(train) [1][1700/4111] lr: 9.9621e-05 eta: 7:52:33 time: 0.6685 data_time: 0.5280 memory: 6319 loss: 0.3815 +2023/06/05 15:42:52 - mmengine - INFO - Epoch(train) [1][1800/4111] lr: 9.9575e-05 eta: 7:50:41 time: 0.6942 data_time: 0.5536 memory: 6319 loss: 0.3952 +2023/06/05 15:44:02 - mmengine - INFO - Epoch(train) [1][1900/4111] lr: 9.9527e-05 eta: 7:48:38 time: 0.6390 data_time: 0.4987 memory: 6319 loss: 0.3808 +2023/06/05 15:45:14 - mmengine - INFO - Exp name: resnet50_2xb256_all_1m_lr1e-4_aug_5e-1_20230605_152040 +2023/06/05 15:45:14 - mmengine - INFO - Epoch(train) [1][2000/4111] lr: 9.9476e-05 eta: 7:47:44 time: 0.7453 data_time: 0.6053 memory: 6319 loss: 0.3601 +2023/06/05 15:46:24 - mmengine - INFO - Epoch(train) [1][2100/4111] lr: 9.9422e-05 eta: 7:46:05 time: 0.7055 data_time: 0.5653 memory: 6319 loss: 0.3705 +2023/06/05 15:47:37 - mmengine - INFO - Epoch(train) [1][2200/4111] lr: 9.9366e-05 eta: 7:45:08 time: 0.6807 data_time: 0.5399 memory: 6319 loss: 0.3434 +2023/06/05 15:48:46 - mmengine - INFO - Epoch(train) [1][2300/4111] lr: 9.9307e-05 eta: 7:43:17 time: 0.6772 data_time: 0.5331 memory: 6319 loss: 0.3587 +2023/06/05 15:49:56 - mmengine - INFO - Epoch(train) [1][2400/4111] lr: 9.9246e-05 eta: 7:41:32 time: 0.6926 data_time: 0.5513 memory: 6319 loss: 0.3472 +2023/06/05 15:51:07 - mmengine - INFO - Epoch(train) [1][2500/4111] lr: 9.9182e-05 eta: 7:40:15 time: 0.7094 data_time: 0.5681 memory: 6319 loss: 0.3582 +2023/06/05 15:52:17 - mmengine - INFO - Epoch(train) [1][2600/4111] lr: 9.9115e-05 eta: 7:38:35 time: 0.6607 data_time: 0.5202 memory: 6319 loss: 0.3443 +2023/06/05 15:55:00 - mmengine - INFO - Epoch(train) [1][2700/4111] lr: 9.9046e-05 eta: 7:59:12 time: 0.6853 data_time: 0.5454 memory: 6319 loss: 0.3530 +2023/06/05 15:56:15 - mmengine - INFO - Epoch(train) [1][2800/4111] lr: 9.8974e-05 eta: 7:57:50 time: 0.7319 data_time: 0.5923 memory: 6319 loss: 0.3269 +2023/06/05 15:57:27 - mmengine - INFO - Epoch(train) [1][2900/4111] lr: 9.8900e-05 eta: 7:56:04 time: 0.7287 data_time: 0.5892 memory: 6319 loss: 0.3432 +2023/06/05 15:58:39 - mmengine - INFO - Exp name: resnet50_2xb256_all_1m_lr1e-4_aug_5e-1_20230605_152040 +2023/06/05 15:58:39 - mmengine - INFO - Epoch(train) [1][3000/4111] lr: 9.8823e-05 eta: 7:54:14 time: 0.7732 data_time: 0.6335 memory: 6319 loss: 0.3493 +2023/06/05 15:59:50 - mmengine - INFO - Epoch(train) [1][3100/4111] lr: 9.8744e-05 eta: 7:52:12 time: 0.6853 data_time: 0.5344 memory: 6319 loss: 0.3424 +2023/06/05 16:01:02 - mmengine - INFO - Epoch(train) [1][3200/4111] lr: 9.8662e-05 eta: 7:50:24 time: 0.6780 data_time: 0.5382 memory: 6319 loss: 0.3396 +2023/06/05 16:02:13 - mmengine - INFO - Epoch(train) [1][3300/4111] lr: 9.8578e-05 eta: 7:48:37 time: 0.7079 data_time: 0.5682 memory: 6319 loss: 0.3353 +2023/06/05 16:03:24 - mmengine - INFO - Epoch(train) [1][3400/4111] lr: 9.8490e-05 eta: 7:46:47 time: 0.8602 data_time: 0.7195 memory: 6319 loss: 0.3274 +2023/06/05 16:04:34 - mmengine - INFO - Epoch(train) [1][3500/4111] lr: 9.8401e-05 eta: 7:44:43 time: 0.7278 data_time: 0.5882 memory: 6319 loss: 0.3180 +2023/06/05 16:05:44 - mmengine - INFO - Epoch(train) [1][3600/4111] lr: 9.8309e-05 eta: 7:42:43 time: 0.6458 data_time: 0.5059 memory: 6319 loss: 0.3155 +2023/06/05 16:06:53 - mmengine - INFO - Epoch(train) [1][3700/4111] lr: 9.8214e-05 eta: 7:40:37 time: 0.6877 data_time: 0.5475 memory: 6319 loss: 0.3263 +2023/06/05 16:08:03 - mmengine - INFO - Epoch(train) [1][3800/4111] lr: 9.8117e-05 eta: 7:38:47 time: 0.7033 data_time: 0.5631 memory: 6319 loss: 0.3218 +2023/06/05 16:09:12 - mmengine - INFO - Epoch(train) [1][3900/4111] lr: 9.8017e-05 eta: 7:36:49 time: 0.6539 data_time: 0.5132 memory: 6319 loss: 0.3055 +2023/06/05 16:10:22 - mmengine - INFO - Exp name: resnet50_2xb256_all_1m_lr1e-4_aug_5e-1_20230605_152040 +2023/06/05 16:10:22 - mmengine - INFO - Epoch(train) [1][4000/4111] lr: 9.7915e-05 eta: 7:35:06 time: 0.6876 data_time: 0.5459 memory: 6319 loss: 0.3265 +2023/06/05 16:11:34 - mmengine - INFO - Epoch(train) [1][4100/4111] lr: 9.7810e-05 eta: 7:33:31 time: 0.8311 data_time: 0.6899 memory: 6319 loss: 0.3121 +2023/06/05 16:11:41 - mmengine - INFO - Exp name: resnet50_2xb256_all_1m_lr1e-4_aug_5e-1_20230605_152040 +2023/06/05 16:11:41 - mmengine - INFO - Saving checkpoint at 1 epochs +2023/06/05 16:12:30 - mmengine - INFO - Epoch(val) [1][100/119] eta: 0:00:08 time: 0.6680 data_time: 0.5794 memory: 8106 +2023/06/05 16:12:56 - mmengine - INFO - Epoch(val) [1][119/119] accuracy/top1: 82.4736 data_time: 0.3936 time: 0.4818 +2023/06/05 16:14:08 - mmengine - INFO - Epoch(train) [2][ 100/4111] lr: 9.7691e-05 eta: 7:31:59 time: 0.7481 data_time: 0.6067 memory: 6318 loss: 0.3242 +2023/06/05 16:15:18 - mmengine - INFO - Epoch(train) [2][ 200/4111] lr: 9.7581e-05 eta: 7:30:11 time: 0.7165 data_time: 0.5759 memory: 6318 loss: 0.3020 +2023/06/05 16:16:28 - mmengine - INFO - Epoch(train) [2][ 300/4111] lr: 9.7469e-05 eta: 7:28:30 time: 0.6634 data_time: 0.3816 memory: 6318 loss: 0.3156 +2023/06/05 16:17:37 - mmengine - INFO - Epoch(train) [2][ 400/4111] lr: 9.7354e-05 eta: 7:26:45 time: 0.6953 data_time: 0.5546 memory: 6318 loss: 0.3117 +2023/06/05 16:18:49 - mmengine - INFO - Epoch(train) [2][ 500/4111] lr: 9.7236e-05 eta: 7:25:15 time: 0.7055 data_time: 0.5643 memory: 6318 loss: 0.3175 +2023/06/05 16:20:00 - mmengine - INFO - Epoch(train) [2][ 600/4111] lr: 9.7116e-05 eta: 7:23:44 time: 0.7150 data_time: 0.5180 memory: 6318 loss: 0.3119 +2023/06/05 16:21:11 - mmengine - INFO - Epoch(train) [2][ 700/4111] lr: 9.6994e-05 eta: 7:22:19 time: 0.6754 data_time: 0.4935 memory: 6318 loss: 0.3098 +2023/06/05 16:22:22 - mmengine - INFO - Epoch(train) [2][ 800/4111] lr: 9.6869e-05 eta: 7:20:49 time: 0.6925 data_time: 0.3180 memory: 6318 loss: 0.3209 +2023/06/05 16:23:24 - mmengine - INFO - Exp name: resnet50_2xb256_all_1m_lr1e-4_aug_5e-1_20230605_152040 +2023/06/05 16:23:31 - mmengine - INFO - Epoch(train) [2][ 900/4111] lr: 9.6742e-05 eta: 7:19:11 time: 0.6876 data_time: 0.4542 memory: 6318 loss: 0.2988 +2023/06/05 16:24:43 - mmengine - INFO - Epoch(train) [2][1000/4111] lr: 9.6612e-05 eta: 7:17:45 time: 0.6937 data_time: 0.5439 memory: 6318 loss: 0.3440 +2023/06/05 16:25:54 - mmengine - INFO - Epoch(train) [2][1100/4111] lr: 9.6480e-05 eta: 7:16:23 time: 0.7457 data_time: 0.4246 memory: 6318 loss: 0.3073 +2023/06/05 16:27:06 - mmengine - INFO - Epoch(train) [2][1200/4111] lr: 9.6346e-05 eta: 7:15:00 time: 0.7637 data_time: 0.1950 memory: 6318 loss: 0.3189 +2023/06/05 16:28:16 - mmengine - INFO - Epoch(train) [2][1300/4111] lr: 9.6209e-05 eta: 7:13:30 time: 0.6712 data_time: 0.0453 memory: 6318 loss: 0.2926 +2023/06/05 16:29:27 - mmengine - INFO - Epoch(train) [2][1400/4111] lr: 9.6069e-05 eta: 7:12:03 time: 0.7660 data_time: 0.2667 memory: 6318 loss: 0.3067 +2023/06/05 16:30:40 - mmengine - INFO - Epoch(train) [2][1500/4111] lr: 9.5928e-05 eta: 7:10:55 time: 1.1053 data_time: 0.7144 memory: 6318 loss: 0.2962 +2023/06/05 16:31:51 - mmengine - INFO - Epoch(train) [2][1600/4111] lr: 9.5783e-05 eta: 7:09:28 time: 0.6608 data_time: 0.4194 memory: 6318 loss: 0.2966 +2023/06/05 16:33:01 - mmengine - INFO - Epoch(train) [2][1700/4111] lr: 9.5637e-05 eta: 7:07:59 time: 0.7694 data_time: 0.6298 memory: 6318 loss: 0.2902 +2023/06/05 16:34:12 - mmengine - INFO - Epoch(train) [2][1800/4111] lr: 9.5488e-05 eta: 7:06:34 time: 0.7698 data_time: 0.6294 memory: 6318 loss: 0.2974 +2023/06/05 16:35:15 - mmengine - INFO - Exp name: resnet50_2xb256_all_1m_lr1e-4_aug_5e-1_20230605_152040 +2023/06/05 16:35:23 - mmengine - INFO - Epoch(train) [2][1900/4111] lr: 9.5337e-05 eta: 7:05:14 time: 0.7797 data_time: 0.6395 memory: 6318 loss: 0.3060 +2023/06/05 16:36:33 - mmengine - INFO - Epoch(train) [2][2000/4111] lr: 9.5183e-05 eta: 7:03:47 time: 0.6529 data_time: 0.5135 memory: 6318 loss: 0.2818 +2023/06/05 16:37:45 - mmengine - INFO - Epoch(train) [2][2100/4111] lr: 9.5027e-05 eta: 7:02:31 time: 0.6340 data_time: 0.4937 memory: 6318 loss: 0.3080 +2023/06/05 16:38:55 - mmengine - INFO - Epoch(train) [2][2200/4111] lr: 9.4869e-05 eta: 7:01:02 time: 0.6890 data_time: 0.5485 memory: 6318 loss: 0.3084 +2023/06/05 16:40:07 - mmengine - INFO - Epoch(train) [2][2300/4111] lr: 9.4708e-05 eta: 6:59:45 time: 0.7857 data_time: 0.6454 memory: 6318 loss: 0.2712 +2023/06/05 16:41:22 - mmengine - INFO - Epoch(train) [2][2400/4111] lr: 9.4545e-05 eta: 6:58:43 time: 0.8974 data_time: 0.7566 memory: 6318 loss: 0.2886 +2023/06/05 16:42:34 - mmengine - INFO - Epoch(train) [2][2500/4111] lr: 9.4380e-05 eta: 6:57:29 time: 0.7038 data_time: 0.5645 memory: 6318 loss: 0.2826 +2023/06/05 16:43:51 - mmengine - INFO - Epoch(train) [2][2600/4111] lr: 9.4212e-05 eta: 6:56:37 time: 0.6865 data_time: 0.5453 memory: 6318 loss: 0.2823 +2023/06/05 16:45:02 - mmengine - INFO - Epoch(train) [2][2700/4111] lr: 9.4043e-05 eta: 6:55:17 time: 0.7029 data_time: 0.5627 memory: 6318 loss: 0.2908 +2023/06/05 16:46:13 - mmengine - INFO - Epoch(train) [2][2800/4111] lr: 9.3870e-05 eta: 6:53:59 time: 0.7291 data_time: 0.5891 memory: 6318 loss: 0.2921 +2023/06/05 16:47:18 - mmengine - INFO - Exp name: resnet50_2xb256_all_1m_lr1e-4_aug_5e-1_20230605_152040 +2023/06/05 16:47:25 - mmengine - INFO - Epoch(train) [2][2900/4111] lr: 9.3696e-05 eta: 6:52:40 time: 0.6954 data_time: 0.5557 memory: 6318 loss: 0.2790 +2023/06/05 16:48:39 - mmengine - INFO - Epoch(train) [2][3000/4111] lr: 9.3519e-05 eta: 6:51:34 time: 0.7108 data_time: 0.5707 memory: 6318 loss: 0.2752 +2023/06/05 16:49:53 - mmengine - INFO - Epoch(train) [2][3100/4111] lr: 9.3340e-05 eta: 6:50:32 time: 0.7575 data_time: 0.6172 memory: 6318 loss: 0.2841 +2023/06/05 16:51:08 - mmengine - INFO - Epoch(train) [2][3200/4111] lr: 9.3159e-05 eta: 6:49:27 time: 0.7836 data_time: 0.6439 memory: 6318 loss: 0.2750 +2023/06/05 16:52:29 - mmengine - INFO - Epoch(train) [2][3300/4111] lr: 9.2976e-05 eta: 6:48:53 time: 0.7227 data_time: 0.5824 memory: 6318 loss: 0.2725 +2023/06/05 16:53:44 - mmengine - INFO - Epoch(train) [2][3400/4111] lr: 9.2790e-05 eta: 6:47:50 time: 0.7616 data_time: 0.6201 memory: 6318 loss: 0.2888 +2023/06/05 16:54:59 - mmengine - INFO - Epoch(train) [2][3500/4111] lr: 9.2603e-05 eta: 6:46:48 time: 0.7335 data_time: 0.5937 memory: 6318 loss: 0.2855 +2023/06/05 16:56:17 - mmengine - INFO - Epoch(train) [2][3600/4111] lr: 9.2413e-05 eta: 6:45:55 time: 0.7825 data_time: 0.6419 memory: 6318 loss: 0.2639 +2023/06/05 16:57:40 - mmengine - INFO - Epoch(train) [2][3700/4111] lr: 9.2220e-05 eta: 6:45:24 time: 0.7304 data_time: 0.5895 memory: 6318 loss: 0.2743 +2023/06/05 16:59:09 - mmengine - INFO - Epoch(train) [2][3800/4111] lr: 9.2026e-05 eta: 6:45:20 time: 0.7432 data_time: 0.6032 memory: 6318 loss: 0.2661 +2023/06/05 17:00:20 - mmengine - INFO - Exp name: resnet50_2xb256_all_1m_lr1e-4_aug_5e-1_20230605_152040 +2023/06/05 17:00:28 - mmengine - INFO - Epoch(train) [2][3900/4111] lr: 9.1829e-05 eta: 6:44:31 time: 0.7515 data_time: 0.6075 memory: 6318 loss: 0.3004 +2023/06/05 17:01:53 - mmengine - INFO - Epoch(train) [2][4000/4111] lr: 9.1631e-05 eta: 6:44:04 time: 0.8359 data_time: 0.6953 memory: 6318 loss: 0.2826 +2023/06/05 17:03:05 - mmengine - INFO - Epoch(train) [2][4100/4111] lr: 9.1430e-05 eta: 6:42:47 time: 0.7300 data_time: 0.5900 memory: 6318 loss: 0.2501 +2023/06/05 17:03:13 - mmengine - INFO - Exp name: resnet50_2xb256_all_1m_lr1e-4_aug_5e-1_20230605_152040 +2023/06/05 17:03:13 - mmengine - INFO - Saving checkpoint at 2 epochs +2023/06/05 17:04:00 - mmengine - INFO - Epoch(val) [2][100/119] eta: 0:00:07 time: 0.5845 data_time: 0.4962 memory: 6318 +2023/06/05 17:04:27 - mmengine - INFO - Epoch(val) [2][119/119] accuracy/top1: 77.8627 data_time: 0.3795 time: 0.4681 +2023/06/05 17:05:45 - mmengine - INFO - Epoch(train) [3][ 100/4111] lr: 9.1205e-05 eta: 6:41:43 time: 0.7583 data_time: 0.6181 memory: 6318 loss: 0.2703 +2023/06/05 17:07:04 - mmengine - INFO - Epoch(train) [3][ 200/4111] lr: 9.0999e-05 eta: 6:40:50 time: 0.7259 data_time: 0.5870 memory: 6318 loss: 0.2637 +2023/06/05 17:08:20 - mmengine - INFO - Epoch(train) [3][ 300/4111] lr: 9.0792e-05 eta: 6:39:45 time: 0.7332 data_time: 0.5927 memory: 6318 loss: 0.3169 +2023/06/05 17:09:38 - mmengine - INFO - Epoch(train) [3][ 400/4111] lr: 9.0582e-05 eta: 6:38:49 time: 0.7841 data_time: 0.6441 memory: 6318 loss: 0.2606 +2023/06/05 17:11:02 - mmengine - INFO - Epoch(train) [3][ 500/4111] lr: 9.0371e-05 eta: 6:38:13 time: 0.7364 data_time: 0.5971 memory: 6318 loss: 0.2841 +2023/06/05 17:12:19 - mmengine - INFO - Epoch(train) [3][ 600/4111] lr: 9.0157e-05 eta: 6:37:12 time: 0.7307 data_time: 0.5914 memory: 6318 loss: 0.2805 +2023/06/05 17:13:36 - mmengine - INFO - Epoch(train) [3][ 700/4111] lr: 8.9942e-05 eta: 6:36:09 time: 0.8299 data_time: 0.6805 memory: 6318 loss: 0.2685 +2023/06/05 17:14:37 - mmengine - INFO - Exp name: resnet50_2xb256_all_1m_lr1e-4_aug_5e-1_20230605_152040 +2023/06/05 17:14:53 - mmengine - INFO - Epoch(train) [3][ 800/4111] lr: 8.9724e-05 eta: 6:35:06 time: 0.7293 data_time: 0.5869 memory: 6318 loss: 0.2602 +2023/06/05 17:16:11 - mmengine - INFO - Epoch(train) [3][ 900/4111] lr: 8.9504e-05 eta: 6:34:08 time: 0.8368 data_time: 0.6840 memory: 6318 loss: 0.2695 +2023/06/05 17:17:32 - mmengine - INFO - Epoch(train) [3][1000/4111] lr: 8.9282e-05 eta: 6:33:18 time: 0.6962 data_time: 0.5555 memory: 6318 loss: 0.2827 +2023/06/05 17:18:48 - mmengine - INFO - Epoch(train) [3][1100/4111] lr: 8.9059e-05 eta: 6:32:10 time: 0.7879 data_time: 0.6395 memory: 6318 loss: 0.2645 +2023/06/05 17:20:06 - mmengine - INFO - Epoch(train) [3][1200/4111] lr: 8.8833e-05 eta: 6:31:10 time: 0.7347 data_time: 0.5917 memory: 6318 loss: 0.2758 +2023/06/05 17:21:22 - mmengine - INFO - Epoch(train) [3][1300/4111] lr: 8.8605e-05 eta: 6:30:03 time: 0.7989 data_time: 0.6540 memory: 6318 loss: 0.2852 +2023/06/05 17:22:38 - mmengine - INFO - Epoch(train) [3][1400/4111] lr: 8.8375e-05 eta: 6:28:55 time: 0.7501 data_time: 0.6082 memory: 6318 loss: 0.2650 +2023/06/05 17:23:52 - mmengine - INFO - Epoch(train) [3][1500/4111] lr: 8.8144e-05 eta: 6:27:42 time: 0.7738 data_time: 0.6312 memory: 6318 loss: 0.2755 +2023/06/05 17:25:06 - mmengine - INFO - Epoch(train) [3][1600/4111] lr: 8.7910e-05 eta: 6:26:27 time: 0.7739 data_time: 0.6327 memory: 6318 loss: 0.2673 +2023/06/05 17:26:21 - mmengine - INFO - Epoch(train) [3][1700/4111] lr: 8.7675e-05 eta: 6:25:15 time: 0.7965 data_time: 0.6557 memory: 6318 loss: 0.2741 +2023/06/05 17:27:21 - mmengine - INFO - Exp name: resnet50_2xb256_all_1m_lr1e-4_aug_5e-1_20230605_152040 +2023/06/05 17:27:37 - mmengine - INFO - Epoch(train) [3][1800/4111] lr: 8.7437e-05 eta: 6:24:06 time: 0.8206 data_time: 0.6807 memory: 6318 loss: 0.2754 +2023/06/05 17:28:57 - mmengine - INFO - Epoch(train) [3][1900/4111] lr: 8.7198e-05 eta: 6:23:09 time: 0.7058 data_time: 0.5637 memory: 6318 loss: 0.2692 +2023/06/05 17:30:13 - mmengine - INFO - Epoch(train) [3][2000/4111] lr: 8.6957e-05 eta: 6:22:02 time: 0.7285 data_time: 0.5882 memory: 6318 loss: 0.2732 +2023/06/05 17:31:28 - mmengine - INFO - Epoch(train) [3][2100/4111] lr: 8.6714e-05 eta: 6:20:48 time: 0.7982 data_time: 0.6565 memory: 6318 loss: 0.2645 +2023/06/05 17:32:43 - mmengine - INFO - Epoch(train) [3][2200/4111] lr: 8.6469e-05 eta: 6:19:36 time: 0.7215 data_time: 0.5810 memory: 6318 loss: 0.2586 +2023/06/05 17:33:55 - mmengine - INFO - Epoch(train) [3][2300/4111] lr: 8.6222e-05 eta: 6:18:18 time: 0.6960 data_time: 0.5559 memory: 6318 loss: 0.2491 +2023/06/05 17:35:09 - mmengine - INFO - Epoch(train) [3][2400/4111] lr: 8.5974e-05 eta: 6:17:03 time: 0.7447 data_time: 0.6048 memory: 6318 loss: 0.2761 +2023/06/05 17:36:27 - mmengine - INFO - Epoch(train) [3][2500/4111] lr: 8.5723e-05 eta: 6:15:58 time: 0.7070 data_time: 0.5659 memory: 6318 loss: 0.2669 +2023/06/05 17:37:43 - mmengine - INFO - Epoch(train) [3][2600/4111] lr: 8.5471e-05 eta: 6:14:48 time: 0.7524 data_time: 0.6131 memory: 6318 loss: 0.2740 +2023/06/05 17:38:58 - mmengine - INFO - Epoch(train) [3][2700/4111] lr: 8.5217e-05 eta: 6:13:38 time: 0.7237 data_time: 0.5844 memory: 6318 loss: 0.2566 +2023/06/05 17:39:58 - mmengine - INFO - Exp name: resnet50_2xb256_all_1m_lr1e-4_aug_5e-1_20230605_152040 +2023/06/05 17:40:13 - mmengine - INFO - Epoch(train) [3][2800/4111] lr: 8.4962e-05 eta: 6:12:24 time: 0.7662 data_time: 0.6260 memory: 6318 loss: 0.2607 +2023/06/05 17:41:27 - mmengine - INFO - Epoch(train) [3][2900/4111] lr: 8.4704e-05 eta: 6:11:10 time: 0.7322 data_time: 0.4468 memory: 6318 loss: 0.2472 +2023/06/05 17:42:46 - mmengine - INFO - Epoch(train) [3][3000/4111] lr: 8.4445e-05 eta: 6:10:07 time: 0.7394 data_time: 0.5983 memory: 6318 loss: 0.2813 +2023/06/05 17:43:58 - mmengine - INFO - Epoch(train) [3][3100/4111] lr: 8.4184e-05 eta: 6:08:48 time: 0.7482 data_time: 0.6078 memory: 6318 loss: 0.2544 +2023/06/05 17:45:23 - mmengine - INFO - Epoch(train) [3][3200/4111] lr: 8.3921e-05 eta: 6:08:02 time: 0.7466 data_time: 0.6049 memory: 6318 loss: 0.2682 +2023/06/05 17:46:36 - mmengine - INFO - Epoch(train) [3][3300/4111] lr: 8.3657e-05 eta: 6:06:42 time: 0.7069 data_time: 0.5665 memory: 6318 loss: 0.2502 +2023/06/05 17:47:52 - mmengine - INFO - Epoch(train) [3][3400/4111] lr: 8.3391e-05 eta: 6:05:33 time: 0.8448 data_time: 0.6967 memory: 6318 loss: 0.2562 +2023/06/05 17:49:08 - mmengine - INFO - Epoch(train) [3][3500/4111] lr: 8.3123e-05 eta: 6:04:23 time: 0.7581 data_time: 0.6184 memory: 6318 loss: 0.2403 +2023/06/05 17:50:23 - mmengine - INFO - Epoch(train) [3][3600/4111] lr: 8.2854e-05 eta: 6:03:11 time: 0.8234 data_time: 0.6712 memory: 6318 loss: 0.2747 +2023/06/05 17:51:50 - mmengine - INFO - Epoch(train) [3][3700/4111] lr: 8.2583e-05 eta: 6:02:25 time: 0.7648 data_time: 0.6249 memory: 6318 loss: 0.2551 +2023/06/05 17:53:00 - mmengine - INFO - Exp name: resnet50_2xb256_all_1m_lr1e-4_aug_5e-1_20230605_152040 +2023/06/05 17:53:16 - mmengine - INFO - Epoch(train) [3][3800/4111] lr: 8.2311e-05 eta: 6:01:41 time: 0.7675 data_time: 0.6217 memory: 6318 loss: 0.2433 +2023/06/05 17:54:35 - mmengine - INFO - Epoch(train) [3][3900/4111] lr: 8.2037e-05 eta: 6:00:35 time: 0.7232 data_time: 0.5821 memory: 6318 loss: 0.2459 +2023/06/05 17:55:51 - mmengine - INFO - Epoch(train) [3][4000/4111] lr: 8.1761e-05 eta: 5:59:25 time: 0.6928 data_time: 0.5440 memory: 6318 loss: 0.2555 +2023/06/05 17:57:11 - mmengine - INFO - Epoch(train) [3][4100/4111] lr: 8.1484e-05 eta: 5:58:21 time: 0.8229 data_time: 0.6836 memory: 6318 loss: 0.2529 +2023/06/05 17:57:18 - mmengine - INFO - Exp name: resnet50_2xb256_all_1m_lr1e-4_aug_5e-1_20230605_152040 +2023/06/05 17:57:18 - mmengine - INFO - Saving checkpoint at 3 epochs +2023/06/05 17:58:07 - mmengine - INFO - Epoch(val) [3][100/119] eta: 0:00:08 time: 0.6279 data_time: 0.5392 memory: 6318 +2023/06/05 17:58:34 - mmengine - INFO - Epoch(val) [3][119/119] accuracy/top1: 79.1470 data_time: 0.3974 time: 0.4855 +2023/06/05 17:59:54 - mmengine - INFO - Epoch(train) [4][ 100/4111] lr: 8.1174e-05 eta: 5:57:11 time: 0.9920 data_time: 0.8513 memory: 6318 loss: 0.2658 +2023/06/05 18:01:17 - mmengine - INFO - Epoch(train) [4][ 200/4111] lr: 8.0894e-05 eta: 5:56:15 time: 0.8346 data_time: 0.6952 memory: 6318 loss: 0.2596 +2023/06/05 18:02:40 - mmengine - INFO - Epoch(train) [4][ 300/4111] lr: 8.0612e-05 eta: 5:55:17 time: 0.7843 data_time: 0.6424 memory: 6318 loss: 0.2486 +2023/06/05 18:03:57 - mmengine - INFO - Epoch(train) [4][ 400/4111] lr: 8.0328e-05 eta: 5:54:08 time: 0.7157 data_time: 0.5741 memory: 6318 loss: 0.2630 +2023/06/05 18:05:18 - mmengine - INFO - Epoch(train) [4][ 500/4111] lr: 8.0043e-05 eta: 5:53:06 time: 0.8058 data_time: 0.6651 memory: 6318 loss: 0.2640 +2023/06/05 18:06:36 - mmengine - INFO - Epoch(train) [4][ 600/4111] lr: 7.9757e-05 eta: 5:51:58 time: 0.8341 data_time: 0.6941 memory: 6318 loss: 0.2509 +2023/06/05 18:07:29 - mmengine - INFO - Exp name: resnet50_2xb256_all_1m_lr1e-4_aug_5e-1_20230605_152040 +2023/06/05 18:07:54 - mmengine - INFO - Epoch(train) [4][ 700/4111] lr: 7.9469e-05 eta: 5:50:49 time: 0.7698 data_time: 0.6279 memory: 6318 loss: 0.2448 +2023/06/05 18:09:11 - mmengine - INFO - Epoch(train) [4][ 800/4111] lr: 7.9180e-05 eta: 5:49:38 time: 0.7797 data_time: 0.6392 memory: 6318 loss: 0.2517 +2023/06/05 18:10:38 - mmengine - INFO - Epoch(train) [4][ 900/4111] lr: 7.8889e-05 eta: 5:48:49 time: 0.8146 data_time: 0.6729 memory: 6318 loss: 0.2784 +2023/06/05 18:12:04 - mmengine - INFO - Epoch(train) [4][1000/4111] lr: 7.8597e-05 eta: 5:47:57 time: 0.7530 data_time: 0.6117 memory: 6318 loss: 0.2623 +2023/06/05 18:13:19 - mmengine - INFO - Epoch(train) [4][1100/4111] lr: 7.8303e-05 eta: 5:46:42 time: 0.7699 data_time: 0.6274 memory: 6318 loss: 0.2496 +2023/06/05 18:14:44 - mmengine - INFO - Epoch(train) [4][1200/4111] lr: 7.8008e-05 eta: 5:45:47 time: 0.7586 data_time: 0.6157 memory: 6318 loss: 0.2473 +2023/06/05 18:16:02 - mmengine - INFO - Epoch(train) [4][1300/4111] lr: 7.7712e-05 eta: 5:44:37 time: 0.7981 data_time: 0.6353 memory: 6318 loss: 0.2408 +2023/06/05 18:17:18 - mmengine - INFO - Epoch(train) [4][1400/4111] lr: 7.7415e-05 eta: 5:43:23 time: 0.7776 data_time: 0.6363 memory: 6318 loss: 0.2399 +2023/06/05 18:18:32 - mmengine - INFO - Epoch(train) [4][1500/4111] lr: 7.7116e-05 eta: 5:42:05 time: 0.7483 data_time: 0.6063 memory: 6318 loss: 0.2610 +2023/06/05 18:19:47 - mmengine - INFO - Epoch(train) [4][1600/4111] lr: 7.6816e-05 eta: 5:40:50 time: 0.7294 data_time: 0.5899 memory: 6318 loss: 0.2485 +2023/06/05 18:20:41 - mmengine - INFO - Exp name: resnet50_2xb256_all_1m_lr1e-4_aug_5e-1_20230605_152040 +2023/06/05 18:21:04 - mmengine - INFO - Epoch(train) [4][1700/4111] lr: 7.6514e-05 eta: 5:39:37 time: 0.7353 data_time: 0.5952 memory: 6318 loss: 0.2620 +2023/06/05 18:22:21 - mmengine - INFO - Epoch(train) [4][1800/4111] lr: 7.6212e-05 eta: 5:38:25 time: 0.7197 data_time: 0.5791 memory: 6318 loss: 0.2430 +2023/06/05 18:23:38 - mmengine - INFO - Epoch(train) [4][1900/4111] lr: 7.5908e-05 eta: 5:37:13 time: 0.7071 data_time: 0.5664 memory: 6318 loss: 0.2509 +2023/06/05 18:24:52 - mmengine - INFO - Epoch(train) [4][2000/4111] lr: 7.5603e-05 eta: 5:35:56 time: 0.7692 data_time: 0.6195 memory: 6318 loss: 0.2576 +2023/06/05 18:26:08 - mmengine - INFO - Epoch(train) [4][2100/4111] lr: 7.5296e-05 eta: 5:34:42 time: 0.7146 data_time: 0.5746 memory: 6318 loss: 0.2374 +2023/06/05 18:27:23 - mmengine - INFO - Epoch(train) [4][2200/4111] lr: 7.4989e-05 eta: 5:33:27 time: 0.7566 data_time: 0.6154 memory: 6318 loss: 0.2282 +2023/06/05 18:28:36 - mmengine - INFO - Epoch(train) [4][2300/4111] lr: 7.4680e-05 eta: 5:32:08 time: 0.7285 data_time: 0.5762 memory: 6318 loss: 0.2506 +2023/06/05 18:29:50 - mmengine - INFO - Epoch(train) [4][2400/4111] lr: 7.4370e-05 eta: 5:30:49 time: 0.7813 data_time: 0.6412 memory: 6318 loss: 0.2487 +2023/06/05 18:31:05 - mmengine - INFO - Epoch(train) [4][2500/4111] lr: 7.4059e-05 eta: 5:29:33 time: 0.7347 data_time: 0.5939 memory: 6318 loss: 0.2330 +2023/06/05 18:32:17 - mmengine - INFO - Epoch(train) [4][2600/4111] lr: 7.3747e-05 eta: 5:28:14 time: 0.7347 data_time: 0.5858 memory: 6318 loss: 0.2689 +2023/06/05 18:33:11 - mmengine - INFO - Exp name: resnet50_2xb256_all_1m_lr1e-4_aug_5e-1_20230605_152040 +2023/06/05 18:33:34 - mmengine - INFO - Epoch(train) [4][2700/4111] lr: 7.3434e-05 eta: 5:27:00 time: 0.7724 data_time: 0.6319 memory: 6318 loss: 0.2383 +2023/06/05 18:34:50 - mmengine - INFO - Epoch(train) [4][2800/4111] lr: 7.3120e-05 eta: 5:25:46 time: 0.7330 data_time: 0.5923 memory: 6318 loss: 0.2503 +2023/06/05 18:36:03 - mmengine - INFO - Epoch(train) [4][2900/4111] lr: 7.2804e-05 eta: 5:24:28 time: 0.7538 data_time: 0.6075 memory: 6318 loss: 0.2364 +2023/06/05 18:37:58 - mmengine - INFO - Epoch(train) [4][3000/4111] lr: 7.2488e-05 eta: 5:24:20 time: 0.7431 data_time: 0.6015 memory: 6318 loss: 0.2382 +2023/06/05 18:39:15 - mmengine - INFO - Epoch(train) [4][3100/4111] lr: 7.2171e-05 eta: 5:23:06 time: 0.7771 data_time: 0.6365 memory: 6318 loss: 0.2418 +2023/06/05 18:40:28 - mmengine - INFO - Epoch(train) [4][3200/4111] lr: 7.1852e-05 eta: 5:21:46 time: 0.6890 data_time: 0.5497 memory: 6318 loss: 0.2319 +2023/06/05 18:41:42 - mmengine - INFO - Epoch(train) [4][3300/4111] lr: 7.1533e-05 eta: 5:20:29 time: 0.7935 data_time: 0.6540 memory: 6318 loss: 0.2338 +2023/06/05 18:42:55 - mmengine - INFO - Epoch(train) [4][3400/4111] lr: 7.1213e-05 eta: 5:19:09 time: 0.7888 data_time: 0.6486 memory: 6318 loss: 0.2326 +2023/06/05 18:44:19 - mmengine - INFO - Epoch(train) [4][3500/4111] lr: 7.0891e-05 eta: 5:18:08 time: 1.0505 data_time: 0.9109 memory: 6318 loss: 0.2285 +2023/06/05 18:45:42 - mmengine - INFO - Epoch(train) [4][3600/4111] lr: 7.0569e-05 eta: 5:17:04 time: 0.7553 data_time: 0.6149 memory: 6318 loss: 0.2507 +2023/06/05 18:46:42 - mmengine - INFO - Exp name: resnet50_2xb256_all_1m_lr1e-4_aug_5e-1_20230605_152040 +2023/06/05 18:47:11 - mmengine - INFO - Epoch(train) [4][3700/4111] lr: 7.0246e-05 eta: 5:16:09 time: 1.0263 data_time: 0.8851 memory: 6318 loss: 0.2303 +2023/06/05 18:48:41 - mmengine - INFO - Epoch(train) [4][3800/4111] lr: 6.9922e-05 eta: 5:15:16 time: 0.8128 data_time: 0.6722 memory: 6318 loss: 0.2276 +2023/06/05 18:50:02 - mmengine - INFO - Epoch(train) [4][3900/4111] lr: 6.9597e-05 eta: 5:14:08 time: 0.8151 data_time: 0.6749 memory: 6318 loss: 0.2389 +2023/06/05 18:51:24 - mmengine - INFO - Epoch(train) [4][4000/4111] lr: 6.9272e-05 eta: 5:13:01 time: 0.7680 data_time: 0.6280 memory: 6318 loss: 0.2316 +2023/06/05 18:52:41 - mmengine - INFO - Epoch(train) [4][4100/4111] lr: 6.8945e-05 eta: 5:11:48 time: 0.8279 data_time: 0.6877 memory: 6318 loss: 0.2460 +2023/06/05 18:52:49 - mmengine - INFO - Exp name: resnet50_2xb256_all_1m_lr1e-4_aug_5e-1_20230605_152040 +2023/06/05 18:52:49 - mmengine - INFO - Saving checkpoint at 4 epochs +2023/06/05 18:53:37 - mmengine - INFO - Epoch(val) [4][100/119] eta: 0:00:07 time: 0.5324 data_time: 0.4424 memory: 6318 +2023/06/05 18:54:03 - mmengine - INFO - Epoch(val) [4][119/119] accuracy/top1: 84.8286 data_time: 0.3858 time: 0.4747 +2023/06/05 18:55:21 - mmengine - INFO - Epoch(train) [5][ 100/4111] lr: 6.8582e-05 eta: 5:10:26 time: 0.7083 data_time: 0.5681 memory: 6318 loss: 0.2401 +2023/06/05 18:56:41 - mmengine - INFO - Epoch(train) [5][ 200/4111] lr: 6.8253e-05 eta: 5:09:16 time: 0.8060 data_time: 0.6666 memory: 6318 loss: 0.2342 +2023/06/05 18:58:17 - mmengine - INFO - Epoch(train) [5][ 300/4111] lr: 6.7924e-05 eta: 5:08:30 time: 0.7743 data_time: 0.6335 memory: 6318 loss: 0.2321 +2023/06/05 18:59:35 - mmengine - INFO - Epoch(train) [5][ 400/4111] lr: 6.7595e-05 eta: 5:07:16 time: 0.7285 data_time: 0.5886 memory: 6318 loss: 0.2339 +2023/06/05 19:01:11 - mmengine - INFO - Epoch(train) [5][ 500/4111] lr: 6.7264e-05 eta: 5:06:28 time: 0.7585 data_time: 0.6156 memory: 6318 loss: 0.2372 +2023/06/05 19:01:57 - mmengine - INFO - Exp name: resnet50_2xb256_all_1m_lr1e-4_aug_5e-1_20230605_152040 +2023/06/05 19:02:29 - mmengine - INFO - Epoch(train) [5][ 600/4111] lr: 6.6933e-05 eta: 5:05:16 time: 0.8711 data_time: 0.7294 memory: 6318 loss: 0.2502 +2023/06/05 19:04:03 - mmengine - INFO - Epoch(train) [5][ 700/4111] lr: 6.6601e-05 eta: 5:04:25 time: 0.7589 data_time: 0.6184 memory: 6318 loss: 0.2335 +2023/06/05 19:05:23 - mmengine - INFO - Epoch(train) [5][ 800/4111] lr: 6.6268e-05 eta: 5:03:13 time: 0.7740 data_time: 0.6342 memory: 6318 loss: 0.2298 +2023/06/05 19:06:43 - mmengine - INFO - Epoch(train) [5][ 900/4111] lr: 6.5935e-05 eta: 5:02:03 time: 0.9358 data_time: 0.7951 memory: 6318 loss: 0.2430 +2023/06/05 19:08:08 - mmengine - INFO - Epoch(train) [5][1000/4111] lr: 6.5601e-05 eta: 5:00:59 time: 0.8132 data_time: 0.6721 memory: 6318 loss: 0.2417 +2023/06/05 19:09:26 - mmengine - INFO - Epoch(train) [5][1100/4111] lr: 6.5267e-05 eta: 4:59:45 time: 0.7892 data_time: 0.6430 memory: 6318 loss: 0.2323 +2023/06/05 19:10:45 - mmengine - INFO - Epoch(train) [5][1200/4111] lr: 6.4932e-05 eta: 4:58:31 time: 0.7581 data_time: 0.6184 memory: 6318 loss: 0.2227 +2023/06/05 19:12:01 - mmengine - INFO - Epoch(train) [5][1300/4111] lr: 6.4596e-05 eta: 4:57:14 time: 0.7561 data_time: 0.5958 memory: 6318 loss: 0.2321 +2023/06/05 19:13:18 - mmengine - INFO - Epoch(train) [5][1400/4111] lr: 6.4260e-05 eta: 4:55:59 time: 0.7444 data_time: 0.6040 memory: 6318 loss: 0.2288 +2023/06/05 19:14:48 - mmengine - INFO - Epoch(train) [5][1500/4111] lr: 6.3923e-05 eta: 4:55:00 time: 0.7444 data_time: 0.6038 memory: 6318 loss: 0.2103 +2023/06/05 19:15:34 - mmengine - INFO - Exp name: resnet50_2xb256_all_1m_lr1e-4_aug_5e-1_20230605_152040 +2023/06/05 19:16:06 - mmengine - INFO - Epoch(train) [5][1600/4111] lr: 6.3585e-05 eta: 4:53:46 time: 0.7831 data_time: 0.6428 memory: 6318 loss: 0.2322 +2023/06/05 19:17:24 - mmengine - INFO - Epoch(train) [5][1700/4111] lr: 6.3248e-05 eta: 4:52:32 time: 0.8304 data_time: 0.6907 memory: 6318 loss: 0.2278 +2023/06/05 19:18:42 - mmengine - INFO - Epoch(train) [5][1800/4111] lr: 6.2909e-05 eta: 4:51:17 time: 0.7770 data_time: 0.6373 memory: 6318 loss: 0.2379 +2023/06/05 19:20:00 - mmengine - INFO - Epoch(train) [5][1900/4111] lr: 6.2571e-05 eta: 4:50:03 time: 0.8290 data_time: 0.6890 memory: 6318 loss: 0.2434 +2023/06/05 19:21:15 - mmengine - INFO - Epoch(train) [5][2000/4111] lr: 6.2231e-05 eta: 4:48:44 time: 0.7519 data_time: 0.6115 memory: 6318 loss: 0.2270 +2023/06/05 19:22:34 - mmengine - INFO - Epoch(train) [5][2100/4111] lr: 6.1892e-05 eta: 4:47:31 time: 0.8247 data_time: 0.6836 memory: 6318 loss: 0.2424 +2023/06/05 19:23:52 - mmengine - INFO - Epoch(train) [5][2200/4111] lr: 6.1552e-05 eta: 4:46:17 time: 0.7765 data_time: 0.6357 memory: 6318 loss: 0.2345 +2023/06/05 19:25:11 - mmengine - INFO - Epoch(train) [5][2300/4111] lr: 6.1211e-05 eta: 4:45:03 time: 0.8519 data_time: 0.7121 memory: 6318 loss: 0.2327 +2023/06/05 19:26:28 - mmengine - INFO - Epoch(train) [5][2400/4111] lr: 6.0871e-05 eta: 4:43:47 time: 0.7622 data_time: 0.6103 memory: 6318 loss: 0.2430 +2023/06/05 19:27:45 - mmengine - INFO - Epoch(train) [5][2500/4111] lr: 6.0529e-05 eta: 4:42:31 time: 0.7704 data_time: 0.6303 memory: 6318 loss: 0.2203 +2023/06/05 19:28:30 - mmengine - INFO - Exp name: resnet50_2xb256_all_1m_lr1e-4_aug_5e-1_20230605_152040 +2023/06/05 19:29:02 - mmengine - INFO - Epoch(train) [5][2600/4111] lr: 6.0188e-05 eta: 4:41:15 time: 0.8023 data_time: 0.6625 memory: 6318 loss: 0.2179 +2023/06/05 19:30:20 - mmengine - INFO - Epoch(train) [5][2700/4111] lr: 5.9846e-05 eta: 4:40:01 time: 0.7628 data_time: 0.6217 memory: 6318 loss: 0.2233 +2023/06/05 19:31:39 - mmengine - INFO - Epoch(train) [5][2800/4111] lr: 5.9504e-05 eta: 4:38:47 time: 0.7977 data_time: 0.6580 memory: 6318 loss: 0.2306 +2023/06/05 19:33:30 - mmengine - INFO - Epoch(train) [5][2900/4111] lr: 5.9162e-05 eta: 4:38:09 time: 0.8246 data_time: 0.6836 memory: 6318 loss: 0.2321 +2023/06/05 19:34:47 - mmengine - INFO - Epoch(train) [5][3000/4111] lr: 5.8819e-05 eta: 4:36:53 time: 0.7552 data_time: 0.6140 memory: 6318 loss: 0.2337 +2023/06/05 19:36:04 - mmengine - INFO - Epoch(train) [5][3100/4111] lr: 5.8477e-05 eta: 4:35:36 time: 0.7281 data_time: 0.5882 memory: 6318 loss: 0.2307 +2023/06/05 19:37:22 - mmengine - INFO - Epoch(train) [5][3200/4111] lr: 5.8134e-05 eta: 4:34:21 time: 0.7822 data_time: 0.6397 memory: 6318 loss: 0.2354 +2023/06/05 19:38:39 - mmengine - INFO - Epoch(train) [5][3300/4111] lr: 5.7791e-05 eta: 4:33:05 time: 0.8323 data_time: 0.6922 memory: 6318 loss: 0.2326 +2023/06/05 19:40:02 - mmengine - INFO - Epoch(train) [5][3400/4111] lr: 5.7447e-05 eta: 4:31:55 time: 0.9836 data_time: 0.8376 memory: 6318 loss: 0.2390 +2023/06/05 19:41:26 - mmengine - INFO - Epoch(train) [5][3500/4111] lr: 5.7104e-05 eta: 4:30:46 time: 0.8006 data_time: 0.6602 memory: 6318 loss: 0.2380 +2023/06/05 19:42:20 - mmengine - INFO - Exp name: resnet50_2xb256_all_1m_lr1e-4_aug_5e-1_20230605_152040 +2023/06/05 19:43:02 - mmengine - INFO - Epoch(train) [5][3600/4111] lr: 5.6760e-05 eta: 4:29:49 time: 0.7596 data_time: 0.6204 memory: 6318 loss: 0.2407 +2023/06/05 19:44:27 - mmengine - INFO - Epoch(train) [5][3700/4111] lr: 5.6417e-05 eta: 4:28:41 time: 1.0260 data_time: 0.8849 memory: 6318 loss: 0.2382 +2023/06/05 19:45:48 - mmengine - INFO - Epoch(train) [5][3800/4111] lr: 5.6073e-05 eta: 4:27:29 time: 0.8182 data_time: 0.6775 memory: 6318 loss: 0.2201 +2023/06/05 19:47:06 - mmengine - INFO - Epoch(train) [5][3900/4111] lr: 5.5729e-05 eta: 4:26:13 time: 0.8179 data_time: 0.6762 memory: 6318 loss: 0.2373 +2023/06/05 19:48:27 - mmengine - INFO - Epoch(train) [5][4000/4111] lr: 5.5385e-05 eta: 4:25:00 time: 0.8244 data_time: 0.6833 memory: 6318 loss: 0.2325 +2023/06/05 19:49:53 - mmengine - INFO - Epoch(train) [5][4100/4111] lr: 5.5041e-05 eta: 4:23:52 time: 0.8194 data_time: 0.6765 memory: 6318 loss: 0.2335 +2023/06/05 19:50:00 - mmengine - INFO - Exp name: resnet50_2xb256_all_1m_lr1e-4_aug_5e-1_20230605_152040 +2023/06/05 19:50:00 - mmengine - INFO - Saving checkpoint at 5 epochs +2023/06/05 19:50:50 - mmengine - INFO - Epoch(val) [5][100/119] eta: 0:00:08 time: 0.6158 data_time: 0.5276 memory: 6318 +2023/06/05 19:51:17 - mmengine - INFO - Epoch(val) [5][119/119] accuracy/top1: 88.0062 data_time: 0.4096 time: 0.4968 +2023/06/05 19:52:46 - mmengine - INFO - Epoch(train) [6][ 100/4111] lr: 5.4660e-05 eta: 4:22:37 time: 0.8418 data_time: 0.5640 memory: 6318 loss: 0.2352 +2023/06/05 19:54:22 - mmengine - INFO - Epoch(train) [6][ 200/4111] lr: 5.4316e-05 eta: 4:21:39 time: 0.9359 data_time: 0.7866 memory: 6318 loss: 0.2412 +2023/06/05 19:56:05 - mmengine - INFO - Epoch(train) [6][ 300/4111] lr: 5.3972e-05 eta: 4:20:47 time: 0.8299 data_time: 0.6890 memory: 6318 loss: 0.2435 +2023/06/05 19:57:39 - mmengine - INFO - Epoch(train) [6][ 400/4111] lr: 5.3628e-05 eta: 4:19:45 time: 0.7582 data_time: 0.6160 memory: 6318 loss: 0.2235 +2023/06/05 19:58:13 - mmengine - INFO - Exp name: resnet50_2xb256_all_1m_lr1e-4_aug_5e-1_20230605_152040 +2023/06/05 19:59:01 - mmengine - INFO - Epoch(train) [6][ 500/4111] lr: 5.3284e-05 eta: 4:18:32 time: 0.8770 data_time: 0.7361 memory: 6318 loss: 0.2315 +2023/06/05 20:00:30 - mmengine - INFO - Epoch(train) [6][ 600/4111] lr: 5.2941e-05 eta: 4:17:26 time: 0.8681 data_time: 0.7282 memory: 6318 loss: 0.2305 +2023/06/05 20:01:59 - mmengine - INFO - Epoch(train) [6][ 700/4111] lr: 5.2597e-05 eta: 4:16:20 time: 0.7799 data_time: 0.6398 memory: 6318 loss: 0.2284 +2023/06/05 20:03:25 - mmengine - INFO - Epoch(train) [6][ 800/4111] lr: 5.2254e-05 eta: 4:15:10 time: 0.7880 data_time: 0.6479 memory: 6318 loss: 0.2305 +2023/06/05 20:04:47 - mmengine - INFO - Epoch(train) [6][ 900/4111] lr: 5.1911e-05 eta: 4:13:57 time: 0.7710 data_time: 0.6064 memory: 6318 loss: 0.2322 +2023/06/05 20:06:06 - mmengine - INFO - Epoch(train) [6][1000/4111] lr: 5.1568e-05 eta: 4:12:40 time: 0.8169 data_time: 0.6771 memory: 6318 loss: 0.2387 +2023/06/05 20:07:25 - mmengine - INFO - Epoch(train) [6][1100/4111] lr: 5.1225e-05 eta: 4:11:25 time: 0.7786 data_time: 0.6384 memory: 6318 loss: 0.2284 +2023/06/05 20:08:46 - mmengine - INFO - Epoch(train) [6][1200/4111] lr: 5.0883e-05 eta: 4:10:10 time: 0.7761 data_time: 0.6358 memory: 6318 loss: 0.2253 +2023/06/05 20:10:05 - mmengine - INFO - Epoch(train) [6][1300/4111] lr: 5.0540e-05 eta: 4:08:54 time: 0.7550 data_time: 0.6141 memory: 6318 loss: 0.2347 +2023/06/05 20:11:26 - mmengine - INFO - Epoch(train) [6][1400/4111] lr: 5.0198e-05 eta: 4:07:39 time: 0.8464 data_time: 0.7070 memory: 6318 loss: 0.2290 +2023/06/05 20:12:00 - mmengine - INFO - Exp name: resnet50_2xb256_all_1m_lr1e-4_aug_5e-1_20230605_152040 +2023/06/05 20:12:47 - mmengine - INFO - Epoch(train) [6][1500/4111] lr: 4.9856e-05 eta: 4:06:24 time: 0.8123 data_time: 0.6718 memory: 6318 loss: 0.2397 +2023/06/05 20:14:07 - mmengine - INFO - Epoch(train) [6][1600/4111] lr: 4.9515e-05 eta: 4:05:09 time: 0.7747 data_time: 0.6345 memory: 6318 loss: 0.2116 +2023/06/05 20:15:29 - mmengine - INFO - Epoch(train) [6][1700/4111] lr: 4.9174e-05 eta: 4:03:55 time: 0.7877 data_time: 0.6453 memory: 6318 loss: 0.2203 +2023/06/05 20:16:52 - mmengine - INFO - Epoch(train) [6][1800/4111] lr: 4.8833e-05 eta: 4:02:42 time: 0.9016 data_time: 0.7558 memory: 6318 loss: 0.2256 +2023/06/05 20:18:16 - mmengine - INFO - Epoch(train) [6][1900/4111] lr: 4.8492e-05 eta: 4:01:29 time: 0.8871 data_time: 0.7456 memory: 6318 loss: 0.2094 +2023/06/05 20:19:40 - mmengine - INFO - Epoch(train) [6][2000/4111] lr: 4.8152e-05 eta: 4:00:17 time: 0.8500 data_time: 0.7094 memory: 6318 loss: 0.2340 +2023/06/05 20:21:04 - mmengine - INFO - Epoch(train) [6][2100/4111] lr: 4.7813e-05 eta: 3:59:04 time: 0.8487 data_time: 0.7075 memory: 6318 loss: 0.2513 +2023/06/05 20:22:26 - mmengine - INFO - Epoch(train) [6][2200/4111] lr: 4.7473e-05 eta: 3:57:50 time: 0.8171 data_time: 0.6699 memory: 6318 loss: 0.2274 +2023/06/05 20:23:47 - mmengine - INFO - Epoch(train) [6][2300/4111] lr: 4.7135e-05 eta: 3:56:35 time: 0.8369 data_time: 0.6963 memory: 6318 loss: 0.2213 +2023/06/05 20:25:09 - mmengine - INFO - Epoch(train) [6][2400/4111] lr: 4.6796e-05 eta: 3:55:21 time: 0.8541 data_time: 0.7062 memory: 6318 loss: 0.2287 +2023/06/05 20:25:42 - mmengine - INFO - Exp name: resnet50_2xb256_all_1m_lr1e-4_aug_5e-1_20230605_152040 +2023/06/05 20:26:29 - mmengine - INFO - Epoch(train) [6][2500/4111] lr: 4.6458e-05 eta: 3:54:04 time: 0.7869 data_time: 0.6461 memory: 6318 loss: 0.2224 +2023/06/05 20:27:57 - mmengine - INFO - Epoch(train) [6][2600/4111] lr: 4.6121e-05 eta: 3:52:54 time: 0.8206 data_time: 0.6813 memory: 6318 loss: 0.2251 +2023/06/05 20:29:15 - mmengine - INFO - Epoch(train) [6][2700/4111] lr: 4.5784e-05 eta: 3:51:37 time: 0.8101 data_time: 0.6693 memory: 6318 loss: 0.2289 +2023/06/05 20:30:35 - mmengine - INFO - Epoch(train) [6][2800/4111] lr: 4.5448e-05 eta: 3:50:21 time: 0.8060 data_time: 0.6656 memory: 6318 loss: 0.2277 +2023/06/05 20:31:55 - mmengine - INFO - Epoch(train) [6][2900/4111] lr: 4.5112e-05 eta: 3:49:04 time: 0.8776 data_time: 0.7370 memory: 6318 loss: 0.2240 +2023/06/05 20:33:13 - mmengine - INFO - Epoch(train) [6][3000/4111] lr: 4.4777e-05 eta: 3:47:47 time: 0.7513 data_time: 0.6091 memory: 6318 loss: 0.2169 +2023/06/05 20:34:35 - mmengine - INFO - Epoch(train) [6][3100/4111] lr: 4.4442e-05 eta: 3:46:32 time: 0.7893 data_time: 0.6490 memory: 6318 loss: 0.2310 +2023/06/05 20:35:57 - mmengine - INFO - Epoch(train) [6][3200/4111] lr: 4.4108e-05 eta: 3:45:17 time: 0.7810 data_time: 0.6413 memory: 6318 loss: 0.2184 +2023/06/05 20:37:32 - mmengine - INFO - Epoch(train) [6][3300/4111] lr: 4.3775e-05 eta: 3:44:12 time: 0.8793 data_time: 0.7388 memory: 6318 loss: 0.2127 +2023/06/05 20:38:53 - mmengine - INFO - Epoch(train) [6][3400/4111] lr: 4.3442e-05 eta: 3:42:55 time: 0.7838 data_time: 0.6426 memory: 6318 loss: 0.2235 +2023/06/05 20:39:26 - mmengine - INFO - Exp name: resnet50_2xb256_all_1m_lr1e-4_aug_5e-1_20230605_152040 +2023/06/05 20:40:13 - mmengine - INFO - Epoch(train) [6][3500/4111] lr: 4.3110e-05 eta: 3:41:39 time: 0.7675 data_time: 0.6268 memory: 6318 loss: 0.2355 +2023/06/05 20:41:32 - mmengine - INFO - Epoch(train) [6][3600/4111] lr: 4.2779e-05 eta: 3:40:22 time: 0.7755 data_time: 0.6346 memory: 6318 loss: 0.2247 +2023/06/05 20:42:52 - mmengine - INFO - Epoch(train) [6][3700/4111] lr: 4.2448e-05 eta: 3:39:05 time: 0.8477 data_time: 0.7075 memory: 6318 loss: 0.2081 +2023/06/05 20:44:15 - mmengine - INFO - Epoch(train) [6][3800/4111] lr: 4.2119e-05 eta: 3:37:51 time: 0.7562 data_time: 0.6137 memory: 6318 loss: 0.2007 +2023/06/05 20:45:36 - mmengine - INFO - Epoch(train) [6][3900/4111] lr: 4.1789e-05 eta: 3:36:35 time: 0.7941 data_time: 0.6522 memory: 6318 loss: 0.2139 +2023/06/05 20:46:57 - mmengine - INFO - Epoch(train) [6][4000/4111] lr: 4.1461e-05 eta: 3:35:19 time: 0.8031 data_time: 0.6624 memory: 6318 loss: 0.2114 +2023/06/05 20:48:19 - mmengine - INFO - Epoch(train) [6][4100/4111] lr: 4.1133e-05 eta: 3:34:03 time: 0.8068 data_time: 0.6593 memory: 6318 loss: 0.1972 +2023/06/05 20:48:26 - mmengine - INFO - Exp name: resnet50_2xb256_all_1m_lr1e-4_aug_5e-1_20230605_152040 +2023/06/05 20:48:26 - mmengine - INFO - Saving checkpoint at 6 epochs +2023/06/05 20:49:16 - mmengine - INFO - Epoch(val) [6][100/119] eta: 0:00:08 time: 0.6186 data_time: 0.5294 memory: 6318 +2023/06/05 20:49:43 - mmengine - INFO - Epoch(val) [6][119/119] accuracy/top1: 90.6178 data_time: 0.4038 time: 0.4922 +2023/06/05 20:51:06 - mmengine - INFO - Epoch(train) [7][ 100/4111] lr: 4.0771e-05 eta: 3:32:39 time: 0.9273 data_time: 0.7648 memory: 6318 loss: 0.2223 +2023/06/05 20:52:44 - mmengine - INFO - Epoch(train) [7][ 200/4111] lr: 4.0445e-05 eta: 3:31:34 time: 0.7607 data_time: 0.6206 memory: 6318 loss: 0.2190 +2023/06/05 20:54:06 - mmengine - INFO - Epoch(train) [7][ 300/4111] lr: 4.0120e-05 eta: 3:30:18 time: 0.7941 data_time: 0.6529 memory: 6318 loss: 0.2335 +2023/06/05 20:54:35 - mmengine - INFO - Exp name: resnet50_2xb256_all_1m_lr1e-4_aug_5e-1_20230605_152040 +2023/06/05 20:55:28 - mmengine - INFO - Epoch(train) [7][ 400/4111] lr: 3.9796e-05 eta: 3:29:03 time: 0.7644 data_time: 0.6245 memory: 6318 loss: 0.2443 +2023/06/05 20:56:53 - mmengine - INFO - Epoch(train) [7][ 500/4111] lr: 3.9473e-05 eta: 3:27:49 time: 0.8562 data_time: 0.7142 memory: 6318 loss: 0.2211 +2023/06/05 20:58:12 - mmengine - INFO - Epoch(train) [7][ 600/4111] lr: 3.9150e-05 eta: 3:26:31 time: 0.8287 data_time: 0.6863 memory: 6318 loss: 0.2090 +2023/06/05 20:59:41 - mmengine - INFO - Epoch(train) [7][ 700/4111] lr: 3.8829e-05 eta: 3:25:19 time: 0.8396 data_time: 0.6973 memory: 6318 loss: 0.2327 +2023/06/05 21:01:06 - mmengine - INFO - Epoch(train) [7][ 800/4111] lr: 3.8509e-05 eta: 3:24:06 time: 0.7590 data_time: 0.6171 memory: 6318 loss: 0.2174 +2023/06/05 21:02:29 - mmengine - INFO - Epoch(train) [7][ 900/4111] lr: 3.8189e-05 eta: 3:22:50 time: 0.8073 data_time: 0.6641 memory: 6318 loss: 0.2343 +2023/06/05 21:03:54 - mmengine - INFO - Epoch(train) [7][1000/4111] lr: 3.7871e-05 eta: 3:21:36 time: 0.7798 data_time: 0.6383 memory: 6318 loss: 0.2221 +2023/06/05 21:05:14 - mmengine - INFO - Epoch(train) [7][1100/4111] lr: 3.7553e-05 eta: 3:20:19 time: 0.8139 data_time: 0.6733 memory: 6318 loss: 0.2249 +2023/06/05 21:06:36 - mmengine - INFO - Epoch(train) [7][1200/4111] lr: 3.7237e-05 eta: 3:19:02 time: 0.7923 data_time: 0.6524 memory: 6318 loss: 0.2299 +2023/06/05 21:07:52 - mmengine - INFO - Epoch(train) [7][1300/4111] lr: 3.6921e-05 eta: 3:17:43 time: 0.7348 data_time: 0.5943 memory: 6318 loss: 0.2116 +2023/06/05 21:08:23 - mmengine - INFO - Exp name: resnet50_2xb256_all_1m_lr1e-4_aug_5e-1_20230605_152040 +2023/06/05 21:09:16 - mmengine - INFO - Epoch(train) [7][1400/4111] lr: 3.6607e-05 eta: 3:16:27 time: 0.7787 data_time: 0.6365 memory: 6318 loss: 0.2309 +2023/06/05 21:10:33 - mmengine - INFO - Epoch(train) [7][1500/4111] lr: 3.6294e-05 eta: 3:15:09 time: 0.7614 data_time: 0.6209 memory: 6318 loss: 0.2391 +2023/06/05 21:11:52 - mmengine - INFO - Epoch(train) [7][1600/4111] lr: 3.5981e-05 eta: 3:13:51 time: 0.7763 data_time: 0.6363 memory: 6318 loss: 0.2250 +2023/06/05 21:13:11 - mmengine - INFO - Epoch(train) [7][1700/4111] lr: 3.5670e-05 eta: 3:12:32 time: 0.7863 data_time: 0.6458 memory: 6318 loss: 0.2065 +2023/06/05 21:14:29 - mmengine - INFO - Epoch(train) [7][1800/4111] lr: 3.5360e-05 eta: 3:11:14 time: 0.7749 data_time: 0.6337 memory: 6318 loss: 0.2124 +2023/06/05 21:15:48 - mmengine - INFO - Epoch(train) [7][1900/4111] lr: 3.5051e-05 eta: 3:09:56 time: 0.7664 data_time: 0.6258 memory: 6318 loss: 0.2132 +2023/06/05 21:17:08 - mmengine - INFO - Epoch(train) [7][2000/4111] lr: 3.4744e-05 eta: 3:08:38 time: 0.8567 data_time: 0.7147 memory: 6318 loss: 0.2223 +2023/06/05 21:18:26 - mmengine - INFO - Epoch(train) [7][2100/4111] lr: 3.4437e-05 eta: 3:07:20 time: 0.7756 data_time: 0.6349 memory: 6318 loss: 0.2282 +2023/06/05 21:19:44 - mmengine - INFO - Epoch(train) [7][2200/4111] lr: 3.4132e-05 eta: 3:06:02 time: 0.8062 data_time: 0.6656 memory: 6318 loss: 0.2173 +2023/06/05 21:21:39 - mmengine - INFO - Epoch(train) [7][2300/4111] lr: 3.3828e-05 eta: 3:05:02 time: 0.7225 data_time: 0.5825 memory: 6318 loss: 0.2266 +2023/06/05 21:22:10 - mmengine - INFO - Exp name: resnet50_2xb256_all_1m_lr1e-4_aug_5e-1_20230605_152040 +2023/06/05 21:22:55 - mmengine - INFO - Epoch(train) [7][2400/4111] lr: 3.3525e-05 eta: 3:03:43 time: 0.7958 data_time: 0.6492 memory: 6318 loss: 0.2297 +2023/06/05 21:24:10 - mmengine - INFO - Epoch(train) [7][2500/4111] lr: 3.3224e-05 eta: 3:02:22 time: 0.7735 data_time: 0.6220 memory: 6318 loss: 0.2032 +2023/06/05 21:25:25 - mmengine - INFO - Epoch(train) [7][2600/4111] lr: 3.2923e-05 eta: 3:01:02 time: 0.7633 data_time: 0.6233 memory: 6318 loss: 0.2184 +2023/06/05 21:26:42 - mmengine - INFO - Epoch(train) [7][2700/4111] lr: 3.2624e-05 eta: 2:59:42 time: 0.7284 data_time: 0.5884 memory: 6318 loss: 0.2244 +2023/06/05 21:27:56 - mmengine - INFO - Epoch(train) [7][2800/4111] lr: 3.2326e-05 eta: 2:58:22 time: 0.7287 data_time: 0.5816 memory: 6318 loss: 0.2136 +2023/06/05 21:29:10 - mmengine - INFO - Epoch(train) [7][2900/4111] lr: 3.2030e-05 eta: 2:57:01 time: 0.7710 data_time: 0.6303 memory: 6318 loss: 0.2312 +2023/06/05 21:30:26 - mmengine - INFO - Epoch(train) [7][3000/4111] lr: 3.1735e-05 eta: 2:55:42 time: 0.7887 data_time: 0.6477 memory: 6318 loss: 0.2057 +2023/06/05 21:31:41 - mmengine - INFO - Epoch(train) [7][3100/4111] lr: 3.1441e-05 eta: 2:54:21 time: 0.7016 data_time: 0.5622 memory: 6318 loss: 0.2073 +2023/06/05 21:32:54 - mmengine - INFO - Epoch(train) [7][3200/4111] lr: 3.1149e-05 eta: 2:53:01 time: 0.7420 data_time: 0.5977 memory: 6318 loss: 0.2091 +2023/06/05 21:34:10 - mmengine - INFO - Epoch(train) [7][3300/4111] lr: 3.0858e-05 eta: 2:51:41 time: 0.7123 data_time: 0.5713 memory: 6318 loss: 0.2025 +2023/06/05 21:34:38 - mmengine - INFO - Exp name: resnet50_2xb256_all_1m_lr1e-4_aug_5e-1_20230605_152040 +2023/06/05 21:35:24 - mmengine - INFO - Epoch(train) [7][3400/4111] lr: 3.0569e-05 eta: 2:50:21 time: 0.7505 data_time: 0.6111 memory: 6318 loss: 0.2176 +2023/06/05 21:36:39 - mmengine - INFO - Epoch(train) [7][3500/4111] lr: 3.0281e-05 eta: 2:49:01 time: 0.7819 data_time: 0.6413 memory: 6318 loss: 0.2117 +2023/06/05 21:37:54 - mmengine - INFO - Epoch(train) [7][3600/4111] lr: 2.9994e-05 eta: 2:47:41 time: 0.7291 data_time: 0.5889 memory: 6318 loss: 0.2219 +2023/06/05 21:39:10 - mmengine - INFO - Epoch(train) [7][3700/4111] lr: 2.9709e-05 eta: 2:46:21 time: 0.7776 data_time: 0.6378 memory: 6318 loss: 0.2253 +2023/06/05 21:40:25 - mmengine - INFO - Epoch(train) [7][3800/4111] lr: 2.9425e-05 eta: 2:45:02 time: 0.7334 data_time: 0.5933 memory: 6318 loss: 0.2275 +2023/06/05 21:41:47 - mmengine - INFO - Epoch(train) [7][3900/4111] lr: 2.9143e-05 eta: 2:43:45 time: 0.7487 data_time: 0.6079 memory: 6318 loss: 0.2068 +2023/06/05 21:43:03 - mmengine - INFO - Epoch(train) [7][4000/4111] lr: 2.8862e-05 eta: 2:42:26 time: 0.7887 data_time: 0.6474 memory: 6318 loss: 0.2244 +2023/06/05 21:44:18 - mmengine - INFO - Epoch(train) [7][4100/4111] lr: 2.8583e-05 eta: 2:41:06 time: 0.7472 data_time: 0.6077 memory: 6318 loss: 0.2238 +2023/06/05 21:44:25 - mmengine - INFO - Exp name: resnet50_2xb256_all_1m_lr1e-4_aug_5e-1_20230605_152040 +2023/06/05 21:44:25 - mmengine - INFO - Saving checkpoint at 7 epochs +2023/06/05 21:45:15 - mmengine - INFO - Epoch(val) [7][100/119] eta: 0:00:08 time: 0.6263 data_time: 0.5383 memory: 6318 +2023/06/05 21:45:43 - mmengine - INFO - Epoch(val) [7][119/119] accuracy/top1: 91.9087 data_time: 0.4005 time: 0.4896 +2023/06/05 21:47:14 - mmengine - INFO - Epoch(train) [8][ 100/4111] lr: 2.8275e-05 eta: 2:39:44 time: 0.7586 data_time: 0.6176 memory: 6318 loss: 0.2144 +2023/06/05 21:48:30 - mmengine - INFO - Epoch(train) [8][ 200/4111] lr: 2.7999e-05 eta: 2:38:25 time: 0.8150 data_time: 0.6749 memory: 6318 loss: 0.2248 +2023/06/05 21:48:51 - mmengine - INFO - Exp name: resnet50_2xb256_all_1m_lr1e-4_aug_5e-1_20230605_152040 +2023/06/05 21:49:45 - mmengine - INFO - Epoch(train) [8][ 300/4111] lr: 2.7725e-05 eta: 2:37:05 time: 0.7365 data_time: 0.5958 memory: 6318 loss: 0.2192 +2023/06/05 21:51:01 - mmengine - INFO - Epoch(train) [8][ 400/4111] lr: 2.7452e-05 eta: 2:35:46 time: 0.7860 data_time: 0.6443 memory: 6318 loss: 0.2184 +2023/06/05 21:52:15 - mmengine - INFO - Epoch(train) [8][ 500/4111] lr: 2.7181e-05 eta: 2:34:25 time: 0.7289 data_time: 0.5866 memory: 6318 loss: 0.2303 +2023/06/05 21:53:32 - mmengine - INFO - Epoch(train) [8][ 600/4111] lr: 2.6911e-05 eta: 2:33:07 time: 0.7699 data_time: 0.6300 memory: 6318 loss: 0.2199 +2023/06/05 21:55:06 - mmengine - INFO - Epoch(train) [8][ 700/4111] lr: 2.6644e-05 eta: 2:31:55 time: 0.7539 data_time: 0.6140 memory: 6318 loss: 0.2198 +2023/06/05 21:56:27 - mmengine - INFO - Epoch(train) [8][ 800/4111] lr: 2.6377e-05 eta: 2:30:37 time: 0.7429 data_time: 0.6002 memory: 6318 loss: 0.2317 +2023/06/05 21:57:45 - mmengine - INFO - Epoch(train) [8][ 900/4111] lr: 2.6113e-05 eta: 2:29:19 time: 0.8370 data_time: 0.6951 memory: 6318 loss: 0.2173 +2023/06/05 21:59:00 - mmengine - INFO - Epoch(train) [8][1000/4111] lr: 2.5850e-05 eta: 2:27:59 time: 0.7417 data_time: 0.6004 memory: 6318 loss: 0.2071 +2023/06/05 22:00:25 - mmengine - INFO - Epoch(train) [8][1100/4111] lr: 2.5589e-05 eta: 2:26:43 time: 0.8688 data_time: 0.7259 memory: 6318 loss: 0.2138 +2023/06/05 22:01:39 - mmengine - INFO - Epoch(train) [8][1200/4111] lr: 2.5330e-05 eta: 2:25:23 time: 0.6813 data_time: 0.5390 memory: 6318 loss: 0.2042 +2023/06/05 22:01:54 - mmengine - INFO - Exp name: resnet50_2xb256_all_1m_lr1e-4_aug_5e-1_20230605_152040 +2023/06/05 22:02:52 - mmengine - INFO - Epoch(train) [8][1300/4111] lr: 2.5072e-05 eta: 2:24:03 time: 0.7317 data_time: 0.5912 memory: 6318 loss: 0.2301 +2023/06/05 22:04:07 - mmengine - INFO - Epoch(train) [8][1400/4111] lr: 2.4816e-05 eta: 2:22:43 time: 0.7430 data_time: 0.6030 memory: 6318 loss: 0.2234 +2023/06/05 22:05:29 - mmengine - INFO - Epoch(train) [8][1500/4111] lr: 2.4562e-05 eta: 2:21:26 time: 0.7748 data_time: 0.6337 memory: 6318 loss: 0.2208 +2023/06/05 22:06:52 - mmengine - INFO - Epoch(train) [8][1600/4111] lr: 2.4309e-05 eta: 2:20:10 time: 0.7498 data_time: 0.6095 memory: 6318 loss: 0.2225 +2023/06/05 22:08:12 - mmengine - INFO - Epoch(train) [8][1700/4111] lr: 2.4059e-05 eta: 2:18:52 time: 0.8155 data_time: 0.6754 memory: 6318 loss: 0.2218 +2023/06/05 22:09:28 - mmengine - INFO - Epoch(train) [8][1800/4111] lr: 2.3810e-05 eta: 2:17:33 time: 0.7130 data_time: 0.5739 memory: 6318 loss: 0.2264 +2023/06/05 22:10:50 - mmengine - INFO - Epoch(train) [8][1900/4111] lr: 2.3563e-05 eta: 2:16:16 time: 0.7513 data_time: 0.6098 memory: 6318 loss: 0.2345 +2023/06/05 22:12:09 - mmengine - INFO - Epoch(train) [8][2000/4111] lr: 2.3318e-05 eta: 2:14:58 time: 0.7409 data_time: 0.5980 memory: 6318 loss: 0.2221 +2023/06/05 22:13:25 - mmengine - INFO - Epoch(train) [8][2100/4111] lr: 2.3075e-05 eta: 2:13:38 time: 0.7459 data_time: 0.6060 memory: 6318 loss: 0.2210 +2023/06/05 22:14:41 - mmengine - INFO - Epoch(train) [8][2200/4111] lr: 2.2833e-05 eta: 2:12:19 time: 0.7767 data_time: 0.6351 memory: 6318 loss: 0.2263 +2023/06/05 22:14:57 - mmengine - INFO - Exp name: resnet50_2xb256_all_1m_lr1e-4_aug_5e-1_20230605_152040 +2023/06/05 22:15:58 - mmengine - INFO - Epoch(train) [8][2300/4111] lr: 2.2594e-05 eta: 2:11:00 time: 0.7762 data_time: 0.6353 memory: 6318 loss: 0.2037 +2023/06/05 22:17:14 - mmengine - INFO - Epoch(train) [8][2400/4111] lr: 2.2356e-05 eta: 2:09:41 time: 0.7605 data_time: 0.6200 memory: 6318 loss: 0.2065 +2023/06/05 22:18:31 - mmengine - INFO - Epoch(train) [8][2500/4111] lr: 2.2120e-05 eta: 2:08:22 time: 0.8336 data_time: 0.6804 memory: 6318 loss: 0.2050 +2023/06/05 22:19:46 - mmengine - INFO - Epoch(train) [8][2600/4111] lr: 2.1886e-05 eta: 2:07:03 time: 0.7369 data_time: 0.5961 memory: 6318 loss: 0.2050 +2023/06/05 22:21:01 - mmengine - INFO - Epoch(train) [8][2700/4111] lr: 2.1655e-05 eta: 2:05:44 time: 0.7749 data_time: 0.6269 memory: 6318 loss: 0.2052 +2023/06/05 22:22:23 - mmengine - INFO - Epoch(train) [8][2800/4111] lr: 2.1425e-05 eta: 2:04:26 time: 0.8055 data_time: 0.6658 memory: 6318 loss: 0.1943 +2023/06/05 22:23:39 - mmengine - INFO - Epoch(train) [8][2900/4111] lr: 2.1197e-05 eta: 2:03:08 time: 0.7821 data_time: 0.6407 memory: 6318 loss: 0.2202 +2023/06/05 22:24:55 - mmengine - INFO - Epoch(train) [8][3000/4111] lr: 2.0971e-05 eta: 2:01:49 time: 0.7509 data_time: 0.6112 memory: 6318 loss: 0.2110 +2023/06/05 22:26:12 - mmengine - INFO - Epoch(train) [8][3100/4111] lr: 2.0747e-05 eta: 2:00:30 time: 0.7708 data_time: 0.6297 memory: 6318 loss: 0.2108 +2023/06/05 22:27:27 - mmengine - INFO - Epoch(train) [8][3200/4111] lr: 2.0525e-05 eta: 1:59:10 time: 0.7457 data_time: 0.6054 memory: 6318 loss: 0.2018 +2023/06/05 22:27:42 - mmengine - INFO - Exp name: resnet50_2xb256_all_1m_lr1e-4_aug_5e-1_20230605_152040 +2023/06/05 22:28:44 - mmengine - INFO - Epoch(train) [8][3300/4111] lr: 2.0305e-05 eta: 1:57:52 time: 0.7944 data_time: 0.6531 memory: 6318 loss: 0.2070 +2023/06/05 22:29:59 - mmengine - INFO - Epoch(train) [8][3400/4111] lr: 2.0087e-05 eta: 1:56:33 time: 0.7608 data_time: 0.6205 memory: 6318 loss: 0.2094 +2023/06/05 22:31:13 - mmengine - INFO - Epoch(train) [8][3500/4111] lr: 1.9871e-05 eta: 1:55:13 time: 0.7134 data_time: 0.5727 memory: 6318 loss: 0.2290 +2023/06/05 22:32:30 - mmengine - INFO - Epoch(train) [8][3600/4111] lr: 1.9657e-05 eta: 1:53:54 time: 0.7433 data_time: 0.6035 memory: 6318 loss: 0.2058 +2023/06/05 22:33:43 - mmengine - INFO - Epoch(train) [8][3700/4111] lr: 1.9445e-05 eta: 1:52:35 time: 0.7022 data_time: 0.5592 memory: 6318 loss: 0.2393 +2023/06/05 22:34:58 - mmengine - INFO - Epoch(train) [8][3800/4111] lr: 1.9235e-05 eta: 1:51:16 time: 0.7248 data_time: 0.5849 memory: 6318 loss: 0.2197 +2023/06/05 22:36:15 - mmengine - INFO - Epoch(train) [8][3900/4111] lr: 1.9028e-05 eta: 1:49:57 time: 0.8582 data_time: 0.7181 memory: 6318 loss: 0.2124 +2023/06/05 22:37:31 - mmengine - INFO - Epoch(train) [8][4000/4111] lr: 1.8822e-05 eta: 1:48:38 time: 0.7843 data_time: 0.6347 memory: 6318 loss: 0.2031 +2023/06/05 22:38:46 - mmengine - INFO - Epoch(train) [8][4100/4111] lr: 1.8619e-05 eta: 1:47:19 time: 0.7361 data_time: 0.5957 memory: 6318 loss: 0.1992 +2023/06/05 22:38:54 - mmengine - INFO - Exp name: resnet50_2xb256_all_1m_lr1e-4_aug_5e-1_20230605_152040 +2023/06/05 22:38:54 - mmengine - INFO - Saving checkpoint at 8 epochs +2023/06/05 22:39:43 - mmengine - INFO - Epoch(val) [8][100/119] eta: 0:00:08 time: 0.5803 data_time: 0.4920 memory: 6318 +2023/06/05 22:40:10 - mmengine - INFO - Epoch(val) [8][119/119] accuracy/top1: 92.7528 data_time: 0.3964 time: 0.4843 +2023/06/05 22:41:27 - mmengine - INFO - Epoch(train) [9][ 100/4111] lr: 1.8395e-05 eta: 1:45:52 time: 0.7618 data_time: 0.6216 memory: 6318 loss: 0.2079 +2023/06/05 22:41:40 - mmengine - INFO - Exp name: resnet50_2xb256_all_1m_lr1e-4_aug_5e-1_20230605_152040 +2023/06/05 22:42:44 - mmengine - INFO - Epoch(train) [9][ 200/4111] lr: 1.8196e-05 eta: 1:44:34 time: 0.7839 data_time: 0.6424 memory: 6318 loss: 0.2028 +2023/06/05 22:44:01 - mmengine - INFO - Epoch(train) [9][ 300/4111] lr: 1.7999e-05 eta: 1:43:15 time: 0.7687 data_time: 0.6288 memory: 6318 loss: 0.2058 +2023/06/05 22:45:18 - mmengine - INFO - Epoch(train) [9][ 400/4111] lr: 1.7805e-05 eta: 1:41:57 time: 0.7503 data_time: 0.6097 memory: 6318 loss: 0.2070 +2023/06/05 22:46:32 - mmengine - INFO - Epoch(train) [9][ 500/4111] lr: 1.7612e-05 eta: 1:40:38 time: 0.7100 data_time: 0.5703 memory: 6318 loss: 0.2184 +2023/06/05 22:47:48 - mmengine - INFO - Epoch(train) [9][ 600/4111] lr: 1.7422e-05 eta: 1:39:19 time: 0.7838 data_time: 0.6435 memory: 6318 loss: 0.2091 +2023/06/05 22:49:03 - mmengine - INFO - Epoch(train) [9][ 700/4111] lr: 1.7234e-05 eta: 1:38:00 time: 0.7484 data_time: 0.6079 memory: 6318 loss: 0.2412 +2023/06/05 22:50:18 - mmengine - INFO - Epoch(train) [9][ 800/4111] lr: 1.7048e-05 eta: 1:36:41 time: 0.7618 data_time: 0.6214 memory: 6318 loss: 0.2058 +2023/06/05 22:51:34 - mmengine - INFO - Epoch(train) [9][ 900/4111] lr: 1.6864e-05 eta: 1:35:22 time: 0.7668 data_time: 0.6261 memory: 6318 loss: 0.2207 +2023/06/05 22:52:49 - mmengine - INFO - Epoch(train) [9][1000/4111] lr: 1.6683e-05 eta: 1:34:03 time: 0.7379 data_time: 0.5972 memory: 6318 loss: 0.2070 +2023/06/05 22:54:06 - mmengine - INFO - Epoch(train) [9][1100/4111] lr: 1.6504e-05 eta: 1:32:45 time: 0.7950 data_time: 0.6551 memory: 6318 loss: 0.2059 +2023/06/05 22:54:20 - mmengine - INFO - Exp name: resnet50_2xb256_all_1m_lr1e-4_aug_5e-1_20230605_152040 +2023/06/05 22:55:24 - mmengine - INFO - Epoch(train) [9][1200/4111] lr: 1.6327e-05 eta: 1:31:27 time: 1.0441 data_time: 0.9051 memory: 6318 loss: 0.2000 +2023/06/05 22:56:40 - mmengine - INFO - Epoch(train) [9][1300/4111] lr: 1.6152e-05 eta: 1:30:08 time: 0.7461 data_time: 0.6062 memory: 6318 loss: 0.2065 +2023/06/05 22:58:15 - mmengine - INFO - Epoch(train) [9][1400/4111] lr: 1.5980e-05 eta: 1:28:54 time: 0.7663 data_time: 0.6238 memory: 6318 loss: 0.2185 +2023/06/05 22:59:30 - mmengine - INFO - Epoch(train) [9][1500/4111] lr: 1.5810e-05 eta: 1:27:35 time: 0.7178 data_time: 0.5701 memory: 6318 loss: 0.2023 +2023/06/05 23:00:46 - mmengine - INFO - Epoch(train) [9][1600/4111] lr: 1.5642e-05 eta: 1:26:16 time: 0.7258 data_time: 0.5865 memory: 6318 loss: 0.2161 +2023/06/05 23:02:01 - mmengine - INFO - Epoch(train) [9][1700/4111] lr: 1.5476e-05 eta: 1:24:57 time: 0.7565 data_time: 0.6171 memory: 6318 loss: 0.2161 +2023/06/05 23:03:18 - mmengine - INFO - Epoch(train) [9][1800/4111] lr: 1.5313e-05 eta: 1:23:39 time: 0.7376 data_time: 0.5973 memory: 6318 loss: 0.2140 +2023/06/05 23:04:33 - mmengine - INFO - Epoch(train) [9][1900/4111] lr: 1.5152e-05 eta: 1:22:20 time: 0.7722 data_time: 0.6321 memory: 6318 loss: 0.2282 +2023/06/05 23:05:49 - mmengine - INFO - Epoch(train) [9][2000/4111] lr: 1.4993e-05 eta: 1:21:02 time: 0.7084 data_time: 0.5628 memory: 6318 loss: 0.2132 +2023/06/05 23:07:21 - mmengine - INFO - Epoch(train) [9][2100/4111] lr: 1.4837e-05 eta: 1:19:46 time: 0.7355 data_time: 0.5949 memory: 6318 loss: 0.2250 +2023/06/05 23:07:29 - mmengine - INFO - Exp name: resnet50_2xb256_all_1m_lr1e-4_aug_5e-1_20230605_152040 +2023/06/05 23:08:36 - mmengine - INFO - Epoch(train) [9][2200/4111] lr: 1.4683e-05 eta: 1:18:27 time: 0.7318 data_time: 0.5912 memory: 6318 loss: 0.2193 +2023/06/05 23:09:53 - mmengine - INFO - Epoch(train) [9][2300/4111] lr: 1.4532e-05 eta: 1:17:09 time: 0.7496 data_time: 0.6078 memory: 6318 loss: 0.2304 +2023/06/05 23:11:12 - mmengine - INFO - Epoch(train) [9][2400/4111] lr: 1.4382e-05 eta: 1:15:51 time: 1.2516 data_time: 1.1011 memory: 6318 loss: 0.2287 +2023/06/05 23:12:33 - mmengine - INFO - Epoch(train) [9][2500/4111] lr: 1.4235e-05 eta: 1:14:33 time: 0.8279 data_time: 0.6855 memory: 6318 loss: 0.2134 +2023/06/05 23:13:51 - mmengine - INFO - Epoch(train) [9][2600/4111] lr: 1.4091e-05 eta: 1:13:15 time: 0.7562 data_time: 0.6148 memory: 6318 loss: 0.2074 +2023/06/05 23:15:11 - mmengine - INFO - Epoch(train) [9][2700/4111] lr: 1.3949e-05 eta: 1:11:57 time: 0.7231 data_time: 0.5825 memory: 6318 loss: 0.2286 +2023/06/05 23:16:34 - mmengine - INFO - Epoch(train) [9][2800/4111] lr: 1.3809e-05 eta: 1:10:40 time: 0.7823 data_time: 0.6417 memory: 6318 loss: 0.1930 +2023/06/05 23:17:51 - mmengine - INFO - Epoch(train) [9][2900/4111] lr: 1.3672e-05 eta: 1:09:21 time: 0.8071 data_time: 0.6662 memory: 6318 loss: 0.2063 +2023/06/05 23:19:10 - mmengine - INFO - Epoch(train) [9][3000/4111] lr: 1.3537e-05 eta: 1:08:03 time: 0.8116 data_time: 0.6649 memory: 6318 loss: 0.2107 +2023/06/05 23:20:31 - mmengine - INFO - Epoch(train) [9][3100/4111] lr: 1.3405e-05 eta: 1:06:45 time: 0.7700 data_time: 0.6291 memory: 6318 loss: 0.2335 +2023/06/05 23:22:08 - mmengine - INFO - Exp name: resnet50_2xb256_all_1m_lr1e-4_aug_5e-1_20230605_152040 +2023/06/05 23:23:15 - mmengine - INFO - Epoch(train) [9][3200/4111] lr: 1.3275e-05 eta: 1:05:39 time: 0.7781 data_time: 0.6371 memory: 6318 loss: 0.2093 +2023/06/05 23:24:33 - mmengine - INFO - Epoch(train) [9][3300/4111] lr: 1.3147e-05 eta: 1:04:21 time: 0.7801 data_time: 0.6383 memory: 6318 loss: 0.2197 +2023/06/05 23:25:52 - mmengine - INFO - Epoch(train) [9][3400/4111] lr: 1.3022e-05 eta: 1:03:02 time: 0.7636 data_time: 0.6223 memory: 6318 loss: 0.2027 +2023/06/05 23:27:07 - mmengine - INFO - Epoch(train) [9][3500/4111] lr: 1.2899e-05 eta: 1:01:43 time: 0.6885 data_time: 0.5479 memory: 6318 loss: 0.2135 +2023/06/05 23:28:20 - mmengine - INFO - Epoch(train) [9][3600/4111] lr: 1.2779e-05 eta: 1:00:24 time: 0.7353 data_time: 0.5945 memory: 6318 loss: 0.2067 +2023/06/05 23:29:40 - mmengine - INFO - Epoch(train) [9][3700/4111] lr: 1.2661e-05 eta: 0:59:06 time: 0.7595 data_time: 0.6184 memory: 6318 loss: 0.2092 +2023/06/05 23:30:58 - mmengine - INFO - Epoch(train) [9][3800/4111] lr: 1.2546e-05 eta: 0:57:47 time: 0.7966 data_time: 0.6569 memory: 6318 loss: 0.2110 +2023/06/05 23:32:16 - mmengine - INFO - Epoch(train) [9][3900/4111] lr: 1.2433e-05 eta: 0:56:29 time: 0.8309 data_time: 0.6890 memory: 6318 loss: 0.2114 +2023/06/05 23:33:32 - mmengine - INFO - Epoch(train) [9][4000/4111] lr: 1.2323e-05 eta: 0:55:10 time: 0.7671 data_time: 0.6255 memory: 6318 loss: 0.2108 +2023/06/05 23:34:49 - mmengine - INFO - Epoch(train) [9][4100/4111] lr: 1.2215e-05 eta: 0:53:52 time: 0.7465 data_time: 0.6040 memory: 6318 loss: 0.2191 +2023/06/05 23:34:56 - mmengine - INFO - Exp name: resnet50_2xb256_all_1m_lr1e-4_aug_5e-1_20230605_152040 +2023/06/05 23:34:56 - mmengine - INFO - Saving checkpoint at 9 epochs +2023/06/05 23:35:45 - mmengine - INFO - Epoch(val) [9][100/119] eta: 0:00:08 time: 0.6058 data_time: 0.5153 memory: 6318 +2023/06/05 23:36:12 - mmengine - INFO - Epoch(val) [9][119/119] accuracy/top1: 93.2691 data_time: 0.4000 time: 0.4894 +2023/06/05 23:36:20 - mmengine - INFO - Exp name: resnet50_2xb256_all_1m_lr1e-4_aug_5e-1_20230605_152040 +2023/06/05 23:37:31 - mmengine - INFO - Epoch(train) [10][ 100/4111] lr: 1.2098e-05 eta: 0:52:24 time: 0.7186 data_time: 0.5794 memory: 6318 loss: 0.2179 +2023/06/05 23:38:48 - mmengine - INFO - Epoch(train) [10][ 200/4111] lr: 1.1996e-05 eta: 0:51:06 time: 0.7585 data_time: 0.6175 memory: 6318 loss: 0.2178 +2023/06/05 23:40:03 - mmengine - INFO - Epoch(train) [10][ 300/4111] lr: 1.1896e-05 eta: 0:49:47 time: 0.7452 data_time: 0.6042 memory: 6318 loss: 0.2118 +2023/06/05 23:41:19 - mmengine - INFO - Epoch(train) [10][ 400/4111] lr: 1.1798e-05 eta: 0:48:29 time: 0.7072 data_time: 0.5670 memory: 6318 loss: 0.2085 +2023/06/05 23:42:36 - mmengine - INFO - Epoch(train) [10][ 500/4111] lr: 1.1703e-05 eta: 0:47:10 time: 0.7323 data_time: 0.5916 memory: 6318 loss: 0.2231 +2023/06/05 23:43:51 - mmengine - INFO - Epoch(train) [10][ 600/4111] lr: 1.1611e-05 eta: 0:45:51 time: 0.7914 data_time: 0.6482 memory: 6318 loss: 0.2028 +2023/06/05 23:45:07 - mmengine - INFO - Epoch(train) [10][ 700/4111] lr: 1.1521e-05 eta: 0:44:33 time: 0.7402 data_time: 0.6006 memory: 6318 loss: 0.2071 +2023/06/05 23:46:22 - mmengine - INFO - Epoch(train) [10][ 800/4111] lr: 1.1434e-05 eta: 0:43:14 time: 0.7175 data_time: 0.5758 memory: 6318 loss: 0.1967 +2023/06/05 23:47:39 - mmengine - INFO - Epoch(train) [10][ 900/4111] lr: 1.1349e-05 eta: 0:41:56 time: 0.7908 data_time: 0.6488 memory: 6318 loss: 0.2157 +2023/06/05 23:48:57 - mmengine - INFO - Epoch(train) [10][1000/4111] lr: 1.1267e-05 eta: 0:40:37 time: 0.7675 data_time: 0.6274 memory: 6318 loss: 0.2113 +2023/06/05 23:48:58 - mmengine - INFO - Exp name: resnet50_2xb256_all_1m_lr1e-4_aug_5e-1_20230605_152040 +2023/06/05 23:50:13 - mmengine - INFO - Epoch(train) [10][1100/4111] lr: 1.1187e-05 eta: 0:39:19 time: 0.7371 data_time: 0.5963 memory: 6318 loss: 0.2048 +2023/06/05 23:51:29 - mmengine - INFO - Epoch(train) [10][1200/4111] lr: 1.1110e-05 eta: 0:38:00 time: 0.7521 data_time: 0.6116 memory: 6318 loss: 0.1992 +2023/06/05 23:52:47 - mmengine - INFO - Epoch(train) [10][1300/4111] lr: 1.1035e-05 eta: 0:36:42 time: 0.8128 data_time: 0.6719 memory: 6318 loss: 0.2173 +2023/06/05 23:54:02 - mmengine - INFO - Epoch(train) [10][1400/4111] lr: 1.0963e-05 eta: 0:35:23 time: 0.6952 data_time: 0.5537 memory: 6318 loss: 0.2113 +2023/06/05 23:55:17 - mmengine - INFO - Epoch(train) [10][1500/4111] lr: 1.0893e-05 eta: 0:34:05 time: 0.7728 data_time: 0.6330 memory: 6318 loss: 0.2066 +2023/06/05 23:56:33 - mmengine - INFO - Epoch(train) [10][1600/4111] lr: 1.0827e-05 eta: 0:32:46 time: 0.7670 data_time: 0.6279 memory: 6318 loss: 0.2006 +2023/06/05 23:57:50 - mmengine - INFO - Epoch(train) [10][1700/4111] lr: 1.0762e-05 eta: 0:31:28 time: 0.7832 data_time: 0.6421 memory: 6318 loss: 0.2029 +2023/06/05 23:59:07 - mmengine - INFO - Epoch(train) [10][1800/4111] lr: 1.0701e-05 eta: 0:30:09 time: 0.7497 data_time: 0.6088 memory: 6318 loss: 0.2127 +2023/06/06 00:00:23 - mmengine - INFO - Epoch(train) [10][1900/4111] lr: 1.0641e-05 eta: 0:28:51 time: 0.8233 data_time: 0.6813 memory: 6318 loss: 0.2037 +2023/06/06 00:01:35 - mmengine - INFO - Epoch(train) [10][2000/4111] lr: 1.0585e-05 eta: 0:27:32 time: 0.7736 data_time: 0.6327 memory: 6318 loss: 0.1878 +2023/06/06 00:01:35 - mmengine - INFO - Exp name: resnet50_2xb256_all_1m_lr1e-4_aug_5e-1_20230605_152040 +2023/06/06 00:02:51 - mmengine - INFO - Epoch(train) [10][2100/4111] lr: 1.0531e-05 eta: 0:26:14 time: 0.8368 data_time: 0.6955 memory: 6318 loss: 0.1975 +2023/06/06 00:04:08 - mmengine - INFO - Epoch(train) [10][2200/4111] lr: 1.0480e-05 eta: 0:24:55 time: 0.7826 data_time: 0.6411 memory: 6318 loss: 0.2140 +2023/06/06 00:05:25 - mmengine - INFO - Epoch(train) [10][2300/4111] lr: 1.0431e-05 eta: 0:23:37 time: 0.7488 data_time: 0.5959 memory: 6318 loss: 0.2210 +2023/06/06 00:06:45 - mmengine - INFO - Epoch(train) [10][2400/4111] lr: 1.0385e-05 eta: 0:22:19 time: 0.7800 data_time: 0.6397 memory: 6318 loss: 0.2126 +2023/06/06 00:08:00 - mmengine - INFO - Epoch(train) [10][2500/4111] lr: 1.0341e-05 eta: 0:21:01 time: 0.7297 data_time: 0.5885 memory: 6318 loss: 0.2287 +2023/06/06 00:09:16 - mmengine - INFO - Epoch(train) [10][2600/4111] lr: 1.0300e-05 eta: 0:19:42 time: 0.7461 data_time: 0.6051 memory: 6318 loss: 0.2171 +2023/06/06 00:10:32 - mmengine - INFO - Epoch(train) [10][2700/4111] lr: 1.0262e-05 eta: 0:18:24 time: 0.7759 data_time: 0.6356 memory: 6318 loss: 0.1918 +2023/06/06 00:11:51 - mmengine - INFO - Epoch(train) [10][2800/4111] lr: 1.0226e-05 eta: 0:17:06 time: 0.7720 data_time: 0.6307 memory: 6318 loss: 0.2038 +2023/06/06 00:13:07 - mmengine - INFO - Epoch(train) [10][2900/4111] lr: 1.0193e-05 eta: 0:15:47 time: 0.7604 data_time: 0.6178 memory: 6318 loss: 0.2056 +2023/06/06 00:14:26 - mmengine - INFO - Epoch(train) [10][3000/4111] lr: 1.0162e-05 eta: 0:14:29 time: 0.7391 data_time: 0.5979 memory: 6318 loss: 0.1937 +2023/06/06 00:14:26 - mmengine - INFO - Exp name: resnet50_2xb256_all_1m_lr1e-4_aug_5e-1_20230605_152040 +2023/06/06 00:15:43 - mmengine - INFO - Epoch(train) [10][3100/4111] lr: 1.0135e-05 eta: 0:13:11 time: 0.7999 data_time: 0.6583 memory: 6318 loss: 0.2047 +2023/06/06 00:16:57 - mmengine - INFO - Epoch(train) [10][3200/4111] lr: 1.0109e-05 eta: 0:11:52 time: 0.7180 data_time: 0.5788 memory: 6318 loss: 0.1988 +2023/06/06 00:18:22 - mmengine - INFO - Epoch(train) [10][3300/4111] lr: 1.0087e-05 eta: 0:10:34 time: 0.8104 data_time: 0.6692 memory: 6318 loss: 0.2237 +2023/06/06 00:19:45 - mmengine - INFO - Epoch(train) [10][3400/4111] lr: 1.0067e-05 eta: 0:09:16 time: 0.8319 data_time: 0.6905 memory: 6318 loss: 0.2012 +2023/06/06 00:21:09 - mmengine - INFO - Epoch(train) [10][3500/4111] lr: 1.0049e-05 eta: 0:07:58 time: 1.1398 data_time: 0.9919 memory: 6318 loss: 0.2020 +2023/06/06 00:22:33 - mmengine - INFO - Epoch(train) [10][3600/4111] lr: 1.0034e-05 eta: 0:06:40 time: 0.7321 data_time: 0.5915 memory: 6318 loss: 0.2074 +2023/06/06 00:23:51 - mmengine - INFO - Epoch(train) [10][3700/4111] lr: 1.0022e-05 eta: 0:05:21 time: 0.7483 data_time: 0.6071 memory: 6318 loss: 0.2059 +2023/06/06 00:25:09 - mmengine - INFO - Epoch(train) [10][3800/4111] lr: 1.0013e-05 eta: 0:04:03 time: 0.8581 data_time: 0.7183 memory: 6318 loss: 0.2115 +2023/06/06 00:26:30 - mmengine - INFO - Epoch(train) [10][3900/4111] lr: 1.0006e-05 eta: 0:02:45 time: 0.8053 data_time: 0.6576 memory: 6318 loss: 0.2119 +2023/06/06 00:27:49 - mmengine - INFO - Epoch(train) [10][4000/4111] lr: 1.0002e-05 eta: 0:01:26 time: 0.8099 data_time: 0.6687 memory: 6318 loss: 0.2107 +2023/06/06 00:27:49 - mmengine - INFO - Exp name: resnet50_2xb256_all_1m_lr1e-4_aug_5e-1_20230605_152040 +2023/06/06 00:29:05 - mmengine - INFO - Epoch(train) [10][4100/4111] lr: 1.0000e-05 eta: 0:00:08 time: 0.7658 data_time: 0.6263 memory: 6318 loss: 0.1976 +2023/06/06 00:29:12 - mmengine - INFO - Exp name: resnet50_2xb256_all_1m_lr1e-4_aug_5e-1_20230605_152040 +2023/06/06 00:29:12 - mmengine - INFO - Saving checkpoint at 10 epochs +2023/06/06 00:30:00 - mmengine - INFO - Epoch(val) [10][100/119] eta: 0:00:08 time: 0.6157 data_time: 0.5277 memory: 6318 +2023/06/06 00:30:28 - mmengine - INFO - Epoch(val) [10][119/119] accuracy/top1: 93.5107 data_time: 0.3998 time: 0.4877 diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/20230605_152040.json b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/20230605_152040.json new file mode 100644 index 0000000000000000000000000000000000000000..e49ebb98be8b4598bdcaf2260aacc506d3abfed9 --- /dev/null +++ b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/20230605_152040.json @@ -0,0 +1,420 @@ +{"lr": 9.999871217862936e-05, "data_time": 0.5505931377410889, "loss": 0.6505624175071716, "time": 0.6934590101242065, "epoch": 1, "memory": 9436, "step": 100} +{"lr": 9.999479662539605e-05, "data_time": 0.5152626276016236, "loss": 0.5974045336246491, "time": 0.6552257299423218, "epoch": 1, "memory": 6319, "step": 200} +{"lr": 9.99882534375654e-05, "data_time": 0.001011061668395996, "loss": 0.5777045428752899, "time": 0.6513210296630859, "epoch": 1, "memory": 6319, "step": 300} +{"lr": 9.997908299725075e-05, "data_time": 0.0008590936660766602, "loss": 0.5506176173686981, "time": 0.6428668737411499, "epoch": 1, "memory": 6319, "step": 400} +{"lr": 9.996728583999318e-05, "data_time": 0.2106086015701294, "loss": 0.5142255157232285, "time": 0.6030692815780639, "epoch": 1, "memory": 6319, "step": 500} +{"lr": 9.995286265473058e-05, "data_time": 0.34703359603881834, "loss": 0.5076296508312226, "time": 0.7148289918899536, "epoch": 1, "memory": 6319, "step": 600} +{"lr": 9.993581428375725e-05, "data_time": 0.46016180515289307, "loss": 0.4696363240480423, "time": 0.7294468402862548, "epoch": 1, "memory": 6319, "step": 700} +{"lr": 9.991614172267466e-05, "data_time": 0.4658113956451416, "loss": 0.4616829603910446, "time": 0.6683710813522339, "epoch": 1, "memory": 6319, "step": 800} +{"lr": 9.989384612033355e-05, "data_time": 0.43298189640045165, "loss": 0.4413948357105255, "time": 0.7080170631408691, "epoch": 1, "memory": 6319, "step": 900} +{"lr": 9.986892877876658e-05, "data_time": 0.43495018482208253, "loss": 0.4226416856050491, "time": 0.6916615962982178, "epoch": 1, "memory": 6319, "step": 1000} +{"lr": 9.984139115311244e-05, "data_time": 0.5044760465621948, "loss": 0.4144183903932571, "time": 0.6448450088500977, "epoch": 1, "memory": 6319, "step": 1100} +{"lr": 9.981123485153081e-05, "data_time": 0.2968530893325806, "loss": 0.4110246986150742, "time": 0.6685676097869873, "epoch": 1, "memory": 6319, "step": 1200} +{"lr": 9.977846163510851e-05, "data_time": 0.19977850914001466, "loss": 0.4085386097431183, "time": 0.6431279659271241, "epoch": 1, "memory": 6319, "step": 1300} +{"lr": 9.974307341775653e-05, "data_time": 0.5203864812850952, "loss": 0.3970152199268341, "time": 0.6990034580230713, "epoch": 1, "memory": 6319, "step": 1400} +{"lr": 9.970507226609833e-05, "data_time": 0.5476192951202392, "loss": 0.3996817052364349, "time": 0.6898721218109131, "epoch": 1, "memory": 6319, "step": 1500} +{"lr": 9.966446039934934e-05, "data_time": 0.5750117301940918, "loss": 0.39513109922409057, "time": 0.7154863595962524, "epoch": 1, "memory": 6319, "step": 1600} +{"lr": 9.962124018918683e-05, "data_time": 0.5280206203460693, "loss": 0.3815469264984131, "time": 0.6684962034225463, "epoch": 1, "memory": 6319, "step": 1700} +{"lr": 9.957541415961212e-05, "data_time": 0.5535875082015991, "loss": 0.39518000185489655, "time": 0.694165563583374, "epoch": 1, "memory": 6319, "step": 1800} +{"lr": 9.952698498680267e-05, "data_time": 0.4987358331680298, "loss": 0.3807822555303574, "time": 0.6390132427215576, "epoch": 1, "memory": 6319, "step": 1900} +{"lr": 9.947595549895583e-05, "data_time": 0.6053460836410522, "loss": 0.36012162566184996, "time": 0.7453249216079711, "epoch": 1, "memory": 6319, "step": 2000} +{"lr": 9.942232867612417e-05, "data_time": 0.5652534246444703, "loss": 0.37053406834602354, "time": 0.7054509878158569, "epoch": 1, "memory": 6319, "step": 2100} +{"lr": 9.93661076500406e-05, "data_time": 0.5398832798004151, "loss": 0.34339838922023774, "time": 0.6806561946868896, "epoch": 1, "memory": 6319, "step": 2200} +{"lr": 9.930729570393622e-05, "data_time": 0.5331053256988525, "loss": 0.35869871973991396, "time": 0.6771989345550538, "epoch": 1, "memory": 6319, "step": 2300} +{"lr": 9.924589627234824e-05, "data_time": 0.5512842416763306, "loss": 0.34724647700786593, "time": 0.6925719976425171, "epoch": 1, "memory": 6319, "step": 2400} +{"lr": 9.918191294091956e-05, "data_time": 0.5681015491485596, "loss": 0.3581976592540741, "time": 0.7094371080398559, "epoch": 1, "memory": 6319, "step": 2500} +{"lr": 9.91153494461892e-05, "data_time": 0.5202316999435425, "loss": 0.34429652988910675, "time": 0.6607050418853759, "epoch": 1, "memory": 6319, "step": 2600} +{"lr": 9.904620967537422e-05, "data_time": 0.545382022857666, "loss": 0.35300026535987855, "time": 0.685313367843628, "epoch": 1, "memory": 6319, "step": 2700} +{"lr": 9.897449766614277e-05, "data_time": 0.5922863721847534, "loss": 0.3269111156463623, "time": 0.7319374084472656, "epoch": 1, "memory": 6319, "step": 2800} +{"lr": 9.890021760637804e-05, "data_time": 0.5891611337661743, "loss": 0.34317596852779386, "time": 0.7286672115325927, "epoch": 1, "memory": 6319, "step": 2900} +{"lr": 9.882337383393409e-05, "data_time": 0.6334947109222412, "loss": 0.3492516875267029, "time": 0.7731748104095459, "epoch": 1, "memory": 6319, "step": 3000} +{"lr": 9.874397083638193e-05, "data_time": 0.5344001054763794, "loss": 0.3423819899559021, "time": 0.6852553844451904, "epoch": 1, "memory": 6319, "step": 3100} +{"lr": 9.866201325074828e-05, "data_time": 0.5381552457809449, "loss": 0.33956805765628817, "time": 0.6779947757720948, "epoch": 1, "memory": 6319, "step": 3200} +{"lr": 9.857750586324393e-05, "data_time": 0.5681901216506958, "loss": 0.3352639853954315, "time": 0.707914400100708, "epoch": 1, "memory": 6319, "step": 3300} +{"lr": 9.84904536089848e-05, "data_time": 0.719521164894104, "loss": 0.32742994725704194, "time": 0.8602179527282715, "epoch": 1, "memory": 6319, "step": 3400} +{"lr": 9.840086157170351e-05, "data_time": 0.5882129669189453, "loss": 0.3180242747068405, "time": 0.7277715682983399, "epoch": 1, "memory": 6319, "step": 3500} +{"lr": 9.830873498345255e-05, "data_time": 0.5058643341064453, "loss": 0.3155341863632202, "time": 0.6458086013793946, "epoch": 1, "memory": 6319, "step": 3600} +{"lr": 9.821407922429864e-05, "data_time": 0.5474719762802124, "loss": 0.3262654036283493, "time": 0.6877282142639161, "epoch": 1, "memory": 6319, "step": 3700} +{"lr": 9.811689982200877e-05, "data_time": 0.5631479740142822, "loss": 0.3217996835708618, "time": 0.7033457279205322, "epoch": 1, "memory": 6319, "step": 3800} +{"lr": 9.801720245172702e-05, "data_time": 0.5131886959075928, "loss": 0.3055334031581879, "time": 0.6539286375045776, "epoch": 1, "memory": 6319, "step": 3900} +{"lr": 9.791499293564358e-05, "data_time": 0.5458756446838379, "loss": 0.3265227496623993, "time": 0.6876251220703125, "epoch": 1, "memory": 6319, "step": 4000} +{"lr": 9.781027724265439e-05, "data_time": 0.6899309873580932, "loss": 0.3120740562677383, "time": 0.831060814857483, "epoch": 1, "memory": 6319, "step": 4100} +{"accuracy/top1": 82.47356414794922, "data_time": 0.393584958645476, "time": 0.48177839327259225, "step": 1} +{"lr": 9.769111539369434e-05, "data_time": 0.6067419528961182, "loss": 0.32421033680438993, "time": 0.748124074935913, "epoch": 2, "memory": 6318, "step": 4211} +{"lr": 9.758113190911623e-05, "data_time": 0.5759464502334595, "loss": 0.302031347155571, "time": 0.7165237665176392, "epoch": 2, "memory": 6318, "step": 4311} +{"lr": 9.746866174465984e-05, "data_time": 0.38157072067260744, "loss": 0.31559916138648986, "time": 0.6634103298187256, "epoch": 2, "memory": 6318, "step": 4411} +{"lr": 9.735371146842894e-05, "data_time": 0.5546189069747924, "loss": 0.31170307099819183, "time": 0.6952664852142334, "epoch": 2, "memory": 6318, "step": 4511} +{"lr": 9.723628779336255e-05, "data_time": 0.5642520904541015, "loss": 0.3174903839826584, "time": 0.7054730653762817, "epoch": 2, "memory": 6318, "step": 4611} +{"lr": 9.711639757684265e-05, "data_time": 0.5179787397384643, "loss": 0.31193001568317413, "time": 0.7149520874023437, "epoch": 2, "memory": 6318, "step": 4711} +{"lr": 9.699404782029395e-05, "data_time": 0.4935267448425293, "loss": 0.3097731113433838, "time": 0.6753772735595703, "epoch": 2, "memory": 6318, "step": 4811} +{"lr": 9.686924566877498e-05, "data_time": 0.3179853200912476, "loss": 0.32091196775436404, "time": 0.692458701133728, "epoch": 2, "memory": 6318, "step": 4911} +{"lr": 9.674199841056078e-05, "data_time": 0.45420470237731936, "loss": 0.2987965703010559, "time": 0.6875965356826782, "epoch": 2, "memory": 6318, "step": 5011} +{"lr": 9.661231347671705e-05, "data_time": 0.5439108610153198, "loss": 0.3439896672964096, "time": 0.69365394115448, "epoch": 2, "memory": 6318, "step": 5111} +{"lr": 9.64801984406668e-05, "data_time": 0.4245661020278931, "loss": 0.30733801126480104, "time": 0.7456601381301879, "epoch": 2, "memory": 6318, "step": 5211} +{"lr": 9.634566101774737e-05, "data_time": 0.19504849910736083, "loss": 0.3189211219549179, "time": 0.7636782407760621, "epoch": 2, "memory": 6318, "step": 5311} +{"lr": 9.62087090647604e-05, "data_time": 0.04529838562011719, "loss": 0.29260619133710863, "time": 0.6711793899536133, "epoch": 2, "memory": 6318, "step": 5411} +{"lr": 9.606935057951247e-05, "data_time": 0.2667135953903198, "loss": 0.3066791981458664, "time": 0.7659513711929321, "epoch": 2, "memory": 6318, "step": 5511} +{"lr": 9.592759370034884e-05, "data_time": 0.7144370317459107, "loss": 0.2962004154920578, "time": 1.105335783958435, "epoch": 2, "memory": 6318, "step": 5611} +{"lr": 9.578344670567721e-05, "data_time": 0.4193567276000977, "loss": 0.2966422110795975, "time": 0.6607670783996582, "epoch": 2, "memory": 6318, "step": 5711} +{"lr": 9.563691801348495e-05, "data_time": 0.6298110485076904, "loss": 0.29018100500106814, "time": 0.7693820238113404, "epoch": 2, "memory": 6318, "step": 5811} +{"lr": 9.548801618084738e-05, "data_time": 0.6293867111206055, "loss": 0.29744812846183777, "time": 0.7697560071945191, "epoch": 2, "memory": 6318, "step": 5911} +{"lr": 9.533674990342786e-05, "data_time": 0.6395065069198609, "loss": 0.30597208589315417, "time": 0.7797405958175659, "epoch": 2, "memory": 6318, "step": 6011} +{"lr": 9.518312801497029e-05, "data_time": 0.5134622812271118, "loss": 0.28178397864103316, "time": 0.6528750896453858, "epoch": 2, "memory": 6318, "step": 6111} +{"lr": 9.502715948678276e-05, "data_time": 0.4936917781829834, "loss": 0.3079785257577896, "time": 0.6340498685836792, "epoch": 2, "memory": 6318, "step": 6211} +{"lr": 9.486885342721427e-05, "data_time": 0.5484623908996582, "loss": 0.30840844213962554, "time": 0.6890472650527955, "epoch": 2, "memory": 6318, "step": 6311} +{"lr": 9.470821908112202e-05, "data_time": 0.645397424697876, "loss": 0.271228489279747, "time": 0.7857234954833985, "epoch": 2, "memory": 6318, "step": 6411} +{"lr": 9.454526582933214e-05, "data_time": 0.756637978553772, "loss": 0.28856208473443984, "time": 0.8973519563674927, "epoch": 2, "memory": 6318, "step": 6511} +{"lr": 9.438000318809176e-05, "data_time": 0.5645488262176513, "loss": 0.28264967501163485, "time": 0.7037885427474976, "epoch": 2, "memory": 6318, "step": 6611} +{"lr": 9.42124408085132e-05, "data_time": 0.5453065872192383, "loss": 0.2823116719722748, "time": 0.6865315675735474, "epoch": 2, "memory": 6318, "step": 6711} +{"lr": 9.404258847601036e-05, "data_time": 0.5627376079559326, "loss": 0.2907982677221298, "time": 0.7028753757476807, "epoch": 2, "memory": 6318, "step": 6811} +{"lr": 9.387045610972717e-05, "data_time": 0.5891398668289185, "loss": 0.29210465848445893, "time": 0.7290917873382569, "epoch": 2, "memory": 6318, "step": 6911} +{"lr": 9.369605376195838e-05, "data_time": 0.555685019493103, "loss": 0.2789680004119873, "time": 0.6954281091690063, "epoch": 2, "memory": 6318, "step": 7011} +{"lr": 9.351939161756269e-05, "data_time": 0.5706737756729126, "loss": 0.2751905009150505, "time": 0.7108277559280396, "epoch": 2, "memory": 6318, "step": 7111} +{"lr": 9.334047999336764e-05, "data_time": 0.6172459363937378, "loss": 0.28409205973148344, "time": 0.7575044393539428, "epoch": 2, "memory": 6318, "step": 7211} +{"lr": 9.315932933756752e-05, "data_time": 0.6438805818557739, "loss": 0.274951645731926, "time": 0.7835977792739868, "epoch": 2, "memory": 6318, "step": 7311} +{"lr": 9.297595022911287e-05, "data_time": 0.5824469566345215, "loss": 0.2724910795688629, "time": 0.7227290153503418, "epoch": 2, "memory": 6318, "step": 7411} +{"lr": 9.279035337709295e-05, "data_time": 0.6200707435607911, "loss": 0.28884666264057157, "time": 0.7615639209747315, "epoch": 2, "memory": 6318, "step": 7511} +{"lr": 9.260254962011016e-05, "data_time": 0.5936978816986084, "loss": 0.2855002909898758, "time": 0.7335208415985107, "epoch": 2, "memory": 6318, "step": 7611} +{"lr": 9.241254992564702e-05, "data_time": 0.6418598651885986, "loss": 0.26392817199230195, "time": 0.7825077533721924, "epoch": 2, "memory": 6318, "step": 7711} +{"lr": 9.22203653894259e-05, "data_time": 0.5894522190093994, "loss": 0.274297884106636, "time": 0.7303678512573242, "epoch": 2, "memory": 6318, "step": 7811} +{"lr": 9.202600723476098e-05, "data_time": 0.603200101852417, "loss": 0.26613368690013883, "time": 0.7431907415390014, "epoch": 2, "memory": 6318, "step": 7911} +{"lr": 9.182948681190268e-05, "data_time": 0.6074742078781128, "loss": 0.3004067957401276, "time": 0.7514505624771118, "epoch": 2, "memory": 6318, "step": 8011} +{"lr": 9.163081559737497e-05, "data_time": 0.6953469753265381, "loss": 0.28262611478567123, "time": 0.8359466791152954, "epoch": 2, "memory": 6318, "step": 8111} +{"lr": 9.143000519330517e-05, "data_time": 0.5899689674377442, "loss": 0.2500761702656746, "time": 0.7300094604492188, "epoch": 2, "memory": 6318, "step": 8211} +{"accuracy/top1": 77.86273956298828, "data_time": 0.37948496341705323, "time": 0.46813223759333294, "step": 2} +{"lr": 9.120461478753376e-05, "data_time": 0.6181094408035278, "loss": 0.27026086002588273, "time": 0.7582679986953735, "epoch": 3, "memory": 6318, "step": 8322} +{"lr": 9.099932932138134e-05, "data_time": 0.5869765520095825, "loss": 0.2636982038617134, "time": 0.7259285688400269, "epoch": 3, "memory": 6318, "step": 8422} +{"lr": 9.079194154362181e-05, "data_time": 0.5926679611206055, "loss": 0.3168958812952042, "time": 0.7332017660140991, "epoch": 3, "memory": 6318, "step": 8522} +{"lr": 9.058246356541787e-05, "data_time": 0.6440795183181762, "loss": 0.2605751916766167, "time": 0.7840817213058472, "epoch": 3, "memory": 6318, "step": 8622} +{"lr": 9.037090761999699e-05, "data_time": 0.5970780611038208, "loss": 0.28410408943891524, "time": 0.7363588571548462, "epoch": 3, "memory": 6318, "step": 8722} +{"lr": 9.015728606193701e-05, "data_time": 0.5913930892944336, "loss": 0.28051196932792666, "time": 0.7306947231292724, "epoch": 3, "memory": 6318, "step": 8822} +{"lr": 8.994161136644488e-05, "data_time": 0.6805069208145141, "loss": 0.2685139149427414, "time": 0.829882001876831, "epoch": 3, "memory": 6318, "step": 8922} +{"lr": 8.972389612862792e-05, "data_time": 0.5869032144546509, "loss": 0.2602032795548439, "time": 0.729338526725769, "epoch": 3, "memory": 6318, "step": 9022} +{"lr": 8.950415306275833e-05, "data_time": 0.6840051174163818, "loss": 0.2694586977362633, "time": 0.8367887496948242, "epoch": 3, "memory": 6318, "step": 9122} +{"lr": 8.928239500153059e-05, "data_time": 0.5554930925369262, "loss": 0.2827488571405411, "time": 0.6961770534515381, "epoch": 3, "memory": 6318, "step": 9222} +{"lr": 8.905863489531232e-05, "data_time": 0.639508056640625, "loss": 0.264517243206501, "time": 0.7878513813018799, "epoch": 3, "memory": 6318, "step": 9322} +{"lr": 8.883288581138775e-05, "data_time": 0.5916643142700195, "loss": 0.2758107051253319, "time": 0.7347429275512696, "epoch": 3, "memory": 6318, "step": 9422} +{"lr": 8.86051609331947e-05, "data_time": 0.6540353775024415, "loss": 0.28519781231880187, "time": 0.7988569736480713, "epoch": 3, "memory": 6318, "step": 9522} +{"lr": 8.837547355955467e-05, "data_time": 0.608176875114441, "loss": 0.26498973965644834, "time": 0.7500752449035645, "epoch": 3, "memory": 6318, "step": 9622} +{"lr": 8.814383710389614e-05, "data_time": 0.631160306930542, "loss": 0.27551338374614714, "time": 0.7737520933151245, "epoch": 3, "memory": 6318, "step": 9722} +{"lr": 8.791026509347146e-05, "data_time": 0.6326789379119873, "loss": 0.2672505512833595, "time": 0.7738667726516724, "epoch": 3, "memory": 6318, "step": 9822} +{"lr": 8.767477116856659e-05, "data_time": 0.6556741952896118, "loss": 0.27411281168460844, "time": 0.7964605093002319, "epoch": 3, "memory": 6318, "step": 9922} +{"lr": 8.743736908170466e-05, "data_time": 0.6806576490402222, "loss": 0.27539069652557374, "time": 0.8206354856491089, "epoch": 3, "memory": 6318, "step": 10022} +{"lr": 8.719807269684305e-05, "data_time": 0.5637223720550537, "loss": 0.26915688514709474, "time": 0.7057634592056274, "epoch": 3, "memory": 6318, "step": 10122} +{"lr": 8.695689598856356e-05, "data_time": 0.5881913900375366, "loss": 0.27323976457118987, "time": 0.7285332441329956, "epoch": 3, "memory": 6318, "step": 10222} +{"lr": 8.67138530412559e-05, "data_time": 0.6564640283584595, "loss": 0.26446687132120134, "time": 0.798209261894226, "epoch": 3, "memory": 6318, "step": 10322} +{"lr": 8.646895804829604e-05, "data_time": 0.5809694528579712, "loss": 0.2585653379559517, "time": 0.7214553356170654, "epoch": 3, "memory": 6318, "step": 10422} +{"lr": 8.622222531121664e-05, "data_time": 0.5559334754943848, "loss": 0.24912258386611938, "time": 0.6959937572479248, "epoch": 3, "memory": 6318, "step": 10522} +{"lr": 8.597366923887211e-05, "data_time": 0.6048402309417724, "loss": 0.27612168788909913, "time": 0.7447296380996704, "epoch": 3, "memory": 6318, "step": 10622} +{"lr": 8.57233043465972e-05, "data_time": 0.5658660888671875, "loss": 0.26694310158491136, "time": 0.7070073366165162, "epoch": 3, "memory": 6318, "step": 10722} +{"lr": 8.547114525535937e-05, "data_time": 0.6130959510803222, "loss": 0.2739654153585434, "time": 0.7524494171142578, "epoch": 3, "memory": 6318, "step": 10822} +{"lr": 8.521720669090474e-05, "data_time": 0.5844237327575683, "loss": 0.25660845935344695, "time": 0.7236541509628296, "epoch": 3, "memory": 6318, "step": 10922} +{"lr": 8.496150348289821e-05, "data_time": 0.6259898900985718, "loss": 0.2606738433241844, "time": 0.766187334060669, "epoch": 3, "memory": 6318, "step": 11022} +{"lr": 8.470405056405736e-05, "data_time": 0.4467794418334961, "loss": 0.24720416367053985, "time": 0.7321699380874633, "epoch": 3, "memory": 6318, "step": 11122} +{"lr": 8.44448629692807e-05, "data_time": 0.5983347415924072, "loss": 0.28130067586898805, "time": 0.7394391775131226, "epoch": 3, "memory": 6318, "step": 11222} +{"lr": 8.418395583476941e-05, "data_time": 0.607829236984253, "loss": 0.2544344007968903, "time": 0.7481575012207031, "epoch": 3, "memory": 6318, "step": 11322} +{"lr": 8.392134439714336e-05, "data_time": 0.604886507987976, "loss": 0.26822121143341066, "time": 0.7466015338897705, "epoch": 3, "memory": 6318, "step": 11422} +{"lr": 8.365704399255133e-05, "data_time": 0.5664644002914428, "loss": 0.25017236173152924, "time": 0.7069388628005981, "epoch": 3, "memory": 6318, "step": 11522} +{"lr": 8.339107005577562e-05, "data_time": 0.6967349529266358, "loss": 0.25618584603071215, "time": 0.844751501083374, "epoch": 3, "memory": 6318, "step": 11622} +{"lr": 8.31234381193305e-05, "data_time": 0.6183933973312378, "loss": 0.24026695638895035, "time": 0.7581479549407959, "epoch": 3, "memory": 6318, "step": 11722} +{"lr": 8.285416381255504e-05, "data_time": 0.6711583614349366, "loss": 0.2746740996837616, "time": 0.8233832359313965, "epoch": 3, "memory": 6318, "step": 11822} +{"lr": 8.258326286070047e-05, "data_time": 0.6249480009078979, "loss": 0.25514610558748246, "time": 0.7648244142532349, "epoch": 3, "memory": 6318, "step": 11922} +{"lr": 8.231075108401202e-05, "data_time": 0.6216845989227295, "loss": 0.2432646408677101, "time": 0.7675186634063721, "epoch": 3, "memory": 6318, "step": 12022} +{"lr": 8.203664439680493e-05, "data_time": 0.5821406126022339, "loss": 0.2458882138133049, "time": 0.7232086896896363, "epoch": 3, "memory": 6318, "step": 12122} +{"lr": 8.176095880653475e-05, "data_time": 0.5440454006195068, "loss": 0.2554523289203644, "time": 0.6927780866622925, "epoch": 3, "memory": 6318, "step": 12222} +{"lr": 8.148371041286308e-05, "data_time": 0.6835984945297241, "loss": 0.25288918167352675, "time": 0.8229263305664063, "epoch": 3, "memory": 6318, "step": 12322} +{"accuracy/top1": 79.14701843261719, "data_time": 0.39744665622711184, "time": 0.48549293478329975, "step": 3} +{"lr": 8.117415423358443e-05, "data_time": 0.8512596845626831, "loss": 0.2658050417900085, "time": 0.991992712020874, "epoch": 4, "memory": 6318, "step": 12433} +{"lr": 8.089366155759266e-05, "data_time": 0.6952439785003662, "loss": 0.25963141620159147, "time": 0.8345962285995483, "epoch": 4, "memory": 6318, "step": 12533} +{"lr": 8.061165672717273e-05, "data_time": 0.6424442529678345, "loss": 0.2485888347029686, "time": 0.7843495368957519, "epoch": 4, "memory": 6318, "step": 12633} +{"lr": 8.032815621102116e-05, "data_time": 0.5740828275680542, "loss": 0.26304204761981964, "time": 0.715699315071106, "epoch": 4, "memory": 6318, "step": 12733} +{"lr": 8.004317656518042e-05, "data_time": 0.6650530815124511, "loss": 0.26400354504585266, "time": 0.8057906150817871, "epoch": 4, "memory": 6318, "step": 12833} +{"lr": 7.975673443207223e-05, "data_time": 0.6941095352172851, "loss": 0.2509127527475357, "time": 0.8341012954711914, "epoch": 4, "memory": 6318, "step": 12933} +{"lr": 7.946884653952537e-05, "data_time": 0.627883791923523, "loss": 0.2448240414261818, "time": 0.7698372602462769, "epoch": 4, "memory": 6318, "step": 13033} +{"lr": 7.91795296997992e-05, "data_time": 0.6392451286315918, "loss": 0.25170914083719254, "time": 0.7796733856201172, "epoch": 4, "memory": 6318, "step": 13133} +{"lr": 7.888880080860151e-05, "data_time": 0.6729235649108887, "loss": 0.27842089384794233, "time": 0.8145662069320678, "epoch": 4, "memory": 6318, "step": 13233} +{"lr": 7.859667684410205e-05, "data_time": 0.611674165725708, "loss": 0.2623119980096817, "time": 0.7529521942138672, "epoch": 4, "memory": 6318, "step": 13333} +{"lr": 7.830317486594082e-05, "data_time": 0.6273994684219361, "loss": 0.2496202439069748, "time": 0.7699385404586792, "epoch": 4, "memory": 6318, "step": 13433} +{"lr": 7.800831201423207e-05, "data_time": 0.6156965255737304, "loss": 0.24734974801540374, "time": 0.7585714817047119, "epoch": 4, "memory": 6318, "step": 13533} +{"lr": 7.771210550856309e-05, "data_time": 0.6352834701538086, "loss": 0.24084002524614334, "time": 0.7980647087097168, "epoch": 4, "memory": 6318, "step": 13633} +{"lr": 7.741457264698884e-05, "data_time": 0.6363005876541138, "loss": 0.23988503217697144, "time": 0.7775790452957153, "epoch": 4, "memory": 6318, "step": 13733} +{"lr": 7.711573080502147e-05, "data_time": 0.6062970399856568, "loss": 0.26099351346492766, "time": 0.7483415842056275, "epoch": 4, "memory": 6318, "step": 13833} +{"lr": 7.681559743461609e-05, "data_time": 0.5898828268051147, "loss": 0.24848497658967972, "time": 0.7293738842010498, "epoch": 4, "memory": 6318, "step": 13933} +{"lr": 7.651419006315099e-05, "data_time": 0.5952056407928467, "loss": 0.26195275634527204, "time": 0.735258960723877, "epoch": 4, "memory": 6318, "step": 14033} +{"lr": 7.621152629240453e-05, "data_time": 0.5790565729141235, "loss": 0.24304310530424117, "time": 0.719720721244812, "epoch": 4, "memory": 6318, "step": 14133} +{"lr": 7.590762379752719e-05, "data_time": 0.5663726329803467, "loss": 0.25091861486434935, "time": 0.7070995569229126, "epoch": 4, "memory": 6318, "step": 14233} +{"lr": 7.560250032600911e-05, "data_time": 0.6194745302200317, "loss": 0.2575841635465622, "time": 0.769227123260498, "epoch": 4, "memory": 6318, "step": 14333} +{"lr": 7.52961736966437e-05, "data_time": 0.5746166706085205, "loss": 0.2373547524213791, "time": 0.7145738363265991, "epoch": 4, "memory": 6318, "step": 14433} +{"lr": 7.498866179848732e-05, "data_time": 0.6154284715652466, "loss": 0.2282430574297905, "time": 0.7566136121749878, "epoch": 4, "memory": 6318, "step": 14533} +{"lr": 7.467998258981415e-05, "data_time": 0.5762117624282836, "loss": 0.25059892386198046, "time": 0.7285410165786743, "epoch": 4, "memory": 6318, "step": 14633} +{"lr": 7.437015409706804e-05, "data_time": 0.6411524772644043, "loss": 0.24868184477090835, "time": 0.7812589406967163, "epoch": 4, "memory": 6318, "step": 14733} +{"lr": 7.405919441380929e-05, "data_time": 0.5939372777938843, "loss": 0.233042611181736, "time": 0.7347374200820923, "epoch": 4, "memory": 6318, "step": 14833} +{"lr": 7.374712169965821e-05, "data_time": 0.5857679843902588, "loss": 0.2688971906900406, "time": 0.7347111463546753, "epoch": 4, "memory": 6318, "step": 14933} +{"lr": 7.343395417923461e-05, "data_time": 0.6318930149078369, "loss": 0.2383084326982498, "time": 0.7723827600479126, "epoch": 4, "memory": 6318, "step": 15033} +{"lr": 7.311971014109342e-05, "data_time": 0.592315936088562, "loss": 0.25025821626186373, "time": 0.7330101251602172, "epoch": 4, "memory": 6318, "step": 15133} +{"lr": 7.280440793665682e-05, "data_time": 0.6075129508972168, "loss": 0.23638832420110703, "time": 0.7537994384765625, "epoch": 4, "memory": 6318, "step": 15233} +{"lr": 7.248806597914234e-05, "data_time": 0.6015305519104004, "loss": 0.23821211755275726, "time": 0.7430759906768799, "epoch": 4, "memory": 6318, "step": 15333} +{"lr": 7.217070274248765e-05, "data_time": 0.6364526271820068, "loss": 0.24179704636335372, "time": 0.7770860671997071, "epoch": 4, "memory": 6318, "step": 15433} +{"lr": 7.185233676027203e-05, "data_time": 0.54974844455719, "loss": 0.23191624581813813, "time": 0.688976240158081, "epoch": 4, "memory": 6318, "step": 15533} +{"lr": 7.153298662463338e-05, "data_time": 0.6540096759796142, "loss": 0.23380161374807357, "time": 0.7934549570083618, "epoch": 4, "memory": 6318, "step": 15633} +{"lr": 7.121267098518291e-05, "data_time": 0.6486236810684204, "loss": 0.23264795392751694, "time": 0.7888083457946777, "epoch": 4, "memory": 6318, "step": 15733} +{"lr": 7.089140854791618e-05, "data_time": 0.9109420537948608, "loss": 0.2285044640302658, "time": 1.0505372047424317, "epoch": 4, "memory": 6318, "step": 15833} +{"lr": 7.056921807412025e-05, "data_time": 0.614934754371643, "loss": 0.25067051500082016, "time": 0.7553423166275024, "epoch": 4, "memory": 6318, "step": 15933} +{"lr": 7.024611837927802e-05, "data_time": 0.8851374626159668, "loss": 0.2302881136536598, "time": 1.0263184309005737, "epoch": 4, "memory": 6318, "step": 16033} +{"lr": 6.992212833196998e-05, "data_time": 0.6721847295761109, "loss": 0.22763000726699828, "time": 0.8127633094787597, "epoch": 4, "memory": 6318, "step": 16133} +{"lr": 6.959726685277174e-05, "data_time": 0.674868130683899, "loss": 0.238856540620327, "time": 0.8151495218276977, "epoch": 4, "memory": 6318, "step": 16233} +{"lr": 6.927155291314949e-05, "data_time": 0.6280160665512085, "loss": 0.2315930336713791, "time": 0.7679660320281982, "epoch": 4, "memory": 6318, "step": 16333} +{"lr": 6.894500553435204e-05, "data_time": 0.6877367496490479, "loss": 0.24604939222335814, "time": 0.8278838634490967, "epoch": 4, "memory": 6318, "step": 16433} +{"accuracy/top1": 84.82862854003906, "data_time": 0.3857512831687927, "time": 0.47472933928171795, "step": 4} +{"lr": 6.858158509719903e-05, "data_time": 0.5680527448654175, "loss": 0.24009820967912673, "time": 0.7082586765289307, "epoch": 5, "memory": 6318, "step": 16544} +{"lr": 6.825334178880171e-05, "data_time": 0.6665577173233033, "loss": 0.2341777890920639, "time": 0.8059651851654053, "epoch": 5, "memory": 6318, "step": 16644} +{"lr": 6.792432450336645e-05, "data_time": 0.6335298299789429, "loss": 0.23208148777484894, "time": 0.7742976188659668, "epoch": 5, "memory": 6318, "step": 16744} +{"lr": 6.759455245505286e-05, "data_time": 0.588647437095642, "loss": 0.23389839380979538, "time": 0.7285048723220825, "epoch": 5, "memory": 6318, "step": 16844} +{"lr": 6.726404490209753e-05, "data_time": 0.615578556060791, "loss": 0.23718324303627014, "time": 0.7584915161132812, "epoch": 5, "memory": 6318, "step": 16944} +{"lr": 6.693282114568972e-05, "data_time": 0.7293593645095825, "loss": 0.25018548518419265, "time": 0.8711184740066529, "epoch": 5, "memory": 6318, "step": 17044} +{"lr": 6.660090052884388e-05, "data_time": 0.6184142351150512, "loss": 0.23349623531103134, "time": 0.7589134693145752, "epoch": 5, "memory": 6318, "step": 17144} +{"lr": 6.626830243527016e-05, "data_time": 0.6342181921005249, "loss": 0.22983630746603012, "time": 0.7739503622055054, "epoch": 5, "memory": 6318, "step": 17244} +{"lr": 6.593504628824236e-05, "data_time": 0.7950935125350952, "loss": 0.24304037392139435, "time": 0.9357823133468628, "epoch": 5, "memory": 6318, "step": 17344} +{"lr": 6.560115154946387e-05, "data_time": 0.6720767736434936, "loss": 0.2416984662413597, "time": 0.8131884336471558, "epoch": 5, "memory": 6318, "step": 17444} +{"lr": 6.526663771793081e-05, "data_time": 0.6429623365402222, "loss": 0.23228186815977098, "time": 0.7892422437667846, "epoch": 5, "memory": 6318, "step": 17544} +{"lr": 6.493152432879363e-05, "data_time": 0.6183753252029419, "loss": 0.22265356332063674, "time": 0.7581408977508545, "epoch": 5, "memory": 6318, "step": 17644} +{"lr": 6.459583095221596e-05, "data_time": 0.5958486080169678, "loss": 0.23214069306850432, "time": 0.7560694456100464, "epoch": 5, "memory": 6318, "step": 17744} +{"lr": 6.425957719223197e-05, "data_time": 0.6039879560470581, "loss": 0.2288201406598091, "time": 0.7443809509277344, "epoch": 5, "memory": 6318, "step": 17844} +{"lr": 6.392278268560153e-05, "data_time": 0.6037734031677247, "loss": 0.2102500855922699, "time": 0.7444466590881348, "epoch": 5, "memory": 6318, "step": 17944} +{"lr": 6.358546710066316e-05, "data_time": 0.6427554130554199, "loss": 0.2321569338440895, "time": 0.7831363439559936, "epoch": 5, "memory": 6318, "step": 18044} +{"lr": 6.324765013618573e-05, "data_time": 0.6907007694244385, "loss": 0.22775797545909882, "time": 0.830433440208435, "epoch": 5, "memory": 6318, "step": 18144} +{"lr": 6.290935152021818e-05, "data_time": 0.6373366832733154, "loss": 0.23793013393878937, "time": 0.777033805847168, "epoch": 5, "memory": 6318, "step": 18244} +{"lr": 6.2570591008937e-05, "data_time": 0.6889883279800415, "loss": 0.24342462569475173, "time": 0.829011344909668, "epoch": 5, "memory": 6318, "step": 18344} +{"lr": 6.223138838549294e-05, "data_time": 0.6115456819534302, "loss": 0.22704339325428008, "time": 0.7518725395202637, "epoch": 5, "memory": 6318, "step": 18444} +{"lr": 6.18917634588553e-05, "data_time": 0.6836014986038208, "loss": 0.2423887297511101, "time": 0.824725341796875, "epoch": 5, "memory": 6318, "step": 18544} +{"lr": 6.155173606265555e-05, "data_time": 0.6356582641601562, "loss": 0.2345250889658928, "time": 0.7764936447143554, "epoch": 5, "memory": 6318, "step": 18644} +{"lr": 6.121132605402843e-05, "data_time": 0.7121028900146484, "loss": 0.2327351987361908, "time": 0.8519127845764161, "epoch": 5, "memory": 6318, "step": 18744} +{"lr": 6.087055331245287e-05, "data_time": 0.6103345394134522, "loss": 0.243044975399971, "time": 0.7622027635574341, "epoch": 5, "memory": 6318, "step": 18844} +{"lr": 6.05294377385911e-05, "data_time": 0.6303061723709107, "loss": 0.2202696442604065, "time": 0.770358681678772, "epoch": 5, "memory": 6318, "step": 18944} +{"lr": 6.0187999253126355e-05, "data_time": 0.6624517917633057, "loss": 0.21790529787540436, "time": 0.8022862672805786, "epoch": 5, "memory": 6318, "step": 19044} +{"lr": 5.9846257795599195e-05, "data_time": 0.6217154026031494, "loss": 0.22332927882671355, "time": 0.7627980232238769, "epoch": 5, "memory": 6318, "step": 19144} +{"lr": 5.950423332324355e-05, "data_time": 0.657962965965271, "loss": 0.23062596917152406, "time": 0.7977495431900025, "epoch": 5, "memory": 6318, "step": 19244} +{"lr": 5.9161945809820855e-05, "data_time": 0.6836307287216187, "loss": 0.2321133390069008, "time": 0.8246405601501465, "epoch": 5, "memory": 6318, "step": 19344} +{"lr": 5.8819415244453975e-05, "data_time": 0.61401846408844, "loss": 0.23368024080991745, "time": 0.7551644802093506, "epoch": 5, "memory": 6318, "step": 19444} +{"lr": 5.8476661630459405e-05, "data_time": 0.5882413148880005, "loss": 0.23067801892757417, "time": 0.7281119585037231, "epoch": 5, "memory": 6318, "step": 19544} +{"lr": 5.81337049841796e-05, "data_time": 0.6396611928939819, "loss": 0.23544860631227493, "time": 0.7821515083312989, "epoch": 5, "memory": 6318, "step": 19644} +{"lr": 5.779056533381373e-05, "data_time": 0.6922419548034668, "loss": 0.23260191977024078, "time": 0.8323213338851929, "epoch": 5, "memory": 6318, "step": 19744} +{"lr": 5.744726271824817e-05, "data_time": 0.837563443183899, "loss": 0.23904459774494172, "time": 0.9835986375808716, "epoch": 5, "memory": 6318, "step": 19844} +{"lr": 5.710381718588622e-05, "data_time": 0.6601776361465455, "loss": 0.2379686340689659, "time": 0.8006458520889282, "epoch": 5, "memory": 6318, "step": 19944} +{"lr": 5.676024879347741e-05, "data_time": 0.6204306125640869, "loss": 0.24066135734319688, "time": 0.7595683336257935, "epoch": 5, "memory": 6318, "step": 20044} +{"lr": 5.6416577604945966e-05, "data_time": 0.8849411010742188, "loss": 0.2382201671600342, "time": 1.026048755645752, "epoch": 5, "memory": 6318, "step": 20144} +{"lr": 5.6072823690219385e-05, "data_time": 0.677523922920227, "loss": 0.22010047286748885, "time": 0.8182397842407226, "epoch": 5, "memory": 6318, "step": 20244} +{"lr": 5.572900712405624e-05, "data_time": 0.6762327194213867, "loss": 0.237260402739048, "time": 0.8178808212280273, "epoch": 5, "memory": 6318, "step": 20344} +{"lr": 5.538514798487397e-05, "data_time": 0.6832957983016967, "loss": 0.23247854113578797, "time": 0.8243798732757568, "epoch": 5, "memory": 6318, "step": 20444} +{"lr": 5.5041266353576e-05, "data_time": 0.6764598846435547, "loss": 0.23345525115728377, "time": 0.8193856477737427, "epoch": 5, "memory": 6318, "step": 20544} +{"accuracy/top1": 88.0062255859375, "data_time": 0.40955922404925027, "time": 0.4967768351236979, "step": 5} +{"lr": 5.4659555783035384e-05, "data_time": 0.5640170097351074, "loss": 0.23523711860179902, "time": 0.8417809724807739, "epoch": 6, "memory": 6318, "step": 20655} +{"lr": 5.431569258426336e-05, "data_time": 0.7865960359573364, "loss": 0.2411901906132698, "time": 0.9359241247177124, "epoch": 6, "memory": 6318, "step": 20755} +{"lr": 5.397186934810864e-05, "data_time": 0.6889936923980713, "loss": 0.24351299554109573, "time": 0.829921817779541, "epoch": 6, "memory": 6318, "step": 20855} +{"lr": 5.362810615337815e-05, "data_time": 0.6160107612609863, "loss": 0.22347162514925004, "time": 0.7581508874893188, "epoch": 6, "memory": 6318, "step": 20955} +{"lr": 5.328442307537233e-05, "data_time": 0.7360903739929199, "loss": 0.23152901828289033, "time": 0.8770427703857422, "epoch": 6, "memory": 6318, "step": 21055} +{"lr": 5.294084018471303e-05, "data_time": 0.7281834840774536, "loss": 0.23047693073749542, "time": 0.8681220293045044, "epoch": 6, "memory": 6318, "step": 21155} +{"lr": 5.259737754617132e-05, "data_time": 0.6397636890411377, "loss": 0.2284434214234352, "time": 0.7799125432968139, "epoch": 6, "memory": 6318, "step": 21255} +{"lr": 5.2254055217495525e-05, "data_time": 0.6479471921920776, "loss": 0.23046401143074036, "time": 0.7879778861999511, "epoch": 6, "memory": 6318, "step": 21355} +{"lr": 5.1910893248240283e-05, "data_time": 0.6063809394836426, "loss": 0.23219931572675706, "time": 0.7710274934768677, "epoch": 6, "memory": 6318, "step": 21455} +{"lr": 5.156791167859532e-05, "data_time": 0.6770941495895386, "loss": 0.23874398320913315, "time": 0.8168701887130737, "epoch": 6, "memory": 6318, "step": 21555} +{"lr": 5.122513053821533e-05, "data_time": 0.6383856773376465, "loss": 0.22835891842842101, "time": 0.7785636186599731, "epoch": 6, "memory": 6318, "step": 21655} +{"lr": 5.0882569845050173e-05, "data_time": 0.6357599973678589, "loss": 0.22527429312467576, "time": 0.7761294603347778, "epoch": 6, "memory": 6318, "step": 21755} +{"lr": 5.054024960417592e-05, "data_time": 0.6140935897827149, "loss": 0.23473118543624877, "time": 0.7549921989440918, "epoch": 6, "memory": 6318, "step": 21855} +{"lr": 5.0198189806626494e-05, "data_time": 0.7070087909698486, "loss": 0.22903325110673906, "time": 0.84640531539917, "epoch": 6, "memory": 6318, "step": 21955} +{"lr": 4.985641042822647e-05, "data_time": 0.6718303918838501, "loss": 0.2397131696343422, "time": 0.8123040437698364, "epoch": 6, "memory": 6318, "step": 22055} +{"lr": 4.951493142842412e-05, "data_time": 0.6344693899154663, "loss": 0.2116445705294609, "time": 0.7747231245040893, "epoch": 6, "memory": 6318, "step": 22155} +{"lr": 4.917377274912603e-05, "data_time": 0.6452999591827393, "loss": 0.22026612162590026, "time": 0.7876631021499634, "epoch": 6, "memory": 6318, "step": 22255} +{"lr": 4.88329543135327e-05, "data_time": 0.7558466672897339, "loss": 0.2256220430135727, "time": 0.901628565788269, "epoch": 6, "memory": 6318, "step": 22355} +{"lr": 4.849249602497467e-05, "data_time": 0.7455522537231445, "loss": 0.20942132472991942, "time": 0.8870739459991455, "epoch": 6, "memory": 6318, "step": 22455} +{"lr": 4.8152417765750456e-05, "data_time": 0.7094234466552735, "loss": 0.23402302861213684, "time": 0.849983024597168, "epoch": 6, "memory": 6318, "step": 22555} +{"lr": 4.7812739395965355e-05, "data_time": 0.7074878692626954, "loss": 0.25134078711271285, "time": 0.848683524131775, "epoch": 6, "memory": 6318, "step": 22655} +{"lr": 4.74734807523717e-05, "data_time": 0.6698830127716064, "loss": 0.22743172198534012, "time": 0.817116093635559, "epoch": 6, "memory": 6318, "step": 22755} +{"lr": 4.7134661647210416e-05, "data_time": 0.6963460922241211, "loss": 0.2213045433163643, "time": 0.8369168519973755, "epoch": 6, "memory": 6318, "step": 22855} +{"lr": 4.679630186705388e-05, "data_time": 0.7061920642852784, "loss": 0.22865137755870818, "time": 0.8540581703186035, "epoch": 6, "memory": 6318, "step": 22955} +{"lr": 4.6458421171650644e-05, "data_time": 0.6461435317993164, "loss": 0.2223941832780838, "time": 0.7868811368942261, "epoch": 6, "memory": 6318, "step": 23055} +{"lr": 4.6121039292771295e-05, "data_time": 0.6812747955322266, "loss": 0.2251240998506546, "time": 0.8205606698989868, "epoch": 6, "memory": 6318, "step": 23155} +{"lr": 4.5784175933056197e-05, "data_time": 0.6692547798156738, "loss": 0.22888384461402894, "time": 0.8101490259170532, "epoch": 6, "memory": 6318, "step": 23255} +{"lr": 4.544785076486494e-05, "data_time": 0.6655930042266845, "loss": 0.22767900228500365, "time": 0.8060162305831909, "epoch": 6, "memory": 6318, "step": 23355} +{"lr": 4.5112083429127475e-05, "data_time": 0.7369659900665283, "loss": 0.22401324063539504, "time": 0.8776190996170044, "epoch": 6, "memory": 6318, "step": 23455} +{"lr": 4.477689353419706e-05, "data_time": 0.6090786695480347, "loss": 0.21689967960119247, "time": 0.7513005256652832, "epoch": 6, "memory": 6318, "step": 23555} +{"lr": 4.4442300654705266e-05, "data_time": 0.6490351438522339, "loss": 0.23096476197242738, "time": 0.7893428802490234, "epoch": 6, "memory": 6318, "step": 23655} +{"lr": 4.410832433041866e-05, "data_time": 0.6412917852401734, "loss": 0.21838558614253997, "time": 0.7809933662414551, "epoch": 6, "memory": 6318, "step": 23755} +{"lr": 4.377498406509792e-05, "data_time": 0.7388076543807983, "loss": 0.21265769451856614, "time": 0.8793262958526611, "epoch": 6, "memory": 6318, "step": 23855} +{"lr": 4.344229932535875e-05, "data_time": 0.6426203966140747, "loss": 0.2235260084271431, "time": 0.7837935209274292, "epoch": 6, "memory": 6318, "step": 23955} +{"lr": 4.3110289539535115e-05, "data_time": 0.6267694711685181, "loss": 0.23553937673568726, "time": 0.7675327062606812, "epoch": 6, "memory": 6318, "step": 24055} +{"lr": 4.2778974096544436e-05, "data_time": 0.6346272230148315, "loss": 0.2247397780418396, "time": 0.7754977941513062, "epoch": 6, "memory": 6318, "step": 24155} +{"lr": 4.2448372344755544e-05, "data_time": 0.7075176477432251, "loss": 0.20807492583990098, "time": 0.8477226257324219, "epoch": 6, "memory": 6318, "step": 24255} +{"lr": 4.211850359085871e-05, "data_time": 0.6136820793151856, "loss": 0.20074809491634368, "time": 0.756170129776001, "epoch": 6, "memory": 6318, "step": 24355} +{"lr": 4.17893870987381e-05, "data_time": 0.652216625213623, "loss": 0.2138784110546112, "time": 0.7941282510757446, "epoch": 6, "memory": 6318, "step": 24455} +{"lr": 4.146104208834679e-05, "data_time": 0.6623796224594116, "loss": 0.21141144633293152, "time": 0.8031316280364991, "epoch": 6, "memory": 6318, "step": 24555} +{"lr": 4.113348773458442e-05, "data_time": 0.6592641592025756, "loss": 0.19716128557920456, "time": 0.80677809715271, "epoch": 6, "memory": 6318, "step": 24655} +{"accuracy/top1": 90.61781311035156, "data_time": 0.40377671718597413, "time": 0.49216635823249816, "step": 6} +{"lr": 4.0770851520835164e-05, "data_time": 0.7648333787918091, "loss": 0.22227001190185547, "time": 0.927307391166687, "epoch": 7, "memory": 6318, "step": 24766} +{"lr": 4.044502815742039e-05, "data_time": 0.620558500289917, "loss": 0.2190093293786049, "time": 0.7607215404510498, "epoch": 7, "memory": 6318, "step": 24866} +{"lr": 4.012005478445896e-05, "data_time": 0.6529367446899415, "loss": 0.23349482268095018, "time": 0.7941431522369384, "epoch": 7, "memory": 6318, "step": 24966} +{"lr": 3.9795950379951394e-05, "data_time": 0.624491548538208, "loss": 0.24427032321691514, "time": 0.764388918876648, "epoch": 7, "memory": 6318, "step": 25066} +{"lr": 3.947273387115169e-05, "data_time": 0.7141632080078125, "loss": 0.22110608369112014, "time": 0.8562009572982788, "epoch": 7, "memory": 6318, "step": 25166} +{"lr": 3.9150424133462044e-05, "data_time": 0.686302661895752, "loss": 0.20895550549030303, "time": 0.8286733150482177, "epoch": 7, "memory": 6318, "step": 25266} +{"lr": 3.882903998933037e-05, "data_time": 0.6973209619522095, "loss": 0.23273418247699737, "time": 0.8395562648773194, "epoch": 7, "memory": 6318, "step": 25366} +{"lr": 3.8508600207151264e-05, "data_time": 0.6170834541320801, "loss": 0.21740475445985794, "time": 0.7590098142623901, "epoch": 7, "memory": 6318, "step": 25466} +{"lr": 3.818912350016978e-05, "data_time": 0.6640692234039307, "loss": 0.2343267560005188, "time": 0.807293438911438, "epoch": 7, "memory": 6318, "step": 25566} +{"lr": 3.787062852538883e-05, "data_time": 0.6382928371429444, "loss": 0.22212339490652083, "time": 0.779767370223999, "epoch": 7, "memory": 6318, "step": 25666} +{"lr": 3.7553133882479346e-05, "data_time": 0.6732906579971314, "loss": 0.22489983290433885, "time": 0.8138845443725586, "epoch": 7, "memory": 6318, "step": 25766} +{"lr": 3.723665811269437e-05, "data_time": 0.6524448156356811, "loss": 0.22993300557136537, "time": 0.7923453330993653, "epoch": 7, "memory": 6318, "step": 25866} +{"lr": 3.6921219697786176e-05, "data_time": 0.5943305969238282, "loss": 0.21163887828588485, "time": 0.7348302364349365, "epoch": 7, "memory": 6318, "step": 25966} +{"lr": 3.660683705892677e-05, "data_time": 0.6365231990814209, "loss": 0.2309290960431099, "time": 0.7787399530410767, "epoch": 7, "memory": 6318, "step": 26066} +{"lr": 3.6293528555632435e-05, "data_time": 0.6208684206008911, "loss": 0.23909600526094438, "time": 0.7614422798156738, "epoch": 7, "memory": 6318, "step": 26166} +{"lr": 3.598131248469134e-05, "data_time": 0.6362852573394775, "loss": 0.22502311319112778, "time": 0.7762999296188354, "epoch": 7, "memory": 6318, "step": 26266} +{"lr": 3.567020707909507e-05, "data_time": 0.6458447217941284, "loss": 0.20648026764392852, "time": 0.7862965345382691, "epoch": 7, "memory": 6318, "step": 26366} +{"lr": 3.5360230506973935e-05, "data_time": 0.6337474107742309, "loss": 0.21238310039043426, "time": 0.7749159574508667, "epoch": 7, "memory": 6318, "step": 26466} +{"lr": 3.5051400870535877e-05, "data_time": 0.6258433341979981, "loss": 0.2131961017847061, "time": 0.7664084196090698, "epoch": 7, "memory": 6318, "step": 26566} +{"lr": 3.47437362050094e-05, "data_time": 0.7146966218948364, "loss": 0.22229647785425186, "time": 0.8566596508026123, "epoch": 7, "memory": 6318, "step": 26666} +{"lr": 3.443725447759023e-05, "data_time": 0.6348914623260498, "loss": 0.22821670323610305, "time": 0.7756235599517822, "epoch": 7, "memory": 6318, "step": 26766} +{"lr": 3.413197358639224e-05, "data_time": 0.665582799911499, "loss": 0.2172778755426407, "time": 0.8062179327011109, "epoch": 7, "memory": 6318, "step": 26866} +{"lr": 3.382791135940194e-05, "data_time": 0.5825109958648682, "loss": 0.22656544148921967, "time": 0.722493028640747, "epoch": 7, "memory": 6318, "step": 26966} +{"lr": 3.35250855534377e-05, "data_time": 0.6492086887359619, "loss": 0.22972712516784669, "time": 0.7958221197128296, "epoch": 7, "memory": 6318, "step": 27066} +{"lr": 3.3223513853112476e-05, "data_time": 0.622047233581543, "loss": 0.20316445976495742, "time": 0.7735379219055176, "epoch": 7, "memory": 6318, "step": 27166} +{"lr": 3.292321386980115e-05, "data_time": 0.6232569217681885, "loss": 0.21836679428815842, "time": 0.7632628202438354, "epoch": 7, "memory": 6318, "step": 27266} +{"lr": 3.262420314061217e-05, "data_time": 0.5884154081344605, "loss": 0.2244155913591385, "time": 0.7284396409988403, "epoch": 7, "memory": 6318, "step": 27366} +{"lr": 3.2326499127363195e-05, "data_time": 0.5815781116485595, "loss": 0.21355385780334474, "time": 0.7286983013153077, "epoch": 7, "memory": 6318, "step": 27466} +{"lr": 3.2030119215561613e-05, "data_time": 0.630339789390564, "loss": 0.23119299113750458, "time": 0.7710340023040771, "epoch": 7, "memory": 6318, "step": 27566} +{"lr": 3.173508071338896e-05, "data_time": 0.647737455368042, "loss": 0.20573912113904952, "time": 0.7886793375015259, "epoch": 7, "memory": 6318, "step": 27666} +{"lr": 3.144140085069037e-05, "data_time": 0.5622187614440918, "loss": 0.20731066912412643, "time": 0.7015899658203125, "epoch": 7, "memory": 6318, "step": 27766} +{"lr": 3.114909677796814e-05, "data_time": 0.5976841926574707, "loss": 0.20905494689941406, "time": 0.741986083984375, "epoch": 7, "memory": 6318, "step": 27866} +{"lr": 3.085818556538059e-05, "data_time": 0.57134530544281, "loss": 0.20247323513031007, "time": 0.712349534034729, "epoch": 7, "memory": 6318, "step": 27966} +{"lr": 3.056868420174455e-05, "data_time": 0.6110909223556519, "loss": 0.21762798875570297, "time": 0.750453519821167, "epoch": 7, "memory": 6318, "step": 28066} +{"lr": 3.028060959354392e-05, "data_time": 0.6412607431411743, "loss": 0.2117123380303383, "time": 0.7819455146789551, "epoch": 7, "memory": 6318, "step": 28166} +{"lr": 2.9993978563941876e-05, "data_time": 0.5889158010482788, "loss": 0.2219421848654747, "time": 0.729128384590149, "epoch": 7, "memory": 6318, "step": 28266} +{"lr": 2.9708807851798587e-05, "data_time": 0.6377652168273926, "loss": 0.2252564862370491, "time": 0.7775874614715577, "epoch": 7, "memory": 6318, "step": 28366} +{"lr": 2.9425114110693717e-05, "data_time": 0.5933243751525878, "loss": 0.22750522345304489, "time": 0.7334260940551758, "epoch": 7, "memory": 6318, "step": 28466} +{"lr": 2.9142913907953844e-05, "data_time": 0.6078749895095825, "loss": 0.20678254663944245, "time": 0.748738431930542, "epoch": 7, "memory": 6318, "step": 28566} +{"lr": 2.8862223723684912e-05, "data_time": 0.6473680734634399, "loss": 0.22438707798719407, "time": 0.7887472629547119, "epoch": 7, "memory": 6318, "step": 28666} +{"lr": 2.8583059949809854e-05, "data_time": 0.6076606035232544, "loss": 0.2237881377339363, "time": 0.7471909284591675, "epoch": 7, "memory": 6318, "step": 28766} +{"accuracy/top1": 91.9087142944336, "data_time": 0.400511501232783, "time": 0.48955397804578143, "step": 7} +{"lr": 2.827499545208893e-05, "data_time": 0.6176312685012817, "loss": 0.21436824053525924, "time": 0.7585988759994506, "epoch": 8, "memory": 6318, "step": 28877} +{"lr": 2.7999105784987406e-05, "data_time": 0.6748634815216065, "loss": 0.22476212233304976, "time": 0.8150183200836182, "epoch": 8, "memory": 6318, "step": 28977} +{"lr": 2.772479293318711e-05, "data_time": 0.5957891464233398, "loss": 0.21923489421606063, "time": 0.7365003108978272, "epoch": 8, "memory": 6318, "step": 29077} +{"lr": 2.7452072916183603e-05, "data_time": 0.6443199872970581, "loss": 0.21840370297431946, "time": 0.7859854698181152, "epoch": 8, "memory": 6318, "step": 29177} +{"lr": 2.7180961660452883e-05, "data_time": 0.5866249084472657, "loss": 0.23032589703798295, "time": 0.7289427280426025, "epoch": 8, "memory": 6318, "step": 29277} +{"lr": 2.6911474998521622e-05, "data_time": 0.6299821853637695, "loss": 0.21991163194179536, "time": 0.7699019432067871, "epoch": 8, "memory": 6318, "step": 29377} +{"lr": 2.664362866804239e-05, "data_time": 0.6139799594879151, "loss": 0.21983575075864792, "time": 0.7538867473602295, "epoch": 8, "memory": 6318, "step": 29477} +{"lr": 2.637743831087459e-05, "data_time": 0.6002455472946167, "loss": 0.23171020299196243, "time": 0.7429261684417725, "epoch": 8, "memory": 6318, "step": 29577} +{"lr": 2.6112919472171114e-05, "data_time": 0.6951230287551879, "loss": 0.21728023290634155, "time": 0.8370468854904175, "epoch": 8, "memory": 6318, "step": 29677} +{"lr": 2.5850087599470434e-05, "data_time": 0.6004352331161499, "loss": 0.20709521770477296, "time": 0.7416739702224732, "epoch": 8, "memory": 6318, "step": 29777} +{"lr": 2.5588958041794503e-05, "data_time": 0.7258542537689209, "loss": 0.2137942522764206, "time": 0.8688105583190918, "epoch": 8, "memory": 6318, "step": 29877} +{"lr": 2.5329546048752453e-05, "data_time": 0.5389542818069458, "loss": 0.2041707828640938, "time": 0.68125319480896, "epoch": 8, "memory": 6318, "step": 29977} +{"lr": 2.5071866769649806e-05, "data_time": 0.591166615486145, "loss": 0.23005905896425247, "time": 0.7317116975784301, "epoch": 8, "memory": 6318, "step": 30077} +{"lr": 2.4815935252604222e-05, "data_time": 0.6030442476272583, "loss": 0.2234228640794754, "time": 0.7429580450057983, "epoch": 8, "memory": 6318, "step": 30177} +{"lr": 2.4561766443666384e-05, "data_time": 0.6336564779281616, "loss": 0.22076471745967866, "time": 0.7747917413711548, "epoch": 8, "memory": 6318, "step": 30277} +{"lr": 2.4309375185947164e-05, "data_time": 0.6094679117202759, "loss": 0.22254387885332108, "time": 0.7498175859451294, "epoch": 8, "memory": 6318, "step": 30377} +{"lr": 2.40587762187509e-05, "data_time": 0.6754100084304809, "loss": 0.22178202420473098, "time": 0.8154917001724243, "epoch": 8, "memory": 6318, "step": 30477} +{"lr": 2.3809984176714634e-05, "data_time": 0.573855972290039, "loss": 0.2264096185564995, "time": 0.7130426406860352, "epoch": 8, "memory": 6318, "step": 30577} +{"lr": 2.3563013588953406e-05, "data_time": 0.6098098993301392, "loss": 0.2345052808523178, "time": 0.7513066291809082, "epoch": 8, "memory": 6318, "step": 30677} +{"lr": 2.331787887821182e-05, "data_time": 0.5980265617370606, "loss": 0.22210220098495484, "time": 0.7409123182296753, "epoch": 8, "memory": 6318, "step": 30777} +{"lr": 2.307459436002179e-05, "data_time": 0.6060364723205567, "loss": 0.22102549225091933, "time": 0.7459308624267578, "epoch": 8, "memory": 6318, "step": 30877} +{"lr": 2.2833174241866527e-05, "data_time": 0.6350964546203614, "loss": 0.22628044486045837, "time": 0.7767293930053711, "epoch": 8, "memory": 6318, "step": 30977} +{"lr": 2.259363262235086e-05, "data_time": 0.6353445768356323, "loss": 0.20365117043256759, "time": 0.7762448072433472, "epoch": 8, "memory": 6318, "step": 31077} +{"lr": 2.2355983490377735e-05, "data_time": 0.6199627637863159, "loss": 0.2064829558134079, "time": 0.7605207681655883, "epoch": 8, "memory": 6318, "step": 31177} +{"lr": 2.2120240724331546e-05, "data_time": 0.680437421798706, "loss": 0.2050203114748001, "time": 0.8336479902267456, "epoch": 8, "memory": 6318, "step": 31277} +{"lr": 2.1886418091267445e-05, "data_time": 0.5961122035980224, "loss": 0.2049966424703598, "time": 0.7368704319000244, "epoch": 8, "memory": 6318, "step": 31377} +{"lr": 2.1654529246107496e-05, "data_time": 0.6269190549850464, "loss": 0.20520494729280472, "time": 0.7749230623245239, "epoch": 8, "memory": 6318, "step": 31477} +{"lr": 2.1424587730843184e-05, "data_time": 0.6658232927322387, "loss": 0.1943259358406067, "time": 0.8055380582809448, "epoch": 8, "memory": 6318, "step": 31577} +{"lr": 2.1196606973744593e-05, "data_time": 0.6407031774520874, "loss": 0.2201676696538925, "time": 0.7820532083511352, "epoch": 8, "memory": 6318, "step": 31677} +{"lr": 2.0970600288576148e-05, "data_time": 0.6111618041992187, "loss": 0.21097415089607238, "time": 0.7509358882904053, "epoch": 8, "memory": 6318, "step": 31777} +{"lr": 2.0746580873819274e-05, "data_time": 0.6296913146972656, "loss": 0.21080616861581802, "time": 0.770757794380188, "epoch": 8, "memory": 6318, "step": 31877} +{"lr": 2.0524561811901426e-05, "data_time": 0.6054296016693115, "loss": 0.20179010927677155, "time": 0.745720100402832, "epoch": 8, "memory": 6318, "step": 31977} +{"lr": 2.0304556068432268e-05, "data_time": 0.6531343698501587, "loss": 0.20696854442358018, "time": 0.7943596124649048, "epoch": 8, "memory": 6318, "step": 32077} +{"lr": 2.0086576491446333e-05, "data_time": 0.6205193281173706, "loss": 0.20936913937330245, "time": 0.7608472824096679, "epoch": 8, "memory": 6318, "step": 32177} +{"lr": 1.987063581065291e-05, "data_time": 0.5727409362792969, "loss": 0.22898492515087127, "time": 0.7134162664413453, "epoch": 8, "memory": 6318, "step": 32277} +{"lr": 1.965674663669245e-05, "data_time": 0.603482699394226, "loss": 0.20578554421663284, "time": 0.7432650804519654, "epoch": 8, "memory": 6318, "step": 32377} +{"lr": 1.9444921460400247e-05, "data_time": 0.5591931104660034, "loss": 0.23926617801189423, "time": 0.7021947622299194, "epoch": 8, "memory": 6318, "step": 32477} +{"lr": 1.923517265207697e-05, "data_time": 0.5848503351211548, "loss": 0.21970323771238326, "time": 0.7247686624526978, "epoch": 8, "memory": 6318, "step": 32577} +{"lr": 1.9027512460766204e-05, "data_time": 0.7181018590927124, "loss": 0.21240825355052947, "time": 0.8582144021987915, "epoch": 8, "memory": 6318, "step": 32677} +{"lr": 1.8821953013539136e-05, "data_time": 0.634683895111084, "loss": 0.20308539122343064, "time": 0.7843214511871338, "epoch": 8, "memory": 6318, "step": 32777} +{"lr": 1.8618506314786433e-05, "data_time": 0.5956550598144531, "loss": 0.19921929389238358, "time": 0.7360635042190552, "epoch": 8, "memory": 6318, "step": 32877} +{"accuracy/top1": 92.75276184082031, "data_time": 0.39636743664741514, "time": 0.48426974217096963, "step": 8} +{"lr": 1.8395169032542028e-05, "data_time": 0.6216209888458252, "loss": 0.20786677449941635, "time": 0.7618060827255249, "epoch": 9, "memory": 6318, "step": 32988} +{"lr": 1.819621906460573e-05, "data_time": 0.6423932075500488, "loss": 0.20277125090360643, "time": 0.7839249849319458, "epoch": 9, "memory": 6318, "step": 33088} +{"lr": 1.799941838715077e-05, "data_time": 0.628791069984436, "loss": 0.20579521358013153, "time": 0.7687147617340088, "epoch": 9, "memory": 6318, "step": 33188} +{"lr": 1.7804778493067625e-05, "data_time": 0.6096827507019043, "loss": 0.20695388615131377, "time": 0.7503035306930542, "epoch": 9, "memory": 6318, "step": 33288} +{"lr": 1.761231074905996e-05, "data_time": 0.570347261428833, "loss": 0.2184254989027977, "time": 0.7099888563156128, "epoch": 9, "memory": 6318, "step": 33388} +{"lr": 1.742202639498083e-05, "data_time": 0.6435333251953125, "loss": 0.20907920897006987, "time": 0.7838011980056763, "epoch": 9, "memory": 6318, "step": 33488} +{"lr": 1.7233936543176347e-05, "data_time": 0.607920479774475, "loss": 0.24120692610740663, "time": 0.7484342098236084, "epoch": 9, "memory": 6318, "step": 33588} +{"lr": 1.704805217783665e-05, "data_time": 0.6213541507720948, "loss": 0.20577476918697357, "time": 0.7618390083312988, "epoch": 9, "memory": 6318, "step": 33688} +{"lr": 1.68643841543545e-05, "data_time": 0.6261132478713989, "loss": 0.22067376375198364, "time": 0.7667516708374024, "epoch": 9, "memory": 6318, "step": 33788} +{"lr": 1.6682943198691327e-05, "data_time": 0.5971660375595093, "loss": 0.20701639652252196, "time": 0.7378833055496216, "epoch": 9, "memory": 6318, "step": 33888} +{"lr": 1.6503739906750822e-05, "data_time": 0.655132532119751, "loss": 0.20590871721506118, "time": 0.7950443029403687, "epoch": 9, "memory": 6318, "step": 33988} +{"lr": 1.6326784743760187e-05, "data_time": 0.9051164388656616, "loss": 0.19997305124998094, "time": 1.0440760612487794, "epoch": 9, "memory": 6318, "step": 34088} +{"lr": 1.6152088043659013e-05, "data_time": 0.6061965942382812, "loss": 0.20650468170642852, "time": 0.7460791587829589, "epoch": 9, "memory": 6318, "step": 34188} +{"lr": 1.59796600084957e-05, "data_time": 0.6238481998443604, "loss": 0.21854153722524644, "time": 0.7663270473480225, "epoch": 9, "memory": 6318, "step": 34288} +{"lr": 1.5809510707831687e-05, "data_time": 0.5700688362121582, "loss": 0.2022887408733368, "time": 0.7177820920944213, "epoch": 9, "memory": 6318, "step": 34388} +{"lr": 1.5641650078153473e-05, "data_time": 0.5865361452102661, "loss": 0.21612299233675003, "time": 0.7258360862731934, "epoch": 9, "memory": 6318, "step": 34488} +{"lr": 1.547608792229235e-05, "data_time": 0.617069149017334, "loss": 0.21610855162143708, "time": 0.7565063953399658, "epoch": 9, "memory": 6318, "step": 34588} +{"lr": 1.5312833908851794e-05, "data_time": 0.5972658872604371, "loss": 0.213951775431633, "time": 0.7376068115234375, "epoch": 9, "memory": 6318, "step": 34688} +{"lr": 1.5151897571642958e-05, "data_time": 0.6320751428604126, "loss": 0.22815759629011154, "time": 0.7721969366073609, "epoch": 9, "memory": 6318, "step": 34788} +{"lr": 1.4993288309127877e-05, "data_time": 0.5628451347351074, "loss": 0.21320310980081558, "time": 0.7084277391433715, "epoch": 9, "memory": 6318, "step": 34888} +{"lr": 1.4837015383870593e-05, "data_time": 0.5949024200439453, "loss": 0.22500751316547393, "time": 0.7355358600616455, "epoch": 9, "memory": 6318, "step": 34988} +{"lr": 1.4683087921996264e-05, "data_time": 0.591160774230957, "loss": 0.21934388875961303, "time": 0.7318089008331299, "epoch": 9, "memory": 6318, "step": 35088} +{"lr": 1.453151491265819e-05, "data_time": 0.6077698707580567, "loss": 0.23041684776544571, "time": 0.7496355295181274, "epoch": 9, "memory": 6318, "step": 35188} +{"lr": 1.4382305207512847e-05, "data_time": 1.101082968711853, "loss": 0.22867132276296614, "time": 1.251568865776062, "epoch": 9, "memory": 6318, "step": 35288} +{"lr": 1.423546752020298e-05, "data_time": 0.6855479717254639, "loss": 0.2133968412876129, "time": 0.8279071569442749, "epoch": 9, "memory": 6318, "step": 35388} +{"lr": 1.4091010425848762e-05, "data_time": 0.6148358345031738, "loss": 0.20738937109708785, "time": 0.7561617374420166, "epoch": 9, "memory": 6318, "step": 35488} +{"lr": 1.3948942360546966e-05, "data_time": 0.5825257539749146, "loss": 0.22857085168361663, "time": 0.7231127262115479, "epoch": 9, "memory": 6318, "step": 35588} +{"lr": 1.3809271620878346e-05, "data_time": 0.6417038917541504, "loss": 0.1929815486073494, "time": 0.782297158241272, "epoch": 9, "memory": 6318, "step": 35688} +{"lr": 1.3672006363423091e-05, "data_time": 0.6662217140197754, "loss": 0.2063335046172142, "time": 0.8071076154708863, "epoch": 9, "memory": 6318, "step": 35788} +{"lr": 1.3537154604284566e-05, "data_time": 0.6649338245391846, "loss": 0.21065299808979035, "time": 0.811578345298767, "epoch": 9, "memory": 6318, "step": 35888} +{"lr": 1.3404724218621087e-05, "data_time": 0.6290776014328003, "loss": 0.23351238667964935, "time": 0.7699763774871826, "epoch": 9, "memory": 6318, "step": 35988} +{"lr": 1.3274722940186121e-05, "data_time": 0.6371079206466674, "loss": 0.2093411773443222, "time": 0.7780585765838623, "epoch": 9, "memory": 6318, "step": 36088} +{"lr": 1.3147158360876541e-05, "data_time": 0.6383229255676269, "loss": 0.21968120634555816, "time": 0.7800687074661254, "epoch": 9, "memory": 6318, "step": 36188} +{"lr": 1.3022037930289352e-05, "data_time": 0.6223465204238892, "loss": 0.20273312032222748, "time": 0.7635986328125, "epoch": 9, "memory": 6318, "step": 36288} +{"lr": 1.2899368955286627e-05, "data_time": 0.5478759050369263, "loss": 0.21345443278551102, "time": 0.6884966850280761, "epoch": 9, "memory": 6318, "step": 36388} +{"lr": 1.2779158599568751e-05, "data_time": 0.5945204973220826, "loss": 0.2066677615046501, "time": 0.7352986812591553, "epoch": 9, "memory": 6318, "step": 36488} +{"lr": 1.2661413883256134e-05, "data_time": 0.6183670043945313, "loss": 0.20916925817728044, "time": 0.7594605922698975, "epoch": 9, "memory": 6318, "step": 36588} +{"lr": 1.2546141682479203e-05, "data_time": 0.6569263219833374, "loss": 0.21101305931806563, "time": 0.7965500831604004, "epoch": 9, "memory": 6318, "step": 36688} +{"lr": 1.2433348728976846e-05, "data_time": 0.6889534711837768, "loss": 0.21142506301403047, "time": 0.8309194564819335, "epoch": 9, "memory": 6318, "step": 36788} +{"lr": 1.2323041609703344e-05, "data_time": 0.625512957572937, "loss": 0.21078932136297227, "time": 0.7670870304107666, "epoch": 9, "memory": 6318, "step": 36888} +{"lr": 1.2215226766443614e-05, "data_time": 0.6039667367935181, "loss": 0.21911162734031678, "time": 0.7465001583099365, "epoch": 9, "memory": 6318, "step": 36988} +{"accuracy/top1": 93.26911926269531, "data_time": 0.39996779163678486, "time": 0.48942285776138306, "step": 9} +{"lr": 1.209847850894437e-05, "data_time": 0.5794030666351319, "loss": 0.21785298883914947, "time": 0.7185676097869873, "epoch": 10, "memory": 6318, "step": 37099} +{"lr": 1.1995942849201055e-05, "data_time": 0.6175326347351074, "loss": 0.21775296032428743, "time": 0.758519983291626, "epoch": 10, "memory": 6318, "step": 37199} +{"lr": 1.189591856759183e-05, "data_time": 0.6041690826416015, "loss": 0.21178694516420365, "time": 0.7452110290527344, "epoch": 10, "memory": 6318, "step": 37299} +{"lr": 1.1798411505398016e-05, "data_time": 0.5669848680496216, "loss": 0.2084959164261818, "time": 0.7071744918823242, "epoch": 10, "memory": 6318, "step": 37399} +{"lr": 1.1703427356898678e-05, "data_time": 0.5915694236755371, "loss": 0.2231340527534485, "time": 0.7323443651199341, "epoch": 10, "memory": 6318, "step": 37499} +{"lr": 1.1610971669038223e-05, "data_time": 0.6481956958770752, "loss": 0.20284702330827714, "time": 0.7913770198822021, "epoch": 10, "memory": 6318, "step": 37599} +{"lr": 1.1521049841102396e-05, "data_time": 0.6005637645721436, "loss": 0.20713708102703093, "time": 0.7401561498641968, "epoch": 10, "memory": 6318, "step": 37699} +{"lr": 1.1433667124402997e-05, "data_time": 0.575803804397583, "loss": 0.19673682302236556, "time": 0.7175272226333618, "epoch": 10, "memory": 6318, "step": 37799} +{"lr": 1.1348828621971181e-05, "data_time": 0.648768138885498, "loss": 0.21565770655870437, "time": 0.790831732749939, "epoch": 10, "memory": 6318, "step": 37899} +{"lr": 1.1266539288259486e-05, "data_time": 0.627412486076355, "loss": 0.21127624362707137, "time": 0.7675125360488891, "epoch": 10, "memory": 6318, "step": 37999} +{"lr": 1.1186803928852477e-05, "data_time": 0.596336054801941, "loss": 0.20482346266508103, "time": 0.7371329069137573, "epoch": 10, "memory": 6318, "step": 38099} +{"lr": 1.1109627200186125e-05, "data_time": 0.6115892171859741, "loss": 0.1991797998547554, "time": 0.7520804166793823, "epoch": 10, "memory": 6318, "step": 38199} +{"lr": 1.1035013609275835e-05, "data_time": 0.6719270706176758, "loss": 0.21726299673318863, "time": 0.8128053426742554, "epoch": 10, "memory": 6318, "step": 38299} +{"lr": 1.0962967513453298e-05, "data_time": 0.5536705493927002, "loss": 0.21128429919481279, "time": 0.6952198266983032, "epoch": 10, "memory": 6318, "step": 38399} +{"lr": 1.0893493120111999e-05, "data_time": 0.6330197334289551, "loss": 0.20664433985948563, "time": 0.7728255987167358, "epoch": 10, "memory": 6318, "step": 38499} +{"lr": 1.0826594486461509e-05, "data_time": 0.6278717279434204, "loss": 0.20064590871334076, "time": 0.7669581413269043, "epoch": 10, "memory": 6318, "step": 38599} +{"lr": 1.0762275519290557e-05, "data_time": 0.6420759916305542, "loss": 0.2029467850923538, "time": 0.7832370281219483, "epoch": 10, "memory": 6318, "step": 38699} +{"lr": 1.0700539974738862e-05, "data_time": 0.608813214302063, "loss": 0.2126883715391159, "time": 0.7497075319290161, "epoch": 10, "memory": 6318, "step": 38799} +{"lr": 1.064139145807783e-05, "data_time": 0.6813237428665161, "loss": 0.2037241443991661, "time": 0.8232875108718872, "epoch": 10, "memory": 6318, "step": 38899} +{"lr": 1.0584833423499947e-05, "data_time": 0.6327476739883423, "loss": 0.18783843368291855, "time": 0.7735593795776368, "epoch": 10, "memory": 6318, "step": 38999} +{"lr": 1.0530869173917099e-05, "data_time": 0.6954510688781739, "loss": 0.1974943086504936, "time": 0.8368297815322876, "epoch": 10, "memory": 6318, "step": 39099} +{"lr": 1.0479501860767654e-05, "data_time": 0.6410568475723266, "loss": 0.21395813673734665, "time": 0.7825519323349, "epoch": 10, "memory": 6318, "step": 39199} +{"lr": 1.043073448383247e-05, "data_time": 0.5959242582321167, "loss": 0.22101271003484727, "time": 0.7487569332122803, "epoch": 10, "memory": 6318, "step": 39299} +{"lr": 1.038456989105968e-05, "data_time": 0.639715600013733, "loss": 0.21259773075580596, "time": 0.7799566745758056, "epoch": 10, "memory": 6318, "step": 39399} +{"lr": 1.034101077839838e-05, "data_time": 0.5884696006774902, "loss": 0.22869202941656114, "time": 0.7296674251556396, "epoch": 10, "memory": 6318, "step": 39499} +{"lr": 1.0300059689641194e-05, "data_time": 0.6050638914108276, "loss": 0.21708722412586212, "time": 0.746091651916504, "epoch": 10, "memory": 6318, "step": 39599} +{"lr": 1.0261719016275683e-05, "data_time": 0.635635232925415, "loss": 0.1917801320552826, "time": 0.7758737802505493, "epoch": 10, "memory": 6318, "step": 39699} +{"lr": 1.0225990997344757e-05, "data_time": 0.6306647062301636, "loss": 0.20383365154266359, "time": 0.7720104932785035, "epoch": 10, "memory": 6318, "step": 39799} +{"lr": 1.019287771931587e-05, "data_time": 0.6178133249282837, "loss": 0.20559848845005035, "time": 0.760369873046875, "epoch": 10, "memory": 6318, "step": 39899} +{"lr": 1.0162381115959187e-05, "data_time": 0.5978616952896119, "loss": 0.19369159936904906, "time": 0.7390765190124512, "epoch": 10, "memory": 6318, "step": 39999} +{"lr": 1.0134502968234636e-05, "data_time": 0.6582518815994263, "loss": 0.20472228899598122, "time": 0.7998521566390991, "epoch": 10, "memory": 6318, "step": 40099} +{"lr": 1.0109244904187918e-05, "data_time": 0.5787601232528686, "loss": 0.19884382784366608, "time": 0.7180487394332886, "epoch": 10, "memory": 6318, "step": 40199} +{"lr": 1.0086608398855446e-05, "data_time": 0.6692479372024536, "loss": 0.22374869287014007, "time": 0.8103873252868652, "epoch": 10, "memory": 6318, "step": 40299} +{"lr": 1.0066594774178176e-05, "data_time": 0.6905025959014892, "loss": 0.20124364793300628, "time": 0.8318718194961547, "epoch": 10, "memory": 6318, "step": 40399} +{"lr": 1.0049205198924413e-05, "data_time": 0.9919116258621216, "loss": 0.20201308280229568, "time": 1.1397632598876952, "epoch": 10, "memory": 6318, "step": 40499} +{"lr": 1.0034440688621577e-05, "data_time": 0.5915333747863769, "loss": 0.20739715546369553, "time": 0.7321382522583008, "epoch": 10, "memory": 6318, "step": 40599} +{"lr": 1.0022302105496896e-05, "data_time": 0.607059907913208, "loss": 0.20587428659200668, "time": 0.7482892751693726, "epoch": 10, "memory": 6318, "step": 40699} +{"lr": 1.0012790158427017e-05, "data_time": 0.7182532787322998, "loss": 0.21151627898216246, "time": 0.8581497192382812, "epoch": 10, "memory": 6318, "step": 40799} +{"lr": 1.000590540289662e-05, "data_time": 0.6576358079910278, "loss": 0.21189197301864623, "time": 0.8053101062774658, "epoch": 10, "memory": 6318, "step": 40899} +{"lr": 1.0001648240966046e-05, "data_time": 0.6686748504638672, "loss": 0.21074877828359603, "time": 0.8098900318145752, "epoch": 10, "memory": 6318, "step": 40999} +{"lr": 1.0000018921247718e-05, "data_time": 0.6262905836105347, "loss": 0.19762154221534728, "time": 0.7657563924789429, "epoch": 10, "memory": 6318, "step": 41099} +{"accuracy/top1": 93.51074981689453, "data_time": 0.3998194396495819, "time": 0.48767611185709636, "step": 10} diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/config.py b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/config.py new file mode 100644 index 0000000000000000000000000000000000000000..b35ee250106aa84d1065073f108965cfa202c34d --- /dev/null +++ b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/config.py @@ -0,0 +1,366 @@ +optim_wrapper = dict( + optimizer=dict( + type='SGD', + lr=0.0001, + momentum=0.9, + weight_decay=0.0001, + _scope_='mmpretrain'), + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=False, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=512) +model = dict( + type='ImageClassifier', + backbone=dict( + frozen_stages=2, + type='ResNet', + depth=50, + num_stages=4, + out_indices=(3, ), + style='pytorch', + init_cfg=dict( + type='Pretrained', + checkpoint= + 'https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth', + prefix='backbone')), + neck=dict(type='GlobalAveragePooling'), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=2048, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + topk=1)) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/all1.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/all2.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/all3.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3fake8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/cc1m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3real8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV2-1-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/if-dpmsolver++-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = dict(type='Accuracy', topk=1) +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV2-1-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/if-dpmsolver++-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = dict(type='Accuracy', topk=1) +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1' diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/events.out.tfevents.1685949654.SH-IDC1-10-140-24-19.149928.0 b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/events.out.tfevents.1685949654.SH-IDC1-10-140-24-19.149928.0 new file mode 100644 index 0000000000000000000000000000000000000000..f9830f5c97b5faf1a7bbac9cd95c7c39240a26ed --- /dev/null +++ b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/events.out.tfevents.1685949654.SH-IDC1-10-140-24-19.149928.0 @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:d4d238f9b15d26c1f3dca4b1bfa6071360dd2bf210d138ef0d93efbee8aed605 +size 5478960 diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/scalars.json b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/scalars.json new file mode 100644 index 0000000000000000000000000000000000000000..e49ebb98be8b4598bdcaf2260aacc506d3abfed9 --- /dev/null +++ b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/scalars.json @@ -0,0 +1,420 @@ +{"lr": 9.999871217862936e-05, "data_time": 0.5505931377410889, "loss": 0.6505624175071716, "time": 0.6934590101242065, "epoch": 1, "memory": 9436, "step": 100} +{"lr": 9.999479662539605e-05, "data_time": 0.5152626276016236, "loss": 0.5974045336246491, "time": 0.6552257299423218, "epoch": 1, "memory": 6319, "step": 200} +{"lr": 9.99882534375654e-05, "data_time": 0.001011061668395996, "loss": 0.5777045428752899, "time": 0.6513210296630859, "epoch": 1, "memory": 6319, "step": 300} +{"lr": 9.997908299725075e-05, "data_time": 0.0008590936660766602, "loss": 0.5506176173686981, "time": 0.6428668737411499, "epoch": 1, "memory": 6319, "step": 400} +{"lr": 9.996728583999318e-05, "data_time": 0.2106086015701294, "loss": 0.5142255157232285, "time": 0.6030692815780639, "epoch": 1, "memory": 6319, "step": 500} +{"lr": 9.995286265473058e-05, "data_time": 0.34703359603881834, "loss": 0.5076296508312226, "time": 0.7148289918899536, "epoch": 1, "memory": 6319, "step": 600} +{"lr": 9.993581428375725e-05, "data_time": 0.46016180515289307, "loss": 0.4696363240480423, "time": 0.7294468402862548, "epoch": 1, "memory": 6319, "step": 700} +{"lr": 9.991614172267466e-05, "data_time": 0.4658113956451416, "loss": 0.4616829603910446, "time": 0.6683710813522339, "epoch": 1, "memory": 6319, "step": 800} +{"lr": 9.989384612033355e-05, "data_time": 0.43298189640045165, "loss": 0.4413948357105255, "time": 0.7080170631408691, "epoch": 1, "memory": 6319, "step": 900} +{"lr": 9.986892877876658e-05, "data_time": 0.43495018482208253, "loss": 0.4226416856050491, "time": 0.6916615962982178, "epoch": 1, "memory": 6319, "step": 1000} +{"lr": 9.984139115311244e-05, "data_time": 0.5044760465621948, "loss": 0.4144183903932571, "time": 0.6448450088500977, "epoch": 1, "memory": 6319, "step": 1100} +{"lr": 9.981123485153081e-05, "data_time": 0.2968530893325806, "loss": 0.4110246986150742, "time": 0.6685676097869873, "epoch": 1, "memory": 6319, "step": 1200} +{"lr": 9.977846163510851e-05, "data_time": 0.19977850914001466, "loss": 0.4085386097431183, "time": 0.6431279659271241, "epoch": 1, "memory": 6319, "step": 1300} +{"lr": 9.974307341775653e-05, "data_time": 0.5203864812850952, "loss": 0.3970152199268341, "time": 0.6990034580230713, "epoch": 1, "memory": 6319, "step": 1400} +{"lr": 9.970507226609833e-05, "data_time": 0.5476192951202392, "loss": 0.3996817052364349, "time": 0.6898721218109131, "epoch": 1, "memory": 6319, "step": 1500} +{"lr": 9.966446039934934e-05, "data_time": 0.5750117301940918, "loss": 0.39513109922409057, "time": 0.7154863595962524, "epoch": 1, "memory": 6319, "step": 1600} +{"lr": 9.962124018918683e-05, "data_time": 0.5280206203460693, "loss": 0.3815469264984131, "time": 0.6684962034225463, "epoch": 1, "memory": 6319, "step": 1700} +{"lr": 9.957541415961212e-05, "data_time": 0.5535875082015991, "loss": 0.39518000185489655, "time": 0.694165563583374, "epoch": 1, "memory": 6319, "step": 1800} +{"lr": 9.952698498680267e-05, "data_time": 0.4987358331680298, "loss": 0.3807822555303574, "time": 0.6390132427215576, "epoch": 1, "memory": 6319, "step": 1900} +{"lr": 9.947595549895583e-05, "data_time": 0.6053460836410522, "loss": 0.36012162566184996, "time": 0.7453249216079711, "epoch": 1, "memory": 6319, "step": 2000} +{"lr": 9.942232867612417e-05, "data_time": 0.5652534246444703, "loss": 0.37053406834602354, "time": 0.7054509878158569, "epoch": 1, "memory": 6319, "step": 2100} +{"lr": 9.93661076500406e-05, "data_time": 0.5398832798004151, "loss": 0.34339838922023774, "time": 0.6806561946868896, "epoch": 1, "memory": 6319, "step": 2200} +{"lr": 9.930729570393622e-05, "data_time": 0.5331053256988525, "loss": 0.35869871973991396, "time": 0.6771989345550538, "epoch": 1, "memory": 6319, "step": 2300} +{"lr": 9.924589627234824e-05, "data_time": 0.5512842416763306, "loss": 0.34724647700786593, "time": 0.6925719976425171, "epoch": 1, "memory": 6319, "step": 2400} +{"lr": 9.918191294091956e-05, "data_time": 0.5681015491485596, "loss": 0.3581976592540741, "time": 0.7094371080398559, "epoch": 1, "memory": 6319, "step": 2500} +{"lr": 9.91153494461892e-05, "data_time": 0.5202316999435425, "loss": 0.34429652988910675, "time": 0.6607050418853759, "epoch": 1, "memory": 6319, "step": 2600} +{"lr": 9.904620967537422e-05, "data_time": 0.545382022857666, "loss": 0.35300026535987855, "time": 0.685313367843628, "epoch": 1, "memory": 6319, "step": 2700} +{"lr": 9.897449766614277e-05, "data_time": 0.5922863721847534, "loss": 0.3269111156463623, "time": 0.7319374084472656, "epoch": 1, "memory": 6319, "step": 2800} +{"lr": 9.890021760637804e-05, "data_time": 0.5891611337661743, "loss": 0.34317596852779386, "time": 0.7286672115325927, "epoch": 1, "memory": 6319, "step": 2900} +{"lr": 9.882337383393409e-05, "data_time": 0.6334947109222412, "loss": 0.3492516875267029, "time": 0.7731748104095459, "epoch": 1, "memory": 6319, "step": 3000} +{"lr": 9.874397083638193e-05, "data_time": 0.5344001054763794, "loss": 0.3423819899559021, "time": 0.6852553844451904, "epoch": 1, "memory": 6319, "step": 3100} +{"lr": 9.866201325074828e-05, "data_time": 0.5381552457809449, "loss": 0.33956805765628817, "time": 0.6779947757720948, "epoch": 1, "memory": 6319, "step": 3200} +{"lr": 9.857750586324393e-05, "data_time": 0.5681901216506958, "loss": 0.3352639853954315, "time": 0.707914400100708, "epoch": 1, "memory": 6319, "step": 3300} +{"lr": 9.84904536089848e-05, "data_time": 0.719521164894104, "loss": 0.32742994725704194, "time": 0.8602179527282715, "epoch": 1, "memory": 6319, "step": 3400} +{"lr": 9.840086157170351e-05, "data_time": 0.5882129669189453, "loss": 0.3180242747068405, "time": 0.7277715682983399, "epoch": 1, "memory": 6319, "step": 3500} +{"lr": 9.830873498345255e-05, "data_time": 0.5058643341064453, "loss": 0.3155341863632202, "time": 0.6458086013793946, "epoch": 1, "memory": 6319, "step": 3600} +{"lr": 9.821407922429864e-05, "data_time": 0.5474719762802124, "loss": 0.3262654036283493, "time": 0.6877282142639161, "epoch": 1, "memory": 6319, "step": 3700} +{"lr": 9.811689982200877e-05, "data_time": 0.5631479740142822, "loss": 0.3217996835708618, "time": 0.7033457279205322, "epoch": 1, "memory": 6319, "step": 3800} +{"lr": 9.801720245172702e-05, "data_time": 0.5131886959075928, "loss": 0.3055334031581879, "time": 0.6539286375045776, "epoch": 1, "memory": 6319, "step": 3900} +{"lr": 9.791499293564358e-05, "data_time": 0.5458756446838379, "loss": 0.3265227496623993, "time": 0.6876251220703125, "epoch": 1, "memory": 6319, "step": 4000} +{"lr": 9.781027724265439e-05, "data_time": 0.6899309873580932, "loss": 0.3120740562677383, "time": 0.831060814857483, "epoch": 1, "memory": 6319, "step": 4100} +{"accuracy/top1": 82.47356414794922, "data_time": 0.393584958645476, "time": 0.48177839327259225, "step": 1} +{"lr": 9.769111539369434e-05, "data_time": 0.6067419528961182, "loss": 0.32421033680438993, "time": 0.748124074935913, "epoch": 2, "memory": 6318, "step": 4211} +{"lr": 9.758113190911623e-05, "data_time": 0.5759464502334595, "loss": 0.302031347155571, "time": 0.7165237665176392, "epoch": 2, "memory": 6318, "step": 4311} +{"lr": 9.746866174465984e-05, "data_time": 0.38157072067260744, "loss": 0.31559916138648986, "time": 0.6634103298187256, "epoch": 2, "memory": 6318, "step": 4411} +{"lr": 9.735371146842894e-05, "data_time": 0.5546189069747924, "loss": 0.31170307099819183, "time": 0.6952664852142334, "epoch": 2, "memory": 6318, "step": 4511} +{"lr": 9.723628779336255e-05, "data_time": 0.5642520904541015, "loss": 0.3174903839826584, "time": 0.7054730653762817, "epoch": 2, "memory": 6318, "step": 4611} +{"lr": 9.711639757684265e-05, "data_time": 0.5179787397384643, "loss": 0.31193001568317413, "time": 0.7149520874023437, "epoch": 2, "memory": 6318, "step": 4711} +{"lr": 9.699404782029395e-05, "data_time": 0.4935267448425293, "loss": 0.3097731113433838, "time": 0.6753772735595703, "epoch": 2, "memory": 6318, "step": 4811} +{"lr": 9.686924566877498e-05, "data_time": 0.3179853200912476, "loss": 0.32091196775436404, "time": 0.692458701133728, "epoch": 2, "memory": 6318, "step": 4911} +{"lr": 9.674199841056078e-05, "data_time": 0.45420470237731936, "loss": 0.2987965703010559, "time": 0.6875965356826782, "epoch": 2, "memory": 6318, "step": 5011} +{"lr": 9.661231347671705e-05, "data_time": 0.5439108610153198, "loss": 0.3439896672964096, "time": 0.69365394115448, "epoch": 2, "memory": 6318, "step": 5111} +{"lr": 9.64801984406668e-05, "data_time": 0.4245661020278931, "loss": 0.30733801126480104, "time": 0.7456601381301879, "epoch": 2, "memory": 6318, "step": 5211} +{"lr": 9.634566101774737e-05, "data_time": 0.19504849910736083, "loss": 0.3189211219549179, "time": 0.7636782407760621, "epoch": 2, "memory": 6318, "step": 5311} +{"lr": 9.62087090647604e-05, "data_time": 0.04529838562011719, "loss": 0.29260619133710863, "time": 0.6711793899536133, "epoch": 2, "memory": 6318, "step": 5411} +{"lr": 9.606935057951247e-05, "data_time": 0.2667135953903198, "loss": 0.3066791981458664, "time": 0.7659513711929321, "epoch": 2, "memory": 6318, "step": 5511} +{"lr": 9.592759370034884e-05, "data_time": 0.7144370317459107, "loss": 0.2962004154920578, "time": 1.105335783958435, "epoch": 2, "memory": 6318, "step": 5611} +{"lr": 9.578344670567721e-05, "data_time": 0.4193567276000977, "loss": 0.2966422110795975, "time": 0.6607670783996582, "epoch": 2, "memory": 6318, "step": 5711} +{"lr": 9.563691801348495e-05, "data_time": 0.6298110485076904, "loss": 0.29018100500106814, "time": 0.7693820238113404, "epoch": 2, "memory": 6318, "step": 5811} +{"lr": 9.548801618084738e-05, "data_time": 0.6293867111206055, "loss": 0.29744812846183777, "time": 0.7697560071945191, "epoch": 2, "memory": 6318, "step": 5911} +{"lr": 9.533674990342786e-05, "data_time": 0.6395065069198609, "loss": 0.30597208589315417, "time": 0.7797405958175659, "epoch": 2, "memory": 6318, "step": 6011} +{"lr": 9.518312801497029e-05, "data_time": 0.5134622812271118, "loss": 0.28178397864103316, "time": 0.6528750896453858, "epoch": 2, "memory": 6318, "step": 6111} +{"lr": 9.502715948678276e-05, "data_time": 0.4936917781829834, "loss": 0.3079785257577896, "time": 0.6340498685836792, "epoch": 2, "memory": 6318, "step": 6211} +{"lr": 9.486885342721427e-05, "data_time": 0.5484623908996582, "loss": 0.30840844213962554, "time": 0.6890472650527955, "epoch": 2, "memory": 6318, "step": 6311} +{"lr": 9.470821908112202e-05, "data_time": 0.645397424697876, "loss": 0.271228489279747, "time": 0.7857234954833985, "epoch": 2, "memory": 6318, "step": 6411} +{"lr": 9.454526582933214e-05, "data_time": 0.756637978553772, "loss": 0.28856208473443984, "time": 0.8973519563674927, "epoch": 2, "memory": 6318, "step": 6511} +{"lr": 9.438000318809176e-05, "data_time": 0.5645488262176513, "loss": 0.28264967501163485, "time": 0.7037885427474976, "epoch": 2, "memory": 6318, "step": 6611} +{"lr": 9.42124408085132e-05, "data_time": 0.5453065872192383, "loss": 0.2823116719722748, "time": 0.6865315675735474, "epoch": 2, "memory": 6318, "step": 6711} +{"lr": 9.404258847601036e-05, "data_time": 0.5627376079559326, "loss": 0.2907982677221298, "time": 0.7028753757476807, "epoch": 2, "memory": 6318, "step": 6811} +{"lr": 9.387045610972717e-05, "data_time": 0.5891398668289185, "loss": 0.29210465848445893, "time": 0.7290917873382569, "epoch": 2, "memory": 6318, "step": 6911} +{"lr": 9.369605376195838e-05, "data_time": 0.555685019493103, "loss": 0.2789680004119873, "time": 0.6954281091690063, "epoch": 2, "memory": 6318, "step": 7011} +{"lr": 9.351939161756269e-05, "data_time": 0.5706737756729126, "loss": 0.2751905009150505, "time": 0.7108277559280396, "epoch": 2, "memory": 6318, "step": 7111} +{"lr": 9.334047999336764e-05, "data_time": 0.6172459363937378, "loss": 0.28409205973148344, "time": 0.7575044393539428, "epoch": 2, "memory": 6318, "step": 7211} +{"lr": 9.315932933756752e-05, "data_time": 0.6438805818557739, "loss": 0.274951645731926, "time": 0.7835977792739868, "epoch": 2, "memory": 6318, "step": 7311} +{"lr": 9.297595022911287e-05, "data_time": 0.5824469566345215, "loss": 0.2724910795688629, "time": 0.7227290153503418, "epoch": 2, "memory": 6318, "step": 7411} +{"lr": 9.279035337709295e-05, "data_time": 0.6200707435607911, "loss": 0.28884666264057157, "time": 0.7615639209747315, "epoch": 2, "memory": 6318, "step": 7511} +{"lr": 9.260254962011016e-05, "data_time": 0.5936978816986084, "loss": 0.2855002909898758, "time": 0.7335208415985107, "epoch": 2, "memory": 6318, "step": 7611} +{"lr": 9.241254992564702e-05, "data_time": 0.6418598651885986, "loss": 0.26392817199230195, "time": 0.7825077533721924, "epoch": 2, "memory": 6318, "step": 7711} +{"lr": 9.22203653894259e-05, "data_time": 0.5894522190093994, "loss": 0.274297884106636, "time": 0.7303678512573242, "epoch": 2, "memory": 6318, "step": 7811} +{"lr": 9.202600723476098e-05, "data_time": 0.603200101852417, "loss": 0.26613368690013883, "time": 0.7431907415390014, "epoch": 2, "memory": 6318, "step": 7911} +{"lr": 9.182948681190268e-05, "data_time": 0.6074742078781128, "loss": 0.3004067957401276, "time": 0.7514505624771118, "epoch": 2, "memory": 6318, "step": 8011} +{"lr": 9.163081559737497e-05, "data_time": 0.6953469753265381, "loss": 0.28262611478567123, "time": 0.8359466791152954, "epoch": 2, "memory": 6318, "step": 8111} +{"lr": 9.143000519330517e-05, "data_time": 0.5899689674377442, "loss": 0.2500761702656746, "time": 0.7300094604492188, "epoch": 2, "memory": 6318, "step": 8211} +{"accuracy/top1": 77.86273956298828, "data_time": 0.37948496341705323, "time": 0.46813223759333294, "step": 2} +{"lr": 9.120461478753376e-05, "data_time": 0.6181094408035278, "loss": 0.27026086002588273, "time": 0.7582679986953735, "epoch": 3, "memory": 6318, "step": 8322} +{"lr": 9.099932932138134e-05, "data_time": 0.5869765520095825, "loss": 0.2636982038617134, "time": 0.7259285688400269, "epoch": 3, "memory": 6318, "step": 8422} +{"lr": 9.079194154362181e-05, "data_time": 0.5926679611206055, "loss": 0.3168958812952042, "time": 0.7332017660140991, "epoch": 3, "memory": 6318, "step": 8522} +{"lr": 9.058246356541787e-05, "data_time": 0.6440795183181762, "loss": 0.2605751916766167, "time": 0.7840817213058472, "epoch": 3, "memory": 6318, "step": 8622} +{"lr": 9.037090761999699e-05, "data_time": 0.5970780611038208, "loss": 0.28410408943891524, "time": 0.7363588571548462, "epoch": 3, "memory": 6318, "step": 8722} +{"lr": 9.015728606193701e-05, "data_time": 0.5913930892944336, "loss": 0.28051196932792666, "time": 0.7306947231292724, "epoch": 3, "memory": 6318, "step": 8822} +{"lr": 8.994161136644488e-05, "data_time": 0.6805069208145141, "loss": 0.2685139149427414, "time": 0.829882001876831, "epoch": 3, "memory": 6318, "step": 8922} +{"lr": 8.972389612862792e-05, "data_time": 0.5869032144546509, "loss": 0.2602032795548439, "time": 0.729338526725769, "epoch": 3, "memory": 6318, "step": 9022} +{"lr": 8.950415306275833e-05, "data_time": 0.6840051174163818, "loss": 0.2694586977362633, "time": 0.8367887496948242, "epoch": 3, "memory": 6318, "step": 9122} +{"lr": 8.928239500153059e-05, "data_time": 0.5554930925369262, "loss": 0.2827488571405411, "time": 0.6961770534515381, "epoch": 3, "memory": 6318, "step": 9222} +{"lr": 8.905863489531232e-05, "data_time": 0.639508056640625, "loss": 0.264517243206501, "time": 0.7878513813018799, "epoch": 3, "memory": 6318, "step": 9322} +{"lr": 8.883288581138775e-05, "data_time": 0.5916643142700195, "loss": 0.2758107051253319, "time": 0.7347429275512696, "epoch": 3, "memory": 6318, "step": 9422} +{"lr": 8.86051609331947e-05, "data_time": 0.6540353775024415, "loss": 0.28519781231880187, "time": 0.7988569736480713, "epoch": 3, "memory": 6318, "step": 9522} +{"lr": 8.837547355955467e-05, "data_time": 0.608176875114441, "loss": 0.26498973965644834, "time": 0.7500752449035645, "epoch": 3, "memory": 6318, "step": 9622} +{"lr": 8.814383710389614e-05, "data_time": 0.631160306930542, "loss": 0.27551338374614714, "time": 0.7737520933151245, "epoch": 3, "memory": 6318, "step": 9722} +{"lr": 8.791026509347146e-05, "data_time": 0.6326789379119873, "loss": 0.2672505512833595, "time": 0.7738667726516724, "epoch": 3, "memory": 6318, "step": 9822} +{"lr": 8.767477116856659e-05, "data_time": 0.6556741952896118, "loss": 0.27411281168460844, "time": 0.7964605093002319, "epoch": 3, "memory": 6318, "step": 9922} +{"lr": 8.743736908170466e-05, "data_time": 0.6806576490402222, "loss": 0.27539069652557374, "time": 0.8206354856491089, "epoch": 3, "memory": 6318, "step": 10022} +{"lr": 8.719807269684305e-05, "data_time": 0.5637223720550537, "loss": 0.26915688514709474, "time": 0.7057634592056274, "epoch": 3, "memory": 6318, "step": 10122} +{"lr": 8.695689598856356e-05, "data_time": 0.5881913900375366, "loss": 0.27323976457118987, "time": 0.7285332441329956, "epoch": 3, "memory": 6318, "step": 10222} +{"lr": 8.67138530412559e-05, "data_time": 0.6564640283584595, "loss": 0.26446687132120134, "time": 0.798209261894226, "epoch": 3, "memory": 6318, "step": 10322} +{"lr": 8.646895804829604e-05, "data_time": 0.5809694528579712, "loss": 0.2585653379559517, "time": 0.7214553356170654, "epoch": 3, "memory": 6318, "step": 10422} +{"lr": 8.622222531121664e-05, "data_time": 0.5559334754943848, "loss": 0.24912258386611938, "time": 0.6959937572479248, "epoch": 3, "memory": 6318, "step": 10522} +{"lr": 8.597366923887211e-05, "data_time": 0.6048402309417724, "loss": 0.27612168788909913, "time": 0.7447296380996704, "epoch": 3, "memory": 6318, "step": 10622} +{"lr": 8.57233043465972e-05, "data_time": 0.5658660888671875, "loss": 0.26694310158491136, "time": 0.7070073366165162, "epoch": 3, "memory": 6318, "step": 10722} +{"lr": 8.547114525535937e-05, "data_time": 0.6130959510803222, "loss": 0.2739654153585434, "time": 0.7524494171142578, "epoch": 3, "memory": 6318, "step": 10822} +{"lr": 8.521720669090474e-05, "data_time": 0.5844237327575683, "loss": 0.25660845935344695, "time": 0.7236541509628296, "epoch": 3, "memory": 6318, "step": 10922} +{"lr": 8.496150348289821e-05, "data_time": 0.6259898900985718, "loss": 0.2606738433241844, "time": 0.766187334060669, "epoch": 3, "memory": 6318, "step": 11022} +{"lr": 8.470405056405736e-05, "data_time": 0.4467794418334961, "loss": 0.24720416367053985, "time": 0.7321699380874633, "epoch": 3, "memory": 6318, "step": 11122} +{"lr": 8.44448629692807e-05, "data_time": 0.5983347415924072, "loss": 0.28130067586898805, "time": 0.7394391775131226, "epoch": 3, "memory": 6318, "step": 11222} +{"lr": 8.418395583476941e-05, "data_time": 0.607829236984253, "loss": 0.2544344007968903, "time": 0.7481575012207031, "epoch": 3, "memory": 6318, "step": 11322} +{"lr": 8.392134439714336e-05, "data_time": 0.604886507987976, "loss": 0.26822121143341066, "time": 0.7466015338897705, "epoch": 3, "memory": 6318, "step": 11422} +{"lr": 8.365704399255133e-05, "data_time": 0.5664644002914428, "loss": 0.25017236173152924, "time": 0.7069388628005981, "epoch": 3, "memory": 6318, "step": 11522} +{"lr": 8.339107005577562e-05, "data_time": 0.6967349529266358, "loss": 0.25618584603071215, "time": 0.844751501083374, "epoch": 3, "memory": 6318, "step": 11622} +{"lr": 8.31234381193305e-05, "data_time": 0.6183933973312378, "loss": 0.24026695638895035, "time": 0.7581479549407959, "epoch": 3, "memory": 6318, "step": 11722} +{"lr": 8.285416381255504e-05, "data_time": 0.6711583614349366, "loss": 0.2746740996837616, "time": 0.8233832359313965, "epoch": 3, "memory": 6318, "step": 11822} +{"lr": 8.258326286070047e-05, "data_time": 0.6249480009078979, "loss": 0.25514610558748246, "time": 0.7648244142532349, "epoch": 3, "memory": 6318, "step": 11922} +{"lr": 8.231075108401202e-05, "data_time": 0.6216845989227295, "loss": 0.2432646408677101, "time": 0.7675186634063721, "epoch": 3, "memory": 6318, "step": 12022} +{"lr": 8.203664439680493e-05, "data_time": 0.5821406126022339, "loss": 0.2458882138133049, "time": 0.7232086896896363, "epoch": 3, "memory": 6318, "step": 12122} +{"lr": 8.176095880653475e-05, "data_time": 0.5440454006195068, "loss": 0.2554523289203644, "time": 0.6927780866622925, "epoch": 3, "memory": 6318, "step": 12222} +{"lr": 8.148371041286308e-05, "data_time": 0.6835984945297241, "loss": 0.25288918167352675, "time": 0.8229263305664063, "epoch": 3, "memory": 6318, "step": 12322} +{"accuracy/top1": 79.14701843261719, "data_time": 0.39744665622711184, "time": 0.48549293478329975, "step": 3} +{"lr": 8.117415423358443e-05, "data_time": 0.8512596845626831, "loss": 0.2658050417900085, "time": 0.991992712020874, "epoch": 4, "memory": 6318, "step": 12433} +{"lr": 8.089366155759266e-05, "data_time": 0.6952439785003662, "loss": 0.25963141620159147, "time": 0.8345962285995483, "epoch": 4, "memory": 6318, "step": 12533} +{"lr": 8.061165672717273e-05, "data_time": 0.6424442529678345, "loss": 0.2485888347029686, "time": 0.7843495368957519, "epoch": 4, "memory": 6318, "step": 12633} +{"lr": 8.032815621102116e-05, "data_time": 0.5740828275680542, "loss": 0.26304204761981964, "time": 0.715699315071106, "epoch": 4, "memory": 6318, "step": 12733} +{"lr": 8.004317656518042e-05, "data_time": 0.6650530815124511, "loss": 0.26400354504585266, "time": 0.8057906150817871, "epoch": 4, "memory": 6318, "step": 12833} +{"lr": 7.975673443207223e-05, "data_time": 0.6941095352172851, "loss": 0.2509127527475357, "time": 0.8341012954711914, "epoch": 4, "memory": 6318, "step": 12933} +{"lr": 7.946884653952537e-05, "data_time": 0.627883791923523, "loss": 0.2448240414261818, "time": 0.7698372602462769, "epoch": 4, "memory": 6318, "step": 13033} +{"lr": 7.91795296997992e-05, "data_time": 0.6392451286315918, "loss": 0.25170914083719254, "time": 0.7796733856201172, "epoch": 4, "memory": 6318, "step": 13133} +{"lr": 7.888880080860151e-05, "data_time": 0.6729235649108887, "loss": 0.27842089384794233, "time": 0.8145662069320678, "epoch": 4, "memory": 6318, "step": 13233} +{"lr": 7.859667684410205e-05, "data_time": 0.611674165725708, "loss": 0.2623119980096817, "time": 0.7529521942138672, "epoch": 4, "memory": 6318, "step": 13333} +{"lr": 7.830317486594082e-05, "data_time": 0.6273994684219361, "loss": 0.2496202439069748, "time": 0.7699385404586792, "epoch": 4, "memory": 6318, "step": 13433} +{"lr": 7.800831201423207e-05, "data_time": 0.6156965255737304, "loss": 0.24734974801540374, "time": 0.7585714817047119, "epoch": 4, "memory": 6318, "step": 13533} +{"lr": 7.771210550856309e-05, "data_time": 0.6352834701538086, "loss": 0.24084002524614334, "time": 0.7980647087097168, "epoch": 4, "memory": 6318, "step": 13633} +{"lr": 7.741457264698884e-05, "data_time": 0.6363005876541138, "loss": 0.23988503217697144, "time": 0.7775790452957153, "epoch": 4, "memory": 6318, "step": 13733} +{"lr": 7.711573080502147e-05, "data_time": 0.6062970399856568, "loss": 0.26099351346492766, "time": 0.7483415842056275, "epoch": 4, "memory": 6318, "step": 13833} +{"lr": 7.681559743461609e-05, "data_time": 0.5898828268051147, "loss": 0.24848497658967972, "time": 0.7293738842010498, "epoch": 4, "memory": 6318, "step": 13933} +{"lr": 7.651419006315099e-05, "data_time": 0.5952056407928467, "loss": 0.26195275634527204, "time": 0.735258960723877, "epoch": 4, "memory": 6318, "step": 14033} +{"lr": 7.621152629240453e-05, "data_time": 0.5790565729141235, "loss": 0.24304310530424117, "time": 0.719720721244812, "epoch": 4, "memory": 6318, "step": 14133} +{"lr": 7.590762379752719e-05, "data_time": 0.5663726329803467, "loss": 0.25091861486434935, "time": 0.7070995569229126, "epoch": 4, "memory": 6318, "step": 14233} +{"lr": 7.560250032600911e-05, "data_time": 0.6194745302200317, "loss": 0.2575841635465622, "time": 0.769227123260498, "epoch": 4, "memory": 6318, "step": 14333} +{"lr": 7.52961736966437e-05, "data_time": 0.5746166706085205, "loss": 0.2373547524213791, "time": 0.7145738363265991, "epoch": 4, "memory": 6318, "step": 14433} +{"lr": 7.498866179848732e-05, "data_time": 0.6154284715652466, "loss": 0.2282430574297905, "time": 0.7566136121749878, "epoch": 4, "memory": 6318, "step": 14533} +{"lr": 7.467998258981415e-05, "data_time": 0.5762117624282836, "loss": 0.25059892386198046, "time": 0.7285410165786743, "epoch": 4, "memory": 6318, "step": 14633} +{"lr": 7.437015409706804e-05, "data_time": 0.6411524772644043, "loss": 0.24868184477090835, "time": 0.7812589406967163, "epoch": 4, "memory": 6318, "step": 14733} +{"lr": 7.405919441380929e-05, "data_time": 0.5939372777938843, "loss": 0.233042611181736, "time": 0.7347374200820923, "epoch": 4, "memory": 6318, "step": 14833} +{"lr": 7.374712169965821e-05, "data_time": 0.5857679843902588, "loss": 0.2688971906900406, "time": 0.7347111463546753, "epoch": 4, "memory": 6318, "step": 14933} +{"lr": 7.343395417923461e-05, "data_time": 0.6318930149078369, "loss": 0.2383084326982498, "time": 0.7723827600479126, "epoch": 4, "memory": 6318, "step": 15033} +{"lr": 7.311971014109342e-05, "data_time": 0.592315936088562, "loss": 0.25025821626186373, "time": 0.7330101251602172, "epoch": 4, "memory": 6318, "step": 15133} +{"lr": 7.280440793665682e-05, "data_time": 0.6075129508972168, "loss": 0.23638832420110703, "time": 0.7537994384765625, "epoch": 4, "memory": 6318, "step": 15233} +{"lr": 7.248806597914234e-05, "data_time": 0.6015305519104004, "loss": 0.23821211755275726, "time": 0.7430759906768799, "epoch": 4, "memory": 6318, "step": 15333} +{"lr": 7.217070274248765e-05, "data_time": 0.6364526271820068, "loss": 0.24179704636335372, "time": 0.7770860671997071, "epoch": 4, "memory": 6318, "step": 15433} +{"lr": 7.185233676027203e-05, "data_time": 0.54974844455719, "loss": 0.23191624581813813, "time": 0.688976240158081, "epoch": 4, "memory": 6318, "step": 15533} +{"lr": 7.153298662463338e-05, "data_time": 0.6540096759796142, "loss": 0.23380161374807357, "time": 0.7934549570083618, "epoch": 4, "memory": 6318, "step": 15633} +{"lr": 7.121267098518291e-05, "data_time": 0.6486236810684204, "loss": 0.23264795392751694, "time": 0.7888083457946777, "epoch": 4, "memory": 6318, "step": 15733} +{"lr": 7.089140854791618e-05, "data_time": 0.9109420537948608, "loss": 0.2285044640302658, "time": 1.0505372047424317, "epoch": 4, "memory": 6318, "step": 15833} +{"lr": 7.056921807412025e-05, "data_time": 0.614934754371643, "loss": 0.25067051500082016, "time": 0.7553423166275024, "epoch": 4, "memory": 6318, "step": 15933} +{"lr": 7.024611837927802e-05, "data_time": 0.8851374626159668, "loss": 0.2302881136536598, "time": 1.0263184309005737, "epoch": 4, "memory": 6318, "step": 16033} +{"lr": 6.992212833196998e-05, "data_time": 0.6721847295761109, "loss": 0.22763000726699828, "time": 0.8127633094787597, "epoch": 4, "memory": 6318, "step": 16133} +{"lr": 6.959726685277174e-05, "data_time": 0.674868130683899, "loss": 0.238856540620327, "time": 0.8151495218276977, "epoch": 4, "memory": 6318, "step": 16233} +{"lr": 6.927155291314949e-05, "data_time": 0.6280160665512085, "loss": 0.2315930336713791, "time": 0.7679660320281982, "epoch": 4, "memory": 6318, "step": 16333} +{"lr": 6.894500553435204e-05, "data_time": 0.6877367496490479, "loss": 0.24604939222335814, "time": 0.8278838634490967, "epoch": 4, "memory": 6318, "step": 16433} +{"accuracy/top1": 84.82862854003906, "data_time": 0.3857512831687927, "time": 0.47472933928171795, "step": 4} +{"lr": 6.858158509719903e-05, "data_time": 0.5680527448654175, "loss": 0.24009820967912673, "time": 0.7082586765289307, "epoch": 5, "memory": 6318, "step": 16544} +{"lr": 6.825334178880171e-05, "data_time": 0.6665577173233033, "loss": 0.2341777890920639, "time": 0.8059651851654053, "epoch": 5, "memory": 6318, "step": 16644} +{"lr": 6.792432450336645e-05, "data_time": 0.6335298299789429, "loss": 0.23208148777484894, "time": 0.7742976188659668, "epoch": 5, "memory": 6318, "step": 16744} +{"lr": 6.759455245505286e-05, "data_time": 0.588647437095642, "loss": 0.23389839380979538, "time": 0.7285048723220825, "epoch": 5, "memory": 6318, "step": 16844} +{"lr": 6.726404490209753e-05, "data_time": 0.615578556060791, "loss": 0.23718324303627014, "time": 0.7584915161132812, "epoch": 5, "memory": 6318, "step": 16944} +{"lr": 6.693282114568972e-05, "data_time": 0.7293593645095825, "loss": 0.25018548518419265, "time": 0.8711184740066529, "epoch": 5, "memory": 6318, "step": 17044} +{"lr": 6.660090052884388e-05, "data_time": 0.6184142351150512, "loss": 0.23349623531103134, "time": 0.7589134693145752, "epoch": 5, "memory": 6318, "step": 17144} +{"lr": 6.626830243527016e-05, "data_time": 0.6342181921005249, "loss": 0.22983630746603012, "time": 0.7739503622055054, "epoch": 5, "memory": 6318, "step": 17244} +{"lr": 6.593504628824236e-05, "data_time": 0.7950935125350952, "loss": 0.24304037392139435, "time": 0.9357823133468628, "epoch": 5, "memory": 6318, "step": 17344} +{"lr": 6.560115154946387e-05, "data_time": 0.6720767736434936, "loss": 0.2416984662413597, "time": 0.8131884336471558, "epoch": 5, "memory": 6318, "step": 17444} +{"lr": 6.526663771793081e-05, "data_time": 0.6429623365402222, "loss": 0.23228186815977098, "time": 0.7892422437667846, "epoch": 5, "memory": 6318, "step": 17544} +{"lr": 6.493152432879363e-05, "data_time": 0.6183753252029419, "loss": 0.22265356332063674, "time": 0.7581408977508545, "epoch": 5, "memory": 6318, "step": 17644} +{"lr": 6.459583095221596e-05, "data_time": 0.5958486080169678, "loss": 0.23214069306850432, "time": 0.7560694456100464, "epoch": 5, "memory": 6318, "step": 17744} +{"lr": 6.425957719223197e-05, "data_time": 0.6039879560470581, "loss": 0.2288201406598091, "time": 0.7443809509277344, "epoch": 5, "memory": 6318, "step": 17844} +{"lr": 6.392278268560153e-05, "data_time": 0.6037734031677247, "loss": 0.2102500855922699, "time": 0.7444466590881348, "epoch": 5, "memory": 6318, "step": 17944} +{"lr": 6.358546710066316e-05, "data_time": 0.6427554130554199, "loss": 0.2321569338440895, "time": 0.7831363439559936, "epoch": 5, "memory": 6318, "step": 18044} +{"lr": 6.324765013618573e-05, "data_time": 0.6907007694244385, "loss": 0.22775797545909882, "time": 0.830433440208435, "epoch": 5, "memory": 6318, "step": 18144} +{"lr": 6.290935152021818e-05, "data_time": 0.6373366832733154, "loss": 0.23793013393878937, "time": 0.777033805847168, "epoch": 5, "memory": 6318, "step": 18244} +{"lr": 6.2570591008937e-05, "data_time": 0.6889883279800415, "loss": 0.24342462569475173, "time": 0.829011344909668, "epoch": 5, "memory": 6318, "step": 18344} +{"lr": 6.223138838549294e-05, "data_time": 0.6115456819534302, "loss": 0.22704339325428008, "time": 0.7518725395202637, "epoch": 5, "memory": 6318, "step": 18444} +{"lr": 6.18917634588553e-05, "data_time": 0.6836014986038208, "loss": 0.2423887297511101, "time": 0.824725341796875, "epoch": 5, "memory": 6318, "step": 18544} +{"lr": 6.155173606265555e-05, "data_time": 0.6356582641601562, "loss": 0.2345250889658928, "time": 0.7764936447143554, "epoch": 5, "memory": 6318, "step": 18644} +{"lr": 6.121132605402843e-05, "data_time": 0.7121028900146484, "loss": 0.2327351987361908, "time": 0.8519127845764161, "epoch": 5, "memory": 6318, "step": 18744} +{"lr": 6.087055331245287e-05, "data_time": 0.6103345394134522, "loss": 0.243044975399971, "time": 0.7622027635574341, "epoch": 5, "memory": 6318, "step": 18844} +{"lr": 6.05294377385911e-05, "data_time": 0.6303061723709107, "loss": 0.2202696442604065, "time": 0.770358681678772, "epoch": 5, "memory": 6318, "step": 18944} +{"lr": 6.0187999253126355e-05, "data_time": 0.6624517917633057, "loss": 0.21790529787540436, "time": 0.8022862672805786, "epoch": 5, "memory": 6318, "step": 19044} +{"lr": 5.9846257795599195e-05, "data_time": 0.6217154026031494, "loss": 0.22332927882671355, "time": 0.7627980232238769, "epoch": 5, "memory": 6318, "step": 19144} +{"lr": 5.950423332324355e-05, "data_time": 0.657962965965271, "loss": 0.23062596917152406, "time": 0.7977495431900025, "epoch": 5, "memory": 6318, "step": 19244} +{"lr": 5.9161945809820855e-05, "data_time": 0.6836307287216187, "loss": 0.2321133390069008, "time": 0.8246405601501465, "epoch": 5, "memory": 6318, "step": 19344} +{"lr": 5.8819415244453975e-05, "data_time": 0.61401846408844, "loss": 0.23368024080991745, "time": 0.7551644802093506, "epoch": 5, "memory": 6318, "step": 19444} +{"lr": 5.8476661630459405e-05, "data_time": 0.5882413148880005, "loss": 0.23067801892757417, "time": 0.7281119585037231, "epoch": 5, "memory": 6318, "step": 19544} +{"lr": 5.81337049841796e-05, "data_time": 0.6396611928939819, "loss": 0.23544860631227493, "time": 0.7821515083312989, "epoch": 5, "memory": 6318, "step": 19644} +{"lr": 5.779056533381373e-05, "data_time": 0.6922419548034668, "loss": 0.23260191977024078, "time": 0.8323213338851929, "epoch": 5, "memory": 6318, "step": 19744} +{"lr": 5.744726271824817e-05, "data_time": 0.837563443183899, "loss": 0.23904459774494172, "time": 0.9835986375808716, "epoch": 5, "memory": 6318, "step": 19844} +{"lr": 5.710381718588622e-05, "data_time": 0.6601776361465455, "loss": 0.2379686340689659, "time": 0.8006458520889282, "epoch": 5, "memory": 6318, "step": 19944} +{"lr": 5.676024879347741e-05, "data_time": 0.6204306125640869, "loss": 0.24066135734319688, "time": 0.7595683336257935, "epoch": 5, "memory": 6318, "step": 20044} +{"lr": 5.6416577604945966e-05, "data_time": 0.8849411010742188, "loss": 0.2382201671600342, "time": 1.026048755645752, "epoch": 5, "memory": 6318, "step": 20144} +{"lr": 5.6072823690219385e-05, "data_time": 0.677523922920227, "loss": 0.22010047286748885, "time": 0.8182397842407226, "epoch": 5, "memory": 6318, "step": 20244} +{"lr": 5.572900712405624e-05, "data_time": 0.6762327194213867, "loss": 0.237260402739048, "time": 0.8178808212280273, "epoch": 5, "memory": 6318, "step": 20344} +{"lr": 5.538514798487397e-05, "data_time": 0.6832957983016967, "loss": 0.23247854113578797, "time": 0.8243798732757568, "epoch": 5, "memory": 6318, "step": 20444} +{"lr": 5.5041266353576e-05, "data_time": 0.6764598846435547, "loss": 0.23345525115728377, "time": 0.8193856477737427, "epoch": 5, "memory": 6318, "step": 20544} +{"accuracy/top1": 88.0062255859375, "data_time": 0.40955922404925027, "time": 0.4967768351236979, "step": 5} +{"lr": 5.4659555783035384e-05, "data_time": 0.5640170097351074, "loss": 0.23523711860179902, "time": 0.8417809724807739, "epoch": 6, "memory": 6318, "step": 20655} +{"lr": 5.431569258426336e-05, "data_time": 0.7865960359573364, "loss": 0.2411901906132698, "time": 0.9359241247177124, "epoch": 6, "memory": 6318, "step": 20755} +{"lr": 5.397186934810864e-05, "data_time": 0.6889936923980713, "loss": 0.24351299554109573, "time": 0.829921817779541, "epoch": 6, "memory": 6318, "step": 20855} +{"lr": 5.362810615337815e-05, "data_time": 0.6160107612609863, "loss": 0.22347162514925004, "time": 0.7581508874893188, "epoch": 6, "memory": 6318, "step": 20955} +{"lr": 5.328442307537233e-05, "data_time": 0.7360903739929199, "loss": 0.23152901828289033, "time": 0.8770427703857422, "epoch": 6, "memory": 6318, "step": 21055} +{"lr": 5.294084018471303e-05, "data_time": 0.7281834840774536, "loss": 0.23047693073749542, "time": 0.8681220293045044, "epoch": 6, "memory": 6318, "step": 21155} +{"lr": 5.259737754617132e-05, "data_time": 0.6397636890411377, "loss": 0.2284434214234352, "time": 0.7799125432968139, "epoch": 6, "memory": 6318, "step": 21255} +{"lr": 5.2254055217495525e-05, "data_time": 0.6479471921920776, "loss": 0.23046401143074036, "time": 0.7879778861999511, "epoch": 6, "memory": 6318, "step": 21355} +{"lr": 5.1910893248240283e-05, "data_time": 0.6063809394836426, "loss": 0.23219931572675706, "time": 0.7710274934768677, "epoch": 6, "memory": 6318, "step": 21455} +{"lr": 5.156791167859532e-05, "data_time": 0.6770941495895386, "loss": 0.23874398320913315, "time": 0.8168701887130737, "epoch": 6, "memory": 6318, "step": 21555} +{"lr": 5.122513053821533e-05, "data_time": 0.6383856773376465, "loss": 0.22835891842842101, "time": 0.7785636186599731, "epoch": 6, "memory": 6318, "step": 21655} +{"lr": 5.0882569845050173e-05, "data_time": 0.6357599973678589, "loss": 0.22527429312467576, "time": 0.7761294603347778, "epoch": 6, "memory": 6318, "step": 21755} +{"lr": 5.054024960417592e-05, "data_time": 0.6140935897827149, "loss": 0.23473118543624877, "time": 0.7549921989440918, "epoch": 6, "memory": 6318, "step": 21855} +{"lr": 5.0198189806626494e-05, "data_time": 0.7070087909698486, "loss": 0.22903325110673906, "time": 0.84640531539917, "epoch": 6, "memory": 6318, "step": 21955} +{"lr": 4.985641042822647e-05, "data_time": 0.6718303918838501, "loss": 0.2397131696343422, "time": 0.8123040437698364, "epoch": 6, "memory": 6318, "step": 22055} +{"lr": 4.951493142842412e-05, "data_time": 0.6344693899154663, "loss": 0.2116445705294609, "time": 0.7747231245040893, "epoch": 6, "memory": 6318, "step": 22155} +{"lr": 4.917377274912603e-05, "data_time": 0.6452999591827393, "loss": 0.22026612162590026, "time": 0.7876631021499634, "epoch": 6, "memory": 6318, "step": 22255} +{"lr": 4.88329543135327e-05, "data_time": 0.7558466672897339, "loss": 0.2256220430135727, "time": 0.901628565788269, "epoch": 6, "memory": 6318, "step": 22355} +{"lr": 4.849249602497467e-05, "data_time": 0.7455522537231445, "loss": 0.20942132472991942, "time": 0.8870739459991455, "epoch": 6, "memory": 6318, "step": 22455} +{"lr": 4.8152417765750456e-05, "data_time": 0.7094234466552735, "loss": 0.23402302861213684, "time": 0.849983024597168, "epoch": 6, "memory": 6318, "step": 22555} +{"lr": 4.7812739395965355e-05, "data_time": 0.7074878692626954, "loss": 0.25134078711271285, "time": 0.848683524131775, "epoch": 6, "memory": 6318, "step": 22655} +{"lr": 4.74734807523717e-05, "data_time": 0.6698830127716064, "loss": 0.22743172198534012, "time": 0.817116093635559, "epoch": 6, "memory": 6318, "step": 22755} +{"lr": 4.7134661647210416e-05, "data_time": 0.6963460922241211, "loss": 0.2213045433163643, "time": 0.8369168519973755, "epoch": 6, "memory": 6318, "step": 22855} +{"lr": 4.679630186705388e-05, "data_time": 0.7061920642852784, "loss": 0.22865137755870818, "time": 0.8540581703186035, "epoch": 6, "memory": 6318, "step": 22955} +{"lr": 4.6458421171650644e-05, "data_time": 0.6461435317993164, "loss": 0.2223941832780838, "time": 0.7868811368942261, "epoch": 6, "memory": 6318, "step": 23055} +{"lr": 4.6121039292771295e-05, "data_time": 0.6812747955322266, "loss": 0.2251240998506546, "time": 0.8205606698989868, "epoch": 6, "memory": 6318, "step": 23155} +{"lr": 4.5784175933056197e-05, "data_time": 0.6692547798156738, "loss": 0.22888384461402894, "time": 0.8101490259170532, "epoch": 6, "memory": 6318, "step": 23255} +{"lr": 4.544785076486494e-05, "data_time": 0.6655930042266845, "loss": 0.22767900228500365, "time": 0.8060162305831909, "epoch": 6, "memory": 6318, "step": 23355} +{"lr": 4.5112083429127475e-05, "data_time": 0.7369659900665283, "loss": 0.22401324063539504, "time": 0.8776190996170044, "epoch": 6, "memory": 6318, "step": 23455} +{"lr": 4.477689353419706e-05, "data_time": 0.6090786695480347, "loss": 0.21689967960119247, "time": 0.7513005256652832, "epoch": 6, "memory": 6318, "step": 23555} +{"lr": 4.4442300654705266e-05, "data_time": 0.6490351438522339, "loss": 0.23096476197242738, "time": 0.7893428802490234, "epoch": 6, "memory": 6318, "step": 23655} +{"lr": 4.410832433041866e-05, "data_time": 0.6412917852401734, "loss": 0.21838558614253997, "time": 0.7809933662414551, "epoch": 6, "memory": 6318, "step": 23755} +{"lr": 4.377498406509792e-05, "data_time": 0.7388076543807983, "loss": 0.21265769451856614, "time": 0.8793262958526611, "epoch": 6, "memory": 6318, "step": 23855} +{"lr": 4.344229932535875e-05, "data_time": 0.6426203966140747, "loss": 0.2235260084271431, "time": 0.7837935209274292, "epoch": 6, "memory": 6318, "step": 23955} +{"lr": 4.3110289539535115e-05, "data_time": 0.6267694711685181, "loss": 0.23553937673568726, "time": 0.7675327062606812, "epoch": 6, "memory": 6318, "step": 24055} +{"lr": 4.2778974096544436e-05, "data_time": 0.6346272230148315, "loss": 0.2247397780418396, "time": 0.7754977941513062, "epoch": 6, "memory": 6318, "step": 24155} +{"lr": 4.2448372344755544e-05, "data_time": 0.7075176477432251, "loss": 0.20807492583990098, "time": 0.8477226257324219, "epoch": 6, "memory": 6318, "step": 24255} +{"lr": 4.211850359085871e-05, "data_time": 0.6136820793151856, "loss": 0.20074809491634368, "time": 0.756170129776001, "epoch": 6, "memory": 6318, "step": 24355} +{"lr": 4.17893870987381e-05, "data_time": 0.652216625213623, "loss": 0.2138784110546112, "time": 0.7941282510757446, "epoch": 6, "memory": 6318, "step": 24455} +{"lr": 4.146104208834679e-05, "data_time": 0.6623796224594116, "loss": 0.21141144633293152, "time": 0.8031316280364991, "epoch": 6, "memory": 6318, "step": 24555} +{"lr": 4.113348773458442e-05, "data_time": 0.6592641592025756, "loss": 0.19716128557920456, "time": 0.80677809715271, "epoch": 6, "memory": 6318, "step": 24655} +{"accuracy/top1": 90.61781311035156, "data_time": 0.40377671718597413, "time": 0.49216635823249816, "step": 6} +{"lr": 4.0770851520835164e-05, "data_time": 0.7648333787918091, "loss": 0.22227001190185547, "time": 0.927307391166687, "epoch": 7, "memory": 6318, "step": 24766} +{"lr": 4.044502815742039e-05, "data_time": 0.620558500289917, "loss": 0.2190093293786049, "time": 0.7607215404510498, "epoch": 7, "memory": 6318, "step": 24866} +{"lr": 4.012005478445896e-05, "data_time": 0.6529367446899415, "loss": 0.23349482268095018, "time": 0.7941431522369384, "epoch": 7, "memory": 6318, "step": 24966} +{"lr": 3.9795950379951394e-05, "data_time": 0.624491548538208, "loss": 0.24427032321691514, "time": 0.764388918876648, "epoch": 7, "memory": 6318, "step": 25066} +{"lr": 3.947273387115169e-05, "data_time": 0.7141632080078125, "loss": 0.22110608369112014, "time": 0.8562009572982788, "epoch": 7, "memory": 6318, "step": 25166} +{"lr": 3.9150424133462044e-05, "data_time": 0.686302661895752, "loss": 0.20895550549030303, "time": 0.8286733150482177, "epoch": 7, "memory": 6318, "step": 25266} +{"lr": 3.882903998933037e-05, "data_time": 0.6973209619522095, "loss": 0.23273418247699737, "time": 0.8395562648773194, "epoch": 7, "memory": 6318, "step": 25366} +{"lr": 3.8508600207151264e-05, "data_time": 0.6170834541320801, "loss": 0.21740475445985794, "time": 0.7590098142623901, "epoch": 7, "memory": 6318, "step": 25466} +{"lr": 3.818912350016978e-05, "data_time": 0.6640692234039307, "loss": 0.2343267560005188, "time": 0.807293438911438, "epoch": 7, "memory": 6318, "step": 25566} +{"lr": 3.787062852538883e-05, "data_time": 0.6382928371429444, "loss": 0.22212339490652083, "time": 0.779767370223999, "epoch": 7, "memory": 6318, "step": 25666} +{"lr": 3.7553133882479346e-05, "data_time": 0.6732906579971314, "loss": 0.22489983290433885, "time": 0.8138845443725586, "epoch": 7, "memory": 6318, "step": 25766} +{"lr": 3.723665811269437e-05, "data_time": 0.6524448156356811, "loss": 0.22993300557136537, "time": 0.7923453330993653, "epoch": 7, "memory": 6318, "step": 25866} +{"lr": 3.6921219697786176e-05, "data_time": 0.5943305969238282, "loss": 0.21163887828588485, "time": 0.7348302364349365, "epoch": 7, "memory": 6318, "step": 25966} +{"lr": 3.660683705892677e-05, "data_time": 0.6365231990814209, "loss": 0.2309290960431099, "time": 0.7787399530410767, "epoch": 7, "memory": 6318, "step": 26066} +{"lr": 3.6293528555632435e-05, "data_time": 0.6208684206008911, "loss": 0.23909600526094438, "time": 0.7614422798156738, "epoch": 7, "memory": 6318, "step": 26166} +{"lr": 3.598131248469134e-05, "data_time": 0.6362852573394775, "loss": 0.22502311319112778, "time": 0.7762999296188354, "epoch": 7, "memory": 6318, "step": 26266} +{"lr": 3.567020707909507e-05, "data_time": 0.6458447217941284, "loss": 0.20648026764392852, "time": 0.7862965345382691, "epoch": 7, "memory": 6318, "step": 26366} +{"lr": 3.5360230506973935e-05, "data_time": 0.6337474107742309, "loss": 0.21238310039043426, "time": 0.7749159574508667, "epoch": 7, "memory": 6318, "step": 26466} +{"lr": 3.5051400870535877e-05, "data_time": 0.6258433341979981, "loss": 0.2131961017847061, "time": 0.7664084196090698, "epoch": 7, "memory": 6318, "step": 26566} +{"lr": 3.47437362050094e-05, "data_time": 0.7146966218948364, "loss": 0.22229647785425186, "time": 0.8566596508026123, "epoch": 7, "memory": 6318, "step": 26666} +{"lr": 3.443725447759023e-05, "data_time": 0.6348914623260498, "loss": 0.22821670323610305, "time": 0.7756235599517822, "epoch": 7, "memory": 6318, "step": 26766} +{"lr": 3.413197358639224e-05, "data_time": 0.665582799911499, "loss": 0.2172778755426407, "time": 0.8062179327011109, "epoch": 7, "memory": 6318, "step": 26866} +{"lr": 3.382791135940194e-05, "data_time": 0.5825109958648682, "loss": 0.22656544148921967, "time": 0.722493028640747, "epoch": 7, "memory": 6318, "step": 26966} +{"lr": 3.35250855534377e-05, "data_time": 0.6492086887359619, "loss": 0.22972712516784669, "time": 0.7958221197128296, "epoch": 7, "memory": 6318, "step": 27066} +{"lr": 3.3223513853112476e-05, "data_time": 0.622047233581543, "loss": 0.20316445976495742, "time": 0.7735379219055176, "epoch": 7, "memory": 6318, "step": 27166} +{"lr": 3.292321386980115e-05, "data_time": 0.6232569217681885, "loss": 0.21836679428815842, "time": 0.7632628202438354, "epoch": 7, "memory": 6318, "step": 27266} +{"lr": 3.262420314061217e-05, "data_time": 0.5884154081344605, "loss": 0.2244155913591385, "time": 0.7284396409988403, "epoch": 7, "memory": 6318, "step": 27366} +{"lr": 3.2326499127363195e-05, "data_time": 0.5815781116485595, "loss": 0.21355385780334474, "time": 0.7286983013153077, "epoch": 7, "memory": 6318, "step": 27466} +{"lr": 3.2030119215561613e-05, "data_time": 0.630339789390564, "loss": 0.23119299113750458, "time": 0.7710340023040771, "epoch": 7, "memory": 6318, "step": 27566} +{"lr": 3.173508071338896e-05, "data_time": 0.647737455368042, "loss": 0.20573912113904952, "time": 0.7886793375015259, "epoch": 7, "memory": 6318, "step": 27666} +{"lr": 3.144140085069037e-05, "data_time": 0.5622187614440918, "loss": 0.20731066912412643, "time": 0.7015899658203125, "epoch": 7, "memory": 6318, "step": 27766} +{"lr": 3.114909677796814e-05, "data_time": 0.5976841926574707, "loss": 0.20905494689941406, "time": 0.741986083984375, "epoch": 7, "memory": 6318, "step": 27866} +{"lr": 3.085818556538059e-05, "data_time": 0.57134530544281, "loss": 0.20247323513031007, "time": 0.712349534034729, "epoch": 7, "memory": 6318, "step": 27966} +{"lr": 3.056868420174455e-05, "data_time": 0.6110909223556519, "loss": 0.21762798875570297, "time": 0.750453519821167, "epoch": 7, "memory": 6318, "step": 28066} +{"lr": 3.028060959354392e-05, "data_time": 0.6412607431411743, "loss": 0.2117123380303383, "time": 0.7819455146789551, "epoch": 7, "memory": 6318, "step": 28166} +{"lr": 2.9993978563941876e-05, "data_time": 0.5889158010482788, "loss": 0.2219421848654747, "time": 0.729128384590149, "epoch": 7, "memory": 6318, "step": 28266} +{"lr": 2.9708807851798587e-05, "data_time": 0.6377652168273926, "loss": 0.2252564862370491, "time": 0.7775874614715577, "epoch": 7, "memory": 6318, "step": 28366} +{"lr": 2.9425114110693717e-05, "data_time": 0.5933243751525878, "loss": 0.22750522345304489, "time": 0.7334260940551758, "epoch": 7, "memory": 6318, "step": 28466} +{"lr": 2.9142913907953844e-05, "data_time": 0.6078749895095825, "loss": 0.20678254663944245, "time": 0.748738431930542, "epoch": 7, "memory": 6318, "step": 28566} +{"lr": 2.8862223723684912e-05, "data_time": 0.6473680734634399, "loss": 0.22438707798719407, "time": 0.7887472629547119, "epoch": 7, "memory": 6318, "step": 28666} +{"lr": 2.8583059949809854e-05, "data_time": 0.6076606035232544, "loss": 0.2237881377339363, "time": 0.7471909284591675, "epoch": 7, "memory": 6318, "step": 28766} +{"accuracy/top1": 91.9087142944336, "data_time": 0.400511501232783, "time": 0.48955397804578143, "step": 7} +{"lr": 2.827499545208893e-05, "data_time": 0.6176312685012817, "loss": 0.21436824053525924, "time": 0.7585988759994506, "epoch": 8, "memory": 6318, "step": 28877} +{"lr": 2.7999105784987406e-05, "data_time": 0.6748634815216065, "loss": 0.22476212233304976, "time": 0.8150183200836182, "epoch": 8, "memory": 6318, "step": 28977} +{"lr": 2.772479293318711e-05, "data_time": 0.5957891464233398, "loss": 0.21923489421606063, "time": 0.7365003108978272, "epoch": 8, "memory": 6318, "step": 29077} +{"lr": 2.7452072916183603e-05, "data_time": 0.6443199872970581, "loss": 0.21840370297431946, "time": 0.7859854698181152, "epoch": 8, "memory": 6318, "step": 29177} +{"lr": 2.7180961660452883e-05, "data_time": 0.5866249084472657, "loss": 0.23032589703798295, "time": 0.7289427280426025, "epoch": 8, "memory": 6318, "step": 29277} +{"lr": 2.6911474998521622e-05, "data_time": 0.6299821853637695, "loss": 0.21991163194179536, "time": 0.7699019432067871, "epoch": 8, "memory": 6318, "step": 29377} +{"lr": 2.664362866804239e-05, "data_time": 0.6139799594879151, "loss": 0.21983575075864792, "time": 0.7538867473602295, "epoch": 8, "memory": 6318, "step": 29477} +{"lr": 2.637743831087459e-05, "data_time": 0.6002455472946167, "loss": 0.23171020299196243, "time": 0.7429261684417725, "epoch": 8, "memory": 6318, "step": 29577} +{"lr": 2.6112919472171114e-05, "data_time": 0.6951230287551879, "loss": 0.21728023290634155, "time": 0.8370468854904175, "epoch": 8, "memory": 6318, "step": 29677} +{"lr": 2.5850087599470434e-05, "data_time": 0.6004352331161499, "loss": 0.20709521770477296, "time": 0.7416739702224732, "epoch": 8, "memory": 6318, "step": 29777} +{"lr": 2.5588958041794503e-05, "data_time": 0.7258542537689209, "loss": 0.2137942522764206, "time": 0.8688105583190918, "epoch": 8, "memory": 6318, "step": 29877} +{"lr": 2.5329546048752453e-05, "data_time": 0.5389542818069458, "loss": 0.2041707828640938, "time": 0.68125319480896, "epoch": 8, "memory": 6318, "step": 29977} +{"lr": 2.5071866769649806e-05, "data_time": 0.591166615486145, "loss": 0.23005905896425247, "time": 0.7317116975784301, "epoch": 8, "memory": 6318, "step": 30077} +{"lr": 2.4815935252604222e-05, "data_time": 0.6030442476272583, "loss": 0.2234228640794754, "time": 0.7429580450057983, "epoch": 8, "memory": 6318, "step": 30177} +{"lr": 2.4561766443666384e-05, "data_time": 0.6336564779281616, "loss": 0.22076471745967866, "time": 0.7747917413711548, "epoch": 8, "memory": 6318, "step": 30277} +{"lr": 2.4309375185947164e-05, "data_time": 0.6094679117202759, "loss": 0.22254387885332108, "time": 0.7498175859451294, "epoch": 8, "memory": 6318, "step": 30377} +{"lr": 2.40587762187509e-05, "data_time": 0.6754100084304809, "loss": 0.22178202420473098, "time": 0.8154917001724243, "epoch": 8, "memory": 6318, "step": 30477} +{"lr": 2.3809984176714634e-05, "data_time": 0.573855972290039, "loss": 0.2264096185564995, "time": 0.7130426406860352, "epoch": 8, "memory": 6318, "step": 30577} +{"lr": 2.3563013588953406e-05, "data_time": 0.6098098993301392, "loss": 0.2345052808523178, "time": 0.7513066291809082, "epoch": 8, "memory": 6318, "step": 30677} +{"lr": 2.331787887821182e-05, "data_time": 0.5980265617370606, "loss": 0.22210220098495484, "time": 0.7409123182296753, "epoch": 8, "memory": 6318, "step": 30777} +{"lr": 2.307459436002179e-05, "data_time": 0.6060364723205567, "loss": 0.22102549225091933, "time": 0.7459308624267578, "epoch": 8, "memory": 6318, "step": 30877} +{"lr": 2.2833174241866527e-05, "data_time": 0.6350964546203614, "loss": 0.22628044486045837, "time": 0.7767293930053711, "epoch": 8, "memory": 6318, "step": 30977} +{"lr": 2.259363262235086e-05, "data_time": 0.6353445768356323, "loss": 0.20365117043256759, "time": 0.7762448072433472, "epoch": 8, "memory": 6318, "step": 31077} +{"lr": 2.2355983490377735e-05, "data_time": 0.6199627637863159, "loss": 0.2064829558134079, "time": 0.7605207681655883, "epoch": 8, "memory": 6318, "step": 31177} +{"lr": 2.2120240724331546e-05, "data_time": 0.680437421798706, "loss": 0.2050203114748001, "time": 0.8336479902267456, "epoch": 8, "memory": 6318, "step": 31277} +{"lr": 2.1886418091267445e-05, "data_time": 0.5961122035980224, "loss": 0.2049966424703598, "time": 0.7368704319000244, "epoch": 8, "memory": 6318, "step": 31377} +{"lr": 2.1654529246107496e-05, "data_time": 0.6269190549850464, "loss": 0.20520494729280472, "time": 0.7749230623245239, "epoch": 8, "memory": 6318, "step": 31477} +{"lr": 2.1424587730843184e-05, "data_time": 0.6658232927322387, "loss": 0.1943259358406067, "time": 0.8055380582809448, "epoch": 8, "memory": 6318, "step": 31577} +{"lr": 2.1196606973744593e-05, "data_time": 0.6407031774520874, "loss": 0.2201676696538925, "time": 0.7820532083511352, "epoch": 8, "memory": 6318, "step": 31677} +{"lr": 2.0970600288576148e-05, "data_time": 0.6111618041992187, "loss": 0.21097415089607238, "time": 0.7509358882904053, "epoch": 8, "memory": 6318, "step": 31777} +{"lr": 2.0746580873819274e-05, "data_time": 0.6296913146972656, "loss": 0.21080616861581802, "time": 0.770757794380188, "epoch": 8, "memory": 6318, "step": 31877} +{"lr": 2.0524561811901426e-05, "data_time": 0.6054296016693115, "loss": 0.20179010927677155, "time": 0.745720100402832, "epoch": 8, "memory": 6318, "step": 31977} +{"lr": 2.0304556068432268e-05, "data_time": 0.6531343698501587, "loss": 0.20696854442358018, "time": 0.7943596124649048, "epoch": 8, "memory": 6318, "step": 32077} +{"lr": 2.0086576491446333e-05, "data_time": 0.6205193281173706, "loss": 0.20936913937330245, "time": 0.7608472824096679, "epoch": 8, "memory": 6318, "step": 32177} +{"lr": 1.987063581065291e-05, "data_time": 0.5727409362792969, "loss": 0.22898492515087127, "time": 0.7134162664413453, "epoch": 8, "memory": 6318, "step": 32277} +{"lr": 1.965674663669245e-05, "data_time": 0.603482699394226, "loss": 0.20578554421663284, "time": 0.7432650804519654, "epoch": 8, "memory": 6318, "step": 32377} +{"lr": 1.9444921460400247e-05, "data_time": 0.5591931104660034, "loss": 0.23926617801189423, "time": 0.7021947622299194, "epoch": 8, "memory": 6318, "step": 32477} +{"lr": 1.923517265207697e-05, "data_time": 0.5848503351211548, "loss": 0.21970323771238326, "time": 0.7247686624526978, "epoch": 8, "memory": 6318, "step": 32577} +{"lr": 1.9027512460766204e-05, "data_time": 0.7181018590927124, "loss": 0.21240825355052947, "time": 0.8582144021987915, "epoch": 8, "memory": 6318, "step": 32677} +{"lr": 1.8821953013539136e-05, "data_time": 0.634683895111084, "loss": 0.20308539122343064, "time": 0.7843214511871338, "epoch": 8, "memory": 6318, "step": 32777} +{"lr": 1.8618506314786433e-05, "data_time": 0.5956550598144531, "loss": 0.19921929389238358, "time": 0.7360635042190552, "epoch": 8, "memory": 6318, "step": 32877} +{"accuracy/top1": 92.75276184082031, "data_time": 0.39636743664741514, "time": 0.48426974217096963, "step": 8} +{"lr": 1.8395169032542028e-05, "data_time": 0.6216209888458252, "loss": 0.20786677449941635, "time": 0.7618060827255249, "epoch": 9, "memory": 6318, "step": 32988} +{"lr": 1.819621906460573e-05, "data_time": 0.6423932075500488, "loss": 0.20277125090360643, "time": 0.7839249849319458, "epoch": 9, "memory": 6318, "step": 33088} +{"lr": 1.799941838715077e-05, "data_time": 0.628791069984436, "loss": 0.20579521358013153, "time": 0.7687147617340088, "epoch": 9, "memory": 6318, "step": 33188} +{"lr": 1.7804778493067625e-05, "data_time": 0.6096827507019043, "loss": 0.20695388615131377, "time": 0.7503035306930542, "epoch": 9, "memory": 6318, "step": 33288} +{"lr": 1.761231074905996e-05, "data_time": 0.570347261428833, "loss": 0.2184254989027977, "time": 0.7099888563156128, "epoch": 9, "memory": 6318, "step": 33388} +{"lr": 1.742202639498083e-05, "data_time": 0.6435333251953125, "loss": 0.20907920897006987, "time": 0.7838011980056763, "epoch": 9, "memory": 6318, "step": 33488} +{"lr": 1.7233936543176347e-05, "data_time": 0.607920479774475, "loss": 0.24120692610740663, "time": 0.7484342098236084, "epoch": 9, "memory": 6318, "step": 33588} +{"lr": 1.704805217783665e-05, "data_time": 0.6213541507720948, "loss": 0.20577476918697357, "time": 0.7618390083312988, "epoch": 9, "memory": 6318, "step": 33688} +{"lr": 1.68643841543545e-05, "data_time": 0.6261132478713989, "loss": 0.22067376375198364, "time": 0.7667516708374024, "epoch": 9, "memory": 6318, "step": 33788} +{"lr": 1.6682943198691327e-05, "data_time": 0.5971660375595093, "loss": 0.20701639652252196, "time": 0.7378833055496216, "epoch": 9, "memory": 6318, "step": 33888} +{"lr": 1.6503739906750822e-05, "data_time": 0.655132532119751, "loss": 0.20590871721506118, "time": 0.7950443029403687, "epoch": 9, "memory": 6318, "step": 33988} +{"lr": 1.6326784743760187e-05, "data_time": 0.9051164388656616, "loss": 0.19997305124998094, "time": 1.0440760612487794, "epoch": 9, "memory": 6318, "step": 34088} +{"lr": 1.6152088043659013e-05, "data_time": 0.6061965942382812, "loss": 0.20650468170642852, "time": 0.7460791587829589, "epoch": 9, "memory": 6318, "step": 34188} +{"lr": 1.59796600084957e-05, "data_time": 0.6238481998443604, "loss": 0.21854153722524644, "time": 0.7663270473480225, "epoch": 9, "memory": 6318, "step": 34288} +{"lr": 1.5809510707831687e-05, "data_time": 0.5700688362121582, "loss": 0.2022887408733368, "time": 0.7177820920944213, "epoch": 9, "memory": 6318, "step": 34388} +{"lr": 1.5641650078153473e-05, "data_time": 0.5865361452102661, "loss": 0.21612299233675003, "time": 0.7258360862731934, "epoch": 9, "memory": 6318, "step": 34488} +{"lr": 1.547608792229235e-05, "data_time": 0.617069149017334, "loss": 0.21610855162143708, "time": 0.7565063953399658, "epoch": 9, "memory": 6318, "step": 34588} +{"lr": 1.5312833908851794e-05, "data_time": 0.5972658872604371, "loss": 0.213951775431633, "time": 0.7376068115234375, "epoch": 9, "memory": 6318, "step": 34688} +{"lr": 1.5151897571642958e-05, "data_time": 0.6320751428604126, "loss": 0.22815759629011154, "time": 0.7721969366073609, "epoch": 9, "memory": 6318, "step": 34788} +{"lr": 1.4993288309127877e-05, "data_time": 0.5628451347351074, "loss": 0.21320310980081558, "time": 0.7084277391433715, "epoch": 9, "memory": 6318, "step": 34888} +{"lr": 1.4837015383870593e-05, "data_time": 0.5949024200439453, "loss": 0.22500751316547393, "time": 0.7355358600616455, "epoch": 9, "memory": 6318, "step": 34988} +{"lr": 1.4683087921996264e-05, "data_time": 0.591160774230957, "loss": 0.21934388875961303, "time": 0.7318089008331299, "epoch": 9, "memory": 6318, "step": 35088} +{"lr": 1.453151491265819e-05, "data_time": 0.6077698707580567, "loss": 0.23041684776544571, "time": 0.7496355295181274, "epoch": 9, "memory": 6318, "step": 35188} +{"lr": 1.4382305207512847e-05, "data_time": 1.101082968711853, "loss": 0.22867132276296614, "time": 1.251568865776062, "epoch": 9, "memory": 6318, "step": 35288} +{"lr": 1.423546752020298e-05, "data_time": 0.6855479717254639, "loss": 0.2133968412876129, "time": 0.8279071569442749, "epoch": 9, "memory": 6318, "step": 35388} +{"lr": 1.4091010425848762e-05, "data_time": 0.6148358345031738, "loss": 0.20738937109708785, "time": 0.7561617374420166, "epoch": 9, "memory": 6318, "step": 35488} +{"lr": 1.3948942360546966e-05, "data_time": 0.5825257539749146, "loss": 0.22857085168361663, "time": 0.7231127262115479, "epoch": 9, "memory": 6318, "step": 35588} +{"lr": 1.3809271620878346e-05, "data_time": 0.6417038917541504, "loss": 0.1929815486073494, "time": 0.782297158241272, "epoch": 9, "memory": 6318, "step": 35688} +{"lr": 1.3672006363423091e-05, "data_time": 0.6662217140197754, "loss": 0.2063335046172142, "time": 0.8071076154708863, "epoch": 9, "memory": 6318, "step": 35788} +{"lr": 1.3537154604284566e-05, "data_time": 0.6649338245391846, "loss": 0.21065299808979035, "time": 0.811578345298767, "epoch": 9, "memory": 6318, "step": 35888} +{"lr": 1.3404724218621087e-05, "data_time": 0.6290776014328003, "loss": 0.23351238667964935, "time": 0.7699763774871826, "epoch": 9, "memory": 6318, "step": 35988} +{"lr": 1.3274722940186121e-05, "data_time": 0.6371079206466674, "loss": 0.2093411773443222, "time": 0.7780585765838623, "epoch": 9, "memory": 6318, "step": 36088} +{"lr": 1.3147158360876541e-05, "data_time": 0.6383229255676269, "loss": 0.21968120634555816, "time": 0.7800687074661254, "epoch": 9, "memory": 6318, "step": 36188} +{"lr": 1.3022037930289352e-05, "data_time": 0.6223465204238892, "loss": 0.20273312032222748, "time": 0.7635986328125, "epoch": 9, "memory": 6318, "step": 36288} +{"lr": 1.2899368955286627e-05, "data_time": 0.5478759050369263, "loss": 0.21345443278551102, "time": 0.6884966850280761, "epoch": 9, "memory": 6318, "step": 36388} +{"lr": 1.2779158599568751e-05, "data_time": 0.5945204973220826, "loss": 0.2066677615046501, "time": 0.7352986812591553, "epoch": 9, "memory": 6318, "step": 36488} +{"lr": 1.2661413883256134e-05, "data_time": 0.6183670043945313, "loss": 0.20916925817728044, "time": 0.7594605922698975, "epoch": 9, "memory": 6318, "step": 36588} +{"lr": 1.2546141682479203e-05, "data_time": 0.6569263219833374, "loss": 0.21101305931806563, "time": 0.7965500831604004, "epoch": 9, "memory": 6318, "step": 36688} +{"lr": 1.2433348728976846e-05, "data_time": 0.6889534711837768, "loss": 0.21142506301403047, "time": 0.8309194564819335, "epoch": 9, "memory": 6318, "step": 36788} +{"lr": 1.2323041609703344e-05, "data_time": 0.625512957572937, "loss": 0.21078932136297227, "time": 0.7670870304107666, "epoch": 9, "memory": 6318, "step": 36888} +{"lr": 1.2215226766443614e-05, "data_time": 0.6039667367935181, "loss": 0.21911162734031678, "time": 0.7465001583099365, "epoch": 9, "memory": 6318, "step": 36988} +{"accuracy/top1": 93.26911926269531, "data_time": 0.39996779163678486, "time": 0.48942285776138306, "step": 9} +{"lr": 1.209847850894437e-05, "data_time": 0.5794030666351319, "loss": 0.21785298883914947, "time": 0.7185676097869873, "epoch": 10, "memory": 6318, "step": 37099} +{"lr": 1.1995942849201055e-05, "data_time": 0.6175326347351074, "loss": 0.21775296032428743, "time": 0.758519983291626, "epoch": 10, "memory": 6318, "step": 37199} +{"lr": 1.189591856759183e-05, "data_time": 0.6041690826416015, "loss": 0.21178694516420365, "time": 0.7452110290527344, "epoch": 10, "memory": 6318, "step": 37299} +{"lr": 1.1798411505398016e-05, "data_time": 0.5669848680496216, "loss": 0.2084959164261818, "time": 0.7071744918823242, "epoch": 10, "memory": 6318, "step": 37399} +{"lr": 1.1703427356898678e-05, "data_time": 0.5915694236755371, "loss": 0.2231340527534485, "time": 0.7323443651199341, "epoch": 10, "memory": 6318, "step": 37499} +{"lr": 1.1610971669038223e-05, "data_time": 0.6481956958770752, "loss": 0.20284702330827714, "time": 0.7913770198822021, "epoch": 10, "memory": 6318, "step": 37599} +{"lr": 1.1521049841102396e-05, "data_time": 0.6005637645721436, "loss": 0.20713708102703093, "time": 0.7401561498641968, "epoch": 10, "memory": 6318, "step": 37699} +{"lr": 1.1433667124402997e-05, "data_time": 0.575803804397583, "loss": 0.19673682302236556, "time": 0.7175272226333618, "epoch": 10, "memory": 6318, "step": 37799} +{"lr": 1.1348828621971181e-05, "data_time": 0.648768138885498, "loss": 0.21565770655870437, "time": 0.790831732749939, "epoch": 10, "memory": 6318, "step": 37899} +{"lr": 1.1266539288259486e-05, "data_time": 0.627412486076355, "loss": 0.21127624362707137, "time": 0.7675125360488891, "epoch": 10, "memory": 6318, "step": 37999} +{"lr": 1.1186803928852477e-05, "data_time": 0.596336054801941, "loss": 0.20482346266508103, "time": 0.7371329069137573, "epoch": 10, "memory": 6318, "step": 38099} +{"lr": 1.1109627200186125e-05, "data_time": 0.6115892171859741, "loss": 0.1991797998547554, "time": 0.7520804166793823, "epoch": 10, "memory": 6318, "step": 38199} +{"lr": 1.1035013609275835e-05, "data_time": 0.6719270706176758, "loss": 0.21726299673318863, "time": 0.8128053426742554, "epoch": 10, "memory": 6318, "step": 38299} +{"lr": 1.0962967513453298e-05, "data_time": 0.5536705493927002, "loss": 0.21128429919481279, "time": 0.6952198266983032, "epoch": 10, "memory": 6318, "step": 38399} +{"lr": 1.0893493120111999e-05, "data_time": 0.6330197334289551, "loss": 0.20664433985948563, "time": 0.7728255987167358, "epoch": 10, "memory": 6318, "step": 38499} +{"lr": 1.0826594486461509e-05, "data_time": 0.6278717279434204, "loss": 0.20064590871334076, "time": 0.7669581413269043, "epoch": 10, "memory": 6318, "step": 38599} +{"lr": 1.0762275519290557e-05, "data_time": 0.6420759916305542, "loss": 0.2029467850923538, "time": 0.7832370281219483, "epoch": 10, "memory": 6318, "step": 38699} +{"lr": 1.0700539974738862e-05, "data_time": 0.608813214302063, "loss": 0.2126883715391159, "time": 0.7497075319290161, "epoch": 10, "memory": 6318, "step": 38799} +{"lr": 1.064139145807783e-05, "data_time": 0.6813237428665161, "loss": 0.2037241443991661, "time": 0.8232875108718872, "epoch": 10, "memory": 6318, "step": 38899} +{"lr": 1.0584833423499947e-05, "data_time": 0.6327476739883423, "loss": 0.18783843368291855, "time": 0.7735593795776368, "epoch": 10, "memory": 6318, "step": 38999} +{"lr": 1.0530869173917099e-05, "data_time": 0.6954510688781739, "loss": 0.1974943086504936, "time": 0.8368297815322876, "epoch": 10, "memory": 6318, "step": 39099} +{"lr": 1.0479501860767654e-05, "data_time": 0.6410568475723266, "loss": 0.21395813673734665, "time": 0.7825519323349, "epoch": 10, "memory": 6318, "step": 39199} +{"lr": 1.043073448383247e-05, "data_time": 0.5959242582321167, "loss": 0.22101271003484727, "time": 0.7487569332122803, "epoch": 10, "memory": 6318, "step": 39299} +{"lr": 1.038456989105968e-05, "data_time": 0.639715600013733, "loss": 0.21259773075580596, "time": 0.7799566745758056, "epoch": 10, "memory": 6318, "step": 39399} +{"lr": 1.034101077839838e-05, "data_time": 0.5884696006774902, "loss": 0.22869202941656114, "time": 0.7296674251556396, "epoch": 10, "memory": 6318, "step": 39499} +{"lr": 1.0300059689641194e-05, "data_time": 0.6050638914108276, "loss": 0.21708722412586212, "time": 0.746091651916504, "epoch": 10, "memory": 6318, "step": 39599} +{"lr": 1.0261719016275683e-05, "data_time": 0.635635232925415, "loss": 0.1917801320552826, "time": 0.7758737802505493, "epoch": 10, "memory": 6318, "step": 39699} +{"lr": 1.0225990997344757e-05, "data_time": 0.6306647062301636, "loss": 0.20383365154266359, "time": 0.7720104932785035, "epoch": 10, "memory": 6318, "step": 39799} +{"lr": 1.019287771931587e-05, "data_time": 0.6178133249282837, "loss": 0.20559848845005035, "time": 0.760369873046875, "epoch": 10, "memory": 6318, "step": 39899} +{"lr": 1.0162381115959187e-05, "data_time": 0.5978616952896119, "loss": 0.19369159936904906, "time": 0.7390765190124512, "epoch": 10, "memory": 6318, "step": 39999} +{"lr": 1.0134502968234636e-05, "data_time": 0.6582518815994263, "loss": 0.20472228899598122, "time": 0.7998521566390991, "epoch": 10, "memory": 6318, "step": 40099} +{"lr": 1.0109244904187918e-05, "data_time": 0.5787601232528686, "loss": 0.19884382784366608, "time": 0.7180487394332886, "epoch": 10, "memory": 6318, "step": 40199} +{"lr": 1.0086608398855446e-05, "data_time": 0.6692479372024536, "loss": 0.22374869287014007, "time": 0.8103873252868652, "epoch": 10, "memory": 6318, "step": 40299} +{"lr": 1.0066594774178176e-05, "data_time": 0.6905025959014892, "loss": 0.20124364793300628, "time": 0.8318718194961547, "epoch": 10, "memory": 6318, "step": 40399} +{"lr": 1.0049205198924413e-05, "data_time": 0.9919116258621216, "loss": 0.20201308280229568, "time": 1.1397632598876952, "epoch": 10, "memory": 6318, "step": 40499} +{"lr": 1.0034440688621577e-05, "data_time": 0.5915333747863769, "loss": 0.20739715546369553, "time": 0.7321382522583008, "epoch": 10, "memory": 6318, "step": 40599} +{"lr": 1.0022302105496896e-05, "data_time": 0.607059907913208, "loss": 0.20587428659200668, "time": 0.7482892751693726, "epoch": 10, "memory": 6318, "step": 40699} +{"lr": 1.0012790158427017e-05, "data_time": 0.7182532787322998, "loss": 0.21151627898216246, "time": 0.8581497192382812, "epoch": 10, "memory": 6318, "step": 40799} +{"lr": 1.000590540289662e-05, "data_time": 0.6576358079910278, "loss": 0.21189197301864623, "time": 0.8053101062774658, "epoch": 10, "memory": 6318, "step": 40899} +{"lr": 1.0001648240966046e-05, "data_time": 0.6686748504638672, "loss": 0.21074877828359603, "time": 0.8098900318145752, "epoch": 10, "memory": 6318, "step": 40999} +{"lr": 1.0000018921247718e-05, "data_time": 0.6262905836105347, "loss": 0.19762154221534728, "time": 0.7657563924789429, "epoch": 10, "memory": 6318, "step": 41099} +{"accuracy/top1": 93.51074981689453, "data_time": 0.3998194396495819, "time": 0.48767611185709636, "step": 10} diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/0000000.png_1.png b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/0000000.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..fd63c327e732cb3cbc1400dd713b4b1e5b0250d1 Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/0000000.png_1.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/0000000.png_10.png b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/0000000.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..021d94dae1c3eff45542920068625c258ef3f9e4 Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/0000000.png_10.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/0000000.png_2.png b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/0000000.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..922a5cf44e40d7de8afe46aaced6e8d0cd0a4d99 Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/0000000.png_2.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/0000000.png_3.png b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/0000000.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..3af84b9ed1e569b656e68f03145ba17c7ee020e2 Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/0000000.png_3.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/0000000.png_4.png b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/0000000.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..2a7338aeaa11f582f52f9227b289e3cff78d679e Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/0000000.png_4.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/0000000.png_5.png b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/0000000.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..f8d698c194001fd929d8227efc072d79209ec7b5 Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/0000000.png_5.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/0000000.png_6.png b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/0000000.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..464feb36faabfa4ff2c89b7bab94d5899475fb61 Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/0000000.png_6.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/0000000.png_7.png b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/0000000.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..ce883668e7c206be24cb62142b9ffa1b20f64735 Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/0000000.png_7.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/0000000.png_8.png b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/0000000.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..f79105ab2be57b6a00da49f26e68d426dfe280af Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/0000000.png_8.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/0000000.png_9.png b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/0000000.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..ab3c43396cc669c37d94b04dd7042997e01ba092 Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/0000000.png_9.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/000003093.jpg_1.png b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/000003093.jpg_1.png new file mode 100644 index 0000000000000000000000000000000000000000..f67405ac603a1d10d5c7cf761dc38ee59ba0d303 Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/000003093.jpg_1.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/000003093.jpg_10.png b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/000003093.jpg_10.png new file mode 100644 index 0000000000000000000000000000000000000000..58c2f7d2766853d99b4dee1b168f63da4423356a Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/000003093.jpg_10.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/000003093.jpg_2.png b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/000003093.jpg_2.png new file mode 100644 index 0000000000000000000000000000000000000000..d26a0c732a25871055bbb61774a8cdd1ec1939b2 Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/000003093.jpg_2.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/000003093.jpg_3.png b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/000003093.jpg_3.png new file mode 100644 index 0000000000000000000000000000000000000000..c8d2762b1de44963e7a18a1208de460c4a3523bb Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/000003093.jpg_3.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/000003093.jpg_4.png b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/000003093.jpg_4.png new file mode 100644 index 0000000000000000000000000000000000000000..984c8730e6abda6f8892deb277621d8aa498727f Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/000003093.jpg_4.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/000003093.jpg_5.png b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/000003093.jpg_5.png new file mode 100644 index 0000000000000000000000000000000000000000..34b740a348c26b22e3c4bdec4eb6565735366b9b Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/000003093.jpg_5.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/000003093.jpg_6.png b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/000003093.jpg_6.png new file mode 100644 index 0000000000000000000000000000000000000000..1c71aecbee6d6cf2de91a45718bc67b496736045 Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/000003093.jpg_6.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/000003093.jpg_7.png b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/000003093.jpg_7.png new file mode 100644 index 0000000000000000000000000000000000000000..cee83c922e25fdb62b63515e10c8332815dc1f60 Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/000003093.jpg_7.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/000003093.jpg_8.png b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/000003093.jpg_8.png new file mode 100644 index 0000000000000000000000000000000000000000..e0a3d9bc2664e1d55a26dcf682bbe351f2ff05c3 Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/000003093.jpg_8.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/000003093.jpg_9.png b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/000003093.jpg_9.png new file mode 100644 index 0000000000000000000000000000000000000000..70ffc605fda869d4ed74b82c08f65021722d1329 Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/000003093.jpg_9.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/000015337.jpg_1.png b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/000015337.jpg_1.png new file mode 100644 index 0000000000000000000000000000000000000000..68044bce31269a3f25706d410b2a4c3c7c738122 Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/000015337.jpg_1.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/000015337.jpg_10.png b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/000015337.jpg_10.png new file mode 100644 index 0000000000000000000000000000000000000000..b2c8870ddea8c4e681ac4953ff1bf0a60a420335 Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/000015337.jpg_10.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/000015337.jpg_2.png b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/000015337.jpg_2.png new file mode 100644 index 0000000000000000000000000000000000000000..07515ef417e6440330a1a96d2d61f749913ed108 Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/000015337.jpg_2.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/000015337.jpg_3.png b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/000015337.jpg_3.png new file mode 100644 index 0000000000000000000000000000000000000000..65b1503b113077c42238070e03db75419c880254 Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/000015337.jpg_3.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/000015337.jpg_4.png b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/000015337.jpg_4.png new file mode 100644 index 0000000000000000000000000000000000000000..264aea4b04283382dd06f98d9110fdc93017e2e2 Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/000015337.jpg_4.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/000015337.jpg_5.png b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/000015337.jpg_5.png new file mode 100644 index 0000000000000000000000000000000000000000..457f323248c944eec4e90d1fcb314384a4a8638b Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/000015337.jpg_5.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/000015337.jpg_6.png b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/000015337.jpg_6.png new file mode 100644 index 0000000000000000000000000000000000000000..c6942ae45f883d0856536e44522f37fe5ac4af48 Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/000015337.jpg_6.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/000015337.jpg_7.png b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/000015337.jpg_7.png new file mode 100644 index 0000000000000000000000000000000000000000..8252787ca572de855a3c49cb3e2ae65a31d40d8c Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/000015337.jpg_7.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/000015337.jpg_8.png b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/000015337.jpg_8.png new file mode 100644 index 0000000000000000000000000000000000000000..f9fa984167c66e9ed5e6f272fae71824ad47eadd Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/000015337.jpg_8.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/000015337.jpg_9.png b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/000015337.jpg_9.png new file mode 100644 index 0000000000000000000000000000000000000000..5841fdd0fe1009d9a1fc035003673415521022e6 Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/000015337.jpg_9.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/0004159.png_1.png b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/0004159.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..4356a177a208c51e7948c63c6cf4d5ac39b2a9dd Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/0004159.png_1.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/0004159.png_10.png b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/0004159.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..f0378b7a973e384a0e70f68a56aaa79fe4ba6707 Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/0004159.png_10.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/0004159.png_2.png b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/0004159.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..4bd3bad6362fc91d9c2627fc7ef46803975c7c03 Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/0004159.png_2.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/0004159.png_3.png b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/0004159.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..d7d7234d8c2f3364d2298078d2727b909b6d910d Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/0004159.png_3.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/0004159.png_4.png b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/0004159.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..b82d7721ba81f6602b797bcb5c0e701a211d49d4 Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/0004159.png_4.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/0004159.png_5.png b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/0004159.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..519dc22972e5381294c5373767634480473bc519 Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/0004159.png_5.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/0004159.png_6.png b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/0004159.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..b01daeae4f9c2e6f58f49a3bd9252451c0f4f9f1 Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/0004159.png_6.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/0004159.png_7.png b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/0004159.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..232d039f334872bbc569428c93954931d2d2da8f Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/0004159.png_7.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/0004159.png_8.png b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/0004159.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..302224ec1ea9a1a016234146ea3d04e160a476a2 Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/0004159.png_8.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/0004159.png_9.png b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/0004159.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..418a90933f53e0abd72ba607ff327e87234b2265 Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/0004159.png_9.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/0008318.png_1.png b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/0008318.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..6f10b26a05d1200793cba1c996345c03ff1637ca Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/0008318.png_1.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/0008318.png_10.png b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/0008318.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..2310ba606a0868ff590e6bcf983379e4999e5e23 Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/0008318.png_10.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/0008318.png_2.png b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/0008318.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..feebff060af4c9b7a6899eb627800fb96ad0881b Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/0008318.png_2.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/0008318.png_3.png b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/0008318.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..cde88115c44ed30f745822b2a2d57e39d946d862 Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/0008318.png_3.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/0008318.png_4.png b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/0008318.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..5eace8938a707fd56f7d3088f6360781743af7a1 Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/0008318.png_4.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/0008318.png_5.png b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/0008318.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..cc22344d3772f632f43357d811e701cb9b806337 Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/0008318.png_5.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/0008318.png_6.png b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/0008318.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..8d6349df770bb8a93900e3276aa27f3895aec515 Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/0008318.png_6.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/0008318.png_7.png b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/0008318.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..8f3dc50897d6dc16277959542b6cd7e5dbef44bf Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/0008318.png_7.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/0008318.png_8.png b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/0008318.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..9ba526ea9244de74a296f205259e35886fcf686b Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/0008318.png_8.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/0008318.png_9.png b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/0008318.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..ff6afba917579a48940aa09b02dc104bad1e21a6 Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/0008318.png_9.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/0010000.png_1.png b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/0010000.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..5cbdc42008497fa777946de0b7bfa426d1325f18 Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/0010000.png_1.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/0010000.png_10.png b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/0010000.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..077c5a3505a382b02a039d574f7e3acdadf247bb Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/0010000.png_10.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/0010000.png_2.png b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/0010000.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..6901c53c45c1dd1cfd74765e4c1ffcce57fdd738 Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/0010000.png_2.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/0010000.png_3.png b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/0010000.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..97fbe983996f043da4e1ad6628930056d4f551f3 Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/0010000.png_3.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/0010000.png_4.png b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/0010000.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..6d20cf08ab84bf064e61d1cd67cd388fa6357b6d Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/0010000.png_4.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/0010000.png_5.png b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/0010000.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..bfd09e9756d134a1a1b4e731a308a06f435ecbd4 Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/0010000.png_5.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/0010000.png_6.png b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/0010000.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..1ad3633397802393b17d10b09a0153434aa31a2d Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/0010000.png_6.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/0010000.png_7.png b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/0010000.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..01d569e6cab497ac6660cb421d2d2a144d51a4e2 Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/0010000.png_7.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/0010000.png_8.png b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/0010000.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..4255180cb090c2a2ed590307c576a05be46bde20 Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/0010000.png_8.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/0010000.png_9.png b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/0010000.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..67322ef79e3ef0f72a569de11373d2bfdf9fee6b Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/0010000.png_9.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/0014159.png_1.png b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/0014159.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..8ae314e00b97c33503b42848415380aeeab02a0a Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/0014159.png_1.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/0014159.png_10.png b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/0014159.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..8f4a68bcad8f6744930c4da0eee20d85a2619b3a Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/0014159.png_10.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/0014159.png_2.png b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/0014159.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..446a52892b41fe18c3221504e1645ff659519603 Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/0014159.png_2.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/0014159.png_3.png b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/0014159.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..3e98995d94c3a0ffef39a581d7dc8e179deb2e5a Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/0014159.png_3.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/0014159.png_4.png b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/0014159.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..33c952fd213f0d343e28952ae477b8f7e8aceff7 Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/0014159.png_4.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/0014159.png_5.png b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/0014159.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..d21dcb6e0a63546497cd883b80a5db000a2d378a Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/0014159.png_5.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/0014159.png_6.png b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/0014159.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..ae67f1cf8e2269b63d0e0eb90b8660d08c83f045 Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/0014159.png_6.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/0014159.png_7.png b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/0014159.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..ef123b740d769917b2fe3760938dc36e9a769619 Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/0014159.png_7.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/0014159.png_8.png b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/0014159.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..569386fb46b3ba2459139ae28edb1e45ebb96be1 Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/0014159.png_8.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/0014159.png_9.png b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/0014159.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..6d4049f488ea475b22b41ff4fdd59473a2450c18 Binary files /dev/null and b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/20230605_152040/vis_data/vis_image/0014159.png_9.png differ diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/epoch_1.pth b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/epoch_1.pth new file mode 100644 index 0000000000000000000000000000000000000000..96a2f518f58173e5ba55c00cdd542a9528e5e81c --- /dev/null +++ b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/epoch_1.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:fd78f063ea138f4141270cfaa002b47fce662751a87d1394cdd09d9fb20e0b81 +size 277359309 diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/epoch_10.pth b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/epoch_10.pth new file mode 100644 index 0000000000000000000000000000000000000000..96652357a1a2dd7cf497a04817602709877d0782 --- /dev/null +++ b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/epoch_10.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:60cd9fea411cc4330f0cbfe9703b44b88ce7a27a5a2661cd612e8680958fab10 +size 280089229 diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/epoch_2.pth b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/epoch_2.pth new file mode 100644 index 0000000000000000000000000000000000000000..0198ae1307d611c726c62b2fdb2a41c0a9027f3c --- /dev/null +++ b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/epoch_2.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:7ca7a61afdee0c327e7f26f9c26ad9fe816dd7bf42e27b2daefa8a71efb243b1 +size 277661581 diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/epoch_3.pth b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/epoch_3.pth new file mode 100644 index 0000000000000000000000000000000000000000..7faecb86c18df4604b1d243aa59ed9adb1e06ee2 --- /dev/null +++ b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/epoch_3.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:bc6f12cee685a7521843e22c58683a62f3a8b11571bd7fce38af10ceea27bc30 +size 277963853 diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/epoch_4.pth b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/epoch_4.pth new file mode 100644 index 0000000000000000000000000000000000000000..cd125aeadf60dda1c1c42e54fbfeefe3ee2669ce --- /dev/null +++ b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/epoch_4.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:9ab69a983eba7be9e0ba76e7fc6b6221bdf469045d738247ec60b29f98c42d13 +size 278265549 diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/epoch_5.pth b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/epoch_5.pth new file mode 100644 index 0000000000000000000000000000000000000000..5840b8d6a74ad102e6b1faf75e4e8855ca54e1d8 --- /dev/null +++ b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/epoch_5.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:52f9af813c413b42ce2fa1c07bb519a4a1c052e85113ce9db2be4745f1c5efbf +size 278567117 diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/epoch_6.pth b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/epoch_6.pth new file mode 100644 index 0000000000000000000000000000000000000000..8b94a7ef8f53725b1ecccc6589f09549209f5c2b --- /dev/null +++ b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/epoch_6.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:8a6ee82da26ec0a13ed27631c7810da88c0b0fb5d0e1bdb2333e4f221d040767 +size 278869133 diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/epoch_7.pth b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/epoch_7.pth new file mode 100644 index 0000000000000000000000000000000000000000..0ff2e8165da4bb893983a1c2e77395ba2acb2f3f --- /dev/null +++ b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/epoch_7.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:cbe26b09bb302c0c698e43d1c4f30f09e715680f92c42ade645b5d1309aed4ee +size 279171725 diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/epoch_8.pth b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/epoch_8.pth new file mode 100644 index 0000000000000000000000000000000000000000..38d44249e746e2b4e296a8e267648371f41a594b --- /dev/null +++ b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/epoch_8.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:77ef5e5fb94a51af5264337f0d99f10756bda19d1d8900eebcb6e525cd8e4165 +size 279477325 diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/epoch_9.pth b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/epoch_9.pth new file mode 100644 index 0000000000000000000000000000000000000000..b42f85f37490dba4b13008b9daafe0d98bd18734 --- /dev/null +++ b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/epoch_9.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c07c90eff65c37d7f3b4f5cacec063a50a9a2d0263c66b8cd0084cf21f89ebd0 +size 279783437 diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/last_checkpoint b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/last_checkpoint new file mode 100644 index 0000000000000000000000000000000000000000..741823d60b7b2acafe9b030dcd003cccd5aae1b1 --- /dev/null +++ b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/last_checkpoint @@ -0,0 +1 @@ +/mnt/petrelfs/luzeyu/workspace/fakebench/mmpretrain/workdir/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/epoch_10.pth \ No newline at end of file diff --git a/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1.py b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1.py new file mode 100644 index 0000000000000000000000000000000000000000..b35ee250106aa84d1065073f108965cfa202c34d --- /dev/null +++ b/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1.py @@ -0,0 +1,366 @@ +optim_wrapper = dict( + optimizer=dict( + type='SGD', + lr=0.0001, + momentum=0.9, + weight_decay=0.0001, + _scope_='mmpretrain'), + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=False, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=512) +model = dict( + type='ImageClassifier', + backbone=dict( + frozen_stages=2, + type='ResNet', + depth=50, + num_stages=4, + out_indices=(3, ), + style='pytorch', + init_cfg=dict( + type='Pretrained', + checkpoint= + 'https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth', + prefix='backbone')), + neck=dict(type='GlobalAveragePooling'), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=2048, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + topk=1)) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/all1.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/all2.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/all3.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3fake8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/cc1m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3real8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV2-1-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/if-dpmsolver++-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = dict(type='Accuracy', topk=1) +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV2-1-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/if-dpmsolver++-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = dict(type='Accuracy', topk=1) +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/resnet50_2xb256_all_1m_lr1e-4_aug_5e-1' diff --git a/resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1/20230604_214254/20230604_214254.log b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1/20230604_214254/20230604_214254.log new file mode 100644 index 0000000000000000000000000000000000000000..0c953b5db19dac330c78c06c0beeadbfe362b742 --- /dev/null +++ b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1/20230604_214254/20230604_214254.log @@ -0,0 +1,361 @@ +2023/06/04 21:42:58 - mmengine - INFO - +------------------------------------------------------------ +System environment: + sys.platform: linux + Python: 3.10.9 (main, Mar 8 2023, 10:47:38) [GCC 11.2.0] + CUDA available: True + numpy_random_seed: 1326369828 + GPU 0,1: NVIDIA A100-SXM4-80GB + CUDA_HOME: /mnt/petrelfs/share/cuda-11.6 + NVCC: Cuda compilation tools, release 11.6, V11.6.124 + GCC: gcc (GCC) 7.5.0 + PyTorch: 1.13.1 + PyTorch compiling details: PyTorch built with: + - GCC 9.3 + - C++ Version: 201402 + - Intel(R) oneAPI Math Kernel Library Version 2021.4-Product Build 20210904 for Intel(R) 64 architecture applications + - Intel(R) MKL-DNN v2.6.0 (Git Hash 52b5f107dd9cf10910aaa19cb47f3abf9b349815) + - OpenMP 201511 (a.k.a. OpenMP 4.5) + - LAPACK is enabled (usually provided by MKL) + - NNPACK is enabled + - CPU capability usage: AVX2 + - CUDA Runtime 11.6 + - NVCC architecture flags: -gencode;arch=compute_37,code=sm_37;-gencode;arch=compute_50,code=sm_50;-gencode;arch=compute_60,code=sm_60;-gencode;arch=compute_61,code=sm_61;-gencode;arch=compute_70,code=sm_70;-gencode;arch=compute_75,code=sm_75;-gencode;arch=compute_80,code=sm_80;-gencode;arch=compute_86,code=sm_86;-gencode;arch=compute_37,code=compute_37 + - CuDNN 8.3.2 (built against CUDA 11.5) + - Magma 2.6.1 + - Build settings: BLAS_INFO=mkl, BUILD_TYPE=Release, CUDA_VERSION=11.6, CUDNN_VERSION=8.3.2, CXX_COMPILER=/opt/rh/devtoolset-9/root/usr/bin/c++, CXX_FLAGS= -fabi-version=11 -Wno-deprecated -fvisibility-inlines-hidden -DUSE_PTHREADPOOL -fopenmp -DNDEBUG -DUSE_KINETO -DUSE_FBGEMM -DUSE_QNNPACK -DUSE_PYTORCH_QNNPACK -DUSE_XNNPACK -DSYMBOLICATE_MOBILE_DEBUG_HANDLE -DEDGE_PROFILER_USE_KINETO -O2 -fPIC -Wno-narrowing -Wall -Wextra -Werror=return-type -Werror=non-virtual-dtor -Wno-missing-field-initializers -Wno-type-limits -Wno-array-bounds -Wno-unknown-pragmas -Wunused-local-typedefs -Wno-unused-parameter -Wno-unused-function -Wno-unused-result -Wno-strict-overflow -Wno-strict-aliasing -Wno-error=deprecated-declarations -Wno-stringop-overflow -Wno-psabi -Wno-error=pedantic -Wno-error=redundant-decls -Wno-error=old-style-cast -fdiagnostics-color=always -faligned-new -Wno-unused-but-set-variable -Wno-maybe-uninitialized -fno-math-errno -fno-trapping-math -Werror=format -Werror=cast-function-type -Wno-stringop-overflow, LAPACK_INFO=mkl, PERF_WITH_AVX=1, PERF_WITH_AVX2=1, PERF_WITH_AVX512=1, TORCH_VERSION=1.13.1, USE_CUDA=ON, USE_CUDNN=ON, USE_EXCEPTION_PTR=1, USE_GFLAGS=OFF, USE_GLOG=OFF, USE_MKL=ON, USE_MKLDNN=ON, USE_MPI=OFF, USE_NCCL=ON, USE_NNPACK=ON, USE_OPENMP=ON, USE_ROCM=OFF, + + TorchVision: 0.14.1 + OpenCV: 4.7.0 + MMEngine: 0.7.3 + +Runtime environment: + cudnn_benchmark: True + mp_cfg: {'mp_start_method': 'fork', 'opencv_num_threads': 0} + dist_cfg: {'backend': 'nccl'} + seed: None + deterministic: False + Distributed launcher: slurm + Distributed training: True + GPU number: 2 +------------------------------------------------------------ + +2023/06/04 21:43:03 - mmengine - INFO - Config: +optim_wrapper = dict( + optimizer=dict( + type='SGD', + lr=0.0001, + momentum=0.9, + weight_decay=0.0001, + _scope_='mmpretrain'), + clip_grad=None) +param_scheduler = dict( + type='MultiStepLR', + by_epoch=True, + milestones=[30, 60, 90], + gamma=0.1, + _scope_='mmpretrain') +train_cfg = dict(by_epoch=False, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=512) +model = dict( + type='ImageClassifier', + backbone=dict( + frozen_stages=2, + type='ResNet', + depth=50, + num_stages=4, + out_indices=(3, ), + style='pytorch', + init_cfg=dict( + type='Pretrained', + checkpoint= + 'https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth', + prefix='backbone')), + neck=dict(type='GlobalAveragePooling'), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=2048, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + topk=1)) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stablediffusionV1-5R2-dpmsolver-25-1m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/cc1m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = dict(type='Accuracy', topk=1) +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = dict(type='Accuracy', topk=1) +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1' + +2023/06/04 21:43:17 - mmengine - INFO - Hooks will be executed in the following order: +before_run: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_load_checkpoint: +(ABOVE_NORMAL) EMAHook + -------------------- +before_train: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_train_epoch: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(NORMAL ) DistSamplerSeedHook + -------------------- +before_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook + -------------------- +after_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +after_train_epoch: +(NORMAL ) IterTimerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_val_epoch: +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook + -------------------- +before_val_iter: +(NORMAL ) IterTimerHook + -------------------- +after_val_iter: +(NORMAL ) IterTimerHook +(NORMAL ) VisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_val_epoch: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_save_checkpoint: +(ABOVE_NORMAL) EMAHook + -------------------- +after_train: +(VERY_LOW ) CheckpointHook + -------------------- +before_test_epoch: +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook + -------------------- +before_test_iter: +(NORMAL ) IterTimerHook + -------------------- +after_test_iter: +(NORMAL ) IterTimerHook +(NORMAL ) VisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_test_epoch: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_run: +(BELOW_NORMAL) LoggerHook + -------------------- diff --git a/resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1/20230604_214254/vis_data/config.py b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1/20230604_214254/vis_data/config.py new file mode 100644 index 0000000000000000000000000000000000000000..c5dbbac04dcfba6be188f3bc510947bbaa43d1ac --- /dev/null +++ b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1/20230604_214254/vis_data/config.py @@ -0,0 +1,233 @@ +optim_wrapper = dict( + optimizer=dict( + type='SGD', + lr=0.0001, + momentum=0.9, + weight_decay=0.0001, + _scope_='mmpretrain'), + clip_grad=None) +param_scheduler = dict( + type='MultiStepLR', + by_epoch=True, + milestones=[30, 60, 90], + gamma=0.1, + _scope_='mmpretrain') +train_cfg = dict(by_epoch=False, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=512) +model = dict( + type='ImageClassifier', + backbone=dict( + frozen_stages=2, + type='ResNet', + depth=50, + num_stages=4, + out_indices=(3, ), + style='pytorch', + init_cfg=dict( + type='Pretrained', + checkpoint= + 'https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth', + prefix='backbone')), + neck=dict(type='GlobalAveragePooling'), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=2048, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + topk=1)) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stablediffusionV1-5R2-dpmsolver-25-1m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/cc1m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = dict(type='Accuracy', topk=1) +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = dict(type='Accuracy', topk=1) +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1' diff --git a/resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1/20230604_214254/vis_data/events.out.tfevents.1685886191.SH-IDC1-10-140-24-129.8612.0 b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1/20230604_214254/vis_data/events.out.tfevents.1685886191.SH-IDC1-10-140-24-129.8612.0 new file mode 100644 index 0000000000000000000000000000000000000000..fef615e5fbdebc8a6e78e125b3f161b43ce67ad7 --- /dev/null +++ b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1/20230604_214254/vis_data/events.out.tfevents.1685886191.SH-IDC1-10-140-24-129.8612.0 @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:872443d1a26e59fa2229248093963b4f9c0594f4ff7354bd6e5d0ca374cdc904 +size 8826 diff --git a/resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1/20230604_215105/20230604_215105.log b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1/20230604_215105/20230604_215105.log new file mode 100644 index 0000000000000000000000000000000000000000..058944929e13a13a96a7dc989d960a764e03ce97 --- /dev/null +++ b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1/20230604_215105/20230604_215105.log @@ -0,0 +1,1305 @@ +2023/06/04 21:51:08 - mmengine - INFO - +------------------------------------------------------------ +System environment: + sys.platform: linux + Python: 3.10.9 (main, Mar 8 2023, 10:47:38) [GCC 11.2.0] + CUDA available: True + numpy_random_seed: 929922210 + GPU 0,1: NVIDIA A100-SXM4-80GB + CUDA_HOME: /mnt/petrelfs/share/cuda-11.6 + NVCC: Cuda compilation tools, release 11.6, V11.6.124 + GCC: gcc (GCC) 7.5.0 + PyTorch: 1.13.1 + PyTorch compiling details: PyTorch built with: + - GCC 9.3 + - C++ Version: 201402 + - Intel(R) oneAPI Math Kernel Library Version 2021.4-Product Build 20210904 for Intel(R) 64 architecture applications + - Intel(R) MKL-DNN v2.6.0 (Git Hash 52b5f107dd9cf10910aaa19cb47f3abf9b349815) + - OpenMP 201511 (a.k.a. OpenMP 4.5) + - LAPACK is enabled (usually provided by MKL) + - NNPACK is enabled + - CPU capability usage: AVX2 + - CUDA Runtime 11.6 + - NVCC architecture flags: -gencode;arch=compute_37,code=sm_37;-gencode;arch=compute_50,code=sm_50;-gencode;arch=compute_60,code=sm_60;-gencode;arch=compute_61,code=sm_61;-gencode;arch=compute_70,code=sm_70;-gencode;arch=compute_75,code=sm_75;-gencode;arch=compute_80,code=sm_80;-gencode;arch=compute_86,code=sm_86;-gencode;arch=compute_37,code=compute_37 + - CuDNN 8.3.2 (built against CUDA 11.5) + - Magma 2.6.1 + - Build settings: BLAS_INFO=mkl, BUILD_TYPE=Release, CUDA_VERSION=11.6, CUDNN_VERSION=8.3.2, CXX_COMPILER=/opt/rh/devtoolset-9/root/usr/bin/c++, CXX_FLAGS= -fabi-version=11 -Wno-deprecated -fvisibility-inlines-hidden -DUSE_PTHREADPOOL -fopenmp -DNDEBUG -DUSE_KINETO -DUSE_FBGEMM -DUSE_QNNPACK -DUSE_PYTORCH_QNNPACK -DUSE_XNNPACK -DSYMBOLICATE_MOBILE_DEBUG_HANDLE -DEDGE_PROFILER_USE_KINETO -O2 -fPIC -Wno-narrowing -Wall -Wextra -Werror=return-type -Werror=non-virtual-dtor -Wno-missing-field-initializers -Wno-type-limits -Wno-array-bounds -Wno-unknown-pragmas -Wunused-local-typedefs -Wno-unused-parameter -Wno-unused-function -Wno-unused-result -Wno-strict-overflow -Wno-strict-aliasing -Wno-error=deprecated-declarations -Wno-stringop-overflow -Wno-psabi -Wno-error=pedantic -Wno-error=redundant-decls -Wno-error=old-style-cast -fdiagnostics-color=always -faligned-new -Wno-unused-but-set-variable -Wno-maybe-uninitialized -fno-math-errno -fno-trapping-math -Werror=format -Werror=cast-function-type -Wno-stringop-overflow, LAPACK_INFO=mkl, PERF_WITH_AVX=1, PERF_WITH_AVX2=1, PERF_WITH_AVX512=1, TORCH_VERSION=1.13.1, USE_CUDA=ON, USE_CUDNN=ON, USE_EXCEPTION_PTR=1, USE_GFLAGS=OFF, USE_GLOG=OFF, USE_MKL=ON, USE_MKLDNN=ON, USE_MPI=OFF, USE_NCCL=ON, USE_NNPACK=ON, USE_OPENMP=ON, USE_ROCM=OFF, + + TorchVision: 0.14.1 + OpenCV: 4.7.0 + MMEngine: 0.7.3 + +Runtime environment: + cudnn_benchmark: True + mp_cfg: {'mp_start_method': 'fork', 'opencv_num_threads': 0} + dist_cfg: {'backend': 'nccl'} + seed: None + deterministic: False + Distributed launcher: slurm + Distributed training: True + GPU number: 2 +------------------------------------------------------------ + +2023/06/04 21:51:13 - mmengine - INFO - Config: +optim_wrapper = dict( + optimizer=dict( + type='SGD', + lr=0.0001, + momentum=0.9, + weight_decay=0.0001, + _scope_='mmpretrain'), + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=False, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=512) +model = dict( + type='ImageClassifier', + backbone=dict( + frozen_stages=2, + type='ResNet', + depth=50, + num_stages=4, + out_indices=(3, ), + style='pytorch', + init_cfg=dict( + type='Pretrained', + checkpoint= + 'https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth', + prefix='backbone')), + neck=dict(type='GlobalAveragePooling'), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=2048, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + topk=1)) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stablediffusionV1-5R2-dpmsolver-25-1m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/cc1m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = dict(type='Accuracy', topk=1) +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = dict(type='Accuracy', topk=1) +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1' + +2023/06/04 21:51:26 - mmengine - INFO - Hooks will be executed in the following order: +before_run: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_load_checkpoint: +(ABOVE_NORMAL) EMAHook + -------------------- +before_train: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_train_epoch: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(NORMAL ) DistSamplerSeedHook + -------------------- +before_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook + -------------------- +after_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +after_train_epoch: +(NORMAL ) IterTimerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_val_epoch: +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook + -------------------- +before_val_iter: +(NORMAL ) IterTimerHook + -------------------- +after_val_iter: +(NORMAL ) IterTimerHook +(NORMAL ) VisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_val_epoch: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_save_checkpoint: +(ABOVE_NORMAL) EMAHook + -------------------- +after_train: +(VERY_LOW ) CheckpointHook + -------------------- +before_test_epoch: +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook + -------------------- +before_test_iter: +(NORMAL ) IterTimerHook + -------------------- +after_test_iter: +(NORMAL ) IterTimerHook +(NORMAL ) VisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_test_epoch: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_run: +(BELOW_NORMAL) LoggerHook + -------------------- +2023/06/04 21:51:45 - mmengine - INFO - load backbone in model from: https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth +Name of parameter - Initialization information + +backbone.conv1.weight - torch.Size([64, 3, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.bn1.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.bn1.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.conv1.weight - torch.Size([64, 64, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.bn1.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.bn1.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.conv2.weight - torch.Size([64, 64, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.bn2.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.bn2.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.conv3.weight - torch.Size([256, 64, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.bn3.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.bn3.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.downsample.0.weight - torch.Size([256, 64, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.downsample.1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.downsample.1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.conv1.weight - torch.Size([64, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.bn1.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.bn1.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.conv2.weight - torch.Size([64, 64, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.bn2.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.bn2.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.conv3.weight - torch.Size([256, 64, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.bn3.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.bn3.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.conv1.weight - torch.Size([64, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.bn1.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.bn1.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.conv2.weight - torch.Size([64, 64, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.bn2.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.bn2.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.conv3.weight - torch.Size([256, 64, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.bn3.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.bn3.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.conv1.weight - torch.Size([128, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.bn1.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.bn1.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.conv2.weight - torch.Size([128, 128, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.bn2.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.bn2.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.conv3.weight - torch.Size([512, 128, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.bn3.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.bn3.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.downsample.0.weight - torch.Size([512, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.downsample.1.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.downsample.1.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.conv1.weight - torch.Size([128, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.bn1.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.bn1.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.conv2.weight - torch.Size([128, 128, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.bn2.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.bn2.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.conv3.weight - torch.Size([512, 128, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.bn3.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.bn3.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.conv1.weight - torch.Size([128, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.bn1.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.bn1.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.conv2.weight - torch.Size([128, 128, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.bn2.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.bn2.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.conv3.weight - torch.Size([512, 128, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.bn3.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.bn3.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.conv1.weight - torch.Size([128, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.bn1.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.bn1.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.conv2.weight - torch.Size([128, 128, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.bn2.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.bn2.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.conv3.weight - torch.Size([512, 128, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.bn3.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.bn3.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.conv1.weight - torch.Size([256, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.bn1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.bn1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.conv2.weight - torch.Size([256, 256, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.bn2.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.bn2.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.conv3.weight - torch.Size([1024, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.bn3.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.bn3.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.downsample.0.weight - torch.Size([1024, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.downsample.1.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.downsample.1.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.conv1.weight - torch.Size([256, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.bn1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.bn1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.conv2.weight - torch.Size([256, 256, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.bn2.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.bn2.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.conv3.weight - torch.Size([1024, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.bn3.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.bn3.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.conv1.weight - torch.Size([256, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.bn1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.bn1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.conv2.weight - torch.Size([256, 256, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.bn2.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.bn2.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.conv3.weight - torch.Size([1024, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.bn3.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.bn3.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.conv1.weight - torch.Size([256, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.bn1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.bn1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.conv2.weight - torch.Size([256, 256, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.bn2.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.bn2.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.conv3.weight - torch.Size([1024, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.bn3.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.bn3.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.conv1.weight - torch.Size([256, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.bn1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.bn1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.conv2.weight - torch.Size([256, 256, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.bn2.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.bn2.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.conv3.weight - torch.Size([1024, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.bn3.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.bn3.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.conv1.weight - torch.Size([256, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.bn1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.bn1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.conv2.weight - torch.Size([256, 256, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.bn2.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.bn2.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.conv3.weight - torch.Size([1024, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.bn3.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.bn3.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.conv1.weight - torch.Size([512, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.bn1.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.bn1.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.conv2.weight - torch.Size([512, 512, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.bn2.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.bn2.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.conv3.weight - torch.Size([2048, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.bn3.weight - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.bn3.bias - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.downsample.0.weight - torch.Size([2048, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.downsample.1.weight - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.downsample.1.bias - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.conv1.weight - torch.Size([512, 2048, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.bn1.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.bn1.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.conv2.weight - torch.Size([512, 512, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.bn2.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.bn2.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.conv3.weight - torch.Size([2048, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.bn3.weight - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.bn3.bias - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.conv1.weight - torch.Size([512, 2048, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.bn1.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.bn1.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.conv2.weight - torch.Size([512, 512, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.bn2.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.bn2.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.conv3.weight - torch.Size([2048, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.bn3.weight - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.bn3.bias - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +head.fc.weight - torch.Size([2, 2048]): +NormalInit: mean=0, std=0.01, bias=0 + +head.fc.bias - torch.Size([2]): +NormalInit: mean=0, std=0.01, bias=0 +2023/06/04 21:52:20 - mmengine - WARNING - "FileClient" will be deprecated in future. Please use io functions in https://mmengine.readthedocs.io/en/latest/api/fileio.html#file-io +2023/06/04 21:52:20 - mmengine - WARNING - "HardDiskBackend" is the alias of "LocalBackend" and the former will be deprecated in future. +2023/06/04 21:52:20 - mmengine - INFO - Checkpoints will be saved to /mnt/petrelfs/luzeyu/workspace/fakebench/mmpretrain/workdir/resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1. +2023/06/04 21:53:49 - mmengine - INFO - Epoch(train) [1][ 100/3907] lr: 9.9999e-05 eta: 9:33:55 time: 0.8556 data_time: 0.6374 memory: 9436 loss: 0.6017 +2023/06/04 21:55:13 - mmengine - INFO - Epoch(train) [1][ 200/3907] lr: 9.9994e-05 eta: 9:19:44 time: 0.8826 data_time: 0.6513 memory: 6319 loss: 0.5355 +2023/06/04 21:56:35 - mmengine - INFO - Epoch(train) [1][ 300/3907] lr: 9.9987e-05 eta: 9:08:23 time: 0.8178 data_time: 0.6206 memory: 6319 loss: 0.4696 +2023/06/04 21:57:58 - mmengine - INFO - Epoch(train) [1][ 400/3907] lr: 9.9977e-05 eta: 9:04:02 time: 0.8552 data_time: 0.7160 memory: 6319 loss: 0.4289 +2023/06/04 21:59:21 - mmengine - INFO - Epoch(train) [1][ 500/3907] lr: 9.9964e-05 eta: 9:01:06 time: 0.8139 data_time: 0.5343 memory: 6319 loss: 0.3963 +2023/06/04 22:00:44 - mmengine - INFO - Epoch(train) [1][ 600/3907] lr: 9.9948e-05 eta: 8:57:47 time: 0.8081 data_time: 0.5578 memory: 6319 loss: 0.3755 +2023/06/04 22:02:06 - mmengine - INFO - Epoch(train) [1][ 700/3907] lr: 9.9929e-05 eta: 8:54:35 time: 0.8240 data_time: 0.4442 memory: 6319 loss: 0.3686 +2023/06/04 22:03:30 - mmengine - INFO - Epoch(train) [1][ 800/3907] lr: 9.9907e-05 eta: 8:53:39 time: 0.8215 data_time: 0.6118 memory: 6319 loss: 0.3186 +2023/06/04 22:04:53 - mmengine - INFO - Epoch(train) [1][ 900/3907] lr: 9.9882e-05 eta: 8:51:42 time: 0.8196 data_time: 0.6151 memory: 6319 loss: 0.3201 +2023/06/04 22:06:15 - mmengine - INFO - Exp name: resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1_20230604_215105 +2023/06/04 22:06:15 - mmengine - INFO - Epoch(train) [1][1000/3907] lr: 9.9855e-05 eta: 8:49:37 time: 0.7850 data_time: 0.6462 memory: 6319 loss: 0.2997 +2023/06/04 22:07:39 - mmengine - INFO - Epoch(train) [1][1100/3907] lr: 9.9824e-05 eta: 8:48:23 time: 0.7865 data_time: 0.6466 memory: 6319 loss: 0.2942 +2023/06/04 22:09:02 - mmengine - INFO - Epoch(train) [1][1200/3907] lr: 9.9791e-05 eta: 8:46:43 time: 0.8114 data_time: 0.6717 memory: 6319 loss: 0.2938 +2023/06/04 22:10:26 - mmengine - INFO - Epoch(train) [1][1300/3907] lr: 9.9755e-05 eta: 8:45:38 time: 0.8235 data_time: 0.6846 memory: 6319 loss: 0.2830 +2023/06/04 22:11:49 - mmengine - INFO - Epoch(train) [1][1400/3907] lr: 9.9716e-05 eta: 8:44:09 time: 0.8342 data_time: 0.6955 memory: 6319 loss: 0.2429 +2023/06/04 22:13:15 - mmengine - INFO - Epoch(train) [1][1500/3907] lr: 9.9674e-05 eta: 8:43:40 time: 0.8962 data_time: 0.7570 memory: 6319 loss: 0.2325 +2023/06/04 22:14:39 - mmengine - INFO - Epoch(train) [1][1600/3907] lr: 9.9629e-05 eta: 8:42:32 time: 0.7987 data_time: 0.6592 memory: 6319 loss: 0.2570 +2023/06/04 22:16:01 - mmengine - INFO - Epoch(train) [1][1700/3907] lr: 9.9581e-05 eta: 8:40:19 time: 0.8404 data_time: 0.7005 memory: 6319 loss: 0.2341 +2023/06/04 22:17:26 - mmengine - INFO - Epoch(train) [1][1800/3907] lr: 9.9530e-05 eta: 8:39:26 time: 0.8508 data_time: 0.4487 memory: 6319 loss: 0.2164 +2023/06/04 22:18:49 - mmengine - INFO - Epoch(train) [1][1900/3907] lr: 9.9476e-05 eta: 8:38:04 time: 0.8518 data_time: 0.4794 memory: 6319 loss: 0.2403 +2023/06/04 22:20:11 - mmengine - INFO - Exp name: resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1_20230604_215105 +2023/06/04 22:20:11 - mmengine - INFO - Epoch(train) [1][2000/3907] lr: 9.9420e-05 eta: 8:36:07 time: 0.8377 data_time: 0.6046 memory: 6319 loss: 0.2153 +2023/06/04 22:21:35 - mmengine - INFO - Epoch(train) [1][2100/3907] lr: 9.9361e-05 eta: 8:34:52 time: 0.8786 data_time: 0.6837 memory: 6319 loss: 0.2249 +2023/06/04 22:22:59 - mmengine - INFO - Epoch(train) [1][2200/3907] lr: 9.9298e-05 eta: 8:33:39 time: 0.9060 data_time: 0.7660 memory: 6319 loss: 0.2216 +2023/06/04 22:24:25 - mmengine - INFO - Epoch(train) [1][2300/3907] lr: 9.9233e-05 eta: 8:32:43 time: 0.8855 data_time: 0.7312 memory: 6319 loss: 0.2242 +2023/06/04 22:25:50 - mmengine - INFO - Epoch(train) [1][2400/3907] lr: 9.9165e-05 eta: 8:31:45 time: 0.8973 data_time: 0.7572 memory: 6319 loss: 0.1893 +2023/06/04 22:27:13 - mmengine - INFO - Epoch(train) [1][2500/3907] lr: 9.9095e-05 eta: 8:30:12 time: 0.8494 data_time: 0.6952 memory: 6319 loss: 0.2068 +2023/06/04 22:28:37 - mmengine - INFO - Epoch(train) [1][2600/3907] lr: 9.9021e-05 eta: 8:28:43 time: 0.8950 data_time: 0.7557 memory: 6319 loss: 0.1941 +2023/06/04 22:30:00 - mmengine - INFO - Epoch(train) [1][2700/3907] lr: 9.8944e-05 eta: 8:27:19 time: 0.8160 data_time: 0.6767 memory: 6319 loss: 0.1944 +2023/06/04 22:31:24 - mmengine - INFO - Epoch(train) [1][2800/3907] lr: 9.8865e-05 eta: 8:26:02 time: 0.9016 data_time: 0.7614 memory: 6319 loss: 0.1921 +2023/06/04 22:32:49 - mmengine - INFO - Epoch(train) [1][2900/3907] lr: 9.8783e-05 eta: 8:24:43 time: 0.8836 data_time: 0.7428 memory: 6319 loss: 0.2043 +2023/06/04 22:34:13 - mmengine - INFO - Exp name: resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1_20230604_215105 +2023/06/04 22:34:13 - mmengine - INFO - Epoch(train) [1][3000/3907] lr: 9.8698e-05 eta: 8:23:26 time: 0.8847 data_time: 0.7450 memory: 6319 loss: 0.1752 +2023/06/04 22:35:40 - mmengine - INFO - Epoch(train) [1][3100/3907] lr: 9.8610e-05 eta: 8:22:39 time: 0.8595 data_time: 0.7172 memory: 6319 loss: 0.1846 +2023/06/04 22:37:08 - mmengine - INFO - Epoch(train) [1][3200/3907] lr: 9.8519e-05 eta: 8:22:07 time: 0.8963 data_time: 0.7557 memory: 6319 loss: 0.1729 +2023/06/04 22:38:34 - mmengine - INFO - Epoch(train) [1][3300/3907] lr: 9.8426e-05 eta: 8:21:01 time: 0.8097 data_time: 0.6702 memory: 6319 loss: 0.1801 +2023/06/04 22:39:54 - mmengine - INFO - Epoch(train) [1][3400/3907] lr: 9.8330e-05 eta: 8:19:00 time: 0.8048 data_time: 0.6524 memory: 6319 loss: 0.1668 +2023/06/04 22:41:18 - mmengine - INFO - Epoch(train) [1][3500/3907] lr: 9.8231e-05 eta: 8:17:33 time: 0.8350 data_time: 0.6932 memory: 6319 loss: 0.1847 +2023/06/04 22:42:45 - mmengine - INFO - Epoch(train) [1][3600/3907] lr: 9.8129e-05 eta: 8:16:38 time: 0.8272 data_time: 0.6723 memory: 6319 loss: 0.1566 +2023/06/04 22:44:08 - mmengine - INFO - Epoch(train) [1][3700/3907] lr: 9.8024e-05 eta: 8:15:01 time: 0.8044 data_time: 0.6637 memory: 6319 loss: 0.1768 +2023/06/04 22:45:32 - mmengine - INFO - Epoch(train) [1][3800/3907] lr: 9.7917e-05 eta: 8:13:37 time: 0.8199 data_time: 0.6011 memory: 6319 loss: 0.1848 +2023/06/04 22:46:54 - mmengine - INFO - Epoch(train) [1][3900/3907] lr: 9.7806e-05 eta: 8:12:01 time: 0.8399 data_time: 0.6998 memory: 6319 loss: 0.1674 +2023/06/04 22:46:57 - mmengine - INFO - Exp name: resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1_20230604_215105 +2023/06/04 22:46:57 - mmengine - INFO - Saving checkpoint at 1 epochs +2023/06/04 22:47:45 - mmengine - INFO - Epoch(val) [1][57/57] accuracy/top1: 81.1210 data_time: 0.5407 time: 0.6341 +2023/06/04 22:49:09 - mmengine - INFO - Exp name: resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1_20230604_215105 +2023/06/04 22:49:12 - mmengine - INFO - Epoch(train) [2][ 100/3907] lr: 9.7685e-05 eta: 8:10:32 time: 0.8102 data_time: 0.4053 memory: 6319 loss: 0.1645 +2023/06/04 22:50:37 - mmengine - INFO - Epoch(train) [2][ 200/3907] lr: 9.7570e-05 eta: 8:09:15 time: 0.8014 data_time: 0.2409 memory: 6319 loss: 0.1662 +2023/06/04 22:51:59 - mmengine - INFO - Epoch(train) [2][ 300/3907] lr: 9.7451e-05 eta: 8:07:40 time: 0.7924 data_time: 0.1317 memory: 6319 loss: 0.1659 +2023/06/04 22:53:21 - mmengine - INFO - Epoch(train) [2][ 400/3907] lr: 9.7329e-05 eta: 8:05:58 time: 0.8041 data_time: 0.3432 memory: 6319 loss: 0.1442 +2023/06/04 22:54:42 - mmengine - INFO - Epoch(train) [2][ 500/3907] lr: 9.7205e-05 eta: 8:04:10 time: 0.7639 data_time: 0.3228 memory: 6319 loss: 0.1593 +2023/06/04 22:56:02 - mmengine - INFO - Epoch(train) [2][ 600/3907] lr: 9.7078e-05 eta: 8:02:22 time: 0.8368 data_time: 0.2667 memory: 6319 loss: 0.1477 +2023/06/04 22:57:26 - mmengine - INFO - Epoch(train) [2][ 700/3907] lr: 9.6949e-05 eta: 8:00:59 time: 0.8618 data_time: 0.3482 memory: 6319 loss: 0.1542 +2023/06/04 22:58:50 - mmengine - INFO - Epoch(train) [2][ 800/3907] lr: 9.6816e-05 eta: 7:59:36 time: 0.8392 data_time: 0.5624 memory: 6319 loss: 0.1587 +2023/06/04 23:00:15 - mmengine - INFO - Epoch(train) [2][ 900/3907] lr: 9.6681e-05 eta: 7:58:23 time: 0.8635 data_time: 0.7023 memory: 6319 loss: 0.1503 +2023/06/04 23:01:40 - mmengine - INFO - Epoch(train) [2][1000/3907] lr: 9.6544e-05 eta: 7:57:06 time: 0.8262 data_time: 0.5626 memory: 6319 loss: 0.1641 +2023/06/04 23:02:58 - mmengine - INFO - Exp name: resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1_20230604_215105 +2023/06/04 23:03:07 - mmengine - INFO - Epoch(train) [2][1100/3907] lr: 9.6403e-05 eta: 7:56:02 time: 0.8973 data_time: 0.7439 memory: 6319 loss: 0.1409 +2023/06/04 23:04:30 - mmengine - INFO - Epoch(train) [2][1200/3907] lr: 9.6260e-05 eta: 7:54:36 time: 0.8961 data_time: 0.6420 memory: 6319 loss: 0.1324 +2023/06/04 23:05:56 - mmengine - INFO - Epoch(train) [2][1300/3907] lr: 9.6114e-05 eta: 7:53:21 time: 0.8394 data_time: 0.2256 memory: 6319 loss: 0.1446 +2023/06/04 23:07:21 - mmengine - INFO - Epoch(train) [2][1400/3907] lr: 9.5966e-05 eta: 7:52:08 time: 0.8950 data_time: 0.0013 memory: 6319 loss: 0.1479 +2023/06/04 23:08:48 - mmengine - INFO - Epoch(train) [2][1500/3907] lr: 9.5815e-05 eta: 7:51:02 time: 0.9058 data_time: 0.0016 memory: 6319 loss: 0.1650 +2023/06/04 23:10:13 - mmengine - INFO - Epoch(train) [2][1600/3907] lr: 9.5661e-05 eta: 7:49:47 time: 0.8428 data_time: 0.0010 memory: 6319 loss: 0.1288 +2023/06/04 23:12:41 - mmengine - INFO - Epoch(train) [2][1700/3907] lr: 9.5505e-05 eta: 7:54:41 time: 0.8919 data_time: 0.5324 memory: 6319 loss: 0.1347 +2023/06/04 23:14:04 - mmengine - INFO - Epoch(train) [2][1800/3907] lr: 9.5346e-05 eta: 7:53:07 time: 0.8369 data_time: 0.3318 memory: 6319 loss: 0.1408 +2023/06/04 23:15:27 - mmengine - INFO - Epoch(train) [2][1900/3907] lr: 9.5184e-05 eta: 7:51:26 time: 0.7978 data_time: 0.3791 memory: 6319 loss: 0.1348 +2023/06/04 23:16:50 - mmengine - INFO - Epoch(train) [2][2000/3907] lr: 9.5020e-05 eta: 7:49:50 time: 0.8452 data_time: 0.2967 memory: 6319 loss: 0.1497 +2023/06/04 23:18:32 - mmengine - INFO - Exp name: resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1_20230604_215105 +2023/06/04 23:18:40 - mmengine - INFO - Epoch(train) [2][2100/3907] lr: 9.4854e-05 eta: 7:50:45 time: 0.8520 data_time: 0.5304 memory: 6319 loss: 0.1324 +2023/06/04 23:20:03 - mmengine - INFO - Epoch(train) [2][2200/3907] lr: 9.4684e-05 eta: 7:49:03 time: 0.8335 data_time: 0.6336 memory: 6319 loss: 0.1286 +2023/06/04 23:21:24 - mmengine - INFO - Epoch(train) [2][2300/3907] lr: 9.4512e-05 eta: 7:47:18 time: 0.8087 data_time: 0.6686 memory: 6319 loss: 0.1406 +2023/06/04 23:22:47 - mmengine - INFO - Epoch(train) [2][2400/3907] lr: 9.4338e-05 eta: 7:45:37 time: 0.7929 data_time: 0.6530 memory: 6319 loss: 0.1464 +2023/06/04 23:24:09 - mmengine - INFO - Epoch(train) [2][2500/3907] lr: 9.4161e-05 eta: 7:43:59 time: 0.8468 data_time: 0.4401 memory: 6319 loss: 0.1287 +2023/06/04 23:25:32 - mmengine - INFO - Epoch(train) [2][2600/3907] lr: 9.3981e-05 eta: 7:42:19 time: 0.8114 data_time: 0.2139 memory: 6319 loss: 0.1295 +2023/06/04 23:26:57 - mmengine - INFO - Epoch(train) [2][2700/3907] lr: 9.3799e-05 eta: 7:40:54 time: 0.8726 data_time: 0.0775 memory: 6319 loss: 0.1248 +2023/06/04 23:28:21 - mmengine - INFO - Epoch(train) [2][2800/3907] lr: 9.3615e-05 eta: 7:39:23 time: 0.7963 data_time: 0.0010 memory: 6319 loss: 0.1352 +2023/06/04 23:29:44 - mmengine - INFO - Epoch(train) [2][2900/3907] lr: 9.3428e-05 eta: 7:37:49 time: 0.8247 data_time: 0.0010 memory: 6319 loss: 0.1253 +2023/06/04 23:31:06 - mmengine - INFO - Epoch(train) [2][3000/3907] lr: 9.3238e-05 eta: 7:36:10 time: 0.7806 data_time: 0.0103 memory: 6319 loss: 0.1147 +2023/06/04 23:32:22 - mmengine - INFO - Exp name: resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1_20230604_215105 +2023/06/04 23:32:27 - mmengine - INFO - Epoch(train) [2][3100/3907] lr: 9.3046e-05 eta: 7:34:27 time: 0.8262 data_time: 0.5238 memory: 6319 loss: 0.1257 +2023/06/04 23:33:50 - mmengine - INFO - Epoch(train) [2][3200/3907] lr: 9.2852e-05 eta: 7:32:49 time: 0.8250 data_time: 0.6844 memory: 6319 loss: 0.1266 +2023/06/04 23:35:11 - mmengine - INFO - Epoch(train) [2][3300/3907] lr: 9.2655e-05 eta: 7:31:07 time: 0.8342 data_time: 0.6880 memory: 6319 loss: 0.1203 +2023/06/04 23:36:34 - mmengine - INFO - Epoch(train) [2][3400/3907] lr: 9.2456e-05 eta: 7:29:35 time: 0.8034 data_time: 0.6625 memory: 6319 loss: 0.1206 +2023/06/04 23:37:55 - mmengine - INFO - Epoch(train) [2][3500/3907] lr: 9.2254e-05 eta: 7:27:55 time: 0.8242 data_time: 0.6826 memory: 6319 loss: 0.1223 +2023/06/04 23:39:17 - mmengine - INFO - Epoch(train) [2][3600/3907] lr: 9.2050e-05 eta: 7:26:18 time: 0.8548 data_time: 0.6546 memory: 6319 loss: 0.1287 +2023/06/04 23:40:41 - mmengine - INFO - Epoch(train) [2][3700/3907] lr: 9.1843e-05 eta: 7:24:49 time: 0.8155 data_time: 0.6769 memory: 6319 loss: 0.1375 +2023/06/04 23:42:04 - mmengine - INFO - Epoch(train) [2][3800/3907] lr: 9.1634e-05 eta: 7:23:16 time: 0.8176 data_time: 0.6778 memory: 6319 loss: 0.1166 +2023/06/04 23:43:30 - mmengine - INFO - Epoch(train) [2][3900/3907] lr: 9.1423e-05 eta: 7:21:54 time: 0.8078 data_time: 0.6675 memory: 6319 loss: 0.1104 +2023/06/04 23:43:30 - mmengine - INFO - Exp name: resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1_20230604_215105 +2023/06/04 23:43:30 - mmengine - INFO - Saving checkpoint at 2 epochs +2023/06/04 23:44:19 - mmengine - INFO - Epoch(val) [2][57/57] accuracy/top1: 87.6309 data_time: 0.5409 time: 0.6319 +2023/06/04 23:45:39 - mmengine - INFO - Epoch(train) [3][ 100/3907] lr: 9.1194e-05 eta: 7:19:42 time: 0.8016 data_time: 0.5991 memory: 6319 loss: 0.1080 +2023/06/04 23:46:52 - mmengine - INFO - Exp name: resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1_20230604_215105 +2023/06/04 23:47:01 - mmengine - INFO - Epoch(train) [3][ 200/3907] lr: 9.0978e-05 eta: 7:18:06 time: 0.8080 data_time: 0.6571 memory: 6319 loss: 0.1194 +2023/06/04 23:48:23 - mmengine - INFO - Epoch(train) [3][ 300/3907] lr: 9.0759e-05 eta: 7:16:32 time: 0.8023 data_time: 0.6612 memory: 6319 loss: 0.1083 +2023/06/04 23:49:47 - mmengine - INFO - Epoch(train) [3][ 400/3907] lr: 9.0539e-05 eta: 7:15:08 time: 0.7928 data_time: 0.6531 memory: 6319 loss: 0.1171 +2023/06/04 23:51:13 - mmengine - INFO - Epoch(train) [3][ 500/3907] lr: 9.0315e-05 eta: 7:13:48 time: 0.8415 data_time: 0.2923 memory: 6319 loss: 0.1012 +2023/06/04 23:52:35 - mmengine - INFO - Epoch(train) [3][ 600/3907] lr: 9.0090e-05 eta: 7:12:14 time: 0.8081 data_time: 0.4353 memory: 6319 loss: 0.1045 +2023/06/04 23:53:57 - mmengine - INFO - Epoch(train) [3][ 700/3907] lr: 8.9862e-05 eta: 7:10:40 time: 0.9185 data_time: 0.7276 memory: 6319 loss: 0.1094 +2023/06/04 23:55:20 - mmengine - INFO - Epoch(train) [3][ 800/3907] lr: 8.9632e-05 eta: 7:09:07 time: 0.8502 data_time: 0.6078 memory: 6319 loss: 0.1209 +2023/06/04 23:56:41 - mmengine - INFO - Epoch(train) [3][ 900/3907] lr: 8.9400e-05 eta: 7:07:31 time: 0.8272 data_time: 0.6877 memory: 6319 loss: 0.1124 +2023/06/04 23:58:04 - mmengine - INFO - Epoch(train) [3][1000/3907] lr: 8.9166e-05 eta: 7:06:01 time: 0.8271 data_time: 0.6879 memory: 6319 loss: 0.1017 +2023/06/04 23:59:25 - mmengine - INFO - Epoch(train) [3][1100/3907] lr: 8.8929e-05 eta: 7:04:27 time: 0.8020 data_time: 0.6629 memory: 6319 loss: 0.1049 +2023/06/05 00:00:37 - mmengine - INFO - Exp name: resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1_20230604_215105 +2023/06/05 00:00:46 - mmengine - INFO - Epoch(train) [3][1200/3907] lr: 8.8691e-05 eta: 7:02:50 time: 0.8449 data_time: 0.7062 memory: 6319 loss: 0.0998 +2023/06/05 00:02:10 - mmengine - INFO - Epoch(train) [3][1300/3907] lr: 8.8450e-05 eta: 7:01:25 time: 0.8525 data_time: 0.7122 memory: 6319 loss: 0.0998 +2023/06/05 00:03:34 - mmengine - INFO - Epoch(train) [3][1400/3907] lr: 8.8206e-05 eta: 6:59:58 time: 0.8061 data_time: 0.6661 memory: 6319 loss: 0.0975 +2023/06/05 00:04:57 - mmengine - INFO - Epoch(train) [3][1500/3907] lr: 8.7961e-05 eta: 6:58:28 time: 0.8054 data_time: 0.6657 memory: 6319 loss: 0.1061 +2023/06/05 00:06:22 - mmengine - INFO - Epoch(train) [3][1600/3907] lr: 8.7714e-05 eta: 6:57:08 time: 1.0598 data_time: 0.9204 memory: 6319 loss: 0.1128 +2023/06/05 00:07:43 - mmengine - INFO - Epoch(train) [3][1700/3907] lr: 8.7464e-05 eta: 6:55:32 time: 0.7843 data_time: 0.6355 memory: 6319 loss: 0.1119 +2023/06/05 00:09:04 - mmengine - INFO - Epoch(train) [3][1800/3907] lr: 8.7213e-05 eta: 6:53:57 time: 0.7983 data_time: 0.6173 memory: 6319 loss: 0.1119 +2023/06/05 00:10:26 - mmengine - INFO - Epoch(train) [3][1900/3907] lr: 8.6959e-05 eta: 6:52:27 time: 0.8474 data_time: 0.7083 memory: 6319 loss: 0.1246 +2023/06/05 00:11:50 - mmengine - INFO - Epoch(train) [3][2000/3907] lr: 8.6703e-05 eta: 6:51:01 time: 0.8266 data_time: 0.6870 memory: 6319 loss: 0.0850 +2023/06/05 00:13:13 - mmengine - INFO - Epoch(train) [3][2100/3907] lr: 8.6445e-05 eta: 6:49:34 time: 0.8448 data_time: 0.7059 memory: 6319 loss: 0.1088 +2023/06/05 00:14:27 - mmengine - INFO - Exp name: resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1_20230604_215105 +2023/06/05 00:14:36 - mmengine - INFO - Epoch(train) [3][2200/3907] lr: 8.6186e-05 eta: 6:48:04 time: 0.8474 data_time: 0.7079 memory: 6319 loss: 0.0951 +2023/06/05 00:15:57 - mmengine - INFO - Epoch(train) [3][2300/3907] lr: 8.5924e-05 eta: 6:46:32 time: 0.8378 data_time: 0.6977 memory: 6319 loss: 0.1166 +2023/06/05 00:17:18 - mmengine - INFO - Epoch(train) [3][2400/3907] lr: 8.5660e-05 eta: 6:44:56 time: 0.8034 data_time: 0.6644 memory: 6319 loss: 0.0951 +2023/06/05 00:18:40 - mmengine - INFO - Epoch(train) [3][2500/3907] lr: 8.5394e-05 eta: 6:43:27 time: 0.8075 data_time: 0.6676 memory: 6319 loss: 0.1133 +2023/06/05 00:20:03 - mmengine - INFO - Epoch(train) [3][2600/3907] lr: 8.5126e-05 eta: 6:41:59 time: 0.8520 data_time: 0.7125 memory: 6319 loss: 0.0913 +2023/06/05 00:21:26 - mmengine - INFO - Epoch(train) [3][2700/3907] lr: 8.4856e-05 eta: 6:40:31 time: 0.8511 data_time: 0.7117 memory: 6319 loss: 0.1074 +2023/06/05 00:22:46 - mmengine - INFO - Epoch(train) [3][2800/3907] lr: 8.4585e-05 eta: 6:38:57 time: 0.8115 data_time: 0.6724 memory: 6319 loss: 0.1015 +2023/06/05 00:24:10 - mmengine - INFO - Epoch(train) [3][2900/3907] lr: 8.4311e-05 eta: 6:37:32 time: 0.7821 data_time: 0.6431 memory: 6319 loss: 0.0994 +2023/06/05 00:25:32 - mmengine - INFO - Epoch(train) [3][3000/3907] lr: 8.4036e-05 eta: 6:36:04 time: 0.8413 data_time: 0.7016 memory: 6319 loss: 0.1060 +2023/06/05 00:26:57 - mmengine - INFO - Epoch(train) [3][3100/3907] lr: 8.3758e-05 eta: 6:34:41 time: 0.8459 data_time: 0.7046 memory: 6319 loss: 0.0905 +2023/06/05 00:28:11 - mmengine - INFO - Exp name: resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1_20230604_215105 +2023/06/05 00:28:20 - mmengine - INFO - Epoch(train) [3][3200/3907] lr: 8.3479e-05 eta: 6:33:14 time: 0.8190 data_time: 0.6797 memory: 6319 loss: 0.0948 +2023/06/05 00:29:41 - mmengine - INFO - Epoch(train) [3][3300/3907] lr: 8.3198e-05 eta: 6:31:43 time: 0.8492 data_time: 0.7089 memory: 6319 loss: 0.1008 +2023/06/05 00:31:04 - mmengine - INFO - Epoch(train) [3][3400/3907] lr: 8.2915e-05 eta: 6:30:15 time: 0.8372 data_time: 0.6973 memory: 6319 loss: 0.0898 +2023/06/05 00:32:34 - mmengine - INFO - Epoch(train) [3][3500/3907] lr: 8.2630e-05 eta: 6:29:06 time: 0.7207 data_time: 0.5805 memory: 6319 loss: 0.0927 +2023/06/05 00:33:52 - mmengine - INFO - Epoch(train) [3][3600/3907] lr: 8.2344e-05 eta: 6:27:28 time: 0.8000 data_time: 0.6612 memory: 6319 loss: 0.0940 +2023/06/05 00:35:14 - mmengine - INFO - Epoch(train) [3][3700/3907] lr: 8.2056e-05 eta: 6:25:57 time: 0.8175 data_time: 0.6755 memory: 6319 loss: 0.0904 +2023/06/05 00:36:35 - mmengine - INFO - Epoch(train) [3][3800/3907] lr: 8.1765e-05 eta: 6:24:26 time: 0.8047 data_time: 0.6652 memory: 6319 loss: 0.0913 +2023/06/05 00:37:56 - mmengine - INFO - Epoch(train) [3][3900/3907] lr: 8.1474e-05 eta: 6:22:56 time: 0.8135 data_time: 0.6747 memory: 6319 loss: 0.0852 +2023/06/05 00:38:04 - mmengine - INFO - Exp name: resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1_20230604_215105 +2023/06/05 00:38:04 - mmengine - INFO - Saving checkpoint at 3 epochs +2023/06/05 00:38:48 - mmengine - INFO - Epoch(val) [3][57/57] accuracy/top1: 91.8409 data_time: 0.4862 time: 0.5769 +2023/06/05 00:40:13 - mmengine - INFO - Epoch(train) [4][ 100/3907] lr: 8.1160e-05 eta: 6:21:31 time: 0.8407 data_time: 0.7006 memory: 6319 loss: 0.0802 +2023/06/05 00:41:42 - mmengine - INFO - Epoch(train) [4][ 200/3907] lr: 8.0864e-05 eta: 6:20:19 time: 0.9453 data_time: 0.7935 memory: 6319 loss: 0.0890 +2023/06/05 00:42:55 - mmengine - INFO - Exp name: resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1_20230604_215105 +2023/06/05 00:43:09 - mmengine - INFO - Epoch(train) [4][ 300/3907] lr: 8.0567e-05 eta: 6:19:01 time: 0.6830 data_time: 0.2867 memory: 6319 loss: 0.0892 +2023/06/05 00:44:27 - mmengine - INFO - Epoch(train) [4][ 400/3907] lr: 8.0269e-05 eta: 6:17:24 time: 0.8337 data_time: 0.4486 memory: 6319 loss: 0.0916 +2023/06/05 00:45:49 - mmengine - INFO - Epoch(train) [4][ 500/3907] lr: 7.9969e-05 eta: 6:15:55 time: 0.8283 data_time: 0.5825 memory: 6319 loss: 0.0874 +2023/06/05 00:47:12 - mmengine - INFO - Epoch(train) [4][ 600/3907] lr: 7.9667e-05 eta: 6:14:29 time: 0.8623 data_time: 0.7231 memory: 6319 loss: 0.0967 +2023/06/05 00:48:36 - mmengine - INFO - Epoch(train) [4][ 700/3907] lr: 7.9363e-05 eta: 6:13:05 time: 0.8511 data_time: 0.7115 memory: 6319 loss: 0.0838 +2023/06/05 00:49:57 - mmengine - INFO - Epoch(train) [4][ 800/3907] lr: 7.9058e-05 eta: 6:11:36 time: 0.8360 data_time: 0.6501 memory: 6319 loss: 0.0824 +2023/06/05 00:51:20 - mmengine - INFO - Epoch(train) [4][ 900/3907] lr: 7.8752e-05 eta: 6:10:09 time: 0.8109 data_time: 0.6011 memory: 6319 loss: 0.0920 +2023/06/05 00:53:00 - mmengine - INFO - Epoch(train) [4][1000/3907] lr: 7.8444e-05 eta: 6:09:18 time: 1.2103 data_time: 1.0703 memory: 6319 loss: 0.0875 +2023/06/05 00:54:41 - mmengine - INFO - Epoch(train) [4][1100/3907] lr: 7.8134e-05 eta: 6:08:27 time: 0.7627 data_time: 0.5997 memory: 6319 loss: 0.0951 +2023/06/05 00:56:01 - mmengine - INFO - Epoch(train) [4][1200/3907] lr: 7.7823e-05 eta: 6:06:55 time: 0.8529 data_time: 0.7129 memory: 6319 loss: 0.0788 +2023/06/05 00:57:08 - mmengine - INFO - Exp name: resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1_20230604_215105 +2023/06/05 00:57:24 - mmengine - INFO - Epoch(train) [4][1300/3907] lr: 7.7510e-05 eta: 6:05:29 time: 0.7780 data_time: 0.6366 memory: 6319 loss: 0.0904 +2023/06/05 00:58:47 - mmengine - INFO - Epoch(train) [4][1400/3907] lr: 7.7196e-05 eta: 6:04:03 time: 0.8584 data_time: 0.7182 memory: 6319 loss: 0.0874 +2023/06/05 01:00:09 - mmengine - INFO - Epoch(train) [4][1500/3907] lr: 7.6881e-05 eta: 6:02:34 time: 0.8351 data_time: 0.6954 memory: 6319 loss: 0.0911 +2023/06/05 01:01:30 - mmengine - INFO - Epoch(train) [4][1600/3907] lr: 7.6564e-05 eta: 6:01:04 time: 0.7912 data_time: 0.6502 memory: 6319 loss: 0.0823 +2023/06/05 01:02:54 - mmengine - INFO - Epoch(train) [4][1700/3907] lr: 7.6246e-05 eta: 5:59:40 time: 0.8093 data_time: 0.6700 memory: 6319 loss: 0.0881 +2023/06/05 01:04:21 - mmengine - INFO - Epoch(train) [4][1800/3907] lr: 7.5926e-05 eta: 5:58:21 time: 0.8603 data_time: 0.7184 memory: 6319 loss: 0.1022 +2023/06/05 01:05:45 - mmengine - INFO - Epoch(train) [4][1900/3907] lr: 7.5605e-05 eta: 5:56:56 time: 0.8956 data_time: 0.7553 memory: 6319 loss: 0.0738 +2023/06/05 01:07:10 - mmengine - INFO - Epoch(train) [4][2000/3907] lr: 7.5283e-05 eta: 5:55:33 time: 0.8788 data_time: 0.7363 memory: 6319 loss: 0.0894 +2023/06/05 01:08:34 - mmengine - INFO - Epoch(train) [4][2100/3907] lr: 7.4959e-05 eta: 5:54:09 time: 0.8677 data_time: 0.7268 memory: 6319 loss: 0.0958 +2023/06/05 01:09:59 - mmengine - INFO - Epoch(train) [4][2200/3907] lr: 7.4634e-05 eta: 5:52:47 time: 0.8484 data_time: 0.7070 memory: 6319 loss: 0.0759 +2023/06/05 01:11:07 - mmengine - INFO - Exp name: resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1_20230604_215105 +2023/06/05 01:11:26 - mmengine - INFO - Epoch(train) [4][2300/3907] lr: 7.4308e-05 eta: 5:51:27 time: 0.9134 data_time: 0.7728 memory: 6319 loss: 0.0849 +2023/06/05 01:12:46 - mmengine - INFO - Epoch(train) [4][2400/3907] lr: 7.3980e-05 eta: 5:49:56 time: 0.8246 data_time: 0.6785 memory: 6319 loss: 0.0892 +2023/06/05 01:14:11 - mmengine - INFO - Epoch(train) [4][2500/3907] lr: 7.3652e-05 eta: 5:48:34 time: 0.8778 data_time: 0.7386 memory: 6319 loss: 0.0804 +2023/06/05 01:15:40 - mmengine - INFO - Epoch(train) [4][2600/3907] lr: 7.3322e-05 eta: 5:47:18 time: 0.8677 data_time: 0.7279 memory: 6319 loss: 0.0739 +2023/06/05 01:17:05 - mmengine - INFO - Epoch(train) [4][2700/3907] lr: 7.2991e-05 eta: 5:45:54 time: 0.8694 data_time: 0.7301 memory: 6319 loss: 0.0753 +2023/06/05 01:18:30 - mmengine - INFO - Epoch(train) [4][2800/3907] lr: 7.2659e-05 eta: 5:44:31 time: 0.8242 data_time: 0.6853 memory: 6319 loss: 0.0867 +2023/06/05 01:19:58 - mmengine - INFO - Epoch(train) [4][2900/3907] lr: 7.2325e-05 eta: 5:43:13 time: 0.8967 data_time: 0.7513 memory: 6319 loss: 0.0914 +2023/06/05 01:21:25 - mmengine - INFO - Epoch(train) [4][3000/3907] lr: 7.1991e-05 eta: 5:41:54 time: 0.8413 data_time: 0.7023 memory: 6319 loss: 0.1017 +2023/06/05 01:22:50 - mmengine - INFO - Epoch(train) [4][3100/3907] lr: 7.1655e-05 eta: 5:40:31 time: 0.8975 data_time: 0.7560 memory: 6319 loss: 0.0768 +2023/06/05 01:24:17 - mmengine - INFO - Epoch(train) [4][3200/3907] lr: 7.1318e-05 eta: 5:39:10 time: 0.8561 data_time: 0.7165 memory: 6319 loss: 0.0812 +2023/06/05 01:25:24 - mmengine - INFO - Exp name: resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1_20230604_215105 +2023/06/05 01:25:41 - mmengine - INFO - Epoch(train) [4][3300/3907] lr: 7.0981e-05 eta: 5:37:47 time: 0.8588 data_time: 0.7158 memory: 6319 loss: 0.0770 +2023/06/05 01:27:08 - mmengine - INFO - Epoch(train) [4][3400/3907] lr: 7.0642e-05 eta: 5:36:26 time: 0.8454 data_time: 0.7051 memory: 6319 loss: 0.0883 +2023/06/05 01:28:33 - mmengine - INFO - Epoch(train) [4][3500/3907] lr: 7.0302e-05 eta: 5:35:03 time: 0.8114 data_time: 0.6706 memory: 6319 loss: 0.0867 +2023/06/05 01:29:58 - mmengine - INFO - Epoch(train) [4][3600/3907] lr: 6.9961e-05 eta: 5:33:40 time: 0.8810 data_time: 0.7404 memory: 6319 loss: 0.0827 +2023/06/05 01:31:23 - mmengine - INFO - Epoch(train) [4][3700/3907] lr: 6.9620e-05 eta: 5:32:17 time: 0.7926 data_time: 0.6521 memory: 6319 loss: 0.0972 +2023/06/05 01:32:48 - mmengine - INFO - Epoch(train) [4][3800/3907] lr: 6.9277e-05 eta: 5:30:53 time: 0.8082 data_time: 0.6675 memory: 6319 loss: 0.0867 +2023/06/05 01:34:12 - mmengine - INFO - Epoch(train) [4][3900/3907] lr: 6.8933e-05 eta: 5:29:28 time: 0.8148 data_time: 0.6737 memory: 6319 loss: 0.0895 +2023/06/05 01:34:14 - mmengine - INFO - Exp name: resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1_20230604_215105 +2023/06/05 01:34:14 - mmengine - INFO - Saving checkpoint at 4 epochs +2023/06/05 01:35:00 - mmengine - INFO - Epoch(val) [4][57/57] accuracy/top1: 94.3495 data_time: 0.5083 time: 0.5965 +2023/06/05 01:36:41 - mmengine - INFO - Epoch(train) [5][ 100/3907] lr: 6.8565e-05 eta: 5:28:18 time: 0.7549 data_time: 0.4127 memory: 6319 loss: 0.0970 +2023/06/05 01:38:00 - mmengine - INFO - Epoch(train) [5][ 200/3907] lr: 6.8219e-05 eta: 5:26:46 time: 0.8272 data_time: 0.3821 memory: 6319 loss: 0.0766 +2023/06/05 01:39:23 - mmengine - INFO - Epoch(train) [5][ 300/3907] lr: 6.7873e-05 eta: 5:25:19 time: 0.8222 data_time: 0.5177 memory: 6319 loss: 0.0816 +2023/06/05 01:40:26 - mmengine - INFO - Exp name: resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1_20230604_215105 +2023/06/05 01:40:47 - mmengine - INFO - Epoch(train) [5][ 400/3907] lr: 6.7526e-05 eta: 5:23:55 time: 0.8675 data_time: 0.7214 memory: 6319 loss: 0.0735 +2023/06/05 01:42:12 - mmengine - INFO - Epoch(train) [5][ 500/3907] lr: 6.7178e-05 eta: 5:22:31 time: 0.8823 data_time: 0.5565 memory: 6319 loss: 0.0755 +2023/06/05 01:43:37 - mmengine - INFO - Epoch(train) [5][ 600/3907] lr: 6.6829e-05 eta: 5:21:07 time: 0.8253 data_time: 0.3923 memory: 6319 loss: 0.0835 +2023/06/05 01:45:02 - mmengine - INFO - Epoch(train) [5][ 700/3907] lr: 6.6480e-05 eta: 5:19:44 time: 0.8386 data_time: 0.1915 memory: 6319 loss: 0.0856 +2023/06/05 01:46:26 - mmengine - INFO - Epoch(train) [5][ 800/3907] lr: 6.6129e-05 eta: 5:18:19 time: 0.8302 data_time: 0.2131 memory: 6319 loss: 0.0802 +2023/06/05 01:47:49 - mmengine - INFO - Epoch(train) [5][ 900/3907] lr: 6.5778e-05 eta: 5:16:53 time: 0.8166 data_time: 0.0225 memory: 6319 loss: 0.0772 +2023/06/05 01:49:28 - mmengine - INFO - Epoch(train) [5][1000/3907] lr: 6.5427e-05 eta: 5:15:49 time: 1.1307 data_time: 0.0010 memory: 6319 loss: 0.0927 +2023/06/05 01:50:49 - mmengine - INFO - Epoch(train) [5][1100/3907] lr: 6.5074e-05 eta: 5:14:19 time: 0.7901 data_time: 0.0009 memory: 6319 loss: 0.0818 +2023/06/05 01:52:13 - mmengine - INFO - Epoch(train) [5][1200/3907] lr: 6.4721e-05 eta: 5:12:54 time: 0.8299 data_time: 0.0009 memory: 6319 loss: 0.0767 +2023/06/05 01:53:36 - mmengine - INFO - Epoch(train) [5][1300/3907] lr: 6.4368e-05 eta: 5:11:28 time: 0.8583 data_time: 0.0523 memory: 6319 loss: 0.0771 +2023/06/05 01:54:41 - mmengine - INFO - Exp name: resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1_20230604_215105 +2023/06/05 01:55:00 - mmengine - INFO - Epoch(train) [5][1400/3907] lr: 6.4014e-05 eta: 5:10:03 time: 0.8422 data_time: 0.1219 memory: 6319 loss: 0.0841 +2023/06/05 01:56:26 - mmengine - INFO - Epoch(train) [5][1500/3907] lr: 6.3659e-05 eta: 5:08:40 time: 0.8982 data_time: 0.2085 memory: 6319 loss: 0.0858 +2023/06/05 01:57:49 - mmengine - INFO - Epoch(train) [5][1600/3907] lr: 6.3303e-05 eta: 5:07:14 time: 0.8958 data_time: 0.1841 memory: 6319 loss: 0.0790 +2023/06/05 01:59:14 - mmengine - INFO - Epoch(train) [5][1700/3907] lr: 6.2948e-05 eta: 5:05:51 time: 0.8508 data_time: 0.4459 memory: 6319 loss: 0.0767 +2023/06/05 02:00:36 - mmengine - INFO - Epoch(train) [5][1800/3907] lr: 6.2591e-05 eta: 5:04:24 time: 0.8304 data_time: 0.2976 memory: 6319 loss: 0.0808 +2023/06/05 02:02:06 - mmengine - INFO - Epoch(train) [5][1900/3907] lr: 6.2234e-05 eta: 5:03:06 time: 1.0674 data_time: 0.4384 memory: 6319 loss: 0.0723 +2023/06/05 02:03:29 - mmengine - INFO - Epoch(train) [5][2000/3907] lr: 6.1877e-05 eta: 5:01:39 time: 0.7913 data_time: 0.2719 memory: 6319 loss: 0.0744 +2023/06/05 02:04:50 - mmengine - INFO - Epoch(train) [5][2100/3907] lr: 6.1519e-05 eta: 5:00:11 time: 0.8297 data_time: 0.3729 memory: 6319 loss: 0.0776 +2023/06/05 02:06:15 - mmengine - INFO - Epoch(train) [5][2200/3907] lr: 6.1161e-05 eta: 4:58:47 time: 0.8959 data_time: 0.4120 memory: 6319 loss: 0.0560 +2023/06/05 02:07:41 - mmengine - INFO - Epoch(train) [5][2300/3907] lr: 6.0802e-05 eta: 4:57:25 time: 0.8639 data_time: 0.7243 memory: 6319 loss: 0.0733 +2023/06/05 02:08:41 - mmengine - INFO - Exp name: resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1_20230604_215105 +2023/06/05 02:09:06 - mmengine - INFO - Epoch(train) [5][2400/3907] lr: 6.0443e-05 eta: 4:56:01 time: 0.8660 data_time: 0.7259 memory: 6319 loss: 0.0529 +2023/06/05 02:10:33 - mmengine - INFO - Epoch(train) [5][2500/3907] lr: 6.0084e-05 eta: 4:54:40 time: 0.8520 data_time: 0.7117 memory: 6319 loss: 0.0676 +2023/06/05 02:12:00 - mmengine - INFO - Epoch(train) [5][2600/3907] lr: 5.9724e-05 eta: 4:53:18 time: 0.8632 data_time: 0.7236 memory: 6319 loss: 0.0729 +2023/06/05 02:13:27 - mmengine - INFO - Epoch(train) [5][2700/3907] lr: 5.9364e-05 eta: 4:51:56 time: 0.8523 data_time: 0.7136 memory: 6319 loss: 0.0760 +2023/06/05 02:14:52 - mmengine - INFO - Epoch(train) [5][2800/3907] lr: 5.9004e-05 eta: 4:50:32 time: 0.8226 data_time: 0.6829 memory: 6319 loss: 0.0676 +2023/06/05 02:16:14 - mmengine - INFO - Epoch(train) [5][2900/3907] lr: 5.8643e-05 eta: 4:49:06 time: 0.8419 data_time: 0.7022 memory: 6319 loss: 0.0744 +2023/06/05 02:17:39 - mmengine - INFO - Epoch(train) [5][3000/3907] lr: 5.8283e-05 eta: 4:47:42 time: 0.8567 data_time: 0.7164 memory: 6319 loss: 0.0823 +2023/06/05 02:19:13 - mmengine - INFO - Epoch(train) [5][3100/3907] lr: 5.7922e-05 eta: 4:46:28 time: 1.1178 data_time: 0.9789 memory: 6319 loss: 0.0633 +2023/06/05 02:20:59 - mmengine - INFO - Epoch(train) [5][3200/3907] lr: 5.7560e-05 eta: 4:45:26 time: 0.7396 data_time: 0.6000 memory: 6319 loss: 0.0869 +2023/06/05 02:22:13 - mmengine - INFO - Epoch(train) [5][3300/3907] lr: 5.7199e-05 eta: 4:43:51 time: 0.8110 data_time: 0.6714 memory: 6319 loss: 0.0845 +2023/06/05 02:23:08 - mmengine - INFO - Exp name: resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1_20230604_215105 +2023/06/05 02:23:32 - mmengine - INFO - Epoch(train) [5][3400/3907] lr: 5.6838e-05 eta: 4:42:20 time: 0.7934 data_time: 0.6545 memory: 6319 loss: 0.0704 +2023/06/05 02:24:55 - mmengine - INFO - Epoch(train) [5][3500/3907] lr: 5.6476e-05 eta: 4:40:54 time: 0.8120 data_time: 0.6725 memory: 6319 loss: 0.0751 +2023/06/05 02:26:18 - mmengine - INFO - Epoch(train) [5][3600/3907] lr: 5.6114e-05 eta: 4:39:28 time: 0.8364 data_time: 0.6955 memory: 6319 loss: 0.0762 +2023/06/05 02:27:43 - mmengine - INFO - Epoch(train) [5][3700/3907] lr: 5.5753e-05 eta: 4:38:03 time: 0.8483 data_time: 0.7081 memory: 6319 loss: 0.0769 +2023/06/05 02:29:06 - mmengine - INFO - Epoch(train) [5][3800/3907] lr: 5.5391e-05 eta: 4:36:37 time: 0.8153 data_time: 0.6760 memory: 6319 loss: 0.0689 +2023/06/05 02:30:29 - mmengine - INFO - Epoch(train) [5][3900/3907] lr: 5.5029e-05 eta: 4:35:12 time: 0.8264 data_time: 0.6872 memory: 6319 loss: 0.0734 +2023/06/05 02:30:37 - mmengine - INFO - Exp name: resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1_20230604_215105 +2023/06/05 02:30:37 - mmengine - INFO - Saving checkpoint at 5 epochs +2023/06/05 02:31:22 - mmengine - INFO - Epoch(val) [5][57/57] accuracy/top1: 95.7552 data_time: 0.4929 time: 0.5795 +2023/06/05 02:32:49 - mmengine - INFO - Epoch(train) [6][ 100/3907] lr: 5.4642e-05 eta: 4:33:46 time: 0.8089 data_time: 0.6675 memory: 6319 loss: 0.0722 +2023/06/05 02:34:13 - mmengine - INFO - Epoch(train) [6][ 200/3907] lr: 5.4280e-05 eta: 4:32:20 time: 0.8361 data_time: 0.6952 memory: 6319 loss: 0.0818 +2023/06/05 02:35:36 - mmengine - INFO - Epoch(train) [6][ 300/3907] lr: 5.3918e-05 eta: 4:30:55 time: 0.8487 data_time: 0.7094 memory: 6319 loss: 0.0731 +2023/06/05 02:37:00 - mmengine - INFO - Epoch(train) [6][ 400/3907] lr: 5.3556e-05 eta: 4:29:30 time: 0.8373 data_time: 0.6976 memory: 6319 loss: 0.0664 +2023/06/05 02:38:05 - mmengine - INFO - Exp name: resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1_20230604_215105 +2023/06/05 02:38:36 - mmengine - INFO - Epoch(train) [6][ 500/3907] lr: 5.3195e-05 eta: 4:28:16 time: 1.1845 data_time: 1.0439 memory: 6319 loss: 0.0747 +2023/06/05 02:40:53 - mmengine - INFO - Epoch(train) [6][ 600/3907] lr: 5.2833e-05 eta: 4:27:40 time: 0.7011 data_time: 0.1094 memory: 6319 loss: 0.0690 +2023/06/05 02:42:04 - mmengine - INFO - Epoch(train) [6][ 700/3907] lr: 5.2472e-05 eta: 4:26:04 time: 0.7453 data_time: 0.0411 memory: 6319 loss: 0.0921 +2023/06/05 02:43:23 - mmengine - INFO - Epoch(train) [6][ 800/3907] lr: 5.2111e-05 eta: 4:24:33 time: 0.8559 data_time: 0.1596 memory: 6319 loss: 0.0757 +2023/06/05 02:44:41 - mmengine - INFO - Epoch(train) [6][ 900/3907] lr: 5.1750e-05 eta: 4:23:03 time: 0.7837 data_time: 0.1250 memory: 6319 loss: 0.0715 +2023/06/05 02:46:03 - mmengine - INFO - Epoch(train) [6][1000/3907] lr: 5.1389e-05 eta: 4:21:35 time: 0.8438 data_time: 0.0907 memory: 6319 loss: 0.0576 +2023/06/05 02:47:29 - mmengine - INFO - Epoch(train) [6][1100/3907] lr: 5.1029e-05 eta: 4:20:12 time: 0.8392 data_time: 0.2350 memory: 6319 loss: 0.0774 +2023/06/05 02:48:51 - mmengine - INFO - Epoch(train) [6][1200/3907] lr: 5.0668e-05 eta: 4:18:44 time: 0.8808 data_time: 0.2394 memory: 6319 loss: 0.0693 +2023/06/05 02:50:14 - mmengine - INFO - Epoch(train) [6][1300/3907] lr: 5.0308e-05 eta: 4:17:18 time: 0.8061 data_time: 0.1434 memory: 6319 loss: 0.0765 +2023/06/05 02:51:37 - mmengine - INFO - Epoch(train) [6][1400/3907] lr: 4.9949e-05 eta: 4:15:53 time: 0.8857 data_time: 0.5710 memory: 6319 loss: 0.0802 +2023/06/05 02:52:29 - mmengine - INFO - Exp name: resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1_20230604_215105 +2023/06/05 02:53:01 - mmengine - INFO - Epoch(train) [6][1500/3907] lr: 4.9589e-05 eta: 4:14:27 time: 0.8002 data_time: 0.6611 memory: 6319 loss: 0.0676 +2023/06/05 02:54:22 - mmengine - INFO - Epoch(train) [6][1600/3907] lr: 4.9230e-05 eta: 4:12:59 time: 0.8128 data_time: 0.6735 memory: 6319 loss: 0.0738 +2023/06/05 02:55:45 - mmengine - INFO - Epoch(train) [6][1700/3907] lr: 4.8871e-05 eta: 4:11:33 time: 0.8157 data_time: 0.6756 memory: 6319 loss: 0.0686 +2023/06/05 02:57:08 - mmengine - INFO - Epoch(train) [6][1800/3907] lr: 4.8513e-05 eta: 4:10:07 time: 0.8167 data_time: 0.6765 memory: 6319 loss: 0.0659 +2023/06/05 02:58:29 - mmengine - INFO - Epoch(train) [6][1900/3907] lr: 4.8155e-05 eta: 4:08:40 time: 0.8136 data_time: 0.6737 memory: 6319 loss: 0.0746 +2023/06/05 02:59:51 - mmengine - INFO - Epoch(train) [6][2000/3907] lr: 4.7798e-05 eta: 4:07:13 time: 0.8305 data_time: 0.6854 memory: 6319 loss: 0.0602 +2023/06/05 03:01:16 - mmengine - INFO - Epoch(train) [6][2100/3907] lr: 4.7441e-05 eta: 4:05:49 time: 0.8643 data_time: 0.7252 memory: 6319 loss: 0.0703 +2023/06/05 03:02:38 - mmengine - INFO - Epoch(train) [6][2200/3907] lr: 4.7084e-05 eta: 4:04:22 time: 0.8169 data_time: 0.6773 memory: 6319 loss: 0.0628 +2023/06/05 03:04:00 - mmengine - INFO - Epoch(train) [6][2300/3907] lr: 4.6729e-05 eta: 4:02:56 time: 0.7715 data_time: 0.6327 memory: 6319 loss: 0.0701 +2023/06/05 03:05:24 - mmengine - INFO - Epoch(train) [6][2400/3907] lr: 4.6373e-05 eta: 4:01:30 time: 0.8264 data_time: 0.6743 memory: 6319 loss: 0.0657 +2023/06/05 03:06:14 - mmengine - INFO - Exp name: resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1_20230604_215105 +2023/06/05 03:06:46 - mmengine - INFO - Epoch(train) [6][2500/3907] lr: 4.6018e-05 eta: 4:00:04 time: 0.8313 data_time: 0.6913 memory: 6319 loss: 0.0773 +2023/06/05 03:08:10 - mmengine - INFO - Epoch(train) [6][2600/3907] lr: 4.5664e-05 eta: 3:58:39 time: 0.8750 data_time: 0.7342 memory: 6319 loss: 0.0729 +2023/06/05 03:09:35 - mmengine - INFO - Epoch(train) [6][2700/3907] lr: 4.5310e-05 eta: 3:57:14 time: 0.8533 data_time: 0.7129 memory: 6319 loss: 0.0873 +2023/06/05 03:10:57 - mmengine - INFO - Epoch(train) [6][2800/3907] lr: 4.4957e-05 eta: 3:55:48 time: 0.8568 data_time: 0.7101 memory: 6319 loss: 0.0750 +2023/06/05 03:12:23 - mmengine - INFO - Epoch(train) [6][2900/3907] lr: 4.4605e-05 eta: 3:54:24 time: 0.8295 data_time: 0.6771 memory: 6319 loss: 0.0782 +2023/06/05 03:13:46 - mmengine - INFO - Epoch(train) [6][3000/3907] lr: 4.4253e-05 eta: 3:52:59 time: 0.8532 data_time: 0.7131 memory: 6319 loss: 0.0717 +2023/06/05 03:15:09 - mmengine - INFO - Epoch(train) [6][3100/3907] lr: 4.3902e-05 eta: 3:51:33 time: 0.8811 data_time: 0.7379 memory: 6319 loss: 0.0704 +2023/06/05 03:16:30 - mmengine - INFO - Epoch(train) [6][3200/3907] lr: 4.3552e-05 eta: 3:50:06 time: 0.8030 data_time: 0.6639 memory: 6319 loss: 0.0680 +2023/06/05 03:17:55 - mmengine - INFO - Epoch(train) [6][3300/3907] lr: 4.3202e-05 eta: 3:48:42 time: 0.8265 data_time: 0.6869 memory: 6319 loss: 0.0707 +2023/06/05 03:19:17 - mmengine - INFO - Epoch(train) [6][3400/3907] lr: 4.2854e-05 eta: 3:47:16 time: 0.8276 data_time: 0.6874 memory: 6319 loss: 0.0793 +2023/06/05 03:20:07 - mmengine - INFO - Exp name: resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1_20230604_215105 +2023/06/05 03:20:40 - mmengine - INFO - Epoch(train) [6][3500/3907] lr: 4.2506e-05 eta: 3:45:50 time: 0.7719 data_time: 0.6331 memory: 6319 loss: 0.0716 +2023/06/05 03:22:03 - mmengine - INFO - Epoch(train) [6][3600/3907] lr: 4.2158e-05 eta: 3:44:24 time: 0.8336 data_time: 0.6937 memory: 6319 loss: 0.0946 +2023/06/05 03:23:25 - mmengine - INFO - Epoch(train) [6][3700/3907] lr: 4.1812e-05 eta: 3:42:58 time: 0.8356 data_time: 0.6967 memory: 6319 loss: 0.0730 +2023/06/05 03:24:49 - mmengine - INFO - Epoch(train) [6][3800/3907] lr: 4.1466e-05 eta: 3:41:34 time: 0.8337 data_time: 0.6929 memory: 6319 loss: 0.0669 +2023/06/05 03:26:11 - mmengine - INFO - Epoch(train) [6][3900/3907] lr: 4.1122e-05 eta: 3:40:07 time: 0.8037 data_time: 0.6638 memory: 6319 loss: 0.0629 +2023/06/05 03:26:12 - mmengine - INFO - Exp name: resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1_20230604_215105 +2023/06/05 03:26:12 - mmengine - INFO - Saving checkpoint at 6 epochs +2023/06/05 03:26:56 - mmengine - INFO - Epoch(val) [6][57/57] accuracy/top1: 96.2527 data_time: 0.4765 time: 0.5658 +2023/06/05 03:28:22 - mmengine - INFO - Epoch(train) [7][ 100/3907] lr: 4.0754e-05 eta: 3:38:35 time: 0.8169 data_time: 0.6782 memory: 6319 loss: 0.0686 +2023/06/05 03:29:47 - mmengine - INFO - Epoch(train) [7][ 200/3907] lr: 4.0411e-05 eta: 3:37:10 time: 0.8342 data_time: 0.6940 memory: 6319 loss: 0.0583 +2023/06/05 03:31:08 - mmengine - INFO - Epoch(train) [7][ 300/3907] lr: 4.0069e-05 eta: 3:35:44 time: 0.8712 data_time: 0.6737 memory: 6319 loss: 0.0575 +2023/06/05 03:32:30 - mmengine - INFO - Epoch(train) [7][ 400/3907] lr: 3.9729e-05 eta: 3:34:17 time: 0.8164 data_time: 0.6282 memory: 6319 loss: 0.0650 +2023/06/05 03:33:51 - mmengine - INFO - Epoch(train) [7][ 500/3907] lr: 3.9389e-05 eta: 3:32:51 time: 0.8275 data_time: 0.6889 memory: 6319 loss: 0.0618 +2023/06/05 03:34:41 - mmengine - INFO - Exp name: resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1_20230604_215105 +2023/06/05 03:35:14 - mmengine - INFO - Epoch(train) [7][ 600/3907] lr: 3.9050e-05 eta: 3:31:25 time: 0.8018 data_time: 0.5287 memory: 6319 loss: 0.0588 +2023/06/05 03:36:37 - mmengine - INFO - Epoch(train) [7][ 700/3907] lr: 3.8712e-05 eta: 3:30:00 time: 0.8472 data_time: 0.4973 memory: 6319 loss: 0.0567 +2023/06/05 03:38:00 - mmengine - INFO - Epoch(train) [7][ 800/3907] lr: 3.8375e-05 eta: 3:28:35 time: 0.7817 data_time: 0.4355 memory: 6319 loss: 0.0645 +2023/06/05 03:39:20 - mmengine - INFO - Epoch(train) [7][ 900/3907] lr: 3.8039e-05 eta: 3:27:08 time: 0.8372 data_time: 0.4605 memory: 6319 loss: 0.0641 +2023/06/05 03:40:43 - mmengine - INFO - Epoch(train) [7][1000/3907] lr: 3.7705e-05 eta: 3:25:43 time: 0.8847 data_time: 0.6041 memory: 6319 loss: 0.0652 +2023/06/05 03:42:05 - mmengine - INFO - Epoch(train) [7][1100/3907] lr: 3.7371e-05 eta: 3:24:17 time: 0.8795 data_time: 0.7392 memory: 6319 loss: 0.0659 +2023/06/05 03:43:31 - mmengine - INFO - Epoch(train) [7][1200/3907] lr: 3.7039e-05 eta: 3:22:54 time: 0.8919 data_time: 0.7530 memory: 6319 loss: 0.0642 +2023/06/05 03:44:53 - mmengine - INFO - Epoch(train) [7][1300/3907] lr: 3.6708e-05 eta: 3:21:28 time: 0.8078 data_time: 0.6683 memory: 6319 loss: 0.0771 +2023/06/05 03:46:16 - mmengine - INFO - Epoch(train) [7][1400/3907] lr: 3.6378e-05 eta: 3:20:02 time: 0.8425 data_time: 0.7031 memory: 6319 loss: 0.0741 +2023/06/05 03:47:38 - mmengine - INFO - Epoch(train) [7][1500/3907] lr: 3.6049e-05 eta: 3:18:37 time: 0.8235 data_time: 0.6847 memory: 6319 loss: 0.0674 +2023/06/05 03:48:27 - mmengine - INFO - Exp name: resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1_20230604_215105 +2023/06/05 03:48:59 - mmengine - INFO - Epoch(train) [7][1600/3907] lr: 3.5721e-05 eta: 3:17:11 time: 0.7755 data_time: 0.6364 memory: 6319 loss: 0.0502 +2023/06/05 03:50:21 - mmengine - INFO - Epoch(train) [7][1700/3907] lr: 3.5395e-05 eta: 3:15:45 time: 0.8080 data_time: 0.6695 memory: 6319 loss: 0.0652 +2023/06/05 03:51:45 - mmengine - INFO - Epoch(train) [7][1800/3907] lr: 3.5070e-05 eta: 3:14:20 time: 0.9209 data_time: 0.7814 memory: 6319 loss: 0.0626 +2023/06/05 03:53:08 - mmengine - INFO - Epoch(train) [7][1900/3907] lr: 3.4746e-05 eta: 3:12:56 time: 0.8842 data_time: 0.7359 memory: 6319 loss: 0.0720 +2023/06/05 03:54:30 - mmengine - INFO - Epoch(train) [7][2000/3907] lr: 3.4424e-05 eta: 3:11:30 time: 0.8189 data_time: 0.6787 memory: 6319 loss: 0.0741 +2023/06/05 03:55:51 - mmengine - INFO - Epoch(train) [7][2100/3907] lr: 3.4103e-05 eta: 3:10:04 time: 0.8333 data_time: 0.6938 memory: 6319 loss: 0.0623 +2023/06/05 03:57:14 - mmengine - INFO - Epoch(train) [7][2200/3907] lr: 3.3783e-05 eta: 3:08:39 time: 0.8399 data_time: 0.7014 memory: 6319 loss: 0.0675 +2023/06/05 03:58:36 - mmengine - INFO - Epoch(train) [7][2300/3907] lr: 3.3465e-05 eta: 3:07:13 time: 0.8338 data_time: 0.6722 memory: 6319 loss: 0.0717 +2023/06/05 03:59:56 - mmengine - INFO - Epoch(train) [7][2400/3907] lr: 3.3148e-05 eta: 3:05:47 time: 0.8269 data_time: 0.6873 memory: 6319 loss: 0.0733 +2023/06/05 04:01:19 - mmengine - INFO - Epoch(train) [7][2500/3907] lr: 3.2832e-05 eta: 3:04:22 time: 0.8303 data_time: 0.6770 memory: 6319 loss: 0.0654 +2023/06/05 04:02:08 - mmengine - INFO - Exp name: resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1_20230604_215105 +2023/06/05 04:02:41 - mmengine - INFO - Epoch(train) [7][2600/3907] lr: 3.2518e-05 eta: 3:02:57 time: 0.7735 data_time: 0.6336 memory: 6319 loss: 0.0649 +2023/06/05 04:04:02 - mmengine - INFO - Epoch(train) [7][2700/3907] lr: 3.2205e-05 eta: 3:01:31 time: 0.7807 data_time: 0.6411 memory: 6319 loss: 0.0847 +2023/06/05 04:05:23 - mmengine - INFO - Epoch(train) [7][2800/3907] lr: 3.1894e-05 eta: 3:00:05 time: 0.7995 data_time: 0.6600 memory: 6319 loss: 0.0547 +2023/06/05 04:06:45 - mmengine - INFO - Epoch(train) [7][2900/3907] lr: 3.1584e-05 eta: 2:58:40 time: 0.8717 data_time: 0.7324 memory: 6319 loss: 0.0627 +2023/06/05 04:08:08 - mmengine - INFO - Epoch(train) [7][3000/3907] lr: 3.1276e-05 eta: 2:57:15 time: 0.7923 data_time: 0.6506 memory: 6319 loss: 0.0627 +2023/06/05 04:09:30 - mmengine - INFO - Epoch(train) [7][3100/3907] lr: 3.0969e-05 eta: 2:55:49 time: 0.9122 data_time: 0.7728 memory: 6319 loss: 0.0731 +2023/06/05 04:10:51 - mmengine - INFO - Epoch(train) [7][3200/3907] lr: 3.0664e-05 eta: 2:54:24 time: 0.8341 data_time: 0.6930 memory: 6319 loss: 0.0687 +2023/06/05 04:12:13 - mmengine - INFO - Epoch(train) [7][3300/3907] lr: 3.0360e-05 eta: 2:52:58 time: 0.8031 data_time: 0.6637 memory: 6319 loss: 0.0607 +2023/06/05 04:13:35 - mmengine - INFO - Epoch(train) [7][3400/3907] lr: 3.0058e-05 eta: 2:51:33 time: 0.7988 data_time: 0.6585 memory: 6319 loss: 0.0704 +2023/06/05 04:14:56 - mmengine - INFO - Epoch(train) [7][3500/3907] lr: 2.9758e-05 eta: 2:50:08 time: 0.8350 data_time: 0.6954 memory: 6319 loss: 0.0622 +2023/06/05 04:15:45 - mmengine - INFO - Exp name: resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1_20230604_215105 +2023/06/05 04:16:19 - mmengine - INFO - Epoch(train) [7][3600/3907] lr: 2.9459e-05 eta: 2:48:43 time: 0.9301 data_time: 0.7898 memory: 6319 loss: 0.0627 +2023/06/05 04:17:43 - mmengine - INFO - Epoch(train) [7][3700/3907] lr: 2.9162e-05 eta: 2:47:19 time: 0.8300 data_time: 0.6892 memory: 6319 loss: 0.0616 +2023/06/05 04:19:05 - mmengine - INFO - Epoch(train) [7][3800/3907] lr: 2.8867e-05 eta: 2:45:54 time: 0.7846 data_time: 0.6450 memory: 6319 loss: 0.0561 +2023/06/05 04:20:29 - mmengine - INFO - Epoch(train) [7][3900/3907] lr: 2.8573e-05 eta: 2:44:29 time: 0.8708 data_time: 0.7308 memory: 6319 loss: 0.0694 +2023/06/05 04:20:36 - mmengine - INFO - Exp name: resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1_20230604_215105 +2023/06/05 04:20:36 - mmengine - INFO - Saving checkpoint at 7 epochs +2023/06/05 04:21:21 - mmengine - INFO - Epoch(val) [7][57/57] accuracy/top1: 96.9904 data_time: 0.4890 time: 0.5752 +2023/06/05 04:22:46 - mmengine - INFO - Epoch(train) [8][ 100/3907] lr: 2.8261e-05 eta: 2:43:00 time: 0.7999 data_time: 0.5401 memory: 6319 loss: 0.0604 +2023/06/05 04:24:09 - mmengine - INFO - Epoch(train) [8][ 200/3907] lr: 2.7971e-05 eta: 2:41:35 time: 0.8246 data_time: 0.4197 memory: 6319 loss: 0.0597 +2023/06/05 04:25:31 - mmengine - INFO - Epoch(train) [8][ 300/3907] lr: 2.7682e-05 eta: 2:40:10 time: 0.8386 data_time: 0.3896 memory: 6319 loss: 0.0603 +2023/06/05 04:26:55 - mmengine - INFO - Epoch(train) [8][ 400/3907] lr: 2.7395e-05 eta: 2:38:46 time: 0.8738 data_time: 0.2583 memory: 6319 loss: 0.0683 +2023/06/05 04:28:20 - mmengine - INFO - Epoch(train) [8][ 500/3907] lr: 2.7111e-05 eta: 2:37:22 time: 0.8239 data_time: 0.0651 memory: 6319 loss: 0.0612 +2023/06/05 04:29:43 - mmengine - INFO - Epoch(train) [8][ 600/3907] lr: 2.6828e-05 eta: 2:35:58 time: 0.8147 data_time: 0.0008 memory: 6319 loss: 0.0672 +2023/06/05 04:30:25 - mmengine - INFO - Exp name: resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1_20230604_215105 +2023/06/05 04:31:07 - mmengine - INFO - Epoch(train) [8][ 700/3907] lr: 2.6546e-05 eta: 2:34:34 time: 0.8189 data_time: 0.0008 memory: 6319 loss: 0.0698 +2023/06/05 04:32:29 - mmengine - INFO - Epoch(train) [8][ 800/3907] lr: 2.6267e-05 eta: 2:33:08 time: 0.8906 data_time: 0.0010 memory: 6319 loss: 0.0612 +2023/06/05 04:33:48 - mmengine - INFO - Epoch(train) [8][ 900/3907] lr: 2.5989e-05 eta: 2:31:43 time: 0.8044 data_time: 0.0009 memory: 6319 loss: 0.0633 +2023/06/05 04:35:07 - mmengine - INFO - Epoch(train) [8][1000/3907] lr: 2.5714e-05 eta: 2:30:16 time: 0.7781 data_time: 0.0009 memory: 6319 loss: 0.0734 +2023/06/05 04:36:30 - mmengine - INFO - Epoch(train) [8][1100/3907] lr: 2.5440e-05 eta: 2:28:52 time: 0.8685 data_time: 0.0011 memory: 6319 loss: 0.0616 +2023/06/05 04:37:57 - mmengine - INFO - Epoch(train) [8][1200/3907] lr: 2.5168e-05 eta: 2:27:29 time: 0.8432 data_time: 0.0008 memory: 6319 loss: 0.0597 +2023/06/05 04:39:16 - mmengine - INFO - Epoch(train) [8][1300/3907] lr: 2.4898e-05 eta: 2:26:03 time: 0.8371 data_time: 0.0012 memory: 6319 loss: 0.0603 +2023/06/05 04:40:36 - mmengine - INFO - Epoch(train) [8][1400/3907] lr: 2.4630e-05 eta: 2:24:37 time: 0.8273 data_time: 0.0011 memory: 6319 loss: 0.0711 +2023/06/05 04:41:57 - mmengine - INFO - Epoch(train) [8][1500/3907] lr: 2.4364e-05 eta: 2:23:12 time: 0.8597 data_time: 0.0008 memory: 6319 loss: 0.0634 +2023/06/05 04:43:19 - mmengine - INFO - Epoch(train) [8][1600/3907] lr: 2.4100e-05 eta: 2:21:47 time: 0.8000 data_time: 0.0008 memory: 6319 loss: 0.0655 +2023/06/05 04:43:58 - mmengine - INFO - Exp name: resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1_20230604_215105 +2023/06/05 04:44:37 - mmengine - INFO - Epoch(train) [8][1700/3907] lr: 2.3838e-05 eta: 2:20:21 time: 0.7644 data_time: 0.0008 memory: 6319 loss: 0.0633 +2023/06/05 04:45:55 - mmengine - INFO - Epoch(train) [8][1800/3907] lr: 2.3578e-05 eta: 2:18:55 time: 0.8001 data_time: 0.1444 memory: 6319 loss: 0.0640 +2023/06/05 04:47:19 - mmengine - INFO - Epoch(train) [8][1900/3907] lr: 2.3320e-05 eta: 2:17:31 time: 0.8051 data_time: 0.4547 memory: 6319 loss: 0.0636 +2023/06/05 04:48:41 - mmengine - INFO - Epoch(train) [8][2000/3907] lr: 2.3064e-05 eta: 2:16:06 time: 0.8390 data_time: 0.4230 memory: 6319 loss: 0.0573 +2023/06/05 04:50:04 - mmengine - INFO - Epoch(train) [8][2100/3907] lr: 2.2810e-05 eta: 2:14:42 time: 0.8745 data_time: 0.3388 memory: 6319 loss: 0.0654 +2023/06/05 04:51:25 - mmengine - INFO - Epoch(train) [8][2200/3907] lr: 2.2558e-05 eta: 2:13:17 time: 0.7765 data_time: 0.5325 memory: 6319 loss: 0.0680 +2023/06/05 04:52:45 - mmengine - INFO - Epoch(train) [8][2300/3907] lr: 2.2309e-05 eta: 2:11:52 time: 0.7840 data_time: 0.5308 memory: 6319 loss: 0.0636 +2023/06/05 04:54:05 - mmengine - INFO - Epoch(train) [8][2400/3907] lr: 2.2061e-05 eta: 2:10:27 time: 0.7755 data_time: 0.5517 memory: 6319 loss: 0.0533 +2023/06/05 04:55:25 - mmengine - INFO - Epoch(train) [8][2500/3907] lr: 2.1816e-05 eta: 2:09:01 time: 0.8122 data_time: 0.6716 memory: 6319 loss: 0.0647 +2023/06/05 04:56:47 - mmengine - INFO - Epoch(train) [8][2600/3907] lr: 2.1572e-05 eta: 2:07:37 time: 0.8186 data_time: 0.6784 memory: 6319 loss: 0.0733 +2023/06/05 04:57:27 - mmengine - INFO - Exp name: resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1_20230604_215105 +2023/06/05 04:58:07 - mmengine - INFO - Epoch(train) [8][2700/3907] lr: 2.1331e-05 eta: 2:06:12 time: 0.7645 data_time: 0.6244 memory: 6319 loss: 0.0669 +2023/06/05 04:59:29 - mmengine - INFO - Epoch(train) [8][2800/3907] lr: 2.1092e-05 eta: 2:04:47 time: 0.8165 data_time: 0.6771 memory: 6319 loss: 0.0550 +2023/06/05 05:00:49 - mmengine - INFO - Epoch(train) [8][2900/3907] lr: 2.0855e-05 eta: 2:03:22 time: 0.8316 data_time: 0.6835 memory: 6319 loss: 0.0624 +2023/06/05 05:02:10 - mmengine - INFO - Epoch(train) [8][3000/3907] lr: 2.0621e-05 eta: 2:01:57 time: 0.8103 data_time: 0.6708 memory: 6319 loss: 0.0638 +2023/06/05 05:03:32 - mmengine - INFO - Epoch(train) [8][3100/3907] lr: 2.0388e-05 eta: 2:00:33 time: 0.7899 data_time: 0.6447 memory: 6319 loss: 0.0755 +2023/06/05 05:04:53 - mmengine - INFO - Epoch(train) [8][3200/3907] lr: 2.0158e-05 eta: 1:59:08 time: 0.8338 data_time: 0.6944 memory: 6319 loss: 0.0700 +2023/06/05 05:06:16 - mmengine - INFO - Epoch(train) [8][3300/3907] lr: 1.9930e-05 eta: 1:57:44 time: 0.8094 data_time: 0.6702 memory: 6319 loss: 0.0658 +2023/06/05 05:07:37 - mmengine - INFO - Epoch(train) [8][3400/3907] lr: 1.9705e-05 eta: 1:56:19 time: 0.8472 data_time: 0.7080 memory: 6319 loss: 0.0665 +2023/06/05 05:08:59 - mmengine - INFO - Epoch(train) [8][3500/3907] lr: 1.9481e-05 eta: 1:54:55 time: 0.8574 data_time: 0.7069 memory: 6319 loss: 0.0550 +2023/06/05 05:10:21 - mmengine - INFO - Epoch(train) [8][3600/3907] lr: 1.9260e-05 eta: 1:53:31 time: 0.7703 data_time: 0.6303 memory: 6319 loss: 0.0516 +2023/06/05 05:11:03 - mmengine - INFO - Exp name: resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1_20230604_215105 +2023/06/05 05:11:45 - mmengine - INFO - Epoch(train) [8][3700/3907] lr: 1.9042e-05 eta: 1:52:07 time: 0.8212 data_time: 0.6811 memory: 6319 loss: 0.0694 +2023/06/05 05:13:07 - mmengine - INFO - Epoch(train) [8][3800/3907] lr: 1.8825e-05 eta: 1:50:42 time: 0.8238 data_time: 0.6838 memory: 6319 loss: 0.0541 +2023/06/05 05:14:27 - mmengine - INFO - Epoch(train) [8][3900/3907] lr: 1.8611e-05 eta: 1:49:17 time: 0.7984 data_time: 0.6598 memory: 6319 loss: 0.0647 +2023/06/05 05:14:27 - mmengine - INFO - Exp name: resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1_20230604_215105 +2023/06/05 05:14:27 - mmengine - INFO - Saving checkpoint at 8 epochs +2023/06/05 05:15:11 - mmengine - INFO - Epoch(val) [8][57/57] accuracy/top1: 97.2096 data_time: 0.4780 time: 0.5643 +2023/06/05 05:16:37 - mmengine - INFO - Epoch(train) [9][ 100/3907] lr: 1.8385e-05 eta: 1:47:47 time: 0.9074 data_time: 0.5645 memory: 6319 loss: 0.0735 +2023/06/05 05:18:01 - mmengine - INFO - Epoch(train) [9][ 200/3907] lr: 1.8176e-05 eta: 1:46:23 time: 0.8670 data_time: 0.1210 memory: 6319 loss: 0.0578 +2023/06/05 05:19:23 - mmengine - INFO - Epoch(train) [9][ 300/3907] lr: 1.7969e-05 eta: 1:44:59 time: 0.8499 data_time: 0.0013 memory: 6319 loss: 0.0513 +2023/06/05 05:20:44 - mmengine - INFO - Epoch(train) [9][ 400/3907] lr: 1.7765e-05 eta: 1:43:34 time: 0.7750 data_time: 0.0008 memory: 6319 loss: 0.0602 +2023/06/05 05:22:03 - mmengine - INFO - Epoch(train) [9][ 500/3907] lr: 1.7563e-05 eta: 1:42:09 time: 0.8156 data_time: 0.1222 memory: 6319 loss: 0.0654 +2023/06/05 05:23:24 - mmengine - INFO - Epoch(train) [9][ 600/3907] lr: 1.7363e-05 eta: 1:40:45 time: 0.7604 data_time: 0.2080 memory: 6319 loss: 0.0599 +2023/06/05 05:24:44 - mmengine - INFO - Epoch(train) [9][ 700/3907] lr: 1.7166e-05 eta: 1:39:20 time: 0.8643 data_time: 0.1439 memory: 6319 loss: 0.0604 +2023/06/05 05:25:24 - mmengine - INFO - Exp name: resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1_20230604_215105 +2023/06/05 05:26:06 - mmengine - INFO - Epoch(train) [9][ 800/3907] lr: 1.6971e-05 eta: 1:37:56 time: 0.7845 data_time: 0.0009 memory: 6319 loss: 0.0621 +2023/06/05 05:27:26 - mmengine - INFO - Epoch(train) [9][ 900/3907] lr: 1.6779e-05 eta: 1:36:32 time: 0.7562 data_time: 0.0558 memory: 6319 loss: 0.0572 +2023/06/05 05:28:47 - mmengine - INFO - Epoch(train) [9][1000/3907] lr: 1.6589e-05 eta: 1:35:07 time: 0.7898 data_time: 0.1641 memory: 6319 loss: 0.0689 +2023/06/05 05:30:08 - mmengine - INFO - Epoch(train) [9][1100/3907] lr: 1.6402e-05 eta: 1:33:43 time: 0.7986 data_time: 0.4035 memory: 6319 loss: 0.0685 +2023/06/05 05:31:30 - mmengine - INFO - Epoch(train) [9][1200/3907] lr: 1.6217e-05 eta: 1:32:19 time: 0.8010 data_time: 0.6586 memory: 6319 loss: 0.0584 +2023/06/05 05:32:52 - mmengine - INFO - Epoch(train) [9][1300/3907] lr: 1.6035e-05 eta: 1:30:55 time: 0.8267 data_time: 0.6859 memory: 6319 loss: 0.0641 +2023/06/05 05:34:12 - mmengine - INFO - Epoch(train) [9][1400/3907] lr: 1.5855e-05 eta: 1:29:30 time: 0.8208 data_time: 0.6793 memory: 6319 loss: 0.0543 +2023/06/05 05:35:32 - mmengine - INFO - Epoch(train) [9][1500/3907] lr: 1.5678e-05 eta: 1:28:06 time: 0.8107 data_time: 0.6697 memory: 6319 loss: 0.0569 +2023/06/05 05:36:54 - mmengine - INFO - Epoch(train) [9][1600/3907] lr: 1.5503e-05 eta: 1:26:42 time: 0.8180 data_time: 0.6770 memory: 6319 loss: 0.0595 +2023/06/05 05:38:16 - mmengine - INFO - Epoch(train) [9][1700/3907] lr: 1.5331e-05 eta: 1:25:18 time: 0.8163 data_time: 0.6756 memory: 6319 loss: 0.0621 +2023/06/05 05:38:50 - mmengine - INFO - Exp name: resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1_20230604_215105 +2023/06/05 05:39:37 - mmengine - INFO - Epoch(train) [9][1800/3907] lr: 1.5162e-05 eta: 1:23:53 time: 0.8037 data_time: 0.6640 memory: 6319 loss: 0.0572 +2023/06/05 05:40:58 - mmengine - INFO - Epoch(train) [9][1900/3907] lr: 1.4995e-05 eta: 1:22:29 time: 0.8550 data_time: 0.7145 memory: 6319 loss: 0.0685 +2023/06/05 05:42:18 - mmengine - INFO - Epoch(train) [9][2000/3907] lr: 1.4830e-05 eta: 1:21:05 time: 0.8321 data_time: 0.6928 memory: 6319 loss: 0.0700 +2023/06/05 05:43:39 - mmengine - INFO - Epoch(train) [9][2100/3907] lr: 1.4668e-05 eta: 1:19:41 time: 0.8226 data_time: 0.6831 memory: 6319 loss: 0.0555 +2023/06/05 05:44:59 - mmengine - INFO - Epoch(train) [9][2200/3907] lr: 1.4509e-05 eta: 1:18:16 time: 0.8092 data_time: 0.6699 memory: 6319 loss: 0.0611 +2023/06/05 05:46:20 - mmengine - INFO - Epoch(train) [9][2300/3907] lr: 1.4353e-05 eta: 1:16:52 time: 0.7856 data_time: 0.6460 memory: 6319 loss: 0.0650 +2023/06/05 05:47:43 - mmengine - INFO - Epoch(train) [9][2400/3907] lr: 1.4199e-05 eta: 1:15:29 time: 0.8703 data_time: 0.7299 memory: 6319 loss: 0.0577 +2023/06/05 05:49:05 - mmengine - INFO - Epoch(train) [9][2500/3907] lr: 1.4047e-05 eta: 1:14:05 time: 0.8197 data_time: 0.6807 memory: 6319 loss: 0.0556 +2023/06/05 05:50:25 - mmengine - INFO - Epoch(train) [9][2600/3907] lr: 1.3899e-05 eta: 1:12:40 time: 0.8049 data_time: 0.6651 memory: 6319 loss: 0.0723 +2023/06/05 05:51:48 - mmengine - INFO - Epoch(train) [9][2700/3907] lr: 1.3753e-05 eta: 1:11:17 time: 0.8191 data_time: 0.6800 memory: 6319 loss: 0.0641 +2023/06/05 05:52:22 - mmengine - INFO - Exp name: resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1_20230604_215105 +2023/06/05 05:53:10 - mmengine - INFO - Epoch(train) [9][2800/3907] lr: 1.3609e-05 eta: 1:09:53 time: 0.8143 data_time: 0.6686 memory: 6319 loss: 0.0636 +2023/06/05 05:54:31 - mmengine - INFO - Epoch(train) [9][2900/3907] lr: 1.3469e-05 eta: 1:08:29 time: 0.7831 data_time: 0.6434 memory: 6319 loss: 0.0642 +2023/06/05 05:55:50 - mmengine - INFO - Epoch(train) [9][3000/3907] lr: 1.3331e-05 eta: 1:07:05 time: 0.8319 data_time: 0.6880 memory: 6319 loss: 0.0646 +2023/06/05 05:57:12 - mmengine - INFO - Epoch(train) [9][3100/3907] lr: 1.3196e-05 eta: 1:05:41 time: 0.8315 data_time: 0.6910 memory: 6319 loss: 0.0646 +2023/06/05 05:58:33 - mmengine - INFO - Epoch(train) [9][3200/3907] lr: 1.3063e-05 eta: 1:04:17 time: 0.7779 data_time: 0.6375 memory: 6319 loss: 0.0592 +2023/06/05 05:59:56 - mmengine - INFO - Epoch(train) [9][3300/3907] lr: 1.2933e-05 eta: 1:02:53 time: 0.8100 data_time: 0.6699 memory: 6319 loss: 0.0524 +2023/06/05 06:01:16 - mmengine - INFO - Epoch(train) [9][3400/3907] lr: 1.2806e-05 eta: 1:01:29 time: 0.8076 data_time: 0.6687 memory: 6319 loss: 0.0622 +2023/06/05 06:02:38 - mmengine - INFO - Epoch(train) [9][3500/3907] lr: 1.2682e-05 eta: 1:00:05 time: 0.8126 data_time: 0.6728 memory: 6319 loss: 0.0543 +2023/06/05 06:04:00 - mmengine - INFO - Epoch(train) [9][3600/3907] lr: 1.2560e-05 eta: 0:58:41 time: 0.8308 data_time: 0.6883 memory: 6319 loss: 0.0541 +2023/06/05 06:05:21 - mmengine - INFO - Epoch(train) [9][3700/3907] lr: 1.2441e-05 eta: 0:57:18 time: 0.8353 data_time: 0.6946 memory: 6319 loss: 0.0833 +2023/06/05 06:05:53 - mmengine - INFO - Exp name: resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1_20230604_215105 +2023/06/05 06:06:41 - mmengine - INFO - Epoch(train) [9][3800/3907] lr: 1.2325e-05 eta: 0:55:54 time: 0.8448 data_time: 0.7056 memory: 6319 loss: 0.0624 +2023/06/05 06:08:05 - mmengine - INFO - Epoch(train) [9][3900/3907] lr: 1.2211e-05 eta: 0:54:30 time: 0.7763 data_time: 0.6372 memory: 6319 loss: 0.0638 +2023/06/05 06:08:06 - mmengine - INFO - Exp name: resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1_20230604_215105 +2023/06/05 06:08:06 - mmengine - INFO - Saving checkpoint at 9 epochs +2023/06/05 06:08:50 - mmengine - INFO - Epoch(val) [9][57/57] accuracy/top1: 97.2513 data_time: 0.4777 time: 0.5636 +2023/06/05 06:10:14 - mmengine - INFO - Epoch(train) [10][ 100/3907] lr: 1.2093e-05 eta: 0:53:00 time: 0.8314 data_time: 0.3991 memory: 6319 loss: 0.0566 +2023/06/05 06:11:35 - mmengine - INFO - Epoch(train) [10][ 200/3907] lr: 1.1985e-05 eta: 0:51:36 time: 0.7888 data_time: 0.3483 memory: 6319 loss: 0.0608 +2023/06/05 06:13:00 - mmengine - INFO - Epoch(train) [10][ 300/3907] lr: 1.1881e-05 eta: 0:50:13 time: 0.7649 data_time: 0.1796 memory: 6319 loss: 0.0553 +2023/06/05 06:14:23 - mmengine - INFO - Epoch(train) [10][ 400/3907] lr: 1.1778e-05 eta: 0:48:49 time: 0.8525 data_time: 0.1940 memory: 6319 loss: 0.0642 +2023/06/05 06:15:45 - mmengine - INFO - Epoch(train) [10][ 500/3907] lr: 1.1679e-05 eta: 0:47:26 time: 0.8114 data_time: 0.0939 memory: 6319 loss: 0.0629 +2023/06/05 06:17:04 - mmengine - INFO - Epoch(train) [10][ 600/3907] lr: 1.1583e-05 eta: 0:46:02 time: 0.7451 data_time: 0.2651 memory: 6319 loss: 0.0684 +2023/06/05 06:18:21 - mmengine - INFO - Epoch(train) [10][ 700/3907] lr: 1.1489e-05 eta: 0:44:38 time: 0.7837 data_time: 0.4019 memory: 6319 loss: 0.0708 +2023/06/05 06:19:42 - mmengine - INFO - Epoch(train) [10][ 800/3907] lr: 1.1398e-05 eta: 0:43:14 time: 0.8084 data_time: 0.6263 memory: 6319 loss: 0.0645 +2023/06/05 06:20:15 - mmengine - INFO - Exp name: resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1_20230604_215105 +2023/06/05 06:21:03 - mmengine - INFO - Epoch(train) [10][ 900/3907] lr: 1.1310e-05 eta: 0:41:50 time: 0.8027 data_time: 0.6167 memory: 6319 loss: 0.0589 +2023/06/05 06:22:24 - mmengine - INFO - Epoch(train) [10][1000/3907] lr: 1.1225e-05 eta: 0:40:27 time: 0.8277 data_time: 0.6860 memory: 6319 loss: 0.0645 +2023/06/05 06:23:46 - mmengine - INFO - Epoch(train) [10][1100/3907] lr: 1.1142e-05 eta: 0:39:03 time: 0.8499 data_time: 0.7088 memory: 6319 loss: 0.0546 +2023/06/05 06:25:06 - mmengine - INFO - Epoch(train) [10][1200/3907] lr: 1.1063e-05 eta: 0:37:39 time: 0.8116 data_time: 0.6648 memory: 6319 loss: 0.0614 +2023/06/05 06:26:27 - mmengine - INFO - Epoch(train) [10][1300/3907] lr: 1.0986e-05 eta: 0:36:15 time: 0.8547 data_time: 0.7137 memory: 6319 loss: 0.0739 +2023/06/05 06:27:49 - mmengine - INFO - Epoch(train) [10][1400/3907] lr: 1.0912e-05 eta: 0:34:52 time: 0.8186 data_time: 0.6773 memory: 6319 loss: 0.0545 +2023/06/05 06:29:11 - mmengine - INFO - Epoch(train) [10][1500/3907] lr: 1.0841e-05 eta: 0:33:28 time: 0.8103 data_time: 0.6711 memory: 6319 loss: 0.0657 +2023/06/05 06:30:33 - mmengine - INFO - Epoch(train) [10][1600/3907] lr: 1.0773e-05 eta: 0:32:05 time: 0.8152 data_time: 0.6750 memory: 6319 loss: 0.0593 +2023/06/05 06:31:55 - mmengine - INFO - Epoch(train) [10][1700/3907] lr: 1.0707e-05 eta: 0:30:41 time: 0.8574 data_time: 0.7159 memory: 6319 loss: 0.0595 +2023/06/05 06:33:17 - mmengine - INFO - Epoch(train) [10][1800/3907] lr: 1.0645e-05 eta: 0:29:18 time: 0.8191 data_time: 0.6792 memory: 6319 loss: 0.0632 +2023/06/05 06:33:43 - mmengine - INFO - Exp name: resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1_20230604_215105 +2023/06/05 06:34:37 - mmengine - INFO - Epoch(train) [10][1900/3907] lr: 1.0585e-05 eta: 0:27:54 time: 0.7543 data_time: 0.6151 memory: 6319 loss: 0.0608 +2023/06/05 06:35:58 - mmengine - INFO - Epoch(train) [10][2000/3907] lr: 1.0529e-05 eta: 0:26:31 time: 0.7631 data_time: 0.6235 memory: 6319 loss: 0.0508 +2023/06/05 06:37:18 - mmengine - INFO - Epoch(train) [10][2100/3907] lr: 1.0475e-05 eta: 0:25:07 time: 0.8186 data_time: 0.6772 memory: 6319 loss: 0.0667 +2023/06/05 06:38:39 - mmengine - INFO - Epoch(train) [10][2200/3907] lr: 1.0424e-05 eta: 0:23:43 time: 0.8144 data_time: 0.6750 memory: 6319 loss: 0.0669 +2023/06/05 06:40:01 - mmengine - INFO - Epoch(train) [10][2300/3907] lr: 1.0376e-05 eta: 0:22:20 time: 0.8303 data_time: 0.6899 memory: 6319 loss: 0.0589 +2023/06/05 06:41:21 - mmengine - INFO - Epoch(train) [10][2400/3907] lr: 1.0330e-05 eta: 0:20:56 time: 0.8459 data_time: 0.7051 memory: 6319 loss: 0.0619 +2023/06/05 06:42:46 - mmengine - INFO - Epoch(train) [10][2500/3907] lr: 1.0288e-05 eta: 0:19:33 time: 0.8917 data_time: 0.7522 memory: 6319 loss: 0.0635 +2023/06/05 06:44:08 - mmengine - INFO - Epoch(train) [10][2600/3907] lr: 1.0249e-05 eta: 0:18:10 time: 0.7884 data_time: 0.6496 memory: 6319 loss: 0.0582 +2023/06/05 06:45:29 - mmengine - INFO - Epoch(train) [10][2700/3907] lr: 1.0212e-05 eta: 0:16:46 time: 0.7970 data_time: 0.6574 memory: 6319 loss: 0.0763 +2023/06/05 06:46:51 - mmengine - INFO - Epoch(train) [10][2800/3907] lr: 1.0178e-05 eta: 0:15:23 time: 0.8415 data_time: 0.7019 memory: 6319 loss: 0.0688 +2023/06/05 06:47:16 - mmengine - INFO - Exp name: resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1_20230604_215105 +2023/06/05 06:48:12 - mmengine - INFO - Epoch(train) [10][2900/3907] lr: 1.0148e-05 eta: 0:13:59 time: 0.8350 data_time: 0.6947 memory: 6319 loss: 0.0669 +2023/06/05 06:49:32 - mmengine - INFO - Epoch(train) [10][3000/3907] lr: 1.0120e-05 eta: 0:12:36 time: 0.7863 data_time: 0.6461 memory: 6319 loss: 0.0625 +2023/06/05 06:50:54 - mmengine - INFO - Epoch(train) [10][3100/3907] lr: 1.0095e-05 eta: 0:11:12 time: 0.8006 data_time: 0.6603 memory: 6319 loss: 0.0596 +2023/06/05 06:52:14 - mmengine - INFO - Epoch(train) [10][3200/3907] lr: 1.0073e-05 eta: 0:09:49 time: 0.7687 data_time: 0.6292 memory: 6319 loss: 0.0703 +2023/06/05 06:53:36 - mmengine - INFO - Epoch(train) [10][3300/3907] lr: 1.0054e-05 eta: 0:08:26 time: 0.8422 data_time: 0.7025 memory: 6319 loss: 0.0695 +2023/06/05 06:54:58 - mmengine - INFO - Epoch(train) [10][3400/3907] lr: 1.0038e-05 eta: 0:07:02 time: 0.8897 data_time: 0.7504 memory: 6319 loss: 0.0671 +2023/06/05 06:56:21 - mmengine - INFO - Epoch(train) [10][3500/3907] lr: 1.0024e-05 eta: 0:05:39 time: 0.8049 data_time: 0.6651 memory: 6319 loss: 0.0618 +2023/06/05 06:57:42 - mmengine - INFO - Epoch(train) [10][3600/3907] lr: 1.0014e-05 eta: 0:04:15 time: 0.8006 data_time: 0.6607 memory: 6319 loss: 0.0554 +2023/06/05 06:59:03 - mmengine - INFO - Epoch(train) [10][3700/3907] lr: 1.0006e-05 eta: 0:02:52 time: 0.7930 data_time: 0.6534 memory: 6319 loss: 0.0582 +2023/06/05 07:00:24 - mmengine - INFO - Epoch(train) [10][3800/3907] lr: 1.0002e-05 eta: 0:01:29 time: 0.8202 data_time: 0.6800 memory: 6319 loss: 0.0546 +2023/06/05 07:00:49 - mmengine - INFO - Exp name: resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1_20230604_215105 +2023/06/05 07:01:46 - mmengine - INFO - Epoch(train) [10][3900/3907] lr: 1.0000e-05 eta: 0:00:05 time: 0.8321 data_time: 0.6927 memory: 6319 loss: 0.0651 +2023/06/05 07:01:47 - mmengine - INFO - Exp name: resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1_20230604_215105 +2023/06/05 07:01:47 - mmengine - INFO - Saving checkpoint at 10 epochs +2023/06/05 07:02:32 - mmengine - INFO - Epoch(val) [10][57/57] accuracy/top1: 97.5297 data_time: 0.5048 time: 0.5928 diff --git a/resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1/20230604_215105/vis_data/20230604_215105.json b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1/20230604_215105/vis_data/20230604_215105.json new file mode 100644 index 0000000000000000000000000000000000000000..82d3284e451bc76603b545c6bab6b12f54bb0c54 --- /dev/null +++ b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1/20230604_215105/vis_data/20230604_215105.json @@ -0,0 +1,400 @@ +{"lr": 9.999857418382631e-05, "data_time": 0.6374414443969727, "loss": 0.601702731847763, "time": 0.8555856943130493, "epoch": 1, "memory": 9436, "step": 100} +{"lr": 9.999423907359506e-05, "data_time": 0.6512850522994995, "loss": 0.5355101227760315, "time": 0.8826247930526734, "epoch": 1, "memory": 6319, "step": 200} +{"lr": 9.998699480412073e-05, "data_time": 0.62061927318573, "loss": 0.46963952779769896, "time": 0.8177884101867676, "epoch": 1, "memory": 6319, "step": 300} +{"lr": 9.997684184379085e-05, "data_time": 0.7160451650619507, "loss": 0.42892860770225527, "time": 0.8552490472793579, "epoch": 1, "memory": 6319, "step": 400} +{"lr": 9.996378084905784e-05, "data_time": 0.5343232154846191, "loss": 0.39631963074207305, "time": 0.8138736248016357, "epoch": 1, "memory": 6319, "step": 500} +{"lr": 9.994781266439667e-05, "data_time": 0.5577570915222168, "loss": 0.37549435496330263, "time": 0.8080990791320801, "epoch": 1, "memory": 6319, "step": 600} +{"lr": 9.992893832225061e-05, "data_time": 0.4441789150238037, "loss": 0.36862215101718904, "time": 0.8239611864089966, "epoch": 1, "memory": 6319, "step": 700} +{"lr": 9.990715904296405e-05, "data_time": 0.611823320388794, "loss": 0.31859688758850097, "time": 0.8215285062789917, "epoch": 1, "memory": 6319, "step": 800} +{"lr": 9.988247623470371e-05, "data_time": 0.6151374340057373, "loss": 0.3201468825340271, "time": 0.8195865154266357, "epoch": 1, "memory": 6319, "step": 900} +{"lr": 9.985489149336749e-05, "data_time": 0.6461711883544922, "loss": 0.2996815800666809, "time": 0.7850286960601807, "epoch": 1, "memory": 6319, "step": 1000} +{"lr": 9.982440660248187e-05, "data_time": 0.6466344833374024, "loss": 0.29422697573900225, "time": 0.7864808320999146, "epoch": 1, "memory": 6319, "step": 1100} +{"lr": 9.979102353308574e-05, "data_time": 0.6717127084732055, "loss": 0.29378020763397217, "time": 0.8114043951034546, "epoch": 1, "memory": 6319, "step": 1200} +{"lr": 9.975474444360358e-05, "data_time": 0.6845787286758422, "loss": 0.2830167576670647, "time": 0.8234544992446899, "epoch": 1, "memory": 6319, "step": 1300} +{"lr": 9.97155716797056e-05, "data_time": 0.6955356121063232, "loss": 0.2428518310189247, "time": 0.8341902494430542, "epoch": 1, "memory": 6319, "step": 1400} +{"lr": 9.967350777415632e-05, "data_time": 0.7570158004760742, "loss": 0.23246523439884187, "time": 0.8962111234664917, "epoch": 1, "memory": 6319, "step": 1500} +{"lr": 9.962855544665059e-05, "data_time": 0.6592184782028199, "loss": 0.25699472427368164, "time": 0.7987259864807129, "epoch": 1, "memory": 6319, "step": 1600} +{"lr": 9.958071760363773e-05, "data_time": 0.700481367111206, "loss": 0.2340873584151268, "time": 0.84044349193573, "epoch": 1, "memory": 6319, "step": 1700} +{"lr": 9.952999733813382e-05, "data_time": 0.44866161346435546, "loss": 0.21638996303081512, "time": 0.8508273839950562, "epoch": 1, "memory": 6319, "step": 1800} +{"lr": 9.947639792952148e-05, "data_time": 0.47938241958618166, "loss": 0.2403170332312584, "time": 0.851843285560608, "epoch": 1, "memory": 6319, "step": 1900} +{"lr": 9.941992284333813e-05, "data_time": 0.6045541524887085, "loss": 0.21530583649873733, "time": 0.8377137899398803, "epoch": 1, "memory": 6319, "step": 2000} +{"lr": 9.93605757310516e-05, "data_time": 0.6837259769439697, "loss": 0.22493914216756822, "time": 0.8786030769348144, "epoch": 1, "memory": 6319, "step": 2100} +{"lr": 9.929836042982421e-05, "data_time": 0.7660118103027344, "loss": 0.22159391790628433, "time": 0.9059905529022216, "epoch": 1, "memory": 6319, "step": 2200} +{"lr": 9.923328096226467e-05, "data_time": 0.7311722278594971, "loss": 0.224177086353302, "time": 0.8855130910873413, "epoch": 1, "memory": 6319, "step": 2300} +{"lr": 9.916534153616819e-05, "data_time": 0.7572393178939819, "loss": 0.18926170617341995, "time": 0.8972986936569214, "epoch": 1, "memory": 6319, "step": 2400} +{"lr": 9.909454654424387e-05, "data_time": 0.6951786756515503, "loss": 0.20676430612802504, "time": 0.8493762731552124, "epoch": 1, "memory": 6319, "step": 2500} +{"lr": 9.902090056383109e-05, "data_time": 0.7557082891464233, "loss": 0.1941266715526581, "time": 0.8950440168380738, "epoch": 1, "memory": 6319, "step": 2600} +{"lr": 9.894440835660365e-05, "data_time": 0.67665593624115, "loss": 0.19437728822231293, "time": 0.8159930229187011, "epoch": 1, "memory": 6319, "step": 2700} +{"lr": 9.886507486826162e-05, "data_time": 0.7613880872726441, "loss": 0.19213476926088333, "time": 0.9016317844390869, "epoch": 1, "memory": 6319, "step": 2800} +{"lr": 9.878290522821154e-05, "data_time": 0.7427852153778076, "loss": 0.20432229340076447, "time": 0.8836068868637085, "epoch": 1, "memory": 6319, "step": 2900} +{"lr": 9.86979047492352e-05, "data_time": 0.7449662208557128, "loss": 0.17515064626932145, "time": 0.8846933603286743, "epoch": 1, "memory": 6319, "step": 3000} +{"lr": 9.861007892714578e-05, "data_time": 0.7171546936035156, "loss": 0.1845693200826645, "time": 0.8594824552536011, "epoch": 1, "memory": 6319, "step": 3100} +{"lr": 9.851943344043239e-05, "data_time": 0.7557272434234619, "loss": 0.1728598140180111, "time": 0.8962688207626343, "epoch": 1, "memory": 6319, "step": 3200} +{"lr": 9.842597414989319e-05, "data_time": 0.6702453374862671, "loss": 0.18007278442382812, "time": 0.8097019672393799, "epoch": 1, "memory": 6319, "step": 3300} +{"lr": 9.832970709825637e-05, "data_time": 0.6524338006973267, "loss": 0.16683334857225418, "time": 0.8048237323760986, "epoch": 1, "memory": 6319, "step": 3400} +{"lr": 9.823063850978968e-05, "data_time": 0.6931561470031739, "loss": 0.18467812538146972, "time": 0.8349983215332031, "epoch": 1, "memory": 6319, "step": 3500} +{"lr": 9.81287747898975e-05, "data_time": 0.6722934484481812, "loss": 0.15657266080379487, "time": 0.827212643623352, "epoch": 1, "memory": 6319, "step": 3600} +{"lr": 9.802412252470715e-05, "data_time": 0.6636826515197753, "loss": 0.1768097162246704, "time": 0.8044445753097534, "epoch": 1, "memory": 6319, "step": 3700} +{"lr": 9.791668848064258e-05, "data_time": 0.601073956489563, "loss": 0.18478747755289077, "time": 0.8198684692382813, "epoch": 1, "memory": 6319, "step": 3800} +{"lr": 9.780647960398751e-05, "data_time": 0.6998269557952881, "loss": 0.16740401089191437, "time": 0.8399232864379883, "epoch": 1, "memory": 6319, "step": 3900} +{"accuracy/top1": 81.12104797363281, "data_time": 0.5407230603067499, "time": 0.6341041807542768, "step": 1} +{"lr": 9.7685491196646e-05, "data_time": 0.4053492069244385, "loss": 0.164532807469368, "time": 0.8102155685424804, "epoch": 2, "memory": 6319, "step": 4007} +{"lr": 9.756956126085676e-05, "data_time": 0.24088940620422364, "loss": 0.16619824916124343, "time": 0.8014201641082763, "epoch": 2, "memory": 6319, "step": 4107} +{"lr": 9.745087893642482e-05, "data_time": 0.1316983222961426, "loss": 0.16588684022426606, "time": 0.7923794269561768, "epoch": 2, "memory": 6319, "step": 4207} +{"lr": 9.73294518969056e-05, "data_time": 0.3431976318359375, "loss": 0.1442151889204979, "time": 0.8041167974472045, "epoch": 2, "memory": 6319, "step": 4307} +{"lr": 9.72052879933172e-05, "data_time": 0.3228480339050293, "loss": 0.15931000784039498, "time": 0.7638721227645874, "epoch": 2, "memory": 6319, "step": 4407} +{"lr": 9.707839525363358e-05, "data_time": 0.26673552989959715, "loss": 0.14772066250443458, "time": 0.8368432760238648, "epoch": 2, "memory": 6319, "step": 4507} +{"lr": 9.69487818822646e-05, "data_time": 0.3482442140579224, "loss": 0.1541784442961216, "time": 0.8618246555328369, "epoch": 2, "memory": 6319, "step": 4607} +{"lr": 9.681645625952616e-05, "data_time": 0.5624294757843018, "loss": 0.15874218046665192, "time": 0.8391940355300903, "epoch": 2, "memory": 6319, "step": 4707} +{"lr": 9.668142694109826e-05, "data_time": 0.7022975206375122, "loss": 0.15026861429214478, "time": 0.863461709022522, "epoch": 2, "memory": 6319, "step": 4807} +{"lr": 9.654370265747162e-05, "data_time": 0.5626144409179688, "loss": 0.16405150443315505, "time": 0.8262191772460937, "epoch": 2, "memory": 6319, "step": 4907} +{"lr": 9.640329231338345e-05, "data_time": 0.7439480781555176, "loss": 0.1409287489950657, "time": 0.8972822427749634, "epoch": 2, "memory": 6319, "step": 5007} +{"lr": 9.626020498724147e-05, "data_time": 0.6420143127441407, "loss": 0.13239647075533867, "time": 0.8961099624633789, "epoch": 2, "memory": 6319, "step": 5107} +{"lr": 9.611444993053717e-05, "data_time": 0.2255570650100708, "loss": 0.14459227547049522, "time": 0.8393664598464966, "epoch": 2, "memory": 6319, "step": 5207} +{"lr": 9.596603656724746e-05, "data_time": 0.0013016700744628907, "loss": 0.14789986237883568, "time": 0.8949613809585572, "epoch": 2, "memory": 6319, "step": 5307} +{"lr": 9.58149744932254e-05, "data_time": 0.0016445159912109376, "loss": 0.16496630012989044, "time": 0.9058290719985962, "epoch": 2, "memory": 6319, "step": 5407} +{"lr": 9.566127347557972e-05, "data_time": 0.0010085344314575196, "loss": 0.12877277731895448, "time": 0.8427849292755127, "epoch": 2, "memory": 6319, "step": 5507} +{"lr": 9.550494345204367e-05, "data_time": 0.5324305772781373, "loss": 0.1346907749772072, "time": 0.8919317960739136, "epoch": 2, "memory": 6319, "step": 5607} +{"lr": 9.534599453033183e-05, "data_time": 0.33181347846984866, "loss": 0.14077757373452188, "time": 0.836850643157959, "epoch": 2, "memory": 6319, "step": 5707} +{"lr": 9.518443698748712e-05, "data_time": 0.37905948162078856, "loss": 0.13480956330895424, "time": 0.7978240966796875, "epoch": 2, "memory": 6319, "step": 5807} +{"lr": 9.50202812692162e-05, "data_time": 0.2967047214508057, "loss": 0.14967381954193115, "time": 0.8451672077178956, "epoch": 2, "memory": 6319, "step": 5907} +{"lr": 9.485353798921386e-05, "data_time": 0.530403995513916, "loss": 0.13241284266114234, "time": 0.8520182132720947, "epoch": 2, "memory": 6319, "step": 6007} +{"lr": 9.468421792847712e-05, "data_time": 0.6336118698120117, "loss": 0.12861084043979645, "time": 0.8335238218307495, "epoch": 2, "memory": 6319, "step": 6107} +{"lr": 9.451233203460794e-05, "data_time": 0.6685770273208618, "loss": 0.14063130617141723, "time": 0.8086844921112061, "epoch": 2, "memory": 6319, "step": 6207} +{"lr": 9.433789142110532e-05, "data_time": 0.6530247211456299, "loss": 0.14641004279255868, "time": 0.7929362297058106, "epoch": 2, "memory": 6319, "step": 6307} +{"lr": 9.416090736664709e-05, "data_time": 0.44009592533111574, "loss": 0.12866621166467668, "time": 0.846757459640503, "epoch": 2, "memory": 6319, "step": 6407} +{"lr": 9.398139131436032e-05, "data_time": 0.21391720771789552, "loss": 0.1294976994395256, "time": 0.8114058256149292, "epoch": 2, "memory": 6319, "step": 6507} +{"lr": 9.379935487108167e-05, "data_time": 0.07746953964233398, "loss": 0.12480120584368706, "time": 0.8726335048675538, "epoch": 2, "memory": 6319, "step": 6607} +{"lr": 9.361480980660697e-05, "data_time": 0.0010189771652221679, "loss": 0.13518597483634948, "time": 0.7963105916976929, "epoch": 2, "memory": 6319, "step": 6707} +{"lr": 9.342776805292993e-05, "data_time": 0.0010326385498046875, "loss": 0.12527859807014466, "time": 0.8246926069259644, "epoch": 2, "memory": 6319, "step": 6807} +{"lr": 9.323824170347083e-05, "data_time": 0.010301589965820312, "loss": 0.11467275694012642, "time": 0.7805991888046264, "epoch": 2, "memory": 6319, "step": 6907} +{"lr": 9.304624301229471e-05, "data_time": 0.5237648010253906, "loss": 0.1257191315293312, "time": 0.826194429397583, "epoch": 2, "memory": 6319, "step": 7007} +{"lr": 9.285178439331897e-05, "data_time": 0.6844304800033569, "loss": 0.1265637643635273, "time": 0.824958086013794, "epoch": 2, "memory": 6319, "step": 7107} +{"lr": 9.265487841951087e-05, "data_time": 0.6880308866500855, "loss": 0.1203043982386589, "time": 0.8342239856719971, "epoch": 2, "memory": 6319, "step": 7207} +{"lr": 9.245553782207421e-05, "data_time": 0.6624742269515991, "loss": 0.12059230878949165, "time": 0.8033913373947144, "epoch": 2, "memory": 6319, "step": 7307} +{"lr": 9.225377548962671e-05, "data_time": 0.6826150178909302, "loss": 0.12225314378738403, "time": 0.8241559743881226, "epoch": 2, "memory": 6319, "step": 7407} +{"lr": 9.204960446736601e-05, "data_time": 0.6545940160751342, "loss": 0.12865836247801782, "time": 0.8548345565795898, "epoch": 2, "memory": 6319, "step": 7507} +{"lr": 9.184303795622715e-05, "data_time": 0.676851463317871, "loss": 0.13751766011118888, "time": 0.8155324935913086, "epoch": 2, "memory": 6319, "step": 7607} +{"lr": 9.163408931202823e-05, "data_time": 0.6777835369110108, "loss": 0.11660075187683105, "time": 0.817630124092102, "epoch": 2, "memory": 6319, "step": 7707} +{"lr": 9.142277204460729e-05, "data_time": 0.6675420522689819, "loss": 0.11041985228657722, "time": 0.8077948331832886, "epoch": 2, "memory": 6319, "step": 7807} +{"accuracy/top1": 87.63091278076172, "data_time": 0.5409339337513365, "time": 0.6319431518686229, "step": 2} +{"lr": 9.119405492383288e-05, "data_time": 0.5990873575210571, "loss": 0.1079963929951191, "time": 0.8015936136245727, "epoch": 3, "memory": 6319, "step": 7914} +{"lr": 9.097787819233205e-05, "data_time": 0.6570791482925415, "loss": 0.11941554769873619, "time": 0.8080140113830566, "epoch": 3, "memory": 6319, "step": 8014} +{"lr": 9.075937526576627e-05, "data_time": 0.6611604452133178, "loss": 0.10831483006477356, "time": 0.8022749423980713, "epoch": 3, "memory": 6319, "step": 8114} +{"lr": 9.053856027171755e-05, "data_time": 0.6530561208724975, "loss": 0.11707084774971008, "time": 0.7928166627883911, "epoch": 3, "memory": 6319, "step": 8214} +{"lr": 9.031544748725785e-05, "data_time": 0.2923057317733765, "loss": 0.10115634873509408, "time": 0.8415463685989379, "epoch": 3, "memory": 6319, "step": 8314} +{"lr": 9.009005133802542e-05, "data_time": 0.4353105306625366, "loss": 0.10452483519911766, "time": 0.8080923080444335, "epoch": 3, "memory": 6319, "step": 8414} +{"lr": 8.986238639729242e-05, "data_time": 0.7275919675827026, "loss": 0.10942161232233047, "time": 0.918468427658081, "epoch": 3, "memory": 6319, "step": 8514} +{"lr": 8.963246738502254e-05, "data_time": 0.6077797412872314, "loss": 0.12090184316039085, "time": 0.8502497434616089, "epoch": 3, "memory": 6319, "step": 8614} +{"lr": 8.940030916691928e-05, "data_time": 0.6876582860946655, "loss": 0.11244189366698265, "time": 0.8271790981292725, "epoch": 3, "memory": 6319, "step": 8714} +{"lr": 8.91659267534649e-05, "data_time": 0.6878661394119263, "loss": 0.10171929523348808, "time": 0.8271135330200196, "epoch": 3, "memory": 6319, "step": 8814} +{"lr": 8.892933529894978e-05, "data_time": 0.6629051685333252, "loss": 0.10485116988420487, "time": 0.8020254850387574, "epoch": 3, "memory": 6319, "step": 8914} +{"lr": 8.869055010049283e-05, "data_time": 0.7062404870986938, "loss": 0.09983178451657296, "time": 0.8448907613754273, "epoch": 3, "memory": 6319, "step": 9014} +{"lr": 8.844958659705188e-05, "data_time": 0.7121747255325317, "loss": 0.09979011118412018, "time": 0.8525150775909424, "epoch": 3, "memory": 6319, "step": 9114} +{"lr": 8.82064603684259e-05, "data_time": 0.6661271572113037, "loss": 0.09753329679369926, "time": 0.8060618162155151, "epoch": 3, "memory": 6319, "step": 9214} +{"lr": 8.796118713424776e-05, "data_time": 0.6657455921173095, "loss": 0.10611129701137542, "time": 0.8054347515106202, "epoch": 3, "memory": 6319, "step": 9314} +{"lr": 8.77137827529676e-05, "data_time": 0.9204483747482299, "loss": 0.11277874186635017, "time": 1.0597905158996581, "epoch": 3, "memory": 6319, "step": 9414} +{"lr": 8.746426322082745e-05, "data_time": 0.6355006217956543, "loss": 0.11189985498785973, "time": 0.7843372821807861, "epoch": 3, "memory": 6319, "step": 9514} +{"lr": 8.721264467082732e-05, "data_time": 0.6172508478164673, "loss": 0.11194451823830605, "time": 0.7983240842819214, "epoch": 3, "memory": 6319, "step": 9614} +{"lr": 8.695894337168168e-05, "data_time": 0.7082661867141724, "loss": 0.1246405228972435, "time": 0.8474427223205566, "epoch": 3, "memory": 6319, "step": 9714} +{"lr": 8.67031757267677e-05, "data_time": 0.6869972229003907, "loss": 0.08495607413351536, "time": 0.826607084274292, "epoch": 3, "memory": 6319, "step": 9814} +{"lr": 8.64453582730651e-05, "data_time": 0.7059129476547241, "loss": 0.10880973637104034, "time": 0.8447617769241333, "epoch": 3, "memory": 6319, "step": 9914} +{"lr": 8.61855076800861e-05, "data_time": 0.7079426765441894, "loss": 0.09506015405058861, "time": 0.8473704576492309, "epoch": 3, "memory": 6319, "step": 10014} +{"lr": 8.592364074879843e-05, "data_time": 0.6977201938629151, "loss": 0.11660193353891372, "time": 0.8378056049346924, "epoch": 3, "memory": 6319, "step": 10114} +{"lr": 8.565977441053854e-05, "data_time": 0.6644403457641601, "loss": 0.09511319696903228, "time": 0.8034246444702149, "epoch": 3, "memory": 6319, "step": 10214} +{"lr": 8.539392572591724e-05, "data_time": 0.667553162574768, "loss": 0.11333511024713516, "time": 0.8074770450592041, "epoch": 3, "memory": 6319, "step": 10314} +{"lr": 8.512611188371633e-05, "data_time": 0.7124860048294067, "loss": 0.09126886427402496, "time": 0.8519822359085083, "epoch": 3, "memory": 6319, "step": 10414} +{"lr": 8.485635019977725e-05, "data_time": 0.7116752147674561, "loss": 0.10744883194565773, "time": 0.8510948181152344, "epoch": 3, "memory": 6319, "step": 10514} +{"lr": 8.45846581158818e-05, "data_time": 0.6724309682846069, "loss": 0.10148456953465938, "time": 0.8115061044692993, "epoch": 3, "memory": 6319, "step": 10614} +{"lr": 8.431105319862416e-05, "data_time": 0.6431312561035156, "loss": 0.0994156077504158, "time": 0.7821248292922973, "epoch": 3, "memory": 6319, "step": 10714} +{"lr": 8.403555313827516e-05, "data_time": 0.7016383171081543, "loss": 0.10597254410386085, "time": 0.8413288354873657, "epoch": 3, "memory": 6319, "step": 10814} +{"lr": 8.375817574763844e-05, "data_time": 0.7045982837677002, "loss": 0.09045120924711228, "time": 0.8459288835525512, "epoch": 3, "memory": 6319, "step": 10914} +{"lr": 8.347893896089871e-05, "data_time": 0.6797439575195312, "loss": 0.09476337805390359, "time": 0.8190370798110962, "epoch": 3, "memory": 6319, "step": 11014} +{"lr": 8.31978608324625e-05, "data_time": 0.7089139461517334, "loss": 0.1007885254919529, "time": 0.8491958379745483, "epoch": 3, "memory": 6319, "step": 11114} +{"lr": 8.291495953579052e-05, "data_time": 0.6972779750823974, "loss": 0.08978517912328243, "time": 0.8372002363204956, "epoch": 3, "memory": 6319, "step": 11214} +{"lr": 8.263025336222267e-05, "data_time": 0.5805172204971314, "loss": 0.09269502758979797, "time": 0.7206877946853638, "epoch": 3, "memory": 6319, "step": 11314} +{"lr": 8.234376071979551e-05, "data_time": 0.6612140417099, "loss": 0.09400350823998452, "time": 0.8000160932540894, "epoch": 3, "memory": 6319, "step": 11414} +{"lr": 8.205550013205187e-05, "data_time": 0.6754632711410522, "loss": 0.09037117287516594, "time": 0.817452073097229, "epoch": 3, "memory": 6319, "step": 11514} +{"lr": 8.17654902368435e-05, "data_time": 0.6652022123336792, "loss": 0.0913025826215744, "time": 0.8047217845916748, "epoch": 3, "memory": 6319, "step": 11614} +{"lr": 8.147374978512582e-05, "data_time": 0.6747350215911865, "loss": 0.08519593179225922, "time": 0.8134832382202148, "epoch": 3, "memory": 6319, "step": 11714} +{"accuracy/top1": 91.84092712402344, "data_time": 0.48619574924995157, "time": 0.5769296843430092, "step": 3} +{"lr": 8.115969237626578e-05, "data_time": 0.70062735080719, "loss": 0.08016534075140953, "time": 0.8406968116760254, "epoch": 4, "memory": 6319, "step": 11821} +{"lr": 8.08644297337245e-05, "data_time": 0.7934768915176391, "loss": 0.08904951065778732, "time": 0.9452500343322754, "epoch": 4, "memory": 6319, "step": 11921} +{"lr": 8.056749479387651e-05, "data_time": 0.28671603202819823, "loss": 0.0891688846051693, "time": 0.6829891443252564, "epoch": 4, "memory": 6319, "step": 12021} +{"lr": 8.026890675542457e-05, "data_time": 0.44856834411621094, "loss": 0.0916030652821064, "time": 0.8336772203445435, "epoch": 4, "memory": 6319, "step": 12121} +{"lr": 7.996868492395457e-05, "data_time": 0.5824866771697998, "loss": 0.08735960982739925, "time": 0.8282800197601319, "epoch": 4, "memory": 6319, "step": 12221} +{"lr": 7.966684871068762e-05, "data_time": 0.7231475353240967, "loss": 0.09666001684963703, "time": 0.862327241897583, "epoch": 4, "memory": 6319, "step": 12321} +{"lr": 7.936341763122449e-05, "data_time": 0.7115377902984619, "loss": 0.08375656008720397, "time": 0.8511252641677857, "epoch": 4, "memory": 6319, "step": 12421} +{"lr": 7.905841130428405e-05, "data_time": 0.6500834465026856, "loss": 0.0824422724545002, "time": 0.836016058921814, "epoch": 4, "memory": 6319, "step": 12521} +{"lr": 7.875184945043487e-05, "data_time": 0.601146149635315, "loss": 0.09197330102324486, "time": 0.8108680963516235, "epoch": 4, "memory": 6319, "step": 12621} +{"lr": 7.844375189082012e-05, "data_time": 1.0703449726104737, "loss": 0.08748614117503166, "time": 1.2102807760238647, "epoch": 4, "memory": 6319, "step": 12721} +{"lr": 7.813413854587572e-05, "data_time": 0.5996574878692627, "loss": 0.09513286128640175, "time": 0.7626878976821899, "epoch": 4, "memory": 6319, "step": 12821} +{"lr": 7.782302943404291e-05, "data_time": 0.7128910541534423, "loss": 0.07880107462406158, "time": 0.8529023885726928, "epoch": 4, "memory": 6319, "step": 12921} +{"lr": 7.751044467047337e-05, "data_time": 0.6366375684738159, "loss": 0.09039798676967621, "time": 0.7779595136642456, "epoch": 4, "memory": 6319, "step": 13021} +{"lr": 7.719640446572907e-05, "data_time": 0.7182265520095825, "loss": 0.08743033334612846, "time": 0.8584444522857666, "epoch": 4, "memory": 6319, "step": 13121} +{"lr": 7.688092912447536e-05, "data_time": 0.6953620672225952, "loss": 0.09112714752554893, "time": 0.8351022481918335, "epoch": 4, "memory": 6319, "step": 13221} +{"lr": 7.656403904416794e-05, "data_time": 0.6501647233963013, "loss": 0.08233190029859543, "time": 0.7912294149398804, "epoch": 4, "memory": 6319, "step": 13321} +{"lr": 7.624575471373434e-05, "data_time": 0.6700344800949096, "loss": 0.08806102760136128, "time": 0.80925452709198, "epoch": 4, "memory": 6319, "step": 13421} +{"lr": 7.592609671224926e-05, "data_time": 0.7183692455291748, "loss": 0.1021886222064495, "time": 0.8603218078613282, "epoch": 4, "memory": 6319, "step": 13521} +{"lr": 7.560508570760375e-05, "data_time": 0.7553340673446656, "loss": 0.07378659248352051, "time": 0.8955759048461914, "epoch": 4, "memory": 6319, "step": 13621} +{"lr": 7.528274245516895e-05, "data_time": 0.7363073825836182, "loss": 0.0893584880977869, "time": 0.8787969350814819, "epoch": 4, "memory": 6319, "step": 13721} +{"lr": 7.495908779645405e-05, "data_time": 0.7267844438552856, "loss": 0.09580430649220943, "time": 0.8677319049835205, "epoch": 4, "memory": 6319, "step": 13821} +{"lr": 7.463414265775906e-05, "data_time": 0.706990122795105, "loss": 0.07591142952442169, "time": 0.8483651638031006, "epoch": 4, "memory": 6319, "step": 13921} +{"lr": 7.430792804882147e-05, "data_time": 0.7727843761444092, "loss": 0.08486550599336624, "time": 0.9133606195449829, "epoch": 4, "memory": 6319, "step": 14021} +{"lr": 7.398046506145803e-05, "data_time": 0.6785132646560669, "loss": 0.08922588638961315, "time": 0.8246493339538574, "epoch": 4, "memory": 6319, "step": 14121} +{"lr": 7.365177486820114e-05, "data_time": 0.7385735511779785, "loss": 0.08043766170740127, "time": 0.8777874946594239, "epoch": 4, "memory": 6319, "step": 14221} +{"lr": 7.332187872092939e-05, "data_time": 0.7279436588287354, "loss": 0.07390545904636384, "time": 0.8677479743957519, "epoch": 4, "memory": 6319, "step": 14321} +{"lr": 7.299079794949401e-05, "data_time": 0.7300907373428345, "loss": 0.07530647069215775, "time": 0.8693580865859986, "epoch": 4, "memory": 6319, "step": 14421} +{"lr": 7.265855396033955e-05, "data_time": 0.6853252649307251, "loss": 0.08673914894461632, "time": 0.8242392778396607, "epoch": 4, "memory": 6319, "step": 14521} +{"lr": 7.232516823512013e-05, "data_time": 0.7513039588928223, "loss": 0.09135455749928952, "time": 0.8967481136322022, "epoch": 4, "memory": 6319, "step": 14621} +{"lr": 7.199066232930995e-05, "data_time": 0.7023496389389038, "loss": 0.10166266560554504, "time": 0.8412851095199585, "epoch": 4, "memory": 6319, "step": 14721} +{"lr": 7.165505787080997e-05, "data_time": 0.756019401550293, "loss": 0.07684456519782543, "time": 0.8975329875946045, "epoch": 4, "memory": 6319, "step": 14821} +{"lr": 7.13183765585495e-05, "data_time": 0.7164599180221558, "loss": 0.08118878044188023, "time": 0.8561276435852051, "epoch": 4, "memory": 6319, "step": 14921} +{"lr": 7.098064016108317e-05, "data_time": 0.7158091068267822, "loss": 0.07695223167538642, "time": 0.8588071823120117, "epoch": 4, "memory": 6319, "step": 15021} +{"lr": 7.064187051518379e-05, "data_time": 0.7051473140716553, "loss": 0.0883091926574707, "time": 0.845363450050354, "epoch": 4, "memory": 6319, "step": 15121} +{"lr": 7.030208952442949e-05, "data_time": 0.6706280946731568, "loss": 0.0866649679839611, "time": 0.8114208936691284, "epoch": 4, "memory": 6319, "step": 15221} +{"lr": 6.996131915778851e-05, "data_time": 0.7403595924377442, "loss": 0.08271320089697838, "time": 0.881003999710083, "epoch": 4, "memory": 6319, "step": 15321} +{"lr": 6.961958144819853e-05, "data_time": 0.6520731925964356, "loss": 0.09721692129969597, "time": 0.792563533782959, "epoch": 4, "memory": 6319, "step": 15421} +{"lr": 6.927689849114193e-05, "data_time": 0.6674933671951294, "loss": 0.08665039278566837, "time": 0.8082375526428223, "epoch": 4, "memory": 6319, "step": 15521} +{"lr": 6.893329244321722e-05, "data_time": 0.6737451553344727, "loss": 0.08945233225822449, "time": 0.8147722959518433, "epoch": 4, "memory": 6319, "step": 15621} +{"accuracy/top1": 94.34954071044922, "data_time": 0.5083364042742499, "time": 0.5965004172818414, "step": 4} +{"lr": 6.856463687361792e-05, "data_time": 0.41266841888427735, "loss": 0.09699749648571014, "time": 0.7549313068389892, "epoch": 5, "memory": 6319, "step": 15728} +{"lr": 6.821919068470432e-05, "data_time": 0.38207876682281494, "loss": 0.07658592164516449, "time": 0.827199149131775, "epoch": 5, "memory": 6319, "step": 15828} +{"lr": 6.787288979235407e-05, "data_time": 0.5177194356918335, "loss": 0.08155790939927102, "time": 0.8222105503082275, "epoch": 5, "memory": 6319, "step": 15928} +{"lr": 6.752575658708793e-05, "data_time": 0.7213511943817139, "loss": 0.07354740388691425, "time": 0.867519760131836, "epoch": 5, "memory": 6319, "step": 16028} +{"lr": 6.71778135132408e-05, "data_time": 0.5564947366714478, "loss": 0.07551858536899089, "time": 0.8823280811309815, "epoch": 5, "memory": 6319, "step": 16128} +{"lr": 6.682908306751074e-05, "data_time": 0.3923478126525879, "loss": 0.08346405848860741, "time": 0.8252520084381103, "epoch": 5, "memory": 6319, "step": 16228} +{"lr": 6.64795877975043e-05, "data_time": 0.19153399467468263, "loss": 0.08557555750012398, "time": 0.8386387825012207, "epoch": 5, "memory": 6319, "step": 16328} +{"lr": 6.61293503002787e-05, "data_time": 0.2131415367126465, "loss": 0.08024964146316052, "time": 0.8301738500595093, "epoch": 5, "memory": 6319, "step": 16428} +{"lr": 6.577839322088084e-05, "data_time": 0.02249758243560791, "loss": 0.07721111588180066, "time": 0.8165907144546509, "epoch": 5, "memory": 6319, "step": 16528} +{"lr": 6.542673925088298e-05, "data_time": 0.0009778738021850586, "loss": 0.09270984455943107, "time": 1.13065664768219, "epoch": 5, "memory": 6319, "step": 16628} +{"lr": 6.507441112691584e-05, "data_time": 0.0008757114410400391, "loss": 0.08181111589074135, "time": 0.7900747060775757, "epoch": 5, "memory": 6319, "step": 16728} +{"lr": 6.47214316291983e-05, "data_time": 0.0009041309356689453, "loss": 0.07667498663067818, "time": 0.8298603534698487, "epoch": 5, "memory": 6319, "step": 16828} +{"lr": 6.436782358006481e-05, "data_time": 0.05229659080505371, "loss": 0.07713332884013653, "time": 0.8582651853561402, "epoch": 5, "memory": 6319, "step": 16928} +{"lr": 6.40136098424896e-05, "data_time": 0.12185480594635009, "loss": 0.0840992011129856, "time": 0.8422068119049072, "epoch": 5, "memory": 6319, "step": 17028} +{"lr": 6.365881331860828e-05, "data_time": 0.20846986770629883, "loss": 0.08576505780220031, "time": 0.898157286643982, "epoch": 5, "memory": 6319, "step": 17128} +{"lr": 6.330345694823739e-05, "data_time": 0.18410565853118896, "loss": 0.07903642058372498, "time": 0.8957612991333008, "epoch": 5, "memory": 6319, "step": 17228} +{"lr": 6.29475637073908e-05, "data_time": 0.44593658447265627, "loss": 0.07665348090231419, "time": 0.8508060693740844, "epoch": 5, "memory": 6319, "step": 17328} +{"lr": 6.259115660679482e-05, "data_time": 0.2975869655609131, "loss": 0.08075075931847095, "time": 0.8303794145584107, "epoch": 5, "memory": 6319, "step": 17428} +{"lr": 6.223425869039963e-05, "data_time": 0.43838350772857665, "loss": 0.07233618497848511, "time": 1.0674143075942992, "epoch": 5, "memory": 6319, "step": 17528} +{"lr": 6.187689303389006e-05, "data_time": 0.27185142040252686, "loss": 0.0743786882609129, "time": 0.7913447141647338, "epoch": 5, "memory": 6319, "step": 17628} +{"lr": 6.151908274319292e-05, "data_time": 0.3729255676269531, "loss": 0.07758879140019417, "time": 0.8297183990478516, "epoch": 5, "memory": 6319, "step": 17728} +{"lr": 6.116085095298373e-05, "data_time": 0.4120022773742676, "loss": 0.056010492518544196, "time": 0.8959392309188843, "epoch": 5, "memory": 6319, "step": 17828} +{"lr": 6.080222082519007e-05, "data_time": 0.7243278741836547, "loss": 0.07328197248280048, "time": 0.8639042139053345, "epoch": 5, "memory": 6319, "step": 17928} +{"lr": 6.0443215547495147e-05, "data_time": 0.7258928298950196, "loss": 0.052864911034703255, "time": 0.8660146474838257, "epoch": 5, "memory": 6319, "step": 18028} +{"lr": 6.0083858331837714e-05, "data_time": 0.7116958379745484, "loss": 0.06759368702769279, "time": 0.8520001888275146, "epoch": 5, "memory": 6319, "step": 18128} +{"lr": 5.97241724129116e-05, "data_time": 0.7235569953918457, "loss": 0.07290937937796116, "time": 0.8631746053695679, "epoch": 5, "memory": 6319, "step": 18228} +{"lr": 5.936418104666328e-05, "data_time": 0.7136447191238403, "loss": 0.07604317106306553, "time": 0.8523468017578125, "epoch": 5, "memory": 6319, "step": 18328} +{"lr": 5.9003907508788295e-05, "data_time": 0.6828764200210571, "loss": 0.06759182997047901, "time": 0.8225975036621094, "epoch": 5, "memory": 6319, "step": 18428} +{"lr": 5.864337509322633e-05, "data_time": 0.7022364139556885, "loss": 0.07443410381674767, "time": 0.8418588638305664, "epoch": 5, "memory": 6319, "step": 18528} +{"lr": 5.8282607110655194e-05, "data_time": 0.7163833618164063, "loss": 0.0823317289352417, "time": 0.8567037582397461, "epoch": 5, "memory": 6319, "step": 18628} +{"lr": 5.7921626886983564e-05, "data_time": 0.978909707069397, "loss": 0.06326876506209374, "time": 1.117817187309265, "epoch": 5, "memory": 6319, "step": 18728} +{"lr": 5.756045776184287e-05, "data_time": 0.5999541521072388, "loss": 0.08691534772515297, "time": 0.739604115486145, "epoch": 5, "memory": 6319, "step": 18828} +{"lr": 5.719912308707808e-05, "data_time": 0.6714261531829834, "loss": 0.08451641723513603, "time": 0.8109571218490601, "epoch": 5, "memory": 6319, "step": 18928} +{"lr": 5.683764622523808e-05, "data_time": 0.6545212030410766, "loss": 0.07043433301150799, "time": 0.7934217929840088, "epoch": 5, "memory": 6319, "step": 19028} +{"lr": 5.647605054806502e-05, "data_time": 0.6724504709243775, "loss": 0.07509004473686218, "time": 0.812038779258728, "epoch": 5, "memory": 6319, "step": 19128} +{"lr": 5.6114359434983206e-05, "data_time": 0.6955249071121216, "loss": 0.07620588913559914, "time": 0.8364404439926147, "epoch": 5, "memory": 6319, "step": 19228} +{"lr": 5.575259627158745e-05, "data_time": 0.7081258535385132, "loss": 0.0768796220421791, "time": 0.8482952117919922, "epoch": 5, "memory": 6319, "step": 19328} +{"lr": 5.539078444813113e-05, "data_time": 0.6759809494018555, "loss": 0.06893887259066105, "time": 0.8153200626373291, "epoch": 5, "memory": 6319, "step": 19428} +{"lr": 5.5028947358013735e-05, "data_time": 0.6872256517410278, "loss": 0.07342792190611362, "time": 0.826401662826538, "epoch": 5, "memory": 6319, "step": 19528} +{"accuracy/top1": 95.75519561767578, "data_time": 0.49287892210072487, "time": 0.579543948173523, "step": 5} +{"lr": 5.464178020340108e-05, "data_time": 0.6675108909606934, "loss": 0.07222603149712085, "time": 0.8089387655258179, "epoch": 6, "memory": 6319, "step": 19635} +{"lr": 5.427996514795046e-05, "data_time": 0.6952463626861572, "loss": 0.08180357217788696, "time": 0.8361425638198853, "epoch": 6, "memory": 6319, "step": 19735} +{"lr": 5.3918196647260834e-05, "data_time": 0.7093801736831665, "loss": 0.07309752069413662, "time": 0.8487287282943725, "epoch": 6, "memory": 6319, "step": 19835} +{"lr": 5.355649809193057e-05, "data_time": 0.6975620985031128, "loss": 0.06639642640948296, "time": 0.8372887372970581, "epoch": 6, "memory": 6319, "step": 19935} +{"lr": 5.3194892868035765e-05, "data_time": 1.0439325332641602, "loss": 0.0747215911746025, "time": 1.1844680786132813, "epoch": 6, "memory": 6319, "step": 20035} +{"lr": 5.283340435561795e-05, "data_time": 0.1093963623046875, "loss": 0.06897138692438602, "time": 0.7011008501052857, "epoch": 6, "memory": 6319, "step": 20135} +{"lr": 5.2472055927172495e-05, "data_time": 0.04112601280212402, "loss": 0.09206668809056281, "time": 0.745328688621521, "epoch": 6, "memory": 6319, "step": 20235} +{"lr": 5.2110870946137605e-05, "data_time": 0.15963370800018312, "loss": 0.07574363686144352, "time": 0.8558576822280883, "epoch": 6, "memory": 6319, "step": 20335} +{"lr": 5.174987276538342e-05, "data_time": 0.1250157594680786, "loss": 0.07149523869156837, "time": 0.7836745977401733, "epoch": 6, "memory": 6319, "step": 20435} +{"lr": 5.1389084725702425e-05, "data_time": 0.09073824882507324, "loss": 0.057606670632958414, "time": 0.843821382522583, "epoch": 6, "memory": 6319, "step": 20535} +{"lr": 5.1028530154300045e-05, "data_time": 0.23503546714782714, "loss": 0.07742604538798332, "time": 0.8392188310623169, "epoch": 6, "memory": 6319, "step": 20635} +{"lr": 5.066823236328663e-05, "data_time": 0.23944566249847413, "loss": 0.0693360522389412, "time": 0.8808053970336914, "epoch": 6, "memory": 6319, "step": 20735} +{"lr": 5.030821464817006e-05, "data_time": 0.1433509111404419, "loss": 0.07649844773113727, "time": 0.8061399936676026, "epoch": 6, "memory": 6319, "step": 20835} +{"lr": 4.99485002863494e-05, "data_time": 0.5709505558013916, "loss": 0.08023917004466057, "time": 0.885708212852478, "epoch": 6, "memory": 6319, "step": 20935} +{"lr": 4.95891125356102e-05, "data_time": 0.661069107055664, "loss": 0.06758230775594712, "time": 0.8002021789550782, "epoch": 6, "memory": 6319, "step": 21035} +{"lr": 4.923007463262053e-05, "data_time": 0.6734744310379028, "loss": 0.0738202977925539, "time": 0.8127961158752441, "epoch": 6, "memory": 6319, "step": 21135} +{"lr": 4.887140979142852e-05, "data_time": 0.6755631923675537, "loss": 0.06861904226243495, "time": 0.8156698703765869, "epoch": 6, "memory": 6319, "step": 21235} +{"lr": 4.851314120196161e-05, "data_time": 0.676516842842102, "loss": 0.06585529148578644, "time": 0.8166579246520996, "epoch": 6, "memory": 6319, "step": 21335} +{"lr": 4.815529202852693e-05, "data_time": 0.6736643314361572, "loss": 0.07461902275681495, "time": 0.8135789155960083, "epoch": 6, "memory": 6319, "step": 21435} +{"lr": 4.779788540831406e-05, "data_time": 0.6853596448898316, "loss": 0.06019394584000111, "time": 0.8304765939712524, "epoch": 6, "memory": 6319, "step": 21535} +{"lr": 4.744094444989843e-05, "data_time": 0.7251887798309327, "loss": 0.0702584121376276, "time": 0.864327335357666, "epoch": 6, "memory": 6319, "step": 21635} +{"lr": 4.708449223174764e-05, "data_time": 0.6773305177688599, "loss": 0.06284310463815927, "time": 0.8168601274490357, "epoch": 6, "memory": 6319, "step": 21735} +{"lr": 4.672855180072916e-05, "data_time": 0.6327378749847412, "loss": 0.07013889662921428, "time": 0.7714540719985962, "epoch": 6, "memory": 6319, "step": 21835} +{"lr": 4.6373146170620256e-05, "data_time": 0.6743086338043213, "loss": 0.06573177762329578, "time": 0.8264220476150512, "epoch": 6, "memory": 6319, "step": 21935} +{"lr": 4.6018298320619915e-05, "data_time": 0.6912712335586548, "loss": 0.0773338109254837, "time": 0.8312756061553955, "epoch": 6, "memory": 6319, "step": 22035} +{"lr": 4.566403119386318e-05, "data_time": 0.7341667413711548, "loss": 0.07290297895669937, "time": 0.8750342130661011, "epoch": 6, "memory": 6319, "step": 22135} +{"lr": 4.531036769593769e-05, "data_time": 0.7128526210784912, "loss": 0.0873205129057169, "time": 0.8532657623291016, "epoch": 6, "memory": 6319, "step": 22235} +{"lr": 4.4957330693402685e-05, "data_time": 0.7100772619247436, "loss": 0.07504116408526898, "time": 0.8568342447280883, "epoch": 6, "memory": 6319, "step": 22335} +{"lr": 4.460494301231059e-05, "data_time": 0.677058744430542, "loss": 0.07817576602101325, "time": 0.8295367240905762, "epoch": 6, "memory": 6319, "step": 22435} +{"lr": 4.425322743673109e-05, "data_time": 0.7131170272827149, "loss": 0.07172847464680672, "time": 0.8532304048538208, "epoch": 6, "memory": 6319, "step": 22535} +{"lr": 4.390220670727812e-05, "data_time": 0.7379009485244751, "loss": 0.07035707458853721, "time": 0.8811063289642334, "epoch": 6, "memory": 6319, "step": 22635} +{"lr": 4.3551903519639325e-05, "data_time": 0.6639108419418335, "loss": 0.06795312836766243, "time": 0.802966833114624, "epoch": 6, "memory": 6319, "step": 22735} +{"lr": 4.3202340523108894e-05, "data_time": 0.6869333028793335, "loss": 0.07074483633041381, "time": 0.8265269279479981, "epoch": 6, "memory": 6319, "step": 22835} +{"lr": 4.285354031912302e-05, "data_time": 0.6874081373214722, "loss": 0.07929525859653949, "time": 0.8276089668273926, "epoch": 6, "memory": 6319, "step": 22935} +{"lr": 4.2505525459798544e-05, "data_time": 0.6330511808395386, "loss": 0.07162248343229294, "time": 0.7719424724578857, "epoch": 6, "memory": 6319, "step": 23035} +{"lr": 4.2158318446474836e-05, "data_time": 0.6937422275543212, "loss": 0.0946463007479906, "time": 0.833633017539978, "epoch": 6, "memory": 6319, "step": 23135} +{"lr": 4.181194172825905e-05, "data_time": 0.6966533184051513, "loss": 0.07296304292976856, "time": 0.8356399297714233, "epoch": 6, "memory": 6319, "step": 23235} +{"lr": 4.146641770057448e-05, "data_time": 0.6928616046905518, "loss": 0.06685053631663322, "time": 0.8336604118347168, "epoch": 6, "memory": 6319, "step": 23335} +{"lr": 4.112176870371267e-05, "data_time": 0.6637640476226807, "loss": 0.0628920566290617, "time": 0.8036967992782593, "epoch": 6, "memory": 6319, "step": 23435} +{"accuracy/top1": 96.25273895263672, "data_time": 0.47652052599808264, "time": 0.5657904353635065, "step": 6} +{"lr": 4.0753988582732786e-05, "data_time": 0.6781873226165771, "loss": 0.06858180090785027, "time": 0.8169117212295532, "epoch": 7, "memory": 6319, "step": 23542} +{"lr": 4.0411221639286815e-05, "data_time": 0.694000506401062, "loss": 0.05826659053564072, "time": 0.8341515302658081, "epoch": 7, "memory": 6319, "step": 23642} +{"lr": 4.006939795169518e-05, "data_time": 0.6736550807952881, "loss": 0.05747688505798578, "time": 0.8711623430252076, "epoch": 7, "memory": 6319, "step": 23742} +{"lr": 3.972853962099913e-05, "data_time": 0.6281713247299194, "loss": 0.0650223795324564, "time": 0.8164336442947387, "epoch": 7, "memory": 6319, "step": 23842} +{"lr": 3.93886686858237e-05, "data_time": 0.688922119140625, "loss": 0.0617504931986332, "time": 0.8274576425552368, "epoch": 7, "memory": 6319, "step": 23942} +{"lr": 3.904980712095259e-05, "data_time": 0.5287313461303711, "loss": 0.0588339526206255, "time": 0.8018426895141602, "epoch": 7, "memory": 6319, "step": 24042} +{"lr": 3.871197683590748e-05, "data_time": 0.49728686809539796, "loss": 0.0567039743065834, "time": 0.8472254037857055, "epoch": 7, "memory": 6319, "step": 24142} +{"lr": 3.837519967353125e-05, "data_time": 0.4354600191116333, "loss": 0.06448316983878613, "time": 0.781667685508728, "epoch": 7, "memory": 6319, "step": 24242} +{"lr": 3.8039497408575855e-05, "data_time": 0.4605488538742065, "loss": 0.06413083449006081, "time": 0.8372302055358887, "epoch": 7, "memory": 6319, "step": 24342} +{"lr": 3.770489174629432e-05, "data_time": 0.60411536693573, "loss": 0.06517446022480726, "time": 0.8847229957580567, "epoch": 7, "memory": 6319, "step": 24442} +{"lr": 3.737140432103754e-05, "data_time": 0.7392241954803467, "loss": 0.06586945131421089, "time": 0.8795247316360474, "epoch": 7, "memory": 6319, "step": 24542} +{"lr": 3.703905669485545e-05, "data_time": 0.7529799222946167, "loss": 0.06417501755058766, "time": 0.891901183128357, "epoch": 7, "memory": 6319, "step": 24642} +{"lr": 3.6707870356102706e-05, "data_time": 0.6683059692382812, "loss": 0.0770888464525342, "time": 0.8078370571136475, "epoch": 7, "memory": 6319, "step": 24742} +{"lr": 3.6377866718049484e-05, "data_time": 0.7030785560607911, "loss": 0.07405411973595619, "time": 0.8424556970596313, "epoch": 7, "memory": 6319, "step": 24842} +{"lr": 3.6049067117496965e-05, "data_time": 0.6847036361694336, "loss": 0.06737223193049431, "time": 0.8235409975051879, "epoch": 7, "memory": 6319, "step": 24942} +{"lr": 3.5721492813397755e-05, "data_time": 0.6364476919174195, "loss": 0.05018443632870913, "time": 0.775453519821167, "epoch": 7, "memory": 6319, "step": 25042} +{"lr": 3.539516498548151e-05, "data_time": 0.6694787502288818, "loss": 0.06519287563860417, "time": 0.8079834938049316, "epoch": 7, "memory": 6319, "step": 25142} +{"lr": 3.507010473288518e-05, "data_time": 0.7814359188079834, "loss": 0.06257222071290017, "time": 0.9209163904190063, "epoch": 7, "memory": 6319, "step": 25242} +{"lr": 3.474633307278909e-05, "data_time": 0.7358900308609009, "loss": 0.07201835811138153, "time": 0.884191370010376, "epoch": 7, "memory": 6319, "step": 25342} +{"lr": 3.4423870939058177e-05, "data_time": 0.6787420034408569, "loss": 0.07409783266484737, "time": 0.8189467430114746, "epoch": 7, "memory": 6319, "step": 25442} +{"lr": 3.410273918088805e-05, "data_time": 0.6938150644302368, "loss": 0.062348130345344546, "time": 0.8333451747894287, "epoch": 7, "memory": 6319, "step": 25542} +{"lr": 3.37829585614574e-05, "data_time": 0.7013892650604248, "loss": 0.06754357106983662, "time": 0.8399263620376587, "epoch": 7, "memory": 6319, "step": 25642} +{"lr": 3.3464549756585216e-05, "data_time": 0.6721920967102051, "loss": 0.07165113016963005, "time": 0.8337712049484253, "epoch": 7, "memory": 6319, "step": 25742} +{"lr": 3.314753335339416e-05, "data_time": 0.6872803688049316, "loss": 0.07331808060407638, "time": 0.8269120216369629, "epoch": 7, "memory": 6319, "step": 25842} +{"lr": 3.2831929848979394e-05, "data_time": 0.6770183324813843, "loss": 0.06544382236897946, "time": 0.8303147554397583, "epoch": 7, "memory": 6319, "step": 25942} +{"lr": 3.251775964908332e-05, "data_time": 0.6336238384246826, "loss": 0.06489804610610009, "time": 0.7735423803329468, "epoch": 7, "memory": 6319, "step": 26042} +{"lr": 3.22050430667762e-05, "data_time": 0.6411094188690185, "loss": 0.08467503376305104, "time": 0.7807039260864258, "epoch": 7, "memory": 6319, "step": 26142} +{"lr": 3.189380032114285e-05, "data_time": 0.660017728805542, "loss": 0.05466986652463675, "time": 0.7994655132293701, "epoch": 7, "memory": 6319, "step": 26242} +{"lr": 3.1584051535975324e-05, "data_time": 0.7323796272277832, "loss": 0.06272181272506713, "time": 0.8717444896697998, "epoch": 7, "memory": 6319, "step": 26342} +{"lr": 3.127581673847175e-05, "data_time": 0.6505621433258056, "loss": 0.06274190135300159, "time": 0.7922843217849731, "epoch": 7, "memory": 6319, "step": 26442} +{"lr": 3.0969115857941555e-05, "data_time": 0.7728345632553101, "loss": 0.07314100936055183, "time": 0.9122422695159912, "epoch": 7, "memory": 6319, "step": 26542} +{"lr": 3.0663968724516654e-05, "data_time": 0.6929701805114746, "loss": 0.06871767975389957, "time": 0.8341165542602539, "epoch": 7, "memory": 6319, "step": 26642} +{"lr": 3.0360395067869668e-05, "data_time": 0.6636749505996704, "loss": 0.06069712303578854, "time": 0.8030887365341186, "epoch": 7, "memory": 6319, "step": 26742} +{"lr": 3.0058414515938063e-05, "data_time": 0.6584934234619141, "loss": 0.0703870963305235, "time": 0.7987941265106201, "epoch": 7, "memory": 6319, "step": 26842} +{"lr": 2.975804659365501e-05, "data_time": 0.6954042434692382, "loss": 0.06219286173582077, "time": 0.8350435972213746, "epoch": 7, "memory": 6319, "step": 26942} +{"lr": 2.945931072168714e-05, "data_time": 0.7897649526596069, "loss": 0.06268918849527835, "time": 0.9300974845886231, "epoch": 7, "memory": 6319, "step": 27042} +{"lr": 2.9162226215178855e-05, "data_time": 0.6892393350601196, "loss": 0.061595524102449416, "time": 0.8299695491790772, "epoch": 7, "memory": 6319, "step": 27142} +{"lr": 2.8866812282503322e-05, "data_time": 0.6449941635131836, "loss": 0.05605282727628946, "time": 0.784614872932434, "epoch": 7, "memory": 6319, "step": 27242} +{"lr": 2.8573088024020838e-05, "data_time": 0.7308354377746582, "loss": 0.06938082985579967, "time": 0.8708070516586304, "epoch": 7, "memory": 6319, "step": 27342} +{"accuracy/top1": 96.99036407470703, "data_time": 0.48901195772762956, "time": 0.5752114304180803, "step": 7} +{"lr": 2.8260695817390092e-05, "data_time": 0.5400987148284913, "loss": 0.06039245910942555, "time": 0.7998890399932861, "epoch": 8, "memory": 6319, "step": 27449} +{"lr": 2.7970529402570967e-05, "data_time": 0.4197249889373779, "loss": 0.059700028225779536, "time": 0.8246399164199829, "epoch": 8, "memory": 6319, "step": 27549} +{"lr": 2.7682110612244214e-05, "data_time": 0.3895702362060547, "loss": 0.060332028195261955, "time": 0.838551926612854, "epoch": 8, "memory": 6319, "step": 27649} +{"lr": 2.7395458094490283e-05, "data_time": 0.258343768119812, "loss": 0.06834733560681343, "time": 0.8738335609436035, "epoch": 8, "memory": 6319, "step": 27749} +{"lr": 2.7110590383189015e-05, "data_time": 0.06507272720336914, "loss": 0.06124198362231255, "time": 0.8238757371902465, "epoch": 8, "memory": 6319, "step": 27849} +{"lr": 2.6827525896821397e-05, "data_time": 0.0008154630661010743, "loss": 0.06720116976648569, "time": 0.8146600723266602, "epoch": 8, "memory": 6319, "step": 27949} +{"lr": 2.6546282937278552e-05, "data_time": 0.000783848762512207, "loss": 0.06980157122015954, "time": 0.8189395427703857, "epoch": 8, "memory": 6319, "step": 28049} +{"lr": 2.6266879688678606e-05, "data_time": 0.0009537220001220703, "loss": 0.06123245526105166, "time": 0.890581202507019, "epoch": 8, "memory": 6319, "step": 28149} +{"lr": 2.5989334216190738e-05, "data_time": 0.0009265899658203125, "loss": 0.06326903328299523, "time": 0.8043756723403931, "epoch": 8, "memory": 6319, "step": 28249} +{"lr": 2.571366446486732e-05, "data_time": 0.0008894205093383789, "loss": 0.07336103729903698, "time": 0.7781476259231568, "epoch": 8, "memory": 6319, "step": 28349} +{"lr": 2.5439888258483665e-05, "data_time": 0.0011068820953369141, "loss": 0.06161957867443561, "time": 0.8684613943099976, "epoch": 8, "memory": 6319, "step": 28449} +{"lr": 2.5168023298385384e-05, "data_time": 0.000771021842956543, "loss": 0.059670351445674896, "time": 0.8432024955749512, "epoch": 8, "memory": 6319, "step": 28549} +{"lr": 2.489808716234412e-05, "data_time": 0.0011826753616333008, "loss": 0.06030804291367531, "time": 0.837065315246582, "epoch": 8, "memory": 6319, "step": 28649} +{"lr": 2.463009730342111e-05, "data_time": 0.001067519187927246, "loss": 0.07112121991813183, "time": 0.8273203372955322, "epoch": 8, "memory": 6319, "step": 28749} +{"lr": 2.436407104883849e-05, "data_time": 0.0007772684097290039, "loss": 0.06335515901446342, "time": 0.8596891164779663, "epoch": 8, "memory": 6319, "step": 28849} +{"lr": 2.4100025598859092e-05, "data_time": 0.0008195638656616211, "loss": 0.0655285321176052, "time": 0.8000187158584595, "epoch": 8, "memory": 6319, "step": 28949} +{"lr": 2.3837978025674284e-05, "data_time": 0.0007653236389160156, "loss": 0.06330592222511769, "time": 0.7643567562103272, "epoch": 8, "memory": 6319, "step": 29049} +{"lr": 2.357794527230027e-05, "data_time": 0.14439218044281005, "loss": 0.06396490409970283, "time": 0.8001121520996094, "epoch": 8, "memory": 6319, "step": 29149} +{"lr": 2.33199441514825e-05, "data_time": 0.4546696662902832, "loss": 0.06362710930407048, "time": 0.805083966255188, "epoch": 8, "memory": 6319, "step": 29249} +{"lr": 2.306399134460859e-05, "data_time": 0.42298116683959963, "loss": 0.05732900165021419, "time": 0.8389808416366578, "epoch": 8, "memory": 6319, "step": 29349} +{"lr": 2.281010340062995e-05, "data_time": 0.3388469934463501, "loss": 0.06541037410497666, "time": 0.8745235919952392, "epoch": 8, "memory": 6319, "step": 29449} +{"lr": 2.2558296734991603e-05, "data_time": 0.532512092590332, "loss": 0.06803475804626942, "time": 0.7765348672866821, "epoch": 8, "memory": 6319, "step": 29549} +{"lr": 2.2308587628570915e-05, "data_time": 0.5307520866394043, "loss": 0.06364604346454143, "time": 0.7839691162109375, "epoch": 8, "memory": 6319, "step": 29649} +{"lr": 2.2060992226624955e-05, "data_time": 0.5516899108886719, "loss": 0.05329091902822256, "time": 0.7754719018936157, "epoch": 8, "memory": 6319, "step": 29749} +{"lr": 2.181552653774648e-05, "data_time": 0.6716019153594971, "loss": 0.0647040979936719, "time": 0.8121907949447632, "epoch": 8, "memory": 6319, "step": 29849} +{"lr": 2.1572206432829035e-05, "data_time": 0.6784110546112061, "loss": 0.07328018620610237, "time": 0.8185736894607544, "epoch": 8, "memory": 6319, "step": 29949} +{"lr": 2.1331047644040754e-05, "data_time": 0.6243602991104126, "loss": 0.06692851856350898, "time": 0.7644826650619507, "epoch": 8, "memory": 6319, "step": 30049} +{"lr": 2.109206576380705e-05, "data_time": 0.677127480506897, "loss": 0.054996407777071, "time": 0.8164906740188599, "epoch": 8, "memory": 6319, "step": 30149} +{"lr": 2.0855276243802687e-05, "data_time": 0.6834668159484864, "loss": 0.062416527047753335, "time": 0.8316063165664673, "epoch": 8, "memory": 6319, "step": 30249} +{"lr": 2.0620694393952487e-05, "data_time": 0.6708291292190551, "loss": 0.0638434112071991, "time": 0.8102986574172973, "epoch": 8, "memory": 6319, "step": 30349} +{"lr": 2.038833538144169e-05, "data_time": 0.6446569204330445, "loss": 0.07554081343114376, "time": 0.7898784875869751, "epoch": 8, "memory": 6319, "step": 30449} +{"lr": 2.0158214229735154e-05, "data_time": 0.6944183826446533, "loss": 0.06999053806066513, "time": 0.8337982892990112, "epoch": 8, "memory": 6319, "step": 30549} +{"lr": 1.9930345817605977e-05, "data_time": 0.6702313899993897, "loss": 0.0658162672072649, "time": 0.8093949079513549, "epoch": 8, "memory": 6319, "step": 30649} +{"lr": 1.970474487817358e-05, "data_time": 0.7079986810684205, "loss": 0.06648847609758377, "time": 0.8471720457077027, "epoch": 8, "memory": 6319, "step": 30749} +{"lr": 1.9481425997951092e-05, "data_time": 0.706857442855835, "loss": 0.05500877685844898, "time": 0.8573665618896484, "epoch": 8, "memory": 6319, "step": 30849} +{"lr": 1.9260403615902187e-05, "data_time": 0.6302733898162842, "loss": 0.05155815072357654, "time": 0.770255708694458, "epoch": 8, "memory": 6319, "step": 30949} +{"lr": 1.9041692022507647e-05, "data_time": 0.681073784828186, "loss": 0.06943475604057311, "time": 0.8212351322174072, "epoch": 8, "memory": 6319, "step": 31049} +{"lr": 1.8825305358841125e-05, "data_time": 0.6838078260421753, "loss": 0.054085373878479004, "time": 0.8238441944122314, "epoch": 8, "memory": 6319, "step": 31149} +{"lr": 1.861125761565518e-05, "data_time": 0.6597942113876343, "loss": 0.064662741497159, "time": 0.7983960390090943, "epoch": 8, "memory": 6319, "step": 31249} +{"accuracy/top1": 97.20956420898438, "data_time": 0.4779896201758549, "time": 0.5642891300135645, "step": 8} +{"lr": 1.8384832449122864e-05, "data_time": 0.5645036935806275, "loss": 0.07345356605947018, "time": 0.9074468851089478, "epoch": 9, "memory": 6319, "step": 31356} +{"lr": 1.817567007332114e-05, "data_time": 0.12101547718048096, "loss": 0.05780891682952642, "time": 0.8669555187225342, "epoch": 9, "memory": 6319, "step": 31456} +{"lr": 1.796888862098825e-05, "data_time": 0.0013144493103027343, "loss": 0.0513185903429985, "time": 0.8499105453491211, "epoch": 9, "memory": 6319, "step": 31556} +{"lr": 1.7764501461839678e-05, "data_time": 0.0007951259613037109, "loss": 0.06022476479411125, "time": 0.775009560585022, "epoch": 9, "memory": 6319, "step": 31656} +{"lr": 1.7562521810784746e-05, "data_time": 0.12217726707458496, "loss": 0.06544761583209038, "time": 0.815608024597168, "epoch": 9, "memory": 6319, "step": 31756} +{"lr": 1.7362962727072444e-05, "data_time": 0.2080303907394409, "loss": 0.05993554592132568, "time": 0.7603644371032715, "epoch": 9, "memory": 6319, "step": 31856} +{"lr": 1.716583711344682e-05, "data_time": 0.14389641284942628, "loss": 0.06042714975774288, "time": 0.8643301486968994, "epoch": 9, "memory": 6319, "step": 31956} +{"lr": 1.6971157715312954e-05, "data_time": 0.0008918046951293945, "loss": 0.06205844022333622, "time": 0.7845324754714966, "epoch": 9, "memory": 6319, "step": 32056} +{"lr": 1.677893711991274e-05, "data_time": 0.05575065612792969, "loss": 0.057245842181146145, "time": 0.756241750717163, "epoch": 9, "memory": 6319, "step": 32156} +{"lr": 1.6589187755511025e-05, "data_time": 0.1641395330429077, "loss": 0.06889153718948364, "time": 0.7898016214370728, "epoch": 9, "memory": 6319, "step": 32256} +{"lr": 1.640192189059216e-05, "data_time": 0.4034645318984985, "loss": 0.0685161579400301, "time": 0.798612093925476, "epoch": 9, "memory": 6319, "step": 32356} +{"lr": 1.6217151633066716e-05, "data_time": 0.658571720123291, "loss": 0.05839485619217157, "time": 0.8009984731674195, "epoch": 9, "memory": 6319, "step": 32456} +{"lr": 1.6034888929488564e-05, "data_time": 0.6858772277832031, "loss": 0.06413462795317174, "time": 0.8266857862472534, "epoch": 9, "memory": 6319, "step": 32556} +{"lr": 1.585514556428258e-05, "data_time": 0.679267168045044, "loss": 0.05429813750088215, "time": 0.8208481311798096, "epoch": 9, "memory": 6319, "step": 32656} +{"lr": 1.567793315898261e-05, "data_time": 0.6696809291839599, "loss": 0.05694551486521959, "time": 0.8106741189956665, "epoch": 9, "memory": 6319, "step": 32756} +{"lr": 1.5503263171480048e-05, "data_time": 0.6769977807998657, "loss": 0.05947272926568985, "time": 0.8180159568786621, "epoch": 9, "memory": 6319, "step": 32856} +{"lr": 1.5331146895283165e-05, "data_time": 0.6756047487258912, "loss": 0.06210103891789913, "time": 0.8163010120391846, "epoch": 9, "memory": 6319, "step": 32956} +{"lr": 1.5161595458786728e-05, "data_time": 0.6639619588851928, "loss": 0.05722990818321705, "time": 0.8037036657333374, "epoch": 9, "memory": 6319, "step": 33056} +{"lr": 1.4994619824552563e-05, "data_time": 0.7145344018936157, "loss": 0.06845532264560461, "time": 0.8550087213516235, "epoch": 9, "memory": 6319, "step": 33156} +{"lr": 1.4830230788600802e-05, "data_time": 0.6928485870361328, "loss": 0.06998468115925789, "time": 0.8321084260940552, "epoch": 9, "memory": 6319, "step": 33256} +{"lr": 1.4668438979711779e-05, "data_time": 0.6831482648849487, "loss": 0.055490275658667085, "time": 0.8225808382034302, "epoch": 9, "memory": 6319, "step": 33356} +{"lr": 1.4509254858738844e-05, "data_time": 0.6699369430541993, "loss": 0.06110903955996037, "time": 0.809241247177124, "epoch": 9, "memory": 6319, "step": 33456} +{"lr": 1.4352688717931966e-05, "data_time": 0.6460357666015625, "loss": 0.0649946790188551, "time": 0.7856189250946045, "epoch": 9, "memory": 6319, "step": 33556} +{"lr": 1.4198750680272348e-05, "data_time": 0.7298983812332154, "loss": 0.05767113342881203, "time": 0.8703232049942017, "epoch": 9, "memory": 6319, "step": 33656} +{"lr": 1.4047450698817857e-05, "data_time": 0.6806979417800904, "loss": 0.05556385312229395, "time": 0.819658613204956, "epoch": 9, "memory": 6319, "step": 33756} +{"lr": 1.389879855605952e-05, "data_time": 0.665081787109375, "loss": 0.07229854948818684, "time": 0.8048717260360718, "epoch": 9, "memory": 6319, "step": 33856} +{"lr": 1.3752803863288976e-05, "data_time": 0.6799609422683716, "loss": 0.06413264237344266, "time": 0.8191285848617553, "epoch": 9, "memory": 6319, "step": 33956} +{"lr": 1.3609476059977129e-05, "data_time": 0.6685647010803223, "loss": 0.0635818213224411, "time": 0.8143080711364746, "epoch": 9, "memory": 6319, "step": 34056} +{"lr": 1.3468824413163754e-05, "data_time": 0.6434451580047608, "loss": 0.06423680298030376, "time": 0.7831258773803711, "epoch": 9, "memory": 6319, "step": 34156} +{"lr": 1.3330858016858395e-05, "data_time": 0.687971568107605, "loss": 0.06455809995532036, "time": 0.8318761348724365, "epoch": 9, "memory": 6319, "step": 34256} +{"lr": 1.3195585791452309e-05, "data_time": 0.6910084247589111, "loss": 0.0646224495023489, "time": 0.8315199851989746, "epoch": 9, "memory": 6319, "step": 34356} +{"lr": 1.3063016483141729e-05, "data_time": 0.6375117540359497, "loss": 0.059206053614616394, "time": 0.777857494354248, "epoch": 9, "memory": 6319, "step": 34456} +{"lr": 1.2933158663362388e-05, "data_time": 0.6698830366134644, "loss": 0.052413994818925856, "time": 0.810014295578003, "epoch": 9, "memory": 6319, "step": 34556} +{"lr": 1.2806020728235311e-05, "data_time": 0.6686711549758911, "loss": 0.06219264827668667, "time": 0.807573938369751, "epoch": 9, "memory": 6319, "step": 34656} +{"lr": 1.2681610898023963e-05, "data_time": 0.6727989435195922, "loss": 0.05427257437258959, "time": 0.8126304149627686, "epoch": 9, "memory": 6319, "step": 34756} +{"lr": 1.2559937216602731e-05, "data_time": 0.6883304119110107, "loss": 0.05407445915043354, "time": 0.8308426141738892, "epoch": 9, "memory": 6319, "step": 34856} +{"lr": 1.2441007550936847e-05, "data_time": 0.6946065187454223, "loss": 0.08329944387078285, "time": 0.8352630376815796, "epoch": 9, "memory": 6319, "step": 34956} +{"lr": 1.232482959057377e-05, "data_time": 0.7055997610092163, "loss": 0.062410608679056165, "time": 0.8447673320770264, "epoch": 9, "memory": 6319, "step": 35056} +{"lr": 1.2211410847145969e-05, "data_time": 0.6372318029403686, "loss": 0.06384058706462384, "time": 0.7763364553451538, "epoch": 9, "memory": 6319, "step": 35156} +{"accuracy/top1": 97.25131225585938, "data_time": 0.47774931480144633, "time": 0.5636117006170338, "step": 9} +{"lr": 1.209311679360931e-05, "data_time": 0.39909961223602297, "loss": 0.05663879550993443, "time": 0.8313610076904296, "epoch": 10, "memory": 6319, "step": 35263} +{"lr": 1.1985432727488666e-05, "data_time": 0.3483341455459595, "loss": 0.06078498288989067, "time": 0.7888491153717041, "epoch": 10, "memory": 6319, "step": 35363} +{"lr": 1.1880529822435289e-05, "data_time": 0.1795570135116577, "loss": 0.05528392493724823, "time": 0.7648987293243408, "epoch": 10, "memory": 6319, "step": 35463} +{"lr": 1.1778414861078726e-05, "data_time": 0.19396381378173827, "loss": 0.06417308263480663, "time": 0.8525235652923584, "epoch": 10, "memory": 6319, "step": 35563} +{"lr": 1.1679094445790538e-05, "data_time": 0.09389820098876953, "loss": 0.06285519935190678, "time": 0.8113935708999633, "epoch": 10, "memory": 6319, "step": 35663} +{"lr": 1.1582574998257377e-05, "data_time": 0.2651108741760254, "loss": 0.06844070069491863, "time": 0.7451489686965942, "epoch": 10, "memory": 6319, "step": 35763} +{"lr": 1.1488862759065762e-05, "data_time": 0.401881742477417, "loss": 0.07080382071435451, "time": 0.7836827039718628, "epoch": 10, "memory": 6319, "step": 35863} +{"lr": 1.1397963787298665e-05, "data_time": 0.6262887716293335, "loss": 0.06453243866562844, "time": 0.8084224462509155, "epoch": 10, "memory": 6319, "step": 35963} +{"lr": 1.1309883960143664e-05, "data_time": 0.6166814804077149, "loss": 0.05892754942178726, "time": 0.8026927947998047, "epoch": 10, "memory": 6319, "step": 36063} +{"lr": 1.122462897251302e-05, "data_time": 0.6859795570373535, "loss": 0.06453697718679904, "time": 0.827719783782959, "epoch": 10, "memory": 6319, "step": 36163} +{"lr": 1.114220433667542e-05, "data_time": 0.7088356971740722, "loss": 0.054554440826177594, "time": 0.8499486446380615, "epoch": 10, "memory": 6319, "step": 36263} +{"lr": 1.10626153818996e-05, "data_time": 0.6648312568664551, "loss": 0.061416537314653394, "time": 0.8116134881973267, "epoch": 10, "memory": 6319, "step": 36363} +{"lr": 1.0985867254109761e-05, "data_time": 0.7136873483657837, "loss": 0.07393808402121067, "time": 0.854666018486023, "epoch": 10, "memory": 6319, "step": 36463} +{"lr": 1.0911964915552823e-05, "data_time": 0.6773239850997925, "loss": 0.05446305610239506, "time": 0.8185667037963867, "epoch": 10, "memory": 6319, "step": 36563} +{"lr": 1.0840913144477643e-05, "data_time": 0.6711352586746215, "loss": 0.06571899875998496, "time": 0.8102775096893311, "epoch": 10, "memory": 6319, "step": 36663} +{"lr": 1.0772716534826064e-05, "data_time": 0.675044584274292, "loss": 0.05931845977902413, "time": 0.8152484178543091, "epoch": 10, "memory": 6319, "step": 36763} +{"lr": 1.0707379495935828e-05, "data_time": 0.7159065485000611, "loss": 0.059493158385157584, "time": 0.8574132919311523, "epoch": 10, "memory": 6319, "step": 36863} +{"lr": 1.0644906252255541e-05, "data_time": 0.6791799306869507, "loss": 0.06316351033747196, "time": 0.8191067218780518, "epoch": 10, "memory": 6319, "step": 36963} +{"lr": 1.0585300843071527e-05, "data_time": 0.6151295185089112, "loss": 0.06075747609138489, "time": 0.7542976856231689, "epoch": 10, "memory": 6319, "step": 37063} +{"lr": 1.0528567122246654e-05, "data_time": 0.6234589576721191, "loss": 0.050756143592298034, "time": 0.7630714893341064, "epoch": 10, "memory": 6319, "step": 37163} +{"lr": 1.0474708757971141e-05, "data_time": 0.6771501779556275, "loss": 0.06674916744232177, "time": 0.8185913801193238, "epoch": 10, "memory": 6319, "step": 37263} +{"lr": 1.042372923252543e-05, "data_time": 0.6750219583511352, "loss": 0.0669028852134943, "time": 0.8143959045410156, "epoch": 10, "memory": 6319, "step": 37363} +{"lr": 1.0375631842054998e-05, "data_time": 0.6899311542510986, "loss": 0.05886037386953831, "time": 0.8302990436553955, "epoch": 10, "memory": 6319, "step": 37463} +{"lr": 1.0330419696357254e-05, "data_time": 0.7050694227218628, "loss": 0.061891605332493785, "time": 0.8458878040313721, "epoch": 10, "memory": 6319, "step": 37563} +{"lr": 1.028809571868047e-05, "data_time": 0.7521899938583374, "loss": 0.06351680010557174, "time": 0.8917283773422241, "epoch": 10, "memory": 6319, "step": 37663} +{"lr": 1.0248662645534766e-05, "data_time": 0.6496395587921142, "loss": 0.058163994923233986, "time": 0.7883558511734009, "epoch": 10, "memory": 6319, "step": 37763} +{"lr": 1.0212123026515194e-05, "data_time": 0.6573646783828735, "loss": 0.07633700519800186, "time": 0.7970118284225464, "epoch": 10, "memory": 6319, "step": 37863} +{"lr": 1.0178479224136881e-05, "data_time": 0.7019353628158569, "loss": 0.06876394785940647, "time": 0.8414695739746094, "epoch": 10, "memory": 6319, "step": 37963} +{"lr": 1.0147733413682289e-05, "data_time": 0.694713544845581, "loss": 0.06688739284873009, "time": 0.8349694728851318, "epoch": 10, "memory": 6319, "step": 38063} +{"lr": 1.0119887583060527e-05, "data_time": 0.6461351394653321, "loss": 0.06250211223959923, "time": 0.7862644672393799, "epoch": 10, "memory": 6319, "step": 38163} +{"lr": 1.0094943532678871e-05, "data_time": 0.660258412361145, "loss": 0.05961154475808143, "time": 0.8006170988082886, "epoch": 10, "memory": 6319, "step": 38263} +{"lr": 1.007290287532637e-05, "data_time": 0.6291829824447632, "loss": 0.07028537392616271, "time": 0.7687488794326782, "epoch": 10, "memory": 6319, "step": 38363} +{"lr": 1.0053767036069495e-05, "data_time": 0.7024988889694214, "loss": 0.06954733319580556, "time": 0.8421658992767334, "epoch": 10, "memory": 6319, "step": 38463} +{"lr": 1.003753725216005e-05, "data_time": 0.7504273176193237, "loss": 0.06711390689015388, "time": 0.8897173404693604, "epoch": 10, "memory": 6319, "step": 38563} +{"lr": 1.0024214572955152e-05, "data_time": 0.6651474237442017, "loss": 0.061823930218815805, "time": 0.8049480676651001, "epoch": 10, "memory": 6319, "step": 38663} +{"lr": 1.0013799859849456e-05, "data_time": 0.6606799364089966, "loss": 0.05535350404679775, "time": 0.8006099700927735, "epoch": 10, "memory": 6319, "step": 38763} +{"lr": 1.0006293786219338e-05, "data_time": 0.653389835357666, "loss": 0.058217883482575415, "time": 0.7929687261581421, "epoch": 10, "memory": 6319, "step": 38863} +{"lr": 1.0001696837379472e-05, "data_time": 0.6799966812133789, "loss": 0.05455840863287449, "time": 0.8202425718307496, "epoch": 10, "memory": 6319, "step": 38963} +{"lr": 1.0000009310551361e-05, "data_time": 0.692695426940918, "loss": 0.06514642164111137, "time": 0.8321401119232178, "epoch": 10, "memory": 6319, "step": 39063} +{"accuracy/top1": 97.5296630859375, "data_time": 0.5047953704307819, "time": 0.5928477418833765, "step": 10} diff --git a/resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1/20230604_215105/vis_data/config.py b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1/20230604_215105/vis_data/config.py new file mode 100644 index 0000000000000000000000000000000000000000..32dc328c1686171e778233c0a8aeab585932a5fe --- /dev/null +++ b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1/20230604_215105/vis_data/config.py @@ -0,0 +1,230 @@ +optim_wrapper = dict( + optimizer=dict( + type='SGD', + lr=0.0001, + momentum=0.9, + weight_decay=0.0001, + _scope_='mmpretrain'), + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=False, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=512) +model = dict( + type='ImageClassifier', + backbone=dict( + frozen_stages=2, + type='ResNet', + depth=50, + num_stages=4, + out_indices=(3, ), + style='pytorch', + init_cfg=dict( + type='Pretrained', + checkpoint= + 'https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth', + prefix='backbone')), + neck=dict(type='GlobalAveragePooling'), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=2048, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + topk=1)) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stablediffusionV1-5R2-dpmsolver-25-1m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/cc1m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = dict(type='Accuracy', topk=1) +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = dict(type='Accuracy', topk=1) +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1' diff --git a/resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1/20230604_215105/vis_data/events.out.tfevents.1685886680.SH-IDC1-10-140-24-129.14946.0 b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1/20230604_215105/vis_data/events.out.tfevents.1685886680.SH-IDC1-10-140-24-129.14946.0 new file mode 100644 index 0000000000000000000000000000000000000000..8209bd934c68ee0ddd49dd8d971fa7738bca8795 --- /dev/null +++ b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1/20230604_215105/vis_data/events.out.tfevents.1685886680.SH-IDC1-10-140-24-129.14946.0 @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:4104444be8e8f024ab6701df29f8d1509876096ad17333abf0969525f5a28445 +size 2317272 diff --git a/resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1/20230604_215105/vis_data/scalars.json b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1/20230604_215105/vis_data/scalars.json new file mode 100644 index 0000000000000000000000000000000000000000..82d3284e451bc76603b545c6bab6b12f54bb0c54 --- /dev/null +++ b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1/20230604_215105/vis_data/scalars.json @@ -0,0 +1,400 @@ +{"lr": 9.999857418382631e-05, "data_time": 0.6374414443969727, "loss": 0.601702731847763, "time": 0.8555856943130493, "epoch": 1, "memory": 9436, "step": 100} +{"lr": 9.999423907359506e-05, "data_time": 0.6512850522994995, "loss": 0.5355101227760315, "time": 0.8826247930526734, "epoch": 1, "memory": 6319, "step": 200} +{"lr": 9.998699480412073e-05, "data_time": 0.62061927318573, "loss": 0.46963952779769896, "time": 0.8177884101867676, "epoch": 1, "memory": 6319, "step": 300} +{"lr": 9.997684184379085e-05, "data_time": 0.7160451650619507, "loss": 0.42892860770225527, "time": 0.8552490472793579, "epoch": 1, "memory": 6319, "step": 400} +{"lr": 9.996378084905784e-05, "data_time": 0.5343232154846191, "loss": 0.39631963074207305, "time": 0.8138736248016357, "epoch": 1, "memory": 6319, "step": 500} +{"lr": 9.994781266439667e-05, "data_time": 0.5577570915222168, "loss": 0.37549435496330263, "time": 0.8080990791320801, "epoch": 1, "memory": 6319, "step": 600} +{"lr": 9.992893832225061e-05, "data_time": 0.4441789150238037, "loss": 0.36862215101718904, "time": 0.8239611864089966, "epoch": 1, "memory": 6319, "step": 700} +{"lr": 9.990715904296405e-05, "data_time": 0.611823320388794, "loss": 0.31859688758850097, "time": 0.8215285062789917, "epoch": 1, "memory": 6319, "step": 800} +{"lr": 9.988247623470371e-05, "data_time": 0.6151374340057373, "loss": 0.3201468825340271, "time": 0.8195865154266357, "epoch": 1, "memory": 6319, "step": 900} +{"lr": 9.985489149336749e-05, "data_time": 0.6461711883544922, "loss": 0.2996815800666809, "time": 0.7850286960601807, "epoch": 1, "memory": 6319, "step": 1000} +{"lr": 9.982440660248187e-05, "data_time": 0.6466344833374024, "loss": 0.29422697573900225, "time": 0.7864808320999146, "epoch": 1, "memory": 6319, "step": 1100} +{"lr": 9.979102353308574e-05, "data_time": 0.6717127084732055, "loss": 0.29378020763397217, "time": 0.8114043951034546, "epoch": 1, "memory": 6319, "step": 1200} +{"lr": 9.975474444360358e-05, "data_time": 0.6845787286758422, "loss": 0.2830167576670647, "time": 0.8234544992446899, "epoch": 1, "memory": 6319, "step": 1300} +{"lr": 9.97155716797056e-05, "data_time": 0.6955356121063232, "loss": 0.2428518310189247, "time": 0.8341902494430542, "epoch": 1, "memory": 6319, "step": 1400} +{"lr": 9.967350777415632e-05, "data_time": 0.7570158004760742, "loss": 0.23246523439884187, "time": 0.8962111234664917, "epoch": 1, "memory": 6319, "step": 1500} +{"lr": 9.962855544665059e-05, "data_time": 0.6592184782028199, "loss": 0.25699472427368164, "time": 0.7987259864807129, "epoch": 1, "memory": 6319, "step": 1600} +{"lr": 9.958071760363773e-05, "data_time": 0.700481367111206, "loss": 0.2340873584151268, "time": 0.84044349193573, "epoch": 1, "memory": 6319, "step": 1700} +{"lr": 9.952999733813382e-05, "data_time": 0.44866161346435546, "loss": 0.21638996303081512, "time": 0.8508273839950562, "epoch": 1, "memory": 6319, "step": 1800} +{"lr": 9.947639792952148e-05, "data_time": 0.47938241958618166, "loss": 0.2403170332312584, "time": 0.851843285560608, "epoch": 1, "memory": 6319, "step": 1900} +{"lr": 9.941992284333813e-05, "data_time": 0.6045541524887085, "loss": 0.21530583649873733, "time": 0.8377137899398803, "epoch": 1, "memory": 6319, "step": 2000} +{"lr": 9.93605757310516e-05, "data_time": 0.6837259769439697, "loss": 0.22493914216756822, "time": 0.8786030769348144, "epoch": 1, "memory": 6319, "step": 2100} +{"lr": 9.929836042982421e-05, "data_time": 0.7660118103027344, "loss": 0.22159391790628433, "time": 0.9059905529022216, "epoch": 1, "memory": 6319, "step": 2200} +{"lr": 9.923328096226467e-05, "data_time": 0.7311722278594971, "loss": 0.224177086353302, "time": 0.8855130910873413, "epoch": 1, "memory": 6319, "step": 2300} +{"lr": 9.916534153616819e-05, "data_time": 0.7572393178939819, "loss": 0.18926170617341995, "time": 0.8972986936569214, "epoch": 1, "memory": 6319, "step": 2400} +{"lr": 9.909454654424387e-05, "data_time": 0.6951786756515503, "loss": 0.20676430612802504, "time": 0.8493762731552124, "epoch": 1, "memory": 6319, "step": 2500} +{"lr": 9.902090056383109e-05, "data_time": 0.7557082891464233, "loss": 0.1941266715526581, "time": 0.8950440168380738, "epoch": 1, "memory": 6319, "step": 2600} +{"lr": 9.894440835660365e-05, "data_time": 0.67665593624115, "loss": 0.19437728822231293, "time": 0.8159930229187011, "epoch": 1, "memory": 6319, "step": 2700} +{"lr": 9.886507486826162e-05, "data_time": 0.7613880872726441, "loss": 0.19213476926088333, "time": 0.9016317844390869, "epoch": 1, "memory": 6319, "step": 2800} +{"lr": 9.878290522821154e-05, "data_time": 0.7427852153778076, "loss": 0.20432229340076447, "time": 0.8836068868637085, "epoch": 1, "memory": 6319, "step": 2900} +{"lr": 9.86979047492352e-05, "data_time": 0.7449662208557128, "loss": 0.17515064626932145, "time": 0.8846933603286743, "epoch": 1, "memory": 6319, "step": 3000} +{"lr": 9.861007892714578e-05, "data_time": 0.7171546936035156, "loss": 0.1845693200826645, "time": 0.8594824552536011, "epoch": 1, "memory": 6319, "step": 3100} +{"lr": 9.851943344043239e-05, "data_time": 0.7557272434234619, "loss": 0.1728598140180111, "time": 0.8962688207626343, "epoch": 1, "memory": 6319, "step": 3200} +{"lr": 9.842597414989319e-05, "data_time": 0.6702453374862671, "loss": 0.18007278442382812, "time": 0.8097019672393799, "epoch": 1, "memory": 6319, "step": 3300} +{"lr": 9.832970709825637e-05, "data_time": 0.6524338006973267, "loss": 0.16683334857225418, "time": 0.8048237323760986, "epoch": 1, "memory": 6319, "step": 3400} +{"lr": 9.823063850978968e-05, "data_time": 0.6931561470031739, "loss": 0.18467812538146972, "time": 0.8349983215332031, "epoch": 1, "memory": 6319, "step": 3500} +{"lr": 9.81287747898975e-05, "data_time": 0.6722934484481812, "loss": 0.15657266080379487, "time": 0.827212643623352, "epoch": 1, "memory": 6319, "step": 3600} +{"lr": 9.802412252470715e-05, "data_time": 0.6636826515197753, "loss": 0.1768097162246704, "time": 0.8044445753097534, "epoch": 1, "memory": 6319, "step": 3700} +{"lr": 9.791668848064258e-05, "data_time": 0.601073956489563, "loss": 0.18478747755289077, "time": 0.8198684692382813, "epoch": 1, "memory": 6319, "step": 3800} +{"lr": 9.780647960398751e-05, "data_time": 0.6998269557952881, "loss": 0.16740401089191437, "time": 0.8399232864379883, "epoch": 1, "memory": 6319, "step": 3900} +{"accuracy/top1": 81.12104797363281, "data_time": 0.5407230603067499, "time": 0.6341041807542768, "step": 1} +{"lr": 9.7685491196646e-05, "data_time": 0.4053492069244385, "loss": 0.164532807469368, "time": 0.8102155685424804, "epoch": 2, "memory": 6319, "step": 4007} +{"lr": 9.756956126085676e-05, "data_time": 0.24088940620422364, "loss": 0.16619824916124343, "time": 0.8014201641082763, "epoch": 2, "memory": 6319, "step": 4107} +{"lr": 9.745087893642482e-05, "data_time": 0.1316983222961426, "loss": 0.16588684022426606, "time": 0.7923794269561768, "epoch": 2, "memory": 6319, "step": 4207} +{"lr": 9.73294518969056e-05, "data_time": 0.3431976318359375, "loss": 0.1442151889204979, "time": 0.8041167974472045, "epoch": 2, "memory": 6319, "step": 4307} +{"lr": 9.72052879933172e-05, "data_time": 0.3228480339050293, "loss": 0.15931000784039498, "time": 0.7638721227645874, "epoch": 2, "memory": 6319, "step": 4407} +{"lr": 9.707839525363358e-05, "data_time": 0.26673552989959715, "loss": 0.14772066250443458, "time": 0.8368432760238648, "epoch": 2, "memory": 6319, "step": 4507} +{"lr": 9.69487818822646e-05, "data_time": 0.3482442140579224, "loss": 0.1541784442961216, "time": 0.8618246555328369, "epoch": 2, "memory": 6319, "step": 4607} +{"lr": 9.681645625952616e-05, "data_time": 0.5624294757843018, "loss": 0.15874218046665192, "time": 0.8391940355300903, "epoch": 2, "memory": 6319, "step": 4707} +{"lr": 9.668142694109826e-05, "data_time": 0.7022975206375122, "loss": 0.15026861429214478, "time": 0.863461709022522, "epoch": 2, "memory": 6319, "step": 4807} +{"lr": 9.654370265747162e-05, "data_time": 0.5626144409179688, "loss": 0.16405150443315505, "time": 0.8262191772460937, "epoch": 2, "memory": 6319, "step": 4907} +{"lr": 9.640329231338345e-05, "data_time": 0.7439480781555176, "loss": 0.1409287489950657, "time": 0.8972822427749634, "epoch": 2, "memory": 6319, "step": 5007} +{"lr": 9.626020498724147e-05, "data_time": 0.6420143127441407, "loss": 0.13239647075533867, "time": 0.8961099624633789, "epoch": 2, "memory": 6319, "step": 5107} +{"lr": 9.611444993053717e-05, "data_time": 0.2255570650100708, "loss": 0.14459227547049522, "time": 0.8393664598464966, "epoch": 2, "memory": 6319, "step": 5207} +{"lr": 9.596603656724746e-05, "data_time": 0.0013016700744628907, "loss": 0.14789986237883568, "time": 0.8949613809585572, "epoch": 2, "memory": 6319, "step": 5307} +{"lr": 9.58149744932254e-05, "data_time": 0.0016445159912109376, "loss": 0.16496630012989044, "time": 0.9058290719985962, "epoch": 2, "memory": 6319, "step": 5407} +{"lr": 9.566127347557972e-05, "data_time": 0.0010085344314575196, "loss": 0.12877277731895448, "time": 0.8427849292755127, "epoch": 2, "memory": 6319, "step": 5507} +{"lr": 9.550494345204367e-05, "data_time": 0.5324305772781373, "loss": 0.1346907749772072, "time": 0.8919317960739136, "epoch": 2, "memory": 6319, "step": 5607} +{"lr": 9.534599453033183e-05, "data_time": 0.33181347846984866, "loss": 0.14077757373452188, "time": 0.836850643157959, "epoch": 2, "memory": 6319, "step": 5707} +{"lr": 9.518443698748712e-05, "data_time": 0.37905948162078856, "loss": 0.13480956330895424, "time": 0.7978240966796875, "epoch": 2, "memory": 6319, "step": 5807} +{"lr": 9.50202812692162e-05, "data_time": 0.2967047214508057, "loss": 0.14967381954193115, "time": 0.8451672077178956, "epoch": 2, "memory": 6319, "step": 5907} +{"lr": 9.485353798921386e-05, "data_time": 0.530403995513916, "loss": 0.13241284266114234, "time": 0.8520182132720947, "epoch": 2, "memory": 6319, "step": 6007} +{"lr": 9.468421792847712e-05, "data_time": 0.6336118698120117, "loss": 0.12861084043979645, "time": 0.8335238218307495, "epoch": 2, "memory": 6319, "step": 6107} +{"lr": 9.451233203460794e-05, "data_time": 0.6685770273208618, "loss": 0.14063130617141723, "time": 0.8086844921112061, "epoch": 2, "memory": 6319, "step": 6207} +{"lr": 9.433789142110532e-05, "data_time": 0.6530247211456299, "loss": 0.14641004279255868, "time": 0.7929362297058106, "epoch": 2, "memory": 6319, "step": 6307} +{"lr": 9.416090736664709e-05, "data_time": 0.44009592533111574, "loss": 0.12866621166467668, "time": 0.846757459640503, "epoch": 2, "memory": 6319, "step": 6407} +{"lr": 9.398139131436032e-05, "data_time": 0.21391720771789552, "loss": 0.1294976994395256, "time": 0.8114058256149292, "epoch": 2, "memory": 6319, "step": 6507} +{"lr": 9.379935487108167e-05, "data_time": 0.07746953964233398, "loss": 0.12480120584368706, "time": 0.8726335048675538, "epoch": 2, "memory": 6319, "step": 6607} +{"lr": 9.361480980660697e-05, "data_time": 0.0010189771652221679, "loss": 0.13518597483634948, "time": 0.7963105916976929, "epoch": 2, "memory": 6319, "step": 6707} +{"lr": 9.342776805292993e-05, "data_time": 0.0010326385498046875, "loss": 0.12527859807014466, "time": 0.8246926069259644, "epoch": 2, "memory": 6319, "step": 6807} +{"lr": 9.323824170347083e-05, "data_time": 0.010301589965820312, "loss": 0.11467275694012642, "time": 0.7805991888046264, "epoch": 2, "memory": 6319, "step": 6907} +{"lr": 9.304624301229471e-05, "data_time": 0.5237648010253906, "loss": 0.1257191315293312, "time": 0.826194429397583, "epoch": 2, "memory": 6319, "step": 7007} +{"lr": 9.285178439331897e-05, "data_time": 0.6844304800033569, "loss": 0.1265637643635273, "time": 0.824958086013794, "epoch": 2, "memory": 6319, "step": 7107} +{"lr": 9.265487841951087e-05, "data_time": 0.6880308866500855, "loss": 0.1203043982386589, "time": 0.8342239856719971, "epoch": 2, "memory": 6319, "step": 7207} +{"lr": 9.245553782207421e-05, "data_time": 0.6624742269515991, "loss": 0.12059230878949165, "time": 0.8033913373947144, "epoch": 2, "memory": 6319, "step": 7307} +{"lr": 9.225377548962671e-05, "data_time": 0.6826150178909302, "loss": 0.12225314378738403, "time": 0.8241559743881226, "epoch": 2, "memory": 6319, "step": 7407} +{"lr": 9.204960446736601e-05, "data_time": 0.6545940160751342, "loss": 0.12865836247801782, "time": 0.8548345565795898, "epoch": 2, "memory": 6319, "step": 7507} +{"lr": 9.184303795622715e-05, "data_time": 0.676851463317871, "loss": 0.13751766011118888, "time": 0.8155324935913086, "epoch": 2, "memory": 6319, "step": 7607} +{"lr": 9.163408931202823e-05, "data_time": 0.6777835369110108, "loss": 0.11660075187683105, "time": 0.817630124092102, "epoch": 2, "memory": 6319, "step": 7707} +{"lr": 9.142277204460729e-05, "data_time": 0.6675420522689819, "loss": 0.11041985228657722, "time": 0.8077948331832886, "epoch": 2, "memory": 6319, "step": 7807} +{"accuracy/top1": 87.63091278076172, "data_time": 0.5409339337513365, "time": 0.6319431518686229, "step": 2} +{"lr": 9.119405492383288e-05, "data_time": 0.5990873575210571, "loss": 0.1079963929951191, "time": 0.8015936136245727, "epoch": 3, "memory": 6319, "step": 7914} +{"lr": 9.097787819233205e-05, "data_time": 0.6570791482925415, "loss": 0.11941554769873619, "time": 0.8080140113830566, "epoch": 3, "memory": 6319, "step": 8014} +{"lr": 9.075937526576627e-05, "data_time": 0.6611604452133178, "loss": 0.10831483006477356, "time": 0.8022749423980713, "epoch": 3, "memory": 6319, "step": 8114} +{"lr": 9.053856027171755e-05, "data_time": 0.6530561208724975, "loss": 0.11707084774971008, "time": 0.7928166627883911, "epoch": 3, "memory": 6319, "step": 8214} +{"lr": 9.031544748725785e-05, "data_time": 0.2923057317733765, "loss": 0.10115634873509408, "time": 0.8415463685989379, "epoch": 3, "memory": 6319, "step": 8314} +{"lr": 9.009005133802542e-05, "data_time": 0.4353105306625366, "loss": 0.10452483519911766, "time": 0.8080923080444335, "epoch": 3, "memory": 6319, "step": 8414} +{"lr": 8.986238639729242e-05, "data_time": 0.7275919675827026, "loss": 0.10942161232233047, "time": 0.918468427658081, "epoch": 3, "memory": 6319, "step": 8514} +{"lr": 8.963246738502254e-05, "data_time": 0.6077797412872314, "loss": 0.12090184316039085, "time": 0.8502497434616089, "epoch": 3, "memory": 6319, "step": 8614} +{"lr": 8.940030916691928e-05, "data_time": 0.6876582860946655, "loss": 0.11244189366698265, "time": 0.8271790981292725, "epoch": 3, "memory": 6319, "step": 8714} +{"lr": 8.91659267534649e-05, "data_time": 0.6878661394119263, "loss": 0.10171929523348808, "time": 0.8271135330200196, "epoch": 3, "memory": 6319, "step": 8814} +{"lr": 8.892933529894978e-05, "data_time": 0.6629051685333252, "loss": 0.10485116988420487, "time": 0.8020254850387574, "epoch": 3, "memory": 6319, "step": 8914} +{"lr": 8.869055010049283e-05, "data_time": 0.7062404870986938, "loss": 0.09983178451657296, "time": 0.8448907613754273, "epoch": 3, "memory": 6319, "step": 9014} +{"lr": 8.844958659705188e-05, "data_time": 0.7121747255325317, "loss": 0.09979011118412018, "time": 0.8525150775909424, "epoch": 3, "memory": 6319, "step": 9114} +{"lr": 8.82064603684259e-05, "data_time": 0.6661271572113037, "loss": 0.09753329679369926, "time": 0.8060618162155151, "epoch": 3, "memory": 6319, "step": 9214} +{"lr": 8.796118713424776e-05, "data_time": 0.6657455921173095, "loss": 0.10611129701137542, "time": 0.8054347515106202, "epoch": 3, "memory": 6319, "step": 9314} +{"lr": 8.77137827529676e-05, "data_time": 0.9204483747482299, "loss": 0.11277874186635017, "time": 1.0597905158996581, "epoch": 3, "memory": 6319, "step": 9414} +{"lr": 8.746426322082745e-05, "data_time": 0.6355006217956543, "loss": 0.11189985498785973, "time": 0.7843372821807861, "epoch": 3, "memory": 6319, "step": 9514} +{"lr": 8.721264467082732e-05, "data_time": 0.6172508478164673, "loss": 0.11194451823830605, "time": 0.7983240842819214, "epoch": 3, "memory": 6319, "step": 9614} +{"lr": 8.695894337168168e-05, "data_time": 0.7082661867141724, "loss": 0.1246405228972435, "time": 0.8474427223205566, "epoch": 3, "memory": 6319, "step": 9714} +{"lr": 8.67031757267677e-05, "data_time": 0.6869972229003907, "loss": 0.08495607413351536, "time": 0.826607084274292, "epoch": 3, "memory": 6319, "step": 9814} +{"lr": 8.64453582730651e-05, "data_time": 0.7059129476547241, "loss": 0.10880973637104034, "time": 0.8447617769241333, "epoch": 3, "memory": 6319, "step": 9914} +{"lr": 8.61855076800861e-05, "data_time": 0.7079426765441894, "loss": 0.09506015405058861, "time": 0.8473704576492309, "epoch": 3, "memory": 6319, "step": 10014} +{"lr": 8.592364074879843e-05, "data_time": 0.6977201938629151, "loss": 0.11660193353891372, "time": 0.8378056049346924, "epoch": 3, "memory": 6319, "step": 10114} +{"lr": 8.565977441053854e-05, "data_time": 0.6644403457641601, "loss": 0.09511319696903228, "time": 0.8034246444702149, "epoch": 3, "memory": 6319, "step": 10214} +{"lr": 8.539392572591724e-05, "data_time": 0.667553162574768, "loss": 0.11333511024713516, "time": 0.8074770450592041, "epoch": 3, "memory": 6319, "step": 10314} +{"lr": 8.512611188371633e-05, "data_time": 0.7124860048294067, "loss": 0.09126886427402496, "time": 0.8519822359085083, "epoch": 3, "memory": 6319, "step": 10414} +{"lr": 8.485635019977725e-05, "data_time": 0.7116752147674561, "loss": 0.10744883194565773, "time": 0.8510948181152344, "epoch": 3, "memory": 6319, "step": 10514} +{"lr": 8.45846581158818e-05, "data_time": 0.6724309682846069, "loss": 0.10148456953465938, "time": 0.8115061044692993, "epoch": 3, "memory": 6319, "step": 10614} +{"lr": 8.431105319862416e-05, "data_time": 0.6431312561035156, "loss": 0.0994156077504158, "time": 0.7821248292922973, "epoch": 3, "memory": 6319, "step": 10714} +{"lr": 8.403555313827516e-05, "data_time": 0.7016383171081543, "loss": 0.10597254410386085, "time": 0.8413288354873657, "epoch": 3, "memory": 6319, "step": 10814} +{"lr": 8.375817574763844e-05, "data_time": 0.7045982837677002, "loss": 0.09045120924711228, "time": 0.8459288835525512, "epoch": 3, "memory": 6319, "step": 10914} +{"lr": 8.347893896089871e-05, "data_time": 0.6797439575195312, "loss": 0.09476337805390359, "time": 0.8190370798110962, "epoch": 3, "memory": 6319, "step": 11014} +{"lr": 8.31978608324625e-05, "data_time": 0.7089139461517334, "loss": 0.1007885254919529, "time": 0.8491958379745483, "epoch": 3, "memory": 6319, "step": 11114} +{"lr": 8.291495953579052e-05, "data_time": 0.6972779750823974, "loss": 0.08978517912328243, "time": 0.8372002363204956, "epoch": 3, "memory": 6319, "step": 11214} +{"lr": 8.263025336222267e-05, "data_time": 0.5805172204971314, "loss": 0.09269502758979797, "time": 0.7206877946853638, "epoch": 3, "memory": 6319, "step": 11314} +{"lr": 8.234376071979551e-05, "data_time": 0.6612140417099, "loss": 0.09400350823998452, "time": 0.8000160932540894, "epoch": 3, "memory": 6319, "step": 11414} +{"lr": 8.205550013205187e-05, "data_time": 0.6754632711410522, "loss": 0.09037117287516594, "time": 0.817452073097229, "epoch": 3, "memory": 6319, "step": 11514} +{"lr": 8.17654902368435e-05, "data_time": 0.6652022123336792, "loss": 0.0913025826215744, "time": 0.8047217845916748, "epoch": 3, "memory": 6319, "step": 11614} +{"lr": 8.147374978512582e-05, "data_time": 0.6747350215911865, "loss": 0.08519593179225922, "time": 0.8134832382202148, "epoch": 3, "memory": 6319, "step": 11714} +{"accuracy/top1": 91.84092712402344, "data_time": 0.48619574924995157, "time": 0.5769296843430092, "step": 3} +{"lr": 8.115969237626578e-05, "data_time": 0.70062735080719, "loss": 0.08016534075140953, "time": 0.8406968116760254, "epoch": 4, "memory": 6319, "step": 11821} +{"lr": 8.08644297337245e-05, "data_time": 0.7934768915176391, "loss": 0.08904951065778732, "time": 0.9452500343322754, "epoch": 4, "memory": 6319, "step": 11921} +{"lr": 8.056749479387651e-05, "data_time": 0.28671603202819823, "loss": 0.0891688846051693, "time": 0.6829891443252564, "epoch": 4, "memory": 6319, "step": 12021} +{"lr": 8.026890675542457e-05, "data_time": 0.44856834411621094, "loss": 0.0916030652821064, "time": 0.8336772203445435, "epoch": 4, "memory": 6319, "step": 12121} +{"lr": 7.996868492395457e-05, "data_time": 0.5824866771697998, "loss": 0.08735960982739925, "time": 0.8282800197601319, "epoch": 4, "memory": 6319, "step": 12221} +{"lr": 7.966684871068762e-05, "data_time": 0.7231475353240967, "loss": 0.09666001684963703, "time": 0.862327241897583, "epoch": 4, "memory": 6319, "step": 12321} +{"lr": 7.936341763122449e-05, "data_time": 0.7115377902984619, "loss": 0.08375656008720397, "time": 0.8511252641677857, "epoch": 4, "memory": 6319, "step": 12421} +{"lr": 7.905841130428405e-05, "data_time": 0.6500834465026856, "loss": 0.0824422724545002, "time": 0.836016058921814, "epoch": 4, "memory": 6319, "step": 12521} +{"lr": 7.875184945043487e-05, "data_time": 0.601146149635315, "loss": 0.09197330102324486, "time": 0.8108680963516235, "epoch": 4, "memory": 6319, "step": 12621} +{"lr": 7.844375189082012e-05, "data_time": 1.0703449726104737, "loss": 0.08748614117503166, "time": 1.2102807760238647, "epoch": 4, "memory": 6319, "step": 12721} +{"lr": 7.813413854587572e-05, "data_time": 0.5996574878692627, "loss": 0.09513286128640175, "time": 0.7626878976821899, "epoch": 4, "memory": 6319, "step": 12821} +{"lr": 7.782302943404291e-05, "data_time": 0.7128910541534423, "loss": 0.07880107462406158, "time": 0.8529023885726928, "epoch": 4, "memory": 6319, "step": 12921} +{"lr": 7.751044467047337e-05, "data_time": 0.6366375684738159, "loss": 0.09039798676967621, "time": 0.7779595136642456, "epoch": 4, "memory": 6319, "step": 13021} +{"lr": 7.719640446572907e-05, "data_time": 0.7182265520095825, "loss": 0.08743033334612846, "time": 0.8584444522857666, "epoch": 4, "memory": 6319, "step": 13121} +{"lr": 7.688092912447536e-05, "data_time": 0.6953620672225952, "loss": 0.09112714752554893, "time": 0.8351022481918335, "epoch": 4, "memory": 6319, "step": 13221} +{"lr": 7.656403904416794e-05, "data_time": 0.6501647233963013, "loss": 0.08233190029859543, "time": 0.7912294149398804, "epoch": 4, "memory": 6319, "step": 13321} +{"lr": 7.624575471373434e-05, "data_time": 0.6700344800949096, "loss": 0.08806102760136128, "time": 0.80925452709198, "epoch": 4, "memory": 6319, "step": 13421} +{"lr": 7.592609671224926e-05, "data_time": 0.7183692455291748, "loss": 0.1021886222064495, "time": 0.8603218078613282, "epoch": 4, "memory": 6319, "step": 13521} +{"lr": 7.560508570760375e-05, "data_time": 0.7553340673446656, "loss": 0.07378659248352051, "time": 0.8955759048461914, "epoch": 4, "memory": 6319, "step": 13621} +{"lr": 7.528274245516895e-05, "data_time": 0.7363073825836182, "loss": 0.0893584880977869, "time": 0.8787969350814819, "epoch": 4, "memory": 6319, "step": 13721} +{"lr": 7.495908779645405e-05, "data_time": 0.7267844438552856, "loss": 0.09580430649220943, "time": 0.8677319049835205, "epoch": 4, "memory": 6319, "step": 13821} +{"lr": 7.463414265775906e-05, "data_time": 0.706990122795105, "loss": 0.07591142952442169, "time": 0.8483651638031006, "epoch": 4, "memory": 6319, "step": 13921} +{"lr": 7.430792804882147e-05, "data_time": 0.7727843761444092, "loss": 0.08486550599336624, "time": 0.9133606195449829, "epoch": 4, "memory": 6319, "step": 14021} +{"lr": 7.398046506145803e-05, "data_time": 0.6785132646560669, "loss": 0.08922588638961315, "time": 0.8246493339538574, "epoch": 4, "memory": 6319, "step": 14121} +{"lr": 7.365177486820114e-05, "data_time": 0.7385735511779785, "loss": 0.08043766170740127, "time": 0.8777874946594239, "epoch": 4, "memory": 6319, "step": 14221} +{"lr": 7.332187872092939e-05, "data_time": 0.7279436588287354, "loss": 0.07390545904636384, "time": 0.8677479743957519, "epoch": 4, "memory": 6319, "step": 14321} +{"lr": 7.299079794949401e-05, "data_time": 0.7300907373428345, "loss": 0.07530647069215775, "time": 0.8693580865859986, "epoch": 4, "memory": 6319, "step": 14421} +{"lr": 7.265855396033955e-05, "data_time": 0.6853252649307251, "loss": 0.08673914894461632, "time": 0.8242392778396607, "epoch": 4, "memory": 6319, "step": 14521} +{"lr": 7.232516823512013e-05, "data_time": 0.7513039588928223, "loss": 0.09135455749928952, "time": 0.8967481136322022, "epoch": 4, "memory": 6319, "step": 14621} +{"lr": 7.199066232930995e-05, "data_time": 0.7023496389389038, "loss": 0.10166266560554504, "time": 0.8412851095199585, "epoch": 4, "memory": 6319, "step": 14721} +{"lr": 7.165505787080997e-05, "data_time": 0.756019401550293, "loss": 0.07684456519782543, "time": 0.8975329875946045, "epoch": 4, "memory": 6319, "step": 14821} +{"lr": 7.13183765585495e-05, "data_time": 0.7164599180221558, "loss": 0.08118878044188023, "time": 0.8561276435852051, "epoch": 4, "memory": 6319, "step": 14921} +{"lr": 7.098064016108317e-05, "data_time": 0.7158091068267822, "loss": 0.07695223167538642, "time": 0.8588071823120117, "epoch": 4, "memory": 6319, "step": 15021} +{"lr": 7.064187051518379e-05, "data_time": 0.7051473140716553, "loss": 0.0883091926574707, "time": 0.845363450050354, "epoch": 4, "memory": 6319, "step": 15121} +{"lr": 7.030208952442949e-05, "data_time": 0.6706280946731568, "loss": 0.0866649679839611, "time": 0.8114208936691284, "epoch": 4, "memory": 6319, "step": 15221} +{"lr": 6.996131915778851e-05, "data_time": 0.7403595924377442, "loss": 0.08271320089697838, "time": 0.881003999710083, "epoch": 4, "memory": 6319, "step": 15321} +{"lr": 6.961958144819853e-05, "data_time": 0.6520731925964356, "loss": 0.09721692129969597, "time": 0.792563533782959, "epoch": 4, "memory": 6319, "step": 15421} +{"lr": 6.927689849114193e-05, "data_time": 0.6674933671951294, "loss": 0.08665039278566837, "time": 0.8082375526428223, "epoch": 4, "memory": 6319, "step": 15521} +{"lr": 6.893329244321722e-05, "data_time": 0.6737451553344727, "loss": 0.08945233225822449, "time": 0.8147722959518433, "epoch": 4, "memory": 6319, "step": 15621} +{"accuracy/top1": 94.34954071044922, "data_time": 0.5083364042742499, "time": 0.5965004172818414, "step": 4} +{"lr": 6.856463687361792e-05, "data_time": 0.41266841888427735, "loss": 0.09699749648571014, "time": 0.7549313068389892, "epoch": 5, "memory": 6319, "step": 15728} +{"lr": 6.821919068470432e-05, "data_time": 0.38207876682281494, "loss": 0.07658592164516449, "time": 0.827199149131775, "epoch": 5, "memory": 6319, "step": 15828} +{"lr": 6.787288979235407e-05, "data_time": 0.5177194356918335, "loss": 0.08155790939927102, "time": 0.8222105503082275, "epoch": 5, "memory": 6319, "step": 15928} +{"lr": 6.752575658708793e-05, "data_time": 0.7213511943817139, "loss": 0.07354740388691425, "time": 0.867519760131836, "epoch": 5, "memory": 6319, "step": 16028} +{"lr": 6.71778135132408e-05, "data_time": 0.5564947366714478, "loss": 0.07551858536899089, "time": 0.8823280811309815, "epoch": 5, "memory": 6319, "step": 16128} +{"lr": 6.682908306751074e-05, "data_time": 0.3923478126525879, "loss": 0.08346405848860741, "time": 0.8252520084381103, "epoch": 5, "memory": 6319, "step": 16228} +{"lr": 6.64795877975043e-05, "data_time": 0.19153399467468263, "loss": 0.08557555750012398, "time": 0.8386387825012207, "epoch": 5, "memory": 6319, "step": 16328} +{"lr": 6.61293503002787e-05, "data_time": 0.2131415367126465, "loss": 0.08024964146316052, "time": 0.8301738500595093, "epoch": 5, "memory": 6319, "step": 16428} +{"lr": 6.577839322088084e-05, "data_time": 0.02249758243560791, "loss": 0.07721111588180066, "time": 0.8165907144546509, "epoch": 5, "memory": 6319, "step": 16528} +{"lr": 6.542673925088298e-05, "data_time": 0.0009778738021850586, "loss": 0.09270984455943107, "time": 1.13065664768219, "epoch": 5, "memory": 6319, "step": 16628} +{"lr": 6.507441112691584e-05, "data_time": 0.0008757114410400391, "loss": 0.08181111589074135, "time": 0.7900747060775757, "epoch": 5, "memory": 6319, "step": 16728} +{"lr": 6.47214316291983e-05, "data_time": 0.0009041309356689453, "loss": 0.07667498663067818, "time": 0.8298603534698487, "epoch": 5, "memory": 6319, "step": 16828} +{"lr": 6.436782358006481e-05, "data_time": 0.05229659080505371, "loss": 0.07713332884013653, "time": 0.8582651853561402, "epoch": 5, "memory": 6319, "step": 16928} +{"lr": 6.40136098424896e-05, "data_time": 0.12185480594635009, "loss": 0.0840992011129856, "time": 0.8422068119049072, "epoch": 5, "memory": 6319, "step": 17028} +{"lr": 6.365881331860828e-05, "data_time": 0.20846986770629883, "loss": 0.08576505780220031, "time": 0.898157286643982, "epoch": 5, "memory": 6319, "step": 17128} +{"lr": 6.330345694823739e-05, "data_time": 0.18410565853118896, "loss": 0.07903642058372498, "time": 0.8957612991333008, "epoch": 5, "memory": 6319, "step": 17228} +{"lr": 6.29475637073908e-05, "data_time": 0.44593658447265627, "loss": 0.07665348090231419, "time": 0.8508060693740844, "epoch": 5, "memory": 6319, "step": 17328} +{"lr": 6.259115660679482e-05, "data_time": 0.2975869655609131, "loss": 0.08075075931847095, "time": 0.8303794145584107, "epoch": 5, "memory": 6319, "step": 17428} +{"lr": 6.223425869039963e-05, "data_time": 0.43838350772857665, "loss": 0.07233618497848511, "time": 1.0674143075942992, "epoch": 5, "memory": 6319, "step": 17528} +{"lr": 6.187689303389006e-05, "data_time": 0.27185142040252686, "loss": 0.0743786882609129, "time": 0.7913447141647338, "epoch": 5, "memory": 6319, "step": 17628} +{"lr": 6.151908274319292e-05, "data_time": 0.3729255676269531, "loss": 0.07758879140019417, "time": 0.8297183990478516, "epoch": 5, "memory": 6319, "step": 17728} +{"lr": 6.116085095298373e-05, "data_time": 0.4120022773742676, "loss": 0.056010492518544196, "time": 0.8959392309188843, "epoch": 5, "memory": 6319, "step": 17828} +{"lr": 6.080222082519007e-05, "data_time": 0.7243278741836547, "loss": 0.07328197248280048, "time": 0.8639042139053345, "epoch": 5, "memory": 6319, "step": 17928} +{"lr": 6.0443215547495147e-05, "data_time": 0.7258928298950196, "loss": 0.052864911034703255, "time": 0.8660146474838257, "epoch": 5, "memory": 6319, "step": 18028} +{"lr": 6.0083858331837714e-05, "data_time": 0.7116958379745484, "loss": 0.06759368702769279, "time": 0.8520001888275146, "epoch": 5, "memory": 6319, "step": 18128} +{"lr": 5.97241724129116e-05, "data_time": 0.7235569953918457, "loss": 0.07290937937796116, "time": 0.8631746053695679, "epoch": 5, "memory": 6319, "step": 18228} +{"lr": 5.936418104666328e-05, "data_time": 0.7136447191238403, "loss": 0.07604317106306553, "time": 0.8523468017578125, "epoch": 5, "memory": 6319, "step": 18328} +{"lr": 5.9003907508788295e-05, "data_time": 0.6828764200210571, "loss": 0.06759182997047901, "time": 0.8225975036621094, "epoch": 5, "memory": 6319, "step": 18428} +{"lr": 5.864337509322633e-05, "data_time": 0.7022364139556885, "loss": 0.07443410381674767, "time": 0.8418588638305664, "epoch": 5, "memory": 6319, "step": 18528} +{"lr": 5.8282607110655194e-05, "data_time": 0.7163833618164063, "loss": 0.0823317289352417, "time": 0.8567037582397461, "epoch": 5, "memory": 6319, "step": 18628} +{"lr": 5.7921626886983564e-05, "data_time": 0.978909707069397, "loss": 0.06326876506209374, "time": 1.117817187309265, "epoch": 5, "memory": 6319, "step": 18728} +{"lr": 5.756045776184287e-05, "data_time": 0.5999541521072388, "loss": 0.08691534772515297, "time": 0.739604115486145, "epoch": 5, "memory": 6319, "step": 18828} +{"lr": 5.719912308707808e-05, "data_time": 0.6714261531829834, "loss": 0.08451641723513603, "time": 0.8109571218490601, "epoch": 5, "memory": 6319, "step": 18928} +{"lr": 5.683764622523808e-05, "data_time": 0.6545212030410766, "loss": 0.07043433301150799, "time": 0.7934217929840088, "epoch": 5, "memory": 6319, "step": 19028} +{"lr": 5.647605054806502e-05, "data_time": 0.6724504709243775, "loss": 0.07509004473686218, "time": 0.812038779258728, "epoch": 5, "memory": 6319, "step": 19128} +{"lr": 5.6114359434983206e-05, "data_time": 0.6955249071121216, "loss": 0.07620588913559914, "time": 0.8364404439926147, "epoch": 5, "memory": 6319, "step": 19228} +{"lr": 5.575259627158745e-05, "data_time": 0.7081258535385132, "loss": 0.0768796220421791, "time": 0.8482952117919922, "epoch": 5, "memory": 6319, "step": 19328} +{"lr": 5.539078444813113e-05, "data_time": 0.6759809494018555, "loss": 0.06893887259066105, "time": 0.8153200626373291, "epoch": 5, "memory": 6319, "step": 19428} +{"lr": 5.5028947358013735e-05, "data_time": 0.6872256517410278, "loss": 0.07342792190611362, "time": 0.826401662826538, "epoch": 5, "memory": 6319, "step": 19528} +{"accuracy/top1": 95.75519561767578, "data_time": 0.49287892210072487, "time": 0.579543948173523, "step": 5} +{"lr": 5.464178020340108e-05, "data_time": 0.6675108909606934, "loss": 0.07222603149712085, "time": 0.8089387655258179, "epoch": 6, "memory": 6319, "step": 19635} +{"lr": 5.427996514795046e-05, "data_time": 0.6952463626861572, "loss": 0.08180357217788696, "time": 0.8361425638198853, "epoch": 6, "memory": 6319, "step": 19735} +{"lr": 5.3918196647260834e-05, "data_time": 0.7093801736831665, "loss": 0.07309752069413662, "time": 0.8487287282943725, "epoch": 6, "memory": 6319, "step": 19835} +{"lr": 5.355649809193057e-05, "data_time": 0.6975620985031128, "loss": 0.06639642640948296, "time": 0.8372887372970581, "epoch": 6, "memory": 6319, "step": 19935} +{"lr": 5.3194892868035765e-05, "data_time": 1.0439325332641602, "loss": 0.0747215911746025, "time": 1.1844680786132813, "epoch": 6, "memory": 6319, "step": 20035} +{"lr": 5.283340435561795e-05, "data_time": 0.1093963623046875, "loss": 0.06897138692438602, "time": 0.7011008501052857, "epoch": 6, "memory": 6319, "step": 20135} +{"lr": 5.2472055927172495e-05, "data_time": 0.04112601280212402, "loss": 0.09206668809056281, "time": 0.745328688621521, "epoch": 6, "memory": 6319, "step": 20235} +{"lr": 5.2110870946137605e-05, "data_time": 0.15963370800018312, "loss": 0.07574363686144352, "time": 0.8558576822280883, "epoch": 6, "memory": 6319, "step": 20335} +{"lr": 5.174987276538342e-05, "data_time": 0.1250157594680786, "loss": 0.07149523869156837, "time": 0.7836745977401733, "epoch": 6, "memory": 6319, "step": 20435} +{"lr": 5.1389084725702425e-05, "data_time": 0.09073824882507324, "loss": 0.057606670632958414, "time": 0.843821382522583, "epoch": 6, "memory": 6319, "step": 20535} +{"lr": 5.1028530154300045e-05, "data_time": 0.23503546714782714, "loss": 0.07742604538798332, "time": 0.8392188310623169, "epoch": 6, "memory": 6319, "step": 20635} +{"lr": 5.066823236328663e-05, "data_time": 0.23944566249847413, "loss": 0.0693360522389412, "time": 0.8808053970336914, "epoch": 6, "memory": 6319, "step": 20735} +{"lr": 5.030821464817006e-05, "data_time": 0.1433509111404419, "loss": 0.07649844773113727, "time": 0.8061399936676026, "epoch": 6, "memory": 6319, "step": 20835} +{"lr": 4.99485002863494e-05, "data_time": 0.5709505558013916, "loss": 0.08023917004466057, "time": 0.885708212852478, "epoch": 6, "memory": 6319, "step": 20935} +{"lr": 4.95891125356102e-05, "data_time": 0.661069107055664, "loss": 0.06758230775594712, "time": 0.8002021789550782, "epoch": 6, "memory": 6319, "step": 21035} +{"lr": 4.923007463262053e-05, "data_time": 0.6734744310379028, "loss": 0.0738202977925539, "time": 0.8127961158752441, "epoch": 6, "memory": 6319, "step": 21135} +{"lr": 4.887140979142852e-05, "data_time": 0.6755631923675537, "loss": 0.06861904226243495, "time": 0.8156698703765869, "epoch": 6, "memory": 6319, "step": 21235} +{"lr": 4.851314120196161e-05, "data_time": 0.676516842842102, "loss": 0.06585529148578644, "time": 0.8166579246520996, "epoch": 6, "memory": 6319, "step": 21335} +{"lr": 4.815529202852693e-05, "data_time": 0.6736643314361572, "loss": 0.07461902275681495, "time": 0.8135789155960083, "epoch": 6, "memory": 6319, "step": 21435} +{"lr": 4.779788540831406e-05, "data_time": 0.6853596448898316, "loss": 0.06019394584000111, "time": 0.8304765939712524, "epoch": 6, "memory": 6319, "step": 21535} +{"lr": 4.744094444989843e-05, "data_time": 0.7251887798309327, "loss": 0.0702584121376276, "time": 0.864327335357666, "epoch": 6, "memory": 6319, "step": 21635} +{"lr": 4.708449223174764e-05, "data_time": 0.6773305177688599, "loss": 0.06284310463815927, "time": 0.8168601274490357, "epoch": 6, "memory": 6319, "step": 21735} +{"lr": 4.672855180072916e-05, "data_time": 0.6327378749847412, "loss": 0.07013889662921428, "time": 0.7714540719985962, "epoch": 6, "memory": 6319, "step": 21835} +{"lr": 4.6373146170620256e-05, "data_time": 0.6743086338043213, "loss": 0.06573177762329578, "time": 0.8264220476150512, "epoch": 6, "memory": 6319, "step": 21935} +{"lr": 4.6018298320619915e-05, "data_time": 0.6912712335586548, "loss": 0.0773338109254837, "time": 0.8312756061553955, "epoch": 6, "memory": 6319, "step": 22035} +{"lr": 4.566403119386318e-05, "data_time": 0.7341667413711548, "loss": 0.07290297895669937, "time": 0.8750342130661011, "epoch": 6, "memory": 6319, "step": 22135} +{"lr": 4.531036769593769e-05, "data_time": 0.7128526210784912, "loss": 0.0873205129057169, "time": 0.8532657623291016, "epoch": 6, "memory": 6319, "step": 22235} +{"lr": 4.4957330693402685e-05, "data_time": 0.7100772619247436, "loss": 0.07504116408526898, "time": 0.8568342447280883, "epoch": 6, "memory": 6319, "step": 22335} +{"lr": 4.460494301231059e-05, "data_time": 0.677058744430542, "loss": 0.07817576602101325, "time": 0.8295367240905762, "epoch": 6, "memory": 6319, "step": 22435} +{"lr": 4.425322743673109e-05, "data_time": 0.7131170272827149, "loss": 0.07172847464680672, "time": 0.8532304048538208, "epoch": 6, "memory": 6319, "step": 22535} +{"lr": 4.390220670727812e-05, "data_time": 0.7379009485244751, "loss": 0.07035707458853721, "time": 0.8811063289642334, "epoch": 6, "memory": 6319, "step": 22635} +{"lr": 4.3551903519639325e-05, "data_time": 0.6639108419418335, "loss": 0.06795312836766243, "time": 0.802966833114624, "epoch": 6, "memory": 6319, "step": 22735} +{"lr": 4.3202340523108894e-05, "data_time": 0.6869333028793335, "loss": 0.07074483633041381, "time": 0.8265269279479981, "epoch": 6, "memory": 6319, "step": 22835} +{"lr": 4.285354031912302e-05, "data_time": 0.6874081373214722, "loss": 0.07929525859653949, "time": 0.8276089668273926, "epoch": 6, "memory": 6319, "step": 22935} +{"lr": 4.2505525459798544e-05, "data_time": 0.6330511808395386, "loss": 0.07162248343229294, "time": 0.7719424724578857, "epoch": 6, "memory": 6319, "step": 23035} +{"lr": 4.2158318446474836e-05, "data_time": 0.6937422275543212, "loss": 0.0946463007479906, "time": 0.833633017539978, "epoch": 6, "memory": 6319, "step": 23135} +{"lr": 4.181194172825905e-05, "data_time": 0.6966533184051513, "loss": 0.07296304292976856, "time": 0.8356399297714233, "epoch": 6, "memory": 6319, "step": 23235} +{"lr": 4.146641770057448e-05, "data_time": 0.6928616046905518, "loss": 0.06685053631663322, "time": 0.8336604118347168, "epoch": 6, "memory": 6319, "step": 23335} +{"lr": 4.112176870371267e-05, "data_time": 0.6637640476226807, "loss": 0.0628920566290617, "time": 0.8036967992782593, "epoch": 6, "memory": 6319, "step": 23435} +{"accuracy/top1": 96.25273895263672, "data_time": 0.47652052599808264, "time": 0.5657904353635065, "step": 6} +{"lr": 4.0753988582732786e-05, "data_time": 0.6781873226165771, "loss": 0.06858180090785027, "time": 0.8169117212295532, "epoch": 7, "memory": 6319, "step": 23542} +{"lr": 4.0411221639286815e-05, "data_time": 0.694000506401062, "loss": 0.05826659053564072, "time": 0.8341515302658081, "epoch": 7, "memory": 6319, "step": 23642} +{"lr": 4.006939795169518e-05, "data_time": 0.6736550807952881, "loss": 0.05747688505798578, "time": 0.8711623430252076, "epoch": 7, "memory": 6319, "step": 23742} +{"lr": 3.972853962099913e-05, "data_time": 0.6281713247299194, "loss": 0.0650223795324564, "time": 0.8164336442947387, "epoch": 7, "memory": 6319, "step": 23842} +{"lr": 3.93886686858237e-05, "data_time": 0.688922119140625, "loss": 0.0617504931986332, "time": 0.8274576425552368, "epoch": 7, "memory": 6319, "step": 23942} +{"lr": 3.904980712095259e-05, "data_time": 0.5287313461303711, "loss": 0.0588339526206255, "time": 0.8018426895141602, "epoch": 7, "memory": 6319, "step": 24042} +{"lr": 3.871197683590748e-05, "data_time": 0.49728686809539796, "loss": 0.0567039743065834, "time": 0.8472254037857055, "epoch": 7, "memory": 6319, "step": 24142} +{"lr": 3.837519967353125e-05, "data_time": 0.4354600191116333, "loss": 0.06448316983878613, "time": 0.781667685508728, "epoch": 7, "memory": 6319, "step": 24242} +{"lr": 3.8039497408575855e-05, "data_time": 0.4605488538742065, "loss": 0.06413083449006081, "time": 0.8372302055358887, "epoch": 7, "memory": 6319, "step": 24342} +{"lr": 3.770489174629432e-05, "data_time": 0.60411536693573, "loss": 0.06517446022480726, "time": 0.8847229957580567, "epoch": 7, "memory": 6319, "step": 24442} +{"lr": 3.737140432103754e-05, "data_time": 0.7392241954803467, "loss": 0.06586945131421089, "time": 0.8795247316360474, "epoch": 7, "memory": 6319, "step": 24542} +{"lr": 3.703905669485545e-05, "data_time": 0.7529799222946167, "loss": 0.06417501755058766, "time": 0.891901183128357, "epoch": 7, "memory": 6319, "step": 24642} +{"lr": 3.6707870356102706e-05, "data_time": 0.6683059692382812, "loss": 0.0770888464525342, "time": 0.8078370571136475, "epoch": 7, "memory": 6319, "step": 24742} +{"lr": 3.6377866718049484e-05, "data_time": 0.7030785560607911, "loss": 0.07405411973595619, "time": 0.8424556970596313, "epoch": 7, "memory": 6319, "step": 24842} +{"lr": 3.6049067117496965e-05, "data_time": 0.6847036361694336, "loss": 0.06737223193049431, "time": 0.8235409975051879, "epoch": 7, "memory": 6319, "step": 24942} +{"lr": 3.5721492813397755e-05, "data_time": 0.6364476919174195, "loss": 0.05018443632870913, "time": 0.775453519821167, "epoch": 7, "memory": 6319, "step": 25042} +{"lr": 3.539516498548151e-05, "data_time": 0.6694787502288818, "loss": 0.06519287563860417, "time": 0.8079834938049316, "epoch": 7, "memory": 6319, "step": 25142} +{"lr": 3.507010473288518e-05, "data_time": 0.7814359188079834, "loss": 0.06257222071290017, "time": 0.9209163904190063, "epoch": 7, "memory": 6319, "step": 25242} +{"lr": 3.474633307278909e-05, "data_time": 0.7358900308609009, "loss": 0.07201835811138153, "time": 0.884191370010376, "epoch": 7, "memory": 6319, "step": 25342} +{"lr": 3.4423870939058177e-05, "data_time": 0.6787420034408569, "loss": 0.07409783266484737, "time": 0.8189467430114746, "epoch": 7, "memory": 6319, "step": 25442} +{"lr": 3.410273918088805e-05, "data_time": 0.6938150644302368, "loss": 0.062348130345344546, "time": 0.8333451747894287, "epoch": 7, "memory": 6319, "step": 25542} +{"lr": 3.37829585614574e-05, "data_time": 0.7013892650604248, "loss": 0.06754357106983662, "time": 0.8399263620376587, "epoch": 7, "memory": 6319, "step": 25642} +{"lr": 3.3464549756585216e-05, "data_time": 0.6721920967102051, "loss": 0.07165113016963005, "time": 0.8337712049484253, "epoch": 7, "memory": 6319, "step": 25742} +{"lr": 3.314753335339416e-05, "data_time": 0.6872803688049316, "loss": 0.07331808060407638, "time": 0.8269120216369629, "epoch": 7, "memory": 6319, "step": 25842} +{"lr": 3.2831929848979394e-05, "data_time": 0.6770183324813843, "loss": 0.06544382236897946, "time": 0.8303147554397583, "epoch": 7, "memory": 6319, "step": 25942} +{"lr": 3.251775964908332e-05, "data_time": 0.6336238384246826, "loss": 0.06489804610610009, "time": 0.7735423803329468, "epoch": 7, "memory": 6319, "step": 26042} +{"lr": 3.22050430667762e-05, "data_time": 0.6411094188690185, "loss": 0.08467503376305104, "time": 0.7807039260864258, "epoch": 7, "memory": 6319, "step": 26142} +{"lr": 3.189380032114285e-05, "data_time": 0.660017728805542, "loss": 0.05466986652463675, "time": 0.7994655132293701, "epoch": 7, "memory": 6319, "step": 26242} +{"lr": 3.1584051535975324e-05, "data_time": 0.7323796272277832, "loss": 0.06272181272506713, "time": 0.8717444896697998, "epoch": 7, "memory": 6319, "step": 26342} +{"lr": 3.127581673847175e-05, "data_time": 0.6505621433258056, "loss": 0.06274190135300159, "time": 0.7922843217849731, "epoch": 7, "memory": 6319, "step": 26442} +{"lr": 3.0969115857941555e-05, "data_time": 0.7728345632553101, "loss": 0.07314100936055183, "time": 0.9122422695159912, "epoch": 7, "memory": 6319, "step": 26542} +{"lr": 3.0663968724516654e-05, "data_time": 0.6929701805114746, "loss": 0.06871767975389957, "time": 0.8341165542602539, "epoch": 7, "memory": 6319, "step": 26642} +{"lr": 3.0360395067869668e-05, "data_time": 0.6636749505996704, "loss": 0.06069712303578854, "time": 0.8030887365341186, "epoch": 7, "memory": 6319, "step": 26742} +{"lr": 3.0058414515938063e-05, "data_time": 0.6584934234619141, "loss": 0.0703870963305235, "time": 0.7987941265106201, "epoch": 7, "memory": 6319, "step": 26842} +{"lr": 2.975804659365501e-05, "data_time": 0.6954042434692382, "loss": 0.06219286173582077, "time": 0.8350435972213746, "epoch": 7, "memory": 6319, "step": 26942} +{"lr": 2.945931072168714e-05, "data_time": 0.7897649526596069, "loss": 0.06268918849527835, "time": 0.9300974845886231, "epoch": 7, "memory": 6319, "step": 27042} +{"lr": 2.9162226215178855e-05, "data_time": 0.6892393350601196, "loss": 0.061595524102449416, "time": 0.8299695491790772, "epoch": 7, "memory": 6319, "step": 27142} +{"lr": 2.8866812282503322e-05, "data_time": 0.6449941635131836, "loss": 0.05605282727628946, "time": 0.784614872932434, "epoch": 7, "memory": 6319, "step": 27242} +{"lr": 2.8573088024020838e-05, "data_time": 0.7308354377746582, "loss": 0.06938082985579967, "time": 0.8708070516586304, "epoch": 7, "memory": 6319, "step": 27342} +{"accuracy/top1": 96.99036407470703, "data_time": 0.48901195772762956, "time": 0.5752114304180803, "step": 7} +{"lr": 2.8260695817390092e-05, "data_time": 0.5400987148284913, "loss": 0.06039245910942555, "time": 0.7998890399932861, "epoch": 8, "memory": 6319, "step": 27449} +{"lr": 2.7970529402570967e-05, "data_time": 0.4197249889373779, "loss": 0.059700028225779536, "time": 0.8246399164199829, "epoch": 8, "memory": 6319, "step": 27549} +{"lr": 2.7682110612244214e-05, "data_time": 0.3895702362060547, "loss": 0.060332028195261955, "time": 0.838551926612854, "epoch": 8, "memory": 6319, "step": 27649} +{"lr": 2.7395458094490283e-05, "data_time": 0.258343768119812, "loss": 0.06834733560681343, "time": 0.8738335609436035, "epoch": 8, "memory": 6319, "step": 27749} +{"lr": 2.7110590383189015e-05, "data_time": 0.06507272720336914, "loss": 0.06124198362231255, "time": 0.8238757371902465, "epoch": 8, "memory": 6319, "step": 27849} +{"lr": 2.6827525896821397e-05, "data_time": 0.0008154630661010743, "loss": 0.06720116976648569, "time": 0.8146600723266602, "epoch": 8, "memory": 6319, "step": 27949} +{"lr": 2.6546282937278552e-05, "data_time": 0.000783848762512207, "loss": 0.06980157122015954, "time": 0.8189395427703857, "epoch": 8, "memory": 6319, "step": 28049} +{"lr": 2.6266879688678606e-05, "data_time": 0.0009537220001220703, "loss": 0.06123245526105166, "time": 0.890581202507019, "epoch": 8, "memory": 6319, "step": 28149} +{"lr": 2.5989334216190738e-05, "data_time": 0.0009265899658203125, "loss": 0.06326903328299523, "time": 0.8043756723403931, "epoch": 8, "memory": 6319, "step": 28249} +{"lr": 2.571366446486732e-05, "data_time": 0.0008894205093383789, "loss": 0.07336103729903698, "time": 0.7781476259231568, "epoch": 8, "memory": 6319, "step": 28349} +{"lr": 2.5439888258483665e-05, "data_time": 0.0011068820953369141, "loss": 0.06161957867443561, "time": 0.8684613943099976, "epoch": 8, "memory": 6319, "step": 28449} +{"lr": 2.5168023298385384e-05, "data_time": 0.000771021842956543, "loss": 0.059670351445674896, "time": 0.8432024955749512, "epoch": 8, "memory": 6319, "step": 28549} +{"lr": 2.489808716234412e-05, "data_time": 0.0011826753616333008, "loss": 0.06030804291367531, "time": 0.837065315246582, "epoch": 8, "memory": 6319, "step": 28649} +{"lr": 2.463009730342111e-05, "data_time": 0.001067519187927246, "loss": 0.07112121991813183, "time": 0.8273203372955322, "epoch": 8, "memory": 6319, "step": 28749} +{"lr": 2.436407104883849e-05, "data_time": 0.0007772684097290039, "loss": 0.06335515901446342, "time": 0.8596891164779663, "epoch": 8, "memory": 6319, "step": 28849} +{"lr": 2.4100025598859092e-05, "data_time": 0.0008195638656616211, "loss": 0.0655285321176052, "time": 0.8000187158584595, "epoch": 8, "memory": 6319, "step": 28949} +{"lr": 2.3837978025674284e-05, "data_time": 0.0007653236389160156, "loss": 0.06330592222511769, "time": 0.7643567562103272, "epoch": 8, "memory": 6319, "step": 29049} +{"lr": 2.357794527230027e-05, "data_time": 0.14439218044281005, "loss": 0.06396490409970283, "time": 0.8001121520996094, "epoch": 8, "memory": 6319, "step": 29149} +{"lr": 2.33199441514825e-05, "data_time": 0.4546696662902832, "loss": 0.06362710930407048, "time": 0.805083966255188, "epoch": 8, "memory": 6319, "step": 29249} +{"lr": 2.306399134460859e-05, "data_time": 0.42298116683959963, "loss": 0.05732900165021419, "time": 0.8389808416366578, "epoch": 8, "memory": 6319, "step": 29349} +{"lr": 2.281010340062995e-05, "data_time": 0.3388469934463501, "loss": 0.06541037410497666, "time": 0.8745235919952392, "epoch": 8, "memory": 6319, "step": 29449} +{"lr": 2.2558296734991603e-05, "data_time": 0.532512092590332, "loss": 0.06803475804626942, "time": 0.7765348672866821, "epoch": 8, "memory": 6319, "step": 29549} +{"lr": 2.2308587628570915e-05, "data_time": 0.5307520866394043, "loss": 0.06364604346454143, "time": 0.7839691162109375, "epoch": 8, "memory": 6319, "step": 29649} +{"lr": 2.2060992226624955e-05, "data_time": 0.5516899108886719, "loss": 0.05329091902822256, "time": 0.7754719018936157, "epoch": 8, "memory": 6319, "step": 29749} +{"lr": 2.181552653774648e-05, "data_time": 0.6716019153594971, "loss": 0.0647040979936719, "time": 0.8121907949447632, "epoch": 8, "memory": 6319, "step": 29849} +{"lr": 2.1572206432829035e-05, "data_time": 0.6784110546112061, "loss": 0.07328018620610237, "time": 0.8185736894607544, "epoch": 8, "memory": 6319, "step": 29949} +{"lr": 2.1331047644040754e-05, "data_time": 0.6243602991104126, "loss": 0.06692851856350898, "time": 0.7644826650619507, "epoch": 8, "memory": 6319, "step": 30049} +{"lr": 2.109206576380705e-05, "data_time": 0.677127480506897, "loss": 0.054996407777071, "time": 0.8164906740188599, "epoch": 8, "memory": 6319, "step": 30149} +{"lr": 2.0855276243802687e-05, "data_time": 0.6834668159484864, "loss": 0.062416527047753335, "time": 0.8316063165664673, "epoch": 8, "memory": 6319, "step": 30249} +{"lr": 2.0620694393952487e-05, "data_time": 0.6708291292190551, "loss": 0.0638434112071991, "time": 0.8102986574172973, "epoch": 8, "memory": 6319, "step": 30349} +{"lr": 2.038833538144169e-05, "data_time": 0.6446569204330445, "loss": 0.07554081343114376, "time": 0.7898784875869751, "epoch": 8, "memory": 6319, "step": 30449} +{"lr": 2.0158214229735154e-05, "data_time": 0.6944183826446533, "loss": 0.06999053806066513, "time": 0.8337982892990112, "epoch": 8, "memory": 6319, "step": 30549} +{"lr": 1.9930345817605977e-05, "data_time": 0.6702313899993897, "loss": 0.0658162672072649, "time": 0.8093949079513549, "epoch": 8, "memory": 6319, "step": 30649} +{"lr": 1.970474487817358e-05, "data_time": 0.7079986810684205, "loss": 0.06648847609758377, "time": 0.8471720457077027, "epoch": 8, "memory": 6319, "step": 30749} +{"lr": 1.9481425997951092e-05, "data_time": 0.706857442855835, "loss": 0.05500877685844898, "time": 0.8573665618896484, "epoch": 8, "memory": 6319, "step": 30849} +{"lr": 1.9260403615902187e-05, "data_time": 0.6302733898162842, "loss": 0.05155815072357654, "time": 0.770255708694458, "epoch": 8, "memory": 6319, "step": 30949} +{"lr": 1.9041692022507647e-05, "data_time": 0.681073784828186, "loss": 0.06943475604057311, "time": 0.8212351322174072, "epoch": 8, "memory": 6319, "step": 31049} +{"lr": 1.8825305358841125e-05, "data_time": 0.6838078260421753, "loss": 0.054085373878479004, "time": 0.8238441944122314, "epoch": 8, "memory": 6319, "step": 31149} +{"lr": 1.861125761565518e-05, "data_time": 0.6597942113876343, "loss": 0.064662741497159, "time": 0.7983960390090943, "epoch": 8, "memory": 6319, "step": 31249} +{"accuracy/top1": 97.20956420898438, "data_time": 0.4779896201758549, "time": 0.5642891300135645, "step": 8} +{"lr": 1.8384832449122864e-05, "data_time": 0.5645036935806275, "loss": 0.07345356605947018, "time": 0.9074468851089478, "epoch": 9, "memory": 6319, "step": 31356} +{"lr": 1.817567007332114e-05, "data_time": 0.12101547718048096, "loss": 0.05780891682952642, "time": 0.8669555187225342, "epoch": 9, "memory": 6319, "step": 31456} +{"lr": 1.796888862098825e-05, "data_time": 0.0013144493103027343, "loss": 0.0513185903429985, "time": 0.8499105453491211, "epoch": 9, "memory": 6319, "step": 31556} +{"lr": 1.7764501461839678e-05, "data_time": 0.0007951259613037109, "loss": 0.06022476479411125, "time": 0.775009560585022, "epoch": 9, "memory": 6319, "step": 31656} +{"lr": 1.7562521810784746e-05, "data_time": 0.12217726707458496, "loss": 0.06544761583209038, "time": 0.815608024597168, "epoch": 9, "memory": 6319, "step": 31756} +{"lr": 1.7362962727072444e-05, "data_time": 0.2080303907394409, "loss": 0.05993554592132568, "time": 0.7603644371032715, "epoch": 9, "memory": 6319, "step": 31856} +{"lr": 1.716583711344682e-05, "data_time": 0.14389641284942628, "loss": 0.06042714975774288, "time": 0.8643301486968994, "epoch": 9, "memory": 6319, "step": 31956} +{"lr": 1.6971157715312954e-05, "data_time": 0.0008918046951293945, "loss": 0.06205844022333622, "time": 0.7845324754714966, "epoch": 9, "memory": 6319, "step": 32056} +{"lr": 1.677893711991274e-05, "data_time": 0.05575065612792969, "loss": 0.057245842181146145, "time": 0.756241750717163, "epoch": 9, "memory": 6319, "step": 32156} +{"lr": 1.6589187755511025e-05, "data_time": 0.1641395330429077, "loss": 0.06889153718948364, "time": 0.7898016214370728, "epoch": 9, "memory": 6319, "step": 32256} +{"lr": 1.640192189059216e-05, "data_time": 0.4034645318984985, "loss": 0.0685161579400301, "time": 0.798612093925476, "epoch": 9, "memory": 6319, "step": 32356} +{"lr": 1.6217151633066716e-05, "data_time": 0.658571720123291, "loss": 0.05839485619217157, "time": 0.8009984731674195, "epoch": 9, "memory": 6319, "step": 32456} +{"lr": 1.6034888929488564e-05, "data_time": 0.6858772277832031, "loss": 0.06413462795317174, "time": 0.8266857862472534, "epoch": 9, "memory": 6319, "step": 32556} +{"lr": 1.585514556428258e-05, "data_time": 0.679267168045044, "loss": 0.05429813750088215, "time": 0.8208481311798096, "epoch": 9, "memory": 6319, "step": 32656} +{"lr": 1.567793315898261e-05, "data_time": 0.6696809291839599, "loss": 0.05694551486521959, "time": 0.8106741189956665, "epoch": 9, "memory": 6319, "step": 32756} +{"lr": 1.5503263171480048e-05, "data_time": 0.6769977807998657, "loss": 0.05947272926568985, "time": 0.8180159568786621, "epoch": 9, "memory": 6319, "step": 32856} +{"lr": 1.5331146895283165e-05, "data_time": 0.6756047487258912, "loss": 0.06210103891789913, "time": 0.8163010120391846, "epoch": 9, "memory": 6319, "step": 32956} +{"lr": 1.5161595458786728e-05, "data_time": 0.6639619588851928, "loss": 0.05722990818321705, "time": 0.8037036657333374, "epoch": 9, "memory": 6319, "step": 33056} +{"lr": 1.4994619824552563e-05, "data_time": 0.7145344018936157, "loss": 0.06845532264560461, "time": 0.8550087213516235, "epoch": 9, "memory": 6319, "step": 33156} +{"lr": 1.4830230788600802e-05, "data_time": 0.6928485870361328, "loss": 0.06998468115925789, "time": 0.8321084260940552, "epoch": 9, "memory": 6319, "step": 33256} +{"lr": 1.4668438979711779e-05, "data_time": 0.6831482648849487, "loss": 0.055490275658667085, "time": 0.8225808382034302, "epoch": 9, "memory": 6319, "step": 33356} +{"lr": 1.4509254858738844e-05, "data_time": 0.6699369430541993, "loss": 0.06110903955996037, "time": 0.809241247177124, "epoch": 9, "memory": 6319, "step": 33456} +{"lr": 1.4352688717931966e-05, "data_time": 0.6460357666015625, "loss": 0.0649946790188551, "time": 0.7856189250946045, "epoch": 9, "memory": 6319, "step": 33556} +{"lr": 1.4198750680272348e-05, "data_time": 0.7298983812332154, "loss": 0.05767113342881203, "time": 0.8703232049942017, "epoch": 9, "memory": 6319, "step": 33656} +{"lr": 1.4047450698817857e-05, "data_time": 0.6806979417800904, "loss": 0.05556385312229395, "time": 0.819658613204956, "epoch": 9, "memory": 6319, "step": 33756} +{"lr": 1.389879855605952e-05, "data_time": 0.665081787109375, "loss": 0.07229854948818684, "time": 0.8048717260360718, "epoch": 9, "memory": 6319, "step": 33856} +{"lr": 1.3752803863288976e-05, "data_time": 0.6799609422683716, "loss": 0.06413264237344266, "time": 0.8191285848617553, "epoch": 9, "memory": 6319, "step": 33956} +{"lr": 1.3609476059977129e-05, "data_time": 0.6685647010803223, "loss": 0.0635818213224411, "time": 0.8143080711364746, "epoch": 9, "memory": 6319, "step": 34056} +{"lr": 1.3468824413163754e-05, "data_time": 0.6434451580047608, "loss": 0.06423680298030376, "time": 0.7831258773803711, "epoch": 9, "memory": 6319, "step": 34156} +{"lr": 1.3330858016858395e-05, "data_time": 0.687971568107605, "loss": 0.06455809995532036, "time": 0.8318761348724365, "epoch": 9, "memory": 6319, "step": 34256} +{"lr": 1.3195585791452309e-05, "data_time": 0.6910084247589111, "loss": 0.0646224495023489, "time": 0.8315199851989746, "epoch": 9, "memory": 6319, "step": 34356} +{"lr": 1.3063016483141729e-05, "data_time": 0.6375117540359497, "loss": 0.059206053614616394, "time": 0.777857494354248, "epoch": 9, "memory": 6319, "step": 34456} +{"lr": 1.2933158663362388e-05, "data_time": 0.6698830366134644, "loss": 0.052413994818925856, "time": 0.810014295578003, "epoch": 9, "memory": 6319, "step": 34556} +{"lr": 1.2806020728235311e-05, "data_time": 0.6686711549758911, "loss": 0.06219264827668667, "time": 0.807573938369751, "epoch": 9, "memory": 6319, "step": 34656} +{"lr": 1.2681610898023963e-05, "data_time": 0.6727989435195922, "loss": 0.05427257437258959, "time": 0.8126304149627686, "epoch": 9, "memory": 6319, "step": 34756} +{"lr": 1.2559937216602731e-05, "data_time": 0.6883304119110107, "loss": 0.05407445915043354, "time": 0.8308426141738892, "epoch": 9, "memory": 6319, "step": 34856} +{"lr": 1.2441007550936847e-05, "data_time": 0.6946065187454223, "loss": 0.08329944387078285, "time": 0.8352630376815796, "epoch": 9, "memory": 6319, "step": 34956} +{"lr": 1.232482959057377e-05, "data_time": 0.7055997610092163, "loss": 0.062410608679056165, "time": 0.8447673320770264, "epoch": 9, "memory": 6319, "step": 35056} +{"lr": 1.2211410847145969e-05, "data_time": 0.6372318029403686, "loss": 0.06384058706462384, "time": 0.7763364553451538, "epoch": 9, "memory": 6319, "step": 35156} +{"accuracy/top1": 97.25131225585938, "data_time": 0.47774931480144633, "time": 0.5636117006170338, "step": 9} +{"lr": 1.209311679360931e-05, "data_time": 0.39909961223602297, "loss": 0.05663879550993443, "time": 0.8313610076904296, "epoch": 10, "memory": 6319, "step": 35263} +{"lr": 1.1985432727488666e-05, "data_time": 0.3483341455459595, "loss": 0.06078498288989067, "time": 0.7888491153717041, "epoch": 10, "memory": 6319, "step": 35363} +{"lr": 1.1880529822435289e-05, "data_time": 0.1795570135116577, "loss": 0.05528392493724823, "time": 0.7648987293243408, "epoch": 10, "memory": 6319, "step": 35463} +{"lr": 1.1778414861078726e-05, "data_time": 0.19396381378173827, "loss": 0.06417308263480663, "time": 0.8525235652923584, "epoch": 10, "memory": 6319, "step": 35563} +{"lr": 1.1679094445790538e-05, "data_time": 0.09389820098876953, "loss": 0.06285519935190678, "time": 0.8113935708999633, "epoch": 10, "memory": 6319, "step": 35663} +{"lr": 1.1582574998257377e-05, "data_time": 0.2651108741760254, "loss": 0.06844070069491863, "time": 0.7451489686965942, "epoch": 10, "memory": 6319, "step": 35763} +{"lr": 1.1488862759065762e-05, "data_time": 0.401881742477417, "loss": 0.07080382071435451, "time": 0.7836827039718628, "epoch": 10, "memory": 6319, "step": 35863} +{"lr": 1.1397963787298665e-05, "data_time": 0.6262887716293335, "loss": 0.06453243866562844, "time": 0.8084224462509155, "epoch": 10, "memory": 6319, "step": 35963} +{"lr": 1.1309883960143664e-05, "data_time": 0.6166814804077149, "loss": 0.05892754942178726, "time": 0.8026927947998047, "epoch": 10, "memory": 6319, "step": 36063} +{"lr": 1.122462897251302e-05, "data_time": 0.6859795570373535, "loss": 0.06453697718679904, "time": 0.827719783782959, "epoch": 10, "memory": 6319, "step": 36163} +{"lr": 1.114220433667542e-05, "data_time": 0.7088356971740722, "loss": 0.054554440826177594, "time": 0.8499486446380615, "epoch": 10, "memory": 6319, "step": 36263} +{"lr": 1.10626153818996e-05, "data_time": 0.6648312568664551, "loss": 0.061416537314653394, "time": 0.8116134881973267, "epoch": 10, "memory": 6319, "step": 36363} +{"lr": 1.0985867254109761e-05, "data_time": 0.7136873483657837, "loss": 0.07393808402121067, "time": 0.854666018486023, "epoch": 10, "memory": 6319, "step": 36463} +{"lr": 1.0911964915552823e-05, "data_time": 0.6773239850997925, "loss": 0.05446305610239506, "time": 0.8185667037963867, "epoch": 10, "memory": 6319, "step": 36563} +{"lr": 1.0840913144477643e-05, "data_time": 0.6711352586746215, "loss": 0.06571899875998496, "time": 0.8102775096893311, "epoch": 10, "memory": 6319, "step": 36663} +{"lr": 1.0772716534826064e-05, "data_time": 0.675044584274292, "loss": 0.05931845977902413, "time": 0.8152484178543091, "epoch": 10, "memory": 6319, "step": 36763} +{"lr": 1.0707379495935828e-05, "data_time": 0.7159065485000611, "loss": 0.059493158385157584, "time": 0.8574132919311523, "epoch": 10, "memory": 6319, "step": 36863} +{"lr": 1.0644906252255541e-05, "data_time": 0.6791799306869507, "loss": 0.06316351033747196, "time": 0.8191067218780518, "epoch": 10, "memory": 6319, "step": 36963} +{"lr": 1.0585300843071527e-05, "data_time": 0.6151295185089112, "loss": 0.06075747609138489, "time": 0.7542976856231689, "epoch": 10, "memory": 6319, "step": 37063} +{"lr": 1.0528567122246654e-05, "data_time": 0.6234589576721191, "loss": 0.050756143592298034, "time": 0.7630714893341064, "epoch": 10, "memory": 6319, "step": 37163} +{"lr": 1.0474708757971141e-05, "data_time": 0.6771501779556275, "loss": 0.06674916744232177, "time": 0.8185913801193238, "epoch": 10, "memory": 6319, "step": 37263} +{"lr": 1.042372923252543e-05, "data_time": 0.6750219583511352, "loss": 0.0669028852134943, "time": 0.8143959045410156, "epoch": 10, "memory": 6319, "step": 37363} +{"lr": 1.0375631842054998e-05, "data_time": 0.6899311542510986, "loss": 0.05886037386953831, "time": 0.8302990436553955, "epoch": 10, "memory": 6319, "step": 37463} +{"lr": 1.0330419696357254e-05, "data_time": 0.7050694227218628, "loss": 0.061891605332493785, "time": 0.8458878040313721, "epoch": 10, "memory": 6319, "step": 37563} +{"lr": 1.028809571868047e-05, "data_time": 0.7521899938583374, "loss": 0.06351680010557174, "time": 0.8917283773422241, "epoch": 10, "memory": 6319, "step": 37663} +{"lr": 1.0248662645534766e-05, "data_time": 0.6496395587921142, "loss": 0.058163994923233986, "time": 0.7883558511734009, "epoch": 10, "memory": 6319, "step": 37763} +{"lr": 1.0212123026515194e-05, "data_time": 0.6573646783828735, "loss": 0.07633700519800186, "time": 0.7970118284225464, "epoch": 10, "memory": 6319, "step": 37863} +{"lr": 1.0178479224136881e-05, "data_time": 0.7019353628158569, "loss": 0.06876394785940647, "time": 0.8414695739746094, "epoch": 10, "memory": 6319, "step": 37963} +{"lr": 1.0147733413682289e-05, "data_time": 0.694713544845581, "loss": 0.06688739284873009, "time": 0.8349694728851318, "epoch": 10, "memory": 6319, "step": 38063} +{"lr": 1.0119887583060527e-05, "data_time": 0.6461351394653321, "loss": 0.06250211223959923, "time": 0.7862644672393799, "epoch": 10, "memory": 6319, "step": 38163} +{"lr": 1.0094943532678871e-05, "data_time": 0.660258412361145, "loss": 0.05961154475808143, "time": 0.8006170988082886, "epoch": 10, "memory": 6319, "step": 38263} +{"lr": 1.007290287532637e-05, "data_time": 0.6291829824447632, "loss": 0.07028537392616271, "time": 0.7687488794326782, "epoch": 10, "memory": 6319, "step": 38363} +{"lr": 1.0053767036069495e-05, "data_time": 0.7024988889694214, "loss": 0.06954733319580556, "time": 0.8421658992767334, "epoch": 10, "memory": 6319, "step": 38463} +{"lr": 1.003753725216005e-05, "data_time": 0.7504273176193237, "loss": 0.06711390689015388, "time": 0.8897173404693604, "epoch": 10, "memory": 6319, "step": 38563} +{"lr": 1.0024214572955152e-05, "data_time": 0.6651474237442017, "loss": 0.061823930218815805, "time": 0.8049480676651001, "epoch": 10, "memory": 6319, "step": 38663} +{"lr": 1.0013799859849456e-05, "data_time": 0.6606799364089966, "loss": 0.05535350404679775, "time": 0.8006099700927735, "epoch": 10, "memory": 6319, "step": 38763} +{"lr": 1.0006293786219338e-05, "data_time": 0.653389835357666, "loss": 0.058217883482575415, "time": 0.7929687261581421, "epoch": 10, "memory": 6319, "step": 38863} +{"lr": 1.0001696837379472e-05, "data_time": 0.6799966812133789, "loss": 0.05455840863287449, "time": 0.8202425718307496, "epoch": 10, "memory": 6319, "step": 38963} +{"lr": 1.0000009310551361e-05, "data_time": 0.692695426940918, "loss": 0.06514642164111137, "time": 0.8321401119232178, "epoch": 10, "memory": 6319, "step": 39063} +{"accuracy/top1": 97.5296630859375, "data_time": 0.5047953704307819, "time": 0.5928477418833765, "step": 10} diff --git a/resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1/20230604_215105/vis_data/vis_image/0000000.png_1.png b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1/20230604_215105/vis_data/vis_image/0000000.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..5986843894fc575660ceb93b1cc9b8e04cf5da7d Binary files /dev/null and b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1/20230604_215105/vis_data/vis_image/0000000.png_1.png differ diff --git a/resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1/20230604_215105/vis_data/vis_image/0000000.png_10.png b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1/20230604_215105/vis_data/vis_image/0000000.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..1650ddb73ae63769b613deb493a985d007b4a002 Binary files /dev/null and b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1/20230604_215105/vis_data/vis_image/0000000.png_10.png differ diff --git a/resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1/20230604_215105/vis_data/vis_image/0000000.png_2.png b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1/20230604_215105/vis_data/vis_image/0000000.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..f3d71f4c96e1c26657069b71fa5c55f352eac8e5 Binary files /dev/null and b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1/20230604_215105/vis_data/vis_image/0000000.png_2.png differ diff --git a/resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1/20230604_215105/vis_data/vis_image/0000000.png_3.png b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1/20230604_215105/vis_data/vis_image/0000000.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..59de2433a6e56fde2020e3e75529ea2bec3e2b5a Binary files /dev/null and b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1/20230604_215105/vis_data/vis_image/0000000.png_3.png differ diff --git a/resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1/20230604_215105/vis_data/vis_image/0000000.png_4.png b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1/20230604_215105/vis_data/vis_image/0000000.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..c85b40ec5c99fd2b1c784e672920378d1732a829 Binary files /dev/null and b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1/20230604_215105/vis_data/vis_image/0000000.png_4.png differ diff --git a/resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1/20230604_215105/vis_data/vis_image/0000000.png_5.png b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1/20230604_215105/vis_data/vis_image/0000000.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..3238f4ee93d65eb0bbaaa2688af0baa75e3a59c4 Binary files /dev/null and b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1/20230604_215105/vis_data/vis_image/0000000.png_5.png differ diff --git a/resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1/20230604_215105/vis_data/vis_image/0000000.png_6.png b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1/20230604_215105/vis_data/vis_image/0000000.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..23aed641249ca8fb200010abfcb664389b4c8a26 Binary files /dev/null and b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1/20230604_215105/vis_data/vis_image/0000000.png_6.png differ diff --git a/resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1/20230604_215105/vis_data/vis_image/0000000.png_7.png b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1/20230604_215105/vis_data/vis_image/0000000.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..efd5b99d5d442fdf47f47ad7a0a791a2edc999ec Binary files /dev/null and b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1/20230604_215105/vis_data/vis_image/0000000.png_7.png differ diff --git a/resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1/20230604_215105/vis_data/vis_image/0000000.png_8.png b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1/20230604_215105/vis_data/vis_image/0000000.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..072d2841d7a32fabecfe44704ebda446d4723db0 Binary files /dev/null and b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1/20230604_215105/vis_data/vis_image/0000000.png_8.png differ diff --git a/resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1/20230604_215105/vis_data/vis_image/0000000.png_9.png b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1/20230604_215105/vis_data/vis_image/0000000.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..3638156778848742bc4de60773b6575875fa5d5e Binary files /dev/null and b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1/20230604_215105/vis_data/vis_image/0000000.png_9.png differ diff --git a/resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1/20230604_215105/vis_data/vis_image/000005174.jpg_1.png b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1/20230604_215105/vis_data/vis_image/000005174.jpg_1.png new file mode 100644 index 0000000000000000000000000000000000000000..339d3c14feb6a0c6cc3160457aa1cbf26abf91e0 Binary files /dev/null and b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1/20230604_215105/vis_data/vis_image/000005174.jpg_1.png differ diff --git a/resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1/20230604_215105/vis_data/vis_image/000005174.jpg_10.png b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1/20230604_215105/vis_data/vis_image/000005174.jpg_10.png new file mode 100644 index 0000000000000000000000000000000000000000..915adeeaeb0aa6405bf7f847614cae014be57b59 Binary files /dev/null and b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1/20230604_215105/vis_data/vis_image/000005174.jpg_10.png differ diff --git a/resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1/20230604_215105/vis_data/vis_image/000005174.jpg_2.png b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1/20230604_215105/vis_data/vis_image/000005174.jpg_2.png new file mode 100644 index 0000000000000000000000000000000000000000..0fb2a5d8435097cea53253759f88f55e9ccbba47 Binary files /dev/null and b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1/20230604_215105/vis_data/vis_image/000005174.jpg_2.png differ diff --git a/resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1/20230604_215105/vis_data/vis_image/000005174.jpg_3.png b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1/20230604_215105/vis_data/vis_image/000005174.jpg_3.png new file mode 100644 index 0000000000000000000000000000000000000000..a302b83fd06c2990af68accd65645cc72ed99424 Binary files /dev/null and b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1/20230604_215105/vis_data/vis_image/000005174.jpg_3.png differ diff --git a/resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1/20230604_215105/vis_data/vis_image/000005174.jpg_4.png b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1/20230604_215105/vis_data/vis_image/000005174.jpg_4.png new file mode 100644 index 0000000000000000000000000000000000000000..e47c76b519f6b9b9e2448b1aa3c00a51ea54b6a8 Binary files /dev/null and b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1/20230604_215105/vis_data/vis_image/000005174.jpg_4.png differ diff --git a/resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1/20230604_215105/vis_data/vis_image/000005174.jpg_5.png b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1/20230604_215105/vis_data/vis_image/000005174.jpg_5.png new file mode 100644 index 0000000000000000000000000000000000000000..7aeaed90e85b789376dd6aac806621facd97dbda Binary files /dev/null and b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1/20230604_215105/vis_data/vis_image/000005174.jpg_5.png differ diff --git a/resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1/20230604_215105/vis_data/vis_image/000005174.jpg_6.png b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1/20230604_215105/vis_data/vis_image/000005174.jpg_6.png new file mode 100644 index 0000000000000000000000000000000000000000..adb3ce9e475a544c181c3072cd2e7a42a3acc0ab Binary files /dev/null and b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1/20230604_215105/vis_data/vis_image/000005174.jpg_6.png differ diff --git a/resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1/20230604_215105/vis_data/vis_image/000005174.jpg_7.png b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1/20230604_215105/vis_data/vis_image/000005174.jpg_7.png new file mode 100644 index 0000000000000000000000000000000000000000..3bc2edc6a7f471e670802cefbabc8aed6b10365e Binary files /dev/null and b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1/20230604_215105/vis_data/vis_image/000005174.jpg_7.png differ diff --git a/resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1/20230604_215105/vis_data/vis_image/000005174.jpg_8.png b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1/20230604_215105/vis_data/vis_image/000005174.jpg_8.png new file mode 100644 index 0000000000000000000000000000000000000000..2eacb2480e1c569b91e409b89064eaf72511679d Binary files /dev/null and b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1/20230604_215105/vis_data/vis_image/000005174.jpg_8.png differ diff --git a/resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1/20230604_215105/vis_data/vis_image/000005174.jpg_9.png b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1/20230604_215105/vis_data/vis_image/000005174.jpg_9.png new file mode 100644 index 0000000000000000000000000000000000000000..91b3cf3ae996be22411bef357b05e34ceb92a58a Binary files /dev/null and b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1/20230604_215105/vis_data/vis_image/000005174.jpg_9.png differ diff --git a/resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1/20230604_215105/vis_data/vis_image/0010000.png_1.png b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1/20230604_215105/vis_data/vis_image/0010000.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..1a111d4f309124a004be1f76c99eafd6e0fed6ad Binary files /dev/null and b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1/20230604_215105/vis_data/vis_image/0010000.png_1.png differ diff --git a/resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1/20230604_215105/vis_data/vis_image/0010000.png_10.png b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1/20230604_215105/vis_data/vis_image/0010000.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..d20469416743f1afefde99d85de46919c9317f54 Binary files /dev/null and b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1/20230604_215105/vis_data/vis_image/0010000.png_10.png differ diff --git a/resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1/20230604_215105/vis_data/vis_image/0010000.png_2.png b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1/20230604_215105/vis_data/vis_image/0010000.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..1cceb8ac507b17138462c231c388e6424f430105 Binary files /dev/null and b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1/20230604_215105/vis_data/vis_image/0010000.png_2.png differ diff --git a/resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1/20230604_215105/vis_data/vis_image/0010000.png_3.png b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1/20230604_215105/vis_data/vis_image/0010000.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..2c028613955ff61d4bf9d74d3b398f78d942fa80 Binary files /dev/null and b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1/20230604_215105/vis_data/vis_image/0010000.png_3.png differ diff --git a/resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1/20230604_215105/vis_data/vis_image/0010000.png_4.png b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1/20230604_215105/vis_data/vis_image/0010000.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..934a512e5cbc7abe4be053a9f4c6506a2b3f09ed Binary files /dev/null and b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1/20230604_215105/vis_data/vis_image/0010000.png_4.png differ diff --git a/resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1/20230604_215105/vis_data/vis_image/0010000.png_5.png b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1/20230604_215105/vis_data/vis_image/0010000.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..48c6bfca0ac7ccdbec4e94cd24a5a5d8e3d730ef Binary files /dev/null and b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1/20230604_215105/vis_data/vis_image/0010000.png_5.png differ diff --git a/resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1/20230604_215105/vis_data/vis_image/0010000.png_6.png b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1/20230604_215105/vis_data/vis_image/0010000.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..9656d8ca5b87865b3dfb01500bb4cced6b6c3291 Binary files /dev/null and b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1/20230604_215105/vis_data/vis_image/0010000.png_6.png differ diff --git a/resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1/20230604_215105/vis_data/vis_image/0010000.png_7.png b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1/20230604_215105/vis_data/vis_image/0010000.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..47d438b26c1c8f81113bfb321d3fdfbb6b1e8d1d Binary files /dev/null and b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1/20230604_215105/vis_data/vis_image/0010000.png_7.png differ diff --git a/resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1/20230604_215105/vis_data/vis_image/0010000.png_8.png b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1/20230604_215105/vis_data/vis_image/0010000.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..bc59409581bda1e65c8ec91039d9b885f831398c Binary files /dev/null and b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1/20230604_215105/vis_data/vis_image/0010000.png_8.png differ diff --git a/resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1/20230604_215105/vis_data/vis_image/0010000.png_9.png b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1/20230604_215105/vis_data/vis_image/0010000.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..463103d1bb0fefa640e4cdc379f1f4a095c6a705 Binary files /dev/null and b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1/20230604_215105/vis_data/vis_image/0010000.png_9.png differ diff --git a/resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1/epoch_1.pth b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1/epoch_1.pth new file mode 100644 index 0000000000000000000000000000000000000000..5be27898c17984058d6cdcd20f15307d817eeedc --- /dev/null +++ b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1/epoch_1.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:442eb9e7b0624195d9867ed338588c66b8fc08903c45fefb08cb833c801efa04 +size 277331661 diff --git a/resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1/epoch_10.pth b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1/epoch_10.pth new file mode 100644 index 0000000000000000000000000000000000000000..36b0d238737eb2d91604b104b2a5a1670fafc08b --- /dev/null +++ b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1/epoch_10.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:821da7db989fd28930276010ca52890370a95e00a36559f457c62766b1dc6975 +size 279909005 diff --git a/resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1/epoch_2.pth b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1/epoch_2.pth new file mode 100644 index 0000000000000000000000000000000000000000..676e877c5d82fdefd8e0e401c0f279fb95404c6f --- /dev/null +++ b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1/epoch_2.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:577a528ab6a9a262b61955652abd2599eb0ae2f820aec16791d3488a57f5ac40 +size 277617293 diff --git a/resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1/epoch_3.pth b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1/epoch_3.pth new file mode 100644 index 0000000000000000000000000000000000000000..0a4e9eefaa188c29a63adb020b661dc34e2d7c1e --- /dev/null +++ b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1/epoch_3.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:17c8178eab37f44b31feeb154dd5b60bfd70601f93f049f00f07d97323f12fda +size 277902541 diff --git a/resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1/epoch_4.pth b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1/epoch_4.pth new file mode 100644 index 0000000000000000000000000000000000000000..981bb323533dd3b67c4c780f69784c610933d87f --- /dev/null +++ b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1/epoch_4.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:65b6c5bd5c85b000588093234152edf953dd4dac9fec77cda7b64133c9351c73 +size 278187469 diff --git a/resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1/epoch_5.pth b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1/epoch_5.pth new file mode 100644 index 0000000000000000000000000000000000000000..8645ea4db15a9844a76fa1d2d74543c1de83f79b --- /dev/null +++ b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1/epoch_5.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:cb4b3af2947c309c491d29176c1b916dac93b2a89c3851734724dbffbf62da86 +size 278472461 diff --git a/resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1/epoch_6.pth b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1/epoch_6.pth new file mode 100644 index 0000000000000000000000000000000000000000..3dfc6131e0846fd1fca537187e87d0aa1105f954 --- /dev/null +++ b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1/epoch_6.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:71f61d6fafcb7e8a38f2a803e188c8738cdee6e44ffe0e18d7802dd1340f09fa +size 278757453 diff --git a/resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1/epoch_7.pth b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1/epoch_7.pth new file mode 100644 index 0000000000000000000000000000000000000000..d7547f675d90c37d8c270a938839dc6a6b5cc64c --- /dev/null +++ b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1/epoch_7.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:1a614455661210b5d78fbcbacd649c015c08b9632998499e86faf0e6287287dc +size 279042893 diff --git a/resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1/epoch_8.pth b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1/epoch_8.pth new file mode 100644 index 0000000000000000000000000000000000000000..aa98bf027a8fee076b9b76e05ac7c7ee7d802f30 --- /dev/null +++ b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1/epoch_8.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:f255674b629bb7afd8c066d27298f7e197d55e880e36359444480b896f2f3c5a +size 279331533 diff --git a/resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1/epoch_9.pth b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1/epoch_9.pth new file mode 100644 index 0000000000000000000000000000000000000000..97b994cab4a14ee291da9019f29c22b72f2d0a80 --- /dev/null +++ b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1/epoch_9.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:69f762b0e66dedfd42a05e124a501b7eeeb5698bc226e41da994a842132f06d7 +size 279620237 diff --git a/resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1/last_checkpoint b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1/last_checkpoint new file mode 100644 index 0000000000000000000000000000000000000000..600a648c26cfcdda0842731a2665d3a7ca479c2c --- /dev/null +++ b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1/last_checkpoint @@ -0,0 +1 @@ +/mnt/petrelfs/luzeyu/workspace/fakebench/mmpretrain/workdir/resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1/epoch_10.pth \ No newline at end of file diff --git a/resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1/resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1.py b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1/resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1.py new file mode 100644 index 0000000000000000000000000000000000000000..32dc328c1686171e778233c0a8aeab585932a5fe --- /dev/null +++ b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1/resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1.py @@ -0,0 +1,230 @@ +optim_wrapper = dict( + optimizer=dict( + type='SGD', + lr=0.0001, + momentum=0.9, + weight_decay=0.0001, + _scope_='mmpretrain'), + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=False, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=512) +model = dict( + type='ImageClassifier', + backbone=dict( + frozen_stages=2, + type='ResNet', + depth=50, + num_stages=4, + out_indices=(3, ), + style='pytorch', + init_cfg=dict( + type='Pretrained', + checkpoint= + 'https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth', + prefix='backbone')), + neck=dict(type='GlobalAveragePooling'), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=2048, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + topk=1)) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stablediffusionV1-5R2-dpmsolver-25-1m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/cc1m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = dict(type='Accuracy', topk=1) +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = dict(type='Accuracy', topk=1) +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/resnet50_2xb256_sdv1_1m_lr1e-4_aug_1e-1' diff --git a/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/20230604_225234/20230604_225234.log b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/20230604_225234/20230604_225234.log new file mode 100644 index 0000000000000000000000000000000000000000..29e1681ebbc6cbdfe6bf12a14219d6de5146297d --- /dev/null +++ b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/20230604_225234/20230604_225234.log @@ -0,0 +1,911 @@ +2023/06/04 22:52:37 - mmengine - INFO - +------------------------------------------------------------ +System environment: + sys.platform: linux + Python: 3.10.9 (main, Mar 8 2023, 10:47:38) [GCC 11.2.0] + CUDA available: True + numpy_random_seed: 358692289 + GPU 0,1: NVIDIA A100-SXM4-80GB + CUDA_HOME: /mnt/petrelfs/share/cuda-11.6 + NVCC: Cuda compilation tools, release 11.6, V11.6.124 + GCC: gcc (GCC) 7.5.0 + PyTorch: 1.13.1 + PyTorch compiling details: PyTorch built with: + - GCC 9.3 + - C++ Version: 201402 + - Intel(R) oneAPI Math Kernel Library Version 2021.4-Product Build 20210904 for Intel(R) 64 architecture applications + - Intel(R) MKL-DNN v2.6.0 (Git Hash 52b5f107dd9cf10910aaa19cb47f3abf9b349815) + - OpenMP 201511 (a.k.a. OpenMP 4.5) + - LAPACK is enabled (usually provided by MKL) + - NNPACK is enabled + - CPU capability usage: AVX2 + - CUDA Runtime 11.6 + - NVCC architecture flags: -gencode;arch=compute_37,code=sm_37;-gencode;arch=compute_50,code=sm_50;-gencode;arch=compute_60,code=sm_60;-gencode;arch=compute_61,code=sm_61;-gencode;arch=compute_70,code=sm_70;-gencode;arch=compute_75,code=sm_75;-gencode;arch=compute_80,code=sm_80;-gencode;arch=compute_86,code=sm_86;-gencode;arch=compute_37,code=compute_37 + - CuDNN 8.3.2 (built against CUDA 11.5) + - Magma 2.6.1 + - Build settings: BLAS_INFO=mkl, BUILD_TYPE=Release, CUDA_VERSION=11.6, CUDNN_VERSION=8.3.2, CXX_COMPILER=/opt/rh/devtoolset-9/root/usr/bin/c++, CXX_FLAGS= -fabi-version=11 -Wno-deprecated -fvisibility-inlines-hidden -DUSE_PTHREADPOOL -fopenmp -DNDEBUG -DUSE_KINETO -DUSE_FBGEMM -DUSE_QNNPACK -DUSE_PYTORCH_QNNPACK -DUSE_XNNPACK -DSYMBOLICATE_MOBILE_DEBUG_HANDLE -DEDGE_PROFILER_USE_KINETO -O2 -fPIC -Wno-narrowing -Wall -Wextra -Werror=return-type -Werror=non-virtual-dtor -Wno-missing-field-initializers -Wno-type-limits -Wno-array-bounds -Wno-unknown-pragmas -Wunused-local-typedefs -Wno-unused-parameter -Wno-unused-function -Wno-unused-result -Wno-strict-overflow -Wno-strict-aliasing -Wno-error=deprecated-declarations -Wno-stringop-overflow -Wno-psabi -Wno-error=pedantic -Wno-error=redundant-decls -Wno-error=old-style-cast -fdiagnostics-color=always -faligned-new -Wno-unused-but-set-variable -Wno-maybe-uninitialized -fno-math-errno -fno-trapping-math -Werror=format -Werror=cast-function-type -Wno-stringop-overflow, LAPACK_INFO=mkl, PERF_WITH_AVX=1, PERF_WITH_AVX2=1, PERF_WITH_AVX512=1, TORCH_VERSION=1.13.1, USE_CUDA=ON, USE_CUDNN=ON, USE_EXCEPTION_PTR=1, USE_GFLAGS=OFF, USE_GLOG=OFF, USE_MKL=ON, USE_MKLDNN=ON, USE_MPI=OFF, USE_NCCL=ON, USE_NNPACK=ON, USE_OPENMP=ON, USE_ROCM=OFF, + + TorchVision: 0.14.1 + OpenCV: 4.7.0 + MMEngine: 0.7.3 + +Runtime environment: + cudnn_benchmark: True + mp_cfg: {'mp_start_method': 'fork', 'opencv_num_threads': 0} + dist_cfg: {'backend': 'nccl'} + seed: None + deterministic: False + Distributed launcher: slurm + Distributed training: True + GPU number: 2 +------------------------------------------------------------ + +2023/06/04 22:52:43 - mmengine - INFO - Config: +optim_wrapper = dict( + optimizer=dict( + type='SGD', + lr=0.0001, + momentum=0.9, + weight_decay=0.0001, + _scope_='mmpretrain'), + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=False, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=512) +model = dict( + type='ImageClassifier', + backbone=dict( + frozen_stages=2, + type='ResNet', + depth=50, + num_stages=4, + out_indices=(3, ), + style='pytorch', + init_cfg=dict( + type='Pretrained', + checkpoint= + 'https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth', + prefix='backbone')), + neck=dict(type='GlobalAveragePooling'), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=2048, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + topk=1)) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stablediffusionV1-5R2-dpmsolver-25-1m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/cc1m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = dict(type='Accuracy', topk=1) +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = dict(type='Accuracy', topk=1) +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1' + +2023/06/04 22:52:56 - mmengine - INFO - Hooks will be executed in the following order: +before_run: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_load_checkpoint: +(ABOVE_NORMAL) EMAHook + -------------------- +before_train: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_train_epoch: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(NORMAL ) DistSamplerSeedHook + -------------------- +before_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook + -------------------- +after_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +after_train_epoch: +(NORMAL ) IterTimerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_val_epoch: +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook + -------------------- +before_val_iter: +(NORMAL ) IterTimerHook + -------------------- +after_val_iter: +(NORMAL ) IterTimerHook +(NORMAL ) VisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_val_epoch: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_save_checkpoint: +(ABOVE_NORMAL) EMAHook + -------------------- +after_train: +(VERY_LOW ) CheckpointHook + -------------------- +before_test_epoch: +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook + -------------------- +before_test_iter: +(NORMAL ) IterTimerHook + -------------------- +after_test_iter: +(NORMAL ) IterTimerHook +(NORMAL ) VisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_test_epoch: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_run: +(BELOW_NORMAL) LoggerHook + -------------------- +2023/06/04 22:53:16 - mmengine - INFO - load backbone in model from: https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth +Name of parameter - Initialization information + +backbone.conv1.weight - torch.Size([64, 3, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.bn1.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.bn1.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.conv1.weight - torch.Size([64, 64, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.bn1.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.bn1.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.conv2.weight - torch.Size([64, 64, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.bn2.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.bn2.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.conv3.weight - torch.Size([256, 64, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.bn3.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.bn3.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.downsample.0.weight - torch.Size([256, 64, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.downsample.1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.downsample.1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.conv1.weight - torch.Size([64, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.bn1.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.bn1.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.conv2.weight - torch.Size([64, 64, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.bn2.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.bn2.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.conv3.weight - torch.Size([256, 64, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.bn3.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.bn3.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.conv1.weight - torch.Size([64, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.bn1.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.bn1.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.conv2.weight - torch.Size([64, 64, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.bn2.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.bn2.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.conv3.weight - torch.Size([256, 64, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.bn3.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.bn3.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.conv1.weight - torch.Size([128, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.bn1.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.bn1.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.conv2.weight - torch.Size([128, 128, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.bn2.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.bn2.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.conv3.weight - torch.Size([512, 128, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.bn3.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.bn3.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.downsample.0.weight - torch.Size([512, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.downsample.1.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.downsample.1.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.conv1.weight - torch.Size([128, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.bn1.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.bn1.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.conv2.weight - torch.Size([128, 128, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.bn2.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.bn2.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.conv3.weight - torch.Size([512, 128, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.bn3.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.bn3.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.conv1.weight - torch.Size([128, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.bn1.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.bn1.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.conv2.weight - torch.Size([128, 128, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.bn2.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.bn2.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.conv3.weight - torch.Size([512, 128, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.bn3.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.bn3.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.conv1.weight - torch.Size([128, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.bn1.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.bn1.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.conv2.weight - torch.Size([128, 128, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.bn2.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.bn2.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.conv3.weight - torch.Size([512, 128, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.bn3.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.bn3.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.conv1.weight - torch.Size([256, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.bn1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.bn1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.conv2.weight - torch.Size([256, 256, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.bn2.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.bn2.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.conv3.weight - torch.Size([1024, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.bn3.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.bn3.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.downsample.0.weight - torch.Size([1024, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.downsample.1.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.downsample.1.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.conv1.weight - torch.Size([256, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.bn1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.bn1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.conv2.weight - torch.Size([256, 256, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.bn2.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.bn2.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.conv3.weight - torch.Size([1024, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.bn3.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.bn3.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.conv1.weight - torch.Size([256, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.bn1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.bn1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.conv2.weight - torch.Size([256, 256, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.bn2.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.bn2.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.conv3.weight - torch.Size([1024, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.bn3.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.bn3.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.conv1.weight - torch.Size([256, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.bn1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.bn1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.conv2.weight - torch.Size([256, 256, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.bn2.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.bn2.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.conv3.weight - torch.Size([1024, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.bn3.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.bn3.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.conv1.weight - torch.Size([256, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.bn1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.bn1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.conv2.weight - torch.Size([256, 256, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.bn2.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.bn2.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.conv3.weight - torch.Size([1024, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.bn3.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.bn3.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.conv1.weight - torch.Size([256, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.bn1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.bn1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.conv2.weight - torch.Size([256, 256, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.bn2.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.bn2.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.conv3.weight - torch.Size([1024, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.bn3.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.bn3.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.conv1.weight - torch.Size([512, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.bn1.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.bn1.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.conv2.weight - torch.Size([512, 512, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.bn2.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.bn2.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.conv3.weight - torch.Size([2048, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.bn3.weight - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.bn3.bias - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.downsample.0.weight - torch.Size([2048, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.downsample.1.weight - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.downsample.1.bias - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.conv1.weight - torch.Size([512, 2048, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.bn1.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.bn1.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.conv2.weight - torch.Size([512, 512, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.bn2.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.bn2.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.conv3.weight - torch.Size([2048, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.bn3.weight - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.bn3.bias - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.conv1.weight - torch.Size([512, 2048, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.bn1.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.bn1.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.conv2.weight - torch.Size([512, 512, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.bn2.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.bn2.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.conv3.weight - torch.Size([2048, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.bn3.weight - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.bn3.bias - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +head.fc.weight - torch.Size([2, 2048]): +NormalInit: mean=0, std=0.01, bias=0 + +head.fc.bias - torch.Size([2]): +NormalInit: mean=0, std=0.01, bias=0 +2023/06/04 22:53:16 - mmengine - WARNING - "FileClient" will be deprecated in future. Please use io functions in https://mmengine.readthedocs.io/en/latest/api/fileio.html#file-io +2023/06/04 22:53:16 - mmengine - WARNING - "HardDiskBackend" is the alias of "LocalBackend" and the former will be deprecated in future. +2023/06/04 22:53:16 - mmengine - INFO - Checkpoints will be saved to /mnt/petrelfs/luzeyu/workspace/fakebench/mmpretrain/workdir/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1. +2023/06/04 22:54:50 - mmengine - INFO - Epoch(train) [1][ 100/3907] lr: 9.9999e-05 eta: 10:06:44 time: 0.8845 data_time: 0.4807 memory: 9436 loss: 0.6215 +2023/06/04 22:56:18 - mmengine - INFO - Epoch(train) [1][ 200/3907] lr: 9.9994e-05 eta: 9:46:36 time: 0.9200 data_time: 0.4065 memory: 6319 loss: 0.5611 +2023/06/04 22:57:47 - mmengine - INFO - Epoch(train) [1][ 300/3907] lr: 9.9987e-05 eta: 9:42:12 time: 0.9063 data_time: 0.2852 memory: 6319 loss: 0.5187 +2023/06/04 22:59:14 - mmengine - INFO - Epoch(train) [1][ 400/3907] lr: 9.9977e-05 eta: 9:36:10 time: 0.8530 data_time: 0.2649 memory: 6319 loss: 0.4674 +2023/06/04 23:00:43 - mmengine - INFO - Epoch(train) [1][ 500/3907] lr: 9.9964e-05 eta: 9:33:51 time: 0.9599 data_time: 0.2566 memory: 6319 loss: 0.4438 +2023/06/04 23:02:12 - mmengine - INFO - Epoch(train) [1][ 600/3907] lr: 9.9948e-05 eta: 9:32:24 time: 0.8880 data_time: 0.1575 memory: 6319 loss: 0.4355 +2023/06/04 23:03:45 - mmengine - INFO - Epoch(train) [1][ 700/3907] lr: 9.9929e-05 eta: 9:34:17 time: 1.0156 data_time: 0.5151 memory: 6319 loss: 0.3920 +2023/06/04 23:05:17 - mmengine - INFO - Epoch(train) [1][ 800/3907] lr: 9.9907e-05 eta: 9:34:07 time: 0.8921 data_time: 0.7522 memory: 6319 loss: 0.3643 +2023/06/04 23:06:47 - mmengine - INFO - Epoch(train) [1][ 900/3907] lr: 9.9882e-05 eta: 9:32:48 time: 0.9407 data_time: 0.8009 memory: 6319 loss: 0.3550 +2023/06/04 23:08:19 - mmengine - INFO - Exp name: resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1_20230604_225234 +2023/06/04 23:08:19 - mmengine - INFO - Epoch(train) [1][1000/3907] lr: 9.9855e-05 eta: 9:32:31 time: 0.8908 data_time: 0.7513 memory: 6319 loss: 0.3280 +2023/06/04 23:09:50 - mmengine - INFO - Epoch(train) [1][1100/3907] lr: 9.9824e-05 eta: 9:31:36 time: 0.9108 data_time: 0.7708 memory: 6319 loss: 0.3226 +2023/06/04 23:11:24 - mmengine - INFO - Epoch(train) [1][1200/3907] lr: 9.9791e-05 eta: 9:32:09 time: 0.9084 data_time: 0.7671 memory: 6319 loss: 0.2961 +2023/06/04 23:12:56 - mmengine - INFO - Epoch(train) [1][1300/3907] lr: 9.9755e-05 eta: 9:30:59 time: 0.9580 data_time: 0.8187 memory: 6319 loss: 0.2984 +2023/06/04 23:14:24 - mmengine - INFO - Epoch(train) [1][1400/3907] lr: 9.9716e-05 eta: 9:28:32 time: 0.8756 data_time: 0.7345 memory: 6319 loss: 0.2771 +2023/06/04 23:15:53 - mmengine - INFO - Epoch(train) [1][1500/3907] lr: 9.9674e-05 eta: 9:26:17 time: 0.9425 data_time: 0.8023 memory: 6319 loss: 0.2787 +2023/06/04 23:17:26 - mmengine - INFO - Epoch(train) [1][1600/3907] lr: 9.9629e-05 eta: 9:25:38 time: 0.9166 data_time: 0.7755 memory: 6319 loss: 0.2630 +2023/06/04 23:18:55 - mmengine - INFO - Epoch(train) [1][1700/3907] lr: 9.9581e-05 eta: 9:23:45 time: 0.9121 data_time: 0.7724 memory: 6319 loss: 0.2665 +2023/06/04 23:20:25 - mmengine - INFO - Epoch(train) [1][1800/3907] lr: 9.9530e-05 eta: 9:21:50 time: 0.9554 data_time: 0.8140 memory: 6319 loss: 0.2633 +2023/06/04 23:21:56 - mmengine - INFO - Epoch(train) [1][1900/3907] lr: 9.9476e-05 eta: 9:20:29 time: 0.9379 data_time: 0.7844 memory: 6319 loss: 0.2642 +2023/06/04 23:23:25 - mmengine - INFO - Exp name: resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1_20230604_225234 +2023/06/04 23:23:25 - mmengine - INFO - Epoch(train) [1][2000/3907] lr: 9.9420e-05 eta: 9:18:47 time: 0.9429 data_time: 0.8033 memory: 6319 loss: 0.2563 +2023/06/04 23:24:54 - mmengine - INFO - Epoch(train) [1][2100/3907] lr: 9.9361e-05 eta: 9:16:42 time: 0.9284 data_time: 0.7860 memory: 6319 loss: 0.2464 +2023/06/04 23:26:27 - mmengine - INFO - Epoch(train) [1][2200/3907] lr: 9.9298e-05 eta: 9:15:51 time: 0.9516 data_time: 0.8111 memory: 6319 loss: 0.2610 +2023/06/04 23:27:57 - mmengine - INFO - Epoch(train) [1][2300/3907] lr: 9.9233e-05 eta: 9:14:23 time: 0.9586 data_time: 0.8178 memory: 6319 loss: 0.2458 +2023/06/04 23:29:27 - mmengine - INFO - Epoch(train) [1][2400/3907] lr: 9.9165e-05 eta: 9:12:39 time: 0.8383 data_time: 0.6977 memory: 6319 loss: 0.2372 +2023/06/04 23:30:57 - mmengine - INFO - Epoch(train) [1][2500/3907] lr: 9.9095e-05 eta: 9:11:04 time: 0.9077 data_time: 0.7598 memory: 6319 loss: 0.2316 +2023/06/04 23:32:26 - mmengine - INFO - Epoch(train) [1][2600/3907] lr: 9.9021e-05 eta: 9:09:21 time: 0.8839 data_time: 0.7438 memory: 6319 loss: 0.2406 +2023/06/04 23:33:57 - mmengine - INFO - Epoch(train) [1][2700/3907] lr: 9.8944e-05 eta: 9:07:47 time: 0.9159 data_time: 0.7758 memory: 6319 loss: 0.2260 +2023/06/04 23:35:27 - mmengine - INFO - Epoch(train) [1][2800/3907] lr: 9.8865e-05 eta: 9:06:12 time: 0.8969 data_time: 0.7563 memory: 6319 loss: 0.2175 +2023/06/04 23:36:57 - mmengine - INFO - Epoch(train) [1][2900/3907] lr: 9.8783e-05 eta: 9:04:46 time: 0.8350 data_time: 0.6903 memory: 6319 loss: 0.2243 +2023/06/04 23:38:27 - mmengine - INFO - Exp name: resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1_20230604_225234 +2023/06/04 23:38:27 - mmengine - INFO - Epoch(train) [1][3000/3907] lr: 9.8698e-05 eta: 9:03:13 time: 0.8761 data_time: 0.7358 memory: 6319 loss: 0.2049 +2023/06/04 23:39:58 - mmengine - INFO - Epoch(train) [1][3100/3907] lr: 9.8610e-05 eta: 9:01:45 time: 0.9651 data_time: 0.8259 memory: 6319 loss: 0.2048 +2023/06/04 23:41:28 - mmengine - INFO - Epoch(train) [1][3200/3907] lr: 9.8519e-05 eta: 9:00:09 time: 0.9377 data_time: 0.7972 memory: 6319 loss: 0.1967 +2023/06/04 23:42:59 - mmengine - INFO - Epoch(train) [1][3300/3907] lr: 9.8426e-05 eta: 8:58:50 time: 0.9831 data_time: 0.8432 memory: 6319 loss: 0.1920 +2023/06/04 23:44:30 - mmengine - INFO - Epoch(train) [1][3400/3907] lr: 9.8330e-05 eta: 8:57:21 time: 0.8967 data_time: 0.7564 memory: 6319 loss: 0.2002 +2023/06/04 23:45:53 - mmengine - INFO - Epoch(train) [1][3500/3907] lr: 9.8231e-05 eta: 8:54:35 time: 0.8935 data_time: 0.7535 memory: 6319 loss: 0.2055 +2023/06/04 23:47:22 - mmengine - INFO - Epoch(train) [1][3600/3907] lr: 9.8129e-05 eta: 8:52:52 time: 0.9404 data_time: 0.8013 memory: 6319 loss: 0.1754 +2023/06/04 23:48:50 - mmengine - INFO - Epoch(train) [1][3700/3907] lr: 9.8024e-05 eta: 8:51:07 time: 0.9046 data_time: 0.7640 memory: 6319 loss: 0.1823 +2023/06/04 23:50:21 - mmengine - INFO - Epoch(train) [1][3800/3907] lr: 9.7917e-05 eta: 8:49:46 time: 0.8511 data_time: 0.7111 memory: 6319 loss: 0.1809 +2023/06/04 23:51:52 - mmengine - INFO - Epoch(train) [1][3900/3907] lr: 9.7806e-05 eta: 8:48:24 time: 0.9282 data_time: 0.7875 memory: 6319 loss: 0.1796 +2023/06/04 23:52:01 - mmengine - INFO - Exp name: resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1_20230604_225234 +2023/06/04 23:52:01 - mmengine - INFO - Saving checkpoint at 1 epochs +2023/06/04 23:52:50 - mmengine - INFO - Epoch(val) [1][57/57] accuracy/top1: 69.8827 data_time: 0.5719 time: 0.6628 +2023/06/04 23:54:21 - mmengine - INFO - Exp name: resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1_20230604_225234 +2023/06/04 23:54:22 - mmengine - INFO - Epoch(train) [2][ 100/3907] lr: 9.7685e-05 eta: 8:47:20 time: 0.8868 data_time: 0.7218 memory: 6319 loss: 0.1896 +2023/06/04 23:55:53 - mmengine - INFO - Epoch(train) [2][ 200/3907] lr: 9.7570e-05 eta: 8:45:51 time: 0.8879 data_time: 0.6016 memory: 6319 loss: 0.1862 +2023/06/04 23:57:22 - mmengine - INFO - Epoch(train) [2][ 300/3907] lr: 9.7451e-05 eta: 8:44:18 time: 0.8896 data_time: 0.6072 memory: 6319 loss: 0.1781 +2023/06/04 23:58:54 - mmengine - INFO - Epoch(train) [2][ 400/3907] lr: 9.7329e-05 eta: 8:42:54 time: 0.8982 data_time: 0.7579 memory: 6319 loss: 0.1914 +2023/06/05 00:00:25 - mmengine - INFO - Epoch(train) [2][ 500/3907] lr: 9.7205e-05 eta: 8:41:31 time: 0.9556 data_time: 0.8162 memory: 6319 loss: 0.1814 +2023/06/05 00:01:54 - mmengine - INFO - Epoch(train) [2][ 600/3907] lr: 9.7078e-05 eta: 8:39:56 time: 0.8879 data_time: 0.7476 memory: 6319 loss: 0.1805 +2023/06/05 00:03:26 - mmengine - INFO - Epoch(train) [2][ 700/3907] lr: 9.6949e-05 eta: 8:38:39 time: 0.9267 data_time: 0.7871 memory: 6319 loss: 0.1815 +2023/06/05 00:04:57 - mmengine - INFO - Epoch(train) [2][ 800/3907] lr: 9.6816e-05 eta: 8:37:11 time: 0.9145 data_time: 0.7730 memory: 6319 loss: 0.1681 +2023/06/05 00:06:29 - mmengine - INFO - Epoch(train) [2][ 900/3907] lr: 9.6681e-05 eta: 8:35:51 time: 1.0679 data_time: 0.9281 memory: 6319 loss: 0.1759 +2023/06/05 00:07:54 - mmengine - INFO - Epoch(train) [2][1000/3907] lr: 9.6544e-05 eta: 8:33:48 time: 0.8482 data_time: 0.7081 memory: 6319 loss: 0.1633 +2023/06/05 00:09:21 - mmengine - INFO - Exp name: resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1_20230604_225234 +2023/06/05 00:09:22 - mmengine - INFO - Epoch(train) [2][1100/3907] lr: 9.6403e-05 eta: 8:31:57 time: 0.8467 data_time: 0.7069 memory: 6319 loss: 0.1815 +2023/06/05 00:10:50 - mmengine - INFO - Epoch(train) [2][1200/3907] lr: 9.6260e-05 eta: 8:30:12 time: 0.8424 data_time: 0.7018 memory: 6319 loss: 0.1666 +2023/06/05 00:12:16 - mmengine - INFO - Epoch(train) [2][1300/3907] lr: 9.6114e-05 eta: 8:28:15 time: 0.8867 data_time: 0.7469 memory: 6319 loss: 0.1600 +2023/06/05 00:13:46 - mmengine - INFO - Epoch(train) [2][1400/3907] lr: 9.5966e-05 eta: 8:26:47 time: 0.9278 data_time: 0.7864 memory: 6319 loss: 0.1625 +2023/06/05 00:15:16 - mmengine - INFO - Epoch(train) [2][1500/3907] lr: 9.5815e-05 eta: 8:25:16 time: 0.8201 data_time: 0.6791 memory: 6319 loss: 0.1751 +2023/06/05 00:16:39 - mmengine - INFO - Epoch(train) [2][1600/3907] lr: 9.5661e-05 eta: 8:23:01 time: 0.8245 data_time: 0.6841 memory: 6319 loss: 0.1717 +2023/06/05 00:18:07 - mmengine - INFO - Epoch(train) [2][1700/3907] lr: 9.5505e-05 eta: 8:21:23 time: 0.8652 data_time: 0.7234 memory: 6319 loss: 0.1549 +2023/06/05 00:19:38 - mmengine - INFO - Epoch(train) [2][1800/3907] lr: 9.5346e-05 eta: 8:19:59 time: 0.9551 data_time: 0.8134 memory: 6319 loss: 0.1533 diff --git a/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/20230604_225234/vis_data/20230604_225234.json b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/20230604_225234/vis_data/20230604_225234.json new file mode 100644 index 0000000000000000000000000000000000000000..af78cb8a3aa2449fd93832f729b9542dd3c3d680 --- /dev/null +++ b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/20230604_225234/vis_data/20230604_225234.json @@ -0,0 +1,58 @@ +{"lr": 9.999857418382631e-05, "data_time": 0.48071534633636476, "loss": 0.6214756667613983, "time": 0.884512209892273, "epoch": 1, "memory": 9436, "step": 100} +{"lr": 9.999423907359506e-05, "data_time": 0.40646538734436033, "loss": 0.5611209154129029, "time": 0.9200191736221314, "epoch": 1, "memory": 6319, "step": 200} +{"lr": 9.998699480412073e-05, "data_time": 0.2852163791656494, "loss": 0.5186954021453858, "time": 0.9063416481018066, "epoch": 1, "memory": 6319, "step": 300} +{"lr": 9.997684184379085e-05, "data_time": 0.2648536205291748, "loss": 0.4674046695232391, "time": 0.8530332326889039, "epoch": 1, "memory": 6319, "step": 400} +{"lr": 9.996378084905784e-05, "data_time": 0.256616473197937, "loss": 0.4437552928924561, "time": 0.9599295377731323, "epoch": 1, "memory": 6319, "step": 500} +{"lr": 9.994781266439667e-05, "data_time": 0.1574784278869629, "loss": 0.4355391889810562, "time": 0.8879710674285889, "epoch": 1, "memory": 6319, "step": 600} +{"lr": 9.992893832225061e-05, "data_time": 0.5151271343231201, "loss": 0.3920274466276169, "time": 1.0155930757522582, "epoch": 1, "memory": 6319, "step": 700} +{"lr": 9.990715904296405e-05, "data_time": 0.752193021774292, "loss": 0.3642873466014862, "time": 0.8921104431152344, "epoch": 1, "memory": 6319, "step": 800} +{"lr": 9.988247623470371e-05, "data_time": 0.8008645057678223, "loss": 0.3549769729375839, "time": 0.9406989336013794, "epoch": 1, "memory": 6319, "step": 900} +{"lr": 9.985489149336749e-05, "data_time": 0.7513064622879029, "loss": 0.32802344858646393, "time": 0.8908206462860108, "epoch": 1, "memory": 6319, "step": 1000} +{"lr": 9.982440660248187e-05, "data_time": 0.7708238363265991, "loss": 0.3226481914520264, "time": 0.9108400106430053, "epoch": 1, "memory": 6319, "step": 1100} +{"lr": 9.979102353308574e-05, "data_time": 0.7671159744262696, "loss": 0.29605636894702914, "time": 0.9083706855773925, "epoch": 1, "memory": 6319, "step": 1200} +{"lr": 9.975474444360358e-05, "data_time": 0.8186986684799195, "loss": 0.2984150111675262, "time": 0.9580048322677612, "epoch": 1, "memory": 6319, "step": 1300} +{"lr": 9.97155716797056e-05, "data_time": 0.734524130821228, "loss": 0.27713678032159805, "time": 0.8755678176879883, "epoch": 1, "memory": 6319, "step": 1400} +{"lr": 9.967350777415632e-05, "data_time": 0.8022580862045288, "loss": 0.2786711513996124, "time": 0.9425456047058105, "epoch": 1, "memory": 6319, "step": 1500} +{"lr": 9.962855544665059e-05, "data_time": 0.775536322593689, "loss": 0.2630007520318031, "time": 0.9165681838989258, "epoch": 1, "memory": 6319, "step": 1600} +{"lr": 9.958071760363773e-05, "data_time": 0.7723695278167725, "loss": 0.2665025681257248, "time": 0.9121029853820801, "epoch": 1, "memory": 6319, "step": 1700} +{"lr": 9.952999733813382e-05, "data_time": 0.813955569267273, "loss": 0.263336718082428, "time": 0.9554113864898681, "epoch": 1, "memory": 6319, "step": 1800} +{"lr": 9.947639792952148e-05, "data_time": 0.7844291448593139, "loss": 0.26416445672512057, "time": 0.937911057472229, "epoch": 1, "memory": 6319, "step": 1900} +{"lr": 9.941992284333813e-05, "data_time": 0.8032824277877808, "loss": 0.2562682881951332, "time": 0.9428738832473755, "epoch": 1, "memory": 6319, "step": 2000} +{"lr": 9.93605757310516e-05, "data_time": 0.7860048055648804, "loss": 0.24636256396770478, "time": 0.9283785343170166, "epoch": 1, "memory": 6319, "step": 2100} +{"lr": 9.929836042982421e-05, "data_time": 0.8111056566238404, "loss": 0.2609648942947388, "time": 0.9515869379043579, "epoch": 1, "memory": 6319, "step": 2200} +{"lr": 9.923328096226467e-05, "data_time": 0.8178458690643311, "loss": 0.24578360915184022, "time": 0.9586335897445679, "epoch": 1, "memory": 6319, "step": 2300} +{"lr": 9.916534153616819e-05, "data_time": 0.6977327823638916, "loss": 0.2371533378958702, "time": 0.8383141994476319, "epoch": 1, "memory": 6319, "step": 2400} +{"lr": 9.909454654424387e-05, "data_time": 0.759843373298645, "loss": 0.2315744549036026, "time": 0.907707929611206, "epoch": 1, "memory": 6319, "step": 2500} +{"lr": 9.902090056383109e-05, "data_time": 0.7438435077667236, "loss": 0.24061515033245087, "time": 0.8839004278182984, "epoch": 1, "memory": 6319, "step": 2600} +{"lr": 9.894440835660365e-05, "data_time": 0.7758362531661988, "loss": 0.22597335577011107, "time": 0.9159365892410278, "epoch": 1, "memory": 6319, "step": 2700} +{"lr": 9.886507486826162e-05, "data_time": 0.7563228845596314, "loss": 0.217546309530735, "time": 0.8968774557113648, "epoch": 1, "memory": 6319, "step": 2800} +{"lr": 9.878290522821154e-05, "data_time": 0.6902622938156128, "loss": 0.22426812946796418, "time": 0.8350137948989869, "epoch": 1, "memory": 6319, "step": 2900} +{"lr": 9.86979047492352e-05, "data_time": 0.7357762098312378, "loss": 0.2048680603504181, "time": 0.8761419057846069, "epoch": 1, "memory": 6319, "step": 3000} +{"lr": 9.861007892714578e-05, "data_time": 0.8259191513061523, "loss": 0.20484017878770827, "time": 0.965129804611206, "epoch": 1, "memory": 6319, "step": 3100} +{"lr": 9.851943344043239e-05, "data_time": 0.7971682786941529, "loss": 0.1966571554541588, "time": 0.9376757860183715, "epoch": 1, "memory": 6319, "step": 3200} +{"lr": 9.842597414989319e-05, "data_time": 0.8432482004165649, "loss": 0.1920214667916298, "time": 0.9830685615539551, "epoch": 1, "memory": 6319, "step": 3300} +{"lr": 9.832970709825637e-05, "data_time": 0.7563623905181884, "loss": 0.20023352652788162, "time": 0.8966696500778198, "epoch": 1, "memory": 6319, "step": 3400} +{"lr": 9.823063850978968e-05, "data_time": 0.7534742593765259, "loss": 0.20545622259378432, "time": 0.8934894323348999, "epoch": 1, "memory": 6319, "step": 3500} +{"lr": 9.81287747898975e-05, "data_time": 0.8013161897659302, "loss": 0.17536355704069137, "time": 0.9403944253921509, "epoch": 1, "memory": 6319, "step": 3600} +{"lr": 9.802412252470715e-05, "data_time": 0.764031457901001, "loss": 0.18228217363357543, "time": 0.9046345233917237, "epoch": 1, "memory": 6319, "step": 3700} +{"lr": 9.791668848064258e-05, "data_time": 0.7111430406570435, "loss": 0.18087174892425537, "time": 0.8511451959609986, "epoch": 1, "memory": 6319, "step": 3800} +{"lr": 9.780647960398751e-05, "data_time": 0.7875319480895996, "loss": 0.17960446923971177, "time": 0.9282076358795166, "epoch": 1, "memory": 6319, "step": 3900} +{"accuracy/top1": 69.88274383544922, "data_time": 0.571912309579682, "time": 0.6627781098349053, "step": 1} +{"lr": 9.7685491196646e-05, "data_time": 0.7218224048614502, "loss": 0.18962292820215226, "time": 0.8868228673934937, "epoch": 2, "memory": 6319, "step": 4007} +{"lr": 9.756956126085676e-05, "data_time": 0.6016045808792114, "loss": 0.18620216101408005, "time": 0.8879056453704834, "epoch": 2, "memory": 6319, "step": 4107} +{"lr": 9.745087893642482e-05, "data_time": 0.6072285175323486, "loss": 0.17807240635156632, "time": 0.8895961046218872, "epoch": 2, "memory": 6319, "step": 4207} +{"lr": 9.73294518969056e-05, "data_time": 0.757879090309143, "loss": 0.19140062034130095, "time": 0.8981546640396119, "epoch": 2, "memory": 6319, "step": 4307} +{"lr": 9.72052879933172e-05, "data_time": 0.8162017345428467, "loss": 0.18137903064489364, "time": 0.9555898427963256, "epoch": 2, "memory": 6319, "step": 4407} +{"lr": 9.707839525363358e-05, "data_time": 0.7476181507110595, "loss": 0.18049391508102416, "time": 0.8879045963287353, "epoch": 2, "memory": 6319, "step": 4507} +{"lr": 9.69487818822646e-05, "data_time": 0.7870512962341308, "loss": 0.18146572560071944, "time": 0.9267127990722657, "epoch": 2, "memory": 6319, "step": 4607} +{"lr": 9.681645625952616e-05, "data_time": 0.7730315685272217, "loss": 0.16808514446020126, "time": 0.9145148754119873, "epoch": 2, "memory": 6319, "step": 4707} +{"lr": 9.668142694109826e-05, "data_time": 0.9281142950057983, "loss": 0.17594008296728134, "time": 1.0678827047348023, "epoch": 2, "memory": 6319, "step": 4807} +{"lr": 9.654370265747162e-05, "data_time": 0.7081193685531616, "loss": 0.16326686292886733, "time": 0.8481757640838623, "epoch": 2, "memory": 6319, "step": 4907} +{"lr": 9.640329231338345e-05, "data_time": 0.7069461345672607, "loss": 0.18151011168956757, "time": 0.8467189550399781, "epoch": 2, "memory": 6319, "step": 5007} +{"lr": 9.626020498724147e-05, "data_time": 0.7017974615097046, "loss": 0.16655913591384888, "time": 0.8423813581466675, "epoch": 2, "memory": 6319, "step": 5107} +{"lr": 9.611444993053717e-05, "data_time": 0.7469312667846679, "loss": 0.1599966548383236, "time": 0.8867108583450317, "epoch": 2, "memory": 6319, "step": 5207} +{"lr": 9.596603656724746e-05, "data_time": 0.7863739013671875, "loss": 0.16251930892467498, "time": 0.9277769804000855, "epoch": 2, "memory": 6319, "step": 5307} +{"lr": 9.58149744932254e-05, "data_time": 0.679094934463501, "loss": 0.1750597044825554, "time": 0.8200845718383789, "epoch": 2, "memory": 6319, "step": 5407} +{"lr": 9.566127347557972e-05, "data_time": 0.684056305885315, "loss": 0.17173441201448442, "time": 0.8245492935180664, "epoch": 2, "memory": 6319, "step": 5507} +{"lr": 9.550494345204367e-05, "data_time": 0.7233998298645019, "loss": 0.15493762791156768, "time": 0.8651554107666015, "epoch": 2, "memory": 6319, "step": 5607} +{"lr": 9.534599453033183e-05, "data_time": 0.8133837461471558, "loss": 0.15334631204605104, "time": 0.9550534248352051, "epoch": 2, "memory": 6319, "step": 5707} diff --git a/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/20230604_225234/vis_data/config.py b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/20230604_225234/vis_data/config.py new file mode 100644 index 0000000000000000000000000000000000000000..661bf63adae66645caa3af55d05b1778159f197c --- /dev/null +++ b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/20230604_225234/vis_data/config.py @@ -0,0 +1,230 @@ +optim_wrapper = dict( + optimizer=dict( + type='SGD', + lr=0.0001, + momentum=0.9, + weight_decay=0.0001, + _scope_='mmpretrain'), + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=False, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=512) +model = dict( + type='ImageClassifier', + backbone=dict( + frozen_stages=2, + type='ResNet', + depth=50, + num_stages=4, + out_indices=(3, ), + style='pytorch', + init_cfg=dict( + type='Pretrained', + checkpoint= + 'https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth', + prefix='backbone')), + neck=dict(type='GlobalAveragePooling'), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=2048, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + topk=1)) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stablediffusionV1-5R2-dpmsolver-25-1m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/cc1m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = dict(type='Accuracy', topk=1) +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = dict(type='Accuracy', topk=1) +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1' diff --git a/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/20230604_225234/vis_data/events.out.tfevents.1685890370.SH-IDC1-10-140-24-17.159322.0 b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/20230604_225234/vis_data/events.out.tfevents.1685890370.SH-IDC1-10-140-24-17.159322.0 new file mode 100644 index 0000000000000000000000000000000000000000..d79f51ff607cee397b2cc415f87c75ba6583c6ae --- /dev/null +++ b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/20230604_225234/vis_data/events.out.tfevents.1685890370.SH-IDC1-10-140-24-17.159322.0 @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:e37a7bf044fe4eae424792888008cac2aeddf5b066d2b0158a9194a1e8393d42 +size 211473 diff --git a/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/20230604_225234/vis_data/scalars.json b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/20230604_225234/vis_data/scalars.json new file mode 100644 index 0000000000000000000000000000000000000000..af78cb8a3aa2449fd93832f729b9542dd3c3d680 --- /dev/null +++ b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/20230604_225234/vis_data/scalars.json @@ -0,0 +1,58 @@ +{"lr": 9.999857418382631e-05, "data_time": 0.48071534633636476, "loss": 0.6214756667613983, "time": 0.884512209892273, "epoch": 1, "memory": 9436, "step": 100} +{"lr": 9.999423907359506e-05, "data_time": 0.40646538734436033, "loss": 0.5611209154129029, "time": 0.9200191736221314, "epoch": 1, "memory": 6319, "step": 200} +{"lr": 9.998699480412073e-05, "data_time": 0.2852163791656494, "loss": 0.5186954021453858, "time": 0.9063416481018066, "epoch": 1, "memory": 6319, "step": 300} +{"lr": 9.997684184379085e-05, "data_time": 0.2648536205291748, "loss": 0.4674046695232391, "time": 0.8530332326889039, "epoch": 1, "memory": 6319, "step": 400} +{"lr": 9.996378084905784e-05, "data_time": 0.256616473197937, "loss": 0.4437552928924561, "time": 0.9599295377731323, "epoch": 1, "memory": 6319, "step": 500} +{"lr": 9.994781266439667e-05, "data_time": 0.1574784278869629, "loss": 0.4355391889810562, "time": 0.8879710674285889, "epoch": 1, "memory": 6319, "step": 600} +{"lr": 9.992893832225061e-05, "data_time": 0.5151271343231201, "loss": 0.3920274466276169, "time": 1.0155930757522582, "epoch": 1, "memory": 6319, "step": 700} +{"lr": 9.990715904296405e-05, "data_time": 0.752193021774292, "loss": 0.3642873466014862, "time": 0.8921104431152344, "epoch": 1, "memory": 6319, "step": 800} +{"lr": 9.988247623470371e-05, "data_time": 0.8008645057678223, "loss": 0.3549769729375839, "time": 0.9406989336013794, "epoch": 1, "memory": 6319, "step": 900} +{"lr": 9.985489149336749e-05, "data_time": 0.7513064622879029, "loss": 0.32802344858646393, "time": 0.8908206462860108, "epoch": 1, "memory": 6319, "step": 1000} +{"lr": 9.982440660248187e-05, "data_time": 0.7708238363265991, "loss": 0.3226481914520264, "time": 0.9108400106430053, "epoch": 1, "memory": 6319, "step": 1100} +{"lr": 9.979102353308574e-05, "data_time": 0.7671159744262696, "loss": 0.29605636894702914, "time": 0.9083706855773925, "epoch": 1, "memory": 6319, "step": 1200} +{"lr": 9.975474444360358e-05, "data_time": 0.8186986684799195, "loss": 0.2984150111675262, "time": 0.9580048322677612, "epoch": 1, "memory": 6319, "step": 1300} +{"lr": 9.97155716797056e-05, "data_time": 0.734524130821228, "loss": 0.27713678032159805, "time": 0.8755678176879883, "epoch": 1, "memory": 6319, "step": 1400} +{"lr": 9.967350777415632e-05, "data_time": 0.8022580862045288, "loss": 0.2786711513996124, "time": 0.9425456047058105, "epoch": 1, "memory": 6319, "step": 1500} +{"lr": 9.962855544665059e-05, "data_time": 0.775536322593689, "loss": 0.2630007520318031, "time": 0.9165681838989258, "epoch": 1, "memory": 6319, "step": 1600} +{"lr": 9.958071760363773e-05, "data_time": 0.7723695278167725, "loss": 0.2665025681257248, "time": 0.9121029853820801, "epoch": 1, "memory": 6319, "step": 1700} +{"lr": 9.952999733813382e-05, "data_time": 0.813955569267273, "loss": 0.263336718082428, "time": 0.9554113864898681, "epoch": 1, "memory": 6319, "step": 1800} +{"lr": 9.947639792952148e-05, "data_time": 0.7844291448593139, "loss": 0.26416445672512057, "time": 0.937911057472229, "epoch": 1, "memory": 6319, "step": 1900} +{"lr": 9.941992284333813e-05, "data_time": 0.8032824277877808, "loss": 0.2562682881951332, "time": 0.9428738832473755, "epoch": 1, "memory": 6319, "step": 2000} +{"lr": 9.93605757310516e-05, "data_time": 0.7860048055648804, "loss": 0.24636256396770478, "time": 0.9283785343170166, "epoch": 1, "memory": 6319, "step": 2100} +{"lr": 9.929836042982421e-05, "data_time": 0.8111056566238404, "loss": 0.2609648942947388, "time": 0.9515869379043579, "epoch": 1, "memory": 6319, "step": 2200} +{"lr": 9.923328096226467e-05, "data_time": 0.8178458690643311, "loss": 0.24578360915184022, "time": 0.9586335897445679, "epoch": 1, "memory": 6319, "step": 2300} +{"lr": 9.916534153616819e-05, "data_time": 0.6977327823638916, "loss": 0.2371533378958702, "time": 0.8383141994476319, "epoch": 1, "memory": 6319, "step": 2400} +{"lr": 9.909454654424387e-05, "data_time": 0.759843373298645, "loss": 0.2315744549036026, "time": 0.907707929611206, "epoch": 1, "memory": 6319, "step": 2500} +{"lr": 9.902090056383109e-05, "data_time": 0.7438435077667236, "loss": 0.24061515033245087, "time": 0.8839004278182984, "epoch": 1, "memory": 6319, "step": 2600} +{"lr": 9.894440835660365e-05, "data_time": 0.7758362531661988, "loss": 0.22597335577011107, "time": 0.9159365892410278, "epoch": 1, "memory": 6319, "step": 2700} +{"lr": 9.886507486826162e-05, "data_time": 0.7563228845596314, "loss": 0.217546309530735, "time": 0.8968774557113648, "epoch": 1, "memory": 6319, "step": 2800} +{"lr": 9.878290522821154e-05, "data_time": 0.6902622938156128, "loss": 0.22426812946796418, "time": 0.8350137948989869, "epoch": 1, "memory": 6319, "step": 2900} +{"lr": 9.86979047492352e-05, "data_time": 0.7357762098312378, "loss": 0.2048680603504181, "time": 0.8761419057846069, "epoch": 1, "memory": 6319, "step": 3000} +{"lr": 9.861007892714578e-05, "data_time": 0.8259191513061523, "loss": 0.20484017878770827, "time": 0.965129804611206, "epoch": 1, "memory": 6319, "step": 3100} +{"lr": 9.851943344043239e-05, "data_time": 0.7971682786941529, "loss": 0.1966571554541588, "time": 0.9376757860183715, "epoch": 1, "memory": 6319, "step": 3200} +{"lr": 9.842597414989319e-05, "data_time": 0.8432482004165649, "loss": 0.1920214667916298, "time": 0.9830685615539551, "epoch": 1, "memory": 6319, "step": 3300} +{"lr": 9.832970709825637e-05, "data_time": 0.7563623905181884, "loss": 0.20023352652788162, "time": 0.8966696500778198, "epoch": 1, "memory": 6319, "step": 3400} +{"lr": 9.823063850978968e-05, "data_time": 0.7534742593765259, "loss": 0.20545622259378432, "time": 0.8934894323348999, "epoch": 1, "memory": 6319, "step": 3500} +{"lr": 9.81287747898975e-05, "data_time": 0.8013161897659302, "loss": 0.17536355704069137, "time": 0.9403944253921509, "epoch": 1, "memory": 6319, "step": 3600} +{"lr": 9.802412252470715e-05, "data_time": 0.764031457901001, "loss": 0.18228217363357543, "time": 0.9046345233917237, "epoch": 1, "memory": 6319, "step": 3700} +{"lr": 9.791668848064258e-05, "data_time": 0.7111430406570435, "loss": 0.18087174892425537, "time": 0.8511451959609986, "epoch": 1, "memory": 6319, "step": 3800} +{"lr": 9.780647960398751e-05, "data_time": 0.7875319480895996, "loss": 0.17960446923971177, "time": 0.9282076358795166, "epoch": 1, "memory": 6319, "step": 3900} +{"accuracy/top1": 69.88274383544922, "data_time": 0.571912309579682, "time": 0.6627781098349053, "step": 1} +{"lr": 9.7685491196646e-05, "data_time": 0.7218224048614502, "loss": 0.18962292820215226, "time": 0.8868228673934937, "epoch": 2, "memory": 6319, "step": 4007} +{"lr": 9.756956126085676e-05, "data_time": 0.6016045808792114, "loss": 0.18620216101408005, "time": 0.8879056453704834, "epoch": 2, "memory": 6319, "step": 4107} +{"lr": 9.745087893642482e-05, "data_time": 0.6072285175323486, "loss": 0.17807240635156632, "time": 0.8895961046218872, "epoch": 2, "memory": 6319, "step": 4207} +{"lr": 9.73294518969056e-05, "data_time": 0.757879090309143, "loss": 0.19140062034130095, "time": 0.8981546640396119, "epoch": 2, "memory": 6319, "step": 4307} +{"lr": 9.72052879933172e-05, "data_time": 0.8162017345428467, "loss": 0.18137903064489364, "time": 0.9555898427963256, "epoch": 2, "memory": 6319, "step": 4407} +{"lr": 9.707839525363358e-05, "data_time": 0.7476181507110595, "loss": 0.18049391508102416, "time": 0.8879045963287353, "epoch": 2, "memory": 6319, "step": 4507} +{"lr": 9.69487818822646e-05, "data_time": 0.7870512962341308, "loss": 0.18146572560071944, "time": 0.9267127990722657, "epoch": 2, "memory": 6319, "step": 4607} +{"lr": 9.681645625952616e-05, "data_time": 0.7730315685272217, "loss": 0.16808514446020126, "time": 0.9145148754119873, "epoch": 2, "memory": 6319, "step": 4707} +{"lr": 9.668142694109826e-05, "data_time": 0.9281142950057983, "loss": 0.17594008296728134, "time": 1.0678827047348023, "epoch": 2, "memory": 6319, "step": 4807} +{"lr": 9.654370265747162e-05, "data_time": 0.7081193685531616, "loss": 0.16326686292886733, "time": 0.8481757640838623, "epoch": 2, "memory": 6319, "step": 4907} +{"lr": 9.640329231338345e-05, "data_time": 0.7069461345672607, "loss": 0.18151011168956757, "time": 0.8467189550399781, "epoch": 2, "memory": 6319, "step": 5007} +{"lr": 9.626020498724147e-05, "data_time": 0.7017974615097046, "loss": 0.16655913591384888, "time": 0.8423813581466675, "epoch": 2, "memory": 6319, "step": 5107} +{"lr": 9.611444993053717e-05, "data_time": 0.7469312667846679, "loss": 0.1599966548383236, "time": 0.8867108583450317, "epoch": 2, "memory": 6319, "step": 5207} +{"lr": 9.596603656724746e-05, "data_time": 0.7863739013671875, "loss": 0.16251930892467498, "time": 0.9277769804000855, "epoch": 2, "memory": 6319, "step": 5307} +{"lr": 9.58149744932254e-05, "data_time": 0.679094934463501, "loss": 0.1750597044825554, "time": 0.8200845718383789, "epoch": 2, "memory": 6319, "step": 5407} +{"lr": 9.566127347557972e-05, "data_time": 0.684056305885315, "loss": 0.17173441201448442, "time": 0.8245492935180664, "epoch": 2, "memory": 6319, "step": 5507} +{"lr": 9.550494345204367e-05, "data_time": 0.7233998298645019, "loss": 0.15493762791156768, "time": 0.8651554107666015, "epoch": 2, "memory": 6319, "step": 5607} +{"lr": 9.534599453033183e-05, "data_time": 0.8133837461471558, "loss": 0.15334631204605104, "time": 0.9550534248352051, "epoch": 2, "memory": 6319, "step": 5707} diff --git a/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/20230604_225234/vis_data/vis_image/0000000.png_1.png b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/20230604_225234/vis_data/vis_image/0000000.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..4e5d3d392f15b23c6c0a3debdc2e0db97137d347 Binary files /dev/null and b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/20230604_225234/vis_data/vis_image/0000000.png_1.png differ diff --git a/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/20230604_225234/vis_data/vis_image/000005174.jpg_1.png b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/20230604_225234/vis_data/vis_image/000005174.jpg_1.png new file mode 100644 index 0000000000000000000000000000000000000000..a8af0ea5b4cd06b9d326d10dfb617177a3c03041 Binary files /dev/null and b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/20230604_225234/vis_data/vis_image/000005174.jpg_1.png differ diff --git a/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/20230604_225234/vis_data/vis_image/0010000.png_1.png b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/20230604_225234/vis_data/vis_image/0010000.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..006c648530a5e8899238bbdb6a146e807e6505b5 Binary files /dev/null and b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/20230604_225234/vis_data/vis_image/0010000.png_1.png differ diff --git a/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/20230605_002310/20230605_002310.log b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/20230605_002310/20230605_002310.log new file mode 100644 index 0000000000000000000000000000000000000000..4938c3da0bbd12241e6a5d34fd8231c2bf10ce3f --- /dev/null +++ b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/20230605_002310/20230605_002310.log @@ -0,0 +1,862 @@ +2023/06/05 00:23:13 - mmengine - INFO - +------------------------------------------------------------ +System environment: + sys.platform: linux + Python: 3.10.9 (main, Mar 8 2023, 10:47:38) [GCC 11.2.0] + CUDA available: True + numpy_random_seed: 1027137554 + GPU 0,1: NVIDIA A100-SXM4-80GB + CUDA_HOME: /mnt/petrelfs/share/cuda-11.6 + NVCC: Cuda compilation tools, release 11.6, V11.6.124 + GCC: gcc (GCC) 7.5.0 + PyTorch: 1.13.1 + PyTorch compiling details: PyTorch built with: + - GCC 9.3 + - C++ Version: 201402 + - Intel(R) oneAPI Math Kernel Library Version 2021.4-Product Build 20210904 for Intel(R) 64 architecture applications + - Intel(R) MKL-DNN v2.6.0 (Git Hash 52b5f107dd9cf10910aaa19cb47f3abf9b349815) + - OpenMP 201511 (a.k.a. OpenMP 4.5) + - LAPACK is enabled (usually provided by MKL) + - NNPACK is enabled + - CPU capability usage: AVX2 + - CUDA Runtime 11.6 + - NVCC architecture flags: -gencode;arch=compute_37,code=sm_37;-gencode;arch=compute_50,code=sm_50;-gencode;arch=compute_60,code=sm_60;-gencode;arch=compute_61,code=sm_61;-gencode;arch=compute_70,code=sm_70;-gencode;arch=compute_75,code=sm_75;-gencode;arch=compute_80,code=sm_80;-gencode;arch=compute_86,code=sm_86;-gencode;arch=compute_37,code=compute_37 + - CuDNN 8.3.2 (built against CUDA 11.5) + - Magma 2.6.1 + - Build settings: BLAS_INFO=mkl, BUILD_TYPE=Release, CUDA_VERSION=11.6, CUDNN_VERSION=8.3.2, CXX_COMPILER=/opt/rh/devtoolset-9/root/usr/bin/c++, CXX_FLAGS= -fabi-version=11 -Wno-deprecated -fvisibility-inlines-hidden -DUSE_PTHREADPOOL -fopenmp -DNDEBUG -DUSE_KINETO -DUSE_FBGEMM -DUSE_QNNPACK -DUSE_PYTORCH_QNNPACK -DUSE_XNNPACK -DSYMBOLICATE_MOBILE_DEBUG_HANDLE -DEDGE_PROFILER_USE_KINETO -O2 -fPIC -Wno-narrowing -Wall -Wextra -Werror=return-type -Werror=non-virtual-dtor -Wno-missing-field-initializers -Wno-type-limits -Wno-array-bounds -Wno-unknown-pragmas -Wunused-local-typedefs -Wno-unused-parameter -Wno-unused-function -Wno-unused-result -Wno-strict-overflow -Wno-strict-aliasing -Wno-error=deprecated-declarations -Wno-stringop-overflow -Wno-psabi -Wno-error=pedantic -Wno-error=redundant-decls -Wno-error=old-style-cast -fdiagnostics-color=always -faligned-new -Wno-unused-but-set-variable -Wno-maybe-uninitialized -fno-math-errno -fno-trapping-math -Werror=format -Werror=cast-function-type -Wno-stringop-overflow, LAPACK_INFO=mkl, PERF_WITH_AVX=1, PERF_WITH_AVX2=1, PERF_WITH_AVX512=1, TORCH_VERSION=1.13.1, USE_CUDA=ON, USE_CUDNN=ON, USE_EXCEPTION_PTR=1, USE_GFLAGS=OFF, USE_GLOG=OFF, USE_MKL=ON, USE_MKLDNN=ON, USE_MPI=OFF, USE_NCCL=ON, USE_NNPACK=ON, USE_OPENMP=ON, USE_ROCM=OFF, + + TorchVision: 0.14.1 + OpenCV: 4.7.0 + MMEngine: 0.7.3 + +Runtime environment: + cudnn_benchmark: True + mp_cfg: {'mp_start_method': 'fork', 'opencv_num_threads': 0} + dist_cfg: {'backend': 'nccl'} + seed: None + deterministic: False + Distributed launcher: slurm + Distributed training: True + GPU number: 2 +------------------------------------------------------------ + +2023/06/05 00:23:18 - mmengine - INFO - Config: +optim_wrapper = dict( + optimizer=dict( + type='SGD', + lr=0.0001, + momentum=0.9, + weight_decay=0.0001, + _scope_='mmpretrain'), + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=False, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=512) +model = dict( + type='ImageClassifier', + backbone=dict( + frozen_stages=2, + type='ResNet', + depth=50, + num_stages=4, + out_indices=(3, ), + style='pytorch', + init_cfg=dict( + type='Pretrained', + checkpoint= + 'https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth', + prefix='backbone')), + neck=dict(type='GlobalAveragePooling'), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=2048, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + topk=1)) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stablediffusionV1-5R2-dpmsolver-25-1m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/cc1m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = dict(type='Accuracy', topk=1) +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = dict(type='Accuracy', topk=1) +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1' + +2023/06/05 00:23:31 - mmengine - INFO - Hooks will be executed in the following order: +before_run: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_load_checkpoint: +(ABOVE_NORMAL) EMAHook + -------------------- +before_train: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_train_epoch: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(NORMAL ) DistSamplerSeedHook + -------------------- +before_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook + -------------------- +after_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +after_train_epoch: +(NORMAL ) IterTimerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_val_epoch: +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook + -------------------- +before_val_iter: +(NORMAL ) IterTimerHook + -------------------- +after_val_iter: +(NORMAL ) IterTimerHook +(NORMAL ) VisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_val_epoch: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_save_checkpoint: +(ABOVE_NORMAL) EMAHook + -------------------- +after_train: +(VERY_LOW ) CheckpointHook + -------------------- +before_test_epoch: +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook + -------------------- +before_test_iter: +(NORMAL ) IterTimerHook + -------------------- +after_test_iter: +(NORMAL ) IterTimerHook +(NORMAL ) VisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_test_epoch: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_run: +(BELOW_NORMAL) LoggerHook + -------------------- +2023/06/05 00:23:51 - mmengine - INFO - load backbone in model from: https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth +Name of parameter - Initialization information + +backbone.conv1.weight - torch.Size([64, 3, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.bn1.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.bn1.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.conv1.weight - torch.Size([64, 64, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.bn1.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.bn1.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.conv2.weight - torch.Size([64, 64, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.bn2.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.bn2.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.conv3.weight - torch.Size([256, 64, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.bn3.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.bn3.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.downsample.0.weight - torch.Size([256, 64, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.downsample.1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.downsample.1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.conv1.weight - torch.Size([64, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.bn1.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.bn1.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.conv2.weight - torch.Size([64, 64, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.bn2.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.bn2.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.conv3.weight - torch.Size([256, 64, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.bn3.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.bn3.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.conv1.weight - torch.Size([64, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.bn1.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.bn1.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.conv2.weight - torch.Size([64, 64, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.bn2.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.bn2.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.conv3.weight - torch.Size([256, 64, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.bn3.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.bn3.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.conv1.weight - torch.Size([128, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.bn1.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.bn1.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.conv2.weight - torch.Size([128, 128, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.bn2.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.bn2.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.conv3.weight - torch.Size([512, 128, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.bn3.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.bn3.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.downsample.0.weight - torch.Size([512, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.downsample.1.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.downsample.1.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.conv1.weight - torch.Size([128, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.bn1.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.bn1.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.conv2.weight - torch.Size([128, 128, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.bn2.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.bn2.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.conv3.weight - torch.Size([512, 128, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.bn3.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.bn3.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.conv1.weight - torch.Size([128, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.bn1.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.bn1.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.conv2.weight - torch.Size([128, 128, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.bn2.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.bn2.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.conv3.weight - torch.Size([512, 128, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.bn3.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.bn3.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.conv1.weight - torch.Size([128, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.bn1.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.bn1.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.conv2.weight - torch.Size([128, 128, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.bn2.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.bn2.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.conv3.weight - torch.Size([512, 128, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.bn3.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.bn3.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.conv1.weight - torch.Size([256, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.bn1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.bn1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.conv2.weight - torch.Size([256, 256, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.bn2.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.bn2.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.conv3.weight - torch.Size([1024, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.bn3.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.bn3.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.downsample.0.weight - torch.Size([1024, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.downsample.1.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.downsample.1.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.conv1.weight - torch.Size([256, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.bn1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.bn1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.conv2.weight - torch.Size([256, 256, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.bn2.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.bn2.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.conv3.weight - torch.Size([1024, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.bn3.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.bn3.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.conv1.weight - torch.Size([256, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.bn1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.bn1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.conv2.weight - torch.Size([256, 256, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.bn2.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.bn2.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.conv3.weight - torch.Size([1024, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.bn3.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.bn3.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.conv1.weight - torch.Size([256, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.bn1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.bn1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.conv2.weight - torch.Size([256, 256, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.bn2.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.bn2.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.conv3.weight - torch.Size([1024, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.bn3.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.bn3.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.conv1.weight - torch.Size([256, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.bn1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.bn1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.conv2.weight - torch.Size([256, 256, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.bn2.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.bn2.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.conv3.weight - torch.Size([1024, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.bn3.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.bn3.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.conv1.weight - torch.Size([256, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.bn1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.bn1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.conv2.weight - torch.Size([256, 256, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.bn2.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.bn2.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.conv3.weight - torch.Size([1024, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.bn3.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.bn3.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.conv1.weight - torch.Size([512, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.bn1.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.bn1.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.conv2.weight - torch.Size([512, 512, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.bn2.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.bn2.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.conv3.weight - torch.Size([2048, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.bn3.weight - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.bn3.bias - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.downsample.0.weight - torch.Size([2048, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.downsample.1.weight - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.downsample.1.bias - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.conv1.weight - torch.Size([512, 2048, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.bn1.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.bn1.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.conv2.weight - torch.Size([512, 512, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.bn2.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.bn2.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.conv3.weight - torch.Size([2048, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.bn3.weight - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.bn3.bias - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.conv1.weight - torch.Size([512, 2048, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.bn1.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.bn1.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.conv2.weight - torch.Size([512, 512, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.bn2.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.bn2.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.conv3.weight - torch.Size([2048, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.bn3.weight - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.bn3.bias - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +head.fc.weight - torch.Size([2, 2048]): +NormalInit: mean=0, std=0.01, bias=0 + +head.fc.bias - torch.Size([2]): +NormalInit: mean=0, std=0.01, bias=0 +2023/06/05 00:23:51 - mmengine - WARNING - "FileClient" will be deprecated in future. Please use io functions in https://mmengine.readthedocs.io/en/latest/api/fileio.html#file-io +2023/06/05 00:23:51 - mmengine - WARNING - "HardDiskBackend" is the alias of "LocalBackend" and the former will be deprecated in future. +2023/06/05 00:23:51 - mmengine - INFO - Checkpoints will be saved to /mnt/petrelfs/luzeyu/workspace/fakebench/mmpretrain/workdir/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1. +2023/06/05 00:25:25 - mmengine - INFO - Epoch(train) [1][ 100/3907] lr: 9.9999e-05 eta: 10:08:59 time: 0.9031 data_time: 0.6710 memory: 9436 loss: 0.6090 +2023/06/05 00:26:56 - mmengine - INFO - Epoch(train) [1][ 200/3907] lr: 9.9994e-05 eta: 9:57:22 time: 0.9145 data_time: 0.7492 memory: 6319 loss: 0.5507 +2023/06/05 00:28:26 - mmengine - INFO - Epoch(train) [1][ 300/3907] lr: 9.9987e-05 eta: 9:50:41 time: 0.8488 data_time: 0.7060 memory: 6319 loss: 0.5090 +2023/06/05 00:29:53 - mmengine - INFO - Epoch(train) [1][ 400/3907] lr: 9.9977e-05 eta: 9:42:18 time: 0.8746 data_time: 0.6556 memory: 6319 loss: 0.4677 +2023/06/05 00:31:23 - mmengine - INFO - Epoch(train) [1][ 500/3907] lr: 9.9964e-05 eta: 9:40:56 time: 1.1397 data_time: 0.7713 memory: 6319 loss: 0.4235 +2023/06/05 00:33:00 - mmengine - INFO - Epoch(train) [1][ 600/3907] lr: 9.9948e-05 eta: 9:45:47 time: 0.8369 data_time: 0.4890 memory: 6319 loss: 0.3909 +2023/06/05 00:34:23 - mmengine - INFO - Epoch(train) [1][ 700/3907] lr: 9.9929e-05 eta: 9:37:03 time: 0.8248 data_time: 0.2733 memory: 6319 loss: 0.3861 +2023/06/05 00:35:50 - mmengine - INFO - Epoch(train) [1][ 800/3907] lr: 9.9907e-05 eta: 9:33:15 time: 0.8611 data_time: 0.1454 memory: 6319 loss: 0.3719 +2023/06/05 00:37:20 - mmengine - INFO - Epoch(train) [1][ 900/3907] lr: 9.9882e-05 eta: 9:31:45 time: 0.9318 data_time: 0.1474 memory: 6319 loss: 0.3623 +2023/06/05 00:38:49 - mmengine - INFO - Exp name: resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1_20230605_002310 +2023/06/05 00:38:49 - mmengine - INFO - Epoch(train) [1][1000/3907] lr: 9.9855e-05 eta: 9:29:14 time: 0.9197 data_time: 0.0009 memory: 6319 loss: 0.3439 +2023/06/05 00:40:15 - mmengine - INFO - Epoch(train) [1][1100/3907] lr: 9.9824e-05 eta: 9:25:54 time: 0.8668 data_time: 0.0666 memory: 6319 loss: 0.3320 +2023/06/05 00:41:52 - mmengine - INFO - Epoch(train) [1][1200/3907] lr: 9.9791e-05 eta: 9:28:27 time: 1.0129 data_time: 0.0009 memory: 6319 loss: 0.3154 +2023/06/05 00:43:21 - mmengine - INFO - Epoch(train) [1][1300/3907] lr: 9.9755e-05 eta: 9:26:08 time: 0.7945 data_time: 0.2863 memory: 6319 loss: 0.3001 +2023/06/05 00:44:47 - mmengine - INFO - Epoch(train) [1][1400/3907] lr: 9.9716e-05 eta: 9:23:13 time: 0.8691 data_time: 0.2437 memory: 6319 loss: 0.2982 +2023/06/05 00:46:15 - mmengine - INFO - Epoch(train) [1][1500/3907] lr: 9.9674e-05 eta: 9:20:44 time: 0.8481 data_time: 0.0349 memory: 6319 loss: 0.2839 diff --git a/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/20230605_002310/vis_data/20230605_002310.json b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/20230605_002310/vis_data/20230605_002310.json new file mode 100644 index 0000000000000000000000000000000000000000..86a613e8a19a7af117980b0c726e936760e3464c --- /dev/null +++ b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/20230605_002310/vis_data/20230605_002310.json @@ -0,0 +1,15 @@ +{"lr": 9.999857418382631e-05, "data_time": 0.6710198879241943, "loss": 0.6089565873146057, "time": 0.9030874490737915, "epoch": 1, "memory": 9436, "step": 100} +{"lr": 9.999423907359506e-05, "data_time": 0.7491933345794678, "loss": 0.5506765961647033, "time": 0.9145426750183105, "epoch": 1, "memory": 6319, "step": 200} +{"lr": 9.998699480412073e-05, "data_time": 0.7059998273849487, "loss": 0.5089826732873917, "time": 0.848846435546875, "epoch": 1, "memory": 6319, "step": 300} +{"lr": 9.997684184379085e-05, "data_time": 0.6556185245513916, "loss": 0.46768696010112765, "time": 0.8745664596557617, "epoch": 1, "memory": 6319, "step": 400} +{"lr": 9.996378084905784e-05, "data_time": 0.7712557077407837, "loss": 0.42347240149974824, "time": 1.1397310495376587, "epoch": 1, "memory": 6319, "step": 500} +{"lr": 9.994781266439667e-05, "data_time": 0.4890120267868042, "loss": 0.3909292876720428, "time": 0.8369115829467774, "epoch": 1, "memory": 6319, "step": 600} +{"lr": 9.992893832225061e-05, "data_time": 0.27330541610717773, "loss": 0.3860609769821167, "time": 0.8247616052627563, "epoch": 1, "memory": 6319, "step": 700} +{"lr": 9.990715904296405e-05, "data_time": 0.14538750648498536, "loss": 0.37194141149520876, "time": 0.8610518932342529, "epoch": 1, "memory": 6319, "step": 800} +{"lr": 9.988247623470371e-05, "data_time": 0.14744775295257567, "loss": 0.3623491764068604, "time": 0.9318323135375977, "epoch": 1, "memory": 6319, "step": 900} +{"lr": 9.985489149336749e-05, "data_time": 0.0008916378021240235, "loss": 0.34388838708400726, "time": 0.9196843862533569, "epoch": 1, "memory": 6319, "step": 1000} +{"lr": 9.982440660248187e-05, "data_time": 0.06658101081848145, "loss": 0.3320307433605194, "time": 0.86680588722229, "epoch": 1, "memory": 6319, "step": 1100} +{"lr": 9.979102353308574e-05, "data_time": 0.0008611917495727539, "loss": 0.31542661488056184, "time": 1.0129417657852173, "epoch": 1, "memory": 6319, "step": 1200} +{"lr": 9.975474444360358e-05, "data_time": 0.286257004737854, "loss": 0.3000659987330437, "time": 0.794460129737854, "epoch": 1, "memory": 6319, "step": 1300} +{"lr": 9.97155716797056e-05, "data_time": 0.2437370777130127, "loss": 0.2982131540775299, "time": 0.8690698862075805, "epoch": 1, "memory": 6319, "step": 1400} +{"lr": 9.967350777415632e-05, "data_time": 0.034885597229003903, "loss": 0.28386528193950655, "time": 0.8481145858764648, "epoch": 1, "memory": 6319, "step": 1500} diff --git a/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/20230605_002310/vis_data/config.py b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/20230605_002310/vis_data/config.py new file mode 100644 index 0000000000000000000000000000000000000000..661bf63adae66645caa3af55d05b1778159f197c --- /dev/null +++ b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/20230605_002310/vis_data/config.py @@ -0,0 +1,230 @@ +optim_wrapper = dict( + optimizer=dict( + type='SGD', + lr=0.0001, + momentum=0.9, + weight_decay=0.0001, + _scope_='mmpretrain'), + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=False, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=512) +model = dict( + type='ImageClassifier', + backbone=dict( + frozen_stages=2, + type='ResNet', + depth=50, + num_stages=4, + out_indices=(3, ), + style='pytorch', + init_cfg=dict( + type='Pretrained', + checkpoint= + 'https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth', + prefix='backbone')), + neck=dict(type='GlobalAveragePooling'), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=2048, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + topk=1)) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stablediffusionV1-5R2-dpmsolver-25-1m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/cc1m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = dict(type='Accuracy', topk=1) +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = dict(type='Accuracy', topk=1) +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1' diff --git a/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/20230605_002310/vis_data/events.out.tfevents.1685895805.SH-IDC1-10-140-24-17.3259.0 b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/20230605_002310/vis_data/events.out.tfevents.1685895805.SH-IDC1-10-140-24-17.3259.0 new file mode 100644 index 0000000000000000000000000000000000000000..b3cf8898c790851a27b860ddbbfdc8120dad98f3 --- /dev/null +++ b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/20230605_002310/vis_data/events.out.tfevents.1685895805.SH-IDC1-10-140-24-17.3259.0 @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:208e3ba1295c5df493b1b14d6fa27d033e0a3b27fc486702eb9cb8f1d145c37e +size 12739 diff --git a/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/20230605_002310/vis_data/scalars.json b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/20230605_002310/vis_data/scalars.json new file mode 100644 index 0000000000000000000000000000000000000000..86a613e8a19a7af117980b0c726e936760e3464c --- /dev/null +++ b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/20230605_002310/vis_data/scalars.json @@ -0,0 +1,15 @@ +{"lr": 9.999857418382631e-05, "data_time": 0.6710198879241943, "loss": 0.6089565873146057, "time": 0.9030874490737915, "epoch": 1, "memory": 9436, "step": 100} +{"lr": 9.999423907359506e-05, "data_time": 0.7491933345794678, "loss": 0.5506765961647033, "time": 0.9145426750183105, "epoch": 1, "memory": 6319, "step": 200} +{"lr": 9.998699480412073e-05, "data_time": 0.7059998273849487, "loss": 0.5089826732873917, "time": 0.848846435546875, "epoch": 1, "memory": 6319, "step": 300} +{"lr": 9.997684184379085e-05, "data_time": 0.6556185245513916, "loss": 0.46768696010112765, "time": 0.8745664596557617, "epoch": 1, "memory": 6319, "step": 400} +{"lr": 9.996378084905784e-05, "data_time": 0.7712557077407837, "loss": 0.42347240149974824, "time": 1.1397310495376587, "epoch": 1, "memory": 6319, "step": 500} +{"lr": 9.994781266439667e-05, "data_time": 0.4890120267868042, "loss": 0.3909292876720428, "time": 0.8369115829467774, "epoch": 1, "memory": 6319, "step": 600} +{"lr": 9.992893832225061e-05, "data_time": 0.27330541610717773, "loss": 0.3860609769821167, "time": 0.8247616052627563, "epoch": 1, "memory": 6319, "step": 700} +{"lr": 9.990715904296405e-05, "data_time": 0.14538750648498536, "loss": 0.37194141149520876, "time": 0.8610518932342529, "epoch": 1, "memory": 6319, "step": 800} +{"lr": 9.988247623470371e-05, "data_time": 0.14744775295257567, "loss": 0.3623491764068604, "time": 0.9318323135375977, "epoch": 1, "memory": 6319, "step": 900} +{"lr": 9.985489149336749e-05, "data_time": 0.0008916378021240235, "loss": 0.34388838708400726, "time": 0.9196843862533569, "epoch": 1, "memory": 6319, "step": 1000} +{"lr": 9.982440660248187e-05, "data_time": 0.06658101081848145, "loss": 0.3320307433605194, "time": 0.86680588722229, "epoch": 1, "memory": 6319, "step": 1100} +{"lr": 9.979102353308574e-05, "data_time": 0.0008611917495727539, "loss": 0.31542661488056184, "time": 1.0129417657852173, "epoch": 1, "memory": 6319, "step": 1200} +{"lr": 9.975474444360358e-05, "data_time": 0.286257004737854, "loss": 0.3000659987330437, "time": 0.794460129737854, "epoch": 1, "memory": 6319, "step": 1300} +{"lr": 9.97155716797056e-05, "data_time": 0.2437370777130127, "loss": 0.2982131540775299, "time": 0.8690698862075805, "epoch": 1, "memory": 6319, "step": 1400} +{"lr": 9.967350777415632e-05, "data_time": 0.034885597229003903, "loss": 0.28386528193950655, "time": 0.8481145858764648, "epoch": 1, "memory": 6319, "step": 1500} diff --git a/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/20230605_005004/20230605_005004.log b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/20230605_005004/20230605_005004.log new file mode 100644 index 0000000000000000000000000000000000000000..0507e92ed342b5417b007e7fb5cfe1e7417db1ba --- /dev/null +++ b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/20230605_005004/20230605_005004.log @@ -0,0 +1,1305 @@ +2023/06/05 00:50:07 - mmengine - INFO - +------------------------------------------------------------ +System environment: + sys.platform: linux + Python: 3.10.9 (main, Mar 8 2023, 10:47:38) [GCC 11.2.0] + CUDA available: True + numpy_random_seed: 1816270783 + GPU 0,1: NVIDIA A100-SXM4-80GB + CUDA_HOME: /mnt/petrelfs/share/cuda-11.6 + NVCC: Cuda compilation tools, release 11.6, V11.6.124 + GCC: gcc (GCC) 7.5.0 + PyTorch: 1.13.1 + PyTorch compiling details: PyTorch built with: + - GCC 9.3 + - C++ Version: 201402 + - Intel(R) oneAPI Math Kernel Library Version 2021.4-Product Build 20210904 for Intel(R) 64 architecture applications + - Intel(R) MKL-DNN v2.6.0 (Git Hash 52b5f107dd9cf10910aaa19cb47f3abf9b349815) + - OpenMP 201511 (a.k.a. OpenMP 4.5) + - LAPACK is enabled (usually provided by MKL) + - NNPACK is enabled + - CPU capability usage: AVX2 + - CUDA Runtime 11.6 + - NVCC architecture flags: -gencode;arch=compute_37,code=sm_37;-gencode;arch=compute_50,code=sm_50;-gencode;arch=compute_60,code=sm_60;-gencode;arch=compute_61,code=sm_61;-gencode;arch=compute_70,code=sm_70;-gencode;arch=compute_75,code=sm_75;-gencode;arch=compute_80,code=sm_80;-gencode;arch=compute_86,code=sm_86;-gencode;arch=compute_37,code=compute_37 + - CuDNN 8.3.2 (built against CUDA 11.5) + - Magma 2.6.1 + - Build settings: BLAS_INFO=mkl, BUILD_TYPE=Release, CUDA_VERSION=11.6, CUDNN_VERSION=8.3.2, CXX_COMPILER=/opt/rh/devtoolset-9/root/usr/bin/c++, CXX_FLAGS= -fabi-version=11 -Wno-deprecated -fvisibility-inlines-hidden -DUSE_PTHREADPOOL -fopenmp -DNDEBUG -DUSE_KINETO -DUSE_FBGEMM -DUSE_QNNPACK -DUSE_PYTORCH_QNNPACK -DUSE_XNNPACK -DSYMBOLICATE_MOBILE_DEBUG_HANDLE -DEDGE_PROFILER_USE_KINETO -O2 -fPIC -Wno-narrowing -Wall -Wextra -Werror=return-type -Werror=non-virtual-dtor -Wno-missing-field-initializers -Wno-type-limits -Wno-array-bounds -Wno-unknown-pragmas -Wunused-local-typedefs -Wno-unused-parameter -Wno-unused-function -Wno-unused-result -Wno-strict-overflow -Wno-strict-aliasing -Wno-error=deprecated-declarations -Wno-stringop-overflow -Wno-psabi -Wno-error=pedantic -Wno-error=redundant-decls -Wno-error=old-style-cast -fdiagnostics-color=always -faligned-new -Wno-unused-but-set-variable -Wno-maybe-uninitialized -fno-math-errno -fno-trapping-math -Werror=format -Werror=cast-function-type -Wno-stringop-overflow, LAPACK_INFO=mkl, PERF_WITH_AVX=1, PERF_WITH_AVX2=1, PERF_WITH_AVX512=1, TORCH_VERSION=1.13.1, USE_CUDA=ON, USE_CUDNN=ON, USE_EXCEPTION_PTR=1, USE_GFLAGS=OFF, USE_GLOG=OFF, USE_MKL=ON, USE_MKLDNN=ON, USE_MPI=OFF, USE_NCCL=ON, USE_NNPACK=ON, USE_OPENMP=ON, USE_ROCM=OFF, + + TorchVision: 0.14.1 + OpenCV: 4.7.0 + MMEngine: 0.7.3 + +Runtime environment: + cudnn_benchmark: True + mp_cfg: {'mp_start_method': 'fork', 'opencv_num_threads': 0} + dist_cfg: {'backend': 'nccl'} + seed: None + deterministic: False + Distributed launcher: slurm + Distributed training: True + GPU number: 2 +------------------------------------------------------------ + +2023/06/05 00:50:13 - mmengine - INFO - Config: +optim_wrapper = dict( + optimizer=dict( + type='SGD', + lr=0.0001, + momentum=0.9, + weight_decay=0.0001, + _scope_='mmpretrain'), + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=False, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=512) +model = dict( + type='ImageClassifier', + backbone=dict( + frozen_stages=2, + type='ResNet', + depth=50, + num_stages=4, + out_indices=(3, ), + style='pytorch', + init_cfg=dict( + type='Pretrained', + checkpoint= + 'https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth', + prefix='backbone')), + neck=dict(type='GlobalAveragePooling'), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=2048, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + topk=1)) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stablediffusionV1-5R2-dpmsolver-25-1m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/cc1m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = dict(type='Accuracy', topk=1) +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = dict(type='Accuracy', topk=1) +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1' + +2023/06/05 00:50:27 - mmengine - INFO - Hooks will be executed in the following order: +before_run: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_load_checkpoint: +(ABOVE_NORMAL) EMAHook + -------------------- +before_train: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_train_epoch: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(NORMAL ) DistSamplerSeedHook + -------------------- +before_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook + -------------------- +after_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +after_train_epoch: +(NORMAL ) IterTimerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_val_epoch: +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook + -------------------- +before_val_iter: +(NORMAL ) IterTimerHook + -------------------- +after_val_iter: +(NORMAL ) IterTimerHook +(NORMAL ) VisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_val_epoch: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_save_checkpoint: +(ABOVE_NORMAL) EMAHook + -------------------- +after_train: +(VERY_LOW ) CheckpointHook + -------------------- +before_test_epoch: +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook + -------------------- +before_test_iter: +(NORMAL ) IterTimerHook + -------------------- +after_test_iter: +(NORMAL ) IterTimerHook +(NORMAL ) VisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_test_epoch: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_run: +(BELOW_NORMAL) LoggerHook + -------------------- +2023/06/05 00:50:48 - mmengine - INFO - load backbone in model from: https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth +Name of parameter - Initialization information + +backbone.conv1.weight - torch.Size([64, 3, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.bn1.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.bn1.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.conv1.weight - torch.Size([64, 64, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.bn1.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.bn1.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.conv2.weight - torch.Size([64, 64, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.bn2.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.bn2.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.conv3.weight - torch.Size([256, 64, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.bn3.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.bn3.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.downsample.0.weight - torch.Size([256, 64, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.downsample.1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.downsample.1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.conv1.weight - torch.Size([64, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.bn1.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.bn1.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.conv2.weight - torch.Size([64, 64, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.bn2.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.bn2.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.conv3.weight - torch.Size([256, 64, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.bn3.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.bn3.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.conv1.weight - torch.Size([64, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.bn1.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.bn1.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.conv2.weight - torch.Size([64, 64, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.bn2.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.bn2.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.conv3.weight - torch.Size([256, 64, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.bn3.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.bn3.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.conv1.weight - torch.Size([128, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.bn1.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.bn1.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.conv2.weight - torch.Size([128, 128, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.bn2.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.bn2.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.conv3.weight - torch.Size([512, 128, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.bn3.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.bn3.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.downsample.0.weight - torch.Size([512, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.downsample.1.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.downsample.1.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.conv1.weight - torch.Size([128, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.bn1.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.bn1.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.conv2.weight - torch.Size([128, 128, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.bn2.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.bn2.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.conv3.weight - torch.Size([512, 128, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.bn3.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.bn3.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.conv1.weight - torch.Size([128, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.bn1.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.bn1.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.conv2.weight - torch.Size([128, 128, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.bn2.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.bn2.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.conv3.weight - torch.Size([512, 128, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.bn3.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.bn3.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.conv1.weight - torch.Size([128, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.bn1.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.bn1.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.conv2.weight - torch.Size([128, 128, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.bn2.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.bn2.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.conv3.weight - torch.Size([512, 128, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.bn3.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.bn3.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.conv1.weight - torch.Size([256, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.bn1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.bn1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.conv2.weight - torch.Size([256, 256, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.bn2.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.bn2.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.conv3.weight - torch.Size([1024, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.bn3.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.bn3.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.downsample.0.weight - torch.Size([1024, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.downsample.1.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.downsample.1.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.conv1.weight - torch.Size([256, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.bn1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.bn1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.conv2.weight - torch.Size([256, 256, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.bn2.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.bn2.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.conv3.weight - torch.Size([1024, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.bn3.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.bn3.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.conv1.weight - torch.Size([256, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.bn1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.bn1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.conv2.weight - torch.Size([256, 256, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.bn2.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.bn2.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.conv3.weight - torch.Size([1024, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.bn3.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.bn3.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.conv1.weight - torch.Size([256, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.bn1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.bn1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.conv2.weight - torch.Size([256, 256, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.bn2.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.bn2.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.conv3.weight - torch.Size([1024, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.bn3.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.bn3.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.conv1.weight - torch.Size([256, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.bn1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.bn1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.conv2.weight - torch.Size([256, 256, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.bn2.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.bn2.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.conv3.weight - torch.Size([1024, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.bn3.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.bn3.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.conv1.weight - torch.Size([256, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.bn1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.bn1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.conv2.weight - torch.Size([256, 256, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.bn2.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.bn2.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.conv3.weight - torch.Size([1024, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.bn3.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.bn3.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.conv1.weight - torch.Size([512, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.bn1.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.bn1.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.conv2.weight - torch.Size([512, 512, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.bn2.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.bn2.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.conv3.weight - torch.Size([2048, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.bn3.weight - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.bn3.bias - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.downsample.0.weight - torch.Size([2048, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.downsample.1.weight - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.downsample.1.bias - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.conv1.weight - torch.Size([512, 2048, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.bn1.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.bn1.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.conv2.weight - torch.Size([512, 512, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.bn2.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.bn2.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.conv3.weight - torch.Size([2048, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.bn3.weight - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.bn3.bias - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.conv1.weight - torch.Size([512, 2048, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.bn1.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.bn1.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.conv2.weight - torch.Size([512, 512, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.bn2.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.bn2.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.conv3.weight - torch.Size([2048, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.bn3.weight - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.bn3.bias - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +head.fc.weight - torch.Size([2, 2048]): +NormalInit: mean=0, std=0.01, bias=0 + +head.fc.bias - torch.Size([2]): +NormalInit: mean=0, std=0.01, bias=0 +2023/06/05 00:50:48 - mmengine - WARNING - "FileClient" will be deprecated in future. Please use io functions in https://mmengine.readthedocs.io/en/latest/api/fileio.html#file-io +2023/06/05 00:50:48 - mmengine - WARNING - "HardDiskBackend" is the alias of "LocalBackend" and the former will be deprecated in future. +2023/06/05 00:50:48 - mmengine - INFO - Checkpoints will be saved to /mnt/petrelfs/luzeyu/workspace/fakebench/mmpretrain/workdir/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1. +2023/06/05 00:52:34 - mmengine - INFO - Epoch(train) [1][ 100/3907] lr: 9.9999e-05 eta: 11:32:09 time: 1.3435 data_time: 0.9104 memory: 9436 loss: 0.6140 +2023/06/05 00:54:32 - mmengine - INFO - Epoch(train) [1][ 200/3907] lr: 9.9994e-05 eta: 12:05:21 time: 0.8297 data_time: 0.5141 memory: 6319 loss: 0.5577 +2023/06/05 00:55:59 - mmengine - INFO - Epoch(train) [1][ 300/3907] lr: 9.9987e-05 eta: 11:10:58 time: 0.8844 data_time: 0.3725 memory: 6319 loss: 0.5053 +2023/06/05 00:57:30 - mmengine - INFO - Epoch(train) [1][ 400/3907] lr: 9.9977e-05 eta: 10:48:03 time: 0.9432 data_time: 0.2932 memory: 6319 loss: 0.4581 +2023/06/05 00:59:01 - mmengine - INFO - Epoch(train) [1][ 500/3907] lr: 9.9964e-05 eta: 10:33:41 time: 0.8859 data_time: 0.4679 memory: 6319 loss: 0.4414 +2023/06/05 01:00:33 - mmengine - INFO - Epoch(train) [1][ 600/3907] lr: 9.9948e-05 eta: 10:25:11 time: 0.8934 data_time: 0.6340 memory: 6319 loss: 0.4045 +2023/06/05 01:02:03 - mmengine - INFO - Epoch(train) [1][ 700/3907] lr: 9.9929e-05 eta: 10:16:23 time: 0.9070 data_time: 0.4763 memory: 6319 loss: 0.3901 +2023/06/05 01:03:35 - mmengine - INFO - Epoch(train) [1][ 800/3907] lr: 9.9907e-05 eta: 10:11:49 time: 0.8604 data_time: 0.5812 memory: 6319 loss: 0.3575 +2023/06/05 01:05:09 - mmengine - INFO - Epoch(train) [1][ 900/3907] lr: 9.9882e-05 eta: 10:08:32 time: 0.9043 data_time: 0.7637 memory: 6319 loss: 0.3411 +2023/06/05 01:06:40 - mmengine - INFO - Exp name: resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1_20230605_005004 +2023/06/05 01:06:40 - mmengine - INFO - Epoch(train) [1][1000/3907] lr: 9.9855e-05 eta: 10:04:02 time: 0.9026 data_time: 0.7349 memory: 6319 loss: 0.3338 +2023/06/05 01:08:15 - mmengine - INFO - Epoch(train) [1][1100/3907] lr: 9.9824e-05 eta: 10:02:20 time: 0.9518 data_time: 0.8113 memory: 6319 loss: 0.3185 +2023/06/05 01:09:48 - mmengine - INFO - Epoch(train) [1][1200/3907] lr: 9.9791e-05 eta: 9:59:50 time: 0.9547 data_time: 0.8124 memory: 6319 loss: 0.3290 +2023/06/05 01:11:25 - mmengine - INFO - Epoch(train) [1][1300/3907] lr: 9.9755e-05 eta: 9:59:12 time: 1.0366 data_time: 0.8950 memory: 6319 loss: 0.3130 +2023/06/05 01:12:54 - mmengine - INFO - Epoch(train) [1][1400/3907] lr: 9.9716e-05 eta: 9:54:36 time: 0.9110 data_time: 0.7602 memory: 6319 loss: 0.2925 +2023/06/05 01:14:29 - mmengine - INFO - Epoch(train) [1][1500/3907] lr: 9.9674e-05 eta: 9:53:13 time: 0.9153 data_time: 0.7745 memory: 6319 loss: 0.2958 +2023/06/05 01:16:04 - mmengine - INFO - Epoch(train) [1][1600/3907] lr: 9.9629e-05 eta: 9:51:54 time: 0.9497 data_time: 0.7992 memory: 6319 loss: 0.2959 +2023/06/05 01:17:37 - mmengine - INFO - Epoch(train) [1][1700/3907] lr: 9.9581e-05 eta: 9:49:36 time: 0.9059 data_time: 0.7656 memory: 6319 loss: 0.2618 +2023/06/05 01:19:09 - mmengine - INFO - Epoch(train) [1][1800/3907] lr: 9.9530e-05 eta: 9:47:07 time: 0.9151 data_time: 0.7734 memory: 6319 loss: 0.2773 +2023/06/05 01:20:44 - mmengine - INFO - Epoch(train) [1][1900/3907] lr: 9.9476e-05 eta: 9:45:36 time: 1.0719 data_time: 0.9314 memory: 6319 loss: 0.2589 +2023/06/05 01:22:17 - mmengine - INFO - Exp name: resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1_20230605_005004 +2023/06/05 01:22:17 - mmengine - INFO - Epoch(train) [1][2000/3907] lr: 9.9420e-05 eta: 9:43:32 time: 0.9504 data_time: 0.8003 memory: 6319 loss: 0.2333 +2023/06/05 01:23:51 - mmengine - INFO - Epoch(train) [1][2100/3907] lr: 9.9361e-05 eta: 9:41:54 time: 0.9295 data_time: 0.7894 memory: 6319 loss: 0.2547 +2023/06/05 01:25:26 - mmengine - INFO - Epoch(train) [1][2200/3907] lr: 9.9298e-05 eta: 9:40:25 time: 0.9649 data_time: 0.7355 memory: 6319 loss: 0.2377 +2023/06/05 01:26:59 - mmengine - INFO - Epoch(train) [1][2300/3907] lr: 9.9233e-05 eta: 9:38:35 time: 0.9442 data_time: 0.8029 memory: 6319 loss: 0.2387 +2023/06/05 01:28:31 - mmengine - INFO - Epoch(train) [1][2400/3907] lr: 9.9165e-05 eta: 9:36:23 time: 0.9128 data_time: 0.7717 memory: 6319 loss: 0.2254 +2023/06/05 01:30:07 - mmengine - INFO - Epoch(train) [1][2500/3907] lr: 9.9095e-05 eta: 9:35:11 time: 0.9164 data_time: 0.7757 memory: 6319 loss: 0.2296 +2023/06/05 01:31:40 - mmengine - INFO - Epoch(train) [1][2600/3907] lr: 9.9021e-05 eta: 9:33:14 time: 0.9202 data_time: 0.7729 memory: 6319 loss: 0.2305 +2023/06/05 01:33:14 - mmengine - INFO - Epoch(train) [1][2700/3907] lr: 9.8944e-05 eta: 9:31:41 time: 0.9397 data_time: 0.7994 memory: 6319 loss: 0.2144 +2023/06/05 01:34:48 - mmengine - INFO - Epoch(train) [1][2800/3907] lr: 9.8865e-05 eta: 9:30:02 time: 0.9213 data_time: 0.7800 memory: 6319 loss: 0.2114 +2023/06/05 01:36:34 - mmengine - INFO - Epoch(train) [1][2900/3907] lr: 9.8783e-05 eta: 9:30:53 time: 1.0847 data_time: 0.9439 memory: 6319 loss: 0.2165 +2023/06/05 01:38:03 - mmengine - INFO - Exp name: resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1_20230605_005004 +2023/06/05 01:38:03 - mmengine - INFO - Epoch(train) [1][3000/3907] lr: 9.8698e-05 eta: 9:28:10 time: 0.9725 data_time: 0.8304 memory: 6319 loss: 0.2230 +2023/06/05 01:39:35 - mmengine - INFO - Epoch(train) [1][3100/3907] lr: 9.8610e-05 eta: 9:26:07 time: 0.8884 data_time: 0.7491 memory: 6319 loss: 0.1944 +2023/06/05 01:41:10 - mmengine - INFO - Epoch(train) [1][3200/3907] lr: 9.8519e-05 eta: 9:24:33 time: 0.9380 data_time: 0.7982 memory: 6319 loss: 0.1902 +2023/06/05 01:42:43 - mmengine - INFO - Epoch(train) [1][3300/3907] lr: 9.8426e-05 eta: 9:22:40 time: 0.9202 data_time: 0.7791 memory: 6319 loss: 0.1994 +2023/06/05 01:44:17 - mmengine - INFO - Epoch(train) [1][3400/3907] lr: 9.8330e-05 eta: 9:21:09 time: 0.9155 data_time: 0.7738 memory: 6319 loss: 0.2073 +2023/06/05 01:45:51 - mmengine - INFO - Epoch(train) [1][3500/3907] lr: 9.8231e-05 eta: 9:19:26 time: 0.9742 data_time: 0.8334 memory: 6319 loss: 0.2061 +2023/06/05 01:47:26 - mmengine - INFO - Epoch(train) [1][3600/3907] lr: 9.8129e-05 eta: 9:17:56 time: 0.9193 data_time: 0.7801 memory: 6319 loss: 0.1972 +2023/06/05 01:49:08 - mmengine - INFO - Epoch(train) [1][3700/3907] lr: 9.8024e-05 eta: 9:17:36 time: 1.1128 data_time: 0.9733 memory: 6319 loss: 0.1906 +2023/06/05 01:50:40 - mmengine - INFO - Epoch(train) [1][3800/3907] lr: 9.7917e-05 eta: 9:15:40 time: 0.9095 data_time: 0.7267 memory: 6319 loss: 0.1877 +2023/06/05 01:52:13 - mmengine - INFO - Epoch(train) [1][3900/3907] lr: 9.7806e-05 eta: 9:13:51 time: 0.9296 data_time: 0.6991 memory: 6319 loss: 0.2030 +2023/06/05 01:52:22 - mmengine - INFO - Exp name: resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1_20230605_005004 +2023/06/05 01:52:22 - mmengine - INFO - Saving checkpoint at 1 epochs +2023/06/05 01:53:16 - mmengine - INFO - Epoch(val) [1][57/57] accuracy/top1: 72.2939 data_time: 0.6344 time: 0.7288 +2023/06/05 01:54:51 - mmengine - INFO - Exp name: resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1_20230605_005004 +2023/06/05 01:54:53 - mmengine - INFO - Epoch(train) [2][ 100/3907] lr: 9.7685e-05 eta: 9:12:53 time: 0.9377 data_time: 0.6942 memory: 6319 loss: 0.1863 +2023/06/05 01:56:27 - mmengine - INFO - Epoch(train) [2][ 200/3907] lr: 9.7570e-05 eta: 9:11:11 time: 0.9124 data_time: 0.4810 memory: 6319 loss: 0.1830 +2023/06/05 01:58:01 - mmengine - INFO - Epoch(train) [2][ 300/3907] lr: 9.7451e-05 eta: 9:09:31 time: 0.9542 data_time: 0.1303 memory: 6319 loss: 0.1873 +2023/06/05 01:59:34 - mmengine - INFO - Epoch(train) [2][ 400/3907] lr: 9.7329e-05 eta: 9:07:44 time: 0.8785 data_time: 0.3912 memory: 6319 loss: 0.1964 +2023/06/05 02:01:08 - mmengine - INFO - Epoch(train) [2][ 500/3907] lr: 9.7205e-05 eta: 9:06:08 time: 1.0425 data_time: 0.4964 memory: 6319 loss: 0.1640 +2023/06/05 02:02:53 - mmengine - INFO - Epoch(train) [2][ 600/3907] lr: 9.7078e-05 eta: 9:05:55 time: 0.8847 data_time: 0.0009 memory: 6319 loss: 0.1859 +2023/06/05 02:04:20 - mmengine - INFO - Epoch(train) [2][ 700/3907] lr: 9.6949e-05 eta: 9:03:20 time: 0.8780 data_time: 0.0008 memory: 6319 loss: 0.1765 +2023/06/05 02:05:53 - mmengine - INFO - Epoch(train) [2][ 800/3907] lr: 9.6816e-05 eta: 9:01:33 time: 0.9277 data_time: 0.0010 memory: 6319 loss: 0.1485 +2023/06/05 02:07:26 - mmengine - INFO - Epoch(train) [2][ 900/3907] lr: 9.6681e-05 eta: 8:59:46 time: 0.9405 data_time: 0.0013 memory: 6319 loss: 0.1635 +2023/06/05 02:08:58 - mmengine - INFO - Epoch(train) [2][1000/3907] lr: 9.6544e-05 eta: 8:57:56 time: 0.9014 data_time: 0.0012 memory: 6319 loss: 0.1782 +2023/06/05 02:10:22 - mmengine - INFO - Exp name: resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1_20230605_005004 +2023/06/05 02:10:30 - mmengine - INFO - Epoch(train) [2][1100/3907] lr: 9.6403e-05 eta: 8:56:08 time: 0.9265 data_time: 0.0008 memory: 6319 loss: 0.1663 +2023/06/05 02:12:03 - mmengine - INFO - Epoch(train) [2][1200/3907] lr: 9.6260e-05 eta: 8:54:19 time: 0.8964 data_time: 0.1195 memory: 6319 loss: 0.1653 +2023/06/05 02:13:36 - mmengine - INFO - Epoch(train) [2][1300/3907] lr: 9.6114e-05 eta: 8:52:36 time: 0.9097 data_time: 0.1018 memory: 6319 loss: 0.1627 +2023/06/05 02:15:10 - mmengine - INFO - Epoch(train) [2][1400/3907] lr: 9.5966e-05 eta: 8:51:02 time: 0.9312 data_time: 0.0009 memory: 6319 loss: 0.1729 +2023/06/05 02:16:42 - mmengine - INFO - Epoch(train) [2][1500/3907] lr: 9.5815e-05 eta: 8:49:15 time: 0.8903 data_time: 0.0010 memory: 6319 loss: 0.1617 +2023/06/05 02:18:13 - mmengine - INFO - Epoch(train) [2][1600/3907] lr: 9.5661e-05 eta: 8:47:15 time: 0.8931 data_time: 0.0009 memory: 6319 loss: 0.1610 +2023/06/05 02:20:09 - mmengine - INFO - Epoch(train) [2][1700/3907] lr: 9.5505e-05 eta: 8:47:53 time: 1.1866 data_time: 0.0009 memory: 6319 loss: 0.1541 +2023/06/05 02:21:40 - mmengine - INFO - Epoch(train) [2][1800/3907] lr: 9.5346e-05 eta: 8:45:59 time: 0.8901 data_time: 0.0010 memory: 6319 loss: 0.1459 +2023/06/05 02:23:07 - mmengine - INFO - Epoch(train) [2][1900/3907] lr: 9.5184e-05 eta: 8:43:39 time: 0.9058 data_time: 0.0010 memory: 6319 loss: 0.1632 +2023/06/05 02:24:41 - mmengine - INFO - Epoch(train) [2][2000/3907] lr: 9.5020e-05 eta: 8:42:00 time: 0.9148 data_time: 0.0012 memory: 6319 loss: 0.1564 +2023/06/05 02:26:05 - mmengine - INFO - Exp name: resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1_20230605_005004 +2023/06/05 02:26:14 - mmengine - INFO - Epoch(train) [2][2100/3907] lr: 9.4854e-05 eta: 8:40:18 time: 0.9513 data_time: 0.0011 memory: 6319 loss: 0.1472 +2023/06/05 02:27:46 - mmengine - INFO - Epoch(train) [2][2200/3907] lr: 9.4684e-05 eta: 8:38:33 time: 0.9603 data_time: 0.0011 memory: 6319 loss: 0.1522 +2023/06/05 02:29:18 - mmengine - INFO - Epoch(train) [2][2300/3907] lr: 9.4512e-05 eta: 8:36:44 time: 0.8811 data_time: 0.0015 memory: 6319 loss: 0.1509 +2023/06/05 02:30:50 - mmengine - INFO - Epoch(train) [2][2400/3907] lr: 9.4338e-05 eta: 8:34:59 time: 0.9293 data_time: 0.0013 memory: 6319 loss: 0.1373 +2023/06/05 02:32:23 - mmengine - INFO - Epoch(train) [2][2500/3907] lr: 9.4161e-05 eta: 8:33:17 time: 0.9071 data_time: 0.0029 memory: 6319 loss: 0.1581 +2023/06/05 02:33:55 - mmengine - INFO - Epoch(train) [2][2600/3907] lr: 9.3981e-05 eta: 8:31:29 time: 0.9530 data_time: 0.0011 memory: 6319 loss: 0.1618 +2023/06/05 02:35:28 - mmengine - INFO - Epoch(train) [2][2700/3907] lr: 9.3799e-05 eta: 8:29:50 time: 0.9390 data_time: 0.0010 memory: 6319 loss: 0.1558 +2023/06/05 02:37:00 - mmengine - INFO - Epoch(train) [2][2800/3907] lr: 9.3615e-05 eta: 8:28:05 time: 0.8889 data_time: 0.0010 memory: 6319 loss: 0.1467 +2023/06/05 02:38:47 - mmengine - INFO - Epoch(train) [2][2900/3907] lr: 9.3428e-05 eta: 8:27:34 time: 1.3877 data_time: 0.3139 memory: 6319 loss: 0.1381 +2023/06/05 02:41:03 - mmengine - INFO - Epoch(train) [2][3000/3907] lr: 9.3238e-05 eta: 8:29:12 time: 0.8234 data_time: 0.0009 memory: 6319 loss: 0.1440 +2023/06/05 02:42:19 - mmengine - INFO - Exp name: resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1_20230605_005004 +2023/06/05 02:42:26 - mmengine - INFO - Epoch(train) [2][3100/3907] lr: 9.3046e-05 eta: 8:26:44 time: 0.9329 data_time: 0.0010 memory: 6319 loss: 0.1390 +2023/06/05 02:43:54 - mmengine - INFO - Epoch(train) [2][3200/3907] lr: 9.2852e-05 eta: 8:24:38 time: 0.9016 data_time: 0.0008 memory: 6319 loss: 0.1575 +2023/06/05 02:45:24 - mmengine - INFO - Epoch(train) [2][3300/3907] lr: 9.2655e-05 eta: 8:22:40 time: 0.8751 data_time: 0.0011 memory: 6319 loss: 0.1643 +2023/06/05 02:46:58 - mmengine - INFO - Epoch(train) [2][3400/3907] lr: 9.2456e-05 eta: 8:21:04 time: 0.8702 data_time: 0.0010 memory: 6319 loss: 0.1424 +2023/06/05 02:48:27 - mmengine - INFO - Epoch(train) [2][3500/3907] lr: 9.2254e-05 eta: 8:19:06 time: 0.9298 data_time: 0.0298 memory: 6319 loss: 0.1589 +2023/06/05 02:49:59 - mmengine - INFO - Epoch(train) [2][3600/3907] lr: 9.2050e-05 eta: 8:17:18 time: 0.9080 data_time: 0.0010 memory: 6319 loss: 0.1362 +2023/06/05 02:51:30 - mmengine - INFO - Epoch(train) [2][3700/3907] lr: 9.1843e-05 eta: 8:15:29 time: 0.8900 data_time: 0.0012 memory: 6319 loss: 0.1292 +2023/06/05 02:53:01 - mmengine - INFO - Epoch(train) [2][3800/3907] lr: 9.1634e-05 eta: 8:13:41 time: 0.9340 data_time: 0.0011 memory: 6319 loss: 0.1407 +2023/06/05 02:54:31 - mmengine - INFO - Epoch(train) [2][3900/3907] lr: 9.1423e-05 eta: 8:11:48 time: 0.8854 data_time: 0.1532 memory: 6319 loss: 0.1329 +2023/06/05 02:54:33 - mmengine - INFO - Exp name: resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1_20230605_005004 +2023/06/05 02:54:33 - mmengine - INFO - Saving checkpoint at 2 epochs +2023/06/05 02:55:24 - mmengine - INFO - Epoch(val) [2][57/57] accuracy/top1: 87.6657 data_time: 0.5516 time: 0.6425 +2023/06/05 02:56:56 - mmengine - INFO - Epoch(train) [3][ 100/3907] lr: 9.1194e-05 eta: 8:09:42 time: 0.9267 data_time: 0.6440 memory: 6319 loss: 0.1410 +2023/06/05 02:58:17 - mmengine - INFO - Exp name: resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1_20230605_005004 +2023/06/05 02:58:27 - mmengine - INFO - Epoch(train) [3][ 200/3907] lr: 9.0978e-05 eta: 8:07:55 time: 0.9242 data_time: 0.7048 memory: 6319 loss: 0.1484 +2023/06/05 02:59:57 - mmengine - INFO - Epoch(train) [3][ 300/3907] lr: 9.0759e-05 eta: 8:06:04 time: 0.8952 data_time: 0.5704 memory: 6319 loss: 0.1361 +2023/06/05 03:01:29 - mmengine - INFO - Epoch(train) [3][ 400/3907] lr: 9.0539e-05 eta: 8:04:20 time: 0.9041 data_time: 0.5876 memory: 6319 loss: 0.1474 +2023/06/05 03:02:59 - mmengine - INFO - Epoch(train) [3][ 500/3907] lr: 9.0315e-05 eta: 8:02:33 time: 0.9061 data_time: 0.4506 memory: 6319 loss: 0.1351 +2023/06/05 03:04:30 - mmengine - INFO - Epoch(train) [3][ 600/3907] lr: 9.0090e-05 eta: 8:00:46 time: 0.8939 data_time: 0.4380 memory: 6319 loss: 0.1333 +2023/06/05 03:06:02 - mmengine - INFO - Epoch(train) [3][ 700/3907] lr: 8.9862e-05 eta: 7:59:03 time: 0.9614 data_time: 0.4666 memory: 6319 loss: 0.1189 +2023/06/05 03:07:33 - mmengine - INFO - Epoch(train) [3][ 800/3907] lr: 8.9632e-05 eta: 7:57:20 time: 0.8900 data_time: 0.4249 memory: 6319 loss: 0.1308 +2023/06/05 03:09:05 - mmengine - INFO - Epoch(train) [3][ 900/3907] lr: 8.9400e-05 eta: 7:55:38 time: 0.9113 data_time: 0.2928 memory: 6319 loss: 0.1369 +2023/06/05 03:10:35 - mmengine - INFO - Epoch(train) [3][1000/3907] lr: 8.9166e-05 eta: 7:53:52 time: 0.9123 data_time: 0.3997 memory: 6319 loss: 0.1307 +2023/06/05 03:12:08 - mmengine - INFO - Epoch(train) [3][1100/3907] lr: 8.8929e-05 eta: 7:52:15 time: 0.8534 data_time: 0.5624 memory: 6319 loss: 0.1373 +2023/06/05 03:13:31 - mmengine - INFO - Exp name: resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1_20230605_005004 +2023/06/05 03:13:41 - mmengine - INFO - Epoch(train) [3][1200/3907] lr: 8.8691e-05 eta: 7:50:35 time: 0.9333 data_time: 0.7919 memory: 6319 loss: 0.1250 +2023/06/05 03:15:13 - mmengine - INFO - Epoch(train) [3][1300/3907] lr: 8.8450e-05 eta: 7:48:57 time: 0.9083 data_time: 0.6926 memory: 6319 loss: 0.1170 +2023/06/05 03:16:44 - mmengine - INFO - Epoch(train) [3][1400/3907] lr: 8.8206e-05 eta: 7:47:14 time: 0.9107 data_time: 0.6411 memory: 6319 loss: 0.1120 +2023/06/05 03:18:12 - mmengine - INFO - Epoch(train) [3][1500/3907] lr: 8.7961e-05 eta: 7:45:20 time: 0.9314 data_time: 0.5771 memory: 6319 loss: 0.1333 +2023/06/05 03:19:45 - mmengine - INFO - Epoch(train) [3][1600/3907] lr: 8.7714e-05 eta: 7:43:43 time: 0.9074 data_time: 0.5237 memory: 6319 loss: 0.1460 +2023/06/05 03:21:18 - mmengine - INFO - Epoch(train) [3][1700/3907] lr: 8.7464e-05 eta: 7:42:06 time: 0.8852 data_time: 0.4846 memory: 6319 loss: 0.1217 +2023/06/05 03:22:50 - mmengine - INFO - Epoch(train) [3][1800/3907] lr: 8.7213e-05 eta: 7:40:27 time: 0.9162 data_time: 0.2292 memory: 6319 loss: 0.1088 +2023/06/05 03:24:24 - mmengine - INFO - Epoch(train) [3][1900/3907] lr: 8.6959e-05 eta: 7:38:55 time: 0.9655 data_time: 0.3407 memory: 6319 loss: 0.1254 +2023/06/05 03:25:55 - mmengine - INFO - Epoch(train) [3][2000/3907] lr: 8.6703e-05 eta: 7:37:12 time: 0.8774 data_time: 0.4117 memory: 6319 loss: 0.1320 +2023/06/05 03:27:44 - mmengine - INFO - Epoch(train) [3][2100/3907] lr: 8.6445e-05 eta: 7:36:25 time: 1.1628 data_time: 0.4863 memory: 6319 loss: 0.1270 +2023/06/05 03:29:20 - mmengine - INFO - Exp name: resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1_20230605_005004 +2023/06/05 03:29:34 - mmengine - INFO - Epoch(train) [3][2200/3907] lr: 8.6186e-05 eta: 7:35:35 time: 0.9102 data_time: 0.3986 memory: 6319 loss: 0.1289 +2023/06/05 03:31:23 - mmengine - INFO - Epoch(train) [3][2300/3907] lr: 8.5924e-05 eta: 7:34:45 time: 1.0618 data_time: 0.7757 memory: 6319 loss: 0.1224 +2023/06/05 03:33:15 - mmengine - INFO - Epoch(train) [3][2400/3907] lr: 8.5660e-05 eta: 7:33:59 time: 1.2811 data_time: 1.1375 memory: 6319 loss: 0.1166 +2023/06/05 03:35:04 - mmengine - INFO - Epoch(train) [3][2500/3907] lr: 8.5394e-05 eta: 7:33:07 time: 0.9341 data_time: 0.7940 memory: 6319 loss: 0.1232 +2023/06/05 03:36:32 - mmengine - INFO - Epoch(train) [3][2600/3907] lr: 8.5126e-05 eta: 7:31:16 time: 0.8765 data_time: 0.7362 memory: 6319 loss: 0.1009 +2023/06/05 03:38:03 - mmengine - INFO - Epoch(train) [3][2700/3907] lr: 8.4856e-05 eta: 7:29:29 time: 0.8936 data_time: 0.7523 memory: 6319 loss: 0.1338 +2023/06/05 03:39:31 - mmengine - INFO - Epoch(train) [3][2800/3907] lr: 8.4585e-05 eta: 7:27:39 time: 0.8997 data_time: 0.7583 memory: 6319 loss: 0.1197 +2023/06/05 03:41:03 - mmengine - INFO - Epoch(train) [3][2900/3907] lr: 8.4311e-05 eta: 7:25:58 time: 0.8732 data_time: 0.7312 memory: 6319 loss: 0.1121 +2023/06/05 03:42:35 - mmengine - INFO - Epoch(train) [3][3000/3907] lr: 8.4036e-05 eta: 7:24:17 time: 0.9675 data_time: 0.8261 memory: 6319 loss: 0.1139 +2023/06/05 03:44:04 - mmengine - INFO - Epoch(train) [3][3100/3907] lr: 8.3758e-05 eta: 7:22:30 time: 0.8432 data_time: 0.7037 memory: 6319 loss: 0.1204 +2023/06/05 03:45:22 - mmengine - INFO - Exp name: resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1_20230605_005004 +2023/06/05 03:45:31 - mmengine - INFO - Epoch(train) [3][3200/3907] lr: 8.3479e-05 eta: 7:20:37 time: 0.8740 data_time: 0.7328 memory: 6319 loss: 0.1229 +2023/06/05 03:47:03 - mmengine - INFO - Epoch(train) [3][3300/3907] lr: 8.3198e-05 eta: 7:18:57 time: 0.9856 data_time: 0.8449 memory: 6319 loss: 0.1154 +2023/06/05 03:48:33 - mmengine - INFO - Epoch(train) [3][3400/3907] lr: 8.2915e-05 eta: 7:17:13 time: 0.8967 data_time: 0.7430 memory: 6319 loss: 0.1148 +2023/06/05 03:50:03 - mmengine - INFO - Epoch(train) [3][3500/3907] lr: 8.2630e-05 eta: 7:15:28 time: 0.8638 data_time: 0.7223 memory: 6319 loss: 0.1155 +2023/06/05 03:51:33 - mmengine - INFO - Epoch(train) [3][3600/3907] lr: 8.2344e-05 eta: 7:13:43 time: 0.9698 data_time: 0.8286 memory: 6319 loss: 0.1265 +2023/06/05 03:53:03 - mmengine - INFO - Epoch(train) [3][3700/3907] lr: 8.2056e-05 eta: 7:11:58 time: 0.9162 data_time: 0.7762 memory: 6319 loss: 0.1178 +2023/06/05 03:54:32 - mmengine - INFO - Epoch(train) [3][3800/3907] lr: 8.1765e-05 eta: 7:10:14 time: 0.8859 data_time: 0.7462 memory: 6319 loss: 0.1229 +2023/06/05 03:56:01 - mmengine - INFO - Epoch(train) [3][3900/3907] lr: 8.1474e-05 eta: 7:08:28 time: 0.8784 data_time: 0.7393 memory: 6319 loss: 0.1254 +2023/06/05 03:56:10 - mmengine - INFO - Exp name: resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1_20230605_005004 +2023/06/05 03:56:10 - mmengine - INFO - Saving checkpoint at 3 epochs +2023/06/05 03:57:01 - mmengine - INFO - Epoch(val) [3][57/57] accuracy/top1: 91.2703 data_time: 0.5698 time: 0.6579 +2023/06/05 03:58:34 - mmengine - INFO - Epoch(train) [4][ 100/3907] lr: 8.1160e-05 eta: 7:06:51 time: 0.9180 data_time: 0.7769 memory: 6319 loss: 0.1149 +2023/06/05 04:00:05 - mmengine - INFO - Epoch(train) [4][ 200/3907] lr: 8.0864e-05 eta: 7:05:12 time: 0.8997 data_time: 0.7582 memory: 6319 loss: 0.1064 +2023/06/05 04:01:18 - mmengine - INFO - Exp name: resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1_20230605_005004 +2023/06/05 04:01:36 - mmengine - INFO - Epoch(train) [4][ 300/3907] lr: 8.0567e-05 eta: 7:03:30 time: 0.8834 data_time: 0.7399 memory: 6319 loss: 0.1249 +2023/06/05 04:03:04 - mmengine - INFO - Epoch(train) [4][ 400/3907] lr: 8.0269e-05 eta: 7:01:44 time: 0.8847 data_time: 0.7430 memory: 6319 loss: 0.1161 +2023/06/05 04:04:36 - mmengine - INFO - Epoch(train) [4][ 500/3907] lr: 7.9969e-05 eta: 7:00:05 time: 0.8698 data_time: 0.7280 memory: 6319 loss: 0.1128 +2023/06/05 04:06:05 - mmengine - INFO - Epoch(train) [4][ 600/3907] lr: 7.9667e-05 eta: 6:58:21 time: 0.8657 data_time: 0.7241 memory: 6319 loss: 0.1136 +2023/06/05 04:07:36 - mmengine - INFO - Epoch(train) [4][ 700/3907] lr: 7.9363e-05 eta: 6:56:41 time: 0.9251 data_time: 0.7847 memory: 6319 loss: 0.1007 +2023/06/05 04:09:09 - mmengine - INFO - Epoch(train) [4][ 800/3907] lr: 7.9058e-05 eta: 6:55:04 time: 0.9326 data_time: 0.7920 memory: 6319 loss: 0.1132 +2023/06/05 04:10:39 - mmengine - INFO - Epoch(train) [4][ 900/3907] lr: 7.8752e-05 eta: 6:53:22 time: 0.8399 data_time: 0.7001 memory: 6319 loss: 0.1143 +2023/06/05 04:12:05 - mmengine - INFO - Epoch(train) [4][1000/3907] lr: 7.8444e-05 eta: 6:51:34 time: 0.8449 data_time: 0.7048 memory: 6319 loss: 0.1035 +2023/06/05 04:13:35 - mmengine - INFO - Epoch(train) [4][1100/3907] lr: 7.8134e-05 eta: 6:49:51 time: 0.8909 data_time: 0.7509 memory: 6319 loss: 0.1028 +2023/06/05 04:15:03 - mmengine - INFO - Epoch(train) [4][1200/3907] lr: 7.7823e-05 eta: 6:48:07 time: 0.8871 data_time: 0.7471 memory: 6319 loss: 0.1146 +2023/06/05 04:16:15 - mmengine - INFO - Exp name: resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1_20230605_005004 +2023/06/05 04:16:33 - mmengine - INFO - Epoch(train) [4][1300/3907] lr: 7.7510e-05 eta: 6:46:25 time: 0.9088 data_time: 0.7491 memory: 6319 loss: 0.1258 +2023/06/05 04:18:07 - mmengine - INFO - Epoch(train) [4][1400/3907] lr: 7.7196e-05 eta: 6:44:53 time: 0.9199 data_time: 0.7791 memory: 6319 loss: 0.1060 +2023/06/05 04:19:38 - mmengine - INFO - Epoch(train) [4][1500/3907] lr: 7.6881e-05 eta: 6:43:14 time: 0.9230 data_time: 0.7820 memory: 6319 loss: 0.1000 +2023/06/05 04:21:09 - mmengine - INFO - Epoch(train) [4][1600/3907] lr: 7.6564e-05 eta: 6:41:36 time: 0.8953 data_time: 0.7549 memory: 6319 loss: 0.0978 +2023/06/05 04:22:36 - mmengine - INFO - Epoch(train) [4][1700/3907] lr: 7.6246e-05 eta: 6:39:50 time: 0.8664 data_time: 0.7260 memory: 6319 loss: 0.1063 +2023/06/05 04:24:06 - mmengine - INFO - Epoch(train) [4][1800/3907] lr: 7.5926e-05 eta: 6:38:10 time: 0.9012 data_time: 0.7611 memory: 6319 loss: 0.1105 +2023/06/05 04:25:35 - mmengine - INFO - Epoch(train) [4][1900/3907] lr: 7.5605e-05 eta: 6:36:28 time: 0.8996 data_time: 0.7587 memory: 6319 loss: 0.0949 +2023/06/05 04:27:03 - mmengine - INFO - Epoch(train) [4][2000/3907] lr: 7.5283e-05 eta: 6:34:44 time: 0.8624 data_time: 0.7230 memory: 6319 loss: 0.0988 +2023/06/05 04:28:42 - mmengine - INFO - Epoch(train) [4][2100/3907] lr: 7.4959e-05 eta: 6:33:21 time: 0.8469 data_time: 0.5954 memory: 6319 loss: 0.1202 +2023/06/05 04:30:14 - mmengine - INFO - Epoch(train) [4][2200/3907] lr: 7.4634e-05 eta: 6:31:44 time: 0.9595 data_time: 0.5913 memory: 6319 loss: 0.0970 +2023/06/05 04:31:26 - mmengine - INFO - Exp name: resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1_20230605_005004 +2023/06/05 04:31:45 - mmengine - INFO - Epoch(train) [4][2300/3907] lr: 7.4308e-05 eta: 6:30:06 time: 0.9349 data_time: 0.7595 memory: 6319 loss: 0.1112 +2023/06/05 04:33:14 - mmengine - INFO - Epoch(train) [4][2400/3907] lr: 7.3980e-05 eta: 6:28:25 time: 0.8592 data_time: 0.7184 memory: 6319 loss: 0.0929 +2023/06/05 04:34:43 - mmengine - INFO - Epoch(train) [4][2500/3907] lr: 7.3652e-05 eta: 6:26:43 time: 0.9228 data_time: 0.7816 memory: 6319 loss: 0.0941 +2023/06/05 04:36:14 - mmengine - INFO - Epoch(train) [4][2600/3907] lr: 7.3322e-05 eta: 6:25:06 time: 0.9041 data_time: 0.7645 memory: 6319 loss: 0.0887 +2023/06/05 04:37:43 - mmengine - INFO - Epoch(train) [4][2700/3907] lr: 7.2991e-05 eta: 6:23:25 time: 0.9211 data_time: 0.7800 memory: 6319 loss: 0.1205 +2023/06/05 04:39:12 - mmengine - INFO - Epoch(train) [4][2800/3907] lr: 7.2659e-05 eta: 6:21:44 time: 0.8854 data_time: 0.7452 memory: 6319 loss: 0.1063 +2023/06/05 04:40:42 - mmengine - INFO - Epoch(train) [4][2900/3907] lr: 7.2325e-05 eta: 6:20:06 time: 0.8378 data_time: 0.6977 memory: 6319 loss: 0.1005 +2023/06/05 04:42:11 - mmengine - INFO - Epoch(train) [4][3000/3907] lr: 7.1991e-05 eta: 6:18:25 time: 0.8737 data_time: 0.7326 memory: 6319 loss: 0.0924 +2023/06/05 04:43:38 - mmengine - INFO - Epoch(train) [4][3100/3907] lr: 7.1655e-05 eta: 6:16:42 time: 0.9047 data_time: 0.7636 memory: 6319 loss: 0.1026 +2023/06/05 04:45:03 - mmengine - INFO - Epoch(train) [4][3200/3907] lr: 7.1318e-05 eta: 6:14:56 time: 0.8481 data_time: 0.7071 memory: 6319 loss: 0.0964 +2023/06/05 04:46:11 - mmengine - INFO - Exp name: resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1_20230605_005004 +2023/06/05 04:46:29 - mmengine - INFO - Epoch(train) [4][3300/3907] lr: 7.0981e-05 eta: 6:13:11 time: 0.8599 data_time: 0.7198 memory: 6319 loss: 0.1067 +2023/06/05 04:47:56 - mmengine - INFO - Epoch(train) [4][3400/3907] lr: 7.0642e-05 eta: 6:11:28 time: 0.9222 data_time: 0.7828 memory: 6319 loss: 0.0976 +2023/06/05 04:49:24 - mmengine - INFO - Epoch(train) [4][3500/3907] lr: 7.0302e-05 eta: 6:09:48 time: 0.8766 data_time: 0.7303 memory: 6319 loss: 0.1024 +2023/06/05 04:50:51 - mmengine - INFO - Epoch(train) [4][3600/3907] lr: 6.9961e-05 eta: 6:08:05 time: 0.8568 data_time: 0.7167 memory: 6319 loss: 0.1080 +2023/06/05 04:52:16 - mmengine - INFO - Epoch(train) [4][3700/3907] lr: 6.9620e-05 eta: 6:06:19 time: 0.8947 data_time: 0.5676 memory: 6319 loss: 0.0982 +2023/06/05 04:53:43 - mmengine - INFO - Epoch(train) [4][3800/3907] lr: 6.9277e-05 eta: 6:04:37 time: 0.8818 data_time: 0.5461 memory: 6319 loss: 0.1020 +2023/06/05 04:55:08 - mmengine - INFO - Epoch(train) [4][3900/3907] lr: 6.8933e-05 eta: 6:02:53 time: 0.8513 data_time: 0.4925 memory: 6319 loss: 0.1061 +2023/06/05 04:55:15 - mmengine - INFO - Exp name: resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1_20230605_005004 +2023/06/05 04:55:15 - mmengine - INFO - Saving checkpoint at 4 epochs +2023/06/05 04:56:08 - mmengine - INFO - Epoch(val) [4][57/57] accuracy/top1: 93.6676 data_time: 0.5767 time: 0.6661 +2023/06/05 04:57:40 - mmengine - INFO - Epoch(train) [5][ 100/3907] lr: 6.8565e-05 eta: 6:01:12 time: 0.8576 data_time: 0.7158 memory: 6319 loss: 0.1007 +2023/06/05 04:59:06 - mmengine - INFO - Epoch(train) [5][ 200/3907] lr: 6.8219e-05 eta: 5:59:30 time: 0.8227 data_time: 0.6828 memory: 6319 loss: 0.1158 +2023/06/05 05:00:31 - mmengine - INFO - Epoch(train) [5][ 300/3907] lr: 6.7873e-05 eta: 5:57:45 time: 0.8216 data_time: 0.6816 memory: 6319 loss: 0.0880 +2023/06/05 05:01:37 - mmengine - INFO - Exp name: resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1_20230605_005004 +2023/06/05 05:01:57 - mmengine - INFO - Epoch(train) [5][ 400/3907] lr: 6.7526e-05 eta: 5:56:03 time: 0.8700 data_time: 0.5739 memory: 6319 loss: 0.0923 +2023/06/05 05:03:24 - mmengine - INFO - Epoch(train) [5][ 500/3907] lr: 6.7178e-05 eta: 5:54:23 time: 0.9017 data_time: 0.3925 memory: 6319 loss: 0.0960 +2023/06/05 05:04:52 - mmengine - INFO - Epoch(train) [5][ 600/3907] lr: 6.6829e-05 eta: 5:52:42 time: 0.9112 data_time: 0.5439 memory: 6319 loss: 0.0961 +2023/06/05 05:06:19 - mmengine - INFO - Epoch(train) [5][ 700/3907] lr: 6.6480e-05 eta: 5:51:02 time: 0.9371 data_time: 0.6819 memory: 6319 loss: 0.0924 +2023/06/05 05:07:46 - mmengine - INFO - Epoch(train) [5][ 800/3907] lr: 6.6129e-05 eta: 5:49:22 time: 0.8458 data_time: 0.4025 memory: 6319 loss: 0.1131 +2023/06/05 05:09:10 - mmengine - INFO - Epoch(train) [5][ 900/3907] lr: 6.5778e-05 eta: 5:47:38 time: 0.8598 data_time: 0.5673 memory: 6319 loss: 0.1025 +2023/06/05 05:10:38 - mmengine - INFO - Epoch(train) [5][1000/3907] lr: 6.5427e-05 eta: 5:45:58 time: 0.8780 data_time: 0.3596 memory: 6319 loss: 0.0974 +2023/06/05 05:12:03 - mmengine - INFO - Epoch(train) [5][1100/3907] lr: 6.5074e-05 eta: 5:44:16 time: 0.8206 data_time: 0.2431 memory: 6319 loss: 0.0874 +2023/06/05 05:13:28 - mmengine - INFO - Epoch(train) [5][1200/3907] lr: 6.4721e-05 eta: 5:42:35 time: 0.8541 data_time: 0.2172 memory: 6319 loss: 0.0927 +2023/06/05 05:14:54 - mmengine - INFO - Epoch(train) [5][1300/3907] lr: 6.4368e-05 eta: 5:40:54 time: 0.8683 data_time: 0.2020 memory: 6319 loss: 0.0937 +2023/06/05 05:15:58 - mmengine - INFO - Exp name: resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1_20230605_005004 +2023/06/05 05:16:23 - mmengine - INFO - Epoch(train) [5][1400/3907] lr: 6.4014e-05 eta: 5:39:16 time: 0.9106 data_time: 0.2511 memory: 6319 loss: 0.0941 +2023/06/05 05:17:51 - mmengine - INFO - Epoch(train) [5][1500/3907] lr: 6.3659e-05 eta: 5:37:39 time: 0.9081 data_time: 0.0791 memory: 6319 loss: 0.1140 +2023/06/05 05:19:18 - mmengine - INFO - Epoch(train) [5][1600/3907] lr: 6.3303e-05 eta: 5:36:00 time: 0.8431 data_time: 0.0013 memory: 6319 loss: 0.0948 +2023/06/05 05:20:45 - mmengine - INFO - Epoch(train) [5][1700/3907] lr: 6.2948e-05 eta: 5:34:20 time: 0.9024 data_time: 0.0009 memory: 6319 loss: 0.0878 +2023/06/05 05:22:13 - mmengine - INFO - Epoch(train) [5][1800/3907] lr: 6.2591e-05 eta: 5:32:42 time: 0.9443 data_time: 0.0014 memory: 6319 loss: 0.0825 +2023/06/05 05:23:38 - mmengine - INFO - Epoch(train) [5][1900/3907] lr: 6.2234e-05 eta: 5:31:02 time: 0.7980 data_time: 0.0014 memory: 6319 loss: 0.0971 +2023/06/05 05:25:05 - mmengine - INFO - Epoch(train) [5][2000/3907] lr: 6.1877e-05 eta: 5:29:23 time: 0.9106 data_time: 0.0015 memory: 6319 loss: 0.0958 +2023/06/05 05:26:32 - mmengine - INFO - Epoch(train) [5][2100/3907] lr: 6.1519e-05 eta: 5:27:45 time: 0.8483 data_time: 0.0012 memory: 6319 loss: 0.0841 +2023/06/05 05:27:59 - mmengine - INFO - Epoch(train) [5][2200/3907] lr: 6.1161e-05 eta: 5:26:07 time: 0.8732 data_time: 0.0011 memory: 6319 loss: 0.1126 +2023/06/05 05:29:29 - mmengine - INFO - Epoch(train) [5][2300/3907] lr: 6.0802e-05 eta: 5:24:32 time: 0.9497 data_time: 0.0010 memory: 6319 loss: 0.0919 +2023/06/05 05:30:30 - mmengine - INFO - Exp name: resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1_20230605_005004 +2023/06/05 05:30:56 - mmengine - INFO - Epoch(train) [5][2400/3907] lr: 6.0443e-05 eta: 5:22:54 time: 0.8957 data_time: 0.0016 memory: 6319 loss: 0.0926 +2023/06/05 05:32:23 - mmengine - INFO - Epoch(train) [5][2500/3907] lr: 6.0084e-05 eta: 5:21:16 time: 0.8653 data_time: 0.0010 memory: 6319 loss: 0.0985 +2023/06/05 05:33:49 - mmengine - INFO - Epoch(train) [5][2600/3907] lr: 5.9724e-05 eta: 5:19:36 time: 0.9047 data_time: 0.0012 memory: 6319 loss: 0.0951 +2023/06/05 05:35:15 - mmengine - INFO - Epoch(train) [5][2700/3907] lr: 5.9364e-05 eta: 5:17:58 time: 0.8602 data_time: 0.0016 memory: 6319 loss: 0.0792 +2023/06/05 05:36:41 - mmengine - INFO - Epoch(train) [5][2800/3907] lr: 5.9004e-05 eta: 5:16:20 time: 0.8848 data_time: 0.0024 memory: 6319 loss: 0.1015 +2023/06/05 05:38:07 - mmengine - INFO - Epoch(train) [5][2900/3907] lr: 5.8643e-05 eta: 5:14:41 time: 0.9009 data_time: 0.0011 memory: 6319 loss: 0.0997 +2023/06/05 05:39:36 - mmengine - INFO - Epoch(train) [5][3000/3907] lr: 5.8283e-05 eta: 5:13:05 time: 0.9279 data_time: 0.0014 memory: 6319 loss: 0.1028 +2023/06/05 05:41:03 - mmengine - INFO - Epoch(train) [5][3100/3907] lr: 5.7922e-05 eta: 5:11:28 time: 0.8469 data_time: 0.0018 memory: 6319 loss: 0.0893 +2023/06/05 05:42:30 - mmengine - INFO - Epoch(train) [5][3200/3907] lr: 5.7560e-05 eta: 5:09:51 time: 0.8649 data_time: 0.0014 memory: 6319 loss: 0.1141 +2023/06/05 05:43:56 - mmengine - INFO - Epoch(train) [5][3300/3907] lr: 5.7199e-05 eta: 5:08:13 time: 0.8548 data_time: 0.0015 memory: 6319 loss: 0.0951 +2023/06/05 05:44:57 - mmengine - INFO - Exp name: resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1_20230605_005004 +2023/06/05 05:45:22 - mmengine - INFO - Epoch(train) [5][3400/3907] lr: 5.6838e-05 eta: 5:06:35 time: 0.8596 data_time: 0.0008 memory: 6319 loss: 0.1077 +2023/06/05 05:46:49 - mmengine - INFO - Epoch(train) [5][3500/3907] lr: 5.6476e-05 eta: 5:04:59 time: 0.8443 data_time: 0.0064 memory: 6319 loss: 0.0961 +2023/06/05 05:48:16 - mmengine - INFO - Epoch(train) [5][3600/3907] lr: 5.6114e-05 eta: 5:03:22 time: 0.8782 data_time: 0.2764 memory: 6319 loss: 0.0891 +2023/06/05 05:49:43 - mmengine - INFO - Epoch(train) [5][3700/3907] lr: 5.5753e-05 eta: 5:01:46 time: 0.8478 data_time: 0.1058 memory: 6319 loss: 0.0875 +2023/06/05 05:51:10 - mmengine - INFO - Epoch(train) [5][3800/3907] lr: 5.5391e-05 eta: 5:00:08 time: 0.8218 data_time: 0.3337 memory: 6319 loss: 0.1057 +2023/06/05 05:52:35 - mmengine - INFO - Epoch(train) [5][3900/3907] lr: 5.5029e-05 eta: 4:58:31 time: 0.8530 data_time: 0.4288 memory: 6319 loss: 0.0873 +2023/06/05 05:52:40 - mmengine - INFO - Exp name: resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1_20230605_005004 +2023/06/05 05:52:40 - mmengine - INFO - Saving checkpoint at 5 epochs +2023/06/05 05:53:31 - mmengine - INFO - Epoch(val) [5][57/57] accuracy/top1: 94.7914 data_time: 0.5716 time: 0.6595 +2023/06/05 05:55:01 - mmengine - INFO - Epoch(train) [6][ 100/3907] lr: 5.4642e-05 eta: 4:56:49 time: 0.8367 data_time: 0.6958 memory: 6319 loss: 0.0874 +2023/06/05 05:56:28 - mmengine - INFO - Epoch(train) [6][ 200/3907] lr: 5.4280e-05 eta: 4:55:13 time: 0.8562 data_time: 0.3697 memory: 6319 loss: 0.1028 +2023/06/05 05:57:56 - mmengine - INFO - Epoch(train) [6][ 300/3907] lr: 5.3918e-05 eta: 4:53:38 time: 0.8948 data_time: 0.0011 memory: 6319 loss: 0.0973 +2023/06/05 05:59:23 - mmengine - INFO - Epoch(train) [6][ 400/3907] lr: 5.3556e-05 eta: 4:52:02 time: 0.9095 data_time: 0.4860 memory: 6319 loss: 0.0862 +2023/06/05 06:00:23 - mmengine - INFO - Exp name: resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1_20230605_005004 +2023/06/05 06:00:50 - mmengine - INFO - Epoch(train) [6][ 500/3907] lr: 5.3195e-05 eta: 4:50:26 time: 0.8785 data_time: 0.3417 memory: 6319 loss: 0.0918 +2023/06/05 06:02:16 - mmengine - INFO - Epoch(train) [6][ 600/3907] lr: 5.2833e-05 eta: 4:48:50 time: 0.8458 data_time: 0.3129 memory: 6319 loss: 0.0935 +2023/06/05 06:03:44 - mmengine - INFO - Epoch(train) [6][ 700/3907] lr: 5.2472e-05 eta: 4:47:14 time: 0.8982 data_time: 0.2621 memory: 6319 loss: 0.0920 +2023/06/05 06:05:10 - mmengine - INFO - Epoch(train) [6][ 800/3907] lr: 5.2111e-05 eta: 4:45:38 time: 0.8591 data_time: 0.2098 memory: 6319 loss: 0.0934 +2023/06/05 06:06:37 - mmengine - INFO - Epoch(train) [6][ 900/3907] lr: 5.1750e-05 eta: 4:44:02 time: 0.8593 data_time: 0.2419 memory: 6319 loss: 0.1009 +2023/06/05 06:08:03 - mmengine - INFO - Epoch(train) [6][1000/3907] lr: 5.1389e-05 eta: 4:42:26 time: 0.8671 data_time: 0.2608 memory: 6319 loss: 0.0899 +2023/06/05 06:09:30 - mmengine - INFO - Epoch(train) [6][1100/3907] lr: 5.1029e-05 eta: 4:40:51 time: 0.8788 data_time: 0.1881 memory: 6319 loss: 0.0930 +2023/06/05 06:10:57 - mmengine - INFO - Epoch(train) [6][1200/3907] lr: 5.0668e-05 eta: 4:39:16 time: 0.8981 data_time: 0.2563 memory: 6319 loss: 0.0854 +2023/06/05 06:12:24 - mmengine - INFO - Epoch(train) [6][1300/3907] lr: 5.0308e-05 eta: 4:37:40 time: 0.8827 data_time: 0.2645 memory: 6319 loss: 0.0879 +2023/06/05 06:13:52 - mmengine - INFO - Epoch(train) [6][1400/3907] lr: 4.9949e-05 eta: 4:36:06 time: 0.8823 data_time: 0.3767 memory: 6319 loss: 0.0930 +2023/06/05 06:14:52 - mmengine - INFO - Exp name: resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1_20230605_005004 +2023/06/05 06:15:20 - mmengine - INFO - Epoch(train) [6][1500/3907] lr: 4.9589e-05 eta: 4:34:32 time: 0.9584 data_time: 0.4078 memory: 6319 loss: 0.0816 +2023/06/05 06:16:47 - mmengine - INFO - Epoch(train) [6][1600/3907] lr: 4.9230e-05 eta: 4:32:57 time: 0.9099 data_time: 0.1006 memory: 6319 loss: 0.0799 +2023/06/05 06:18:11 - mmengine - INFO - Epoch(train) [6][1700/3907] lr: 4.8871e-05 eta: 4:31:19 time: 0.8498 data_time: 0.0818 memory: 6319 loss: 0.0818 +2023/06/05 06:19:38 - mmengine - INFO - Epoch(train) [6][1800/3907] lr: 4.8513e-05 eta: 4:29:44 time: 0.8746 data_time: 0.0120 memory: 6319 loss: 0.0815 +2023/06/05 06:21:05 - mmengine - INFO - Epoch(train) [6][1900/3907] lr: 4.8155e-05 eta: 4:28:09 time: 0.9074 data_time: 0.1346 memory: 6319 loss: 0.0855 +2023/06/05 06:22:32 - mmengine - INFO - Epoch(train) [6][2000/3907] lr: 4.7798e-05 eta: 4:26:35 time: 0.8949 data_time: 0.1818 memory: 6319 loss: 0.0932 +2023/06/05 06:24:00 - mmengine - INFO - Epoch(train) [6][2100/3907] lr: 4.7441e-05 eta: 4:25:01 time: 0.8991 data_time: 0.2333 memory: 6319 loss: 0.0890 +2023/06/05 06:25:25 - mmengine - INFO - Epoch(train) [6][2200/3907] lr: 4.7084e-05 eta: 4:23:25 time: 0.8605 data_time: 0.1925 memory: 6319 loss: 0.0845 +2023/06/05 06:26:51 - mmengine - INFO - Epoch(train) [6][2300/3907] lr: 4.6729e-05 eta: 4:21:50 time: 0.8764 data_time: 0.0011 memory: 6319 loss: 0.0838 +2023/06/05 06:28:18 - mmengine - INFO - Epoch(train) [6][2400/3907] lr: 4.6373e-05 eta: 4:20:15 time: 0.8523 data_time: 0.0008 memory: 6319 loss: 0.0984 +2023/06/05 06:29:18 - mmengine - INFO - Exp name: resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1_20230605_005004 +2023/06/05 06:29:45 - mmengine - INFO - Epoch(train) [6][2500/3907] lr: 4.6018e-05 eta: 4:18:41 time: 0.8351 data_time: 0.0378 memory: 6319 loss: 0.0833 +2023/06/05 06:31:12 - mmengine - INFO - Epoch(train) [6][2600/3907] lr: 4.5664e-05 eta: 4:17:06 time: 0.8598 data_time: 0.1655 memory: 6319 loss: 0.0821 +2023/06/05 06:32:40 - mmengine - INFO - Epoch(train) [6][2700/3907] lr: 4.5310e-05 eta: 4:15:33 time: 0.9598 data_time: 0.0013 memory: 6319 loss: 0.0791 +2023/06/05 06:34:07 - mmengine - INFO - Epoch(train) [6][2800/3907] lr: 4.4957e-05 eta: 4:13:59 time: 0.9162 data_time: 0.0013 memory: 6319 loss: 0.0800 +2023/06/05 06:35:35 - mmengine - INFO - Epoch(train) [6][2900/3907] lr: 4.4605e-05 eta: 4:12:25 time: 0.8416 data_time: 0.0012 memory: 6319 loss: 0.0832 +2023/06/05 06:37:01 - mmengine - INFO - Epoch(train) [6][3000/3907] lr: 4.4253e-05 eta: 4:10:51 time: 0.8518 data_time: 0.0013 memory: 6319 loss: 0.0836 +2023/06/05 06:38:27 - mmengine - INFO - Epoch(train) [6][3100/3907] lr: 4.3902e-05 eta: 4:09:16 time: 0.8203 data_time: 0.0011 memory: 6319 loss: 0.0914 +2023/06/05 06:39:54 - mmengine - INFO - Epoch(train) [6][3200/3907] lr: 4.3552e-05 eta: 4:07:42 time: 0.9106 data_time: 0.0016 memory: 6319 loss: 0.0798 +2023/06/05 06:41:19 - mmengine - INFO - Epoch(train) [6][3300/3907] lr: 4.3202e-05 eta: 4:06:07 time: 0.8781 data_time: 0.0015 memory: 6319 loss: 0.0804 +2023/06/05 06:42:46 - mmengine - INFO - Epoch(train) [6][3400/3907] lr: 4.2854e-05 eta: 4:04:33 time: 0.8196 data_time: 0.0011 memory: 6319 loss: 0.0834 +2023/06/05 06:43:47 - mmengine - INFO - Exp name: resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1_20230605_005004 +2023/06/05 06:44:13 - mmengine - INFO - Epoch(train) [6][3500/3907] lr: 4.2506e-05 eta: 4:02:59 time: 0.8233 data_time: 0.0014 memory: 6319 loss: 0.0834 +2023/06/05 06:45:44 - mmengine - INFO - Epoch(train) [6][3600/3907] lr: 4.2158e-05 eta: 4:01:28 time: 0.8593 data_time: 0.0014 memory: 6319 loss: 0.0796 +2023/06/05 06:47:14 - mmengine - INFO - Epoch(train) [6][3700/3907] lr: 4.1812e-05 eta: 3:59:57 time: 0.9151 data_time: 0.0013 memory: 6319 loss: 0.0748 +2023/06/05 06:48:42 - mmengine - INFO - Epoch(train) [6][3800/3907] lr: 4.1466e-05 eta: 3:58:24 time: 0.9031 data_time: 0.0019 memory: 6319 loss: 0.0789 +2023/06/05 06:50:08 - mmengine - INFO - Epoch(train) [6][3900/3907] lr: 4.1122e-05 eta: 3:56:50 time: 0.8818 data_time: 0.0014 memory: 6319 loss: 0.0829 +2023/06/05 06:50:16 - mmengine - INFO - Exp name: resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1_20230605_005004 +2023/06/05 06:50:16 - mmengine - INFO - Saving checkpoint at 6 epochs +2023/06/05 06:51:07 - mmengine - INFO - Epoch(val) [6][57/57] accuracy/top1: 95.6891 data_time: 0.5603 time: 0.6511 +2023/06/05 06:52:37 - mmengine - INFO - Epoch(train) [7][ 100/3907] lr: 4.0754e-05 eta: 3:55:13 time: 0.8788 data_time: 0.6458 memory: 6319 loss: 0.0882 +2023/06/05 06:54:05 - mmengine - INFO - Epoch(train) [7][ 200/3907] lr: 4.0411e-05 eta: 3:53:40 time: 0.9160 data_time: 0.7044 memory: 6319 loss: 0.0914 +2023/06/05 06:55:33 - mmengine - INFO - Epoch(train) [7][ 300/3907] lr: 4.0069e-05 eta: 3:52:08 time: 0.8741 data_time: 0.4020 memory: 6319 loss: 0.0734 +2023/06/05 06:57:01 - mmengine - INFO - Epoch(train) [7][ 400/3907] lr: 3.9729e-05 eta: 3:50:35 time: 0.9005 data_time: 0.2172 memory: 6319 loss: 0.0990 +2023/06/05 06:58:28 - mmengine - INFO - Epoch(train) [7][ 500/3907] lr: 3.9389e-05 eta: 3:49:02 time: 0.8485 data_time: 0.2860 memory: 6319 loss: 0.1028 +2023/06/05 06:59:19 - mmengine - INFO - Exp name: resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1_20230605_005004 +2023/06/05 06:59:55 - mmengine - INFO - Epoch(train) [7][ 600/3907] lr: 3.9050e-05 eta: 3:47:28 time: 0.8925 data_time: 0.3062 memory: 6319 loss: 0.0784 +2023/06/05 07:01:21 - mmengine - INFO - Epoch(train) [7][ 700/3907] lr: 3.8712e-05 eta: 3:45:55 time: 0.8458 data_time: 0.2190 memory: 6319 loss: 0.0798 +2023/06/05 07:02:49 - mmengine - INFO - Epoch(train) [7][ 800/3907] lr: 3.8375e-05 eta: 3:44:22 time: 0.8858 data_time: 0.2450 memory: 6319 loss: 0.0856 +2023/06/05 07:04:16 - mmengine - INFO - Epoch(train) [7][ 900/3907] lr: 3.8039e-05 eta: 3:42:49 time: 0.8840 data_time: 0.1595 memory: 6319 loss: 0.0864 +2023/06/05 07:05:44 - mmengine - INFO - Epoch(train) [7][1000/3907] lr: 3.7705e-05 eta: 3:41:17 time: 0.8697 data_time: 0.1218 memory: 6319 loss: 0.0722 +2023/06/05 07:07:13 - mmengine - INFO - Epoch(train) [7][1100/3907] lr: 3.7371e-05 eta: 3:39:45 time: 0.9085 data_time: 0.1945 memory: 6319 loss: 0.0812 +2023/06/05 07:08:42 - mmengine - INFO - Epoch(train) [7][1200/3907] lr: 3.7039e-05 eta: 3:38:13 time: 0.8779 data_time: 0.1543 memory: 6319 loss: 0.0822 +2023/06/05 07:10:11 - mmengine - INFO - Epoch(train) [7][1300/3907] lr: 3.6708e-05 eta: 3:36:41 time: 0.9229 data_time: 0.1841 memory: 6319 loss: 0.0815 +2023/06/05 07:11:44 - mmengine - INFO - Epoch(train) [7][1400/3907] lr: 3.6378e-05 eta: 3:35:12 time: 0.9016 data_time: 0.0011 memory: 6319 loss: 0.0907 +2023/06/05 07:13:14 - mmengine - INFO - Epoch(train) [7][1500/3907] lr: 3.6049e-05 eta: 3:33:41 time: 0.8719 data_time: 0.0011 memory: 6319 loss: 0.0810 +2023/06/05 07:14:06 - mmengine - INFO - Exp name: resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1_20230605_005004 +2023/06/05 07:14:44 - mmengine - INFO - Epoch(train) [7][1600/3907] lr: 3.5721e-05 eta: 3:32:09 time: 0.8803 data_time: 0.0018 memory: 6319 loss: 0.0843 +2023/06/05 07:16:13 - mmengine - INFO - Epoch(train) [7][1700/3907] lr: 3.5395e-05 eta: 3:30:37 time: 0.9024 data_time: 0.0012 memory: 6319 loss: 0.0738 +2023/06/05 07:17:40 - mmengine - INFO - Epoch(train) [7][1800/3907] lr: 3.5070e-05 eta: 3:29:05 time: 0.8492 data_time: 0.0011 memory: 6319 loss: 0.0899 +2023/06/05 07:19:07 - mmengine - INFO - Epoch(train) [7][1900/3907] lr: 3.4746e-05 eta: 3:27:32 time: 0.8629 data_time: 0.0315 memory: 6319 loss: 0.0995 +2023/06/05 07:20:35 - mmengine - INFO - Epoch(train) [7][2000/3907] lr: 3.4424e-05 eta: 3:26:00 time: 0.9037 data_time: 0.4481 memory: 6319 loss: 0.0822 +2023/06/05 07:22:05 - mmengine - INFO - Epoch(train) [7][2100/3907] lr: 3.4103e-05 eta: 3:24:29 time: 0.8510 data_time: 0.3498 memory: 6319 loss: 0.0742 +2023/06/05 07:23:37 - mmengine - INFO - Epoch(train) [7][2200/3907] lr: 3.3783e-05 eta: 3:22:59 time: 0.9196 data_time: 0.2011 memory: 6319 loss: 0.0834 +2023/06/05 07:25:07 - mmengine - INFO - Epoch(train) [7][2300/3907] lr: 3.3465e-05 eta: 3:21:28 time: 0.9185 data_time: 0.0012 memory: 6319 loss: 0.0714 +2023/06/05 07:26:39 - mmengine - INFO - Epoch(train) [7][2400/3907] lr: 3.3148e-05 eta: 3:19:57 time: 0.9016 data_time: 0.0012 memory: 6319 loss: 0.0760 +2023/06/05 07:28:07 - mmengine - INFO - Epoch(train) [7][2500/3907] lr: 3.2832e-05 eta: 3:18:26 time: 0.9291 data_time: 0.0014 memory: 6319 loss: 0.0824 +2023/06/05 07:28:52 - mmengine - INFO - Exp name: resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1_20230605_005004 +2023/06/05 07:29:34 - mmengine - INFO - Epoch(train) [7][2600/3907] lr: 3.2518e-05 eta: 3:16:53 time: 0.8514 data_time: 0.0009 memory: 6319 loss: 0.0952 +2023/06/05 07:31:03 - mmengine - INFO - Epoch(train) [7][2700/3907] lr: 3.2205e-05 eta: 3:15:21 time: 0.8465 data_time: 0.0010 memory: 6319 loss: 0.0764 +2023/06/05 07:32:30 - mmengine - INFO - Epoch(train) [7][2800/3907] lr: 3.1894e-05 eta: 3:13:49 time: 0.9169 data_time: 0.1621 memory: 6319 loss: 0.0731 +2023/06/05 07:33:59 - mmengine - INFO - Epoch(train) [7][2900/3907] lr: 3.1584e-05 eta: 3:12:18 time: 0.9133 data_time: 0.5428 memory: 6319 loss: 0.0927 +2023/06/05 07:35:28 - mmengine - INFO - Epoch(train) [7][3000/3907] lr: 3.1276e-05 eta: 3:10:46 time: 0.9186 data_time: 0.5765 memory: 6319 loss: 0.0860 +2023/06/05 07:36:58 - mmengine - INFO - Epoch(train) [7][3100/3907] lr: 3.0969e-05 eta: 3:09:15 time: 0.8728 data_time: 0.2776 memory: 6319 loss: 0.1000 +2023/06/05 07:38:29 - mmengine - INFO - Epoch(train) [7][3200/3907] lr: 3.0664e-05 eta: 3:07:45 time: 0.8892 data_time: 0.4183 memory: 6319 loss: 0.0922 +2023/06/05 07:39:56 - mmengine - INFO - Epoch(train) [7][3300/3907] lr: 3.0360e-05 eta: 3:06:13 time: 0.8984 data_time: 0.4561 memory: 6319 loss: 0.0751 +2023/06/05 07:41:26 - mmengine - INFO - Epoch(train) [7][3400/3907] lr: 3.0058e-05 eta: 3:04:41 time: 0.8724 data_time: 0.0297 memory: 6319 loss: 0.0830 +2023/06/05 07:42:55 - mmengine - INFO - Epoch(train) [7][3500/3907] lr: 2.9758e-05 eta: 3:03:10 time: 0.8737 data_time: 0.0014 memory: 6319 loss: 0.0820 +2023/06/05 07:43:41 - mmengine - INFO - Exp name: resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1_20230605_005004 +2023/06/05 07:44:25 - mmengine - INFO - Epoch(train) [7][3600/3907] lr: 2.9459e-05 eta: 3:01:39 time: 0.8833 data_time: 0.0012 memory: 6319 loss: 0.0936 +2023/06/05 07:45:55 - mmengine - INFO - Epoch(train) [7][3700/3907] lr: 2.9162e-05 eta: 3:00:08 time: 0.9197 data_time: 0.0012 memory: 6319 loss: 0.0718 +2023/06/05 07:47:25 - mmengine - INFO - Epoch(train) [7][3800/3907] lr: 2.8867e-05 eta: 2:58:38 time: 0.8804 data_time: 0.0012 memory: 6319 loss: 0.0737 +2023/06/05 07:48:54 - mmengine - INFO - Epoch(train) [7][3900/3907] lr: 2.8573e-05 eta: 2:57:06 time: 0.8879 data_time: 0.0009 memory: 6319 loss: 0.0697 +2023/06/05 07:48:55 - mmengine - INFO - Exp name: resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1_20230605_005004 +2023/06/05 07:48:55 - mmengine - INFO - Saving checkpoint at 7 epochs +2023/06/05 07:49:46 - mmengine - INFO - Epoch(val) [7][57/57] accuracy/top1: 96.1553 data_time: 0.5790 time: 0.6666 +2023/06/05 07:51:19 - mmengine - INFO - Epoch(train) [8][ 100/3907] lr: 2.8261e-05 eta: 2:55:28 time: 0.9134 data_time: 0.7732 memory: 6319 loss: 0.0677 +2023/06/05 07:52:49 - mmengine - INFO - Epoch(train) [8][ 200/3907] lr: 2.7971e-05 eta: 2:53:57 time: 0.8653 data_time: 0.7258 memory: 6319 loss: 0.0866 +2023/06/05 07:54:17 - mmengine - INFO - Epoch(train) [8][ 300/3907] lr: 2.7682e-05 eta: 2:52:26 time: 0.8934 data_time: 0.7519 memory: 6319 loss: 0.0832 +2023/06/05 07:55:47 - mmengine - INFO - Epoch(train) [8][ 400/3907] lr: 2.7395e-05 eta: 2:50:55 time: 0.8481 data_time: 0.7072 memory: 6319 loss: 0.0847 +2023/06/05 07:57:18 - mmengine - INFO - Epoch(train) [8][ 500/3907] lr: 2.7111e-05 eta: 2:49:24 time: 0.9298 data_time: 0.7882 memory: 6319 loss: 0.0849 +2023/06/05 07:58:49 - mmengine - INFO - Epoch(train) [8][ 600/3907] lr: 2.6828e-05 eta: 2:47:54 time: 1.0543 data_time: 0.4759 memory: 6319 loss: 0.0830 +2023/06/05 07:59:35 - mmengine - INFO - Exp name: resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1_20230605_005004 +2023/06/05 08:00:18 - mmengine - INFO - Epoch(train) [8][ 700/3907] lr: 2.6546e-05 eta: 2:46:23 time: 0.8654 data_time: 0.0672 memory: 6319 loss: 0.0805 +2023/06/05 08:01:46 - mmengine - INFO - Epoch(train) [8][ 800/3907] lr: 2.6267e-05 eta: 2:44:51 time: 0.8753 data_time: 0.1237 memory: 6319 loss: 0.0873 +2023/06/05 08:03:14 - mmengine - INFO - Epoch(train) [8][ 900/3907] lr: 2.5989e-05 eta: 2:43:20 time: 0.8810 data_time: 0.4160 memory: 6319 loss: 0.0751 +2023/06/05 08:04:45 - mmengine - INFO - Epoch(train) [8][1000/3907] lr: 2.5714e-05 eta: 2:41:49 time: 0.8699 data_time: 0.7109 memory: 6319 loss: 0.0780 +2023/06/05 08:06:15 - mmengine - INFO - Epoch(train) [8][1100/3907] lr: 2.5440e-05 eta: 2:40:18 time: 0.8340 data_time: 0.5895 memory: 6319 loss: 0.0803 +2023/06/05 08:07:44 - mmengine - INFO - Epoch(train) [8][1200/3907] lr: 2.5168e-05 eta: 2:38:47 time: 0.9264 data_time: 0.6977 memory: 6319 loss: 0.0870 +2023/06/05 08:09:12 - mmengine - INFO - Epoch(train) [8][1300/3907] lr: 2.4898e-05 eta: 2:37:16 time: 0.8869 data_time: 0.5857 memory: 6319 loss: 0.0749 +2023/06/05 08:10:42 - mmengine - INFO - Epoch(train) [8][1400/3907] lr: 2.4630e-05 eta: 2:35:45 time: 0.8788 data_time: 0.6904 memory: 6319 loss: 0.0905 +2023/06/05 08:12:12 - mmengine - INFO - Epoch(train) [8][1500/3907] lr: 2.4364e-05 eta: 2:34:14 time: 0.9071 data_time: 0.7645 memory: 6319 loss: 0.0735 +2023/06/05 08:13:41 - mmengine - INFO - Epoch(train) [8][1600/3907] lr: 2.4100e-05 eta: 2:32:43 time: 0.8953 data_time: 0.7549 memory: 6319 loss: 0.0857 +2023/06/05 08:14:26 - mmengine - INFO - Exp name: resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1_20230605_005004 +2023/06/05 08:15:10 - mmengine - INFO - Epoch(train) [8][1700/3907] lr: 2.3838e-05 eta: 2:31:12 time: 0.8755 data_time: 0.6345 memory: 6319 loss: 0.0785 +2023/06/05 08:16:40 - mmengine - INFO - Epoch(train) [8][1800/3907] lr: 2.3578e-05 eta: 2:29:41 time: 0.8833 data_time: 0.7431 memory: 6319 loss: 0.0772 +2023/06/05 08:18:10 - mmengine - INFO - Epoch(train) [8][1900/3907] lr: 2.3320e-05 eta: 2:28:10 time: 0.9384 data_time: 0.7974 memory: 6319 loss: 0.0850 +2023/06/05 08:19:40 - mmengine - INFO - Epoch(train) [8][2000/3907] lr: 2.3064e-05 eta: 2:26:40 time: 0.8765 data_time: 0.7361 memory: 6319 loss: 0.0824 +2023/06/05 08:21:10 - mmengine - INFO - Epoch(train) [8][2100/3907] lr: 2.2810e-05 eta: 2:25:09 time: 0.8945 data_time: 0.7424 memory: 6319 loss: 0.0933 +2023/06/05 08:22:41 - mmengine - INFO - Epoch(train) [8][2200/3907] lr: 2.2558e-05 eta: 2:23:39 time: 0.8897 data_time: 0.7501 memory: 6319 loss: 0.0735 +2023/06/05 08:24:08 - mmengine - INFO - Epoch(train) [8][2300/3907] lr: 2.2309e-05 eta: 2:22:07 time: 0.8905 data_time: 0.7491 memory: 6319 loss: 0.0822 +2023/06/05 08:25:38 - mmengine - INFO - Epoch(train) [8][2400/3907] lr: 2.2061e-05 eta: 2:20:36 time: 0.8569 data_time: 0.7169 memory: 6319 loss: 0.0715 +2023/06/05 08:27:07 - mmengine - INFO - Epoch(train) [8][2500/3907] lr: 2.1816e-05 eta: 2:19:05 time: 0.8695 data_time: 0.7214 memory: 6319 loss: 0.0790 +2023/06/05 08:28:35 - mmengine - INFO - Epoch(train) [8][2600/3907] lr: 2.1572e-05 eta: 2:17:34 time: 0.8898 data_time: 0.7498 memory: 6319 loss: 0.0734 +2023/06/05 08:29:19 - mmengine - INFO - Exp name: resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1_20230605_005004 +2023/06/05 08:30:02 - mmengine - INFO - Epoch(train) [8][2700/3907] lr: 2.1331e-05 eta: 2:16:03 time: 0.8631 data_time: 0.7228 memory: 6319 loss: 0.0838 +2023/06/05 08:31:28 - mmengine - INFO - Epoch(train) [8][2800/3907] lr: 2.1092e-05 eta: 2:14:31 time: 0.9125 data_time: 0.7716 memory: 6319 loss: 0.0836 +2023/06/05 08:32:58 - mmengine - INFO - Epoch(train) [8][2900/3907] lr: 2.0855e-05 eta: 2:13:00 time: 0.8625 data_time: 0.7222 memory: 6319 loss: 0.0745 +2023/06/05 08:34:26 - mmengine - INFO - Epoch(train) [8][3000/3907] lr: 2.0621e-05 eta: 2:11:29 time: 0.9054 data_time: 0.6083 memory: 6319 loss: 0.0637 +2023/06/05 08:35:55 - mmengine - INFO - Epoch(train) [8][3100/3907] lr: 2.0388e-05 eta: 2:09:58 time: 0.8785 data_time: 0.6378 memory: 6319 loss: 0.0722 +2023/06/05 08:37:24 - mmengine - INFO - Epoch(train) [8][3200/3907] lr: 2.0158e-05 eta: 2:08:27 time: 0.8364 data_time: 0.6941 memory: 6319 loss: 0.0777 +2023/06/05 08:38:52 - mmengine - INFO - Epoch(train) [8][3300/3907] lr: 1.9930e-05 eta: 2:06:56 time: 0.9249 data_time: 0.7178 memory: 6319 loss: 0.0864 +2023/06/05 08:40:20 - mmengine - INFO - Epoch(train) [8][3400/3907] lr: 1.9705e-05 eta: 2:05:25 time: 0.8730 data_time: 0.7322 memory: 6319 loss: 0.0825 +2023/06/05 08:41:48 - mmengine - INFO - Epoch(train) [8][3500/3907] lr: 1.9481e-05 eta: 2:03:54 time: 0.8992 data_time: 0.7571 memory: 6319 loss: 0.0840 +2023/06/05 08:43:15 - mmengine - INFO - Epoch(train) [8][3600/3907] lr: 1.9260e-05 eta: 2:02:23 time: 0.8475 data_time: 0.7064 memory: 6319 loss: 0.0755 +2023/06/05 08:44:05 - mmengine - INFO - Exp name: resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1_20230605_005004 +2023/06/05 08:44:45 - mmengine - INFO - Epoch(train) [8][3700/3907] lr: 1.9042e-05 eta: 2:00:52 time: 0.8525 data_time: 0.7113 memory: 6319 loss: 0.0776 +2023/06/05 08:46:11 - mmengine - INFO - Epoch(train) [8][3800/3907] lr: 1.8825e-05 eta: 1:59:21 time: 0.8883 data_time: 0.7467 memory: 6319 loss: 0.0744 +2023/06/05 08:47:40 - mmengine - INFO - Epoch(train) [8][3900/3907] lr: 1.8611e-05 eta: 1:57:50 time: 0.8617 data_time: 0.5244 memory: 6319 loss: 0.0709 +2023/06/05 08:47:46 - mmengine - INFO - Exp name: resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1_20230605_005004 +2023/06/05 08:47:46 - mmengine - INFO - Saving checkpoint at 8 epochs +2023/06/05 08:48:37 - mmengine - INFO - Epoch(val) [8][57/57] accuracy/top1: 96.4267 data_time: 0.5711 time: 0.6589 +2023/06/05 08:50:09 - mmengine - INFO - Epoch(train) [9][ 100/3907] lr: 1.8385e-05 eta: 1:56:13 time: 0.9066 data_time: 0.4078 memory: 6319 loss: 0.0778 +2023/06/05 08:51:36 - mmengine - INFO - Epoch(train) [9][ 200/3907] lr: 1.8176e-05 eta: 1:54:42 time: 0.9141 data_time: 0.4129 memory: 6319 loss: 0.0786 +2023/06/05 08:53:07 - mmengine - INFO - Epoch(train) [9][ 300/3907] lr: 1.7969e-05 eta: 1:53:12 time: 0.8531 data_time: 0.4945 memory: 6319 loss: 0.0816 +2023/06/05 08:54:38 - mmengine - INFO - Epoch(train) [9][ 400/3907] lr: 1.7765e-05 eta: 1:51:41 time: 0.8830 data_time: 0.7421 memory: 6319 loss: 0.0811 +2023/06/05 08:56:05 - mmengine - INFO - Epoch(train) [9][ 500/3907] lr: 1.7563e-05 eta: 1:50:10 time: 0.8515 data_time: 0.7099 memory: 6319 loss: 0.0618 +2023/06/05 08:57:34 - mmengine - INFO - Epoch(train) [9][ 600/3907] lr: 1.7363e-05 eta: 1:48:39 time: 0.8905 data_time: 0.4393 memory: 6319 loss: 0.0796 +2023/06/05 08:59:02 - mmengine - INFO - Epoch(train) [9][ 700/3907] lr: 1.7166e-05 eta: 1:47:09 time: 0.8681 data_time: 0.2928 memory: 6319 loss: 0.0669 +2023/06/05 08:59:42 - mmengine - INFO - Exp name: resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1_20230605_005004 +2023/06/05 09:00:29 - mmengine - INFO - Epoch(train) [9][ 800/3907] lr: 1.6971e-05 eta: 1:45:38 time: 0.8866 data_time: 0.2831 memory: 6319 loss: 0.1002 +2023/06/05 09:01:58 - mmengine - INFO - Epoch(train) [9][ 900/3907] lr: 1.6779e-05 eta: 1:44:07 time: 0.9002 data_time: 0.3116 memory: 6319 loss: 0.0696 +2023/06/05 09:03:28 - mmengine - INFO - Epoch(train) [9][1000/3907] lr: 1.6589e-05 eta: 1:42:36 time: 0.8633 data_time: 0.3040 memory: 6319 loss: 0.0750 +2023/06/05 09:04:57 - mmengine - INFO - Epoch(train) [9][1100/3907] lr: 1.6402e-05 eta: 1:41:06 time: 0.8771 data_time: 0.0012 memory: 6319 loss: 0.0726 +2023/06/05 09:06:26 - mmengine - INFO - Epoch(train) [9][1200/3907] lr: 1.6217e-05 eta: 1:39:35 time: 0.9277 data_time: 0.0017 memory: 6319 loss: 0.0714 +2023/06/05 09:07:53 - mmengine - INFO - Epoch(train) [9][1300/3907] lr: 1.6035e-05 eta: 1:38:04 time: 0.8678 data_time: 0.0017 memory: 6319 loss: 0.0764 +2023/06/05 09:09:22 - mmengine - INFO - Epoch(train) [9][1400/3907] lr: 1.5855e-05 eta: 1:36:34 time: 0.8993 data_time: 0.0013 memory: 6319 loss: 0.0691 +2023/06/05 09:10:51 - mmengine - INFO - Epoch(train) [9][1500/3907] lr: 1.5678e-05 eta: 1:35:03 time: 0.8712 data_time: 0.0018 memory: 6319 loss: 0.0773 +2023/06/05 09:12:17 - mmengine - INFO - Epoch(train) [9][1600/3907] lr: 1.5503e-05 eta: 1:33:32 time: 0.8848 data_time: 0.0913 memory: 6319 loss: 0.0754 +2023/06/05 09:13:45 - mmengine - INFO - Epoch(train) [9][1700/3907] lr: 1.5331e-05 eta: 1:32:01 time: 0.9211 data_time: 0.5145 memory: 6319 loss: 0.0820 +2023/06/05 09:14:24 - mmengine - INFO - Exp name: resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1_20230605_005004 +2023/06/05 09:15:13 - mmengine - INFO - Epoch(train) [9][1800/3907] lr: 1.5162e-05 eta: 1:30:30 time: 0.8309 data_time: 0.3576 memory: 6319 loss: 0.0826 +2023/06/05 09:16:42 - mmengine - INFO - Epoch(train) [9][1900/3907] lr: 1.4995e-05 eta: 1:29:00 time: 0.8745 data_time: 0.5439 memory: 6319 loss: 0.0774 +2023/06/05 09:18:09 - mmengine - INFO - Epoch(train) [9][2000/3907] lr: 1.4830e-05 eta: 1:27:29 time: 0.8381 data_time: 0.6360 memory: 6319 loss: 0.0765 +2023/06/05 09:19:38 - mmengine - INFO - Epoch(train) [9][2100/3907] lr: 1.4668e-05 eta: 1:25:58 time: 0.9381 data_time: 0.7989 memory: 6319 loss: 0.0797 +2023/06/05 09:21:06 - mmengine - INFO - Epoch(train) [9][2200/3907] lr: 1.4509e-05 eta: 1:24:28 time: 0.8539 data_time: 0.7137 memory: 6319 loss: 0.0820 +2023/06/05 09:22:35 - mmengine - INFO - Epoch(train) [9][2300/3907] lr: 1.4353e-05 eta: 1:22:57 time: 0.8746 data_time: 0.7314 memory: 6319 loss: 0.0795 +2023/06/05 09:24:03 - mmengine - INFO - Epoch(train) [9][2400/3907] lr: 1.4199e-05 eta: 1:21:27 time: 0.8790 data_time: 0.6613 memory: 6319 loss: 0.0904 +2023/06/05 09:25:33 - mmengine - INFO - Epoch(train) [9][2500/3907] lr: 1.4047e-05 eta: 1:19:56 time: 0.9183 data_time: 0.7778 memory: 6319 loss: 0.0719 +2023/06/05 09:27:01 - mmengine - INFO - Epoch(train) [9][2600/3907] lr: 1.3899e-05 eta: 1:18:26 time: 0.8572 data_time: 0.7127 memory: 6319 loss: 0.0718 +2023/06/05 09:28:31 - mmengine - INFO - Epoch(train) [9][2700/3907] lr: 1.3753e-05 eta: 1:16:55 time: 0.9158 data_time: 0.7749 memory: 6319 loss: 0.0729 +2023/06/05 09:29:17 - mmengine - INFO - Exp name: resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1_20230605_005004 +2023/06/05 09:30:03 - mmengine - INFO - Epoch(train) [9][2800/3907] lr: 1.3609e-05 eta: 1:15:25 time: 0.9341 data_time: 0.7825 memory: 6319 loss: 0.0712 +2023/06/05 09:31:31 - mmengine - INFO - Epoch(train) [9][2900/3907] lr: 1.3469e-05 eta: 1:13:55 time: 0.8848 data_time: 0.7434 memory: 6319 loss: 0.0807 +2023/06/05 09:33:00 - mmengine - INFO - Epoch(train) [9][3000/3907] lr: 1.3331e-05 eta: 1:12:24 time: 0.8445 data_time: 0.7036 memory: 6319 loss: 0.0859 +2023/06/05 09:34:27 - mmengine - INFO - Epoch(train) [9][3100/3907] lr: 1.3196e-05 eta: 1:10:54 time: 0.8498 data_time: 0.7089 memory: 6319 loss: 0.0866 +2023/06/05 09:35:57 - mmengine - INFO - Epoch(train) [9][3200/3907] lr: 1.3063e-05 eta: 1:09:23 time: 0.8963 data_time: 0.6240 memory: 6319 loss: 0.0869 +2023/06/05 09:37:30 - mmengine - INFO - Epoch(train) [9][3300/3907] lr: 1.2933e-05 eta: 1:07:53 time: 1.0171 data_time: 0.7880 memory: 6319 loss: 0.0740 +2023/06/05 09:39:01 - mmengine - INFO - Epoch(train) [9][3400/3907] lr: 1.2806e-05 eta: 1:06:23 time: 0.8917 data_time: 0.6992 memory: 6319 loss: 0.0656 +2023/06/05 09:40:28 - mmengine - INFO - Epoch(train) [9][3500/3907] lr: 1.2682e-05 eta: 1:04:53 time: 0.9182 data_time: 0.5775 memory: 6319 loss: 0.0734 +2023/06/05 09:41:59 - mmengine - INFO - Epoch(train) [9][3600/3907] lr: 1.2560e-05 eta: 1:03:22 time: 0.9139 data_time: 0.7716 memory: 6319 loss: 0.0627 +2023/06/05 09:43:30 - mmengine - INFO - Epoch(train) [9][3700/3907] lr: 1.2441e-05 eta: 1:01:52 time: 0.9312 data_time: 0.7893 memory: 6319 loss: 0.0815 +2023/06/05 09:44:10 - mmengine - INFO - Exp name: resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1_20230605_005004 +2023/06/05 09:44:57 - mmengine - INFO - Epoch(train) [9][3800/3907] lr: 1.2325e-05 eta: 1:00:22 time: 0.8679 data_time: 0.7252 memory: 6319 loss: 0.0742 +2023/06/05 09:46:26 - mmengine - INFO - Epoch(train) [9][3900/3907] lr: 1.2211e-05 eta: 0:58:51 time: 0.8572 data_time: 0.6608 memory: 6319 loss: 0.0857 +2023/06/05 09:46:34 - mmengine - INFO - Exp name: resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1_20230605_005004 +2023/06/05 09:46:34 - mmengine - INFO - Saving checkpoint at 9 epochs +2023/06/05 09:47:25 - mmengine - INFO - Epoch(val) [9][57/57] accuracy/top1: 96.7816 data_time: 0.5831 time: 0.6737 +2023/06/05 09:48:58 - mmengine - INFO - Epoch(train) [10][ 100/3907] lr: 1.2093e-05 eta: 0:57:15 time: 0.9578 data_time: 0.8175 memory: 6319 loss: 0.0888 +2023/06/05 09:50:29 - mmengine - INFO - Epoch(train) [10][ 200/3907] lr: 1.1985e-05 eta: 0:55:45 time: 0.9332 data_time: 0.7931 memory: 6319 loss: 0.0752 +2023/06/05 09:51:59 - mmengine - INFO - Epoch(train) [10][ 300/3907] lr: 1.1881e-05 eta: 0:54:15 time: 0.8997 data_time: 0.7582 memory: 6319 loss: 0.0756 +2023/06/05 09:53:27 - mmengine - INFO - Epoch(train) [10][ 400/3907] lr: 1.1778e-05 eta: 0:52:44 time: 0.8914 data_time: 0.7513 memory: 6319 loss: 0.0721 +2023/06/05 09:54:53 - mmengine - INFO - Epoch(train) [10][ 500/3907] lr: 1.1679e-05 eta: 0:51:14 time: 0.8105 data_time: 0.6705 memory: 6319 loss: 0.0816 +2023/06/05 09:56:20 - mmengine - INFO - Epoch(train) [10][ 600/3907] lr: 1.1583e-05 eta: 0:49:43 time: 0.9008 data_time: 0.7607 memory: 6319 loss: 0.0755 +2023/06/05 09:57:48 - mmengine - INFO - Epoch(train) [10][ 700/3907] lr: 1.1489e-05 eta: 0:48:13 time: 0.9043 data_time: 0.7633 memory: 6319 loss: 0.0741 +2023/06/05 09:59:15 - mmengine - INFO - Epoch(train) [10][ 800/3907] lr: 1.1398e-05 eta: 0:46:42 time: 0.8891 data_time: 0.7482 memory: 6319 loss: 0.0723 +2023/06/05 09:59:50 - mmengine - INFO - Exp name: resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1_20230605_005004 +2023/06/05 10:00:43 - mmengine - INFO - Epoch(train) [10][ 900/3907] lr: 1.1310e-05 eta: 0:45:12 time: 0.8934 data_time: 0.7256 memory: 6319 loss: 0.0837 +2023/06/05 10:02:13 - mmengine - INFO - Epoch(train) [10][1000/3907] lr: 1.1225e-05 eta: 0:43:42 time: 0.9017 data_time: 0.7597 memory: 6319 loss: 0.0775 +2023/06/05 10:03:45 - mmengine - INFO - Epoch(train) [10][1100/3907] lr: 1.1142e-05 eta: 0:42:12 time: 0.9326 data_time: 0.7524 memory: 6319 loss: 0.0744 +2023/06/05 10:05:15 - mmengine - INFO - Epoch(train) [10][1200/3907] lr: 1.1063e-05 eta: 0:40:41 time: 0.9052 data_time: 0.6474 memory: 6319 loss: 0.0840 +2023/06/05 10:06:44 - mmengine - INFO - Epoch(train) [10][1300/3907] lr: 1.0986e-05 eta: 0:39:11 time: 0.8492 data_time: 0.5261 memory: 6319 loss: 0.0979 +2023/06/05 10:08:16 - mmengine - INFO - Epoch(train) [10][1400/3907] lr: 1.0912e-05 eta: 0:37:41 time: 0.9602 data_time: 0.2852 memory: 6319 loss: 0.0918 +2023/06/05 10:09:45 - mmengine - INFO - Epoch(train) [10][1500/3907] lr: 1.0841e-05 eta: 0:36:11 time: 0.8757 data_time: 0.1406 memory: 6319 loss: 0.0646 +2023/06/05 10:11:14 - mmengine - INFO - Epoch(train) [10][1600/3907] lr: 1.0773e-05 eta: 0:34:40 time: 0.8481 data_time: 0.0012 memory: 6319 loss: 0.0777 +2023/06/05 10:12:43 - mmengine - INFO - Epoch(train) [10][1700/3907] lr: 1.0707e-05 eta: 0:33:10 time: 0.9160 data_time: 0.0012 memory: 6319 loss: 0.0776 +2023/06/05 10:14:11 - mmengine - INFO - Epoch(train) [10][1800/3907] lr: 1.0645e-05 eta: 0:31:40 time: 0.8501 data_time: 0.0021 memory: 6319 loss: 0.0653 +2023/06/05 10:14:39 - mmengine - INFO - Exp name: resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1_20230605_005004 +2023/06/05 10:15:41 - mmengine - INFO - Epoch(train) [10][1900/3907] lr: 1.0585e-05 eta: 0:30:10 time: 0.8961 data_time: 0.0017 memory: 6319 loss: 0.0780 +2023/06/05 10:17:08 - mmengine - INFO - Epoch(train) [10][2000/3907] lr: 1.0529e-05 eta: 0:28:39 time: 0.8929 data_time: 0.1272 memory: 6319 loss: 0.0728 +2023/06/05 10:18:38 - mmengine - INFO - Epoch(train) [10][2100/3907] lr: 1.0475e-05 eta: 0:27:09 time: 0.9187 data_time: 0.0072 memory: 6319 loss: 0.0661 +2023/06/05 10:20:07 - mmengine - INFO - Epoch(train) [10][2200/3907] lr: 1.0424e-05 eta: 0:25:39 time: 0.9101 data_time: 0.0018 memory: 6319 loss: 0.0852 +2023/06/05 10:21:37 - mmengine - INFO - Epoch(train) [10][2300/3907] lr: 1.0376e-05 eta: 0:24:09 time: 0.8600 data_time: 0.1332 memory: 6319 loss: 0.0835 +2023/06/05 10:23:08 - mmengine - INFO - Epoch(train) [10][2400/3907] lr: 1.0330e-05 eta: 0:22:38 time: 0.9033 data_time: 0.0012 memory: 6319 loss: 0.0755 +2023/06/05 10:24:39 - mmengine - INFO - Epoch(train) [10][2500/3907] lr: 1.0288e-05 eta: 0:21:08 time: 0.9313 data_time: 0.0014 memory: 6319 loss: 0.0860 +2023/06/05 10:26:10 - mmengine - INFO - Epoch(train) [10][2600/3907] lr: 1.0249e-05 eta: 0:19:38 time: 0.9517 data_time: 0.0010 memory: 6319 loss: 0.0686 +2023/06/05 10:27:42 - mmengine - INFO - Epoch(train) [10][2700/3907] lr: 1.0212e-05 eta: 0:18:08 time: 0.9567 data_time: 0.0015 memory: 6319 loss: 0.0771 +2023/06/05 10:29:13 - mmengine - INFO - Epoch(train) [10][2800/3907] lr: 1.0178e-05 eta: 0:16:38 time: 0.8813 data_time: 0.0022 memory: 6319 loss: 0.0742 +2023/06/05 10:29:41 - mmengine - INFO - Exp name: resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1_20230605_005004 +2023/06/05 10:30:43 - mmengine - INFO - Epoch(train) [10][2900/3907] lr: 1.0148e-05 eta: 0:15:08 time: 0.8484 data_time: 0.0011 memory: 6319 loss: 0.0781 +2023/06/05 10:32:13 - mmengine - INFO - Epoch(train) [10][3000/3907] lr: 1.0120e-05 eta: 0:13:38 time: 0.8971 data_time: 0.0015 memory: 6319 loss: 0.0726 +2023/06/05 10:33:42 - mmengine - INFO - Epoch(train) [10][3100/3907] lr: 1.0095e-05 eta: 0:12:07 time: 0.9233 data_time: 0.0012 memory: 6319 loss: 0.0812 +2023/06/05 10:35:08 - mmengine - INFO - Epoch(train) [10][3200/3907] lr: 1.0073e-05 eta: 0:10:37 time: 0.8691 data_time: 0.0010 memory: 6319 loss: 0.0837 +2023/06/05 10:36:37 - mmengine - INFO - Epoch(train) [10][3300/3907] lr: 1.0054e-05 eta: 0:09:07 time: 0.9090 data_time: 0.0058 memory: 6319 loss: 0.0831 +2023/06/05 10:38:05 - mmengine - INFO - Epoch(train) [10][3400/3907] lr: 1.0038e-05 eta: 0:07:37 time: 0.8885 data_time: 0.1609 memory: 6319 loss: 0.0785 +2023/06/05 10:39:34 - mmengine - INFO - Epoch(train) [10][3500/3907] lr: 1.0024e-05 eta: 0:06:06 time: 0.9101 data_time: 0.0816 memory: 6319 loss: 0.0716 +2023/06/05 10:41:02 - mmengine - INFO - Epoch(train) [10][3600/3907] lr: 1.0014e-05 eta: 0:04:36 time: 0.8365 data_time: 0.0013 memory: 6319 loss: 0.0748 +2023/06/05 10:42:27 - mmengine - INFO - Epoch(train) [10][3700/3907] lr: 1.0006e-05 eta: 0:03:06 time: 0.7640 data_time: 0.0010 memory: 6319 loss: 0.0685 +2023/06/05 10:44:44 - mmengine - INFO - Epoch(train) [10][3800/3907] lr: 1.0002e-05 eta: 0:01:36 time: 0.8526 data_time: 0.6855 memory: 6319 loss: 0.0862 +2023/06/05 10:45:16 - mmengine - INFO - Exp name: resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1_20230605_005004 +2023/06/05 10:46:06 - mmengine - INFO - Epoch(train) [10][3900/3907] lr: 1.0000e-05 eta: 0:00:06 time: 0.8021 data_time: 0.3220 memory: 6319 loss: 0.0761 +2023/06/05 10:46:13 - mmengine - INFO - Exp name: resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1_20230605_005004 +2023/06/05 10:46:13 - mmengine - INFO - Saving checkpoint at 10 epochs +2023/06/05 10:47:03 - mmengine - INFO - Epoch(val) [10][57/57] accuracy/top1: 96.8755 data_time: 0.5468 time: 0.6364 diff --git a/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/20230605_005004/vis_data/20230605_005004.json b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/20230605_005004/vis_data/20230605_005004.json new file mode 100644 index 0000000000000000000000000000000000000000..776df04618cd13133ff5b7f986084c52a340a3ff --- /dev/null +++ b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/20230605_005004/vis_data/20230605_005004.json @@ -0,0 +1,400 @@ +{"lr": 9.999857418382631e-05, "data_time": 0.9103787899017334, "loss": 0.6140454709529877, "time": 1.3434776306152343, "epoch": 1, "memory": 9436, "step": 100} +{"lr": 9.999423907359506e-05, "data_time": 0.5140883684158325, "loss": 0.5577386796474457, "time": 0.8296503305435181, "epoch": 1, "memory": 6319, "step": 200} +{"lr": 9.998699480412073e-05, "data_time": 0.3724829912185669, "loss": 0.505340787768364, "time": 0.884411072731018, "epoch": 1, "memory": 6319, "step": 300} +{"lr": 9.997684184379085e-05, "data_time": 0.2931915521621704, "loss": 0.45813901722431183, "time": 0.9431975364685059, "epoch": 1, "memory": 6319, "step": 400} +{"lr": 9.996378084905784e-05, "data_time": 0.4679435968399048, "loss": 0.44143602550029754, "time": 0.8859490156173706, "epoch": 1, "memory": 6319, "step": 500} +{"lr": 9.994781266439667e-05, "data_time": 0.6340442419052124, "loss": 0.4044747710227966, "time": 0.8933575868606567, "epoch": 1, "memory": 6319, "step": 600} +{"lr": 9.992893832225061e-05, "data_time": 0.47628321647644045, "loss": 0.3901466906070709, "time": 0.9069812536239624, "epoch": 1, "memory": 6319, "step": 700} +{"lr": 9.990715904296405e-05, "data_time": 0.581229305267334, "loss": 0.3575277209281921, "time": 0.8604262828826904, "epoch": 1, "memory": 6319, "step": 800} +{"lr": 9.988247623470371e-05, "data_time": 0.7636813163757324, "loss": 0.3410681664943695, "time": 0.9043498992919922, "epoch": 1, "memory": 6319, "step": 900} +{"lr": 9.985489149336749e-05, "data_time": 0.734875226020813, "loss": 0.3338106542825699, "time": 0.9026185274124146, "epoch": 1, "memory": 6319, "step": 1000} +{"lr": 9.982440660248187e-05, "data_time": 0.8113263130187989, "loss": 0.318480509519577, "time": 0.9517826318740845, "epoch": 1, "memory": 6319, "step": 1100} +{"lr": 9.979102353308574e-05, "data_time": 0.8123575448989868, "loss": 0.3289756804704666, "time": 0.954686951637268, "epoch": 1, "memory": 6319, "step": 1200} +{"lr": 9.975474444360358e-05, "data_time": 0.8950042724609375, "loss": 0.3130018025636673, "time": 1.036578369140625, "epoch": 1, "memory": 6319, "step": 1300} +{"lr": 9.97155716797056e-05, "data_time": 0.7601632118225098, "loss": 0.2924617141485214, "time": 0.9109964609146118, "epoch": 1, "memory": 6319, "step": 1400} +{"lr": 9.967350777415632e-05, "data_time": 0.7745493650436401, "loss": 0.29582877159118653, "time": 0.9152874946594238, "epoch": 1, "memory": 6319, "step": 1500} +{"lr": 9.962855544665059e-05, "data_time": 0.7992411613464355, "loss": 0.2959482431411743, "time": 0.9496608257293702, "epoch": 1, "memory": 6319, "step": 1600} +{"lr": 9.958071760363773e-05, "data_time": 0.7655555248260498, "loss": 0.26180612593889235, "time": 0.9058728694915772, "epoch": 1, "memory": 6319, "step": 1700} +{"lr": 9.952999733813382e-05, "data_time": 0.7733849287033081, "loss": 0.27730225026607513, "time": 0.9151493787765503, "epoch": 1, "memory": 6319, "step": 1800} +{"lr": 9.947639792952148e-05, "data_time": 0.9313913822174072, "loss": 0.25891817510128023, "time": 1.0718937873840333, "epoch": 1, "memory": 6319, "step": 1900} +{"lr": 9.941992284333813e-05, "data_time": 0.8002794265747071, "loss": 0.23325824439525605, "time": 0.9503692388534546, "epoch": 1, "memory": 6319, "step": 2000} +{"lr": 9.93605757310516e-05, "data_time": 0.7893903017044067, "loss": 0.2547219946980476, "time": 0.9295383214950561, "epoch": 1, "memory": 6319, "step": 2100} +{"lr": 9.929836042982421e-05, "data_time": 0.73553946018219, "loss": 0.23771324306726455, "time": 0.9649042844772339, "epoch": 1, "memory": 6319, "step": 2200} +{"lr": 9.923328096226467e-05, "data_time": 0.8029179096221923, "loss": 0.23873921036720275, "time": 0.9442267894744873, "epoch": 1, "memory": 6319, "step": 2300} +{"lr": 9.916534153616819e-05, "data_time": 0.7717267990112304, "loss": 0.22541094273328782, "time": 0.9127573490142822, "epoch": 1, "memory": 6319, "step": 2400} +{"lr": 9.909454654424387e-05, "data_time": 0.7756819725036621, "loss": 0.22962120026350022, "time": 0.9164350986480713, "epoch": 1, "memory": 6319, "step": 2500} +{"lr": 9.902090056383109e-05, "data_time": 0.7729017257690429, "loss": 0.2304988294839859, "time": 0.9201807975769043, "epoch": 1, "memory": 6319, "step": 2600} +{"lr": 9.894440835660365e-05, "data_time": 0.7993639230728149, "loss": 0.21435047537088395, "time": 0.9397423028945923, "epoch": 1, "memory": 6319, "step": 2700} +{"lr": 9.886507486826162e-05, "data_time": 0.7800388097763061, "loss": 0.21141017377376556, "time": 0.9213169574737549, "epoch": 1, "memory": 6319, "step": 2800} +{"lr": 9.878290522821154e-05, "data_time": 0.9438929796218872, "loss": 0.21645904332399368, "time": 1.0847433805465698, "epoch": 1, "memory": 6319, "step": 2900} +{"lr": 9.86979047492352e-05, "data_time": 0.8303739547729492, "loss": 0.22295447587966918, "time": 0.9725280284881592, "epoch": 1, "memory": 6319, "step": 3000} +{"lr": 9.861007892714578e-05, "data_time": 0.7491269111633301, "loss": 0.19436087012290953, "time": 0.8884036779403687, "epoch": 1, "memory": 6319, "step": 3100} +{"lr": 9.851943344043239e-05, "data_time": 0.7981921672821045, "loss": 0.1901811569929123, "time": 0.9379892587661743, "epoch": 1, "memory": 6319, "step": 3200} +{"lr": 9.842597414989319e-05, "data_time": 0.7791166067123413, "loss": 0.19938674718141555, "time": 0.9201548576354981, "epoch": 1, "memory": 6319, "step": 3300} +{"lr": 9.832970709825637e-05, "data_time": 0.773836088180542, "loss": 0.20726506412029266, "time": 0.9154900312423706, "epoch": 1, "memory": 6319, "step": 3400} +{"lr": 9.823063850978968e-05, "data_time": 0.8334040880203247, "loss": 0.20610593855381013, "time": 0.9741550207138061, "epoch": 1, "memory": 6319, "step": 3500} +{"lr": 9.81287747898975e-05, "data_time": 0.7800627708435058, "loss": 0.19718588441610335, "time": 0.9192874908447266, "epoch": 1, "memory": 6319, "step": 3600} +{"lr": 9.802412252470715e-05, "data_time": 0.9732753992080688, "loss": 0.1905989408493042, "time": 1.1127555608749389, "epoch": 1, "memory": 6319, "step": 3700} +{"lr": 9.791668848064258e-05, "data_time": 0.7266729831695556, "loss": 0.1876543715596199, "time": 0.9095073938369751, "epoch": 1, "memory": 6319, "step": 3800} +{"lr": 9.780647960398751e-05, "data_time": 0.6990907192230225, "loss": 0.20302069634199144, "time": 0.9295929431915283, "epoch": 1, "memory": 6319, "step": 3900} +{"accuracy/top1": 72.29393768310547, "data_time": 0.6343517387122438, "time": 0.7288378288871363, "step": 1} +{"lr": 9.7685491196646e-05, "data_time": 0.6941997289657593, "loss": 0.1862989366054535, "time": 0.9376872777938843, "epoch": 2, "memory": 6319, "step": 4007} +{"lr": 9.756956126085676e-05, "data_time": 0.4809847116470337, "loss": 0.1829976722598076, "time": 0.9123958349227905, "epoch": 2, "memory": 6319, "step": 4107} +{"lr": 9.745087893642482e-05, "data_time": 0.130255126953125, "loss": 0.1872863069176674, "time": 0.954208779335022, "epoch": 2, "memory": 6319, "step": 4207} +{"lr": 9.73294518969056e-05, "data_time": 0.39117271900177003, "loss": 0.19641478657722472, "time": 0.8785030126571656, "epoch": 2, "memory": 6319, "step": 4307} +{"lr": 9.72052879933172e-05, "data_time": 0.4963539123535156, "loss": 0.16399107426404952, "time": 1.0425027370452882, "epoch": 2, "memory": 6319, "step": 4407} +{"lr": 9.707839525363358e-05, "data_time": 0.0008583784103393555, "loss": 0.18590614348649978, "time": 0.8847098588943482, "epoch": 2, "memory": 6319, "step": 4507} +{"lr": 9.69487818822646e-05, "data_time": 0.0008333444595336914, "loss": 0.17653323858976364, "time": 0.8779830694198608, "epoch": 2, "memory": 6319, "step": 4607} +{"lr": 9.681645625952616e-05, "data_time": 0.0010425567626953125, "loss": 0.14852667599916458, "time": 0.9276500225067139, "epoch": 2, "memory": 6319, "step": 4707} +{"lr": 9.668142694109826e-05, "data_time": 0.001341700553894043, "loss": 0.16353578865528107, "time": 0.9405390739440918, "epoch": 2, "memory": 6319, "step": 4807} +{"lr": 9.654370265747162e-05, "data_time": 0.0012044191360473632, "loss": 0.17817180901765822, "time": 0.901382851600647, "epoch": 2, "memory": 6319, "step": 4907} +{"lr": 9.640329231338345e-05, "data_time": 0.0008122682571411133, "loss": 0.16629842966794967, "time": 0.9265341281890869, "epoch": 2, "memory": 6319, "step": 5007} +{"lr": 9.626020498724147e-05, "data_time": 0.11947834491729736, "loss": 0.16525886654853822, "time": 0.8963955163955688, "epoch": 2, "memory": 6319, "step": 5107} +{"lr": 9.611444993053717e-05, "data_time": 0.10180156230926514, "loss": 0.16267255395650865, "time": 0.9097028255462647, "epoch": 2, "memory": 6319, "step": 5207} +{"lr": 9.596603656724746e-05, "data_time": 0.0008769512176513671, "loss": 0.17286299914121628, "time": 0.9312175512313843, "epoch": 2, "memory": 6319, "step": 5307} +{"lr": 9.58149744932254e-05, "data_time": 0.0009878158569335937, "loss": 0.16170311644673346, "time": 0.8902554273605346, "epoch": 2, "memory": 6319, "step": 5407} +{"lr": 9.566127347557972e-05, "data_time": 0.0008677005767822266, "loss": 0.16101982742547988, "time": 0.8930779933929444, "epoch": 2, "memory": 6319, "step": 5507} +{"lr": 9.550494345204367e-05, "data_time": 0.0009014368057250977, "loss": 0.15405538603663443, "time": 1.1866081476211547, "epoch": 2, "memory": 6319, "step": 5607} +{"lr": 9.534599453033183e-05, "data_time": 0.000959014892578125, "loss": 0.14588873237371444, "time": 0.8901443719863892, "epoch": 2, "memory": 6319, "step": 5707} +{"lr": 9.518443698748712e-05, "data_time": 0.0010004997253417968, "loss": 0.16322704702615737, "time": 0.905848503112793, "epoch": 2, "memory": 6319, "step": 5807} +{"lr": 9.50202812692162e-05, "data_time": 0.001230001449584961, "loss": 0.15641508996486664, "time": 0.9147838354110718, "epoch": 2, "memory": 6319, "step": 5907} +{"lr": 9.485353798921386e-05, "data_time": 0.0010570526123046876, "loss": 0.14724608659744262, "time": 0.951305603981018, "epoch": 2, "memory": 6319, "step": 6007} +{"lr": 9.468421792847712e-05, "data_time": 0.0010634660720825195, "loss": 0.1521591491997242, "time": 0.9603242874145508, "epoch": 2, "memory": 6319, "step": 6107} +{"lr": 9.451233203460794e-05, "data_time": 0.0015326261520385743, "loss": 0.1508932925760746, "time": 0.8811447620391846, "epoch": 2, "memory": 6319, "step": 6207} +{"lr": 9.433789142110532e-05, "data_time": 0.001342010498046875, "loss": 0.1372598946094513, "time": 0.9293378829956055, "epoch": 2, "memory": 6319, "step": 6307} +{"lr": 9.416090736664709e-05, "data_time": 0.0029036998748779297, "loss": 0.1580960802733898, "time": 0.9070582628250122, "epoch": 2, "memory": 6319, "step": 6407} +{"lr": 9.398139131436032e-05, "data_time": 0.0010502338409423828, "loss": 0.1617794707417488, "time": 0.9530377388000488, "epoch": 2, "memory": 6319, "step": 6507} +{"lr": 9.379935487108167e-05, "data_time": 0.0010036706924438476, "loss": 0.15584279000759124, "time": 0.9389506578445435, "epoch": 2, "memory": 6319, "step": 6607} +{"lr": 9.361480980660697e-05, "data_time": 0.001021265983581543, "loss": 0.14667166024446487, "time": 0.8889129161834717, "epoch": 2, "memory": 6319, "step": 6707} +{"lr": 9.342776805292993e-05, "data_time": 0.3138540506362915, "loss": 0.13809967413544655, "time": 1.387740921974182, "epoch": 2, "memory": 6319, "step": 6807} +{"lr": 9.323824170347083e-05, "data_time": 0.0008918046951293945, "loss": 0.14395274966955185, "time": 0.8233531951904297, "epoch": 2, "memory": 6319, "step": 6907} +{"lr": 9.304624301229471e-05, "data_time": 0.0010257720947265624, "loss": 0.1390126757323742, "time": 0.9329187154769898, "epoch": 2, "memory": 6319, "step": 7007} +{"lr": 9.285178439331897e-05, "data_time": 0.000758957862854004, "loss": 0.1575406163930893, "time": 0.9016340970993042, "epoch": 2, "memory": 6319, "step": 7107} +{"lr": 9.265487841951087e-05, "data_time": 0.0011159658432006836, "loss": 0.16433390974998474, "time": 0.8750843286514283, "epoch": 2, "memory": 6319, "step": 7207} +{"lr": 9.245553782207421e-05, "data_time": 0.0009636163711547851, "loss": 0.14237954020500182, "time": 0.8701907634735108, "epoch": 2, "memory": 6319, "step": 7307} +{"lr": 9.225377548962671e-05, "data_time": 0.029822874069213866, "loss": 0.1589060455560684, "time": 0.929782772064209, "epoch": 2, "memory": 6319, "step": 7407} +{"lr": 9.204960446736601e-05, "data_time": 0.0010097503662109375, "loss": 0.13618902415037154, "time": 0.9080040931701661, "epoch": 2, "memory": 6319, "step": 7507} +{"lr": 9.184303795622715e-05, "data_time": 0.0011968612670898438, "loss": 0.12923629358410835, "time": 0.890028429031372, "epoch": 2, "memory": 6319, "step": 7607} +{"lr": 9.163408931202823e-05, "data_time": 0.001125335693359375, "loss": 0.1407058507204056, "time": 0.933996033668518, "epoch": 2, "memory": 6319, "step": 7707} +{"lr": 9.142277204460729e-05, "data_time": 0.1532210350036621, "loss": 0.13286348655819893, "time": 0.8853833913803101, "epoch": 2, "memory": 6319, "step": 7807} +{"accuracy/top1": 87.66570281982422, "data_time": 0.5515617872106617, "time": 0.6425289491127277, "step": 2} +{"lr": 9.119405492383288e-05, "data_time": 0.6440482378005982, "loss": 0.1410001888871193, "time": 0.9266921520233155, "epoch": 3, "memory": 6319, "step": 7914} +{"lr": 9.097787819233205e-05, "data_time": 0.7047796249389648, "loss": 0.14838272482156753, "time": 0.9241950035095214, "epoch": 3, "memory": 6319, "step": 8014} +{"lr": 9.075937526576627e-05, "data_time": 0.5703942060470581, "loss": 0.1360948972404003, "time": 0.8951602697372436, "epoch": 3, "memory": 6319, "step": 8114} +{"lr": 9.053856027171755e-05, "data_time": 0.5875956535339355, "loss": 0.14739920645952226, "time": 0.9040512561798095, "epoch": 3, "memory": 6319, "step": 8214} +{"lr": 9.031544748725785e-05, "data_time": 0.4505643129348755, "loss": 0.1350906141102314, "time": 0.9061188220977783, "epoch": 3, "memory": 6319, "step": 8314} +{"lr": 9.009005133802542e-05, "data_time": 0.4379525661468506, "loss": 0.13327923268079758, "time": 0.8938762903213501, "epoch": 3, "memory": 6319, "step": 8414} +{"lr": 8.986238639729242e-05, "data_time": 0.4665829181671143, "loss": 0.11894499957561493, "time": 0.9614094972610474, "epoch": 3, "memory": 6319, "step": 8514} +{"lr": 8.963246738502254e-05, "data_time": 0.42489118576049806, "loss": 0.13079854100942612, "time": 0.890029001235962, "epoch": 3, "memory": 6319, "step": 8614} +{"lr": 8.940030916691928e-05, "data_time": 0.2928466320037842, "loss": 0.13693108186125755, "time": 0.9113356351852417, "epoch": 3, "memory": 6319, "step": 8714} +{"lr": 8.91659267534649e-05, "data_time": 0.3996603965759277, "loss": 0.13073570877313614, "time": 0.9123415231704712, "epoch": 3, "memory": 6319, "step": 8814} +{"lr": 8.892933529894978e-05, "data_time": 0.5623992681503296, "loss": 0.1372727259993553, "time": 0.8533718585968018, "epoch": 3, "memory": 6319, "step": 8914} +{"lr": 8.869055010049283e-05, "data_time": 0.791880989074707, "loss": 0.12498634904623032, "time": 0.9333408594131469, "epoch": 3, "memory": 6319, "step": 9014} +{"lr": 8.844958659705188e-05, "data_time": 0.6925709486007691, "loss": 0.11703649088740349, "time": 0.9083391189575195, "epoch": 3, "memory": 6319, "step": 9114} +{"lr": 8.82064603684259e-05, "data_time": 0.6411150932312012, "loss": 0.11198518797755241, "time": 0.9106740236282349, "epoch": 3, "memory": 6319, "step": 9214} +{"lr": 8.796118713424776e-05, "data_time": 0.5770849227905274, "loss": 0.13326073661446572, "time": 0.9313672304153442, "epoch": 3, "memory": 6319, "step": 9314} +{"lr": 8.77137827529676e-05, "data_time": 0.5237188339233398, "loss": 0.1459883339703083, "time": 0.9073834180831909, "epoch": 3, "memory": 6319, "step": 9414} +{"lr": 8.746426322082745e-05, "data_time": 0.4845975399017334, "loss": 0.12167832925915718, "time": 0.8851630210876464, "epoch": 3, "memory": 6319, "step": 9514} +{"lr": 8.721264467082732e-05, "data_time": 0.22923238277435304, "loss": 0.10880739465355874, "time": 0.9161917448043824, "epoch": 3, "memory": 6319, "step": 9614} +{"lr": 8.695894337168168e-05, "data_time": 0.34072372913360593, "loss": 0.1253657318651676, "time": 0.9655329465866089, "epoch": 3, "memory": 6319, "step": 9714} +{"lr": 8.67031757267677e-05, "data_time": 0.41166934967041013, "loss": 0.13195557817816733, "time": 0.8773979425430298, "epoch": 3, "memory": 6319, "step": 9814} +{"lr": 8.64453582730651e-05, "data_time": 0.4863105535507202, "loss": 0.12704012244939805, "time": 1.162819743156433, "epoch": 3, "memory": 6319, "step": 9914} +{"lr": 8.61855076800861e-05, "data_time": 0.3985645294189453, "loss": 0.12885515242815018, "time": 0.9102446794509887, "epoch": 3, "memory": 6319, "step": 10014} +{"lr": 8.592364074879843e-05, "data_time": 0.775734543800354, "loss": 0.12236005663871766, "time": 1.061826515197754, "epoch": 3, "memory": 6319, "step": 10114} +{"lr": 8.565977441053854e-05, "data_time": 1.1374670028686524, "loss": 0.11657613888382912, "time": 1.2810584545135497, "epoch": 3, "memory": 6319, "step": 10214} +{"lr": 8.539392572591724e-05, "data_time": 0.7939578771591187, "loss": 0.1231521837413311, "time": 0.9340907573699951, "epoch": 3, "memory": 6319, "step": 10314} +{"lr": 8.512611188371633e-05, "data_time": 0.7361741065979004, "loss": 0.10086353421211243, "time": 0.8765170097351074, "epoch": 3, "memory": 6319, "step": 10414} +{"lr": 8.485635019977725e-05, "data_time": 0.7522613286972046, "loss": 0.13376380875706673, "time": 0.893633770942688, "epoch": 3, "memory": 6319, "step": 10514} +{"lr": 8.45846581158818e-05, "data_time": 0.7583223342895508, "loss": 0.1197498731315136, "time": 0.899726390838623, "epoch": 3, "memory": 6319, "step": 10614} +{"lr": 8.431105319862416e-05, "data_time": 0.7312403678894043, "loss": 0.11213548332452775, "time": 0.873240303993225, "epoch": 3, "memory": 6319, "step": 10714} +{"lr": 8.403555313827516e-05, "data_time": 0.826082444190979, "loss": 0.1139491319656372, "time": 0.9674846649169921, "epoch": 3, "memory": 6319, "step": 10814} +{"lr": 8.375817574763844e-05, "data_time": 0.7036544561386109, "loss": 0.12039667442440986, "time": 0.8431573867797851, "epoch": 3, "memory": 6319, "step": 10914} +{"lr": 8.347893896089871e-05, "data_time": 0.7328249931335449, "loss": 0.12288720905780792, "time": 0.8740258932113647, "epoch": 3, "memory": 6319, "step": 11014} +{"lr": 8.31978608324625e-05, "data_time": 0.8448924541473388, "loss": 0.11540027558803559, "time": 0.9855837106704712, "epoch": 3, "memory": 6319, "step": 11114} +{"lr": 8.291495953579052e-05, "data_time": 0.7429500102996827, "loss": 0.11476954445242882, "time": 0.8967215538024902, "epoch": 3, "memory": 6319, "step": 11214} +{"lr": 8.263025336222267e-05, "data_time": 0.7222877740859985, "loss": 0.11551371961832047, "time": 0.8638005971908569, "epoch": 3, "memory": 6319, "step": 11314} +{"lr": 8.234376071979551e-05, "data_time": 0.8286003828048706, "loss": 0.12651703283190727, "time": 0.9698094606399537, "epoch": 3, "memory": 6319, "step": 11414} +{"lr": 8.205550013205187e-05, "data_time": 0.7761763572692871, "loss": 0.11782191470265388, "time": 0.9162150859832764, "epoch": 3, "memory": 6319, "step": 11514} +{"lr": 8.17654902368435e-05, "data_time": 0.7461871862411499, "loss": 0.1229425273835659, "time": 0.8858694314956665, "epoch": 3, "memory": 6319, "step": 11614} +{"lr": 8.147374978512582e-05, "data_time": 0.739277458190918, "loss": 0.12535810321569443, "time": 0.8783728122711182, "epoch": 3, "memory": 6319, "step": 11714} +{"accuracy/top1": 91.27030944824219, "data_time": 0.5697703238191276, "time": 0.6578957047955744, "step": 3} +{"lr": 8.115969237626578e-05, "data_time": 0.776940393447876, "loss": 0.11489329561591148, "time": 0.9180193424224854, "epoch": 4, "memory": 6319, "step": 11821} +{"lr": 8.08644297337245e-05, "data_time": 0.758247685432434, "loss": 0.10636537447571755, "time": 0.8997403860092164, "epoch": 4, "memory": 6319, "step": 11921} +{"lr": 8.056749479387651e-05, "data_time": 0.7399152278900146, "loss": 0.12493508234620095, "time": 0.8834031820297241, "epoch": 4, "memory": 6319, "step": 12021} +{"lr": 8.026890675542457e-05, "data_time": 0.7429991960525513, "loss": 0.11613981425762177, "time": 0.884716796875, "epoch": 4, "memory": 6319, "step": 12121} +{"lr": 7.996868492395457e-05, "data_time": 0.7280483961105346, "loss": 0.11278834193944931, "time": 0.8697750568389893, "epoch": 4, "memory": 6319, "step": 12221} +{"lr": 7.966684871068762e-05, "data_time": 0.7241075992584228, "loss": 0.11355518251657486, "time": 0.8657039165496826, "epoch": 4, "memory": 6319, "step": 12321} +{"lr": 7.936341763122449e-05, "data_time": 0.784734058380127, "loss": 0.10069649256765842, "time": 0.925127363204956, "epoch": 4, "memory": 6319, "step": 12421} +{"lr": 7.905841130428405e-05, "data_time": 0.7919846534729004, "loss": 0.11319094225764274, "time": 0.9325732946395874, "epoch": 4, "memory": 6319, "step": 12521} +{"lr": 7.875184945043487e-05, "data_time": 0.7001120090484619, "loss": 0.1142881266772747, "time": 0.8399037599563599, "epoch": 4, "memory": 6319, "step": 12621} +{"lr": 7.844375189082012e-05, "data_time": 0.7048356771469116, "loss": 0.10352867394685746, "time": 0.8449249744415284, "epoch": 4, "memory": 6319, "step": 12721} +{"lr": 7.813413854587572e-05, "data_time": 0.7509239912033081, "loss": 0.10277210175991058, "time": 0.8908655405044555, "epoch": 4, "memory": 6319, "step": 12821} +{"lr": 7.782302943404291e-05, "data_time": 0.7470978260040283, "loss": 0.11459473595023155, "time": 0.8870609760284424, "epoch": 4, "memory": 6319, "step": 12921} +{"lr": 7.751044467047337e-05, "data_time": 0.7490935325622559, "loss": 0.12576298043131828, "time": 0.908761715888977, "epoch": 4, "memory": 6319, "step": 13021} +{"lr": 7.719640446572907e-05, "data_time": 0.7791104793548584, "loss": 0.10602714270353317, "time": 0.9199146747589111, "epoch": 4, "memory": 6319, "step": 13121} +{"lr": 7.688092912447536e-05, "data_time": 0.7819726705551148, "loss": 0.09999264776706696, "time": 0.9230399847030639, "epoch": 4, "memory": 6319, "step": 13221} +{"lr": 7.656403904416794e-05, "data_time": 0.7548886775970459, "loss": 0.09779963791370391, "time": 0.8953191041946411, "epoch": 4, "memory": 6319, "step": 13321} +{"lr": 7.624575471373434e-05, "data_time": 0.7260256767272949, "loss": 0.10628859400749206, "time": 0.8663904190063476, "epoch": 4, "memory": 6319, "step": 13421} +{"lr": 7.592609671224926e-05, "data_time": 0.761096715927124, "loss": 0.11050747781991958, "time": 0.9012486934661865, "epoch": 4, "memory": 6319, "step": 13521} +{"lr": 7.560508570760375e-05, "data_time": 0.7586737394332885, "loss": 0.09491719752550125, "time": 0.8995528221130371, "epoch": 4, "memory": 6319, "step": 13621} +{"lr": 7.528274245516895e-05, "data_time": 0.7230385780334473, "loss": 0.09878298342227936, "time": 0.8623899221420288, "epoch": 4, "memory": 6319, "step": 13721} +{"lr": 7.495908779645405e-05, "data_time": 0.5954001665115356, "loss": 0.12024580836296081, "time": 0.8468973159790039, "epoch": 4, "memory": 6319, "step": 13821} +{"lr": 7.463414265775906e-05, "data_time": 0.591273832321167, "loss": 0.09703774675726891, "time": 0.9595482349395752, "epoch": 4, "memory": 6319, "step": 13921} +{"lr": 7.430792804882147e-05, "data_time": 0.7594960451126098, "loss": 0.11120865792036057, "time": 0.9348649978637695, "epoch": 4, "memory": 6319, "step": 14021} +{"lr": 7.398046506145803e-05, "data_time": 0.7183664321899415, "loss": 0.09288974180817604, "time": 0.8592118740081787, "epoch": 4, "memory": 6319, "step": 14121} +{"lr": 7.365177486820114e-05, "data_time": 0.7816154479980468, "loss": 0.09406102634966373, "time": 0.9227701425552368, "epoch": 4, "memory": 6319, "step": 14221} +{"lr": 7.332187872092939e-05, "data_time": 0.7644946575164795, "loss": 0.08868381567299366, "time": 0.9041457891464233, "epoch": 4, "memory": 6319, "step": 14321} +{"lr": 7.299079794949401e-05, "data_time": 0.779997730255127, "loss": 0.12051670849323273, "time": 0.9211164474487304, "epoch": 4, "memory": 6319, "step": 14421} +{"lr": 7.265855396033955e-05, "data_time": 0.7451975107192993, "loss": 0.10627755373716355, "time": 0.8853612899780273, "epoch": 4, "memory": 6319, "step": 14521} +{"lr": 7.232516823512013e-05, "data_time": 0.6976796150207519, "loss": 0.10052933394908906, "time": 0.8378207921981812, "epoch": 4, "memory": 6319, "step": 14621} +{"lr": 7.199066232930995e-05, "data_time": 0.7325936794281006, "loss": 0.09236596971750259, "time": 0.8737106323242188, "epoch": 4, "memory": 6319, "step": 14721} +{"lr": 7.165505787080997e-05, "data_time": 0.7636126756668091, "loss": 0.10255631729960442, "time": 0.9047358512878418, "epoch": 4, "memory": 6319, "step": 14821} +{"lr": 7.13183765585495e-05, "data_time": 0.7070759534835815, "loss": 0.09638598337769508, "time": 0.8480723619461059, "epoch": 4, "memory": 6319, "step": 14921} +{"lr": 7.098064016108317e-05, "data_time": 0.7198184728622437, "loss": 0.1066634014248848, "time": 0.8599213123321533, "epoch": 4, "memory": 6319, "step": 15021} +{"lr": 7.064187051518379e-05, "data_time": 0.7827658891677857, "loss": 0.09756147637963294, "time": 0.9222442626953125, "epoch": 4, "memory": 6319, "step": 15121} +{"lr": 7.030208952442949e-05, "data_time": 0.730298638343811, "loss": 0.10235938653349877, "time": 0.8765639781951904, "epoch": 4, "memory": 6319, "step": 15221} +{"lr": 6.996131915778851e-05, "data_time": 0.7166630506515503, "loss": 0.1080032080411911, "time": 0.8568323850631714, "epoch": 4, "memory": 6319, "step": 15321} +{"lr": 6.961958144819853e-05, "data_time": 0.5675545692443847, "loss": 0.09821044877171517, "time": 0.894742226600647, "epoch": 4, "memory": 6319, "step": 15421} +{"lr": 6.927689849114193e-05, "data_time": 0.5460972309112548, "loss": 0.10202130004763603, "time": 0.8817622900009155, "epoch": 4, "memory": 6319, "step": 15521} +{"lr": 6.893329244321722e-05, "data_time": 0.4924583911895752, "loss": 0.10610084086656571, "time": 0.8512593269348144, "epoch": 4, "memory": 6319, "step": 15621} +{"accuracy/top1": 93.66758728027344, "data_time": 0.5766637530820123, "time": 0.6660525757690956, "step": 4} +{"lr": 6.856463687361792e-05, "data_time": 0.7157703638076782, "loss": 0.10067639276385307, "time": 0.8576092004776001, "epoch": 5, "memory": 6319, "step": 15728} +{"lr": 6.821919068470432e-05, "data_time": 0.6827544450759888, "loss": 0.1158062532544136, "time": 0.8226525545120239, "epoch": 5, "memory": 6319, "step": 15828} +{"lr": 6.787288979235407e-05, "data_time": 0.6815675735473633, "loss": 0.08799593523144722, "time": 0.821623969078064, "epoch": 5, "memory": 6319, "step": 15928} +{"lr": 6.752575658708793e-05, "data_time": 0.5739075183868408, "loss": 0.09230509996414185, "time": 0.8700387477874756, "epoch": 5, "memory": 6319, "step": 16028} +{"lr": 6.71778135132408e-05, "data_time": 0.3925312519073486, "loss": 0.09604379832744599, "time": 0.9017087697982789, "epoch": 5, "memory": 6319, "step": 16128} +{"lr": 6.682908306751074e-05, "data_time": 0.5438744783401489, "loss": 0.0960699513554573, "time": 0.9112030029296875, "epoch": 5, "memory": 6319, "step": 16228} +{"lr": 6.64795877975043e-05, "data_time": 0.6819400072097779, "loss": 0.09235613346099854, "time": 0.9370819807052613, "epoch": 5, "memory": 6319, "step": 16328} +{"lr": 6.61293503002787e-05, "data_time": 0.4025144100189209, "loss": 0.1131122749298811, "time": 0.8457817077636719, "epoch": 5, "memory": 6319, "step": 16428} +{"lr": 6.577839322088084e-05, "data_time": 0.5672608852386475, "loss": 0.10250139832496644, "time": 0.8598496437072753, "epoch": 5, "memory": 6319, "step": 16528} +{"lr": 6.542673925088298e-05, "data_time": 0.359615159034729, "loss": 0.09744763374328613, "time": 0.8780105590820313, "epoch": 5, "memory": 6319, "step": 16628} +{"lr": 6.507441112691584e-05, "data_time": 0.24308438301086427, "loss": 0.08741888999938965, "time": 0.820630669593811, "epoch": 5, "memory": 6319, "step": 16728} +{"lr": 6.47214316291983e-05, "data_time": 0.21722354888916015, "loss": 0.09268408082425594, "time": 0.8541023254394531, "epoch": 5, "memory": 6319, "step": 16828} +{"lr": 6.436782358006481e-05, "data_time": 0.20202741622924805, "loss": 0.09366943910717965, "time": 0.8682555913925171, "epoch": 5, "memory": 6319, "step": 16928} +{"lr": 6.40136098424896e-05, "data_time": 0.2511094808578491, "loss": 0.09410080313682556, "time": 0.9106420516967774, "epoch": 5, "memory": 6319, "step": 17028} +{"lr": 6.365881331860828e-05, "data_time": 0.07914295196533203, "loss": 0.11396202370524407, "time": 0.9080969095230103, "epoch": 5, "memory": 6319, "step": 17128} +{"lr": 6.330345694823739e-05, "data_time": 0.00127866268157959, "loss": 0.09479823037981987, "time": 0.8430700778961182, "epoch": 5, "memory": 6319, "step": 17228} +{"lr": 6.29475637073908e-05, "data_time": 0.0009033679962158203, "loss": 0.08776037395000458, "time": 0.9023783206939697, "epoch": 5, "memory": 6319, "step": 17328} +{"lr": 6.259115660679482e-05, "data_time": 0.0014479637145996093, "loss": 0.08252049572765827, "time": 0.9442789077758789, "epoch": 5, "memory": 6319, "step": 17428} +{"lr": 6.223425869039963e-05, "data_time": 0.00141448974609375, "loss": 0.09708419069647789, "time": 0.7980317115783692, "epoch": 5, "memory": 6319, "step": 17528} +{"lr": 6.187689303389006e-05, "data_time": 0.001473402976989746, "loss": 0.09579034522175789, "time": 0.9105540990829468, "epoch": 5, "memory": 6319, "step": 17628} +{"lr": 6.151908274319292e-05, "data_time": 0.0012321710586547852, "loss": 0.08410843573510647, "time": 0.8482635021209717, "epoch": 5, "memory": 6319, "step": 17728} +{"lr": 6.116085095298373e-05, "data_time": 0.0011002063751220704, "loss": 0.11261978149414062, "time": 0.8731829166412354, "epoch": 5, "memory": 6319, "step": 17828} +{"lr": 6.080222082519007e-05, "data_time": 0.00096435546875, "loss": 0.09185517057776452, "time": 0.9497179985046387, "epoch": 5, "memory": 6319, "step": 17928} +{"lr": 6.0443215547495147e-05, "data_time": 0.0015970706939697266, "loss": 0.09255757033824921, "time": 0.8956833839416504, "epoch": 5, "memory": 6319, "step": 18028} +{"lr": 6.0083858331837714e-05, "data_time": 0.000991487503051758, "loss": 0.09854256957769394, "time": 0.8653445720672608, "epoch": 5, "memory": 6319, "step": 18128} +{"lr": 5.97241724129116e-05, "data_time": 0.0011955022811889649, "loss": 0.09508971720933915, "time": 0.9046830415725708, "epoch": 5, "memory": 6319, "step": 18228} +{"lr": 5.936418104666328e-05, "data_time": 0.0016076326370239257, "loss": 0.07920684032142163, "time": 0.8601744413375855, "epoch": 5, "memory": 6319, "step": 18328} +{"lr": 5.9003907508788295e-05, "data_time": 0.0023520708084106444, "loss": 0.1014905646443367, "time": 0.884795355796814, "epoch": 5, "memory": 6319, "step": 18428} +{"lr": 5.864337509322633e-05, "data_time": 0.001106405258178711, "loss": 0.09967180341482162, "time": 0.9008808851242065, "epoch": 5, "memory": 6319, "step": 18528} +{"lr": 5.8282607110655194e-05, "data_time": 0.0013938426971435546, "loss": 0.1027616836130619, "time": 0.9279417276382447, "epoch": 5, "memory": 6319, "step": 18628} +{"lr": 5.7921626886983564e-05, "data_time": 0.0018150091171264648, "loss": 0.08932366222143173, "time": 0.8468700170516967, "epoch": 5, "memory": 6319, "step": 18728} +{"lr": 5.756045776184287e-05, "data_time": 0.0013753652572631836, "loss": 0.11413226500153542, "time": 0.8648529291152954, "epoch": 5, "memory": 6319, "step": 18828} +{"lr": 5.719912308707808e-05, "data_time": 0.0014557361602783204, "loss": 0.09509230181574821, "time": 0.8548434019088745, "epoch": 5, "memory": 6319, "step": 18928} +{"lr": 5.683764622523808e-05, "data_time": 0.0008451461791992188, "loss": 0.10767360329627991, "time": 0.8595911979675293, "epoch": 5, "memory": 6319, "step": 19028} +{"lr": 5.647605054806502e-05, "data_time": 0.006430959701538086, "loss": 0.09613983780145645, "time": 0.8443287372589111, "epoch": 5, "memory": 6319, "step": 19128} +{"lr": 5.6114359434983206e-05, "data_time": 0.2763890027999878, "loss": 0.08913504853844642, "time": 0.8781718492507935, "epoch": 5, "memory": 6319, "step": 19228} +{"lr": 5.575259627158745e-05, "data_time": 0.10579869747161866, "loss": 0.08754120543599128, "time": 0.847752594947815, "epoch": 5, "memory": 6319, "step": 19328} +{"lr": 5.539078444813113e-05, "data_time": 0.33369197845458987, "loss": 0.1057322308421135, "time": 0.82184317111969, "epoch": 5, "memory": 6319, "step": 19428} +{"lr": 5.5028947358013735e-05, "data_time": 0.42878286838531493, "loss": 0.08728051409125329, "time": 0.8530410528182983, "epoch": 5, "memory": 6319, "step": 19528} +{"accuracy/top1": 94.79141235351562, "data_time": 0.5715527205631651, "time": 0.6595231459058565, "step": 5} +{"lr": 5.464178020340108e-05, "data_time": 0.6957848787307739, "loss": 0.08735099732875824, "time": 0.8367028951644897, "epoch": 6, "memory": 6319, "step": 19635} +{"lr": 5.427996514795046e-05, "data_time": 0.3696676015853882, "loss": 0.10282282382249833, "time": 0.8561572790145874, "epoch": 6, "memory": 6319, "step": 19735} +{"lr": 5.3918196647260834e-05, "data_time": 0.0011491060256958007, "loss": 0.09733559787273408, "time": 0.8947566270828247, "epoch": 6, "memory": 6319, "step": 19835} +{"lr": 5.355649809193057e-05, "data_time": 0.4860093116760254, "loss": 0.08619643189013004, "time": 0.9094852924346923, "epoch": 6, "memory": 6319, "step": 19935} +{"lr": 5.3194892868035765e-05, "data_time": 0.3416721343994141, "loss": 0.0918333400040865, "time": 0.8785401821136475, "epoch": 6, "memory": 6319, "step": 20035} +{"lr": 5.283340435561795e-05, "data_time": 0.31291096210479735, "loss": 0.09346374906599522, "time": 0.8457836389541626, "epoch": 6, "memory": 6319, "step": 20135} +{"lr": 5.2472055927172495e-05, "data_time": 0.2621103048324585, "loss": 0.09200282096862793, "time": 0.8982357501983642, "epoch": 6, "memory": 6319, "step": 20235} +{"lr": 5.2110870946137605e-05, "data_time": 0.20984821319580077, "loss": 0.0933717668056488, "time": 0.8591134548187256, "epoch": 6, "memory": 6319, "step": 20335} +{"lr": 5.174987276538342e-05, "data_time": 0.2419429302215576, "loss": 0.10093883015215396, "time": 0.859332537651062, "epoch": 6, "memory": 6319, "step": 20435} +{"lr": 5.1389084725702425e-05, "data_time": 0.2607812643051147, "loss": 0.08994136080145836, "time": 0.8670778751373291, "epoch": 6, "memory": 6319, "step": 20535} +{"lr": 5.1028530154300045e-05, "data_time": 0.18812243938446044, "loss": 0.09303448684513568, "time": 0.8788151025772095, "epoch": 6, "memory": 6319, "step": 20635} +{"lr": 5.066823236328663e-05, "data_time": 0.2563325881958008, "loss": 0.08541717454791069, "time": 0.8980855226516724, "epoch": 6, "memory": 6319, "step": 20735} +{"lr": 5.030821464817006e-05, "data_time": 0.2645404577255249, "loss": 0.08788195103406907, "time": 0.8827181339263916, "epoch": 6, "memory": 6319, "step": 20835} +{"lr": 4.99485002863494e-05, "data_time": 0.3766984701156616, "loss": 0.09297304302453994, "time": 0.8823019742965699, "epoch": 6, "memory": 6319, "step": 20935} +{"lr": 4.95891125356102e-05, "data_time": 0.4078476905822754, "loss": 0.08156409449875354, "time": 0.9584446907043457, "epoch": 6, "memory": 6319, "step": 21035} +{"lr": 4.923007463262053e-05, "data_time": 0.10058321952819824, "loss": 0.07987777292728424, "time": 0.9098809242248536, "epoch": 6, "memory": 6319, "step": 21135} +{"lr": 4.887140979142852e-05, "data_time": 0.08175497055053711, "loss": 0.0818291261792183, "time": 0.8497950792312622, "epoch": 6, "memory": 6319, "step": 21235} +{"lr": 4.851314120196161e-05, "data_time": 0.011955714225769043, "loss": 0.08152264468371868, "time": 0.8745648622512817, "epoch": 6, "memory": 6319, "step": 21335} +{"lr": 4.815529202852693e-05, "data_time": 0.13461427688598632, "loss": 0.08553754091262818, "time": 0.9074079513549804, "epoch": 6, "memory": 6319, "step": 21435} +{"lr": 4.779788540831406e-05, "data_time": 0.18175103664398193, "loss": 0.09324220716953277, "time": 0.8948556423187256, "epoch": 6, "memory": 6319, "step": 21535} +{"lr": 4.744094444989843e-05, "data_time": 0.23326785564422609, "loss": 0.08899538591504097, "time": 0.8990905284881592, "epoch": 6, "memory": 6319, "step": 21635} +{"lr": 4.708449223174764e-05, "data_time": 0.1924884557723999, "loss": 0.08449041098356247, "time": 0.8605105876922607, "epoch": 6, "memory": 6319, "step": 21735} +{"lr": 4.672855180072916e-05, "data_time": 0.0010923385620117188, "loss": 0.0837862677872181, "time": 0.8763978004455566, "epoch": 6, "memory": 6319, "step": 21835} +{"lr": 4.6373146170620256e-05, "data_time": 0.0008182525634765625, "loss": 0.09838488847017288, "time": 0.8523350954055786, "epoch": 6, "memory": 6319, "step": 21935} +{"lr": 4.6018298320619915e-05, "data_time": 0.03775930404663086, "loss": 0.0832609124481678, "time": 0.835089898109436, "epoch": 6, "memory": 6319, "step": 22035} +{"lr": 4.566403119386318e-05, "data_time": 0.1655141830444336, "loss": 0.08210644125938416, "time": 0.8597707271575927, "epoch": 6, "memory": 6319, "step": 22135} +{"lr": 4.531036769593769e-05, "data_time": 0.0012761592864990235, "loss": 0.07907106652855873, "time": 0.9598047971725464, "epoch": 6, "memory": 6319, "step": 22235} +{"lr": 4.4957330693402685e-05, "data_time": 0.0012713432312011718, "loss": 0.07999893091619015, "time": 0.9161706924438476, "epoch": 6, "memory": 6319, "step": 22335} +{"lr": 4.460494301231059e-05, "data_time": 0.0011863231658935547, "loss": 0.08322003372013569, "time": 0.8416136026382446, "epoch": 6, "memory": 6319, "step": 22435} +{"lr": 4.425322743673109e-05, "data_time": 0.0013361215591430665, "loss": 0.0836469754576683, "time": 0.8517721652984619, "epoch": 6, "memory": 6319, "step": 22535} +{"lr": 4.390220670727812e-05, "data_time": 0.0010993242263793944, "loss": 0.09141123667359352, "time": 0.8202883481979371, "epoch": 6, "memory": 6319, "step": 22635} +{"lr": 4.3551903519639325e-05, "data_time": 0.0015716791152954102, "loss": 0.07976200245320797, "time": 0.9105865955352783, "epoch": 6, "memory": 6319, "step": 22735} +{"lr": 4.3202340523108894e-05, "data_time": 0.0014548778533935546, "loss": 0.08041631653904915, "time": 0.8781120300292968, "epoch": 6, "memory": 6319, "step": 22835} +{"lr": 4.285354031912302e-05, "data_time": 0.0010716915130615234, "loss": 0.08340962044894695, "time": 0.8195673227310181, "epoch": 6, "memory": 6319, "step": 22935} +{"lr": 4.2505525459798544e-05, "data_time": 0.0014203786849975586, "loss": 0.08337981961667537, "time": 0.823271656036377, "epoch": 6, "memory": 6319, "step": 23035} +{"lr": 4.2158318446474836e-05, "data_time": 0.001408839225769043, "loss": 0.07958810180425643, "time": 0.8592772245407104, "epoch": 6, "memory": 6319, "step": 23135} +{"lr": 4.181194172825905e-05, "data_time": 0.001300811767578125, "loss": 0.07475061267614365, "time": 0.915073823928833, "epoch": 6, "memory": 6319, "step": 23235} +{"lr": 4.146641770057448e-05, "data_time": 0.0018882513046264648, "loss": 0.0789183296263218, "time": 0.9030552387237549, "epoch": 6, "memory": 6319, "step": 23335} +{"lr": 4.112176870371267e-05, "data_time": 0.0014145851135253906, "loss": 0.08289809748530388, "time": 0.8818341016769409, "epoch": 6, "memory": 6319, "step": 23435} +{"accuracy/top1": 95.6890869140625, "data_time": 0.5603013326381815, "time": 0.6510588382852489, "step": 6} +{"lr": 4.0753988582732786e-05, "data_time": 0.6458329916000366, "loss": 0.08821750245988369, "time": 0.8787811040878296, "epoch": 7, "memory": 6319, "step": 23542} +{"lr": 4.0411221639286815e-05, "data_time": 0.7044112920761109, "loss": 0.0914335660636425, "time": 0.9160228729248047, "epoch": 7, "memory": 6319, "step": 23642} +{"lr": 4.006939795169518e-05, "data_time": 0.4019924163818359, "loss": 0.07337699197232724, "time": 0.8741212368011475, "epoch": 7, "memory": 6319, "step": 23742} +{"lr": 3.972853962099913e-05, "data_time": 0.2172248363494873, "loss": 0.0989815104752779, "time": 0.9004807472229004, "epoch": 7, "memory": 6319, "step": 23842} +{"lr": 3.93886686858237e-05, "data_time": 0.28595924377441406, "loss": 0.10283050090074539, "time": 0.8484663724899292, "epoch": 7, "memory": 6319, "step": 23942} +{"lr": 3.904980712095259e-05, "data_time": 0.3061556100845337, "loss": 0.07841564007103444, "time": 0.8925023555755616, "epoch": 7, "memory": 6319, "step": 24042} +{"lr": 3.871197683590748e-05, "data_time": 0.21903626918792723, "loss": 0.07981203086674213, "time": 0.8458019733428955, "epoch": 7, "memory": 6319, "step": 24142} +{"lr": 3.837519967353125e-05, "data_time": 0.2449690580368042, "loss": 0.08555740900337697, "time": 0.8858100652694703, "epoch": 7, "memory": 6319, "step": 24242} +{"lr": 3.8039497408575855e-05, "data_time": 0.1595212697982788, "loss": 0.08635051213204861, "time": 0.8840290307998657, "epoch": 7, "memory": 6319, "step": 24342} +{"lr": 3.770489174629432e-05, "data_time": 0.12182822227478027, "loss": 0.07224807813763619, "time": 0.869728422164917, "epoch": 7, "memory": 6319, "step": 24442} +{"lr": 3.737140432103754e-05, "data_time": 0.19448225498199462, "loss": 0.08117571622133254, "time": 0.9084516048431397, "epoch": 7, "memory": 6319, "step": 24542} +{"lr": 3.703905669485545e-05, "data_time": 0.15427806377410888, "loss": 0.08219698742032051, "time": 0.877875804901123, "epoch": 7, "memory": 6319, "step": 24642} +{"lr": 3.6707870356102706e-05, "data_time": 0.18405275344848632, "loss": 0.08146934397518635, "time": 0.9229199647903442, "epoch": 7, "memory": 6319, "step": 24742} +{"lr": 3.6377866718049484e-05, "data_time": 0.0011170387268066406, "loss": 0.09072392359375954, "time": 0.9015771865844726, "epoch": 7, "memory": 6319, "step": 24842} +{"lr": 3.6049067117496965e-05, "data_time": 0.0011453628540039062, "loss": 0.08101946152746678, "time": 0.8718708515167236, "epoch": 7, "memory": 6319, "step": 24942} +{"lr": 3.5721492813397755e-05, "data_time": 0.0017980337142944336, "loss": 0.08431394286453724, "time": 0.880272102355957, "epoch": 7, "memory": 6319, "step": 25042} +{"lr": 3.539516498548151e-05, "data_time": 0.0012185096740722657, "loss": 0.07376923486590385, "time": 0.9023571014404297, "epoch": 7, "memory": 6319, "step": 25142} +{"lr": 3.507010473288518e-05, "data_time": 0.0011278867721557617, "loss": 0.08985093757510185, "time": 0.8491546630859375, "epoch": 7, "memory": 6319, "step": 25242} +{"lr": 3.474633307278909e-05, "data_time": 0.031475830078125, "loss": 0.09947454854846001, "time": 0.8629065990447998, "epoch": 7, "memory": 6319, "step": 25342} +{"lr": 3.4423870939058177e-05, "data_time": 0.4481214046478271, "loss": 0.08215671852231025, "time": 0.9036730527877808, "epoch": 7, "memory": 6319, "step": 25442} +{"lr": 3.410273918088805e-05, "data_time": 0.3497700452804565, "loss": 0.07424916923046113, "time": 0.8509632587432862, "epoch": 7, "memory": 6319, "step": 25542} +{"lr": 3.37829585614574e-05, "data_time": 0.20111565589904784, "loss": 0.08338887766003608, "time": 0.9196459531784058, "epoch": 7, "memory": 6319, "step": 25642} +{"lr": 3.3464549756585216e-05, "data_time": 0.001168227195739746, "loss": 0.07143484726548195, "time": 0.9184943675994873, "epoch": 7, "memory": 6319, "step": 25742} +{"lr": 3.314753335339416e-05, "data_time": 0.0012459754943847656, "loss": 0.07596002370119095, "time": 0.9016108989715577, "epoch": 7, "memory": 6319, "step": 25842} +{"lr": 3.2831929848979394e-05, "data_time": 0.00139310359954834, "loss": 0.08242843970656395, "time": 0.9291356086730957, "epoch": 7, "memory": 6319, "step": 25942} +{"lr": 3.251775964908332e-05, "data_time": 0.0008750200271606445, "loss": 0.09521238580346107, "time": 0.8514190673828125, "epoch": 7, "memory": 6319, "step": 26042} +{"lr": 3.22050430667762e-05, "data_time": 0.0010126352310180664, "loss": 0.07639812268316745, "time": 0.8465380907058716, "epoch": 7, "memory": 6319, "step": 26142} +{"lr": 3.189380032114285e-05, "data_time": 0.16205165386199952, "loss": 0.07312794998288155, "time": 0.9169226884841919, "epoch": 7, "memory": 6319, "step": 26242} +{"lr": 3.1584051535975324e-05, "data_time": 0.5428041934967041, "loss": 0.09266364872455597, "time": 0.9132752656936646, "epoch": 7, "memory": 6319, "step": 26342} +{"lr": 3.127581673847175e-05, "data_time": 0.5764893770217896, "loss": 0.08598207011818886, "time": 0.9185611724853515, "epoch": 7, "memory": 6319, "step": 26442} +{"lr": 3.0969115857941555e-05, "data_time": 0.27761924266815186, "loss": 0.09996403455734253, "time": 0.8727656841278076, "epoch": 7, "memory": 6319, "step": 26542} +{"lr": 3.0663968724516654e-05, "data_time": 0.41826543807983396, "loss": 0.0922010935842991, "time": 0.889160418510437, "epoch": 7, "memory": 6319, "step": 26642} +{"lr": 3.0360395067869668e-05, "data_time": 0.4560523271560669, "loss": 0.07513608299195766, "time": 0.8984469413757324, "epoch": 7, "memory": 6319, "step": 26742} +{"lr": 3.0058414515938063e-05, "data_time": 0.029741072654724122, "loss": 0.0829878356307745, "time": 0.8723999500274658, "epoch": 7, "memory": 6319, "step": 26842} +{"lr": 2.975804659365501e-05, "data_time": 0.0014225244522094727, "loss": 0.08196965903043747, "time": 0.8736865758895874, "epoch": 7, "memory": 6319, "step": 26942} +{"lr": 2.945931072168714e-05, "data_time": 0.0012315034866333008, "loss": 0.09356333091855049, "time": 0.8833344459533692, "epoch": 7, "memory": 6319, "step": 27042} +{"lr": 2.9162226215178855e-05, "data_time": 0.0011543512344360351, "loss": 0.07178382985293866, "time": 0.9196656227111817, "epoch": 7, "memory": 6319, "step": 27142} +{"lr": 2.8866812282503322e-05, "data_time": 0.0011996030807495117, "loss": 0.07366007119417191, "time": 0.8804421663284302, "epoch": 7, "memory": 6319, "step": 27242} +{"lr": 2.8573088024020838e-05, "data_time": 0.0008530139923095704, "loss": 0.06969595476984977, "time": 0.8878621578216552, "epoch": 7, "memory": 6319, "step": 27342} +{"accuracy/top1": 96.15531921386719, "data_time": 0.5790184900678438, "time": 0.6665845451683834, "step": 7} +{"lr": 2.8260695817390092e-05, "data_time": 0.773190975189209, "loss": 0.0676898017525673, "time": 0.9133787631988526, "epoch": 8, "memory": 6319, "step": 27449} +{"lr": 2.7970529402570967e-05, "data_time": 0.7257717609405517, "loss": 0.08663739562034607, "time": 0.8652721643447876, "epoch": 8, "memory": 6319, "step": 27549} +{"lr": 2.7682110612244214e-05, "data_time": 0.7518941640853882, "loss": 0.08318739645183086, "time": 0.8933749675750733, "epoch": 8, "memory": 6319, "step": 27649} +{"lr": 2.7395458094490283e-05, "data_time": 0.7072431564331054, "loss": 0.08465914279222489, "time": 0.8481247425079346, "epoch": 8, "memory": 6319, "step": 27749} +{"lr": 2.7110590383189015e-05, "data_time": 0.7882378816604614, "loss": 0.08485641404986381, "time": 0.9298220157623291, "epoch": 8, "memory": 6319, "step": 27849} +{"lr": 2.6827525896821397e-05, "data_time": 0.4758735656738281, "loss": 0.08296169936656952, "time": 1.0543368577957153, "epoch": 8, "memory": 6319, "step": 27949} +{"lr": 2.6546282937278552e-05, "data_time": 0.06723496913909913, "loss": 0.08053266108036042, "time": 0.8654393196105957, "epoch": 8, "memory": 6319, "step": 28049} +{"lr": 2.6266879688678606e-05, "data_time": 0.12370157241821289, "loss": 0.0872628390789032, "time": 0.8752964973449707, "epoch": 8, "memory": 6319, "step": 28149} +{"lr": 2.5989334216190738e-05, "data_time": 0.4160438537597656, "loss": 0.07510889694094658, "time": 0.8809774398803711, "epoch": 8, "memory": 6319, "step": 28249} +{"lr": 2.571366446486732e-05, "data_time": 0.7109492301940918, "loss": 0.0779775895178318, "time": 0.8698572635650634, "epoch": 8, "memory": 6319, "step": 28349} +{"lr": 2.5439888258483665e-05, "data_time": 0.5894846439361572, "loss": 0.08027538731694221, "time": 0.8340277433395386, "epoch": 8, "memory": 6319, "step": 28449} +{"lr": 2.5168023298385384e-05, "data_time": 0.6976788520812989, "loss": 0.0870208889245987, "time": 0.9264285564422607, "epoch": 8, "memory": 6319, "step": 28549} +{"lr": 2.489808716234412e-05, "data_time": 0.5857330560684204, "loss": 0.07486210726201534, "time": 0.886945366859436, "epoch": 8, "memory": 6319, "step": 28649} +{"lr": 2.463009730342111e-05, "data_time": 0.6903819084167481, "loss": 0.0905293308198452, "time": 0.8788272619247437, "epoch": 8, "memory": 6319, "step": 28749} +{"lr": 2.436407104883849e-05, "data_time": 0.7645008563995361, "loss": 0.0734794195741415, "time": 0.9070576906204224, "epoch": 8, "memory": 6319, "step": 28849} +{"lr": 2.4100025598859092e-05, "data_time": 0.7549132585525513, "loss": 0.0857387587428093, "time": 0.8952568292617797, "epoch": 8, "memory": 6319, "step": 28949} +{"lr": 2.3837978025674284e-05, "data_time": 0.634507131576538, "loss": 0.07850695848464966, "time": 0.8755037307739257, "epoch": 8, "memory": 6319, "step": 29049} +{"lr": 2.357794527230027e-05, "data_time": 0.7431282997131348, "loss": 0.0772418100386858, "time": 0.8832640409469604, "epoch": 8, "memory": 6319, "step": 29149} +{"lr": 2.33199441514825e-05, "data_time": 0.7974025964736938, "loss": 0.08501079007983207, "time": 0.9383984088897706, "epoch": 8, "memory": 6319, "step": 29249} +{"lr": 2.306399134460859e-05, "data_time": 0.7360717058181763, "loss": 0.08238439373672009, "time": 0.8765291929244995, "epoch": 8, "memory": 6319, "step": 29349} +{"lr": 2.281010340062995e-05, "data_time": 0.7424033403396606, "loss": 0.09331198707222939, "time": 0.8944615840911865, "epoch": 8, "memory": 6319, "step": 29449} +{"lr": 2.2558296734991603e-05, "data_time": 0.7500584602355957, "loss": 0.07346097975969315, "time": 0.8896821498870849, "epoch": 8, "memory": 6319, "step": 29549} +{"lr": 2.2308587628570915e-05, "data_time": 0.7491367340087891, "loss": 0.08215555846691132, "time": 0.8904971122741699, "epoch": 8, "memory": 6319, "step": 29649} +{"lr": 2.2060992226624955e-05, "data_time": 0.7169220209121704, "loss": 0.0714668869972229, "time": 0.8568846464157105, "epoch": 8, "memory": 6319, "step": 29749} +{"lr": 2.181552653774648e-05, "data_time": 0.7214415788650512, "loss": 0.07901271656155587, "time": 0.8694997549057006, "epoch": 8, "memory": 6319, "step": 29849} +{"lr": 2.1572206432829035e-05, "data_time": 0.7497676610946655, "loss": 0.07336163073778153, "time": 0.8897875070571899, "epoch": 8, "memory": 6319, "step": 29949} +{"lr": 2.1331047644040754e-05, "data_time": 0.7227826356887818, "loss": 0.08376354165375233, "time": 0.8631266593933106, "epoch": 8, "memory": 6319, "step": 30049} +{"lr": 2.109206576380705e-05, "data_time": 0.7716203212738038, "loss": 0.08364129476249219, "time": 0.9124569177627564, "epoch": 8, "memory": 6319, "step": 30149} +{"lr": 2.0855276243802687e-05, "data_time": 0.7222044229507446, "loss": 0.07447917945683002, "time": 0.8624828338623047, "epoch": 8, "memory": 6319, "step": 30249} +{"lr": 2.0620694393952487e-05, "data_time": 0.6082808494567871, "loss": 0.0637462567538023, "time": 0.9054466009140014, "epoch": 8, "memory": 6319, "step": 30349} +{"lr": 2.038833538144169e-05, "data_time": 0.6377532243728637, "loss": 0.07222930155694485, "time": 0.8785322666168213, "epoch": 8, "memory": 6319, "step": 30449} +{"lr": 2.0158214229735154e-05, "data_time": 0.6940939903259278, "loss": 0.07766427472233772, "time": 0.8363755941390991, "epoch": 8, "memory": 6319, "step": 30549} +{"lr": 1.9930345817605977e-05, "data_time": 0.717761754989624, "loss": 0.08640623912215233, "time": 0.9249245882034302, "epoch": 8, "memory": 6319, "step": 30649} +{"lr": 1.970474487817358e-05, "data_time": 0.7321541547775269, "loss": 0.0824542447924614, "time": 0.8729692220687866, "epoch": 8, "memory": 6319, "step": 30749} +{"lr": 1.9481425997951092e-05, "data_time": 0.7570785045623779, "loss": 0.08402719981968403, "time": 0.8991684675216675, "epoch": 8, "memory": 6319, "step": 30849} +{"lr": 1.9260403615902187e-05, "data_time": 0.7063649892807007, "loss": 0.07551782429218293, "time": 0.8475235223770141, "epoch": 8, "memory": 6319, "step": 30949} +{"lr": 1.9041692022507647e-05, "data_time": 0.7113152265548706, "loss": 0.07760469317436218, "time": 0.8525068998336792, "epoch": 8, "memory": 6319, "step": 31049} +{"lr": 1.8825305358841125e-05, "data_time": 0.7466799259185791, "loss": 0.0744364645332098, "time": 0.8883065938949585, "epoch": 8, "memory": 6319, "step": 31149} +{"lr": 1.861125761565518e-05, "data_time": 0.5244179010391236, "loss": 0.07091340497136116, "time": 0.8617112398147583, "epoch": 8, "memory": 6319, "step": 31249} +{"accuracy/top1": 96.42671203613281, "data_time": 0.5710797926475262, "time": 0.6589143235107948, "step": 8} +{"lr": 1.8384832449122864e-05, "data_time": 0.40781588554382325, "loss": 0.07775134034454823, "time": 0.9066422462463379, "epoch": 9, "memory": 6319, "step": 31356} +{"lr": 1.817567007332114e-05, "data_time": 0.41288764476776124, "loss": 0.07856393307447433, "time": 0.9140744686126709, "epoch": 9, "memory": 6319, "step": 31456} +{"lr": 1.796888862098825e-05, "data_time": 0.49449379444122316, "loss": 0.08160603269934655, "time": 0.853122067451477, "epoch": 9, "memory": 6319, "step": 31556} +{"lr": 1.7764501461839678e-05, "data_time": 0.7420816659927368, "loss": 0.08109548687934875, "time": 0.8830325365066528, "epoch": 9, "memory": 6319, "step": 31656} +{"lr": 1.7562521810784746e-05, "data_time": 0.709919023513794, "loss": 0.06176501996815205, "time": 0.8515275001525879, "epoch": 9, "memory": 6319, "step": 31756} +{"lr": 1.7362962727072444e-05, "data_time": 0.43931331634521487, "loss": 0.07960855662822723, "time": 0.8904636859893799, "epoch": 9, "memory": 6319, "step": 31856} +{"lr": 1.716583711344682e-05, "data_time": 0.29276366233825685, "loss": 0.06693844497203827, "time": 0.8681150674819946, "epoch": 9, "memory": 6319, "step": 31956} +{"lr": 1.6971157715312954e-05, "data_time": 0.28310532569885255, "loss": 0.100167166441679, "time": 0.8865897417068481, "epoch": 9, "memory": 6319, "step": 32056} +{"lr": 1.677893711991274e-05, "data_time": 0.3116255044937134, "loss": 0.0696358185261488, "time": 0.9002179384231568, "epoch": 9, "memory": 6319, "step": 32156} +{"lr": 1.6589187755511025e-05, "data_time": 0.3039934396743774, "loss": 0.07503429688513279, "time": 0.8633481740951539, "epoch": 9, "memory": 6319, "step": 32256} +{"lr": 1.640192189059216e-05, "data_time": 0.0011800289154052734, "loss": 0.07262629978358745, "time": 0.8771113634109498, "epoch": 9, "memory": 6319, "step": 32356} +{"lr": 1.6217151633066716e-05, "data_time": 0.0017085790634155274, "loss": 0.0713750522583723, "time": 0.9277477264404297, "epoch": 9, "memory": 6319, "step": 32456} +{"lr": 1.6034888929488564e-05, "data_time": 0.0017104148864746094, "loss": 0.07639891132712365, "time": 0.8678189516067505, "epoch": 9, "memory": 6319, "step": 32556} +{"lr": 1.585514556428258e-05, "data_time": 0.0013008356094360352, "loss": 0.06912901028990745, "time": 0.8992761611938477, "epoch": 9, "memory": 6319, "step": 32656} +{"lr": 1.567793315898261e-05, "data_time": 0.0017913579940795898, "loss": 0.07725722081959248, "time": 0.8712166547775269, "epoch": 9, "memory": 6319, "step": 32756} +{"lr": 1.5503263171480048e-05, "data_time": 0.09125230312347413, "loss": 0.07543345391750336, "time": 0.8847838401794433, "epoch": 9, "memory": 6319, "step": 32856} +{"lr": 1.5331146895283165e-05, "data_time": 0.5145464181900025, "loss": 0.08202039748430252, "time": 0.9210999011993408, "epoch": 9, "memory": 6319, "step": 32956} +{"lr": 1.5161595458786728e-05, "data_time": 0.3576303720474243, "loss": 0.08256775140762329, "time": 0.8308727979660034, "epoch": 9, "memory": 6319, "step": 33056} +{"lr": 1.4994619824552563e-05, "data_time": 0.5439150333404541, "loss": 0.07742651663720608, "time": 0.8744504928588868, "epoch": 9, "memory": 6319, "step": 33156} +{"lr": 1.4830230788600802e-05, "data_time": 0.6360100984573365, "loss": 0.07649558186531066, "time": 0.8381055355072021, "epoch": 9, "memory": 6319, "step": 33256} +{"lr": 1.4668438979711779e-05, "data_time": 0.7989260911941528, "loss": 0.07968887686729431, "time": 0.9380785226821899, "epoch": 9, "memory": 6319, "step": 33356} +{"lr": 1.4509254858738844e-05, "data_time": 0.7136792421340943, "loss": 0.08204310238361359, "time": 0.8539438724517823, "epoch": 9, "memory": 6319, "step": 33456} +{"lr": 1.4352688717931966e-05, "data_time": 0.7313532590866089, "loss": 0.07945706844329833, "time": 0.874568772315979, "epoch": 9, "memory": 6319, "step": 33556} +{"lr": 1.4198750680272348e-05, "data_time": 0.6613088607788086, "loss": 0.0904395893216133, "time": 0.8789570808410645, "epoch": 9, "memory": 6319, "step": 33656} +{"lr": 1.4047450698817857e-05, "data_time": 0.7777715921401978, "loss": 0.07187089733779431, "time": 0.9182924032211304, "epoch": 9, "memory": 6319, "step": 33756} +{"lr": 1.389879855605952e-05, "data_time": 0.7126684188842773, "loss": 0.0718031745404005, "time": 0.857210898399353, "epoch": 9, "memory": 6319, "step": 33856} +{"lr": 1.3752803863288976e-05, "data_time": 0.7748696327209472, "loss": 0.0728682342916727, "time": 0.9158221006393432, "epoch": 9, "memory": 6319, "step": 33956} +{"lr": 1.3609476059977129e-05, "data_time": 0.7824553489685059, "loss": 0.07123819030821324, "time": 0.9341459274291992, "epoch": 9, "memory": 6319, "step": 34056} +{"lr": 1.3468824413163754e-05, "data_time": 0.7434473514556885, "loss": 0.0806659147143364, "time": 0.8847616910934448, "epoch": 9, "memory": 6319, "step": 34156} +{"lr": 1.3330858016858395e-05, "data_time": 0.7036133050918579, "loss": 0.08593320660293102, "time": 0.844521164894104, "epoch": 9, "memory": 6319, "step": 34256} +{"lr": 1.3195585791452309e-05, "data_time": 0.7089182615280152, "loss": 0.08662436828017235, "time": 0.8497709274291992, "epoch": 9, "memory": 6319, "step": 34356} +{"lr": 1.3063016483141729e-05, "data_time": 0.6240370035171509, "loss": 0.08690226525068283, "time": 0.8963229179382324, "epoch": 9, "memory": 6319, "step": 34456} +{"lr": 1.2933158663362388e-05, "data_time": 0.7880263805389405, "loss": 0.07403073199093342, "time": 1.017116951942444, "epoch": 9, "memory": 6319, "step": 34556} +{"lr": 1.2806020728235311e-05, "data_time": 0.6992140769958496, "loss": 0.06560872867703438, "time": 0.8916797637939453, "epoch": 9, "memory": 6319, "step": 34656} +{"lr": 1.2681610898023963e-05, "data_time": 0.5774627685546875, "loss": 0.07340772338211536, "time": 0.9182001352310181, "epoch": 9, "memory": 6319, "step": 34756} +{"lr": 1.2559937216602731e-05, "data_time": 0.7716176509857178, "loss": 0.06270115822553635, "time": 0.9139045953750611, "epoch": 9, "memory": 6319, "step": 34856} +{"lr": 1.2441007550936847e-05, "data_time": 0.7892608642578125, "loss": 0.08152935728430748, "time": 0.9311720371246338, "epoch": 9, "memory": 6319, "step": 34956} +{"lr": 1.232482959057377e-05, "data_time": 0.725196647644043, "loss": 0.07424784190952778, "time": 0.8679368495941162, "epoch": 9, "memory": 6319, "step": 35056} +{"lr": 1.2211410847145969e-05, "data_time": 0.6607789278030396, "loss": 0.08571186661720276, "time": 0.8571977138519287, "epoch": 9, "memory": 6319, "step": 35156} +{"accuracy/top1": 96.78160095214844, "data_time": 0.5831122521696419, "time": 0.6737202118183004, "step": 9} +{"lr": 1.209311679360931e-05, "data_time": 0.8175110578536987, "loss": 0.08883111737668514, "time": 0.9578233480453491, "epoch": 10, "memory": 6319, "step": 35263} +{"lr": 1.1985432727488666e-05, "data_time": 0.7931027412414551, "loss": 0.07517887912690639, "time": 0.9331846237182617, "epoch": 10, "memory": 6319, "step": 35363} +{"lr": 1.1880529822435289e-05, "data_time": 0.7581690788269043, "loss": 0.0755938570946455, "time": 0.8997335195541382, "epoch": 10, "memory": 6319, "step": 35463} +{"lr": 1.1778414861078726e-05, "data_time": 0.7512569665908814, "loss": 0.0721186900511384, "time": 0.8914328336715698, "epoch": 10, "memory": 6319, "step": 35563} +{"lr": 1.1679094445790538e-05, "data_time": 0.67048020362854, "loss": 0.08164873197674752, "time": 0.8105042934417724, "epoch": 10, "memory": 6319, "step": 35663} +{"lr": 1.1582574998257377e-05, "data_time": 0.7606892585754395, "loss": 0.07552231177687645, "time": 0.9008347272872925, "epoch": 10, "memory": 6319, "step": 35763} +{"lr": 1.1488862759065762e-05, "data_time": 0.7632877588272095, "loss": 0.07405286356806755, "time": 0.9042915582656861, "epoch": 10, "memory": 6319, "step": 35863} +{"lr": 1.1397963787298665e-05, "data_time": 0.7482366323471069, "loss": 0.07233749367296696, "time": 0.8890695095062255, "epoch": 10, "memory": 6319, "step": 35963} +{"lr": 1.1309883960143664e-05, "data_time": 0.7255772352218628, "loss": 0.0837469134479761, "time": 0.8934251070022583, "epoch": 10, "memory": 6319, "step": 36063} +{"lr": 1.122462897251302e-05, "data_time": 0.7596636295318604, "loss": 0.07752110138535499, "time": 0.9017051696777344, "epoch": 10, "memory": 6319, "step": 36163} +{"lr": 1.114220433667542e-05, "data_time": 0.752418041229248, "loss": 0.07443158216774463, "time": 0.9325988292694092, "epoch": 10, "memory": 6319, "step": 36263} +{"lr": 1.10626153818996e-05, "data_time": 0.6474128723144531, "loss": 0.0840250864624977, "time": 0.9051640748977661, "epoch": 10, "memory": 6319, "step": 36363} +{"lr": 1.0985867254109761e-05, "data_time": 0.5261065483093261, "loss": 0.09787520878016949, "time": 0.849211311340332, "epoch": 10, "memory": 6319, "step": 36463} +{"lr": 1.0911964915552823e-05, "data_time": 0.2851930379867554, "loss": 0.09178892150521278, "time": 0.960207176208496, "epoch": 10, "memory": 6319, "step": 36563} +{"lr": 1.0840913144477643e-05, "data_time": 0.14056668281555176, "loss": 0.06464039199054242, "time": 0.8757411241531372, "epoch": 10, "memory": 6319, "step": 36663} +{"lr": 1.0772716534826064e-05, "data_time": 0.001236724853515625, "loss": 0.07767030596733093, "time": 0.8480842351913452, "epoch": 10, "memory": 6319, "step": 36763} +{"lr": 1.0707379495935828e-05, "data_time": 0.0012414216995239257, "loss": 0.07756403498351574, "time": 0.9160326242446899, "epoch": 10, "memory": 6319, "step": 36863} +{"lr": 1.0644906252255541e-05, "data_time": 0.002115345001220703, "loss": 0.06526481397449971, "time": 0.850139856338501, "epoch": 10, "memory": 6319, "step": 36963} +{"lr": 1.0585300843071527e-05, "data_time": 0.0017118215560913085, "loss": 0.07799581438302994, "time": 0.8961062908172608, "epoch": 10, "memory": 6319, "step": 37063} +{"lr": 1.0528567122246654e-05, "data_time": 0.12719011306762695, "loss": 0.0727805256843567, "time": 0.8929010629653931, "epoch": 10, "memory": 6319, "step": 37163} +{"lr": 1.0474708757971141e-05, "data_time": 0.007215547561645508, "loss": 0.0660861760377884, "time": 0.9187443494796753, "epoch": 10, "memory": 6319, "step": 37263} +{"lr": 1.042372923252543e-05, "data_time": 0.001841592788696289, "loss": 0.08520641326904296, "time": 0.9100911855697632, "epoch": 10, "memory": 6319, "step": 37363} +{"lr": 1.0375631842054998e-05, "data_time": 0.13324143886566162, "loss": 0.0835212867707014, "time": 0.859965467453003, "epoch": 10, "memory": 6319, "step": 37463} +{"lr": 1.0330419696357254e-05, "data_time": 0.0011774063110351562, "loss": 0.07549788057804108, "time": 0.9032527923583984, "epoch": 10, "memory": 6319, "step": 37563} +{"lr": 1.028809571868047e-05, "data_time": 0.001439356803894043, "loss": 0.08595659658312797, "time": 0.9312840700149536, "epoch": 10, "memory": 6319, "step": 37663} +{"lr": 1.0248662645534766e-05, "data_time": 0.0009754419326782227, "loss": 0.0685815777629614, "time": 0.9517098188400268, "epoch": 10, "memory": 6319, "step": 37763} +{"lr": 1.0212123026515194e-05, "data_time": 0.0015410661697387695, "loss": 0.07711313255131244, "time": 0.9567354917526245, "epoch": 10, "memory": 6319, "step": 37863} +{"lr": 1.0178479224136881e-05, "data_time": 0.002238559722900391, "loss": 0.07415566071867943, "time": 0.8813257455825806, "epoch": 10, "memory": 6319, "step": 37963} +{"lr": 1.0147733413682289e-05, "data_time": 0.001111006736755371, "loss": 0.07807697914540768, "time": 0.8484270811080933, "epoch": 10, "memory": 6319, "step": 38063} +{"lr": 1.0119887583060527e-05, "data_time": 0.0015458345413208007, "loss": 0.07257316969335079, "time": 0.8970549821853637, "epoch": 10, "memory": 6319, "step": 38163} +{"lr": 1.0094943532678871e-05, "data_time": 0.0012021780014038086, "loss": 0.08120028041303158, "time": 0.9232643127441407, "epoch": 10, "memory": 6319, "step": 38263} +{"lr": 1.007290287532637e-05, "data_time": 0.0009762763977050782, "loss": 0.08372033014893532, "time": 0.8691381931304931, "epoch": 10, "memory": 6319, "step": 38363} +{"lr": 1.0053767036069495e-05, "data_time": 0.005805325508117676, "loss": 0.08308596312999725, "time": 0.9090308427810669, "epoch": 10, "memory": 6319, "step": 38463} +{"lr": 1.003753725216005e-05, "data_time": 0.16092276573181152, "loss": 0.07852649614214897, "time": 0.888532018661499, "epoch": 10, "memory": 6319, "step": 38563} +{"lr": 1.0024214572955152e-05, "data_time": 0.08160817623138428, "loss": 0.07156525179743767, "time": 0.9100588083267211, "epoch": 10, "memory": 6319, "step": 38663} +{"lr": 1.0013799859849456e-05, "data_time": 0.0012605905532836914, "loss": 0.07482991330325603, "time": 0.836520004272461, "epoch": 10, "memory": 6319, "step": 38763} +{"lr": 1.0006293786219338e-05, "data_time": 0.0010182619094848632, "loss": 0.06851191893219948, "time": 0.7639588356018067, "epoch": 10, "memory": 6319, "step": 38863} +{"lr": 1.0001696837379472e-05, "data_time": 0.685526442527771, "loss": 0.08617636188864708, "time": 0.852595043182373, "epoch": 10, "memory": 6319, "step": 38963} +{"lr": 1.0000009310551361e-05, "data_time": 0.3220003366470337, "loss": 0.07605950310826301, "time": 0.8021457433700562, "epoch": 10, "memory": 6319, "step": 39063} +{"accuracy/top1": 96.87554931640625, "data_time": 0.5467592231158552, "time": 0.6363690187191141, "step": 10} diff --git a/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/20230605_005004/vis_data/config.py b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/20230605_005004/vis_data/config.py new file mode 100644 index 0000000000000000000000000000000000000000..661bf63adae66645caa3af55d05b1778159f197c --- /dev/null +++ b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/20230605_005004/vis_data/config.py @@ -0,0 +1,230 @@ +optim_wrapper = dict( + optimizer=dict( + type='SGD', + lr=0.0001, + momentum=0.9, + weight_decay=0.0001, + _scope_='mmpretrain'), + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=False, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=512) +model = dict( + type='ImageClassifier', + backbone=dict( + frozen_stages=2, + type='ResNet', + depth=50, + num_stages=4, + out_indices=(3, ), + style='pytorch', + init_cfg=dict( + type='Pretrained', + checkpoint= + 'https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth', + prefix='backbone')), + neck=dict(type='GlobalAveragePooling'), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=2048, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + topk=1)) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stablediffusionV1-5R2-dpmsolver-25-1m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/cc1m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = dict(type='Accuracy', topk=1) +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = dict(type='Accuracy', topk=1) +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1' diff --git a/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/20230605_005004/vis_data/events.out.tfevents.1685897421.SH-IDC1-10-140-24-100.35638.0 b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/20230605_005004/vis_data/events.out.tfevents.1685897421.SH-IDC1-10-140-24-100.35638.0 new file mode 100644 index 0000000000000000000000000000000000000000..daab48caca1493d8538c1cc081b8b8fa67609c69 --- /dev/null +++ b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/20230605_005004/vis_data/events.out.tfevents.1685897421.SH-IDC1-10-140-24-100.35638.0 @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:a668a53a21f1261fd4975e82483e9f14313b323f6cf2029aba11c35cee3f897f +size 2239802 diff --git a/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/20230605_005004/vis_data/scalars.json b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/20230605_005004/vis_data/scalars.json new file mode 100644 index 0000000000000000000000000000000000000000..776df04618cd13133ff5b7f986084c52a340a3ff --- /dev/null +++ b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/20230605_005004/vis_data/scalars.json @@ -0,0 +1,400 @@ +{"lr": 9.999857418382631e-05, "data_time": 0.9103787899017334, "loss": 0.6140454709529877, "time": 1.3434776306152343, "epoch": 1, "memory": 9436, "step": 100} +{"lr": 9.999423907359506e-05, "data_time": 0.5140883684158325, "loss": 0.5577386796474457, "time": 0.8296503305435181, "epoch": 1, "memory": 6319, "step": 200} +{"lr": 9.998699480412073e-05, "data_time": 0.3724829912185669, "loss": 0.505340787768364, "time": 0.884411072731018, "epoch": 1, "memory": 6319, "step": 300} +{"lr": 9.997684184379085e-05, "data_time": 0.2931915521621704, "loss": 0.45813901722431183, "time": 0.9431975364685059, "epoch": 1, "memory": 6319, "step": 400} +{"lr": 9.996378084905784e-05, "data_time": 0.4679435968399048, "loss": 0.44143602550029754, "time": 0.8859490156173706, "epoch": 1, "memory": 6319, "step": 500} +{"lr": 9.994781266439667e-05, "data_time": 0.6340442419052124, "loss": 0.4044747710227966, "time": 0.8933575868606567, "epoch": 1, "memory": 6319, "step": 600} +{"lr": 9.992893832225061e-05, "data_time": 0.47628321647644045, "loss": 0.3901466906070709, "time": 0.9069812536239624, "epoch": 1, "memory": 6319, "step": 700} +{"lr": 9.990715904296405e-05, "data_time": 0.581229305267334, "loss": 0.3575277209281921, "time": 0.8604262828826904, "epoch": 1, "memory": 6319, "step": 800} +{"lr": 9.988247623470371e-05, "data_time": 0.7636813163757324, "loss": 0.3410681664943695, "time": 0.9043498992919922, "epoch": 1, "memory": 6319, "step": 900} +{"lr": 9.985489149336749e-05, "data_time": 0.734875226020813, "loss": 0.3338106542825699, "time": 0.9026185274124146, "epoch": 1, "memory": 6319, "step": 1000} +{"lr": 9.982440660248187e-05, "data_time": 0.8113263130187989, "loss": 0.318480509519577, "time": 0.9517826318740845, "epoch": 1, "memory": 6319, "step": 1100} +{"lr": 9.979102353308574e-05, "data_time": 0.8123575448989868, "loss": 0.3289756804704666, "time": 0.954686951637268, "epoch": 1, "memory": 6319, "step": 1200} +{"lr": 9.975474444360358e-05, "data_time": 0.8950042724609375, "loss": 0.3130018025636673, "time": 1.036578369140625, "epoch": 1, "memory": 6319, "step": 1300} +{"lr": 9.97155716797056e-05, "data_time": 0.7601632118225098, "loss": 0.2924617141485214, "time": 0.9109964609146118, "epoch": 1, "memory": 6319, "step": 1400} +{"lr": 9.967350777415632e-05, "data_time": 0.7745493650436401, "loss": 0.29582877159118653, "time": 0.9152874946594238, "epoch": 1, "memory": 6319, "step": 1500} +{"lr": 9.962855544665059e-05, "data_time": 0.7992411613464355, "loss": 0.2959482431411743, "time": 0.9496608257293702, "epoch": 1, "memory": 6319, "step": 1600} +{"lr": 9.958071760363773e-05, "data_time": 0.7655555248260498, "loss": 0.26180612593889235, "time": 0.9058728694915772, "epoch": 1, "memory": 6319, "step": 1700} +{"lr": 9.952999733813382e-05, "data_time": 0.7733849287033081, "loss": 0.27730225026607513, "time": 0.9151493787765503, "epoch": 1, "memory": 6319, "step": 1800} +{"lr": 9.947639792952148e-05, "data_time": 0.9313913822174072, "loss": 0.25891817510128023, "time": 1.0718937873840333, "epoch": 1, "memory": 6319, "step": 1900} +{"lr": 9.941992284333813e-05, "data_time": 0.8002794265747071, "loss": 0.23325824439525605, "time": 0.9503692388534546, "epoch": 1, "memory": 6319, "step": 2000} +{"lr": 9.93605757310516e-05, "data_time": 0.7893903017044067, "loss": 0.2547219946980476, "time": 0.9295383214950561, "epoch": 1, "memory": 6319, "step": 2100} +{"lr": 9.929836042982421e-05, "data_time": 0.73553946018219, "loss": 0.23771324306726455, "time": 0.9649042844772339, "epoch": 1, "memory": 6319, "step": 2200} +{"lr": 9.923328096226467e-05, "data_time": 0.8029179096221923, "loss": 0.23873921036720275, "time": 0.9442267894744873, "epoch": 1, "memory": 6319, "step": 2300} +{"lr": 9.916534153616819e-05, "data_time": 0.7717267990112304, "loss": 0.22541094273328782, "time": 0.9127573490142822, "epoch": 1, "memory": 6319, "step": 2400} +{"lr": 9.909454654424387e-05, "data_time": 0.7756819725036621, "loss": 0.22962120026350022, "time": 0.9164350986480713, "epoch": 1, "memory": 6319, "step": 2500} +{"lr": 9.902090056383109e-05, "data_time": 0.7729017257690429, "loss": 0.2304988294839859, "time": 0.9201807975769043, "epoch": 1, "memory": 6319, "step": 2600} +{"lr": 9.894440835660365e-05, "data_time": 0.7993639230728149, "loss": 0.21435047537088395, "time": 0.9397423028945923, "epoch": 1, "memory": 6319, "step": 2700} +{"lr": 9.886507486826162e-05, "data_time": 0.7800388097763061, "loss": 0.21141017377376556, "time": 0.9213169574737549, "epoch": 1, "memory": 6319, "step": 2800} +{"lr": 9.878290522821154e-05, "data_time": 0.9438929796218872, "loss": 0.21645904332399368, "time": 1.0847433805465698, "epoch": 1, "memory": 6319, "step": 2900} +{"lr": 9.86979047492352e-05, "data_time": 0.8303739547729492, "loss": 0.22295447587966918, "time": 0.9725280284881592, "epoch": 1, "memory": 6319, "step": 3000} +{"lr": 9.861007892714578e-05, "data_time": 0.7491269111633301, "loss": 0.19436087012290953, "time": 0.8884036779403687, "epoch": 1, "memory": 6319, "step": 3100} +{"lr": 9.851943344043239e-05, "data_time": 0.7981921672821045, "loss": 0.1901811569929123, "time": 0.9379892587661743, "epoch": 1, "memory": 6319, "step": 3200} +{"lr": 9.842597414989319e-05, "data_time": 0.7791166067123413, "loss": 0.19938674718141555, "time": 0.9201548576354981, "epoch": 1, "memory": 6319, "step": 3300} +{"lr": 9.832970709825637e-05, "data_time": 0.773836088180542, "loss": 0.20726506412029266, "time": 0.9154900312423706, "epoch": 1, "memory": 6319, "step": 3400} +{"lr": 9.823063850978968e-05, "data_time": 0.8334040880203247, "loss": 0.20610593855381013, "time": 0.9741550207138061, "epoch": 1, "memory": 6319, "step": 3500} +{"lr": 9.81287747898975e-05, "data_time": 0.7800627708435058, "loss": 0.19718588441610335, "time": 0.9192874908447266, "epoch": 1, "memory": 6319, "step": 3600} +{"lr": 9.802412252470715e-05, "data_time": 0.9732753992080688, "loss": 0.1905989408493042, "time": 1.1127555608749389, "epoch": 1, "memory": 6319, "step": 3700} +{"lr": 9.791668848064258e-05, "data_time": 0.7266729831695556, "loss": 0.1876543715596199, "time": 0.9095073938369751, "epoch": 1, "memory": 6319, "step": 3800} +{"lr": 9.780647960398751e-05, "data_time": 0.6990907192230225, "loss": 0.20302069634199144, "time": 0.9295929431915283, "epoch": 1, "memory": 6319, "step": 3900} +{"accuracy/top1": 72.29393768310547, "data_time": 0.6343517387122438, "time": 0.7288378288871363, "step": 1} +{"lr": 9.7685491196646e-05, "data_time": 0.6941997289657593, "loss": 0.1862989366054535, "time": 0.9376872777938843, "epoch": 2, "memory": 6319, "step": 4007} +{"lr": 9.756956126085676e-05, "data_time": 0.4809847116470337, "loss": 0.1829976722598076, "time": 0.9123958349227905, "epoch": 2, "memory": 6319, "step": 4107} +{"lr": 9.745087893642482e-05, "data_time": 0.130255126953125, "loss": 0.1872863069176674, "time": 0.954208779335022, "epoch": 2, "memory": 6319, "step": 4207} +{"lr": 9.73294518969056e-05, "data_time": 0.39117271900177003, "loss": 0.19641478657722472, "time": 0.8785030126571656, "epoch": 2, "memory": 6319, "step": 4307} +{"lr": 9.72052879933172e-05, "data_time": 0.4963539123535156, "loss": 0.16399107426404952, "time": 1.0425027370452882, "epoch": 2, "memory": 6319, "step": 4407} +{"lr": 9.707839525363358e-05, "data_time": 0.0008583784103393555, "loss": 0.18590614348649978, "time": 0.8847098588943482, "epoch": 2, "memory": 6319, "step": 4507} +{"lr": 9.69487818822646e-05, "data_time": 0.0008333444595336914, "loss": 0.17653323858976364, "time": 0.8779830694198608, "epoch": 2, "memory": 6319, "step": 4607} +{"lr": 9.681645625952616e-05, "data_time": 0.0010425567626953125, "loss": 0.14852667599916458, "time": 0.9276500225067139, "epoch": 2, "memory": 6319, "step": 4707} +{"lr": 9.668142694109826e-05, "data_time": 0.001341700553894043, "loss": 0.16353578865528107, "time": 0.9405390739440918, "epoch": 2, "memory": 6319, "step": 4807} +{"lr": 9.654370265747162e-05, "data_time": 0.0012044191360473632, "loss": 0.17817180901765822, "time": 0.901382851600647, "epoch": 2, "memory": 6319, "step": 4907} +{"lr": 9.640329231338345e-05, "data_time": 0.0008122682571411133, "loss": 0.16629842966794967, "time": 0.9265341281890869, "epoch": 2, "memory": 6319, "step": 5007} +{"lr": 9.626020498724147e-05, "data_time": 0.11947834491729736, "loss": 0.16525886654853822, "time": 0.8963955163955688, "epoch": 2, "memory": 6319, "step": 5107} +{"lr": 9.611444993053717e-05, "data_time": 0.10180156230926514, "loss": 0.16267255395650865, "time": 0.9097028255462647, "epoch": 2, "memory": 6319, "step": 5207} +{"lr": 9.596603656724746e-05, "data_time": 0.0008769512176513671, "loss": 0.17286299914121628, "time": 0.9312175512313843, "epoch": 2, "memory": 6319, "step": 5307} +{"lr": 9.58149744932254e-05, "data_time": 0.0009878158569335937, "loss": 0.16170311644673346, "time": 0.8902554273605346, "epoch": 2, "memory": 6319, "step": 5407} +{"lr": 9.566127347557972e-05, "data_time": 0.0008677005767822266, "loss": 0.16101982742547988, "time": 0.8930779933929444, "epoch": 2, "memory": 6319, "step": 5507} +{"lr": 9.550494345204367e-05, "data_time": 0.0009014368057250977, "loss": 0.15405538603663443, "time": 1.1866081476211547, "epoch": 2, "memory": 6319, "step": 5607} +{"lr": 9.534599453033183e-05, "data_time": 0.000959014892578125, "loss": 0.14588873237371444, "time": 0.8901443719863892, "epoch": 2, "memory": 6319, "step": 5707} +{"lr": 9.518443698748712e-05, "data_time": 0.0010004997253417968, "loss": 0.16322704702615737, "time": 0.905848503112793, "epoch": 2, "memory": 6319, "step": 5807} +{"lr": 9.50202812692162e-05, "data_time": 0.001230001449584961, "loss": 0.15641508996486664, "time": 0.9147838354110718, "epoch": 2, "memory": 6319, "step": 5907} +{"lr": 9.485353798921386e-05, "data_time": 0.0010570526123046876, "loss": 0.14724608659744262, "time": 0.951305603981018, "epoch": 2, "memory": 6319, "step": 6007} +{"lr": 9.468421792847712e-05, "data_time": 0.0010634660720825195, "loss": 0.1521591491997242, "time": 0.9603242874145508, "epoch": 2, "memory": 6319, "step": 6107} +{"lr": 9.451233203460794e-05, "data_time": 0.0015326261520385743, "loss": 0.1508932925760746, "time": 0.8811447620391846, "epoch": 2, "memory": 6319, "step": 6207} +{"lr": 9.433789142110532e-05, "data_time": 0.001342010498046875, "loss": 0.1372598946094513, "time": 0.9293378829956055, "epoch": 2, "memory": 6319, "step": 6307} +{"lr": 9.416090736664709e-05, "data_time": 0.0029036998748779297, "loss": 0.1580960802733898, "time": 0.9070582628250122, "epoch": 2, "memory": 6319, "step": 6407} +{"lr": 9.398139131436032e-05, "data_time": 0.0010502338409423828, "loss": 0.1617794707417488, "time": 0.9530377388000488, "epoch": 2, "memory": 6319, "step": 6507} +{"lr": 9.379935487108167e-05, "data_time": 0.0010036706924438476, "loss": 0.15584279000759124, "time": 0.9389506578445435, "epoch": 2, "memory": 6319, "step": 6607} +{"lr": 9.361480980660697e-05, "data_time": 0.001021265983581543, "loss": 0.14667166024446487, "time": 0.8889129161834717, "epoch": 2, "memory": 6319, "step": 6707} +{"lr": 9.342776805292993e-05, "data_time": 0.3138540506362915, "loss": 0.13809967413544655, "time": 1.387740921974182, "epoch": 2, "memory": 6319, "step": 6807} +{"lr": 9.323824170347083e-05, "data_time": 0.0008918046951293945, "loss": 0.14395274966955185, "time": 0.8233531951904297, "epoch": 2, "memory": 6319, "step": 6907} +{"lr": 9.304624301229471e-05, "data_time": 0.0010257720947265624, "loss": 0.1390126757323742, "time": 0.9329187154769898, "epoch": 2, "memory": 6319, "step": 7007} +{"lr": 9.285178439331897e-05, "data_time": 0.000758957862854004, "loss": 0.1575406163930893, "time": 0.9016340970993042, "epoch": 2, "memory": 6319, "step": 7107} +{"lr": 9.265487841951087e-05, "data_time": 0.0011159658432006836, "loss": 0.16433390974998474, "time": 0.8750843286514283, "epoch": 2, "memory": 6319, "step": 7207} +{"lr": 9.245553782207421e-05, "data_time": 0.0009636163711547851, "loss": 0.14237954020500182, "time": 0.8701907634735108, "epoch": 2, "memory": 6319, "step": 7307} +{"lr": 9.225377548962671e-05, "data_time": 0.029822874069213866, "loss": 0.1589060455560684, "time": 0.929782772064209, "epoch": 2, "memory": 6319, "step": 7407} +{"lr": 9.204960446736601e-05, "data_time": 0.0010097503662109375, "loss": 0.13618902415037154, "time": 0.9080040931701661, "epoch": 2, "memory": 6319, "step": 7507} +{"lr": 9.184303795622715e-05, "data_time": 0.0011968612670898438, "loss": 0.12923629358410835, "time": 0.890028429031372, "epoch": 2, "memory": 6319, "step": 7607} +{"lr": 9.163408931202823e-05, "data_time": 0.001125335693359375, "loss": 0.1407058507204056, "time": 0.933996033668518, "epoch": 2, "memory": 6319, "step": 7707} +{"lr": 9.142277204460729e-05, "data_time": 0.1532210350036621, "loss": 0.13286348655819893, "time": 0.8853833913803101, "epoch": 2, "memory": 6319, "step": 7807} +{"accuracy/top1": 87.66570281982422, "data_time": 0.5515617872106617, "time": 0.6425289491127277, "step": 2} +{"lr": 9.119405492383288e-05, "data_time": 0.6440482378005982, "loss": 0.1410001888871193, "time": 0.9266921520233155, "epoch": 3, "memory": 6319, "step": 7914} +{"lr": 9.097787819233205e-05, "data_time": 0.7047796249389648, "loss": 0.14838272482156753, "time": 0.9241950035095214, "epoch": 3, "memory": 6319, "step": 8014} +{"lr": 9.075937526576627e-05, "data_time": 0.5703942060470581, "loss": 0.1360948972404003, "time": 0.8951602697372436, "epoch": 3, "memory": 6319, "step": 8114} +{"lr": 9.053856027171755e-05, "data_time": 0.5875956535339355, "loss": 0.14739920645952226, "time": 0.9040512561798095, "epoch": 3, "memory": 6319, "step": 8214} +{"lr": 9.031544748725785e-05, "data_time": 0.4505643129348755, "loss": 0.1350906141102314, "time": 0.9061188220977783, "epoch": 3, "memory": 6319, "step": 8314} +{"lr": 9.009005133802542e-05, "data_time": 0.4379525661468506, "loss": 0.13327923268079758, "time": 0.8938762903213501, "epoch": 3, "memory": 6319, "step": 8414} +{"lr": 8.986238639729242e-05, "data_time": 0.4665829181671143, "loss": 0.11894499957561493, "time": 0.9614094972610474, "epoch": 3, "memory": 6319, "step": 8514} +{"lr": 8.963246738502254e-05, "data_time": 0.42489118576049806, "loss": 0.13079854100942612, "time": 0.890029001235962, "epoch": 3, "memory": 6319, "step": 8614} +{"lr": 8.940030916691928e-05, "data_time": 0.2928466320037842, "loss": 0.13693108186125755, "time": 0.9113356351852417, "epoch": 3, "memory": 6319, "step": 8714} +{"lr": 8.91659267534649e-05, "data_time": 0.3996603965759277, "loss": 0.13073570877313614, "time": 0.9123415231704712, "epoch": 3, "memory": 6319, "step": 8814} +{"lr": 8.892933529894978e-05, "data_time": 0.5623992681503296, "loss": 0.1372727259993553, "time": 0.8533718585968018, "epoch": 3, "memory": 6319, "step": 8914} +{"lr": 8.869055010049283e-05, "data_time": 0.791880989074707, "loss": 0.12498634904623032, "time": 0.9333408594131469, "epoch": 3, "memory": 6319, "step": 9014} +{"lr": 8.844958659705188e-05, "data_time": 0.6925709486007691, "loss": 0.11703649088740349, "time": 0.9083391189575195, "epoch": 3, "memory": 6319, "step": 9114} +{"lr": 8.82064603684259e-05, "data_time": 0.6411150932312012, "loss": 0.11198518797755241, "time": 0.9106740236282349, "epoch": 3, "memory": 6319, "step": 9214} +{"lr": 8.796118713424776e-05, "data_time": 0.5770849227905274, "loss": 0.13326073661446572, "time": 0.9313672304153442, "epoch": 3, "memory": 6319, "step": 9314} +{"lr": 8.77137827529676e-05, "data_time": 0.5237188339233398, "loss": 0.1459883339703083, "time": 0.9073834180831909, "epoch": 3, "memory": 6319, "step": 9414} +{"lr": 8.746426322082745e-05, "data_time": 0.4845975399017334, "loss": 0.12167832925915718, "time": 0.8851630210876464, "epoch": 3, "memory": 6319, "step": 9514} +{"lr": 8.721264467082732e-05, "data_time": 0.22923238277435304, "loss": 0.10880739465355874, "time": 0.9161917448043824, "epoch": 3, "memory": 6319, "step": 9614} +{"lr": 8.695894337168168e-05, "data_time": 0.34072372913360593, "loss": 0.1253657318651676, "time": 0.9655329465866089, "epoch": 3, "memory": 6319, "step": 9714} +{"lr": 8.67031757267677e-05, "data_time": 0.41166934967041013, "loss": 0.13195557817816733, "time": 0.8773979425430298, "epoch": 3, "memory": 6319, "step": 9814} +{"lr": 8.64453582730651e-05, "data_time": 0.4863105535507202, "loss": 0.12704012244939805, "time": 1.162819743156433, "epoch": 3, "memory": 6319, "step": 9914} +{"lr": 8.61855076800861e-05, "data_time": 0.3985645294189453, "loss": 0.12885515242815018, "time": 0.9102446794509887, "epoch": 3, "memory": 6319, "step": 10014} +{"lr": 8.592364074879843e-05, "data_time": 0.775734543800354, "loss": 0.12236005663871766, "time": 1.061826515197754, "epoch": 3, "memory": 6319, "step": 10114} +{"lr": 8.565977441053854e-05, "data_time": 1.1374670028686524, "loss": 0.11657613888382912, "time": 1.2810584545135497, "epoch": 3, "memory": 6319, "step": 10214} +{"lr": 8.539392572591724e-05, "data_time": 0.7939578771591187, "loss": 0.1231521837413311, "time": 0.9340907573699951, "epoch": 3, "memory": 6319, "step": 10314} +{"lr": 8.512611188371633e-05, "data_time": 0.7361741065979004, "loss": 0.10086353421211243, "time": 0.8765170097351074, "epoch": 3, "memory": 6319, "step": 10414} +{"lr": 8.485635019977725e-05, "data_time": 0.7522613286972046, "loss": 0.13376380875706673, "time": 0.893633770942688, "epoch": 3, "memory": 6319, "step": 10514} +{"lr": 8.45846581158818e-05, "data_time": 0.7583223342895508, "loss": 0.1197498731315136, "time": 0.899726390838623, "epoch": 3, "memory": 6319, "step": 10614} +{"lr": 8.431105319862416e-05, "data_time": 0.7312403678894043, "loss": 0.11213548332452775, "time": 0.873240303993225, "epoch": 3, "memory": 6319, "step": 10714} +{"lr": 8.403555313827516e-05, "data_time": 0.826082444190979, "loss": 0.1139491319656372, "time": 0.9674846649169921, "epoch": 3, "memory": 6319, "step": 10814} +{"lr": 8.375817574763844e-05, "data_time": 0.7036544561386109, "loss": 0.12039667442440986, "time": 0.8431573867797851, "epoch": 3, "memory": 6319, "step": 10914} +{"lr": 8.347893896089871e-05, "data_time": 0.7328249931335449, "loss": 0.12288720905780792, "time": 0.8740258932113647, "epoch": 3, "memory": 6319, "step": 11014} +{"lr": 8.31978608324625e-05, "data_time": 0.8448924541473388, "loss": 0.11540027558803559, "time": 0.9855837106704712, "epoch": 3, "memory": 6319, "step": 11114} +{"lr": 8.291495953579052e-05, "data_time": 0.7429500102996827, "loss": 0.11476954445242882, "time": 0.8967215538024902, "epoch": 3, "memory": 6319, "step": 11214} +{"lr": 8.263025336222267e-05, "data_time": 0.7222877740859985, "loss": 0.11551371961832047, "time": 0.8638005971908569, "epoch": 3, "memory": 6319, "step": 11314} +{"lr": 8.234376071979551e-05, "data_time": 0.8286003828048706, "loss": 0.12651703283190727, "time": 0.9698094606399537, "epoch": 3, "memory": 6319, "step": 11414} +{"lr": 8.205550013205187e-05, "data_time": 0.7761763572692871, "loss": 0.11782191470265388, "time": 0.9162150859832764, "epoch": 3, "memory": 6319, "step": 11514} +{"lr": 8.17654902368435e-05, "data_time": 0.7461871862411499, "loss": 0.1229425273835659, "time": 0.8858694314956665, "epoch": 3, "memory": 6319, "step": 11614} +{"lr": 8.147374978512582e-05, "data_time": 0.739277458190918, "loss": 0.12535810321569443, "time": 0.8783728122711182, "epoch": 3, "memory": 6319, "step": 11714} +{"accuracy/top1": 91.27030944824219, "data_time": 0.5697703238191276, "time": 0.6578957047955744, "step": 3} +{"lr": 8.115969237626578e-05, "data_time": 0.776940393447876, "loss": 0.11489329561591148, "time": 0.9180193424224854, "epoch": 4, "memory": 6319, "step": 11821} +{"lr": 8.08644297337245e-05, "data_time": 0.758247685432434, "loss": 0.10636537447571755, "time": 0.8997403860092164, "epoch": 4, "memory": 6319, "step": 11921} +{"lr": 8.056749479387651e-05, "data_time": 0.7399152278900146, "loss": 0.12493508234620095, "time": 0.8834031820297241, "epoch": 4, "memory": 6319, "step": 12021} +{"lr": 8.026890675542457e-05, "data_time": 0.7429991960525513, "loss": 0.11613981425762177, "time": 0.884716796875, "epoch": 4, "memory": 6319, "step": 12121} +{"lr": 7.996868492395457e-05, "data_time": 0.7280483961105346, "loss": 0.11278834193944931, "time": 0.8697750568389893, "epoch": 4, "memory": 6319, "step": 12221} +{"lr": 7.966684871068762e-05, "data_time": 0.7241075992584228, "loss": 0.11355518251657486, "time": 0.8657039165496826, "epoch": 4, "memory": 6319, "step": 12321} +{"lr": 7.936341763122449e-05, "data_time": 0.784734058380127, "loss": 0.10069649256765842, "time": 0.925127363204956, "epoch": 4, "memory": 6319, "step": 12421} +{"lr": 7.905841130428405e-05, "data_time": 0.7919846534729004, "loss": 0.11319094225764274, "time": 0.9325732946395874, "epoch": 4, "memory": 6319, "step": 12521} +{"lr": 7.875184945043487e-05, "data_time": 0.7001120090484619, "loss": 0.1142881266772747, "time": 0.8399037599563599, "epoch": 4, "memory": 6319, "step": 12621} +{"lr": 7.844375189082012e-05, "data_time": 0.7048356771469116, "loss": 0.10352867394685746, "time": 0.8449249744415284, "epoch": 4, "memory": 6319, "step": 12721} +{"lr": 7.813413854587572e-05, "data_time": 0.7509239912033081, "loss": 0.10277210175991058, "time": 0.8908655405044555, "epoch": 4, "memory": 6319, "step": 12821} +{"lr": 7.782302943404291e-05, "data_time": 0.7470978260040283, "loss": 0.11459473595023155, "time": 0.8870609760284424, "epoch": 4, "memory": 6319, "step": 12921} +{"lr": 7.751044467047337e-05, "data_time": 0.7490935325622559, "loss": 0.12576298043131828, "time": 0.908761715888977, "epoch": 4, "memory": 6319, "step": 13021} +{"lr": 7.719640446572907e-05, "data_time": 0.7791104793548584, "loss": 0.10602714270353317, "time": 0.9199146747589111, "epoch": 4, "memory": 6319, "step": 13121} +{"lr": 7.688092912447536e-05, "data_time": 0.7819726705551148, "loss": 0.09999264776706696, "time": 0.9230399847030639, "epoch": 4, "memory": 6319, "step": 13221} +{"lr": 7.656403904416794e-05, "data_time": 0.7548886775970459, "loss": 0.09779963791370391, "time": 0.8953191041946411, "epoch": 4, "memory": 6319, "step": 13321} +{"lr": 7.624575471373434e-05, "data_time": 0.7260256767272949, "loss": 0.10628859400749206, "time": 0.8663904190063476, "epoch": 4, "memory": 6319, "step": 13421} +{"lr": 7.592609671224926e-05, "data_time": 0.761096715927124, "loss": 0.11050747781991958, "time": 0.9012486934661865, "epoch": 4, "memory": 6319, "step": 13521} +{"lr": 7.560508570760375e-05, "data_time": 0.7586737394332885, "loss": 0.09491719752550125, "time": 0.8995528221130371, "epoch": 4, "memory": 6319, "step": 13621} +{"lr": 7.528274245516895e-05, "data_time": 0.7230385780334473, "loss": 0.09878298342227936, "time": 0.8623899221420288, "epoch": 4, "memory": 6319, "step": 13721} +{"lr": 7.495908779645405e-05, "data_time": 0.5954001665115356, "loss": 0.12024580836296081, "time": 0.8468973159790039, "epoch": 4, "memory": 6319, "step": 13821} +{"lr": 7.463414265775906e-05, "data_time": 0.591273832321167, "loss": 0.09703774675726891, "time": 0.9595482349395752, "epoch": 4, "memory": 6319, "step": 13921} +{"lr": 7.430792804882147e-05, "data_time": 0.7594960451126098, "loss": 0.11120865792036057, "time": 0.9348649978637695, "epoch": 4, "memory": 6319, "step": 14021} +{"lr": 7.398046506145803e-05, "data_time": 0.7183664321899415, "loss": 0.09288974180817604, "time": 0.8592118740081787, "epoch": 4, "memory": 6319, "step": 14121} +{"lr": 7.365177486820114e-05, "data_time": 0.7816154479980468, "loss": 0.09406102634966373, "time": 0.9227701425552368, "epoch": 4, "memory": 6319, "step": 14221} +{"lr": 7.332187872092939e-05, "data_time": 0.7644946575164795, "loss": 0.08868381567299366, "time": 0.9041457891464233, "epoch": 4, "memory": 6319, "step": 14321} +{"lr": 7.299079794949401e-05, "data_time": 0.779997730255127, "loss": 0.12051670849323273, "time": 0.9211164474487304, "epoch": 4, "memory": 6319, "step": 14421} +{"lr": 7.265855396033955e-05, "data_time": 0.7451975107192993, "loss": 0.10627755373716355, "time": 0.8853612899780273, "epoch": 4, "memory": 6319, "step": 14521} +{"lr": 7.232516823512013e-05, "data_time": 0.6976796150207519, "loss": 0.10052933394908906, "time": 0.8378207921981812, "epoch": 4, "memory": 6319, "step": 14621} +{"lr": 7.199066232930995e-05, "data_time": 0.7325936794281006, "loss": 0.09236596971750259, "time": 0.8737106323242188, "epoch": 4, "memory": 6319, "step": 14721} +{"lr": 7.165505787080997e-05, "data_time": 0.7636126756668091, "loss": 0.10255631729960442, "time": 0.9047358512878418, "epoch": 4, "memory": 6319, "step": 14821} +{"lr": 7.13183765585495e-05, "data_time": 0.7070759534835815, "loss": 0.09638598337769508, "time": 0.8480723619461059, "epoch": 4, "memory": 6319, "step": 14921} +{"lr": 7.098064016108317e-05, "data_time": 0.7198184728622437, "loss": 0.1066634014248848, "time": 0.8599213123321533, "epoch": 4, "memory": 6319, "step": 15021} +{"lr": 7.064187051518379e-05, "data_time": 0.7827658891677857, "loss": 0.09756147637963294, "time": 0.9222442626953125, "epoch": 4, "memory": 6319, "step": 15121} +{"lr": 7.030208952442949e-05, "data_time": 0.730298638343811, "loss": 0.10235938653349877, "time": 0.8765639781951904, "epoch": 4, "memory": 6319, "step": 15221} +{"lr": 6.996131915778851e-05, "data_time": 0.7166630506515503, "loss": 0.1080032080411911, "time": 0.8568323850631714, "epoch": 4, "memory": 6319, "step": 15321} +{"lr": 6.961958144819853e-05, "data_time": 0.5675545692443847, "loss": 0.09821044877171517, "time": 0.894742226600647, "epoch": 4, "memory": 6319, "step": 15421} +{"lr": 6.927689849114193e-05, "data_time": 0.5460972309112548, "loss": 0.10202130004763603, "time": 0.8817622900009155, "epoch": 4, "memory": 6319, "step": 15521} +{"lr": 6.893329244321722e-05, "data_time": 0.4924583911895752, "loss": 0.10610084086656571, "time": 0.8512593269348144, "epoch": 4, "memory": 6319, "step": 15621} +{"accuracy/top1": 93.66758728027344, "data_time": 0.5766637530820123, "time": 0.6660525757690956, "step": 4} +{"lr": 6.856463687361792e-05, "data_time": 0.7157703638076782, "loss": 0.10067639276385307, "time": 0.8576092004776001, "epoch": 5, "memory": 6319, "step": 15728} +{"lr": 6.821919068470432e-05, "data_time": 0.6827544450759888, "loss": 0.1158062532544136, "time": 0.8226525545120239, "epoch": 5, "memory": 6319, "step": 15828} +{"lr": 6.787288979235407e-05, "data_time": 0.6815675735473633, "loss": 0.08799593523144722, "time": 0.821623969078064, "epoch": 5, "memory": 6319, "step": 15928} +{"lr": 6.752575658708793e-05, "data_time": 0.5739075183868408, "loss": 0.09230509996414185, "time": 0.8700387477874756, "epoch": 5, "memory": 6319, "step": 16028} +{"lr": 6.71778135132408e-05, "data_time": 0.3925312519073486, "loss": 0.09604379832744599, "time": 0.9017087697982789, "epoch": 5, "memory": 6319, "step": 16128} +{"lr": 6.682908306751074e-05, "data_time": 0.5438744783401489, "loss": 0.0960699513554573, "time": 0.9112030029296875, "epoch": 5, "memory": 6319, "step": 16228} +{"lr": 6.64795877975043e-05, "data_time": 0.6819400072097779, "loss": 0.09235613346099854, "time": 0.9370819807052613, "epoch": 5, "memory": 6319, "step": 16328} +{"lr": 6.61293503002787e-05, "data_time": 0.4025144100189209, "loss": 0.1131122749298811, "time": 0.8457817077636719, "epoch": 5, "memory": 6319, "step": 16428} +{"lr": 6.577839322088084e-05, "data_time": 0.5672608852386475, "loss": 0.10250139832496644, "time": 0.8598496437072753, "epoch": 5, "memory": 6319, "step": 16528} +{"lr": 6.542673925088298e-05, "data_time": 0.359615159034729, "loss": 0.09744763374328613, "time": 0.8780105590820313, "epoch": 5, "memory": 6319, "step": 16628} +{"lr": 6.507441112691584e-05, "data_time": 0.24308438301086427, "loss": 0.08741888999938965, "time": 0.820630669593811, "epoch": 5, "memory": 6319, "step": 16728} +{"lr": 6.47214316291983e-05, "data_time": 0.21722354888916015, "loss": 0.09268408082425594, "time": 0.8541023254394531, "epoch": 5, "memory": 6319, "step": 16828} +{"lr": 6.436782358006481e-05, "data_time": 0.20202741622924805, "loss": 0.09366943910717965, "time": 0.8682555913925171, "epoch": 5, "memory": 6319, "step": 16928} +{"lr": 6.40136098424896e-05, "data_time": 0.2511094808578491, "loss": 0.09410080313682556, "time": 0.9106420516967774, "epoch": 5, "memory": 6319, "step": 17028} +{"lr": 6.365881331860828e-05, "data_time": 0.07914295196533203, "loss": 0.11396202370524407, "time": 0.9080969095230103, "epoch": 5, "memory": 6319, "step": 17128} +{"lr": 6.330345694823739e-05, "data_time": 0.00127866268157959, "loss": 0.09479823037981987, "time": 0.8430700778961182, "epoch": 5, "memory": 6319, "step": 17228} +{"lr": 6.29475637073908e-05, "data_time": 0.0009033679962158203, "loss": 0.08776037395000458, "time": 0.9023783206939697, "epoch": 5, "memory": 6319, "step": 17328} +{"lr": 6.259115660679482e-05, "data_time": 0.0014479637145996093, "loss": 0.08252049572765827, "time": 0.9442789077758789, "epoch": 5, "memory": 6319, "step": 17428} +{"lr": 6.223425869039963e-05, "data_time": 0.00141448974609375, "loss": 0.09708419069647789, "time": 0.7980317115783692, "epoch": 5, "memory": 6319, "step": 17528} +{"lr": 6.187689303389006e-05, "data_time": 0.001473402976989746, "loss": 0.09579034522175789, "time": 0.9105540990829468, "epoch": 5, "memory": 6319, "step": 17628} +{"lr": 6.151908274319292e-05, "data_time": 0.0012321710586547852, "loss": 0.08410843573510647, "time": 0.8482635021209717, "epoch": 5, "memory": 6319, "step": 17728} +{"lr": 6.116085095298373e-05, "data_time": 0.0011002063751220704, "loss": 0.11261978149414062, "time": 0.8731829166412354, "epoch": 5, "memory": 6319, "step": 17828} +{"lr": 6.080222082519007e-05, "data_time": 0.00096435546875, "loss": 0.09185517057776452, "time": 0.9497179985046387, "epoch": 5, "memory": 6319, "step": 17928} +{"lr": 6.0443215547495147e-05, "data_time": 0.0015970706939697266, "loss": 0.09255757033824921, "time": 0.8956833839416504, "epoch": 5, "memory": 6319, "step": 18028} +{"lr": 6.0083858331837714e-05, "data_time": 0.000991487503051758, "loss": 0.09854256957769394, "time": 0.8653445720672608, "epoch": 5, "memory": 6319, "step": 18128} +{"lr": 5.97241724129116e-05, "data_time": 0.0011955022811889649, "loss": 0.09508971720933915, "time": 0.9046830415725708, "epoch": 5, "memory": 6319, "step": 18228} +{"lr": 5.936418104666328e-05, "data_time": 0.0016076326370239257, "loss": 0.07920684032142163, "time": 0.8601744413375855, "epoch": 5, "memory": 6319, "step": 18328} +{"lr": 5.9003907508788295e-05, "data_time": 0.0023520708084106444, "loss": 0.1014905646443367, "time": 0.884795355796814, "epoch": 5, "memory": 6319, "step": 18428} +{"lr": 5.864337509322633e-05, "data_time": 0.001106405258178711, "loss": 0.09967180341482162, "time": 0.9008808851242065, "epoch": 5, "memory": 6319, "step": 18528} +{"lr": 5.8282607110655194e-05, "data_time": 0.0013938426971435546, "loss": 0.1027616836130619, "time": 0.9279417276382447, "epoch": 5, "memory": 6319, "step": 18628} +{"lr": 5.7921626886983564e-05, "data_time": 0.0018150091171264648, "loss": 0.08932366222143173, "time": 0.8468700170516967, "epoch": 5, "memory": 6319, "step": 18728} +{"lr": 5.756045776184287e-05, "data_time": 0.0013753652572631836, "loss": 0.11413226500153542, "time": 0.8648529291152954, "epoch": 5, "memory": 6319, "step": 18828} +{"lr": 5.719912308707808e-05, "data_time": 0.0014557361602783204, "loss": 0.09509230181574821, "time": 0.8548434019088745, "epoch": 5, "memory": 6319, "step": 18928} +{"lr": 5.683764622523808e-05, "data_time": 0.0008451461791992188, "loss": 0.10767360329627991, "time": 0.8595911979675293, "epoch": 5, "memory": 6319, "step": 19028} +{"lr": 5.647605054806502e-05, "data_time": 0.006430959701538086, "loss": 0.09613983780145645, "time": 0.8443287372589111, "epoch": 5, "memory": 6319, "step": 19128} +{"lr": 5.6114359434983206e-05, "data_time": 0.2763890027999878, "loss": 0.08913504853844642, "time": 0.8781718492507935, "epoch": 5, "memory": 6319, "step": 19228} +{"lr": 5.575259627158745e-05, "data_time": 0.10579869747161866, "loss": 0.08754120543599128, "time": 0.847752594947815, "epoch": 5, "memory": 6319, "step": 19328} +{"lr": 5.539078444813113e-05, "data_time": 0.33369197845458987, "loss": 0.1057322308421135, "time": 0.82184317111969, "epoch": 5, "memory": 6319, "step": 19428} +{"lr": 5.5028947358013735e-05, "data_time": 0.42878286838531493, "loss": 0.08728051409125329, "time": 0.8530410528182983, "epoch": 5, "memory": 6319, "step": 19528} +{"accuracy/top1": 94.79141235351562, "data_time": 0.5715527205631651, "time": 0.6595231459058565, "step": 5} +{"lr": 5.464178020340108e-05, "data_time": 0.6957848787307739, "loss": 0.08735099732875824, "time": 0.8367028951644897, "epoch": 6, "memory": 6319, "step": 19635} +{"lr": 5.427996514795046e-05, "data_time": 0.3696676015853882, "loss": 0.10282282382249833, "time": 0.8561572790145874, "epoch": 6, "memory": 6319, "step": 19735} +{"lr": 5.3918196647260834e-05, "data_time": 0.0011491060256958007, "loss": 0.09733559787273408, "time": 0.8947566270828247, "epoch": 6, "memory": 6319, "step": 19835} +{"lr": 5.355649809193057e-05, "data_time": 0.4860093116760254, "loss": 0.08619643189013004, "time": 0.9094852924346923, "epoch": 6, "memory": 6319, "step": 19935} +{"lr": 5.3194892868035765e-05, "data_time": 0.3416721343994141, "loss": 0.0918333400040865, "time": 0.8785401821136475, "epoch": 6, "memory": 6319, "step": 20035} +{"lr": 5.283340435561795e-05, "data_time": 0.31291096210479735, "loss": 0.09346374906599522, "time": 0.8457836389541626, "epoch": 6, "memory": 6319, "step": 20135} +{"lr": 5.2472055927172495e-05, "data_time": 0.2621103048324585, "loss": 0.09200282096862793, "time": 0.8982357501983642, "epoch": 6, "memory": 6319, "step": 20235} +{"lr": 5.2110870946137605e-05, "data_time": 0.20984821319580077, "loss": 0.0933717668056488, "time": 0.8591134548187256, "epoch": 6, "memory": 6319, "step": 20335} +{"lr": 5.174987276538342e-05, "data_time": 0.2419429302215576, "loss": 0.10093883015215396, "time": 0.859332537651062, "epoch": 6, "memory": 6319, "step": 20435} +{"lr": 5.1389084725702425e-05, "data_time": 0.2607812643051147, "loss": 0.08994136080145836, "time": 0.8670778751373291, "epoch": 6, "memory": 6319, "step": 20535} +{"lr": 5.1028530154300045e-05, "data_time": 0.18812243938446044, "loss": 0.09303448684513568, "time": 0.8788151025772095, "epoch": 6, "memory": 6319, "step": 20635} +{"lr": 5.066823236328663e-05, "data_time": 0.2563325881958008, "loss": 0.08541717454791069, "time": 0.8980855226516724, "epoch": 6, "memory": 6319, "step": 20735} +{"lr": 5.030821464817006e-05, "data_time": 0.2645404577255249, "loss": 0.08788195103406907, "time": 0.8827181339263916, "epoch": 6, "memory": 6319, "step": 20835} +{"lr": 4.99485002863494e-05, "data_time": 0.3766984701156616, "loss": 0.09297304302453994, "time": 0.8823019742965699, "epoch": 6, "memory": 6319, "step": 20935} +{"lr": 4.95891125356102e-05, "data_time": 0.4078476905822754, "loss": 0.08156409449875354, "time": 0.9584446907043457, "epoch": 6, "memory": 6319, "step": 21035} +{"lr": 4.923007463262053e-05, "data_time": 0.10058321952819824, "loss": 0.07987777292728424, "time": 0.9098809242248536, "epoch": 6, "memory": 6319, "step": 21135} +{"lr": 4.887140979142852e-05, "data_time": 0.08175497055053711, "loss": 0.0818291261792183, "time": 0.8497950792312622, "epoch": 6, "memory": 6319, "step": 21235} +{"lr": 4.851314120196161e-05, "data_time": 0.011955714225769043, "loss": 0.08152264468371868, "time": 0.8745648622512817, "epoch": 6, "memory": 6319, "step": 21335} +{"lr": 4.815529202852693e-05, "data_time": 0.13461427688598632, "loss": 0.08553754091262818, "time": 0.9074079513549804, "epoch": 6, "memory": 6319, "step": 21435} +{"lr": 4.779788540831406e-05, "data_time": 0.18175103664398193, "loss": 0.09324220716953277, "time": 0.8948556423187256, "epoch": 6, "memory": 6319, "step": 21535} +{"lr": 4.744094444989843e-05, "data_time": 0.23326785564422609, "loss": 0.08899538591504097, "time": 0.8990905284881592, "epoch": 6, "memory": 6319, "step": 21635} +{"lr": 4.708449223174764e-05, "data_time": 0.1924884557723999, "loss": 0.08449041098356247, "time": 0.8605105876922607, "epoch": 6, "memory": 6319, "step": 21735} +{"lr": 4.672855180072916e-05, "data_time": 0.0010923385620117188, "loss": 0.0837862677872181, "time": 0.8763978004455566, "epoch": 6, "memory": 6319, "step": 21835} +{"lr": 4.6373146170620256e-05, "data_time": 0.0008182525634765625, "loss": 0.09838488847017288, "time": 0.8523350954055786, "epoch": 6, "memory": 6319, "step": 21935} +{"lr": 4.6018298320619915e-05, "data_time": 0.03775930404663086, "loss": 0.0832609124481678, "time": 0.835089898109436, "epoch": 6, "memory": 6319, "step": 22035} +{"lr": 4.566403119386318e-05, "data_time": 0.1655141830444336, "loss": 0.08210644125938416, "time": 0.8597707271575927, "epoch": 6, "memory": 6319, "step": 22135} +{"lr": 4.531036769593769e-05, "data_time": 0.0012761592864990235, "loss": 0.07907106652855873, "time": 0.9598047971725464, "epoch": 6, "memory": 6319, "step": 22235} +{"lr": 4.4957330693402685e-05, "data_time": 0.0012713432312011718, "loss": 0.07999893091619015, "time": 0.9161706924438476, "epoch": 6, "memory": 6319, "step": 22335} +{"lr": 4.460494301231059e-05, "data_time": 0.0011863231658935547, "loss": 0.08322003372013569, "time": 0.8416136026382446, "epoch": 6, "memory": 6319, "step": 22435} +{"lr": 4.425322743673109e-05, "data_time": 0.0013361215591430665, "loss": 0.0836469754576683, "time": 0.8517721652984619, "epoch": 6, "memory": 6319, "step": 22535} +{"lr": 4.390220670727812e-05, "data_time": 0.0010993242263793944, "loss": 0.09141123667359352, "time": 0.8202883481979371, "epoch": 6, "memory": 6319, "step": 22635} +{"lr": 4.3551903519639325e-05, "data_time": 0.0015716791152954102, "loss": 0.07976200245320797, "time": 0.9105865955352783, "epoch": 6, "memory": 6319, "step": 22735} +{"lr": 4.3202340523108894e-05, "data_time": 0.0014548778533935546, "loss": 0.08041631653904915, "time": 0.8781120300292968, "epoch": 6, "memory": 6319, "step": 22835} +{"lr": 4.285354031912302e-05, "data_time": 0.0010716915130615234, "loss": 0.08340962044894695, "time": 0.8195673227310181, "epoch": 6, "memory": 6319, "step": 22935} +{"lr": 4.2505525459798544e-05, "data_time": 0.0014203786849975586, "loss": 0.08337981961667537, "time": 0.823271656036377, "epoch": 6, "memory": 6319, "step": 23035} +{"lr": 4.2158318446474836e-05, "data_time": 0.001408839225769043, "loss": 0.07958810180425643, "time": 0.8592772245407104, "epoch": 6, "memory": 6319, "step": 23135} +{"lr": 4.181194172825905e-05, "data_time": 0.001300811767578125, "loss": 0.07475061267614365, "time": 0.915073823928833, "epoch": 6, "memory": 6319, "step": 23235} +{"lr": 4.146641770057448e-05, "data_time": 0.0018882513046264648, "loss": 0.0789183296263218, "time": 0.9030552387237549, "epoch": 6, "memory": 6319, "step": 23335} +{"lr": 4.112176870371267e-05, "data_time": 0.0014145851135253906, "loss": 0.08289809748530388, "time": 0.8818341016769409, "epoch": 6, "memory": 6319, "step": 23435} +{"accuracy/top1": 95.6890869140625, "data_time": 0.5603013326381815, "time": 0.6510588382852489, "step": 6} +{"lr": 4.0753988582732786e-05, "data_time": 0.6458329916000366, "loss": 0.08821750245988369, "time": 0.8787811040878296, "epoch": 7, "memory": 6319, "step": 23542} +{"lr": 4.0411221639286815e-05, "data_time": 0.7044112920761109, "loss": 0.0914335660636425, "time": 0.9160228729248047, "epoch": 7, "memory": 6319, "step": 23642} +{"lr": 4.006939795169518e-05, "data_time": 0.4019924163818359, "loss": 0.07337699197232724, "time": 0.8741212368011475, "epoch": 7, "memory": 6319, "step": 23742} +{"lr": 3.972853962099913e-05, "data_time": 0.2172248363494873, "loss": 0.0989815104752779, "time": 0.9004807472229004, "epoch": 7, "memory": 6319, "step": 23842} +{"lr": 3.93886686858237e-05, "data_time": 0.28595924377441406, "loss": 0.10283050090074539, "time": 0.8484663724899292, "epoch": 7, "memory": 6319, "step": 23942} +{"lr": 3.904980712095259e-05, "data_time": 0.3061556100845337, "loss": 0.07841564007103444, "time": 0.8925023555755616, "epoch": 7, "memory": 6319, "step": 24042} +{"lr": 3.871197683590748e-05, "data_time": 0.21903626918792723, "loss": 0.07981203086674213, "time": 0.8458019733428955, "epoch": 7, "memory": 6319, "step": 24142} +{"lr": 3.837519967353125e-05, "data_time": 0.2449690580368042, "loss": 0.08555740900337697, "time": 0.8858100652694703, "epoch": 7, "memory": 6319, "step": 24242} +{"lr": 3.8039497408575855e-05, "data_time": 0.1595212697982788, "loss": 0.08635051213204861, "time": 0.8840290307998657, "epoch": 7, "memory": 6319, "step": 24342} +{"lr": 3.770489174629432e-05, "data_time": 0.12182822227478027, "loss": 0.07224807813763619, "time": 0.869728422164917, "epoch": 7, "memory": 6319, "step": 24442} +{"lr": 3.737140432103754e-05, "data_time": 0.19448225498199462, "loss": 0.08117571622133254, "time": 0.9084516048431397, "epoch": 7, "memory": 6319, "step": 24542} +{"lr": 3.703905669485545e-05, "data_time": 0.15427806377410888, "loss": 0.08219698742032051, "time": 0.877875804901123, "epoch": 7, "memory": 6319, "step": 24642} +{"lr": 3.6707870356102706e-05, "data_time": 0.18405275344848632, "loss": 0.08146934397518635, "time": 0.9229199647903442, "epoch": 7, "memory": 6319, "step": 24742} +{"lr": 3.6377866718049484e-05, "data_time": 0.0011170387268066406, "loss": 0.09072392359375954, "time": 0.9015771865844726, "epoch": 7, "memory": 6319, "step": 24842} +{"lr": 3.6049067117496965e-05, "data_time": 0.0011453628540039062, "loss": 0.08101946152746678, "time": 0.8718708515167236, "epoch": 7, "memory": 6319, "step": 24942} +{"lr": 3.5721492813397755e-05, "data_time": 0.0017980337142944336, "loss": 0.08431394286453724, "time": 0.880272102355957, "epoch": 7, "memory": 6319, "step": 25042} +{"lr": 3.539516498548151e-05, "data_time": 0.0012185096740722657, "loss": 0.07376923486590385, "time": 0.9023571014404297, "epoch": 7, "memory": 6319, "step": 25142} +{"lr": 3.507010473288518e-05, "data_time": 0.0011278867721557617, "loss": 0.08985093757510185, "time": 0.8491546630859375, "epoch": 7, "memory": 6319, "step": 25242} +{"lr": 3.474633307278909e-05, "data_time": 0.031475830078125, "loss": 0.09947454854846001, "time": 0.8629065990447998, "epoch": 7, "memory": 6319, "step": 25342} +{"lr": 3.4423870939058177e-05, "data_time": 0.4481214046478271, "loss": 0.08215671852231025, "time": 0.9036730527877808, "epoch": 7, "memory": 6319, "step": 25442} +{"lr": 3.410273918088805e-05, "data_time": 0.3497700452804565, "loss": 0.07424916923046113, "time": 0.8509632587432862, "epoch": 7, "memory": 6319, "step": 25542} +{"lr": 3.37829585614574e-05, "data_time": 0.20111565589904784, "loss": 0.08338887766003608, "time": 0.9196459531784058, "epoch": 7, "memory": 6319, "step": 25642} +{"lr": 3.3464549756585216e-05, "data_time": 0.001168227195739746, "loss": 0.07143484726548195, "time": 0.9184943675994873, "epoch": 7, "memory": 6319, "step": 25742} +{"lr": 3.314753335339416e-05, "data_time": 0.0012459754943847656, "loss": 0.07596002370119095, "time": 0.9016108989715577, "epoch": 7, "memory": 6319, "step": 25842} +{"lr": 3.2831929848979394e-05, "data_time": 0.00139310359954834, "loss": 0.08242843970656395, "time": 0.9291356086730957, "epoch": 7, "memory": 6319, "step": 25942} +{"lr": 3.251775964908332e-05, "data_time": 0.0008750200271606445, "loss": 0.09521238580346107, "time": 0.8514190673828125, "epoch": 7, "memory": 6319, "step": 26042} +{"lr": 3.22050430667762e-05, "data_time": 0.0010126352310180664, "loss": 0.07639812268316745, "time": 0.8465380907058716, "epoch": 7, "memory": 6319, "step": 26142} +{"lr": 3.189380032114285e-05, "data_time": 0.16205165386199952, "loss": 0.07312794998288155, "time": 0.9169226884841919, "epoch": 7, "memory": 6319, "step": 26242} +{"lr": 3.1584051535975324e-05, "data_time": 0.5428041934967041, "loss": 0.09266364872455597, "time": 0.9132752656936646, "epoch": 7, "memory": 6319, "step": 26342} +{"lr": 3.127581673847175e-05, "data_time": 0.5764893770217896, "loss": 0.08598207011818886, "time": 0.9185611724853515, "epoch": 7, "memory": 6319, "step": 26442} +{"lr": 3.0969115857941555e-05, "data_time": 0.27761924266815186, "loss": 0.09996403455734253, "time": 0.8727656841278076, "epoch": 7, "memory": 6319, "step": 26542} +{"lr": 3.0663968724516654e-05, "data_time": 0.41826543807983396, "loss": 0.0922010935842991, "time": 0.889160418510437, "epoch": 7, "memory": 6319, "step": 26642} +{"lr": 3.0360395067869668e-05, "data_time": 0.4560523271560669, "loss": 0.07513608299195766, "time": 0.8984469413757324, "epoch": 7, "memory": 6319, "step": 26742} +{"lr": 3.0058414515938063e-05, "data_time": 0.029741072654724122, "loss": 0.0829878356307745, "time": 0.8723999500274658, "epoch": 7, "memory": 6319, "step": 26842} +{"lr": 2.975804659365501e-05, "data_time": 0.0014225244522094727, "loss": 0.08196965903043747, "time": 0.8736865758895874, "epoch": 7, "memory": 6319, "step": 26942} +{"lr": 2.945931072168714e-05, "data_time": 0.0012315034866333008, "loss": 0.09356333091855049, "time": 0.8833344459533692, "epoch": 7, "memory": 6319, "step": 27042} +{"lr": 2.9162226215178855e-05, "data_time": 0.0011543512344360351, "loss": 0.07178382985293866, "time": 0.9196656227111817, "epoch": 7, "memory": 6319, "step": 27142} +{"lr": 2.8866812282503322e-05, "data_time": 0.0011996030807495117, "loss": 0.07366007119417191, "time": 0.8804421663284302, "epoch": 7, "memory": 6319, "step": 27242} +{"lr": 2.8573088024020838e-05, "data_time": 0.0008530139923095704, "loss": 0.06969595476984977, "time": 0.8878621578216552, "epoch": 7, "memory": 6319, "step": 27342} +{"accuracy/top1": 96.15531921386719, "data_time": 0.5790184900678438, "time": 0.6665845451683834, "step": 7} +{"lr": 2.8260695817390092e-05, "data_time": 0.773190975189209, "loss": 0.0676898017525673, "time": 0.9133787631988526, "epoch": 8, "memory": 6319, "step": 27449} +{"lr": 2.7970529402570967e-05, "data_time": 0.7257717609405517, "loss": 0.08663739562034607, "time": 0.8652721643447876, "epoch": 8, "memory": 6319, "step": 27549} +{"lr": 2.7682110612244214e-05, "data_time": 0.7518941640853882, "loss": 0.08318739645183086, "time": 0.8933749675750733, "epoch": 8, "memory": 6319, "step": 27649} +{"lr": 2.7395458094490283e-05, "data_time": 0.7072431564331054, "loss": 0.08465914279222489, "time": 0.8481247425079346, "epoch": 8, "memory": 6319, "step": 27749} +{"lr": 2.7110590383189015e-05, "data_time": 0.7882378816604614, "loss": 0.08485641404986381, "time": 0.9298220157623291, "epoch": 8, "memory": 6319, "step": 27849} +{"lr": 2.6827525896821397e-05, "data_time": 0.4758735656738281, "loss": 0.08296169936656952, "time": 1.0543368577957153, "epoch": 8, "memory": 6319, "step": 27949} +{"lr": 2.6546282937278552e-05, "data_time": 0.06723496913909913, "loss": 0.08053266108036042, "time": 0.8654393196105957, "epoch": 8, "memory": 6319, "step": 28049} +{"lr": 2.6266879688678606e-05, "data_time": 0.12370157241821289, "loss": 0.0872628390789032, "time": 0.8752964973449707, "epoch": 8, "memory": 6319, "step": 28149} +{"lr": 2.5989334216190738e-05, "data_time": 0.4160438537597656, "loss": 0.07510889694094658, "time": 0.8809774398803711, "epoch": 8, "memory": 6319, "step": 28249} +{"lr": 2.571366446486732e-05, "data_time": 0.7109492301940918, "loss": 0.0779775895178318, "time": 0.8698572635650634, "epoch": 8, "memory": 6319, "step": 28349} +{"lr": 2.5439888258483665e-05, "data_time": 0.5894846439361572, "loss": 0.08027538731694221, "time": 0.8340277433395386, "epoch": 8, "memory": 6319, "step": 28449} +{"lr": 2.5168023298385384e-05, "data_time": 0.6976788520812989, "loss": 0.0870208889245987, "time": 0.9264285564422607, "epoch": 8, "memory": 6319, "step": 28549} +{"lr": 2.489808716234412e-05, "data_time": 0.5857330560684204, "loss": 0.07486210726201534, "time": 0.886945366859436, "epoch": 8, "memory": 6319, "step": 28649} +{"lr": 2.463009730342111e-05, "data_time": 0.6903819084167481, "loss": 0.0905293308198452, "time": 0.8788272619247437, "epoch": 8, "memory": 6319, "step": 28749} +{"lr": 2.436407104883849e-05, "data_time": 0.7645008563995361, "loss": 0.0734794195741415, "time": 0.9070576906204224, "epoch": 8, "memory": 6319, "step": 28849} +{"lr": 2.4100025598859092e-05, "data_time": 0.7549132585525513, "loss": 0.0857387587428093, "time": 0.8952568292617797, "epoch": 8, "memory": 6319, "step": 28949} +{"lr": 2.3837978025674284e-05, "data_time": 0.634507131576538, "loss": 0.07850695848464966, "time": 0.8755037307739257, "epoch": 8, "memory": 6319, "step": 29049} +{"lr": 2.357794527230027e-05, "data_time": 0.7431282997131348, "loss": 0.0772418100386858, "time": 0.8832640409469604, "epoch": 8, "memory": 6319, "step": 29149} +{"lr": 2.33199441514825e-05, "data_time": 0.7974025964736938, "loss": 0.08501079007983207, "time": 0.9383984088897706, "epoch": 8, "memory": 6319, "step": 29249} +{"lr": 2.306399134460859e-05, "data_time": 0.7360717058181763, "loss": 0.08238439373672009, "time": 0.8765291929244995, "epoch": 8, "memory": 6319, "step": 29349} +{"lr": 2.281010340062995e-05, "data_time": 0.7424033403396606, "loss": 0.09331198707222939, "time": 0.8944615840911865, "epoch": 8, "memory": 6319, "step": 29449} +{"lr": 2.2558296734991603e-05, "data_time": 0.7500584602355957, "loss": 0.07346097975969315, "time": 0.8896821498870849, "epoch": 8, "memory": 6319, "step": 29549} +{"lr": 2.2308587628570915e-05, "data_time": 0.7491367340087891, "loss": 0.08215555846691132, "time": 0.8904971122741699, "epoch": 8, "memory": 6319, "step": 29649} +{"lr": 2.2060992226624955e-05, "data_time": 0.7169220209121704, "loss": 0.0714668869972229, "time": 0.8568846464157105, "epoch": 8, "memory": 6319, "step": 29749} +{"lr": 2.181552653774648e-05, "data_time": 0.7214415788650512, "loss": 0.07901271656155587, "time": 0.8694997549057006, "epoch": 8, "memory": 6319, "step": 29849} +{"lr": 2.1572206432829035e-05, "data_time": 0.7497676610946655, "loss": 0.07336163073778153, "time": 0.8897875070571899, "epoch": 8, "memory": 6319, "step": 29949} +{"lr": 2.1331047644040754e-05, "data_time": 0.7227826356887818, "loss": 0.08376354165375233, "time": 0.8631266593933106, "epoch": 8, "memory": 6319, "step": 30049} +{"lr": 2.109206576380705e-05, "data_time": 0.7716203212738038, "loss": 0.08364129476249219, "time": 0.9124569177627564, "epoch": 8, "memory": 6319, "step": 30149} +{"lr": 2.0855276243802687e-05, "data_time": 0.7222044229507446, "loss": 0.07447917945683002, "time": 0.8624828338623047, "epoch": 8, "memory": 6319, "step": 30249} +{"lr": 2.0620694393952487e-05, "data_time": 0.6082808494567871, "loss": 0.0637462567538023, "time": 0.9054466009140014, "epoch": 8, "memory": 6319, "step": 30349} +{"lr": 2.038833538144169e-05, "data_time": 0.6377532243728637, "loss": 0.07222930155694485, "time": 0.8785322666168213, "epoch": 8, "memory": 6319, "step": 30449} +{"lr": 2.0158214229735154e-05, "data_time": 0.6940939903259278, "loss": 0.07766427472233772, "time": 0.8363755941390991, "epoch": 8, "memory": 6319, "step": 30549} +{"lr": 1.9930345817605977e-05, "data_time": 0.717761754989624, "loss": 0.08640623912215233, "time": 0.9249245882034302, "epoch": 8, "memory": 6319, "step": 30649} +{"lr": 1.970474487817358e-05, "data_time": 0.7321541547775269, "loss": 0.0824542447924614, "time": 0.8729692220687866, "epoch": 8, "memory": 6319, "step": 30749} +{"lr": 1.9481425997951092e-05, "data_time": 0.7570785045623779, "loss": 0.08402719981968403, "time": 0.8991684675216675, "epoch": 8, "memory": 6319, "step": 30849} +{"lr": 1.9260403615902187e-05, "data_time": 0.7063649892807007, "loss": 0.07551782429218293, "time": 0.8475235223770141, "epoch": 8, "memory": 6319, "step": 30949} +{"lr": 1.9041692022507647e-05, "data_time": 0.7113152265548706, "loss": 0.07760469317436218, "time": 0.8525068998336792, "epoch": 8, "memory": 6319, "step": 31049} +{"lr": 1.8825305358841125e-05, "data_time": 0.7466799259185791, "loss": 0.0744364645332098, "time": 0.8883065938949585, "epoch": 8, "memory": 6319, "step": 31149} +{"lr": 1.861125761565518e-05, "data_time": 0.5244179010391236, "loss": 0.07091340497136116, "time": 0.8617112398147583, "epoch": 8, "memory": 6319, "step": 31249} +{"accuracy/top1": 96.42671203613281, "data_time": 0.5710797926475262, "time": 0.6589143235107948, "step": 8} +{"lr": 1.8384832449122864e-05, "data_time": 0.40781588554382325, "loss": 0.07775134034454823, "time": 0.9066422462463379, "epoch": 9, "memory": 6319, "step": 31356} +{"lr": 1.817567007332114e-05, "data_time": 0.41288764476776124, "loss": 0.07856393307447433, "time": 0.9140744686126709, "epoch": 9, "memory": 6319, "step": 31456} +{"lr": 1.796888862098825e-05, "data_time": 0.49449379444122316, "loss": 0.08160603269934655, "time": 0.853122067451477, "epoch": 9, "memory": 6319, "step": 31556} +{"lr": 1.7764501461839678e-05, "data_time": 0.7420816659927368, "loss": 0.08109548687934875, "time": 0.8830325365066528, "epoch": 9, "memory": 6319, "step": 31656} +{"lr": 1.7562521810784746e-05, "data_time": 0.709919023513794, "loss": 0.06176501996815205, "time": 0.8515275001525879, "epoch": 9, "memory": 6319, "step": 31756} +{"lr": 1.7362962727072444e-05, "data_time": 0.43931331634521487, "loss": 0.07960855662822723, "time": 0.8904636859893799, "epoch": 9, "memory": 6319, "step": 31856} +{"lr": 1.716583711344682e-05, "data_time": 0.29276366233825685, "loss": 0.06693844497203827, "time": 0.8681150674819946, "epoch": 9, "memory": 6319, "step": 31956} +{"lr": 1.6971157715312954e-05, "data_time": 0.28310532569885255, "loss": 0.100167166441679, "time": 0.8865897417068481, "epoch": 9, "memory": 6319, "step": 32056} +{"lr": 1.677893711991274e-05, "data_time": 0.3116255044937134, "loss": 0.0696358185261488, "time": 0.9002179384231568, "epoch": 9, "memory": 6319, "step": 32156} +{"lr": 1.6589187755511025e-05, "data_time": 0.3039934396743774, "loss": 0.07503429688513279, "time": 0.8633481740951539, "epoch": 9, "memory": 6319, "step": 32256} +{"lr": 1.640192189059216e-05, "data_time": 0.0011800289154052734, "loss": 0.07262629978358745, "time": 0.8771113634109498, "epoch": 9, "memory": 6319, "step": 32356} +{"lr": 1.6217151633066716e-05, "data_time": 0.0017085790634155274, "loss": 0.0713750522583723, "time": 0.9277477264404297, "epoch": 9, "memory": 6319, "step": 32456} +{"lr": 1.6034888929488564e-05, "data_time": 0.0017104148864746094, "loss": 0.07639891132712365, "time": 0.8678189516067505, "epoch": 9, "memory": 6319, "step": 32556} +{"lr": 1.585514556428258e-05, "data_time": 0.0013008356094360352, "loss": 0.06912901028990745, "time": 0.8992761611938477, "epoch": 9, "memory": 6319, "step": 32656} +{"lr": 1.567793315898261e-05, "data_time": 0.0017913579940795898, "loss": 0.07725722081959248, "time": 0.8712166547775269, "epoch": 9, "memory": 6319, "step": 32756} +{"lr": 1.5503263171480048e-05, "data_time": 0.09125230312347413, "loss": 0.07543345391750336, "time": 0.8847838401794433, "epoch": 9, "memory": 6319, "step": 32856} +{"lr": 1.5331146895283165e-05, "data_time": 0.5145464181900025, "loss": 0.08202039748430252, "time": 0.9210999011993408, "epoch": 9, "memory": 6319, "step": 32956} +{"lr": 1.5161595458786728e-05, "data_time": 0.3576303720474243, "loss": 0.08256775140762329, "time": 0.8308727979660034, "epoch": 9, "memory": 6319, "step": 33056} +{"lr": 1.4994619824552563e-05, "data_time": 0.5439150333404541, "loss": 0.07742651663720608, "time": 0.8744504928588868, "epoch": 9, "memory": 6319, "step": 33156} +{"lr": 1.4830230788600802e-05, "data_time": 0.6360100984573365, "loss": 0.07649558186531066, "time": 0.8381055355072021, "epoch": 9, "memory": 6319, "step": 33256} +{"lr": 1.4668438979711779e-05, "data_time": 0.7989260911941528, "loss": 0.07968887686729431, "time": 0.9380785226821899, "epoch": 9, "memory": 6319, "step": 33356} +{"lr": 1.4509254858738844e-05, "data_time": 0.7136792421340943, "loss": 0.08204310238361359, "time": 0.8539438724517823, "epoch": 9, "memory": 6319, "step": 33456} +{"lr": 1.4352688717931966e-05, "data_time": 0.7313532590866089, "loss": 0.07945706844329833, "time": 0.874568772315979, "epoch": 9, "memory": 6319, "step": 33556} +{"lr": 1.4198750680272348e-05, "data_time": 0.6613088607788086, "loss": 0.0904395893216133, "time": 0.8789570808410645, "epoch": 9, "memory": 6319, "step": 33656} +{"lr": 1.4047450698817857e-05, "data_time": 0.7777715921401978, "loss": 0.07187089733779431, "time": 0.9182924032211304, "epoch": 9, "memory": 6319, "step": 33756} +{"lr": 1.389879855605952e-05, "data_time": 0.7126684188842773, "loss": 0.0718031745404005, "time": 0.857210898399353, "epoch": 9, "memory": 6319, "step": 33856} +{"lr": 1.3752803863288976e-05, "data_time": 0.7748696327209472, "loss": 0.0728682342916727, "time": 0.9158221006393432, "epoch": 9, "memory": 6319, "step": 33956} +{"lr": 1.3609476059977129e-05, "data_time": 0.7824553489685059, "loss": 0.07123819030821324, "time": 0.9341459274291992, "epoch": 9, "memory": 6319, "step": 34056} +{"lr": 1.3468824413163754e-05, "data_time": 0.7434473514556885, "loss": 0.0806659147143364, "time": 0.8847616910934448, "epoch": 9, "memory": 6319, "step": 34156} +{"lr": 1.3330858016858395e-05, "data_time": 0.7036133050918579, "loss": 0.08593320660293102, "time": 0.844521164894104, "epoch": 9, "memory": 6319, "step": 34256} +{"lr": 1.3195585791452309e-05, "data_time": 0.7089182615280152, "loss": 0.08662436828017235, "time": 0.8497709274291992, "epoch": 9, "memory": 6319, "step": 34356} +{"lr": 1.3063016483141729e-05, "data_time": 0.6240370035171509, "loss": 0.08690226525068283, "time": 0.8963229179382324, "epoch": 9, "memory": 6319, "step": 34456} +{"lr": 1.2933158663362388e-05, "data_time": 0.7880263805389405, "loss": 0.07403073199093342, "time": 1.017116951942444, "epoch": 9, "memory": 6319, "step": 34556} +{"lr": 1.2806020728235311e-05, "data_time": 0.6992140769958496, "loss": 0.06560872867703438, "time": 0.8916797637939453, "epoch": 9, "memory": 6319, "step": 34656} +{"lr": 1.2681610898023963e-05, "data_time": 0.5774627685546875, "loss": 0.07340772338211536, "time": 0.9182001352310181, "epoch": 9, "memory": 6319, "step": 34756} +{"lr": 1.2559937216602731e-05, "data_time": 0.7716176509857178, "loss": 0.06270115822553635, "time": 0.9139045953750611, "epoch": 9, "memory": 6319, "step": 34856} +{"lr": 1.2441007550936847e-05, "data_time": 0.7892608642578125, "loss": 0.08152935728430748, "time": 0.9311720371246338, "epoch": 9, "memory": 6319, "step": 34956} +{"lr": 1.232482959057377e-05, "data_time": 0.725196647644043, "loss": 0.07424784190952778, "time": 0.8679368495941162, "epoch": 9, "memory": 6319, "step": 35056} +{"lr": 1.2211410847145969e-05, "data_time": 0.6607789278030396, "loss": 0.08571186661720276, "time": 0.8571977138519287, "epoch": 9, "memory": 6319, "step": 35156} +{"accuracy/top1": 96.78160095214844, "data_time": 0.5831122521696419, "time": 0.6737202118183004, "step": 9} +{"lr": 1.209311679360931e-05, "data_time": 0.8175110578536987, "loss": 0.08883111737668514, "time": 0.9578233480453491, "epoch": 10, "memory": 6319, "step": 35263} +{"lr": 1.1985432727488666e-05, "data_time": 0.7931027412414551, "loss": 0.07517887912690639, "time": 0.9331846237182617, "epoch": 10, "memory": 6319, "step": 35363} +{"lr": 1.1880529822435289e-05, "data_time": 0.7581690788269043, "loss": 0.0755938570946455, "time": 0.8997335195541382, "epoch": 10, "memory": 6319, "step": 35463} +{"lr": 1.1778414861078726e-05, "data_time": 0.7512569665908814, "loss": 0.0721186900511384, "time": 0.8914328336715698, "epoch": 10, "memory": 6319, "step": 35563} +{"lr": 1.1679094445790538e-05, "data_time": 0.67048020362854, "loss": 0.08164873197674752, "time": 0.8105042934417724, "epoch": 10, "memory": 6319, "step": 35663} +{"lr": 1.1582574998257377e-05, "data_time": 0.7606892585754395, "loss": 0.07552231177687645, "time": 0.9008347272872925, "epoch": 10, "memory": 6319, "step": 35763} +{"lr": 1.1488862759065762e-05, "data_time": 0.7632877588272095, "loss": 0.07405286356806755, "time": 0.9042915582656861, "epoch": 10, "memory": 6319, "step": 35863} +{"lr": 1.1397963787298665e-05, "data_time": 0.7482366323471069, "loss": 0.07233749367296696, "time": 0.8890695095062255, "epoch": 10, "memory": 6319, "step": 35963} +{"lr": 1.1309883960143664e-05, "data_time": 0.7255772352218628, "loss": 0.0837469134479761, "time": 0.8934251070022583, "epoch": 10, "memory": 6319, "step": 36063} +{"lr": 1.122462897251302e-05, "data_time": 0.7596636295318604, "loss": 0.07752110138535499, "time": 0.9017051696777344, "epoch": 10, "memory": 6319, "step": 36163} +{"lr": 1.114220433667542e-05, "data_time": 0.752418041229248, "loss": 0.07443158216774463, "time": 0.9325988292694092, "epoch": 10, "memory": 6319, "step": 36263} +{"lr": 1.10626153818996e-05, "data_time": 0.6474128723144531, "loss": 0.0840250864624977, "time": 0.9051640748977661, "epoch": 10, "memory": 6319, "step": 36363} +{"lr": 1.0985867254109761e-05, "data_time": 0.5261065483093261, "loss": 0.09787520878016949, "time": 0.849211311340332, "epoch": 10, "memory": 6319, "step": 36463} +{"lr": 1.0911964915552823e-05, "data_time": 0.2851930379867554, "loss": 0.09178892150521278, "time": 0.960207176208496, "epoch": 10, "memory": 6319, "step": 36563} +{"lr": 1.0840913144477643e-05, "data_time": 0.14056668281555176, "loss": 0.06464039199054242, "time": 0.8757411241531372, "epoch": 10, "memory": 6319, "step": 36663} +{"lr": 1.0772716534826064e-05, "data_time": 0.001236724853515625, "loss": 0.07767030596733093, "time": 0.8480842351913452, "epoch": 10, "memory": 6319, "step": 36763} +{"lr": 1.0707379495935828e-05, "data_time": 0.0012414216995239257, "loss": 0.07756403498351574, "time": 0.9160326242446899, "epoch": 10, "memory": 6319, "step": 36863} +{"lr": 1.0644906252255541e-05, "data_time": 0.002115345001220703, "loss": 0.06526481397449971, "time": 0.850139856338501, "epoch": 10, "memory": 6319, "step": 36963} +{"lr": 1.0585300843071527e-05, "data_time": 0.0017118215560913085, "loss": 0.07799581438302994, "time": 0.8961062908172608, "epoch": 10, "memory": 6319, "step": 37063} +{"lr": 1.0528567122246654e-05, "data_time": 0.12719011306762695, "loss": 0.0727805256843567, "time": 0.8929010629653931, "epoch": 10, "memory": 6319, "step": 37163} +{"lr": 1.0474708757971141e-05, "data_time": 0.007215547561645508, "loss": 0.0660861760377884, "time": 0.9187443494796753, "epoch": 10, "memory": 6319, "step": 37263} +{"lr": 1.042372923252543e-05, "data_time": 0.001841592788696289, "loss": 0.08520641326904296, "time": 0.9100911855697632, "epoch": 10, "memory": 6319, "step": 37363} +{"lr": 1.0375631842054998e-05, "data_time": 0.13324143886566162, "loss": 0.0835212867707014, "time": 0.859965467453003, "epoch": 10, "memory": 6319, "step": 37463} +{"lr": 1.0330419696357254e-05, "data_time": 0.0011774063110351562, "loss": 0.07549788057804108, "time": 0.9032527923583984, "epoch": 10, "memory": 6319, "step": 37563} +{"lr": 1.028809571868047e-05, "data_time": 0.001439356803894043, "loss": 0.08595659658312797, "time": 0.9312840700149536, "epoch": 10, "memory": 6319, "step": 37663} +{"lr": 1.0248662645534766e-05, "data_time": 0.0009754419326782227, "loss": 0.0685815777629614, "time": 0.9517098188400268, "epoch": 10, "memory": 6319, "step": 37763} +{"lr": 1.0212123026515194e-05, "data_time": 0.0015410661697387695, "loss": 0.07711313255131244, "time": 0.9567354917526245, "epoch": 10, "memory": 6319, "step": 37863} +{"lr": 1.0178479224136881e-05, "data_time": 0.002238559722900391, "loss": 0.07415566071867943, "time": 0.8813257455825806, "epoch": 10, "memory": 6319, "step": 37963} +{"lr": 1.0147733413682289e-05, "data_time": 0.001111006736755371, "loss": 0.07807697914540768, "time": 0.8484270811080933, "epoch": 10, "memory": 6319, "step": 38063} +{"lr": 1.0119887583060527e-05, "data_time": 0.0015458345413208007, "loss": 0.07257316969335079, "time": 0.8970549821853637, "epoch": 10, "memory": 6319, "step": 38163} +{"lr": 1.0094943532678871e-05, "data_time": 0.0012021780014038086, "loss": 0.08120028041303158, "time": 0.9232643127441407, "epoch": 10, "memory": 6319, "step": 38263} +{"lr": 1.007290287532637e-05, "data_time": 0.0009762763977050782, "loss": 0.08372033014893532, "time": 0.8691381931304931, "epoch": 10, "memory": 6319, "step": 38363} +{"lr": 1.0053767036069495e-05, "data_time": 0.005805325508117676, "loss": 0.08308596312999725, "time": 0.9090308427810669, "epoch": 10, "memory": 6319, "step": 38463} +{"lr": 1.003753725216005e-05, "data_time": 0.16092276573181152, "loss": 0.07852649614214897, "time": 0.888532018661499, "epoch": 10, "memory": 6319, "step": 38563} +{"lr": 1.0024214572955152e-05, "data_time": 0.08160817623138428, "loss": 0.07156525179743767, "time": 0.9100588083267211, "epoch": 10, "memory": 6319, "step": 38663} +{"lr": 1.0013799859849456e-05, "data_time": 0.0012605905532836914, "loss": 0.07482991330325603, "time": 0.836520004272461, "epoch": 10, "memory": 6319, "step": 38763} +{"lr": 1.0006293786219338e-05, "data_time": 0.0010182619094848632, "loss": 0.06851191893219948, "time": 0.7639588356018067, "epoch": 10, "memory": 6319, "step": 38863} +{"lr": 1.0001696837379472e-05, "data_time": 0.685526442527771, "loss": 0.08617636188864708, "time": 0.852595043182373, "epoch": 10, "memory": 6319, "step": 38963} +{"lr": 1.0000009310551361e-05, "data_time": 0.3220003366470337, "loss": 0.07605950310826301, "time": 0.8021457433700562, "epoch": 10, "memory": 6319, "step": 39063} +{"accuracy/top1": 96.87554931640625, "data_time": 0.5467592231158552, "time": 0.6363690187191141, "step": 10} diff --git a/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/20230605_005004/vis_data/vis_image/0000000.png_1.png b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/20230605_005004/vis_data/vis_image/0000000.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..ac3d17062602ef75488cedba63e72feb167f940a Binary files /dev/null and b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/20230605_005004/vis_data/vis_image/0000000.png_1.png differ diff --git a/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/20230605_005004/vis_data/vis_image/0000000.png_10.png b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/20230605_005004/vis_data/vis_image/0000000.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..1e54b266646ae76241588359b9a79e1eedff6790 Binary files /dev/null and b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/20230605_005004/vis_data/vis_image/0000000.png_10.png differ diff --git a/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/20230605_005004/vis_data/vis_image/0000000.png_2.png b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/20230605_005004/vis_data/vis_image/0000000.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..470d671e19e511d2d559ee558e4a6718514d21d0 Binary files /dev/null and b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/20230605_005004/vis_data/vis_image/0000000.png_2.png differ diff --git a/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/20230605_005004/vis_data/vis_image/0000000.png_3.png b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/20230605_005004/vis_data/vis_image/0000000.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..e71aa252f3df9f869b91eaa49ea7191e9ca46b48 Binary files /dev/null and b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/20230605_005004/vis_data/vis_image/0000000.png_3.png differ diff --git a/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/20230605_005004/vis_data/vis_image/0000000.png_4.png b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/20230605_005004/vis_data/vis_image/0000000.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..447f8e695cb35d0a69dc1f2642eaeb8e6282c2e3 Binary files /dev/null and b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/20230605_005004/vis_data/vis_image/0000000.png_4.png differ diff --git a/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/20230605_005004/vis_data/vis_image/0000000.png_5.png b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/20230605_005004/vis_data/vis_image/0000000.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..b389ba61f38e3bc1931f519b28ec0142e17c2d22 Binary files /dev/null and b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/20230605_005004/vis_data/vis_image/0000000.png_5.png differ diff --git a/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/20230605_005004/vis_data/vis_image/0000000.png_6.png b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/20230605_005004/vis_data/vis_image/0000000.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..6a59882454f8f450b3486a6fd22127d7af24a280 Binary files /dev/null and b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/20230605_005004/vis_data/vis_image/0000000.png_6.png differ diff --git a/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/20230605_005004/vis_data/vis_image/0000000.png_7.png b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/20230605_005004/vis_data/vis_image/0000000.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..13e5fec23c8ed8e4ccaf5236283bc556d9cafbf8 Binary files /dev/null and b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/20230605_005004/vis_data/vis_image/0000000.png_7.png differ diff --git a/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/20230605_005004/vis_data/vis_image/0000000.png_8.png b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/20230605_005004/vis_data/vis_image/0000000.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..4ea1e8e80a6007997a12331a8d1ee6ddf2d80f3e Binary files /dev/null and b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/20230605_005004/vis_data/vis_image/0000000.png_8.png differ diff --git a/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/20230605_005004/vis_data/vis_image/0000000.png_9.png b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/20230605_005004/vis_data/vis_image/0000000.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..8c01636ba7e36ce729aff52cec83a7807e30629f Binary files /dev/null and b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/20230605_005004/vis_data/vis_image/0000000.png_9.png differ diff --git a/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/20230605_005004/vis_data/vis_image/000005174.jpg_1.png b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/20230605_005004/vis_data/vis_image/000005174.jpg_1.png new file mode 100644 index 0000000000000000000000000000000000000000..f8ed54368a9864d452e07c579084382a5d38fa05 Binary files /dev/null and b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/20230605_005004/vis_data/vis_image/000005174.jpg_1.png differ diff --git a/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/20230605_005004/vis_data/vis_image/000005174.jpg_10.png b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/20230605_005004/vis_data/vis_image/000005174.jpg_10.png new file mode 100644 index 0000000000000000000000000000000000000000..6a5c4442c95e8bb24b5bd2b284aea56c916dab87 Binary files /dev/null and b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/20230605_005004/vis_data/vis_image/000005174.jpg_10.png differ diff --git a/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/20230605_005004/vis_data/vis_image/000005174.jpg_2.png b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/20230605_005004/vis_data/vis_image/000005174.jpg_2.png new file mode 100644 index 0000000000000000000000000000000000000000..59aea93e8943edcfc6b37ca100df056dd67eab9d Binary files /dev/null and b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/20230605_005004/vis_data/vis_image/000005174.jpg_2.png differ diff --git a/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/20230605_005004/vis_data/vis_image/000005174.jpg_3.png b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/20230605_005004/vis_data/vis_image/000005174.jpg_3.png new file mode 100644 index 0000000000000000000000000000000000000000..49958fc271b9726b3e4542c3b0c1543f8672a7b0 Binary files /dev/null and b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/20230605_005004/vis_data/vis_image/000005174.jpg_3.png differ diff --git a/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/20230605_005004/vis_data/vis_image/000005174.jpg_4.png b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/20230605_005004/vis_data/vis_image/000005174.jpg_4.png new file mode 100644 index 0000000000000000000000000000000000000000..d426dd7192736751513e574b511855a54702f4d5 Binary files /dev/null and b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/20230605_005004/vis_data/vis_image/000005174.jpg_4.png differ diff --git a/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/20230605_005004/vis_data/vis_image/000005174.jpg_5.png b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/20230605_005004/vis_data/vis_image/000005174.jpg_5.png new file mode 100644 index 0000000000000000000000000000000000000000..aae68ce160d3daaedddb4e344d488662527267ef Binary files /dev/null and b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/20230605_005004/vis_data/vis_image/000005174.jpg_5.png differ diff --git a/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/20230605_005004/vis_data/vis_image/000005174.jpg_6.png b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/20230605_005004/vis_data/vis_image/000005174.jpg_6.png new file mode 100644 index 0000000000000000000000000000000000000000..d31b6a7803550d13b8a6e6079a2e5dbe594f543f Binary files /dev/null and b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/20230605_005004/vis_data/vis_image/000005174.jpg_6.png differ diff --git a/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/20230605_005004/vis_data/vis_image/000005174.jpg_7.png b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/20230605_005004/vis_data/vis_image/000005174.jpg_7.png new file mode 100644 index 0000000000000000000000000000000000000000..c5ea544a917758786d968d8cc9a88289c6c7b2a5 Binary files /dev/null and b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/20230605_005004/vis_data/vis_image/000005174.jpg_7.png differ diff --git a/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/20230605_005004/vis_data/vis_image/000005174.jpg_8.png b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/20230605_005004/vis_data/vis_image/000005174.jpg_8.png new file mode 100644 index 0000000000000000000000000000000000000000..0a2617d3da347405c0165fb54afd5078cec888b1 Binary files /dev/null and b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/20230605_005004/vis_data/vis_image/000005174.jpg_8.png differ diff --git a/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/20230605_005004/vis_data/vis_image/000005174.jpg_9.png b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/20230605_005004/vis_data/vis_image/000005174.jpg_9.png new file mode 100644 index 0000000000000000000000000000000000000000..a42bc7df8fecf64156609030f9ace87250373b61 Binary files /dev/null and b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/20230605_005004/vis_data/vis_image/000005174.jpg_9.png differ diff --git a/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/20230605_005004/vis_data/vis_image/0010000.png_1.png b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/20230605_005004/vis_data/vis_image/0010000.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..285c8dcddc179cb2c03e12f056d902636ba96a65 Binary files /dev/null and b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/20230605_005004/vis_data/vis_image/0010000.png_1.png differ diff --git a/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/20230605_005004/vis_data/vis_image/0010000.png_10.png b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/20230605_005004/vis_data/vis_image/0010000.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..9061e26e9c867671a82a56d1d8b6e8740675e844 Binary files /dev/null and b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/20230605_005004/vis_data/vis_image/0010000.png_10.png differ diff --git a/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/20230605_005004/vis_data/vis_image/0010000.png_2.png b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/20230605_005004/vis_data/vis_image/0010000.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..9f63bdd92fb9e6abc71358b9f1e3195125c016b1 Binary files /dev/null and b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/20230605_005004/vis_data/vis_image/0010000.png_2.png differ diff --git a/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/20230605_005004/vis_data/vis_image/0010000.png_3.png b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/20230605_005004/vis_data/vis_image/0010000.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..edfa922e5508ba7597c8c33fa6820536a5d91214 Binary files /dev/null and b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/20230605_005004/vis_data/vis_image/0010000.png_3.png differ diff --git a/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/20230605_005004/vis_data/vis_image/0010000.png_4.png b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/20230605_005004/vis_data/vis_image/0010000.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..fa02779967c0587d40a98d8ec92c4037e4f66d02 Binary files /dev/null and b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/20230605_005004/vis_data/vis_image/0010000.png_4.png differ diff --git a/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/20230605_005004/vis_data/vis_image/0010000.png_5.png b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/20230605_005004/vis_data/vis_image/0010000.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..143c00aa960e073d075b0345e1776da2704252f5 Binary files /dev/null and b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/20230605_005004/vis_data/vis_image/0010000.png_5.png differ diff --git a/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/20230605_005004/vis_data/vis_image/0010000.png_6.png b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/20230605_005004/vis_data/vis_image/0010000.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..70b52b5c82daaeaca44c03403963611d295a74ad Binary files /dev/null and b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/20230605_005004/vis_data/vis_image/0010000.png_6.png differ diff --git a/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/20230605_005004/vis_data/vis_image/0010000.png_7.png b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/20230605_005004/vis_data/vis_image/0010000.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..2bfe03f3d4236c3a8d58925d91773b8a8a8d9599 Binary files /dev/null and b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/20230605_005004/vis_data/vis_image/0010000.png_7.png differ diff --git a/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/20230605_005004/vis_data/vis_image/0010000.png_8.png b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/20230605_005004/vis_data/vis_image/0010000.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..7a53ba53c091f2377a00563d2976a7326394d22b Binary files /dev/null and b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/20230605_005004/vis_data/vis_image/0010000.png_8.png differ diff --git a/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/20230605_005004/vis_data/vis_image/0010000.png_9.png b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/20230605_005004/vis_data/vis_image/0010000.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..b86a4db6cac9bb5c1063b91bf868a22c76e25145 Binary files /dev/null and b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/20230605_005004/vis_data/vis_image/0010000.png_9.png differ diff --git a/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/epoch_1.pth b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/epoch_1.pth new file mode 100644 index 0000000000000000000000000000000000000000..b892ffe1c69619d90bba01664d8b589f438dcc97 --- /dev/null +++ b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/epoch_1.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:71a7c7e5cc7217386f790b3ec274ab0f8b19a1326476c8b6010a6a066bbac3c1 +size 277331341 diff --git a/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/epoch_10.pth b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/epoch_10.pth new file mode 100644 index 0000000000000000000000000000000000000000..12bfa99c49fb835a54b49d887cdb2d0eaccde3f7 --- /dev/null +++ b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/epoch_10.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:7a34d2b4952b89590ce6bb3ecd6be4342b5c5ed75e854e7ba24f2caacc528843 +size 279903309 diff --git a/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/epoch_2.pth b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/epoch_2.pth new file mode 100644 index 0000000000000000000000000000000000000000..72af493d6029ea3cc2f2da6b42952d4ad029c86d --- /dev/null +++ b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/epoch_2.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:8ccc600b39964bd585c0a87ae1533eb05d9ad67ae3e065e52298b9b96ba4770a +size 277616205 diff --git a/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/epoch_3.pth b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/epoch_3.pth new file mode 100644 index 0000000000000000000000000000000000000000..004cbbb6eefa7a64b3ea57b1499ab4788ccec3cd --- /dev/null +++ b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/epoch_3.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:7f49038c91d0734684857a9013054e4dd6d483988470d9d6cae483a12329197a +size 277900941 diff --git a/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/epoch_4.pth b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/epoch_4.pth new file mode 100644 index 0000000000000000000000000000000000000000..f98bee8a85c34d03a59b46d70fbda6468672139d --- /dev/null +++ b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/epoch_4.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:f5935adcf7d0cb5d777c3fc425ef731ba6882de8bb8b0290f708c169ad0270be +size 278185229 diff --git a/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/epoch_5.pth b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/epoch_5.pth new file mode 100644 index 0000000000000000000000000000000000000000..b05c5e1b34371cb8f5aac1414af3b7e447bd8fcd --- /dev/null +++ b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/epoch_5.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ebc127f037cc04d3a4af495040198aa24f0799397c242198da276f30755596fc +size 278469453 diff --git a/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/epoch_6.pth b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/epoch_6.pth new file mode 100644 index 0000000000000000000000000000000000000000..f482e489d1771c6bbf1dbd766cee2bd58dbee8b8 --- /dev/null +++ b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/epoch_6.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:018957bdc295f9b7cd09ff6d11a32881e1a58a046738528a28af547a325e07a0 +size 278753613 diff --git a/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/epoch_7.pth b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/epoch_7.pth new file mode 100644 index 0000000000000000000000000000000000000000..ea4dd7e1fa786acd14d6b7241d5575b2ddd32b9f --- /dev/null +++ b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/epoch_7.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:4066310443714f4c2c371a2e7ecd1056dc5dac7f13da9d490d7056ac34023bba +size 279038349 diff --git a/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/epoch_8.pth b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/epoch_8.pth new file mode 100644 index 0000000000000000000000000000000000000000..2913d531c6618ecb37365c283ee3762eb2ff92f6 --- /dev/null +++ b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/epoch_8.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:beb451096568bc94fc48e55409f1daf7e7878563935b04cc8b0943925e5329f9 +size 279327053 diff --git a/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/epoch_9.pth b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/epoch_9.pth new file mode 100644 index 0000000000000000000000000000000000000000..10b9c5fb4a5c64da2c0afa62a7c0a2b1c8b7f7bd --- /dev/null +++ b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/epoch_9.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:358686ad5f298e7b2d25a4097cfe8d5aa3a6a5c796ba4ce2926decda16453c26 +size 279615373 diff --git a/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/last_checkpoint b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/last_checkpoint new file mode 100644 index 0000000000000000000000000000000000000000..bd7b0e68a4d6fb7725edb722ab6e8872e12bfd90 --- /dev/null +++ b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/last_checkpoint @@ -0,0 +1 @@ +/mnt/petrelfs/luzeyu/workspace/fakebench/mmpretrain/workdir/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/epoch_10.pth \ No newline at end of file diff --git a/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1.py b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1.py new file mode 100644 index 0000000000000000000000000000000000000000..661bf63adae66645caa3af55d05b1778159f197c --- /dev/null +++ b/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1.py @@ -0,0 +1,230 @@ +optim_wrapper = dict( + optimizer=dict( + type='SGD', + lr=0.0001, + momentum=0.9, + weight_decay=0.0001, + _scope_='mmpretrain'), + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=False, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=512) +model = dict( + type='ImageClassifier', + backbone=dict( + frozen_stages=2, + type='ResNet', + depth=50, + num_stages=4, + out_indices=(3, ), + style='pytorch', + init_cfg=dict( + type='Pretrained', + checkpoint= + 'https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth', + prefix='backbone')), + neck=dict(type='GlobalAveragePooling'), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=2048, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + topk=1)) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stablediffusionV1-5R2-dpmsolver-25-1m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/cc1m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = dict(type='Accuracy', topk=1) +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = dict(type='Accuracy', topk=1) +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/resnet50_2xb256_sdv1_1m_lr1e-4_aug_5e-1' diff --git a/resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1/20230604_225317/20230604_225317.log b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1/20230604_225317/20230604_225317.log new file mode 100644 index 0000000000000000000000000000000000000000..70f75a745ad06fcb631bca6ba75f260b062b1622 --- /dev/null +++ b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1/20230604_225317/20230604_225317.log @@ -0,0 +1,1305 @@ +2023/06/04 22:53:21 - mmengine - INFO - +------------------------------------------------------------ +System environment: + sys.platform: linux + Python: 3.10.9 (main, Mar 8 2023, 10:47:38) [GCC 11.2.0] + CUDA available: True + numpy_random_seed: 708732946 + GPU 0,1: NVIDIA A100-SXM4-80GB + CUDA_HOME: /mnt/petrelfs/share/cuda-11.6 + NVCC: Cuda compilation tools, release 11.6, V11.6.124 + GCC: gcc (GCC) 7.5.0 + PyTorch: 1.13.1 + PyTorch compiling details: PyTorch built with: + - GCC 9.3 + - C++ Version: 201402 + - Intel(R) oneAPI Math Kernel Library Version 2021.4-Product Build 20210904 for Intel(R) 64 architecture applications + - Intel(R) MKL-DNN v2.6.0 (Git Hash 52b5f107dd9cf10910aaa19cb47f3abf9b349815) + - OpenMP 201511 (a.k.a. OpenMP 4.5) + - LAPACK is enabled (usually provided by MKL) + - NNPACK is enabled + - CPU capability usage: AVX2 + - CUDA Runtime 11.6 + - NVCC architecture flags: -gencode;arch=compute_37,code=sm_37;-gencode;arch=compute_50,code=sm_50;-gencode;arch=compute_60,code=sm_60;-gencode;arch=compute_61,code=sm_61;-gencode;arch=compute_70,code=sm_70;-gencode;arch=compute_75,code=sm_75;-gencode;arch=compute_80,code=sm_80;-gencode;arch=compute_86,code=sm_86;-gencode;arch=compute_37,code=compute_37 + - CuDNN 8.3.2 (built against CUDA 11.5) + - Magma 2.6.1 + - Build settings: BLAS_INFO=mkl, BUILD_TYPE=Release, CUDA_VERSION=11.6, CUDNN_VERSION=8.3.2, CXX_COMPILER=/opt/rh/devtoolset-9/root/usr/bin/c++, CXX_FLAGS= -fabi-version=11 -Wno-deprecated -fvisibility-inlines-hidden -DUSE_PTHREADPOOL -fopenmp -DNDEBUG -DUSE_KINETO -DUSE_FBGEMM -DUSE_QNNPACK -DUSE_PYTORCH_QNNPACK -DUSE_XNNPACK -DSYMBOLICATE_MOBILE_DEBUG_HANDLE -DEDGE_PROFILER_USE_KINETO -O2 -fPIC -Wno-narrowing -Wall -Wextra -Werror=return-type -Werror=non-virtual-dtor -Wno-missing-field-initializers -Wno-type-limits -Wno-array-bounds -Wno-unknown-pragmas -Wunused-local-typedefs -Wno-unused-parameter -Wno-unused-function -Wno-unused-result -Wno-strict-overflow -Wno-strict-aliasing -Wno-error=deprecated-declarations -Wno-stringop-overflow -Wno-psabi -Wno-error=pedantic -Wno-error=redundant-decls -Wno-error=old-style-cast -fdiagnostics-color=always -faligned-new -Wno-unused-but-set-variable -Wno-maybe-uninitialized -fno-math-errno -fno-trapping-math -Werror=format -Werror=cast-function-type -Wno-stringop-overflow, LAPACK_INFO=mkl, PERF_WITH_AVX=1, PERF_WITH_AVX2=1, PERF_WITH_AVX512=1, TORCH_VERSION=1.13.1, USE_CUDA=ON, USE_CUDNN=ON, USE_EXCEPTION_PTR=1, USE_GFLAGS=OFF, USE_GLOG=OFF, USE_MKL=ON, USE_MKLDNN=ON, USE_MPI=OFF, USE_NCCL=ON, USE_NNPACK=ON, USE_OPENMP=ON, USE_ROCM=OFF, + + TorchVision: 0.14.1 + OpenCV: 4.7.0 + MMEngine: 0.7.3 + +Runtime environment: + cudnn_benchmark: True + mp_cfg: {'mp_start_method': 'fork', 'opencv_num_threads': 0} + dist_cfg: {'backend': 'nccl'} + seed: None + deterministic: False + Distributed launcher: slurm + Distributed training: True + GPU number: 2 +------------------------------------------------------------ + +2023/06/04 22:53:26 - mmengine - INFO - Config: +optim_wrapper = dict( + optimizer=dict( + type='SGD', + lr=0.0001, + momentum=0.9, + weight_decay=0.0001, + _scope_='mmpretrain'), + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=False, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=512) +model = dict( + type='ImageClassifier', + backbone=dict( + frozen_stages=2, + type='ResNet', + depth=50, + num_stages=4, + out_indices=(3, ), + style='pytorch', + init_cfg=dict( + type='Pretrained', + checkpoint= + 'https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth', + prefix='backbone')), + neck=dict(type='GlobalAveragePooling'), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=2048, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + topk=1)) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stablediffusionV2-1-dpmsolver-25-1m.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/cc1m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV2-1-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = dict(type='Accuracy', topk=1) +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV2-1-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = dict(type='Accuracy', topk=1) +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1' + +2023/06/04 22:53:39 - mmengine - INFO - Hooks will be executed in the following order: +before_run: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_load_checkpoint: +(ABOVE_NORMAL) EMAHook + -------------------- +before_train: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_train_epoch: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(NORMAL ) DistSamplerSeedHook + -------------------- +before_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook + -------------------- +after_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +after_train_epoch: +(NORMAL ) IterTimerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_val_epoch: +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook + -------------------- +before_val_iter: +(NORMAL ) IterTimerHook + -------------------- +after_val_iter: +(NORMAL ) IterTimerHook +(NORMAL ) VisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_val_epoch: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_save_checkpoint: +(ABOVE_NORMAL) EMAHook + -------------------- +after_train: +(VERY_LOW ) CheckpointHook + -------------------- +before_test_epoch: +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook + -------------------- +before_test_iter: +(NORMAL ) IterTimerHook + -------------------- +after_test_iter: +(NORMAL ) IterTimerHook +(NORMAL ) VisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_test_epoch: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_run: +(BELOW_NORMAL) LoggerHook + -------------------- +2023/06/04 22:54:01 - mmengine - INFO - load backbone in model from: https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth +Name of parameter - Initialization information + +backbone.conv1.weight - torch.Size([64, 3, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.bn1.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.bn1.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.conv1.weight - torch.Size([64, 64, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.bn1.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.bn1.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.conv2.weight - torch.Size([64, 64, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.bn2.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.bn2.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.conv3.weight - torch.Size([256, 64, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.bn3.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.bn3.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.downsample.0.weight - torch.Size([256, 64, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.downsample.1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.downsample.1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.conv1.weight - torch.Size([64, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.bn1.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.bn1.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.conv2.weight - torch.Size([64, 64, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.bn2.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.bn2.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.conv3.weight - torch.Size([256, 64, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.bn3.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.bn3.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.conv1.weight - torch.Size([64, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.bn1.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.bn1.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.conv2.weight - torch.Size([64, 64, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.bn2.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.bn2.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.conv3.weight - torch.Size([256, 64, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.bn3.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.bn3.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.conv1.weight - torch.Size([128, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.bn1.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.bn1.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.conv2.weight - torch.Size([128, 128, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.bn2.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.bn2.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.conv3.weight - torch.Size([512, 128, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.bn3.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.bn3.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.downsample.0.weight - torch.Size([512, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.downsample.1.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.downsample.1.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.conv1.weight - torch.Size([128, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.bn1.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.bn1.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.conv2.weight - torch.Size([128, 128, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.bn2.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.bn2.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.conv3.weight - torch.Size([512, 128, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.bn3.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.bn3.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.conv1.weight - torch.Size([128, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.bn1.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.bn1.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.conv2.weight - torch.Size([128, 128, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.bn2.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.bn2.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.conv3.weight - torch.Size([512, 128, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.bn3.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.bn3.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.conv1.weight - torch.Size([128, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.bn1.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.bn1.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.conv2.weight - torch.Size([128, 128, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.bn2.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.bn2.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.conv3.weight - torch.Size([512, 128, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.bn3.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.bn3.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.conv1.weight - torch.Size([256, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.bn1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.bn1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.conv2.weight - torch.Size([256, 256, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.bn2.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.bn2.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.conv3.weight - torch.Size([1024, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.bn3.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.bn3.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.downsample.0.weight - torch.Size([1024, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.downsample.1.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.downsample.1.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.conv1.weight - torch.Size([256, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.bn1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.bn1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.conv2.weight - torch.Size([256, 256, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.bn2.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.bn2.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.conv3.weight - torch.Size([1024, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.bn3.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.bn3.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.conv1.weight - torch.Size([256, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.bn1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.bn1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.conv2.weight - torch.Size([256, 256, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.bn2.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.bn2.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.conv3.weight - torch.Size([1024, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.bn3.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.bn3.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.conv1.weight - torch.Size([256, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.bn1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.bn1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.conv2.weight - torch.Size([256, 256, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.bn2.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.bn2.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.conv3.weight - torch.Size([1024, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.bn3.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.bn3.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.conv1.weight - torch.Size([256, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.bn1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.bn1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.conv2.weight - torch.Size([256, 256, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.bn2.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.bn2.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.conv3.weight - torch.Size([1024, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.bn3.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.bn3.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.conv1.weight - torch.Size([256, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.bn1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.bn1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.conv2.weight - torch.Size([256, 256, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.bn2.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.bn2.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.conv3.weight - torch.Size([1024, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.bn3.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.bn3.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.conv1.weight - torch.Size([512, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.bn1.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.bn1.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.conv2.weight - torch.Size([512, 512, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.bn2.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.bn2.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.conv3.weight - torch.Size([2048, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.bn3.weight - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.bn3.bias - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.downsample.0.weight - torch.Size([2048, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.downsample.1.weight - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.downsample.1.bias - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.conv1.weight - torch.Size([512, 2048, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.bn1.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.bn1.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.conv2.weight - torch.Size([512, 512, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.bn2.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.bn2.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.conv3.weight - torch.Size([2048, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.bn3.weight - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.bn3.bias - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.conv1.weight - torch.Size([512, 2048, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.bn1.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.bn1.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.conv2.weight - torch.Size([512, 512, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.bn2.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.bn2.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.conv3.weight - torch.Size([2048, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.bn3.weight - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.bn3.bias - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +head.fc.weight - torch.Size([2, 2048]): +NormalInit: mean=0, std=0.01, bias=0 + +head.fc.bias - torch.Size([2]): +NormalInit: mean=0, std=0.01, bias=0 +2023/06/04 22:54:01 - mmengine - WARNING - "FileClient" will be deprecated in future. Please use io functions in https://mmengine.readthedocs.io/en/latest/api/fileio.html#file-io +2023/06/04 22:54:01 - mmengine - WARNING - "HardDiskBackend" is the alias of "LocalBackend" and the former will be deprecated in future. +2023/06/04 22:54:01 - mmengine - INFO - Checkpoints will be saved to /mnt/petrelfs/luzeyu/workspace/fakebench/mmpretrain/workdir/resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1. +2023/06/04 22:55:09 - mmengine - INFO - Epoch(train) [1][ 100/3907] lr: 9.9999e-05 eta: 7:20:08 time: 0.5942 data_time: 0.0433 memory: 9436 loss: 0.6144 +2023/06/04 22:56:11 - mmengine - INFO - Epoch(train) [1][ 200/3907] lr: 9.9994e-05 eta: 7:01:05 time: 0.6243 data_time: 0.0018 memory: 6319 loss: 0.5405 +2023/06/04 22:57:14 - mmengine - INFO - Epoch(train) [1][ 300/3907] lr: 9.9987e-05 eta: 6:55:18 time: 0.6351 data_time: 0.0014 memory: 6319 loss: 0.4861 +2023/06/04 22:58:16 - mmengine - INFO - Epoch(train) [1][ 400/3907] lr: 9.9977e-05 eta: 6:51:10 time: 0.5895 data_time: 0.0015 memory: 6319 loss: 0.4412 +2023/06/04 22:59:20 - mmengine - INFO - Epoch(train) [1][ 500/3907] lr: 9.9964e-05 eta: 6:50:10 time: 0.5980 data_time: 0.0015 memory: 6319 loss: 0.3898 +2023/06/04 23:00:22 - mmengine - INFO - Epoch(train) [1][ 600/3907] lr: 9.9948e-05 eta: 6:47:16 time: 0.6539 data_time: 0.0011 memory: 6319 loss: 0.3696 +2023/06/04 23:01:25 - mmengine - INFO - Epoch(train) [1][ 700/3907] lr: 9.9929e-05 eta: 6:45:31 time: 0.6491 data_time: 0.0011 memory: 6319 loss: 0.3306 +2023/06/04 23:02:27 - mmengine - INFO - Epoch(train) [1][ 800/3907] lr: 9.9907e-05 eta: 6:43:21 time: 0.5923 data_time: 0.0016 memory: 6319 loss: 0.3094 +2023/06/04 23:03:30 - mmengine - INFO - Epoch(train) [1][ 900/3907] lr: 9.9882e-05 eta: 6:42:21 time: 0.6268 data_time: 0.0008 memory: 6319 loss: 0.2967 +2023/06/04 23:04:32 - mmengine - INFO - Exp name: resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1_20230604_225317 +2023/06/04 23:04:32 - mmengine - INFO - Epoch(train) [1][1000/3907] lr: 9.9855e-05 eta: 6:40:26 time: 0.6295 data_time: 0.0011 memory: 6319 loss: 0.2710 +2023/06/04 23:05:35 - mmengine - INFO - Epoch(train) [1][1100/3907] lr: 9.9824e-05 eta: 6:39:12 time: 0.6169 data_time: 0.0008 memory: 6319 loss: 0.2720 +2023/06/04 23:06:37 - mmengine - INFO - Epoch(train) [1][1200/3907] lr: 9.9791e-05 eta: 6:37:38 time: 0.6338 data_time: 0.0009 memory: 6319 loss: 0.2541 +2023/06/04 23:07:42 - mmengine - INFO - Epoch(train) [1][1300/3907] lr: 9.9755e-05 eta: 6:37:27 time: 0.7134 data_time: 0.0011 memory: 6319 loss: 0.2377 +2023/06/04 23:08:42 - mmengine - INFO - Epoch(train) [1][1400/3907] lr: 9.9716e-05 eta: 6:35:17 time: 0.5884 data_time: 0.0012 memory: 6319 loss: 0.2348 +2023/06/04 23:09:46 - mmengine - INFO - Epoch(train) [1][1500/3907] lr: 9.9674e-05 eta: 6:34:23 time: 0.6164 data_time: 0.0010 memory: 6319 loss: 0.2446 +2023/06/04 23:10:48 - mmengine - INFO - Epoch(train) [1][1600/3907] lr: 9.9629e-05 eta: 6:33:00 time: 0.6497 data_time: 0.0011 memory: 6319 loss: 0.2131 +2023/06/04 23:11:51 - mmengine - INFO - Epoch(train) [1][1700/3907] lr: 9.9581e-05 eta: 6:31:54 time: 0.6395 data_time: 0.0015 memory: 6319 loss: 0.2084 +2023/06/04 23:12:54 - mmengine - INFO - Epoch(train) [1][1800/3907] lr: 9.9530e-05 eta: 6:30:53 time: 0.6245 data_time: 0.0013 memory: 6319 loss: 0.2089 +2023/06/04 23:13:56 - mmengine - INFO - Epoch(train) [1][1900/3907] lr: 9.9476e-05 eta: 6:29:39 time: 0.6109 data_time: 0.0010 memory: 6319 loss: 0.2113 +2023/06/04 23:14:58 - mmengine - INFO - Exp name: resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1_20230604_225317 +2023/06/04 23:14:58 - mmengine - INFO - Epoch(train) [1][2000/3907] lr: 9.9420e-05 eta: 6:28:14 time: 0.6053 data_time: 0.0011 memory: 6319 loss: 0.2004 +2023/06/04 23:15:59 - mmengine - INFO - Epoch(train) [1][2100/3907] lr: 9.9361e-05 eta: 6:26:48 time: 0.6134 data_time: 0.0008 memory: 6319 loss: 0.1954 +2023/06/04 23:17:02 - mmengine - INFO - Epoch(train) [1][2200/3907] lr: 9.9298e-05 eta: 6:25:47 time: 0.6748 data_time: 0.0012 memory: 6319 loss: 0.1850 +2023/06/04 23:18:04 - mmengine - INFO - Epoch(train) [1][2300/3907] lr: 9.9233e-05 eta: 6:24:22 time: 0.6820 data_time: 0.0011 memory: 6319 loss: 0.1764 +2023/06/04 23:19:10 - mmengine - INFO - Epoch(train) [1][2400/3907] lr: 9.9165e-05 eta: 6:24:11 time: 0.6188 data_time: 0.0010 memory: 6319 loss: 0.1709 +2023/06/04 23:20:10 - mmengine - INFO - Epoch(train) [1][2500/3907] lr: 9.9095e-05 eta: 6:22:34 time: 0.6502 data_time: 0.0013 memory: 6319 loss: 0.1771 +2023/06/04 23:21:14 - mmengine - INFO - Epoch(train) [1][2600/3907] lr: 9.9021e-05 eta: 6:21:42 time: 0.6513 data_time: 0.0010 memory: 6319 loss: 0.1946 +2023/06/04 23:22:19 - mmengine - INFO - Epoch(train) [1][2700/3907] lr: 9.8944e-05 eta: 6:21:11 time: 0.6358 data_time: 0.0013 memory: 6319 loss: 0.1769 +2023/06/04 23:23:23 - mmengine - INFO - Epoch(train) [1][2800/3907] lr: 9.8865e-05 eta: 6:20:28 time: 0.6681 data_time: 0.0010 memory: 6319 loss: 0.1762 +2023/06/04 23:24:26 - mmengine - INFO - Epoch(train) [1][2900/3907] lr: 9.8783e-05 eta: 6:19:19 time: 0.6001 data_time: 0.0009 memory: 6319 loss: 0.1681 +2023/06/04 23:25:28 - mmengine - INFO - Exp name: resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1_20230604_225317 +2023/06/04 23:25:28 - mmengine - INFO - Epoch(train) [1][3000/3907] lr: 9.8698e-05 eta: 6:18:03 time: 0.6313 data_time: 0.0015 memory: 6319 loss: 0.1718 +2023/06/04 23:26:29 - mmengine - INFO - Epoch(train) [1][3100/3907] lr: 9.8610e-05 eta: 6:16:48 time: 0.6330 data_time: 0.0008 memory: 6319 loss: 0.1642 +2023/06/04 23:27:32 - mmengine - INFO - Epoch(train) [1][3200/3907] lr: 9.8519e-05 eta: 6:15:37 time: 0.6067 data_time: 0.0012 memory: 6319 loss: 0.1780 +2023/06/04 23:28:32 - mmengine - INFO - Epoch(train) [1][3300/3907] lr: 9.8426e-05 eta: 6:14:09 time: 0.6164 data_time: 0.0011 memory: 6319 loss: 0.1483 +2023/06/04 23:29:34 - mmengine - INFO - Epoch(train) [1][3400/3907] lr: 9.8330e-05 eta: 6:12:54 time: 0.6566 data_time: 0.0012 memory: 6319 loss: 0.1571 +2023/06/04 23:30:37 - mmengine - INFO - Epoch(train) [1][3500/3907] lr: 9.8231e-05 eta: 6:11:51 time: 0.5866 data_time: 0.0014 memory: 6319 loss: 0.1466 +2023/06/04 23:31:37 - mmengine - INFO - Epoch(train) [1][3600/3907] lr: 9.8129e-05 eta: 6:10:25 time: 0.6107 data_time: 0.0011 memory: 6319 loss: 0.1623 +2023/06/04 23:32:39 - mmengine - INFO - Epoch(train) [1][3700/3907] lr: 9.8024e-05 eta: 6:09:17 time: 0.6640 data_time: 0.0010 memory: 6319 loss: 0.1488 +2023/06/04 23:33:40 - mmengine - INFO - Epoch(train) [1][3800/3907] lr: 9.7917e-05 eta: 6:07:56 time: 0.6098 data_time: 0.0012 memory: 6319 loss: 0.1414 +2023/06/04 23:34:43 - mmengine - INFO - Epoch(train) [1][3900/3907] lr: 9.7806e-05 eta: 6:06:57 time: 0.6231 data_time: 0.0009 memory: 6319 loss: 0.1574 +2023/06/04 23:34:48 - mmengine - INFO - Exp name: resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1_20230604_225317 +2023/06/04 23:34:48 - mmengine - INFO - Saving checkpoint at 1 epochs +2023/06/04 23:35:42 - mmengine - INFO - Epoch(val) [1][57/57] accuracy/top1: 80.4217 data_time: 0.6049 time: 0.6993 +2023/06/04 23:36:46 - mmengine - INFO - Exp name: resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1_20230604_225317 +2023/06/04 23:36:47 - mmengine - INFO - Epoch(train) [2][ 100/3907] lr: 9.7685e-05 eta: 6:06:21 time: 0.6566 data_time: 0.0013 memory: 6319 loss: 0.1497 +2023/06/04 23:37:49 - mmengine - INFO - Epoch(train) [2][ 200/3907] lr: 9.7570e-05 eta: 6:05:14 time: 0.6339 data_time: 0.0014 memory: 6319 loss: 0.1545 +2023/06/04 23:38:51 - mmengine - INFO - Epoch(train) [2][ 300/3907] lr: 9.7451e-05 eta: 6:04:06 time: 0.6071 data_time: 0.0011 memory: 6319 loss: 0.1540 +2023/06/04 23:39:55 - mmengine - INFO - Epoch(train) [2][ 400/3907] lr: 9.7329e-05 eta: 6:03:08 time: 0.6508 data_time: 0.0012 memory: 6319 loss: 0.1377 +2023/06/04 23:40:56 - mmengine - INFO - Epoch(train) [2][ 500/3907] lr: 9.7205e-05 eta: 6:01:55 time: 0.6095 data_time: 0.0011 memory: 6319 loss: 0.1483 +2023/06/04 23:41:57 - mmengine - INFO - Epoch(train) [2][ 600/3907] lr: 9.7078e-05 eta: 6:00:42 time: 0.6026 data_time: 0.0011 memory: 6319 loss: 0.1317 +2023/06/04 23:42:59 - mmengine - INFO - Epoch(train) [2][ 700/3907] lr: 9.6949e-05 eta: 5:59:31 time: 0.6099 data_time: 0.0010 memory: 6319 loss: 0.1532 +2023/06/04 23:44:01 - mmengine - INFO - Epoch(train) [2][ 800/3907] lr: 9.6816e-05 eta: 5:58:25 time: 0.6574 data_time: 0.0013 memory: 6319 loss: 0.1308 +2023/06/04 23:45:02 - mmengine - INFO - Epoch(train) [2][ 900/3907] lr: 9.6681e-05 eta: 5:57:09 time: 0.5588 data_time: 0.0012 memory: 6319 loss: 0.1337 +2023/06/04 23:46:02 - mmengine - INFO - Epoch(train) [2][1000/3907] lr: 9.6544e-05 eta: 5:55:51 time: 0.6176 data_time: 0.0013 memory: 6319 loss: 0.1303 +2023/06/04 23:47:01 - mmengine - INFO - Exp name: resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1_20230604_225317 +2023/06/04 23:47:04 - mmengine - INFO - Epoch(train) [2][1100/3907] lr: 9.6403e-05 eta: 5:54:42 time: 0.6701 data_time: 0.0013 memory: 6319 loss: 0.1229 +2023/06/04 23:48:04 - mmengine - INFO - Epoch(train) [2][1200/3907] lr: 9.6260e-05 eta: 5:53:23 time: 0.6172 data_time: 0.0011 memory: 6319 loss: 0.1405 +2023/06/04 23:49:06 - mmengine - INFO - Epoch(train) [2][1300/3907] lr: 9.6114e-05 eta: 5:52:22 time: 0.6054 data_time: 0.0016 memory: 6319 loss: 0.1391 +2023/06/04 23:50:08 - mmengine - INFO - Epoch(train) [2][1400/3907] lr: 9.5966e-05 eta: 5:51:14 time: 0.6270 data_time: 0.0010 memory: 6319 loss: 0.1204 +2023/06/04 23:51:08 - mmengine - INFO - Epoch(train) [2][1500/3907] lr: 9.5815e-05 eta: 5:49:59 time: 0.6327 data_time: 0.0011 memory: 6319 loss: 0.1474 +2023/06/04 23:52:11 - mmengine - INFO - Epoch(train) [2][1600/3907] lr: 9.5661e-05 eta: 5:49:02 time: 0.6522 data_time: 0.0015 memory: 6319 loss: 0.1289 +2023/06/04 23:53:15 - mmengine - INFO - Epoch(train) [2][1700/3907] lr: 9.5505e-05 eta: 5:48:04 time: 0.6179 data_time: 0.0013 memory: 6319 loss: 0.1204 +2023/06/04 23:54:17 - mmengine - INFO - Epoch(train) [2][1800/3907] lr: 9.5346e-05 eta: 5:47:02 time: 0.5668 data_time: 0.0013 memory: 6319 loss: 0.1174 +2023/06/04 23:55:20 - mmengine - INFO - Epoch(train) [2][1900/3907] lr: 9.5184e-05 eta: 5:46:04 time: 0.6009 data_time: 0.0011 memory: 6319 loss: 0.1267 +2023/06/04 23:56:23 - mmengine - INFO - Epoch(train) [2][2000/3907] lr: 9.5020e-05 eta: 5:45:01 time: 0.6914 data_time: 0.0011 memory: 6319 loss: 0.1327 +2023/06/04 23:57:23 - mmengine - INFO - Exp name: resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1_20230604_225317 +2023/06/04 23:57:24 - mmengine - INFO - Epoch(train) [2][2100/3907] lr: 9.4854e-05 eta: 5:43:55 time: 0.5970 data_time: 0.0009 memory: 6319 loss: 0.1313 +2023/06/04 23:58:26 - mmengine - INFO - Epoch(train) [2][2200/3907] lr: 9.4684e-05 eta: 5:42:51 time: 0.6438 data_time: 0.0010 memory: 6319 loss: 0.1189 +2023/06/04 23:59:28 - mmengine - INFO - Epoch(train) [2][2300/3907] lr: 9.4512e-05 eta: 5:41:46 time: 0.6048 data_time: 0.0012 memory: 6319 loss: 0.1218 +2023/06/05 00:00:30 - mmengine - INFO - Epoch(train) [2][2400/3907] lr: 9.4338e-05 eta: 5:40:38 time: 0.6326 data_time: 0.0009 memory: 6319 loss: 0.1213 +2023/06/05 00:01:30 - mmengine - INFO - Epoch(train) [2][2500/3907] lr: 9.4161e-05 eta: 5:39:27 time: 0.6049 data_time: 0.0009 memory: 6319 loss: 0.1269 +2023/06/05 00:02:33 - mmengine - INFO - Epoch(train) [2][2600/3907] lr: 9.3981e-05 eta: 5:38:28 time: 0.6421 data_time: 0.0010 memory: 6319 loss: 0.1067 +2023/06/05 00:03:36 - mmengine - INFO - Epoch(train) [2][2700/3907] lr: 9.3799e-05 eta: 5:37:28 time: 0.6008 data_time: 0.0010 memory: 6319 loss: 0.1204 +2023/06/05 00:04:37 - mmengine - INFO - Epoch(train) [2][2800/3907] lr: 9.3615e-05 eta: 5:36:19 time: 0.5981 data_time: 0.0241 memory: 6319 loss: 0.1317 +2023/06/05 00:05:37 - mmengine - INFO - Epoch(train) [2][2900/3907] lr: 9.3428e-05 eta: 5:35:06 time: 0.5673 data_time: 0.0010 memory: 6319 loss: 0.1162 +2023/06/05 00:06:42 - mmengine - INFO - Epoch(train) [2][3000/3907] lr: 9.3238e-05 eta: 5:34:12 time: 0.5922 data_time: 0.0013 memory: 6319 loss: 0.1247 +2023/06/05 00:07:42 - mmengine - INFO - Exp name: resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1_20230604_225317 +2023/06/05 00:07:43 - mmengine - INFO - Epoch(train) [2][3100/3907] lr: 9.3046e-05 eta: 5:33:06 time: 0.6147 data_time: 0.0011 memory: 6319 loss: 0.0999 +2023/06/05 00:08:44 - mmengine - INFO - Epoch(train) [2][3200/3907] lr: 9.2852e-05 eta: 5:31:58 time: 0.6372 data_time: 0.0016 memory: 6319 loss: 0.1218 +2023/06/05 00:09:48 - mmengine - INFO - Epoch(train) [2][3300/3907] lr: 9.2655e-05 eta: 5:31:01 time: 0.6388 data_time: 0.0013 memory: 6319 loss: 0.1103 +2023/06/05 00:10:51 - mmengine - INFO - Epoch(train) [2][3400/3907] lr: 9.2456e-05 eta: 5:30:02 time: 0.6173 data_time: 0.0012 memory: 6319 loss: 0.1180 +2023/06/05 00:11:52 - mmengine - INFO - Epoch(train) [2][3500/3907] lr: 9.2254e-05 eta: 5:28:56 time: 0.6374 data_time: 0.0011 memory: 6319 loss: 0.1124 +2023/06/05 00:12:55 - mmengine - INFO - Epoch(train) [2][3600/3907] lr: 9.2050e-05 eta: 5:27:54 time: 0.6131 data_time: 0.0011 memory: 6319 loss: 0.1274 +2023/06/05 00:13:58 - mmengine - INFO - Epoch(train) [2][3700/3907] lr: 9.1843e-05 eta: 5:26:54 time: 0.5851 data_time: 0.0014 memory: 6319 loss: 0.1237 +2023/06/05 00:15:00 - mmengine - INFO - Epoch(train) [2][3800/3907] lr: 9.1634e-05 eta: 5:25:53 time: 0.5574 data_time: 0.0009 memory: 6319 loss: 0.1099 +2023/06/05 00:15:59 - mmengine - INFO - Epoch(train) [2][3900/3907] lr: 9.1423e-05 eta: 5:24:34 time: 0.5945 data_time: 0.0009 memory: 6319 loss: 0.1304 +2023/06/05 00:16:04 - mmengine - INFO - Exp name: resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1_20230604_225317 +2023/06/05 00:16:04 - mmengine - INFO - Saving checkpoint at 2 epochs +2023/06/05 00:16:53 - mmengine - INFO - Epoch(val) [2][57/57] accuracy/top1: 78.6472 data_time: 0.5207 time: 0.6123 +2023/06/05 00:17:57 - mmengine - INFO - Epoch(train) [3][ 100/3907] lr: 9.1194e-05 eta: 5:23:40 time: 0.6440 data_time: 0.1597 memory: 6319 loss: 0.1029 +2023/06/05 00:18:53 - mmengine - INFO - Exp name: resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1_20230604_225317 +2023/06/05 00:18:59 - mmengine - INFO - Epoch(train) [3][ 200/3907] lr: 9.0978e-05 eta: 5:22:39 time: 0.5980 data_time: 0.0016 memory: 6319 loss: 0.1145 +2023/06/05 00:20:01 - mmengine - INFO - Epoch(train) [3][ 300/3907] lr: 9.0759e-05 eta: 5:21:34 time: 0.6024 data_time: 0.0014 memory: 6319 loss: 0.1096 +2023/06/05 00:21:02 - mmengine - INFO - Epoch(train) [3][ 400/3907] lr: 9.0539e-05 eta: 5:20:26 time: 0.5998 data_time: 0.0015 memory: 6319 loss: 0.1104 +2023/06/05 00:22:04 - mmengine - INFO - Epoch(train) [3][ 500/3907] lr: 9.0315e-05 eta: 5:19:21 time: 0.6282 data_time: 0.0012 memory: 6319 loss: 0.0986 +2023/06/05 00:23:05 - mmengine - INFO - Epoch(train) [3][ 600/3907] lr: 9.0090e-05 eta: 5:18:18 time: 0.6023 data_time: 0.0011 memory: 6319 loss: 0.1137 +2023/06/05 00:24:09 - mmengine - INFO - Epoch(train) [3][ 700/3907] lr: 8.9862e-05 eta: 5:17:19 time: 0.6255 data_time: 0.0017 memory: 6319 loss: 0.1075 +2023/06/05 00:25:11 - mmengine - INFO - Epoch(train) [3][ 800/3907] lr: 8.9632e-05 eta: 5:16:17 time: 0.5797 data_time: 0.0010 memory: 6319 loss: 0.1133 +2023/06/05 00:26:14 - mmengine - INFO - Epoch(train) [3][ 900/3907] lr: 8.9400e-05 eta: 5:15:16 time: 0.5906 data_time: 0.0012 memory: 6319 loss: 0.1130 +2023/06/05 00:27:14 - mmengine - INFO - Epoch(train) [3][1000/3907] lr: 8.9166e-05 eta: 5:14:07 time: 0.5974 data_time: 0.0011 memory: 6319 loss: 0.1090 +2023/06/05 00:28:17 - mmengine - INFO - Epoch(train) [3][1100/3907] lr: 8.8929e-05 eta: 5:13:06 time: 0.6059 data_time: 0.0010 memory: 6319 loss: 0.1151 +2023/06/05 00:29:11 - mmengine - INFO - Exp name: resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1_20230604_225317 +2023/06/05 00:29:20 - mmengine - INFO - Epoch(train) [3][1200/3907] lr: 8.8691e-05 eta: 5:12:05 time: 0.6163 data_time: 0.0012 memory: 6319 loss: 0.0966 +2023/06/05 00:30:21 - mmengine - INFO - Epoch(train) [3][1300/3907] lr: 8.8450e-05 eta: 5:11:00 time: 0.6097 data_time: 0.0010 memory: 6319 loss: 0.1111 +2023/06/05 00:31:25 - mmengine - INFO - Epoch(train) [3][1400/3907] lr: 8.8206e-05 eta: 5:10:01 time: 0.6459 data_time: 0.0012 memory: 6319 loss: 0.1011 +2023/06/05 00:32:31 - mmengine - INFO - Epoch(train) [3][1500/3907] lr: 8.7961e-05 eta: 5:09:12 time: 0.5573 data_time: 0.0010 memory: 6319 loss: 0.0956 +2023/06/05 00:33:30 - mmengine - INFO - Epoch(train) [3][1600/3907] lr: 8.7714e-05 eta: 5:07:59 time: 0.6570 data_time: 0.0016 memory: 6319 loss: 0.1045 +2023/06/05 00:34:31 - mmengine - INFO - Epoch(train) [3][1700/3907] lr: 8.7464e-05 eta: 5:06:52 time: 0.6292 data_time: 0.0018 memory: 6319 loss: 0.0916 +2023/06/05 00:35:33 - mmengine - INFO - Epoch(train) [3][1800/3907] lr: 8.7213e-05 eta: 5:05:49 time: 0.6127 data_time: 0.0012 memory: 6319 loss: 0.0995 +2023/06/05 00:36:34 - mmengine - INFO - Epoch(train) [3][1900/3907] lr: 8.6959e-05 eta: 5:04:42 time: 0.6106 data_time: 0.0013 memory: 6319 loss: 0.0963 +2023/06/05 00:37:35 - mmengine - INFO - Epoch(train) [3][2000/3907] lr: 8.6703e-05 eta: 5:03:38 time: 0.6230 data_time: 0.0013 memory: 6319 loss: 0.0992 +2023/06/05 00:38:36 - mmengine - INFO - Epoch(train) [3][2100/3907] lr: 8.6445e-05 eta: 5:02:32 time: 0.6017 data_time: 0.0014 memory: 6319 loss: 0.1029 +2023/06/05 00:39:30 - mmengine - INFO - Exp name: resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1_20230604_225317 +2023/06/05 00:39:37 - mmengine - INFO - Epoch(train) [3][2200/3907] lr: 8.6186e-05 eta: 5:01:26 time: 0.6403 data_time: 0.0012 memory: 6319 loss: 0.0951 +2023/06/05 00:40:40 - mmengine - INFO - Epoch(train) [3][2300/3907] lr: 8.5924e-05 eta: 5:00:26 time: 0.6266 data_time: 0.0012 memory: 6319 loss: 0.1064 +2023/06/05 00:41:49 - mmengine - INFO - Epoch(train) [3][2400/3907] lr: 8.5660e-05 eta: 4:59:43 time: 0.7791 data_time: 0.0016 memory: 6319 loss: 0.1105 +2023/06/05 00:42:54 - mmengine - INFO - Epoch(train) [3][2500/3907] lr: 8.5394e-05 eta: 4:58:48 time: 0.5468 data_time: 0.0011 memory: 6319 loss: 0.1047 +2023/06/05 00:43:53 - mmengine - INFO - Epoch(train) [3][2600/3907] lr: 8.5126e-05 eta: 4:57:35 time: 0.5611 data_time: 0.0012 memory: 6319 loss: 0.1030 +2023/06/05 00:44:52 - mmengine - INFO - Epoch(train) [3][2700/3907] lr: 8.4856e-05 eta: 4:56:23 time: 0.5413 data_time: 0.0012 memory: 6319 loss: 0.1051 +2023/06/05 00:45:50 - mmengine - INFO - Epoch(train) [3][2800/3907] lr: 8.4585e-05 eta: 4:55:10 time: 0.6266 data_time: 0.0011 memory: 6319 loss: 0.1068 +2023/06/05 00:46:50 - mmengine - INFO - Epoch(train) [3][2900/3907] lr: 8.4311e-05 eta: 4:54:02 time: 0.6155 data_time: 0.0010 memory: 6319 loss: 0.0897 +2023/06/05 00:47:51 - mmengine - INFO - Epoch(train) [3][3000/3907] lr: 8.4036e-05 eta: 4:52:57 time: 0.6194 data_time: 0.0011 memory: 6319 loss: 0.1030 +2023/06/05 00:48:52 - mmengine - INFO - Epoch(train) [3][3100/3907] lr: 8.3758e-05 eta: 4:51:51 time: 0.5691 data_time: 0.0012 memory: 6319 loss: 0.0950 +2023/06/05 00:49:49 - mmengine - INFO - Exp name: resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1_20230604_225317 +2023/06/05 00:49:55 - mmengine - INFO - Epoch(train) [3][3200/3907] lr: 8.3479e-05 eta: 4:50:52 time: 0.5626 data_time: 0.0012 memory: 6319 loss: 0.0938 +2023/06/05 00:50:56 - mmengine - INFO - Epoch(train) [3][3300/3907] lr: 8.3198e-05 eta: 4:49:46 time: 0.5902 data_time: 0.0011 memory: 6319 loss: 0.0888 +2023/06/05 00:51:59 - mmengine - INFO - Epoch(train) [3][3400/3907] lr: 8.2915e-05 eta: 4:48:45 time: 0.6066 data_time: 0.0014 memory: 6319 loss: 0.0924 +2023/06/05 00:53:17 - mmengine - INFO - Epoch(train) [3][3500/3907] lr: 8.2630e-05 eta: 4:48:24 time: 0.8397 data_time: 0.0009 memory: 6319 loss: 0.1119 +2023/06/05 00:54:32 - mmengine - INFO - Epoch(train) [3][3600/3907] lr: 8.2344e-05 eta: 4:47:50 time: 0.5562 data_time: 0.0009 memory: 6319 loss: 0.1063 +2023/06/05 00:55:30 - mmengine - INFO - Epoch(train) [3][3700/3907] lr: 8.2056e-05 eta: 4:46:39 time: 0.6229 data_time: 0.0013 memory: 6319 loss: 0.1074 +2023/06/05 00:56:30 - mmengine - INFO - Epoch(train) [3][3800/3907] lr: 8.1765e-05 eta: 4:45:30 time: 0.5701 data_time: 0.0010 memory: 6319 loss: 0.0977 +2023/06/05 00:57:35 - mmengine - INFO - Epoch(train) [3][3900/3907] lr: 8.1474e-05 eta: 4:44:33 time: 0.6194 data_time: 0.0011 memory: 6319 loss: 0.0957 +2023/06/05 00:57:39 - mmengine - INFO - Exp name: resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1_20230604_225317 +2023/06/05 00:57:39 - mmengine - INFO - Saving checkpoint at 3 epochs +2023/06/05 00:58:29 - mmengine - INFO - Epoch(val) [3][57/57] accuracy/top1: 82.1823 data_time: 0.5462 time: 0.6366 +2023/06/05 00:59:35 - mmengine - INFO - Epoch(train) [4][ 100/3907] lr: 8.1160e-05 eta: 4:43:34 time: 0.5875 data_time: 0.1213 memory: 6319 loss: 0.1053 +2023/06/05 01:00:38 - mmengine - INFO - Epoch(train) [4][ 200/3907] lr: 8.0864e-05 eta: 4:42:32 time: 0.6172 data_time: 0.2184 memory: 6319 loss: 0.0958 +2023/06/05 01:01:26 - mmengine - INFO - Exp name: resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1_20230604_225317 +2023/06/05 01:01:38 - mmengine - INFO - Epoch(train) [4][ 300/3907] lr: 8.0567e-05 eta: 4:41:26 time: 0.6325 data_time: 0.0015 memory: 6319 loss: 0.0914 +2023/06/05 01:02:42 - mmengine - INFO - Epoch(train) [4][ 400/3907] lr: 8.0269e-05 eta: 4:40:26 time: 0.5816 data_time: 0.0014 memory: 6319 loss: 0.1018 +2023/06/05 01:03:43 - mmengine - INFO - Epoch(train) [4][ 500/3907] lr: 7.9969e-05 eta: 4:39:21 time: 0.6351 data_time: 0.0012 memory: 6319 loss: 0.0961 +2023/06/05 01:04:47 - mmengine - INFO - Epoch(train) [4][ 600/3907] lr: 7.9667e-05 eta: 4:38:20 time: 0.6687 data_time: 0.0013 memory: 6319 loss: 0.0994 +2023/06/05 01:05:49 - mmengine - INFO - Epoch(train) [4][ 700/3907] lr: 7.9363e-05 eta: 4:37:17 time: 0.6186 data_time: 0.0008 memory: 6319 loss: 0.0862 +2023/06/05 01:06:48 - mmengine - INFO - Epoch(train) [4][ 800/3907] lr: 7.9058e-05 eta: 4:36:09 time: 0.5616 data_time: 0.0008 memory: 6319 loss: 0.0925 +2023/06/05 01:07:50 - mmengine - INFO - Epoch(train) [4][ 900/3907] lr: 7.8752e-05 eta: 4:35:04 time: 0.6511 data_time: 0.0010 memory: 6319 loss: 0.0991 +2023/06/05 01:08:52 - mmengine - INFO - Epoch(train) [4][1000/3907] lr: 7.8444e-05 eta: 4:34:02 time: 0.6311 data_time: 0.0012 memory: 6319 loss: 0.0778 +2023/06/05 01:09:55 - mmengine - INFO - Epoch(train) [4][1100/3907] lr: 7.8134e-05 eta: 4:33:01 time: 0.5927 data_time: 0.0009 memory: 6319 loss: 0.1054 +2023/06/05 01:10:57 - mmengine - INFO - Epoch(train) [4][1200/3907] lr: 7.7823e-05 eta: 4:31:57 time: 0.6521 data_time: 0.0013 memory: 6319 loss: 0.1059 +2023/06/05 01:11:47 - mmengine - INFO - Exp name: resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1_20230604_225317 +2023/06/05 01:11:59 - mmengine - INFO - Epoch(train) [4][1300/3907] lr: 7.7510e-05 eta: 4:30:54 time: 0.6131 data_time: 0.0012 memory: 6319 loss: 0.0938 +2023/06/05 01:13:02 - mmengine - INFO - Epoch(train) [4][1400/3907] lr: 7.7196e-05 eta: 4:29:52 time: 0.6381 data_time: 0.0015 memory: 6319 loss: 0.0910 +2023/06/05 01:14:06 - mmengine - INFO - Epoch(train) [4][1500/3907] lr: 7.6881e-05 eta: 4:28:53 time: 0.6155 data_time: 0.0010 memory: 6319 loss: 0.0993 +2023/06/05 01:15:12 - mmengine - INFO - Epoch(train) [4][1600/3907] lr: 7.6564e-05 eta: 4:27:57 time: 0.6361 data_time: 0.0012 memory: 6319 loss: 0.0995 +2023/06/05 01:16:16 - mmengine - INFO - Epoch(train) [4][1700/3907] lr: 7.6246e-05 eta: 4:26:59 time: 0.5898 data_time: 0.0013 memory: 6319 loss: 0.1058 +2023/06/05 01:17:18 - mmengine - INFO - Epoch(train) [4][1800/3907] lr: 7.5926e-05 eta: 4:25:56 time: 0.6397 data_time: 0.0012 memory: 6319 loss: 0.0935 +2023/06/05 01:18:22 - mmengine - INFO - Epoch(train) [4][1900/3907] lr: 7.5605e-05 eta: 4:24:56 time: 0.5972 data_time: 0.0009 memory: 6319 loss: 0.0946 +2023/06/05 01:19:26 - mmengine - INFO - Epoch(train) [4][2000/3907] lr: 7.5283e-05 eta: 4:23:56 time: 0.6742 data_time: 0.0009 memory: 6319 loss: 0.0984 +2023/06/05 01:20:29 - mmengine - INFO - Epoch(train) [4][2100/3907] lr: 7.4959e-05 eta: 4:22:55 time: 0.6965 data_time: 0.0014 memory: 6319 loss: 0.0927 +2023/06/05 01:21:34 - mmengine - INFO - Epoch(train) [4][2200/3907] lr: 7.4634e-05 eta: 4:21:57 time: 0.6147 data_time: 0.0014 memory: 6319 loss: 0.0853 +2023/06/05 01:22:27 - mmengine - INFO - Exp name: resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1_20230604_225317 +2023/06/05 01:22:40 - mmengine - INFO - Epoch(train) [4][2300/3907] lr: 7.4308e-05 eta: 4:21:00 time: 0.6109 data_time: 0.0013 memory: 6319 loss: 0.0944 +2023/06/05 01:23:42 - mmengine - INFO - Epoch(train) [4][2400/3907] lr: 7.3980e-05 eta: 4:19:56 time: 0.6191 data_time: 0.0011 memory: 6319 loss: 0.0850 +2023/06/05 01:24:46 - mmengine - INFO - Epoch(train) [4][2500/3907] lr: 7.3652e-05 eta: 4:18:57 time: 0.6604 data_time: 0.0010 memory: 6319 loss: 0.0950 +2023/06/05 01:25:50 - mmengine - INFO - Epoch(train) [4][2600/3907] lr: 7.3322e-05 eta: 4:17:57 time: 0.6484 data_time: 0.0012 memory: 6319 loss: 0.1010 +2023/06/05 01:26:54 - mmengine - INFO - Epoch(train) [4][2700/3907] lr: 7.2991e-05 eta: 4:16:57 time: 0.6595 data_time: 0.0013 memory: 6319 loss: 0.0707 +2023/06/05 01:27:57 - mmengine - INFO - Epoch(train) [4][2800/3907] lr: 7.2659e-05 eta: 4:15:55 time: 0.6921 data_time: 0.0008 memory: 6319 loss: 0.0982 +2023/06/05 01:29:02 - mmengine - INFO - Epoch(train) [4][2900/3907] lr: 7.2325e-05 eta: 4:14:57 time: 0.7043 data_time: 0.0011 memory: 6319 loss: 0.0809 +2023/06/05 01:30:03 - mmengine - INFO - Epoch(train) [4][3000/3907] lr: 7.1991e-05 eta: 4:13:51 time: 0.5801 data_time: 0.0009 memory: 6319 loss: 0.0837 +2023/06/05 01:31:05 - mmengine - INFO - Epoch(train) [4][3100/3907] lr: 7.1655e-05 eta: 4:12:47 time: 0.6220 data_time: 0.0011 memory: 6319 loss: 0.1000 +2023/06/05 01:32:08 - mmengine - INFO - Epoch(train) [4][3200/3907] lr: 7.1318e-05 eta: 4:11:46 time: 0.6225 data_time: 0.0013 memory: 6319 loss: 0.0965 +2023/06/05 01:32:57 - mmengine - INFO - Exp name: resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1_20230604_225317 +2023/06/05 01:33:10 - mmengine - INFO - Epoch(train) [4][3300/3907] lr: 7.0981e-05 eta: 4:10:43 time: 0.6231 data_time: 0.0010 memory: 6319 loss: 0.0949 +2023/06/05 01:34:11 - mmengine - INFO - Epoch(train) [4][3400/3907] lr: 7.0642e-05 eta: 4:09:37 time: 0.6057 data_time: 0.0010 memory: 6319 loss: 0.1001 +2023/06/05 01:35:12 - mmengine - INFO - Epoch(train) [4][3500/3907] lr: 7.0302e-05 eta: 4:08:33 time: 0.6111 data_time: 0.0014 memory: 6319 loss: 0.0895 +2023/06/05 01:36:26 - mmengine - INFO - Epoch(train) [4][3600/3907] lr: 6.9961e-05 eta: 4:07:48 time: 0.6950 data_time: 0.0009 memory: 6319 loss: 0.0923 +2023/06/05 01:37:25 - mmengine - INFO - Epoch(train) [4][3700/3907] lr: 6.9620e-05 eta: 4:06:40 time: 0.6230 data_time: 0.0010 memory: 6319 loss: 0.0875 +2023/06/05 01:38:28 - mmengine - INFO - Epoch(train) [4][3800/3907] lr: 6.9277e-05 eta: 4:05:38 time: 0.6202 data_time: 0.0011 memory: 6319 loss: 0.0921 +2023/06/05 01:39:30 - mmengine - INFO - Epoch(train) [4][3900/3907] lr: 6.8933e-05 eta: 4:04:34 time: 0.5700 data_time: 0.0006 memory: 6319 loss: 0.0955 +2023/06/05 01:39:31 - mmengine - INFO - Exp name: resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1_20230604_225317 +2023/06/05 01:39:31 - mmengine - INFO - Saving checkpoint at 4 epochs +2023/06/05 01:40:19 - mmengine - INFO - Epoch(val) [4][57/57] accuracy/top1: 89.2592 data_time: 0.5374 time: 0.6265 +2023/06/05 01:41:22 - mmengine - INFO - Epoch(train) [5][ 100/3907] lr: 6.8565e-05 eta: 4:03:22 time: 0.6087 data_time: 0.2433 memory: 6319 loss: 0.0906 +2023/06/05 01:42:25 - mmengine - INFO - Epoch(train) [5][ 200/3907] lr: 6.8219e-05 eta: 4:02:21 time: 0.6210 data_time: 0.0011 memory: 6319 loss: 0.0801 +2023/06/05 01:43:25 - mmengine - INFO - Epoch(train) [5][ 300/3907] lr: 6.7873e-05 eta: 4:01:14 time: 0.5748 data_time: 0.0012 memory: 6319 loss: 0.0855 +2023/06/05 01:44:13 - mmengine - INFO - Exp name: resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1_20230604_225317 +2023/06/05 01:44:28 - mmengine - INFO - Epoch(train) [5][ 400/3907] lr: 6.7526e-05 eta: 4:00:12 time: 0.6464 data_time: 0.0013 memory: 6319 loss: 0.0860 +2023/06/05 01:45:32 - mmengine - INFO - Epoch(train) [5][ 500/3907] lr: 6.7178e-05 eta: 3:59:12 time: 0.7036 data_time: 0.0012 memory: 6319 loss: 0.0851 +2023/06/05 01:46:34 - mmengine - INFO - Epoch(train) [5][ 600/3907] lr: 6.6829e-05 eta: 3:58:09 time: 0.5974 data_time: 0.0013 memory: 6319 loss: 0.0835 +2023/06/05 01:47:37 - mmengine - INFO - Epoch(train) [5][ 700/3907] lr: 6.6480e-05 eta: 3:57:07 time: 0.6252 data_time: 0.0014 memory: 6319 loss: 0.0940 +2023/06/05 01:48:45 - mmengine - INFO - Epoch(train) [5][ 800/3907] lr: 6.6129e-05 eta: 3:56:11 time: 0.7113 data_time: 0.0012 memory: 6319 loss: 0.0816 +2023/06/05 01:49:58 - mmengine - INFO - Epoch(train) [5][ 900/3907] lr: 6.5778e-05 eta: 3:55:23 time: 0.5497 data_time: 0.0011 memory: 6319 loss: 0.0895 +2023/06/05 01:50:57 - mmengine - INFO - Epoch(train) [5][1000/3907] lr: 6.5427e-05 eta: 3:54:15 time: 0.6087 data_time: 0.0010 memory: 6319 loss: 0.0858 +2023/06/05 01:51:57 - mmengine - INFO - Epoch(train) [5][1100/3907] lr: 6.5074e-05 eta: 3:53:09 time: 0.5784 data_time: 0.0009 memory: 6319 loss: 0.0990 +2023/06/05 01:52:59 - mmengine - INFO - Epoch(train) [5][1200/3907] lr: 6.4721e-05 eta: 3:52:06 time: 0.6415 data_time: 0.0019 memory: 6319 loss: 0.0831 +2023/06/05 01:54:03 - mmengine - INFO - Epoch(train) [5][1300/3907] lr: 6.4368e-05 eta: 3:51:04 time: 0.6327 data_time: 0.0011 memory: 6319 loss: 0.0898 +2023/06/05 01:54:46 - mmengine - INFO - Exp name: resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1_20230604_225317 +2023/06/05 01:55:04 - mmengine - INFO - Epoch(train) [5][1400/3907] lr: 6.4014e-05 eta: 3:50:00 time: 0.5824 data_time: 0.0015 memory: 6319 loss: 0.0866 +2023/06/05 01:56:07 - mmengine - INFO - Epoch(train) [5][1500/3907] lr: 6.3659e-05 eta: 3:48:58 time: 0.6218 data_time: 0.0013 memory: 6319 loss: 0.0776 +2023/06/05 01:57:09 - mmengine - INFO - Epoch(train) [5][1600/3907] lr: 6.3303e-05 eta: 3:47:55 time: 0.6022 data_time: 0.0011 memory: 6319 loss: 0.0895 +2023/06/05 01:58:11 - mmengine - INFO - Epoch(train) [5][1700/3907] lr: 6.2948e-05 eta: 3:46:51 time: 0.6077 data_time: 0.0014 memory: 6319 loss: 0.0736 +2023/06/05 01:59:13 - mmengine - INFO - Epoch(train) [5][1800/3907] lr: 6.2591e-05 eta: 3:45:48 time: 0.6006 data_time: 0.0012 memory: 6319 loss: 0.0840 +2023/06/05 02:00:13 - mmengine - INFO - Epoch(train) [5][1900/3907] lr: 6.2234e-05 eta: 3:44:43 time: 0.5905 data_time: 0.0014 memory: 6319 loss: 0.0748 +2023/06/05 02:01:16 - mmengine - INFO - Epoch(train) [5][2000/3907] lr: 6.1877e-05 eta: 3:43:40 time: 0.6500 data_time: 0.0012 memory: 6319 loss: 0.0920 +2023/06/05 02:02:27 - mmengine - INFO - Epoch(train) [5][2100/3907] lr: 6.1519e-05 eta: 3:42:48 time: 0.7826 data_time: 0.0010 memory: 6319 loss: 0.0937 +2023/06/05 02:03:27 - mmengine - INFO - Epoch(train) [5][2200/3907] lr: 6.1161e-05 eta: 3:41:41 time: 0.5486 data_time: 0.0008 memory: 6319 loss: 0.0733 +2023/06/05 02:04:27 - mmengine - INFO - Epoch(train) [5][2300/3907] lr: 6.0802e-05 eta: 3:40:36 time: 0.6194 data_time: 0.0010 memory: 6319 loss: 0.0858 +2023/06/05 02:05:11 - mmengine - INFO - Exp name: resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1_20230604_225317 +2023/06/05 02:05:31 - mmengine - INFO - Epoch(train) [5][2400/3907] lr: 6.0443e-05 eta: 3:39:35 time: 0.6452 data_time: 0.0008 memory: 6319 loss: 0.0875 +2023/06/05 02:06:32 - mmengine - INFO - Epoch(train) [5][2500/3907] lr: 6.0084e-05 eta: 3:38:30 time: 0.6099 data_time: 0.0008 memory: 6319 loss: 0.0854 +2023/06/05 02:07:33 - mmengine - INFO - Epoch(train) [5][2600/3907] lr: 5.9724e-05 eta: 3:37:26 time: 0.6139 data_time: 0.0013 memory: 6319 loss: 0.0867 +2023/06/05 02:08:36 - mmengine - INFO - Epoch(train) [5][2700/3907] lr: 5.9364e-05 eta: 3:36:24 time: 0.6391 data_time: 0.0013 memory: 6319 loss: 0.0976 +2023/06/05 02:09:39 - mmengine - INFO - Epoch(train) [5][2800/3907] lr: 5.9004e-05 eta: 3:35:22 time: 0.6239 data_time: 0.0013 memory: 6319 loss: 0.0817 +2023/06/05 02:10:43 - mmengine - INFO - Epoch(train) [5][2900/3907] lr: 5.8643e-05 eta: 3:34:21 time: 0.7048 data_time: 0.0012 memory: 6319 loss: 0.0827 +2023/06/05 02:11:45 - mmengine - INFO - Epoch(train) [5][3000/3907] lr: 5.8283e-05 eta: 3:33:18 time: 0.6219 data_time: 0.0010 memory: 6319 loss: 0.0901 +2023/06/05 02:12:45 - mmengine - INFO - Epoch(train) [5][3100/3907] lr: 5.7922e-05 eta: 3:32:12 time: 0.6343 data_time: 0.0014 memory: 6319 loss: 0.0880 +2023/06/05 02:13:46 - mmengine - INFO - Epoch(train) [5][3200/3907] lr: 5.7560e-05 eta: 3:31:08 time: 0.6153 data_time: 0.0014 memory: 6319 loss: 0.0820 +2023/06/05 02:14:50 - mmengine - INFO - Epoch(train) [5][3300/3907] lr: 5.7199e-05 eta: 3:30:07 time: 0.6185 data_time: 0.0013 memory: 6319 loss: 0.0713 +2023/06/05 02:15:35 - mmengine - INFO - Exp name: resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1_20230604_225317 +2023/06/05 02:15:53 - mmengine - INFO - Epoch(train) [5][3400/3907] lr: 5.6838e-05 eta: 3:29:04 time: 0.5832 data_time: 0.0010 memory: 6319 loss: 0.0640 +2023/06/05 02:16:55 - mmengine - INFO - Epoch(train) [5][3500/3907] lr: 5.6476e-05 eta: 3:28:01 time: 0.6259 data_time: 0.0015 memory: 6319 loss: 0.0796 +2023/06/05 02:17:58 - mmengine - INFO - Epoch(train) [5][3600/3907] lr: 5.6114e-05 eta: 3:26:59 time: 0.6208 data_time: 0.0017 memory: 6319 loss: 0.0774 +2023/06/05 02:19:05 - mmengine - INFO - Epoch(train) [5][3700/3907] lr: 5.5753e-05 eta: 3:26:01 time: 0.6512 data_time: 0.0452 memory: 6319 loss: 0.0744 +2023/06/05 02:20:24 - mmengine - INFO - Epoch(train) [5][3800/3907] lr: 5.5391e-05 eta: 3:25:15 time: 0.6746 data_time: 0.0018 memory: 6319 loss: 0.0869 +2023/06/05 02:21:21 - mmengine - INFO - Epoch(train) [5][3900/3907] lr: 5.5029e-05 eta: 3:24:07 time: 0.5811 data_time: 0.0008 memory: 6319 loss: 0.0817 +2023/06/05 02:21:24 - mmengine - INFO - Exp name: resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1_20230604_225317 +2023/06/05 02:21:24 - mmengine - INFO - Saving checkpoint at 5 epochs +2023/06/05 02:22:11 - mmengine - INFO - Epoch(val) [5][57/57] accuracy/top1: 93.6850 data_time: 0.4932 time: 0.5811 +2023/06/05 02:23:14 - mmengine - INFO - Epoch(train) [6][ 100/3907] lr: 5.4642e-05 eta: 3:22:58 time: 0.5858 data_time: 0.2116 memory: 6319 loss: 0.0871 +2023/06/05 02:24:15 - mmengine - INFO - Epoch(train) [6][ 200/3907] lr: 5.4280e-05 eta: 3:21:54 time: 0.6000 data_time: 0.0548 memory: 6319 loss: 0.0986 +2023/06/05 02:25:18 - mmengine - INFO - Epoch(train) [6][ 300/3907] lr: 5.3918e-05 eta: 3:20:51 time: 0.5921 data_time: 0.0012 memory: 6319 loss: 0.0761 +2023/06/05 02:26:16 - mmengine - INFO - Epoch(train) [6][ 400/3907] lr: 5.3556e-05 eta: 3:19:44 time: 0.5497 data_time: 0.0011 memory: 6319 loss: 0.0839 +2023/06/05 02:26:58 - mmengine - INFO - Exp name: resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1_20230604_225317 +2023/06/05 02:27:16 - mmengine - INFO - Epoch(train) [6][ 500/3907] lr: 5.3195e-05 eta: 3:18:39 time: 0.5999 data_time: 0.0013 memory: 6319 loss: 0.0799 +2023/06/05 02:28:18 - mmengine - INFO - Epoch(train) [6][ 600/3907] lr: 5.2833e-05 eta: 3:17:36 time: 0.5771 data_time: 0.0013 memory: 6319 loss: 0.0872 +2023/06/05 02:29:21 - mmengine - INFO - Epoch(train) [6][ 700/3907] lr: 5.2472e-05 eta: 3:16:33 time: 0.6362 data_time: 0.0009 memory: 6319 loss: 0.0750 +2023/06/05 02:30:22 - mmengine - INFO - Epoch(train) [6][ 800/3907] lr: 5.2111e-05 eta: 3:15:29 time: 0.6204 data_time: 0.0011 memory: 6319 loss: 0.0781 +2023/06/05 02:31:24 - mmengine - INFO - Epoch(train) [6][ 900/3907] lr: 5.1750e-05 eta: 3:14:26 time: 0.6397 data_time: 0.0010 memory: 6319 loss: 0.0802 +2023/06/05 02:32:24 - mmengine - INFO - Epoch(train) [6][1000/3907] lr: 5.1389e-05 eta: 3:13:22 time: 0.5844 data_time: 0.0012 memory: 6319 loss: 0.0882 +2023/06/05 02:33:25 - mmengine - INFO - Epoch(train) [6][1100/3907] lr: 5.1029e-05 eta: 3:12:18 time: 0.6115 data_time: 0.0010 memory: 6319 loss: 0.0808 +2023/06/05 02:34:28 - mmengine - INFO - Epoch(train) [6][1200/3907] lr: 5.0668e-05 eta: 3:11:15 time: 0.6076 data_time: 0.0012 memory: 6319 loss: 0.0678 +2023/06/05 02:35:30 - mmengine - INFO - Epoch(train) [6][1300/3907] lr: 5.0308e-05 eta: 3:10:12 time: 0.5770 data_time: 0.0011 memory: 6319 loss: 0.0786 +2023/06/05 02:36:31 - mmengine - INFO - Epoch(train) [6][1400/3907] lr: 4.9949e-05 eta: 3:09:08 time: 0.6137 data_time: 0.0008 memory: 6319 loss: 0.0833 +2023/06/05 02:37:12 - mmengine - INFO - Exp name: resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1_20230604_225317 +2023/06/05 02:37:33 - mmengine - INFO - Epoch(train) [6][1500/3907] lr: 4.9589e-05 eta: 3:08:05 time: 0.7235 data_time: 0.0014 memory: 6319 loss: 0.0815 +2023/06/05 02:38:44 - mmengine - INFO - Epoch(train) [6][1600/3907] lr: 4.9230e-05 eta: 3:07:09 time: 0.6747 data_time: 0.0011 memory: 6319 loss: 0.0848 +2023/06/05 02:40:21 - mmengine - INFO - Epoch(train) [6][1700/3907] lr: 4.8871e-05 eta: 3:06:36 time: 0.8727 data_time: 0.0011 memory: 6319 loss: 0.0740 +2023/06/05 02:41:22 - mmengine - INFO - Epoch(train) [6][1800/3907] lr: 4.8513e-05 eta: 3:05:32 time: 0.5724 data_time: 0.0008 memory: 6319 loss: 0.0774 +2023/06/05 02:42:21 - mmengine - INFO - Epoch(train) [6][1900/3907] lr: 4.8155e-05 eta: 3:04:25 time: 0.6192 data_time: 0.0012 memory: 6319 loss: 0.0956 +2023/06/05 02:43:16 - mmengine - INFO - Epoch(train) [6][2000/3907] lr: 4.7798e-05 eta: 3:03:16 time: 0.5283 data_time: 0.0010 memory: 6319 loss: 0.0726 +2023/06/05 02:44:13 - mmengine - INFO - Epoch(train) [6][2100/3907] lr: 4.7441e-05 eta: 3:02:09 time: 0.6080 data_time: 0.0207 memory: 6319 loss: 0.0802 +2023/06/05 02:45:14 - mmengine - INFO - Epoch(train) [6][2200/3907] lr: 4.7084e-05 eta: 3:01:05 time: 0.6035 data_time: 0.0013 memory: 6319 loss: 0.0750 +2023/06/05 02:46:16 - mmengine - INFO - Epoch(train) [6][2300/3907] lr: 4.6729e-05 eta: 3:00:02 time: 0.6789 data_time: 0.0012 memory: 6319 loss: 0.0635 +2023/06/05 02:47:18 - mmengine - INFO - Epoch(train) [6][2400/3907] lr: 4.6373e-05 eta: 2:58:59 time: 0.6389 data_time: 0.0008 memory: 6319 loss: 0.0692 +2023/06/05 02:47:56 - mmengine - INFO - Exp name: resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1_20230604_225317 +2023/06/05 02:48:19 - mmengine - INFO - Epoch(train) [6][2500/3907] lr: 4.6018e-05 eta: 2:57:55 time: 0.6203 data_time: 0.0008 memory: 6319 loss: 0.0742 +2023/06/05 02:49:18 - mmengine - INFO - Epoch(train) [6][2600/3907] lr: 4.5664e-05 eta: 2:56:50 time: 0.5737 data_time: 0.0011 memory: 6319 loss: 0.0653 +2023/06/05 02:50:20 - mmengine - INFO - Epoch(train) [6][2700/3907] lr: 4.5310e-05 eta: 2:55:46 time: 0.6477 data_time: 0.0013 memory: 6319 loss: 0.0740 +2023/06/05 02:51:22 - mmengine - INFO - Epoch(train) [6][2800/3907] lr: 4.4957e-05 eta: 2:54:43 time: 0.6272 data_time: 0.0014 memory: 6319 loss: 0.0849 +2023/06/05 02:52:24 - mmengine - INFO - Epoch(train) [6][2900/3907] lr: 4.4605e-05 eta: 2:53:40 time: 0.6080 data_time: 0.0011 memory: 6319 loss: 0.0758 +2023/06/05 02:53:24 - mmengine - INFO - Epoch(train) [6][3000/3907] lr: 4.4253e-05 eta: 2:52:35 time: 0.6142 data_time: 0.0013 memory: 6319 loss: 0.0712 +2023/06/05 02:54:22 - mmengine - INFO - Epoch(train) [6][3100/3907] lr: 4.3902e-05 eta: 2:51:29 time: 0.5848 data_time: 0.0010 memory: 6319 loss: 0.0815 +2023/06/05 02:55:25 - mmengine - INFO - Epoch(train) [6][3200/3907] lr: 4.3552e-05 eta: 2:50:27 time: 0.6074 data_time: 0.0010 memory: 6319 loss: 0.0827 +2023/06/05 02:56:25 - mmengine - INFO - Epoch(train) [6][3300/3907] lr: 4.3202e-05 eta: 2:49:23 time: 0.6092 data_time: 0.0009 memory: 6319 loss: 0.0630 +2023/06/05 02:57:24 - mmengine - INFO - Epoch(train) [6][3400/3907] lr: 4.2854e-05 eta: 2:48:17 time: 0.6172 data_time: 0.0014 memory: 6319 loss: 0.0864 +2023/06/05 02:58:06 - mmengine - INFO - Exp name: resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1_20230604_225317 +2023/06/05 02:58:25 - mmengine - INFO - Epoch(train) [6][3500/3907] lr: 4.2506e-05 eta: 2:47:13 time: 0.6004 data_time: 0.0011 memory: 6319 loss: 0.0703 +2023/06/05 02:59:27 - mmengine - INFO - Epoch(train) [6][3600/3907] lr: 4.2158e-05 eta: 2:46:11 time: 0.5945 data_time: 0.0008 memory: 6319 loss: 0.0810 +2023/06/05 03:00:28 - mmengine - INFO - Epoch(train) [6][3700/3907] lr: 4.1812e-05 eta: 2:45:07 time: 0.6618 data_time: 0.0011 memory: 6319 loss: 0.0744 +2023/06/05 03:01:30 - mmengine - INFO - Epoch(train) [6][3800/3907] lr: 4.1466e-05 eta: 2:44:04 time: 0.6104 data_time: 0.0013 memory: 6319 loss: 0.0714 +2023/06/05 03:02:31 - mmengine - INFO - Epoch(train) [6][3900/3907] lr: 4.1122e-05 eta: 2:43:01 time: 0.6612 data_time: 0.0009 memory: 6319 loss: 0.0888 +2023/06/05 03:02:32 - mmengine - INFO - Exp name: resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1_20230604_225317 +2023/06/05 03:02:32 - mmengine - INFO - Saving checkpoint at 6 epochs +2023/06/05 03:03:20 - mmengine - INFO - Epoch(val) [6][57/57] accuracy/top1: 94.7845 data_time: 0.5004 time: 0.5899 +2023/06/05 03:04:22 - mmengine - INFO - Epoch(train) [7][ 100/3907] lr: 4.0754e-05 eta: 2:41:51 time: 0.6164 data_time: 0.4744 memory: 6319 loss: 0.0669 +2023/06/05 03:05:24 - mmengine - INFO - Epoch(train) [7][ 200/3907] lr: 4.0411e-05 eta: 2:40:48 time: 0.6700 data_time: 0.5253 memory: 6319 loss: 0.0630 +2023/06/05 03:06:25 - mmengine - INFO - Epoch(train) [7][ 300/3907] lr: 4.0069e-05 eta: 2:39:45 time: 0.5889 data_time: 0.4356 memory: 6319 loss: 0.0716 +2023/06/05 03:07:23 - mmengine - INFO - Epoch(train) [7][ 400/3907] lr: 3.9729e-05 eta: 2:38:39 time: 0.5881 data_time: 0.4349 memory: 6319 loss: 0.0887 +2023/06/05 03:08:27 - mmengine - INFO - Epoch(train) [7][ 500/3907] lr: 3.9389e-05 eta: 2:37:38 time: 0.6514 data_time: 0.4960 memory: 6319 loss: 0.0866 +2023/06/05 03:09:02 - mmengine - INFO - Exp name: resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1_20230604_225317 +2023/06/05 03:09:27 - mmengine - INFO - Epoch(train) [7][ 600/3907] lr: 3.9050e-05 eta: 2:36:34 time: 0.5642 data_time: 0.4207 memory: 6319 loss: 0.0793 +2023/06/05 03:10:31 - mmengine - INFO - Epoch(train) [7][ 700/3907] lr: 3.8712e-05 eta: 2:35:32 time: 0.6644 data_time: 0.5221 memory: 6319 loss: 0.0802 +2023/06/05 03:11:32 - mmengine - INFO - Epoch(train) [7][ 800/3907] lr: 3.8375e-05 eta: 2:34:29 time: 0.6276 data_time: 0.4823 memory: 6319 loss: 0.0816 +2023/06/05 03:12:35 - mmengine - INFO - Epoch(train) [7][ 900/3907] lr: 3.8039e-05 eta: 2:33:26 time: 0.6439 data_time: 0.4872 memory: 6319 loss: 0.0778 +2023/06/05 03:13:35 - mmengine - INFO - Epoch(train) [7][1000/3907] lr: 3.7705e-05 eta: 2:32:22 time: 0.6047 data_time: 0.4618 memory: 6319 loss: 0.0706 +2023/06/05 03:14:36 - mmengine - INFO - Epoch(train) [7][1100/3907] lr: 3.7371e-05 eta: 2:31:19 time: 0.6124 data_time: 0.4566 memory: 6319 loss: 0.0922 +2023/06/05 03:15:34 - mmengine - INFO - Epoch(train) [7][1200/3907] lr: 3.7039e-05 eta: 2:30:14 time: 0.6098 data_time: 0.4667 memory: 6319 loss: 0.0824 +2023/06/05 03:16:33 - mmengine - INFO - Epoch(train) [7][1300/3907] lr: 3.6708e-05 eta: 2:29:09 time: 0.5636 data_time: 0.4211 memory: 6319 loss: 0.0805 +2023/06/05 03:17:30 - mmengine - INFO - Epoch(train) [7][1400/3907] lr: 3.6378e-05 eta: 2:28:04 time: 0.5435 data_time: 0.4015 memory: 6319 loss: 0.0690 +2023/06/05 03:18:30 - mmengine - INFO - Epoch(train) [7][1500/3907] lr: 3.6049e-05 eta: 2:27:00 time: 0.6432 data_time: 0.5016 memory: 6319 loss: 0.0660 +2023/06/05 03:19:08 - mmengine - INFO - Exp name: resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1_20230604_225317 +2023/06/05 03:19:33 - mmengine - INFO - Epoch(train) [7][1600/3907] lr: 3.5721e-05 eta: 2:25:58 time: 0.6021 data_time: 0.4593 memory: 6319 loss: 0.0747 +2023/06/05 03:20:34 - mmengine - INFO - Epoch(train) [7][1700/3907] lr: 3.5395e-05 eta: 2:24:54 time: 0.5701 data_time: 0.4278 memory: 6319 loss: 0.0733 +2023/06/05 03:21:33 - mmengine - INFO - Epoch(train) [7][1800/3907] lr: 3.5070e-05 eta: 2:23:50 time: 0.5632 data_time: 0.4210 memory: 6319 loss: 0.0741 +2023/06/05 03:22:33 - mmengine - INFO - Epoch(train) [7][1900/3907] lr: 3.4746e-05 eta: 2:22:47 time: 0.6304 data_time: 0.4880 memory: 6319 loss: 0.0916 +2023/06/05 03:23:34 - mmengine - INFO - Epoch(train) [7][2000/3907] lr: 3.4424e-05 eta: 2:21:43 time: 0.6336 data_time: 0.4624 memory: 6319 loss: 0.0936 +2023/06/05 03:24:36 - mmengine - INFO - Epoch(train) [7][2100/3907] lr: 3.4103e-05 eta: 2:20:41 time: 0.6018 data_time: 0.4577 memory: 6319 loss: 0.0751 +2023/06/05 03:25:36 - mmengine - INFO - Epoch(train) [7][2200/3907] lr: 3.3783e-05 eta: 2:19:37 time: 0.6132 data_time: 0.4712 memory: 6319 loss: 0.0968 +2023/06/05 03:26:36 - mmengine - INFO - Epoch(train) [7][2300/3907] lr: 3.3465e-05 eta: 2:18:34 time: 0.5796 data_time: 0.4360 memory: 6319 loss: 0.0864 +2023/06/05 03:27:39 - mmengine - INFO - Epoch(train) [7][2400/3907] lr: 3.3148e-05 eta: 2:17:31 time: 0.6460 data_time: 0.5020 memory: 6319 loss: 0.0851 +2023/06/05 03:28:42 - mmengine - INFO - Epoch(train) [7][2500/3907] lr: 3.2832e-05 eta: 2:16:29 time: 0.6405 data_time: 0.4878 memory: 6319 loss: 0.0772 +2023/06/05 03:29:13 - mmengine - INFO - Exp name: resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1_20230604_225317 +2023/06/05 03:29:42 - mmengine - INFO - Epoch(train) [7][2600/3907] lr: 3.2518e-05 eta: 2:15:26 time: 0.6298 data_time: 0.4859 memory: 6319 loss: 0.0843 +2023/06/05 03:30:45 - mmengine - INFO - Epoch(train) [7][2700/3907] lr: 3.2205e-05 eta: 2:14:24 time: 0.6201 data_time: 0.4764 memory: 6319 loss: 0.0819 +2023/06/05 03:31:47 - mmengine - INFO - Epoch(train) [7][2800/3907] lr: 3.1894e-05 eta: 2:13:21 time: 0.6483 data_time: 0.5053 memory: 6319 loss: 0.0904 +2023/06/05 03:32:48 - mmengine - INFO - Epoch(train) [7][2900/3907] lr: 3.1584e-05 eta: 2:12:18 time: 0.6086 data_time: 0.4550 memory: 6319 loss: 0.0751 +2023/06/05 03:33:51 - mmengine - INFO - Epoch(train) [7][3000/3907] lr: 3.1276e-05 eta: 2:11:16 time: 0.5821 data_time: 0.4395 memory: 6319 loss: 0.0826 +2023/06/05 03:34:51 - mmengine - INFO - Epoch(train) [7][3100/3907] lr: 3.0969e-05 eta: 2:10:13 time: 0.6294 data_time: 0.4849 memory: 6319 loss: 0.0919 +2023/06/05 03:35:51 - mmengine - INFO - Epoch(train) [7][3200/3907] lr: 3.0664e-05 eta: 2:09:09 time: 0.5464 data_time: 0.4047 memory: 6319 loss: 0.0836 +2023/06/05 03:36:50 - mmengine - INFO - Epoch(train) [7][3300/3907] lr: 3.0360e-05 eta: 2:08:05 time: 0.5855 data_time: 0.4322 memory: 6319 loss: 0.0874 +2023/06/05 03:37:50 - mmengine - INFO - Epoch(train) [7][3400/3907] lr: 3.0058e-05 eta: 2:07:02 time: 0.6322 data_time: 0.4880 memory: 6319 loss: 0.0883 +2023/06/05 03:38:51 - mmengine - INFO - Epoch(train) [7][3500/3907] lr: 2.9758e-05 eta: 2:05:59 time: 0.5850 data_time: 0.4412 memory: 6319 loss: 0.0903 +2023/06/05 03:39:23 - mmengine - INFO - Exp name: resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1_20230604_225317 +2023/06/05 03:39:53 - mmengine - INFO - Epoch(train) [7][3600/3907] lr: 2.9459e-05 eta: 2:04:56 time: 0.6256 data_time: 0.4815 memory: 6319 loss: 0.0793 +2023/06/05 03:40:55 - mmengine - INFO - Epoch(train) [7][3700/3907] lr: 2.9162e-05 eta: 2:03:54 time: 0.5588 data_time: 0.4176 memory: 6319 loss: 0.0818 +2023/06/05 03:41:54 - mmengine - INFO - Epoch(train) [7][3800/3907] lr: 2.8867e-05 eta: 2:02:50 time: 0.6088 data_time: 0.4661 memory: 6319 loss: 0.0839 +2023/06/05 03:42:55 - mmengine - INFO - Epoch(train) [7][3900/3907] lr: 2.8573e-05 eta: 2:01:47 time: 0.5906 data_time: 0.4384 memory: 6319 loss: 0.0712 +2023/06/05 03:42:58 - mmengine - INFO - Exp name: resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1_20230604_225317 +2023/06/05 03:42:58 - mmengine - INFO - Saving checkpoint at 7 epochs +2023/06/05 03:43:46 - mmengine - INFO - Epoch(val) [7][57/57] accuracy/top1: 96.3119 data_time: 0.5100 time: 0.6020 +2023/06/05 03:44:50 - mmengine - INFO - Epoch(train) [8][ 100/3907] lr: 2.8261e-05 eta: 2:00:40 time: 0.5920 data_time: 0.4505 memory: 6319 loss: 0.0879 +2023/06/05 03:45:51 - mmengine - INFO - Epoch(train) [8][ 200/3907] lr: 2.7971e-05 eta: 1:59:37 time: 0.6259 data_time: 0.4833 memory: 6319 loss: 0.0642 +2023/06/05 03:46:52 - mmengine - INFO - Epoch(train) [8][ 300/3907] lr: 2.7682e-05 eta: 1:58:35 time: 0.6036 data_time: 0.4600 memory: 6319 loss: 0.0862 +2023/06/05 03:47:54 - mmengine - INFO - Epoch(train) [8][ 400/3907] lr: 2.7395e-05 eta: 1:57:32 time: 0.6247 data_time: 0.4710 memory: 6319 loss: 0.0751 +2023/06/05 03:48:53 - mmengine - INFO - Epoch(train) [8][ 500/3907] lr: 2.7111e-05 eta: 1:56:29 time: 0.5760 data_time: 0.4342 memory: 6319 loss: 0.0685 +2023/06/05 03:49:51 - mmengine - INFO - Epoch(train) [8][ 600/3907] lr: 2.6828e-05 eta: 1:55:25 time: 0.5643 data_time: 0.4144 memory: 6319 loss: 0.0699 +2023/06/05 03:50:22 - mmengine - INFO - Exp name: resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1_20230604_225317 +2023/06/05 03:50:52 - mmengine - INFO - Epoch(train) [8][ 700/3907] lr: 2.6546e-05 eta: 1:54:22 time: 0.5827 data_time: 0.4420 memory: 6319 loss: 0.0742 +2023/06/05 03:51:51 - mmengine - INFO - Epoch(train) [8][ 800/3907] lr: 2.6267e-05 eta: 1:53:19 time: 0.6737 data_time: 0.5173 memory: 6319 loss: 0.1046 +2023/06/05 03:52:53 - mmengine - INFO - Epoch(train) [8][ 900/3907] lr: 2.5989e-05 eta: 1:52:16 time: 0.6133 data_time: 0.4718 memory: 6319 loss: 0.0815 +2023/06/05 03:53:55 - mmengine - INFO - Epoch(train) [8][1000/3907] lr: 2.5714e-05 eta: 1:51:14 time: 0.6003 data_time: 0.4434 memory: 6319 loss: 0.0845 +2023/06/05 03:54:56 - mmengine - INFO - Epoch(train) [8][1100/3907] lr: 2.5440e-05 eta: 1:50:11 time: 0.6659 data_time: 0.5231 memory: 6319 loss: 0.0773 +2023/06/05 03:55:58 - mmengine - INFO - Epoch(train) [8][1200/3907] lr: 2.5168e-05 eta: 1:49:08 time: 0.5979 data_time: 0.4535 memory: 6319 loss: 0.0728 +2023/06/05 03:56:59 - mmengine - INFO - Epoch(train) [8][1300/3907] lr: 2.4898e-05 eta: 1:48:06 time: 0.6376 data_time: 0.4939 memory: 6319 loss: 0.0855 +2023/06/05 03:57:58 - mmengine - INFO - Epoch(train) [8][1400/3907] lr: 2.4630e-05 eta: 1:47:03 time: 0.6086 data_time: 0.4644 memory: 6319 loss: 0.0738 +2023/06/05 03:58:59 - mmengine - INFO - Epoch(train) [8][1500/3907] lr: 2.4364e-05 eta: 1:46:00 time: 0.6236 data_time: 0.4821 memory: 6319 loss: 0.0770 +2023/06/05 04:00:01 - mmengine - INFO - Epoch(train) [8][1600/3907] lr: 2.4100e-05 eta: 1:44:58 time: 0.6405 data_time: 0.4964 memory: 6319 loss: 0.0698 +2023/06/05 04:00:32 - mmengine - INFO - Exp name: resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1_20230604_225317 +2023/06/05 04:01:02 - mmengine - INFO - Epoch(train) [8][1700/3907] lr: 2.3838e-05 eta: 1:43:55 time: 0.5912 data_time: 0.4483 memory: 6319 loss: 0.0646 +2023/06/05 04:02:04 - mmengine - INFO - Epoch(train) [8][1800/3907] lr: 2.3578e-05 eta: 1:42:53 time: 0.5862 data_time: 0.4443 memory: 6319 loss: 0.0721 +2023/06/05 04:03:06 - mmengine - INFO - Epoch(train) [8][1900/3907] lr: 2.3320e-05 eta: 1:41:50 time: 0.6124 data_time: 0.4700 memory: 6319 loss: 0.0770 +2023/06/05 04:04:09 - mmengine - INFO - Epoch(train) [8][2000/3907] lr: 2.3064e-05 eta: 1:40:48 time: 0.7024 data_time: 0.5597 memory: 6319 loss: 0.0865 +2023/06/05 04:05:09 - mmengine - INFO - Epoch(train) [8][2100/3907] lr: 2.2810e-05 eta: 1:39:45 time: 0.5932 data_time: 0.4523 memory: 6319 loss: 0.0876 +2023/06/05 04:06:09 - mmengine - INFO - Epoch(train) [8][2200/3907] lr: 2.2558e-05 eta: 1:38:42 time: 0.5839 data_time: 0.4421 memory: 6319 loss: 0.0860 +2023/06/05 04:07:09 - mmengine - INFO - Epoch(train) [8][2300/3907] lr: 2.2309e-05 eta: 1:37:39 time: 0.6095 data_time: 0.4542 memory: 6319 loss: 0.0769 +2023/06/05 04:08:06 - mmengine - INFO - Epoch(train) [8][2400/3907] lr: 2.2061e-05 eta: 1:36:36 time: 0.5770 data_time: 0.4348 memory: 6319 loss: 0.0752 +2023/06/05 04:09:04 - mmengine - INFO - Epoch(train) [8][2500/3907] lr: 2.1816e-05 eta: 1:35:32 time: 0.5602 data_time: 0.4176 memory: 6319 loss: 0.0701 +2023/06/05 04:10:03 - mmengine - INFO - Epoch(train) [8][2600/3907] lr: 2.1572e-05 eta: 1:34:29 time: 0.5870 data_time: 0.4450 memory: 6319 loss: 0.0730 +2023/06/05 04:10:34 - mmengine - INFO - Exp name: resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1_20230604_225317 +2023/06/05 04:11:04 - mmengine - INFO - Epoch(train) [8][2700/3907] lr: 2.1331e-05 eta: 1:33:27 time: 0.6086 data_time: 0.4658 memory: 6319 loss: 0.0836 +2023/06/05 04:12:03 - mmengine - INFO - Epoch(train) [8][2800/3907] lr: 2.1092e-05 eta: 1:32:24 time: 0.5645 data_time: 0.4225 memory: 6319 loss: 0.0875 +2023/06/05 04:13:02 - mmengine - INFO - Epoch(train) [8][2900/3907] lr: 2.0855e-05 eta: 1:31:20 time: 0.6183 data_time: 0.4734 memory: 6319 loss: 0.0757 +2023/06/05 04:14:04 - mmengine - INFO - Epoch(train) [8][3000/3907] lr: 2.0621e-05 eta: 1:30:18 time: 0.6540 data_time: 0.5105 memory: 6319 loss: 0.0856 +2023/06/05 04:15:09 - mmengine - INFO - Epoch(train) [8][3100/3907] lr: 2.0388e-05 eta: 1:29:17 time: 0.6890 data_time: 0.5449 memory: 6319 loss: 0.0844 +2023/06/05 04:16:10 - mmengine - INFO - Epoch(train) [8][3200/3907] lr: 2.0158e-05 eta: 1:28:14 time: 0.6061 data_time: 0.4630 memory: 6319 loss: 0.0747 +2023/06/05 04:17:14 - mmengine - INFO - Epoch(train) [8][3300/3907] lr: 1.9930e-05 eta: 1:27:13 time: 0.6578 data_time: 0.5144 memory: 6319 loss: 0.0883 +2023/06/05 04:18:15 - mmengine - INFO - Epoch(train) [8][3400/3907] lr: 1.9705e-05 eta: 1:26:10 time: 0.6399 data_time: 0.4972 memory: 6319 loss: 0.0772 +2023/06/05 04:19:17 - mmengine - INFO - Epoch(train) [8][3500/3907] lr: 1.9481e-05 eta: 1:25:08 time: 0.6035 data_time: 0.4575 memory: 6319 loss: 0.0775 +2023/06/05 04:20:17 - mmengine - INFO - Epoch(train) [8][3600/3907] lr: 1.9260e-05 eta: 1:24:05 time: 0.6433 data_time: 0.5013 memory: 6319 loss: 0.0758 +2023/06/05 04:20:47 - mmengine - INFO - Exp name: resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1_20230604_225317 +2023/06/05 04:21:18 - mmengine - INFO - Epoch(train) [8][3700/3907] lr: 1.9042e-05 eta: 1:23:03 time: 0.7127 data_time: 0.5531 memory: 6319 loss: 0.0728 +2023/06/05 04:22:19 - mmengine - INFO - Epoch(train) [8][3800/3907] lr: 1.8825e-05 eta: 1:22:01 time: 0.6105 data_time: 0.4671 memory: 6319 loss: 0.0895 +2023/06/05 04:23:16 - mmengine - INFO - Epoch(train) [8][3900/3907] lr: 1.8611e-05 eta: 1:20:57 time: 0.5426 data_time: 0.3985 memory: 6319 loss: 0.0749 +2023/06/05 04:23:21 - mmengine - INFO - Exp name: resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1_20230604_225317 +2023/06/05 04:23:21 - mmengine - INFO - Saving checkpoint at 8 epochs +2023/06/05 04:24:08 - mmengine - INFO - Epoch(val) [8][57/57] accuracy/top1: 96.7225 data_time: 0.4933 time: 0.5810 +2023/06/05 04:25:11 - mmengine - INFO - Epoch(train) [9][ 100/3907] lr: 1.8385e-05 eta: 1:19:51 time: 0.6189 data_time: 0.4759 memory: 6319 loss: 0.0755 +2023/06/05 04:26:11 - mmengine - INFO - Epoch(train) [9][ 200/3907] lr: 1.8176e-05 eta: 1:18:48 time: 0.6066 data_time: 0.4504 memory: 6319 loss: 0.0818 +2023/06/05 04:27:12 - mmengine - INFO - Epoch(train) [9][ 300/3907] lr: 1.7969e-05 eta: 1:17:46 time: 0.6234 data_time: 0.4804 memory: 6319 loss: 0.0670 +2023/06/05 04:28:12 - mmengine - INFO - Epoch(train) [9][ 400/3907] lr: 1.7765e-05 eta: 1:16:43 time: 0.6136 data_time: 0.4701 memory: 6319 loss: 0.0769 +2023/06/05 04:29:11 - mmengine - INFO - Epoch(train) [9][ 500/3907] lr: 1.7563e-05 eta: 1:15:41 time: 0.5967 data_time: 0.4518 memory: 6319 loss: 0.0811 +2023/06/05 04:30:10 - mmengine - INFO - Epoch(train) [9][ 600/3907] lr: 1.7363e-05 eta: 1:14:38 time: 0.5911 data_time: 0.4487 memory: 6319 loss: 0.0706 +2023/06/05 04:31:11 - mmengine - INFO - Epoch(train) [9][ 700/3907] lr: 1.7166e-05 eta: 1:13:36 time: 0.6231 data_time: 0.4813 memory: 6319 loss: 0.0746 +2023/06/05 04:31:38 - mmengine - INFO - Exp name: resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1_20230604_225317 +2023/06/05 04:32:11 - mmengine - INFO - Epoch(train) [9][ 800/3907] lr: 1.6971e-05 eta: 1:12:33 time: 0.6046 data_time: 0.4628 memory: 6319 loss: 0.0700 +2023/06/05 04:33:10 - mmengine - INFO - Epoch(train) [9][ 900/3907] lr: 1.6779e-05 eta: 1:11:30 time: 0.6288 data_time: 0.4872 memory: 6319 loss: 0.0780 +2023/06/05 04:34:11 - mmengine - INFO - Epoch(train) [9][1000/3907] lr: 1.6589e-05 eta: 1:10:28 time: 0.5982 data_time: 0.4406 memory: 6319 loss: 0.0653 +2023/06/05 04:35:11 - mmengine - INFO - Epoch(train) [9][1100/3907] lr: 1.6402e-05 eta: 1:09:25 time: 0.5596 data_time: 0.4182 memory: 6319 loss: 0.0775 +2023/06/05 04:36:10 - mmengine - INFO - Epoch(train) [9][1200/3907] lr: 1.6217e-05 eta: 1:08:23 time: 0.6044 data_time: 0.4525 memory: 6319 loss: 0.0811 +2023/06/05 04:37:08 - mmengine - INFO - Epoch(train) [9][1300/3907] lr: 1.6035e-05 eta: 1:07:20 time: 0.6150 data_time: 0.4695 memory: 6319 loss: 0.0680 +2023/06/05 04:38:10 - mmengine - INFO - Epoch(train) [9][1400/3907] lr: 1.5855e-05 eta: 1:06:18 time: 0.6378 data_time: 0.4838 memory: 6319 loss: 0.0654 +2023/06/05 04:39:08 - mmengine - INFO - Epoch(train) [9][1500/3907] lr: 1.5678e-05 eta: 1:05:15 time: 0.5450 data_time: 0.4008 memory: 6319 loss: 0.0785 +2023/06/05 04:40:05 - mmengine - INFO - Epoch(train) [9][1600/3907] lr: 1.5503e-05 eta: 1:04:12 time: 0.6424 data_time: 0.4750 memory: 6319 loss: 0.0752 +2023/06/05 04:41:05 - mmengine - INFO - Epoch(train) [9][1700/3907] lr: 1.5331e-05 eta: 1:03:10 time: 0.5732 data_time: 0.4319 memory: 6319 loss: 0.0764 +2023/06/05 04:41:32 - mmengine - INFO - Exp name: resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1_20230604_225317 +2023/06/05 04:42:03 - mmengine - INFO - Epoch(train) [9][1800/3907] lr: 1.5162e-05 eta: 1:02:07 time: 0.5952 data_time: 0.4367 memory: 6319 loss: 0.0812 +2023/06/05 04:43:04 - mmengine - INFO - Epoch(train) [9][1900/3907] lr: 1.4995e-05 eta: 1:01:05 time: 0.5903 data_time: 0.4457 memory: 6319 loss: 0.0740 +2023/06/05 04:44:03 - mmengine - INFO - Epoch(train) [9][2000/3907] lr: 1.4830e-05 eta: 1:00:02 time: 0.6237 data_time: 0.4651 memory: 6319 loss: 0.0756 +2023/06/05 04:45:04 - mmengine - INFO - Epoch(train) [9][2100/3907] lr: 1.4668e-05 eta: 0:59:00 time: 0.5542 data_time: 0.4104 memory: 6319 loss: 0.0783 +2023/06/05 04:46:04 - mmengine - INFO - Epoch(train) [9][2200/3907] lr: 1.4509e-05 eta: 0:57:58 time: 0.6137 data_time: 0.4575 memory: 6319 loss: 0.0677 +2023/06/05 04:47:03 - mmengine - INFO - Epoch(train) [9][2300/3907] lr: 1.4353e-05 eta: 0:56:55 time: 0.5726 data_time: 0.4299 memory: 6319 loss: 0.0775 +2023/06/05 04:48:03 - mmengine - INFO - Epoch(train) [9][2400/3907] lr: 1.4199e-05 eta: 0:55:53 time: 0.6064 data_time: 0.4645 memory: 6319 loss: 0.0789 +2023/06/05 04:49:03 - mmengine - INFO - Epoch(train) [9][2500/3907] lr: 1.4047e-05 eta: 0:54:51 time: 0.6004 data_time: 0.4596 memory: 6319 loss: 0.0895 +2023/06/05 04:50:01 - mmengine - INFO - Epoch(train) [9][2600/3907] lr: 1.3899e-05 eta: 0:53:48 time: 0.5478 data_time: 0.4055 memory: 6319 loss: 0.0702 +2023/06/05 04:51:03 - mmengine - INFO - Epoch(train) [9][2700/3907] lr: 1.3753e-05 eta: 0:52:46 time: 0.5985 data_time: 0.4569 memory: 6319 loss: 0.0654 +2023/06/05 04:51:27 - mmengine - INFO - Exp name: resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1_20230604_225317 +2023/06/05 04:52:03 - mmengine - INFO - Epoch(train) [9][2800/3907] lr: 1.3609e-05 eta: 0:51:44 time: 0.6840 data_time: 0.5412 memory: 6319 loss: 0.0702 +2023/06/05 04:53:02 - mmengine - INFO - Epoch(train) [9][2900/3907] lr: 1.3469e-05 eta: 0:50:42 time: 0.5631 data_time: 0.4203 memory: 6319 loss: 0.0635 +2023/06/05 04:54:03 - mmengine - INFO - Epoch(train) [9][3000/3907] lr: 1.3331e-05 eta: 0:49:40 time: 0.5699 data_time: 0.4271 memory: 6319 loss: 0.0938 +2023/06/05 04:55:03 - mmengine - INFO - Epoch(train) [9][3100/3907] lr: 1.3196e-05 eta: 0:48:38 time: 0.6232 data_time: 0.4813 memory: 6319 loss: 0.0716 +2023/06/05 04:56:01 - mmengine - INFO - Epoch(train) [9][3200/3907] lr: 1.3063e-05 eta: 0:47:35 time: 0.5956 data_time: 0.4542 memory: 6319 loss: 0.0660 +2023/06/05 04:57:01 - mmengine - INFO - Epoch(train) [9][3300/3907] lr: 1.2933e-05 eta: 0:46:33 time: 0.5846 data_time: 0.4255 memory: 6319 loss: 0.0805 +2023/06/05 04:58:02 - mmengine - INFO - Epoch(train) [9][3400/3907] lr: 1.2806e-05 eta: 0:45:31 time: 0.5871 data_time: 0.4452 memory: 6319 loss: 0.0723 +2023/06/05 04:59:02 - mmengine - INFO - Epoch(train) [9][3500/3907] lr: 1.2682e-05 eta: 0:44:29 time: 0.5854 data_time: 0.4433 memory: 6319 loss: 0.0767 +2023/06/05 05:00:01 - mmengine - INFO - Epoch(train) [9][3600/3907] lr: 1.2560e-05 eta: 0:43:27 time: 0.6187 data_time: 0.4758 memory: 6319 loss: 0.0802 +2023/06/05 05:01:05 - mmengine - INFO - Epoch(train) [9][3700/3907] lr: 1.2441e-05 eta: 0:42:25 time: 0.6557 data_time: 0.5139 memory: 6319 loss: 0.0782 +2023/06/05 05:01:30 - mmengine - INFO - Exp name: resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1_20230604_225317 +2023/06/05 05:02:03 - mmengine - INFO - Epoch(train) [9][3800/3907] lr: 1.2325e-05 eta: 0:41:23 time: 0.5939 data_time: 0.4509 memory: 6319 loss: 0.0830 +2023/06/05 05:03:03 - mmengine - INFO - Epoch(train) [9][3900/3907] lr: 1.2211e-05 eta: 0:40:21 time: 0.6212 data_time: 0.4791 memory: 6319 loss: 0.0770 +2023/06/05 05:03:05 - mmengine - INFO - Exp name: resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1_20230604_225317 +2023/06/05 05:03:05 - mmengine - INFO - Saving checkpoint at 9 epochs +2023/06/05 05:03:52 - mmengine - INFO - Epoch(val) [9][57/57] accuracy/top1: 96.9834 data_time: 0.5080 time: 0.5982 +2023/06/05 05:04:52 - mmengine - INFO - Epoch(train) [10][ 100/3907] lr: 1.2093e-05 eta: 0:39:14 time: 0.5621 data_time: 0.4199 memory: 6319 loss: 0.0748 +2023/06/05 05:05:51 - mmengine - INFO - Epoch(train) [10][ 200/3907] lr: 1.1985e-05 eta: 0:38:12 time: 0.5691 data_time: 0.4259 memory: 6319 loss: 0.0837 +2023/06/05 05:06:51 - mmengine - INFO - Epoch(train) [10][ 300/3907] lr: 1.1881e-05 eta: 0:37:10 time: 0.6307 data_time: 0.4882 memory: 6319 loss: 0.0876 +2023/06/05 05:07:52 - mmengine - INFO - Epoch(train) [10][ 400/3907] lr: 1.1778e-05 eta: 0:36:08 time: 0.5951 data_time: 0.4422 memory: 6319 loss: 0.0720 +2023/06/05 05:08:53 - mmengine - INFO - Epoch(train) [10][ 500/3907] lr: 1.1679e-05 eta: 0:35:06 time: 0.5566 data_time: 0.4146 memory: 6319 loss: 0.0775 +2023/06/05 05:09:53 - mmengine - INFO - Epoch(train) [10][ 600/3907] lr: 1.1583e-05 eta: 0:34:04 time: 0.6031 data_time: 0.4592 memory: 6319 loss: 0.0647 +2023/06/05 05:10:50 - mmengine - INFO - Epoch(train) [10][ 700/3907] lr: 1.1489e-05 eta: 0:33:02 time: 0.5468 data_time: 0.4054 memory: 6319 loss: 0.0781 +2023/06/05 05:11:47 - mmengine - INFO - Epoch(train) [10][ 800/3907] lr: 1.1398e-05 eta: 0:32:00 time: 0.6019 data_time: 0.4425 memory: 6319 loss: 0.0828 +2023/06/05 05:12:12 - mmengine - INFO - Exp name: resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1_20230604_225317 +2023/06/05 05:12:49 - mmengine - INFO - Epoch(train) [10][ 900/3907] lr: 1.1310e-05 eta: 0:30:58 time: 0.6152 data_time: 0.4726 memory: 6319 loss: 0.0835 +2023/06/05 05:13:48 - mmengine - INFO - Epoch(train) [10][1000/3907] lr: 1.1225e-05 eta: 0:29:56 time: 0.6446 data_time: 0.4820 memory: 6319 loss: 0.0697 +2023/06/05 05:14:49 - mmengine - INFO - Epoch(train) [10][1100/3907] lr: 1.1142e-05 eta: 0:28:54 time: 0.5972 data_time: 0.4400 memory: 6319 loss: 0.0900 +2023/06/05 05:15:50 - mmengine - INFO - Epoch(train) [10][1200/3907] lr: 1.1063e-05 eta: 0:27:52 time: 0.5879 data_time: 0.4460 memory: 6319 loss: 0.0645 +2023/06/05 05:16:49 - mmengine - INFO - Epoch(train) [10][1300/3907] lr: 1.0986e-05 eta: 0:26:50 time: 0.6340 data_time: 0.4896 memory: 6319 loss: 0.0746 +2023/06/05 05:17:49 - mmengine - INFO - Epoch(train) [10][1400/3907] lr: 1.0912e-05 eta: 0:25:48 time: 0.5552 data_time: 0.3942 memory: 6319 loss: 0.0850 +2023/06/05 05:18:45 - mmengine - INFO - Epoch(train) [10][1500/3907] lr: 1.0841e-05 eta: 0:24:46 time: 0.6212 data_time: 0.4789 memory: 6319 loss: 0.0711 +2023/06/05 05:19:44 - mmengine - INFO - Epoch(train) [10][1600/3907] lr: 1.0773e-05 eta: 0:23:44 time: 0.6043 data_time: 0.4616 memory: 6319 loss: 0.0719 +2023/06/05 05:20:42 - mmengine - INFO - Epoch(train) [10][1700/3907] lr: 1.0707e-05 eta: 0:22:42 time: 0.6120 data_time: 0.4548 memory: 6319 loss: 0.0703 +2023/06/05 05:21:37 - mmengine - INFO - Epoch(train) [10][1800/3907] lr: 1.0645e-05 eta: 0:21:40 time: 0.5521 data_time: 0.3992 memory: 6319 loss: 0.0715 +2023/06/05 05:22:00 - mmengine - INFO - Exp name: resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1_20230604_225317 +2023/06/05 05:22:35 - mmengine - INFO - Epoch(train) [10][1900/3907] lr: 1.0585e-05 eta: 0:20:38 time: 0.6041 data_time: 0.4598 memory: 6319 loss: 0.0677 +2023/06/05 05:23:35 - mmengine - INFO - Epoch(train) [10][2000/3907] lr: 1.0529e-05 eta: 0:19:36 time: 0.5876 data_time: 0.4464 memory: 6319 loss: 0.0760 +2023/06/05 05:24:35 - mmengine - INFO - Epoch(train) [10][2100/3907] lr: 1.0475e-05 eta: 0:18:35 time: 0.6209 data_time: 0.4786 memory: 6319 loss: 0.0667 +2023/06/05 05:25:34 - mmengine - INFO - Epoch(train) [10][2200/3907] lr: 1.0424e-05 eta: 0:17:33 time: 0.5541 data_time: 0.4120 memory: 6319 loss: 0.0729 +2023/06/05 05:26:31 - mmengine - INFO - Epoch(train) [10][2300/3907] lr: 1.0376e-05 eta: 0:16:31 time: 0.6067 data_time: 0.4626 memory: 6319 loss: 0.0744 +2023/06/05 05:27:30 - mmengine - INFO - Epoch(train) [10][2400/3907] lr: 1.0330e-05 eta: 0:15:29 time: 0.5708 data_time: 0.4272 memory: 6319 loss: 0.0633 +2023/06/05 05:28:31 - mmengine - INFO - Epoch(train) [10][2500/3907] lr: 1.0288e-05 eta: 0:14:27 time: 0.6081 data_time: 0.4651 memory: 6319 loss: 0.0665 +2023/06/05 05:29:29 - mmengine - INFO - Epoch(train) [10][2600/3907] lr: 1.0249e-05 eta: 0:13:25 time: 0.5578 data_time: 0.4137 memory: 6319 loss: 0.0777 +2023/06/05 05:30:27 - mmengine - INFO - Epoch(train) [10][2700/3907] lr: 1.0212e-05 eta: 0:12:24 time: 0.5640 data_time: 0.4224 memory: 6319 loss: 0.0665 +2023/06/05 05:31:27 - mmengine - INFO - Epoch(train) [10][2800/3907] lr: 1.0178e-05 eta: 0:11:22 time: 0.5590 data_time: 0.4165 memory: 6319 loss: 0.0783 +2023/06/05 05:31:46 - mmengine - INFO - Exp name: resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1_20230604_225317 +2023/06/05 05:32:29 - mmengine - INFO - Epoch(train) [10][2900/3907] lr: 1.0148e-05 eta: 0:10:20 time: 0.6525 data_time: 0.5093 memory: 6319 loss: 0.0665 +2023/06/05 05:33:32 - mmengine - INFO - Epoch(train) [10][3000/3907] lr: 1.0120e-05 eta: 0:09:19 time: 0.5783 data_time: 0.4358 memory: 6319 loss: 0.0697 +2023/06/05 05:34:33 - mmengine - INFO - Epoch(train) [10][3100/3907] lr: 1.0095e-05 eta: 0:08:17 time: 0.6042 data_time: 0.4608 memory: 6319 loss: 0.0807 +2023/06/05 05:35:28 - mmengine - INFO - Epoch(train) [10][3200/3907] lr: 1.0073e-05 eta: 0:07:15 time: 0.5526 data_time: 0.4108 memory: 6319 loss: 0.0778 +2023/06/05 05:36:28 - mmengine - INFO - Epoch(train) [10][3300/3907] lr: 1.0054e-05 eta: 0:06:14 time: 0.5733 data_time: 0.4312 memory: 6319 loss: 0.0591 +2023/06/05 05:37:28 - mmengine - INFO - Epoch(train) [10][3400/3907] lr: 1.0038e-05 eta: 0:05:12 time: 0.6162 data_time: 0.4748 memory: 6319 loss: 0.0725 +2023/06/05 05:38:28 - mmengine - INFO - Epoch(train) [10][3500/3907] lr: 1.0024e-05 eta: 0:04:10 time: 0.6028 data_time: 0.4609 memory: 6319 loss: 0.0655 +2023/06/05 05:39:28 - mmengine - INFO - Epoch(train) [10][3600/3907] lr: 1.0014e-05 eta: 0:03:09 time: 0.5531 data_time: 0.4115 memory: 6319 loss: 0.0963 +2023/06/05 05:40:29 - mmengine - INFO - Epoch(train) [10][3700/3907] lr: 1.0006e-05 eta: 0:02:07 time: 0.6058 data_time: 0.4628 memory: 6319 loss: 0.0800 +2023/06/05 05:41:28 - mmengine - INFO - Epoch(train) [10][3800/3907] lr: 1.0002e-05 eta: 0:01:05 time: 0.6612 data_time: 0.5197 memory: 6319 loss: 0.0754 +2023/06/05 05:41:50 - mmengine - INFO - Exp name: resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1_20230604_225317 +2023/06/05 05:42:26 - mmengine - INFO - Epoch(train) [10][3900/3907] lr: 1.0000e-05 eta: 0:00:04 time: 0.5697 data_time: 0.4274 memory: 6319 loss: 0.0756 +2023/06/05 05:42:31 - mmengine - INFO - Exp name: resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1_20230604_225317 +2023/06/05 05:42:31 - mmengine - INFO - Saving checkpoint at 10 epochs +2023/06/05 05:43:17 - mmengine - INFO - Epoch(val) [10][57/57] accuracy/top1: 97.0460 data_time: 0.4989 time: 0.5858 diff --git a/resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1/20230604_225317/vis_data/20230604_225317.json b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1/20230604_225317/vis_data/20230604_225317.json new file mode 100644 index 0000000000000000000000000000000000000000..ebe3ff4317681eded55cb9cbd32f881b168bdcb1 --- /dev/null +++ b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1/20230604_225317/vis_data/20230604_225317.json @@ -0,0 +1,400 @@ +{"lr": 9.999857418382631e-05, "data_time": 0.04329502582550049, "loss": 0.6144014835357666, "time": 0.5942174673080445, "epoch": 1, "memory": 9436, "step": 100} +{"lr": 9.999423907359506e-05, "data_time": 0.0017529964447021485, "loss": 0.540549498796463, "time": 0.6243215084075928, "epoch": 1, "memory": 6319, "step": 200} +{"lr": 9.998699480412073e-05, "data_time": 0.001414918899536133, "loss": 0.48605347573757174, "time": 0.6351176261901855, "epoch": 1, "memory": 6319, "step": 300} +{"lr": 9.997684184379085e-05, "data_time": 0.0015470027923583985, "loss": 0.4411657601594925, "time": 0.5894553184509277, "epoch": 1, "memory": 6319, "step": 400} +{"lr": 9.996378084905784e-05, "data_time": 0.0015391826629638672, "loss": 0.3897904396057129, "time": 0.5979857444763184, "epoch": 1, "memory": 6319, "step": 500} +{"lr": 9.994781266439667e-05, "data_time": 0.0010723114013671876, "loss": 0.3695789337158203, "time": 0.6539291620254517, "epoch": 1, "memory": 6319, "step": 600} +{"lr": 9.992893832225061e-05, "data_time": 0.0011087179183959961, "loss": 0.33060954213142396, "time": 0.6491123914718628, "epoch": 1, "memory": 6319, "step": 700} +{"lr": 9.990715904296405e-05, "data_time": 0.0016474723815917969, "loss": 0.3094102323055267, "time": 0.5923159122467041, "epoch": 1, "memory": 6319, "step": 800} +{"lr": 9.988247623470371e-05, "data_time": 0.000794672966003418, "loss": 0.29671770334243774, "time": 0.6268330574035644, "epoch": 1, "memory": 6319, "step": 900} +{"lr": 9.985489149336749e-05, "data_time": 0.001099562644958496, "loss": 0.2709860235452652, "time": 0.6295077800750732, "epoch": 1, "memory": 6319, "step": 1000} +{"lr": 9.982440660248187e-05, "data_time": 0.0008476495742797852, "loss": 0.27203600108623505, "time": 0.6169278621673584, "epoch": 1, "memory": 6319, "step": 1100} +{"lr": 9.979102353308574e-05, "data_time": 0.0008696317672729492, "loss": 0.2540829986333847, "time": 0.633787989616394, "epoch": 1, "memory": 6319, "step": 1200} +{"lr": 9.975474444360358e-05, "data_time": 0.0011459827423095704, "loss": 0.23767077326774597, "time": 0.7133777379989624, "epoch": 1, "memory": 6319, "step": 1300} +{"lr": 9.97155716797056e-05, "data_time": 0.001165318489074707, "loss": 0.2348430186510086, "time": 0.5883892059326172, "epoch": 1, "memory": 6319, "step": 1400} +{"lr": 9.967350777415632e-05, "data_time": 0.0010182857513427734, "loss": 0.2445591375231743, "time": 0.6163933038711548, "epoch": 1, "memory": 6319, "step": 1500} +{"lr": 9.962855544665059e-05, "data_time": 0.001074671745300293, "loss": 0.2130959838628769, "time": 0.6496816873550415, "epoch": 1, "memory": 6319, "step": 1600} +{"lr": 9.958071760363773e-05, "data_time": 0.0015396356582641601, "loss": 0.20838867425918578, "time": 0.639490270614624, "epoch": 1, "memory": 6319, "step": 1700} +{"lr": 9.952999733813382e-05, "data_time": 0.0012651205062866211, "loss": 0.20887960642576217, "time": 0.6244978427886962, "epoch": 1, "memory": 6319, "step": 1800} +{"lr": 9.947639792952148e-05, "data_time": 0.0009616613388061523, "loss": 0.2113117352128029, "time": 0.610913872718811, "epoch": 1, "memory": 6319, "step": 1900} +{"lr": 9.941992284333813e-05, "data_time": 0.001140928268432617, "loss": 0.20041645765304567, "time": 0.6053268194198609, "epoch": 1, "memory": 6319, "step": 2000} +{"lr": 9.93605757310516e-05, "data_time": 0.0008172035217285156, "loss": 0.19543420225381852, "time": 0.6133771657943725, "epoch": 1, "memory": 6319, "step": 2100} +{"lr": 9.929836042982421e-05, "data_time": 0.001177501678466797, "loss": 0.18500066697597503, "time": 0.6748033761978149, "epoch": 1, "memory": 6319, "step": 2200} +{"lr": 9.923328096226467e-05, "data_time": 0.0011435508728027343, "loss": 0.17640187293291093, "time": 0.6820239067077637, "epoch": 1, "memory": 6319, "step": 2300} +{"lr": 9.916534153616819e-05, "data_time": 0.0010052919387817383, "loss": 0.17090136110782622, "time": 0.6188471555709839, "epoch": 1, "memory": 6319, "step": 2400} +{"lr": 9.909454654424387e-05, "data_time": 0.0013057708740234375, "loss": 0.17705828100442886, "time": 0.6501886367797851, "epoch": 1, "memory": 6319, "step": 2500} +{"lr": 9.902090056383109e-05, "data_time": 0.0010050058364868164, "loss": 0.19458670020103455, "time": 0.6512877702713012, "epoch": 1, "memory": 6319, "step": 2600} +{"lr": 9.894440835660365e-05, "data_time": 0.0013167381286621094, "loss": 0.17685725688934326, "time": 0.6358255624771119, "epoch": 1, "memory": 6319, "step": 2700} +{"lr": 9.886507486826162e-05, "data_time": 0.0009731054306030273, "loss": 0.1762370079755783, "time": 0.6680647611618042, "epoch": 1, "memory": 6319, "step": 2800} +{"lr": 9.878290522821154e-05, "data_time": 0.0008811235427856445, "loss": 0.16809552758932114, "time": 0.6001085996627807, "epoch": 1, "memory": 6319, "step": 2900} +{"lr": 9.86979047492352e-05, "data_time": 0.0014863252639770509, "loss": 0.17183809280395507, "time": 0.6313298463821411, "epoch": 1, "memory": 6319, "step": 3000} +{"lr": 9.861007892714578e-05, "data_time": 0.0008005142211914062, "loss": 0.16417725533246993, "time": 0.6330173254013062, "epoch": 1, "memory": 6319, "step": 3100} +{"lr": 9.851943344043239e-05, "data_time": 0.001177072525024414, "loss": 0.17803086191415787, "time": 0.6067479848861694, "epoch": 1, "memory": 6319, "step": 3200} +{"lr": 9.842597414989319e-05, "data_time": 0.0010962724685668946, "loss": 0.14825334995985032, "time": 0.616440725326538, "epoch": 1, "memory": 6319, "step": 3300} +{"lr": 9.832970709825637e-05, "data_time": 0.0012169837951660155, "loss": 0.15710122585296632, "time": 0.656633996963501, "epoch": 1, "memory": 6319, "step": 3400} +{"lr": 9.823063850978968e-05, "data_time": 0.001428532600402832, "loss": 0.14660579934716225, "time": 0.5866067171096802, "epoch": 1, "memory": 6319, "step": 3500} +{"lr": 9.81287747898975e-05, "data_time": 0.001101231575012207, "loss": 0.16227754950523376, "time": 0.6107003211975097, "epoch": 1, "memory": 6319, "step": 3600} +{"lr": 9.802412252470715e-05, "data_time": 0.001011037826538086, "loss": 0.14880369156599044, "time": 0.6640010595321655, "epoch": 1, "memory": 6319, "step": 3700} +{"lr": 9.791668848064258e-05, "data_time": 0.0011841058731079102, "loss": 0.14143735468387603, "time": 0.6097638368606567, "epoch": 1, "memory": 6319, "step": 3800} +{"lr": 9.780647960398751e-05, "data_time": 0.0008812665939331055, "loss": 0.1573878437280655, "time": 0.6231289625167846, "epoch": 1, "memory": 6319, "step": 3900} +{"accuracy/top1": 80.42169952392578, "data_time": 0.6049195339805201, "time": 0.6992707377985904, "step": 1} +{"lr": 9.7685491196646e-05, "data_time": 0.0012707710266113281, "loss": 0.14969086796045303, "time": 0.6566398620605469, "epoch": 2, "memory": 6319, "step": 4007} +{"lr": 9.756956126085676e-05, "data_time": 0.0014353275299072265, "loss": 0.1544960506260395, "time": 0.6339057445526123, "epoch": 2, "memory": 6319, "step": 4107} +{"lr": 9.745087893642482e-05, "data_time": 0.0010685920715332031, "loss": 0.15404153391718864, "time": 0.6071049690246582, "epoch": 2, "memory": 6319, "step": 4207} +{"lr": 9.73294518969056e-05, "data_time": 0.0011847972869873046, "loss": 0.13772937953472136, "time": 0.6508279800415039, "epoch": 2, "memory": 6319, "step": 4307} +{"lr": 9.72052879933172e-05, "data_time": 0.0010708093643188477, "loss": 0.14833975285291673, "time": 0.6094993352890015, "epoch": 2, "memory": 6319, "step": 4407} +{"lr": 9.707839525363358e-05, "data_time": 0.0011272430419921875, "loss": 0.13168747276067733, "time": 0.6026272058486939, "epoch": 2, "memory": 6319, "step": 4507} +{"lr": 9.69487818822646e-05, "data_time": 0.001011037826538086, "loss": 0.15316015034914016, "time": 0.6099385261535645, "epoch": 2, "memory": 6319, "step": 4607} +{"lr": 9.681645625952616e-05, "data_time": 0.0012804508209228516, "loss": 0.13080811202526094, "time": 0.6574455738067627, "epoch": 2, "memory": 6319, "step": 4707} +{"lr": 9.668142694109826e-05, "data_time": 0.0012222766876220704, "loss": 0.1337391011416912, "time": 0.558815860748291, "epoch": 2, "memory": 6319, "step": 4807} +{"lr": 9.654370265747162e-05, "data_time": 0.0013068914413452148, "loss": 0.1303473137319088, "time": 0.6175517082214356, "epoch": 2, "memory": 6319, "step": 4907} +{"lr": 9.640329231338345e-05, "data_time": 0.0012620449066162109, "loss": 0.1228546142578125, "time": 0.6700607061386108, "epoch": 2, "memory": 6319, "step": 5007} +{"lr": 9.626020498724147e-05, "data_time": 0.001144719123840332, "loss": 0.1405038818717003, "time": 0.6171956300735474, "epoch": 2, "memory": 6319, "step": 5107} +{"lr": 9.611444993053717e-05, "data_time": 0.0016186952590942383, "loss": 0.13908120915293692, "time": 0.6053791284561157, "epoch": 2, "memory": 6319, "step": 5207} +{"lr": 9.596603656724746e-05, "data_time": 0.0009818315505981446, "loss": 0.12037205770611763, "time": 0.6270071983337402, "epoch": 2, "memory": 6319, "step": 5307} +{"lr": 9.58149744932254e-05, "data_time": 0.0011154413223266602, "loss": 0.14744246304035186, "time": 0.632693862915039, "epoch": 2, "memory": 6319, "step": 5407} +{"lr": 9.566127347557972e-05, "data_time": 0.001504063606262207, "loss": 0.12885030657052993, "time": 0.6522293090820312, "epoch": 2, "memory": 6319, "step": 5507} +{"lr": 9.550494345204367e-05, "data_time": 0.0012865066528320312, "loss": 0.12040070071816444, "time": 0.6179014205932617, "epoch": 2, "memory": 6319, "step": 5607} +{"lr": 9.534599453033183e-05, "data_time": 0.0012835979461669922, "loss": 0.11735203564167022, "time": 0.5667521238327027, "epoch": 2, "memory": 6319, "step": 5707} +{"lr": 9.518443698748712e-05, "data_time": 0.0011459112167358399, "loss": 0.12665610015392303, "time": 0.6008880376815796, "epoch": 2, "memory": 6319, "step": 5807} +{"lr": 9.50202812692162e-05, "data_time": 0.0010967254638671875, "loss": 0.1326982371509075, "time": 0.6913531303405762, "epoch": 2, "memory": 6319, "step": 5907} +{"lr": 9.485353798921386e-05, "data_time": 0.000941777229309082, "loss": 0.1312527395784855, "time": 0.5970057249069214, "epoch": 2, "memory": 6319, "step": 6007} +{"lr": 9.468421792847712e-05, "data_time": 0.0009882926940917968, "loss": 0.11886056512594223, "time": 0.6437604427337646, "epoch": 2, "memory": 6319, "step": 6107} +{"lr": 9.451233203460794e-05, "data_time": 0.001193690299987793, "loss": 0.12179807424545289, "time": 0.6047776699066162, "epoch": 2, "memory": 6319, "step": 6207} +{"lr": 9.433789142110532e-05, "data_time": 0.0008589029312133789, "loss": 0.12132530137896538, "time": 0.6325850009918212, "epoch": 2, "memory": 6319, "step": 6307} +{"lr": 9.416090736664709e-05, "data_time": 0.0008988857269287109, "loss": 0.1269269645214081, "time": 0.6048826694488525, "epoch": 2, "memory": 6319, "step": 6407} +{"lr": 9.398139131436032e-05, "data_time": 0.0009505748748779297, "loss": 0.10670752972364425, "time": 0.6421198368072509, "epoch": 2, "memory": 6319, "step": 6507} +{"lr": 9.379935487108167e-05, "data_time": 0.0009618759155273438, "loss": 0.12039944976568222, "time": 0.600836992263794, "epoch": 2, "memory": 6319, "step": 6607} +{"lr": 9.361480980660697e-05, "data_time": 0.024141836166381835, "loss": 0.13166357204318047, "time": 0.5980572462081909, "epoch": 2, "memory": 6319, "step": 6707} +{"lr": 9.342776805292993e-05, "data_time": 0.0010269880294799805, "loss": 0.11618372052907944, "time": 0.5672625541687012, "epoch": 2, "memory": 6319, "step": 6807} +{"lr": 9.323824170347083e-05, "data_time": 0.0013018369674682618, "loss": 0.12470941171050072, "time": 0.5922423839569092, "epoch": 2, "memory": 6319, "step": 6907} +{"lr": 9.304624301229471e-05, "data_time": 0.0011385202407836914, "loss": 0.09987586140632629, "time": 0.6147332191467285, "epoch": 2, "memory": 6319, "step": 7007} +{"lr": 9.285178439331897e-05, "data_time": 0.0015858173370361327, "loss": 0.1218492977321148, "time": 0.6371518611907959, "epoch": 2, "memory": 6319, "step": 7107} +{"lr": 9.265487841951087e-05, "data_time": 0.0012993812561035156, "loss": 0.11031495705246926, "time": 0.6388080835342407, "epoch": 2, "memory": 6319, "step": 7207} +{"lr": 9.245553782207421e-05, "data_time": 0.0012434720993041992, "loss": 0.11798710525035858, "time": 0.6172866344451904, "epoch": 2, "memory": 6319, "step": 7307} +{"lr": 9.225377548962671e-05, "data_time": 0.0010834455490112305, "loss": 0.1123718038201332, "time": 0.6374030828475952, "epoch": 2, "memory": 6319, "step": 7407} +{"lr": 9.204960446736601e-05, "data_time": 0.0010797739028930663, "loss": 0.12735430300235748, "time": 0.6130917787551879, "epoch": 2, "memory": 6319, "step": 7507} +{"lr": 9.184303795622715e-05, "data_time": 0.0013530254364013672, "loss": 0.12367594540119171, "time": 0.5850515604019165, "epoch": 2, "memory": 6319, "step": 7607} +{"lr": 9.163408931202823e-05, "data_time": 0.0009424686431884766, "loss": 0.109854356944561, "time": 0.55736825466156, "epoch": 2, "memory": 6319, "step": 7707} +{"lr": 9.142277204460729e-05, "data_time": 0.0008727788925170898, "loss": 0.1303892105817795, "time": 0.5944935798645019, "epoch": 2, "memory": 6319, "step": 7807} +{"accuracy/top1": 78.64723205566406, "data_time": 0.5207227797343813, "time": 0.6123008316960828, "step": 2} +{"lr": 9.119405492383288e-05, "data_time": 0.15969440937042237, "loss": 0.10292618237435817, "time": 0.644035029411316, "epoch": 3, "memory": 6319, "step": 7914} +{"lr": 9.097787819233205e-05, "data_time": 0.001564788818359375, "loss": 0.11451506689190864, "time": 0.5979749441146851, "epoch": 3, "memory": 6319, "step": 8014} +{"lr": 9.075937526576627e-05, "data_time": 0.0013667106628417968, "loss": 0.10959412232041359, "time": 0.6024105310440063, "epoch": 3, "memory": 6319, "step": 8114} +{"lr": 9.053856027171755e-05, "data_time": 0.0014506101608276368, "loss": 0.1103747770190239, "time": 0.5998272657394409, "epoch": 3, "memory": 6319, "step": 8214} +{"lr": 9.031544748725785e-05, "data_time": 0.0011715888977050781, "loss": 0.0986387774348259, "time": 0.6282076597213745, "epoch": 3, "memory": 6319, "step": 8314} +{"lr": 9.009005133802542e-05, "data_time": 0.0011480093002319337, "loss": 0.11372050642967224, "time": 0.6023003101348877, "epoch": 3, "memory": 6319, "step": 8414} +{"lr": 8.986238639729242e-05, "data_time": 0.0017016172409057618, "loss": 0.10751230046153068, "time": 0.625498366355896, "epoch": 3, "memory": 6319, "step": 8514} +{"lr": 8.963246738502254e-05, "data_time": 0.0009747982025146484, "loss": 0.11333387643098831, "time": 0.5796732902526855, "epoch": 3, "memory": 6319, "step": 8614} +{"lr": 8.940030916691928e-05, "data_time": 0.00117340087890625, "loss": 0.11304366365075111, "time": 0.5906398296356201, "epoch": 3, "memory": 6319, "step": 8714} +{"lr": 8.91659267534649e-05, "data_time": 0.001058197021484375, "loss": 0.1090011790394783, "time": 0.5973772287368775, "epoch": 3, "memory": 6319, "step": 8814} +{"lr": 8.892933529894978e-05, "data_time": 0.0010092020034790038, "loss": 0.1150609865784645, "time": 0.6059458971023559, "epoch": 3, "memory": 6319, "step": 8914} +{"lr": 8.869055010049283e-05, "data_time": 0.001197481155395508, "loss": 0.0966479279100895, "time": 0.6162628173828125, "epoch": 3, "memory": 6319, "step": 9014} +{"lr": 8.844958659705188e-05, "data_time": 0.0010462284088134765, "loss": 0.11113687679171562, "time": 0.6096754789352417, "epoch": 3, "memory": 6319, "step": 9114} +{"lr": 8.82064603684259e-05, "data_time": 0.0011551618576049806, "loss": 0.10108362510800362, "time": 0.6459286689758301, "epoch": 3, "memory": 6319, "step": 9214} +{"lr": 8.796118713424776e-05, "data_time": 0.0009507179260253906, "loss": 0.09558529742062091, "time": 0.5573125123977661, "epoch": 3, "memory": 6319, "step": 9314} +{"lr": 8.77137827529676e-05, "data_time": 0.0015813350677490235, "loss": 0.10445983856916427, "time": 0.6569721221923828, "epoch": 3, "memory": 6319, "step": 9414} +{"lr": 8.746426322082745e-05, "data_time": 0.00175018310546875, "loss": 0.09164994284510612, "time": 0.6291789770126343, "epoch": 3, "memory": 6319, "step": 9514} +{"lr": 8.721264467082732e-05, "data_time": 0.001194596290588379, "loss": 0.09945103079080582, "time": 0.6127496957778931, "epoch": 3, "memory": 6319, "step": 9614} +{"lr": 8.695894337168168e-05, "data_time": 0.0013375282287597656, "loss": 0.09632962122559548, "time": 0.6105611085891723, "epoch": 3, "memory": 6319, "step": 9714} +{"lr": 8.67031757267677e-05, "data_time": 0.0013091325759887694, "loss": 0.09921496212482453, "time": 0.6229627847671508, "epoch": 3, "memory": 6319, "step": 9814} +{"lr": 8.64453582730651e-05, "data_time": 0.001412367820739746, "loss": 0.10290349721908569, "time": 0.6016729831695556, "epoch": 3, "memory": 6319, "step": 9914} +{"lr": 8.61855076800861e-05, "data_time": 0.0011766910552978515, "loss": 0.09510720074176789, "time": 0.6402531623840332, "epoch": 3, "memory": 6319, "step": 10014} +{"lr": 8.592364074879843e-05, "data_time": 0.001172351837158203, "loss": 0.10642112605273724, "time": 0.6265649795532227, "epoch": 3, "memory": 6319, "step": 10114} +{"lr": 8.565977441053854e-05, "data_time": 0.0015697240829467773, "loss": 0.11050756648182869, "time": 0.779146385192871, "epoch": 3, "memory": 6319, "step": 10214} +{"lr": 8.539392572591724e-05, "data_time": 0.0010590076446533204, "loss": 0.10465100929141044, "time": 0.5467787504196167, "epoch": 3, "memory": 6319, "step": 10314} +{"lr": 8.512611188371633e-05, "data_time": 0.0012309551239013672, "loss": 0.10303699672222137, "time": 0.5611342430114746, "epoch": 3, "memory": 6319, "step": 10414} +{"lr": 8.485635019977725e-05, "data_time": 0.0011582136154174804, "loss": 0.10508915334939957, "time": 0.5413308620452881, "epoch": 3, "memory": 6319, "step": 10514} +{"lr": 8.45846581158818e-05, "data_time": 0.0011284828186035156, "loss": 0.10682142674922943, "time": 0.6266285657882691, "epoch": 3, "memory": 6319, "step": 10614} +{"lr": 8.431105319862416e-05, "data_time": 0.0010250329971313477, "loss": 0.08967098034918308, "time": 0.6155362606048584, "epoch": 3, "memory": 6319, "step": 10714} +{"lr": 8.403555313827516e-05, "data_time": 0.0010985374450683595, "loss": 0.10303728207945824, "time": 0.6193816900253296, "epoch": 3, "memory": 6319, "step": 10814} +{"lr": 8.375817574763844e-05, "data_time": 0.001214146614074707, "loss": 0.09502370059490203, "time": 0.5690989971160889, "epoch": 3, "memory": 6319, "step": 10914} +{"lr": 8.347893896089871e-05, "data_time": 0.001226329803466797, "loss": 0.0937697671353817, "time": 0.5626082420349121, "epoch": 3, "memory": 6319, "step": 11014} +{"lr": 8.31978608324625e-05, "data_time": 0.0011416196823120118, "loss": 0.0888057790696621, "time": 0.5901943683624268, "epoch": 3, "memory": 6319, "step": 11114} +{"lr": 8.291495953579052e-05, "data_time": 0.0014118909835815429, "loss": 0.09242230579257012, "time": 0.6065904855728149, "epoch": 3, "memory": 6319, "step": 11214} +{"lr": 8.263025336222267e-05, "data_time": 0.0009230136871337891, "loss": 0.11187078952789306, "time": 0.8397389650344849, "epoch": 3, "memory": 6319, "step": 11314} +{"lr": 8.234376071979551e-05, "data_time": 0.000915980339050293, "loss": 0.1063428483903408, "time": 0.5562231540679932, "epoch": 3, "memory": 6319, "step": 11414} +{"lr": 8.205550013205187e-05, "data_time": 0.0012593984603881836, "loss": 0.10742152333259583, "time": 0.6229443073272705, "epoch": 3, "memory": 6319, "step": 11514} +{"lr": 8.17654902368435e-05, "data_time": 0.001011800765991211, "loss": 0.09773860052227974, "time": 0.5700891733169555, "epoch": 3, "memory": 6319, "step": 11614} +{"lr": 8.147374978512582e-05, "data_time": 0.0011054039001464843, "loss": 0.09565523415803909, "time": 0.619389533996582, "epoch": 3, "memory": 6319, "step": 11714} +{"accuracy/top1": 82.1822509765625, "data_time": 0.5461730874817947, "time": 0.6365683078765869, "step": 3} +{"lr": 8.115969237626578e-05, "data_time": 0.12131602764129638, "loss": 0.10533967837691308, "time": 0.5874639987945557, "epoch": 4, "memory": 6319, "step": 11821} +{"lr": 8.08644297337245e-05, "data_time": 0.2184462308883667, "loss": 0.09577877409756183, "time": 0.6172070026397705, "epoch": 4, "memory": 6319, "step": 11921} +{"lr": 8.056749479387651e-05, "data_time": 0.0014777660369873046, "loss": 0.0914388693869114, "time": 0.6325190544128418, "epoch": 4, "memory": 6319, "step": 12021} +{"lr": 8.026890675542457e-05, "data_time": 0.001353621482849121, "loss": 0.1017810121178627, "time": 0.5816426753997803, "epoch": 4, "memory": 6319, "step": 12121} +{"lr": 7.996868492395457e-05, "data_time": 0.0011809110641479493, "loss": 0.09612506739795208, "time": 0.6351016759872437, "epoch": 4, "memory": 6319, "step": 12221} +{"lr": 7.966684871068762e-05, "data_time": 0.0012735843658447266, "loss": 0.0993888571858406, "time": 0.6687037706375122, "epoch": 4, "memory": 6319, "step": 12321} +{"lr": 7.936341763122449e-05, "data_time": 0.0008375644683837891, "loss": 0.08615965768694878, "time": 0.6186165571212768, "epoch": 4, "memory": 6319, "step": 12421} +{"lr": 7.905841130428405e-05, "data_time": 0.0008159875869750977, "loss": 0.09248109087347985, "time": 0.5616381406784058, "epoch": 4, "memory": 6319, "step": 12521} +{"lr": 7.875184945043487e-05, "data_time": 0.0009601831436157227, "loss": 0.099121855199337, "time": 0.6511362314224243, "epoch": 4, "memory": 6319, "step": 12621} +{"lr": 7.844375189082012e-05, "data_time": 0.0011842489242553712, "loss": 0.0778490923345089, "time": 0.6310737371444702, "epoch": 4, "memory": 6319, "step": 12721} +{"lr": 7.813413854587572e-05, "data_time": 0.0009321212768554687, "loss": 0.1054179273545742, "time": 0.5926837205886841, "epoch": 4, "memory": 6319, "step": 12821} +{"lr": 7.782302943404291e-05, "data_time": 0.0012653827667236327, "loss": 0.105924541503191, "time": 0.6521082401275635, "epoch": 4, "memory": 6319, "step": 12921} +{"lr": 7.751044467047337e-05, "data_time": 0.0011621475219726562, "loss": 0.09382660947740078, "time": 0.6130581378936768, "epoch": 4, "memory": 6319, "step": 13021} +{"lr": 7.719640446572907e-05, "data_time": 0.0015146970748901368, "loss": 0.09103393107652664, "time": 0.6381123304367066, "epoch": 4, "memory": 6319, "step": 13121} +{"lr": 7.688092912447536e-05, "data_time": 0.0009727716445922851, "loss": 0.09927580878138542, "time": 0.615462851524353, "epoch": 4, "memory": 6319, "step": 13221} +{"lr": 7.656403904416794e-05, "data_time": 0.0011687040328979491, "loss": 0.099492472037673, "time": 0.636084508895874, "epoch": 4, "memory": 6319, "step": 13321} +{"lr": 7.624575471373434e-05, "data_time": 0.001276254653930664, "loss": 0.10581168681383132, "time": 0.5897547245025635, "epoch": 4, "memory": 6319, "step": 13421} +{"lr": 7.592609671224926e-05, "data_time": 0.0012411355972290039, "loss": 0.09347496926784515, "time": 0.6397350549697876, "epoch": 4, "memory": 6319, "step": 13521} +{"lr": 7.560508570760375e-05, "data_time": 0.0009140968322753906, "loss": 0.09461783841252328, "time": 0.5972483634948731, "epoch": 4, "memory": 6319, "step": 13621} +{"lr": 7.528274245516895e-05, "data_time": 0.0008981466293334961, "loss": 0.09843641594052314, "time": 0.6742244005203247, "epoch": 4, "memory": 6319, "step": 13721} +{"lr": 7.495908779645405e-05, "data_time": 0.0013815879821777344, "loss": 0.09268547669053077, "time": 0.6965377807617188, "epoch": 4, "memory": 6319, "step": 13821} +{"lr": 7.463414265775906e-05, "data_time": 0.0014081478118896484, "loss": 0.08534050807356834, "time": 0.6147332191467285, "epoch": 4, "memory": 6319, "step": 13921} +{"lr": 7.430792804882147e-05, "data_time": 0.0013270139694213866, "loss": 0.09435487315058708, "time": 0.6108735799789429, "epoch": 4, "memory": 6319, "step": 14021} +{"lr": 7.398046506145803e-05, "data_time": 0.0011119842529296875, "loss": 0.08499853499233723, "time": 0.6191097974777222, "epoch": 4, "memory": 6319, "step": 14121} +{"lr": 7.365177486820114e-05, "data_time": 0.0009569644927978516, "loss": 0.09503398910164833, "time": 0.6604063034057617, "epoch": 4, "memory": 6319, "step": 14221} +{"lr": 7.332187872092939e-05, "data_time": 0.0012288570404052734, "loss": 0.10102737545967103, "time": 0.6483676195144653, "epoch": 4, "memory": 6319, "step": 14321} +{"lr": 7.299079794949401e-05, "data_time": 0.0013399600982666015, "loss": 0.07072069868445396, "time": 0.6595494985580445, "epoch": 4, "memory": 6319, "step": 14421} +{"lr": 7.265855396033955e-05, "data_time": 0.0008251667022705078, "loss": 0.09817356318235397, "time": 0.6921434879302979, "epoch": 4, "memory": 6319, "step": 14521} +{"lr": 7.232516823512013e-05, "data_time": 0.001130533218383789, "loss": 0.08089447729289531, "time": 0.7042551755905151, "epoch": 4, "memory": 6319, "step": 14621} +{"lr": 7.199066232930995e-05, "data_time": 0.0008875131607055664, "loss": 0.08366571292281151, "time": 0.5800805568695069, "epoch": 4, "memory": 6319, "step": 14721} +{"lr": 7.165505787080997e-05, "data_time": 0.001098942756652832, "loss": 0.09997397139668465, "time": 0.6220454454421998, "epoch": 4, "memory": 6319, "step": 14821} +{"lr": 7.13183765585495e-05, "data_time": 0.0012734651565551758, "loss": 0.09645765498280526, "time": 0.6224927186965943, "epoch": 4, "memory": 6319, "step": 14921} +{"lr": 7.098064016108317e-05, "data_time": 0.0010340213775634766, "loss": 0.09494763761758804, "time": 0.6230634450912476, "epoch": 4, "memory": 6319, "step": 15021} +{"lr": 7.064187051518379e-05, "data_time": 0.0009942770004272461, "loss": 0.10012876093387604, "time": 0.6057145833969116, "epoch": 4, "memory": 6319, "step": 15121} +{"lr": 7.030208952442949e-05, "data_time": 0.0014056682586669922, "loss": 0.08946731351315976, "time": 0.6111350059509277, "epoch": 4, "memory": 6319, "step": 15221} +{"lr": 6.996131915778851e-05, "data_time": 0.0009274482727050781, "loss": 0.09231140837073326, "time": 0.6949593305587769, "epoch": 4, "memory": 6319, "step": 15321} +{"lr": 6.961958144819853e-05, "data_time": 0.001043248176574707, "loss": 0.0874828852713108, "time": 0.6230496644973755, "epoch": 4, "memory": 6319, "step": 15421} +{"lr": 6.927689849114193e-05, "data_time": 0.001140284538269043, "loss": 0.09214382246136665, "time": 0.6202146768569946, "epoch": 4, "memory": 6319, "step": 15521} +{"lr": 6.893329244321722e-05, "data_time": 0.0006139516830444336, "loss": 0.09554443918168545, "time": 0.5699955224990845, "epoch": 4, "memory": 6319, "step": 15621} +{"accuracy/top1": 89.25924682617188, "data_time": 0.5373759228607704, "time": 0.626487530511001, "step": 4} +{"lr": 6.856463687361792e-05, "data_time": 0.24332633018493652, "loss": 0.09055482372641563, "time": 0.6087357759475708, "epoch": 5, "memory": 6319, "step": 15728} +{"lr": 6.821919068470432e-05, "data_time": 0.0010714292526245116, "loss": 0.08013462312519551, "time": 0.6210313558578491, "epoch": 5, "memory": 6319, "step": 15828} +{"lr": 6.787288979235407e-05, "data_time": 0.0011994123458862304, "loss": 0.08548355773091317, "time": 0.5748008966445923, "epoch": 5, "memory": 6319, "step": 15928} +{"lr": 6.752575658708793e-05, "data_time": 0.0012803792953491211, "loss": 0.08600676655769349, "time": 0.6464286327362061, "epoch": 5, "memory": 6319, "step": 16028} +{"lr": 6.71778135132408e-05, "data_time": 0.0011774539947509766, "loss": 0.08509748727083206, "time": 0.7035858869552613, "epoch": 5, "memory": 6319, "step": 16128} +{"lr": 6.682908306751074e-05, "data_time": 0.0012784481048583984, "loss": 0.08347053341567516, "time": 0.5974216938018799, "epoch": 5, "memory": 6319, "step": 16228} +{"lr": 6.64795877975043e-05, "data_time": 0.0013874530792236327, "loss": 0.09401379153132439, "time": 0.6251525402069091, "epoch": 5, "memory": 6319, "step": 16328} +{"lr": 6.61293503002787e-05, "data_time": 0.0012379169464111328, "loss": 0.08163192085921764, "time": 0.7113333225250245, "epoch": 5, "memory": 6319, "step": 16428} +{"lr": 6.577839322088084e-05, "data_time": 0.0011114835739135741, "loss": 0.08945540450513363, "time": 0.5497198104858398, "epoch": 5, "memory": 6319, "step": 16528} +{"lr": 6.542673925088298e-05, "data_time": 0.0009516000747680664, "loss": 0.0858134638518095, "time": 0.6087232112884522, "epoch": 5, "memory": 6319, "step": 16628} +{"lr": 6.507441112691584e-05, "data_time": 0.0009476184844970703, "loss": 0.09904510974884033, "time": 0.5784173727035522, "epoch": 5, "memory": 6319, "step": 16728} +{"lr": 6.47214316291983e-05, "data_time": 0.0019447803497314453, "loss": 0.08314869739115238, "time": 0.6415282249450683, "epoch": 5, "memory": 6319, "step": 16828} +{"lr": 6.436782358006481e-05, "data_time": 0.0011305570602416991, "loss": 0.08982470408082008, "time": 0.6326637506484986, "epoch": 5, "memory": 6319, "step": 16928} +{"lr": 6.40136098424896e-05, "data_time": 0.0015105724334716797, "loss": 0.08660290613770485, "time": 0.5824389696121216, "epoch": 5, "memory": 6319, "step": 17028} +{"lr": 6.365881331860828e-05, "data_time": 0.0012520551681518555, "loss": 0.07760576084256172, "time": 0.6218335151672363, "epoch": 5, "memory": 6319, "step": 17128} +{"lr": 6.330345694823739e-05, "data_time": 0.0010503530502319336, "loss": 0.08954662829637527, "time": 0.6022094488143921, "epoch": 5, "memory": 6319, "step": 17228} +{"lr": 6.29475637073908e-05, "data_time": 0.0014076948165893555, "loss": 0.07359700575470925, "time": 0.6077304363250733, "epoch": 5, "memory": 6319, "step": 17328} +{"lr": 6.259115660679482e-05, "data_time": 0.0012317180633544921, "loss": 0.0839787632226944, "time": 0.6005590438842774, "epoch": 5, "memory": 6319, "step": 17428} +{"lr": 6.223425869039963e-05, "data_time": 0.001350569725036621, "loss": 0.0748495526611805, "time": 0.5905034065246582, "epoch": 5, "memory": 6319, "step": 17528} +{"lr": 6.187689303389006e-05, "data_time": 0.001175689697265625, "loss": 0.09200113415718078, "time": 0.6499512434005738, "epoch": 5, "memory": 6319, "step": 17628} +{"lr": 6.151908274319292e-05, "data_time": 0.0010364294052124024, "loss": 0.09374405071139336, "time": 0.7825910091400147, "epoch": 5, "memory": 6319, "step": 17728} +{"lr": 6.116085095298373e-05, "data_time": 0.0008411169052124024, "loss": 0.07329643815755844, "time": 0.5486382722854615, "epoch": 5, "memory": 6319, "step": 17828} +{"lr": 6.080222082519007e-05, "data_time": 0.0009914159774780274, "loss": 0.08581202402710915, "time": 0.6193802118301391, "epoch": 5, "memory": 6319, "step": 17928} +{"lr": 6.0443215547495147e-05, "data_time": 0.0008234262466430664, "loss": 0.0874987043440342, "time": 0.6451940298080444, "epoch": 5, "memory": 6319, "step": 18028} +{"lr": 6.0083858331837714e-05, "data_time": 0.0007999897003173828, "loss": 0.08538103513419629, "time": 0.6099106550216675, "epoch": 5, "memory": 6319, "step": 18128} +{"lr": 5.97241724129116e-05, "data_time": 0.0013352394104003905, "loss": 0.08670755960047245, "time": 0.61390061378479, "epoch": 5, "memory": 6319, "step": 18228} +{"lr": 5.936418104666328e-05, "data_time": 0.0012809276580810548, "loss": 0.09755421504378319, "time": 0.6391181945800781, "epoch": 5, "memory": 6319, "step": 18328} +{"lr": 5.9003907508788295e-05, "data_time": 0.0012899398803710937, "loss": 0.08166025988757611, "time": 0.6239294052124024, "epoch": 5, "memory": 6319, "step": 18428} +{"lr": 5.864337509322633e-05, "data_time": 0.001182866096496582, "loss": 0.08269152827560902, "time": 0.7048297643661499, "epoch": 5, "memory": 6319, "step": 18528} +{"lr": 5.8282607110655194e-05, "data_time": 0.0009640216827392578, "loss": 0.09006138816475869, "time": 0.6218681097030639, "epoch": 5, "memory": 6319, "step": 18628} +{"lr": 5.7921626886983564e-05, "data_time": 0.001384139060974121, "loss": 0.08800899088382722, "time": 0.6342753410339356, "epoch": 5, "memory": 6319, "step": 18728} +{"lr": 5.756045776184287e-05, "data_time": 0.0013786077499389649, "loss": 0.08197480849921704, "time": 0.6153100728988647, "epoch": 5, "memory": 6319, "step": 18828} +{"lr": 5.719912308707808e-05, "data_time": 0.0013013362884521484, "loss": 0.071322251111269, "time": 0.618512749671936, "epoch": 5, "memory": 6319, "step": 18928} +{"lr": 5.683764622523808e-05, "data_time": 0.000980710983276367, "loss": 0.06396436169743538, "time": 0.5832077741622925, "epoch": 5, "memory": 6319, "step": 19028} +{"lr": 5.647605054806502e-05, "data_time": 0.0015149831771850586, "loss": 0.07964058071374894, "time": 0.6258903741836548, "epoch": 5, "memory": 6319, "step": 19128} +{"lr": 5.6114359434983206e-05, "data_time": 0.001742386817932129, "loss": 0.07741358056664467, "time": 0.6207758665084839, "epoch": 5, "memory": 6319, "step": 19228} +{"lr": 5.575259627158745e-05, "data_time": 0.04519574642181397, "loss": 0.07444062791764736, "time": 0.6512256145477295, "epoch": 5, "memory": 6319, "step": 19328} +{"lr": 5.539078444813113e-05, "data_time": 0.0018009424209594726, "loss": 0.08687329590320587, "time": 0.6746395587921142, "epoch": 5, "memory": 6319, "step": 19428} +{"lr": 5.5028947358013735e-05, "data_time": 0.0007738590240478516, "loss": 0.08167134039103985, "time": 0.5811173439025878, "epoch": 5, "memory": 6319, "step": 19528} +{"accuracy/top1": 93.68498229980469, "data_time": 0.4931796049249583, "time": 0.5811454838719862, "step": 5} +{"lr": 5.464178020340108e-05, "data_time": 0.2116227149963379, "loss": 0.08711279481649399, "time": 0.5857839107513427, "epoch": 6, "memory": 6319, "step": 19635} +{"lr": 5.427996514795046e-05, "data_time": 0.05478310585021973, "loss": 0.09860560074448585, "time": 0.6000259399414063, "epoch": 6, "memory": 6319, "step": 19735} +{"lr": 5.3918196647260834e-05, "data_time": 0.0011778831481933593, "loss": 0.07608385905623435, "time": 0.5921300888061524, "epoch": 6, "memory": 6319, "step": 19835} +{"lr": 5.355649809193057e-05, "data_time": 0.0011354446411132812, "loss": 0.08391941040754318, "time": 0.5497043371200562, "epoch": 6, "memory": 6319, "step": 19935} +{"lr": 5.3194892868035765e-05, "data_time": 0.001302337646484375, "loss": 0.07991163469851018, "time": 0.5998502731323242, "epoch": 6, "memory": 6319, "step": 20035} +{"lr": 5.283340435561795e-05, "data_time": 0.0012693166732788085, "loss": 0.0871560662984848, "time": 0.577147650718689, "epoch": 6, "memory": 6319, "step": 20135} +{"lr": 5.2472055927172495e-05, "data_time": 0.00094451904296875, "loss": 0.07504497133195401, "time": 0.6361608505249023, "epoch": 6, "memory": 6319, "step": 20235} +{"lr": 5.2110870946137605e-05, "data_time": 0.0010939598083496093, "loss": 0.07807315681129694, "time": 0.6203902244567872, "epoch": 6, "memory": 6319, "step": 20335} +{"lr": 5.174987276538342e-05, "data_time": 0.0010441303253173827, "loss": 0.08019522801041604, "time": 0.639669394493103, "epoch": 6, "memory": 6319, "step": 20435} +{"lr": 5.1389084725702425e-05, "data_time": 0.0011540651321411133, "loss": 0.08820452354848385, "time": 0.5844098329544067, "epoch": 6, "memory": 6319, "step": 20535} +{"lr": 5.1028530154300045e-05, "data_time": 0.0009647369384765625, "loss": 0.08078234754502774, "time": 0.6114591121673584, "epoch": 6, "memory": 6319, "step": 20635} +{"lr": 5.066823236328663e-05, "data_time": 0.0011961221694946288, "loss": 0.06776001751422882, "time": 0.6076150178909302, "epoch": 6, "memory": 6319, "step": 20735} +{"lr": 5.030821464817006e-05, "data_time": 0.0010621070861816407, "loss": 0.07857253141701222, "time": 0.5769826889038085, "epoch": 6, "memory": 6319, "step": 20835} +{"lr": 4.99485002863494e-05, "data_time": 0.000817108154296875, "loss": 0.08330013453960419, "time": 0.6137124538421631, "epoch": 6, "memory": 6319, "step": 20935} +{"lr": 4.95891125356102e-05, "data_time": 0.0013603448867797851, "loss": 0.08146739304065705, "time": 0.7235180854797363, "epoch": 6, "memory": 6319, "step": 21035} +{"lr": 4.923007463262053e-05, "data_time": 0.0011154413223266602, "loss": 0.08484194241464138, "time": 0.6746965646743774, "epoch": 6, "memory": 6319, "step": 21135} +{"lr": 4.887140979142852e-05, "data_time": 0.0011101007461547852, "loss": 0.07396330647170543, "time": 0.8727160215377807, "epoch": 6, "memory": 6319, "step": 21235} +{"lr": 4.851314120196161e-05, "data_time": 0.0008193731307983398, "loss": 0.07735928781330585, "time": 0.5724152326583862, "epoch": 6, "memory": 6319, "step": 21335} +{"lr": 4.815529202852693e-05, "data_time": 0.0011681795120239257, "loss": 0.09563387855887413, "time": 0.6191658496856689, "epoch": 6, "memory": 6319, "step": 21435} +{"lr": 4.779788540831406e-05, "data_time": 0.0009724378585815429, "loss": 0.07255338467657566, "time": 0.5283006191253662, "epoch": 6, "memory": 6319, "step": 21535} +{"lr": 4.744094444989843e-05, "data_time": 0.020735740661621094, "loss": 0.08015012815594673, "time": 0.6079918384552002, "epoch": 6, "memory": 6319, "step": 21635} +{"lr": 4.708449223174764e-05, "data_time": 0.0012578010559082032, "loss": 0.07504624053835869, "time": 0.603456997871399, "epoch": 6, "memory": 6319, "step": 21735} +{"lr": 4.672855180072916e-05, "data_time": 0.0011845111846923828, "loss": 0.06350935623049736, "time": 0.6788998365402221, "epoch": 6, "memory": 6319, "step": 21835} +{"lr": 4.6373146170620256e-05, "data_time": 0.000836491584777832, "loss": 0.06916520148515701, "time": 0.6389220952987671, "epoch": 6, "memory": 6319, "step": 21935} +{"lr": 4.6018298320619915e-05, "data_time": 0.0008458375930786132, "loss": 0.07418507188558579, "time": 0.620337438583374, "epoch": 6, "memory": 6319, "step": 22035} +{"lr": 4.566403119386318e-05, "data_time": 0.0010722875595092773, "loss": 0.06531587988138199, "time": 0.5736503839492798, "epoch": 6, "memory": 6319, "step": 22135} +{"lr": 4.531036769593769e-05, "data_time": 0.0012825965881347657, "loss": 0.07403096482157707, "time": 0.6477460145950318, "epoch": 6, "memory": 6319, "step": 22235} +{"lr": 4.4957330693402685e-05, "data_time": 0.0013995170593261719, "loss": 0.08489035218954086, "time": 0.6271989107131958, "epoch": 6, "memory": 6319, "step": 22335} +{"lr": 4.460494301231059e-05, "data_time": 0.0011365652084350587, "loss": 0.0758124865591526, "time": 0.6079889297485351, "epoch": 6, "memory": 6319, "step": 22435} +{"lr": 4.425322743673109e-05, "data_time": 0.0012985944747924804, "loss": 0.07123874612152577, "time": 0.6142471313476563, "epoch": 6, "memory": 6319, "step": 22535} +{"lr": 4.390220670727812e-05, "data_time": 0.0010303497314453126, "loss": 0.08152568638324738, "time": 0.5848260164260864, "epoch": 6, "memory": 6319, "step": 22635} +{"lr": 4.3551903519639325e-05, "data_time": 0.0010384321212768555, "loss": 0.08272028975188732, "time": 0.6074117660522461, "epoch": 6, "memory": 6319, "step": 22735} +{"lr": 4.3202340523108894e-05, "data_time": 0.0009337425231933594, "loss": 0.06299005895853042, "time": 0.6091707944869995, "epoch": 6, "memory": 6319, "step": 22835} +{"lr": 4.285354031912302e-05, "data_time": 0.001403045654296875, "loss": 0.08641164749860764, "time": 0.6172260999679565, "epoch": 6, "memory": 6319, "step": 22935} +{"lr": 4.2505525459798544e-05, "data_time": 0.0010636806488037109, "loss": 0.07034889832139016, "time": 0.6003542184829712, "epoch": 6, "memory": 6319, "step": 23035} +{"lr": 4.2158318446474836e-05, "data_time": 0.0008357524871826172, "loss": 0.0810418576002121, "time": 0.5945213556289672, "epoch": 6, "memory": 6319, "step": 23135} +{"lr": 4.181194172825905e-05, "data_time": 0.001106858253479004, "loss": 0.07436636798083782, "time": 0.661811375617981, "epoch": 6, "memory": 6319, "step": 23235} +{"lr": 4.146641770057448e-05, "data_time": 0.00128629207611084, "loss": 0.07141970098018646, "time": 0.6103588342666626, "epoch": 6, "memory": 6319, "step": 23335} +{"lr": 4.112176870371267e-05, "data_time": 0.0008770465850830078, "loss": 0.08884702026844024, "time": 0.6611963748931885, "epoch": 6, "memory": 6319, "step": 23435} +{"accuracy/top1": 94.78445434570312, "data_time": 0.5004175860306312, "time": 0.5898567602552217, "step": 6} +{"lr": 4.0753988582732786e-05, "data_time": 0.474355149269104, "loss": 0.06689820997416973, "time": 0.6163939237594604, "epoch": 7, "memory": 6319, "step": 23542} +{"lr": 4.0411221639286815e-05, "data_time": 0.525312352180481, "loss": 0.0629511084407568, "time": 0.6700114965438843, "epoch": 7, "memory": 6319, "step": 23642} +{"lr": 4.006939795169518e-05, "data_time": 0.4356280565261841, "loss": 0.07155358083546162, "time": 0.5888641119003296, "epoch": 7, "memory": 6319, "step": 23742} +{"lr": 3.972853962099913e-05, "data_time": 0.4349459886550903, "loss": 0.08872487097978592, "time": 0.5881338596343995, "epoch": 7, "memory": 6319, "step": 23842} +{"lr": 3.93886686858237e-05, "data_time": 0.49597482681274413, "loss": 0.08656098246574402, "time": 0.6514045715332031, "epoch": 7, "memory": 6319, "step": 23942} +{"lr": 3.904980712095259e-05, "data_time": 0.42073636054992675, "loss": 0.07925464436411858, "time": 0.564240574836731, "epoch": 7, "memory": 6319, "step": 24042} +{"lr": 3.871197683590748e-05, "data_time": 0.5220647573471069, "loss": 0.08019356690347194, "time": 0.6643844366073608, "epoch": 7, "memory": 6319, "step": 24142} +{"lr": 3.837519967353125e-05, "data_time": 0.4823348045349121, "loss": 0.08155170679092408, "time": 0.6275966882705688, "epoch": 7, "memory": 6319, "step": 24242} +{"lr": 3.8039497408575855e-05, "data_time": 0.4871502876281738, "loss": 0.07777530662715434, "time": 0.6438503742218018, "epoch": 7, "memory": 6319, "step": 24342} +{"lr": 3.770489174629432e-05, "data_time": 0.4617809534072876, "loss": 0.07056905478239059, "time": 0.6046681642532349, "epoch": 7, "memory": 6319, "step": 24442} +{"lr": 3.737140432103754e-05, "data_time": 0.45655364990234376, "loss": 0.0921558927744627, "time": 0.6123908042907715, "epoch": 7, "memory": 6319, "step": 24542} +{"lr": 3.703905669485545e-05, "data_time": 0.46669831275939944, "loss": 0.08235384598374366, "time": 0.6098385810852051, "epoch": 7, "memory": 6319, "step": 24642} +{"lr": 3.6707870356102706e-05, "data_time": 0.4210800647735596, "loss": 0.08048867210745811, "time": 0.563645315170288, "epoch": 7, "memory": 6319, "step": 24742} +{"lr": 3.6377866718049484e-05, "data_time": 0.4014909267425537, "loss": 0.06895597763359547, "time": 0.5435338258743286, "epoch": 7, "memory": 6319, "step": 24842} +{"lr": 3.6049067117496965e-05, "data_time": 0.5016248464584351, "loss": 0.06598893962800503, "time": 0.6431945562362671, "epoch": 7, "memory": 6319, "step": 24942} +{"lr": 3.5721492813397755e-05, "data_time": 0.45930278301239014, "loss": 0.07469790577888488, "time": 0.6020509719848632, "epoch": 7, "memory": 6319, "step": 25042} +{"lr": 3.539516498548151e-05, "data_time": 0.4277693510055542, "loss": 0.07332868464291095, "time": 0.5700596570968628, "epoch": 7, "memory": 6319, "step": 25142} +{"lr": 3.507010473288518e-05, "data_time": 0.4210413694381714, "loss": 0.07406840659677982, "time": 0.5631956100463867, "epoch": 7, "memory": 6319, "step": 25242} +{"lr": 3.474633307278909e-05, "data_time": 0.48797500133514404, "loss": 0.0916257381439209, "time": 0.6304330587387085, "epoch": 7, "memory": 6319, "step": 25342} +{"lr": 3.4423870939058177e-05, "data_time": 0.462441611289978, "loss": 0.09362282380461692, "time": 0.633586072921753, "epoch": 7, "memory": 6319, "step": 25442} +{"lr": 3.410273918088805e-05, "data_time": 0.45768418312072756, "loss": 0.07509752325713634, "time": 0.6017839193344117, "epoch": 7, "memory": 6319, "step": 25542} +{"lr": 3.37829585614574e-05, "data_time": 0.4711894989013672, "loss": 0.09683760479092599, "time": 0.61318039894104, "epoch": 7, "memory": 6319, "step": 25642} +{"lr": 3.3464549756585216e-05, "data_time": 0.43604578971862795, "loss": 0.08644391037523746, "time": 0.5796491622924804, "epoch": 7, "memory": 6319, "step": 25742} +{"lr": 3.314753335339416e-05, "data_time": 0.502040696144104, "loss": 0.0850952073931694, "time": 0.6460355997085572, "epoch": 7, "memory": 6319, "step": 25842} +{"lr": 3.2831929848979394e-05, "data_time": 0.48782057762145997, "loss": 0.07723802626132965, "time": 0.6405158042907715, "epoch": 7, "memory": 6319, "step": 25942} +{"lr": 3.251775964908332e-05, "data_time": 0.48589468002319336, "loss": 0.08428933471441269, "time": 0.6298029899597168, "epoch": 7, "memory": 6319, "step": 26042} +{"lr": 3.22050430667762e-05, "data_time": 0.47635655403137206, "loss": 0.08189189061522484, "time": 0.6201178073883057, "epoch": 7, "memory": 6319, "step": 26142} +{"lr": 3.189380032114285e-05, "data_time": 0.5052550315856934, "loss": 0.09040755890309811, "time": 0.6483135938644409, "epoch": 7, "memory": 6319, "step": 26242} +{"lr": 3.1584051535975324e-05, "data_time": 0.4549750328063965, "loss": 0.07514316774904728, "time": 0.6086362361907959, "epoch": 7, "memory": 6319, "step": 26342} +{"lr": 3.127581673847175e-05, "data_time": 0.4394742250442505, "loss": 0.08260803520679474, "time": 0.5820672035217285, "epoch": 7, "memory": 6319, "step": 26442} +{"lr": 3.0969115857941555e-05, "data_time": 0.48485751152038575, "loss": 0.09186746180057526, "time": 0.629362678527832, "epoch": 7, "memory": 6319, "step": 26542} +{"lr": 3.0663968724516654e-05, "data_time": 0.4046575307846069, "loss": 0.0835926216095686, "time": 0.5463639497756958, "epoch": 7, "memory": 6319, "step": 26642} +{"lr": 3.0360395067869668e-05, "data_time": 0.43217408657073975, "loss": 0.08738039694726467, "time": 0.5855432271957397, "epoch": 7, "memory": 6319, "step": 26742} +{"lr": 3.0058414515938063e-05, "data_time": 0.48802919387817384, "loss": 0.08832942582666874, "time": 0.6322428226470947, "epoch": 7, "memory": 6319, "step": 26842} +{"lr": 2.975804659365501e-05, "data_time": 0.4412156343460083, "loss": 0.0902914434671402, "time": 0.5849912881851196, "epoch": 7, "memory": 6319, "step": 26942} +{"lr": 2.945931072168714e-05, "data_time": 0.48147096633911135, "loss": 0.07930332496762275, "time": 0.6256494283676147, "epoch": 7, "memory": 6319, "step": 27042} +{"lr": 2.9162226215178855e-05, "data_time": 0.4175581693649292, "loss": 0.08182027153670787, "time": 0.5587935209274292, "epoch": 7, "memory": 6319, "step": 27142} +{"lr": 2.8866812282503322e-05, "data_time": 0.4660873651504517, "loss": 0.08387458957731724, "time": 0.6087932825088501, "epoch": 7, "memory": 6319, "step": 27242} +{"lr": 2.8573088024020838e-05, "data_time": 0.4384413242340088, "loss": 0.07118411213159562, "time": 0.5906196594238281, "epoch": 7, "memory": 6319, "step": 27342} +{"accuracy/top1": 96.3118896484375, "data_time": 0.5100329908831366, "time": 0.6019841309251457, "step": 7} +{"lr": 2.8260695817390092e-05, "data_time": 0.4504760980606079, "loss": 0.08787877596914768, "time": 0.5920387983322144, "epoch": 8, "memory": 6319, "step": 27449} +{"lr": 2.7970529402570967e-05, "data_time": 0.4833108186721802, "loss": 0.06415975615382194, "time": 0.6258732080459595, "epoch": 8, "memory": 6319, "step": 27549} +{"lr": 2.7682110612244214e-05, "data_time": 0.45999147891998293, "loss": 0.08622431494295597, "time": 0.6035938501358032, "epoch": 8, "memory": 6319, "step": 27649} +{"lr": 2.7395458094490283e-05, "data_time": 0.4709547281265259, "loss": 0.07511433474719524, "time": 0.6246785402297974, "epoch": 8, "memory": 6319, "step": 27749} +{"lr": 2.7110590383189015e-05, "data_time": 0.4341695547103882, "loss": 0.06846614442765712, "time": 0.5759546518325805, "epoch": 8, "memory": 6319, "step": 27849} +{"lr": 2.6827525896821397e-05, "data_time": 0.4143645763397217, "loss": 0.0698573287576437, "time": 0.5643430471420288, "epoch": 8, "memory": 6319, "step": 27949} +{"lr": 2.6546282937278552e-05, "data_time": 0.4419599771499634, "loss": 0.07415164783596992, "time": 0.5827390670776367, "epoch": 8, "memory": 6319, "step": 28049} +{"lr": 2.6266879688678606e-05, "data_time": 0.5172754049301147, "loss": 0.10456358268857002, "time": 0.6736511707305908, "epoch": 8, "memory": 6319, "step": 28149} +{"lr": 2.5989334216190738e-05, "data_time": 0.4718210458755493, "loss": 0.0814783986657858, "time": 0.6133327007293701, "epoch": 8, "memory": 6319, "step": 28249} +{"lr": 2.571366446486732e-05, "data_time": 0.44338967800140383, "loss": 0.08454127684235573, "time": 0.6003024578094482, "epoch": 8, "memory": 6319, "step": 28349} +{"lr": 2.5439888258483665e-05, "data_time": 0.5230553865432739, "loss": 0.07729196920990944, "time": 0.6659160852432251, "epoch": 8, "memory": 6319, "step": 28449} +{"lr": 2.5168023298385384e-05, "data_time": 0.4534576892852783, "loss": 0.07279450930655003, "time": 0.5979245901107788, "epoch": 8, "memory": 6319, "step": 28549} +{"lr": 2.489808716234412e-05, "data_time": 0.4939014434814453, "loss": 0.0854671698063612, "time": 0.6376083612442016, "epoch": 8, "memory": 6319, "step": 28649} +{"lr": 2.463009730342111e-05, "data_time": 0.4644150733947754, "loss": 0.07376628667116165, "time": 0.6085563898086548, "epoch": 8, "memory": 6319, "step": 28749} +{"lr": 2.436407104883849e-05, "data_time": 0.48210346698760986, "loss": 0.07696508876979351, "time": 0.6235823392868042, "epoch": 8, "memory": 6319, "step": 28849} +{"lr": 2.4100025598859092e-05, "data_time": 0.49641640186309816, "loss": 0.0698467107489705, "time": 0.6405101776123047, "epoch": 8, "memory": 6319, "step": 28949} +{"lr": 2.3837978025674284e-05, "data_time": 0.4483215570449829, "loss": 0.0646116517484188, "time": 0.5912272214889527, "epoch": 8, "memory": 6319, "step": 29049} +{"lr": 2.357794527230027e-05, "data_time": 0.4442706823348999, "loss": 0.07210826762020588, "time": 0.5862396717071533, "epoch": 8, "memory": 6319, "step": 29149} +{"lr": 2.33199441514825e-05, "data_time": 0.46998779773712157, "loss": 0.07695260904729366, "time": 0.6124399900436401, "epoch": 8, "memory": 6319, "step": 29249} +{"lr": 2.306399134460859e-05, "data_time": 0.5597031116485596, "loss": 0.0864983320236206, "time": 0.7024467706680297, "epoch": 8, "memory": 6319, "step": 29349} +{"lr": 2.281010340062995e-05, "data_time": 0.45230419635772706, "loss": 0.0875934824347496, "time": 0.5931612968444824, "epoch": 8, "memory": 6319, "step": 29449} +{"lr": 2.2558296734991603e-05, "data_time": 0.4420763969421387, "loss": 0.08598154541105033, "time": 0.5839328050613404, "epoch": 8, "memory": 6319, "step": 29549} +{"lr": 2.2308587628570915e-05, "data_time": 0.45420012474060056, "loss": 0.07689887657761574, "time": 0.6094831943511962, "epoch": 8, "memory": 6319, "step": 29649} +{"lr": 2.2060992226624955e-05, "data_time": 0.43480234146118163, "loss": 0.07518543973565102, "time": 0.5770165681838989, "epoch": 8, "memory": 6319, "step": 29749} +{"lr": 2.181552653774648e-05, "data_time": 0.41755697727203367, "loss": 0.07007209099829197, "time": 0.5601857662200928, "epoch": 8, "memory": 6319, "step": 29849} +{"lr": 2.1572206432829035e-05, "data_time": 0.44503648281097413, "loss": 0.07302001379430294, "time": 0.5869835376739502, "epoch": 8, "memory": 6319, "step": 29949} +{"lr": 2.1331047644040754e-05, "data_time": 0.4657976388931274, "loss": 0.08360778503119945, "time": 0.6086403846740722, "epoch": 8, "memory": 6319, "step": 30049} +{"lr": 2.109206576380705e-05, "data_time": 0.42251613140106203, "loss": 0.08748757466673851, "time": 0.5644537448883057, "epoch": 8, "memory": 6319, "step": 30149} +{"lr": 2.0855276243802687e-05, "data_time": 0.4733867645263672, "loss": 0.07566434890031815, "time": 0.6182521343231201, "epoch": 8, "memory": 6319, "step": 30249} +{"lr": 2.0620694393952487e-05, "data_time": 0.5105309247970581, "loss": 0.08560679405927658, "time": 0.6539804697036743, "epoch": 8, "memory": 6319, "step": 30349} +{"lr": 2.038833538144169e-05, "data_time": 0.544872498512268, "loss": 0.0843689925968647, "time": 0.6889699935913086, "epoch": 8, "memory": 6319, "step": 30449} +{"lr": 2.0158214229735154e-05, "data_time": 0.4630296230316162, "loss": 0.07467749454081059, "time": 0.6060987949371338, "epoch": 8, "memory": 6319, "step": 30549} +{"lr": 1.9930345817605977e-05, "data_time": 0.5143996000289917, "loss": 0.08833044432103634, "time": 0.6577551364898682, "epoch": 8, "memory": 6319, "step": 30649} +{"lr": 1.970474487817358e-05, "data_time": 0.4972402572631836, "loss": 0.07718662656843663, "time": 0.6398980617523193, "epoch": 8, "memory": 6319, "step": 30749} +{"lr": 1.9481425997951092e-05, "data_time": 0.45746264457702634, "loss": 0.07747239619493484, "time": 0.6035328388214112, "epoch": 8, "memory": 6319, "step": 30849} +{"lr": 1.9260403615902187e-05, "data_time": 0.5013140916824341, "loss": 0.07578352279961109, "time": 0.6433354377746582, "epoch": 8, "memory": 6319, "step": 30949} +{"lr": 1.9041692022507647e-05, "data_time": 0.5530886173248291, "loss": 0.072789004445076, "time": 0.7126932859420776, "epoch": 8, "memory": 6319, "step": 31049} +{"lr": 1.8825305358841125e-05, "data_time": 0.46707730293273925, "loss": 0.08952586576342583, "time": 0.6104723930358886, "epoch": 8, "memory": 6319, "step": 31149} +{"lr": 1.861125761565518e-05, "data_time": 0.39845116138458253, "loss": 0.07494150102138519, "time": 0.5425600528717041, "epoch": 8, "memory": 6319, "step": 31249} +{"accuracy/top1": 96.72245788574219, "data_time": 0.4933110845500025, "time": 0.5810148099373127, "step": 8} +{"lr": 1.8384832449122864e-05, "data_time": 0.4758762359619141, "loss": 0.0754853405058384, "time": 0.6189368009567261, "epoch": 9, "memory": 6319, "step": 31356} +{"lr": 1.817567007332114e-05, "data_time": 0.45037035942077636, "loss": 0.08184201121330262, "time": 0.606569504737854, "epoch": 9, "memory": 6319, "step": 31456} +{"lr": 1.796888862098825e-05, "data_time": 0.4804424285888672, "loss": 0.06698336489498616, "time": 0.623430871963501, "epoch": 9, "memory": 6319, "step": 31556} +{"lr": 1.7764501461839678e-05, "data_time": 0.47014167308807375, "loss": 0.07689600102603436, "time": 0.6136189460754394, "epoch": 9, "memory": 6319, "step": 31656} +{"lr": 1.7562521810784746e-05, "data_time": 0.45176260471343993, "loss": 0.08106597140431404, "time": 0.5967425346374512, "epoch": 9, "memory": 6319, "step": 31756} +{"lr": 1.7362962727072444e-05, "data_time": 0.4487202882766724, "loss": 0.07057645507156848, "time": 0.5911329746246338, "epoch": 9, "memory": 6319, "step": 31856} +{"lr": 1.716583711344682e-05, "data_time": 0.48130006790161134, "loss": 0.07462608590722083, "time": 0.6231300115585328, "epoch": 9, "memory": 6319, "step": 31956} +{"lr": 1.6971157715312954e-05, "data_time": 0.46283550262451173, "loss": 0.06999175287783146, "time": 0.6046403408050537, "epoch": 9, "memory": 6319, "step": 32056} +{"lr": 1.677893711991274e-05, "data_time": 0.4871793508529663, "loss": 0.07801001891493797, "time": 0.6288370609283447, "epoch": 9, "memory": 6319, "step": 32156} +{"lr": 1.6589187755511025e-05, "data_time": 0.4406203985214233, "loss": 0.06530537940561772, "time": 0.5981998920440674, "epoch": 9, "memory": 6319, "step": 32256} +{"lr": 1.640192189059216e-05, "data_time": 0.41819570064544676, "loss": 0.0774762861430645, "time": 0.5595851421356202, "epoch": 9, "memory": 6319, "step": 32356} +{"lr": 1.6217151633066716e-05, "data_time": 0.4525012731552124, "loss": 0.08113399259746075, "time": 0.6044313192367554, "epoch": 9, "memory": 6319, "step": 32456} +{"lr": 1.6034888929488564e-05, "data_time": 0.4695448398590088, "loss": 0.06804250255227089, "time": 0.6150140285491943, "epoch": 9, "memory": 6319, "step": 32556} +{"lr": 1.585514556428258e-05, "data_time": 0.48380489349365235, "loss": 0.065376490727067, "time": 0.6378083467483521, "epoch": 9, "memory": 6319, "step": 32656} +{"lr": 1.567793315898261e-05, "data_time": 0.4007999897003174, "loss": 0.07852859795093536, "time": 0.5449673891067505, "epoch": 9, "memory": 6319, "step": 32756} +{"lr": 1.5503263171480048e-05, "data_time": 0.47496657371520995, "loss": 0.07522008717060089, "time": 0.6424149751663208, "epoch": 9, "memory": 6319, "step": 32856} +{"lr": 1.5331146895283165e-05, "data_time": 0.43189353942871095, "loss": 0.07639147564768792, "time": 0.5732213973999023, "epoch": 9, "memory": 6319, "step": 32956} +{"lr": 1.5161595458786728e-05, "data_time": 0.4367440938949585, "loss": 0.08115535750985145, "time": 0.595166802406311, "epoch": 9, "memory": 6319, "step": 33056} +{"lr": 1.4994619824552563e-05, "data_time": 0.4457152605056763, "loss": 0.07400755509734154, "time": 0.5902722597122192, "epoch": 9, "memory": 6319, "step": 33156} +{"lr": 1.4830230788600802e-05, "data_time": 0.46508469581604006, "loss": 0.0755990494042635, "time": 0.623738956451416, "epoch": 9, "memory": 6319, "step": 33256} +{"lr": 1.4668438979711779e-05, "data_time": 0.4103731870651245, "loss": 0.07834706865251065, "time": 0.554237675666809, "epoch": 9, "memory": 6319, "step": 33356} +{"lr": 1.4509254858738844e-05, "data_time": 0.45750463008880615, "loss": 0.06773686297237873, "time": 0.613737964630127, "epoch": 9, "memory": 6319, "step": 33456} +{"lr": 1.4352688717931966e-05, "data_time": 0.4298782587051392, "loss": 0.07748924121260643, "time": 0.5725626707077026, "epoch": 9, "memory": 6319, "step": 33556} +{"lr": 1.4198750680272348e-05, "data_time": 0.4645204782485962, "loss": 0.0789119977504015, "time": 0.6064340353012085, "epoch": 9, "memory": 6319, "step": 33656} +{"lr": 1.4047450698817857e-05, "data_time": 0.4595818519592285, "loss": 0.08945306725800037, "time": 0.6004274606704711, "epoch": 9, "memory": 6319, "step": 33756} +{"lr": 1.389879855605952e-05, "data_time": 0.4054666519165039, "loss": 0.07022984363138676, "time": 0.5477522850036621, "epoch": 9, "memory": 6319, "step": 33856} +{"lr": 1.3752803863288976e-05, "data_time": 0.456856369972229, "loss": 0.06537099406123162, "time": 0.5985014200210571, "epoch": 9, "memory": 6319, "step": 33956} +{"lr": 1.3609476059977129e-05, "data_time": 0.5411594390869141, "loss": 0.07017961330711842, "time": 0.6840475559234619, "epoch": 9, "memory": 6319, "step": 34056} +{"lr": 1.3468824413163754e-05, "data_time": 0.42032756805419924, "loss": 0.06352031752467155, "time": 0.5630666494369507, "epoch": 9, "memory": 6319, "step": 34156} +{"lr": 1.3330858016858395e-05, "data_time": 0.4271347999572754, "loss": 0.0937989518046379, "time": 0.5699228286743164, "epoch": 9, "memory": 6319, "step": 34256} +{"lr": 1.3195585791452309e-05, "data_time": 0.4812811851501465, "loss": 0.07161642089486123, "time": 0.6232021808624267, "epoch": 9, "memory": 6319, "step": 34356} +{"lr": 1.3063016483141729e-05, "data_time": 0.45418920516967776, "loss": 0.06601140499114991, "time": 0.5956439256668091, "epoch": 9, "memory": 6319, "step": 34456} +{"lr": 1.2933158663362388e-05, "data_time": 0.42545571327209475, "loss": 0.08046995252370834, "time": 0.5846264362335205, "epoch": 9, "memory": 6319, "step": 34556} +{"lr": 1.2806020728235311e-05, "data_time": 0.4451966047286987, "loss": 0.07233460396528243, "time": 0.5870979309082032, "epoch": 9, "memory": 6319, "step": 34656} +{"lr": 1.2681610898023963e-05, "data_time": 0.4433310508728027, "loss": 0.07668758258223533, "time": 0.585440993309021, "epoch": 9, "memory": 6319, "step": 34756} +{"lr": 1.2559937216602731e-05, "data_time": 0.47575376033782957, "loss": 0.08021556288003921, "time": 0.6187443256378173, "epoch": 9, "memory": 6319, "step": 34856} +{"lr": 1.2441007550936847e-05, "data_time": 0.5138709545135498, "loss": 0.07820392809808255, "time": 0.6556874990463257, "epoch": 9, "memory": 6319, "step": 34956} +{"lr": 1.232482959057377e-05, "data_time": 0.45088682174682615, "loss": 0.0829607080668211, "time": 0.5938930034637451, "epoch": 9, "memory": 6319, "step": 35056} +{"lr": 1.2211410847145969e-05, "data_time": 0.479123592376709, "loss": 0.07704100422561169, "time": 0.6211544513702393, "epoch": 9, "memory": 6319, "step": 35156} +{"accuracy/top1": 96.98340606689453, "data_time": 0.50804053092825, "time": 0.5981968764601082, "step": 9} +{"lr": 1.209311679360931e-05, "data_time": 0.41988067626953124, "loss": 0.07475814446806908, "time": 0.5620572805404663, "epoch": 10, "memory": 6319, "step": 35263} +{"lr": 1.1985432727488666e-05, "data_time": 0.42589197158813474, "loss": 0.08367883190512657, "time": 0.569089674949646, "epoch": 10, "memory": 6319, "step": 35363} +{"lr": 1.1880529822435289e-05, "data_time": 0.48822770118713377, "loss": 0.08758219256997109, "time": 0.6306748628616333, "epoch": 10, "memory": 6319, "step": 35463} +{"lr": 1.1778414861078726e-05, "data_time": 0.44216940402984617, "loss": 0.07201689779758454, "time": 0.5951197385787964, "epoch": 10, "memory": 6319, "step": 35563} +{"lr": 1.1679094445790538e-05, "data_time": 0.41455626487731934, "loss": 0.07749701552093029, "time": 0.5566478490829467, "epoch": 10, "memory": 6319, "step": 35663} +{"lr": 1.1582574998257377e-05, "data_time": 0.45920236110687257, "loss": 0.06470660604536534, "time": 0.6031032562255859, "epoch": 10, "memory": 6319, "step": 35763} +{"lr": 1.1488862759065762e-05, "data_time": 0.40541141033172606, "loss": 0.07805955074727536, "time": 0.5468392610549927, "epoch": 10, "memory": 6319, "step": 35863} +{"lr": 1.1397963787298665e-05, "data_time": 0.4425260066986084, "loss": 0.08282766602933407, "time": 0.6018846511840821, "epoch": 10, "memory": 6319, "step": 35963} +{"lr": 1.1309883960143664e-05, "data_time": 0.4725736379623413, "loss": 0.08345601595938205, "time": 0.6152339220046997, "epoch": 10, "memory": 6319, "step": 36063} +{"lr": 1.122462897251302e-05, "data_time": 0.4819929599761963, "loss": 0.06973667442798615, "time": 0.6446127891540527, "epoch": 10, "memory": 6319, "step": 36163} +{"lr": 1.114220433667542e-05, "data_time": 0.43996443748474123, "loss": 0.08997197225689887, "time": 0.5972410678863526, "epoch": 10, "memory": 6319, "step": 36263} +{"lr": 1.10626153818996e-05, "data_time": 0.4460367441177368, "loss": 0.06445202976465225, "time": 0.5879318714141846, "epoch": 10, "memory": 6319, "step": 36363} +{"lr": 1.0985867254109761e-05, "data_time": 0.48964581489562986, "loss": 0.0746229462325573, "time": 0.6339953899383545, "epoch": 10, "memory": 6319, "step": 36463} +{"lr": 1.0911964915552823e-05, "data_time": 0.3942296266555786, "loss": 0.08499046415090561, "time": 0.5552056550979614, "epoch": 10, "memory": 6319, "step": 36563} +{"lr": 1.0840913144477643e-05, "data_time": 0.4788695812225342, "loss": 0.07111723981797695, "time": 0.6212372064590455, "epoch": 10, "memory": 6319, "step": 36663} +{"lr": 1.0772716534826064e-05, "data_time": 0.4616294622421265, "loss": 0.07191082239151, "time": 0.6043037414550781, "epoch": 10, "memory": 6319, "step": 36763} +{"lr": 1.0707379495935828e-05, "data_time": 0.4547978639602661, "loss": 0.07027343362569809, "time": 0.6120363950729371, "epoch": 10, "memory": 6319, "step": 36863} +{"lr": 1.0644906252255541e-05, "data_time": 0.39924206733703616, "loss": 0.0715204045176506, "time": 0.5521424293518067, "epoch": 10, "memory": 6319, "step": 36963} +{"lr": 1.0585300843071527e-05, "data_time": 0.45977783203125, "loss": 0.06767462603747845, "time": 0.6041017055511475, "epoch": 10, "memory": 6319, "step": 37063} +{"lr": 1.0528567122246654e-05, "data_time": 0.44636244773864747, "loss": 0.07604262344539166, "time": 0.5876028537750244, "epoch": 10, "memory": 6319, "step": 37163} +{"lr": 1.0474708757971141e-05, "data_time": 0.4786412715911865, "loss": 0.06674356833100319, "time": 0.6208931922912597, "epoch": 10, "memory": 6319, "step": 37263} +{"lr": 1.042372923252543e-05, "data_time": 0.4120201826095581, "loss": 0.07286047264933586, "time": 0.5540947198867798, "epoch": 10, "memory": 6319, "step": 37363} +{"lr": 1.0375631842054998e-05, "data_time": 0.46259894371032717, "loss": 0.07435069568455219, "time": 0.606714916229248, "epoch": 10, "memory": 6319, "step": 37463} +{"lr": 1.0330419696357254e-05, "data_time": 0.4271500825881958, "loss": 0.06328838840126991, "time": 0.5708273649215698, "epoch": 10, "memory": 6319, "step": 37563} +{"lr": 1.028809571868047e-05, "data_time": 0.4651278734207153, "loss": 0.06647076942026615, "time": 0.6080742359161377, "epoch": 10, "memory": 6319, "step": 37663} +{"lr": 1.0248662645534766e-05, "data_time": 0.41365871429443357, "loss": 0.07770975790917874, "time": 0.5578250169754029, "epoch": 10, "memory": 6319, "step": 37763} +{"lr": 1.0212123026515194e-05, "data_time": 0.4224325895309448, "loss": 0.0664729431271553, "time": 0.5639659404754639, "epoch": 10, "memory": 6319, "step": 37863} +{"lr": 1.0178479224136881e-05, "data_time": 0.416499400138855, "loss": 0.07827116660773754, "time": 0.5590309381484986, "epoch": 10, "memory": 6319, "step": 37963} +{"lr": 1.0147733413682289e-05, "data_time": 0.5093073368072509, "loss": 0.06647942066192628, "time": 0.6525231122970581, "epoch": 10, "memory": 6319, "step": 38063} +{"lr": 1.0119887583060527e-05, "data_time": 0.435845947265625, "loss": 0.06967916637659073, "time": 0.5783462762832642, "epoch": 10, "memory": 6319, "step": 38163} +{"lr": 1.0094943532678871e-05, "data_time": 0.4608129024505615, "loss": 0.08074088096618652, "time": 0.60422682762146, "epoch": 10, "memory": 6319, "step": 38263} +{"lr": 1.007290287532637e-05, "data_time": 0.41082403659820554, "loss": 0.07776946760714054, "time": 0.5526329278945923, "epoch": 10, "memory": 6319, "step": 38363} +{"lr": 1.0053767036069495e-05, "data_time": 0.43117775917053225, "loss": 0.059059906750917435, "time": 0.5733058452606201, "epoch": 10, "memory": 6319, "step": 38463} +{"lr": 1.003753725216005e-05, "data_time": 0.4748003721237183, "loss": 0.07253213152289391, "time": 0.6162424802780151, "epoch": 10, "memory": 6319, "step": 38563} +{"lr": 1.0024214572955152e-05, "data_time": 0.4608904838562012, "loss": 0.06550565175712109, "time": 0.6028083801269531, "epoch": 10, "memory": 6319, "step": 38663} +{"lr": 1.0013799859849456e-05, "data_time": 0.4114959239959717, "loss": 0.09634218141436576, "time": 0.5531365394592285, "epoch": 10, "memory": 6319, "step": 38763} +{"lr": 1.0006293786219338e-05, "data_time": 0.46281936168670657, "loss": 0.07998348549008369, "time": 0.6058304071426391, "epoch": 10, "memory": 6319, "step": 38863} +{"lr": 1.0001696837379472e-05, "data_time": 0.519748306274414, "loss": 0.07544913105666637, "time": 0.6611512660980224, "epoch": 10, "memory": 6319, "step": 38963} +{"lr": 1.0000009310551361e-05, "data_time": 0.4274256944656372, "loss": 0.07560225427150727, "time": 0.5697463035583497, "epoch": 10, "memory": 6319, "step": 39063} +{"accuracy/top1": 97.04603576660156, "data_time": 0.4988717580663747, "time": 0.585812913960424, "step": 10} diff --git a/resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1/20230604_225317/vis_data/config.py b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1/20230604_225317/vis_data/config.py new file mode 100644 index 0000000000000000000000000000000000000000..ab532bdd5585a688820d77a34f991bc9b1d4859b --- /dev/null +++ b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1/20230604_225317/vis_data/config.py @@ -0,0 +1,230 @@ +optim_wrapper = dict( + optimizer=dict( + type='SGD', + lr=0.0001, + momentum=0.9, + weight_decay=0.0001, + _scope_='mmpretrain'), + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=False, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=512) +model = dict( + type='ImageClassifier', + backbone=dict( + frozen_stages=2, + type='ResNet', + depth=50, + num_stages=4, + out_indices=(3, ), + style='pytorch', + init_cfg=dict( + type='Pretrained', + checkpoint= + 'https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth', + prefix='backbone')), + neck=dict(type='GlobalAveragePooling'), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=2048, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + topk=1)) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stablediffusionV2-1-dpmsolver-25-1m.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/cc1m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV2-1-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = dict(type='Accuracy', topk=1) +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV2-1-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = dict(type='Accuracy', topk=1) +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1' diff --git a/resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1/20230604_225317/vis_data/events.out.tfevents.1685890413.SH-IDC1-10-140-24-80.153248.0 b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1/20230604_225317/vis_data/events.out.tfevents.1685890413.SH-IDC1-10-140-24-80.153248.0 new file mode 100644 index 0000000000000000000000000000000000000000..ae40ee4c6d4b29f3864e0f116ca325f72ade75ca --- /dev/null +++ b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1/20230604_225317/vis_data/events.out.tfevents.1685890413.SH-IDC1-10-140-24-80.153248.0 @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:05e28fc63305af87963bd6c381698f5b707a6917822497a0ec40336f50963d32 +size 2655214 diff --git a/resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1/20230604_225317/vis_data/scalars.json b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1/20230604_225317/vis_data/scalars.json new file mode 100644 index 0000000000000000000000000000000000000000..ebe3ff4317681eded55cb9cbd32f881b168bdcb1 --- /dev/null +++ b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1/20230604_225317/vis_data/scalars.json @@ -0,0 +1,400 @@ +{"lr": 9.999857418382631e-05, "data_time": 0.04329502582550049, "loss": 0.6144014835357666, "time": 0.5942174673080445, "epoch": 1, "memory": 9436, "step": 100} +{"lr": 9.999423907359506e-05, "data_time": 0.0017529964447021485, "loss": 0.540549498796463, "time": 0.6243215084075928, "epoch": 1, "memory": 6319, "step": 200} +{"lr": 9.998699480412073e-05, "data_time": 0.001414918899536133, "loss": 0.48605347573757174, "time": 0.6351176261901855, "epoch": 1, "memory": 6319, "step": 300} +{"lr": 9.997684184379085e-05, "data_time": 0.0015470027923583985, "loss": 0.4411657601594925, "time": 0.5894553184509277, "epoch": 1, "memory": 6319, "step": 400} +{"lr": 9.996378084905784e-05, "data_time": 0.0015391826629638672, "loss": 0.3897904396057129, "time": 0.5979857444763184, "epoch": 1, "memory": 6319, "step": 500} +{"lr": 9.994781266439667e-05, "data_time": 0.0010723114013671876, "loss": 0.3695789337158203, "time": 0.6539291620254517, "epoch": 1, "memory": 6319, "step": 600} +{"lr": 9.992893832225061e-05, "data_time": 0.0011087179183959961, "loss": 0.33060954213142396, "time": 0.6491123914718628, "epoch": 1, "memory": 6319, "step": 700} +{"lr": 9.990715904296405e-05, "data_time": 0.0016474723815917969, "loss": 0.3094102323055267, "time": 0.5923159122467041, "epoch": 1, "memory": 6319, "step": 800} +{"lr": 9.988247623470371e-05, "data_time": 0.000794672966003418, "loss": 0.29671770334243774, "time": 0.6268330574035644, "epoch": 1, "memory": 6319, "step": 900} +{"lr": 9.985489149336749e-05, "data_time": 0.001099562644958496, "loss": 0.2709860235452652, "time": 0.6295077800750732, "epoch": 1, "memory": 6319, "step": 1000} +{"lr": 9.982440660248187e-05, "data_time": 0.0008476495742797852, "loss": 0.27203600108623505, "time": 0.6169278621673584, "epoch": 1, "memory": 6319, "step": 1100} +{"lr": 9.979102353308574e-05, "data_time": 0.0008696317672729492, "loss": 0.2540829986333847, "time": 0.633787989616394, "epoch": 1, "memory": 6319, "step": 1200} +{"lr": 9.975474444360358e-05, "data_time": 0.0011459827423095704, "loss": 0.23767077326774597, "time": 0.7133777379989624, "epoch": 1, "memory": 6319, "step": 1300} +{"lr": 9.97155716797056e-05, "data_time": 0.001165318489074707, "loss": 0.2348430186510086, "time": 0.5883892059326172, "epoch": 1, "memory": 6319, "step": 1400} +{"lr": 9.967350777415632e-05, "data_time": 0.0010182857513427734, "loss": 0.2445591375231743, "time": 0.6163933038711548, "epoch": 1, "memory": 6319, "step": 1500} +{"lr": 9.962855544665059e-05, "data_time": 0.001074671745300293, "loss": 0.2130959838628769, "time": 0.6496816873550415, "epoch": 1, "memory": 6319, "step": 1600} +{"lr": 9.958071760363773e-05, "data_time": 0.0015396356582641601, "loss": 0.20838867425918578, "time": 0.639490270614624, "epoch": 1, "memory": 6319, "step": 1700} +{"lr": 9.952999733813382e-05, "data_time": 0.0012651205062866211, "loss": 0.20887960642576217, "time": 0.6244978427886962, "epoch": 1, "memory": 6319, "step": 1800} +{"lr": 9.947639792952148e-05, "data_time": 0.0009616613388061523, "loss": 0.2113117352128029, "time": 0.610913872718811, "epoch": 1, "memory": 6319, "step": 1900} +{"lr": 9.941992284333813e-05, "data_time": 0.001140928268432617, "loss": 0.20041645765304567, "time": 0.6053268194198609, "epoch": 1, "memory": 6319, "step": 2000} +{"lr": 9.93605757310516e-05, "data_time": 0.0008172035217285156, "loss": 0.19543420225381852, "time": 0.6133771657943725, "epoch": 1, "memory": 6319, "step": 2100} +{"lr": 9.929836042982421e-05, "data_time": 0.001177501678466797, "loss": 0.18500066697597503, "time": 0.6748033761978149, "epoch": 1, "memory": 6319, "step": 2200} +{"lr": 9.923328096226467e-05, "data_time": 0.0011435508728027343, "loss": 0.17640187293291093, "time": 0.6820239067077637, "epoch": 1, "memory": 6319, "step": 2300} +{"lr": 9.916534153616819e-05, "data_time": 0.0010052919387817383, "loss": 0.17090136110782622, "time": 0.6188471555709839, "epoch": 1, "memory": 6319, "step": 2400} +{"lr": 9.909454654424387e-05, "data_time": 0.0013057708740234375, "loss": 0.17705828100442886, "time": 0.6501886367797851, "epoch": 1, "memory": 6319, "step": 2500} +{"lr": 9.902090056383109e-05, "data_time": 0.0010050058364868164, "loss": 0.19458670020103455, "time": 0.6512877702713012, "epoch": 1, "memory": 6319, "step": 2600} +{"lr": 9.894440835660365e-05, "data_time": 0.0013167381286621094, "loss": 0.17685725688934326, "time": 0.6358255624771119, "epoch": 1, "memory": 6319, "step": 2700} +{"lr": 9.886507486826162e-05, "data_time": 0.0009731054306030273, "loss": 0.1762370079755783, "time": 0.6680647611618042, "epoch": 1, "memory": 6319, "step": 2800} +{"lr": 9.878290522821154e-05, "data_time": 0.0008811235427856445, "loss": 0.16809552758932114, "time": 0.6001085996627807, "epoch": 1, "memory": 6319, "step": 2900} +{"lr": 9.86979047492352e-05, "data_time": 0.0014863252639770509, "loss": 0.17183809280395507, "time": 0.6313298463821411, "epoch": 1, "memory": 6319, "step": 3000} +{"lr": 9.861007892714578e-05, "data_time": 0.0008005142211914062, "loss": 0.16417725533246993, "time": 0.6330173254013062, "epoch": 1, "memory": 6319, "step": 3100} +{"lr": 9.851943344043239e-05, "data_time": 0.001177072525024414, "loss": 0.17803086191415787, "time": 0.6067479848861694, "epoch": 1, "memory": 6319, "step": 3200} +{"lr": 9.842597414989319e-05, "data_time": 0.0010962724685668946, "loss": 0.14825334995985032, "time": 0.616440725326538, "epoch": 1, "memory": 6319, "step": 3300} +{"lr": 9.832970709825637e-05, "data_time": 0.0012169837951660155, "loss": 0.15710122585296632, "time": 0.656633996963501, "epoch": 1, "memory": 6319, "step": 3400} +{"lr": 9.823063850978968e-05, "data_time": 0.001428532600402832, "loss": 0.14660579934716225, "time": 0.5866067171096802, "epoch": 1, "memory": 6319, "step": 3500} +{"lr": 9.81287747898975e-05, "data_time": 0.001101231575012207, "loss": 0.16227754950523376, "time": 0.6107003211975097, "epoch": 1, "memory": 6319, "step": 3600} +{"lr": 9.802412252470715e-05, "data_time": 0.001011037826538086, "loss": 0.14880369156599044, "time": 0.6640010595321655, "epoch": 1, "memory": 6319, "step": 3700} +{"lr": 9.791668848064258e-05, "data_time": 0.0011841058731079102, "loss": 0.14143735468387603, "time": 0.6097638368606567, "epoch": 1, "memory": 6319, "step": 3800} +{"lr": 9.780647960398751e-05, "data_time": 0.0008812665939331055, "loss": 0.1573878437280655, "time": 0.6231289625167846, "epoch": 1, "memory": 6319, "step": 3900} +{"accuracy/top1": 80.42169952392578, "data_time": 0.6049195339805201, "time": 0.6992707377985904, "step": 1} +{"lr": 9.7685491196646e-05, "data_time": 0.0012707710266113281, "loss": 0.14969086796045303, "time": 0.6566398620605469, "epoch": 2, "memory": 6319, "step": 4007} +{"lr": 9.756956126085676e-05, "data_time": 0.0014353275299072265, "loss": 0.1544960506260395, "time": 0.6339057445526123, "epoch": 2, "memory": 6319, "step": 4107} +{"lr": 9.745087893642482e-05, "data_time": 0.0010685920715332031, "loss": 0.15404153391718864, "time": 0.6071049690246582, "epoch": 2, "memory": 6319, "step": 4207} +{"lr": 9.73294518969056e-05, "data_time": 0.0011847972869873046, "loss": 0.13772937953472136, "time": 0.6508279800415039, "epoch": 2, "memory": 6319, "step": 4307} +{"lr": 9.72052879933172e-05, "data_time": 0.0010708093643188477, "loss": 0.14833975285291673, "time": 0.6094993352890015, "epoch": 2, "memory": 6319, "step": 4407} +{"lr": 9.707839525363358e-05, "data_time": 0.0011272430419921875, "loss": 0.13168747276067733, "time": 0.6026272058486939, "epoch": 2, "memory": 6319, "step": 4507} +{"lr": 9.69487818822646e-05, "data_time": 0.001011037826538086, "loss": 0.15316015034914016, "time": 0.6099385261535645, "epoch": 2, "memory": 6319, "step": 4607} +{"lr": 9.681645625952616e-05, "data_time": 0.0012804508209228516, "loss": 0.13080811202526094, "time": 0.6574455738067627, "epoch": 2, "memory": 6319, "step": 4707} +{"lr": 9.668142694109826e-05, "data_time": 0.0012222766876220704, "loss": 0.1337391011416912, "time": 0.558815860748291, "epoch": 2, "memory": 6319, "step": 4807} +{"lr": 9.654370265747162e-05, "data_time": 0.0013068914413452148, "loss": 0.1303473137319088, "time": 0.6175517082214356, "epoch": 2, "memory": 6319, "step": 4907} +{"lr": 9.640329231338345e-05, "data_time": 0.0012620449066162109, "loss": 0.1228546142578125, "time": 0.6700607061386108, "epoch": 2, "memory": 6319, "step": 5007} +{"lr": 9.626020498724147e-05, "data_time": 0.001144719123840332, "loss": 0.1405038818717003, "time": 0.6171956300735474, "epoch": 2, "memory": 6319, "step": 5107} +{"lr": 9.611444993053717e-05, "data_time": 0.0016186952590942383, "loss": 0.13908120915293692, "time": 0.6053791284561157, "epoch": 2, "memory": 6319, "step": 5207} +{"lr": 9.596603656724746e-05, "data_time": 0.0009818315505981446, "loss": 0.12037205770611763, "time": 0.6270071983337402, "epoch": 2, "memory": 6319, "step": 5307} +{"lr": 9.58149744932254e-05, "data_time": 0.0011154413223266602, "loss": 0.14744246304035186, "time": 0.632693862915039, "epoch": 2, "memory": 6319, "step": 5407} +{"lr": 9.566127347557972e-05, "data_time": 0.001504063606262207, "loss": 0.12885030657052993, "time": 0.6522293090820312, "epoch": 2, "memory": 6319, "step": 5507} +{"lr": 9.550494345204367e-05, "data_time": 0.0012865066528320312, "loss": 0.12040070071816444, "time": 0.6179014205932617, "epoch": 2, "memory": 6319, "step": 5607} +{"lr": 9.534599453033183e-05, "data_time": 0.0012835979461669922, "loss": 0.11735203564167022, "time": 0.5667521238327027, "epoch": 2, "memory": 6319, "step": 5707} +{"lr": 9.518443698748712e-05, "data_time": 0.0011459112167358399, "loss": 0.12665610015392303, "time": 0.6008880376815796, "epoch": 2, "memory": 6319, "step": 5807} +{"lr": 9.50202812692162e-05, "data_time": 0.0010967254638671875, "loss": 0.1326982371509075, "time": 0.6913531303405762, "epoch": 2, "memory": 6319, "step": 5907} +{"lr": 9.485353798921386e-05, "data_time": 0.000941777229309082, "loss": 0.1312527395784855, "time": 0.5970057249069214, "epoch": 2, "memory": 6319, "step": 6007} +{"lr": 9.468421792847712e-05, "data_time": 0.0009882926940917968, "loss": 0.11886056512594223, "time": 0.6437604427337646, "epoch": 2, "memory": 6319, "step": 6107} +{"lr": 9.451233203460794e-05, "data_time": 0.001193690299987793, "loss": 0.12179807424545289, "time": 0.6047776699066162, "epoch": 2, "memory": 6319, "step": 6207} +{"lr": 9.433789142110532e-05, "data_time": 0.0008589029312133789, "loss": 0.12132530137896538, "time": 0.6325850009918212, "epoch": 2, "memory": 6319, "step": 6307} +{"lr": 9.416090736664709e-05, "data_time": 0.0008988857269287109, "loss": 0.1269269645214081, "time": 0.6048826694488525, "epoch": 2, "memory": 6319, "step": 6407} +{"lr": 9.398139131436032e-05, "data_time": 0.0009505748748779297, "loss": 0.10670752972364425, "time": 0.6421198368072509, "epoch": 2, "memory": 6319, "step": 6507} +{"lr": 9.379935487108167e-05, "data_time": 0.0009618759155273438, "loss": 0.12039944976568222, "time": 0.600836992263794, "epoch": 2, "memory": 6319, "step": 6607} +{"lr": 9.361480980660697e-05, "data_time": 0.024141836166381835, "loss": 0.13166357204318047, "time": 0.5980572462081909, "epoch": 2, "memory": 6319, "step": 6707} +{"lr": 9.342776805292993e-05, "data_time": 0.0010269880294799805, "loss": 0.11618372052907944, "time": 0.5672625541687012, "epoch": 2, "memory": 6319, "step": 6807} +{"lr": 9.323824170347083e-05, "data_time": 0.0013018369674682618, "loss": 0.12470941171050072, "time": 0.5922423839569092, "epoch": 2, "memory": 6319, "step": 6907} +{"lr": 9.304624301229471e-05, "data_time": 0.0011385202407836914, "loss": 0.09987586140632629, "time": 0.6147332191467285, "epoch": 2, "memory": 6319, "step": 7007} +{"lr": 9.285178439331897e-05, "data_time": 0.0015858173370361327, "loss": 0.1218492977321148, "time": 0.6371518611907959, "epoch": 2, "memory": 6319, "step": 7107} +{"lr": 9.265487841951087e-05, "data_time": 0.0012993812561035156, "loss": 0.11031495705246926, "time": 0.6388080835342407, "epoch": 2, "memory": 6319, "step": 7207} +{"lr": 9.245553782207421e-05, "data_time": 0.0012434720993041992, "loss": 0.11798710525035858, "time": 0.6172866344451904, "epoch": 2, "memory": 6319, "step": 7307} +{"lr": 9.225377548962671e-05, "data_time": 0.0010834455490112305, "loss": 0.1123718038201332, "time": 0.6374030828475952, "epoch": 2, "memory": 6319, "step": 7407} +{"lr": 9.204960446736601e-05, "data_time": 0.0010797739028930663, "loss": 0.12735430300235748, "time": 0.6130917787551879, "epoch": 2, "memory": 6319, "step": 7507} +{"lr": 9.184303795622715e-05, "data_time": 0.0013530254364013672, "loss": 0.12367594540119171, "time": 0.5850515604019165, "epoch": 2, "memory": 6319, "step": 7607} +{"lr": 9.163408931202823e-05, "data_time": 0.0009424686431884766, "loss": 0.109854356944561, "time": 0.55736825466156, "epoch": 2, "memory": 6319, "step": 7707} +{"lr": 9.142277204460729e-05, "data_time": 0.0008727788925170898, "loss": 0.1303892105817795, "time": 0.5944935798645019, "epoch": 2, "memory": 6319, "step": 7807} +{"accuracy/top1": 78.64723205566406, "data_time": 0.5207227797343813, "time": 0.6123008316960828, "step": 2} +{"lr": 9.119405492383288e-05, "data_time": 0.15969440937042237, "loss": 0.10292618237435817, "time": 0.644035029411316, "epoch": 3, "memory": 6319, "step": 7914} +{"lr": 9.097787819233205e-05, "data_time": 0.001564788818359375, "loss": 0.11451506689190864, "time": 0.5979749441146851, "epoch": 3, "memory": 6319, "step": 8014} +{"lr": 9.075937526576627e-05, "data_time": 0.0013667106628417968, "loss": 0.10959412232041359, "time": 0.6024105310440063, "epoch": 3, "memory": 6319, "step": 8114} +{"lr": 9.053856027171755e-05, "data_time": 0.0014506101608276368, "loss": 0.1103747770190239, "time": 0.5998272657394409, "epoch": 3, "memory": 6319, "step": 8214} +{"lr": 9.031544748725785e-05, "data_time": 0.0011715888977050781, "loss": 0.0986387774348259, "time": 0.6282076597213745, "epoch": 3, "memory": 6319, "step": 8314} +{"lr": 9.009005133802542e-05, "data_time": 0.0011480093002319337, "loss": 0.11372050642967224, "time": 0.6023003101348877, "epoch": 3, "memory": 6319, "step": 8414} +{"lr": 8.986238639729242e-05, "data_time": 0.0017016172409057618, "loss": 0.10751230046153068, "time": 0.625498366355896, "epoch": 3, "memory": 6319, "step": 8514} +{"lr": 8.963246738502254e-05, "data_time": 0.0009747982025146484, "loss": 0.11333387643098831, "time": 0.5796732902526855, "epoch": 3, "memory": 6319, "step": 8614} +{"lr": 8.940030916691928e-05, "data_time": 0.00117340087890625, "loss": 0.11304366365075111, "time": 0.5906398296356201, "epoch": 3, "memory": 6319, "step": 8714} +{"lr": 8.91659267534649e-05, "data_time": 0.001058197021484375, "loss": 0.1090011790394783, "time": 0.5973772287368775, "epoch": 3, "memory": 6319, "step": 8814} +{"lr": 8.892933529894978e-05, "data_time": 0.0010092020034790038, "loss": 0.1150609865784645, "time": 0.6059458971023559, "epoch": 3, "memory": 6319, "step": 8914} +{"lr": 8.869055010049283e-05, "data_time": 0.001197481155395508, "loss": 0.0966479279100895, "time": 0.6162628173828125, "epoch": 3, "memory": 6319, "step": 9014} +{"lr": 8.844958659705188e-05, "data_time": 0.0010462284088134765, "loss": 0.11113687679171562, "time": 0.6096754789352417, "epoch": 3, "memory": 6319, "step": 9114} +{"lr": 8.82064603684259e-05, "data_time": 0.0011551618576049806, "loss": 0.10108362510800362, "time": 0.6459286689758301, "epoch": 3, "memory": 6319, "step": 9214} +{"lr": 8.796118713424776e-05, "data_time": 0.0009507179260253906, "loss": 0.09558529742062091, "time": 0.5573125123977661, "epoch": 3, "memory": 6319, "step": 9314} +{"lr": 8.77137827529676e-05, "data_time": 0.0015813350677490235, "loss": 0.10445983856916427, "time": 0.6569721221923828, "epoch": 3, "memory": 6319, "step": 9414} +{"lr": 8.746426322082745e-05, "data_time": 0.00175018310546875, "loss": 0.09164994284510612, "time": 0.6291789770126343, "epoch": 3, "memory": 6319, "step": 9514} +{"lr": 8.721264467082732e-05, "data_time": 0.001194596290588379, "loss": 0.09945103079080582, "time": 0.6127496957778931, "epoch": 3, "memory": 6319, "step": 9614} +{"lr": 8.695894337168168e-05, "data_time": 0.0013375282287597656, "loss": 0.09632962122559548, "time": 0.6105611085891723, "epoch": 3, "memory": 6319, "step": 9714} +{"lr": 8.67031757267677e-05, "data_time": 0.0013091325759887694, "loss": 0.09921496212482453, "time": 0.6229627847671508, "epoch": 3, "memory": 6319, "step": 9814} +{"lr": 8.64453582730651e-05, "data_time": 0.001412367820739746, "loss": 0.10290349721908569, "time": 0.6016729831695556, "epoch": 3, "memory": 6319, "step": 9914} +{"lr": 8.61855076800861e-05, "data_time": 0.0011766910552978515, "loss": 0.09510720074176789, "time": 0.6402531623840332, "epoch": 3, "memory": 6319, "step": 10014} +{"lr": 8.592364074879843e-05, "data_time": 0.001172351837158203, "loss": 0.10642112605273724, "time": 0.6265649795532227, "epoch": 3, "memory": 6319, "step": 10114} +{"lr": 8.565977441053854e-05, "data_time": 0.0015697240829467773, "loss": 0.11050756648182869, "time": 0.779146385192871, "epoch": 3, "memory": 6319, "step": 10214} +{"lr": 8.539392572591724e-05, "data_time": 0.0010590076446533204, "loss": 0.10465100929141044, "time": 0.5467787504196167, "epoch": 3, "memory": 6319, "step": 10314} +{"lr": 8.512611188371633e-05, "data_time": 0.0012309551239013672, "loss": 0.10303699672222137, "time": 0.5611342430114746, "epoch": 3, "memory": 6319, "step": 10414} +{"lr": 8.485635019977725e-05, "data_time": 0.0011582136154174804, "loss": 0.10508915334939957, "time": 0.5413308620452881, "epoch": 3, "memory": 6319, "step": 10514} +{"lr": 8.45846581158818e-05, "data_time": 0.0011284828186035156, "loss": 0.10682142674922943, "time": 0.6266285657882691, "epoch": 3, "memory": 6319, "step": 10614} +{"lr": 8.431105319862416e-05, "data_time": 0.0010250329971313477, "loss": 0.08967098034918308, "time": 0.6155362606048584, "epoch": 3, "memory": 6319, "step": 10714} +{"lr": 8.403555313827516e-05, "data_time": 0.0010985374450683595, "loss": 0.10303728207945824, "time": 0.6193816900253296, "epoch": 3, "memory": 6319, "step": 10814} +{"lr": 8.375817574763844e-05, "data_time": 0.001214146614074707, "loss": 0.09502370059490203, "time": 0.5690989971160889, "epoch": 3, "memory": 6319, "step": 10914} +{"lr": 8.347893896089871e-05, "data_time": 0.001226329803466797, "loss": 0.0937697671353817, "time": 0.5626082420349121, "epoch": 3, "memory": 6319, "step": 11014} +{"lr": 8.31978608324625e-05, "data_time": 0.0011416196823120118, "loss": 0.0888057790696621, "time": 0.5901943683624268, "epoch": 3, "memory": 6319, "step": 11114} +{"lr": 8.291495953579052e-05, "data_time": 0.0014118909835815429, "loss": 0.09242230579257012, "time": 0.6065904855728149, "epoch": 3, "memory": 6319, "step": 11214} +{"lr": 8.263025336222267e-05, "data_time": 0.0009230136871337891, "loss": 0.11187078952789306, "time": 0.8397389650344849, "epoch": 3, "memory": 6319, "step": 11314} +{"lr": 8.234376071979551e-05, "data_time": 0.000915980339050293, "loss": 0.1063428483903408, "time": 0.5562231540679932, "epoch": 3, "memory": 6319, "step": 11414} +{"lr": 8.205550013205187e-05, "data_time": 0.0012593984603881836, "loss": 0.10742152333259583, "time": 0.6229443073272705, "epoch": 3, "memory": 6319, "step": 11514} +{"lr": 8.17654902368435e-05, "data_time": 0.001011800765991211, "loss": 0.09773860052227974, "time": 0.5700891733169555, "epoch": 3, "memory": 6319, "step": 11614} +{"lr": 8.147374978512582e-05, "data_time": 0.0011054039001464843, "loss": 0.09565523415803909, "time": 0.619389533996582, "epoch": 3, "memory": 6319, "step": 11714} +{"accuracy/top1": 82.1822509765625, "data_time": 0.5461730874817947, "time": 0.6365683078765869, "step": 3} +{"lr": 8.115969237626578e-05, "data_time": 0.12131602764129638, "loss": 0.10533967837691308, "time": 0.5874639987945557, "epoch": 4, "memory": 6319, "step": 11821} +{"lr": 8.08644297337245e-05, "data_time": 0.2184462308883667, "loss": 0.09577877409756183, "time": 0.6172070026397705, "epoch": 4, "memory": 6319, "step": 11921} +{"lr": 8.056749479387651e-05, "data_time": 0.0014777660369873046, "loss": 0.0914388693869114, "time": 0.6325190544128418, "epoch": 4, "memory": 6319, "step": 12021} +{"lr": 8.026890675542457e-05, "data_time": 0.001353621482849121, "loss": 0.1017810121178627, "time": 0.5816426753997803, "epoch": 4, "memory": 6319, "step": 12121} +{"lr": 7.996868492395457e-05, "data_time": 0.0011809110641479493, "loss": 0.09612506739795208, "time": 0.6351016759872437, "epoch": 4, "memory": 6319, "step": 12221} +{"lr": 7.966684871068762e-05, "data_time": 0.0012735843658447266, "loss": 0.0993888571858406, "time": 0.6687037706375122, "epoch": 4, "memory": 6319, "step": 12321} +{"lr": 7.936341763122449e-05, "data_time": 0.0008375644683837891, "loss": 0.08615965768694878, "time": 0.6186165571212768, "epoch": 4, "memory": 6319, "step": 12421} +{"lr": 7.905841130428405e-05, "data_time": 0.0008159875869750977, "loss": 0.09248109087347985, "time": 0.5616381406784058, "epoch": 4, "memory": 6319, "step": 12521} +{"lr": 7.875184945043487e-05, "data_time": 0.0009601831436157227, "loss": 0.099121855199337, "time": 0.6511362314224243, "epoch": 4, "memory": 6319, "step": 12621} +{"lr": 7.844375189082012e-05, "data_time": 0.0011842489242553712, "loss": 0.0778490923345089, "time": 0.6310737371444702, "epoch": 4, "memory": 6319, "step": 12721} +{"lr": 7.813413854587572e-05, "data_time": 0.0009321212768554687, "loss": 0.1054179273545742, "time": 0.5926837205886841, "epoch": 4, "memory": 6319, "step": 12821} +{"lr": 7.782302943404291e-05, "data_time": 0.0012653827667236327, "loss": 0.105924541503191, "time": 0.6521082401275635, "epoch": 4, "memory": 6319, "step": 12921} +{"lr": 7.751044467047337e-05, "data_time": 0.0011621475219726562, "loss": 0.09382660947740078, "time": 0.6130581378936768, "epoch": 4, "memory": 6319, "step": 13021} +{"lr": 7.719640446572907e-05, "data_time": 0.0015146970748901368, "loss": 0.09103393107652664, "time": 0.6381123304367066, "epoch": 4, "memory": 6319, "step": 13121} +{"lr": 7.688092912447536e-05, "data_time": 0.0009727716445922851, "loss": 0.09927580878138542, "time": 0.615462851524353, "epoch": 4, "memory": 6319, "step": 13221} +{"lr": 7.656403904416794e-05, "data_time": 0.0011687040328979491, "loss": 0.099492472037673, "time": 0.636084508895874, "epoch": 4, "memory": 6319, "step": 13321} +{"lr": 7.624575471373434e-05, "data_time": 0.001276254653930664, "loss": 0.10581168681383132, "time": 0.5897547245025635, "epoch": 4, "memory": 6319, "step": 13421} +{"lr": 7.592609671224926e-05, "data_time": 0.0012411355972290039, "loss": 0.09347496926784515, "time": 0.6397350549697876, "epoch": 4, "memory": 6319, "step": 13521} +{"lr": 7.560508570760375e-05, "data_time": 0.0009140968322753906, "loss": 0.09461783841252328, "time": 0.5972483634948731, "epoch": 4, "memory": 6319, "step": 13621} +{"lr": 7.528274245516895e-05, "data_time": 0.0008981466293334961, "loss": 0.09843641594052314, "time": 0.6742244005203247, "epoch": 4, "memory": 6319, "step": 13721} +{"lr": 7.495908779645405e-05, "data_time": 0.0013815879821777344, "loss": 0.09268547669053077, "time": 0.6965377807617188, "epoch": 4, "memory": 6319, "step": 13821} +{"lr": 7.463414265775906e-05, "data_time": 0.0014081478118896484, "loss": 0.08534050807356834, "time": 0.6147332191467285, "epoch": 4, "memory": 6319, "step": 13921} +{"lr": 7.430792804882147e-05, "data_time": 0.0013270139694213866, "loss": 0.09435487315058708, "time": 0.6108735799789429, "epoch": 4, "memory": 6319, "step": 14021} +{"lr": 7.398046506145803e-05, "data_time": 0.0011119842529296875, "loss": 0.08499853499233723, "time": 0.6191097974777222, "epoch": 4, "memory": 6319, "step": 14121} +{"lr": 7.365177486820114e-05, "data_time": 0.0009569644927978516, "loss": 0.09503398910164833, "time": 0.6604063034057617, "epoch": 4, "memory": 6319, "step": 14221} +{"lr": 7.332187872092939e-05, "data_time": 0.0012288570404052734, "loss": 0.10102737545967103, "time": 0.6483676195144653, "epoch": 4, "memory": 6319, "step": 14321} +{"lr": 7.299079794949401e-05, "data_time": 0.0013399600982666015, "loss": 0.07072069868445396, "time": 0.6595494985580445, "epoch": 4, "memory": 6319, "step": 14421} +{"lr": 7.265855396033955e-05, "data_time": 0.0008251667022705078, "loss": 0.09817356318235397, "time": 0.6921434879302979, "epoch": 4, "memory": 6319, "step": 14521} +{"lr": 7.232516823512013e-05, "data_time": 0.001130533218383789, "loss": 0.08089447729289531, "time": 0.7042551755905151, "epoch": 4, "memory": 6319, "step": 14621} +{"lr": 7.199066232930995e-05, "data_time": 0.0008875131607055664, "loss": 0.08366571292281151, "time": 0.5800805568695069, "epoch": 4, "memory": 6319, "step": 14721} +{"lr": 7.165505787080997e-05, "data_time": 0.001098942756652832, "loss": 0.09997397139668465, "time": 0.6220454454421998, "epoch": 4, "memory": 6319, "step": 14821} +{"lr": 7.13183765585495e-05, "data_time": 0.0012734651565551758, "loss": 0.09645765498280526, "time": 0.6224927186965943, "epoch": 4, "memory": 6319, "step": 14921} +{"lr": 7.098064016108317e-05, "data_time": 0.0010340213775634766, "loss": 0.09494763761758804, "time": 0.6230634450912476, "epoch": 4, "memory": 6319, "step": 15021} +{"lr": 7.064187051518379e-05, "data_time": 0.0009942770004272461, "loss": 0.10012876093387604, "time": 0.6057145833969116, "epoch": 4, "memory": 6319, "step": 15121} +{"lr": 7.030208952442949e-05, "data_time": 0.0014056682586669922, "loss": 0.08946731351315976, "time": 0.6111350059509277, "epoch": 4, "memory": 6319, "step": 15221} +{"lr": 6.996131915778851e-05, "data_time": 0.0009274482727050781, "loss": 0.09231140837073326, "time": 0.6949593305587769, "epoch": 4, "memory": 6319, "step": 15321} +{"lr": 6.961958144819853e-05, "data_time": 0.001043248176574707, "loss": 0.0874828852713108, "time": 0.6230496644973755, "epoch": 4, "memory": 6319, "step": 15421} +{"lr": 6.927689849114193e-05, "data_time": 0.001140284538269043, "loss": 0.09214382246136665, "time": 0.6202146768569946, "epoch": 4, "memory": 6319, "step": 15521} +{"lr": 6.893329244321722e-05, "data_time": 0.0006139516830444336, "loss": 0.09554443918168545, "time": 0.5699955224990845, "epoch": 4, "memory": 6319, "step": 15621} +{"accuracy/top1": 89.25924682617188, "data_time": 0.5373759228607704, "time": 0.626487530511001, "step": 4} +{"lr": 6.856463687361792e-05, "data_time": 0.24332633018493652, "loss": 0.09055482372641563, "time": 0.6087357759475708, "epoch": 5, "memory": 6319, "step": 15728} +{"lr": 6.821919068470432e-05, "data_time": 0.0010714292526245116, "loss": 0.08013462312519551, "time": 0.6210313558578491, "epoch": 5, "memory": 6319, "step": 15828} +{"lr": 6.787288979235407e-05, "data_time": 0.0011994123458862304, "loss": 0.08548355773091317, "time": 0.5748008966445923, "epoch": 5, "memory": 6319, "step": 15928} +{"lr": 6.752575658708793e-05, "data_time": 0.0012803792953491211, "loss": 0.08600676655769349, "time": 0.6464286327362061, "epoch": 5, "memory": 6319, "step": 16028} +{"lr": 6.71778135132408e-05, "data_time": 0.0011774539947509766, "loss": 0.08509748727083206, "time": 0.7035858869552613, "epoch": 5, "memory": 6319, "step": 16128} +{"lr": 6.682908306751074e-05, "data_time": 0.0012784481048583984, "loss": 0.08347053341567516, "time": 0.5974216938018799, "epoch": 5, "memory": 6319, "step": 16228} +{"lr": 6.64795877975043e-05, "data_time": 0.0013874530792236327, "loss": 0.09401379153132439, "time": 0.6251525402069091, "epoch": 5, "memory": 6319, "step": 16328} +{"lr": 6.61293503002787e-05, "data_time": 0.0012379169464111328, "loss": 0.08163192085921764, "time": 0.7113333225250245, "epoch": 5, "memory": 6319, "step": 16428} +{"lr": 6.577839322088084e-05, "data_time": 0.0011114835739135741, "loss": 0.08945540450513363, "time": 0.5497198104858398, "epoch": 5, "memory": 6319, "step": 16528} +{"lr": 6.542673925088298e-05, "data_time": 0.0009516000747680664, "loss": 0.0858134638518095, "time": 0.6087232112884522, "epoch": 5, "memory": 6319, "step": 16628} +{"lr": 6.507441112691584e-05, "data_time": 0.0009476184844970703, "loss": 0.09904510974884033, "time": 0.5784173727035522, "epoch": 5, "memory": 6319, "step": 16728} +{"lr": 6.47214316291983e-05, "data_time": 0.0019447803497314453, "loss": 0.08314869739115238, "time": 0.6415282249450683, "epoch": 5, "memory": 6319, "step": 16828} +{"lr": 6.436782358006481e-05, "data_time": 0.0011305570602416991, "loss": 0.08982470408082008, "time": 0.6326637506484986, "epoch": 5, "memory": 6319, "step": 16928} +{"lr": 6.40136098424896e-05, "data_time": 0.0015105724334716797, "loss": 0.08660290613770485, "time": 0.5824389696121216, "epoch": 5, "memory": 6319, "step": 17028} +{"lr": 6.365881331860828e-05, "data_time": 0.0012520551681518555, "loss": 0.07760576084256172, "time": 0.6218335151672363, "epoch": 5, "memory": 6319, "step": 17128} +{"lr": 6.330345694823739e-05, "data_time": 0.0010503530502319336, "loss": 0.08954662829637527, "time": 0.6022094488143921, "epoch": 5, "memory": 6319, "step": 17228} +{"lr": 6.29475637073908e-05, "data_time": 0.0014076948165893555, "loss": 0.07359700575470925, "time": 0.6077304363250733, "epoch": 5, "memory": 6319, "step": 17328} +{"lr": 6.259115660679482e-05, "data_time": 0.0012317180633544921, "loss": 0.0839787632226944, "time": 0.6005590438842774, "epoch": 5, "memory": 6319, "step": 17428} +{"lr": 6.223425869039963e-05, "data_time": 0.001350569725036621, "loss": 0.0748495526611805, "time": 0.5905034065246582, "epoch": 5, "memory": 6319, "step": 17528} +{"lr": 6.187689303389006e-05, "data_time": 0.001175689697265625, "loss": 0.09200113415718078, "time": 0.6499512434005738, "epoch": 5, "memory": 6319, "step": 17628} +{"lr": 6.151908274319292e-05, "data_time": 0.0010364294052124024, "loss": 0.09374405071139336, "time": 0.7825910091400147, "epoch": 5, "memory": 6319, "step": 17728} +{"lr": 6.116085095298373e-05, "data_time": 0.0008411169052124024, "loss": 0.07329643815755844, "time": 0.5486382722854615, "epoch": 5, "memory": 6319, "step": 17828} +{"lr": 6.080222082519007e-05, "data_time": 0.0009914159774780274, "loss": 0.08581202402710915, "time": 0.6193802118301391, "epoch": 5, "memory": 6319, "step": 17928} +{"lr": 6.0443215547495147e-05, "data_time": 0.0008234262466430664, "loss": 0.0874987043440342, "time": 0.6451940298080444, "epoch": 5, "memory": 6319, "step": 18028} +{"lr": 6.0083858331837714e-05, "data_time": 0.0007999897003173828, "loss": 0.08538103513419629, "time": 0.6099106550216675, "epoch": 5, "memory": 6319, "step": 18128} +{"lr": 5.97241724129116e-05, "data_time": 0.0013352394104003905, "loss": 0.08670755960047245, "time": 0.61390061378479, "epoch": 5, "memory": 6319, "step": 18228} +{"lr": 5.936418104666328e-05, "data_time": 0.0012809276580810548, "loss": 0.09755421504378319, "time": 0.6391181945800781, "epoch": 5, "memory": 6319, "step": 18328} +{"lr": 5.9003907508788295e-05, "data_time": 0.0012899398803710937, "loss": 0.08166025988757611, "time": 0.6239294052124024, "epoch": 5, "memory": 6319, "step": 18428} +{"lr": 5.864337509322633e-05, "data_time": 0.001182866096496582, "loss": 0.08269152827560902, "time": 0.7048297643661499, "epoch": 5, "memory": 6319, "step": 18528} +{"lr": 5.8282607110655194e-05, "data_time": 0.0009640216827392578, "loss": 0.09006138816475869, "time": 0.6218681097030639, "epoch": 5, "memory": 6319, "step": 18628} +{"lr": 5.7921626886983564e-05, "data_time": 0.001384139060974121, "loss": 0.08800899088382722, "time": 0.6342753410339356, "epoch": 5, "memory": 6319, "step": 18728} +{"lr": 5.756045776184287e-05, "data_time": 0.0013786077499389649, "loss": 0.08197480849921704, "time": 0.6153100728988647, "epoch": 5, "memory": 6319, "step": 18828} +{"lr": 5.719912308707808e-05, "data_time": 0.0013013362884521484, "loss": 0.071322251111269, "time": 0.618512749671936, "epoch": 5, "memory": 6319, "step": 18928} +{"lr": 5.683764622523808e-05, "data_time": 0.000980710983276367, "loss": 0.06396436169743538, "time": 0.5832077741622925, "epoch": 5, "memory": 6319, "step": 19028} +{"lr": 5.647605054806502e-05, "data_time": 0.0015149831771850586, "loss": 0.07964058071374894, "time": 0.6258903741836548, "epoch": 5, "memory": 6319, "step": 19128} +{"lr": 5.6114359434983206e-05, "data_time": 0.001742386817932129, "loss": 0.07741358056664467, "time": 0.6207758665084839, "epoch": 5, "memory": 6319, "step": 19228} +{"lr": 5.575259627158745e-05, "data_time": 0.04519574642181397, "loss": 0.07444062791764736, "time": 0.6512256145477295, "epoch": 5, "memory": 6319, "step": 19328} +{"lr": 5.539078444813113e-05, "data_time": 0.0018009424209594726, "loss": 0.08687329590320587, "time": 0.6746395587921142, "epoch": 5, "memory": 6319, "step": 19428} +{"lr": 5.5028947358013735e-05, "data_time": 0.0007738590240478516, "loss": 0.08167134039103985, "time": 0.5811173439025878, "epoch": 5, "memory": 6319, "step": 19528} +{"accuracy/top1": 93.68498229980469, "data_time": 0.4931796049249583, "time": 0.5811454838719862, "step": 5} +{"lr": 5.464178020340108e-05, "data_time": 0.2116227149963379, "loss": 0.08711279481649399, "time": 0.5857839107513427, "epoch": 6, "memory": 6319, "step": 19635} +{"lr": 5.427996514795046e-05, "data_time": 0.05478310585021973, "loss": 0.09860560074448585, "time": 0.6000259399414063, "epoch": 6, "memory": 6319, "step": 19735} +{"lr": 5.3918196647260834e-05, "data_time": 0.0011778831481933593, "loss": 0.07608385905623435, "time": 0.5921300888061524, "epoch": 6, "memory": 6319, "step": 19835} +{"lr": 5.355649809193057e-05, "data_time": 0.0011354446411132812, "loss": 0.08391941040754318, "time": 0.5497043371200562, "epoch": 6, "memory": 6319, "step": 19935} +{"lr": 5.3194892868035765e-05, "data_time": 0.001302337646484375, "loss": 0.07991163469851018, "time": 0.5998502731323242, "epoch": 6, "memory": 6319, "step": 20035} +{"lr": 5.283340435561795e-05, "data_time": 0.0012693166732788085, "loss": 0.0871560662984848, "time": 0.577147650718689, "epoch": 6, "memory": 6319, "step": 20135} +{"lr": 5.2472055927172495e-05, "data_time": 0.00094451904296875, "loss": 0.07504497133195401, "time": 0.6361608505249023, "epoch": 6, "memory": 6319, "step": 20235} +{"lr": 5.2110870946137605e-05, "data_time": 0.0010939598083496093, "loss": 0.07807315681129694, "time": 0.6203902244567872, "epoch": 6, "memory": 6319, "step": 20335} +{"lr": 5.174987276538342e-05, "data_time": 0.0010441303253173827, "loss": 0.08019522801041604, "time": 0.639669394493103, "epoch": 6, "memory": 6319, "step": 20435} +{"lr": 5.1389084725702425e-05, "data_time": 0.0011540651321411133, "loss": 0.08820452354848385, "time": 0.5844098329544067, "epoch": 6, "memory": 6319, "step": 20535} +{"lr": 5.1028530154300045e-05, "data_time": 0.0009647369384765625, "loss": 0.08078234754502774, "time": 0.6114591121673584, "epoch": 6, "memory": 6319, "step": 20635} +{"lr": 5.066823236328663e-05, "data_time": 0.0011961221694946288, "loss": 0.06776001751422882, "time": 0.6076150178909302, "epoch": 6, "memory": 6319, "step": 20735} +{"lr": 5.030821464817006e-05, "data_time": 0.0010621070861816407, "loss": 0.07857253141701222, "time": 0.5769826889038085, "epoch": 6, "memory": 6319, "step": 20835} +{"lr": 4.99485002863494e-05, "data_time": 0.000817108154296875, "loss": 0.08330013453960419, "time": 0.6137124538421631, "epoch": 6, "memory": 6319, "step": 20935} +{"lr": 4.95891125356102e-05, "data_time": 0.0013603448867797851, "loss": 0.08146739304065705, "time": 0.7235180854797363, "epoch": 6, "memory": 6319, "step": 21035} +{"lr": 4.923007463262053e-05, "data_time": 0.0011154413223266602, "loss": 0.08484194241464138, "time": 0.6746965646743774, "epoch": 6, "memory": 6319, "step": 21135} +{"lr": 4.887140979142852e-05, "data_time": 0.0011101007461547852, "loss": 0.07396330647170543, "time": 0.8727160215377807, "epoch": 6, "memory": 6319, "step": 21235} +{"lr": 4.851314120196161e-05, "data_time": 0.0008193731307983398, "loss": 0.07735928781330585, "time": 0.5724152326583862, "epoch": 6, "memory": 6319, "step": 21335} +{"lr": 4.815529202852693e-05, "data_time": 0.0011681795120239257, "loss": 0.09563387855887413, "time": 0.6191658496856689, "epoch": 6, "memory": 6319, "step": 21435} +{"lr": 4.779788540831406e-05, "data_time": 0.0009724378585815429, "loss": 0.07255338467657566, "time": 0.5283006191253662, "epoch": 6, "memory": 6319, "step": 21535} +{"lr": 4.744094444989843e-05, "data_time": 0.020735740661621094, "loss": 0.08015012815594673, "time": 0.6079918384552002, "epoch": 6, "memory": 6319, "step": 21635} +{"lr": 4.708449223174764e-05, "data_time": 0.0012578010559082032, "loss": 0.07504624053835869, "time": 0.603456997871399, "epoch": 6, "memory": 6319, "step": 21735} +{"lr": 4.672855180072916e-05, "data_time": 0.0011845111846923828, "loss": 0.06350935623049736, "time": 0.6788998365402221, "epoch": 6, "memory": 6319, "step": 21835} +{"lr": 4.6373146170620256e-05, "data_time": 0.000836491584777832, "loss": 0.06916520148515701, "time": 0.6389220952987671, "epoch": 6, "memory": 6319, "step": 21935} +{"lr": 4.6018298320619915e-05, "data_time": 0.0008458375930786132, "loss": 0.07418507188558579, "time": 0.620337438583374, "epoch": 6, "memory": 6319, "step": 22035} +{"lr": 4.566403119386318e-05, "data_time": 0.0010722875595092773, "loss": 0.06531587988138199, "time": 0.5736503839492798, "epoch": 6, "memory": 6319, "step": 22135} +{"lr": 4.531036769593769e-05, "data_time": 0.0012825965881347657, "loss": 0.07403096482157707, "time": 0.6477460145950318, "epoch": 6, "memory": 6319, "step": 22235} +{"lr": 4.4957330693402685e-05, "data_time": 0.0013995170593261719, "loss": 0.08489035218954086, "time": 0.6271989107131958, "epoch": 6, "memory": 6319, "step": 22335} +{"lr": 4.460494301231059e-05, "data_time": 0.0011365652084350587, "loss": 0.0758124865591526, "time": 0.6079889297485351, "epoch": 6, "memory": 6319, "step": 22435} +{"lr": 4.425322743673109e-05, "data_time": 0.0012985944747924804, "loss": 0.07123874612152577, "time": 0.6142471313476563, "epoch": 6, "memory": 6319, "step": 22535} +{"lr": 4.390220670727812e-05, "data_time": 0.0010303497314453126, "loss": 0.08152568638324738, "time": 0.5848260164260864, "epoch": 6, "memory": 6319, "step": 22635} +{"lr": 4.3551903519639325e-05, "data_time": 0.0010384321212768555, "loss": 0.08272028975188732, "time": 0.6074117660522461, "epoch": 6, "memory": 6319, "step": 22735} +{"lr": 4.3202340523108894e-05, "data_time": 0.0009337425231933594, "loss": 0.06299005895853042, "time": 0.6091707944869995, "epoch": 6, "memory": 6319, "step": 22835} +{"lr": 4.285354031912302e-05, "data_time": 0.001403045654296875, "loss": 0.08641164749860764, "time": 0.6172260999679565, "epoch": 6, "memory": 6319, "step": 22935} +{"lr": 4.2505525459798544e-05, "data_time": 0.0010636806488037109, "loss": 0.07034889832139016, "time": 0.6003542184829712, "epoch": 6, "memory": 6319, "step": 23035} +{"lr": 4.2158318446474836e-05, "data_time": 0.0008357524871826172, "loss": 0.0810418576002121, "time": 0.5945213556289672, "epoch": 6, "memory": 6319, "step": 23135} +{"lr": 4.181194172825905e-05, "data_time": 0.001106858253479004, "loss": 0.07436636798083782, "time": 0.661811375617981, "epoch": 6, "memory": 6319, "step": 23235} +{"lr": 4.146641770057448e-05, "data_time": 0.00128629207611084, "loss": 0.07141970098018646, "time": 0.6103588342666626, "epoch": 6, "memory": 6319, "step": 23335} +{"lr": 4.112176870371267e-05, "data_time": 0.0008770465850830078, "loss": 0.08884702026844024, "time": 0.6611963748931885, "epoch": 6, "memory": 6319, "step": 23435} +{"accuracy/top1": 94.78445434570312, "data_time": 0.5004175860306312, "time": 0.5898567602552217, "step": 6} +{"lr": 4.0753988582732786e-05, "data_time": 0.474355149269104, "loss": 0.06689820997416973, "time": 0.6163939237594604, "epoch": 7, "memory": 6319, "step": 23542} +{"lr": 4.0411221639286815e-05, "data_time": 0.525312352180481, "loss": 0.0629511084407568, "time": 0.6700114965438843, "epoch": 7, "memory": 6319, "step": 23642} +{"lr": 4.006939795169518e-05, "data_time": 0.4356280565261841, "loss": 0.07155358083546162, "time": 0.5888641119003296, "epoch": 7, "memory": 6319, "step": 23742} +{"lr": 3.972853962099913e-05, "data_time": 0.4349459886550903, "loss": 0.08872487097978592, "time": 0.5881338596343995, "epoch": 7, "memory": 6319, "step": 23842} +{"lr": 3.93886686858237e-05, "data_time": 0.49597482681274413, "loss": 0.08656098246574402, "time": 0.6514045715332031, "epoch": 7, "memory": 6319, "step": 23942} +{"lr": 3.904980712095259e-05, "data_time": 0.42073636054992675, "loss": 0.07925464436411858, "time": 0.564240574836731, "epoch": 7, "memory": 6319, "step": 24042} +{"lr": 3.871197683590748e-05, "data_time": 0.5220647573471069, "loss": 0.08019356690347194, "time": 0.6643844366073608, "epoch": 7, "memory": 6319, "step": 24142} +{"lr": 3.837519967353125e-05, "data_time": 0.4823348045349121, "loss": 0.08155170679092408, "time": 0.6275966882705688, "epoch": 7, "memory": 6319, "step": 24242} +{"lr": 3.8039497408575855e-05, "data_time": 0.4871502876281738, "loss": 0.07777530662715434, "time": 0.6438503742218018, "epoch": 7, "memory": 6319, "step": 24342} +{"lr": 3.770489174629432e-05, "data_time": 0.4617809534072876, "loss": 0.07056905478239059, "time": 0.6046681642532349, "epoch": 7, "memory": 6319, "step": 24442} +{"lr": 3.737140432103754e-05, "data_time": 0.45655364990234376, "loss": 0.0921558927744627, "time": 0.6123908042907715, "epoch": 7, "memory": 6319, "step": 24542} +{"lr": 3.703905669485545e-05, "data_time": 0.46669831275939944, "loss": 0.08235384598374366, "time": 0.6098385810852051, "epoch": 7, "memory": 6319, "step": 24642} +{"lr": 3.6707870356102706e-05, "data_time": 0.4210800647735596, "loss": 0.08048867210745811, "time": 0.563645315170288, "epoch": 7, "memory": 6319, "step": 24742} +{"lr": 3.6377866718049484e-05, "data_time": 0.4014909267425537, "loss": 0.06895597763359547, "time": 0.5435338258743286, "epoch": 7, "memory": 6319, "step": 24842} +{"lr": 3.6049067117496965e-05, "data_time": 0.5016248464584351, "loss": 0.06598893962800503, "time": 0.6431945562362671, "epoch": 7, "memory": 6319, "step": 24942} +{"lr": 3.5721492813397755e-05, "data_time": 0.45930278301239014, "loss": 0.07469790577888488, "time": 0.6020509719848632, "epoch": 7, "memory": 6319, "step": 25042} +{"lr": 3.539516498548151e-05, "data_time": 0.4277693510055542, "loss": 0.07332868464291095, "time": 0.5700596570968628, "epoch": 7, "memory": 6319, "step": 25142} +{"lr": 3.507010473288518e-05, "data_time": 0.4210413694381714, "loss": 0.07406840659677982, "time": 0.5631956100463867, "epoch": 7, "memory": 6319, "step": 25242} +{"lr": 3.474633307278909e-05, "data_time": 0.48797500133514404, "loss": 0.0916257381439209, "time": 0.6304330587387085, "epoch": 7, "memory": 6319, "step": 25342} +{"lr": 3.4423870939058177e-05, "data_time": 0.462441611289978, "loss": 0.09362282380461692, "time": 0.633586072921753, "epoch": 7, "memory": 6319, "step": 25442} +{"lr": 3.410273918088805e-05, "data_time": 0.45768418312072756, "loss": 0.07509752325713634, "time": 0.6017839193344117, "epoch": 7, "memory": 6319, "step": 25542} +{"lr": 3.37829585614574e-05, "data_time": 0.4711894989013672, "loss": 0.09683760479092599, "time": 0.61318039894104, "epoch": 7, "memory": 6319, "step": 25642} +{"lr": 3.3464549756585216e-05, "data_time": 0.43604578971862795, "loss": 0.08644391037523746, "time": 0.5796491622924804, "epoch": 7, "memory": 6319, "step": 25742} +{"lr": 3.314753335339416e-05, "data_time": 0.502040696144104, "loss": 0.0850952073931694, "time": 0.6460355997085572, "epoch": 7, "memory": 6319, "step": 25842} +{"lr": 3.2831929848979394e-05, "data_time": 0.48782057762145997, "loss": 0.07723802626132965, "time": 0.6405158042907715, "epoch": 7, "memory": 6319, "step": 25942} +{"lr": 3.251775964908332e-05, "data_time": 0.48589468002319336, "loss": 0.08428933471441269, "time": 0.6298029899597168, "epoch": 7, "memory": 6319, "step": 26042} +{"lr": 3.22050430667762e-05, "data_time": 0.47635655403137206, "loss": 0.08189189061522484, "time": 0.6201178073883057, "epoch": 7, "memory": 6319, "step": 26142} +{"lr": 3.189380032114285e-05, "data_time": 0.5052550315856934, "loss": 0.09040755890309811, "time": 0.6483135938644409, "epoch": 7, "memory": 6319, "step": 26242} +{"lr": 3.1584051535975324e-05, "data_time": 0.4549750328063965, "loss": 0.07514316774904728, "time": 0.6086362361907959, "epoch": 7, "memory": 6319, "step": 26342} +{"lr": 3.127581673847175e-05, "data_time": 0.4394742250442505, "loss": 0.08260803520679474, "time": 0.5820672035217285, "epoch": 7, "memory": 6319, "step": 26442} +{"lr": 3.0969115857941555e-05, "data_time": 0.48485751152038575, "loss": 0.09186746180057526, "time": 0.629362678527832, "epoch": 7, "memory": 6319, "step": 26542} +{"lr": 3.0663968724516654e-05, "data_time": 0.4046575307846069, "loss": 0.0835926216095686, "time": 0.5463639497756958, "epoch": 7, "memory": 6319, "step": 26642} +{"lr": 3.0360395067869668e-05, "data_time": 0.43217408657073975, "loss": 0.08738039694726467, "time": 0.5855432271957397, "epoch": 7, "memory": 6319, "step": 26742} +{"lr": 3.0058414515938063e-05, "data_time": 0.48802919387817384, "loss": 0.08832942582666874, "time": 0.6322428226470947, "epoch": 7, "memory": 6319, "step": 26842} +{"lr": 2.975804659365501e-05, "data_time": 0.4412156343460083, "loss": 0.0902914434671402, "time": 0.5849912881851196, "epoch": 7, "memory": 6319, "step": 26942} +{"lr": 2.945931072168714e-05, "data_time": 0.48147096633911135, "loss": 0.07930332496762275, "time": 0.6256494283676147, "epoch": 7, "memory": 6319, "step": 27042} +{"lr": 2.9162226215178855e-05, "data_time": 0.4175581693649292, "loss": 0.08182027153670787, "time": 0.5587935209274292, "epoch": 7, "memory": 6319, "step": 27142} +{"lr": 2.8866812282503322e-05, "data_time": 0.4660873651504517, "loss": 0.08387458957731724, "time": 0.6087932825088501, "epoch": 7, "memory": 6319, "step": 27242} +{"lr": 2.8573088024020838e-05, "data_time": 0.4384413242340088, "loss": 0.07118411213159562, "time": 0.5906196594238281, "epoch": 7, "memory": 6319, "step": 27342} +{"accuracy/top1": 96.3118896484375, "data_time": 0.5100329908831366, "time": 0.6019841309251457, "step": 7} +{"lr": 2.8260695817390092e-05, "data_time": 0.4504760980606079, "loss": 0.08787877596914768, "time": 0.5920387983322144, "epoch": 8, "memory": 6319, "step": 27449} +{"lr": 2.7970529402570967e-05, "data_time": 0.4833108186721802, "loss": 0.06415975615382194, "time": 0.6258732080459595, "epoch": 8, "memory": 6319, "step": 27549} +{"lr": 2.7682110612244214e-05, "data_time": 0.45999147891998293, "loss": 0.08622431494295597, "time": 0.6035938501358032, "epoch": 8, "memory": 6319, "step": 27649} +{"lr": 2.7395458094490283e-05, "data_time": 0.4709547281265259, "loss": 0.07511433474719524, "time": 0.6246785402297974, "epoch": 8, "memory": 6319, "step": 27749} +{"lr": 2.7110590383189015e-05, "data_time": 0.4341695547103882, "loss": 0.06846614442765712, "time": 0.5759546518325805, "epoch": 8, "memory": 6319, "step": 27849} +{"lr": 2.6827525896821397e-05, "data_time": 0.4143645763397217, "loss": 0.0698573287576437, "time": 0.5643430471420288, "epoch": 8, "memory": 6319, "step": 27949} +{"lr": 2.6546282937278552e-05, "data_time": 0.4419599771499634, "loss": 0.07415164783596992, "time": 0.5827390670776367, "epoch": 8, "memory": 6319, "step": 28049} +{"lr": 2.6266879688678606e-05, "data_time": 0.5172754049301147, "loss": 0.10456358268857002, "time": 0.6736511707305908, "epoch": 8, "memory": 6319, "step": 28149} +{"lr": 2.5989334216190738e-05, "data_time": 0.4718210458755493, "loss": 0.0814783986657858, "time": 0.6133327007293701, "epoch": 8, "memory": 6319, "step": 28249} +{"lr": 2.571366446486732e-05, "data_time": 0.44338967800140383, "loss": 0.08454127684235573, "time": 0.6003024578094482, "epoch": 8, "memory": 6319, "step": 28349} +{"lr": 2.5439888258483665e-05, "data_time": 0.5230553865432739, "loss": 0.07729196920990944, "time": 0.6659160852432251, "epoch": 8, "memory": 6319, "step": 28449} +{"lr": 2.5168023298385384e-05, "data_time": 0.4534576892852783, "loss": 0.07279450930655003, "time": 0.5979245901107788, "epoch": 8, "memory": 6319, "step": 28549} +{"lr": 2.489808716234412e-05, "data_time": 0.4939014434814453, "loss": 0.0854671698063612, "time": 0.6376083612442016, "epoch": 8, "memory": 6319, "step": 28649} +{"lr": 2.463009730342111e-05, "data_time": 0.4644150733947754, "loss": 0.07376628667116165, "time": 0.6085563898086548, "epoch": 8, "memory": 6319, "step": 28749} +{"lr": 2.436407104883849e-05, "data_time": 0.48210346698760986, "loss": 0.07696508876979351, "time": 0.6235823392868042, "epoch": 8, "memory": 6319, "step": 28849} +{"lr": 2.4100025598859092e-05, "data_time": 0.49641640186309816, "loss": 0.0698467107489705, "time": 0.6405101776123047, "epoch": 8, "memory": 6319, "step": 28949} +{"lr": 2.3837978025674284e-05, "data_time": 0.4483215570449829, "loss": 0.0646116517484188, "time": 0.5912272214889527, "epoch": 8, "memory": 6319, "step": 29049} +{"lr": 2.357794527230027e-05, "data_time": 0.4442706823348999, "loss": 0.07210826762020588, "time": 0.5862396717071533, "epoch": 8, "memory": 6319, "step": 29149} +{"lr": 2.33199441514825e-05, "data_time": 0.46998779773712157, "loss": 0.07695260904729366, "time": 0.6124399900436401, "epoch": 8, "memory": 6319, "step": 29249} +{"lr": 2.306399134460859e-05, "data_time": 0.5597031116485596, "loss": 0.0864983320236206, "time": 0.7024467706680297, "epoch": 8, "memory": 6319, "step": 29349} +{"lr": 2.281010340062995e-05, "data_time": 0.45230419635772706, "loss": 0.0875934824347496, "time": 0.5931612968444824, "epoch": 8, "memory": 6319, "step": 29449} +{"lr": 2.2558296734991603e-05, "data_time": 0.4420763969421387, "loss": 0.08598154541105033, "time": 0.5839328050613404, "epoch": 8, "memory": 6319, "step": 29549} +{"lr": 2.2308587628570915e-05, "data_time": 0.45420012474060056, "loss": 0.07689887657761574, "time": 0.6094831943511962, "epoch": 8, "memory": 6319, "step": 29649} +{"lr": 2.2060992226624955e-05, "data_time": 0.43480234146118163, "loss": 0.07518543973565102, "time": 0.5770165681838989, "epoch": 8, "memory": 6319, "step": 29749} +{"lr": 2.181552653774648e-05, "data_time": 0.41755697727203367, "loss": 0.07007209099829197, "time": 0.5601857662200928, "epoch": 8, "memory": 6319, "step": 29849} +{"lr": 2.1572206432829035e-05, "data_time": 0.44503648281097413, "loss": 0.07302001379430294, "time": 0.5869835376739502, "epoch": 8, "memory": 6319, "step": 29949} +{"lr": 2.1331047644040754e-05, "data_time": 0.4657976388931274, "loss": 0.08360778503119945, "time": 0.6086403846740722, "epoch": 8, "memory": 6319, "step": 30049} +{"lr": 2.109206576380705e-05, "data_time": 0.42251613140106203, "loss": 0.08748757466673851, "time": 0.5644537448883057, "epoch": 8, "memory": 6319, "step": 30149} +{"lr": 2.0855276243802687e-05, "data_time": 0.4733867645263672, "loss": 0.07566434890031815, "time": 0.6182521343231201, "epoch": 8, "memory": 6319, "step": 30249} +{"lr": 2.0620694393952487e-05, "data_time": 0.5105309247970581, "loss": 0.08560679405927658, "time": 0.6539804697036743, "epoch": 8, "memory": 6319, "step": 30349} +{"lr": 2.038833538144169e-05, "data_time": 0.544872498512268, "loss": 0.0843689925968647, "time": 0.6889699935913086, "epoch": 8, "memory": 6319, "step": 30449} +{"lr": 2.0158214229735154e-05, "data_time": 0.4630296230316162, "loss": 0.07467749454081059, "time": 0.6060987949371338, "epoch": 8, "memory": 6319, "step": 30549} +{"lr": 1.9930345817605977e-05, "data_time": 0.5143996000289917, "loss": 0.08833044432103634, "time": 0.6577551364898682, "epoch": 8, "memory": 6319, "step": 30649} +{"lr": 1.970474487817358e-05, "data_time": 0.4972402572631836, "loss": 0.07718662656843663, "time": 0.6398980617523193, "epoch": 8, "memory": 6319, "step": 30749} +{"lr": 1.9481425997951092e-05, "data_time": 0.45746264457702634, "loss": 0.07747239619493484, "time": 0.6035328388214112, "epoch": 8, "memory": 6319, "step": 30849} +{"lr": 1.9260403615902187e-05, "data_time": 0.5013140916824341, "loss": 0.07578352279961109, "time": 0.6433354377746582, "epoch": 8, "memory": 6319, "step": 30949} +{"lr": 1.9041692022507647e-05, "data_time": 0.5530886173248291, "loss": 0.072789004445076, "time": 0.7126932859420776, "epoch": 8, "memory": 6319, "step": 31049} +{"lr": 1.8825305358841125e-05, "data_time": 0.46707730293273925, "loss": 0.08952586576342583, "time": 0.6104723930358886, "epoch": 8, "memory": 6319, "step": 31149} +{"lr": 1.861125761565518e-05, "data_time": 0.39845116138458253, "loss": 0.07494150102138519, "time": 0.5425600528717041, "epoch": 8, "memory": 6319, "step": 31249} +{"accuracy/top1": 96.72245788574219, "data_time": 0.4933110845500025, "time": 0.5810148099373127, "step": 8} +{"lr": 1.8384832449122864e-05, "data_time": 0.4758762359619141, "loss": 0.0754853405058384, "time": 0.6189368009567261, "epoch": 9, "memory": 6319, "step": 31356} +{"lr": 1.817567007332114e-05, "data_time": 0.45037035942077636, "loss": 0.08184201121330262, "time": 0.606569504737854, "epoch": 9, "memory": 6319, "step": 31456} +{"lr": 1.796888862098825e-05, "data_time": 0.4804424285888672, "loss": 0.06698336489498616, "time": 0.623430871963501, "epoch": 9, "memory": 6319, "step": 31556} +{"lr": 1.7764501461839678e-05, "data_time": 0.47014167308807375, "loss": 0.07689600102603436, "time": 0.6136189460754394, "epoch": 9, "memory": 6319, "step": 31656} +{"lr": 1.7562521810784746e-05, "data_time": 0.45176260471343993, "loss": 0.08106597140431404, "time": 0.5967425346374512, "epoch": 9, "memory": 6319, "step": 31756} +{"lr": 1.7362962727072444e-05, "data_time": 0.4487202882766724, "loss": 0.07057645507156848, "time": 0.5911329746246338, "epoch": 9, "memory": 6319, "step": 31856} +{"lr": 1.716583711344682e-05, "data_time": 0.48130006790161134, "loss": 0.07462608590722083, "time": 0.6231300115585328, "epoch": 9, "memory": 6319, "step": 31956} +{"lr": 1.6971157715312954e-05, "data_time": 0.46283550262451173, "loss": 0.06999175287783146, "time": 0.6046403408050537, "epoch": 9, "memory": 6319, "step": 32056} +{"lr": 1.677893711991274e-05, "data_time": 0.4871793508529663, "loss": 0.07801001891493797, "time": 0.6288370609283447, "epoch": 9, "memory": 6319, "step": 32156} +{"lr": 1.6589187755511025e-05, "data_time": 0.4406203985214233, "loss": 0.06530537940561772, "time": 0.5981998920440674, "epoch": 9, "memory": 6319, "step": 32256} +{"lr": 1.640192189059216e-05, "data_time": 0.41819570064544676, "loss": 0.0774762861430645, "time": 0.5595851421356202, "epoch": 9, "memory": 6319, "step": 32356} +{"lr": 1.6217151633066716e-05, "data_time": 0.4525012731552124, "loss": 0.08113399259746075, "time": 0.6044313192367554, "epoch": 9, "memory": 6319, "step": 32456} +{"lr": 1.6034888929488564e-05, "data_time": 0.4695448398590088, "loss": 0.06804250255227089, "time": 0.6150140285491943, "epoch": 9, "memory": 6319, "step": 32556} +{"lr": 1.585514556428258e-05, "data_time": 0.48380489349365235, "loss": 0.065376490727067, "time": 0.6378083467483521, "epoch": 9, "memory": 6319, "step": 32656} +{"lr": 1.567793315898261e-05, "data_time": 0.4007999897003174, "loss": 0.07852859795093536, "time": 0.5449673891067505, "epoch": 9, "memory": 6319, "step": 32756} +{"lr": 1.5503263171480048e-05, "data_time": 0.47496657371520995, "loss": 0.07522008717060089, "time": 0.6424149751663208, "epoch": 9, "memory": 6319, "step": 32856} +{"lr": 1.5331146895283165e-05, "data_time": 0.43189353942871095, "loss": 0.07639147564768792, "time": 0.5732213973999023, "epoch": 9, "memory": 6319, "step": 32956} +{"lr": 1.5161595458786728e-05, "data_time": 0.4367440938949585, "loss": 0.08115535750985145, "time": 0.595166802406311, "epoch": 9, "memory": 6319, "step": 33056} +{"lr": 1.4994619824552563e-05, "data_time": 0.4457152605056763, "loss": 0.07400755509734154, "time": 0.5902722597122192, "epoch": 9, "memory": 6319, "step": 33156} +{"lr": 1.4830230788600802e-05, "data_time": 0.46508469581604006, "loss": 0.0755990494042635, "time": 0.623738956451416, "epoch": 9, "memory": 6319, "step": 33256} +{"lr": 1.4668438979711779e-05, "data_time": 0.4103731870651245, "loss": 0.07834706865251065, "time": 0.554237675666809, "epoch": 9, "memory": 6319, "step": 33356} +{"lr": 1.4509254858738844e-05, "data_time": 0.45750463008880615, "loss": 0.06773686297237873, "time": 0.613737964630127, "epoch": 9, "memory": 6319, "step": 33456} +{"lr": 1.4352688717931966e-05, "data_time": 0.4298782587051392, "loss": 0.07748924121260643, "time": 0.5725626707077026, "epoch": 9, "memory": 6319, "step": 33556} +{"lr": 1.4198750680272348e-05, "data_time": 0.4645204782485962, "loss": 0.0789119977504015, "time": 0.6064340353012085, "epoch": 9, "memory": 6319, "step": 33656} +{"lr": 1.4047450698817857e-05, "data_time": 0.4595818519592285, "loss": 0.08945306725800037, "time": 0.6004274606704711, "epoch": 9, "memory": 6319, "step": 33756} +{"lr": 1.389879855605952e-05, "data_time": 0.4054666519165039, "loss": 0.07022984363138676, "time": 0.5477522850036621, "epoch": 9, "memory": 6319, "step": 33856} +{"lr": 1.3752803863288976e-05, "data_time": 0.456856369972229, "loss": 0.06537099406123162, "time": 0.5985014200210571, "epoch": 9, "memory": 6319, "step": 33956} +{"lr": 1.3609476059977129e-05, "data_time": 0.5411594390869141, "loss": 0.07017961330711842, "time": 0.6840475559234619, "epoch": 9, "memory": 6319, "step": 34056} +{"lr": 1.3468824413163754e-05, "data_time": 0.42032756805419924, "loss": 0.06352031752467155, "time": 0.5630666494369507, "epoch": 9, "memory": 6319, "step": 34156} +{"lr": 1.3330858016858395e-05, "data_time": 0.4271347999572754, "loss": 0.0937989518046379, "time": 0.5699228286743164, "epoch": 9, "memory": 6319, "step": 34256} +{"lr": 1.3195585791452309e-05, "data_time": 0.4812811851501465, "loss": 0.07161642089486123, "time": 0.6232021808624267, "epoch": 9, "memory": 6319, "step": 34356} +{"lr": 1.3063016483141729e-05, "data_time": 0.45418920516967776, "loss": 0.06601140499114991, "time": 0.5956439256668091, "epoch": 9, "memory": 6319, "step": 34456} +{"lr": 1.2933158663362388e-05, "data_time": 0.42545571327209475, "loss": 0.08046995252370834, "time": 0.5846264362335205, "epoch": 9, "memory": 6319, "step": 34556} +{"lr": 1.2806020728235311e-05, "data_time": 0.4451966047286987, "loss": 0.07233460396528243, "time": 0.5870979309082032, "epoch": 9, "memory": 6319, "step": 34656} +{"lr": 1.2681610898023963e-05, "data_time": 0.4433310508728027, "loss": 0.07668758258223533, "time": 0.585440993309021, "epoch": 9, "memory": 6319, "step": 34756} +{"lr": 1.2559937216602731e-05, "data_time": 0.47575376033782957, "loss": 0.08021556288003921, "time": 0.6187443256378173, "epoch": 9, "memory": 6319, "step": 34856} +{"lr": 1.2441007550936847e-05, "data_time": 0.5138709545135498, "loss": 0.07820392809808255, "time": 0.6556874990463257, "epoch": 9, "memory": 6319, "step": 34956} +{"lr": 1.232482959057377e-05, "data_time": 0.45088682174682615, "loss": 0.0829607080668211, "time": 0.5938930034637451, "epoch": 9, "memory": 6319, "step": 35056} +{"lr": 1.2211410847145969e-05, "data_time": 0.479123592376709, "loss": 0.07704100422561169, "time": 0.6211544513702393, "epoch": 9, "memory": 6319, "step": 35156} +{"accuracy/top1": 96.98340606689453, "data_time": 0.50804053092825, "time": 0.5981968764601082, "step": 9} +{"lr": 1.209311679360931e-05, "data_time": 0.41988067626953124, "loss": 0.07475814446806908, "time": 0.5620572805404663, "epoch": 10, "memory": 6319, "step": 35263} +{"lr": 1.1985432727488666e-05, "data_time": 0.42589197158813474, "loss": 0.08367883190512657, "time": 0.569089674949646, "epoch": 10, "memory": 6319, "step": 35363} +{"lr": 1.1880529822435289e-05, "data_time": 0.48822770118713377, "loss": 0.08758219256997109, "time": 0.6306748628616333, "epoch": 10, "memory": 6319, "step": 35463} +{"lr": 1.1778414861078726e-05, "data_time": 0.44216940402984617, "loss": 0.07201689779758454, "time": 0.5951197385787964, "epoch": 10, "memory": 6319, "step": 35563} +{"lr": 1.1679094445790538e-05, "data_time": 0.41455626487731934, "loss": 0.07749701552093029, "time": 0.5566478490829467, "epoch": 10, "memory": 6319, "step": 35663} +{"lr": 1.1582574998257377e-05, "data_time": 0.45920236110687257, "loss": 0.06470660604536534, "time": 0.6031032562255859, "epoch": 10, "memory": 6319, "step": 35763} +{"lr": 1.1488862759065762e-05, "data_time": 0.40541141033172606, "loss": 0.07805955074727536, "time": 0.5468392610549927, "epoch": 10, "memory": 6319, "step": 35863} +{"lr": 1.1397963787298665e-05, "data_time": 0.4425260066986084, "loss": 0.08282766602933407, "time": 0.6018846511840821, "epoch": 10, "memory": 6319, "step": 35963} +{"lr": 1.1309883960143664e-05, "data_time": 0.4725736379623413, "loss": 0.08345601595938205, "time": 0.6152339220046997, "epoch": 10, "memory": 6319, "step": 36063} +{"lr": 1.122462897251302e-05, "data_time": 0.4819929599761963, "loss": 0.06973667442798615, "time": 0.6446127891540527, "epoch": 10, "memory": 6319, "step": 36163} +{"lr": 1.114220433667542e-05, "data_time": 0.43996443748474123, "loss": 0.08997197225689887, "time": 0.5972410678863526, "epoch": 10, "memory": 6319, "step": 36263} +{"lr": 1.10626153818996e-05, "data_time": 0.4460367441177368, "loss": 0.06445202976465225, "time": 0.5879318714141846, "epoch": 10, "memory": 6319, "step": 36363} +{"lr": 1.0985867254109761e-05, "data_time": 0.48964581489562986, "loss": 0.0746229462325573, "time": 0.6339953899383545, "epoch": 10, "memory": 6319, "step": 36463} +{"lr": 1.0911964915552823e-05, "data_time": 0.3942296266555786, "loss": 0.08499046415090561, "time": 0.5552056550979614, "epoch": 10, "memory": 6319, "step": 36563} +{"lr": 1.0840913144477643e-05, "data_time": 0.4788695812225342, "loss": 0.07111723981797695, "time": 0.6212372064590455, "epoch": 10, "memory": 6319, "step": 36663} +{"lr": 1.0772716534826064e-05, "data_time": 0.4616294622421265, "loss": 0.07191082239151, "time": 0.6043037414550781, "epoch": 10, "memory": 6319, "step": 36763} +{"lr": 1.0707379495935828e-05, "data_time": 0.4547978639602661, "loss": 0.07027343362569809, "time": 0.6120363950729371, "epoch": 10, "memory": 6319, "step": 36863} +{"lr": 1.0644906252255541e-05, "data_time": 0.39924206733703616, "loss": 0.0715204045176506, "time": 0.5521424293518067, "epoch": 10, "memory": 6319, "step": 36963} +{"lr": 1.0585300843071527e-05, "data_time": 0.45977783203125, "loss": 0.06767462603747845, "time": 0.6041017055511475, "epoch": 10, "memory": 6319, "step": 37063} +{"lr": 1.0528567122246654e-05, "data_time": 0.44636244773864747, "loss": 0.07604262344539166, "time": 0.5876028537750244, "epoch": 10, "memory": 6319, "step": 37163} +{"lr": 1.0474708757971141e-05, "data_time": 0.4786412715911865, "loss": 0.06674356833100319, "time": 0.6208931922912597, "epoch": 10, "memory": 6319, "step": 37263} +{"lr": 1.042372923252543e-05, "data_time": 0.4120201826095581, "loss": 0.07286047264933586, "time": 0.5540947198867798, "epoch": 10, "memory": 6319, "step": 37363} +{"lr": 1.0375631842054998e-05, "data_time": 0.46259894371032717, "loss": 0.07435069568455219, "time": 0.606714916229248, "epoch": 10, "memory": 6319, "step": 37463} +{"lr": 1.0330419696357254e-05, "data_time": 0.4271500825881958, "loss": 0.06328838840126991, "time": 0.5708273649215698, "epoch": 10, "memory": 6319, "step": 37563} +{"lr": 1.028809571868047e-05, "data_time": 0.4651278734207153, "loss": 0.06647076942026615, "time": 0.6080742359161377, "epoch": 10, "memory": 6319, "step": 37663} +{"lr": 1.0248662645534766e-05, "data_time": 0.41365871429443357, "loss": 0.07770975790917874, "time": 0.5578250169754029, "epoch": 10, "memory": 6319, "step": 37763} +{"lr": 1.0212123026515194e-05, "data_time": 0.4224325895309448, "loss": 0.0664729431271553, "time": 0.5639659404754639, "epoch": 10, "memory": 6319, "step": 37863} +{"lr": 1.0178479224136881e-05, "data_time": 0.416499400138855, "loss": 0.07827116660773754, "time": 0.5590309381484986, "epoch": 10, "memory": 6319, "step": 37963} +{"lr": 1.0147733413682289e-05, "data_time": 0.5093073368072509, "loss": 0.06647942066192628, "time": 0.6525231122970581, "epoch": 10, "memory": 6319, "step": 38063} +{"lr": 1.0119887583060527e-05, "data_time": 0.435845947265625, "loss": 0.06967916637659073, "time": 0.5783462762832642, "epoch": 10, "memory": 6319, "step": 38163} +{"lr": 1.0094943532678871e-05, "data_time": 0.4608129024505615, "loss": 0.08074088096618652, "time": 0.60422682762146, "epoch": 10, "memory": 6319, "step": 38263} +{"lr": 1.007290287532637e-05, "data_time": 0.41082403659820554, "loss": 0.07776946760714054, "time": 0.5526329278945923, "epoch": 10, "memory": 6319, "step": 38363} +{"lr": 1.0053767036069495e-05, "data_time": 0.43117775917053225, "loss": 0.059059906750917435, "time": 0.5733058452606201, "epoch": 10, "memory": 6319, "step": 38463} +{"lr": 1.003753725216005e-05, "data_time": 0.4748003721237183, "loss": 0.07253213152289391, "time": 0.6162424802780151, "epoch": 10, "memory": 6319, "step": 38563} +{"lr": 1.0024214572955152e-05, "data_time": 0.4608904838562012, "loss": 0.06550565175712109, "time": 0.6028083801269531, "epoch": 10, "memory": 6319, "step": 38663} +{"lr": 1.0013799859849456e-05, "data_time": 0.4114959239959717, "loss": 0.09634218141436576, "time": 0.5531365394592285, "epoch": 10, "memory": 6319, "step": 38763} +{"lr": 1.0006293786219338e-05, "data_time": 0.46281936168670657, "loss": 0.07998348549008369, "time": 0.6058304071426391, "epoch": 10, "memory": 6319, "step": 38863} +{"lr": 1.0001696837379472e-05, "data_time": 0.519748306274414, "loss": 0.07544913105666637, "time": 0.6611512660980224, "epoch": 10, "memory": 6319, "step": 38963} +{"lr": 1.0000009310551361e-05, "data_time": 0.4274256944656372, "loss": 0.07560225427150727, "time": 0.5697463035583497, "epoch": 10, "memory": 6319, "step": 39063} +{"accuracy/top1": 97.04603576660156, "data_time": 0.4988717580663747, "time": 0.585812913960424, "step": 10} diff --git a/resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1/20230604_225317/vis_data/vis_image/0000000.png_1.png b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1/20230604_225317/vis_data/vis_image/0000000.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..a9b7781aa1320e4fb5e0dab670bcbf30e14c6376 Binary files /dev/null and b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1/20230604_225317/vis_data/vis_image/0000000.png_1.png differ diff --git a/resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1/20230604_225317/vis_data/vis_image/0000000.png_10.png b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1/20230604_225317/vis_data/vis_image/0000000.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..a8557ecb920f58442b3166e48a118c4d91b8b1fc Binary files /dev/null and b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1/20230604_225317/vis_data/vis_image/0000000.png_10.png differ diff --git a/resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1/20230604_225317/vis_data/vis_image/0000000.png_2.png b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1/20230604_225317/vis_data/vis_image/0000000.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..0cbc0bf68778b193047657b7bf3b61e22a1b6a3a Binary files /dev/null and b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1/20230604_225317/vis_data/vis_image/0000000.png_2.png differ diff --git a/resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1/20230604_225317/vis_data/vis_image/0000000.png_3.png b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1/20230604_225317/vis_data/vis_image/0000000.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..1edebc8bce44e9138680f33f8e9d67641d822c6b Binary files /dev/null and b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1/20230604_225317/vis_data/vis_image/0000000.png_3.png differ diff --git a/resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1/20230604_225317/vis_data/vis_image/0000000.png_4.png b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1/20230604_225317/vis_data/vis_image/0000000.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..bedf853cac3acaf8df2db8f81a683b9a1becb07e Binary files /dev/null and b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1/20230604_225317/vis_data/vis_image/0000000.png_4.png differ diff --git a/resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1/20230604_225317/vis_data/vis_image/0000000.png_5.png b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1/20230604_225317/vis_data/vis_image/0000000.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..5bb2070185c6b14950d1bc5572b8dfdf6dd3240a Binary files /dev/null and b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1/20230604_225317/vis_data/vis_image/0000000.png_5.png differ diff --git a/resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1/20230604_225317/vis_data/vis_image/0000000.png_6.png b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1/20230604_225317/vis_data/vis_image/0000000.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..4e2d701ed604e99cb6b615b1deb567f281df772c Binary files /dev/null and b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1/20230604_225317/vis_data/vis_image/0000000.png_6.png differ diff --git a/resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1/20230604_225317/vis_data/vis_image/0000000.png_7.png b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1/20230604_225317/vis_data/vis_image/0000000.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..43576a946efa40aa41efef443a2594709422155d Binary files /dev/null and b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1/20230604_225317/vis_data/vis_image/0000000.png_7.png differ diff --git a/resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1/20230604_225317/vis_data/vis_image/0000000.png_8.png b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1/20230604_225317/vis_data/vis_image/0000000.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..4019a9da8d4b1fdf84e349cc8e96f6d496cfc315 Binary files /dev/null and b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1/20230604_225317/vis_data/vis_image/0000000.png_8.png differ diff --git a/resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1/20230604_225317/vis_data/vis_image/0000000.png_9.png b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1/20230604_225317/vis_data/vis_image/0000000.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..93f928d3ed43f51f64af8d2190cc8b97c72b0c27 Binary files /dev/null and b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1/20230604_225317/vis_data/vis_image/0000000.png_9.png differ diff --git a/resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1/20230604_225317/vis_data/vis_image/000005174.jpg_1.png b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1/20230604_225317/vis_data/vis_image/000005174.jpg_1.png new file mode 100644 index 0000000000000000000000000000000000000000..c51385d5b6cfde6c6b0737cdb54edd3533ccc833 Binary files /dev/null and b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1/20230604_225317/vis_data/vis_image/000005174.jpg_1.png differ diff --git a/resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1/20230604_225317/vis_data/vis_image/000005174.jpg_10.png b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1/20230604_225317/vis_data/vis_image/000005174.jpg_10.png new file mode 100644 index 0000000000000000000000000000000000000000..17351605b5db7e06c81f8afdb62e07a5deafbc9f Binary files /dev/null and b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1/20230604_225317/vis_data/vis_image/000005174.jpg_10.png differ diff --git a/resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1/20230604_225317/vis_data/vis_image/000005174.jpg_2.png b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1/20230604_225317/vis_data/vis_image/000005174.jpg_2.png new file mode 100644 index 0000000000000000000000000000000000000000..bfa43a38610e2793a68debd67ee22c794fe8d09b Binary files /dev/null and b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1/20230604_225317/vis_data/vis_image/000005174.jpg_2.png differ diff --git a/resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1/20230604_225317/vis_data/vis_image/000005174.jpg_3.png b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1/20230604_225317/vis_data/vis_image/000005174.jpg_3.png new file mode 100644 index 0000000000000000000000000000000000000000..dbd77f01e260f86b8057ab625a57e752e06e28d8 Binary files /dev/null and b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1/20230604_225317/vis_data/vis_image/000005174.jpg_3.png differ diff --git a/resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1/20230604_225317/vis_data/vis_image/000005174.jpg_4.png b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1/20230604_225317/vis_data/vis_image/000005174.jpg_4.png new file mode 100644 index 0000000000000000000000000000000000000000..93e220ffa63a1087129dd447bfa2d04b6ff4e3bd Binary files /dev/null and b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1/20230604_225317/vis_data/vis_image/000005174.jpg_4.png differ diff --git a/resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1/20230604_225317/vis_data/vis_image/000005174.jpg_5.png b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1/20230604_225317/vis_data/vis_image/000005174.jpg_5.png new file mode 100644 index 0000000000000000000000000000000000000000..bedca059b1905d2307957486769cdab78bc07b0f Binary files /dev/null and b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1/20230604_225317/vis_data/vis_image/000005174.jpg_5.png differ diff --git a/resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1/20230604_225317/vis_data/vis_image/000005174.jpg_6.png b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1/20230604_225317/vis_data/vis_image/000005174.jpg_6.png new file mode 100644 index 0000000000000000000000000000000000000000..287dd94b672ac11d233e0723732963d8efc00fc1 Binary files /dev/null and b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1/20230604_225317/vis_data/vis_image/000005174.jpg_6.png differ diff --git a/resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1/20230604_225317/vis_data/vis_image/000005174.jpg_7.png b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1/20230604_225317/vis_data/vis_image/000005174.jpg_7.png new file mode 100644 index 0000000000000000000000000000000000000000..f14c4e97d9eefbc2e6628d8153e21f6288e6ffc4 Binary files /dev/null and b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1/20230604_225317/vis_data/vis_image/000005174.jpg_7.png differ diff --git a/resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1/20230604_225317/vis_data/vis_image/000005174.jpg_8.png b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1/20230604_225317/vis_data/vis_image/000005174.jpg_8.png new file mode 100644 index 0000000000000000000000000000000000000000..3dde7dc78bf1363be48bc5fa96fbfceaab42ef1e Binary files /dev/null and b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1/20230604_225317/vis_data/vis_image/000005174.jpg_8.png differ diff --git a/resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1/20230604_225317/vis_data/vis_image/000005174.jpg_9.png b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1/20230604_225317/vis_data/vis_image/000005174.jpg_9.png new file mode 100644 index 0000000000000000000000000000000000000000..f9fcbaa8bc1990cb98c6ad0439bb5f64ce37ac00 Binary files /dev/null and b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1/20230604_225317/vis_data/vis_image/000005174.jpg_9.png differ diff --git a/resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1/20230604_225317/vis_data/vis_image/0010000.png_1.png b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1/20230604_225317/vis_data/vis_image/0010000.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..898aad4fde4129544ce09761826b4384508bee4c Binary files /dev/null and b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1/20230604_225317/vis_data/vis_image/0010000.png_1.png differ diff --git a/resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1/20230604_225317/vis_data/vis_image/0010000.png_10.png b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1/20230604_225317/vis_data/vis_image/0010000.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..40da1cb3b80d86cb73ff728a1e42a610b851be83 Binary files /dev/null and b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1/20230604_225317/vis_data/vis_image/0010000.png_10.png differ diff --git a/resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1/20230604_225317/vis_data/vis_image/0010000.png_2.png b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1/20230604_225317/vis_data/vis_image/0010000.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..c92f26ea6de6d44942e425079a72a9ac9bb1825f Binary files /dev/null and b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1/20230604_225317/vis_data/vis_image/0010000.png_2.png differ diff --git a/resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1/20230604_225317/vis_data/vis_image/0010000.png_3.png b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1/20230604_225317/vis_data/vis_image/0010000.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..a2d562cb7a2564bf7164bb4618cf1f678fe2f39a Binary files /dev/null and b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1/20230604_225317/vis_data/vis_image/0010000.png_3.png differ diff --git a/resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1/20230604_225317/vis_data/vis_image/0010000.png_4.png b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1/20230604_225317/vis_data/vis_image/0010000.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..c72a00be6dd98ff25478efaf0795a287a4fdfbac Binary files /dev/null and b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1/20230604_225317/vis_data/vis_image/0010000.png_4.png differ diff --git a/resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1/20230604_225317/vis_data/vis_image/0010000.png_5.png b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1/20230604_225317/vis_data/vis_image/0010000.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..fe188771fcd40f1af543a59dc68c635abbbf57a2 Binary files /dev/null and b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1/20230604_225317/vis_data/vis_image/0010000.png_5.png differ diff --git a/resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1/20230604_225317/vis_data/vis_image/0010000.png_6.png b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1/20230604_225317/vis_data/vis_image/0010000.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..91894c46be01890eb5d58615d55ed0362a7a7aa7 Binary files /dev/null and b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1/20230604_225317/vis_data/vis_image/0010000.png_6.png differ diff --git a/resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1/20230604_225317/vis_data/vis_image/0010000.png_7.png b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1/20230604_225317/vis_data/vis_image/0010000.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..3fd2a0b9c4fedd74e1a6e8b8c5e294d085e59965 Binary files /dev/null and b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1/20230604_225317/vis_data/vis_image/0010000.png_7.png differ diff --git a/resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1/20230604_225317/vis_data/vis_image/0010000.png_8.png b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1/20230604_225317/vis_data/vis_image/0010000.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..5f961f31964767ea5025a8acbb1f990f4a453852 Binary files /dev/null and b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1/20230604_225317/vis_data/vis_image/0010000.png_8.png differ diff --git a/resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1/20230604_225317/vis_data/vis_image/0010000.png_9.png b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1/20230604_225317/vis_data/vis_image/0010000.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..bd45c3214131bb4cfe9e35b6c30fb5090af597d3 Binary files /dev/null and b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1/20230604_225317/vis_data/vis_image/0010000.png_9.png differ diff --git a/resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1/epoch_1.pth b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1/epoch_1.pth new file mode 100644 index 0000000000000000000000000000000000000000..b70462e9c6f8739a3cabe32b1683861c2de2c4c9 --- /dev/null +++ b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1/epoch_1.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:fd6539f87d9841c4a7c1c76f20d60f94c3a3c85ae65706f967eba7224d14df67 +size 277329549 diff --git a/resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1/epoch_10.pth b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1/epoch_10.pth new file mode 100644 index 0000000000000000000000000000000000000000..6b2b8174575f563346bcdb0d1c684b58fb7b31a9 --- /dev/null +++ b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1/epoch_10.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:de8f25e306c3249615e33553e9d5570302f85635dd9972b8659b31ff740da8f4 +size 279888653 diff --git a/resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1/epoch_2.pth b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1/epoch_2.pth new file mode 100644 index 0000000000000000000000000000000000000000..58ef39ba6c85f6dd4009ddf63c3c5e4a26164cae --- /dev/null +++ b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1/epoch_2.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:a04387006b7dad78d9dbee70be6eac74a16a5386729f0c7309ce834dce39daa4 +size 277612941 diff --git a/resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1/epoch_3.pth b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1/epoch_3.pth new file mode 100644 index 0000000000000000000000000000000000000000..96a9475f93db6689eba960de3c516b1eb186d0ba --- /dev/null +++ b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1/epoch_3.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:f8b88a565eda2e0394164c5b7997ec1381292541b5ea0bb6c5af165164e18714 +size 277896077 diff --git a/resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1/epoch_4.pth b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1/epoch_4.pth new file mode 100644 index 0000000000000000000000000000000000000000..3aaa62513907834c0bd92f7e8c84623f20622fcf --- /dev/null +++ b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1/epoch_4.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:771b8f8391803c3f9f0994038e9bba0e359025bdaf83ab3d5aba02cbf340fd01 +size 278178637 diff --git a/resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1/epoch_5.pth b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1/epoch_5.pth new file mode 100644 index 0000000000000000000000000000000000000000..7fa059fddfbc722e90a9ad97959d4dbabd7891de --- /dev/null +++ b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1/epoch_5.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:a1f12b033e3ac9c98dd2ed8e54740796254b484c730dde6899423f1e01e0b56b +size 278461197 diff --git a/resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1/epoch_6.pth b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1/epoch_6.pth new file mode 100644 index 0000000000000000000000000000000000000000..34e123aa951049624e11a6583af030a103c222c4 --- /dev/null +++ b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1/epoch_6.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:5421a136015aaa29bd4ae4ba936c76c1edf3498b2a4ad364432ae173ecd7e662 +size 278743885 diff --git a/resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1/epoch_7.pth b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1/epoch_7.pth new file mode 100644 index 0000000000000000000000000000000000000000..d5c77a0bdcf1dd13288bcb9eb724773f9b662804 --- /dev/null +++ b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1/epoch_7.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:e46832bb352d2901d8c06060107ca57f76528b9a195f63b34f6f4564bfc6f7bc +size 279027597 diff --git a/resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1/epoch_8.pth b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1/epoch_8.pth new file mode 100644 index 0000000000000000000000000000000000000000..646cc0d81e337f54457bb5b8784bfaf3a05c30a6 --- /dev/null +++ b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1/epoch_8.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:bf34af7dd6521014019eb8d5270602f8bb3c67dca765a179b749a0b0f794c0ad +size 279314765 diff --git a/resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1/epoch_9.pth b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1/epoch_9.pth new file mode 100644 index 0000000000000000000000000000000000000000..ace56ff8d3415aea68c5d038bb8aa40c677b5646 --- /dev/null +++ b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1/epoch_9.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:cfba25ecfb4b7043f8c00c37e90ffb717cf2ba757261595a8f877eb4e9a8cc46 +size 279601869 diff --git a/resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1/last_checkpoint b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1/last_checkpoint new file mode 100644 index 0000000000000000000000000000000000000000..a8af92fc8a3e46da7bad802e67d5f60c97554d25 --- /dev/null +++ b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1/last_checkpoint @@ -0,0 +1 @@ +/mnt/petrelfs/luzeyu/workspace/fakebench/mmpretrain/workdir/resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1/epoch_10.pth \ No newline at end of file diff --git a/resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1/resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1.py b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1/resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1.py new file mode 100644 index 0000000000000000000000000000000000000000..ab532bdd5585a688820d77a34f991bc9b1d4859b --- /dev/null +++ b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1/resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1.py @@ -0,0 +1,230 @@ +optim_wrapper = dict( + optimizer=dict( + type='SGD', + lr=0.0001, + momentum=0.9, + weight_decay=0.0001, + _scope_='mmpretrain'), + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=False, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=512) +model = dict( + type='ImageClassifier', + backbone=dict( + frozen_stages=2, + type='ResNet', + depth=50, + num_stages=4, + out_indices=(3, ), + style='pytorch', + init_cfg=dict( + type='Pretrained', + checkpoint= + 'https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth', + prefix='backbone')), + neck=dict(type='GlobalAveragePooling'), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=2048, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + topk=1)) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stablediffusionV2-1-dpmsolver-25-1m.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/cc1m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV2-1-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = dict(type='Accuracy', topk=1) +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV2-1-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = dict(type='Accuracy', topk=1) +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/resnet50_2xb256_sdv2_1m_lr1e-4_aug_1e-1' diff --git a/resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1/20230604_225351/20230604_225351.log b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1/20230604_225351/20230604_225351.log new file mode 100644 index 0000000000000000000000000000000000000000..9e995554015746eee374a07a9f7b28e218a639ce --- /dev/null +++ b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1/20230604_225351/20230604_225351.log @@ -0,0 +1,1305 @@ +2023/06/04 22:53:54 - mmengine - INFO - +------------------------------------------------------------ +System environment: + sys.platform: linux + Python: 3.10.9 (main, Mar 8 2023, 10:47:38) [GCC 11.2.0] + CUDA available: True + numpy_random_seed: 1263609607 + GPU 0,1: NVIDIA A100-SXM4-80GB + CUDA_HOME: /mnt/petrelfs/share/cuda-11.6 + NVCC: Cuda compilation tools, release 11.6, V11.6.124 + GCC: gcc (GCC) 7.5.0 + PyTorch: 1.13.1 + PyTorch compiling details: PyTorch built with: + - GCC 9.3 + - C++ Version: 201402 + - Intel(R) oneAPI Math Kernel Library Version 2021.4-Product Build 20210904 for Intel(R) 64 architecture applications + - Intel(R) MKL-DNN v2.6.0 (Git Hash 52b5f107dd9cf10910aaa19cb47f3abf9b349815) + - OpenMP 201511 (a.k.a. OpenMP 4.5) + - LAPACK is enabled (usually provided by MKL) + - NNPACK is enabled + - CPU capability usage: AVX2 + - CUDA Runtime 11.6 + - NVCC architecture flags: -gencode;arch=compute_37,code=sm_37;-gencode;arch=compute_50,code=sm_50;-gencode;arch=compute_60,code=sm_60;-gencode;arch=compute_61,code=sm_61;-gencode;arch=compute_70,code=sm_70;-gencode;arch=compute_75,code=sm_75;-gencode;arch=compute_80,code=sm_80;-gencode;arch=compute_86,code=sm_86;-gencode;arch=compute_37,code=compute_37 + - CuDNN 8.3.2 (built against CUDA 11.5) + - Magma 2.6.1 + - Build settings: BLAS_INFO=mkl, BUILD_TYPE=Release, CUDA_VERSION=11.6, CUDNN_VERSION=8.3.2, CXX_COMPILER=/opt/rh/devtoolset-9/root/usr/bin/c++, CXX_FLAGS= -fabi-version=11 -Wno-deprecated -fvisibility-inlines-hidden -DUSE_PTHREADPOOL -fopenmp -DNDEBUG -DUSE_KINETO -DUSE_FBGEMM -DUSE_QNNPACK -DUSE_PYTORCH_QNNPACK -DUSE_XNNPACK -DSYMBOLICATE_MOBILE_DEBUG_HANDLE -DEDGE_PROFILER_USE_KINETO -O2 -fPIC -Wno-narrowing -Wall -Wextra -Werror=return-type -Werror=non-virtual-dtor -Wno-missing-field-initializers -Wno-type-limits -Wno-array-bounds -Wno-unknown-pragmas -Wunused-local-typedefs -Wno-unused-parameter -Wno-unused-function -Wno-unused-result -Wno-strict-overflow -Wno-strict-aliasing -Wno-error=deprecated-declarations -Wno-stringop-overflow -Wno-psabi -Wno-error=pedantic -Wno-error=redundant-decls -Wno-error=old-style-cast -fdiagnostics-color=always -faligned-new -Wno-unused-but-set-variable -Wno-maybe-uninitialized -fno-math-errno -fno-trapping-math -Werror=format -Werror=cast-function-type -Wno-stringop-overflow, LAPACK_INFO=mkl, PERF_WITH_AVX=1, PERF_WITH_AVX2=1, PERF_WITH_AVX512=1, TORCH_VERSION=1.13.1, USE_CUDA=ON, USE_CUDNN=ON, USE_EXCEPTION_PTR=1, USE_GFLAGS=OFF, USE_GLOG=OFF, USE_MKL=ON, USE_MKLDNN=ON, USE_MPI=OFF, USE_NCCL=ON, USE_NNPACK=ON, USE_OPENMP=ON, USE_ROCM=OFF, + + TorchVision: 0.14.1 + OpenCV: 4.7.0 + MMEngine: 0.7.3 + +Runtime environment: + cudnn_benchmark: True + mp_cfg: {'mp_start_method': 'fork', 'opencv_num_threads': 0} + dist_cfg: {'backend': 'nccl'} + seed: None + deterministic: False + Distributed launcher: slurm + Distributed training: True + GPU number: 2 +------------------------------------------------------------ + +2023/06/04 22:54:00 - mmengine - INFO - Config: +optim_wrapper = dict( + optimizer=dict( + type='SGD', + lr=0.0001, + momentum=0.9, + weight_decay=0.0001, + _scope_='mmpretrain'), + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=False, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=512) +model = dict( + type='ImageClassifier', + backbone=dict( + frozen_stages=2, + type='ResNet', + depth=50, + num_stages=4, + out_indices=(3, ), + style='pytorch', + init_cfg=dict( + type='Pretrained', + checkpoint= + 'https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth', + prefix='backbone')), + neck=dict(type='GlobalAveragePooling'), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=2048, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + topk=1)) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stablediffusionV2-1-dpmsolver-25-1m.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/cc1m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV2-1-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = dict(type='Accuracy', topk=1) +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV2-1-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = dict(type='Accuracy', topk=1) +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1' + +2023/06/04 22:54:13 - mmengine - INFO - Hooks will be executed in the following order: +before_run: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_load_checkpoint: +(ABOVE_NORMAL) EMAHook + -------------------- +before_train: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_train_epoch: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(NORMAL ) DistSamplerSeedHook + -------------------- +before_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook + -------------------- +after_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +after_train_epoch: +(NORMAL ) IterTimerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_val_epoch: +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook + -------------------- +before_val_iter: +(NORMAL ) IterTimerHook + -------------------- +after_val_iter: +(NORMAL ) IterTimerHook +(NORMAL ) VisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_val_epoch: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_save_checkpoint: +(ABOVE_NORMAL) EMAHook + -------------------- +after_train: +(VERY_LOW ) CheckpointHook + -------------------- +before_test_epoch: +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook + -------------------- +before_test_iter: +(NORMAL ) IterTimerHook + -------------------- +after_test_iter: +(NORMAL ) IterTimerHook +(NORMAL ) VisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_test_epoch: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_run: +(BELOW_NORMAL) LoggerHook + -------------------- +2023/06/04 22:54:34 - mmengine - INFO - load backbone in model from: https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth +Name of parameter - Initialization information + +backbone.conv1.weight - torch.Size([64, 3, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.bn1.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.bn1.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.conv1.weight - torch.Size([64, 64, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.bn1.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.bn1.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.conv2.weight - torch.Size([64, 64, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.bn2.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.bn2.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.conv3.weight - torch.Size([256, 64, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.bn3.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.bn3.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.downsample.0.weight - torch.Size([256, 64, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.downsample.1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.downsample.1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.conv1.weight - torch.Size([64, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.bn1.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.bn1.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.conv2.weight - torch.Size([64, 64, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.bn2.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.bn2.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.conv3.weight - torch.Size([256, 64, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.bn3.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.bn3.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.conv1.weight - torch.Size([64, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.bn1.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.bn1.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.conv2.weight - torch.Size([64, 64, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.bn2.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.bn2.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.conv3.weight - torch.Size([256, 64, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.bn3.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.bn3.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.conv1.weight - torch.Size([128, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.bn1.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.bn1.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.conv2.weight - torch.Size([128, 128, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.bn2.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.bn2.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.conv3.weight - torch.Size([512, 128, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.bn3.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.bn3.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.downsample.0.weight - torch.Size([512, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.downsample.1.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.downsample.1.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.conv1.weight - torch.Size([128, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.bn1.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.bn1.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.conv2.weight - torch.Size([128, 128, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.bn2.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.bn2.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.conv3.weight - torch.Size([512, 128, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.bn3.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.bn3.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.conv1.weight - torch.Size([128, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.bn1.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.bn1.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.conv2.weight - torch.Size([128, 128, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.bn2.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.bn2.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.conv3.weight - torch.Size([512, 128, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.bn3.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.bn3.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.conv1.weight - torch.Size([128, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.bn1.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.bn1.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.conv2.weight - torch.Size([128, 128, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.bn2.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.bn2.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.conv3.weight - torch.Size([512, 128, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.bn3.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.bn3.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.conv1.weight - torch.Size([256, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.bn1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.bn1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.conv2.weight - torch.Size([256, 256, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.bn2.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.bn2.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.conv3.weight - torch.Size([1024, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.bn3.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.bn3.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.downsample.0.weight - torch.Size([1024, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.downsample.1.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.downsample.1.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.conv1.weight - torch.Size([256, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.bn1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.bn1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.conv2.weight - torch.Size([256, 256, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.bn2.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.bn2.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.conv3.weight - torch.Size([1024, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.bn3.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.bn3.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.conv1.weight - torch.Size([256, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.bn1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.bn1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.conv2.weight - torch.Size([256, 256, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.bn2.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.bn2.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.conv3.weight - torch.Size([1024, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.bn3.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.bn3.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.conv1.weight - torch.Size([256, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.bn1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.bn1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.conv2.weight - torch.Size([256, 256, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.bn2.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.bn2.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.conv3.weight - torch.Size([1024, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.bn3.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.bn3.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.conv1.weight - torch.Size([256, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.bn1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.bn1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.conv2.weight - torch.Size([256, 256, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.bn2.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.bn2.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.conv3.weight - torch.Size([1024, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.bn3.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.bn3.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.conv1.weight - torch.Size([256, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.bn1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.bn1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.conv2.weight - torch.Size([256, 256, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.bn2.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.bn2.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.conv3.weight - torch.Size([1024, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.bn3.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.bn3.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.conv1.weight - torch.Size([512, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.bn1.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.bn1.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.conv2.weight - torch.Size([512, 512, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.bn2.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.bn2.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.conv3.weight - torch.Size([2048, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.bn3.weight - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.bn3.bias - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.downsample.0.weight - torch.Size([2048, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.downsample.1.weight - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.downsample.1.bias - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.conv1.weight - torch.Size([512, 2048, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.bn1.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.bn1.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.conv2.weight - torch.Size([512, 512, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.bn2.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.bn2.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.conv3.weight - torch.Size([2048, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.bn3.weight - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.bn3.bias - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.conv1.weight - torch.Size([512, 2048, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.bn1.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.bn1.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.conv2.weight - torch.Size([512, 512, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.bn2.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.bn2.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.conv3.weight - torch.Size([2048, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.bn3.weight - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.bn3.bias - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +head.fc.weight - torch.Size([2, 2048]): +NormalInit: mean=0, std=0.01, bias=0 + +head.fc.bias - torch.Size([2]): +NormalInit: mean=0, std=0.01, bias=0 +2023/06/04 22:54:34 - mmengine - WARNING - "FileClient" will be deprecated in future. Please use io functions in https://mmengine.readthedocs.io/en/latest/api/fileio.html#file-io +2023/06/04 22:54:34 - mmengine - WARNING - "HardDiskBackend" is the alias of "LocalBackend" and the former will be deprecated in future. +2023/06/04 22:54:34 - mmengine - INFO - Checkpoints will be saved to /mnt/petrelfs/luzeyu/workspace/fakebench/mmpretrain/workdir/resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1. +2023/06/04 22:55:43 - mmengine - INFO - Epoch(train) [1][ 100/3907] lr: 9.9999e-05 eta: 7:26:57 time: 0.6773 data_time: 0.2523 memory: 9436 loss: 0.6043 +2023/06/04 22:56:50 - mmengine - INFO - Epoch(train) [1][ 200/3907] lr: 9.9994e-05 eta: 7:21:19 time: 0.7268 data_time: 0.0753 memory: 6319 loss: 0.5358 +2023/06/04 22:57:58 - mmengine - INFO - Epoch(train) [1][ 300/3907] lr: 9.9987e-05 eta: 7:18:53 time: 0.6230 data_time: 0.0010 memory: 6319 loss: 0.4941 +2023/06/04 22:59:05 - mmengine - INFO - Epoch(train) [1][ 400/3907] lr: 9.9977e-05 eta: 7:17:19 time: 0.6354 data_time: 0.0009 memory: 6319 loss: 0.4298 +2023/06/04 23:00:12 - mmengine - INFO - Epoch(train) [1][ 500/3907] lr: 9.9964e-05 eta: 7:15:02 time: 0.7114 data_time: 0.0008 memory: 6319 loss: 0.4119 +2023/06/04 23:01:19 - mmengine - INFO - Epoch(train) [1][ 600/3907] lr: 9.9948e-05 eta: 7:12:36 time: 0.6574 data_time: 0.0012 memory: 6319 loss: 0.3756 +2023/06/04 23:02:24 - mmengine - INFO - Epoch(train) [1][ 700/3907] lr: 9.9929e-05 eta: 7:09:32 time: 0.6608 data_time: 0.0009 memory: 6319 loss: 0.3355 +2023/06/04 23:03:33 - mmengine - INFO - Epoch(train) [1][ 800/3907] lr: 9.9907e-05 eta: 7:09:44 time: 0.6427 data_time: 0.0013 memory: 6319 loss: 0.3327 +2023/06/04 23:04:39 - mmengine - INFO - Epoch(train) [1][ 900/3907] lr: 9.9882e-05 eta: 7:08:03 time: 0.6512 data_time: 0.0009 memory: 6319 loss: 0.3026 +2023/06/04 23:05:47 - mmengine - INFO - Exp name: resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1_20230604_225351 +2023/06/04 23:05:47 - mmengine - INFO - Epoch(train) [1][1000/3907] lr: 9.9855e-05 eta: 7:07:24 time: 0.6902 data_time: 0.0010 memory: 6319 loss: 0.2757 +2023/06/04 23:06:57 - mmengine - INFO - Epoch(train) [1][1100/3907] lr: 9.9824e-05 eta: 7:07:15 time: 0.7048 data_time: 0.0009 memory: 6319 loss: 0.2721 +2023/06/04 23:08:03 - mmengine - INFO - Epoch(train) [1][1200/3907] lr: 9.9791e-05 eta: 7:05:21 time: 0.6478 data_time: 0.0010 memory: 6319 loss: 0.2623 +2023/06/04 23:09:08 - mmengine - INFO - Epoch(train) [1][1300/3907] lr: 9.9755e-05 eta: 7:03:21 time: 0.6527 data_time: 0.0009 memory: 6319 loss: 0.2532 +2023/06/04 23:10:15 - mmengine - INFO - Epoch(train) [1][1400/3907] lr: 9.9716e-05 eta: 7:02:12 time: 0.7147 data_time: 0.0009 memory: 6319 loss: 0.2702 +2023/06/04 23:11:21 - mmengine - INFO - Epoch(train) [1][1500/3907] lr: 9.9674e-05 eta: 7:00:27 time: 0.6704 data_time: 0.0008 memory: 6319 loss: 0.2465 +2023/06/04 23:12:29 - mmengine - INFO - Epoch(train) [1][1600/3907] lr: 9.9629e-05 eta: 6:59:41 time: 0.6794 data_time: 0.0011 memory: 6319 loss: 0.2376 +2023/06/04 23:13:30 - mmengine - INFO - Epoch(train) [1][1700/3907] lr: 9.9581e-05 eta: 6:56:22 time: 0.6408 data_time: 0.0008 memory: 6319 loss: 0.2211 +2023/06/04 23:14:35 - mmengine - INFO - Epoch(train) [1][1800/3907] lr: 9.9530e-05 eta: 6:54:22 time: 0.6415 data_time: 0.1043 memory: 6319 loss: 0.2324 +2023/06/04 23:15:40 - mmengine - INFO - Epoch(train) [1][1900/3907] lr: 9.9476e-05 eta: 6:52:54 time: 0.6706 data_time: 0.5309 memory: 6319 loss: 0.2141 +2023/06/04 23:16:48 - mmengine - INFO - Exp name: resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1_20230604_225351 +2023/06/04 23:16:48 - mmengine - INFO - Epoch(train) [1][2000/3907] lr: 9.9420e-05 eta: 6:52:10 time: 0.7201 data_time: 0.5010 memory: 6319 loss: 0.2260 +2023/06/04 23:17:55 - mmengine - INFO - Epoch(train) [1][2100/3907] lr: 9.9361e-05 eta: 6:51:10 time: 0.7120 data_time: 0.5712 memory: 6319 loss: 0.2068 +2023/06/04 23:19:04 - mmengine - INFO - Epoch(train) [1][2200/3907] lr: 9.9298e-05 eta: 6:50:30 time: 0.7229 data_time: 0.5833 memory: 6319 loss: 0.1956 +2023/06/04 23:20:12 - mmengine - INFO - Epoch(train) [1][2300/3907] lr: 9.9233e-05 eta: 6:49:52 time: 0.6740 data_time: 0.5348 memory: 6319 loss: 0.2159 +2023/06/04 23:21:21 - mmengine - INFO - Epoch(train) [1][2400/3907] lr: 9.9165e-05 eta: 6:49:13 time: 0.6373 data_time: 0.4979 memory: 6319 loss: 0.1922 +2023/06/04 23:22:27 - mmengine - INFO - Epoch(train) [1][2500/3907] lr: 9.9095e-05 eta: 6:48:00 time: 0.6930 data_time: 0.5531 memory: 6319 loss: 0.2031 +2023/06/04 23:23:35 - mmengine - INFO - Epoch(train) [1][2600/3907] lr: 9.9021e-05 eta: 6:47:03 time: 0.6640 data_time: 0.5237 memory: 6319 loss: 0.1767 +2023/06/04 23:24:43 - mmengine - INFO - Epoch(train) [1][2700/3907] lr: 9.8944e-05 eta: 6:46:08 time: 0.6682 data_time: 0.5284 memory: 6319 loss: 0.2076 +2023/06/04 23:25:51 - mmengine - INFO - Epoch(train) [1][2800/3907] lr: 9.8865e-05 eta: 6:45:11 time: 0.6659 data_time: 0.5254 memory: 6319 loss: 0.1828 +2023/06/04 23:27:00 - mmengine - INFO - Epoch(train) [1][2900/3907] lr: 9.8783e-05 eta: 6:44:35 time: 0.6735 data_time: 0.5331 memory: 6319 loss: 0.1638 +2023/06/04 23:28:07 - mmengine - INFO - Exp name: resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1_20230604_225351 +2023/06/04 23:28:07 - mmengine - INFO - Epoch(train) [1][3000/3907] lr: 9.8698e-05 eta: 6:43:29 time: 0.6812 data_time: 0.5406 memory: 6319 loss: 0.1844 +2023/06/04 23:29:15 - mmengine - INFO - Epoch(train) [1][3100/3907] lr: 9.8610e-05 eta: 6:42:25 time: 0.6715 data_time: 0.5321 memory: 6319 loss: 0.1827 +2023/06/04 23:30:21 - mmengine - INFO - Epoch(train) [1][3200/3907] lr: 9.8519e-05 eta: 6:41:11 time: 0.6751 data_time: 0.5345 memory: 6319 loss: 0.1714 +2023/06/04 23:31:28 - mmengine - INFO - Epoch(train) [1][3300/3907] lr: 9.8426e-05 eta: 6:39:54 time: 0.6606 data_time: 0.5207 memory: 6319 loss: 0.1628 +2023/06/04 23:32:35 - mmengine - INFO - Epoch(train) [1][3400/3907] lr: 9.8330e-05 eta: 6:38:50 time: 0.6689 data_time: 0.5295 memory: 6319 loss: 0.1740 +2023/06/04 23:33:41 - mmengine - INFO - Epoch(train) [1][3500/3907] lr: 9.8231e-05 eta: 6:37:30 time: 0.6999 data_time: 0.5588 memory: 6319 loss: 0.1812 +2023/06/04 23:34:47 - mmengine - INFO - Epoch(train) [1][3600/3907] lr: 9.8129e-05 eta: 6:36:16 time: 0.6636 data_time: 0.5235 memory: 6319 loss: 0.1638 +2023/06/04 23:35:53 - mmengine - INFO - Epoch(train) [1][3700/3907] lr: 9.8024e-05 eta: 6:35:01 time: 0.6513 data_time: 0.5098 memory: 6319 loss: 0.1713 +2023/06/04 23:37:00 - mmengine - INFO - Epoch(train) [1][3800/3907] lr: 9.7917e-05 eta: 6:33:52 time: 0.6676 data_time: 0.5158 memory: 6319 loss: 0.1612 +2023/06/04 23:38:06 - mmengine - INFO - Epoch(train) [1][3900/3907] lr: 9.7806e-05 eta: 6:32:32 time: 0.6452 data_time: 0.5063 memory: 6319 loss: 0.1737 +2023/06/04 23:38:12 - mmengine - INFO - Exp name: resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1_20230604_225351 +2023/06/04 23:38:12 - mmengine - INFO - Saving checkpoint at 1 epochs +2023/06/04 23:39:03 - mmengine - INFO - Epoch(val) [1][57/57] accuracy/top1: 79.2874 data_time: 0.5959 time: 0.6884 +2023/06/04 23:40:12 - mmengine - INFO - Exp name: resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1_20230604_225351 +2023/06/04 23:40:13 - mmengine - INFO - Epoch(train) [2][ 100/3907] lr: 9.7685e-05 eta: 6:31:58 time: 0.6768 data_time: 0.5371 memory: 6319 loss: 0.1528 +2023/06/04 23:41:21 - mmengine - INFO - Epoch(train) [2][ 200/3907] lr: 9.7570e-05 eta: 6:30:56 time: 0.6940 data_time: 0.5544 memory: 6319 loss: 0.1413 +2023/06/04 23:42:28 - mmengine - INFO - Epoch(train) [2][ 300/3907] lr: 9.7451e-05 eta: 6:29:47 time: 0.6467 data_time: 0.5068 memory: 6319 loss: 0.1545 +2023/06/04 23:43:35 - mmengine - INFO - Epoch(train) [2][ 400/3907] lr: 9.7329e-05 eta: 6:28:42 time: 0.6443 data_time: 0.5037 memory: 6319 loss: 0.1405 +2023/06/04 23:44:42 - mmengine - INFO - Epoch(train) [2][ 500/3907] lr: 9.7205e-05 eta: 6:27:34 time: 0.5851 data_time: 0.4447 memory: 6319 loss: 0.1506 +2023/06/04 23:45:47 - mmengine - INFO - Epoch(train) [2][ 600/3907] lr: 9.7078e-05 eta: 6:26:12 time: 0.6615 data_time: 0.5212 memory: 6319 loss: 0.1310 +2023/06/04 23:46:54 - mmengine - INFO - Epoch(train) [2][ 700/3907] lr: 9.6949e-05 eta: 6:25:03 time: 0.6660 data_time: 0.5258 memory: 6319 loss: 0.1483 +2023/06/04 23:48:00 - mmengine - INFO - Epoch(train) [2][ 800/3907] lr: 9.6816e-05 eta: 6:23:50 time: 0.6336 data_time: 0.4931 memory: 6319 loss: 0.1387 +2023/06/04 23:49:07 - mmengine - INFO - Epoch(train) [2][ 900/3907] lr: 9.6681e-05 eta: 6:22:38 time: 0.6974 data_time: 0.5573 memory: 6319 loss: 0.1431 +2023/06/04 23:50:14 - mmengine - INFO - Epoch(train) [2][1000/3907] lr: 9.6544e-05 eta: 6:21:33 time: 0.6310 data_time: 0.4908 memory: 6319 loss: 0.1592 +2023/06/04 23:51:20 - mmengine - INFO - Exp name: resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1_20230604_225351 +2023/06/04 23:51:21 - mmengine - INFO - Epoch(train) [2][1100/3907] lr: 9.6403e-05 eta: 6:20:29 time: 0.6652 data_time: 0.5240 memory: 6319 loss: 0.1495 +2023/06/04 23:52:28 - mmengine - INFO - Epoch(train) [2][1200/3907] lr: 9.6260e-05 eta: 6:19:21 time: 0.6592 data_time: 0.5177 memory: 6319 loss: 0.1486 +2023/06/04 23:53:35 - mmengine - INFO - Epoch(train) [2][1300/3907] lr: 9.6114e-05 eta: 6:18:10 time: 0.7580 data_time: 0.6181 memory: 6319 loss: 0.1456 +2023/06/04 23:54:41 - mmengine - INFO - Epoch(train) [2][1400/3907] lr: 9.5966e-05 eta: 6:16:57 time: 0.6575 data_time: 0.5168 memory: 6319 loss: 0.1564 +2023/06/04 23:55:48 - mmengine - INFO - Epoch(train) [2][1500/3907] lr: 9.5815e-05 eta: 6:15:52 time: 0.6685 data_time: 0.5288 memory: 6319 loss: 0.1406 +2023/06/04 23:56:54 - mmengine - INFO - Epoch(train) [2][1600/3907] lr: 9.5661e-05 eta: 6:14:36 time: 0.7404 data_time: 0.6000 memory: 6319 loss: 0.1410 +2023/06/04 23:57:58 - mmengine - INFO - Epoch(train) [2][1700/3907] lr: 9.5505e-05 eta: 6:13:15 time: 0.6693 data_time: 0.5292 memory: 6319 loss: 0.1464 +2023/06/04 23:59:04 - mmengine - INFO - Epoch(train) [2][1800/3907] lr: 9.5346e-05 eta: 6:12:01 time: 0.6595 data_time: 0.5193 memory: 6319 loss: 0.1251 +2023/06/05 00:00:10 - mmengine - INFO - Epoch(train) [2][1900/3907] lr: 9.5184e-05 eta: 6:10:51 time: 0.6812 data_time: 0.5405 memory: 6319 loss: 0.1406 +2023/06/05 00:01:17 - mmengine - INFO - Epoch(train) [2][2000/3907] lr: 9.5020e-05 eta: 6:09:43 time: 0.6625 data_time: 0.5032 memory: 6319 loss: 0.1317 +2023/06/05 00:02:22 - mmengine - INFO - Exp name: resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1_20230604_225351 +2023/06/05 00:02:23 - mmengine - INFO - Epoch(train) [2][2100/3907] lr: 9.4854e-05 eta: 6:08:34 time: 0.6838 data_time: 0.5442 memory: 6319 loss: 0.1416 +2023/06/05 00:03:30 - mmengine - INFO - Epoch(train) [2][2200/3907] lr: 9.4684e-05 eta: 6:07:27 time: 0.6704 data_time: 0.5202 memory: 6319 loss: 0.1402 +2023/06/05 00:04:36 - mmengine - INFO - Epoch(train) [2][2300/3907] lr: 9.4512e-05 eta: 6:06:12 time: 0.6754 data_time: 0.5359 memory: 6319 loss: 0.1390 +2023/06/05 00:05:42 - mmengine - INFO - Epoch(train) [2][2400/3907] lr: 9.4338e-05 eta: 6:05:01 time: 0.6629 data_time: 0.4885 memory: 6319 loss: 0.1325 +2023/06/05 00:06:49 - mmengine - INFO - Epoch(train) [2][2500/3907] lr: 9.4161e-05 eta: 6:03:57 time: 0.6119 data_time: 0.4711 memory: 6319 loss: 0.1328 +2023/06/05 00:07:55 - mmengine - INFO - Epoch(train) [2][2600/3907] lr: 9.3981e-05 eta: 6:02:43 time: 0.6861 data_time: 0.5423 memory: 6319 loss: 0.1291 +2023/06/05 00:09:00 - mmengine - INFO - Epoch(train) [2][2700/3907] lr: 9.3799e-05 eta: 6:01:30 time: 0.6600 data_time: 0.5194 memory: 6319 loss: 0.1384 +2023/06/05 00:10:06 - mmengine - INFO - Epoch(train) [2][2800/3907] lr: 9.3615e-05 eta: 6:00:19 time: 0.6556 data_time: 0.5155 memory: 6319 loss: 0.1473 +2023/06/05 00:11:11 - mmengine - INFO - Epoch(train) [2][2900/3907] lr: 9.3428e-05 eta: 5:59:04 time: 0.6514 data_time: 0.5120 memory: 6319 loss: 0.1209 +2023/06/05 00:12:19 - mmengine - INFO - Epoch(train) [2][3000/3907] lr: 9.3238e-05 eta: 5:58:00 time: 0.7055 data_time: 0.5649 memory: 6319 loss: 0.1435 +2023/06/05 00:13:24 - mmengine - INFO - Exp name: resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1_20230604_225351 +2023/06/05 00:13:25 - mmengine - INFO - Epoch(train) [2][3100/3907] lr: 9.3046e-05 eta: 5:56:49 time: 0.6247 data_time: 0.4848 memory: 6319 loss: 0.1390 +2023/06/05 00:14:31 - mmengine - INFO - Epoch(train) [2][3200/3907] lr: 9.2852e-05 eta: 5:55:41 time: 0.6478 data_time: 0.5071 memory: 6319 loss: 0.1205 +2023/06/05 00:15:39 - mmengine - INFO - Epoch(train) [2][3300/3907] lr: 9.2655e-05 eta: 5:54:39 time: 0.6395 data_time: 0.4996 memory: 6319 loss: 0.1324 +2023/06/05 00:16:44 - mmengine - INFO - Epoch(train) [2][3400/3907] lr: 9.2456e-05 eta: 5:53:25 time: 0.6907 data_time: 0.5493 memory: 6319 loss: 0.1269 +2023/06/05 00:17:52 - mmengine - INFO - Epoch(train) [2][3500/3907] lr: 9.2254e-05 eta: 5:52:22 time: 0.7101 data_time: 0.5684 memory: 6319 loss: 0.1229 +2023/06/05 00:19:00 - mmengine - INFO - Epoch(train) [2][3600/3907] lr: 9.2050e-05 eta: 5:51:23 time: 0.6303 data_time: 0.4899 memory: 6319 loss: 0.1143 +2023/06/05 00:20:08 - mmengine - INFO - Epoch(train) [2][3700/3907] lr: 9.1843e-05 eta: 5:50:20 time: 0.6717 data_time: 0.5326 memory: 6319 loss: 0.1283 +2023/06/05 00:21:13 - mmengine - INFO - Epoch(train) [2][3800/3907] lr: 9.1634e-05 eta: 5:49:07 time: 0.6564 data_time: 0.5166 memory: 6319 loss: 0.1219 +2023/06/05 00:22:19 - mmengine - INFO - Epoch(train) [2][3900/3907] lr: 9.1423e-05 eta: 5:47:55 time: 0.6266 data_time: 0.4867 memory: 6319 loss: 0.1200 +2023/06/05 00:22:24 - mmengine - INFO - Exp name: resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1_20230604_225351 +2023/06/05 00:22:24 - mmengine - INFO - Saving checkpoint at 2 epochs +2023/06/05 00:23:09 - mmengine - INFO - Epoch(val) [2][57/57] accuracy/top1: 84.4160 data_time: 0.4754 time: 0.5640 +2023/06/05 00:24:18 - mmengine - INFO - Epoch(train) [3][ 100/3907] lr: 9.1194e-05 eta: 5:46:55 time: 0.6764 data_time: 0.5369 memory: 6319 loss: 0.1321 +2023/06/05 00:25:16 - mmengine - INFO - Exp name: resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1_20230604_225351 +2023/06/05 00:25:23 - mmengine - INFO - Epoch(train) [3][ 200/3907] lr: 9.0978e-05 eta: 5:45:43 time: 0.6464 data_time: 0.4976 memory: 6319 loss: 0.1009 +2023/06/05 00:26:30 - mmengine - INFO - Epoch(train) [3][ 300/3907] lr: 9.0759e-05 eta: 5:44:35 time: 0.6661 data_time: 0.5266 memory: 6319 loss: 0.1469 +2023/06/05 00:27:36 - mmengine - INFO - Epoch(train) [3][ 400/3907] lr: 9.0539e-05 eta: 5:43:27 time: 0.6567 data_time: 0.5115 memory: 6319 loss: 0.1193 +2023/06/05 00:28:42 - mmengine - INFO - Epoch(train) [3][ 500/3907] lr: 9.0315e-05 eta: 5:42:17 time: 0.6791 data_time: 0.5386 memory: 6319 loss: 0.1172 +2023/06/05 00:29:49 - mmengine - INFO - Epoch(train) [3][ 600/3907] lr: 9.0090e-05 eta: 5:41:11 time: 0.6794 data_time: 0.5395 memory: 6319 loss: 0.1280 +2023/06/05 00:30:55 - mmengine - INFO - Epoch(train) [3][ 700/3907] lr: 8.9862e-05 eta: 5:40:01 time: 0.6367 data_time: 0.4958 memory: 6319 loss: 0.1217 +2023/06/05 00:32:11 - mmengine - INFO - Epoch(train) [3][ 800/3907] lr: 8.9632e-05 eta: 5:39:25 time: 0.7412 data_time: 0.6008 memory: 6319 loss: 0.1163 +2023/06/05 00:33:15 - mmengine - INFO - Epoch(train) [3][ 900/3907] lr: 8.9400e-05 eta: 5:38:08 time: 0.6671 data_time: 0.5272 memory: 6319 loss: 0.1225 +2023/06/05 00:34:19 - mmengine - INFO - Epoch(train) [3][1000/3907] lr: 8.9166e-05 eta: 5:36:52 time: 0.6386 data_time: 0.4976 memory: 6319 loss: 0.1387 +2023/06/05 00:35:24 - mmengine - INFO - Epoch(train) [3][1100/3907] lr: 8.8929e-05 eta: 5:35:41 time: 0.6426 data_time: 0.5019 memory: 6319 loss: 0.1089 +2023/06/05 00:36:23 - mmengine - INFO - Exp name: resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1_20230604_225351 +2023/06/05 00:36:30 - mmengine - INFO - Epoch(train) [3][1200/3907] lr: 8.8691e-05 eta: 5:34:30 time: 0.6732 data_time: 0.5331 memory: 6319 loss: 0.1227 +2023/06/05 00:37:36 - mmengine - INFO - Epoch(train) [3][1300/3907] lr: 8.8450e-05 eta: 5:33:22 time: 0.6398 data_time: 0.4999 memory: 6319 loss: 0.1164 +2023/06/05 00:38:42 - mmengine - INFO - Epoch(train) [3][1400/3907] lr: 8.8206e-05 eta: 5:32:12 time: 0.6419 data_time: 0.5013 memory: 6319 loss: 0.1125 +2023/06/05 00:39:49 - mmengine - INFO - Epoch(train) [3][1500/3907] lr: 8.7961e-05 eta: 5:31:06 time: 0.6775 data_time: 0.5387 memory: 6319 loss: 0.1149 +2023/06/05 00:40:56 - mmengine - INFO - Epoch(train) [3][1600/3907] lr: 8.7714e-05 eta: 5:29:59 time: 0.6409 data_time: 0.5014 memory: 6319 loss: 0.1108 +2023/06/05 00:42:12 - mmengine - INFO - Epoch(train) [3][1700/3907] lr: 8.7464e-05 eta: 5:29:20 time: 0.8585 data_time: 0.7175 memory: 6319 loss: 0.1080 +2023/06/05 00:43:18 - mmengine - INFO - Epoch(train) [3][1800/3907] lr: 8.7213e-05 eta: 5:28:11 time: 0.6469 data_time: 0.5071 memory: 6319 loss: 0.1144 +2023/06/05 00:44:22 - mmengine - INFO - Epoch(train) [3][1900/3907] lr: 8.6959e-05 eta: 5:26:56 time: 0.6303 data_time: 0.4912 memory: 6319 loss: 0.1140 +2023/06/05 00:45:28 - mmengine - INFO - Epoch(train) [3][2000/3907] lr: 8.6703e-05 eta: 5:25:48 time: 0.6445 data_time: 0.5045 memory: 6319 loss: 0.1196 +2023/06/05 00:46:35 - mmengine - INFO - Epoch(train) [3][2100/3907] lr: 8.6445e-05 eta: 5:24:41 time: 0.6851 data_time: 0.5450 memory: 6319 loss: 0.1173 +2023/06/05 00:47:35 - mmengine - INFO - Exp name: resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1_20230604_225351 +2023/06/05 00:47:43 - mmengine - INFO - Epoch(train) [3][2200/3907] lr: 8.6186e-05 eta: 5:23:36 time: 0.6610 data_time: 0.5212 memory: 6319 loss: 0.1034 +2023/06/05 00:48:48 - mmengine - INFO - Epoch(train) [3][2300/3907] lr: 8.5924e-05 eta: 5:22:27 time: 0.6864 data_time: 0.5474 memory: 6319 loss: 0.1198 +2023/06/05 00:49:55 - mmengine - INFO - Epoch(train) [3][2400/3907] lr: 8.5660e-05 eta: 5:21:20 time: 0.7293 data_time: 0.5904 memory: 6319 loss: 0.1248 +2023/06/05 00:51:02 - mmengine - INFO - Epoch(train) [3][2500/3907] lr: 8.5394e-05 eta: 5:20:14 time: 0.6492 data_time: 0.5097 memory: 6319 loss: 0.1360 +2023/06/05 00:52:12 - mmengine - INFO - Epoch(train) [3][2600/3907] lr: 8.5126e-05 eta: 5:19:14 time: 0.8287 data_time: 0.6894 memory: 6319 loss: 0.1009 +2023/06/05 00:53:28 - mmengine - INFO - Epoch(train) [3][2700/3907] lr: 8.4856e-05 eta: 5:18:33 time: 0.7133 data_time: 0.5736 memory: 6319 loss: 0.1160 +2023/06/05 00:54:39 - mmengine - INFO - Epoch(train) [3][2800/3907] lr: 8.4585e-05 eta: 5:17:37 time: 0.6500 data_time: 0.5090 memory: 6319 loss: 0.0941 +2023/06/05 00:55:45 - mmengine - INFO - Epoch(train) [3][2900/3907] lr: 8.4311e-05 eta: 5:16:26 time: 0.6542 data_time: 0.5090 memory: 6319 loss: 0.1214 +2023/06/05 00:56:51 - mmengine - INFO - Epoch(train) [3][3000/3907] lr: 8.4036e-05 eta: 5:15:17 time: 0.6606 data_time: 0.5095 memory: 6319 loss: 0.1028 +2023/06/05 00:57:57 - mmengine - INFO - Epoch(train) [3][3100/3907] lr: 8.3758e-05 eta: 5:14:09 time: 0.6595 data_time: 0.5198 memory: 6319 loss: 0.1071 +2023/06/05 00:58:57 - mmengine - INFO - Exp name: resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1_20230604_225351 +2023/06/05 00:59:04 - mmengine - INFO - Epoch(train) [3][3200/3907] lr: 8.3479e-05 eta: 5:13:02 time: 0.6474 data_time: 0.4932 memory: 6319 loss: 0.1060 +2023/06/05 01:00:12 - mmengine - INFO - Epoch(train) [3][3300/3907] lr: 8.3198e-05 eta: 5:11:56 time: 0.6125 data_time: 0.4724 memory: 6319 loss: 0.1152 +2023/06/05 01:01:17 - mmengine - INFO - Epoch(train) [3][3400/3907] lr: 8.2915e-05 eta: 5:10:45 time: 0.6834 data_time: 0.5436 memory: 6319 loss: 0.0990 +2023/06/05 01:02:25 - mmengine - INFO - Epoch(train) [3][3500/3907] lr: 8.2630e-05 eta: 5:09:42 time: 0.6846 data_time: 0.5444 memory: 6319 loss: 0.1035 +2023/06/05 01:03:33 - mmengine - INFO - Epoch(train) [3][3600/3907] lr: 8.2344e-05 eta: 5:08:37 time: 0.6770 data_time: 0.5355 memory: 6319 loss: 0.1068 +2023/06/05 01:04:42 - mmengine - INFO - Epoch(train) [3][3700/3907] lr: 8.2056e-05 eta: 5:07:34 time: 0.6740 data_time: 0.5349 memory: 6319 loss: 0.1099 +2023/06/05 01:05:50 - mmengine - INFO - Epoch(train) [3][3800/3907] lr: 8.1765e-05 eta: 5:06:29 time: 0.6752 data_time: 0.5361 memory: 6319 loss: 0.1061 +2023/06/05 01:06:57 - mmengine - INFO - Epoch(train) [3][3900/3907] lr: 8.1474e-05 eta: 5:05:22 time: 0.6158 data_time: 0.4763 memory: 6319 loss: 0.1085 +2023/06/05 01:07:02 - mmengine - INFO - Exp name: resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1_20230604_225351 +2023/06/05 01:07:02 - mmengine - INFO - Saving checkpoint at 3 epochs +2023/06/05 01:07:51 - mmengine - INFO - Epoch(val) [3][57/57] accuracy/top1: 87.3421 data_time: 0.5443 time: 0.6337 +2023/06/05 01:09:01 - mmengine - INFO - Epoch(train) [4][ 100/3907] lr: 8.1160e-05 eta: 5:04:19 time: 0.7443 data_time: 0.5888 memory: 6319 loss: 0.1141 +2023/06/05 01:10:08 - mmengine - INFO - Epoch(train) [4][ 200/3907] lr: 8.0864e-05 eta: 5:03:12 time: 0.6592 data_time: 0.5061 memory: 6319 loss: 0.1029 +2023/06/05 01:11:02 - mmengine - INFO - Exp name: resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1_20230604_225351 +2023/06/05 01:11:17 - mmengine - INFO - Epoch(train) [4][ 300/3907] lr: 8.0567e-05 eta: 5:02:09 time: 0.6956 data_time: 0.4117 memory: 6319 loss: 0.1024 +2023/06/05 01:12:23 - mmengine - INFO - Epoch(train) [4][ 400/3907] lr: 8.0269e-05 eta: 5:01:00 time: 0.6918 data_time: 0.3840 memory: 6319 loss: 0.1088 +2023/06/05 01:13:31 - mmengine - INFO - Epoch(train) [4][ 500/3907] lr: 7.9969e-05 eta: 4:59:55 time: 0.6833 data_time: 0.3068 memory: 6319 loss: 0.1142 +2023/06/05 01:14:39 - mmengine - INFO - Epoch(train) [4][ 600/3907] lr: 7.9667e-05 eta: 4:58:51 time: 0.6850 data_time: 0.3399 memory: 6319 loss: 0.1026 +2023/06/05 01:15:47 - mmengine - INFO - Epoch(train) [4][ 700/3907] lr: 7.9363e-05 eta: 4:57:46 time: 0.6532 data_time: 0.3745 memory: 6319 loss: 0.1052 +2023/06/05 01:16:55 - mmengine - INFO - Epoch(train) [4][ 800/3907] lr: 7.9058e-05 eta: 4:56:40 time: 0.6609 data_time: 0.4166 memory: 6319 loss: 0.1049 +2023/06/05 01:18:03 - mmengine - INFO - Epoch(train) [4][ 900/3907] lr: 7.8752e-05 eta: 4:55:35 time: 0.6547 data_time: 0.3144 memory: 6319 loss: 0.0911 +2023/06/05 01:19:10 - mmengine - INFO - Epoch(train) [4][1000/3907] lr: 7.8444e-05 eta: 4:54:29 time: 0.7190 data_time: 0.2799 memory: 6319 loss: 0.1041 +2023/06/05 01:20:17 - mmengine - INFO - Epoch(train) [4][1100/3907] lr: 7.8134e-05 eta: 4:53:21 time: 0.6120 data_time: 0.3072 memory: 6319 loss: 0.0889 +2023/06/05 01:21:22 - mmengine - INFO - Epoch(train) [4][1200/3907] lr: 7.7823e-05 eta: 4:52:10 time: 0.6498 data_time: 0.5094 memory: 6319 loss: 0.1102 +2023/06/05 01:22:15 - mmengine - INFO - Exp name: resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1_20230604_225351 +2023/06/05 01:22:29 - mmengine - INFO - Epoch(train) [4][1300/3907] lr: 7.7510e-05 eta: 4:51:02 time: 0.6814 data_time: 0.5414 memory: 6319 loss: 0.1199 +2023/06/05 01:23:37 - mmengine - INFO - Epoch(train) [4][1400/3907] lr: 7.7196e-05 eta: 4:49:58 time: 0.7173 data_time: 0.5784 memory: 6319 loss: 0.1049 +2023/06/05 01:24:45 - mmengine - INFO - Epoch(train) [4][1500/3907] lr: 7.6881e-05 eta: 4:48:52 time: 0.6452 data_time: 0.5041 memory: 6319 loss: 0.1167 +2023/06/05 01:25:52 - mmengine - INFO - Epoch(train) [4][1600/3907] lr: 7.6564e-05 eta: 4:47:46 time: 0.6744 data_time: 0.5350 memory: 6319 loss: 0.1014 +2023/06/05 01:27:00 - mmengine - INFO - Epoch(train) [4][1700/3907] lr: 7.6246e-05 eta: 4:46:40 time: 0.7068 data_time: 0.5671 memory: 6319 loss: 0.1028 +2023/06/05 01:28:07 - mmengine - INFO - Epoch(train) [4][1800/3907] lr: 7.5926e-05 eta: 4:45:34 time: 0.6507 data_time: 0.5110 memory: 6319 loss: 0.1068 +2023/06/05 01:29:17 - mmengine - INFO - Epoch(train) [4][1900/3907] lr: 7.5605e-05 eta: 4:44:31 time: 0.7270 data_time: 0.5814 memory: 6319 loss: 0.1104 +2023/06/05 01:30:25 - mmengine - INFO - Epoch(train) [4][2000/3907] lr: 7.5283e-05 eta: 4:43:26 time: 0.7088 data_time: 0.5673 memory: 6319 loss: 0.1058 +2023/06/05 01:31:34 - mmengine - INFO - Epoch(train) [4][2100/3907] lr: 7.4959e-05 eta: 4:42:22 time: 0.7210 data_time: 0.5114 memory: 6319 loss: 0.0918 +2023/06/05 01:32:41 - mmengine - INFO - Epoch(train) [4][2200/3907] lr: 7.4634e-05 eta: 4:41:16 time: 0.6922 data_time: 0.5522 memory: 6319 loss: 0.1060 +2023/06/05 01:33:37 - mmengine - INFO - Exp name: resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1_20230604_225351 +2023/06/05 01:33:51 - mmengine - INFO - Epoch(train) [4][2300/3907] lr: 7.4308e-05 eta: 4:40:13 time: 0.6872 data_time: 0.5371 memory: 6319 loss: 0.1012 +2023/06/05 01:34:57 - mmengine - INFO - Epoch(train) [4][2400/3907] lr: 7.3980e-05 eta: 4:39:04 time: 0.6490 data_time: 0.5098 memory: 6319 loss: 0.1126 +2023/06/05 01:36:13 - mmengine - INFO - Epoch(train) [4][2500/3907] lr: 7.3652e-05 eta: 4:38:13 time: 0.7904 data_time: 0.6447 memory: 6319 loss: 0.0988 +2023/06/05 01:37:20 - mmengine - INFO - Epoch(train) [4][2600/3907] lr: 7.3322e-05 eta: 4:37:06 time: 0.6249 data_time: 0.4853 memory: 6319 loss: 0.1030 +2023/06/05 01:38:24 - mmengine - INFO - Epoch(train) [4][2700/3907] lr: 7.2991e-05 eta: 4:35:53 time: 0.6624 data_time: 0.5232 memory: 6319 loss: 0.0952 +2023/06/05 01:39:32 - mmengine - INFO - Epoch(train) [4][2800/3907] lr: 7.2659e-05 eta: 4:34:47 time: 0.6820 data_time: 0.5393 memory: 6319 loss: 0.0991 +2023/06/05 01:40:41 - mmengine - INFO - Epoch(train) [4][2900/3907] lr: 7.2325e-05 eta: 4:33:43 time: 0.6450 data_time: 0.5042 memory: 6319 loss: 0.0953 +2023/06/05 01:41:49 - mmengine - INFO - Epoch(train) [4][3000/3907] lr: 7.1991e-05 eta: 4:32:37 time: 0.6776 data_time: 0.5369 memory: 6319 loss: 0.0873 +2023/06/05 01:42:56 - mmengine - INFO - Epoch(train) [4][3100/3907] lr: 7.1655e-05 eta: 4:31:30 time: 0.7061 data_time: 0.5643 memory: 6319 loss: 0.1050 +2023/06/05 01:44:04 - mmengine - INFO - Epoch(train) [4][3200/3907] lr: 7.1318e-05 eta: 4:30:23 time: 0.6815 data_time: 0.5413 memory: 6319 loss: 0.0882 +2023/06/05 01:44:58 - mmengine - INFO - Exp name: resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1_20230604_225351 +2023/06/05 01:45:11 - mmengine - INFO - Epoch(train) [4][3300/3907] lr: 7.0981e-05 eta: 4:29:17 time: 0.6491 data_time: 0.5089 memory: 6319 loss: 0.0940 +2023/06/05 01:46:19 - mmengine - INFO - Epoch(train) [4][3400/3907] lr: 7.0642e-05 eta: 4:28:10 time: 0.6593 data_time: 0.5177 memory: 6319 loss: 0.0992 +2023/06/05 01:47:26 - mmengine - INFO - Epoch(train) [4][3500/3907] lr: 7.0302e-05 eta: 4:27:03 time: 0.6801 data_time: 0.5397 memory: 6319 loss: 0.0944 +2023/06/05 01:48:37 - mmengine - INFO - Epoch(train) [4][3600/3907] lr: 6.9961e-05 eta: 4:26:02 time: 0.7637 data_time: 0.6162 memory: 6319 loss: 0.0895 +2023/06/05 01:49:50 - mmengine - INFO - Epoch(train) [4][3700/3907] lr: 6.9620e-05 eta: 4:25:03 time: 0.6629 data_time: 0.5222 memory: 6319 loss: 0.0996 +2023/06/05 01:50:56 - mmengine - INFO - Epoch(train) [4][3800/3907] lr: 6.9277e-05 eta: 4:23:54 time: 0.6424 data_time: 0.4960 memory: 6319 loss: 0.0930 +2023/06/05 01:52:02 - mmengine - INFO - Epoch(train) [4][3900/3907] lr: 6.8933e-05 eta: 4:22:45 time: 0.6563 data_time: 0.5162 memory: 6319 loss: 0.0981 +2023/06/05 01:52:08 - mmengine - INFO - Exp name: resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1_20230604_225351 +2023/06/05 01:52:08 - mmengine - INFO - Saving checkpoint at 4 epochs +2023/06/05 01:52:58 - mmengine - INFO - Epoch(val) [4][57/57] accuracy/top1: 90.2509 data_time: 0.5514 time: 0.6400 +2023/06/05 01:54:07 - mmengine - INFO - Epoch(train) [5][ 100/3907] lr: 6.8565e-05 eta: 4:21:37 time: 0.7089 data_time: 0.5262 memory: 6319 loss: 0.0847 +2023/06/05 01:55:15 - mmengine - INFO - Epoch(train) [5][ 200/3907] lr: 6.8219e-05 eta: 4:20:31 time: 0.6575 data_time: 0.5166 memory: 6319 loss: 0.0910 +2023/06/05 01:56:22 - mmengine - INFO - Epoch(train) [5][ 300/3907] lr: 6.7873e-05 eta: 4:19:24 time: 0.6636 data_time: 0.5228 memory: 6319 loss: 0.1058 +2023/06/05 01:57:16 - mmengine - INFO - Exp name: resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1_20230604_225351 +2023/06/05 01:57:31 - mmengine - INFO - Epoch(train) [5][ 400/3907] lr: 6.7526e-05 eta: 4:18:19 time: 0.7193 data_time: 0.5789 memory: 6319 loss: 0.0884 +2023/06/05 01:58:37 - mmengine - INFO - Epoch(train) [5][ 500/3907] lr: 6.7178e-05 eta: 4:17:10 time: 0.6873 data_time: 0.5475 memory: 6319 loss: 0.0965 +2023/06/05 01:59:45 - mmengine - INFO - Epoch(train) [5][ 600/3907] lr: 6.6829e-05 eta: 4:16:03 time: 0.6608 data_time: 0.5068 memory: 6319 loss: 0.1029 +2023/06/05 02:00:53 - mmengine - INFO - Epoch(train) [5][ 700/3907] lr: 6.6480e-05 eta: 4:14:57 time: 0.6656 data_time: 0.5259 memory: 6319 loss: 0.0936 +2023/06/05 02:02:04 - mmengine - INFO - Epoch(train) [5][ 800/3907] lr: 6.6129e-05 eta: 4:13:55 time: 0.8690 data_time: 0.7285 memory: 6319 loss: 0.1000 +2023/06/05 02:03:15 - mmengine - INFO - Epoch(train) [5][ 900/3907] lr: 6.5778e-05 eta: 4:12:53 time: 0.6115 data_time: 0.4692 memory: 6319 loss: 0.0940 +2023/06/05 02:04:19 - mmengine - INFO - Epoch(train) [5][1000/3907] lr: 6.5427e-05 eta: 4:11:42 time: 0.6595 data_time: 0.5046 memory: 6319 loss: 0.0937 +2023/06/05 02:05:25 - mmengine - INFO - Epoch(train) [5][1100/3907] lr: 6.5074e-05 eta: 4:10:32 time: 0.6369 data_time: 0.4959 memory: 6319 loss: 0.0941 +2023/06/05 02:06:33 - mmengine - INFO - Epoch(train) [5][1200/3907] lr: 6.4721e-05 eta: 4:09:26 time: 0.6507 data_time: 0.4961 memory: 6319 loss: 0.1007 +2023/06/05 02:07:40 - mmengine - INFO - Epoch(train) [5][1300/3907] lr: 6.4368e-05 eta: 4:08:17 time: 0.6899 data_time: 0.5493 memory: 6319 loss: 0.0979 +2023/06/05 02:08:33 - mmengine - INFO - Exp name: resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1_20230604_225351 +2023/06/05 02:08:48 - mmengine - INFO - Epoch(train) [5][1400/3907] lr: 6.4014e-05 eta: 4:07:11 time: 0.6991 data_time: 0.5583 memory: 6319 loss: 0.0972 +2023/06/05 02:09:55 - mmengine - INFO - Epoch(train) [5][1500/3907] lr: 6.3659e-05 eta: 4:06:04 time: 0.7026 data_time: 0.5615 memory: 6319 loss: 0.0773 +2023/06/05 02:11:04 - mmengine - INFO - Epoch(train) [5][1600/3907] lr: 6.3303e-05 eta: 4:04:59 time: 0.6768 data_time: 0.5357 memory: 6319 loss: 0.0957 +2023/06/05 02:12:13 - mmengine - INFO - Epoch(train) [5][1700/3907] lr: 6.2948e-05 eta: 4:03:54 time: 0.6474 data_time: 0.5066 memory: 6319 loss: 0.0933 +2023/06/05 02:13:21 - mmengine - INFO - Epoch(train) [5][1800/3907] lr: 6.2591e-05 eta: 4:02:47 time: 0.6274 data_time: 0.4877 memory: 6319 loss: 0.1000 +2023/06/05 02:14:27 - mmengine - INFO - Epoch(train) [5][1900/3907] lr: 6.2234e-05 eta: 4:01:39 time: 0.6816 data_time: 0.5397 memory: 6319 loss: 0.0879 +2023/06/05 02:15:35 - mmengine - INFO - Epoch(train) [5][2000/3907] lr: 6.1877e-05 eta: 4:00:32 time: 0.6615 data_time: 0.5211 memory: 6319 loss: 0.0972 +2023/06/05 02:16:42 - mmengine - INFO - Epoch(train) [5][2100/3907] lr: 6.1519e-05 eta: 3:59:24 time: 0.6226 data_time: 0.4726 memory: 6319 loss: 0.0888 +2023/06/05 02:17:49 - mmengine - INFO - Epoch(train) [5][2200/3907] lr: 6.1161e-05 eta: 3:58:17 time: 0.6426 data_time: 0.5019 memory: 6319 loss: 0.0908 +2023/06/05 02:19:01 - mmengine - INFO - Epoch(train) [5][2300/3907] lr: 6.0802e-05 eta: 3:57:15 time: 0.7770 data_time: 0.6354 memory: 6319 loss: 0.1021 +2023/06/05 02:20:04 - mmengine - INFO - Exp name: resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1_20230604_225351 +2023/06/05 02:20:20 - mmengine - INFO - Epoch(train) [5][2400/3907] lr: 6.0443e-05 eta: 3:56:21 time: 0.7664 data_time: 0.6275 memory: 6319 loss: 0.0956 +2023/06/05 02:21:24 - mmengine - INFO - Epoch(train) [5][2500/3907] lr: 6.0084e-05 eta: 3:55:10 time: 0.6135 data_time: 0.4720 memory: 6319 loss: 0.0871 +2023/06/05 02:22:27 - mmengine - INFO - Epoch(train) [5][2600/3907] lr: 5.9724e-05 eta: 3:53:58 time: 0.6515 data_time: 0.5098 memory: 6319 loss: 0.0901 +2023/06/05 02:23:34 - mmengine - INFO - Epoch(train) [5][2700/3907] lr: 5.9364e-05 eta: 3:52:49 time: 0.6207 data_time: 0.4803 memory: 6319 loss: 0.0891 +2023/06/05 02:24:39 - mmengine - INFO - Epoch(train) [5][2800/3907] lr: 5.9004e-05 eta: 3:51:40 time: 0.6623 data_time: 0.5215 memory: 6319 loss: 0.1049 +2023/06/05 02:25:46 - mmengine - INFO - Epoch(train) [5][2900/3907] lr: 5.8643e-05 eta: 3:50:32 time: 0.6878 data_time: 0.5449 memory: 6319 loss: 0.0943 +2023/06/05 02:26:53 - mmengine - INFO - Epoch(train) [5][3000/3907] lr: 5.8283e-05 eta: 3:49:25 time: 0.7121 data_time: 0.5710 memory: 6319 loss: 0.0935 +2023/06/05 02:28:01 - mmengine - INFO - Epoch(train) [5][3100/3907] lr: 5.7922e-05 eta: 3:48:18 time: 0.6972 data_time: 0.5558 memory: 6319 loss: 0.0996 +2023/06/05 02:29:09 - mmengine - INFO - Epoch(train) [5][3200/3907] lr: 5.7560e-05 eta: 3:47:10 time: 0.6685 data_time: 0.5280 memory: 6319 loss: 0.1042 +2023/06/05 02:30:16 - mmengine - INFO - Epoch(train) [5][3300/3907] lr: 5.7199e-05 eta: 3:46:03 time: 0.6613 data_time: 0.5208 memory: 6319 loss: 0.1008 +2023/06/05 02:31:09 - mmengine - INFO - Exp name: resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1_20230604_225351 +2023/06/05 02:31:24 - mmengine - INFO - Epoch(train) [5][3400/3907] lr: 5.6838e-05 eta: 3:44:56 time: 0.6708 data_time: 0.5302 memory: 6319 loss: 0.0858 +2023/06/05 02:32:31 - mmengine - INFO - Epoch(train) [5][3500/3907] lr: 5.6476e-05 eta: 3:43:48 time: 0.6429 data_time: 0.5027 memory: 6319 loss: 0.1171 +2023/06/05 02:33:37 - mmengine - INFO - Epoch(train) [5][3600/3907] lr: 5.6114e-05 eta: 3:42:40 time: 0.6569 data_time: 0.5160 memory: 6319 loss: 0.0920 +2023/06/05 02:34:45 - mmengine - INFO - Epoch(train) [5][3700/3907] lr: 5.5753e-05 eta: 3:41:33 time: 0.6830 data_time: 0.5427 memory: 6319 loss: 0.1029 +2023/06/05 02:35:52 - mmengine - INFO - Epoch(train) [5][3800/3907] lr: 5.5391e-05 eta: 3:40:25 time: 0.6381 data_time: 0.4972 memory: 6319 loss: 0.0870 +2023/06/05 02:36:59 - mmengine - INFO - Epoch(train) [5][3900/3907] lr: 5.5029e-05 eta: 3:39:18 time: 0.6577 data_time: 0.5175 memory: 6319 loss: 0.1026 +2023/06/05 02:37:04 - mmengine - INFO - Exp name: resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1_20230604_225351 +2023/06/05 02:37:04 - mmengine - INFO - Saving checkpoint at 5 epochs +2023/06/05 02:37:59 - mmengine - INFO - Epoch(val) [5][57/57] accuracy/top1: 93.7302 data_time: 0.6467 time: 0.7327 +2023/06/05 02:39:32 - mmengine - INFO - Epoch(train) [6][ 100/3907] lr: 5.4642e-05 eta: 3:38:32 time: 1.0088 data_time: 0.0313 memory: 6319 loss: 0.1072 +2023/06/05 02:40:54 - mmengine - INFO - Epoch(train) [6][ 200/3907] lr: 5.4280e-05 eta: 3:37:39 time: 0.5342 data_time: 0.1992 memory: 6319 loss: 0.1047 +2023/06/05 02:41:53 - mmengine - INFO - Epoch(train) [6][ 300/3907] lr: 5.3918e-05 eta: 3:36:23 time: 0.6258 data_time: 0.4796 memory: 6319 loss: 0.0882 +2023/06/05 02:42:59 - mmengine - INFO - Epoch(train) [6][ 400/3907] lr: 5.3556e-05 eta: 3:35:14 time: 0.6749 data_time: 0.4100 memory: 6319 loss: 0.0793 +2023/06/05 02:43:42 - mmengine - INFO - Exp name: resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1_20230604_225351 +2023/06/05 02:44:04 - mmengine - INFO - Epoch(train) [6][ 500/3907] lr: 5.3195e-05 eta: 3:34:04 time: 0.6482 data_time: 0.3807 memory: 6319 loss: 0.0931 +2023/06/05 02:45:10 - mmengine - INFO - Epoch(train) [6][ 600/3907] lr: 5.2833e-05 eta: 3:32:56 time: 0.7213 data_time: 0.5792 memory: 6319 loss: 0.0903 +2023/06/05 02:46:19 - mmengine - INFO - Epoch(train) [6][ 700/3907] lr: 5.2472e-05 eta: 3:31:50 time: 0.8090 data_time: 0.6684 memory: 6319 loss: 0.0896 +2023/06/05 02:47:26 - mmengine - INFO - Epoch(train) [6][ 800/3907] lr: 5.2111e-05 eta: 3:30:42 time: 0.6541 data_time: 0.5131 memory: 6319 loss: 0.0865 +2023/06/05 02:48:33 - mmengine - INFO - Epoch(train) [6][ 900/3907] lr: 5.1750e-05 eta: 3:29:34 time: 0.6661 data_time: 0.5252 memory: 6319 loss: 0.0908 +2023/06/05 02:49:40 - mmengine - INFO - Epoch(train) [6][1000/3907] lr: 5.1389e-05 eta: 3:28:26 time: 0.6543 data_time: 0.5123 memory: 6319 loss: 0.0951 +2023/06/05 02:50:47 - mmengine - INFO - Epoch(train) [6][1100/3907] lr: 5.1029e-05 eta: 3:27:18 time: 0.6567 data_time: 0.5156 memory: 6319 loss: 0.1042 +2023/06/05 02:51:54 - mmengine - INFO - Epoch(train) [6][1200/3907] lr: 5.0668e-05 eta: 3:26:10 time: 0.6560 data_time: 0.5148 memory: 6319 loss: 0.0892 +2023/06/05 02:53:01 - mmengine - INFO - Epoch(train) [6][1300/3907] lr: 5.0308e-05 eta: 3:25:02 time: 0.6526 data_time: 0.5123 memory: 6319 loss: 0.0874 +2023/06/05 02:54:07 - mmengine - INFO - Epoch(train) [6][1400/3907] lr: 4.9949e-05 eta: 3:23:53 time: 0.6498 data_time: 0.5093 memory: 6319 loss: 0.0833 +2023/06/05 02:54:53 - mmengine - INFO - Exp name: resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1_20230604_225351 +2023/06/05 02:55:13 - mmengine - INFO - Epoch(train) [6][1500/3907] lr: 4.9589e-05 eta: 3:22:45 time: 0.6572 data_time: 0.5166 memory: 6319 loss: 0.0813 +2023/06/05 02:56:19 - mmengine - INFO - Epoch(train) [6][1600/3907] lr: 4.9230e-05 eta: 3:21:36 time: 0.6470 data_time: 0.5068 memory: 6319 loss: 0.1099 +2023/06/05 02:57:26 - mmengine - INFO - Epoch(train) [6][1700/3907] lr: 4.8871e-05 eta: 3:20:28 time: 0.7002 data_time: 0.5566 memory: 6319 loss: 0.0949 +2023/06/05 02:58:33 - mmengine - INFO - Epoch(train) [6][1800/3907] lr: 4.8513e-05 eta: 3:19:20 time: 0.6985 data_time: 0.5592 memory: 6319 loss: 0.0785 +2023/06/05 02:59:39 - mmengine - INFO - Epoch(train) [6][1900/3907] lr: 4.8155e-05 eta: 3:18:11 time: 0.6720 data_time: 0.5205 memory: 6319 loss: 0.0885 +2023/06/05 03:00:44 - mmengine - INFO - Epoch(train) [6][2000/3907] lr: 4.7798e-05 eta: 3:17:02 time: 0.6435 data_time: 0.5039 memory: 6319 loss: 0.0847 +2023/06/05 03:01:51 - mmengine - INFO - Epoch(train) [6][2100/3907] lr: 4.7441e-05 eta: 3:15:55 time: 0.6574 data_time: 0.5169 memory: 6319 loss: 0.0869 +2023/06/05 03:02:57 - mmengine - INFO - Epoch(train) [6][2200/3907] lr: 4.7084e-05 eta: 3:14:46 time: 0.6593 data_time: 0.5160 memory: 6319 loss: 0.0849 +2023/06/05 03:04:05 - mmengine - INFO - Epoch(train) [6][2300/3907] lr: 4.6729e-05 eta: 3:13:39 time: 0.6877 data_time: 0.5460 memory: 6319 loss: 0.0828 +2023/06/05 03:05:11 - mmengine - INFO - Epoch(train) [6][2400/3907] lr: 4.6373e-05 eta: 3:12:30 time: 0.6736 data_time: 0.5129 memory: 6319 loss: 0.0953 +2023/06/05 03:05:57 - mmengine - INFO - Exp name: resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1_20230604_225351 +2023/06/05 03:06:17 - mmengine - INFO - Epoch(train) [6][2500/3907] lr: 4.6018e-05 eta: 3:11:22 time: 0.6585 data_time: 0.5153 memory: 6319 loss: 0.0898 +2023/06/05 03:07:22 - mmengine - INFO - Epoch(train) [6][2600/3907] lr: 4.5664e-05 eta: 3:10:13 time: 0.6338 data_time: 0.4810 memory: 6319 loss: 0.1028 +2023/06/05 03:08:29 - mmengine - INFO - Epoch(train) [6][2700/3907] lr: 4.5310e-05 eta: 3:09:05 time: 0.6423 data_time: 0.5007 memory: 6319 loss: 0.0992 +2023/06/05 03:09:35 - mmengine - INFO - Epoch(train) [6][2800/3907] lr: 4.4957e-05 eta: 3:07:57 time: 0.6834 data_time: 0.5392 memory: 6319 loss: 0.0924 +2023/06/05 03:10:42 - mmengine - INFO - Epoch(train) [6][2900/3907] lr: 4.4605e-05 eta: 3:06:49 time: 0.6495 data_time: 0.5094 memory: 6319 loss: 0.0798 +2023/06/05 03:11:50 - mmengine - INFO - Epoch(train) [6][3000/3907] lr: 4.4253e-05 eta: 3:05:42 time: 0.6808 data_time: 0.5398 memory: 6319 loss: 0.0837 +2023/06/05 03:12:56 - mmengine - INFO - Epoch(train) [6][3100/3907] lr: 4.3902e-05 eta: 3:04:33 time: 0.6207 data_time: 0.4800 memory: 6319 loss: 0.0891 +2023/06/05 03:14:03 - mmengine - INFO - Epoch(train) [6][3200/3907] lr: 4.3552e-05 eta: 3:03:26 time: 0.6707 data_time: 0.5293 memory: 6319 loss: 0.0951 +2023/06/05 03:15:11 - mmengine - INFO - Epoch(train) [6][3300/3907] lr: 4.3202e-05 eta: 3:02:19 time: 0.6986 data_time: 0.5575 memory: 6319 loss: 0.0866 +2023/06/05 03:16:16 - mmengine - INFO - Epoch(train) [6][3400/3907] lr: 4.2854e-05 eta: 3:01:10 time: 0.6644 data_time: 0.5238 memory: 6319 loss: 0.0825 +2023/06/05 03:17:02 - mmengine - INFO - Exp name: resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1_20230604_225351 +2023/06/05 03:17:23 - mmengine - INFO - Epoch(train) [6][3500/3907] lr: 4.2506e-05 eta: 3:00:02 time: 0.6671 data_time: 0.5263 memory: 6319 loss: 0.0943 +2023/06/05 03:18:31 - mmengine - INFO - Epoch(train) [6][3600/3907] lr: 4.2158e-05 eta: 2:58:55 time: 0.6680 data_time: 0.5269 memory: 6319 loss: 0.1002 +2023/06/05 03:19:38 - mmengine - INFO - Epoch(train) [6][3700/3907] lr: 4.1812e-05 eta: 2:57:47 time: 0.6261 data_time: 0.4859 memory: 6319 loss: 0.0942 +2023/06/05 03:20:45 - mmengine - INFO - Epoch(train) [6][3800/3907] lr: 4.1466e-05 eta: 2:56:40 time: 0.6174 data_time: 0.4777 memory: 6319 loss: 0.0874 +2023/06/05 03:21:51 - mmengine - INFO - Epoch(train) [6][3900/3907] lr: 4.1122e-05 eta: 2:55:32 time: 0.6479 data_time: 0.5053 memory: 6319 loss: 0.0733 +2023/06/05 03:21:56 - mmengine - INFO - Exp name: resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1_20230604_225351 +2023/06/05 03:21:56 - mmengine - INFO - Saving checkpoint at 6 epochs +2023/06/05 03:22:45 - mmengine - INFO - Epoch(val) [6][57/57] accuracy/top1: 95.1428 data_time: 0.5447 time: 0.6323 +2023/06/05 03:23:54 - mmengine - INFO - Epoch(train) [7][ 100/3907] lr: 4.0754e-05 eta: 2:54:21 time: 0.6833 data_time: 0.4709 memory: 6319 loss: 0.0830 +2023/06/05 03:25:01 - mmengine - INFO - Epoch(train) [7][ 200/3907] lr: 4.0411e-05 eta: 2:53:13 time: 0.6511 data_time: 0.4820 memory: 6319 loss: 0.0863 +2023/06/05 03:26:06 - mmengine - INFO - Epoch(train) [7][ 300/3907] lr: 4.0069e-05 eta: 2:52:05 time: 0.6624 data_time: 0.5034 memory: 6319 loss: 0.0899 +2023/06/05 03:27:13 - mmengine - INFO - Epoch(train) [7][ 400/3907] lr: 3.9729e-05 eta: 2:50:57 time: 0.6492 data_time: 0.3304 memory: 6319 loss: 0.0917 +2023/06/05 03:28:20 - mmengine - INFO - Epoch(train) [7][ 500/3907] lr: 3.9389e-05 eta: 2:49:50 time: 0.6958 data_time: 0.2444 memory: 6319 loss: 0.0931 +2023/06/05 03:29:00 - mmengine - INFO - Exp name: resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1_20230604_225351 +2023/06/05 03:29:27 - mmengine - INFO - Epoch(train) [7][ 600/3907] lr: 3.9050e-05 eta: 2:48:42 time: 0.6798 data_time: 0.3008 memory: 6319 loss: 0.0904 +2023/06/05 03:30:35 - mmengine - INFO - Epoch(train) [7][ 700/3907] lr: 3.8712e-05 eta: 2:47:35 time: 0.6649 data_time: 0.1281 memory: 6319 loss: 0.0933 +2023/06/05 03:31:41 - mmengine - INFO - Epoch(train) [7][ 800/3907] lr: 3.8375e-05 eta: 2:46:27 time: 0.6324 data_time: 0.2489 memory: 6319 loss: 0.0722 +2023/06/05 03:32:48 - mmengine - INFO - Epoch(train) [7][ 900/3907] lr: 3.8039e-05 eta: 2:45:19 time: 0.6473 data_time: 0.3247 memory: 6319 loss: 0.0875 +2023/06/05 03:33:53 - mmengine - INFO - Epoch(train) [7][1000/3907] lr: 3.7705e-05 eta: 2:44:11 time: 0.6407 data_time: 0.2841 memory: 6319 loss: 0.1027 +2023/06/05 03:35:00 - mmengine - INFO - Epoch(train) [7][1100/3907] lr: 3.7371e-05 eta: 2:43:03 time: 0.7046 data_time: 0.3224 memory: 6319 loss: 0.0939 +2023/06/05 03:36:05 - mmengine - INFO - Epoch(train) [7][1200/3907] lr: 3.7039e-05 eta: 2:41:54 time: 0.6629 data_time: 0.3817 memory: 6319 loss: 0.0884 +2023/06/05 03:37:11 - mmengine - INFO - Epoch(train) [7][1300/3907] lr: 3.6708e-05 eta: 2:40:46 time: 0.6343 data_time: 0.3548 memory: 6319 loss: 0.0907 +2023/06/05 03:38:17 - mmengine - INFO - Epoch(train) [7][1400/3907] lr: 3.6378e-05 eta: 2:39:38 time: 0.6865 data_time: 0.5104 memory: 6319 loss: 0.0993 +2023/06/05 03:39:24 - mmengine - INFO - Epoch(train) [7][1500/3907] lr: 3.6049e-05 eta: 2:38:31 time: 0.6687 data_time: 0.5267 memory: 6319 loss: 0.0951 +2023/06/05 03:40:04 - mmengine - INFO - Exp name: resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1_20230604_225351 +2023/06/05 03:40:31 - mmengine - INFO - Epoch(train) [7][1600/3907] lr: 3.5721e-05 eta: 2:37:23 time: 0.6518 data_time: 0.5122 memory: 6319 loss: 0.0814 +2023/06/05 03:41:38 - mmengine - INFO - Epoch(train) [7][1700/3907] lr: 3.5395e-05 eta: 2:36:15 time: 0.6637 data_time: 0.5005 memory: 6319 loss: 0.0833 +2023/06/05 03:42:45 - mmengine - INFO - Epoch(train) [7][1800/3907] lr: 3.5070e-05 eta: 2:35:08 time: 0.6965 data_time: 0.4331 memory: 6319 loss: 0.0814 +2023/06/05 03:43:53 - mmengine - INFO - Epoch(train) [7][1900/3907] lr: 3.4746e-05 eta: 2:34:01 time: 0.6273 data_time: 0.2228 memory: 6319 loss: 0.0851 +2023/06/05 03:44:59 - mmengine - INFO - Epoch(train) [7][2000/3907] lr: 3.4424e-05 eta: 2:32:53 time: 0.6682 data_time: 0.3571 memory: 6319 loss: 0.0795 +2023/06/05 03:46:05 - mmengine - INFO - Epoch(train) [7][2100/3907] lr: 3.4103e-05 eta: 2:31:45 time: 0.6555 data_time: 0.2025 memory: 6319 loss: 0.0941 +2023/06/05 03:47:10 - mmengine - INFO - Epoch(train) [7][2200/3907] lr: 3.3783e-05 eta: 2:30:36 time: 0.6488 data_time: 0.1938 memory: 6319 loss: 0.0736 +2023/06/05 03:48:17 - mmengine - INFO - Epoch(train) [7][2300/3907] lr: 3.3465e-05 eta: 2:29:29 time: 0.6764 data_time: 0.0210 memory: 6319 loss: 0.0829 +2023/06/05 03:49:23 - mmengine - INFO - Epoch(train) [7][2400/3907] lr: 3.3148e-05 eta: 2:28:21 time: 0.6985 data_time: 0.0012 memory: 6319 loss: 0.0913 +2023/06/05 03:50:30 - mmengine - INFO - Epoch(train) [7][2500/3907] lr: 3.2832e-05 eta: 2:27:13 time: 0.6535 data_time: 0.0012 memory: 6319 loss: 0.0767 +2023/06/05 03:51:10 - mmengine - INFO - Exp name: resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1_20230604_225351 +2023/06/05 03:51:37 - mmengine - INFO - Epoch(train) [7][2600/3907] lr: 3.2518e-05 eta: 2:26:06 time: 0.6932 data_time: 0.0014 memory: 6319 loss: 0.0775 +2023/06/05 03:52:42 - mmengine - INFO - Epoch(train) [7][2700/3907] lr: 3.2205e-05 eta: 2:24:57 time: 0.6336 data_time: 0.0013 memory: 6319 loss: 0.0764 +2023/06/05 03:53:46 - mmengine - INFO - Epoch(train) [7][2800/3907] lr: 3.1894e-05 eta: 2:23:49 time: 0.6317 data_time: 0.1095 memory: 6319 loss: 0.0750 +2023/06/05 03:54:52 - mmengine - INFO - Epoch(train) [7][2900/3907] lr: 3.1584e-05 eta: 2:22:41 time: 0.6450 data_time: 0.2127 memory: 6319 loss: 0.0994 +2023/06/05 03:55:58 - mmengine - INFO - Epoch(train) [7][3000/3907] lr: 3.1276e-05 eta: 2:21:33 time: 0.6719 data_time: 0.2100 memory: 6319 loss: 0.0889 +2023/06/05 03:57:03 - mmengine - INFO - Epoch(train) [7][3100/3907] lr: 3.0969e-05 eta: 2:20:25 time: 0.6451 data_time: 0.2017 memory: 6319 loss: 0.0774 +2023/06/05 03:58:10 - mmengine - INFO - Epoch(train) [7][3200/3907] lr: 3.0664e-05 eta: 2:19:17 time: 0.6313 data_time: 0.0578 memory: 6319 loss: 0.0933 +2023/06/05 03:59:16 - mmengine - INFO - Epoch(train) [7][3300/3907] lr: 3.0360e-05 eta: 2:18:09 time: 0.6147 data_time: 0.0367 memory: 6319 loss: 0.0919 +2023/06/05 04:00:23 - mmengine - INFO - Epoch(train) [7][3400/3907] lr: 3.0058e-05 eta: 2:17:02 time: 0.6930 data_time: 0.0251 memory: 6319 loss: 0.0838 +2023/06/05 04:01:29 - mmengine - INFO - Epoch(train) [7][3500/3907] lr: 2.9758e-05 eta: 2:15:54 time: 0.6763 data_time: 0.0023 memory: 6319 loss: 0.0886 +2023/06/05 04:02:08 - mmengine - INFO - Exp name: resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1_20230604_225351 +2023/06/05 04:02:35 - mmengine - INFO - Epoch(train) [7][3600/3907] lr: 2.9459e-05 eta: 2:14:46 time: 0.6263 data_time: 0.0017 memory: 6319 loss: 0.0850 +2023/06/05 04:03:43 - mmengine - INFO - Epoch(train) [7][3700/3907] lr: 2.9162e-05 eta: 2:13:40 time: 0.6992 data_time: 0.0011 memory: 6319 loss: 0.0823 +2023/06/05 04:04:49 - mmengine - INFO - Epoch(train) [7][3800/3907] lr: 2.8867e-05 eta: 2:12:32 time: 0.6734 data_time: 0.0016 memory: 6319 loss: 0.0807 +2023/06/05 04:05:53 - mmengine - INFO - Epoch(train) [7][3900/3907] lr: 2.8573e-05 eta: 2:11:23 time: 0.6396 data_time: 0.0008 memory: 6319 loss: 0.0829 +2023/06/05 04:05:58 - mmengine - INFO - Exp name: resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1_20230604_225351 +2023/06/05 04:05:58 - mmengine - INFO - Saving checkpoint at 7 epochs +2023/06/05 04:06:47 - mmengine - INFO - Epoch(val) [7][57/57] accuracy/top1: 95.8770 data_time: 0.5131 time: 0.6030 +2023/06/05 04:07:54 - mmengine - INFO - Epoch(train) [8][ 100/3907] lr: 2.8261e-05 eta: 2:10:11 time: 0.6539 data_time: 0.5128 memory: 6319 loss: 0.0746 +2023/06/05 04:09:00 - mmengine - INFO - Epoch(train) [8][ 200/3907] lr: 2.7971e-05 eta: 2:09:04 time: 0.6469 data_time: 0.4983 memory: 6319 loss: 0.0821 +2023/06/05 04:10:07 - mmengine - INFO - Epoch(train) [8][ 300/3907] lr: 2.7682e-05 eta: 2:07:57 time: 0.6515 data_time: 0.5047 memory: 6319 loss: 0.0866 +2023/06/05 04:11:13 - mmengine - INFO - Epoch(train) [8][ 400/3907] lr: 2.7395e-05 eta: 2:06:49 time: 0.6198 data_time: 0.4775 memory: 6319 loss: 0.1056 +2023/06/05 04:12:19 - mmengine - INFO - Epoch(train) [8][ 500/3907] lr: 2.7111e-05 eta: 2:05:41 time: 0.6706 data_time: 0.5294 memory: 6319 loss: 0.0904 +2023/06/05 04:13:25 - mmengine - INFO - Epoch(train) [8][ 600/3907] lr: 2.6828e-05 eta: 2:04:33 time: 0.6807 data_time: 0.5398 memory: 6319 loss: 0.0981 +2023/06/05 04:13:59 - mmengine - INFO - Exp name: resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1_20230604_225351 +2023/06/05 04:14:33 - mmengine - INFO - Epoch(train) [8][ 700/3907] lr: 2.6546e-05 eta: 2:03:26 time: 0.6687 data_time: 0.4495 memory: 6319 loss: 0.0908 +2023/06/05 04:15:39 - mmengine - INFO - Epoch(train) [8][ 800/3907] lr: 2.6267e-05 eta: 2:02:19 time: 0.6814 data_time: 0.5386 memory: 6319 loss: 0.0912 +2023/06/05 04:16:45 - mmengine - INFO - Epoch(train) [8][ 900/3907] lr: 2.5989e-05 eta: 2:01:11 time: 0.6589 data_time: 0.5184 memory: 6319 loss: 0.1057 +2023/06/05 04:17:51 - mmengine - INFO - Epoch(train) [8][1000/3907] lr: 2.5714e-05 eta: 2:00:04 time: 0.6964 data_time: 0.5535 memory: 6319 loss: 0.1015 +2023/06/05 04:18:59 - mmengine - INFO - Epoch(train) [8][1100/3907] lr: 2.5440e-05 eta: 1:58:56 time: 0.6736 data_time: 0.5336 memory: 6319 loss: 0.0967 +2023/06/05 04:20:05 - mmengine - INFO - Epoch(train) [8][1200/3907] lr: 2.5168e-05 eta: 1:57:49 time: 0.6763 data_time: 0.5342 memory: 6319 loss: 0.0763 +2023/06/05 04:21:12 - mmengine - INFO - Epoch(train) [8][1300/3907] lr: 2.4898e-05 eta: 1:56:42 time: 0.6497 data_time: 0.5070 memory: 6319 loss: 0.0863 +2023/06/05 04:22:18 - mmengine - INFO - Epoch(train) [8][1400/3907] lr: 2.4630e-05 eta: 1:55:34 time: 0.6296 data_time: 0.4846 memory: 6319 loss: 0.0780 +2023/06/05 04:23:24 - mmengine - INFO - Epoch(train) [8][1500/3907] lr: 2.4364e-05 eta: 1:54:26 time: 0.6721 data_time: 0.5297 memory: 6319 loss: 0.0711 +2023/06/05 04:24:31 - mmengine - INFO - Epoch(train) [8][1600/3907] lr: 2.4100e-05 eta: 1:53:19 time: 0.6428 data_time: 0.5019 memory: 6319 loss: 0.0866 +2023/06/05 04:25:04 - mmengine - INFO - Exp name: resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1_20230604_225351 +2023/06/05 04:25:36 - mmengine - INFO - Epoch(train) [8][1700/3907] lr: 2.3838e-05 eta: 1:52:11 time: 0.6370 data_time: 0.4974 memory: 6319 loss: 0.0868 +2023/06/05 04:26:42 - mmengine - INFO - Epoch(train) [8][1800/3907] lr: 2.3578e-05 eta: 1:51:04 time: 0.6740 data_time: 0.5338 memory: 6319 loss: 0.0866 +2023/06/05 04:27:49 - mmengine - INFO - Epoch(train) [8][1900/3907] lr: 2.3320e-05 eta: 1:49:56 time: 0.6760 data_time: 0.5322 memory: 6319 loss: 0.0691 +2023/06/05 04:28:55 - mmengine - INFO - Epoch(train) [8][2000/3907] lr: 2.3064e-05 eta: 1:48:49 time: 0.6792 data_time: 0.5385 memory: 6319 loss: 0.0872 +2023/06/05 04:30:00 - mmengine - INFO - Epoch(train) [8][2100/3907] lr: 2.2810e-05 eta: 1:47:41 time: 0.6652 data_time: 0.5235 memory: 6319 loss: 0.0788 +2023/06/05 04:31:08 - mmengine - INFO - Epoch(train) [8][2200/3907] lr: 2.2558e-05 eta: 1:46:34 time: 0.6289 data_time: 0.4888 memory: 6319 loss: 0.0879 +2023/06/05 04:32:13 - mmengine - INFO - Epoch(train) [8][2300/3907] lr: 2.2309e-05 eta: 1:45:26 time: 0.6296 data_time: 0.4862 memory: 6319 loss: 0.0855 +2023/06/05 04:33:19 - mmengine - INFO - Epoch(train) [8][2400/3907] lr: 2.2061e-05 eta: 1:44:19 time: 0.6510 data_time: 0.5056 memory: 6319 loss: 0.0799 +2023/06/05 04:34:24 - mmengine - INFO - Epoch(train) [8][2500/3907] lr: 2.1816e-05 eta: 1:43:11 time: 0.6673 data_time: 0.5047 memory: 6319 loss: 0.0879 +2023/06/05 04:35:30 - mmengine - INFO - Epoch(train) [8][2600/3907] lr: 2.1572e-05 eta: 1:42:03 time: 0.6555 data_time: 0.5138 memory: 6319 loss: 0.0794 +2023/06/05 04:36:03 - mmengine - INFO - Exp name: resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1_20230604_225351 +2023/06/05 04:36:35 - mmengine - INFO - Epoch(train) [8][2700/3907] lr: 2.1331e-05 eta: 1:40:56 time: 0.6874 data_time: 0.5457 memory: 6319 loss: 0.0835 +2023/06/05 04:37:41 - mmengine - INFO - Epoch(train) [8][2800/3907] lr: 2.1092e-05 eta: 1:39:48 time: 0.6198 data_time: 0.4772 memory: 6319 loss: 0.0770 +2023/06/05 04:38:46 - mmengine - INFO - Epoch(train) [8][2900/3907] lr: 2.0855e-05 eta: 1:38:40 time: 0.6416 data_time: 0.4771 memory: 6319 loss: 0.0769 +2023/06/05 04:39:50 - mmengine - INFO - Epoch(train) [8][3000/3907] lr: 2.0621e-05 eta: 1:37:32 time: 0.5951 data_time: 0.4539 memory: 6319 loss: 0.0880 +2023/06/05 04:40:56 - mmengine - INFO - Epoch(train) [8][3100/3907] lr: 2.0388e-05 eta: 1:36:25 time: 0.6554 data_time: 0.5135 memory: 6319 loss: 0.0979 +2023/06/05 04:42:01 - mmengine - INFO - Epoch(train) [8][3200/3907] lr: 2.0158e-05 eta: 1:35:17 time: 0.6210 data_time: 0.4787 memory: 6319 loss: 0.0819 +2023/06/05 04:43:07 - mmengine - INFO - Epoch(train) [8][3300/3907] lr: 1.9930e-05 eta: 1:34:10 time: 0.6211 data_time: 0.4796 memory: 6319 loss: 0.0671 +2023/06/05 04:44:13 - mmengine - INFO - Epoch(train) [8][3400/3907] lr: 1.9705e-05 eta: 1:33:02 time: 0.7155 data_time: 0.5479 memory: 6319 loss: 0.0779 +2023/06/05 04:45:18 - mmengine - INFO - Epoch(train) [8][3500/3907] lr: 1.9481e-05 eta: 1:31:55 time: 0.6287 data_time: 0.4846 memory: 6319 loss: 0.0861 +2023/06/05 04:46:23 - mmengine - INFO - Epoch(train) [8][3600/3907] lr: 1.9260e-05 eta: 1:30:47 time: 0.6590 data_time: 0.5063 memory: 6319 loss: 0.0887 +2023/06/05 04:46:56 - mmengine - INFO - Exp name: resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1_20230604_225351 +2023/06/05 04:47:28 - mmengine - INFO - Epoch(train) [8][3700/3907] lr: 1.9042e-05 eta: 1:29:40 time: 0.6220 data_time: 0.4814 memory: 6319 loss: 0.0826 +2023/06/05 04:48:35 - mmengine - INFO - Epoch(train) [8][3800/3907] lr: 1.8825e-05 eta: 1:28:32 time: 0.6637 data_time: 0.5231 memory: 6319 loss: 0.0693 +2023/06/05 04:49:40 - mmengine - INFO - Epoch(train) [8][3900/3907] lr: 1.8611e-05 eta: 1:27:25 time: 0.6353 data_time: 0.4940 memory: 6319 loss: 0.0868 +2023/06/05 04:49:46 - mmengine - INFO - Exp name: resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1_20230604_225351 +2023/06/05 04:49:46 - mmengine - INFO - Saving checkpoint at 8 epochs +2023/06/05 04:50:35 - mmengine - INFO - Epoch(val) [8][57/57] accuracy/top1: 96.5067 data_time: 0.5388 time: 0.6272 +2023/06/05 04:51:42 - mmengine - INFO - Epoch(train) [9][ 100/3907] lr: 1.8385e-05 eta: 1:26:13 time: 0.5908 data_time: 0.4504 memory: 6319 loss: 0.0774 +2023/06/05 04:52:48 - mmengine - INFO - Epoch(train) [9][ 200/3907] lr: 1.8176e-05 eta: 1:25:06 time: 0.6101 data_time: 0.4683 memory: 6319 loss: 0.0846 +2023/06/05 04:53:53 - mmengine - INFO - Epoch(train) [9][ 300/3907] lr: 1.7969e-05 eta: 1:23:58 time: 0.6208 data_time: 0.4809 memory: 6319 loss: 0.0809 +2023/06/05 04:54:58 - mmengine - INFO - Epoch(train) [9][ 400/3907] lr: 1.7765e-05 eta: 1:22:51 time: 0.6318 data_time: 0.4907 memory: 6319 loss: 0.0857 +2023/06/05 04:56:03 - mmengine - INFO - Epoch(train) [9][ 500/3907] lr: 1.7563e-05 eta: 1:21:43 time: 0.6543 data_time: 0.5134 memory: 6319 loss: 0.0786 +2023/06/05 04:57:09 - mmengine - INFO - Epoch(train) [9][ 600/3907] lr: 1.7363e-05 eta: 1:20:36 time: 0.6742 data_time: 0.4887 memory: 6319 loss: 0.0888 +2023/06/05 04:58:14 - mmengine - INFO - Epoch(train) [9][ 700/3907] lr: 1.7166e-05 eta: 1:19:29 time: 0.6330 data_time: 0.4915 memory: 6319 loss: 0.0832 +2023/06/05 04:58:46 - mmengine - INFO - Exp name: resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1_20230604_225351 +2023/06/05 04:59:20 - mmengine - INFO - Epoch(train) [9][ 800/3907] lr: 1.6971e-05 eta: 1:18:21 time: 0.6416 data_time: 0.5001 memory: 6319 loss: 0.0735 +2023/06/05 05:00:25 - mmengine - INFO - Epoch(train) [9][ 900/3907] lr: 1.6779e-05 eta: 1:17:14 time: 0.6127 data_time: 0.4580 memory: 6319 loss: 0.0807 +2023/06/05 05:01:30 - mmengine - INFO - Epoch(train) [9][1000/3907] lr: 1.6589e-05 eta: 1:16:06 time: 0.6638 data_time: 0.5103 memory: 6319 loss: 0.0787 +2023/06/05 05:02:36 - mmengine - INFO - Epoch(train) [9][1100/3907] lr: 1.6402e-05 eta: 1:14:59 time: 0.7057 data_time: 0.4995 memory: 6319 loss: 0.0781 +2023/06/05 05:03:42 - mmengine - INFO - Epoch(train) [9][1200/3907] lr: 1.6217e-05 eta: 1:13:52 time: 0.7084 data_time: 0.4486 memory: 6319 loss: 0.0836 +2023/06/05 05:04:47 - mmengine - INFO - Epoch(train) [9][1300/3907] lr: 1.6035e-05 eta: 1:12:44 time: 0.7299 data_time: 0.5895 memory: 6319 loss: 0.0897 +2023/06/05 05:05:51 - mmengine - INFO - Epoch(train) [9][1400/3907] lr: 1.5855e-05 eta: 1:11:37 time: 0.6415 data_time: 0.5001 memory: 6319 loss: 0.0734 +2023/06/05 05:06:59 - mmengine - INFO - Epoch(train) [9][1500/3907] lr: 1.5678e-05 eta: 1:10:30 time: 0.7241 data_time: 0.5831 memory: 6319 loss: 0.0838 +2023/06/05 05:08:06 - mmengine - INFO - Epoch(train) [9][1600/3907] lr: 1.5503e-05 eta: 1:09:23 time: 0.6703 data_time: 0.5291 memory: 6319 loss: 0.0905 +2023/06/05 05:09:13 - mmengine - INFO - Epoch(train) [9][1700/3907] lr: 1.5331e-05 eta: 1:08:16 time: 0.6597 data_time: 0.5190 memory: 6319 loss: 0.0885 +2023/06/05 05:09:45 - mmengine - INFO - Exp name: resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1_20230604_225351 +2023/06/05 05:10:20 - mmengine - INFO - Epoch(train) [9][1800/3907] lr: 1.5162e-05 eta: 1:07:09 time: 0.6580 data_time: 0.5171 memory: 6319 loss: 0.0948 +2023/06/05 05:11:25 - mmengine - INFO - Epoch(train) [9][1900/3907] lr: 1.4995e-05 eta: 1:06:02 time: 0.6293 data_time: 0.4896 memory: 6319 loss: 0.0805 +2023/06/05 05:12:30 - mmengine - INFO - Epoch(train) [9][2000/3907] lr: 1.4830e-05 eta: 1:04:54 time: 0.6350 data_time: 0.4890 memory: 6319 loss: 0.0831 +2023/06/05 05:13:34 - mmengine - INFO - Epoch(train) [9][2100/3907] lr: 1.4668e-05 eta: 1:03:47 time: 0.6452 data_time: 0.5043 memory: 6319 loss: 0.1006 +2023/06/05 05:14:40 - mmengine - INFO - Epoch(train) [9][2200/3907] lr: 1.4509e-05 eta: 1:02:40 time: 0.6242 data_time: 0.4838 memory: 6319 loss: 0.0748 +2023/06/05 05:15:45 - mmengine - INFO - Epoch(train) [9][2300/3907] lr: 1.4353e-05 eta: 1:01:32 time: 0.6371 data_time: 0.4971 memory: 6319 loss: 0.0890 +2023/06/05 05:16:51 - mmengine - INFO - Epoch(train) [9][2400/3907] lr: 1.4199e-05 eta: 1:00:25 time: 0.6479 data_time: 0.5083 memory: 6319 loss: 0.0932 +2023/06/05 05:17:56 - mmengine - INFO - Epoch(train) [9][2500/3907] lr: 1.4047e-05 eta: 0:59:18 time: 0.6291 data_time: 0.4894 memory: 6319 loss: 0.0795 +2023/06/05 05:19:00 - mmengine - INFO - Epoch(train) [9][2600/3907] lr: 1.3899e-05 eta: 0:58:10 time: 0.6274 data_time: 0.4867 memory: 6319 loss: 0.0787 +2023/06/05 05:20:05 - mmengine - INFO - Epoch(train) [9][2700/3907] lr: 1.3753e-05 eta: 0:57:03 time: 0.6523 data_time: 0.5119 memory: 6319 loss: 0.0862 +2023/06/05 05:20:37 - mmengine - INFO - Exp name: resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1_20230604_225351 +2023/06/05 05:21:09 - mmengine - INFO - Epoch(train) [9][2800/3907] lr: 1.3609e-05 eta: 0:55:56 time: 0.6593 data_time: 0.5193 memory: 6319 loss: 0.0787 +2023/06/05 05:22:13 - mmengine - INFO - Epoch(train) [9][2900/3907] lr: 1.3469e-05 eta: 0:54:49 time: 0.6560 data_time: 0.5041 memory: 6319 loss: 0.0808 +2023/06/05 05:23:16 - mmengine - INFO - Epoch(train) [9][3000/3907] lr: 1.3331e-05 eta: 0:53:41 time: 0.5918 data_time: 0.4517 memory: 6319 loss: 0.0821 +2023/06/05 05:24:23 - mmengine - INFO - Epoch(train) [9][3100/3907] lr: 1.3196e-05 eta: 0:52:34 time: 0.6678 data_time: 0.5188 memory: 6319 loss: 0.0774 +2023/06/05 05:25:28 - mmengine - INFO - Epoch(train) [9][3200/3907] lr: 1.3063e-05 eta: 0:51:27 time: 0.6356 data_time: 0.4950 memory: 6319 loss: 0.0837 +2023/06/05 05:26:33 - mmengine - INFO - Epoch(train) [9][3300/3907] lr: 1.2933e-05 eta: 0:50:20 time: 0.6414 data_time: 0.4888 memory: 6319 loss: 0.0862 +2023/06/05 05:27:39 - mmengine - INFO - Epoch(train) [9][3400/3907] lr: 1.2806e-05 eta: 0:49:13 time: 0.7346 data_time: 0.5944 memory: 6319 loss: 0.0854 +2023/06/05 05:28:46 - mmengine - INFO - Epoch(train) [9][3500/3907] lr: 1.2682e-05 eta: 0:48:06 time: 0.6734 data_time: 0.5329 memory: 6319 loss: 0.0963 +2023/06/05 05:29:52 - mmengine - INFO - Epoch(train) [9][3600/3907] lr: 1.2560e-05 eta: 0:46:59 time: 0.6625 data_time: 0.5211 memory: 6319 loss: 0.0763 +2023/06/05 05:30:58 - mmengine - INFO - Epoch(train) [9][3700/3907] lr: 1.2441e-05 eta: 0:45:52 time: 0.6801 data_time: 0.5393 memory: 6319 loss: 0.0787 +2023/06/05 05:31:30 - mmengine - INFO - Exp name: resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1_20230604_225351 +2023/06/05 05:32:05 - mmengine - INFO - Epoch(train) [9][3800/3907] lr: 1.2325e-05 eta: 0:44:45 time: 0.6657 data_time: 0.5266 memory: 6319 loss: 0.0835 +2023/06/05 05:33:10 - mmengine - INFO - Epoch(train) [9][3900/3907] lr: 1.2211e-05 eta: 0:43:38 time: 0.6446 data_time: 0.5052 memory: 6319 loss: 0.0826 +2023/06/05 05:33:16 - mmengine - INFO - Exp name: resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1_20230604_225351 +2023/06/05 05:33:16 - mmengine - INFO - Saving checkpoint at 9 epochs +2023/06/05 05:34:04 - mmengine - INFO - Epoch(val) [9][57/57] accuracy/top1: 96.4754 data_time: 0.5188 time: 0.6089 +2023/06/05 05:35:11 - mmengine - INFO - Epoch(train) [10][ 100/3907] lr: 1.2093e-05 eta: 0:42:26 time: 0.6700 data_time: 0.5301 memory: 6319 loss: 0.0877 +2023/06/05 05:36:18 - mmengine - INFO - Epoch(train) [10][ 200/3907] lr: 1.1985e-05 eta: 0:41:20 time: 0.6757 data_time: 0.5218 memory: 6319 loss: 0.0822 +2023/06/05 05:37:25 - mmengine - INFO - Epoch(train) [10][ 300/3907] lr: 1.1881e-05 eta: 0:40:13 time: 0.6407 data_time: 0.5014 memory: 6319 loss: 0.0712 +2023/06/05 05:38:31 - mmengine - INFO - Epoch(train) [10][ 400/3907] lr: 1.1778e-05 eta: 0:39:06 time: 0.6248 data_time: 0.4735 memory: 6319 loss: 0.0819 +2023/06/05 05:39:36 - mmengine - INFO - Epoch(train) [10][ 500/3907] lr: 1.1679e-05 eta: 0:37:59 time: 0.6288 data_time: 0.4895 memory: 6319 loss: 0.0824 +2023/06/05 05:40:42 - mmengine - INFO - Epoch(train) [10][ 600/3907] lr: 1.1583e-05 eta: 0:36:52 time: 0.6897 data_time: 0.5363 memory: 6319 loss: 0.0834 +2023/06/05 05:41:49 - mmengine - INFO - Epoch(train) [10][ 700/3907] lr: 1.1489e-05 eta: 0:35:45 time: 0.6838 data_time: 0.5400 memory: 6319 loss: 0.0853 +2023/06/05 05:42:54 - mmengine - INFO - Epoch(train) [10][ 800/3907] lr: 1.1398e-05 eta: 0:34:38 time: 0.7402 data_time: 0.5999 memory: 6319 loss: 0.0971 +2023/06/05 05:43:20 - mmengine - INFO - Exp name: resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1_20230604_225351 +2023/06/05 05:43:59 - mmengine - INFO - Epoch(train) [10][ 900/3907] lr: 1.1310e-05 eta: 0:33:31 time: 0.6521 data_time: 0.5114 memory: 6319 loss: 0.0844 +2023/06/05 05:45:04 - mmengine - INFO - Epoch(train) [10][1000/3907] lr: 1.1225e-05 eta: 0:32:24 time: 0.7041 data_time: 0.5624 memory: 6319 loss: 0.0777 +2023/06/05 05:46:11 - mmengine - INFO - Epoch(train) [10][1100/3907] lr: 1.1142e-05 eta: 0:31:17 time: 0.6838 data_time: 0.5429 memory: 6319 loss: 0.0802 +2023/06/05 05:47:16 - mmengine - INFO - Epoch(train) [10][1200/3907] lr: 1.1063e-05 eta: 0:30:10 time: 0.6186 data_time: 0.4776 memory: 6319 loss: 0.0894 +2023/06/05 05:48:22 - mmengine - INFO - Epoch(train) [10][1300/3907] lr: 1.0986e-05 eta: 0:29:03 time: 0.6485 data_time: 0.5084 memory: 6319 loss: 0.0824 +2023/06/05 05:49:27 - mmengine - INFO - Epoch(train) [10][1400/3907] lr: 1.0912e-05 eta: 0:27:56 time: 0.6370 data_time: 0.4960 memory: 6319 loss: 0.0967 +2023/06/05 05:50:33 - mmengine - INFO - Epoch(train) [10][1500/3907] lr: 1.0841e-05 eta: 0:26:49 time: 0.6530 data_time: 0.5139 memory: 6319 loss: 0.0790 +2023/06/05 05:51:38 - mmengine - INFO - Epoch(train) [10][1600/3907] lr: 1.0773e-05 eta: 0:25:42 time: 0.6111 data_time: 0.4700 memory: 6319 loss: 0.0812 +2023/06/05 05:52:44 - mmengine - INFO - Epoch(train) [10][1700/3907] lr: 1.0707e-05 eta: 0:24:35 time: 0.6769 data_time: 0.5374 memory: 6319 loss: 0.0772 +2023/06/05 05:53:50 - mmengine - INFO - Epoch(train) [10][1800/3907] lr: 1.0645e-05 eta: 0:23:28 time: 0.6857 data_time: 0.5459 memory: 6319 loss: 0.0683 +2023/06/05 05:54:15 - mmengine - INFO - Exp name: resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1_20230604_225351 +2023/06/05 05:54:55 - mmengine - INFO - Epoch(train) [10][1900/3907] lr: 1.0585e-05 eta: 0:22:21 time: 0.6731 data_time: 0.5325 memory: 6319 loss: 0.0693 +2023/06/05 05:56:01 - mmengine - INFO - Epoch(train) [10][2000/3907] lr: 1.0529e-05 eta: 0:21:14 time: 0.6679 data_time: 0.5285 memory: 6319 loss: 0.0746 +2023/06/05 05:57:08 - mmengine - INFO - Epoch(train) [10][2100/3907] lr: 1.0475e-05 eta: 0:20:07 time: 0.6382 data_time: 0.4974 memory: 6319 loss: 0.0836 +2023/06/05 05:58:13 - mmengine - INFO - Epoch(train) [10][2200/3907] lr: 1.0424e-05 eta: 0:19:00 time: 0.6249 data_time: 0.4841 memory: 6319 loss: 0.0870 +2023/06/05 05:59:21 - mmengine - INFO - Epoch(train) [10][2300/3907] lr: 1.0376e-05 eta: 0:17:54 time: 0.7029 data_time: 0.5610 memory: 6319 loss: 0.0942 +2023/06/05 06:00:28 - mmengine - INFO - Epoch(train) [10][2400/3907] lr: 1.0330e-05 eta: 0:16:47 time: 0.6636 data_time: 0.5243 memory: 6319 loss: 0.0782 +2023/06/05 06:01:34 - mmengine - INFO - Epoch(train) [10][2500/3907] lr: 1.0288e-05 eta: 0:15:40 time: 0.6514 data_time: 0.5105 memory: 6319 loss: 0.0888 +2023/06/05 06:02:40 - mmengine - INFO - Epoch(train) [10][2600/3907] lr: 1.0249e-05 eta: 0:14:33 time: 0.6535 data_time: 0.5124 memory: 6319 loss: 0.0817 +2023/06/05 06:03:46 - mmengine - INFO - Epoch(train) [10][2700/3907] lr: 1.0212e-05 eta: 0:13:26 time: 0.6815 data_time: 0.5410 memory: 6319 loss: 0.0876 +2023/06/05 06:04:52 - mmengine - INFO - Epoch(train) [10][2800/3907] lr: 1.0178e-05 eta: 0:12:19 time: 0.6742 data_time: 0.5327 memory: 6319 loss: 0.0873 +2023/06/05 06:05:18 - mmengine - INFO - Exp name: resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1_20230604_225351 +2023/06/05 06:05:57 - mmengine - INFO - Epoch(train) [10][2900/3907] lr: 1.0148e-05 eta: 0:11:12 time: 0.6848 data_time: 0.5445 memory: 6319 loss: 0.0851 +2023/06/05 06:07:05 - mmengine - INFO - Epoch(train) [10][3000/3907] lr: 1.0120e-05 eta: 0:10:06 time: 0.7787 data_time: 0.6379 memory: 6319 loss: 0.0841 +2023/06/05 06:08:11 - mmengine - INFO - Epoch(train) [10][3100/3907] lr: 1.0095e-05 eta: 0:08:59 time: 0.6587 data_time: 0.5178 memory: 6319 loss: 0.0948 +2023/06/05 06:09:17 - mmengine - INFO - Epoch(train) [10][3200/3907] lr: 1.0073e-05 eta: 0:07:52 time: 0.6501 data_time: 0.5099 memory: 6319 loss: 0.0908 +2023/06/05 06:10:24 - mmengine - INFO - Epoch(train) [10][3300/3907] lr: 1.0054e-05 eta: 0:06:45 time: 0.6800 data_time: 0.5393 memory: 6319 loss: 0.0735 +2023/06/05 06:11:29 - mmengine - INFO - Epoch(train) [10][3400/3907] lr: 1.0038e-05 eta: 0:05:38 time: 0.6487 data_time: 0.5076 memory: 6319 loss: 0.0899 +2023/06/05 06:12:35 - mmengine - INFO - Epoch(train) [10][3500/3907] lr: 1.0024e-05 eta: 0:04:31 time: 0.6477 data_time: 0.5074 memory: 6319 loss: 0.0803 +2023/06/05 06:13:41 - mmengine - INFO - Epoch(train) [10][3600/3907] lr: 1.0014e-05 eta: 0:03:25 time: 0.6441 data_time: 0.5041 memory: 6319 loss: 0.0884 +2023/06/05 06:14:47 - mmengine - INFO - Epoch(train) [10][3700/3907] lr: 1.0006e-05 eta: 0:02:18 time: 0.6623 data_time: 0.5217 memory: 6319 loss: 0.0832 +2023/06/05 06:15:53 - mmengine - INFO - Epoch(train) [10][3800/3907] lr: 1.0002e-05 eta: 0:01:11 time: 0.6792 data_time: 0.5377 memory: 6319 loss: 0.0842 +2023/06/05 06:16:20 - mmengine - INFO - Exp name: resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1_20230604_225351 +2023/06/05 06:17:00 - mmengine - INFO - Epoch(train) [10][3900/3907] lr: 1.0000e-05 eta: 0:00:04 time: 0.6661 data_time: 0.5254 memory: 6319 loss: 0.0833 +2023/06/05 06:17:05 - mmengine - INFO - Exp name: resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1_20230604_225351 +2023/06/05 06:17:05 - mmengine - INFO - Saving checkpoint at 10 epochs +2023/06/05 06:17:54 - mmengine - INFO - Epoch(val) [10][57/57] accuracy/top1: 96.8686 data_time: 0.5283 time: 0.6142 diff --git a/resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1/20230604_225351/vis_data/20230604_225351.json b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1/20230604_225351/vis_data/20230604_225351.json new file mode 100644 index 0000000000000000000000000000000000000000..304f3697b6d295f35f2ceef96ae889bd834b54c7 --- /dev/null +++ b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1/20230604_225351/vis_data/20230604_225351.json @@ -0,0 +1,400 @@ +{"lr": 9.999857418382631e-05, "data_time": 0.25231518745422366, "loss": 0.6043034374713898, "time": 0.6773250818252563, "epoch": 1, "memory": 9436, "step": 100} +{"lr": 9.999423907359506e-05, "data_time": 0.0753288745880127, "loss": 0.53575439453125, "time": 0.7267685651779174, "epoch": 1, "memory": 6319, "step": 200} +{"lr": 9.998699480412073e-05, "data_time": 0.0010489463806152345, "loss": 0.49411348402500155, "time": 0.6230251789093018, "epoch": 1, "memory": 6319, "step": 300} +{"lr": 9.997684184379085e-05, "data_time": 0.000860142707824707, "loss": 0.42975725531578063, "time": 0.6354247808456421, "epoch": 1, "memory": 6319, "step": 400} +{"lr": 9.996378084905784e-05, "data_time": 0.0007945775985717773, "loss": 0.41185124814510343, "time": 0.7114021062850953, "epoch": 1, "memory": 6319, "step": 500} +{"lr": 9.994781266439667e-05, "data_time": 0.0011701345443725585, "loss": 0.3755955159664154, "time": 0.6573834180831909, "epoch": 1, "memory": 6319, "step": 600} +{"lr": 9.992893832225061e-05, "data_time": 0.0008877992630004882, "loss": 0.3354937732219696, "time": 0.6607741355895996, "epoch": 1, "memory": 6319, "step": 700} +{"lr": 9.990715904296405e-05, "data_time": 0.001269364356994629, "loss": 0.3326795011758804, "time": 0.6426759719848633, "epoch": 1, "memory": 6319, "step": 800} +{"lr": 9.988247623470371e-05, "data_time": 0.0008656740188598633, "loss": 0.30257115662097933, "time": 0.6511765241622924, "epoch": 1, "memory": 6319, "step": 900} +{"lr": 9.985489149336749e-05, "data_time": 0.000989818572998047, "loss": 0.2756628647446632, "time": 0.690179967880249, "epoch": 1, "memory": 6319, "step": 1000} +{"lr": 9.982440660248187e-05, "data_time": 0.0009242057800292969, "loss": 0.2721254274249077, "time": 0.7048046827316284, "epoch": 1, "memory": 6319, "step": 1100} +{"lr": 9.979102353308574e-05, "data_time": 0.0010079145431518555, "loss": 0.26231861114501953, "time": 0.6478001832962036, "epoch": 1, "memory": 6319, "step": 1200} +{"lr": 9.975474444360358e-05, "data_time": 0.0008924007415771484, "loss": 0.25315001159906386, "time": 0.6526968002319335, "epoch": 1, "memory": 6319, "step": 1300} +{"lr": 9.97155716797056e-05, "data_time": 0.0009036779403686524, "loss": 0.27018743604421613, "time": 0.7147150754928588, "epoch": 1, "memory": 6319, "step": 1400} +{"lr": 9.967350777415632e-05, "data_time": 0.0007965564727783203, "loss": 0.24654453098773957, "time": 0.6703690528869629, "epoch": 1, "memory": 6319, "step": 1500} +{"lr": 9.962855544665059e-05, "data_time": 0.001060152053833008, "loss": 0.23762782663106918, "time": 0.6793582677841187, "epoch": 1, "memory": 6319, "step": 1600} +{"lr": 9.958071760363773e-05, "data_time": 0.0008116006851196289, "loss": 0.22109235376119613, "time": 0.6407901525497437, "epoch": 1, "memory": 6319, "step": 1700} +{"lr": 9.952999733813382e-05, "data_time": 0.10425570011138915, "loss": 0.23239289820194245, "time": 0.6414692878723145, "epoch": 1, "memory": 6319, "step": 1800} +{"lr": 9.947639792952148e-05, "data_time": 0.5308783292770386, "loss": 0.21411006152629852, "time": 0.6705950498580933, "epoch": 1, "memory": 6319, "step": 1900} +{"lr": 9.941992284333813e-05, "data_time": 0.500979995727539, "loss": 0.22601257115602494, "time": 0.7200746774673462, "epoch": 1, "memory": 6319, "step": 2000} +{"lr": 9.93605757310516e-05, "data_time": 0.5712074756622314, "loss": 0.2068329483270645, "time": 0.7120099782943725, "epoch": 1, "memory": 6319, "step": 2100} +{"lr": 9.929836042982421e-05, "data_time": 0.5833227157592773, "loss": 0.19560886919498444, "time": 0.7229428052902221, "epoch": 1, "memory": 6319, "step": 2200} +{"lr": 9.923328096226467e-05, "data_time": 0.5348201990127563, "loss": 0.2159124195575714, "time": 0.6739555835723877, "epoch": 1, "memory": 6319, "step": 2300} +{"lr": 9.916534153616819e-05, "data_time": 0.49794604778289797, "loss": 0.19221736639738082, "time": 0.637341833114624, "epoch": 1, "memory": 6319, "step": 2400} +{"lr": 9.909454654424387e-05, "data_time": 0.5531359910964966, "loss": 0.20310077518224717, "time": 0.6929550647735596, "epoch": 1, "memory": 6319, "step": 2500} +{"lr": 9.902090056383109e-05, "data_time": 0.5236783504486084, "loss": 0.17665813416242598, "time": 0.6639738559722901, "epoch": 1, "memory": 6319, "step": 2600} +{"lr": 9.894440835660365e-05, "data_time": 0.5283995389938354, "loss": 0.20761384516954423, "time": 0.6682494163513184, "epoch": 1, "memory": 6319, "step": 2700} +{"lr": 9.886507486826162e-05, "data_time": 0.5253950595855713, "loss": 0.18284522891044616, "time": 0.6658698797225953, "epoch": 1, "memory": 6319, "step": 2800} +{"lr": 9.878290522821154e-05, "data_time": 0.5330732345581055, "loss": 0.16379252076148987, "time": 0.6735464811325074, "epoch": 1, "memory": 6319, "step": 2900} +{"lr": 9.86979047492352e-05, "data_time": 0.5406192064285278, "loss": 0.1843602553009987, "time": 0.6811985731124878, "epoch": 1, "memory": 6319, "step": 3000} +{"lr": 9.861007892714578e-05, "data_time": 0.5320868730545044, "loss": 0.1826549530029297, "time": 0.6714630842208862, "epoch": 1, "memory": 6319, "step": 3100} +{"lr": 9.851943344043239e-05, "data_time": 0.5344639778137207, "loss": 0.17140807658433915, "time": 0.6751244544982911, "epoch": 1, "memory": 6319, "step": 3200} +{"lr": 9.842597414989319e-05, "data_time": 0.5207000017166138, "loss": 0.16282510459423066, "time": 0.6606178760528565, "epoch": 1, "memory": 6319, "step": 3300} +{"lr": 9.832970709825637e-05, "data_time": 0.5294976711273194, "loss": 0.1740340769290924, "time": 0.668919587135315, "epoch": 1, "memory": 6319, "step": 3400} +{"lr": 9.823063850978968e-05, "data_time": 0.5587979078292846, "loss": 0.18116564899683, "time": 0.6999265432357789, "epoch": 1, "memory": 6319, "step": 3500} +{"lr": 9.81287747898975e-05, "data_time": 0.5234679460525513, "loss": 0.16375539302825928, "time": 0.663607382774353, "epoch": 1, "memory": 6319, "step": 3600} +{"lr": 9.802412252470715e-05, "data_time": 0.5097539186477661, "loss": 0.17133889049291612, "time": 0.6512763977050782, "epoch": 1, "memory": 6319, "step": 3700} +{"lr": 9.791668848064258e-05, "data_time": 0.5157545566558838, "loss": 0.1612338401377201, "time": 0.6675833225250244, "epoch": 1, "memory": 6319, "step": 3800} +{"lr": 9.780647960398751e-05, "data_time": 0.5062771320343018, "loss": 0.17370904386043548, "time": 0.6451803684234619, "epoch": 1, "memory": 6319, "step": 3900} +{"accuracy/top1": 79.28742980957031, "data_time": 0.5958551900428638, "time": 0.6883839389734101, "step": 1} +{"lr": 9.7685491196646e-05, "data_time": 0.5370930910110474, "loss": 0.15279597342014312, "time": 0.6767707109451294, "epoch": 2, "memory": 6319, "step": 4007} +{"lr": 9.756956126085676e-05, "data_time": 0.5544443845748901, "loss": 0.14128805696964264, "time": 0.6940245151519775, "epoch": 2, "memory": 6319, "step": 4107} +{"lr": 9.745087893642482e-05, "data_time": 0.5067760944366455, "loss": 0.15452901273965836, "time": 0.646721601486206, "epoch": 2, "memory": 6319, "step": 4207} +{"lr": 9.73294518969056e-05, "data_time": 0.5036807775497436, "loss": 0.14054470509290695, "time": 0.6443322420120239, "epoch": 2, "memory": 6319, "step": 4307} +{"lr": 9.72052879933172e-05, "data_time": 0.4447112321853638, "loss": 0.15058355033397675, "time": 0.5851428747177124, "epoch": 2, "memory": 6319, "step": 4407} +{"lr": 9.707839525363358e-05, "data_time": 0.5212276458740235, "loss": 0.13102484717965127, "time": 0.6614755630493164, "epoch": 2, "memory": 6319, "step": 4507} +{"lr": 9.69487818822646e-05, "data_time": 0.5257565021514893, "loss": 0.14825795441865922, "time": 0.6660473108291626, "epoch": 2, "memory": 6319, "step": 4607} +{"lr": 9.681645625952616e-05, "data_time": 0.49306907653808596, "loss": 0.13866764903068543, "time": 0.6335996627807617, "epoch": 2, "memory": 6319, "step": 4707} +{"lr": 9.668142694109826e-05, "data_time": 0.5573489904403687, "loss": 0.14306616485118867, "time": 0.6974190473556519, "epoch": 2, "memory": 6319, "step": 4807} +{"lr": 9.654370265747162e-05, "data_time": 0.49083871841430665, "loss": 0.15921448469161986, "time": 0.6309617757797241, "epoch": 2, "memory": 6319, "step": 4907} +{"lr": 9.640329231338345e-05, "data_time": 0.5240086793899537, "loss": 0.1494707390666008, "time": 0.6651803970336914, "epoch": 2, "memory": 6319, "step": 5007} +{"lr": 9.626020498724147e-05, "data_time": 0.5176722526550293, "loss": 0.14862466901540755, "time": 0.6592450618743897, "epoch": 2, "memory": 6319, "step": 5107} +{"lr": 9.611444993053717e-05, "data_time": 0.6180542945861817, "loss": 0.1456143379211426, "time": 0.7580090999603272, "epoch": 2, "memory": 6319, "step": 5207} +{"lr": 9.596603656724746e-05, "data_time": 0.5168131828308106, "loss": 0.15639145374298097, "time": 0.6575065135955811, "epoch": 2, "memory": 6319, "step": 5307} +{"lr": 9.58149744932254e-05, "data_time": 0.5288204193115235, "loss": 0.1405955284833908, "time": 0.6684650659561158, "epoch": 2, "memory": 6319, "step": 5407} +{"lr": 9.566127347557972e-05, "data_time": 0.6000181674957276, "loss": 0.14103675931692122, "time": 0.7403709411621093, "epoch": 2, "memory": 6319, "step": 5507} +{"lr": 9.550494345204367e-05, "data_time": 0.5291576623916626, "loss": 0.14637183845043183, "time": 0.669269609451294, "epoch": 2, "memory": 6319, "step": 5607} +{"lr": 9.534599453033183e-05, "data_time": 0.5193047523498535, "loss": 0.12512431144714356, "time": 0.6595483541488647, "epoch": 2, "memory": 6319, "step": 5707} +{"lr": 9.518443698748712e-05, "data_time": 0.5405029296875, "loss": 0.14058603942394257, "time": 0.6811832666397095, "epoch": 2, "memory": 6319, "step": 5807} +{"lr": 9.50202812692162e-05, "data_time": 0.5031807899475098, "loss": 0.1317460559308529, "time": 0.6625309467315674, "epoch": 2, "memory": 6319, "step": 5907} +{"lr": 9.485353798921386e-05, "data_time": 0.5442139387130738, "loss": 0.14158946871757508, "time": 0.683808445930481, "epoch": 2, "memory": 6319, "step": 6007} +{"lr": 9.468421792847712e-05, "data_time": 0.5201704502105713, "loss": 0.1402334488928318, "time": 0.6704195499420166, "epoch": 2, "memory": 6319, "step": 6107} +{"lr": 9.451233203460794e-05, "data_time": 0.5358564853668213, "loss": 0.13895688578486443, "time": 0.6754046201705932, "epoch": 2, "memory": 6319, "step": 6207} +{"lr": 9.433789142110532e-05, "data_time": 0.48852086067199707, "loss": 0.13254403844475746, "time": 0.6629394054412842, "epoch": 2, "memory": 6319, "step": 6307} +{"lr": 9.416090736664709e-05, "data_time": 0.4711421251296997, "loss": 0.13277369290590285, "time": 0.611909294128418, "epoch": 2, "memory": 6319, "step": 6407} +{"lr": 9.398139131436032e-05, "data_time": 0.5422593593597412, "loss": 0.1290692836046219, "time": 0.6860929727554321, "epoch": 2, "memory": 6319, "step": 6507} +{"lr": 9.379935487108167e-05, "data_time": 0.5193511486053467, "loss": 0.1384017914533615, "time": 0.6600296020507812, "epoch": 2, "memory": 6319, "step": 6607} +{"lr": 9.361480980660697e-05, "data_time": 0.5154844999313355, "loss": 0.14731456190347672, "time": 0.6555843830108643, "epoch": 2, "memory": 6319, "step": 6707} +{"lr": 9.342776805292993e-05, "data_time": 0.511954951286316, "loss": 0.12090265229344369, "time": 0.6514381885528564, "epoch": 2, "memory": 6319, "step": 6807} +{"lr": 9.323824170347083e-05, "data_time": 0.5648843050003052, "loss": 0.1435110352933407, "time": 0.7055478811264038, "epoch": 2, "memory": 6319, "step": 6907} +{"lr": 9.304624301229471e-05, "data_time": 0.48480408191680907, "loss": 0.13903704956173896, "time": 0.6247333288192749, "epoch": 2, "memory": 6319, "step": 7007} +{"lr": 9.285178439331897e-05, "data_time": 0.5070621728897095, "loss": 0.12053157761693001, "time": 0.6478193283081055, "epoch": 2, "memory": 6319, "step": 7107} +{"lr": 9.265487841951087e-05, "data_time": 0.49961085319519044, "loss": 0.13237974122166635, "time": 0.6394788503646851, "epoch": 2, "memory": 6319, "step": 7207} +{"lr": 9.245553782207421e-05, "data_time": 0.5493011474609375, "loss": 0.12692570984363555, "time": 0.6906674385070801, "epoch": 2, "memory": 6319, "step": 7307} +{"lr": 9.225377548962671e-05, "data_time": 0.5684386014938354, "loss": 0.12293680310249329, "time": 0.7101089239120484, "epoch": 2, "memory": 6319, "step": 7407} +{"lr": 9.204960446736601e-05, "data_time": 0.4899008750915527, "loss": 0.11427125558257104, "time": 0.6302706718444824, "epoch": 2, "memory": 6319, "step": 7507} +{"lr": 9.184303795622715e-05, "data_time": 0.5326293230056762, "loss": 0.12831226736307144, "time": 0.6717268705368042, "epoch": 2, "memory": 6319, "step": 7607} +{"lr": 9.163408931202823e-05, "data_time": 0.5166270971298218, "loss": 0.12190170288085937, "time": 0.6563956499099731, "epoch": 2, "memory": 6319, "step": 7707} +{"lr": 9.142277204460729e-05, "data_time": 0.4866673707962036, "loss": 0.12000658884644508, "time": 0.6266154050827026, "epoch": 2, "memory": 6319, "step": 7807} +{"accuracy/top1": 84.4159927368164, "data_time": 0.47536983983270054, "time": 0.5640446399820263, "step": 2} +{"lr": 9.119405492383288e-05, "data_time": 0.5369272470474243, "loss": 0.1321154534816742, "time": 0.6764336824417114, "epoch": 3, "memory": 6319, "step": 7914} +{"lr": 9.097787819233205e-05, "data_time": 0.4976217269897461, "loss": 0.10088460966944694, "time": 0.6464096069335937, "epoch": 3, "memory": 6319, "step": 8014} +{"lr": 9.075937526576627e-05, "data_time": 0.5266026496887207, "loss": 0.14688998758792876, "time": 0.6661184072494507, "epoch": 3, "memory": 6319, "step": 8114} +{"lr": 9.053856027171755e-05, "data_time": 0.511517333984375, "loss": 0.11930367425084114, "time": 0.6567173957824707, "epoch": 3, "memory": 6319, "step": 8214} +{"lr": 9.031544748725785e-05, "data_time": 0.5386279821395874, "loss": 0.11722368896007537, "time": 0.6791386604309082, "epoch": 3, "memory": 6319, "step": 8314} +{"lr": 9.009005133802542e-05, "data_time": 0.5394606590270996, "loss": 0.12796138003468513, "time": 0.6794095516204834, "epoch": 3, "memory": 6319, "step": 8414} +{"lr": 8.986238639729242e-05, "data_time": 0.49581429958343504, "loss": 0.12170281857252122, "time": 0.636667799949646, "epoch": 3, "memory": 6319, "step": 8514} +{"lr": 8.963246738502254e-05, "data_time": 0.6008338689804077, "loss": 0.11625181287527084, "time": 0.741152572631836, "epoch": 3, "memory": 6319, "step": 8614} +{"lr": 8.940030916691928e-05, "data_time": 0.5272071599960327, "loss": 0.12248632237315178, "time": 0.6671080112457275, "epoch": 3, "memory": 6319, "step": 8714} +{"lr": 8.91659267534649e-05, "data_time": 0.497597861289978, "loss": 0.13870976641774177, "time": 0.6385526418685913, "epoch": 3, "memory": 6319, "step": 8814} +{"lr": 8.892933529894978e-05, "data_time": 0.5019416570663452, "loss": 0.10886357426643371, "time": 0.6426058292388916, "epoch": 3, "memory": 6319, "step": 8914} +{"lr": 8.869055010049283e-05, "data_time": 0.5331192493438721, "loss": 0.1227203406393528, "time": 0.6732116222381592, "epoch": 3, "memory": 6319, "step": 9014} +{"lr": 8.844958659705188e-05, "data_time": 0.49990861415863036, "loss": 0.11642178148031235, "time": 0.6398486137390137, "epoch": 3, "memory": 6319, "step": 9114} +{"lr": 8.82064603684259e-05, "data_time": 0.501291298866272, "loss": 0.11247068196535111, "time": 0.6419290065765381, "epoch": 3, "memory": 6319, "step": 9214} +{"lr": 8.796118713424776e-05, "data_time": 0.5387366771697998, "loss": 0.11492721736431122, "time": 0.6775113105773926, "epoch": 3, "memory": 6319, "step": 9314} +{"lr": 8.77137827529676e-05, "data_time": 0.5013673305511475, "loss": 0.11084105968475341, "time": 0.6408612489700317, "epoch": 3, "memory": 6319, "step": 9414} +{"lr": 8.746426322082745e-05, "data_time": 0.7175061464309692, "loss": 0.10803602561354637, "time": 0.8585130214691162, "epoch": 3, "memory": 6319, "step": 9514} +{"lr": 8.721264467082732e-05, "data_time": 0.507056450843811, "loss": 0.11435580402612686, "time": 0.6468563318252564, "epoch": 3, "memory": 6319, "step": 9614} +{"lr": 8.695894337168168e-05, "data_time": 0.4911908388137817, "loss": 0.11398598477244377, "time": 0.6302866458892822, "epoch": 3, "memory": 6319, "step": 9714} +{"lr": 8.67031757267677e-05, "data_time": 0.504483413696289, "loss": 0.11963612660765648, "time": 0.6445215225219727, "epoch": 3, "memory": 6319, "step": 9814} +{"lr": 8.64453582730651e-05, "data_time": 0.5450257301330567, "loss": 0.1172746978700161, "time": 0.6851329326629638, "epoch": 3, "memory": 6319, "step": 9914} +{"lr": 8.61855076800861e-05, "data_time": 0.521156644821167, "loss": 0.10335676074028015, "time": 0.6610426425933837, "epoch": 3, "memory": 6319, "step": 10014} +{"lr": 8.592364074879843e-05, "data_time": 0.5473639488220214, "loss": 0.11976106017827988, "time": 0.6864054679870606, "epoch": 3, "memory": 6319, "step": 10114} +{"lr": 8.565977441053854e-05, "data_time": 0.5903863906860352, "loss": 0.1248409979045391, "time": 0.7293467283248901, "epoch": 3, "memory": 6319, "step": 10214} +{"lr": 8.539392572591724e-05, "data_time": 0.5097359180450439, "loss": 0.13603954315185546, "time": 0.6492294788360595, "epoch": 3, "memory": 6319, "step": 10314} +{"lr": 8.512611188371633e-05, "data_time": 0.6894086122512817, "loss": 0.10091322213411331, "time": 0.8286743879318237, "epoch": 3, "memory": 6319, "step": 10414} +{"lr": 8.485635019977725e-05, "data_time": 0.5736372947692872, "loss": 0.11598249152302742, "time": 0.7133435726165771, "epoch": 3, "memory": 6319, "step": 10514} +{"lr": 8.45846581158818e-05, "data_time": 0.5090338945388794, "loss": 0.09406849667429924, "time": 0.6500195264816284, "epoch": 3, "memory": 6319, "step": 10614} +{"lr": 8.431105319862416e-05, "data_time": 0.5089647054672242, "loss": 0.12137909457087517, "time": 0.6541881322860718, "epoch": 3, "memory": 6319, "step": 10714} +{"lr": 8.403555313827516e-05, "data_time": 0.5094683170318604, "loss": 0.10281328037381172, "time": 0.6606269598007202, "epoch": 3, "memory": 6319, "step": 10814} +{"lr": 8.375817574763844e-05, "data_time": 0.5197919845581055, "loss": 0.10709384977817535, "time": 0.659517502784729, "epoch": 3, "memory": 6319, "step": 10914} +{"lr": 8.347893896089871e-05, "data_time": 0.49315719604492186, "loss": 0.10601055026054382, "time": 0.6474194049835205, "epoch": 3, "memory": 6319, "step": 11014} +{"lr": 8.31978608324625e-05, "data_time": 0.4724468946456909, "loss": 0.11521460637450218, "time": 0.6125427961349488, "epoch": 3, "memory": 6319, "step": 11114} +{"lr": 8.291495953579052e-05, "data_time": 0.5435532093048095, "loss": 0.0990497924387455, "time": 0.6833879709243774, "epoch": 3, "memory": 6319, "step": 11214} +{"lr": 8.263025336222267e-05, "data_time": 0.5444221019744873, "loss": 0.10346187576651573, "time": 0.684584665298462, "epoch": 3, "memory": 6319, "step": 11314} +{"lr": 8.234376071979551e-05, "data_time": 0.5355138540267944, "loss": 0.10680332705378533, "time": 0.6770189762115478, "epoch": 3, "memory": 6319, "step": 11414} +{"lr": 8.205550013205187e-05, "data_time": 0.5348822355270386, "loss": 0.10994913391768932, "time": 0.6740388154983521, "epoch": 3, "memory": 6319, "step": 11514} +{"lr": 8.17654902368435e-05, "data_time": 0.5360685110092163, "loss": 0.10609661005437374, "time": 0.6752196073532104, "epoch": 3, "memory": 6319, "step": 11614} +{"lr": 8.147374978512582e-05, "data_time": 0.4762966394424438, "loss": 0.10849932357668876, "time": 0.6158416271209717, "epoch": 3, "memory": 6319, "step": 11714} +{"accuracy/top1": 87.34212493896484, "data_time": 0.5443065618646556, "time": 0.6336771126451164, "step": 3} +{"lr": 8.115969237626578e-05, "data_time": 0.588813328742981, "loss": 0.11412608325481415, "time": 0.7443063259124756, "epoch": 4, "memory": 6319, "step": 11821} +{"lr": 8.08644297337245e-05, "data_time": 0.506064486503601, "loss": 0.10286526530981063, "time": 0.6592431783676147, "epoch": 4, "memory": 6319, "step": 11921} +{"lr": 8.056749479387651e-05, "data_time": 0.4116933345794678, "loss": 0.10236260145902634, "time": 0.6955714464187622, "epoch": 4, "memory": 6319, "step": 12021} +{"lr": 8.026890675542457e-05, "data_time": 0.3840444326400757, "loss": 0.1088383249938488, "time": 0.6918084859848023, "epoch": 4, "memory": 6319, "step": 12121} +{"lr": 7.996868492395457e-05, "data_time": 0.30678324699401854, "loss": 0.11419989466667176, "time": 0.6832905769348144, "epoch": 4, "memory": 6319, "step": 12221} +{"lr": 7.966684871068762e-05, "data_time": 0.3399131059646606, "loss": 0.1026128277182579, "time": 0.6849857330322265, "epoch": 4, "memory": 6319, "step": 12321} +{"lr": 7.936341763122449e-05, "data_time": 0.37445812225341796, "loss": 0.10519614145159721, "time": 0.653217077255249, "epoch": 4, "memory": 6319, "step": 12421} +{"lr": 7.905841130428405e-05, "data_time": 0.4166030168533325, "loss": 0.10487660765647888, "time": 0.6608822822570801, "epoch": 4, "memory": 6319, "step": 12521} +{"lr": 7.875184945043487e-05, "data_time": 0.31440300941467286, "loss": 0.0910740464925766, "time": 0.6547337293624877, "epoch": 4, "memory": 6319, "step": 12621} +{"lr": 7.844375189082012e-05, "data_time": 0.27991909980773927, "loss": 0.1040660209953785, "time": 0.7189595699310303, "epoch": 4, "memory": 6319, "step": 12721} +{"lr": 7.813413854587572e-05, "data_time": 0.3071895599365234, "loss": 0.08887388631701469, "time": 0.6119935035705566, "epoch": 4, "memory": 6319, "step": 12821} +{"lr": 7.782302943404291e-05, "data_time": 0.5093741893768311, "loss": 0.11017532199621201, "time": 0.649790620803833, "epoch": 4, "memory": 6319, "step": 12921} +{"lr": 7.751044467047337e-05, "data_time": 0.541435956954956, "loss": 0.11994380503892899, "time": 0.68137047290802, "epoch": 4, "memory": 6319, "step": 13021} +{"lr": 7.719640446572907e-05, "data_time": 0.5784233808517456, "loss": 0.10489736124873161, "time": 0.717302942276001, "epoch": 4, "memory": 6319, "step": 13121} +{"lr": 7.688092912447536e-05, "data_time": 0.5041449308395386, "loss": 0.11665222495794296, "time": 0.6451606035232544, "epoch": 4, "memory": 6319, "step": 13221} +{"lr": 7.656403904416794e-05, "data_time": 0.53504958152771, "loss": 0.1014468491077423, "time": 0.6744351625442505, "epoch": 4, "memory": 6319, "step": 13321} +{"lr": 7.624575471373434e-05, "data_time": 0.5670973300933838, "loss": 0.10275198072195053, "time": 0.7068441867828369, "epoch": 4, "memory": 6319, "step": 13421} +{"lr": 7.592609671224926e-05, "data_time": 0.5109803676605225, "loss": 0.10678192600607872, "time": 0.6506503343582153, "epoch": 4, "memory": 6319, "step": 13521} +{"lr": 7.560508570760375e-05, "data_time": 0.5814297676086426, "loss": 0.11037224233150482, "time": 0.7270254135131836, "epoch": 4, "memory": 6319, "step": 13621} +{"lr": 7.528274245516895e-05, "data_time": 0.5672618627548218, "loss": 0.10577552020549774, "time": 0.7087949752807617, "epoch": 4, "memory": 6319, "step": 13721} +{"lr": 7.495908779645405e-05, "data_time": 0.5114144086837769, "loss": 0.0917666207998991, "time": 0.720955753326416, "epoch": 4, "memory": 6319, "step": 13821} +{"lr": 7.463414265775906e-05, "data_time": 0.5522169828414917, "loss": 0.10598318874835969, "time": 0.6921687364578247, "epoch": 4, "memory": 6319, "step": 13921} +{"lr": 7.430792804882147e-05, "data_time": 0.5370861291885376, "loss": 0.10121533870697022, "time": 0.6871623277664185, "epoch": 4, "memory": 6319, "step": 14021} +{"lr": 7.398046506145803e-05, "data_time": 0.5097856283187866, "loss": 0.11258571781218052, "time": 0.6490369081497193, "epoch": 4, "memory": 6319, "step": 14121} +{"lr": 7.365177486820114e-05, "data_time": 0.6446864604949951, "loss": 0.09884856045246124, "time": 0.7903539896011352, "epoch": 4, "memory": 6319, "step": 14221} +{"lr": 7.332187872092939e-05, "data_time": 0.48531346321105956, "loss": 0.10299100801348686, "time": 0.6248530864715576, "epoch": 4, "memory": 6319, "step": 14321} +{"lr": 7.299079794949401e-05, "data_time": 0.5231744766235351, "loss": 0.09523547291755677, "time": 0.6623619794845581, "epoch": 4, "memory": 6319, "step": 14421} +{"lr": 7.265855396033955e-05, "data_time": 0.5392854452133179, "loss": 0.09905967712402344, "time": 0.6819924354553223, "epoch": 4, "memory": 6319, "step": 14521} +{"lr": 7.232516823512013e-05, "data_time": 0.5042264461517334, "loss": 0.09532521292567253, "time": 0.6450309753417969, "epoch": 4, "memory": 6319, "step": 14621} +{"lr": 7.199066232930995e-05, "data_time": 0.5368626594543457, "loss": 0.08734327033162118, "time": 0.6775792598724365, "epoch": 4, "memory": 6319, "step": 14721} +{"lr": 7.165505787080997e-05, "data_time": 0.5642958164215088, "loss": 0.10501666218042374, "time": 0.7061097145080566, "epoch": 4, "memory": 6319, "step": 14821} +{"lr": 7.13183765585495e-05, "data_time": 0.5412783622741699, "loss": 0.08821191489696503, "time": 0.6815030813217163, "epoch": 4, "memory": 6319, "step": 14921} +{"lr": 7.098064016108317e-05, "data_time": 0.5088694095611572, "loss": 0.09396752193570138, "time": 0.6490758419036865, "epoch": 4, "memory": 6319, "step": 15021} +{"lr": 7.064187051518379e-05, "data_time": 0.5177159547805786, "loss": 0.09922549277544021, "time": 0.659341287612915, "epoch": 4, "memory": 6319, "step": 15121} +{"lr": 7.030208952442949e-05, "data_time": 0.5396681308746338, "loss": 0.09435384050011635, "time": 0.6801180362701416, "epoch": 4, "memory": 6319, "step": 15221} +{"lr": 6.996131915778851e-05, "data_time": 0.6161983013153076, "loss": 0.0895415049046278, "time": 0.7637013912200927, "epoch": 4, "memory": 6319, "step": 15321} +{"lr": 6.961958144819853e-05, "data_time": 0.5221543073654175, "loss": 0.09963801801204682, "time": 0.6628645181655883, "epoch": 4, "memory": 6319, "step": 15421} +{"lr": 6.927689849114193e-05, "data_time": 0.49598758220672606, "loss": 0.09301925748586655, "time": 0.6423600673675537, "epoch": 4, "memory": 6319, "step": 15521} +{"lr": 6.893329244321722e-05, "data_time": 0.5162488698959351, "loss": 0.09806651324033737, "time": 0.6562772035598755, "epoch": 4, "memory": 6319, "step": 15621} +{"accuracy/top1": 90.25086212158203, "data_time": 0.5513947051146935, "time": 0.6400111543721166, "step": 4} +{"lr": 6.856463687361792e-05, "data_time": 0.5262357473373414, "loss": 0.08470112420618534, "time": 0.708863353729248, "epoch": 5, "memory": 6319, "step": 15728} +{"lr": 6.821919068470432e-05, "data_time": 0.5165777683258057, "loss": 0.09095192849636077, "time": 0.6574634790420533, "epoch": 5, "memory": 6319, "step": 15828} +{"lr": 6.787288979235407e-05, "data_time": 0.5227962255477905, "loss": 0.1057699330151081, "time": 0.6635576248168945, "epoch": 5, "memory": 6319, "step": 15928} +{"lr": 6.752575658708793e-05, "data_time": 0.5789114236831665, "loss": 0.08842909708619118, "time": 0.719330883026123, "epoch": 5, "memory": 6319, "step": 16028} +{"lr": 6.71778135132408e-05, "data_time": 0.5475069046020508, "loss": 0.09648694843053818, "time": 0.6873472929000854, "epoch": 5, "memory": 6319, "step": 16128} +{"lr": 6.682908306751074e-05, "data_time": 0.5067532300949097, "loss": 0.10286273434758186, "time": 0.6608107566833497, "epoch": 5, "memory": 6319, "step": 16228} +{"lr": 6.64795877975043e-05, "data_time": 0.5259492635726929, "loss": 0.09357130825519562, "time": 0.6656451463699341, "epoch": 5, "memory": 6319, "step": 16328} +{"lr": 6.61293503002787e-05, "data_time": 0.7284724712371826, "loss": 0.10000606030225753, "time": 0.8689975500106811, "epoch": 5, "memory": 6319, "step": 16428} +{"lr": 6.577839322088084e-05, "data_time": 0.46921329498291015, "loss": 0.0939922533929348, "time": 0.6115054368972779, "epoch": 5, "memory": 6319, "step": 16528} +{"lr": 6.542673925088298e-05, "data_time": 0.5045658111572265, "loss": 0.09371010065078736, "time": 0.6595198392868042, "epoch": 5, "memory": 6319, "step": 16628} +{"lr": 6.507441112691584e-05, "data_time": 0.495917820930481, "loss": 0.09412411749362945, "time": 0.6368811845779419, "epoch": 5, "memory": 6319, "step": 16728} +{"lr": 6.47214316291983e-05, "data_time": 0.49607903957366944, "loss": 0.10073281228542327, "time": 0.6506678104400635, "epoch": 5, "memory": 6319, "step": 16828} +{"lr": 6.436782358006481e-05, "data_time": 0.5493082523345947, "loss": 0.09790346249938012, "time": 0.6898935794830322, "epoch": 5, "memory": 6319, "step": 16928} +{"lr": 6.40136098424896e-05, "data_time": 0.5583202600479126, "loss": 0.09719995222985744, "time": 0.6991142988204956, "epoch": 5, "memory": 6319, "step": 17028} +{"lr": 6.365881331860828e-05, "data_time": 0.5614637851715087, "loss": 0.07730818539857864, "time": 0.7026409149169922, "epoch": 5, "memory": 6319, "step": 17128} +{"lr": 6.330345694823739e-05, "data_time": 0.5357184648513794, "loss": 0.09566049724817276, "time": 0.6767968893051147, "epoch": 5, "memory": 6319, "step": 17228} +{"lr": 6.29475637073908e-05, "data_time": 0.5066222429275513, "loss": 0.09327691644430161, "time": 0.6473726987838745, "epoch": 5, "memory": 6319, "step": 17328} +{"lr": 6.259115660679482e-05, "data_time": 0.4877237558364868, "loss": 0.0999925646930933, "time": 0.6273519992828369, "epoch": 5, "memory": 6319, "step": 17428} +{"lr": 6.223425869039963e-05, "data_time": 0.5396599292755127, "loss": 0.08791594877839089, "time": 0.6816283702850342, "epoch": 5, "memory": 6319, "step": 17528} +{"lr": 6.187689303389006e-05, "data_time": 0.5210930585861206, "loss": 0.09718217775225639, "time": 0.6614705085754394, "epoch": 5, "memory": 6319, "step": 17628} +{"lr": 6.151908274319292e-05, "data_time": 0.4726128339767456, "loss": 0.08879563137888909, "time": 0.6225663185119629, "epoch": 5, "memory": 6319, "step": 17728} +{"lr": 6.116085095298373e-05, "data_time": 0.501851224899292, "loss": 0.09075678363442422, "time": 0.6425661563873291, "epoch": 5, "memory": 6319, "step": 17828} +{"lr": 6.080222082519007e-05, "data_time": 0.6354441165924072, "loss": 0.10212242342531681, "time": 0.7770376443862915, "epoch": 5, "memory": 6319, "step": 17928} +{"lr": 6.0443215547495147e-05, "data_time": 0.6274643421173096, "loss": 0.09559036567807197, "time": 0.7664098501205444, "epoch": 5, "memory": 6319, "step": 18028} +{"lr": 6.0083858331837714e-05, "data_time": 0.47197983264923093, "loss": 0.08705018796026706, "time": 0.6134809732437134, "epoch": 5, "memory": 6319, "step": 18128} +{"lr": 5.97241724129116e-05, "data_time": 0.5098498582839965, "loss": 0.09013766944408416, "time": 0.6514616966247558, "epoch": 5, "memory": 6319, "step": 18228} +{"lr": 5.936418104666328e-05, "data_time": 0.48030169010162355, "loss": 0.08908655494451523, "time": 0.6206804037094116, "epoch": 5, "memory": 6319, "step": 18328} +{"lr": 5.9003907508788295e-05, "data_time": 0.5214685916900634, "loss": 0.10488158017396927, "time": 0.662332558631897, "epoch": 5, "memory": 6319, "step": 18428} +{"lr": 5.864337509322633e-05, "data_time": 0.5449373722076416, "loss": 0.09434689357876777, "time": 0.6877973794937133, "epoch": 5, "memory": 6319, "step": 18528} +{"lr": 5.8282607110655194e-05, "data_time": 0.5710325956344604, "loss": 0.09347037933766841, "time": 0.7120609760284424, "epoch": 5, "memory": 6319, "step": 18628} +{"lr": 5.7921626886983564e-05, "data_time": 0.5557557106018066, "loss": 0.09960277751088142, "time": 0.6971974611282349, "epoch": 5, "memory": 6319, "step": 18728} +{"lr": 5.756045776184287e-05, "data_time": 0.5279939889907836, "loss": 0.10416858866810799, "time": 0.6684570789337159, "epoch": 5, "memory": 6319, "step": 18828} +{"lr": 5.719912308707808e-05, "data_time": 0.5208239555358887, "loss": 0.10076755955815316, "time": 0.6613128900527954, "epoch": 5, "memory": 6319, "step": 18928} +{"lr": 5.683764622523808e-05, "data_time": 0.5302015542984009, "loss": 0.08578926399350166, "time": 0.6708161354064941, "epoch": 5, "memory": 6319, "step": 19028} +{"lr": 5.647605054806502e-05, "data_time": 0.5026536464691163, "loss": 0.1170566976070404, "time": 0.6428820848464966, "epoch": 5, "memory": 6319, "step": 19128} +{"lr": 5.6114359434983206e-05, "data_time": 0.516049861907959, "loss": 0.09198833629488945, "time": 0.6569303035736084, "epoch": 5, "memory": 6319, "step": 19228} +{"lr": 5.575259627158745e-05, "data_time": 0.5426882743835449, "loss": 0.10285206660628318, "time": 0.6830023527145386, "epoch": 5, "memory": 6319, "step": 19328} +{"lr": 5.539078444813113e-05, "data_time": 0.49720287322998047, "loss": 0.08695899248123169, "time": 0.6380908966064454, "epoch": 5, "memory": 6319, "step": 19428} +{"lr": 5.5028947358013735e-05, "data_time": 0.5174664974212646, "loss": 0.10256444588303566, "time": 0.6577479362487793, "epoch": 5, "memory": 6319, "step": 19528} +{"accuracy/top1": 93.73020935058594, "data_time": 0.6467497266572098, "time": 0.7327331181230217, "step": 5} +{"lr": 5.464178020340108e-05, "data_time": 0.03125379085540771, "loss": 0.10717456564307212, "time": 1.0087543964385985, "epoch": 6, "memory": 6319, "step": 19635} +{"lr": 5.427996514795046e-05, "data_time": 0.19920940399169923, "loss": 0.10471052080392837, "time": 0.534230399131775, "epoch": 6, "memory": 6319, "step": 19735} +{"lr": 5.3918196647260834e-05, "data_time": 0.4795753717422485, "loss": 0.08822899051010609, "time": 0.6257527112960816, "epoch": 6, "memory": 6319, "step": 19835} +{"lr": 5.355649809193057e-05, "data_time": 0.41004362106323244, "loss": 0.07934464812278748, "time": 0.6748523235321044, "epoch": 6, "memory": 6319, "step": 19935} +{"lr": 5.3194892868035765e-05, "data_time": 0.3806777477264404, "loss": 0.0930691808462143, "time": 0.6481860637664795, "epoch": 6, "memory": 6319, "step": 20035} +{"lr": 5.283340435561795e-05, "data_time": 0.5792163610458374, "loss": 0.09030508734285832, "time": 0.7213134765625, "epoch": 6, "memory": 6319, "step": 20135} +{"lr": 5.2472055927172495e-05, "data_time": 0.6683616399765014, "loss": 0.0896085411310196, "time": 0.8090466022491455, "epoch": 6, "memory": 6319, "step": 20235} +{"lr": 5.2110870946137605e-05, "data_time": 0.5130580425262451, "loss": 0.08650345578789712, "time": 0.6541225671768188, "epoch": 6, "memory": 6319, "step": 20335} +{"lr": 5.174987276538342e-05, "data_time": 0.5252459287643433, "loss": 0.09079801067709922, "time": 0.666130542755127, "epoch": 6, "memory": 6319, "step": 20435} +{"lr": 5.1389084725702425e-05, "data_time": 0.5122568368911743, "loss": 0.0950685203075409, "time": 0.6543164253234863, "epoch": 6, "memory": 6319, "step": 20535} +{"lr": 5.1028530154300045e-05, "data_time": 0.5155791997909546, "loss": 0.10416885912418365, "time": 0.6567234992980957, "epoch": 6, "memory": 6319, "step": 20635} +{"lr": 5.066823236328663e-05, "data_time": 0.5147588968276977, "loss": 0.08922100961208343, "time": 0.6560417890548706, "epoch": 6, "memory": 6319, "step": 20735} +{"lr": 5.030821464817006e-05, "data_time": 0.5122957944869995, "loss": 0.08737279959022999, "time": 0.6525972843170166, "epoch": 6, "memory": 6319, "step": 20835} +{"lr": 4.99485002863494e-05, "data_time": 0.5093324661254883, "loss": 0.08328028954565525, "time": 0.6497575759887695, "epoch": 6, "memory": 6319, "step": 20935} +{"lr": 4.95891125356102e-05, "data_time": 0.516583514213562, "loss": 0.08127425834536553, "time": 0.6572388172149658, "epoch": 6, "memory": 6319, "step": 21035} +{"lr": 4.923007463262053e-05, "data_time": 0.5068090915679931, "loss": 0.10989921689033508, "time": 0.6470248222351074, "epoch": 6, "memory": 6319, "step": 21135} +{"lr": 4.887140979142852e-05, "data_time": 0.556620454788208, "loss": 0.09486041627824307, "time": 0.7001865625381469, "epoch": 6, "memory": 6319, "step": 21235} +{"lr": 4.851314120196161e-05, "data_time": 0.5591617822647095, "loss": 0.07852182500064372, "time": 0.6984658002853393, "epoch": 6, "memory": 6319, "step": 21335} +{"lr": 4.815529202852693e-05, "data_time": 0.5204687595367432, "loss": 0.08852827474474907, "time": 0.6720494985580444, "epoch": 6, "memory": 6319, "step": 21435} +{"lr": 4.779788540831406e-05, "data_time": 0.5039268970489502, "loss": 0.08470682874321937, "time": 0.6435454845428467, "epoch": 6, "memory": 6319, "step": 21535} +{"lr": 4.744094444989843e-05, "data_time": 0.5169116020202636, "loss": 0.08685770258307457, "time": 0.6574362277984619, "epoch": 6, "memory": 6319, "step": 21635} +{"lr": 4.708449223174764e-05, "data_time": 0.5160007953643799, "loss": 0.08491603136062623, "time": 0.6593384504318237, "epoch": 6, "memory": 6319, "step": 21735} +{"lr": 4.672855180072916e-05, "data_time": 0.5460444211959838, "loss": 0.08279446847736835, "time": 0.6877310514450073, "epoch": 6, "memory": 6319, "step": 21835} +{"lr": 4.6373146170620256e-05, "data_time": 0.5128842115402221, "loss": 0.09533453360199928, "time": 0.6736371517181396, "epoch": 6, "memory": 6319, "step": 21935} +{"lr": 4.6018298320619915e-05, "data_time": 0.5153392791748047, "loss": 0.0897812344133854, "time": 0.6585339784622193, "epoch": 6, "memory": 6319, "step": 22035} +{"lr": 4.566403119386318e-05, "data_time": 0.48095753192901614, "loss": 0.10277087241411209, "time": 0.6338201284408569, "epoch": 6, "memory": 6319, "step": 22135} +{"lr": 4.531036769593769e-05, "data_time": 0.5006656169891357, "loss": 0.09919351935386658, "time": 0.6422847270965576, "epoch": 6, "memory": 6319, "step": 22235} +{"lr": 4.4957330693402685e-05, "data_time": 0.539228868484497, "loss": 0.0924248743802309, "time": 0.6834451913833618, "epoch": 6, "memory": 6319, "step": 22335} +{"lr": 4.460494301231059e-05, "data_time": 0.5093908548355103, "loss": 0.07978219091892243, "time": 0.6495115995407105, "epoch": 6, "memory": 6319, "step": 22435} +{"lr": 4.425322743673109e-05, "data_time": 0.5398077249526978, "loss": 0.08374065943062306, "time": 0.680783486366272, "epoch": 6, "memory": 6319, "step": 22535} +{"lr": 4.390220670727812e-05, "data_time": 0.4800394058227539, "loss": 0.08907215818762779, "time": 0.6207040786743164, "epoch": 6, "memory": 6319, "step": 22635} +{"lr": 4.3551903519639325e-05, "data_time": 0.5293129444122314, "loss": 0.09509703516960144, "time": 0.6707393646240234, "epoch": 6, "memory": 6319, "step": 22735} +{"lr": 4.3202340523108894e-05, "data_time": 0.5574806690216064, "loss": 0.08660216294229031, "time": 0.6985911130905151, "epoch": 6, "memory": 6319, "step": 22835} +{"lr": 4.285354031912302e-05, "data_time": 0.5237831830978393, "loss": 0.08252441361546517, "time": 0.6644094705581665, "epoch": 6, "memory": 6319, "step": 22935} +{"lr": 4.2505525459798544e-05, "data_time": 0.5262906312942505, "loss": 0.0942851398140192, "time": 0.6670958280563355, "epoch": 6, "memory": 6319, "step": 23035} +{"lr": 4.2158318446474836e-05, "data_time": 0.5269134998321533, "loss": 0.10015843287110329, "time": 0.6679558753967285, "epoch": 6, "memory": 6319, "step": 23135} +{"lr": 4.181194172825905e-05, "data_time": 0.4859327793121338, "loss": 0.09421113654971122, "time": 0.6261427640914917, "epoch": 6, "memory": 6319, "step": 23235} +{"lr": 4.146641770057448e-05, "data_time": 0.4777074813842773, "loss": 0.08744788654148579, "time": 0.6174055576324463, "epoch": 6, "memory": 6319, "step": 23335} +{"lr": 4.112176870371267e-05, "data_time": 0.505276894569397, "loss": 0.07330489903688431, "time": 0.6478984355926514, "epoch": 6, "memory": 6319, "step": 23435} +{"accuracy/top1": 95.14282989501953, "data_time": 0.5447075613613787, "time": 0.6323356463991362, "step": 6} +{"lr": 4.0753988582732786e-05, "data_time": 0.470862889289856, "loss": 0.08295015580952167, "time": 0.6833272933959961, "epoch": 7, "memory": 6319, "step": 23542} +{"lr": 4.0411221639286815e-05, "data_time": 0.48199923038482667, "loss": 0.08625544793903828, "time": 0.6510698318481445, "epoch": 7, "memory": 6319, "step": 23642} +{"lr": 4.006939795169518e-05, "data_time": 0.5034197807312012, "loss": 0.08994064182043075, "time": 0.6623530626296997, "epoch": 7, "memory": 6319, "step": 23742} +{"lr": 3.972853962099913e-05, "data_time": 0.33036906719207765, "loss": 0.09170535504817963, "time": 0.6491644144058227, "epoch": 7, "memory": 6319, "step": 23842} +{"lr": 3.93886686858237e-05, "data_time": 0.24437239170074462, "loss": 0.09314442239701748, "time": 0.6958123683929444, "epoch": 7, "memory": 6319, "step": 23942} +{"lr": 3.904980712095259e-05, "data_time": 0.30084683895111086, "loss": 0.09044926017522811, "time": 0.6798213958740235, "epoch": 7, "memory": 6319, "step": 24042} +{"lr": 3.871197683590748e-05, "data_time": 0.128113317489624, "loss": 0.09334088489413261, "time": 0.6648510932922364, "epoch": 7, "memory": 6319, "step": 24142} +{"lr": 3.837519967353125e-05, "data_time": 0.24887685775756835, "loss": 0.07223951257765293, "time": 0.6324395418167115, "epoch": 7, "memory": 6319, "step": 24242} +{"lr": 3.8039497408575855e-05, "data_time": 0.3246570348739624, "loss": 0.08753116950392723, "time": 0.6473471879959106, "epoch": 7, "memory": 6319, "step": 24342} +{"lr": 3.770489174629432e-05, "data_time": 0.28405938148498533, "loss": 0.10265554189682007, "time": 0.6407233476638794, "epoch": 7, "memory": 6319, "step": 24442} +{"lr": 3.737140432103754e-05, "data_time": 0.322407603263855, "loss": 0.09392819106578827, "time": 0.7046275615692139, "epoch": 7, "memory": 6319, "step": 24542} +{"lr": 3.703905669485545e-05, "data_time": 0.3817084074020386, "loss": 0.08836689218878746, "time": 0.662895917892456, "epoch": 7, "memory": 6319, "step": 24642} +{"lr": 3.6707870356102706e-05, "data_time": 0.35480873584747313, "loss": 0.09067027866840363, "time": 0.6342610359191895, "epoch": 7, "memory": 6319, "step": 24742} +{"lr": 3.6377866718049484e-05, "data_time": 0.5103625059127808, "loss": 0.09932314157485962, "time": 0.6865311145782471, "epoch": 7, "memory": 6319, "step": 24842} +{"lr": 3.6049067117496965e-05, "data_time": 0.5267199277877808, "loss": 0.09514525346457958, "time": 0.6687266588211059, "epoch": 7, "memory": 6319, "step": 24942} +{"lr": 3.5721492813397755e-05, "data_time": 0.5122140407562256, "loss": 0.0813816886395216, "time": 0.6517822027206421, "epoch": 7, "memory": 6319, "step": 25042} +{"lr": 3.539516498548151e-05, "data_time": 0.5005153179168701, "loss": 0.0832541186362505, "time": 0.6636749267578125, "epoch": 7, "memory": 6319, "step": 25142} +{"lr": 3.507010473288518e-05, "data_time": 0.4330591678619385, "loss": 0.0813992377370596, "time": 0.6964970588684082, "epoch": 7, "memory": 6319, "step": 25242} +{"lr": 3.474633307278909e-05, "data_time": 0.22276041507720948, "loss": 0.08509134873747826, "time": 0.6273354291915894, "epoch": 7, "memory": 6319, "step": 25342} +{"lr": 3.4423870939058177e-05, "data_time": 0.3570723533630371, "loss": 0.07953443638980388, "time": 0.6682219982147217, "epoch": 7, "memory": 6319, "step": 25442} +{"lr": 3.410273918088805e-05, "data_time": 0.20250792503356935, "loss": 0.09414939917623996, "time": 0.6555251598358154, "epoch": 7, "memory": 6319, "step": 25542} +{"lr": 3.37829585614574e-05, "data_time": 0.1938471794128418, "loss": 0.07364090755581856, "time": 0.6488155364990235, "epoch": 7, "memory": 6319, "step": 25642} +{"lr": 3.3464549756585216e-05, "data_time": 0.02102198600769043, "loss": 0.08286008350551129, "time": 0.6763957977294922, "epoch": 7, "memory": 6319, "step": 25742} +{"lr": 3.314753335339416e-05, "data_time": 0.0011744976043701171, "loss": 0.09127445928752423, "time": 0.6984846353530884, "epoch": 7, "memory": 6319, "step": 25842} +{"lr": 3.2831929848979394e-05, "data_time": 0.0012300491333007812, "loss": 0.07671383023262024, "time": 0.6535136699676514, "epoch": 7, "memory": 6319, "step": 25942} +{"lr": 3.251775964908332e-05, "data_time": 0.001391458511352539, "loss": 0.07746944688260556, "time": 0.6931942939758301, "epoch": 7, "memory": 6319, "step": 26042} +{"lr": 3.22050430667762e-05, "data_time": 0.0012549161911010742, "loss": 0.07638381943106651, "time": 0.633611798286438, "epoch": 7, "memory": 6319, "step": 26142} +{"lr": 3.189380032114285e-05, "data_time": 0.10946221351623535, "loss": 0.07504312098026275, "time": 0.6317307710647583, "epoch": 7, "memory": 6319, "step": 26242} +{"lr": 3.1584051535975324e-05, "data_time": 0.21273183822631836, "loss": 0.09941074028611183, "time": 0.6449794054031373, "epoch": 7, "memory": 6319, "step": 26342} +{"lr": 3.127581673847175e-05, "data_time": 0.21001415252685546, "loss": 0.08890110403299331, "time": 0.671927809715271, "epoch": 7, "memory": 6319, "step": 26442} +{"lr": 3.0969115857941555e-05, "data_time": 0.20174038410186768, "loss": 0.0774132028222084, "time": 0.6450658798217773, "epoch": 7, "memory": 6319, "step": 26542} +{"lr": 3.0663968724516654e-05, "data_time": 0.05780491828918457, "loss": 0.09331877008080483, "time": 0.6313152551651001, "epoch": 7, "memory": 6319, "step": 26642} +{"lr": 3.0360395067869668e-05, "data_time": 0.03669705390930176, "loss": 0.09191939383745193, "time": 0.614667558670044, "epoch": 7, "memory": 6319, "step": 26742} +{"lr": 3.0058414515938063e-05, "data_time": 0.02508423328399658, "loss": 0.08378423005342484, "time": 0.6930113077163697, "epoch": 7, "memory": 6319, "step": 26842} +{"lr": 2.975804659365501e-05, "data_time": 0.0022589445114135744, "loss": 0.08855830878019333, "time": 0.6762532711029052, "epoch": 7, "memory": 6319, "step": 26942} +{"lr": 2.945931072168714e-05, "data_time": 0.0016698360443115235, "loss": 0.08498649261891841, "time": 0.6263153076171875, "epoch": 7, "memory": 6319, "step": 27042} +{"lr": 2.9162226215178855e-05, "data_time": 0.0010523557662963866, "loss": 0.08232737556099892, "time": 0.6991947889328003, "epoch": 7, "memory": 6319, "step": 27142} +{"lr": 2.8866812282503322e-05, "data_time": 0.001619267463684082, "loss": 0.08074983358383178, "time": 0.6734080791473389, "epoch": 7, "memory": 6319, "step": 27242} +{"lr": 2.8573088024020838e-05, "data_time": 0.0008196592330932617, "loss": 0.08293840140104294, "time": 0.6396318435668945, "epoch": 7, "memory": 6319, "step": 27342} +{"accuracy/top1": 95.87696838378906, "data_time": 0.513101314676219, "time": 0.6030006696437967, "step": 7} +{"lr": 2.8260695817390092e-05, "data_time": 0.5127811431884766, "loss": 0.07460867390036582, "time": 0.6538702487945557, "epoch": 8, "memory": 6319, "step": 27449} +{"lr": 2.7970529402570967e-05, "data_time": 0.4982527494430542, "loss": 0.08210007287561893, "time": 0.6468517303466796, "epoch": 8, "memory": 6319, "step": 27549} +{"lr": 2.7682110612244214e-05, "data_time": 0.5047098875045777, "loss": 0.08655336350202561, "time": 0.651490330696106, "epoch": 8, "memory": 6319, "step": 27649} +{"lr": 2.7395458094490283e-05, "data_time": 0.4774590015411377, "loss": 0.10557109713554383, "time": 0.6198263168334961, "epoch": 8, "memory": 6319, "step": 27749} +{"lr": 2.7110590383189015e-05, "data_time": 0.5294427871704102, "loss": 0.09038270860910416, "time": 0.6706011295318604, "epoch": 8, "memory": 6319, "step": 27849} +{"lr": 2.6827525896821397e-05, "data_time": 0.5397517681121826, "loss": 0.09813364967703819, "time": 0.6807192325592041, "epoch": 8, "memory": 6319, "step": 27949} +{"lr": 2.6546282937278552e-05, "data_time": 0.44946396350860596, "loss": 0.09082960337400436, "time": 0.6686970710754394, "epoch": 8, "memory": 6319, "step": 28049} +{"lr": 2.6266879688678606e-05, "data_time": 0.5386229515075683, "loss": 0.09124983251094818, "time": 0.6814395904541015, "epoch": 8, "memory": 6319, "step": 28149} +{"lr": 2.5989334216190738e-05, "data_time": 0.5184274673461914, "loss": 0.10565513707697391, "time": 0.6589367389678955, "epoch": 8, "memory": 6319, "step": 28249} +{"lr": 2.571366446486732e-05, "data_time": 0.5535402059555053, "loss": 0.10147601626813411, "time": 0.6964458227157593, "epoch": 8, "memory": 6319, "step": 28349} +{"lr": 2.5439888258483665e-05, "data_time": 0.5335720777511597, "loss": 0.0966638259589672, "time": 0.673550796508789, "epoch": 8, "memory": 6319, "step": 28449} +{"lr": 2.5168023298385384e-05, "data_time": 0.5342338562011719, "loss": 0.07630592696368695, "time": 0.6762902975082398, "epoch": 8, "memory": 6319, "step": 28549} +{"lr": 2.489808716234412e-05, "data_time": 0.5069514274597168, "loss": 0.08626412823796273, "time": 0.6497405290603637, "epoch": 8, "memory": 6319, "step": 28649} +{"lr": 2.463009730342111e-05, "data_time": 0.4845763921737671, "loss": 0.0779623631387949, "time": 0.6296336412429809, "epoch": 8, "memory": 6319, "step": 28749} +{"lr": 2.436407104883849e-05, "data_time": 0.5296821832656861, "loss": 0.07111451178789138, "time": 0.6721384048461914, "epoch": 8, "memory": 6319, "step": 28849} +{"lr": 2.4100025598859092e-05, "data_time": 0.5019245386123657, "loss": 0.08655482605099678, "time": 0.6427699089050293, "epoch": 8, "memory": 6319, "step": 28949} +{"lr": 2.3837978025674284e-05, "data_time": 0.4973599910736084, "loss": 0.0867620412260294, "time": 0.6369682073593139, "epoch": 8, "memory": 6319, "step": 29049} +{"lr": 2.357794527230027e-05, "data_time": 0.5338386535644531, "loss": 0.08663633055984973, "time": 0.6740134716033935, "epoch": 8, "memory": 6319, "step": 29149} +{"lr": 2.33199441514825e-05, "data_time": 0.5322182655334473, "loss": 0.06905041374266148, "time": 0.6760332584381104, "epoch": 8, "memory": 6319, "step": 29249} +{"lr": 2.306399134460859e-05, "data_time": 0.5384789705276489, "loss": 0.08724874705076217, "time": 0.679179048538208, "epoch": 8, "memory": 6319, "step": 29349} +{"lr": 2.281010340062995e-05, "data_time": 0.5234551906585694, "loss": 0.07879753299057483, "time": 0.6651687622070312, "epoch": 8, "memory": 6319, "step": 29449} +{"lr": 2.2558296734991603e-05, "data_time": 0.48877928256988523, "loss": 0.08786786571145058, "time": 0.6288945436477661, "epoch": 8, "memory": 6319, "step": 29549} +{"lr": 2.2308587628570915e-05, "data_time": 0.4862042427062988, "loss": 0.08550646640360356, "time": 0.6295605897903442, "epoch": 8, "memory": 6319, "step": 29649} +{"lr": 2.2060992226624955e-05, "data_time": 0.5055670261383056, "loss": 0.07987292930483818, "time": 0.6510406017303467, "epoch": 8, "memory": 6319, "step": 29749} +{"lr": 2.181552653774648e-05, "data_time": 0.504708981513977, "loss": 0.08791458755731582, "time": 0.6673235893249512, "epoch": 8, "memory": 6319, "step": 29849} +{"lr": 2.1572206432829035e-05, "data_time": 0.5137610673904419, "loss": 0.07938118763267994, "time": 0.6555030345916748, "epoch": 8, "memory": 6319, "step": 29949} +{"lr": 2.1331047644040754e-05, "data_time": 0.5457321643829346, "loss": 0.08348729908466339, "time": 0.6873914003372192, "epoch": 8, "memory": 6319, "step": 30049} +{"lr": 2.109206576380705e-05, "data_time": 0.47716259956359863, "loss": 0.07699331119656563, "time": 0.6198265314102173, "epoch": 8, "memory": 6319, "step": 30149} +{"lr": 2.0855276243802687e-05, "data_time": 0.4770580530166626, "loss": 0.0768574696034193, "time": 0.6415684223175049, "epoch": 8, "memory": 6319, "step": 30249} +{"lr": 2.0620694393952487e-05, "data_time": 0.4538614511489868, "loss": 0.08800110891461373, "time": 0.5951207399368286, "epoch": 8, "memory": 6319, "step": 30349} +{"lr": 2.038833538144169e-05, "data_time": 0.5135487794876099, "loss": 0.09790711887180806, "time": 0.655436372756958, "epoch": 8, "memory": 6319, "step": 30449} +{"lr": 2.0158214229735154e-05, "data_time": 0.4786839485168457, "loss": 0.0819152183830738, "time": 0.621023154258728, "epoch": 8, "memory": 6319, "step": 30549} +{"lr": 1.9930345817605977e-05, "data_time": 0.4796322822570801, "loss": 0.06708877757191659, "time": 0.6211246967315673, "epoch": 8, "memory": 6319, "step": 30649} +{"lr": 1.970474487817358e-05, "data_time": 0.5479300737380981, "loss": 0.07790469005703926, "time": 0.7155184984207154, "epoch": 8, "memory": 6319, "step": 30749} +{"lr": 1.9481425997951092e-05, "data_time": 0.48464412689208985, "loss": 0.08609076626598836, "time": 0.6287136554718018, "epoch": 8, "memory": 6319, "step": 30849} +{"lr": 1.9260403615902187e-05, "data_time": 0.50627760887146, "loss": 0.08874175995588303, "time": 0.6590127229690552, "epoch": 8, "memory": 6319, "step": 30949} +{"lr": 1.9041692022507647e-05, "data_time": 0.481411337852478, "loss": 0.08259556405246257, "time": 0.6220230340957642, "epoch": 8, "memory": 6319, "step": 31049} +{"lr": 1.8825305358841125e-05, "data_time": 0.5230817794799805, "loss": 0.06932475790381432, "time": 0.6637158632278443, "epoch": 8, "memory": 6319, "step": 31149} +{"lr": 1.861125761565518e-05, "data_time": 0.49395802021026614, "loss": 0.08681600689888, "time": 0.6353115797042846, "epoch": 8, "memory": 6319, "step": 31249} +{"accuracy/top1": 96.5067367553711, "data_time": 0.5387871429837984, "time": 0.6271591474269999, "step": 8} +{"lr": 1.8384832449122864e-05, "data_time": 0.45036723613739016, "loss": 0.0774447601288557, "time": 0.5908416509628296, "epoch": 9, "memory": 6319, "step": 31356} +{"lr": 1.817567007332114e-05, "data_time": 0.4683086395263672, "loss": 0.08455916084349155, "time": 0.6101473569869995, "epoch": 9, "memory": 6319, "step": 31456} +{"lr": 1.796888862098825e-05, "data_time": 0.4808666229248047, "loss": 0.08091538585722446, "time": 0.620784068107605, "epoch": 9, "memory": 6319, "step": 31556} +{"lr": 1.7764501461839678e-05, "data_time": 0.49072086811065674, "loss": 0.08568033501505852, "time": 0.6317744016647339, "epoch": 9, "memory": 6319, "step": 31656} +{"lr": 1.7562521810784746e-05, "data_time": 0.5133663654327393, "loss": 0.07859389409422875, "time": 0.654290747642517, "epoch": 9, "memory": 6319, "step": 31756} +{"lr": 1.7362962727072444e-05, "data_time": 0.48868684768676757, "loss": 0.0888303942978382, "time": 0.6742319822311401, "epoch": 9, "memory": 6319, "step": 31856} +{"lr": 1.716583711344682e-05, "data_time": 0.49153151512146, "loss": 0.08321366459131241, "time": 0.6329559087753296, "epoch": 9, "memory": 6319, "step": 31956} +{"lr": 1.6971157715312954e-05, "data_time": 0.500112795829773, "loss": 0.07347173951566219, "time": 0.6415749311447143, "epoch": 9, "memory": 6319, "step": 32056} +{"lr": 1.677893711991274e-05, "data_time": 0.4580115795135498, "loss": 0.0807027492672205, "time": 0.6126995801925659, "epoch": 9, "memory": 6319, "step": 32156} +{"lr": 1.6589187755511025e-05, "data_time": 0.5103238582611084, "loss": 0.078684401512146, "time": 0.66378173828125, "epoch": 9, "memory": 6319, "step": 32256} +{"lr": 1.640192189059216e-05, "data_time": 0.49948337078094485, "loss": 0.0780689612030983, "time": 0.705699110031128, "epoch": 9, "memory": 6319, "step": 32356} +{"lr": 1.6217151633066716e-05, "data_time": 0.44858882427215574, "loss": 0.08355204723775386, "time": 0.708412766456604, "epoch": 9, "memory": 6319, "step": 32456} +{"lr": 1.6034888929488564e-05, "data_time": 0.5895434856414795, "loss": 0.08971339166164398, "time": 0.7298603057861328, "epoch": 9, "memory": 6319, "step": 32556} +{"lr": 1.585514556428258e-05, "data_time": 0.5000896453857422, "loss": 0.07340927235782146, "time": 0.6414830207824707, "epoch": 9, "memory": 6319, "step": 32656} +{"lr": 1.567793315898261e-05, "data_time": 0.5831059217453003, "loss": 0.0837759081274271, "time": 0.7241213321685791, "epoch": 9, "memory": 6319, "step": 32756} +{"lr": 1.5503263171480048e-05, "data_time": 0.5291210174560547, "loss": 0.09053281843662261, "time": 0.6703447818756103, "epoch": 9, "memory": 6319, "step": 32856} +{"lr": 1.5331146895283165e-05, "data_time": 0.519048810005188, "loss": 0.08845474012196064, "time": 0.6597179651260376, "epoch": 9, "memory": 6319, "step": 32956} +{"lr": 1.5161595458786728e-05, "data_time": 0.5170748472213745, "loss": 0.09481598138809204, "time": 0.657964038848877, "epoch": 9, "memory": 6319, "step": 33056} +{"lr": 1.4994619824552563e-05, "data_time": 0.4895815849304199, "loss": 0.08049409575760365, "time": 0.6292577266693116, "epoch": 9, "memory": 6319, "step": 33156} +{"lr": 1.4830230788600802e-05, "data_time": 0.48902390003204343, "loss": 0.08306508734822274, "time": 0.6349939823150634, "epoch": 9, "memory": 6319, "step": 33256} +{"lr": 1.4668438979711779e-05, "data_time": 0.5043458700180053, "loss": 0.10058392062783242, "time": 0.6451935052871705, "epoch": 9, "memory": 6319, "step": 33356} +{"lr": 1.4509254858738844e-05, "data_time": 0.48384382724761965, "loss": 0.07483227849006653, "time": 0.6241837739944458, "epoch": 9, "memory": 6319, "step": 33456} +{"lr": 1.4352688717931966e-05, "data_time": 0.4971457481384277, "loss": 0.08895487524569035, "time": 0.6370618104934692, "epoch": 9, "memory": 6319, "step": 33556} +{"lr": 1.4198750680272348e-05, "data_time": 0.5083073616027832, "loss": 0.09322233945131302, "time": 0.6478856086730957, "epoch": 9, "memory": 6319, "step": 33656} +{"lr": 1.4047450698817857e-05, "data_time": 0.4894073963165283, "loss": 0.07951398901641368, "time": 0.6290619373321533, "epoch": 9, "memory": 6319, "step": 33756} +{"lr": 1.389879855605952e-05, "data_time": 0.4867011308670044, "loss": 0.0786565899848938, "time": 0.6274277210235596, "epoch": 9, "memory": 6319, "step": 33856} +{"lr": 1.3752803863288976e-05, "data_time": 0.5118851661682129, "loss": 0.08616845607757569, "time": 0.6523105144500733, "epoch": 9, "memory": 6319, "step": 33956} +{"lr": 1.3609476059977129e-05, "data_time": 0.5192718029022216, "loss": 0.07865990065038204, "time": 0.6592938423156738, "epoch": 9, "memory": 6319, "step": 34056} +{"lr": 1.3468824413163754e-05, "data_time": 0.5041179180145263, "loss": 0.08080047145485877, "time": 0.6560468435287475, "epoch": 9, "memory": 6319, "step": 34156} +{"lr": 1.3330858016858395e-05, "data_time": 0.45170934200286866, "loss": 0.08205296695232392, "time": 0.5917628526687622, "epoch": 9, "memory": 6319, "step": 34256} +{"lr": 1.3195585791452309e-05, "data_time": 0.5187618970870972, "loss": 0.07742132656276227, "time": 0.6678142786026001, "epoch": 9, "memory": 6319, "step": 34356} +{"lr": 1.3063016483141729e-05, "data_time": 0.4949553728103638, "loss": 0.08371957205235958, "time": 0.6356351613998413, "epoch": 9, "memory": 6319, "step": 34456} +{"lr": 1.2933158663362388e-05, "data_time": 0.4888258934020996, "loss": 0.08621455803513527, "time": 0.6414131641387939, "epoch": 9, "memory": 6319, "step": 34556} +{"lr": 1.2806020728235311e-05, "data_time": 0.594412112236023, "loss": 0.08536653108894825, "time": 0.7345845222473144, "epoch": 9, "memory": 6319, "step": 34656} +{"lr": 1.2681610898023963e-05, "data_time": 0.5329106092453003, "loss": 0.09632372669875622, "time": 0.6734464645385743, "epoch": 9, "memory": 6319, "step": 34756} +{"lr": 1.2559937216602731e-05, "data_time": 0.5210883617401123, "loss": 0.07627506218850613, "time": 0.662468695640564, "epoch": 9, "memory": 6319, "step": 34856} +{"lr": 1.2441007550936847e-05, "data_time": 0.5392831802368164, "loss": 0.07872199155390262, "time": 0.6801229238510131, "epoch": 9, "memory": 6319, "step": 34956} +{"lr": 1.232482959057377e-05, "data_time": 0.5266360759735107, "loss": 0.0835390716791153, "time": 0.6657192230224609, "epoch": 9, "memory": 6319, "step": 35056} +{"lr": 1.2211410847145969e-05, "data_time": 0.5051618576049804, "loss": 0.08262104205787182, "time": 0.6445920467376709, "epoch": 9, "memory": 6319, "step": 35156} +{"accuracy/top1": 96.47541809082031, "data_time": 0.518794454377273, "time": 0.608899108294783, "step": 9} +{"lr": 1.209311679360931e-05, "data_time": 0.5300727128982544, "loss": 0.08774049505591393, "time": 0.6700115919113159, "epoch": 10, "memory": 6319, "step": 35263} +{"lr": 1.1985432727488666e-05, "data_time": 0.5218451499938965, "loss": 0.08216948658227921, "time": 0.6757283926010131, "epoch": 10, "memory": 6319, "step": 35363} +{"lr": 1.1880529822435289e-05, "data_time": 0.5013918161392212, "loss": 0.07124878317117692, "time": 0.6406710147857666, "epoch": 10, "memory": 6319, "step": 35463} +{"lr": 1.1778414861078726e-05, "data_time": 0.47354414463043215, "loss": 0.08191103041172028, "time": 0.6248251914978027, "epoch": 10, "memory": 6319, "step": 35563} +{"lr": 1.1679094445790538e-05, "data_time": 0.4894792795181274, "loss": 0.08240576237440109, "time": 0.6288114786148071, "epoch": 10, "memory": 6319, "step": 35663} +{"lr": 1.1582574998257377e-05, "data_time": 0.536252498626709, "loss": 0.08340882994234562, "time": 0.6897312641143799, "epoch": 10, "memory": 6319, "step": 35763} +{"lr": 1.1488862759065762e-05, "data_time": 0.5399514675140381, "loss": 0.08530356548726559, "time": 0.6838498592376709, "epoch": 10, "memory": 6319, "step": 35863} +{"lr": 1.1397963787298665e-05, "data_time": 0.599949049949646, "loss": 0.09712032303214073, "time": 0.7402030467987061, "epoch": 10, "memory": 6319, "step": 35963} +{"lr": 1.1309883960143664e-05, "data_time": 0.5113953351974487, "loss": 0.08439582176506519, "time": 0.6521112442016601, "epoch": 10, "memory": 6319, "step": 36063} +{"lr": 1.122462897251302e-05, "data_time": 0.5624451398849487, "loss": 0.07773884236812592, "time": 0.7040619611740112, "epoch": 10, "memory": 6319, "step": 36163} +{"lr": 1.114220433667542e-05, "data_time": 0.5429295301437378, "loss": 0.08018908873200417, "time": 0.683778953552246, "epoch": 10, "memory": 6319, "step": 36263} +{"lr": 1.10626153818996e-05, "data_time": 0.4776179790496826, "loss": 0.089403111115098, "time": 0.618587875366211, "epoch": 10, "memory": 6319, "step": 36363} +{"lr": 1.0985867254109761e-05, "data_time": 0.5083762407302856, "loss": 0.08239412233233452, "time": 0.6485461473464966, "epoch": 10, "memory": 6319, "step": 36463} +{"lr": 1.0911964915552823e-05, "data_time": 0.4959545612335205, "loss": 0.09672980047762395, "time": 0.6369888305664062, "epoch": 10, "memory": 6319, "step": 36563} +{"lr": 1.0840913144477643e-05, "data_time": 0.5138628721237183, "loss": 0.07901668697595596, "time": 0.6529916286468506, "epoch": 10, "memory": 6319, "step": 36663} +{"lr": 1.0772716534826064e-05, "data_time": 0.4700314998626709, "loss": 0.08116241209208966, "time": 0.6110960245132446, "epoch": 10, "memory": 6319, "step": 36763} +{"lr": 1.0707379495935828e-05, "data_time": 0.537388277053833, "loss": 0.0771650567650795, "time": 0.6768936395645142, "epoch": 10, "memory": 6319, "step": 36863} +{"lr": 1.0644906252255541e-05, "data_time": 0.5458524703979493, "loss": 0.06829882338643074, "time": 0.6856985569000245, "epoch": 10, "memory": 6319, "step": 36963} +{"lr": 1.0585300843071527e-05, "data_time": 0.5325053215026856, "loss": 0.06929968297481537, "time": 0.6731296062469483, "epoch": 10, "memory": 6319, "step": 37063} +{"lr": 1.0528567122246654e-05, "data_time": 0.5284550189971924, "loss": 0.07455556318163872, "time": 0.6679382801055909, "epoch": 10, "memory": 6319, "step": 37163} +{"lr": 1.0474708757971141e-05, "data_time": 0.4974109172821045, "loss": 0.08358317576348781, "time": 0.6382056713104248, "epoch": 10, "memory": 6319, "step": 37263} +{"lr": 1.042372923252543e-05, "data_time": 0.4840532302856445, "loss": 0.08695609383285045, "time": 0.624877643585205, "epoch": 10, "memory": 6319, "step": 37363} +{"lr": 1.0375631842054998e-05, "data_time": 0.5609840869903564, "loss": 0.0942134540528059, "time": 0.702863335609436, "epoch": 10, "memory": 6319, "step": 37463} +{"lr": 1.0330419696357254e-05, "data_time": 0.5242627382278442, "loss": 0.07821050956845284, "time": 0.6635823011398315, "epoch": 10, "memory": 6319, "step": 37563} +{"lr": 1.028809571868047e-05, "data_time": 0.5104862451553345, "loss": 0.08879930265247822, "time": 0.6513925552368164, "epoch": 10, "memory": 6319, "step": 37663} +{"lr": 1.0248662645534766e-05, "data_time": 0.5124234437942505, "loss": 0.08167126402258873, "time": 0.6535289525985718, "epoch": 10, "memory": 6319, "step": 37763} +{"lr": 1.0212123026515194e-05, "data_time": 0.5410007953643798, "loss": 0.08757947683334351, "time": 0.6814837694168091, "epoch": 10, "memory": 6319, "step": 37863} +{"lr": 1.0178479224136881e-05, "data_time": 0.532672643661499, "loss": 0.08730786256492137, "time": 0.6741935253143311, "epoch": 10, "memory": 6319, "step": 37963} +{"lr": 1.0147733413682289e-05, "data_time": 0.5444990396499634, "loss": 0.08508696109056473, "time": 0.6847944974899292, "epoch": 10, "memory": 6319, "step": 38063} +{"lr": 1.0119887583060527e-05, "data_time": 0.6378956317901612, "loss": 0.08408012986183167, "time": 0.7786592006683349, "epoch": 10, "memory": 6319, "step": 38163} +{"lr": 1.0094943532678871e-05, "data_time": 0.5177739858627319, "loss": 0.09481669887900353, "time": 0.6586864709854126, "epoch": 10, "memory": 6319, "step": 38263} +{"lr": 1.007290287532637e-05, "data_time": 0.5099339962005616, "loss": 0.09078535623848438, "time": 0.6501465559005737, "epoch": 10, "memory": 6319, "step": 38363} +{"lr": 1.0053767036069495e-05, "data_time": 0.5393283367156982, "loss": 0.07345308773219586, "time": 0.679974102973938, "epoch": 10, "memory": 6319, "step": 38463} +{"lr": 1.003753725216005e-05, "data_time": 0.5076316356658935, "loss": 0.08990654423832893, "time": 0.648727560043335, "epoch": 10, "memory": 6319, "step": 38563} +{"lr": 1.0024214572955152e-05, "data_time": 0.5073602676391602, "loss": 0.08032020516693592, "time": 0.6476858854293823, "epoch": 10, "memory": 6319, "step": 38663} +{"lr": 1.0013799859849456e-05, "data_time": 0.5041482448577881, "loss": 0.08838330432772637, "time": 0.6441276550292969, "epoch": 10, "memory": 6319, "step": 38763} +{"lr": 1.0006293786219338e-05, "data_time": 0.5216606855392456, "loss": 0.0832119818776846, "time": 0.6623240947723389, "epoch": 10, "memory": 6319, "step": 38863} +{"lr": 1.0001696837379472e-05, "data_time": 0.5376900196075439, "loss": 0.08422540500760078, "time": 0.6791685342788696, "epoch": 10, "memory": 6319, "step": 38963} +{"lr": 1.0000009310551361e-05, "data_time": 0.5254184484481812, "loss": 0.0833253562450409, "time": 0.6661343574523926, "epoch": 10, "memory": 6319, "step": 39063} +{"accuracy/top1": 96.86858367919922, "data_time": 0.5282502462124002, "time": 0.6142095245164017, "step": 10} diff --git a/resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1/20230604_225351/vis_data/config.py b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1/20230604_225351/vis_data/config.py new file mode 100644 index 0000000000000000000000000000000000000000..6664519846f1271ef814a9610846d4ebd7a6cc83 --- /dev/null +++ b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1/20230604_225351/vis_data/config.py @@ -0,0 +1,230 @@ +optim_wrapper = dict( + optimizer=dict( + type='SGD', + lr=0.0001, + momentum=0.9, + weight_decay=0.0001, + _scope_='mmpretrain'), + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=False, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=512) +model = dict( + type='ImageClassifier', + backbone=dict( + frozen_stages=2, + type='ResNet', + depth=50, + num_stages=4, + out_indices=(3, ), + style='pytorch', + init_cfg=dict( + type='Pretrained', + checkpoint= + 'https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth', + prefix='backbone')), + neck=dict(type='GlobalAveragePooling'), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=2048, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + topk=1)) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stablediffusionV2-1-dpmsolver-25-1m.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/cc1m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV2-1-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = dict(type='Accuracy', topk=1) +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV2-1-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = dict(type='Accuracy', topk=1) +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1' diff --git a/resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1/20230604_225351/vis_data/events.out.tfevents.1685890446.SH-IDC1-10-140-24-88.145983.0 b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1/20230604_225351/vis_data/events.out.tfevents.1685890446.SH-IDC1-10-140-24-88.145983.0 new file mode 100644 index 0000000000000000000000000000000000000000..5f87c18ec2bad91fdede425511eadc2d79f0b896 --- /dev/null +++ b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1/20230604_225351/vis_data/events.out.tfevents.1685890446.SH-IDC1-10-140-24-88.145983.0 @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:4ca54aea0b7b8703719eb66a0c2a45ab07c1f8be8694e87c2db9dbdf40356245 +size 2584499 diff --git a/resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1/20230604_225351/vis_data/scalars.json b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1/20230604_225351/vis_data/scalars.json new file mode 100644 index 0000000000000000000000000000000000000000..304f3697b6d295f35f2ceef96ae889bd834b54c7 --- /dev/null +++ b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1/20230604_225351/vis_data/scalars.json @@ -0,0 +1,400 @@ +{"lr": 9.999857418382631e-05, "data_time": 0.25231518745422366, "loss": 0.6043034374713898, "time": 0.6773250818252563, "epoch": 1, "memory": 9436, "step": 100} +{"lr": 9.999423907359506e-05, "data_time": 0.0753288745880127, "loss": 0.53575439453125, "time": 0.7267685651779174, "epoch": 1, "memory": 6319, "step": 200} +{"lr": 9.998699480412073e-05, "data_time": 0.0010489463806152345, "loss": 0.49411348402500155, "time": 0.6230251789093018, "epoch": 1, "memory": 6319, "step": 300} +{"lr": 9.997684184379085e-05, "data_time": 0.000860142707824707, "loss": 0.42975725531578063, "time": 0.6354247808456421, "epoch": 1, "memory": 6319, "step": 400} +{"lr": 9.996378084905784e-05, "data_time": 0.0007945775985717773, "loss": 0.41185124814510343, "time": 0.7114021062850953, "epoch": 1, "memory": 6319, "step": 500} +{"lr": 9.994781266439667e-05, "data_time": 0.0011701345443725585, "loss": 0.3755955159664154, "time": 0.6573834180831909, "epoch": 1, "memory": 6319, "step": 600} +{"lr": 9.992893832225061e-05, "data_time": 0.0008877992630004882, "loss": 0.3354937732219696, "time": 0.6607741355895996, "epoch": 1, "memory": 6319, "step": 700} +{"lr": 9.990715904296405e-05, "data_time": 0.001269364356994629, "loss": 0.3326795011758804, "time": 0.6426759719848633, "epoch": 1, "memory": 6319, "step": 800} +{"lr": 9.988247623470371e-05, "data_time": 0.0008656740188598633, "loss": 0.30257115662097933, "time": 0.6511765241622924, "epoch": 1, "memory": 6319, "step": 900} +{"lr": 9.985489149336749e-05, "data_time": 0.000989818572998047, "loss": 0.2756628647446632, "time": 0.690179967880249, "epoch": 1, "memory": 6319, "step": 1000} +{"lr": 9.982440660248187e-05, "data_time": 0.0009242057800292969, "loss": 0.2721254274249077, "time": 0.7048046827316284, "epoch": 1, "memory": 6319, "step": 1100} +{"lr": 9.979102353308574e-05, "data_time": 0.0010079145431518555, "loss": 0.26231861114501953, "time": 0.6478001832962036, "epoch": 1, "memory": 6319, "step": 1200} +{"lr": 9.975474444360358e-05, "data_time": 0.0008924007415771484, "loss": 0.25315001159906386, "time": 0.6526968002319335, "epoch": 1, "memory": 6319, "step": 1300} +{"lr": 9.97155716797056e-05, "data_time": 0.0009036779403686524, "loss": 0.27018743604421613, "time": 0.7147150754928588, "epoch": 1, "memory": 6319, "step": 1400} +{"lr": 9.967350777415632e-05, "data_time": 0.0007965564727783203, "loss": 0.24654453098773957, "time": 0.6703690528869629, "epoch": 1, "memory": 6319, "step": 1500} +{"lr": 9.962855544665059e-05, "data_time": 0.001060152053833008, "loss": 0.23762782663106918, "time": 0.6793582677841187, "epoch": 1, "memory": 6319, "step": 1600} +{"lr": 9.958071760363773e-05, "data_time": 0.0008116006851196289, "loss": 0.22109235376119613, "time": 0.6407901525497437, "epoch": 1, "memory": 6319, "step": 1700} +{"lr": 9.952999733813382e-05, "data_time": 0.10425570011138915, "loss": 0.23239289820194245, "time": 0.6414692878723145, "epoch": 1, "memory": 6319, "step": 1800} +{"lr": 9.947639792952148e-05, "data_time": 0.5308783292770386, "loss": 0.21411006152629852, "time": 0.6705950498580933, "epoch": 1, "memory": 6319, "step": 1900} +{"lr": 9.941992284333813e-05, "data_time": 0.500979995727539, "loss": 0.22601257115602494, "time": 0.7200746774673462, "epoch": 1, "memory": 6319, "step": 2000} +{"lr": 9.93605757310516e-05, "data_time": 0.5712074756622314, "loss": 0.2068329483270645, "time": 0.7120099782943725, "epoch": 1, "memory": 6319, "step": 2100} +{"lr": 9.929836042982421e-05, "data_time": 0.5833227157592773, "loss": 0.19560886919498444, "time": 0.7229428052902221, "epoch": 1, "memory": 6319, "step": 2200} +{"lr": 9.923328096226467e-05, "data_time": 0.5348201990127563, "loss": 0.2159124195575714, "time": 0.6739555835723877, "epoch": 1, "memory": 6319, "step": 2300} +{"lr": 9.916534153616819e-05, "data_time": 0.49794604778289797, "loss": 0.19221736639738082, "time": 0.637341833114624, "epoch": 1, "memory": 6319, "step": 2400} +{"lr": 9.909454654424387e-05, "data_time": 0.5531359910964966, "loss": 0.20310077518224717, "time": 0.6929550647735596, "epoch": 1, "memory": 6319, "step": 2500} +{"lr": 9.902090056383109e-05, "data_time": 0.5236783504486084, "loss": 0.17665813416242598, "time": 0.6639738559722901, "epoch": 1, "memory": 6319, "step": 2600} +{"lr": 9.894440835660365e-05, "data_time": 0.5283995389938354, "loss": 0.20761384516954423, "time": 0.6682494163513184, "epoch": 1, "memory": 6319, "step": 2700} +{"lr": 9.886507486826162e-05, "data_time": 0.5253950595855713, "loss": 0.18284522891044616, "time": 0.6658698797225953, "epoch": 1, "memory": 6319, "step": 2800} +{"lr": 9.878290522821154e-05, "data_time": 0.5330732345581055, "loss": 0.16379252076148987, "time": 0.6735464811325074, "epoch": 1, "memory": 6319, "step": 2900} +{"lr": 9.86979047492352e-05, "data_time": 0.5406192064285278, "loss": 0.1843602553009987, "time": 0.6811985731124878, "epoch": 1, "memory": 6319, "step": 3000} +{"lr": 9.861007892714578e-05, "data_time": 0.5320868730545044, "loss": 0.1826549530029297, "time": 0.6714630842208862, "epoch": 1, "memory": 6319, "step": 3100} +{"lr": 9.851943344043239e-05, "data_time": 0.5344639778137207, "loss": 0.17140807658433915, "time": 0.6751244544982911, "epoch": 1, "memory": 6319, "step": 3200} +{"lr": 9.842597414989319e-05, "data_time": 0.5207000017166138, "loss": 0.16282510459423066, "time": 0.6606178760528565, "epoch": 1, "memory": 6319, "step": 3300} +{"lr": 9.832970709825637e-05, "data_time": 0.5294976711273194, "loss": 0.1740340769290924, "time": 0.668919587135315, "epoch": 1, "memory": 6319, "step": 3400} +{"lr": 9.823063850978968e-05, "data_time": 0.5587979078292846, "loss": 0.18116564899683, "time": 0.6999265432357789, "epoch": 1, "memory": 6319, "step": 3500} +{"lr": 9.81287747898975e-05, "data_time": 0.5234679460525513, "loss": 0.16375539302825928, "time": 0.663607382774353, "epoch": 1, "memory": 6319, "step": 3600} +{"lr": 9.802412252470715e-05, "data_time": 0.5097539186477661, "loss": 0.17133889049291612, "time": 0.6512763977050782, "epoch": 1, "memory": 6319, "step": 3700} +{"lr": 9.791668848064258e-05, "data_time": 0.5157545566558838, "loss": 0.1612338401377201, "time": 0.6675833225250244, "epoch": 1, "memory": 6319, "step": 3800} +{"lr": 9.780647960398751e-05, "data_time": 0.5062771320343018, "loss": 0.17370904386043548, "time": 0.6451803684234619, "epoch": 1, "memory": 6319, "step": 3900} +{"accuracy/top1": 79.28742980957031, "data_time": 0.5958551900428638, "time": 0.6883839389734101, "step": 1} +{"lr": 9.7685491196646e-05, "data_time": 0.5370930910110474, "loss": 0.15279597342014312, "time": 0.6767707109451294, "epoch": 2, "memory": 6319, "step": 4007} +{"lr": 9.756956126085676e-05, "data_time": 0.5544443845748901, "loss": 0.14128805696964264, "time": 0.6940245151519775, "epoch": 2, "memory": 6319, "step": 4107} +{"lr": 9.745087893642482e-05, "data_time": 0.5067760944366455, "loss": 0.15452901273965836, "time": 0.646721601486206, "epoch": 2, "memory": 6319, "step": 4207} +{"lr": 9.73294518969056e-05, "data_time": 0.5036807775497436, "loss": 0.14054470509290695, "time": 0.6443322420120239, "epoch": 2, "memory": 6319, "step": 4307} +{"lr": 9.72052879933172e-05, "data_time": 0.4447112321853638, "loss": 0.15058355033397675, "time": 0.5851428747177124, "epoch": 2, "memory": 6319, "step": 4407} +{"lr": 9.707839525363358e-05, "data_time": 0.5212276458740235, "loss": 0.13102484717965127, "time": 0.6614755630493164, "epoch": 2, "memory": 6319, "step": 4507} +{"lr": 9.69487818822646e-05, "data_time": 0.5257565021514893, "loss": 0.14825795441865922, "time": 0.6660473108291626, "epoch": 2, "memory": 6319, "step": 4607} +{"lr": 9.681645625952616e-05, "data_time": 0.49306907653808596, "loss": 0.13866764903068543, "time": 0.6335996627807617, "epoch": 2, "memory": 6319, "step": 4707} +{"lr": 9.668142694109826e-05, "data_time": 0.5573489904403687, "loss": 0.14306616485118867, "time": 0.6974190473556519, "epoch": 2, "memory": 6319, "step": 4807} +{"lr": 9.654370265747162e-05, "data_time": 0.49083871841430665, "loss": 0.15921448469161986, "time": 0.6309617757797241, "epoch": 2, "memory": 6319, "step": 4907} +{"lr": 9.640329231338345e-05, "data_time": 0.5240086793899537, "loss": 0.1494707390666008, "time": 0.6651803970336914, "epoch": 2, "memory": 6319, "step": 5007} +{"lr": 9.626020498724147e-05, "data_time": 0.5176722526550293, "loss": 0.14862466901540755, "time": 0.6592450618743897, "epoch": 2, "memory": 6319, "step": 5107} +{"lr": 9.611444993053717e-05, "data_time": 0.6180542945861817, "loss": 0.1456143379211426, "time": 0.7580090999603272, "epoch": 2, "memory": 6319, "step": 5207} +{"lr": 9.596603656724746e-05, "data_time": 0.5168131828308106, "loss": 0.15639145374298097, "time": 0.6575065135955811, "epoch": 2, "memory": 6319, "step": 5307} +{"lr": 9.58149744932254e-05, "data_time": 0.5288204193115235, "loss": 0.1405955284833908, "time": 0.6684650659561158, "epoch": 2, "memory": 6319, "step": 5407} +{"lr": 9.566127347557972e-05, "data_time": 0.6000181674957276, "loss": 0.14103675931692122, "time": 0.7403709411621093, "epoch": 2, "memory": 6319, "step": 5507} +{"lr": 9.550494345204367e-05, "data_time": 0.5291576623916626, "loss": 0.14637183845043183, "time": 0.669269609451294, "epoch": 2, "memory": 6319, "step": 5607} +{"lr": 9.534599453033183e-05, "data_time": 0.5193047523498535, "loss": 0.12512431144714356, "time": 0.6595483541488647, "epoch": 2, "memory": 6319, "step": 5707} +{"lr": 9.518443698748712e-05, "data_time": 0.5405029296875, "loss": 0.14058603942394257, "time": 0.6811832666397095, "epoch": 2, "memory": 6319, "step": 5807} +{"lr": 9.50202812692162e-05, "data_time": 0.5031807899475098, "loss": 0.1317460559308529, "time": 0.6625309467315674, "epoch": 2, "memory": 6319, "step": 5907} +{"lr": 9.485353798921386e-05, "data_time": 0.5442139387130738, "loss": 0.14158946871757508, "time": 0.683808445930481, "epoch": 2, "memory": 6319, "step": 6007} +{"lr": 9.468421792847712e-05, "data_time": 0.5201704502105713, "loss": 0.1402334488928318, "time": 0.6704195499420166, "epoch": 2, "memory": 6319, "step": 6107} +{"lr": 9.451233203460794e-05, "data_time": 0.5358564853668213, "loss": 0.13895688578486443, "time": 0.6754046201705932, "epoch": 2, "memory": 6319, "step": 6207} +{"lr": 9.433789142110532e-05, "data_time": 0.48852086067199707, "loss": 0.13254403844475746, "time": 0.6629394054412842, "epoch": 2, "memory": 6319, "step": 6307} +{"lr": 9.416090736664709e-05, "data_time": 0.4711421251296997, "loss": 0.13277369290590285, "time": 0.611909294128418, "epoch": 2, "memory": 6319, "step": 6407} +{"lr": 9.398139131436032e-05, "data_time": 0.5422593593597412, "loss": 0.1290692836046219, "time": 0.6860929727554321, "epoch": 2, "memory": 6319, "step": 6507} +{"lr": 9.379935487108167e-05, "data_time": 0.5193511486053467, "loss": 0.1384017914533615, "time": 0.6600296020507812, "epoch": 2, "memory": 6319, "step": 6607} +{"lr": 9.361480980660697e-05, "data_time": 0.5154844999313355, "loss": 0.14731456190347672, "time": 0.6555843830108643, "epoch": 2, "memory": 6319, "step": 6707} +{"lr": 9.342776805292993e-05, "data_time": 0.511954951286316, "loss": 0.12090265229344369, "time": 0.6514381885528564, "epoch": 2, "memory": 6319, "step": 6807} +{"lr": 9.323824170347083e-05, "data_time": 0.5648843050003052, "loss": 0.1435110352933407, "time": 0.7055478811264038, "epoch": 2, "memory": 6319, "step": 6907} +{"lr": 9.304624301229471e-05, "data_time": 0.48480408191680907, "loss": 0.13903704956173896, "time": 0.6247333288192749, "epoch": 2, "memory": 6319, "step": 7007} +{"lr": 9.285178439331897e-05, "data_time": 0.5070621728897095, "loss": 0.12053157761693001, "time": 0.6478193283081055, "epoch": 2, "memory": 6319, "step": 7107} +{"lr": 9.265487841951087e-05, "data_time": 0.49961085319519044, "loss": 0.13237974122166635, "time": 0.6394788503646851, "epoch": 2, "memory": 6319, "step": 7207} +{"lr": 9.245553782207421e-05, "data_time": 0.5493011474609375, "loss": 0.12692570984363555, "time": 0.6906674385070801, "epoch": 2, "memory": 6319, "step": 7307} +{"lr": 9.225377548962671e-05, "data_time": 0.5684386014938354, "loss": 0.12293680310249329, "time": 0.7101089239120484, "epoch": 2, "memory": 6319, "step": 7407} +{"lr": 9.204960446736601e-05, "data_time": 0.4899008750915527, "loss": 0.11427125558257104, "time": 0.6302706718444824, "epoch": 2, "memory": 6319, "step": 7507} +{"lr": 9.184303795622715e-05, "data_time": 0.5326293230056762, "loss": 0.12831226736307144, "time": 0.6717268705368042, "epoch": 2, "memory": 6319, "step": 7607} +{"lr": 9.163408931202823e-05, "data_time": 0.5166270971298218, "loss": 0.12190170288085937, "time": 0.6563956499099731, "epoch": 2, "memory": 6319, "step": 7707} +{"lr": 9.142277204460729e-05, "data_time": 0.4866673707962036, "loss": 0.12000658884644508, "time": 0.6266154050827026, "epoch": 2, "memory": 6319, "step": 7807} +{"accuracy/top1": 84.4159927368164, "data_time": 0.47536983983270054, "time": 0.5640446399820263, "step": 2} +{"lr": 9.119405492383288e-05, "data_time": 0.5369272470474243, "loss": 0.1321154534816742, "time": 0.6764336824417114, "epoch": 3, "memory": 6319, "step": 7914} +{"lr": 9.097787819233205e-05, "data_time": 0.4976217269897461, "loss": 0.10088460966944694, "time": 0.6464096069335937, "epoch": 3, "memory": 6319, "step": 8014} +{"lr": 9.075937526576627e-05, "data_time": 0.5266026496887207, "loss": 0.14688998758792876, "time": 0.6661184072494507, "epoch": 3, "memory": 6319, "step": 8114} +{"lr": 9.053856027171755e-05, "data_time": 0.511517333984375, "loss": 0.11930367425084114, "time": 0.6567173957824707, "epoch": 3, "memory": 6319, "step": 8214} +{"lr": 9.031544748725785e-05, "data_time": 0.5386279821395874, "loss": 0.11722368896007537, "time": 0.6791386604309082, "epoch": 3, "memory": 6319, "step": 8314} +{"lr": 9.009005133802542e-05, "data_time": 0.5394606590270996, "loss": 0.12796138003468513, "time": 0.6794095516204834, "epoch": 3, "memory": 6319, "step": 8414} +{"lr": 8.986238639729242e-05, "data_time": 0.49581429958343504, "loss": 0.12170281857252122, "time": 0.636667799949646, "epoch": 3, "memory": 6319, "step": 8514} +{"lr": 8.963246738502254e-05, "data_time": 0.6008338689804077, "loss": 0.11625181287527084, "time": 0.741152572631836, "epoch": 3, "memory": 6319, "step": 8614} +{"lr": 8.940030916691928e-05, "data_time": 0.5272071599960327, "loss": 0.12248632237315178, "time": 0.6671080112457275, "epoch": 3, "memory": 6319, "step": 8714} +{"lr": 8.91659267534649e-05, "data_time": 0.497597861289978, "loss": 0.13870976641774177, "time": 0.6385526418685913, "epoch": 3, "memory": 6319, "step": 8814} +{"lr": 8.892933529894978e-05, "data_time": 0.5019416570663452, "loss": 0.10886357426643371, "time": 0.6426058292388916, "epoch": 3, "memory": 6319, "step": 8914} +{"lr": 8.869055010049283e-05, "data_time": 0.5331192493438721, "loss": 0.1227203406393528, "time": 0.6732116222381592, "epoch": 3, "memory": 6319, "step": 9014} +{"lr": 8.844958659705188e-05, "data_time": 0.49990861415863036, "loss": 0.11642178148031235, "time": 0.6398486137390137, "epoch": 3, "memory": 6319, "step": 9114} +{"lr": 8.82064603684259e-05, "data_time": 0.501291298866272, "loss": 0.11247068196535111, "time": 0.6419290065765381, "epoch": 3, "memory": 6319, "step": 9214} +{"lr": 8.796118713424776e-05, "data_time": 0.5387366771697998, "loss": 0.11492721736431122, "time": 0.6775113105773926, "epoch": 3, "memory": 6319, "step": 9314} +{"lr": 8.77137827529676e-05, "data_time": 0.5013673305511475, "loss": 0.11084105968475341, "time": 0.6408612489700317, "epoch": 3, "memory": 6319, "step": 9414} +{"lr": 8.746426322082745e-05, "data_time": 0.7175061464309692, "loss": 0.10803602561354637, "time": 0.8585130214691162, "epoch": 3, "memory": 6319, "step": 9514} +{"lr": 8.721264467082732e-05, "data_time": 0.507056450843811, "loss": 0.11435580402612686, "time": 0.6468563318252564, "epoch": 3, "memory": 6319, "step": 9614} +{"lr": 8.695894337168168e-05, "data_time": 0.4911908388137817, "loss": 0.11398598477244377, "time": 0.6302866458892822, "epoch": 3, "memory": 6319, "step": 9714} +{"lr": 8.67031757267677e-05, "data_time": 0.504483413696289, "loss": 0.11963612660765648, "time": 0.6445215225219727, "epoch": 3, "memory": 6319, "step": 9814} +{"lr": 8.64453582730651e-05, "data_time": 0.5450257301330567, "loss": 0.1172746978700161, "time": 0.6851329326629638, "epoch": 3, "memory": 6319, "step": 9914} +{"lr": 8.61855076800861e-05, "data_time": 0.521156644821167, "loss": 0.10335676074028015, "time": 0.6610426425933837, "epoch": 3, "memory": 6319, "step": 10014} +{"lr": 8.592364074879843e-05, "data_time": 0.5473639488220214, "loss": 0.11976106017827988, "time": 0.6864054679870606, "epoch": 3, "memory": 6319, "step": 10114} +{"lr": 8.565977441053854e-05, "data_time": 0.5903863906860352, "loss": 0.1248409979045391, "time": 0.7293467283248901, "epoch": 3, "memory": 6319, "step": 10214} +{"lr": 8.539392572591724e-05, "data_time": 0.5097359180450439, "loss": 0.13603954315185546, "time": 0.6492294788360595, "epoch": 3, "memory": 6319, "step": 10314} +{"lr": 8.512611188371633e-05, "data_time": 0.6894086122512817, "loss": 0.10091322213411331, "time": 0.8286743879318237, "epoch": 3, "memory": 6319, "step": 10414} +{"lr": 8.485635019977725e-05, "data_time": 0.5736372947692872, "loss": 0.11598249152302742, "time": 0.7133435726165771, "epoch": 3, "memory": 6319, "step": 10514} +{"lr": 8.45846581158818e-05, "data_time": 0.5090338945388794, "loss": 0.09406849667429924, "time": 0.6500195264816284, "epoch": 3, "memory": 6319, "step": 10614} +{"lr": 8.431105319862416e-05, "data_time": 0.5089647054672242, "loss": 0.12137909457087517, "time": 0.6541881322860718, "epoch": 3, "memory": 6319, "step": 10714} +{"lr": 8.403555313827516e-05, "data_time": 0.5094683170318604, "loss": 0.10281328037381172, "time": 0.6606269598007202, "epoch": 3, "memory": 6319, "step": 10814} +{"lr": 8.375817574763844e-05, "data_time": 0.5197919845581055, "loss": 0.10709384977817535, "time": 0.659517502784729, "epoch": 3, "memory": 6319, "step": 10914} +{"lr": 8.347893896089871e-05, "data_time": 0.49315719604492186, "loss": 0.10601055026054382, "time": 0.6474194049835205, "epoch": 3, "memory": 6319, "step": 11014} +{"lr": 8.31978608324625e-05, "data_time": 0.4724468946456909, "loss": 0.11521460637450218, "time": 0.6125427961349488, "epoch": 3, "memory": 6319, "step": 11114} +{"lr": 8.291495953579052e-05, "data_time": 0.5435532093048095, "loss": 0.0990497924387455, "time": 0.6833879709243774, "epoch": 3, "memory": 6319, "step": 11214} +{"lr": 8.263025336222267e-05, "data_time": 0.5444221019744873, "loss": 0.10346187576651573, "time": 0.684584665298462, "epoch": 3, "memory": 6319, "step": 11314} +{"lr": 8.234376071979551e-05, "data_time": 0.5355138540267944, "loss": 0.10680332705378533, "time": 0.6770189762115478, "epoch": 3, "memory": 6319, "step": 11414} +{"lr": 8.205550013205187e-05, "data_time": 0.5348822355270386, "loss": 0.10994913391768932, "time": 0.6740388154983521, "epoch": 3, "memory": 6319, "step": 11514} +{"lr": 8.17654902368435e-05, "data_time": 0.5360685110092163, "loss": 0.10609661005437374, "time": 0.6752196073532104, "epoch": 3, "memory": 6319, "step": 11614} +{"lr": 8.147374978512582e-05, "data_time": 0.4762966394424438, "loss": 0.10849932357668876, "time": 0.6158416271209717, "epoch": 3, "memory": 6319, "step": 11714} +{"accuracy/top1": 87.34212493896484, "data_time": 0.5443065618646556, "time": 0.6336771126451164, "step": 3} +{"lr": 8.115969237626578e-05, "data_time": 0.588813328742981, "loss": 0.11412608325481415, "time": 0.7443063259124756, "epoch": 4, "memory": 6319, "step": 11821} +{"lr": 8.08644297337245e-05, "data_time": 0.506064486503601, "loss": 0.10286526530981063, "time": 0.6592431783676147, "epoch": 4, "memory": 6319, "step": 11921} +{"lr": 8.056749479387651e-05, "data_time": 0.4116933345794678, "loss": 0.10236260145902634, "time": 0.6955714464187622, "epoch": 4, "memory": 6319, "step": 12021} +{"lr": 8.026890675542457e-05, "data_time": 0.3840444326400757, "loss": 0.1088383249938488, "time": 0.6918084859848023, "epoch": 4, "memory": 6319, "step": 12121} +{"lr": 7.996868492395457e-05, "data_time": 0.30678324699401854, "loss": 0.11419989466667176, "time": 0.6832905769348144, "epoch": 4, "memory": 6319, "step": 12221} +{"lr": 7.966684871068762e-05, "data_time": 0.3399131059646606, "loss": 0.1026128277182579, "time": 0.6849857330322265, "epoch": 4, "memory": 6319, "step": 12321} +{"lr": 7.936341763122449e-05, "data_time": 0.37445812225341796, "loss": 0.10519614145159721, "time": 0.653217077255249, "epoch": 4, "memory": 6319, "step": 12421} +{"lr": 7.905841130428405e-05, "data_time": 0.4166030168533325, "loss": 0.10487660765647888, "time": 0.6608822822570801, "epoch": 4, "memory": 6319, "step": 12521} +{"lr": 7.875184945043487e-05, "data_time": 0.31440300941467286, "loss": 0.0910740464925766, "time": 0.6547337293624877, "epoch": 4, "memory": 6319, "step": 12621} +{"lr": 7.844375189082012e-05, "data_time": 0.27991909980773927, "loss": 0.1040660209953785, "time": 0.7189595699310303, "epoch": 4, "memory": 6319, "step": 12721} +{"lr": 7.813413854587572e-05, "data_time": 0.3071895599365234, "loss": 0.08887388631701469, "time": 0.6119935035705566, "epoch": 4, "memory": 6319, "step": 12821} +{"lr": 7.782302943404291e-05, "data_time": 0.5093741893768311, "loss": 0.11017532199621201, "time": 0.649790620803833, "epoch": 4, "memory": 6319, "step": 12921} +{"lr": 7.751044467047337e-05, "data_time": 0.541435956954956, "loss": 0.11994380503892899, "time": 0.68137047290802, "epoch": 4, "memory": 6319, "step": 13021} +{"lr": 7.719640446572907e-05, "data_time": 0.5784233808517456, "loss": 0.10489736124873161, "time": 0.717302942276001, "epoch": 4, "memory": 6319, "step": 13121} +{"lr": 7.688092912447536e-05, "data_time": 0.5041449308395386, "loss": 0.11665222495794296, "time": 0.6451606035232544, "epoch": 4, "memory": 6319, "step": 13221} +{"lr": 7.656403904416794e-05, "data_time": 0.53504958152771, "loss": 0.1014468491077423, "time": 0.6744351625442505, "epoch": 4, "memory": 6319, "step": 13321} +{"lr": 7.624575471373434e-05, "data_time": 0.5670973300933838, "loss": 0.10275198072195053, "time": 0.7068441867828369, "epoch": 4, "memory": 6319, "step": 13421} +{"lr": 7.592609671224926e-05, "data_time": 0.5109803676605225, "loss": 0.10678192600607872, "time": 0.6506503343582153, "epoch": 4, "memory": 6319, "step": 13521} +{"lr": 7.560508570760375e-05, "data_time": 0.5814297676086426, "loss": 0.11037224233150482, "time": 0.7270254135131836, "epoch": 4, "memory": 6319, "step": 13621} +{"lr": 7.528274245516895e-05, "data_time": 0.5672618627548218, "loss": 0.10577552020549774, "time": 0.7087949752807617, "epoch": 4, "memory": 6319, "step": 13721} +{"lr": 7.495908779645405e-05, "data_time": 0.5114144086837769, "loss": 0.0917666207998991, "time": 0.720955753326416, "epoch": 4, "memory": 6319, "step": 13821} +{"lr": 7.463414265775906e-05, "data_time": 0.5522169828414917, "loss": 0.10598318874835969, "time": 0.6921687364578247, "epoch": 4, "memory": 6319, "step": 13921} +{"lr": 7.430792804882147e-05, "data_time": 0.5370861291885376, "loss": 0.10121533870697022, "time": 0.6871623277664185, "epoch": 4, "memory": 6319, "step": 14021} +{"lr": 7.398046506145803e-05, "data_time": 0.5097856283187866, "loss": 0.11258571781218052, "time": 0.6490369081497193, "epoch": 4, "memory": 6319, "step": 14121} +{"lr": 7.365177486820114e-05, "data_time": 0.6446864604949951, "loss": 0.09884856045246124, "time": 0.7903539896011352, "epoch": 4, "memory": 6319, "step": 14221} +{"lr": 7.332187872092939e-05, "data_time": 0.48531346321105956, "loss": 0.10299100801348686, "time": 0.6248530864715576, "epoch": 4, "memory": 6319, "step": 14321} +{"lr": 7.299079794949401e-05, "data_time": 0.5231744766235351, "loss": 0.09523547291755677, "time": 0.6623619794845581, "epoch": 4, "memory": 6319, "step": 14421} +{"lr": 7.265855396033955e-05, "data_time": 0.5392854452133179, "loss": 0.09905967712402344, "time": 0.6819924354553223, "epoch": 4, "memory": 6319, "step": 14521} +{"lr": 7.232516823512013e-05, "data_time": 0.5042264461517334, "loss": 0.09532521292567253, "time": 0.6450309753417969, "epoch": 4, "memory": 6319, "step": 14621} +{"lr": 7.199066232930995e-05, "data_time": 0.5368626594543457, "loss": 0.08734327033162118, "time": 0.6775792598724365, "epoch": 4, "memory": 6319, "step": 14721} +{"lr": 7.165505787080997e-05, "data_time": 0.5642958164215088, "loss": 0.10501666218042374, "time": 0.7061097145080566, "epoch": 4, "memory": 6319, "step": 14821} +{"lr": 7.13183765585495e-05, "data_time": 0.5412783622741699, "loss": 0.08821191489696503, "time": 0.6815030813217163, "epoch": 4, "memory": 6319, "step": 14921} +{"lr": 7.098064016108317e-05, "data_time": 0.5088694095611572, "loss": 0.09396752193570138, "time": 0.6490758419036865, "epoch": 4, "memory": 6319, "step": 15021} +{"lr": 7.064187051518379e-05, "data_time": 0.5177159547805786, "loss": 0.09922549277544021, "time": 0.659341287612915, "epoch": 4, "memory": 6319, "step": 15121} +{"lr": 7.030208952442949e-05, "data_time": 0.5396681308746338, "loss": 0.09435384050011635, "time": 0.6801180362701416, "epoch": 4, "memory": 6319, "step": 15221} +{"lr": 6.996131915778851e-05, "data_time": 0.6161983013153076, "loss": 0.0895415049046278, "time": 0.7637013912200927, "epoch": 4, "memory": 6319, "step": 15321} +{"lr": 6.961958144819853e-05, "data_time": 0.5221543073654175, "loss": 0.09963801801204682, "time": 0.6628645181655883, "epoch": 4, "memory": 6319, "step": 15421} +{"lr": 6.927689849114193e-05, "data_time": 0.49598758220672606, "loss": 0.09301925748586655, "time": 0.6423600673675537, "epoch": 4, "memory": 6319, "step": 15521} +{"lr": 6.893329244321722e-05, "data_time": 0.5162488698959351, "loss": 0.09806651324033737, "time": 0.6562772035598755, "epoch": 4, "memory": 6319, "step": 15621} +{"accuracy/top1": 90.25086212158203, "data_time": 0.5513947051146935, "time": 0.6400111543721166, "step": 4} +{"lr": 6.856463687361792e-05, "data_time": 0.5262357473373414, "loss": 0.08470112420618534, "time": 0.708863353729248, "epoch": 5, "memory": 6319, "step": 15728} +{"lr": 6.821919068470432e-05, "data_time": 0.5165777683258057, "loss": 0.09095192849636077, "time": 0.6574634790420533, "epoch": 5, "memory": 6319, "step": 15828} +{"lr": 6.787288979235407e-05, "data_time": 0.5227962255477905, "loss": 0.1057699330151081, "time": 0.6635576248168945, "epoch": 5, "memory": 6319, "step": 15928} +{"lr": 6.752575658708793e-05, "data_time": 0.5789114236831665, "loss": 0.08842909708619118, "time": 0.719330883026123, "epoch": 5, "memory": 6319, "step": 16028} +{"lr": 6.71778135132408e-05, "data_time": 0.5475069046020508, "loss": 0.09648694843053818, "time": 0.6873472929000854, "epoch": 5, "memory": 6319, "step": 16128} +{"lr": 6.682908306751074e-05, "data_time": 0.5067532300949097, "loss": 0.10286273434758186, "time": 0.6608107566833497, "epoch": 5, "memory": 6319, "step": 16228} +{"lr": 6.64795877975043e-05, "data_time": 0.5259492635726929, "loss": 0.09357130825519562, "time": 0.6656451463699341, "epoch": 5, "memory": 6319, "step": 16328} +{"lr": 6.61293503002787e-05, "data_time": 0.7284724712371826, "loss": 0.10000606030225753, "time": 0.8689975500106811, "epoch": 5, "memory": 6319, "step": 16428} +{"lr": 6.577839322088084e-05, "data_time": 0.46921329498291015, "loss": 0.0939922533929348, "time": 0.6115054368972779, "epoch": 5, "memory": 6319, "step": 16528} +{"lr": 6.542673925088298e-05, "data_time": 0.5045658111572265, "loss": 0.09371010065078736, "time": 0.6595198392868042, "epoch": 5, "memory": 6319, "step": 16628} +{"lr": 6.507441112691584e-05, "data_time": 0.495917820930481, "loss": 0.09412411749362945, "time": 0.6368811845779419, "epoch": 5, "memory": 6319, "step": 16728} +{"lr": 6.47214316291983e-05, "data_time": 0.49607903957366944, "loss": 0.10073281228542327, "time": 0.6506678104400635, "epoch": 5, "memory": 6319, "step": 16828} +{"lr": 6.436782358006481e-05, "data_time": 0.5493082523345947, "loss": 0.09790346249938012, "time": 0.6898935794830322, "epoch": 5, "memory": 6319, "step": 16928} +{"lr": 6.40136098424896e-05, "data_time": 0.5583202600479126, "loss": 0.09719995222985744, "time": 0.6991142988204956, "epoch": 5, "memory": 6319, "step": 17028} +{"lr": 6.365881331860828e-05, "data_time": 0.5614637851715087, "loss": 0.07730818539857864, "time": 0.7026409149169922, "epoch": 5, "memory": 6319, "step": 17128} +{"lr": 6.330345694823739e-05, "data_time": 0.5357184648513794, "loss": 0.09566049724817276, "time": 0.6767968893051147, "epoch": 5, "memory": 6319, "step": 17228} +{"lr": 6.29475637073908e-05, "data_time": 0.5066222429275513, "loss": 0.09327691644430161, "time": 0.6473726987838745, "epoch": 5, "memory": 6319, "step": 17328} +{"lr": 6.259115660679482e-05, "data_time": 0.4877237558364868, "loss": 0.0999925646930933, "time": 0.6273519992828369, "epoch": 5, "memory": 6319, "step": 17428} +{"lr": 6.223425869039963e-05, "data_time": 0.5396599292755127, "loss": 0.08791594877839089, "time": 0.6816283702850342, "epoch": 5, "memory": 6319, "step": 17528} +{"lr": 6.187689303389006e-05, "data_time": 0.5210930585861206, "loss": 0.09718217775225639, "time": 0.6614705085754394, "epoch": 5, "memory": 6319, "step": 17628} +{"lr": 6.151908274319292e-05, "data_time": 0.4726128339767456, "loss": 0.08879563137888909, "time": 0.6225663185119629, "epoch": 5, "memory": 6319, "step": 17728} +{"lr": 6.116085095298373e-05, "data_time": 0.501851224899292, "loss": 0.09075678363442422, "time": 0.6425661563873291, "epoch": 5, "memory": 6319, "step": 17828} +{"lr": 6.080222082519007e-05, "data_time": 0.6354441165924072, "loss": 0.10212242342531681, "time": 0.7770376443862915, "epoch": 5, "memory": 6319, "step": 17928} +{"lr": 6.0443215547495147e-05, "data_time": 0.6274643421173096, "loss": 0.09559036567807197, "time": 0.7664098501205444, "epoch": 5, "memory": 6319, "step": 18028} +{"lr": 6.0083858331837714e-05, "data_time": 0.47197983264923093, "loss": 0.08705018796026706, "time": 0.6134809732437134, "epoch": 5, "memory": 6319, "step": 18128} +{"lr": 5.97241724129116e-05, "data_time": 0.5098498582839965, "loss": 0.09013766944408416, "time": 0.6514616966247558, "epoch": 5, "memory": 6319, "step": 18228} +{"lr": 5.936418104666328e-05, "data_time": 0.48030169010162355, "loss": 0.08908655494451523, "time": 0.6206804037094116, "epoch": 5, "memory": 6319, "step": 18328} +{"lr": 5.9003907508788295e-05, "data_time": 0.5214685916900634, "loss": 0.10488158017396927, "time": 0.662332558631897, "epoch": 5, "memory": 6319, "step": 18428} +{"lr": 5.864337509322633e-05, "data_time": 0.5449373722076416, "loss": 0.09434689357876777, "time": 0.6877973794937133, "epoch": 5, "memory": 6319, "step": 18528} +{"lr": 5.8282607110655194e-05, "data_time": 0.5710325956344604, "loss": 0.09347037933766841, "time": 0.7120609760284424, "epoch": 5, "memory": 6319, "step": 18628} +{"lr": 5.7921626886983564e-05, "data_time": 0.5557557106018066, "loss": 0.09960277751088142, "time": 0.6971974611282349, "epoch": 5, "memory": 6319, "step": 18728} +{"lr": 5.756045776184287e-05, "data_time": 0.5279939889907836, "loss": 0.10416858866810799, "time": 0.6684570789337159, "epoch": 5, "memory": 6319, "step": 18828} +{"lr": 5.719912308707808e-05, "data_time": 0.5208239555358887, "loss": 0.10076755955815316, "time": 0.6613128900527954, "epoch": 5, "memory": 6319, "step": 18928} +{"lr": 5.683764622523808e-05, "data_time": 0.5302015542984009, "loss": 0.08578926399350166, "time": 0.6708161354064941, "epoch": 5, "memory": 6319, "step": 19028} +{"lr": 5.647605054806502e-05, "data_time": 0.5026536464691163, "loss": 0.1170566976070404, "time": 0.6428820848464966, "epoch": 5, "memory": 6319, "step": 19128} +{"lr": 5.6114359434983206e-05, "data_time": 0.516049861907959, "loss": 0.09198833629488945, "time": 0.6569303035736084, "epoch": 5, "memory": 6319, "step": 19228} +{"lr": 5.575259627158745e-05, "data_time": 0.5426882743835449, "loss": 0.10285206660628318, "time": 0.6830023527145386, "epoch": 5, "memory": 6319, "step": 19328} +{"lr": 5.539078444813113e-05, "data_time": 0.49720287322998047, "loss": 0.08695899248123169, "time": 0.6380908966064454, "epoch": 5, "memory": 6319, "step": 19428} +{"lr": 5.5028947358013735e-05, "data_time": 0.5174664974212646, "loss": 0.10256444588303566, "time": 0.6577479362487793, "epoch": 5, "memory": 6319, "step": 19528} +{"accuracy/top1": 93.73020935058594, "data_time": 0.6467497266572098, "time": 0.7327331181230217, "step": 5} +{"lr": 5.464178020340108e-05, "data_time": 0.03125379085540771, "loss": 0.10717456564307212, "time": 1.0087543964385985, "epoch": 6, "memory": 6319, "step": 19635} +{"lr": 5.427996514795046e-05, "data_time": 0.19920940399169923, "loss": 0.10471052080392837, "time": 0.534230399131775, "epoch": 6, "memory": 6319, "step": 19735} +{"lr": 5.3918196647260834e-05, "data_time": 0.4795753717422485, "loss": 0.08822899051010609, "time": 0.6257527112960816, "epoch": 6, "memory": 6319, "step": 19835} +{"lr": 5.355649809193057e-05, "data_time": 0.41004362106323244, "loss": 0.07934464812278748, "time": 0.6748523235321044, "epoch": 6, "memory": 6319, "step": 19935} +{"lr": 5.3194892868035765e-05, "data_time": 0.3806777477264404, "loss": 0.0930691808462143, "time": 0.6481860637664795, "epoch": 6, "memory": 6319, "step": 20035} +{"lr": 5.283340435561795e-05, "data_time": 0.5792163610458374, "loss": 0.09030508734285832, "time": 0.7213134765625, "epoch": 6, "memory": 6319, "step": 20135} +{"lr": 5.2472055927172495e-05, "data_time": 0.6683616399765014, "loss": 0.0896085411310196, "time": 0.8090466022491455, "epoch": 6, "memory": 6319, "step": 20235} +{"lr": 5.2110870946137605e-05, "data_time": 0.5130580425262451, "loss": 0.08650345578789712, "time": 0.6541225671768188, "epoch": 6, "memory": 6319, "step": 20335} +{"lr": 5.174987276538342e-05, "data_time": 0.5252459287643433, "loss": 0.09079801067709922, "time": 0.666130542755127, "epoch": 6, "memory": 6319, "step": 20435} +{"lr": 5.1389084725702425e-05, "data_time": 0.5122568368911743, "loss": 0.0950685203075409, "time": 0.6543164253234863, "epoch": 6, "memory": 6319, "step": 20535} +{"lr": 5.1028530154300045e-05, "data_time": 0.5155791997909546, "loss": 0.10416885912418365, "time": 0.6567234992980957, "epoch": 6, "memory": 6319, "step": 20635} +{"lr": 5.066823236328663e-05, "data_time": 0.5147588968276977, "loss": 0.08922100961208343, "time": 0.6560417890548706, "epoch": 6, "memory": 6319, "step": 20735} +{"lr": 5.030821464817006e-05, "data_time": 0.5122957944869995, "loss": 0.08737279959022999, "time": 0.6525972843170166, "epoch": 6, "memory": 6319, "step": 20835} +{"lr": 4.99485002863494e-05, "data_time": 0.5093324661254883, "loss": 0.08328028954565525, "time": 0.6497575759887695, "epoch": 6, "memory": 6319, "step": 20935} +{"lr": 4.95891125356102e-05, "data_time": 0.516583514213562, "loss": 0.08127425834536553, "time": 0.6572388172149658, "epoch": 6, "memory": 6319, "step": 21035} +{"lr": 4.923007463262053e-05, "data_time": 0.5068090915679931, "loss": 0.10989921689033508, "time": 0.6470248222351074, "epoch": 6, "memory": 6319, "step": 21135} +{"lr": 4.887140979142852e-05, "data_time": 0.556620454788208, "loss": 0.09486041627824307, "time": 0.7001865625381469, "epoch": 6, "memory": 6319, "step": 21235} +{"lr": 4.851314120196161e-05, "data_time": 0.5591617822647095, "loss": 0.07852182500064372, "time": 0.6984658002853393, "epoch": 6, "memory": 6319, "step": 21335} +{"lr": 4.815529202852693e-05, "data_time": 0.5204687595367432, "loss": 0.08852827474474907, "time": 0.6720494985580444, "epoch": 6, "memory": 6319, "step": 21435} +{"lr": 4.779788540831406e-05, "data_time": 0.5039268970489502, "loss": 0.08470682874321937, "time": 0.6435454845428467, "epoch": 6, "memory": 6319, "step": 21535} +{"lr": 4.744094444989843e-05, "data_time": 0.5169116020202636, "loss": 0.08685770258307457, "time": 0.6574362277984619, "epoch": 6, "memory": 6319, "step": 21635} +{"lr": 4.708449223174764e-05, "data_time": 0.5160007953643799, "loss": 0.08491603136062623, "time": 0.6593384504318237, "epoch": 6, "memory": 6319, "step": 21735} +{"lr": 4.672855180072916e-05, "data_time": 0.5460444211959838, "loss": 0.08279446847736835, "time": 0.6877310514450073, "epoch": 6, "memory": 6319, "step": 21835} +{"lr": 4.6373146170620256e-05, "data_time": 0.5128842115402221, "loss": 0.09533453360199928, "time": 0.6736371517181396, "epoch": 6, "memory": 6319, "step": 21935} +{"lr": 4.6018298320619915e-05, "data_time": 0.5153392791748047, "loss": 0.0897812344133854, "time": 0.6585339784622193, "epoch": 6, "memory": 6319, "step": 22035} +{"lr": 4.566403119386318e-05, "data_time": 0.48095753192901614, "loss": 0.10277087241411209, "time": 0.6338201284408569, "epoch": 6, "memory": 6319, "step": 22135} +{"lr": 4.531036769593769e-05, "data_time": 0.5006656169891357, "loss": 0.09919351935386658, "time": 0.6422847270965576, "epoch": 6, "memory": 6319, "step": 22235} +{"lr": 4.4957330693402685e-05, "data_time": 0.539228868484497, "loss": 0.0924248743802309, "time": 0.6834451913833618, "epoch": 6, "memory": 6319, "step": 22335} +{"lr": 4.460494301231059e-05, "data_time": 0.5093908548355103, "loss": 0.07978219091892243, "time": 0.6495115995407105, "epoch": 6, "memory": 6319, "step": 22435} +{"lr": 4.425322743673109e-05, "data_time": 0.5398077249526978, "loss": 0.08374065943062306, "time": 0.680783486366272, "epoch": 6, "memory": 6319, "step": 22535} +{"lr": 4.390220670727812e-05, "data_time": 0.4800394058227539, "loss": 0.08907215818762779, "time": 0.6207040786743164, "epoch": 6, "memory": 6319, "step": 22635} +{"lr": 4.3551903519639325e-05, "data_time": 0.5293129444122314, "loss": 0.09509703516960144, "time": 0.6707393646240234, "epoch": 6, "memory": 6319, "step": 22735} +{"lr": 4.3202340523108894e-05, "data_time": 0.5574806690216064, "loss": 0.08660216294229031, "time": 0.6985911130905151, "epoch": 6, "memory": 6319, "step": 22835} +{"lr": 4.285354031912302e-05, "data_time": 0.5237831830978393, "loss": 0.08252441361546517, "time": 0.6644094705581665, "epoch": 6, "memory": 6319, "step": 22935} +{"lr": 4.2505525459798544e-05, "data_time": 0.5262906312942505, "loss": 0.0942851398140192, "time": 0.6670958280563355, "epoch": 6, "memory": 6319, "step": 23035} +{"lr": 4.2158318446474836e-05, "data_time": 0.5269134998321533, "loss": 0.10015843287110329, "time": 0.6679558753967285, "epoch": 6, "memory": 6319, "step": 23135} +{"lr": 4.181194172825905e-05, "data_time": 0.4859327793121338, "loss": 0.09421113654971122, "time": 0.6261427640914917, "epoch": 6, "memory": 6319, "step": 23235} +{"lr": 4.146641770057448e-05, "data_time": 0.4777074813842773, "loss": 0.08744788654148579, "time": 0.6174055576324463, "epoch": 6, "memory": 6319, "step": 23335} +{"lr": 4.112176870371267e-05, "data_time": 0.505276894569397, "loss": 0.07330489903688431, "time": 0.6478984355926514, "epoch": 6, "memory": 6319, "step": 23435} +{"accuracy/top1": 95.14282989501953, "data_time": 0.5447075613613787, "time": 0.6323356463991362, "step": 6} +{"lr": 4.0753988582732786e-05, "data_time": 0.470862889289856, "loss": 0.08295015580952167, "time": 0.6833272933959961, "epoch": 7, "memory": 6319, "step": 23542} +{"lr": 4.0411221639286815e-05, "data_time": 0.48199923038482667, "loss": 0.08625544793903828, "time": 0.6510698318481445, "epoch": 7, "memory": 6319, "step": 23642} +{"lr": 4.006939795169518e-05, "data_time": 0.5034197807312012, "loss": 0.08994064182043075, "time": 0.6623530626296997, "epoch": 7, "memory": 6319, "step": 23742} +{"lr": 3.972853962099913e-05, "data_time": 0.33036906719207765, "loss": 0.09170535504817963, "time": 0.6491644144058227, "epoch": 7, "memory": 6319, "step": 23842} +{"lr": 3.93886686858237e-05, "data_time": 0.24437239170074462, "loss": 0.09314442239701748, "time": 0.6958123683929444, "epoch": 7, "memory": 6319, "step": 23942} +{"lr": 3.904980712095259e-05, "data_time": 0.30084683895111086, "loss": 0.09044926017522811, "time": 0.6798213958740235, "epoch": 7, "memory": 6319, "step": 24042} +{"lr": 3.871197683590748e-05, "data_time": 0.128113317489624, "loss": 0.09334088489413261, "time": 0.6648510932922364, "epoch": 7, "memory": 6319, "step": 24142} +{"lr": 3.837519967353125e-05, "data_time": 0.24887685775756835, "loss": 0.07223951257765293, "time": 0.6324395418167115, "epoch": 7, "memory": 6319, "step": 24242} +{"lr": 3.8039497408575855e-05, "data_time": 0.3246570348739624, "loss": 0.08753116950392723, "time": 0.6473471879959106, "epoch": 7, "memory": 6319, "step": 24342} +{"lr": 3.770489174629432e-05, "data_time": 0.28405938148498533, "loss": 0.10265554189682007, "time": 0.6407233476638794, "epoch": 7, "memory": 6319, "step": 24442} +{"lr": 3.737140432103754e-05, "data_time": 0.322407603263855, "loss": 0.09392819106578827, "time": 0.7046275615692139, "epoch": 7, "memory": 6319, "step": 24542} +{"lr": 3.703905669485545e-05, "data_time": 0.3817084074020386, "loss": 0.08836689218878746, "time": 0.662895917892456, "epoch": 7, "memory": 6319, "step": 24642} +{"lr": 3.6707870356102706e-05, "data_time": 0.35480873584747313, "loss": 0.09067027866840363, "time": 0.6342610359191895, "epoch": 7, "memory": 6319, "step": 24742} +{"lr": 3.6377866718049484e-05, "data_time": 0.5103625059127808, "loss": 0.09932314157485962, "time": 0.6865311145782471, "epoch": 7, "memory": 6319, "step": 24842} +{"lr": 3.6049067117496965e-05, "data_time": 0.5267199277877808, "loss": 0.09514525346457958, "time": 0.6687266588211059, "epoch": 7, "memory": 6319, "step": 24942} +{"lr": 3.5721492813397755e-05, "data_time": 0.5122140407562256, "loss": 0.0813816886395216, "time": 0.6517822027206421, "epoch": 7, "memory": 6319, "step": 25042} +{"lr": 3.539516498548151e-05, "data_time": 0.5005153179168701, "loss": 0.0832541186362505, "time": 0.6636749267578125, "epoch": 7, "memory": 6319, "step": 25142} +{"lr": 3.507010473288518e-05, "data_time": 0.4330591678619385, "loss": 0.0813992377370596, "time": 0.6964970588684082, "epoch": 7, "memory": 6319, "step": 25242} +{"lr": 3.474633307278909e-05, "data_time": 0.22276041507720948, "loss": 0.08509134873747826, "time": 0.6273354291915894, "epoch": 7, "memory": 6319, "step": 25342} +{"lr": 3.4423870939058177e-05, "data_time": 0.3570723533630371, "loss": 0.07953443638980388, "time": 0.6682219982147217, "epoch": 7, "memory": 6319, "step": 25442} +{"lr": 3.410273918088805e-05, "data_time": 0.20250792503356935, "loss": 0.09414939917623996, "time": 0.6555251598358154, "epoch": 7, "memory": 6319, "step": 25542} +{"lr": 3.37829585614574e-05, "data_time": 0.1938471794128418, "loss": 0.07364090755581856, "time": 0.6488155364990235, "epoch": 7, "memory": 6319, "step": 25642} +{"lr": 3.3464549756585216e-05, "data_time": 0.02102198600769043, "loss": 0.08286008350551129, "time": 0.6763957977294922, "epoch": 7, "memory": 6319, "step": 25742} +{"lr": 3.314753335339416e-05, "data_time": 0.0011744976043701171, "loss": 0.09127445928752423, "time": 0.6984846353530884, "epoch": 7, "memory": 6319, "step": 25842} +{"lr": 3.2831929848979394e-05, "data_time": 0.0012300491333007812, "loss": 0.07671383023262024, "time": 0.6535136699676514, "epoch": 7, "memory": 6319, "step": 25942} +{"lr": 3.251775964908332e-05, "data_time": 0.001391458511352539, "loss": 0.07746944688260556, "time": 0.6931942939758301, "epoch": 7, "memory": 6319, "step": 26042} +{"lr": 3.22050430667762e-05, "data_time": 0.0012549161911010742, "loss": 0.07638381943106651, "time": 0.633611798286438, "epoch": 7, "memory": 6319, "step": 26142} +{"lr": 3.189380032114285e-05, "data_time": 0.10946221351623535, "loss": 0.07504312098026275, "time": 0.6317307710647583, "epoch": 7, "memory": 6319, "step": 26242} +{"lr": 3.1584051535975324e-05, "data_time": 0.21273183822631836, "loss": 0.09941074028611183, "time": 0.6449794054031373, "epoch": 7, "memory": 6319, "step": 26342} +{"lr": 3.127581673847175e-05, "data_time": 0.21001415252685546, "loss": 0.08890110403299331, "time": 0.671927809715271, "epoch": 7, "memory": 6319, "step": 26442} +{"lr": 3.0969115857941555e-05, "data_time": 0.20174038410186768, "loss": 0.0774132028222084, "time": 0.6450658798217773, "epoch": 7, "memory": 6319, "step": 26542} +{"lr": 3.0663968724516654e-05, "data_time": 0.05780491828918457, "loss": 0.09331877008080483, "time": 0.6313152551651001, "epoch": 7, "memory": 6319, "step": 26642} +{"lr": 3.0360395067869668e-05, "data_time": 0.03669705390930176, "loss": 0.09191939383745193, "time": 0.614667558670044, "epoch": 7, "memory": 6319, "step": 26742} +{"lr": 3.0058414515938063e-05, "data_time": 0.02508423328399658, "loss": 0.08378423005342484, "time": 0.6930113077163697, "epoch": 7, "memory": 6319, "step": 26842} +{"lr": 2.975804659365501e-05, "data_time": 0.0022589445114135744, "loss": 0.08855830878019333, "time": 0.6762532711029052, "epoch": 7, "memory": 6319, "step": 26942} +{"lr": 2.945931072168714e-05, "data_time": 0.0016698360443115235, "loss": 0.08498649261891841, "time": 0.6263153076171875, "epoch": 7, "memory": 6319, "step": 27042} +{"lr": 2.9162226215178855e-05, "data_time": 0.0010523557662963866, "loss": 0.08232737556099892, "time": 0.6991947889328003, "epoch": 7, "memory": 6319, "step": 27142} +{"lr": 2.8866812282503322e-05, "data_time": 0.001619267463684082, "loss": 0.08074983358383178, "time": 0.6734080791473389, "epoch": 7, "memory": 6319, "step": 27242} +{"lr": 2.8573088024020838e-05, "data_time": 0.0008196592330932617, "loss": 0.08293840140104294, "time": 0.6396318435668945, "epoch": 7, "memory": 6319, "step": 27342} +{"accuracy/top1": 95.87696838378906, "data_time": 0.513101314676219, "time": 0.6030006696437967, "step": 7} +{"lr": 2.8260695817390092e-05, "data_time": 0.5127811431884766, "loss": 0.07460867390036582, "time": 0.6538702487945557, "epoch": 8, "memory": 6319, "step": 27449} +{"lr": 2.7970529402570967e-05, "data_time": 0.4982527494430542, "loss": 0.08210007287561893, "time": 0.6468517303466796, "epoch": 8, "memory": 6319, "step": 27549} +{"lr": 2.7682110612244214e-05, "data_time": 0.5047098875045777, "loss": 0.08655336350202561, "time": 0.651490330696106, "epoch": 8, "memory": 6319, "step": 27649} +{"lr": 2.7395458094490283e-05, "data_time": 0.4774590015411377, "loss": 0.10557109713554383, "time": 0.6198263168334961, "epoch": 8, "memory": 6319, "step": 27749} +{"lr": 2.7110590383189015e-05, "data_time": 0.5294427871704102, "loss": 0.09038270860910416, "time": 0.6706011295318604, "epoch": 8, "memory": 6319, "step": 27849} +{"lr": 2.6827525896821397e-05, "data_time": 0.5397517681121826, "loss": 0.09813364967703819, "time": 0.6807192325592041, "epoch": 8, "memory": 6319, "step": 27949} +{"lr": 2.6546282937278552e-05, "data_time": 0.44946396350860596, "loss": 0.09082960337400436, "time": 0.6686970710754394, "epoch": 8, "memory": 6319, "step": 28049} +{"lr": 2.6266879688678606e-05, "data_time": 0.5386229515075683, "loss": 0.09124983251094818, "time": 0.6814395904541015, "epoch": 8, "memory": 6319, "step": 28149} +{"lr": 2.5989334216190738e-05, "data_time": 0.5184274673461914, "loss": 0.10565513707697391, "time": 0.6589367389678955, "epoch": 8, "memory": 6319, "step": 28249} +{"lr": 2.571366446486732e-05, "data_time": 0.5535402059555053, "loss": 0.10147601626813411, "time": 0.6964458227157593, "epoch": 8, "memory": 6319, "step": 28349} +{"lr": 2.5439888258483665e-05, "data_time": 0.5335720777511597, "loss": 0.0966638259589672, "time": 0.673550796508789, "epoch": 8, "memory": 6319, "step": 28449} +{"lr": 2.5168023298385384e-05, "data_time": 0.5342338562011719, "loss": 0.07630592696368695, "time": 0.6762902975082398, "epoch": 8, "memory": 6319, "step": 28549} +{"lr": 2.489808716234412e-05, "data_time": 0.5069514274597168, "loss": 0.08626412823796273, "time": 0.6497405290603637, "epoch": 8, "memory": 6319, "step": 28649} +{"lr": 2.463009730342111e-05, "data_time": 0.4845763921737671, "loss": 0.0779623631387949, "time": 0.6296336412429809, "epoch": 8, "memory": 6319, "step": 28749} +{"lr": 2.436407104883849e-05, "data_time": 0.5296821832656861, "loss": 0.07111451178789138, "time": 0.6721384048461914, "epoch": 8, "memory": 6319, "step": 28849} +{"lr": 2.4100025598859092e-05, "data_time": 0.5019245386123657, "loss": 0.08655482605099678, "time": 0.6427699089050293, "epoch": 8, "memory": 6319, "step": 28949} +{"lr": 2.3837978025674284e-05, "data_time": 0.4973599910736084, "loss": 0.0867620412260294, "time": 0.6369682073593139, "epoch": 8, "memory": 6319, "step": 29049} +{"lr": 2.357794527230027e-05, "data_time": 0.5338386535644531, "loss": 0.08663633055984973, "time": 0.6740134716033935, "epoch": 8, "memory": 6319, "step": 29149} +{"lr": 2.33199441514825e-05, "data_time": 0.5322182655334473, "loss": 0.06905041374266148, "time": 0.6760332584381104, "epoch": 8, "memory": 6319, "step": 29249} +{"lr": 2.306399134460859e-05, "data_time": 0.5384789705276489, "loss": 0.08724874705076217, "time": 0.679179048538208, "epoch": 8, "memory": 6319, "step": 29349} +{"lr": 2.281010340062995e-05, "data_time": 0.5234551906585694, "loss": 0.07879753299057483, "time": 0.6651687622070312, "epoch": 8, "memory": 6319, "step": 29449} +{"lr": 2.2558296734991603e-05, "data_time": 0.48877928256988523, "loss": 0.08786786571145058, "time": 0.6288945436477661, "epoch": 8, "memory": 6319, "step": 29549} +{"lr": 2.2308587628570915e-05, "data_time": 0.4862042427062988, "loss": 0.08550646640360356, "time": 0.6295605897903442, "epoch": 8, "memory": 6319, "step": 29649} +{"lr": 2.2060992226624955e-05, "data_time": 0.5055670261383056, "loss": 0.07987292930483818, "time": 0.6510406017303467, "epoch": 8, "memory": 6319, "step": 29749} +{"lr": 2.181552653774648e-05, "data_time": 0.504708981513977, "loss": 0.08791458755731582, "time": 0.6673235893249512, "epoch": 8, "memory": 6319, "step": 29849} +{"lr": 2.1572206432829035e-05, "data_time": 0.5137610673904419, "loss": 0.07938118763267994, "time": 0.6555030345916748, "epoch": 8, "memory": 6319, "step": 29949} +{"lr": 2.1331047644040754e-05, "data_time": 0.5457321643829346, "loss": 0.08348729908466339, "time": 0.6873914003372192, "epoch": 8, "memory": 6319, "step": 30049} +{"lr": 2.109206576380705e-05, "data_time": 0.47716259956359863, "loss": 0.07699331119656563, "time": 0.6198265314102173, "epoch": 8, "memory": 6319, "step": 30149} +{"lr": 2.0855276243802687e-05, "data_time": 0.4770580530166626, "loss": 0.0768574696034193, "time": 0.6415684223175049, "epoch": 8, "memory": 6319, "step": 30249} +{"lr": 2.0620694393952487e-05, "data_time": 0.4538614511489868, "loss": 0.08800110891461373, "time": 0.5951207399368286, "epoch": 8, "memory": 6319, "step": 30349} +{"lr": 2.038833538144169e-05, "data_time": 0.5135487794876099, "loss": 0.09790711887180806, "time": 0.655436372756958, "epoch": 8, "memory": 6319, "step": 30449} +{"lr": 2.0158214229735154e-05, "data_time": 0.4786839485168457, "loss": 0.0819152183830738, "time": 0.621023154258728, "epoch": 8, "memory": 6319, "step": 30549} +{"lr": 1.9930345817605977e-05, "data_time": 0.4796322822570801, "loss": 0.06708877757191659, "time": 0.6211246967315673, "epoch": 8, "memory": 6319, "step": 30649} +{"lr": 1.970474487817358e-05, "data_time": 0.5479300737380981, "loss": 0.07790469005703926, "time": 0.7155184984207154, "epoch": 8, "memory": 6319, "step": 30749} +{"lr": 1.9481425997951092e-05, "data_time": 0.48464412689208985, "loss": 0.08609076626598836, "time": 0.6287136554718018, "epoch": 8, "memory": 6319, "step": 30849} +{"lr": 1.9260403615902187e-05, "data_time": 0.50627760887146, "loss": 0.08874175995588303, "time": 0.6590127229690552, "epoch": 8, "memory": 6319, "step": 30949} +{"lr": 1.9041692022507647e-05, "data_time": 0.481411337852478, "loss": 0.08259556405246257, "time": 0.6220230340957642, "epoch": 8, "memory": 6319, "step": 31049} +{"lr": 1.8825305358841125e-05, "data_time": 0.5230817794799805, "loss": 0.06932475790381432, "time": 0.6637158632278443, "epoch": 8, "memory": 6319, "step": 31149} +{"lr": 1.861125761565518e-05, "data_time": 0.49395802021026614, "loss": 0.08681600689888, "time": 0.6353115797042846, "epoch": 8, "memory": 6319, "step": 31249} +{"accuracy/top1": 96.5067367553711, "data_time": 0.5387871429837984, "time": 0.6271591474269999, "step": 8} +{"lr": 1.8384832449122864e-05, "data_time": 0.45036723613739016, "loss": 0.0774447601288557, "time": 0.5908416509628296, "epoch": 9, "memory": 6319, "step": 31356} +{"lr": 1.817567007332114e-05, "data_time": 0.4683086395263672, "loss": 0.08455916084349155, "time": 0.6101473569869995, "epoch": 9, "memory": 6319, "step": 31456} +{"lr": 1.796888862098825e-05, "data_time": 0.4808666229248047, "loss": 0.08091538585722446, "time": 0.620784068107605, "epoch": 9, "memory": 6319, "step": 31556} +{"lr": 1.7764501461839678e-05, "data_time": 0.49072086811065674, "loss": 0.08568033501505852, "time": 0.6317744016647339, "epoch": 9, "memory": 6319, "step": 31656} +{"lr": 1.7562521810784746e-05, "data_time": 0.5133663654327393, "loss": 0.07859389409422875, "time": 0.654290747642517, "epoch": 9, "memory": 6319, "step": 31756} +{"lr": 1.7362962727072444e-05, "data_time": 0.48868684768676757, "loss": 0.0888303942978382, "time": 0.6742319822311401, "epoch": 9, "memory": 6319, "step": 31856} +{"lr": 1.716583711344682e-05, "data_time": 0.49153151512146, "loss": 0.08321366459131241, "time": 0.6329559087753296, "epoch": 9, "memory": 6319, "step": 31956} +{"lr": 1.6971157715312954e-05, "data_time": 0.500112795829773, "loss": 0.07347173951566219, "time": 0.6415749311447143, "epoch": 9, "memory": 6319, "step": 32056} +{"lr": 1.677893711991274e-05, "data_time": 0.4580115795135498, "loss": 0.0807027492672205, "time": 0.6126995801925659, "epoch": 9, "memory": 6319, "step": 32156} +{"lr": 1.6589187755511025e-05, "data_time": 0.5103238582611084, "loss": 0.078684401512146, "time": 0.66378173828125, "epoch": 9, "memory": 6319, "step": 32256} +{"lr": 1.640192189059216e-05, "data_time": 0.49948337078094485, "loss": 0.0780689612030983, "time": 0.705699110031128, "epoch": 9, "memory": 6319, "step": 32356} +{"lr": 1.6217151633066716e-05, "data_time": 0.44858882427215574, "loss": 0.08355204723775386, "time": 0.708412766456604, "epoch": 9, "memory": 6319, "step": 32456} +{"lr": 1.6034888929488564e-05, "data_time": 0.5895434856414795, "loss": 0.08971339166164398, "time": 0.7298603057861328, "epoch": 9, "memory": 6319, "step": 32556} +{"lr": 1.585514556428258e-05, "data_time": 0.5000896453857422, "loss": 0.07340927235782146, "time": 0.6414830207824707, "epoch": 9, "memory": 6319, "step": 32656} +{"lr": 1.567793315898261e-05, "data_time": 0.5831059217453003, "loss": 0.0837759081274271, "time": 0.7241213321685791, "epoch": 9, "memory": 6319, "step": 32756} +{"lr": 1.5503263171480048e-05, "data_time": 0.5291210174560547, "loss": 0.09053281843662261, "time": 0.6703447818756103, "epoch": 9, "memory": 6319, "step": 32856} +{"lr": 1.5331146895283165e-05, "data_time": 0.519048810005188, "loss": 0.08845474012196064, "time": 0.6597179651260376, "epoch": 9, "memory": 6319, "step": 32956} +{"lr": 1.5161595458786728e-05, "data_time": 0.5170748472213745, "loss": 0.09481598138809204, "time": 0.657964038848877, "epoch": 9, "memory": 6319, "step": 33056} +{"lr": 1.4994619824552563e-05, "data_time": 0.4895815849304199, "loss": 0.08049409575760365, "time": 0.6292577266693116, "epoch": 9, "memory": 6319, "step": 33156} +{"lr": 1.4830230788600802e-05, "data_time": 0.48902390003204343, "loss": 0.08306508734822274, "time": 0.6349939823150634, "epoch": 9, "memory": 6319, "step": 33256} +{"lr": 1.4668438979711779e-05, "data_time": 0.5043458700180053, "loss": 0.10058392062783242, "time": 0.6451935052871705, "epoch": 9, "memory": 6319, "step": 33356} +{"lr": 1.4509254858738844e-05, "data_time": 0.48384382724761965, "loss": 0.07483227849006653, "time": 0.6241837739944458, "epoch": 9, "memory": 6319, "step": 33456} +{"lr": 1.4352688717931966e-05, "data_time": 0.4971457481384277, "loss": 0.08895487524569035, "time": 0.6370618104934692, "epoch": 9, "memory": 6319, "step": 33556} +{"lr": 1.4198750680272348e-05, "data_time": 0.5083073616027832, "loss": 0.09322233945131302, "time": 0.6478856086730957, "epoch": 9, "memory": 6319, "step": 33656} +{"lr": 1.4047450698817857e-05, "data_time": 0.4894073963165283, "loss": 0.07951398901641368, "time": 0.6290619373321533, "epoch": 9, "memory": 6319, "step": 33756} +{"lr": 1.389879855605952e-05, "data_time": 0.4867011308670044, "loss": 0.0786565899848938, "time": 0.6274277210235596, "epoch": 9, "memory": 6319, "step": 33856} +{"lr": 1.3752803863288976e-05, "data_time": 0.5118851661682129, "loss": 0.08616845607757569, "time": 0.6523105144500733, "epoch": 9, "memory": 6319, "step": 33956} +{"lr": 1.3609476059977129e-05, "data_time": 0.5192718029022216, "loss": 0.07865990065038204, "time": 0.6592938423156738, "epoch": 9, "memory": 6319, "step": 34056} +{"lr": 1.3468824413163754e-05, "data_time": 0.5041179180145263, "loss": 0.08080047145485877, "time": 0.6560468435287475, "epoch": 9, "memory": 6319, "step": 34156} +{"lr": 1.3330858016858395e-05, "data_time": 0.45170934200286866, "loss": 0.08205296695232392, "time": 0.5917628526687622, "epoch": 9, "memory": 6319, "step": 34256} +{"lr": 1.3195585791452309e-05, "data_time": 0.5187618970870972, "loss": 0.07742132656276227, "time": 0.6678142786026001, "epoch": 9, "memory": 6319, "step": 34356} +{"lr": 1.3063016483141729e-05, "data_time": 0.4949553728103638, "loss": 0.08371957205235958, "time": 0.6356351613998413, "epoch": 9, "memory": 6319, "step": 34456} +{"lr": 1.2933158663362388e-05, "data_time": 0.4888258934020996, "loss": 0.08621455803513527, "time": 0.6414131641387939, "epoch": 9, "memory": 6319, "step": 34556} +{"lr": 1.2806020728235311e-05, "data_time": 0.594412112236023, "loss": 0.08536653108894825, "time": 0.7345845222473144, "epoch": 9, "memory": 6319, "step": 34656} +{"lr": 1.2681610898023963e-05, "data_time": 0.5329106092453003, "loss": 0.09632372669875622, "time": 0.6734464645385743, "epoch": 9, "memory": 6319, "step": 34756} +{"lr": 1.2559937216602731e-05, "data_time": 0.5210883617401123, "loss": 0.07627506218850613, "time": 0.662468695640564, "epoch": 9, "memory": 6319, "step": 34856} +{"lr": 1.2441007550936847e-05, "data_time": 0.5392831802368164, "loss": 0.07872199155390262, "time": 0.6801229238510131, "epoch": 9, "memory": 6319, "step": 34956} +{"lr": 1.232482959057377e-05, "data_time": 0.5266360759735107, "loss": 0.0835390716791153, "time": 0.6657192230224609, "epoch": 9, "memory": 6319, "step": 35056} +{"lr": 1.2211410847145969e-05, "data_time": 0.5051618576049804, "loss": 0.08262104205787182, "time": 0.6445920467376709, "epoch": 9, "memory": 6319, "step": 35156} +{"accuracy/top1": 96.47541809082031, "data_time": 0.518794454377273, "time": 0.608899108294783, "step": 9} +{"lr": 1.209311679360931e-05, "data_time": 0.5300727128982544, "loss": 0.08774049505591393, "time": 0.6700115919113159, "epoch": 10, "memory": 6319, "step": 35263} +{"lr": 1.1985432727488666e-05, "data_time": 0.5218451499938965, "loss": 0.08216948658227921, "time": 0.6757283926010131, "epoch": 10, "memory": 6319, "step": 35363} +{"lr": 1.1880529822435289e-05, "data_time": 0.5013918161392212, "loss": 0.07124878317117692, "time": 0.6406710147857666, "epoch": 10, "memory": 6319, "step": 35463} +{"lr": 1.1778414861078726e-05, "data_time": 0.47354414463043215, "loss": 0.08191103041172028, "time": 0.6248251914978027, "epoch": 10, "memory": 6319, "step": 35563} +{"lr": 1.1679094445790538e-05, "data_time": 0.4894792795181274, "loss": 0.08240576237440109, "time": 0.6288114786148071, "epoch": 10, "memory": 6319, "step": 35663} +{"lr": 1.1582574998257377e-05, "data_time": 0.536252498626709, "loss": 0.08340882994234562, "time": 0.6897312641143799, "epoch": 10, "memory": 6319, "step": 35763} +{"lr": 1.1488862759065762e-05, "data_time": 0.5399514675140381, "loss": 0.08530356548726559, "time": 0.6838498592376709, "epoch": 10, "memory": 6319, "step": 35863} +{"lr": 1.1397963787298665e-05, "data_time": 0.599949049949646, "loss": 0.09712032303214073, "time": 0.7402030467987061, "epoch": 10, "memory": 6319, "step": 35963} +{"lr": 1.1309883960143664e-05, "data_time": 0.5113953351974487, "loss": 0.08439582176506519, "time": 0.6521112442016601, "epoch": 10, "memory": 6319, "step": 36063} +{"lr": 1.122462897251302e-05, "data_time": 0.5624451398849487, "loss": 0.07773884236812592, "time": 0.7040619611740112, "epoch": 10, "memory": 6319, "step": 36163} +{"lr": 1.114220433667542e-05, "data_time": 0.5429295301437378, "loss": 0.08018908873200417, "time": 0.683778953552246, "epoch": 10, "memory": 6319, "step": 36263} +{"lr": 1.10626153818996e-05, "data_time": 0.4776179790496826, "loss": 0.089403111115098, "time": 0.618587875366211, "epoch": 10, "memory": 6319, "step": 36363} +{"lr": 1.0985867254109761e-05, "data_time": 0.5083762407302856, "loss": 0.08239412233233452, "time": 0.6485461473464966, "epoch": 10, "memory": 6319, "step": 36463} +{"lr": 1.0911964915552823e-05, "data_time": 0.4959545612335205, "loss": 0.09672980047762395, "time": 0.6369888305664062, "epoch": 10, "memory": 6319, "step": 36563} +{"lr": 1.0840913144477643e-05, "data_time": 0.5138628721237183, "loss": 0.07901668697595596, "time": 0.6529916286468506, "epoch": 10, "memory": 6319, "step": 36663} +{"lr": 1.0772716534826064e-05, "data_time": 0.4700314998626709, "loss": 0.08116241209208966, "time": 0.6110960245132446, "epoch": 10, "memory": 6319, "step": 36763} +{"lr": 1.0707379495935828e-05, "data_time": 0.537388277053833, "loss": 0.0771650567650795, "time": 0.6768936395645142, "epoch": 10, "memory": 6319, "step": 36863} +{"lr": 1.0644906252255541e-05, "data_time": 0.5458524703979493, "loss": 0.06829882338643074, "time": 0.6856985569000245, "epoch": 10, "memory": 6319, "step": 36963} +{"lr": 1.0585300843071527e-05, "data_time": 0.5325053215026856, "loss": 0.06929968297481537, "time": 0.6731296062469483, "epoch": 10, "memory": 6319, "step": 37063} +{"lr": 1.0528567122246654e-05, "data_time": 0.5284550189971924, "loss": 0.07455556318163872, "time": 0.6679382801055909, "epoch": 10, "memory": 6319, "step": 37163} +{"lr": 1.0474708757971141e-05, "data_time": 0.4974109172821045, "loss": 0.08358317576348781, "time": 0.6382056713104248, "epoch": 10, "memory": 6319, "step": 37263} +{"lr": 1.042372923252543e-05, "data_time": 0.4840532302856445, "loss": 0.08695609383285045, "time": 0.624877643585205, "epoch": 10, "memory": 6319, "step": 37363} +{"lr": 1.0375631842054998e-05, "data_time": 0.5609840869903564, "loss": 0.0942134540528059, "time": 0.702863335609436, "epoch": 10, "memory": 6319, "step": 37463} +{"lr": 1.0330419696357254e-05, "data_time": 0.5242627382278442, "loss": 0.07821050956845284, "time": 0.6635823011398315, "epoch": 10, "memory": 6319, "step": 37563} +{"lr": 1.028809571868047e-05, "data_time": 0.5104862451553345, "loss": 0.08879930265247822, "time": 0.6513925552368164, "epoch": 10, "memory": 6319, "step": 37663} +{"lr": 1.0248662645534766e-05, "data_time": 0.5124234437942505, "loss": 0.08167126402258873, "time": 0.6535289525985718, "epoch": 10, "memory": 6319, "step": 37763} +{"lr": 1.0212123026515194e-05, "data_time": 0.5410007953643798, "loss": 0.08757947683334351, "time": 0.6814837694168091, "epoch": 10, "memory": 6319, "step": 37863} +{"lr": 1.0178479224136881e-05, "data_time": 0.532672643661499, "loss": 0.08730786256492137, "time": 0.6741935253143311, "epoch": 10, "memory": 6319, "step": 37963} +{"lr": 1.0147733413682289e-05, "data_time": 0.5444990396499634, "loss": 0.08508696109056473, "time": 0.6847944974899292, "epoch": 10, "memory": 6319, "step": 38063} +{"lr": 1.0119887583060527e-05, "data_time": 0.6378956317901612, "loss": 0.08408012986183167, "time": 0.7786592006683349, "epoch": 10, "memory": 6319, "step": 38163} +{"lr": 1.0094943532678871e-05, "data_time": 0.5177739858627319, "loss": 0.09481669887900353, "time": 0.6586864709854126, "epoch": 10, "memory": 6319, "step": 38263} +{"lr": 1.007290287532637e-05, "data_time": 0.5099339962005616, "loss": 0.09078535623848438, "time": 0.6501465559005737, "epoch": 10, "memory": 6319, "step": 38363} +{"lr": 1.0053767036069495e-05, "data_time": 0.5393283367156982, "loss": 0.07345308773219586, "time": 0.679974102973938, "epoch": 10, "memory": 6319, "step": 38463} +{"lr": 1.003753725216005e-05, "data_time": 0.5076316356658935, "loss": 0.08990654423832893, "time": 0.648727560043335, "epoch": 10, "memory": 6319, "step": 38563} +{"lr": 1.0024214572955152e-05, "data_time": 0.5073602676391602, "loss": 0.08032020516693592, "time": 0.6476858854293823, "epoch": 10, "memory": 6319, "step": 38663} +{"lr": 1.0013799859849456e-05, "data_time": 0.5041482448577881, "loss": 0.08838330432772637, "time": 0.6441276550292969, "epoch": 10, "memory": 6319, "step": 38763} +{"lr": 1.0006293786219338e-05, "data_time": 0.5216606855392456, "loss": 0.0832119818776846, "time": 0.6623240947723389, "epoch": 10, "memory": 6319, "step": 38863} +{"lr": 1.0001696837379472e-05, "data_time": 0.5376900196075439, "loss": 0.08422540500760078, "time": 0.6791685342788696, "epoch": 10, "memory": 6319, "step": 38963} +{"lr": 1.0000009310551361e-05, "data_time": 0.5254184484481812, "loss": 0.0833253562450409, "time": 0.6661343574523926, "epoch": 10, "memory": 6319, "step": 39063} +{"accuracy/top1": 96.86858367919922, "data_time": 0.5282502462124002, "time": 0.6142095245164017, "step": 10} diff --git a/resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1/20230604_225351/vis_data/vis_image/0000000.png_1.png b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1/20230604_225351/vis_data/vis_image/0000000.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..60da8a5cb3dbe6f10487d04a2acc99020c4aef62 Binary files /dev/null and b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1/20230604_225351/vis_data/vis_image/0000000.png_1.png differ diff --git a/resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1/20230604_225351/vis_data/vis_image/0000000.png_10.png b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1/20230604_225351/vis_data/vis_image/0000000.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..1b8dbeeaa406dba8934dcc5cef313e77a1e569de Binary files /dev/null and b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1/20230604_225351/vis_data/vis_image/0000000.png_10.png differ diff --git a/resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1/20230604_225351/vis_data/vis_image/0000000.png_2.png b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1/20230604_225351/vis_data/vis_image/0000000.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..f6203e5c37ca49428789a05411c8d3ef63723487 Binary files /dev/null and b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1/20230604_225351/vis_data/vis_image/0000000.png_2.png differ diff --git a/resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1/20230604_225351/vis_data/vis_image/0000000.png_3.png b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1/20230604_225351/vis_data/vis_image/0000000.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..9165f941801c6315e2065cfd3b53ca675c9c3757 Binary files /dev/null and b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1/20230604_225351/vis_data/vis_image/0000000.png_3.png differ diff --git a/resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1/20230604_225351/vis_data/vis_image/0000000.png_4.png b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1/20230604_225351/vis_data/vis_image/0000000.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..e8d322b7fe496c15f44f5d13008685a1d0efc0c3 Binary files /dev/null and b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1/20230604_225351/vis_data/vis_image/0000000.png_4.png differ diff --git a/resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1/20230604_225351/vis_data/vis_image/0000000.png_5.png b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1/20230604_225351/vis_data/vis_image/0000000.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..1ce7cb346c4c360172e0a4b29d7fb12b974b6834 Binary files /dev/null and b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1/20230604_225351/vis_data/vis_image/0000000.png_5.png differ diff --git a/resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1/20230604_225351/vis_data/vis_image/0000000.png_6.png b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1/20230604_225351/vis_data/vis_image/0000000.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..21da43171daa0eece2def205125ad3812f332259 Binary files /dev/null and b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1/20230604_225351/vis_data/vis_image/0000000.png_6.png differ diff --git a/resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1/20230604_225351/vis_data/vis_image/0000000.png_7.png b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1/20230604_225351/vis_data/vis_image/0000000.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..d2239dae4bfdf10116ed84e5e721f6a6580e0cd9 Binary files /dev/null and b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1/20230604_225351/vis_data/vis_image/0000000.png_7.png differ diff --git a/resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1/20230604_225351/vis_data/vis_image/0000000.png_8.png b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1/20230604_225351/vis_data/vis_image/0000000.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..4d1ae491b691944431b445ed8fa3dcd8e34d1583 Binary files /dev/null and b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1/20230604_225351/vis_data/vis_image/0000000.png_8.png differ diff --git a/resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1/20230604_225351/vis_data/vis_image/0000000.png_9.png b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1/20230604_225351/vis_data/vis_image/0000000.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..ec70419c0ba5e1587f3eb044798d5ea28e3b6cc7 Binary files /dev/null and b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1/20230604_225351/vis_data/vis_image/0000000.png_9.png differ diff --git a/resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1/20230604_225351/vis_data/vis_image/000005174.jpg_1.png b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1/20230604_225351/vis_data/vis_image/000005174.jpg_1.png new file mode 100644 index 0000000000000000000000000000000000000000..6cfb84c0b460f7b7dc1fd77d831e64902c0c700e Binary files /dev/null and b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1/20230604_225351/vis_data/vis_image/000005174.jpg_1.png differ diff --git a/resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1/20230604_225351/vis_data/vis_image/000005174.jpg_10.png b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1/20230604_225351/vis_data/vis_image/000005174.jpg_10.png new file mode 100644 index 0000000000000000000000000000000000000000..8e9e50e0a0e8293c4b96fd188a1b1ba8a01f0c96 Binary files /dev/null and b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1/20230604_225351/vis_data/vis_image/000005174.jpg_10.png differ diff --git a/resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1/20230604_225351/vis_data/vis_image/000005174.jpg_2.png b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1/20230604_225351/vis_data/vis_image/000005174.jpg_2.png new file mode 100644 index 0000000000000000000000000000000000000000..fe90290da97f29a126f0370571cabf81bd439856 Binary files /dev/null and b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1/20230604_225351/vis_data/vis_image/000005174.jpg_2.png differ diff --git a/resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1/20230604_225351/vis_data/vis_image/000005174.jpg_3.png b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1/20230604_225351/vis_data/vis_image/000005174.jpg_3.png new file mode 100644 index 0000000000000000000000000000000000000000..4ae5cb7c6f2981be41de3a21ffcaef762179bdea Binary files /dev/null and b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1/20230604_225351/vis_data/vis_image/000005174.jpg_3.png differ diff --git a/resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1/20230604_225351/vis_data/vis_image/000005174.jpg_4.png b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1/20230604_225351/vis_data/vis_image/000005174.jpg_4.png new file mode 100644 index 0000000000000000000000000000000000000000..a7925943a508ca88c53be91bb799338aaabf2b61 Binary files /dev/null and b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1/20230604_225351/vis_data/vis_image/000005174.jpg_4.png differ diff --git a/resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1/20230604_225351/vis_data/vis_image/000005174.jpg_5.png b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1/20230604_225351/vis_data/vis_image/000005174.jpg_5.png new file mode 100644 index 0000000000000000000000000000000000000000..73439492f5389eeb6e441ef68612ddd51fdbb59f Binary files /dev/null and b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1/20230604_225351/vis_data/vis_image/000005174.jpg_5.png differ diff --git a/resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1/20230604_225351/vis_data/vis_image/000005174.jpg_6.png b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1/20230604_225351/vis_data/vis_image/000005174.jpg_6.png new file mode 100644 index 0000000000000000000000000000000000000000..5b1e9cd1db838b93ca0bf94f21705e1e6fdbdb5f Binary files /dev/null and b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1/20230604_225351/vis_data/vis_image/000005174.jpg_6.png differ diff --git a/resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1/20230604_225351/vis_data/vis_image/000005174.jpg_7.png b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1/20230604_225351/vis_data/vis_image/000005174.jpg_7.png new file mode 100644 index 0000000000000000000000000000000000000000..46e409c0c7733929896e7d989869612772d39d9d Binary files /dev/null and b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1/20230604_225351/vis_data/vis_image/000005174.jpg_7.png differ diff --git a/resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1/20230604_225351/vis_data/vis_image/000005174.jpg_8.png b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1/20230604_225351/vis_data/vis_image/000005174.jpg_8.png new file mode 100644 index 0000000000000000000000000000000000000000..137ff2577b03ec3da5327acabc5777c2d09f25ea Binary files /dev/null and b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1/20230604_225351/vis_data/vis_image/000005174.jpg_8.png differ diff --git a/resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1/20230604_225351/vis_data/vis_image/000005174.jpg_9.png b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1/20230604_225351/vis_data/vis_image/000005174.jpg_9.png new file mode 100644 index 0000000000000000000000000000000000000000..49fde26d2eda4a8c9b8433082469b9dfb8b6a06e Binary files /dev/null and b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1/20230604_225351/vis_data/vis_image/000005174.jpg_9.png differ diff --git a/resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1/20230604_225351/vis_data/vis_image/0010000.png_1.png b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1/20230604_225351/vis_data/vis_image/0010000.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..9d68b04d11d413910ef41c88ec7686dcb9e444c5 Binary files /dev/null and b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1/20230604_225351/vis_data/vis_image/0010000.png_1.png differ diff --git a/resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1/20230604_225351/vis_data/vis_image/0010000.png_10.png b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1/20230604_225351/vis_data/vis_image/0010000.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..05bc7171e3329c68b7610f7c4e8a96c618ae3945 Binary files /dev/null and b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1/20230604_225351/vis_data/vis_image/0010000.png_10.png differ diff --git a/resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1/20230604_225351/vis_data/vis_image/0010000.png_2.png b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1/20230604_225351/vis_data/vis_image/0010000.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..e6ae2dcfea0cb2f31c50b6f1d76f73fe6f5a60b0 Binary files /dev/null and b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1/20230604_225351/vis_data/vis_image/0010000.png_2.png differ diff --git a/resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1/20230604_225351/vis_data/vis_image/0010000.png_3.png b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1/20230604_225351/vis_data/vis_image/0010000.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..8573db94cadbb6bb2fde1b86cd7415c1f3a56976 Binary files /dev/null and b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1/20230604_225351/vis_data/vis_image/0010000.png_3.png differ diff --git a/resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1/20230604_225351/vis_data/vis_image/0010000.png_4.png b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1/20230604_225351/vis_data/vis_image/0010000.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..3eb13a1355a503ea6fe69d45bb821a3d4811df7f Binary files /dev/null and b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1/20230604_225351/vis_data/vis_image/0010000.png_4.png differ diff --git a/resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1/20230604_225351/vis_data/vis_image/0010000.png_5.png b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1/20230604_225351/vis_data/vis_image/0010000.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..8f3612ea507ff4697ff6eae96fea1d99bf0d1337 Binary files /dev/null and b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1/20230604_225351/vis_data/vis_image/0010000.png_5.png differ diff --git a/resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1/20230604_225351/vis_data/vis_image/0010000.png_6.png b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1/20230604_225351/vis_data/vis_image/0010000.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..fa7bafdd73ac4308474f575978d65cab45b06e8e Binary files /dev/null and b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1/20230604_225351/vis_data/vis_image/0010000.png_6.png differ diff --git a/resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1/20230604_225351/vis_data/vis_image/0010000.png_7.png b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1/20230604_225351/vis_data/vis_image/0010000.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..5af604dfd059d65d85f9b1a0b7ad9f03c774bba9 Binary files /dev/null and b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1/20230604_225351/vis_data/vis_image/0010000.png_7.png differ diff --git a/resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1/20230604_225351/vis_data/vis_image/0010000.png_8.png b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1/20230604_225351/vis_data/vis_image/0010000.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..8ed0e6bf21f7211b960a039051204ffcc93ebfd2 Binary files /dev/null and b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1/20230604_225351/vis_data/vis_image/0010000.png_8.png differ diff --git a/resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1/20230604_225351/vis_data/vis_image/0010000.png_9.png b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1/20230604_225351/vis_data/vis_image/0010000.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..4b058c21a7d706f8c7ab680d79de7751bbaaa0cc Binary files /dev/null and b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1/20230604_225351/vis_data/vis_image/0010000.png_9.png differ diff --git a/resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1/epoch_1.pth b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1/epoch_1.pth new file mode 100644 index 0000000000000000000000000000000000000000..afaca4683c6df8c9d8f2046e780781ac910f109a --- /dev/null +++ b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1/epoch_1.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:a584958da36396bd0d77ffe2b344fd2777a9c0f3a964eb677879e5a54e7e487d +size 277331277 diff --git a/resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1/epoch_10.pth b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1/epoch_10.pth new file mode 100644 index 0000000000000000000000000000000000000000..706bf9c90a0108aef5302c41ed1297bea1f6b540 --- /dev/null +++ b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1/epoch_10.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:26eee75387027866bdfce589a77546c6c0b9ff5221d2ad9cd80a0f6008df115d +size 279901517 diff --git a/resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1/epoch_2.pth b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1/epoch_2.pth new file mode 100644 index 0000000000000000000000000000000000000000..ea301b5755e6fdef42b91124e5037384026c211b --- /dev/null +++ b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1/epoch_2.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ea45a0ce8f53ead968689ca24eeb1ed4f693920670f1580ab6348c40c947f52b +size 277616205 diff --git a/resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1/epoch_3.pth b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1/epoch_3.pth new file mode 100644 index 0000000000000000000000000000000000000000..98bcc121b7f97238e3bda00eecbd535ea507e0a0 --- /dev/null +++ b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1/epoch_3.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ccdae9690dbf33133bdff4ab654a401e0fefebf70057de6f3cb8444dfd872315 +size 277900941 diff --git a/resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1/epoch_4.pth b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1/epoch_4.pth new file mode 100644 index 0000000000000000000000000000000000000000..a6218045ab30e500c6c76e803bde63eda8b9f47b --- /dev/null +++ b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1/epoch_4.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:dde72f98187f08e55e07caa47b4647b0466f4bf64d703a639bf55f434676b430 +size 278185037 diff --git a/resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1/epoch_5.pth b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1/epoch_5.pth new file mode 100644 index 0000000000000000000000000000000000000000..dadc80d85b2772f07a73cd0a50c06bcf6aa8bc45 --- /dev/null +++ b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1/epoch_5.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:556df6df41e293ca1906f93835ddf4675ff17da33d4d9abe56f1a7a8a0093912 +size 278468749 diff --git a/resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1/epoch_6.pth b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1/epoch_6.pth new file mode 100644 index 0000000000000000000000000000000000000000..a006f6f69afe9e30a89634abba139809505b0a47 --- /dev/null +++ b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1/epoch_6.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ef15c2d429e9bcd51922e028e5b1e3f152ebce8a3f14b8170ef514ccea3ecb12 +size 278752717 diff --git a/resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1/epoch_7.pth b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1/epoch_7.pth new file mode 100644 index 0000000000000000000000000000000000000000..53a14514ba78c187e179c43e7a4687f04d2ec39a --- /dev/null +++ b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1/epoch_7.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:623f0f5dbba6a9d3be84863afbd9f264d3d351539a6e346e21359999023fd17c +size 279037389 diff --git a/resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1/epoch_8.pth b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1/epoch_8.pth new file mode 100644 index 0000000000000000000000000000000000000000..365f409094d612a99a4a9c28f3f5f7f81badedbf --- /dev/null +++ b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1/epoch_8.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:b034d7ed40b55d31324e443b26a3ca68740834598781ac65b3975181001e95b4 +size 279325261 diff --git a/resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1/epoch_9.pth b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1/epoch_9.pth new file mode 100644 index 0000000000000000000000000000000000000000..56534c7fb4a313c9ed09acb68dbf1085b419c78c --- /dev/null +++ b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1/epoch_9.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:f59cc4b2265916c72961a0bb7ab87efe129585995c158d4d0549f1b1d6a0a974 +size 279613389 diff --git a/resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1/last_checkpoint b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1/last_checkpoint new file mode 100644 index 0000000000000000000000000000000000000000..11f4d19daf3f4655c2139fec7e30bd8223b3be1d --- /dev/null +++ b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1/last_checkpoint @@ -0,0 +1 @@ +/mnt/petrelfs/luzeyu/workspace/fakebench/mmpretrain/workdir/resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1/epoch_10.pth \ No newline at end of file diff --git a/resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1/resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1.py b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1/resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1.py new file mode 100644 index 0000000000000000000000000000000000000000..6664519846f1271ef814a9610846d4ebd7a6cc83 --- /dev/null +++ b/resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1/resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1.py @@ -0,0 +1,230 @@ +optim_wrapper = dict( + optimizer=dict( + type='SGD', + lr=0.0001, + momentum=0.9, + weight_decay=0.0001, + _scope_='mmpretrain'), + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=False, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=512) +model = dict( + type='ImageClassifier', + backbone=dict( + frozen_stages=2, + type='ResNet', + depth=50, + num_stages=4, + out_indices=(3, ), + style='pytorch', + init_cfg=dict( + type='Pretrained', + checkpoint= + 'https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth', + prefix='backbone')), + neck=dict(type='GlobalAveragePooling'), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=2048, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + topk=1)) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stablediffusionV2-1-dpmsolver-25-1m.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/cc1m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV2-1-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = dict(type='Accuracy', topk=1) +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV2-1-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/cc1w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = dict(type='Accuracy', topk=1) +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/resnet50_2xb256_sdv2_1m_lr1e-4_aug_5e-1' diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/20230605_010029.log b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/20230605_010029.log new file mode 100644 index 0000000000000000000000000000000000000000..08a11e73b369869c417d75c10f573f30fa028efc --- /dev/null +++ b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/20230605_010029.log @@ -0,0 +1,939 @@ +2023/06/05 01:00:32 - mmengine - INFO - +------------------------------------------------------------ +System environment: + sys.platform: linux + Python: 3.10.9 (main, Mar 8 2023, 10:47:38) [GCC 11.2.0] + CUDA available: True + numpy_random_seed: 1449923474 + GPU 0,1: NVIDIA A100-SXM4-80GB + CUDA_HOME: /mnt/petrelfs/share/cuda-11.6 + NVCC: Cuda compilation tools, release 11.6, V11.6.124 + GCC: gcc (GCC) 7.5.0 + PyTorch: 1.13.1 + PyTorch compiling details: PyTorch built with: + - GCC 9.3 + - C++ Version: 201402 + - Intel(R) oneAPI Math Kernel Library Version 2021.4-Product Build 20210904 for Intel(R) 64 architecture applications + - Intel(R) MKL-DNN v2.6.0 (Git Hash 52b5f107dd9cf10910aaa19cb47f3abf9b349815) + - OpenMP 201511 (a.k.a. OpenMP 4.5) + - LAPACK is enabled (usually provided by MKL) + - NNPACK is enabled + - CPU capability usage: AVX2 + - CUDA Runtime 11.6 + - NVCC architecture flags: -gencode;arch=compute_37,code=sm_37;-gencode;arch=compute_50,code=sm_50;-gencode;arch=compute_60,code=sm_60;-gencode;arch=compute_61,code=sm_61;-gencode;arch=compute_70,code=sm_70;-gencode;arch=compute_75,code=sm_75;-gencode;arch=compute_80,code=sm_80;-gencode;arch=compute_86,code=sm_86;-gencode;arch=compute_37,code=compute_37 + - CuDNN 8.3.2 (built against CUDA 11.5) + - Magma 2.6.1 + - Build settings: BLAS_INFO=mkl, BUILD_TYPE=Release, CUDA_VERSION=11.6, CUDNN_VERSION=8.3.2, CXX_COMPILER=/opt/rh/devtoolset-9/root/usr/bin/c++, CXX_FLAGS= -fabi-version=11 -Wno-deprecated -fvisibility-inlines-hidden -DUSE_PTHREADPOOL -fopenmp -DNDEBUG -DUSE_KINETO -DUSE_FBGEMM -DUSE_QNNPACK -DUSE_PYTORCH_QNNPACK -DUSE_XNNPACK -DSYMBOLICATE_MOBILE_DEBUG_HANDLE -DEDGE_PROFILER_USE_KINETO -O2 -fPIC -Wno-narrowing -Wall -Wextra -Werror=return-type -Werror=non-virtual-dtor -Wno-missing-field-initializers -Wno-type-limits -Wno-array-bounds -Wno-unknown-pragmas -Wunused-local-typedefs -Wno-unused-parameter -Wno-unused-function -Wno-unused-result -Wno-strict-overflow -Wno-strict-aliasing -Wno-error=deprecated-declarations -Wno-stringop-overflow -Wno-psabi -Wno-error=pedantic -Wno-error=redundant-decls -Wno-error=old-style-cast -fdiagnostics-color=always -faligned-new -Wno-unused-but-set-variable -Wno-maybe-uninitialized -fno-math-errno -fno-trapping-math -Werror=format -Werror=cast-function-type -Wno-stringop-overflow, LAPACK_INFO=mkl, PERF_WITH_AVX=1, PERF_WITH_AVX2=1, PERF_WITH_AVX512=1, TORCH_VERSION=1.13.1, USE_CUDA=ON, USE_CUDNN=ON, USE_EXCEPTION_PTR=1, USE_GFLAGS=OFF, USE_GLOG=OFF, USE_MKL=ON, USE_MKLDNN=ON, USE_MPI=OFF, USE_NCCL=ON, USE_NNPACK=ON, USE_OPENMP=ON, USE_ROCM=OFF, + + TorchVision: 0.14.1 + OpenCV: 4.7.0 + MMEngine: 0.7.3 + +Runtime environment: + cudnn_benchmark: True + mp_cfg: {'mp_start_method': 'fork', 'opencv_num_threads': 0} + dist_cfg: {'backend': 'nccl'} + seed: None + deterministic: False + Distributed launcher: slurm + Distributed training: True + GPU number: 2 +------------------------------------------------------------ + +2023/06/05 01:00:37 - mmengine - INFO - Config: +optim_wrapper = dict( + optimizer=dict( + type='SGD', + lr=0.0001, + momentum=0.9, + weight_decay=0.0001, + _scope_='mmpretrain'), + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=False, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=512) +model = dict( + type='ImageClassifier', + backbone=dict( + frozen_stages=2, + type='ResNet', + depth=50, + num_stages=4, + out_indices=(3, ), + style='pytorch', + init_cfg=dict( + type='Pretrained', + checkpoint= + 'https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth', + prefix='backbone')), + neck=dict(type='GlobalAveragePooling'), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=2048, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + topk=1)) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3fake8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3real8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3fake8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3real8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = dict(type='Accuracy', topk=1) +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3fake8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3real8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = dict(type='Accuracy', topk=1) +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1' + +2023/06/05 01:00:49 - mmengine - INFO - Hooks will be executed in the following order: +before_run: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_load_checkpoint: +(ABOVE_NORMAL) EMAHook + -------------------- +before_train: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_train_epoch: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(NORMAL ) DistSamplerSeedHook + -------------------- +before_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook + -------------------- +after_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +after_train_epoch: +(NORMAL ) IterTimerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_val_epoch: +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook + -------------------- +before_val_iter: +(NORMAL ) IterTimerHook + -------------------- +after_val_iter: +(NORMAL ) IterTimerHook +(NORMAL ) VisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_val_epoch: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_save_checkpoint: +(ABOVE_NORMAL) EMAHook + -------------------- +after_train: +(VERY_LOW ) CheckpointHook + -------------------- +before_test_epoch: +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook + -------------------- +before_test_iter: +(NORMAL ) IterTimerHook + -------------------- +after_test_iter: +(NORMAL ) IterTimerHook +(NORMAL ) VisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_test_epoch: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_run: +(BELOW_NORMAL) LoggerHook + -------------------- +2023/06/05 01:01:00 - mmengine - INFO - load backbone in model from: https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth +Name of parameter - Initialization information + +backbone.conv1.weight - torch.Size([64, 3, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.bn1.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.bn1.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.conv1.weight - torch.Size([64, 64, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.bn1.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.bn1.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.conv2.weight - torch.Size([64, 64, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.bn2.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.bn2.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.conv3.weight - torch.Size([256, 64, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.bn3.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.bn3.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.downsample.0.weight - torch.Size([256, 64, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.downsample.1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.downsample.1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.conv1.weight - torch.Size([64, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.bn1.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.bn1.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.conv2.weight - torch.Size([64, 64, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.bn2.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.bn2.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.conv3.weight - torch.Size([256, 64, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.bn3.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.bn3.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.conv1.weight - torch.Size([64, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.bn1.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.bn1.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.conv2.weight - torch.Size([64, 64, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.bn2.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.bn2.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.conv3.weight - torch.Size([256, 64, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.bn3.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.bn3.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.conv1.weight - torch.Size([128, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.bn1.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.bn1.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.conv2.weight - torch.Size([128, 128, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.bn2.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.bn2.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.conv3.weight - torch.Size([512, 128, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.bn3.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.bn3.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.downsample.0.weight - torch.Size([512, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.downsample.1.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.downsample.1.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.conv1.weight - torch.Size([128, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.bn1.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.bn1.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.conv2.weight - torch.Size([128, 128, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.bn2.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.bn2.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.conv3.weight - torch.Size([512, 128, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.bn3.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.bn3.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.conv1.weight - torch.Size([128, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.bn1.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.bn1.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.conv2.weight - torch.Size([128, 128, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.bn2.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.bn2.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.conv3.weight - torch.Size([512, 128, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.bn3.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.bn3.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.conv1.weight - torch.Size([128, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.bn1.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.bn1.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.conv2.weight - torch.Size([128, 128, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.bn2.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.bn2.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.conv3.weight - torch.Size([512, 128, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.bn3.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.bn3.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.conv1.weight - torch.Size([256, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.bn1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.bn1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.conv2.weight - torch.Size([256, 256, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.bn2.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.bn2.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.conv3.weight - torch.Size([1024, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.bn3.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.bn3.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.downsample.0.weight - torch.Size([1024, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.downsample.1.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.downsample.1.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.conv1.weight - torch.Size([256, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.bn1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.bn1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.conv2.weight - torch.Size([256, 256, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.bn2.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.bn2.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.conv3.weight - torch.Size([1024, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.bn3.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.bn3.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.conv1.weight - torch.Size([256, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.bn1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.bn1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.conv2.weight - torch.Size([256, 256, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.bn2.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.bn2.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.conv3.weight - torch.Size([1024, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.bn3.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.bn3.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.conv1.weight - torch.Size([256, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.bn1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.bn1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.conv2.weight - torch.Size([256, 256, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.bn2.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.bn2.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.conv3.weight - torch.Size([1024, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.bn3.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.bn3.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.conv1.weight - torch.Size([256, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.bn1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.bn1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.conv2.weight - torch.Size([256, 256, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.bn2.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.bn2.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.conv3.weight - torch.Size([1024, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.bn3.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.bn3.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.conv1.weight - torch.Size([256, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.bn1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.bn1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.conv2.weight - torch.Size([256, 256, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.bn2.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.bn2.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.conv3.weight - torch.Size([1024, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.bn3.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.bn3.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.conv1.weight - torch.Size([512, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.bn1.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.bn1.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.conv2.weight - torch.Size([512, 512, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.bn2.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.bn2.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.conv3.weight - torch.Size([2048, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.bn3.weight - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.bn3.bias - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.downsample.0.weight - torch.Size([2048, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.downsample.1.weight - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.downsample.1.bias - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.conv1.weight - torch.Size([512, 2048, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.bn1.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.bn1.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.conv2.weight - torch.Size([512, 512, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.bn2.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.bn2.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.conv3.weight - torch.Size([2048, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.bn3.weight - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.bn3.bias - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.conv1.weight - torch.Size([512, 2048, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.bn1.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.bn1.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.conv2.weight - torch.Size([512, 512, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.bn2.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.bn2.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.conv3.weight - torch.Size([2048, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.bn3.weight - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.bn3.bias - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +head.fc.weight - torch.Size([2, 2048]): +NormalInit: mean=0, std=0.01, bias=0 + +head.fc.bias - torch.Size([2]): +NormalInit: mean=0, std=0.01, bias=0 +2023/06/05 01:01:00 - mmengine - WARNING - "FileClient" will be deprecated in future. Please use io functions in https://mmengine.readthedocs.io/en/latest/api/fileio.html#file-io +2023/06/05 01:01:00 - mmengine - WARNING - "HardDiskBackend" is the alias of "LocalBackend" and the former will be deprecated in future. +2023/06/05 01:01:00 - mmengine - INFO - Checkpoints will be saved to /mnt/petrelfs/luzeyu/workspace/fakebench/mmpretrain/workdir/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1. +2023/06/05 01:03:00 - mmengine - INFO - Epoch(train) [1][100/342] lr: 9.9814e-05 eta: 1:06:20 time: 1.1679 data_time: 1.0271 memory: 9436 loss: 0.6834 +2023/06/05 01:04:56 - mmengine - INFO - Epoch(train) [1][200/342] lr: 9.9250e-05 eta: 1:03:13 time: 1.1562 data_time: 1.0170 memory: 6319 loss: 0.6645 +2023/06/05 01:06:52 - mmengine - INFO - Epoch(train) [1][300/342] lr: 9.8313e-05 eta: 1:00:54 time: 1.1148 data_time: 0.9758 memory: 6319 loss: 0.6610 +2023/06/05 01:07:45 - mmengine - INFO - Exp name: resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1_20230605_010029 +2023/06/05 01:07:45 - mmengine - INFO - Saving checkpoint at 1 epochs +2023/06/05 01:09:40 - mmengine - INFO - Epoch(val) [1][100/342] eta: 0:04:26 time: 0.4016 data_time: 0.3114 memory: 6319 +2023/06/05 01:11:17 - mmengine - INFO - Epoch(val) [1][200/342] eta: 0:02:26 time: 0.2278 data_time: 0.1332 memory: 3133 +2023/06/05 01:13:19 - mmengine - INFO - Epoch(val) [1][300/342] eta: 0:00:45 time: 1.1740 data_time: 1.0847 memory: 3133 +2023/06/05 01:15:53 - mmengine - INFO - Epoch(val) [1][342/342] accuracy/top1: 49.9027 data_time: 1.2237 time: 1.3139 +2023/06/05 01:17:50 - mmengine - INFO - Epoch(train) [2][100/342] lr: 9.6358e-05 eta: 0:58:38 time: 1.0999 data_time: 0.9593 memory: 6319 loss: 0.6385 +2023/06/05 01:19:45 - mmengine - INFO - Epoch(train) [2][200/342] lr: 9.4557e-05 eta: 0:56:20 time: 1.1884 data_time: 1.0444 memory: 6319 loss: 0.6376 +2023/06/05 01:21:41 - mmengine - INFO - Epoch(train) [2][300/342] lr: 9.2422e-05 eta: 0:54:16 time: 1.0861 data_time: 0.9449 memory: 6319 loss: 0.6336 +2023/06/05 01:22:30 - mmengine - INFO - Exp name: resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1_20230605_010029 +2023/06/05 01:22:30 - mmengine - INFO - Saving checkpoint at 2 epochs +2023/06/05 01:24:25 - mmengine - INFO - Epoch(val) [2][100/342] eta: 0:04:25 time: 0.3972 data_time: 0.3094 memory: 6319 +2023/06/05 01:26:04 - mmengine - INFO - Epoch(val) [2][200/342] eta: 0:02:27 time: 0.2260 data_time: 0.1382 memory: 3133 +2023/06/05 01:28:06 - mmengine - INFO - Epoch(val) [2][300/342] eta: 0:00:46 time: 1.2512 data_time: 1.1626 memory: 3133 +2023/06/05 01:30:45 - mmengine - INFO - Epoch(val) [2][342/342] accuracy/top1: 50.6954 data_time: 1.2377 time: 1.3283 +2023/06/05 01:32:43 - mmengine - INFO - Epoch(train) [3][100/342] lr: 8.8853e-05 eta: 0:51:35 time: 1.2035 data_time: 1.0648 memory: 6319 loss: 0.6195 +2023/06/05 01:34:38 - mmengine - INFO - Epoch(train) [3][200/342] lr: 8.5991e-05 eta: 0:49:30 time: 1.1861 data_time: 1.0464 memory: 6319 loss: 0.6067 +2023/06/05 01:36:33 - mmengine - INFO - Epoch(train) [3][300/342] lr: 8.2867e-05 eta: 0:47:28 time: 1.0694 data_time: 0.9306 memory: 6319 loss: 0.6024 +2023/06/05 01:36:57 - mmengine - INFO - Exp name: resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1_20230605_010029 +2023/06/05 01:37:29 - mmengine - INFO - Exp name: resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1_20230605_010029 +2023/06/05 01:37:29 - mmengine - INFO - Saving checkpoint at 3 epochs +2023/06/05 01:39:23 - mmengine - INFO - Epoch(val) [3][100/342] eta: 0:04:22 time: 0.4047 data_time: 0.3157 memory: 6319 +2023/06/05 01:41:01 - mmengine - INFO - Epoch(val) [3][200/342] eta: 0:02:26 time: 0.2308 data_time: 0.1435 memory: 3133 +2023/06/05 01:43:06 - mmengine - INFO - Epoch(val) [3][300/342] eta: 0:00:46 time: 1.3063 data_time: 1.2168 memory: 3133 +2023/06/05 01:45:42 - mmengine - INFO - Epoch(val) [3][342/342] accuracy/top1: 51.7817 data_time: 1.2334 time: 1.3234 +2023/06/05 01:47:43 - mmengine - INFO - Epoch(train) [4][100/342] lr: 7.8035e-05 eta: 0:45:04 time: 1.1833 data_time: 1.0419 memory: 6319 loss: 0.5972 +2023/06/05 01:49:40 - mmengine - INFO - Epoch(train) [4][200/342] lr: 7.4392e-05 eta: 0:43:04 time: 1.3588 data_time: 1.2166 memory: 6319 loss: 0.5935 +2023/06/05 01:51:37 - mmengine - INFO - Epoch(train) [4][300/342] lr: 7.0585e-05 eta: 0:41:05 time: 1.1882 data_time: 1.0475 memory: 6319 loss: 0.5727 +2023/06/05 01:52:24 - mmengine - INFO - Exp name: resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1_20230605_010029 +2023/06/05 01:52:24 - mmengine - INFO - Saving checkpoint at 4 epochs +2023/06/05 01:54:18 - mmengine - INFO - Epoch(val) [4][100/342] eta: 0:04:21 time: 0.4173 data_time: 0.3294 memory: 6319 +2023/06/05 01:55:56 - mmengine - INFO - Epoch(val) [4][200/342] eta: 0:02:26 time: 0.2102 data_time: 0.1233 memory: 3133 +2023/06/05 01:58:02 - mmengine - INFO - Epoch(val) [4][300/342] eta: 0:00:46 time: 1.3082 data_time: 1.2188 memory: 3133 +2023/06/05 02:00:39 - mmengine - INFO - Epoch(val) [4][342/342] accuracy/top1: 52.7364 data_time: 1.2381 time: 1.3284 +2023/06/05 02:02:38 - mmengine - INFO - Epoch(train) [5][100/342] lr: 6.4962e-05 eta: 0:38:18 time: 1.2908 data_time: 1.1481 memory: 6319 loss: 0.5702 +2023/06/05 02:04:34 - mmengine - INFO - Epoch(train) [5][200/342] lr: 6.0894e-05 eta: 0:36:18 time: 1.1949 data_time: 1.0548 memory: 6319 loss: 0.5646 +2023/06/05 02:06:31 - mmengine - INFO - Epoch(train) [5][300/342] lr: 5.6777e-05 eta: 0:34:20 time: 1.1639 data_time: 1.0214 memory: 6319 loss: 0.5818 +2023/06/05 02:07:21 - mmengine - INFO - Exp name: resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1_20230605_010029 +2023/06/05 02:07:21 - mmengine - INFO - Saving checkpoint at 5 epochs +2023/06/05 02:09:16 - mmengine - INFO - Epoch(val) [5][100/342] eta: 0:04:26 time: 0.4118 data_time: 0.3230 memory: 6319 +2023/06/05 02:10:54 - mmengine - INFO - Epoch(val) [5][200/342] eta: 0:02:27 time: 0.2274 data_time: 0.1406 memory: 3133 +2023/06/05 02:12:57 - mmengine - INFO - Epoch(val) [5][300/342] eta: 0:00:46 time: 1.2464 data_time: 1.1581 memory: 3133 +2023/06/05 02:15:26 - mmengine - INFO - Epoch(val) [5][342/342] accuracy/top1: 53.2750 data_time: 1.2156 time: 1.3059 +2023/06/05 02:17:26 - mmengine - INFO - Epoch(train) [6][100/342] lr: 5.0913e-05 eta: 0:31:35 time: 1.1552 data_time: 1.0126 memory: 6319 loss: 0.5454 +2023/06/05 02:19:20 - mmengine - INFO - Epoch(train) [6][200/342] lr: 4.6820e-05 eta: 0:29:34 time: 1.1261 data_time: 0.9853 memory: 6319 loss: 0.5420 +2023/06/05 02:21:07 - mmengine - INFO - Exp name: resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1_20230605_010029 +2023/06/05 02:21:19 - mmengine - INFO - Epoch(train) [6][300/342] lr: 4.2795e-05 eta: 0:27:38 time: 1.1874 data_time: 1.0425 memory: 6319 loss: 0.5400 +2023/06/05 02:22:13 - mmengine - INFO - Exp name: resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1_20230605_010029 +2023/06/05 02:22:13 - mmengine - INFO - Saving checkpoint at 6 epochs +2023/06/05 02:24:07 - mmengine - INFO - Epoch(val) [6][100/342] eta: 0:04:22 time: 0.4121 data_time: 0.3234 memory: 6319 +2023/06/05 02:25:45 - mmengine - INFO - Epoch(val) [6][200/342] eta: 0:02:26 time: 0.2273 data_time: 0.1401 memory: 3133 +2023/06/05 02:27:51 - mmengine - INFO - Epoch(val) [6][300/342] eta: 0:00:46 time: 1.2495 data_time: 1.1605 memory: 3133 +2023/06/05 02:30:23 - mmengine - INFO - Epoch(val) [6][342/342] accuracy/top1: 53.4781 data_time: 1.2244 time: 1.3152 +2023/06/05 02:32:21 - mmengine - INFO - Epoch(train) [7][100/342] lr: 3.7265e-05 eta: 0:24:54 time: 1.2204 data_time: 1.0790 memory: 6319 loss: 0.5369 +2023/06/05 02:34:17 - mmengine - INFO - Epoch(train) [7][200/342] lr: 3.3546e-05 eta: 0:22:55 time: 1.1635 data_time: 1.0226 memory: 6319 loss: 0.5399 +2023/06/05 02:36:11 - mmengine - INFO - Epoch(train) [7][300/342] lr: 3.0008e-05 eta: 0:20:55 time: 1.1193 data_time: 0.9791 memory: 6319 loss: 0.5242 +2023/06/05 02:37:00 - mmengine - INFO - Exp name: resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1_20230605_010029 +2023/06/05 02:37:00 - mmengine - INFO - Saving checkpoint at 7 epochs +2023/06/05 02:38:54 - mmengine - INFO - Epoch(val) [7][100/342] eta: 0:04:24 time: 0.4059 data_time: 0.3119 memory: 6319 +2023/06/05 02:40:32 - mmengine - INFO - Epoch(val) [7][200/342] eta: 0:02:26 time: 0.2178 data_time: 0.1299 memory: 3133 +2023/06/05 02:42:35 - mmengine - INFO - Epoch(val) [7][300/342] eta: 0:00:46 time: 1.2517 data_time: 1.1632 memory: 3133 +2023/06/05 02:45:07 - mmengine - INFO - Epoch(val) [7][342/342] accuracy/top1: 53.6676 data_time: 1.2182 time: 1.3087 +2023/06/05 02:47:06 - mmengine - INFO - Epoch(train) [8][100/342] lr: 2.5353e-05 eta: 0:18:09 time: 1.2057 data_time: 1.0666 memory: 6319 loss: 0.5182 +2023/06/05 02:49:03 - mmengine - INFO - Epoch(train) [8][200/342] lr: 2.2372e-05 eta: 0:16:11 time: 1.2061 data_time: 1.0665 memory: 6319 loss: 0.5216 +2023/06/05 02:51:00 - mmengine - INFO - Epoch(train) [8][300/342] lr: 1.9667e-05 eta: 0:14:13 time: 1.1136 data_time: 0.9733 memory: 6319 loss: 0.5161 +2023/06/05 02:51:55 - mmengine - INFO - Exp name: resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1_20230605_010029 +2023/06/05 02:51:55 - mmengine - INFO - Saving checkpoint at 8 epochs +2023/06/05 02:53:49 - mmengine - INFO - Epoch(val) [8][100/342] eta: 0:04:22 time: 0.4135 data_time: 0.3130 memory: 6319 +2023/06/05 02:55:26 - mmengine - INFO - Epoch(val) [8][200/342] eta: 0:02:25 time: 0.2042 data_time: 0.1175 memory: 3133 +2023/06/05 02:57:30 - mmengine - INFO - Epoch(val) [8][300/342] eta: 0:00:46 time: 1.2102 data_time: 1.1209 memory: 3133 +2023/06/05 02:59:57 - mmengine - INFO - Epoch(val) [8][342/342] accuracy/top1: 53.7941 data_time: 1.2053 time: 1.2954 +2023/06/05 03:01:54 - mmengine - INFO - Epoch(train) [9][100/342] lr: 1.6343e-05 eta: 0:11:27 time: 1.1193 data_time: 0.9801 memory: 6319 loss: 0.5225 +2023/06/05 03:03:49 - mmengine - INFO - Epoch(train) [9][200/342] lr: 1.4393e-05 eta: 0:09:29 time: 1.1506 data_time: 1.0099 memory: 6319 loss: 0.5106 +2023/06/05 03:05:15 - mmengine - INFO - Exp name: resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1_20230605_010029 +2023/06/05 03:05:52 - mmengine - INFO - Epoch(train) [9][300/342] lr: 1.2785e-05 eta: 0:07:32 time: 1.2044 data_time: 1.0639 memory: 6319 loss: 0.5137 +2023/06/05 03:06:48 - mmengine - INFO - Exp name: resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1_20230605_010029 +2023/06/05 03:06:48 - mmengine - INFO - Saving checkpoint at 9 epochs +2023/06/05 03:08:41 - mmengine - INFO - Epoch(val) [9][100/342] eta: 0:04:22 time: 0.4250 data_time: 0.3369 memory: 6319 +2023/06/05 03:10:19 - mmengine - INFO - Epoch(val) [9][200/342] eta: 0:02:26 time: 0.2294 data_time: 0.1354 memory: 3133 +2023/06/05 03:12:20 - mmengine - INFO - Epoch(val) [9][300/342] eta: 0:00:45 time: 1.1748 data_time: 1.0864 memory: 3133 +2023/06/05 03:14:54 - mmengine - INFO - Epoch(val) [9][342/342] accuracy/top1: 53.7105 data_time: 1.2163 time: 1.3066 +2023/06/05 03:16:52 - mmengine - INFO - Epoch(train) [10][100/342] lr: 1.1116e-05 eta: 0:04:45 time: 1.1325 data_time: 0.9899 memory: 6319 loss: 0.5064 +2023/06/05 03:18:46 - mmengine - INFO - Epoch(train) [10][200/342] lr: 1.0388e-05 eta: 0:02:47 time: 1.0886 data_time: 0.9468 memory: 6319 loss: 0.5054 +2023/06/05 03:20:40 - mmengine - INFO - Epoch(train) [10][300/342] lr: 1.0035e-05 eta: 0:00:49 time: 1.1395 data_time: 0.9989 memory: 6319 loss: 0.5024 +2023/06/05 03:21:30 - mmengine - INFO - Exp name: resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1_20230605_010029 +2023/06/05 03:21:30 - mmengine - INFO - Saving checkpoint at 10 epochs +2023/06/05 03:23:23 - mmengine - INFO - Epoch(val) [10][100/342] eta: 0:04:22 time: 0.4172 data_time: 0.3268 memory: 6319 +2023/06/05 03:25:01 - mmengine - INFO - Epoch(val) [10][200/342] eta: 0:02:26 time: 0.2240 data_time: 0.1370 memory: 3133 +2023/06/05 03:27:04 - mmengine - INFO - Epoch(val) [10][300/342] eta: 0:00:46 time: 1.2277 data_time: 1.1386 memory: 3133 +2023/06/05 03:29:32 - mmengine - INFO - Epoch(val) [10][342/342] accuracy/top1: 53.8090 data_time: 1.2051 time: 1.2951 diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/20230605_010029.json b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/20230605_010029.json new file mode 100644 index 0000000000000000000000000000000000000000..ed88c6704260df3848815f98240ed1c3c7c70931 --- /dev/null +++ b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/20230605_010029.json @@ -0,0 +1,40 @@ +{"lr": 9.981404789023513e-05, "data_time": 1.0271079301834107, "loss": 0.6833862006664276, "time": 1.1679381608963013, "epoch": 1, "memory": 9436, "step": 100} +{"lr": 9.925023268502683e-05, "data_time": 1.0169579267501831, "loss": 0.6645129442214965, "time": 1.1561670303344727, "epoch": 1, "memory": 6319, "step": 200} +{"lr": 9.831328961451971e-05, "data_time": 0.9758212566375732, "loss": 0.6609851539134979, "time": 1.1147706985473633, "epoch": 1, "memory": 6319, "step": 300} +{"accuracy/top1": 49.90270233154297, "data_time": 1.223721544644986, "time": 1.313911643641734, "step": 1} +{"lr": 9.635783271319408e-05, "data_time": 0.9593441724777222, "loss": 0.6384726643562317, "time": 1.0999414920806885, "epoch": 2, "memory": 6319, "step": 442} +{"lr": 9.455662799327337e-05, "data_time": 1.0444109201431275, "loss": 0.6376423180103302, "time": 1.1883981227874756, "epoch": 2, "memory": 6319, "step": 542} +{"lr": 9.242187293325357e-05, "data_time": 0.9449031829833985, "loss": 0.6336082577705383, "time": 1.0861395359039308, "epoch": 2, "memory": 6319, "step": 642} +{"accuracy/top1": 50.69540023803711, "data_time": 1.2376895360974458, "time": 1.3283191234655352, "step": 2} +{"lr": 8.885322471322101e-05, "data_time": 1.0647746801376343, "loss": 0.6194638431072235, "time": 1.2034748315811157, "epoch": 3, "memory": 6319, "step": 784} +{"lr": 8.59909449463049e-05, "data_time": 1.0464270830154419, "loss": 0.6067410171031952, "time": 1.1861287832260132, "epoch": 3, "memory": 6319, "step": 884} +{"lr": 8.28673425957603e-05, "data_time": 0.9305630207061768, "loss": 0.602357542514801, "time": 1.0694422006607056, "epoch": 3, "memory": 6319, "step": 984} +{"accuracy/top1": 51.78171157836914, "data_time": 1.2333615625564975, "time": 1.323392610160672, "step": 3} +{"lr": 7.803482720903195e-05, "data_time": 1.0419308423995972, "loss": 0.5972189784049988, "time": 1.183283805847168, "epoch": 4, "memory": 6319, "step": 1126} +{"lr": 7.439165228138202e-05, "data_time": 1.216626238822937, "loss": 0.5935378432273865, "time": 1.3588289976119996, "epoch": 4, "memory": 6319, "step": 1226} +{"lr": 7.058496260180133e-05, "data_time": 1.0475386142730714, "loss": 0.5726555049419403, "time": 1.188169574737549, "epoch": 4, "memory": 6319, "step": 1326} +{"accuracy/top1": 52.73638153076172, "data_time": 1.2381047020848222, "time": 1.3284218074976528, "step": 4} +{"lr": 6.496162032454504e-05, "data_time": 1.148072862625122, "loss": 0.5702099859714508, "time": 1.2908209323883058, "epoch": 5, "memory": 6319, "step": 1468} +{"lr": 6.089416958157208e-05, "data_time": 1.0547701358795165, "loss": 0.5645862698554993, "time": 1.1949419021606444, "epoch": 5, "memory": 6319, "step": 1568} +{"lr": 5.677701788155902e-05, "data_time": 1.0214194536209107, "loss": 0.5818164110183716, "time": 1.1639350175857544, "epoch": 5, "memory": 6319, "step": 1668} +{"accuracy/top1": 53.27495574951172, "data_time": 1.21558803163534, "time": 1.3059362058389292, "step": 5} +{"lr": 5.091330063600123e-05, "data_time": 1.0125902891159058, "loss": 0.5454251766204834, "time": 1.1552147150039673, "epoch": 6, "memory": 6319, "step": 1810} +{"lr": 4.681972449602271e-05, "data_time": 0.985312557220459, "loss": 0.5420195996761322, "time": 1.1260746240615844, "epoch": 6, "memory": 6319, "step": 1910} +{"lr": 4.27951262698573e-05, "data_time": 1.0424684524536132, "loss": 0.5400137960910797, "time": 1.1873953342437744, "epoch": 6, "memory": 6319, "step": 2010} +{"accuracy/top1": 53.4781379699707, "data_time": 1.2243830427831532, "time": 1.3152291163063605, "step": 6} +{"lr": 3.726501555491419e-05, "data_time": 1.0790399074554444, "loss": 0.5368733644485474, "time": 1.2204342842102052, "epoch": 7, "memory": 6319, "step": 2152} +{"lr": 3.354602177213381e-05, "data_time": 1.0225852727890015, "loss": 0.5399499863386155, "time": 1.163516068458557, "epoch": 7, "memory": 6319, "step": 2252} +{"lr": 3.0007932735217966e-05, "data_time": 0.9791089057922363, "loss": 0.5241945743560791, "time": 1.1192760229110719, "epoch": 7, "memory": 6319, "step": 2352} +{"accuracy/top1": 53.66758346557617, "data_time": 1.2182474115499602, "time": 1.3086557235384126, "step": 7} +{"lr": 2.5352754318215113e-05, "data_time": 1.0666207790374755, "loss": 0.5181624472141266, "time": 1.2057128429412842, "epoch": 8, "memory": 6319, "step": 2494} +{"lr": 2.2372383915844294e-05, "data_time": 1.0664579629898072, "loss": 0.5216139316558838, "time": 1.2061090469360352, "epoch": 8, "memory": 6319, "step": 2594} +{"lr": 1.9667136874426842e-05, "data_time": 0.9732812166213989, "loss": 0.5160503000020981, "time": 1.1135627508163453, "epoch": 8, "memory": 6319, "step": 2694} +{"accuracy/top1": 53.794071197509766, "data_time": 1.2053193170197156, "time": 1.2953845613551904, "step": 8} +{"lr": 1.6342572053355464e-05, "data_time": 0.9800638437271119, "loss": 0.5225351750850677, "time": 1.119275736808777, "epoch": 9, "memory": 6319, "step": 2836} +{"lr": 1.4392564451840171e-05, "data_time": 1.0098625898361206, "loss": 0.5105531573295593, "time": 1.1505921840667725, "epoch": 9, "memory": 6319, "step": 2936} +{"lr": 1.2784967834899793e-05, "data_time": 1.0639393329620361, "loss": 0.5137149631977082, "time": 1.204388976097107, "epoch": 9, "memory": 6319, "step": 3036} +{"accuracy/top1": 53.71051025390625, "data_time": 1.2163165213415297, "time": 1.3065960428110017, "step": 9} +{"lr": 1.1116448178051759e-05, "data_time": 0.9899445533752441, "loss": 0.5064258933067322, "time": 1.1324551105499268, "epoch": 10, "memory": 6319, "step": 3178} +{"lr": 1.0387683707930187e-05, "data_time": 0.9467991352081299, "loss": 0.5054368585348129, "time": 1.0886245012283324, "epoch": 10, "memory": 6319, "step": 3278} +{"lr": 1.0035100273117234e-05, "data_time": 0.9988670110702514, "loss": 0.502412137389183, "time": 1.1394683122634888, "epoch": 10, "memory": 6319, "step": 3378} +{"accuracy/top1": 53.80895233154297, "data_time": 1.2051049773269076, "time": 1.295068000565465, "step": 10} diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/config.py b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/config.py new file mode 100644 index 0000000000000000000000000000000000000000..f37931e24e882a176c705bb4cec9cc535f363f49 --- /dev/null +++ b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/config.py @@ -0,0 +1,230 @@ +optim_wrapper = dict( + optimizer=dict( + type='SGD', + lr=0.0001, + momentum=0.9, + weight_decay=0.0001, + _scope_='mmpretrain'), + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=False, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=512) +model = dict( + type='ImageClassifier', + backbone=dict( + frozen_stages=2, + type='ResNet', + depth=50, + num_stages=4, + out_indices=(3, ), + style='pytorch', + init_cfg=dict( + type='Pretrained', + checkpoint= + 'https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth', + prefix='backbone')), + neck=dict(type='GlobalAveragePooling'), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=2048, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + topk=1)) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3fake8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3real8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3fake8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3real8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = dict(type='Accuracy', topk=1) +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3fake8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3real8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = dict(type='Accuracy', topk=1) +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1' diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/events.out.tfevents.1685898043.SH-IDC1-10-140-24-106.218263.0 b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/events.out.tfevents.1685898043.SH-IDC1-10-140-24-106.218263.0 new file mode 100644 index 0000000000000000000000000000000000000000..3b05eeb2a9ea5618b0d8482c3d57949074e844f0 --- /dev/null +++ b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/events.out.tfevents.1685898043.SH-IDC1-10-140-24-106.218263.0 @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:37f31d49b3380d4186b58b6234bb4a242c999be7b2f4035cbc033a640d9254e9 +size 14445258 diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/scalars.json b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/scalars.json new file mode 100644 index 0000000000000000000000000000000000000000..ed88c6704260df3848815f98240ed1c3c7c70931 --- /dev/null +++ b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/scalars.json @@ -0,0 +1,40 @@ +{"lr": 9.981404789023513e-05, "data_time": 1.0271079301834107, "loss": 0.6833862006664276, "time": 1.1679381608963013, "epoch": 1, "memory": 9436, "step": 100} +{"lr": 9.925023268502683e-05, "data_time": 1.0169579267501831, "loss": 0.6645129442214965, "time": 1.1561670303344727, "epoch": 1, "memory": 6319, "step": 200} +{"lr": 9.831328961451971e-05, "data_time": 0.9758212566375732, "loss": 0.6609851539134979, "time": 1.1147706985473633, "epoch": 1, "memory": 6319, "step": 300} +{"accuracy/top1": 49.90270233154297, "data_time": 1.223721544644986, "time": 1.313911643641734, "step": 1} +{"lr": 9.635783271319408e-05, "data_time": 0.9593441724777222, "loss": 0.6384726643562317, "time": 1.0999414920806885, "epoch": 2, "memory": 6319, "step": 442} +{"lr": 9.455662799327337e-05, "data_time": 1.0444109201431275, "loss": 0.6376423180103302, "time": 1.1883981227874756, "epoch": 2, "memory": 6319, "step": 542} +{"lr": 9.242187293325357e-05, "data_time": 0.9449031829833985, "loss": 0.6336082577705383, "time": 1.0861395359039308, "epoch": 2, "memory": 6319, "step": 642} +{"accuracy/top1": 50.69540023803711, "data_time": 1.2376895360974458, "time": 1.3283191234655352, "step": 2} +{"lr": 8.885322471322101e-05, "data_time": 1.0647746801376343, "loss": 0.6194638431072235, "time": 1.2034748315811157, "epoch": 3, "memory": 6319, "step": 784} +{"lr": 8.59909449463049e-05, "data_time": 1.0464270830154419, "loss": 0.6067410171031952, "time": 1.1861287832260132, "epoch": 3, "memory": 6319, "step": 884} +{"lr": 8.28673425957603e-05, "data_time": 0.9305630207061768, "loss": 0.602357542514801, "time": 1.0694422006607056, "epoch": 3, "memory": 6319, "step": 984} +{"accuracy/top1": 51.78171157836914, "data_time": 1.2333615625564975, "time": 1.323392610160672, "step": 3} +{"lr": 7.803482720903195e-05, "data_time": 1.0419308423995972, "loss": 0.5972189784049988, "time": 1.183283805847168, "epoch": 4, "memory": 6319, "step": 1126} +{"lr": 7.439165228138202e-05, "data_time": 1.216626238822937, "loss": 0.5935378432273865, "time": 1.3588289976119996, "epoch": 4, "memory": 6319, "step": 1226} +{"lr": 7.058496260180133e-05, "data_time": 1.0475386142730714, "loss": 0.5726555049419403, "time": 1.188169574737549, "epoch": 4, "memory": 6319, "step": 1326} +{"accuracy/top1": 52.73638153076172, "data_time": 1.2381047020848222, "time": 1.3284218074976528, "step": 4} +{"lr": 6.496162032454504e-05, "data_time": 1.148072862625122, "loss": 0.5702099859714508, "time": 1.2908209323883058, "epoch": 5, "memory": 6319, "step": 1468} +{"lr": 6.089416958157208e-05, "data_time": 1.0547701358795165, "loss": 0.5645862698554993, "time": 1.1949419021606444, "epoch": 5, "memory": 6319, "step": 1568} +{"lr": 5.677701788155902e-05, "data_time": 1.0214194536209107, "loss": 0.5818164110183716, "time": 1.1639350175857544, "epoch": 5, "memory": 6319, "step": 1668} +{"accuracy/top1": 53.27495574951172, "data_time": 1.21558803163534, "time": 1.3059362058389292, "step": 5} +{"lr": 5.091330063600123e-05, "data_time": 1.0125902891159058, "loss": 0.5454251766204834, "time": 1.1552147150039673, "epoch": 6, "memory": 6319, "step": 1810} +{"lr": 4.681972449602271e-05, "data_time": 0.985312557220459, "loss": 0.5420195996761322, "time": 1.1260746240615844, "epoch": 6, "memory": 6319, "step": 1910} +{"lr": 4.27951262698573e-05, "data_time": 1.0424684524536132, "loss": 0.5400137960910797, "time": 1.1873953342437744, "epoch": 6, "memory": 6319, "step": 2010} +{"accuracy/top1": 53.4781379699707, "data_time": 1.2243830427831532, "time": 1.3152291163063605, "step": 6} +{"lr": 3.726501555491419e-05, "data_time": 1.0790399074554444, "loss": 0.5368733644485474, "time": 1.2204342842102052, "epoch": 7, "memory": 6319, "step": 2152} +{"lr": 3.354602177213381e-05, "data_time": 1.0225852727890015, "loss": 0.5399499863386155, "time": 1.163516068458557, "epoch": 7, "memory": 6319, "step": 2252} +{"lr": 3.0007932735217966e-05, "data_time": 0.9791089057922363, "loss": 0.5241945743560791, "time": 1.1192760229110719, "epoch": 7, "memory": 6319, "step": 2352} +{"accuracy/top1": 53.66758346557617, "data_time": 1.2182474115499602, "time": 1.3086557235384126, "step": 7} +{"lr": 2.5352754318215113e-05, "data_time": 1.0666207790374755, "loss": 0.5181624472141266, "time": 1.2057128429412842, "epoch": 8, "memory": 6319, "step": 2494} +{"lr": 2.2372383915844294e-05, "data_time": 1.0664579629898072, "loss": 0.5216139316558838, "time": 1.2061090469360352, "epoch": 8, "memory": 6319, "step": 2594} +{"lr": 1.9667136874426842e-05, "data_time": 0.9732812166213989, "loss": 0.5160503000020981, "time": 1.1135627508163453, "epoch": 8, "memory": 6319, "step": 2694} +{"accuracy/top1": 53.794071197509766, "data_time": 1.2053193170197156, "time": 1.2953845613551904, "step": 8} +{"lr": 1.6342572053355464e-05, "data_time": 0.9800638437271119, "loss": 0.5225351750850677, "time": 1.119275736808777, "epoch": 9, "memory": 6319, "step": 2836} +{"lr": 1.4392564451840171e-05, "data_time": 1.0098625898361206, "loss": 0.5105531573295593, "time": 1.1505921840667725, "epoch": 9, "memory": 6319, "step": 2936} +{"lr": 1.2784967834899793e-05, "data_time": 1.0639393329620361, "loss": 0.5137149631977082, "time": 1.204388976097107, "epoch": 9, "memory": 6319, "step": 3036} +{"accuracy/top1": 53.71051025390625, "data_time": 1.2163165213415297, "time": 1.3065960428110017, "step": 9} +{"lr": 1.1116448178051759e-05, "data_time": 0.9899445533752441, "loss": 0.5064258933067322, "time": 1.1324551105499268, "epoch": 10, "memory": 6319, "step": 3178} +{"lr": 1.0387683707930187e-05, "data_time": 0.9467991352081299, "loss": 0.5054368585348129, "time": 1.0886245012283324, "epoch": 10, "memory": 6319, "step": 3278} +{"lr": 1.0035100273117234e-05, "data_time": 0.9988670110702514, "loss": 0.502412137389183, "time": 1.1394683122634888, "epoch": 10, "memory": 6319, "step": 3378} +{"accuracy/top1": 53.80895233154297, "data_time": 1.2051049773269076, "time": 1.295068000565465, "step": 10} diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/35624.png_1.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/35624.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..27523089c6d19f629532aa03185aac91ef2ce0e9 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/35624.png_1.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/35624.png_10.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/35624.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..fae2ec2223eb3e4caa37cd2b56f12c3c5c10ba31 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/35624.png_10.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/35624.png_2.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/35624.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..01dc140b4a64601bb9b7a9d63404f6d45009553c Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/35624.png_2.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/35624.png_3.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/35624.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..58523b23ae00b99366e2e6475373a71774b46315 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/35624.png_3.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/35624.png_4.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/35624.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..d566b370dd95d5385cbac6e87c59acb2fac6190d Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/35624.png_4.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/35624.png_5.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/35624.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..9037d362bb1feed7b5aaabfd49c8869c1740c3ca Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/35624.png_5.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/35624.png_6.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/35624.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..3bac659997f34738a390c2458a17de3bbedd3c76 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/35624.png_6.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/35624.png_7.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/35624.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..c87740d4b802f3349765ad4ff93f488d6e62562c Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/35624.png_7.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/35624.png_8.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/35624.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..9e93d8d8d6c4c792cfd74d25e2593520bd355fd6 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/35624.png_8.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/35624.png_9.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/35624.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..1019492b875f4be4beef66009e3c0229cdf8e908 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/35624.png_9.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/38442.png_1.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/38442.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..f000a7585cbed69fa074f814a0cf4e2a9abfb872 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/38442.png_1.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/38442.png_10.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/38442.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..6354b69a0f295db2b33482cd3a0191404ed85d28 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/38442.png_10.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/38442.png_2.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/38442.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..f09fab4b6af9d67bdf87155d14b79624b3dd6e1f Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/38442.png_2.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/38442.png_3.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/38442.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..bb26b8d8475e20a1a12b64298e5716a90d068def Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/38442.png_3.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/38442.png_4.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/38442.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..54b1527ff470fcf8e56921340a4ce5d8661930a9 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/38442.png_4.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/38442.png_5.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/38442.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..d530f04655acea36321da149d17865b98b9ec50c Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/38442.png_5.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/38442.png_6.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/38442.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..f085f9a206d59284eb13cc3af9ab5f36c67f4abe Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/38442.png_6.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/38442.png_7.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/38442.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..e4090c170f2a9d6ce5552d6df6b84f925937caa8 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/38442.png_7.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/38442.png_8.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/38442.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..e69c93b9f0b5fdf9d603a60ab874b9b48b528b5f Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/38442.png_8.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/38442.png_9.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/38442.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..5b97baa748e9cb79299b0917653137009150c5a3 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/38442.png_9.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/43283.png_1.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/43283.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..1b6bf1a6a243b8843aa8b286d0a9fbc7b393c143 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/43283.png_1.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/43283.png_10.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/43283.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..21338e342cc939f9688bfcbcc88b5c9b316333eb Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/43283.png_10.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/43283.png_2.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/43283.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..db1d1f7a7db8a50667140481744c4ca4398a84d0 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/43283.png_2.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/43283.png_3.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/43283.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..919210cac8288912a9cce39ff107e77c1c4e87c8 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/43283.png_3.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/43283.png_4.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/43283.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..219cbddf5d302ddcbe1a40530ddd16946539b200 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/43283.png_4.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/43283.png_5.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/43283.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..7f64475084b9e89d38b03b77d3c4e23c179adc01 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/43283.png_5.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/43283.png_6.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/43283.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..bb0d44abe44f4f829f205ca07282ed9c98478e44 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/43283.png_6.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/43283.png_7.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/43283.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..5ba4065d75bf8663947e770bedf67bce00e8dbab Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/43283.png_7.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/43283.png_8.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/43283.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..97664c06b0261ebb202ec2126dda809cf37a6d67 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/43283.png_8.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/43283.png_9.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/43283.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..297511d18cd3e41bf4faba64a08157c4d87904a9 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/43283.png_9.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/44982.png_1.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/44982.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..97ba0937c8a4866080fc464eb1e96a6c4804eee5 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/44982.png_1.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/44982.png_10.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/44982.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..590824ed8c17c684e40223f02aa584741bccf234 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/44982.png_10.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/44982.png_2.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/44982.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..8a9f673345ea7420afaece1fc81f2c7cc4ad9ad2 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/44982.png_2.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/44982.png_3.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/44982.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..3fe0434c20b867a75cd2c9cabb7d9031e85557ea Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/44982.png_3.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/44982.png_4.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/44982.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..2f0e4e036c98c06cdc3c50443c7b9d375e1f071e Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/44982.png_4.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/44982.png_5.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/44982.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..6a89ec34f8357c6a616a6519abbcab8eb4ff3990 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/44982.png_5.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/44982.png_6.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/44982.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..d7f05d232f5f51244d90049a6a6475423bf49d7f Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/44982.png_6.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/44982.png_7.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/44982.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..b71e8c4cceceb3f22e24671eeb224904724836a0 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/44982.png_7.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/44982.png_8.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/44982.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..a6abf6ca31e290f972e24853a65602a498033e4c Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/44982.png_8.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/44982.png_9.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/44982.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..8e0d01ceced08e4d3c5b3bdcb5e80eaea296b7cb Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/44982.png_9.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/56423.png_1.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/56423.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..936132a1e4cb34d56b00309afa516fd3514b6adc Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/56423.png_1.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/56423.png_10.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/56423.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..caf0ac08b5af00dc2db6bdc1c38cf44298a51ed1 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/56423.png_10.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/56423.png_2.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/56423.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..225d5b7b801d114b291043083fd64d0bfa7384d3 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/56423.png_2.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/56423.png_3.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/56423.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..d5ddb30c74468dbbc17dfc68f93b002b2471221e Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/56423.png_3.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/56423.png_4.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/56423.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..b75e2422d50e9af291b9b5a5d06ccc7ed7127348 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/56423.png_4.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/56423.png_5.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/56423.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..e1d9ada39ef40538db0f762fd86fc47808e27a82 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/56423.png_5.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/56423.png_6.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/56423.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..de57043da37c4535e457da77df456868c125c935 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/56423.png_6.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/56423.png_7.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/56423.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..4f78521b4c8eed6cc0db4c1cce8e4b0c0eb83bcc Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/56423.png_7.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/56423.png_8.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/56423.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..1e4763879a08cfd25f1e95fe30821b1f758d8fce Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/56423.png_8.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/56423.png_9.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/56423.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..358e64f89c2285589151ec860e543e3a9055cf69 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/56423.png_9.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/61024.png_1.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/61024.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..ee3ec7f2e2bec2b0a313bd4b99d882b50a7a5f10 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/61024.png_1.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/61024.png_10.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/61024.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..1040191617cdb37e4bdc8de44db5b642547d309c Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/61024.png_10.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/61024.png_2.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/61024.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..6fef139b693b00376e4d20abe15d5168fa136a76 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/61024.png_2.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/61024.png_3.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/61024.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..ec5ce5c136469ee2d31ddd0b2009dbdb50e65b6c Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/61024.png_3.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/61024.png_4.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/61024.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..8bd6de920b6f5178093002a18fd484cb4572d0a5 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/61024.png_4.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/61024.png_5.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/61024.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..8b6d3ccc0f65d846c0d8b39f7cca71390f44eb4c Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/61024.png_5.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/61024.png_6.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/61024.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..2b69a41d806172a6b93460560994ce7d8cfadc35 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/61024.png_6.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/61024.png_7.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/61024.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..cc1f1721cc165b7641cef06be6105eec67eafa55 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/61024.png_7.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/61024.png_8.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/61024.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..9b2865ff4124ff55170b0f5b5b856d09f60ce323 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/61024.png_8.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/61024.png_9.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/61024.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..fc0fb7fb4843c828b03417517667def92d003328 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/61024.png_9.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/66390.png_1.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/66390.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..103d3d5e9885bb28d504415aa8956b81ef2cd576 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/66390.png_1.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/66390.png_10.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/66390.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..4ee0b12410a95b778c470f62dbae4df63552231e Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/66390.png_10.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/66390.png_2.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/66390.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..a6839b306792b461e634f53515a099162363c25f Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/66390.png_2.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/66390.png_3.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/66390.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..93e25f0b4e0b357c29655096a76cb8883c41f50f Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/66390.png_3.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/66390.png_4.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/66390.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..c2eba064a9dce70211391290de81246636f32a6c Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/66390.png_4.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/66390.png_5.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/66390.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..a2174ef64081d631572708203b0a47ff69d0e6ba Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/66390.png_5.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/66390.png_6.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/66390.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..4b8d7c90aaa39c3b5ba9daedafe282fe429ffb7f Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/66390.png_6.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/66390.png_7.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/66390.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..9b736d49ef6511200a4d83a01cbe7e7f201e0d9d Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/66390.png_7.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/66390.png_8.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/66390.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..1fcdfb0a1d2299caf2bf182b84c47b3f5c32855a Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/66390.png_8.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/66390.png_9.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/66390.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..1cb260ab8377eb789e3347b68ca1fd06ddc3a015 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/66390.png_9.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/flickr_wild_000179.jpg_1.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/flickr_wild_000179.jpg_1.png new file mode 100644 index 0000000000000000000000000000000000000000..5fd845b5c98e209d2129dfc2729b3b2f7f35b7dd Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/flickr_wild_000179.jpg_1.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/flickr_wild_000179.jpg_10.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/flickr_wild_000179.jpg_10.png new file mode 100644 index 0000000000000000000000000000000000000000..56e775d6582face800cb35f58294bfdd38b03b14 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/flickr_wild_000179.jpg_10.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/flickr_wild_000179.jpg_2.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/flickr_wild_000179.jpg_2.png new file mode 100644 index 0000000000000000000000000000000000000000..801864e13fb8513163b3529c06bdb7b536de4a8f Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/flickr_wild_000179.jpg_2.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/flickr_wild_000179.jpg_3.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/flickr_wild_000179.jpg_3.png new file mode 100644 index 0000000000000000000000000000000000000000..7128f1b12870adac540974b43afea9935f864bdd Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/flickr_wild_000179.jpg_3.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/flickr_wild_000179.jpg_4.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/flickr_wild_000179.jpg_4.png new file mode 100644 index 0000000000000000000000000000000000000000..a7bf4d46bccff13b3ca2f21435fc4d0c5a96f51e Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/flickr_wild_000179.jpg_4.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/flickr_wild_000179.jpg_5.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/flickr_wild_000179.jpg_5.png new file mode 100644 index 0000000000000000000000000000000000000000..82d5046a7ad6d615ec226c94cb5f8c5f92ed285e Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/flickr_wild_000179.jpg_5.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/flickr_wild_000179.jpg_6.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/flickr_wild_000179.jpg_6.png new file mode 100644 index 0000000000000000000000000000000000000000..a966378dc7e65eb7f37030a3e0035e6de403ce39 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/flickr_wild_000179.jpg_6.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/flickr_wild_000179.jpg_7.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/flickr_wild_000179.jpg_7.png new file mode 100644 index 0000000000000000000000000000000000000000..670502ee8d0748683f1bf8ffd3d88041cb5907e2 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/flickr_wild_000179.jpg_7.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/flickr_wild_000179.jpg_8.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/flickr_wild_000179.jpg_8.png new file mode 100644 index 0000000000000000000000000000000000000000..0fed2caeca3368838367e8948de67837b2a1b005 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/flickr_wild_000179.jpg_8.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/flickr_wild_000179.jpg_9.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/flickr_wild_000179.jpg_9.png new file mode 100644 index 0000000000000000000000000000000000000000..cb5b4135f5baeb912e83a0399e1d7e485eee9a37 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/flickr_wild_000179.jpg_9.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/pixabay_dog_002352.jpg_1.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/pixabay_dog_002352.jpg_1.png new file mode 100644 index 0000000000000000000000000000000000000000..0b2cc58a2e2c1ce0c9e2e9ae4515496ee5ffaac9 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/pixabay_dog_002352.jpg_1.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/pixabay_dog_002352.jpg_10.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/pixabay_dog_002352.jpg_10.png new file mode 100644 index 0000000000000000000000000000000000000000..6620cc258d30eff6755fbe3d1a68bb015989b586 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/pixabay_dog_002352.jpg_10.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/pixabay_dog_002352.jpg_2.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/pixabay_dog_002352.jpg_2.png new file mode 100644 index 0000000000000000000000000000000000000000..b74483af885670bc709656cc6e46aadf8cee9c22 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/pixabay_dog_002352.jpg_2.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/pixabay_dog_002352.jpg_3.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/pixabay_dog_002352.jpg_3.png new file mode 100644 index 0000000000000000000000000000000000000000..978e9a77c1cfcaa2cd6c94009880995e7f5f6d20 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/pixabay_dog_002352.jpg_3.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/pixabay_dog_002352.jpg_4.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/pixabay_dog_002352.jpg_4.png new file mode 100644 index 0000000000000000000000000000000000000000..a2533c37deae4e0ec41c1ccbbf07a8f6c5eac18b Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/pixabay_dog_002352.jpg_4.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/pixabay_dog_002352.jpg_5.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/pixabay_dog_002352.jpg_5.png new file mode 100644 index 0000000000000000000000000000000000000000..8bd87c885980168270027ab476ab623208c8448e Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/pixabay_dog_002352.jpg_5.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/pixabay_dog_002352.jpg_6.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/pixabay_dog_002352.jpg_6.png new file mode 100644 index 0000000000000000000000000000000000000000..5520c7b6a5c134988afde2a36c073a417ba14a1a Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/pixabay_dog_002352.jpg_6.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/pixabay_dog_002352.jpg_7.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/pixabay_dog_002352.jpg_7.png new file mode 100644 index 0000000000000000000000000000000000000000..53dee72dcbbbd223dc6ee2dcc5d778c4f64ea4ea Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/pixabay_dog_002352.jpg_7.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/pixabay_dog_002352.jpg_8.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/pixabay_dog_002352.jpg_8.png new file mode 100644 index 0000000000000000000000000000000000000000..db5a4e96e1411322eeba99f4d9b30b0726a6aef8 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/pixabay_dog_002352.jpg_8.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/pixabay_dog_002352.jpg_9.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/pixabay_dog_002352.jpg_9.png new file mode 100644 index 0000000000000000000000000000000000000000..6f7e03d81041380825024da69c9279338453a2dd Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/pixabay_dog_002352.jpg_9.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed14088.png_1.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed14088.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..50d900a758ff9b490c6b63da5646ba730276b681 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed14088.png_1.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed14088.png_10.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed14088.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..45a4f7da15d43866e235bc57afc8041d633a0c3a Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed14088.png_10.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed14088.png_2.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed14088.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..e633484d2e7255ae32776fffa220fd2ef0405701 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed14088.png_2.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed14088.png_3.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed14088.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..dcf884a32e662e1e69d953a31dc24fda4cf321f6 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed14088.png_3.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed14088.png_4.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed14088.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..282584594177f3720c9c12c4c4b9503f8c2051e3 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed14088.png_4.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed14088.png_5.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed14088.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..388214b4e3a2acc0b01f2084de607fb4bdc4bc2d Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed14088.png_5.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed14088.png_6.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed14088.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..aefd042b0c1b8be9ff5e125f54553c4f37e1d722 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed14088.png_6.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed14088.png_7.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed14088.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..a55aea5f0a6b0b1cef848982ce68458a55e01d97 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed14088.png_7.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed14088.png_8.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed14088.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..c1f64113068b52e3ca00e21399472d587add9916 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed14088.png_8.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed14088.png_9.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed14088.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..461a91bb70985058ea9e321c99c75366866d2b19 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed14088.png_9.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed16124.png_1.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed16124.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..e75b23a126657ff72649ad8b11e67c147e5d9bc1 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed16124.png_1.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed16124.png_10.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed16124.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..97ebe9490efd2f735464b5a87672ef13edad0f76 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed16124.png_10.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed16124.png_2.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed16124.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..44c6bacb836723f0843d611a1a5fb3f459e204a7 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed16124.png_2.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed16124.png_3.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed16124.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..a11a0362d6dc58399d60f75109182c9566964fe0 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed16124.png_3.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed16124.png_4.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed16124.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..c05d9793b970819fde452b50efe1ff9101acf08f Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed16124.png_4.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed16124.png_5.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed16124.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..218ff5d999b456dc43cce6fb35c8731d76eb663c Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed16124.png_5.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed16124.png_6.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed16124.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..7170988175a0b3d02022f9c2e14518397d08b913 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed16124.png_6.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed16124.png_7.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed16124.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..5d0daf25da4e4e5e84464ebd87d50532cd960547 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed16124.png_7.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed16124.png_8.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed16124.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..7158048764c1e0aefe0770f597bfea9e0e6d6219 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed16124.png_8.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed16124.png_9.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed16124.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..2192af4fe4cf70bcbfc4aa04d66a3e3cf9739f11 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed16124.png_9.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed17282.png_1.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed17282.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..2894e33f07099866f7bd317e8655a5900266f1a0 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed17282.png_1.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed17282.png_10.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed17282.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..fbb99abed46468ce5268c4c1d8d0a87030a8215b Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed17282.png_10.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed17282.png_2.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed17282.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..0009d840e8f8a037fa7aa6be468898139135f4fe Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed17282.png_2.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed17282.png_3.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed17282.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..c9eec123d878311cf39c42490f1b6d918836b8a1 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed17282.png_3.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed17282.png_4.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed17282.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..f30679c31eb3b49e8f23fb4928e1a20c486a578e Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed17282.png_4.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed17282.png_5.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed17282.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..78848b74e7d3f94597b0d3a7ab1a31e1f240eb20 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed17282.png_5.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed17282.png_6.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed17282.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..0dc1042573b68ea75b3f003eba67c4c01ba18299 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed17282.png_6.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed17282.png_7.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed17282.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..2288521024ad83cf43f2aadf1c985816657c6c1e Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed17282.png_7.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed17282.png_8.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed17282.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..c452c309726e5e0ebbb9b4a12f5e5f89db03c5bc Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed17282.png_8.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed17282.png_9.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed17282.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..c97ee8ff28c80d3cddba638fa1f05d24925db7c0 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed17282.png_9.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed21815.png_1.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed21815.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..9b4302a08eca70b6019bf541a46c70aed757b4f7 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed21815.png_1.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed21815.png_10.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed21815.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..40557d1e9cdd45a305cabe9447efee4dcbe99a95 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed21815.png_10.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed21815.png_2.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed21815.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..d0641d02e5c378e086f7f0c5429e63042e3c1d05 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed21815.png_2.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed21815.png_3.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed21815.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..fde50886f580465f2f2f9ad67243342e1760f72f Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed21815.png_3.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed21815.png_4.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed21815.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..f8dc48b30e10821a3194088097a26e3a31447357 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed21815.png_4.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed21815.png_5.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed21815.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..395642edc28f4a8bda2ebd430443461f08a8b0a6 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed21815.png_5.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed21815.png_6.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed21815.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..6c296fe6b14c630d9546b42f5d5eb1727115f66a Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed21815.png_6.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed21815.png_7.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed21815.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..d59814accbba12290c72239a257c3f05413544f0 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed21815.png_7.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed21815.png_8.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed21815.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..63deb59fdb39888ecf77993b5db9083723cc167b Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed21815.png_8.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed21815.png_9.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed21815.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..1ed5cc31137194bb25d80b30ca0bc115d08653b6 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed21815.png_9.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed22624.png_1.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed22624.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..21422269a87f5a444c60347fd247835e4481bc56 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed22624.png_1.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed22624.png_10.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed22624.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..89ef1d2c3bb028575535d6fdca59ecf64983ff28 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed22624.png_10.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed22624.png_2.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed22624.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..5d132ad79addd126783c98087cb7ab9581d772a1 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed22624.png_2.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed22624.png_3.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed22624.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..2b47361060fb8568d52dd19b4b457bb7d4ce7920 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed22624.png_3.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed22624.png_4.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed22624.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..4f6ce7ed85f2e4786cc5a8e6315da848fee6aee9 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed22624.png_4.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed22624.png_5.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed22624.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..de850658b5846a1624aa00b5203fd7fa26674ab8 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed22624.png_5.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed22624.png_6.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed22624.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..3f1977be5a0986da98a390a369e37c6f354124d1 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed22624.png_6.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed22624.png_7.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed22624.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..40009d0a5e574bf2028866ba2a68d65735b60569 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed22624.png_7.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed22624.png_8.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed22624.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..27a18c9e4bcbddea10ea2c2a629987c63bf1ab88 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed22624.png_8.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed22624.png_9.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed22624.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..109b9d04216f76ced322e12c370364d2a9b03cf2 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed22624.png_9.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed23602.png_1.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed23602.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..2999845c88b1b385b1ec7187562e4ba4dfb50b21 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed23602.png_1.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed23602.png_10.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed23602.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..89f8e663f8c6e178a77a4c078e4e3e3ca462e7f2 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed23602.png_10.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed23602.png_2.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed23602.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..f0fb49980afaaa22a1e5687b18536a05b0ad5b90 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed23602.png_2.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed23602.png_3.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed23602.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..694f1b569cbd659f9c48e33082aecd711f2c314d Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed23602.png_3.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed23602.png_4.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed23602.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..db5ce82bc0357c49bae080405b76767a449f8143 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed23602.png_4.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed23602.png_5.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed23602.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..e6748ed07c2e9a9d5a4a7bca84058422235921ed Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed23602.png_5.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed23602.png_6.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed23602.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..09e1178e5dbfad4586e705fb1c6b61a3ef04cc85 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed23602.png_6.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed23602.png_7.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed23602.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..92e416ca0d8a7e9ad83fc2b11c32cf2e0beb6c25 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed23602.png_7.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed23602.png_8.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed23602.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..8dce179f064a88f99fd45d862b505e7b9eaaf141 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed23602.png_8.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed23602.png_9.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed23602.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..c685b41ee82264b63bb51d01d286c5360fedcb96 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed23602.png_9.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed24366.png_1.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed24366.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..dfbf20aa27f178cc1bcbd2e4d977034bfb2229f1 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed24366.png_1.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed24366.png_10.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed24366.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..402ee61e548dcd697027a971ed09f8f43f4f602b Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed24366.png_10.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed24366.png_2.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed24366.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..e48d771649042efb9bfae7ce71c317262ed41cf3 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed24366.png_2.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed24366.png_3.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed24366.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..cf488d09cd2a6f811c3bc78940bacb7e5dfb998f Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed24366.png_3.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed24366.png_4.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed24366.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..d39e26da1356034e333b86289f63812accaddccf Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed24366.png_4.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed24366.png_5.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed24366.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..836ff935f5a5583408d10970561df8a3924a6872 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed24366.png_5.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed24366.png_6.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed24366.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..78f5cdb32d47d043c5b65b393ee0c384dad3695f Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed24366.png_6.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed24366.png_7.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed24366.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..e38cbd04621c4bec781b4655e624f914003b3a37 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed24366.png_7.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed24366.png_8.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed24366.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..b4222ffc1c13fac60ff6235e84e186c3a5be8d77 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed24366.png_8.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed24366.png_9.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed24366.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..cddebb9fb4028a3fadfdee128c49f9c8deaff49e Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed24366.png_9.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed26848.png_1.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed26848.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..fa690e003cea6c683a57b25027cf230abbc0618e Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed26848.png_1.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed26848.png_10.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed26848.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..4d9c7a2ba38e74b30b1fec3b8e6f4b50b6726b76 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed26848.png_10.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed26848.png_2.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed26848.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..ab2b3f85d0d1b51ba3a6386c81186f7fbd72e6e1 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed26848.png_2.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed26848.png_3.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed26848.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..052a0290fce2a449da492ba5d8a7e0bd1157d7bd Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed26848.png_3.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed26848.png_4.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed26848.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..2493e54e2b0f031e7b636e78611e8cddb7131be7 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed26848.png_4.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed26848.png_5.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed26848.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..baf19ddf3173d2372a48a91ef7c1f49fda514421 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed26848.png_5.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed26848.png_6.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed26848.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..f02f3180d6d8226980867b7a917b2c1e36b2f6d9 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed26848.png_6.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed26848.png_7.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed26848.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..2547b8a36e1f75f1885f11259a44d36543dc7289 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed26848.png_7.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed26848.png_8.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed26848.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..c08871cd52082408e16981a99b921a7f17c7aa72 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed26848.png_8.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed26848.png_9.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed26848.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..ac6370474d5490cec3da67b916f6dc1aa82f5d7d Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed26848.png_9.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed44086.png_1.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed44086.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..3e40d42ad6e84b86947c6da03bf62d040ab6bd99 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed44086.png_1.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed44086.png_10.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed44086.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..a7a0aaf2cd3775df7cd8f347453e82495c6953b2 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed44086.png_10.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed44086.png_2.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed44086.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..6db676c7e43dfb9b31ecbaed815ee879665d006e Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed44086.png_2.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed44086.png_3.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed44086.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..f554f077898ca068c92fe0475821bc4c161a4c86 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed44086.png_3.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed44086.png_4.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed44086.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..fbd6940026b50bfd5b970470ba9308ec4acea1d9 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed44086.png_4.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed44086.png_5.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed44086.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..21356b86c4641352330917df1ed799b68333eba4 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed44086.png_5.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed44086.png_6.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed44086.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..12b7646ff4fe70b02c9cbd3464e7573366ee0424 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed44086.png_6.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed44086.png_7.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed44086.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..e6643278397955bbd5b580b6e86eeb763674eea9 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed44086.png_7.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed44086.png_8.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed44086.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..47e279c53b64085069eca5856c146ea2e8f7ff8e Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed44086.png_8.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed44086.png_9.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed44086.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..50621dcb6a6fe6a3e980692a64627f1c32b5dfc8 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/20230605_010029/vis_data/vis_image/seed44086.png_9.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/epoch_1.pth b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/epoch_1.pth new file mode 100644 index 0000000000000000000000000000000000000000..12589c0b919b97583b75de23bd95e2c2f794b636 --- /dev/null +++ b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/epoch_1.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:54c5db2e3455bbf231a661f3acdc5a3297597e2d95f4c2faabbc3c859aa80dd5 +size 277074189 diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/epoch_10.pth b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/epoch_10.pth new file mode 100644 index 0000000000000000000000000000000000000000..9646d67a755f7b6a03032800de996cd090802558 --- /dev/null +++ b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/epoch_10.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:fdbeba30371e28ffd0e9156209cea6c3ab65295f01ae53ba3541b53e1887b19e +size 277406861 diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/epoch_2.pth b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/epoch_2.pth new file mode 100644 index 0000000000000000000000000000000000000000..70adba4ca740b658ab8015727ddd0fee3fa51096 --- /dev/null +++ b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/epoch_2.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:5fd302a62ab5cc9688945c9b4bc6a1e159de68ee4bf3f39b507fd77c5b978a03 +size 277111885 diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/epoch_3.pth b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/epoch_3.pth new file mode 100644 index 0000000000000000000000000000000000000000..b9217332b93421b717466ae3e55e687c20889ece --- /dev/null +++ b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/epoch_3.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c6c4cdd00874d6b298e175cf76bb89d29a7a6cd3180cc1430ec8400c5c4dd025 +size 277148557 diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/epoch_4.pth b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/epoch_4.pth new file mode 100644 index 0000000000000000000000000000000000000000..ac0127b672f96910731052c61cef4a97bd590f50 --- /dev/null +++ b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/epoch_4.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:62da4411477ba37a54445ddd7264c1ae0d893b9d980accf2f298e9be3f79b2d4 +size 277185293 diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/epoch_5.pth b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/epoch_5.pth new file mode 100644 index 0000000000000000000000000000000000000000..4c9ed535709b3f26c1d4495773bfc7200e6b6d45 --- /dev/null +++ b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/epoch_5.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:09ae97eb7db29bd72fead16e9936023d031bc588986559a4dc0d2030389101dc +size 277222157 diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/epoch_6.pth b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/epoch_6.pth new file mode 100644 index 0000000000000000000000000000000000000000..1ab6a790e614e4362f9d87d7d8986b7117799519 --- /dev/null +++ b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/epoch_6.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c3a797308dd611e5010b5cc1cd22ef33bea9bc1d10fbcc32cdbc2199bf2caf35 +size 277258957 diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/epoch_7.pth b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/epoch_7.pth new file mode 100644 index 0000000000000000000000000000000000000000..abdffb3bc858ee2b780c30abdcba448465ba53df --- /dev/null +++ b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/epoch_7.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:b525535219bb9d523be04f6dccbbccb60c9b4e1df28e1be12c0b3816c0d222fc +size 277295757 diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/epoch_8.pth b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/epoch_8.pth new file mode 100644 index 0000000000000000000000000000000000000000..82307da5ab15a281e77879505ddac755e493cd3d --- /dev/null +++ b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/epoch_8.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:69326be03ee37ba8ec846a95ab6112a8149ff70a199a44f1c76cbd70082bdc99 +size 277332749 diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/epoch_9.pth b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/epoch_9.pth new file mode 100644 index 0000000000000000000000000000000000000000..1b8d955c78e552b7b6cc24f457dc2babd4ca5bd3 --- /dev/null +++ b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/epoch_9.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:0afa08cf07d99f45391950dd0c4f41e2592e2cec44abace43dbd36b1d3d698a9 +size 277369805 diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/last_checkpoint b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/last_checkpoint new file mode 100644 index 0000000000000000000000000000000000000000..2cf934e40eb2361702350c2eaa6a65ae1bc514c5 --- /dev/null +++ b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/last_checkpoint @@ -0,0 +1 @@ +/mnt/petrelfs/luzeyu/workspace/fakebench/mmpretrain/workdir/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/epoch_10.pth \ No newline at end of file diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1.py b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1.py new file mode 100644 index 0000000000000000000000000000000000000000..f37931e24e882a176c705bb4cec9cc535f363f49 --- /dev/null +++ b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1.py @@ -0,0 +1,230 @@ +optim_wrapper = dict( + optimizer=dict( + type='SGD', + lr=0.0001, + momentum=0.9, + weight_decay=0.0001, + _scope_='mmpretrain'), + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=False, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=512) +model = dict( + type='ImageClassifier', + backbone=dict( + frozen_stages=2, + type='ResNet', + depth=50, + num_stages=4, + out_indices=(3, ), + style='pytorch', + init_cfg=dict( + type='Pretrained', + checkpoint= + 'https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth', + prefix='backbone')), + neck=dict(type='GlobalAveragePooling'), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=2048, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + topk=1)) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3fake8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3real8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3fake8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3real8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = dict(type='Accuracy', topk=1) +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3fake8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3real8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = dict(type='Accuracy', topk=1) +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_1e-1' diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/20230604_232954.log b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/20230604_232954.log new file mode 100644 index 0000000000000000000000000000000000000000..79401416a2a96fa8cd740694b72bdfd99a3fd210 --- /dev/null +++ b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/20230604_232954.log @@ -0,0 +1,939 @@ +2023/06/04 23:29:58 - mmengine - INFO - +------------------------------------------------------------ +System environment: + sys.platform: linux + Python: 3.10.9 (main, Mar 8 2023, 10:47:38) [GCC 11.2.0] + CUDA available: True + numpy_random_seed: 361995161 + GPU 0,1: NVIDIA A100-SXM4-80GB + CUDA_HOME: /mnt/petrelfs/share/cuda-11.6 + NVCC: Cuda compilation tools, release 11.6, V11.6.124 + GCC: gcc (GCC) 7.5.0 + PyTorch: 1.13.1 + PyTorch compiling details: PyTorch built with: + - GCC 9.3 + - C++ Version: 201402 + - Intel(R) oneAPI Math Kernel Library Version 2021.4-Product Build 20210904 for Intel(R) 64 architecture applications + - Intel(R) MKL-DNN v2.6.0 (Git Hash 52b5f107dd9cf10910aaa19cb47f3abf9b349815) + - OpenMP 201511 (a.k.a. OpenMP 4.5) + - LAPACK is enabled (usually provided by MKL) + - NNPACK is enabled + - CPU capability usage: AVX2 + - CUDA Runtime 11.6 + - NVCC architecture flags: -gencode;arch=compute_37,code=sm_37;-gencode;arch=compute_50,code=sm_50;-gencode;arch=compute_60,code=sm_60;-gencode;arch=compute_61,code=sm_61;-gencode;arch=compute_70,code=sm_70;-gencode;arch=compute_75,code=sm_75;-gencode;arch=compute_80,code=sm_80;-gencode;arch=compute_86,code=sm_86;-gencode;arch=compute_37,code=compute_37 + - CuDNN 8.3.2 (built against CUDA 11.5) + - Magma 2.6.1 + - Build settings: BLAS_INFO=mkl, BUILD_TYPE=Release, CUDA_VERSION=11.6, CUDNN_VERSION=8.3.2, CXX_COMPILER=/opt/rh/devtoolset-9/root/usr/bin/c++, CXX_FLAGS= -fabi-version=11 -Wno-deprecated -fvisibility-inlines-hidden -DUSE_PTHREADPOOL -fopenmp -DNDEBUG -DUSE_KINETO -DUSE_FBGEMM -DUSE_QNNPACK -DUSE_PYTORCH_QNNPACK -DUSE_XNNPACK -DSYMBOLICATE_MOBILE_DEBUG_HANDLE -DEDGE_PROFILER_USE_KINETO -O2 -fPIC -Wno-narrowing -Wall -Wextra -Werror=return-type -Werror=non-virtual-dtor -Wno-missing-field-initializers -Wno-type-limits -Wno-array-bounds -Wno-unknown-pragmas -Wunused-local-typedefs -Wno-unused-parameter -Wno-unused-function -Wno-unused-result -Wno-strict-overflow -Wno-strict-aliasing -Wno-error=deprecated-declarations -Wno-stringop-overflow -Wno-psabi -Wno-error=pedantic -Wno-error=redundant-decls -Wno-error=old-style-cast -fdiagnostics-color=always -faligned-new -Wno-unused-but-set-variable -Wno-maybe-uninitialized -fno-math-errno -fno-trapping-math -Werror=format -Werror=cast-function-type -Wno-stringop-overflow, LAPACK_INFO=mkl, PERF_WITH_AVX=1, PERF_WITH_AVX2=1, PERF_WITH_AVX512=1, TORCH_VERSION=1.13.1, USE_CUDA=ON, USE_CUDNN=ON, USE_EXCEPTION_PTR=1, USE_GFLAGS=OFF, USE_GLOG=OFF, USE_MKL=ON, USE_MKLDNN=ON, USE_MPI=OFF, USE_NCCL=ON, USE_NNPACK=ON, USE_OPENMP=ON, USE_ROCM=OFF, + + TorchVision: 0.14.1 + OpenCV: 4.7.0 + MMEngine: 0.7.3 + +Runtime environment: + cudnn_benchmark: True + mp_cfg: {'mp_start_method': 'fork', 'opencv_num_threads': 0} + dist_cfg: {'backend': 'nccl'} + seed: None + deterministic: False + Distributed launcher: slurm + Distributed training: True + GPU number: 2 +------------------------------------------------------------ + +2023/06/04 23:30:02 - mmengine - INFO - Config: +optim_wrapper = dict( + optimizer=dict( + type='SGD', + lr=0.0001, + momentum=0.9, + weight_decay=0.0001, + _scope_='mmpretrain'), + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=False, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=512) +model = dict( + type='ImageClassifier', + backbone=dict( + frozen_stages=2, + type='ResNet', + depth=50, + num_stages=4, + out_indices=(3, ), + style='pytorch', + init_cfg=dict( + type='Pretrained', + checkpoint= + 'https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth', + prefix='backbone')), + neck=dict(type='GlobalAveragePooling'), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=2048, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + topk=1)) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3fake8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3real8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3fake8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3real8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = dict(type='Accuracy', topk=1) +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3fake8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3real8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = dict(type='Accuracy', topk=1) +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1' + +2023/06/04 23:30:14 - mmengine - INFO - Hooks will be executed in the following order: +before_run: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_load_checkpoint: +(ABOVE_NORMAL) EMAHook + -------------------- +before_train: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_train_epoch: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(NORMAL ) DistSamplerSeedHook + -------------------- +before_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook + -------------------- +after_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +after_train_epoch: +(NORMAL ) IterTimerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_val_epoch: +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook + -------------------- +before_val_iter: +(NORMAL ) IterTimerHook + -------------------- +after_val_iter: +(NORMAL ) IterTimerHook +(NORMAL ) VisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_val_epoch: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_save_checkpoint: +(ABOVE_NORMAL) EMAHook + -------------------- +after_train: +(VERY_LOW ) CheckpointHook + -------------------- +before_test_epoch: +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook + -------------------- +before_test_iter: +(NORMAL ) IterTimerHook + -------------------- +after_test_iter: +(NORMAL ) IterTimerHook +(NORMAL ) VisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_test_epoch: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_run: +(BELOW_NORMAL) LoggerHook + -------------------- +2023/06/04 23:30:25 - mmengine - INFO - load backbone in model from: https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth +Name of parameter - Initialization information + +backbone.conv1.weight - torch.Size([64, 3, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.bn1.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.bn1.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.conv1.weight - torch.Size([64, 64, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.bn1.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.bn1.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.conv2.weight - torch.Size([64, 64, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.bn2.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.bn2.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.conv3.weight - torch.Size([256, 64, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.bn3.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.bn3.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.downsample.0.weight - torch.Size([256, 64, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.downsample.1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.downsample.1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.conv1.weight - torch.Size([64, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.bn1.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.bn1.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.conv2.weight - torch.Size([64, 64, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.bn2.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.bn2.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.conv3.weight - torch.Size([256, 64, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.bn3.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.bn3.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.conv1.weight - torch.Size([64, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.bn1.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.bn1.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.conv2.weight - torch.Size([64, 64, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.bn2.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.bn2.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.conv3.weight - torch.Size([256, 64, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.bn3.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.bn3.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.conv1.weight - torch.Size([128, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.bn1.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.bn1.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.conv2.weight - torch.Size([128, 128, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.bn2.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.bn2.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.conv3.weight - torch.Size([512, 128, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.bn3.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.bn3.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.downsample.0.weight - torch.Size([512, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.downsample.1.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.downsample.1.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.conv1.weight - torch.Size([128, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.bn1.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.bn1.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.conv2.weight - torch.Size([128, 128, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.bn2.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.bn2.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.conv3.weight - torch.Size([512, 128, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.bn3.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.bn3.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.conv1.weight - torch.Size([128, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.bn1.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.bn1.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.conv2.weight - torch.Size([128, 128, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.bn2.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.bn2.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.conv3.weight - torch.Size([512, 128, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.bn3.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.bn3.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.conv1.weight - torch.Size([128, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.bn1.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.bn1.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.conv2.weight - torch.Size([128, 128, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.bn2.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.bn2.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.conv3.weight - torch.Size([512, 128, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.bn3.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.bn3.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.conv1.weight - torch.Size([256, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.bn1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.bn1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.conv2.weight - torch.Size([256, 256, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.bn2.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.bn2.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.conv3.weight - torch.Size([1024, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.bn3.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.bn3.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.downsample.0.weight - torch.Size([1024, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.downsample.1.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.downsample.1.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.conv1.weight - torch.Size([256, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.bn1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.bn1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.conv2.weight - torch.Size([256, 256, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.bn2.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.bn2.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.conv3.weight - torch.Size([1024, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.bn3.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.bn3.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.conv1.weight - torch.Size([256, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.bn1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.bn1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.conv2.weight - torch.Size([256, 256, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.bn2.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.bn2.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.conv3.weight - torch.Size([1024, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.bn3.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.bn3.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.conv1.weight - torch.Size([256, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.bn1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.bn1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.conv2.weight - torch.Size([256, 256, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.bn2.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.bn2.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.conv3.weight - torch.Size([1024, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.bn3.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.bn3.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.conv1.weight - torch.Size([256, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.bn1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.bn1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.conv2.weight - torch.Size([256, 256, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.bn2.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.bn2.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.conv3.weight - torch.Size([1024, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.bn3.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.bn3.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.conv1.weight - torch.Size([256, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.bn1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.bn1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.conv2.weight - torch.Size([256, 256, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.bn2.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.bn2.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.conv3.weight - torch.Size([1024, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.bn3.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.bn3.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.conv1.weight - torch.Size([512, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.bn1.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.bn1.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.conv2.weight - torch.Size([512, 512, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.bn2.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.bn2.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.conv3.weight - torch.Size([2048, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.bn3.weight - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.bn3.bias - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.downsample.0.weight - torch.Size([2048, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.downsample.1.weight - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.downsample.1.bias - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.conv1.weight - torch.Size([512, 2048, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.bn1.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.bn1.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.conv2.weight - torch.Size([512, 512, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.bn2.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.bn2.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.conv3.weight - torch.Size([2048, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.bn3.weight - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.bn3.bias - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.conv1.weight - torch.Size([512, 2048, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.bn1.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.bn1.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.conv2.weight - torch.Size([512, 512, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.bn2.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.bn2.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.conv3.weight - torch.Size([2048, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.bn3.weight - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.bn3.bias - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +head.fc.weight - torch.Size([2, 2048]): +NormalInit: mean=0, std=0.01, bias=0 + +head.fc.bias - torch.Size([2]): +NormalInit: mean=0, std=0.01, bias=0 +2023/06/04 23:30:25 - mmengine - WARNING - "FileClient" will be deprecated in future. Please use io functions in https://mmengine.readthedocs.io/en/latest/api/fileio.html#file-io +2023/06/04 23:30:25 - mmengine - WARNING - "HardDiskBackend" is the alias of "LocalBackend" and the former will be deprecated in future. +2023/06/04 23:30:25 - mmengine - INFO - Checkpoints will be saved to /mnt/petrelfs/luzeyu/workspace/fakebench/mmpretrain/workdir/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1. +2023/06/04 23:32:30 - mmengine - INFO - Epoch(train) [1][100/342] lr: 9.9814e-05 eta: 1:09:16 time: 1.2160 data_time: 0.1523 memory: 9436 loss: 0.6878 +2023/06/04 23:34:34 - mmengine - INFO - Epoch(train) [1][200/342] lr: 9.9250e-05 eta: 1:06:50 time: 1.2280 data_time: 0.0268 memory: 6319 loss: 0.6727 +2023/06/04 23:36:36 - mmengine - INFO - Epoch(train) [1][300/342] lr: 9.8313e-05 eta: 1:04:23 time: 1.1674 data_time: 0.2385 memory: 6319 loss: 0.6611 +2023/06/04 23:37:25 - mmengine - INFO - Exp name: resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1_20230604_232954 +2023/06/04 23:37:25 - mmengine - INFO - Saving checkpoint at 1 epochs +2023/06/04 23:39:25 - mmengine - INFO - Epoch(val) [1][100/342] eta: 0:04:37 time: 0.3940 data_time: 0.3053 memory: 6319 +2023/06/04 23:41:10 - mmengine - INFO - Epoch(val) [1][200/342] eta: 0:02:36 time: 0.3081 data_time: 0.2198 memory: 3133 +2023/06/04 23:43:24 - mmengine - INFO - Epoch(val) [1][300/342] eta: 0:00:49 time: 1.2829 data_time: 1.1923 memory: 3133 +2023/06/04 23:45:59 - mmengine - INFO - Epoch(val) [1][342/342] accuracy/top1: 49.6640 data_time: 1.2918 time: 1.3832 +2023/06/04 23:48:05 - mmengine - INFO - Epoch(train) [2][100/342] lr: 9.6358e-05 eta: 1:01:18 time: 1.1442 data_time: 0.2228 memory: 6319 loss: 0.6464 +2023/06/04 23:50:07 - mmengine - INFO - Epoch(train) [2][200/342] lr: 9.4557e-05 eta: 0:59:08 time: 1.1230 data_time: 0.0009 memory: 6319 loss: 0.6405 +2023/06/04 23:52:10 - mmengine - INFO - Epoch(train) [2][300/342] lr: 9.2422e-05 eta: 0:57:03 time: 1.3448 data_time: 0.0011 memory: 6319 loss: 0.6363 +2023/06/04 23:52:58 - mmengine - INFO - Exp name: resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1_20230604_232954 +2023/06/04 23:52:58 - mmengine - INFO - Saving checkpoint at 2 epochs +2023/06/04 23:54:59 - mmengine - INFO - Epoch(val) [2][100/342] eta: 0:04:40 time: 0.4653 data_time: 0.3769 memory: 6319 +2023/06/04 23:56:46 - mmengine - INFO - Epoch(val) [2][200/342] eta: 0:02:38 time: 0.2734 data_time: 0.1844 memory: 3133 +2023/06/04 23:59:00 - mmengine - INFO - Epoch(val) [2][300/342] eta: 0:00:49 time: 1.2595 data_time: 1.1696 memory: 3133 +2023/06/05 00:01:34 - mmengine - INFO - Epoch(val) [2][342/342] accuracy/top1: 50.5454 data_time: 1.2973 time: 1.3883 +2023/06/05 00:03:38 - mmengine - INFO - Epoch(train) [3][100/342] lr: 8.8853e-05 eta: 0:53:55 time: 1.1317 data_time: 0.0235 memory: 6319 loss: 0.6221 +2023/06/05 00:05:39 - mmengine - INFO - Epoch(train) [3][200/342] lr: 8.5991e-05 eta: 0:51:50 time: 1.2449 data_time: 0.0010 memory: 6319 loss: 0.6250 +2023/06/05 00:07:38 - mmengine - INFO - Epoch(train) [3][300/342] lr: 8.2867e-05 eta: 0:49:38 time: 1.2113 data_time: 0.0020 memory: 6319 loss: 0.6090 +2023/06/05 00:08:01 - mmengine - INFO - Exp name: resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1_20230604_232954 +2023/06/05 00:08:27 - mmengine - INFO - Exp name: resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1_20230604_232954 +2023/06/05 00:08:27 - mmengine - INFO - Saving checkpoint at 3 epochs +2023/06/05 00:10:27 - mmengine - INFO - Epoch(val) [3][100/342] eta: 0:04:37 time: 0.3948 data_time: 0.3052 memory: 6319 +2023/06/05 00:12:14 - mmengine - INFO - Epoch(val) [3][200/342] eta: 0:02:37 time: 0.2803 data_time: 0.1900 memory: 3133 +2023/06/05 00:14:26 - mmengine - INFO - Epoch(val) [3][300/342] eta: 0:00:49 time: 1.3311 data_time: 1.2409 memory: 3133 +2023/06/05 00:17:02 - mmengine - INFO - Epoch(val) [3][342/342] accuracy/top1: 51.4389 data_time: 1.2900 time: 1.3825 +2023/06/05 00:19:07 - mmengine - INFO - Epoch(train) [4][100/342] lr: 7.8035e-05 eta: 0:46:45 time: 1.2685 data_time: 0.3636 memory: 6319 loss: 0.6026 +2023/06/05 00:21:09 - mmengine - INFO - Epoch(train) [4][200/342] lr: 7.4392e-05 eta: 0:44:42 time: 1.1637 data_time: 0.0023 memory: 6319 loss: 0.5981 +2023/06/05 00:23:09 - mmengine - INFO - Epoch(train) [4][300/342] lr: 7.0585e-05 eta: 0:42:37 time: 1.2687 data_time: 0.0012 memory: 6319 loss: 0.5935 +2023/06/05 00:23:58 - mmengine - INFO - Exp name: resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1_20230604_232954 +2023/06/05 00:23:58 - mmengine - INFO - Saving checkpoint at 4 epochs +2023/06/05 00:26:00 - mmengine - INFO - Epoch(val) [4][100/342] eta: 0:04:41 time: 0.4612 data_time: 0.3726 memory: 6319 +2023/06/05 00:27:47 - mmengine - INFO - Epoch(val) [4][200/342] eta: 0:02:38 time: 0.2494 data_time: 0.1618 memory: 3133 +2023/06/05 00:29:58 - mmengine - INFO - Epoch(val) [4][300/342] eta: 0:00:49 time: 1.2658 data_time: 1.1757 memory: 3133 +2023/06/05 00:32:35 - mmengine - INFO - Epoch(val) [4][342/342] accuracy/top1: 52.4634 data_time: 1.2972 time: 1.3892 +2023/06/05 00:34:38 - mmengine - INFO - Epoch(train) [5][100/342] lr: 6.4962e-05 eta: 0:39:42 time: 1.1538 data_time: 0.0712 memory: 6319 loss: 0.5890 +2023/06/05 00:36:39 - mmengine - INFO - Epoch(train) [5][200/342] lr: 6.0894e-05 eta: 0:37:38 time: 1.1599 data_time: 0.0010 memory: 6319 loss: 0.5847 +2023/06/05 00:38:41 - mmengine - INFO - Epoch(train) [5][300/342] lr: 5.6777e-05 eta: 0:35:36 time: 1.1869 data_time: 0.0011 memory: 6319 loss: 0.5784 +2023/06/05 00:39:29 - mmengine - INFO - Exp name: resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1_20230604_232954 +2023/06/05 00:39:29 - mmengine - INFO - Saving checkpoint at 5 epochs +2023/06/05 00:41:33 - mmengine - INFO - Epoch(val) [5][100/342] eta: 0:04:45 time: 0.4191 data_time: 0.3284 memory: 6319 +2023/06/05 00:43:17 - mmengine - INFO - Epoch(val) [5][200/342] eta: 0:02:37 time: 0.2824 data_time: 0.1935 memory: 3133 +2023/06/05 00:45:29 - mmengine - INFO - Epoch(val) [5][300/342] eta: 0:00:49 time: 1.2892 data_time: 1.1990 memory: 3133 +2023/06/05 00:48:09 - mmengine - INFO - Epoch(val) [5][342/342] accuracy/top1: 53.5022 data_time: 1.3029 time: 1.3943 +2023/06/05 00:50:16 - mmengine - INFO - Epoch(train) [6][100/342] lr: 5.0913e-05 eta: 0:32:44 time: 1.2461 data_time: 0.0009 memory: 6319 loss: 0.5700 +2023/06/05 00:52:16 - mmengine - INFO - Epoch(train) [6][200/342] lr: 4.6820e-05 eta: 0:30:41 time: 1.1725 data_time: 0.0009 memory: 6319 loss: 0.5646 +2023/06/05 00:54:08 - mmengine - INFO - Exp name: resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1_20230604_232954 +2023/06/05 00:54:19 - mmengine - INFO - Epoch(train) [6][300/342] lr: 4.2795e-05 eta: 0:28:40 time: 1.1770 data_time: 0.0011 memory: 6319 loss: 0.5663 +2023/06/05 00:55:17 - mmengine - INFO - Exp name: resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1_20230604_232954 +2023/06/05 00:55:17 - mmengine - INFO - Saving checkpoint at 6 epochs +2023/06/05 00:57:19 - mmengine - INFO - Epoch(val) [6][100/342] eta: 0:04:41 time: 0.3983 data_time: 0.3094 memory: 6319 +2023/06/05 00:59:07 - mmengine - INFO - Epoch(val) [6][200/342] eta: 0:02:39 time: 0.3016 data_time: 0.2112 memory: 3133 +2023/06/05 01:01:17 - mmengine - INFO - Epoch(val) [6][300/342] eta: 0:00:49 time: 1.2450 data_time: 1.1546 memory: 3133 +2023/06/05 01:03:54 - mmengine - INFO - Epoch(val) [6][342/342] accuracy/top1: 54.3412 data_time: 1.2997 time: 1.3911 +2023/06/05 01:06:00 - mmengine - INFO - Epoch(train) [7][100/342] lr: 3.7265e-05 eta: 0:25:53 time: 1.2421 data_time: 0.2574 memory: 6319 loss: 0.5655 +2023/06/05 01:08:01 - mmengine - INFO - Epoch(train) [7][200/342] lr: 3.3546e-05 eta: 0:23:49 time: 1.1933 data_time: 0.0009 memory: 6319 loss: 0.5553 +2023/06/05 01:10:03 - mmengine - INFO - Epoch(train) [7][300/342] lr: 3.0008e-05 eta: 0:21:47 time: 1.1059 data_time: 0.0010 memory: 6319 loss: 0.5656 +2023/06/05 01:10:59 - mmengine - INFO - Exp name: resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1_20230604_232954 +2023/06/05 01:10:59 - mmengine - INFO - Saving checkpoint at 7 epochs +2023/06/05 01:13:02 - mmengine - INFO - Epoch(val) [7][100/342] eta: 0:04:41 time: 0.4008 data_time: 0.3130 memory: 6319 +2023/06/05 01:14:49 - mmengine - INFO - Epoch(val) [7][200/342] eta: 0:02:38 time: 0.2896 data_time: 0.1996 memory: 3133 +2023/06/05 01:17:01 - mmengine - INFO - Epoch(val) [7][300/342] eta: 0:00:49 time: 1.2823 data_time: 1.1930 memory: 3133 +2023/06/05 01:19:40 - mmengine - INFO - Epoch(val) [7][342/342] accuracy/top1: 54.9319 data_time: 1.3065 time: 1.3970 +2023/06/05 01:21:45 - mmengine - INFO - Epoch(train) [8][100/342] lr: 2.5353e-05 eta: 0:18:56 time: 1.2050 data_time: 0.0008 memory: 6319 loss: 0.5701 +2023/06/05 01:23:46 - mmengine - INFO - Epoch(train) [8][200/342] lr: 2.2372e-05 eta: 0:16:52 time: 1.1828 data_time: 0.0010 memory: 6319 loss: 0.5533 +2023/06/05 01:25:49 - mmengine - INFO - Epoch(train) [8][300/342] lr: 1.9667e-05 eta: 0:14:50 time: 1.2497 data_time: 0.0011 memory: 6319 loss: 0.5500 +2023/06/05 01:26:47 - mmengine - INFO - Exp name: resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1_20230604_232954 +2023/06/05 01:26:47 - mmengine - INFO - Saving checkpoint at 8 epochs +2023/06/05 01:28:50 - mmengine - INFO - Epoch(val) [8][100/342] eta: 0:04:43 time: 0.3776 data_time: 0.2890 memory: 6319 +2023/06/05 01:30:37 - mmengine - INFO - Epoch(val) [8][200/342] eta: 0:02:39 time: 0.2778 data_time: 0.1889 memory: 3133 +2023/06/05 01:32:51 - mmengine - INFO - Epoch(val) [8][300/342] eta: 0:00:50 time: 1.2904 data_time: 1.2004 memory: 3133 +2023/06/05 01:35:29 - mmengine - INFO - Epoch(val) [8][342/342] accuracy/top1: 55.4957 data_time: 1.3102 time: 1.4009 +2023/06/05 01:37:33 - mmengine - INFO - Epoch(train) [9][100/342] lr: 1.6343e-05 eta: 0:11:57 time: 1.2294 data_time: 0.5104 memory: 6319 loss: 0.5514 +2023/06/05 01:39:34 - mmengine - INFO - Epoch(train) [9][200/342] lr: 1.4393e-05 eta: 0:09:54 time: 1.2118 data_time: 0.0010 memory: 6319 loss: 0.5432 +2023/06/05 01:40:51 - mmengine - INFO - Exp name: resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1_20230604_232954 +2023/06/05 01:41:34 - mmengine - INFO - Epoch(train) [9][300/342] lr: 1.2785e-05 eta: 0:07:51 time: 1.2039 data_time: 0.0009 memory: 6319 loss: 0.5493 +2023/06/05 01:42:23 - mmengine - INFO - Exp name: resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1_20230604_232954 +2023/06/05 01:42:23 - mmengine - INFO - Saving checkpoint at 9 epochs +2023/06/05 01:44:25 - mmengine - INFO - Epoch(val) [9][100/342] eta: 0:04:38 time: 0.4043 data_time: 0.3151 memory: 6319 +2023/06/05 01:46:10 - mmengine - INFO - Epoch(val) [9][200/342] eta: 0:02:36 time: 0.2646 data_time: 0.1768 memory: 3133 +2023/06/05 01:48:25 - mmengine - INFO - Epoch(val) [9][300/342] eta: 0:00:49 time: 1.3143 data_time: 1.2235 memory: 3133 +2023/06/05 01:51:01 - mmengine - INFO - Epoch(val) [9][342/342] accuracy/top1: 55.8156 data_time: 1.2947 time: 1.3871 +2023/06/05 01:53:04 - mmengine - INFO - Epoch(train) [10][100/342] lr: 1.1116e-05 eta: 0:04:56 time: 1.2743 data_time: 0.3279 memory: 6319 loss: 0.5498 +2023/06/05 01:55:06 - mmengine - INFO - Epoch(train) [10][200/342] lr: 1.0388e-05 eta: 0:02:54 time: 1.4397 data_time: 0.0010 memory: 6319 loss: 0.5461 +2023/06/05 01:57:10 - mmengine - INFO - Epoch(train) [10][300/342] lr: 1.0035e-05 eta: 0:00:51 time: 1.2579 data_time: 0.0021 memory: 6319 loss: 0.5386 +2023/06/05 01:57:55 - mmengine - INFO - Exp name: resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1_20230604_232954 +2023/06/05 01:57:55 - mmengine - INFO - Saving checkpoint at 10 epochs +2023/06/05 01:59:54 - mmengine - INFO - Epoch(val) [10][100/342] eta: 0:04:33 time: 0.4075 data_time: 0.3189 memory: 6319 +2023/06/05 02:01:38 - mmengine - INFO - Epoch(val) [10][200/342] eta: 0:02:34 time: 0.2569 data_time: 0.1676 memory: 3133 +2023/06/05 02:03:54 - mmengine - INFO - Epoch(val) [10][300/342] eta: 0:00:49 time: 1.2783 data_time: 1.1882 memory: 3133 +2023/06/05 02:06:29 - mmengine - INFO - Epoch(val) [10][342/342] accuracy/top1: 56.2117 data_time: 1.2867 time: 1.3777 diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/20230604_232954.json b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/20230604_232954.json new file mode 100644 index 0000000000000000000000000000000000000000..73a1a0fa44b07ec9511fc8ac47de62df229a5d0e --- /dev/null +++ b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/20230604_232954.json @@ -0,0 +1,40 @@ +{"lr": 9.981404789023513e-05, "data_time": 0.1522911310195923, "loss": 0.6877720952033997, "time": 1.2160188198089599, "epoch": 1, "memory": 9436, "step": 100} +{"lr": 9.925023268502683e-05, "data_time": 0.02678549289703369, "loss": 0.6727290153503418, "time": 1.2280450344085694, "epoch": 1, "memory": 6319, "step": 200} +{"lr": 9.831328961451971e-05, "data_time": 0.23851115703582765, "loss": 0.6611114025115967, "time": 1.1674016475677491, "epoch": 1, "memory": 6319, "step": 300} +{"accuracy/top1": 49.66403579711914, "data_time": 1.2918182067703783, "time": 1.3832352468144824, "step": 1} +{"lr": 9.635783271319408e-05, "data_time": 0.22280569076538087, "loss": 0.6463721930980683, "time": 1.1441802978515625, "epoch": 2, "memory": 6319, "step": 442} +{"lr": 9.455662799327337e-05, "data_time": 0.0008972644805908203, "loss": 0.6404570877552033, "time": 1.122991156578064, "epoch": 2, "memory": 6319, "step": 542} +{"lr": 9.242187293325357e-05, "data_time": 0.0011354684829711914, "loss": 0.6362692892551423, "time": 1.3448488473892213, "epoch": 2, "memory": 6319, "step": 642} +{"accuracy/top1": 50.54544448852539, "data_time": 1.2973375619326666, "time": 1.3883099667184902, "step": 2} +{"lr": 8.885322471322101e-05, "data_time": 0.023528909683227538, "loss": 0.6221254885196685, "time": 1.1317203283309936, "epoch": 3, "memory": 6319, "step": 784} +{"lr": 8.59909449463049e-05, "data_time": 0.0010162115097045899, "loss": 0.6249741315841675, "time": 1.2448975086212157, "epoch": 3, "memory": 6319, "step": 884} +{"lr": 8.28673425957603e-05, "data_time": 0.0019779443740844727, "loss": 0.6089748561382293, "time": 1.211303210258484, "epoch": 3, "memory": 6319, "step": 984} +{"accuracy/top1": 51.43887710571289, "data_time": 1.2900429536580345, "time": 1.3825226525176024, "step": 3} +{"lr": 7.803482720903195e-05, "data_time": 0.3636435270309448, "loss": 0.6026146233081817, "time": 1.2685081720352174, "epoch": 4, "memory": 6319, "step": 1126} +{"lr": 7.439165228138202e-05, "data_time": 0.0023336410522460938, "loss": 0.5980541050434113, "time": 1.163676118850708, "epoch": 4, "memory": 6319, "step": 1226} +{"lr": 7.058496260180133e-05, "data_time": 0.001241612434387207, "loss": 0.5935293853282928, "time": 1.2686659097671509, "epoch": 4, "memory": 6319, "step": 1326} +{"accuracy/top1": 52.46337127685547, "data_time": 1.2971795717411765, "time": 1.3891826628248476, "step": 4} +{"lr": 6.496162032454504e-05, "data_time": 0.07118077278137207, "loss": 0.5890477299690247, "time": 1.1538291931152345, "epoch": 5, "memory": 6319, "step": 1468} +{"lr": 6.089416958157208e-05, "data_time": 0.0009602546691894531, "loss": 0.584687077999115, "time": 1.1599042654037475, "epoch": 5, "memory": 6319, "step": 1568} +{"lr": 5.677701788155902e-05, "data_time": 0.0011432647705078125, "loss": 0.5784062802791595, "time": 1.186885905265808, "epoch": 5, "memory": 6319, "step": 1668} +{"accuracy/top1": 53.50217819213867, "data_time": 1.302898420189282, "time": 1.3943057240967154, "step": 5} +{"lr": 5.091330063600123e-05, "data_time": 0.0008756875991821289, "loss": 0.5700162708759308, "time": 1.2461036443710327, "epoch": 6, "memory": 6319, "step": 1810} +{"lr": 4.681972449602271e-05, "data_time": 0.0008860111236572265, "loss": 0.5646239936351776, "time": 1.1724539041519164, "epoch": 6, "memory": 6319, "step": 1910} +{"lr": 4.27951262698573e-05, "data_time": 0.001126861572265625, "loss": 0.5662783503532409, "time": 1.1770079135894775, "epoch": 6, "memory": 6319, "step": 2010} +{"accuracy/top1": 54.34123229980469, "data_time": 1.2996568373966495, "time": 1.3910835685952412, "step": 6} +{"lr": 3.726501555491419e-05, "data_time": 0.2574317455291748, "loss": 0.5655214548110962, "time": 1.2420902729034424, "epoch": 7, "memory": 6319, "step": 2152} +{"lr": 3.354602177213381e-05, "data_time": 0.000925898551940918, "loss": 0.5553143858909607, "time": 1.1933157920837403, "epoch": 7, "memory": 6319, "step": 2252} +{"lr": 3.0007932735217966e-05, "data_time": 0.0010452032089233398, "loss": 0.5655701816082, "time": 1.105875825881958, "epoch": 7, "memory": 6319, "step": 2352} +{"accuracy/top1": 54.93189239501953, "data_time": 1.3064878821025436, "time": 1.3970393342109881, "step": 7} +{"lr": 2.5352754318215113e-05, "data_time": 0.0008061408996582031, "loss": 0.5700551688671112, "time": 1.2050061702728272, "epoch": 8, "memory": 6319, "step": 2494} +{"lr": 2.2372383915844294e-05, "data_time": 0.000982189178466797, "loss": 0.5533104658126831, "time": 1.182770037651062, "epoch": 8, "memory": 6319, "step": 2594} +{"lr": 1.9667136874426842e-05, "data_time": 0.001096343994140625, "loss": 0.5499626159667969, "time": 1.2496567964553833, "epoch": 8, "memory": 6319, "step": 2694} +{"accuracy/top1": 55.49565124511719, "data_time": 1.3102325628519753, "time": 1.400934366026008, "step": 8} +{"lr": 1.6342572053355464e-05, "data_time": 0.5104479551315307, "loss": 0.5513653874397277, "time": 1.2293860912322998, "epoch": 9, "memory": 6319, "step": 2836} +{"lr": 1.4392564451840171e-05, "data_time": 0.0009862661361694336, "loss": 0.5432322919368744, "time": 1.2118443250656128, "epoch": 9, "memory": 6319, "step": 2936} +{"lr": 1.2784967834899793e-05, "data_time": 0.0008760213851928711, "loss": 0.5493425965309143, "time": 1.203943133354187, "epoch": 9, "memory": 6319, "step": 3036} +{"accuracy/top1": 55.81559371948242, "data_time": 1.2946712817812105, "time": 1.3870672303803113, "step": 9} +{"lr": 1.1116448178051759e-05, "data_time": 0.3279048204421997, "loss": 0.5497923195362091, "time": 1.2743196964263916, "epoch": 10, "memory": 6319, "step": 3178} +{"lr": 1.0387683707930187e-05, "data_time": 0.0009604215621948243, "loss": 0.5461299002170563, "time": 1.439703345298767, "epoch": 10, "memory": 6319, "step": 3278} +{"lr": 1.0035100273117234e-05, "data_time": 0.0020652055740356446, "loss": 0.5385670244693757, "time": 1.257918119430542, "epoch": 10, "memory": 6319, "step": 3378} +{"accuracy/top1": 56.21165466308594, "data_time": 1.286743437235974, "time": 1.3777424100536646, "step": 10} diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/config.py b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/config.py new file mode 100644 index 0000000000000000000000000000000000000000..6a57017db12d13dd92877da9c6fa10e7190c05b5 --- /dev/null +++ b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/config.py @@ -0,0 +1,230 @@ +optim_wrapper = dict( + optimizer=dict( + type='SGD', + lr=0.0001, + momentum=0.9, + weight_decay=0.0001, + _scope_='mmpretrain'), + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=False, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=512) +model = dict( + type='ImageClassifier', + backbone=dict( + frozen_stages=2, + type='ResNet', + depth=50, + num_stages=4, + out_indices=(3, ), + style='pytorch', + init_cfg=dict( + type='Pretrained', + checkpoint= + 'https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth', + prefix='backbone')), + neck=dict(type='GlobalAveragePooling'), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=2048, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + topk=1)) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3fake8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3real8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3fake8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3real8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = dict(type='Accuracy', topk=1) +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3fake8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3real8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = dict(type='Accuracy', topk=1) +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1' diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/events.out.tfevents.1685892608.SH-IDC1-10-140-24-127.107193.0 b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/events.out.tfevents.1685892608.SH-IDC1-10-140-24-127.107193.0 new file mode 100644 index 0000000000000000000000000000000000000000..cb2bf43a01a9757439ca0bf8a13adaee0d64df8d --- /dev/null +++ b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/events.out.tfevents.1685892608.SH-IDC1-10-140-24-127.107193.0 @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:d697f92567f7da0022c761ecc098a62f118538915f4ffaf292d8befb77bbde41 +size 13374186 diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/scalars.json b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/scalars.json new file mode 100644 index 0000000000000000000000000000000000000000..73a1a0fa44b07ec9511fc8ac47de62df229a5d0e --- /dev/null +++ b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/scalars.json @@ -0,0 +1,40 @@ +{"lr": 9.981404789023513e-05, "data_time": 0.1522911310195923, "loss": 0.6877720952033997, "time": 1.2160188198089599, "epoch": 1, "memory": 9436, "step": 100} +{"lr": 9.925023268502683e-05, "data_time": 0.02678549289703369, "loss": 0.6727290153503418, "time": 1.2280450344085694, "epoch": 1, "memory": 6319, "step": 200} +{"lr": 9.831328961451971e-05, "data_time": 0.23851115703582765, "loss": 0.6611114025115967, "time": 1.1674016475677491, "epoch": 1, "memory": 6319, "step": 300} +{"accuracy/top1": 49.66403579711914, "data_time": 1.2918182067703783, "time": 1.3832352468144824, "step": 1} +{"lr": 9.635783271319408e-05, "data_time": 0.22280569076538087, "loss": 0.6463721930980683, "time": 1.1441802978515625, "epoch": 2, "memory": 6319, "step": 442} +{"lr": 9.455662799327337e-05, "data_time": 0.0008972644805908203, "loss": 0.6404570877552033, "time": 1.122991156578064, "epoch": 2, "memory": 6319, "step": 542} +{"lr": 9.242187293325357e-05, "data_time": 0.0011354684829711914, "loss": 0.6362692892551423, "time": 1.3448488473892213, "epoch": 2, "memory": 6319, "step": 642} +{"accuracy/top1": 50.54544448852539, "data_time": 1.2973375619326666, "time": 1.3883099667184902, "step": 2} +{"lr": 8.885322471322101e-05, "data_time": 0.023528909683227538, "loss": 0.6221254885196685, "time": 1.1317203283309936, "epoch": 3, "memory": 6319, "step": 784} +{"lr": 8.59909449463049e-05, "data_time": 0.0010162115097045899, "loss": 0.6249741315841675, "time": 1.2448975086212157, "epoch": 3, "memory": 6319, "step": 884} +{"lr": 8.28673425957603e-05, "data_time": 0.0019779443740844727, "loss": 0.6089748561382293, "time": 1.211303210258484, "epoch": 3, "memory": 6319, "step": 984} +{"accuracy/top1": 51.43887710571289, "data_time": 1.2900429536580345, "time": 1.3825226525176024, "step": 3} +{"lr": 7.803482720903195e-05, "data_time": 0.3636435270309448, "loss": 0.6026146233081817, "time": 1.2685081720352174, "epoch": 4, "memory": 6319, "step": 1126} +{"lr": 7.439165228138202e-05, "data_time": 0.0023336410522460938, "loss": 0.5980541050434113, "time": 1.163676118850708, "epoch": 4, "memory": 6319, "step": 1226} +{"lr": 7.058496260180133e-05, "data_time": 0.001241612434387207, "loss": 0.5935293853282928, "time": 1.2686659097671509, "epoch": 4, "memory": 6319, "step": 1326} +{"accuracy/top1": 52.46337127685547, "data_time": 1.2971795717411765, "time": 1.3891826628248476, "step": 4} +{"lr": 6.496162032454504e-05, "data_time": 0.07118077278137207, "loss": 0.5890477299690247, "time": 1.1538291931152345, "epoch": 5, "memory": 6319, "step": 1468} +{"lr": 6.089416958157208e-05, "data_time": 0.0009602546691894531, "loss": 0.584687077999115, "time": 1.1599042654037475, "epoch": 5, "memory": 6319, "step": 1568} +{"lr": 5.677701788155902e-05, "data_time": 0.0011432647705078125, "loss": 0.5784062802791595, "time": 1.186885905265808, "epoch": 5, "memory": 6319, "step": 1668} +{"accuracy/top1": 53.50217819213867, "data_time": 1.302898420189282, "time": 1.3943057240967154, "step": 5} +{"lr": 5.091330063600123e-05, "data_time": 0.0008756875991821289, "loss": 0.5700162708759308, "time": 1.2461036443710327, "epoch": 6, "memory": 6319, "step": 1810} +{"lr": 4.681972449602271e-05, "data_time": 0.0008860111236572265, "loss": 0.5646239936351776, "time": 1.1724539041519164, "epoch": 6, "memory": 6319, "step": 1910} +{"lr": 4.27951262698573e-05, "data_time": 0.001126861572265625, "loss": 0.5662783503532409, "time": 1.1770079135894775, "epoch": 6, "memory": 6319, "step": 2010} +{"accuracy/top1": 54.34123229980469, "data_time": 1.2996568373966495, "time": 1.3910835685952412, "step": 6} +{"lr": 3.726501555491419e-05, "data_time": 0.2574317455291748, "loss": 0.5655214548110962, "time": 1.2420902729034424, "epoch": 7, "memory": 6319, "step": 2152} +{"lr": 3.354602177213381e-05, "data_time": 0.000925898551940918, "loss": 0.5553143858909607, "time": 1.1933157920837403, "epoch": 7, "memory": 6319, "step": 2252} +{"lr": 3.0007932735217966e-05, "data_time": 0.0010452032089233398, "loss": 0.5655701816082, "time": 1.105875825881958, "epoch": 7, "memory": 6319, "step": 2352} +{"accuracy/top1": 54.93189239501953, "data_time": 1.3064878821025436, "time": 1.3970393342109881, "step": 7} +{"lr": 2.5352754318215113e-05, "data_time": 0.0008061408996582031, "loss": 0.5700551688671112, "time": 1.2050061702728272, "epoch": 8, "memory": 6319, "step": 2494} +{"lr": 2.2372383915844294e-05, "data_time": 0.000982189178466797, "loss": 0.5533104658126831, "time": 1.182770037651062, "epoch": 8, "memory": 6319, "step": 2594} +{"lr": 1.9667136874426842e-05, "data_time": 0.001096343994140625, "loss": 0.5499626159667969, "time": 1.2496567964553833, "epoch": 8, "memory": 6319, "step": 2694} +{"accuracy/top1": 55.49565124511719, "data_time": 1.3102325628519753, "time": 1.400934366026008, "step": 8} +{"lr": 1.6342572053355464e-05, "data_time": 0.5104479551315307, "loss": 0.5513653874397277, "time": 1.2293860912322998, "epoch": 9, "memory": 6319, "step": 2836} +{"lr": 1.4392564451840171e-05, "data_time": 0.0009862661361694336, "loss": 0.5432322919368744, "time": 1.2118443250656128, "epoch": 9, "memory": 6319, "step": 2936} +{"lr": 1.2784967834899793e-05, "data_time": 0.0008760213851928711, "loss": 0.5493425965309143, "time": 1.203943133354187, "epoch": 9, "memory": 6319, "step": 3036} +{"accuracy/top1": 55.81559371948242, "data_time": 1.2946712817812105, "time": 1.3870672303803113, "step": 9} +{"lr": 1.1116448178051759e-05, "data_time": 0.3279048204421997, "loss": 0.5497923195362091, "time": 1.2743196964263916, "epoch": 10, "memory": 6319, "step": 3178} +{"lr": 1.0387683707930187e-05, "data_time": 0.0009604215621948243, "loss": 0.5461299002170563, "time": 1.439703345298767, "epoch": 10, "memory": 6319, "step": 3278} +{"lr": 1.0035100273117234e-05, "data_time": 0.0020652055740356446, "loss": 0.5385670244693757, "time": 1.257918119430542, "epoch": 10, "memory": 6319, "step": 3378} +{"accuracy/top1": 56.21165466308594, "data_time": 1.286743437235974, "time": 1.3777424100536646, "step": 10} diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/35624.png_1.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/35624.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..1107faa750f3dc7f4e35aa32af572cf3c1c797fd Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/35624.png_1.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/35624.png_10.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/35624.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..c4cf3e8501875b82a6c0e435e6c4a6d5621a591d Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/35624.png_10.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/35624.png_2.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/35624.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..afc3ccea1ddbb24895aecd4cc68095eabf64f733 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/35624.png_2.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/35624.png_3.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/35624.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..5abe8dccbb08c159f9022dc0e0deb50dc34d4a46 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/35624.png_3.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/35624.png_4.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/35624.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..0ab835c1eda355ce68acfb89e36643b2a243f87a Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/35624.png_4.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/35624.png_5.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/35624.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..15c35350ef7e5a3cbaf494992835fba0da22625e Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/35624.png_5.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/35624.png_6.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/35624.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..1479b8052f880a36bcaf8eb91320e0192dacb014 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/35624.png_6.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/35624.png_7.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/35624.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..37fae2b3153a451473923dce25870b208a11db0e Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/35624.png_7.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/35624.png_8.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/35624.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..946272ba2fa92e21e180905116408f2c9f580735 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/35624.png_8.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/35624.png_9.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/35624.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..3844c080c574d2548073cf6704fc196200a7f64d Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/35624.png_9.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/38442.png_1.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/38442.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..91af4d88126ed9797cce8006ca2a512277fb7d02 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/38442.png_1.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/38442.png_10.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/38442.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..e1172a2a4f929092a4d7d0050068f54c6d26fe20 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/38442.png_10.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/38442.png_2.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/38442.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..f55722c141963b8edd8f64e370a869c58c378215 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/38442.png_2.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/38442.png_3.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/38442.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..24f6fca6766ecc3da9642dddd15f227635e93e3d Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/38442.png_3.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/38442.png_4.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/38442.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..91d95f230ba5bd63141d95e9d1d83bfde73c4db9 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/38442.png_4.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/38442.png_5.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/38442.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..42b611b0168418b43d822573f324275e7d360fb9 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/38442.png_5.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/38442.png_6.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/38442.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..ea9785cabd2758ec7c826dc8fb699163a867575d Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/38442.png_6.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/38442.png_7.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/38442.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..a34fafe0a4707dd8ffb8cf24fb0b404790030e77 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/38442.png_7.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/38442.png_8.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/38442.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..519ffe841fc9782359cd3c902ef3f6b4e4b0912a Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/38442.png_8.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/38442.png_9.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/38442.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..0b46d101a2394e32adbb88e8b42be13fa53889d8 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/38442.png_9.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/43283.png_1.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/43283.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..2b50b5a8f2ea60ae20bef28460d088c537bceadb Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/43283.png_1.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/43283.png_10.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/43283.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..8ef37479f20be47d84a0879cb8acc18122f8f5c3 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/43283.png_10.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/43283.png_2.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/43283.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..e2366301c7e97aa5025e8086e4d2f5e8bd8841c8 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/43283.png_2.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/43283.png_3.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/43283.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..9c950385e7c8231a36d15699fcec79f7d80fd7f7 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/43283.png_3.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/43283.png_4.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/43283.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..ba39a33754e14f84c9667877050548957d6bf43b Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/43283.png_4.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/43283.png_5.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/43283.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..d9dd2fe7e92d3a7e86812c96800985512969a843 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/43283.png_5.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/43283.png_6.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/43283.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..c83e53b888c19a9685bbd6676e97a1b85256e852 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/43283.png_6.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/43283.png_7.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/43283.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..fac64af32156c30cbc3ea8dd3c589cf9ab1fd9ad Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/43283.png_7.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/43283.png_8.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/43283.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..b0a878f216aae15502c0a703cb5206e877bf9072 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/43283.png_8.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/43283.png_9.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/43283.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..6e3860e9ec4aaa5c01386f65700fc80807f19465 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/43283.png_9.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/44982.png_1.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/44982.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..af40752c3a8c0bbda5d673b6c5a97bc88859f12f Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/44982.png_1.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/44982.png_10.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/44982.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..67ba8fbe6f05b810f597faa96ad961536d638f39 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/44982.png_10.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/44982.png_2.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/44982.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..2c7cb2a0b06014cae9d96f388dbc338aabc824de Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/44982.png_2.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/44982.png_3.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/44982.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..7bc6c38fb67c7aace75a4cd71b2cd7ed41353663 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/44982.png_3.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/44982.png_4.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/44982.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..11cdd07e5851e95186bfbcfc8486bd1709c8dbd2 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/44982.png_4.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/44982.png_5.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/44982.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..4a5f5330b781325dfa8459f73ac52226ef12f91e Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/44982.png_5.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/44982.png_6.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/44982.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..2e519da1f20767ae45b12861c34117d48ad5f4bf Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/44982.png_6.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/44982.png_7.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/44982.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..b740d37dd1dfebd20ab02e386104511c222f1da1 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/44982.png_7.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/44982.png_8.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/44982.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..56515015200eb7f7f8f856094428410602366783 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/44982.png_8.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/44982.png_9.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/44982.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..1f47afda16314df7e01695f12e84efe44a17cfa8 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/44982.png_9.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/56423.png_1.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/56423.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..d7ad8a93a8c99bd6bb5caf5f2bb1fb4e7c15ff1b Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/56423.png_1.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/56423.png_10.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/56423.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..f5396e28d42550f5a5256f938c18907bbf31ab99 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/56423.png_10.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/56423.png_2.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/56423.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..f27fdcd4f130b1e4e7fed6f7d763f4378b6e28cb Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/56423.png_2.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/56423.png_3.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/56423.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..984963f8f10534760c935aadf39b0a9d38cd9f50 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/56423.png_3.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/56423.png_4.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/56423.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..bea26c4b59aadc74021d1b02451cbea3ca878303 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/56423.png_4.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/56423.png_5.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/56423.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..4e602e28a14ff2bda8b03c81832f3e06d3ed3e83 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/56423.png_5.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/56423.png_6.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/56423.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..3de0bfd47f1fbd35f4e47f14996d5f169847e61b Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/56423.png_6.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/56423.png_7.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/56423.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..06fc37862758ca3dea471d51026bfc8d09de560d Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/56423.png_7.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/56423.png_8.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/56423.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..9763d4edb2d0198a0e7103c3dfbde06ab8bd650d Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/56423.png_8.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/56423.png_9.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/56423.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..b635dcfa459e0b9555f0b5946ba0e839bec69539 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/56423.png_9.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/61024.png_1.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/61024.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..bd07d0e01dac5c5a6dc124e3f6746550db925c0b Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/61024.png_1.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/61024.png_10.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/61024.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..d563829418bc4bafb162a1338fa4d4ff582c4911 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/61024.png_10.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/61024.png_2.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/61024.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..5f088fb0d51791b2dcc0a1d4d231ca4975f6fe09 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/61024.png_2.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/61024.png_3.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/61024.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..4d837a14d46aca96bad9cc51e7801c872c94b8e4 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/61024.png_3.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/61024.png_4.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/61024.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..0bb82de97113ca1063880ec2a03a93b54b85cb62 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/61024.png_4.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/61024.png_5.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/61024.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..5b50ace6718bbcb68e72d66ebd2ef79a3e6c1206 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/61024.png_5.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/61024.png_6.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/61024.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..fda8df6101d9bd5522f1211712dd9dad1b550259 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/61024.png_6.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/61024.png_7.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/61024.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..a9398f4a9a96b2b93e07a92c144f171c081a022e Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/61024.png_7.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/61024.png_8.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/61024.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..4d1faa40f78939b6c13b13ed26c061ddd5594fbf Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/61024.png_8.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/61024.png_9.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/61024.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..ec09d606a95a4cc944da1345e67a1829f35db6ed Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/61024.png_9.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/66390.png_1.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/66390.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..171c48d29caf2d5547201c010d4c2f4f4bfc37af Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/66390.png_1.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/66390.png_10.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/66390.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..c01306b0212c69bc71ad01ff4eda90365a7aa46b Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/66390.png_10.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/66390.png_2.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/66390.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..e25613ffa620fb4f05d6b7a3c3c766dbd93d6c98 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/66390.png_2.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/66390.png_3.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/66390.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..08e28b45e4334e0c5efe78ce13f70ee1c3d63847 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/66390.png_3.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/66390.png_4.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/66390.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..59102f0d6b4b4644806505c708b99632bbbead32 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/66390.png_4.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/66390.png_5.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/66390.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..5bbe82ff643354b59b655d143b8c2692b954aff9 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/66390.png_5.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/66390.png_6.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/66390.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..a9ddea36a0921c39048cde521baa1c67df8a2bec Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/66390.png_6.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/66390.png_7.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/66390.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..ceef7e823f675cfbaea4a7d7b133c6cd6ccff56c Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/66390.png_7.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/66390.png_8.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/66390.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..87bace7158012409809fef952815d4594c9970d8 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/66390.png_8.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/66390.png_9.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/66390.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..79e1f48e9f85ada24bc6ee4f2426544232454f02 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/66390.png_9.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/flickr_wild_000179.jpg_1.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/flickr_wild_000179.jpg_1.png new file mode 100644 index 0000000000000000000000000000000000000000..69a8109a5b10213cd715122414ccfb16f903a1db Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/flickr_wild_000179.jpg_1.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/flickr_wild_000179.jpg_10.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/flickr_wild_000179.jpg_10.png new file mode 100644 index 0000000000000000000000000000000000000000..cf62529b23d551b26040f711a9b7bd9bc85630d6 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/flickr_wild_000179.jpg_10.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/flickr_wild_000179.jpg_2.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/flickr_wild_000179.jpg_2.png new file mode 100644 index 0000000000000000000000000000000000000000..4aaea38945bcead382f47ea6f5d51a14993811a2 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/flickr_wild_000179.jpg_2.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/flickr_wild_000179.jpg_3.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/flickr_wild_000179.jpg_3.png new file mode 100644 index 0000000000000000000000000000000000000000..c4a5234843d825bf2f9799d3795148dff8fd6c52 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/flickr_wild_000179.jpg_3.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/flickr_wild_000179.jpg_4.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/flickr_wild_000179.jpg_4.png new file mode 100644 index 0000000000000000000000000000000000000000..4ba5023f8cef1368d084a193269c5ef37421a775 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/flickr_wild_000179.jpg_4.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/flickr_wild_000179.jpg_5.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/flickr_wild_000179.jpg_5.png new file mode 100644 index 0000000000000000000000000000000000000000..51ad47c6dd33b2baf2616506b51b7fb33dbc585c Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/flickr_wild_000179.jpg_5.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/flickr_wild_000179.jpg_6.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/flickr_wild_000179.jpg_6.png new file mode 100644 index 0000000000000000000000000000000000000000..834d05d8a7d74decd3f1babf6b975666a0d39f7d Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/flickr_wild_000179.jpg_6.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/flickr_wild_000179.jpg_7.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/flickr_wild_000179.jpg_7.png new file mode 100644 index 0000000000000000000000000000000000000000..c7d4eb1b0c8dd27ad26a0b51d76aab14553bc1ed Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/flickr_wild_000179.jpg_7.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/flickr_wild_000179.jpg_8.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/flickr_wild_000179.jpg_8.png new file mode 100644 index 0000000000000000000000000000000000000000..3cead0a55b2f67ef492f0d6d96dacbaebd581c83 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/flickr_wild_000179.jpg_8.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/flickr_wild_000179.jpg_9.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/flickr_wild_000179.jpg_9.png new file mode 100644 index 0000000000000000000000000000000000000000..9c33b916f53aeccafc5f73813033985cd28d48e5 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/flickr_wild_000179.jpg_9.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/pixabay_dog_002352.jpg_1.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/pixabay_dog_002352.jpg_1.png new file mode 100644 index 0000000000000000000000000000000000000000..4d9003cca3d62bfda6a8299e9ea762251b7a49d8 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/pixabay_dog_002352.jpg_1.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/pixabay_dog_002352.jpg_10.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/pixabay_dog_002352.jpg_10.png new file mode 100644 index 0000000000000000000000000000000000000000..8b4e0362d6e2c6cef994202e4ffdd8e59ae623c4 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/pixabay_dog_002352.jpg_10.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/pixabay_dog_002352.jpg_2.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/pixabay_dog_002352.jpg_2.png new file mode 100644 index 0000000000000000000000000000000000000000..dfbf064822dc6848bf78968ca17631879d3f4ab5 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/pixabay_dog_002352.jpg_2.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/pixabay_dog_002352.jpg_3.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/pixabay_dog_002352.jpg_3.png new file mode 100644 index 0000000000000000000000000000000000000000..5b6cec22a7c57851eeb754ac47bb0f47fe96d571 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/pixabay_dog_002352.jpg_3.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/pixabay_dog_002352.jpg_4.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/pixabay_dog_002352.jpg_4.png new file mode 100644 index 0000000000000000000000000000000000000000..73c6ee519f735f107df4fec46e85971e8fff9681 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/pixabay_dog_002352.jpg_4.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/pixabay_dog_002352.jpg_5.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/pixabay_dog_002352.jpg_5.png new file mode 100644 index 0000000000000000000000000000000000000000..2b17b7502e85ac3d921d8c0c32d081612cffa699 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/pixabay_dog_002352.jpg_5.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/pixabay_dog_002352.jpg_6.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/pixabay_dog_002352.jpg_6.png new file mode 100644 index 0000000000000000000000000000000000000000..3a2016fc0ae5f242bcc30735aecf676dba873096 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/pixabay_dog_002352.jpg_6.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/pixabay_dog_002352.jpg_7.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/pixabay_dog_002352.jpg_7.png new file mode 100644 index 0000000000000000000000000000000000000000..ad95b4aa6f3eed8f7ecabaf4a2cfe8437b409a6f Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/pixabay_dog_002352.jpg_7.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/pixabay_dog_002352.jpg_8.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/pixabay_dog_002352.jpg_8.png new file mode 100644 index 0000000000000000000000000000000000000000..b780754fd5030e7051b9cb93ee998eb1b5415b2d Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/pixabay_dog_002352.jpg_8.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/pixabay_dog_002352.jpg_9.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/pixabay_dog_002352.jpg_9.png new file mode 100644 index 0000000000000000000000000000000000000000..e13bd722d16cabcc74186c6a3733a32e0c5a64bc Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/pixabay_dog_002352.jpg_9.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed14088.png_1.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed14088.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..d8027e1d800d9d854c5e22c56e04ec8624dd03a6 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed14088.png_1.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed14088.png_10.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed14088.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..96fac8665779c004fb6e641e8f5808148b3a607e Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed14088.png_10.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed14088.png_2.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed14088.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..c978b5f0491d0d00c7e324feeb0f5da286725ea7 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed14088.png_2.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed14088.png_3.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed14088.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..7efe078fe1022acd951d6e2916478e1e6c62d8e1 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed14088.png_3.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed14088.png_4.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed14088.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..6b302e44b96619f1523fb4cb7eaad2c599b2e0e5 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed14088.png_4.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed14088.png_5.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed14088.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..948033ebbe9980709cccc63ffd6f5c81cb44daaa Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed14088.png_5.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed14088.png_6.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed14088.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..ff5993ec4120a264cdd51d6904d90f2815900da2 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed14088.png_6.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed14088.png_7.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed14088.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..6d4a62f4c6cb3ec663d5e6175def78114099fb66 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed14088.png_7.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed14088.png_8.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed14088.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..f94d36e91b28fa3e0ef16ee8a968746a5cbfbc1c Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed14088.png_8.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed14088.png_9.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed14088.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..51858303c778436817137d4a2450dfadad49801a Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed14088.png_9.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed16124.png_1.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed16124.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..8229d890efca1049176ac67e79d21ba101ea470a Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed16124.png_1.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed16124.png_10.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed16124.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..39919eb556a7029f518d37bc35f0e5c78229e03a Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed16124.png_10.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed16124.png_2.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed16124.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..5a80c2ddbdd4dc9bec41c7615b5097b82e3503ef Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed16124.png_2.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed16124.png_3.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed16124.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..93e1b505c01c3ca86d1feb91143250d974d13028 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed16124.png_3.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed16124.png_4.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed16124.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..21519159777dcced4ae614575fbd27048f1df76c Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed16124.png_4.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed16124.png_5.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed16124.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..c537bd1048e02123fe1b923c7a334c284eeb6afe Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed16124.png_5.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed16124.png_6.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed16124.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..30a34179b7ba1aabd299fbff7b67681d67563231 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed16124.png_6.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed16124.png_7.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed16124.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..52ec226413e3d87037168ad8f09a594ad90bf985 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed16124.png_7.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed16124.png_8.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed16124.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..9840b907adaa840116344123de81e68d391c162a Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed16124.png_8.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed16124.png_9.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed16124.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..adbf722833cd4e3199e013b64c78a40d04a0b8be Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed16124.png_9.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed17282.png_1.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed17282.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..ba00f1de60bc4e8bc45fd7c81924628a57258cc3 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed17282.png_1.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed17282.png_10.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed17282.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..192e491a4fe94e522577188d15d9599bdd4a897c Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed17282.png_10.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed17282.png_2.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed17282.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..8cbbd17eda97b82741f19ca7a267102a17ea1509 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed17282.png_2.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed17282.png_3.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed17282.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..6a2ba3c98b6db9916f80c98adc984f291d2be1cb Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed17282.png_3.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed17282.png_4.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed17282.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..a58f7fa08ad0a0e4cfd6c696d3e1c24254f8b744 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed17282.png_4.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed17282.png_5.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed17282.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..5ea5ad66482427066f9ada9379e3111d8125b553 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed17282.png_5.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed17282.png_6.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed17282.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..2c0c482f85222b62cb52ad1cfba7429294ed3ba0 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed17282.png_6.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed17282.png_7.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed17282.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..4c4adb38a2525a2f1c94e64a793b01c142bc207d Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed17282.png_7.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed17282.png_8.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed17282.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..e81ac59ad5283a589dddd63b76d4910fc16b4b8a Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed17282.png_8.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed17282.png_9.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed17282.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..be5f11f138a2014c01b76ba4da63b394c9db50d9 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed17282.png_9.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed21815.png_1.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed21815.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..2b1a36e15b93cb8b3700cc106acc61a74f981ee5 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed21815.png_1.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed21815.png_10.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed21815.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..d6250af5304635ff0ab83040f38d8ac69daafbde Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed21815.png_10.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed21815.png_2.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed21815.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..012e7bab813b8ee52f187a94bf59b2937bcec1d3 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed21815.png_2.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed21815.png_3.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed21815.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..0ff797d52b0275639008986a65b8f02fb589ddbd Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed21815.png_3.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed21815.png_4.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed21815.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..0584ef70c17f6aaec828390b81dd1bf35adfd30d Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed21815.png_4.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed21815.png_5.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed21815.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..dd7fe8bae185c3cb42b932523f41b52b2fddf0af Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed21815.png_5.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed21815.png_6.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed21815.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..4123682aac86b33805e7225efe4009484b2fb656 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed21815.png_6.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed21815.png_7.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed21815.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..106ee6c651342f023d0e3421aeacbf0281e8b812 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed21815.png_7.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed21815.png_8.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed21815.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..79992c800f2156283fe42a663d5d9842a78ca1e3 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed21815.png_8.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed21815.png_9.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed21815.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..872cd0784068960bdba8d5890787cfa265f4b64c Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed21815.png_9.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed22624.png_1.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed22624.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..b77e09a7b6cbc7a2ac23e963e38292d29b3b01d6 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed22624.png_1.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed22624.png_10.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed22624.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..6a0534e332ca7557a3c251a3abf071c1f07b9a7e Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed22624.png_10.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed22624.png_2.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed22624.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..4ba25773178ca3a4758c23b438fd7b9c00652f8c Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed22624.png_2.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed22624.png_3.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed22624.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..95f7a989cc5ccad4f73b15140f1dfe4f9a203966 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed22624.png_3.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed22624.png_4.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed22624.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..2d01e8fd8eda1e79cdc65e0e76b1d0586526a2e5 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed22624.png_4.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed22624.png_5.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed22624.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..caad0abddd5d91e80c5dab9b21b3dff92019d026 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed22624.png_5.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed22624.png_6.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed22624.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..7bfe5fcf1893355f6327df9052fe85d7c972c9a0 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed22624.png_6.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed22624.png_7.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed22624.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..efcb9d9afae559f0346a927fc30a296d5d4f0d27 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed22624.png_7.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed22624.png_8.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed22624.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..b13098d725f2e7670a5934af2f506d74533ac556 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed22624.png_8.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed22624.png_9.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed22624.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..2534467d72a363efb8d5968f2740f8a7917f0ec1 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed22624.png_9.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed23602.png_1.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed23602.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..28f904ef2412b594a3a09d73ed842e6cc56d49a5 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed23602.png_1.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed23602.png_10.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed23602.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..e6e8c5e82978db18735c75a7b40dffd39c5e766c Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed23602.png_10.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed23602.png_2.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed23602.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..5be8c1dca5e52655933b515d57ab333b3262b481 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed23602.png_2.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed23602.png_3.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed23602.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..39c4ab9e6dc61e749bfb411abd29dfba22d3f677 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed23602.png_3.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed23602.png_4.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed23602.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..5a89c833ac0a28bac000898196076d41245d82da Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed23602.png_4.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed23602.png_5.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed23602.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..34dc3c243758471c2013aacc41b6e596ef3ee3cd Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed23602.png_5.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed23602.png_6.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed23602.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..976eee81fa3914d282b39aadd23f0fcc86f935f1 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed23602.png_6.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed23602.png_7.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed23602.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..52e4ebe78b076c018d0da5352ca290c338831a6d Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed23602.png_7.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed23602.png_8.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed23602.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..43dc3798e8769069cff4e26ea6ba9221df8c2cd9 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed23602.png_8.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed23602.png_9.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed23602.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..a83a800f6e150dd0649d2e373242028c92b0fac3 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed23602.png_9.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed24366.png_1.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed24366.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..4937a0cbc4fb23bfb743164ae15d1769f9727a7d Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed24366.png_1.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed24366.png_10.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed24366.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..0ff3ca2027528a2b391cf31a5217305c553561ed Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed24366.png_10.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed24366.png_2.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed24366.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..9b67b46efe578d5fc32de6d341cc4d56eec39d1a Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed24366.png_2.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed24366.png_3.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed24366.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..4be0b03d362d2b0b0a27589ef83a759ab7f06398 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed24366.png_3.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed24366.png_4.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed24366.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..92796e8843da4f51e2a71a2d94bb7e1533e4977f Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed24366.png_4.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed24366.png_5.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed24366.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..1220c148076b4d983486856caf53e94f1585cbb5 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed24366.png_5.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed24366.png_6.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed24366.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..a65bfef1cba9bc7872087c88e01275c5e5c5a848 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed24366.png_6.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed24366.png_7.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed24366.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..3a4848f61f79760c545d1df344276560228bc344 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed24366.png_7.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed24366.png_8.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed24366.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..e508049ebf12e4c96831b396d567ae2381d3c353 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed24366.png_8.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed24366.png_9.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed24366.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..832fb7aec893ff04eca236769c469b80c3d2004b Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed24366.png_9.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed26848.png_1.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed26848.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..795a57ef21dfa55c23e1d4d91df4375f5302c603 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed26848.png_1.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed26848.png_10.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed26848.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..96bb46cd687a0698342e4b75eddee72593565673 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed26848.png_10.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed26848.png_2.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed26848.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..5a1976c43fcc5c772b2725474fc5f5a533a93b21 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed26848.png_2.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed26848.png_3.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed26848.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..e2e52a29d061297340276eb26ba206d5d464b403 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed26848.png_3.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed26848.png_4.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed26848.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..dac2eaf05ad300b65c74e5d897727d73f6c4e633 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed26848.png_4.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed26848.png_5.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed26848.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..9fa541519ca1cca6837169940f5f5c2ef42a8eb2 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed26848.png_5.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed26848.png_6.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed26848.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..e38e0015d9caf5805c023524f1a714302b04d0b1 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed26848.png_6.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed26848.png_7.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed26848.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..f247c7db730261423614ab3c7582003b8a718423 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed26848.png_7.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed26848.png_8.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed26848.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..1ee66e4b561a466302472bdd022229e047ec0b74 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed26848.png_8.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed26848.png_9.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed26848.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..fdc1a971248a6d1294322a4286c6e4d749d6295d Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed26848.png_9.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed44086.png_1.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed44086.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..38b1584fae4b30e73a95772fb4313c0c3b1c8135 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed44086.png_1.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed44086.png_10.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed44086.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..162ae887c1b2947dca85ebd2800f9b9459d52dca Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed44086.png_10.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed44086.png_2.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed44086.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..3da4ef065486292196249d91693dad9853ba5e2b Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed44086.png_2.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed44086.png_3.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed44086.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..adca0e7c841f2e608faa0595ddbc8f15ede0f554 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed44086.png_3.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed44086.png_4.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed44086.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..03f661fdb496e0d9cabdae4b12eda5801834eded Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed44086.png_4.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed44086.png_5.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed44086.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..48f0a9f3680390289883affbee64b0d321d29fd3 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed44086.png_5.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed44086.png_6.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed44086.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..ed06d88e282b6e94d186a2b7c51845e8b6e0d4b3 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed44086.png_6.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed44086.png_7.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed44086.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..98ac6b68763963daa1f639fc3aa961215dd20c51 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed44086.png_7.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed44086.png_8.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed44086.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..62f310665d89d276b4064936a84565b1c045695a Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed44086.png_8.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed44086.png_9.png b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed44086.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..ecf07ab811ded04a8ec9410f8ddf45821f418a45 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/20230604_232954/vis_data/vis_image/seed44086.png_9.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/epoch_1.pth b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/epoch_1.pth new file mode 100644 index 0000000000000000000000000000000000000000..7a4f13377a36cfda407cd35c6997001989c36c1f --- /dev/null +++ b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/epoch_1.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:0f8511cfaeda985a459eba45e962ff31e628f8b4493354ffc7931ccbc188b8e4 +size 277074189 diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/epoch_10.pth b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/epoch_10.pth new file mode 100644 index 0000000000000000000000000000000000000000..2646b33b09f95c053d79f847e140065c18d9e2a3 --- /dev/null +++ b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/epoch_10.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:23c4c900d7fc5c014ab3e81d4e4d99c7884c75e219bb5eb9168941123f012492 +size 277405901 diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/epoch_2.pth b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/epoch_2.pth new file mode 100644 index 0000000000000000000000000000000000000000..a16b3792680d9a88b4d0f6bd35ffd557ddbcd5f7 --- /dev/null +++ b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/epoch_2.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:d54883044233970a8fa6bdba5411a62181c82a7fdd4e2b1936d0296c7a9e04fc +size 277111693 diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/epoch_3.pth b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/epoch_3.pth new file mode 100644 index 0000000000000000000000000000000000000000..15b540ca51b5a185878bf93b2ba09e84c0cc2efb --- /dev/null +++ b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/epoch_3.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:d92f29002cbf57b4ce15755d8e896408ed5461b60b33b2c0c18f611421d5fd19 +size 277148429 diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/epoch_4.pth b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/epoch_4.pth new file mode 100644 index 0000000000000000000000000000000000000000..27d0404928ba31155999e66160b2155b84c71c04 --- /dev/null +++ b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/epoch_4.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:d2aa47d89b304972abfb1f0d662cff5cee339199170a61b31918bc19f19e3c42 +size 277185037 diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/epoch_5.pth b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/epoch_5.pth new file mode 100644 index 0000000000000000000000000000000000000000..ff1794e1eccc0647deaf135fa8770ea782d27274 --- /dev/null +++ b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/epoch_5.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:16b3b0350314d267bfe243f3b9d48537e58ce2bff522925c6d0fd71bb1a1432e +size 277221645 diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/epoch_6.pth b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/epoch_6.pth new file mode 100644 index 0000000000000000000000000000000000000000..b05f8e92acd5c6614b4ca61247549cfaaaddb234 --- /dev/null +++ b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/epoch_6.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:3ba1f978e8574e972405796145896db4794af7dec2228bfd0630ad8be23ba38a +size 277258317 diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/epoch_7.pth b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/epoch_7.pth new file mode 100644 index 0000000000000000000000000000000000000000..86af345a1a1b87c695c8b22ffe4e8c01940a9f36 --- /dev/null +++ b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/epoch_7.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:6b1b481a47b183d9e9d041fa8ee4ecc92b24db8f220a2a16187ded48da5e0c2a +size 277295053 diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/epoch_8.pth b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/epoch_8.pth new file mode 100644 index 0000000000000000000000000000000000000000..1a19387000cfb5d81bcf19d3ea566036743d6c28 --- /dev/null +++ b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/epoch_8.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:20b86031927a3d4aa8d0ee2a42d6b00f7daeee14bffd5b768a8d3663a530fb27 +size 277331981 diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/epoch_9.pth b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/epoch_9.pth new file mode 100644 index 0000000000000000000000000000000000000000..46d5aa7f7fb2558322bdefa47c82d3504fbd9491 --- /dev/null +++ b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/epoch_9.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c69bceb8133b07b860b60d20b52d9e8f9d172ae77e03bde7e9c351e859d30cdf +size 277368909 diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/last_checkpoint b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/last_checkpoint new file mode 100644 index 0000000000000000000000000000000000000000..99e9997610d6bcc2e61f3ad12fb58147936afbb8 --- /dev/null +++ b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/last_checkpoint @@ -0,0 +1 @@ +/mnt/petrelfs/luzeyu/workspace/fakebench/mmpretrain/workdir/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/epoch_10.pth \ No newline at end of file diff --git a/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1.py b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1.py new file mode 100644 index 0000000000000000000000000000000000000000..6a57017db12d13dd92877da9c6fa10e7190c05b5 --- /dev/null +++ b/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1.py @@ -0,0 +1,230 @@ +optim_wrapper = dict( + optimizer=dict( + type='SGD', + lr=0.0001, + momentum=0.9, + weight_decay=0.0001, + _scope_='mmpretrain'), + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=False, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=512) +model = dict( + type='ImageClassifier', + backbone=dict( + frozen_stages=2, + type='ResNet', + depth=50, + num_stages=4, + out_indices=(3, ), + style='pytorch', + init_cfg=dict( + type='Pretrained', + checkpoint= + 'https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth', + prefix='backbone')), + neck=dict(type='GlobalAveragePooling'), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=2048, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + topk=1)) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3fake8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3real8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3fake8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3real8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = dict(type='Accuracy', topk=1) +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3fake8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3real8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = dict(type='Accuracy', topk=1) +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/resnet50_2xb256_stylegan3_1m_lr1e-4_aug_5e-1' diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/20230605_152245.log b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/20230605_152245.log new file mode 100644 index 0000000000000000000000000000000000000000..44c511c22704937741e0d42fbcacfa6edc8483e1 --- /dev/null +++ b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/20230605_152245.log @@ -0,0 +1,939 @@ +2023/06/05 15:22:48 - mmengine - INFO - +------------------------------------------------------------ +System environment: + sys.platform: linux + Python: 3.10.9 (main, Mar 8 2023, 10:47:38) [GCC 11.2.0] + CUDA available: True + numpy_random_seed: 1528626734 + GPU 0,1: NVIDIA A100-SXM4-80GB + CUDA_HOME: /mnt/petrelfs/share/cuda-11.6 + NVCC: Cuda compilation tools, release 11.6, V11.6.124 + GCC: gcc (GCC) 7.5.0 + PyTorch: 1.13.1 + PyTorch compiling details: PyTorch built with: + - GCC 9.3 + - C++ Version: 201402 + - Intel(R) oneAPI Math Kernel Library Version 2021.4-Product Build 20210904 for Intel(R) 64 architecture applications + - Intel(R) MKL-DNN v2.6.0 (Git Hash 52b5f107dd9cf10910aaa19cb47f3abf9b349815) + - OpenMP 201511 (a.k.a. OpenMP 4.5) + - LAPACK is enabled (usually provided by MKL) + - NNPACK is enabled + - CPU capability usage: AVX2 + - CUDA Runtime 11.6 + - NVCC architecture flags: -gencode;arch=compute_37,code=sm_37;-gencode;arch=compute_50,code=sm_50;-gencode;arch=compute_60,code=sm_60;-gencode;arch=compute_61,code=sm_61;-gencode;arch=compute_70,code=sm_70;-gencode;arch=compute_75,code=sm_75;-gencode;arch=compute_80,code=sm_80;-gencode;arch=compute_86,code=sm_86;-gencode;arch=compute_37,code=compute_37 + - CuDNN 8.3.2 (built against CUDA 11.5) + - Magma 2.6.1 + - Build settings: BLAS_INFO=mkl, BUILD_TYPE=Release, CUDA_VERSION=11.6, CUDNN_VERSION=8.3.2, CXX_COMPILER=/opt/rh/devtoolset-9/root/usr/bin/c++, CXX_FLAGS= -fabi-version=11 -Wno-deprecated -fvisibility-inlines-hidden -DUSE_PTHREADPOOL -fopenmp -DNDEBUG -DUSE_KINETO -DUSE_FBGEMM -DUSE_QNNPACK -DUSE_PYTORCH_QNNPACK -DUSE_XNNPACK -DSYMBOLICATE_MOBILE_DEBUG_HANDLE -DEDGE_PROFILER_USE_KINETO -O2 -fPIC -Wno-narrowing -Wall -Wextra -Werror=return-type -Werror=non-virtual-dtor -Wno-missing-field-initializers -Wno-type-limits -Wno-array-bounds -Wno-unknown-pragmas -Wunused-local-typedefs -Wno-unused-parameter -Wno-unused-function -Wno-unused-result -Wno-strict-overflow -Wno-strict-aliasing -Wno-error=deprecated-declarations -Wno-stringop-overflow -Wno-psabi -Wno-error=pedantic -Wno-error=redundant-decls -Wno-error=old-style-cast -fdiagnostics-color=always -faligned-new -Wno-unused-but-set-variable -Wno-maybe-uninitialized -fno-math-errno -fno-trapping-math -Werror=format -Werror=cast-function-type -Wno-stringop-overflow, LAPACK_INFO=mkl, PERF_WITH_AVX=1, PERF_WITH_AVX2=1, PERF_WITH_AVX512=1, TORCH_VERSION=1.13.1, USE_CUDA=ON, USE_CUDNN=ON, USE_EXCEPTION_PTR=1, USE_GFLAGS=OFF, USE_GLOG=OFF, USE_MKL=ON, USE_MKLDNN=ON, USE_MPI=OFF, USE_NCCL=ON, USE_NNPACK=ON, USE_OPENMP=ON, USE_ROCM=OFF, + + TorchVision: 0.14.1 + OpenCV: 4.7.0 + MMEngine: 0.7.3 + +Runtime environment: + cudnn_benchmark: True + mp_cfg: {'mp_start_method': 'fork', 'opencv_num_threads': 0} + dist_cfg: {'backend': 'nccl'} + seed: None + deterministic: False + Distributed launcher: slurm + Distributed training: True + GPU number: 2 +------------------------------------------------------------ + +2023/06/05 15:22:52 - mmengine - INFO - Config: +optim_wrapper = dict( + optimizer=dict( + type='SGD', + lr=0.0001, + momentum=0.9, + weight_decay=0.0001, + _scope_='mmpretrain'), + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=False, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=512) +model = dict( + type='ImageClassifier', + backbone=dict( + frozen_stages=2, + type='ResNet', + depth=50, + num_stages=4, + out_indices=(3, ), + style='pytorch', + init_cfg=dict( + type='Pretrained', + checkpoint= + 'https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth', + prefix='backbone')), + neck=dict(type='GlobalAveragePooling'), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=2048, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + topk=1)) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3fake8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3real8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3fake8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3real8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = dict(type='Accuracy', topk=1) +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3fake8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3real8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = dict(type='Accuracy', topk=1) +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1' + +2023/06/05 15:23:04 - mmengine - INFO - Hooks will be executed in the following order: +before_run: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_load_checkpoint: +(ABOVE_NORMAL) EMAHook + -------------------- +before_train: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_train_epoch: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(NORMAL ) DistSamplerSeedHook + -------------------- +before_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook + -------------------- +after_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +after_train_epoch: +(NORMAL ) IterTimerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_val_epoch: +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook + -------------------- +before_val_iter: +(NORMAL ) IterTimerHook + -------------------- +after_val_iter: +(NORMAL ) IterTimerHook +(NORMAL ) VisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_val_epoch: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_save_checkpoint: +(ABOVE_NORMAL) EMAHook + -------------------- +after_train: +(VERY_LOW ) CheckpointHook + -------------------- +before_test_epoch: +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook + -------------------- +before_test_iter: +(NORMAL ) IterTimerHook + -------------------- +after_test_iter: +(NORMAL ) IterTimerHook +(NORMAL ) VisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_test_epoch: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_run: +(BELOW_NORMAL) LoggerHook + -------------------- +2023/06/05 15:23:16 - mmengine - INFO - load backbone in model from: https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth +Name of parameter - Initialization information + +backbone.conv1.weight - torch.Size([64, 3, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.bn1.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.bn1.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.conv1.weight - torch.Size([64, 64, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.bn1.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.bn1.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.conv2.weight - torch.Size([64, 64, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.bn2.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.bn2.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.conv3.weight - torch.Size([256, 64, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.bn3.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.bn3.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.downsample.0.weight - torch.Size([256, 64, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.downsample.1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.downsample.1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.conv1.weight - torch.Size([64, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.bn1.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.bn1.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.conv2.weight - torch.Size([64, 64, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.bn2.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.bn2.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.conv3.weight - torch.Size([256, 64, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.bn3.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.bn3.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.conv1.weight - torch.Size([64, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.bn1.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.bn1.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.conv2.weight - torch.Size([64, 64, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.bn2.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.bn2.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.conv3.weight - torch.Size([256, 64, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.bn3.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.bn3.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.conv1.weight - torch.Size([128, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.bn1.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.bn1.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.conv2.weight - torch.Size([128, 128, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.bn2.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.bn2.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.conv3.weight - torch.Size([512, 128, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.bn3.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.bn3.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.downsample.0.weight - torch.Size([512, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.downsample.1.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.downsample.1.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.conv1.weight - torch.Size([128, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.bn1.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.bn1.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.conv2.weight - torch.Size([128, 128, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.bn2.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.bn2.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.conv3.weight - torch.Size([512, 128, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.bn3.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.bn3.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.conv1.weight - torch.Size([128, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.bn1.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.bn1.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.conv2.weight - torch.Size([128, 128, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.bn2.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.bn2.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.conv3.weight - torch.Size([512, 128, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.bn3.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.bn3.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.conv1.weight - torch.Size([128, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.bn1.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.bn1.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.conv2.weight - torch.Size([128, 128, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.bn2.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.bn2.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.conv3.weight - torch.Size([512, 128, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.bn3.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.bn3.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.conv1.weight - torch.Size([256, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.bn1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.bn1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.conv2.weight - torch.Size([256, 256, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.bn2.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.bn2.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.conv3.weight - torch.Size([1024, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.bn3.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.bn3.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.downsample.0.weight - torch.Size([1024, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.downsample.1.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.downsample.1.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.conv1.weight - torch.Size([256, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.bn1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.bn1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.conv2.weight - torch.Size([256, 256, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.bn2.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.bn2.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.conv3.weight - torch.Size([1024, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.bn3.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.bn3.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.conv1.weight - torch.Size([256, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.bn1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.bn1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.conv2.weight - torch.Size([256, 256, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.bn2.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.bn2.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.conv3.weight - torch.Size([1024, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.bn3.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.bn3.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.conv1.weight - torch.Size([256, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.bn1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.bn1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.conv2.weight - torch.Size([256, 256, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.bn2.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.bn2.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.conv3.weight - torch.Size([1024, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.bn3.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.bn3.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.conv1.weight - torch.Size([256, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.bn1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.bn1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.conv2.weight - torch.Size([256, 256, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.bn2.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.bn2.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.conv3.weight - torch.Size([1024, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.bn3.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.bn3.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.conv1.weight - torch.Size([256, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.bn1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.bn1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.conv2.weight - torch.Size([256, 256, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.bn2.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.bn2.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.conv3.weight - torch.Size([1024, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.bn3.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.bn3.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.conv1.weight - torch.Size([512, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.bn1.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.bn1.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.conv2.weight - torch.Size([512, 512, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.bn2.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.bn2.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.conv3.weight - torch.Size([2048, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.bn3.weight - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.bn3.bias - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.downsample.0.weight - torch.Size([2048, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.downsample.1.weight - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.downsample.1.bias - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.conv1.weight - torch.Size([512, 2048, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.bn1.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.bn1.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.conv2.weight - torch.Size([512, 512, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.bn2.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.bn2.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.conv3.weight - torch.Size([2048, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.bn3.weight - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.bn3.bias - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.conv1.weight - torch.Size([512, 2048, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.bn1.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.bn1.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.conv2.weight - torch.Size([512, 512, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.bn2.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.bn2.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.conv3.weight - torch.Size([2048, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.bn3.weight - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.bn3.bias - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +head.fc.weight - torch.Size([2, 2048]): +NormalInit: mean=0, std=0.01, bias=0 + +head.fc.bias - torch.Size([2]): +NormalInit: mean=0, std=0.01, bias=0 +2023/06/05 15:23:16 - mmengine - WARNING - "FileClient" will be deprecated in future. Please use io functions in https://mmengine.readthedocs.io/en/latest/api/fileio.html#file-io +2023/06/05 15:23:16 - mmengine - WARNING - "HardDiskBackend" is the alias of "LocalBackend" and the former will be deprecated in future. +2023/06/05 15:23:16 - mmengine - INFO - Checkpoints will be saved to /mnt/petrelfs/luzeyu/workspace/fakebench/mmpretrain/workdir/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1. +2023/06/05 15:25:17 - mmengine - INFO - Epoch(train) [1][100/342] lr: 9.9814e-05 eta: 1:06:59 time: 1.1890 data_time: 0.8950 memory: 9436 loss: 0.6825 +2023/06/05 15:27:13 - mmengine - INFO - Epoch(train) [1][200/342] lr: 9.9250e-05 eta: 1:03:40 time: 1.1525 data_time: 1.0062 memory: 6319 loss: 0.6692 +2023/06/05 15:29:12 - mmengine - INFO - Epoch(train) [1][300/342] lr: 9.8313e-05 eta: 1:01:45 time: 1.2061 data_time: 1.0674 memory: 6319 loss: 0.6599 +2023/06/05 15:30:11 - mmengine - INFO - Exp name: resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1_20230605_152245 +2023/06/05 15:30:11 - mmengine - INFO - Saving checkpoint at 1 epochs +2023/06/05 15:32:08 - mmengine - INFO - Epoch(val) [1][100/342] eta: 0:04:29 time: 0.3609 data_time: 0.2718 memory: 6319 +2023/06/05 15:33:49 - mmengine - INFO - Epoch(val) [1][200/342] eta: 0:02:30 time: 0.2470 data_time: 0.1592 memory: 3133 +2023/06/05 15:36:00 - mmengine - INFO - Epoch(val) [1][300/342] eta: 0:00:48 time: 1.3476 data_time: 1.2591 memory: 3133 +2023/06/05 15:38:44 - mmengine - INFO - Epoch(val) [1][342/342] accuracy/top1: 50.6496 data_time: 1.2936 time: 1.3836 +2023/06/05 15:40:58 - mmengine - INFO - Epoch(train) [2][100/342] lr: 9.6358e-05 eta: 1:01:38 time: 1.1650 data_time: 1.0241 memory: 6318 loss: 0.6411 +2023/06/05 15:43:01 - mmengine - INFO - Epoch(train) [2][200/342] lr: 9.4557e-05 eta: 0:59:27 time: 1.2714 data_time: 1.1312 memory: 6318 loss: 0.6296 +2023/06/05 15:45:00 - mmengine - INFO - Epoch(train) [2][300/342] lr: 9.2422e-05 eta: 0:57:03 time: 1.2035 data_time: 1.0602 memory: 6318 loss: 0.6186 +2023/06/05 15:45:56 - mmengine - INFO - Exp name: resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1_20230605_152245 +2023/06/05 15:45:56 - mmengine - INFO - Saving checkpoint at 2 epochs +2023/06/05 15:47:54 - mmengine - INFO - Epoch(val) [2][100/342] eta: 0:04:34 time: 0.4133 data_time: 0.3145 memory: 6318 +2023/06/05 15:49:37 - mmengine - INFO - Epoch(val) [2][200/342] eta: 0:02:33 time: 0.2325 data_time: 0.1457 memory: 3133 +2023/06/05 15:51:42 - mmengine - INFO - Epoch(val) [2][300/342] eta: 0:00:47 time: 1.2278 data_time: 1.1398 memory: 3133 +2023/06/05 15:55:51 - mmengine - INFO - Epoch(val) [2][342/342] accuracy/top1: 50.9959 data_time: 1.5341 time: 1.6239 +2023/06/05 15:57:52 - mmengine - INFO - Epoch(train) [3][100/342] lr: 8.8853e-05 eta: 0:54:13 time: 1.2978 data_time: 1.1576 memory: 6318 loss: 0.6150 +2023/06/05 15:59:48 - mmengine - INFO - Epoch(train) [3][200/342] lr: 8.5991e-05 eta: 0:51:47 time: 1.1297 data_time: 0.9878 memory: 6318 loss: 0.6050 +2023/06/05 16:01:45 - mmengine - INFO - Epoch(train) [3][300/342] lr: 8.2867e-05 eta: 0:49:31 time: 1.2984 data_time: 0.8756 memory: 6318 loss: 0.6017 +2023/06/05 16:02:08 - mmengine - INFO - Exp name: resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1_20230605_152245 +2023/06/05 16:02:38 - mmengine - INFO - Exp name: resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1_20230605_152245 +2023/06/05 16:02:38 - mmengine - INFO - Saving checkpoint at 3 epochs +2023/06/05 16:04:35 - mmengine - INFO - Epoch(val) [3][100/342] eta: 0:04:29 time: 0.3697 data_time: 0.2815 memory: 6318 +2023/06/05 16:06:15 - mmengine - INFO - Epoch(val) [3][200/342] eta: 0:02:30 time: 0.2354 data_time: 0.1489 memory: 3133 +2023/06/05 16:08:18 - mmengine - INFO - Epoch(val) [3][300/342] eta: 0:00:46 time: 1.1983 data_time: 1.1099 memory: 3133 +2023/06/05 16:10:56 - mmengine - INFO - Epoch(val) [3][342/342] accuracy/top1: 51.7210 data_time: 1.2516 time: 1.3406 +2023/06/05 16:12:54 - mmengine - INFO - Epoch(train) [4][100/342] lr: 7.8035e-05 eta: 0:46:33 time: 1.1460 data_time: 1.0065 memory: 6318 loss: 0.5950 +2023/06/05 16:14:51 - mmengine - INFO - Epoch(train) [4][200/342] lr: 7.4392e-05 eta: 0:44:22 time: 1.1317 data_time: 0.9614 memory: 6318 loss: 0.5726 +2023/06/05 16:16:49 - mmengine - INFO - Epoch(train) [4][300/342] lr: 7.0585e-05 eta: 0:42:15 time: 1.2142 data_time: 1.0538 memory: 6318 loss: 0.5772 +2023/06/05 16:17:37 - mmengine - INFO - Exp name: resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1_20230605_152245 +2023/06/05 16:17:37 - mmengine - INFO - Saving checkpoint at 4 epochs +2023/06/05 16:19:33 - mmengine - INFO - Epoch(val) [4][100/342] eta: 0:04:27 time: 0.3797 data_time: 0.2916 memory: 6318 +2023/06/05 16:21:11 - mmengine - INFO - Epoch(val) [4][200/342] eta: 0:02:27 time: 0.2393 data_time: 0.1530 memory: 3133 +2023/06/05 16:23:15 - mmengine - INFO - Epoch(val) [4][300/342] eta: 0:00:46 time: 1.2091 data_time: 1.1123 memory: 3133 +2023/06/05 16:25:55 - mmengine - INFO - Epoch(val) [4][342/342] accuracy/top1: 52.5778 data_time: 1.2498 time: 1.3390 +2023/06/05 16:27:54 - mmengine - INFO - Epoch(train) [5][100/342] lr: 6.4962e-05 eta: 0:39:17 time: 1.1165 data_time: 0.5891 memory: 6318 loss: 0.5707 +2023/06/05 16:29:53 - mmengine - INFO - Epoch(train) [5][200/342] lr: 6.0894e-05 eta: 0:37:14 time: 1.1354 data_time: 0.0009 memory: 6318 loss: 0.5622 +2023/06/05 16:31:52 - mmengine - INFO - Epoch(train) [5][300/342] lr: 5.6777e-05 eta: 0:35:11 time: 1.1498 data_time: 0.0009 memory: 6318 loss: 0.5554 +2023/06/05 16:32:39 - mmengine - INFO - Exp name: resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1_20230605_152245 +2023/06/05 16:32:39 - mmengine - INFO - Saving checkpoint at 5 epochs +2023/06/05 16:34:34 - mmengine - INFO - Epoch(val) [5][100/342] eta: 0:04:26 time: 0.3705 data_time: 0.2824 memory: 6318 +2023/06/05 16:36:15 - mmengine - INFO - Epoch(val) [5][200/342] eta: 0:02:29 time: 0.2497 data_time: 0.1628 memory: 3133 +2023/06/05 16:38:18 - mmengine - INFO - Epoch(val) [5][300/342] eta: 0:00:46 time: 1.2158 data_time: 1.1257 memory: 3133 +2023/06/05 16:40:58 - mmengine - INFO - Epoch(val) [5][342/342] accuracy/top1: 54.2697 data_time: 1.2538 time: 1.3430 +2023/06/05 16:42:59 - mmengine - INFO - Epoch(train) [6][100/342] lr: 5.0913e-05 eta: 0:32:17 time: 1.2449 data_time: 0.0008 memory: 6318 loss: 0.5514 +2023/06/05 16:45:03 - mmengine - INFO - Epoch(train) [6][200/342] lr: 4.6820e-05 eta: 0:30:19 time: 1.1295 data_time: 0.1760 memory: 6318 loss: 0.5461 +2023/06/05 16:46:45 - mmengine - INFO - Exp name: resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1_20230605_152245 +2023/06/05 16:46:56 - mmengine - INFO - Epoch(train) [6][300/342] lr: 4.2795e-05 eta: 0:28:14 time: 1.0794 data_time: 0.0480 memory: 6318 loss: 0.5393 +2023/06/05 16:47:53 - mmengine - INFO - Exp name: resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1_20230605_152245 +2023/06/05 16:47:53 - mmengine - INFO - Saving checkpoint at 6 epochs +2023/06/05 16:49:49 - mmengine - INFO - Epoch(val) [6][100/342] eta: 0:04:28 time: 0.3673 data_time: 0.2752 memory: 6318 +2023/06/05 16:51:28 - mmengine - INFO - Epoch(val) [6][200/342] eta: 0:02:28 time: 0.2362 data_time: 0.1498 memory: 3133 +2023/06/05 16:53:36 - mmengine - INFO - Epoch(val) [6][300/342] eta: 0:00:47 time: 1.2055 data_time: 1.1157 memory: 3133 +2023/06/05 16:56:18 - mmengine - INFO - Epoch(val) [6][342/342] accuracy/top1: 55.9684 data_time: 1.2709 time: 1.3599 +2023/06/05 16:58:43 - mmengine - INFO - Epoch(train) [7][100/342] lr: 3.7265e-05 eta: 0:25:42 time: 1.3398 data_time: 0.3791 memory: 6318 loss: 0.5337 +2023/06/05 17:00:40 - mmengine - INFO - Epoch(train) [7][200/342] lr: 3.3546e-05 eta: 0:23:37 time: 1.2089 data_time: 0.0008 memory: 6318 loss: 0.5213 +2023/06/05 17:02:40 - mmengine - INFO - Epoch(train) [7][300/342] lr: 3.0008e-05 eta: 0:21:35 time: 1.1896 data_time: 0.0010 memory: 6318 loss: 0.5286 +2023/06/05 17:03:34 - mmengine - INFO - Exp name: resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1_20230605_152245 +2023/06/05 17:03:34 - mmengine - INFO - Saving checkpoint at 7 epochs +2023/06/05 17:05:31 - mmengine - INFO - Epoch(val) [7][100/342] eta: 0:04:28 time: 0.3842 data_time: 0.2954 memory: 6318 +2023/06/05 17:07:13 - mmengine - INFO - Epoch(val) [7][200/342] eta: 0:02:31 time: 0.2442 data_time: 0.1443 memory: 3133 +2023/06/05 17:09:22 - mmengine - INFO - Epoch(val) [7][300/342] eta: 0:00:47 time: 1.2277 data_time: 1.1381 memory: 3133 +2023/06/05 17:12:14 - mmengine - INFO - Epoch(val) [7][342/342] accuracy/top1: 57.9018 data_time: 1.3131 time: 1.4027 +2023/06/05 17:14:16 - mmengine - INFO - Epoch(train) [8][100/342] lr: 2.5353e-05 eta: 0:18:44 time: 1.2066 data_time: 0.2822 memory: 6318 loss: 0.5239 +2023/06/05 17:16:18 - mmengine - INFO - Epoch(train) [8][200/342] lr: 2.2372e-05 eta: 0:16:43 time: 1.1655 data_time: 0.0008 memory: 6318 loss: 0.5149 +2023/06/05 17:18:20 - mmengine - INFO - Epoch(train) [8][300/342] lr: 1.9667e-05 eta: 0:14:42 time: 1.1396 data_time: 0.0008 memory: 6318 loss: 0.5259 +2023/06/05 17:19:16 - mmengine - INFO - Exp name: resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1_20230605_152245 +2023/06/05 17:19:16 - mmengine - INFO - Saving checkpoint at 8 epochs +2023/06/05 17:21:12 - mmengine - INFO - Epoch(val) [8][100/342] eta: 0:04:27 time: 0.3493 data_time: 0.2612 memory: 6318 +2023/06/05 17:22:51 - mmengine - INFO - Epoch(val) [8][200/342] eta: 0:02:28 time: 0.2387 data_time: 0.1470 memory: 3133 +2023/06/05 17:25:00 - mmengine - INFO - Epoch(val) [8][300/342] eta: 0:00:47 time: 1.2511 data_time: 1.1614 memory: 3133 +2023/06/05 17:27:42 - mmengine - INFO - Epoch(val) [8][342/342] accuracy/top1: 59.5919 data_time: 1.2731 time: 1.3627 +2023/06/05 17:29:46 - mmengine - INFO - Epoch(train) [9][100/342] lr: 1.6343e-05 eta: 0:11:51 time: 1.4554 data_time: 0.7423 memory: 6318 loss: 0.5241 +2023/06/05 17:31:48 - mmengine - INFO - Epoch(train) [9][200/342] lr: 1.4393e-05 eta: 0:09:49 time: 1.1975 data_time: 0.3691 memory: 6318 loss: 0.5109 +2023/06/05 17:33:06 - mmengine - INFO - Exp name: resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1_20230605_152245 +2023/06/05 17:33:46 - mmengine - INFO - Epoch(train) [9][300/342] lr: 1.2785e-05 eta: 0:07:47 time: 1.2325 data_time: 0.0252 memory: 6318 loss: 0.5198 +2023/06/05 17:34:32 - mmengine - INFO - Exp name: resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1_20230605_152245 +2023/06/05 17:34:32 - mmengine - INFO - Saving checkpoint at 9 epochs +2023/06/05 17:36:26 - mmengine - INFO - Epoch(val) [9][100/342] eta: 0:04:21 time: 0.3462 data_time: 0.2529 memory: 6318 +2023/06/05 17:37:53 - mmengine - INFO - Epoch(val) [9][200/342] eta: 0:02:19 time: 0.2109 data_time: 0.1242 memory: 3133 +2023/06/05 17:39:58 - mmengine - INFO - Epoch(val) [9][300/342] eta: 0:00:44 time: 1.1651 data_time: 1.0761 memory: 3133 +2023/06/05 17:43:00 - mmengine - INFO - Epoch(val) [9][342/342] accuracy/top1: 60.9770 data_time: 1.2203 time: 1.3100 +2023/06/05 17:45:07 - mmengine - INFO - Epoch(train) [10][100/342] lr: 1.1116e-05 eta: 0:04:54 time: 1.1144 data_time: 0.4077 memory: 6318 loss: 0.5053 +2023/06/05 17:47:05 - mmengine - INFO - Epoch(train) [10][200/342] lr: 1.0388e-05 eta: 0:02:52 time: 1.1674 data_time: 0.0009 memory: 6318 loss: 0.5073 +2023/06/05 17:49:02 - mmengine - INFO - Epoch(train) [10][300/342] lr: 1.0035e-05 eta: 0:00:51 time: 1.1998 data_time: 0.0009 memory: 6318 loss: 0.5161 +2023/06/05 17:49:59 - mmengine - INFO - Exp name: resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1_20230605_152245 +2023/06/05 17:49:59 - mmengine - INFO - Saving checkpoint at 10 epochs +2023/06/05 17:51:49 - mmengine - INFO - Epoch(val) [10][100/342] eta: 0:04:12 time: 0.3383 data_time: 0.2504 memory: 6318 +2023/06/05 17:53:23 - mmengine - INFO - Epoch(val) [10][200/342] eta: 0:02:20 time: 0.2193 data_time: 0.1319 memory: 3133 +2023/06/05 17:55:29 - mmengine - INFO - Epoch(val) [10][300/342] eta: 0:00:45 time: 1.1584 data_time: 1.0694 memory: 3133 +2023/06/05 17:58:36 - mmengine - INFO - Epoch(val) [10][342/342] accuracy/top1: 62.3369 data_time: 1.2060 time: 1.2960 diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/20230605_152245.json b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/20230605_152245.json new file mode 100644 index 0000000000000000000000000000000000000000..6155e401866ee38fb9012909c63fbc9f953d56c5 --- /dev/null +++ b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/20230605_152245.json @@ -0,0 +1,40 @@ +{"lr": 9.981404789023513e-05, "data_time": 0.8950233459472656, "loss": 0.6824808180332184, "time": 1.1890023708343507, "epoch": 1, "memory": 9436, "step": 100} +{"lr": 9.925023268502683e-05, "data_time": 1.0062472581863404, "loss": 0.66917724609375, "time": 1.1525463342666626, "epoch": 1, "memory": 6319, "step": 200} +{"lr": 9.831328961451971e-05, "data_time": 1.067365860939026, "loss": 0.6599426209926605, "time": 1.206129217147827, "epoch": 1, "memory": 6319, "step": 300} +{"accuracy/top1": 50.64961242675781, "data_time": 1.2936465328896951, "time": 1.38361062083328, "step": 1} +{"lr": 9.635783271319408e-05, "data_time": 1.0240814924240111, "loss": 0.6410930991172791, "time": 1.164955997467041, "epoch": 2, "memory": 6318, "step": 442} +{"lr": 9.455662799327337e-05, "data_time": 1.1311914443969726, "loss": 0.6296148359775543, "time": 1.2714224338531495, "epoch": 2, "memory": 6318, "step": 542} +{"lr": 9.242187293325357e-05, "data_time": 1.0601815700531005, "loss": 0.6186129808425903, "time": 1.2034532070159911, "epoch": 2, "memory": 6318, "step": 642} +{"accuracy/top1": 50.995880126953125, "data_time": 1.5341123637930645, "time": 1.623861736180831, "step": 2} +{"lr": 8.885322471322101e-05, "data_time": 1.157559084892273, "loss": 0.6149831771850586, "time": 1.2978172302246094, "epoch": 3, "memory": 6318, "step": 784} +{"lr": 8.59909449463049e-05, "data_time": 0.987814474105835, "loss": 0.6049974083900451, "time": 1.1296810150146483, "epoch": 3, "memory": 6318, "step": 884} +{"lr": 8.28673425957603e-05, "data_time": 0.875587272644043, "loss": 0.6016679406166077, "time": 1.2983975172042848, "epoch": 3, "memory": 6318, "step": 984} +{"accuracy/top1": 51.72104263305664, "data_time": 1.2516064463134409, "time": 1.340595806305332, "step": 3} +{"lr": 7.803482720903195e-05, "data_time": 1.0065465927124024, "loss": 0.5950481772422791, "time": 1.1460498809814452, "epoch": 4, "memory": 6318, "step": 1126} +{"lr": 7.439165228138202e-05, "data_time": 0.9614474296569824, "loss": 0.5725913643836975, "time": 1.1317323207855225, "epoch": 4, "memory": 6318, "step": 1226} +{"lr": 7.058496260180133e-05, "data_time": 1.0538496017456054, "loss": 0.5771694421768189, "time": 1.214171290397644, "epoch": 4, "memory": 6318, "step": 1326} +{"accuracy/top1": 52.577842712402344, "data_time": 1.2498030342791588, "time": 1.3390403148384205, "step": 4} +{"lr": 6.496162032454504e-05, "data_time": 0.5890519618988037, "loss": 0.5706973016262055, "time": 1.1165104389190674, "epoch": 5, "memory": 6318, "step": 1468} +{"lr": 6.089416958157208e-05, "data_time": 0.0009218692779541015, "loss": 0.5621558129787445, "time": 1.1353834390640258, "epoch": 5, "memory": 6318, "step": 1568} +{"lr": 5.677701788155902e-05, "data_time": 0.0008935928344726562, "loss": 0.5553753674030304, "time": 1.1497874021530152, "epoch": 5, "memory": 6318, "step": 1668} +{"accuracy/top1": 54.269691467285156, "data_time": 1.2537985028747916, "time": 1.3430342264147612, "step": 5} +{"lr": 5.091330063600123e-05, "data_time": 0.0008427858352661133, "loss": 0.5514010399580002, "time": 1.2448705434799194, "epoch": 6, "memory": 6318, "step": 1810} +{"lr": 4.681972449602271e-05, "data_time": 0.17596094608306884, "loss": 0.5461299479007721, "time": 1.1294851064682008, "epoch": 6, "memory": 6318, "step": 1910} +{"lr": 4.27951262698573e-05, "data_time": 0.047959637641906736, "loss": 0.5393110156059265, "time": 1.0794288396835328, "epoch": 6, "memory": 6318, "step": 2010} +{"accuracy/top1": 55.96841049194336, "data_time": 1.2708633140641816, "time": 1.3598824075637683, "step": 6} +{"lr": 3.726501555491419e-05, "data_time": 0.37908642292022704, "loss": 0.533661812543869, "time": 1.3398460626602173, "epoch": 7, "memory": 6318, "step": 2152} +{"lr": 3.354602177213381e-05, "data_time": 0.0007644176483154296, "loss": 0.5213438749313355, "time": 1.208915376663208, "epoch": 7, "memory": 6318, "step": 2252} +{"lr": 3.0007932735217966e-05, "data_time": 0.000972747802734375, "loss": 0.5285512804985046, "time": 1.1895843505859376, "epoch": 7, "memory": 6318, "step": 2352} +{"accuracy/top1": 57.90178680419922, "data_time": 1.3131155474192895, "time": 1.4026674828098398, "step": 7} +{"lr": 2.5352754318215113e-05, "data_time": 0.28219668865203856, "loss": 0.5238977313041687, "time": 1.2065652370452882, "epoch": 8, "memory": 6318, "step": 2494} +{"lr": 2.2372383915844294e-05, "data_time": 0.0008071660995483398, "loss": 0.5148650944232941, "time": 1.1654731035232544, "epoch": 8, "memory": 6318, "step": 2594} +{"lr": 1.9667136874426842e-05, "data_time": 0.0008148193359375, "loss": 0.5259059607982636, "time": 1.1395622730255126, "epoch": 8, "memory": 6318, "step": 2694} +{"accuracy/top1": 59.591922760009766, "data_time": 1.2730592485767065, "time": 1.3626738053369105, "step": 8} +{"lr": 1.6342572053355464e-05, "data_time": 0.7423277378082276, "loss": 0.5241424888372421, "time": 1.4553632974624633, "epoch": 9, "memory": 6318, "step": 2836} +{"lr": 1.4392564451840171e-05, "data_time": 0.3691322088241577, "loss": 0.5108548015356064, "time": 1.1974761486053467, "epoch": 9, "memory": 6318, "step": 2936} +{"lr": 1.2784967834899793e-05, "data_time": 0.02521207332611084, "loss": 0.5198489099740982, "time": 1.2324589014053344, "epoch": 9, "memory": 6318, "step": 3036} +{"accuracy/top1": 60.976993560791016, "data_time": 1.2203224629771954, "time": 1.3099683610065338, "step": 9} +{"lr": 1.1116448178051759e-05, "data_time": 0.40768578052520754, "loss": 0.5053461819887162, "time": 1.1143648862838744, "epoch": 10, "memory": 6318, "step": 3178} +{"lr": 1.0387683707930187e-05, "data_time": 0.0008753776550292969, "loss": 0.5073316335678101, "time": 1.1674257516860962, "epoch": 10, "memory": 6318, "step": 3278} +{"lr": 1.0035100273117234e-05, "data_time": 0.0009264707565307617, "loss": 0.5160931169986724, "time": 1.1998209714889527, "epoch": 10, "memory": 6318, "step": 3378} +{"accuracy/top1": 62.336883544921875, "data_time": 1.2060417058516522, "time": 1.295999117564877, "step": 10} diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/config.py b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/config.py new file mode 100644 index 0000000000000000000000000000000000000000..78ef787cd2fb0c90720cc58721897600a5b7dd79 --- /dev/null +++ b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/config.py @@ -0,0 +1,230 @@ +optim_wrapper = dict( + optimizer=dict( + type='SGD', + lr=0.0001, + momentum=0.9, + weight_decay=0.0001, + _scope_='mmpretrain'), + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=False, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=512) +model = dict( + type='ImageClassifier', + backbone=dict( + frozen_stages=2, + type='ResNet', + depth=50, + num_stages=4, + out_indices=(3, ), + style='pytorch', + init_cfg=dict( + type='Pretrained', + checkpoint= + 'https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth', + prefix='backbone')), + neck=dict(type='GlobalAveragePooling'), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=2048, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + topk=1)) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3fake8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3real8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3fake8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3real8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = dict(type='Accuracy', topk=1) +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3fake8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3real8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = dict(type='Accuracy', topk=1) +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1' diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/events.out.tfevents.1685949779.SH-IDC1-10-140-24-103.232700.0 b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/events.out.tfevents.1685949779.SH-IDC1-10-140-24-103.232700.0 new file mode 100644 index 0000000000000000000000000000000000000000..e5b6f661f8f73abd0d91ba44e73bae61fdad9e69 --- /dev/null +++ b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/events.out.tfevents.1685949779.SH-IDC1-10-140-24-103.232700.0 @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:2af665c22ca54f5706fd8f10a948447770f29de6db95cd26cb1dcddbfab2ae83 +size 14575804 diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/scalars.json b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/scalars.json new file mode 100644 index 0000000000000000000000000000000000000000..6155e401866ee38fb9012909c63fbc9f953d56c5 --- /dev/null +++ b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/scalars.json @@ -0,0 +1,40 @@ +{"lr": 9.981404789023513e-05, "data_time": 0.8950233459472656, "loss": 0.6824808180332184, "time": 1.1890023708343507, "epoch": 1, "memory": 9436, "step": 100} +{"lr": 9.925023268502683e-05, "data_time": 1.0062472581863404, "loss": 0.66917724609375, "time": 1.1525463342666626, "epoch": 1, "memory": 6319, "step": 200} +{"lr": 9.831328961451971e-05, "data_time": 1.067365860939026, "loss": 0.6599426209926605, "time": 1.206129217147827, "epoch": 1, "memory": 6319, "step": 300} +{"accuracy/top1": 50.64961242675781, "data_time": 1.2936465328896951, "time": 1.38361062083328, "step": 1} +{"lr": 9.635783271319408e-05, "data_time": 1.0240814924240111, "loss": 0.6410930991172791, "time": 1.164955997467041, "epoch": 2, "memory": 6318, "step": 442} +{"lr": 9.455662799327337e-05, "data_time": 1.1311914443969726, "loss": 0.6296148359775543, "time": 1.2714224338531495, "epoch": 2, "memory": 6318, "step": 542} +{"lr": 9.242187293325357e-05, "data_time": 1.0601815700531005, "loss": 0.6186129808425903, "time": 1.2034532070159911, "epoch": 2, "memory": 6318, "step": 642} +{"accuracy/top1": 50.995880126953125, "data_time": 1.5341123637930645, "time": 1.623861736180831, "step": 2} +{"lr": 8.885322471322101e-05, "data_time": 1.157559084892273, "loss": 0.6149831771850586, "time": 1.2978172302246094, "epoch": 3, "memory": 6318, "step": 784} +{"lr": 8.59909449463049e-05, "data_time": 0.987814474105835, "loss": 0.6049974083900451, "time": 1.1296810150146483, "epoch": 3, "memory": 6318, "step": 884} +{"lr": 8.28673425957603e-05, "data_time": 0.875587272644043, "loss": 0.6016679406166077, "time": 1.2983975172042848, "epoch": 3, "memory": 6318, "step": 984} +{"accuracy/top1": 51.72104263305664, "data_time": 1.2516064463134409, "time": 1.340595806305332, "step": 3} +{"lr": 7.803482720903195e-05, "data_time": 1.0065465927124024, "loss": 0.5950481772422791, "time": 1.1460498809814452, "epoch": 4, "memory": 6318, "step": 1126} +{"lr": 7.439165228138202e-05, "data_time": 0.9614474296569824, "loss": 0.5725913643836975, "time": 1.1317323207855225, "epoch": 4, "memory": 6318, "step": 1226} +{"lr": 7.058496260180133e-05, "data_time": 1.0538496017456054, "loss": 0.5771694421768189, "time": 1.214171290397644, "epoch": 4, "memory": 6318, "step": 1326} +{"accuracy/top1": 52.577842712402344, "data_time": 1.2498030342791588, "time": 1.3390403148384205, "step": 4} +{"lr": 6.496162032454504e-05, "data_time": 0.5890519618988037, "loss": 0.5706973016262055, "time": 1.1165104389190674, "epoch": 5, "memory": 6318, "step": 1468} +{"lr": 6.089416958157208e-05, "data_time": 0.0009218692779541015, "loss": 0.5621558129787445, "time": 1.1353834390640258, "epoch": 5, "memory": 6318, "step": 1568} +{"lr": 5.677701788155902e-05, "data_time": 0.0008935928344726562, "loss": 0.5553753674030304, "time": 1.1497874021530152, "epoch": 5, "memory": 6318, "step": 1668} +{"accuracy/top1": 54.269691467285156, "data_time": 1.2537985028747916, "time": 1.3430342264147612, "step": 5} +{"lr": 5.091330063600123e-05, "data_time": 0.0008427858352661133, "loss": 0.5514010399580002, "time": 1.2448705434799194, "epoch": 6, "memory": 6318, "step": 1810} +{"lr": 4.681972449602271e-05, "data_time": 0.17596094608306884, "loss": 0.5461299479007721, "time": 1.1294851064682008, "epoch": 6, "memory": 6318, "step": 1910} +{"lr": 4.27951262698573e-05, "data_time": 0.047959637641906736, "loss": 0.5393110156059265, "time": 1.0794288396835328, "epoch": 6, "memory": 6318, "step": 2010} +{"accuracy/top1": 55.96841049194336, "data_time": 1.2708633140641816, "time": 1.3598824075637683, "step": 6} +{"lr": 3.726501555491419e-05, "data_time": 0.37908642292022704, "loss": 0.533661812543869, "time": 1.3398460626602173, "epoch": 7, "memory": 6318, "step": 2152} +{"lr": 3.354602177213381e-05, "data_time": 0.0007644176483154296, "loss": 0.5213438749313355, "time": 1.208915376663208, "epoch": 7, "memory": 6318, "step": 2252} +{"lr": 3.0007932735217966e-05, "data_time": 0.000972747802734375, "loss": 0.5285512804985046, "time": 1.1895843505859376, "epoch": 7, "memory": 6318, "step": 2352} +{"accuracy/top1": 57.90178680419922, "data_time": 1.3131155474192895, "time": 1.4026674828098398, "step": 7} +{"lr": 2.5352754318215113e-05, "data_time": 0.28219668865203856, "loss": 0.5238977313041687, "time": 1.2065652370452882, "epoch": 8, "memory": 6318, "step": 2494} +{"lr": 2.2372383915844294e-05, "data_time": 0.0008071660995483398, "loss": 0.5148650944232941, "time": 1.1654731035232544, "epoch": 8, "memory": 6318, "step": 2594} +{"lr": 1.9667136874426842e-05, "data_time": 0.0008148193359375, "loss": 0.5259059607982636, "time": 1.1395622730255126, "epoch": 8, "memory": 6318, "step": 2694} +{"accuracy/top1": 59.591922760009766, "data_time": 1.2730592485767065, "time": 1.3626738053369105, "step": 8} +{"lr": 1.6342572053355464e-05, "data_time": 0.7423277378082276, "loss": 0.5241424888372421, "time": 1.4553632974624633, "epoch": 9, "memory": 6318, "step": 2836} +{"lr": 1.4392564451840171e-05, "data_time": 0.3691322088241577, "loss": 0.5108548015356064, "time": 1.1974761486053467, "epoch": 9, "memory": 6318, "step": 2936} +{"lr": 1.2784967834899793e-05, "data_time": 0.02521207332611084, "loss": 0.5198489099740982, "time": 1.2324589014053344, "epoch": 9, "memory": 6318, "step": 3036} +{"accuracy/top1": 60.976993560791016, "data_time": 1.2203224629771954, "time": 1.3099683610065338, "step": 9} +{"lr": 1.1116448178051759e-05, "data_time": 0.40768578052520754, "loss": 0.5053461819887162, "time": 1.1143648862838744, "epoch": 10, "memory": 6318, "step": 3178} +{"lr": 1.0387683707930187e-05, "data_time": 0.0008753776550292969, "loss": 0.5073316335678101, "time": 1.1674257516860962, "epoch": 10, "memory": 6318, "step": 3278} +{"lr": 1.0035100273117234e-05, "data_time": 0.0009264707565307617, "loss": 0.5160931169986724, "time": 1.1998209714889527, "epoch": 10, "memory": 6318, "step": 3378} +{"accuracy/top1": 62.336883544921875, "data_time": 1.2060417058516522, "time": 1.295999117564877, "step": 10} diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/35624.png_1.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/35624.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..e1e6585859d5bde0382e2817cc8cda8a790eff4d Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/35624.png_1.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/35624.png_10.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/35624.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..a0a9259bf78683344180bd502bf7dee769e4b329 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/35624.png_10.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/35624.png_2.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/35624.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..d41b92086ad70bd1bdfc59e68d875737c98081ee Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/35624.png_2.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/35624.png_3.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/35624.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..aca95e85df344894680cd9caac666523bdd27add Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/35624.png_3.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/35624.png_4.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/35624.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..a4b1dfed80b69c4e6728d99d3fd70c3e61072acd Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/35624.png_4.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/35624.png_5.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/35624.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..24c2bacc1ad13961fa7233251f53a82796e25bb9 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/35624.png_5.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/35624.png_6.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/35624.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..5b0592e9f721ef514756dee524ba5ea9d408f263 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/35624.png_6.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/35624.png_7.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/35624.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..68b3312dc7515b91b252cbaf27e8d095bc19f85d Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/35624.png_7.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/35624.png_8.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/35624.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..b570d1c1a1dcaa54dc07f82ac1fa1d3eb0757dd4 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/35624.png_8.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/35624.png_9.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/35624.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..8bfcf4c6baf8ba30fa966f89f6d15ee9c3fa229d Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/35624.png_9.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/38442.png_1.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/38442.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..73dce409a3d03fb5786865fa6aca8dc101023fbd Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/38442.png_1.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/38442.png_10.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/38442.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..6afc867115cea73568058ab044d67a5824186bde Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/38442.png_10.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/38442.png_2.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/38442.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..1e2c2b6ea646ecc4d8067ba61b947bc9c66f4ede Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/38442.png_2.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/38442.png_3.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/38442.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..4cb2f2a0ebc4e086a97da3ef83931da0890bb8d6 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/38442.png_3.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/38442.png_4.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/38442.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..d23f2fbb0c9ba2441e736d6a2aa85d3c53a3db90 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/38442.png_4.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/38442.png_5.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/38442.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..c9a1a046502f195aceac98e64c00aec3edb0280b Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/38442.png_5.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/38442.png_6.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/38442.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..00d608b6d10fa267355a2da5745a9dadf26bb9b3 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/38442.png_6.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/38442.png_7.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/38442.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..5b2a7637e7410babcc523fa9a7a36baa5b4271d2 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/38442.png_7.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/38442.png_8.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/38442.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..719d4b63c448a6ecbe2d5c58c2ed6abe5bfc0b8a Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/38442.png_8.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/38442.png_9.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/38442.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..69ad1ae0f4ef5c5d39640d5053ea3b6e2390522e Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/38442.png_9.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/43283.png_1.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/43283.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..fa530ed7ee24877c6abba6d33eb6b685f6f6d005 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/43283.png_1.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/43283.png_10.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/43283.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..c03a8032397bd07a26f5861df9bf85c75baf59ef Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/43283.png_10.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/43283.png_2.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/43283.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..3b3f562758ee53bb0dfc10c706d874d681da3b40 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/43283.png_2.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/43283.png_3.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/43283.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..4c8e29cff12883ba4a5ae6e249f7a854c9628dde Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/43283.png_3.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/43283.png_4.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/43283.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..502216d2b839841c778f80bdcd5757d1f3d54c17 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/43283.png_4.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/43283.png_5.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/43283.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..b7e84ceb41ce9651baf3f6d0617ca874f0d68cf9 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/43283.png_5.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/43283.png_6.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/43283.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..88b2506f4e5f31667c6872001583e6c63d8355bc Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/43283.png_6.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/43283.png_7.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/43283.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..6ae6fcdf4bcacf074df74c8d7a7d988bfe2f8601 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/43283.png_7.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/43283.png_8.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/43283.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..210d259f0de703405ba490e74175533116f58e7c Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/43283.png_8.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/43283.png_9.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/43283.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..3ff4f62181159861eba0fe94de82ef241c0f1583 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/43283.png_9.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/44982.png_1.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/44982.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..a1785ba702d19a929589f5abb7bee278774d5af8 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/44982.png_1.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/44982.png_10.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/44982.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..d83d8bf43b6a48529e158d76aa1ce9bc13dd1514 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/44982.png_10.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/44982.png_2.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/44982.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..019bfc930513690f63a78663e21c5b757600689e Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/44982.png_2.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/44982.png_3.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/44982.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..634c6ae98d2ac908228bc2fb5f95272f1dc6cfd8 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/44982.png_3.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/44982.png_4.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/44982.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..96509c1a6031cd6f4c79dfecc25c0754bc635d0f Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/44982.png_4.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/44982.png_5.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/44982.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..e89ce5bb2b35ee2efa4cc9d3422221cc77710b20 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/44982.png_5.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/44982.png_6.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/44982.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..ebbed0d52cd82c1233a7046079253a6083f2cff7 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/44982.png_6.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/44982.png_7.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/44982.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..2ea4dbb49380e5a286b271b8fba9d533a4221a4b Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/44982.png_7.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/44982.png_8.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/44982.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..b1e6b59641a116073bfd94ef60575c22f92fd5a3 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/44982.png_8.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/44982.png_9.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/44982.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..9baf1338c63d62e0247ce49dac2d669b6475dc9b Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/44982.png_9.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/56423.png_1.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/56423.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..0c501b80a03f945acf85d796efe0a30ca6ed54f1 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/56423.png_1.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/56423.png_10.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/56423.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..ebe5582f07504fe01997539c6ef464df842f5b67 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/56423.png_10.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/56423.png_2.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/56423.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..2d9b598f5b7a0a70c26c6b04a2582828b4c667ab Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/56423.png_2.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/56423.png_3.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/56423.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..84cc970d098f2c4b50a8ed07ee312f77bb3c5ec8 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/56423.png_3.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/56423.png_4.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/56423.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..6d782120b36ba05bd41cb711ed15126242a22b52 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/56423.png_4.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/56423.png_5.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/56423.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..21ba103f68b97d6bc5306af3ed9e414f974dc140 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/56423.png_5.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/56423.png_6.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/56423.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..3010823ddb104d8060c6c2b537ba61d7fd6ec94b Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/56423.png_6.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/56423.png_7.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/56423.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..7055dd5c1c5350735814fdfa5a26aa20286afa59 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/56423.png_7.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/56423.png_8.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/56423.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..4e938e3cda2c5e08fa9a947914d65d37460b49d0 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/56423.png_8.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/56423.png_9.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/56423.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..966b2e0652a58be4ff21b4361e8b137e9851c59b Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/56423.png_9.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/61024.png_1.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/61024.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..fbb1e97b6e11da4e9ac23f6061ae4271ff1e9f2f Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/61024.png_1.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/61024.png_10.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/61024.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..8134847e9823a075a7b7bd8db9509cfdc4945423 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/61024.png_10.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/61024.png_2.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/61024.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..3e7af242fed943383540c8fa129f232e62c4d55e Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/61024.png_2.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/61024.png_3.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/61024.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..311793b01ce22e208f4b0c1102d56c8490d20602 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/61024.png_3.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/61024.png_4.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/61024.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..d3932e7f15170bab6cadc70df0adfd5b59fb8493 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/61024.png_4.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/61024.png_5.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/61024.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..b75649f34f2ebc6970093e5a5a63cdca127052b5 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/61024.png_5.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/61024.png_6.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/61024.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..0af4358cc22af553bdbe12755a37f010c3685f8e Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/61024.png_6.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/61024.png_7.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/61024.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..0968eb53984d2be63b859e7f6dedee3af2687e4e Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/61024.png_7.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/61024.png_8.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/61024.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..90617224df326b74afe934ef58712c84fbadb14e Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/61024.png_8.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/61024.png_9.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/61024.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..52ba7003f114a40c38888078f48b11965b90dd24 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/61024.png_9.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/66390.png_1.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/66390.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..a4b8056758aa75e083ce7f1007a5e94abcbefc15 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/66390.png_1.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/66390.png_10.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/66390.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..d818df0602f45ee8cd96a80b074ccbaaf12098f0 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/66390.png_10.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/66390.png_2.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/66390.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..4ccd86da120c9068d8f77ccdda96f9c5271ead48 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/66390.png_2.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/66390.png_3.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/66390.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..fa46cb321f9f59cf92027d4ca428fc0fb1fec571 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/66390.png_3.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/66390.png_4.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/66390.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..c0d898b33e8fcc1e353483f3fb7d74b7faf15459 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/66390.png_4.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/66390.png_5.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/66390.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..caa81296b62dd9cb7d7345f3afd3dd9e5de46ea0 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/66390.png_5.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/66390.png_6.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/66390.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..2defc957010b508c5fae805a18c6b5f654e4cc19 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/66390.png_6.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/66390.png_7.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/66390.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..afe5bccecf80beda4cc3086690c6eff618165f9a Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/66390.png_7.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/66390.png_8.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/66390.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..54bd7205885337bbf34b8255c04ff322872c6439 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/66390.png_8.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/66390.png_9.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/66390.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..b29fba4aa10e82c5b5022dfa66d927f43f4d2949 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/66390.png_9.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/flickr_wild_000179.jpg_1.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/flickr_wild_000179.jpg_1.png new file mode 100644 index 0000000000000000000000000000000000000000..364b441013f6f898b5a0541c3f4ffa1d099657ca Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/flickr_wild_000179.jpg_1.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/flickr_wild_000179.jpg_10.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/flickr_wild_000179.jpg_10.png new file mode 100644 index 0000000000000000000000000000000000000000..08be685a60ffe0dff3eb05bf0230c3ec63fb6b10 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/flickr_wild_000179.jpg_10.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/flickr_wild_000179.jpg_2.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/flickr_wild_000179.jpg_2.png new file mode 100644 index 0000000000000000000000000000000000000000..77d850d8725606bfd66ee64fb56963d62f4f1aa4 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/flickr_wild_000179.jpg_2.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/flickr_wild_000179.jpg_3.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/flickr_wild_000179.jpg_3.png new file mode 100644 index 0000000000000000000000000000000000000000..e60f56ee067015a99d52d9fbdd6c649015bd2403 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/flickr_wild_000179.jpg_3.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/flickr_wild_000179.jpg_4.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/flickr_wild_000179.jpg_4.png new file mode 100644 index 0000000000000000000000000000000000000000..72626305f1b057398b8c20f3da87c792619d71e9 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/flickr_wild_000179.jpg_4.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/flickr_wild_000179.jpg_5.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/flickr_wild_000179.jpg_5.png new file mode 100644 index 0000000000000000000000000000000000000000..c5556fb508fd87f95d769b240d65ac87a95d676b Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/flickr_wild_000179.jpg_5.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/flickr_wild_000179.jpg_6.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/flickr_wild_000179.jpg_6.png new file mode 100644 index 0000000000000000000000000000000000000000..a27e63141583bfe4c62b04ab2917940df20af734 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/flickr_wild_000179.jpg_6.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/flickr_wild_000179.jpg_7.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/flickr_wild_000179.jpg_7.png new file mode 100644 index 0000000000000000000000000000000000000000..9f8e52f214341ed80c51226f2051a110d24c7a6b Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/flickr_wild_000179.jpg_7.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/flickr_wild_000179.jpg_8.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/flickr_wild_000179.jpg_8.png new file mode 100644 index 0000000000000000000000000000000000000000..92cf4a18b061b3973e16ecc74d91a3cc49b605dc Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/flickr_wild_000179.jpg_8.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/flickr_wild_000179.jpg_9.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/flickr_wild_000179.jpg_9.png new file mode 100644 index 0000000000000000000000000000000000000000..9c1b93274fd81ac74c8dd5fcbf4175e2c3312617 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/flickr_wild_000179.jpg_9.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/pixabay_dog_002352.jpg_1.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/pixabay_dog_002352.jpg_1.png new file mode 100644 index 0000000000000000000000000000000000000000..af4deb92a243af95f9894df314f8a7ceca30851a Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/pixabay_dog_002352.jpg_1.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/pixabay_dog_002352.jpg_10.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/pixabay_dog_002352.jpg_10.png new file mode 100644 index 0000000000000000000000000000000000000000..3522c844f48acc6917520dab26c2fd1a7b3471fd Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/pixabay_dog_002352.jpg_10.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/pixabay_dog_002352.jpg_2.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/pixabay_dog_002352.jpg_2.png new file mode 100644 index 0000000000000000000000000000000000000000..45acfcdf4596bfbc76c4b982e768be61aedb244a Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/pixabay_dog_002352.jpg_2.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/pixabay_dog_002352.jpg_3.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/pixabay_dog_002352.jpg_3.png new file mode 100644 index 0000000000000000000000000000000000000000..6361d65dc434940a9d9937915aa623f27be29135 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/pixabay_dog_002352.jpg_3.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/pixabay_dog_002352.jpg_4.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/pixabay_dog_002352.jpg_4.png new file mode 100644 index 0000000000000000000000000000000000000000..16486087d971fe757f0c491836d1af6cf41876f9 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/pixabay_dog_002352.jpg_4.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/pixabay_dog_002352.jpg_5.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/pixabay_dog_002352.jpg_5.png new file mode 100644 index 0000000000000000000000000000000000000000..d7868f653c652589bf165fef096fad04bc0305cf Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/pixabay_dog_002352.jpg_5.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/pixabay_dog_002352.jpg_6.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/pixabay_dog_002352.jpg_6.png new file mode 100644 index 0000000000000000000000000000000000000000..4df0e17a2bf48643106e3ef5ec30342b58c5bc1e Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/pixabay_dog_002352.jpg_6.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/pixabay_dog_002352.jpg_7.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/pixabay_dog_002352.jpg_7.png new file mode 100644 index 0000000000000000000000000000000000000000..3d6e1ae9dd432f4bd2fca6ababd307856e58d2d9 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/pixabay_dog_002352.jpg_7.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/pixabay_dog_002352.jpg_8.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/pixabay_dog_002352.jpg_8.png new file mode 100644 index 0000000000000000000000000000000000000000..2f44e60b71d5170f99690920441d4bf88ed5e892 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/pixabay_dog_002352.jpg_8.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/pixabay_dog_002352.jpg_9.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/pixabay_dog_002352.jpg_9.png new file mode 100644 index 0000000000000000000000000000000000000000..d150eda6834270fff8a26651da801b30ed1505c6 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/pixabay_dog_002352.jpg_9.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed14088.png_1.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed14088.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..a41cbf53ee526f93a8675acb90a9e2cd7bbcfd4d Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed14088.png_1.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed14088.png_10.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed14088.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..21692031dbfe4696409881040878c75c533a47c8 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed14088.png_10.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed14088.png_2.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed14088.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..e31d01e26c1ab5c33b4169855ebfaea079fc06c7 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed14088.png_2.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed14088.png_3.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed14088.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..fa6d7462114e6ffc917273e715a4adf53cfbb8f3 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed14088.png_3.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed14088.png_4.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed14088.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..cda1833d7f0aaac02286463d1f95edb4ef57190a Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed14088.png_4.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed14088.png_5.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed14088.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..2c439ba87c6dab4e7d3726fdcad3015cd3472df6 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed14088.png_5.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed14088.png_6.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed14088.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..fe380e3267f465bdc33e3299b5c003b5ab3cbae1 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed14088.png_6.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed14088.png_7.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed14088.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..41f84298dc48c91255d9fc807093ce280f9a8212 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed14088.png_7.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed14088.png_8.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed14088.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..332d5a4ff0b63a5f085b93c7640db3cec8da09bc Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed14088.png_8.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed14088.png_9.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed14088.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..77a8188c1ad01e7b1679f156e67da5a5328f3861 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed14088.png_9.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed16124.png_1.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed16124.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..c78877df29247984f6115dbd41d1294985734890 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed16124.png_1.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed16124.png_10.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed16124.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..db5777746096db15350f88acc7f70a9940355707 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed16124.png_10.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed16124.png_2.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed16124.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..837ac023e97750c42bd70de0d42a570e7c8037e5 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed16124.png_2.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed16124.png_3.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed16124.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..3d73d57557038da858f052d409eee716da34f46e Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed16124.png_3.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed16124.png_4.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed16124.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..e81a2b537e2df6c4e266c148f4c4c9a4bfcd0158 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed16124.png_4.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed16124.png_5.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed16124.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..67c0acc2436f84453959987593c1ee16ffde94d8 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed16124.png_5.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed16124.png_6.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed16124.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..5a41c248f61bb7dfca43ffc5ad5cd37c6820968b Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed16124.png_6.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed16124.png_7.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed16124.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..bf4b2aafa50effea96e65533a48c786c14077e3f Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed16124.png_7.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed16124.png_8.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed16124.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..bdeedd7814618fb08424fc7fc067b519ac5cd4f2 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed16124.png_8.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed16124.png_9.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed16124.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..b821e1f58b5dc04af37a9d507a2b316ad7490b98 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed16124.png_9.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed17282.png_1.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed17282.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..ed3d4dbd0de67bd3b7e8c0d9720c76ed3c971145 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed17282.png_1.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed17282.png_10.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed17282.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..11885c3be5f96b09d002253f4ac594ec8bd2d506 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed17282.png_10.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed17282.png_2.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed17282.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..5d84407dd15f3c7be7a7cedd5c5fe3f85a4232b8 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed17282.png_2.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed17282.png_3.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed17282.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..3043f8346d835a6503c5566185650cf87df58aad Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed17282.png_3.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed17282.png_4.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed17282.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..7368eacb604aeabd31e4bfde800636a9a05a3046 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed17282.png_4.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed17282.png_5.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed17282.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..3bbd188d098b0b9ddab56295a5300389deb656dc Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed17282.png_5.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed17282.png_6.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed17282.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..fcae28161eb038c96c1c72c518d567617c40188c Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed17282.png_6.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed17282.png_7.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed17282.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..a2d7f8e643e24acb63da103848764d88a729e25d Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed17282.png_7.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed17282.png_8.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed17282.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..52f48582d3c89a1fc74d7b44c5516c075ba175a1 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed17282.png_8.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed17282.png_9.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed17282.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..f465b37e253e4af4f06a8c9e09806e38b8a2f3ab Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed17282.png_9.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed21815.png_1.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed21815.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..2bd70ab97a31c4833e67f05cca629b96a9437113 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed21815.png_1.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed21815.png_10.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed21815.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..cd494c7ff1a12fd595f078db62b53bf85f0e7da2 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed21815.png_10.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed21815.png_2.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed21815.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..f9248d3191fa63f441136d9c4f112d5bde190513 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed21815.png_2.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed21815.png_3.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed21815.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..72ce496b350c71b0b440a29713a8b356ef4d704e Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed21815.png_3.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed21815.png_4.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed21815.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..c99f4346172a59ae6f189c98acbf67ae66e0636b Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed21815.png_4.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed21815.png_5.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed21815.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..3bb56202bca7c51f7b3cfd82eaa9704c71df90f9 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed21815.png_5.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed21815.png_6.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed21815.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..201502e7c6a5bd662cc4c13f071294132870defd Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed21815.png_6.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed21815.png_7.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed21815.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..56e4401ecc301798f47e0108adf6ddb0d5f8fb2b Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed21815.png_7.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed21815.png_8.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed21815.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..96f48c490b52e8acefb0c6890b3f59e7b4b5354f Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed21815.png_8.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed21815.png_9.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed21815.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..bc27cead90ab94e97be1b2084d114e9075664008 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed21815.png_9.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed22624.png_1.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed22624.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..84afd310a471596f3caf0d0fb439aadd3e9b50f0 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed22624.png_1.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed22624.png_10.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed22624.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..7991d79407577be1771814e9c6c78357b0f4be00 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed22624.png_10.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed22624.png_2.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed22624.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..c51ee6c791884006a13990c04aa1e89c80066635 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed22624.png_2.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed22624.png_3.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed22624.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..984ed0393079818e4624113081344fe30cb6e403 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed22624.png_3.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed22624.png_4.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed22624.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..ca09853649d3e17f82c07182ed8ead00e24c12cb Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed22624.png_4.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed22624.png_5.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed22624.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..72b7da96c7961fc9ba29ad414743a4df8e938b8e Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed22624.png_5.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed22624.png_6.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed22624.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..a36253e966dfd07fecf27a11043bd4084be94c4b Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed22624.png_6.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed22624.png_7.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed22624.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..25d4b325352ad7bededc8a90eede2eef394d4cdc Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed22624.png_7.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed22624.png_8.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed22624.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..3484c22475c79b162f2685c947135c8444b5c3fa Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed22624.png_8.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed22624.png_9.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed22624.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..cade1814d7c68e2519f18bbc978d3c93ede087f0 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed22624.png_9.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed23602.png_1.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed23602.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..eb89de2bb6aa33467794d15190f031db051ebb38 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed23602.png_1.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed23602.png_10.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed23602.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..41583be2dc19f3d2078e63c2aecbcd2d972d4c9a Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed23602.png_10.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed23602.png_2.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed23602.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..16cac509907da63a99e64ded7a5ef9ad7cc0a8c5 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed23602.png_2.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed23602.png_3.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed23602.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..593036947937d882e45749a378d75bef8c54a988 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed23602.png_3.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed23602.png_4.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed23602.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..f71f55fddb7d66aaaa7cabdb9e6c083c0f73bd4a Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed23602.png_4.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed23602.png_5.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed23602.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..081b17b9c756b34e6fbcf430b2abb5f0290de567 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed23602.png_5.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed23602.png_6.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed23602.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..b6b4c7a71d99ca71a7070e9db6c638d134bce6f7 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed23602.png_6.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed23602.png_7.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed23602.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..d9ba62928aab622a4d7e0b155739443e37e06430 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed23602.png_7.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed23602.png_8.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed23602.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..10d2c034be3791e0de7189632f3722b1ef34ad4c Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed23602.png_8.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed23602.png_9.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed23602.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..87fdd552d7f02f758664cc4f02d605846f9b02ad Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed23602.png_9.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed24366.png_1.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed24366.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..0735101911c068d380b1b88c45ba1aa0cb716056 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed24366.png_1.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed24366.png_10.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed24366.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..756daf80f923939bea3e41abd930a5a95bfaf929 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed24366.png_10.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed24366.png_2.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed24366.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..5e7923d1d711c016612874dee9bfdf87a052e7b7 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed24366.png_2.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed24366.png_3.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed24366.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..2687475f67437523dc0c0c57d64731f3b9b39f09 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed24366.png_3.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed24366.png_4.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed24366.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..f6dbf6059f70b0414f40fea3c9d206975caa94a7 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed24366.png_4.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed24366.png_5.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed24366.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..52c41d49e044cdd107f2195f41fced56660bc05d Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed24366.png_5.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed24366.png_6.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed24366.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..7c79af71c36e773f7990b2301ed4abb4d6a6e753 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed24366.png_6.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed24366.png_7.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed24366.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..943973b3eb6683ecf1a33803b95f516496c834fe Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed24366.png_7.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed24366.png_8.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed24366.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..b59952acdd31ee2d636c392d0f385f8f281f8efa Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed24366.png_8.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed24366.png_9.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed24366.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..ff2f9cd8f152f6e65120a6f94db260ac57e5be7b Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed24366.png_9.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed26848.png_1.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed26848.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..61163ba7314cca8832849a6a7fca250d364b4498 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed26848.png_1.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed26848.png_10.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed26848.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..7563ba03d87302607923d67a14b283b21153e873 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed26848.png_10.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed26848.png_2.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed26848.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..794c3fc846eac79a40a17c816b024ee763ae4154 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed26848.png_2.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed26848.png_3.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed26848.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..ed1a2dca223dc6400498945ec200d8e6557798f9 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed26848.png_3.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed26848.png_4.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed26848.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..27f57b9abb09529deb5294a5c01e104971943ac0 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed26848.png_4.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed26848.png_5.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed26848.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..0bc339a9e6277b04bbb1771fca2f4c9936f2d57a Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed26848.png_5.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed26848.png_6.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed26848.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..b5b2967d00ffcd0990dd04e78dfcd2b84b0b8c8a Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed26848.png_6.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed26848.png_7.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed26848.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..0969ce2f4b62878e3d6ef1bc85ff9d2288ac0e18 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed26848.png_7.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed26848.png_8.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed26848.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..1cfad3dd456d5cb76841d3fd8bf8e1776e846a20 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed26848.png_8.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed26848.png_9.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed26848.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..c82d1754c65cdb494f252e2b635dc76aed0aace3 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed26848.png_9.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed44086.png_1.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed44086.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..8aa69e247617fd8753885261dc8c4dddb8a8bd75 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed44086.png_1.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed44086.png_10.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed44086.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..f7f85eff48d000baddde9011bdb7ebcb8161825e Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed44086.png_10.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed44086.png_2.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed44086.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..5f4b5e7f44dca38be0bcae50838d2a82ad026dc8 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed44086.png_2.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed44086.png_3.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed44086.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..7221e51083067d3e4ec759691ae4e9d9fdeda355 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed44086.png_3.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed44086.png_4.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed44086.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..9a4b8ace76458f6b8c3e5e7fc8c407d53c96097a Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed44086.png_4.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed44086.png_5.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed44086.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..618e87d6c26e3c07ced437b689d04a4bf8891139 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed44086.png_5.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed44086.png_6.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed44086.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..1dc55b67fe33d4f647e2245e0e438415012abce3 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed44086.png_6.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed44086.png_7.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed44086.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..b820da0b2c4a568d6cd17c8129a8653800539bf2 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed44086.png_7.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed44086.png_8.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed44086.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..8d1b765edd464d1c27ca861a5e8250d3f479dcb7 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed44086.png_8.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed44086.png_9.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed44086.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..3734f436c3f051d8c71f5934c3e783610011b4ae Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/20230605_152245/vis_data/vis_image/seed44086.png_9.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/epoch_1.pth b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/epoch_1.pth new file mode 100644 index 0000000000000000000000000000000000000000..e215dc802eba9ce9e0070dfed973690a137df689 --- /dev/null +++ b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/epoch_1.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c149288c4e88a54d60ae906f6129e74dea4067498a07229f96acfd9b1ded6633 +size 277074125 diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/epoch_10.pth b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/epoch_10.pth new file mode 100644 index 0000000000000000000000000000000000000000..cfcf72cda1bc066dd99dae61a239f997f9373bf0 --- /dev/null +++ b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/epoch_10.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:acc3f942114c0cd46d458f47d9eb2cdef903a9ff3c9f9d8b65ff2c61cf5b58a9 +size 277405517 diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/epoch_2.pth b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/epoch_2.pth new file mode 100644 index 0000000000000000000000000000000000000000..bf69a9e16150b19fd2c20618242bfc9c94cf603a --- /dev/null +++ b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/epoch_2.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:536a024806bd522a95b6f71bb69ab1a7ba67b8a1959258b394569adfdc75e1c2 +size 277111693 diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/epoch_3.pth b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/epoch_3.pth new file mode 100644 index 0000000000000000000000000000000000000000..1a7c78f95e4a3230e8c1499485ff154f4ea2bc1f --- /dev/null +++ b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/epoch_3.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:f05e907765842438399012a2225da3482371fbab613905b1f8477241c27182d0 +size 277148365 diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/epoch_4.pth b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/epoch_4.pth new file mode 100644 index 0000000000000000000000000000000000000000..14a14216596428c37d2e60dbc823ad617234f289 --- /dev/null +++ b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/epoch_4.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:eab9e061cb40f977a18a0eee7c62cc1acb290fb907d7ad305535c8234ef42019 +size 277185037 diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/epoch_5.pth b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/epoch_5.pth new file mode 100644 index 0000000000000000000000000000000000000000..97fb0cc0cac80a2e5e54890d4e7e06296ebd9409 --- /dev/null +++ b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/epoch_5.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:9da185b97967ccc21ee7e9cc33301a7e8ef0c03fd64651d6202c4e5fc4ddd508 +size 277221517 diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/epoch_6.pth b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/epoch_6.pth new file mode 100644 index 0000000000000000000000000000000000000000..f8cd755e444ff32f18a1060f90f45ae15566da8c --- /dev/null +++ b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/epoch_6.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:2c0a51dc5fd3dedbd7e39438b9f12d152f784987ab84579f955957543a033f72 +size 277258125 diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/epoch_7.pth b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/epoch_7.pth new file mode 100644 index 0000000000000000000000000000000000000000..c90d9766f174666052422348dea1d5c7b4d50a90 --- /dev/null +++ b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/epoch_7.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c95a80fbfffcbe5037f195a49333180293628058fc70449dbf5c73c373542a0c +size 277294733 diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/epoch_8.pth b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/epoch_8.pth new file mode 100644 index 0000000000000000000000000000000000000000..19291ec0b84541f698f6858c4425edf77cd3da21 --- /dev/null +++ b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/epoch_8.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:0f2aa945d24ff7221d1f24c583cf7635921f3a4527683b70b955e61fbca12c7f +size 277331597 diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/epoch_9.pth b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/epoch_9.pth new file mode 100644 index 0000000000000000000000000000000000000000..fad1fdbbf9123de2b7b925fc717090df2c484f6a --- /dev/null +++ b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/epoch_9.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:f09180f1c4178e3dac3e76ec2d061a182c410eb91b9ac9da2083f6ff5bbb4596 +size 277368525 diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/last_checkpoint b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/last_checkpoint new file mode 100644 index 0000000000000000000000000000000000000000..a8e30c9042e9d60ada2514e53beed00b4b55e84e --- /dev/null +++ b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/last_checkpoint @@ -0,0 +1 @@ +/mnt/petrelfs/luzeyu/workspace/fakebench/mmpretrain/workdir/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/epoch_10.pth \ No newline at end of file diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1.py b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1.py new file mode 100644 index 0000000000000000000000000000000000000000..78ef787cd2fb0c90720cc58721897600a5b7dd79 --- /dev/null +++ b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1.py @@ -0,0 +1,230 @@ +optim_wrapper = dict( + optimizer=dict( + type='SGD', + lr=0.0001, + momentum=0.9, + weight_decay=0.0001, + _scope_='mmpretrain'), + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=False, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=512) +model = dict( + type='ImageClassifier', + backbone=dict( + frozen_stages=2, + type='ResNet', + depth=50, + num_stages=4, + out_indices=(3, ), + style='pytorch', + init_cfg=dict( + type='Pretrained', + checkpoint= + 'https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth', + prefix='backbone')), + neck=dict(type='GlobalAveragePooling'), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=2048, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + topk=1)) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3fake8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3real8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3fake8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3real8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = dict(type='Accuracy', topk=1) +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3fake8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3real8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = dict(type='Accuracy', topk=1) +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_1e-1' diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/20230605_153011.log b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/20230605_153011.log new file mode 100644 index 0000000000000000000000000000000000000000..b3751bfb8b20a3570aed8854c373a0b32c2407c3 --- /dev/null +++ b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/20230605_153011.log @@ -0,0 +1,939 @@ +2023/06/05 15:30:15 - mmengine - INFO - +------------------------------------------------------------ +System environment: + sys.platform: linux + Python: 3.10.9 (main, Mar 8 2023, 10:47:38) [GCC 11.2.0] + CUDA available: True + numpy_random_seed: 1945260712 + GPU 0,1: NVIDIA A100-SXM4-80GB + CUDA_HOME: /mnt/petrelfs/share/cuda-11.6 + NVCC: Cuda compilation tools, release 11.6, V11.6.124 + GCC: gcc (GCC) 7.5.0 + PyTorch: 1.13.1 + PyTorch compiling details: PyTorch built with: + - GCC 9.3 + - C++ Version: 201402 + - Intel(R) oneAPI Math Kernel Library Version 2021.4-Product Build 20210904 for Intel(R) 64 architecture applications + - Intel(R) MKL-DNN v2.6.0 (Git Hash 52b5f107dd9cf10910aaa19cb47f3abf9b349815) + - OpenMP 201511 (a.k.a. OpenMP 4.5) + - LAPACK is enabled (usually provided by MKL) + - NNPACK is enabled + - CPU capability usage: AVX2 + - CUDA Runtime 11.6 + - NVCC architecture flags: -gencode;arch=compute_37,code=sm_37;-gencode;arch=compute_50,code=sm_50;-gencode;arch=compute_60,code=sm_60;-gencode;arch=compute_61,code=sm_61;-gencode;arch=compute_70,code=sm_70;-gencode;arch=compute_75,code=sm_75;-gencode;arch=compute_80,code=sm_80;-gencode;arch=compute_86,code=sm_86;-gencode;arch=compute_37,code=compute_37 + - CuDNN 8.3.2 (built against CUDA 11.5) + - Magma 2.6.1 + - Build settings: BLAS_INFO=mkl, BUILD_TYPE=Release, CUDA_VERSION=11.6, CUDNN_VERSION=8.3.2, CXX_COMPILER=/opt/rh/devtoolset-9/root/usr/bin/c++, CXX_FLAGS= -fabi-version=11 -Wno-deprecated -fvisibility-inlines-hidden -DUSE_PTHREADPOOL -fopenmp -DNDEBUG -DUSE_KINETO -DUSE_FBGEMM -DUSE_QNNPACK -DUSE_PYTORCH_QNNPACK -DUSE_XNNPACK -DSYMBOLICATE_MOBILE_DEBUG_HANDLE -DEDGE_PROFILER_USE_KINETO -O2 -fPIC -Wno-narrowing -Wall -Wextra -Werror=return-type -Werror=non-virtual-dtor -Wno-missing-field-initializers -Wno-type-limits -Wno-array-bounds -Wno-unknown-pragmas -Wunused-local-typedefs -Wno-unused-parameter -Wno-unused-function -Wno-unused-result -Wno-strict-overflow -Wno-strict-aliasing -Wno-error=deprecated-declarations -Wno-stringop-overflow -Wno-psabi -Wno-error=pedantic -Wno-error=redundant-decls -Wno-error=old-style-cast -fdiagnostics-color=always -faligned-new -Wno-unused-but-set-variable -Wno-maybe-uninitialized -fno-math-errno -fno-trapping-math -Werror=format -Werror=cast-function-type -Wno-stringop-overflow, LAPACK_INFO=mkl, PERF_WITH_AVX=1, PERF_WITH_AVX2=1, PERF_WITH_AVX512=1, TORCH_VERSION=1.13.1, USE_CUDA=ON, USE_CUDNN=ON, USE_EXCEPTION_PTR=1, USE_GFLAGS=OFF, USE_GLOG=OFF, USE_MKL=ON, USE_MKLDNN=ON, USE_MPI=OFF, USE_NCCL=ON, USE_NNPACK=ON, USE_OPENMP=ON, USE_ROCM=OFF, + + TorchVision: 0.14.1 + OpenCV: 4.7.0 + MMEngine: 0.7.3 + +Runtime environment: + cudnn_benchmark: True + mp_cfg: {'mp_start_method': 'fork', 'opencv_num_threads': 0} + dist_cfg: {'backend': 'nccl'} + seed: None + deterministic: False + Distributed launcher: slurm + Distributed training: True + GPU number: 2 +------------------------------------------------------------ + +2023/06/05 15:30:19 - mmengine - INFO - Config: +optim_wrapper = dict( + optimizer=dict( + type='SGD', + lr=0.0001, + momentum=0.9, + weight_decay=0.0001, + _scope_='mmpretrain'), + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=False, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=512) +model = dict( + type='ImageClassifier', + backbone=dict( + frozen_stages=2, + type='ResNet', + depth=50, + num_stages=4, + out_indices=(3, ), + style='pytorch', + init_cfg=dict( + type='Pretrained', + checkpoint= + 'https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth', + prefix='backbone')), + neck=dict(type='GlobalAveragePooling'), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=2048, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + topk=1)) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3fake8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3real8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3fake8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3real8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = dict(type='Accuracy', topk=1) +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3fake8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3real8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = dict(type='Accuracy', topk=1) +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1' + +2023/06/05 15:30:30 - mmengine - INFO - Hooks will be executed in the following order: +before_run: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_load_checkpoint: +(ABOVE_NORMAL) EMAHook + -------------------- +before_train: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_train_epoch: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(NORMAL ) DistSamplerSeedHook + -------------------- +before_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook + -------------------- +after_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +after_train_epoch: +(NORMAL ) IterTimerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_val_epoch: +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook + -------------------- +before_val_iter: +(NORMAL ) IterTimerHook + -------------------- +after_val_iter: +(NORMAL ) IterTimerHook +(NORMAL ) VisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_val_epoch: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_save_checkpoint: +(ABOVE_NORMAL) EMAHook + -------------------- +after_train: +(VERY_LOW ) CheckpointHook + -------------------- +before_test_epoch: +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook + -------------------- +before_test_iter: +(NORMAL ) IterTimerHook + -------------------- +after_test_iter: +(NORMAL ) IterTimerHook +(NORMAL ) VisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_test_epoch: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_run: +(BELOW_NORMAL) LoggerHook + -------------------- +2023/06/05 15:30:41 - mmengine - INFO - load backbone in model from: https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth +Name of parameter - Initialization information + +backbone.conv1.weight - torch.Size([64, 3, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.bn1.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.bn1.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.conv1.weight - torch.Size([64, 64, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.bn1.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.bn1.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.conv2.weight - torch.Size([64, 64, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.bn2.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.bn2.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.conv3.weight - torch.Size([256, 64, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.bn3.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.bn3.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.downsample.0.weight - torch.Size([256, 64, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.downsample.1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.downsample.1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.conv1.weight - torch.Size([64, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.bn1.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.bn1.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.conv2.weight - torch.Size([64, 64, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.bn2.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.bn2.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.conv3.weight - torch.Size([256, 64, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.bn3.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.bn3.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.conv1.weight - torch.Size([64, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.bn1.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.bn1.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.conv2.weight - torch.Size([64, 64, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.bn2.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.bn2.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.conv3.weight - torch.Size([256, 64, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.bn3.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.bn3.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.conv1.weight - torch.Size([128, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.bn1.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.bn1.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.conv2.weight - torch.Size([128, 128, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.bn2.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.bn2.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.conv3.weight - torch.Size([512, 128, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.bn3.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.bn3.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.downsample.0.weight - torch.Size([512, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.downsample.1.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.downsample.1.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.conv1.weight - torch.Size([128, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.bn1.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.bn1.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.conv2.weight - torch.Size([128, 128, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.bn2.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.bn2.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.conv3.weight - torch.Size([512, 128, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.bn3.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.bn3.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.conv1.weight - torch.Size([128, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.bn1.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.bn1.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.conv2.weight - torch.Size([128, 128, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.bn2.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.bn2.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.conv3.weight - torch.Size([512, 128, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.bn3.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.bn3.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.conv1.weight - torch.Size([128, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.bn1.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.bn1.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.conv2.weight - torch.Size([128, 128, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.bn2.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.bn2.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.conv3.weight - torch.Size([512, 128, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.bn3.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.bn3.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.conv1.weight - torch.Size([256, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.bn1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.bn1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.conv2.weight - torch.Size([256, 256, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.bn2.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.bn2.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.conv3.weight - torch.Size([1024, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.bn3.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.bn3.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.downsample.0.weight - torch.Size([1024, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.downsample.1.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.downsample.1.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.conv1.weight - torch.Size([256, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.bn1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.bn1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.conv2.weight - torch.Size([256, 256, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.bn2.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.bn2.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.conv3.weight - torch.Size([1024, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.bn3.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.bn3.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.conv1.weight - torch.Size([256, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.bn1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.bn1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.conv2.weight - torch.Size([256, 256, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.bn2.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.bn2.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.conv3.weight - torch.Size([1024, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.bn3.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.bn3.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.conv1.weight - torch.Size([256, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.bn1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.bn1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.conv2.weight - torch.Size([256, 256, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.bn2.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.bn2.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.conv3.weight - torch.Size([1024, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.bn3.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.bn3.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.conv1.weight - torch.Size([256, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.bn1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.bn1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.conv2.weight - torch.Size([256, 256, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.bn2.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.bn2.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.conv3.weight - torch.Size([1024, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.bn3.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.bn3.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.conv1.weight - torch.Size([256, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.bn1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.bn1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.conv2.weight - torch.Size([256, 256, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.bn2.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.bn2.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.conv3.weight - torch.Size([1024, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.bn3.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.bn3.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.conv1.weight - torch.Size([512, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.bn1.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.bn1.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.conv2.weight - torch.Size([512, 512, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.bn2.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.bn2.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.conv3.weight - torch.Size([2048, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.bn3.weight - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.bn3.bias - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.downsample.0.weight - torch.Size([2048, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.downsample.1.weight - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.downsample.1.bias - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.conv1.weight - torch.Size([512, 2048, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.bn1.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.bn1.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.conv2.weight - torch.Size([512, 512, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.bn2.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.bn2.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.conv3.weight - torch.Size([2048, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.bn3.weight - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.bn3.bias - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.conv1.weight - torch.Size([512, 2048, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.bn1.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.bn1.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.conv2.weight - torch.Size([512, 512, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.bn2.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.bn2.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.conv3.weight - torch.Size([2048, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.bn3.weight - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.bn3.bias - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +head.fc.weight - torch.Size([2, 2048]): +NormalInit: mean=0, std=0.01, bias=0 + +head.fc.bias - torch.Size([2]): +NormalInit: mean=0, std=0.01, bias=0 +2023/06/05 15:30:41 - mmengine - WARNING - "FileClient" will be deprecated in future. Please use io functions in https://mmengine.readthedocs.io/en/latest/api/fileio.html#file-io +2023/06/05 15:30:41 - mmengine - WARNING - "HardDiskBackend" is the alias of "LocalBackend" and the former will be deprecated in future. +2023/06/05 15:30:41 - mmengine - INFO - Checkpoints will be saved to /mnt/petrelfs/luzeyu/workspace/fakebench/mmpretrain/workdir/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1. +2023/06/05 15:32:40 - mmengine - INFO - Epoch(train) [1][100/342] lr: 9.9814e-05 eta: 1:06:00 time: 1.2049 data_time: 0.8983 memory: 9436 loss: 0.6787 +2023/06/05 15:34:35 - mmengine - INFO - Epoch(train) [1][200/342] lr: 9.9250e-05 eta: 1:02:47 time: 1.1238 data_time: 0.8004 memory: 6319 loss: 0.6656 +2023/06/05 15:36:30 - mmengine - INFO - Epoch(train) [1][300/342] lr: 9.8313e-05 eta: 1:00:31 time: 1.1811 data_time: 0.6303 memory: 6319 loss: 0.6573 +2023/06/05 15:37:19 - mmengine - INFO - Exp name: resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1_20230605_153011 +2023/06/05 15:37:19 - mmengine - INFO - Saving checkpoint at 1 epochs +2023/06/05 15:39:09 - mmengine - INFO - Epoch(val) [1][100/342] eta: 0:04:11 time: 0.3784 data_time: 0.2897 memory: 6319 +2023/06/05 15:40:44 - mmengine - INFO - Epoch(val) [1][200/342] eta: 0:02:21 time: 0.2651 data_time: 0.1774 memory: 3133 +2023/06/05 15:42:53 - mmengine - INFO - Epoch(val) [1][300/342] eta: 0:00:45 time: 1.2960 data_time: 1.2066 memory: 3133 +2023/06/05 15:45:27 - mmengine - INFO - Epoch(val) [1][342/342] accuracy/top1: 50.2106 data_time: 1.2180 time: 1.3094 +2023/06/05 15:47:26 - mmengine - INFO - Epoch(train) [2][100/342] lr: 9.6358e-05 eta: 0:58:09 time: 1.2105 data_time: 1.0696 memory: 6319 loss: 0.6448 +2023/06/05 15:49:24 - mmengine - INFO - Epoch(train) [2][200/342] lr: 9.4557e-05 eta: 0:56:12 time: 1.1774 data_time: 1.0370 memory: 6319 loss: 0.6401 +2023/06/05 15:51:24 - mmengine - INFO - Epoch(train) [2][300/342] lr: 9.2422e-05 eta: 0:54:28 time: 1.0953 data_time: 0.9550 memory: 6319 loss: 0.6284 +2023/06/05 15:52:09 - mmengine - INFO - Exp name: resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1_20230605_153011 +2023/06/05 15:52:09 - mmengine - INFO - Saving checkpoint at 2 epochs +2023/06/05 15:53:59 - mmengine - INFO - Epoch(val) [2][100/342] eta: 0:04:12 time: 0.3998 data_time: 0.3109 memory: 6319 +2023/06/05 15:55:37 - mmengine - INFO - Epoch(val) [2][200/342] eta: 0:02:23 time: 0.2780 data_time: 0.1895 memory: 3133 +2023/06/05 15:57:51 - mmengine - INFO - Epoch(val) [2][300/342] eta: 0:00:47 time: 1.2933 data_time: 1.2040 memory: 3133 +2023/06/05 16:00:21 - mmengine - INFO - Epoch(val) [2][342/342] accuracy/top1: 50.7217 data_time: 1.2331 time: 1.3235 +2023/06/05 16:02:27 - mmengine - INFO - Epoch(train) [3][100/342] lr: 8.8853e-05 eta: 0:51:54 time: 1.2071 data_time: 1.0630 memory: 6319 loss: 0.6251 +2023/06/05 16:04:26 - mmengine - INFO - Epoch(train) [3][200/342] lr: 8.5991e-05 eta: 0:49:58 time: 1.1982 data_time: 1.0558 memory: 6319 loss: 0.6139 +2023/06/05 16:06:24 - mmengine - INFO - Epoch(train) [3][300/342] lr: 8.2867e-05 eta: 0:48:00 time: 1.2471 data_time: 1.1042 memory: 6319 loss: 0.6115 +2023/06/05 16:06:44 - mmengine - INFO - Exp name: resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1_20230605_153011 +2023/06/05 16:07:11 - mmengine - INFO - Exp name: resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1_20230605_153011 +2023/06/05 16:07:11 - mmengine - INFO - Saving checkpoint at 3 epochs +2023/06/05 16:09:01 - mmengine - INFO - Epoch(val) [3][100/342] eta: 0:04:14 time: 0.3745 data_time: 0.2866 memory: 6319 +2023/06/05 16:10:40 - mmengine - INFO - Epoch(val) [3][200/342] eta: 0:02:24 time: 0.2844 data_time: 0.1969 memory: 3133 +2023/06/05 16:12:52 - mmengine - INFO - Epoch(val) [3][300/342] eta: 0:00:47 time: 1.2410 data_time: 1.1524 memory: 3133 +2023/06/05 16:15:19 - mmengine - INFO - Epoch(val) [3][342/342] accuracy/top1: 51.3696 data_time: 1.2191 time: 1.3094 +2023/06/05 16:17:21 - mmengine - INFO - Epoch(train) [4][100/342] lr: 7.8035e-05 eta: 0:45:14 time: 1.1143 data_time: 0.9709 memory: 6319 loss: 0.6028 +2023/06/05 16:19:19 - mmengine - INFO - Epoch(train) [4][200/342] lr: 7.4392e-05 eta: 0:43:15 time: 1.2439 data_time: 1.1045 memory: 6319 loss: 0.5947 +2023/06/05 16:21:18 - mmengine - INFO - Epoch(train) [4][300/342] lr: 7.0585e-05 eta: 0:41:17 time: 1.1688 data_time: 1.0215 memory: 6319 loss: 0.5964 +2023/06/05 16:22:07 - mmengine - INFO - Exp name: resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1_20230605_153011 +2023/06/05 16:22:07 - mmengine - INFO - Saving checkpoint at 4 epochs +2023/06/05 16:24:02 - mmengine - INFO - Epoch(val) [4][100/342] eta: 0:04:24 time: 0.3786 data_time: 0.2913 memory: 6319 +2023/06/05 16:25:43 - mmengine - INFO - Epoch(val) [4][200/342] eta: 0:02:29 time: 0.2911 data_time: 0.2023 memory: 3133 +2023/06/05 16:27:56 - mmengine - INFO - Epoch(val) [4][300/342] eta: 0:00:48 time: 1.2596 data_time: 1.1519 memory: 3133 +2023/06/05 16:30:22 - mmengine - INFO - Epoch(val) [4][342/342] accuracy/top1: 51.6867 data_time: 1.2391 time: 1.3299 +2023/06/05 16:32:24 - mmengine - INFO - Epoch(train) [5][100/342] lr: 6.4962e-05 eta: 0:38:34 time: 1.2703 data_time: 0.9684 memory: 6319 loss: 0.5867 +2023/06/05 16:34:23 - mmengine - INFO - Epoch(train) [5][200/342] lr: 6.0894e-05 eta: 0:36:36 time: 1.1602 data_time: 1.0207 memory: 6319 loss: 0.5730 +2023/06/05 16:36:23 - mmengine - INFO - Epoch(train) [5][300/342] lr: 5.6777e-05 eta: 0:34:38 time: 1.2297 data_time: 1.0876 memory: 6319 loss: 0.5703 +2023/06/05 16:37:14 - mmengine - INFO - Exp name: resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1_20230605_153011 +2023/06/05 16:37:14 - mmengine - INFO - Saving checkpoint at 5 epochs +2023/06/05 16:39:09 - mmengine - INFO - Epoch(val) [5][100/342] eta: 0:04:25 time: 0.3783 data_time: 0.2895 memory: 6319 +2023/06/05 16:40:49 - mmengine - INFO - Epoch(val) [5][200/342] eta: 0:02:29 time: 0.2879 data_time: 0.1992 memory: 3133 +2023/06/05 16:43:03 - mmengine - INFO - Epoch(val) [5][300/342] eta: 0:00:48 time: 1.2321 data_time: 1.1431 memory: 3133 +2023/06/05 16:45:37 - mmengine - INFO - Epoch(val) [5][342/342] accuracy/top1: 51.8126 data_time: 1.2675 time: 1.3576 +2023/06/05 16:47:41 - mmengine - INFO - Epoch(train) [6][100/342] lr: 5.0913e-05 eta: 0:31:55 time: 1.1654 data_time: 1.0260 memory: 6319 loss: 0.5698 +2023/06/05 16:49:40 - mmengine - INFO - Epoch(train) [6][200/342] lr: 4.6820e-05 eta: 0:29:56 time: 1.1347 data_time: 0.9944 memory: 6319 loss: 0.5772 +2023/06/05 16:51:29 - mmengine - INFO - Exp name: resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1_20230605_153011 +2023/06/05 16:51:40 - mmengine - INFO - Epoch(train) [6][300/342] lr: 4.2795e-05 eta: 0:27:58 time: 1.1354 data_time: 0.9951 memory: 6319 loss: 0.5719 +2023/06/05 16:52:33 - mmengine - INFO - Exp name: resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1_20230605_153011 +2023/06/05 16:52:33 - mmengine - INFO - Saving checkpoint at 6 epochs +2023/06/05 16:54:28 - mmengine - INFO - Epoch(val) [6][100/342] eta: 0:04:24 time: 0.3954 data_time: 0.3073 memory: 6319 +2023/06/05 16:56:56 - mmengine - INFO - Epoch(val) [6][200/342] eta: 0:03:03 time: 0.6211 data_time: 0.5270 memory: 3133 +2023/06/05 17:00:38 - mmengine - INFO - Epoch(val) [6][300/342] eta: 0:01:07 time: 2.0907 data_time: 1.9988 memory: 3133 +2023/06/05 17:04:03 - mmengine - INFO - Epoch(val) [6][342/342] accuracy/top1: 51.9637 data_time: 1.7389 time: 1.8296 +2023/06/05 17:06:51 - mmengine - INFO - Epoch(train) [7][100/342] lr: 3.7265e-05 eta: 0:25:40 time: 1.9474 data_time: 1.8078 memory: 6319 loss: 0.5599 +2023/06/05 17:09:46 - mmengine - INFO - Epoch(train) [7][200/342] lr: 3.3546e-05 eta: 0:24:06 time: 1.8817 data_time: 1.7402 memory: 6319 loss: 0.5564 +2023/06/05 17:12:50 - mmengine - INFO - Epoch(train) [7][300/342] lr: 3.0008e-05 eta: 0:22:29 time: 1.8208 data_time: 1.6773 memory: 6319 loss: 0.5530 +2023/06/05 17:13:42 - mmengine - INFO - Exp name: resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1_20230605_153011 +2023/06/05 17:13:42 - mmengine - INFO - Saving checkpoint at 7 epochs +2023/06/05 17:15:41 - mmengine - INFO - Epoch(val) [7][100/342] eta: 0:04:35 time: 0.3869 data_time: 0.2981 memory: 6319 +2023/06/05 17:17:27 - mmengine - INFO - Epoch(val) [7][200/342] eta: 0:02:35 time: 0.2746 data_time: 0.1862 memory: 3133 +2023/06/05 17:19:47 - mmengine - INFO - Epoch(val) [7][300/342] eta: 0:00:50 time: 1.3018 data_time: 1.2139 memory: 3133 +2023/06/05 17:22:23 - mmengine - INFO - Epoch(val) [7][342/342] accuracy/top1: 52.2379 data_time: 1.2955 time: 1.3855 +2023/06/05 17:24:27 - mmengine - INFO - Epoch(train) [8][100/342] lr: 2.5353e-05 eta: 0:19:29 time: 1.2690 data_time: 0.5319 memory: 6319 loss: 0.5469 +2023/06/05 17:26:28 - mmengine - INFO - Epoch(train) [8][200/342] lr: 2.2372e-05 eta: 0:17:21 time: 1.2214 data_time: 0.2534 memory: 6319 loss: 0.5319 +2023/06/05 17:28:29 - mmengine - INFO - Epoch(train) [8][300/342] lr: 1.9667e-05 eta: 0:15:13 time: 1.2270 data_time: 0.5351 memory: 6319 loss: 0.5496 +2023/06/05 17:29:23 - mmengine - INFO - Exp name: resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1_20230605_153011 +2023/06/05 17:29:23 - mmengine - INFO - Saving checkpoint at 8 epochs +2023/06/05 17:31:17 - mmengine - INFO - Epoch(val) [8][100/342] eta: 0:04:24 time: 0.4101 data_time: 0.3232 memory: 6319 +2023/06/05 17:32:56 - mmengine - INFO - Epoch(val) [8][200/342] eta: 0:02:27 time: 0.2743 data_time: 0.1875 memory: 3133 +2023/06/05 17:35:15 - mmengine - INFO - Epoch(val) [8][300/342] eta: 0:00:48 time: 1.4263 data_time: 1.3382 memory: 3133 +2023/06/05 17:37:49 - mmengine - INFO - Epoch(val) [8][342/342] accuracy/top1: 52.2321 data_time: 1.2488 time: 1.3374 +2023/06/05 17:39:52 - mmengine - INFO - Epoch(train) [9][100/342] lr: 1.6343e-05 eta: 0:12:14 time: 1.1555 data_time: 1.0151 memory: 6319 loss: 0.5447 +2023/06/05 17:41:58 - mmengine - INFO - Epoch(train) [9][200/342] lr: 1.4393e-05 eta: 0:10:08 time: 1.1952 data_time: 1.0507 memory: 6319 loss: 0.5306 +2023/06/05 17:43:17 - mmengine - INFO - Exp name: resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1_20230605_153011 +2023/06/05 17:44:01 - mmengine - INFO - Epoch(train) [9][300/342] lr: 1.2785e-05 eta: 0:08:02 time: 1.2326 data_time: 1.0923 memory: 6319 loss: 0.5441 +2023/06/05 17:44:54 - mmengine - INFO - Exp name: resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1_20230605_153011 +2023/06/05 17:44:54 - mmengine - INFO - Saving checkpoint at 9 epochs +2023/06/05 17:46:47 - mmengine - INFO - Epoch(val) [9][100/342] eta: 0:04:18 time: 0.3998 data_time: 0.3117 memory: 6319 +2023/06/05 17:48:30 - mmengine - INFO - Epoch(val) [9][200/342] eta: 0:02:29 time: 0.2737 data_time: 0.1850 memory: 3133 +2023/06/05 17:50:42 - mmengine - INFO - Epoch(val) [9][300/342] eta: 0:00:47 time: 1.3435 data_time: 1.2551 memory: 3133 +2023/06/05 17:53:37 - mmengine - INFO - Epoch(val) [9][342/342] accuracy/top1: 52.3913 data_time: 1.2917 time: 1.3814 +2023/06/05 17:55:40 - mmengine - INFO - Epoch(train) [10][100/342] lr: 1.1116e-05 eta: 0:05:04 time: 1.1965 data_time: 1.0550 memory: 6319 loss: 0.5496 +2023/06/05 17:57:49 - mmengine - INFO - Epoch(train) [10][200/342] lr: 1.0388e-05 eta: 0:02:58 time: 1.3288 data_time: 0.8463 memory: 6319 loss: 0.5357 +2023/06/05 17:59:49 - mmengine - INFO - Epoch(train) [10][300/342] lr: 1.0035e-05 eta: 0:00:52 time: 1.4149 data_time: 0.6167 memory: 6319 loss: 0.5258 +2023/06/05 18:00:49 - mmengine - INFO - Exp name: resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1_20230605_153011 +2023/06/05 18:00:49 - mmengine - INFO - Saving checkpoint at 10 epochs +2023/06/05 18:02:45 - mmengine - INFO - Epoch(val) [10][100/342] eta: 0:04:25 time: 0.4074 data_time: 0.3209 memory: 6319 +2023/06/05 18:04:27 - mmengine - INFO - Epoch(val) [10][200/342] eta: 0:02:30 time: 0.2618 data_time: 0.1740 memory: 3133 +2023/06/05 18:06:45 - mmengine - INFO - Epoch(val) [10][300/342] eta: 0:00:48 time: 1.2870 data_time: 1.1983 memory: 3133 +2023/06/05 18:09:21 - mmengine - INFO - Epoch(val) [10][342/342] accuracy/top1: 52.5590 data_time: 1.2733 time: 1.3625 diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/20230605_153011.json b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/20230605_153011.json new file mode 100644 index 0000000000000000000000000000000000000000..dead89868e085de6e724b2c46bb10b1ccd96d01f --- /dev/null +++ b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/20230605_153011.json @@ -0,0 +1,40 @@ +{"lr": 9.981404789023513e-05, "data_time": 0.8983277320861817, "loss": 0.6786872267723083, "time": 1.204912495613098, "epoch": 1, "memory": 9436, "step": 100} +{"lr": 9.925023268502683e-05, "data_time": 0.8003745079040527, "loss": 0.6656096994876861, "time": 1.1238023281097411, "epoch": 1, "memory": 6319, "step": 200} +{"lr": 9.831328961451971e-05, "data_time": 0.6303098917007446, "loss": 0.6573303699493408, "time": 1.1811432123184205, "epoch": 1, "memory": 6319, "step": 300} +{"accuracy/top1": 50.21062469482422, "data_time": 1.2180426671491031, "time": 1.3094420851322643, "step": 1} +{"lr": 9.635783271319408e-05, "data_time": 1.0695750951766967, "loss": 0.6448290288448334, "time": 1.2105227947235107, "epoch": 2, "memory": 6319, "step": 442} +{"lr": 9.455662799327337e-05, "data_time": 1.0369538068771362, "loss": 0.6401268422603608, "time": 1.1773675441741944, "epoch": 2, "memory": 6319, "step": 542} +{"lr": 9.242187293325357e-05, "data_time": 0.9550090789794922, "loss": 0.6284441351890564, "time": 1.0953048944473267, "epoch": 2, "memory": 6319, "step": 642} +{"accuracy/top1": 50.72172927856445, "data_time": 1.2330823938631108, "time": 1.323526049494396, "step": 2} +{"lr": 8.885322471322101e-05, "data_time": 1.0629539489746094, "loss": 0.6251482427120209, "time": 1.2071065425872802, "epoch": 3, "memory": 6319, "step": 784} +{"lr": 8.59909449463049e-05, "data_time": 1.0558144807815553, "loss": 0.6138731718063355, "time": 1.198172664642334, "epoch": 3, "memory": 6319, "step": 884} +{"lr": 8.28673425957603e-05, "data_time": 1.1041572332382201, "loss": 0.6115147233009338, "time": 1.2470949649810792, "epoch": 3, "memory": 6319, "step": 984} +{"accuracy/top1": 51.36962127685547, "data_time": 1.2190984991479545, "time": 1.3094370796103518, "step": 3} +{"lr": 7.803482720903195e-05, "data_time": 0.970897626876831, "loss": 0.6027706265449524, "time": 1.114265513420105, "epoch": 4, "memory": 6319, "step": 1126} +{"lr": 7.439165228138202e-05, "data_time": 1.1045328855514527, "loss": 0.594668060541153, "time": 1.243904948234558, "epoch": 4, "memory": 6319, "step": 1226} +{"lr": 7.058496260180133e-05, "data_time": 1.0215199708938598, "loss": 0.5963524401187896, "time": 1.168829584121704, "epoch": 4, "memory": 6319, "step": 1326} +{"accuracy/top1": 51.686702728271484, "data_time": 1.2391408337796048, "time": 1.3298753531264147, "step": 4} +{"lr": 6.496162032454504e-05, "data_time": 0.9684446811676025, "loss": 0.5866643965244294, "time": 1.2702704429626466, "epoch": 5, "memory": 6319, "step": 1468} +{"lr": 6.089416958157208e-05, "data_time": 1.020720362663269, "loss": 0.5730401337146759, "time": 1.1602105855941773, "epoch": 5, "memory": 6319, "step": 1568} +{"lr": 5.677701788155902e-05, "data_time": 1.087578296661377, "loss": 0.5702815890312195, "time": 1.2296816825866699, "epoch": 5, "memory": 6319, "step": 1668} +{"accuracy/top1": 51.812618255615234, "data_time": 1.267533813204084, "time": 1.3576031586171586, "step": 5} +{"lr": 5.091330063600123e-05, "data_time": 1.0259903907775878, "loss": 0.5698013365268707, "time": 1.1654181480407715, "epoch": 6, "memory": 6319, "step": 1810} +{"lr": 4.681972449602271e-05, "data_time": 0.9943993330001831, "loss": 0.5771673381328583, "time": 1.13474702835083, "epoch": 6, "memory": 6319, "step": 1910} +{"lr": 4.27951262698573e-05, "data_time": 0.9950903654098511, "loss": 0.5718912839889526, "time": 1.1354322195053101, "epoch": 6, "memory": 6319, "step": 2010} +{"accuracy/top1": 51.963714599609375, "data_time": 1.738925643982067, "time": 1.8295878154543328, "step": 6} +{"lr": 3.726501555491419e-05, "data_time": 1.807818388938904, "loss": 0.5598736822605133, "time": 1.9473816871643066, "epoch": 7, "memory": 6319, "step": 2152} +{"lr": 3.354602177213381e-05, "data_time": 1.740167212486267, "loss": 0.5563906371593476, "time": 1.8816517114639282, "epoch": 7, "memory": 6319, "step": 2252} +{"lr": 3.0007932735217966e-05, "data_time": 1.6773326873779297, "loss": 0.553026819229126, "time": 1.8207677364349366, "epoch": 7, "memory": 6319, "step": 2352} +{"accuracy/top1": 52.23786926269531, "data_time": 1.2954849925750198, "time": 1.3854857992500327, "step": 7} +{"lr": 2.5352754318215113e-05, "data_time": 0.5319002628326416, "loss": 0.5469098925590515, "time": 1.2690301895141602, "epoch": 8, "memory": 6319, "step": 2494} +{"lr": 2.2372383915844294e-05, "data_time": 0.2533545732498169, "loss": 0.5318661510944367, "time": 1.221434473991394, "epoch": 8, "memory": 6319, "step": 2594} +{"lr": 1.9667136874426842e-05, "data_time": 0.5351297378540039, "loss": 0.5496402144432068, "time": 1.2270087718963623, "epoch": 8, "memory": 6319, "step": 2694} +{"accuracy/top1": 52.23214340209961, "data_time": 1.2487522596520515, "time": 1.337363676844116, "step": 8} +{"lr": 1.6342572053355464e-05, "data_time": 1.0150771617889405, "loss": 0.5446509897708893, "time": 1.155514645576477, "epoch": 9, "memory": 6319, "step": 2836} +{"lr": 1.4392564451840171e-05, "data_time": 1.0506842851638794, "loss": 0.5305792450904846, "time": 1.1952463865280152, "epoch": 9, "memory": 6319, "step": 2936} +{"lr": 1.2784967834899793e-05, "data_time": 1.0922734975814818, "loss": 0.5441479206085205, "time": 1.232629084587097, "epoch": 9, "memory": 6319, "step": 3036} +{"accuracy/top1": 52.391258239746094, "data_time": 1.2917077145145517, "time": 1.3813747428248992, "step": 9} +{"lr": 1.1116448178051759e-05, "data_time": 1.0549811363220214, "loss": 0.5495900869369507, "time": 1.1964537620544433, "epoch": 10, "memory": 6319, "step": 3178} +{"lr": 1.0387683707930187e-05, "data_time": 0.8462581872940064, "loss": 0.5357032209634781, "time": 1.328784203529358, "epoch": 10, "memory": 6319, "step": 3278} +{"lr": 1.0035100273117234e-05, "data_time": 0.6166882276535034, "loss": 0.5258055031299591, "time": 1.4149172067642213, "epoch": 10, "memory": 6319, "step": 3378} +{"accuracy/top1": 52.55895233154297, "data_time": 1.2733482125896753, "time": 1.3625113748600461, "step": 10} diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/config.py b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/config.py new file mode 100644 index 0000000000000000000000000000000000000000..9c6aa9196e597dae584979a8296cf7085929b605 --- /dev/null +++ b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/config.py @@ -0,0 +1,230 @@ +optim_wrapper = dict( + optimizer=dict( + type='SGD', + lr=0.0001, + momentum=0.9, + weight_decay=0.0001, + _scope_='mmpretrain'), + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=False, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=512) +model = dict( + type='ImageClassifier', + backbone=dict( + frozen_stages=2, + type='ResNet', + depth=50, + num_stages=4, + out_indices=(3, ), + style='pytorch', + init_cfg=dict( + type='Pretrained', + checkpoint= + 'https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth', + prefix='backbone')), + neck=dict(type='GlobalAveragePooling'), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=2048, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + topk=1)) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3fake8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3real8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3fake8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3real8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = dict(type='Accuracy', topk=1) +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3fake8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3real8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = dict(type='Accuracy', topk=1) +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1' diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/events.out.tfevents.1685950225.SH-IDC1-10-140-24-15.130656.0 b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/events.out.tfevents.1685950225.SH-IDC1-10-140-24-15.130656.0 new file mode 100644 index 0000000000000000000000000000000000000000..0d1e61842afd2f193e8cbafff04f5f815476e573 --- /dev/null +++ b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/events.out.tfevents.1685950225.SH-IDC1-10-140-24-15.130656.0 @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:7dcadc545a5a9cc65eec56b6070171bc094c966a1a72c4b59a5ed6c33c10c956 +size 13209399 diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/scalars.json b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/scalars.json new file mode 100644 index 0000000000000000000000000000000000000000..dead89868e085de6e724b2c46bb10b1ccd96d01f --- /dev/null +++ b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/scalars.json @@ -0,0 +1,40 @@ +{"lr": 9.981404789023513e-05, "data_time": 0.8983277320861817, "loss": 0.6786872267723083, "time": 1.204912495613098, "epoch": 1, "memory": 9436, "step": 100} +{"lr": 9.925023268502683e-05, "data_time": 0.8003745079040527, "loss": 0.6656096994876861, "time": 1.1238023281097411, "epoch": 1, "memory": 6319, "step": 200} +{"lr": 9.831328961451971e-05, "data_time": 0.6303098917007446, "loss": 0.6573303699493408, "time": 1.1811432123184205, "epoch": 1, "memory": 6319, "step": 300} +{"accuracy/top1": 50.21062469482422, "data_time": 1.2180426671491031, "time": 1.3094420851322643, "step": 1} +{"lr": 9.635783271319408e-05, "data_time": 1.0695750951766967, "loss": 0.6448290288448334, "time": 1.2105227947235107, "epoch": 2, "memory": 6319, "step": 442} +{"lr": 9.455662799327337e-05, "data_time": 1.0369538068771362, "loss": 0.6401268422603608, "time": 1.1773675441741944, "epoch": 2, "memory": 6319, "step": 542} +{"lr": 9.242187293325357e-05, "data_time": 0.9550090789794922, "loss": 0.6284441351890564, "time": 1.0953048944473267, "epoch": 2, "memory": 6319, "step": 642} +{"accuracy/top1": 50.72172927856445, "data_time": 1.2330823938631108, "time": 1.323526049494396, "step": 2} +{"lr": 8.885322471322101e-05, "data_time": 1.0629539489746094, "loss": 0.6251482427120209, "time": 1.2071065425872802, "epoch": 3, "memory": 6319, "step": 784} +{"lr": 8.59909449463049e-05, "data_time": 1.0558144807815553, "loss": 0.6138731718063355, "time": 1.198172664642334, "epoch": 3, "memory": 6319, "step": 884} +{"lr": 8.28673425957603e-05, "data_time": 1.1041572332382201, "loss": 0.6115147233009338, "time": 1.2470949649810792, "epoch": 3, "memory": 6319, "step": 984} +{"accuracy/top1": 51.36962127685547, "data_time": 1.2190984991479545, "time": 1.3094370796103518, "step": 3} +{"lr": 7.803482720903195e-05, "data_time": 0.970897626876831, "loss": 0.6027706265449524, "time": 1.114265513420105, "epoch": 4, "memory": 6319, "step": 1126} +{"lr": 7.439165228138202e-05, "data_time": 1.1045328855514527, "loss": 0.594668060541153, "time": 1.243904948234558, "epoch": 4, "memory": 6319, "step": 1226} +{"lr": 7.058496260180133e-05, "data_time": 1.0215199708938598, "loss": 0.5963524401187896, "time": 1.168829584121704, "epoch": 4, "memory": 6319, "step": 1326} +{"accuracy/top1": 51.686702728271484, "data_time": 1.2391408337796048, "time": 1.3298753531264147, "step": 4} +{"lr": 6.496162032454504e-05, "data_time": 0.9684446811676025, "loss": 0.5866643965244294, "time": 1.2702704429626466, "epoch": 5, "memory": 6319, "step": 1468} +{"lr": 6.089416958157208e-05, "data_time": 1.020720362663269, "loss": 0.5730401337146759, "time": 1.1602105855941773, "epoch": 5, "memory": 6319, "step": 1568} +{"lr": 5.677701788155902e-05, "data_time": 1.087578296661377, "loss": 0.5702815890312195, "time": 1.2296816825866699, "epoch": 5, "memory": 6319, "step": 1668} +{"accuracy/top1": 51.812618255615234, "data_time": 1.267533813204084, "time": 1.3576031586171586, "step": 5} +{"lr": 5.091330063600123e-05, "data_time": 1.0259903907775878, "loss": 0.5698013365268707, "time": 1.1654181480407715, "epoch": 6, "memory": 6319, "step": 1810} +{"lr": 4.681972449602271e-05, "data_time": 0.9943993330001831, "loss": 0.5771673381328583, "time": 1.13474702835083, "epoch": 6, "memory": 6319, "step": 1910} +{"lr": 4.27951262698573e-05, "data_time": 0.9950903654098511, "loss": 0.5718912839889526, "time": 1.1354322195053101, "epoch": 6, "memory": 6319, "step": 2010} +{"accuracy/top1": 51.963714599609375, "data_time": 1.738925643982067, "time": 1.8295878154543328, "step": 6} +{"lr": 3.726501555491419e-05, "data_time": 1.807818388938904, "loss": 0.5598736822605133, "time": 1.9473816871643066, "epoch": 7, "memory": 6319, "step": 2152} +{"lr": 3.354602177213381e-05, "data_time": 1.740167212486267, "loss": 0.5563906371593476, "time": 1.8816517114639282, "epoch": 7, "memory": 6319, "step": 2252} +{"lr": 3.0007932735217966e-05, "data_time": 1.6773326873779297, "loss": 0.553026819229126, "time": 1.8207677364349366, "epoch": 7, "memory": 6319, "step": 2352} +{"accuracy/top1": 52.23786926269531, "data_time": 1.2954849925750198, "time": 1.3854857992500327, "step": 7} +{"lr": 2.5352754318215113e-05, "data_time": 0.5319002628326416, "loss": 0.5469098925590515, "time": 1.2690301895141602, "epoch": 8, "memory": 6319, "step": 2494} +{"lr": 2.2372383915844294e-05, "data_time": 0.2533545732498169, "loss": 0.5318661510944367, "time": 1.221434473991394, "epoch": 8, "memory": 6319, "step": 2594} +{"lr": 1.9667136874426842e-05, "data_time": 0.5351297378540039, "loss": 0.5496402144432068, "time": 1.2270087718963623, "epoch": 8, "memory": 6319, "step": 2694} +{"accuracy/top1": 52.23214340209961, "data_time": 1.2487522596520515, "time": 1.337363676844116, "step": 8} +{"lr": 1.6342572053355464e-05, "data_time": 1.0150771617889405, "loss": 0.5446509897708893, "time": 1.155514645576477, "epoch": 9, "memory": 6319, "step": 2836} +{"lr": 1.4392564451840171e-05, "data_time": 1.0506842851638794, "loss": 0.5305792450904846, "time": 1.1952463865280152, "epoch": 9, "memory": 6319, "step": 2936} +{"lr": 1.2784967834899793e-05, "data_time": 1.0922734975814818, "loss": 0.5441479206085205, "time": 1.232629084587097, "epoch": 9, "memory": 6319, "step": 3036} +{"accuracy/top1": 52.391258239746094, "data_time": 1.2917077145145517, "time": 1.3813747428248992, "step": 9} +{"lr": 1.1116448178051759e-05, "data_time": 1.0549811363220214, "loss": 0.5495900869369507, "time": 1.1964537620544433, "epoch": 10, "memory": 6319, "step": 3178} +{"lr": 1.0387683707930187e-05, "data_time": 0.8462581872940064, "loss": 0.5357032209634781, "time": 1.328784203529358, "epoch": 10, "memory": 6319, "step": 3278} +{"lr": 1.0035100273117234e-05, "data_time": 0.6166882276535034, "loss": 0.5258055031299591, "time": 1.4149172067642213, "epoch": 10, "memory": 6319, "step": 3378} +{"accuracy/top1": 52.55895233154297, "data_time": 1.2733482125896753, "time": 1.3625113748600461, "step": 10} diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/35624.png_1.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/35624.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..69beed0830e65ebd75c27d4dd93b1e98450b566d Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/35624.png_1.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/35624.png_10.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/35624.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..4368446c001ed05090491dd85c38ec5cccff09cd Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/35624.png_10.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/35624.png_2.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/35624.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..d259c211f3080d4f23b62ac6f3d9597d6bae3fb5 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/35624.png_2.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/35624.png_3.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/35624.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..773b4bafc1de4d0b9ffc2280d6cdc006e90d09d2 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/35624.png_3.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/35624.png_4.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/35624.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..7e021b9627ba5304e2377def7e5a8a5e29716830 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/35624.png_4.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/35624.png_5.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/35624.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..787680ca8327f39edb0fd368ae0c314999473f5d Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/35624.png_5.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/35624.png_6.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/35624.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..8526d32d4c12a6acb373efaad239f284cd690fd9 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/35624.png_6.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/35624.png_7.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/35624.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..cf4485ff1998e9f2b29554f8f00183c4ee469430 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/35624.png_7.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/35624.png_8.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/35624.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..18b051e506c99562152238e09cc3ef3aef93d751 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/35624.png_8.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/35624.png_9.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/35624.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..ad771c38696a85b0b8e72b10d316bca1d7c69324 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/35624.png_9.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/38442.png_1.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/38442.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..5a0e83c11e31a0c6ff769af25e615b7b26825a67 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/38442.png_1.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/38442.png_10.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/38442.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..aea0580ea24baa0a155ddf9c560a0f16a4411b85 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/38442.png_10.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/38442.png_2.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/38442.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..bb60558e88532c731e5170cc068b642a4e67a113 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/38442.png_2.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/38442.png_3.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/38442.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..202cd1d2e376a5f7eceb8e693e60b6a257389368 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/38442.png_3.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/38442.png_4.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/38442.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..1921db67cb362a794af1dca87c82ca34dcc7f69f Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/38442.png_4.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/38442.png_5.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/38442.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..2c7c4bc8c3a2b4bc27a5f8f7d4968867fdd040f6 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/38442.png_5.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/38442.png_6.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/38442.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..25b722338c8fe9c0c847879b9bb6c60508de152a Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/38442.png_6.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/38442.png_7.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/38442.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..84be2c743e330b1b35674b92a9319a265eba2f3c Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/38442.png_7.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/38442.png_8.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/38442.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..ce2d54dc7478d79a6eb492f74e701642a78eb6f5 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/38442.png_8.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/38442.png_9.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/38442.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..85fc32e5be078796547067e2e8b6db01e05ee962 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/38442.png_9.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/43283.png_1.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/43283.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..ad67b032a47ec4e081b47a1d4ade0eb0c66efac1 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/43283.png_1.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/43283.png_10.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/43283.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..36cf9bb6587aea652b3b7e93411836447839db99 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/43283.png_10.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/43283.png_2.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/43283.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..e0471abf6b47e1c4a6d50d763af45717e8cf6ab7 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/43283.png_2.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/43283.png_3.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/43283.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..673fc1a44551429d4110fa53125846b36a157675 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/43283.png_3.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/43283.png_4.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/43283.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..a2ed3d6677ffef9d14718e4ec0b5d7dec9342637 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/43283.png_4.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/43283.png_5.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/43283.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..70550c41750a701426c994d9c0f2ac4717cd7cf0 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/43283.png_5.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/43283.png_6.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/43283.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..6b62648c016afeda7266372b8c84bdb33665cba2 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/43283.png_6.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/43283.png_7.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/43283.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..d3f59528f68009ba36f630f9c7c7b51eb5472993 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/43283.png_7.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/43283.png_8.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/43283.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..1f83b3d3e57a5add0c3b6f359f52cebbd4040dc7 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/43283.png_8.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/43283.png_9.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/43283.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..156127561c65602890561b36c22ff5587ba21da0 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/43283.png_9.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/44982.png_1.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/44982.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..971a657deaf052102027f073cd5d3acb4c8a1f8a Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/44982.png_1.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/44982.png_10.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/44982.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..d111dcb234e89873fe5526fc8e4e29253fcb6368 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/44982.png_10.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/44982.png_2.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/44982.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..11616b2fa17f0b834235f83a54030726cf2f6676 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/44982.png_2.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/44982.png_3.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/44982.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..08cd5dece8b93f53ae154e97fde4e99c27d522f7 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/44982.png_3.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/44982.png_4.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/44982.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..a61327f4407b1a425a2840d6df155e6433414bde Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/44982.png_4.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/44982.png_5.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/44982.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..fe1f973a4dfecf3da322a25142b9bc4f261f5ed9 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/44982.png_5.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/44982.png_6.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/44982.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..29c49919f9420a71c4a63867b260a53fbe2529bb Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/44982.png_6.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/44982.png_7.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/44982.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..ffdd941cecb09fb93197a4a01ca9738e35ea0f96 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/44982.png_7.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/44982.png_8.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/44982.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..110d8a9cf78759c3aba08beb6423c60d30fbba46 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/44982.png_8.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/44982.png_9.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/44982.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..31be76fc1b3912e391cd1c6f5ad848095213c47f Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/44982.png_9.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/56423.png_1.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/56423.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..a9ad280162a772600d42f33c9919b1b1d2bef083 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/56423.png_1.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/56423.png_10.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/56423.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..750d1c169b0adc1e1432eaa91468eb1c0a19bedf Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/56423.png_10.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/56423.png_2.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/56423.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..84c690452e604ae7f4025c755b93e043ef4944f8 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/56423.png_2.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/56423.png_3.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/56423.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..d3a0f4014c4907b5de20f3b769b2b9f18a680e2a Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/56423.png_3.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/56423.png_4.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/56423.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..7c768e9fa4a7115a038c8b00c384f19c7b2bf86c Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/56423.png_4.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/56423.png_5.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/56423.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..e45a5cb7de67ef3b622c2b4e124ea6c89e07bdd9 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/56423.png_5.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/56423.png_6.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/56423.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..000fd7588774bee5f4d4aaf85ad3fe97f73a5ee9 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/56423.png_6.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/56423.png_7.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/56423.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..16c67ef900b6c30933db452bd445285b485a12af Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/56423.png_7.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/56423.png_8.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/56423.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..309d8bceb15165362e988426973a40ab9698803f Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/56423.png_8.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/56423.png_9.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/56423.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..43c127316d4b27696582ecc13d8047ea3e238112 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/56423.png_9.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/61024.png_1.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/61024.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..d544d6b0c2a1619bbcdaa2a3a363a2a7d7a824d2 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/61024.png_1.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/61024.png_10.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/61024.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..8c5134a8d6f9b616c0a4bf8c1d90be9f30bf5862 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/61024.png_10.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/61024.png_2.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/61024.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..375ff82ef454754c07279ab8e3a2ef1f1007bf67 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/61024.png_2.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/61024.png_3.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/61024.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..95462f6d64701952fbfa55510df657f481729866 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/61024.png_3.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/61024.png_4.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/61024.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..97c4834f8dd885e3d2148a1a0fc0640a63f7be9c Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/61024.png_4.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/61024.png_5.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/61024.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..5be132256274a7146492dce30441f3c728aabba3 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/61024.png_5.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/61024.png_6.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/61024.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..54b0920eaceb21835c96c17b5c1203986c1e3b69 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/61024.png_6.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/61024.png_7.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/61024.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..069293b6604f6b48314dde9ca69047cf832c260d Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/61024.png_7.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/61024.png_8.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/61024.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..9e6367f9b958023be0691faae81afb09c63541c5 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/61024.png_8.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/61024.png_9.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/61024.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..a5da49f3d1bead5e9ffd4bf13b03065c45ffa787 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/61024.png_9.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/66390.png_1.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/66390.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..9010b6e3041cd2f89fb97a73381d4f94a91a99eb Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/66390.png_1.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/66390.png_10.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/66390.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..0125891bde8a4c1a20d48919bc327edad7afb259 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/66390.png_10.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/66390.png_2.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/66390.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..932b96edf4ce6e417c3e6d9d1ebf51207dadb3dc Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/66390.png_2.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/66390.png_3.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/66390.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..e7876484aa3734be2b847d51fb40c9b948671c17 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/66390.png_3.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/66390.png_4.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/66390.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..e0b062a655be49b0b67c12b137e8eb4fe43d8c24 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/66390.png_4.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/66390.png_5.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/66390.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..f5e7d744908b52f1eefb6c4eb87a7bcf5dac50b2 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/66390.png_5.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/66390.png_6.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/66390.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..d35e4d7192bd002dfb3a0dbc3af090b8e57f92dd Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/66390.png_6.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/66390.png_7.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/66390.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..f9b650c765ea1946a45be221d9d18e3d72faca91 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/66390.png_7.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/66390.png_8.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/66390.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..1599eb8463ecec7a722e6b43b0d5d86021eacea4 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/66390.png_8.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/66390.png_9.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/66390.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..0bd2993284174270a41ba0ba1edd35e849db5d64 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/66390.png_9.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/flickr_wild_000179.jpg_1.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/flickr_wild_000179.jpg_1.png new file mode 100644 index 0000000000000000000000000000000000000000..7312674b733ccef093c9065f1d7af0e328c53f2b Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/flickr_wild_000179.jpg_1.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/flickr_wild_000179.jpg_10.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/flickr_wild_000179.jpg_10.png new file mode 100644 index 0000000000000000000000000000000000000000..399651dda2dd65400b407e394ee10c42dcd5a898 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/flickr_wild_000179.jpg_10.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/flickr_wild_000179.jpg_2.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/flickr_wild_000179.jpg_2.png new file mode 100644 index 0000000000000000000000000000000000000000..d400e0e26fc2a7521b165350b8f691c0ded53bff Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/flickr_wild_000179.jpg_2.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/flickr_wild_000179.jpg_3.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/flickr_wild_000179.jpg_3.png new file mode 100644 index 0000000000000000000000000000000000000000..f801a46c6bd9113e30b5f2443f73dea0f3664f7a Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/flickr_wild_000179.jpg_3.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/flickr_wild_000179.jpg_4.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/flickr_wild_000179.jpg_4.png new file mode 100644 index 0000000000000000000000000000000000000000..bbc1ffc73ebe2f58713d9b74ee0e5d95852936c0 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/flickr_wild_000179.jpg_4.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/flickr_wild_000179.jpg_5.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/flickr_wild_000179.jpg_5.png new file mode 100644 index 0000000000000000000000000000000000000000..a51ab0ff342f66ba7616d269b8a765252dc01fde Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/flickr_wild_000179.jpg_5.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/flickr_wild_000179.jpg_6.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/flickr_wild_000179.jpg_6.png new file mode 100644 index 0000000000000000000000000000000000000000..a49f3b6c02f5cae470a1add7f09e38d5fe77693e Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/flickr_wild_000179.jpg_6.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/flickr_wild_000179.jpg_7.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/flickr_wild_000179.jpg_7.png new file mode 100644 index 0000000000000000000000000000000000000000..8cafdb15e8541008fd4b96572441d6f4bc310dbf Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/flickr_wild_000179.jpg_7.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/flickr_wild_000179.jpg_8.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/flickr_wild_000179.jpg_8.png new file mode 100644 index 0000000000000000000000000000000000000000..6c7dd098673810a0c5ebf9ac64bd8ed60c522ee0 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/flickr_wild_000179.jpg_8.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/flickr_wild_000179.jpg_9.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/flickr_wild_000179.jpg_9.png new file mode 100644 index 0000000000000000000000000000000000000000..2803fe221cf79337b7b41065594de46e5742c23b Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/flickr_wild_000179.jpg_9.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/pixabay_dog_002352.jpg_1.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/pixabay_dog_002352.jpg_1.png new file mode 100644 index 0000000000000000000000000000000000000000..f46610c6615781f9e6276ae35ec6b89994fe336f Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/pixabay_dog_002352.jpg_1.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/pixabay_dog_002352.jpg_10.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/pixabay_dog_002352.jpg_10.png new file mode 100644 index 0000000000000000000000000000000000000000..61b7e9dd9da0e6a816dd0a1acce5d869c851f15c Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/pixabay_dog_002352.jpg_10.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/pixabay_dog_002352.jpg_2.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/pixabay_dog_002352.jpg_2.png new file mode 100644 index 0000000000000000000000000000000000000000..0e120393dcce587a7117927ed3a389f375e096fc Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/pixabay_dog_002352.jpg_2.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/pixabay_dog_002352.jpg_3.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/pixabay_dog_002352.jpg_3.png new file mode 100644 index 0000000000000000000000000000000000000000..e894eafd5f345f3714a36176ffe986d0a8a394ab Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/pixabay_dog_002352.jpg_3.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/pixabay_dog_002352.jpg_4.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/pixabay_dog_002352.jpg_4.png new file mode 100644 index 0000000000000000000000000000000000000000..33c26439224505ed141af68daa8765a587d7e948 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/pixabay_dog_002352.jpg_4.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/pixabay_dog_002352.jpg_5.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/pixabay_dog_002352.jpg_5.png new file mode 100644 index 0000000000000000000000000000000000000000..7efbcc4ec922a67352506bfcd8f95d6fe558c140 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/pixabay_dog_002352.jpg_5.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/pixabay_dog_002352.jpg_6.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/pixabay_dog_002352.jpg_6.png new file mode 100644 index 0000000000000000000000000000000000000000..7d7fd34cd7dc243803e9e3fa306d796b74f64e3c Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/pixabay_dog_002352.jpg_6.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/pixabay_dog_002352.jpg_7.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/pixabay_dog_002352.jpg_7.png new file mode 100644 index 0000000000000000000000000000000000000000..c01fa49b4ee019356ff2d48c8bf2610fd83fdec3 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/pixabay_dog_002352.jpg_7.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/pixabay_dog_002352.jpg_8.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/pixabay_dog_002352.jpg_8.png new file mode 100644 index 0000000000000000000000000000000000000000..9b1f036f58097bb51f4dd09a1785f88183a49d17 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/pixabay_dog_002352.jpg_8.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/pixabay_dog_002352.jpg_9.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/pixabay_dog_002352.jpg_9.png new file mode 100644 index 0000000000000000000000000000000000000000..3bdf13ecec87b1184fce21ef6764c3bb3dd772a1 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/pixabay_dog_002352.jpg_9.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed14088.png_1.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed14088.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..7654fe2e3e10e213b2ec30e2dcd825583ac66363 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed14088.png_1.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed14088.png_10.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed14088.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..e7ee397de0cb34a5c883e4ac75462121953cec05 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed14088.png_10.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed14088.png_2.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed14088.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..eeb2e191742e3f3ea6d986cc5a748c49926bf6c0 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed14088.png_2.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed14088.png_3.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed14088.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..c4b10f400fdfae13602389bc361c2ba8e401d30d Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed14088.png_3.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed14088.png_4.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed14088.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..6af91f1ed31cea40a67b9df21f4a69b201eee2b6 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed14088.png_4.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed14088.png_5.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed14088.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..32bc23f4715135dbc4db5e22b04464601ebf324d Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed14088.png_5.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed14088.png_6.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed14088.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..ead83ade132c1052ccd6428a7e74ef57e9915fa9 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed14088.png_6.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed14088.png_7.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed14088.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..7c65688d9b330afe90686df282c0a146c9d986e2 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed14088.png_7.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed14088.png_8.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed14088.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..02e7553e82ad75b205685082e02cacc20f1c1f93 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed14088.png_8.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed14088.png_9.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed14088.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..db20a3364dc9c774a224237c3b7998e73bc8a4ff Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed14088.png_9.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed16124.png_1.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed16124.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..69cb92cf757060a968737556185fae9ed1abd647 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed16124.png_1.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed16124.png_10.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed16124.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..9e9ea3b10d1931c303498d5b9934f4744c4bd5b7 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed16124.png_10.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed16124.png_2.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed16124.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..df9a727b4938b4b81434ad93f0afab3f0340eb14 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed16124.png_2.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed16124.png_3.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed16124.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..591df1abc3b88e2ea8d2b571b8761c090e8fa679 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed16124.png_3.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed16124.png_4.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed16124.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..b94c56970f4b7cf1109c3891465c62ed046512ed Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed16124.png_4.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed16124.png_5.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed16124.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..16c599c839a9e2d767d938ec3636b64c218da240 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed16124.png_5.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed16124.png_6.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed16124.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..137c3e153bdf9178feb4c6eacc4c0c414aea02c5 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed16124.png_6.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed16124.png_7.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed16124.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..f3daa861257a582ace9a202b2aaa31fd39d63789 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed16124.png_7.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed16124.png_8.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed16124.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..c8db26e875fb4fcfdc33887d103dd374d722fa36 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed16124.png_8.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed16124.png_9.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed16124.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..c193c6bb050efee48b72f613a0ebe3764839d366 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed16124.png_9.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed17282.png_1.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed17282.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..60b7ea2b33f00416f111b614273f67907a0c4605 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed17282.png_1.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed17282.png_10.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed17282.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..cbc004813f19d5ed34652d807936fb6e522d58e7 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed17282.png_10.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed17282.png_2.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed17282.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..dae0e332011b28c337e43e666f8abbc333c35c67 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed17282.png_2.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed17282.png_3.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed17282.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..66a1447dac0ae040dd9933f1729a81f1b95e89da Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed17282.png_3.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed17282.png_4.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed17282.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..0353d27a10d1b676d625b1edbcb3dd28543c9960 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed17282.png_4.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed17282.png_5.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed17282.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..9244fbab9ca1aaf3c6413be034a7ffed09f5ce5f Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed17282.png_5.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed17282.png_6.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed17282.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..845b1614c938b79c495baa3af653b0d3558e1d01 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed17282.png_6.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed17282.png_7.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed17282.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..be5d9f5aefebb61c6095e37d4fcab45bf370a8b7 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed17282.png_7.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed17282.png_8.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed17282.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..d64745d9e1dce62872fe36f96c85b2af455eabac Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed17282.png_8.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed17282.png_9.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed17282.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..dfa0a7c7d4f61ccbfdc0c6da90636955adfd1464 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed17282.png_9.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed21815.png_1.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed21815.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..de4195d10ed7920646ebeb0ab8e9aab610b9ff58 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed21815.png_1.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed21815.png_10.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed21815.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..b481c467eb6549776a790546881fd17e90f618a9 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed21815.png_10.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed21815.png_2.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed21815.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..6612b136776afec8e7755cb87885ef53862c7c6c Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed21815.png_2.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed21815.png_3.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed21815.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..cdd0cae0b769ac405de37cfdc6a883e1c604c52c Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed21815.png_3.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed21815.png_4.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed21815.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..bbd02bcbb875b3671f543249d377644f91a5f2d6 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed21815.png_4.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed21815.png_5.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed21815.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..b0874037bdb7081513aeec2155f73d1e19ca65e5 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed21815.png_5.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed21815.png_6.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed21815.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..00de5f84cf6dbfba90aa69dbda71825a35c75ee7 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed21815.png_6.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed21815.png_7.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed21815.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..8699ba8a0462f20981103ec3c7bb883200cffe7f Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed21815.png_7.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed21815.png_8.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed21815.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..c16624b449c7a7d5ead8391407beb3b01985f2fb Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed21815.png_8.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed21815.png_9.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed21815.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..0c804705aa143133353efae4c465e4a3e65070dd Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed21815.png_9.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed22624.png_1.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed22624.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..0b81f03259ed2228722bf3f08b303d835b14f2a9 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed22624.png_1.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed22624.png_10.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed22624.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..a127de83fd3eaf53ab8d6a4e2819c2cb05638f0b Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed22624.png_10.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed22624.png_2.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed22624.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..d15fc36523c1b9182338a46e68cbaeda1bea9288 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed22624.png_2.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed22624.png_3.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed22624.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..d1b25aac7b650782d4c080a862a5d00009a5528c Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed22624.png_3.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed22624.png_4.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed22624.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..8eeeebc89b449fb596fe13a55a7e0e5c1c8f1ad5 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed22624.png_4.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed22624.png_5.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed22624.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..05653e10ac4577dfd27d1ee26c1119a984a061fc Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed22624.png_5.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed22624.png_6.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed22624.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..f40e20c80eb5637decea4014e6ce94e00fc0c5d5 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed22624.png_6.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed22624.png_7.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed22624.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..655232a749a675aaae993c7d7528bced090da027 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed22624.png_7.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed22624.png_8.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed22624.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..9d801483f21a060d6af19ec7eb634b5f48834439 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed22624.png_8.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed22624.png_9.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed22624.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..86d8e6db83df4e79120baef2b6d2732f6abb24d6 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed22624.png_9.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed23602.png_1.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed23602.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..11dd14923457a19a18fd0c91b5288d9b74ed6549 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed23602.png_1.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed23602.png_10.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed23602.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..cb842af47a2bab462a2753f90d006fc7371e37bc Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed23602.png_10.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed23602.png_2.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed23602.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..56ea33488893a9ccd9cd83c78df5568ef893a41f Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed23602.png_2.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed23602.png_3.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed23602.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..715a5c8e4d52fe6e65e500e76957d899a18d6951 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed23602.png_3.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed23602.png_4.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed23602.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..10eb8599c7ac3dde8b0fc645d55fba543d446287 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed23602.png_4.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed23602.png_5.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed23602.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..766c4c71b52850583ea37058cf2775f661e60968 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed23602.png_5.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed23602.png_6.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed23602.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..7cb739bb3062ef7f9ceeb8b3024630d20f7f218f Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed23602.png_6.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed23602.png_7.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed23602.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..a7a696de3bb18456cd0e7f3aa191a969d7c62b4f Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed23602.png_7.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed23602.png_8.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed23602.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..39fe608ec636f9b5e0999947ed272521a6e3b41e Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed23602.png_8.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed23602.png_9.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed23602.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..b12bf328a19e309c6bf33e3c5ac7535391a9e495 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed23602.png_9.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed24366.png_1.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed24366.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..4e1a69744757464b779836b216fed8d13924e84b Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed24366.png_1.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed24366.png_10.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed24366.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..90a9c6771ae549ee71f5fd70fab82bd7c4dddc5a Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed24366.png_10.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed24366.png_2.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed24366.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..24d890b32a400f70465e60baeb45155cf7a8d1f9 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed24366.png_2.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed24366.png_3.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed24366.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..bd5ac09a64fecba884146e05574006bc0973416e Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed24366.png_3.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed24366.png_4.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed24366.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..6113a52e6fc8ebc8ab60c77b42f4f0be2e57470e Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed24366.png_4.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed24366.png_5.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed24366.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..8405179f913133ae56cb4dadf640eb4a0344c1aa Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed24366.png_5.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed24366.png_6.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed24366.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..d2d10ce8048018e4226434e0e8088072c6ffde1a Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed24366.png_6.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed24366.png_7.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed24366.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..74ef3e0abd7f1a53ef5dd8f4b0823db1258fbab7 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed24366.png_7.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed24366.png_8.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed24366.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..cf28d8205aeca2f2c406d9a0c47b89655904a908 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed24366.png_8.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed24366.png_9.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed24366.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..e1afe53746d6dd4cd2d9251b0a552c660d57e088 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed24366.png_9.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed26848.png_1.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed26848.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..2382613f4df5b43ae5066e60bf6097bb4c14408e Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed26848.png_1.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed26848.png_10.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed26848.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..f08ceed862a20bd26f868f1f2c259ebfad9b431b Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed26848.png_10.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed26848.png_2.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed26848.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..31310dbb3fffb66b2cb5b19d644879541a3ea952 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed26848.png_2.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed26848.png_3.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed26848.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..bec62b43ee527a3e6ed3c04eac4e836cb211f403 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed26848.png_3.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed26848.png_4.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed26848.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..0cc26d4da59bea0f1b25766c0e7a97bd17d6ace3 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed26848.png_4.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed26848.png_5.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed26848.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..72a8ae53350ead09ed59cf41ab697b71718578e8 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed26848.png_5.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed26848.png_6.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed26848.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..d692c17a6c8d894ff90db74da86dae2cec62c38a Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed26848.png_6.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed26848.png_7.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed26848.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..54d0414a6588a99144a7197b168bb87b136a1f4e Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed26848.png_7.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed26848.png_8.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed26848.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..caca51aff37fec4069802805197305f834ca9116 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed26848.png_8.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed26848.png_9.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed26848.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..c3e95df068af97a4c61ab019730e89a93229c6eb Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed26848.png_9.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed44086.png_1.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed44086.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..f5fe57f66932cdfb87ea35840999ff6a5eec7d52 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed44086.png_1.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed44086.png_10.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed44086.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..c8136e02a0a3a33c8d38706f1c372d15a11a7994 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed44086.png_10.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed44086.png_2.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed44086.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..81646277686efe306e5b17b6360c09124e4bf126 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed44086.png_2.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed44086.png_3.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed44086.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..dae34d946fcda8c53901e20e1c9545f6ede46f0f Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed44086.png_3.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed44086.png_4.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed44086.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..cbe142689b660951cdb1711d41b2b45f0fbbb9ff Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed44086.png_4.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed44086.png_5.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed44086.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..8c9788b0231a0cb0ae8136da655092c0a1c7d67f Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed44086.png_5.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed44086.png_6.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed44086.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..2cb60ae9f0cc367359f2d93857545b6737a68cfe Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed44086.png_6.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed44086.png_7.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed44086.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..1b6801a3f8fc0a4938aa968210f03c144c332ff0 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed44086.png_7.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed44086.png_8.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed44086.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..5dd85ba0171ef42a3f6483504ac18475fb7716a9 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed44086.png_8.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed44086.png_9.png b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed44086.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..6d15e15a34bf3496bee790af5e1de49f9058e994 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/20230605_153011/vis_data/vis_image/seed44086.png_9.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/epoch_1.pth b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/epoch_1.pth new file mode 100644 index 0000000000000000000000000000000000000000..ac44027f8dd8eddf3eed4034a93a7cd7b75b91e9 --- /dev/null +++ b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/epoch_1.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:4f07dcd0a7517f520b553b164ebed24a14320481a88620405179cc9f6ebf9c76 +size 277074189 diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/epoch_10.pth b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/epoch_10.pth new file mode 100644 index 0000000000000000000000000000000000000000..6fe543b44ef1d1a9ec4073a18dde0d513cbc507f --- /dev/null +++ b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/epoch_10.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:413ad4fa1b269c5b8ebf09da27bc7d9d017c8fdaa86ac7329fac5c33dd21f302 +size 277406349 diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/epoch_2.pth b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/epoch_2.pth new file mode 100644 index 0000000000000000000000000000000000000000..9a7e73fb787f1d0eedf186302086568504097081 --- /dev/null +++ b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/epoch_2.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:033c2e25a91a0cde70f92622430ffa666f78955a5ddf80ad137c1b00bd5bed00 +size 277111757 diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/epoch_3.pth b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/epoch_3.pth new file mode 100644 index 0000000000000000000000000000000000000000..3ccfabb4f6f7caef4bec8b4b9d2dd477e7f07c31 --- /dev/null +++ b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/epoch_3.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:3b6eb792298254dfce2da0498c474994ca1ae4db3dbd7275db438b025b00974e +size 277148493 diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/epoch_4.pth b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/epoch_4.pth new file mode 100644 index 0000000000000000000000000000000000000000..72e26ef4e63ba6c5bafb8f96d2285a5c7360c013 --- /dev/null +++ b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/epoch_4.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:1334e9a21c550b8d76f15123af852c62517532e84f0f7664cbc86199360da21b +size 277185229 diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/epoch_5.pth b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/epoch_5.pth new file mode 100644 index 0000000000000000000000000000000000000000..5b44fcc347f39aabf15a8c0c7673a9e09094e5c2 --- /dev/null +++ b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/epoch_5.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:8dbb6b79f5e5cfac705ac736434bf3ab0b2bd9615e5458b3797cfc99602388e8 +size 277221965 diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/epoch_6.pth b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/epoch_6.pth new file mode 100644 index 0000000000000000000000000000000000000000..450c31b93299803f28bf0e858606970763eb9002 --- /dev/null +++ b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/epoch_6.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:325e8f6589ef07202dfa16569b80b1863ac58f8decd182cac35679aa9f262821 +size 277258573 diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/epoch_7.pth b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/epoch_7.pth new file mode 100644 index 0000000000000000000000000000000000000000..99a20c78432ade8dce919e7a84a00a3a52c09594 --- /dev/null +++ b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/epoch_7.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:8fe37f6c9418e191da37d445822f6c454c06b996000c0daa3fcc99682b8d706b +size 277295373 diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/epoch_8.pth b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/epoch_8.pth new file mode 100644 index 0000000000000000000000000000000000000000..0cfc7c1844db7c568f8de2bc35fbc09f4f84ad6f --- /dev/null +++ b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/epoch_8.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:37c76c45f2eadf18de238880cd09f14c5c46b1abfd829878139ea7a4e7db74aa +size 277332365 diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/epoch_9.pth b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/epoch_9.pth new file mode 100644 index 0000000000000000000000000000000000000000..4eb9460656584ffa09d2e38144bdcd16c0c4afb3 --- /dev/null +++ b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/epoch_9.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:412052642fa93d79cbfbf31d0ff1fd0ca44bb9e8af07f3500261e787b1c4779b +size 277369421 diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/last_checkpoint b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/last_checkpoint new file mode 100644 index 0000000000000000000000000000000000000000..6b8f5ca3eb736d41d8b98352bdcab050738a47a7 --- /dev/null +++ b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/last_checkpoint @@ -0,0 +1 @@ +/mnt/petrelfs/luzeyu/workspace/fakebench/mmpretrain/workdir/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/epoch_10.pth \ No newline at end of file diff --git a/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1.py b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1.py new file mode 100644 index 0000000000000000000000000000000000000000..9c6aa9196e597dae584979a8296cf7085929b605 --- /dev/null +++ b/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1.py @@ -0,0 +1,230 @@ +optim_wrapper = dict( + optimizer=dict( + type='SGD', + lr=0.0001, + momentum=0.9, + weight_decay=0.0001, + _scope_='mmpretrain'), + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=False, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=512) +model = dict( + type='ImageClassifier', + backbone=dict( + frozen_stages=2, + type='ResNet', + depth=50, + num_stages=4, + out_indices=(3, ), + style='pytorch', + init_cfg=dict( + type='Pretrained', + checkpoint= + 'https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth', + prefix='backbone')), + neck=dict(type='GlobalAveragePooling'), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=2048, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + topk=1)) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3fake8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3real8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3fake8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3real8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = dict(type='Accuracy', topk=1) +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3fake8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3real8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = dict(type='Accuracy', topk=1) +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/resnet50_2xb256_stylegan3_1m_lr3e-5_aug_5e-1' diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/20230605_201824.log b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/20230605_201824.log new file mode 100644 index 0000000000000000000000000000000000000000..dbf93cb86573235ec7c0683ce6a7a9d299d4e393 --- /dev/null +++ b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/20230605_201824.log @@ -0,0 +1,939 @@ +2023/06/05 20:18:27 - mmengine - INFO - +------------------------------------------------------------ +System environment: + sys.platform: linux + Python: 3.10.9 (main, Mar 8 2023, 10:47:38) [GCC 11.2.0] + CUDA available: True + numpy_random_seed: 829172272 + GPU 0,1: NVIDIA A100-SXM4-80GB + CUDA_HOME: /mnt/petrelfs/share/cuda-11.6 + NVCC: Cuda compilation tools, release 11.6, V11.6.124 + GCC: gcc (GCC) 7.5.0 + PyTorch: 1.13.1 + PyTorch compiling details: PyTorch built with: + - GCC 9.3 + - C++ Version: 201402 + - Intel(R) oneAPI Math Kernel Library Version 2021.4-Product Build 20210904 for Intel(R) 64 architecture applications + - Intel(R) MKL-DNN v2.6.0 (Git Hash 52b5f107dd9cf10910aaa19cb47f3abf9b349815) + - OpenMP 201511 (a.k.a. OpenMP 4.5) + - LAPACK is enabled (usually provided by MKL) + - NNPACK is enabled + - CPU capability usage: AVX2 + - CUDA Runtime 11.6 + - NVCC architecture flags: -gencode;arch=compute_37,code=sm_37;-gencode;arch=compute_50,code=sm_50;-gencode;arch=compute_60,code=sm_60;-gencode;arch=compute_61,code=sm_61;-gencode;arch=compute_70,code=sm_70;-gencode;arch=compute_75,code=sm_75;-gencode;arch=compute_80,code=sm_80;-gencode;arch=compute_86,code=sm_86;-gencode;arch=compute_37,code=compute_37 + - CuDNN 8.3.2 (built against CUDA 11.5) + - Magma 2.6.1 + - Build settings: BLAS_INFO=mkl, BUILD_TYPE=Release, CUDA_VERSION=11.6, CUDNN_VERSION=8.3.2, CXX_COMPILER=/opt/rh/devtoolset-9/root/usr/bin/c++, CXX_FLAGS= -fabi-version=11 -Wno-deprecated -fvisibility-inlines-hidden -DUSE_PTHREADPOOL -fopenmp -DNDEBUG -DUSE_KINETO -DUSE_FBGEMM -DUSE_QNNPACK -DUSE_PYTORCH_QNNPACK -DUSE_XNNPACK -DSYMBOLICATE_MOBILE_DEBUG_HANDLE -DEDGE_PROFILER_USE_KINETO -O2 -fPIC -Wno-narrowing -Wall -Wextra -Werror=return-type -Werror=non-virtual-dtor -Wno-missing-field-initializers -Wno-type-limits -Wno-array-bounds -Wno-unknown-pragmas -Wunused-local-typedefs -Wno-unused-parameter -Wno-unused-function -Wno-unused-result -Wno-strict-overflow -Wno-strict-aliasing -Wno-error=deprecated-declarations -Wno-stringop-overflow -Wno-psabi -Wno-error=pedantic -Wno-error=redundant-decls -Wno-error=old-style-cast -fdiagnostics-color=always -faligned-new -Wno-unused-but-set-variable -Wno-maybe-uninitialized -fno-math-errno -fno-trapping-math -Werror=format -Werror=cast-function-type -Wno-stringop-overflow, LAPACK_INFO=mkl, PERF_WITH_AVX=1, PERF_WITH_AVX2=1, PERF_WITH_AVX512=1, TORCH_VERSION=1.13.1, USE_CUDA=ON, USE_CUDNN=ON, USE_EXCEPTION_PTR=1, USE_GFLAGS=OFF, USE_GLOG=OFF, USE_MKL=ON, USE_MKLDNN=ON, USE_MPI=OFF, USE_NCCL=ON, USE_NNPACK=ON, USE_OPENMP=ON, USE_ROCM=OFF, + + TorchVision: 0.14.1 + OpenCV: 4.7.0 + MMEngine: 0.7.3 + +Runtime environment: + cudnn_benchmark: True + mp_cfg: {'mp_start_method': 'fork', 'opencv_num_threads': 0} + dist_cfg: {'backend': 'nccl'} + seed: None + deterministic: False + Distributed launcher: slurm + Distributed training: True + GPU number: 2 +------------------------------------------------------------ + +2023/06/05 20:18:32 - mmengine - INFO - Config: +optim_wrapper = dict( + optimizer=dict( + type='SGD', + lr=0.0005, + momentum=0.9, + weight_decay=0.0001, + _scope_='mmpretrain'), + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=False, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=512) +model = dict( + type='ImageClassifier', + backbone=dict( + frozen_stages=2, + type='ResNet', + depth=50, + num_stages=4, + out_indices=(3, ), + style='pytorch', + init_cfg=dict( + type='Pretrained', + checkpoint= + 'https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth', + prefix='backbone')), + neck=dict(type='GlobalAveragePooling'), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=2048, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + topk=1)) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3fake8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3real8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3fake8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3real8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = dict(type='Accuracy', topk=1) +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3fake8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3real8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = dict(type='Accuracy', topk=1) +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1' + +2023/06/05 20:18:43 - mmengine - INFO - Hooks will be executed in the following order: +before_run: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_load_checkpoint: +(ABOVE_NORMAL) EMAHook + -------------------- +before_train: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_train_epoch: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(NORMAL ) DistSamplerSeedHook + -------------------- +before_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook + -------------------- +after_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +after_train_epoch: +(NORMAL ) IterTimerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_val_epoch: +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook + -------------------- +before_val_iter: +(NORMAL ) IterTimerHook + -------------------- +after_val_iter: +(NORMAL ) IterTimerHook +(NORMAL ) VisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_val_epoch: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_save_checkpoint: +(ABOVE_NORMAL) EMAHook + -------------------- +after_train: +(VERY_LOW ) CheckpointHook + -------------------- +before_test_epoch: +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook + -------------------- +before_test_iter: +(NORMAL ) IterTimerHook + -------------------- +after_test_iter: +(NORMAL ) IterTimerHook +(NORMAL ) VisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_test_epoch: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_run: +(BELOW_NORMAL) LoggerHook + -------------------- +2023/06/05 20:18:54 - mmengine - INFO - load backbone in model from: https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth +Name of parameter - Initialization information + +backbone.conv1.weight - torch.Size([64, 3, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.bn1.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.bn1.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.conv1.weight - torch.Size([64, 64, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.bn1.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.bn1.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.conv2.weight - torch.Size([64, 64, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.bn2.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.bn2.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.conv3.weight - torch.Size([256, 64, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.bn3.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.bn3.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.downsample.0.weight - torch.Size([256, 64, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.downsample.1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.downsample.1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.conv1.weight - torch.Size([64, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.bn1.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.bn1.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.conv2.weight - torch.Size([64, 64, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.bn2.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.bn2.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.conv3.weight - torch.Size([256, 64, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.bn3.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.bn3.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.conv1.weight - torch.Size([64, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.bn1.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.bn1.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.conv2.weight - torch.Size([64, 64, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.bn2.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.bn2.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.conv3.weight - torch.Size([256, 64, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.bn3.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.bn3.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.conv1.weight - torch.Size([128, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.bn1.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.bn1.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.conv2.weight - torch.Size([128, 128, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.bn2.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.bn2.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.conv3.weight - torch.Size([512, 128, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.bn3.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.bn3.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.downsample.0.weight - torch.Size([512, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.downsample.1.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.downsample.1.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.conv1.weight - torch.Size([128, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.bn1.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.bn1.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.conv2.weight - torch.Size([128, 128, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.bn2.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.bn2.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.conv3.weight - torch.Size([512, 128, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.bn3.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.bn3.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.conv1.weight - torch.Size([128, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.bn1.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.bn1.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.conv2.weight - torch.Size([128, 128, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.bn2.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.bn2.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.conv3.weight - torch.Size([512, 128, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.bn3.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.bn3.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.conv1.weight - torch.Size([128, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.bn1.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.bn1.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.conv2.weight - torch.Size([128, 128, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.bn2.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.bn2.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.conv3.weight - torch.Size([512, 128, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.bn3.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.bn3.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.conv1.weight - torch.Size([256, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.bn1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.bn1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.conv2.weight - torch.Size([256, 256, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.bn2.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.bn2.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.conv3.weight - torch.Size([1024, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.bn3.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.bn3.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.downsample.0.weight - torch.Size([1024, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.downsample.1.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.downsample.1.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.conv1.weight - torch.Size([256, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.bn1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.bn1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.conv2.weight - torch.Size([256, 256, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.bn2.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.bn2.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.conv3.weight - torch.Size([1024, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.bn3.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.bn3.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.conv1.weight - torch.Size([256, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.bn1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.bn1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.conv2.weight - torch.Size([256, 256, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.bn2.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.bn2.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.conv3.weight - torch.Size([1024, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.bn3.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.bn3.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.conv1.weight - torch.Size([256, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.bn1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.bn1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.conv2.weight - torch.Size([256, 256, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.bn2.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.bn2.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.conv3.weight - torch.Size([1024, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.bn3.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.bn3.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.conv1.weight - torch.Size([256, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.bn1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.bn1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.conv2.weight - torch.Size([256, 256, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.bn2.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.bn2.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.conv3.weight - torch.Size([1024, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.bn3.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.bn3.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.conv1.weight - torch.Size([256, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.bn1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.bn1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.conv2.weight - torch.Size([256, 256, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.bn2.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.bn2.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.conv3.weight - torch.Size([1024, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.bn3.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.bn3.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.conv1.weight - torch.Size([512, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.bn1.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.bn1.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.conv2.weight - torch.Size([512, 512, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.bn2.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.bn2.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.conv3.weight - torch.Size([2048, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.bn3.weight - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.bn3.bias - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.downsample.0.weight - torch.Size([2048, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.downsample.1.weight - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.downsample.1.bias - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.conv1.weight - torch.Size([512, 2048, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.bn1.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.bn1.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.conv2.weight - torch.Size([512, 512, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.bn2.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.bn2.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.conv3.weight - torch.Size([2048, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.bn3.weight - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.bn3.bias - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.conv1.weight - torch.Size([512, 2048, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.bn1.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.bn1.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.conv2.weight - torch.Size([512, 512, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.bn2.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.bn2.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.conv3.weight - torch.Size([2048, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.bn3.weight - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.bn3.bias - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +head.fc.weight - torch.Size([2, 2048]): +NormalInit: mean=0, std=0.01, bias=0 + +head.fc.bias - torch.Size([2]): +NormalInit: mean=0, std=0.01, bias=0 +2023/06/05 20:18:54 - mmengine - WARNING - "FileClient" will be deprecated in future. Please use io functions in https://mmengine.readthedocs.io/en/latest/api/fileio.html#file-io +2023/06/05 20:18:54 - mmengine - WARNING - "HardDiskBackend" is the alias of "LocalBackend" and the former will be deprecated in future. +2023/06/05 20:18:54 - mmengine - INFO - Checkpoints will be saved to /mnt/petrelfs/luzeyu/workspace/fakebench/mmpretrain/workdir/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1. +2023/06/05 20:20:55 - mmengine - INFO - Epoch(train) [1][100/342] lr: 4.9899e-04 eta: 1:07:02 time: 1.2894 data_time: 0.0462 memory: 9436 loss: 0.6466 +2023/06/05 20:22:52 - mmengine - INFO - Epoch(train) [1][200/342] lr: 4.9592e-04 eta: 1:04:00 time: 1.1408 data_time: 0.0374 memory: 6319 loss: 0.5994 +2023/06/05 20:24:48 - mmengine - INFO - Epoch(train) [1][300/342] lr: 4.9082e-04 eta: 1:01:19 time: 1.1450 data_time: 0.0338 memory: 6319 loss: 0.5529 +2023/06/05 20:25:33 - mmengine - INFO - Exp name: resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1_20230605_201824 +2023/06/05 20:25:33 - mmengine - INFO - Saving checkpoint at 1 epochs +2023/06/05 20:27:32 - mmengine - INFO - Epoch(val) [1][100/342] eta: 0:04:35 time: 0.3524 data_time: 0.2598 memory: 6319 +2023/06/05 20:29:08 - mmengine - INFO - Epoch(val) [1][200/342] eta: 0:02:29 time: 0.2520 data_time: 0.1639 memory: 3133 +2023/06/05 20:31:17 - mmengine - INFO - Epoch(val) [1][300/342] eta: 0:00:47 time: 1.2765 data_time: 1.1885 memory: 3133 +2023/06/05 20:33:48 - mmengine - INFO - Epoch(val) [1][342/342] accuracy/top1: 52.5023 data_time: 1.2363 time: 1.3259 +2023/06/05 20:35:48 - mmengine - INFO - Epoch(train) [2][100/342] lr: 4.8017e-04 eta: 0:58:13 time: 1.1271 data_time: 0.3758 memory: 6319 loss: 0.4926 +2023/06/05 20:37:58 - mmengine - INFO - Epoch(train) [2][200/342] lr: 4.7036e-04 eta: 0:57:23 time: 1.2324 data_time: 0.0280 memory: 6319 loss: 0.4379 +2023/06/05 20:39:51 - mmengine - INFO - Epoch(train) [2][300/342] lr: 4.5874e-04 eta: 0:54:58 time: 1.1658 data_time: 0.0007 memory: 6319 loss: 0.4117 +2023/06/05 20:40:38 - mmengine - INFO - Exp name: resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1_20230605_201824 +2023/06/05 20:40:38 - mmengine - INFO - Saving checkpoint at 2 epochs +2023/06/05 20:42:34 - mmengine - INFO - Epoch(val) [2][100/342] eta: 0:04:25 time: 0.3601 data_time: 0.2723 memory: 6319 +2023/06/05 20:44:11 - mmengine - INFO - Epoch(val) [2][200/342] eta: 0:02:26 time: 0.2166 data_time: 0.1296 memory: 3133 +2023/06/05 20:46:18 - mmengine - INFO - Epoch(val) [2][300/342] eta: 0:00:46 time: 1.1879 data_time: 1.0999 memory: 3133 +2023/06/05 20:48:49 - mmengine - INFO - Epoch(val) [2][342/342] accuracy/top1: 54.9073 data_time: 1.2266 time: 1.3162 +2023/06/05 20:50:50 - mmengine - INFO - Epoch(train) [3][100/342] lr: 4.3931e-04 eta: 0:52:07 time: 1.2552 data_time: 0.5291 memory: 6319 loss: 0.3391 +2023/06/05 20:53:01 - mmengine - INFO - Epoch(train) [3][200/342] lr: 4.2373e-04 eta: 0:50:45 time: 1.1545 data_time: 0.5126 memory: 6319 loss: 0.3379 +2023/06/05 20:54:56 - mmengine - INFO - Epoch(train) [3][300/342] lr: 4.0672e-04 eta: 0:48:33 time: 1.0924 data_time: 0.5283 memory: 6319 loss: 0.3355 +2023/06/05 20:55:20 - mmengine - INFO - Exp name: resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1_20230605_201824 +2023/06/05 20:55:43 - mmengine - INFO - Exp name: resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1_20230605_201824 +2023/06/05 20:55:43 - mmengine - INFO - Saving checkpoint at 3 epochs +2023/06/05 20:57:40 - mmengine - INFO - Epoch(val) [3][100/342] eta: 0:04:28 time: 0.3432 data_time: 0.2546 memory: 6319 +2023/06/05 20:59:18 - mmengine - INFO - Epoch(val) [3][200/342] eta: 0:02:28 time: 0.2374 data_time: 0.1503 memory: 3133 +2023/06/05 21:01:23 - mmengine - INFO - Epoch(val) [3][300/342] eta: 0:00:46 time: 1.1826 data_time: 1.0944 memory: 3133 +2023/06/05 21:04:00 - mmengine - INFO - Epoch(val) [3][342/342] accuracy/top1: 56.8017 data_time: 1.2417 time: 1.3311 +2023/06/05 21:05:59 - mmengine - INFO - Epoch(train) [4][100/342] lr: 3.8041e-04 eta: 0:45:34 time: 1.1012 data_time: 0.6531 memory: 6319 loss: 0.2856 +2023/06/05 21:07:53 - mmengine - INFO - Epoch(train) [4][200/342] lr: 3.6058e-04 eta: 0:43:27 time: 1.1526 data_time: 0.0985 memory: 6319 loss: 0.2939 +2023/06/05 21:09:56 - mmengine - INFO - Epoch(train) [4][300/342] lr: 3.3985e-04 eta: 0:41:35 time: 1.1228 data_time: 0.1639 memory: 6319 loss: 0.2874 +2023/06/05 21:10:41 - mmengine - INFO - Exp name: resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1_20230605_201824 +2023/06/05 21:10:41 - mmengine - INFO - Saving checkpoint at 4 epochs +2023/06/05 21:12:34 - mmengine - INFO - Epoch(val) [4][100/342] eta: 0:04:20 time: 0.3319 data_time: 0.2435 memory: 6319 +2023/06/05 21:14:11 - mmengine - INFO - Epoch(val) [4][200/342] eta: 0:02:25 time: 0.2262 data_time: 0.1397 memory: 3133 +2023/06/05 21:16:16 - mmengine - INFO - Epoch(val) [4][300/342] eta: 0:00:46 time: 1.1651 data_time: 1.0761 memory: 3133 +2023/06/05 21:18:53 - mmengine - INFO - Epoch(val) [4][342/342] accuracy/top1: 58.0077 data_time: 1.2350 time: 1.3242 +2023/06/05 21:21:25 - mmengine - INFO - Epoch(train) [5][100/342] lr: 3.0924e-04 eta: 0:39:22 time: 1.0484 data_time: 0.0007 memory: 6319 loss: 0.2582 +2023/06/05 21:23:20 - mmengine - INFO - Epoch(train) [5][200/342] lr: 2.8709e-04 eta: 0:37:13 time: 1.1158 data_time: 0.0008 memory: 6319 loss: 0.2382 +2023/06/05 21:25:14 - mmengine - INFO - Epoch(train) [5][300/342] lr: 2.6467e-04 eta: 0:35:06 time: 1.2034 data_time: 0.0032 memory: 6319 loss: 0.2546 +2023/06/05 21:25:58 - mmengine - INFO - Exp name: resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1_20230605_201824 +2023/06/05 21:25:58 - mmengine - INFO - Saving checkpoint at 5 epochs +2023/06/05 21:27:51 - mmengine - INFO - Epoch(val) [5][100/342] eta: 0:04:20 time: 0.3397 data_time: 0.2517 memory: 6319 +2023/06/05 21:29:28 - mmengine - INFO - Epoch(val) [5][200/342] eta: 0:02:24 time: 0.2111 data_time: 0.1246 memory: 3133 +2023/06/05 21:31:31 - mmengine - INFO - Epoch(val) [5][300/342] eta: 0:00:45 time: 1.2263 data_time: 1.1381 memory: 3133 +2023/06/05 21:34:04 - mmengine - INFO - Epoch(val) [5][342/342] accuracy/top1: 58.4781 data_time: 1.2147 time: 1.3046 +2023/06/05 21:36:03 - mmengine - INFO - Epoch(train) [6][100/342] lr: 2.3275e-04 eta: 0:32:08 time: 1.2094 data_time: 0.0141 memory: 6319 loss: 0.2229 +2023/06/05 21:37:57 - mmengine - INFO - Epoch(train) [6][200/342] lr: 2.1046e-04 eta: 0:30:04 time: 1.1259 data_time: 0.0032 memory: 6319 loss: 0.2249 +2023/06/05 21:39:40 - mmengine - INFO - Exp name: resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1_20230605_201824 +2023/06/05 21:39:51 - mmengine - INFO - Epoch(train) [6][300/342] lr: 1.8855e-04 eta: 0:28:01 time: 1.1113 data_time: 0.0009 memory: 6319 loss: 0.2035 +2023/06/05 21:40:36 - mmengine - INFO - Exp name: resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1_20230605_201824 +2023/06/05 21:40:36 - mmengine - INFO - Saving checkpoint at 6 epochs +2023/06/05 21:42:29 - mmengine - INFO - Epoch(val) [6][100/342] eta: 0:04:20 time: 0.3264 data_time: 0.2388 memory: 6319 +2023/06/05 21:44:05 - mmengine - INFO - Epoch(val) [6][200/342] eta: 0:02:24 time: 0.2174 data_time: 0.1302 memory: 3133 +2023/06/05 21:46:08 - mmengine - INFO - Epoch(val) [6][300/342] eta: 0:00:45 time: 1.1842 data_time: 1.0961 memory: 3133 +2023/06/05 21:48:58 - mmengine - INFO - Epoch(val) [6][342/342] accuracy/top1: 58.6636 data_time: 1.2616 time: 1.3514 +2023/06/05 21:50:54 - mmengine - INFO - Epoch(train) [7][100/342] lr: 1.5844e-04 eta: 0:25:06 time: 1.1492 data_time: 0.5207 memory: 6319 loss: 0.2196 +2023/06/05 21:52:48 - mmengine - INFO - Epoch(train) [7][200/342] lr: 1.3820e-04 eta: 0:23:05 time: 1.2282 data_time: 0.2014 memory: 6319 loss: 0.2140 +2023/06/05 21:54:57 - mmengine - INFO - Epoch(train) [7][300/342] lr: 1.1893e-04 eta: 0:21:11 time: 1.1372 data_time: 0.0008 memory: 6319 loss: 0.2145 +2023/06/05 21:55:43 - mmengine - INFO - Exp name: resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1_20230605_201824 +2023/06/05 21:55:43 - mmengine - INFO - Saving checkpoint at 7 epochs +2023/06/05 21:57:37 - mmengine - INFO - Epoch(val) [7][100/342] eta: 0:04:21 time: 0.3393 data_time: 0.2504 memory: 6319 +2023/06/05 21:59:15 - mmengine - INFO - Epoch(val) [7][200/342] eta: 0:02:25 time: 0.2251 data_time: 0.1374 memory: 3133 +2023/06/05 22:01:14 - mmengine - INFO - Epoch(val) [7][300/342] eta: 0:00:45 time: 1.1680 data_time: 1.0796 memory: 3133 +2023/06/05 22:03:49 - mmengine - INFO - Epoch(val) [7][342/342] accuracy/top1: 57.8239 data_time: 1.2072 time: 1.2968 +2023/06/05 22:05:46 - mmengine - INFO - Epoch(train) [8][100/342] lr: 9.3587e-05 eta: 0:18:20 time: 1.0941 data_time: 0.0016 memory: 6319 loss: 0.2031 +2023/06/05 22:07:44 - mmengine - INFO - Epoch(train) [8][200/342] lr: 7.7361e-05 eta: 0:16:21 time: 1.1903 data_time: 0.0012 memory: 6319 loss: 0.2023 +2023/06/05 22:09:39 - mmengine - INFO - Epoch(train) [8][300/342] lr: 6.2632e-05 eta: 0:14:21 time: 1.2062 data_time: 0.0015 memory: 6319 loss: 0.2108 +2023/06/05 22:10:25 - mmengine - INFO - Exp name: resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1_20230605_201824 +2023/06/05 22:10:25 - mmengine - INFO - Saving checkpoint at 8 epochs +2023/06/05 22:12:20 - mmengine - INFO - Epoch(val) [8][100/342] eta: 0:04:23 time: 0.3376 data_time: 0.2488 memory: 6319 +2023/06/05 22:13:59 - mmengine - INFO - Epoch(val) [8][200/342] eta: 0:02:27 time: 0.2099 data_time: 0.1225 memory: 3133 +2023/06/05 22:16:02 - mmengine - INFO - Epoch(val) [8][300/342] eta: 0:00:46 time: 1.1863 data_time: 1.0969 memory: 3133 +2023/06/05 22:18:34 - mmengine - INFO - Epoch(val) [8][342/342] accuracy/top1: 57.4348 data_time: 1.2203 time: 1.3109 +2023/06/05 22:20:30 - mmengine - INFO - Epoch(train) [9][100/342] lr: 4.4532e-05 eta: 0:11:31 time: 1.1255 data_time: 0.2349 memory: 6319 loss: 0.2043 +2023/06/05 22:22:29 - mmengine - INFO - Epoch(train) [9][200/342] lr: 3.3915e-05 eta: 0:09:33 time: 1.1331 data_time: 0.0008 memory: 6319 loss: 0.2061 +2023/06/05 22:23:44 - mmengine - INFO - Exp name: resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1_20230605_201824 +2023/06/05 22:24:22 - mmengine - INFO - Epoch(train) [9][300/342] lr: 2.5163e-05 eta: 0:07:34 time: 1.2186 data_time: 0.0009 memory: 6319 loss: 0.1880 +2023/06/05 22:25:10 - mmengine - INFO - Exp name: resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1_20230605_201824 +2023/06/05 22:25:10 - mmengine - INFO - Saving checkpoint at 9 epochs +2023/06/05 22:27:03 - mmengine - INFO - Epoch(val) [9][100/342] eta: 0:04:20 time: 0.3659 data_time: 0.2787 memory: 6319 +2023/06/05 22:28:40 - mmengine - INFO - Epoch(val) [9][200/342] eta: 0:02:25 time: 0.2184 data_time: 0.1318 memory: 3133 +2023/06/05 22:30:43 - mmengine - INFO - Epoch(val) [9][300/342] eta: 0:00:45 time: 1.2018 data_time: 1.1051 memory: 3133 +2023/06/05 22:33:16 - mmengine - INFO - Epoch(val) [9][342/342] accuracy/top1: 57.2224 data_time: 1.2147 time: 1.3041 +2023/06/05 22:35:11 - mmengine - INFO - Epoch(train) [10][100/342] lr: 1.6078e-05 eta: 0:04:45 time: 1.1201 data_time: 0.1549 memory: 6319 loss: 0.1912 +2023/06/05 22:37:04 - mmengine - INFO - Epoch(train) [10][200/342] lr: 1.2111e-05 eta: 0:02:47 time: 1.1655 data_time: 0.0007 memory: 6319 loss: 0.2117 +2023/06/05 22:38:57 - mmengine - INFO - Epoch(train) [10][300/342] lr: 1.0191e-05 eta: 0:00:49 time: 1.0937 data_time: 0.0011 memory: 6319 loss: 0.1982 +2023/06/05 22:39:49 - mmengine - INFO - Exp name: resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1_20230605_201824 +2023/06/05 22:39:49 - mmengine - INFO - Saving checkpoint at 10 epochs +2023/06/05 22:41:43 - mmengine - INFO - Epoch(val) [10][100/342] eta: 0:04:22 time: 0.3482 data_time: 0.2601 memory: 6319 +2023/06/05 22:43:21 - mmengine - INFO - Epoch(val) [10][200/342] eta: 0:02:26 time: 0.2168 data_time: 0.1303 memory: 3133 +2023/06/05 22:45:24 - mmengine - INFO - Epoch(val) [10][300/342] eta: 0:00:46 time: 1.2010 data_time: 1.1128 memory: 3133 +2023/06/05 22:47:58 - mmengine - INFO - Epoch(val) [10][342/342] accuracy/top1: 56.7966 data_time: 1.2155 time: 1.3052 diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/20230605_201824.json b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/20230605_201824.json new file mode 100644 index 0000000000000000000000000000000000000000..1dd31e3ff3824802ee15c3411574561c69a6a9bb --- /dev/null +++ b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/20230605_201824.json @@ -0,0 +1,40 @@ +{"lr": 0.0004989875940690582, "data_time": 0.04618525505065918, "loss": 0.6466095805168152, "time": 1.2894049167633057, "epoch": 1, "memory": 9436, "step": 100} +{"lr": 0.0004959179335073666, "data_time": 0.037417149543762206, "loss": 0.5993919551372529, "time": 1.1407800674438477, "epoch": 1, "memory": 6319, "step": 200} +{"lr": 0.0004908167990123805, "data_time": 0.03383724689483643, "loss": 0.5529384315013885, "time": 1.1449747562408448, "epoch": 1, "memory": 6319, "step": 300} +{"accuracy/top1": 52.50229263305664, "data_time": 1.2362775990837498, "time": 1.325923048264799, "step": 1} +{"lr": 0.00048017042254960683, "data_time": 0.375758957862854, "loss": 0.4925939321517944, "time": 1.1270631074905395, "epoch": 2, "memory": 6319, "step": 442} +{"lr": 0.0004703638635189273, "data_time": 0.02797234058380127, "loss": 0.43789932429790496, "time": 1.2324337244033814, "epoch": 2, "memory": 6319, "step": 542} +{"lr": 0.00045874130819215336, "data_time": 0.0006974935531616211, "loss": 0.411741504073143, "time": 1.1657596111297608, "epoch": 2, "memory": 6319, "step": 642} +{"accuracy/top1": 54.907283782958984, "data_time": 1.2266088713709884, "time": 1.3161629441875757, "step": 2} +{"lr": 0.000439312001216421, "data_time": 0.5290867805480957, "loss": 0.3391281634569168, "time": 1.2551673650741577, "epoch": 3, "memory": 6319, "step": 784} +{"lr": 0.00042372847804098874, "data_time": 0.5125674724578857, "loss": 0.337921142578125, "time": 1.1545050859451294, "epoch": 3, "memory": 6319, "step": 884} +{"lr": 0.0004067221985769131, "data_time": 0.5282731533050538, "loss": 0.33552381694316863, "time": 1.0923629999160767, "epoch": 3, "memory": 6319, "step": 984} +{"accuracy/top1": 56.80174255371094, "data_time": 1.2417100045840872, "time": 1.3311393691916507, "step": 3} +{"lr": 0.0003804118370269479, "data_time": 0.6530909538269043, "loss": 0.2856471180915833, "time": 1.1011572360992432, "epoch": 4, "memory": 6319, "step": 1126} +{"lr": 0.00036057677353196443, "data_time": 0.09853157997131348, "loss": 0.29394196420907975, "time": 1.1526427030563355, "epoch": 4, "memory": 6319, "step": 1226} +{"lr": 0.00033985146305424703, "data_time": 0.16394495964050293, "loss": 0.2873724430799484, "time": 1.1228408813476562, "epoch": 4, "memory": 6319, "step": 1326} +{"accuracy/top1": 58.00767135620117, "data_time": 1.2349844861656167, "time": 1.3242476021235607, "step": 4} +{"lr": 0.0003092354884336273, "data_time": 0.0007134914398193359, "loss": 0.25816421359777453, "time": 1.0484312772750854, "epoch": 5, "memory": 6319, "step": 1468} +{"lr": 0.00028709047883299587, "data_time": 0.0008438825607299805, "loss": 0.23823048174381256, "time": 1.1158048391342164, "epoch": 5, "memory": 6319, "step": 1568} +{"lr": 0.0002646748751329253, "data_time": 0.0032169342041015623, "loss": 0.25457300394773485, "time": 1.203367280960083, "epoch": 5, "memory": 6319, "step": 1668} +{"accuracy/top1": 58.4781379699707, "data_time": 1.2147304664200318, "time": 1.3045718822812895, "step": 5} +{"lr": 0.00023275019235155644, "data_time": 0.014092826843261718, "loss": 0.2228940784931183, "time": 1.209360957145691, "epoch": 6, "memory": 6319, "step": 1810} +{"lr": 0.00021046294447834082, "data_time": 0.003190898895263672, "loss": 0.22490936517715454, "time": 1.125851082801819, "epoch": 6, "memory": 6319, "step": 1910} +{"lr": 0.00018855124302477407, "data_time": 0.0009340047836303711, "loss": 0.20353734046220778, "time": 1.111295962333679, "epoch": 6, "memory": 6319, "step": 2010} +{"accuracy/top1": 58.663578033447266, "data_time": 1.261570965235852, "time": 1.3513825113502258, "step": 6} +{"lr": 0.0001584428624656395, "data_time": 0.5207465887069702, "loss": 0.219617123901844, "time": 1.1492493152618408, "epoch": 7, "memory": 6319, "step": 2152} +{"lr": 0.00013819500742605714, "data_time": 0.20136706829071044, "loss": 0.2140459045767784, "time": 1.2281822443008423, "epoch": 7, "memory": 6319, "step": 2252} +{"lr": 0.00011893207822507105, "data_time": 0.0008205652236938476, "loss": 0.21445338279008866, "time": 1.137204885482788, "epoch": 7, "memory": 6319, "step": 2352} +{"accuracy/top1": 57.82394790649414, "data_time": 1.2071522554225198, "time": 1.2967924272353726, "step": 7} +{"lr": 9.35872179547221e-05, "data_time": 0.001580953598022461, "loss": 0.20313074439764023, "time": 1.0940540313720704, "epoch": 8, "memory": 6319, "step": 2494} +{"lr": 7.736075687514847e-05, "data_time": 0.0011932134628295898, "loss": 0.2022942677140236, "time": 1.1903136491775512, "epoch": 8, "memory": 6319, "step": 2594} +{"lr": 6.263218964965453e-05, "data_time": 0.0014899253845214843, "loss": 0.2107890158891678, "time": 1.206168532371521, "epoch": 8, "memory": 6319, "step": 2694} +{"accuracy/top1": 57.43475341796875, "data_time": 1.2203194003758555, "time": 1.310949471532082, "step": 8} +{"lr": 4.453178117937791e-05, "data_time": 0.234869647026062, "loss": 0.20425462871789932, "time": 1.125475001335144, "epoch": 9, "memory": 6319, "step": 2836} +{"lr": 3.391507312668422e-05, "data_time": 0.0007771730422973632, "loss": 0.20605545192956926, "time": 1.1331078290939331, "epoch": 9, "memory": 6319, "step": 2936} +{"lr": 2.5162602656676073e-05, "data_time": 0.0009218931198120117, "loss": 0.18800390362739564, "time": 1.2185774803161622, "epoch": 9, "memory": 6319, "step": 3036} +{"accuracy/top1": 57.222415924072266, "data_time": 1.2146767958260138, "time": 1.3041407776991063, "step": 9} +{"lr": 1.607844008050381e-05, "data_time": 0.1549124002456665, "loss": 0.19119564443826675, "time": 1.1201043605804444, "epoch": 10, "memory": 6319, "step": 3178} +{"lr": 1.2110722409842075e-05, "data_time": 0.0007314205169677735, "loss": 0.21169835329055786, "time": 1.1654816389083862, "epoch": 10, "memory": 6319, "step": 3278} +{"lr": 1.019110148697161e-05, "data_time": 0.0010805130004882812, "loss": 0.19820658564567567, "time": 1.093675184249878, "epoch": 10, "memory": 6319, "step": 3378} +{"accuracy/top1": 56.796592712402344, "data_time": 1.2154598110966364, "time": 1.3052300419821337, "step": 10} diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/config.py b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/config.py new file mode 100644 index 0000000000000000000000000000000000000000..86eb2c86d955d97ec259c3c8897f93ba1a2234a9 --- /dev/null +++ b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/config.py @@ -0,0 +1,230 @@ +optim_wrapper = dict( + optimizer=dict( + type='SGD', + lr=0.0005, + momentum=0.9, + weight_decay=0.0001, + _scope_='mmpretrain'), + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=False, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=512) +model = dict( + type='ImageClassifier', + backbone=dict( + frozen_stages=2, + type='ResNet', + depth=50, + num_stages=4, + out_indices=(3, ), + style='pytorch', + init_cfg=dict( + type='Pretrained', + checkpoint= + 'https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth', + prefix='backbone')), + neck=dict(type='GlobalAveragePooling'), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=2048, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + topk=1)) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3fake8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3real8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3fake8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3real8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = dict(type='Accuracy', topk=1) +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3fake8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3real8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = dict(type='Accuracy', topk=1) +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1' diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/events.out.tfevents.1685967518.SH-IDC1-10-140-24-118.86374.0 b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/events.out.tfevents.1685967518.SH-IDC1-10-140-24-118.86374.0 new file mode 100644 index 0000000000000000000000000000000000000000..906581b7afdaa299a00f1b07208106c68887ab43 --- /dev/null +++ b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/events.out.tfevents.1685967518.SH-IDC1-10-140-24-118.86374.0 @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:8a32c47e6dbf713533176c7e98706b47f4d29b6a4235865d1204b3b8e3b9c8e7 +size 14376131 diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/scalars.json b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/scalars.json new file mode 100644 index 0000000000000000000000000000000000000000..1dd31e3ff3824802ee15c3411574561c69a6a9bb --- /dev/null +++ b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/scalars.json @@ -0,0 +1,40 @@ +{"lr": 0.0004989875940690582, "data_time": 0.04618525505065918, "loss": 0.6466095805168152, "time": 1.2894049167633057, "epoch": 1, "memory": 9436, "step": 100} +{"lr": 0.0004959179335073666, "data_time": 0.037417149543762206, "loss": 0.5993919551372529, "time": 1.1407800674438477, "epoch": 1, "memory": 6319, "step": 200} +{"lr": 0.0004908167990123805, "data_time": 0.03383724689483643, "loss": 0.5529384315013885, "time": 1.1449747562408448, "epoch": 1, "memory": 6319, "step": 300} +{"accuracy/top1": 52.50229263305664, "data_time": 1.2362775990837498, "time": 1.325923048264799, "step": 1} +{"lr": 0.00048017042254960683, "data_time": 0.375758957862854, "loss": 0.4925939321517944, "time": 1.1270631074905395, "epoch": 2, "memory": 6319, "step": 442} +{"lr": 0.0004703638635189273, "data_time": 0.02797234058380127, "loss": 0.43789932429790496, "time": 1.2324337244033814, "epoch": 2, "memory": 6319, "step": 542} +{"lr": 0.00045874130819215336, "data_time": 0.0006974935531616211, "loss": 0.411741504073143, "time": 1.1657596111297608, "epoch": 2, "memory": 6319, "step": 642} +{"accuracy/top1": 54.907283782958984, "data_time": 1.2266088713709884, "time": 1.3161629441875757, "step": 2} +{"lr": 0.000439312001216421, "data_time": 0.5290867805480957, "loss": 0.3391281634569168, "time": 1.2551673650741577, "epoch": 3, "memory": 6319, "step": 784} +{"lr": 0.00042372847804098874, "data_time": 0.5125674724578857, "loss": 0.337921142578125, "time": 1.1545050859451294, "epoch": 3, "memory": 6319, "step": 884} +{"lr": 0.0004067221985769131, "data_time": 0.5282731533050538, "loss": 0.33552381694316863, "time": 1.0923629999160767, "epoch": 3, "memory": 6319, "step": 984} +{"accuracy/top1": 56.80174255371094, "data_time": 1.2417100045840872, "time": 1.3311393691916507, "step": 3} +{"lr": 0.0003804118370269479, "data_time": 0.6530909538269043, "loss": 0.2856471180915833, "time": 1.1011572360992432, "epoch": 4, "memory": 6319, "step": 1126} +{"lr": 0.00036057677353196443, "data_time": 0.09853157997131348, "loss": 0.29394196420907975, "time": 1.1526427030563355, "epoch": 4, "memory": 6319, "step": 1226} +{"lr": 0.00033985146305424703, "data_time": 0.16394495964050293, "loss": 0.2873724430799484, "time": 1.1228408813476562, "epoch": 4, "memory": 6319, "step": 1326} +{"accuracy/top1": 58.00767135620117, "data_time": 1.2349844861656167, "time": 1.3242476021235607, "step": 4} +{"lr": 0.0003092354884336273, "data_time": 0.0007134914398193359, "loss": 0.25816421359777453, "time": 1.0484312772750854, "epoch": 5, "memory": 6319, "step": 1468} +{"lr": 0.00028709047883299587, "data_time": 0.0008438825607299805, "loss": 0.23823048174381256, "time": 1.1158048391342164, "epoch": 5, "memory": 6319, "step": 1568} +{"lr": 0.0002646748751329253, "data_time": 0.0032169342041015623, "loss": 0.25457300394773485, "time": 1.203367280960083, "epoch": 5, "memory": 6319, "step": 1668} +{"accuracy/top1": 58.4781379699707, "data_time": 1.2147304664200318, "time": 1.3045718822812895, "step": 5} +{"lr": 0.00023275019235155644, "data_time": 0.014092826843261718, "loss": 0.2228940784931183, "time": 1.209360957145691, "epoch": 6, "memory": 6319, "step": 1810} +{"lr": 0.00021046294447834082, "data_time": 0.003190898895263672, "loss": 0.22490936517715454, "time": 1.125851082801819, "epoch": 6, "memory": 6319, "step": 1910} +{"lr": 0.00018855124302477407, "data_time": 0.0009340047836303711, "loss": 0.20353734046220778, "time": 1.111295962333679, "epoch": 6, "memory": 6319, "step": 2010} +{"accuracy/top1": 58.663578033447266, "data_time": 1.261570965235852, "time": 1.3513825113502258, "step": 6} +{"lr": 0.0001584428624656395, "data_time": 0.5207465887069702, "loss": 0.219617123901844, "time": 1.1492493152618408, "epoch": 7, "memory": 6319, "step": 2152} +{"lr": 0.00013819500742605714, "data_time": 0.20136706829071044, "loss": 0.2140459045767784, "time": 1.2281822443008423, "epoch": 7, "memory": 6319, "step": 2252} +{"lr": 0.00011893207822507105, "data_time": 0.0008205652236938476, "loss": 0.21445338279008866, "time": 1.137204885482788, "epoch": 7, "memory": 6319, "step": 2352} +{"accuracy/top1": 57.82394790649414, "data_time": 1.2071522554225198, "time": 1.2967924272353726, "step": 7} +{"lr": 9.35872179547221e-05, "data_time": 0.001580953598022461, "loss": 0.20313074439764023, "time": 1.0940540313720704, "epoch": 8, "memory": 6319, "step": 2494} +{"lr": 7.736075687514847e-05, "data_time": 0.0011932134628295898, "loss": 0.2022942677140236, "time": 1.1903136491775512, "epoch": 8, "memory": 6319, "step": 2594} +{"lr": 6.263218964965453e-05, "data_time": 0.0014899253845214843, "loss": 0.2107890158891678, "time": 1.206168532371521, "epoch": 8, "memory": 6319, "step": 2694} +{"accuracy/top1": 57.43475341796875, "data_time": 1.2203194003758555, "time": 1.310949471532082, "step": 8} +{"lr": 4.453178117937791e-05, "data_time": 0.234869647026062, "loss": 0.20425462871789932, "time": 1.125475001335144, "epoch": 9, "memory": 6319, "step": 2836} +{"lr": 3.391507312668422e-05, "data_time": 0.0007771730422973632, "loss": 0.20605545192956926, "time": 1.1331078290939331, "epoch": 9, "memory": 6319, "step": 2936} +{"lr": 2.5162602656676073e-05, "data_time": 0.0009218931198120117, "loss": 0.18800390362739564, "time": 1.2185774803161622, "epoch": 9, "memory": 6319, "step": 3036} +{"accuracy/top1": 57.222415924072266, "data_time": 1.2146767958260138, "time": 1.3041407776991063, "step": 9} +{"lr": 1.607844008050381e-05, "data_time": 0.1549124002456665, "loss": 0.19119564443826675, "time": 1.1201043605804444, "epoch": 10, "memory": 6319, "step": 3178} +{"lr": 1.2110722409842075e-05, "data_time": 0.0007314205169677735, "loss": 0.21169835329055786, "time": 1.1654816389083862, "epoch": 10, "memory": 6319, "step": 3278} +{"lr": 1.019110148697161e-05, "data_time": 0.0010805130004882812, "loss": 0.19820658564567567, "time": 1.093675184249878, "epoch": 10, "memory": 6319, "step": 3378} +{"accuracy/top1": 56.796592712402344, "data_time": 1.2154598110966364, "time": 1.3052300419821337, "step": 10} diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/35624.png_1.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/35624.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..87505dbde103cb1a2aa04a1807311f14c5d1c2de Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/35624.png_1.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/35624.png_10.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/35624.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..c673f40bc20487b63c76a1a716cff036ec9fc644 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/35624.png_10.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/35624.png_2.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/35624.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..191cd19dafcdbab9e33ca3cb1397406cc4df88a0 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/35624.png_2.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/35624.png_3.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/35624.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..a98308b0ee57c78b8a54db06f4edf8463b531dcd Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/35624.png_3.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/35624.png_4.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/35624.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..063673e6999231915d68fd142a9ccd45b97698af Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/35624.png_4.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/35624.png_5.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/35624.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..21b2c3733f9646b75e8181952145b0c138b63d01 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/35624.png_5.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/35624.png_6.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/35624.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..04525047760260743911992b20d9c765d3ff632e Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/35624.png_6.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/35624.png_7.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/35624.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..bbc82628c5960ffe5d19b5078bf276e5850d29ed Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/35624.png_7.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/35624.png_8.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/35624.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..0149cce1bafd989c1959cf526c88bdaa5a33e4ba Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/35624.png_8.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/35624.png_9.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/35624.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..4309cf667a7cb36bb6db40e1f57ffbbf3086d7f6 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/35624.png_9.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/38442.png_1.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/38442.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..ac4683bb741bed6fbaa35618e45c9d94aa3d482d Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/38442.png_1.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/38442.png_10.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/38442.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..f4e7a7bf4085c61279045dfe8eef5e2098a43c58 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/38442.png_10.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/38442.png_2.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/38442.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..7c6eeb7967ee35f3484dcaca4930602c692a878e Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/38442.png_2.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/38442.png_3.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/38442.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..f7e14ade238f94fdb952a65f29c16b0a63e5aa9b Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/38442.png_3.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/38442.png_4.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/38442.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..9df979c5f8305fdf3ff6aca3238a6687fe8a781d Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/38442.png_4.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/38442.png_5.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/38442.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..62be838b82e547c910e4cf3fd3cf9cb85518093e Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/38442.png_5.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/38442.png_6.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/38442.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..b708f01a4b49debd88697fa649b6be263141cc4a Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/38442.png_6.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/38442.png_7.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/38442.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..da4dc24c1d579781194520900281a144498833fb Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/38442.png_7.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/38442.png_8.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/38442.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..6c32fe3f1723e10f7e8ee7cc9cc7d88a6995d0a1 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/38442.png_8.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/38442.png_9.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/38442.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..9b6d59c8c17f7fe9981ce983f6654818ffce46f2 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/38442.png_9.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/43283.png_1.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/43283.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..2ef13bf1147f11614bc71215e146cca4ec62d673 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/43283.png_1.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/43283.png_10.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/43283.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..66f16136e2ea7bd09766efd0d3889550f8ad5b47 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/43283.png_10.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/43283.png_2.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/43283.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..013b4b6752345915c515f9389ad2f4dce5ff7140 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/43283.png_2.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/43283.png_3.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/43283.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..495fde9374a2c356424b7dd188c58844b71e973b Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/43283.png_3.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/43283.png_4.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/43283.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..e7924f7e7d6d34020060505ae4ad24d64f9d0415 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/43283.png_4.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/43283.png_5.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/43283.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..304f335f9611ee3f06ebdae54ab33b48e2b5c945 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/43283.png_5.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/43283.png_6.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/43283.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..3ea98b214dacb6bef26e9c9b1e61375cc442ac7c Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/43283.png_6.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/43283.png_7.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/43283.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..ebcdabbb1e3da5b5c8b58b90dd5a686311eca800 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/43283.png_7.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/43283.png_8.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/43283.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..708bb7eb9b9262b749091b9d06bb9fdb12a7f40a Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/43283.png_8.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/43283.png_9.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/43283.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..b920c71bc244deef8f6887c93e340fab4920f7bd Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/43283.png_9.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/44982.png_1.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/44982.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..ac58b37f3895a6684e327b0acb44c09ab7d8d06c Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/44982.png_1.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/44982.png_10.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/44982.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..1cf13aeb2c4c8d6ef6a8f1c9ebe51fa684944d76 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/44982.png_10.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/44982.png_2.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/44982.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..1ef53443f9d4812a72e57cc2cac258a56c949fc0 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/44982.png_2.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/44982.png_3.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/44982.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..c47c618a5cc4dd9cd2d18eff720942c8f86211c4 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/44982.png_3.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/44982.png_4.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/44982.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..5ae95b913ecf7bcf8b935b462f7223b399061995 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/44982.png_4.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/44982.png_5.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/44982.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..9fb9227eb4ef123d9c5ba50fed99bf72b08fa0e1 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/44982.png_5.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/44982.png_6.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/44982.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..72c4654fa105d6017d5734a05006c6c364e409cb Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/44982.png_6.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/44982.png_7.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/44982.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..5d3c2f7e82b36bbadb89cd83e2c73104389e2b93 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/44982.png_7.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/44982.png_8.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/44982.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..ad99442feb2e9e8fa5f61563669fda105d9e9277 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/44982.png_8.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/44982.png_9.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/44982.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..2e69e0d9895180d4e59008f68481e78efcb9c83f Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/44982.png_9.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/56423.png_1.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/56423.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..c134bf9beb9a99d614d0435a24d1b6a9f85c6b9b Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/56423.png_1.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/56423.png_10.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/56423.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..d5e502a3b08d0c35ba9db75bed0630bc227ba7c2 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/56423.png_10.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/56423.png_2.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/56423.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..dcdb05923859d4b626f0b30fb06bebe61dd9ab93 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/56423.png_2.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/56423.png_3.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/56423.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..744e5c738ec7271e2e5ff70a40318b80d6df7926 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/56423.png_3.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/56423.png_4.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/56423.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..4d4dc10aa92b382bfbe431cccecd114a6ad42dec Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/56423.png_4.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/56423.png_5.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/56423.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..a1fe9f7f154bc2c42ba5b5dad101cd01ce63244a Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/56423.png_5.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/56423.png_6.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/56423.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..d7d785ab1c420ceb45114b047e1c912a72fe1501 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/56423.png_6.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/56423.png_7.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/56423.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..f9ce5912faf6696862f6ca1bc4d1815e1ef0a37c Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/56423.png_7.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/56423.png_8.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/56423.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..9ed53da00956620ab9234dd6a04a59c13eabd132 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/56423.png_8.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/56423.png_9.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/56423.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..bf5a7b3021325fc5c7564e883876977e1750a778 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/56423.png_9.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/61024.png_1.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/61024.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..44beeb916c0211b35fa2df500b804e178b3109aa Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/61024.png_1.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/61024.png_10.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/61024.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..5adc43abe7db40fc14cc621dec81e367fde0c40d Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/61024.png_10.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/61024.png_2.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/61024.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..80afd1a692c5d763d256d8f958d8b19624609dc2 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/61024.png_2.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/61024.png_3.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/61024.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..2f2e1e5cbf64d44e960f7772f21ccdf3f2b2e925 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/61024.png_3.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/61024.png_4.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/61024.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..ccbb29380eb712665e7a85c96f3e86bd1a80d405 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/61024.png_4.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/61024.png_5.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/61024.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..a476aa917ffbf6378fa4a9291ee054dfbbd77d56 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/61024.png_5.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/61024.png_6.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/61024.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..64154a2431a29cd7c842fd75e1418b2a0fcf30ca Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/61024.png_6.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/61024.png_7.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/61024.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..89897cfc74b7e74f708c34e3f54023ecdc0b02d8 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/61024.png_7.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/61024.png_8.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/61024.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..c506f5ce2c38b374a3c795eb63a4adafd58775ac Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/61024.png_8.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/61024.png_9.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/61024.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..41912a9c0d88f170e032941fec6ea1a701cc743c Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/61024.png_9.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/66390.png_1.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/66390.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..db511ad1e71eb2978e51ace0afdc55f903b2d1b9 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/66390.png_1.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/66390.png_10.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/66390.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..7e2c22878a6210426a1adcd9997e5f9e52c86dca Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/66390.png_10.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/66390.png_2.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/66390.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..732720b8a8373283014d88a3fb2e4b2790a2da5c Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/66390.png_2.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/66390.png_3.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/66390.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..c03b610800a1d54be6138aa48a310461432f7f0d Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/66390.png_3.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/66390.png_4.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/66390.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..acf4aba09aedcea57c98d5fe9c2ae787bd068926 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/66390.png_4.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/66390.png_5.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/66390.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..75e2b22a481c82363762df78dd9f47adaa8a7cff Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/66390.png_5.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/66390.png_6.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/66390.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..c1ade768af95eda3c47facca7b789d903bb68a12 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/66390.png_6.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/66390.png_7.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/66390.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..e29e4eec38c5458d63b7a5039fa91d03ca7e8b3e Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/66390.png_7.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/66390.png_8.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/66390.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..7d25fb25f39bcf0a6e531c665dffadcc03d9e4e2 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/66390.png_8.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/66390.png_9.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/66390.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..d3163518c94ccbd251136495bc54e56d265b58dd Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/66390.png_9.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/flickr_wild_000179.jpg_1.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/flickr_wild_000179.jpg_1.png new file mode 100644 index 0000000000000000000000000000000000000000..36f13e20466b160bb942a7744bd4a6948917e47a Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/flickr_wild_000179.jpg_1.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/flickr_wild_000179.jpg_10.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/flickr_wild_000179.jpg_10.png new file mode 100644 index 0000000000000000000000000000000000000000..c99d122b0faea01109a00e9fcbd5a5cc5507f98b Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/flickr_wild_000179.jpg_10.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/flickr_wild_000179.jpg_2.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/flickr_wild_000179.jpg_2.png new file mode 100644 index 0000000000000000000000000000000000000000..f219a89b074b4e89cd154dcd31e6749b615356a5 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/flickr_wild_000179.jpg_2.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/flickr_wild_000179.jpg_3.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/flickr_wild_000179.jpg_3.png new file mode 100644 index 0000000000000000000000000000000000000000..385b8258edc9f7464416f7fc4d29d14b0d75f9fd Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/flickr_wild_000179.jpg_3.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/flickr_wild_000179.jpg_4.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/flickr_wild_000179.jpg_4.png new file mode 100644 index 0000000000000000000000000000000000000000..686049af4411b3c5c06895f1267b335dd9eb99e9 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/flickr_wild_000179.jpg_4.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/flickr_wild_000179.jpg_5.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/flickr_wild_000179.jpg_5.png new file mode 100644 index 0000000000000000000000000000000000000000..f4b725d539cff6d2ca2491815e19948436088eb1 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/flickr_wild_000179.jpg_5.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/flickr_wild_000179.jpg_6.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/flickr_wild_000179.jpg_6.png new file mode 100644 index 0000000000000000000000000000000000000000..ec80e06ac0d37d4a89fabf883d0ed78ff8ec6989 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/flickr_wild_000179.jpg_6.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/flickr_wild_000179.jpg_7.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/flickr_wild_000179.jpg_7.png new file mode 100644 index 0000000000000000000000000000000000000000..909023077949870a8950d844d944c1d49ab0db7a Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/flickr_wild_000179.jpg_7.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/flickr_wild_000179.jpg_8.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/flickr_wild_000179.jpg_8.png new file mode 100644 index 0000000000000000000000000000000000000000..03dad56c9e8686326ce189d86039b969f938e355 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/flickr_wild_000179.jpg_8.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/flickr_wild_000179.jpg_9.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/flickr_wild_000179.jpg_9.png new file mode 100644 index 0000000000000000000000000000000000000000..1bb5534d65e713039ec96bad0f6422dcfe90bfc9 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/flickr_wild_000179.jpg_9.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/pixabay_dog_002352.jpg_1.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/pixabay_dog_002352.jpg_1.png new file mode 100644 index 0000000000000000000000000000000000000000..f8b7e57c212d3971582f6daef73041f56d2d17e1 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/pixabay_dog_002352.jpg_1.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/pixabay_dog_002352.jpg_10.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/pixabay_dog_002352.jpg_10.png new file mode 100644 index 0000000000000000000000000000000000000000..3987680d3f9cba94e47448aca11940a87df8cd27 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/pixabay_dog_002352.jpg_10.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/pixabay_dog_002352.jpg_2.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/pixabay_dog_002352.jpg_2.png new file mode 100644 index 0000000000000000000000000000000000000000..7814508a1e841f23c5f3aa675bb261eb16126f92 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/pixabay_dog_002352.jpg_2.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/pixabay_dog_002352.jpg_3.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/pixabay_dog_002352.jpg_3.png new file mode 100644 index 0000000000000000000000000000000000000000..5b8130443e1194d86a9c6f69559863f1778b233c Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/pixabay_dog_002352.jpg_3.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/pixabay_dog_002352.jpg_4.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/pixabay_dog_002352.jpg_4.png new file mode 100644 index 0000000000000000000000000000000000000000..205b5c6aeb925ea59b46fe9e477173b3468075c3 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/pixabay_dog_002352.jpg_4.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/pixabay_dog_002352.jpg_5.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/pixabay_dog_002352.jpg_5.png new file mode 100644 index 0000000000000000000000000000000000000000..a03a7ec0e8c4b86b07b5fceb9eb5e81ac4528376 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/pixabay_dog_002352.jpg_5.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/pixabay_dog_002352.jpg_6.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/pixabay_dog_002352.jpg_6.png new file mode 100644 index 0000000000000000000000000000000000000000..039ded2b9ebd6d2c094c873dd794db7a9d1c4f43 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/pixabay_dog_002352.jpg_6.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/pixabay_dog_002352.jpg_7.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/pixabay_dog_002352.jpg_7.png new file mode 100644 index 0000000000000000000000000000000000000000..5f03f861add5af4a9f3b2192bcbcf615d9ba205f Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/pixabay_dog_002352.jpg_7.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/pixabay_dog_002352.jpg_8.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/pixabay_dog_002352.jpg_8.png new file mode 100644 index 0000000000000000000000000000000000000000..1974798213db7d727e9e9901729967919e5b19b0 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/pixabay_dog_002352.jpg_8.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/pixabay_dog_002352.jpg_9.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/pixabay_dog_002352.jpg_9.png new file mode 100644 index 0000000000000000000000000000000000000000..104eb65720799a2c57146b126ece7456a5cf7476 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/pixabay_dog_002352.jpg_9.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed14088.png_1.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed14088.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..01f4c1136e721dcaaa01af9c5757b3b5591cbafe Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed14088.png_1.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed14088.png_10.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed14088.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..d960680fda4dd6553b63a3830685e2933ac5e1eb Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed14088.png_10.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed14088.png_2.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed14088.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..574ccaf2038b14d4e0f8d9075724be803e7c2c79 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed14088.png_2.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed14088.png_3.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed14088.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..b52d8e640b373547ac86d6deffbe9fa5950d392e Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed14088.png_3.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed14088.png_4.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed14088.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..82275b49979a33c34bc7e60f4ea644c2c3be34c8 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed14088.png_4.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed14088.png_5.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed14088.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..0c72afa90d67965efd9bf154361fdc0c3e8b8dd1 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed14088.png_5.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed14088.png_6.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed14088.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..7ba70d29f2a622183a7441220b8d707e4efd1dd2 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed14088.png_6.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed14088.png_7.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed14088.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..19a65922e4985ab270f99be24fb78cfa988a2139 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed14088.png_7.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed14088.png_8.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed14088.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..2066de9e6cc0531122663fe9e5c8a3e036f04c3a Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed14088.png_8.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed14088.png_9.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed14088.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..5af02ed422d00f455fed890e35d280c8ac6bfc01 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed14088.png_9.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed16124.png_1.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed16124.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..5899aafb284d9bdb4731b54b016ad40f635110bb Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed16124.png_1.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed16124.png_10.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed16124.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..c77ad9ac76695b891723a326c22011a504d33539 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed16124.png_10.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed16124.png_2.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed16124.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..e2533dcd596ea5d3750aff2401d37eb4c7d19655 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed16124.png_2.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed16124.png_3.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed16124.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..54cc72c0515cd9961152b79478d720ea3cff35d7 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed16124.png_3.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed16124.png_4.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed16124.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..f48df5634585e6ef9f722e13c443225744fddf76 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed16124.png_4.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed16124.png_5.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed16124.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..9e9b875499327f7c79bf3fa89c3bd1e1f0f7a470 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed16124.png_5.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed16124.png_6.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed16124.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..183e832de039ead3e672084bdc7674704c58d40a Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed16124.png_6.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed16124.png_7.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed16124.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..862750d7b0d3a3f5a6ddddd7d3c39c708184a907 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed16124.png_7.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed16124.png_8.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed16124.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..1f32fd2fae140d37025829c011f3cfebbfac0fd8 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed16124.png_8.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed16124.png_9.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed16124.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..a56d0278eb839e646e8f5ebf11d0437cb60df753 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed16124.png_9.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed17282.png_1.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed17282.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..694b1f3019b153648443185bb5ce374da8384183 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed17282.png_1.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed17282.png_10.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed17282.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..2f5b4ca2be7db0b3ffaae1e88e7819f8e5bb5ae6 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed17282.png_10.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed17282.png_2.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed17282.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..c063fdd30a0b2de95c58f5cc254ee8c0a97a2b1b Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed17282.png_2.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed17282.png_3.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed17282.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..ec88379ade272d657720bf90f8c7aa6b0b42ea0a Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed17282.png_3.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed17282.png_4.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed17282.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..fc917c28432a95e5d0e7d60d348dcf3fefffe080 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed17282.png_4.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed17282.png_5.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed17282.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..aa80a6ccc7f1b72a48bef4fb48e3d2349daaa373 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed17282.png_5.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed17282.png_6.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed17282.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..cb3090ff817e00276af4fb98568e15fdd043b7de Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed17282.png_6.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed17282.png_7.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed17282.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..bccc4458366c54fce38c711997484c0feffdc788 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed17282.png_7.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed17282.png_8.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed17282.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..893c2b3e095e507d555ef10f9befbc5e54063ef1 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed17282.png_8.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed17282.png_9.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed17282.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..d636c562f51fed3052a42e87cf04278f45352c38 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed17282.png_9.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed21815.png_1.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed21815.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..87d8fbe21968c2a790f83aff5d33329fa4a17de3 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed21815.png_1.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed21815.png_10.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed21815.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..a0dc388f73dee78a4b07ba9f3871d3c7702872ab Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed21815.png_10.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed21815.png_2.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed21815.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..f71f4bba4a80968afef521a92f63f363a1a1c7d3 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed21815.png_2.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed21815.png_3.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed21815.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..02623b75f54ac0a539ce623f3766e2fc814be198 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed21815.png_3.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed21815.png_4.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed21815.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..6406922e9792a9020d4a7a28af95dbda15c74236 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed21815.png_4.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed21815.png_5.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed21815.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..e42cca737c546e124d14894822b3d8fc4fed593c Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed21815.png_5.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed21815.png_6.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed21815.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..b3d2e0585125347622a8d194cea0cc0fad867f89 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed21815.png_6.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed21815.png_7.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed21815.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..f15972b44f93bc6906139772a7d925e6b6487801 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed21815.png_7.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed21815.png_8.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed21815.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..bd951e4d4209586f58f2baeb0b009f0038b8afcb Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed21815.png_8.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed21815.png_9.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed21815.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..290f4beb720ae48a06e91169e984689cca31109e Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed21815.png_9.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed22624.png_1.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed22624.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..cf29133aa3721ce197accbc3227b1d379191fc11 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed22624.png_1.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed22624.png_10.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed22624.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..38198264bbd3008f5f0ab3bd846df9ae5d39a449 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed22624.png_10.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed22624.png_2.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed22624.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..4346c1631f86705eed5699e0e00c83edaddd5cfc Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed22624.png_2.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed22624.png_3.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed22624.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..ccfc4dc5d288fd73c57fcbfa42a8d1427dd66f72 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed22624.png_3.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed22624.png_4.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed22624.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..80f7ec278d6072be6255af5eada70da7c3416f0b Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed22624.png_4.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed22624.png_5.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed22624.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..941668e0b5d5dc31776d827059b9f5d3059aa69d Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed22624.png_5.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed22624.png_6.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed22624.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..417ef2e93bfce6bbcc3e2697bedf4b5e8847fb56 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed22624.png_6.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed22624.png_7.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed22624.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..52f02f858a25087f8e29bec2010f6bf54a139df4 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed22624.png_7.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed22624.png_8.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed22624.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..6075a78ca1346826fa30cde827909b610340e122 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed22624.png_8.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed22624.png_9.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed22624.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..c5e87fb0936e02ed0734f760b05a105a57b025d9 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed22624.png_9.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed23602.png_1.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed23602.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..06b75cedd1d5d89fedc28af3cb4b41856bc78dfb Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed23602.png_1.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed23602.png_10.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed23602.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..b2f96c09d2bdd36f66a17369ad2e1aa69788ed4f Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed23602.png_10.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed23602.png_2.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed23602.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..63205f780df8118b4b126caf958a0f531b11d5c5 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed23602.png_2.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed23602.png_3.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed23602.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..c8f96ca6f1587175732144cf28adcc97b9eedafd Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed23602.png_3.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed23602.png_4.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed23602.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..c0f5cc80aaa27ce8e3f1ec8f78106a5a99e788ef Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed23602.png_4.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed23602.png_5.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed23602.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..3d23bdfd98722884ebed70b2d5204ef914b0388d Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed23602.png_5.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed23602.png_6.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed23602.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..abfb57ca2df11fafbc56d6423bc80707ea121023 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed23602.png_6.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed23602.png_7.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed23602.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..2155da747294b7f149e966958bcb82990d8810a8 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed23602.png_7.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed23602.png_8.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed23602.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..317022bfc04db33f2fd7e07be795570e3a38e91c Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed23602.png_8.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed23602.png_9.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed23602.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..b7dea0f488d110b8eb31c9c8f573b6023f5f4169 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed23602.png_9.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed24366.png_1.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed24366.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..77e05184541b0493d058d40840a9edfdfae60f6b Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed24366.png_1.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed24366.png_10.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed24366.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..0eca9a744f734c487a5a3a0d1aeff230e82f49da Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed24366.png_10.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed24366.png_2.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed24366.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..01af7a1dde41dff53c456ca8f074df4dc536ce3c Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed24366.png_2.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed24366.png_3.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed24366.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..b2fc90cbb82c6541ba1ca143c656c0fdf33a496d Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed24366.png_3.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed24366.png_4.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed24366.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..331501cbeebf87a4fc4f870136e17ce2934c4e63 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed24366.png_4.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed24366.png_5.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed24366.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..08713cfa73b3d800db90426ec2bebe5fe862656a Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed24366.png_5.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed24366.png_6.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed24366.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..e254fc4943e378136e040f1b4158e731a5bbd666 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed24366.png_6.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed24366.png_7.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed24366.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..2be038989fcfb20cff07504a1758b44ca94ff01d Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed24366.png_7.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed24366.png_8.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed24366.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..c264c509dacf18996f8723b84ce49286de25d17c Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed24366.png_8.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed24366.png_9.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed24366.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..cd159c5941fdd3f0b14668ee17dd217225865a64 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed24366.png_9.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed26848.png_1.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed26848.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..eff458544fa7f0c2cff69664b625ac1e113e1eef Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed26848.png_1.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed26848.png_10.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed26848.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..6c223d1559c5e324d46c2c6a78538622a2ccab27 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed26848.png_10.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed26848.png_2.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed26848.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..a146baf7a49200bce72ac8b3ed7d08cfe7cdd155 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed26848.png_2.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed26848.png_3.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed26848.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..4a6a4727a56a00018edfda84f92512d6939faa9a Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed26848.png_3.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed26848.png_4.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed26848.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..9ca8c7cfa346938cc75c4a30037e4099c9d5d91e Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed26848.png_4.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed26848.png_5.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed26848.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..9eac2f99043308fbe004397ba6eec471237c4dc9 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed26848.png_5.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed26848.png_6.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed26848.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..2a09047406135e612ccda562e4e8214391daf4de Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed26848.png_6.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed26848.png_7.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed26848.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..39b0bc61aa7c12b5319499711fe0b3397e04601d Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed26848.png_7.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed26848.png_8.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed26848.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..1d266aa95c04b5c16e5c7763b7683a1a551c9072 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed26848.png_8.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed26848.png_9.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed26848.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..7567dc67a98bf5a1a201006ad406a9cb955b20b5 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed26848.png_9.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed44086.png_1.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed44086.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..17148f864a7525fb3e956ae678f1083d4e2a0ee9 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed44086.png_1.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed44086.png_10.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed44086.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..a061fb735d010bf22e17167f72b12bd155048e93 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed44086.png_10.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed44086.png_2.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed44086.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..4a51cdd0887345a113cccf66b62e8e77cf3e4ada Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed44086.png_2.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed44086.png_3.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed44086.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..6c5747b2a00e672ac0e4e045d71ce85fabe097fc Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed44086.png_3.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed44086.png_4.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed44086.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..fbbb55283dd47c8b67c9e57940dc9f6355d1928b Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed44086.png_4.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed44086.png_5.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed44086.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..d15db9337d238cda793bc55e1728f2cb883c060e Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed44086.png_5.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed44086.png_6.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed44086.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..95c5054e798ef23ed0397d2808f3218ec48a1e21 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed44086.png_6.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed44086.png_7.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed44086.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..c3b8d37e5671693ad0c7cdaaca645238a071cb58 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed44086.png_7.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed44086.png_8.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed44086.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..126ee7dd82afae263e426b569cf1c8b6581b4eda Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed44086.png_8.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed44086.png_9.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed44086.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..9426e5959bceceb4147301ba161779401e0950c9 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/20230605_201824/vis_data/vis_image/seed44086.png_9.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/epoch_1.pth b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/epoch_1.pth new file mode 100644 index 0000000000000000000000000000000000000000..d14a1a5731b797f4617f85728f992960d4d6b335 --- /dev/null +++ b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/epoch_1.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c24a6ea63c15a885562eb14d2e37bd02e47174b14c3c69d0de6cf3c13864be00 +size 277073997 diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/epoch_10.pth b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/epoch_10.pth new file mode 100644 index 0000000000000000000000000000000000000000..77a3515b16e23407ecbb69ea1604da402c54474d --- /dev/null +++ b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/epoch_10.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:2a6524ca5809ccb90e5f474e16e8ab03ce3c6c04f0f303be4f946ba56aec1d77 +size 277404941 diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/epoch_2.pth b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/epoch_2.pth new file mode 100644 index 0000000000000000000000000000000000000000..d01135ee0cb5026166bc7a9f72994ccab3ac0ec5 --- /dev/null +++ b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/epoch_2.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:9f4783533f364383ba668e080c0fd6b13acf636df45969b8273cd3977651c3f0 +size 277111629 diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/epoch_3.pth b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/epoch_3.pth new file mode 100644 index 0000000000000000000000000000000000000000..8cfd2bf1521ea459843decd863aa283c9dfb38d8 --- /dev/null +++ b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/epoch_3.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ea45b74f7938a5b96bcd772fe12949303cf68b5a754fccf34ece58a409dee0c4 +size 277148365 diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/epoch_4.pth b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/epoch_4.pth new file mode 100644 index 0000000000000000000000000000000000000000..da80ab17aa658dfc6f17c9ab95c23066baf7a6f1 --- /dev/null +++ b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/epoch_4.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:0a1327336b02e1cb6444d7e86a92d82f15813cf0b1ca4d521979098e541f3fdd +size 277184909 diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/epoch_5.pth b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/epoch_5.pth new file mode 100644 index 0000000000000000000000000000000000000000..35d820b46e00ec7d6fda39d4de977096d753f80a --- /dev/null +++ b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/epoch_5.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:b8acfc8daf314317a72ed736f543bbd2c6209a3d0f7e266b71b0d220a360fc24 +size 277221581 diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/epoch_6.pth b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/epoch_6.pth new file mode 100644 index 0000000000000000000000000000000000000000..771141eb8e3ae7c6b6b3ffd8a1d02e2d81c31f55 --- /dev/null +++ b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/epoch_6.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:05393a8b54456be252ec878592011e0d073a925fb342a90813628fbc9e6e09e4 +size 277258253 diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/epoch_7.pth b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/epoch_7.pth new file mode 100644 index 0000000000000000000000000000000000000000..d73386deb2b442b2be613ebde7c3092a8349e2c8 --- /dev/null +++ b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/epoch_7.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:a3306f4cb3e90048e98b84509b0e184e380ea45336eead2a60f4a42bbfe2fd57 +size 277294797 diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/epoch_8.pth b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/epoch_8.pth new file mode 100644 index 0000000000000000000000000000000000000000..961685ada044ead78facdd1f1fe3d34ab30f10b6 --- /dev/null +++ b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/epoch_8.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:a46ed2978926a30a3fcc786b4df60eaa021ab06d1e2ddb170338d9410bcfc2f1 +size 277331405 diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/epoch_9.pth b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/epoch_9.pth new file mode 100644 index 0000000000000000000000000000000000000000..2694dbd7089b40d9c99db84a2af620017dddce8b --- /dev/null +++ b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/epoch_9.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:9fa44e01259b25fe8d990a2ecd615a93de65bf570bfd14d245b997d47710707b +size 277368077 diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/last_checkpoint b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/last_checkpoint new file mode 100644 index 0000000000000000000000000000000000000000..12647aae85c0ada288a58c72401247b020dc7039 --- /dev/null +++ b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/last_checkpoint @@ -0,0 +1 @@ +/mnt/petrelfs/luzeyu/workspace/fakebench/mmpretrain/workdir/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/epoch_10.pth \ No newline at end of file diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1.py b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1.py new file mode 100644 index 0000000000000000000000000000000000000000..86eb2c86d955d97ec259c3c8897f93ba1a2234a9 --- /dev/null +++ b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1.py @@ -0,0 +1,230 @@ +optim_wrapper = dict( + optimizer=dict( + type='SGD', + lr=0.0005, + momentum=0.9, + weight_decay=0.0001, + _scope_='mmpretrain'), + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=False, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=512) +model = dict( + type='ImageClassifier', + backbone=dict( + frozen_stages=2, + type='ResNet', + depth=50, + num_stages=4, + out_indices=(3, ), + style='pytorch', + init_cfg=dict( + type='Pretrained', + checkpoint= + 'https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth', + prefix='backbone')), + neck=dict(type='GlobalAveragePooling'), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=2048, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + topk=1)) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3fake8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3real8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3fake8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3real8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = dict(type='Accuracy', topk=1) +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3fake8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3real8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.1), + dict(type='GaussianBlur', radius=1.5, prob=0.1), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = dict(type='Accuracy', topk=1) +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_1e-1' diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/20230605_201824.log b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/20230605_201824.log new file mode 100644 index 0000000000000000000000000000000000000000..16416b1e65eff401472e3a30162dfcfc8ed445e2 --- /dev/null +++ b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/20230605_201824.log @@ -0,0 +1,939 @@ +2023/06/05 20:18:27 - mmengine - INFO - +------------------------------------------------------------ +System environment: + sys.platform: linux + Python: 3.10.9 (main, Mar 8 2023, 10:47:38) [GCC 11.2.0] + CUDA available: True + numpy_random_seed: 1882164705 + GPU 0,1: NVIDIA A100-SXM4-80GB + CUDA_HOME: /mnt/petrelfs/share/cuda-11.6 + NVCC: Cuda compilation tools, release 11.6, V11.6.124 + GCC: gcc (GCC) 7.5.0 + PyTorch: 1.13.1 + PyTorch compiling details: PyTorch built with: + - GCC 9.3 + - C++ Version: 201402 + - Intel(R) oneAPI Math Kernel Library Version 2021.4-Product Build 20210904 for Intel(R) 64 architecture applications + - Intel(R) MKL-DNN v2.6.0 (Git Hash 52b5f107dd9cf10910aaa19cb47f3abf9b349815) + - OpenMP 201511 (a.k.a. OpenMP 4.5) + - LAPACK is enabled (usually provided by MKL) + - NNPACK is enabled + - CPU capability usage: AVX2 + - CUDA Runtime 11.6 + - NVCC architecture flags: -gencode;arch=compute_37,code=sm_37;-gencode;arch=compute_50,code=sm_50;-gencode;arch=compute_60,code=sm_60;-gencode;arch=compute_61,code=sm_61;-gencode;arch=compute_70,code=sm_70;-gencode;arch=compute_75,code=sm_75;-gencode;arch=compute_80,code=sm_80;-gencode;arch=compute_86,code=sm_86;-gencode;arch=compute_37,code=compute_37 + - CuDNN 8.3.2 (built against CUDA 11.5) + - Magma 2.6.1 + - Build settings: BLAS_INFO=mkl, BUILD_TYPE=Release, CUDA_VERSION=11.6, CUDNN_VERSION=8.3.2, CXX_COMPILER=/opt/rh/devtoolset-9/root/usr/bin/c++, CXX_FLAGS= -fabi-version=11 -Wno-deprecated -fvisibility-inlines-hidden -DUSE_PTHREADPOOL -fopenmp -DNDEBUG -DUSE_KINETO -DUSE_FBGEMM -DUSE_QNNPACK -DUSE_PYTORCH_QNNPACK -DUSE_XNNPACK -DSYMBOLICATE_MOBILE_DEBUG_HANDLE -DEDGE_PROFILER_USE_KINETO -O2 -fPIC -Wno-narrowing -Wall -Wextra -Werror=return-type -Werror=non-virtual-dtor -Wno-missing-field-initializers -Wno-type-limits -Wno-array-bounds -Wno-unknown-pragmas -Wunused-local-typedefs -Wno-unused-parameter -Wno-unused-function -Wno-unused-result -Wno-strict-overflow -Wno-strict-aliasing -Wno-error=deprecated-declarations -Wno-stringop-overflow -Wno-psabi -Wno-error=pedantic -Wno-error=redundant-decls -Wno-error=old-style-cast -fdiagnostics-color=always -faligned-new -Wno-unused-but-set-variable -Wno-maybe-uninitialized -fno-math-errno -fno-trapping-math -Werror=format -Werror=cast-function-type -Wno-stringop-overflow, LAPACK_INFO=mkl, PERF_WITH_AVX=1, PERF_WITH_AVX2=1, PERF_WITH_AVX512=1, TORCH_VERSION=1.13.1, USE_CUDA=ON, USE_CUDNN=ON, USE_EXCEPTION_PTR=1, USE_GFLAGS=OFF, USE_GLOG=OFF, USE_MKL=ON, USE_MKLDNN=ON, USE_MPI=OFF, USE_NCCL=ON, USE_NNPACK=ON, USE_OPENMP=ON, USE_ROCM=OFF, + + TorchVision: 0.14.1 + OpenCV: 4.7.0 + MMEngine: 0.7.3 + +Runtime environment: + cudnn_benchmark: True + mp_cfg: {'mp_start_method': 'fork', 'opencv_num_threads': 0} + dist_cfg: {'backend': 'nccl'} + seed: None + deterministic: False + Distributed launcher: slurm + Distributed training: True + GPU number: 2 +------------------------------------------------------------ + +2023/06/05 20:18:32 - mmengine - INFO - Config: +optim_wrapper = dict( + optimizer=dict( + type='SGD', + lr=0.0005, + momentum=0.9, + weight_decay=0.0001, + _scope_='mmpretrain'), + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=False, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=512) +model = dict( + type='ImageClassifier', + backbone=dict( + frozen_stages=2, + type='ResNet', + depth=50, + num_stages=4, + out_indices=(3, ), + style='pytorch', + init_cfg=dict( + type='Pretrained', + checkpoint= + 'https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth', + prefix='backbone')), + neck=dict(type='GlobalAveragePooling'), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=2048, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + topk=1)) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3fake8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3real8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3fake8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3real8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = dict(type='Accuracy', topk=1) +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3fake8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3real8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = dict(type='Accuracy', topk=1) +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1' + +2023/06/05 20:18:43 - mmengine - INFO - Hooks will be executed in the following order: +before_run: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_load_checkpoint: +(ABOVE_NORMAL) EMAHook + -------------------- +before_train: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_train_epoch: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(NORMAL ) DistSamplerSeedHook + -------------------- +before_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook + -------------------- +after_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +after_train_epoch: +(NORMAL ) IterTimerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_val_epoch: +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook + -------------------- +before_val_iter: +(NORMAL ) IterTimerHook + -------------------- +after_val_iter: +(NORMAL ) IterTimerHook +(NORMAL ) VisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_val_epoch: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_save_checkpoint: +(ABOVE_NORMAL) EMAHook + -------------------- +after_train: +(VERY_LOW ) CheckpointHook + -------------------- +before_test_epoch: +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook + -------------------- +before_test_iter: +(NORMAL ) IterTimerHook + -------------------- +after_test_iter: +(NORMAL ) IterTimerHook +(NORMAL ) VisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_test_epoch: +(VERY_HIGH ) RuntimeInfoHook +(ABOVE_NORMAL) EMAHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_run: +(BELOW_NORMAL) LoggerHook + -------------------- +2023/06/05 20:18:54 - mmengine - INFO - load backbone in model from: https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth +Name of parameter - Initialization information + +backbone.conv1.weight - torch.Size([64, 3, 7, 7]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.bn1.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.bn1.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.conv1.weight - torch.Size([64, 64, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.bn1.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.bn1.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.conv2.weight - torch.Size([64, 64, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.bn2.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.bn2.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.conv3.weight - torch.Size([256, 64, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.bn3.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.bn3.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.downsample.0.weight - torch.Size([256, 64, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.downsample.1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.0.downsample.1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.conv1.weight - torch.Size([64, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.bn1.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.bn1.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.conv2.weight - torch.Size([64, 64, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.bn2.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.bn2.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.conv3.weight - torch.Size([256, 64, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.bn3.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.1.bn3.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.conv1.weight - torch.Size([64, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.bn1.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.bn1.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.conv2.weight - torch.Size([64, 64, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.bn2.weight - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.bn2.bias - torch.Size([64]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.conv3.weight - torch.Size([256, 64, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.bn3.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer1.2.bn3.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.conv1.weight - torch.Size([128, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.bn1.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.bn1.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.conv2.weight - torch.Size([128, 128, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.bn2.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.bn2.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.conv3.weight - torch.Size([512, 128, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.bn3.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.bn3.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.downsample.0.weight - torch.Size([512, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.downsample.1.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.0.downsample.1.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.conv1.weight - torch.Size([128, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.bn1.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.bn1.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.conv2.weight - torch.Size([128, 128, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.bn2.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.bn2.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.conv3.weight - torch.Size([512, 128, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.bn3.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.1.bn3.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.conv1.weight - torch.Size([128, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.bn1.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.bn1.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.conv2.weight - torch.Size([128, 128, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.bn2.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.bn2.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.conv3.weight - torch.Size([512, 128, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.bn3.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.2.bn3.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.conv1.weight - torch.Size([128, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.bn1.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.bn1.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.conv2.weight - torch.Size([128, 128, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.bn2.weight - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.bn2.bias - torch.Size([128]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.conv3.weight - torch.Size([512, 128, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.bn3.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer2.3.bn3.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.conv1.weight - torch.Size([256, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.bn1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.bn1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.conv2.weight - torch.Size([256, 256, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.bn2.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.bn2.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.conv3.weight - torch.Size([1024, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.bn3.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.bn3.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.downsample.0.weight - torch.Size([1024, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.downsample.1.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.0.downsample.1.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.conv1.weight - torch.Size([256, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.bn1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.bn1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.conv2.weight - torch.Size([256, 256, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.bn2.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.bn2.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.conv3.weight - torch.Size([1024, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.bn3.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.1.bn3.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.conv1.weight - torch.Size([256, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.bn1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.bn1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.conv2.weight - torch.Size([256, 256, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.bn2.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.bn2.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.conv3.weight - torch.Size([1024, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.bn3.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.2.bn3.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.conv1.weight - torch.Size([256, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.bn1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.bn1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.conv2.weight - torch.Size([256, 256, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.bn2.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.bn2.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.conv3.weight - torch.Size([1024, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.bn3.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.3.bn3.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.conv1.weight - torch.Size([256, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.bn1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.bn1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.conv2.weight - torch.Size([256, 256, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.bn2.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.bn2.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.conv3.weight - torch.Size([1024, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.bn3.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.4.bn3.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.conv1.weight - torch.Size([256, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.bn1.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.bn1.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.conv2.weight - torch.Size([256, 256, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.bn2.weight - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.bn2.bias - torch.Size([256]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.conv3.weight - torch.Size([1024, 256, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.bn3.weight - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer3.5.bn3.bias - torch.Size([1024]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.conv1.weight - torch.Size([512, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.bn1.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.bn1.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.conv2.weight - torch.Size([512, 512, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.bn2.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.bn2.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.conv3.weight - torch.Size([2048, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.bn3.weight - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.bn3.bias - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.downsample.0.weight - torch.Size([2048, 1024, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.downsample.1.weight - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.0.downsample.1.bias - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.conv1.weight - torch.Size([512, 2048, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.bn1.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.bn1.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.conv2.weight - torch.Size([512, 512, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.bn2.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.bn2.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.conv3.weight - torch.Size([2048, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.bn3.weight - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.1.bn3.bias - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.conv1.weight - torch.Size([512, 2048, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.bn1.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.bn1.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.conv2.weight - torch.Size([512, 512, 3, 3]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.bn2.weight - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.bn2.bias - torch.Size([512]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.conv3.weight - torch.Size([2048, 512, 1, 1]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.bn3.weight - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +backbone.layer4.2.bn3.bias - torch.Size([2048]): +PretrainedInit: load from https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth + +head.fc.weight - torch.Size([2, 2048]): +NormalInit: mean=0, std=0.01, bias=0 + +head.fc.bias - torch.Size([2]): +NormalInit: mean=0, std=0.01, bias=0 +2023/06/05 20:18:54 - mmengine - WARNING - "FileClient" will be deprecated in future. Please use io functions in https://mmengine.readthedocs.io/en/latest/api/fileio.html#file-io +2023/06/05 20:18:54 - mmengine - WARNING - "HardDiskBackend" is the alias of "LocalBackend" and the former will be deprecated in future. +2023/06/05 20:18:54 - mmengine - INFO - Checkpoints will be saved to /mnt/petrelfs/luzeyu/workspace/fakebench/mmpretrain/workdir/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1. +2023/06/05 20:20:58 - mmengine - INFO - Epoch(train) [1][100/342] lr: 4.9899e-04 eta: 1:08:46 time: 1.2849 data_time: 0.5975 memory: 9436 loss: 0.6495 +2023/06/05 20:23:00 - mmengine - INFO - Epoch(train) [1][200/342] lr: 4.9592e-04 eta: 1:05:59 time: 1.1534 data_time: 0.0010 memory: 6319 loss: 0.6212 +2023/06/05 20:25:03 - mmengine - INFO - Epoch(train) [1][300/342] lr: 4.9082e-04 eta: 1:03:52 time: 1.1771 data_time: 0.0010 memory: 6319 loss: 0.5789 +2023/06/05 20:26:03 - mmengine - INFO - Exp name: resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1_20230605_201824 +2023/06/05 20:26:03 - mmengine - INFO - Saving checkpoint at 1 epochs +2023/06/05 20:28:08 - mmengine - INFO - Epoch(val) [1][100/342] eta: 0:04:48 time: 0.3926 data_time: 0.3047 memory: 6319 +2023/06/05 20:29:51 - mmengine - INFO - Epoch(val) [1][200/342] eta: 0:02:38 time: 0.2632 data_time: 0.1767 memory: 3133 +2023/06/05 20:32:02 - mmengine - INFO - Epoch(val) [1][300/342] eta: 0:00:49 time: 1.2990 data_time: 1.2044 memory: 3133 +2023/06/05 20:34:41 - mmengine - INFO - Epoch(val) [1][342/342] accuracy/top1: 50.0418 data_time: 1.2935 time: 1.3827 +2023/06/05 20:36:46 - mmengine - INFO - Epoch(train) [2][100/342] lr: 4.8017e-04 eta: 1:02:11 time: 1.1686 data_time: 1.0274 memory: 6319 loss: 0.5387 +2023/06/05 20:39:16 - mmengine - INFO - Epoch(train) [2][200/342] lr: 4.7036e-04 eta: 1:02:18 time: 1.5408 data_time: 1.1892 memory: 6319 loss: 0.4997 +2023/06/05 20:41:21 - mmengine - INFO - Epoch(train) [2][300/342] lr: 4.5874e-04 eta: 0:59:48 time: 1.3369 data_time: 1.1039 memory: 6319 loss: 0.4653 +2023/06/05 20:42:14 - mmengine - INFO - Exp name: resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1_20230605_201824 +2023/06/05 20:42:14 - mmengine - INFO - Saving checkpoint at 2 epochs +2023/06/05 20:44:12 - mmengine - INFO - Epoch(val) [2][100/342] eta: 0:04:34 time: 0.4002 data_time: 0.3128 memory: 6319 +2023/06/05 20:45:56 - mmengine - INFO - Epoch(val) [2][200/342] eta: 0:02:34 time: 0.2944 data_time: 0.2061 memory: 3133 +2023/06/05 20:48:09 - mmengine - INFO - Epoch(val) [2][300/342] eta: 0:00:48 time: 1.2405 data_time: 1.1527 memory: 3133 +2023/06/05 20:50:48 - mmengine - INFO - Epoch(val) [2][342/342] accuracy/top1: 50.1025 data_time: 1.2862 time: 1.3758 +2023/06/05 20:53:11 - mmengine - INFO - Epoch(train) [3][100/342] lr: 4.3931e-04 eta: 0:57:24 time: 1.1695 data_time: 1.0277 memory: 6319 loss: 0.4386 +2023/06/05 20:55:12 - mmengine - INFO - Epoch(train) [3][200/342] lr: 4.2373e-04 eta: 0:54:45 time: 1.1963 data_time: 0.8555 memory: 6319 loss: 0.4121 +2023/06/05 20:57:12 - mmengine - INFO - Epoch(train) [3][300/342] lr: 4.0672e-04 eta: 0:52:12 time: 1.1999 data_time: 1.0597 memory: 6319 loss: 0.3897 +2023/06/05 20:57:34 - mmengine - INFO - Exp name: resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1_20230605_201824 +2023/06/05 20:58:07 - mmengine - INFO - Exp name: resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1_20230605_201824 +2023/06/05 20:58:07 - mmengine - INFO - Saving checkpoint at 3 epochs +2023/06/05 21:00:06 - mmengine - INFO - Epoch(val) [3][100/342] eta: 0:04:36 time: 0.3877 data_time: 0.2995 memory: 6319 +2023/06/05 21:01:49 - mmengine - INFO - Epoch(val) [3][200/342] eta: 0:02:34 time: 0.2764 data_time: 0.1889 memory: 3133 +2023/06/05 21:03:58 - mmengine - INFO - Epoch(val) [3][300/342] eta: 0:00:48 time: 1.2306 data_time: 1.1430 memory: 3133 +2023/06/05 21:06:32 - mmengine - INFO - Epoch(val) [3][342/342] accuracy/top1: 50.3514 data_time: 1.2670 time: 1.3561 +2023/06/05 21:08:41 - mmengine - INFO - Epoch(train) [4][100/342] lr: 3.8041e-04 eta: 0:49:13 time: 1.2096 data_time: 1.0692 memory: 6319 loss: 0.3662 +2023/06/05 21:10:42 - mmengine - INFO - Epoch(train) [4][200/342] lr: 3.6058e-04 eta: 0:46:50 time: 1.1554 data_time: 0.7792 memory: 6319 loss: 0.3346 +2023/06/05 21:12:45 - mmengine - INFO - Epoch(train) [4][300/342] lr: 3.3985e-04 eta: 0:44:33 time: 1.2434 data_time: 0.5548 memory: 6319 loss: 0.3336 +2023/06/05 21:13:36 - mmengine - INFO - Exp name: resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1_20230605_201824 +2023/06/05 21:13:36 - mmengine - INFO - Saving checkpoint at 4 epochs +2023/06/05 21:15:37 - mmengine - INFO - Epoch(val) [4][100/342] eta: 0:04:38 time: 0.3925 data_time: 0.3045 memory: 6319 +2023/06/05 21:17:20 - mmengine - INFO - Epoch(val) [4][200/342] eta: 0:02:35 time: 0.2717 data_time: 0.1841 memory: 3133 +2023/06/05 21:19:29 - mmengine - INFO - Epoch(val) [4][300/342] eta: 0:00:48 time: 1.2387 data_time: 1.1516 memory: 3133 +2023/06/05 21:22:41 - mmengine - INFO - Epoch(val) [4][342/342] accuracy/top1: 51.3788 data_time: 1.3843 time: 1.4735 +2023/06/05 21:24:43 - mmengine - INFO - Epoch(train) [5][100/342] lr: 3.0924e-04 eta: 0:41:22 time: 1.2301 data_time: 0.8734 memory: 6319 loss: 0.3272 +2023/06/05 21:26:42 - mmengine - INFO - Epoch(train) [5][200/342] lr: 2.8709e-04 eta: 0:39:05 time: 1.1990 data_time: 0.8177 memory: 6319 loss: 0.3046 +2023/06/05 21:28:40 - mmengine - INFO - Epoch(train) [5][300/342] lr: 2.6467e-04 eta: 0:36:49 time: 1.1245 data_time: 0.9713 memory: 6319 loss: 0.3211 +2023/06/05 21:29:37 - mmengine - INFO - Exp name: resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1_20230605_201824 +2023/06/05 21:29:37 - mmengine - INFO - Saving checkpoint at 5 epochs +2023/06/05 21:31:37 - mmengine - INFO - Epoch(val) [5][100/342] eta: 0:04:37 time: 0.3860 data_time: 0.2988 memory: 6319 +2023/06/05 21:33:21 - mmengine - INFO - Epoch(val) [5][200/342] eta: 0:02:35 time: 0.2676 data_time: 0.1793 memory: 3133 +2023/06/05 21:35:27 - mmengine - INFO - Epoch(val) [5][300/342] eta: 0:00:48 time: 1.2500 data_time: 1.1624 memory: 3133 +2023/06/05 21:38:01 - mmengine - INFO - Epoch(val) [5][342/342] accuracy/top1: 54.6057 data_time: 1.2660 time: 1.3563 +2023/06/05 21:40:04 - mmengine - INFO - Epoch(train) [6][100/342] lr: 2.3275e-04 eta: 0:33:50 time: 1.1654 data_time: 1.0251 memory: 6319 loss: 0.2789 +2023/06/05 21:42:14 - mmengine - INFO - Epoch(train) [6][200/342] lr: 2.1046e-04 eta: 0:31:47 time: 1.4526 data_time: 1.3120 memory: 6319 loss: 0.2919 +2023/06/05 21:44:11 - mmengine - INFO - Exp name: resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1_20230605_201824 +2023/06/05 21:44:23 - mmengine - INFO - Epoch(train) [6][300/342] lr: 1.8855e-04 eta: 0:29:43 time: 1.1946 data_time: 1.0530 memory: 6319 loss: 0.2822 +2023/06/05 21:45:22 - mmengine - INFO - Exp name: resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1_20230605_201824 +2023/06/05 21:45:22 - mmengine - INFO - Saving checkpoint at 6 epochs +2023/06/05 21:47:26 - mmengine - INFO - Epoch(val) [6][100/342] eta: 0:04:45 time: 0.4808 data_time: 0.3928 memory: 6319 +2023/06/05 21:49:16 - mmengine - INFO - Epoch(val) [6][200/342] eta: 0:02:42 time: 0.3201 data_time: 0.2319 memory: 3133 +2023/06/05 21:51:43 - mmengine - INFO - Epoch(val) [6][300/342] eta: 0:00:52 time: 1.3505 data_time: 1.2626 memory: 3133 +2023/06/05 21:54:17 - mmengine - INFO - Epoch(val) [6][342/342] accuracy/top1: 57.4096 data_time: 1.3534 time: 1.4438 +2023/06/05 21:56:33 - mmengine - INFO - Epoch(train) [7][100/342] lr: 1.5844e-04 eta: 0:26:53 time: 1.3771 data_time: 1.1431 memory: 6319 loss: 0.2906 +2023/06/05 21:58:49 - mmengine - INFO - Epoch(train) [7][200/342] lr: 1.3820e-04 eta: 0:24:50 time: 1.2198 data_time: 0.8385 memory: 6319 loss: 0.2867 +2023/06/05 22:01:01 - mmengine - INFO - Epoch(train) [7][300/342] lr: 1.1893e-04 eta: 0:22:44 time: 1.1376 data_time: 0.7595 memory: 6319 loss: 0.2879 +2023/06/05 22:01:51 - mmengine - INFO - Exp name: resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1_20230605_201824 +2023/06/05 22:01:51 - mmengine - INFO - Saving checkpoint at 7 epochs +2023/06/05 22:03:49 - mmengine - INFO - Epoch(val) [7][100/342] eta: 0:04:31 time: 0.3966 data_time: 0.3082 memory: 6319 +2023/06/05 22:05:32 - mmengine - INFO - Epoch(val) [7][200/342] eta: 0:02:32 time: 0.2680 data_time: 0.1805 memory: 3133 +2023/06/05 22:07:38 - mmengine - INFO - Epoch(val) [7][300/342] eta: 0:00:47 time: 1.2322 data_time: 1.1445 memory: 3133 +2023/06/05 22:10:08 - mmengine - INFO - Epoch(val) [7][342/342] accuracy/top1: 61.2506 data_time: 1.2450 time: 1.3354 +2023/06/05 22:12:10 - mmengine - INFO - Epoch(train) [8][100/342] lr: 9.3587e-05 eta: 0:19:39 time: 1.1975 data_time: 1.0346 memory: 6319 loss: 0.2823 +2023/06/05 22:14:08 - mmengine - INFO - Epoch(train) [8][200/342] lr: 7.7361e-05 eta: 0:17:29 time: 1.2037 data_time: 1.0451 memory: 6319 loss: 0.2687 +2023/06/05 22:16:07 - mmengine - INFO - Epoch(train) [8][300/342] lr: 6.2632e-05 eta: 0:15:20 time: 1.1284 data_time: 0.9485 memory: 6319 loss: 0.2843 +2023/06/05 22:17:07 - mmengine - INFO - Exp name: resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1_20230605_201824 +2023/06/05 22:17:07 - mmengine - INFO - Saving checkpoint at 8 epochs +2023/06/05 22:19:05 - mmengine - INFO - Epoch(val) [8][100/342] eta: 0:04:33 time: 0.3895 data_time: 0.3023 memory: 6319 +2023/06/05 22:20:48 - mmengine - INFO - Epoch(val) [8][200/342] eta: 0:02:33 time: 0.2742 data_time: 0.1872 memory: 3133 +2023/06/05 22:22:54 - mmengine - INFO - Epoch(val) [8][300/342] eta: 0:00:47 time: 1.2206 data_time: 1.1333 memory: 3133 +2023/06/05 22:25:25 - mmengine - INFO - Epoch(val) [8][342/342] accuracy/top1: 63.6636 data_time: 1.2523 time: 1.3414 +2023/06/05 22:27:24 - mmengine - INFO - Epoch(train) [9][100/342] lr: 4.4532e-05 eta: 0:12:20 time: 1.1546 data_time: 1.0153 memory: 6319 loss: 0.2839 +2023/06/05 22:29:24 - mmengine - INFO - Epoch(train) [9][200/342] lr: 3.3915e-05 eta: 0:10:12 time: 1.1568 data_time: 1.0171 memory: 6319 loss: 0.2801 +2023/06/05 22:30:47 - mmengine - INFO - Exp name: resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1_20230605_201824 +2023/06/05 22:31:24 - mmengine - INFO - Epoch(train) [9][300/342] lr: 2.5163e-05 eta: 0:08:04 time: 1.2109 data_time: 1.0695 memory: 6319 loss: 0.2601 +2023/06/05 22:32:20 - mmengine - INFO - Exp name: resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1_20230605_201824 +2023/06/05 22:32:20 - mmengine - INFO - Saving checkpoint at 9 epochs +2023/06/05 22:34:18 - mmengine - INFO - Epoch(val) [9][100/342] eta: 0:04:34 time: 0.3927 data_time: 0.3045 memory: 6319 +2023/06/05 22:36:01 - mmengine - INFO - Epoch(val) [9][200/342] eta: 0:02:33 time: 0.2709 data_time: 0.1835 memory: 3133 +2023/06/05 22:38:07 - mmengine - INFO - Epoch(val) [9][300/342] eta: 0:00:47 time: 1.2099 data_time: 1.1229 memory: 3133 +2023/06/05 22:40:37 - mmengine - INFO - Epoch(val) [9][342/342] accuracy/top1: 64.7819 data_time: 1.2482 time: 1.3377 +2023/06/05 22:42:41 - mmengine - INFO - Epoch(train) [10][100/342] lr: 1.6078e-05 eta: 0:05:05 time: 1.2366 data_time: 1.0948 memory: 6319 loss: 0.2446 +2023/06/05 22:44:38 - mmengine - INFO - Epoch(train) [10][200/342] lr: 1.2111e-05 eta: 0:02:58 time: 1.1288 data_time: 0.9895 memory: 6319 loss: 0.2731 +2023/06/05 22:46:37 - mmengine - INFO - Epoch(train) [10][300/342] lr: 1.0191e-05 eta: 0:00:52 time: 1.1453 data_time: 0.9601 memory: 6319 loss: 0.2515 +2023/06/05 22:47:34 - mmengine - INFO - Exp name: resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1_20230605_201824 +2023/06/05 22:47:34 - mmengine - INFO - Saving checkpoint at 10 epochs +2023/06/05 22:49:34 - mmengine - INFO - Epoch(val) [10][100/342] eta: 0:04:35 time: 0.3990 data_time: 0.3106 memory: 6319 +2023/06/05 22:51:16 - mmengine - INFO - Epoch(val) [10][200/342] eta: 0:02:33 time: 0.2628 data_time: 0.1764 memory: 3133 +2023/06/05 22:53:26 - mmengine - INFO - Epoch(val) [10][300/342] eta: 0:00:48 time: 1.2678 data_time: 1.1803 memory: 3133 +2023/06/05 22:56:02 - mmengine - INFO - Epoch(val) [10][342/342] accuracy/top1: 65.0790 data_time: 1.2776 time: 1.3668 diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/20230605_201824.json b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/20230605_201824.json new file mode 100644 index 0000000000000000000000000000000000000000..ecc35dbddb0b5d99056abdbcb33d0d464bad75f1 --- /dev/null +++ b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/20230605_201824.json @@ -0,0 +1,40 @@ +{"lr": 0.0004989875940690582, "data_time": 0.5975338459014893, "loss": 0.6494718611240387, "time": 1.284948205947876, "epoch": 1, "memory": 9436, "step": 100} +{"lr": 0.0004959179335073666, "data_time": 0.0010037899017333984, "loss": 0.6211717963218689, "time": 1.1534340620040893, "epoch": 1, "memory": 6319, "step": 200} +{"lr": 0.0004908167990123805, "data_time": 0.0010053396224975586, "loss": 0.5789063513278961, "time": 1.1771417140960694, "epoch": 1, "memory": 6319, "step": 300} +{"accuracy/top1": 50.04178237915039, "data_time": 1.2934649394966706, "time": 1.3826614360363163, "step": 1} +{"lr": 0.00048017042254960683, "data_time": 1.0274295091629029, "loss": 0.5386920869350433, "time": 1.168610405921936, "epoch": 2, "memory": 6319, "step": 442} +{"lr": 0.0004703638635189273, "data_time": 1.1891530990600585, "loss": 0.4996999889612198, "time": 1.5407658338546752, "epoch": 2, "memory": 6319, "step": 542} +{"lr": 0.00045874130819215336, "data_time": 1.1039084672927857, "loss": 0.4652951151132584, "time": 1.336899757385254, "epoch": 2, "memory": 6319, "step": 642} +{"accuracy/top1": 50.10245132446289, "data_time": 1.2861871406566296, "time": 1.3758409273520156, "step": 2} +{"lr": 0.000439312001216421, "data_time": 1.0276948451995849, "loss": 0.4386217385530472, "time": 1.1694526195526123, "epoch": 3, "memory": 6319, "step": 784} +{"lr": 0.00042372847804098874, "data_time": 0.8555418491363526, "loss": 0.4121491193771362, "time": 1.1963413953781128, "epoch": 3, "memory": 6319, "step": 884} +{"lr": 0.0004067221985769131, "data_time": 1.059733510017395, "loss": 0.3896999448537827, "time": 1.1998631477355957, "epoch": 3, "memory": 6319, "step": 984} +{"accuracy/top1": 50.35142135620117, "data_time": 1.2669503640155404, "time": 1.356149530271747, "step": 3} +{"lr": 0.0003804118370269479, "data_time": 1.0692113161087036, "loss": 0.3662323087453842, "time": 1.2095635890960694, "epoch": 4, "memory": 6319, "step": 1126} +{"lr": 0.00036057677353196443, "data_time": 0.7792315244674682, "loss": 0.3345847100019455, "time": 1.1554383754730224, "epoch": 4, "memory": 6319, "step": 1226} +{"lr": 0.00033985146305424703, "data_time": 0.5548282146453858, "loss": 0.3336219251155853, "time": 1.2434006690979005, "epoch": 4, "memory": 6319, "step": 1326} +{"accuracy/top1": 51.378780364990234, "data_time": 1.3842955240355288, "time": 1.4735116673628026, "step": 4} +{"lr": 0.0003092354884336273, "data_time": 0.8734158277511597, "loss": 0.327228906750679, "time": 1.2301369190216065, "epoch": 5, "memory": 6319, "step": 1468} +{"lr": 0.00028709047883299587, "data_time": 0.8177111625671387, "loss": 0.3045926094055176, "time": 1.1990467071533204, "epoch": 5, "memory": 6319, "step": 1568} +{"lr": 0.0002646748751329253, "data_time": 0.9713254928588867, "loss": 0.32107959091663363, "time": 1.1244938850402832, "epoch": 5, "memory": 6319, "step": 1668} +{"accuracy/top1": 54.605655670166016, "data_time": 1.2660299883639499, "time": 1.356262774231135, "step": 5} +{"lr": 0.00023275019235155644, "data_time": 1.0250571966171265, "loss": 0.27891314327716826, "time": 1.1653618574142457, "epoch": 6, "memory": 6319, "step": 1810} +{"lr": 0.00021046294447834082, "data_time": 1.3120041131973266, "loss": 0.2919475883245468, "time": 1.452617335319519, "epoch": 6, "memory": 6319, "step": 1910} +{"lr": 0.00018855124302477407, "data_time": 1.0530446290969848, "loss": 0.28223230242729186, "time": 1.1946197271347045, "epoch": 6, "memory": 6319, "step": 2010} +{"accuracy/top1": 57.40957260131836, "data_time": 1.3534343409468759, "time": 1.4438216588935073, "step": 6} +{"lr": 0.0001584428624656395, "data_time": 1.14311683177948, "loss": 0.29064355939626696, "time": 1.3770950078964233, "epoch": 7, "memory": 6319, "step": 2152} +{"lr": 0.00013819500742605714, "data_time": 0.8385151386260986, "loss": 0.2866679698228836, "time": 1.2198035717010498, "epoch": 7, "memory": 6319, "step": 2252} +{"lr": 0.00011893207822507105, "data_time": 0.7594561338424682, "loss": 0.2878702521324158, "time": 1.137579870223999, "epoch": 7, "memory": 6319, "step": 2352} +{"accuracy/top1": 61.25057601928711, "data_time": 1.2449571154903045, "time": 1.3353959867627558, "step": 7} +{"lr": 9.35872179547221e-05, "data_time": 1.0346049070358276, "loss": 0.2823109865188599, "time": 1.1975229024887084, "epoch": 8, "memory": 6319, "step": 2494} +{"lr": 7.736075687514847e-05, "data_time": 1.0450668334960938, "loss": 0.268718022108078, "time": 1.2036758422851563, "epoch": 8, "memory": 6319, "step": 2594} +{"lr": 6.263218964965453e-05, "data_time": 0.9484605073928833, "loss": 0.28429160118103025, "time": 1.1283856868743896, "epoch": 8, "memory": 6319, "step": 2694} +{"accuracy/top1": 63.663578033447266, "data_time": 1.2523446569637375, "time": 1.3413701015728208, "step": 8} +{"lr": 4.453178117937791e-05, "data_time": 1.0152947902679443, "loss": 0.2838896796107292, "time": 1.1546067953109742, "epoch": 9, "memory": 6319, "step": 2836} +{"lr": 3.391507312668422e-05, "data_time": 1.0170658349990844, "loss": 0.2800527557730675, "time": 1.1567620992660523, "epoch": 9, "memory": 6319, "step": 2936} +{"lr": 2.5162602656676073e-05, "data_time": 1.0694981336593627, "loss": 0.26014820486307144, "time": 1.2109471797943114, "epoch": 9, "memory": 6319, "step": 3036} +{"accuracy/top1": 64.78193664550781, "data_time": 1.2481763460198227, "time": 1.337702832486122, "step": 9} +{"lr": 1.607844008050381e-05, "data_time": 1.0947700023651123, "loss": 0.24455928653478623, "time": 1.2365663051605225, "epoch": 10, "memory": 6319, "step": 3178} +{"lr": 1.2110722409842075e-05, "data_time": 0.9895391225814819, "loss": 0.2730581223964691, "time": 1.1288204193115234, "epoch": 10, "memory": 6319, "step": 3278} +{"lr": 1.019110148697161e-05, "data_time": 0.9600519895553589, "loss": 0.25153474062681197, "time": 1.1452585220336915, "epoch": 10, "memory": 6319, "step": 3378} +{"accuracy/top1": 65.07898712158203, "data_time": 1.2776176227425, "time": 1.3667749017042599, "step": 10} diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/config.py b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/config.py new file mode 100644 index 0000000000000000000000000000000000000000..5a015cf83c3e0bc25c3f53ad1ff46ae4ceceabeb --- /dev/null +++ b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/config.py @@ -0,0 +1,230 @@ +optim_wrapper = dict( + optimizer=dict( + type='SGD', + lr=0.0005, + momentum=0.9, + weight_decay=0.0001, + _scope_='mmpretrain'), + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=False, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=512) +model = dict( + type='ImageClassifier', + backbone=dict( + frozen_stages=2, + type='ResNet', + depth=50, + num_stages=4, + out_indices=(3, ), + style='pytorch', + init_cfg=dict( + type='Pretrained', + checkpoint= + 'https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth', + prefix='backbone')), + neck=dict(type='GlobalAveragePooling'), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=2048, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + topk=1)) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3fake8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3real8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3fake8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3real8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = dict(type='Accuracy', topk=1) +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3fake8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3real8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = dict(type='Accuracy', topk=1) +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1' diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/events.out.tfevents.1685967518.SH-IDC1-10-140-24-114.79597.0 b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/events.out.tfevents.1685967518.SH-IDC1-10-140-24-114.79597.0 new file mode 100644 index 0000000000000000000000000000000000000000..fb28ae3f2c30c62b07f02e501e317e5f3193c817 --- /dev/null +++ b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/events.out.tfevents.1685967518.SH-IDC1-10-140-24-114.79597.0 @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:aaeed248cedcc0977cf0d290a46d06f5bcdafaa637325f614c492e98edfb5924 +size 13220345 diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/scalars.json b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/scalars.json new file mode 100644 index 0000000000000000000000000000000000000000..ecc35dbddb0b5d99056abdbcb33d0d464bad75f1 --- /dev/null +++ b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/scalars.json @@ -0,0 +1,40 @@ +{"lr": 0.0004989875940690582, "data_time": 0.5975338459014893, "loss": 0.6494718611240387, "time": 1.284948205947876, "epoch": 1, "memory": 9436, "step": 100} +{"lr": 0.0004959179335073666, "data_time": 0.0010037899017333984, "loss": 0.6211717963218689, "time": 1.1534340620040893, "epoch": 1, "memory": 6319, "step": 200} +{"lr": 0.0004908167990123805, "data_time": 0.0010053396224975586, "loss": 0.5789063513278961, "time": 1.1771417140960694, "epoch": 1, "memory": 6319, "step": 300} +{"accuracy/top1": 50.04178237915039, "data_time": 1.2934649394966706, "time": 1.3826614360363163, "step": 1} +{"lr": 0.00048017042254960683, "data_time": 1.0274295091629029, "loss": 0.5386920869350433, "time": 1.168610405921936, "epoch": 2, "memory": 6319, "step": 442} +{"lr": 0.0004703638635189273, "data_time": 1.1891530990600585, "loss": 0.4996999889612198, "time": 1.5407658338546752, "epoch": 2, "memory": 6319, "step": 542} +{"lr": 0.00045874130819215336, "data_time": 1.1039084672927857, "loss": 0.4652951151132584, "time": 1.336899757385254, "epoch": 2, "memory": 6319, "step": 642} +{"accuracy/top1": 50.10245132446289, "data_time": 1.2861871406566296, "time": 1.3758409273520156, "step": 2} +{"lr": 0.000439312001216421, "data_time": 1.0276948451995849, "loss": 0.4386217385530472, "time": 1.1694526195526123, "epoch": 3, "memory": 6319, "step": 784} +{"lr": 0.00042372847804098874, "data_time": 0.8555418491363526, "loss": 0.4121491193771362, "time": 1.1963413953781128, "epoch": 3, "memory": 6319, "step": 884} +{"lr": 0.0004067221985769131, "data_time": 1.059733510017395, "loss": 0.3896999448537827, "time": 1.1998631477355957, "epoch": 3, "memory": 6319, "step": 984} +{"accuracy/top1": 50.35142135620117, "data_time": 1.2669503640155404, "time": 1.356149530271747, "step": 3} +{"lr": 0.0003804118370269479, "data_time": 1.0692113161087036, "loss": 0.3662323087453842, "time": 1.2095635890960694, "epoch": 4, "memory": 6319, "step": 1126} +{"lr": 0.00036057677353196443, "data_time": 0.7792315244674682, "loss": 0.3345847100019455, "time": 1.1554383754730224, "epoch": 4, "memory": 6319, "step": 1226} +{"lr": 0.00033985146305424703, "data_time": 0.5548282146453858, "loss": 0.3336219251155853, "time": 1.2434006690979005, "epoch": 4, "memory": 6319, "step": 1326} +{"accuracy/top1": 51.378780364990234, "data_time": 1.3842955240355288, "time": 1.4735116673628026, "step": 4} +{"lr": 0.0003092354884336273, "data_time": 0.8734158277511597, "loss": 0.327228906750679, "time": 1.2301369190216065, "epoch": 5, "memory": 6319, "step": 1468} +{"lr": 0.00028709047883299587, "data_time": 0.8177111625671387, "loss": 0.3045926094055176, "time": 1.1990467071533204, "epoch": 5, "memory": 6319, "step": 1568} +{"lr": 0.0002646748751329253, "data_time": 0.9713254928588867, "loss": 0.32107959091663363, "time": 1.1244938850402832, "epoch": 5, "memory": 6319, "step": 1668} +{"accuracy/top1": 54.605655670166016, "data_time": 1.2660299883639499, "time": 1.356262774231135, "step": 5} +{"lr": 0.00023275019235155644, "data_time": 1.0250571966171265, "loss": 0.27891314327716826, "time": 1.1653618574142457, "epoch": 6, "memory": 6319, "step": 1810} +{"lr": 0.00021046294447834082, "data_time": 1.3120041131973266, "loss": 0.2919475883245468, "time": 1.452617335319519, "epoch": 6, "memory": 6319, "step": 1910} +{"lr": 0.00018855124302477407, "data_time": 1.0530446290969848, "loss": 0.28223230242729186, "time": 1.1946197271347045, "epoch": 6, "memory": 6319, "step": 2010} +{"accuracy/top1": 57.40957260131836, "data_time": 1.3534343409468759, "time": 1.4438216588935073, "step": 6} +{"lr": 0.0001584428624656395, "data_time": 1.14311683177948, "loss": 0.29064355939626696, "time": 1.3770950078964233, "epoch": 7, "memory": 6319, "step": 2152} +{"lr": 0.00013819500742605714, "data_time": 0.8385151386260986, "loss": 0.2866679698228836, "time": 1.2198035717010498, "epoch": 7, "memory": 6319, "step": 2252} +{"lr": 0.00011893207822507105, "data_time": 0.7594561338424682, "loss": 0.2878702521324158, "time": 1.137579870223999, "epoch": 7, "memory": 6319, "step": 2352} +{"accuracy/top1": 61.25057601928711, "data_time": 1.2449571154903045, "time": 1.3353959867627558, "step": 7} +{"lr": 9.35872179547221e-05, "data_time": 1.0346049070358276, "loss": 0.2823109865188599, "time": 1.1975229024887084, "epoch": 8, "memory": 6319, "step": 2494} +{"lr": 7.736075687514847e-05, "data_time": 1.0450668334960938, "loss": 0.268718022108078, "time": 1.2036758422851563, "epoch": 8, "memory": 6319, "step": 2594} +{"lr": 6.263218964965453e-05, "data_time": 0.9484605073928833, "loss": 0.28429160118103025, "time": 1.1283856868743896, "epoch": 8, "memory": 6319, "step": 2694} +{"accuracy/top1": 63.663578033447266, "data_time": 1.2523446569637375, "time": 1.3413701015728208, "step": 8} +{"lr": 4.453178117937791e-05, "data_time": 1.0152947902679443, "loss": 0.2838896796107292, "time": 1.1546067953109742, "epoch": 9, "memory": 6319, "step": 2836} +{"lr": 3.391507312668422e-05, "data_time": 1.0170658349990844, "loss": 0.2800527557730675, "time": 1.1567620992660523, "epoch": 9, "memory": 6319, "step": 2936} +{"lr": 2.5162602656676073e-05, "data_time": 1.0694981336593627, "loss": 0.26014820486307144, "time": 1.2109471797943114, "epoch": 9, "memory": 6319, "step": 3036} +{"accuracy/top1": 64.78193664550781, "data_time": 1.2481763460198227, "time": 1.337702832486122, "step": 9} +{"lr": 1.607844008050381e-05, "data_time": 1.0947700023651123, "loss": 0.24455928653478623, "time": 1.2365663051605225, "epoch": 10, "memory": 6319, "step": 3178} +{"lr": 1.2110722409842075e-05, "data_time": 0.9895391225814819, "loss": 0.2730581223964691, "time": 1.1288204193115234, "epoch": 10, "memory": 6319, "step": 3278} +{"lr": 1.019110148697161e-05, "data_time": 0.9600519895553589, "loss": 0.25153474062681197, "time": 1.1452585220336915, "epoch": 10, "memory": 6319, "step": 3378} +{"accuracy/top1": 65.07898712158203, "data_time": 1.2776176227425, "time": 1.3667749017042599, "step": 10} diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/35624.png_1.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/35624.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..0226fc92c86d6b610a1f01a0e0bfee8070c2ac57 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/35624.png_1.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/35624.png_10.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/35624.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..f4e94c69f4eab855e398f1cbbbd2f392f321d899 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/35624.png_10.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/35624.png_2.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/35624.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..47489d7865d13de4f65d2e2b29a6293473d5683f Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/35624.png_2.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/35624.png_3.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/35624.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..84c290f69ba0e45c19a0e891ad3f0ae75d949e0c Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/35624.png_3.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/35624.png_4.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/35624.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..8a3eea422cc0fb622655fc5a47ddc555c65b4fe7 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/35624.png_4.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/35624.png_5.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/35624.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..23d584f88888a28bd4504c6d2df764d856a36a4d Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/35624.png_5.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/35624.png_6.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/35624.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..2e083c5c33277c8c7e46ec61b459d95760486a0c Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/35624.png_6.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/35624.png_7.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/35624.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..9bf0bf451db18080454ddcfc93dffb5d5684a392 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/35624.png_7.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/35624.png_8.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/35624.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..6e0e7a42176d1f507ce45f570967394d4442c71d Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/35624.png_8.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/35624.png_9.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/35624.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..07a1c0ee6c1dd1a31a0520dc83ccebee75540d1e Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/35624.png_9.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/38442.png_1.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/38442.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..3f8894d0e5c0522f0cb5a657ef3f06974691090c Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/38442.png_1.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/38442.png_10.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/38442.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..1d5bad856a0de753b2f6c776da356f9bfe742902 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/38442.png_10.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/38442.png_2.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/38442.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..9e19111949fc0dc907f9fb0e35b78333f78969ee Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/38442.png_2.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/38442.png_3.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/38442.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..b7c6cba86a550591b5a67bf8c34b85439572c35e Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/38442.png_3.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/38442.png_4.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/38442.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..66d8f55a94c3a63e428752500a0cd95d9d32c586 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/38442.png_4.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/38442.png_5.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/38442.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..082283d5e9e21338570af52a7a6a01416ce9fb51 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/38442.png_5.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/38442.png_6.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/38442.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..28ee5a9ef99e86bd07ec6998a300c156f1a8417d Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/38442.png_6.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/38442.png_7.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/38442.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..46dfe7412f85abd542c7f4c51d0c8c17fe0e13c7 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/38442.png_7.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/38442.png_8.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/38442.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..ccac8dba33057aefef4bcb3237ddc7fb78f5d443 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/38442.png_8.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/38442.png_9.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/38442.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..285ae4f0ddc174283bb3f8626f89315ca3b7a647 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/38442.png_9.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/43283.png_1.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/43283.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..58098cbae7abc57ea2748cf8341336327b76e648 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/43283.png_1.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/43283.png_10.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/43283.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..792d641a3228710dfff7a476b5e27a4e431b7250 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/43283.png_10.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/43283.png_2.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/43283.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..946d4e91c87a5d9aacfff6d3acc604d4126b0c0c Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/43283.png_2.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/43283.png_3.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/43283.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..34fbdc74e22d6410cf8fcb9d6758b0d70ccbf856 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/43283.png_3.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/43283.png_4.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/43283.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..8e5d2c7ae9c18956fde929be387957a1bd33964a Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/43283.png_4.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/43283.png_5.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/43283.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..4cb21e60471459b4e62dc0298f64093d6be0d787 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/43283.png_5.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/43283.png_6.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/43283.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..3602fa274b86c4d5cd57c94a03b30f93cfae82ea Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/43283.png_6.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/43283.png_7.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/43283.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..af421bc2ad1bd4f465b3cd9f7bc7fa5d30f2004a Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/43283.png_7.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/43283.png_8.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/43283.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..678f0dbff5575d6aaeeb63f38cc78478214591ac Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/43283.png_8.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/43283.png_9.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/43283.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..294224731b913882974b634bcaa428e999ea99a4 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/43283.png_9.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/44982.png_1.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/44982.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..9959886ae55cea609a86fe6a44ac0bf80ebaef89 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/44982.png_1.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/44982.png_10.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/44982.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..5f2f43d4649244a1b89ffe9266931284c1a58a3c Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/44982.png_10.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/44982.png_2.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/44982.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..74863c1d34f4adef5f0b16cc1172a025aceb7f6b Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/44982.png_2.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/44982.png_3.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/44982.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..43746e478247c789d43139d9ec1ccdc0d6da7cc6 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/44982.png_3.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/44982.png_4.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/44982.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..14c497e5c59672939a7710b79176db93a55a3071 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/44982.png_4.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/44982.png_5.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/44982.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..219d12b5227a67aa3a4449697d98d7299783f577 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/44982.png_5.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/44982.png_6.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/44982.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..412c11de2b4922c0b4bbfed718758cc19fa9effa Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/44982.png_6.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/44982.png_7.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/44982.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..4e0669f548bb4ad5fc36c35153b6885b2e7f8ac1 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/44982.png_7.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/44982.png_8.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/44982.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..ec663c98154de6e53eb55c3feb0f84b77bfe0278 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/44982.png_8.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/44982.png_9.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/44982.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..ce8d9add5a3b7e80d5210f2cabef801fdf5591f2 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/44982.png_9.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/56423.png_1.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/56423.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..ba055c569d591d57c4d5ce5ea741bd5110e413f0 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/56423.png_1.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/56423.png_10.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/56423.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..b8a70b3f98314175ea0e9d5dd46a377893d28e6e Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/56423.png_10.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/56423.png_2.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/56423.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..567ffba1a669781ab94eff95804d825410f7cf4e Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/56423.png_2.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/56423.png_3.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/56423.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..b7e409bd18bf77b2256b2e1eb6a261c11b9d22cf Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/56423.png_3.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/56423.png_4.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/56423.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..454d7114fb348064dbbe7fb26441d01a5888c8b6 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/56423.png_4.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/56423.png_5.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/56423.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..fa688c148d4361c51bcac55c7e52a4ba637e985d Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/56423.png_5.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/56423.png_6.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/56423.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..510a0ea9cec9bf73783a605814be93d07ce2dbc9 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/56423.png_6.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/56423.png_7.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/56423.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..9a8af3d60e8a300013cc603f1193bd206df48942 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/56423.png_7.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/56423.png_8.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/56423.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..003bfca60672a6c0d9168a29acec54bc86f69c33 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/56423.png_8.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/56423.png_9.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/56423.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..38dc1539a54fb81747390af6bbe3947f41390998 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/56423.png_9.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/61024.png_1.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/61024.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..12ad4951da86b6f955bd3efbca648ad9eeb23f67 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/61024.png_1.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/61024.png_10.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/61024.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..b5775e0372da66e934ec508820acbe4fc9f24865 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/61024.png_10.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/61024.png_2.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/61024.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..36f6676d5e1b2be0aacc6afeed26ee5b4a0f0b62 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/61024.png_2.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/61024.png_3.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/61024.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..ee6c29ee453b9422b6cb99ca834c48ccb247c921 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/61024.png_3.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/61024.png_4.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/61024.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..11c544f08c9bfb52d946c7239447a675d85a92da Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/61024.png_4.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/61024.png_5.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/61024.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..90a7fcd4126bd0978c4a4528b2a69fb3ee83a36a Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/61024.png_5.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/61024.png_6.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/61024.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..cde95d224106f03d8799746d99b049720b30eae9 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/61024.png_6.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/61024.png_7.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/61024.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..75bb80f57bfa6a81c8a06142c76e472714563594 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/61024.png_7.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/61024.png_8.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/61024.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..0bb96027d3d121e90d765a1d055a574f538b33ba Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/61024.png_8.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/61024.png_9.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/61024.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..31d8506d4690a59b8ccb6a9de9fee09c5cef3e3d Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/61024.png_9.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/66390.png_1.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/66390.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..8a51e73112433cb2c742b456f387c3a5db1886a0 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/66390.png_1.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/66390.png_10.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/66390.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..fabdfb8377302751268a2e69a9080c5df90093ba Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/66390.png_10.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/66390.png_2.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/66390.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..c3c53a4d026326476e124270d6680a381c6c05a4 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/66390.png_2.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/66390.png_3.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/66390.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..8707ef0306dc0abcb1f8086153f859c1361a42a8 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/66390.png_3.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/66390.png_4.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/66390.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..67598f8d6f1b9a4bcf9bb1c3cf5b8432983da1bc Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/66390.png_4.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/66390.png_5.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/66390.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..13120dcc34a58e6518fe2f308ff62489d99caebf Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/66390.png_5.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/66390.png_6.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/66390.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..a7e38e9cee8b2c8091dde7a2b8912b4f355a6e96 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/66390.png_6.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/66390.png_7.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/66390.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..2634e201c886c463773e70ee6c2c03bc9c693774 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/66390.png_7.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/66390.png_8.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/66390.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..f486f762fcbdcd868170f979eea2bdef9d18a905 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/66390.png_8.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/66390.png_9.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/66390.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..64e306d312d6f285c538b213b89dfeaa4e20e944 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/66390.png_9.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/flickr_wild_000179.jpg_1.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/flickr_wild_000179.jpg_1.png new file mode 100644 index 0000000000000000000000000000000000000000..9f45a2f6547256397c9928114fc974ed4e9780d0 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/flickr_wild_000179.jpg_1.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/flickr_wild_000179.jpg_10.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/flickr_wild_000179.jpg_10.png new file mode 100644 index 0000000000000000000000000000000000000000..0da2367f6efbaa47d4faea4f4d7e2ec0d75beaa4 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/flickr_wild_000179.jpg_10.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/flickr_wild_000179.jpg_2.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/flickr_wild_000179.jpg_2.png new file mode 100644 index 0000000000000000000000000000000000000000..7cff388c3c34df173690712d242b760b117f18bf Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/flickr_wild_000179.jpg_2.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/flickr_wild_000179.jpg_3.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/flickr_wild_000179.jpg_3.png new file mode 100644 index 0000000000000000000000000000000000000000..34eb1938aeab46243218891d397f28df78e1c700 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/flickr_wild_000179.jpg_3.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/flickr_wild_000179.jpg_4.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/flickr_wild_000179.jpg_4.png new file mode 100644 index 0000000000000000000000000000000000000000..d45006e63b8d0d342e7d95cc88e4c0aafb5c75fb Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/flickr_wild_000179.jpg_4.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/flickr_wild_000179.jpg_5.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/flickr_wild_000179.jpg_5.png new file mode 100644 index 0000000000000000000000000000000000000000..3ff96fd1a4337118380648987a3524375edc10e5 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/flickr_wild_000179.jpg_5.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/flickr_wild_000179.jpg_6.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/flickr_wild_000179.jpg_6.png new file mode 100644 index 0000000000000000000000000000000000000000..88e3b43c54f487cdc0fe95e1277a6a99dac726d4 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/flickr_wild_000179.jpg_6.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/flickr_wild_000179.jpg_7.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/flickr_wild_000179.jpg_7.png new file mode 100644 index 0000000000000000000000000000000000000000..6e418adb20b1deb6d7a8bf18a6944f6f78667659 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/flickr_wild_000179.jpg_7.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/flickr_wild_000179.jpg_8.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/flickr_wild_000179.jpg_8.png new file mode 100644 index 0000000000000000000000000000000000000000..024c85838e2aefc1257ab294554d27779007e0e4 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/flickr_wild_000179.jpg_8.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/flickr_wild_000179.jpg_9.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/flickr_wild_000179.jpg_9.png new file mode 100644 index 0000000000000000000000000000000000000000..c9c2b4c0ac627ed00e04adabf7acaa23756548dd Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/flickr_wild_000179.jpg_9.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/pixabay_dog_002352.jpg_1.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/pixabay_dog_002352.jpg_1.png new file mode 100644 index 0000000000000000000000000000000000000000..25d93ab9f66e9787549e2ad3ed3a9f46663ec161 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/pixabay_dog_002352.jpg_1.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/pixabay_dog_002352.jpg_10.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/pixabay_dog_002352.jpg_10.png new file mode 100644 index 0000000000000000000000000000000000000000..9a804b88b26587aec35947d08bbc5419b9c98bf4 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/pixabay_dog_002352.jpg_10.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/pixabay_dog_002352.jpg_2.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/pixabay_dog_002352.jpg_2.png new file mode 100644 index 0000000000000000000000000000000000000000..ffba495174b32a51ebdb8e4029b5c406851424db Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/pixabay_dog_002352.jpg_2.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/pixabay_dog_002352.jpg_3.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/pixabay_dog_002352.jpg_3.png new file mode 100644 index 0000000000000000000000000000000000000000..d46266f445981e1bde8b837fbd05283f196b1a29 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/pixabay_dog_002352.jpg_3.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/pixabay_dog_002352.jpg_4.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/pixabay_dog_002352.jpg_4.png new file mode 100644 index 0000000000000000000000000000000000000000..bc9fcb9ee555070596d2042a43deda51923941e0 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/pixabay_dog_002352.jpg_4.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/pixabay_dog_002352.jpg_5.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/pixabay_dog_002352.jpg_5.png new file mode 100644 index 0000000000000000000000000000000000000000..7fc9a8ac5f06a85e256f47eb7413b2665ac5bec2 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/pixabay_dog_002352.jpg_5.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/pixabay_dog_002352.jpg_6.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/pixabay_dog_002352.jpg_6.png new file mode 100644 index 0000000000000000000000000000000000000000..82a7df0507bb116f9414b92ab05ddfdc19b04c1e Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/pixabay_dog_002352.jpg_6.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/pixabay_dog_002352.jpg_7.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/pixabay_dog_002352.jpg_7.png new file mode 100644 index 0000000000000000000000000000000000000000..ac718d7779828ae037af7c78c32516cd89b9e1cc Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/pixabay_dog_002352.jpg_7.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/pixabay_dog_002352.jpg_8.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/pixabay_dog_002352.jpg_8.png new file mode 100644 index 0000000000000000000000000000000000000000..73797b4eca8073a230607f690fcc79d255104c90 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/pixabay_dog_002352.jpg_8.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/pixabay_dog_002352.jpg_9.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/pixabay_dog_002352.jpg_9.png new file mode 100644 index 0000000000000000000000000000000000000000..57127acf6134730e129301141f3cc076a9f44230 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/pixabay_dog_002352.jpg_9.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed14088.png_1.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed14088.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..638f6ee0323cd3a5db8c44d351641e4be3d3a6c3 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed14088.png_1.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed14088.png_10.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed14088.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..075767ee7dff1b2b7d4d851e514b7ef7f53dcf82 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed14088.png_10.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed14088.png_2.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed14088.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..1ccc18578ed51bc1ec6347e6a5ae52037f433ac2 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed14088.png_2.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed14088.png_3.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed14088.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..1522f0331b4732ec71639f966720462cab97a887 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed14088.png_3.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed14088.png_4.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed14088.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..177071a5be14febc7ab1af52f692f675af63fec0 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed14088.png_4.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed14088.png_5.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed14088.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..6545c6b35aaa2328ecc20cc45100837e9c29f5b4 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed14088.png_5.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed14088.png_6.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed14088.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..b213d6354cdcf29efa98fb4e05fa4a8176c39763 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed14088.png_6.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed14088.png_7.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed14088.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..ade16abbd47c38ebe3a641ccd9549ed9d56e2307 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed14088.png_7.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed14088.png_8.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed14088.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..3b5e6a1a6e65e3b4c53548e0db4976a5d6014a86 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed14088.png_8.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed14088.png_9.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed14088.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..65c5cb854f80d532b8df1748be8911480fa2e7cc Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed14088.png_9.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed16124.png_1.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed16124.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..aaefb6b6bbb9e0a16548a8f341662c831703308a Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed16124.png_1.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed16124.png_10.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed16124.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..97a6d1172d8ba1c7001b5bfec91eab98f147e02f Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed16124.png_10.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed16124.png_2.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed16124.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..19c9ab4cf1a7e09c9606023d04da4ab920fc18fd Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed16124.png_2.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed16124.png_3.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed16124.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..aff73e0d848353cb39690c0c9b7d78827d5fa8c9 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed16124.png_3.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed16124.png_4.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed16124.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..43d1d312bd5a1776e607e2f3668c541072c21193 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed16124.png_4.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed16124.png_5.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed16124.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..753df0bbd8c132fe1f23c89c5a89329c0ece45e7 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed16124.png_5.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed16124.png_6.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed16124.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..d30c4097854182b8aabf414cdb2891472976c042 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed16124.png_6.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed16124.png_7.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed16124.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..9efa6cbdcfefd8648827352a960f2af388901a43 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed16124.png_7.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed16124.png_8.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed16124.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..f183e2f9ad7063fcff59f377fa38c24b08ace2a1 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed16124.png_8.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed16124.png_9.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed16124.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..8731b441b8a6a7ddb1b254c8d423a0fb5487f384 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed16124.png_9.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed17282.png_1.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed17282.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..997d45c79d9dc82bd00ce59348664289f5ccd06c Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed17282.png_1.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed17282.png_10.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed17282.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..56a167aae9e0efc6030a528d73f52004e38ead62 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed17282.png_10.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed17282.png_2.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed17282.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..7cdf690c28965650d3dd5c6b11d2b08c421b50dc Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed17282.png_2.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed17282.png_3.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed17282.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..e1ebbaefdedc794a1dffeb0b6a56efd707335bf1 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed17282.png_3.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed17282.png_4.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed17282.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..5c4b12eafcc7ec0faae492f66f9d25877bf839c7 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed17282.png_4.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed17282.png_5.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed17282.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..5266282e21f3d5a5c9c99d41ac63c5634931453a Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed17282.png_5.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed17282.png_6.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed17282.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..17231630be5f80e1f7de047e1b881504bf384a16 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed17282.png_6.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed17282.png_7.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed17282.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..8195b63f463eb48a149f50de6ff4d624c6fb4a17 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed17282.png_7.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed17282.png_8.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed17282.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..5307eb4991cbf9e6a81b147b49f8fa05e3c13174 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed17282.png_8.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed17282.png_9.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed17282.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..b8fab8f556eaa18160b2944b763b34c51edd0f67 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed17282.png_9.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed21815.png_1.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed21815.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..6da916cb4042495cacc424b95a5ab0814ea410d7 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed21815.png_1.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed21815.png_10.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed21815.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..c868da51c052f58d3f29f4f0cdeba152ff328d65 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed21815.png_10.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed21815.png_2.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed21815.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..86e0a1773e04d5a75700b04f93af8887e74b2228 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed21815.png_2.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed21815.png_3.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed21815.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..45f8351e7217d0a25285e8209c9ec3e6a67a5ecd Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed21815.png_3.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed21815.png_4.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed21815.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..1ecf06a166c9725ed6ec1fb6a84ef20baf15b16f Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed21815.png_4.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed21815.png_5.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed21815.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..1c1af83438eeb5c87663867fd3ee25f9faa15555 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed21815.png_5.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed21815.png_6.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed21815.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..644dc03aa057cd2222046921d5b648435d61cd5e Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed21815.png_6.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed21815.png_7.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed21815.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..24dcf4030b0ecc56653f2205b99a08bf2b6af200 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed21815.png_7.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed21815.png_8.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed21815.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..8477a640bbe8539daddf8f7966906bded7a70863 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed21815.png_8.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed21815.png_9.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed21815.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..762d92cb2afc1cccff661c8ba3b056caf58d9df0 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed21815.png_9.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed22624.png_1.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed22624.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..228de51605d38e316f194c14628b70fdc526622d Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed22624.png_1.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed22624.png_10.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed22624.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..8dd1af2d77bd60fed3bd23642cbadf3e1e977dd8 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed22624.png_10.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed22624.png_2.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed22624.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..e017309c0275140c2c9768f4e7ca0ac04cd2beb8 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed22624.png_2.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed22624.png_3.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed22624.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..828fdf6748a62193c7b080775c3751c3eb8390eb Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed22624.png_3.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed22624.png_4.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed22624.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..4db9c8f34f73c547592852b4e2774d89007ed72a Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed22624.png_4.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed22624.png_5.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed22624.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..ef51d3bc372413a8632e17b97ad505dc54f0c837 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed22624.png_5.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed22624.png_6.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed22624.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..b4c677ad3f01247ee51924a2e221ded07e541fa7 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed22624.png_6.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed22624.png_7.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed22624.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..d9803bbb5247bcbaca6f9550b1fea67175fb1d60 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed22624.png_7.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed22624.png_8.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed22624.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..0feabc855bfb814f8b53256c9f84aedad8548e6b Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed22624.png_8.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed22624.png_9.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed22624.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..038c4a2d81d25b541e3a59d978351b11cb3d8bfc Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed22624.png_9.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed23602.png_1.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed23602.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..404b09a28b1b3c8cab7b7af76d90fb5b6cdb7ab0 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed23602.png_1.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed23602.png_10.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed23602.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..477aa48698f9e9b41a1c01b58965005a49bb27e3 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed23602.png_10.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed23602.png_2.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed23602.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..1321ec871a7095655c603427cf93dda86a6c9150 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed23602.png_2.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed23602.png_3.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed23602.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..a779ac5dd5f88a4a79c2d77c8877107f1a24f768 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed23602.png_3.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed23602.png_4.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed23602.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..e86fc30edb8daa9f885bfaa26f3a55901ad2ae58 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed23602.png_4.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed23602.png_5.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed23602.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..285692e0589621fb51d8a5cb24373fa8a093cd5b Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed23602.png_5.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed23602.png_6.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed23602.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..6a29fc348831433e6580991bfd3c39141b03f9ce Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed23602.png_6.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed23602.png_7.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed23602.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..a3c0549e65d20992dc55344f47ac7da0a03f4fc2 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed23602.png_7.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed23602.png_8.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed23602.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..5eec67e9e3bf9d6d5f9c788b901d6b9395463fd2 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed23602.png_8.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed23602.png_9.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed23602.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..03b6011c85ac63c45f7e5a7dd083827d705156ff Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed23602.png_9.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed24366.png_1.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed24366.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..c9831a2846f9d1118213054142da7a4f9d3c5922 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed24366.png_1.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed24366.png_10.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed24366.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..c24aef68733d7974bac2504e09ab71769ed42ce8 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed24366.png_10.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed24366.png_2.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed24366.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..22f43f93d868951f01b60ef92d36f91a12bd33b0 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed24366.png_2.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed24366.png_3.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed24366.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..7625fa1e351866b18ea61a0b593f7719af581959 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed24366.png_3.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed24366.png_4.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed24366.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..0900e27f42c941b69a5f649104216327355a6026 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed24366.png_4.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed24366.png_5.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed24366.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..e911cf7b4e027a292b0d03b584c139442ce6d233 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed24366.png_5.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed24366.png_6.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed24366.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..d69097752816b89f18064c6bf0d505ce80e9574b Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed24366.png_6.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed24366.png_7.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed24366.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..5c87757551ede50747d99bc8348f08bd3524804f Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed24366.png_7.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed24366.png_8.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed24366.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..a543f4beff90a294a5ceb7cc1e187f3533058cd8 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed24366.png_8.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed24366.png_9.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed24366.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..fe3dcf6614715ac367f27a548b50841217df2fbf Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed24366.png_9.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed26848.png_1.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed26848.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..cb6d8d3a4888d1a2b4241ad0368f9f6e7396ce25 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed26848.png_1.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed26848.png_10.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed26848.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..7745768a8dc4cc7a15849e72f06670f1227569b4 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed26848.png_10.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed26848.png_2.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed26848.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..7b59c875553d460506e92bf76bd4177c24fdfb5c Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed26848.png_2.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed26848.png_3.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed26848.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..31e368f80d2c620732c51d2eaf3014b56755ef83 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed26848.png_3.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed26848.png_4.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed26848.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..ed95d288f834d4ad601a595f6dde1c08c79d3ec0 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed26848.png_4.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed26848.png_5.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed26848.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..87423af3d6bcd70e2e3ff3f35eca7a2525a37d86 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed26848.png_5.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed26848.png_6.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed26848.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..d37c8f3be7d22ff2332f0a8b8116de085c9eb793 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed26848.png_6.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed26848.png_7.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed26848.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..3c3e5f3e604a2755f967af8ea77a47b31bd72858 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed26848.png_7.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed26848.png_8.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed26848.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..3974b8b7087076c641c26c14dd363a774a9193a0 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed26848.png_8.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed26848.png_9.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed26848.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..dd11373e336aae748a6db12013dd44f46af81486 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed26848.png_9.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed44086.png_1.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed44086.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..8a660b1870cd1fb07eb17cbe97d3ceb1e5293f62 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed44086.png_1.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed44086.png_10.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed44086.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..748fbe2b03469fa71f29578c6a9d53dffd916f8f Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed44086.png_10.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed44086.png_2.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed44086.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..a978e08b1d4a93187fb38b5bf0890965dfab32b4 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed44086.png_2.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed44086.png_3.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed44086.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..7257507b9e802cd434d57d8eaaa2d12761ff4bbc Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed44086.png_3.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed44086.png_4.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed44086.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..73cc2b0f6f8e3f77705e218da1684407aecefedd Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed44086.png_4.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed44086.png_5.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed44086.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..cb706b4621c7ced23ed6b3df54b56c5d1af21933 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed44086.png_5.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed44086.png_6.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed44086.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..8153d7f109ab4207a91c3abbee92b239c3106ef2 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed44086.png_6.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed44086.png_7.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed44086.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..32b73f9faf316ef56464535fedea9e580cb7089e Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed44086.png_7.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed44086.png_8.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed44086.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..72525db80484dc9cefff9a7e2d21aabea7c231d3 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed44086.png_8.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed44086.png_9.png b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed44086.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..e15f1fb75b3dcdcd3b8016596220a265b399b339 Binary files /dev/null and b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/20230605_201824/vis_data/vis_image/seed44086.png_9.png differ diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/epoch_1.pth b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/epoch_1.pth new file mode 100644 index 0000000000000000000000000000000000000000..1f4408408dc7592c31e4fc2d4cbb46ce76da75e9 --- /dev/null +++ b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/epoch_1.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:04b0c0b48ad17f0db26c89ccbf0702ef8a6434d7b9cca631e77bdbe96ab52b75 +size 277073997 diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/epoch_10.pth b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/epoch_10.pth new file mode 100644 index 0000000000000000000000000000000000000000..a47dc153c85dd0c5526f0a8462efcc193f5349ca --- /dev/null +++ b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/epoch_10.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:cb07cc93fdab08bc76b447f4aa9c7534cfe5cb4a039e31e9a8fb6f096c09cd63 +size 277405069 diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/epoch_2.pth b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/epoch_2.pth new file mode 100644 index 0000000000000000000000000000000000000000..e1ffa2a6c5853d4529dfe0edf042e3c3202f8c7f --- /dev/null +++ b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/epoch_2.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:5b4cb4d77d296df68cab45f321b327d8b3c992944e55048dc0506fb73f3b564d +size 277111565 diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/epoch_3.pth b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/epoch_3.pth new file mode 100644 index 0000000000000000000000000000000000000000..83f960bb838f62d0786b48c8a26af4e7b4c56eb2 --- /dev/null +++ b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/epoch_3.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:d3690783318ac4bec216ee16c0a38c1cb0e57ca14e28a82df9158a3a95860644 +size 277148237 diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/epoch_4.pth b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/epoch_4.pth new file mode 100644 index 0000000000000000000000000000000000000000..88a82c1d8c10cef7f7af6445eb877ffd41310067 --- /dev/null +++ b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/epoch_4.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:4536c0ade85122594ab6295ef94cb3984d3b487e15493aac3392b93a4d5d6fd1 +size 277184973 diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/epoch_5.pth b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/epoch_5.pth new file mode 100644 index 0000000000000000000000000000000000000000..d129f9bac1201b61545589d7cba929c12f2cc64c --- /dev/null +++ b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/epoch_5.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:51c6f23165ea179a360fed3fe03fe05ed528ab286f7424d68168e42d4fce809c +size 277221517 diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/epoch_6.pth b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/epoch_6.pth new file mode 100644 index 0000000000000000000000000000000000000000..b3c62c1970205f76306a97a559f7e508dd7ebf2d --- /dev/null +++ b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/epoch_6.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:0f5d1283b36443803a2188cbfa0f98b654a65f0481337217fb47e8866a7c3b96 +size 277258189 diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/epoch_7.pth b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/epoch_7.pth new file mode 100644 index 0000000000000000000000000000000000000000..838a160a5fdee82c6481895925ecf4e8be13b769 --- /dev/null +++ b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/epoch_7.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:d0475ce9817678888a767e886c5e4f68419298721f069b05b846541b32bdbec9 +size 277294733 diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/epoch_8.pth b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/epoch_8.pth new file mode 100644 index 0000000000000000000000000000000000000000..ef768daf1f13e5dd6a8e191ada3ac4ba97bce718 --- /dev/null +++ b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/epoch_8.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:3cc13e2c27772cd87f1b7649e66856b79d35e257d56b7bcd14064a99ef453841 +size 277331405 diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/epoch_9.pth b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/epoch_9.pth new file mode 100644 index 0000000000000000000000000000000000000000..58625743a77002ae57b12b83b6efe83691349889 --- /dev/null +++ b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/epoch_9.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:8662271c92825621488c1870409d179665816999c0860acf321363bad93ed180 +size 277368141 diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/last_checkpoint b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/last_checkpoint new file mode 100644 index 0000000000000000000000000000000000000000..fd7404d2aca48547d1f33c46be77c304283cab20 --- /dev/null +++ b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/last_checkpoint @@ -0,0 +1 @@ +/mnt/petrelfs/luzeyu/workspace/fakebench/mmpretrain/workdir/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/epoch_10.pth \ No newline at end of file diff --git a/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1.py b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1.py new file mode 100644 index 0000000000000000000000000000000000000000..5a015cf83c3e0bc25c3f53ad1ff46ae4ceceabeb --- /dev/null +++ b/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1.py @@ -0,0 +1,230 @@ +optim_wrapper = dict( + optimizer=dict( + type='SGD', + lr=0.0005, + momentum=0.9, + weight_decay=0.0001, + _scope_='mmpretrain'), + clip_grad=None) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=False, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=512) +model = dict( + type='ImageClassifier', + backbone=dict( + frozen_stages=2, + type='ResNet', + depth=50, + num_stages=4, + out_indices=(3, ), + style='pytorch', + init_cfg=dict( + type='Pretrained', + checkpoint= + 'https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth', + prefix='backbone')), + neck=dict(type='GlobalAveragePooling'), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=2048, + loss=dict(type='CrossEntropyLoss', loss_weight=1.0), + topk=1)) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3fake8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3real8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3fake8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3real8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = dict(type='Accuracy', topk=1) +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='ConcatDataset', + datasets=[ + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3fake8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]), + dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stylegan3real8w.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='JPEG', compress_val=65, prob=0.5), + dict(type='GaussianBlur', radius=1.5, prob=0.5), + dict(type='PackInputs') + ]) + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = dict(type='Accuracy', topk=1) +custom_hooks = [dict(type='EMAHook', momentum=0.0001, priority='ABOVE_NORMAL')] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/resnet50_2xb256_stylegan3_1m_lr5e-4_aug_5e-1' diff --git a/swin_base_8xb128_1e-3lr_5m/20230531_183241/20230531_183241.log b/swin_base_8xb128_1e-3lr_5m/20230531_183241/20230531_183241.log new file mode 100644 index 0000000000000000000000000000000000000000..8d7f3fc842f5e244e9e1a87d6149cee1ba7bb285 --- /dev/null +++ b/swin_base_8xb128_1e-3lr_5m/20230531_183241/20230531_183241.log @@ -0,0 +1,2871 @@ +2023/05/31 18:32:47 - mmengine - INFO - +------------------------------------------------------------ +System environment: + sys.platform: linux + Python: 3.10.9 (main, Mar 8 2023, 10:47:38) [GCC 11.2.0] + CUDA available: True + numpy_random_seed: 530487989 + GPU 0,1,2,3,4,5,6,7: NVIDIA A100-SXM4-80GB + CUDA_HOME: /mnt/petrelfs/share/cuda-11.6 + NVCC: Cuda compilation tools, release 11.6, V11.6.124 + GCC: gcc (GCC) 4.8.5 20150623 (Red Hat 4.8.5-44) + PyTorch: 1.13.1 + PyTorch compiling details: PyTorch built with: + - GCC 9.3 + - C++ Version: 201402 + - Intel(R) oneAPI Math Kernel Library Version 2021.4-Product Build 20210904 for Intel(R) 64 architecture applications + - Intel(R) MKL-DNN v2.6.0 (Git Hash 52b5f107dd9cf10910aaa19cb47f3abf9b349815) + - OpenMP 201511 (a.k.a. OpenMP 4.5) + - LAPACK is enabled (usually provided by MKL) + - NNPACK is enabled + - CPU capability usage: AVX2 + - CUDA Runtime 11.6 + - NVCC architecture flags: -gencode;arch=compute_37,code=sm_37;-gencode;arch=compute_50,code=sm_50;-gencode;arch=compute_60,code=sm_60;-gencode;arch=compute_61,code=sm_61;-gencode;arch=compute_70,code=sm_70;-gencode;arch=compute_75,code=sm_75;-gencode;arch=compute_80,code=sm_80;-gencode;arch=compute_86,code=sm_86;-gencode;arch=compute_37,code=compute_37 + - CuDNN 8.3.2 (built against CUDA 11.5) + - Magma 2.6.1 + - Build settings: BLAS_INFO=mkl, BUILD_TYPE=Release, CUDA_VERSION=11.6, CUDNN_VERSION=8.3.2, CXX_COMPILER=/opt/rh/devtoolset-9/root/usr/bin/c++, CXX_FLAGS= -fabi-version=11 -Wno-deprecated -fvisibility-inlines-hidden -DUSE_PTHREADPOOL -fopenmp -DNDEBUG -DUSE_KINETO -DUSE_FBGEMM -DUSE_QNNPACK -DUSE_PYTORCH_QNNPACK -DUSE_XNNPACK -DSYMBOLICATE_MOBILE_DEBUG_HANDLE -DEDGE_PROFILER_USE_KINETO -O2 -fPIC -Wno-narrowing -Wall -Wextra -Werror=return-type -Werror=non-virtual-dtor -Wno-missing-field-initializers -Wno-type-limits -Wno-array-bounds -Wno-unknown-pragmas -Wunused-local-typedefs -Wno-unused-parameter -Wno-unused-function -Wno-unused-result -Wno-strict-overflow -Wno-strict-aliasing -Wno-error=deprecated-declarations -Wno-stringop-overflow -Wno-psabi -Wno-error=pedantic -Wno-error=redundant-decls -Wno-error=old-style-cast -fdiagnostics-color=always -faligned-new -Wno-unused-but-set-variable -Wno-maybe-uninitialized -fno-math-errno -fno-trapping-math -Werror=format -Werror=cast-function-type -Wno-stringop-overflow, LAPACK_INFO=mkl, PERF_WITH_AVX=1, PERF_WITH_AVX2=1, PERF_WITH_AVX512=1, TORCH_VERSION=1.13.1, USE_CUDA=ON, USE_CUDNN=ON, USE_EXCEPTION_PTR=1, USE_GFLAGS=OFF, USE_GLOG=OFF, USE_MKL=ON, USE_MKLDNN=ON, USE_MPI=OFF, USE_NCCL=ON, USE_NNPACK=ON, USE_OPENMP=ON, USE_ROCM=OFF, + + TorchVision: 0.14.1 + OpenCV: 4.7.0 + MMEngine: 0.7.3 + +Runtime environment: + cudnn_benchmark: True + mp_cfg: {'mp_start_method': 'fork', 'opencv_num_threads': 0} + dist_cfg: {'backend': 'nccl'} + seed: None + deterministic: False + Distributed launcher: slurm + Distributed training: True + GPU number: 8 +------------------------------------------------------------ + +2023/05/31 18:32:48 - mmengine - INFO - Config: +optim_wrapper = dict( + optimizer=dict( + type='AdamW', + lr=0.001, + weight_decay=0.05, + eps=1e-08, + betas=(0.9, 0.999), + _scope_='mmpretrain'), + paramwise_cfg=dict( + norm_decay_mult=0.0, + bias_decay_mult=0.0, + flat_decay_mult=0.0, + custom_keys=dict({ + '.absolute_pos_embed': dict(decay_mult=0.0), + '.relative_position_bias_table': dict(decay_mult=0.0) + })), + type='AmpOptimWrapper', + dtype='bfloat16', + clip_grad=dict(max_norm=5.0)) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=True, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=20, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=1024) +model = dict( + type='ImageClassifier', + backbone=dict( + type='SwinTransformer', arch='base', img_size=224, drop_path_rate=0.5), + neck=dict(type='GlobalAveragePooling'), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=1024, + init_cfg=None, + loss=dict( + type='LabelSmoothLoss', label_smooth_val=0.1, mode='original'), + cal_acc=False), + init_cfg=[ + dict(type='TruncNormal', layer='Linear', std=0.02, bias=0.0), + dict(type='Constant', layer='LayerNorm', val=1.0, bias=0.0) + ], + train_cfg=None) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=128, + num_workers=10, + dataset=dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stablediffusionV1-5R2-dpmsolver-25-5m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/swin_base_8xb128_1e-3lr_5m' + +2023/05/31 18:32:52 - mmengine - INFO - Hooks will be executed in the following order: +before_run: +(VERY_HIGH ) RuntimeInfoHook +(BELOW_NORMAL) LoggerHook + -------------------- +before_train: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_train_epoch: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(NORMAL ) DistSamplerSeedHook + -------------------- +before_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook + -------------------- +after_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +after_train_epoch: +(NORMAL ) IterTimerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_val_epoch: +(NORMAL ) IterTimerHook + -------------------- +before_val_iter: +(NORMAL ) IterTimerHook + -------------------- +after_val_iter: +(NORMAL ) IterTimerHook +(NORMAL ) VisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_val_epoch: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +after_train: +(VERY_LOW ) CheckpointHook + -------------------- +before_test_epoch: +(NORMAL ) IterTimerHook + -------------------- +before_test_iter: +(NORMAL ) IterTimerHook + -------------------- +after_test_iter: +(NORMAL ) IterTimerHook +(NORMAL ) VisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_test_epoch: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_run: +(BELOW_NORMAL) LoggerHook + -------------------- +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.patch_embed.projection.bias:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.patch_embed.norm.weight:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.patch_embed.norm.bias:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.0.blocks.0.norm1.weight:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.0.blocks.0.norm1.bias:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.0.blocks.0.attn.w_msa.relative_position_bias_table:lr=0.001 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.0.blocks.0.attn.w_msa.relative_position_bias_table:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.0.blocks.0.attn.w_msa.relative_position_bias_table:decay_mult=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.0.blocks.0.attn.w_msa.qkv.bias:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.0.blocks.0.attn.w_msa.proj.bias:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.0.blocks.0.norm2.weight:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.0.blocks.0.norm2.bias:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.0.blocks.0.ffn.layers.0.0.bias:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.0.blocks.0.ffn.layers.1.bias:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.0.blocks.1.norm1.weight:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.0.blocks.1.norm1.bias:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.0.blocks.1.attn.w_msa.relative_position_bias_table:lr=0.001 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.0.blocks.1.attn.w_msa.relative_position_bias_table:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.0.blocks.1.attn.w_msa.relative_position_bias_table:decay_mult=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.0.blocks.1.attn.w_msa.qkv.bias:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.0.blocks.1.attn.w_msa.proj.bias:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.0.blocks.1.norm2.weight:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.0.blocks.1.norm2.bias:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.0.blocks.1.ffn.layers.0.0.bias:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.0.blocks.1.ffn.layers.1.bias:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.0.downsample.norm.weight:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.0.downsample.norm.bias:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.1.blocks.0.norm1.weight:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.1.blocks.0.norm1.bias:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.1.blocks.0.attn.w_msa.relative_position_bias_table:lr=0.001 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.1.blocks.0.attn.w_msa.relative_position_bias_table:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.1.blocks.0.attn.w_msa.relative_position_bias_table:decay_mult=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.1.blocks.0.attn.w_msa.qkv.bias:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.1.blocks.0.attn.w_msa.proj.bias:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.1.blocks.0.norm2.weight:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.1.blocks.0.norm2.bias:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.1.blocks.0.ffn.layers.0.0.bias:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.1.blocks.0.ffn.layers.1.bias:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.1.blocks.1.norm1.weight:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.1.blocks.1.norm1.bias:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.1.blocks.1.attn.w_msa.relative_position_bias_table:lr=0.001 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.1.blocks.1.attn.w_msa.relative_position_bias_table:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.1.blocks.1.attn.w_msa.relative_position_bias_table:decay_mult=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.1.blocks.1.attn.w_msa.qkv.bias:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.1.blocks.1.attn.w_msa.proj.bias:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.1.blocks.1.norm2.weight:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.1.blocks.1.norm2.bias:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.1.blocks.1.ffn.layers.0.0.bias:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.1.blocks.1.ffn.layers.1.bias:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.1.downsample.norm.weight:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.1.downsample.norm.bias:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.0.norm1.weight:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.0.norm1.bias:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.0.attn.w_msa.relative_position_bias_table:lr=0.001 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.0.attn.w_msa.relative_position_bias_table:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.0.attn.w_msa.relative_position_bias_table:decay_mult=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.0.attn.w_msa.qkv.bias:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.0.attn.w_msa.proj.bias:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.0.norm2.weight:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.0.norm2.bias:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.0.ffn.layers.0.0.bias:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.0.ffn.layers.1.bias:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.1.norm1.weight:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.1.norm1.bias:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.1.attn.w_msa.relative_position_bias_table:lr=0.001 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.1.attn.w_msa.relative_position_bias_table:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.1.attn.w_msa.relative_position_bias_table:decay_mult=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.1.attn.w_msa.qkv.bias:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.1.attn.w_msa.proj.bias:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.1.norm2.weight:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.1.norm2.bias:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.1.ffn.layers.0.0.bias:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.1.ffn.layers.1.bias:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.2.norm1.weight:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.2.norm1.bias:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.2.attn.w_msa.relative_position_bias_table:lr=0.001 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.2.attn.w_msa.relative_position_bias_table:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.2.attn.w_msa.relative_position_bias_table:decay_mult=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.2.attn.w_msa.qkv.bias:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.2.attn.w_msa.proj.bias:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.2.norm2.weight:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.2.norm2.bias:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.2.ffn.layers.0.0.bias:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.2.ffn.layers.1.bias:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.3.norm1.weight:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.3.norm1.bias:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.3.attn.w_msa.relative_position_bias_table:lr=0.001 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.3.attn.w_msa.relative_position_bias_table:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.3.attn.w_msa.relative_position_bias_table:decay_mult=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.3.attn.w_msa.qkv.bias:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.3.attn.w_msa.proj.bias:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.3.norm2.weight:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.3.norm2.bias:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.3.ffn.layers.0.0.bias:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.3.ffn.layers.1.bias:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.4.norm1.weight:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.4.norm1.bias:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.4.attn.w_msa.relative_position_bias_table:lr=0.001 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.4.attn.w_msa.relative_position_bias_table:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.4.attn.w_msa.relative_position_bias_table:decay_mult=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.4.attn.w_msa.qkv.bias:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.4.attn.w_msa.proj.bias:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.4.norm2.weight:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.4.norm2.bias:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.4.ffn.layers.0.0.bias:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.4.ffn.layers.1.bias:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.5.norm1.weight:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.5.norm1.bias:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.5.attn.w_msa.relative_position_bias_table:lr=0.001 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.5.attn.w_msa.relative_position_bias_table:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.5.attn.w_msa.relative_position_bias_table:decay_mult=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.5.attn.w_msa.qkv.bias:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.5.attn.w_msa.proj.bias:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.5.norm2.weight:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.5.norm2.bias:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.5.ffn.layers.0.0.bias:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.5.ffn.layers.1.bias:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.6.norm1.weight:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.6.norm1.bias:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.6.attn.w_msa.relative_position_bias_table:lr=0.001 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.6.attn.w_msa.relative_position_bias_table:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.6.attn.w_msa.relative_position_bias_table:decay_mult=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.6.attn.w_msa.qkv.bias:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.6.attn.w_msa.proj.bias:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.6.norm2.weight:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.6.norm2.bias:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.6.ffn.layers.0.0.bias:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.6.ffn.layers.1.bias:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.7.norm1.weight:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.7.norm1.bias:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.7.attn.w_msa.relative_position_bias_table:lr=0.001 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.7.attn.w_msa.relative_position_bias_table:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.7.attn.w_msa.relative_position_bias_table:decay_mult=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.7.attn.w_msa.qkv.bias:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.7.attn.w_msa.proj.bias:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.7.norm2.weight:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.7.norm2.bias:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.7.ffn.layers.0.0.bias:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.7.ffn.layers.1.bias:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.8.norm1.weight:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.8.norm1.bias:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.8.attn.w_msa.relative_position_bias_table:lr=0.001 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.8.attn.w_msa.relative_position_bias_table:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.8.attn.w_msa.relative_position_bias_table:decay_mult=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.8.attn.w_msa.qkv.bias:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.8.attn.w_msa.proj.bias:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.8.norm2.weight:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.8.norm2.bias:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.8.ffn.layers.0.0.bias:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.8.ffn.layers.1.bias:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.9.norm1.weight:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.9.norm1.bias:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.9.attn.w_msa.relative_position_bias_table:lr=0.001 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.9.attn.w_msa.relative_position_bias_table:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.9.attn.w_msa.relative_position_bias_table:decay_mult=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.9.attn.w_msa.qkv.bias:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.9.attn.w_msa.proj.bias:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.9.norm2.weight:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.9.norm2.bias:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.9.ffn.layers.0.0.bias:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.9.ffn.layers.1.bias:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.10.norm1.weight:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.10.norm1.bias:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.10.attn.w_msa.relative_position_bias_table:lr=0.001 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.10.attn.w_msa.relative_position_bias_table:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.10.attn.w_msa.relative_position_bias_table:decay_mult=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.10.attn.w_msa.qkv.bias:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.10.attn.w_msa.proj.bias:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.10.norm2.weight:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.10.norm2.bias:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.10.ffn.layers.0.0.bias:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.10.ffn.layers.1.bias:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.11.norm1.weight:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.11.norm1.bias:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.11.attn.w_msa.relative_position_bias_table:lr=0.001 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.11.attn.w_msa.relative_position_bias_table:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.11.attn.w_msa.relative_position_bias_table:decay_mult=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.11.attn.w_msa.qkv.bias:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.11.attn.w_msa.proj.bias:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.11.norm2.weight:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.11.norm2.bias:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.11.ffn.layers.0.0.bias:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.11.ffn.layers.1.bias:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.12.norm1.weight:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.12.norm1.bias:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.12.attn.w_msa.relative_position_bias_table:lr=0.001 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.12.attn.w_msa.relative_position_bias_table:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.12.attn.w_msa.relative_position_bias_table:decay_mult=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.12.attn.w_msa.qkv.bias:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.12.attn.w_msa.proj.bias:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.12.norm2.weight:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.12.norm2.bias:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.12.ffn.layers.0.0.bias:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.12.ffn.layers.1.bias:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.13.norm1.weight:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.13.norm1.bias:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.13.attn.w_msa.relative_position_bias_table:lr=0.001 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.13.attn.w_msa.relative_position_bias_table:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.13.attn.w_msa.relative_position_bias_table:decay_mult=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.13.attn.w_msa.qkv.bias:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.13.attn.w_msa.proj.bias:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.13.norm2.weight:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.13.norm2.bias:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.13.ffn.layers.0.0.bias:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.13.ffn.layers.1.bias:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.14.norm1.weight:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.14.norm1.bias:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.14.attn.w_msa.relative_position_bias_table:lr=0.001 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.14.attn.w_msa.relative_position_bias_table:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.14.attn.w_msa.relative_position_bias_table:decay_mult=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.14.attn.w_msa.qkv.bias:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.14.attn.w_msa.proj.bias:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.14.norm2.weight:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.14.norm2.bias:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.14.ffn.layers.0.0.bias:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.14.ffn.layers.1.bias:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.15.norm1.weight:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.15.norm1.bias:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.15.attn.w_msa.relative_position_bias_table:lr=0.001 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.15.attn.w_msa.relative_position_bias_table:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.15.attn.w_msa.relative_position_bias_table:decay_mult=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.15.attn.w_msa.qkv.bias:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.15.attn.w_msa.proj.bias:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.15.norm2.weight:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.15.norm2.bias:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.15.ffn.layers.0.0.bias:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.15.ffn.layers.1.bias:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.16.norm1.weight:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.16.norm1.bias:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.16.attn.w_msa.relative_position_bias_table:lr=0.001 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.16.attn.w_msa.relative_position_bias_table:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.16.attn.w_msa.relative_position_bias_table:decay_mult=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.16.attn.w_msa.qkv.bias:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.16.attn.w_msa.proj.bias:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.16.norm2.weight:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.16.norm2.bias:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.16.ffn.layers.0.0.bias:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.16.ffn.layers.1.bias:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.17.norm1.weight:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.17.norm1.bias:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.17.attn.w_msa.relative_position_bias_table:lr=0.001 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.17.attn.w_msa.relative_position_bias_table:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.17.attn.w_msa.relative_position_bias_table:decay_mult=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.17.attn.w_msa.qkv.bias:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.17.attn.w_msa.proj.bias:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.17.norm2.weight:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.17.norm2.bias:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.17.ffn.layers.0.0.bias:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.17.ffn.layers.1.bias:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.downsample.norm.weight:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.2.downsample.norm.bias:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.3.blocks.0.norm1.weight:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.3.blocks.0.norm1.bias:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.3.blocks.0.attn.w_msa.relative_position_bias_table:lr=0.001 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.3.blocks.0.attn.w_msa.relative_position_bias_table:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.3.blocks.0.attn.w_msa.relative_position_bias_table:decay_mult=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.3.blocks.0.attn.w_msa.qkv.bias:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.3.blocks.0.attn.w_msa.proj.bias:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.3.blocks.0.norm2.weight:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.3.blocks.0.norm2.bias:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.3.blocks.0.ffn.layers.0.0.bias:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.3.blocks.0.ffn.layers.1.bias:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.3.blocks.1.norm1.weight:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.3.blocks.1.norm1.bias:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.3.blocks.1.attn.w_msa.relative_position_bias_table:lr=0.001 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.3.blocks.1.attn.w_msa.relative_position_bias_table:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.3.blocks.1.attn.w_msa.relative_position_bias_table:decay_mult=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.3.blocks.1.attn.w_msa.qkv.bias:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.3.blocks.1.attn.w_msa.proj.bias:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.3.blocks.1.norm2.weight:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.3.blocks.1.norm2.bias:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.3.blocks.1.ffn.layers.0.0.bias:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.stages.3.blocks.1.ffn.layers.1.bias:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.norm3.weight:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- backbone.norm3.bias:weight_decay=0.0 +2023/05/31 18:33:19 - mmengine - INFO - paramwise_options -- head.fc.bias:weight_decay=0.0 +Name of parameter - Initialization information + +backbone.patch_embed.projection.weight - torch.Size([128, 3, 4, 4]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.patch_embed.projection.bias - torch.Size([128]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.patch_embed.norm.weight - torch.Size([128]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.patch_embed.norm.bias - torch.Size([128]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.0.blocks.0.norm1.weight - torch.Size([128]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.0.blocks.0.norm1.bias - torch.Size([128]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.0.blocks.0.attn.w_msa.relative_position_bias_table - torch.Size([169, 4]): +Initialized by user-defined `init_weights` in WindowMSA + +backbone.stages.0.blocks.0.attn.w_msa.qkv.weight - torch.Size([384, 128]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.blocks.0.attn.w_msa.qkv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.blocks.0.attn.w_msa.proj.weight - torch.Size([128, 128]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.blocks.0.attn.w_msa.proj.bias - torch.Size([128]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.blocks.0.norm2.weight - torch.Size([128]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.0.blocks.0.norm2.bias - torch.Size([128]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.0.blocks.0.ffn.layers.0.0.weight - torch.Size([512, 128]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.blocks.0.ffn.layers.0.0.bias - torch.Size([512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.blocks.0.ffn.layers.1.weight - torch.Size([128, 512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.blocks.0.ffn.layers.1.bias - torch.Size([128]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.blocks.1.norm1.weight - torch.Size([128]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.0.blocks.1.norm1.bias - torch.Size([128]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.0.blocks.1.attn.w_msa.relative_position_bias_table - torch.Size([169, 4]): +Initialized by user-defined `init_weights` in WindowMSA + +backbone.stages.0.blocks.1.attn.w_msa.qkv.weight - torch.Size([384, 128]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.blocks.1.attn.w_msa.qkv.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.blocks.1.attn.w_msa.proj.weight - torch.Size([128, 128]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.blocks.1.attn.w_msa.proj.bias - torch.Size([128]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.blocks.1.norm2.weight - torch.Size([128]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.0.blocks.1.norm2.bias - torch.Size([128]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.0.blocks.1.ffn.layers.0.0.weight - torch.Size([512, 128]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.blocks.1.ffn.layers.0.0.bias - torch.Size([512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.blocks.1.ffn.layers.1.weight - torch.Size([128, 512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.blocks.1.ffn.layers.1.bias - torch.Size([128]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.downsample.norm.weight - torch.Size([512]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.0.downsample.norm.bias - torch.Size([512]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.0.downsample.reduction.weight - torch.Size([256, 512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.blocks.0.norm1.weight - torch.Size([256]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.1.blocks.0.norm1.bias - torch.Size([256]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.1.blocks.0.attn.w_msa.relative_position_bias_table - torch.Size([169, 8]): +Initialized by user-defined `init_weights` in WindowMSA + +backbone.stages.1.blocks.0.attn.w_msa.qkv.weight - torch.Size([768, 256]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.blocks.0.attn.w_msa.qkv.bias - torch.Size([768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.blocks.0.attn.w_msa.proj.weight - torch.Size([256, 256]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.blocks.0.attn.w_msa.proj.bias - torch.Size([256]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.blocks.0.norm2.weight - torch.Size([256]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.1.blocks.0.norm2.bias - torch.Size([256]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.1.blocks.0.ffn.layers.0.0.weight - torch.Size([1024, 256]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.blocks.0.ffn.layers.0.0.bias - torch.Size([1024]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.blocks.0.ffn.layers.1.weight - torch.Size([256, 1024]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.blocks.0.ffn.layers.1.bias - torch.Size([256]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.blocks.1.norm1.weight - torch.Size([256]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.1.blocks.1.norm1.bias - torch.Size([256]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.1.blocks.1.attn.w_msa.relative_position_bias_table - torch.Size([169, 8]): +Initialized by user-defined `init_weights` in WindowMSA + +backbone.stages.1.blocks.1.attn.w_msa.qkv.weight - torch.Size([768, 256]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.blocks.1.attn.w_msa.qkv.bias - torch.Size([768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.blocks.1.attn.w_msa.proj.weight - torch.Size([256, 256]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.blocks.1.attn.w_msa.proj.bias - torch.Size([256]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.blocks.1.norm2.weight - torch.Size([256]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.1.blocks.1.norm2.bias - torch.Size([256]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.1.blocks.1.ffn.layers.0.0.weight - torch.Size([1024, 256]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.blocks.1.ffn.layers.0.0.bias - torch.Size([1024]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.blocks.1.ffn.layers.1.weight - torch.Size([256, 1024]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.blocks.1.ffn.layers.1.bias - torch.Size([256]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.downsample.norm.weight - torch.Size([1024]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.1.downsample.norm.bias - torch.Size([1024]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.1.downsample.reduction.weight - torch.Size([512, 1024]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.0.norm1.weight - torch.Size([512]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.0.norm1.bias - torch.Size([512]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.0.attn.w_msa.relative_position_bias_table - torch.Size([169, 16]): +Initialized by user-defined `init_weights` in WindowMSA + +backbone.stages.2.blocks.0.attn.w_msa.qkv.weight - torch.Size([1536, 512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.0.attn.w_msa.qkv.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.0.attn.w_msa.proj.weight - torch.Size([512, 512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.0.attn.w_msa.proj.bias - torch.Size([512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.0.norm2.weight - torch.Size([512]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.0.norm2.bias - torch.Size([512]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.0.ffn.layers.0.0.weight - torch.Size([2048, 512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.0.ffn.layers.0.0.bias - torch.Size([2048]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.0.ffn.layers.1.weight - torch.Size([512, 2048]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.0.ffn.layers.1.bias - torch.Size([512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.1.norm1.weight - torch.Size([512]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.1.norm1.bias - torch.Size([512]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.1.attn.w_msa.relative_position_bias_table - torch.Size([169, 16]): +Initialized by user-defined `init_weights` in WindowMSA + +backbone.stages.2.blocks.1.attn.w_msa.qkv.weight - torch.Size([1536, 512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.1.attn.w_msa.qkv.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.1.attn.w_msa.proj.weight - torch.Size([512, 512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.1.attn.w_msa.proj.bias - torch.Size([512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.1.norm2.weight - torch.Size([512]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.1.norm2.bias - torch.Size([512]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.1.ffn.layers.0.0.weight - torch.Size([2048, 512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.1.ffn.layers.0.0.bias - torch.Size([2048]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.1.ffn.layers.1.weight - torch.Size([512, 2048]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.1.ffn.layers.1.bias - torch.Size([512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.2.norm1.weight - torch.Size([512]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.2.norm1.bias - torch.Size([512]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.2.attn.w_msa.relative_position_bias_table - torch.Size([169, 16]): +Initialized by user-defined `init_weights` in WindowMSA + +backbone.stages.2.blocks.2.attn.w_msa.qkv.weight - torch.Size([1536, 512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.2.attn.w_msa.qkv.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.2.attn.w_msa.proj.weight - torch.Size([512, 512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.2.attn.w_msa.proj.bias - torch.Size([512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.2.norm2.weight - torch.Size([512]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.2.norm2.bias - torch.Size([512]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.2.ffn.layers.0.0.weight - torch.Size([2048, 512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.2.ffn.layers.0.0.bias - torch.Size([2048]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.2.ffn.layers.1.weight - torch.Size([512, 2048]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.2.ffn.layers.1.bias - torch.Size([512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.3.norm1.weight - torch.Size([512]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.3.norm1.bias - torch.Size([512]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.3.attn.w_msa.relative_position_bias_table - torch.Size([169, 16]): +Initialized by user-defined `init_weights` in WindowMSA + +backbone.stages.2.blocks.3.attn.w_msa.qkv.weight - torch.Size([1536, 512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.3.attn.w_msa.qkv.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.3.attn.w_msa.proj.weight - torch.Size([512, 512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.3.attn.w_msa.proj.bias - torch.Size([512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.3.norm2.weight - torch.Size([512]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.3.norm2.bias - torch.Size([512]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.3.ffn.layers.0.0.weight - torch.Size([2048, 512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.3.ffn.layers.0.0.bias - torch.Size([2048]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.3.ffn.layers.1.weight - torch.Size([512, 2048]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.3.ffn.layers.1.bias - torch.Size([512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.4.norm1.weight - torch.Size([512]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.4.norm1.bias - torch.Size([512]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.4.attn.w_msa.relative_position_bias_table - torch.Size([169, 16]): +Initialized by user-defined `init_weights` in WindowMSA + +backbone.stages.2.blocks.4.attn.w_msa.qkv.weight - torch.Size([1536, 512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.4.attn.w_msa.qkv.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.4.attn.w_msa.proj.weight - torch.Size([512, 512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.4.attn.w_msa.proj.bias - torch.Size([512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.4.norm2.weight - torch.Size([512]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.4.norm2.bias - torch.Size([512]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.4.ffn.layers.0.0.weight - torch.Size([2048, 512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.4.ffn.layers.0.0.bias - torch.Size([2048]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.4.ffn.layers.1.weight - torch.Size([512, 2048]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.4.ffn.layers.1.bias - torch.Size([512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.5.norm1.weight - torch.Size([512]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.5.norm1.bias - torch.Size([512]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.5.attn.w_msa.relative_position_bias_table - torch.Size([169, 16]): +Initialized by user-defined `init_weights` in WindowMSA + +backbone.stages.2.blocks.5.attn.w_msa.qkv.weight - torch.Size([1536, 512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.5.attn.w_msa.qkv.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.5.attn.w_msa.proj.weight - torch.Size([512, 512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.5.attn.w_msa.proj.bias - torch.Size([512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.5.norm2.weight - torch.Size([512]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.5.norm2.bias - torch.Size([512]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.5.ffn.layers.0.0.weight - torch.Size([2048, 512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.5.ffn.layers.0.0.bias - torch.Size([2048]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.5.ffn.layers.1.weight - torch.Size([512, 2048]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.5.ffn.layers.1.bias - torch.Size([512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.6.norm1.weight - torch.Size([512]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.6.norm1.bias - torch.Size([512]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.6.attn.w_msa.relative_position_bias_table - torch.Size([169, 16]): +Initialized by user-defined `init_weights` in WindowMSA + +backbone.stages.2.blocks.6.attn.w_msa.qkv.weight - torch.Size([1536, 512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.6.attn.w_msa.qkv.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.6.attn.w_msa.proj.weight - torch.Size([512, 512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.6.attn.w_msa.proj.bias - torch.Size([512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.6.norm2.weight - torch.Size([512]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.6.norm2.bias - torch.Size([512]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.6.ffn.layers.0.0.weight - torch.Size([2048, 512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.6.ffn.layers.0.0.bias - torch.Size([2048]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.6.ffn.layers.1.weight - torch.Size([512, 2048]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.6.ffn.layers.1.bias - torch.Size([512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.7.norm1.weight - torch.Size([512]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.7.norm1.bias - torch.Size([512]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.7.attn.w_msa.relative_position_bias_table - torch.Size([169, 16]): +Initialized by user-defined `init_weights` in WindowMSA + +backbone.stages.2.blocks.7.attn.w_msa.qkv.weight - torch.Size([1536, 512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.7.attn.w_msa.qkv.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.7.attn.w_msa.proj.weight - torch.Size([512, 512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.7.attn.w_msa.proj.bias - torch.Size([512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.7.norm2.weight - torch.Size([512]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.7.norm2.bias - torch.Size([512]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.7.ffn.layers.0.0.weight - torch.Size([2048, 512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.7.ffn.layers.0.0.bias - torch.Size([2048]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.7.ffn.layers.1.weight - torch.Size([512, 2048]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.7.ffn.layers.1.bias - torch.Size([512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.8.norm1.weight - torch.Size([512]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.8.norm1.bias - torch.Size([512]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.8.attn.w_msa.relative_position_bias_table - torch.Size([169, 16]): +Initialized by user-defined `init_weights` in WindowMSA + +backbone.stages.2.blocks.8.attn.w_msa.qkv.weight - torch.Size([1536, 512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.8.attn.w_msa.qkv.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.8.attn.w_msa.proj.weight - torch.Size([512, 512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.8.attn.w_msa.proj.bias - torch.Size([512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.8.norm2.weight - torch.Size([512]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.8.norm2.bias - torch.Size([512]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.8.ffn.layers.0.0.weight - torch.Size([2048, 512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.8.ffn.layers.0.0.bias - torch.Size([2048]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.8.ffn.layers.1.weight - torch.Size([512, 2048]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.8.ffn.layers.1.bias - torch.Size([512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.9.norm1.weight - torch.Size([512]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.9.norm1.bias - torch.Size([512]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.9.attn.w_msa.relative_position_bias_table - torch.Size([169, 16]): +Initialized by user-defined `init_weights` in WindowMSA + +backbone.stages.2.blocks.9.attn.w_msa.qkv.weight - torch.Size([1536, 512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.9.attn.w_msa.qkv.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.9.attn.w_msa.proj.weight - torch.Size([512, 512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.9.attn.w_msa.proj.bias - torch.Size([512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.9.norm2.weight - torch.Size([512]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.9.norm2.bias - torch.Size([512]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.9.ffn.layers.0.0.weight - torch.Size([2048, 512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.9.ffn.layers.0.0.bias - torch.Size([2048]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.9.ffn.layers.1.weight - torch.Size([512, 2048]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.9.ffn.layers.1.bias - torch.Size([512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.10.norm1.weight - torch.Size([512]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.10.norm1.bias - torch.Size([512]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.10.attn.w_msa.relative_position_bias_table - torch.Size([169, 16]): +Initialized by user-defined `init_weights` in WindowMSA + +backbone.stages.2.blocks.10.attn.w_msa.qkv.weight - torch.Size([1536, 512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.10.attn.w_msa.qkv.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.10.attn.w_msa.proj.weight - torch.Size([512, 512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.10.attn.w_msa.proj.bias - torch.Size([512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.10.norm2.weight - torch.Size([512]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.10.norm2.bias - torch.Size([512]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.10.ffn.layers.0.0.weight - torch.Size([2048, 512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.10.ffn.layers.0.0.bias - torch.Size([2048]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.10.ffn.layers.1.weight - torch.Size([512, 2048]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.10.ffn.layers.1.bias - torch.Size([512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.11.norm1.weight - torch.Size([512]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.11.norm1.bias - torch.Size([512]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.11.attn.w_msa.relative_position_bias_table - torch.Size([169, 16]): +Initialized by user-defined `init_weights` in WindowMSA + +backbone.stages.2.blocks.11.attn.w_msa.qkv.weight - torch.Size([1536, 512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.11.attn.w_msa.qkv.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.11.attn.w_msa.proj.weight - torch.Size([512, 512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.11.attn.w_msa.proj.bias - torch.Size([512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.11.norm2.weight - torch.Size([512]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.11.norm2.bias - torch.Size([512]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.11.ffn.layers.0.0.weight - torch.Size([2048, 512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.11.ffn.layers.0.0.bias - torch.Size([2048]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.11.ffn.layers.1.weight - torch.Size([512, 2048]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.11.ffn.layers.1.bias - torch.Size([512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.12.norm1.weight - torch.Size([512]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.12.norm1.bias - torch.Size([512]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.12.attn.w_msa.relative_position_bias_table - torch.Size([169, 16]): +Initialized by user-defined `init_weights` in WindowMSA + +backbone.stages.2.blocks.12.attn.w_msa.qkv.weight - torch.Size([1536, 512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.12.attn.w_msa.qkv.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.12.attn.w_msa.proj.weight - torch.Size([512, 512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.12.attn.w_msa.proj.bias - torch.Size([512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.12.norm2.weight - torch.Size([512]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.12.norm2.bias - torch.Size([512]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.12.ffn.layers.0.0.weight - torch.Size([2048, 512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.12.ffn.layers.0.0.bias - torch.Size([2048]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.12.ffn.layers.1.weight - torch.Size([512, 2048]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.12.ffn.layers.1.bias - torch.Size([512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.13.norm1.weight - torch.Size([512]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.13.norm1.bias - torch.Size([512]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.13.attn.w_msa.relative_position_bias_table - torch.Size([169, 16]): +Initialized by user-defined `init_weights` in WindowMSA + +backbone.stages.2.blocks.13.attn.w_msa.qkv.weight - torch.Size([1536, 512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.13.attn.w_msa.qkv.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.13.attn.w_msa.proj.weight - torch.Size([512, 512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.13.attn.w_msa.proj.bias - torch.Size([512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.13.norm2.weight - torch.Size([512]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.13.norm2.bias - torch.Size([512]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.13.ffn.layers.0.0.weight - torch.Size([2048, 512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.13.ffn.layers.0.0.bias - torch.Size([2048]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.13.ffn.layers.1.weight - torch.Size([512, 2048]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.13.ffn.layers.1.bias - torch.Size([512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.14.norm1.weight - torch.Size([512]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.14.norm1.bias - torch.Size([512]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.14.attn.w_msa.relative_position_bias_table - torch.Size([169, 16]): +Initialized by user-defined `init_weights` in WindowMSA + +backbone.stages.2.blocks.14.attn.w_msa.qkv.weight - torch.Size([1536, 512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.14.attn.w_msa.qkv.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.14.attn.w_msa.proj.weight - torch.Size([512, 512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.14.attn.w_msa.proj.bias - torch.Size([512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.14.norm2.weight - torch.Size([512]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.14.norm2.bias - torch.Size([512]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.14.ffn.layers.0.0.weight - torch.Size([2048, 512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.14.ffn.layers.0.0.bias - torch.Size([2048]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.14.ffn.layers.1.weight - torch.Size([512, 2048]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.14.ffn.layers.1.bias - torch.Size([512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.15.norm1.weight - torch.Size([512]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.15.norm1.bias - torch.Size([512]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.15.attn.w_msa.relative_position_bias_table - torch.Size([169, 16]): +Initialized by user-defined `init_weights` in WindowMSA + +backbone.stages.2.blocks.15.attn.w_msa.qkv.weight - torch.Size([1536, 512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.15.attn.w_msa.qkv.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.15.attn.w_msa.proj.weight - torch.Size([512, 512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.15.attn.w_msa.proj.bias - torch.Size([512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.15.norm2.weight - torch.Size([512]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.15.norm2.bias - torch.Size([512]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.15.ffn.layers.0.0.weight - torch.Size([2048, 512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.15.ffn.layers.0.0.bias - torch.Size([2048]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.15.ffn.layers.1.weight - torch.Size([512, 2048]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.15.ffn.layers.1.bias - torch.Size([512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.16.norm1.weight - torch.Size([512]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.16.norm1.bias - torch.Size([512]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.16.attn.w_msa.relative_position_bias_table - torch.Size([169, 16]): +Initialized by user-defined `init_weights` in WindowMSA + +backbone.stages.2.blocks.16.attn.w_msa.qkv.weight - torch.Size([1536, 512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.16.attn.w_msa.qkv.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.16.attn.w_msa.proj.weight - torch.Size([512, 512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.16.attn.w_msa.proj.bias - torch.Size([512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.16.norm2.weight - torch.Size([512]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.16.norm2.bias - torch.Size([512]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.16.ffn.layers.0.0.weight - torch.Size([2048, 512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.16.ffn.layers.0.0.bias - torch.Size([2048]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.16.ffn.layers.1.weight - torch.Size([512, 2048]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.16.ffn.layers.1.bias - torch.Size([512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.17.norm1.weight - torch.Size([512]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.17.norm1.bias - torch.Size([512]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.17.attn.w_msa.relative_position_bias_table - torch.Size([169, 16]): +Initialized by user-defined `init_weights` in WindowMSA + +backbone.stages.2.blocks.17.attn.w_msa.qkv.weight - torch.Size([1536, 512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.17.attn.w_msa.qkv.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.17.attn.w_msa.proj.weight - torch.Size([512, 512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.17.attn.w_msa.proj.bias - torch.Size([512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.17.norm2.weight - torch.Size([512]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.17.norm2.bias - torch.Size([512]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.17.ffn.layers.0.0.weight - torch.Size([2048, 512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.17.ffn.layers.0.0.bias - torch.Size([2048]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.17.ffn.layers.1.weight - torch.Size([512, 2048]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.17.ffn.layers.1.bias - torch.Size([512]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.downsample.norm.weight - torch.Size([2048]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.downsample.norm.bias - torch.Size([2048]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.downsample.reduction.weight - torch.Size([1024, 2048]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.blocks.0.norm1.weight - torch.Size([1024]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.3.blocks.0.norm1.bias - torch.Size([1024]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.3.blocks.0.attn.w_msa.relative_position_bias_table - torch.Size([169, 32]): +Initialized by user-defined `init_weights` in WindowMSA + +backbone.stages.3.blocks.0.attn.w_msa.qkv.weight - torch.Size([3072, 1024]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.blocks.0.attn.w_msa.qkv.bias - torch.Size([3072]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.blocks.0.attn.w_msa.proj.weight - torch.Size([1024, 1024]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.blocks.0.attn.w_msa.proj.bias - torch.Size([1024]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.blocks.0.norm2.weight - torch.Size([1024]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.3.blocks.0.norm2.bias - torch.Size([1024]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.3.blocks.0.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.blocks.0.ffn.layers.0.0.bias - torch.Size([4096]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.blocks.0.ffn.layers.1.weight - torch.Size([1024, 4096]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.blocks.0.ffn.layers.1.bias - torch.Size([1024]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.blocks.1.norm1.weight - torch.Size([1024]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.3.blocks.1.norm1.bias - torch.Size([1024]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.3.blocks.1.attn.w_msa.relative_position_bias_table - torch.Size([169, 32]): +Initialized by user-defined `init_weights` in WindowMSA + +backbone.stages.3.blocks.1.attn.w_msa.qkv.weight - torch.Size([3072, 1024]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.blocks.1.attn.w_msa.qkv.bias - torch.Size([3072]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.blocks.1.attn.w_msa.proj.weight - torch.Size([1024, 1024]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.blocks.1.attn.w_msa.proj.bias - torch.Size([1024]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.blocks.1.norm2.weight - torch.Size([1024]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.3.blocks.1.norm2.bias - torch.Size([1024]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.3.blocks.1.ffn.layers.0.0.weight - torch.Size([4096, 1024]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.blocks.1.ffn.layers.0.0.bias - torch.Size([4096]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.blocks.1.ffn.layers.1.weight - torch.Size([1024, 4096]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.blocks.1.ffn.layers.1.bias - torch.Size([1024]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.norm3.weight - torch.Size([1024]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.norm3.bias - torch.Size([1024]): +The value is the same before and after calling `init_weights` of ImageClassifier + +head.fc.weight - torch.Size([2, 1024]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +head.fc.bias - torch.Size([2]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 +2023/05/31 18:33:19 - mmengine - WARNING - "FileClient" will be deprecated in future. Please use io functions in https://mmengine.readthedocs.io/en/latest/api/fileio.html#file-io +2023/05/31 18:33:19 - mmengine - WARNING - "HardDiskBackend" is the alias of "LocalBackend" and the former will be deprecated in future. +2023/05/31 18:33:19 - mmengine - INFO - Checkpoints will be saved to /mnt/petrelfs/luzeyu/workspace/fakebench/mmpretrain/workdir/swin_base_8xb128_1e-3lr_5m. +2023/05/31 18:34:13 - mmengine - INFO - Epoch(train) [1][ 100/5758] lr: 1.0000e-03 eta: 17:01:09 time: 0.4505 data_time: 0.0015 memory: 20328 grad_norm: 1.2535 loss: 0.6617 +2023/05/31 18:34:59 - mmengine - INFO - Epoch(train) [1][ 200/5758] lr: 1.0000e-03 eta: 15:50:29 time: 0.4186 data_time: 0.0021 memory: 20327 grad_norm: 1.3020 loss: 0.6787 +2023/05/31 18:35:45 - mmengine - INFO - Epoch(train) [1][ 300/5758] lr: 1.0000e-03 eta: 15:31:14 time: 0.4118 data_time: 0.0016 memory: 20327 grad_norm: 1.6164 loss: 0.6962 +2023/05/31 18:36:32 - mmengine - INFO - Epoch(train) [1][ 400/5758] lr: 1.0000e-03 eta: 15:21:04 time: 0.4402 data_time: 0.0017 memory: 20327 grad_norm: 0.9162 loss: 0.6910 +2023/05/31 18:37:18 - mmengine - INFO - Epoch(train) [1][ 500/5758] lr: 1.0000e-03 eta: 15:10:26 time: 0.4888 data_time: 0.0025 memory: 20327 grad_norm: 0.7051 loss: 0.6896 +2023/05/31 18:38:04 - mmengine - INFO - Epoch(train) [1][ 600/5758] lr: 1.0000e-03 eta: 15:04:36 time: 0.4830 data_time: 0.0025 memory: 20327 grad_norm: 1.0663 loss: 0.6936 +2023/05/31 18:38:50 - mmengine - INFO - Epoch(train) [1][ 700/5758] lr: 1.0000e-03 eta: 15:00:17 time: 0.4459 data_time: 0.0024 memory: 20327 grad_norm: 0.6849 loss: 0.6929 +2023/05/31 18:39:35 - mmengine - INFO - Epoch(train) [1][ 800/5758] lr: 1.0000e-03 eta: 14:55:29 time: 0.4738 data_time: 0.0020 memory: 20327 grad_norm: 0.4387 loss: 0.6870 +2023/05/31 18:40:20 - mmengine - INFO - Epoch(train) [1][ 900/5758] lr: 1.0000e-03 eta: 14:49:50 time: 0.4850 data_time: 0.0022 memory: 20327 grad_norm: 0.7461 loss: 0.6842 +2023/05/31 18:41:05 - mmengine - INFO - Exp name: swin_base_8xb128_fake5m_20230531_183241 +2023/05/31 18:41:05 - mmengine - INFO - Epoch(train) [1][1000/5758] lr: 1.0000e-03 eta: 14:46:26 time: 0.5281 data_time: 0.0018 memory: 20327 grad_norm: 0.5114 loss: 0.6821 +2023/05/31 18:41:50 - mmengine - INFO - Epoch(train) [1][1100/5758] lr: 1.0000e-03 eta: 14:42:37 time: 0.4209 data_time: 0.0019 memory: 20327 grad_norm: 0.4384 loss: 0.6854 +2023/05/31 18:42:36 - mmengine - INFO - Epoch(train) [1][1200/5758] lr: 1.0000e-03 eta: 14:40:45 time: 0.4580 data_time: 0.0018 memory: 20327 grad_norm: 0.5817 loss: 0.6899 +2023/05/31 18:43:21 - mmengine - INFO - Epoch(train) [1][1300/5758] lr: 1.0000e-03 eta: 14:38:28 time: 0.4686 data_time: 0.0014 memory: 20327 grad_norm: 0.4683 loss: 0.6840 +2023/05/31 18:44:07 - mmengine - INFO - Epoch(train) [1][1400/5758] lr: 1.0000e-03 eta: 14:36:32 time: 0.4447 data_time: 0.0020 memory: 20327 grad_norm: 0.4274 loss: 0.6886 +2023/05/31 18:44:51 - mmengine - INFO - Epoch(train) [1][1500/5758] lr: 1.0000e-03 eta: 14:32:48 time: 0.4260 data_time: 0.0025 memory: 20327 grad_norm: 0.3759 loss: 0.6896 +2023/05/31 18:45:35 - mmengine - INFO - Epoch(train) [1][1600/5758] lr: 1.0000e-03 eta: 14:30:42 time: 0.4184 data_time: 0.0017 memory: 20327 grad_norm: 0.4764 loss: 0.6906 +2023/05/31 18:46:21 - mmengine - INFO - Epoch(train) [1][1700/5758] lr: 1.0000e-03 eta: 14:29:02 time: 0.4273 data_time: 0.0023 memory: 20327 grad_norm: 0.3174 loss: 0.6887 +2023/05/31 18:47:08 - mmengine - INFO - Epoch(train) [1][1800/5758] lr: 1.0000e-03 eta: 14:29:39 time: 0.4459 data_time: 0.0526 memory: 20327 grad_norm: 0.2392 loss: 0.6858 +2023/05/31 18:47:56 - mmengine - INFO - Epoch(train) [1][1900/5758] lr: 1.0000e-03 eta: 14:31:11 time: 0.5703 data_time: 0.0018 memory: 20327 grad_norm: 0.3035 loss: 0.6881 +2023/05/31 18:48:46 - mmengine - INFO - Exp name: swin_base_8xb128_fake5m_20230531_183241 +2023/05/31 18:48:46 - mmengine - INFO - Epoch(train) [1][2000/5758] lr: 1.0000e-03 eta: 14:33:21 time: 0.5233 data_time: 0.0017 memory: 20327 grad_norm: 0.2468 loss: 0.6844 +2023/05/31 18:49:38 - mmengine - INFO - Epoch(train) [1][2100/5758] lr: 1.0000e-03 eta: 14:38:07 time: 0.4493 data_time: 0.0016 memory: 20327 grad_norm: 0.2431 loss: 0.6869 +2023/05/31 18:50:28 - mmengine - INFO - Epoch(train) [1][2200/5758] lr: 1.0000e-03 eta: 14:40:04 time: 0.4997 data_time: 0.1156 memory: 20327 grad_norm: 0.2144 loss: 0.6881 +2023/05/31 18:51:19 - mmengine - INFO - Epoch(train) [1][2300/5758] lr: 1.0000e-03 eta: 14:42:47 time: 0.4611 data_time: 0.0017 memory: 20327 grad_norm: 0.3295 loss: 0.6890 +2023/05/31 18:52:16 - mmengine - INFO - Epoch(train) [1][2400/5758] lr: 1.0000e-03 eta: 14:50:09 time: 0.4768 data_time: 0.0015 memory: 20327 grad_norm: 0.3206 loss: 0.6834 +2023/05/31 18:53:12 - mmengine - INFO - Epoch(train) [1][2500/5758] lr: 1.0000e-03 eta: 14:55:23 time: 0.4431 data_time: 0.0020 memory: 20327 grad_norm: 0.1802 loss: 0.6857 +2023/05/31 18:54:03 - mmengine - INFO - Epoch(train) [1][2600/5758] lr: 1.0000e-03 eta: 14:57:08 time: 0.5280 data_time: 0.0015 memory: 20327 grad_norm: 0.2828 loss: 0.6845 +2023/05/31 18:54:50 - mmengine - INFO - Epoch(train) [1][2700/5758] lr: 1.0000e-03 eta: 14:55:37 time: 0.4529 data_time: 0.0015 memory: 20327 grad_norm: 0.2345 loss: 0.6897 +2023/05/31 18:55:38 - mmengine - INFO - Epoch(train) [1][2800/5758] lr: 1.0000e-03 eta: 14:55:23 time: 0.5384 data_time: 0.0018 memory: 20327 grad_norm: 0.1539 loss: 0.6886 +2023/05/31 18:56:27 - mmengine - INFO - Epoch(train) [1][2900/5758] lr: 1.0000e-03 eta: 14:55:04 time: 0.4375 data_time: 0.0015 memory: 20327 grad_norm: 0.2427 loss: 0.6880 +2023/05/31 18:57:15 - mmengine - INFO - Exp name: swin_base_8xb128_fake5m_20230531_183241 +2023/05/31 18:57:15 - mmengine - INFO - Epoch(train) [1][3000/5758] lr: 1.0000e-03 eta: 14:54:23 time: 0.4969 data_time: 0.0014 memory: 20327 grad_norm: 0.2302 loss: 0.6911 +2023/05/31 18:58:05 - mmengine - INFO - Epoch(train) [1][3100/5758] lr: 1.0000e-03 eta: 14:55:14 time: 0.5301 data_time: 0.0023 memory: 20327 grad_norm: 0.2076 loss: 0.6825 +2023/05/31 18:58:55 - mmengine - INFO - Epoch(train) [1][3200/5758] lr: 1.0000e-03 eta: 14:55:35 time: 0.5860 data_time: 0.0015 memory: 20327 grad_norm: 0.2537 loss: 0.6854 +2023/05/31 18:59:46 - mmengine - INFO - Epoch(train) [1][3300/5758] lr: 1.0000e-03 eta: 14:56:15 time: 0.4651 data_time: 0.0025 memory: 20327 grad_norm: 0.1809 loss: 0.6864 +2023/05/31 19:00:35 - mmengine - INFO - Epoch(train) [1][3400/5758] lr: 1.0000e-03 eta: 14:55:55 time: 0.4548 data_time: 0.0018 memory: 20327 grad_norm: 0.2350 loss: 0.6890 +2023/05/31 19:01:20 - mmengine - INFO - Epoch(train) [1][3500/5758] lr: 1.0000e-03 eta: 14:53:40 time: 0.4564 data_time: 0.0017 memory: 20327 grad_norm: 0.1906 loss: 0.6882 +2023/05/31 19:02:07 - mmengine - INFO - Epoch(train) [1][3600/5758] lr: 1.0000e-03 eta: 14:52:00 time: 0.4628 data_time: 0.0016 memory: 20327 grad_norm: 0.2133 loss: 0.6878 +2023/05/31 19:02:52 - mmengine - INFO - Epoch(train) [1][3700/5758] lr: 1.0000e-03 eta: 14:50:04 time: 0.4189 data_time: 0.0022 memory: 20327 grad_norm: 0.2188 loss: 0.6868 +2023/05/31 19:03:38 - mmengine - INFO - Epoch(train) [1][3800/5758] lr: 1.0000e-03 eta: 14:48:01 time: 0.4756 data_time: 0.0015 memory: 20327 grad_norm: 0.1568 loss: 0.6829 +2023/05/31 19:04:23 - mmengine - INFO - Epoch(train) [1][3900/5758] lr: 1.0000e-03 eta: 14:46:13 time: 0.5234 data_time: 0.0014 memory: 20327 grad_norm: 0.2446 loss: 0.6878 +2023/05/31 19:05:10 - mmengine - INFO - Exp name: swin_base_8xb128_fake5m_20230531_183241 +2023/05/31 19:05:10 - mmengine - INFO - Epoch(train) [1][4000/5758] lr: 1.0000e-03 eta: 14:44:57 time: 0.5318 data_time: 0.0016 memory: 20327 grad_norm: 0.2752 loss: 0.6842 +2023/05/31 19:05:57 - mmengine - INFO - Epoch(train) [1][4100/5758] lr: 1.0000e-03 eta: 14:43:44 time: 0.4641 data_time: 0.0014 memory: 20327 grad_norm: 0.1678 loss: 0.6814 +2023/05/31 19:06:48 - mmengine - INFO - Epoch(train) [1][4200/5758] lr: 1.0000e-03 eta: 14:44:16 time: 0.6016 data_time: 0.0015 memory: 20327 grad_norm: 0.2206 loss: 0.6909 +2023/05/31 19:07:39 - mmengine - INFO - Epoch(train) [1][4300/5758] lr: 1.0000e-03 eta: 14:44:53 time: 0.4643 data_time: 0.0018 memory: 20327 grad_norm: 0.1305 loss: 0.6935 +2023/05/31 19:08:32 - mmengine - INFO - Epoch(train) [1][4400/5758] lr: 1.0000e-03 eta: 14:46:22 time: 0.5172 data_time: 0.0018 memory: 20327 grad_norm: 0.1892 loss: 0.6804 +2023/05/31 19:09:26 - mmengine - INFO - Epoch(train) [1][4500/5758] lr: 1.0000e-03 eta: 14:47:47 time: 0.5839 data_time: 0.0019 memory: 20327 grad_norm: 4.1434 loss: 0.6887 +2023/05/31 19:10:18 - mmengine - INFO - Epoch(train) [1][4600/5758] lr: 1.0000e-03 eta: 14:48:38 time: 0.4826 data_time: 0.0444 memory: 20327 grad_norm: 0.2898 loss: 0.6888 +2023/05/31 19:11:11 - mmengine - INFO - Epoch(train) [1][4700/5758] lr: 1.0000e-03 eta: 14:49:48 time: 0.5790 data_time: 0.0118 memory: 20327 grad_norm: 0.2279 loss: 0.6894 +2023/05/31 19:11:59 - mmengine - INFO - Epoch(train) [1][4800/5758] lr: 1.0000e-03 eta: 14:48:54 time: 0.4071 data_time: 0.0430 memory: 20327 grad_norm: 0.1455 loss: 0.6874 +2023/05/31 19:12:45 - mmengine - INFO - Epoch(train) [1][4900/5758] lr: 1.0000e-03 eta: 14:47:08 time: 0.4641 data_time: 0.0017 memory: 20327 grad_norm: 0.2417 loss: 0.6905 +2023/05/31 19:13:31 - mmengine - INFO - Exp name: swin_base_8xb128_fake5m_20230531_183241 +2023/05/31 19:13:31 - mmengine - INFO - Epoch(train) [1][5000/5758] lr: 1.0000e-03 eta: 14:45:29 time: 0.4728 data_time: 0.0017 memory: 20327 grad_norm: 0.1863 loss: 0.6864 +2023/05/31 19:14:18 - mmengine - INFO - Epoch(train) [1][5100/5758] lr: 1.0000e-03 eta: 14:44:15 time: 0.4919 data_time: 0.0020 memory: 20327 grad_norm: 0.1691 loss: 0.6881 +2023/05/31 19:15:03 - mmengine - INFO - Epoch(train) [1][5200/5758] lr: 1.0000e-03 eta: 14:42:21 time: 0.4347 data_time: 0.0016 memory: 20327 grad_norm: 0.2515 loss: 0.6828 +2023/05/31 19:15:50 - mmengine - INFO - Epoch(train) [1][5300/5758] lr: 1.0000e-03 eta: 14:41:10 time: 0.4721 data_time: 0.0015 memory: 20327 grad_norm: 0.1911 loss: 0.6846 +2023/05/31 19:16:37 - mmengine - INFO - Epoch(train) [1][5400/5758] lr: 1.0000e-03 eta: 14:39:55 time: 0.4641 data_time: 0.0014 memory: 20327 grad_norm: 0.1328 loss: 0.6849 +2023/05/31 19:17:24 - mmengine - INFO - Epoch(train) [1][5500/5758] lr: 1.0000e-03 eta: 14:38:56 time: 0.5697 data_time: 0.0017 memory: 20327 grad_norm: 0.1780 loss: 0.6843 +2023/05/31 19:18:10 - mmengine - INFO - Epoch(train) [1][5600/5758] lr: 1.0000e-03 eta: 14:37:20 time: 0.3915 data_time: 0.0016 memory: 20327 grad_norm: 0.1942 loss: 0.6819 +2023/05/31 19:18:58 - mmengine - INFO - Epoch(train) [1][5700/5758] lr: 1.0000e-03 eta: 14:36:35 time: 0.5306 data_time: 0.0016 memory: 20327 grad_norm: 0.2729 loss: 0.6899 +2023/05/31 19:19:24 - mmengine - INFO - Exp name: swin_base_8xb128_fake5m_20230531_183241 +2023/05/31 19:19:24 - mmengine - INFO - Saving checkpoint at 1 epochs +2023/05/31 19:19:43 - mmengine - INFO - Epoch(val) [1][8/8] accuracy/top1: 100.0000 single-label/precision_classwise: [100.0, 0.0] single-label/recall_classwise: [100.0, 0.0] single-label/f1-score_classwise: [100.0, 0.0] data_time: 0.5274 time: 1.4208 +2023/05/31 19:20:35 - mmengine - INFO - Epoch(train) [2][ 100/5758] lr: 9.9391e-04 eta: 14:35:41 time: 0.4719 data_time: 0.0019 memory: 20338 grad_norm: 0.1390 loss: 0.6857 +2023/05/31 19:21:21 - mmengine - INFO - Epoch(train) [2][ 200/5758] lr: 9.9391e-04 eta: 14:34:21 time: 0.3786 data_time: 0.0015 memory: 20334 grad_norm: 0.2039 loss: 0.6875 +2023/05/31 19:21:39 - mmengine - INFO - Exp name: swin_base_8xb128_fake5m_20230531_183241 +2023/05/31 19:22:05 - mmengine - INFO - Epoch(train) [2][ 300/5758] lr: 9.9391e-04 eta: 14:32:13 time: 0.4775 data_time: 0.0023 memory: 20334 grad_norm: 0.1824 loss: 0.6888 +2023/05/31 19:22:48 - mmengine - INFO - Epoch(train) [2][ 400/5758] lr: 9.9391e-04 eta: 14:30:01 time: 0.4025 data_time: 0.0020 memory: 20334 grad_norm: 0.1098 loss: 0.6871 +2023/05/31 19:23:34 - mmengine - INFO - Epoch(train) [2][ 500/5758] lr: 9.9391e-04 eta: 14:28:41 time: 0.4650 data_time: 0.0020 memory: 20334 grad_norm: 0.1699 loss: 0.6880 +2023/05/31 19:24:20 - mmengine - INFO - Epoch(train) [2][ 600/5758] lr: 9.9391e-04 eta: 14:27:18 time: 0.4857 data_time: 0.0022 memory: 20334 grad_norm: 0.2250 loss: 0.6882 +2023/05/31 19:25:07 - mmengine - INFO - Epoch(train) [2][ 700/5758] lr: 9.9391e-04 eta: 14:26:28 time: 0.5223 data_time: 0.0028 memory: 20334 grad_norm: 0.0980 loss: 0.6891 +2023/05/31 19:25:55 - mmengine - INFO - Epoch(train) [2][ 800/5758] lr: 9.9391e-04 eta: 14:25:44 time: 0.4998 data_time: 0.0017 memory: 20334 grad_norm: 0.2047 loss: 0.6881 +2023/05/31 19:26:43 - mmengine - INFO - Epoch(train) [2][ 900/5758] lr: 9.9391e-04 eta: 14:24:48 time: 0.4293 data_time: 0.0015 memory: 20334 grad_norm: 0.1334 loss: 0.6904 +2023/05/31 19:27:30 - mmengine - INFO - Epoch(train) [2][1000/5758] lr: 9.9391e-04 eta: 14:23:42 time: 0.5042 data_time: 0.0023 memory: 20334 grad_norm: 0.1210 loss: 0.6860 +2023/05/31 19:28:16 - mmengine - INFO - Epoch(train) [2][1100/5758] lr: 9.9391e-04 eta: 14:22:41 time: 0.4424 data_time: 0.0016 memory: 20334 grad_norm: 0.1209 loss: 0.6867 +2023/05/31 19:29:04 - mmengine - INFO - Epoch(train) [2][1200/5758] lr: 9.9391e-04 eta: 14:21:42 time: 0.4532 data_time: 0.0015 memory: 20334 grad_norm: 0.1007 loss: 0.6839 +2023/05/31 19:29:23 - mmengine - INFO - Exp name: swin_base_8xb128_fake5m_20230531_183241 +2023/05/31 19:29:51 - mmengine - INFO - Epoch(train) [2][1300/5758] lr: 9.9391e-04 eta: 14:20:46 time: 0.4828 data_time: 0.0014 memory: 20334 grad_norm: 0.2354 loss: 0.6881 +2023/05/31 19:30:35 - mmengine - INFO - Epoch(train) [2][1400/5758] lr: 9.9391e-04 eta: 14:19:09 time: 0.4426 data_time: 0.0014 memory: 20334 grad_norm: 0.1225 loss: 0.6852 +2023/05/31 19:31:21 - mmengine - INFO - Epoch(train) [2][1500/5758] lr: 9.9391e-04 eta: 14:17:49 time: 0.4048 data_time: 0.0016 memory: 20334 grad_norm: 0.1605 loss: 0.6865 +2023/05/31 19:32:06 - mmengine - INFO - Epoch(train) [2][1600/5758] lr: 9.9391e-04 eta: 14:16:20 time: 0.3803 data_time: 0.0014 memory: 20334 grad_norm: 0.0794 loss: 0.6857 +2023/05/31 19:32:50 - mmengine - INFO - Epoch(train) [2][1700/5758] lr: 9.9391e-04 eta: 14:14:44 time: 0.4127 data_time: 0.0015 memory: 20334 grad_norm: 0.1306 loss: 0.6866 +2023/05/31 19:33:35 - mmengine - INFO - Epoch(train) [2][1800/5758] lr: 9.9391e-04 eta: 14:13:21 time: 0.4324 data_time: 0.0015 memory: 20334 grad_norm: 0.1867 loss: 0.6870 +2023/05/31 19:34:20 - mmengine - INFO - Epoch(train) [2][1900/5758] lr: 9.9391e-04 eta: 14:11:48 time: 0.4226 data_time: 0.0015 memory: 20334 grad_norm: 0.1899 loss: 0.6858 +2023/05/31 19:35:05 - mmengine - INFO - Epoch(train) [2][2000/5758] lr: 9.9391e-04 eta: 14:10:28 time: 0.4057 data_time: 0.0016 memory: 20334 grad_norm: 0.1243 loss: 0.6885 +2023/05/31 19:35:50 - mmengine - INFO - Epoch(train) [2][2100/5758] lr: 9.9391e-04 eta: 14:09:09 time: 0.4477 data_time: 0.0015 memory: 20334 grad_norm: 0.0797 loss: 0.6879 +2023/05/31 19:36:32 - mmengine - INFO - Epoch(train) [2][2200/5758] lr: 9.9391e-04 eta: 14:07:08 time: 0.5015 data_time: 0.0018 memory: 20334 grad_norm: 0.1389 loss: 0.6910 +2023/05/31 19:36:51 - mmengine - INFO - Exp name: swin_base_8xb128_fake5m_20230531_183241 +2023/05/31 19:37:15 - mmengine - INFO - Epoch(train) [2][2300/5758] lr: 9.9391e-04 eta: 14:05:21 time: 0.3787 data_time: 0.0016 memory: 20334 grad_norm: 0.0999 loss: 0.6889 +2023/05/31 19:37:58 - mmengine - INFO - Epoch(train) [2][2400/5758] lr: 9.9391e-04 eta: 14:03:31 time: 0.4515 data_time: 0.0015 memory: 20334 grad_norm: 0.0948 loss: 0.6863 +2023/05/31 19:38:40 - mmengine - INFO - Epoch(train) [2][2500/5758] lr: 9.9391e-04 eta: 14:01:44 time: 0.4425 data_time: 0.0016 memory: 20334 grad_norm: 0.1123 loss: 0.6916 +2023/05/31 19:39:22 - mmengine - INFO - Epoch(train) [2][2600/5758] lr: 9.9391e-04 eta: 13:59:44 time: 0.4332 data_time: 0.0016 memory: 20334 grad_norm: 0.1154 loss: 0.6823 +2023/05/31 19:40:04 - mmengine - INFO - Epoch(train) [2][2700/5758] lr: 9.9391e-04 eta: 13:57:53 time: 0.4113 data_time: 0.0015 memory: 20334 grad_norm: 0.1471 loss: 0.6900 +2023/05/31 19:40:46 - mmengine - INFO - Epoch(train) [2][2800/5758] lr: 9.9391e-04 eta: 13:56:04 time: 0.3838 data_time: 0.0015 memory: 20334 grad_norm: 0.1072 loss: 0.6860 +2023/05/31 19:41:29 - mmengine - INFO - Epoch(train) [2][2900/5758] lr: 9.9391e-04 eta: 13:54:28 time: 0.4390 data_time: 0.0016 memory: 20334 grad_norm: 0.1075 loss: 0.6898 +2023/05/31 19:42:10 - mmengine - INFO - Epoch(train) [2][3000/5758] lr: 9.9391e-04 eta: 13:52:26 time: 0.3877 data_time: 0.0016 memory: 20334 grad_norm: 0.1138 loss: 0.6888 +2023/05/31 19:42:52 - mmengine - INFO - Epoch(train) [2][3100/5758] lr: 9.9391e-04 eta: 13:50:39 time: 0.3935 data_time: 0.0017 memory: 20334 grad_norm: 0.0867 loss: 0.6898 +2023/05/31 19:43:34 - mmengine - INFO - Epoch(train) [2][3200/5758] lr: 9.9391e-04 eta: 13:48:50 time: 0.4575 data_time: 0.0016 memory: 20334 grad_norm: 0.1330 loss: 0.6889 +2023/05/31 19:43:53 - mmengine - INFO - Exp name: swin_base_8xb128_fake5m_20230531_183241 +2023/05/31 19:44:17 - mmengine - INFO - Epoch(train) [2][3300/5758] lr: 9.9391e-04 eta: 13:47:20 time: 0.4360 data_time: 0.0018 memory: 20334 grad_norm: 0.0948 loss: 0.6842 +2023/05/31 19:44:59 - mmengine - INFO - Epoch(train) [2][3400/5758] lr: 9.9391e-04 eta: 13:45:46 time: 0.4034 data_time: 0.0017 memory: 20334 grad_norm: 0.1201 loss: 0.6888 +2023/05/31 19:45:39 - mmengine - INFO - Epoch(train) [2][3500/5758] lr: 9.9391e-04 eta: 13:43:43 time: 0.3864 data_time: 0.0017 memory: 20334 grad_norm: 0.1380 loss: 0.6826 +2023/05/31 19:46:19 - mmengine - INFO - Epoch(train) [2][3600/5758] lr: 9.9391e-04 eta: 13:41:41 time: 0.4217 data_time: 0.0016 memory: 20334 grad_norm: 0.0715 loss: 0.6875 +2023/05/31 19:46:58 - mmengine - INFO - Epoch(train) [2][3700/5758] lr: 9.9391e-04 eta: 13:39:27 time: 0.4025 data_time: 0.0016 memory: 20334 grad_norm: 0.0821 loss: 0.6869 +2023/05/31 19:47:38 - mmengine - INFO - Epoch(train) [2][3800/5758] lr: 9.9391e-04 eta: 13:37:21 time: 0.3974 data_time: 0.0014 memory: 20334 grad_norm: 0.0815 loss: 0.6842 +2023/05/31 19:48:17 - mmengine - INFO - Epoch(train) [2][3900/5758] lr: 9.9391e-04 eta: 13:35:14 time: 0.4269 data_time: 0.0016 memory: 20334 grad_norm: 0.0614 loss: 0.6825 +2023/05/31 19:48:58 - mmengine - INFO - Epoch(train) [2][4000/5758] lr: 9.9391e-04 eta: 13:33:29 time: 0.4035 data_time: 0.0018 memory: 20334 grad_norm: 0.0564 loss: 0.6930 +2023/05/31 19:49:36 - mmengine - INFO - Epoch(train) [2][4100/5758] lr: 9.9391e-04 eta: 13:31:19 time: 0.4068 data_time: 0.0016 memory: 20334 grad_norm: 0.0691 loss: 0.6821 +2023/05/31 19:50:15 - mmengine - INFO - Epoch(train) [2][4200/5758] lr: 9.9391e-04 eta: 13:29:18 time: 0.4096 data_time: 0.0016 memory: 20334 grad_norm: 0.0648 loss: 0.6846 +2023/05/31 19:50:32 - mmengine - INFO - Exp name: swin_base_8xb128_fake5m_20230531_183241 +2023/05/31 19:50:55 - mmengine - INFO - Epoch(train) [2][4300/5758] lr: 9.9391e-04 eta: 13:27:23 time: 0.4189 data_time: 0.0018 memory: 20334 grad_norm: 0.0824 loss: 0.6868 +2023/05/31 19:51:35 - mmengine - INFO - Epoch(train) [2][4400/5758] lr: 9.9391e-04 eta: 13:25:34 time: 0.4179 data_time: 0.0019 memory: 20334 grad_norm: 0.0776 loss: 0.6857 +2023/05/31 19:52:15 - mmengine - INFO - Epoch(train) [2][4500/5758] lr: 9.9391e-04 eta: 13:23:51 time: 0.4282 data_time: 0.0015 memory: 20334 grad_norm: 0.0703 loss: 0.6881 +2023/05/31 19:52:56 - mmengine - INFO - Epoch(train) [2][4600/5758] lr: 9.9391e-04 eta: 13:22:08 time: 0.3999 data_time: 0.0016 memory: 20334 grad_norm: 0.0567 loss: 0.6862 +2023/05/31 19:53:35 - mmengine - INFO - Epoch(train) [2][4700/5758] lr: 9.9391e-04 eta: 13:20:16 time: 0.3663 data_time: 0.0019 memory: 20334 grad_norm: 0.0345 loss: 0.6859 +2023/05/31 19:54:14 - mmengine - INFO - Epoch(train) [2][4800/5758] lr: 9.9391e-04 eta: 13:18:27 time: 0.3887 data_time: 0.0015 memory: 20334 grad_norm: 0.0623 loss: 0.6874 +2023/05/31 19:54:56 - mmengine - INFO - Epoch(train) [2][4900/5758] lr: 9.9391e-04 eta: 13:17:01 time: 0.4403 data_time: 0.0027 memory: 20334 grad_norm: 0.0594 loss: 0.6838 +2023/05/31 19:55:36 - mmengine - INFO - Epoch(train) [2][5000/5758] lr: 9.9391e-04 eta: 13:15:20 time: 0.3890 data_time: 0.0016 memory: 20334 grad_norm: 0.0604 loss: 0.6927 +2023/05/31 19:56:18 - mmengine - INFO - Epoch(train) [2][5100/5758] lr: 9.9391e-04 eta: 13:13:55 time: 0.4493 data_time: 0.0017 memory: 20334 grad_norm: 0.0538 loss: 0.6867 +2023/05/31 19:56:59 - mmengine - INFO - Epoch(train) [2][5200/5758] lr: 9.9391e-04 eta: 13:12:23 time: 0.4407 data_time: 0.0017 memory: 20334 grad_norm: 0.0643 loss: 0.6885 +2023/05/31 19:57:16 - mmengine - INFO - Exp name: swin_base_8xb128_fake5m_20230531_183241 +2023/05/31 19:57:39 - mmengine - INFO - Epoch(train) [2][5300/5758] lr: 9.9391e-04 eta: 13:10:52 time: 0.3635 data_time: 0.0022 memory: 20334 grad_norm: 0.0312 loss: 0.6879 +2023/05/31 19:58:21 - mmengine - INFO - Epoch(train) [2][5400/5758] lr: 9.9391e-04 eta: 13:09:27 time: 0.4347 data_time: 0.0023 memory: 20334 grad_norm: 0.0441 loss: 0.6896 +2023/05/31 19:59:03 - mmengine - INFO - Epoch(train) [2][5500/5758] lr: 9.9391e-04 eta: 13:08:13 time: 0.4383 data_time: 0.0019 memory: 20334 grad_norm: 0.0362 loss: 0.6920 +2023/05/31 19:59:45 - mmengine - INFO - Epoch(train) [2][5600/5758] lr: 9.9391e-04 eta: 13:06:57 time: 0.3910 data_time: 0.0016 memory: 20334 grad_norm: 0.0308 loss: 0.6834 +2023/05/31 20:00:26 - mmengine - INFO - Epoch(train) [2][5700/5758] lr: 9.9391e-04 eta: 13:05:30 time: 0.3994 data_time: 0.0016 memory: 20334 grad_norm: 0.0362 loss: 0.6887 +2023/05/31 20:00:51 - mmengine - INFO - Exp name: swin_base_8xb128_fake5m_20230531_183241 +2023/05/31 20:00:51 - mmengine - INFO - Saving checkpoint at 2 epochs +2023/05/31 20:01:08 - mmengine - INFO - Epoch(val) [2][8/8] accuracy/top1: 100.0000 single-label/precision_classwise: [100.0, 0.0] single-label/recall_classwise: [100.0, 0.0] single-label/f1-score_classwise: [100.0, 0.0] data_time: 0.3736 time: 1.0013 +2023/05/31 20:01:54 - mmengine - INFO - Epoch(train) [3][ 100/5758] lr: 9.7577e-04 eta: 13:04:09 time: 0.4018 data_time: 0.0015 memory: 20334 grad_norm: 0.0310 loss: 0.6865 +2023/05/31 20:02:36 - mmengine - INFO - Epoch(train) [3][ 200/5758] lr: 9.7577e-04 eta: 13:02:53 time: 0.3977 data_time: 0.0015 memory: 20334 grad_norm: 0.0339 loss: 0.6903 +2023/05/31 20:03:18 - mmengine - INFO - Epoch(train) [3][ 300/5758] lr: 9.7577e-04 eta: 13:01:41 time: 0.3921 data_time: 0.0015 memory: 20334 grad_norm: 0.0301 loss: 0.6797 +2023/05/31 20:03:59 - mmengine - INFO - Epoch(train) [3][ 400/5758] lr: 9.7577e-04 eta: 13:00:18 time: 0.4322 data_time: 0.0016 memory: 20334 grad_norm: 0.0222 loss: 0.6907 +2023/05/31 20:04:34 - mmengine - INFO - Exp name: swin_base_8xb128_fake5m_20230531_183241 +2023/05/31 20:04:40 - mmengine - INFO - Epoch(train) [3][ 500/5758] lr: 9.7577e-04 eta: 12:58:54 time: 0.3707 data_time: 0.0014 memory: 20334 grad_norm: 0.0252 loss: 0.6878 +2023/05/31 20:05:21 - mmengine - INFO - Epoch(train) [3][ 600/5758] lr: 9.7577e-04 eta: 12:57:29 time: 0.4078 data_time: 0.0014 memory: 20334 grad_norm: 0.0266 loss: 0.6892 +2023/05/31 20:06:03 - mmengine - INFO - Epoch(train) [3][ 700/5758] lr: 9.7577e-04 eta: 12:56:16 time: 0.4164 data_time: 0.0014 memory: 20334 grad_norm: 0.0289 loss: 0.6871 +2023/05/31 20:06:46 - mmengine - INFO - Epoch(train) [3][ 800/5758] lr: 9.7577e-04 eta: 12:55:19 time: 0.4127 data_time: 0.0015 memory: 20334 grad_norm: 0.0276 loss: 0.6887 +2023/05/31 20:07:31 - mmengine - INFO - Epoch(train) [3][ 900/5758] lr: 9.7577e-04 eta: 12:54:27 time: 0.4864 data_time: 0.0014 memory: 20334 grad_norm: 0.0165 loss: 0.6819 +2023/05/31 20:08:17 - mmengine - INFO - Epoch(train) [3][1000/5758] lr: 9.7577e-04 eta: 12:53:46 time: 0.4091 data_time: 0.0016 memory: 20334 grad_norm: 0.0219 loss: 0.6840 +2023/05/31 20:09:03 - mmengine - INFO - Epoch(train) [3][1100/5758] lr: 9.7577e-04 eta: 12:53:09 time: 0.4353 data_time: 0.0014 memory: 20334 grad_norm: 0.0235 loss: 0.6887 +2023/05/31 20:09:47 - mmengine - INFO - Epoch(train) [3][1200/5758] lr: 9.7577e-04 eta: 12:52:14 time: 0.4666 data_time: 0.0014 memory: 20334 grad_norm: 0.0147 loss: 0.6851 +2023/05/31 20:10:34 - mmengine - INFO - Epoch(train) [3][1300/5758] lr: 9.7577e-04 eta: 12:51:40 time: 0.4219 data_time: 0.0014 memory: 20334 grad_norm: 0.0252 loss: 0.6859 +2023/05/31 20:11:17 - mmengine - INFO - Epoch(train) [3][1400/5758] lr: 9.7577e-04 eta: 12:50:37 time: 0.4625 data_time: 0.0015 memory: 20334 grad_norm: 0.0157 loss: 0.6885 +2023/05/31 20:11:54 - mmengine - INFO - Exp name: swin_base_8xb128_fake5m_20230531_183241 +2023/05/31 20:12:01 - mmengine - INFO - Epoch(train) [3][1500/5758] lr: 9.7577e-04 eta: 12:49:44 time: 0.4381 data_time: 0.0014 memory: 20334 grad_norm: 0.0259 loss: 0.6954 +2023/05/31 20:12:45 - mmengine - INFO - Epoch(train) [3][1600/5758] lr: 9.7577e-04 eta: 12:48:48 time: 0.4287 data_time: 0.0023 memory: 20334 grad_norm: 0.0237 loss: 0.6878 +2023/05/31 20:13:30 - mmengine - INFO - Epoch(train) [3][1700/5758] lr: 9.7577e-04 eta: 12:48:03 time: 0.4157 data_time: 0.0018 memory: 20334 grad_norm: 0.0196 loss: 0.6908 +2023/05/31 20:14:14 - mmengine - INFO - Epoch(train) [3][1800/5758] lr: 9.7577e-04 eta: 12:47:07 time: 0.4315 data_time: 0.0019 memory: 20334 grad_norm: 0.0223 loss: 0.6813 +2023/05/31 20:14:59 - mmengine - INFO - Epoch(train) [3][1900/5758] lr: 9.7577e-04 eta: 12:46:20 time: 0.4805 data_time: 0.0016 memory: 20334 grad_norm: 0.0273 loss: 0.6873 +2023/05/31 20:15:43 - mmengine - INFO - Epoch(train) [3][2000/5758] lr: 9.7577e-04 eta: 12:45:30 time: 0.4310 data_time: 0.0015 memory: 20334 grad_norm: 0.0265 loss: 0.6899 +2023/05/31 20:16:26 - mmengine - INFO - Epoch(train) [3][2100/5758] lr: 9.7577e-04 eta: 12:44:28 time: 0.4255 data_time: 0.0015 memory: 20334 grad_norm: 0.0124 loss: 0.6818 +2023/05/31 20:17:10 - mmengine - INFO - Epoch(train) [3][2200/5758] lr: 9.7577e-04 eta: 12:43:30 time: 0.4454 data_time: 0.0020 memory: 20334 grad_norm: 0.0120 loss: 0.6839 +2023/05/31 20:17:54 - mmengine - INFO - Epoch(train) [3][2300/5758] lr: 9.7577e-04 eta: 12:42:41 time: 0.4625 data_time: 0.0014 memory: 20334 grad_norm: 0.0236 loss: 0.6908 +2023/05/31 20:18:37 - mmengine - INFO - Epoch(train) [3][2400/5758] lr: 9.7577e-04 eta: 12:41:41 time: 0.4348 data_time: 0.0016 memory: 20334 grad_norm: 0.0181 loss: 0.6885 +2023/05/31 20:19:16 - mmengine - INFO - Exp name: swin_base_8xb128_fake5m_20230531_183241 +2023/05/31 20:19:22 - mmengine - INFO - Epoch(train) [3][2500/5758] lr: 9.7577e-04 eta: 12:40:54 time: 0.4458 data_time: 0.0022 memory: 20334 grad_norm: 0.0273 loss: 0.6876 +2023/05/31 20:20:06 - mmengine - INFO - Epoch(train) [3][2600/5758] lr: 9.7577e-04 eta: 12:40:01 time: 0.3965 data_time: 0.0015 memory: 20334 grad_norm: 0.0133 loss: 0.6822 +2023/05/31 20:20:50 - mmengine - INFO - Epoch(train) [3][2700/5758] lr: 9.7577e-04 eta: 12:39:03 time: 0.4620 data_time: 0.0013 memory: 20334 grad_norm: 0.0275 loss: 0.6798 +2023/05/31 20:21:36 - mmengine - INFO - Epoch(train) [3][2800/5758] lr: 9.7577e-04 eta: 12:38:26 time: 0.4705 data_time: 0.0019 memory: 20334 grad_norm: 0.0217 loss: 0.6842 +2023/05/31 20:22:20 - mmengine - INFO - Epoch(train) [3][2900/5758] lr: 9.7577e-04 eta: 12:37:34 time: 0.4299 data_time: 0.0017 memory: 20334 grad_norm: 0.0169 loss: 0.6871 +2023/05/31 20:23:03 - mmengine - INFO - Epoch(train) [3][3000/5758] lr: 9.7577e-04 eta: 12:36:34 time: 0.4040 data_time: 0.0015 memory: 20334 grad_norm: 0.0239 loss: 0.6868 +2023/05/31 20:23:46 - mmengine - INFO - Epoch(train) [3][3100/5758] lr: 9.7577e-04 eta: 12:35:34 time: 0.3936 data_time: 0.0017 memory: 20334 grad_norm: 0.0167 loss: 0.6868 +2023/05/31 20:24:31 - mmengine - INFO - Epoch(train) [3][3200/5758] lr: 9.7577e-04 eta: 12:34:47 time: 0.4512 data_time: 0.0015 memory: 20334 grad_norm: 0.0154 loss: 0.6861 +2023/05/31 20:25:15 - mmengine - INFO - Epoch(train) [3][3300/5758] lr: 9.7577e-04 eta: 12:33:56 time: 0.4681 data_time: 0.0016 memory: 20334 grad_norm: 0.0145 loss: 0.6834 +2023/05/31 20:26:01 - mmengine - INFO - Epoch(train) [3][3400/5758] lr: 9.7577e-04 eta: 12:33:19 time: 0.4666 data_time: 0.0016 memory: 20334 grad_norm: 0.0221 loss: 0.6830 +2023/05/31 20:26:38 - mmengine - INFO - Exp name: swin_base_8xb128_fake5m_20230531_183241 +2023/05/31 20:26:44 - mmengine - INFO - Epoch(train) [3][3500/5758] lr: 9.7577e-04 eta: 12:32:20 time: 0.4276 data_time: 0.0025 memory: 20334 grad_norm: 0.0118 loss: 0.6831 +2023/05/31 20:27:27 - mmengine - INFO - Epoch(train) [3][3600/5758] lr: 9.7577e-04 eta: 12:31:23 time: 0.3959 data_time: 0.0022 memory: 20334 grad_norm: 0.0161 loss: 0.6844 +2023/05/31 20:28:12 - mmengine - INFO - Epoch(train) [3][3700/5758] lr: 9.7577e-04 eta: 12:30:34 time: 0.3996 data_time: 0.0014 memory: 20334 grad_norm: 0.0217 loss: 0.6889 +2023/05/31 20:28:57 - mmengine - INFO - Epoch(train) [3][3800/5758] lr: 9.7577e-04 eta: 12:29:49 time: 0.4285 data_time: 0.0013 memory: 20334 grad_norm: 0.0112 loss: 0.6836 +2023/05/31 20:29:41 - mmengine - INFO - Epoch(train) [3][3900/5758] lr: 9.7577e-04 eta: 12:29:00 time: 0.4196 data_time: 0.0020 memory: 20334 grad_norm: 0.0170 loss: 0.6858 +2023/05/31 20:30:26 - mmengine - INFO - Epoch(train) [3][4000/5758] lr: 9.7577e-04 eta: 12:28:09 time: 0.4298 data_time: 0.0015 memory: 20334 grad_norm: 0.0314 loss: 0.6842 +2023/05/31 20:31:09 - mmengine - INFO - Epoch(train) [3][4100/5758] lr: 9.7577e-04 eta: 12:27:10 time: 0.4046 data_time: 0.0021 memory: 20334 grad_norm: 0.0170 loss: 0.6856 +2023/05/31 20:31:53 - mmengine - INFO - Epoch(train) [3][4200/5758] lr: 9.7577e-04 eta: 12:26:19 time: 0.4321 data_time: 0.0015 memory: 20334 grad_norm: 0.0161 loss: 0.6863 +2023/05/31 20:32:35 - mmengine - INFO - Epoch(train) [3][4300/5758] lr: 9.7577e-04 eta: 12:25:20 time: 0.4370 data_time: 0.0015 memory: 20334 grad_norm: 0.0134 loss: 0.6926 +2023/05/31 20:33:21 - mmengine - INFO - Epoch(train) [3][4400/5758] lr: 9.7577e-04 eta: 12:24:38 time: 0.4092 data_time: 0.0459 memory: 20334 grad_norm: 0.0179 loss: 0.6789 +2023/05/31 20:33:57 - mmengine - INFO - Exp name: swin_base_8xb128_fake5m_20230531_183241 +2023/05/31 20:34:05 - mmengine - INFO - Epoch(train) [3][4500/5758] lr: 9.7577e-04 eta: 12:23:46 time: 0.4261 data_time: 0.0647 memory: 20334 grad_norm: 0.0180 loss: 0.6856 +2023/05/31 20:34:49 - mmengine - INFO - Epoch(train) [3][4600/5758] lr: 9.7577e-04 eta: 12:22:55 time: 0.4407 data_time: 0.0775 memory: 20334 grad_norm: 0.0139 loss: 0.6850 +2023/05/31 20:35:33 - mmengine - INFO - Epoch(train) [3][4700/5758] lr: 9.7577e-04 eta: 12:22:03 time: 0.4256 data_time: 0.0435 memory: 20334 grad_norm: 0.0240 loss: 0.6887 +2023/05/31 20:36:16 - mmengine - INFO - Epoch(train) [3][4800/5758] lr: 9.7577e-04 eta: 12:21:10 time: 0.4609 data_time: 0.0015 memory: 20334 grad_norm: 0.0186 loss: 0.6877 +2023/05/31 20:37:01 - mmengine - INFO - Epoch(train) [3][4900/5758] lr: 9.7577e-04 eta: 12:20:20 time: 0.4639 data_time: 0.0017 memory: 20334 grad_norm: 0.0175 loss: 0.6846 +2023/05/31 20:37:44 - mmengine - INFO - Epoch(train) [3][5000/5758] lr: 9.7577e-04 eta: 12:19:28 time: 0.4122 data_time: 0.0012 memory: 20334 grad_norm: 0.0167 loss: 0.6921 +2023/05/31 20:38:27 - mmengine - INFO - Epoch(train) [3][5100/5758] lr: 9.7577e-04 eta: 12:18:29 time: 0.4096 data_time: 0.0013 memory: 20334 grad_norm: 0.0165 loss: 0.6891 +2023/05/31 20:39:11 - mmengine - INFO - Epoch(train) [3][5200/5758] lr: 9.7577e-04 eta: 12:17:38 time: 0.4865 data_time: 0.0020 memory: 20334 grad_norm: 0.0147 loss: 0.6927 +2023/05/31 20:39:53 - mmengine - INFO - Epoch(train) [3][5300/5758] lr: 9.7577e-04 eta: 12:16:38 time: 0.4180 data_time: 0.0017 memory: 20334 grad_norm: 0.0146 loss: 0.6871 +2023/05/31 20:40:37 - mmengine - INFO - Epoch(train) [3][5400/5758] lr: 9.7577e-04 eta: 12:15:43 time: 0.4236 data_time: 0.0023 memory: 20334 grad_norm: 0.0146 loss: 0.6873 +2023/05/31 20:41:15 - mmengine - INFO - Exp name: swin_base_8xb128_fake5m_20230531_183241 +2023/05/31 20:41:22 - mmengine - INFO - Epoch(train) [3][5500/5758] lr: 9.7577e-04 eta: 12:14:59 time: 0.4132 data_time: 0.0020 memory: 20334 grad_norm: 0.0168 loss: 0.6844 +2023/05/31 20:42:06 - mmengine - INFO - Epoch(train) [3][5600/5758] lr: 9.7577e-04 eta: 12:14:13 time: 0.4342 data_time: 0.0019 memory: 20334 grad_norm: 0.0145 loss: 0.6877 +2023/05/31 20:42:49 - mmengine - INFO - Epoch(train) [3][5700/5758] lr: 9.7577e-04 eta: 12:13:17 time: 0.4276 data_time: 0.0016 memory: 20334 grad_norm: 0.0230 loss: 0.6922 +2023/05/31 20:43:14 - mmengine - INFO - Exp name: swin_base_8xb128_fake5m_20230531_183241 +2023/05/31 20:43:14 - mmengine - INFO - Saving checkpoint at 3 epochs +2023/05/31 20:43:31 - mmengine - INFO - Epoch(val) [3][8/8] accuracy/top1: 100.0000 single-label/precision_classwise: [100.0, 0.0] single-label/recall_classwise: [100.0, 0.0] single-label/f1-score_classwise: [100.0, 0.0] data_time: 0.3651 time: 0.9904 +2023/05/31 20:44:19 - mmengine - INFO - Epoch(train) [4][ 100/5758] lr: 9.4605e-04 eta: 12:12:19 time: 0.3723 data_time: 0.0018 memory: 20334 grad_norm: 0.0164 loss: 0.6823 +2023/05/31 20:45:04 - mmengine - INFO - Epoch(train) [4][ 200/5758] lr: 9.4605e-04 eta: 12:11:34 time: 0.4839 data_time: 0.0016 memory: 20334 grad_norm: 0.0149 loss: 0.6836 +2023/05/31 20:45:46 - mmengine - INFO - Epoch(train) [4][ 300/5758] lr: 9.4605e-04 eta: 12:10:31 time: 0.4000 data_time: 0.0018 memory: 20334 grad_norm: 0.0210 loss: 0.6859 +2023/05/31 20:46:30 - mmengine - INFO - Epoch(train) [4][ 400/5758] lr: 9.4605e-04 eta: 12:09:42 time: 0.4330 data_time: 0.0017 memory: 20334 grad_norm: 0.0166 loss: 0.6917 +2023/05/31 20:47:13 - mmengine - INFO - Epoch(train) [4][ 500/5758] lr: 9.4605e-04 eta: 12:08:48 time: 0.4329 data_time: 0.0018 memory: 20334 grad_norm: 0.0118 loss: 0.6859 +2023/05/31 20:47:58 - mmengine - INFO - Epoch(train) [4][ 600/5758] lr: 9.4605e-04 eta: 12:08:00 time: 0.4122 data_time: 0.0017 memory: 20334 grad_norm: 0.0178 loss: 0.6884 +2023/05/31 20:48:41 - mmengine - INFO - Epoch(train) [4][ 700/5758] lr: 9.4605e-04 eta: 12:07:05 time: 0.4330 data_time: 0.0016 memory: 20334 grad_norm: 0.0183 loss: 0.6913 +2023/05/31 20:48:53 - mmengine - INFO - Exp name: swin_base_8xb128_fake5m_20230531_183241 +2023/05/31 20:49:24 - mmengine - INFO - Epoch(train) [4][ 800/5758] lr: 9.4605e-04 eta: 12:06:13 time: 0.4226 data_time: 0.0017 memory: 20334 grad_norm: 0.0258 loss: 0.6846 +2023/05/31 20:50:07 - mmengine - INFO - Epoch(train) [4][ 900/5758] lr: 9.4605e-04 eta: 12:05:18 time: 0.4490 data_time: 0.0016 memory: 20334 grad_norm: 0.0205 loss: 0.6850 +2023/05/31 20:50:51 - mmengine - INFO - Epoch(train) [4][1000/5758] lr: 9.4605e-04 eta: 12:04:25 time: 0.3926 data_time: 0.0018 memory: 20334 grad_norm: 0.0270 loss: 0.6846 +2023/05/31 20:51:35 - mmengine - INFO - Epoch(train) [4][1100/5758] lr: 9.4605e-04 eta: 12:03:36 time: 0.4475 data_time: 0.0016 memory: 20334 grad_norm: 0.0234 loss: 0.6881 +2023/05/31 20:52:18 - mmengine - INFO - Epoch(train) [4][1200/5758] lr: 9.4605e-04 eta: 12:02:40 time: 0.4378 data_time: 0.0019 memory: 20334 grad_norm: 0.0179 loss: 0.6856 +2023/05/31 20:53:01 - mmengine - INFO - Epoch(train) [4][1300/5758] lr: 9.4605e-04 eta: 12:01:46 time: 0.4295 data_time: 0.0015 memory: 20334 grad_norm: 0.0126 loss: 0.6849 +2023/05/31 20:53:45 - mmengine - INFO - Epoch(train) [4][1400/5758] lr: 9.4605e-04 eta: 12:00:57 time: 0.4935 data_time: 0.0018 memory: 20334 grad_norm: 0.0144 loss: 0.6878 +2023/05/31 20:54:27 - mmengine - INFO - Epoch(train) [4][1500/5758] lr: 9.4605e-04 eta: 12:00:00 time: 0.4450 data_time: 0.0019 memory: 20334 grad_norm: 0.0175 loss: 0.6796 +2023/05/31 20:55:11 - mmengine - INFO - Epoch(train) [4][1600/5758] lr: 9.4605e-04 eta: 11:59:09 time: 0.3642 data_time: 0.0019 memory: 20334 grad_norm: 0.0237 loss: 0.6832 +2023/05/31 20:55:54 - mmengine - INFO - Epoch(train) [4][1700/5758] lr: 9.4605e-04 eta: 11:58:16 time: 0.4355 data_time: 0.0015 memory: 20334 grad_norm: 0.0191 loss: 0.6884 +2023/05/31 20:56:05 - mmengine - INFO - Exp name: swin_base_8xb128_fake5m_20230531_183241 +2023/05/31 20:56:37 - mmengine - INFO - Epoch(train) [4][1800/5758] lr: 9.4605e-04 eta: 11:57:21 time: 0.3870 data_time: 0.0015 memory: 20334 grad_norm: 0.0135 loss: 0.6882 +2023/05/31 20:57:19 - mmengine - INFO - Epoch(train) [4][1900/5758] lr: 9.4605e-04 eta: 11:56:26 time: 0.3704 data_time: 0.0014 memory: 20334 grad_norm: 0.0217 loss: 0.6863 +2023/05/31 20:58:02 - mmengine - INFO - Epoch(train) [4][2000/5758] lr: 9.4605e-04 eta: 11:55:33 time: 0.4416 data_time: 0.0015 memory: 20334 grad_norm: 0.0081 loss: 0.6860 +2023/05/31 20:58:45 - mmengine - INFO - Epoch(train) [4][2100/5758] lr: 9.4605e-04 eta: 11:54:36 time: 0.4142 data_time: 0.0017 memory: 20334 grad_norm: 0.0168 loss: 0.6825 +2023/05/31 20:59:28 - mmengine - INFO - Epoch(train) [4][2200/5758] lr: 9.4605e-04 eta: 11:53:45 time: 0.4347 data_time: 0.0018 memory: 20334 grad_norm: 0.0182 loss: 0.6867 +2023/05/31 21:00:13 - mmengine - INFO - Epoch(train) [4][2300/5758] lr: 9.4605e-04 eta: 11:52:59 time: 0.4444 data_time: 0.0015 memory: 20334 grad_norm: 0.0137 loss: 0.6875 +2023/05/31 21:00:55 - mmengine - INFO - Epoch(train) [4][2400/5758] lr: 9.4605e-04 eta: 11:52:01 time: 0.4392 data_time: 0.0014 memory: 20334 grad_norm: 0.0134 loss: 0.6847 +2023/05/31 21:01:37 - mmengine - INFO - Epoch(train) [4][2500/5758] lr: 9.4605e-04 eta: 11:51:04 time: 0.4118 data_time: 0.0015 memory: 20334 grad_norm: 0.0210 loss: 0.6840 +2023/05/31 21:02:21 - mmengine - INFO - Epoch(train) [4][2600/5758] lr: 9.4605e-04 eta: 11:50:16 time: 0.4177 data_time: 0.0023 memory: 20334 grad_norm: 0.0131 loss: 0.6845 +2023/05/31 21:03:02 - mmengine - INFO - Epoch(train) [4][2700/5758] lr: 9.4605e-04 eta: 11:49:13 time: 0.4065 data_time: 0.0017 memory: 20334 grad_norm: 0.0162 loss: 0.6870 +2023/05/31 21:03:13 - mmengine - INFO - Exp name: swin_base_8xb128_fake5m_20230531_183241 +2023/05/31 21:03:45 - mmengine - INFO - Epoch(train) [4][2800/5758] lr: 9.4605e-04 eta: 11:48:19 time: 0.4555 data_time: 0.0017 memory: 20334 grad_norm: 0.0206 loss: 0.6928 +2023/05/31 21:04:28 - mmengine - INFO - Epoch(train) [4][2900/5758] lr: 9.4605e-04 eta: 11:47:27 time: 0.4271 data_time: 0.0016 memory: 20334 grad_norm: 0.0206 loss: 0.6828 +2023/05/31 21:05:11 - mmengine - INFO - Epoch(train) [4][3000/5758] lr: 9.4605e-04 eta: 11:46:35 time: 0.4085 data_time: 0.0021 memory: 20334 grad_norm: 0.0146 loss: 0.6862 +2023/05/31 21:05:54 - mmengine - INFO - Epoch(train) [4][3100/5758] lr: 9.4605e-04 eta: 11:45:41 time: 0.4297 data_time: 0.0016 memory: 20334 grad_norm: 0.0215 loss: 0.6816 +2023/05/31 21:06:36 - mmengine - INFO - Epoch(train) [4][3200/5758] lr: 9.4605e-04 eta: 11:44:44 time: 0.4052 data_time: 0.0016 memory: 20334 grad_norm: 0.0207 loss: 0.6868 +2023/05/31 21:07:20 - mmengine - INFO - Epoch(train) [4][3300/5758] lr: 9.4605e-04 eta: 11:44:00 time: 0.4284 data_time: 0.0016 memory: 20334 grad_norm: 0.0207 loss: 0.6845 +2023/05/31 21:08:05 - mmengine - INFO - Epoch(train) [4][3400/5758] lr: 9.4605e-04 eta: 11:43:13 time: 0.5038 data_time: 0.0015 memory: 20334 grad_norm: 0.0181 loss: 0.6863 +2023/05/31 21:08:48 - mmengine - INFO - Epoch(train) [4][3500/5758] lr: 9.4605e-04 eta: 11:42:20 time: 0.4160 data_time: 0.0016 memory: 20334 grad_norm: 0.0179 loss: 0.6884 +2023/05/31 21:09:29 - mmengine - INFO - Epoch(train) [4][3600/5758] lr: 9.4605e-04 eta: 11:41:22 time: 0.4228 data_time: 0.0023 memory: 20334 grad_norm: 0.0198 loss: 0.6881 +2023/05/31 21:10:11 - mmengine - INFO - Epoch(train) [4][3700/5758] lr: 9.4605e-04 eta: 11:40:26 time: 0.4068 data_time: 0.0012 memory: 20334 grad_norm: 0.0236 loss: 0.6876 +2023/05/31 21:10:23 - mmengine - INFO - Exp name: swin_base_8xb128_fake5m_20230531_183241 +2023/05/31 21:10:55 - mmengine - INFO - Epoch(train) [4][3800/5758] lr: 9.4605e-04 eta: 11:39:37 time: 0.4778 data_time: 0.0018 memory: 20334 grad_norm: 0.0132 loss: 0.6891 +2023/05/31 21:11:39 - mmengine - INFO - Epoch(train) [4][3900/5758] lr: 9.4605e-04 eta: 11:38:50 time: 0.4253 data_time: 0.0021 memory: 20334 grad_norm: 0.0147 loss: 0.6857 +2023/05/31 21:12:23 - mmengine - INFO - Epoch(train) [4][4000/5758] lr: 9.4605e-04 eta: 11:38:04 time: 0.4027 data_time: 0.0020 memory: 20334 grad_norm: 0.0128 loss: 0.6858 +2023/05/31 21:13:06 - mmengine - INFO - Epoch(train) [4][4100/5758] lr: 9.4605e-04 eta: 11:37:09 time: 0.4351 data_time: 0.0017 memory: 20334 grad_norm: 0.0162 loss: 0.6872 +2023/05/31 21:13:48 - mmengine - INFO - Epoch(train) [4][4200/5758] lr: 9.4605e-04 eta: 11:36:14 time: 0.3973 data_time: 0.0018 memory: 20334 grad_norm: 0.0114 loss: 0.6922 +2023/05/31 21:14:30 - mmengine - INFO - Epoch(train) [4][4300/5758] lr: 9.4605e-04 eta: 11:35:18 time: 0.4285 data_time: 0.0018 memory: 20334 grad_norm: 0.0213 loss: 0.6886 +2023/05/31 21:15:12 - mmengine - INFO - Epoch(train) [4][4400/5758] lr: 9.4605e-04 eta: 11:34:25 time: 0.4498 data_time: 0.0017 memory: 20334 grad_norm: 0.0167 loss: 0.6875 +2023/05/31 21:15:55 - mmengine - INFO - Epoch(train) [4][4500/5758] lr: 9.4605e-04 eta: 11:33:30 time: 0.4027 data_time: 0.0018 memory: 20334 grad_norm: 0.0458 loss: 0.6877 +2023/05/31 21:16:38 - mmengine - INFO - Epoch(train) [4][4600/5758] lr: 9.4605e-04 eta: 11:32:41 time: 0.4618 data_time: 0.0017 memory: 20334 grad_norm: 0.0162 loss: 0.6881 +2023/05/31 21:17:19 - mmengine - INFO - Epoch(train) [4][4700/5758] lr: 9.4605e-04 eta: 11:31:43 time: 0.4060 data_time: 0.0017 memory: 20334 grad_norm: 0.0261 loss: 0.6868 +2023/05/31 21:17:31 - mmengine - INFO - Exp name: swin_base_8xb128_fake5m_20230531_183241 +2023/05/31 21:18:03 - mmengine - INFO - Epoch(train) [4][4800/5758] lr: 9.4605e-04 eta: 11:30:52 time: 0.4904 data_time: 0.0020 memory: 20334 grad_norm: 0.0265 loss: 0.6937 +2023/05/31 21:18:44 - mmengine - INFO - Epoch(train) [4][4900/5758] lr: 9.4605e-04 eta: 11:29:53 time: 0.3855 data_time: 0.0020 memory: 20334 grad_norm: 0.0134 loss: 0.6906 +2023/05/31 21:19:27 - mmengine - INFO - Epoch(train) [4][5000/5758] lr: 9.4605e-04 eta: 11:29:05 time: 0.4084 data_time: 0.0027 memory: 20334 grad_norm: 0.0191 loss: 0.6885 +2023/05/31 21:20:11 - mmengine - INFO - Epoch(train) [4][5100/5758] lr: 9.4605e-04 eta: 11:28:15 time: 0.4244 data_time: 0.0017 memory: 20334 grad_norm: 0.0107 loss: 0.6882 +2023/05/31 21:20:54 - mmengine - INFO - Epoch(train) [4][5200/5758] lr: 9.4605e-04 eta: 11:27:28 time: 0.4063 data_time: 0.0021 memory: 20334 grad_norm: 0.0165 loss: 0.6871 +2023/05/31 21:21:38 - mmengine - INFO - Epoch(train) [4][5300/5758] lr: 9.4605e-04 eta: 11:26:38 time: 0.4249 data_time: 0.0021 memory: 20334 grad_norm: 0.0172 loss: 0.6871 +2023/05/31 21:22:20 - mmengine - INFO - Epoch(train) [4][5400/5758] lr: 9.4605e-04 eta: 11:25:46 time: 0.4152 data_time: 0.0017 memory: 20334 grad_norm: 0.0170 loss: 0.6895 +2023/05/31 21:23:03 - mmengine - INFO - Epoch(train) [4][5500/5758] lr: 9.4605e-04 eta: 11:24:54 time: 0.4291 data_time: 0.0018 memory: 20334 grad_norm: 0.0172 loss: 0.6827 +2023/05/31 21:23:45 - mmengine - INFO - Epoch(train) [4][5600/5758] lr: 9.4605e-04 eta: 11:24:02 time: 0.4585 data_time: 0.0016 memory: 20334 grad_norm: 0.0167 loss: 0.6921 +2023/05/31 21:24:29 - mmengine - INFO - Epoch(train) [4][5700/5758] lr: 9.4605e-04 eta: 11:23:15 time: 0.4407 data_time: 0.0019 memory: 20334 grad_norm: 0.0169 loss: 0.6857 +2023/05/31 21:24:40 - mmengine - INFO - Exp name: swin_base_8xb128_fake5m_20230531_183241 +2023/05/31 21:24:53 - mmengine - INFO - Exp name: swin_base_8xb128_fake5m_20230531_183241 +2023/05/31 21:24:53 - mmengine - INFO - Saving checkpoint at 4 epochs +2023/05/31 21:25:11 - mmengine - INFO - Epoch(val) [4][8/8] accuracy/top1: 100.0000 single-label/precision_classwise: [100.0, 0.0] single-label/recall_classwise: [100.0, 0.0] single-label/f1-score_classwise: [100.0, 0.0] data_time: 0.3748 time: 1.0005 +2023/05/31 21:26:00 - mmengine - INFO - Epoch(train) [5][ 100/5758] lr: 9.0546e-04 eta: 11:22:13 time: 0.4626 data_time: 0.0017 memory: 20334 grad_norm: 0.0135 loss: 0.6892 +2023/05/31 21:26:46 - mmengine - INFO - Epoch(train) [5][ 200/5758] lr: 9.0546e-04 eta: 11:21:33 time: 0.5576 data_time: 0.0018 memory: 20334 grad_norm: 0.0138 loss: 0.6833 +2023/05/31 21:27:29 - mmengine - INFO - Epoch(train) [5][ 300/5758] lr: 9.0546e-04 eta: 11:20:46 time: 0.4598 data_time: 0.0018 memory: 20334 grad_norm: 0.0153 loss: 0.6862 +2023/05/31 21:28:12 - mmengine - INFO - Epoch(train) [5][ 400/5758] lr: 9.0546e-04 eta: 11:19:56 time: 0.4185 data_time: 0.0015 memory: 20334 grad_norm: 0.0148 loss: 0.6823 +2023/05/31 21:28:56 - mmengine - INFO - Epoch(train) [5][ 500/5758] lr: 9.0546e-04 eta: 11:19:08 time: 0.4824 data_time: 0.0016 memory: 20334 grad_norm: 0.0160 loss: 0.6872 +2023/05/31 21:29:39 - mmengine - INFO - Epoch(train) [5][ 600/5758] lr: 9.0546e-04 eta: 11:18:19 time: 0.4359 data_time: 0.0017 memory: 20334 grad_norm: 0.0169 loss: 0.6870 +2023/05/31 21:30:24 - mmengine - INFO - Epoch(train) [5][ 700/5758] lr: 9.0546e-04 eta: 11:17:35 time: 0.4727 data_time: 0.0016 memory: 20334 grad_norm: 0.0149 loss: 0.6889 +2023/05/31 21:31:08 - mmengine - INFO - Epoch(train) [5][ 800/5758] lr: 9.0546e-04 eta: 11:16:48 time: 0.3930 data_time: 0.0017 memory: 20334 grad_norm: 0.0179 loss: 0.6852 +2023/05/31 21:31:51 - mmengine - INFO - Epoch(train) [5][ 900/5758] lr: 9.0546e-04 eta: 11:15:59 time: 0.4594 data_time: 0.0018 memory: 20334 grad_norm: 0.0157 loss: 0.6877 +2023/05/31 21:32:21 - mmengine - INFO - Exp name: swin_base_8xb128_fake5m_20230531_183241 +2023/05/31 21:32:35 - mmengine - INFO - Epoch(train) [5][1000/5758] lr: 9.0546e-04 eta: 11:15:14 time: 0.4944 data_time: 0.0013 memory: 20334 grad_norm: 0.0222 loss: 0.6885 +2023/05/31 21:33:18 - mmengine - INFO - Epoch(train) [5][1100/5758] lr: 9.0546e-04 eta: 11:14:22 time: 0.4181 data_time: 0.0017 memory: 20334 grad_norm: 0.0180 loss: 0.6851 +2023/05/31 21:34:03 - mmengine - INFO - Epoch(train) [5][1200/5758] lr: 9.0546e-04 eta: 11:13:41 time: 0.4728 data_time: 0.0016 memory: 20334 grad_norm: 0.0201 loss: 0.6893 +2023/05/31 21:34:47 - mmengine - INFO - Epoch(train) [5][1300/5758] lr: 9.0546e-04 eta: 11:12:54 time: 0.4514 data_time: 0.0024 memory: 20334 grad_norm: 0.0127 loss: 0.6831 +2023/05/31 21:35:31 - mmengine - INFO - Epoch(train) [5][1400/5758] lr: 9.0546e-04 eta: 11:12:08 time: 0.4225 data_time: 0.0016 memory: 20334 grad_norm: 0.0154 loss: 0.6874 +2023/05/31 21:36:14 - mmengine - INFO - Epoch(train) [5][1500/5758] lr: 9.0546e-04 eta: 11:11:16 time: 0.4359 data_time: 0.0013 memory: 20334 grad_norm: 0.0270 loss: 0.6900 +2023/05/31 21:36:58 - mmengine - INFO - Epoch(train) [5][1600/5758] lr: 9.0546e-04 eta: 11:10:34 time: 0.4071 data_time: 0.0019 memory: 20334 grad_norm: 0.0152 loss: 0.6914 +2023/05/31 21:37:39 - mmengine - INFO - Epoch(train) [5][1700/5758] lr: 9.0546e-04 eta: 11:09:34 time: 0.4208 data_time: 0.0020 memory: 20334 grad_norm: 0.0142 loss: 0.6832 +2023/05/31 21:38:21 - mmengine - INFO - Epoch(train) [5][1800/5758] lr: 9.0546e-04 eta: 11:08:40 time: 0.4055 data_time: 0.0018 memory: 20334 grad_norm: 0.0170 loss: 0.6816 +2023/05/31 21:39:04 - mmengine - INFO - Epoch(train) [5][1900/5758] lr: 9.0546e-04 eta: 11:07:54 time: 0.4407 data_time: 0.0018 memory: 20334 grad_norm: 0.0153 loss: 0.6831 +2023/05/31 21:39:33 - mmengine - INFO - Exp name: swin_base_8xb128_fake5m_20230531_183241 +2023/05/31 21:39:47 - mmengine - INFO - Epoch(train) [5][2000/5758] lr: 9.0546e-04 eta: 11:07:02 time: 0.4170 data_time: 0.0020 memory: 20334 grad_norm: 0.0161 loss: 0.6893 +2023/05/31 21:40:29 - mmengine - INFO - Epoch(train) [5][2100/5758] lr: 9.0546e-04 eta: 11:06:11 time: 0.4147 data_time: 0.0024 memory: 20334 grad_norm: 0.0106 loss: 0.6865 +2023/05/31 21:41:11 - mmengine - INFO - Epoch(train) [5][2200/5758] lr: 9.0546e-04 eta: 11:05:18 time: 0.3871 data_time: 0.0017 memory: 20334 grad_norm: 0.0140 loss: 0.6822 +2023/05/31 21:41:56 - mmengine - INFO - Epoch(train) [5][2300/5758] lr: 9.0546e-04 eta: 11:04:34 time: 0.4810 data_time: 0.0025 memory: 20334 grad_norm: 0.0115 loss: 0.6844 +2023/05/31 21:42:40 - mmengine - INFO - Epoch(train) [5][2400/5758] lr: 9.0546e-04 eta: 11:03:47 time: 0.4656 data_time: 0.0023 memory: 20334 grad_norm: 0.0259 loss: 0.6852 +2023/05/31 21:43:24 - mmengine - INFO - Epoch(train) [5][2500/5758] lr: 9.0546e-04 eta: 11:03:02 time: 0.4012 data_time: 0.0022 memory: 20334 grad_norm: 0.0133 loss: 0.6867 +2023/05/31 21:44:07 - mmengine - INFO - Epoch(train) [5][2600/5758] lr: 9.0546e-04 eta: 11:02:13 time: 0.3986 data_time: 0.0025 memory: 20334 grad_norm: 0.0169 loss: 0.6808 +2023/05/31 21:44:50 - mmengine - INFO - Epoch(train) [5][2700/5758] lr: 9.0546e-04 eta: 11:01:25 time: 0.4566 data_time: 0.0017 memory: 20334 grad_norm: 0.0106 loss: 0.6883 +2023/05/31 21:45:32 - mmengine - INFO - Epoch(train) [5][2800/5758] lr: 9.0546e-04 eta: 11:00:30 time: 0.3927 data_time: 0.0016 memory: 20334 grad_norm: 0.0138 loss: 0.6906 +2023/05/31 21:46:13 - mmengine - INFO - Epoch(train) [5][2900/5758] lr: 9.0546e-04 eta: 10:59:36 time: 0.4240 data_time: 0.0020 memory: 20334 grad_norm: 0.0143 loss: 0.6879 +2023/05/31 21:46:43 - mmengine - INFO - Exp name: swin_base_8xb128_fake5m_20230531_183241 +2023/05/31 21:46:56 - mmengine - INFO - Epoch(train) [5][3000/5758] lr: 9.0546e-04 eta: 10:58:48 time: 0.4904 data_time: 0.0023 memory: 20334 grad_norm: 0.0263 loss: 0.6910 +2023/05/31 21:47:38 - mmengine - INFO - Epoch(train) [5][3100/5758] lr: 9.0546e-04 eta: 10:57:56 time: 0.4185 data_time: 0.0018 memory: 20334 grad_norm: 0.0194 loss: 0.6845 +2023/05/31 21:48:21 - mmengine - INFO - Epoch(train) [5][3200/5758] lr: 9.0546e-04 eta: 10:57:05 time: 0.4449 data_time: 0.0017 memory: 20334 grad_norm: 0.0220 loss: 0.6875 +2023/05/31 21:49:05 - mmengine - INFO - Epoch(train) [5][3300/5758] lr: 9.0546e-04 eta: 10:56:20 time: 0.3759 data_time: 0.0014 memory: 20334 grad_norm: 0.0263 loss: 0.6842 +2023/05/31 21:49:49 - mmengine - INFO - Epoch(train) [5][3400/5758] lr: 9.0546e-04 eta: 10:55:33 time: 0.5078 data_time: 0.0015 memory: 20334 grad_norm: 0.0196 loss: 0.6860 +2023/05/31 21:50:32 - mmengine - INFO - Epoch(train) [5][3500/5758] lr: 9.0546e-04 eta: 10:54:44 time: 0.5131 data_time: 0.0022 memory: 20334 grad_norm: 0.0172 loss: 0.6866 +2023/05/31 21:51:15 - mmengine - INFO - Epoch(train) [5][3600/5758] lr: 9.0546e-04 eta: 10:53:59 time: 0.5038 data_time: 0.0016 memory: 20334 grad_norm: 0.0099 loss: 0.6881 +2023/05/31 21:51:59 - mmengine - INFO - Epoch(train) [5][3700/5758] lr: 9.0546e-04 eta: 10:53:12 time: 0.4277 data_time: 0.0022 memory: 20334 grad_norm: 0.0164 loss: 0.6874 +2023/05/31 21:52:41 - mmengine - INFO - Epoch(train) [5][3800/5758] lr: 9.0546e-04 eta: 10:52:20 time: 0.4236 data_time: 0.0025 memory: 20334 grad_norm: 0.0197 loss: 0.6891 +2023/05/31 21:53:26 - mmengine - INFO - Epoch(train) [5][3900/5758] lr: 9.0546e-04 eta: 10:51:37 time: 0.4664 data_time: 0.0020 memory: 20334 grad_norm: 0.0167 loss: 0.6835 +2023/05/31 21:53:59 - mmengine - INFO - Exp name: swin_base_8xb128_fake5m_20230531_183241 +2023/05/31 21:54:12 - mmengine - INFO - Epoch(train) [5][4000/5758] lr: 9.0546e-04 eta: 10:50:58 time: 0.3931 data_time: 0.0024 memory: 20334 grad_norm: 0.0188 loss: 0.6852 +2023/05/31 21:54:53 - mmengine - INFO - Epoch(train) [5][4100/5758] lr: 9.0546e-04 eta: 10:50:04 time: 0.3859 data_time: 0.0015 memory: 20334 grad_norm: 0.0138 loss: 0.6881 +2023/05/31 21:55:37 - mmengine - INFO - Epoch(train) [5][4200/5758] lr: 9.0546e-04 eta: 10:49:17 time: 0.4127 data_time: 0.0015 memory: 20334 grad_norm: 0.0129 loss: 0.6901 +2023/05/31 21:56:22 - mmengine - INFO - Epoch(train) [5][4300/5758] lr: 9.0546e-04 eta: 10:48:37 time: 0.4177 data_time: 0.0017 memory: 20334 grad_norm: 0.0163 loss: 0.6870 +2023/05/31 21:57:05 - mmengine - INFO - Epoch(train) [5][4400/5758] lr: 9.0546e-04 eta: 10:47:47 time: 0.4095 data_time: 0.0016 memory: 20334 grad_norm: 0.0224 loss: 0.6888 +2023/05/31 21:57:47 - mmengine - INFO - Epoch(train) [5][4500/5758] lr: 9.0546e-04 eta: 10:46:57 time: 0.4457 data_time: 0.0017 memory: 20334 grad_norm: 0.0121 loss: 0.6857 +2023/05/31 21:58:33 - mmengine - INFO - Epoch(train) [5][4600/5758] lr: 9.0546e-04 eta: 10:46:16 time: 0.5404 data_time: 0.0016 memory: 20334 grad_norm: 0.0148 loss: 0.6922 +2023/05/31 21:59:16 - mmengine - INFO - Epoch(train) [5][4700/5758] lr: 9.0546e-04 eta: 10:45:27 time: 0.4003 data_time: 0.0026 memory: 20334 grad_norm: 0.0248 loss: 0.6810 +2023/05/31 22:00:00 - mmengine - INFO - Epoch(train) [5][4800/5758] lr: 9.0546e-04 eta: 10:44:43 time: 0.4443 data_time: 0.0026 memory: 20334 grad_norm: 0.0229 loss: 0.6823 +2023/05/31 22:00:44 - mmengine - INFO - Epoch(train) [5][4900/5758] lr: 9.0546e-04 eta: 10:43:57 time: 0.4572 data_time: 0.0023 memory: 20334 grad_norm: 0.0174 loss: 0.6865 +2023/05/31 22:01:14 - mmengine - INFO - Exp name: swin_base_8xb128_fake5m_20230531_183241 +2023/05/31 22:01:29 - mmengine - INFO - Epoch(train) [5][5000/5758] lr: 9.0546e-04 eta: 10:43:15 time: 0.4226 data_time: 0.0025 memory: 20334 grad_norm: 0.0185 loss: 0.6900 +2023/05/31 22:02:13 - mmengine - INFO - Epoch(train) [5][5100/5758] lr: 9.0546e-04 eta: 10:42:31 time: 0.4421 data_time: 0.0024 memory: 20334 grad_norm: 0.0099 loss: 0.6917 +2023/05/31 22:02:59 - mmengine - INFO - Epoch(train) [5][5200/5758] lr: 9.0546e-04 eta: 10:41:52 time: 0.4703 data_time: 0.0024 memory: 20334 grad_norm: 0.0149 loss: 0.6890 +2023/05/31 22:03:43 - mmengine - INFO - Epoch(train) [5][5300/5758] lr: 9.0546e-04 eta: 10:41:06 time: 0.4200 data_time: 0.0026 memory: 20334 grad_norm: 0.0187 loss: 0.6927 +2023/05/31 22:04:27 - mmengine - INFO - Epoch(train) [5][5400/5758] lr: 9.0546e-04 eta: 10:40:21 time: 0.4258 data_time: 0.0025 memory: 20334 grad_norm: 0.0158 loss: 0.6880 +2023/05/31 22:05:11 - mmengine - INFO - Epoch(train) [5][5500/5758] lr: 9.0546e-04 eta: 10:39:37 time: 0.4851 data_time: 0.0023 memory: 20334 grad_norm: 0.0193 loss: 0.6876 +2023/05/31 22:05:56 - mmengine - INFO - Epoch(train) [5][5600/5758] lr: 9.0546e-04 eta: 10:38:54 time: 0.4599 data_time: 0.0018 memory: 20334 grad_norm: 0.0201 loss: 0.6833 +2023/05/31 22:06:41 - mmengine - INFO - Epoch(train) [5][5700/5758] lr: 9.0546e-04 eta: 10:38:10 time: 0.4305 data_time: 0.0022 memory: 20334 grad_norm: 0.0176 loss: 0.6856 +2023/05/31 22:07:05 - mmengine - INFO - Exp name: swin_base_8xb128_fake5m_20230531_183241 +2023/05/31 22:07:05 - mmengine - INFO - Saving checkpoint at 5 epochs +2023/05/31 22:07:22 - mmengine - INFO - Epoch(val) [5][8/8] accuracy/top1: 100.0000 single-label/precision_classwise: [100.0, 0.0] single-label/recall_classwise: [100.0, 0.0] single-label/f1-score_classwise: [100.0, 0.0] data_time: 0.3672 time: 0.9941 +2023/05/31 22:08:08 - mmengine - INFO - Epoch(train) [6][ 100/5758] lr: 8.5502e-04 eta: 10:37:02 time: 0.4215 data_time: 0.0017 memory: 20334 grad_norm: 0.0107 loss: 0.6819 +2023/05/31 22:08:50 - mmengine - INFO - Epoch(train) [6][ 200/5758] lr: 8.5502e-04 eta: 10:36:11 time: 0.4055 data_time: 0.0017 memory: 20334 grad_norm: 0.0142 loss: 0.6858 +2023/05/31 22:08:55 - mmengine - INFO - Exp name: swin_base_8xb128_fake5m_20230531_183241 +2023/05/31 22:09:31 - mmengine - INFO - Epoch(train) [6][ 300/5758] lr: 8.5502e-04 eta: 10:35:16 time: 0.4040 data_time: 0.0022 memory: 20334 grad_norm: 0.0184 loss: 0.6853 +2023/05/31 22:10:10 - mmengine - INFO - Epoch(train) [6][ 400/5758] lr: 8.5502e-04 eta: 10:34:17 time: 0.3814 data_time: 0.0017 memory: 20334 grad_norm: 0.0178 loss: 0.6879 +2023/05/31 22:10:52 - mmengine - INFO - Epoch(train) [6][ 500/5758] lr: 8.5502e-04 eta: 10:33:25 time: 0.4531 data_time: 0.0023 memory: 20334 grad_norm: 0.0173 loss: 0.6870 +2023/05/31 22:11:31 - mmengine - INFO - Epoch(train) [6][ 600/5758] lr: 8.5502e-04 eta: 10:32:27 time: 0.3860 data_time: 0.0020 memory: 20334 grad_norm: 0.0177 loss: 0.6863 +2023/05/31 22:12:13 - mmengine - INFO - Epoch(train) [6][ 700/5758] lr: 8.5502e-04 eta: 10:31:36 time: 0.4217 data_time: 0.0020 memory: 20334 grad_norm: 0.0179 loss: 0.6891 +2023/05/31 22:12:53 - mmengine - INFO - Epoch(train) [6][ 800/5758] lr: 8.5502e-04 eta: 10:30:40 time: 0.3811 data_time: 0.0017 memory: 20334 grad_norm: 0.0186 loss: 0.6859 +2023/05/31 22:13:35 - mmengine - INFO - Epoch(train) [6][ 900/5758] lr: 8.5502e-04 eta: 10:29:47 time: 0.4309 data_time: 0.0031 memory: 20334 grad_norm: 0.0150 loss: 0.6860 +2023/05/31 22:14:16 - mmengine - INFO - Epoch(train) [6][1000/5758] lr: 8.5502e-04 eta: 10:28:54 time: 0.4587 data_time: 0.0024 memory: 20334 grad_norm: 0.0151 loss: 0.6870 +2023/05/31 22:14:58 - mmengine - INFO - Epoch(train) [6][1100/5758] lr: 8.5502e-04 eta: 10:28:04 time: 0.4721 data_time: 0.0026 memory: 20334 grad_norm: 0.0219 loss: 0.6868 +2023/05/31 22:15:39 - mmengine - INFO - Epoch(train) [6][1200/5758] lr: 8.5502e-04 eta: 10:27:10 time: 0.4290 data_time: 0.0030 memory: 20334 grad_norm: 0.0190 loss: 0.6873 +2023/05/31 22:15:43 - mmengine - INFO - Exp name: swin_base_8xb128_fake5m_20230531_183241 +2023/05/31 22:16:19 - mmengine - INFO - Epoch(train) [6][1300/5758] lr: 8.5502e-04 eta: 10:26:16 time: 0.4148 data_time: 0.0020 memory: 20334 grad_norm: 0.0140 loss: 0.6891 +2023/05/31 22:17:00 - mmengine - INFO - Epoch(train) [6][1400/5758] lr: 8.5502e-04 eta: 10:25:22 time: 0.3663 data_time: 0.0024 memory: 20334 grad_norm: 0.0121 loss: 0.6891 +2023/05/31 22:17:41 - mmengine - INFO - Epoch(train) [6][1500/5758] lr: 8.5502e-04 eta: 10:24:29 time: 0.4630 data_time: 0.0023 memory: 20334 grad_norm: 0.0190 loss: 0.6904 +2023/05/31 22:18:23 - mmengine - INFO - Epoch(train) [6][1600/5758] lr: 8.5502e-04 eta: 10:23:39 time: 0.4389 data_time: 0.0030 memory: 20334 grad_norm: 0.0143 loss: 0.6874 +2023/05/31 22:19:03 - mmengine - INFO - Epoch(train) [6][1700/5758] lr: 8.5502e-04 eta: 10:22:44 time: 0.3718 data_time: 0.0019 memory: 20334 grad_norm: 0.0181 loss: 0.6867 +2023/05/31 22:19:45 - mmengine - INFO - Epoch(train) [6][1800/5758] lr: 8.5502e-04 eta: 10:21:53 time: 0.3911 data_time: 0.0017 memory: 20334 grad_norm: 0.0164 loss: 0.6881 +2023/05/31 22:20:26 - mmengine - INFO - Epoch(train) [6][1900/5758] lr: 8.5502e-04 eta: 10:21:01 time: 0.4190 data_time: 0.0026 memory: 20334 grad_norm: 0.0180 loss: 0.6916 +2023/05/31 22:21:06 - mmengine - INFO - Epoch(train) [6][2000/5758] lr: 8.5502e-04 eta: 10:20:05 time: 0.3862 data_time: 0.0022 memory: 20334 grad_norm: 0.0198 loss: 0.6882 +2023/05/31 22:21:47 - mmengine - INFO - Epoch(train) [6][2100/5758] lr: 8.5502e-04 eta: 10:19:12 time: 0.4260 data_time: 0.0032 memory: 20334 grad_norm: 0.0146 loss: 0.6861 +2023/05/31 22:22:29 - mmengine - INFO - Epoch(train) [6][2200/5758] lr: 8.5502e-04 eta: 10:18:23 time: 0.4317 data_time: 0.0025 memory: 20334 grad_norm: 0.0162 loss: 0.6831 +2023/05/31 22:22:33 - mmengine - INFO - Exp name: swin_base_8xb128_fake5m_20230531_183241 +2023/05/31 22:23:08 - mmengine - INFO - Epoch(train) [6][2300/5758] lr: 8.5502e-04 eta: 10:17:26 time: 0.3788 data_time: 0.0024 memory: 20334 grad_norm: 0.0135 loss: 0.6834 +2023/05/31 22:23:50 - mmengine - INFO - Epoch(train) [6][2400/5758] lr: 8.5502e-04 eta: 10:16:34 time: 0.4986 data_time: 0.0027 memory: 20334 grad_norm: 0.0187 loss: 0.6832 +2023/05/31 22:24:31 - mmengine - INFO - Epoch(train) [6][2500/5758] lr: 8.5502e-04 eta: 10:15:43 time: 0.4379 data_time: 0.0024 memory: 20334 grad_norm: 0.0160 loss: 0.6868 +2023/05/31 22:25:12 - mmengine - INFO - Epoch(train) [6][2600/5758] lr: 8.5502e-04 eta: 10:14:51 time: 0.4220 data_time: 0.0025 memory: 20334 grad_norm: 0.0152 loss: 0.6876 +2023/05/31 22:25:53 - mmengine - INFO - Epoch(train) [6][2700/5758] lr: 8.5502e-04 eta: 10:14:00 time: 0.3924 data_time: 0.0022 memory: 20334 grad_norm: 0.0156 loss: 0.6884 +2023/05/31 22:26:33 - mmengine - INFO - Epoch(train) [6][2800/5758] lr: 8.5502e-04 eta: 10:13:04 time: 0.3723 data_time: 0.0024 memory: 20334 grad_norm: 0.0140 loss: 0.6841 +2023/05/31 22:27:16 - mmengine - INFO - Epoch(train) [6][2900/5758] lr: 8.5502e-04 eta: 10:12:16 time: 0.4717 data_time: 0.0028 memory: 20334 grad_norm: 0.0248 loss: 0.6864 +2023/05/31 22:27:56 - mmengine - INFO - Epoch(train) [6][3000/5758] lr: 8.5502e-04 eta: 10:11:23 time: 0.4016 data_time: 0.0021 memory: 20334 grad_norm: 0.0141 loss: 0.6857 +2023/05/31 22:28:38 - mmengine - INFO - Epoch(train) [6][3100/5758] lr: 8.5502e-04 eta: 10:10:33 time: 0.4071 data_time: 0.0019 memory: 20334 grad_norm: 0.0214 loss: 0.6921 +2023/05/31 22:29:19 - mmengine - INFO - Epoch(train) [6][3200/5758] lr: 8.5502e-04 eta: 10:09:41 time: 0.4218 data_time: 0.0025 memory: 20334 grad_norm: 0.0125 loss: 0.6927 +2023/05/31 22:29:23 - mmengine - INFO - Exp name: swin_base_8xb128_fake5m_20230531_183241 +2023/05/31 22:30:00 - mmengine - INFO - Epoch(train) [6][3300/5758] lr: 8.5502e-04 eta: 10:08:51 time: 0.3823 data_time: 0.0025 memory: 20334 grad_norm: 0.0199 loss: 0.6829 +2023/05/31 22:30:41 - mmengine - INFO - Epoch(train) [6][3400/5758] lr: 8.5502e-04 eta: 10:07:57 time: 0.4505 data_time: 0.0018 memory: 20334 grad_norm: 0.0151 loss: 0.6896 +2023/05/31 22:31:21 - mmengine - INFO - Epoch(train) [6][3500/5758] lr: 8.5502e-04 eta: 10:07:05 time: 0.4410 data_time: 0.0022 memory: 20334 grad_norm: 0.0186 loss: 0.6846 +2023/05/31 22:32:02 - mmengine - INFO - Epoch(train) [6][3600/5758] lr: 8.5502e-04 eta: 10:06:13 time: 0.3710 data_time: 0.0017 memory: 20334 grad_norm: 0.0135 loss: 0.6896 +2023/05/31 22:32:43 - mmengine - INFO - Epoch(train) [6][3700/5758] lr: 8.5502e-04 eta: 10:05:21 time: 0.4106 data_time: 0.0021 memory: 20334 grad_norm: 0.0225 loss: 0.6931 +2023/05/31 22:33:23 - mmengine - INFO - Epoch(train) [6][3800/5758] lr: 8.5502e-04 eta: 10:04:27 time: 0.4088 data_time: 0.0025 memory: 20334 grad_norm: 0.0137 loss: 0.6867 +2023/05/31 22:34:04 - mmengine - INFO - Epoch(train) [6][3900/5758] lr: 8.5502e-04 eta: 10:03:35 time: 0.4119 data_time: 0.0017 memory: 20334 grad_norm: 0.0203 loss: 0.6895 +2023/05/31 22:34:44 - mmengine - INFO - Epoch(train) [6][4000/5758] lr: 8.5502e-04 eta: 10:02:41 time: 0.3971 data_time: 0.0028 memory: 20334 grad_norm: 0.0173 loss: 0.6862 +2023/05/31 22:35:25 - mmengine - INFO - Epoch(train) [6][4100/5758] lr: 8.5502e-04 eta: 10:01:51 time: 0.3677 data_time: 0.0017 memory: 20334 grad_norm: 0.0246 loss: 0.6876 +2023/05/31 22:36:06 - mmengine - INFO - Epoch(train) [6][4200/5758] lr: 8.5502e-04 eta: 10:01:00 time: 0.4329 data_time: 0.0016 memory: 20334 grad_norm: 0.0176 loss: 0.6919 +2023/05/31 22:36:10 - mmengine - INFO - Exp name: swin_base_8xb128_fake5m_20230531_183241 +2023/05/31 22:36:47 - mmengine - INFO - Epoch(train) [6][4300/5758] lr: 8.5502e-04 eta: 10:00:09 time: 0.4061 data_time: 0.0029 memory: 20334 grad_norm: 0.0139 loss: 0.6934 +2023/05/31 22:37:27 - mmengine - INFO - Epoch(train) [6][4400/5758] lr: 8.5502e-04 eta: 9:59:16 time: 0.4042 data_time: 0.0019 memory: 20334 grad_norm: 0.0120 loss: 0.6842 +2023/05/31 22:38:08 - mmengine - INFO - Epoch(train) [6][4500/5758] lr: 8.5502e-04 eta: 9:58:25 time: 0.3866 data_time: 0.0019 memory: 20334 grad_norm: 0.0176 loss: 0.6855 +2023/05/31 22:38:50 - mmengine - INFO - Epoch(train) [6][4600/5758] lr: 8.5502e-04 eta: 9:57:36 time: 0.3988 data_time: 0.0020 memory: 20334 grad_norm: 0.0179 loss: 0.6893 +2023/05/31 22:39:30 - mmengine - INFO - Epoch(train) [6][4700/5758] lr: 8.5502e-04 eta: 9:56:43 time: 0.4070 data_time: 0.0023 memory: 20334 grad_norm: 0.0114 loss: 0.6911 +2023/05/31 22:40:10 - mmengine - INFO - Epoch(train) [6][4800/5758] lr: 8.5502e-04 eta: 9:55:50 time: 0.3735 data_time: 0.0022 memory: 20334 grad_norm: 0.0144 loss: 0.6865 +2023/05/31 22:40:50 - mmengine - INFO - Epoch(train) [6][4900/5758] lr: 8.5502e-04 eta: 9:54:57 time: 0.3939 data_time: 0.0024 memory: 20334 grad_norm: 0.0183 loss: 0.6883 +2023/05/31 22:41:30 - mmengine - INFO - Epoch(train) [6][5000/5758] lr: 8.5502e-04 eta: 9:54:04 time: 0.4602 data_time: 0.0022 memory: 20334 grad_norm: 0.0245 loss: 0.6903 +2023/05/31 22:42:11 - mmengine - INFO - Epoch(train) [6][5100/5758] lr: 8.5502e-04 eta: 9:53:13 time: 0.4275 data_time: 0.0022 memory: 20334 grad_norm: 0.0258 loss: 0.6845 +2023/05/31 22:42:52 - mmengine - INFO - Epoch(train) [6][5200/5758] lr: 8.5502e-04 eta: 9:52:21 time: 0.4513 data_time: 0.0023 memory: 20334 grad_norm: 0.0158 loss: 0.6855 +2023/05/31 22:42:56 - mmengine - INFO - Exp name: swin_base_8xb128_fake5m_20230531_183241 +2023/05/31 22:43:34 - mmengine - INFO - Epoch(train) [6][5300/5758] lr: 8.5502e-04 eta: 9:51:34 time: 0.4971 data_time: 0.0019 memory: 20334 grad_norm: 0.0185 loss: 0.6863 +2023/05/31 22:44:14 - mmengine - INFO - Epoch(train) [6][5400/5758] lr: 8.5502e-04 eta: 9:50:42 time: 0.4245 data_time: 0.0024 memory: 20334 grad_norm: 0.0111 loss: 0.6906 +2023/05/31 22:44:55 - mmengine - INFO - Epoch(train) [6][5500/5758] lr: 8.5502e-04 eta: 9:49:51 time: 0.4005 data_time: 0.0028 memory: 20334 grad_norm: 0.0163 loss: 0.6868 +2023/05/31 22:45:42 - mmengine - INFO - Epoch(train) [6][5600/5758] lr: 8.5502e-04 eta: 9:49:15 time: 0.5972 data_time: 0.0018 memory: 20334 grad_norm: 0.0120 loss: 0.6849 +2023/05/31 22:46:22 - mmengine - INFO - Epoch(train) [6][5700/5758] lr: 8.5502e-04 eta: 9:48:22 time: 0.3656 data_time: 0.0018 memory: 20334 grad_norm: 0.0100 loss: 0.6855 +2023/05/31 22:46:44 - mmengine - INFO - Exp name: swin_base_8xb128_fake5m_20230531_183241 +2023/05/31 22:46:44 - mmengine - INFO - Saving checkpoint at 6 epochs +2023/05/31 22:47:03 - mmengine - INFO - Epoch(val) [6][8/8] accuracy/top1: 100.0000 single-label/precision_classwise: [100.0, 0.0] single-label/recall_classwise: [100.0, 0.0] single-label/f1-score_classwise: [100.0, 0.0] data_time: 0.3767 time: 1.0327 +2023/05/31 22:47:51 - mmengine - INFO - Epoch(train) [7][ 100/5758] lr: 7.9595e-04 eta: 9:47:15 time: 0.3915 data_time: 0.0027 memory: 20334 grad_norm: 0.0088 loss: 0.6865 +2023/05/31 22:48:32 - mmengine - INFO - Epoch(train) [7][ 200/5758] lr: 7.9595e-04 eta: 9:46:25 time: 0.3915 data_time: 0.0023 memory: 20334 grad_norm: 0.0178 loss: 0.6882 +2023/05/31 22:49:13 - mmengine - INFO - Epoch(train) [7][ 300/5758] lr: 7.9595e-04 eta: 9:45:36 time: 0.4006 data_time: 0.0028 memory: 20334 grad_norm: 0.0171 loss: 0.6869 +2023/05/31 22:49:55 - mmengine - INFO - Epoch(train) [7][ 400/5758] lr: 7.9595e-04 eta: 9:44:48 time: 0.4888 data_time: 0.0027 memory: 20334 grad_norm: 0.0208 loss: 0.6847 +2023/05/31 22:50:17 - mmengine - INFO - Exp name: swin_base_8xb128_fake5m_20230531_183241 +2023/05/31 22:50:38 - mmengine - INFO - Epoch(train) [7][ 500/5758] lr: 7.9595e-04 eta: 9:44:01 time: 0.4256 data_time: 0.0025 memory: 20334 grad_norm: 0.0154 loss: 0.6861 +2023/05/31 22:51:18 - mmengine - INFO - Epoch(train) [7][ 600/5758] lr: 7.9595e-04 eta: 9:43:11 time: 0.4233 data_time: 0.0031 memory: 20334 grad_norm: 0.0174 loss: 0.6872 +2023/05/31 22:52:00 - mmengine - INFO - Epoch(train) [7][ 700/5758] lr: 7.9595e-04 eta: 9:42:22 time: 0.4026 data_time: 0.0020 memory: 20334 grad_norm: 0.0144 loss: 0.6850 +2023/05/31 22:53:08 - mmengine - INFO - Epoch(train) [7][ 800/5758] lr: 7.9595e-04 eta: 9:42:33 time: 1.0925 data_time: 0.0024 memory: 20334 grad_norm: 0.0130 loss: 0.6876 +2023/05/31 22:53:45 - mmengine - INFO - Epoch(train) [7][ 900/5758] lr: 7.9595e-04 eta: 9:41:35 time: 0.3737 data_time: 0.0025 memory: 20334 grad_norm: 0.0184 loss: 0.6845 +2023/05/31 22:54:24 - mmengine - INFO - Epoch(train) [7][1000/5758] lr: 7.9595e-04 eta: 9:40:41 time: 0.3978 data_time: 0.0023 memory: 20334 grad_norm: 0.0258 loss: 0.6840 +2023/05/31 22:55:04 - mmengine - INFO - Epoch(train) [7][1100/5758] lr: 7.9595e-04 eta: 9:39:48 time: 0.3873 data_time: 0.0025 memory: 20334 grad_norm: 0.0184 loss: 0.6846 +2023/05/31 22:55:45 - mmengine - INFO - Epoch(train) [7][1200/5758] lr: 7.9595e-04 eta: 9:38:58 time: 0.4038 data_time: 0.0028 memory: 20334 grad_norm: 0.0246 loss: 0.6853 +2023/05/31 22:56:26 - mmengine - INFO - Epoch(train) [7][1300/5758] lr: 7.9595e-04 eta: 9:38:07 time: 0.4370 data_time: 0.0019 memory: 20334 grad_norm: 0.0178 loss: 0.6870 +2023/05/31 22:57:05 - mmengine - INFO - Epoch(train) [7][1400/5758] lr: 7.9595e-04 eta: 9:37:14 time: 0.3937 data_time: 0.0030 memory: 20334 grad_norm: 0.0242 loss: 0.6844 +2023/05/31 22:57:25 - mmengine - INFO - Exp name: swin_base_8xb128_fake5m_20230531_183241 +2023/05/31 22:57:45 - mmengine - INFO - Epoch(train) [7][1500/5758] lr: 7.9595e-04 eta: 9:36:21 time: 0.4178 data_time: 0.0019 memory: 20334 grad_norm: 0.0087 loss: 0.6863 +2023/05/31 22:58:24 - mmengine - INFO - Epoch(train) [7][1600/5758] lr: 7.9595e-04 eta: 9:35:28 time: 0.4099 data_time: 0.0021 memory: 20334 grad_norm: 0.0165 loss: 0.6887 +2023/05/31 22:59:05 - mmengine - INFO - Epoch(train) [7][1700/5758] lr: 7.9595e-04 eta: 9:34:39 time: 0.4031 data_time: 0.0026 memory: 20334 grad_norm: 0.0192 loss: 0.6890 +2023/05/31 22:59:46 - mmengine - INFO - Epoch(train) [7][1800/5758] lr: 7.9595e-04 eta: 9:33:48 time: 0.3943 data_time: 0.0024 memory: 20334 grad_norm: 0.0139 loss: 0.6847 +2023/05/31 23:00:27 - mmengine - INFO - Epoch(train) [7][1900/5758] lr: 7.9595e-04 eta: 9:32:59 time: 0.4035 data_time: 0.0023 memory: 20334 grad_norm: 0.0245 loss: 0.6876 +2023/05/31 23:01:09 - mmengine - INFO - Epoch(train) [7][2000/5758] lr: 7.9595e-04 eta: 9:32:11 time: 0.4083 data_time: 0.0030 memory: 20334 grad_norm: 0.0135 loss: 0.6861 +2023/05/31 23:01:49 - mmengine - INFO - Epoch(train) [7][2100/5758] lr: 7.9595e-04 eta: 9:31:20 time: 0.3640 data_time: 0.0024 memory: 20334 grad_norm: 0.0168 loss: 0.6863 +2023/05/31 23:02:29 - mmengine - INFO - Epoch(train) [7][2200/5758] lr: 7.9595e-04 eta: 9:30:29 time: 0.4015 data_time: 0.0023 memory: 20334 grad_norm: 0.0144 loss: 0.6854 +2023/05/31 23:03:09 - mmengine - INFO - Epoch(train) [7][2300/5758] lr: 7.9595e-04 eta: 9:29:38 time: 0.3630 data_time: 0.0021 memory: 20334 grad_norm: 0.0163 loss: 0.6862 +2023/05/31 23:03:50 - mmengine - INFO - Epoch(train) [7][2400/5758] lr: 7.9595e-04 eta: 9:28:47 time: 0.3996 data_time: 0.0017 memory: 20334 grad_norm: 0.0198 loss: 0.6823 +2023/05/31 23:04:10 - mmengine - INFO - Exp name: swin_base_8xb128_fake5m_20230531_183241 +2023/05/31 23:04:30 - mmengine - INFO - Epoch(train) [7][2500/5758] lr: 7.9595e-04 eta: 9:27:56 time: 0.4455 data_time: 0.0023 memory: 20334 grad_norm: 0.0218 loss: 0.6903 +2023/05/31 23:05:10 - mmengine - INFO - Epoch(train) [7][2600/5758] lr: 7.9595e-04 eta: 9:27:06 time: 0.4206 data_time: 0.0026 memory: 20334 grad_norm: 0.0181 loss: 0.6891 +2023/05/31 23:05:51 - mmengine - INFO - Epoch(train) [7][2700/5758] lr: 7.9595e-04 eta: 9:26:16 time: 0.3833 data_time: 0.0017 memory: 20334 grad_norm: 0.0279 loss: 0.6872 +2023/05/31 23:06:32 - mmengine - INFO - Epoch(train) [7][2800/5758] lr: 7.9595e-04 eta: 9:25:27 time: 0.4038 data_time: 0.0017 memory: 20334 grad_norm: 0.0139 loss: 0.6853 +2023/05/31 23:07:12 - mmengine - INFO - Epoch(train) [7][2900/5758] lr: 7.9595e-04 eta: 9:24:36 time: 0.3934 data_time: 0.0025 memory: 20334 grad_norm: 0.0113 loss: 0.6854 +2023/05/31 23:07:52 - mmengine - INFO - Epoch(train) [7][3000/5758] lr: 7.9595e-04 eta: 9:23:45 time: 0.4036 data_time: 0.0023 memory: 20334 grad_norm: 0.0171 loss: 0.6871 +2023/05/31 23:08:32 - mmengine - INFO - Epoch(train) [7][3100/5758] lr: 7.9595e-04 eta: 9:22:55 time: 0.4592 data_time: 0.0017 memory: 20334 grad_norm: 0.0211 loss: 0.6862 +2023/05/31 23:09:13 - mmengine - INFO - Epoch(train) [7][3200/5758] lr: 7.9595e-04 eta: 9:22:05 time: 0.3824 data_time: 0.0029 memory: 20334 grad_norm: 0.0181 loss: 0.6914 +2023/05/31 23:09:53 - mmengine - INFO - Epoch(train) [7][3300/5758] lr: 7.9595e-04 eta: 9:21:15 time: 0.4048 data_time: 0.0025 memory: 20334 grad_norm: 0.0138 loss: 0.6873 +2023/05/31 23:10:33 - mmengine - INFO - Epoch(train) [7][3400/5758] lr: 7.9595e-04 eta: 9:20:23 time: 0.3673 data_time: 0.0021 memory: 20334 grad_norm: 0.0177 loss: 0.6882 +2023/05/31 23:10:54 - mmengine - INFO - Exp name: swin_base_8xb128_fake5m_20230531_183241 +2023/05/31 23:11:13 - mmengine - INFO - Epoch(train) [7][3500/5758] lr: 7.9595e-04 eta: 9:19:32 time: 0.4064 data_time: 0.0025 memory: 20334 grad_norm: 0.0139 loss: 0.6901 +2023/05/31 23:11:54 - mmengine - INFO - Epoch(train) [7][3600/5758] lr: 7.9595e-04 eta: 9:18:43 time: 0.4259 data_time: 0.0017 memory: 20334 grad_norm: 0.0126 loss: 0.6852 +2023/05/31 23:12:34 - mmengine - INFO - Epoch(train) [7][3700/5758] lr: 7.9595e-04 eta: 9:17:53 time: 0.4189 data_time: 0.0026 memory: 20334 grad_norm: 0.0186 loss: 0.6862 +2023/05/31 23:13:15 - mmengine - INFO - Epoch(train) [7][3800/5758] lr: 7.9595e-04 eta: 9:17:05 time: 0.3871 data_time: 0.0015 memory: 20334 grad_norm: 0.0215 loss: 0.6917 +2023/05/31 23:13:55 - mmengine - INFO - Epoch(train) [7][3900/5758] lr: 7.9595e-04 eta: 9:16:15 time: 0.4560 data_time: 0.0015 memory: 20334 grad_norm: 0.0234 loss: 0.6864 +2023/05/31 23:14:35 - mmengine - INFO - Epoch(train) [7][4000/5758] lr: 7.9595e-04 eta: 9:15:24 time: 0.3990 data_time: 0.0019 memory: 20334 grad_norm: 0.0159 loss: 0.6854 +2023/05/31 23:15:16 - mmengine - INFO - Epoch(train) [7][4100/5758] lr: 7.9595e-04 eta: 9:14:35 time: 0.3968 data_time: 0.0016 memory: 20334 grad_norm: 0.0163 loss: 0.6881 +2023/05/31 23:15:55 - mmengine - INFO - Epoch(train) [7][4200/5758] lr: 7.9595e-04 eta: 9:13:43 time: 0.4082 data_time: 0.0016 memory: 20334 grad_norm: 0.0161 loss: 0.6845 +2023/05/31 23:16:37 - mmengine - INFO - Epoch(train) [7][4300/5758] lr: 7.9595e-04 eta: 9:12:57 time: 0.3830 data_time: 0.0019 memory: 20334 grad_norm: 0.0181 loss: 0.6873 +2023/05/31 23:17:18 - mmengine - INFO - Epoch(train) [7][4400/5758] lr: 7.9595e-04 eta: 9:12:08 time: 0.4112 data_time: 0.0026 memory: 20334 grad_norm: 0.0142 loss: 0.6873 +2023/05/31 23:17:38 - mmengine - INFO - Exp name: swin_base_8xb128_fake5m_20230531_183241 +2023/05/31 23:17:58 - mmengine - INFO - Epoch(train) [7][4500/5758] lr: 7.9595e-04 eta: 9:11:19 time: 0.3989 data_time: 0.0018 memory: 20334 grad_norm: 0.0202 loss: 0.6883 +2023/05/31 23:18:40 - mmengine - INFO - Epoch(train) [7][4600/5758] lr: 7.9595e-04 eta: 9:10:31 time: 0.4141 data_time: 0.0017 memory: 20334 grad_norm: 0.0102 loss: 0.6852 +2023/05/31 23:19:20 - mmengine - INFO - Epoch(train) [7][4700/5758] lr: 7.9595e-04 eta: 9:09:41 time: 0.5179 data_time: 0.0019 memory: 20334 grad_norm: 0.0136 loss: 0.6882 +2023/05/31 23:20:01 - mmengine - INFO - Epoch(train) [7][4800/5758] lr: 7.9595e-04 eta: 9:08:53 time: 0.4319 data_time: 0.0017 memory: 20334 grad_norm: 0.0147 loss: 0.6873 +2023/05/31 23:20:42 - mmengine - INFO - Epoch(train) [7][4900/5758] lr: 7.9595e-04 eta: 9:08:05 time: 0.3954 data_time: 0.0017 memory: 20334 grad_norm: 0.0126 loss: 0.6876 +2023/05/31 23:21:22 - mmengine - INFO - Epoch(train) [7][5000/5758] lr: 7.9595e-04 eta: 9:07:16 time: 0.3781 data_time: 0.0016 memory: 20334 grad_norm: 0.0188 loss: 0.6844 +2023/05/31 23:22:03 - mmengine - INFO - Epoch(train) [7][5100/5758] lr: 7.9595e-04 eta: 9:06:26 time: 0.4298 data_time: 0.0018 memory: 20334 grad_norm: 0.0119 loss: 0.6844 +2023/05/31 23:22:44 - mmengine - INFO - Epoch(train) [7][5200/5758] lr: 7.9595e-04 eta: 9:05:39 time: 0.4400 data_time: 0.0017 memory: 20334 grad_norm: 0.0133 loss: 0.6878 +2023/05/31 23:23:26 - mmengine - INFO - Epoch(train) [7][5300/5758] lr: 7.9595e-04 eta: 9:04:52 time: 0.4161 data_time: 0.0020 memory: 20334 grad_norm: 0.0077 loss: 0.6867 +2023/05/31 23:24:08 - mmengine - INFO - Epoch(train) [7][5400/5758] lr: 7.9595e-04 eta: 9:04:08 time: 0.4377 data_time: 0.0017 memory: 20334 grad_norm: 0.0167 loss: 0.6833 +2023/05/31 23:24:31 - mmengine - INFO - Exp name: swin_base_8xb128_fake5m_20230531_183241 +2023/05/31 23:24:51 - mmengine - INFO - Epoch(train) [7][5500/5758] lr: 7.9595e-04 eta: 9:03:22 time: 0.3777 data_time: 0.0021 memory: 20334 grad_norm: 0.0198 loss: 0.6925 +2023/05/31 23:25:29 - mmengine - INFO - Epoch(train) [7][5600/5758] lr: 7.9595e-04 eta: 9:02:30 time: 0.4018 data_time: 0.0017 memory: 20334 grad_norm: 0.0204 loss: 0.6840 +2023/05/31 23:26:09 - mmengine - INFO - Epoch(train) [7][5700/5758] lr: 7.9595e-04 eta: 9:01:40 time: 0.3764 data_time: 0.0021 memory: 20334 grad_norm: 0.0142 loss: 0.6851 +2023/05/31 23:26:34 - mmengine - INFO - Exp name: swin_base_8xb128_fake5m_20230531_183241 +2023/05/31 23:26:34 - mmengine - INFO - Saving checkpoint at 7 epochs +2023/05/31 23:26:51 - mmengine - INFO - Epoch(val) [7][8/8] accuracy/top1: 100.0000 single-label/precision_classwise: [100.0, 0.0] single-label/recall_classwise: [100.0, 0.0] single-label/f1-score_classwise: [100.0, 0.0] data_time: 0.3704 time: 1.0113 +2023/05/31 23:27:36 - mmengine - INFO - Epoch(train) [8][ 100/5758] lr: 7.2973e-04 eta: 9:00:33 time: 0.4223 data_time: 0.0014 memory: 20334 grad_norm: 0.0150 loss: 0.6873 +2023/05/31 23:28:16 - mmengine - INFO - Epoch(train) [8][ 200/5758] lr: 7.2973e-04 eta: 8:59:43 time: 0.3914 data_time: 0.0015 memory: 20334 grad_norm: 0.0126 loss: 0.6861 +2023/05/31 23:28:56 - mmengine - INFO - Epoch(train) [8][ 300/5758] lr: 7.2973e-04 eta: 8:58:55 time: 0.4302 data_time: 0.0018 memory: 20334 grad_norm: 0.0213 loss: 0.6850 +2023/05/31 23:29:37 - mmengine - INFO - Epoch(train) [8][ 400/5758] lr: 7.2973e-04 eta: 8:58:06 time: 0.4036 data_time: 0.0015 memory: 20334 grad_norm: 0.0211 loss: 0.6903 +2023/05/31 23:30:18 - mmengine - INFO - Epoch(train) [8][ 500/5758] lr: 7.2973e-04 eta: 8:57:20 time: 0.4276 data_time: 0.0016 memory: 20334 grad_norm: 0.0141 loss: 0.6874 +2023/05/31 23:30:59 - mmengine - INFO - Epoch(train) [8][ 600/5758] lr: 7.2973e-04 eta: 8:56:31 time: 0.3991 data_time: 0.0015 memory: 20334 grad_norm: 0.0171 loss: 0.6926 +2023/05/31 23:31:38 - mmengine - INFO - Exp name: swin_base_8xb128_fake5m_20230531_183241 +2023/05/31 23:31:41 - mmengine - INFO - Epoch(train) [8][ 700/5758] lr: 7.2973e-04 eta: 8:55:45 time: 0.4612 data_time: 0.0016 memory: 20334 grad_norm: 0.0161 loss: 0.6854 +2023/05/31 23:32:22 - mmengine - INFO - Epoch(train) [8][ 800/5758] lr: 7.2973e-04 eta: 8:54:59 time: 0.3963 data_time: 0.0015 memory: 20334 grad_norm: 0.0195 loss: 0.6854 +2023/05/31 23:33:03 - mmengine - INFO - Epoch(train) [8][ 900/5758] lr: 7.2973e-04 eta: 8:54:10 time: 0.4091 data_time: 0.0015 memory: 20334 grad_norm: 0.0093 loss: 0.6858 +2023/05/31 23:33:43 - mmengine - INFO - Epoch(train) [8][1000/5758] lr: 7.2973e-04 eta: 8:53:21 time: 0.3946 data_time: 0.0014 memory: 20334 grad_norm: 0.0173 loss: 0.6882 +2023/05/31 23:34:22 - mmengine - INFO - Epoch(train) [8][1100/5758] lr: 7.2973e-04 eta: 8:52:31 time: 0.3819 data_time: 0.0015 memory: 20334 grad_norm: 0.0206 loss: 0.6901 +2023/05/31 23:35:03 - mmengine - INFO - Epoch(train) [8][1200/5758] lr: 7.2973e-04 eta: 8:51:42 time: 0.3753 data_time: 0.0015 memory: 20334 grad_norm: 0.0139 loss: 0.6842 +2023/05/31 23:35:43 - mmengine - INFO - Epoch(train) [8][1300/5758] lr: 7.2973e-04 eta: 8:50:53 time: 0.3834 data_time: 0.0014 memory: 20334 grad_norm: 0.0207 loss: 0.6848 +2023/05/31 23:36:23 - mmengine - INFO - Epoch(train) [8][1400/5758] lr: 7.2973e-04 eta: 8:50:05 time: 0.4026 data_time: 0.0015 memory: 20334 grad_norm: 0.0194 loss: 0.6852 +2023/05/31 23:37:05 - mmengine - INFO - Epoch(train) [8][1500/5758] lr: 7.2973e-04 eta: 8:49:19 time: 0.4379 data_time: 0.0015 memory: 20334 grad_norm: 0.0174 loss: 0.6905 +2023/05/31 23:37:46 - mmengine - INFO - Epoch(train) [8][1600/5758] lr: 7.2973e-04 eta: 8:48:32 time: 0.4068 data_time: 0.0015 memory: 20334 grad_norm: 0.0078 loss: 0.6846 +2023/05/31 23:38:25 - mmengine - INFO - Exp name: swin_base_8xb128_fake5m_20230531_183241 +2023/05/31 23:38:27 - mmengine - INFO - Epoch(train) [8][1700/5758] lr: 7.2973e-04 eta: 8:47:45 time: 0.3627 data_time: 0.0016 memory: 20334 grad_norm: 0.0138 loss: 0.6831 +2023/05/31 23:39:08 - mmengine - INFO - Epoch(train) [8][1800/5758] lr: 7.2973e-04 eta: 8:46:57 time: 0.4403 data_time: 0.0015 memory: 20334 grad_norm: 0.0195 loss: 0.6897 +2023/05/31 23:39:49 - mmengine - INFO - Epoch(train) [8][1900/5758] lr: 7.2973e-04 eta: 8:46:09 time: 0.4031 data_time: 0.0015 memory: 20334 grad_norm: 0.0151 loss: 0.6837 +2023/05/31 23:40:28 - mmengine - INFO - Epoch(train) [8][2000/5758] lr: 7.2973e-04 eta: 8:45:20 time: 0.4056 data_time: 0.0015 memory: 20334 grad_norm: 0.0154 loss: 0.6881 +2023/05/31 23:41:10 - mmengine - INFO - Epoch(train) [8][2100/5758] lr: 7.2973e-04 eta: 8:44:33 time: 0.4114 data_time: 0.0015 memory: 20334 grad_norm: 0.0176 loss: 0.6852 +2023/05/31 23:41:49 - mmengine - INFO - Epoch(train) [8][2200/5758] lr: 7.2973e-04 eta: 8:43:43 time: 0.4120 data_time: 0.0012 memory: 20334 grad_norm: 0.0169 loss: 0.6877 +2023/05/31 23:42:30 - mmengine - INFO - Epoch(train) [8][2300/5758] lr: 7.2973e-04 eta: 8:42:56 time: 0.4703 data_time: 0.0013 memory: 20334 grad_norm: 0.0109 loss: 0.6843 +2023/05/31 23:43:10 - mmengine - INFO - Epoch(train) [8][2400/5758] lr: 7.2973e-04 eta: 8:42:08 time: 0.4309 data_time: 0.0013 memory: 20334 grad_norm: 0.0137 loss: 0.6947 +2023/05/31 23:43:50 - mmengine - INFO - Epoch(train) [8][2500/5758] lr: 7.2973e-04 eta: 8:41:19 time: 0.4007 data_time: 0.0012 memory: 20334 grad_norm: 0.0156 loss: 0.6836 +2023/05/31 23:44:32 - mmengine - INFO - Epoch(train) [8][2600/5758] lr: 7.2973e-04 eta: 8:40:34 time: 0.4241 data_time: 0.0017 memory: 20334 grad_norm: 0.0155 loss: 0.6884 +2023/05/31 23:45:12 - mmengine - INFO - Exp name: swin_base_8xb128_fake5m_20230531_183241 +2023/05/31 23:45:14 - mmengine - INFO - Epoch(train) [8][2700/5758] lr: 7.2973e-04 eta: 8:39:48 time: 0.4685 data_time: 0.0017 memory: 20334 grad_norm: 0.0120 loss: 0.6840 +2023/05/31 23:45:54 - mmengine - INFO - Epoch(train) [8][2800/5758] lr: 7.2973e-04 eta: 8:38:59 time: 0.4118 data_time: 0.0014 memory: 20334 grad_norm: 0.0096 loss: 0.6828 +2023/05/31 23:46:35 - mmengine - INFO - Epoch(train) [8][2900/5758] lr: 7.2973e-04 eta: 8:38:12 time: 0.4031 data_time: 0.0019 memory: 20334 grad_norm: 0.0122 loss: 0.6836 +2023/05/31 23:47:14 - mmengine - INFO - Epoch(train) [8][3000/5758] lr: 7.2973e-04 eta: 8:37:23 time: 0.3638 data_time: 0.0018 memory: 20334 grad_norm: 0.0161 loss: 0.6854 +2023/05/31 23:47:54 - mmengine - INFO - Epoch(train) [8][3100/5758] lr: 7.2973e-04 eta: 8:36:33 time: 0.4331 data_time: 0.0020 memory: 20334 grad_norm: 0.0202 loss: 0.6902 +2023/05/31 23:48:33 - mmengine - INFO - Epoch(train) [8][3200/5758] lr: 7.2973e-04 eta: 8:35:44 time: 0.3863 data_time: 0.0021 memory: 20334 grad_norm: 0.0199 loss: 0.6832 +2023/05/31 23:49:14 - mmengine - INFO - Epoch(train) [8][3300/5758] lr: 7.2973e-04 eta: 8:34:57 time: 0.4603 data_time: 0.0018 memory: 20334 grad_norm: 0.0144 loss: 0.6889 +2023/05/31 23:49:53 - mmengine - INFO - Epoch(train) [8][3400/5758] lr: 7.2973e-04 eta: 8:34:07 time: 0.3926 data_time: 0.0016 memory: 20334 grad_norm: 0.0170 loss: 0.6835 +2023/05/31 23:50:34 - mmengine - INFO - Epoch(train) [8][3500/5758] lr: 7.2973e-04 eta: 8:33:19 time: 0.4093 data_time: 0.0018 memory: 20334 grad_norm: 0.0176 loss: 0.6804 +2023/05/31 23:51:14 - mmengine - INFO - Epoch(train) [8][3600/5758] lr: 7.2973e-04 eta: 8:32:32 time: 0.4041 data_time: 0.0016 memory: 20334 grad_norm: 0.0166 loss: 0.6864 +2023/05/31 23:51:51 - mmengine - INFO - Exp name: swin_base_8xb128_fake5m_20230531_183241 +2023/05/31 23:51:53 - mmengine - INFO - Epoch(train) [8][3700/5758] lr: 7.2973e-04 eta: 8:31:42 time: 0.3867 data_time: 0.0019 memory: 20334 grad_norm: 0.0128 loss: 0.6867 +2023/05/31 23:52:33 - mmengine - INFO - Epoch(train) [8][3800/5758] lr: 7.2973e-04 eta: 8:30:53 time: 0.4175 data_time: 0.0023 memory: 20334 grad_norm: 0.0122 loss: 0.6845 +2023/05/31 23:53:12 - mmengine - INFO - Epoch(train) [8][3900/5758] lr: 7.2973e-04 eta: 8:30:04 time: 0.3992 data_time: 0.0015 memory: 20334 grad_norm: 0.0148 loss: 0.6846 +2023/05/31 23:53:52 - mmengine - INFO - Epoch(train) [8][4000/5758] lr: 7.2973e-04 eta: 8:29:15 time: 0.4085 data_time: 0.0017 memory: 20334 grad_norm: 0.0194 loss: 0.6884 +2023/05/31 23:54:32 - mmengine - INFO - Epoch(train) [8][4100/5758] lr: 7.2973e-04 eta: 8:28:27 time: 0.4341 data_time: 0.0019 memory: 20334 grad_norm: 0.0218 loss: 0.6902 +2023/05/31 23:55:12 - mmengine - INFO - Epoch(train) [8][4200/5758] lr: 7.2973e-04 eta: 8:27:39 time: 0.4425 data_time: 0.0026 memory: 20334 grad_norm: 0.0092 loss: 0.6831 +2023/05/31 23:55:53 - mmengine - INFO - Epoch(train) [8][4300/5758] lr: 7.2973e-04 eta: 8:26:52 time: 0.3826 data_time: 0.0031 memory: 20334 grad_norm: 0.0142 loss: 0.6839 +2023/05/31 23:56:32 - mmengine - INFO - Epoch(train) [8][4400/5758] lr: 7.2973e-04 eta: 8:26:03 time: 0.3903 data_time: 0.0020 memory: 20334 grad_norm: 0.0141 loss: 0.6895 +2023/05/31 23:57:11 - mmengine - INFO - Epoch(train) [8][4500/5758] lr: 7.2973e-04 eta: 8:25:14 time: 0.4005 data_time: 0.0021 memory: 20334 grad_norm: 0.0095 loss: 0.6898 +2023/05/31 23:57:51 - mmengine - INFO - Epoch(train) [8][4600/5758] lr: 7.2973e-04 eta: 8:24:25 time: 0.3952 data_time: 0.0024 memory: 20334 grad_norm: 0.0182 loss: 0.6843 +2023/05/31 23:58:29 - mmengine - INFO - Exp name: swin_base_8xb128_fake5m_20230531_183241 +2023/05/31 23:58:31 - mmengine - INFO - Epoch(train) [8][4700/5758] lr: 7.2973e-04 eta: 8:23:39 time: 0.3863 data_time: 0.0017 memory: 20334 grad_norm: 0.0133 loss: 0.6857 +2023/05/31 23:59:12 - mmengine - INFO - Epoch(train) [8][4800/5758] lr: 7.2973e-04 eta: 8:22:52 time: 0.4085 data_time: 0.0020 memory: 20334 grad_norm: 0.0106 loss: 0.6831 +2023/05/31 23:59:52 - mmengine - INFO - Epoch(train) [8][4900/5758] lr: 7.2973e-04 eta: 8:22:04 time: 0.3887 data_time: 0.0021 memory: 20334 grad_norm: 0.0196 loss: 0.6865 +2023/06/01 00:00:34 - mmengine - INFO - Epoch(train) [8][5000/5758] lr: 7.2973e-04 eta: 8:21:19 time: 0.4114 data_time: 0.0019 memory: 20334 grad_norm: 0.0168 loss: 0.6831 +2023/06/01 00:01:13 - mmengine - INFO - Epoch(train) [8][5100/5758] lr: 7.2973e-04 eta: 8:20:30 time: 0.4419 data_time: 0.0017 memory: 20334 grad_norm: 0.0174 loss: 0.6849 +2023/06/01 00:01:53 - mmengine - INFO - Epoch(train) [8][5200/5758] lr: 7.2973e-04 eta: 8:19:42 time: 0.4285 data_time: 0.0015 memory: 20334 grad_norm: 0.0248 loss: 0.6894 +2023/06/01 00:02:33 - mmengine - INFO - Epoch(train) [8][5300/5758] lr: 7.2973e-04 eta: 8:18:55 time: 0.3812 data_time: 0.0017 memory: 20334 grad_norm: 0.0194 loss: 0.6841 +2023/06/01 00:03:13 - mmengine - INFO - Epoch(train) [8][5400/5758] lr: 7.2973e-04 eta: 8:18:07 time: 0.4205 data_time: 0.0017 memory: 20334 grad_norm: 0.0108 loss: 0.6873 +2023/06/01 00:03:53 - mmengine - INFO - Epoch(train) [8][5500/5758] lr: 7.2973e-04 eta: 8:17:19 time: 0.4115 data_time: 0.0016 memory: 20334 grad_norm: 0.0099 loss: 0.6845 +2023/06/01 00:04:33 - mmengine - INFO - Epoch(train) [8][5600/5758] lr: 7.2973e-04 eta: 8:16:31 time: 0.3997 data_time: 0.0018 memory: 20334 grad_norm: 0.0236 loss: 0.6868 +2023/06/01 00:05:10 - mmengine - INFO - Exp name: swin_base_8xb128_fake5m_20230531_183241 +2023/06/01 00:05:12 - mmengine - INFO - Epoch(train) [8][5700/5758] lr: 7.2973e-04 eta: 8:15:43 time: 0.3803 data_time: 0.0017 memory: 20334 grad_norm: 0.0141 loss: 0.6926 +2023/06/01 00:05:35 - mmengine - INFO - Exp name: swin_base_8xb128_fake5m_20230531_183241 +2023/06/01 00:05:35 - mmengine - INFO - Saving checkpoint at 8 epochs +2023/06/01 00:05:52 - mmengine - INFO - Epoch(val) [8][8/8] accuracy/top1: 100.0000 single-label/precision_classwise: [100.0, 0.0] single-label/recall_classwise: [100.0, 0.0] single-label/f1-score_classwise: [100.0, 0.0] data_time: 0.3851 time: 1.0094 +2023/06/01 00:06:37 - mmengine - INFO - Epoch(train) [9][ 100/5758] lr: 6.5796e-04 eta: 8:14:34 time: 0.3657 data_time: 0.0016 memory: 20334 grad_norm: 0.0142 loss: 0.6866 +2023/06/01 00:07:16 - mmengine - INFO - Epoch(train) [9][ 200/5758] lr: 6.5796e-04 eta: 8:13:46 time: 0.4105 data_time: 0.0016 memory: 20334 grad_norm: 0.0199 loss: 0.6880 +2023/06/01 00:07:57 - mmengine - INFO - Epoch(train) [9][ 300/5758] lr: 6.5796e-04 eta: 8:12:59 time: 0.3799 data_time: 0.0017 memory: 20334 grad_norm: 0.0085 loss: 0.6842 +2023/06/01 00:08:36 - mmengine - INFO - Epoch(train) [9][ 400/5758] lr: 6.5796e-04 eta: 8:12:11 time: 0.3870 data_time: 0.0016 memory: 20334 grad_norm: 0.0180 loss: 0.6870 +2023/06/01 00:09:17 - mmengine - INFO - Epoch(train) [9][ 500/5758] lr: 6.5796e-04 eta: 8:11:24 time: 0.4097 data_time: 0.0016 memory: 20334 grad_norm: 0.0165 loss: 0.6872 +2023/06/01 00:09:56 - mmengine - INFO - Epoch(train) [9][ 600/5758] lr: 6.5796e-04 eta: 8:10:36 time: 0.4078 data_time: 0.0015 memory: 20334 grad_norm: 0.0171 loss: 0.6877 +2023/06/01 00:10:37 - mmengine - INFO - Epoch(train) [9][ 700/5758] lr: 6.5796e-04 eta: 8:09:50 time: 0.4025 data_time: 0.0015 memory: 20334 grad_norm: 0.0262 loss: 0.6899 +2023/06/01 00:11:17 - mmengine - INFO - Epoch(train) [9][ 800/5758] lr: 6.5796e-04 eta: 8:09:03 time: 0.4089 data_time: 0.0016 memory: 20334 grad_norm: 0.0134 loss: 0.6825 +2023/06/01 00:11:59 - mmengine - INFO - Epoch(train) [9][ 900/5758] lr: 6.5796e-04 eta: 8:08:18 time: 0.4349 data_time: 0.0017 memory: 20334 grad_norm: 0.0166 loss: 0.6853 +2023/06/01 00:12:14 - mmengine - INFO - Exp name: swin_base_8xb128_fake5m_20230531_183241 +2023/06/01 00:12:39 - mmengine - INFO - Epoch(train) [9][1000/5758] lr: 6.5796e-04 eta: 8:07:31 time: 0.4156 data_time: 0.0016 memory: 20334 grad_norm: 0.0297 loss: 0.6826 +2023/06/01 00:13:18 - mmengine - INFO - Epoch(train) [9][1100/5758] lr: 6.5796e-04 eta: 8:06:43 time: 0.4004 data_time: 0.0017 memory: 20334 grad_norm: 0.0241 loss: 0.6842 +2023/06/01 00:13:58 - mmengine - INFO - Epoch(train) [9][1200/5758] lr: 6.5796e-04 eta: 8:05:55 time: 0.3890 data_time: 0.0015 memory: 20334 grad_norm: 0.0166 loss: 0.6880 +2023/06/01 00:14:39 - mmengine - INFO - Epoch(train) [9][1300/5758] lr: 6.5796e-04 eta: 8:05:09 time: 0.4182 data_time: 0.0019 memory: 20334 grad_norm: 0.0172 loss: 0.6901 +2023/06/01 00:15:19 - mmengine - INFO - Epoch(train) [9][1400/5758] lr: 6.5796e-04 eta: 8:04:22 time: 0.3768 data_time: 0.0018 memory: 20334 grad_norm: 0.0236 loss: 0.6926 +2023/06/01 00:16:00 - mmengine - INFO - Epoch(train) [9][1500/5758] lr: 6.5796e-04 eta: 8:03:36 time: 0.3780 data_time: 0.0016 memory: 20334 grad_norm: 0.0203 loss: 0.6883 +2023/06/01 00:16:40 - mmengine - INFO - Epoch(train) [9][1600/5758] lr: 6.5796e-04 eta: 8:02:49 time: 0.4377 data_time: 0.0018 memory: 20334 grad_norm: 0.0207 loss: 0.6895 +2023/06/01 00:17:19 - mmengine - INFO - Epoch(train) [9][1700/5758] lr: 6.5796e-04 eta: 8:02:01 time: 0.4236 data_time: 0.0015 memory: 20334 grad_norm: 0.0263 loss: 0.6902 +2023/06/01 00:17:58 - mmengine - INFO - Epoch(train) [9][1800/5758] lr: 6.5796e-04 eta: 8:01:12 time: 0.3900 data_time: 0.0015 memory: 20334 grad_norm: 0.0167 loss: 0.6927 +2023/06/01 00:18:38 - mmengine - INFO - Epoch(train) [9][1900/5758] lr: 6.5796e-04 eta: 8:00:25 time: 0.4081 data_time: 0.0014 memory: 20334 grad_norm: 0.0168 loss: 0.6866 +2023/06/01 00:18:53 - mmengine - INFO - Exp name: swin_base_8xb128_fake5m_20230531_183241 +2023/06/01 00:19:18 - mmengine - INFO - Epoch(train) [9][2000/5758] lr: 6.5796e-04 eta: 7:59:39 time: 0.4291 data_time: 0.0015 memory: 20334 grad_norm: 0.0119 loss: 0.6865 +2023/06/01 00:19:57 - mmengine - INFO - Epoch(train) [9][2100/5758] lr: 6.5796e-04 eta: 7:58:50 time: 0.3696 data_time: 0.0015 memory: 20334 grad_norm: 0.0185 loss: 0.6884 +2023/06/01 00:20:38 - mmengine - INFO - Epoch(train) [9][2200/5758] lr: 6.5796e-04 eta: 7:58:04 time: 0.4319 data_time: 0.0014 memory: 20334 grad_norm: 0.0166 loss: 0.6853 +2023/06/01 00:21:18 - mmengine - INFO - Epoch(train) [9][2300/5758] lr: 6.5796e-04 eta: 7:57:17 time: 0.4174 data_time: 0.0015 memory: 20334 grad_norm: 0.0148 loss: 0.6844 +2023/06/01 00:21:58 - mmengine - INFO - Epoch(train) [9][2400/5758] lr: 6.5796e-04 eta: 7:56:30 time: 0.4347 data_time: 0.0015 memory: 20334 grad_norm: 0.0178 loss: 0.6882 +2023/06/01 00:22:38 - mmengine - INFO - Epoch(train) [9][2500/5758] lr: 6.5796e-04 eta: 7:55:44 time: 0.3834 data_time: 0.0017 memory: 20334 grad_norm: 0.0168 loss: 0.6823 +2023/06/01 00:23:18 - mmengine - INFO - Epoch(train) [9][2600/5758] lr: 6.5796e-04 eta: 7:54:57 time: 0.3810 data_time: 0.0017 memory: 20334 grad_norm: 0.0126 loss: 0.6860 +2023/06/01 00:23:58 - mmengine - INFO - Epoch(train) [9][2700/5758] lr: 6.5796e-04 eta: 7:54:10 time: 0.3961 data_time: 0.0017 memory: 20334 grad_norm: 0.0197 loss: 0.6837 +2023/06/01 00:24:39 - mmengine - INFO - Epoch(train) [9][2800/5758] lr: 6.5796e-04 eta: 7:53:26 time: 0.4435 data_time: 0.0018 memory: 20334 grad_norm: 0.0157 loss: 0.6875 +2023/06/01 00:25:19 - mmengine - INFO - Epoch(train) [9][2900/5758] lr: 6.5796e-04 eta: 7:52:38 time: 0.3820 data_time: 0.0018 memory: 20334 grad_norm: 0.0180 loss: 0.6824 +2023/06/01 00:25:33 - mmengine - INFO - Exp name: swin_base_8xb128_fake5m_20230531_183241 +2023/06/01 00:25:59 - mmengine - INFO - Epoch(train) [9][3000/5758] lr: 6.5796e-04 eta: 7:51:51 time: 0.4053 data_time: 0.0016 memory: 20334 grad_norm: 0.0163 loss: 0.6910 +2023/06/01 00:26:38 - mmengine - INFO - Epoch(train) [9][3100/5758] lr: 6.5796e-04 eta: 7:51:04 time: 0.3820 data_time: 0.0016 memory: 20334 grad_norm: 0.0174 loss: 0.6902 +2023/06/01 00:27:18 - mmengine - INFO - Epoch(train) [9][3200/5758] lr: 6.5796e-04 eta: 7:50:18 time: 0.4061 data_time: 0.0018 memory: 20334 grad_norm: 0.0130 loss: 0.6888 +2023/06/01 00:27:58 - mmengine - INFO - Epoch(train) [9][3300/5758] lr: 6.5796e-04 eta: 7:49:31 time: 0.3685 data_time: 0.0017 memory: 20334 grad_norm: 0.0157 loss: 0.6879 +2023/06/01 00:28:39 - mmengine - INFO - Epoch(train) [9][3400/5758] lr: 6.5796e-04 eta: 7:48:45 time: 0.4373 data_time: 0.0017 memory: 20334 grad_norm: 0.0146 loss: 0.6809 +2023/06/01 00:29:19 - mmengine - INFO - Epoch(train) [9][3500/5758] lr: 6.5796e-04 eta: 7:47:59 time: 0.3634 data_time: 0.0019 memory: 20334 grad_norm: 0.0162 loss: 0.6866 +2023/06/01 00:29:59 - mmengine - INFO - Epoch(train) [9][3600/5758] lr: 6.5796e-04 eta: 7:47:13 time: 0.3802 data_time: 0.0017 memory: 20334 grad_norm: 0.0205 loss: 0.6849 +2023/06/01 00:30:39 - mmengine - INFO - Epoch(train) [9][3700/5758] lr: 6.5796e-04 eta: 7:46:25 time: 0.3969 data_time: 0.0016 memory: 20334 grad_norm: 0.0171 loss: 0.6884 +2023/06/01 00:31:19 - mmengine - INFO - Epoch(train) [9][3800/5758] lr: 6.5796e-04 eta: 7:45:40 time: 0.4089 data_time: 0.0017 memory: 20334 grad_norm: 0.0101 loss: 0.6838 +2023/06/01 00:31:59 - mmengine - INFO - Epoch(train) [9][3900/5758] lr: 6.5796e-04 eta: 7:44:53 time: 0.3888 data_time: 0.0018 memory: 20334 grad_norm: 0.0148 loss: 0.6896 +2023/06/01 00:32:14 - mmengine - INFO - Exp name: swin_base_8xb128_fake5m_20230531_183241 +2023/06/01 00:32:40 - mmengine - INFO - Epoch(train) [9][4000/5758] lr: 6.5796e-04 eta: 7:44:08 time: 0.4657 data_time: 0.0015 memory: 20334 grad_norm: 0.0133 loss: 0.6859 +2023/06/01 00:33:21 - mmengine - INFO - Epoch(train) [9][4100/5758] lr: 6.5796e-04 eta: 7:43:22 time: 0.4193 data_time: 0.0016 memory: 20334 grad_norm: 0.0126 loss: 0.6846 +2023/06/01 00:34:01 - mmengine - INFO - Epoch(train) [9][4200/5758] lr: 6.5796e-04 eta: 7:42:36 time: 0.4574 data_time: 0.0016 memory: 20334 grad_norm: 0.0164 loss: 0.6868 +2023/06/01 00:34:40 - mmengine - INFO - Epoch(train) [9][4300/5758] lr: 6.5796e-04 eta: 7:41:49 time: 0.4001 data_time: 0.0016 memory: 20334 grad_norm: 0.0199 loss: 0.6844 +2023/06/01 00:35:20 - mmengine - INFO - Epoch(train) [9][4400/5758] lr: 6.5796e-04 eta: 7:41:03 time: 0.4158 data_time: 0.0014 memory: 20334 grad_norm: 0.0163 loss: 0.6905 +2023/06/01 00:36:00 - mmengine - INFO - Epoch(train) [9][4500/5758] lr: 6.5796e-04 eta: 7:40:16 time: 0.4090 data_time: 0.0013 memory: 20334 grad_norm: 0.0154 loss: 0.6873 +2023/06/01 00:36:40 - mmengine - INFO - Epoch(train) [9][4600/5758] lr: 6.5796e-04 eta: 7:39:30 time: 0.4068 data_time: 0.0014 memory: 20334 grad_norm: 0.0163 loss: 0.6849 +2023/06/01 00:37:20 - mmengine - INFO - Epoch(train) [9][4700/5758] lr: 6.5796e-04 eta: 7:38:43 time: 0.3742 data_time: 0.0013 memory: 20334 grad_norm: 0.0120 loss: 0.6895 +2023/06/01 00:38:00 - mmengine - INFO - Epoch(train) [9][4800/5758] lr: 6.5796e-04 eta: 7:37:57 time: 0.3626 data_time: 0.0013 memory: 20334 grad_norm: 0.0179 loss: 0.6838 +2023/06/01 00:38:40 - mmengine - INFO - Epoch(train) [9][4900/5758] lr: 6.5796e-04 eta: 7:37:11 time: 0.4155 data_time: 0.0016 memory: 20334 grad_norm: 0.0104 loss: 0.6882 +2023/06/01 00:38:55 - mmengine - INFO - Exp name: swin_base_8xb128_fake5m_20230531_183241 +2023/06/01 00:39:21 - mmengine - INFO - Epoch(train) [9][5000/5758] lr: 6.5796e-04 eta: 7:36:26 time: 0.4163 data_time: 0.0016 memory: 20334 grad_norm: 0.0163 loss: 0.6855 +2023/06/01 00:40:00 - mmengine - INFO - Epoch(train) [9][5100/5758] lr: 6.5796e-04 eta: 7:35:39 time: 0.3880 data_time: 0.0014 memory: 20334 grad_norm: 0.0136 loss: 0.6881 +2023/06/01 00:40:40 - mmengine - INFO - Epoch(train) [9][5200/5758] lr: 6.5796e-04 eta: 7:34:53 time: 0.3931 data_time: 0.0015 memory: 20334 grad_norm: 0.0151 loss: 0.6864 +2023/06/01 00:41:20 - mmengine - INFO - Epoch(train) [9][5300/5758] lr: 6.5796e-04 eta: 7:34:06 time: 0.3680 data_time: 0.0015 memory: 20334 grad_norm: 0.0214 loss: 0.6849 +2023/06/01 00:42:00 - mmengine - INFO - Epoch(train) [9][5400/5758] lr: 6.5796e-04 eta: 7:33:20 time: 0.4489 data_time: 0.0017 memory: 20334 grad_norm: 0.0175 loss: 0.6825 +2023/06/01 00:42:39 - mmengine - INFO - Epoch(train) [9][5500/5758] lr: 6.5796e-04 eta: 7:32:33 time: 0.4165 data_time: 0.0017 memory: 20334 grad_norm: 0.0185 loss: 0.6859 +2023/06/01 00:43:18 - mmengine - INFO - Epoch(train) [9][5600/5758] lr: 6.5796e-04 eta: 7:31:46 time: 0.3864 data_time: 0.0018 memory: 20334 grad_norm: 0.0111 loss: 0.6867 +2023/06/01 00:43:58 - mmengine - INFO - Epoch(train) [9][5700/5758] lr: 6.5796e-04 eta: 7:31:00 time: 0.3709 data_time: 0.0015 memory: 20334 grad_norm: 0.0168 loss: 0.6819 +2023/06/01 00:44:21 - mmengine - INFO - Exp name: swin_base_8xb128_fake5m_20230531_183241 +2023/06/01 00:44:21 - mmengine - INFO - Saving checkpoint at 9 epochs +2023/06/01 00:44:38 - mmengine - INFO - Epoch(val) [9][8/8] accuracy/top1: 100.0000 single-label/precision_classwise: [100.0, 0.0] single-label/recall_classwise: [100.0, 0.0] single-label/f1-score_classwise: [100.0, 0.0] data_time: 0.3715 time: 0.9948 +2023/06/01 00:45:22 - mmengine - INFO - Epoch(train) [10][ 100/5758] lr: 5.8244e-04 eta: 7:29:53 time: 0.4262 data_time: 0.0016 memory: 20334 grad_norm: 0.0136 loss: 0.6862 +2023/06/01 00:45:54 - mmengine - INFO - Exp name: swin_base_8xb128_fake5m_20230531_183241 +2023/06/01 00:46:03 - mmengine - INFO - Epoch(train) [10][ 200/5758] lr: 5.8244e-04 eta: 7:29:07 time: 0.3890 data_time: 0.0016 memory: 20334 grad_norm: 0.0158 loss: 0.6849 +2023/06/01 00:46:41 - mmengine - INFO - Epoch(train) [10][ 300/5758] lr: 5.8244e-04 eta: 7:28:19 time: 0.4332 data_time: 0.0018 memory: 20334 grad_norm: 0.0155 loss: 0.6848 +2023/06/01 00:47:20 - mmengine - INFO - Epoch(train) [10][ 400/5758] lr: 5.8244e-04 eta: 7:27:32 time: 0.3700 data_time: 0.0018 memory: 20334 grad_norm: 0.0152 loss: 0.6876 +2023/06/01 00:48:00 - mmengine - INFO - Epoch(train) [10][ 500/5758] lr: 5.8244e-04 eta: 7:26:47 time: 0.4105 data_time: 0.0016 memory: 20334 grad_norm: 0.0105 loss: 0.6874 +2023/06/01 00:48:41 - mmengine - INFO - Epoch(train) [10][ 600/5758] lr: 5.8244e-04 eta: 7:26:01 time: 0.4195 data_time: 0.0016 memory: 20334 grad_norm: 0.0116 loss: 0.6903 +2023/06/01 00:49:20 - mmengine - INFO - Epoch(train) [10][ 700/5758] lr: 5.8244e-04 eta: 7:25:15 time: 0.3901 data_time: 0.0018 memory: 20334 grad_norm: 0.0214 loss: 0.6843 +2023/06/01 00:50:00 - mmengine - INFO - Epoch(train) [10][ 800/5758] lr: 5.8244e-04 eta: 7:24:29 time: 0.4323 data_time: 0.0014 memory: 20334 grad_norm: 0.0161 loss: 0.6893 +2023/06/01 00:50:39 - mmengine - INFO - Epoch(train) [10][ 900/5758] lr: 5.8244e-04 eta: 7:23:42 time: 0.3841 data_time: 0.0017 memory: 20334 grad_norm: 0.0155 loss: 0.6848 +2023/06/01 00:51:19 - mmengine - INFO - Epoch(train) [10][1000/5758] lr: 5.8244e-04 eta: 7:22:56 time: 0.3833 data_time: 0.0019 memory: 20334 grad_norm: 0.0183 loss: 0.6794 +2023/06/01 00:51:59 - mmengine - INFO - Epoch(train) [10][1100/5758] lr: 5.8244e-04 eta: 7:22:10 time: 0.3641 data_time: 0.0018 memory: 20334 grad_norm: 0.0127 loss: 0.6832 +2023/06/01 00:52:31 - mmengine - INFO - Exp name: swin_base_8xb128_fake5m_20230531_183241 +2023/06/01 00:52:39 - mmengine - INFO - Epoch(train) [10][1200/5758] lr: 5.8244e-04 eta: 7:21:25 time: 0.3908 data_time: 0.0017 memory: 20334 grad_norm: 0.0186 loss: 0.6863 +2023/06/01 00:53:20 - mmengine - INFO - Epoch(train) [10][1300/5758] lr: 5.8244e-04 eta: 7:20:40 time: 0.4295 data_time: 0.0015 memory: 20334 grad_norm: 0.0171 loss: 0.6921 +2023/06/01 00:53:58 - mmengine - INFO - Epoch(train) [10][1400/5758] lr: 5.8244e-04 eta: 7:19:52 time: 0.3710 data_time: 0.0017 memory: 20334 grad_norm: 0.0203 loss: 0.6852 +2023/06/01 00:54:38 - mmengine - INFO - Epoch(train) [10][1500/5758] lr: 5.8244e-04 eta: 7:19:06 time: 0.3865 data_time: 0.0018 memory: 20334 grad_norm: 0.0182 loss: 0.6863 +2023/06/01 00:55:18 - mmengine - INFO - Epoch(train) [10][1600/5758] lr: 5.8244e-04 eta: 7:18:21 time: 0.4194 data_time: 0.0015 memory: 20334 grad_norm: 0.0177 loss: 0.6906 +2023/06/01 00:55:58 - mmengine - INFO - Epoch(train) [10][1700/5758] lr: 5.8244e-04 eta: 7:17:35 time: 0.4043 data_time: 0.0015 memory: 20334 grad_norm: 0.0141 loss: 0.6836 +2023/06/01 00:56:38 - mmengine - INFO - Epoch(train) [10][1800/5758] lr: 5.8244e-04 eta: 7:16:50 time: 0.3795 data_time: 0.0016 memory: 20334 grad_norm: 0.0189 loss: 0.6863 +2023/06/01 00:57:18 - mmengine - INFO - Epoch(train) [10][1900/5758] lr: 5.8244e-04 eta: 7:16:04 time: 0.4205 data_time: 0.0014 memory: 20334 grad_norm: 0.0117 loss: 0.6885 +2023/06/01 00:57:59 - mmengine - INFO - Epoch(train) [10][2000/5758] lr: 5.8244e-04 eta: 7:15:19 time: 0.4040 data_time: 0.0016 memory: 20334 grad_norm: 0.0186 loss: 0.6876 +2023/06/01 00:58:38 - mmengine - INFO - Epoch(train) [10][2100/5758] lr: 5.8244e-04 eta: 7:14:33 time: 0.3708 data_time: 0.0015 memory: 20334 grad_norm: 0.0202 loss: 0.6900 +2023/06/01 00:59:09 - mmengine - INFO - Exp name: swin_base_8xb128_fake5m_20230531_183241 +2023/06/01 00:59:17 - mmengine - INFO - Epoch(train) [10][2200/5758] lr: 5.8244e-04 eta: 7:13:46 time: 0.4004 data_time: 0.0016 memory: 20334 grad_norm: 0.0131 loss: 0.6886 +2023/06/01 00:59:56 - mmengine - INFO - Epoch(train) [10][2300/5758] lr: 5.8244e-04 eta: 7:13:00 time: 0.3760 data_time: 0.0019 memory: 20334 grad_norm: 0.0103 loss: 0.6827 +2023/06/01 01:00:37 - mmengine - INFO - Epoch(train) [10][2400/5758] lr: 5.8244e-04 eta: 7:12:15 time: 0.4136 data_time: 0.0017 memory: 20334 grad_norm: 0.0192 loss: 0.6844 +2023/06/01 01:01:17 - mmengine - INFO - Epoch(train) [10][2500/5758] lr: 5.8244e-04 eta: 7:11:30 time: 0.4294 data_time: 0.0020 memory: 20334 grad_norm: 0.0161 loss: 0.6878 +2023/06/01 01:01:57 - mmengine - INFO - Epoch(train) [10][2600/5758] lr: 5.8244e-04 eta: 7:10:44 time: 0.4233 data_time: 0.0013 memory: 20334 grad_norm: 0.0128 loss: 0.6899 +2023/06/01 01:02:36 - mmengine - INFO - Epoch(train) [10][2700/5758] lr: 5.8244e-04 eta: 7:09:58 time: 0.4120 data_time: 0.0015 memory: 20334 grad_norm: 0.0128 loss: 0.6877 +2023/06/01 01:03:15 - mmengine - INFO - Epoch(train) [10][2800/5758] lr: 5.8244e-04 eta: 7:09:12 time: 0.4175 data_time: 0.0019 memory: 20334 grad_norm: 0.0157 loss: 0.6882 +2023/06/01 01:03:55 - mmengine - INFO - Epoch(train) [10][2900/5758] lr: 5.8244e-04 eta: 7:08:26 time: 0.4090 data_time: 0.0015 memory: 20334 grad_norm: 0.0175 loss: 0.6817 +2023/06/01 01:04:34 - mmengine - INFO - Epoch(train) [10][3000/5758] lr: 5.8244e-04 eta: 7:07:40 time: 0.3908 data_time: 0.0016 memory: 20334 grad_norm: 0.0069 loss: 0.6842 +2023/06/01 01:05:13 - mmengine - INFO - Epoch(train) [10][3100/5758] lr: 5.8244e-04 eta: 7:06:54 time: 0.4035 data_time: 0.0017 memory: 20334 grad_norm: 0.0181 loss: 0.6884 +2023/06/01 01:05:46 - mmengine - INFO - Exp name: swin_base_8xb128_fake5m_20230531_183241 +2023/06/01 01:05:54 - mmengine - INFO - Epoch(train) [10][3200/5758] lr: 5.8244e-04 eta: 7:06:09 time: 0.3929 data_time: 0.0016 memory: 20334 grad_norm: 0.0088 loss: 0.6831 +2023/06/01 01:06:34 - mmengine - INFO - Epoch(train) [10][3300/5758] lr: 5.8244e-04 eta: 7:05:23 time: 0.4198 data_time: 0.0016 memory: 20334 grad_norm: 0.0117 loss: 0.6844 +2023/06/01 01:08:24 - mmengine - INFO - Epoch(train) [10][3400/5758] lr: 5.8244e-04 eta: 7:05:54 time: 0.4055 data_time: 0.0017 memory: 20334 grad_norm: 0.0115 loss: 0.6901 +2023/06/01 01:09:02 - mmengine - INFO - Epoch(train) [10][3500/5758] lr: 5.8244e-04 eta: 7:05:07 time: 0.3676 data_time: 0.0017 memory: 20334 grad_norm: 0.0141 loss: 0.6913 +2023/06/01 01:09:41 - mmengine - INFO - Epoch(train) [10][3600/5758] lr: 5.8244e-04 eta: 7:04:20 time: 0.3884 data_time: 0.0018 memory: 20334 grad_norm: 0.0159 loss: 0.6866 +2023/06/01 01:10:21 - mmengine - INFO - Epoch(train) [10][3700/5758] lr: 5.8244e-04 eta: 7:03:35 time: 0.4087 data_time: 0.0018 memory: 20334 grad_norm: 0.0216 loss: 0.6877 +2023/06/01 01:11:01 - mmengine - INFO - Epoch(train) [10][3800/5758] lr: 5.8244e-04 eta: 7:02:49 time: 0.4177 data_time: 0.0016 memory: 20334 grad_norm: 0.0146 loss: 0.6817 +2023/06/01 01:11:41 - mmengine - INFO - Epoch(train) [10][3900/5758] lr: 5.8244e-04 eta: 7:02:04 time: 0.3804 data_time: 0.0019 memory: 20334 grad_norm: 0.0110 loss: 0.6878 +2023/06/01 01:12:21 - mmengine - INFO - Epoch(train) [10][4000/5758] lr: 5.8244e-04 eta: 7:01:18 time: 0.3712 data_time: 0.0016 memory: 20334 grad_norm: 0.0101 loss: 0.6867 +2023/06/01 01:13:02 - mmengine - INFO - Epoch(train) [10][4100/5758] lr: 5.8244e-04 eta: 7:00:34 time: 0.3736 data_time: 0.0019 memory: 20334 grad_norm: 0.0115 loss: 0.6877 +2023/06/01 01:13:32 - mmengine - INFO - Exp name: swin_base_8xb128_fake5m_20230531_183241 +2023/06/01 01:13:41 - mmengine - INFO - Epoch(train) [10][4200/5758] lr: 5.8244e-04 eta: 6:59:48 time: 0.3945 data_time: 0.0018 memory: 20334 grad_norm: 0.0124 loss: 0.6830 +2023/06/01 01:14:21 - mmengine - INFO - Epoch(train) [10][4300/5758] lr: 5.8244e-04 eta: 6:59:02 time: 0.3841 data_time: 0.0020 memory: 20334 grad_norm: 0.0208 loss: 0.6918 +2023/06/01 01:15:00 - mmengine - INFO - Epoch(train) [10][4400/5758] lr: 5.8244e-04 eta: 6:58:16 time: 0.4067 data_time: 0.0017 memory: 20334 grad_norm: 0.0135 loss: 0.6818 +2023/06/01 01:15:39 - mmengine - INFO - Epoch(train) [10][4500/5758] lr: 5.8244e-04 eta: 6:57:30 time: 0.3725 data_time: 0.0019 memory: 20334 grad_norm: 0.0127 loss: 0.6808 +2023/06/01 01:16:20 - mmengine - INFO - Epoch(train) [10][4600/5758] lr: 5.8244e-04 eta: 6:56:45 time: 0.3779 data_time: 0.0015 memory: 20334 grad_norm: 0.0210 loss: 0.6907 +2023/06/01 01:17:01 - mmengine - INFO - Epoch(train) [10][4700/5758] lr: 5.8244e-04 eta: 6:56:02 time: 0.4157 data_time: 0.0017 memory: 20334 grad_norm: 0.0121 loss: 0.6793 +2023/06/01 01:17:40 - mmengine - INFO - Epoch(train) [10][4800/5758] lr: 5.8244e-04 eta: 6:55:15 time: 0.3864 data_time: 0.0017 memory: 20334 grad_norm: 0.0156 loss: 0.6881 +2023/06/01 01:18:20 - mmengine - INFO - Epoch(train) [10][4900/5758] lr: 5.8244e-04 eta: 6:54:30 time: 0.4160 data_time: 0.0017 memory: 20334 grad_norm: 0.0154 loss: 0.6878 +2023/06/01 01:19:00 - mmengine - INFO - Epoch(train) [10][5000/5758] lr: 5.8244e-04 eta: 6:53:44 time: 0.4001 data_time: 0.0017 memory: 20334 grad_norm: 0.0165 loss: 0.6827 +2023/06/01 01:19:40 - mmengine - INFO - Epoch(train) [10][5100/5758] lr: 5.8244e-04 eta: 6:52:59 time: 0.3830 data_time: 0.0016 memory: 20334 grad_norm: 0.0135 loss: 0.6873 +2023/06/01 01:20:12 - mmengine - INFO - Exp name: swin_base_8xb128_fake5m_20230531_183241 +2023/06/01 01:20:21 - mmengine - INFO - Epoch(train) [10][5200/5758] lr: 5.8244e-04 eta: 6:52:16 time: 0.4249 data_time: 0.0016 memory: 20334 grad_norm: 0.0196 loss: 0.6893 +2023/06/01 01:21:01 - mmengine - INFO - Epoch(train) [10][5300/5758] lr: 5.8244e-04 eta: 6:51:31 time: 0.4225 data_time: 0.0016 memory: 20334 grad_norm: 0.0183 loss: 0.6847 +2023/06/01 01:21:42 - mmengine - INFO - Epoch(train) [10][5400/5758] lr: 5.8244e-04 eta: 6:50:46 time: 0.3960 data_time: 0.0016 memory: 20334 grad_norm: 0.0171 loss: 0.6869 +2023/06/01 01:22:22 - mmengine - INFO - Epoch(train) [10][5500/5758] lr: 5.8244e-04 eta: 6:50:01 time: 0.4215 data_time: 0.0019 memory: 20334 grad_norm: 0.0119 loss: 0.6865 +2023/06/01 01:23:01 - mmengine - INFO - Epoch(train) [10][5600/5758] lr: 5.8244e-04 eta: 6:49:15 time: 0.3706 data_time: 0.0018 memory: 20334 grad_norm: 0.0249 loss: 0.6873 +2023/06/01 01:23:41 - mmengine - INFO - Epoch(train) [10][5700/5758] lr: 5.8244e-04 eta: 6:48:30 time: 0.4217 data_time: 0.0019 memory: 20334 grad_norm: 0.0146 loss: 0.6857 +2023/06/01 01:24:04 - mmengine - INFO - Exp name: swin_base_8xb128_fake5m_20230531_183241 +2023/06/01 01:24:04 - mmengine - INFO - Saving checkpoint at 10 epochs +2023/06/01 01:24:21 - mmengine - INFO - Epoch(val) [10][8/8] accuracy/top1: 100.0000 single-label/precision_classwise: [100.0, 0.0] single-label/recall_classwise: [100.0, 0.0] single-label/f1-score_classwise: [100.0, 0.0] data_time: 0.3693 time: 1.0086 +2023/06/01 01:25:05 - mmengine - INFO - Epoch(train) [11][ 100/5758] lr: 5.0500e-04 eta: 6:47:23 time: 0.4090 data_time: 0.0017 memory: 20334 grad_norm: 0.0156 loss: 0.6873 +2023/06/01 01:25:45 - mmengine - INFO - Epoch(train) [11][ 200/5758] lr: 5.0500e-04 eta: 6:46:38 time: 0.3714 data_time: 0.0017 memory: 20334 grad_norm: 0.0134 loss: 0.6865 +2023/06/01 01:26:25 - mmengine - INFO - Epoch(train) [11][ 300/5758] lr: 5.0500e-04 eta: 6:45:53 time: 0.3854 data_time: 0.0017 memory: 20334 grad_norm: 0.0179 loss: 0.6903 +2023/06/01 01:27:06 - mmengine - INFO - Epoch(train) [11][ 400/5758] lr: 5.0500e-04 eta: 6:45:09 time: 0.4526 data_time: 0.0016 memory: 20334 grad_norm: 0.0126 loss: 0.6907 +2023/06/01 01:27:14 - mmengine - INFO - Exp name: swin_base_8xb128_fake5m_20230531_183241 +2023/06/01 01:27:46 - mmengine - INFO - Epoch(train) [11][ 500/5758] lr: 5.0500e-04 eta: 6:44:24 time: 0.4103 data_time: 0.0018 memory: 20334 grad_norm: 0.0132 loss: 0.6839 +2023/06/01 01:28:25 - mmengine - INFO - Epoch(train) [11][ 600/5758] lr: 5.0500e-04 eta: 6:43:38 time: 0.3825 data_time: 0.0016 memory: 20334 grad_norm: 0.0110 loss: 0.6890 +2023/06/01 01:29:05 - mmengine - INFO - Epoch(train) [11][ 700/5758] lr: 5.0500e-04 eta: 6:42:53 time: 0.4215 data_time: 0.0017 memory: 20334 grad_norm: 0.0168 loss: 0.6823 +2023/06/01 01:29:45 - mmengine - INFO - Epoch(train) [11][ 800/5758] lr: 5.0500e-04 eta: 6:42:08 time: 0.3628 data_time: 0.0017 memory: 20334 grad_norm: 0.0158 loss: 0.6853 +2023/06/01 01:30:24 - mmengine - INFO - Epoch(train) [11][ 900/5758] lr: 5.0500e-04 eta: 6:41:23 time: 0.3953 data_time: 0.0017 memory: 20334 grad_norm: 0.0144 loss: 0.6872 +2023/06/01 01:31:04 - mmengine - INFO - Epoch(train) [11][1000/5758] lr: 5.0500e-04 eta: 6:40:37 time: 0.3968 data_time: 0.0016 memory: 20334 grad_norm: 0.0158 loss: 0.6875 +2023/06/01 01:31:43 - mmengine - INFO - Epoch(train) [11][1100/5758] lr: 5.0500e-04 eta: 6:39:51 time: 0.3885 data_time: 0.0017 memory: 20334 grad_norm: 0.0135 loss: 0.6835 +2023/06/01 01:32:23 - mmengine - INFO - Epoch(train) [11][1200/5758] lr: 5.0500e-04 eta: 6:39:07 time: 0.4766 data_time: 0.0016 memory: 20334 grad_norm: 0.0178 loss: 0.6883 +2023/06/01 01:33:04 - mmengine - INFO - Epoch(train) [11][1300/5758] lr: 5.0500e-04 eta: 6:38:22 time: 0.3716 data_time: 0.0018 memory: 20334 grad_norm: 0.0212 loss: 0.6876 +2023/06/01 01:33:45 - mmengine - INFO - Epoch(train) [11][1400/5758] lr: 5.0500e-04 eta: 6:37:39 time: 0.4112 data_time: 0.0020 memory: 20334 grad_norm: 0.0115 loss: 0.6881 +2023/06/01 01:33:53 - mmengine - INFO - Exp name: swin_base_8xb128_fake5m_20230531_183241 +2023/06/01 01:34:26 - mmengine - INFO - Epoch(train) [11][1500/5758] lr: 5.0500e-04 eta: 6:36:55 time: 0.4026 data_time: 0.0020 memory: 20334 grad_norm: 0.0114 loss: 0.6862 +2023/06/01 01:35:05 - mmengine - INFO - Epoch(train) [11][1600/5758] lr: 5.0500e-04 eta: 6:36:09 time: 0.4091 data_time: 0.0018 memory: 20334 grad_norm: 0.0144 loss: 0.6901 +2023/06/01 01:35:45 - mmengine - INFO - Epoch(train) [11][1700/5758] lr: 5.0500e-04 eta: 6:35:25 time: 0.4363 data_time: 0.0016 memory: 20334 grad_norm: 0.0201 loss: 0.6836 +2023/06/01 01:36:26 - mmengine - INFO - Epoch(train) [11][1800/5758] lr: 5.0500e-04 eta: 6:34:41 time: 0.3936 data_time: 0.0017 memory: 20334 grad_norm: 0.0148 loss: 0.6905 +2023/06/01 01:37:07 - mmengine - INFO - Epoch(train) [11][1900/5758] lr: 5.0500e-04 eta: 6:33:56 time: 0.4215 data_time: 0.0016 memory: 20334 grad_norm: 0.0212 loss: 0.6817 +2023/06/01 01:37:46 - mmengine - INFO - Epoch(train) [11][2000/5758] lr: 5.0500e-04 eta: 6:33:11 time: 0.3910 data_time: 0.0015 memory: 20334 grad_norm: 0.0194 loss: 0.6845 +2023/06/01 01:38:26 - mmengine - INFO - Epoch(train) [11][2100/5758] lr: 5.0500e-04 eta: 6:32:27 time: 0.4174 data_time: 0.0015 memory: 20334 grad_norm: 0.0126 loss: 0.6814 +2023/06/01 01:39:06 - mmengine - INFO - Epoch(train) [11][2200/5758] lr: 5.0500e-04 eta: 6:31:42 time: 0.3951 data_time: 0.0015 memory: 20334 grad_norm: 0.0177 loss: 0.6864 +2023/06/01 01:39:46 - mmengine - INFO - Epoch(train) [11][2300/5758] lr: 5.0500e-04 eta: 6:30:57 time: 0.4045 data_time: 0.0015 memory: 20334 grad_norm: 0.0183 loss: 0.6857 +2023/06/01 01:40:27 - mmengine - INFO - Epoch(train) [11][2400/5758] lr: 5.0500e-04 eta: 6:30:13 time: 0.4253 data_time: 0.0013 memory: 20334 grad_norm: 0.0133 loss: 0.6857 +2023/06/01 01:40:36 - mmengine - INFO - Exp name: swin_base_8xb128_fake5m_20230531_183241 +2023/06/01 01:41:08 - mmengine - INFO - Epoch(train) [11][2500/5758] lr: 5.0500e-04 eta: 6:29:29 time: 0.3719 data_time: 0.0014 memory: 20334 grad_norm: 0.0134 loss: 0.6874 +2023/06/01 01:41:48 - mmengine - INFO - Epoch(train) [11][2600/5758] lr: 5.0500e-04 eta: 6:28:45 time: 0.4071 data_time: 0.0013 memory: 20334 grad_norm: 0.0163 loss: 0.6870 +2023/06/01 01:42:29 - mmengine - INFO - Epoch(train) [11][2700/5758] lr: 5.0500e-04 eta: 6:28:01 time: 0.4196 data_time: 0.0014 memory: 20334 grad_norm: 0.0130 loss: 0.6926 +2023/06/01 01:43:09 - mmengine - INFO - Epoch(train) [11][2800/5758] lr: 5.0500e-04 eta: 6:27:16 time: 0.4231 data_time: 0.0015 memory: 20334 grad_norm: 0.0163 loss: 0.6841 +2023/06/01 01:43:48 - mmengine - INFO - Epoch(train) [11][2900/5758] lr: 5.0500e-04 eta: 6:26:31 time: 0.4027 data_time: 0.0013 memory: 20334 grad_norm: 0.0160 loss: 0.6814 +2023/06/01 01:44:28 - mmengine - INFO - Epoch(train) [11][3000/5758] lr: 5.0500e-04 eta: 6:25:47 time: 0.3883 data_time: 0.0014 memory: 20334 grad_norm: 0.0186 loss: 0.6886 +2023/06/01 01:45:09 - mmengine - INFO - Epoch(train) [11][3100/5758] lr: 5.0500e-04 eta: 6:25:02 time: 0.4226 data_time: 0.0019 memory: 20334 grad_norm: 0.0186 loss: 0.6887 +2023/06/01 01:45:49 - mmengine - INFO - Epoch(train) [11][3200/5758] lr: 5.0500e-04 eta: 6:24:18 time: 0.3969 data_time: 0.0016 memory: 20334 grad_norm: 0.0239 loss: 0.6888 +2023/06/01 01:46:29 - mmengine - INFO - Epoch(train) [11][3300/5758] lr: 5.0500e-04 eta: 6:23:33 time: 0.3683 data_time: 0.0017 memory: 20334 grad_norm: 0.0195 loss: 0.6818 +2023/06/01 01:47:10 - mmengine - INFO - Epoch(train) [11][3400/5758] lr: 5.0500e-04 eta: 6:22:49 time: 0.4013 data_time: 0.0015 memory: 20334 grad_norm: 0.0218 loss: 0.6876 +2023/06/01 01:47:18 - mmengine - INFO - Exp name: swin_base_8xb128_fake5m_20230531_183241 +2023/06/01 01:47:49 - mmengine - INFO - Epoch(train) [11][3500/5758] lr: 5.0500e-04 eta: 6:22:05 time: 0.3984 data_time: 0.0017 memory: 20334 grad_norm: 0.0136 loss: 0.6868 +2023/06/01 01:48:32 - mmengine - INFO - Epoch(train) [11][3600/5758] lr: 5.0500e-04 eta: 6:21:22 time: 0.4510 data_time: 0.0019 memory: 20334 grad_norm: 0.0144 loss: 0.6916 +2023/06/01 01:49:13 - mmengine - INFO - Epoch(train) [11][3700/5758] lr: 5.0500e-04 eta: 6:20:39 time: 0.4095 data_time: 0.0019 memory: 20334 grad_norm: 0.0127 loss: 0.6799 +2023/06/01 01:49:53 - mmengine - INFO - Epoch(train) [11][3800/5758] lr: 5.0500e-04 eta: 6:19:55 time: 0.4010 data_time: 0.0017 memory: 20334 grad_norm: 0.0263 loss: 0.6835 +2023/06/01 01:50:34 - mmengine - INFO - Epoch(train) [11][3900/5758] lr: 5.0500e-04 eta: 6:19:11 time: 0.4133 data_time: 0.0018 memory: 20334 grad_norm: 0.0257 loss: 0.6871 +2023/06/01 01:51:14 - mmengine - INFO - Epoch(train) [11][4000/5758] lr: 5.0500e-04 eta: 6:18:27 time: 0.3668 data_time: 0.0019 memory: 20334 grad_norm: 0.0124 loss: 0.6869 +2023/06/01 01:51:56 - mmengine - INFO - Epoch(train) [11][4100/5758] lr: 5.0500e-04 eta: 6:17:44 time: 0.4262 data_time: 0.0015 memory: 20334 grad_norm: 0.0131 loss: 0.6843 +2023/06/01 01:52:36 - mmengine - INFO - Epoch(train) [11][4200/5758] lr: 5.0500e-04 eta: 6:16:59 time: 0.4180 data_time: 0.0018 memory: 20334 grad_norm: 0.0095 loss: 0.6876 +2023/06/01 01:53:17 - mmengine - INFO - Epoch(train) [11][4300/5758] lr: 5.0500e-04 eta: 6:16:16 time: 0.4135 data_time: 0.0015 memory: 20334 grad_norm: 0.0148 loss: 0.6903 +2023/06/01 01:53:58 - mmengine - INFO - Epoch(train) [11][4400/5758] lr: 5.0500e-04 eta: 6:15:32 time: 0.4193 data_time: 0.0018 memory: 20334 grad_norm: 0.0247 loss: 0.6850 +2023/06/01 01:54:06 - mmengine - INFO - Exp name: swin_base_8xb128_fake5m_20230531_183241 +2023/06/01 01:54:39 - mmengine - INFO - Epoch(train) [11][4500/5758] lr: 5.0500e-04 eta: 6:14:48 time: 0.3970 data_time: 0.0018 memory: 20334 grad_norm: 0.0213 loss: 0.6857 +2023/06/01 01:55:19 - mmengine - INFO - Epoch(train) [11][4600/5758] lr: 5.0500e-04 eta: 6:14:04 time: 0.4347 data_time: 0.0017 memory: 20334 grad_norm: 0.0156 loss: 0.6863 +2023/06/01 01:55:58 - mmengine - INFO - Epoch(train) [11][4700/5758] lr: 5.0500e-04 eta: 6:13:19 time: 0.3953 data_time: 0.0018 memory: 20334 grad_norm: 0.0190 loss: 0.6859 +2023/06/01 01:56:39 - mmengine - INFO - Epoch(train) [11][4800/5758] lr: 5.0500e-04 eta: 6:12:35 time: 0.4231 data_time: 0.0019 memory: 20334 grad_norm: 0.0218 loss: 0.6859 +2023/06/01 01:57:20 - mmengine - INFO - Epoch(train) [11][4900/5758] lr: 5.0500e-04 eta: 6:11:52 time: 0.3895 data_time: 0.0018 memory: 20334 grad_norm: 0.0144 loss: 0.6904 +2023/06/01 01:58:00 - mmengine - INFO - Epoch(train) [11][5000/5758] lr: 5.0500e-04 eta: 6:11:07 time: 0.3818 data_time: 0.0027 memory: 20334 grad_norm: 0.0133 loss: 0.6810 +2023/06/01 01:58:40 - mmengine - INFO - Epoch(train) [11][5100/5758] lr: 5.0500e-04 eta: 6:10:24 time: 0.3831 data_time: 0.0017 memory: 20334 grad_norm: 0.0147 loss: 0.6874 +2023/06/01 01:59:22 - mmengine - INFO - Epoch(train) [11][5200/5758] lr: 5.0500e-04 eta: 6:09:41 time: 0.3923 data_time: 0.0023 memory: 20334 grad_norm: 0.0130 loss: 0.6864 +2023/06/01 02:00:03 - mmengine - INFO - Epoch(train) [11][5300/5758] lr: 5.0500e-04 eta: 6:08:57 time: 0.4071 data_time: 0.0016 memory: 20334 grad_norm: 0.0164 loss: 0.6879 +2023/06/01 02:00:43 - mmengine - INFO - Epoch(train) [11][5400/5758] lr: 5.0500e-04 eta: 6:08:13 time: 0.3822 data_time: 0.0017 memory: 20334 grad_norm: 0.0173 loss: 0.6896 +2023/06/01 02:00:51 - mmengine - INFO - Exp name: swin_base_8xb128_fake5m_20230531_183241 +2023/06/01 02:01:23 - mmengine - INFO - Epoch(train) [11][5500/5758] lr: 5.0500e-04 eta: 6:07:29 time: 0.4003 data_time: 0.0018 memory: 20334 grad_norm: 0.0177 loss: 0.6919 +2023/06/01 02:02:04 - mmengine - INFO - Epoch(train) [11][5600/5758] lr: 5.0500e-04 eta: 6:06:45 time: 0.3750 data_time: 0.0019 memory: 20334 grad_norm: 0.0145 loss: 0.6873 +2023/06/01 02:02:45 - mmengine - INFO - Epoch(train) [11][5700/5758] lr: 5.0500e-04 eta: 6:06:01 time: 0.4209 data_time: 0.0016 memory: 20334 grad_norm: 0.0182 loss: 0.6873 +2023/06/01 02:03:08 - mmengine - INFO - Exp name: swin_base_8xb128_fake5m_20230531_183241 +2023/06/01 02:03:08 - mmengine - INFO - Saving checkpoint at 11 epochs +2023/06/01 02:03:25 - mmengine - INFO - Epoch(val) [11][8/8] accuracy/top1: 100.0000 single-label/precision_classwise: [100.0, 0.0] single-label/recall_classwise: [100.0, 0.0] single-label/f1-score_classwise: [100.0, 0.0] data_time: 0.3622 time: 0.9878 +2023/06/01 02:04:11 - mmengine - INFO - Epoch(train) [12][ 100/5758] lr: 4.2756e-04 eta: 6:04:56 time: 0.4174 data_time: 0.0016 memory: 20334 grad_norm: 0.0167 loss: 0.6899 +2023/06/01 02:04:52 - mmengine - INFO - Epoch(train) [12][ 200/5758] lr: 4.2756e-04 eta: 6:04:13 time: 0.3964 data_time: 0.0016 memory: 20334 grad_norm: 0.0208 loss: 0.6836 +2023/06/01 02:05:32 - mmengine - INFO - Epoch(train) [12][ 300/5758] lr: 4.2756e-04 eta: 6:03:28 time: 0.3718 data_time: 0.0018 memory: 20334 grad_norm: 0.0154 loss: 0.6921 +2023/06/01 02:06:12 - mmengine - INFO - Epoch(train) [12][ 400/5758] lr: 4.2756e-04 eta: 6:02:44 time: 0.3979 data_time: 0.0017 memory: 20334 grad_norm: 0.0131 loss: 0.6874 +2023/06/01 02:06:53 - mmengine - INFO - Epoch(train) [12][ 500/5758] lr: 4.2756e-04 eta: 6:02:01 time: 0.3903 data_time: 0.0015 memory: 20334 grad_norm: 0.0158 loss: 0.6895 +2023/06/01 02:07:34 - mmengine - INFO - Epoch(train) [12][ 600/5758] lr: 4.2756e-04 eta: 6:01:18 time: 0.3752 data_time: 0.0017 memory: 20334 grad_norm: 0.0149 loss: 0.6878 +2023/06/01 02:08:00 - mmengine - INFO - Exp name: swin_base_8xb128_fake5m_20230531_183241 +2023/06/01 02:08:15 - mmengine - INFO - Epoch(train) [12][ 700/5758] lr: 4.2756e-04 eta: 6:00:34 time: 0.4368 data_time: 0.0018 memory: 20334 grad_norm: 0.0146 loss: 0.6890 +2023/06/01 02:08:56 - mmengine - INFO - Epoch(train) [12][ 800/5758] lr: 4.2756e-04 eta: 5:59:51 time: 0.4049 data_time: 0.0017 memory: 20334 grad_norm: 0.0140 loss: 0.6842 +2023/06/01 02:09:37 - mmengine - INFO - Epoch(train) [12][ 900/5758] lr: 4.2756e-04 eta: 5:59:07 time: 0.3810 data_time: 0.0029 memory: 20334 grad_norm: 0.0170 loss: 0.6824 +2023/06/01 02:10:17 - mmengine - INFO - Epoch(train) [12][1000/5758] lr: 4.2756e-04 eta: 5:58:23 time: 0.3715 data_time: 0.0018 memory: 20334 grad_norm: 0.0112 loss: 0.6865 +2023/06/01 02:10:58 - mmengine - INFO - Epoch(train) [12][1100/5758] lr: 4.2756e-04 eta: 5:57:40 time: 0.4021 data_time: 0.0017 memory: 20334 grad_norm: 0.0156 loss: 0.6854 +2023/06/01 02:11:41 - mmengine - INFO - Epoch(train) [12][1200/5758] lr: 4.2756e-04 eta: 5:56:58 time: 0.4122 data_time: 0.0023 memory: 20334 grad_norm: 0.0186 loss: 0.6856 +2023/06/01 02:12:23 - mmengine - INFO - Epoch(train) [12][1300/5758] lr: 4.2756e-04 eta: 5:56:15 time: 0.4182 data_time: 0.0014 memory: 20334 grad_norm: 0.0197 loss: 0.6861 +2023/06/01 02:13:04 - mmengine - INFO - Epoch(train) [12][1400/5758] lr: 4.2756e-04 eta: 5:55:32 time: 0.3827 data_time: 0.0015 memory: 20334 grad_norm: 0.0152 loss: 0.6856 +2023/06/01 02:13:45 - mmengine - INFO - Epoch(train) [12][1500/5758] lr: 4.2756e-04 eta: 5:54:49 time: 0.4138 data_time: 0.0015 memory: 20334 grad_norm: 0.0193 loss: 0.6876 +2023/06/01 02:14:25 - mmengine - INFO - Epoch(train) [12][1600/5758] lr: 4.2756e-04 eta: 5:54:05 time: 0.3940 data_time: 0.0020 memory: 20334 grad_norm: 0.0134 loss: 0.6887 +2023/06/01 02:14:51 - mmengine - INFO - Exp name: swin_base_8xb128_fake5m_20230531_183241 +2023/06/01 02:15:07 - mmengine - INFO - Epoch(train) [12][1700/5758] lr: 4.2756e-04 eta: 5:53:22 time: 0.4305 data_time: 0.0019 memory: 20334 grad_norm: 0.0235 loss: 0.6858 +2023/06/01 02:15:47 - mmengine - INFO - Epoch(train) [12][1800/5758] lr: 4.2756e-04 eta: 5:52:38 time: 0.4081 data_time: 0.0018 memory: 20334 grad_norm: 0.0164 loss: 0.6866 +2023/06/01 02:16:28 - mmengine - INFO - Epoch(train) [12][1900/5758] lr: 4.2756e-04 eta: 5:51:55 time: 0.4181 data_time: 0.0020 memory: 20334 grad_norm: 0.0173 loss: 0.6868 +2023/06/01 02:17:10 - mmengine - INFO - Epoch(train) [12][2000/5758] lr: 4.2756e-04 eta: 5:51:12 time: 0.4066 data_time: 0.0016 memory: 20334 grad_norm: 0.0169 loss: 0.6861 +2023/06/01 02:17:51 - mmengine - INFO - Epoch(train) [12][2100/5758] lr: 4.2756e-04 eta: 5:50:29 time: 0.3917 data_time: 0.0016 memory: 20334 grad_norm: 0.0211 loss: 0.6913 +2023/06/01 02:18:32 - mmengine - INFO - Epoch(train) [12][2200/5758] lr: 4.2756e-04 eta: 5:49:45 time: 0.4172 data_time: 0.0017 memory: 20334 grad_norm: 0.0162 loss: 0.6850 +2023/06/01 02:19:12 - mmengine - INFO - Epoch(train) [12][2300/5758] lr: 4.2756e-04 eta: 5:49:02 time: 0.3820 data_time: 0.0019 memory: 20334 grad_norm: 0.0124 loss: 0.6903 +2023/06/01 02:19:53 - mmengine - INFO - Epoch(train) [12][2400/5758] lr: 4.2756e-04 eta: 5:48:19 time: 0.3993 data_time: 0.0019 memory: 20334 grad_norm: 0.0221 loss: 0.6847 +2023/06/01 02:20:34 - mmengine - INFO - Epoch(train) [12][2500/5758] lr: 4.2756e-04 eta: 5:47:35 time: 0.4088 data_time: 0.0020 memory: 20334 grad_norm: 0.0126 loss: 0.6928 +2023/06/01 02:21:14 - mmengine - INFO - Epoch(train) [12][2600/5758] lr: 4.2756e-04 eta: 5:46:51 time: 0.3915 data_time: 0.0018 memory: 20334 grad_norm: 0.0179 loss: 0.6878 +2023/06/01 02:21:39 - mmengine - INFO - Exp name: swin_base_8xb128_fake5m_20230531_183241 +2023/06/01 02:21:55 - mmengine - INFO - Epoch(train) [12][2700/5758] lr: 4.2756e-04 eta: 5:46:08 time: 0.3912 data_time: 0.0021 memory: 20334 grad_norm: 0.0172 loss: 0.6884 +2023/06/01 02:22:37 - mmengine - INFO - Epoch(train) [12][2800/5758] lr: 4.2756e-04 eta: 5:45:25 time: 0.4016 data_time: 0.0018 memory: 20334 grad_norm: 0.0187 loss: 0.6887 +2023/06/01 02:23:18 - mmengine - INFO - Epoch(train) [12][2900/5758] lr: 4.2756e-04 eta: 5:44:42 time: 0.3667 data_time: 0.0017 memory: 20334 grad_norm: 0.0153 loss: 0.6836 +2023/06/01 02:23:59 - mmengine - INFO - Epoch(train) [12][3000/5758] lr: 4.2756e-04 eta: 5:43:58 time: 0.4098 data_time: 0.0017 memory: 20334 grad_norm: 0.0124 loss: 0.6891 +2023/06/01 02:24:38 - mmengine - INFO - Epoch(train) [12][3100/5758] lr: 4.2756e-04 eta: 5:43:14 time: 0.4134 data_time: 0.0026 memory: 20334 grad_norm: 0.0183 loss: 0.6834 +2023/06/01 02:25:19 - mmengine - INFO - Epoch(train) [12][3200/5758] lr: 4.2756e-04 eta: 5:42:31 time: 0.4088 data_time: 0.0018 memory: 20334 grad_norm: 0.0214 loss: 0.6866 +2023/06/01 02:26:00 - mmengine - INFO - Epoch(train) [12][3300/5758] lr: 4.2756e-04 eta: 5:41:48 time: 0.4331 data_time: 0.0018 memory: 20334 grad_norm: 0.0172 loss: 0.6874 +2023/06/01 02:26:41 - mmengine - INFO - Epoch(train) [12][3400/5758] lr: 4.2756e-04 eta: 5:41:05 time: 0.4540 data_time: 0.0016 memory: 20334 grad_norm: 0.0108 loss: 0.6891 +2023/06/01 02:27:23 - mmengine - INFO - Epoch(train) [12][3500/5758] lr: 4.2756e-04 eta: 5:40:22 time: 0.4149 data_time: 0.0018 memory: 20334 grad_norm: 0.0150 loss: 0.6788 +2023/06/01 02:28:03 - mmengine - INFO - Epoch(train) [12][3600/5758] lr: 4.2756e-04 eta: 5:39:38 time: 0.4236 data_time: 0.0017 memory: 20334 grad_norm: 0.0179 loss: 0.6834 +2023/06/01 02:28:30 - mmengine - INFO - Exp name: swin_base_8xb128_fake5m_20230531_183241 +2023/06/01 02:28:46 - mmengine - INFO - Epoch(train) [12][3700/5758] lr: 4.2756e-04 eta: 5:38:56 time: 0.4062 data_time: 0.0017 memory: 20334 grad_norm: 0.0107 loss: 0.6892 +2023/06/01 02:29:26 - mmengine - INFO - Epoch(train) [12][3800/5758] lr: 4.2756e-04 eta: 5:38:13 time: 0.4004 data_time: 0.0018 memory: 20334 grad_norm: 0.0169 loss: 0.6833 +2023/06/01 02:30:07 - mmengine - INFO - Epoch(train) [12][3900/5758] lr: 4.2756e-04 eta: 5:37:29 time: 0.4242 data_time: 0.0019 memory: 20334 grad_norm: 0.0192 loss: 0.6863 +2023/06/01 02:30:48 - mmengine - INFO - Epoch(train) [12][4000/5758] lr: 4.2756e-04 eta: 5:36:46 time: 0.4238 data_time: 0.0017 memory: 20334 grad_norm: 0.0180 loss: 0.6898 +2023/06/01 02:31:30 - mmengine - INFO - Epoch(train) [12][4100/5758] lr: 4.2756e-04 eta: 5:36:04 time: 0.3922 data_time: 0.0018 memory: 20334 grad_norm: 0.0174 loss: 0.6843 +2023/06/01 02:32:11 - mmengine - INFO - Epoch(train) [12][4200/5758] lr: 4.2756e-04 eta: 5:35:21 time: 0.4629 data_time: 0.0018 memory: 20334 grad_norm: 0.0163 loss: 0.6855 +2023/06/01 02:32:52 - mmengine - INFO - Epoch(train) [12][4300/5758] lr: 4.2756e-04 eta: 5:34:38 time: 0.4016 data_time: 0.0020 memory: 20334 grad_norm: 0.0091 loss: 0.6874 +2023/06/01 02:33:34 - mmengine - INFO - Epoch(train) [12][4400/5758] lr: 4.2756e-04 eta: 5:33:55 time: 0.3769 data_time: 0.0018 memory: 20334 grad_norm: 0.0186 loss: 0.6868 +2023/06/01 02:34:14 - mmengine - INFO - Epoch(train) [12][4500/5758] lr: 4.2756e-04 eta: 5:33:11 time: 0.4596 data_time: 0.0017 memory: 20334 grad_norm: 0.0145 loss: 0.6851 +2023/06/01 02:34:54 - mmengine - INFO - Epoch(train) [12][4600/5758] lr: 4.2756e-04 eta: 5:32:27 time: 0.3866 data_time: 0.0019 memory: 20334 grad_norm: 0.0203 loss: 0.6899 +2023/06/01 02:35:19 - mmengine - INFO - Exp name: swin_base_8xb128_fake5m_20230531_183241 +2023/06/01 02:35:34 - mmengine - INFO - Epoch(train) [12][4700/5758] lr: 4.2756e-04 eta: 5:31:44 time: 0.3940 data_time: 0.0017 memory: 20334 grad_norm: 0.0242 loss: 0.6864 +2023/06/01 02:36:15 - mmengine - INFO - Epoch(train) [12][4800/5758] lr: 4.2756e-04 eta: 5:31:01 time: 0.4072 data_time: 0.0018 memory: 20334 grad_norm: 0.0172 loss: 0.6876 +2023/06/01 02:36:57 - mmengine - INFO - Epoch(train) [12][4900/5758] lr: 4.2756e-04 eta: 5:30:18 time: 0.4369 data_time: 0.0021 memory: 20334 grad_norm: 0.0201 loss: 0.6837 +2023/06/01 02:37:37 - mmengine - INFO - Epoch(train) [12][5000/5758] lr: 4.2756e-04 eta: 5:29:35 time: 0.4183 data_time: 0.0017 memory: 20334 grad_norm: 0.0112 loss: 0.6864 +2023/06/01 02:38:19 - mmengine - INFO - Epoch(train) [12][5100/5758] lr: 4.2756e-04 eta: 5:28:52 time: 0.4049 data_time: 0.0019 memory: 20334 grad_norm: 0.0135 loss: 0.6894 +2023/06/01 02:38:59 - mmengine - INFO - Epoch(train) [12][5200/5758] lr: 4.2756e-04 eta: 5:28:09 time: 0.3898 data_time: 0.0018 memory: 20334 grad_norm: 0.0130 loss: 0.6878 +2023/06/01 02:39:39 - mmengine - INFO - Epoch(train) [12][5300/5758] lr: 4.2756e-04 eta: 5:27:25 time: 0.3955 data_time: 0.0018 memory: 20334 grad_norm: 0.0145 loss: 0.6886 +2023/06/01 02:40:20 - mmengine - INFO - Epoch(train) [12][5400/5758] lr: 4.2756e-04 eta: 5:26:41 time: 0.3866 data_time: 0.0016 memory: 20334 grad_norm: 0.0206 loss: 0.6910 +2023/06/01 02:41:00 - mmengine - INFO - Epoch(train) [12][5500/5758] lr: 4.2756e-04 eta: 5:25:58 time: 0.4252 data_time: 0.0020 memory: 20334 grad_norm: 0.0150 loss: 0.6893 +2023/06/01 02:41:41 - mmengine - INFO - Epoch(train) [12][5600/5758] lr: 4.2756e-04 eta: 5:25:15 time: 0.4106 data_time: 0.0020 memory: 20334 grad_norm: 0.0154 loss: 0.6877 +2023/06/01 02:42:07 - mmengine - INFO - Exp name: swin_base_8xb128_fake5m_20230531_183241 +2023/06/01 02:42:21 - mmengine - INFO - Epoch(train) [12][5700/5758] lr: 4.2756e-04 eta: 5:24:31 time: 0.3690 data_time: 0.0016 memory: 20334 grad_norm: 0.0150 loss: 0.6824 +2023/06/01 02:42:45 - mmengine - INFO - Exp name: swin_base_8xb128_fake5m_20230531_183241 +2023/06/01 02:42:45 - mmengine - INFO - Saving checkpoint at 12 epochs +2023/06/01 02:43:03 - mmengine - INFO - Epoch(val) [12][8/8] accuracy/top1: 100.0000 single-label/precision_classwise: [100.0, 0.0] single-label/recall_classwise: [100.0, 0.0] single-label/f1-score_classwise: [100.0, 0.0] data_time: 0.3588 time: 0.9851 +2023/06/01 02:43:49 - mmengine - INFO - Epoch(train) [13][ 100/5758] lr: 3.5204e-04 eta: 5:23:26 time: 0.4269 data_time: 0.0025 memory: 20334 grad_norm: 0.0141 loss: 0.6865 +2023/06/01 02:44:30 - mmengine - INFO - Epoch(train) [13][ 200/5758] lr: 3.5204e-04 eta: 5:22:43 time: 0.4002 data_time: 0.0017 memory: 20334 grad_norm: 0.0101 loss: 0.6878 +2023/06/01 02:45:10 - mmengine - INFO - Epoch(train) [13][ 300/5758] lr: 3.5204e-04 eta: 5:21:59 time: 0.4012 data_time: 0.0014 memory: 20334 grad_norm: 0.0100 loss: 0.6862 +2023/06/01 02:45:49 - mmengine - INFO - Epoch(train) [13][ 400/5758] lr: 3.5204e-04 eta: 5:21:15 time: 0.3791 data_time: 0.0024 memory: 20334 grad_norm: 0.0128 loss: 0.6865 +2023/06/01 02:46:30 - mmengine - INFO - Epoch(train) [13][ 500/5758] lr: 3.5204e-04 eta: 5:20:32 time: 0.4206 data_time: 0.0020 memory: 20334 grad_norm: 0.0171 loss: 0.6865 +2023/06/01 02:47:11 - mmengine - INFO - Epoch(train) [13][ 600/5758] lr: 3.5204e-04 eta: 5:19:49 time: 0.4063 data_time: 0.0023 memory: 20334 grad_norm: 0.0095 loss: 0.6879 +2023/06/01 02:47:51 - mmengine - INFO - Epoch(train) [13][ 700/5758] lr: 3.5204e-04 eta: 5:19:05 time: 0.4161 data_time: 0.0020 memory: 20334 grad_norm: 0.0248 loss: 0.6804 +2023/06/01 02:48:32 - mmengine - INFO - Epoch(train) [13][ 800/5758] lr: 3.5204e-04 eta: 5:18:22 time: 0.3798 data_time: 0.0020 memory: 20334 grad_norm: 0.0139 loss: 0.6857 +2023/06/01 02:49:14 - mmengine - INFO - Epoch(train) [13][ 900/5758] lr: 3.5204e-04 eta: 5:17:40 time: 0.4143 data_time: 0.0019 memory: 20334 grad_norm: 0.0180 loss: 0.6901 +2023/06/01 02:49:15 - mmengine - INFO - Exp name: swin_base_8xb128_fake5m_20230531_183241 +2023/06/01 02:49:53 - mmengine - INFO - Epoch(train) [13][1000/5758] lr: 3.5204e-04 eta: 5:16:56 time: 0.3758 data_time: 0.0020 memory: 20334 grad_norm: 0.0149 loss: 0.6854 +2023/06/01 02:50:34 - mmengine - INFO - Epoch(train) [13][1100/5758] lr: 3.5204e-04 eta: 5:16:13 time: 0.4092 data_time: 0.0018 memory: 20334 grad_norm: 0.0154 loss: 0.6901 +2023/06/01 02:51:15 - mmengine - INFO - Epoch(train) [13][1200/5758] lr: 3.5204e-04 eta: 5:15:30 time: 0.4044 data_time: 0.0018 memory: 20334 grad_norm: 0.0099 loss: 0.6862 +2023/06/01 02:51:56 - mmengine - INFO - Epoch(train) [13][1300/5758] lr: 3.5204e-04 eta: 5:14:47 time: 0.3941 data_time: 0.0017 memory: 20334 grad_norm: 0.0154 loss: 0.6870 +2023/06/01 02:52:37 - mmengine - INFO - Epoch(train) [13][1400/5758] lr: 3.5204e-04 eta: 5:14:04 time: 0.3964 data_time: 0.0024 memory: 20334 grad_norm: 0.0230 loss: 0.6842 +2023/06/01 02:53:17 - mmengine - INFO - Epoch(train) [13][1500/5758] lr: 3.5204e-04 eta: 5:13:21 time: 0.3983 data_time: 0.0016 memory: 20334 grad_norm: 0.0152 loss: 0.6912 +2023/06/01 02:53:58 - mmengine - INFO - Epoch(train) [13][1600/5758] lr: 3.5204e-04 eta: 5:12:38 time: 0.3649 data_time: 0.0017 memory: 20334 grad_norm: 0.0111 loss: 0.6871 +2023/06/01 02:54:40 - mmengine - INFO - Epoch(train) [13][1700/5758] lr: 3.5204e-04 eta: 5:11:55 time: 0.3884 data_time: 0.0018 memory: 20334 grad_norm: 0.0183 loss: 0.6873 +2023/06/01 02:55:21 - mmengine - INFO - Epoch(train) [13][1800/5758] lr: 3.5204e-04 eta: 5:11:12 time: 0.4034 data_time: 0.0017 memory: 20334 grad_norm: 0.0086 loss: 0.6863 +2023/06/01 02:56:03 - mmengine - INFO - Epoch(train) [13][1900/5758] lr: 3.5204e-04 eta: 5:10:30 time: 0.3939 data_time: 0.0017 memory: 20334 grad_norm: 0.0165 loss: 0.6863 +2023/06/01 02:56:04 - mmengine - INFO - Exp name: swin_base_8xb128_fake5m_20230531_183241 +2023/06/01 02:56:44 - mmengine - INFO - Epoch(train) [13][2000/5758] lr: 3.5204e-04 eta: 5:09:47 time: 0.4203 data_time: 0.0017 memory: 20334 grad_norm: 0.0131 loss: 0.6839 +2023/06/01 02:57:25 - mmengine - INFO - Epoch(train) [13][2100/5758] lr: 3.5204e-04 eta: 5:09:04 time: 0.4262 data_time: 0.0021 memory: 20334 grad_norm: 0.0209 loss: 0.6896 +2023/06/01 02:58:06 - mmengine - INFO - Epoch(train) [13][2200/5758] lr: 3.5204e-04 eta: 5:08:21 time: 0.3909 data_time: 0.0019 memory: 20334 grad_norm: 0.0167 loss: 0.6838 +2023/06/01 02:58:46 - mmengine - INFO - Epoch(train) [13][2300/5758] lr: 3.5204e-04 eta: 5:07:38 time: 0.3971 data_time: 0.0020 memory: 20334 grad_norm: 0.0200 loss: 0.6866 +2023/06/01 02:59:27 - mmengine - INFO - Epoch(train) [13][2400/5758] lr: 3.5204e-04 eta: 5:06:54 time: 0.4094 data_time: 0.0019 memory: 20334 grad_norm: 0.0117 loss: 0.6851 +2023/06/01 03:00:08 - mmengine - INFO - Epoch(train) [13][2500/5758] lr: 3.5204e-04 eta: 5:06:12 time: 0.3848 data_time: 0.0025 memory: 20334 grad_norm: 0.0142 loss: 0.6870 +2023/06/01 03:00:48 - mmengine - INFO - Epoch(train) [13][2600/5758] lr: 3.5204e-04 eta: 5:05:28 time: 0.4538 data_time: 0.0018 memory: 20334 grad_norm: 0.0174 loss: 0.6851 +2023/06/01 03:01:28 - mmengine - INFO - Epoch(train) [13][2700/5758] lr: 3.5204e-04 eta: 5:04:45 time: 0.3970 data_time: 0.0018 memory: 20334 grad_norm: 0.0168 loss: 0.6873 +2023/06/01 03:02:08 - mmengine - INFO - Epoch(train) [13][2800/5758] lr: 3.5204e-04 eta: 5:04:01 time: 0.4020 data_time: 0.0014 memory: 20334 grad_norm: 0.0135 loss: 0.6868 +2023/06/01 03:02:49 - mmengine - INFO - Epoch(train) [13][2900/5758] lr: 3.5204e-04 eta: 5:03:19 time: 0.3962 data_time: 0.0013 memory: 20334 grad_norm: 0.0152 loss: 0.6831 +2023/06/01 03:02:51 - mmengine - INFO - Exp name: swin_base_8xb128_fake5m_20230531_183241 +2023/06/01 03:03:30 - mmengine - INFO - Epoch(train) [13][3000/5758] lr: 3.5204e-04 eta: 5:02:36 time: 0.4417 data_time: 0.0021 memory: 20334 grad_norm: 0.0159 loss: 0.6846 +2023/06/01 03:04:11 - mmengine - INFO - Epoch(train) [13][3100/5758] lr: 3.5204e-04 eta: 5:01:53 time: 0.4251 data_time: 0.0020 memory: 20334 grad_norm: 0.0134 loss: 0.6819 +2023/06/01 03:04:52 - mmengine - INFO - Epoch(train) [13][3200/5758] lr: 3.5204e-04 eta: 5:01:10 time: 0.4020 data_time: 0.0019 memory: 20334 grad_norm: 0.0189 loss: 0.6857 +2023/06/01 03:05:32 - mmengine - INFO - Epoch(train) [13][3300/5758] lr: 3.5204e-04 eta: 5:00:27 time: 0.4200 data_time: 0.0016 memory: 20334 grad_norm: 0.0112 loss: 0.6902 +2023/06/01 03:06:12 - mmengine - INFO - Epoch(train) [13][3400/5758] lr: 3.5204e-04 eta: 4:59:43 time: 0.3886 data_time: 0.0016 memory: 20334 grad_norm: 0.0155 loss: 0.6882 +2023/06/01 03:06:52 - mmengine - INFO - Epoch(train) [13][3500/5758] lr: 3.5204e-04 eta: 4:59:00 time: 0.4287 data_time: 0.0019 memory: 20334 grad_norm: 0.0182 loss: 0.6857 +2023/06/01 03:07:33 - mmengine - INFO - Epoch(train) [13][3600/5758] lr: 3.5204e-04 eta: 4:58:17 time: 0.4204 data_time: 0.0018 memory: 20334 grad_norm: 0.0161 loss: 0.6881 +2023/06/01 03:08:14 - mmengine - INFO - Epoch(train) [13][3700/5758] lr: 3.5204e-04 eta: 4:57:34 time: 0.4773 data_time: 0.0016 memory: 20334 grad_norm: 0.0174 loss: 0.6847 +2023/06/01 03:08:55 - mmengine - INFO - Epoch(train) [13][3800/5758] lr: 3.5204e-04 eta: 4:56:51 time: 0.4264 data_time: 0.0019 memory: 20334 grad_norm: 0.0183 loss: 0.6820 +2023/06/01 03:09:36 - mmengine - INFO - Epoch(train) [13][3900/5758] lr: 3.5204e-04 eta: 4:56:08 time: 0.4156 data_time: 0.0018 memory: 20334 grad_norm: 0.0165 loss: 0.6912 +2023/06/01 03:09:37 - mmengine - INFO - Exp name: swin_base_8xb128_fake5m_20230531_183241 +2023/06/01 03:10:16 - mmengine - INFO - Epoch(train) [13][4000/5758] lr: 3.5204e-04 eta: 4:55:25 time: 0.4144 data_time: 0.0017 memory: 20334 grad_norm: 0.0192 loss: 0.6915 +2023/06/01 03:10:58 - mmengine - INFO - Epoch(train) [13][4100/5758] lr: 3.5204e-04 eta: 4:54:43 time: 0.4298 data_time: 0.0016 memory: 20334 grad_norm: 0.0150 loss: 0.6840 +2023/06/01 03:11:38 - mmengine - INFO - Epoch(train) [13][4200/5758] lr: 3.5204e-04 eta: 4:54:00 time: 0.4044 data_time: 0.0017 memory: 20334 grad_norm: 0.0147 loss: 0.6888 +2023/06/01 03:12:19 - mmengine - INFO - Epoch(train) [13][4300/5758] lr: 3.5204e-04 eta: 4:53:16 time: 0.4214 data_time: 0.0016 memory: 20334 grad_norm: 0.0135 loss: 0.6862 +2023/06/01 03:12:59 - mmengine - INFO - Epoch(train) [13][4400/5758] lr: 3.5204e-04 eta: 4:52:33 time: 0.4440 data_time: 0.0018 memory: 20334 grad_norm: 0.0154 loss: 0.6865 +2023/06/01 03:13:40 - mmengine - INFO - Epoch(train) [13][4500/5758] lr: 3.5204e-04 eta: 4:51:50 time: 0.4069 data_time: 0.0016 memory: 20334 grad_norm: 0.0135 loss: 0.6880 +2023/06/01 03:14:20 - mmengine - INFO - Epoch(train) [13][4600/5758] lr: 3.5204e-04 eta: 4:51:07 time: 0.3917 data_time: 0.0018 memory: 20334 grad_norm: 0.0148 loss: 0.6874 +2023/06/01 03:14:59 - mmengine - INFO - Epoch(train) [13][4700/5758] lr: 3.5204e-04 eta: 4:50:23 time: 0.3876 data_time: 0.0019 memory: 20334 grad_norm: 0.0158 loss: 0.6909 +2023/06/01 03:15:40 - mmengine - INFO - Epoch(train) [13][4800/5758] lr: 3.5204e-04 eta: 4:49:40 time: 0.3883 data_time: 0.0017 memory: 20334 grad_norm: 0.0157 loss: 0.6887 +2023/06/01 03:16:21 - mmengine - INFO - Epoch(train) [13][4900/5758] lr: 3.5204e-04 eta: 4:48:58 time: 0.3957 data_time: 0.0019 memory: 20334 grad_norm: 0.0104 loss: 0.6860 +2023/06/01 03:16:23 - mmengine - INFO - Exp name: swin_base_8xb128_fake5m_20230531_183241 +2023/06/01 03:17:03 - mmengine - INFO - Epoch(train) [13][5000/5758] lr: 3.5204e-04 eta: 4:48:16 time: 0.4279 data_time: 0.0016 memory: 20334 grad_norm: 0.0171 loss: 0.6907 +2023/06/01 03:17:44 - mmengine - INFO - Epoch(train) [13][5100/5758] lr: 3.5204e-04 eta: 4:47:33 time: 0.4068 data_time: 0.0018 memory: 20334 grad_norm: 0.0140 loss: 0.6830 +2023/06/01 03:18:25 - mmengine - INFO - Epoch(train) [13][5200/5758] lr: 3.5204e-04 eta: 4:46:50 time: 0.3933 data_time: 0.0017 memory: 20334 grad_norm: 0.0163 loss: 0.6807 +2023/06/01 03:19:07 - mmengine - INFO - Epoch(train) [13][5300/5758] lr: 3.5204e-04 eta: 4:46:08 time: 0.3786 data_time: 0.0017 memory: 20334 grad_norm: 0.0129 loss: 0.6895 +2023/06/01 03:19:48 - mmengine - INFO - Epoch(train) [13][5400/5758] lr: 3.5204e-04 eta: 4:45:25 time: 0.3915 data_time: 0.0019 memory: 20334 grad_norm: 0.0120 loss: 0.6844 +2023/06/01 03:20:28 - mmengine - INFO - Epoch(train) [13][5500/5758] lr: 3.5204e-04 eta: 4:44:42 time: 0.4118 data_time: 0.0019 memory: 20334 grad_norm: 0.0198 loss: 0.6846 +2023/06/01 03:21:10 - mmengine - INFO - Epoch(train) [13][5600/5758] lr: 3.5204e-04 eta: 4:44:00 time: 0.4156 data_time: 0.0020 memory: 20334 grad_norm: 0.0192 loss: 0.6796 +2023/06/01 03:21:50 - mmengine - INFO - Epoch(train) [13][5700/5758] lr: 3.5204e-04 eta: 4:43:17 time: 0.4015 data_time: 0.0025 memory: 20334 grad_norm: 0.0195 loss: 0.6879 +2023/06/01 03:22:13 - mmengine - INFO - Exp name: swin_base_8xb128_fake5m_20230531_183241 +2023/06/01 03:22:13 - mmengine - INFO - Saving checkpoint at 13 epochs +2023/06/01 03:22:30 - mmengine - INFO - Epoch(val) [13][8/8] accuracy/top1: 100.0000 single-label/precision_classwise: [100.0, 0.0] single-label/recall_classwise: [100.0, 0.0] single-label/f1-score_classwise: [100.0, 0.0] data_time: 0.3619 time: 0.9856 +2023/06/01 03:23:15 - mmengine - INFO - Epoch(train) [14][ 100/5758] lr: 2.8027e-04 eta: 4:42:11 time: 0.4051 data_time: 0.0016 memory: 20334 grad_norm: 0.0117 loss: 0.6840 +2023/06/01 03:23:33 - mmengine - INFO - Exp name: swin_base_8xb128_fake5m_20230531_183241 +2023/06/01 03:23:56 - mmengine - INFO - Epoch(train) [14][ 200/5758] lr: 2.8027e-04 eta: 4:41:28 time: 0.4049 data_time: 0.0018 memory: 20334 grad_norm: 0.0189 loss: 0.6888 +2023/06/01 03:24:35 - mmengine - INFO - Epoch(train) [14][ 300/5758] lr: 2.8027e-04 eta: 4:40:45 time: 0.4188 data_time: 0.0016 memory: 20334 grad_norm: 0.0146 loss: 0.6866 +2023/06/01 03:25:16 - mmengine - INFO - Epoch(train) [14][ 400/5758] lr: 2.8027e-04 eta: 4:40:02 time: 0.4036 data_time: 0.0017 memory: 20334 grad_norm: 0.0139 loss: 0.6841 +2023/06/01 03:25:57 - mmengine - INFO - Epoch(train) [14][ 500/5758] lr: 2.8027e-04 eta: 4:39:19 time: 0.4008 data_time: 0.0016 memory: 20334 grad_norm: 0.0176 loss: 0.6901 +2023/06/01 03:26:38 - mmengine - INFO - Epoch(train) [14][ 600/5758] lr: 2.8027e-04 eta: 4:38:37 time: 0.3748 data_time: 0.0018 memory: 20334 grad_norm: 0.0164 loss: 0.6843 +2023/06/01 03:27:18 - mmengine - INFO - Epoch(train) [14][ 700/5758] lr: 2.8027e-04 eta: 4:37:53 time: 0.3973 data_time: 0.0017 memory: 20334 grad_norm: 0.0147 loss: 0.6854 +2023/06/01 03:27:59 - mmengine - INFO - Epoch(train) [14][ 800/5758] lr: 2.8027e-04 eta: 4:37:11 time: 0.3918 data_time: 0.0020 memory: 20334 grad_norm: 0.0120 loss: 0.6881 +2023/06/01 03:28:39 - mmengine - INFO - Epoch(train) [14][ 900/5758] lr: 2.8027e-04 eta: 4:36:27 time: 0.4074 data_time: 0.0015 memory: 20334 grad_norm: 0.0124 loss: 0.6873 +2023/06/01 03:29:20 - mmengine - INFO - Epoch(train) [14][1000/5758] lr: 2.8027e-04 eta: 4:35:45 time: 0.4161 data_time: 0.0017 memory: 20334 grad_norm: 0.0130 loss: 0.6828 +2023/06/01 03:30:00 - mmengine - INFO - Epoch(train) [14][1100/5758] lr: 2.8027e-04 eta: 4:35:02 time: 0.4034 data_time: 0.0017 memory: 20334 grad_norm: 0.0188 loss: 0.6833 +2023/06/01 03:30:20 - mmengine - INFO - Exp name: swin_base_8xb128_fake5m_20230531_183241 +2023/06/01 03:30:42 - mmengine - INFO - Epoch(train) [14][1200/5758] lr: 2.8027e-04 eta: 4:34:19 time: 0.4017 data_time: 0.0015 memory: 20334 grad_norm: 0.0133 loss: 0.6858 +2023/06/01 03:31:23 - mmengine - INFO - Epoch(train) [14][1300/5758] lr: 2.8027e-04 eta: 4:33:37 time: 0.3840 data_time: 0.0017 memory: 20334 grad_norm: 0.0218 loss: 0.6842 +2023/06/01 03:32:03 - mmengine - INFO - Epoch(train) [14][1400/5758] lr: 2.8027e-04 eta: 4:32:53 time: 0.3650 data_time: 0.0025 memory: 20334 grad_norm: 0.0217 loss: 0.6864 +2023/06/01 03:32:43 - mmengine - INFO - Epoch(train) [14][1500/5758] lr: 2.8027e-04 eta: 4:32:10 time: 0.3948 data_time: 0.0022 memory: 20334 grad_norm: 0.0172 loss: 0.6841 +2023/06/01 03:33:23 - mmengine - INFO - Epoch(train) [14][1600/5758] lr: 2.8027e-04 eta: 4:31:27 time: 0.4120 data_time: 0.0018 memory: 20334 grad_norm: 0.0167 loss: 0.6881 +2023/06/01 03:34:04 - mmengine - INFO - Epoch(train) [14][1700/5758] lr: 2.8027e-04 eta: 4:30:45 time: 0.4402 data_time: 0.0019 memory: 20334 grad_norm: 0.0161 loss: 0.6885 +2023/06/01 03:34:44 - mmengine - INFO - Epoch(train) [14][1800/5758] lr: 2.8027e-04 eta: 4:30:01 time: 0.3780 data_time: 0.0028 memory: 20334 grad_norm: 0.0162 loss: 0.6849 +2023/06/01 03:35:25 - mmengine - INFO - Epoch(train) [14][1900/5758] lr: 2.8027e-04 eta: 4:29:19 time: 0.4140 data_time: 0.0017 memory: 20334 grad_norm: 0.0153 loss: 0.6846 +2023/06/01 03:36:05 - mmengine - INFO - Epoch(train) [14][2000/5758] lr: 2.8027e-04 eta: 4:28:36 time: 0.3821 data_time: 0.0019 memory: 20334 grad_norm: 0.0220 loss: 0.6852 +2023/06/01 03:36:46 - mmengine - INFO - Epoch(train) [14][2100/5758] lr: 2.8027e-04 eta: 4:27:53 time: 0.4418 data_time: 0.0032 memory: 20334 grad_norm: 0.0167 loss: 0.6878 +2023/06/01 03:37:05 - mmengine - INFO - Exp name: swin_base_8xb128_fake5m_20230531_183241 +2023/06/01 03:37:26 - mmengine - INFO - Epoch(train) [14][2200/5758] lr: 2.8027e-04 eta: 4:27:10 time: 0.3728 data_time: 0.0019 memory: 20334 grad_norm: 0.0161 loss: 0.6832 +2023/06/01 03:38:07 - mmengine - INFO - Epoch(train) [14][2300/5758] lr: 2.8027e-04 eta: 4:26:27 time: 0.3868 data_time: 0.0021 memory: 20334 grad_norm: 0.0172 loss: 0.6870 +2023/06/01 03:38:47 - mmengine - INFO - Epoch(train) [14][2400/5758] lr: 2.8027e-04 eta: 4:25:45 time: 0.3791 data_time: 0.0019 memory: 20334 grad_norm: 0.0179 loss: 0.6842 +2023/06/01 03:39:28 - mmengine - INFO - Epoch(train) [14][2500/5758] lr: 2.8027e-04 eta: 4:25:02 time: 0.4111 data_time: 0.0023 memory: 20334 grad_norm: 0.0167 loss: 0.6866 +2023/06/01 03:40:09 - mmengine - INFO - Epoch(train) [14][2600/5758] lr: 2.8027e-04 eta: 4:24:19 time: 0.4093 data_time: 0.0028 memory: 20334 grad_norm: 0.0161 loss: 0.6886 +2023/06/01 03:40:49 - mmengine - INFO - Epoch(train) [14][2700/5758] lr: 2.8027e-04 eta: 4:23:36 time: 0.3964 data_time: 0.0019 memory: 20334 grad_norm: 0.0148 loss: 0.6856 +2023/06/01 03:41:30 - mmengine - INFO - Epoch(train) [14][2800/5758] lr: 2.8027e-04 eta: 4:22:54 time: 0.4002 data_time: 0.0022 memory: 20334 grad_norm: 0.0129 loss: 0.6868 +2023/06/01 03:42:10 - mmengine - INFO - Epoch(train) [14][2900/5758] lr: 2.8027e-04 eta: 4:22:11 time: 0.4013 data_time: 0.0016 memory: 20334 grad_norm: 0.0194 loss: 0.6817 +2023/06/01 03:42:51 - mmengine - INFO - Epoch(train) [14][3000/5758] lr: 2.8027e-04 eta: 4:21:28 time: 0.3891 data_time: 0.0030 memory: 20334 grad_norm: 0.0112 loss: 0.6808 +2023/06/01 03:43:30 - mmengine - INFO - Epoch(train) [14][3100/5758] lr: 2.8027e-04 eta: 4:20:45 time: 0.3894 data_time: 0.0022 memory: 20334 grad_norm: 0.0140 loss: 0.6876 +2023/06/01 03:43:50 - mmengine - INFO - Exp name: swin_base_8xb128_fake5m_20230531_183241 +2023/06/01 03:44:12 - mmengine - INFO - Epoch(train) [14][3200/5758] lr: 2.8027e-04 eta: 4:20:02 time: 0.4051 data_time: 0.0018 memory: 20334 grad_norm: 0.0148 loss: 0.6881 +2023/06/01 03:44:53 - mmengine - INFO - Epoch(train) [14][3300/5758] lr: 2.8027e-04 eta: 4:19:20 time: 0.4410 data_time: 0.0026 memory: 20334 grad_norm: 0.0166 loss: 0.6841 +2023/06/01 03:45:33 - mmengine - INFO - Epoch(train) [14][3400/5758] lr: 2.8027e-04 eta: 4:18:37 time: 0.3974 data_time: 0.0017 memory: 20334 grad_norm: 0.0212 loss: 0.6881 +2023/06/01 03:46:13 - mmengine - INFO - Epoch(train) [14][3500/5758] lr: 2.8027e-04 eta: 4:17:54 time: 0.4076 data_time: 0.0025 memory: 20334 grad_norm: 0.0132 loss: 0.6879 +2023/06/01 03:46:53 - mmengine - INFO - Epoch(train) [14][3600/5758] lr: 2.8027e-04 eta: 4:17:11 time: 0.4075 data_time: 0.0023 memory: 20334 grad_norm: 0.0202 loss: 0.6878 +2023/06/01 03:47:34 - mmengine - INFO - Epoch(train) [14][3700/5758] lr: 2.8027e-04 eta: 4:16:28 time: 0.4230 data_time: 0.0018 memory: 20334 grad_norm: 0.0168 loss: 0.6904 +2023/06/01 03:48:14 - mmengine - INFO - Epoch(train) [14][3800/5758] lr: 2.8027e-04 eta: 4:15:45 time: 0.4008 data_time: 0.0020 memory: 20334 grad_norm: 0.0209 loss: 0.6921 +2023/06/01 03:48:56 - mmengine - INFO - Epoch(train) [14][3900/5758] lr: 2.8027e-04 eta: 4:15:03 time: 0.4062 data_time: 0.0026 memory: 20334 grad_norm: 0.0159 loss: 0.6849 +2023/06/01 03:49:37 - mmengine - INFO - Epoch(train) [14][4000/5758] lr: 2.8027e-04 eta: 4:14:20 time: 0.4196 data_time: 0.0024 memory: 20334 grad_norm: 0.0141 loss: 0.6823 +2023/06/01 03:50:19 - mmengine - INFO - Epoch(train) [14][4100/5758] lr: 2.8027e-04 eta: 4:13:38 time: 0.4289 data_time: 0.0028 memory: 20334 grad_norm: 0.0155 loss: 0.6923 +2023/06/01 03:50:37 - mmengine - INFO - Exp name: swin_base_8xb128_fake5m_20230531_183241 +2023/06/01 03:51:00 - mmengine - INFO - Epoch(train) [14][4200/5758] lr: 2.8027e-04 eta: 4:12:56 time: 0.4182 data_time: 0.0023 memory: 20334 grad_norm: 0.0169 loss: 0.6882 +2023/06/01 03:51:41 - mmengine - INFO - Epoch(train) [14][4300/5758] lr: 2.8027e-04 eta: 4:12:14 time: 0.4597 data_time: 0.0022 memory: 20334 grad_norm: 0.0163 loss: 0.6863 +2023/06/01 03:52:21 - mmengine - INFO - Epoch(train) [14][4400/5758] lr: 2.8027e-04 eta: 4:11:30 time: 0.4341 data_time: 0.0023 memory: 20334 grad_norm: 0.0165 loss: 0.6871 +2023/06/01 03:53:02 - mmengine - INFO - Epoch(train) [14][4500/5758] lr: 2.8027e-04 eta: 4:10:48 time: 0.3826 data_time: 0.0024 memory: 20334 grad_norm: 0.0174 loss: 0.6883 +2023/06/01 03:53:42 - mmengine - INFO - Epoch(train) [14][4600/5758] lr: 2.8027e-04 eta: 4:10:05 time: 0.4040 data_time: 0.0028 memory: 20334 grad_norm: 0.0177 loss: 0.6873 +2023/06/01 03:54:23 - mmengine - INFO - Epoch(train) [14][4700/5758] lr: 2.8027e-04 eta: 4:09:23 time: 0.4057 data_time: 0.0029 memory: 20334 grad_norm: 0.0138 loss: 0.6849 +2023/06/01 03:55:02 - mmengine - INFO - Epoch(train) [14][4800/5758] lr: 2.8027e-04 eta: 4:08:39 time: 0.3935 data_time: 0.0018 memory: 20334 grad_norm: 0.0143 loss: 0.6844 +2023/06/01 03:55:43 - mmengine - INFO - Epoch(train) [14][4900/5758] lr: 2.8027e-04 eta: 4:07:57 time: 0.3880 data_time: 0.0022 memory: 20334 grad_norm: 0.0140 loss: 0.6846 +2023/06/01 03:56:24 - mmengine - INFO - Epoch(train) [14][5000/5758] lr: 2.8027e-04 eta: 4:07:14 time: 0.4062 data_time: 0.0018 memory: 20334 grad_norm: 0.0130 loss: 0.6860 +2023/06/01 03:57:04 - mmengine - INFO - Epoch(train) [14][5100/5758] lr: 2.8027e-04 eta: 4:06:31 time: 0.4120 data_time: 0.0016 memory: 20334 grad_norm: 0.0183 loss: 0.6916 +2023/06/01 03:57:23 - mmengine - INFO - Exp name: swin_base_8xb128_fake5m_20230531_183241 +2023/06/01 03:57:44 - mmengine - INFO - Epoch(train) [14][5200/5758] lr: 2.8027e-04 eta: 4:05:49 time: 0.3808 data_time: 0.0028 memory: 20334 grad_norm: 0.0202 loss: 0.6902 +2023/06/01 03:58:25 - mmengine - INFO - Epoch(train) [14][5300/5758] lr: 2.8027e-04 eta: 4:05:06 time: 0.4348 data_time: 0.0020 memory: 20334 grad_norm: 0.0183 loss: 0.6889 +2023/06/01 03:59:05 - mmengine - INFO - Epoch(train) [14][5400/5758] lr: 2.8027e-04 eta: 4:04:23 time: 0.4004 data_time: 0.0017 memory: 20334 grad_norm: 0.0187 loss: 0.6876 +2023/06/01 03:59:47 - mmengine - INFO - Epoch(train) [14][5500/5758] lr: 2.8027e-04 eta: 4:03:41 time: 0.4484 data_time: 0.0020 memory: 20334 grad_norm: 0.0207 loss: 0.6857 +2023/06/01 04:00:27 - mmengine - INFO - Epoch(train) [14][5600/5758] lr: 2.8027e-04 eta: 4:02:58 time: 0.3978 data_time: 0.0018 memory: 20334 grad_norm: 0.0182 loss: 0.6878 +2023/06/01 04:01:06 - mmengine - INFO - Epoch(train) [14][5700/5758] lr: 2.8027e-04 eta: 4:02:15 time: 0.4089 data_time: 0.0019 memory: 20334 grad_norm: 0.0149 loss: 0.6852 +2023/06/01 04:01:30 - mmengine - INFO - Exp name: swin_base_8xb128_fake5m_20230531_183241 +2023/06/01 04:01:30 - mmengine - INFO - Saving checkpoint at 14 epochs +2023/06/01 04:01:47 - mmengine - INFO - Epoch(val) [14][8/8] accuracy/top1: 100.0000 single-label/precision_classwise: [100.0, 0.0] single-label/recall_classwise: [100.0, 0.0] single-label/f1-score_classwise: [100.0, 0.0] data_time: 0.3668 time: 0.9949 +2023/06/01 04:02:32 - mmengine - INFO - Epoch(train) [15][ 100/5758] lr: 2.1405e-04 eta: 4:01:10 time: 0.3993 data_time: 0.0020 memory: 20334 grad_norm: 0.0204 loss: 0.6880 +2023/06/01 04:03:13 - mmengine - INFO - Epoch(train) [15][ 200/5758] lr: 2.1405e-04 eta: 4:00:27 time: 0.4345 data_time: 0.0018 memory: 20334 grad_norm: 0.0132 loss: 0.6877 +2023/06/01 04:03:54 - mmengine - INFO - Epoch(train) [15][ 300/5758] lr: 2.1405e-04 eta: 3:59:44 time: 0.4268 data_time: 0.0021 memory: 20334 grad_norm: 0.0208 loss: 0.6855 +2023/06/01 04:04:30 - mmengine - INFO - Exp name: swin_base_8xb128_fake5m_20230531_183241 +2023/06/01 04:04:34 - mmengine - INFO - Epoch(train) [15][ 400/5758] lr: 2.1405e-04 eta: 3:59:02 time: 0.4004 data_time: 0.0031 memory: 20334 grad_norm: 0.0173 loss: 0.6879 +2023/06/01 04:05:16 - mmengine - INFO - Epoch(train) [15][ 500/5758] lr: 2.1405e-04 eta: 3:58:20 time: 0.3764 data_time: 0.0019 memory: 20334 grad_norm: 0.0120 loss: 0.6882 +2023/06/01 04:05:56 - mmengine - INFO - Epoch(train) [15][ 600/5758] lr: 2.1405e-04 eta: 3:57:37 time: 0.3963 data_time: 0.0019 memory: 20334 grad_norm: 0.0103 loss: 0.6887 +2023/06/01 04:06:37 - mmengine - INFO - Epoch(train) [15][ 700/5758] lr: 2.1405e-04 eta: 3:56:54 time: 0.4071 data_time: 0.0019 memory: 20334 grad_norm: 0.0194 loss: 0.6878 +2023/06/01 04:07:18 - mmengine - INFO - Epoch(train) [15][ 800/5758] lr: 2.1405e-04 eta: 3:56:12 time: 0.4206 data_time: 0.0020 memory: 20334 grad_norm: 0.0173 loss: 0.6901 +2023/06/01 04:08:01 - mmengine - INFO - Epoch(train) [15][ 900/5758] lr: 2.1405e-04 eta: 3:55:30 time: 0.3930 data_time: 0.0017 memory: 20334 grad_norm: 0.0251 loss: 0.6883 +2023/06/01 04:08:41 - mmengine - INFO - Epoch(train) [15][1000/5758] lr: 2.1405e-04 eta: 3:54:48 time: 0.3753 data_time: 0.0016 memory: 20334 grad_norm: 0.0246 loss: 0.6852 +2023/06/01 04:09:22 - mmengine - INFO - Epoch(train) [15][1100/5758] lr: 2.1405e-04 eta: 3:54:05 time: 0.4012 data_time: 0.0021 memory: 20334 grad_norm: 0.0125 loss: 0.6830 +2023/06/01 04:10:03 - mmengine - INFO - Epoch(train) [15][1200/5758] lr: 2.1405e-04 eta: 3:53:23 time: 0.3802 data_time: 0.0017 memory: 20334 grad_norm: 0.0187 loss: 0.6915 +2023/06/01 04:10:44 - mmengine - INFO - Epoch(train) [15][1300/5758] lr: 2.1405e-04 eta: 3:52:41 time: 0.4221 data_time: 0.0018 memory: 20334 grad_norm: 0.0111 loss: 0.6896 +2023/06/01 04:11:20 - mmengine - INFO - Exp name: swin_base_8xb128_fake5m_20230531_183241 +2023/06/01 04:11:25 - mmengine - INFO - Epoch(train) [15][1400/5758] lr: 2.1405e-04 eta: 3:51:58 time: 0.4112 data_time: 0.0020 memory: 20334 grad_norm: 0.0220 loss: 0.6879 +2023/06/01 04:12:05 - mmengine - INFO - Epoch(train) [15][1500/5758] lr: 2.1405e-04 eta: 3:51:15 time: 0.4390 data_time: 0.0026 memory: 20334 grad_norm: 0.0219 loss: 0.6829 +2023/06/01 04:12:45 - mmengine - INFO - Epoch(train) [15][1600/5758] lr: 2.1405e-04 eta: 3:50:33 time: 0.3933 data_time: 0.0015 memory: 20334 grad_norm: 0.7235 loss: 0.6865 +2023/06/01 04:13:26 - mmengine - INFO - Epoch(train) [15][1700/5758] lr: 2.1405e-04 eta: 3:49:50 time: 0.3944 data_time: 0.0015 memory: 20334 grad_norm: 0.0194 loss: 0.6884 +2023/06/01 04:14:08 - mmengine - INFO - Epoch(train) [15][1800/5758] lr: 2.1405e-04 eta: 3:49:08 time: 0.4042 data_time: 0.0019 memory: 20334 grad_norm: 0.0194 loss: 0.6874 +2023/06/01 04:14:47 - mmengine - INFO - Epoch(train) [15][1900/5758] lr: 2.1405e-04 eta: 3:48:25 time: 0.4023 data_time: 0.0018 memory: 20334 grad_norm: 0.0235 loss: 0.6824 +2023/06/01 04:15:28 - mmengine - INFO - Epoch(train) [15][2000/5758] lr: 2.1405e-04 eta: 3:47:43 time: 0.4261 data_time: 0.0028 memory: 20334 grad_norm: 0.0168 loss: 0.6863 +2023/06/01 04:16:09 - mmengine - INFO - Epoch(train) [15][2100/5758] lr: 2.1405e-04 eta: 3:47:00 time: 0.4147 data_time: 0.0019 memory: 20334 grad_norm: 0.0195 loss: 0.6848 +2023/06/01 04:16:49 - mmengine - INFO - Epoch(train) [15][2200/5758] lr: 2.1405e-04 eta: 3:46:17 time: 0.4664 data_time: 0.0023 memory: 20334 grad_norm: 0.0115 loss: 0.6897 +2023/06/01 04:17:29 - mmengine - INFO - Epoch(train) [15][2300/5758] lr: 2.1405e-04 eta: 3:45:35 time: 0.3852 data_time: 0.0019 memory: 20334 grad_norm: 0.0186 loss: 0.6862 +2023/06/01 04:18:05 - mmengine - INFO - Exp name: swin_base_8xb128_fake5m_20230531_183241 +2023/06/01 04:18:10 - mmengine - INFO - Epoch(train) [15][2400/5758] lr: 2.1405e-04 eta: 3:44:53 time: 0.4318 data_time: 0.0021 memory: 20334 grad_norm: 0.0214 loss: 0.6847 +2023/06/01 04:18:50 - mmengine - INFO - Epoch(train) [15][2500/5758] lr: 2.1405e-04 eta: 3:44:10 time: 0.3894 data_time: 0.0020 memory: 20334 grad_norm: 0.0150 loss: 0.6843 +2023/06/01 04:19:31 - mmengine - INFO - Epoch(train) [15][2600/5758] lr: 2.1405e-04 eta: 3:43:27 time: 0.4034 data_time: 0.0023 memory: 20334 grad_norm: 0.0115 loss: 0.6823 +2023/06/01 04:20:11 - mmengine - INFO - Epoch(train) [15][2700/5758] lr: 2.1405e-04 eta: 3:42:45 time: 0.3971 data_time: 0.0018 memory: 20334 grad_norm: 0.0240 loss: 0.6818 +2023/06/01 04:20:52 - mmengine - INFO - Epoch(train) [15][2800/5758] lr: 2.1405e-04 eta: 3:42:02 time: 0.3992 data_time: 0.0021 memory: 20334 grad_norm: 0.0106 loss: 0.6865 +2023/06/01 04:21:32 - mmengine - INFO - Epoch(train) [15][2900/5758] lr: 2.1405e-04 eta: 3:41:20 time: 0.3959 data_time: 0.0019 memory: 20334 grad_norm: 0.0105 loss: 0.6884 +2023/06/01 04:22:14 - mmengine - INFO - Epoch(train) [15][3000/5758] lr: 2.1405e-04 eta: 3:40:38 time: 0.4292 data_time: 0.0017 memory: 20334 grad_norm: 0.0187 loss: 0.6831 +2023/06/01 04:22:55 - mmengine - INFO - Epoch(train) [15][3100/5758] lr: 2.1405e-04 eta: 3:39:55 time: 0.4178 data_time: 0.0017 memory: 20334 grad_norm: 0.0196 loss: 0.6889 +2023/06/01 04:23:35 - mmengine - INFO - Epoch(train) [15][3200/5758] lr: 2.1405e-04 eta: 3:39:13 time: 0.3990 data_time: 0.0018 memory: 20334 grad_norm: 0.0221 loss: 0.6835 +2023/06/01 04:24:17 - mmengine - INFO - Epoch(train) [15][3300/5758] lr: 2.1405e-04 eta: 3:38:30 time: 0.4686 data_time: 0.0017 memory: 20334 grad_norm: 8.7150 loss: 0.6825 +2023/06/01 04:24:53 - mmengine - INFO - Exp name: swin_base_8xb128_fake5m_20230531_183241 +2023/06/01 04:24:58 - mmengine - INFO - Epoch(train) [15][3400/5758] lr: 2.1405e-04 eta: 3:37:48 time: 0.4047 data_time: 0.0016 memory: 20334 grad_norm: 0.0140 loss: 0.6852 +2023/06/01 04:25:39 - mmengine - INFO - Epoch(train) [15][3500/5758] lr: 2.1405e-04 eta: 3:37:06 time: 0.4381 data_time: 0.0018 memory: 20334 grad_norm: 0.0253 loss: 0.6840 +2023/06/01 04:26:19 - mmengine - INFO - Epoch(train) [15][3600/5758] lr: 2.1405e-04 eta: 3:36:23 time: 0.3851 data_time: 0.0017 memory: 20334 grad_norm: 0.0284 loss: 0.6848 +2023/06/01 04:27:00 - mmengine - INFO - Epoch(train) [15][3700/5758] lr: 2.1405e-04 eta: 3:35:41 time: 0.4300 data_time: 0.0018 memory: 20334 grad_norm: 0.0278 loss: 0.6871 +2023/06/01 04:27:41 - mmengine - INFO - Epoch(train) [15][3800/5758] lr: 2.1405e-04 eta: 3:34:59 time: 0.4136 data_time: 0.0021 memory: 20334 grad_norm: 0.0438 loss: 0.6851 +2023/06/01 04:28:21 - mmengine - INFO - Epoch(train) [15][3900/5758] lr: 2.1405e-04 eta: 3:34:16 time: 0.3976 data_time: 0.0018 memory: 20334 grad_norm: 0.0293 loss: 0.6853 +2023/06/01 04:29:03 - mmengine - INFO - Epoch(train) [15][4000/5758] lr: 2.1405e-04 eta: 3:33:34 time: 0.4272 data_time: 0.0030 memory: 20334 grad_norm: 0.0542 loss: 0.6835 +2023/06/01 04:29:47 - mmengine - INFO - Epoch(train) [15][4100/5758] lr: 2.1405e-04 eta: 3:32:53 time: 0.4178 data_time: 0.0025 memory: 20334 grad_norm: 0.1979 loss: 0.6902 +2023/06/01 04:30:27 - mmengine - INFO - Epoch(train) [15][4200/5758] lr: 2.1405e-04 eta: 3:32:10 time: 0.4293 data_time: 0.0020 memory: 20334 grad_norm: 0.0218 loss: 0.6847 +2023/06/01 04:31:07 - mmengine - INFO - Epoch(train) [15][4300/5758] lr: 2.1405e-04 eta: 3:31:28 time: 0.3864 data_time: 0.0019 memory: 20334 grad_norm: 0.1379 loss: 0.6823 +2023/06/01 04:31:42 - mmengine - INFO - Exp name: swin_base_8xb128_fake5m_20230531_183241 +2023/06/01 04:31:47 - mmengine - INFO - Epoch(train) [15][4400/5758] lr: 2.1405e-04 eta: 3:30:45 time: 0.3929 data_time: 0.0020 memory: 20334 grad_norm: 0.0296 loss: 0.6803 +2023/06/01 04:32:28 - mmengine - INFO - Epoch(train) [15][4500/5758] lr: 2.1405e-04 eta: 3:30:02 time: 0.4120 data_time: 0.0018 memory: 20334 grad_norm: 0.1280 loss: 0.6833 +2023/06/01 04:33:08 - mmengine - INFO - Epoch(train) [15][4600/5758] lr: 2.1405e-04 eta: 3:29:20 time: 0.4210 data_time: 0.0020 memory: 20334 grad_norm: 0.0831 loss: 0.6848 +2023/06/01 04:33:48 - mmengine - INFO - Epoch(train) [15][4700/5758] lr: 2.1405e-04 eta: 3:28:37 time: 0.4033 data_time: 0.0015 memory: 20334 grad_norm: 0.0285 loss: 0.6901 +2023/06/01 04:34:30 - mmengine - INFO - Epoch(train) [15][4800/5758] lr: 2.1405e-04 eta: 3:27:55 time: 0.4446 data_time: 0.0015 memory: 20334 grad_norm: 0.0201 loss: 0.6823 +2023/06/01 04:35:10 - mmengine - INFO - Epoch(train) [15][4900/5758] lr: 2.1405e-04 eta: 3:27:13 time: 0.4031 data_time: 0.0019 memory: 20334 grad_norm: 0.0154 loss: 0.6880 +2023/06/01 04:35:50 - mmengine - INFO - Epoch(train) [15][5000/5758] lr: 2.1405e-04 eta: 3:26:30 time: 0.4032 data_time: 0.0016 memory: 20334 grad_norm: 8.6955 loss: 0.6833 +2023/06/01 04:36:30 - mmengine - INFO - Epoch(train) [15][5100/5758] lr: 2.1405e-04 eta: 3:25:48 time: 0.3882 data_time: 0.0020 memory: 20334 grad_norm: 0.0649 loss: 0.6860 +2023/06/01 04:37:11 - mmengine - INFO - Epoch(train) [15][5200/5758] lr: 2.1405e-04 eta: 3:25:05 time: 0.3962 data_time: 0.0024 memory: 20334 grad_norm: 0.0877 loss: 0.6858 +2023/06/01 04:37:52 - mmengine - INFO - Epoch(train) [15][5300/5758] lr: 2.1405e-04 eta: 3:24:23 time: 0.4034 data_time: 0.0021 memory: 20334 grad_norm: 3.8150 loss: 0.6732 +2023/06/01 04:38:28 - mmengine - INFO - Exp name: swin_base_8xb128_fake5m_20230531_183241 +2023/06/01 04:38:32 - mmengine - INFO - Epoch(train) [15][5400/5758] lr: 2.1405e-04 eta: 3:23:41 time: 0.3718 data_time: 0.0022 memory: 20334 grad_norm: 2.4243 loss: 0.6876 +2023/06/01 04:39:15 - mmengine - INFO - Epoch(train) [15][5500/5758] lr: 2.1405e-04 eta: 3:22:59 time: 0.4638 data_time: 0.0023 memory: 20334 grad_norm: 1.6639 loss: 0.6721 +2023/06/01 04:39:57 - mmengine - INFO - Epoch(train) [15][5600/5758] lr: 2.1405e-04 eta: 3:22:17 time: 0.4014 data_time: 0.0020 memory: 20334 grad_norm: 46.6717 loss: 0.6611 +2023/06/01 04:40:36 - mmengine - INFO - Epoch(train) [15][5700/5758] lr: 2.1405e-04 eta: 3:21:34 time: 0.4040 data_time: 0.0022 memory: 20334 grad_norm: 2.5040 loss: 0.6639 +2023/06/01 04:40:59 - mmengine - INFO - Exp name: swin_base_8xb128_fake5m_20230531_183241 +2023/06/01 04:40:59 - mmengine - INFO - Saving checkpoint at 15 epochs +2023/06/01 04:41:16 - mmengine - INFO - Epoch(val) [15][8/8] accuracy/top1: 93.8956 single-label/precision_classwise: [100.0, 0.0] single-label/recall_classwise: [93.89559173583984, 0.0] single-label/f1-score_classwise: [96.8517074584961, 0.0] data_time: 0.3732 time: 0.9995 +2023/06/01 04:42:01 - mmengine - INFO - Epoch(train) [16][ 100/5758] lr: 1.5498e-04 eta: 3:20:29 time: 0.4125 data_time: 0.0017 memory: 20334 grad_norm: 12.9511 loss: 0.6620 +2023/06/01 04:42:42 - mmengine - INFO - Epoch(train) [16][ 200/5758] lr: 1.5498e-04 eta: 3:19:46 time: 0.4402 data_time: 0.0018 memory: 20334 grad_norm: 20.3987 loss: 0.6426 +2023/06/01 04:43:22 - mmengine - INFO - Epoch(train) [16][ 300/5758] lr: 1.5498e-04 eta: 3:19:04 time: 0.3781 data_time: 0.0025 memory: 20334 grad_norm: 166.1285 loss: 0.6846 +2023/06/01 04:44:03 - mmengine - INFO - Epoch(train) [16][ 400/5758] lr: 1.5498e-04 eta: 3:18:22 time: 0.4268 data_time: 0.0031 memory: 20334 grad_norm: 17.4556 loss: 0.6753 +2023/06/01 04:44:43 - mmengine - INFO - Epoch(train) [16][ 500/5758] lr: 1.5498e-04 eta: 3:17:39 time: 0.4338 data_time: 0.0017 memory: 20334 grad_norm: 153.9520 loss: 0.6711 +2023/06/01 04:45:23 - mmengine - INFO - Epoch(train) [16][ 600/5758] lr: 1.5498e-04 eta: 3:16:56 time: 0.4157 data_time: 0.0019 memory: 20334 grad_norm: 16.2066 loss: 0.6689 +2023/06/01 04:45:35 - mmengine - INFO - Exp name: swin_base_8xb128_fake5m_20230531_183241 +2023/06/01 04:46:05 - mmengine - INFO - Epoch(train) [16][ 700/5758] lr: 1.5498e-04 eta: 3:16:14 time: 0.4042 data_time: 0.0025 memory: 20334 grad_norm: 5.0856 loss: 0.6964 +2023/06/01 04:46:45 - mmengine - INFO - Epoch(train) [16][ 800/5758] lr: 1.5498e-04 eta: 3:15:32 time: 0.4223 data_time: 0.0034 memory: 20334 grad_norm: 6.3401 loss: 0.6841 +2023/06/01 04:47:27 - mmengine - INFO - Epoch(train) [16][ 900/5758] lr: 1.5498e-04 eta: 3:14:50 time: 0.3991 data_time: 0.0016 memory: 20334 grad_norm: 0.9693 loss: 0.6880 +2023/06/01 04:48:07 - mmengine - INFO - Epoch(train) [16][1000/5758] lr: 1.5498e-04 eta: 3:14:08 time: 0.4078 data_time: 0.0020 memory: 20334 grad_norm: 1.2147 loss: 0.6861 +2023/06/01 04:48:49 - mmengine - INFO - Epoch(train) [16][1100/5758] lr: 1.5498e-04 eta: 3:13:26 time: 0.4397 data_time: 0.0017 memory: 20334 grad_norm: 0.8336 loss: 0.6850 +2023/06/01 04:49:30 - mmengine - INFO - Epoch(train) [16][1200/5758] lr: 1.5498e-04 eta: 3:12:44 time: 0.4458 data_time: 0.0019 memory: 20334 grad_norm: 1.7692 loss: 0.6863 +2023/06/01 04:50:11 - mmengine - INFO - Epoch(train) [16][1300/5758] lr: 1.5498e-04 eta: 3:12:01 time: 0.4181 data_time: 0.0022 memory: 20334 grad_norm: 1.9187 loss: 0.6828 +2023/06/01 04:50:51 - mmengine - INFO - Epoch(train) [16][1400/5758] lr: 1.5498e-04 eta: 3:11:19 time: 0.3825 data_time: 0.0031 memory: 20334 grad_norm: 10.5395 loss: 0.6742 +2023/06/01 04:51:32 - mmengine - INFO - Epoch(train) [16][1500/5758] lr: 1.5498e-04 eta: 3:10:36 time: 0.4437 data_time: 0.0026 memory: 20334 grad_norm: 0.7334 loss: 0.6746 +2023/06/01 04:52:14 - mmengine - INFO - Epoch(train) [16][1600/5758] lr: 1.5498e-04 eta: 3:09:55 time: 0.4179 data_time: 0.0022 memory: 20334 grad_norm: 0.8798 loss: 0.6776 +2023/06/01 04:52:26 - mmengine - INFO - Exp name: swin_base_8xb128_fake5m_20230531_183241 +2023/06/01 04:52:55 - mmengine - INFO - Epoch(train) [16][1700/5758] lr: 1.5498e-04 eta: 3:09:13 time: 0.4212 data_time: 0.0036 memory: 20334 grad_norm: 3.3916 loss: 0.6739 +2023/06/01 04:53:34 - mmengine - INFO - Epoch(train) [16][1800/5758] lr: 1.5498e-04 eta: 3:08:30 time: 0.3852 data_time: 0.0019 memory: 20334 grad_norm: 0.2993 loss: 0.6853 +2023/06/01 04:54:14 - mmengine - INFO - Epoch(train) [16][1900/5758] lr: 1.5498e-04 eta: 3:07:47 time: 0.4051 data_time: 0.0019 memory: 20334 grad_norm: 0.3990 loss: 0.6886 +2023/06/01 04:54:55 - mmengine - INFO - Epoch(train) [16][2000/5758] lr: 1.5498e-04 eta: 3:07:05 time: 0.4105 data_time: 0.0028 memory: 20334 grad_norm: 0.1702 loss: 0.6884 +2023/06/01 04:55:36 - mmengine - INFO - Epoch(train) [16][2100/5758] lr: 1.5498e-04 eta: 3:06:23 time: 0.4038 data_time: 0.0029 memory: 20334 grad_norm: 0.4629 loss: 0.6927 +2023/06/01 04:56:17 - mmengine - INFO - Epoch(train) [16][2200/5758] lr: 1.5498e-04 eta: 3:05:41 time: 0.4490 data_time: 0.0022 memory: 20334 grad_norm: 17.8738 loss: 0.6876 +2023/06/01 04:56:59 - mmengine - INFO - Epoch(train) [16][2300/5758] lr: 1.5498e-04 eta: 3:04:59 time: 0.4195 data_time: 0.0023 memory: 20334 grad_norm: 11.7259 loss: 0.6912 +2023/06/01 04:57:40 - mmengine - INFO - Epoch(train) [16][2400/5758] lr: 1.5498e-04 eta: 3:04:16 time: 0.4291 data_time: 0.0018 memory: 20334 grad_norm: 1.5582 loss: 0.6845 +2023/06/01 04:58:20 - mmengine - INFO - Epoch(train) [16][2500/5758] lr: 1.5498e-04 eta: 3:03:34 time: 0.3938 data_time: 0.0028 memory: 20334 grad_norm: 4.2360 loss: 0.6870 +2023/06/01 04:59:01 - mmengine - INFO - Epoch(train) [16][2600/5758] lr: 1.5498e-04 eta: 3:02:52 time: 0.4125 data_time: 0.0018 memory: 20334 grad_norm: 0.7279 loss: 0.6894 +2023/06/01 04:59:12 - mmengine - INFO - Exp name: swin_base_8xb128_fake5m_20230531_183241 +2023/06/01 04:59:40 - mmengine - INFO - Epoch(train) [16][2700/5758] lr: 1.5498e-04 eta: 3:02:09 time: 0.4068 data_time: 0.0019 memory: 20334 grad_norm: 0.4202 loss: 0.6843 +2023/06/01 05:00:20 - mmengine - INFO - Epoch(train) [16][2800/5758] lr: 1.5498e-04 eta: 3:01:27 time: 0.4023 data_time: 0.0018 memory: 20334 grad_norm: 0.1388 loss: 0.6857 +2023/06/01 05:01:01 - mmengine - INFO - Epoch(train) [16][2900/5758] lr: 1.5498e-04 eta: 3:00:45 time: 0.4056 data_time: 0.0019 memory: 20334 grad_norm: 0.0492 loss: 0.6896 +2023/06/01 05:01:42 - mmengine - INFO - Epoch(train) [16][3000/5758] lr: 1.5498e-04 eta: 3:00:02 time: 0.3845 data_time: 0.0018 memory: 20334 grad_norm: 2.3024 loss: 0.6834 +2023/06/01 05:02:22 - mmengine - INFO - Epoch(train) [16][3100/5758] lr: 1.5498e-04 eta: 2:59:20 time: 0.3881 data_time: 0.0021 memory: 20334 grad_norm: 0.8119 loss: 0.6922 +2023/06/01 05:03:03 - mmengine - INFO - Epoch(train) [16][3200/5758] lr: 1.5498e-04 eta: 2:58:38 time: 0.4399 data_time: 0.0018 memory: 20334 grad_norm: 0.2433 loss: 0.6882 +2023/06/01 05:03:43 - mmengine - INFO - Epoch(train) [16][3300/5758] lr: 1.5498e-04 eta: 2:57:56 time: 0.4182 data_time: 0.0018 memory: 20334 grad_norm: 0.3354 loss: 0.6832 +2023/06/01 05:04:24 - mmengine - INFO - Epoch(train) [16][3400/5758] lr: 1.5498e-04 eta: 2:57:13 time: 0.4243 data_time: 0.0017 memory: 20334 grad_norm: 0.2514 loss: 0.6861 +2023/06/01 05:05:04 - mmengine - INFO - Epoch(train) [16][3500/5758] lr: 1.5498e-04 eta: 2:56:31 time: 0.4271 data_time: 0.0015 memory: 20334 grad_norm: 0.6776 loss: 0.6859 +2023/06/01 05:05:45 - mmengine - INFO - Epoch(train) [16][3600/5758] lr: 1.5498e-04 eta: 2:55:49 time: 0.3950 data_time: 0.0016 memory: 20334 grad_norm: 0.6123 loss: 0.6826 +2023/06/01 05:05:57 - mmengine - INFO - Exp name: swin_base_8xb128_fake5m_20230531_183241 +2023/06/01 05:06:27 - mmengine - INFO - Epoch(train) [16][3700/5758] lr: 1.5498e-04 eta: 2:55:07 time: 0.4651 data_time: 0.0020 memory: 20334 grad_norm: 0.2018 loss: 0.6864 +2023/06/01 05:07:08 - mmengine - INFO - Epoch(train) [16][3800/5758] lr: 1.5498e-04 eta: 2:54:25 time: 0.4244 data_time: 0.0019 memory: 20334 grad_norm: 0.2864 loss: 0.6877 +2023/06/01 05:07:49 - mmengine - INFO - Epoch(train) [16][3900/5758] lr: 1.5498e-04 eta: 2:53:43 time: 0.4155 data_time: 0.0017 memory: 20334 grad_norm: 0.1539 loss: 0.6884 +2023/06/01 05:08:30 - mmengine - INFO - Epoch(train) [16][4000/5758] lr: 1.5498e-04 eta: 2:53:01 time: 0.4340 data_time: 0.0017 memory: 20334 grad_norm: 0.1950 loss: 0.6865 +2023/06/01 05:09:11 - mmengine - INFO - Epoch(train) [16][4100/5758] lr: 1.5498e-04 eta: 2:52:18 time: 0.4267 data_time: 0.0016 memory: 20334 grad_norm: 0.5413 loss: 0.6927 +2023/06/01 05:09:51 - mmengine - INFO - Epoch(train) [16][4200/5758] lr: 1.5498e-04 eta: 2:51:36 time: 0.3973 data_time: 0.0017 memory: 20334 grad_norm: 0.7295 loss: 0.6888 +2023/06/01 05:10:32 - mmengine - INFO - Epoch(train) [16][4300/5758] lr: 1.5498e-04 eta: 2:50:54 time: 0.3891 data_time: 0.0019 memory: 20334 grad_norm: 1.2996 loss: 0.6843 +2023/06/01 05:11:13 - mmengine - INFO - Epoch(train) [16][4400/5758] lr: 1.5498e-04 eta: 2:50:12 time: 0.3753 data_time: 0.0016 memory: 20334 grad_norm: 0.5797 loss: 0.6847 +2023/06/01 05:11:52 - mmengine - INFO - Epoch(train) [16][4500/5758] lr: 1.5498e-04 eta: 2:49:29 time: 0.3714 data_time: 0.0016 memory: 20334 grad_norm: 4.0275 loss: 0.6887 +2023/06/01 05:12:34 - mmengine - INFO - Epoch(train) [16][4600/5758] lr: 1.5498e-04 eta: 2:48:47 time: 0.4044 data_time: 0.0016 memory: 20334 grad_norm: 2.0650 loss: 0.6807 +2023/06/01 05:12:46 - mmengine - INFO - Exp name: swin_base_8xb128_fake5m_20230531_183241 +2023/06/01 05:13:14 - mmengine - INFO - Epoch(train) [16][4700/5758] lr: 1.5498e-04 eta: 2:48:05 time: 0.3974 data_time: 0.0016 memory: 20334 grad_norm: 2.3120 loss: 0.6872 +2023/06/01 05:13:56 - mmengine - INFO - Epoch(train) [16][4800/5758] lr: 1.5498e-04 eta: 2:47:23 time: 0.4134 data_time: 0.0016 memory: 20334 grad_norm: 6.7637 loss: 0.6780 +2023/06/01 05:14:38 - mmengine - INFO - Epoch(train) [16][4900/5758] lr: 1.5498e-04 eta: 2:46:41 time: 0.4374 data_time: 0.0017 memory: 20334 grad_norm: 16.9621 loss: 0.6605 +2023/06/01 05:15:18 - mmengine - INFO - Epoch(train) [16][5000/5758] lr: 1.5498e-04 eta: 2:45:59 time: 0.3812 data_time: 0.0016 memory: 20334 grad_norm: 7.3323 loss: 0.6797 +2023/06/01 05:15:58 - mmengine - INFO - Epoch(train) [16][5100/5758] lr: 1.5498e-04 eta: 2:45:17 time: 0.3892 data_time: 0.0017 memory: 20334 grad_norm: 18.0871 loss: 0.6598 +2023/06/01 05:16:39 - mmengine - INFO - Epoch(train) [16][5200/5758] lr: 1.5498e-04 eta: 2:44:35 time: 0.4083 data_time: 0.0015 memory: 20334 grad_norm: 422.3003 loss: 0.6643 +2023/06/01 05:17:20 - mmengine - INFO - Epoch(train) [16][5300/5758] lr: 1.5498e-04 eta: 2:43:52 time: 0.3825 data_time: 0.0017 memory: 20334 grad_norm: 293.7801 loss: 0.6675 +2023/06/01 05:18:00 - mmengine - INFO - Epoch(train) [16][5400/5758] lr: 1.5498e-04 eta: 2:43:10 time: 0.3898 data_time: 0.0016 memory: 20334 grad_norm: 63.2483 loss: 0.6756 +2023/06/01 05:18:40 - mmengine - INFO - Epoch(train) [16][5500/5758] lr: 1.5498e-04 eta: 2:42:28 time: 0.4151 data_time: 0.0018 memory: 20334 grad_norm: 40.1539 loss: 0.6775 +2023/06/01 05:19:21 - mmengine - INFO - Epoch(train) [16][5600/5758] lr: 1.5498e-04 eta: 2:41:46 time: 0.3952 data_time: 0.0020 memory: 20334 grad_norm: 77.3026 loss: 0.6732 +2023/06/01 05:19:33 - mmengine - INFO - Exp name: swin_base_8xb128_fake5m_20230531_183241 +2023/06/01 05:20:01 - mmengine - INFO - Epoch(train) [16][5700/5758] lr: 1.5498e-04 eta: 2:41:03 time: 0.3873 data_time: 0.0017 memory: 20334 grad_norm: 126.4350 loss: 0.6708 +2023/06/01 05:20:25 - mmengine - INFO - Exp name: swin_base_8xb128_fake5m_20230531_183241 +2023/06/01 05:20:25 - mmengine - INFO - Saving checkpoint at 16 epochs +2023/06/01 05:20:41 - mmengine - INFO - Epoch(val) [16][8/8] accuracy/top1: 98.0999 single-label/precision_classwise: [100.0, 0.0] single-label/recall_classwise: [98.09986877441406, 0.0] single-label/f1-score_classwise: [99.04082489013672, 0.0] data_time: 0.3740 time: 1.0006 +2023/06/01 05:21:25 - mmengine - INFO - Epoch(train) [17][ 100/5758] lr: 1.0454e-04 eta: 2:39:58 time: 0.4126 data_time: 0.0017 memory: 20334 grad_norm: 3.2546 loss: 0.6642 +2023/06/01 05:22:06 - mmengine - INFO - Epoch(train) [17][ 200/5758] lr: 1.0454e-04 eta: 2:39:16 time: 0.4131 data_time: 0.0016 memory: 20334 grad_norm: 9.8711 loss: 0.6585 +2023/06/01 05:22:48 - mmengine - INFO - Epoch(train) [17][ 300/5758] lr: 1.0454e-04 eta: 2:38:34 time: 0.4243 data_time: 0.0016 memory: 20334 grad_norm: 16.2878 loss: 0.6528 +2023/06/01 05:23:29 - mmengine - INFO - Epoch(train) [17][ 400/5758] lr: 1.0454e-04 eta: 2:37:52 time: 0.4310 data_time: 0.0016 memory: 20334 grad_norm: 15.2030 loss: 0.6579 +2023/06/01 05:24:10 - mmengine - INFO - Epoch(train) [17][ 500/5758] lr: 1.0454e-04 eta: 2:37:09 time: 0.4290 data_time: 0.0016 memory: 20334 grad_norm: 11.4554 loss: 0.6541 +2023/06/01 05:24:50 - mmengine - INFO - Epoch(train) [17][ 600/5758] lr: 1.0454e-04 eta: 2:36:27 time: 0.4141 data_time: 0.0016 memory: 20334 grad_norm: 12.4674 loss: 0.6616 +2023/06/01 05:25:31 - mmengine - INFO - Epoch(train) [17][ 700/5758] lr: 1.0454e-04 eta: 2:35:45 time: 0.4093 data_time: 0.0015 memory: 20334 grad_norm: 15.2415 loss: 0.6654 +2023/06/01 05:26:12 - mmengine - INFO - Epoch(train) [17][ 800/5758] lr: 1.0454e-04 eta: 2:35:03 time: 0.4104 data_time: 0.0017 memory: 20334 grad_norm: 6.9286 loss: 0.6621 +2023/06/01 05:26:41 - mmengine - INFO - Exp name: swin_base_8xb128_fake5m_20230531_183241 +2023/06/01 05:26:53 - mmengine - INFO - Epoch(train) [17][ 900/5758] lr: 1.0454e-04 eta: 2:34:21 time: 0.4183 data_time: 0.0016 memory: 20334 grad_norm: 11.7854 loss: 0.6657 +2023/06/01 05:27:33 - mmengine - INFO - Epoch(train) [17][1000/5758] lr: 1.0454e-04 eta: 2:33:39 time: 0.4516 data_time: 0.0019 memory: 20334 grad_norm: 19.8003 loss: 0.6556 +2023/06/01 05:28:13 - mmengine - INFO - Epoch(train) [17][1100/5758] lr: 1.0454e-04 eta: 2:32:57 time: 0.3782 data_time: 0.0018 memory: 20334 grad_norm: 13.8714 loss: 0.6472 +2023/06/01 05:28:54 - mmengine - INFO - Epoch(train) [17][1200/5758] lr: 1.0454e-04 eta: 2:32:15 time: 0.4360 data_time: 0.0033 memory: 20334 grad_norm: 22.1152 loss: 0.6549 +2023/06/01 05:29:35 - mmengine - INFO - Epoch(train) [17][1300/5758] lr: 1.0454e-04 eta: 2:31:32 time: 0.4103 data_time: 0.0019 memory: 20334 grad_norm: 5.6001 loss: 0.6542 +2023/06/01 05:30:16 - mmengine - INFO - Epoch(train) [17][1400/5758] lr: 1.0454e-04 eta: 2:30:50 time: 0.4596 data_time: 0.0024 memory: 20334 grad_norm: 11.6462 loss: 0.6577 +2023/06/01 05:30:56 - mmengine - INFO - Epoch(train) [17][1500/5758] lr: 1.0454e-04 eta: 2:30:08 time: 0.3715 data_time: 0.0031 memory: 20334 grad_norm: 129.4207 loss: 0.6680 +2023/06/01 05:31:36 - mmengine - INFO - Epoch(train) [17][1600/5758] lr: 1.0454e-04 eta: 2:29:26 time: 0.3841 data_time: 0.0024 memory: 20334 grad_norm: 32.7622 loss: 0.6641 +2023/06/01 05:32:16 - mmengine - INFO - Epoch(train) [17][1700/5758] lr: 1.0454e-04 eta: 2:28:44 time: 0.3952 data_time: 0.0017 memory: 20334 grad_norm: 4.8323 loss: 0.6561 +2023/06/01 05:32:56 - mmengine - INFO - Epoch(train) [17][1800/5758] lr: 1.0454e-04 eta: 2:28:01 time: 0.4079 data_time: 0.0021 memory: 20334 grad_norm: 3108.9795 loss: 0.6573 +2023/06/01 05:33:26 - mmengine - INFO - Exp name: swin_base_8xb128_fake5m_20230531_183241 +2023/06/01 05:33:38 - mmengine - INFO - Epoch(train) [17][1900/5758] lr: 1.0454e-04 eta: 2:27:20 time: 0.4417 data_time: 0.0018 memory: 20334 grad_norm: 7.6875 loss: 0.6517 +2023/06/01 05:34:18 - mmengine - INFO - Epoch(train) [17][2000/5758] lr: 1.0454e-04 eta: 2:26:37 time: 0.4202 data_time: 0.0017 memory: 20334 grad_norm: 86.2196 loss: 0.6473 +2023/06/01 05:34:59 - mmengine - INFO - Epoch(train) [17][2100/5758] lr: 1.0454e-04 eta: 2:25:55 time: 0.4056 data_time: 0.0027 memory: 20334 grad_norm: 60.8420 loss: 0.6557 +2023/06/01 05:35:40 - mmengine - INFO - Epoch(train) [17][2200/5758] lr: 1.0454e-04 eta: 2:25:13 time: 0.3857 data_time: 0.0016 memory: 20334 grad_norm: 19.7107 loss: 0.6611 +2023/06/01 05:36:20 - mmengine - INFO - Epoch(train) [17][2300/5758] lr: 1.0454e-04 eta: 2:24:31 time: 0.4219 data_time: 0.0028 memory: 20334 grad_norm: 8.5472 loss: 0.6529 +2023/06/01 05:37:01 - mmengine - INFO - Epoch(train) [17][2400/5758] lr: 1.0454e-04 eta: 2:23:49 time: 0.4285 data_time: 0.0025 memory: 20334 grad_norm: 21.4679 loss: 0.6664 +2023/06/01 05:37:41 - mmengine - INFO - Epoch(train) [17][2500/5758] lr: 1.0454e-04 eta: 2:23:07 time: 0.4013 data_time: 0.0024 memory: 20334 grad_norm: 29.7841 loss: 0.6476 +2023/06/01 05:38:23 - mmengine - INFO - Epoch(train) [17][2600/5758] lr: 1.0454e-04 eta: 2:22:25 time: 0.4078 data_time: 0.0025 memory: 20334 grad_norm: 49.7052 loss: 0.6484 +2023/06/01 05:39:05 - mmengine - INFO - Epoch(train) [17][2700/5758] lr: 1.0454e-04 eta: 2:21:43 time: 0.3936 data_time: 0.0023 memory: 20334 grad_norm: 8.6435 loss: 0.6472 +2023/06/01 05:39:45 - mmengine - INFO - Epoch(train) [17][2800/5758] lr: 1.0454e-04 eta: 2:21:01 time: 0.4298 data_time: 0.0020 memory: 20334 grad_norm: 51.1484 loss: 0.6423 +2023/06/01 05:40:14 - mmengine - INFO - Exp name: swin_base_8xb128_fake5m_20230531_183241 +2023/06/01 05:40:25 - mmengine - INFO - Epoch(train) [17][2900/5758] lr: 1.0454e-04 eta: 2:20:19 time: 0.3986 data_time: 0.0033 memory: 20334 grad_norm: 19.7588 loss: 0.6388 +2023/06/01 05:41:05 - mmengine - INFO - Epoch(train) [17][3000/5758] lr: 1.0454e-04 eta: 2:19:37 time: 0.4048 data_time: 0.0017 memory: 20334 grad_norm: 4.2124 loss: 0.6483 +2023/06/01 05:41:45 - mmengine - INFO - Epoch(train) [17][3100/5758] lr: 1.0454e-04 eta: 2:18:54 time: 0.3725 data_time: 0.0019 memory: 20334 grad_norm: 3.8446 loss: 0.6351 +2023/06/01 05:42:26 - mmengine - INFO - Epoch(train) [17][3200/5758] lr: 1.0454e-04 eta: 2:18:12 time: 0.4294 data_time: 0.0017 memory: 20334 grad_norm: 18.1920 loss: 0.6391 +2023/06/01 05:43:06 - mmengine - INFO - Epoch(train) [17][3300/5758] lr: 1.0454e-04 eta: 2:17:30 time: 0.4056 data_time: 0.0030 memory: 20334 grad_norm: 9.1583 loss: 0.6451 +2023/06/01 05:43:47 - mmengine - INFO - Epoch(train) [17][3400/5758] lr: 1.0454e-04 eta: 2:16:48 time: 0.3879 data_time: 0.0024 memory: 20334 grad_norm: 41.2997 loss: 0.6426 +2023/06/01 05:44:27 - mmengine - INFO - Epoch(train) [17][3500/5758] lr: 1.0454e-04 eta: 2:16:06 time: 0.3872 data_time: 0.0018 memory: 20334 grad_norm: 37.7454 loss: 0.6478 +2023/06/01 05:45:07 - mmengine - INFO - Epoch(train) [17][3600/5758] lr: 1.0454e-04 eta: 2:15:24 time: 0.3866 data_time: 0.0028 memory: 20334 grad_norm: 18.1691 loss: 0.6497 +2023/06/01 05:45:47 - mmengine - INFO - Epoch(train) [17][3700/5758] lr: 1.0454e-04 eta: 2:14:42 time: 0.4066 data_time: 0.0017 memory: 20334 grad_norm: 63.3308 loss: 0.6567 +2023/06/01 05:46:27 - mmengine - INFO - Epoch(train) [17][3800/5758] lr: 1.0454e-04 eta: 2:14:00 time: 0.3708 data_time: 0.0019 memory: 20334 grad_norm: 37.7261 loss: 0.6425 +2023/06/01 05:46:57 - mmengine - INFO - Exp name: swin_base_8xb128_fake5m_20230531_183241 +2023/06/01 05:47:07 - mmengine - INFO - Epoch(train) [17][3900/5758] lr: 1.0454e-04 eta: 2:13:17 time: 0.3940 data_time: 0.0021 memory: 20334 grad_norm: 221.2627 loss: 0.6479 +2023/06/01 05:47:48 - mmengine - INFO - Epoch(train) [17][4000/5758] lr: 1.0454e-04 eta: 2:12:35 time: 0.3877 data_time: 0.0024 memory: 20334 grad_norm: 16.5339 loss: 0.6504 +2023/06/01 05:48:29 - mmengine - INFO - Epoch(train) [17][4100/5758] lr: 1.0454e-04 eta: 2:11:53 time: 0.3799 data_time: 0.0022 memory: 20334 grad_norm: 46.5200 loss: 0.6470 +2023/06/01 05:49:10 - mmengine - INFO - Epoch(train) [17][4200/5758] lr: 1.0454e-04 eta: 2:11:11 time: 0.4132 data_time: 0.0019 memory: 20334 grad_norm: 45.2016 loss: 0.6488 +2023/06/01 05:49:50 - mmengine - INFO - Epoch(train) [17][4300/5758] lr: 1.0454e-04 eta: 2:10:29 time: 0.4007 data_time: 0.0017 memory: 20334 grad_norm: 82.1602 loss: 0.6456 +2023/06/01 05:50:30 - mmengine - INFO - Epoch(train) [17][4400/5758] lr: 1.0454e-04 eta: 2:09:47 time: 0.3833 data_time: 0.0016 memory: 20334 grad_norm: 13.5351 loss: 0.6291 +2023/06/01 05:51:09 - mmengine - INFO - Epoch(train) [17][4500/5758] lr: 1.0454e-04 eta: 2:09:05 time: 0.3848 data_time: 0.0022 memory: 20334 grad_norm: 8.1859 loss: 0.6446 +2023/06/01 05:51:50 - mmengine - INFO - Epoch(train) [17][4600/5758] lr: 1.0454e-04 eta: 2:08:23 time: 0.3876 data_time: 0.0016 memory: 20334 grad_norm: 12.5262 loss: 0.6456 +2023/06/01 05:52:31 - mmengine - INFO - Epoch(train) [17][4700/5758] lr: 1.0454e-04 eta: 2:07:41 time: 0.4495 data_time: 0.0024 memory: 20334 grad_norm: 54.9705 loss: 0.6526 +2023/06/01 05:53:12 - mmengine - INFO - Epoch(train) [17][4800/5758] lr: 1.0454e-04 eta: 2:06:59 time: 0.4094 data_time: 0.0026 memory: 20334 grad_norm: 27.8509 loss: 0.6469 +2023/06/01 05:53:42 - mmengine - INFO - Exp name: swin_base_8xb128_fake5m_20230531_183241 +2023/06/01 05:53:52 - mmengine - INFO - Epoch(train) [17][4900/5758] lr: 1.0454e-04 eta: 2:06:17 time: 0.4024 data_time: 0.0023 memory: 20334 grad_norm: 15.3742 loss: 0.6598 +2023/06/01 05:54:33 - mmengine - INFO - Epoch(train) [17][5000/5758] lr: 1.0454e-04 eta: 2:05:35 time: 0.4276 data_time: 0.0020 memory: 20334 grad_norm: 23.7485 loss: 0.6477 +2023/06/01 05:55:13 - mmengine - INFO - Epoch(train) [17][5100/5758] lr: 1.0454e-04 eta: 2:04:53 time: 0.3852 data_time: 0.0021 memory: 20334 grad_norm: 8.8197 loss: 0.6474 +2023/06/01 05:55:53 - mmengine - INFO - Epoch(train) [17][5200/5758] lr: 1.0454e-04 eta: 2:04:11 time: 0.4023 data_time: 0.0018 memory: 20334 grad_norm: 37.9932 loss: 0.6425 +2023/06/01 05:56:33 - mmengine - INFO - Epoch(train) [17][5300/5758] lr: 1.0454e-04 eta: 2:03:29 time: 0.4000 data_time: 0.0023 memory: 20334 grad_norm: 34.1676 loss: 0.6537 +2023/06/01 05:57:14 - mmengine - INFO - Epoch(train) [17][5400/5758] lr: 1.0454e-04 eta: 2:02:47 time: 0.4286 data_time: 0.0026 memory: 20334 grad_norm: 48.1832 loss: 0.6532 +2023/06/01 05:57:54 - mmengine - INFO - Epoch(train) [17][5500/5758] lr: 1.0454e-04 eta: 2:02:04 time: 0.3951 data_time: 0.0017 memory: 20334 grad_norm: 16.8611 loss: 0.6360 +2023/06/01 05:58:34 - mmengine - INFO - Epoch(train) [17][5600/5758] lr: 1.0454e-04 eta: 2:01:22 time: 0.4180 data_time: 0.0026 memory: 20334 grad_norm: 43.3313 loss: 0.6361 +2023/06/01 05:59:15 - mmengine - INFO - Epoch(train) [17][5700/5758] lr: 1.0454e-04 eta: 2:00:40 time: 0.4013 data_time: 0.0019 memory: 20334 grad_norm: 32.4200 loss: 0.6425 +2023/06/01 05:59:38 - mmengine - INFO - Exp name: swin_base_8xb128_fake5m_20230531_183241 +2023/06/01 05:59:38 - mmengine - INFO - Saving checkpoint at 17 epochs +2023/06/01 05:59:55 - mmengine - INFO - Epoch(val) [17][8/8] accuracy/top1: 88.1131 single-label/precision_classwise: [100.0, 0.0] single-label/recall_classwise: [88.11312103271484, 0.0] single-label/f1-score_classwise: [93.68099212646484, 0.0] data_time: 0.3581 time: 0.9834 +2023/06/01 06:00:41 - mmengine - INFO - Epoch(train) [18][ 100/5758] lr: 6.3952e-05 eta: 1:59:35 time: 0.4257 data_time: 0.0021 memory: 20334 grad_norm: 63.2827 loss: 0.6445 +2023/06/01 06:00:47 - mmengine - INFO - Exp name: swin_base_8xb128_fake5m_20230531_183241 +2023/06/01 06:01:22 - mmengine - INFO - Epoch(train) [18][ 200/5758] lr: 6.3952e-05 eta: 1:58:53 time: 0.3952 data_time: 0.0027 memory: 20334 grad_norm: 71.9475 loss: 0.6376 +2023/06/01 06:02:03 - mmengine - INFO - Epoch(train) [18][ 300/5758] lr: 6.3952e-05 eta: 1:58:11 time: 0.4254 data_time: 0.0019 memory: 20334 grad_norm: 27.9691 loss: 0.6340 +2023/06/01 06:02:43 - mmengine - INFO - Epoch(train) [18][ 400/5758] lr: 6.3952e-05 eta: 1:57:29 time: 0.4136 data_time: 0.0029 memory: 20334 grad_norm: 33.1160 loss: 0.6605 +2023/06/01 06:03:22 - mmengine - INFO - Epoch(train) [18][ 500/5758] lr: 6.3952e-05 eta: 1:56:47 time: 0.3749 data_time: 0.0022 memory: 20334 grad_norm: 21.2225 loss: 0.6445 +2023/06/01 06:04:03 - mmengine - INFO - Epoch(train) [18][ 600/5758] lr: 6.3952e-05 eta: 1:56:05 time: 0.4187 data_time: 0.0020 memory: 20334 grad_norm: 75.3499 loss: 0.6529 +2023/06/01 06:04:43 - mmengine - INFO - Epoch(train) [18][ 700/5758] lr: 6.3952e-05 eta: 1:55:23 time: 0.4211 data_time: 0.0018 memory: 20334 grad_norm: 131.6462 loss: 0.6250 +2023/06/01 06:05:23 - mmengine - INFO - Epoch(train) [18][ 800/5758] lr: 6.3952e-05 eta: 1:54:41 time: 0.3788 data_time: 0.0018 memory: 20334 grad_norm: 71.8548 loss: 0.6417 +2023/06/01 06:06:04 - mmengine - INFO - Epoch(train) [18][ 900/5758] lr: 6.3952e-05 eta: 1:53:59 time: 0.3884 data_time: 0.0018 memory: 20334 grad_norm: 14.1204 loss: 0.6533 +2023/06/01 06:06:44 - mmengine - INFO - Epoch(train) [18][1000/5758] lr: 6.3952e-05 eta: 1:53:17 time: 0.4122 data_time: 0.0025 memory: 20334 grad_norm: 82.0788 loss: 0.6442 +2023/06/01 06:07:24 - mmengine - INFO - Epoch(train) [18][1100/5758] lr: 6.3952e-05 eta: 1:52:35 time: 0.4319 data_time: 0.0026 memory: 20334 grad_norm: 17.0746 loss: 0.6485 +2023/06/01 06:07:30 - mmengine - INFO - Exp name: swin_base_8xb128_fake5m_20230531_183241 +2023/06/01 06:08:05 - mmengine - INFO - Epoch(train) [18][1200/5758] lr: 6.3952e-05 eta: 1:51:53 time: 0.3998 data_time: 0.0019 memory: 20334 grad_norm: 12.5568 loss: 0.6545 +2023/06/01 06:08:45 - mmengine - INFO - Epoch(train) [18][1300/5758] lr: 6.3952e-05 eta: 1:51:11 time: 0.3827 data_time: 0.0024 memory: 20334 grad_norm: 43.7911 loss: 0.6472 +2023/06/01 06:09:26 - mmengine - INFO - Epoch(train) [18][1400/5758] lr: 6.3952e-05 eta: 1:50:29 time: 0.3796 data_time: 0.0030 memory: 20334 grad_norm: 19.4557 loss: 0.6481 +2023/06/01 06:10:07 - mmengine - INFO - Epoch(train) [18][1500/5758] lr: 6.3952e-05 eta: 1:49:47 time: 0.4128 data_time: 0.0015 memory: 20334 grad_norm: 19.8311 loss: 0.6462 +2023/06/01 06:10:49 - mmengine - INFO - Epoch(train) [18][1600/5758] lr: 6.3952e-05 eta: 1:49:05 time: 0.4163 data_time: 0.0018 memory: 20334 grad_norm: 30.9645 loss: 0.6447 +2023/06/01 06:11:30 - mmengine - INFO - Epoch(train) [18][1700/5758] lr: 6.3952e-05 eta: 1:48:23 time: 0.4512 data_time: 0.0023 memory: 20334 grad_norm: 20.1364 loss: 0.6417 +2023/06/01 06:12:10 - mmengine - INFO - Epoch(train) [18][1800/5758] lr: 6.3952e-05 eta: 1:47:41 time: 0.3844 data_time: 0.0025 memory: 20334 grad_norm: 32.4031 loss: 0.6447 +2023/06/01 06:12:50 - mmengine - INFO - Epoch(train) [18][1900/5758] lr: 6.3952e-05 eta: 1:46:59 time: 0.3970 data_time: 0.0023 memory: 20334 grad_norm: 14.2658 loss: 0.6529 +2023/06/01 06:13:31 - mmengine - INFO - Epoch(train) [18][2000/5758] lr: 6.3952e-05 eta: 1:46:17 time: 0.3929 data_time: 0.0021 memory: 20334 grad_norm: 20.6992 loss: 0.6427 +2023/06/01 06:14:12 - mmengine - INFO - Epoch(train) [18][2100/5758] lr: 6.3952e-05 eta: 1:45:36 time: 0.4156 data_time: 0.0026 memory: 20334 grad_norm: 70.5440 loss: 0.6593 +2023/06/01 06:14:18 - mmengine - INFO - Exp name: swin_base_8xb128_fake5m_20230531_183241 +2023/06/01 06:14:52 - mmengine - INFO - Epoch(train) [18][2200/5758] lr: 6.3952e-05 eta: 1:44:54 time: 0.3908 data_time: 0.0024 memory: 20334 grad_norm: 21.3339 loss: 0.6493 +2023/06/01 06:15:33 - mmengine - INFO - Epoch(train) [18][2300/5758] lr: 6.3952e-05 eta: 1:44:12 time: 0.3997 data_time: 0.0027 memory: 20334 grad_norm: 27.2712 loss: 0.6560 +2023/06/01 06:16:13 - mmengine - INFO - Epoch(train) [18][2400/5758] lr: 6.3952e-05 eta: 1:43:30 time: 0.3754 data_time: 0.0022 memory: 20334 grad_norm: 25.0591 loss: 0.6598 +2023/06/01 06:16:54 - mmengine - INFO - Epoch(train) [18][2500/5758] lr: 6.3952e-05 eta: 1:42:48 time: 0.3689 data_time: 0.0018 memory: 20334 grad_norm: 26.6567 loss: 0.6417 +2023/06/01 06:17:35 - mmengine - INFO - Epoch(train) [18][2600/5758] lr: 6.3952e-05 eta: 1:42:06 time: 0.3946 data_time: 0.0029 memory: 20334 grad_norm: 44.6973 loss: 0.6490 +2023/06/01 06:18:15 - mmengine - INFO - Epoch(train) [18][2700/5758] lr: 6.3952e-05 eta: 1:41:24 time: 0.3943 data_time: 0.0018 memory: 20334 grad_norm: 225.2489 loss: 0.6620 +2023/06/01 06:18:56 - mmengine - INFO - Epoch(train) [18][2800/5758] lr: 6.3952e-05 eta: 1:40:42 time: 0.3994 data_time: 0.0018 memory: 20334 grad_norm: 83.4851 loss: 0.6607 +2023/06/01 06:19:37 - mmengine - INFO - Epoch(train) [18][2900/5758] lr: 6.3952e-05 eta: 1:40:00 time: 0.3779 data_time: 0.0022 memory: 20334 grad_norm: 44.0406 loss: 0.6465 +2023/06/01 06:20:18 - mmengine - INFO - Epoch(train) [18][3000/5758] lr: 6.3952e-05 eta: 1:39:18 time: 0.4228 data_time: 0.0032 memory: 20334 grad_norm: 24.5456 loss: 0.6401 +2023/06/01 06:20:59 - mmengine - INFO - Epoch(train) [18][3100/5758] lr: 6.3952e-05 eta: 1:38:37 time: 0.3933 data_time: 0.0018 memory: 20334 grad_norm: 128.9199 loss: 0.6634 +2023/06/01 06:21:05 - mmengine - INFO - Exp name: swin_base_8xb128_fake5m_20230531_183241 +2023/06/01 06:21:41 - mmengine - INFO - Epoch(train) [18][3200/5758] lr: 6.3952e-05 eta: 1:37:55 time: 0.4023 data_time: 0.0022 memory: 20334 grad_norm: 152.1927 loss: 0.6575 +2023/06/01 06:22:23 - mmengine - INFO - Epoch(train) [18][3300/5758] lr: 6.3952e-05 eta: 1:37:13 time: 0.3871 data_time: 0.0017 memory: 20334 grad_norm: 34.5417 loss: 0.6583 +2023/06/01 06:23:04 - mmengine - INFO - Epoch(train) [18][3400/5758] lr: 6.3952e-05 eta: 1:36:31 time: 0.3989 data_time: 0.0022 memory: 20334 grad_norm: 10.8451 loss: 0.6594 +2023/06/01 06:23:45 - mmengine - INFO - Epoch(train) [18][3500/5758] lr: 6.3952e-05 eta: 1:35:49 time: 0.4142 data_time: 0.0020 memory: 20334 grad_norm: 136.5663 loss: 0.6515 +2023/06/01 06:24:25 - mmengine - INFO - Epoch(train) [18][3600/5758] lr: 6.3952e-05 eta: 1:35:08 time: 0.3946 data_time: 0.0016 memory: 20334 grad_norm: 26.9938 loss: 0.6600 +2023/06/01 06:25:06 - mmengine - INFO - Epoch(train) [18][3700/5758] lr: 6.3952e-05 eta: 1:34:26 time: 0.3926 data_time: 0.0021 memory: 20334 grad_norm: 32.2152 loss: 0.6732 +2023/06/01 06:25:47 - mmengine - INFO - Epoch(train) [18][3800/5758] lr: 6.3952e-05 eta: 1:33:44 time: 0.4148 data_time: 0.0030 memory: 20334 grad_norm: 63.9369 loss: 0.6618 +2023/06/01 06:26:28 - mmengine - INFO - Epoch(train) [18][3900/5758] lr: 6.3952e-05 eta: 1:33:02 time: 0.4113 data_time: 0.0028 memory: 20334 grad_norm: 25.9544 loss: 0.6741 +2023/06/01 06:27:08 - mmengine - INFO - Epoch(train) [18][4000/5758] lr: 6.3952e-05 eta: 1:32:20 time: 0.4400 data_time: 0.0018 memory: 20334 grad_norm: 71.6971 loss: 0.6596 +2023/06/01 06:27:49 - mmengine - INFO - Epoch(train) [18][4100/5758] lr: 6.3952e-05 eta: 1:31:38 time: 0.3902 data_time: 0.0021 memory: 20334 grad_norm: 46.5575 loss: 0.6668 +2023/06/01 06:27:55 - mmengine - INFO - Exp name: swin_base_8xb128_fake5m_20230531_183241 +2023/06/01 06:28:31 - mmengine - INFO - Epoch(train) [18][4200/5758] lr: 6.3952e-05 eta: 1:30:56 time: 0.4282 data_time: 0.0022 memory: 20334 grad_norm: 102.1191 loss: 0.6648 +2023/06/01 06:29:12 - mmengine - INFO - Epoch(train) [18][4300/5758] lr: 6.3952e-05 eta: 1:30:15 time: 0.3854 data_time: 0.0017 memory: 20334 grad_norm: 91.0550 loss: 0.6600 +2023/06/01 06:29:53 - mmengine - INFO - Epoch(train) [18][4400/5758] lr: 6.3952e-05 eta: 1:29:33 time: 0.3878 data_time: 0.0019 memory: 20334 grad_norm: 40.9992 loss: 0.6546 +2023/06/01 06:30:34 - mmengine - INFO - Epoch(train) [18][4500/5758] lr: 6.3952e-05 eta: 1:28:51 time: 0.4111 data_time: 0.0018 memory: 20334 grad_norm: 58.0564 loss: 0.6499 +2023/06/01 06:31:15 - mmengine - INFO - Epoch(train) [18][4600/5758] lr: 6.3952e-05 eta: 1:28:09 time: 0.4121 data_time: 0.0021 memory: 20334 grad_norm: 38.4158 loss: 0.6537 +2023/06/01 06:31:56 - mmengine - INFO - Epoch(train) [18][4700/5758] lr: 6.3952e-05 eta: 1:27:27 time: 0.4433 data_time: 0.0022 memory: 20334 grad_norm: 47.8747 loss: 0.6522 +2023/06/01 06:32:37 - mmengine - INFO - Epoch(train) [18][4800/5758] lr: 6.3952e-05 eta: 1:26:45 time: 0.4109 data_time: 0.0029 memory: 20334 grad_norm: 42.6568 loss: 0.6453 +2023/06/01 06:33:18 - mmengine - INFO - Epoch(train) [18][4900/5758] lr: 6.3952e-05 eta: 1:26:04 time: 0.4342 data_time: 0.0020 memory: 20334 grad_norm: 59.3426 loss: 0.6423 +2023/06/01 06:33:58 - mmengine - INFO - Epoch(train) [18][5000/5758] lr: 6.3952e-05 eta: 1:25:22 time: 0.3906 data_time: 0.0017 memory: 20334 grad_norm: 15.6146 loss: 0.6566 +2023/06/01 06:34:37 - mmengine - INFO - Epoch(train) [18][5100/5758] lr: 6.3952e-05 eta: 1:24:40 time: 0.3962 data_time: 0.0027 memory: 20334 grad_norm: 21.6596 loss: 0.6484 +2023/06/01 06:34:44 - mmengine - INFO - Exp name: swin_base_8xb128_fake5m_20230531_183241 +2023/06/01 06:35:18 - mmengine - INFO - Epoch(train) [18][5200/5758] lr: 6.3952e-05 eta: 1:23:58 time: 0.3806 data_time: 0.0023 memory: 20334 grad_norm: 296.3698 loss: 0.6488 +2023/06/01 06:35:59 - mmengine - INFO - Epoch(train) [18][5300/5758] lr: 6.3952e-05 eta: 1:23:16 time: 0.4201 data_time: 0.0024 memory: 20334 grad_norm: 171.4270 loss: 0.6475 +2023/06/01 06:36:42 - mmengine - INFO - Epoch(train) [18][5400/5758] lr: 6.3952e-05 eta: 1:22:34 time: 0.4169 data_time: 0.0020 memory: 20334 grad_norm: 15.7709 loss: 0.6602 +2023/06/01 06:37:22 - mmengine - INFO - Epoch(train) [18][5500/5758] lr: 6.3952e-05 eta: 1:21:52 time: 0.4175 data_time: 0.0030 memory: 20334 grad_norm: 15.7511 loss: 0.6604 +2023/06/01 06:38:03 - mmengine - INFO - Epoch(train) [18][5600/5758] lr: 6.3952e-05 eta: 1:21:11 time: 0.4094 data_time: 0.0024 memory: 20334 grad_norm: 16.8459 loss: 0.6537 +2023/06/01 06:38:44 - mmengine - INFO - Epoch(train) [18][5700/5758] lr: 6.3952e-05 eta: 1:20:29 time: 0.4027 data_time: 0.0019 memory: 20334 grad_norm: 27.6579 loss: 0.6491 +2023/06/01 06:39:07 - mmengine - INFO - Exp name: swin_base_8xb128_fake5m_20230531_183241 +2023/06/01 06:39:07 - mmengine - INFO - Saving checkpoint at 18 epochs +2023/06/01 06:39:24 - mmengine - INFO - Epoch(val) [18][8/8] accuracy/top1: 86.9895 single-label/precision_classwise: [100.0, 0.0] single-label/recall_classwise: [86.98945617675781, 0.0] single-label/f1-score_classwise: [93.0420913696289, 0.0] data_time: 0.3680 time: 0.9957 +2023/06/01 06:40:09 - mmengine - INFO - Epoch(train) [19][ 100/5758] lr: 3.4227e-05 eta: 1:19:23 time: 0.4029 data_time: 0.0020 memory: 20334 grad_norm: 66.6539 loss: 0.6515 +2023/06/01 06:40:49 - mmengine - INFO - Epoch(train) [19][ 200/5758] lr: 3.4227e-05 eta: 1:18:41 time: 0.4083 data_time: 0.0019 memory: 20334 grad_norm: 34.7455 loss: 0.6644 +2023/06/01 06:41:31 - mmengine - INFO - Epoch(train) [19][ 300/5758] lr: 3.4227e-05 eta: 1:18:00 time: 0.3937 data_time: 0.0020 memory: 20334 grad_norm: 213.3387 loss: 0.6749 +2023/06/01 06:41:54 - mmengine - INFO - Exp name: swin_base_8xb128_fake5m_20230531_183241 +2023/06/01 06:42:12 - mmengine - INFO - Epoch(train) [19][ 400/5758] lr: 3.4227e-05 eta: 1:17:18 time: 0.4123 data_time: 0.0018 memory: 20334 grad_norm: 74.3776 loss: 0.6606 +2023/06/01 06:42:53 - mmengine - INFO - Epoch(train) [19][ 500/5758] lr: 3.4227e-05 eta: 1:16:36 time: 0.4405 data_time: 0.0019 memory: 20334 grad_norm: 89.7455 loss: 0.6422 +2023/06/01 06:43:34 - mmengine - INFO - Epoch(train) [19][ 600/5758] lr: 3.4227e-05 eta: 1:15:54 time: 0.4268 data_time: 0.0017 memory: 20334 grad_norm: 40.8558 loss: 0.6535 +2023/06/01 06:44:14 - mmengine - INFO - Epoch(train) [19][ 700/5758] lr: 3.4227e-05 eta: 1:15:12 time: 0.3833 data_time: 0.0018 memory: 20334 grad_norm: 15.9707 loss: 0.6586 +2023/06/01 06:44:54 - mmengine - INFO - Epoch(train) [19][ 800/5758] lr: 3.4227e-05 eta: 1:14:30 time: 0.4180 data_time: 0.0018 memory: 20334 grad_norm: 18.2879 loss: 0.6529 +2023/06/01 06:45:35 - mmengine - INFO - Epoch(train) [19][ 900/5758] lr: 3.4227e-05 eta: 1:13:49 time: 0.3997 data_time: 0.0018 memory: 20334 grad_norm: 14.3042 loss: 0.6278 +2023/06/01 06:46:16 - mmengine - INFO - Epoch(train) [19][1000/5758] lr: 3.4227e-05 eta: 1:13:07 time: 0.3913 data_time: 0.0018 memory: 20334 grad_norm: 9.4428 loss: 0.6414 +2023/06/01 06:46:58 - mmengine - INFO - Epoch(train) [19][1100/5758] lr: 3.4227e-05 eta: 1:12:25 time: 0.3959 data_time: 0.0019 memory: 20334 grad_norm: 40.7144 loss: 0.6593 +2023/06/01 06:47:38 - mmengine - INFO - Epoch(train) [19][1200/5758] lr: 3.4227e-05 eta: 1:11:43 time: 0.3841 data_time: 0.0023 memory: 20334 grad_norm: 13.6276 loss: 0.6553 +2023/06/01 06:48:19 - mmengine - INFO - Epoch(train) [19][1300/5758] lr: 3.4227e-05 eta: 1:11:01 time: 0.4204 data_time: 0.0023 memory: 20334 grad_norm: 14.2798 loss: 0.6425 +2023/06/01 06:48:43 - mmengine - INFO - Exp name: swin_base_8xb128_fake5m_20230531_183241 +2023/06/01 06:49:01 - mmengine - INFO - Epoch(train) [19][1400/5758] lr: 3.4227e-05 eta: 1:10:20 time: 0.4097 data_time: 0.0019 memory: 20334 grad_norm: 61.6538 loss: 0.6414 +2023/06/01 06:49:43 - mmengine - INFO - Epoch(train) [19][1500/5758] lr: 3.4227e-05 eta: 1:09:38 time: 0.3920 data_time: 0.0017 memory: 20334 grad_norm: 19.8346 loss: 0.6525 +2023/06/01 06:50:24 - mmengine - INFO - Epoch(train) [19][1600/5758] lr: 3.4227e-05 eta: 1:08:56 time: 0.3671 data_time: 0.0021 memory: 20334 grad_norm: 21.9790 loss: 0.6535 +2023/06/01 06:51:05 - mmengine - INFO - Epoch(train) [19][1700/5758] lr: 3.4227e-05 eta: 1:08:14 time: 0.4107 data_time: 0.0027 memory: 20334 grad_norm: 9.2302 loss: 0.6392 +2023/06/01 06:51:47 - mmengine - INFO - Epoch(train) [19][1800/5758] lr: 3.4227e-05 eta: 1:07:33 time: 0.4140 data_time: 0.0027 memory: 20334 grad_norm: 25.4786 loss: 0.6501 +2023/06/01 06:52:29 - mmengine - INFO - Epoch(train) [19][1900/5758] lr: 3.4227e-05 eta: 1:06:51 time: 0.4219 data_time: 0.0018 memory: 20334 grad_norm: 31.8239 loss: 0.6490 +2023/06/01 06:53:10 - mmengine - INFO - Epoch(train) [19][2000/5758] lr: 3.4227e-05 eta: 1:06:09 time: 0.3887 data_time: 0.0018 memory: 20334 grad_norm: 42.6173 loss: 0.6465 +2023/06/01 06:53:51 - mmengine - INFO - Epoch(train) [19][2100/5758] lr: 3.4227e-05 eta: 1:05:27 time: 0.4121 data_time: 0.0020 memory: 20334 grad_norm: 23.1560 loss: 0.6574 +2023/06/01 06:54:31 - mmengine - INFO - Epoch(train) [19][2200/5758] lr: 3.4227e-05 eta: 1:04:46 time: 0.4002 data_time: 0.0023 memory: 20334 grad_norm: 22.8971 loss: 0.6566 +2023/06/01 06:55:12 - mmengine - INFO - Epoch(train) [19][2300/5758] lr: 3.4227e-05 eta: 1:04:04 time: 0.3821 data_time: 0.0017 memory: 20334 grad_norm: 30.9176 loss: 0.6556 +2023/06/01 06:55:34 - mmengine - INFO - Exp name: swin_base_8xb128_fake5m_20230531_183241 +2023/06/01 06:55:52 - mmengine - INFO - Epoch(train) [19][2400/5758] lr: 3.4227e-05 eta: 1:03:22 time: 0.3650 data_time: 0.0019 memory: 20334 grad_norm: 41.5924 loss: 0.6587 +2023/06/01 06:56:33 - mmengine - INFO - Epoch(train) [19][2500/5758] lr: 3.4227e-05 eta: 1:02:40 time: 0.4692 data_time: 0.0019 memory: 20334 grad_norm: 33.9543 loss: 0.6456 +2023/06/01 06:57:12 - mmengine - INFO - Epoch(train) [19][2600/5758] lr: 3.4227e-05 eta: 1:01:58 time: 0.3927 data_time: 0.0018 memory: 20334 grad_norm: 40.9530 loss: 0.6533 +2023/06/01 06:57:54 - mmengine - INFO - Epoch(train) [19][2700/5758] lr: 3.4227e-05 eta: 1:01:17 time: 0.4043 data_time: 0.0018 memory: 20334 grad_norm: 14.2156 loss: 0.6593 +2023/06/01 06:58:34 - mmengine - INFO - Epoch(train) [19][2800/5758] lr: 3.4227e-05 eta: 1:00:35 time: 0.4019 data_time: 0.0020 memory: 20334 grad_norm: 143.3173 loss: 0.6481 +2023/06/01 06:59:16 - mmengine - INFO - Epoch(train) [19][2900/5758] lr: 3.4227e-05 eta: 0:59:53 time: 0.4302 data_time: 0.0020 memory: 20334 grad_norm: 40.7847 loss: 0.6658 +2023/06/01 06:59:57 - mmengine - INFO - Epoch(train) [19][3000/5758] lr: 3.4227e-05 eta: 0:59:11 time: 0.4114 data_time: 0.0019 memory: 20334 grad_norm: 19.6428 loss: 0.6577 +2023/06/01 07:00:37 - mmengine - INFO - Epoch(train) [19][3100/5758] lr: 3.4227e-05 eta: 0:58:29 time: 0.4089 data_time: 0.0015 memory: 20334 grad_norm: 75.8778 loss: 0.6442 +2023/06/01 07:01:18 - mmengine - INFO - Epoch(train) [19][3200/5758] lr: 3.4227e-05 eta: 0:57:48 time: 0.4103 data_time: 0.0015 memory: 20334 grad_norm: 12.3155 loss: 0.6522 +2023/06/01 07:01:58 - mmengine - INFO - Epoch(train) [19][3300/5758] lr: 3.4227e-05 eta: 0:57:06 time: 0.4080 data_time: 0.0017 memory: 20334 grad_norm: 14.5197 loss: 0.6573 +2023/06/01 07:02:21 - mmengine - INFO - Exp name: swin_base_8xb128_fake5m_20230531_183241 +2023/06/01 07:02:38 - mmengine - INFO - Epoch(train) [19][3400/5758] lr: 3.4227e-05 eta: 0:56:24 time: 0.3837 data_time: 0.0022 memory: 20334 grad_norm: 49.5584 loss: 0.6521 +2023/06/01 07:03:19 - mmengine - INFO - Epoch(train) [19][3500/5758] lr: 3.4227e-05 eta: 0:55:42 time: 0.3867 data_time: 0.0017 memory: 20334 grad_norm: 34.7282 loss: 0.6657 +2023/06/01 07:04:00 - mmengine - INFO - Epoch(train) [19][3600/5758] lr: 3.4227e-05 eta: 0:55:00 time: 0.3831 data_time: 0.0016 memory: 20334 grad_norm: 17.0216 loss: 0.6684 +2023/06/01 07:04:41 - mmengine - INFO - Epoch(train) [19][3700/5758] lr: 3.4227e-05 eta: 0:54:19 time: 0.3945 data_time: 0.0017 memory: 20334 grad_norm: 30.0264 loss: 0.6665 +2023/06/01 07:05:22 - mmengine - INFO - Epoch(train) [19][3800/5758] lr: 3.4227e-05 eta: 0:53:37 time: 0.3869 data_time: 0.0016 memory: 20334 grad_norm: 19.3257 loss: 0.6644 +2023/06/01 07:06:03 - mmengine - INFO - Epoch(train) [19][3900/5758] lr: 3.4227e-05 eta: 0:52:55 time: 0.4218 data_time: 0.0016 memory: 20334 grad_norm: 36.1686 loss: 0.6727 +2023/06/01 07:06:43 - mmengine - INFO - Epoch(train) [19][4000/5758] lr: 3.4227e-05 eta: 0:52:13 time: 0.4407 data_time: 0.0017 memory: 20334 grad_norm: 12.6983 loss: 0.6665 +2023/06/01 07:07:23 - mmengine - INFO - Epoch(train) [19][4100/5758] lr: 3.4227e-05 eta: 0:51:32 time: 0.4180 data_time: 0.0016 memory: 20334 grad_norm: 18.5814 loss: 0.6727 +2023/06/01 07:08:05 - mmengine - INFO - Epoch(train) [19][4200/5758] lr: 3.4227e-05 eta: 0:50:50 time: 0.4268 data_time: 0.0016 memory: 20334 grad_norm: 40.6030 loss: 0.6681 +2023/06/01 07:08:44 - mmengine - INFO - Epoch(train) [19][4300/5758] lr: 3.4227e-05 eta: 0:50:08 time: 0.4036 data_time: 0.0016 memory: 20334 grad_norm: 85.0285 loss: 0.6686 +2023/06/01 07:09:08 - mmengine - INFO - Exp name: swin_base_8xb128_fake5m_20230531_183241 +2023/06/01 07:09:26 - mmengine - INFO - Epoch(train) [19][4400/5758] lr: 3.4227e-05 eta: 0:49:26 time: 0.4441 data_time: 0.0016 memory: 20334 grad_norm: 62.7240 loss: 0.6558 +2023/06/01 07:10:07 - mmengine - INFO - Epoch(train) [19][4500/5758] lr: 3.4227e-05 eta: 0:48:45 time: 0.4474 data_time: 0.0018 memory: 20334 grad_norm: 99.8403 loss: 0.6742 +2023/06/01 07:10:47 - mmengine - INFO - Epoch(train) [19][4600/5758] lr: 3.4227e-05 eta: 0:48:03 time: 0.4304 data_time: 0.0016 memory: 20334 grad_norm: 57.7155 loss: 0.6722 +2023/06/01 07:11:28 - mmengine - INFO - Epoch(train) [19][4700/5758] lr: 3.4227e-05 eta: 0:47:21 time: 0.4323 data_time: 0.0017 memory: 20334 grad_norm: 80.3763 loss: 0.6625 +2023/06/01 07:12:08 - mmengine - INFO - Epoch(train) [19][4800/5758] lr: 3.4227e-05 eta: 0:46:39 time: 0.3881 data_time: 0.0020 memory: 20334 grad_norm: 62.8092 loss: 0.6744 +2023/06/01 07:12:50 - mmengine - INFO - Epoch(train) [19][4900/5758] lr: 3.4227e-05 eta: 0:45:58 time: 0.3931 data_time: 0.0020 memory: 20334 grad_norm: 31.4577 loss: 0.6706 +2023/06/01 07:13:31 - mmengine - INFO - Epoch(train) [19][5000/5758] lr: 3.4227e-05 eta: 0:45:16 time: 0.4111 data_time: 0.0020 memory: 20334 grad_norm: 56.2091 loss: 0.6599 +2023/06/01 07:14:12 - mmengine - INFO - Epoch(train) [19][5100/5758] lr: 3.4227e-05 eta: 0:44:34 time: 0.4001 data_time: 0.0014 memory: 20334 grad_norm: 206.1507 loss: 0.6731 +2023/06/01 07:14:53 - mmengine - INFO - Epoch(train) [19][5200/5758] lr: 3.4227e-05 eta: 0:43:52 time: 0.3867 data_time: 0.0036 memory: 20334 grad_norm: 44.8787 loss: 0.6735 +2023/06/01 07:15:33 - mmengine - INFO - Epoch(train) [19][5300/5758] lr: 3.4227e-05 eta: 0:43:11 time: 0.4130 data_time: 0.0024 memory: 20334 grad_norm: 79.7090 loss: 0.6719 +2023/06/01 07:15:57 - mmengine - INFO - Exp name: swin_base_8xb128_fake5m_20230531_183241 +2023/06/01 07:16:15 - mmengine - INFO - Epoch(train) [19][5400/5758] lr: 3.4227e-05 eta: 0:42:29 time: 0.4001 data_time: 0.0019 memory: 20334 grad_norm: 58.0159 loss: 0.6580 +2023/06/01 07:16:55 - mmengine - INFO - Epoch(train) [19][5500/5758] lr: 3.4227e-05 eta: 0:41:47 time: 0.3854 data_time: 0.0022 memory: 20334 grad_norm: 86.6426 loss: 0.6738 +2023/06/01 07:17:36 - mmengine - INFO - Epoch(train) [19][5600/5758] lr: 3.4227e-05 eta: 0:41:05 time: 0.3885 data_time: 0.0017 memory: 20334 grad_norm: 38.4154 loss: 0.6705 +2023/06/01 07:18:17 - mmengine - INFO - Epoch(train) [19][5700/5758] lr: 3.4227e-05 eta: 0:40:24 time: 0.3995 data_time: 0.0018 memory: 20334 grad_norm: 203.1712 loss: 0.6734 +2023/06/01 07:18:40 - mmengine - INFO - Exp name: swin_base_8xb128_fake5m_20230531_183241 +2023/06/01 07:18:40 - mmengine - INFO - Saving checkpoint at 19 epochs +2023/06/01 07:18:56 - mmengine - INFO - Epoch(val) [19][8/8] accuracy/top1: 85.8342 single-label/precision_classwise: [100.0, 0.0] single-label/recall_classwise: [85.834228515625, 0.0] single-label/f1-score_classwise: [92.377197265625, 0.0] data_time: 0.3582 time: 0.9835 +2023/06/01 07:19:41 - mmengine - INFO - Epoch(train) [20][ 100/5758] lr: 1.6094e-05 eta: 0:39:18 time: 0.3837 data_time: 0.0018 memory: 20334 grad_norm: 37.8316 loss: 0.6657 +2023/06/01 07:20:23 - mmengine - INFO - Epoch(train) [20][ 200/5758] lr: 1.6094e-05 eta: 0:38:36 time: 0.3624 data_time: 0.0023 memory: 20334 grad_norm: 131.8389 loss: 0.6685 +2023/06/01 07:21:04 - mmengine - INFO - Epoch(train) [20][ 300/5758] lr: 1.6094e-05 eta: 0:37:55 time: 0.4057 data_time: 0.0032 memory: 20334 grad_norm: 73.4937 loss: 0.6725 +2023/06/01 07:21:44 - mmengine - INFO - Epoch(train) [20][ 400/5758] lr: 1.6094e-05 eta: 0:37:13 time: 0.3943 data_time: 0.0029 memory: 20334 grad_norm: 48.6913 loss: 0.6698 +2023/06/01 07:22:26 - mmengine - INFO - Epoch(train) [20][ 500/5758] lr: 1.6094e-05 eta: 0:36:31 time: 0.3894 data_time: 0.0024 memory: 20334 grad_norm: 47.0496 loss: 0.6681 +2023/06/01 07:23:05 - mmengine - INFO - Exp name: swin_base_8xb128_fake5m_20230531_183241 +2023/06/01 07:23:06 - mmengine - INFO - Epoch(train) [20][ 600/5758] lr: 1.6094e-05 eta: 0:35:49 time: 0.3859 data_time: 0.0024 memory: 20334 grad_norm: 1127.3899 loss: 0.6733 +2023/06/01 07:23:47 - mmengine - INFO - Epoch(train) [20][ 700/5758] lr: 1.6094e-05 eta: 0:35:08 time: 0.4176 data_time: 0.0024 memory: 20334 grad_norm: 82.1451 loss: 0.6699 +2023/06/01 07:24:29 - mmengine - INFO - Epoch(train) [20][ 800/5758] lr: 1.6094e-05 eta: 0:34:26 time: 0.4404 data_time: 0.0017 memory: 20334 grad_norm: 11.3550 loss: 0.6541 +2023/06/01 07:25:09 - mmengine - INFO - Epoch(train) [20][ 900/5758] lr: 1.6094e-05 eta: 0:33:44 time: 0.3815 data_time: 0.0021 memory: 20334 grad_norm: 44.5827 loss: 0.6699 +2023/06/01 07:25:49 - mmengine - INFO - Epoch(train) [20][1000/5758] lr: 1.6094e-05 eta: 0:33:03 time: 0.3839 data_time: 0.0020 memory: 20334 grad_norm: 181.1916 loss: 0.6704 +2023/06/01 07:26:31 - mmengine - INFO - Epoch(train) [20][1100/5758] lr: 1.6094e-05 eta: 0:32:21 time: 0.4080 data_time: 0.0026 memory: 20334 grad_norm: 29.3792 loss: 0.6695 +2023/06/01 07:27:11 - mmengine - INFO - Epoch(train) [20][1200/5758] lr: 1.6094e-05 eta: 0:31:39 time: 0.3870 data_time: 0.0019 memory: 20334 grad_norm: 112.6470 loss: 0.6738 +2023/06/01 07:27:51 - mmengine - INFO - Epoch(train) [20][1300/5758] lr: 1.6094e-05 eta: 0:30:57 time: 0.3776 data_time: 0.0025 memory: 20334 grad_norm: 97.6848 loss: 0.6756 +2023/06/01 07:28:31 - mmengine - INFO - Epoch(train) [20][1400/5758] lr: 1.6094e-05 eta: 0:30:16 time: 0.3918 data_time: 0.0017 memory: 20334 grad_norm: 71.3823 loss: 0.6758 +2023/06/01 07:29:13 - mmengine - INFO - Epoch(train) [20][1500/5758] lr: 1.6094e-05 eta: 0:29:34 time: 0.4175 data_time: 0.0031 memory: 20334 grad_norm: 12.3704 loss: 0.6698 +2023/06/01 07:29:53 - mmengine - INFO - Exp name: swin_base_8xb128_fake5m_20230531_183241 +2023/06/01 07:29:54 - mmengine - INFO - Epoch(train) [20][1600/5758] lr: 1.6094e-05 eta: 0:28:52 time: 0.4094 data_time: 0.0020 memory: 20334 grad_norm: 217.7896 loss: 0.6712 +2023/06/01 07:30:35 - mmengine - INFO - Epoch(train) [20][1700/5758] lr: 1.6094e-05 eta: 0:28:11 time: 0.4169 data_time: 0.0020 memory: 20334 grad_norm: 75.2327 loss: 0.6697 +2023/06/01 07:31:16 - mmengine - INFO - Epoch(train) [20][1800/5758] lr: 1.6094e-05 eta: 0:27:29 time: 0.3831 data_time: 0.0026 memory: 20334 grad_norm: 49.7328 loss: 0.6667 +2023/06/01 07:31:56 - mmengine - INFO - Epoch(train) [20][1900/5758] lr: 1.6094e-05 eta: 0:26:47 time: 0.3818 data_time: 0.0018 memory: 20334 grad_norm: 33.9477 loss: 0.6709 +2023/06/01 07:32:37 - mmengine - INFO - Epoch(train) [20][2000/5758] lr: 1.6094e-05 eta: 0:26:05 time: 0.3901 data_time: 0.0018 memory: 20334 grad_norm: 148.2320 loss: 0.6696 +2023/06/01 07:33:17 - mmengine - INFO - Epoch(train) [20][2100/5758] lr: 1.6094e-05 eta: 0:25:24 time: 0.3885 data_time: 0.0021 memory: 20334 grad_norm: 24.1435 loss: 0.6764 +2023/06/01 07:33:57 - mmengine - INFO - Epoch(train) [20][2200/5758] lr: 1.6094e-05 eta: 0:24:42 time: 0.3702 data_time: 0.0018 memory: 20334 grad_norm: 29.8747 loss: 0.6745 +2023/06/01 07:34:39 - mmengine - INFO - Epoch(train) [20][2300/5758] lr: 1.6094e-05 eta: 0:24:00 time: 0.4365 data_time: 0.0020 memory: 20334 grad_norm: 91.6214 loss: 0.6710 +2023/06/01 07:35:20 - mmengine - INFO - Epoch(train) [20][2400/5758] lr: 1.6094e-05 eta: 0:23:19 time: 0.4153 data_time: 0.0017 memory: 20334 grad_norm: 86.2144 loss: 0.6671 +2023/06/01 07:36:01 - mmengine - INFO - Epoch(train) [20][2500/5758] lr: 1.6094e-05 eta: 0:22:37 time: 0.4084 data_time: 0.0026 memory: 20334 grad_norm: 304.8029 loss: 0.6622 +2023/06/01 07:36:41 - mmengine - INFO - Exp name: swin_base_8xb128_fake5m_20230531_183241 +2023/06/01 07:36:41 - mmengine - INFO - Epoch(train) [20][2600/5758] lr: 1.6094e-05 eta: 0:21:55 time: 0.4042 data_time: 0.0022 memory: 20334 grad_norm: 97.6951 loss: 0.6740 +2023/06/01 07:37:22 - mmengine - INFO - Epoch(train) [20][2700/5758] lr: 1.6094e-05 eta: 0:21:14 time: 0.3819 data_time: 0.0019 memory: 20334 grad_norm: 93.4661 loss: 0.6579 +2023/06/01 07:38:02 - mmengine - INFO - Epoch(train) [20][2800/5758] lr: 1.6094e-05 eta: 0:20:32 time: 0.3765 data_time: 0.0019 memory: 20334 grad_norm: 38.1375 loss: 0.6646 +2023/06/01 07:38:45 - mmengine - INFO - Epoch(train) [20][2900/5758] lr: 1.6094e-05 eta: 0:19:50 time: 0.4129 data_time: 0.0021 memory: 20334 grad_norm: 38.6878 loss: 0.6653 +2023/06/01 07:39:25 - mmengine - INFO - Epoch(train) [20][3000/5758] lr: 1.6094e-05 eta: 0:19:09 time: 0.3859 data_time: 0.0021 memory: 20334 grad_norm: 22.5367 loss: 0.6759 +2023/06/01 07:40:06 - mmengine - INFO - Epoch(train) [20][3100/5758] lr: 1.6094e-05 eta: 0:18:27 time: 0.4300 data_time: 0.0021 memory: 20334 grad_norm: 51.0495 loss: 0.6673 +2023/06/01 07:40:48 - mmengine - INFO - Epoch(train) [20][3200/5758] lr: 1.6094e-05 eta: 0:17:45 time: 0.4007 data_time: 0.0033 memory: 20334 grad_norm: 46.9795 loss: 0.6686 +2023/06/01 07:41:27 - mmengine - INFO - Epoch(train) [20][3300/5758] lr: 1.6094e-05 eta: 0:17:03 time: 0.3918 data_time: 0.0025 memory: 20334 grad_norm: 94.8533 loss: 0.6783 +2023/06/01 07:42:09 - mmengine - INFO - Epoch(train) [20][3400/5758] lr: 1.6094e-05 eta: 0:16:22 time: 0.3831 data_time: 0.0033 memory: 20334 grad_norm: 36.0965 loss: 0.6774 +2023/06/01 07:42:50 - mmengine - INFO - Epoch(train) [20][3500/5758] lr: 1.6094e-05 eta: 0:15:40 time: 0.3858 data_time: 0.0018 memory: 20334 grad_norm: 14.7771 loss: 0.6693 +2023/06/01 07:43:31 - mmengine - INFO - Exp name: swin_base_8xb128_fake5m_20230531_183241 +2023/06/01 07:43:32 - mmengine - INFO - Epoch(train) [20][3600/5758] lr: 1.6094e-05 eta: 0:14:58 time: 0.4007 data_time: 0.0025 memory: 20334 grad_norm: 48.9178 loss: 0.6821 +2023/06/01 07:44:12 - mmengine - INFO - Epoch(train) [20][3700/5758] lr: 1.6094e-05 eta: 0:14:17 time: 0.4178 data_time: 0.0030 memory: 20334 grad_norm: 41.1830 loss: 0.6624 +2023/06/01 07:44:52 - mmengine - INFO - Epoch(train) [20][3800/5758] lr: 1.6094e-05 eta: 0:13:35 time: 0.3991 data_time: 0.0019 memory: 20334 grad_norm: 49.4230 loss: 0.6683 +2023/06/01 07:45:33 - mmengine - INFO - Epoch(train) [20][3900/5758] lr: 1.6094e-05 eta: 0:12:53 time: 0.3863 data_time: 0.0018 memory: 20334 grad_norm: 61.3406 loss: 0.6697 +2023/06/01 07:46:14 - mmengine - INFO - Epoch(train) [20][4000/5758] lr: 1.6094e-05 eta: 0:12:12 time: 0.4347 data_time: 0.0020 memory: 20334 grad_norm: 20.7913 loss: 0.6625 +2023/06/01 07:46:55 - mmengine - INFO - Epoch(train) [20][4100/5758] lr: 1.6094e-05 eta: 0:11:30 time: 0.4153 data_time: 0.0023 memory: 20334 grad_norm: 58.9290 loss: 0.6697 +2023/06/01 07:47:36 - mmengine - INFO - Epoch(train) [20][4200/5758] lr: 1.6094e-05 eta: 0:10:48 time: 0.3967 data_time: 0.0021 memory: 20334 grad_norm: 18.9529 loss: 0.6848 +2023/06/01 07:48:16 - mmengine - INFO - Epoch(train) [20][4300/5758] lr: 1.6094e-05 eta: 0:10:07 time: 0.3866 data_time: 0.0022 memory: 20334 grad_norm: 20.6123 loss: 0.6800 +2023/06/01 07:48:57 - mmengine - INFO - Epoch(train) [20][4400/5758] lr: 1.6094e-05 eta: 0:09:25 time: 0.4148 data_time: 0.0017 memory: 20334 grad_norm: 46.0141 loss: 0.6571 +2023/06/01 07:49:39 - mmengine - INFO - Epoch(train) [20][4500/5758] lr: 1.6094e-05 eta: 0:08:43 time: 0.4024 data_time: 0.0020 memory: 20334 grad_norm: 135.2398 loss: 0.6718 +2023/06/01 07:50:20 - mmengine - INFO - Exp name: swin_base_8xb128_fake5m_20230531_183241 +2023/06/01 07:50:20 - mmengine - INFO - Epoch(train) [20][4600/5758] lr: 1.6094e-05 eta: 0:08:02 time: 0.4077 data_time: 0.0019 memory: 20334 grad_norm: 42.5326 loss: 0.6675 +2023/06/01 07:51:00 - mmengine - INFO - Epoch(train) [20][4700/5758] lr: 1.6094e-05 eta: 0:07:20 time: 0.4156 data_time: 0.0026 memory: 20334 grad_norm: 20.7560 loss: 0.6798 +2023/06/01 07:51:41 - mmengine - INFO - Epoch(train) [20][4800/5758] lr: 1.6094e-05 eta: 0:06:38 time: 0.3626 data_time: 0.0022 memory: 20334 grad_norm: 12.9377 loss: 0.6682 +2023/06/01 07:52:24 - mmengine - INFO - Epoch(train) [20][4900/5758] lr: 1.6094e-05 eta: 0:05:57 time: 0.4467 data_time: 0.0019 memory: 20334 grad_norm: 40.2078 loss: 0.6652 +2023/06/01 07:53:05 - mmengine - INFO - Epoch(train) [20][5000/5758] lr: 1.6094e-05 eta: 0:05:15 time: 0.3934 data_time: 0.0022 memory: 20334 grad_norm: 925.7427 loss: 0.6716 +2023/06/01 07:53:45 - mmengine - INFO - Epoch(train) [20][5100/5758] lr: 1.6094e-05 eta: 0:04:34 time: 0.3918 data_time: 0.0017 memory: 20334 grad_norm: 31.5509 loss: 0.6629 +2023/06/01 07:54:25 - mmengine - INFO - Epoch(train) [20][5200/5758] lr: 1.6094e-05 eta: 0:03:52 time: 0.4472 data_time: 0.0027 memory: 20334 grad_norm: 53.2940 loss: 0.6778 +2023/06/01 07:55:06 - mmengine - INFO - Epoch(train) [20][5300/5758] lr: 1.6094e-05 eta: 0:03:10 time: 0.3977 data_time: 0.0023 memory: 20334 grad_norm: 202.7078 loss: 0.6775 +2023/06/01 07:55:47 - mmengine - INFO - Epoch(train) [20][5400/5758] lr: 1.6094e-05 eta: 0:02:29 time: 0.4490 data_time: 0.0017 memory: 20334 grad_norm: 117.0472 loss: 0.6648 +2023/06/01 07:56:28 - mmengine - INFO - Epoch(train) [20][5500/5758] lr: 1.6094e-05 eta: 0:01:47 time: 0.4497 data_time: 0.0024 memory: 20334 grad_norm: 166.6944 loss: 0.6743 +2023/06/01 07:57:08 - mmengine - INFO - Exp name: swin_base_8xb128_fake5m_20230531_183241 +2023/06/01 07:57:09 - mmengine - INFO - Epoch(train) [20][5600/5758] lr: 1.6094e-05 eta: 0:01:05 time: 0.4039 data_time: 0.0032 memory: 20334 grad_norm: 60.1827 loss: 0.6722 +2023/06/01 07:57:51 - mmengine - INFO - Epoch(train) [20][5700/5758] lr: 1.6094e-05 eta: 0:00:24 time: 0.3753 data_time: 0.0018 memory: 20334 grad_norm: 116.3336 loss: 0.6590 +2023/06/01 07:58:14 - mmengine - INFO - Exp name: swin_base_8xb128_fake5m_20230531_183241 +2023/06/01 07:58:14 - mmengine - INFO - Saving checkpoint at 20 epochs +2023/06/01 07:58:30 - mmengine - INFO - Epoch(val) [20][8/8] accuracy/top1: 80.5252 single-label/precision_classwise: [100.0, 0.0] single-label/recall_classwise: [80.52522277832031, 0.0] single-label/f1-score_classwise: [89.212158203125, 0.0] data_time: 0.3744 time: 1.0008 diff --git a/swin_base_8xb128_1e-3lr_5m/20230531_183241/vis_data/20230531_183241.json b/swin_base_8xb128_1e-3lr_5m/20230531_183241/vis_data/20230531_183241.json new file mode 100644 index 0000000000000000000000000000000000000000..1103fa3896a89a2b820058511435d2bec4d37862 --- /dev/null +++ b/swin_base_8xb128_1e-3lr_5m/20230531_183241/vis_data/20230531_183241.json @@ -0,0 +1,1160 @@ +{"lr": 0.001, "data_time": 0.001547527313232422, "grad_norm": 1.2535260617733002, "loss": 0.6617416918277741, "time": 0.4504673480987549, "epoch": 1, "memory": 20328, "step": 100} +{"lr": 0.001, "data_time": 0.0021373510360717775, "grad_norm": 1.3020053178071975, "loss": 0.6786544740200042, "time": 0.41860456466674806, "epoch": 1, "memory": 20327, "step": 200} +{"lr": 0.001, "data_time": 0.0015830039978027345, "grad_norm": 1.6163756191730498, "loss": 0.6961738705635071, "time": 0.41177544593811033, "epoch": 1, "memory": 20327, "step": 300} +{"lr": 0.001, "data_time": 0.001738595962524414, "grad_norm": 0.916207692027092, "loss": 0.6909652471542358, "time": 0.44023077487945556, "epoch": 1, "memory": 20327, "step": 400} +{"lr": 0.001, "data_time": 0.002464485168457031, "grad_norm": 0.7051077991724014, "loss": 0.6895924627780914, "time": 0.48884766101837157, "epoch": 1, "memory": 20327, "step": 500} +{"lr": 0.001, "data_time": 0.0024768590927124025, "grad_norm": 1.066253486275673, "loss": 0.6935589075088501, "time": 0.4829955816268921, "epoch": 1, "memory": 20327, "step": 600} +{"lr": 0.001, "data_time": 0.0024304866790771486, "grad_norm": 0.6848722219467163, "loss": 0.692876398563385, "time": 0.44585633277893066, "epoch": 1, "memory": 20327, "step": 700} +{"lr": 0.001, "data_time": 0.002019524574279785, "grad_norm": 0.4387118622660637, "loss": 0.6869839131832123, "time": 0.47382638454437254, "epoch": 1, "memory": 20327, "step": 800} +{"lr": 0.001, "data_time": 0.0021958351135253906, "grad_norm": 0.7460969805717468, "loss": 0.6842294037342072, "time": 0.48500847816467285, "epoch": 1, "memory": 20327, "step": 900} +{"lr": 0.001, "data_time": 0.0018077850341796874, "grad_norm": 0.5114121817052364, "loss": 0.6820880651474, "time": 0.5280540943145752, "epoch": 1, "memory": 20327, "step": 1000} +{"lr": 0.001, "data_time": 0.0019245386123657227, "grad_norm": 0.43836204409599305, "loss": 0.6854023098945617, "time": 0.42091100215911864, "epoch": 1, "memory": 20327, "step": 1100} +{"lr": 0.001, "data_time": 0.0017555475234985352, "grad_norm": 0.581676198542118, "loss": 0.6898906469345093, "time": 0.4580332040786743, "epoch": 1, "memory": 20327, "step": 1200} +{"lr": 0.001, "data_time": 0.0014124155044555665, "grad_norm": 0.4683442488312721, "loss": 0.684032940864563, "time": 0.4685969829559326, "epoch": 1, "memory": 20327, "step": 1300} +{"lr": 0.001, "data_time": 0.001987648010253906, "grad_norm": 0.4273607209324837, "loss": 0.6885849595069885, "time": 0.44467649459838865, "epoch": 1, "memory": 20327, "step": 1400} +{"lr": 0.001, "data_time": 0.0024909496307373045, "grad_norm": 0.3759254410862923, "loss": 0.6895965993404388, "time": 0.4259899616241455, "epoch": 1, "memory": 20327, "step": 1500} +{"lr": 0.001, "data_time": 0.0016693115234375, "grad_norm": 0.47642067819833755, "loss": 0.6906299531459809, "time": 0.4184162139892578, "epoch": 1, "memory": 20327, "step": 1600} +{"lr": 0.001, "data_time": 0.002308964729309082, "grad_norm": 0.31735817939043043, "loss": 0.688661515712738, "time": 0.42727036476135255, "epoch": 1, "memory": 20327, "step": 1700} +{"lr": 0.001, "data_time": 0.05264580249786377, "grad_norm": 0.23922348469495774, "loss": 0.6857641935348511, "time": 0.44586069583892823, "epoch": 1, "memory": 20327, "step": 1800} +{"lr": 0.001, "data_time": 0.0017943620681762696, "grad_norm": 0.3035322234034538, "loss": 0.6880684614181518, "time": 0.5702909708023072, "epoch": 1, "memory": 20327, "step": 1900} +{"lr": 0.001, "data_time": 0.0017334461212158204, "grad_norm": 0.24675946086645126, "loss": 0.6844491779804229, "time": 0.5232514619827271, "epoch": 1, "memory": 20327, "step": 2000} +{"lr": 0.001, "data_time": 0.0015720605850219726, "grad_norm": 0.24305076003074647, "loss": 0.6868813693523407, "time": 0.4493051528930664, "epoch": 1, "memory": 20327, "step": 2100} +{"lr": 0.001, "data_time": 0.11564030647277831, "grad_norm": 0.21443298310041428, "loss": 0.6881125211715698, "time": 0.4997014760971069, "epoch": 1, "memory": 20327, "step": 2200} +{"lr": 0.001, "data_time": 0.0016659259796142577, "grad_norm": 0.32948400005698203, "loss": 0.6890442490577697, "time": 0.46114354133605956, "epoch": 1, "memory": 20327, "step": 2300} +{"lr": 0.001, "data_time": 0.0015170812606811524, "grad_norm": 0.32062850557267664, "loss": 0.683380401134491, "time": 0.4767552137374878, "epoch": 1, "memory": 20327, "step": 2400} +{"lr": 0.001, "data_time": 0.0020232439041137696, "grad_norm": 0.18023623824119567, "loss": 0.6857121706008911, "time": 0.4431389093399048, "epoch": 1, "memory": 20327, "step": 2500} +{"lr": 0.001, "data_time": 0.001457381248474121, "grad_norm": 0.2827820084989071, "loss": 0.6844555914402009, "time": 0.5280491590499878, "epoch": 1, "memory": 20327, "step": 2600} +{"lr": 0.001, "data_time": 0.001476883888244629, "grad_norm": 0.23454785868525505, "loss": 0.6897346138954162, "time": 0.45286059379577637, "epoch": 1, "memory": 20327, "step": 2700} +{"lr": 0.001, "data_time": 0.0018162965774536134, "grad_norm": 0.15394328869879245, "loss": 0.6885932683944702, "time": 0.5384252548217774, "epoch": 1, "memory": 20327, "step": 2800} +{"lr": 0.001, "data_time": 0.001485729217529297, "grad_norm": 0.24269349724054337, "loss": 0.6879509806632995, "time": 0.43753461837768554, "epoch": 1, "memory": 20327, "step": 2900} +{"lr": 0.001, "data_time": 0.0013791561126708985, "grad_norm": 0.23015224523842334, "loss": 0.691087681055069, "time": 0.49692587852478026, "epoch": 1, "memory": 20327, "step": 3000} +{"lr": 0.001, "data_time": 0.0022708892822265623, "grad_norm": 0.20756464637815952, "loss": 0.6824763238430023, "time": 0.5300716400146485, "epoch": 1, "memory": 20327, "step": 3100} +{"lr": 0.001, "data_time": 0.0015095949172973632, "grad_norm": 0.25365260392427447, "loss": 0.6853608310222625, "time": 0.5860336065292359, "epoch": 1, "memory": 20327, "step": 3200} +{"lr": 0.001, "data_time": 0.0024956226348876952, "grad_norm": 0.18091317750513553, "loss": 0.6864110231399536, "time": 0.46506948471069337, "epoch": 1, "memory": 20327, "step": 3300} +{"lr": 0.001, "data_time": 0.0018155813217163087, "grad_norm": 0.2349975012242794, "loss": 0.6889593541622162, "time": 0.4547833204269409, "epoch": 1, "memory": 20327, "step": 3400} +{"lr": 0.001, "data_time": 0.0017111063003540038, "grad_norm": 0.19061683416366576, "loss": 0.6881612062454223, "time": 0.4564051628112793, "epoch": 1, "memory": 20327, "step": 3500} +{"lr": 0.001, "data_time": 0.00162198543548584, "grad_norm": 0.21332599148154258, "loss": 0.6878034114837647, "time": 0.46280708312988283, "epoch": 1, "memory": 20327, "step": 3600} +{"lr": 0.001, "data_time": 0.0021987199783325196, "grad_norm": 0.21882116943597793, "loss": 0.6868311047554017, "time": 0.4189113140106201, "epoch": 1, "memory": 20327, "step": 3700} +{"lr": 0.001, "data_time": 0.0015400409698486327, "grad_norm": 0.15679723285138608, "loss": 0.6828992605209351, "time": 0.47562575340270996, "epoch": 1, "memory": 20327, "step": 3800} +{"lr": 0.001, "data_time": 0.0014181852340698243, "grad_norm": 0.24460803493857383, "loss": 0.6878409683704376, "time": 0.5233757495880127, "epoch": 1, "memory": 20327, "step": 3900} +{"lr": 0.001, "data_time": 0.0016360282897949219, "grad_norm": 0.2751976024359465, "loss": 0.6841594159603119, "time": 0.5318449974060059, "epoch": 1, "memory": 20327, "step": 4000} +{"lr": 0.001, "data_time": 0.0014498233795166016, "grad_norm": 0.1677925668656826, "loss": 0.6814287006855011, "time": 0.4640952110290527, "epoch": 1, "memory": 20327, "step": 4100} +{"lr": 0.001, "data_time": 0.0015105962753295898, "grad_norm": 0.22063578739762307, "loss": 0.6908972799777985, "time": 0.6015638113021851, "epoch": 1, "memory": 20327, "step": 4200} +{"lr": 0.001, "data_time": 0.001835179328918457, "grad_norm": 0.13054346553981305, "loss": 0.6935339093208313, "time": 0.4642556428909302, "epoch": 1, "memory": 20327, "step": 4300} +{"lr": 0.001, "data_time": 0.0017675638198852539, "grad_norm": 0.18920982331037522, "loss": 0.680409848690033, "time": 0.5172134637832642, "epoch": 1, "memory": 20327, "step": 4400} +{"lr": 0.001, "data_time": 0.001902294158935547, "grad_norm": 4.143417645990849, "loss": 0.6886740803718567, "time": 0.5838793039321899, "epoch": 1, "memory": 20327, "step": 4500} +{"lr": 0.001, "data_time": 0.04444584846496582, "grad_norm": 0.28976195603609084, "loss": 0.6888139069080352, "time": 0.48262581825256345, "epoch": 1, "memory": 20327, "step": 4600} +{"lr": 0.001, "data_time": 0.011837482452392578, "grad_norm": 0.22792382091283797, "loss": 0.6893905997276306, "time": 0.5790437221527099, "epoch": 1, "memory": 20327, "step": 4700} +{"lr": 0.001, "data_time": 0.04299075603485107, "grad_norm": 0.14554290305823087, "loss": 0.6873553156852722, "time": 0.40713043212890626, "epoch": 1, "memory": 20327, "step": 4800} +{"lr": 0.001, "data_time": 0.0017435789108276368, "grad_norm": 0.24170147106051446, "loss": 0.6905454933643341, "time": 0.46413068771362304, "epoch": 1, "memory": 20327, "step": 4900} +{"lr": 0.001, "data_time": 0.0017269372940063477, "grad_norm": 0.18632605150341988, "loss": 0.6864282250404358, "time": 0.47283387184143066, "epoch": 1, "memory": 20327, "step": 5000} +{"lr": 0.001, "data_time": 0.001972675323486328, "grad_norm": 0.16910027228295804, "loss": 0.6881202518939972, "time": 0.4918528079986572, "epoch": 1, "memory": 20327, "step": 5100} +{"lr": 0.001, "data_time": 0.0016173601150512695, "grad_norm": 0.2514938943088055, "loss": 0.6828402757644654, "time": 0.434708309173584, "epoch": 1, "memory": 20327, "step": 5200} +{"lr": 0.001, "data_time": 0.0014672040939331054, "grad_norm": 0.19111468642950058, "loss": 0.684564745426178, "time": 0.47212269306182864, "epoch": 1, "memory": 20327, "step": 5300} +{"lr": 0.001, "data_time": 0.0014446258544921875, "grad_norm": 0.1328372661024332, "loss": 0.6848789632320404, "time": 0.46407668590545653, "epoch": 1, "memory": 20327, "step": 5400} +{"lr": 0.001, "data_time": 0.0016707897186279297, "grad_norm": 0.1780287515372038, "loss": 0.684296190738678, "time": 0.5697210550308227, "epoch": 1, "memory": 20327, "step": 5500} +{"lr": 0.001, "data_time": 0.0016327142715454102, "grad_norm": 0.19417341612279415, "loss": 0.6819146037101745, "time": 0.3914592981338501, "epoch": 1, "memory": 20327, "step": 5600} +{"lr": 0.001, "data_time": 0.0016374826431274415, "grad_norm": 0.2729407835751772, "loss": 0.6899257600307465, "time": 0.5306423902511597, "epoch": 1, "memory": 20327, "step": 5700} +{"accuracy/top1": 100.0, "data_time": 0.5273530781269073, "time": 1.4208164513111115, "step": 1} +{"lr": 0.0009939057285945933, "data_time": 0.0018706560134887696, "grad_norm": 0.1389899268746376, "loss": 0.6856710553169251, "time": 0.47192771434783937, "epoch": 2, "memory": 20338, "step": 5858} +{"lr": 0.0009939057285945933, "data_time": 0.0014982223510742188, "grad_norm": 0.203866333514452, "loss": 0.6875182628631592, "time": 0.37861616611480714, "epoch": 2, "memory": 20334, "step": 5958} +{"lr": 0.0009939057285945933, "data_time": 0.002318286895751953, "grad_norm": 0.18244499415159227, "loss": 0.6888192117214202, "time": 0.4775179386138916, "epoch": 2, "memory": 20334, "step": 6058} +{"lr": 0.0009939057285945933, "data_time": 0.0020490169525146486, "grad_norm": 0.10983743183314801, "loss": 0.687062245607376, "time": 0.40245158672332765, "epoch": 2, "memory": 20334, "step": 6158} +{"lr": 0.0009939057285945933, "data_time": 0.002003383636474609, "grad_norm": 0.16991456001996993, "loss": 0.6880364120006561, "time": 0.4649769067764282, "epoch": 2, "memory": 20334, "step": 6258} +{"lr": 0.0009939057285945933, "data_time": 0.002188229560852051, "grad_norm": 0.2249813586473465, "loss": 0.6882048785686493, "time": 0.4856530427932739, "epoch": 2, "memory": 20334, "step": 6358} +{"lr": 0.0009939057285945933, "data_time": 0.002771282196044922, "grad_norm": 0.09802362509071827, "loss": 0.6891048014163971, "time": 0.5223216056823731, "epoch": 2, "memory": 20334, "step": 6458} +{"lr": 0.0009939057285945933, "data_time": 0.0017272233963012695, "grad_norm": 0.20471495091915132, "loss": 0.6880715608596801, "time": 0.4997502088546753, "epoch": 2, "memory": 20334, "step": 6558} +{"lr": 0.0009939057285945933, "data_time": 0.0015089988708496093, "grad_norm": 0.13341220766305922, "loss": 0.6904010236263275, "time": 0.4293144941329956, "epoch": 2, "memory": 20334, "step": 6658} +{"lr": 0.0009939057285945933, "data_time": 0.0023128747940063476, "grad_norm": 0.1210234124213457, "loss": 0.6860248982906342, "time": 0.5041555881500244, "epoch": 2, "memory": 20334, "step": 6758} +{"lr": 0.0009939057285945933, "data_time": 0.0016096115112304687, "grad_norm": 0.12085049711167813, "loss": 0.6867158532142639, "time": 0.4423695087432861, "epoch": 2, "memory": 20334, "step": 6858} +{"lr": 0.0009939057285945933, "data_time": 0.001529359817504883, "grad_norm": 0.10074920188635587, "loss": 0.6838590800762177, "time": 0.45317051410675047, "epoch": 2, "memory": 20334, "step": 6958} +{"lr": 0.0009939057285945933, "data_time": 0.0014278888702392578, "grad_norm": 0.2354393795132637, "loss": 0.688062334060669, "time": 0.4827596187591553, "epoch": 2, "memory": 20334, "step": 7058} +{"lr": 0.0009939057285945933, "data_time": 0.0013913154602050782, "grad_norm": 0.12252147868275642, "loss": 0.6852106213569641, "time": 0.44258482456207277, "epoch": 2, "memory": 20334, "step": 7158} +{"lr": 0.0009939057285945933, "data_time": 0.0016436576843261719, "grad_norm": 0.1605440594255924, "loss": 0.6864913821220398, "time": 0.40481493473052976, "epoch": 2, "memory": 20334, "step": 7258} +{"lr": 0.0009939057285945933, "data_time": 0.0013869285583496093, "grad_norm": 0.079430715367198, "loss": 0.6856729388237, "time": 0.38026976585388184, "epoch": 2, "memory": 20334, "step": 7358} +{"lr": 0.0009939057285945933, "data_time": 0.001480865478515625, "grad_norm": 0.13062963932752608, "loss": 0.686570119857788, "time": 0.4127319097518921, "epoch": 2, "memory": 20334, "step": 7458} +{"lr": 0.0009939057285945933, "data_time": 0.0014569520950317382, "grad_norm": 0.18666400890797377, "loss": 0.6869749307632447, "time": 0.43243656158447263, "epoch": 2, "memory": 20334, "step": 7558} +{"lr": 0.0009939057285945933, "data_time": 0.0015127420425415038, "grad_norm": 0.1899154122918844, "loss": 0.6858054041862488, "time": 0.4226412534713745, "epoch": 2, "memory": 20334, "step": 7658} +{"lr": 0.0009939057285945933, "data_time": 0.0015762805938720702, "grad_norm": 0.12433419600129128, "loss": 0.6885032474994659, "time": 0.40570530891418455, "epoch": 2, "memory": 20334, "step": 7758} +{"lr": 0.0009939057285945933, "data_time": 0.0014954090118408203, "grad_norm": 0.0796710742637515, "loss": 0.6879362523555755, "time": 0.4477119207382202, "epoch": 2, "memory": 20334, "step": 7858} +{"lr": 0.0009939057285945933, "data_time": 0.0017971992492675781, "grad_norm": 0.13888155594468116, "loss": 0.690966647863388, "time": 0.5014991283416748, "epoch": 2, "memory": 20334, "step": 7958} +{"lr": 0.0009939057285945933, "data_time": 0.0016384124755859375, "grad_norm": 0.0999188207089901, "loss": 0.6889211356639862, "time": 0.3787181615829468, "epoch": 2, "memory": 20334, "step": 8058} +{"lr": 0.0009939057285945933, "data_time": 0.0015300989151000976, "grad_norm": 0.09481384158134461, "loss": 0.6863315820693969, "time": 0.45148138999938964, "epoch": 2, "memory": 20334, "step": 8158} +{"lr": 0.0009939057285945933, "data_time": 0.0015707969665527343, "grad_norm": 0.11229593753814697, "loss": 0.6916408360004425, "time": 0.44246342182159426, "epoch": 2, "memory": 20334, "step": 8258} +{"lr": 0.0009939057285945933, "data_time": 0.0016265630722045899, "grad_norm": 0.11539451573044061, "loss": 0.6822717428207398, "time": 0.43319680690765383, "epoch": 2, "memory": 20334, "step": 8358} +{"lr": 0.0009939057285945933, "data_time": 0.0014976739883422851, "grad_norm": 0.1470975536853075, "loss": 0.6900055825710296, "time": 0.411281418800354, "epoch": 2, "memory": 20334, "step": 8458} +{"lr": 0.0009939057285945933, "data_time": 0.001532769203186035, "grad_norm": 0.10724248215556145, "loss": 0.6859984517097473, "time": 0.38384811878204345, "epoch": 2, "memory": 20334, "step": 8558} +{"lr": 0.0009939057285945933, "data_time": 0.001632833480834961, "grad_norm": 0.1074720073491335, "loss": 0.6898310124874115, "time": 0.4390037298202515, "epoch": 2, "memory": 20334, "step": 8658} +{"lr": 0.0009939057285945933, "data_time": 0.0015862941741943359, "grad_norm": 0.1138488307595253, "loss": 0.6887602984905243, "time": 0.3876753091812134, "epoch": 2, "memory": 20334, "step": 8758} +{"lr": 0.0009939057285945933, "data_time": 0.0017073392868041993, "grad_norm": 0.08672575345262885, "loss": 0.6898120701313019, "time": 0.3934889554977417, "epoch": 2, "memory": 20334, "step": 8858} +{"lr": 0.0009939057285945933, "data_time": 0.0016271829605102538, "grad_norm": 0.1329948291182518, "loss": 0.6889298319816589, "time": 0.45748813152313234, "epoch": 2, "memory": 20334, "step": 8958} +{"lr": 0.0009939057285945933, "data_time": 0.00175931453704834, "grad_norm": 0.0947910262271762, "loss": 0.6842213153839112, "time": 0.43602006435394286, "epoch": 2, "memory": 20334, "step": 9058} +{"lr": 0.0009939057285945933, "data_time": 0.0017358064651489258, "grad_norm": 0.12013679184019566, "loss": 0.6887552499771118, "time": 0.40340964794158934, "epoch": 2, "memory": 20334, "step": 9158} +{"lr": 0.0009939057285945933, "data_time": 0.001702094078063965, "grad_norm": 0.1379565104842186, "loss": 0.682598215341568, "time": 0.3863911390304565, "epoch": 2, "memory": 20334, "step": 9258} +{"lr": 0.0009939057285945933, "data_time": 0.001566028594970703, "grad_norm": 0.07153388736769557, "loss": 0.6875362396240234, "time": 0.42173769474029543, "epoch": 2, "memory": 20334, "step": 9358} +{"lr": 0.0009939057285945933, "data_time": 0.001628279685974121, "grad_norm": 0.08207729160785675, "loss": 0.6868714213371276, "time": 0.402485728263855, "epoch": 2, "memory": 20334, "step": 9458} +{"lr": 0.0009939057285945933, "data_time": 0.0014389991760253907, "grad_norm": 0.08153223311528564, "loss": 0.6841646730899811, "time": 0.397409725189209, "epoch": 2, "memory": 20334, "step": 9558} +{"lr": 0.0009939057285945933, "data_time": 0.001594686508178711, "grad_norm": 0.061381729785352945, "loss": 0.6824875712394715, "time": 0.42694969177246095, "epoch": 2, "memory": 20334, "step": 9658} +{"lr": 0.0009939057285945933, "data_time": 0.001752185821533203, "grad_norm": 0.05635301554575563, "loss": 0.6929938197135925, "time": 0.40352165699005127, "epoch": 2, "memory": 20334, "step": 9758} +{"lr": 0.0009939057285945933, "data_time": 0.0016124248504638672, "grad_norm": 0.0691122176591307, "loss": 0.6821423649787903, "time": 0.4068138837814331, "epoch": 2, "memory": 20334, "step": 9858} +{"lr": 0.0009939057285945933, "data_time": 0.00162811279296875, "grad_norm": 0.06478845039382577, "loss": 0.6845524370670318, "time": 0.4095698595046997, "epoch": 2, "memory": 20334, "step": 9958} +{"lr": 0.0009939057285945933, "data_time": 0.0017833948135375977, "grad_norm": 0.08239487018436194, "loss": 0.6868374109268188, "time": 0.41894381046295165, "epoch": 2, "memory": 20334, "step": 10058} +{"lr": 0.0009939057285945933, "data_time": 0.0018874168395996093, "grad_norm": 0.07758684400469065, "loss": 0.6857448697090149, "time": 0.41786601543426516, "epoch": 2, "memory": 20334, "step": 10158} +{"lr": 0.0009939057285945933, "data_time": 0.001524806022644043, "grad_norm": 0.07034221142530442, "loss": 0.6880883395671844, "time": 0.4282370090484619, "epoch": 2, "memory": 20334, "step": 10258} +{"lr": 0.0009939057285945933, "data_time": 0.001620316505432129, "grad_norm": 0.05666370280086994, "loss": 0.6862131476402282, "time": 0.3999110460281372, "epoch": 2, "memory": 20334, "step": 10358} +{"lr": 0.0009939057285945933, "data_time": 0.0018820524215698241, "grad_norm": 0.03454902684316039, "loss": 0.6858645379543304, "time": 0.3662744045257568, "epoch": 2, "memory": 20334, "step": 10458} +{"lr": 0.0009939057285945933, "data_time": 0.001518416404724121, "grad_norm": 0.06227385513484478, "loss": 0.6873582184314728, "time": 0.3886637449264526, "epoch": 2, "memory": 20334, "step": 10558} +{"lr": 0.0009939057285945933, "data_time": 0.0026936769485473634, "grad_norm": 0.05943831531330943, "loss": 0.6837661266326904, "time": 0.4403040647506714, "epoch": 2, "memory": 20334, "step": 10658} +{"lr": 0.0009939057285945933, "data_time": 0.0015952348709106444, "grad_norm": 0.06036773715168238, "loss": 0.6927081525325776, "time": 0.38903734683990476, "epoch": 2, "memory": 20334, "step": 10758} +{"lr": 0.0009939057285945933, "data_time": 0.0017139673233032226, "grad_norm": 0.05382824447005987, "loss": 0.6866870164871216, "time": 0.4493489503860474, "epoch": 2, "memory": 20334, "step": 10858} +{"lr": 0.0009939057285945933, "data_time": 0.0016783475875854492, "grad_norm": 0.06428167028352619, "loss": 0.6885355353355408, "time": 0.440744686126709, "epoch": 2, "memory": 20334, "step": 10958} +{"lr": 0.0009939057285945933, "data_time": 0.002162528038024902, "grad_norm": 0.03117213575169444, "loss": 0.6879472315311432, "time": 0.363454532623291, "epoch": 2, "memory": 20334, "step": 11058} +{"lr": 0.0009939057285945933, "data_time": 0.002293848991394043, "grad_norm": 0.04414070025086403, "loss": 0.6896017730236054, "time": 0.43469820022583006, "epoch": 2, "memory": 20334, "step": 11158} +{"lr": 0.0009939057285945933, "data_time": 0.0018540382385253905, "grad_norm": 0.036172684840857984, "loss": 0.6920473754405976, "time": 0.43832666873931886, "epoch": 2, "memory": 20334, "step": 11258} +{"lr": 0.0009939057285945933, "data_time": 0.001553797721862793, "grad_norm": 0.03078562580049038, "loss": 0.6833684384822846, "time": 0.3910383224487305, "epoch": 2, "memory": 20334, "step": 11358} +{"lr": 0.0009939057285945933, "data_time": 0.001596689224243164, "grad_norm": 0.03620063448324799, "loss": 0.6886676490306854, "time": 0.39936027526855467, "epoch": 2, "memory": 20334, "step": 11458} +{"accuracy/top1": 100.0, "data_time": 0.37360231081644696, "time": 1.0013271967569988, "step": 2} +{"lr": 0.0009757729755661012, "data_time": 0.001520848274230957, "grad_norm": 0.03098635757341981, "loss": 0.686473685503006, "time": 0.4017822504043579, "epoch": 3, "memory": 20334, "step": 11616} +{"lr": 0.0009757729755661012, "data_time": 0.0015061616897583008, "grad_norm": 0.033890397474169734, "loss": 0.6903098583221435, "time": 0.39765985012054444, "epoch": 3, "memory": 20334, "step": 11716} +{"lr": 0.0009757729755661012, "data_time": 0.0015069961547851563, "grad_norm": 0.030056470725685357, "loss": 0.6796724200248718, "time": 0.3921332836151123, "epoch": 3, "memory": 20334, "step": 11816} +{"lr": 0.0009757729755661012, "data_time": 0.0016312599182128906, "grad_norm": 0.022210765909403563, "loss": 0.6907181024551392, "time": 0.4321902751922607, "epoch": 3, "memory": 20334, "step": 11916} +{"lr": 0.0009757729755661012, "data_time": 0.0014042377471923829, "grad_norm": 0.025225077383220197, "loss": 0.6878110349178315, "time": 0.3707267761230469, "epoch": 3, "memory": 20334, "step": 12016} +{"lr": 0.0009757729755661012, "data_time": 0.0014371633529663085, "grad_norm": 0.026616774778813124, "loss": 0.6892135739326477, "time": 0.4078239440917969, "epoch": 3, "memory": 20334, "step": 12116} +{"lr": 0.0009757729755661012, "data_time": 0.0014354944229125976, "grad_norm": 0.028887146292254327, "loss": 0.6870896816253662, "time": 0.41640093326568606, "epoch": 3, "memory": 20334, "step": 12216} +{"lr": 0.0009757729755661012, "data_time": 0.0015312910079956054, "grad_norm": 0.027596437186002732, "loss": 0.6887373507022858, "time": 0.4126712799072266, "epoch": 3, "memory": 20334, "step": 12316} +{"lr": 0.0009757729755661012, "data_time": 0.0013911724090576172, "grad_norm": 0.016476803366094826, "loss": 0.6819482147693634, "time": 0.4864032506942749, "epoch": 3, "memory": 20334, "step": 12416} +{"lr": 0.0009757729755661012, "data_time": 0.0016367435455322266, "grad_norm": 0.021859406493604183, "loss": 0.6840069830417633, "time": 0.4090930938720703, "epoch": 3, "memory": 20334, "step": 12516} +{"lr": 0.0009757729755661012, "data_time": 0.0014048099517822265, "grad_norm": 0.023460118751972913, "loss": 0.6887216448783875, "time": 0.43534467220306394, "epoch": 3, "memory": 20334, "step": 12616} +{"lr": 0.0009757729755661012, "data_time": 0.0014093399047851562, "grad_norm": 0.014717079047113658, "loss": 0.685056871175766, "time": 0.466562819480896, "epoch": 3, "memory": 20334, "step": 12716} +{"lr": 0.0009757729755661012, "data_time": 0.001378011703491211, "grad_norm": 0.025230715004727244, "loss": 0.6858988761901855, "time": 0.4218915939331055, "epoch": 3, "memory": 20334, "step": 12816} +{"lr": 0.0009757729755661012, "data_time": 0.0014751911163330077, "grad_norm": 0.015667925914749505, "loss": 0.688463419675827, "time": 0.46251440048217773, "epoch": 3, "memory": 20334, "step": 12916} +{"lr": 0.0009757729755661012, "data_time": 0.0014195680618286134, "grad_norm": 0.025939735118299723, "loss": 0.6954469442367553, "time": 0.4380846738815308, "epoch": 3, "memory": 20334, "step": 13016} +{"lr": 0.0009757729755661012, "data_time": 0.0023333072662353516, "grad_norm": 0.023735856171697378, "loss": 0.687801194190979, "time": 0.4287290573120117, "epoch": 3, "memory": 20334, "step": 13116} +{"lr": 0.0009757729755661012, "data_time": 0.0018094539642333984, "grad_norm": 0.019573861034587026, "loss": 0.6907796859741211, "time": 0.4156827211380005, "epoch": 3, "memory": 20334, "step": 13216} +{"lr": 0.0009757729755661012, "data_time": 0.0019041538238525391, "grad_norm": 0.022345282882452012, "loss": 0.6813410520553589, "time": 0.43147850036621094, "epoch": 3, "memory": 20334, "step": 13316} +{"lr": 0.0009757729755661012, "data_time": 0.001631474494934082, "grad_norm": 0.02729429369792342, "loss": 0.6872678399085999, "time": 0.4804619073867798, "epoch": 3, "memory": 20334, "step": 13416} +{"lr": 0.0009757729755661012, "data_time": 0.0014719724655151366, "grad_norm": 0.02647479922743514, "loss": 0.6898800075054169, "time": 0.4309577941894531, "epoch": 3, "memory": 20334, "step": 13516} +{"lr": 0.0009757729755661012, "data_time": 0.0015274763107299804, "grad_norm": 0.012423992273397744, "loss": 0.6817545771598816, "time": 0.42549293041229247, "epoch": 3, "memory": 20334, "step": 13616} +{"lr": 0.0009757729755661012, "data_time": 0.002018260955810547, "grad_norm": 0.011961333989165723, "loss": 0.6839232504367828, "time": 0.4453761339187622, "epoch": 3, "memory": 20334, "step": 13716} +{"lr": 0.0009757729755661012, "data_time": 0.0013747215270996094, "grad_norm": 0.023575666639953852, "loss": 0.690808218717575, "time": 0.46250483989715574, "epoch": 3, "memory": 20334, "step": 13816} +{"lr": 0.0009757729755661012, "data_time": 0.001609635353088379, "grad_norm": 0.018087553139775992, "loss": 0.6885170996189117, "time": 0.4347598791122437, "epoch": 3, "memory": 20334, "step": 13916} +{"lr": 0.0009757729755661012, "data_time": 0.002219557762145996, "grad_norm": 0.027281677862629295, "loss": 0.6876326441764832, "time": 0.4458295822143555, "epoch": 3, "memory": 20334, "step": 14016} +{"lr": 0.0009757729755661012, "data_time": 0.0015475749969482422, "grad_norm": 0.013312716549262405, "loss": 0.6822128236293793, "time": 0.396474027633667, "epoch": 3, "memory": 20334, "step": 14116} +{"lr": 0.0009757729755661012, "data_time": 0.0012688636779785156, "grad_norm": 0.027546300599351526, "loss": 0.6798486113548279, "time": 0.4620030641555786, "epoch": 3, "memory": 20334, "step": 14216} +{"lr": 0.0009757729755661012, "data_time": 0.0019216299057006835, "grad_norm": 0.021650880412198602, "loss": 0.6841916501522064, "time": 0.47051043510437013, "epoch": 3, "memory": 20334, "step": 14316} +{"lr": 0.0009757729755661012, "data_time": 0.0017227649688720704, "grad_norm": 0.01692313519306481, "loss": 0.6871234238147735, "time": 0.42992634773254396, "epoch": 3, "memory": 20334, "step": 14416} +{"lr": 0.0009757729755661012, "data_time": 0.0015279293060302735, "grad_norm": 0.02390893567353487, "loss": 0.6868479728698731, "time": 0.4040129899978638, "epoch": 3, "memory": 20334, "step": 14516} +{"lr": 0.0009757729755661012, "data_time": 0.0016637802124023437, "grad_norm": 0.016689633810892703, "loss": 0.686801153421402, "time": 0.39360096454620364, "epoch": 3, "memory": 20334, "step": 14616} +{"lr": 0.0009757729755661012, "data_time": 0.0015065193176269532, "grad_norm": 0.015438964287750424, "loss": 0.6861332893371582, "time": 0.4511989116668701, "epoch": 3, "memory": 20334, "step": 14716} +{"lr": 0.0009757729755661012, "data_time": 0.0016159296035766601, "grad_norm": 0.014466957421973348, "loss": 0.683395254611969, "time": 0.4681440591812134, "epoch": 3, "memory": 20334, "step": 14816} +{"lr": 0.0009757729755661012, "data_time": 0.0015751838684082032, "grad_norm": 0.022110573854297398, "loss": 0.683032101392746, "time": 0.4665671825408936, "epoch": 3, "memory": 20334, "step": 14916} +{"lr": 0.0009757729755661012, "data_time": 0.0024945974349975587, "grad_norm": 0.011839405028149485, "loss": 0.6831228733062744, "time": 0.4275668621063232, "epoch": 3, "memory": 20334, "step": 15016} +{"lr": 0.0009757729755661012, "data_time": 0.0022376298904418944, "grad_norm": 0.016086345398798586, "loss": 0.6844076752662659, "time": 0.395918869972229, "epoch": 3, "memory": 20334, "step": 15116} +{"lr": 0.0009757729755661012, "data_time": 0.0014168024063110352, "grad_norm": 0.021715958369895816, "loss": 0.6888582885265351, "time": 0.39960379600524903, "epoch": 3, "memory": 20334, "step": 15216} +{"lr": 0.0009757729755661012, "data_time": 0.0012727022171020509, "grad_norm": 0.011211237497627735, "loss": 0.6836376905441284, "time": 0.4284745454788208, "epoch": 3, "memory": 20334, "step": 15316} +{"lr": 0.0009757729755661012, "data_time": 0.0020062923431396484, "grad_norm": 0.016982131195254623, "loss": 0.6858044803142548, "time": 0.4195528984069824, "epoch": 3, "memory": 20334, "step": 15416} +{"lr": 0.0009757729755661012, "data_time": 0.0015186309814453126, "grad_norm": 0.03138001998886466, "loss": 0.6842008352279663, "time": 0.429775595664978, "epoch": 3, "memory": 20334, "step": 15516} +{"lr": 0.0009757729755661012, "data_time": 0.0020823240280151366, "grad_norm": 0.016965086571872234, "loss": 0.6856278061866761, "time": 0.4045754909515381, "epoch": 3, "memory": 20334, "step": 15616} +{"lr": 0.0009757729755661012, "data_time": 0.00151824951171875, "grad_norm": 0.01606045940425247, "loss": 0.6863158762454986, "time": 0.43207101821899413, "epoch": 3, "memory": 20334, "step": 15716} +{"lr": 0.0009757729755661012, "data_time": 0.0015492677688598634, "grad_norm": 0.01343541257083416, "loss": 0.6925902724266052, "time": 0.43695497512817383, "epoch": 3, "memory": 20334, "step": 15816} +{"lr": 0.0009757729755661012, "data_time": 0.04586305618286133, "grad_norm": 0.017851545987650753, "loss": 0.678933173418045, "time": 0.4091578722000122, "epoch": 3, "memory": 20334, "step": 15916} +{"lr": 0.0009757729755661012, "data_time": 0.06472687721252442, "grad_norm": 0.01796461995691061, "loss": 0.685563576221466, "time": 0.4260616064071655, "epoch": 3, "memory": 20334, "step": 16016} +{"lr": 0.0009757729755661012, "data_time": 0.07754015922546387, "grad_norm": 0.013906694715842605, "loss": 0.6849724590778351, "time": 0.44074790477752684, "epoch": 3, "memory": 20334, "step": 16116} +{"lr": 0.0009757729755661012, "data_time": 0.04347686767578125, "grad_norm": 0.02395407101139426, "loss": 0.6887342035770416, "time": 0.42560787200927735, "epoch": 3, "memory": 20334, "step": 16216} +{"lr": 0.0009757729755661012, "data_time": 0.0015156269073486328, "grad_norm": 0.018618854158557952, "loss": 0.6877447664737701, "time": 0.4609185218811035, "epoch": 3, "memory": 20334, "step": 16316} +{"lr": 0.0009757729755661012, "data_time": 0.0016674757003784179, "grad_norm": 0.01747428609523922, "loss": 0.6846291959285736, "time": 0.46385653018951417, "epoch": 3, "memory": 20334, "step": 16416} +{"lr": 0.0009757729755661012, "data_time": 0.0012150049209594727, "grad_norm": 0.016686558863148092, "loss": 0.6920796573162079, "time": 0.41220109462738036, "epoch": 3, "memory": 20334, "step": 16516} +{"lr": 0.0009757729755661012, "data_time": 0.0013314247131347655, "grad_norm": 0.01650532289640978, "loss": 0.6890563368797302, "time": 0.40956666469573977, "epoch": 3, "memory": 20334, "step": 16616} +{"lr": 0.0009757729755661012, "data_time": 0.002044057846069336, "grad_norm": 0.014679540507495403, "loss": 0.692743718624115, "time": 0.4864722967147827, "epoch": 3, "memory": 20334, "step": 16716} +{"lr": 0.0009757729755661012, "data_time": 0.0016607999801635741, "grad_norm": 0.014594785962253809, "loss": 0.6871274709701538, "time": 0.4180482864379883, "epoch": 3, "memory": 20334, "step": 16816} +{"lr": 0.0009757729755661012, "data_time": 0.0023061752319335936, "grad_norm": 0.014580905344337225, "loss": 0.6873303294181824, "time": 0.42358126640319826, "epoch": 3, "memory": 20334, "step": 16916} +{"lr": 0.0009757729755661012, "data_time": 0.0020044565200805662, "grad_norm": 0.016765149729326367, "loss": 0.6843859195709229, "time": 0.4132028818130493, "epoch": 3, "memory": 20334, "step": 17016} +{"lr": 0.0009757729755661012, "data_time": 0.0019243240356445312, "grad_norm": 0.014524851349415257, "loss": 0.6876846849918365, "time": 0.43416016101837157, "epoch": 3, "memory": 20334, "step": 17116} +{"lr": 0.0009757729755661012, "data_time": 0.0016091823577880859, "grad_norm": 0.023040681355632843, "loss": 0.6922148585319519, "time": 0.427603554725647, "epoch": 3, "memory": 20334, "step": 17216} +{"accuracy/top1": 100.0, "data_time": 0.36508538987901473, "time": 0.9904248184627957, "step": 3} +{"lr": 0.0009460482294732422, "data_time": 0.0017638683319091796, "grad_norm": 0.01643019961193204, "loss": 0.6822502732276916, "time": 0.3723294734954834, "epoch": 4, "memory": 20334, "step": 17374} +{"lr": 0.0009460482294732422, "data_time": 0.001596522331237793, "grad_norm": 0.0149491639691405, "loss": 0.6835873544216156, "time": 0.48390424251556396, "epoch": 4, "memory": 20334, "step": 17474} +{"lr": 0.0009460482294732422, "data_time": 0.001820540428161621, "grad_norm": 0.021044259099289774, "loss": 0.6859103977680207, "time": 0.3999966621398926, "epoch": 4, "memory": 20334, "step": 17574} +{"lr": 0.0009460482294732422, "data_time": 0.0017197370529174806, "grad_norm": 0.016584765748120844, "loss": 0.6917169094085693, "time": 0.4330127477645874, "epoch": 4, "memory": 20334, "step": 17674} +{"lr": 0.0009460482294732422, "data_time": 0.0018273115158081055, "grad_norm": 0.01176101916935295, "loss": 0.6858778119087219, "time": 0.432884669303894, "epoch": 4, "memory": 20334, "step": 17774} +{"lr": 0.0009460482294732422, "data_time": 0.0016796350479125976, "grad_norm": 0.01780250105075538, "loss": 0.6884469509124755, "time": 0.41221568584442136, "epoch": 4, "memory": 20334, "step": 17874} +{"lr": 0.0009460482294732422, "data_time": 0.0015875339508056641, "grad_norm": 0.018298603734001516, "loss": 0.6912717163562775, "time": 0.43295817375183104, "epoch": 4, "memory": 20334, "step": 17974} +{"lr": 0.0009460482294732422, "data_time": 0.0016761064529418946, "grad_norm": 0.02583385189063847, "loss": 0.6846343040466308, "time": 0.4225667953491211, "epoch": 4, "memory": 20334, "step": 18074} +{"lr": 0.0009460482294732422, "data_time": 0.0016031980514526368, "grad_norm": 0.020494342735037207, "loss": 0.6850262105464935, "time": 0.4490375280380249, "epoch": 4, "memory": 20334, "step": 18174} +{"lr": 0.0009460482294732422, "data_time": 0.0017838001251220704, "grad_norm": 0.026996927242726086, "loss": 0.6846012651920319, "time": 0.39264416694641113, "epoch": 4, "memory": 20334, "step": 18274} +{"lr": 0.0009460482294732422, "data_time": 0.0016048908233642577, "grad_norm": 0.023436758480966092, "loss": 0.6880956172943116, "time": 0.4475454330444336, "epoch": 4, "memory": 20334, "step": 18374} +{"lr": 0.0009460482294732422, "data_time": 0.0018970966339111328, "grad_norm": 0.017902054672595112, "loss": 0.6856350541114807, "time": 0.4378422737121582, "epoch": 4, "memory": 20334, "step": 18474} +{"lr": 0.0009460482294732422, "data_time": 0.001533842086791992, "grad_norm": 0.012605548347346484, "loss": 0.6849187016487122, "time": 0.42951369285583496, "epoch": 4, "memory": 20334, "step": 18574} +{"lr": 0.0009460482294732422, "data_time": 0.001800847053527832, "grad_norm": 0.0144324409076944, "loss": 0.6878259658813477, "time": 0.49349207878112794, "epoch": 4, "memory": 20334, "step": 18674} +{"lr": 0.0009460482294732422, "data_time": 0.001868867874145508, "grad_norm": 0.017530618351884186, "loss": 0.6795748353004456, "time": 0.4449542284011841, "epoch": 4, "memory": 20334, "step": 18774} +{"lr": 0.0009460482294732422, "data_time": 0.0018942117691040038, "grad_norm": 0.023734983801841737, "loss": 0.6832464158535003, "time": 0.36419076919555665, "epoch": 4, "memory": 20334, "step": 18874} +{"lr": 0.0009460482294732422, "data_time": 0.0015221118927001953, "grad_norm": 0.019056403124704958, "loss": 0.6883776783943176, "time": 0.43550407886505127, "epoch": 4, "memory": 20334, "step": 18974} +{"lr": 0.0009460482294732422, "data_time": 0.0015488624572753905, "grad_norm": 0.013475373107939958, "loss": 0.6881610572338104, "time": 0.3869898796081543, "epoch": 4, "memory": 20334, "step": 19074} +{"lr": 0.0009460482294732422, "data_time": 0.0014147520065307616, "grad_norm": 0.02170384577475488, "loss": 0.6862749814987182, "time": 0.3704317331314087, "epoch": 4, "memory": 20334, "step": 19174} +{"lr": 0.0009460482294732422, "data_time": 0.0015421390533447265, "grad_norm": 0.008144068927504123, "loss": 0.6860386788845062, "time": 0.441575026512146, "epoch": 4, "memory": 20334, "step": 19274} +{"lr": 0.0009460482294732422, "data_time": 0.001698470115661621, "grad_norm": 0.016787175508216023, "loss": 0.682527631521225, "time": 0.4142254114151001, "epoch": 4, "memory": 20334, "step": 19374} +{"lr": 0.0009460482294732422, "data_time": 0.0017727136611938477, "grad_norm": 0.018197024380788206, "loss": 0.6867428302764893, "time": 0.43469235897064207, "epoch": 4, "memory": 20334, "step": 19474} +{"lr": 0.0009460482294732422, "data_time": 0.001530003547668457, "grad_norm": 0.01373957390896976, "loss": 0.6874535799026489, "time": 0.4444230794906616, "epoch": 4, "memory": 20334, "step": 19574} +{"lr": 0.0009460482294732422, "data_time": 0.0014381647109985352, "grad_norm": 0.013393371040001512, "loss": 0.6846566259860992, "time": 0.43920044898986815, "epoch": 4, "memory": 20334, "step": 19674} +{"lr": 0.0009460482294732422, "data_time": 0.001518559455871582, "grad_norm": 0.020969915320165455, "loss": 0.6840063452720642, "time": 0.41179990768432617, "epoch": 4, "memory": 20334, "step": 19774} +{"lr": 0.0009460482294732422, "data_time": 0.0022547006607055663, "grad_norm": 0.013067789108026773, "loss": 0.684538459777832, "time": 0.4177358627319336, "epoch": 4, "memory": 20334, "step": 19874} +{"lr": 0.0009460482294732422, "data_time": 0.0016644716262817383, "grad_norm": 0.01624538628384471, "loss": 0.6869847476482391, "time": 0.40649797916412356, "epoch": 4, "memory": 20334, "step": 19974} +{"lr": 0.0009460482294732422, "data_time": 0.001677536964416504, "grad_norm": 0.020637770625762643, "loss": 0.6928331136703492, "time": 0.4554502248764038, "epoch": 4, "memory": 20334, "step": 20074} +{"lr": 0.0009460482294732422, "data_time": 0.0015866994857788087, "grad_norm": 0.020646632369607686, "loss": 0.6828271448612213, "time": 0.4271300554275513, "epoch": 4, "memory": 20334, "step": 20174} +{"lr": 0.0009460482294732422, "data_time": 0.0021440267562866213, "grad_norm": 0.014571310882456601, "loss": 0.6862425327301025, "time": 0.4084958076477051, "epoch": 4, "memory": 20334, "step": 20274} +{"lr": 0.0009460482294732422, "data_time": 0.0016492605209350586, "grad_norm": 0.021540999668650328, "loss": 0.6815922021865845, "time": 0.4297044038772583, "epoch": 4, "memory": 20334, "step": 20374} +{"lr": 0.0009460482294732422, "data_time": 0.0016182899475097657, "grad_norm": 0.02073707969393581, "loss": 0.6867773354053497, "time": 0.405154824256897, "epoch": 4, "memory": 20334, "step": 20474} +{"lr": 0.0009460482294732422, "data_time": 0.001638031005859375, "grad_norm": 0.020680293254554272, "loss": 0.6845305323600769, "time": 0.42842884063720704, "epoch": 4, "memory": 20334, "step": 20574} +{"lr": 0.0009460482294732422, "data_time": 0.001453065872192383, "grad_norm": 0.01812839702470228, "loss": 0.6863468766212464, "time": 0.503788423538208, "epoch": 4, "memory": 20334, "step": 20674} +{"lr": 0.0009460482294732422, "data_time": 0.0016042232513427735, "grad_norm": 0.017933058692142366, "loss": 0.688391637802124, "time": 0.4160242319107056, "epoch": 4, "memory": 20334, "step": 20774} +{"lr": 0.0009460482294732422, "data_time": 0.0022802114486694335, "grad_norm": 0.01980747119523585, "loss": 0.6880821645259857, "time": 0.42278614044189455, "epoch": 4, "memory": 20334, "step": 20874} +{"lr": 0.0009460482294732422, "data_time": 0.0011942863464355468, "grad_norm": 0.023618259513750673, "loss": 0.6876172184944153, "time": 0.40677585601806643, "epoch": 4, "memory": 20334, "step": 20974} +{"lr": 0.0009460482294732422, "data_time": 0.0017892122268676758, "grad_norm": 0.013236486550886184, "loss": 0.6890556931495666, "time": 0.47781803607940676, "epoch": 4, "memory": 20334, "step": 21074} +{"lr": 0.0009460482294732422, "data_time": 0.0021160602569580077, "grad_norm": 0.014703218964859844, "loss": 0.6856659710407257, "time": 0.4252608060836792, "epoch": 4, "memory": 20334, "step": 21174} +{"lr": 0.0009460482294732422, "data_time": 0.0019513845443725586, "grad_norm": 0.012840284616686404, "loss": 0.6858396232128143, "time": 0.40274677276611326, "epoch": 4, "memory": 20334, "step": 21274} +{"lr": 0.0009460482294732422, "data_time": 0.001669168472290039, "grad_norm": 0.016186735965311526, "loss": 0.6872314870357513, "time": 0.4350972890853882, "epoch": 4, "memory": 20334, "step": 21374} +{"lr": 0.0009460482294732422, "data_time": 0.0018370628356933593, "grad_norm": 0.011378039745613933, "loss": 0.6922192335128784, "time": 0.3972689390182495, "epoch": 4, "memory": 20334, "step": 21474} +{"lr": 0.0009460482294732422, "data_time": 0.0018207311630249023, "grad_norm": 0.021262407768517733, "loss": 0.6886372029781341, "time": 0.4285183668136597, "epoch": 4, "memory": 20334, "step": 21574} +{"lr": 0.0009460482294732422, "data_time": 0.0016820907592773437, "grad_norm": 0.0167135022347793, "loss": 0.6874531209468842, "time": 0.4497814655303955, "epoch": 4, "memory": 20334, "step": 21674} +{"lr": 0.0009460482294732422, "data_time": 0.0018252849578857422, "grad_norm": 0.045796560030430554, "loss": 0.6876559019088745, "time": 0.4027076005935669, "epoch": 4, "memory": 20334, "step": 21774} +{"lr": 0.0009460482294732422, "data_time": 0.0016832828521728515, "grad_norm": 0.016230330849066377, "loss": 0.6880554735660553, "time": 0.46181378364562986, "epoch": 4, "memory": 20334, "step": 21874} +{"lr": 0.0009460482294732422, "data_time": 0.001659727096557617, "grad_norm": 0.02607950558885932, "loss": 0.6868024289608001, "time": 0.4059682130813599, "epoch": 4, "memory": 20334, "step": 21974} +{"lr": 0.0009460482294732422, "data_time": 0.0020300626754760744, "grad_norm": 0.02645618636161089, "loss": 0.6936902642250061, "time": 0.49038636684417725, "epoch": 4, "memory": 20334, "step": 22074} +{"lr": 0.0009460482294732422, "data_time": 0.0020058155059814453, "grad_norm": 0.013353989459574223, "loss": 0.6905680000782013, "time": 0.3854936361312866, "epoch": 4, "memory": 20334, "step": 22174} +{"lr": 0.0009460482294732422, "data_time": 0.002733778953552246, "grad_norm": 0.019059557921718807, "loss": 0.6884543716907501, "time": 0.4084145545959473, "epoch": 4, "memory": 20334, "step": 22274} +{"lr": 0.0009460482294732422, "data_time": 0.001704859733581543, "grad_norm": 0.010697181103751063, "loss": 0.6882004916667939, "time": 0.42439606189727785, "epoch": 4, "memory": 20334, "step": 22374} +{"lr": 0.0009460482294732422, "data_time": 0.002095484733581543, "grad_norm": 0.016525850305333732, "loss": 0.6870546221733094, "time": 0.406258749961853, "epoch": 4, "memory": 20334, "step": 22474} +{"lr": 0.0009460482294732422, "data_time": 0.002139568328857422, "grad_norm": 0.01720315821003169, "loss": 0.687104606628418, "time": 0.4248741626739502, "epoch": 4, "memory": 20334, "step": 22574} +{"lr": 0.0009460482294732422, "data_time": 0.001678633689880371, "grad_norm": 0.01700702402740717, "loss": 0.6894726574420929, "time": 0.41515426635742186, "epoch": 4, "memory": 20334, "step": 22674} +{"lr": 0.0009460482294732422, "data_time": 0.001830887794494629, "grad_norm": 0.017228703980799766, "loss": 0.682729709148407, "time": 0.4291451215744019, "epoch": 4, "memory": 20334, "step": 22774} +{"lr": 0.0009460482294732422, "data_time": 0.0016281604766845703, "grad_norm": 0.016697622137144208, "loss": 0.6920957148075104, "time": 0.45853500366210936, "epoch": 4, "memory": 20334, "step": 22874} +{"lr": 0.0009460482294732422, "data_time": 0.0018860340118408204, "grad_norm": 0.016866310266777873, "loss": 0.6857346296310425, "time": 0.4406775712966919, "epoch": 4, "memory": 20334, "step": 22974} +{"accuracy/top1": 100.0, "data_time": 0.3747745354970296, "time": 1.0004582669999864, "step": 4} +{"lr": 0.0009054634122155991, "data_time": 0.0016640186309814452, "grad_norm": 0.013474358816165478, "loss": 0.6892355859279633, "time": 0.46261348724365237, "epoch": 5, "memory": 20334, "step": 23132} +{"lr": 0.0009054634122155991, "data_time": 0.0018067359924316406, "grad_norm": 0.013750681735109538, "loss": 0.6833201110363006, "time": 0.5576366901397705, "epoch": 5, "memory": 20334, "step": 23232} +{"lr": 0.0009054634122155991, "data_time": 0.0017942667007446289, "grad_norm": 0.015265438344795258, "loss": 0.6861961781978607, "time": 0.4597959041595459, "epoch": 5, "memory": 20334, "step": 23332} +{"lr": 0.0009054634122155991, "data_time": 0.0015351295471191407, "grad_norm": 0.014767507091164589, "loss": 0.6823357164859771, "time": 0.418464732170105, "epoch": 5, "memory": 20334, "step": 23432} +{"lr": 0.0009054634122155991, "data_time": 0.0015769720077514649, "grad_norm": 0.01598195082042366, "loss": 0.6872386693954468, "time": 0.482387113571167, "epoch": 5, "memory": 20334, "step": 23532} +{"lr": 0.0009054634122155991, "data_time": 0.001737356185913086, "grad_norm": 0.016936160705517976, "loss": 0.6869838953018188, "time": 0.43592736721038816, "epoch": 5, "memory": 20334, "step": 23632} +{"lr": 0.0009054634122155991, "data_time": 0.001629638671875, "grad_norm": 0.014898409554734825, "loss": 0.6889171481132508, "time": 0.4727465152740479, "epoch": 5, "memory": 20334, "step": 23732} +{"lr": 0.0009054634122155991, "data_time": 0.0016838312149047852, "grad_norm": 0.017872030846774578, "loss": 0.6852244794368744, "time": 0.39304389953613283, "epoch": 5, "memory": 20334, "step": 23832} +{"lr": 0.0009054634122155991, "data_time": 0.0018454551696777343, "grad_norm": 0.015729168988764287, "loss": 0.6876865446567535, "time": 0.45937998294830323, "epoch": 5, "memory": 20334, "step": 23932} +{"lr": 0.0009054634122155991, "data_time": 0.0013447999954223633, "grad_norm": 0.02219186588190496, "loss": 0.6884686589241028, "time": 0.49438884258270266, "epoch": 5, "memory": 20334, "step": 24032} +{"lr": 0.0009054634122155991, "data_time": 0.001679396629333496, "grad_norm": 0.017965504340827464, "loss": 0.6851427257061005, "time": 0.41814842224121096, "epoch": 5, "memory": 20334, "step": 24132} +{"lr": 0.0009054634122155991, "data_time": 0.0015549182891845704, "grad_norm": 0.020059690158814193, "loss": 0.6893116295337677, "time": 0.4728104114532471, "epoch": 5, "memory": 20334, "step": 24232} +{"lr": 0.0009054634122155991, "data_time": 0.002398848533630371, "grad_norm": 0.012723038648255169, "loss": 0.6831086933612823, "time": 0.4514086484909058, "epoch": 5, "memory": 20334, "step": 24332} +{"lr": 0.0009054634122155991, "data_time": 0.0016026258468627929, "grad_norm": 0.015413007733877748, "loss": 0.6874089002609253, "time": 0.42252435684204104, "epoch": 5, "memory": 20334, "step": 24432} +{"lr": 0.0009054634122155991, "data_time": 0.0012562036514282226, "grad_norm": 0.027014520857483147, "loss": 0.6899824917316437, "time": 0.43594205379486084, "epoch": 5, "memory": 20334, "step": 24532} +{"lr": 0.0009054634122155991, "data_time": 0.0019109487533569337, "grad_norm": 0.015215306938625873, "loss": 0.6913561344146728, "time": 0.4070683240890503, "epoch": 5, "memory": 20334, "step": 24632} +{"lr": 0.0009054634122155991, "data_time": 0.0019548654556274412, "grad_norm": 0.014206824082066305, "loss": 0.6832255184650421, "time": 0.4208178758621216, "epoch": 5, "memory": 20334, "step": 24732} +{"lr": 0.0009054634122155991, "data_time": 0.0018374204635620117, "grad_norm": 0.016956628498155622, "loss": 0.6816364228725433, "time": 0.4055312633514404, "epoch": 5, "memory": 20334, "step": 24832} +{"lr": 0.0009054634122155991, "data_time": 0.001784205436706543, "grad_norm": 0.015278805885463953, "loss": 0.6831121444702148, "time": 0.4406696319580078, "epoch": 5, "memory": 20334, "step": 24932} +{"lr": 0.0009054634122155991, "data_time": 0.001952362060546875, "grad_norm": 0.016142445389414208, "loss": 0.6892522752285004, "time": 0.41704297065734863, "epoch": 5, "memory": 20334, "step": 25032} +{"lr": 0.0009054634122155991, "data_time": 0.0023880243301391602, "grad_norm": 0.010578433435875922, "loss": 0.6865203738212585, "time": 0.41465351581573484, "epoch": 5, "memory": 20334, "step": 25132} +{"lr": 0.0009054634122155991, "data_time": 0.0017028331756591796, "grad_norm": 0.014046089618932455, "loss": 0.6822193622589111, "time": 0.3870721340179443, "epoch": 5, "memory": 20334, "step": 25232} +{"lr": 0.0009054634122155991, "data_time": 0.0024978160858154298, "grad_norm": 0.011540906503796577, "loss": 0.6844033598899841, "time": 0.48096332550048826, "epoch": 5, "memory": 20334, "step": 25332} +{"lr": 0.0009054634122155991, "data_time": 0.0022904634475708007, "grad_norm": 0.02594159790314734, "loss": 0.6852489113807678, "time": 0.4656040906906128, "epoch": 5, "memory": 20334, "step": 25432} +{"lr": 0.0009054634122155991, "data_time": 0.002164196968078613, "grad_norm": 0.013253998104482889, "loss": 0.686679071187973, "time": 0.40122935771942136, "epoch": 5, "memory": 20334, "step": 25532} +{"lr": 0.0009054634122155991, "data_time": 0.0025037527084350586, "grad_norm": 0.01694425903260708, "loss": 0.6807831466197968, "time": 0.39859178066253664, "epoch": 5, "memory": 20334, "step": 25632} +{"lr": 0.0009054634122155991, "data_time": 0.0017140865325927734, "grad_norm": 0.010635975142940879, "loss": 0.688271290063858, "time": 0.4565638780593872, "epoch": 5, "memory": 20334, "step": 25732} +{"lr": 0.0009054634122155991, "data_time": 0.0015695810317993164, "grad_norm": 0.013788261217996478, "loss": 0.6906117022037506, "time": 0.39268429279327394, "epoch": 5, "memory": 20334, "step": 25832} +{"lr": 0.0009054634122155991, "data_time": 0.0020319461822509766, "grad_norm": 0.014317479357123375, "loss": 0.6879059672355652, "time": 0.4240111351013184, "epoch": 5, "memory": 20334, "step": 25932} +{"lr": 0.0009054634122155991, "data_time": 0.002335071563720703, "grad_norm": 0.026299254782497884, "loss": 0.6909553349018097, "time": 0.49040796756744387, "epoch": 5, "memory": 20334, "step": 26032} +{"lr": 0.0009054634122155991, "data_time": 0.0017626047134399413, "grad_norm": 0.019424308091402054, "loss": 0.6844892084598542, "time": 0.4185448169708252, "epoch": 5, "memory": 20334, "step": 26132} +{"lr": 0.0009054634122155991, "data_time": 0.0016578197479248046, "grad_norm": 0.02198729747906327, "loss": 0.6875029325485229, "time": 0.44490096569061277, "epoch": 5, "memory": 20334, "step": 26232} +{"lr": 0.0009054634122155991, "data_time": 0.0014395952224731446, "grad_norm": 0.026266485848464072, "loss": 0.6841505408287049, "time": 0.3759397745132446, "epoch": 5, "memory": 20334, "step": 26332} +{"lr": 0.0009054634122155991, "data_time": 0.0014629602432250977, "grad_norm": 0.01955609628930688, "loss": 0.6860485792160034, "time": 0.5077569246292114, "epoch": 5, "memory": 20334, "step": 26432} +{"lr": 0.0009054634122155991, "data_time": 0.0021701335906982424, "grad_norm": 0.01715301611693576, "loss": 0.686602920293808, "time": 0.5131395101547241, "epoch": 5, "memory": 20334, "step": 26532} +{"lr": 0.0009054634122155991, "data_time": 0.001564621925354004, "grad_norm": 0.009940954123158008, "loss": 0.6881152570247651, "time": 0.5038388490676879, "epoch": 5, "memory": 20334, "step": 26632} +{"lr": 0.0009054634122155991, "data_time": 0.0022216081619262696, "grad_norm": 0.016357232863083482, "loss": 0.6874220490455627, "time": 0.4277378559112549, "epoch": 5, "memory": 20334, "step": 26732} +{"lr": 0.0009054634122155991, "data_time": 0.0025137901306152345, "grad_norm": 0.019747086614370347, "loss": 0.6891478955745697, "time": 0.4235680103302002, "epoch": 5, "memory": 20334, "step": 26832} +{"lr": 0.0009054634122155991, "data_time": 0.0020124197006225588, "grad_norm": 0.016667066141963005, "loss": 0.683452981710434, "time": 0.4664349317550659, "epoch": 5, "memory": 20334, "step": 26932} +{"lr": 0.0009054634122155991, "data_time": 0.002366971969604492, "grad_norm": 0.01884008216438815, "loss": 0.6852039217948913, "time": 0.39306495189666746, "epoch": 5, "memory": 20334, "step": 27032} +{"lr": 0.0009054634122155991, "data_time": 0.0014747142791748046, "grad_norm": 0.013768781418912112, "loss": 0.6881043314933777, "time": 0.38590786457061765, "epoch": 5, "memory": 20334, "step": 27132} +{"lr": 0.0009054634122155991, "data_time": 0.0015051841735839843, "grad_norm": 0.012897312315180898, "loss": 0.690074110031128, "time": 0.4127305030822754, "epoch": 5, "memory": 20334, "step": 27232} +{"lr": 0.0009054634122155991, "data_time": 0.0016739368438720703, "grad_norm": 0.016327945189550518, "loss": 0.6870486497879028, "time": 0.4177191972732544, "epoch": 5, "memory": 20334, "step": 27332} +{"lr": 0.0009054634122155991, "data_time": 0.0015619516372680664, "grad_norm": 0.02242490865755826, "loss": 0.6888010859489441, "time": 0.40947961807250977, "epoch": 5, "memory": 20334, "step": 27432} +{"lr": 0.0009054634122155991, "data_time": 0.0016671419143676758, "grad_norm": 0.01205872044665739, "loss": 0.6856835782527924, "time": 0.44570968151092527, "epoch": 5, "memory": 20334, "step": 27532} +{"lr": 0.0009054634122155991, "data_time": 0.0016495227813720704, "grad_norm": 0.01477744576986879, "loss": 0.6921510756015777, "time": 0.5403522729873658, "epoch": 5, "memory": 20334, "step": 27632} +{"lr": 0.0009054634122155991, "data_time": 0.002578067779541016, "grad_norm": 0.024773371312767266, "loss": 0.6809918463230134, "time": 0.40028655529022217, "epoch": 5, "memory": 20334, "step": 27732} +{"lr": 0.0009054634122155991, "data_time": 0.0025525808334350584, "grad_norm": 0.022941923746839166, "loss": 0.6822918355464935, "time": 0.4442912101745605, "epoch": 5, "memory": 20334, "step": 27832} +{"lr": 0.0009054634122155991, "data_time": 0.0022899866104125976, "grad_norm": 0.01740312119945884, "loss": 0.6865120649337768, "time": 0.45723905563354494, "epoch": 5, "memory": 20334, "step": 27932} +{"lr": 0.0009054634122155991, "data_time": 0.002504301071166992, "grad_norm": 0.01854991556610912, "loss": 0.6900014460086823, "time": 0.42259061336517334, "epoch": 5, "memory": 20334, "step": 28032} +{"lr": 0.0009054634122155991, "data_time": 0.0023766517639160155, "grad_norm": 0.00993919875472784, "loss": 0.6916679263114929, "time": 0.4421380996704102, "epoch": 5, "memory": 20334, "step": 28132} +{"lr": 0.0009054634122155991, "data_time": 0.0023939847946166993, "grad_norm": 0.014894899516366421, "loss": 0.688961535692215, "time": 0.4703490972518921, "epoch": 5, "memory": 20334, "step": 28232} +{"lr": 0.0009054634122155991, "data_time": 0.002553510665893555, "grad_norm": 0.018669087439775467, "loss": 0.6927083015441895, "time": 0.4199769735336304, "epoch": 5, "memory": 20334, "step": 28332} +{"lr": 0.0009054634122155991, "data_time": 0.00251924991607666, "grad_norm": 0.01581809011986479, "loss": 0.6880093038082122, "time": 0.4258387804031372, "epoch": 5, "memory": 20334, "step": 28432} +{"lr": 0.0009054634122155991, "data_time": 0.0022903203964233397, "grad_norm": 0.019251885660924017, "loss": 0.6875609874725341, "time": 0.4850937843322754, "epoch": 5, "memory": 20334, "step": 28532} +{"lr": 0.0009054634122155991, "data_time": 0.001758575439453125, "grad_norm": 0.020095613319426775, "loss": 0.6832868814468384, "time": 0.45992026329040525, "epoch": 5, "memory": 20334, "step": 28632} +{"lr": 0.0009054634122155991, "data_time": 0.0021857500076293947, "grad_norm": 0.017635805904865264, "loss": 0.6856019020080566, "time": 0.43054590225219724, "epoch": 5, "memory": 20334, "step": 28732} +{"accuracy/top1": 100.0, "data_time": 0.3672407203250461, "time": 0.9940706094106039, "step": 5} +{"lr": 0.0008550178566873411, "data_time": 0.0016501426696777343, "grad_norm": 0.010740619990974665, "loss": 0.6819038808345794, "time": 0.4215155839920044, "epoch": 6, "memory": 20334, "step": 28890} +{"lr": 0.0008550178566873411, "data_time": 0.0017414093017578125, "grad_norm": 0.014249167079105974, "loss": 0.6858422577381134, "time": 0.4054927110671997, "epoch": 6, "memory": 20334, "step": 28990} +{"lr": 0.0008550178566873411, "data_time": 0.002213859558105469, "grad_norm": 0.018390496261417866, "loss": 0.6852671027183532, "time": 0.40395450592041016, "epoch": 6, "memory": 20334, "step": 29090} +{"lr": 0.0008550178566873411, "data_time": 0.00170443058013916, "grad_norm": 0.017776680213864894, "loss": 0.6879237771034241, "time": 0.38143167495727537, "epoch": 6, "memory": 20334, "step": 29190} +{"lr": 0.0008550178566873411, "data_time": 0.002321171760559082, "grad_norm": 0.017277160147204994, "loss": 0.6869894623756408, "time": 0.4530991315841675, "epoch": 6, "memory": 20334, "step": 29290} +{"lr": 0.0008550178566873411, "data_time": 0.0020491361618041994, "grad_norm": 0.017743013601284476, "loss": 0.6863463342189788, "time": 0.38599770069122313, "epoch": 6, "memory": 20334, "step": 29390} +{"lr": 0.0008550178566873411, "data_time": 0.0020192146301269533, "grad_norm": 0.017861474899109452, "loss": 0.6891350388526917, "time": 0.42172553539276125, "epoch": 6, "memory": 20334, "step": 29490} +{"lr": 0.0008550178566873411, "data_time": 0.0017431259155273437, "grad_norm": 0.01864182265708223, "loss": 0.685890394449234, "time": 0.3811058044433594, "epoch": 6, "memory": 20334, "step": 29590} +{"lr": 0.0008550178566873411, "data_time": 0.0030803918838500977, "grad_norm": 0.015005982713773847, "loss": 0.6859933257102966, "time": 0.43090860843658446, "epoch": 6, "memory": 20334, "step": 29690} +{"lr": 0.0008550178566873411, "data_time": 0.00235443115234375, "grad_norm": 0.015143247367814184, "loss": 0.6870425939559937, "time": 0.45867719650268557, "epoch": 6, "memory": 20334, "step": 29790} +{"lr": 0.0008550178566873411, "data_time": 0.002558255195617676, "grad_norm": 0.021923623932525515, "loss": 0.686844676733017, "time": 0.4720849275588989, "epoch": 6, "memory": 20334, "step": 29890} +{"lr": 0.0008550178566873411, "data_time": 0.003036165237426758, "grad_norm": 0.018959448114037512, "loss": 0.6872752964496612, "time": 0.42896265983581544, "epoch": 6, "memory": 20334, "step": 29990} +{"lr": 0.0008550178566873411, "data_time": 0.002033638954162598, "grad_norm": 0.014010517811402678, "loss": 0.6890656590461731, "time": 0.41484718322753905, "epoch": 6, "memory": 20334, "step": 30090} +{"lr": 0.0008550178566873411, "data_time": 0.0024161577224731446, "grad_norm": 0.012129091611132025, "loss": 0.689091432094574, "time": 0.36630136966705323, "epoch": 6, "memory": 20334, "step": 30190} +{"lr": 0.0008550178566873411, "data_time": 0.002329564094543457, "grad_norm": 0.01896277009509504, "loss": 0.6904054403305053, "time": 0.4629948616027832, "epoch": 6, "memory": 20334, "step": 30290} +{"lr": 0.0008550178566873411, "data_time": 0.002977895736694336, "grad_norm": 0.014263168442994356, "loss": 0.6873790085315704, "time": 0.43889336585998534, "epoch": 6, "memory": 20334, "step": 30390} +{"lr": 0.0008550178566873411, "data_time": 0.0018604755401611327, "grad_norm": 0.018118901655543595, "loss": 0.6866663873195649, "time": 0.3718499898910522, "epoch": 6, "memory": 20334, "step": 30490} +{"lr": 0.0008550178566873411, "data_time": 0.0017019271850585937, "grad_norm": 0.016365794534794986, "loss": 0.6880724191665649, "time": 0.3911444187164307, "epoch": 6, "memory": 20334, "step": 30590} +{"lr": 0.0008550178566873411, "data_time": 0.0026001930236816406, "grad_norm": 0.01800455499906093, "loss": 0.6916402876377106, "time": 0.41898887157440184, "epoch": 6, "memory": 20334, "step": 30690} +{"lr": 0.0008550178566873411, "data_time": 0.0021966218948364256, "grad_norm": 0.019803097785916178, "loss": 0.6882297456264496, "time": 0.3862194776535034, "epoch": 6, "memory": 20334, "step": 30790} +{"lr": 0.0008550178566873411, "data_time": 0.003175950050354004, "grad_norm": 0.014639736915705725, "loss": 0.6861068546772003, "time": 0.4259798049926758, "epoch": 6, "memory": 20334, "step": 30890} +{"lr": 0.0008550178566873411, "data_time": 0.00249025821685791, "grad_norm": 0.0161762215255294, "loss": 0.6831121385097504, "time": 0.4316707134246826, "epoch": 6, "memory": 20334, "step": 30990} +{"lr": 0.0008550178566873411, "data_time": 0.0024130821228027345, "grad_norm": 0.013457054318860174, "loss": 0.6833941161632537, "time": 0.37877702713012695, "epoch": 6, "memory": 20334, "step": 31090} +{"lr": 0.0008550178566873411, "data_time": 0.0026801347732543944, "grad_norm": 0.01865406068973243, "loss": 0.6831690788269043, "time": 0.49864356517791747, "epoch": 6, "memory": 20334, "step": 31190} +{"lr": 0.0008550178566873411, "data_time": 0.0023834228515625, "grad_norm": 0.016033722483552993, "loss": 0.6868420302867889, "time": 0.4378886461257935, "epoch": 6, "memory": 20334, "step": 31290} +{"lr": 0.0008550178566873411, "data_time": 0.0024738073348999023, "grad_norm": 0.015218574134632946, "loss": 0.6876056492328644, "time": 0.42201330661773684, "epoch": 6, "memory": 20334, "step": 31390} +{"lr": 0.0008550178566873411, "data_time": 0.0022174835205078123, "grad_norm": 0.015551392734050751, "loss": 0.6883639395236969, "time": 0.3924338102340698, "epoch": 6, "memory": 20334, "step": 31490} +{"lr": 0.0008550178566873411, "data_time": 0.0023947238922119142, "grad_norm": 0.014032244007103145, "loss": 0.6840821266174316, "time": 0.3723433017730713, "epoch": 6, "memory": 20334, "step": 31590} +{"lr": 0.0008550178566873411, "data_time": 0.002815508842468262, "grad_norm": 0.024828650563722476, "loss": 0.6863940179347991, "time": 0.4717421054840088, "epoch": 6, "memory": 20334, "step": 31690} +{"lr": 0.0008550178566873411, "data_time": 0.0021286725997924803, "grad_norm": 0.014141946626477876, "loss": 0.6857439279556274, "time": 0.40163042545318606, "epoch": 6, "memory": 20334, "step": 31790} +{"lr": 0.0008550178566873411, "data_time": 0.0019434213638305664, "grad_norm": 0.02141375997744035, "loss": 0.6920516550540924, "time": 0.40711610317230223, "epoch": 6, "memory": 20334, "step": 31890} +{"lr": 0.0008550178566873411, "data_time": 0.0025219440460205076, "grad_norm": 0.012545603339094668, "loss": 0.6926559031009674, "time": 0.4218295097351074, "epoch": 6, "memory": 20334, "step": 31990} +{"lr": 0.0008550178566873411, "data_time": 0.002497553825378418, "grad_norm": 0.019899757322855295, "loss": 0.682876181602478, "time": 0.38234996795654297, "epoch": 6, "memory": 20334, "step": 32090} +{"lr": 0.0008550178566873411, "data_time": 0.0017943382263183594, "grad_norm": 0.015066596562974155, "loss": 0.689560842514038, "time": 0.4505463123321533, "epoch": 6, "memory": 20334, "step": 32190} +{"lr": 0.0008550178566873411, "data_time": 0.002192401885986328, "grad_norm": 0.01855681617744267, "loss": 0.6845539927482605, "time": 0.4410197019577026, "epoch": 6, "memory": 20334, "step": 32290} +{"lr": 0.0008550178566873411, "data_time": 0.0016761541366577149, "grad_norm": 0.013517347583547234, "loss": 0.6895796895027161, "time": 0.37102541923522947, "epoch": 6, "memory": 20334, "step": 32390} +{"lr": 0.0008550178566873411, "data_time": 0.0021207571029663087, "grad_norm": 0.022510111820884048, "loss": 0.6930544078350067, "time": 0.4105675458908081, "epoch": 6, "memory": 20334, "step": 32490} +{"lr": 0.0008550178566873411, "data_time": 0.0024774789810180662, "grad_norm": 0.013739681034348906, "loss": 0.6867192447185516, "time": 0.40883698463439944, "epoch": 6, "memory": 20334, "step": 32590} +{"lr": 0.0008550178566873411, "data_time": 0.0016768455505371093, "grad_norm": 0.02028878660639748, "loss": 0.6895137846469879, "time": 0.41194050312042235, "epoch": 6, "memory": 20334, "step": 32690} +{"lr": 0.0008550178566873411, "data_time": 0.0028486013412475585, "grad_norm": 0.017298220843076705, "loss": 0.6861627876758576, "time": 0.3971180200576782, "epoch": 6, "memory": 20334, "step": 32790} +{"lr": 0.0008550178566873411, "data_time": 0.001663064956665039, "grad_norm": 0.0246057340875268, "loss": 0.6876162230968476, "time": 0.3676790952682495, "epoch": 6, "memory": 20334, "step": 32890} +{"lr": 0.0008550178566873411, "data_time": 0.0016487598419189452, "grad_norm": 0.01762840843293816, "loss": 0.6918885767459869, "time": 0.43290071487426757, "epoch": 6, "memory": 20334, "step": 32990} +{"lr": 0.0008550178566873411, "data_time": 0.0028705835342407227, "grad_norm": 0.01391700147651136, "loss": 0.6933708608150482, "time": 0.40606689453125, "epoch": 6, "memory": 20334, "step": 33090} +{"lr": 0.0008550178566873411, "data_time": 0.0019050836563110352, "grad_norm": 0.012038194620981813, "loss": 0.6841579973697662, "time": 0.404231858253479, "epoch": 6, "memory": 20334, "step": 33190} +{"lr": 0.0008550178566873411, "data_time": 0.0018760204315185548, "grad_norm": 0.017551461001858115, "loss": 0.6854634702205658, "time": 0.38661041259765627, "epoch": 6, "memory": 20334, "step": 33290} +{"lr": 0.0008550178566873411, "data_time": 0.0020201444625854493, "grad_norm": 0.01785781476646662, "loss": 0.6892797350883484, "time": 0.3988152027130127, "epoch": 6, "memory": 20334, "step": 33390} +{"lr": 0.0008550178566873411, "data_time": 0.002326774597167969, "grad_norm": 0.01143301121192053, "loss": 0.6911220073699951, "time": 0.4070104122161865, "epoch": 6, "memory": 20334, "step": 33490} +{"lr": 0.0008550178566873411, "data_time": 0.002182292938232422, "grad_norm": 0.014409385481849312, "loss": 0.6864995062351227, "time": 0.3734931230545044, "epoch": 6, "memory": 20334, "step": 33590} +{"lr": 0.0008550178566873411, "data_time": 0.002413010597229004, "grad_norm": 0.018322595208883286, "loss": 0.6882865190505981, "time": 0.39390385150909424, "epoch": 6, "memory": 20334, "step": 33690} +{"lr": 0.0008550178566873411, "data_time": 0.0022130489349365236, "grad_norm": 0.0244600351434201, "loss": 0.6902685821056366, "time": 0.46015329360961915, "epoch": 6, "memory": 20334, "step": 33790} +{"lr": 0.0008550178566873411, "data_time": 0.002182483673095703, "grad_norm": 0.025804482540115713, "loss": 0.6844572007656098, "time": 0.42754712104797366, "epoch": 6, "memory": 20334, "step": 33890} +{"lr": 0.0008550178566873411, "data_time": 0.002279186248779297, "grad_norm": 0.015815614216262475, "loss": 0.6855115056037903, "time": 0.4512783527374268, "epoch": 6, "memory": 20334, "step": 33990} +{"lr": 0.0008550178566873411, "data_time": 0.0018991470336914063, "grad_norm": 0.01853583964984864, "loss": 0.6862693428993225, "time": 0.4971250057220459, "epoch": 6, "memory": 20334, "step": 34090} +{"lr": 0.0008550178566873411, "data_time": 0.0023725271224975587, "grad_norm": 0.011092688061762602, "loss": 0.6905861556529999, "time": 0.4244656562805176, "epoch": 6, "memory": 20334, "step": 34190} +{"lr": 0.0008550178566873411, "data_time": 0.0028261423110961916, "grad_norm": 0.016301624092739074, "loss": 0.6868258774280548, "time": 0.400516939163208, "epoch": 6, "memory": 20334, "step": 34290} +{"lr": 0.0008550178566873411, "data_time": 0.0018138408660888672, "grad_norm": 0.01195098765892908, "loss": 0.6848946690559388, "time": 0.5971655130386353, "epoch": 6, "memory": 20334, "step": 34390} +{"lr": 0.0008550178566873411, "data_time": 0.0017888784408569337, "grad_norm": 0.009971192036755384, "loss": 0.6855127394199372, "time": 0.36561877727508546, "epoch": 6, "memory": 20334, "step": 34490} +{"accuracy/top1": 100.0, "data_time": 0.37670813666449654, "time": 1.0326961941189237, "step": 6} +{"lr": 0.0007959536998847743, "data_time": 0.002660822868347168, "grad_norm": 0.008818201615940779, "loss": 0.686515724658966, "time": 0.3914986848831177, "epoch": 7, "memory": 20334, "step": 34648} +{"lr": 0.0007959536998847743, "data_time": 0.002326178550720215, "grad_norm": 0.017777158366516232, "loss": 0.6882440745830536, "time": 0.39154982566833496, "epoch": 7, "memory": 20334, "step": 34748} +{"lr": 0.0007959536998847743, "data_time": 0.0028423070907592773, "grad_norm": 0.017145266756415367, "loss": 0.6869498729705811, "time": 0.4006332874298096, "epoch": 7, "memory": 20334, "step": 34848} +{"lr": 0.0007959536998847743, "data_time": 0.0026952028274536133, "grad_norm": 0.020782969333231448, "loss": 0.6846798479557037, "time": 0.48883905410766604, "epoch": 7, "memory": 20334, "step": 34948} +{"lr": 0.0007959536998847743, "data_time": 0.002477884292602539, "grad_norm": 0.015397327730897813, "loss": 0.6860618412494659, "time": 0.42564547061920166, "epoch": 7, "memory": 20334, "step": 35048} +{"lr": 0.0007959536998847743, "data_time": 0.0030760765075683594, "grad_norm": 0.017429115390405057, "loss": 0.6871620416641235, "time": 0.42325582504272463, "epoch": 7, "memory": 20334, "step": 35148} +{"lr": 0.0007959536998847743, "data_time": 0.001963019371032715, "grad_norm": 0.014444156154058873, "loss": 0.6850408792495728, "time": 0.4025650262832642, "epoch": 7, "memory": 20334, "step": 35248} +{"lr": 0.0007959536998847743, "data_time": 0.0023630380630493162, "grad_norm": 0.012954378291033209, "loss": 0.6875571370124817, "time": 1.0924967527389526, "epoch": 7, "memory": 20334, "step": 35348} +{"lr": 0.0007959536998847743, "data_time": 0.0024686098098754884, "grad_norm": 0.01838813836220652, "loss": 0.6845176458358765, "time": 0.3736931085586548, "epoch": 7, "memory": 20334, "step": 35448} +{"lr": 0.0007959536998847743, "data_time": 0.002348494529724121, "grad_norm": 0.025828354945406317, "loss": 0.6839582920074463, "time": 0.3978057861328125, "epoch": 7, "memory": 20334, "step": 35548} +{"lr": 0.0007959536998847743, "data_time": 0.0025232315063476564, "grad_norm": 0.018405025405809284, "loss": 0.6846412360668183, "time": 0.38733506202697754, "epoch": 7, "memory": 20334, "step": 35648} +{"lr": 0.0007959536998847743, "data_time": 0.0027555227279663086, "grad_norm": 0.024616753216832876, "loss": 0.6853302597999573, "time": 0.4038097858428955, "epoch": 7, "memory": 20334, "step": 35748} +{"lr": 0.0007959536998847743, "data_time": 0.0018779277801513673, "grad_norm": 0.01780391177162528, "loss": 0.6869785249233246, "time": 0.43698692321777344, "epoch": 7, "memory": 20334, "step": 35848} +{"lr": 0.0007959536998847743, "data_time": 0.0030397653579711916, "grad_norm": 0.024214513879269362, "loss": 0.6843997478485108, "time": 0.39369449615478513, "epoch": 7, "memory": 20334, "step": 35948} +{"lr": 0.0007959536998847743, "data_time": 0.0019219636917114258, "grad_norm": 0.008677515428280458, "loss": 0.6863185703754425, "time": 0.41775286197662354, "epoch": 7, "memory": 20334, "step": 36048} +{"lr": 0.0007959536998847743, "data_time": 0.0021116256713867186, "grad_norm": 0.016486375685781242, "loss": 0.6887442231178283, "time": 0.40992734432220457, "epoch": 7, "memory": 20334, "step": 36148} +{"lr": 0.0007959536998847743, "data_time": 0.0025710821151733398, "grad_norm": 0.019156168890185656, "loss": 0.6890320301055908, "time": 0.40306859016418456, "epoch": 7, "memory": 20334, "step": 36248} +{"lr": 0.0007959536998847743, "data_time": 0.0024237871170043946, "grad_norm": 0.01392492586746812, "loss": 0.6847045421600342, "time": 0.39427847862243653, "epoch": 7, "memory": 20334, "step": 36348} +{"lr": 0.0007959536998847743, "data_time": 0.0023135900497436523, "grad_norm": 0.024465126590803264, "loss": 0.6876348078250885, "time": 0.4035454511642456, "epoch": 7, "memory": 20334, "step": 36448} +{"lr": 0.0007959536998847743, "data_time": 0.002964615821838379, "grad_norm": 0.01351309111341834, "loss": 0.6860770404338836, "time": 0.4082977294921875, "epoch": 7, "memory": 20334, "step": 36548} +{"lr": 0.0007959536998847743, "data_time": 0.0023701667785644533, "grad_norm": 0.01681348168058321, "loss": 0.6863355100154876, "time": 0.36402928829193115, "epoch": 7, "memory": 20334, "step": 36648} +{"lr": 0.0007959536998847743, "data_time": 0.0023248910903930662, "grad_norm": 0.014404264773475006, "loss": 0.6853826642036438, "time": 0.40149860382080077, "epoch": 7, "memory": 20334, "step": 36748} +{"lr": 0.0007959536998847743, "data_time": 0.0020804405212402344, "grad_norm": 0.0163409007829614, "loss": 0.6862274885177613, "time": 0.3629596471786499, "epoch": 7, "memory": 20334, "step": 36848} +{"lr": 0.0007959536998847743, "data_time": 0.0017332315444946289, "grad_norm": 0.01984357591718435, "loss": 0.6822782814502716, "time": 0.3995716333389282, "epoch": 7, "memory": 20334, "step": 36948} +{"lr": 0.0007959536998847743, "data_time": 0.0022834300994873046, "grad_norm": 0.021805821312591435, "loss": 0.6903250575065613, "time": 0.4455390214920044, "epoch": 7, "memory": 20334, "step": 37048} +{"lr": 0.0007959536998847743, "data_time": 0.0025606870651245115, "grad_norm": 0.01806363833602518, "loss": 0.6891472280025482, "time": 0.4206372261047363, "epoch": 7, "memory": 20334, "step": 37148} +{"lr": 0.0007959536998847743, "data_time": 0.0016620397567749024, "grad_norm": 0.027936227293685077, "loss": 0.6872438967227936, "time": 0.38329253196716306, "epoch": 7, "memory": 20334, "step": 37248} +{"lr": 0.0007959536998847743, "data_time": 0.0017488479614257812, "grad_norm": 0.013940639398060738, "loss": 0.6853080928325653, "time": 0.4037781238555908, "epoch": 7, "memory": 20334, "step": 37348} +{"lr": 0.0007959536998847743, "data_time": 0.0024611949920654297, "grad_norm": 0.011252750316634775, "loss": 0.6853863060474396, "time": 0.39341936111450193, "epoch": 7, "memory": 20334, "step": 37448} +{"lr": 0.0007959536998847743, "data_time": 0.002257633209228516, "grad_norm": 0.01710908564273268, "loss": 0.6870525062084198, "time": 0.40363810062408445, "epoch": 7, "memory": 20334, "step": 37548} +{"lr": 0.0007959536998847743, "data_time": 0.0017189979553222656, "grad_norm": 0.021097694453783334, "loss": 0.6861779630184174, "time": 0.45919251441955566, "epoch": 7, "memory": 20334, "step": 37648} +{"lr": 0.0007959536998847743, "data_time": 0.0028703927993774412, "grad_norm": 0.018132218648679553, "loss": 0.691362488269806, "time": 0.38237967491149905, "epoch": 7, "memory": 20334, "step": 37748} +{"lr": 0.0007959536998847743, "data_time": 0.0025453090667724608, "grad_norm": 0.013848072150722146, "loss": 0.6873121321201324, "time": 0.4048325061798096, "epoch": 7, "memory": 20334, "step": 37848} +{"lr": 0.0007959536998847743, "data_time": 0.00212557315826416, "grad_norm": 0.0176771626342088, "loss": 0.6882309913635254, "time": 0.36730663776397704, "epoch": 7, "memory": 20334, "step": 37948} +{"lr": 0.0007959536998847743, "data_time": 0.0025019407272338866, "grad_norm": 0.01385203015524894, "loss": 0.6901014924049378, "time": 0.40636818408966063, "epoch": 7, "memory": 20334, "step": 38048} +{"lr": 0.0007959536998847743, "data_time": 0.0016633272171020508, "grad_norm": 0.01257945482066134, "loss": 0.6852293372154236, "time": 0.4259000539779663, "epoch": 7, "memory": 20334, "step": 38148} +{"lr": 0.0007959536998847743, "data_time": 0.0025504350662231444, "grad_norm": 0.01858710205415264, "loss": 0.6862414479255676, "time": 0.4188873052597046, "epoch": 7, "memory": 20334, "step": 38248} +{"lr": 0.0007959536998847743, "data_time": 0.0014546871185302734, "grad_norm": 0.021473313542082907, "loss": 0.6916547000408173, "time": 0.38711209297180177, "epoch": 7, "memory": 20334, "step": 38348} +{"lr": 0.0007959536998847743, "data_time": 0.0015260934829711913, "grad_norm": 0.02340732077136636, "loss": 0.6864424705505371, "time": 0.45596649646759035, "epoch": 7, "memory": 20334, "step": 38448} +{"lr": 0.0007959536998847743, "data_time": 0.001941514015197754, "grad_norm": 0.015903096972033383, "loss": 0.6854012429714202, "time": 0.3989575386047363, "epoch": 7, "memory": 20334, "step": 38548} +{"lr": 0.0007959536998847743, "data_time": 0.0016055583953857421, "grad_norm": 0.016300834156572817, "loss": 0.6881219208240509, "time": 0.39681453704833985, "epoch": 7, "memory": 20334, "step": 38648} +{"lr": 0.0007959536998847743, "data_time": 0.0015841484069824218, "grad_norm": 0.0161306019872427, "loss": 0.6845422387123108, "time": 0.40819635391235354, "epoch": 7, "memory": 20334, "step": 38748} +{"lr": 0.0007959536998847743, "data_time": 0.001938605308532715, "grad_norm": 0.01806643467862159, "loss": 0.687273907661438, "time": 0.3830427885055542, "epoch": 7, "memory": 20334, "step": 38848} +{"lr": 0.0007959536998847743, "data_time": 0.0025661230087280274, "grad_norm": 0.014198095246683806, "loss": 0.6872689723968506, "time": 0.4111591577529907, "epoch": 7, "memory": 20334, "step": 38948} +{"lr": 0.0007959536998847743, "data_time": 0.0017806529998779298, "grad_norm": 0.02024270365945995, "loss": 0.6882555246353149, "time": 0.39887540340423583, "epoch": 7, "memory": 20334, "step": 39048} +{"lr": 0.0007959536998847743, "data_time": 0.0017499685287475585, "grad_norm": 0.010240830824477598, "loss": 0.6851606249809266, "time": 0.41409270763397216, "epoch": 7, "memory": 20334, "step": 39148} +{"lr": 0.0007959536998847743, "data_time": 0.0018649816513061524, "grad_norm": 0.013648881413973867, "loss": 0.6881942272186279, "time": 0.5179283618927002, "epoch": 7, "memory": 20334, "step": 39248} +{"lr": 0.0007959536998847743, "data_time": 0.00165557861328125, "grad_norm": 0.014657158497720956, "loss": 0.6872871160507202, "time": 0.4319024085998535, "epoch": 7, "memory": 20334, "step": 39348} +{"lr": 0.0007959536998847743, "data_time": 0.0017303466796875, "grad_norm": 0.012619608407840132, "loss": 0.687632167339325, "time": 0.3954183340072632, "epoch": 7, "memory": 20334, "step": 39448} +{"lr": 0.0007959536998847743, "data_time": 0.0016219139099121094, "grad_norm": 0.018782898550853133, "loss": 0.6843981087207794, "time": 0.3781104564666748, "epoch": 7, "memory": 20334, "step": 39548} +{"lr": 0.0007959536998847743, "data_time": 0.0017826557159423828, "grad_norm": 0.011858153762295843, "loss": 0.6843623101711274, "time": 0.4297770977020264, "epoch": 7, "memory": 20334, "step": 39648} +{"lr": 0.0007959536998847743, "data_time": 0.001736903190612793, "grad_norm": 0.013285747787449509, "loss": 0.6878089725971221, "time": 0.4400207757949829, "epoch": 7, "memory": 20334, "step": 39748} +{"lr": 0.0007959536998847743, "data_time": 0.001980900764465332, "grad_norm": 0.0076624172565061596, "loss": 0.6867228329181672, "time": 0.4161480665206909, "epoch": 7, "memory": 20334, "step": 39848} +{"lr": 0.0007959536998847743, "data_time": 0.0017470359802246095, "grad_norm": 0.016743032447993755, "loss": 0.6832575142383576, "time": 0.4377040147781372, "epoch": 7, "memory": 20334, "step": 39948} +{"lr": 0.0007959536998847743, "data_time": 0.002133679389953613, "grad_norm": 0.01983685716986656, "loss": 0.692478609085083, "time": 0.37770495414733884, "epoch": 7, "memory": 20334, "step": 40048} +{"lr": 0.0007959536998847743, "data_time": 0.0016900062561035155, "grad_norm": 0.020389390457421542, "loss": 0.6839831888675689, "time": 0.4018282175064087, "epoch": 7, "memory": 20334, "step": 40148} +{"lr": 0.0007959536998847743, "data_time": 0.0020685195922851562, "grad_norm": 0.014187499857507646, "loss": 0.6850757956504822, "time": 0.3764138460159302, "epoch": 7, "memory": 20334, "step": 40248} +{"accuracy/top1": 100.0, "data_time": 0.3704343107011583, "time": 1.011265966627333, "step": 7} +{"lr": 0.0007297252973710758, "data_time": 0.0014069557189941406, "grad_norm": 0.01504069536458701, "loss": 0.6872755467891694, "time": 0.4223238468170166, "epoch": 8, "memory": 20334, "step": 40406} +{"lr": 0.0007297252973710758, "data_time": 0.001478719711303711, "grad_norm": 0.01260184432612732, "loss": 0.6861163735389709, "time": 0.39136905670166017, "epoch": 8, "memory": 20334, "step": 40506} +{"lr": 0.0007297252973710758, "data_time": 0.0018114566802978516, "grad_norm": 0.021287666726857425, "loss": 0.6850366771221161, "time": 0.4301972627639771, "epoch": 8, "memory": 20334, "step": 40606} +{"lr": 0.0007297252973710758, "data_time": 0.0015496492385864257, "grad_norm": 0.02114067629445344, "loss": 0.6902939260005951, "time": 0.40356786251068116, "epoch": 8, "memory": 20334, "step": 40706} +{"lr": 0.0007297252973710758, "data_time": 0.0015987396240234376, "grad_norm": 0.014086356712505221, "loss": 0.6874367773532868, "time": 0.4276456356048584, "epoch": 8, "memory": 20334, "step": 40806} +{"lr": 0.0007297252973710758, "data_time": 0.0014924287796020507, "grad_norm": 0.017130320169962944, "loss": 0.6926200211048126, "time": 0.3991257905960083, "epoch": 8, "memory": 20334, "step": 40906} +{"lr": 0.0007297252973710758, "data_time": 0.0015770435333251954, "grad_norm": 0.01609284762525931, "loss": 0.6854384422302247, "time": 0.46119678020477295, "epoch": 8, "memory": 20334, "step": 41006} +{"lr": 0.0007297252973710758, "data_time": 0.0014980554580688477, "grad_norm": 0.019533733511343597, "loss": 0.6853923678398133, "time": 0.39627504348754883, "epoch": 8, "memory": 20334, "step": 41106} +{"lr": 0.0007297252973710758, "data_time": 0.0015324115753173827, "grad_norm": 0.009302004193887114, "loss": 0.6858465254306794, "time": 0.4091492652893066, "epoch": 8, "memory": 20334, "step": 41206} +{"lr": 0.0007297252973710758, "data_time": 0.0014226436614990234, "grad_norm": 0.017282978142611684, "loss": 0.6882459640502929, "time": 0.39461569786071776, "epoch": 8, "memory": 20334, "step": 41306} +{"lr": 0.0007297252973710758, "data_time": 0.0015200138092041015, "grad_norm": 0.02055193189298734, "loss": 0.6900627791881562, "time": 0.3819331884384155, "epoch": 8, "memory": 20334, "step": 41406} +{"lr": 0.0007297252973710758, "data_time": 0.001488494873046875, "grad_norm": 0.013897991087287665, "loss": 0.6842069685459137, "time": 0.3753032922744751, "epoch": 8, "memory": 20334, "step": 41506} +{"lr": 0.0007297252973710758, "data_time": 0.0014233589172363281, "grad_norm": 0.020723226061090828, "loss": 0.6847513914108276, "time": 0.3833609104156494, "epoch": 8, "memory": 20334, "step": 41606} +{"lr": 0.0007297252973710758, "data_time": 0.0015473127365112304, "grad_norm": 0.019423927972093226, "loss": 0.6852060377597808, "time": 0.4026209354400635, "epoch": 8, "memory": 20334, "step": 41706} +{"lr": 0.0007297252973710758, "data_time": 0.0015175819396972656, "grad_norm": 0.017389740380167495, "loss": 0.690499484539032, "time": 0.437898063659668, "epoch": 8, "memory": 20334, "step": 41806} +{"lr": 0.0007297252973710758, "data_time": 0.0015148878097534179, "grad_norm": 0.007821018889080733, "loss": 0.6846396505832673, "time": 0.4067711114883423, "epoch": 8, "memory": 20334, "step": 41906} +{"lr": 0.0007297252973710758, "data_time": 0.0015543460845947265, "grad_norm": 0.013813624647445977, "loss": 0.6831416726112366, "time": 0.3627128601074219, "epoch": 8, "memory": 20334, "step": 42006} +{"lr": 0.0007297252973710758, "data_time": 0.001451730728149414, "grad_norm": 0.019529200764372943, "loss": 0.689741575717926, "time": 0.44032466411590576, "epoch": 8, "memory": 20334, "step": 42106} +{"lr": 0.0007297252973710758, "data_time": 0.0014779329299926757, "grad_norm": 0.015099528920836746, "loss": 0.6836788535118103, "time": 0.4030714511871338, "epoch": 8, "memory": 20334, "step": 42206} +{"lr": 0.0007297252973710758, "data_time": 0.0015130043029785156, "grad_norm": 0.015384985768469051, "loss": 0.6880769610404969, "time": 0.40560460090637207, "epoch": 8, "memory": 20334, "step": 42306} +{"lr": 0.0007297252973710758, "data_time": 0.0015192508697509765, "grad_norm": 0.017572984064463525, "loss": 0.6852215945720672, "time": 0.4113664150238037, "epoch": 8, "memory": 20334, "step": 42406} +{"lr": 0.0007297252973710758, "data_time": 0.0012242555618286132, "grad_norm": 0.01690259254537523, "loss": 0.6877039253711701, "time": 0.4119797945022583, "epoch": 8, "memory": 20334, "step": 42506} +{"lr": 0.0007297252973710758, "data_time": 0.001332402229309082, "grad_norm": 0.010942086321301758, "loss": 0.6843304991722107, "time": 0.47026662826538085, "epoch": 8, "memory": 20334, "step": 42606} +{"lr": 0.0007297252973710758, "data_time": 0.001267719268798828, "grad_norm": 0.01368114808574319, "loss": 0.6946668744087219, "time": 0.4309420347213745, "epoch": 8, "memory": 20334, "step": 42706} +{"lr": 0.0007297252973710758, "data_time": 0.0012215852737426757, "grad_norm": 0.015644904031069017, "loss": 0.6836284339427948, "time": 0.4006821870803833, "epoch": 8, "memory": 20334, "step": 42806} +{"lr": 0.0007297252973710758, "data_time": 0.0016925573348999024, "grad_norm": 0.01545888283289969, "loss": 0.6884254932403564, "time": 0.4241483688354492, "epoch": 8, "memory": 20334, "step": 42906} +{"lr": 0.0007297252973710758, "data_time": 0.0017265558242797851, "grad_norm": 0.012041582120582461, "loss": 0.6839982450008393, "time": 0.46854605674743655, "epoch": 8, "memory": 20334, "step": 43006} +{"lr": 0.0007297252973710758, "data_time": 0.0014457464218139648, "grad_norm": 0.009560292994137853, "loss": 0.682794165611267, "time": 0.41179053783416747, "epoch": 8, "memory": 20334, "step": 43106} +{"lr": 0.0007297252973710758, "data_time": 0.001885390281677246, "grad_norm": 0.012177361303474753, "loss": 0.6835753798484803, "time": 0.4030990839004517, "epoch": 8, "memory": 20334, "step": 43206} +{"lr": 0.0007297252973710758, "data_time": 0.0017806291580200195, "grad_norm": 0.016058540809899567, "loss": 0.6853805541992187, "time": 0.36375586986541747, "epoch": 8, "memory": 20334, "step": 43306} +{"lr": 0.0007297252973710758, "data_time": 0.0019548892974853515, "grad_norm": 0.02024875421775505, "loss": 0.6902428925037384, "time": 0.43312194347381594, "epoch": 8, "memory": 20334, "step": 43406} +{"lr": 0.0007297252973710758, "data_time": 0.002050614356994629, "grad_norm": 0.019893907569348813, "loss": 0.6832017123699188, "time": 0.38632769584655763, "epoch": 8, "memory": 20334, "step": 43506} +{"lr": 0.0007297252973710758, "data_time": 0.001772904396057129, "grad_norm": 0.014353366824798286, "loss": 0.6889331459999084, "time": 0.4603415489196777, "epoch": 8, "memory": 20334, "step": 43606} +{"lr": 0.0007297252973710758, "data_time": 0.001581120491027832, "grad_norm": 0.0169849916594103, "loss": 0.6835180044174194, "time": 0.39262301921844484, "epoch": 8, "memory": 20334, "step": 43706} +{"lr": 0.0007297252973710758, "data_time": 0.0018399715423583984, "grad_norm": 0.017583088134415448, "loss": 0.6803718268871307, "time": 0.40932228565216067, "epoch": 8, "memory": 20334, "step": 43806} +{"lr": 0.0007297252973710758, "data_time": 0.0015604019165039063, "grad_norm": 0.016612075013108553, "loss": 0.6864450633525848, "time": 0.40414042472839357, "epoch": 8, "memory": 20334, "step": 43906} +{"lr": 0.0007297252973710758, "data_time": 0.001879429817199707, "grad_norm": 0.012774892989546061, "loss": 0.686683738231659, "time": 0.38671696186065674, "epoch": 8, "memory": 20334, "step": 44006} +{"lr": 0.0007297252973710758, "data_time": 0.0022849082946777345, "grad_norm": 0.01221808884292841, "loss": 0.6845331132411957, "time": 0.4175445556640625, "epoch": 8, "memory": 20334, "step": 44106} +{"lr": 0.0007297252973710758, "data_time": 0.001493215560913086, "grad_norm": 0.014803660998586566, "loss": 0.6846046149730682, "time": 0.3991935968399048, "epoch": 8, "memory": 20334, "step": 44206} +{"lr": 0.0007297252973710758, "data_time": 0.0016632080078125, "grad_norm": 0.019390276959165932, "loss": 0.6883752167224884, "time": 0.4085487127304077, "epoch": 8, "memory": 20334, "step": 44306} +{"lr": 0.0007297252973710758, "data_time": 0.0018727540969848632, "grad_norm": 0.021830077411141246, "loss": 0.6901873409748077, "time": 0.43409912586212157, "epoch": 8, "memory": 20334, "step": 44406} +{"lr": 0.0007297252973710758, "data_time": 0.0026200056076049806, "grad_norm": 0.009195657703094185, "loss": 0.6831157267093658, "time": 0.4425152540206909, "epoch": 8, "memory": 20334, "step": 44506} +{"lr": 0.0007297252973710758, "data_time": 0.003099370002746582, "grad_norm": 0.014189627242740244, "loss": 0.6839025855064392, "time": 0.38259291648864746, "epoch": 8, "memory": 20334, "step": 44606} +{"lr": 0.0007297252973710758, "data_time": 0.0019882917404174805, "grad_norm": 0.014124078035820276, "loss": 0.6895446300506591, "time": 0.39026198387145994, "epoch": 8, "memory": 20334, "step": 44706} +{"lr": 0.0007297252973710758, "data_time": 0.00212252140045166, "grad_norm": 0.009498244698625058, "loss": 0.6897653877735138, "time": 0.40053367614746094, "epoch": 8, "memory": 20334, "step": 44806} +{"lr": 0.0007297252973710758, "data_time": 0.0024294376373291014, "grad_norm": 0.018176346668042243, "loss": 0.6842960834503173, "time": 0.3951588153839111, "epoch": 8, "memory": 20334, "step": 44906} +{"lr": 0.0007297252973710758, "data_time": 0.0017308712005615235, "grad_norm": 0.013260247011203319, "loss": 0.6856667935848236, "time": 0.38628082275390624, "epoch": 8, "memory": 20334, "step": 45006} +{"lr": 0.0007297252973710758, "data_time": 0.0019758462905883787, "grad_norm": 0.010604234255151824, "loss": 0.6831212699413299, "time": 0.4084931373596191, "epoch": 8, "memory": 20334, "step": 45106} +{"lr": 0.0007297252973710758, "data_time": 0.00211794376373291, "grad_norm": 0.019600277161225678, "loss": 0.6864668607711792, "time": 0.38871335983276367, "epoch": 8, "memory": 20334, "step": 45206} +{"lr": 0.0007297252973710758, "data_time": 0.0019192457199096679, "grad_norm": 0.016822970937937498, "loss": 0.6830528199672699, "time": 0.41142303943634034, "epoch": 8, "memory": 20334, "step": 45306} +{"lr": 0.0007297252973710758, "data_time": 0.0017076253890991211, "grad_norm": 0.017361124770832247, "loss": 0.6849460542201996, "time": 0.4418827295303345, "epoch": 8, "memory": 20334, "step": 45406} +{"lr": 0.0007297252973710758, "data_time": 0.0015312671661376954, "grad_norm": 0.02476479150354862, "loss": 0.6893586575984955, "time": 0.4284776210784912, "epoch": 8, "memory": 20334, "step": 45506} +{"lr": 0.0007297252973710758, "data_time": 0.0016729116439819336, "grad_norm": 0.01935760343912989, "loss": 0.6840833008289338, "time": 0.38116753101348877, "epoch": 8, "memory": 20334, "step": 45606} +{"lr": 0.0007297252973710758, "data_time": 0.0016595125198364258, "grad_norm": 0.01084623575443402, "loss": 0.6872931957244873, "time": 0.42051172256469727, "epoch": 8, "memory": 20334, "step": 45706} +{"lr": 0.0007297252973710758, "data_time": 0.0016260623931884765, "grad_norm": 0.009914619935443625, "loss": 0.6845249891281128, "time": 0.41153945922851565, "epoch": 8, "memory": 20334, "step": 45806} +{"lr": 0.0007297252973710758, "data_time": 0.0017794370651245117, "grad_norm": 0.023571905423887075, "loss": 0.6867764115333557, "time": 0.3996510744094849, "epoch": 8, "memory": 20334, "step": 45906} +{"lr": 0.0007297252973710758, "data_time": 0.0017148971557617188, "grad_norm": 0.014147424115799367, "loss": 0.6926047623157501, "time": 0.38025755882263185, "epoch": 8, "memory": 20334, "step": 46006} +{"accuracy/top1": 100.0, "data_time": 0.3850647078620063, "time": 1.0093521806928847, "step": 8} +{"lr": 0.0006579634122155991, "data_time": 0.0016479969024658203, "grad_norm": 0.01419377648853697, "loss": 0.6865527033805847, "time": 0.36565732955932617, "epoch": 9, "memory": 20334, "step": 46164} +{"lr": 0.0006579634122155991, "data_time": 0.0016137361526489258, "grad_norm": 0.019851939799264074, "loss": 0.6879981100559235, "time": 0.41054272651672363, "epoch": 9, "memory": 20334, "step": 46264} +{"lr": 0.0006579634122155991, "data_time": 0.0016621828079223632, "grad_norm": 0.008498499635607004, "loss": 0.6841829657554627, "time": 0.3799438953399658, "epoch": 9, "memory": 20334, "step": 46364} +{"lr": 0.0006579634122155991, "data_time": 0.0015968561172485351, "grad_norm": 0.01799968988634646, "loss": 0.6869735836982727, "time": 0.38701753616333007, "epoch": 9, "memory": 20334, "step": 46464} +{"lr": 0.0006579634122155991, "data_time": 0.0016284704208374024, "grad_norm": 0.01654924225003924, "loss": 0.6871818482875824, "time": 0.4096548318862915, "epoch": 9, "memory": 20334, "step": 46564} +{"lr": 0.0006579634122155991, "data_time": 0.0015188217163085937, "grad_norm": 0.017102921684272587, "loss": 0.6877378702163697, "time": 0.40784056186676027, "epoch": 9, "memory": 20334, "step": 46664} +{"lr": 0.0006579634122155991, "data_time": 0.0014745235443115235, "grad_norm": 0.026159213646315037, "loss": 0.6898949027061463, "time": 0.4024815082550049, "epoch": 9, "memory": 20334, "step": 46764} +{"lr": 0.0006579634122155991, "data_time": 0.0015573501586914062, "grad_norm": 0.013409408857114613, "loss": 0.6825119733810425, "time": 0.40892596244812013, "epoch": 9, "memory": 20334, "step": 46864} +{"lr": 0.0006579634122155991, "data_time": 0.0016701936721801758, "grad_norm": 0.016575712710618973, "loss": 0.6853305220603942, "time": 0.43486199378967283, "epoch": 9, "memory": 20334, "step": 46964} +{"lr": 0.0006579634122155991, "data_time": 0.001623249053955078, "grad_norm": 0.029668823443353178, "loss": 0.6825747966766358, "time": 0.4156445264816284, "epoch": 9, "memory": 20334, "step": 47064} +{"lr": 0.0006579634122155991, "data_time": 0.001653432846069336, "grad_norm": 0.024123814527411015, "loss": 0.6842483103275299, "time": 0.400403356552124, "epoch": 9, "memory": 20334, "step": 47164} +{"lr": 0.0006579634122155991, "data_time": 0.0015319108963012696, "grad_norm": 0.016646164143458007, "loss": 0.6879630446434021, "time": 0.3889906883239746, "epoch": 9, "memory": 20334, "step": 47264} +{"lr": 0.0006579634122155991, "data_time": 0.0018725395202636719, "grad_norm": 0.017204239362035877, "loss": 0.6901099979877472, "time": 0.41818020343780515, "epoch": 9, "memory": 20334, "step": 47364} +{"lr": 0.0006579634122155991, "data_time": 0.0017591953277587891, "grad_norm": 0.023627828760072588, "loss": 0.6925867855548858, "time": 0.37683274745941164, "epoch": 9, "memory": 20334, "step": 47464} +{"lr": 0.0006579634122155991, "data_time": 0.0015821456909179688, "grad_norm": 0.020276833092793822, "loss": 0.6883210062980651, "time": 0.378000283241272, "epoch": 9, "memory": 20334, "step": 47564} +{"lr": 0.0006579634122155991, "data_time": 0.0018356800079345702, "grad_norm": 0.020691714878194035, "loss": 0.6894786536693573, "time": 0.43771560192108155, "epoch": 9, "memory": 20334, "step": 47664} +{"lr": 0.0006579634122155991, "data_time": 0.0015072107315063476, "grad_norm": 0.02632945142686367, "loss": 0.6901565432548523, "time": 0.423551344871521, "epoch": 9, "memory": 20334, "step": 47764} +{"lr": 0.0006579634122155991, "data_time": 0.0014970541000366212, "grad_norm": 0.016661419306183235, "loss": 0.6927356719970703, "time": 0.3900083065032959, "epoch": 9, "memory": 20334, "step": 47864} +{"lr": 0.0006579634122155991, "data_time": 0.0014163017272949218, "grad_norm": 0.01677846600068733, "loss": 0.6866316437721253, "time": 0.4081490755081177, "epoch": 9, "memory": 20334, "step": 47964} +{"lr": 0.0006579634122155991, "data_time": 0.0014652252197265626, "grad_norm": 0.011928746732883156, "loss": 0.6864598095417023, "time": 0.4291290283203125, "epoch": 9, "memory": 20334, "step": 48064} +{"lr": 0.0006579634122155991, "data_time": 0.0014700174331665039, "grad_norm": 0.018529455410316586, "loss": 0.6883900165557861, "time": 0.36959547996520997, "epoch": 9, "memory": 20334, "step": 48164} +{"lr": 0.0006579634122155991, "data_time": 0.0014411687850952148, "grad_norm": 0.016606410639360546, "loss": 0.6852875113487243, "time": 0.43194167613983153, "epoch": 9, "memory": 20334, "step": 48264} +{"lr": 0.0006579634122155991, "data_time": 0.001485157012939453, "grad_norm": 0.014752424554899335, "loss": 0.6843710303306579, "time": 0.41739075183868407, "epoch": 9, "memory": 20334, "step": 48364} +{"lr": 0.0006579634122155991, "data_time": 0.001543140411376953, "grad_norm": 0.01776666350197047, "loss": 0.6881593048572541, "time": 0.43468174934387205, "epoch": 9, "memory": 20334, "step": 48464} +{"lr": 0.0006579634122155991, "data_time": 0.0017083406448364258, "grad_norm": 0.016836788272485137, "loss": 0.6823172032833099, "time": 0.38342211246490476, "epoch": 9, "memory": 20334, "step": 48564} +{"lr": 0.0006579634122155991, "data_time": 0.0016509532928466798, "grad_norm": 0.012628317018970847, "loss": 0.6859584987163544, "time": 0.38100109100341795, "epoch": 9, "memory": 20334, "step": 48664} +{"lr": 0.0006579634122155991, "data_time": 0.001735687255859375, "grad_norm": 0.019650646939408035, "loss": 0.6837439060211181, "time": 0.39613118171691897, "epoch": 9, "memory": 20334, "step": 48764} +{"lr": 0.0006579634122155991, "data_time": 0.0017971038818359376, "grad_norm": 0.01567758454475552, "loss": 0.6874577343463898, "time": 0.44351351261138916, "epoch": 9, "memory": 20334, "step": 48864} +{"lr": 0.0006579634122155991, "data_time": 0.00178220272064209, "grad_norm": 0.01802654191851616, "loss": 0.6824274778366088, "time": 0.3819967031478882, "epoch": 9, "memory": 20334, "step": 48964} +{"lr": 0.0006579634122155991, "data_time": 0.0016217470169067384, "grad_norm": 0.01629552954691462, "loss": 0.6909584164619446, "time": 0.40527799129486086, "epoch": 9, "memory": 20334, "step": 49064} +{"lr": 0.0006579634122155991, "data_time": 0.001608872413635254, "grad_norm": 0.017447477905079722, "loss": 0.6902225911617279, "time": 0.38200912475585935, "epoch": 9, "memory": 20334, "step": 49164} +{"lr": 0.0006579634122155991, "data_time": 0.0017590761184692384, "grad_norm": 0.012975326050946024, "loss": 0.6888095498085022, "time": 0.4060666561126709, "epoch": 9, "memory": 20334, "step": 49264} +{"lr": 0.0006579634122155991, "data_time": 0.0017446041107177734, "grad_norm": 0.015692339441739023, "loss": 0.6879356622695922, "time": 0.3684777498245239, "epoch": 9, "memory": 20334, "step": 49364} +{"lr": 0.0006579634122155991, "data_time": 0.001659083366394043, "grad_norm": 0.01458758725784719, "loss": 0.6808813750743866, "time": 0.43731138706207273, "epoch": 9, "memory": 20334, "step": 49464} +{"lr": 0.0006579634122155991, "data_time": 0.001909947395324707, "grad_norm": 0.016200729040428997, "loss": 0.6865518748760223, "time": 0.36342535018920896, "epoch": 9, "memory": 20334, "step": 49564} +{"lr": 0.0006579634122155991, "data_time": 0.0017020702362060547, "grad_norm": 0.020468140457523987, "loss": 0.6849137902259826, "time": 0.3801534652709961, "epoch": 9, "memory": 20334, "step": 49664} +{"lr": 0.0006579634122155991, "data_time": 0.0016418933868408204, "grad_norm": 0.01708055417984724, "loss": 0.6884378314018249, "time": 0.39692373275756837, "epoch": 9, "memory": 20334, "step": 49764} +{"lr": 0.0006579634122155991, "data_time": 0.001673579216003418, "grad_norm": 0.010124088363954797, "loss": 0.6838084518909454, "time": 0.4089380979537964, "epoch": 9, "memory": 20334, "step": 49864} +{"lr": 0.0006579634122155991, "data_time": 0.0017638683319091796, "grad_norm": 0.014845739683369174, "loss": 0.689606660604477, "time": 0.3887681484222412, "epoch": 9, "memory": 20334, "step": 49964} +{"lr": 0.0006579634122155991, "data_time": 0.0015306711196899415, "grad_norm": 0.013253023452125489, "loss": 0.6858506262302398, "time": 0.4657459259033203, "epoch": 9, "memory": 20334, "step": 50064} +{"lr": 0.0006579634122155991, "data_time": 0.0015999317169189454, "grad_norm": 0.012637721397913993, "loss": 0.6846360683441162, "time": 0.4193205118179321, "epoch": 9, "memory": 20334, "step": 50164} +{"lr": 0.0006579634122155991, "data_time": 0.0015675544738769532, "grad_norm": 0.01643711608485319, "loss": 0.6868250489234924, "time": 0.45740585327148436, "epoch": 9, "memory": 20334, "step": 50264} +{"lr": 0.0006579634122155991, "data_time": 0.0015584230422973633, "grad_norm": 0.019852726720273495, "loss": 0.6844405114650727, "time": 0.40009050369262694, "epoch": 9, "memory": 20334, "step": 50364} +{"lr": 0.0006579634122155991, "data_time": 0.0014082908630371094, "grad_norm": 0.016296138672623785, "loss": 0.6904605507850647, "time": 0.41584053039550783, "epoch": 9, "memory": 20334, "step": 50464} +{"lr": 0.0006579634122155991, "data_time": 0.0012562990188598633, "grad_norm": 0.015375867043621838, "loss": 0.6873063504695892, "time": 0.4089855194091797, "epoch": 9, "memory": 20334, "step": 50564} +{"lr": 0.0006579634122155991, "data_time": 0.0013644933700561524, "grad_norm": 0.016306512081064282, "loss": 0.6848997890949249, "time": 0.4068164587020874, "epoch": 9, "memory": 20334, "step": 50664} +{"lr": 0.0006579634122155991, "data_time": 0.001303553581237793, "grad_norm": 0.012046275520697236, "loss": 0.6894914329051971, "time": 0.3742116928100586, "epoch": 9, "memory": 20334, "step": 50764} +{"lr": 0.0006579634122155991, "data_time": 0.0013122320175170898, "grad_norm": 0.01793232560157776, "loss": 0.6837822079658509, "time": 0.3626039981842041, "epoch": 9, "memory": 20334, "step": 50864} +{"lr": 0.0006579634122155991, "data_time": 0.0015569210052490234, "grad_norm": 0.010402942029759287, "loss": 0.6881684243679047, "time": 0.41551783084869387, "epoch": 9, "memory": 20334, "step": 50964} +{"lr": 0.0006579634122155991, "data_time": 0.0015505313873291015, "grad_norm": 0.01626515844836831, "loss": 0.6854903101921082, "time": 0.41629719734191895, "epoch": 9, "memory": 20334, "step": 51064} +{"lr": 0.0006579634122155991, "data_time": 0.0014333486557006835, "grad_norm": 0.01355631654150784, "loss": 0.6880523145198822, "time": 0.3880305767059326, "epoch": 9, "memory": 20334, "step": 51164} +{"lr": 0.0006579634122155991, "data_time": 0.0014967679977416991, "grad_norm": 0.015107460401486606, "loss": 0.6864442765712738, "time": 0.3931441783905029, "epoch": 9, "memory": 20334, "step": 51264} +{"lr": 0.0006579634122155991, "data_time": 0.0015359163284301759, "grad_norm": 0.021372124808840452, "loss": 0.684925252199173, "time": 0.36798503398895266, "epoch": 9, "memory": 20334, "step": 51364} +{"lr": 0.0006579634122155991, "data_time": 0.0017138481140136718, "grad_norm": 0.017495215917006136, "loss": 0.6825269877910614, "time": 0.4488940715789795, "epoch": 9, "memory": 20334, "step": 51464} +{"lr": 0.0006579634122155991, "data_time": 0.0016608238220214844, "grad_norm": 0.01851346460171044, "loss": 0.6859010696411133, "time": 0.41650211811065674, "epoch": 9, "memory": 20334, "step": 51564} +{"lr": 0.0006579634122155991, "data_time": 0.0017618656158447266, "grad_norm": 0.011144491762388497, "loss": 0.6866799414157867, "time": 0.3864098072052002, "epoch": 9, "memory": 20334, "step": 51664} +{"lr": 0.0006579634122155991, "data_time": 0.0015483856201171874, "grad_norm": 0.016839176486246287, "loss": 0.6818654358386993, "time": 0.3709238052368164, "epoch": 9, "memory": 20334, "step": 51764} +{"accuracy/top1": 100.0, "data_time": 0.371474027633667, "time": 0.9948434034983317, "step": 9} +{"lr": 0.0005824350601949144, "data_time": 0.0015693187713623046, "grad_norm": 0.013630917656701057, "loss": 0.6861572086811065, "time": 0.4262171983718872, "epoch": 10, "memory": 20334, "step": 51922} +{"lr": 0.0005824350601949144, "data_time": 0.0016273021697998046, "grad_norm": 0.015809801197610794, "loss": 0.6848917543888092, "time": 0.38904197216033937, "epoch": 10, "memory": 20334, "step": 52022} +{"lr": 0.0005824350601949144, "data_time": 0.0018444538116455078, "grad_norm": 0.015528548881411553, "loss": 0.6847606837749481, "time": 0.43321115970611573, "epoch": 10, "memory": 20334, "step": 52122} +{"lr": 0.0005824350601949144, "data_time": 0.0017694473266601563, "grad_norm": 0.015170224756002427, "loss": 0.6875871837139129, "time": 0.36998488903045657, "epoch": 10, "memory": 20334, "step": 52222} +{"lr": 0.0005824350601949144, "data_time": 0.0016355276107788085, "grad_norm": 0.010488854415598325, "loss": 0.6874406039714813, "time": 0.410483193397522, "epoch": 10, "memory": 20334, "step": 52322} +{"lr": 0.0005824350601949144, "data_time": 0.0015836238861083984, "grad_norm": 0.011591467773541808, "loss": 0.6903205394744873, "time": 0.41949594020843506, "epoch": 10, "memory": 20334, "step": 52422} +{"lr": 0.0005824350601949144, "data_time": 0.0017516374588012694, "grad_norm": 0.02135088196955621, "loss": 0.6842758774757385, "time": 0.39008872509002684, "epoch": 10, "memory": 20334, "step": 52522} +{"lr": 0.0005824350601949144, "data_time": 0.0014495372772216797, "grad_norm": 0.016125445498619228, "loss": 0.6893079340457916, "time": 0.4323333740234375, "epoch": 10, "memory": 20334, "step": 52622} +{"lr": 0.0005824350601949144, "data_time": 0.001698470115661621, "grad_norm": 0.01552496430813335, "loss": 0.6847705245018005, "time": 0.38407206535339355, "epoch": 10, "memory": 20334, "step": 52722} +{"lr": 0.0005824350601949144, "data_time": 0.0018587350845336915, "grad_norm": 0.01834776420146227, "loss": 0.6794097065925598, "time": 0.38328773975372316, "epoch": 10, "memory": 20334, "step": 52822} +{"lr": 0.0005824350601949144, "data_time": 0.0017519235610961915, "grad_norm": 0.01273553364444524, "loss": 0.6832436561584473, "time": 0.3640961885452271, "epoch": 10, "memory": 20334, "step": 52922} +{"lr": 0.0005824350601949144, "data_time": 0.0017412185668945312, "grad_norm": 0.018623796047177166, "loss": 0.6863396167755127, "time": 0.3907542943954468, "epoch": 10, "memory": 20334, "step": 53022} +{"lr": 0.0005824350601949144, "data_time": 0.001468515396118164, "grad_norm": 0.017051333433482797, "loss": 0.6921289026737213, "time": 0.4294722318649292, "epoch": 10, "memory": 20334, "step": 53122} +{"lr": 0.0005824350601949144, "data_time": 0.0016848087310791016, "grad_norm": 0.02026559254154563, "loss": 0.685213041305542, "time": 0.3710071325302124, "epoch": 10, "memory": 20334, "step": 53222} +{"lr": 0.0005824350601949144, "data_time": 0.001784515380859375, "grad_norm": 0.01818013987503946, "loss": 0.6863143682479859, "time": 0.3864701747894287, "epoch": 10, "memory": 20334, "step": 53322} +{"lr": 0.0005824350601949144, "data_time": 0.001483583450317383, "grad_norm": 0.017659070342779158, "loss": 0.6906317472457886, "time": 0.41943793296813964, "epoch": 10, "memory": 20334, "step": 53422} +{"lr": 0.0005824350601949144, "data_time": 0.0014837026596069337, "grad_norm": 0.014083898643730209, "loss": 0.6835853934288025, "time": 0.40425686836242675, "epoch": 10, "memory": 20334, "step": 53522} +{"lr": 0.0005824350601949144, "data_time": 0.0016284465789794921, "grad_norm": 0.01893379450775683, "loss": 0.6863393545150757, "time": 0.37946789264678954, "epoch": 10, "memory": 20334, "step": 53622} +{"lr": 0.0005824350601949144, "data_time": 0.0014427900314331055, "grad_norm": 0.011715705273672938, "loss": 0.6884918987751008, "time": 0.42046072483062746, "epoch": 10, "memory": 20334, "step": 53722} +{"lr": 0.0005824350601949144, "data_time": 0.0016260147094726562, "grad_norm": 0.018639779184013605, "loss": 0.6875562071800232, "time": 0.40404956340789794, "epoch": 10, "memory": 20334, "step": 53822} +{"lr": 0.0005824350601949144, "data_time": 0.0015421390533447265, "grad_norm": 0.02021653435076587, "loss": 0.6900019109249115, "time": 0.3707756519317627, "epoch": 10, "memory": 20334, "step": 53922} +{"lr": 0.0005824350601949144, "data_time": 0.0015859127044677735, "grad_norm": 0.013106706569669769, "loss": 0.6885911643505096, "time": 0.40043487548828127, "epoch": 10, "memory": 20334, "step": 54022} +{"lr": 0.0005824350601949144, "data_time": 0.001854681968688965, "grad_norm": 0.01030412825057283, "loss": 0.6827254593372345, "time": 0.37603662014007566, "epoch": 10, "memory": 20334, "step": 54122} +{"lr": 0.0005824350601949144, "data_time": 0.0016684532165527344, "grad_norm": 0.019184313947334886, "loss": 0.6844351291656494, "time": 0.413557767868042, "epoch": 10, "memory": 20334, "step": 54222} +{"lr": 0.0005824350601949144, "data_time": 0.00196535587310791, "grad_norm": 0.016056846117135138, "loss": 0.6877652645111084, "time": 0.4293826103210449, "epoch": 10, "memory": 20334, "step": 54322} +{"lr": 0.0005824350601949144, "data_time": 0.0013401269912719726, "grad_norm": 0.012847650586627424, "loss": 0.6899170875549316, "time": 0.42327816486358644, "epoch": 10, "memory": 20334, "step": 54422} +{"lr": 0.0005824350601949144, "data_time": 0.0014577150344848634, "grad_norm": 0.012815133947879077, "loss": 0.6877005279064179, "time": 0.41201415061950686, "epoch": 10, "memory": 20334, "step": 54522} +{"lr": 0.0005824350601949144, "data_time": 0.001948237419128418, "grad_norm": 0.01572614754550159, "loss": 0.6882017374038696, "time": 0.41745293140411377, "epoch": 10, "memory": 20334, "step": 54622} +{"lr": 0.0005824350601949144, "data_time": 0.0015142440795898437, "grad_norm": 0.017514981690328568, "loss": 0.6816524267196655, "time": 0.4090061902999878, "epoch": 10, "memory": 20334, "step": 54722} +{"lr": 0.0005824350601949144, "data_time": 0.0015898942947387695, "grad_norm": 0.0068565423134714365, "loss": 0.6841822147369385, "time": 0.39079787731170657, "epoch": 10, "memory": 20334, "step": 54822} +{"lr": 0.0005824350601949144, "data_time": 0.0017063617706298828, "grad_norm": 0.018116200400982052, "loss": 0.6883988082408905, "time": 0.40350868701934817, "epoch": 10, "memory": 20334, "step": 54922} +{"lr": 0.0005824350601949144, "data_time": 0.0016272783279418946, "grad_norm": 0.008793468365911394, "loss": 0.6830611050128936, "time": 0.39291794300079347, "epoch": 10, "memory": 20334, "step": 55022} +{"lr": 0.0005824350601949144, "data_time": 0.0015559673309326171, "grad_norm": 0.011733010620810091, "loss": 0.6843908667564392, "time": 0.4198174476623535, "epoch": 10, "memory": 20334, "step": 55122} +{"lr": 0.0005824350601949144, "data_time": 0.0017212390899658202, "grad_norm": 0.011460137815447524, "loss": 0.6900747001171113, "time": 0.40550053119659424, "epoch": 10, "memory": 20334, "step": 55222} +{"lr": 0.0005824350601949144, "data_time": 0.001712965965270996, "grad_norm": 0.014100943400990218, "loss": 0.6912566542625427, "time": 0.3676278114318848, "epoch": 10, "memory": 20334, "step": 55322} +{"lr": 0.0005824350601949144, "data_time": 0.0017971754074096679, "grad_norm": 0.015873079164884986, "loss": 0.6866163492202759, "time": 0.3884373426437378, "epoch": 10, "memory": 20334, "step": 55422} +{"lr": 0.0005824350601949144, "data_time": 0.001832723617553711, "grad_norm": 0.021550805680453776, "loss": 0.6876806855201721, "time": 0.40866105556488036, "epoch": 10, "memory": 20334, "step": 55522} +{"lr": 0.0005824350601949144, "data_time": 0.0016146183013916015, "grad_norm": 0.014574577275197953, "loss": 0.6817083835601807, "time": 0.41771717071533204, "epoch": 10, "memory": 20334, "step": 55622} +{"lr": 0.0005824350601949144, "data_time": 0.0019276857376098633, "grad_norm": 0.010971394274383784, "loss": 0.6877998650074005, "time": 0.38044607639312744, "epoch": 10, "memory": 20334, "step": 55722} +{"lr": 0.0005824350601949144, "data_time": 0.0015934467315673827, "grad_norm": 0.01014065844938159, "loss": 0.6867319226264954, "time": 0.3712127447128296, "epoch": 10, "memory": 20334, "step": 55822} +{"lr": 0.0005824350601949144, "data_time": 0.0019123077392578125, "grad_norm": 0.011535261198878289, "loss": 0.68774453997612, "time": 0.3735557794570923, "epoch": 10, "memory": 20334, "step": 55922} +{"lr": 0.0005824350601949144, "data_time": 0.0018356800079345702, "grad_norm": 0.012448557512834668, "loss": 0.6829745411872864, "time": 0.39452495574951174, "epoch": 10, "memory": 20334, "step": 56022} +{"lr": 0.0005824350601949144, "data_time": 0.001998615264892578, "grad_norm": 0.020786867733113466, "loss": 0.6918170928955079, "time": 0.3841336488723755, "epoch": 10, "memory": 20334, "step": 56122} +{"lr": 0.0005824350601949144, "data_time": 0.0017288446426391602, "grad_norm": 0.013509073189925402, "loss": 0.6818134009838104, "time": 0.40670504570007326, "epoch": 10, "memory": 20334, "step": 56222} +{"lr": 0.0005824350601949144, "data_time": 0.0018560409545898438, "grad_norm": 0.012721085408702492, "loss": 0.68084916472435, "time": 0.3725078344345093, "epoch": 10, "memory": 20334, "step": 56322} +{"lr": 0.0005824350601949144, "data_time": 0.0015450716018676758, "grad_norm": 0.02103833546862006, "loss": 0.6907468199729919, "time": 0.3778859853744507, "epoch": 10, "memory": 20334, "step": 56422} +{"lr": 0.0005824350601949144, "data_time": 0.0017362117767333984, "grad_norm": 0.012111650290898978, "loss": 0.6793406724929809, "time": 0.415705132484436, "epoch": 10, "memory": 20334, "step": 56522} +{"lr": 0.0005824350601949144, "data_time": 0.001720738410949707, "grad_norm": 0.015598732838407159, "loss": 0.6881059110164642, "time": 0.38643875122070315, "epoch": 10, "memory": 20334, "step": 56622} +{"lr": 0.0005824350601949144, "data_time": 0.001687026023864746, "grad_norm": 0.015401056152768433, "loss": 0.6878041982650757, "time": 0.4160081624984741, "epoch": 10, "memory": 20334, "step": 56722} +{"lr": 0.0005824350601949144, "data_time": 0.0016798019409179688, "grad_norm": 0.016470563248731196, "loss": 0.6827340066432953, "time": 0.40014288425445554, "epoch": 10, "memory": 20334, "step": 56822} +{"lr": 0.0005824350601949144, "data_time": 0.0015673160552978516, "grad_norm": 0.013484801992308348, "loss": 0.6872635900974273, "time": 0.38296396732330323, "epoch": 10, "memory": 20334, "step": 56922} +{"lr": 0.0005824350601949144, "data_time": 0.0015890359878540038, "grad_norm": 0.019567966926842928, "loss": 0.6892824411392212, "time": 0.4249274253845215, "epoch": 10, "memory": 20334, "step": 57022} +{"lr": 0.0005824350601949144, "data_time": 0.0016318559646606445, "grad_norm": 0.018295177374966443, "loss": 0.6846611738204956, "time": 0.4224911451339722, "epoch": 10, "memory": 20334, "step": 57122} +{"lr": 0.0005824350601949144, "data_time": 0.0015627622604370117, "grad_norm": 0.01712101136799902, "loss": 0.6868502974510193, "time": 0.396042537689209, "epoch": 10, "memory": 20334, "step": 57222} +{"lr": 0.0005824350601949144, "data_time": 0.0019245147705078125, "grad_norm": 0.011905309010762722, "loss": 0.6864865183830261, "time": 0.421490740776062, "epoch": 10, "memory": 20334, "step": 57322} +{"lr": 0.0005824350601949144, "data_time": 0.0018316984176635742, "grad_norm": 0.02494481534231454, "loss": 0.6872658133506775, "time": 0.37062802314758303, "epoch": 10, "memory": 20334, "step": 57422} +{"lr": 0.0005824350601949144, "data_time": 0.0019359111785888672, "grad_norm": 0.014627020398620516, "loss": 0.6856526255607605, "time": 0.4216599464416504, "epoch": 10, "memory": 20334, "step": 57522} +{"accuracy/top1": 100.0, "data_time": 0.3692814244164361, "time": 1.0086095598008897, "step": 10} +{"lr": 0.0005050000000000001, "data_time": 0.0016867876052856444, "grad_norm": 0.01556721068918705, "loss": 0.6872677981853486, "time": 0.40900297164916993, "epoch": 11, "memory": 20334, "step": 57680} +{"lr": 0.0005050000000000001, "data_time": 0.0017450094223022462, "grad_norm": 0.013444986648391933, "loss": 0.6864603221416473, "time": 0.3714394807815552, "epoch": 11, "memory": 20334, "step": 57780} +{"lr": 0.0005050000000000001, "data_time": 0.0017448663711547852, "grad_norm": 0.01790513030719012, "loss": 0.690277487039566, "time": 0.3853780746459961, "epoch": 11, "memory": 20334, "step": 57880} +{"lr": 0.0005050000000000001, "data_time": 0.0016040325164794922, "grad_norm": 0.012645078659988939, "loss": 0.6907342135906219, "time": 0.4526124715805054, "epoch": 11, "memory": 20334, "step": 57980} +{"lr": 0.0005050000000000001, "data_time": 0.0018157005310058595, "grad_norm": 0.01319127669557929, "loss": 0.6838784575462341, "time": 0.4102788925170898, "epoch": 11, "memory": 20334, "step": 58080} +{"lr": 0.0005050000000000001, "data_time": 0.0016178131103515626, "grad_norm": 0.010991183505393565, "loss": 0.6889654695987701, "time": 0.3825472354888916, "epoch": 11, "memory": 20334, "step": 58180} +{"lr": 0.0005050000000000001, "data_time": 0.0017494440078735351, "grad_norm": 0.016774243616964668, "loss": 0.6823360204696656, "time": 0.4215251445770264, "epoch": 11, "memory": 20334, "step": 58280} +{"lr": 0.0005050000000000001, "data_time": 0.001706075668334961, "grad_norm": 0.01579231631476432, "loss": 0.6852897703647614, "time": 0.3628375053405762, "epoch": 11, "memory": 20334, "step": 58380} +{"lr": 0.0005050000000000001, "data_time": 0.0016942739486694336, "grad_norm": 0.014355609053745865, "loss": 0.6871700406074523, "time": 0.39527525901794436, "epoch": 11, "memory": 20334, "step": 58480} +{"lr": 0.0005050000000000001, "data_time": 0.001577615737915039, "grad_norm": 0.01576843245420605, "loss": 0.6874830901622773, "time": 0.39677345752716064, "epoch": 11, "memory": 20334, "step": 58580} +{"lr": 0.0005050000000000001, "data_time": 0.0016754627227783202, "grad_norm": 0.013481079041957856, "loss": 0.6834925889968873, "time": 0.3885331392288208, "epoch": 11, "memory": 20334, "step": 58680} +{"lr": 0.0005050000000000001, "data_time": 0.0016225576400756836, "grad_norm": 0.01784527823328972, "loss": 0.6883037149906158, "time": 0.47656280994415284, "epoch": 11, "memory": 20334, "step": 58780} +{"lr": 0.0005050000000000001, "data_time": 0.0018278598785400391, "grad_norm": 0.021230232785455883, "loss": 0.687637597322464, "time": 0.37162487506866454, "epoch": 11, "memory": 20334, "step": 58880} +{"lr": 0.0005050000000000001, "data_time": 0.002013278007507324, "grad_norm": 0.011489739397075028, "loss": 0.6880789995193481, "time": 0.4111733675003052, "epoch": 11, "memory": 20334, "step": 58980} +{"lr": 0.0005050000000000001, "data_time": 0.0019643545150756837, "grad_norm": 0.011365625041071326, "loss": 0.6861910879611969, "time": 0.402612042427063, "epoch": 11, "memory": 20334, "step": 59080} +{"lr": 0.0005050000000000001, "data_time": 0.0018092870712280273, "grad_norm": 0.014367542136460543, "loss": 0.6900720775127411, "time": 0.4091456890106201, "epoch": 11, "memory": 20334, "step": 59180} +{"lr": 0.0005050000000000001, "data_time": 0.0016371250152587891, "grad_norm": 0.020088522182777523, "loss": 0.6836402118206024, "time": 0.43627243041992186, "epoch": 11, "memory": 20334, "step": 59280} +{"lr": 0.0005050000000000001, "data_time": 0.0017477035522460937, "grad_norm": 0.014833125588484108, "loss": 0.6904846727848053, "time": 0.39355833530426027, "epoch": 11, "memory": 20334, "step": 59380} +{"lr": 0.0005050000000000001, "data_time": 0.0016386270523071288, "grad_norm": 0.02122778237098828, "loss": 0.6817389607429505, "time": 0.42151169776916503, "epoch": 11, "memory": 20334, "step": 59480} +{"lr": 0.0005050000000000001, "data_time": 0.0015434503555297851, "grad_norm": 0.019401673693209887, "loss": 0.6845398485660553, "time": 0.39098615646362306, "epoch": 11, "memory": 20334, "step": 59580} +{"lr": 0.0005050000000000001, "data_time": 0.0015362262725830077, "grad_norm": 0.012637698696926236, "loss": 0.6813634753227233, "time": 0.4173738956451416, "epoch": 11, "memory": 20334, "step": 59680} +{"lr": 0.0005050000000000001, "data_time": 0.001519465446472168, "grad_norm": 0.017714382335543632, "loss": 0.6863687574863434, "time": 0.3951176881790161, "epoch": 11, "memory": 20334, "step": 59780} +{"lr": 0.0005050000000000001, "data_time": 0.0014509439468383789, "grad_norm": 0.01826527458615601, "loss": 0.685668557882309, "time": 0.40448362827301027, "epoch": 11, "memory": 20334, "step": 59880} +{"lr": 0.0005050000000000001, "data_time": 0.0012654781341552735, "grad_norm": 0.01331446843687445, "loss": 0.6856652140617371, "time": 0.4253491163253784, "epoch": 11, "memory": 20334, "step": 59980} +{"lr": 0.0005050000000000001, "data_time": 0.001400899887084961, "grad_norm": 0.01342430289951153, "loss": 0.6874310374259949, "time": 0.3719109773635864, "epoch": 11, "memory": 20334, "step": 60080} +{"lr": 0.0005050000000000001, "data_time": 0.0012639999389648438, "grad_norm": 0.01627713702619076, "loss": 0.6869747042655945, "time": 0.40706191062927244, "epoch": 11, "memory": 20334, "step": 60180} +{"lr": 0.0005050000000000001, "data_time": 0.0014351367950439452, "grad_norm": 0.013047590979840606, "loss": 0.6925990104675293, "time": 0.41955182552337644, "epoch": 11, "memory": 20334, "step": 60280} +{"lr": 0.0005050000000000001, "data_time": 0.0014681577682495116, "grad_norm": 0.0162918713176623, "loss": 0.6841265261173248, "time": 0.42308735847473145, "epoch": 11, "memory": 20334, "step": 60380} +{"lr": 0.0005050000000000001, "data_time": 0.001253199577331543, "grad_norm": 0.015976412466261535, "loss": 0.6813515782356262, "time": 0.4027266025543213, "epoch": 11, "memory": 20334, "step": 60480} +{"lr": 0.0005050000000000001, "data_time": 0.0013670206069946289, "grad_norm": 0.018589975498616695, "loss": 0.6886433362960815, "time": 0.3883127927780151, "epoch": 11, "memory": 20334, "step": 60580} +{"lr": 0.0005050000000000001, "data_time": 0.0019369125366210938, "grad_norm": 0.018638608464971184, "loss": 0.6886873841285706, "time": 0.4225748062133789, "epoch": 11, "memory": 20334, "step": 60680} +{"lr": 0.0005050000000000001, "data_time": 0.0016449213027954102, "grad_norm": 0.023885917756706475, "loss": 0.6888404905796051, "time": 0.3968960762023926, "epoch": 11, "memory": 20334, "step": 60780} +{"lr": 0.0005050000000000001, "data_time": 0.0016598701477050781, "grad_norm": 0.019483834481798114, "loss": 0.6818107306957245, "time": 0.3683042526245117, "epoch": 11, "memory": 20334, "step": 60880} +{"lr": 0.0005050000000000001, "data_time": 0.0015219926834106445, "grad_norm": 0.021793618984520435, "loss": 0.6875822424888611, "time": 0.40129795074462893, "epoch": 11, "memory": 20334, "step": 60980} +{"lr": 0.0005050000000000001, "data_time": 0.0016986370086669923, "grad_norm": 0.01363177023595199, "loss": 0.6868061244487762, "time": 0.39838347434997556, "epoch": 11, "memory": 20334, "step": 61080} +{"lr": 0.0005050000000000001, "data_time": 0.0018582344055175781, "grad_norm": 0.014399600576143711, "loss": 0.6916409134864807, "time": 0.45100481510162355, "epoch": 11, "memory": 20334, "step": 61180} +{"lr": 0.0005050000000000001, "data_time": 0.0019231796264648437, "grad_norm": 0.012664830044377596, "loss": 0.6798730075359345, "time": 0.40946035385131835, "epoch": 11, "memory": 20334, "step": 61280} +{"lr": 0.0005050000000000001, "data_time": 0.001731133460998535, "grad_norm": 0.026317989779636265, "loss": 0.6835351705551147, "time": 0.4009954214096069, "epoch": 11, "memory": 20334, "step": 61380} +{"lr": 0.0005050000000000001, "data_time": 0.0018470048904418944, "grad_norm": 0.02570729167200625, "loss": 0.6870627582073212, "time": 0.41326916217803955, "epoch": 11, "memory": 20334, "step": 61480} +{"lr": 0.0005050000000000001, "data_time": 0.0018560647964477538, "grad_norm": 0.012357833934947849, "loss": 0.686946451663971, "time": 0.36675879955291746, "epoch": 11, "memory": 20334, "step": 61580} +{"lr": 0.0005050000000000001, "data_time": 0.0015499353408813476, "grad_norm": 0.013128361222334206, "loss": 0.6843057632446289, "time": 0.4262408971786499, "epoch": 11, "memory": 20334, "step": 61680} +{"lr": 0.0005050000000000001, "data_time": 0.0018424510955810547, "grad_norm": 0.009496481786482037, "loss": 0.6876179873943329, "time": 0.41797921657562254, "epoch": 11, "memory": 20334, "step": 61780} +{"lr": 0.0005050000000000001, "data_time": 0.0015399932861328125, "grad_norm": 0.01483668175060302, "loss": 0.6902819156646729, "time": 0.41352152824401855, "epoch": 11, "memory": 20334, "step": 61880} +{"lr": 0.0005050000000000001, "data_time": 0.0017910480499267578, "grad_norm": 0.02472144942730665, "loss": 0.6849947333335876, "time": 0.4193438529968262, "epoch": 11, "memory": 20334, "step": 61980} +{"lr": 0.0005050000000000001, "data_time": 0.0018019437789916991, "grad_norm": 0.021300900680944323, "loss": 0.6857230484485626, "time": 0.3970093488693237, "epoch": 11, "memory": 20334, "step": 62080} +{"lr": 0.0005050000000000001, "data_time": 0.0017255544662475586, "grad_norm": 0.01563761797733605, "loss": 0.6862986862659455, "time": 0.4347070217132568, "epoch": 11, "memory": 20334, "step": 62180} +{"lr": 0.0005050000000000001, "data_time": 0.0018317222595214844, "grad_norm": 0.0190139097860083, "loss": 0.6858514189720154, "time": 0.39530124664306643, "epoch": 11, "memory": 20334, "step": 62280} +{"lr": 0.0005050000000000001, "data_time": 0.0018554449081420899, "grad_norm": 0.02175544567871839, "loss": 0.6858910858631134, "time": 0.42312207221984866, "epoch": 11, "memory": 20334, "step": 62380} +{"lr": 0.0005050000000000001, "data_time": 0.0017745256423950194, "grad_norm": 0.014436323108384386, "loss": 0.6903681099414826, "time": 0.3894811153411865, "epoch": 11, "memory": 20334, "step": 62480} +{"lr": 0.0005050000000000001, "data_time": 0.002657032012939453, "grad_norm": 0.01334753039991483, "loss": 0.6810441672801971, "time": 0.3817819833755493, "epoch": 11, "memory": 20334, "step": 62580} +{"lr": 0.0005050000000000001, "data_time": 0.0017103433609008789, "grad_norm": 0.01473056636750698, "loss": 0.6874148428440094, "time": 0.3831214189529419, "epoch": 11, "memory": 20334, "step": 62680} +{"lr": 0.0005050000000000001, "data_time": 0.002344846725463867, "grad_norm": 0.013010151410708205, "loss": 0.6864040493965149, "time": 0.3922775983810425, "epoch": 11, "memory": 20334, "step": 62780} +{"lr": 0.0005050000000000001, "data_time": 0.0015987396240234376, "grad_norm": 0.016429151548072694, "loss": 0.6879156172275543, "time": 0.40706620216369627, "epoch": 11, "memory": 20334, "step": 62880} +{"lr": 0.0005050000000000001, "data_time": 0.0017150402069091796, "grad_norm": 0.017281638015992938, "loss": 0.6896399796009064, "time": 0.3822291851043701, "epoch": 11, "memory": 20334, "step": 62980} +{"lr": 0.0005050000000000001, "data_time": 0.0017846107482910156, "grad_norm": 0.017714999988675116, "loss": 0.6918750524520874, "time": 0.40033955574035646, "epoch": 11, "memory": 20334, "step": 63080} +{"lr": 0.0005050000000000001, "data_time": 0.0018810987472534179, "grad_norm": 0.01451815478503704, "loss": 0.687272983789444, "time": 0.37504868507385253, "epoch": 11, "memory": 20334, "step": 63180} +{"lr": 0.0005050000000000001, "data_time": 0.0016415119171142578, "grad_norm": 0.01824620590778068, "loss": 0.6873231112957001, "time": 0.42091591358184816, "epoch": 11, "memory": 20334, "step": 63280} +{"accuracy/top1": 100.0, "data_time": 0.362206326590644, "time": 0.9878199895222982, "step": 11} +{"lr": 0.000427564939805086, "data_time": 0.001617121696472168, "grad_norm": 0.0166913520777598, "loss": 0.6899495005607605, "time": 0.417389440536499, "epoch": 12, "memory": 20334, "step": 63438} +{"lr": 0.000427564939805086, "data_time": 0.0016464471817016601, "grad_norm": 0.020788821484893562, "loss": 0.6836094558238983, "time": 0.39636070728302003, "epoch": 12, "memory": 20334, "step": 63538} +{"lr": 0.000427564939805086, "data_time": 0.0018493175506591798, "grad_norm": 0.015353060129564255, "loss": 0.6920715034008026, "time": 0.37181313037872316, "epoch": 12, "memory": 20334, "step": 63638} +{"lr": 0.000427564939805086, "data_time": 0.0017371416091918946, "grad_norm": 0.013098863716004416, "loss": 0.6874368488788605, "time": 0.39792301654815676, "epoch": 12, "memory": 20334, "step": 63738} +{"lr": 0.000427564939805086, "data_time": 0.0014843225479125976, "grad_norm": 0.01581031907116994, "loss": 0.6895247638225556, "time": 0.39028539657592776, "epoch": 12, "memory": 20334, "step": 63838} +{"lr": 0.000427564939805086, "data_time": 0.0016892433166503906, "grad_norm": 0.014902688539586961, "loss": 0.6877779304981232, "time": 0.3751551151275635, "epoch": 12, "memory": 20334, "step": 63938} +{"lr": 0.000427564939805086, "data_time": 0.001846170425415039, "grad_norm": 0.014566584955900908, "loss": 0.689047884941101, "time": 0.4367671966552734, "epoch": 12, "memory": 20334, "step": 64038} +{"lr": 0.000427564939805086, "data_time": 0.0017380237579345704, "grad_norm": 0.013999722059816122, "loss": 0.6841715455055237, "time": 0.4049127817153931, "epoch": 12, "memory": 20334, "step": 64138} +{"lr": 0.000427564939805086, "data_time": 0.0029029130935668947, "grad_norm": 0.017001294903457163, "loss": 0.68236603140831, "time": 0.3810136318206787, "epoch": 12, "memory": 20334, "step": 64238} +{"lr": 0.000427564939805086, "data_time": 0.001759958267211914, "grad_norm": 0.011180007155053318, "loss": 0.6864933133125305, "time": 0.3714949607849121, "epoch": 12, "memory": 20334, "step": 64338} +{"lr": 0.000427564939805086, "data_time": 0.0016855239868164063, "grad_norm": 0.0156432211631909, "loss": 0.6853608846664428, "time": 0.40211546421051025, "epoch": 12, "memory": 20334, "step": 64438} +{"lr": 0.000427564939805086, "data_time": 0.002300453186035156, "grad_norm": 0.018553055147640406, "loss": 0.6856110215187072, "time": 0.41217448711395266, "epoch": 12, "memory": 20334, "step": 64538} +{"lr": 0.000427564939805086, "data_time": 0.0014149904251098632, "grad_norm": 0.019725331536028533, "loss": 0.6861101984977722, "time": 0.41823592185974123, "epoch": 12, "memory": 20334, "step": 64638} +{"lr": 0.000427564939805086, "data_time": 0.0014590024948120117, "grad_norm": 0.015236181626096367, "loss": 0.6855660200119018, "time": 0.38270413875579834, "epoch": 12, "memory": 20334, "step": 64738} +{"lr": 0.000427564939805086, "data_time": 0.0014870166778564453, "grad_norm": 0.019289453083183618, "loss": 0.6875917494297028, "time": 0.4137756109237671, "epoch": 12, "memory": 20334, "step": 64838} +{"lr": 0.000427564939805086, "data_time": 0.002022981643676758, "grad_norm": 0.013446417148225009, "loss": 0.6886761307716369, "time": 0.3940241813659668, "epoch": 12, "memory": 20334, "step": 64938} +{"lr": 0.000427564939805086, "data_time": 0.0019438743591308593, "grad_norm": 0.023530560499057173, "loss": 0.6858498275279998, "time": 0.4304816484451294, "epoch": 12, "memory": 20334, "step": 65038} +{"lr": 0.000427564939805086, "data_time": 0.0018252134323120117, "grad_norm": 0.016393794189207257, "loss": 0.6866360604763031, "time": 0.408117938041687, "epoch": 12, "memory": 20334, "step": 65138} +{"lr": 0.000427564939805086, "data_time": 0.0020288467407226563, "grad_norm": 0.017317197937518358, "loss": 0.6868013143539429, "time": 0.41807188987731936, "epoch": 12, "memory": 20334, "step": 65238} +{"lr": 0.000427564939805086, "data_time": 0.001623225212097168, "grad_norm": 0.016890868052723818, "loss": 0.6861487030982971, "time": 0.40656893253326415, "epoch": 12, "memory": 20334, "step": 65338} +{"lr": 0.000427564939805086, "data_time": 0.0015668630599975585, "grad_norm": 0.021100576501339674, "loss": 0.6913497745990753, "time": 0.39169039726257326, "epoch": 12, "memory": 20334, "step": 65438} +{"lr": 0.000427564939805086, "data_time": 0.0016763925552368164, "grad_norm": 0.016157165495678784, "loss": 0.6850412070751191, "time": 0.4171649217605591, "epoch": 12, "memory": 20334, "step": 65538} +{"lr": 0.000427564939805086, "data_time": 0.0018799543380737306, "grad_norm": 0.012369199434760957, "loss": 0.6903493583202363, "time": 0.3820366144180298, "epoch": 12, "memory": 20334, "step": 65638} +{"lr": 0.000427564939805086, "data_time": 0.001929020881652832, "grad_norm": 0.022135058417916298, "loss": 0.6847064614295959, "time": 0.3993083477020264, "epoch": 12, "memory": 20334, "step": 65738} +{"lr": 0.000427564939805086, "data_time": 0.002015423774719238, "grad_norm": 0.012583296530647204, "loss": 0.6927791237831116, "time": 0.4088080406188965, "epoch": 12, "memory": 20334, "step": 65838} +{"lr": 0.000427564939805086, "data_time": 0.0018355131149291992, "grad_norm": 0.01793795293197036, "loss": 0.6878406882286072, "time": 0.3914525508880615, "epoch": 12, "memory": 20334, "step": 65938} +{"lr": 0.000427564939805086, "data_time": 0.002055716514587402, "grad_norm": 0.017172524356283247, "loss": 0.6883574783802032, "time": 0.39122986793518066, "epoch": 12, "memory": 20334, "step": 66038} +{"lr": 0.000427564939805086, "data_time": 0.0017808437347412109, "grad_norm": 0.018730479036457838, "loss": 0.6886959552764893, "time": 0.40156447887420654, "epoch": 12, "memory": 20334, "step": 66138} +{"lr": 0.000427564939805086, "data_time": 0.0017096281051635742, "grad_norm": 0.015269418933894486, "loss": 0.6835977256298065, "time": 0.3667266607284546, "epoch": 12, "memory": 20334, "step": 66238} +{"lr": 0.000427564939805086, "data_time": 0.0017436027526855468, "grad_norm": 0.012353344610892237, "loss": 0.689087700843811, "time": 0.4098337411880493, "epoch": 12, "memory": 20334, "step": 66338} +{"lr": 0.000427564939805086, "data_time": 0.002637624740600586, "grad_norm": 0.0182646015426144, "loss": 0.6834400355815887, "time": 0.413388991355896, "epoch": 12, "memory": 20334, "step": 66438} +{"lr": 0.000427564939805086, "data_time": 0.0018443584442138672, "grad_norm": 0.021364059206098317, "loss": 0.6865598857402802, "time": 0.40879254341125487, "epoch": 12, "memory": 20334, "step": 66538} +{"lr": 0.000427564939805086, "data_time": 0.0018166542053222657, "grad_norm": 0.017236367636360227, "loss": 0.6873547673225403, "time": 0.4330552339553833, "epoch": 12, "memory": 20334, "step": 66638} +{"lr": 0.000427564939805086, "data_time": 0.0015700340270996093, "grad_norm": 0.010844798694597558, "loss": 0.6891121685504913, "time": 0.4540198802947998, "epoch": 12, "memory": 20334, "step": 66738} +{"lr": 0.000427564939805086, "data_time": 0.0018373250961303711, "grad_norm": 0.015047562075778842, "loss": 0.6787995755672455, "time": 0.4149184226989746, "epoch": 12, "memory": 20334, "step": 66838} +{"lr": 0.000427564939805086, "data_time": 0.0017477989196777344, "grad_norm": 0.01790905058151111, "loss": 0.6833564043045044, "time": 0.42359721660614014, "epoch": 12, "memory": 20334, "step": 66938} +{"lr": 0.000427564939805086, "data_time": 0.0016760826110839844, "grad_norm": 0.010735026933252812, "loss": 0.689172500371933, "time": 0.4061639070510864, "epoch": 12, "memory": 20334, "step": 67038} +{"lr": 0.000427564939805086, "data_time": 0.0018133163452148438, "grad_norm": 0.016866380511783064, "loss": 0.6832768201828003, "time": 0.4003861904144287, "epoch": 12, "memory": 20334, "step": 67138} +{"lr": 0.000427564939805086, "data_time": 0.0018765926361083984, "grad_norm": 0.019166821404360233, "loss": 0.6863247394561768, "time": 0.4241545915603638, "epoch": 12, "memory": 20334, "step": 67238} +{"lr": 0.000427564939805086, "data_time": 0.001663970947265625, "grad_norm": 0.018017438339302316, "loss": 0.6898203909397125, "time": 0.4237964630126953, "epoch": 12, "memory": 20334, "step": 67338} +{"lr": 0.000427564939805086, "data_time": 0.0018258333206176759, "grad_norm": 0.017352794483304023, "loss": 0.684254378080368, "time": 0.3921781539916992, "epoch": 12, "memory": 20334, "step": 67438} +{"lr": 0.000427564939805086, "data_time": 0.0017937183380126952, "grad_norm": 0.016346360580064355, "loss": 0.6855483055114746, "time": 0.462936544418335, "epoch": 12, "memory": 20334, "step": 67538} +{"lr": 0.000427564939805086, "data_time": 0.0019580602645874025, "grad_norm": 0.00908320113667287, "loss": 0.6874360084533692, "time": 0.4015927791595459, "epoch": 12, "memory": 20334, "step": 67638} +{"lr": 0.000427564939805086, "data_time": 0.0017687320709228516, "grad_norm": 0.018595830723643302, "loss": 0.6868012547492981, "time": 0.3768711805343628, "epoch": 12, "memory": 20334, "step": 67738} +{"lr": 0.000427564939805086, "data_time": 0.0017295122146606446, "grad_norm": 0.01453996414784342, "loss": 0.6850512862205506, "time": 0.4595906972885132, "epoch": 12, "memory": 20334, "step": 67838} +{"lr": 0.000427564939805086, "data_time": 0.0018616676330566405, "grad_norm": 0.020302220573648812, "loss": 0.6899466037750244, "time": 0.38662493228912354, "epoch": 12, "memory": 20334, "step": 67938} +{"lr": 0.000427564939805086, "data_time": 0.0016736030578613282, "grad_norm": 0.024154767947038636, "loss": 0.6864497721195221, "time": 0.39399864673614504, "epoch": 12, "memory": 20334, "step": 68038} +{"lr": 0.000427564939805086, "data_time": 0.0018260717391967774, "grad_norm": 0.01718618874438107, "loss": 0.6876361727714538, "time": 0.4071578741073608, "epoch": 12, "memory": 20334, "step": 68138} +{"lr": 0.000427564939805086, "data_time": 0.0021019935607910155, "grad_norm": 0.02005452015437186, "loss": 0.6836853265762329, "time": 0.4368730068206787, "epoch": 12, "memory": 20334, "step": 68238} +{"lr": 0.000427564939805086, "data_time": 0.0017063140869140625, "grad_norm": 0.011229679710231722, "loss": 0.6864103615283966, "time": 0.4182861328125, "epoch": 12, "memory": 20334, "step": 68338} +{"lr": 0.000427564939805086, "data_time": 0.001926589012145996, "grad_norm": 0.01349624409340322, "loss": 0.6893961191177368, "time": 0.40487518310546877, "epoch": 12, "memory": 20334, "step": 68438} +{"lr": 0.000427564939805086, "data_time": 0.0017902612686157226, "grad_norm": 0.013032496441155672, "loss": 0.6877645134925843, "time": 0.38978354930877684, "epoch": 12, "memory": 20334, "step": 68538} +{"lr": 0.000427564939805086, "data_time": 0.0018377065658569335, "grad_norm": 0.014521173760294914, "loss": 0.6885558724403381, "time": 0.3954671621322632, "epoch": 12, "memory": 20334, "step": 68638} +{"lr": 0.000427564939805086, "data_time": 0.001640653610229492, "grad_norm": 0.02059962465427816, "loss": 0.690975034236908, "time": 0.3865856885910034, "epoch": 12, "memory": 20334, "step": 68738} +{"lr": 0.000427564939805086, "data_time": 0.002003645896911621, "grad_norm": 0.015047602856066078, "loss": 0.6892555713653564, "time": 0.42522027492523196, "epoch": 12, "memory": 20334, "step": 68838} +{"lr": 0.000427564939805086, "data_time": 0.0020319461822509766, "grad_norm": 0.015423443308100105, "loss": 0.6877327144145966, "time": 0.4105898141860962, "epoch": 12, "memory": 20334, "step": 68938} +{"lr": 0.000427564939805086, "data_time": 0.0016184329986572265, "grad_norm": 0.014972370746545494, "loss": 0.682448822259903, "time": 0.3690106153488159, "epoch": 12, "memory": 20334, "step": 69038} +{"accuracy/top1": 100.0, "data_time": 0.35881156391567653, "time": 0.9850935935974121, "step": 12} +{"lr": 0.0003520365877844012, "data_time": 0.002499818801879883, "grad_norm": 0.01408783111255616, "loss": 0.686473298072815, "time": 0.42690041065216067, "epoch": 13, "memory": 20334, "step": 69196} +{"lr": 0.0003520365877844012, "data_time": 0.0016981601715087891, "grad_norm": 0.010116408905014395, "loss": 0.6877528607845307, "time": 0.40020408630371096, "epoch": 13, "memory": 20334, "step": 69296} +{"lr": 0.0003520365877844012, "data_time": 0.0014430046081542968, "grad_norm": 0.009993621555622667, "loss": 0.686174887418747, "time": 0.4011565923690796, "epoch": 13, "memory": 20334, "step": 69396} +{"lr": 0.0003520365877844012, "data_time": 0.0024167299270629883, "grad_norm": 0.012839860911481083, "loss": 0.6864809036254883, "time": 0.3791086196899414, "epoch": 13, "memory": 20334, "step": 69496} +{"lr": 0.0003520365877844012, "data_time": 0.001960611343383789, "grad_norm": 0.017129956278949976, "loss": 0.6864925742149353, "time": 0.42064590454101564, "epoch": 13, "memory": 20334, "step": 69596} +{"lr": 0.0003520365877844012, "data_time": 0.0022573947906494142, "grad_norm": 0.009476678224746137, "loss": 0.6879218339920044, "time": 0.40629425048828127, "epoch": 13, "memory": 20334, "step": 69696} +{"lr": 0.0003520365877844012, "data_time": 0.0020205497741699217, "grad_norm": 0.02477117571979761, "loss": 0.6804328858852386, "time": 0.4160814046859741, "epoch": 13, "memory": 20334, "step": 69796} +{"lr": 0.0003520365877844012, "data_time": 0.002030086517333984, "grad_norm": 0.013945743715157732, "loss": 0.6856885552406311, "time": 0.37984271049499513, "epoch": 13, "memory": 20334, "step": 69896} +{"lr": 0.0003520365877844012, "data_time": 0.001855158805847168, "grad_norm": 0.017967581457924098, "loss": 0.6900890290737152, "time": 0.41427340507507326, "epoch": 13, "memory": 20334, "step": 69996} +{"lr": 0.0003520365877844012, "data_time": 0.001960158348083496, "grad_norm": 0.014855450368486345, "loss": 0.6853561997413635, "time": 0.3758256196975708, "epoch": 13, "memory": 20334, "step": 70096} +{"lr": 0.0003520365877844012, "data_time": 0.0017795562744140625, "grad_norm": 0.015406798385083676, "loss": 0.6900947272777558, "time": 0.40915162563323976, "epoch": 13, "memory": 20334, "step": 70196} +{"lr": 0.0003520365877844012, "data_time": 0.0017820358276367187, "grad_norm": 0.009939678665250539, "loss": 0.686232042312622, "time": 0.40437347888946534, "epoch": 13, "memory": 20334, "step": 70296} +{"lr": 0.0003520365877844012, "data_time": 0.0017084360122680664, "grad_norm": 0.015425244870129973, "loss": 0.6870243191719055, "time": 0.3940648078918457, "epoch": 13, "memory": 20334, "step": 70396} +{"lr": 0.0003520365877844012, "data_time": 0.00237889289855957, "grad_norm": 0.022950517875142396, "loss": 0.6841873228549957, "time": 0.396422815322876, "epoch": 13, "memory": 20334, "step": 70496} +{"lr": 0.0003520365877844012, "data_time": 0.001589202880859375, "grad_norm": 0.01524299723096192, "loss": 0.6911931157112121, "time": 0.39834606647491455, "epoch": 13, "memory": 20334, "step": 70596} +{"lr": 0.0003520365877844012, "data_time": 0.0016993999481201172, "grad_norm": 0.011118559376336634, "loss": 0.687118124961853, "time": 0.3648622274398804, "epoch": 13, "memory": 20334, "step": 70696} +{"lr": 0.0003520365877844012, "data_time": 0.0017743349075317384, "grad_norm": 0.018272021901793778, "loss": 0.6872730374336242, "time": 0.3884437084197998, "epoch": 13, "memory": 20334, "step": 70796} +{"lr": 0.0003520365877844012, "data_time": 0.0017374753952026367, "grad_norm": 0.00859234919771552, "loss": 0.6863044977188111, "time": 0.40339174270629885, "epoch": 13, "memory": 20334, "step": 70896} +{"lr": 0.0003520365877844012, "data_time": 0.0016998767852783204, "grad_norm": 0.01648919393774122, "loss": 0.6863160371780396, "time": 0.39385557174682617, "epoch": 13, "memory": 20334, "step": 70996} +{"lr": 0.0003520365877844012, "data_time": 0.0017043828964233398, "grad_norm": 0.013088628451805561, "loss": 0.6838694214820862, "time": 0.4203041076660156, "epoch": 13, "memory": 20334, "step": 71096} +{"lr": 0.0003520365877844012, "data_time": 0.0020523309707641602, "grad_norm": 0.020921247568912803, "loss": 0.6896158695220947, "time": 0.42620530128479006, "epoch": 13, "memory": 20334, "step": 71196} +{"lr": 0.0003520365877844012, "data_time": 0.0018886327743530273, "grad_norm": 0.01668210669886321, "loss": 0.6838137149810791, "time": 0.39089982509613036, "epoch": 13, "memory": 20334, "step": 71296} +{"lr": 0.0003520365877844012, "data_time": 0.0019776582717895507, "grad_norm": 0.020049914065748454, "loss": 0.6866389572620392, "time": 0.39708170890808103, "epoch": 13, "memory": 20334, "step": 71396} +{"lr": 0.0003520365877844012, "data_time": 0.0019255399703979493, "grad_norm": 0.011692477995529771, "loss": 0.685070377588272, "time": 0.40939884185791015, "epoch": 13, "memory": 20334, "step": 71496} +{"lr": 0.0003520365877844012, "data_time": 0.0024569034576416016, "grad_norm": 0.014183701551519334, "loss": 0.6869688928127289, "time": 0.38482298851013186, "epoch": 13, "memory": 20334, "step": 71596} +{"lr": 0.0003520365877844012, "data_time": 0.0018474102020263673, "grad_norm": 0.017426353041082622, "loss": 0.6850690066814422, "time": 0.4538242340087891, "epoch": 13, "memory": 20334, "step": 71696} +{"lr": 0.0003520365877844012, "data_time": 0.0017541885375976563, "grad_norm": 0.016810492976219393, "loss": 0.6873281717300415, "time": 0.3970121622085571, "epoch": 13, "memory": 20334, "step": 71796} +{"lr": 0.0003520365877844012, "data_time": 0.0013822078704833984, "grad_norm": 0.01345356865786016, "loss": 0.6868139803409576, "time": 0.40201380252838137, "epoch": 13, "memory": 20334, "step": 71896} +{"lr": 0.0003520365877844012, "data_time": 0.0013040542602539063, "grad_norm": 0.015152580151334406, "loss": 0.683145421743393, "time": 0.39615478515625, "epoch": 13, "memory": 20334, "step": 71996} +{"lr": 0.0003520365877844012, "data_time": 0.002103257179260254, "grad_norm": 0.015875039622187614, "loss": 0.6846316158771515, "time": 0.4416754722595215, "epoch": 13, "memory": 20334, "step": 72096} +{"lr": 0.0003520365877844012, "data_time": 0.0019967317581176757, "grad_norm": 0.013362546951975673, "loss": 0.6818894207477569, "time": 0.42512590885162355, "epoch": 13, "memory": 20334, "step": 72196} +{"lr": 0.0003520365877844012, "data_time": 0.0018580436706542968, "grad_norm": 0.018948213150724768, "loss": 0.685666698217392, "time": 0.40195112228393554, "epoch": 13, "memory": 20334, "step": 72296} +{"lr": 0.0003520365877844012, "data_time": 0.0016201972961425782, "grad_norm": 0.011216734326444567, "loss": 0.6902041018009186, "time": 0.4200406789779663, "epoch": 13, "memory": 20334, "step": 72396} +{"lr": 0.0003520365877844012, "data_time": 0.0016074657440185546, "grad_norm": 0.015482333645923062, "loss": 0.6882450044155121, "time": 0.38861067295074464, "epoch": 13, "memory": 20334, "step": 72496} +{"lr": 0.0003520365877844012, "data_time": 0.0019302129745483398, "grad_norm": 0.018153125431854277, "loss": 0.6857194900512695, "time": 0.4287428617477417, "epoch": 13, "memory": 20334, "step": 72596} +{"lr": 0.0003520365877844012, "data_time": 0.0017689228057861327, "grad_norm": 0.016061562206596137, "loss": 0.6880880415439605, "time": 0.4203546762466431, "epoch": 13, "memory": 20334, "step": 72696} +{"lr": 0.0003520365877844012, "data_time": 0.001614546775817871, "grad_norm": 0.01742428051074967, "loss": 0.6847258329391479, "time": 0.47726438045501707, "epoch": 13, "memory": 20334, "step": 72796} +{"lr": 0.0003520365877844012, "data_time": 0.001859736442565918, "grad_norm": 0.018289669835940003, "loss": 0.6820207595825195, "time": 0.42640645503997804, "epoch": 13, "memory": 20334, "step": 72896} +{"lr": 0.0003520365877844012, "data_time": 0.0017515659332275391, "grad_norm": 0.016496698174159975, "loss": 0.6912059962749482, "time": 0.41557989120483396, "epoch": 13, "memory": 20334, "step": 72996} +{"lr": 0.0003520365877844012, "data_time": 0.001747441291809082, "grad_norm": 0.019221892394125462, "loss": 0.6915319621562958, "time": 0.41441380977630615, "epoch": 13, "memory": 20334, "step": 73096} +{"lr": 0.0003520365877844012, "data_time": 0.0016184091567993165, "grad_norm": 0.015028754225932061, "loss": 0.6839554131031036, "time": 0.42982704639434816, "epoch": 13, "memory": 20334, "step": 73196} +{"lr": 0.0003520365877844012, "data_time": 0.0016668319702148437, "grad_norm": 0.014722906588576733, "loss": 0.6887838482856751, "time": 0.4043821096420288, "epoch": 13, "memory": 20334, "step": 73296} +{"lr": 0.0003520365877844012, "data_time": 0.0016119956970214843, "grad_norm": 0.013450360629940405, "loss": 0.6861825823783875, "time": 0.4213541030883789, "epoch": 13, "memory": 20334, "step": 73396} +{"lr": 0.0003520365877844012, "data_time": 0.0018059968948364257, "grad_norm": 0.015405056823510677, "loss": 0.6865025579929351, "time": 0.4439688682556152, "epoch": 13, "memory": 20334, "step": 73496} +{"lr": 0.0003520365877844012, "data_time": 0.0015944719314575195, "grad_norm": 0.013549680833239108, "loss": 0.6879826962947846, "time": 0.40688927173614503, "epoch": 13, "memory": 20334, "step": 73596} +{"lr": 0.0003520365877844012, "data_time": 0.0017857074737548829, "grad_norm": 0.014822281093802302, "loss": 0.6874355852603913, "time": 0.39173383712768556, "epoch": 13, "memory": 20334, "step": 73696} +{"lr": 0.0003520365877844012, "data_time": 0.0019335746765136719, "grad_norm": 0.01581891911337152, "loss": 0.6909354984760284, "time": 0.38756699562072755, "epoch": 13, "memory": 20334, "step": 73796} +{"lr": 0.0003520365877844012, "data_time": 0.0016683340072631836, "grad_norm": 0.015744145837379618, "loss": 0.6887045919895172, "time": 0.388281512260437, "epoch": 13, "memory": 20334, "step": 73896} +{"lr": 0.0003520365877844012, "data_time": 0.0018504858016967773, "grad_norm": 0.010446198028512299, "loss": 0.6860404908657074, "time": 0.39574370384216306, "epoch": 13, "memory": 20334, "step": 73996} +{"lr": 0.0003520365877844012, "data_time": 0.0015665531158447266, "grad_norm": 0.017069025873206557, "loss": 0.6907217621803283, "time": 0.4279362440109253, "epoch": 13, "memory": 20334, "step": 74096} +{"lr": 0.0003520365877844012, "data_time": 0.0017962932586669921, "grad_norm": 0.013988114194944501, "loss": 0.6830148696899414, "time": 0.40684354305267334, "epoch": 13, "memory": 20334, "step": 74196} +{"lr": 0.0003520365877844012, "data_time": 0.0016621828079223632, "grad_norm": 0.016329382685944437, "loss": 0.6807142078876496, "time": 0.3933048486709595, "epoch": 13, "memory": 20334, "step": 74296} +{"lr": 0.0003520365877844012, "data_time": 0.001727461814880371, "grad_norm": 0.012931671366095543, "loss": 0.6895102798938751, "time": 0.3785771608352661, "epoch": 13, "memory": 20334, "step": 74396} +{"lr": 0.0003520365877844012, "data_time": 0.0019137144088745117, "grad_norm": 0.012048048479482532, "loss": 0.6844245970249176, "time": 0.3915132522583008, "epoch": 13, "memory": 20334, "step": 74496} +{"lr": 0.0003520365877844012, "data_time": 0.0018544197082519531, "grad_norm": 0.019758155196905137, "loss": 0.6845782697200775, "time": 0.4117830514907837, "epoch": 13, "memory": 20334, "step": 74596} +{"lr": 0.0003520365877844012, "data_time": 0.002025890350341797, "grad_norm": 0.0192345077637583, "loss": 0.6796214640140533, "time": 0.4155980348587036, "epoch": 13, "memory": 20334, "step": 74696} +{"lr": 0.0003520365877844012, "data_time": 0.002515244483947754, "grad_norm": 0.01954756546765566, "loss": 0.6879429399967194, "time": 0.40151567459106446, "epoch": 13, "memory": 20334, "step": 74796} +{"accuracy/top1": 100.0, "data_time": 0.3618520365820991, "time": 0.9855987760755751, "step": 13} +{"lr": 0.00028027470262892445, "data_time": 0.0015807390213012696, "grad_norm": 0.011704668845050037, "loss": 0.6839596807956696, "time": 0.4051476716995239, "epoch": 14, "memory": 20334, "step": 74954} +{"lr": 0.00028027470262892445, "data_time": 0.0017504453659057616, "grad_norm": 0.018872018868569283, "loss": 0.6887854754924774, "time": 0.40489964485168456, "epoch": 14, "memory": 20334, "step": 75054} +{"lr": 0.00028027470262892445, "data_time": 0.0016129255294799806, "grad_norm": 0.014565573446452618, "loss": 0.6866371989250183, "time": 0.41884598731994627, "epoch": 14, "memory": 20334, "step": 75154} +{"lr": 0.00028027470262892445, "data_time": 0.001651167869567871, "grad_norm": 0.01385259844828397, "loss": 0.6840680539608002, "time": 0.40360150337219236, "epoch": 14, "memory": 20334, "step": 75254} +{"lr": 0.00028027470262892445, "data_time": 0.001611018180847168, "grad_norm": 0.01756176436319947, "loss": 0.6900853455066681, "time": 0.4007642984390259, "epoch": 14, "memory": 20334, "step": 75354} +{"lr": 0.00028027470262892445, "data_time": 0.0017755746841430664, "grad_norm": 0.016391893127001823, "loss": 0.6842569053173065, "time": 0.37480669021606444, "epoch": 14, "memory": 20334, "step": 75454} +{"lr": 0.00028027470262892445, "data_time": 0.0017299652099609375, "grad_norm": 0.014681549975648522, "loss": 0.6853827178478241, "time": 0.39731364250183104, "epoch": 14, "memory": 20334, "step": 75554} +{"lr": 0.00028027470262892445, "data_time": 0.0020262956619262694, "grad_norm": 0.01204390093917027, "loss": 0.6880824983119964, "time": 0.39176321029663086, "epoch": 14, "memory": 20334, "step": 75654} +{"lr": 0.00028027470262892445, "data_time": 0.001528453826904297, "grad_norm": 0.012430070911068469, "loss": 0.6872732996940613, "time": 0.4073626518249512, "epoch": 14, "memory": 20334, "step": 75754} +{"lr": 0.00028027470262892445, "data_time": 0.0016721010208129883, "grad_norm": 0.012954751891084016, "loss": 0.6828380167484284, "time": 0.4160911083221436, "epoch": 14, "memory": 20334, "step": 75854} +{"lr": 0.00028027470262892445, "data_time": 0.0016779184341430664, "grad_norm": 0.018801445607095957, "loss": 0.6833130300045014, "time": 0.4033613920211792, "epoch": 14, "memory": 20334, "step": 75954} +{"lr": 0.00028027470262892445, "data_time": 0.0014740943908691407, "grad_norm": 0.013314147177152336, "loss": 0.6858446300029755, "time": 0.40171449184417723, "epoch": 14, "memory": 20334, "step": 76054} +{"lr": 0.00028027470262892445, "data_time": 0.0016918420791625977, "grad_norm": 0.021815078053623437, "loss": 0.6842347383499146, "time": 0.38399443626403806, "epoch": 14, "memory": 20334, "step": 76154} +{"lr": 0.00028027470262892445, "data_time": 0.002535080909729004, "grad_norm": 0.021698733745142817, "loss": 0.6863724827766419, "time": 0.3649530649185181, "epoch": 14, "memory": 20334, "step": 76254} +{"lr": 0.00028027470262892445, "data_time": 0.0022377729415893554, "grad_norm": 0.017168973269872368, "loss": 0.6841258049011231, "time": 0.39482784271240234, "epoch": 14, "memory": 20334, "step": 76354} +{"lr": 0.00028027470262892445, "data_time": 0.00183560848236084, "grad_norm": 0.016725882625905797, "loss": 0.6881155669689178, "time": 0.41203043460845945, "epoch": 14, "memory": 20334, "step": 76454} +{"lr": 0.00028027470262892445, "data_time": 0.0018941640853881836, "grad_norm": 0.01606611717143096, "loss": 0.6885165214538574, "time": 0.44018542766571045, "epoch": 14, "memory": 20334, "step": 76554} +{"lr": 0.00028027470262892445, "data_time": 0.002838468551635742, "grad_norm": 0.016217493638396262, "loss": 0.684929758310318, "time": 0.37801151275634765, "epoch": 14, "memory": 20334, "step": 76654} +{"lr": 0.00028027470262892445, "data_time": 0.0017352342605590821, "grad_norm": 0.015280984248965979, "loss": 0.6845894157886505, "time": 0.41396141052246094, "epoch": 14, "memory": 20334, "step": 76754} +{"lr": 0.00028027470262892445, "data_time": 0.001906132698059082, "grad_norm": 0.021990169165655972, "loss": 0.6852270126342773, "time": 0.38214874267578125, "epoch": 14, "memory": 20334, "step": 76854} +{"lr": 0.00028027470262892445, "data_time": 0.0031803131103515627, "grad_norm": 0.016700853826478124, "loss": 0.6877544343471527, "time": 0.4418198585510254, "epoch": 14, "memory": 20334, "step": 76954} +{"lr": 0.00028027470262892445, "data_time": 0.001935577392578125, "grad_norm": 0.01613506133435294, "loss": 0.6831979990005493, "time": 0.3728139638900757, "epoch": 14, "memory": 20334, "step": 77054} +{"lr": 0.00028027470262892445, "data_time": 0.0020687103271484373, "grad_norm": 0.017249015718698503, "loss": 0.6869511902332306, "time": 0.386832594871521, "epoch": 14, "memory": 20334, "step": 77154} +{"lr": 0.00028027470262892445, "data_time": 0.001937556266784668, "grad_norm": 0.01789760193787515, "loss": 0.6842350125312805, "time": 0.3791035652160645, "epoch": 14, "memory": 20334, "step": 77254} +{"lr": 0.00028027470262892445, "data_time": 0.002295565605163574, "grad_norm": 0.016737726907012983, "loss": 0.6866466999053955, "time": 0.41111068725585936, "epoch": 14, "memory": 20334, "step": 77354} +{"lr": 0.00028027470262892445, "data_time": 0.002791619300842285, "grad_norm": 0.016068597487173975, "loss": 0.6885709464550018, "time": 0.4092714309692383, "epoch": 14, "memory": 20334, "step": 77454} +{"lr": 0.00028027470262892445, "data_time": 0.0019040584564208984, "grad_norm": 0.014783918834291398, "loss": 0.6855606198310852, "time": 0.3963749647140503, "epoch": 14, "memory": 20334, "step": 77554} +{"lr": 0.00028027470262892445, "data_time": 0.0022172927856445312, "grad_norm": 0.012944332987535745, "loss": 0.6867958188056946, "time": 0.40024442672729493, "epoch": 14, "memory": 20334, "step": 77654} +{"lr": 0.00028027470262892445, "data_time": 0.001626729965209961, "grad_norm": 0.019447324564680457, "loss": 0.6816729784011841, "time": 0.4012739896774292, "epoch": 14, "memory": 20334, "step": 77754} +{"lr": 0.00028027470262892445, "data_time": 0.0030068159103393555, "grad_norm": 0.011175620427820832, "loss": 0.680833488702774, "time": 0.3890573024749756, "epoch": 14, "memory": 20334, "step": 77854} +{"lr": 0.00028027470262892445, "data_time": 0.0022169828414916994, "grad_norm": 0.013994936249218881, "loss": 0.6875779330730438, "time": 0.3894374847412109, "epoch": 14, "memory": 20334, "step": 77954} +{"lr": 0.00028027470262892445, "data_time": 0.0018175125122070312, "grad_norm": 0.014813157729804516, "loss": 0.6880999803543091, "time": 0.4051444292068481, "epoch": 14, "memory": 20334, "step": 78054} +{"lr": 0.00028027470262892445, "data_time": 0.0026232719421386717, "grad_norm": 0.016581385885365306, "loss": 0.6841295242309571, "time": 0.44101567268371583, "epoch": 14, "memory": 20334, "step": 78154} +{"lr": 0.00028027470262892445, "data_time": 0.0017164230346679687, "grad_norm": 0.021206063264980913, "loss": 0.688108503818512, "time": 0.39741859436035154, "epoch": 14, "memory": 20334, "step": 78254} +{"lr": 0.00028027470262892445, "data_time": 0.002466726303100586, "grad_norm": 0.013205208000726998, "loss": 0.6878928959369659, "time": 0.4076138734817505, "epoch": 14, "memory": 20334, "step": 78354} +{"lr": 0.00028027470262892445, "data_time": 0.002306079864501953, "grad_norm": 0.02017670273780823, "loss": 0.6877880394458771, "time": 0.4074725151062012, "epoch": 14, "memory": 20334, "step": 78454} +{"lr": 0.00028027470262892445, "data_time": 0.0017530441284179688, "grad_norm": 0.016754168551415205, "loss": 0.6903860211372376, "time": 0.422964072227478, "epoch": 14, "memory": 20334, "step": 78554} +{"lr": 0.00028027470262892445, "data_time": 0.001977968215942383, "grad_norm": 0.020933930505998434, "loss": 0.6920530259609222, "time": 0.40083434581756594, "epoch": 14, "memory": 20334, "step": 78654} +{"lr": 0.00028027470262892445, "data_time": 0.002583169937133789, "grad_norm": 0.01593464156612754, "loss": 0.6848901391029358, "time": 0.4061759948730469, "epoch": 14, "memory": 20334, "step": 78754} +{"lr": 0.00028027470262892445, "data_time": 0.002393364906311035, "grad_norm": 0.014148953440599143, "loss": 0.6822631180286407, "time": 0.4196335792541504, "epoch": 14, "memory": 20334, "step": 78854} +{"lr": 0.00028027470262892445, "data_time": 0.0028033018112182616, "grad_norm": 0.015457310376223176, "loss": 0.6922883689403534, "time": 0.42890777587890627, "epoch": 14, "memory": 20334, "step": 78954} +{"lr": 0.00028027470262892445, "data_time": 0.0022582769393920898, "grad_norm": 0.016887045186012983, "loss": 0.6882308006286622, "time": 0.41818015575408934, "epoch": 14, "memory": 20334, "step": 79054} +{"lr": 0.00028027470262892445, "data_time": 0.0021661996841430666, "grad_norm": 0.016308948502410204, "loss": 0.6863297164440155, "time": 0.45968387126922605, "epoch": 14, "memory": 20334, "step": 79154} +{"lr": 0.00028027470262892445, "data_time": 0.002317214012145996, "grad_norm": 0.016493112477473915, "loss": 0.6870655953884125, "time": 0.43414759635925293, "epoch": 14, "memory": 20334, "step": 79254} +{"lr": 0.00028027470262892445, "data_time": 0.0023792743682861327, "grad_norm": 0.017419931944459677, "loss": 0.6883424043655395, "time": 0.382624077796936, "epoch": 14, "memory": 20334, "step": 79354} +{"lr": 0.00028027470262892445, "data_time": 0.0028496026992797852, "grad_norm": 0.01771250255405903, "loss": 0.6872944593429565, "time": 0.403963565826416, "epoch": 14, "memory": 20334, "step": 79454} +{"lr": 0.00028027470262892445, "data_time": 0.0028537511825561523, "grad_norm": 0.013814081810414791, "loss": 0.684880667924881, "time": 0.40574567317962645, "epoch": 14, "memory": 20334, "step": 79554} +{"lr": 0.00028027470262892445, "data_time": 0.0017877340316772461, "grad_norm": 0.014335823338478803, "loss": 0.684401661157608, "time": 0.3935187578201294, "epoch": 14, "memory": 20334, "step": 79654} +{"lr": 0.00028027470262892445, "data_time": 0.002208518981933594, "grad_norm": 0.014017370890360325, "loss": 0.6845682501792908, "time": 0.38803040981292725, "epoch": 14, "memory": 20334, "step": 79754} +{"lr": 0.00028027470262892445, "data_time": 0.0017626762390136718, "grad_norm": 0.013043110590660944, "loss": 0.6860217750072479, "time": 0.4062116861343384, "epoch": 14, "memory": 20334, "step": 79854} +{"lr": 0.00028027470262892445, "data_time": 0.0015872955322265626, "grad_norm": 0.01832923444453627, "loss": 0.6916180849075317, "time": 0.41204280853271485, "epoch": 14, "memory": 20334, "step": 79954} +{"lr": 0.00028027470262892445, "data_time": 0.0027703046798706055, "grad_norm": 0.020153836393728853, "loss": 0.6901819586753846, "time": 0.38084139823913576, "epoch": 14, "memory": 20334, "step": 80054} +{"lr": 0.00028027470262892445, "data_time": 0.002006077766418457, "grad_norm": 0.018345507606863976, "loss": 0.6888871848583221, "time": 0.4347590684890747, "epoch": 14, "memory": 20334, "step": 80154} +{"lr": 0.00028027470262892445, "data_time": 0.0017123222351074219, "grad_norm": 0.01874561405275017, "loss": 0.6875796914100647, "time": 0.40040853023529055, "epoch": 14, "memory": 20334, "step": 80254} +{"lr": 0.00028027470262892445, "data_time": 0.0020253896713256837, "grad_norm": 0.020681688887998463, "loss": 0.6857459127902985, "time": 0.4483737230300903, "epoch": 14, "memory": 20334, "step": 80354} +{"lr": 0.00028027470262892445, "data_time": 0.0018351078033447266, "grad_norm": 0.018248191615566612, "loss": 0.6877990126609802, "time": 0.39775097370147705, "epoch": 14, "memory": 20334, "step": 80454} +{"lr": 0.00028027470262892445, "data_time": 0.001861405372619629, "grad_norm": 0.014884584629908205, "loss": 0.685151606798172, "time": 0.4089222431182861, "epoch": 14, "memory": 20334, "step": 80554} +{"accuracy/top1": 100.0, "data_time": 0.36675238609313965, "time": 0.9948916170332167, "step": 14} +{"lr": 0.0002140463001152259, "data_time": 0.002021336555480957, "grad_norm": 0.020388060808181764, "loss": 0.6880136609077454, "time": 0.39930334091186526, "epoch": 15, "memory": 20334, "step": 80712} +{"lr": 0.0002140463001152259, "data_time": 0.0017746210098266602, "grad_norm": 0.013172862562350928, "loss": 0.6877367615699768, "time": 0.4345247745513916, "epoch": 15, "memory": 20334, "step": 80812} +{"lr": 0.0002140463001152259, "data_time": 0.002066946029663086, "grad_norm": 0.020787785411812365, "loss": 0.6855421900749207, "time": 0.4268107652664185, "epoch": 15, "memory": 20334, "step": 80912} +{"lr": 0.0002140463001152259, "data_time": 0.003052163124084473, "grad_norm": 0.01730718093458563, "loss": 0.6878939151763916, "time": 0.40036122798919677, "epoch": 15, "memory": 20334, "step": 81012} +{"lr": 0.0002140463001152259, "data_time": 0.0019379854202270508, "grad_norm": 0.012013878964353353, "loss": 0.6882242381572723, "time": 0.37637958526611326, "epoch": 15, "memory": 20334, "step": 81112} +{"lr": 0.0002140463001152259, "data_time": 0.0019441604614257812, "grad_norm": 0.010277168377069757, "loss": 0.6887440383434296, "time": 0.39629502296447755, "epoch": 15, "memory": 20334, "step": 81212} +{"lr": 0.0002140463001152259, "data_time": 0.0019289255142211914, "grad_norm": 0.01941108452156186, "loss": 0.687767618894577, "time": 0.4071494102478027, "epoch": 15, "memory": 20334, "step": 81312} +{"lr": 0.0002140463001152259, "data_time": 0.0019929170608520507, "grad_norm": 0.017288946313783526, "loss": 0.6901023149490356, "time": 0.42064311504364016, "epoch": 15, "memory": 20334, "step": 81412} +{"lr": 0.0002140463001152259, "data_time": 0.0017107486724853515, "grad_norm": 0.025121022341772915, "loss": 0.6882579863071442, "time": 0.39304091930389407, "epoch": 15, "memory": 20334, "step": 81512} +{"lr": 0.0002140463001152259, "data_time": 0.0016469955444335938, "grad_norm": 0.024632679484784602, "loss": 0.6852174937725067, "time": 0.3752723455429077, "epoch": 15, "memory": 20334, "step": 81612} +{"lr": 0.0002140463001152259, "data_time": 0.002080368995666504, "grad_norm": 0.012507924216333777, "loss": 0.6829984188079834, "time": 0.4012298583984375, "epoch": 15, "memory": 20334, "step": 81712} +{"lr": 0.0002140463001152259, "data_time": 0.0017494440078735351, "grad_norm": 0.018673277169000357, "loss": 0.6915401458740235, "time": 0.3802367687225342, "epoch": 15, "memory": 20334, "step": 81812} +{"lr": 0.0002140463001152259, "data_time": 0.0018441438674926757, "grad_norm": 0.011053581692976877, "loss": 0.6896061718463897, "time": 0.4221057415008545, "epoch": 15, "memory": 20334, "step": 81912} +{"lr": 0.0002140463001152259, "data_time": 0.0019703149795532227, "grad_norm": 0.02195824042428285, "loss": 0.687947416305542, "time": 0.4112419128417969, "epoch": 15, "memory": 20334, "step": 82012} +{"lr": 0.0002140463001152259, "data_time": 0.0026426315307617188, "grad_norm": 0.02192495160270482, "loss": 0.6829129338264466, "time": 0.4389740705490112, "epoch": 15, "memory": 20334, "step": 82112} +{"lr": 0.0002140463001152259, "data_time": 0.0015482664108276366, "grad_norm": 0.72354477616027, "loss": 0.6864808142185211, "time": 0.3933025121688843, "epoch": 15, "memory": 20334, "step": 82212} +{"lr": 0.0002140463001152259, "data_time": 0.0014982223510742188, "grad_norm": 0.019350684783421456, "loss": 0.6884127199649811, "time": 0.39441399574279784, "epoch": 15, "memory": 20334, "step": 82312} +{"lr": 0.0002140463001152259, "data_time": 0.0019357919692993164, "grad_norm": 0.01935142130823806, "loss": 0.6874332964420319, "time": 0.4041867971420288, "epoch": 15, "memory": 20334, "step": 82412} +{"lr": 0.0002140463001152259, "data_time": 0.0018247604370117188, "grad_norm": 0.0234906857018359, "loss": 0.6824127256870269, "time": 0.40228424072265623, "epoch": 15, "memory": 20334, "step": 82512} +{"lr": 0.0002140463001152259, "data_time": 0.002836275100708008, "grad_norm": 0.016757755377329885, "loss": 0.6862969815731048, "time": 0.426112699508667, "epoch": 15, "memory": 20334, "step": 82612} +{"lr": 0.0002140463001152259, "data_time": 0.0019039869308471679, "grad_norm": 0.019472584035247563, "loss": 0.6847643315792084, "time": 0.4146775484085083, "epoch": 15, "memory": 20334, "step": 82712} +{"lr": 0.0002140463001152259, "data_time": 0.0023277759552001952, "grad_norm": 0.011498039378784597, "loss": 0.6896952092647552, "time": 0.4664032220840454, "epoch": 15, "memory": 20334, "step": 82812} +{"lr": 0.0002140463001152259, "data_time": 0.001938939094543457, "grad_norm": 0.018599987821653487, "loss": 0.686225414276123, "time": 0.3851919412612915, "epoch": 15, "memory": 20334, "step": 82912} +{"lr": 0.0002140463001152259, "data_time": 0.002094864845275879, "grad_norm": 0.021390922949649394, "loss": 0.6847345352172851, "time": 0.4318429708480835, "epoch": 15, "memory": 20334, "step": 83012} +{"lr": 0.0002140463001152259, "data_time": 0.0019536495208740236, "grad_norm": 0.015049825771711767, "loss": 0.6843174219131469, "time": 0.38935074806213377, "epoch": 15, "memory": 20334, "step": 83112} +{"lr": 0.0002140463001152259, "data_time": 0.0023333072662353516, "grad_norm": 0.011506680818274617, "loss": 0.6823466122150421, "time": 0.4033582925796509, "epoch": 15, "memory": 20334, "step": 83212} +{"lr": 0.0002140463001152259, "data_time": 0.0017803192138671875, "grad_norm": 0.02404228476807475, "loss": 0.681760984659195, "time": 0.39705166816711424, "epoch": 15, "memory": 20334, "step": 83312} +{"lr": 0.0002140463001152259, "data_time": 0.002094674110412598, "grad_norm": 0.010580299817956985, "loss": 0.6865424275398254, "time": 0.3992241144180298, "epoch": 15, "memory": 20334, "step": 83412} +{"lr": 0.0002140463001152259, "data_time": 0.0019010305404663086, "grad_norm": 0.010495419334620237, "loss": 0.6883788228034973, "time": 0.3959256410598755, "epoch": 15, "memory": 20334, "step": 83512} +{"lr": 0.0002140463001152259, "data_time": 0.0016830921173095702, "grad_norm": 0.018749719858169554, "loss": 0.683070284128189, "time": 0.429161810874939, "epoch": 15, "memory": 20334, "step": 83612} +{"lr": 0.0002140463001152259, "data_time": 0.0017444133758544923, "grad_norm": 0.019595989398658274, "loss": 0.6889229655265808, "time": 0.41781799793243407, "epoch": 15, "memory": 20334, "step": 83712} +{"lr": 0.0002140463001152259, "data_time": 0.001847219467163086, "grad_norm": 0.022146631916984915, "loss": 0.6834630370140076, "time": 0.399010157585144, "epoch": 15, "memory": 20334, "step": 83812} +{"lr": 0.0002140463001152259, "data_time": 0.0016989946365356446, "grad_norm": 8.715001306682826, "loss": 0.682533997297287, "time": 0.46860029697418215, "epoch": 15, "memory": 20334, "step": 83912} +{"lr": 0.0002140463001152259, "data_time": 0.0016447067260742187, "grad_norm": 0.01404360574670136, "loss": 0.6851846218109131, "time": 0.40465679168701174, "epoch": 15, "memory": 20334, "step": 84012} +{"lr": 0.0002140463001152259, "data_time": 0.0017577648162841798, "grad_norm": 0.02531041847541928, "loss": 0.6839929401874543, "time": 0.43806498050689696, "epoch": 15, "memory": 20334, "step": 84112} +{"lr": 0.0002140463001152259, "data_time": 0.0017374277114868164, "grad_norm": 0.028425779100507496, "loss": 0.6847677826881409, "time": 0.3850772619247437, "epoch": 15, "memory": 20334, "step": 84212} +{"lr": 0.0002140463001152259, "data_time": 0.0017788410186767578, "grad_norm": 0.02780276567209512, "loss": 0.6871353566646576, "time": 0.4300373554229736, "epoch": 15, "memory": 20334, "step": 84312} +{"lr": 0.0002140463001152259, "data_time": 0.002084779739379883, "grad_norm": 0.04376483880914748, "loss": 0.6851118683815003, "time": 0.41359691619873046, "epoch": 15, "memory": 20334, "step": 84412} +{"lr": 0.0002140463001152259, "data_time": 0.0018017292022705078, "grad_norm": 0.02930031605064869, "loss": 0.6853167831897735, "time": 0.39764833450317383, "epoch": 15, "memory": 20334, "step": 84512} +{"lr": 0.0002140463001152259, "data_time": 0.0030493974685668946, "grad_norm": 0.05422095921821892, "loss": 0.6835175812244415, "time": 0.4271942377090454, "epoch": 15, "memory": 20334, "step": 84612} +{"lr": 0.0002140463001152259, "data_time": 0.0025160551071166993, "grad_norm": 0.19785308777354657, "loss": 0.6901568472385406, "time": 0.41784348487854006, "epoch": 15, "memory": 20334, "step": 84712} +{"lr": 0.0002140463001152259, "data_time": 0.0020036220550537108, "grad_norm": 0.021754285646602513, "loss": 0.6847085654735565, "time": 0.42932806015014646, "epoch": 15, "memory": 20334, "step": 84812} +{"lr": 0.0002140463001152259, "data_time": 0.0018838882446289063, "grad_norm": 0.13789198875892908, "loss": 0.6823304831981659, "time": 0.3864087343215942, "epoch": 15, "memory": 20334, "step": 84912} +{"lr": 0.0002140463001152259, "data_time": 0.0019930124282836912, "grad_norm": 0.029603640688583253, "loss": 0.6802552163600921, "time": 0.39294002056121824, "epoch": 15, "memory": 20334, "step": 85012} +{"lr": 0.0002140463001152259, "data_time": 0.0018412351608276367, "grad_norm": 0.12799951825290917, "loss": 0.683345890045166, "time": 0.4119910717010498, "epoch": 15, "memory": 20334, "step": 85112} +{"lr": 0.0002140463001152259, "data_time": 0.0019691944122314452, "grad_norm": 0.08311668909154832, "loss": 0.6847773671150208, "time": 0.42104787826538087, "epoch": 15, "memory": 20334, "step": 85212} +{"lr": 0.0002140463001152259, "data_time": 0.0014868497848510743, "grad_norm": 0.028482846869155764, "loss": 0.6900527358055115, "time": 0.40334813594818114, "epoch": 15, "memory": 20334, "step": 85312} +{"lr": 0.0002140463001152259, "data_time": 0.0015172481536865235, "grad_norm": 0.02009584875777364, "loss": 0.6823492228984833, "time": 0.4446099042892456, "epoch": 15, "memory": 20334, "step": 85412} +{"lr": 0.0002140463001152259, "data_time": 0.0018595218658447265, "grad_norm": 0.015387416537851095, "loss": 0.6880171895027161, "time": 0.40314862728118894, "epoch": 15, "memory": 20334, "step": 85512} +{"lr": 0.0002140463001152259, "data_time": 0.0016479253768920898, "grad_norm": 8.695492133498192, "loss": 0.683335691690445, "time": 0.4031576156616211, "epoch": 15, "memory": 20334, "step": 85612} +{"lr": 0.0002140463001152259, "data_time": 0.0019529342651367187, "grad_norm": 0.06487121218815446, "loss": 0.685955011844635, "time": 0.3881665229797363, "epoch": 15, "memory": 20334, "step": 85712} +{"lr": 0.0002140463001152259, "data_time": 0.0023677587509155274, "grad_norm": 0.08772307783365249, "loss": 0.6858329474925995, "time": 0.3961667537689209, "epoch": 15, "memory": 20334, "step": 85812} +{"lr": 0.0002140463001152259, "data_time": 0.0021058082580566405, "grad_norm": 3.8150058895349503, "loss": 0.673198688030243, "time": 0.4034181833267212, "epoch": 15, "memory": 20334, "step": 85912} +{"lr": 0.0002140463001152259, "data_time": 0.002176332473754883, "grad_norm": 2.4243196934461593, "loss": 0.6876299798488616, "time": 0.37183878421783445, "epoch": 15, "memory": 20334, "step": 86012} +{"lr": 0.0002140463001152259, "data_time": 0.0022976160049438476, "grad_norm": 1.663897656649351, "loss": 0.6721494734287262, "time": 0.46376795768737794, "epoch": 15, "memory": 20334, "step": 86112} +{"lr": 0.0002140463001152259, "data_time": 0.002008962631225586, "grad_norm": 46.67166759073734, "loss": 0.6610705494880676, "time": 0.4014244079589844, "epoch": 15, "memory": 20334, "step": 86212} +{"lr": 0.0002140463001152259, "data_time": 0.002184939384460449, "grad_norm": 2.503987103700638, "loss": 0.6638953685760498, "time": 0.4040172815322876, "epoch": 15, "memory": 20334, "step": 86312} +{"accuracy/top1": 93.89558410644531, "data_time": 0.37315506405300564, "time": 0.9995271894666884, "step": 15} +{"lr": 0.00015498214331265904, "data_time": 0.0016907691955566407, "grad_norm": 12.951112318038941, "loss": 0.6619929432868957, "time": 0.4124996423721313, "epoch": 16, "memory": 20334, "step": 86470} +{"lr": 0.00015498214331265904, "data_time": 0.0018224239349365235, "grad_norm": 20.398714852333068, "loss": 0.6426031589508057, "time": 0.44016361236572266, "epoch": 16, "memory": 20334, "step": 86570} +{"lr": 0.00015498214331265904, "data_time": 0.0025153160095214844, "grad_norm": 166.12853293418885, "loss": 0.6845923960208893, "time": 0.3781388282775879, "epoch": 16, "memory": 20334, "step": 86670} +{"lr": 0.00015498214331265904, "data_time": 0.0031164407730102537, "grad_norm": 17.45560750961304, "loss": 0.6753440737724304, "time": 0.4268238067626953, "epoch": 16, "memory": 20334, "step": 86770} +{"lr": 0.00015498214331265904, "data_time": 0.0017195940017700195, "grad_norm": 153.9520194530487, "loss": 0.6710609078407288, "time": 0.43384950160980223, "epoch": 16, "memory": 20334, "step": 86870} +{"lr": 0.00015498214331265904, "data_time": 0.0018800973892211914, "grad_norm": 16.20660207271576, "loss": 0.6688604831695557, "time": 0.4156578540802002, "epoch": 16, "memory": 20334, "step": 86970} +{"lr": 0.00015498214331265904, "data_time": 0.0024732351303100586, "grad_norm": 5.085641115903854, "loss": 0.696414589881897, "time": 0.4041528940200806, "epoch": 16, "memory": 20334, "step": 87070} +{"lr": 0.00015498214331265904, "data_time": 0.003353714942932129, "grad_norm": 6.340100446715951, "loss": 0.6840851962566376, "time": 0.4223287582397461, "epoch": 16, "memory": 20334, "step": 87170} +{"lr": 0.00015498214331265904, "data_time": 0.0015853643417358398, "grad_norm": 0.969308928400278, "loss": 0.6880277335643769, "time": 0.3990727663040161, "epoch": 16, "memory": 20334, "step": 87270} +{"lr": 0.00015498214331265904, "data_time": 0.0019559621810913085, "grad_norm": 1.2147032178938388, "loss": 0.6861448228359223, "time": 0.40780184268951414, "epoch": 16, "memory": 20334, "step": 87370} +{"lr": 0.00015498214331265904, "data_time": 0.0017110347747802735, "grad_norm": 0.833619886636734, "loss": 0.6849690914154053, "time": 0.4397338628768921, "epoch": 16, "memory": 20334, "step": 87470} +{"lr": 0.00015498214331265904, "data_time": 0.0019336938858032227, "grad_norm": 1.7692050963640213, "loss": 0.6862511754035949, "time": 0.44584016799926757, "epoch": 16, "memory": 20334, "step": 87570} +{"lr": 0.00015498214331265904, "data_time": 0.002180600166320801, "grad_norm": 1.9187275230884553, "loss": 0.6827983617782593, "time": 0.41813228130340574, "epoch": 16, "memory": 20334, "step": 87670} +{"lr": 0.00015498214331265904, "data_time": 0.0031360387802124023, "grad_norm": 10.539541080594063, "loss": 0.6742065012454986, "time": 0.3825331926345825, "epoch": 16, "memory": 20334, "step": 87770} +{"lr": 0.00015498214331265904, "data_time": 0.0026463747024536135, "grad_norm": 0.7334457784891129, "loss": 0.6746126055717468, "time": 0.44365291595458983, "epoch": 16, "memory": 20334, "step": 87870} +{"lr": 0.00015498214331265904, "data_time": 0.0021526098251342775, "grad_norm": 0.8798057943582535, "loss": 0.6776270508766175, "time": 0.41785004138946535, "epoch": 16, "memory": 20334, "step": 87970} +{"lr": 0.00015498214331265904, "data_time": 0.003639531135559082, "grad_norm": 3.3916256219148635, "loss": 0.6738600790500641, "time": 0.4211766481399536, "epoch": 16, "memory": 20334, "step": 88070} +{"lr": 0.00015498214331265904, "data_time": 0.0019486665725708008, "grad_norm": 0.2992755934596062, "loss": 0.685307115316391, "time": 0.38522896766662595, "epoch": 16, "memory": 20334, "step": 88170} +{"lr": 0.00015498214331265904, "data_time": 0.0019204378128051757, "grad_norm": 0.398986155167222, "loss": 0.6886320233345031, "time": 0.4050882816314697, "epoch": 16, "memory": 20334, "step": 88270} +{"lr": 0.00015498214331265904, "data_time": 0.0028360843658447265, "grad_norm": 0.17018628530204297, "loss": 0.6884448051452636, "time": 0.4105250597000122, "epoch": 16, "memory": 20334, "step": 88370} +{"lr": 0.00015498214331265904, "data_time": 0.002909278869628906, "grad_norm": 0.462920555472374, "loss": 0.6926625609397888, "time": 0.4037681818008423, "epoch": 16, "memory": 20334, "step": 88470} +{"lr": 0.00015498214331265904, "data_time": 0.00221707820892334, "grad_norm": 17.873842865228653, "loss": 0.6875955760478973, "time": 0.44903762340545655, "epoch": 16, "memory": 20334, "step": 88570} +{"lr": 0.00015498214331265904, "data_time": 0.002274298667907715, "grad_norm": 11.725898110866547, "loss": 0.6912346422672272, "time": 0.41946699619293215, "epoch": 16, "memory": 20334, "step": 88670} +{"lr": 0.00015498214331265904, "data_time": 0.0017960548400878906, "grad_norm": 1.5581879802048206, "loss": 0.6845430731773376, "time": 0.4291127443313599, "epoch": 16, "memory": 20334, "step": 88770} +{"lr": 0.00015498214331265904, "data_time": 0.0027998685836791992, "grad_norm": 4.23603507168591, "loss": 0.6869994580745697, "time": 0.39378364086151124, "epoch": 16, "memory": 20334, "step": 88870} +{"lr": 0.00015498214331265904, "data_time": 0.0017722845077514648, "grad_norm": 0.727909404411912, "loss": 0.6893838763236999, "time": 0.412534499168396, "epoch": 16, "memory": 20334, "step": 88970} +{"lr": 0.00015498214331265904, "data_time": 0.0018594741821289062, "grad_norm": 0.42019427847117186, "loss": 0.6843425452709198, "time": 0.40675690174102785, "epoch": 16, "memory": 20334, "step": 89070} +{"lr": 0.00015498214331265904, "data_time": 0.0017817497253417968, "grad_norm": 0.13882692344486713, "loss": 0.6857464551925659, "time": 0.4022836685180664, "epoch": 16, "memory": 20334, "step": 89170} +{"lr": 0.00015498214331265904, "data_time": 0.001861262321472168, "grad_norm": 0.04924918580800295, "loss": 0.6896380245685577, "time": 0.405580997467041, "epoch": 16, "memory": 20334, "step": 89270} +{"lr": 0.00015498214331265904, "data_time": 0.0018180131912231446, "grad_norm": 2.302377102896571, "loss": 0.6833753228187561, "time": 0.3844506025314331, "epoch": 16, "memory": 20334, "step": 89370} +{"lr": 0.00015498214331265904, "data_time": 0.0020698070526123046, "grad_norm": 0.8118695575743914, "loss": 0.6921520233154297, "time": 0.3881402492523193, "epoch": 16, "memory": 20334, "step": 89470} +{"lr": 0.00015498214331265904, "data_time": 0.0017733573913574219, "grad_norm": 0.24332242012023925, "loss": 0.6881895542144776, "time": 0.43993773460388186, "epoch": 16, "memory": 20334, "step": 89570} +{"lr": 0.00015498214331265904, "data_time": 0.0017596006393432618, "grad_norm": 0.3353728521615267, "loss": 0.6832014620304108, "time": 0.41824162006378174, "epoch": 16, "memory": 20334, "step": 89670} +{"lr": 0.00015498214331265904, "data_time": 0.0016766786575317383, "grad_norm": 0.251370819658041, "loss": 0.686100322008133, "time": 0.42429065704345703, "epoch": 16, "memory": 20334, "step": 89770} +{"lr": 0.00015498214331265904, "data_time": 0.001540803909301758, "grad_norm": 0.6775854665786027, "loss": 0.6858800172805786, "time": 0.4270820140838623, "epoch": 16, "memory": 20334, "step": 89870} +{"lr": 0.00015498214331265904, "data_time": 0.0016483306884765626, "grad_norm": 0.6123087488114833, "loss": 0.6825789332389831, "time": 0.39502439498901365, "epoch": 16, "memory": 20334, "step": 89970} +{"lr": 0.00015498214331265904, "data_time": 0.00198209285736084, "grad_norm": 0.20175361707806588, "loss": 0.6863687098026275, "time": 0.46511433124542234, "epoch": 16, "memory": 20334, "step": 90070} +{"lr": 0.00015498214331265904, "data_time": 0.0018755674362182617, "grad_norm": 0.2863535758107901, "loss": 0.687699693441391, "time": 0.4243708848953247, "epoch": 16, "memory": 20334, "step": 90170} +{"lr": 0.00015498214331265904, "data_time": 0.001748061180114746, "grad_norm": 0.1539195215329528, "loss": 0.6884487390518188, "time": 0.4154834747314453, "epoch": 16, "memory": 20334, "step": 90270} +{"lr": 0.00015498214331265904, "data_time": 0.001672840118408203, "grad_norm": 0.19501777943223714, "loss": 0.6865182340145111, "time": 0.43398594856262207, "epoch": 16, "memory": 20334, "step": 90370} +{"lr": 0.00015498214331265904, "data_time": 0.0016072988510131836, "grad_norm": 0.5412828426808118, "loss": 0.6927432000637055, "time": 0.42673544883728026, "epoch": 16, "memory": 20334, "step": 90470} +{"lr": 0.00015498214331265904, "data_time": 0.0017229318618774414, "grad_norm": 0.7294685713946819, "loss": 0.688781064748764, "time": 0.39732863903045657, "epoch": 16, "memory": 20334, "step": 90570} +{"lr": 0.00015498214331265904, "data_time": 0.001889514923095703, "grad_norm": 1.2996361970901489, "loss": 0.6843153655529022, "time": 0.38910818099975586, "epoch": 16, "memory": 20334, "step": 90670} +{"lr": 0.00015498214331265904, "data_time": 0.0015711784362792969, "grad_norm": 0.5796617537736892, "loss": 0.6846848845481872, "time": 0.37526464462280273, "epoch": 16, "memory": 20334, "step": 90770} +{"lr": 0.00015498214331265904, "data_time": 0.0015628814697265625, "grad_norm": 4.027517405152321, "loss": 0.6886925756931305, "time": 0.371440052986145, "epoch": 16, "memory": 20334, "step": 90870} +{"lr": 0.00015498214331265904, "data_time": 0.0015830516815185547, "grad_norm": 2.0649502158164976, "loss": 0.6806890785694122, "time": 0.4043705701828003, "epoch": 16, "memory": 20334, "step": 90970} +{"lr": 0.00015498214331265904, "data_time": 0.0015935182571411132, "grad_norm": 2.3120406031608582, "loss": 0.6872101962566376, "time": 0.39736437797546387, "epoch": 16, "memory": 20334, "step": 91070} +{"lr": 0.00015498214331265904, "data_time": 0.001550149917602539, "grad_norm": 6.763669490814209, "loss": 0.6779651641845703, "time": 0.4133500099182129, "epoch": 16, "memory": 20334, "step": 91170} +{"lr": 0.00015498214331265904, "data_time": 0.0016917705535888672, "grad_norm": 16.962061190605162, "loss": 0.6605131447315216, "time": 0.4373743534088135, "epoch": 16, "memory": 20334, "step": 91270} +{"lr": 0.00015498214331265904, "data_time": 0.0015970230102539062, "grad_norm": 7.3323180198669435, "loss": 0.6797120928764343, "time": 0.3811880350112915, "epoch": 16, "memory": 20334, "step": 91370} +{"lr": 0.00015498214331265904, "data_time": 0.001712632179260254, "grad_norm": 18.087129247188567, "loss": 0.6597596287727356, "time": 0.3891608238220215, "epoch": 16, "memory": 20334, "step": 91470} +{"lr": 0.00015498214331265904, "data_time": 0.0015247821807861327, "grad_norm": 422.3002927303314, "loss": 0.6642901420593261, "time": 0.40826027393341063, "epoch": 16, "memory": 20334, "step": 91570} +{"lr": 0.00015498214331265904, "data_time": 0.0017327308654785157, "grad_norm": 293.7801417350769, "loss": 0.6675473213195801, "time": 0.38245701789855957, "epoch": 16, "memory": 20334, "step": 91670} +{"lr": 0.00015498214331265904, "data_time": 0.0016161441802978516, "grad_norm": 63.248290061950684, "loss": 0.6756110072135926, "time": 0.38981170654296876, "epoch": 16, "memory": 20334, "step": 91770} +{"lr": 0.00015498214331265904, "data_time": 0.0018299579620361327, "grad_norm": 40.153921604156494, "loss": 0.6774940490722656, "time": 0.41505343914031984, "epoch": 16, "memory": 20334, "step": 91870} +{"lr": 0.00015498214331265904, "data_time": 0.0020282506942749024, "grad_norm": 77.30255718231201, "loss": 0.6731551468372345, "time": 0.3951940298080444, "epoch": 16, "memory": 20334, "step": 91970} +{"lr": 0.00015498214331265904, "data_time": 0.0017313957214355469, "grad_norm": 126.43501660823821, "loss": 0.670775580406189, "time": 0.3873365163803101, "epoch": 16, "memory": 20334, "step": 92070} +{"accuracy/top1": 98.09986114501953, "data_time": 0.3739733960893419, "time": 1.0005699263678656, "step": 16} +{"lr": 0.00010453658778440108, "data_time": 0.0016625165939331056, "grad_norm": 3.2546398401260377, "loss": 0.6641832888126373, "time": 0.41259312629699707, "epoch": 17, "memory": 20334, "step": 92228} +{"lr": 0.00010453658778440108, "data_time": 0.0016060352325439453, "grad_norm": 9.871051055192947, "loss": 0.6584961235523223, "time": 0.41314380168914794, "epoch": 17, "memory": 20334, "step": 92328} +{"lr": 0.00010453658778440108, "data_time": 0.0015558719635009766, "grad_norm": 16.2878378868103, "loss": 0.6527981042861939, "time": 0.4243000507354736, "epoch": 17, "memory": 20334, "step": 92428} +{"lr": 0.00010453658778440108, "data_time": 0.0016339778900146484, "grad_norm": 15.20300098657608, "loss": 0.6578659296035767, "time": 0.4310301780700684, "epoch": 17, "memory": 20334, "step": 92528} +{"lr": 0.00010453658778440108, "data_time": 0.0015978336334228516, "grad_norm": 11.455358362197876, "loss": 0.6541256964206695, "time": 0.42904951572418215, "epoch": 17, "memory": 20334, "step": 92628} +{"lr": 0.00010453658778440108, "data_time": 0.0016177892684936523, "grad_norm": 12.467400646209716, "loss": 0.6616435170173645, "time": 0.414095139503479, "epoch": 17, "memory": 20334, "step": 92728} +{"lr": 0.00010453658778440108, "data_time": 0.001538991928100586, "grad_norm": 15.241450905799866, "loss": 0.6654034435749054, "time": 0.40929033756256106, "epoch": 17, "memory": 20334, "step": 92828} +{"lr": 0.00010453658778440108, "data_time": 0.0016882658004760743, "grad_norm": 6.928589969873428, "loss": 0.6621437847614289, "time": 0.4103843688964844, "epoch": 17, "memory": 20334, "step": 92928} +{"lr": 0.00010453658778440108, "data_time": 0.0016207218170166016, "grad_norm": 11.785401248931885, "loss": 0.6657221257686615, "time": 0.4183415174484253, "epoch": 17, "memory": 20334, "step": 93028} +{"lr": 0.00010453658778440108, "data_time": 0.0018795251846313477, "grad_norm": 19.80033379793167, "loss": 0.6555939137935638, "time": 0.45161216259002684, "epoch": 17, "memory": 20334, "step": 93128} +{"lr": 0.00010453658778440108, "data_time": 0.0018337249755859375, "grad_norm": 13.87143907546997, "loss": 0.647191196680069, "time": 0.3781932592391968, "epoch": 17, "memory": 20334, "step": 93228} +{"lr": 0.00010453658778440108, "data_time": 0.003337192535400391, "grad_norm": 22.115187096595765, "loss": 0.6549151837825775, "time": 0.4360326290130615, "epoch": 17, "memory": 20334, "step": 93328} +{"lr": 0.00010453658778440108, "data_time": 0.0018713235855102538, "grad_norm": 5.600091993808746, "loss": 0.6542356610298157, "time": 0.4102866888046265, "epoch": 17, "memory": 20334, "step": 93428} +{"lr": 0.00010453658778440108, "data_time": 0.002435731887817383, "grad_norm": 11.64617965221405, "loss": 0.6576518476009369, "time": 0.4596485376358032, "epoch": 17, "memory": 20334, "step": 93528} +{"lr": 0.00010453658778440108, "data_time": 0.003143620491027832, "grad_norm": 129.42072924375535, "loss": 0.6680347144603729, "time": 0.3715360164642334, "epoch": 17, "memory": 20334, "step": 93628} +{"lr": 0.00010453658778440108, "data_time": 0.0023673295974731444, "grad_norm": 32.76223933696747, "loss": 0.6641122877597809, "time": 0.38414716720581055, "epoch": 17, "memory": 20334, "step": 93728} +{"lr": 0.00010453658778440108, "data_time": 0.0017449617385864257, "grad_norm": 4.832263457775116, "loss": 0.6561089336872101, "time": 0.3951707363128662, "epoch": 17, "memory": 20334, "step": 93828} +{"lr": 0.00010453658778440108, "data_time": 0.0020903348922729492, "grad_norm": 3108.9794964671137, "loss": 0.6573094308376313, "time": 0.4078761339187622, "epoch": 17, "memory": 20334, "step": 93928} +{"lr": 0.00010453658778440108, "data_time": 0.0017760992050170898, "grad_norm": 7.687473797798157, "loss": 0.6516622245311737, "time": 0.44172675609588624, "epoch": 17, "memory": 20334, "step": 94028} +{"lr": 0.00010453658778440108, "data_time": 0.0016998052597045898, "grad_norm": 86.21956859827041, "loss": 0.6472983598709107, "time": 0.42015249729156495, "epoch": 17, "memory": 20334, "step": 94128} +{"lr": 0.00010453658778440108, "data_time": 0.002736973762512207, "grad_norm": 60.84196796417236, "loss": 0.6557434618473053, "time": 0.40561344623565676, "epoch": 17, "memory": 20334, "step": 94228} +{"lr": 0.00010453658778440108, "data_time": 0.001611948013305664, "grad_norm": 19.71068468093872, "loss": 0.6611088514328003, "time": 0.38568027019500734, "epoch": 17, "memory": 20334, "step": 94328} +{"lr": 0.00010453658778440108, "data_time": 0.002765393257141113, "grad_norm": 8.547202229499817, "loss": 0.6529266476631165, "time": 0.4219272375106812, "epoch": 17, "memory": 20334, "step": 94428} +{"lr": 0.00010453658778440108, "data_time": 0.002468371391296387, "grad_norm": 21.46789469718933, "loss": 0.666370689868927, "time": 0.4284600496292114, "epoch": 17, "memory": 20334, "step": 94528} +{"lr": 0.00010453658778440108, "data_time": 0.002414369583129883, "grad_norm": 29.784071826934813, "loss": 0.6476145446300506, "time": 0.40134098529815676, "epoch": 17, "memory": 20334, "step": 94628} +{"lr": 0.00010453658778440108, "data_time": 0.0025254964828491213, "grad_norm": 49.705214047431944, "loss": 0.648444402217865, "time": 0.40775332450866697, "epoch": 17, "memory": 20334, "step": 94728} +{"lr": 0.00010453658778440108, "data_time": 0.002287602424621582, "grad_norm": 8.643514037132263, "loss": 0.6472416996955872, "time": 0.3936166763305664, "epoch": 17, "memory": 20334, "step": 94828} +{"lr": 0.00010453658778440108, "data_time": 0.0020152807235717775, "grad_norm": 51.14836868047714, "loss": 0.6423110663890839, "time": 0.4298069000244141, "epoch": 17, "memory": 20334, "step": 94928} +{"lr": 0.00010453658778440108, "data_time": 0.0033156156539916994, "grad_norm": 19.75882931947708, "loss": 0.6387811541557312, "time": 0.3985825300216675, "epoch": 17, "memory": 20334, "step": 95028} +{"lr": 0.00010453658778440108, "data_time": 0.001702737808227539, "grad_norm": 4.2123747825622555, "loss": 0.648260486125946, "time": 0.4048029899597168, "epoch": 17, "memory": 20334, "step": 95128} +{"lr": 0.00010453658778440108, "data_time": 0.0018829345703125, "grad_norm": 3.844643288850784, "loss": 0.6351286172866821, "time": 0.372525691986084, "epoch": 17, "memory": 20334, "step": 95228} +{"lr": 0.00010453658778440108, "data_time": 0.001734161376953125, "grad_norm": 18.19198731184006, "loss": 0.6390858173370362, "time": 0.42943248748779295, "epoch": 17, "memory": 20334, "step": 95328} +{"lr": 0.00010453658778440108, "data_time": 0.0030065059661865236, "grad_norm": 9.158291417360306, "loss": 0.6450623452663422, "time": 0.4055814504623413, "epoch": 17, "memory": 20334, "step": 95428} +{"lr": 0.00010453658778440108, "data_time": 0.0024463891983032226, "grad_norm": 41.29972470998764, "loss": 0.6425542116165162, "time": 0.3878720998764038, "epoch": 17, "memory": 20334, "step": 95528} +{"lr": 0.00010453658778440108, "data_time": 0.0018125534057617187, "grad_norm": 37.74541869163513, "loss": 0.6478318452835083, "time": 0.3871973276138306, "epoch": 17, "memory": 20334, "step": 95628} +{"lr": 0.00010453658778440108, "data_time": 0.002763652801513672, "grad_norm": 18.169064903259276, "loss": 0.649678373336792, "time": 0.3866436004638672, "epoch": 17, "memory": 20334, "step": 95728} +{"lr": 0.00010453658778440108, "data_time": 0.0017476081848144531, "grad_norm": 63.33078880310059, "loss": 0.656665587425232, "time": 0.4066095113754272, "epoch": 17, "memory": 20334, "step": 95828} +{"lr": 0.00010453658778440108, "data_time": 0.001861405372619629, "grad_norm": 37.726093196868895, "loss": 0.6425377547740936, "time": 0.37084503173828126, "epoch": 17, "memory": 20334, "step": 95928} +{"lr": 0.00010453658778440108, "data_time": 0.002059483528137207, "grad_norm": 221.2627161026001, "loss": 0.6478916466236114, "time": 0.39398856163024903, "epoch": 17, "memory": 20334, "step": 96028} +{"lr": 0.00010453658778440108, "data_time": 0.002388763427734375, "grad_norm": 16.53392014503479, "loss": 0.650449150800705, "time": 0.3876559495925903, "epoch": 17, "memory": 20334, "step": 96128} +{"lr": 0.00010453658778440108, "data_time": 0.0021607398986816405, "grad_norm": 46.52003974914551, "loss": 0.6469671666622162, "time": 0.37990684509277345, "epoch": 17, "memory": 20334, "step": 96228} +{"lr": 0.00010453658778440108, "data_time": 0.0019446134567260743, "grad_norm": 45.20160779953003, "loss": 0.6488111853599549, "time": 0.41321752071380613, "epoch": 17, "memory": 20334, "step": 96328} +{"lr": 0.00010453658778440108, "data_time": 0.0016960620880126954, "grad_norm": 82.16019706726074, "loss": 0.6456238508224488, "time": 0.4006854772567749, "epoch": 17, "memory": 20334, "step": 96428} +{"lr": 0.00010453658778440108, "data_time": 0.001622319221496582, "grad_norm": 13.535050964355468, "loss": 0.6291361391544342, "time": 0.38329644203186036, "epoch": 17, "memory": 20334, "step": 96528} +{"lr": 0.00010453658778440108, "data_time": 0.0021713733673095702, "grad_norm": 8.185927581787109, "loss": 0.6445947825908661, "time": 0.38475635051727297, "epoch": 17, "memory": 20334, "step": 96628} +{"lr": 0.00010453658778440108, "data_time": 0.0016494512557983399, "grad_norm": 12.52618865966797, "loss": 0.6456365406513214, "time": 0.3875510931015015, "epoch": 17, "memory": 20334, "step": 96728} +{"lr": 0.00010453658778440108, "data_time": 0.0023557662963867186, "grad_norm": 54.97046413421631, "loss": 0.6526306748390198, "time": 0.44949591159820557, "epoch": 17, "memory": 20334, "step": 96828} +{"lr": 0.00010453658778440108, "data_time": 0.0025568485260009767, "grad_norm": 27.850926280021667, "loss": 0.6469048321247101, "time": 0.40942893028259275, "epoch": 17, "memory": 20334, "step": 96928} +{"lr": 0.00010453658778440108, "data_time": 0.002293086051940918, "grad_norm": 15.37418692111969, "loss": 0.6597590148448944, "time": 0.40243828296661377, "epoch": 17, "memory": 20334, "step": 97028} +{"lr": 0.00010453658778440108, "data_time": 0.0019539594650268555, "grad_norm": 23.74850389957428, "loss": 0.6477436661720276, "time": 0.4275946140289307, "epoch": 17, "memory": 20334, "step": 97128} +{"lr": 0.00010453658778440108, "data_time": 0.0021381616592407227, "grad_norm": 8.819652533531189, "loss": 0.6474440455436706, "time": 0.38517258167266843, "epoch": 17, "memory": 20334, "step": 97228} +{"lr": 0.00010453658778440108, "data_time": 0.0017972946166992187, "grad_norm": 37.99319822788239, "loss": 0.6425372898578644, "time": 0.4023005962371826, "epoch": 17, "memory": 20334, "step": 97328} +{"lr": 0.00010453658778440108, "data_time": 0.0022548437118530273, "grad_norm": 34.167623949050906, "loss": 0.6537059545516968, "time": 0.39999263286590575, "epoch": 17, "memory": 20334, "step": 97428} +{"lr": 0.00010453658778440108, "data_time": 0.002572035789489746, "grad_norm": 48.18318336009979, "loss": 0.653168547153473, "time": 0.4285785675048828, "epoch": 17, "memory": 20334, "step": 97528} +{"lr": 0.00010453658778440108, "data_time": 0.0016847372055053711, "grad_norm": 16.86105582714081, "loss": 0.6359785616397857, "time": 0.3951293706893921, "epoch": 17, "memory": 20334, "step": 97628} +{"lr": 0.00010453658778440108, "data_time": 0.002587628364562988, "grad_norm": 43.33133211135864, "loss": 0.6361349105834961, "time": 0.418013596534729, "epoch": 17, "memory": 20334, "step": 97728} +{"lr": 0.00010453658778440108, "data_time": 0.0019317388534545899, "grad_norm": 32.4199652671814, "loss": 0.642455643415451, "time": 0.40132341384887693, "epoch": 17, "memory": 20334, "step": 97828} +{"accuracy/top1": 88.11312103271484, "data_time": 0.3581188784705268, "time": 0.9833702511257596, "step": 17} +{"lr": 6.395177052675796e-05, "data_time": 0.002099752426147461, "grad_norm": 63.28272061347961, "loss": 0.6444731771945953, "time": 0.4256767272949219, "epoch": 18, "memory": 20334, "step": 97986} +{"lr": 6.395177052675796e-05, "data_time": 0.002711629867553711, "grad_norm": 71.94749855995178, "loss": 0.6376009404659271, "time": 0.39515902996063235, "epoch": 18, "memory": 20334, "step": 98086} +{"lr": 6.395177052675796e-05, "data_time": 0.0019078969955444336, "grad_norm": 27.969100666046142, "loss": 0.6339970886707306, "time": 0.4254409074783325, "epoch": 18, "memory": 20334, "step": 98186} +{"lr": 6.395177052675796e-05, "data_time": 0.0029084444046020507, "grad_norm": 33.11596373319626, "loss": 0.6604844689369201, "time": 0.4136345624923706, "epoch": 18, "memory": 20334, "step": 98286} +{"lr": 6.395177052675796e-05, "data_time": 0.002246594429016113, "grad_norm": 21.222491908073426, "loss": 0.6445057034492493, "time": 0.37490484714508054, "epoch": 18, "memory": 20334, "step": 98386} +{"lr": 6.395177052675796e-05, "data_time": 0.0019988059997558595, "grad_norm": 75.3499174118042, "loss": 0.6528640747070312, "time": 0.4187065601348877, "epoch": 18, "memory": 20334, "step": 98486} +{"lr": 6.395177052675796e-05, "data_time": 0.0018184900283813477, "grad_norm": 131.6462007522583, "loss": 0.6250163376331329, "time": 0.4211371421813965, "epoch": 18, "memory": 20334, "step": 98586} +{"lr": 6.395177052675796e-05, "data_time": 0.0017574310302734375, "grad_norm": 71.85484886169434, "loss": 0.6417450666427612, "time": 0.37882812023162843, "epoch": 18, "memory": 20334, "step": 98686} +{"lr": 6.395177052675796e-05, "data_time": 0.0017748355865478515, "grad_norm": 14.120350122451782, "loss": 0.6532712817192078, "time": 0.3884144306182861, "epoch": 18, "memory": 20334, "step": 98786} +{"lr": 6.395177052675796e-05, "data_time": 0.002509307861328125, "grad_norm": 82.07881389856338, "loss": 0.644244772195816, "time": 0.4121937036514282, "epoch": 18, "memory": 20334, "step": 98886} +{"lr": 6.395177052675796e-05, "data_time": 0.0025970935821533203, "grad_norm": 17.074625968933105, "loss": 0.648538988828659, "time": 0.4318960905075073, "epoch": 18, "memory": 20334, "step": 98986} +{"lr": 6.395177052675796e-05, "data_time": 0.0019234180450439452, "grad_norm": 12.55678596496582, "loss": 0.6545386016368866, "time": 0.39982435703277586, "epoch": 18, "memory": 20334, "step": 99086} +{"lr": 6.395177052675796e-05, "data_time": 0.0024192333221435547, "grad_norm": 43.79109058380127, "loss": 0.6471724450588227, "time": 0.3826723337173462, "epoch": 18, "memory": 20334, "step": 99186} +{"lr": 6.395177052675796e-05, "data_time": 0.0030316591262817385, "grad_norm": 19.45570340156555, "loss": 0.6481290102005005, "time": 0.3795777320861816, "epoch": 18, "memory": 20334, "step": 99286} +{"lr": 6.395177052675796e-05, "data_time": 0.0015474319458007812, "grad_norm": 19.831133008003235, "loss": 0.6461710214614869, "time": 0.4128087520599365, "epoch": 18, "memory": 20334, "step": 99386} +{"lr": 6.395177052675796e-05, "data_time": 0.0017799615859985351, "grad_norm": 30.964549469947816, "loss": 0.6446856379508972, "time": 0.4162802219390869, "epoch": 18, "memory": 20334, "step": 99486} +{"lr": 6.395177052675796e-05, "data_time": 0.0023103952407836914, "grad_norm": 20.136412382125854, "loss": 0.6416927456855774, "time": 0.45120725631713865, "epoch": 18, "memory": 20334, "step": 99586} +{"lr": 6.395177052675796e-05, "data_time": 0.0024947404861450197, "grad_norm": 32.40307134389877, "loss": 0.644729620218277, "time": 0.3843658924102783, "epoch": 18, "memory": 20334, "step": 99686} +{"lr": 6.395177052675796e-05, "data_time": 0.0022699832916259766, "grad_norm": 14.265842127799989, "loss": 0.6528850615024566, "time": 0.3970250368118286, "epoch": 18, "memory": 20334, "step": 99786} +{"lr": 6.395177052675796e-05, "data_time": 0.0020752191543579102, "grad_norm": 20.699154913425446, "loss": 0.6426553785800934, "time": 0.3929337501525879, "epoch": 18, "memory": 20334, "step": 99886} +{"lr": 6.395177052675796e-05, "data_time": 0.0026435613632202148, "grad_norm": 70.5440259218216, "loss": 0.6593203127384186, "time": 0.415566611289978, "epoch": 18, "memory": 20334, "step": 99986} +{"lr": 6.395177052675796e-05, "data_time": 0.002365422248840332, "grad_norm": 21.333938694000246, "loss": 0.6492511808872223, "time": 0.3907930374145508, "epoch": 18, "memory": 20334, "step": 100086} +{"lr": 6.395177052675796e-05, "data_time": 0.002691984176635742, "grad_norm": 27.271159529685974, "loss": 0.6560031712055207, "time": 0.3996977090835571, "epoch": 18, "memory": 20334, "step": 100186} +{"lr": 6.395177052675796e-05, "data_time": 0.0022272109985351563, "grad_norm": 25.059144473075868, "loss": 0.6597813546657563, "time": 0.3754384756088257, "epoch": 18, "memory": 20334, "step": 100286} +{"lr": 6.395177052675796e-05, "data_time": 0.001793241500854492, "grad_norm": 26.656696462631224, "loss": 0.6416965544223785, "time": 0.3689321994781494, "epoch": 18, "memory": 20334, "step": 100386} +{"lr": 6.395177052675796e-05, "data_time": 0.0029100418090820313, "grad_norm": 44.697294569015504, "loss": 0.6489736020565033, "time": 0.3945737361907959, "epoch": 18, "memory": 20334, "step": 100486} +{"lr": 6.395177052675796e-05, "data_time": 0.0018138647079467773, "grad_norm": 225.24888908863068, "loss": 0.6620129823684693, "time": 0.39425723552703856, "epoch": 18, "memory": 20334, "step": 100586} +{"lr": 6.395177052675796e-05, "data_time": 0.001793074607849121, "grad_norm": 83.48510959148408, "loss": 0.6606607139110565, "time": 0.3993523359298706, "epoch": 18, "memory": 20334, "step": 100686} +{"lr": 6.395177052675796e-05, "data_time": 0.0022084951400756837, "grad_norm": 44.040564727783206, "loss": 0.6465081810951233, "time": 0.3778827667236328, "epoch": 18, "memory": 20334, "step": 100786} +{"lr": 6.395177052675796e-05, "data_time": 0.003248929977416992, "grad_norm": 24.54559905529022, "loss": 0.6401390194892883, "time": 0.4228460073471069, "epoch": 18, "memory": 20334, "step": 100886} +{"lr": 6.395177052675796e-05, "data_time": 0.0018243074417114257, "grad_norm": 128.91987414360045, "loss": 0.6633709013462067, "time": 0.3932612657546997, "epoch": 18, "memory": 20334, "step": 100986} +{"lr": 6.395177052675796e-05, "data_time": 0.002174234390258789, "grad_norm": 152.1927219390869, "loss": 0.6574654817581177, "time": 0.40231847763061523, "epoch": 18, "memory": 20334, "step": 101086} +{"lr": 6.395177052675796e-05, "data_time": 0.0017493724822998046, "grad_norm": 34.541743755340576, "loss": 0.6583094596862793, "time": 0.3871058702468872, "epoch": 18, "memory": 20334, "step": 101186} +{"lr": 6.395177052675796e-05, "data_time": 0.0021985054016113283, "grad_norm": 10.845146322250367, "loss": 0.6594466626644134, "time": 0.3989195108413696, "epoch": 18, "memory": 20334, "step": 101286} +{"lr": 6.395177052675796e-05, "data_time": 0.002009272575378418, "grad_norm": 136.56633160114288, "loss": 0.6515347182750701, "time": 0.4141709327697754, "epoch": 18, "memory": 20334, "step": 101386} +{"lr": 6.395177052675796e-05, "data_time": 0.001636648178100586, "grad_norm": 26.99376971721649, "loss": 0.6600113630294799, "time": 0.39459292888641356, "epoch": 18, "memory": 20334, "step": 101486} +{"lr": 6.395177052675796e-05, "data_time": 0.002132105827331543, "grad_norm": 32.21519119739533, "loss": 0.6731671214103698, "time": 0.39255034923553467, "epoch": 18, "memory": 20334, "step": 101586} +{"lr": 6.395177052675796e-05, "data_time": 0.0029949665069580076, "grad_norm": 63.9368953704834, "loss": 0.661777114868164, "time": 0.41479313373565674, "epoch": 18, "memory": 20334, "step": 101686} +{"lr": 6.395177052675796e-05, "data_time": 0.0028112649917602537, "grad_norm": 25.9544114112854, "loss": 0.6740711569786072, "time": 0.41132607460021975, "epoch": 18, "memory": 20334, "step": 101786} +{"lr": 6.395177052675796e-05, "data_time": 0.001844930648803711, "grad_norm": 71.69706101417542, "loss": 0.6596127629280091, "time": 0.43996922969818114, "epoch": 18, "memory": 20334, "step": 101886} +{"lr": 6.395177052675796e-05, "data_time": 0.002105236053466797, "grad_norm": 46.557479810714725, "loss": 0.6667856931686401, "time": 0.39024529457092283, "epoch": 18, "memory": 20334, "step": 101986} +{"lr": 6.395177052675796e-05, "data_time": 0.002191972732543945, "grad_norm": 102.11906833648682, "loss": 0.6647959530353547, "time": 0.4282243728637695, "epoch": 18, "memory": 20334, "step": 102086} +{"lr": 6.395177052675796e-05, "data_time": 0.0016839027404785157, "grad_norm": 91.0549749135971, "loss": 0.6600245058536529, "time": 0.3853638172149658, "epoch": 18, "memory": 20334, "step": 102186} +{"lr": 6.395177052675796e-05, "data_time": 0.0019200563430786133, "grad_norm": 40.99918479919434, "loss": 0.654580557346344, "time": 0.38777430057525636, "epoch": 18, "memory": 20334, "step": 102286} +{"lr": 6.395177052675796e-05, "data_time": 0.0018175363540649414, "grad_norm": 58.056387066841125, "loss": 0.6499052882194519, "time": 0.41110901832580565, "epoch": 18, "memory": 20334, "step": 102386} +{"lr": 6.395177052675796e-05, "data_time": 0.002111387252807617, "grad_norm": 38.41582188606262, "loss": 0.6536898493766785, "time": 0.4120710611343384, "epoch": 18, "memory": 20334, "step": 102486} +{"lr": 6.395177052675796e-05, "data_time": 0.002192568778991699, "grad_norm": 47.87467775344849, "loss": 0.6521516799926758, "time": 0.4433310508728027, "epoch": 18, "memory": 20334, "step": 102586} +{"lr": 6.395177052675796e-05, "data_time": 0.002946758270263672, "grad_norm": 42.65680613517761, "loss": 0.6453225374221802, "time": 0.4109247446060181, "epoch": 18, "memory": 20334, "step": 102686} +{"lr": 6.395177052675796e-05, "data_time": 0.0020146846771240236, "grad_norm": 59.34259493350983, "loss": 0.6423458158969879, "time": 0.43424973487854, "epoch": 18, "memory": 20334, "step": 102786} +{"lr": 6.395177052675796e-05, "data_time": 0.0016765117645263672, "grad_norm": 15.614595317840577, "loss": 0.6565988421440124, "time": 0.39060814380645753, "epoch": 18, "memory": 20334, "step": 102886} +{"lr": 6.395177052675796e-05, "data_time": 0.00270843505859375, "grad_norm": 21.659555768966676, "loss": 0.6484413981437683, "time": 0.3961986780166626, "epoch": 18, "memory": 20334, "step": 102986} +{"lr": 6.395177052675796e-05, "data_time": 0.002336454391479492, "grad_norm": 296.36976046562194, "loss": 0.6487852931022644, "time": 0.380599308013916, "epoch": 18, "memory": 20334, "step": 103086} +{"lr": 6.395177052675796e-05, "data_time": 0.002427506446838379, "grad_norm": 171.42702980041503, "loss": 0.6475272834300995, "time": 0.4200688123703003, "epoch": 18, "memory": 20334, "step": 103186} +{"lr": 6.395177052675796e-05, "data_time": 0.0019548416137695314, "grad_norm": 15.770891356468201, "loss": 0.6601642727851867, "time": 0.4168720722198486, "epoch": 18, "memory": 20334, "step": 103286} +{"lr": 6.395177052675796e-05, "data_time": 0.002973628044128418, "grad_norm": 15.751076912879943, "loss": 0.6604206025600433, "time": 0.41751086711883545, "epoch": 18, "memory": 20334, "step": 103386} +{"lr": 6.395177052675796e-05, "data_time": 0.002408409118652344, "grad_norm": 16.845925784111024, "loss": 0.6536760210990906, "time": 0.40938050746917726, "epoch": 18, "memory": 20334, "step": 103486} +{"lr": 6.395177052675796e-05, "data_time": 0.0019280433654785157, "grad_norm": 27.657923913002016, "loss": 0.6490884304046631, "time": 0.40270171165466306, "epoch": 18, "memory": 20334, "step": 103586} +{"accuracy/top1": 86.98945617675781, "data_time": 0.36803950203789604, "time": 0.9957342147827148, "step": 18} +{"lr": 3.422702443389901e-05, "data_time": 0.0019544124603271484, "grad_norm": 66.65392987728119, "loss": 0.651543265581131, "time": 0.40294613838195803, "epoch": 19, "memory": 20334, "step": 103744} +{"lr": 3.422702443389901e-05, "data_time": 0.0018543004989624023, "grad_norm": 34.74554295539856, "loss": 0.6643685579299927, "time": 0.4082931041717529, "epoch": 19, "memory": 20334, "step": 103844} +{"lr": 3.422702443389901e-05, "data_time": 0.0020318031311035156, "grad_norm": 213.33866076469423, "loss": 0.6748937785625457, "time": 0.39371323585510254, "epoch": 19, "memory": 20334, "step": 103944} +{"lr": 3.422702443389901e-05, "data_time": 0.001800370216369629, "grad_norm": 74.37764189243316, "loss": 0.6605765461921692, "time": 0.41226332187652587, "epoch": 19, "memory": 20334, "step": 104044} +{"lr": 3.422702443389901e-05, "data_time": 0.0019049644470214844, "grad_norm": 89.74551241397857, "loss": 0.6422040402889252, "time": 0.44047772884368896, "epoch": 19, "memory": 20334, "step": 104144} +{"lr": 3.422702443389901e-05, "data_time": 0.0017130851745605468, "grad_norm": 40.855762004852295, "loss": 0.6535026669502259, "time": 0.4267627716064453, "epoch": 19, "memory": 20334, "step": 104244} +{"lr": 3.422702443389901e-05, "data_time": 0.0017953872680664062, "grad_norm": 15.970692420005799, "loss": 0.6586434721946717, "time": 0.3833063364028931, "epoch": 19, "memory": 20334, "step": 104344} +{"lr": 3.422702443389901e-05, "data_time": 0.0017811059951782227, "grad_norm": 18.287867546081543, "loss": 0.6528598010540009, "time": 0.41804656982421873, "epoch": 19, "memory": 20334, "step": 104444} +{"lr": 3.422702443389901e-05, "data_time": 0.0017637491226196289, "grad_norm": 14.304183197021484, "loss": 0.6278103768825531, "time": 0.3997077941894531, "epoch": 19, "memory": 20334, "step": 104544} +{"lr": 3.422702443389901e-05, "data_time": 0.0018087387084960937, "grad_norm": 9.44283344745636, "loss": 0.6413831651210785, "time": 0.39129114151000977, "epoch": 19, "memory": 20334, "step": 104644} +{"lr": 3.422702443389901e-05, "data_time": 0.0018803596496582032, "grad_norm": 40.7143635392189, "loss": 0.6593172371387481, "time": 0.39586400985717773, "epoch": 19, "memory": 20334, "step": 104744} +{"lr": 3.422702443389901e-05, "data_time": 0.0023250341415405273, "grad_norm": 13.627595686912537, "loss": 0.6552824079990387, "time": 0.3840524911880493, "epoch": 19, "memory": 20334, "step": 104844} +{"lr": 3.422702443389901e-05, "data_time": 0.002281975746154785, "grad_norm": 14.279818868637085, "loss": 0.6424823582172394, "time": 0.4203862190246582, "epoch": 19, "memory": 20334, "step": 104944} +{"lr": 3.422702443389901e-05, "data_time": 0.0019212007522583009, "grad_norm": 61.65380027294159, "loss": 0.6414035320281982, "time": 0.40967159271240233, "epoch": 19, "memory": 20334, "step": 105044} +{"lr": 3.422702443389901e-05, "data_time": 0.0016926765441894532, "grad_norm": 19.834644961357117, "loss": 0.6525056838989258, "time": 0.3920203447341919, "epoch": 19, "memory": 20334, "step": 105144} +{"lr": 3.422702443389901e-05, "data_time": 0.002080535888671875, "grad_norm": 21.97901842594147, "loss": 0.6534707963466644, "time": 0.36706583499908446, "epoch": 19, "memory": 20334, "step": 105244} +{"lr": 3.422702443389901e-05, "data_time": 0.002734875679016113, "grad_norm": 9.230246782302856, "loss": 0.639169204235077, "time": 0.4106771230697632, "epoch": 19, "memory": 20334, "step": 105344} +{"lr": 3.422702443389901e-05, "data_time": 0.0027134895324707033, "grad_norm": 25.478613805770873, "loss": 0.6500572264194489, "time": 0.41395514011383056, "epoch": 19, "memory": 20334, "step": 105444} +{"lr": 3.422702443389901e-05, "data_time": 0.0018291950225830078, "grad_norm": 31.823949766159057, "loss": 0.6490461707115174, "time": 0.4218700885772705, "epoch": 19, "memory": 20334, "step": 105544} +{"lr": 3.422702443389901e-05, "data_time": 0.0017975330352783202, "grad_norm": 42.61731996536255, "loss": 0.6465230882167816, "time": 0.3886647939682007, "epoch": 19, "memory": 20334, "step": 105644} +{"lr": 3.422702443389901e-05, "data_time": 0.002008199691772461, "grad_norm": 23.15601212978363, "loss": 0.657354474067688, "time": 0.4120891809463501, "epoch": 19, "memory": 20334, "step": 105744} +{"lr": 3.422702443389901e-05, "data_time": 0.0023223400115966798, "grad_norm": 22.897148180007935, "loss": 0.6566392123699188, "time": 0.400227427482605, "epoch": 19, "memory": 20334, "step": 105844} +{"lr": 3.422702443389901e-05, "data_time": 0.0017266035079956054, "grad_norm": 30.917552328109743, "loss": 0.655597847700119, "time": 0.38207459449768066, "epoch": 19, "memory": 20334, "step": 105944} +{"lr": 3.422702443389901e-05, "data_time": 0.0018705606460571289, "grad_norm": 41.59244341850281, "loss": 0.6586847841739655, "time": 0.36503379344940184, "epoch": 19, "memory": 20334, "step": 106044} +{"lr": 3.422702443389901e-05, "data_time": 0.0018585443496704102, "grad_norm": 33.95430791378021, "loss": 0.6456314325332642, "time": 0.46921491622924805, "epoch": 19, "memory": 20334, "step": 106144} +{"lr": 3.422702443389901e-05, "data_time": 0.0017628669738769531, "grad_norm": 40.95303258895874, "loss": 0.6533273994922638, "time": 0.39267683029174805, "epoch": 19, "memory": 20334, "step": 106244} +{"lr": 3.422702443389901e-05, "data_time": 0.0017978429794311523, "grad_norm": 14.215553760528564, "loss": 0.6592978179454804, "time": 0.4042861223220825, "epoch": 19, "memory": 20334, "step": 106344} +{"lr": 3.422702443389901e-05, "data_time": 0.0019912958145141602, "grad_norm": 143.317347240448, "loss": 0.648093169927597, "time": 0.401861572265625, "epoch": 19, "memory": 20334, "step": 106444} +{"lr": 3.422702443389901e-05, "data_time": 0.0020437002182006835, "grad_norm": 40.78470206260681, "loss": 0.6658326268196106, "time": 0.43019914627075195, "epoch": 19, "memory": 20334, "step": 106544} +{"lr": 3.422702443389901e-05, "data_time": 0.001868128776550293, "grad_norm": 19.64283421039581, "loss": 0.6576982378959656, "time": 0.41136808395385743, "epoch": 19, "memory": 20334, "step": 106644} +{"lr": 3.422702443389901e-05, "data_time": 0.0015358209609985351, "grad_norm": 75.87776384353637, "loss": 0.6442136585712432, "time": 0.40893604755401614, "epoch": 19, "memory": 20334, "step": 106744} +{"lr": 3.422702443389901e-05, "data_time": 0.0015146255493164063, "grad_norm": 12.315547037124634, "loss": 0.6521753191947937, "time": 0.4102999448776245, "epoch": 19, "memory": 20334, "step": 106844} +{"lr": 3.422702443389901e-05, "data_time": 0.0017220020294189454, "grad_norm": 14.519731771945953, "loss": 0.6572687864303589, "time": 0.4080089330673218, "epoch": 19, "memory": 20334, "step": 106944} +{"lr": 3.422702443389901e-05, "data_time": 0.002236604690551758, "grad_norm": 49.558367705345155, "loss": 0.6520628035068512, "time": 0.3836507320404053, "epoch": 19, "memory": 20334, "step": 107044} +{"lr": 3.422702443389901e-05, "data_time": 0.0017132282257080079, "grad_norm": 34.72822256088257, "loss": 0.6657291710376739, "time": 0.3867108106613159, "epoch": 19, "memory": 20334, "step": 107144} +{"lr": 3.422702443389901e-05, "data_time": 0.0016370773315429687, "grad_norm": 17.021578669548035, "loss": 0.6683961629867554, "time": 0.38309948444366454, "epoch": 19, "memory": 20334, "step": 107244} +{"lr": 3.422702443389901e-05, "data_time": 0.0016996622085571288, "grad_norm": 30.026414847373964, "loss": 0.6664615154266358, "time": 0.39449048042297363, "epoch": 19, "memory": 20334, "step": 107344} +{"lr": 3.422702443389901e-05, "data_time": 0.0016473054885864258, "grad_norm": 19.32567903995514, "loss": 0.6644317507743835, "time": 0.3868776559829712, "epoch": 19, "memory": 20334, "step": 107444} +{"lr": 3.422702443389901e-05, "data_time": 0.0016041278839111328, "grad_norm": 36.1685825586319, "loss": 0.672652804851532, "time": 0.4217507839202881, "epoch": 19, "memory": 20334, "step": 107544} +{"lr": 3.422702443389901e-05, "data_time": 0.0017470598220825195, "grad_norm": 12.698266100883483, "loss": 0.6664849102497101, "time": 0.44067671298980715, "epoch": 19, "memory": 20334, "step": 107644} +{"lr": 3.422702443389901e-05, "data_time": 0.0016474962234497071, "grad_norm": 18.58142523765564, "loss": 0.6727229356765747, "time": 0.4180203676223755, "epoch": 19, "memory": 20334, "step": 107744} +{"lr": 3.422702443389901e-05, "data_time": 0.001585698127746582, "grad_norm": 40.603049755096436, "loss": 0.6680902898311615, "time": 0.4268414735794067, "epoch": 19, "memory": 20334, "step": 107844} +{"lr": 3.422702443389901e-05, "data_time": 0.0016061782836914063, "grad_norm": 85.02847175598144, "loss": 0.6685926496982575, "time": 0.4036363124847412, "epoch": 19, "memory": 20334, "step": 107944} +{"lr": 3.422702443389901e-05, "data_time": 0.0015764951705932618, "grad_norm": 62.724023127555846, "loss": 0.6558150768280029, "time": 0.44412014484405515, "epoch": 19, "memory": 20334, "step": 108044} +{"lr": 3.422702443389901e-05, "data_time": 0.0017894506454467773, "grad_norm": 99.84025325775147, "loss": 0.6741968870162964, "time": 0.4474414587020874, "epoch": 19, "memory": 20334, "step": 108144} +{"lr": 3.422702443389901e-05, "data_time": 0.0016388177871704101, "grad_norm": 57.71546268463135, "loss": 0.6721916258335113, "time": 0.43043463230133056, "epoch": 19, "memory": 20334, "step": 108244} +{"lr": 3.422702443389901e-05, "data_time": 0.0017370462417602539, "grad_norm": 80.37630853652954, "loss": 0.6624933481216431, "time": 0.4322895765304565, "epoch": 19, "memory": 20334, "step": 108344} +{"lr": 3.422702443389901e-05, "data_time": 0.0020455837249755858, "grad_norm": 62.80916078090668, "loss": 0.6744026124477387, "time": 0.3881272077560425, "epoch": 19, "memory": 20334, "step": 108444} +{"lr": 3.422702443389901e-05, "data_time": 0.0019963979721069336, "grad_norm": 31.45765166282654, "loss": 0.6706329762935639, "time": 0.3930949687957764, "epoch": 19, "memory": 20334, "step": 108544} +{"lr": 3.422702443389901e-05, "data_time": 0.001952385902404785, "grad_norm": 56.2090705871582, "loss": 0.6598998725414276, "time": 0.41112775802612306, "epoch": 19, "memory": 20334, "step": 108644} +{"lr": 3.422702443389901e-05, "data_time": 0.001448655128479004, "grad_norm": 206.15071606636047, "loss": 0.673072224855423, "time": 0.4001233339309692, "epoch": 19, "memory": 20334, "step": 108744} +{"lr": 3.422702443389901e-05, "data_time": 0.003635668754577637, "grad_norm": 44.87873637676239, "loss": 0.6734757721424103, "time": 0.3866569995880127, "epoch": 19, "memory": 20334, "step": 108844} +{"lr": 3.422702443389901e-05, "data_time": 0.002407217025756836, "grad_norm": 79.70904369354248, "loss": 0.6719061970710755, "time": 0.4129964351654053, "epoch": 19, "memory": 20334, "step": 108944} +{"lr": 3.422702443389901e-05, "data_time": 0.0019218921661376953, "grad_norm": 58.01592597961426, "loss": 0.6579764366149903, "time": 0.400122857093811, "epoch": 19, "memory": 20334, "step": 109044} +{"lr": 3.422702443389901e-05, "data_time": 0.0021574020385742186, "grad_norm": 86.64260444641113, "loss": 0.6737904250621796, "time": 0.385427188873291, "epoch": 19, "memory": 20334, "step": 109144} +{"lr": 3.422702443389901e-05, "data_time": 0.0017372608184814454, "grad_norm": 38.41540155410767, "loss": 0.6705297827720642, "time": 0.3884838581085205, "epoch": 19, "memory": 20334, "step": 109244} +{"lr": 3.422702443389901e-05, "data_time": 0.0017900943756103515, "grad_norm": 203.17119026184082, "loss": 0.6733665406703949, "time": 0.39952883720397947, "epoch": 19, "memory": 20334, "step": 109344} +{"accuracy/top1": 85.834228515625, "data_time": 0.35817890697055393, "time": 0.9835113949245877, "step": 19} +{"lr": 1.6094271405406862e-05, "data_time": 0.0018102407455444336, "grad_norm": 37.83160743713379, "loss": 0.6657364547252655, "time": 0.3837099313735962, "epoch": 20, "memory": 20334, "step": 109502} +{"lr": 1.6094271405406862e-05, "data_time": 0.0023318052291870115, "grad_norm": 131.83887491226196, "loss": 0.6685264825820922, "time": 0.36237781047821044, "epoch": 20, "memory": 20334, "step": 109602} +{"lr": 1.6094271405406862e-05, "data_time": 0.0031822681427001952, "grad_norm": 73.49366936683654, "loss": 0.6725376188755036, "time": 0.4056878089904785, "epoch": 20, "memory": 20334, "step": 109702} +{"lr": 1.6094271405406862e-05, "data_time": 0.00291903018951416, "grad_norm": 48.69134254455567, "loss": 0.6697592377662659, "time": 0.3942695617675781, "epoch": 20, "memory": 20334, "step": 109802} +{"lr": 1.6094271405406862e-05, "data_time": 0.0024376153945922852, "grad_norm": 47.049586868286134, "loss": 0.6680775642395019, "time": 0.38937761783599856, "epoch": 20, "memory": 20334, "step": 109902} +{"lr": 1.6094271405406862e-05, "data_time": 0.0023961782455444334, "grad_norm": 1127.3899322509765, "loss": 0.673331344127655, "time": 0.38590729236602783, "epoch": 20, "memory": 20334, "step": 110002} +{"lr": 1.6094271405406862e-05, "data_time": 0.0023725748062133787, "grad_norm": 82.14511423110962, "loss": 0.6698859632015228, "time": 0.41763460636138916, "epoch": 20, "memory": 20334, "step": 110102} +{"lr": 1.6094271405406862e-05, "data_time": 0.0016741037368774414, "grad_norm": 11.355046510696411, "loss": 0.6541194677352905, "time": 0.4403804063796997, "epoch": 20, "memory": 20334, "step": 110202} +{"lr": 1.6094271405406862e-05, "data_time": 0.00211186408996582, "grad_norm": 44.58271446228027, "loss": 0.6699179112911224, "time": 0.38148767948150636, "epoch": 20, "memory": 20334, "step": 110302} +{"lr": 1.6094271405406862e-05, "data_time": 0.0019808530807495115, "grad_norm": 181.1916173696518, "loss": 0.6703812181949615, "time": 0.38389732837677004, "epoch": 20, "memory": 20334, "step": 110402} +{"lr": 1.6094271405406862e-05, "data_time": 0.0025669097900390624, "grad_norm": 29.379246282577515, "loss": 0.6694718360900879, "time": 0.40802459716796874, "epoch": 20, "memory": 20334, "step": 110502} +{"lr": 1.6094271405406862e-05, "data_time": 0.0019008636474609375, "grad_norm": 112.64701852798461, "loss": 0.6737649381160736, "time": 0.3869785308837891, "epoch": 20, "memory": 20334, "step": 110602} +{"lr": 1.6094271405406862e-05, "data_time": 0.002530479431152344, "grad_norm": 97.6847596168518, "loss": 0.6756111741065979, "time": 0.3776059865951538, "epoch": 20, "memory": 20334, "step": 110702} +{"lr": 1.6094271405406862e-05, "data_time": 0.0016866683959960937, "grad_norm": 71.38228597640992, "loss": 0.6757996320724488, "time": 0.3918351173400879, "epoch": 20, "memory": 20334, "step": 110802} +{"lr": 1.6094271405406862e-05, "data_time": 0.0030757904052734373, "grad_norm": 12.370378637313843, "loss": 0.6697602152824402, "time": 0.4175167798995972, "epoch": 20, "memory": 20334, "step": 110902} +{"lr": 1.6094271405406862e-05, "data_time": 0.0020238876342773436, "grad_norm": 217.78964290618896, "loss": 0.6711635291576385, "time": 0.40938963890075686, "epoch": 20, "memory": 20334, "step": 111002} +{"lr": 1.6094271405406862e-05, "data_time": 0.0020236730575561523, "grad_norm": 75.23270325660705, "loss": 0.6697175443172455, "time": 0.4169180393218994, "epoch": 20, "memory": 20334, "step": 111102} +{"lr": 1.6094271405406862e-05, "data_time": 0.002639508247375488, "grad_norm": 49.73284821510315, "loss": 0.666726416349411, "time": 0.38310723304748534, "epoch": 20, "memory": 20334, "step": 111202} +{"lr": 1.6094271405406862e-05, "data_time": 0.0017825603485107423, "grad_norm": 33.947696113586424, "loss": 0.6709316611289978, "time": 0.3818256378173828, "epoch": 20, "memory": 20334, "step": 111302} +{"lr": 1.6094271405406862e-05, "data_time": 0.0017880678176879882, "grad_norm": 148.23197169303893, "loss": 0.6696382522583008, "time": 0.39011049270629883, "epoch": 20, "memory": 20334, "step": 111402} +{"lr": 1.6094271405406862e-05, "data_time": 0.002052927017211914, "grad_norm": 24.143499755859374, "loss": 0.6763786494731903, "time": 0.38851616382598875, "epoch": 20, "memory": 20334, "step": 111502} +{"lr": 1.6094271405406862e-05, "data_time": 0.0018058538436889649, "grad_norm": 29.874695301055908, "loss": 0.6745240867137909, "time": 0.3702345848083496, "epoch": 20, "memory": 20334, "step": 111602} +{"lr": 1.6094271405406862e-05, "data_time": 0.0020065069198608398, "grad_norm": 91.62142691612243, "loss": 0.6710270464420318, "time": 0.43645074367523196, "epoch": 20, "memory": 20334, "step": 111702} +{"lr": 1.6094271405406862e-05, "data_time": 0.0017354965209960937, "grad_norm": 86.21442022323609, "loss": 0.6670799493789673, "time": 0.41526498794555666, "epoch": 20, "memory": 20334, "step": 111802} +{"lr": 1.6094271405406862e-05, "data_time": 0.0025794506072998047, "grad_norm": 304.8029176235199, "loss": 0.6622025430202484, "time": 0.4083908796310425, "epoch": 20, "memory": 20334, "step": 111902} +{"lr": 1.6094271405406862e-05, "data_time": 0.002175760269165039, "grad_norm": 97.69507846832275, "loss": 0.6739909708499908, "time": 0.4042088508605957, "epoch": 20, "memory": 20334, "step": 112002} +{"lr": 1.6094271405406862e-05, "data_time": 0.0019194126129150391, "grad_norm": 93.46606841087342, "loss": 0.6579058706760407, "time": 0.3818637847900391, "epoch": 20, "memory": 20334, "step": 112102} +{"lr": 1.6094271405406862e-05, "data_time": 0.0018822431564331054, "grad_norm": 38.137491941452026, "loss": 0.664556086063385, "time": 0.37654929161071776, "epoch": 20, "memory": 20334, "step": 112202} +{"lr": 1.6094271405406862e-05, "data_time": 0.002132606506347656, "grad_norm": 38.68783128261566, "loss": 0.6653063118457794, "time": 0.4129112482070923, "epoch": 20, "memory": 20334, "step": 112302} +{"lr": 1.6094271405406862e-05, "data_time": 0.0021467685699462892, "grad_norm": 22.53668384552002, "loss": 0.675944036245346, "time": 0.3858873128890991, "epoch": 20, "memory": 20334, "step": 112402} +{"lr": 1.6094271405406862e-05, "data_time": 0.0020844697952270507, "grad_norm": 51.04946417808533, "loss": 0.6672535479068756, "time": 0.4299804210662842, "epoch": 20, "memory": 20334, "step": 112502} +{"lr": 1.6094271405406862e-05, "data_time": 0.0032830238342285156, "grad_norm": 46.97947225570679, "loss": 0.6686098754405976, "time": 0.40066864490509035, "epoch": 20, "memory": 20334, "step": 112602} +{"lr": 1.6094271405406862e-05, "data_time": 0.0025171756744384764, "grad_norm": 94.8532820224762, "loss": 0.6783220827579498, "time": 0.3917621850967407, "epoch": 20, "memory": 20334, "step": 112702} +{"lr": 1.6094271405406862e-05, "data_time": 0.003272390365600586, "grad_norm": 36.096519947052, "loss": 0.6773834824562073, "time": 0.38307321071624756, "epoch": 20, "memory": 20334, "step": 112802} +{"lr": 1.6094271405406862e-05, "data_time": 0.001750349998474121, "grad_norm": 14.777117347717285, "loss": 0.669330495595932, "time": 0.38584651947021487, "epoch": 20, "memory": 20334, "step": 112902} +{"lr": 1.6094271405406862e-05, "data_time": 0.0025300025939941407, "grad_norm": 48.91781497001648, "loss": 0.682118809223175, "time": 0.40066051483154297, "epoch": 20, "memory": 20334, "step": 113002} +{"lr": 1.6094271405406862e-05, "data_time": 0.002970170974731445, "grad_norm": 41.183049011230466, "loss": 0.6624325454235077, "time": 0.41779313087463377, "epoch": 20, "memory": 20334, "step": 113102} +{"lr": 1.6094271405406862e-05, "data_time": 0.0019422292709350586, "grad_norm": 49.42296929359436, "loss": 0.6682762026786804, "time": 0.3991267681121826, "epoch": 20, "memory": 20334, "step": 113202} +{"lr": 1.6094271405406862e-05, "data_time": 0.0018179893493652343, "grad_norm": 61.340555572509764, "loss": 0.6697454810142517, "time": 0.38625288009643555, "epoch": 20, "memory": 20334, "step": 113302} +{"lr": 1.6094271405406862e-05, "data_time": 0.0020168542861938475, "grad_norm": 20.791267371177675, "loss": 0.6624826729297638, "time": 0.43474817276000977, "epoch": 20, "memory": 20334, "step": 113402} +{"lr": 1.6094271405406862e-05, "data_time": 0.0022972583770751952, "grad_norm": 58.92897605895996, "loss": 0.6696829795837402, "time": 0.41529862880706786, "epoch": 20, "memory": 20334, "step": 113502} +{"lr": 1.6094271405406862e-05, "data_time": 0.002086949348449707, "grad_norm": 18.95285291671753, "loss": 0.6847931981086731, "time": 0.39669802188873293, "epoch": 20, "memory": 20334, "step": 113602} +{"lr": 1.6094271405406862e-05, "data_time": 0.0021726131439208985, "grad_norm": 20.612304973602296, "loss": 0.679996770620346, "time": 0.386553692817688, "epoch": 20, "memory": 20334, "step": 113702} +{"lr": 1.6094271405406862e-05, "data_time": 0.001727747917175293, "grad_norm": 46.01411848068237, "loss": 0.6571497023105621, "time": 0.4148284435272217, "epoch": 20, "memory": 20334, "step": 113802} +{"lr": 1.6094271405406862e-05, "data_time": 0.0020232439041137696, "grad_norm": 135.2398318529129, "loss": 0.6718491733074188, "time": 0.4023731708526611, "epoch": 20, "memory": 20334, "step": 113902} +{"lr": 1.6094271405406862e-05, "data_time": 0.001869368553161621, "grad_norm": 42.532611989974974, "loss": 0.6674555003643036, "time": 0.407744026184082, "epoch": 20, "memory": 20334, "step": 114002} +{"lr": 1.6094271405406862e-05, "data_time": 0.0025879621505737306, "grad_norm": 20.755958843231202, "loss": 0.6798424124717712, "time": 0.4156376600265503, "epoch": 20, "memory": 20334, "step": 114102} +{"lr": 1.6094271405406862e-05, "data_time": 0.0021868467330932615, "grad_norm": 12.93769965171814, "loss": 0.6681676089763642, "time": 0.36259660720825193, "epoch": 20, "memory": 20334, "step": 114202} +{"lr": 1.6094271405406862e-05, "data_time": 0.001917719841003418, "grad_norm": 40.20782213211059, "loss": 0.6652441084384918, "time": 0.4466781377792358, "epoch": 20, "memory": 20334, "step": 114302} +{"lr": 1.6094271405406862e-05, "data_time": 0.002218341827392578, "grad_norm": 925.7426535129547, "loss": 0.6715802431106568, "time": 0.39341087341308595, "epoch": 20, "memory": 20334, "step": 114402} +{"lr": 1.6094271405406862e-05, "data_time": 0.0016837835311889649, "grad_norm": 31.550852966308593, "loss": 0.662868195772171, "time": 0.3918366193771362, "epoch": 20, "memory": 20334, "step": 114502} +{"lr": 1.6094271405406862e-05, "data_time": 0.002666282653808594, "grad_norm": 53.29404301643372, "loss": 0.6778385102748871, "time": 0.4472174167633057, "epoch": 20, "memory": 20334, "step": 114602} +{"lr": 1.6094271405406862e-05, "data_time": 0.0022686004638671877, "grad_norm": 202.7078077316284, "loss": 0.6774605691432953, "time": 0.3977050304412842, "epoch": 20, "memory": 20334, "step": 114702} +{"lr": 1.6094271405406862e-05, "data_time": 0.0017225503921508788, "grad_norm": 117.04723069667816, "loss": 0.6648397505283355, "time": 0.44902782440185546, "epoch": 20, "memory": 20334, "step": 114802} +{"lr": 1.6094271405406862e-05, "data_time": 0.002423000335693359, "grad_norm": 166.69438486099244, "loss": 0.674306458234787, "time": 0.4497149705886841, "epoch": 20, "memory": 20334, "step": 114902} +{"lr": 1.6094271405406862e-05, "data_time": 0.003200793266296387, "grad_norm": 60.18272271156311, "loss": 0.6722406625747681, "time": 0.4039085626602173, "epoch": 20, "memory": 20334, "step": 115002} +{"lr": 1.6094271405406862e-05, "data_time": 0.001801276206970215, "grad_norm": 116.333566904068, "loss": 0.6590107202529907, "time": 0.37529284954071046, "epoch": 20, "memory": 20334, "step": 115102} +{"accuracy/top1": 80.52521514892578, "data_time": 0.37442196740044487, "time": 1.0008376439412434, "step": 20} diff --git a/swin_base_8xb128_1e-3lr_5m/20230531_183241/vis_data/config.py b/swin_base_8xb128_1e-3lr_5m/20230531_183241/vis_data/config.py new file mode 100644 index 0000000000000000000000000000000000000000..bd4397cc02b636fc0a458413dc88abb66e856cd9 --- /dev/null +++ b/swin_base_8xb128_1e-3lr_5m/20230531_183241/vis_data/config.py @@ -0,0 +1,174 @@ +optim_wrapper = dict( + optimizer=dict( + type='AdamW', + lr=0.001, + weight_decay=0.05, + eps=1e-08, + betas=(0.9, 0.999), + _scope_='mmpretrain'), + paramwise_cfg=dict( + norm_decay_mult=0.0, + bias_decay_mult=0.0, + flat_decay_mult=0.0, + custom_keys=dict({ + '.absolute_pos_embed': dict(decay_mult=0.0), + '.relative_position_bias_table': dict(decay_mult=0.0) + })), + type='AmpOptimWrapper', + dtype='bfloat16', + clip_grad=dict(max_norm=5.0)) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=True, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=20, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=1024) +model = dict( + type='ImageClassifier', + backbone=dict( + type='SwinTransformer', arch='base', img_size=224, drop_path_rate=0.5), + neck=dict(type='GlobalAveragePooling'), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=1024, + init_cfg=None, + loss=dict( + type='LabelSmoothLoss', label_smooth_val=0.1, mode='original'), + cal_acc=False), + init_cfg=[ + dict(type='TruncNormal', layer='Linear', std=0.02, bias=0.0), + dict(type='Constant', layer='LayerNorm', val=1.0, bias=0.0) + ], + train_cfg=None) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=128, + num_workers=10, + dataset=dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stablediffusionV1-5R2-dpmsolver-25-5m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/swin_base_8xb128_1e-3lr_5m' diff --git a/swin_base_8xb128_1e-3lr_5m/20230531_183241/vis_data/events.out.tfevents.1685529170.SH-IDC1-10-140-24-131.256722.0 b/swin_base_8xb128_1e-3lr_5m/20230531_183241/vis_data/events.out.tfevents.1685529170.SH-IDC1-10-140-24-131.256722.0 new file mode 100644 index 0000000000000000000000000000000000000000..bfa0acdd59d4a8a2f13de97e2c0edf861167c922 --- /dev/null +++ b/swin_base_8xb128_1e-3lr_5m/20230531_183241/vis_data/events.out.tfevents.1685529170.SH-IDC1-10-140-24-131.256722.0 @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:b3172d7aa03f14d6d077a6a797987ed357a7b717586e9568f468363b0769fc62 +size 1789178 diff --git a/swin_base_8xb128_1e-3lr_5m/20230531_183241/vis_data/scalars.json b/swin_base_8xb128_1e-3lr_5m/20230531_183241/vis_data/scalars.json new file mode 100644 index 0000000000000000000000000000000000000000..1103fa3896a89a2b820058511435d2bec4d37862 --- /dev/null +++ b/swin_base_8xb128_1e-3lr_5m/20230531_183241/vis_data/scalars.json @@ -0,0 +1,1160 @@ +{"lr": 0.001, "data_time": 0.001547527313232422, "grad_norm": 1.2535260617733002, "loss": 0.6617416918277741, "time": 0.4504673480987549, "epoch": 1, "memory": 20328, "step": 100} +{"lr": 0.001, "data_time": 0.0021373510360717775, "grad_norm": 1.3020053178071975, "loss": 0.6786544740200042, "time": 0.41860456466674806, "epoch": 1, "memory": 20327, "step": 200} +{"lr": 0.001, "data_time": 0.0015830039978027345, "grad_norm": 1.6163756191730498, "loss": 0.6961738705635071, "time": 0.41177544593811033, "epoch": 1, "memory": 20327, "step": 300} +{"lr": 0.001, "data_time": 0.001738595962524414, "grad_norm": 0.916207692027092, "loss": 0.6909652471542358, "time": 0.44023077487945556, "epoch": 1, "memory": 20327, "step": 400} +{"lr": 0.001, "data_time": 0.002464485168457031, "grad_norm": 0.7051077991724014, "loss": 0.6895924627780914, "time": 0.48884766101837157, "epoch": 1, "memory": 20327, "step": 500} +{"lr": 0.001, "data_time": 0.0024768590927124025, "grad_norm": 1.066253486275673, "loss": 0.6935589075088501, "time": 0.4829955816268921, "epoch": 1, "memory": 20327, "step": 600} +{"lr": 0.001, "data_time": 0.0024304866790771486, "grad_norm": 0.6848722219467163, "loss": 0.692876398563385, "time": 0.44585633277893066, "epoch": 1, "memory": 20327, "step": 700} +{"lr": 0.001, "data_time": 0.002019524574279785, "grad_norm": 0.4387118622660637, "loss": 0.6869839131832123, "time": 0.47382638454437254, "epoch": 1, "memory": 20327, "step": 800} +{"lr": 0.001, "data_time": 0.0021958351135253906, "grad_norm": 0.7460969805717468, "loss": 0.6842294037342072, "time": 0.48500847816467285, "epoch": 1, "memory": 20327, "step": 900} +{"lr": 0.001, "data_time": 0.0018077850341796874, "grad_norm": 0.5114121817052364, "loss": 0.6820880651474, "time": 0.5280540943145752, "epoch": 1, "memory": 20327, "step": 1000} +{"lr": 0.001, "data_time": 0.0019245386123657227, "grad_norm": 0.43836204409599305, "loss": 0.6854023098945617, "time": 0.42091100215911864, "epoch": 1, "memory": 20327, "step": 1100} +{"lr": 0.001, "data_time": 0.0017555475234985352, "grad_norm": 0.581676198542118, "loss": 0.6898906469345093, "time": 0.4580332040786743, "epoch": 1, "memory": 20327, "step": 1200} +{"lr": 0.001, "data_time": 0.0014124155044555665, "grad_norm": 0.4683442488312721, "loss": 0.684032940864563, "time": 0.4685969829559326, "epoch": 1, "memory": 20327, "step": 1300} +{"lr": 0.001, "data_time": 0.001987648010253906, "grad_norm": 0.4273607209324837, "loss": 0.6885849595069885, "time": 0.44467649459838865, "epoch": 1, "memory": 20327, "step": 1400} +{"lr": 0.001, "data_time": 0.0024909496307373045, "grad_norm": 0.3759254410862923, "loss": 0.6895965993404388, "time": 0.4259899616241455, "epoch": 1, "memory": 20327, "step": 1500} +{"lr": 0.001, "data_time": 0.0016693115234375, "grad_norm": 0.47642067819833755, "loss": 0.6906299531459809, "time": 0.4184162139892578, "epoch": 1, "memory": 20327, "step": 1600} +{"lr": 0.001, "data_time": 0.002308964729309082, "grad_norm": 0.31735817939043043, "loss": 0.688661515712738, "time": 0.42727036476135255, "epoch": 1, "memory": 20327, "step": 1700} +{"lr": 0.001, "data_time": 0.05264580249786377, "grad_norm": 0.23922348469495774, "loss": 0.6857641935348511, "time": 0.44586069583892823, "epoch": 1, "memory": 20327, "step": 1800} +{"lr": 0.001, "data_time": 0.0017943620681762696, "grad_norm": 0.3035322234034538, "loss": 0.6880684614181518, "time": 0.5702909708023072, "epoch": 1, "memory": 20327, "step": 1900} +{"lr": 0.001, "data_time": 0.0017334461212158204, "grad_norm": 0.24675946086645126, "loss": 0.6844491779804229, "time": 0.5232514619827271, "epoch": 1, "memory": 20327, "step": 2000} +{"lr": 0.001, "data_time": 0.0015720605850219726, "grad_norm": 0.24305076003074647, "loss": 0.6868813693523407, "time": 0.4493051528930664, "epoch": 1, "memory": 20327, "step": 2100} +{"lr": 0.001, "data_time": 0.11564030647277831, "grad_norm": 0.21443298310041428, "loss": 0.6881125211715698, "time": 0.4997014760971069, "epoch": 1, "memory": 20327, "step": 2200} +{"lr": 0.001, "data_time": 0.0016659259796142577, "grad_norm": 0.32948400005698203, "loss": 0.6890442490577697, "time": 0.46114354133605956, "epoch": 1, "memory": 20327, "step": 2300} +{"lr": 0.001, "data_time": 0.0015170812606811524, "grad_norm": 0.32062850557267664, "loss": 0.683380401134491, "time": 0.4767552137374878, "epoch": 1, "memory": 20327, "step": 2400} +{"lr": 0.001, "data_time": 0.0020232439041137696, "grad_norm": 0.18023623824119567, "loss": 0.6857121706008911, "time": 0.4431389093399048, "epoch": 1, "memory": 20327, "step": 2500} +{"lr": 0.001, "data_time": 0.001457381248474121, "grad_norm": 0.2827820084989071, "loss": 0.6844555914402009, "time": 0.5280491590499878, "epoch": 1, "memory": 20327, "step": 2600} +{"lr": 0.001, "data_time": 0.001476883888244629, "grad_norm": 0.23454785868525505, "loss": 0.6897346138954162, "time": 0.45286059379577637, "epoch": 1, "memory": 20327, "step": 2700} +{"lr": 0.001, "data_time": 0.0018162965774536134, "grad_norm": 0.15394328869879245, "loss": 0.6885932683944702, "time": 0.5384252548217774, "epoch": 1, "memory": 20327, "step": 2800} +{"lr": 0.001, "data_time": 0.001485729217529297, "grad_norm": 0.24269349724054337, "loss": 0.6879509806632995, "time": 0.43753461837768554, "epoch": 1, "memory": 20327, "step": 2900} +{"lr": 0.001, "data_time": 0.0013791561126708985, "grad_norm": 0.23015224523842334, "loss": 0.691087681055069, "time": 0.49692587852478026, "epoch": 1, "memory": 20327, "step": 3000} +{"lr": 0.001, "data_time": 0.0022708892822265623, "grad_norm": 0.20756464637815952, "loss": 0.6824763238430023, "time": 0.5300716400146485, "epoch": 1, "memory": 20327, "step": 3100} +{"lr": 0.001, "data_time": 0.0015095949172973632, "grad_norm": 0.25365260392427447, "loss": 0.6853608310222625, "time": 0.5860336065292359, "epoch": 1, "memory": 20327, "step": 3200} +{"lr": 0.001, "data_time": 0.0024956226348876952, "grad_norm": 0.18091317750513553, "loss": 0.6864110231399536, "time": 0.46506948471069337, "epoch": 1, "memory": 20327, "step": 3300} +{"lr": 0.001, "data_time": 0.0018155813217163087, "grad_norm": 0.2349975012242794, "loss": 0.6889593541622162, "time": 0.4547833204269409, "epoch": 1, "memory": 20327, "step": 3400} +{"lr": 0.001, "data_time": 0.0017111063003540038, "grad_norm": 0.19061683416366576, "loss": 0.6881612062454223, "time": 0.4564051628112793, "epoch": 1, "memory": 20327, "step": 3500} +{"lr": 0.001, "data_time": 0.00162198543548584, "grad_norm": 0.21332599148154258, "loss": 0.6878034114837647, "time": 0.46280708312988283, "epoch": 1, "memory": 20327, "step": 3600} +{"lr": 0.001, "data_time": 0.0021987199783325196, "grad_norm": 0.21882116943597793, "loss": 0.6868311047554017, "time": 0.4189113140106201, "epoch": 1, "memory": 20327, "step": 3700} +{"lr": 0.001, "data_time": 0.0015400409698486327, "grad_norm": 0.15679723285138608, "loss": 0.6828992605209351, "time": 0.47562575340270996, "epoch": 1, "memory": 20327, "step": 3800} +{"lr": 0.001, "data_time": 0.0014181852340698243, "grad_norm": 0.24460803493857383, "loss": 0.6878409683704376, "time": 0.5233757495880127, "epoch": 1, "memory": 20327, "step": 3900} +{"lr": 0.001, "data_time": 0.0016360282897949219, "grad_norm": 0.2751976024359465, "loss": 0.6841594159603119, "time": 0.5318449974060059, "epoch": 1, "memory": 20327, "step": 4000} +{"lr": 0.001, "data_time": 0.0014498233795166016, "grad_norm": 0.1677925668656826, "loss": 0.6814287006855011, "time": 0.4640952110290527, "epoch": 1, "memory": 20327, "step": 4100} +{"lr": 0.001, "data_time": 0.0015105962753295898, "grad_norm": 0.22063578739762307, "loss": 0.6908972799777985, "time": 0.6015638113021851, "epoch": 1, "memory": 20327, "step": 4200} +{"lr": 0.001, "data_time": 0.001835179328918457, "grad_norm": 0.13054346553981305, "loss": 0.6935339093208313, "time": 0.4642556428909302, "epoch": 1, "memory": 20327, "step": 4300} +{"lr": 0.001, "data_time": 0.0017675638198852539, "grad_norm": 0.18920982331037522, "loss": 0.680409848690033, "time": 0.5172134637832642, "epoch": 1, "memory": 20327, "step": 4400} +{"lr": 0.001, "data_time": 0.001902294158935547, "grad_norm": 4.143417645990849, "loss": 0.6886740803718567, "time": 0.5838793039321899, "epoch": 1, "memory": 20327, "step": 4500} +{"lr": 0.001, "data_time": 0.04444584846496582, "grad_norm": 0.28976195603609084, "loss": 0.6888139069080352, "time": 0.48262581825256345, "epoch": 1, "memory": 20327, "step": 4600} +{"lr": 0.001, "data_time": 0.011837482452392578, "grad_norm": 0.22792382091283797, "loss": 0.6893905997276306, "time": 0.5790437221527099, "epoch": 1, "memory": 20327, "step": 4700} +{"lr": 0.001, "data_time": 0.04299075603485107, "grad_norm": 0.14554290305823087, "loss": 0.6873553156852722, "time": 0.40713043212890626, "epoch": 1, "memory": 20327, "step": 4800} +{"lr": 0.001, "data_time": 0.0017435789108276368, "grad_norm": 0.24170147106051446, "loss": 0.6905454933643341, "time": 0.46413068771362304, "epoch": 1, "memory": 20327, "step": 4900} +{"lr": 0.001, "data_time": 0.0017269372940063477, "grad_norm": 0.18632605150341988, "loss": 0.6864282250404358, "time": 0.47283387184143066, "epoch": 1, "memory": 20327, "step": 5000} +{"lr": 0.001, "data_time": 0.001972675323486328, "grad_norm": 0.16910027228295804, "loss": 0.6881202518939972, "time": 0.4918528079986572, "epoch": 1, "memory": 20327, "step": 5100} +{"lr": 0.001, "data_time": 0.0016173601150512695, "grad_norm": 0.2514938943088055, "loss": 0.6828402757644654, "time": 0.434708309173584, "epoch": 1, "memory": 20327, "step": 5200} +{"lr": 0.001, "data_time": 0.0014672040939331054, "grad_norm": 0.19111468642950058, "loss": 0.684564745426178, "time": 0.47212269306182864, "epoch": 1, "memory": 20327, "step": 5300} +{"lr": 0.001, "data_time": 0.0014446258544921875, "grad_norm": 0.1328372661024332, "loss": 0.6848789632320404, "time": 0.46407668590545653, "epoch": 1, "memory": 20327, "step": 5400} +{"lr": 0.001, "data_time": 0.0016707897186279297, "grad_norm": 0.1780287515372038, "loss": 0.684296190738678, "time": 0.5697210550308227, "epoch": 1, "memory": 20327, "step": 5500} +{"lr": 0.001, "data_time": 0.0016327142715454102, "grad_norm": 0.19417341612279415, "loss": 0.6819146037101745, "time": 0.3914592981338501, "epoch": 1, "memory": 20327, "step": 5600} +{"lr": 0.001, "data_time": 0.0016374826431274415, "grad_norm": 0.2729407835751772, "loss": 0.6899257600307465, "time": 0.5306423902511597, "epoch": 1, "memory": 20327, "step": 5700} +{"accuracy/top1": 100.0, "data_time": 0.5273530781269073, "time": 1.4208164513111115, "step": 1} +{"lr": 0.0009939057285945933, "data_time": 0.0018706560134887696, "grad_norm": 0.1389899268746376, "loss": 0.6856710553169251, "time": 0.47192771434783937, "epoch": 2, "memory": 20338, "step": 5858} +{"lr": 0.0009939057285945933, "data_time": 0.0014982223510742188, "grad_norm": 0.203866333514452, "loss": 0.6875182628631592, "time": 0.37861616611480714, "epoch": 2, "memory": 20334, "step": 5958} +{"lr": 0.0009939057285945933, "data_time": 0.002318286895751953, "grad_norm": 0.18244499415159227, "loss": 0.6888192117214202, "time": 0.4775179386138916, "epoch": 2, "memory": 20334, "step": 6058} +{"lr": 0.0009939057285945933, "data_time": 0.0020490169525146486, "grad_norm": 0.10983743183314801, "loss": 0.687062245607376, "time": 0.40245158672332765, "epoch": 2, "memory": 20334, "step": 6158} +{"lr": 0.0009939057285945933, "data_time": 0.002003383636474609, "grad_norm": 0.16991456001996993, "loss": 0.6880364120006561, "time": 0.4649769067764282, "epoch": 2, "memory": 20334, "step": 6258} +{"lr": 0.0009939057285945933, "data_time": 0.002188229560852051, "grad_norm": 0.2249813586473465, "loss": 0.6882048785686493, "time": 0.4856530427932739, "epoch": 2, "memory": 20334, "step": 6358} +{"lr": 0.0009939057285945933, "data_time": 0.002771282196044922, "grad_norm": 0.09802362509071827, "loss": 0.6891048014163971, "time": 0.5223216056823731, "epoch": 2, "memory": 20334, "step": 6458} +{"lr": 0.0009939057285945933, "data_time": 0.0017272233963012695, "grad_norm": 0.20471495091915132, "loss": 0.6880715608596801, "time": 0.4997502088546753, "epoch": 2, "memory": 20334, "step": 6558} +{"lr": 0.0009939057285945933, "data_time": 0.0015089988708496093, "grad_norm": 0.13341220766305922, "loss": 0.6904010236263275, "time": 0.4293144941329956, "epoch": 2, "memory": 20334, "step": 6658} +{"lr": 0.0009939057285945933, "data_time": 0.0023128747940063476, "grad_norm": 0.1210234124213457, "loss": 0.6860248982906342, "time": 0.5041555881500244, "epoch": 2, "memory": 20334, "step": 6758} +{"lr": 0.0009939057285945933, "data_time": 0.0016096115112304687, "grad_norm": 0.12085049711167813, "loss": 0.6867158532142639, "time": 0.4423695087432861, "epoch": 2, "memory": 20334, "step": 6858} +{"lr": 0.0009939057285945933, "data_time": 0.001529359817504883, "grad_norm": 0.10074920188635587, "loss": 0.6838590800762177, "time": 0.45317051410675047, "epoch": 2, "memory": 20334, "step": 6958} +{"lr": 0.0009939057285945933, "data_time": 0.0014278888702392578, "grad_norm": 0.2354393795132637, "loss": 0.688062334060669, "time": 0.4827596187591553, "epoch": 2, "memory": 20334, "step": 7058} +{"lr": 0.0009939057285945933, "data_time": 0.0013913154602050782, "grad_norm": 0.12252147868275642, "loss": 0.6852106213569641, "time": 0.44258482456207277, "epoch": 2, "memory": 20334, "step": 7158} +{"lr": 0.0009939057285945933, "data_time": 0.0016436576843261719, "grad_norm": 0.1605440594255924, "loss": 0.6864913821220398, "time": 0.40481493473052976, "epoch": 2, "memory": 20334, "step": 7258} +{"lr": 0.0009939057285945933, "data_time": 0.0013869285583496093, "grad_norm": 0.079430715367198, "loss": 0.6856729388237, "time": 0.38026976585388184, "epoch": 2, "memory": 20334, "step": 7358} +{"lr": 0.0009939057285945933, "data_time": 0.001480865478515625, "grad_norm": 0.13062963932752608, "loss": 0.686570119857788, "time": 0.4127319097518921, "epoch": 2, "memory": 20334, "step": 7458} +{"lr": 0.0009939057285945933, "data_time": 0.0014569520950317382, "grad_norm": 0.18666400890797377, "loss": 0.6869749307632447, "time": 0.43243656158447263, "epoch": 2, "memory": 20334, "step": 7558} +{"lr": 0.0009939057285945933, "data_time": 0.0015127420425415038, "grad_norm": 0.1899154122918844, "loss": 0.6858054041862488, "time": 0.4226412534713745, "epoch": 2, "memory": 20334, "step": 7658} +{"lr": 0.0009939057285945933, "data_time": 0.0015762805938720702, "grad_norm": 0.12433419600129128, "loss": 0.6885032474994659, "time": 0.40570530891418455, "epoch": 2, "memory": 20334, "step": 7758} +{"lr": 0.0009939057285945933, "data_time": 0.0014954090118408203, "grad_norm": 0.0796710742637515, "loss": 0.6879362523555755, "time": 0.4477119207382202, "epoch": 2, "memory": 20334, "step": 7858} +{"lr": 0.0009939057285945933, "data_time": 0.0017971992492675781, "grad_norm": 0.13888155594468116, "loss": 0.690966647863388, "time": 0.5014991283416748, "epoch": 2, "memory": 20334, "step": 7958} +{"lr": 0.0009939057285945933, "data_time": 0.0016384124755859375, "grad_norm": 0.0999188207089901, "loss": 0.6889211356639862, "time": 0.3787181615829468, "epoch": 2, "memory": 20334, "step": 8058} +{"lr": 0.0009939057285945933, "data_time": 0.0015300989151000976, "grad_norm": 0.09481384158134461, "loss": 0.6863315820693969, "time": 0.45148138999938964, "epoch": 2, "memory": 20334, "step": 8158} +{"lr": 0.0009939057285945933, "data_time": 0.0015707969665527343, "grad_norm": 0.11229593753814697, "loss": 0.6916408360004425, "time": 0.44246342182159426, "epoch": 2, "memory": 20334, "step": 8258} +{"lr": 0.0009939057285945933, "data_time": 0.0016265630722045899, "grad_norm": 0.11539451573044061, "loss": 0.6822717428207398, "time": 0.43319680690765383, "epoch": 2, "memory": 20334, "step": 8358} +{"lr": 0.0009939057285945933, "data_time": 0.0014976739883422851, "grad_norm": 0.1470975536853075, "loss": 0.6900055825710296, "time": 0.411281418800354, "epoch": 2, "memory": 20334, "step": 8458} +{"lr": 0.0009939057285945933, "data_time": 0.001532769203186035, "grad_norm": 0.10724248215556145, "loss": 0.6859984517097473, "time": 0.38384811878204345, "epoch": 2, "memory": 20334, "step": 8558} +{"lr": 0.0009939057285945933, "data_time": 0.001632833480834961, "grad_norm": 0.1074720073491335, "loss": 0.6898310124874115, "time": 0.4390037298202515, "epoch": 2, "memory": 20334, "step": 8658} +{"lr": 0.0009939057285945933, "data_time": 0.0015862941741943359, "grad_norm": 0.1138488307595253, "loss": 0.6887602984905243, "time": 0.3876753091812134, "epoch": 2, "memory": 20334, "step": 8758} +{"lr": 0.0009939057285945933, "data_time": 0.0017073392868041993, "grad_norm": 0.08672575345262885, "loss": 0.6898120701313019, "time": 0.3934889554977417, "epoch": 2, "memory": 20334, "step": 8858} +{"lr": 0.0009939057285945933, "data_time": 0.0016271829605102538, "grad_norm": 0.1329948291182518, "loss": 0.6889298319816589, "time": 0.45748813152313234, "epoch": 2, "memory": 20334, "step": 8958} +{"lr": 0.0009939057285945933, "data_time": 0.00175931453704834, "grad_norm": 0.0947910262271762, "loss": 0.6842213153839112, "time": 0.43602006435394286, "epoch": 2, "memory": 20334, "step": 9058} +{"lr": 0.0009939057285945933, "data_time": 0.0017358064651489258, "grad_norm": 0.12013679184019566, "loss": 0.6887552499771118, "time": 0.40340964794158934, "epoch": 2, "memory": 20334, "step": 9158} +{"lr": 0.0009939057285945933, "data_time": 0.001702094078063965, "grad_norm": 0.1379565104842186, "loss": 0.682598215341568, "time": 0.3863911390304565, "epoch": 2, "memory": 20334, "step": 9258} +{"lr": 0.0009939057285945933, "data_time": 0.001566028594970703, "grad_norm": 0.07153388736769557, "loss": 0.6875362396240234, "time": 0.42173769474029543, "epoch": 2, "memory": 20334, "step": 9358} +{"lr": 0.0009939057285945933, "data_time": 0.001628279685974121, "grad_norm": 0.08207729160785675, "loss": 0.6868714213371276, "time": 0.402485728263855, "epoch": 2, "memory": 20334, "step": 9458} +{"lr": 0.0009939057285945933, "data_time": 0.0014389991760253907, "grad_norm": 0.08153223311528564, "loss": 0.6841646730899811, "time": 0.397409725189209, "epoch": 2, "memory": 20334, "step": 9558} +{"lr": 0.0009939057285945933, "data_time": 0.001594686508178711, "grad_norm": 0.061381729785352945, "loss": 0.6824875712394715, "time": 0.42694969177246095, "epoch": 2, "memory": 20334, "step": 9658} +{"lr": 0.0009939057285945933, "data_time": 0.001752185821533203, "grad_norm": 0.05635301554575563, "loss": 0.6929938197135925, "time": 0.40352165699005127, "epoch": 2, "memory": 20334, "step": 9758} +{"lr": 0.0009939057285945933, "data_time": 0.0016124248504638672, "grad_norm": 0.0691122176591307, "loss": 0.6821423649787903, "time": 0.4068138837814331, "epoch": 2, "memory": 20334, "step": 9858} +{"lr": 0.0009939057285945933, "data_time": 0.00162811279296875, "grad_norm": 0.06478845039382577, "loss": 0.6845524370670318, "time": 0.4095698595046997, "epoch": 2, "memory": 20334, "step": 9958} +{"lr": 0.0009939057285945933, "data_time": 0.0017833948135375977, "grad_norm": 0.08239487018436194, "loss": 0.6868374109268188, "time": 0.41894381046295165, "epoch": 2, "memory": 20334, "step": 10058} +{"lr": 0.0009939057285945933, "data_time": 0.0018874168395996093, "grad_norm": 0.07758684400469065, "loss": 0.6857448697090149, "time": 0.41786601543426516, "epoch": 2, "memory": 20334, "step": 10158} +{"lr": 0.0009939057285945933, "data_time": 0.001524806022644043, "grad_norm": 0.07034221142530442, "loss": 0.6880883395671844, "time": 0.4282370090484619, "epoch": 2, "memory": 20334, "step": 10258} +{"lr": 0.0009939057285945933, "data_time": 0.001620316505432129, "grad_norm": 0.05666370280086994, "loss": 0.6862131476402282, "time": 0.3999110460281372, "epoch": 2, "memory": 20334, "step": 10358} +{"lr": 0.0009939057285945933, "data_time": 0.0018820524215698241, "grad_norm": 0.03454902684316039, "loss": 0.6858645379543304, "time": 0.3662744045257568, "epoch": 2, "memory": 20334, "step": 10458} +{"lr": 0.0009939057285945933, "data_time": 0.001518416404724121, "grad_norm": 0.06227385513484478, "loss": 0.6873582184314728, "time": 0.3886637449264526, "epoch": 2, "memory": 20334, "step": 10558} +{"lr": 0.0009939057285945933, "data_time": 0.0026936769485473634, "grad_norm": 0.05943831531330943, "loss": 0.6837661266326904, "time": 0.4403040647506714, "epoch": 2, "memory": 20334, "step": 10658} +{"lr": 0.0009939057285945933, "data_time": 0.0015952348709106444, "grad_norm": 0.06036773715168238, "loss": 0.6927081525325776, "time": 0.38903734683990476, "epoch": 2, "memory": 20334, "step": 10758} +{"lr": 0.0009939057285945933, "data_time": 0.0017139673233032226, "grad_norm": 0.05382824447005987, "loss": 0.6866870164871216, "time": 0.4493489503860474, "epoch": 2, "memory": 20334, "step": 10858} +{"lr": 0.0009939057285945933, "data_time": 0.0016783475875854492, "grad_norm": 0.06428167028352619, "loss": 0.6885355353355408, "time": 0.440744686126709, "epoch": 2, "memory": 20334, "step": 10958} +{"lr": 0.0009939057285945933, "data_time": 0.002162528038024902, "grad_norm": 0.03117213575169444, "loss": 0.6879472315311432, "time": 0.363454532623291, "epoch": 2, "memory": 20334, "step": 11058} +{"lr": 0.0009939057285945933, "data_time": 0.002293848991394043, "grad_norm": 0.04414070025086403, "loss": 0.6896017730236054, "time": 0.43469820022583006, "epoch": 2, "memory": 20334, "step": 11158} +{"lr": 0.0009939057285945933, "data_time": 0.0018540382385253905, "grad_norm": 0.036172684840857984, "loss": 0.6920473754405976, "time": 0.43832666873931886, "epoch": 2, "memory": 20334, "step": 11258} +{"lr": 0.0009939057285945933, "data_time": 0.001553797721862793, "grad_norm": 0.03078562580049038, "loss": 0.6833684384822846, "time": 0.3910383224487305, "epoch": 2, "memory": 20334, "step": 11358} +{"lr": 0.0009939057285945933, "data_time": 0.001596689224243164, "grad_norm": 0.03620063448324799, "loss": 0.6886676490306854, "time": 0.39936027526855467, "epoch": 2, "memory": 20334, "step": 11458} +{"accuracy/top1": 100.0, "data_time": 0.37360231081644696, "time": 1.0013271967569988, "step": 2} +{"lr": 0.0009757729755661012, "data_time": 0.001520848274230957, "grad_norm": 0.03098635757341981, "loss": 0.686473685503006, "time": 0.4017822504043579, "epoch": 3, "memory": 20334, "step": 11616} +{"lr": 0.0009757729755661012, "data_time": 0.0015061616897583008, "grad_norm": 0.033890397474169734, "loss": 0.6903098583221435, "time": 0.39765985012054444, "epoch": 3, "memory": 20334, "step": 11716} +{"lr": 0.0009757729755661012, "data_time": 0.0015069961547851563, "grad_norm": 0.030056470725685357, "loss": 0.6796724200248718, "time": 0.3921332836151123, "epoch": 3, "memory": 20334, "step": 11816} +{"lr": 0.0009757729755661012, "data_time": 0.0016312599182128906, "grad_norm": 0.022210765909403563, "loss": 0.6907181024551392, "time": 0.4321902751922607, "epoch": 3, "memory": 20334, "step": 11916} +{"lr": 0.0009757729755661012, "data_time": 0.0014042377471923829, "grad_norm": 0.025225077383220197, "loss": 0.6878110349178315, "time": 0.3707267761230469, "epoch": 3, "memory": 20334, "step": 12016} +{"lr": 0.0009757729755661012, "data_time": 0.0014371633529663085, "grad_norm": 0.026616774778813124, "loss": 0.6892135739326477, "time": 0.4078239440917969, "epoch": 3, "memory": 20334, "step": 12116} +{"lr": 0.0009757729755661012, "data_time": 0.0014354944229125976, "grad_norm": 0.028887146292254327, "loss": 0.6870896816253662, "time": 0.41640093326568606, "epoch": 3, "memory": 20334, "step": 12216} +{"lr": 0.0009757729755661012, "data_time": 0.0015312910079956054, "grad_norm": 0.027596437186002732, "loss": 0.6887373507022858, "time": 0.4126712799072266, "epoch": 3, "memory": 20334, "step": 12316} +{"lr": 0.0009757729755661012, "data_time": 0.0013911724090576172, "grad_norm": 0.016476803366094826, "loss": 0.6819482147693634, "time": 0.4864032506942749, "epoch": 3, "memory": 20334, "step": 12416} +{"lr": 0.0009757729755661012, "data_time": 0.0016367435455322266, "grad_norm": 0.021859406493604183, "loss": 0.6840069830417633, "time": 0.4090930938720703, "epoch": 3, "memory": 20334, "step": 12516} +{"lr": 0.0009757729755661012, "data_time": 0.0014048099517822265, "grad_norm": 0.023460118751972913, "loss": 0.6887216448783875, "time": 0.43534467220306394, "epoch": 3, "memory": 20334, "step": 12616} +{"lr": 0.0009757729755661012, "data_time": 0.0014093399047851562, "grad_norm": 0.014717079047113658, "loss": 0.685056871175766, "time": 0.466562819480896, "epoch": 3, "memory": 20334, "step": 12716} +{"lr": 0.0009757729755661012, "data_time": 0.001378011703491211, "grad_norm": 0.025230715004727244, "loss": 0.6858988761901855, "time": 0.4218915939331055, "epoch": 3, "memory": 20334, "step": 12816} +{"lr": 0.0009757729755661012, "data_time": 0.0014751911163330077, "grad_norm": 0.015667925914749505, "loss": 0.688463419675827, "time": 0.46251440048217773, "epoch": 3, "memory": 20334, "step": 12916} +{"lr": 0.0009757729755661012, "data_time": 0.0014195680618286134, "grad_norm": 0.025939735118299723, "loss": 0.6954469442367553, "time": 0.4380846738815308, "epoch": 3, "memory": 20334, "step": 13016} +{"lr": 0.0009757729755661012, "data_time": 0.0023333072662353516, "grad_norm": 0.023735856171697378, "loss": 0.687801194190979, "time": 0.4287290573120117, "epoch": 3, "memory": 20334, "step": 13116} +{"lr": 0.0009757729755661012, "data_time": 0.0018094539642333984, "grad_norm": 0.019573861034587026, "loss": 0.6907796859741211, "time": 0.4156827211380005, "epoch": 3, "memory": 20334, "step": 13216} +{"lr": 0.0009757729755661012, "data_time": 0.0019041538238525391, "grad_norm": 0.022345282882452012, "loss": 0.6813410520553589, "time": 0.43147850036621094, "epoch": 3, "memory": 20334, "step": 13316} +{"lr": 0.0009757729755661012, "data_time": 0.001631474494934082, "grad_norm": 0.02729429369792342, "loss": 0.6872678399085999, "time": 0.4804619073867798, "epoch": 3, "memory": 20334, "step": 13416} +{"lr": 0.0009757729755661012, "data_time": 0.0014719724655151366, "grad_norm": 0.02647479922743514, "loss": 0.6898800075054169, "time": 0.4309577941894531, "epoch": 3, "memory": 20334, "step": 13516} +{"lr": 0.0009757729755661012, "data_time": 0.0015274763107299804, "grad_norm": 0.012423992273397744, "loss": 0.6817545771598816, "time": 0.42549293041229247, "epoch": 3, "memory": 20334, "step": 13616} +{"lr": 0.0009757729755661012, "data_time": 0.002018260955810547, "grad_norm": 0.011961333989165723, "loss": 0.6839232504367828, "time": 0.4453761339187622, "epoch": 3, "memory": 20334, "step": 13716} +{"lr": 0.0009757729755661012, "data_time": 0.0013747215270996094, "grad_norm": 0.023575666639953852, "loss": 0.690808218717575, "time": 0.46250483989715574, "epoch": 3, "memory": 20334, "step": 13816} +{"lr": 0.0009757729755661012, "data_time": 0.001609635353088379, "grad_norm": 0.018087553139775992, "loss": 0.6885170996189117, "time": 0.4347598791122437, "epoch": 3, "memory": 20334, "step": 13916} +{"lr": 0.0009757729755661012, "data_time": 0.002219557762145996, "grad_norm": 0.027281677862629295, "loss": 0.6876326441764832, "time": 0.4458295822143555, "epoch": 3, "memory": 20334, "step": 14016} +{"lr": 0.0009757729755661012, "data_time": 0.0015475749969482422, "grad_norm": 0.013312716549262405, "loss": 0.6822128236293793, "time": 0.396474027633667, "epoch": 3, "memory": 20334, "step": 14116} +{"lr": 0.0009757729755661012, "data_time": 0.0012688636779785156, "grad_norm": 0.027546300599351526, "loss": 0.6798486113548279, "time": 0.4620030641555786, "epoch": 3, "memory": 20334, "step": 14216} +{"lr": 0.0009757729755661012, "data_time": 0.0019216299057006835, "grad_norm": 0.021650880412198602, "loss": 0.6841916501522064, "time": 0.47051043510437013, "epoch": 3, "memory": 20334, "step": 14316} +{"lr": 0.0009757729755661012, "data_time": 0.0017227649688720704, "grad_norm": 0.01692313519306481, "loss": 0.6871234238147735, "time": 0.42992634773254396, "epoch": 3, "memory": 20334, "step": 14416} +{"lr": 0.0009757729755661012, "data_time": 0.0015279293060302735, "grad_norm": 0.02390893567353487, "loss": 0.6868479728698731, "time": 0.4040129899978638, "epoch": 3, "memory": 20334, "step": 14516} +{"lr": 0.0009757729755661012, "data_time": 0.0016637802124023437, "grad_norm": 0.016689633810892703, "loss": 0.686801153421402, "time": 0.39360096454620364, "epoch": 3, "memory": 20334, "step": 14616} +{"lr": 0.0009757729755661012, "data_time": 0.0015065193176269532, "grad_norm": 0.015438964287750424, "loss": 0.6861332893371582, "time": 0.4511989116668701, "epoch": 3, "memory": 20334, "step": 14716} +{"lr": 0.0009757729755661012, "data_time": 0.0016159296035766601, "grad_norm": 0.014466957421973348, "loss": 0.683395254611969, "time": 0.4681440591812134, "epoch": 3, "memory": 20334, "step": 14816} +{"lr": 0.0009757729755661012, "data_time": 0.0015751838684082032, "grad_norm": 0.022110573854297398, "loss": 0.683032101392746, "time": 0.4665671825408936, "epoch": 3, "memory": 20334, "step": 14916} +{"lr": 0.0009757729755661012, "data_time": 0.0024945974349975587, "grad_norm": 0.011839405028149485, "loss": 0.6831228733062744, "time": 0.4275668621063232, "epoch": 3, "memory": 20334, "step": 15016} +{"lr": 0.0009757729755661012, "data_time": 0.0022376298904418944, "grad_norm": 0.016086345398798586, "loss": 0.6844076752662659, "time": 0.395918869972229, "epoch": 3, "memory": 20334, "step": 15116} +{"lr": 0.0009757729755661012, "data_time": 0.0014168024063110352, "grad_norm": 0.021715958369895816, "loss": 0.6888582885265351, "time": 0.39960379600524903, "epoch": 3, "memory": 20334, "step": 15216} +{"lr": 0.0009757729755661012, "data_time": 0.0012727022171020509, "grad_norm": 0.011211237497627735, "loss": 0.6836376905441284, "time": 0.4284745454788208, "epoch": 3, "memory": 20334, "step": 15316} +{"lr": 0.0009757729755661012, "data_time": 0.0020062923431396484, "grad_norm": 0.016982131195254623, "loss": 0.6858044803142548, "time": 0.4195528984069824, "epoch": 3, "memory": 20334, "step": 15416} +{"lr": 0.0009757729755661012, "data_time": 0.0015186309814453126, "grad_norm": 0.03138001998886466, "loss": 0.6842008352279663, "time": 0.429775595664978, "epoch": 3, "memory": 20334, "step": 15516} +{"lr": 0.0009757729755661012, "data_time": 0.0020823240280151366, "grad_norm": 0.016965086571872234, "loss": 0.6856278061866761, "time": 0.4045754909515381, "epoch": 3, "memory": 20334, "step": 15616} +{"lr": 0.0009757729755661012, "data_time": 0.00151824951171875, "grad_norm": 0.01606045940425247, "loss": 0.6863158762454986, "time": 0.43207101821899413, "epoch": 3, "memory": 20334, "step": 15716} +{"lr": 0.0009757729755661012, "data_time": 0.0015492677688598634, "grad_norm": 0.01343541257083416, "loss": 0.6925902724266052, "time": 0.43695497512817383, "epoch": 3, "memory": 20334, "step": 15816} +{"lr": 0.0009757729755661012, "data_time": 0.04586305618286133, "grad_norm": 0.017851545987650753, "loss": 0.678933173418045, "time": 0.4091578722000122, "epoch": 3, "memory": 20334, "step": 15916} +{"lr": 0.0009757729755661012, "data_time": 0.06472687721252442, "grad_norm": 0.01796461995691061, "loss": 0.685563576221466, "time": 0.4260616064071655, "epoch": 3, "memory": 20334, "step": 16016} +{"lr": 0.0009757729755661012, "data_time": 0.07754015922546387, "grad_norm": 0.013906694715842605, "loss": 0.6849724590778351, "time": 0.44074790477752684, "epoch": 3, "memory": 20334, "step": 16116} +{"lr": 0.0009757729755661012, "data_time": 0.04347686767578125, "grad_norm": 0.02395407101139426, "loss": 0.6887342035770416, "time": 0.42560787200927735, "epoch": 3, "memory": 20334, "step": 16216} +{"lr": 0.0009757729755661012, "data_time": 0.0015156269073486328, "grad_norm": 0.018618854158557952, "loss": 0.6877447664737701, "time": 0.4609185218811035, "epoch": 3, "memory": 20334, "step": 16316} +{"lr": 0.0009757729755661012, "data_time": 0.0016674757003784179, "grad_norm": 0.01747428609523922, "loss": 0.6846291959285736, "time": 0.46385653018951417, "epoch": 3, "memory": 20334, "step": 16416} +{"lr": 0.0009757729755661012, "data_time": 0.0012150049209594727, "grad_norm": 0.016686558863148092, "loss": 0.6920796573162079, "time": 0.41220109462738036, "epoch": 3, "memory": 20334, "step": 16516} +{"lr": 0.0009757729755661012, "data_time": 0.0013314247131347655, "grad_norm": 0.01650532289640978, "loss": 0.6890563368797302, "time": 0.40956666469573977, "epoch": 3, "memory": 20334, "step": 16616} +{"lr": 0.0009757729755661012, "data_time": 0.002044057846069336, "grad_norm": 0.014679540507495403, "loss": 0.692743718624115, "time": 0.4864722967147827, "epoch": 3, "memory": 20334, "step": 16716} +{"lr": 0.0009757729755661012, "data_time": 0.0016607999801635741, "grad_norm": 0.014594785962253809, "loss": 0.6871274709701538, "time": 0.4180482864379883, "epoch": 3, "memory": 20334, "step": 16816} +{"lr": 0.0009757729755661012, "data_time": 0.0023061752319335936, "grad_norm": 0.014580905344337225, "loss": 0.6873303294181824, "time": 0.42358126640319826, "epoch": 3, "memory": 20334, "step": 16916} +{"lr": 0.0009757729755661012, "data_time": 0.0020044565200805662, "grad_norm": 0.016765149729326367, "loss": 0.6843859195709229, "time": 0.4132028818130493, "epoch": 3, "memory": 20334, "step": 17016} +{"lr": 0.0009757729755661012, "data_time": 0.0019243240356445312, "grad_norm": 0.014524851349415257, "loss": 0.6876846849918365, "time": 0.43416016101837157, "epoch": 3, "memory": 20334, "step": 17116} +{"lr": 0.0009757729755661012, "data_time": 0.0016091823577880859, "grad_norm": 0.023040681355632843, "loss": 0.6922148585319519, "time": 0.427603554725647, "epoch": 3, "memory": 20334, "step": 17216} +{"accuracy/top1": 100.0, "data_time": 0.36508538987901473, "time": 0.9904248184627957, "step": 3} +{"lr": 0.0009460482294732422, "data_time": 0.0017638683319091796, "grad_norm": 0.01643019961193204, "loss": 0.6822502732276916, "time": 0.3723294734954834, "epoch": 4, "memory": 20334, "step": 17374} +{"lr": 0.0009460482294732422, "data_time": 0.001596522331237793, "grad_norm": 0.0149491639691405, "loss": 0.6835873544216156, "time": 0.48390424251556396, "epoch": 4, "memory": 20334, "step": 17474} +{"lr": 0.0009460482294732422, "data_time": 0.001820540428161621, "grad_norm": 0.021044259099289774, "loss": 0.6859103977680207, "time": 0.3999966621398926, "epoch": 4, "memory": 20334, "step": 17574} +{"lr": 0.0009460482294732422, "data_time": 0.0017197370529174806, "grad_norm": 0.016584765748120844, "loss": 0.6917169094085693, "time": 0.4330127477645874, "epoch": 4, "memory": 20334, "step": 17674} +{"lr": 0.0009460482294732422, "data_time": 0.0018273115158081055, "grad_norm": 0.01176101916935295, "loss": 0.6858778119087219, "time": 0.432884669303894, "epoch": 4, "memory": 20334, "step": 17774} +{"lr": 0.0009460482294732422, "data_time": 0.0016796350479125976, "grad_norm": 0.01780250105075538, "loss": 0.6884469509124755, "time": 0.41221568584442136, "epoch": 4, "memory": 20334, "step": 17874} +{"lr": 0.0009460482294732422, "data_time": 0.0015875339508056641, "grad_norm": 0.018298603734001516, "loss": 0.6912717163562775, "time": 0.43295817375183104, "epoch": 4, "memory": 20334, "step": 17974} +{"lr": 0.0009460482294732422, "data_time": 0.0016761064529418946, "grad_norm": 0.02583385189063847, "loss": 0.6846343040466308, "time": 0.4225667953491211, "epoch": 4, "memory": 20334, "step": 18074} +{"lr": 0.0009460482294732422, "data_time": 0.0016031980514526368, "grad_norm": 0.020494342735037207, "loss": 0.6850262105464935, "time": 0.4490375280380249, "epoch": 4, "memory": 20334, "step": 18174} +{"lr": 0.0009460482294732422, "data_time": 0.0017838001251220704, "grad_norm": 0.026996927242726086, "loss": 0.6846012651920319, "time": 0.39264416694641113, "epoch": 4, "memory": 20334, "step": 18274} +{"lr": 0.0009460482294732422, "data_time": 0.0016048908233642577, "grad_norm": 0.023436758480966092, "loss": 0.6880956172943116, "time": 0.4475454330444336, "epoch": 4, "memory": 20334, "step": 18374} +{"lr": 0.0009460482294732422, "data_time": 0.0018970966339111328, "grad_norm": 0.017902054672595112, "loss": 0.6856350541114807, "time": 0.4378422737121582, "epoch": 4, "memory": 20334, "step": 18474} +{"lr": 0.0009460482294732422, "data_time": 0.001533842086791992, "grad_norm": 0.012605548347346484, "loss": 0.6849187016487122, "time": 0.42951369285583496, "epoch": 4, "memory": 20334, "step": 18574} +{"lr": 0.0009460482294732422, "data_time": 0.001800847053527832, "grad_norm": 0.0144324409076944, "loss": 0.6878259658813477, "time": 0.49349207878112794, "epoch": 4, "memory": 20334, "step": 18674} +{"lr": 0.0009460482294732422, "data_time": 0.001868867874145508, "grad_norm": 0.017530618351884186, "loss": 0.6795748353004456, "time": 0.4449542284011841, "epoch": 4, "memory": 20334, "step": 18774} +{"lr": 0.0009460482294732422, "data_time": 0.0018942117691040038, "grad_norm": 0.023734983801841737, "loss": 0.6832464158535003, "time": 0.36419076919555665, "epoch": 4, "memory": 20334, "step": 18874} +{"lr": 0.0009460482294732422, "data_time": 0.0015221118927001953, "grad_norm": 0.019056403124704958, "loss": 0.6883776783943176, "time": 0.43550407886505127, "epoch": 4, "memory": 20334, "step": 18974} +{"lr": 0.0009460482294732422, "data_time": 0.0015488624572753905, "grad_norm": 0.013475373107939958, "loss": 0.6881610572338104, "time": 0.3869898796081543, "epoch": 4, "memory": 20334, "step": 19074} +{"lr": 0.0009460482294732422, "data_time": 0.0014147520065307616, "grad_norm": 0.02170384577475488, "loss": 0.6862749814987182, "time": 0.3704317331314087, "epoch": 4, "memory": 20334, "step": 19174} +{"lr": 0.0009460482294732422, "data_time": 0.0015421390533447265, "grad_norm": 0.008144068927504123, "loss": 0.6860386788845062, "time": 0.441575026512146, "epoch": 4, "memory": 20334, "step": 19274} +{"lr": 0.0009460482294732422, "data_time": 0.001698470115661621, "grad_norm": 0.016787175508216023, "loss": 0.682527631521225, "time": 0.4142254114151001, "epoch": 4, "memory": 20334, "step": 19374} +{"lr": 0.0009460482294732422, "data_time": 0.0017727136611938477, "grad_norm": 0.018197024380788206, "loss": 0.6867428302764893, "time": 0.43469235897064207, "epoch": 4, "memory": 20334, "step": 19474} +{"lr": 0.0009460482294732422, "data_time": 0.001530003547668457, "grad_norm": 0.01373957390896976, "loss": 0.6874535799026489, "time": 0.4444230794906616, "epoch": 4, "memory": 20334, "step": 19574} +{"lr": 0.0009460482294732422, "data_time": 0.0014381647109985352, "grad_norm": 0.013393371040001512, "loss": 0.6846566259860992, "time": 0.43920044898986815, "epoch": 4, "memory": 20334, "step": 19674} +{"lr": 0.0009460482294732422, "data_time": 0.001518559455871582, "grad_norm": 0.020969915320165455, "loss": 0.6840063452720642, "time": 0.41179990768432617, "epoch": 4, "memory": 20334, "step": 19774} +{"lr": 0.0009460482294732422, "data_time": 0.0022547006607055663, "grad_norm": 0.013067789108026773, "loss": 0.684538459777832, "time": 0.4177358627319336, "epoch": 4, "memory": 20334, "step": 19874} +{"lr": 0.0009460482294732422, "data_time": 0.0016644716262817383, "grad_norm": 0.01624538628384471, "loss": 0.6869847476482391, "time": 0.40649797916412356, "epoch": 4, "memory": 20334, "step": 19974} +{"lr": 0.0009460482294732422, "data_time": 0.001677536964416504, "grad_norm": 0.020637770625762643, "loss": 0.6928331136703492, "time": 0.4554502248764038, "epoch": 4, "memory": 20334, "step": 20074} +{"lr": 0.0009460482294732422, "data_time": 0.0015866994857788087, "grad_norm": 0.020646632369607686, "loss": 0.6828271448612213, "time": 0.4271300554275513, "epoch": 4, "memory": 20334, "step": 20174} +{"lr": 0.0009460482294732422, "data_time": 0.0021440267562866213, "grad_norm": 0.014571310882456601, "loss": 0.6862425327301025, "time": 0.4084958076477051, "epoch": 4, "memory": 20334, "step": 20274} +{"lr": 0.0009460482294732422, "data_time": 0.0016492605209350586, "grad_norm": 0.021540999668650328, "loss": 0.6815922021865845, "time": 0.4297044038772583, "epoch": 4, "memory": 20334, "step": 20374} +{"lr": 0.0009460482294732422, "data_time": 0.0016182899475097657, "grad_norm": 0.02073707969393581, "loss": 0.6867773354053497, "time": 0.405154824256897, "epoch": 4, "memory": 20334, "step": 20474} +{"lr": 0.0009460482294732422, "data_time": 0.001638031005859375, "grad_norm": 0.020680293254554272, "loss": 0.6845305323600769, "time": 0.42842884063720704, "epoch": 4, "memory": 20334, "step": 20574} +{"lr": 0.0009460482294732422, "data_time": 0.001453065872192383, "grad_norm": 0.01812839702470228, "loss": 0.6863468766212464, "time": 0.503788423538208, "epoch": 4, "memory": 20334, "step": 20674} +{"lr": 0.0009460482294732422, "data_time": 0.0016042232513427735, "grad_norm": 0.017933058692142366, "loss": 0.688391637802124, "time": 0.4160242319107056, "epoch": 4, "memory": 20334, "step": 20774} +{"lr": 0.0009460482294732422, "data_time": 0.0022802114486694335, "grad_norm": 0.01980747119523585, "loss": 0.6880821645259857, "time": 0.42278614044189455, "epoch": 4, "memory": 20334, "step": 20874} +{"lr": 0.0009460482294732422, "data_time": 0.0011942863464355468, "grad_norm": 0.023618259513750673, "loss": 0.6876172184944153, "time": 0.40677585601806643, "epoch": 4, "memory": 20334, "step": 20974} +{"lr": 0.0009460482294732422, "data_time": 0.0017892122268676758, "grad_norm": 0.013236486550886184, "loss": 0.6890556931495666, "time": 0.47781803607940676, "epoch": 4, "memory": 20334, "step": 21074} +{"lr": 0.0009460482294732422, "data_time": 0.0021160602569580077, "grad_norm": 0.014703218964859844, "loss": 0.6856659710407257, "time": 0.4252608060836792, "epoch": 4, "memory": 20334, "step": 21174} +{"lr": 0.0009460482294732422, "data_time": 0.0019513845443725586, "grad_norm": 0.012840284616686404, "loss": 0.6858396232128143, "time": 0.40274677276611326, "epoch": 4, "memory": 20334, "step": 21274} +{"lr": 0.0009460482294732422, "data_time": 0.001669168472290039, "grad_norm": 0.016186735965311526, "loss": 0.6872314870357513, "time": 0.4350972890853882, "epoch": 4, "memory": 20334, "step": 21374} +{"lr": 0.0009460482294732422, "data_time": 0.0018370628356933593, "grad_norm": 0.011378039745613933, "loss": 0.6922192335128784, "time": 0.3972689390182495, "epoch": 4, "memory": 20334, "step": 21474} +{"lr": 0.0009460482294732422, "data_time": 0.0018207311630249023, "grad_norm": 0.021262407768517733, "loss": 0.6886372029781341, "time": 0.4285183668136597, "epoch": 4, "memory": 20334, "step": 21574} +{"lr": 0.0009460482294732422, "data_time": 0.0016820907592773437, "grad_norm": 0.0167135022347793, "loss": 0.6874531209468842, "time": 0.4497814655303955, "epoch": 4, "memory": 20334, "step": 21674} +{"lr": 0.0009460482294732422, "data_time": 0.0018252849578857422, "grad_norm": 0.045796560030430554, "loss": 0.6876559019088745, "time": 0.4027076005935669, "epoch": 4, "memory": 20334, "step": 21774} +{"lr": 0.0009460482294732422, "data_time": 0.0016832828521728515, "grad_norm": 0.016230330849066377, "loss": 0.6880554735660553, "time": 0.46181378364562986, "epoch": 4, "memory": 20334, "step": 21874} +{"lr": 0.0009460482294732422, "data_time": 0.001659727096557617, "grad_norm": 0.02607950558885932, "loss": 0.6868024289608001, "time": 0.4059682130813599, "epoch": 4, "memory": 20334, "step": 21974} +{"lr": 0.0009460482294732422, "data_time": 0.0020300626754760744, "grad_norm": 0.02645618636161089, "loss": 0.6936902642250061, "time": 0.49038636684417725, "epoch": 4, "memory": 20334, "step": 22074} +{"lr": 0.0009460482294732422, "data_time": 0.0020058155059814453, "grad_norm": 0.013353989459574223, "loss": 0.6905680000782013, "time": 0.3854936361312866, "epoch": 4, "memory": 20334, "step": 22174} +{"lr": 0.0009460482294732422, "data_time": 0.002733778953552246, "grad_norm": 0.019059557921718807, "loss": 0.6884543716907501, "time": 0.4084145545959473, "epoch": 4, "memory": 20334, "step": 22274} +{"lr": 0.0009460482294732422, "data_time": 0.001704859733581543, "grad_norm": 0.010697181103751063, "loss": 0.6882004916667939, "time": 0.42439606189727785, "epoch": 4, "memory": 20334, "step": 22374} +{"lr": 0.0009460482294732422, "data_time": 0.002095484733581543, "grad_norm": 0.016525850305333732, "loss": 0.6870546221733094, "time": 0.406258749961853, "epoch": 4, "memory": 20334, "step": 22474} +{"lr": 0.0009460482294732422, "data_time": 0.002139568328857422, "grad_norm": 0.01720315821003169, "loss": 0.687104606628418, "time": 0.4248741626739502, "epoch": 4, "memory": 20334, "step": 22574} +{"lr": 0.0009460482294732422, "data_time": 0.001678633689880371, "grad_norm": 0.01700702402740717, "loss": 0.6894726574420929, "time": 0.41515426635742186, "epoch": 4, "memory": 20334, "step": 22674} +{"lr": 0.0009460482294732422, "data_time": 0.001830887794494629, "grad_norm": 0.017228703980799766, "loss": 0.682729709148407, "time": 0.4291451215744019, "epoch": 4, "memory": 20334, "step": 22774} +{"lr": 0.0009460482294732422, "data_time": 0.0016281604766845703, "grad_norm": 0.016697622137144208, "loss": 0.6920957148075104, "time": 0.45853500366210936, "epoch": 4, "memory": 20334, "step": 22874} +{"lr": 0.0009460482294732422, "data_time": 0.0018860340118408204, "grad_norm": 0.016866310266777873, "loss": 0.6857346296310425, "time": 0.4406775712966919, "epoch": 4, "memory": 20334, "step": 22974} +{"accuracy/top1": 100.0, "data_time": 0.3747745354970296, "time": 1.0004582669999864, "step": 4} +{"lr": 0.0009054634122155991, "data_time": 0.0016640186309814452, "grad_norm": 0.013474358816165478, "loss": 0.6892355859279633, "time": 0.46261348724365237, "epoch": 5, "memory": 20334, "step": 23132} +{"lr": 0.0009054634122155991, "data_time": 0.0018067359924316406, "grad_norm": 0.013750681735109538, "loss": 0.6833201110363006, "time": 0.5576366901397705, "epoch": 5, "memory": 20334, "step": 23232} +{"lr": 0.0009054634122155991, "data_time": 0.0017942667007446289, "grad_norm": 0.015265438344795258, "loss": 0.6861961781978607, "time": 0.4597959041595459, "epoch": 5, "memory": 20334, "step": 23332} +{"lr": 0.0009054634122155991, "data_time": 0.0015351295471191407, "grad_norm": 0.014767507091164589, "loss": 0.6823357164859771, "time": 0.418464732170105, "epoch": 5, "memory": 20334, "step": 23432} +{"lr": 0.0009054634122155991, "data_time": 0.0015769720077514649, "grad_norm": 0.01598195082042366, "loss": 0.6872386693954468, "time": 0.482387113571167, "epoch": 5, "memory": 20334, "step": 23532} +{"lr": 0.0009054634122155991, "data_time": 0.001737356185913086, "grad_norm": 0.016936160705517976, "loss": 0.6869838953018188, "time": 0.43592736721038816, "epoch": 5, "memory": 20334, "step": 23632} +{"lr": 0.0009054634122155991, "data_time": 0.001629638671875, "grad_norm": 0.014898409554734825, "loss": 0.6889171481132508, "time": 0.4727465152740479, "epoch": 5, "memory": 20334, "step": 23732} +{"lr": 0.0009054634122155991, "data_time": 0.0016838312149047852, "grad_norm": 0.017872030846774578, "loss": 0.6852244794368744, "time": 0.39304389953613283, "epoch": 5, "memory": 20334, "step": 23832} +{"lr": 0.0009054634122155991, "data_time": 0.0018454551696777343, "grad_norm": 0.015729168988764287, "loss": 0.6876865446567535, "time": 0.45937998294830323, "epoch": 5, "memory": 20334, "step": 23932} +{"lr": 0.0009054634122155991, "data_time": 0.0013447999954223633, "grad_norm": 0.02219186588190496, "loss": 0.6884686589241028, "time": 0.49438884258270266, "epoch": 5, "memory": 20334, "step": 24032} +{"lr": 0.0009054634122155991, "data_time": 0.001679396629333496, "grad_norm": 0.017965504340827464, "loss": 0.6851427257061005, "time": 0.41814842224121096, "epoch": 5, "memory": 20334, "step": 24132} +{"lr": 0.0009054634122155991, "data_time": 0.0015549182891845704, "grad_norm": 0.020059690158814193, "loss": 0.6893116295337677, "time": 0.4728104114532471, "epoch": 5, "memory": 20334, "step": 24232} +{"lr": 0.0009054634122155991, "data_time": 0.002398848533630371, "grad_norm": 0.012723038648255169, "loss": 0.6831086933612823, "time": 0.4514086484909058, "epoch": 5, "memory": 20334, "step": 24332} +{"lr": 0.0009054634122155991, "data_time": 0.0016026258468627929, "grad_norm": 0.015413007733877748, "loss": 0.6874089002609253, "time": 0.42252435684204104, "epoch": 5, "memory": 20334, "step": 24432} +{"lr": 0.0009054634122155991, "data_time": 0.0012562036514282226, "grad_norm": 0.027014520857483147, "loss": 0.6899824917316437, "time": 0.43594205379486084, "epoch": 5, "memory": 20334, "step": 24532} +{"lr": 0.0009054634122155991, "data_time": 0.0019109487533569337, "grad_norm": 0.015215306938625873, "loss": 0.6913561344146728, "time": 0.4070683240890503, "epoch": 5, "memory": 20334, "step": 24632} +{"lr": 0.0009054634122155991, "data_time": 0.0019548654556274412, "grad_norm": 0.014206824082066305, "loss": 0.6832255184650421, "time": 0.4208178758621216, "epoch": 5, "memory": 20334, "step": 24732} +{"lr": 0.0009054634122155991, "data_time": 0.0018374204635620117, "grad_norm": 0.016956628498155622, "loss": 0.6816364228725433, "time": 0.4055312633514404, "epoch": 5, "memory": 20334, "step": 24832} +{"lr": 0.0009054634122155991, "data_time": 0.001784205436706543, "grad_norm": 0.015278805885463953, "loss": 0.6831121444702148, "time": 0.4406696319580078, "epoch": 5, "memory": 20334, "step": 24932} +{"lr": 0.0009054634122155991, "data_time": 0.001952362060546875, "grad_norm": 0.016142445389414208, "loss": 0.6892522752285004, "time": 0.41704297065734863, "epoch": 5, "memory": 20334, "step": 25032} +{"lr": 0.0009054634122155991, "data_time": 0.0023880243301391602, "grad_norm": 0.010578433435875922, "loss": 0.6865203738212585, "time": 0.41465351581573484, "epoch": 5, "memory": 20334, "step": 25132} +{"lr": 0.0009054634122155991, "data_time": 0.0017028331756591796, "grad_norm": 0.014046089618932455, "loss": 0.6822193622589111, "time": 0.3870721340179443, "epoch": 5, "memory": 20334, "step": 25232} +{"lr": 0.0009054634122155991, "data_time": 0.0024978160858154298, "grad_norm": 0.011540906503796577, "loss": 0.6844033598899841, "time": 0.48096332550048826, "epoch": 5, "memory": 20334, "step": 25332} +{"lr": 0.0009054634122155991, "data_time": 0.0022904634475708007, "grad_norm": 0.02594159790314734, "loss": 0.6852489113807678, "time": 0.4656040906906128, "epoch": 5, "memory": 20334, "step": 25432} +{"lr": 0.0009054634122155991, "data_time": 0.002164196968078613, "grad_norm": 0.013253998104482889, "loss": 0.686679071187973, "time": 0.40122935771942136, "epoch": 5, "memory": 20334, "step": 25532} +{"lr": 0.0009054634122155991, "data_time": 0.0025037527084350586, "grad_norm": 0.01694425903260708, "loss": 0.6807831466197968, "time": 0.39859178066253664, "epoch": 5, "memory": 20334, "step": 25632} +{"lr": 0.0009054634122155991, "data_time": 0.0017140865325927734, "grad_norm": 0.010635975142940879, "loss": 0.688271290063858, "time": 0.4565638780593872, "epoch": 5, "memory": 20334, "step": 25732} +{"lr": 0.0009054634122155991, "data_time": 0.0015695810317993164, "grad_norm": 0.013788261217996478, "loss": 0.6906117022037506, "time": 0.39268429279327394, "epoch": 5, "memory": 20334, "step": 25832} +{"lr": 0.0009054634122155991, "data_time": 0.0020319461822509766, "grad_norm": 0.014317479357123375, "loss": 0.6879059672355652, "time": 0.4240111351013184, "epoch": 5, "memory": 20334, "step": 25932} +{"lr": 0.0009054634122155991, "data_time": 0.002335071563720703, "grad_norm": 0.026299254782497884, "loss": 0.6909553349018097, "time": 0.49040796756744387, "epoch": 5, "memory": 20334, "step": 26032} +{"lr": 0.0009054634122155991, "data_time": 0.0017626047134399413, "grad_norm": 0.019424308091402054, "loss": 0.6844892084598542, "time": 0.4185448169708252, "epoch": 5, "memory": 20334, "step": 26132} +{"lr": 0.0009054634122155991, "data_time": 0.0016578197479248046, "grad_norm": 0.02198729747906327, "loss": 0.6875029325485229, "time": 0.44490096569061277, "epoch": 5, "memory": 20334, "step": 26232} +{"lr": 0.0009054634122155991, "data_time": 0.0014395952224731446, "grad_norm": 0.026266485848464072, "loss": 0.6841505408287049, "time": 0.3759397745132446, "epoch": 5, "memory": 20334, "step": 26332} +{"lr": 0.0009054634122155991, "data_time": 0.0014629602432250977, "grad_norm": 0.01955609628930688, "loss": 0.6860485792160034, "time": 0.5077569246292114, "epoch": 5, "memory": 20334, "step": 26432} +{"lr": 0.0009054634122155991, "data_time": 0.0021701335906982424, "grad_norm": 0.01715301611693576, "loss": 0.686602920293808, "time": 0.5131395101547241, "epoch": 5, "memory": 20334, "step": 26532} +{"lr": 0.0009054634122155991, "data_time": 0.001564621925354004, "grad_norm": 0.009940954123158008, "loss": 0.6881152570247651, "time": 0.5038388490676879, "epoch": 5, "memory": 20334, "step": 26632} +{"lr": 0.0009054634122155991, "data_time": 0.0022216081619262696, "grad_norm": 0.016357232863083482, "loss": 0.6874220490455627, "time": 0.4277378559112549, "epoch": 5, "memory": 20334, "step": 26732} +{"lr": 0.0009054634122155991, "data_time": 0.0025137901306152345, "grad_norm": 0.019747086614370347, "loss": 0.6891478955745697, "time": 0.4235680103302002, "epoch": 5, "memory": 20334, "step": 26832} +{"lr": 0.0009054634122155991, "data_time": 0.0020124197006225588, "grad_norm": 0.016667066141963005, "loss": 0.683452981710434, "time": 0.4664349317550659, "epoch": 5, "memory": 20334, "step": 26932} +{"lr": 0.0009054634122155991, "data_time": 0.002366971969604492, "grad_norm": 0.01884008216438815, "loss": 0.6852039217948913, "time": 0.39306495189666746, "epoch": 5, "memory": 20334, "step": 27032} +{"lr": 0.0009054634122155991, "data_time": 0.0014747142791748046, "grad_norm": 0.013768781418912112, "loss": 0.6881043314933777, "time": 0.38590786457061765, "epoch": 5, "memory": 20334, "step": 27132} +{"lr": 0.0009054634122155991, "data_time": 0.0015051841735839843, "grad_norm": 0.012897312315180898, "loss": 0.690074110031128, "time": 0.4127305030822754, "epoch": 5, "memory": 20334, "step": 27232} +{"lr": 0.0009054634122155991, "data_time": 0.0016739368438720703, "grad_norm": 0.016327945189550518, "loss": 0.6870486497879028, "time": 0.4177191972732544, "epoch": 5, "memory": 20334, "step": 27332} +{"lr": 0.0009054634122155991, "data_time": 0.0015619516372680664, "grad_norm": 0.02242490865755826, "loss": 0.6888010859489441, "time": 0.40947961807250977, "epoch": 5, "memory": 20334, "step": 27432} +{"lr": 0.0009054634122155991, "data_time": 0.0016671419143676758, "grad_norm": 0.01205872044665739, "loss": 0.6856835782527924, "time": 0.44570968151092527, "epoch": 5, "memory": 20334, "step": 27532} +{"lr": 0.0009054634122155991, "data_time": 0.0016495227813720704, "grad_norm": 0.01477744576986879, "loss": 0.6921510756015777, "time": 0.5403522729873658, "epoch": 5, "memory": 20334, "step": 27632} +{"lr": 0.0009054634122155991, "data_time": 0.002578067779541016, "grad_norm": 0.024773371312767266, "loss": 0.6809918463230134, "time": 0.40028655529022217, "epoch": 5, "memory": 20334, "step": 27732} +{"lr": 0.0009054634122155991, "data_time": 0.0025525808334350584, "grad_norm": 0.022941923746839166, "loss": 0.6822918355464935, "time": 0.4442912101745605, "epoch": 5, "memory": 20334, "step": 27832} +{"lr": 0.0009054634122155991, "data_time": 0.0022899866104125976, "grad_norm": 0.01740312119945884, "loss": 0.6865120649337768, "time": 0.45723905563354494, "epoch": 5, "memory": 20334, "step": 27932} +{"lr": 0.0009054634122155991, "data_time": 0.002504301071166992, "grad_norm": 0.01854991556610912, "loss": 0.6900014460086823, "time": 0.42259061336517334, "epoch": 5, "memory": 20334, "step": 28032} +{"lr": 0.0009054634122155991, "data_time": 0.0023766517639160155, "grad_norm": 0.00993919875472784, "loss": 0.6916679263114929, "time": 0.4421380996704102, "epoch": 5, "memory": 20334, "step": 28132} +{"lr": 0.0009054634122155991, "data_time": 0.0023939847946166993, "grad_norm": 0.014894899516366421, "loss": 0.688961535692215, "time": 0.4703490972518921, "epoch": 5, "memory": 20334, "step": 28232} +{"lr": 0.0009054634122155991, "data_time": 0.002553510665893555, "grad_norm": 0.018669087439775467, "loss": 0.6927083015441895, "time": 0.4199769735336304, "epoch": 5, "memory": 20334, "step": 28332} +{"lr": 0.0009054634122155991, "data_time": 0.00251924991607666, "grad_norm": 0.01581809011986479, "loss": 0.6880093038082122, "time": 0.4258387804031372, "epoch": 5, "memory": 20334, "step": 28432} +{"lr": 0.0009054634122155991, "data_time": 0.0022903203964233397, "grad_norm": 0.019251885660924017, "loss": 0.6875609874725341, "time": 0.4850937843322754, "epoch": 5, "memory": 20334, "step": 28532} +{"lr": 0.0009054634122155991, "data_time": 0.001758575439453125, "grad_norm": 0.020095613319426775, "loss": 0.6832868814468384, "time": 0.45992026329040525, "epoch": 5, "memory": 20334, "step": 28632} +{"lr": 0.0009054634122155991, "data_time": 0.0021857500076293947, "grad_norm": 0.017635805904865264, "loss": 0.6856019020080566, "time": 0.43054590225219724, "epoch": 5, "memory": 20334, "step": 28732} +{"accuracy/top1": 100.0, "data_time": 0.3672407203250461, "time": 0.9940706094106039, "step": 5} +{"lr": 0.0008550178566873411, "data_time": 0.0016501426696777343, "grad_norm": 0.010740619990974665, "loss": 0.6819038808345794, "time": 0.4215155839920044, "epoch": 6, "memory": 20334, "step": 28890} +{"lr": 0.0008550178566873411, "data_time": 0.0017414093017578125, "grad_norm": 0.014249167079105974, "loss": 0.6858422577381134, "time": 0.4054927110671997, "epoch": 6, "memory": 20334, "step": 28990} +{"lr": 0.0008550178566873411, "data_time": 0.002213859558105469, "grad_norm": 0.018390496261417866, "loss": 0.6852671027183532, "time": 0.40395450592041016, "epoch": 6, "memory": 20334, "step": 29090} +{"lr": 0.0008550178566873411, "data_time": 0.00170443058013916, "grad_norm": 0.017776680213864894, "loss": 0.6879237771034241, "time": 0.38143167495727537, "epoch": 6, "memory": 20334, "step": 29190} +{"lr": 0.0008550178566873411, "data_time": 0.002321171760559082, "grad_norm": 0.017277160147204994, "loss": 0.6869894623756408, "time": 0.4530991315841675, "epoch": 6, "memory": 20334, "step": 29290} +{"lr": 0.0008550178566873411, "data_time": 0.0020491361618041994, "grad_norm": 0.017743013601284476, "loss": 0.6863463342189788, "time": 0.38599770069122313, "epoch": 6, "memory": 20334, "step": 29390} +{"lr": 0.0008550178566873411, "data_time": 0.0020192146301269533, "grad_norm": 0.017861474899109452, "loss": 0.6891350388526917, "time": 0.42172553539276125, "epoch": 6, "memory": 20334, "step": 29490} +{"lr": 0.0008550178566873411, "data_time": 0.0017431259155273437, "grad_norm": 0.01864182265708223, "loss": 0.685890394449234, "time": 0.3811058044433594, "epoch": 6, "memory": 20334, "step": 29590} +{"lr": 0.0008550178566873411, "data_time": 0.0030803918838500977, "grad_norm": 0.015005982713773847, "loss": 0.6859933257102966, "time": 0.43090860843658446, "epoch": 6, "memory": 20334, "step": 29690} +{"lr": 0.0008550178566873411, "data_time": 0.00235443115234375, "grad_norm": 0.015143247367814184, "loss": 0.6870425939559937, "time": 0.45867719650268557, "epoch": 6, "memory": 20334, "step": 29790} +{"lr": 0.0008550178566873411, "data_time": 0.002558255195617676, "grad_norm": 0.021923623932525515, "loss": 0.686844676733017, "time": 0.4720849275588989, "epoch": 6, "memory": 20334, "step": 29890} +{"lr": 0.0008550178566873411, "data_time": 0.003036165237426758, "grad_norm": 0.018959448114037512, "loss": 0.6872752964496612, "time": 0.42896265983581544, "epoch": 6, "memory": 20334, "step": 29990} +{"lr": 0.0008550178566873411, "data_time": 0.002033638954162598, "grad_norm": 0.014010517811402678, "loss": 0.6890656590461731, "time": 0.41484718322753905, "epoch": 6, "memory": 20334, "step": 30090} +{"lr": 0.0008550178566873411, "data_time": 0.0024161577224731446, "grad_norm": 0.012129091611132025, "loss": 0.689091432094574, "time": 0.36630136966705323, "epoch": 6, "memory": 20334, "step": 30190} +{"lr": 0.0008550178566873411, "data_time": 0.002329564094543457, "grad_norm": 0.01896277009509504, "loss": 0.6904054403305053, "time": 0.4629948616027832, "epoch": 6, "memory": 20334, "step": 30290} +{"lr": 0.0008550178566873411, "data_time": 0.002977895736694336, "grad_norm": 0.014263168442994356, "loss": 0.6873790085315704, "time": 0.43889336585998534, "epoch": 6, "memory": 20334, "step": 30390} +{"lr": 0.0008550178566873411, "data_time": 0.0018604755401611327, "grad_norm": 0.018118901655543595, "loss": 0.6866663873195649, "time": 0.3718499898910522, "epoch": 6, "memory": 20334, "step": 30490} +{"lr": 0.0008550178566873411, "data_time": 0.0017019271850585937, "grad_norm": 0.016365794534794986, "loss": 0.6880724191665649, "time": 0.3911444187164307, "epoch": 6, "memory": 20334, "step": 30590} +{"lr": 0.0008550178566873411, "data_time": 0.0026001930236816406, "grad_norm": 0.01800455499906093, "loss": 0.6916402876377106, "time": 0.41898887157440184, "epoch": 6, "memory": 20334, "step": 30690} +{"lr": 0.0008550178566873411, "data_time": 0.0021966218948364256, "grad_norm": 0.019803097785916178, "loss": 0.6882297456264496, "time": 0.3862194776535034, "epoch": 6, "memory": 20334, "step": 30790} +{"lr": 0.0008550178566873411, "data_time": 0.003175950050354004, "grad_norm": 0.014639736915705725, "loss": 0.6861068546772003, "time": 0.4259798049926758, "epoch": 6, "memory": 20334, "step": 30890} +{"lr": 0.0008550178566873411, "data_time": 0.00249025821685791, "grad_norm": 0.0161762215255294, "loss": 0.6831121385097504, "time": 0.4316707134246826, "epoch": 6, "memory": 20334, "step": 30990} +{"lr": 0.0008550178566873411, "data_time": 0.0024130821228027345, "grad_norm": 0.013457054318860174, "loss": 0.6833941161632537, "time": 0.37877702713012695, "epoch": 6, "memory": 20334, "step": 31090} +{"lr": 0.0008550178566873411, "data_time": 0.0026801347732543944, "grad_norm": 0.01865406068973243, "loss": 0.6831690788269043, "time": 0.49864356517791747, "epoch": 6, "memory": 20334, "step": 31190} +{"lr": 0.0008550178566873411, "data_time": 0.0023834228515625, "grad_norm": 0.016033722483552993, "loss": 0.6868420302867889, "time": 0.4378886461257935, "epoch": 6, "memory": 20334, "step": 31290} +{"lr": 0.0008550178566873411, "data_time": 0.0024738073348999023, "grad_norm": 0.015218574134632946, "loss": 0.6876056492328644, "time": 0.42201330661773684, "epoch": 6, "memory": 20334, "step": 31390} +{"lr": 0.0008550178566873411, "data_time": 0.0022174835205078123, "grad_norm": 0.015551392734050751, "loss": 0.6883639395236969, "time": 0.3924338102340698, "epoch": 6, "memory": 20334, "step": 31490} +{"lr": 0.0008550178566873411, "data_time": 0.0023947238922119142, "grad_norm": 0.014032244007103145, "loss": 0.6840821266174316, "time": 0.3723433017730713, "epoch": 6, "memory": 20334, "step": 31590} +{"lr": 0.0008550178566873411, "data_time": 0.002815508842468262, "grad_norm": 0.024828650563722476, "loss": 0.6863940179347991, "time": 0.4717421054840088, "epoch": 6, "memory": 20334, "step": 31690} +{"lr": 0.0008550178566873411, "data_time": 0.0021286725997924803, "grad_norm": 0.014141946626477876, "loss": 0.6857439279556274, "time": 0.40163042545318606, "epoch": 6, "memory": 20334, "step": 31790} +{"lr": 0.0008550178566873411, "data_time": 0.0019434213638305664, "grad_norm": 0.02141375997744035, "loss": 0.6920516550540924, "time": 0.40711610317230223, "epoch": 6, "memory": 20334, "step": 31890} +{"lr": 0.0008550178566873411, "data_time": 0.0025219440460205076, "grad_norm": 0.012545603339094668, "loss": 0.6926559031009674, "time": 0.4218295097351074, "epoch": 6, "memory": 20334, "step": 31990} +{"lr": 0.0008550178566873411, "data_time": 0.002497553825378418, "grad_norm": 0.019899757322855295, "loss": 0.682876181602478, "time": 0.38234996795654297, "epoch": 6, "memory": 20334, "step": 32090} +{"lr": 0.0008550178566873411, "data_time": 0.0017943382263183594, "grad_norm": 0.015066596562974155, "loss": 0.689560842514038, "time": 0.4505463123321533, "epoch": 6, "memory": 20334, "step": 32190} +{"lr": 0.0008550178566873411, "data_time": 0.002192401885986328, "grad_norm": 0.01855681617744267, "loss": 0.6845539927482605, "time": 0.4410197019577026, "epoch": 6, "memory": 20334, "step": 32290} +{"lr": 0.0008550178566873411, "data_time": 0.0016761541366577149, "grad_norm": 0.013517347583547234, "loss": 0.6895796895027161, "time": 0.37102541923522947, "epoch": 6, "memory": 20334, "step": 32390} +{"lr": 0.0008550178566873411, "data_time": 0.0021207571029663087, "grad_norm": 0.022510111820884048, "loss": 0.6930544078350067, "time": 0.4105675458908081, "epoch": 6, "memory": 20334, "step": 32490} +{"lr": 0.0008550178566873411, "data_time": 0.0024774789810180662, "grad_norm": 0.013739681034348906, "loss": 0.6867192447185516, "time": 0.40883698463439944, "epoch": 6, "memory": 20334, "step": 32590} +{"lr": 0.0008550178566873411, "data_time": 0.0016768455505371093, "grad_norm": 0.02028878660639748, "loss": 0.6895137846469879, "time": 0.41194050312042235, "epoch": 6, "memory": 20334, "step": 32690} +{"lr": 0.0008550178566873411, "data_time": 0.0028486013412475585, "grad_norm": 0.017298220843076705, "loss": 0.6861627876758576, "time": 0.3971180200576782, "epoch": 6, "memory": 20334, "step": 32790} +{"lr": 0.0008550178566873411, "data_time": 0.001663064956665039, "grad_norm": 0.0246057340875268, "loss": 0.6876162230968476, "time": 0.3676790952682495, "epoch": 6, "memory": 20334, "step": 32890} +{"lr": 0.0008550178566873411, "data_time": 0.0016487598419189452, "grad_norm": 0.01762840843293816, "loss": 0.6918885767459869, "time": 0.43290071487426757, "epoch": 6, "memory": 20334, "step": 32990} +{"lr": 0.0008550178566873411, "data_time": 0.0028705835342407227, "grad_norm": 0.01391700147651136, "loss": 0.6933708608150482, "time": 0.40606689453125, "epoch": 6, "memory": 20334, "step": 33090} +{"lr": 0.0008550178566873411, "data_time": 0.0019050836563110352, "grad_norm": 0.012038194620981813, "loss": 0.6841579973697662, "time": 0.404231858253479, "epoch": 6, "memory": 20334, "step": 33190} +{"lr": 0.0008550178566873411, "data_time": 0.0018760204315185548, "grad_norm": 0.017551461001858115, "loss": 0.6854634702205658, "time": 0.38661041259765627, "epoch": 6, "memory": 20334, "step": 33290} +{"lr": 0.0008550178566873411, "data_time": 0.0020201444625854493, "grad_norm": 0.01785781476646662, "loss": 0.6892797350883484, "time": 0.3988152027130127, "epoch": 6, "memory": 20334, "step": 33390} +{"lr": 0.0008550178566873411, "data_time": 0.002326774597167969, "grad_norm": 0.01143301121192053, "loss": 0.6911220073699951, "time": 0.4070104122161865, "epoch": 6, "memory": 20334, "step": 33490} +{"lr": 0.0008550178566873411, "data_time": 0.002182292938232422, "grad_norm": 0.014409385481849312, "loss": 0.6864995062351227, "time": 0.3734931230545044, "epoch": 6, "memory": 20334, "step": 33590} +{"lr": 0.0008550178566873411, "data_time": 0.002413010597229004, "grad_norm": 0.018322595208883286, "loss": 0.6882865190505981, "time": 0.39390385150909424, "epoch": 6, "memory": 20334, "step": 33690} +{"lr": 0.0008550178566873411, "data_time": 0.0022130489349365236, "grad_norm": 0.0244600351434201, "loss": 0.6902685821056366, "time": 0.46015329360961915, "epoch": 6, "memory": 20334, "step": 33790} +{"lr": 0.0008550178566873411, "data_time": 0.002182483673095703, "grad_norm": 0.025804482540115713, "loss": 0.6844572007656098, "time": 0.42754712104797366, "epoch": 6, "memory": 20334, "step": 33890} +{"lr": 0.0008550178566873411, "data_time": 0.002279186248779297, "grad_norm": 0.015815614216262475, "loss": 0.6855115056037903, "time": 0.4512783527374268, "epoch": 6, "memory": 20334, "step": 33990} +{"lr": 0.0008550178566873411, "data_time": 0.0018991470336914063, "grad_norm": 0.01853583964984864, "loss": 0.6862693428993225, "time": 0.4971250057220459, "epoch": 6, "memory": 20334, "step": 34090} +{"lr": 0.0008550178566873411, "data_time": 0.0023725271224975587, "grad_norm": 0.011092688061762602, "loss": 0.6905861556529999, "time": 0.4244656562805176, "epoch": 6, "memory": 20334, "step": 34190} +{"lr": 0.0008550178566873411, "data_time": 0.0028261423110961916, "grad_norm": 0.016301624092739074, "loss": 0.6868258774280548, "time": 0.400516939163208, "epoch": 6, "memory": 20334, "step": 34290} +{"lr": 0.0008550178566873411, "data_time": 0.0018138408660888672, "grad_norm": 0.01195098765892908, "loss": 0.6848946690559388, "time": 0.5971655130386353, "epoch": 6, "memory": 20334, "step": 34390} +{"lr": 0.0008550178566873411, "data_time": 0.0017888784408569337, "grad_norm": 0.009971192036755384, "loss": 0.6855127394199372, "time": 0.36561877727508546, "epoch": 6, "memory": 20334, "step": 34490} +{"accuracy/top1": 100.0, "data_time": 0.37670813666449654, "time": 1.0326961941189237, "step": 6} +{"lr": 0.0007959536998847743, "data_time": 0.002660822868347168, "grad_norm": 0.008818201615940779, "loss": 0.686515724658966, "time": 0.3914986848831177, "epoch": 7, "memory": 20334, "step": 34648} +{"lr": 0.0007959536998847743, "data_time": 0.002326178550720215, "grad_norm": 0.017777158366516232, "loss": 0.6882440745830536, "time": 0.39154982566833496, "epoch": 7, "memory": 20334, "step": 34748} +{"lr": 0.0007959536998847743, "data_time": 0.0028423070907592773, "grad_norm": 0.017145266756415367, "loss": 0.6869498729705811, "time": 0.4006332874298096, "epoch": 7, "memory": 20334, "step": 34848} +{"lr": 0.0007959536998847743, "data_time": 0.0026952028274536133, "grad_norm": 0.020782969333231448, "loss": 0.6846798479557037, "time": 0.48883905410766604, "epoch": 7, "memory": 20334, "step": 34948} +{"lr": 0.0007959536998847743, "data_time": 0.002477884292602539, "grad_norm": 0.015397327730897813, "loss": 0.6860618412494659, "time": 0.42564547061920166, "epoch": 7, "memory": 20334, "step": 35048} +{"lr": 0.0007959536998847743, "data_time": 0.0030760765075683594, "grad_norm": 0.017429115390405057, "loss": 0.6871620416641235, "time": 0.42325582504272463, "epoch": 7, "memory": 20334, "step": 35148} +{"lr": 0.0007959536998847743, "data_time": 0.001963019371032715, "grad_norm": 0.014444156154058873, "loss": 0.6850408792495728, "time": 0.4025650262832642, "epoch": 7, "memory": 20334, "step": 35248} +{"lr": 0.0007959536998847743, "data_time": 0.0023630380630493162, "grad_norm": 0.012954378291033209, "loss": 0.6875571370124817, "time": 1.0924967527389526, "epoch": 7, "memory": 20334, "step": 35348} +{"lr": 0.0007959536998847743, "data_time": 0.0024686098098754884, "grad_norm": 0.01838813836220652, "loss": 0.6845176458358765, "time": 0.3736931085586548, "epoch": 7, "memory": 20334, "step": 35448} +{"lr": 0.0007959536998847743, "data_time": 0.002348494529724121, "grad_norm": 0.025828354945406317, "loss": 0.6839582920074463, "time": 0.3978057861328125, "epoch": 7, "memory": 20334, "step": 35548} +{"lr": 0.0007959536998847743, "data_time": 0.0025232315063476564, "grad_norm": 0.018405025405809284, "loss": 0.6846412360668183, "time": 0.38733506202697754, "epoch": 7, "memory": 20334, "step": 35648} +{"lr": 0.0007959536998847743, "data_time": 0.0027555227279663086, "grad_norm": 0.024616753216832876, "loss": 0.6853302597999573, "time": 0.4038097858428955, "epoch": 7, "memory": 20334, "step": 35748} +{"lr": 0.0007959536998847743, "data_time": 0.0018779277801513673, "grad_norm": 0.01780391177162528, "loss": 0.6869785249233246, "time": 0.43698692321777344, "epoch": 7, "memory": 20334, "step": 35848} +{"lr": 0.0007959536998847743, "data_time": 0.0030397653579711916, "grad_norm": 0.024214513879269362, "loss": 0.6843997478485108, "time": 0.39369449615478513, "epoch": 7, "memory": 20334, "step": 35948} +{"lr": 0.0007959536998847743, "data_time": 0.0019219636917114258, "grad_norm": 0.008677515428280458, "loss": 0.6863185703754425, "time": 0.41775286197662354, "epoch": 7, "memory": 20334, "step": 36048} +{"lr": 0.0007959536998847743, "data_time": 0.0021116256713867186, "grad_norm": 0.016486375685781242, "loss": 0.6887442231178283, "time": 0.40992734432220457, "epoch": 7, "memory": 20334, "step": 36148} +{"lr": 0.0007959536998847743, "data_time": 0.0025710821151733398, "grad_norm": 0.019156168890185656, "loss": 0.6890320301055908, "time": 0.40306859016418456, "epoch": 7, "memory": 20334, "step": 36248} +{"lr": 0.0007959536998847743, "data_time": 0.0024237871170043946, "grad_norm": 0.01392492586746812, "loss": 0.6847045421600342, "time": 0.39427847862243653, "epoch": 7, "memory": 20334, "step": 36348} +{"lr": 0.0007959536998847743, "data_time": 0.0023135900497436523, "grad_norm": 0.024465126590803264, "loss": 0.6876348078250885, "time": 0.4035454511642456, "epoch": 7, "memory": 20334, "step": 36448} +{"lr": 0.0007959536998847743, "data_time": 0.002964615821838379, "grad_norm": 0.01351309111341834, "loss": 0.6860770404338836, "time": 0.4082977294921875, "epoch": 7, "memory": 20334, "step": 36548} +{"lr": 0.0007959536998847743, "data_time": 0.0023701667785644533, "grad_norm": 0.01681348168058321, "loss": 0.6863355100154876, "time": 0.36402928829193115, "epoch": 7, "memory": 20334, "step": 36648} +{"lr": 0.0007959536998847743, "data_time": 0.0023248910903930662, "grad_norm": 0.014404264773475006, "loss": 0.6853826642036438, "time": 0.40149860382080077, "epoch": 7, "memory": 20334, "step": 36748} +{"lr": 0.0007959536998847743, "data_time": 0.0020804405212402344, "grad_norm": 0.0163409007829614, "loss": 0.6862274885177613, "time": 0.3629596471786499, "epoch": 7, "memory": 20334, "step": 36848} +{"lr": 0.0007959536998847743, "data_time": 0.0017332315444946289, "grad_norm": 0.01984357591718435, "loss": 0.6822782814502716, "time": 0.3995716333389282, "epoch": 7, "memory": 20334, "step": 36948} +{"lr": 0.0007959536998847743, "data_time": 0.0022834300994873046, "grad_norm": 0.021805821312591435, "loss": 0.6903250575065613, "time": 0.4455390214920044, "epoch": 7, "memory": 20334, "step": 37048} +{"lr": 0.0007959536998847743, "data_time": 0.0025606870651245115, "grad_norm": 0.01806363833602518, "loss": 0.6891472280025482, "time": 0.4206372261047363, "epoch": 7, "memory": 20334, "step": 37148} +{"lr": 0.0007959536998847743, "data_time": 0.0016620397567749024, "grad_norm": 0.027936227293685077, "loss": 0.6872438967227936, "time": 0.38329253196716306, "epoch": 7, "memory": 20334, "step": 37248} +{"lr": 0.0007959536998847743, "data_time": 0.0017488479614257812, "grad_norm": 0.013940639398060738, "loss": 0.6853080928325653, "time": 0.4037781238555908, "epoch": 7, "memory": 20334, "step": 37348} +{"lr": 0.0007959536998847743, "data_time": 0.0024611949920654297, "grad_norm": 0.011252750316634775, "loss": 0.6853863060474396, "time": 0.39341936111450193, "epoch": 7, "memory": 20334, "step": 37448} +{"lr": 0.0007959536998847743, "data_time": 0.002257633209228516, "grad_norm": 0.01710908564273268, "loss": 0.6870525062084198, "time": 0.40363810062408445, "epoch": 7, "memory": 20334, "step": 37548} +{"lr": 0.0007959536998847743, "data_time": 0.0017189979553222656, "grad_norm": 0.021097694453783334, "loss": 0.6861779630184174, "time": 0.45919251441955566, "epoch": 7, "memory": 20334, "step": 37648} +{"lr": 0.0007959536998847743, "data_time": 0.0028703927993774412, "grad_norm": 0.018132218648679553, "loss": 0.691362488269806, "time": 0.38237967491149905, "epoch": 7, "memory": 20334, "step": 37748} +{"lr": 0.0007959536998847743, "data_time": 0.0025453090667724608, "grad_norm": 0.013848072150722146, "loss": 0.6873121321201324, "time": 0.4048325061798096, "epoch": 7, "memory": 20334, "step": 37848} +{"lr": 0.0007959536998847743, "data_time": 0.00212557315826416, "grad_norm": 0.0176771626342088, "loss": 0.6882309913635254, "time": 0.36730663776397704, "epoch": 7, "memory": 20334, "step": 37948} +{"lr": 0.0007959536998847743, "data_time": 0.0025019407272338866, "grad_norm": 0.01385203015524894, "loss": 0.6901014924049378, "time": 0.40636818408966063, "epoch": 7, "memory": 20334, "step": 38048} +{"lr": 0.0007959536998847743, "data_time": 0.0016633272171020508, "grad_norm": 0.01257945482066134, "loss": 0.6852293372154236, "time": 0.4259000539779663, "epoch": 7, "memory": 20334, "step": 38148} +{"lr": 0.0007959536998847743, "data_time": 0.0025504350662231444, "grad_norm": 0.01858710205415264, "loss": 0.6862414479255676, "time": 0.4188873052597046, "epoch": 7, "memory": 20334, "step": 38248} +{"lr": 0.0007959536998847743, "data_time": 0.0014546871185302734, "grad_norm": 0.021473313542082907, "loss": 0.6916547000408173, "time": 0.38711209297180177, "epoch": 7, "memory": 20334, "step": 38348} +{"lr": 0.0007959536998847743, "data_time": 0.0015260934829711913, "grad_norm": 0.02340732077136636, "loss": 0.6864424705505371, "time": 0.45596649646759035, "epoch": 7, "memory": 20334, "step": 38448} +{"lr": 0.0007959536998847743, "data_time": 0.001941514015197754, "grad_norm": 0.015903096972033383, "loss": 0.6854012429714202, "time": 0.3989575386047363, "epoch": 7, "memory": 20334, "step": 38548} +{"lr": 0.0007959536998847743, "data_time": 0.0016055583953857421, "grad_norm": 0.016300834156572817, "loss": 0.6881219208240509, "time": 0.39681453704833985, "epoch": 7, "memory": 20334, "step": 38648} +{"lr": 0.0007959536998847743, "data_time": 0.0015841484069824218, "grad_norm": 0.0161306019872427, "loss": 0.6845422387123108, "time": 0.40819635391235354, "epoch": 7, "memory": 20334, "step": 38748} +{"lr": 0.0007959536998847743, "data_time": 0.001938605308532715, "grad_norm": 0.01806643467862159, "loss": 0.687273907661438, "time": 0.3830427885055542, "epoch": 7, "memory": 20334, "step": 38848} +{"lr": 0.0007959536998847743, "data_time": 0.0025661230087280274, "grad_norm": 0.014198095246683806, "loss": 0.6872689723968506, "time": 0.4111591577529907, "epoch": 7, "memory": 20334, "step": 38948} +{"lr": 0.0007959536998847743, "data_time": 0.0017806529998779298, "grad_norm": 0.02024270365945995, "loss": 0.6882555246353149, "time": 0.39887540340423583, "epoch": 7, "memory": 20334, "step": 39048} +{"lr": 0.0007959536998847743, "data_time": 0.0017499685287475585, "grad_norm": 0.010240830824477598, "loss": 0.6851606249809266, "time": 0.41409270763397216, "epoch": 7, "memory": 20334, "step": 39148} +{"lr": 0.0007959536998847743, "data_time": 0.0018649816513061524, "grad_norm": 0.013648881413973867, "loss": 0.6881942272186279, "time": 0.5179283618927002, "epoch": 7, "memory": 20334, "step": 39248} +{"lr": 0.0007959536998847743, "data_time": 0.00165557861328125, "grad_norm": 0.014657158497720956, "loss": 0.6872871160507202, "time": 0.4319024085998535, "epoch": 7, "memory": 20334, "step": 39348} +{"lr": 0.0007959536998847743, "data_time": 0.0017303466796875, "grad_norm": 0.012619608407840132, "loss": 0.687632167339325, "time": 0.3954183340072632, "epoch": 7, "memory": 20334, "step": 39448} +{"lr": 0.0007959536998847743, "data_time": 0.0016219139099121094, "grad_norm": 0.018782898550853133, "loss": 0.6843981087207794, "time": 0.3781104564666748, "epoch": 7, "memory": 20334, "step": 39548} +{"lr": 0.0007959536998847743, "data_time": 0.0017826557159423828, "grad_norm": 0.011858153762295843, "loss": 0.6843623101711274, "time": 0.4297770977020264, "epoch": 7, "memory": 20334, "step": 39648} +{"lr": 0.0007959536998847743, "data_time": 0.001736903190612793, "grad_norm": 0.013285747787449509, "loss": 0.6878089725971221, "time": 0.4400207757949829, "epoch": 7, "memory": 20334, "step": 39748} +{"lr": 0.0007959536998847743, "data_time": 0.001980900764465332, "grad_norm": 0.0076624172565061596, "loss": 0.6867228329181672, "time": 0.4161480665206909, "epoch": 7, "memory": 20334, "step": 39848} +{"lr": 0.0007959536998847743, "data_time": 0.0017470359802246095, "grad_norm": 0.016743032447993755, "loss": 0.6832575142383576, "time": 0.4377040147781372, "epoch": 7, "memory": 20334, "step": 39948} +{"lr": 0.0007959536998847743, "data_time": 0.002133679389953613, "grad_norm": 0.01983685716986656, "loss": 0.692478609085083, "time": 0.37770495414733884, "epoch": 7, "memory": 20334, "step": 40048} +{"lr": 0.0007959536998847743, "data_time": 0.0016900062561035155, "grad_norm": 0.020389390457421542, "loss": 0.6839831888675689, "time": 0.4018282175064087, "epoch": 7, "memory": 20334, "step": 40148} +{"lr": 0.0007959536998847743, "data_time": 0.0020685195922851562, "grad_norm": 0.014187499857507646, "loss": 0.6850757956504822, "time": 0.3764138460159302, "epoch": 7, "memory": 20334, "step": 40248} +{"accuracy/top1": 100.0, "data_time": 0.3704343107011583, "time": 1.011265966627333, "step": 7} +{"lr": 0.0007297252973710758, "data_time": 0.0014069557189941406, "grad_norm": 0.01504069536458701, "loss": 0.6872755467891694, "time": 0.4223238468170166, "epoch": 8, "memory": 20334, "step": 40406} +{"lr": 0.0007297252973710758, "data_time": 0.001478719711303711, "grad_norm": 0.01260184432612732, "loss": 0.6861163735389709, "time": 0.39136905670166017, "epoch": 8, "memory": 20334, "step": 40506} +{"lr": 0.0007297252973710758, "data_time": 0.0018114566802978516, "grad_norm": 0.021287666726857425, "loss": 0.6850366771221161, "time": 0.4301972627639771, "epoch": 8, "memory": 20334, "step": 40606} +{"lr": 0.0007297252973710758, "data_time": 0.0015496492385864257, "grad_norm": 0.02114067629445344, "loss": 0.6902939260005951, "time": 0.40356786251068116, "epoch": 8, "memory": 20334, "step": 40706} +{"lr": 0.0007297252973710758, "data_time": 0.0015987396240234376, "grad_norm": 0.014086356712505221, "loss": 0.6874367773532868, "time": 0.4276456356048584, "epoch": 8, "memory": 20334, "step": 40806} +{"lr": 0.0007297252973710758, "data_time": 0.0014924287796020507, "grad_norm": 0.017130320169962944, "loss": 0.6926200211048126, "time": 0.3991257905960083, "epoch": 8, "memory": 20334, "step": 40906} +{"lr": 0.0007297252973710758, "data_time": 0.0015770435333251954, "grad_norm": 0.01609284762525931, "loss": 0.6854384422302247, "time": 0.46119678020477295, "epoch": 8, "memory": 20334, "step": 41006} +{"lr": 0.0007297252973710758, "data_time": 0.0014980554580688477, "grad_norm": 0.019533733511343597, "loss": 0.6853923678398133, "time": 0.39627504348754883, "epoch": 8, "memory": 20334, "step": 41106} +{"lr": 0.0007297252973710758, "data_time": 0.0015324115753173827, "grad_norm": 0.009302004193887114, "loss": 0.6858465254306794, "time": 0.4091492652893066, "epoch": 8, "memory": 20334, "step": 41206} +{"lr": 0.0007297252973710758, "data_time": 0.0014226436614990234, "grad_norm": 0.017282978142611684, "loss": 0.6882459640502929, "time": 0.39461569786071776, "epoch": 8, "memory": 20334, "step": 41306} +{"lr": 0.0007297252973710758, "data_time": 0.0015200138092041015, "grad_norm": 0.02055193189298734, "loss": 0.6900627791881562, "time": 0.3819331884384155, "epoch": 8, "memory": 20334, "step": 41406} +{"lr": 0.0007297252973710758, "data_time": 0.001488494873046875, "grad_norm": 0.013897991087287665, "loss": 0.6842069685459137, "time": 0.3753032922744751, "epoch": 8, "memory": 20334, "step": 41506} +{"lr": 0.0007297252973710758, "data_time": 0.0014233589172363281, "grad_norm": 0.020723226061090828, "loss": 0.6847513914108276, "time": 0.3833609104156494, "epoch": 8, "memory": 20334, "step": 41606} +{"lr": 0.0007297252973710758, "data_time": 0.0015473127365112304, "grad_norm": 0.019423927972093226, "loss": 0.6852060377597808, "time": 0.4026209354400635, "epoch": 8, "memory": 20334, "step": 41706} +{"lr": 0.0007297252973710758, "data_time": 0.0015175819396972656, "grad_norm": 0.017389740380167495, "loss": 0.690499484539032, "time": 0.437898063659668, "epoch": 8, "memory": 20334, "step": 41806} +{"lr": 0.0007297252973710758, "data_time": 0.0015148878097534179, "grad_norm": 0.007821018889080733, "loss": 0.6846396505832673, "time": 0.4067711114883423, "epoch": 8, "memory": 20334, "step": 41906} +{"lr": 0.0007297252973710758, "data_time": 0.0015543460845947265, "grad_norm": 0.013813624647445977, "loss": 0.6831416726112366, "time": 0.3627128601074219, "epoch": 8, "memory": 20334, "step": 42006} +{"lr": 0.0007297252973710758, "data_time": 0.001451730728149414, "grad_norm": 0.019529200764372943, "loss": 0.689741575717926, "time": 0.44032466411590576, "epoch": 8, "memory": 20334, "step": 42106} +{"lr": 0.0007297252973710758, "data_time": 0.0014779329299926757, "grad_norm": 0.015099528920836746, "loss": 0.6836788535118103, "time": 0.4030714511871338, "epoch": 8, "memory": 20334, "step": 42206} +{"lr": 0.0007297252973710758, "data_time": 0.0015130043029785156, "grad_norm": 0.015384985768469051, "loss": 0.6880769610404969, "time": 0.40560460090637207, "epoch": 8, "memory": 20334, "step": 42306} +{"lr": 0.0007297252973710758, "data_time": 0.0015192508697509765, "grad_norm": 0.017572984064463525, "loss": 0.6852215945720672, "time": 0.4113664150238037, "epoch": 8, "memory": 20334, "step": 42406} +{"lr": 0.0007297252973710758, "data_time": 0.0012242555618286132, "grad_norm": 0.01690259254537523, "loss": 0.6877039253711701, "time": 0.4119797945022583, "epoch": 8, "memory": 20334, "step": 42506} +{"lr": 0.0007297252973710758, "data_time": 0.001332402229309082, "grad_norm": 0.010942086321301758, "loss": 0.6843304991722107, "time": 0.47026662826538085, "epoch": 8, "memory": 20334, "step": 42606} +{"lr": 0.0007297252973710758, "data_time": 0.001267719268798828, "grad_norm": 0.01368114808574319, "loss": 0.6946668744087219, "time": 0.4309420347213745, "epoch": 8, "memory": 20334, "step": 42706} +{"lr": 0.0007297252973710758, "data_time": 0.0012215852737426757, "grad_norm": 0.015644904031069017, "loss": 0.6836284339427948, "time": 0.4006821870803833, "epoch": 8, "memory": 20334, "step": 42806} +{"lr": 0.0007297252973710758, "data_time": 0.0016925573348999024, "grad_norm": 0.01545888283289969, "loss": 0.6884254932403564, "time": 0.4241483688354492, "epoch": 8, "memory": 20334, "step": 42906} +{"lr": 0.0007297252973710758, "data_time": 0.0017265558242797851, "grad_norm": 0.012041582120582461, "loss": 0.6839982450008393, "time": 0.46854605674743655, "epoch": 8, "memory": 20334, "step": 43006} +{"lr": 0.0007297252973710758, "data_time": 0.0014457464218139648, "grad_norm": 0.009560292994137853, "loss": 0.682794165611267, "time": 0.41179053783416747, "epoch": 8, "memory": 20334, "step": 43106} +{"lr": 0.0007297252973710758, "data_time": 0.001885390281677246, "grad_norm": 0.012177361303474753, "loss": 0.6835753798484803, "time": 0.4030990839004517, "epoch": 8, "memory": 20334, "step": 43206} +{"lr": 0.0007297252973710758, "data_time": 0.0017806291580200195, "grad_norm": 0.016058540809899567, "loss": 0.6853805541992187, "time": 0.36375586986541747, "epoch": 8, "memory": 20334, "step": 43306} +{"lr": 0.0007297252973710758, "data_time": 0.0019548892974853515, "grad_norm": 0.02024875421775505, "loss": 0.6902428925037384, "time": 0.43312194347381594, "epoch": 8, "memory": 20334, "step": 43406} +{"lr": 0.0007297252973710758, "data_time": 0.002050614356994629, "grad_norm": 0.019893907569348813, "loss": 0.6832017123699188, "time": 0.38632769584655763, "epoch": 8, "memory": 20334, "step": 43506} +{"lr": 0.0007297252973710758, "data_time": 0.001772904396057129, "grad_norm": 0.014353366824798286, "loss": 0.6889331459999084, "time": 0.4603415489196777, "epoch": 8, "memory": 20334, "step": 43606} +{"lr": 0.0007297252973710758, "data_time": 0.001581120491027832, "grad_norm": 0.0169849916594103, "loss": 0.6835180044174194, "time": 0.39262301921844484, "epoch": 8, "memory": 20334, "step": 43706} +{"lr": 0.0007297252973710758, "data_time": 0.0018399715423583984, "grad_norm": 0.017583088134415448, "loss": 0.6803718268871307, "time": 0.40932228565216067, "epoch": 8, "memory": 20334, "step": 43806} +{"lr": 0.0007297252973710758, "data_time": 0.0015604019165039063, "grad_norm": 0.016612075013108553, "loss": 0.6864450633525848, "time": 0.40414042472839357, "epoch": 8, "memory": 20334, "step": 43906} +{"lr": 0.0007297252973710758, "data_time": 0.001879429817199707, "grad_norm": 0.012774892989546061, "loss": 0.686683738231659, "time": 0.38671696186065674, "epoch": 8, "memory": 20334, "step": 44006} +{"lr": 0.0007297252973710758, "data_time": 0.0022849082946777345, "grad_norm": 0.01221808884292841, "loss": 0.6845331132411957, "time": 0.4175445556640625, "epoch": 8, "memory": 20334, "step": 44106} +{"lr": 0.0007297252973710758, "data_time": 0.001493215560913086, "grad_norm": 0.014803660998586566, "loss": 0.6846046149730682, "time": 0.3991935968399048, "epoch": 8, "memory": 20334, "step": 44206} +{"lr": 0.0007297252973710758, "data_time": 0.0016632080078125, "grad_norm": 0.019390276959165932, "loss": 0.6883752167224884, "time": 0.4085487127304077, "epoch": 8, "memory": 20334, "step": 44306} +{"lr": 0.0007297252973710758, "data_time": 0.0018727540969848632, "grad_norm": 0.021830077411141246, "loss": 0.6901873409748077, "time": 0.43409912586212157, "epoch": 8, "memory": 20334, "step": 44406} +{"lr": 0.0007297252973710758, "data_time": 0.0026200056076049806, "grad_norm": 0.009195657703094185, "loss": 0.6831157267093658, "time": 0.4425152540206909, "epoch": 8, "memory": 20334, "step": 44506} +{"lr": 0.0007297252973710758, "data_time": 0.003099370002746582, "grad_norm": 0.014189627242740244, "loss": 0.6839025855064392, "time": 0.38259291648864746, "epoch": 8, "memory": 20334, "step": 44606} +{"lr": 0.0007297252973710758, "data_time": 0.0019882917404174805, "grad_norm": 0.014124078035820276, "loss": 0.6895446300506591, "time": 0.39026198387145994, "epoch": 8, "memory": 20334, "step": 44706} +{"lr": 0.0007297252973710758, "data_time": 0.00212252140045166, "grad_norm": 0.009498244698625058, "loss": 0.6897653877735138, "time": 0.40053367614746094, "epoch": 8, "memory": 20334, "step": 44806} +{"lr": 0.0007297252973710758, "data_time": 0.0024294376373291014, "grad_norm": 0.018176346668042243, "loss": 0.6842960834503173, "time": 0.3951588153839111, "epoch": 8, "memory": 20334, "step": 44906} +{"lr": 0.0007297252973710758, "data_time": 0.0017308712005615235, "grad_norm": 0.013260247011203319, "loss": 0.6856667935848236, "time": 0.38628082275390624, "epoch": 8, "memory": 20334, "step": 45006} +{"lr": 0.0007297252973710758, "data_time": 0.0019758462905883787, "grad_norm": 0.010604234255151824, "loss": 0.6831212699413299, "time": 0.4084931373596191, "epoch": 8, "memory": 20334, "step": 45106} +{"lr": 0.0007297252973710758, "data_time": 0.00211794376373291, "grad_norm": 0.019600277161225678, "loss": 0.6864668607711792, "time": 0.38871335983276367, "epoch": 8, "memory": 20334, "step": 45206} +{"lr": 0.0007297252973710758, "data_time": 0.0019192457199096679, "grad_norm": 0.016822970937937498, "loss": 0.6830528199672699, "time": 0.41142303943634034, "epoch": 8, "memory": 20334, "step": 45306} +{"lr": 0.0007297252973710758, "data_time": 0.0017076253890991211, "grad_norm": 0.017361124770832247, "loss": 0.6849460542201996, "time": 0.4418827295303345, "epoch": 8, "memory": 20334, "step": 45406} +{"lr": 0.0007297252973710758, "data_time": 0.0015312671661376954, "grad_norm": 0.02476479150354862, "loss": 0.6893586575984955, "time": 0.4284776210784912, "epoch": 8, "memory": 20334, "step": 45506} +{"lr": 0.0007297252973710758, "data_time": 0.0016729116439819336, "grad_norm": 0.01935760343912989, "loss": 0.6840833008289338, "time": 0.38116753101348877, "epoch": 8, "memory": 20334, "step": 45606} +{"lr": 0.0007297252973710758, "data_time": 0.0016595125198364258, "grad_norm": 0.01084623575443402, "loss": 0.6872931957244873, "time": 0.42051172256469727, "epoch": 8, "memory": 20334, "step": 45706} +{"lr": 0.0007297252973710758, "data_time": 0.0016260623931884765, "grad_norm": 0.009914619935443625, "loss": 0.6845249891281128, "time": 0.41153945922851565, "epoch": 8, "memory": 20334, "step": 45806} +{"lr": 0.0007297252973710758, "data_time": 0.0017794370651245117, "grad_norm": 0.023571905423887075, "loss": 0.6867764115333557, "time": 0.3996510744094849, "epoch": 8, "memory": 20334, "step": 45906} +{"lr": 0.0007297252973710758, "data_time": 0.0017148971557617188, "grad_norm": 0.014147424115799367, "loss": 0.6926047623157501, "time": 0.38025755882263185, "epoch": 8, "memory": 20334, "step": 46006} +{"accuracy/top1": 100.0, "data_time": 0.3850647078620063, "time": 1.0093521806928847, "step": 8} +{"lr": 0.0006579634122155991, "data_time": 0.0016479969024658203, "grad_norm": 0.01419377648853697, "loss": 0.6865527033805847, "time": 0.36565732955932617, "epoch": 9, "memory": 20334, "step": 46164} +{"lr": 0.0006579634122155991, "data_time": 0.0016137361526489258, "grad_norm": 0.019851939799264074, "loss": 0.6879981100559235, "time": 0.41054272651672363, "epoch": 9, "memory": 20334, "step": 46264} +{"lr": 0.0006579634122155991, "data_time": 0.0016621828079223632, "grad_norm": 0.008498499635607004, "loss": 0.6841829657554627, "time": 0.3799438953399658, "epoch": 9, "memory": 20334, "step": 46364} +{"lr": 0.0006579634122155991, "data_time": 0.0015968561172485351, "grad_norm": 0.01799968988634646, "loss": 0.6869735836982727, "time": 0.38701753616333007, "epoch": 9, "memory": 20334, "step": 46464} +{"lr": 0.0006579634122155991, "data_time": 0.0016284704208374024, "grad_norm": 0.01654924225003924, "loss": 0.6871818482875824, "time": 0.4096548318862915, "epoch": 9, "memory": 20334, "step": 46564} +{"lr": 0.0006579634122155991, "data_time": 0.0015188217163085937, "grad_norm": 0.017102921684272587, "loss": 0.6877378702163697, "time": 0.40784056186676027, "epoch": 9, "memory": 20334, "step": 46664} +{"lr": 0.0006579634122155991, "data_time": 0.0014745235443115235, "grad_norm": 0.026159213646315037, "loss": 0.6898949027061463, "time": 0.4024815082550049, "epoch": 9, "memory": 20334, "step": 46764} +{"lr": 0.0006579634122155991, "data_time": 0.0015573501586914062, "grad_norm": 0.013409408857114613, "loss": 0.6825119733810425, "time": 0.40892596244812013, "epoch": 9, "memory": 20334, "step": 46864} +{"lr": 0.0006579634122155991, "data_time": 0.0016701936721801758, "grad_norm": 0.016575712710618973, "loss": 0.6853305220603942, "time": 0.43486199378967283, "epoch": 9, "memory": 20334, "step": 46964} +{"lr": 0.0006579634122155991, "data_time": 0.001623249053955078, "grad_norm": 0.029668823443353178, "loss": 0.6825747966766358, "time": 0.4156445264816284, "epoch": 9, "memory": 20334, "step": 47064} +{"lr": 0.0006579634122155991, "data_time": 0.001653432846069336, "grad_norm": 0.024123814527411015, "loss": 0.6842483103275299, "time": 0.400403356552124, "epoch": 9, "memory": 20334, "step": 47164} +{"lr": 0.0006579634122155991, "data_time": 0.0015319108963012696, "grad_norm": 0.016646164143458007, "loss": 0.6879630446434021, "time": 0.3889906883239746, "epoch": 9, "memory": 20334, "step": 47264} +{"lr": 0.0006579634122155991, "data_time": 0.0018725395202636719, "grad_norm": 0.017204239362035877, "loss": 0.6901099979877472, "time": 0.41818020343780515, "epoch": 9, "memory": 20334, "step": 47364} +{"lr": 0.0006579634122155991, "data_time": 0.0017591953277587891, "grad_norm": 0.023627828760072588, "loss": 0.6925867855548858, "time": 0.37683274745941164, "epoch": 9, "memory": 20334, "step": 47464} +{"lr": 0.0006579634122155991, "data_time": 0.0015821456909179688, "grad_norm": 0.020276833092793822, "loss": 0.6883210062980651, "time": 0.378000283241272, "epoch": 9, "memory": 20334, "step": 47564} +{"lr": 0.0006579634122155991, "data_time": 0.0018356800079345702, "grad_norm": 0.020691714878194035, "loss": 0.6894786536693573, "time": 0.43771560192108155, "epoch": 9, "memory": 20334, "step": 47664} +{"lr": 0.0006579634122155991, "data_time": 0.0015072107315063476, "grad_norm": 0.02632945142686367, "loss": 0.6901565432548523, "time": 0.423551344871521, "epoch": 9, "memory": 20334, "step": 47764} +{"lr": 0.0006579634122155991, "data_time": 0.0014970541000366212, "grad_norm": 0.016661419306183235, "loss": 0.6927356719970703, "time": 0.3900083065032959, "epoch": 9, "memory": 20334, "step": 47864} +{"lr": 0.0006579634122155991, "data_time": 0.0014163017272949218, "grad_norm": 0.01677846600068733, "loss": 0.6866316437721253, "time": 0.4081490755081177, "epoch": 9, "memory": 20334, "step": 47964} +{"lr": 0.0006579634122155991, "data_time": 0.0014652252197265626, "grad_norm": 0.011928746732883156, "loss": 0.6864598095417023, "time": 0.4291290283203125, "epoch": 9, "memory": 20334, "step": 48064} +{"lr": 0.0006579634122155991, "data_time": 0.0014700174331665039, "grad_norm": 0.018529455410316586, "loss": 0.6883900165557861, "time": 0.36959547996520997, "epoch": 9, "memory": 20334, "step": 48164} +{"lr": 0.0006579634122155991, "data_time": 0.0014411687850952148, "grad_norm": 0.016606410639360546, "loss": 0.6852875113487243, "time": 0.43194167613983153, "epoch": 9, "memory": 20334, "step": 48264} +{"lr": 0.0006579634122155991, "data_time": 0.001485157012939453, "grad_norm": 0.014752424554899335, "loss": 0.6843710303306579, "time": 0.41739075183868407, "epoch": 9, "memory": 20334, "step": 48364} +{"lr": 0.0006579634122155991, "data_time": 0.001543140411376953, "grad_norm": 0.01776666350197047, "loss": 0.6881593048572541, "time": 0.43468174934387205, "epoch": 9, "memory": 20334, "step": 48464} +{"lr": 0.0006579634122155991, "data_time": 0.0017083406448364258, "grad_norm": 0.016836788272485137, "loss": 0.6823172032833099, "time": 0.38342211246490476, "epoch": 9, "memory": 20334, "step": 48564} +{"lr": 0.0006579634122155991, "data_time": 0.0016509532928466798, "grad_norm": 0.012628317018970847, "loss": 0.6859584987163544, "time": 0.38100109100341795, "epoch": 9, "memory": 20334, "step": 48664} +{"lr": 0.0006579634122155991, "data_time": 0.001735687255859375, "grad_norm": 0.019650646939408035, "loss": 0.6837439060211181, "time": 0.39613118171691897, "epoch": 9, "memory": 20334, "step": 48764} +{"lr": 0.0006579634122155991, "data_time": 0.0017971038818359376, "grad_norm": 0.01567758454475552, "loss": 0.6874577343463898, "time": 0.44351351261138916, "epoch": 9, "memory": 20334, "step": 48864} +{"lr": 0.0006579634122155991, "data_time": 0.00178220272064209, "grad_norm": 0.01802654191851616, "loss": 0.6824274778366088, "time": 0.3819967031478882, "epoch": 9, "memory": 20334, "step": 48964} +{"lr": 0.0006579634122155991, "data_time": 0.0016217470169067384, "grad_norm": 0.01629552954691462, "loss": 0.6909584164619446, "time": 0.40527799129486086, "epoch": 9, "memory": 20334, "step": 49064} +{"lr": 0.0006579634122155991, "data_time": 0.001608872413635254, "grad_norm": 0.017447477905079722, "loss": 0.6902225911617279, "time": 0.38200912475585935, "epoch": 9, "memory": 20334, "step": 49164} +{"lr": 0.0006579634122155991, "data_time": 0.0017590761184692384, "grad_norm": 0.012975326050946024, "loss": 0.6888095498085022, "time": 0.4060666561126709, "epoch": 9, "memory": 20334, "step": 49264} +{"lr": 0.0006579634122155991, "data_time": 0.0017446041107177734, "grad_norm": 0.015692339441739023, "loss": 0.6879356622695922, "time": 0.3684777498245239, "epoch": 9, "memory": 20334, "step": 49364} +{"lr": 0.0006579634122155991, "data_time": 0.001659083366394043, "grad_norm": 0.01458758725784719, "loss": 0.6808813750743866, "time": 0.43731138706207273, "epoch": 9, "memory": 20334, "step": 49464} +{"lr": 0.0006579634122155991, "data_time": 0.001909947395324707, "grad_norm": 0.016200729040428997, "loss": 0.6865518748760223, "time": 0.36342535018920896, "epoch": 9, "memory": 20334, "step": 49564} +{"lr": 0.0006579634122155991, "data_time": 0.0017020702362060547, "grad_norm": 0.020468140457523987, "loss": 0.6849137902259826, "time": 0.3801534652709961, "epoch": 9, "memory": 20334, "step": 49664} +{"lr": 0.0006579634122155991, "data_time": 0.0016418933868408204, "grad_norm": 0.01708055417984724, "loss": 0.6884378314018249, "time": 0.39692373275756837, "epoch": 9, "memory": 20334, "step": 49764} +{"lr": 0.0006579634122155991, "data_time": 0.001673579216003418, "grad_norm": 0.010124088363954797, "loss": 0.6838084518909454, "time": 0.4089380979537964, "epoch": 9, "memory": 20334, "step": 49864} +{"lr": 0.0006579634122155991, "data_time": 0.0017638683319091796, "grad_norm": 0.014845739683369174, "loss": 0.689606660604477, "time": 0.3887681484222412, "epoch": 9, "memory": 20334, "step": 49964} +{"lr": 0.0006579634122155991, "data_time": 0.0015306711196899415, "grad_norm": 0.013253023452125489, "loss": 0.6858506262302398, "time": 0.4657459259033203, "epoch": 9, "memory": 20334, "step": 50064} +{"lr": 0.0006579634122155991, "data_time": 0.0015999317169189454, "grad_norm": 0.012637721397913993, "loss": 0.6846360683441162, "time": 0.4193205118179321, "epoch": 9, "memory": 20334, "step": 50164} +{"lr": 0.0006579634122155991, "data_time": 0.0015675544738769532, "grad_norm": 0.01643711608485319, "loss": 0.6868250489234924, "time": 0.45740585327148436, "epoch": 9, "memory": 20334, "step": 50264} +{"lr": 0.0006579634122155991, "data_time": 0.0015584230422973633, "grad_norm": 0.019852726720273495, "loss": 0.6844405114650727, "time": 0.40009050369262694, "epoch": 9, "memory": 20334, "step": 50364} +{"lr": 0.0006579634122155991, "data_time": 0.0014082908630371094, "grad_norm": 0.016296138672623785, "loss": 0.6904605507850647, "time": 0.41584053039550783, "epoch": 9, "memory": 20334, "step": 50464} +{"lr": 0.0006579634122155991, "data_time": 0.0012562990188598633, "grad_norm": 0.015375867043621838, "loss": 0.6873063504695892, "time": 0.4089855194091797, "epoch": 9, "memory": 20334, "step": 50564} +{"lr": 0.0006579634122155991, "data_time": 0.0013644933700561524, "grad_norm": 0.016306512081064282, "loss": 0.6848997890949249, "time": 0.4068164587020874, "epoch": 9, "memory": 20334, "step": 50664} +{"lr": 0.0006579634122155991, "data_time": 0.001303553581237793, "grad_norm": 0.012046275520697236, "loss": 0.6894914329051971, "time": 0.3742116928100586, "epoch": 9, "memory": 20334, "step": 50764} +{"lr": 0.0006579634122155991, "data_time": 0.0013122320175170898, "grad_norm": 0.01793232560157776, "loss": 0.6837822079658509, "time": 0.3626039981842041, "epoch": 9, "memory": 20334, "step": 50864} +{"lr": 0.0006579634122155991, "data_time": 0.0015569210052490234, "grad_norm": 0.010402942029759287, "loss": 0.6881684243679047, "time": 0.41551783084869387, "epoch": 9, "memory": 20334, "step": 50964} +{"lr": 0.0006579634122155991, "data_time": 0.0015505313873291015, "grad_norm": 0.01626515844836831, "loss": 0.6854903101921082, "time": 0.41629719734191895, "epoch": 9, "memory": 20334, "step": 51064} +{"lr": 0.0006579634122155991, "data_time": 0.0014333486557006835, "grad_norm": 0.01355631654150784, "loss": 0.6880523145198822, "time": 0.3880305767059326, "epoch": 9, "memory": 20334, "step": 51164} +{"lr": 0.0006579634122155991, "data_time": 0.0014967679977416991, "grad_norm": 0.015107460401486606, "loss": 0.6864442765712738, "time": 0.3931441783905029, "epoch": 9, "memory": 20334, "step": 51264} +{"lr": 0.0006579634122155991, "data_time": 0.0015359163284301759, "grad_norm": 0.021372124808840452, "loss": 0.684925252199173, "time": 0.36798503398895266, "epoch": 9, "memory": 20334, "step": 51364} +{"lr": 0.0006579634122155991, "data_time": 0.0017138481140136718, "grad_norm": 0.017495215917006136, "loss": 0.6825269877910614, "time": 0.4488940715789795, "epoch": 9, "memory": 20334, "step": 51464} +{"lr": 0.0006579634122155991, "data_time": 0.0016608238220214844, "grad_norm": 0.01851346460171044, "loss": 0.6859010696411133, "time": 0.41650211811065674, "epoch": 9, "memory": 20334, "step": 51564} +{"lr": 0.0006579634122155991, "data_time": 0.0017618656158447266, "grad_norm": 0.011144491762388497, "loss": 0.6866799414157867, "time": 0.3864098072052002, "epoch": 9, "memory": 20334, "step": 51664} +{"lr": 0.0006579634122155991, "data_time": 0.0015483856201171874, "grad_norm": 0.016839176486246287, "loss": 0.6818654358386993, "time": 0.3709238052368164, "epoch": 9, "memory": 20334, "step": 51764} +{"accuracy/top1": 100.0, "data_time": 0.371474027633667, "time": 0.9948434034983317, "step": 9} +{"lr": 0.0005824350601949144, "data_time": 0.0015693187713623046, "grad_norm": 0.013630917656701057, "loss": 0.6861572086811065, "time": 0.4262171983718872, "epoch": 10, "memory": 20334, "step": 51922} +{"lr": 0.0005824350601949144, "data_time": 0.0016273021697998046, "grad_norm": 0.015809801197610794, "loss": 0.6848917543888092, "time": 0.38904197216033937, "epoch": 10, "memory": 20334, "step": 52022} +{"lr": 0.0005824350601949144, "data_time": 0.0018444538116455078, "grad_norm": 0.015528548881411553, "loss": 0.6847606837749481, "time": 0.43321115970611573, "epoch": 10, "memory": 20334, "step": 52122} +{"lr": 0.0005824350601949144, "data_time": 0.0017694473266601563, "grad_norm": 0.015170224756002427, "loss": 0.6875871837139129, "time": 0.36998488903045657, "epoch": 10, "memory": 20334, "step": 52222} +{"lr": 0.0005824350601949144, "data_time": 0.0016355276107788085, "grad_norm": 0.010488854415598325, "loss": 0.6874406039714813, "time": 0.410483193397522, "epoch": 10, "memory": 20334, "step": 52322} +{"lr": 0.0005824350601949144, "data_time": 0.0015836238861083984, "grad_norm": 0.011591467773541808, "loss": 0.6903205394744873, "time": 0.41949594020843506, "epoch": 10, "memory": 20334, "step": 52422} +{"lr": 0.0005824350601949144, "data_time": 0.0017516374588012694, "grad_norm": 0.02135088196955621, "loss": 0.6842758774757385, "time": 0.39008872509002684, "epoch": 10, "memory": 20334, "step": 52522} +{"lr": 0.0005824350601949144, "data_time": 0.0014495372772216797, "grad_norm": 0.016125445498619228, "loss": 0.6893079340457916, "time": 0.4323333740234375, "epoch": 10, "memory": 20334, "step": 52622} +{"lr": 0.0005824350601949144, "data_time": 0.001698470115661621, "grad_norm": 0.01552496430813335, "loss": 0.6847705245018005, "time": 0.38407206535339355, "epoch": 10, "memory": 20334, "step": 52722} +{"lr": 0.0005824350601949144, "data_time": 0.0018587350845336915, "grad_norm": 0.01834776420146227, "loss": 0.6794097065925598, "time": 0.38328773975372316, "epoch": 10, "memory": 20334, "step": 52822} +{"lr": 0.0005824350601949144, "data_time": 0.0017519235610961915, "grad_norm": 0.01273553364444524, "loss": 0.6832436561584473, "time": 0.3640961885452271, "epoch": 10, "memory": 20334, "step": 52922} +{"lr": 0.0005824350601949144, "data_time": 0.0017412185668945312, "grad_norm": 0.018623796047177166, "loss": 0.6863396167755127, "time": 0.3907542943954468, "epoch": 10, "memory": 20334, "step": 53022} +{"lr": 0.0005824350601949144, "data_time": 0.001468515396118164, "grad_norm": 0.017051333433482797, "loss": 0.6921289026737213, "time": 0.4294722318649292, "epoch": 10, "memory": 20334, "step": 53122} +{"lr": 0.0005824350601949144, "data_time": 0.0016848087310791016, "grad_norm": 0.02026559254154563, "loss": 0.685213041305542, "time": 0.3710071325302124, "epoch": 10, "memory": 20334, "step": 53222} +{"lr": 0.0005824350601949144, "data_time": 0.001784515380859375, "grad_norm": 0.01818013987503946, "loss": 0.6863143682479859, "time": 0.3864701747894287, "epoch": 10, "memory": 20334, "step": 53322} +{"lr": 0.0005824350601949144, "data_time": 0.001483583450317383, "grad_norm": 0.017659070342779158, "loss": 0.6906317472457886, "time": 0.41943793296813964, "epoch": 10, "memory": 20334, "step": 53422} +{"lr": 0.0005824350601949144, "data_time": 0.0014837026596069337, "grad_norm": 0.014083898643730209, "loss": 0.6835853934288025, "time": 0.40425686836242675, "epoch": 10, "memory": 20334, "step": 53522} +{"lr": 0.0005824350601949144, "data_time": 0.0016284465789794921, "grad_norm": 0.01893379450775683, "loss": 0.6863393545150757, "time": 0.37946789264678954, "epoch": 10, "memory": 20334, "step": 53622} +{"lr": 0.0005824350601949144, "data_time": 0.0014427900314331055, "grad_norm": 0.011715705273672938, "loss": 0.6884918987751008, "time": 0.42046072483062746, "epoch": 10, "memory": 20334, "step": 53722} +{"lr": 0.0005824350601949144, "data_time": 0.0016260147094726562, "grad_norm": 0.018639779184013605, "loss": 0.6875562071800232, "time": 0.40404956340789794, "epoch": 10, "memory": 20334, "step": 53822} +{"lr": 0.0005824350601949144, "data_time": 0.0015421390533447265, "grad_norm": 0.02021653435076587, "loss": 0.6900019109249115, "time": 0.3707756519317627, "epoch": 10, "memory": 20334, "step": 53922} +{"lr": 0.0005824350601949144, "data_time": 0.0015859127044677735, "grad_norm": 0.013106706569669769, "loss": 0.6885911643505096, "time": 0.40043487548828127, "epoch": 10, "memory": 20334, "step": 54022} +{"lr": 0.0005824350601949144, "data_time": 0.001854681968688965, "grad_norm": 0.01030412825057283, "loss": 0.6827254593372345, "time": 0.37603662014007566, "epoch": 10, "memory": 20334, "step": 54122} +{"lr": 0.0005824350601949144, "data_time": 0.0016684532165527344, "grad_norm": 0.019184313947334886, "loss": 0.6844351291656494, "time": 0.413557767868042, "epoch": 10, "memory": 20334, "step": 54222} +{"lr": 0.0005824350601949144, "data_time": 0.00196535587310791, "grad_norm": 0.016056846117135138, "loss": 0.6877652645111084, "time": 0.4293826103210449, "epoch": 10, "memory": 20334, "step": 54322} +{"lr": 0.0005824350601949144, "data_time": 0.0013401269912719726, "grad_norm": 0.012847650586627424, "loss": 0.6899170875549316, "time": 0.42327816486358644, "epoch": 10, "memory": 20334, "step": 54422} +{"lr": 0.0005824350601949144, "data_time": 0.0014577150344848634, "grad_norm": 0.012815133947879077, "loss": 0.6877005279064179, "time": 0.41201415061950686, "epoch": 10, "memory": 20334, "step": 54522} +{"lr": 0.0005824350601949144, "data_time": 0.001948237419128418, "grad_norm": 0.01572614754550159, "loss": 0.6882017374038696, "time": 0.41745293140411377, "epoch": 10, "memory": 20334, "step": 54622} +{"lr": 0.0005824350601949144, "data_time": 0.0015142440795898437, "grad_norm": 0.017514981690328568, "loss": 0.6816524267196655, "time": 0.4090061902999878, "epoch": 10, "memory": 20334, "step": 54722} +{"lr": 0.0005824350601949144, "data_time": 0.0015898942947387695, "grad_norm": 0.0068565423134714365, "loss": 0.6841822147369385, "time": 0.39079787731170657, "epoch": 10, "memory": 20334, "step": 54822} +{"lr": 0.0005824350601949144, "data_time": 0.0017063617706298828, "grad_norm": 0.018116200400982052, "loss": 0.6883988082408905, "time": 0.40350868701934817, "epoch": 10, "memory": 20334, "step": 54922} +{"lr": 0.0005824350601949144, "data_time": 0.0016272783279418946, "grad_norm": 0.008793468365911394, "loss": 0.6830611050128936, "time": 0.39291794300079347, "epoch": 10, "memory": 20334, "step": 55022} +{"lr": 0.0005824350601949144, "data_time": 0.0015559673309326171, "grad_norm": 0.011733010620810091, "loss": 0.6843908667564392, "time": 0.4198174476623535, "epoch": 10, "memory": 20334, "step": 55122} +{"lr": 0.0005824350601949144, "data_time": 0.0017212390899658202, "grad_norm": 0.011460137815447524, "loss": 0.6900747001171113, "time": 0.40550053119659424, "epoch": 10, "memory": 20334, "step": 55222} +{"lr": 0.0005824350601949144, "data_time": 0.001712965965270996, "grad_norm": 0.014100943400990218, "loss": 0.6912566542625427, "time": 0.3676278114318848, "epoch": 10, "memory": 20334, "step": 55322} +{"lr": 0.0005824350601949144, "data_time": 0.0017971754074096679, "grad_norm": 0.015873079164884986, "loss": 0.6866163492202759, "time": 0.3884373426437378, "epoch": 10, "memory": 20334, "step": 55422} +{"lr": 0.0005824350601949144, "data_time": 0.001832723617553711, "grad_norm": 0.021550805680453776, "loss": 0.6876806855201721, "time": 0.40866105556488036, "epoch": 10, "memory": 20334, "step": 55522} +{"lr": 0.0005824350601949144, "data_time": 0.0016146183013916015, "grad_norm": 0.014574577275197953, "loss": 0.6817083835601807, "time": 0.41771717071533204, "epoch": 10, "memory": 20334, "step": 55622} +{"lr": 0.0005824350601949144, "data_time": 0.0019276857376098633, "grad_norm": 0.010971394274383784, "loss": 0.6877998650074005, "time": 0.38044607639312744, "epoch": 10, "memory": 20334, "step": 55722} +{"lr": 0.0005824350601949144, "data_time": 0.0015934467315673827, "grad_norm": 0.01014065844938159, "loss": 0.6867319226264954, "time": 0.3712127447128296, "epoch": 10, "memory": 20334, "step": 55822} +{"lr": 0.0005824350601949144, "data_time": 0.0019123077392578125, "grad_norm": 0.011535261198878289, "loss": 0.68774453997612, "time": 0.3735557794570923, "epoch": 10, "memory": 20334, "step": 55922} +{"lr": 0.0005824350601949144, "data_time": 0.0018356800079345702, "grad_norm": 0.012448557512834668, "loss": 0.6829745411872864, "time": 0.39452495574951174, "epoch": 10, "memory": 20334, "step": 56022} +{"lr": 0.0005824350601949144, "data_time": 0.001998615264892578, "grad_norm": 0.020786867733113466, "loss": 0.6918170928955079, "time": 0.3841336488723755, "epoch": 10, "memory": 20334, "step": 56122} +{"lr": 0.0005824350601949144, "data_time": 0.0017288446426391602, "grad_norm": 0.013509073189925402, "loss": 0.6818134009838104, "time": 0.40670504570007326, "epoch": 10, "memory": 20334, "step": 56222} +{"lr": 0.0005824350601949144, "data_time": 0.0018560409545898438, "grad_norm": 0.012721085408702492, "loss": 0.68084916472435, "time": 0.3725078344345093, "epoch": 10, "memory": 20334, "step": 56322} +{"lr": 0.0005824350601949144, "data_time": 0.0015450716018676758, "grad_norm": 0.02103833546862006, "loss": 0.6907468199729919, "time": 0.3778859853744507, "epoch": 10, "memory": 20334, "step": 56422} +{"lr": 0.0005824350601949144, "data_time": 0.0017362117767333984, "grad_norm": 0.012111650290898978, "loss": 0.6793406724929809, "time": 0.415705132484436, "epoch": 10, "memory": 20334, "step": 56522} +{"lr": 0.0005824350601949144, "data_time": 0.001720738410949707, "grad_norm": 0.015598732838407159, "loss": 0.6881059110164642, "time": 0.38643875122070315, "epoch": 10, "memory": 20334, "step": 56622} +{"lr": 0.0005824350601949144, "data_time": 0.001687026023864746, "grad_norm": 0.015401056152768433, "loss": 0.6878041982650757, "time": 0.4160081624984741, "epoch": 10, "memory": 20334, "step": 56722} +{"lr": 0.0005824350601949144, "data_time": 0.0016798019409179688, "grad_norm": 0.016470563248731196, "loss": 0.6827340066432953, "time": 0.40014288425445554, "epoch": 10, "memory": 20334, "step": 56822} +{"lr": 0.0005824350601949144, "data_time": 0.0015673160552978516, "grad_norm": 0.013484801992308348, "loss": 0.6872635900974273, "time": 0.38296396732330323, "epoch": 10, "memory": 20334, "step": 56922} +{"lr": 0.0005824350601949144, "data_time": 0.0015890359878540038, "grad_norm": 0.019567966926842928, "loss": 0.6892824411392212, "time": 0.4249274253845215, "epoch": 10, "memory": 20334, "step": 57022} +{"lr": 0.0005824350601949144, "data_time": 0.0016318559646606445, "grad_norm": 0.018295177374966443, "loss": 0.6846611738204956, "time": 0.4224911451339722, "epoch": 10, "memory": 20334, "step": 57122} +{"lr": 0.0005824350601949144, "data_time": 0.0015627622604370117, "grad_norm": 0.01712101136799902, "loss": 0.6868502974510193, "time": 0.396042537689209, "epoch": 10, "memory": 20334, "step": 57222} +{"lr": 0.0005824350601949144, "data_time": 0.0019245147705078125, "grad_norm": 0.011905309010762722, "loss": 0.6864865183830261, "time": 0.421490740776062, "epoch": 10, "memory": 20334, "step": 57322} +{"lr": 0.0005824350601949144, "data_time": 0.0018316984176635742, "grad_norm": 0.02494481534231454, "loss": 0.6872658133506775, "time": 0.37062802314758303, "epoch": 10, "memory": 20334, "step": 57422} +{"lr": 0.0005824350601949144, "data_time": 0.0019359111785888672, "grad_norm": 0.014627020398620516, "loss": 0.6856526255607605, "time": 0.4216599464416504, "epoch": 10, "memory": 20334, "step": 57522} +{"accuracy/top1": 100.0, "data_time": 0.3692814244164361, "time": 1.0086095598008897, "step": 10} +{"lr": 0.0005050000000000001, "data_time": 0.0016867876052856444, "grad_norm": 0.01556721068918705, "loss": 0.6872677981853486, "time": 0.40900297164916993, "epoch": 11, "memory": 20334, "step": 57680} +{"lr": 0.0005050000000000001, "data_time": 0.0017450094223022462, "grad_norm": 0.013444986648391933, "loss": 0.6864603221416473, "time": 0.3714394807815552, "epoch": 11, "memory": 20334, "step": 57780} +{"lr": 0.0005050000000000001, "data_time": 0.0017448663711547852, "grad_norm": 0.01790513030719012, "loss": 0.690277487039566, "time": 0.3853780746459961, "epoch": 11, "memory": 20334, "step": 57880} +{"lr": 0.0005050000000000001, "data_time": 0.0016040325164794922, "grad_norm": 0.012645078659988939, "loss": 0.6907342135906219, "time": 0.4526124715805054, "epoch": 11, "memory": 20334, "step": 57980} +{"lr": 0.0005050000000000001, "data_time": 0.0018157005310058595, "grad_norm": 0.01319127669557929, "loss": 0.6838784575462341, "time": 0.4102788925170898, "epoch": 11, "memory": 20334, "step": 58080} +{"lr": 0.0005050000000000001, "data_time": 0.0016178131103515626, "grad_norm": 0.010991183505393565, "loss": 0.6889654695987701, "time": 0.3825472354888916, "epoch": 11, "memory": 20334, "step": 58180} +{"lr": 0.0005050000000000001, "data_time": 0.0017494440078735351, "grad_norm": 0.016774243616964668, "loss": 0.6823360204696656, "time": 0.4215251445770264, "epoch": 11, "memory": 20334, "step": 58280} +{"lr": 0.0005050000000000001, "data_time": 0.001706075668334961, "grad_norm": 0.01579231631476432, "loss": 0.6852897703647614, "time": 0.3628375053405762, "epoch": 11, "memory": 20334, "step": 58380} +{"lr": 0.0005050000000000001, "data_time": 0.0016942739486694336, "grad_norm": 0.014355609053745865, "loss": 0.6871700406074523, "time": 0.39527525901794436, "epoch": 11, "memory": 20334, "step": 58480} +{"lr": 0.0005050000000000001, "data_time": 0.001577615737915039, "grad_norm": 0.01576843245420605, "loss": 0.6874830901622773, "time": 0.39677345752716064, "epoch": 11, "memory": 20334, "step": 58580} +{"lr": 0.0005050000000000001, "data_time": 0.0016754627227783202, "grad_norm": 0.013481079041957856, "loss": 0.6834925889968873, "time": 0.3885331392288208, "epoch": 11, "memory": 20334, "step": 58680} +{"lr": 0.0005050000000000001, "data_time": 0.0016225576400756836, "grad_norm": 0.01784527823328972, "loss": 0.6883037149906158, "time": 0.47656280994415284, "epoch": 11, "memory": 20334, "step": 58780} +{"lr": 0.0005050000000000001, "data_time": 0.0018278598785400391, "grad_norm": 0.021230232785455883, "loss": 0.687637597322464, "time": 0.37162487506866454, "epoch": 11, "memory": 20334, "step": 58880} +{"lr": 0.0005050000000000001, "data_time": 0.002013278007507324, "grad_norm": 0.011489739397075028, "loss": 0.6880789995193481, "time": 0.4111733675003052, "epoch": 11, "memory": 20334, "step": 58980} +{"lr": 0.0005050000000000001, "data_time": 0.0019643545150756837, "grad_norm": 0.011365625041071326, "loss": 0.6861910879611969, "time": 0.402612042427063, "epoch": 11, "memory": 20334, "step": 59080} +{"lr": 0.0005050000000000001, "data_time": 0.0018092870712280273, "grad_norm": 0.014367542136460543, "loss": 0.6900720775127411, "time": 0.4091456890106201, "epoch": 11, "memory": 20334, "step": 59180} +{"lr": 0.0005050000000000001, "data_time": 0.0016371250152587891, "grad_norm": 0.020088522182777523, "loss": 0.6836402118206024, "time": 0.43627243041992186, "epoch": 11, "memory": 20334, "step": 59280} +{"lr": 0.0005050000000000001, "data_time": 0.0017477035522460937, "grad_norm": 0.014833125588484108, "loss": 0.6904846727848053, "time": 0.39355833530426027, "epoch": 11, "memory": 20334, "step": 59380} +{"lr": 0.0005050000000000001, "data_time": 0.0016386270523071288, "grad_norm": 0.02122778237098828, "loss": 0.6817389607429505, "time": 0.42151169776916503, "epoch": 11, "memory": 20334, "step": 59480} +{"lr": 0.0005050000000000001, "data_time": 0.0015434503555297851, "grad_norm": 0.019401673693209887, "loss": 0.6845398485660553, "time": 0.39098615646362306, "epoch": 11, "memory": 20334, "step": 59580} +{"lr": 0.0005050000000000001, "data_time": 0.0015362262725830077, "grad_norm": 0.012637698696926236, "loss": 0.6813634753227233, "time": 0.4173738956451416, "epoch": 11, "memory": 20334, "step": 59680} +{"lr": 0.0005050000000000001, "data_time": 0.001519465446472168, "grad_norm": 0.017714382335543632, "loss": 0.6863687574863434, "time": 0.3951176881790161, "epoch": 11, "memory": 20334, "step": 59780} +{"lr": 0.0005050000000000001, "data_time": 0.0014509439468383789, "grad_norm": 0.01826527458615601, "loss": 0.685668557882309, "time": 0.40448362827301027, "epoch": 11, "memory": 20334, "step": 59880} +{"lr": 0.0005050000000000001, "data_time": 0.0012654781341552735, "grad_norm": 0.01331446843687445, "loss": 0.6856652140617371, "time": 0.4253491163253784, "epoch": 11, "memory": 20334, "step": 59980} +{"lr": 0.0005050000000000001, "data_time": 0.001400899887084961, "grad_norm": 0.01342430289951153, "loss": 0.6874310374259949, "time": 0.3719109773635864, "epoch": 11, "memory": 20334, "step": 60080} +{"lr": 0.0005050000000000001, "data_time": 0.0012639999389648438, "grad_norm": 0.01627713702619076, "loss": 0.6869747042655945, "time": 0.40706191062927244, "epoch": 11, "memory": 20334, "step": 60180} +{"lr": 0.0005050000000000001, "data_time": 0.0014351367950439452, "grad_norm": 0.013047590979840606, "loss": 0.6925990104675293, "time": 0.41955182552337644, "epoch": 11, "memory": 20334, "step": 60280} +{"lr": 0.0005050000000000001, "data_time": 0.0014681577682495116, "grad_norm": 0.0162918713176623, "loss": 0.6841265261173248, "time": 0.42308735847473145, "epoch": 11, "memory": 20334, "step": 60380} +{"lr": 0.0005050000000000001, "data_time": 0.001253199577331543, "grad_norm": 0.015976412466261535, "loss": 0.6813515782356262, "time": 0.4027266025543213, "epoch": 11, "memory": 20334, "step": 60480} +{"lr": 0.0005050000000000001, "data_time": 0.0013670206069946289, "grad_norm": 0.018589975498616695, "loss": 0.6886433362960815, "time": 0.3883127927780151, "epoch": 11, "memory": 20334, "step": 60580} +{"lr": 0.0005050000000000001, "data_time": 0.0019369125366210938, "grad_norm": 0.018638608464971184, "loss": 0.6886873841285706, "time": 0.4225748062133789, "epoch": 11, "memory": 20334, "step": 60680} +{"lr": 0.0005050000000000001, "data_time": 0.0016449213027954102, "grad_norm": 0.023885917756706475, "loss": 0.6888404905796051, "time": 0.3968960762023926, "epoch": 11, "memory": 20334, "step": 60780} +{"lr": 0.0005050000000000001, "data_time": 0.0016598701477050781, "grad_norm": 0.019483834481798114, "loss": 0.6818107306957245, "time": 0.3683042526245117, "epoch": 11, "memory": 20334, "step": 60880} +{"lr": 0.0005050000000000001, "data_time": 0.0015219926834106445, "grad_norm": 0.021793618984520435, "loss": 0.6875822424888611, "time": 0.40129795074462893, "epoch": 11, "memory": 20334, "step": 60980} +{"lr": 0.0005050000000000001, "data_time": 0.0016986370086669923, "grad_norm": 0.01363177023595199, "loss": 0.6868061244487762, "time": 0.39838347434997556, "epoch": 11, "memory": 20334, "step": 61080} +{"lr": 0.0005050000000000001, "data_time": 0.0018582344055175781, "grad_norm": 0.014399600576143711, "loss": 0.6916409134864807, "time": 0.45100481510162355, "epoch": 11, "memory": 20334, "step": 61180} +{"lr": 0.0005050000000000001, "data_time": 0.0019231796264648437, "grad_norm": 0.012664830044377596, "loss": 0.6798730075359345, "time": 0.40946035385131835, "epoch": 11, "memory": 20334, "step": 61280} +{"lr": 0.0005050000000000001, "data_time": 0.001731133460998535, "grad_norm": 0.026317989779636265, "loss": 0.6835351705551147, "time": 0.4009954214096069, "epoch": 11, "memory": 20334, "step": 61380} +{"lr": 0.0005050000000000001, "data_time": 0.0018470048904418944, "grad_norm": 0.02570729167200625, "loss": 0.6870627582073212, "time": 0.41326916217803955, "epoch": 11, "memory": 20334, "step": 61480} +{"lr": 0.0005050000000000001, "data_time": 0.0018560647964477538, "grad_norm": 0.012357833934947849, "loss": 0.686946451663971, "time": 0.36675879955291746, "epoch": 11, "memory": 20334, "step": 61580} +{"lr": 0.0005050000000000001, "data_time": 0.0015499353408813476, "grad_norm": 0.013128361222334206, "loss": 0.6843057632446289, "time": 0.4262408971786499, "epoch": 11, "memory": 20334, "step": 61680} +{"lr": 0.0005050000000000001, "data_time": 0.0018424510955810547, "grad_norm": 0.009496481786482037, "loss": 0.6876179873943329, "time": 0.41797921657562254, "epoch": 11, "memory": 20334, "step": 61780} +{"lr": 0.0005050000000000001, "data_time": 0.0015399932861328125, "grad_norm": 0.01483668175060302, "loss": 0.6902819156646729, "time": 0.41352152824401855, "epoch": 11, "memory": 20334, "step": 61880} +{"lr": 0.0005050000000000001, "data_time": 0.0017910480499267578, "grad_norm": 0.02472144942730665, "loss": 0.6849947333335876, "time": 0.4193438529968262, "epoch": 11, "memory": 20334, "step": 61980} +{"lr": 0.0005050000000000001, "data_time": 0.0018019437789916991, "grad_norm": 0.021300900680944323, "loss": 0.6857230484485626, "time": 0.3970093488693237, "epoch": 11, "memory": 20334, "step": 62080} +{"lr": 0.0005050000000000001, "data_time": 0.0017255544662475586, "grad_norm": 0.01563761797733605, "loss": 0.6862986862659455, "time": 0.4347070217132568, "epoch": 11, "memory": 20334, "step": 62180} +{"lr": 0.0005050000000000001, "data_time": 0.0018317222595214844, "grad_norm": 0.0190139097860083, "loss": 0.6858514189720154, "time": 0.39530124664306643, "epoch": 11, "memory": 20334, "step": 62280} +{"lr": 0.0005050000000000001, "data_time": 0.0018554449081420899, "grad_norm": 0.02175544567871839, "loss": 0.6858910858631134, "time": 0.42312207221984866, "epoch": 11, "memory": 20334, "step": 62380} +{"lr": 0.0005050000000000001, "data_time": 0.0017745256423950194, "grad_norm": 0.014436323108384386, "loss": 0.6903681099414826, "time": 0.3894811153411865, "epoch": 11, "memory": 20334, "step": 62480} +{"lr": 0.0005050000000000001, "data_time": 0.002657032012939453, "grad_norm": 0.01334753039991483, "loss": 0.6810441672801971, "time": 0.3817819833755493, "epoch": 11, "memory": 20334, "step": 62580} +{"lr": 0.0005050000000000001, "data_time": 0.0017103433609008789, "grad_norm": 0.01473056636750698, "loss": 0.6874148428440094, "time": 0.3831214189529419, "epoch": 11, "memory": 20334, "step": 62680} +{"lr": 0.0005050000000000001, "data_time": 0.002344846725463867, "grad_norm": 0.013010151410708205, "loss": 0.6864040493965149, "time": 0.3922775983810425, "epoch": 11, "memory": 20334, "step": 62780} +{"lr": 0.0005050000000000001, "data_time": 0.0015987396240234376, "grad_norm": 0.016429151548072694, "loss": 0.6879156172275543, "time": 0.40706620216369627, "epoch": 11, "memory": 20334, "step": 62880} +{"lr": 0.0005050000000000001, "data_time": 0.0017150402069091796, "grad_norm": 0.017281638015992938, "loss": 0.6896399796009064, "time": 0.3822291851043701, "epoch": 11, "memory": 20334, "step": 62980} +{"lr": 0.0005050000000000001, "data_time": 0.0017846107482910156, "grad_norm": 0.017714999988675116, "loss": 0.6918750524520874, "time": 0.40033955574035646, "epoch": 11, "memory": 20334, "step": 63080} +{"lr": 0.0005050000000000001, "data_time": 0.0018810987472534179, "grad_norm": 0.01451815478503704, "loss": 0.687272983789444, "time": 0.37504868507385253, "epoch": 11, "memory": 20334, "step": 63180} +{"lr": 0.0005050000000000001, "data_time": 0.0016415119171142578, "grad_norm": 0.01824620590778068, "loss": 0.6873231112957001, "time": 0.42091591358184816, "epoch": 11, "memory": 20334, "step": 63280} +{"accuracy/top1": 100.0, "data_time": 0.362206326590644, "time": 0.9878199895222982, "step": 11} +{"lr": 0.000427564939805086, "data_time": 0.001617121696472168, "grad_norm": 0.0166913520777598, "loss": 0.6899495005607605, "time": 0.417389440536499, "epoch": 12, "memory": 20334, "step": 63438} +{"lr": 0.000427564939805086, "data_time": 0.0016464471817016601, "grad_norm": 0.020788821484893562, "loss": 0.6836094558238983, "time": 0.39636070728302003, "epoch": 12, "memory": 20334, "step": 63538} +{"lr": 0.000427564939805086, "data_time": 0.0018493175506591798, "grad_norm": 0.015353060129564255, "loss": 0.6920715034008026, "time": 0.37181313037872316, "epoch": 12, "memory": 20334, "step": 63638} +{"lr": 0.000427564939805086, "data_time": 0.0017371416091918946, "grad_norm": 0.013098863716004416, "loss": 0.6874368488788605, "time": 0.39792301654815676, "epoch": 12, "memory": 20334, "step": 63738} +{"lr": 0.000427564939805086, "data_time": 0.0014843225479125976, "grad_norm": 0.01581031907116994, "loss": 0.6895247638225556, "time": 0.39028539657592776, "epoch": 12, "memory": 20334, "step": 63838} +{"lr": 0.000427564939805086, "data_time": 0.0016892433166503906, "grad_norm": 0.014902688539586961, "loss": 0.6877779304981232, "time": 0.3751551151275635, "epoch": 12, "memory": 20334, "step": 63938} +{"lr": 0.000427564939805086, "data_time": 0.001846170425415039, "grad_norm": 0.014566584955900908, "loss": 0.689047884941101, "time": 0.4367671966552734, "epoch": 12, "memory": 20334, "step": 64038} +{"lr": 0.000427564939805086, "data_time": 0.0017380237579345704, "grad_norm": 0.013999722059816122, "loss": 0.6841715455055237, "time": 0.4049127817153931, "epoch": 12, "memory": 20334, "step": 64138} +{"lr": 0.000427564939805086, "data_time": 0.0029029130935668947, "grad_norm": 0.017001294903457163, "loss": 0.68236603140831, "time": 0.3810136318206787, "epoch": 12, "memory": 20334, "step": 64238} +{"lr": 0.000427564939805086, "data_time": 0.001759958267211914, "grad_norm": 0.011180007155053318, "loss": 0.6864933133125305, "time": 0.3714949607849121, "epoch": 12, "memory": 20334, "step": 64338} +{"lr": 0.000427564939805086, "data_time": 0.0016855239868164063, "grad_norm": 0.0156432211631909, "loss": 0.6853608846664428, "time": 0.40211546421051025, "epoch": 12, "memory": 20334, "step": 64438} +{"lr": 0.000427564939805086, "data_time": 0.002300453186035156, "grad_norm": 0.018553055147640406, "loss": 0.6856110215187072, "time": 0.41217448711395266, "epoch": 12, "memory": 20334, "step": 64538} +{"lr": 0.000427564939805086, "data_time": 0.0014149904251098632, "grad_norm": 0.019725331536028533, "loss": 0.6861101984977722, "time": 0.41823592185974123, "epoch": 12, "memory": 20334, "step": 64638} +{"lr": 0.000427564939805086, "data_time": 0.0014590024948120117, "grad_norm": 0.015236181626096367, "loss": 0.6855660200119018, "time": 0.38270413875579834, "epoch": 12, "memory": 20334, "step": 64738} +{"lr": 0.000427564939805086, "data_time": 0.0014870166778564453, "grad_norm": 0.019289453083183618, "loss": 0.6875917494297028, "time": 0.4137756109237671, "epoch": 12, "memory": 20334, "step": 64838} +{"lr": 0.000427564939805086, "data_time": 0.002022981643676758, "grad_norm": 0.013446417148225009, "loss": 0.6886761307716369, "time": 0.3940241813659668, "epoch": 12, "memory": 20334, "step": 64938} +{"lr": 0.000427564939805086, "data_time": 0.0019438743591308593, "grad_norm": 0.023530560499057173, "loss": 0.6858498275279998, "time": 0.4304816484451294, "epoch": 12, "memory": 20334, "step": 65038} +{"lr": 0.000427564939805086, "data_time": 0.0018252134323120117, "grad_norm": 0.016393794189207257, "loss": 0.6866360604763031, "time": 0.408117938041687, "epoch": 12, "memory": 20334, "step": 65138} +{"lr": 0.000427564939805086, "data_time": 0.0020288467407226563, "grad_norm": 0.017317197937518358, "loss": 0.6868013143539429, "time": 0.41807188987731936, "epoch": 12, "memory": 20334, "step": 65238} +{"lr": 0.000427564939805086, "data_time": 0.001623225212097168, "grad_norm": 0.016890868052723818, "loss": 0.6861487030982971, "time": 0.40656893253326415, "epoch": 12, "memory": 20334, "step": 65338} +{"lr": 0.000427564939805086, "data_time": 0.0015668630599975585, "grad_norm": 0.021100576501339674, "loss": 0.6913497745990753, "time": 0.39169039726257326, "epoch": 12, "memory": 20334, "step": 65438} +{"lr": 0.000427564939805086, "data_time": 0.0016763925552368164, "grad_norm": 0.016157165495678784, "loss": 0.6850412070751191, "time": 0.4171649217605591, "epoch": 12, "memory": 20334, "step": 65538} +{"lr": 0.000427564939805086, "data_time": 0.0018799543380737306, "grad_norm": 0.012369199434760957, "loss": 0.6903493583202363, "time": 0.3820366144180298, "epoch": 12, "memory": 20334, "step": 65638} +{"lr": 0.000427564939805086, "data_time": 0.001929020881652832, "grad_norm": 0.022135058417916298, "loss": 0.6847064614295959, "time": 0.3993083477020264, "epoch": 12, "memory": 20334, "step": 65738} +{"lr": 0.000427564939805086, "data_time": 0.002015423774719238, "grad_norm": 0.012583296530647204, "loss": 0.6927791237831116, "time": 0.4088080406188965, "epoch": 12, "memory": 20334, "step": 65838} +{"lr": 0.000427564939805086, "data_time": 0.0018355131149291992, "grad_norm": 0.01793795293197036, "loss": 0.6878406882286072, "time": 0.3914525508880615, "epoch": 12, "memory": 20334, "step": 65938} +{"lr": 0.000427564939805086, "data_time": 0.002055716514587402, "grad_norm": 0.017172524356283247, "loss": 0.6883574783802032, "time": 0.39122986793518066, "epoch": 12, "memory": 20334, "step": 66038} +{"lr": 0.000427564939805086, "data_time": 0.0017808437347412109, "grad_norm": 0.018730479036457838, "loss": 0.6886959552764893, "time": 0.40156447887420654, "epoch": 12, "memory": 20334, "step": 66138} +{"lr": 0.000427564939805086, "data_time": 0.0017096281051635742, "grad_norm": 0.015269418933894486, "loss": 0.6835977256298065, "time": 0.3667266607284546, "epoch": 12, "memory": 20334, "step": 66238} +{"lr": 0.000427564939805086, "data_time": 0.0017436027526855468, "grad_norm": 0.012353344610892237, "loss": 0.689087700843811, "time": 0.4098337411880493, "epoch": 12, "memory": 20334, "step": 66338} +{"lr": 0.000427564939805086, "data_time": 0.002637624740600586, "grad_norm": 0.0182646015426144, "loss": 0.6834400355815887, "time": 0.413388991355896, "epoch": 12, "memory": 20334, "step": 66438} +{"lr": 0.000427564939805086, "data_time": 0.0018443584442138672, "grad_norm": 0.021364059206098317, "loss": 0.6865598857402802, "time": 0.40879254341125487, "epoch": 12, "memory": 20334, "step": 66538} +{"lr": 0.000427564939805086, "data_time": 0.0018166542053222657, "grad_norm": 0.017236367636360227, "loss": 0.6873547673225403, "time": 0.4330552339553833, "epoch": 12, "memory": 20334, "step": 66638} +{"lr": 0.000427564939805086, "data_time": 0.0015700340270996093, "grad_norm": 0.010844798694597558, "loss": 0.6891121685504913, "time": 0.4540198802947998, "epoch": 12, "memory": 20334, "step": 66738} +{"lr": 0.000427564939805086, "data_time": 0.0018373250961303711, "grad_norm": 0.015047562075778842, "loss": 0.6787995755672455, "time": 0.4149184226989746, "epoch": 12, "memory": 20334, "step": 66838} +{"lr": 0.000427564939805086, "data_time": 0.0017477989196777344, "grad_norm": 0.01790905058151111, "loss": 0.6833564043045044, "time": 0.42359721660614014, "epoch": 12, "memory": 20334, "step": 66938} +{"lr": 0.000427564939805086, "data_time": 0.0016760826110839844, "grad_norm": 0.010735026933252812, "loss": 0.689172500371933, "time": 0.4061639070510864, "epoch": 12, "memory": 20334, "step": 67038} +{"lr": 0.000427564939805086, "data_time": 0.0018133163452148438, "grad_norm": 0.016866380511783064, "loss": 0.6832768201828003, "time": 0.4003861904144287, "epoch": 12, "memory": 20334, "step": 67138} +{"lr": 0.000427564939805086, "data_time": 0.0018765926361083984, "grad_norm": 0.019166821404360233, "loss": 0.6863247394561768, "time": 0.4241545915603638, "epoch": 12, "memory": 20334, "step": 67238} +{"lr": 0.000427564939805086, "data_time": 0.001663970947265625, "grad_norm": 0.018017438339302316, "loss": 0.6898203909397125, "time": 0.4237964630126953, "epoch": 12, "memory": 20334, "step": 67338} +{"lr": 0.000427564939805086, "data_time": 0.0018258333206176759, "grad_norm": 0.017352794483304023, "loss": 0.684254378080368, "time": 0.3921781539916992, "epoch": 12, "memory": 20334, "step": 67438} +{"lr": 0.000427564939805086, "data_time": 0.0017937183380126952, "grad_norm": 0.016346360580064355, "loss": 0.6855483055114746, "time": 0.462936544418335, "epoch": 12, "memory": 20334, "step": 67538} +{"lr": 0.000427564939805086, "data_time": 0.0019580602645874025, "grad_norm": 0.00908320113667287, "loss": 0.6874360084533692, "time": 0.4015927791595459, "epoch": 12, "memory": 20334, "step": 67638} +{"lr": 0.000427564939805086, "data_time": 0.0017687320709228516, "grad_norm": 0.018595830723643302, "loss": 0.6868012547492981, "time": 0.3768711805343628, "epoch": 12, "memory": 20334, "step": 67738} +{"lr": 0.000427564939805086, "data_time": 0.0017295122146606446, "grad_norm": 0.01453996414784342, "loss": 0.6850512862205506, "time": 0.4595906972885132, "epoch": 12, "memory": 20334, "step": 67838} +{"lr": 0.000427564939805086, "data_time": 0.0018616676330566405, "grad_norm": 0.020302220573648812, "loss": 0.6899466037750244, "time": 0.38662493228912354, "epoch": 12, "memory": 20334, "step": 67938} +{"lr": 0.000427564939805086, "data_time": 0.0016736030578613282, "grad_norm": 0.024154767947038636, "loss": 0.6864497721195221, "time": 0.39399864673614504, "epoch": 12, "memory": 20334, "step": 68038} +{"lr": 0.000427564939805086, "data_time": 0.0018260717391967774, "grad_norm": 0.01718618874438107, "loss": 0.6876361727714538, "time": 0.4071578741073608, "epoch": 12, "memory": 20334, "step": 68138} +{"lr": 0.000427564939805086, "data_time": 0.0021019935607910155, "grad_norm": 0.02005452015437186, "loss": 0.6836853265762329, "time": 0.4368730068206787, "epoch": 12, "memory": 20334, "step": 68238} +{"lr": 0.000427564939805086, "data_time": 0.0017063140869140625, "grad_norm": 0.011229679710231722, "loss": 0.6864103615283966, "time": 0.4182861328125, "epoch": 12, "memory": 20334, "step": 68338} +{"lr": 0.000427564939805086, "data_time": 0.001926589012145996, "grad_norm": 0.01349624409340322, "loss": 0.6893961191177368, "time": 0.40487518310546877, "epoch": 12, "memory": 20334, "step": 68438} +{"lr": 0.000427564939805086, "data_time": 0.0017902612686157226, "grad_norm": 0.013032496441155672, "loss": 0.6877645134925843, "time": 0.38978354930877684, "epoch": 12, "memory": 20334, "step": 68538} +{"lr": 0.000427564939805086, "data_time": 0.0018377065658569335, "grad_norm": 0.014521173760294914, "loss": 0.6885558724403381, "time": 0.3954671621322632, "epoch": 12, "memory": 20334, "step": 68638} +{"lr": 0.000427564939805086, "data_time": 0.001640653610229492, "grad_norm": 0.02059962465427816, "loss": 0.690975034236908, "time": 0.3865856885910034, "epoch": 12, "memory": 20334, "step": 68738} +{"lr": 0.000427564939805086, "data_time": 0.002003645896911621, "grad_norm": 0.015047602856066078, "loss": 0.6892555713653564, "time": 0.42522027492523196, "epoch": 12, "memory": 20334, "step": 68838} +{"lr": 0.000427564939805086, "data_time": 0.0020319461822509766, "grad_norm": 0.015423443308100105, "loss": 0.6877327144145966, "time": 0.4105898141860962, "epoch": 12, "memory": 20334, "step": 68938} +{"lr": 0.000427564939805086, "data_time": 0.0016184329986572265, "grad_norm": 0.014972370746545494, "loss": 0.682448822259903, "time": 0.3690106153488159, "epoch": 12, "memory": 20334, "step": 69038} +{"accuracy/top1": 100.0, "data_time": 0.35881156391567653, "time": 0.9850935935974121, "step": 12} +{"lr": 0.0003520365877844012, "data_time": 0.002499818801879883, "grad_norm": 0.01408783111255616, "loss": 0.686473298072815, "time": 0.42690041065216067, "epoch": 13, "memory": 20334, "step": 69196} +{"lr": 0.0003520365877844012, "data_time": 0.0016981601715087891, "grad_norm": 0.010116408905014395, "loss": 0.6877528607845307, "time": 0.40020408630371096, "epoch": 13, "memory": 20334, "step": 69296} +{"lr": 0.0003520365877844012, "data_time": 0.0014430046081542968, "grad_norm": 0.009993621555622667, "loss": 0.686174887418747, "time": 0.4011565923690796, "epoch": 13, "memory": 20334, "step": 69396} +{"lr": 0.0003520365877844012, "data_time": 0.0024167299270629883, "grad_norm": 0.012839860911481083, "loss": 0.6864809036254883, "time": 0.3791086196899414, "epoch": 13, "memory": 20334, "step": 69496} +{"lr": 0.0003520365877844012, "data_time": 0.001960611343383789, "grad_norm": 0.017129956278949976, "loss": 0.6864925742149353, "time": 0.42064590454101564, "epoch": 13, "memory": 20334, "step": 69596} +{"lr": 0.0003520365877844012, "data_time": 0.0022573947906494142, "grad_norm": 0.009476678224746137, "loss": 0.6879218339920044, "time": 0.40629425048828127, "epoch": 13, "memory": 20334, "step": 69696} +{"lr": 0.0003520365877844012, "data_time": 0.0020205497741699217, "grad_norm": 0.02477117571979761, "loss": 0.6804328858852386, "time": 0.4160814046859741, "epoch": 13, "memory": 20334, "step": 69796} +{"lr": 0.0003520365877844012, "data_time": 0.002030086517333984, "grad_norm": 0.013945743715157732, "loss": 0.6856885552406311, "time": 0.37984271049499513, "epoch": 13, "memory": 20334, "step": 69896} +{"lr": 0.0003520365877844012, "data_time": 0.001855158805847168, "grad_norm": 0.017967581457924098, "loss": 0.6900890290737152, "time": 0.41427340507507326, "epoch": 13, "memory": 20334, "step": 69996} +{"lr": 0.0003520365877844012, "data_time": 0.001960158348083496, "grad_norm": 0.014855450368486345, "loss": 0.6853561997413635, "time": 0.3758256196975708, "epoch": 13, "memory": 20334, "step": 70096} +{"lr": 0.0003520365877844012, "data_time": 0.0017795562744140625, "grad_norm": 0.015406798385083676, "loss": 0.6900947272777558, "time": 0.40915162563323976, "epoch": 13, "memory": 20334, "step": 70196} +{"lr": 0.0003520365877844012, "data_time": 0.0017820358276367187, "grad_norm": 0.009939678665250539, "loss": 0.686232042312622, "time": 0.40437347888946534, "epoch": 13, "memory": 20334, "step": 70296} +{"lr": 0.0003520365877844012, "data_time": 0.0017084360122680664, "grad_norm": 0.015425244870129973, "loss": 0.6870243191719055, "time": 0.3940648078918457, "epoch": 13, "memory": 20334, "step": 70396} +{"lr": 0.0003520365877844012, "data_time": 0.00237889289855957, "grad_norm": 0.022950517875142396, "loss": 0.6841873228549957, "time": 0.396422815322876, "epoch": 13, "memory": 20334, "step": 70496} +{"lr": 0.0003520365877844012, "data_time": 0.001589202880859375, "grad_norm": 0.01524299723096192, "loss": 0.6911931157112121, "time": 0.39834606647491455, "epoch": 13, "memory": 20334, "step": 70596} +{"lr": 0.0003520365877844012, "data_time": 0.0016993999481201172, "grad_norm": 0.011118559376336634, "loss": 0.687118124961853, "time": 0.3648622274398804, "epoch": 13, "memory": 20334, "step": 70696} +{"lr": 0.0003520365877844012, "data_time": 0.0017743349075317384, "grad_norm": 0.018272021901793778, "loss": 0.6872730374336242, "time": 0.3884437084197998, "epoch": 13, "memory": 20334, "step": 70796} +{"lr": 0.0003520365877844012, "data_time": 0.0017374753952026367, "grad_norm": 0.00859234919771552, "loss": 0.6863044977188111, "time": 0.40339174270629885, "epoch": 13, "memory": 20334, "step": 70896} +{"lr": 0.0003520365877844012, "data_time": 0.0016998767852783204, "grad_norm": 0.01648919393774122, "loss": 0.6863160371780396, "time": 0.39385557174682617, "epoch": 13, "memory": 20334, "step": 70996} +{"lr": 0.0003520365877844012, "data_time": 0.0017043828964233398, "grad_norm": 0.013088628451805561, "loss": 0.6838694214820862, "time": 0.4203041076660156, "epoch": 13, "memory": 20334, "step": 71096} +{"lr": 0.0003520365877844012, "data_time": 0.0020523309707641602, "grad_norm": 0.020921247568912803, "loss": 0.6896158695220947, "time": 0.42620530128479006, "epoch": 13, "memory": 20334, "step": 71196} +{"lr": 0.0003520365877844012, "data_time": 0.0018886327743530273, "grad_norm": 0.01668210669886321, "loss": 0.6838137149810791, "time": 0.39089982509613036, "epoch": 13, "memory": 20334, "step": 71296} +{"lr": 0.0003520365877844012, "data_time": 0.0019776582717895507, "grad_norm": 0.020049914065748454, "loss": 0.6866389572620392, "time": 0.39708170890808103, "epoch": 13, "memory": 20334, "step": 71396} +{"lr": 0.0003520365877844012, "data_time": 0.0019255399703979493, "grad_norm": 0.011692477995529771, "loss": 0.685070377588272, "time": 0.40939884185791015, "epoch": 13, "memory": 20334, "step": 71496} +{"lr": 0.0003520365877844012, "data_time": 0.0024569034576416016, "grad_norm": 0.014183701551519334, "loss": 0.6869688928127289, "time": 0.38482298851013186, "epoch": 13, "memory": 20334, "step": 71596} +{"lr": 0.0003520365877844012, "data_time": 0.0018474102020263673, "grad_norm": 0.017426353041082622, "loss": 0.6850690066814422, "time": 0.4538242340087891, "epoch": 13, "memory": 20334, "step": 71696} +{"lr": 0.0003520365877844012, "data_time": 0.0017541885375976563, "grad_norm": 0.016810492976219393, "loss": 0.6873281717300415, "time": 0.3970121622085571, "epoch": 13, "memory": 20334, "step": 71796} +{"lr": 0.0003520365877844012, "data_time": 0.0013822078704833984, "grad_norm": 0.01345356865786016, "loss": 0.6868139803409576, "time": 0.40201380252838137, "epoch": 13, "memory": 20334, "step": 71896} +{"lr": 0.0003520365877844012, "data_time": 0.0013040542602539063, "grad_norm": 0.015152580151334406, "loss": 0.683145421743393, "time": 0.39615478515625, "epoch": 13, "memory": 20334, "step": 71996} +{"lr": 0.0003520365877844012, "data_time": 0.002103257179260254, "grad_norm": 0.015875039622187614, "loss": 0.6846316158771515, "time": 0.4416754722595215, "epoch": 13, "memory": 20334, "step": 72096} +{"lr": 0.0003520365877844012, "data_time": 0.0019967317581176757, "grad_norm": 0.013362546951975673, "loss": 0.6818894207477569, "time": 0.42512590885162355, "epoch": 13, "memory": 20334, "step": 72196} +{"lr": 0.0003520365877844012, "data_time": 0.0018580436706542968, "grad_norm": 0.018948213150724768, "loss": 0.685666698217392, "time": 0.40195112228393554, "epoch": 13, "memory": 20334, "step": 72296} +{"lr": 0.0003520365877844012, "data_time": 0.0016201972961425782, "grad_norm": 0.011216734326444567, "loss": 0.6902041018009186, "time": 0.4200406789779663, "epoch": 13, "memory": 20334, "step": 72396} +{"lr": 0.0003520365877844012, "data_time": 0.0016074657440185546, "grad_norm": 0.015482333645923062, "loss": 0.6882450044155121, "time": 0.38861067295074464, "epoch": 13, "memory": 20334, "step": 72496} +{"lr": 0.0003520365877844012, "data_time": 0.0019302129745483398, "grad_norm": 0.018153125431854277, "loss": 0.6857194900512695, "time": 0.4287428617477417, "epoch": 13, "memory": 20334, "step": 72596} +{"lr": 0.0003520365877844012, "data_time": 0.0017689228057861327, "grad_norm": 0.016061562206596137, "loss": 0.6880880415439605, "time": 0.4203546762466431, "epoch": 13, "memory": 20334, "step": 72696} +{"lr": 0.0003520365877844012, "data_time": 0.001614546775817871, "grad_norm": 0.01742428051074967, "loss": 0.6847258329391479, "time": 0.47726438045501707, "epoch": 13, "memory": 20334, "step": 72796} +{"lr": 0.0003520365877844012, "data_time": 0.001859736442565918, "grad_norm": 0.018289669835940003, "loss": 0.6820207595825195, "time": 0.42640645503997804, "epoch": 13, "memory": 20334, "step": 72896} +{"lr": 0.0003520365877844012, "data_time": 0.0017515659332275391, "grad_norm": 0.016496698174159975, "loss": 0.6912059962749482, "time": 0.41557989120483396, "epoch": 13, "memory": 20334, "step": 72996} +{"lr": 0.0003520365877844012, "data_time": 0.001747441291809082, "grad_norm": 0.019221892394125462, "loss": 0.6915319621562958, "time": 0.41441380977630615, "epoch": 13, "memory": 20334, "step": 73096} +{"lr": 0.0003520365877844012, "data_time": 0.0016184091567993165, "grad_norm": 0.015028754225932061, "loss": 0.6839554131031036, "time": 0.42982704639434816, "epoch": 13, "memory": 20334, "step": 73196} +{"lr": 0.0003520365877844012, "data_time": 0.0016668319702148437, "grad_norm": 0.014722906588576733, "loss": 0.6887838482856751, "time": 0.4043821096420288, "epoch": 13, "memory": 20334, "step": 73296} +{"lr": 0.0003520365877844012, "data_time": 0.0016119956970214843, "grad_norm": 0.013450360629940405, "loss": 0.6861825823783875, "time": 0.4213541030883789, "epoch": 13, "memory": 20334, "step": 73396} +{"lr": 0.0003520365877844012, "data_time": 0.0018059968948364257, "grad_norm": 0.015405056823510677, "loss": 0.6865025579929351, "time": 0.4439688682556152, "epoch": 13, "memory": 20334, "step": 73496} +{"lr": 0.0003520365877844012, "data_time": 0.0015944719314575195, "grad_norm": 0.013549680833239108, "loss": 0.6879826962947846, "time": 0.40688927173614503, "epoch": 13, "memory": 20334, "step": 73596} +{"lr": 0.0003520365877844012, "data_time": 0.0017857074737548829, "grad_norm": 0.014822281093802302, "loss": 0.6874355852603913, "time": 0.39173383712768556, "epoch": 13, "memory": 20334, "step": 73696} +{"lr": 0.0003520365877844012, "data_time": 0.0019335746765136719, "grad_norm": 0.01581891911337152, "loss": 0.6909354984760284, "time": 0.38756699562072755, "epoch": 13, "memory": 20334, "step": 73796} +{"lr": 0.0003520365877844012, "data_time": 0.0016683340072631836, "grad_norm": 0.015744145837379618, "loss": 0.6887045919895172, "time": 0.388281512260437, "epoch": 13, "memory": 20334, "step": 73896} +{"lr": 0.0003520365877844012, "data_time": 0.0018504858016967773, "grad_norm": 0.010446198028512299, "loss": 0.6860404908657074, "time": 0.39574370384216306, "epoch": 13, "memory": 20334, "step": 73996} +{"lr": 0.0003520365877844012, "data_time": 0.0015665531158447266, "grad_norm": 0.017069025873206557, "loss": 0.6907217621803283, "time": 0.4279362440109253, "epoch": 13, "memory": 20334, "step": 74096} +{"lr": 0.0003520365877844012, "data_time": 0.0017962932586669921, "grad_norm": 0.013988114194944501, "loss": 0.6830148696899414, "time": 0.40684354305267334, "epoch": 13, "memory": 20334, "step": 74196} +{"lr": 0.0003520365877844012, "data_time": 0.0016621828079223632, "grad_norm": 0.016329382685944437, "loss": 0.6807142078876496, "time": 0.3933048486709595, "epoch": 13, "memory": 20334, "step": 74296} +{"lr": 0.0003520365877844012, "data_time": 0.001727461814880371, "grad_norm": 0.012931671366095543, "loss": 0.6895102798938751, "time": 0.3785771608352661, "epoch": 13, "memory": 20334, "step": 74396} +{"lr": 0.0003520365877844012, "data_time": 0.0019137144088745117, "grad_norm": 0.012048048479482532, "loss": 0.6844245970249176, "time": 0.3915132522583008, "epoch": 13, "memory": 20334, "step": 74496} +{"lr": 0.0003520365877844012, "data_time": 0.0018544197082519531, "grad_norm": 0.019758155196905137, "loss": 0.6845782697200775, "time": 0.4117830514907837, "epoch": 13, "memory": 20334, "step": 74596} +{"lr": 0.0003520365877844012, "data_time": 0.002025890350341797, "grad_norm": 0.0192345077637583, "loss": 0.6796214640140533, "time": 0.4155980348587036, "epoch": 13, "memory": 20334, "step": 74696} +{"lr": 0.0003520365877844012, "data_time": 0.002515244483947754, "grad_norm": 0.01954756546765566, "loss": 0.6879429399967194, "time": 0.40151567459106446, "epoch": 13, "memory": 20334, "step": 74796} +{"accuracy/top1": 100.0, "data_time": 0.3618520365820991, "time": 0.9855987760755751, "step": 13} +{"lr": 0.00028027470262892445, "data_time": 0.0015807390213012696, "grad_norm": 0.011704668845050037, "loss": 0.6839596807956696, "time": 0.4051476716995239, "epoch": 14, "memory": 20334, "step": 74954} +{"lr": 0.00028027470262892445, "data_time": 0.0017504453659057616, "grad_norm": 0.018872018868569283, "loss": 0.6887854754924774, "time": 0.40489964485168456, "epoch": 14, "memory": 20334, "step": 75054} +{"lr": 0.00028027470262892445, "data_time": 0.0016129255294799806, "grad_norm": 0.014565573446452618, "loss": 0.6866371989250183, "time": 0.41884598731994627, "epoch": 14, "memory": 20334, "step": 75154} +{"lr": 0.00028027470262892445, "data_time": 0.001651167869567871, "grad_norm": 0.01385259844828397, "loss": 0.6840680539608002, "time": 0.40360150337219236, "epoch": 14, "memory": 20334, "step": 75254} +{"lr": 0.00028027470262892445, "data_time": 0.001611018180847168, "grad_norm": 0.01756176436319947, "loss": 0.6900853455066681, "time": 0.4007642984390259, "epoch": 14, "memory": 20334, "step": 75354} +{"lr": 0.00028027470262892445, "data_time": 0.0017755746841430664, "grad_norm": 0.016391893127001823, "loss": 0.6842569053173065, "time": 0.37480669021606444, "epoch": 14, "memory": 20334, "step": 75454} +{"lr": 0.00028027470262892445, "data_time": 0.0017299652099609375, "grad_norm": 0.014681549975648522, "loss": 0.6853827178478241, "time": 0.39731364250183104, "epoch": 14, "memory": 20334, "step": 75554} +{"lr": 0.00028027470262892445, "data_time": 0.0020262956619262694, "grad_norm": 0.01204390093917027, "loss": 0.6880824983119964, "time": 0.39176321029663086, "epoch": 14, "memory": 20334, "step": 75654} +{"lr": 0.00028027470262892445, "data_time": 0.001528453826904297, "grad_norm": 0.012430070911068469, "loss": 0.6872732996940613, "time": 0.4073626518249512, "epoch": 14, "memory": 20334, "step": 75754} +{"lr": 0.00028027470262892445, "data_time": 0.0016721010208129883, "grad_norm": 0.012954751891084016, "loss": 0.6828380167484284, "time": 0.4160911083221436, "epoch": 14, "memory": 20334, "step": 75854} +{"lr": 0.00028027470262892445, "data_time": 0.0016779184341430664, "grad_norm": 0.018801445607095957, "loss": 0.6833130300045014, "time": 0.4033613920211792, "epoch": 14, "memory": 20334, "step": 75954} +{"lr": 0.00028027470262892445, "data_time": 0.0014740943908691407, "grad_norm": 0.013314147177152336, "loss": 0.6858446300029755, "time": 0.40171449184417723, "epoch": 14, "memory": 20334, "step": 76054} +{"lr": 0.00028027470262892445, "data_time": 0.0016918420791625977, "grad_norm": 0.021815078053623437, "loss": 0.6842347383499146, "time": 0.38399443626403806, "epoch": 14, "memory": 20334, "step": 76154} +{"lr": 0.00028027470262892445, "data_time": 0.002535080909729004, "grad_norm": 0.021698733745142817, "loss": 0.6863724827766419, "time": 0.3649530649185181, "epoch": 14, "memory": 20334, "step": 76254} +{"lr": 0.00028027470262892445, "data_time": 0.0022377729415893554, "grad_norm": 0.017168973269872368, "loss": 0.6841258049011231, "time": 0.39482784271240234, "epoch": 14, "memory": 20334, "step": 76354} +{"lr": 0.00028027470262892445, "data_time": 0.00183560848236084, "grad_norm": 0.016725882625905797, "loss": 0.6881155669689178, "time": 0.41203043460845945, "epoch": 14, "memory": 20334, "step": 76454} +{"lr": 0.00028027470262892445, "data_time": 0.0018941640853881836, "grad_norm": 0.01606611717143096, "loss": 0.6885165214538574, "time": 0.44018542766571045, "epoch": 14, "memory": 20334, "step": 76554} +{"lr": 0.00028027470262892445, "data_time": 0.002838468551635742, "grad_norm": 0.016217493638396262, "loss": 0.684929758310318, "time": 0.37801151275634765, "epoch": 14, "memory": 20334, "step": 76654} +{"lr": 0.00028027470262892445, "data_time": 0.0017352342605590821, "grad_norm": 0.015280984248965979, "loss": 0.6845894157886505, "time": 0.41396141052246094, "epoch": 14, "memory": 20334, "step": 76754} +{"lr": 0.00028027470262892445, "data_time": 0.001906132698059082, "grad_norm": 0.021990169165655972, "loss": 0.6852270126342773, "time": 0.38214874267578125, "epoch": 14, "memory": 20334, "step": 76854} +{"lr": 0.00028027470262892445, "data_time": 0.0031803131103515627, "grad_norm": 0.016700853826478124, "loss": 0.6877544343471527, "time": 0.4418198585510254, "epoch": 14, "memory": 20334, "step": 76954} +{"lr": 0.00028027470262892445, "data_time": 0.001935577392578125, "grad_norm": 0.01613506133435294, "loss": 0.6831979990005493, "time": 0.3728139638900757, "epoch": 14, "memory": 20334, "step": 77054} +{"lr": 0.00028027470262892445, "data_time": 0.0020687103271484373, "grad_norm": 0.017249015718698503, "loss": 0.6869511902332306, "time": 0.386832594871521, "epoch": 14, "memory": 20334, "step": 77154} +{"lr": 0.00028027470262892445, "data_time": 0.001937556266784668, "grad_norm": 0.01789760193787515, "loss": 0.6842350125312805, "time": 0.3791035652160645, "epoch": 14, "memory": 20334, "step": 77254} +{"lr": 0.00028027470262892445, "data_time": 0.002295565605163574, "grad_norm": 0.016737726907012983, "loss": 0.6866466999053955, "time": 0.41111068725585936, "epoch": 14, "memory": 20334, "step": 77354} +{"lr": 0.00028027470262892445, "data_time": 0.002791619300842285, "grad_norm": 0.016068597487173975, "loss": 0.6885709464550018, "time": 0.4092714309692383, "epoch": 14, "memory": 20334, "step": 77454} +{"lr": 0.00028027470262892445, "data_time": 0.0019040584564208984, "grad_norm": 0.014783918834291398, "loss": 0.6855606198310852, "time": 0.3963749647140503, "epoch": 14, "memory": 20334, "step": 77554} +{"lr": 0.00028027470262892445, "data_time": 0.0022172927856445312, "grad_norm": 0.012944332987535745, "loss": 0.6867958188056946, "time": 0.40024442672729493, "epoch": 14, "memory": 20334, "step": 77654} +{"lr": 0.00028027470262892445, "data_time": 0.001626729965209961, "grad_norm": 0.019447324564680457, "loss": 0.6816729784011841, "time": 0.4012739896774292, "epoch": 14, "memory": 20334, "step": 77754} +{"lr": 0.00028027470262892445, "data_time": 0.0030068159103393555, "grad_norm": 0.011175620427820832, "loss": 0.680833488702774, "time": 0.3890573024749756, "epoch": 14, "memory": 20334, "step": 77854} +{"lr": 0.00028027470262892445, "data_time": 0.0022169828414916994, "grad_norm": 0.013994936249218881, "loss": 0.6875779330730438, "time": 0.3894374847412109, "epoch": 14, "memory": 20334, "step": 77954} +{"lr": 0.00028027470262892445, "data_time": 0.0018175125122070312, "grad_norm": 0.014813157729804516, "loss": 0.6880999803543091, "time": 0.4051444292068481, "epoch": 14, "memory": 20334, "step": 78054} +{"lr": 0.00028027470262892445, "data_time": 0.0026232719421386717, "grad_norm": 0.016581385885365306, "loss": 0.6841295242309571, "time": 0.44101567268371583, "epoch": 14, "memory": 20334, "step": 78154} +{"lr": 0.00028027470262892445, "data_time": 0.0017164230346679687, "grad_norm": 0.021206063264980913, "loss": 0.688108503818512, "time": 0.39741859436035154, "epoch": 14, "memory": 20334, "step": 78254} +{"lr": 0.00028027470262892445, "data_time": 0.002466726303100586, "grad_norm": 0.013205208000726998, "loss": 0.6878928959369659, "time": 0.4076138734817505, "epoch": 14, "memory": 20334, "step": 78354} +{"lr": 0.00028027470262892445, "data_time": 0.002306079864501953, "grad_norm": 0.02017670273780823, "loss": 0.6877880394458771, "time": 0.4074725151062012, "epoch": 14, "memory": 20334, "step": 78454} +{"lr": 0.00028027470262892445, "data_time": 0.0017530441284179688, "grad_norm": 0.016754168551415205, "loss": 0.6903860211372376, "time": 0.422964072227478, "epoch": 14, "memory": 20334, "step": 78554} +{"lr": 0.00028027470262892445, "data_time": 0.001977968215942383, "grad_norm": 0.020933930505998434, "loss": 0.6920530259609222, "time": 0.40083434581756594, "epoch": 14, "memory": 20334, "step": 78654} +{"lr": 0.00028027470262892445, "data_time": 0.002583169937133789, "grad_norm": 0.01593464156612754, "loss": 0.6848901391029358, "time": 0.4061759948730469, "epoch": 14, "memory": 20334, "step": 78754} +{"lr": 0.00028027470262892445, "data_time": 0.002393364906311035, "grad_norm": 0.014148953440599143, "loss": 0.6822631180286407, "time": 0.4196335792541504, "epoch": 14, "memory": 20334, "step": 78854} +{"lr": 0.00028027470262892445, "data_time": 0.0028033018112182616, "grad_norm": 0.015457310376223176, "loss": 0.6922883689403534, "time": 0.42890777587890627, "epoch": 14, "memory": 20334, "step": 78954} +{"lr": 0.00028027470262892445, "data_time": 0.0022582769393920898, "grad_norm": 0.016887045186012983, "loss": 0.6882308006286622, "time": 0.41818015575408934, "epoch": 14, "memory": 20334, "step": 79054} +{"lr": 0.00028027470262892445, "data_time": 0.0021661996841430666, "grad_norm": 0.016308948502410204, "loss": 0.6863297164440155, "time": 0.45968387126922605, "epoch": 14, "memory": 20334, "step": 79154} +{"lr": 0.00028027470262892445, "data_time": 0.002317214012145996, "grad_norm": 0.016493112477473915, "loss": 0.6870655953884125, "time": 0.43414759635925293, "epoch": 14, "memory": 20334, "step": 79254} +{"lr": 0.00028027470262892445, "data_time": 0.0023792743682861327, "grad_norm": 0.017419931944459677, "loss": 0.6883424043655395, "time": 0.382624077796936, "epoch": 14, "memory": 20334, "step": 79354} +{"lr": 0.00028027470262892445, "data_time": 0.0028496026992797852, "grad_norm": 0.01771250255405903, "loss": 0.6872944593429565, "time": 0.403963565826416, "epoch": 14, "memory": 20334, "step": 79454} +{"lr": 0.00028027470262892445, "data_time": 0.0028537511825561523, "grad_norm": 0.013814081810414791, "loss": 0.684880667924881, "time": 0.40574567317962645, "epoch": 14, "memory": 20334, "step": 79554} +{"lr": 0.00028027470262892445, "data_time": 0.0017877340316772461, "grad_norm": 0.014335823338478803, "loss": 0.684401661157608, "time": 0.3935187578201294, "epoch": 14, "memory": 20334, "step": 79654} +{"lr": 0.00028027470262892445, "data_time": 0.002208518981933594, "grad_norm": 0.014017370890360325, "loss": 0.6845682501792908, "time": 0.38803040981292725, "epoch": 14, "memory": 20334, "step": 79754} +{"lr": 0.00028027470262892445, "data_time": 0.0017626762390136718, "grad_norm": 0.013043110590660944, "loss": 0.6860217750072479, "time": 0.4062116861343384, "epoch": 14, "memory": 20334, "step": 79854} +{"lr": 0.00028027470262892445, "data_time": 0.0015872955322265626, "grad_norm": 0.01832923444453627, "loss": 0.6916180849075317, "time": 0.41204280853271485, "epoch": 14, "memory": 20334, "step": 79954} +{"lr": 0.00028027470262892445, "data_time": 0.0027703046798706055, "grad_norm": 0.020153836393728853, "loss": 0.6901819586753846, "time": 0.38084139823913576, "epoch": 14, "memory": 20334, "step": 80054} +{"lr": 0.00028027470262892445, "data_time": 0.002006077766418457, "grad_norm": 0.018345507606863976, "loss": 0.6888871848583221, "time": 0.4347590684890747, "epoch": 14, "memory": 20334, "step": 80154} +{"lr": 0.00028027470262892445, "data_time": 0.0017123222351074219, "grad_norm": 0.01874561405275017, "loss": 0.6875796914100647, "time": 0.40040853023529055, "epoch": 14, "memory": 20334, "step": 80254} +{"lr": 0.00028027470262892445, "data_time": 0.0020253896713256837, "grad_norm": 0.020681688887998463, "loss": 0.6857459127902985, "time": 0.4483737230300903, "epoch": 14, "memory": 20334, "step": 80354} +{"lr": 0.00028027470262892445, "data_time": 0.0018351078033447266, "grad_norm": 0.018248191615566612, "loss": 0.6877990126609802, "time": 0.39775097370147705, "epoch": 14, "memory": 20334, "step": 80454} +{"lr": 0.00028027470262892445, "data_time": 0.001861405372619629, "grad_norm": 0.014884584629908205, "loss": 0.685151606798172, "time": 0.4089222431182861, "epoch": 14, "memory": 20334, "step": 80554} +{"accuracy/top1": 100.0, "data_time": 0.36675238609313965, "time": 0.9948916170332167, "step": 14} +{"lr": 0.0002140463001152259, "data_time": 0.002021336555480957, "grad_norm": 0.020388060808181764, "loss": 0.6880136609077454, "time": 0.39930334091186526, "epoch": 15, "memory": 20334, "step": 80712} +{"lr": 0.0002140463001152259, "data_time": 0.0017746210098266602, "grad_norm": 0.013172862562350928, "loss": 0.6877367615699768, "time": 0.4345247745513916, "epoch": 15, "memory": 20334, "step": 80812} +{"lr": 0.0002140463001152259, "data_time": 0.002066946029663086, "grad_norm": 0.020787785411812365, "loss": 0.6855421900749207, "time": 0.4268107652664185, "epoch": 15, "memory": 20334, "step": 80912} +{"lr": 0.0002140463001152259, "data_time": 0.003052163124084473, "grad_norm": 0.01730718093458563, "loss": 0.6878939151763916, "time": 0.40036122798919677, "epoch": 15, "memory": 20334, "step": 81012} +{"lr": 0.0002140463001152259, "data_time": 0.0019379854202270508, "grad_norm": 0.012013878964353353, "loss": 0.6882242381572723, "time": 0.37637958526611326, "epoch": 15, "memory": 20334, "step": 81112} +{"lr": 0.0002140463001152259, "data_time": 0.0019441604614257812, "grad_norm": 0.010277168377069757, "loss": 0.6887440383434296, "time": 0.39629502296447755, "epoch": 15, "memory": 20334, "step": 81212} +{"lr": 0.0002140463001152259, "data_time": 0.0019289255142211914, "grad_norm": 0.01941108452156186, "loss": 0.687767618894577, "time": 0.4071494102478027, "epoch": 15, "memory": 20334, "step": 81312} +{"lr": 0.0002140463001152259, "data_time": 0.0019929170608520507, "grad_norm": 0.017288946313783526, "loss": 0.6901023149490356, "time": 0.42064311504364016, "epoch": 15, "memory": 20334, "step": 81412} +{"lr": 0.0002140463001152259, "data_time": 0.0017107486724853515, "grad_norm": 0.025121022341772915, "loss": 0.6882579863071442, "time": 0.39304091930389407, "epoch": 15, "memory": 20334, "step": 81512} +{"lr": 0.0002140463001152259, "data_time": 0.0016469955444335938, "grad_norm": 0.024632679484784602, "loss": 0.6852174937725067, "time": 0.3752723455429077, "epoch": 15, "memory": 20334, "step": 81612} +{"lr": 0.0002140463001152259, "data_time": 0.002080368995666504, "grad_norm": 0.012507924216333777, "loss": 0.6829984188079834, "time": 0.4012298583984375, "epoch": 15, "memory": 20334, "step": 81712} +{"lr": 0.0002140463001152259, "data_time": 0.0017494440078735351, "grad_norm": 0.018673277169000357, "loss": 0.6915401458740235, "time": 0.3802367687225342, "epoch": 15, "memory": 20334, "step": 81812} +{"lr": 0.0002140463001152259, "data_time": 0.0018441438674926757, "grad_norm": 0.011053581692976877, "loss": 0.6896061718463897, "time": 0.4221057415008545, "epoch": 15, "memory": 20334, "step": 81912} +{"lr": 0.0002140463001152259, "data_time": 0.0019703149795532227, "grad_norm": 0.02195824042428285, "loss": 0.687947416305542, "time": 0.4112419128417969, "epoch": 15, "memory": 20334, "step": 82012} +{"lr": 0.0002140463001152259, "data_time": 0.0026426315307617188, "grad_norm": 0.02192495160270482, "loss": 0.6829129338264466, "time": 0.4389740705490112, "epoch": 15, "memory": 20334, "step": 82112} +{"lr": 0.0002140463001152259, "data_time": 0.0015482664108276366, "grad_norm": 0.72354477616027, "loss": 0.6864808142185211, "time": 0.3933025121688843, "epoch": 15, "memory": 20334, "step": 82212} +{"lr": 0.0002140463001152259, "data_time": 0.0014982223510742188, "grad_norm": 0.019350684783421456, "loss": 0.6884127199649811, "time": 0.39441399574279784, "epoch": 15, "memory": 20334, "step": 82312} +{"lr": 0.0002140463001152259, "data_time": 0.0019357919692993164, "grad_norm": 0.01935142130823806, "loss": 0.6874332964420319, "time": 0.4041867971420288, "epoch": 15, "memory": 20334, "step": 82412} +{"lr": 0.0002140463001152259, "data_time": 0.0018247604370117188, "grad_norm": 0.0234906857018359, "loss": 0.6824127256870269, "time": 0.40228424072265623, "epoch": 15, "memory": 20334, "step": 82512} +{"lr": 0.0002140463001152259, "data_time": 0.002836275100708008, "grad_norm": 0.016757755377329885, "loss": 0.6862969815731048, "time": 0.426112699508667, "epoch": 15, "memory": 20334, "step": 82612} +{"lr": 0.0002140463001152259, "data_time": 0.0019039869308471679, "grad_norm": 0.019472584035247563, "loss": 0.6847643315792084, "time": 0.4146775484085083, "epoch": 15, "memory": 20334, "step": 82712} +{"lr": 0.0002140463001152259, "data_time": 0.0023277759552001952, "grad_norm": 0.011498039378784597, "loss": 0.6896952092647552, "time": 0.4664032220840454, "epoch": 15, "memory": 20334, "step": 82812} +{"lr": 0.0002140463001152259, "data_time": 0.001938939094543457, "grad_norm": 0.018599987821653487, "loss": 0.686225414276123, "time": 0.3851919412612915, "epoch": 15, "memory": 20334, "step": 82912} +{"lr": 0.0002140463001152259, "data_time": 0.002094864845275879, "grad_norm": 0.021390922949649394, "loss": 0.6847345352172851, "time": 0.4318429708480835, "epoch": 15, "memory": 20334, "step": 83012} +{"lr": 0.0002140463001152259, "data_time": 0.0019536495208740236, "grad_norm": 0.015049825771711767, "loss": 0.6843174219131469, "time": 0.38935074806213377, "epoch": 15, "memory": 20334, "step": 83112} +{"lr": 0.0002140463001152259, "data_time": 0.0023333072662353516, "grad_norm": 0.011506680818274617, "loss": 0.6823466122150421, "time": 0.4033582925796509, "epoch": 15, "memory": 20334, "step": 83212} +{"lr": 0.0002140463001152259, "data_time": 0.0017803192138671875, "grad_norm": 0.02404228476807475, "loss": 0.681760984659195, "time": 0.39705166816711424, "epoch": 15, "memory": 20334, "step": 83312} +{"lr": 0.0002140463001152259, "data_time": 0.002094674110412598, "grad_norm": 0.010580299817956985, "loss": 0.6865424275398254, "time": 0.3992241144180298, "epoch": 15, "memory": 20334, "step": 83412} +{"lr": 0.0002140463001152259, "data_time": 0.0019010305404663086, "grad_norm": 0.010495419334620237, "loss": 0.6883788228034973, "time": 0.3959256410598755, "epoch": 15, "memory": 20334, "step": 83512} +{"lr": 0.0002140463001152259, "data_time": 0.0016830921173095702, "grad_norm": 0.018749719858169554, "loss": 0.683070284128189, "time": 0.429161810874939, "epoch": 15, "memory": 20334, "step": 83612} +{"lr": 0.0002140463001152259, "data_time": 0.0017444133758544923, "grad_norm": 0.019595989398658274, "loss": 0.6889229655265808, "time": 0.41781799793243407, "epoch": 15, "memory": 20334, "step": 83712} +{"lr": 0.0002140463001152259, "data_time": 0.001847219467163086, "grad_norm": 0.022146631916984915, "loss": 0.6834630370140076, "time": 0.399010157585144, "epoch": 15, "memory": 20334, "step": 83812} +{"lr": 0.0002140463001152259, "data_time": 0.0016989946365356446, "grad_norm": 8.715001306682826, "loss": 0.682533997297287, "time": 0.46860029697418215, "epoch": 15, "memory": 20334, "step": 83912} +{"lr": 0.0002140463001152259, "data_time": 0.0016447067260742187, "grad_norm": 0.01404360574670136, "loss": 0.6851846218109131, "time": 0.40465679168701174, "epoch": 15, "memory": 20334, "step": 84012} +{"lr": 0.0002140463001152259, "data_time": 0.0017577648162841798, "grad_norm": 0.02531041847541928, "loss": 0.6839929401874543, "time": 0.43806498050689696, "epoch": 15, "memory": 20334, "step": 84112} +{"lr": 0.0002140463001152259, "data_time": 0.0017374277114868164, "grad_norm": 0.028425779100507496, "loss": 0.6847677826881409, "time": 0.3850772619247437, "epoch": 15, "memory": 20334, "step": 84212} +{"lr": 0.0002140463001152259, "data_time": 0.0017788410186767578, "grad_norm": 0.02780276567209512, "loss": 0.6871353566646576, "time": 0.4300373554229736, "epoch": 15, "memory": 20334, "step": 84312} +{"lr": 0.0002140463001152259, "data_time": 0.002084779739379883, "grad_norm": 0.04376483880914748, "loss": 0.6851118683815003, "time": 0.41359691619873046, "epoch": 15, "memory": 20334, "step": 84412} +{"lr": 0.0002140463001152259, "data_time": 0.0018017292022705078, "grad_norm": 0.02930031605064869, "loss": 0.6853167831897735, "time": 0.39764833450317383, "epoch": 15, "memory": 20334, "step": 84512} +{"lr": 0.0002140463001152259, "data_time": 0.0030493974685668946, "grad_norm": 0.05422095921821892, "loss": 0.6835175812244415, "time": 0.4271942377090454, "epoch": 15, "memory": 20334, "step": 84612} +{"lr": 0.0002140463001152259, "data_time": 0.0025160551071166993, "grad_norm": 0.19785308777354657, "loss": 0.6901568472385406, "time": 0.41784348487854006, "epoch": 15, "memory": 20334, "step": 84712} +{"lr": 0.0002140463001152259, "data_time": 0.0020036220550537108, "grad_norm": 0.021754285646602513, "loss": 0.6847085654735565, "time": 0.42932806015014646, "epoch": 15, "memory": 20334, "step": 84812} +{"lr": 0.0002140463001152259, "data_time": 0.0018838882446289063, "grad_norm": 0.13789198875892908, "loss": 0.6823304831981659, "time": 0.3864087343215942, "epoch": 15, "memory": 20334, "step": 84912} +{"lr": 0.0002140463001152259, "data_time": 0.0019930124282836912, "grad_norm": 0.029603640688583253, "loss": 0.6802552163600921, "time": 0.39294002056121824, "epoch": 15, "memory": 20334, "step": 85012} +{"lr": 0.0002140463001152259, "data_time": 0.0018412351608276367, "grad_norm": 0.12799951825290917, "loss": 0.683345890045166, "time": 0.4119910717010498, "epoch": 15, "memory": 20334, "step": 85112} +{"lr": 0.0002140463001152259, "data_time": 0.0019691944122314452, "grad_norm": 0.08311668909154832, "loss": 0.6847773671150208, "time": 0.42104787826538087, "epoch": 15, "memory": 20334, "step": 85212} +{"lr": 0.0002140463001152259, "data_time": 0.0014868497848510743, "grad_norm": 0.028482846869155764, "loss": 0.6900527358055115, "time": 0.40334813594818114, "epoch": 15, "memory": 20334, "step": 85312} +{"lr": 0.0002140463001152259, "data_time": 0.0015172481536865235, "grad_norm": 0.02009584875777364, "loss": 0.6823492228984833, "time": 0.4446099042892456, "epoch": 15, "memory": 20334, "step": 85412} +{"lr": 0.0002140463001152259, "data_time": 0.0018595218658447265, "grad_norm": 0.015387416537851095, "loss": 0.6880171895027161, "time": 0.40314862728118894, "epoch": 15, "memory": 20334, "step": 85512} +{"lr": 0.0002140463001152259, "data_time": 0.0016479253768920898, "grad_norm": 8.695492133498192, "loss": 0.683335691690445, "time": 0.4031576156616211, "epoch": 15, "memory": 20334, "step": 85612} +{"lr": 0.0002140463001152259, "data_time": 0.0019529342651367187, "grad_norm": 0.06487121218815446, "loss": 0.685955011844635, "time": 0.3881665229797363, "epoch": 15, "memory": 20334, "step": 85712} +{"lr": 0.0002140463001152259, "data_time": 0.0023677587509155274, "grad_norm": 0.08772307783365249, "loss": 0.6858329474925995, "time": 0.3961667537689209, "epoch": 15, "memory": 20334, "step": 85812} +{"lr": 0.0002140463001152259, "data_time": 0.0021058082580566405, "grad_norm": 3.8150058895349503, "loss": 0.673198688030243, "time": 0.4034181833267212, "epoch": 15, "memory": 20334, "step": 85912} +{"lr": 0.0002140463001152259, "data_time": 0.002176332473754883, "grad_norm": 2.4243196934461593, "loss": 0.6876299798488616, "time": 0.37183878421783445, "epoch": 15, "memory": 20334, "step": 86012} +{"lr": 0.0002140463001152259, "data_time": 0.0022976160049438476, "grad_norm": 1.663897656649351, "loss": 0.6721494734287262, "time": 0.46376795768737794, "epoch": 15, "memory": 20334, "step": 86112} +{"lr": 0.0002140463001152259, "data_time": 0.002008962631225586, "grad_norm": 46.67166759073734, "loss": 0.6610705494880676, "time": 0.4014244079589844, "epoch": 15, "memory": 20334, "step": 86212} +{"lr": 0.0002140463001152259, "data_time": 0.002184939384460449, "grad_norm": 2.503987103700638, "loss": 0.6638953685760498, "time": 0.4040172815322876, "epoch": 15, "memory": 20334, "step": 86312} +{"accuracy/top1": 93.89558410644531, "data_time": 0.37315506405300564, "time": 0.9995271894666884, "step": 15} +{"lr": 0.00015498214331265904, "data_time": 0.0016907691955566407, "grad_norm": 12.951112318038941, "loss": 0.6619929432868957, "time": 0.4124996423721313, "epoch": 16, "memory": 20334, "step": 86470} +{"lr": 0.00015498214331265904, "data_time": 0.0018224239349365235, "grad_norm": 20.398714852333068, "loss": 0.6426031589508057, "time": 0.44016361236572266, "epoch": 16, "memory": 20334, "step": 86570} +{"lr": 0.00015498214331265904, "data_time": 0.0025153160095214844, "grad_norm": 166.12853293418885, "loss": 0.6845923960208893, "time": 0.3781388282775879, "epoch": 16, "memory": 20334, "step": 86670} +{"lr": 0.00015498214331265904, "data_time": 0.0031164407730102537, "grad_norm": 17.45560750961304, "loss": 0.6753440737724304, "time": 0.4268238067626953, "epoch": 16, "memory": 20334, "step": 86770} +{"lr": 0.00015498214331265904, "data_time": 0.0017195940017700195, "grad_norm": 153.9520194530487, "loss": 0.6710609078407288, "time": 0.43384950160980223, "epoch": 16, "memory": 20334, "step": 86870} +{"lr": 0.00015498214331265904, "data_time": 0.0018800973892211914, "grad_norm": 16.20660207271576, "loss": 0.6688604831695557, "time": 0.4156578540802002, "epoch": 16, "memory": 20334, "step": 86970} +{"lr": 0.00015498214331265904, "data_time": 0.0024732351303100586, "grad_norm": 5.085641115903854, "loss": 0.696414589881897, "time": 0.4041528940200806, "epoch": 16, "memory": 20334, "step": 87070} +{"lr": 0.00015498214331265904, "data_time": 0.003353714942932129, "grad_norm": 6.340100446715951, "loss": 0.6840851962566376, "time": 0.4223287582397461, "epoch": 16, "memory": 20334, "step": 87170} +{"lr": 0.00015498214331265904, "data_time": 0.0015853643417358398, "grad_norm": 0.969308928400278, "loss": 0.6880277335643769, "time": 0.3990727663040161, "epoch": 16, "memory": 20334, "step": 87270} +{"lr": 0.00015498214331265904, "data_time": 0.0019559621810913085, "grad_norm": 1.2147032178938388, "loss": 0.6861448228359223, "time": 0.40780184268951414, "epoch": 16, "memory": 20334, "step": 87370} +{"lr": 0.00015498214331265904, "data_time": 0.0017110347747802735, "grad_norm": 0.833619886636734, "loss": 0.6849690914154053, "time": 0.4397338628768921, "epoch": 16, "memory": 20334, "step": 87470} +{"lr": 0.00015498214331265904, "data_time": 0.0019336938858032227, "grad_norm": 1.7692050963640213, "loss": 0.6862511754035949, "time": 0.44584016799926757, "epoch": 16, "memory": 20334, "step": 87570} +{"lr": 0.00015498214331265904, "data_time": 0.002180600166320801, "grad_norm": 1.9187275230884553, "loss": 0.6827983617782593, "time": 0.41813228130340574, "epoch": 16, "memory": 20334, "step": 87670} +{"lr": 0.00015498214331265904, "data_time": 0.0031360387802124023, "grad_norm": 10.539541080594063, "loss": 0.6742065012454986, "time": 0.3825331926345825, "epoch": 16, "memory": 20334, "step": 87770} +{"lr": 0.00015498214331265904, "data_time": 0.0026463747024536135, "grad_norm": 0.7334457784891129, "loss": 0.6746126055717468, "time": 0.44365291595458983, "epoch": 16, "memory": 20334, "step": 87870} +{"lr": 0.00015498214331265904, "data_time": 0.0021526098251342775, "grad_norm": 0.8798057943582535, "loss": 0.6776270508766175, "time": 0.41785004138946535, "epoch": 16, "memory": 20334, "step": 87970} +{"lr": 0.00015498214331265904, "data_time": 0.003639531135559082, "grad_norm": 3.3916256219148635, "loss": 0.6738600790500641, "time": 0.4211766481399536, "epoch": 16, "memory": 20334, "step": 88070} +{"lr": 0.00015498214331265904, "data_time": 0.0019486665725708008, "grad_norm": 0.2992755934596062, "loss": 0.685307115316391, "time": 0.38522896766662595, "epoch": 16, "memory": 20334, "step": 88170} +{"lr": 0.00015498214331265904, "data_time": 0.0019204378128051757, "grad_norm": 0.398986155167222, "loss": 0.6886320233345031, "time": 0.4050882816314697, "epoch": 16, "memory": 20334, "step": 88270} +{"lr": 0.00015498214331265904, "data_time": 0.0028360843658447265, "grad_norm": 0.17018628530204297, "loss": 0.6884448051452636, "time": 0.4105250597000122, "epoch": 16, "memory": 20334, "step": 88370} +{"lr": 0.00015498214331265904, "data_time": 0.002909278869628906, "grad_norm": 0.462920555472374, "loss": 0.6926625609397888, "time": 0.4037681818008423, "epoch": 16, "memory": 20334, "step": 88470} +{"lr": 0.00015498214331265904, "data_time": 0.00221707820892334, "grad_norm": 17.873842865228653, "loss": 0.6875955760478973, "time": 0.44903762340545655, "epoch": 16, "memory": 20334, "step": 88570} +{"lr": 0.00015498214331265904, "data_time": 0.002274298667907715, "grad_norm": 11.725898110866547, "loss": 0.6912346422672272, "time": 0.41946699619293215, "epoch": 16, "memory": 20334, "step": 88670} +{"lr": 0.00015498214331265904, "data_time": 0.0017960548400878906, "grad_norm": 1.5581879802048206, "loss": 0.6845430731773376, "time": 0.4291127443313599, "epoch": 16, "memory": 20334, "step": 88770} +{"lr": 0.00015498214331265904, "data_time": 0.0027998685836791992, "grad_norm": 4.23603507168591, "loss": 0.6869994580745697, "time": 0.39378364086151124, "epoch": 16, "memory": 20334, "step": 88870} +{"lr": 0.00015498214331265904, "data_time": 0.0017722845077514648, "grad_norm": 0.727909404411912, "loss": 0.6893838763236999, "time": 0.412534499168396, "epoch": 16, "memory": 20334, "step": 88970} +{"lr": 0.00015498214331265904, "data_time": 0.0018594741821289062, "grad_norm": 0.42019427847117186, "loss": 0.6843425452709198, "time": 0.40675690174102785, "epoch": 16, "memory": 20334, "step": 89070} +{"lr": 0.00015498214331265904, "data_time": 0.0017817497253417968, "grad_norm": 0.13882692344486713, "loss": 0.6857464551925659, "time": 0.4022836685180664, "epoch": 16, "memory": 20334, "step": 89170} +{"lr": 0.00015498214331265904, "data_time": 0.001861262321472168, "grad_norm": 0.04924918580800295, "loss": 0.6896380245685577, "time": 0.405580997467041, "epoch": 16, "memory": 20334, "step": 89270} +{"lr": 0.00015498214331265904, "data_time": 0.0018180131912231446, "grad_norm": 2.302377102896571, "loss": 0.6833753228187561, "time": 0.3844506025314331, "epoch": 16, "memory": 20334, "step": 89370} +{"lr": 0.00015498214331265904, "data_time": 0.0020698070526123046, "grad_norm": 0.8118695575743914, "loss": 0.6921520233154297, "time": 0.3881402492523193, "epoch": 16, "memory": 20334, "step": 89470} +{"lr": 0.00015498214331265904, "data_time": 0.0017733573913574219, "grad_norm": 0.24332242012023925, "loss": 0.6881895542144776, "time": 0.43993773460388186, "epoch": 16, "memory": 20334, "step": 89570} +{"lr": 0.00015498214331265904, "data_time": 0.0017596006393432618, "grad_norm": 0.3353728521615267, "loss": 0.6832014620304108, "time": 0.41824162006378174, "epoch": 16, "memory": 20334, "step": 89670} +{"lr": 0.00015498214331265904, "data_time": 0.0016766786575317383, "grad_norm": 0.251370819658041, "loss": 0.686100322008133, "time": 0.42429065704345703, "epoch": 16, "memory": 20334, "step": 89770} +{"lr": 0.00015498214331265904, "data_time": 0.001540803909301758, "grad_norm": 0.6775854665786027, "loss": 0.6858800172805786, "time": 0.4270820140838623, "epoch": 16, "memory": 20334, "step": 89870} +{"lr": 0.00015498214331265904, "data_time": 0.0016483306884765626, "grad_norm": 0.6123087488114833, "loss": 0.6825789332389831, "time": 0.39502439498901365, "epoch": 16, "memory": 20334, "step": 89970} +{"lr": 0.00015498214331265904, "data_time": 0.00198209285736084, "grad_norm": 0.20175361707806588, "loss": 0.6863687098026275, "time": 0.46511433124542234, "epoch": 16, "memory": 20334, "step": 90070} +{"lr": 0.00015498214331265904, "data_time": 0.0018755674362182617, "grad_norm": 0.2863535758107901, "loss": 0.687699693441391, "time": 0.4243708848953247, "epoch": 16, "memory": 20334, "step": 90170} +{"lr": 0.00015498214331265904, "data_time": 0.001748061180114746, "grad_norm": 0.1539195215329528, "loss": 0.6884487390518188, "time": 0.4154834747314453, "epoch": 16, "memory": 20334, "step": 90270} +{"lr": 0.00015498214331265904, "data_time": 0.001672840118408203, "grad_norm": 0.19501777943223714, "loss": 0.6865182340145111, "time": 0.43398594856262207, "epoch": 16, "memory": 20334, "step": 90370} +{"lr": 0.00015498214331265904, "data_time": 0.0016072988510131836, "grad_norm": 0.5412828426808118, "loss": 0.6927432000637055, "time": 0.42673544883728026, "epoch": 16, "memory": 20334, "step": 90470} +{"lr": 0.00015498214331265904, "data_time": 0.0017229318618774414, "grad_norm": 0.7294685713946819, "loss": 0.688781064748764, "time": 0.39732863903045657, "epoch": 16, "memory": 20334, "step": 90570} +{"lr": 0.00015498214331265904, "data_time": 0.001889514923095703, "grad_norm": 1.2996361970901489, "loss": 0.6843153655529022, "time": 0.38910818099975586, "epoch": 16, "memory": 20334, "step": 90670} +{"lr": 0.00015498214331265904, "data_time": 0.0015711784362792969, "grad_norm": 0.5796617537736892, "loss": 0.6846848845481872, "time": 0.37526464462280273, "epoch": 16, "memory": 20334, "step": 90770} +{"lr": 0.00015498214331265904, "data_time": 0.0015628814697265625, "grad_norm": 4.027517405152321, "loss": 0.6886925756931305, "time": 0.371440052986145, "epoch": 16, "memory": 20334, "step": 90870} +{"lr": 0.00015498214331265904, "data_time": 0.0015830516815185547, "grad_norm": 2.0649502158164976, "loss": 0.6806890785694122, "time": 0.4043705701828003, "epoch": 16, "memory": 20334, "step": 90970} +{"lr": 0.00015498214331265904, "data_time": 0.0015935182571411132, "grad_norm": 2.3120406031608582, "loss": 0.6872101962566376, "time": 0.39736437797546387, "epoch": 16, "memory": 20334, "step": 91070} +{"lr": 0.00015498214331265904, "data_time": 0.001550149917602539, "grad_norm": 6.763669490814209, "loss": 0.6779651641845703, "time": 0.4133500099182129, "epoch": 16, "memory": 20334, "step": 91170} +{"lr": 0.00015498214331265904, "data_time": 0.0016917705535888672, "grad_norm": 16.962061190605162, "loss": 0.6605131447315216, "time": 0.4373743534088135, "epoch": 16, "memory": 20334, "step": 91270} +{"lr": 0.00015498214331265904, "data_time": 0.0015970230102539062, "grad_norm": 7.3323180198669435, "loss": 0.6797120928764343, "time": 0.3811880350112915, "epoch": 16, "memory": 20334, "step": 91370} +{"lr": 0.00015498214331265904, "data_time": 0.001712632179260254, "grad_norm": 18.087129247188567, "loss": 0.6597596287727356, "time": 0.3891608238220215, "epoch": 16, "memory": 20334, "step": 91470} +{"lr": 0.00015498214331265904, "data_time": 0.0015247821807861327, "grad_norm": 422.3002927303314, "loss": 0.6642901420593261, "time": 0.40826027393341063, "epoch": 16, "memory": 20334, "step": 91570} +{"lr": 0.00015498214331265904, "data_time": 0.0017327308654785157, "grad_norm": 293.7801417350769, "loss": 0.6675473213195801, "time": 0.38245701789855957, "epoch": 16, "memory": 20334, "step": 91670} +{"lr": 0.00015498214331265904, "data_time": 0.0016161441802978516, "grad_norm": 63.248290061950684, "loss": 0.6756110072135926, "time": 0.38981170654296876, "epoch": 16, "memory": 20334, "step": 91770} +{"lr": 0.00015498214331265904, "data_time": 0.0018299579620361327, "grad_norm": 40.153921604156494, "loss": 0.6774940490722656, "time": 0.41505343914031984, "epoch": 16, "memory": 20334, "step": 91870} +{"lr": 0.00015498214331265904, "data_time": 0.0020282506942749024, "grad_norm": 77.30255718231201, "loss": 0.6731551468372345, "time": 0.3951940298080444, "epoch": 16, "memory": 20334, "step": 91970} +{"lr": 0.00015498214331265904, "data_time": 0.0017313957214355469, "grad_norm": 126.43501660823821, "loss": 0.670775580406189, "time": 0.3873365163803101, "epoch": 16, "memory": 20334, "step": 92070} +{"accuracy/top1": 98.09986114501953, "data_time": 0.3739733960893419, "time": 1.0005699263678656, "step": 16} +{"lr": 0.00010453658778440108, "data_time": 0.0016625165939331056, "grad_norm": 3.2546398401260377, "loss": 0.6641832888126373, "time": 0.41259312629699707, "epoch": 17, "memory": 20334, "step": 92228} +{"lr": 0.00010453658778440108, "data_time": 0.0016060352325439453, "grad_norm": 9.871051055192947, "loss": 0.6584961235523223, "time": 0.41314380168914794, "epoch": 17, "memory": 20334, "step": 92328} +{"lr": 0.00010453658778440108, "data_time": 0.0015558719635009766, "grad_norm": 16.2878378868103, "loss": 0.6527981042861939, "time": 0.4243000507354736, "epoch": 17, "memory": 20334, "step": 92428} +{"lr": 0.00010453658778440108, "data_time": 0.0016339778900146484, "grad_norm": 15.20300098657608, "loss": 0.6578659296035767, "time": 0.4310301780700684, "epoch": 17, "memory": 20334, "step": 92528} +{"lr": 0.00010453658778440108, "data_time": 0.0015978336334228516, "grad_norm": 11.455358362197876, "loss": 0.6541256964206695, "time": 0.42904951572418215, "epoch": 17, "memory": 20334, "step": 92628} +{"lr": 0.00010453658778440108, "data_time": 0.0016177892684936523, "grad_norm": 12.467400646209716, "loss": 0.6616435170173645, "time": 0.414095139503479, "epoch": 17, "memory": 20334, "step": 92728} +{"lr": 0.00010453658778440108, "data_time": 0.001538991928100586, "grad_norm": 15.241450905799866, "loss": 0.6654034435749054, "time": 0.40929033756256106, "epoch": 17, "memory": 20334, "step": 92828} +{"lr": 0.00010453658778440108, "data_time": 0.0016882658004760743, "grad_norm": 6.928589969873428, "loss": 0.6621437847614289, "time": 0.4103843688964844, "epoch": 17, "memory": 20334, "step": 92928} +{"lr": 0.00010453658778440108, "data_time": 0.0016207218170166016, "grad_norm": 11.785401248931885, "loss": 0.6657221257686615, "time": 0.4183415174484253, "epoch": 17, "memory": 20334, "step": 93028} +{"lr": 0.00010453658778440108, "data_time": 0.0018795251846313477, "grad_norm": 19.80033379793167, "loss": 0.6555939137935638, "time": 0.45161216259002684, "epoch": 17, "memory": 20334, "step": 93128} +{"lr": 0.00010453658778440108, "data_time": 0.0018337249755859375, "grad_norm": 13.87143907546997, "loss": 0.647191196680069, "time": 0.3781932592391968, "epoch": 17, "memory": 20334, "step": 93228} +{"lr": 0.00010453658778440108, "data_time": 0.003337192535400391, "grad_norm": 22.115187096595765, "loss": 0.6549151837825775, "time": 0.4360326290130615, "epoch": 17, "memory": 20334, "step": 93328} +{"lr": 0.00010453658778440108, "data_time": 0.0018713235855102538, "grad_norm": 5.600091993808746, "loss": 0.6542356610298157, "time": 0.4102866888046265, "epoch": 17, "memory": 20334, "step": 93428} +{"lr": 0.00010453658778440108, "data_time": 0.002435731887817383, "grad_norm": 11.64617965221405, "loss": 0.6576518476009369, "time": 0.4596485376358032, "epoch": 17, "memory": 20334, "step": 93528} +{"lr": 0.00010453658778440108, "data_time": 0.003143620491027832, "grad_norm": 129.42072924375535, "loss": 0.6680347144603729, "time": 0.3715360164642334, "epoch": 17, "memory": 20334, "step": 93628} +{"lr": 0.00010453658778440108, "data_time": 0.0023673295974731444, "grad_norm": 32.76223933696747, "loss": 0.6641122877597809, "time": 0.38414716720581055, "epoch": 17, "memory": 20334, "step": 93728} +{"lr": 0.00010453658778440108, "data_time": 0.0017449617385864257, "grad_norm": 4.832263457775116, "loss": 0.6561089336872101, "time": 0.3951707363128662, "epoch": 17, "memory": 20334, "step": 93828} +{"lr": 0.00010453658778440108, "data_time": 0.0020903348922729492, "grad_norm": 3108.9794964671137, "loss": 0.6573094308376313, "time": 0.4078761339187622, "epoch": 17, "memory": 20334, "step": 93928} +{"lr": 0.00010453658778440108, "data_time": 0.0017760992050170898, "grad_norm": 7.687473797798157, "loss": 0.6516622245311737, "time": 0.44172675609588624, "epoch": 17, "memory": 20334, "step": 94028} +{"lr": 0.00010453658778440108, "data_time": 0.0016998052597045898, "grad_norm": 86.21956859827041, "loss": 0.6472983598709107, "time": 0.42015249729156495, "epoch": 17, "memory": 20334, "step": 94128} +{"lr": 0.00010453658778440108, "data_time": 0.002736973762512207, "grad_norm": 60.84196796417236, "loss": 0.6557434618473053, "time": 0.40561344623565676, "epoch": 17, "memory": 20334, "step": 94228} +{"lr": 0.00010453658778440108, "data_time": 0.001611948013305664, "grad_norm": 19.71068468093872, "loss": 0.6611088514328003, "time": 0.38568027019500734, "epoch": 17, "memory": 20334, "step": 94328} +{"lr": 0.00010453658778440108, "data_time": 0.002765393257141113, "grad_norm": 8.547202229499817, "loss": 0.6529266476631165, "time": 0.4219272375106812, "epoch": 17, "memory": 20334, "step": 94428} +{"lr": 0.00010453658778440108, "data_time": 0.002468371391296387, "grad_norm": 21.46789469718933, "loss": 0.666370689868927, "time": 0.4284600496292114, "epoch": 17, "memory": 20334, "step": 94528} +{"lr": 0.00010453658778440108, "data_time": 0.002414369583129883, "grad_norm": 29.784071826934813, "loss": 0.6476145446300506, "time": 0.40134098529815676, "epoch": 17, "memory": 20334, "step": 94628} +{"lr": 0.00010453658778440108, "data_time": 0.0025254964828491213, "grad_norm": 49.705214047431944, "loss": 0.648444402217865, "time": 0.40775332450866697, "epoch": 17, "memory": 20334, "step": 94728} +{"lr": 0.00010453658778440108, "data_time": 0.002287602424621582, "grad_norm": 8.643514037132263, "loss": 0.6472416996955872, "time": 0.3936166763305664, "epoch": 17, "memory": 20334, "step": 94828} +{"lr": 0.00010453658778440108, "data_time": 0.0020152807235717775, "grad_norm": 51.14836868047714, "loss": 0.6423110663890839, "time": 0.4298069000244141, "epoch": 17, "memory": 20334, "step": 94928} +{"lr": 0.00010453658778440108, "data_time": 0.0033156156539916994, "grad_norm": 19.75882931947708, "loss": 0.6387811541557312, "time": 0.3985825300216675, "epoch": 17, "memory": 20334, "step": 95028} +{"lr": 0.00010453658778440108, "data_time": 0.001702737808227539, "grad_norm": 4.2123747825622555, "loss": 0.648260486125946, "time": 0.4048029899597168, "epoch": 17, "memory": 20334, "step": 95128} +{"lr": 0.00010453658778440108, "data_time": 0.0018829345703125, "grad_norm": 3.844643288850784, "loss": 0.6351286172866821, "time": 0.372525691986084, "epoch": 17, "memory": 20334, "step": 95228} +{"lr": 0.00010453658778440108, "data_time": 0.001734161376953125, "grad_norm": 18.19198731184006, "loss": 0.6390858173370362, "time": 0.42943248748779295, "epoch": 17, "memory": 20334, "step": 95328} +{"lr": 0.00010453658778440108, "data_time": 0.0030065059661865236, "grad_norm": 9.158291417360306, "loss": 0.6450623452663422, "time": 0.4055814504623413, "epoch": 17, "memory": 20334, "step": 95428} +{"lr": 0.00010453658778440108, "data_time": 0.0024463891983032226, "grad_norm": 41.29972470998764, "loss": 0.6425542116165162, "time": 0.3878720998764038, "epoch": 17, "memory": 20334, "step": 95528} +{"lr": 0.00010453658778440108, "data_time": 0.0018125534057617187, "grad_norm": 37.74541869163513, "loss": 0.6478318452835083, "time": 0.3871973276138306, "epoch": 17, "memory": 20334, "step": 95628} +{"lr": 0.00010453658778440108, "data_time": 0.002763652801513672, "grad_norm": 18.169064903259276, "loss": 0.649678373336792, "time": 0.3866436004638672, "epoch": 17, "memory": 20334, "step": 95728} +{"lr": 0.00010453658778440108, "data_time": 0.0017476081848144531, "grad_norm": 63.33078880310059, "loss": 0.656665587425232, "time": 0.4066095113754272, "epoch": 17, "memory": 20334, "step": 95828} +{"lr": 0.00010453658778440108, "data_time": 0.001861405372619629, "grad_norm": 37.726093196868895, "loss": 0.6425377547740936, "time": 0.37084503173828126, "epoch": 17, "memory": 20334, "step": 95928} +{"lr": 0.00010453658778440108, "data_time": 0.002059483528137207, "grad_norm": 221.2627161026001, "loss": 0.6478916466236114, "time": 0.39398856163024903, "epoch": 17, "memory": 20334, "step": 96028} +{"lr": 0.00010453658778440108, "data_time": 0.002388763427734375, "grad_norm": 16.53392014503479, "loss": 0.650449150800705, "time": 0.3876559495925903, "epoch": 17, "memory": 20334, "step": 96128} +{"lr": 0.00010453658778440108, "data_time": 0.0021607398986816405, "grad_norm": 46.52003974914551, "loss": 0.6469671666622162, "time": 0.37990684509277345, "epoch": 17, "memory": 20334, "step": 96228} +{"lr": 0.00010453658778440108, "data_time": 0.0019446134567260743, "grad_norm": 45.20160779953003, "loss": 0.6488111853599549, "time": 0.41321752071380613, "epoch": 17, "memory": 20334, "step": 96328} +{"lr": 0.00010453658778440108, "data_time": 0.0016960620880126954, "grad_norm": 82.16019706726074, "loss": 0.6456238508224488, "time": 0.4006854772567749, "epoch": 17, "memory": 20334, "step": 96428} +{"lr": 0.00010453658778440108, "data_time": 0.001622319221496582, "grad_norm": 13.535050964355468, "loss": 0.6291361391544342, "time": 0.38329644203186036, "epoch": 17, "memory": 20334, "step": 96528} +{"lr": 0.00010453658778440108, "data_time": 0.0021713733673095702, "grad_norm": 8.185927581787109, "loss": 0.6445947825908661, "time": 0.38475635051727297, "epoch": 17, "memory": 20334, "step": 96628} +{"lr": 0.00010453658778440108, "data_time": 0.0016494512557983399, "grad_norm": 12.52618865966797, "loss": 0.6456365406513214, "time": 0.3875510931015015, "epoch": 17, "memory": 20334, "step": 96728} +{"lr": 0.00010453658778440108, "data_time": 0.0023557662963867186, "grad_norm": 54.97046413421631, "loss": 0.6526306748390198, "time": 0.44949591159820557, "epoch": 17, "memory": 20334, "step": 96828} +{"lr": 0.00010453658778440108, "data_time": 0.0025568485260009767, "grad_norm": 27.850926280021667, "loss": 0.6469048321247101, "time": 0.40942893028259275, "epoch": 17, "memory": 20334, "step": 96928} +{"lr": 0.00010453658778440108, "data_time": 0.002293086051940918, "grad_norm": 15.37418692111969, "loss": 0.6597590148448944, "time": 0.40243828296661377, "epoch": 17, "memory": 20334, "step": 97028} +{"lr": 0.00010453658778440108, "data_time": 0.0019539594650268555, "grad_norm": 23.74850389957428, "loss": 0.6477436661720276, "time": 0.4275946140289307, "epoch": 17, "memory": 20334, "step": 97128} +{"lr": 0.00010453658778440108, "data_time": 0.0021381616592407227, "grad_norm": 8.819652533531189, "loss": 0.6474440455436706, "time": 0.38517258167266843, "epoch": 17, "memory": 20334, "step": 97228} +{"lr": 0.00010453658778440108, "data_time": 0.0017972946166992187, "grad_norm": 37.99319822788239, "loss": 0.6425372898578644, "time": 0.4023005962371826, "epoch": 17, "memory": 20334, "step": 97328} +{"lr": 0.00010453658778440108, "data_time": 0.0022548437118530273, "grad_norm": 34.167623949050906, "loss": 0.6537059545516968, "time": 0.39999263286590575, "epoch": 17, "memory": 20334, "step": 97428} +{"lr": 0.00010453658778440108, "data_time": 0.002572035789489746, "grad_norm": 48.18318336009979, "loss": 0.653168547153473, "time": 0.4285785675048828, "epoch": 17, "memory": 20334, "step": 97528} +{"lr": 0.00010453658778440108, "data_time": 0.0016847372055053711, "grad_norm": 16.86105582714081, "loss": 0.6359785616397857, "time": 0.3951293706893921, "epoch": 17, "memory": 20334, "step": 97628} +{"lr": 0.00010453658778440108, "data_time": 0.002587628364562988, "grad_norm": 43.33133211135864, "loss": 0.6361349105834961, "time": 0.418013596534729, "epoch": 17, "memory": 20334, "step": 97728} +{"lr": 0.00010453658778440108, "data_time": 0.0019317388534545899, "grad_norm": 32.4199652671814, "loss": 0.642455643415451, "time": 0.40132341384887693, "epoch": 17, "memory": 20334, "step": 97828} +{"accuracy/top1": 88.11312103271484, "data_time": 0.3581188784705268, "time": 0.9833702511257596, "step": 17} +{"lr": 6.395177052675796e-05, "data_time": 0.002099752426147461, "grad_norm": 63.28272061347961, "loss": 0.6444731771945953, "time": 0.4256767272949219, "epoch": 18, "memory": 20334, "step": 97986} +{"lr": 6.395177052675796e-05, "data_time": 0.002711629867553711, "grad_norm": 71.94749855995178, "loss": 0.6376009404659271, "time": 0.39515902996063235, "epoch": 18, "memory": 20334, "step": 98086} +{"lr": 6.395177052675796e-05, "data_time": 0.0019078969955444336, "grad_norm": 27.969100666046142, "loss": 0.6339970886707306, "time": 0.4254409074783325, "epoch": 18, "memory": 20334, "step": 98186} +{"lr": 6.395177052675796e-05, "data_time": 0.0029084444046020507, "grad_norm": 33.11596373319626, "loss": 0.6604844689369201, "time": 0.4136345624923706, "epoch": 18, "memory": 20334, "step": 98286} +{"lr": 6.395177052675796e-05, "data_time": 0.002246594429016113, "grad_norm": 21.222491908073426, "loss": 0.6445057034492493, "time": 0.37490484714508054, "epoch": 18, "memory": 20334, "step": 98386} +{"lr": 6.395177052675796e-05, "data_time": 0.0019988059997558595, "grad_norm": 75.3499174118042, "loss": 0.6528640747070312, "time": 0.4187065601348877, "epoch": 18, "memory": 20334, "step": 98486} +{"lr": 6.395177052675796e-05, "data_time": 0.0018184900283813477, "grad_norm": 131.6462007522583, "loss": 0.6250163376331329, "time": 0.4211371421813965, "epoch": 18, "memory": 20334, "step": 98586} +{"lr": 6.395177052675796e-05, "data_time": 0.0017574310302734375, "grad_norm": 71.85484886169434, "loss": 0.6417450666427612, "time": 0.37882812023162843, "epoch": 18, "memory": 20334, "step": 98686} +{"lr": 6.395177052675796e-05, "data_time": 0.0017748355865478515, "grad_norm": 14.120350122451782, "loss": 0.6532712817192078, "time": 0.3884144306182861, "epoch": 18, "memory": 20334, "step": 98786} +{"lr": 6.395177052675796e-05, "data_time": 0.002509307861328125, "grad_norm": 82.07881389856338, "loss": 0.644244772195816, "time": 0.4121937036514282, "epoch": 18, "memory": 20334, "step": 98886} +{"lr": 6.395177052675796e-05, "data_time": 0.0025970935821533203, "grad_norm": 17.074625968933105, "loss": 0.648538988828659, "time": 0.4318960905075073, "epoch": 18, "memory": 20334, "step": 98986} +{"lr": 6.395177052675796e-05, "data_time": 0.0019234180450439452, "grad_norm": 12.55678596496582, "loss": 0.6545386016368866, "time": 0.39982435703277586, "epoch": 18, "memory": 20334, "step": 99086} +{"lr": 6.395177052675796e-05, "data_time": 0.0024192333221435547, "grad_norm": 43.79109058380127, "loss": 0.6471724450588227, "time": 0.3826723337173462, "epoch": 18, "memory": 20334, "step": 99186} +{"lr": 6.395177052675796e-05, "data_time": 0.0030316591262817385, "grad_norm": 19.45570340156555, "loss": 0.6481290102005005, "time": 0.3795777320861816, "epoch": 18, "memory": 20334, "step": 99286} +{"lr": 6.395177052675796e-05, "data_time": 0.0015474319458007812, "grad_norm": 19.831133008003235, "loss": 0.6461710214614869, "time": 0.4128087520599365, "epoch": 18, "memory": 20334, "step": 99386} +{"lr": 6.395177052675796e-05, "data_time": 0.0017799615859985351, "grad_norm": 30.964549469947816, "loss": 0.6446856379508972, "time": 0.4162802219390869, "epoch": 18, "memory": 20334, "step": 99486} +{"lr": 6.395177052675796e-05, "data_time": 0.0023103952407836914, "grad_norm": 20.136412382125854, "loss": 0.6416927456855774, "time": 0.45120725631713865, "epoch": 18, "memory": 20334, "step": 99586} +{"lr": 6.395177052675796e-05, "data_time": 0.0024947404861450197, "grad_norm": 32.40307134389877, "loss": 0.644729620218277, "time": 0.3843658924102783, "epoch": 18, "memory": 20334, "step": 99686} +{"lr": 6.395177052675796e-05, "data_time": 0.0022699832916259766, "grad_norm": 14.265842127799989, "loss": 0.6528850615024566, "time": 0.3970250368118286, "epoch": 18, "memory": 20334, "step": 99786} +{"lr": 6.395177052675796e-05, "data_time": 0.0020752191543579102, "grad_norm": 20.699154913425446, "loss": 0.6426553785800934, "time": 0.3929337501525879, "epoch": 18, "memory": 20334, "step": 99886} +{"lr": 6.395177052675796e-05, "data_time": 0.0026435613632202148, "grad_norm": 70.5440259218216, "loss": 0.6593203127384186, "time": 0.415566611289978, "epoch": 18, "memory": 20334, "step": 99986} +{"lr": 6.395177052675796e-05, "data_time": 0.002365422248840332, "grad_norm": 21.333938694000246, "loss": 0.6492511808872223, "time": 0.3907930374145508, "epoch": 18, "memory": 20334, "step": 100086} +{"lr": 6.395177052675796e-05, "data_time": 0.002691984176635742, "grad_norm": 27.271159529685974, "loss": 0.6560031712055207, "time": 0.3996977090835571, "epoch": 18, "memory": 20334, "step": 100186} +{"lr": 6.395177052675796e-05, "data_time": 0.0022272109985351563, "grad_norm": 25.059144473075868, "loss": 0.6597813546657563, "time": 0.3754384756088257, "epoch": 18, "memory": 20334, "step": 100286} +{"lr": 6.395177052675796e-05, "data_time": 0.001793241500854492, "grad_norm": 26.656696462631224, "loss": 0.6416965544223785, "time": 0.3689321994781494, "epoch": 18, "memory": 20334, "step": 100386} +{"lr": 6.395177052675796e-05, "data_time": 0.0029100418090820313, "grad_norm": 44.697294569015504, "loss": 0.6489736020565033, "time": 0.3945737361907959, "epoch": 18, "memory": 20334, "step": 100486} +{"lr": 6.395177052675796e-05, "data_time": 0.0018138647079467773, "grad_norm": 225.24888908863068, "loss": 0.6620129823684693, "time": 0.39425723552703856, "epoch": 18, "memory": 20334, "step": 100586} +{"lr": 6.395177052675796e-05, "data_time": 0.001793074607849121, "grad_norm": 83.48510959148408, "loss": 0.6606607139110565, "time": 0.3993523359298706, "epoch": 18, "memory": 20334, "step": 100686} +{"lr": 6.395177052675796e-05, "data_time": 0.0022084951400756837, "grad_norm": 44.040564727783206, "loss": 0.6465081810951233, "time": 0.3778827667236328, "epoch": 18, "memory": 20334, "step": 100786} +{"lr": 6.395177052675796e-05, "data_time": 0.003248929977416992, "grad_norm": 24.54559905529022, "loss": 0.6401390194892883, "time": 0.4228460073471069, "epoch": 18, "memory": 20334, "step": 100886} +{"lr": 6.395177052675796e-05, "data_time": 0.0018243074417114257, "grad_norm": 128.91987414360045, "loss": 0.6633709013462067, "time": 0.3932612657546997, "epoch": 18, "memory": 20334, "step": 100986} +{"lr": 6.395177052675796e-05, "data_time": 0.002174234390258789, "grad_norm": 152.1927219390869, "loss": 0.6574654817581177, "time": 0.40231847763061523, "epoch": 18, "memory": 20334, "step": 101086} +{"lr": 6.395177052675796e-05, "data_time": 0.0017493724822998046, "grad_norm": 34.541743755340576, "loss": 0.6583094596862793, "time": 0.3871058702468872, "epoch": 18, "memory": 20334, "step": 101186} +{"lr": 6.395177052675796e-05, "data_time": 0.0021985054016113283, "grad_norm": 10.845146322250367, "loss": 0.6594466626644134, "time": 0.3989195108413696, "epoch": 18, "memory": 20334, "step": 101286} +{"lr": 6.395177052675796e-05, "data_time": 0.002009272575378418, "grad_norm": 136.56633160114288, "loss": 0.6515347182750701, "time": 0.4141709327697754, "epoch": 18, "memory": 20334, "step": 101386} +{"lr": 6.395177052675796e-05, "data_time": 0.001636648178100586, "grad_norm": 26.99376971721649, "loss": 0.6600113630294799, "time": 0.39459292888641356, "epoch": 18, "memory": 20334, "step": 101486} +{"lr": 6.395177052675796e-05, "data_time": 0.002132105827331543, "grad_norm": 32.21519119739533, "loss": 0.6731671214103698, "time": 0.39255034923553467, "epoch": 18, "memory": 20334, "step": 101586} +{"lr": 6.395177052675796e-05, "data_time": 0.0029949665069580076, "grad_norm": 63.9368953704834, "loss": 0.661777114868164, "time": 0.41479313373565674, "epoch": 18, "memory": 20334, "step": 101686} +{"lr": 6.395177052675796e-05, "data_time": 0.0028112649917602537, "grad_norm": 25.9544114112854, "loss": 0.6740711569786072, "time": 0.41132607460021975, "epoch": 18, "memory": 20334, "step": 101786} +{"lr": 6.395177052675796e-05, "data_time": 0.001844930648803711, "grad_norm": 71.69706101417542, "loss": 0.6596127629280091, "time": 0.43996922969818114, "epoch": 18, "memory": 20334, "step": 101886} +{"lr": 6.395177052675796e-05, "data_time": 0.002105236053466797, "grad_norm": 46.557479810714725, "loss": 0.6667856931686401, "time": 0.39024529457092283, "epoch": 18, "memory": 20334, "step": 101986} +{"lr": 6.395177052675796e-05, "data_time": 0.002191972732543945, "grad_norm": 102.11906833648682, "loss": 0.6647959530353547, "time": 0.4282243728637695, "epoch": 18, "memory": 20334, "step": 102086} +{"lr": 6.395177052675796e-05, "data_time": 0.0016839027404785157, "grad_norm": 91.0549749135971, "loss": 0.6600245058536529, "time": 0.3853638172149658, "epoch": 18, "memory": 20334, "step": 102186} +{"lr": 6.395177052675796e-05, "data_time": 0.0019200563430786133, "grad_norm": 40.99918479919434, "loss": 0.654580557346344, "time": 0.38777430057525636, "epoch": 18, "memory": 20334, "step": 102286} +{"lr": 6.395177052675796e-05, "data_time": 0.0018175363540649414, "grad_norm": 58.056387066841125, "loss": 0.6499052882194519, "time": 0.41110901832580565, "epoch": 18, "memory": 20334, "step": 102386} +{"lr": 6.395177052675796e-05, "data_time": 0.002111387252807617, "grad_norm": 38.41582188606262, "loss": 0.6536898493766785, "time": 0.4120710611343384, "epoch": 18, "memory": 20334, "step": 102486} +{"lr": 6.395177052675796e-05, "data_time": 0.002192568778991699, "grad_norm": 47.87467775344849, "loss": 0.6521516799926758, "time": 0.4433310508728027, "epoch": 18, "memory": 20334, "step": 102586} +{"lr": 6.395177052675796e-05, "data_time": 0.002946758270263672, "grad_norm": 42.65680613517761, "loss": 0.6453225374221802, "time": 0.4109247446060181, "epoch": 18, "memory": 20334, "step": 102686} +{"lr": 6.395177052675796e-05, "data_time": 0.0020146846771240236, "grad_norm": 59.34259493350983, "loss": 0.6423458158969879, "time": 0.43424973487854, "epoch": 18, "memory": 20334, "step": 102786} +{"lr": 6.395177052675796e-05, "data_time": 0.0016765117645263672, "grad_norm": 15.614595317840577, "loss": 0.6565988421440124, "time": 0.39060814380645753, "epoch": 18, "memory": 20334, "step": 102886} +{"lr": 6.395177052675796e-05, "data_time": 0.00270843505859375, "grad_norm": 21.659555768966676, "loss": 0.6484413981437683, "time": 0.3961986780166626, "epoch": 18, "memory": 20334, "step": 102986} +{"lr": 6.395177052675796e-05, "data_time": 0.002336454391479492, "grad_norm": 296.36976046562194, "loss": 0.6487852931022644, "time": 0.380599308013916, "epoch": 18, "memory": 20334, "step": 103086} +{"lr": 6.395177052675796e-05, "data_time": 0.002427506446838379, "grad_norm": 171.42702980041503, "loss": 0.6475272834300995, "time": 0.4200688123703003, "epoch": 18, "memory": 20334, "step": 103186} +{"lr": 6.395177052675796e-05, "data_time": 0.0019548416137695314, "grad_norm": 15.770891356468201, "loss": 0.6601642727851867, "time": 0.4168720722198486, "epoch": 18, "memory": 20334, "step": 103286} +{"lr": 6.395177052675796e-05, "data_time": 0.002973628044128418, "grad_norm": 15.751076912879943, "loss": 0.6604206025600433, "time": 0.41751086711883545, "epoch": 18, "memory": 20334, "step": 103386} +{"lr": 6.395177052675796e-05, "data_time": 0.002408409118652344, "grad_norm": 16.845925784111024, "loss": 0.6536760210990906, "time": 0.40938050746917726, "epoch": 18, "memory": 20334, "step": 103486} +{"lr": 6.395177052675796e-05, "data_time": 0.0019280433654785157, "grad_norm": 27.657923913002016, "loss": 0.6490884304046631, "time": 0.40270171165466306, "epoch": 18, "memory": 20334, "step": 103586} +{"accuracy/top1": 86.98945617675781, "data_time": 0.36803950203789604, "time": 0.9957342147827148, "step": 18} +{"lr": 3.422702443389901e-05, "data_time": 0.0019544124603271484, "grad_norm": 66.65392987728119, "loss": 0.651543265581131, "time": 0.40294613838195803, "epoch": 19, "memory": 20334, "step": 103744} +{"lr": 3.422702443389901e-05, "data_time": 0.0018543004989624023, "grad_norm": 34.74554295539856, "loss": 0.6643685579299927, "time": 0.4082931041717529, "epoch": 19, "memory": 20334, "step": 103844} +{"lr": 3.422702443389901e-05, "data_time": 0.0020318031311035156, "grad_norm": 213.33866076469423, "loss": 0.6748937785625457, "time": 0.39371323585510254, "epoch": 19, "memory": 20334, "step": 103944} +{"lr": 3.422702443389901e-05, "data_time": 0.001800370216369629, "grad_norm": 74.37764189243316, "loss": 0.6605765461921692, "time": 0.41226332187652587, "epoch": 19, "memory": 20334, "step": 104044} +{"lr": 3.422702443389901e-05, "data_time": 0.0019049644470214844, "grad_norm": 89.74551241397857, "loss": 0.6422040402889252, "time": 0.44047772884368896, "epoch": 19, "memory": 20334, "step": 104144} +{"lr": 3.422702443389901e-05, "data_time": 0.0017130851745605468, "grad_norm": 40.855762004852295, "loss": 0.6535026669502259, "time": 0.4267627716064453, "epoch": 19, "memory": 20334, "step": 104244} +{"lr": 3.422702443389901e-05, "data_time": 0.0017953872680664062, "grad_norm": 15.970692420005799, "loss": 0.6586434721946717, "time": 0.3833063364028931, "epoch": 19, "memory": 20334, "step": 104344} +{"lr": 3.422702443389901e-05, "data_time": 0.0017811059951782227, "grad_norm": 18.287867546081543, "loss": 0.6528598010540009, "time": 0.41804656982421873, "epoch": 19, "memory": 20334, "step": 104444} +{"lr": 3.422702443389901e-05, "data_time": 0.0017637491226196289, "grad_norm": 14.304183197021484, "loss": 0.6278103768825531, "time": 0.3997077941894531, "epoch": 19, "memory": 20334, "step": 104544} +{"lr": 3.422702443389901e-05, "data_time": 0.0018087387084960937, "grad_norm": 9.44283344745636, "loss": 0.6413831651210785, "time": 0.39129114151000977, "epoch": 19, "memory": 20334, "step": 104644} +{"lr": 3.422702443389901e-05, "data_time": 0.0018803596496582032, "grad_norm": 40.7143635392189, "loss": 0.6593172371387481, "time": 0.39586400985717773, "epoch": 19, "memory": 20334, "step": 104744} +{"lr": 3.422702443389901e-05, "data_time": 0.0023250341415405273, "grad_norm": 13.627595686912537, "loss": 0.6552824079990387, "time": 0.3840524911880493, "epoch": 19, "memory": 20334, "step": 104844} +{"lr": 3.422702443389901e-05, "data_time": 0.002281975746154785, "grad_norm": 14.279818868637085, "loss": 0.6424823582172394, "time": 0.4203862190246582, "epoch": 19, "memory": 20334, "step": 104944} +{"lr": 3.422702443389901e-05, "data_time": 0.0019212007522583009, "grad_norm": 61.65380027294159, "loss": 0.6414035320281982, "time": 0.40967159271240233, "epoch": 19, "memory": 20334, "step": 105044} +{"lr": 3.422702443389901e-05, "data_time": 0.0016926765441894532, "grad_norm": 19.834644961357117, "loss": 0.6525056838989258, "time": 0.3920203447341919, "epoch": 19, "memory": 20334, "step": 105144} +{"lr": 3.422702443389901e-05, "data_time": 0.002080535888671875, "grad_norm": 21.97901842594147, "loss": 0.6534707963466644, "time": 0.36706583499908446, "epoch": 19, "memory": 20334, "step": 105244} +{"lr": 3.422702443389901e-05, "data_time": 0.002734875679016113, "grad_norm": 9.230246782302856, "loss": 0.639169204235077, "time": 0.4106771230697632, "epoch": 19, "memory": 20334, "step": 105344} +{"lr": 3.422702443389901e-05, "data_time": 0.0027134895324707033, "grad_norm": 25.478613805770873, "loss": 0.6500572264194489, "time": 0.41395514011383056, "epoch": 19, "memory": 20334, "step": 105444} +{"lr": 3.422702443389901e-05, "data_time": 0.0018291950225830078, "grad_norm": 31.823949766159057, "loss": 0.6490461707115174, "time": 0.4218700885772705, "epoch": 19, "memory": 20334, "step": 105544} +{"lr": 3.422702443389901e-05, "data_time": 0.0017975330352783202, "grad_norm": 42.61731996536255, "loss": 0.6465230882167816, "time": 0.3886647939682007, "epoch": 19, "memory": 20334, "step": 105644} +{"lr": 3.422702443389901e-05, "data_time": 0.002008199691772461, "grad_norm": 23.15601212978363, "loss": 0.657354474067688, "time": 0.4120891809463501, "epoch": 19, "memory": 20334, "step": 105744} +{"lr": 3.422702443389901e-05, "data_time": 0.0023223400115966798, "grad_norm": 22.897148180007935, "loss": 0.6566392123699188, "time": 0.400227427482605, "epoch": 19, "memory": 20334, "step": 105844} +{"lr": 3.422702443389901e-05, "data_time": 0.0017266035079956054, "grad_norm": 30.917552328109743, "loss": 0.655597847700119, "time": 0.38207459449768066, "epoch": 19, "memory": 20334, "step": 105944} +{"lr": 3.422702443389901e-05, "data_time": 0.0018705606460571289, "grad_norm": 41.59244341850281, "loss": 0.6586847841739655, "time": 0.36503379344940184, "epoch": 19, "memory": 20334, "step": 106044} +{"lr": 3.422702443389901e-05, "data_time": 0.0018585443496704102, "grad_norm": 33.95430791378021, "loss": 0.6456314325332642, "time": 0.46921491622924805, "epoch": 19, "memory": 20334, "step": 106144} +{"lr": 3.422702443389901e-05, "data_time": 0.0017628669738769531, "grad_norm": 40.95303258895874, "loss": 0.6533273994922638, "time": 0.39267683029174805, "epoch": 19, "memory": 20334, "step": 106244} +{"lr": 3.422702443389901e-05, "data_time": 0.0017978429794311523, "grad_norm": 14.215553760528564, "loss": 0.6592978179454804, "time": 0.4042861223220825, "epoch": 19, "memory": 20334, "step": 106344} +{"lr": 3.422702443389901e-05, "data_time": 0.0019912958145141602, "grad_norm": 143.317347240448, "loss": 0.648093169927597, "time": 0.401861572265625, "epoch": 19, "memory": 20334, "step": 106444} +{"lr": 3.422702443389901e-05, "data_time": 0.0020437002182006835, "grad_norm": 40.78470206260681, "loss": 0.6658326268196106, "time": 0.43019914627075195, "epoch": 19, "memory": 20334, "step": 106544} +{"lr": 3.422702443389901e-05, "data_time": 0.001868128776550293, "grad_norm": 19.64283421039581, "loss": 0.6576982378959656, "time": 0.41136808395385743, "epoch": 19, "memory": 20334, "step": 106644} +{"lr": 3.422702443389901e-05, "data_time": 0.0015358209609985351, "grad_norm": 75.87776384353637, "loss": 0.6442136585712432, "time": 0.40893604755401614, "epoch": 19, "memory": 20334, "step": 106744} +{"lr": 3.422702443389901e-05, "data_time": 0.0015146255493164063, "grad_norm": 12.315547037124634, "loss": 0.6521753191947937, "time": 0.4102999448776245, "epoch": 19, "memory": 20334, "step": 106844} +{"lr": 3.422702443389901e-05, "data_time": 0.0017220020294189454, "grad_norm": 14.519731771945953, "loss": 0.6572687864303589, "time": 0.4080089330673218, "epoch": 19, "memory": 20334, "step": 106944} +{"lr": 3.422702443389901e-05, "data_time": 0.002236604690551758, "grad_norm": 49.558367705345155, "loss": 0.6520628035068512, "time": 0.3836507320404053, "epoch": 19, "memory": 20334, "step": 107044} +{"lr": 3.422702443389901e-05, "data_time": 0.0017132282257080079, "grad_norm": 34.72822256088257, "loss": 0.6657291710376739, "time": 0.3867108106613159, "epoch": 19, "memory": 20334, "step": 107144} +{"lr": 3.422702443389901e-05, "data_time": 0.0016370773315429687, "grad_norm": 17.021578669548035, "loss": 0.6683961629867554, "time": 0.38309948444366454, "epoch": 19, "memory": 20334, "step": 107244} +{"lr": 3.422702443389901e-05, "data_time": 0.0016996622085571288, "grad_norm": 30.026414847373964, "loss": 0.6664615154266358, "time": 0.39449048042297363, "epoch": 19, "memory": 20334, "step": 107344} +{"lr": 3.422702443389901e-05, "data_time": 0.0016473054885864258, "grad_norm": 19.32567903995514, "loss": 0.6644317507743835, "time": 0.3868776559829712, "epoch": 19, "memory": 20334, "step": 107444} +{"lr": 3.422702443389901e-05, "data_time": 0.0016041278839111328, "grad_norm": 36.1685825586319, "loss": 0.672652804851532, "time": 0.4217507839202881, "epoch": 19, "memory": 20334, "step": 107544} +{"lr": 3.422702443389901e-05, "data_time": 0.0017470598220825195, "grad_norm": 12.698266100883483, "loss": 0.6664849102497101, "time": 0.44067671298980715, "epoch": 19, "memory": 20334, "step": 107644} +{"lr": 3.422702443389901e-05, "data_time": 0.0016474962234497071, "grad_norm": 18.58142523765564, "loss": 0.6727229356765747, "time": 0.4180203676223755, "epoch": 19, "memory": 20334, "step": 107744} +{"lr": 3.422702443389901e-05, "data_time": 0.001585698127746582, "grad_norm": 40.603049755096436, "loss": 0.6680902898311615, "time": 0.4268414735794067, "epoch": 19, "memory": 20334, "step": 107844} +{"lr": 3.422702443389901e-05, "data_time": 0.0016061782836914063, "grad_norm": 85.02847175598144, "loss": 0.6685926496982575, "time": 0.4036363124847412, "epoch": 19, "memory": 20334, "step": 107944} +{"lr": 3.422702443389901e-05, "data_time": 0.0015764951705932618, "grad_norm": 62.724023127555846, "loss": 0.6558150768280029, "time": 0.44412014484405515, "epoch": 19, "memory": 20334, "step": 108044} +{"lr": 3.422702443389901e-05, "data_time": 0.0017894506454467773, "grad_norm": 99.84025325775147, "loss": 0.6741968870162964, "time": 0.4474414587020874, "epoch": 19, "memory": 20334, "step": 108144} +{"lr": 3.422702443389901e-05, "data_time": 0.0016388177871704101, "grad_norm": 57.71546268463135, "loss": 0.6721916258335113, "time": 0.43043463230133056, "epoch": 19, "memory": 20334, "step": 108244} +{"lr": 3.422702443389901e-05, "data_time": 0.0017370462417602539, "grad_norm": 80.37630853652954, "loss": 0.6624933481216431, "time": 0.4322895765304565, "epoch": 19, "memory": 20334, "step": 108344} +{"lr": 3.422702443389901e-05, "data_time": 0.0020455837249755858, "grad_norm": 62.80916078090668, "loss": 0.6744026124477387, "time": 0.3881272077560425, "epoch": 19, "memory": 20334, "step": 108444} +{"lr": 3.422702443389901e-05, "data_time": 0.0019963979721069336, "grad_norm": 31.45765166282654, "loss": 0.6706329762935639, "time": 0.3930949687957764, "epoch": 19, "memory": 20334, "step": 108544} +{"lr": 3.422702443389901e-05, "data_time": 0.001952385902404785, "grad_norm": 56.2090705871582, "loss": 0.6598998725414276, "time": 0.41112775802612306, "epoch": 19, "memory": 20334, "step": 108644} +{"lr": 3.422702443389901e-05, "data_time": 0.001448655128479004, "grad_norm": 206.15071606636047, "loss": 0.673072224855423, "time": 0.4001233339309692, "epoch": 19, "memory": 20334, "step": 108744} +{"lr": 3.422702443389901e-05, "data_time": 0.003635668754577637, "grad_norm": 44.87873637676239, "loss": 0.6734757721424103, "time": 0.3866569995880127, "epoch": 19, "memory": 20334, "step": 108844} +{"lr": 3.422702443389901e-05, "data_time": 0.002407217025756836, "grad_norm": 79.70904369354248, "loss": 0.6719061970710755, "time": 0.4129964351654053, "epoch": 19, "memory": 20334, "step": 108944} +{"lr": 3.422702443389901e-05, "data_time": 0.0019218921661376953, "grad_norm": 58.01592597961426, "loss": 0.6579764366149903, "time": 0.400122857093811, "epoch": 19, "memory": 20334, "step": 109044} +{"lr": 3.422702443389901e-05, "data_time": 0.0021574020385742186, "grad_norm": 86.64260444641113, "loss": 0.6737904250621796, "time": 0.385427188873291, "epoch": 19, "memory": 20334, "step": 109144} +{"lr": 3.422702443389901e-05, "data_time": 0.0017372608184814454, "grad_norm": 38.41540155410767, "loss": 0.6705297827720642, "time": 0.3884838581085205, "epoch": 19, "memory": 20334, "step": 109244} +{"lr": 3.422702443389901e-05, "data_time": 0.0017900943756103515, "grad_norm": 203.17119026184082, "loss": 0.6733665406703949, "time": 0.39952883720397947, "epoch": 19, "memory": 20334, "step": 109344} +{"accuracy/top1": 85.834228515625, "data_time": 0.35817890697055393, "time": 0.9835113949245877, "step": 19} +{"lr": 1.6094271405406862e-05, "data_time": 0.0018102407455444336, "grad_norm": 37.83160743713379, "loss": 0.6657364547252655, "time": 0.3837099313735962, "epoch": 20, "memory": 20334, "step": 109502} +{"lr": 1.6094271405406862e-05, "data_time": 0.0023318052291870115, "grad_norm": 131.83887491226196, "loss": 0.6685264825820922, "time": 0.36237781047821044, "epoch": 20, "memory": 20334, "step": 109602} +{"lr": 1.6094271405406862e-05, "data_time": 0.0031822681427001952, "grad_norm": 73.49366936683654, "loss": 0.6725376188755036, "time": 0.4056878089904785, "epoch": 20, "memory": 20334, "step": 109702} +{"lr": 1.6094271405406862e-05, "data_time": 0.00291903018951416, "grad_norm": 48.69134254455567, "loss": 0.6697592377662659, "time": 0.3942695617675781, "epoch": 20, "memory": 20334, "step": 109802} +{"lr": 1.6094271405406862e-05, "data_time": 0.0024376153945922852, "grad_norm": 47.049586868286134, "loss": 0.6680775642395019, "time": 0.38937761783599856, "epoch": 20, "memory": 20334, "step": 109902} +{"lr": 1.6094271405406862e-05, "data_time": 0.0023961782455444334, "grad_norm": 1127.3899322509765, "loss": 0.673331344127655, "time": 0.38590729236602783, "epoch": 20, "memory": 20334, "step": 110002} +{"lr": 1.6094271405406862e-05, "data_time": 0.0023725748062133787, "grad_norm": 82.14511423110962, "loss": 0.6698859632015228, "time": 0.41763460636138916, "epoch": 20, "memory": 20334, "step": 110102} +{"lr": 1.6094271405406862e-05, "data_time": 0.0016741037368774414, "grad_norm": 11.355046510696411, "loss": 0.6541194677352905, "time": 0.4403804063796997, "epoch": 20, "memory": 20334, "step": 110202} +{"lr": 1.6094271405406862e-05, "data_time": 0.00211186408996582, "grad_norm": 44.58271446228027, "loss": 0.6699179112911224, "time": 0.38148767948150636, "epoch": 20, "memory": 20334, "step": 110302} +{"lr": 1.6094271405406862e-05, "data_time": 0.0019808530807495115, "grad_norm": 181.1916173696518, "loss": 0.6703812181949615, "time": 0.38389732837677004, "epoch": 20, "memory": 20334, "step": 110402} +{"lr": 1.6094271405406862e-05, "data_time": 0.0025669097900390624, "grad_norm": 29.379246282577515, "loss": 0.6694718360900879, "time": 0.40802459716796874, "epoch": 20, "memory": 20334, "step": 110502} +{"lr": 1.6094271405406862e-05, "data_time": 0.0019008636474609375, "grad_norm": 112.64701852798461, "loss": 0.6737649381160736, "time": 0.3869785308837891, "epoch": 20, "memory": 20334, "step": 110602} +{"lr": 1.6094271405406862e-05, "data_time": 0.002530479431152344, "grad_norm": 97.6847596168518, "loss": 0.6756111741065979, "time": 0.3776059865951538, "epoch": 20, "memory": 20334, "step": 110702} +{"lr": 1.6094271405406862e-05, "data_time": 0.0016866683959960937, "grad_norm": 71.38228597640992, "loss": 0.6757996320724488, "time": 0.3918351173400879, "epoch": 20, "memory": 20334, "step": 110802} +{"lr": 1.6094271405406862e-05, "data_time": 0.0030757904052734373, "grad_norm": 12.370378637313843, "loss": 0.6697602152824402, "time": 0.4175167798995972, "epoch": 20, "memory": 20334, "step": 110902} +{"lr": 1.6094271405406862e-05, "data_time": 0.0020238876342773436, "grad_norm": 217.78964290618896, "loss": 0.6711635291576385, "time": 0.40938963890075686, "epoch": 20, "memory": 20334, "step": 111002} +{"lr": 1.6094271405406862e-05, "data_time": 0.0020236730575561523, "grad_norm": 75.23270325660705, "loss": 0.6697175443172455, "time": 0.4169180393218994, "epoch": 20, "memory": 20334, "step": 111102} +{"lr": 1.6094271405406862e-05, "data_time": 0.002639508247375488, "grad_norm": 49.73284821510315, "loss": 0.666726416349411, "time": 0.38310723304748534, "epoch": 20, "memory": 20334, "step": 111202} +{"lr": 1.6094271405406862e-05, "data_time": 0.0017825603485107423, "grad_norm": 33.947696113586424, "loss": 0.6709316611289978, "time": 0.3818256378173828, "epoch": 20, "memory": 20334, "step": 111302} +{"lr": 1.6094271405406862e-05, "data_time": 0.0017880678176879882, "grad_norm": 148.23197169303893, "loss": 0.6696382522583008, "time": 0.39011049270629883, "epoch": 20, "memory": 20334, "step": 111402} +{"lr": 1.6094271405406862e-05, "data_time": 0.002052927017211914, "grad_norm": 24.143499755859374, "loss": 0.6763786494731903, "time": 0.38851616382598875, "epoch": 20, "memory": 20334, "step": 111502} +{"lr": 1.6094271405406862e-05, "data_time": 0.0018058538436889649, "grad_norm": 29.874695301055908, "loss": 0.6745240867137909, "time": 0.3702345848083496, "epoch": 20, "memory": 20334, "step": 111602} +{"lr": 1.6094271405406862e-05, "data_time": 0.0020065069198608398, "grad_norm": 91.62142691612243, "loss": 0.6710270464420318, "time": 0.43645074367523196, "epoch": 20, "memory": 20334, "step": 111702} +{"lr": 1.6094271405406862e-05, "data_time": 0.0017354965209960937, "grad_norm": 86.21442022323609, "loss": 0.6670799493789673, "time": 0.41526498794555666, "epoch": 20, "memory": 20334, "step": 111802} +{"lr": 1.6094271405406862e-05, "data_time": 0.0025794506072998047, "grad_norm": 304.8029176235199, "loss": 0.6622025430202484, "time": 0.4083908796310425, "epoch": 20, "memory": 20334, "step": 111902} +{"lr": 1.6094271405406862e-05, "data_time": 0.002175760269165039, "grad_norm": 97.69507846832275, "loss": 0.6739909708499908, "time": 0.4042088508605957, "epoch": 20, "memory": 20334, "step": 112002} +{"lr": 1.6094271405406862e-05, "data_time": 0.0019194126129150391, "grad_norm": 93.46606841087342, "loss": 0.6579058706760407, "time": 0.3818637847900391, "epoch": 20, "memory": 20334, "step": 112102} +{"lr": 1.6094271405406862e-05, "data_time": 0.0018822431564331054, "grad_norm": 38.137491941452026, "loss": 0.664556086063385, "time": 0.37654929161071776, "epoch": 20, "memory": 20334, "step": 112202} +{"lr": 1.6094271405406862e-05, "data_time": 0.002132606506347656, "grad_norm": 38.68783128261566, "loss": 0.6653063118457794, "time": 0.4129112482070923, "epoch": 20, "memory": 20334, "step": 112302} +{"lr": 1.6094271405406862e-05, "data_time": 0.0021467685699462892, "grad_norm": 22.53668384552002, "loss": 0.675944036245346, "time": 0.3858873128890991, "epoch": 20, "memory": 20334, "step": 112402} +{"lr": 1.6094271405406862e-05, "data_time": 0.0020844697952270507, "grad_norm": 51.04946417808533, "loss": 0.6672535479068756, "time": 0.4299804210662842, "epoch": 20, "memory": 20334, "step": 112502} +{"lr": 1.6094271405406862e-05, "data_time": 0.0032830238342285156, "grad_norm": 46.97947225570679, "loss": 0.6686098754405976, "time": 0.40066864490509035, "epoch": 20, "memory": 20334, "step": 112602} +{"lr": 1.6094271405406862e-05, "data_time": 0.0025171756744384764, "grad_norm": 94.8532820224762, "loss": 0.6783220827579498, "time": 0.3917621850967407, "epoch": 20, "memory": 20334, "step": 112702} +{"lr": 1.6094271405406862e-05, "data_time": 0.003272390365600586, "grad_norm": 36.096519947052, "loss": 0.6773834824562073, "time": 0.38307321071624756, "epoch": 20, "memory": 20334, "step": 112802} +{"lr": 1.6094271405406862e-05, "data_time": 0.001750349998474121, "grad_norm": 14.777117347717285, "loss": 0.669330495595932, "time": 0.38584651947021487, "epoch": 20, "memory": 20334, "step": 112902} +{"lr": 1.6094271405406862e-05, "data_time": 0.0025300025939941407, "grad_norm": 48.91781497001648, "loss": 0.682118809223175, "time": 0.40066051483154297, "epoch": 20, "memory": 20334, "step": 113002} +{"lr": 1.6094271405406862e-05, "data_time": 0.002970170974731445, "grad_norm": 41.183049011230466, "loss": 0.6624325454235077, "time": 0.41779313087463377, "epoch": 20, "memory": 20334, "step": 113102} +{"lr": 1.6094271405406862e-05, "data_time": 0.0019422292709350586, "grad_norm": 49.42296929359436, "loss": 0.6682762026786804, "time": 0.3991267681121826, "epoch": 20, "memory": 20334, "step": 113202} +{"lr": 1.6094271405406862e-05, "data_time": 0.0018179893493652343, "grad_norm": 61.340555572509764, "loss": 0.6697454810142517, "time": 0.38625288009643555, "epoch": 20, "memory": 20334, "step": 113302} +{"lr": 1.6094271405406862e-05, "data_time": 0.0020168542861938475, "grad_norm": 20.791267371177675, "loss": 0.6624826729297638, "time": 0.43474817276000977, "epoch": 20, "memory": 20334, "step": 113402} +{"lr": 1.6094271405406862e-05, "data_time": 0.0022972583770751952, "grad_norm": 58.92897605895996, "loss": 0.6696829795837402, "time": 0.41529862880706786, "epoch": 20, "memory": 20334, "step": 113502} +{"lr": 1.6094271405406862e-05, "data_time": 0.002086949348449707, "grad_norm": 18.95285291671753, "loss": 0.6847931981086731, "time": 0.39669802188873293, "epoch": 20, "memory": 20334, "step": 113602} +{"lr": 1.6094271405406862e-05, "data_time": 0.0021726131439208985, "grad_norm": 20.612304973602296, "loss": 0.679996770620346, "time": 0.386553692817688, "epoch": 20, "memory": 20334, "step": 113702} +{"lr": 1.6094271405406862e-05, "data_time": 0.001727747917175293, "grad_norm": 46.01411848068237, "loss": 0.6571497023105621, "time": 0.4148284435272217, "epoch": 20, "memory": 20334, "step": 113802} +{"lr": 1.6094271405406862e-05, "data_time": 0.0020232439041137696, "grad_norm": 135.2398318529129, "loss": 0.6718491733074188, "time": 0.4023731708526611, "epoch": 20, "memory": 20334, "step": 113902} +{"lr": 1.6094271405406862e-05, "data_time": 0.001869368553161621, "grad_norm": 42.532611989974974, "loss": 0.6674555003643036, "time": 0.407744026184082, "epoch": 20, "memory": 20334, "step": 114002} +{"lr": 1.6094271405406862e-05, "data_time": 0.0025879621505737306, "grad_norm": 20.755958843231202, "loss": 0.6798424124717712, "time": 0.4156376600265503, "epoch": 20, "memory": 20334, "step": 114102} +{"lr": 1.6094271405406862e-05, "data_time": 0.0021868467330932615, "grad_norm": 12.93769965171814, "loss": 0.6681676089763642, "time": 0.36259660720825193, "epoch": 20, "memory": 20334, "step": 114202} +{"lr": 1.6094271405406862e-05, "data_time": 0.001917719841003418, "grad_norm": 40.20782213211059, "loss": 0.6652441084384918, "time": 0.4466781377792358, "epoch": 20, "memory": 20334, "step": 114302} +{"lr": 1.6094271405406862e-05, "data_time": 0.002218341827392578, "grad_norm": 925.7426535129547, "loss": 0.6715802431106568, "time": 0.39341087341308595, "epoch": 20, "memory": 20334, "step": 114402} +{"lr": 1.6094271405406862e-05, "data_time": 0.0016837835311889649, "grad_norm": 31.550852966308593, "loss": 0.662868195772171, "time": 0.3918366193771362, "epoch": 20, "memory": 20334, "step": 114502} +{"lr": 1.6094271405406862e-05, "data_time": 0.002666282653808594, "grad_norm": 53.29404301643372, "loss": 0.6778385102748871, "time": 0.4472174167633057, "epoch": 20, "memory": 20334, "step": 114602} +{"lr": 1.6094271405406862e-05, "data_time": 0.0022686004638671877, "grad_norm": 202.7078077316284, "loss": 0.6774605691432953, "time": 0.3977050304412842, "epoch": 20, "memory": 20334, "step": 114702} +{"lr": 1.6094271405406862e-05, "data_time": 0.0017225503921508788, "grad_norm": 117.04723069667816, "loss": 0.6648397505283355, "time": 0.44902782440185546, "epoch": 20, "memory": 20334, "step": 114802} +{"lr": 1.6094271405406862e-05, "data_time": 0.002423000335693359, "grad_norm": 166.69438486099244, "loss": 0.674306458234787, "time": 0.4497149705886841, "epoch": 20, "memory": 20334, "step": 114902} +{"lr": 1.6094271405406862e-05, "data_time": 0.003200793266296387, "grad_norm": 60.18272271156311, "loss": 0.6722406625747681, "time": 0.4039085626602173, "epoch": 20, "memory": 20334, "step": 115002} +{"lr": 1.6094271405406862e-05, "data_time": 0.001801276206970215, "grad_norm": 116.333566904068, "loss": 0.6590107202529907, "time": 0.37529284954071046, "epoch": 20, "memory": 20334, "step": 115102} +{"accuracy/top1": 80.52521514892578, "data_time": 0.37442196740044487, "time": 1.0008376439412434, "step": 20} diff --git a/swin_base_8xb128_1e-3lr_5m/20230531_183241/vis_data/vis_image/0000000.png_1.png b/swin_base_8xb128_1e-3lr_5m/20230531_183241/vis_data/vis_image/0000000.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..b1d5db9b91db71ddcbbba2e61273dd6141d70f3d Binary files /dev/null and b/swin_base_8xb128_1e-3lr_5m/20230531_183241/vis_data/vis_image/0000000.png_1.png differ diff --git a/swin_base_8xb128_1e-3lr_5m/20230531_183241/vis_data/vis_image/0000000.png_10.png b/swin_base_8xb128_1e-3lr_5m/20230531_183241/vis_data/vis_image/0000000.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..934769c1f2bb2402694533d16138c8dbc1ce4a97 Binary files /dev/null and b/swin_base_8xb128_1e-3lr_5m/20230531_183241/vis_data/vis_image/0000000.png_10.png differ diff --git a/swin_base_8xb128_1e-3lr_5m/20230531_183241/vis_data/vis_image/0000000.png_11.png b/swin_base_8xb128_1e-3lr_5m/20230531_183241/vis_data/vis_image/0000000.png_11.png new file mode 100644 index 0000000000000000000000000000000000000000..934769c1f2bb2402694533d16138c8dbc1ce4a97 Binary files /dev/null and b/swin_base_8xb128_1e-3lr_5m/20230531_183241/vis_data/vis_image/0000000.png_11.png differ diff --git a/swin_base_8xb128_1e-3lr_5m/20230531_183241/vis_data/vis_image/0000000.png_12.png b/swin_base_8xb128_1e-3lr_5m/20230531_183241/vis_data/vis_image/0000000.png_12.png new file mode 100644 index 0000000000000000000000000000000000000000..934769c1f2bb2402694533d16138c8dbc1ce4a97 Binary files /dev/null and b/swin_base_8xb128_1e-3lr_5m/20230531_183241/vis_data/vis_image/0000000.png_12.png differ diff --git a/swin_base_8xb128_1e-3lr_5m/20230531_183241/vis_data/vis_image/0000000.png_13.png b/swin_base_8xb128_1e-3lr_5m/20230531_183241/vis_data/vis_image/0000000.png_13.png new file mode 100644 index 0000000000000000000000000000000000000000..934769c1f2bb2402694533d16138c8dbc1ce4a97 Binary files /dev/null and b/swin_base_8xb128_1e-3lr_5m/20230531_183241/vis_data/vis_image/0000000.png_13.png differ diff --git a/swin_base_8xb128_1e-3lr_5m/20230531_183241/vis_data/vis_image/0000000.png_14.png b/swin_base_8xb128_1e-3lr_5m/20230531_183241/vis_data/vis_image/0000000.png_14.png new file mode 100644 index 0000000000000000000000000000000000000000..934769c1f2bb2402694533d16138c8dbc1ce4a97 Binary files /dev/null and b/swin_base_8xb128_1e-3lr_5m/20230531_183241/vis_data/vis_image/0000000.png_14.png differ diff --git a/swin_base_8xb128_1e-3lr_5m/20230531_183241/vis_data/vis_image/0000000.png_15.png b/swin_base_8xb128_1e-3lr_5m/20230531_183241/vis_data/vis_image/0000000.png_15.png new file mode 100644 index 0000000000000000000000000000000000000000..1be5dd4998bb6baa8c3ce36c1a4d861cdaf6b2c7 Binary files /dev/null and b/swin_base_8xb128_1e-3lr_5m/20230531_183241/vis_data/vis_image/0000000.png_15.png differ diff --git a/swin_base_8xb128_1e-3lr_5m/20230531_183241/vis_data/vis_image/0000000.png_16.png b/swin_base_8xb128_1e-3lr_5m/20230531_183241/vis_data/vis_image/0000000.png_16.png new file mode 100644 index 0000000000000000000000000000000000000000..dbd00a21bdda4f44b04908f3673b5c490c33ef18 Binary files /dev/null and b/swin_base_8xb128_1e-3lr_5m/20230531_183241/vis_data/vis_image/0000000.png_16.png differ diff --git a/swin_base_8xb128_1e-3lr_5m/20230531_183241/vis_data/vis_image/0000000.png_17.png b/swin_base_8xb128_1e-3lr_5m/20230531_183241/vis_data/vis_image/0000000.png_17.png new file mode 100644 index 0000000000000000000000000000000000000000..dbd00a21bdda4f44b04908f3673b5c490c33ef18 Binary files /dev/null and b/swin_base_8xb128_1e-3lr_5m/20230531_183241/vis_data/vis_image/0000000.png_17.png differ diff --git a/swin_base_8xb128_1e-3lr_5m/20230531_183241/vis_data/vis_image/0000000.png_18.png b/swin_base_8xb128_1e-3lr_5m/20230531_183241/vis_data/vis_image/0000000.png_18.png new file mode 100644 index 0000000000000000000000000000000000000000..78086a4ac43d9c8da460aa6d9b24e940253194a5 Binary files /dev/null and b/swin_base_8xb128_1e-3lr_5m/20230531_183241/vis_data/vis_image/0000000.png_18.png differ diff --git a/swin_base_8xb128_1e-3lr_5m/20230531_183241/vis_data/vis_image/0000000.png_19.png b/swin_base_8xb128_1e-3lr_5m/20230531_183241/vis_data/vis_image/0000000.png_19.png new file mode 100644 index 0000000000000000000000000000000000000000..794f60aac9ffc45f4c9afead90738ca04204b4f6 Binary files /dev/null and b/swin_base_8xb128_1e-3lr_5m/20230531_183241/vis_data/vis_image/0000000.png_19.png differ diff --git a/swin_base_8xb128_1e-3lr_5m/20230531_183241/vis_data/vis_image/0000000.png_2.png b/swin_base_8xb128_1e-3lr_5m/20230531_183241/vis_data/vis_image/0000000.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..b1d5db9b91db71ddcbbba2e61273dd6141d70f3d Binary files /dev/null and b/swin_base_8xb128_1e-3lr_5m/20230531_183241/vis_data/vis_image/0000000.png_2.png differ diff --git a/swin_base_8xb128_1e-3lr_5m/20230531_183241/vis_data/vis_image/0000000.png_20.png b/swin_base_8xb128_1e-3lr_5m/20230531_183241/vis_data/vis_image/0000000.png_20.png new file mode 100644 index 0000000000000000000000000000000000000000..ae323dfb2f99afa256eed926b072e5d438f42b32 Binary files /dev/null and b/swin_base_8xb128_1e-3lr_5m/20230531_183241/vis_data/vis_image/0000000.png_20.png differ diff --git a/swin_base_8xb128_1e-3lr_5m/20230531_183241/vis_data/vis_image/0000000.png_3.png b/swin_base_8xb128_1e-3lr_5m/20230531_183241/vis_data/vis_image/0000000.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..b1d5db9b91db71ddcbbba2e61273dd6141d70f3d Binary files /dev/null and b/swin_base_8xb128_1e-3lr_5m/20230531_183241/vis_data/vis_image/0000000.png_3.png differ diff --git a/swin_base_8xb128_1e-3lr_5m/20230531_183241/vis_data/vis_image/0000000.png_4.png b/swin_base_8xb128_1e-3lr_5m/20230531_183241/vis_data/vis_image/0000000.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..934769c1f2bb2402694533d16138c8dbc1ce4a97 Binary files /dev/null and b/swin_base_8xb128_1e-3lr_5m/20230531_183241/vis_data/vis_image/0000000.png_4.png differ diff --git a/swin_base_8xb128_1e-3lr_5m/20230531_183241/vis_data/vis_image/0000000.png_5.png b/swin_base_8xb128_1e-3lr_5m/20230531_183241/vis_data/vis_image/0000000.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..934769c1f2bb2402694533d16138c8dbc1ce4a97 Binary files /dev/null and b/swin_base_8xb128_1e-3lr_5m/20230531_183241/vis_data/vis_image/0000000.png_5.png differ diff --git a/swin_base_8xb128_1e-3lr_5m/20230531_183241/vis_data/vis_image/0000000.png_6.png b/swin_base_8xb128_1e-3lr_5m/20230531_183241/vis_data/vis_image/0000000.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..934769c1f2bb2402694533d16138c8dbc1ce4a97 Binary files /dev/null and b/swin_base_8xb128_1e-3lr_5m/20230531_183241/vis_data/vis_image/0000000.png_6.png differ diff --git a/swin_base_8xb128_1e-3lr_5m/20230531_183241/vis_data/vis_image/0000000.png_7.png b/swin_base_8xb128_1e-3lr_5m/20230531_183241/vis_data/vis_image/0000000.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..934769c1f2bb2402694533d16138c8dbc1ce4a97 Binary files /dev/null and b/swin_base_8xb128_1e-3lr_5m/20230531_183241/vis_data/vis_image/0000000.png_7.png differ diff --git a/swin_base_8xb128_1e-3lr_5m/20230531_183241/vis_data/vis_image/0000000.png_8.png b/swin_base_8xb128_1e-3lr_5m/20230531_183241/vis_data/vis_image/0000000.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..934769c1f2bb2402694533d16138c8dbc1ce4a97 Binary files /dev/null and b/swin_base_8xb128_1e-3lr_5m/20230531_183241/vis_data/vis_image/0000000.png_8.png differ diff --git a/swin_base_8xb128_1e-3lr_5m/20230531_183241/vis_data/vis_image/0000000.png_9.png b/swin_base_8xb128_1e-3lr_5m/20230531_183241/vis_data/vis_image/0000000.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..ae323dfb2f99afa256eed926b072e5d438f42b32 Binary files /dev/null and b/swin_base_8xb128_1e-3lr_5m/20230531_183241/vis_data/vis_image/0000000.png_9.png differ diff --git a/swin_base_8xb128_1e-3lr_5m/epoch_1.pth b/swin_base_8xb128_1e-3lr_5m/epoch_1.pth new file mode 100644 index 0000000000000000000000000000000000000000..55fe7d9a65777061575985d91e4092021a0edde6 --- /dev/null +++ b/swin_base_8xb128_1e-3lr_5m/epoch_1.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:87d1d6a034e789db150bf4bd5962edd61dbd7abb023b07c6361c9afaa0c3b7ae +size 1042378397 diff --git a/swin_base_8xb128_1e-3lr_5m/epoch_10.pth b/swin_base_8xb128_1e-3lr_5m/epoch_10.pth new file mode 100644 index 0000000000000000000000000000000000000000..19663b5f6105fc52c8397c0780e78797ae81b9b9 --- /dev/null +++ b/swin_base_8xb128_1e-3lr_5m/epoch_10.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ad086485790556bd9b00150b88f7dcc1a7ea5cff80723fbe1c6c7089a2ef90a9 +size 1047063069 diff --git a/swin_base_8xb128_1e-3lr_5m/epoch_11.pth b/swin_base_8xb128_1e-3lr_5m/epoch_11.pth new file mode 100644 index 0000000000000000000000000000000000000000..9fd386105c7e23501848e9cbdb8e83f76344a9f8 --- /dev/null +++ b/swin_base_8xb128_1e-3lr_5m/epoch_11.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:989acc9c5c7d464afd681800ac04066ca4e09cf8b062a1d0d3df7ee4c8e4abf9 +size 1047576221 diff --git a/swin_base_8xb128_1e-3lr_5m/epoch_12.pth b/swin_base_8xb128_1e-3lr_5m/epoch_12.pth new file mode 100644 index 0000000000000000000000000000000000000000..a715625d134c72c4f063fc299ce8557642c321a7 --- /dev/null +++ b/swin_base_8xb128_1e-3lr_5m/epoch_12.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:e8900339f2d363d98259397190f2171d632e5f90f2a620e495487a6103276f89 +size 1048095581 diff --git a/swin_base_8xb128_1e-3lr_5m/epoch_13.pth b/swin_base_8xb128_1e-3lr_5m/epoch_13.pth new file mode 100644 index 0000000000000000000000000000000000000000..8e22da1d45c947c9dc54e3fe7f77b8cb3f3311f4 --- /dev/null +++ b/swin_base_8xb128_1e-3lr_5m/epoch_13.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:f7df57c5d5be862d366e61f5221a8d05e6a014913c31c0fea7c4e671909e6083 +size 1048620829 diff --git a/swin_base_8xb128_1e-3lr_5m/epoch_14.pth b/swin_base_8xb128_1e-3lr_5m/epoch_14.pth new file mode 100644 index 0000000000000000000000000000000000000000..3b167b75ae79f02d12b2c3a749103c0cedadb760 --- /dev/null +++ b/swin_base_8xb128_1e-3lr_5m/epoch_14.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:f42727796448a05b2afebd4d10c97bd5c3b248be465b79b05adfb98420e58b6c +size 1049134941 diff --git a/swin_base_8xb128_1e-3lr_5m/epoch_15.pth b/swin_base_8xb128_1e-3lr_5m/epoch_15.pth new file mode 100644 index 0000000000000000000000000000000000000000..e3918d3346e33fddb04a83e6923f683f3caf047a --- /dev/null +++ b/swin_base_8xb128_1e-3lr_5m/epoch_15.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:af0b337ff0019ba454b115b2bf2cb625c4171bdfd515139ef003595cf363f465 +size 1049660893 diff --git a/swin_base_8xb128_1e-3lr_5m/epoch_16.pth b/swin_base_8xb128_1e-3lr_5m/epoch_16.pth new file mode 100644 index 0000000000000000000000000000000000000000..57e7491478226d19d800e9be1988a272bd076540 --- /dev/null +++ b/swin_base_8xb128_1e-3lr_5m/epoch_16.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:3e0a6ad3afc3aa19489daa4f60e47e7e4e4c8477d11dc243b985956731de84d3 +size 1050168349 diff --git a/swin_base_8xb128_1e-3lr_5m/epoch_17.pth b/swin_base_8xb128_1e-3lr_5m/epoch_17.pth new file mode 100644 index 0000000000000000000000000000000000000000..cf21b167dbafc077747f483ffa24f60267745386 --- /dev/null +++ b/swin_base_8xb128_1e-3lr_5m/epoch_17.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:0e72177a7ff3296f60d7571e8af8ffeed6e0f0ff7dd8ecbc32af3299e9642004 +size 1050673309 diff --git a/swin_base_8xb128_1e-3lr_5m/epoch_18.pth b/swin_base_8xb128_1e-3lr_5m/epoch_18.pth new file mode 100644 index 0000000000000000000000000000000000000000..5f58ea57beaee58c87eca80f724c2362d103e95a --- /dev/null +++ b/swin_base_8xb128_1e-3lr_5m/epoch_18.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:6d7f8396ad8ab337cedeac685c1ca5ab7b8fc6728532faed8bc88dd6f36f1041 +size 1051183965 diff --git a/swin_base_8xb128_1e-3lr_5m/epoch_19.pth b/swin_base_8xb128_1e-3lr_5m/epoch_19.pth new file mode 100644 index 0000000000000000000000000000000000000000..0ecc2c9ef8389ad89c551ce1b2cbeb4da9552def --- /dev/null +++ b/swin_base_8xb128_1e-3lr_5m/epoch_19.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c7628e55f582987b85ad6c50fe2e1bb501d58325557a590e1ba98bdc1b1ea21d +size 1051705373 diff --git a/swin_base_8xb128_1e-3lr_5m/epoch_2.pth b/swin_base_8xb128_1e-3lr_5m/epoch_2.pth new file mode 100644 index 0000000000000000000000000000000000000000..a9c1912f92f6673e48c1b1a1d865bae14cdc7f1f --- /dev/null +++ b/swin_base_8xb128_1e-3lr_5m/epoch_2.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:4a5bbdbe933595e78c4fbffe4d86576a51a765e1c685f53c157b0ae434fe0581 +size 1042889181 diff --git a/swin_base_8xb128_1e-3lr_5m/epoch_20.pth b/swin_base_8xb128_1e-3lr_5m/epoch_20.pth new file mode 100644 index 0000000000000000000000000000000000000000..33a4830de1241de5e587dc3a33b5a62d37dc7028 --- /dev/null +++ b/swin_base_8xb128_1e-3lr_5m/epoch_20.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:312899ff9a08c56f25c756bf1959018d451f8dcaa013727b8af324f1e0650b49 +size 1052233053 diff --git a/swin_base_8xb128_1e-3lr_5m/epoch_3.pth b/swin_base_8xb128_1e-3lr_5m/epoch_3.pth new file mode 100644 index 0000000000000000000000000000000000000000..b70052e40d9bb34b41871664c965cccf1cc8717c --- /dev/null +++ b/swin_base_8xb128_1e-3lr_5m/epoch_3.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:a7513c0e115fac84e4150bdd075457f1325a63b43b794401d911845e15749220 +size 1043409117 diff --git a/swin_base_8xb128_1e-3lr_5m/epoch_4.pth b/swin_base_8xb128_1e-3lr_5m/epoch_4.pth new file mode 100644 index 0000000000000000000000000000000000000000..21cd5c25ccd2e727166148438d1ed8dc5b60f503 --- /dev/null +++ b/swin_base_8xb128_1e-3lr_5m/epoch_4.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:01611b1a482218c0f266c40cacdf9dac51db926ed064f26f773ebf9a866f3cab +size 1043916957 diff --git a/swin_base_8xb128_1e-3lr_5m/epoch_5.pth b/swin_base_8xb128_1e-3lr_5m/epoch_5.pth new file mode 100644 index 0000000000000000000000000000000000000000..0c2cd3d4a510e250e835f371f10c9b916c416738 --- /dev/null +++ b/swin_base_8xb128_1e-3lr_5m/epoch_5.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:2ad11e2c38d224b37eaea4e66ecbca136cfe778dc32dddadecc825071c9af65c +size 1044442653 diff --git a/swin_base_8xb128_1e-3lr_5m/epoch_6.pth b/swin_base_8xb128_1e-3lr_5m/epoch_6.pth new file mode 100644 index 0000000000000000000000000000000000000000..0cd2411922ad273a523bee7ea69d180929b694c8 --- /dev/null +++ b/swin_base_8xb128_1e-3lr_5m/epoch_6.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:98e408d9616dfae95f6995b237d87fdaf696749b6572eab8ac5e450096fd1c51 +size 1044963165 diff --git a/swin_base_8xb128_1e-3lr_5m/epoch_7.pth b/swin_base_8xb128_1e-3lr_5m/epoch_7.pth new file mode 100644 index 0000000000000000000000000000000000000000..fc309c5ba916d593f3a4a88feda58d58f21971a9 --- /dev/null +++ b/swin_base_8xb128_1e-3lr_5m/epoch_7.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:8d5cbbc28738b47d60183cc0b1e0770f1c1988a85283664a2c2d448e2e0a7ba8 +size 1045489181 diff --git a/swin_base_8xb128_1e-3lr_5m/epoch_8.pth b/swin_base_8xb128_1e-3lr_5m/epoch_8.pth new file mode 100644 index 0000000000000000000000000000000000000000..ebfa61c3c9c1000741dbba281cc75c81fa222d98 --- /dev/null +++ b/swin_base_8xb128_1e-3lr_5m/epoch_8.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:bad246539ab79131f5891d253ebe9ff4970abd2b70b6edd769f643aa0f872cd4 +size 1046008349 diff --git a/swin_base_8xb128_1e-3lr_5m/epoch_9.pth b/swin_base_8xb128_1e-3lr_5m/epoch_9.pth new file mode 100644 index 0000000000000000000000000000000000000000..613485cb12d6f6bcdea389bb05b80bda720d2f49 --- /dev/null +++ b/swin_base_8xb128_1e-3lr_5m/epoch_9.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:84ec09f20fe41e1e2f45e9588bfea5e9cdd888783f87009d9f3fcb19cc2338a3 +size 1046532701 diff --git a/swin_base_8xb128_1e-3lr_5m/last_checkpoint b/swin_base_8xb128_1e-3lr_5m/last_checkpoint new file mode 100644 index 0000000000000000000000000000000000000000..fa9332e9f10ac5bf1caebc83bba47d325ce5b6a3 --- /dev/null +++ b/swin_base_8xb128_1e-3lr_5m/last_checkpoint @@ -0,0 +1 @@ +/mnt/petrelfs/luzeyu/workspace/fakebench/mmpretrain/workdir/swin_base_8xb128_1e-3lr_5m/epoch_20.pth \ No newline at end of file diff --git a/swin_base_8xb128_1e-3lr_5m/swin_base_8xb128_fake5m.py b/swin_base_8xb128_1e-3lr_5m/swin_base_8xb128_fake5m.py new file mode 100644 index 0000000000000000000000000000000000000000..bd4397cc02b636fc0a458413dc88abb66e856cd9 --- /dev/null +++ b/swin_base_8xb128_1e-3lr_5m/swin_base_8xb128_fake5m.py @@ -0,0 +1,174 @@ +optim_wrapper = dict( + optimizer=dict( + type='AdamW', + lr=0.001, + weight_decay=0.05, + eps=1e-08, + betas=(0.9, 0.999), + _scope_='mmpretrain'), + paramwise_cfg=dict( + norm_decay_mult=0.0, + bias_decay_mult=0.0, + flat_decay_mult=0.0, + custom_keys=dict({ + '.absolute_pos_embed': dict(decay_mult=0.0), + '.relative_position_bias_table': dict(decay_mult=0.0) + })), + type='AmpOptimWrapper', + dtype='bfloat16', + clip_grad=dict(max_norm=5.0)) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=True, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=20, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=1024) +model = dict( + type='ImageClassifier', + backbone=dict( + type='SwinTransformer', arch='base', img_size=224, drop_path_rate=0.5), + neck=dict(type='GlobalAveragePooling'), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=1024, + init_cfg=None, + loss=dict( + type='LabelSmoothLoss', label_smooth_val=0.1, mode='original'), + cal_acc=False), + init_cfg=[ + dict(type='TruncNormal', layer='Linear', std=0.02, bias=0.0), + dict(type='Constant', layer='LayerNorm', val=1.0, bias=0.0) + ], + train_cfg=None) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=128, + num_workers=10, + dataset=dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stablediffusionV1-5R2-dpmsolver-25-5m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/swin_base_8xb128_1e-3lr_5m' diff --git a/swin_small_4xb256_1e-3lr_5m/20230531_183239/20230531_183239.log b/swin_small_4xb256_1e-3lr_5m/20230531_183239/20230531_183239.log new file mode 100644 index 0000000000000000000000000000000000000000..78e15d69e92e63bcb063a0a691c861760eddf58e --- /dev/null +++ b/swin_small_4xb256_1e-3lr_5m/20230531_183239/20230531_183239.log @@ -0,0 +1,2564 @@ +2023/05/31 18:32:43 - mmengine - INFO - +------------------------------------------------------------ +System environment: + sys.platform: linux + Python: 3.10.9 (main, Mar 8 2023, 10:47:38) [GCC 11.2.0] + CUDA available: True + numpy_random_seed: 2001471778 + GPU 0,1,2,3: NVIDIA A100-SXM4-80GB + CUDA_HOME: /mnt/petrelfs/share/cuda-11.6 + NVCC: Cuda compilation tools, release 11.6, V11.6.124 + GCC: gcc (GCC) 4.8.5 20150623 (Red Hat 4.8.5-44) + PyTorch: 1.13.1 + PyTorch compiling details: PyTorch built with: + - GCC 9.3 + - C++ Version: 201402 + - Intel(R) oneAPI Math Kernel Library Version 2021.4-Product Build 20210904 for Intel(R) 64 architecture applications + - Intel(R) MKL-DNN v2.6.0 (Git Hash 52b5f107dd9cf10910aaa19cb47f3abf9b349815) + - OpenMP 201511 (a.k.a. OpenMP 4.5) + - LAPACK is enabled (usually provided by MKL) + - NNPACK is enabled + - CPU capability usage: AVX2 + - CUDA Runtime 11.6 + - NVCC architecture flags: -gencode;arch=compute_37,code=sm_37;-gencode;arch=compute_50,code=sm_50;-gencode;arch=compute_60,code=sm_60;-gencode;arch=compute_61,code=sm_61;-gencode;arch=compute_70,code=sm_70;-gencode;arch=compute_75,code=sm_75;-gencode;arch=compute_80,code=sm_80;-gencode;arch=compute_86,code=sm_86;-gencode;arch=compute_37,code=compute_37 + - CuDNN 8.3.2 (built against CUDA 11.5) + - Magma 2.6.1 + - Build settings: BLAS_INFO=mkl, BUILD_TYPE=Release, CUDA_VERSION=11.6, CUDNN_VERSION=8.3.2, CXX_COMPILER=/opt/rh/devtoolset-9/root/usr/bin/c++, CXX_FLAGS= -fabi-version=11 -Wno-deprecated -fvisibility-inlines-hidden -DUSE_PTHREADPOOL -fopenmp -DNDEBUG -DUSE_KINETO -DUSE_FBGEMM -DUSE_QNNPACK -DUSE_PYTORCH_QNNPACK -DUSE_XNNPACK -DSYMBOLICATE_MOBILE_DEBUG_HANDLE -DEDGE_PROFILER_USE_KINETO -O2 -fPIC -Wno-narrowing -Wall -Wextra -Werror=return-type -Werror=non-virtual-dtor -Wno-missing-field-initializers -Wno-type-limits -Wno-array-bounds -Wno-unknown-pragmas -Wunused-local-typedefs -Wno-unused-parameter -Wno-unused-function -Wno-unused-result -Wno-strict-overflow -Wno-strict-aliasing -Wno-error=deprecated-declarations -Wno-stringop-overflow -Wno-psabi -Wno-error=pedantic -Wno-error=redundant-decls -Wno-error=old-style-cast -fdiagnostics-color=always -faligned-new -Wno-unused-but-set-variable -Wno-maybe-uninitialized -fno-math-errno -fno-trapping-math -Werror=format -Werror=cast-function-type -Wno-stringop-overflow, LAPACK_INFO=mkl, PERF_WITH_AVX=1, PERF_WITH_AVX2=1, PERF_WITH_AVX512=1, TORCH_VERSION=1.13.1, USE_CUDA=ON, USE_CUDNN=ON, USE_EXCEPTION_PTR=1, USE_GFLAGS=OFF, USE_GLOG=OFF, USE_MKL=ON, USE_MKLDNN=ON, USE_MPI=OFF, USE_NCCL=ON, USE_NNPACK=ON, USE_OPENMP=ON, USE_ROCM=OFF, + + TorchVision: 0.14.1 + OpenCV: 4.7.0 + MMEngine: 0.7.3 + +Runtime environment: + cudnn_benchmark: True + mp_cfg: {'mp_start_method': 'fork', 'opencv_num_threads': 0} + dist_cfg: {'backend': 'nccl'} + seed: None + deterministic: False + Distributed launcher: slurm + Distributed training: True + GPU number: 4 +------------------------------------------------------------ + +2023/05/31 18:32:44 - mmengine - INFO - Config: +optim_wrapper = dict( + optimizer=dict( + type='AdamW', + lr=0.001, + weight_decay=0.05, + eps=1e-08, + betas=(0.9, 0.999), + _scope_='mmpretrain'), + paramwise_cfg=dict( + norm_decay_mult=0.0, + bias_decay_mult=0.0, + flat_decay_mult=0.0, + custom_keys=dict({ + '.absolute_pos_embed': dict(decay_mult=0.0), + '.relative_position_bias_table': dict(decay_mult=0.0) + })), + type='AmpOptimWrapper', + dtype='bfloat16', + clip_grad=dict(max_norm=5.0)) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=True, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=20, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=1024) +model = dict( + type='ImageClassifier', + backbone=dict( + type='SwinTransformer', arch='small', img_size=224, + drop_path_rate=0.3), + neck=dict(type='GlobalAveragePooling'), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=768, + init_cfg=None, + loss=dict( + type='LabelSmoothLoss', label_smooth_val=0.1, mode='original'), + cal_acc=False), + init_cfg=[ + dict(type='TruncNormal', layer='Linear', std=0.02, bias=0.0), + dict(type='Constant', layer='LayerNorm', val=1.0, bias=0.0) + ], + train_cfg=None) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stablediffusionV1-5R2-dpmsolver-25-5m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/swin_small_4xb256_1e-3lr_5m' + +2023/05/31 18:32:48 - mmengine - INFO - Hooks will be executed in the following order: +before_run: +(VERY_HIGH ) RuntimeInfoHook +(BELOW_NORMAL) LoggerHook + -------------------- +before_train: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_train_epoch: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(NORMAL ) DistSamplerSeedHook + -------------------- +before_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook + -------------------- +after_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +after_train_epoch: +(NORMAL ) IterTimerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_val_epoch: +(NORMAL ) IterTimerHook + -------------------- +before_val_iter: +(NORMAL ) IterTimerHook + -------------------- +after_val_iter: +(NORMAL ) IterTimerHook +(NORMAL ) VisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_val_epoch: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +after_train: +(VERY_LOW ) CheckpointHook + -------------------- +before_test_epoch: +(NORMAL ) IterTimerHook + -------------------- +before_test_iter: +(NORMAL ) IterTimerHook + -------------------- +after_test_iter: +(NORMAL ) IterTimerHook +(NORMAL ) VisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_test_epoch: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_run: +(BELOW_NORMAL) LoggerHook + -------------------- +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.patch_embed.projection.bias:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.patch_embed.norm.weight:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.patch_embed.norm.bias:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.0.blocks.0.norm1.weight:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.0.blocks.0.norm1.bias:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.0.blocks.0.attn.w_msa.relative_position_bias_table:lr=0.001 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.0.blocks.0.attn.w_msa.relative_position_bias_table:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.0.blocks.0.attn.w_msa.relative_position_bias_table:decay_mult=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.0.blocks.0.attn.w_msa.qkv.bias:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.0.blocks.0.attn.w_msa.proj.bias:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.0.blocks.0.norm2.weight:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.0.blocks.0.norm2.bias:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.0.blocks.0.ffn.layers.0.0.bias:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.0.blocks.0.ffn.layers.1.bias:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.0.blocks.1.norm1.weight:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.0.blocks.1.norm1.bias:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.0.blocks.1.attn.w_msa.relative_position_bias_table:lr=0.001 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.0.blocks.1.attn.w_msa.relative_position_bias_table:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.0.blocks.1.attn.w_msa.relative_position_bias_table:decay_mult=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.0.blocks.1.attn.w_msa.qkv.bias:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.0.blocks.1.attn.w_msa.proj.bias:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.0.blocks.1.norm2.weight:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.0.blocks.1.norm2.bias:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.0.blocks.1.ffn.layers.0.0.bias:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.0.blocks.1.ffn.layers.1.bias:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.0.downsample.norm.weight:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.0.downsample.norm.bias:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.1.blocks.0.norm1.weight:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.1.blocks.0.norm1.bias:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.1.blocks.0.attn.w_msa.relative_position_bias_table:lr=0.001 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.1.blocks.0.attn.w_msa.relative_position_bias_table:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.1.blocks.0.attn.w_msa.relative_position_bias_table:decay_mult=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.1.blocks.0.attn.w_msa.qkv.bias:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.1.blocks.0.attn.w_msa.proj.bias:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.1.blocks.0.norm2.weight:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.1.blocks.0.norm2.bias:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.1.blocks.0.ffn.layers.0.0.bias:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.1.blocks.0.ffn.layers.1.bias:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.1.blocks.1.norm1.weight:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.1.blocks.1.norm1.bias:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.1.blocks.1.attn.w_msa.relative_position_bias_table:lr=0.001 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.1.blocks.1.attn.w_msa.relative_position_bias_table:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.1.blocks.1.attn.w_msa.relative_position_bias_table:decay_mult=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.1.blocks.1.attn.w_msa.qkv.bias:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.1.blocks.1.attn.w_msa.proj.bias:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.1.blocks.1.norm2.weight:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.1.blocks.1.norm2.bias:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.1.blocks.1.ffn.layers.0.0.bias:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.1.blocks.1.ffn.layers.1.bias:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.1.downsample.norm.weight:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.1.downsample.norm.bias:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.0.norm1.weight:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.0.norm1.bias:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.0.attn.w_msa.relative_position_bias_table:lr=0.001 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.0.attn.w_msa.relative_position_bias_table:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.0.attn.w_msa.relative_position_bias_table:decay_mult=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.0.attn.w_msa.qkv.bias:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.0.attn.w_msa.proj.bias:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.0.norm2.weight:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.0.norm2.bias:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.0.ffn.layers.0.0.bias:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.0.ffn.layers.1.bias:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.1.norm1.weight:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.1.norm1.bias:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.1.attn.w_msa.relative_position_bias_table:lr=0.001 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.1.attn.w_msa.relative_position_bias_table:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.1.attn.w_msa.relative_position_bias_table:decay_mult=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.1.attn.w_msa.qkv.bias:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.1.attn.w_msa.proj.bias:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.1.norm2.weight:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.1.norm2.bias:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.1.ffn.layers.0.0.bias:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.1.ffn.layers.1.bias:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.2.norm1.weight:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.2.norm1.bias:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.2.attn.w_msa.relative_position_bias_table:lr=0.001 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.2.attn.w_msa.relative_position_bias_table:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.2.attn.w_msa.relative_position_bias_table:decay_mult=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.2.attn.w_msa.qkv.bias:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.2.attn.w_msa.proj.bias:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.2.norm2.weight:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.2.norm2.bias:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.2.ffn.layers.0.0.bias:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.2.ffn.layers.1.bias:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.3.norm1.weight:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.3.norm1.bias:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.3.attn.w_msa.relative_position_bias_table:lr=0.001 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.3.attn.w_msa.relative_position_bias_table:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.3.attn.w_msa.relative_position_bias_table:decay_mult=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.3.attn.w_msa.qkv.bias:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.3.attn.w_msa.proj.bias:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.3.norm2.weight:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.3.norm2.bias:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.3.ffn.layers.0.0.bias:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.3.ffn.layers.1.bias:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.4.norm1.weight:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.4.norm1.bias:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.4.attn.w_msa.relative_position_bias_table:lr=0.001 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.4.attn.w_msa.relative_position_bias_table:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.4.attn.w_msa.relative_position_bias_table:decay_mult=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.4.attn.w_msa.qkv.bias:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.4.attn.w_msa.proj.bias:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.4.norm2.weight:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.4.norm2.bias:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.4.ffn.layers.0.0.bias:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.4.ffn.layers.1.bias:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.5.norm1.weight:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.5.norm1.bias:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.5.attn.w_msa.relative_position_bias_table:lr=0.001 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.5.attn.w_msa.relative_position_bias_table:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.5.attn.w_msa.relative_position_bias_table:decay_mult=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.5.attn.w_msa.qkv.bias:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.5.attn.w_msa.proj.bias:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.5.norm2.weight:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.5.norm2.bias:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.5.ffn.layers.0.0.bias:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.5.ffn.layers.1.bias:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.6.norm1.weight:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.6.norm1.bias:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.6.attn.w_msa.relative_position_bias_table:lr=0.001 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.6.attn.w_msa.relative_position_bias_table:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.6.attn.w_msa.relative_position_bias_table:decay_mult=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.6.attn.w_msa.qkv.bias:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.6.attn.w_msa.proj.bias:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.6.norm2.weight:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.6.norm2.bias:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.6.ffn.layers.0.0.bias:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.6.ffn.layers.1.bias:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.7.norm1.weight:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.7.norm1.bias:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.7.attn.w_msa.relative_position_bias_table:lr=0.001 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.7.attn.w_msa.relative_position_bias_table:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.7.attn.w_msa.relative_position_bias_table:decay_mult=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.7.attn.w_msa.qkv.bias:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.7.attn.w_msa.proj.bias:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.7.norm2.weight:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.7.norm2.bias:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.7.ffn.layers.0.0.bias:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.7.ffn.layers.1.bias:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.8.norm1.weight:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.8.norm1.bias:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.8.attn.w_msa.relative_position_bias_table:lr=0.001 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.8.attn.w_msa.relative_position_bias_table:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.8.attn.w_msa.relative_position_bias_table:decay_mult=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.8.attn.w_msa.qkv.bias:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.8.attn.w_msa.proj.bias:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.8.norm2.weight:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.8.norm2.bias:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.8.ffn.layers.0.0.bias:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.8.ffn.layers.1.bias:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.9.norm1.weight:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.9.norm1.bias:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.9.attn.w_msa.relative_position_bias_table:lr=0.001 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.9.attn.w_msa.relative_position_bias_table:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.9.attn.w_msa.relative_position_bias_table:decay_mult=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.9.attn.w_msa.qkv.bias:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.9.attn.w_msa.proj.bias:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.9.norm2.weight:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.9.norm2.bias:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.9.ffn.layers.0.0.bias:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.9.ffn.layers.1.bias:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.10.norm1.weight:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.10.norm1.bias:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.10.attn.w_msa.relative_position_bias_table:lr=0.001 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.10.attn.w_msa.relative_position_bias_table:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.10.attn.w_msa.relative_position_bias_table:decay_mult=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.10.attn.w_msa.qkv.bias:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.10.attn.w_msa.proj.bias:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.10.norm2.weight:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.10.norm2.bias:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.10.ffn.layers.0.0.bias:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.10.ffn.layers.1.bias:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.11.norm1.weight:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.11.norm1.bias:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.11.attn.w_msa.relative_position_bias_table:lr=0.001 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.11.attn.w_msa.relative_position_bias_table:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.11.attn.w_msa.relative_position_bias_table:decay_mult=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.11.attn.w_msa.qkv.bias:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.11.attn.w_msa.proj.bias:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.11.norm2.weight:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.11.norm2.bias:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.11.ffn.layers.0.0.bias:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.11.ffn.layers.1.bias:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.12.norm1.weight:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.12.norm1.bias:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.12.attn.w_msa.relative_position_bias_table:lr=0.001 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.12.attn.w_msa.relative_position_bias_table:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.12.attn.w_msa.relative_position_bias_table:decay_mult=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.12.attn.w_msa.qkv.bias:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.12.attn.w_msa.proj.bias:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.12.norm2.weight:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.12.norm2.bias:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.12.ffn.layers.0.0.bias:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.12.ffn.layers.1.bias:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.13.norm1.weight:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.13.norm1.bias:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.13.attn.w_msa.relative_position_bias_table:lr=0.001 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.13.attn.w_msa.relative_position_bias_table:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.13.attn.w_msa.relative_position_bias_table:decay_mult=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.13.attn.w_msa.qkv.bias:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.13.attn.w_msa.proj.bias:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.13.norm2.weight:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.13.norm2.bias:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.13.ffn.layers.0.0.bias:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.13.ffn.layers.1.bias:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.14.norm1.weight:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.14.norm1.bias:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.14.attn.w_msa.relative_position_bias_table:lr=0.001 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.14.attn.w_msa.relative_position_bias_table:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.14.attn.w_msa.relative_position_bias_table:decay_mult=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.14.attn.w_msa.qkv.bias:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.14.attn.w_msa.proj.bias:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.14.norm2.weight:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.14.norm2.bias:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.14.ffn.layers.0.0.bias:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.14.ffn.layers.1.bias:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.15.norm1.weight:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.15.norm1.bias:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.15.attn.w_msa.relative_position_bias_table:lr=0.001 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.15.attn.w_msa.relative_position_bias_table:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.15.attn.w_msa.relative_position_bias_table:decay_mult=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.15.attn.w_msa.qkv.bias:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.15.attn.w_msa.proj.bias:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.15.norm2.weight:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.15.norm2.bias:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.15.ffn.layers.0.0.bias:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.15.ffn.layers.1.bias:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.16.norm1.weight:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.16.norm1.bias:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.16.attn.w_msa.relative_position_bias_table:lr=0.001 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.16.attn.w_msa.relative_position_bias_table:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.16.attn.w_msa.relative_position_bias_table:decay_mult=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.16.attn.w_msa.qkv.bias:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.16.attn.w_msa.proj.bias:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.16.norm2.weight:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.16.norm2.bias:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.16.ffn.layers.0.0.bias:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.16.ffn.layers.1.bias:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.17.norm1.weight:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.17.norm1.bias:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.17.attn.w_msa.relative_position_bias_table:lr=0.001 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.17.attn.w_msa.relative_position_bias_table:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.17.attn.w_msa.relative_position_bias_table:decay_mult=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.17.attn.w_msa.qkv.bias:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.17.attn.w_msa.proj.bias:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.17.norm2.weight:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.17.norm2.bias:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.17.ffn.layers.0.0.bias:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.17.ffn.layers.1.bias:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.downsample.norm.weight:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.2.downsample.norm.bias:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.3.blocks.0.norm1.weight:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.3.blocks.0.norm1.bias:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.3.blocks.0.attn.w_msa.relative_position_bias_table:lr=0.001 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.3.blocks.0.attn.w_msa.relative_position_bias_table:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.3.blocks.0.attn.w_msa.relative_position_bias_table:decay_mult=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.3.blocks.0.attn.w_msa.qkv.bias:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.3.blocks.0.attn.w_msa.proj.bias:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.3.blocks.0.norm2.weight:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.3.blocks.0.norm2.bias:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.3.blocks.0.ffn.layers.0.0.bias:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.3.blocks.0.ffn.layers.1.bias:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.3.blocks.1.norm1.weight:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.3.blocks.1.norm1.bias:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.3.blocks.1.attn.w_msa.relative_position_bias_table:lr=0.001 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.3.blocks.1.attn.w_msa.relative_position_bias_table:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.3.blocks.1.attn.w_msa.relative_position_bias_table:decay_mult=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.3.blocks.1.attn.w_msa.qkv.bias:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.3.blocks.1.attn.w_msa.proj.bias:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.3.blocks.1.norm2.weight:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.3.blocks.1.norm2.bias:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.3.blocks.1.ffn.layers.0.0.bias:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.stages.3.blocks.1.ffn.layers.1.bias:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.norm3.weight:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- backbone.norm3.bias:weight_decay=0.0 +2023/05/31 18:33:21 - mmengine - INFO - paramwise_options -- head.fc.bias:weight_decay=0.0 +Name of parameter - Initialization information + +backbone.patch_embed.projection.weight - torch.Size([96, 3, 4, 4]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.patch_embed.projection.bias - torch.Size([96]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.patch_embed.norm.weight - torch.Size([96]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.patch_embed.norm.bias - torch.Size([96]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.0.blocks.0.norm1.weight - torch.Size([96]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.0.blocks.0.norm1.bias - torch.Size([96]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.0.blocks.0.attn.w_msa.relative_position_bias_table - torch.Size([169, 3]): +Initialized by user-defined `init_weights` in WindowMSA + +backbone.stages.0.blocks.0.attn.w_msa.qkv.weight - torch.Size([288, 96]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.blocks.0.attn.w_msa.qkv.bias - torch.Size([288]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.blocks.0.attn.w_msa.proj.weight - torch.Size([96, 96]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.blocks.0.attn.w_msa.proj.bias - torch.Size([96]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.blocks.0.norm2.weight - torch.Size([96]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.0.blocks.0.norm2.bias - torch.Size([96]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.0.blocks.0.ffn.layers.0.0.weight - torch.Size([384, 96]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.blocks.0.ffn.layers.0.0.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.blocks.0.ffn.layers.1.weight - torch.Size([96, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.blocks.0.ffn.layers.1.bias - torch.Size([96]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.blocks.1.norm1.weight - torch.Size([96]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.0.blocks.1.norm1.bias - torch.Size([96]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.0.blocks.1.attn.w_msa.relative_position_bias_table - torch.Size([169, 3]): +Initialized by user-defined `init_weights` in WindowMSA + +backbone.stages.0.blocks.1.attn.w_msa.qkv.weight - torch.Size([288, 96]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.blocks.1.attn.w_msa.qkv.bias - torch.Size([288]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.blocks.1.attn.w_msa.proj.weight - torch.Size([96, 96]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.blocks.1.attn.w_msa.proj.bias - torch.Size([96]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.blocks.1.norm2.weight - torch.Size([96]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.0.blocks.1.norm2.bias - torch.Size([96]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.0.blocks.1.ffn.layers.0.0.weight - torch.Size([384, 96]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.blocks.1.ffn.layers.0.0.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.blocks.1.ffn.layers.1.weight - torch.Size([96, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.blocks.1.ffn.layers.1.bias - torch.Size([96]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.downsample.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.0.downsample.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.0.downsample.reduction.weight - torch.Size([192, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.blocks.0.norm1.weight - torch.Size([192]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.1.blocks.0.norm1.bias - torch.Size([192]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.1.blocks.0.attn.w_msa.relative_position_bias_table - torch.Size([169, 6]): +Initialized by user-defined `init_weights` in WindowMSA + +backbone.stages.1.blocks.0.attn.w_msa.qkv.weight - torch.Size([576, 192]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.blocks.0.attn.w_msa.qkv.bias - torch.Size([576]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.blocks.0.attn.w_msa.proj.weight - torch.Size([192, 192]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.blocks.0.attn.w_msa.proj.bias - torch.Size([192]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.blocks.0.norm2.weight - torch.Size([192]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.1.blocks.0.norm2.bias - torch.Size([192]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.1.blocks.0.ffn.layers.0.0.weight - torch.Size([768, 192]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.blocks.0.ffn.layers.0.0.bias - torch.Size([768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.blocks.0.ffn.layers.1.weight - torch.Size([192, 768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.blocks.0.ffn.layers.1.bias - torch.Size([192]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.blocks.1.norm1.weight - torch.Size([192]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.1.blocks.1.norm1.bias - torch.Size([192]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.1.blocks.1.attn.w_msa.relative_position_bias_table - torch.Size([169, 6]): +Initialized by user-defined `init_weights` in WindowMSA + +backbone.stages.1.blocks.1.attn.w_msa.qkv.weight - torch.Size([576, 192]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.blocks.1.attn.w_msa.qkv.bias - torch.Size([576]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.blocks.1.attn.w_msa.proj.weight - torch.Size([192, 192]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.blocks.1.attn.w_msa.proj.bias - torch.Size([192]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.blocks.1.norm2.weight - torch.Size([192]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.1.blocks.1.norm2.bias - torch.Size([192]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.1.blocks.1.ffn.layers.0.0.weight - torch.Size([768, 192]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.blocks.1.ffn.layers.0.0.bias - torch.Size([768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.blocks.1.ffn.layers.1.weight - torch.Size([192, 768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.blocks.1.ffn.layers.1.bias - torch.Size([192]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.downsample.norm.weight - torch.Size([768]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.1.downsample.norm.bias - torch.Size([768]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.1.downsample.reduction.weight - torch.Size([384, 768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.0.norm1.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.0.norm1.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.0.attn.w_msa.relative_position_bias_table - torch.Size([169, 12]): +Initialized by user-defined `init_weights` in WindowMSA + +backbone.stages.2.blocks.0.attn.w_msa.qkv.weight - torch.Size([1152, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.0.attn.w_msa.qkv.bias - torch.Size([1152]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.0.attn.w_msa.proj.weight - torch.Size([384, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.0.attn.w_msa.proj.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.0.norm2.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.0.norm2.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.0.ffn.layers.0.0.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.0.ffn.layers.0.0.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.0.ffn.layers.1.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.0.ffn.layers.1.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.1.norm1.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.1.norm1.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.1.attn.w_msa.relative_position_bias_table - torch.Size([169, 12]): +Initialized by user-defined `init_weights` in WindowMSA + +backbone.stages.2.blocks.1.attn.w_msa.qkv.weight - torch.Size([1152, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.1.attn.w_msa.qkv.bias - torch.Size([1152]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.1.attn.w_msa.proj.weight - torch.Size([384, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.1.attn.w_msa.proj.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.1.norm2.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.1.norm2.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.1.ffn.layers.0.0.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.1.ffn.layers.0.0.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.1.ffn.layers.1.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.1.ffn.layers.1.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.2.norm1.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.2.norm1.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.2.attn.w_msa.relative_position_bias_table - torch.Size([169, 12]): +Initialized by user-defined `init_weights` in WindowMSA + +backbone.stages.2.blocks.2.attn.w_msa.qkv.weight - torch.Size([1152, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.2.attn.w_msa.qkv.bias - torch.Size([1152]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.2.attn.w_msa.proj.weight - torch.Size([384, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.2.attn.w_msa.proj.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.2.norm2.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.2.norm2.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.2.ffn.layers.0.0.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.2.ffn.layers.0.0.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.2.ffn.layers.1.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.2.ffn.layers.1.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.3.norm1.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.3.norm1.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.3.attn.w_msa.relative_position_bias_table - torch.Size([169, 12]): +Initialized by user-defined `init_weights` in WindowMSA + +backbone.stages.2.blocks.3.attn.w_msa.qkv.weight - torch.Size([1152, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.3.attn.w_msa.qkv.bias - torch.Size([1152]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.3.attn.w_msa.proj.weight - torch.Size([384, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.3.attn.w_msa.proj.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.3.norm2.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.3.norm2.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.3.ffn.layers.0.0.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.3.ffn.layers.0.0.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.3.ffn.layers.1.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.3.ffn.layers.1.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.4.norm1.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.4.norm1.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.4.attn.w_msa.relative_position_bias_table - torch.Size([169, 12]): +Initialized by user-defined `init_weights` in WindowMSA + +backbone.stages.2.blocks.4.attn.w_msa.qkv.weight - torch.Size([1152, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.4.attn.w_msa.qkv.bias - torch.Size([1152]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.4.attn.w_msa.proj.weight - torch.Size([384, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.4.attn.w_msa.proj.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.4.norm2.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.4.norm2.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.4.ffn.layers.0.0.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.4.ffn.layers.0.0.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.4.ffn.layers.1.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.4.ffn.layers.1.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.5.norm1.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.5.norm1.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.5.attn.w_msa.relative_position_bias_table - torch.Size([169, 12]): +Initialized by user-defined `init_weights` in WindowMSA + +backbone.stages.2.blocks.5.attn.w_msa.qkv.weight - torch.Size([1152, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.5.attn.w_msa.qkv.bias - torch.Size([1152]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.5.attn.w_msa.proj.weight - torch.Size([384, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.5.attn.w_msa.proj.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.5.norm2.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.5.norm2.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.5.ffn.layers.0.0.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.5.ffn.layers.0.0.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.5.ffn.layers.1.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.5.ffn.layers.1.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.6.norm1.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.6.norm1.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.6.attn.w_msa.relative_position_bias_table - torch.Size([169, 12]): +Initialized by user-defined `init_weights` in WindowMSA + +backbone.stages.2.blocks.6.attn.w_msa.qkv.weight - torch.Size([1152, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.6.attn.w_msa.qkv.bias - torch.Size([1152]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.6.attn.w_msa.proj.weight - torch.Size([384, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.6.attn.w_msa.proj.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.6.norm2.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.6.norm2.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.6.ffn.layers.0.0.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.6.ffn.layers.0.0.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.6.ffn.layers.1.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.6.ffn.layers.1.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.7.norm1.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.7.norm1.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.7.attn.w_msa.relative_position_bias_table - torch.Size([169, 12]): +Initialized by user-defined `init_weights` in WindowMSA + +backbone.stages.2.blocks.7.attn.w_msa.qkv.weight - torch.Size([1152, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.7.attn.w_msa.qkv.bias - torch.Size([1152]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.7.attn.w_msa.proj.weight - torch.Size([384, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.7.attn.w_msa.proj.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.7.norm2.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.7.norm2.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.7.ffn.layers.0.0.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.7.ffn.layers.0.0.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.7.ffn.layers.1.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.7.ffn.layers.1.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.8.norm1.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.8.norm1.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.8.attn.w_msa.relative_position_bias_table - torch.Size([169, 12]): +Initialized by user-defined `init_weights` in WindowMSA + +backbone.stages.2.blocks.8.attn.w_msa.qkv.weight - torch.Size([1152, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.8.attn.w_msa.qkv.bias - torch.Size([1152]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.8.attn.w_msa.proj.weight - torch.Size([384, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.8.attn.w_msa.proj.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.8.norm2.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.8.norm2.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.8.ffn.layers.0.0.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.8.ffn.layers.0.0.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.8.ffn.layers.1.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.8.ffn.layers.1.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.9.norm1.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.9.norm1.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.9.attn.w_msa.relative_position_bias_table - torch.Size([169, 12]): +Initialized by user-defined `init_weights` in WindowMSA + +backbone.stages.2.blocks.9.attn.w_msa.qkv.weight - torch.Size([1152, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.9.attn.w_msa.qkv.bias - torch.Size([1152]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.9.attn.w_msa.proj.weight - torch.Size([384, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.9.attn.w_msa.proj.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.9.norm2.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.9.norm2.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.9.ffn.layers.0.0.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.9.ffn.layers.0.0.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.9.ffn.layers.1.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.9.ffn.layers.1.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.10.norm1.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.10.norm1.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.10.attn.w_msa.relative_position_bias_table - torch.Size([169, 12]): +Initialized by user-defined `init_weights` in WindowMSA + +backbone.stages.2.blocks.10.attn.w_msa.qkv.weight - torch.Size([1152, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.10.attn.w_msa.qkv.bias - torch.Size([1152]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.10.attn.w_msa.proj.weight - torch.Size([384, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.10.attn.w_msa.proj.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.10.norm2.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.10.norm2.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.10.ffn.layers.0.0.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.10.ffn.layers.0.0.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.10.ffn.layers.1.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.10.ffn.layers.1.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.11.norm1.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.11.norm1.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.11.attn.w_msa.relative_position_bias_table - torch.Size([169, 12]): +Initialized by user-defined `init_weights` in WindowMSA + +backbone.stages.2.blocks.11.attn.w_msa.qkv.weight - torch.Size([1152, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.11.attn.w_msa.qkv.bias - torch.Size([1152]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.11.attn.w_msa.proj.weight - torch.Size([384, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.11.attn.w_msa.proj.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.11.norm2.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.11.norm2.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.11.ffn.layers.0.0.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.11.ffn.layers.0.0.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.11.ffn.layers.1.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.11.ffn.layers.1.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.12.norm1.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.12.norm1.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.12.attn.w_msa.relative_position_bias_table - torch.Size([169, 12]): +Initialized by user-defined `init_weights` in WindowMSA + +backbone.stages.2.blocks.12.attn.w_msa.qkv.weight - torch.Size([1152, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.12.attn.w_msa.qkv.bias - torch.Size([1152]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.12.attn.w_msa.proj.weight - torch.Size([384, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.12.attn.w_msa.proj.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.12.norm2.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.12.norm2.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.12.ffn.layers.0.0.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.12.ffn.layers.0.0.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.12.ffn.layers.1.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.12.ffn.layers.1.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.13.norm1.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.13.norm1.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.13.attn.w_msa.relative_position_bias_table - torch.Size([169, 12]): +Initialized by user-defined `init_weights` in WindowMSA + +backbone.stages.2.blocks.13.attn.w_msa.qkv.weight - torch.Size([1152, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.13.attn.w_msa.qkv.bias - torch.Size([1152]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.13.attn.w_msa.proj.weight - torch.Size([384, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.13.attn.w_msa.proj.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.13.norm2.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.13.norm2.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.13.ffn.layers.0.0.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.13.ffn.layers.0.0.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.13.ffn.layers.1.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.13.ffn.layers.1.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.14.norm1.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.14.norm1.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.14.attn.w_msa.relative_position_bias_table - torch.Size([169, 12]): +Initialized by user-defined `init_weights` in WindowMSA + +backbone.stages.2.blocks.14.attn.w_msa.qkv.weight - torch.Size([1152, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.14.attn.w_msa.qkv.bias - torch.Size([1152]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.14.attn.w_msa.proj.weight - torch.Size([384, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.14.attn.w_msa.proj.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.14.norm2.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.14.norm2.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.14.ffn.layers.0.0.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.14.ffn.layers.0.0.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.14.ffn.layers.1.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.14.ffn.layers.1.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.15.norm1.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.15.norm1.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.15.attn.w_msa.relative_position_bias_table - torch.Size([169, 12]): +Initialized by user-defined `init_weights` in WindowMSA + +backbone.stages.2.blocks.15.attn.w_msa.qkv.weight - torch.Size([1152, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.15.attn.w_msa.qkv.bias - torch.Size([1152]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.15.attn.w_msa.proj.weight - torch.Size([384, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.15.attn.w_msa.proj.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.15.norm2.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.15.norm2.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.15.ffn.layers.0.0.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.15.ffn.layers.0.0.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.15.ffn.layers.1.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.15.ffn.layers.1.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.16.norm1.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.16.norm1.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.16.attn.w_msa.relative_position_bias_table - torch.Size([169, 12]): +Initialized by user-defined `init_weights` in WindowMSA + +backbone.stages.2.blocks.16.attn.w_msa.qkv.weight - torch.Size([1152, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.16.attn.w_msa.qkv.bias - torch.Size([1152]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.16.attn.w_msa.proj.weight - torch.Size([384, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.16.attn.w_msa.proj.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.16.norm2.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.16.norm2.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.16.ffn.layers.0.0.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.16.ffn.layers.0.0.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.16.ffn.layers.1.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.16.ffn.layers.1.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.17.norm1.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.17.norm1.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.17.attn.w_msa.relative_position_bias_table - torch.Size([169, 12]): +Initialized by user-defined `init_weights` in WindowMSA + +backbone.stages.2.blocks.17.attn.w_msa.qkv.weight - torch.Size([1152, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.17.attn.w_msa.qkv.bias - torch.Size([1152]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.17.attn.w_msa.proj.weight - torch.Size([384, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.17.attn.w_msa.proj.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.17.norm2.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.17.norm2.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.17.ffn.layers.0.0.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.17.ffn.layers.0.0.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.17.ffn.layers.1.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.17.ffn.layers.1.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.downsample.norm.weight - torch.Size([1536]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.downsample.norm.bias - torch.Size([1536]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.downsample.reduction.weight - torch.Size([768, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.blocks.0.norm1.weight - torch.Size([768]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.3.blocks.0.norm1.bias - torch.Size([768]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.3.blocks.0.attn.w_msa.relative_position_bias_table - torch.Size([169, 24]): +Initialized by user-defined `init_weights` in WindowMSA + +backbone.stages.3.blocks.0.attn.w_msa.qkv.weight - torch.Size([2304, 768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.blocks.0.attn.w_msa.qkv.bias - torch.Size([2304]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.blocks.0.attn.w_msa.proj.weight - torch.Size([768, 768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.blocks.0.attn.w_msa.proj.bias - torch.Size([768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.blocks.0.norm2.weight - torch.Size([768]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.3.blocks.0.norm2.bias - torch.Size([768]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.3.blocks.0.ffn.layers.0.0.weight - torch.Size([3072, 768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.blocks.0.ffn.layers.0.0.bias - torch.Size([3072]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.blocks.0.ffn.layers.1.weight - torch.Size([768, 3072]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.blocks.0.ffn.layers.1.bias - torch.Size([768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.blocks.1.norm1.weight - torch.Size([768]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.3.blocks.1.norm1.bias - torch.Size([768]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.3.blocks.1.attn.w_msa.relative_position_bias_table - torch.Size([169, 24]): +Initialized by user-defined `init_weights` in WindowMSA + +backbone.stages.3.blocks.1.attn.w_msa.qkv.weight - torch.Size([2304, 768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.blocks.1.attn.w_msa.qkv.bias - torch.Size([2304]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.blocks.1.attn.w_msa.proj.weight - torch.Size([768, 768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.blocks.1.attn.w_msa.proj.bias - torch.Size([768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.blocks.1.norm2.weight - torch.Size([768]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.3.blocks.1.norm2.bias - torch.Size([768]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.3.blocks.1.ffn.layers.0.0.weight - torch.Size([3072, 768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.blocks.1.ffn.layers.0.0.bias - torch.Size([3072]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.blocks.1.ffn.layers.1.weight - torch.Size([768, 3072]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.blocks.1.ffn.layers.1.bias - torch.Size([768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.norm3.weight - torch.Size([768]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.norm3.bias - torch.Size([768]): +The value is the same before and after calling `init_weights` of ImageClassifier + +head.fc.weight - torch.Size([2, 768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +head.fc.bias - torch.Size([2]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 +2023/05/31 18:33:22 - mmengine - WARNING - "FileClient" will be deprecated in future. Please use io functions in https://mmengine.readthedocs.io/en/latest/api/fileio.html#file-io +2023/05/31 18:33:22 - mmengine - WARNING - "HardDiskBackend" is the alias of "LocalBackend" and the former will be deprecated in future. +2023/05/31 18:33:22 - mmengine - INFO - Checkpoints will be saved to /mnt/petrelfs/luzeyu/workspace/fakebench/mmpretrain/workdir/swin_small_4xb256_1e-3lr_5m. +2023/05/31 18:34:59 - mmengine - INFO - Epoch(train) [1][ 100/5758] lr: 1.0000e-03 eta: 1 day, 6:58:15 time: 0.8413 data_time: 0.3359 memory: 28798 grad_norm: 1.1872 loss: 0.6847 +2023/05/31 18:36:27 - mmengine - INFO - Epoch(train) [1][ 200/5758] lr: 1.0000e-03 eta: 1 day, 5:36:42 time: 0.8396 data_time: 0.2166 memory: 28798 grad_norm: 1.7218 loss: 0.6908 +2023/05/31 18:37:57 - mmengine - INFO - Epoch(train) [1][ 300/5758] lr: 1.0000e-03 eta: 1 day, 5:15:55 time: 0.8651 data_time: 0.0020 memory: 28798 grad_norm: 0.7792 loss: 0.6860 +2023/05/31 18:39:30 - mmengine - INFO - Epoch(train) [1][ 400/5758] lr: 1.0000e-03 eta: 1 day, 5:21:34 time: 0.9165 data_time: 0.0023 memory: 28798 grad_norm: 1.8431 loss: 0.6948 +2023/05/31 18:41:03 - mmengine - INFO - Epoch(train) [1][ 500/5758] lr: 1.0000e-03 eta: 1 day, 5:21:07 time: 0.9577 data_time: 0.0019 memory: 28798 grad_norm: 0.6886 loss: 0.6912 +2023/05/31 18:42:33 - mmengine - INFO - Epoch(train) [1][ 600/5758] lr: 1.0000e-03 eta: 1 day, 5:15:34 time: 0.8940 data_time: 0.0020 memory: 28798 grad_norm: 0.6635 loss: 0.6864 +2023/05/31 18:44:09 - mmengine - INFO - Epoch(train) [1][ 700/5758] lr: 1.0000e-03 eta: 1 day, 5:24:36 time: 0.9112 data_time: 0.0022 memory: 28798 grad_norm: 0.6898 loss: 0.6847 +2023/05/31 18:45:41 - mmengine - INFO - Epoch(train) [1][ 800/5758] lr: 1.0000e-03 eta: 1 day, 5:22:22 time: 0.9439 data_time: 0.0016 memory: 28798 grad_norm: 1.3765 loss: 0.6887 +2023/05/31 18:47:14 - mmengine - INFO - Epoch(train) [1][ 900/5758] lr: 1.0000e-03 eta: 1 day, 5:20:06 time: 0.9014 data_time: 0.0017 memory: 28798 grad_norm: 0.8592 loss: 0.6897 +2023/05/31 18:48:55 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_20230531_183239 +2023/05/31 18:48:55 - mmengine - INFO - Epoch(train) [1][1000/5758] lr: 1.0000e-03 eta: 1 day, 5:36:22 time: 1.1501 data_time: 0.0018 memory: 28798 grad_norm: 0.5187 loss: 0.6888 +2023/05/31 18:50:46 - mmengine - INFO - Epoch(train) [1][1100/5758] lr: 1.0000e-03 eta: 1 day, 6:05:03 time: 1.0909 data_time: 0.0016 memory: 28798 grad_norm: 0.5395 loss: 0.6882 +2023/05/31 18:52:39 - mmengine - INFO - Epoch(train) [1][1200/5758] lr: 1.0000e-03 eta: 1 day, 6:32:10 time: 1.7967 data_time: 0.0017 memory: 28798 grad_norm: 0.9257 loss: 0.6898 +2023/05/31 18:54:21 - mmengine - INFO - Epoch(train) [1][1300/5758] lr: 1.0000e-03 eta: 1 day, 6:37:43 time: 0.9751 data_time: 0.0017 memory: 28798 grad_norm: 0.3737 loss: 0.6880 +2023/05/31 18:55:57 - mmengine - INFO - Epoch(train) [1][1400/5758] lr: 1.0000e-03 eta: 1 day, 6:35:46 time: 0.8699 data_time: 0.0017 memory: 28798 grad_norm: 0.7636 loss: 0.6904 +2023/05/31 18:57:34 - mmengine - INFO - Epoch(train) [1][1500/5758] lr: 1.0000e-03 eta: 1 day, 6:34:32 time: 0.9833 data_time: 0.0016 memory: 28798 grad_norm: 0.4469 loss: 0.6884 +2023/05/31 18:59:11 - mmengine - INFO - Epoch(train) [1][1600/5758] lr: 1.0000e-03 eta: 1 day, 6:32:52 time: 0.8429 data_time: 0.0017 memory: 28798 grad_norm: 0.2130 loss: 0.6856 +2023/05/31 19:00:46 - mmengine - INFO - Epoch(train) [1][1700/5758] lr: 1.0000e-03 eta: 1 day, 6:28:39 time: 0.9024 data_time: 0.0014 memory: 28798 grad_norm: 0.3191 loss: 0.6899 +2023/05/31 19:02:21 - mmengine - INFO - Epoch(train) [1][1800/5758] lr: 1.0000e-03 eta: 1 day, 6:25:45 time: 0.9407 data_time: 0.0016 memory: 28798 grad_norm: 0.3840 loss: 0.6844 +2023/05/31 19:03:54 - mmengine - INFO - Epoch(train) [1][1900/5758] lr: 1.0000e-03 eta: 1 day, 6:20:10 time: 0.9031 data_time: 0.0016 memory: 28798 grad_norm: 0.2089 loss: 0.6874 +2023/05/31 19:05:29 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_20230531_183239 +2023/05/31 19:05:29 - mmengine - INFO - Epoch(train) [1][2000/5758] lr: 1.0000e-03 eta: 1 day, 6:16:59 time: 1.0162 data_time: 0.0015 memory: 28798 grad_norm: 0.2961 loss: 0.6883 +2023/05/31 19:07:09 - mmengine - INFO - Epoch(train) [1][2100/5758] lr: 1.0000e-03 eta: 1 day, 6:19:18 time: 0.9673 data_time: 0.0019 memory: 28798 grad_norm: 0.2335 loss: 0.6890 +2023/05/31 19:08:57 - mmengine - INFO - Epoch(train) [1][2200/5758] lr: 1.0000e-03 eta: 1 day, 6:27:16 time: 1.1158 data_time: 0.0017 memory: 28798 grad_norm: 0.2770 loss: 0.6872 +2023/05/31 19:10:37 - mmengine - INFO - Epoch(train) [1][2300/5758] lr: 1.0000e-03 eta: 1 day, 6:28:18 time: 1.0251 data_time: 0.0018 memory: 28798 grad_norm: 0.1662 loss: 0.6853 +2023/05/31 19:12:15 - mmengine - INFO - Epoch(train) [1][2400/5758] lr: 1.0000e-03 eta: 1 day, 6:27:16 time: 0.9310 data_time: 0.0013 memory: 28798 grad_norm: 0.1667 loss: 0.6864 +2023/05/31 19:13:47 - mmengine - INFO - Epoch(train) [1][2500/5758] lr: 1.0000e-03 eta: 1 day, 6:21:32 time: 0.8551 data_time: 0.0020 memory: 28798 grad_norm: 0.2844 loss: 0.6850 +2023/05/31 19:15:17 - mmengine - INFO - Epoch(train) [1][2600/5758] lr: 1.0000e-03 eta: 1 day, 6:14:58 time: 0.9001 data_time: 0.0017 memory: 28798 grad_norm: 0.2358 loss: 0.6867 +2023/05/31 19:16:49 - mmengine - INFO - Epoch(train) [1][2700/5758] lr: 1.0000e-03 eta: 1 day, 6:10:12 time: 0.8460 data_time: 0.0016 memory: 28798 grad_norm: 0.2686 loss: 0.6846 +2023/05/31 19:18:21 - mmengine - INFO - Epoch(train) [1][2800/5758] lr: 1.0000e-03 eta: 1 day, 6:05:28 time: 0.9115 data_time: 0.0017 memory: 28798 grad_norm: 0.1900 loss: 0.6850 +2023/05/31 19:19:53 - mmengine - INFO - Epoch(train) [1][2900/5758] lr: 1.0000e-03 eta: 1 day, 6:00:52 time: 0.9324 data_time: 0.0020 memory: 28798 grad_norm: 0.2707 loss: 0.6882 +2023/05/31 19:21:25 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_20230531_183239 +2023/05/31 19:21:25 - mmengine - INFO - Epoch(train) [1][3000/5758] lr: 1.0000e-03 eta: 1 day, 5:56:16 time: 0.8894 data_time: 0.0017 memory: 28798 grad_norm: 0.1762 loss: 0.6838 +2023/05/31 19:22:56 - mmengine - INFO - Epoch(train) [1][3100/5758] lr: 1.0000e-03 eta: 1 day, 5:51:57 time: 0.9973 data_time: 0.0018 memory: 28798 grad_norm: 0.1975 loss: 0.6861 +2023/05/31 19:24:27 - mmengine - INFO - Epoch(train) [1][3200/5758] lr: 1.0000e-03 eta: 1 day, 5:47:20 time: 0.8815 data_time: 0.0017 memory: 28798 grad_norm: 0.1854 loss: 0.6902 +2023/05/31 19:26:03 - mmengine - INFO - Epoch(train) [1][3300/5758] lr: 1.0000e-03 eta: 1 day, 5:46:05 time: 0.9324 data_time: 0.0018 memory: 28798 grad_norm: 0.0923 loss: 0.6834 +2023/05/31 19:27:39 - mmengine - INFO - Epoch(train) [1][3400/5758] lr: 1.0000e-03 eta: 1 day, 5:44:31 time: 1.0112 data_time: 0.0016 memory: 28798 grad_norm: 0.1994 loss: 0.6885 +2023/05/31 19:29:10 - mmengine - INFO - Epoch(train) [1][3500/5758] lr: 1.0000e-03 eta: 1 day, 5:40:29 time: 0.8954 data_time: 0.0018 memory: 28798 grad_norm: 0.1607 loss: 0.6889 +2023/05/31 19:30:41 - mmengine - INFO - Epoch(train) [1][3600/5758] lr: 1.0000e-03 eta: 1 day, 5:36:10 time: 0.9707 data_time: 0.0020 memory: 28798 grad_norm: 0.1494 loss: 0.6931 +2023/05/31 19:32:13 - mmengine - INFO - Epoch(train) [1][3700/5758] lr: 1.0000e-03 eta: 1 day, 5:32:49 time: 0.9076 data_time: 0.0019 memory: 28798 grad_norm: 0.1205 loss: 0.6883 +2023/05/31 19:33:40 - mmengine - INFO - Epoch(train) [1][3800/5758] lr: 1.0000e-03 eta: 1 day, 5:26:59 time: 0.7688 data_time: 0.0018 memory: 28798 grad_norm: 0.1381 loss: 0.6835 +2023/05/31 19:35:08 - mmengine - INFO - Epoch(train) [1][3900/5758] lr: 1.0000e-03 eta: 1 day, 5:22:03 time: 0.8670 data_time: 0.0017 memory: 28798 grad_norm: 0.1866 loss: 0.6844 +2023/05/31 19:36:35 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_20230531_183239 +2023/05/31 19:36:35 - mmengine - INFO - Epoch(train) [1][4000/5758] lr: 1.0000e-03 eta: 1 day, 5:16:53 time: 0.8676 data_time: 0.0017 memory: 28798 grad_norm: 0.1825 loss: 0.6877 +2023/05/31 19:37:59 - mmengine - INFO - Epoch(train) [1][4100/5758] lr: 1.0000e-03 eta: 1 day, 5:10:29 time: 0.7381 data_time: 0.0021 memory: 28798 grad_norm: 0.1300 loss: 0.6866 +2023/05/31 19:39:25 - mmengine - INFO - Epoch(train) [1][4200/5758] lr: 1.0000e-03 eta: 1 day, 5:04:56 time: 0.7909 data_time: 0.0016 memory: 28798 grad_norm: 0.1363 loss: 0.6868 +2023/05/31 19:40:48 - mmengine - INFO - Epoch(train) [1][4300/5758] lr: 1.0000e-03 eta: 1 day, 4:58:44 time: 0.8414 data_time: 0.0018 memory: 28798 grad_norm: 0.1306 loss: 0.6877 +2023/05/31 19:42:11 - mmengine - INFO - Epoch(train) [1][4400/5758] lr: 1.0000e-03 eta: 1 day, 4:52:22 time: 0.8178 data_time: 0.0017 memory: 28798 grad_norm: 0.7684 loss: 0.6813 +2023/05/31 19:43:37 - mmengine - INFO - Epoch(train) [1][4500/5758] lr: 1.0000e-03 eta: 1 day, 4:47:30 time: 0.8225 data_time: 0.0015 memory: 28798 grad_norm: 0.1537 loss: 0.6842 +2023/05/31 19:45:02 - mmengine - INFO - Epoch(train) [1][4600/5758] lr: 1.0000e-03 eta: 1 day, 4:42:30 time: 0.8346 data_time: 0.0015 memory: 28798 grad_norm: 0.1001 loss: 0.6867 +2023/05/31 19:46:20 - mmengine - INFO - Epoch(train) [1][4700/5758] lr: 1.0000e-03 eta: 1 day, 4:35:04 time: 0.7684 data_time: 0.0015 memory: 28798 grad_norm: 0.0917 loss: 0.6881 +2023/05/31 19:47:40 - mmengine - INFO - Epoch(train) [1][4800/5758] lr: 1.0000e-03 eta: 1 day, 4:28:19 time: 0.8191 data_time: 0.0016 memory: 28798 grad_norm: 0.0973 loss: 0.6853 +2023/05/31 19:48:59 - mmengine - INFO - Epoch(train) [1][4900/5758] lr: 1.0000e-03 eta: 1 day, 4:21:46 time: 0.7956 data_time: 0.0018 memory: 28798 grad_norm: 0.1100 loss: 0.6863 +2023/05/31 19:50:19 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_20230531_183239 +2023/05/31 19:50:19 - mmengine - INFO - Epoch(train) [1][5000/5758] lr: 1.0000e-03 eta: 1 day, 4:15:34 time: 0.8014 data_time: 0.0016 memory: 28798 grad_norm: 0.1635 loss: 0.6862 +2023/05/31 19:51:40 - mmengine - INFO - Epoch(train) [1][5100/5758] lr: 1.0000e-03 eta: 1 day, 4:09:44 time: 0.7832 data_time: 0.0019 memory: 28798 grad_norm: 0.1212 loss: 0.6847 +2023/05/31 19:53:00 - mmengine - INFO - Epoch(train) [1][5200/5758] lr: 1.0000e-03 eta: 1 day, 4:03:54 time: 0.7723 data_time: 0.0019 memory: 28798 grad_norm: 0.1053 loss: 0.6855 +2023/05/31 19:54:19 - mmengine - INFO - Epoch(train) [1][5300/5758] lr: 1.0000e-03 eta: 1 day, 3:58:04 time: 0.8393 data_time: 0.0020 memory: 28798 grad_norm: 0.1145 loss: 0.6891 +2023/05/31 19:55:40 - mmengine - INFO - Epoch(train) [1][5400/5758] lr: 1.0000e-03 eta: 1 day, 3:52:49 time: 0.8280 data_time: 0.0025 memory: 28798 grad_norm: 0.1033 loss: 0.6884 +2023/05/31 19:57:00 - mmengine - INFO - Epoch(train) [1][5500/5758] lr: 1.0000e-03 eta: 1 day, 3:47:31 time: 0.8244 data_time: 0.0019 memory: 28798 grad_norm: 0.1317 loss: 0.6850 +2023/05/31 19:58:21 - mmengine - INFO - Epoch(train) [1][5600/5758] lr: 1.0000e-03 eta: 1 day, 3:42:49 time: 0.8310 data_time: 0.0017 memory: 28798 grad_norm: 0.0903 loss: 0.6849 +2023/05/31 19:59:45 - mmengine - INFO - Epoch(train) [1][5700/5758] lr: 1.0000e-03 eta: 1 day, 3:38:48 time: 0.8125 data_time: 0.0016 memory: 28798 grad_norm: 0.1151 loss: 0.6874 +2023/05/31 20:00:33 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_20230531_183239 +2023/05/31 20:00:33 - mmengine - INFO - Saving checkpoint at 1 epochs +2023/05/31 20:00:54 - mmengine - INFO - Epoch(val) [1][16/16] accuracy/top1: 100.0000 single-label/precision_classwise: [100.0, 0.0] single-label/recall_classwise: [100.0, 0.0] single-label/f1-score_classwise: [100.0, 0.0] data_time: 0.2905 time: 0.8415 +2023/05/31 20:02:19 - mmengine - INFO - Epoch(train) [2][ 100/5758] lr: 9.9391e-04 eta: 1 day, 3:33:18 time: 0.7386 data_time: 0.1149 memory: 28798 grad_norm: 0.1119 loss: 0.6852 +2023/05/31 20:03:43 - mmengine - INFO - Epoch(train) [2][ 200/5758] lr: 9.9391e-04 eta: 1 day, 3:29:35 time: 0.7840 data_time: 0.2354 memory: 28783 grad_norm: 0.1253 loss: 0.6893 +2023/05/31 20:04:18 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_20230531_183239 +2023/05/31 20:05:07 - mmengine - INFO - Epoch(train) [2][ 300/5758] lr: 9.9391e-04 eta: 1 day, 3:26:04 time: 0.8326 data_time: 0.0019 memory: 28783 grad_norm: 0.0912 loss: 0.6850 +2023/05/31 20:06:33 - mmengine - INFO - Epoch(train) [2][ 400/5758] lr: 9.9391e-04 eta: 1 day, 3:23:17 time: 0.8944 data_time: 0.0017 memory: 28783 grad_norm: 0.0847 loss: 0.6855 +2023/05/31 20:08:02 - mmengine - INFO - Epoch(train) [2][ 500/5758] lr: 9.9391e-04 eta: 1 day, 3:21:27 time: 0.8275 data_time: 0.0020 memory: 28783 grad_norm: 0.0976 loss: 0.6897 +2023/05/31 20:09:34 - mmengine - INFO - Epoch(train) [2][ 600/5758] lr: 9.9391e-04 eta: 1 day, 3:20:10 time: 0.9148 data_time: 0.0027 memory: 28783 grad_norm: 0.0642 loss: 0.6863 +2023/05/31 20:11:02 - mmengine - INFO - Epoch(train) [2][ 700/5758] lr: 9.9391e-04 eta: 1 day, 3:18:11 time: 0.8873 data_time: 0.0020 memory: 28783 grad_norm: 0.0912 loss: 0.6880 +2023/05/31 20:12:28 - mmengine - INFO - Epoch(train) [2][ 800/5758] lr: 9.9391e-04 eta: 1 day, 3:15:26 time: 0.8819 data_time: 0.0025 memory: 28783 grad_norm: 0.1290 loss: 0.6882 +2023/05/31 20:13:56 - mmengine - INFO - Epoch(train) [2][ 900/5758] lr: 9.9391e-04 eta: 1 day, 3:13:17 time: 0.8586 data_time: 0.0326 memory: 28783 grad_norm: 0.1375 loss: 0.6855 +2023/05/31 20:15:24 - mmengine - INFO - Epoch(train) [2][1000/5758] lr: 9.9391e-04 eta: 1 day, 3:11:00 time: 0.8951 data_time: 0.0013 memory: 28783 grad_norm: 0.0986 loss: 0.6866 +2023/05/31 20:16:51 - mmengine - INFO - Epoch(train) [2][1100/5758] lr: 9.9391e-04 eta: 1 day, 3:08:45 time: 0.7790 data_time: 0.0019 memory: 28783 grad_norm: 0.1306 loss: 0.6851 +2023/05/31 20:18:20 - mmengine - INFO - Epoch(train) [2][1200/5758] lr: 9.9391e-04 eta: 1 day, 3:06:49 time: 0.9762 data_time: 0.0017 memory: 28783 grad_norm: 0.0645 loss: 0.6846 +2023/05/31 20:18:56 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_20230531_183239 +2023/05/31 20:19:47 - mmengine - INFO - Epoch(train) [2][1300/5758] lr: 9.9391e-04 eta: 1 day, 3:04:32 time: 0.8225 data_time: 0.0025 memory: 28783 grad_norm: 0.0656 loss: 0.6870 +2023/05/31 20:21:17 - mmengine - INFO - Epoch(train) [2][1400/5758] lr: 9.9391e-04 eta: 1 day, 3:02:56 time: 0.8734 data_time: 0.0017 memory: 28783 grad_norm: 0.0958 loss: 0.6843 +2023/05/31 20:22:43 - mmengine - INFO - Epoch(train) [2][1500/5758] lr: 9.9391e-04 eta: 1 day, 3:00:24 time: 0.8259 data_time: 0.0024 memory: 28783 grad_norm: 0.1318 loss: 0.6872 +2023/05/31 20:24:11 - mmengine - INFO - Epoch(train) [2][1600/5758] lr: 9.9391e-04 eta: 1 day, 2:58:21 time: 0.8509 data_time: 0.0020 memory: 28783 grad_norm: 0.0897 loss: 0.6851 +2023/05/31 20:25:39 - mmengine - INFO - Epoch(train) [2][1700/5758] lr: 9.9391e-04 eta: 1 day, 2:56:26 time: 0.8398 data_time: 0.0026 memory: 28783 grad_norm: 0.0946 loss: 0.6866 +2023/05/31 20:27:06 - mmengine - INFO - Epoch(train) [2][1800/5758] lr: 9.9391e-04 eta: 1 day, 2:54:11 time: 0.8577 data_time: 0.0017 memory: 28783 grad_norm: 0.0437 loss: 0.6842 +2023/05/31 20:28:34 - mmengine - INFO - Epoch(train) [2][1900/5758] lr: 9.9391e-04 eta: 1 day, 2:52:23 time: 0.9476 data_time: 0.0019 memory: 28783 grad_norm: 0.1016 loss: 0.6883 +2023/05/31 20:30:04 - mmengine - INFO - Epoch(train) [2][2000/5758] lr: 9.9391e-04 eta: 1 day, 2:50:43 time: 0.8512 data_time: 0.0017 memory: 28783 grad_norm: 0.0500 loss: 0.6900 +2023/05/31 20:31:34 - mmengine - INFO - Epoch(train) [2][2100/5758] lr: 9.9391e-04 eta: 1 day, 2:49:17 time: 0.8802 data_time: 0.0016 memory: 28783 grad_norm: 0.1089 loss: 0.6876 +2023/05/31 20:33:04 - mmengine - INFO - Epoch(train) [2][2200/5758] lr: 9.9391e-04 eta: 1 day, 2:47:49 time: 0.8773 data_time: 0.0016 memory: 28783 grad_norm: 0.0597 loss: 0.6860 +2023/05/31 20:33:40 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_20230531_183239 +2023/05/31 20:34:32 - mmengine - INFO - Epoch(train) [2][2300/5758] lr: 9.9391e-04 eta: 1 day, 2:45:49 time: 0.8825 data_time: 0.0016 memory: 28783 grad_norm: 0.0889 loss: 0.6890 +2023/05/31 20:36:00 - mmengine - INFO - Epoch(train) [2][2400/5758] lr: 9.9391e-04 eta: 1 day, 2:44:00 time: 0.8731 data_time: 0.0020 memory: 28783 grad_norm: 0.0611 loss: 0.6835 +2023/05/31 20:37:31 - mmengine - INFO - Epoch(train) [2][2500/5758] lr: 9.9391e-04 eta: 1 day, 2:42:34 time: 0.8559 data_time: 0.0018 memory: 28783 grad_norm: 0.0475 loss: 0.6855 +2023/05/31 20:38:57 - mmengine - INFO - Epoch(train) [2][2600/5758] lr: 9.9391e-04 eta: 1 day, 2:40:14 time: 0.8678 data_time: 0.0023 memory: 28783 grad_norm: 0.0290 loss: 0.6850 +2023/05/31 20:40:23 - mmengine - INFO - Epoch(train) [2][2700/5758] lr: 9.9391e-04 eta: 1 day, 2:38:05 time: 0.8611 data_time: 0.0016 memory: 28783 grad_norm: 0.0631 loss: 0.6903 +2023/05/31 20:41:51 - mmengine - INFO - Epoch(train) [2][2800/5758] lr: 9.9391e-04 eta: 1 day, 2:36:03 time: 0.8659 data_time: 0.0018 memory: 28783 grad_norm: 0.0540 loss: 0.6853 +2023/05/31 20:43:17 - mmengine - INFO - Epoch(train) [2][2900/5758] lr: 9.9391e-04 eta: 1 day, 2:33:50 time: 0.8212 data_time: 0.0019 memory: 28783 grad_norm: 0.0654 loss: 0.6853 +2023/05/31 20:44:44 - mmengine - INFO - Epoch(train) [2][3000/5758] lr: 9.9391e-04 eta: 1 day, 2:31:40 time: 0.9224 data_time: 0.0033 memory: 28783 grad_norm: 0.0573 loss: 0.6872 +2023/05/31 20:46:13 - mmengine - INFO - Epoch(train) [2][3100/5758] lr: 9.9391e-04 eta: 1 day, 2:30:04 time: 0.9416 data_time: 0.0020 memory: 28783 grad_norm: 0.0750 loss: 0.6859 +2023/05/31 20:47:40 - mmengine - INFO - Epoch(train) [2][3200/5758] lr: 9.9391e-04 eta: 1 day, 2:28:08 time: 0.8827 data_time: 0.0025 memory: 28783 grad_norm: 0.0529 loss: 0.6839 +2023/05/31 20:48:15 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_20230531_183239 +2023/05/31 20:49:07 - mmengine - INFO - Epoch(train) [2][3300/5758] lr: 9.9391e-04 eta: 1 day, 2:26:00 time: 0.8703 data_time: 0.0020 memory: 28783 grad_norm: 0.0385 loss: 0.6882 +2023/05/31 20:50:31 - mmengine - INFO - Epoch(train) [2][3400/5758] lr: 9.9391e-04 eta: 1 day, 2:23:24 time: 0.7982 data_time: 0.0023 memory: 28783 grad_norm: 0.0430 loss: 0.6866 +2023/05/31 20:51:57 - mmengine - INFO - Epoch(train) [2][3500/5758] lr: 9.9391e-04 eta: 1 day, 2:21:19 time: 0.8314 data_time: 0.0027 memory: 28783 grad_norm: 0.0509 loss: 0.6848 +2023/05/31 20:53:23 - mmengine - INFO - Epoch(train) [2][3600/5758] lr: 9.9391e-04 eta: 1 day, 2:19:07 time: 0.8449 data_time: 0.0026 memory: 28783 grad_norm: 0.0554 loss: 0.6863 +2023/05/31 20:54:50 - mmengine - INFO - Epoch(train) [2][3700/5758] lr: 9.9391e-04 eta: 1 day, 2:17:05 time: 0.8194 data_time: 0.0018 memory: 28783 grad_norm: 0.0331 loss: 0.6851 +2023/05/31 20:56:17 - mmengine - INFO - Epoch(train) [2][3800/5758] lr: 9.9391e-04 eta: 1 day, 2:15:08 time: 0.8861 data_time: 0.0023 memory: 28783 grad_norm: 0.0578 loss: 0.6863 +2023/05/31 20:57:44 - mmengine - INFO - Epoch(train) [2][3900/5758] lr: 9.9391e-04 eta: 1 day, 2:13:13 time: 0.8959 data_time: 0.0019 memory: 28783 grad_norm: 0.0615 loss: 0.6837 +2023/05/31 20:59:09 - mmengine - INFO - Epoch(train) [2][4000/5758] lr: 9.9391e-04 eta: 1 day, 2:10:56 time: 0.8687 data_time: 0.0028 memory: 28783 grad_norm: 0.0449 loss: 0.6884 +2023/05/31 21:00:36 - mmengine - INFO - Epoch(train) [2][4100/5758] lr: 9.9391e-04 eta: 1 day, 2:09:03 time: 0.8510 data_time: 0.0029 memory: 28783 grad_norm: 0.0378 loss: 0.6857 +2023/05/31 21:02:01 - mmengine - INFO - Epoch(train) [2][4200/5758] lr: 9.9391e-04 eta: 1 day, 2:06:42 time: 0.8748 data_time: 0.0018 memory: 28783 grad_norm: 0.0520 loss: 0.6859 +2023/05/31 21:02:36 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_20230531_183239 +2023/05/31 21:03:25 - mmengine - INFO - Epoch(train) [2][4300/5758] lr: 9.9391e-04 eta: 1 day, 2:04:22 time: 0.9334 data_time: 0.0026 memory: 28783 grad_norm: 0.0385 loss: 0.6874 +2023/05/31 21:04:51 - mmengine - INFO - Epoch(train) [2][4400/5758] lr: 9.9391e-04 eta: 1 day, 2:02:13 time: 0.8259 data_time: 0.0024 memory: 28783 grad_norm: 0.0380 loss: 0.6890 +2023/05/31 21:06:17 - mmengine - INFO - Epoch(train) [2][4500/5758] lr: 9.9391e-04 eta: 1 day, 2:00:08 time: 0.8748 data_time: 0.0021 memory: 28783 grad_norm: 0.0496 loss: 0.6844 +2023/05/31 21:07:45 - mmengine - INFO - Epoch(train) [2][4600/5758] lr: 9.9391e-04 eta: 1 day, 1:58:34 time: 0.9434 data_time: 0.0018 memory: 28783 grad_norm: 0.0517 loss: 0.6877 +2023/05/31 21:09:11 - mmengine - INFO - Epoch(train) [2][4700/5758] lr: 9.9391e-04 eta: 1 day, 1:56:30 time: 0.7991 data_time: 0.0017 memory: 28783 grad_norm: 0.0274 loss: 0.6842 +2023/05/31 21:10:35 - mmengine - INFO - Epoch(train) [2][4800/5758] lr: 9.9391e-04 eta: 1 day, 1:54:06 time: 0.8281 data_time: 0.0026 memory: 28783 grad_norm: 0.0509 loss: 0.6882 +2023/05/31 21:11:59 - mmengine - INFO - Epoch(train) [2][4900/5758] lr: 9.9391e-04 eta: 1 day, 1:51:51 time: 0.8703 data_time: 0.0023 memory: 28783 grad_norm: 0.0358 loss: 0.6883 +2023/05/31 21:13:24 - mmengine - INFO - Epoch(train) [2][5000/5758] lr: 9.9391e-04 eta: 1 day, 1:49:38 time: 0.8387 data_time: 0.0020 memory: 28783 grad_norm: 0.0305 loss: 0.6890 +2023/05/31 21:14:48 - mmengine - INFO - Epoch(train) [2][5100/5758] lr: 9.9391e-04 eta: 1 day, 1:47:18 time: 0.8687 data_time: 0.0016 memory: 28783 grad_norm: 0.0382 loss: 0.6871 +2023/05/31 21:16:14 - mmengine - INFO - Epoch(train) [2][5200/5758] lr: 9.9391e-04 eta: 1 day, 1:45:20 time: 0.8054 data_time: 0.0016 memory: 28783 grad_norm: 0.0416 loss: 0.6893 +2023/05/31 21:16:48 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_20230531_183239 +2023/05/31 21:17:39 - mmengine - INFO - Epoch(train) [2][5300/5758] lr: 9.9391e-04 eta: 1 day, 1:43:20 time: 0.8645 data_time: 0.0027 memory: 28783 grad_norm: 0.0331 loss: 0.6858 +2023/05/31 21:19:04 - mmengine - INFO - Epoch(train) [2][5400/5758] lr: 9.9391e-04 eta: 1 day, 1:41:12 time: 0.8388 data_time: 0.0016 memory: 28783 grad_norm: 0.0434 loss: 0.6840 +2023/05/31 21:20:28 - mmengine - INFO - Epoch(train) [2][5500/5758] lr: 9.9391e-04 eta: 1 day, 1:39:01 time: 0.8524 data_time: 0.0031 memory: 28783 grad_norm: 0.0518 loss: 0.6873 +2023/05/31 21:21:54 - mmengine - INFO - Epoch(train) [2][5600/5758] lr: 9.9391e-04 eta: 1 day, 1:37:01 time: 0.8564 data_time: 0.0021 memory: 28783 grad_norm: 0.0398 loss: 0.6871 +2023/05/31 21:23:19 - mmengine - INFO - Epoch(train) [2][5700/5758] lr: 9.9391e-04 eta: 1 day, 1:35:03 time: 0.8307 data_time: 0.0024 memory: 28783 grad_norm: 0.0301 loss: 0.6841 +2023/05/31 21:24:09 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_20230531_183239 +2023/05/31 21:24:09 - mmengine - INFO - Saving checkpoint at 2 epochs +2023/05/31 21:24:27 - mmengine - INFO - Epoch(val) [2][16/16] accuracy/top1: 100.0000 single-label/precision_classwise: [100.0, 0.0] single-label/recall_classwise: [100.0, 0.0] single-label/f1-score_classwise: [100.0, 0.0] data_time: 0.2362 time: 0.6698 +2023/05/31 21:25:59 - mmengine - INFO - Epoch(train) [3][ 100/5758] lr: 9.7577e-04 eta: 1 day, 1:32:45 time: 0.8802 data_time: 0.1560 memory: 28783 grad_norm: 0.0320 loss: 0.6842 +2023/05/31 21:27:27 - mmengine - INFO - Epoch(train) [3][ 200/5758] lr: 9.7577e-04 eta: 1 day, 1:31:12 time: 0.8316 data_time: 0.0021 memory: 28783 grad_norm: 0.0348 loss: 0.6889 +2023/05/31 21:28:52 - mmengine - INFO - Epoch(train) [3][ 300/5758] lr: 9.7577e-04 eta: 1 day, 1:29:14 time: 0.8582 data_time: 0.0023 memory: 28783 grad_norm: 0.0401 loss: 0.6884 +2023/05/31 21:30:19 - mmengine - INFO - Epoch(train) [3][ 400/5758] lr: 9.7577e-04 eta: 1 day, 1:27:30 time: 0.8027 data_time: 0.0015 memory: 28783 grad_norm: 0.0261 loss: 0.6869 +2023/05/31 21:31:30 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_20230531_183239 +2023/05/31 21:31:45 - mmengine - INFO - Epoch(train) [3][ 500/5758] lr: 9.7577e-04 eta: 1 day, 1:25:35 time: 0.8198 data_time: 0.0018 memory: 28783 grad_norm: 0.0352 loss: 0.6843 +2023/05/31 21:33:14 - mmengine - INFO - Epoch(train) [3][ 600/5758] lr: 9.7577e-04 eta: 1 day, 1:24:04 time: 0.8314 data_time: 0.0027 memory: 28783 grad_norm: 0.0305 loss: 0.6916 +2023/05/31 21:34:41 - mmengine - INFO - Epoch(train) [3][ 700/5758] lr: 9.7577e-04 eta: 1 day, 1:22:26 time: 0.8887 data_time: 0.0022 memory: 28783 grad_norm: 0.0331 loss: 0.6845 +2023/05/31 21:36:06 - mmengine - INFO - Epoch(train) [3][ 800/5758] lr: 9.7577e-04 eta: 1 day, 1:20:23 time: 0.8935 data_time: 0.0021 memory: 28783 grad_norm: 0.0307 loss: 0.6866 +2023/05/31 21:37:35 - mmengine - INFO - Epoch(train) [3][ 900/5758] lr: 9.7577e-04 eta: 1 day, 1:18:53 time: 0.8315 data_time: 0.0022 memory: 28783 grad_norm: 0.0290 loss: 0.6870 +2023/05/31 21:38:58 - mmengine - INFO - Epoch(train) [3][1000/5758] lr: 9.7577e-04 eta: 1 day, 1:16:41 time: 0.8499 data_time: 0.0016 memory: 28783 grad_norm: 0.0304 loss: 0.6877 +2023/05/31 21:40:24 - mmengine - INFO - Epoch(train) [3][1100/5758] lr: 9.7577e-04 eta: 1 day, 1:14:51 time: 0.8727 data_time: 0.0024 memory: 28783 grad_norm: 0.0321 loss: 0.6809 +2023/05/31 21:41:52 - mmengine - INFO - Epoch(train) [3][1200/5758] lr: 9.7577e-04 eta: 1 day, 1:13:13 time: 0.9144 data_time: 0.0026 memory: 28783 grad_norm: 0.0399 loss: 0.6894 +2023/05/31 21:43:17 - mmengine - INFO - Epoch(train) [3][1300/5758] lr: 9.7577e-04 eta: 1 day, 1:11:22 time: 0.8784 data_time: 0.0019 memory: 28783 grad_norm: 0.0309 loss: 0.6871 +2023/05/31 21:44:43 - mmengine - INFO - Epoch(train) [3][1400/5758] lr: 9.7577e-04 eta: 1 day, 1:09:30 time: 0.8042 data_time: 0.0024 memory: 28783 grad_norm: 0.0253 loss: 0.6866 +2023/05/31 21:45:53 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_20230531_183239 +2023/05/31 21:46:08 - mmengine - INFO - Epoch(train) [3][1500/5758] lr: 9.7577e-04 eta: 1 day, 1:07:34 time: 0.8169 data_time: 0.0020 memory: 28783 grad_norm: 0.0331 loss: 0.6878 +2023/05/31 21:47:31 - mmengine - INFO - Epoch(train) [3][1600/5758] lr: 9.7577e-04 eta: 1 day, 1:05:24 time: 0.8307 data_time: 0.0018 memory: 28783 grad_norm: 0.0192 loss: 0.6881 +2023/05/31 21:48:57 - mmengine - INFO - Epoch(train) [3][1700/5758] lr: 9.7577e-04 eta: 1 day, 1:03:33 time: 0.8646 data_time: 0.0018 memory: 28783 grad_norm: 0.0232 loss: 0.6882 +2023/05/31 21:50:25 - mmengine - INFO - Epoch(train) [3][1800/5758] lr: 9.7577e-04 eta: 1 day, 1:02:01 time: 0.9349 data_time: 0.0022 memory: 28783 grad_norm: 0.0356 loss: 0.6882 +2023/05/31 21:51:49 - mmengine - INFO - Epoch(train) [3][1900/5758] lr: 9.7577e-04 eta: 1 day, 1:00:00 time: 0.8742 data_time: 0.0027 memory: 28783 grad_norm: 0.0397 loss: 0.6897 +2023/05/31 21:53:16 - mmengine - INFO - Epoch(train) [3][2000/5758] lr: 9.7577e-04 eta: 1 day, 0:58:19 time: 0.9291 data_time: 0.0025 memory: 28783 grad_norm: 0.0339 loss: 0.6882 +2023/05/31 21:54:43 - mmengine - INFO - Epoch(train) [3][2100/5758] lr: 9.7577e-04 eta: 1 day, 0:56:40 time: 0.9029 data_time: 0.0025 memory: 28783 grad_norm: 0.0360 loss: 0.6903 +2023/05/31 21:56:12 - mmengine - INFO - Epoch(train) [3][2200/5758] lr: 9.7577e-04 eta: 1 day, 0:55:18 time: 0.8466 data_time: 0.0034 memory: 28783 grad_norm: 0.0267 loss: 0.6836 +2023/05/31 21:57:39 - mmengine - INFO - Epoch(train) [3][2300/5758] lr: 9.7577e-04 eta: 1 day, 0:53:36 time: 0.8440 data_time: 0.0024 memory: 28783 grad_norm: 0.0195 loss: 0.6842 +2023/05/31 21:59:07 - mmengine - INFO - Epoch(train) [3][2400/5758] lr: 9.7577e-04 eta: 1 day, 0:52:05 time: 0.8161 data_time: 0.0016 memory: 28783 grad_norm: 0.0417 loss: 0.6889 +2023/05/31 22:00:18 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_20230531_183239 +2023/05/31 22:00:35 - mmengine - INFO - Epoch(train) [3][2500/5758] lr: 9.7577e-04 eta: 1 day, 0:50:32 time: 0.9336 data_time: 0.0018 memory: 28783 grad_norm: 0.0321 loss: 0.6887 +2023/05/31 22:02:03 - mmengine - INFO - Epoch(train) [3][2600/5758] lr: 9.7577e-04 eta: 1 day, 0:49:05 time: 0.8904 data_time: 0.0025 memory: 28783 grad_norm: 0.0290 loss: 0.6877 +2023/05/31 22:03:38 - mmengine - INFO - Epoch(train) [3][2700/5758] lr: 9.7577e-04 eta: 1 day, 0:48:17 time: 0.8838 data_time: 0.0016 memory: 28783 grad_norm: 0.0283 loss: 0.6899 +2023/05/31 22:05:06 - mmengine - INFO - Epoch(train) [3][2800/5758] lr: 9.7577e-04 eta: 1 day, 0:46:52 time: 0.8419 data_time: 0.0019 memory: 28783 grad_norm: 0.0192 loss: 0.6879 +2023/05/31 22:06:35 - mmengine - INFO - Epoch(train) [3][2900/5758] lr: 9.7577e-04 eta: 1 day, 0:45:25 time: 0.9080 data_time: 0.0017 memory: 28783 grad_norm: 0.0296 loss: 0.6893 +2023/05/31 22:07:59 - mmengine - INFO - Epoch(train) [3][3000/5758] lr: 9.7577e-04 eta: 1 day, 0:43:27 time: 0.8397 data_time: 0.0026 memory: 28783 grad_norm: 0.0297 loss: 0.6883 +2023/05/31 22:09:20 - mmengine - INFO - Epoch(train) [3][3100/5758] lr: 9.7577e-04 eta: 1 day, 0:41:02 time: 0.8690 data_time: 0.0025 memory: 28783 grad_norm: 0.0197 loss: 0.6836 +2023/05/31 22:10:42 - mmengine - INFO - Epoch(train) [3][3200/5758] lr: 9.7577e-04 eta: 1 day, 0:38:51 time: 0.7857 data_time: 0.0024 memory: 28783 grad_norm: 0.0257 loss: 0.6879 +2023/05/31 22:12:06 - mmengine - INFO - Epoch(train) [3][3300/5758] lr: 9.7577e-04 eta: 1 day, 0:36:57 time: 0.8270 data_time: 0.0020 memory: 28783 grad_norm: 0.0334 loss: 0.6883 +2023/05/31 22:13:27 - mmengine - INFO - Epoch(train) [3][3400/5758] lr: 9.7577e-04 eta: 1 day, 0:34:35 time: 0.7631 data_time: 0.0019 memory: 28783 grad_norm: 0.0193 loss: 0.6824 +2023/05/31 22:14:33 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_20230531_183239 +2023/05/31 22:14:47 - mmengine - INFO - Epoch(train) [3][3500/5758] lr: 9.7577e-04 eta: 1 day, 0:32:16 time: 0.8477 data_time: 0.0026 memory: 28783 grad_norm: 0.0233 loss: 0.6884 +2023/05/31 22:16:08 - mmengine - INFO - Epoch(train) [3][3600/5758] lr: 9.7577e-04 eta: 1 day, 0:30:01 time: 0.8013 data_time: 0.0018 memory: 28783 grad_norm: 0.0217 loss: 0.6863 +2023/05/31 22:17:29 - mmengine - INFO - Epoch(train) [3][3700/5758] lr: 9.7577e-04 eta: 1 day, 0:27:44 time: 0.8590 data_time: 0.0024 memory: 28783 grad_norm: 0.0247 loss: 0.6866 +2023/05/31 22:18:49 - mmengine - INFO - Epoch(train) [3][3800/5758] lr: 9.7577e-04 eta: 1 day, 0:25:19 time: 0.7917 data_time: 0.0023 memory: 28783 grad_norm: 0.0179 loss: 0.6886 +2023/05/31 22:20:11 - mmengine - INFO - Epoch(train) [3][3900/5758] lr: 9.7577e-04 eta: 1 day, 0:23:12 time: 0.7782 data_time: 0.0017 memory: 28783 grad_norm: 0.0256 loss: 0.6898 +2023/05/31 22:21:35 - mmengine - INFO - Epoch(train) [3][4000/5758] lr: 9.7577e-04 eta: 1 day, 0:21:20 time: 0.8618 data_time: 0.0025 memory: 28783 grad_norm: 0.0183 loss: 0.6882 +2023/05/31 22:22:57 - mmengine - INFO - Epoch(train) [3][4100/5758] lr: 9.7577e-04 eta: 1 day, 0:19:12 time: 0.7991 data_time: 0.0024 memory: 28783 grad_norm: 0.0243 loss: 0.6846 +2023/05/31 22:24:20 - mmengine - INFO - Epoch(train) [3][4200/5758] lr: 9.7577e-04 eta: 1 day, 0:17:14 time: 0.8461 data_time: 0.0024 memory: 28783 grad_norm: 0.0179 loss: 0.6848 +2023/05/31 22:25:42 - mmengine - INFO - Epoch(train) [3][4300/5758] lr: 9.7577e-04 eta: 1 day, 0:15:07 time: 0.8860 data_time: 0.0017 memory: 28783 grad_norm: 0.0269 loss: 0.6859 +2023/05/31 22:27:03 - mmengine - INFO - Epoch(train) [3][4400/5758] lr: 9.7577e-04 eta: 1 day, 0:12:58 time: 0.7866 data_time: 0.0023 memory: 28783 grad_norm: 0.0204 loss: 0.6883 +2023/05/31 22:28:09 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_20230531_183239 +2023/05/31 22:28:24 - mmengine - INFO - Epoch(train) [3][4500/5758] lr: 9.7577e-04 eta: 1 day, 0:10:48 time: 0.8771 data_time: 0.0026 memory: 28783 grad_norm: 0.0239 loss: 0.6896 +2023/05/31 22:29:50 - mmengine - INFO - Epoch(train) [3][4600/5758] lr: 9.7577e-04 eta: 1 day, 0:09:10 time: 0.8408 data_time: 0.0023 memory: 28783 grad_norm: 0.0427 loss: 0.6882 +2023/05/31 22:31:11 - mmengine - INFO - Epoch(train) [3][4700/5758] lr: 9.7577e-04 eta: 1 day, 0:07:03 time: 0.8630 data_time: 0.0029 memory: 28783 grad_norm: 0.0408 loss: 0.6856 +2023/05/31 22:32:35 - mmengine - INFO - Epoch(train) [3][4800/5758] lr: 9.7577e-04 eta: 1 day, 0:05:10 time: 0.7796 data_time: 0.0020 memory: 28783 grad_norm: 0.0396 loss: 0.6861 +2023/05/31 22:33:56 - mmengine - INFO - Epoch(train) [3][4900/5758] lr: 9.7577e-04 eta: 1 day, 0:03:03 time: 0.8268 data_time: 0.0019 memory: 28783 grad_norm: 0.0333 loss: 0.6891 +2023/05/31 22:35:18 - mmengine - INFO - Epoch(train) [3][5000/5758] lr: 9.7577e-04 eta: 1 day, 0:01:00 time: 0.7782 data_time: 0.0022 memory: 28783 grad_norm: 0.0272 loss: 0.6853 +2023/05/31 22:36:39 - mmengine - INFO - Epoch(train) [3][5100/5758] lr: 9.7577e-04 eta: 23:58:56 time: 0.7931 data_time: 0.0024 memory: 28783 grad_norm: 0.0202 loss: 0.6882 +2023/05/31 22:38:03 - mmengine - INFO - Epoch(train) [3][5200/5758] lr: 9.7577e-04 eta: 23:57:02 time: 0.9098 data_time: 0.0024 memory: 28783 grad_norm: 0.0230 loss: 0.6872 +2023/05/31 22:39:25 - mmengine - INFO - Epoch(train) [3][5300/5758] lr: 9.7577e-04 eta: 23:55:05 time: 0.8082 data_time: 0.0026 memory: 28783 grad_norm: 0.0261 loss: 0.6864 +2023/05/31 22:40:46 - mmengine - INFO - Epoch(train) [3][5400/5758] lr: 9.7577e-04 eta: 23:52:57 time: 0.8320 data_time: 0.0021 memory: 28783 grad_norm: 0.0226 loss: 0.6847 +2023/05/31 22:41:53 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_20230531_183239 +2023/05/31 22:42:07 - mmengine - INFO - Epoch(train) [3][5500/5758] lr: 9.7577e-04 eta: 23:50:56 time: 0.8610 data_time: 0.0025 memory: 28783 grad_norm: 0.0312 loss: 0.6863 +2023/05/31 22:43:32 - mmengine - INFO - Epoch(train) [3][5600/5758] lr: 9.7577e-04 eta: 23:49:11 time: 0.8509 data_time: 0.0021 memory: 28783 grad_norm: 0.0319 loss: 0.6834 +2023/05/31 22:44:55 - mmengine - INFO - Epoch(train) [3][5700/5758] lr: 9.7577e-04 eta: 23:47:19 time: 0.7737 data_time: 0.0022 memory: 28783 grad_norm: 0.0327 loss: 0.6867 +2023/05/31 22:45:42 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_20230531_183239 +2023/05/31 22:45:42 - mmengine - INFO - Saving checkpoint at 3 epochs +2023/05/31 22:46:01 - mmengine - INFO - Epoch(val) [3][16/16] accuracy/top1: 100.0000 single-label/precision_classwise: [100.0, 0.0] single-label/recall_classwise: [100.0, 0.0] single-label/f1-score_classwise: [100.0, 0.0] data_time: 0.2414 time: 0.6570 +2023/05/31 22:47:30 - mmengine - INFO - Epoch(train) [4][ 100/5758] lr: 9.4605e-04 eta: 23:44:49 time: 0.8160 data_time: 0.2718 memory: 28783 grad_norm: 0.0252 loss: 0.6857 +2023/05/31 22:48:51 - mmengine - INFO - Epoch(train) [4][ 200/5758] lr: 9.4605e-04 eta: 23:42:46 time: 0.7802 data_time: 0.1557 memory: 28783 grad_norm: 0.0300 loss: 0.6872 +2023/05/31 22:50:14 - mmengine - INFO - Epoch(train) [4][ 300/5758] lr: 9.4605e-04 eta: 23:40:56 time: 0.7861 data_time: 0.0017 memory: 28783 grad_norm: 0.0224 loss: 0.6870 +2023/05/31 22:51:37 - mmengine - INFO - Epoch(train) [4][ 400/5758] lr: 9.4605e-04 eta: 23:39:03 time: 0.7784 data_time: 0.0022 memory: 28783 grad_norm: 0.0265 loss: 0.6866 +2023/05/31 22:53:18 - mmengine - INFO - Epoch(train) [4][ 500/5758] lr: 9.4605e-04 eta: 23:38:53 time: 0.6624 data_time: 0.0022 memory: 28783 grad_norm: 0.0153 loss: 0.6857 +2023/05/31 22:54:28 - mmengine - INFO - Epoch(train) [4][ 600/5758] lr: 9.4605e-04 eta: 23:35:51 time: 0.7903 data_time: 0.0025 memory: 28783 grad_norm: 0.0361 loss: 0.6892 +2023/05/31 22:55:49 - mmengine - INFO - Epoch(train) [4][ 700/5758] lr: 9.4605e-04 eta: 23:33:48 time: 0.8287 data_time: 0.0016 memory: 28783 grad_norm: 0.0207 loss: 0.6884 +2023/05/31 22:56:08 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_20230531_183239 +2023/05/31 22:57:09 - mmengine - INFO - Epoch(train) [4][ 800/5758] lr: 9.4605e-04 eta: 23:31:41 time: 0.8033 data_time: 0.0016 memory: 28783 grad_norm: 0.0297 loss: 0.6866 +2023/05/31 22:58:27 - mmengine - INFO - Epoch(train) [4][ 900/5758] lr: 9.4605e-04 eta: 23:29:24 time: 0.7449 data_time: 0.0023 memory: 28783 grad_norm: 0.0168 loss: 0.6898 +2023/05/31 22:59:49 - mmengine - INFO - Epoch(train) [4][1000/5758] lr: 9.4605e-04 eta: 23:27:34 time: 0.8102 data_time: 0.0019 memory: 28783 grad_norm: 0.0238 loss: 0.6864 +2023/05/31 23:01:14 - mmengine - INFO - Epoch(train) [4][1100/5758] lr: 9.4605e-04 eta: 23:25:50 time: 0.8246 data_time: 0.0030 memory: 28783 grad_norm: 0.0277 loss: 0.6844 +2023/05/31 23:02:34 - mmengine - INFO - Epoch(train) [4][1200/5758] lr: 9.4605e-04 eta: 23:23:50 time: 0.8021 data_time: 0.0023 memory: 28783 grad_norm: 0.0136 loss: 0.6841 +2023/05/31 23:03:57 - mmengine - INFO - Epoch(train) [4][1300/5758] lr: 9.4605e-04 eta: 23:21:59 time: 0.8011 data_time: 0.0021 memory: 28783 grad_norm: 0.0222 loss: 0.6860 +2023/05/31 23:05:20 - mmengine - INFO - Epoch(train) [4][1400/5758] lr: 9.4605e-04 eta: 23:20:13 time: 0.8457 data_time: 0.0019 memory: 28783 grad_norm: 0.0223 loss: 0.6874 +2023/05/31 23:06:44 - mmengine - INFO - Epoch(train) [4][1500/5758] lr: 9.4605e-04 eta: 23:18:30 time: 0.8185 data_time: 0.0018 memory: 28783 grad_norm: 0.0196 loss: 0.6865 +2023/05/31 23:08:06 - mmengine - INFO - Epoch(train) [4][1600/5758] lr: 9.4605e-04 eta: 23:16:38 time: 0.8482 data_time: 0.0020 memory: 28783 grad_norm: 0.0163 loss: 0.6878 +2023/05/31 23:09:27 - mmengine - INFO - Epoch(train) [4][1700/5758] lr: 9.4605e-04 eta: 23:14:37 time: 0.7707 data_time: 0.0025 memory: 28783 grad_norm: 0.0154 loss: 0.6862 +2023/05/31 23:09:45 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_20230531_183239 +2023/05/31 23:10:46 - mmengine - INFO - Epoch(train) [4][1800/5758] lr: 9.4605e-04 eta: 23:12:32 time: 0.7614 data_time: 0.0017 memory: 28783 grad_norm: 0.0256 loss: 0.6847 +2023/05/31 23:12:07 - mmengine - INFO - Epoch(train) [4][1900/5758] lr: 9.4605e-04 eta: 23:10:35 time: 0.7657 data_time: 0.0017 memory: 28783 grad_norm: 0.0161 loss: 0.6865 +2023/05/31 23:13:30 - mmengine - INFO - Epoch(train) [4][2000/5758] lr: 9.4605e-04 eta: 23:08:46 time: 0.8767 data_time: 0.0023 memory: 28783 grad_norm: 0.0223 loss: 0.6868 +2023/05/31 23:14:50 - mmengine - INFO - Epoch(train) [4][2100/5758] lr: 9.4605e-04 eta: 23:06:49 time: 0.8380 data_time: 0.0024 memory: 28783 grad_norm: 0.0159 loss: 0.6905 +2023/05/31 23:16:13 - mmengine - INFO - Epoch(train) [4][2200/5758] lr: 9.4605e-04 eta: 23:05:00 time: 0.9253 data_time: 0.0019 memory: 28783 grad_norm: 0.0230 loss: 0.6866 +2023/05/31 23:17:36 - mmengine - INFO - Epoch(train) [4][2300/5758] lr: 9.4605e-04 eta: 23:03:13 time: 0.9055 data_time: 0.0018 memory: 28783 grad_norm: 0.0209 loss: 0.6846 +2023/05/31 23:18:58 - mmengine - INFO - Epoch(train) [4][2400/5758] lr: 9.4605e-04 eta: 23:01:24 time: 0.7952 data_time: 0.0022 memory: 28783 grad_norm: 0.0186 loss: 0.6884 +2023/05/31 23:20:19 - mmengine - INFO - Epoch(train) [4][2500/5758] lr: 9.4605e-04 eta: 22:59:31 time: 0.8281 data_time: 0.0019 memory: 28783 grad_norm: 0.0171 loss: 0.6888 +2023/05/31 23:21:43 - mmengine - INFO - Epoch(train) [4][2600/5758] lr: 9.4605e-04 eta: 22:57:49 time: 0.9085 data_time: 0.0026 memory: 28783 grad_norm: 0.0171 loss: 0.6890 +2023/05/31 23:23:05 - mmengine - INFO - Epoch(train) [4][2700/5758] lr: 9.4605e-04 eta: 22:56:01 time: 0.8443 data_time: 0.0017 memory: 28783 grad_norm: 0.0228 loss: 0.6880 +2023/05/31 23:23:25 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_20230531_183239 +2023/05/31 23:24:28 - mmengine - INFO - Epoch(train) [4][2800/5758] lr: 9.4605e-04 eta: 22:54:13 time: 0.7941 data_time: 0.0018 memory: 28783 grad_norm: 0.0155 loss: 0.6910 +2023/05/31 23:25:47 - mmengine - INFO - Epoch(train) [4][2900/5758] lr: 9.4605e-04 eta: 22:52:13 time: 0.7402 data_time: 0.0016 memory: 28783 grad_norm: 0.0112 loss: 0.6877 +2023/05/31 23:27:09 - mmengine - INFO - Epoch(train) [4][3000/5758] lr: 9.4605e-04 eta: 22:50:22 time: 0.7656 data_time: 0.0018 memory: 28783 grad_norm: 0.0256 loss: 0.6874 +2023/05/31 23:28:31 - mmengine - INFO - Epoch(train) [4][3100/5758] lr: 9.4605e-04 eta: 22:48:36 time: 0.8255 data_time: 0.0015 memory: 28783 grad_norm: 0.0152 loss: 0.6888 +2023/05/31 23:29:54 - mmengine - INFO - Epoch(train) [4][3200/5758] lr: 9.4605e-04 eta: 22:46:51 time: 0.8347 data_time: 0.0018 memory: 28783 grad_norm: 0.0256 loss: 0.6851 +2023/05/31 23:31:17 - mmengine - INFO - Epoch(train) [4][3300/5758] lr: 9.4605e-04 eta: 22:45:09 time: 0.8248 data_time: 0.0018 memory: 28783 grad_norm: 0.0158 loss: 0.6898 +2023/05/31 23:32:39 - mmengine - INFO - Epoch(train) [4][3400/5758] lr: 9.4605e-04 eta: 22:43:19 time: 0.7931 data_time: 0.0020 memory: 28783 grad_norm: 0.0210 loss: 0.6828 +2023/05/31 23:34:00 - mmengine - INFO - Epoch(train) [4][3500/5758] lr: 9.4605e-04 eta: 22:41:31 time: 0.8832 data_time: 0.0021 memory: 28783 grad_norm: 0.0189 loss: 0.6877 +2023/05/31 23:35:22 - mmengine - INFO - Epoch(train) [4][3600/5758] lr: 9.4605e-04 eta: 22:39:41 time: 0.8106 data_time: 0.0019 memory: 28783 grad_norm: 0.0212 loss: 0.6866 +2023/05/31 23:36:43 - mmengine - INFO - Epoch(train) [4][3700/5758] lr: 9.4605e-04 eta: 22:37:52 time: 0.7847 data_time: 0.0020 memory: 28783 grad_norm: 0.0163 loss: 0.6873 +2023/05/31 23:37:03 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_20230531_183239 +2023/05/31 23:38:05 - mmengine - INFO - Epoch(train) [4][3800/5758] lr: 9.4605e-04 eta: 22:36:05 time: 0.8509 data_time: 0.0017 memory: 28783 grad_norm: 0.0201 loss: 0.6868 +2023/05/31 23:39:26 - mmengine - INFO - Epoch(train) [4][3900/5758] lr: 9.4605e-04 eta: 22:34:15 time: 0.8274 data_time: 0.0018 memory: 28783 grad_norm: 0.0246 loss: 0.6874 +2023/05/31 23:40:48 - mmengine - INFO - Epoch(train) [4][4000/5758] lr: 9.4605e-04 eta: 22:32:29 time: 0.7816 data_time: 0.0019 memory: 28783 grad_norm: 0.0176 loss: 0.6860 +2023/05/31 23:42:10 - mmengine - INFO - Epoch(train) [4][4100/5758] lr: 9.4605e-04 eta: 22:30:41 time: 0.8441 data_time: 0.0019 memory: 28783 grad_norm: 0.0150 loss: 0.6863 +2023/05/31 23:43:29 - mmengine - INFO - Epoch(train) [4][4200/5758] lr: 9.4605e-04 eta: 22:28:43 time: 0.8000 data_time: 0.0017 memory: 28783 grad_norm: 0.0163 loss: 0.6894 +2023/05/31 23:44:54 - mmengine - INFO - Epoch(train) [4][4300/5758] lr: 9.4605e-04 eta: 22:27:09 time: 0.8927 data_time: 0.0019 memory: 28783 grad_norm: 0.0176 loss: 0.6894 +2023/05/31 23:46:16 - mmengine - INFO - Epoch(train) [4][4400/5758] lr: 9.4605e-04 eta: 22:25:24 time: 0.8131 data_time: 0.0526 memory: 28783 grad_norm: 0.0240 loss: 0.6868 +2023/05/31 23:47:35 - mmengine - INFO - Epoch(train) [4][4500/5758] lr: 9.4605e-04 eta: 22:23:28 time: 0.7483 data_time: 0.1344 memory: 28783 grad_norm: 0.0087 loss: 0.6896 +2023/05/31 23:48:57 - mmengine - INFO - Epoch(train) [4][4600/5758] lr: 9.4605e-04 eta: 22:21:43 time: 0.8646 data_time: 0.3608 memory: 28783 grad_norm: 0.0194 loss: 0.6859 +2023/05/31 23:50:20 - mmengine - INFO - Epoch(train) [4][4700/5758] lr: 9.4605e-04 eta: 22:20:01 time: 0.8043 data_time: 0.2884 memory: 28783 grad_norm: 0.0161 loss: 0.6821 +2023/05/31 23:50:41 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_20230531_183239 +2023/05/31 23:51:39 - mmengine - INFO - Epoch(train) [4][4800/5758] lr: 9.4605e-04 eta: 22:18:05 time: 0.8259 data_time: 0.3223 memory: 28783 grad_norm: 0.0155 loss: 0.6854 +2023/05/31 23:53:01 - mmengine - INFO - Epoch(train) [4][4900/5758] lr: 9.4605e-04 eta: 22:16:21 time: 0.8290 data_time: 0.3203 memory: 28783 grad_norm: 0.0167 loss: 0.6851 +2023/05/31 23:54:25 - mmengine - INFO - Epoch(train) [4][5000/5758] lr: 9.4605e-04 eta: 22:14:43 time: 0.8130 data_time: 0.3064 memory: 28783 grad_norm: 0.0142 loss: 0.6861 +2023/05/31 23:55:46 - mmengine - INFO - Epoch(train) [4][5100/5758] lr: 9.4605e-04 eta: 22:12:58 time: 0.8231 data_time: 0.3194 memory: 28783 grad_norm: 0.0113 loss: 0.6856 +2023/05/31 23:57:07 - mmengine - INFO - Epoch(train) [4][5200/5758] lr: 9.4605e-04 eta: 22:11:09 time: 0.8611 data_time: 0.3567 memory: 28783 grad_norm: 0.0215 loss: 0.6860 +2023/05/31 23:58:30 - mmengine - INFO - Epoch(train) [4][5300/5758] lr: 9.4605e-04 eta: 22:09:29 time: 0.8114 data_time: 0.3067 memory: 28783 grad_norm: 0.0220 loss: 0.6880 +2023/05/31 23:59:47 - mmengine - INFO - Epoch(train) [4][5400/5758] lr: 9.4605e-04 eta: 22:07:26 time: 0.7842 data_time: 0.2771 memory: 28783 grad_norm: 0.0132 loss: 0.6867 +2023/06/01 00:01:04 - mmengine - INFO - Epoch(train) [4][5500/5758] lr: 9.4605e-04 eta: 22:05:23 time: 0.7771 data_time: 0.1559 memory: 28783 grad_norm: 0.0191 loss: 0.6883 +2023/06/01 00:02:25 - mmengine - INFO - Epoch(train) [4][5600/5758] lr: 9.4605e-04 eta: 22:03:38 time: 0.8027 data_time: 0.0017 memory: 28783 grad_norm: 0.0151 loss: 0.6915 +2023/06/01 00:03:43 - mmengine - INFO - Epoch(train) [4][5700/5758] lr: 9.4605e-04 eta: 22:01:39 time: 0.7684 data_time: 0.0018 memory: 28783 grad_norm: 0.0200 loss: 0.6877 +2023/06/01 00:04:02 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_20230531_183239 +2023/06/01 00:04:29 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_20230531_183239 +2023/06/01 00:04:29 - mmengine - INFO - Saving checkpoint at 4 epochs +2023/06/01 00:04:47 - mmengine - INFO - Epoch(val) [4][16/16] accuracy/top1: 100.0000 single-label/precision_classwise: [100.0, 0.0] single-label/recall_classwise: [100.0, 0.0] single-label/f1-score_classwise: [100.0, 0.0] data_time: 0.2107 time: 0.6256 +2023/06/01 00:06:09 - mmengine - INFO - Epoch(train) [5][ 100/5758] lr: 9.0546e-04 eta: 21:58:53 time: 0.7441 data_time: 0.0023 memory: 28783 grad_norm: 0.0198 loss: 0.6858 +2023/06/01 00:07:29 - mmengine - INFO - Epoch(train) [5][ 200/5758] lr: 9.0546e-04 eta: 21:57:02 time: 0.8215 data_time: 0.0016 memory: 28783 grad_norm: 0.0162 loss: 0.6885 +2023/06/01 00:08:51 - mmengine - INFO - Epoch(train) [5][ 300/5758] lr: 9.0546e-04 eta: 21:55:21 time: 0.7937 data_time: 0.0018 memory: 28783 grad_norm: 0.0154 loss: 0.6878 +2023/06/01 00:10:09 - mmengine - INFO - Epoch(train) [5][ 400/5758] lr: 9.0546e-04 eta: 21:53:27 time: 0.7927 data_time: 0.0019 memory: 28783 grad_norm: 0.0160 loss: 0.6848 +2023/06/01 00:11:31 - mmengine - INFO - Epoch(train) [5][ 500/5758] lr: 9.0546e-04 eta: 21:51:43 time: 0.8547 data_time: 0.0017 memory: 28783 grad_norm: 0.0165 loss: 0.6901 +2023/06/01 00:12:50 - mmengine - INFO - Epoch(train) [5][ 600/5758] lr: 9.0546e-04 eta: 21:49:53 time: 0.8249 data_time: 0.0015 memory: 28783 grad_norm: 0.0198 loss: 0.6886 +2023/06/01 00:14:10 - mmengine - INFO - Epoch(train) [5][ 700/5758] lr: 9.0546e-04 eta: 21:48:02 time: 0.7773 data_time: 0.0023 memory: 28783 grad_norm: 0.0198 loss: 0.6862 +2023/06/01 00:15:31 - mmengine - INFO - Epoch(train) [5][ 800/5758] lr: 9.0546e-04 eta: 21:46:17 time: 0.8398 data_time: 0.0018 memory: 28783 grad_norm: 0.0184 loss: 0.6863 +2023/06/01 00:16:51 - mmengine - INFO - Epoch(train) [5][ 900/5758] lr: 9.0546e-04 eta: 21:44:29 time: 0.8179 data_time: 0.0020 memory: 28783 grad_norm: 0.0244 loss: 0.6902 +2023/06/01 00:17:44 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_20230531_183239 +2023/06/01 00:18:09 - mmengine - INFO - Epoch(train) [5][1000/5758] lr: 9.0546e-04 eta: 21:42:33 time: 0.7620 data_time: 0.0017 memory: 28783 grad_norm: 0.0196 loss: 0.6887 +2023/06/01 00:19:29 - mmengine - INFO - Epoch(train) [5][1100/5758] lr: 9.0546e-04 eta: 21:40:48 time: 0.7844 data_time: 0.0023 memory: 28783 grad_norm: 0.0163 loss: 0.6862 +2023/06/01 00:20:48 - mmengine - INFO - Epoch(train) [5][1200/5758] lr: 9.0546e-04 eta: 21:38:58 time: 0.8334 data_time: 0.0028 memory: 28783 grad_norm: 0.0137 loss: 0.6811 +2023/06/01 00:22:07 - mmengine - INFO - Epoch(train) [5][1300/5758] lr: 9.0546e-04 eta: 21:37:06 time: 0.8194 data_time: 0.0025 memory: 28783 grad_norm: 0.0219 loss: 0.6858 +2023/06/01 00:23:27 - mmengine - INFO - Epoch(train) [5][1400/5758] lr: 9.0546e-04 eta: 21:35:20 time: 0.7778 data_time: 0.0018 memory: 28783 grad_norm: 0.0175 loss: 0.6847 +2023/06/01 00:24:47 - mmengine - INFO - Epoch(train) [5][1500/5758] lr: 9.0546e-04 eta: 21:33:33 time: 0.8111 data_time: 0.0017 memory: 28783 grad_norm: 0.0192 loss: 0.6864 +2023/06/01 00:26:07 - mmengine - INFO - Epoch(train) [5][1600/5758] lr: 9.0546e-04 eta: 21:31:46 time: 0.8756 data_time: 0.0019 memory: 28783 grad_norm: 0.0158 loss: 0.6858 +2023/06/01 00:27:28 - mmengine - INFO - Epoch(train) [5][1700/5758] lr: 9.0546e-04 eta: 21:30:02 time: 0.7886 data_time: 0.0016 memory: 28783 grad_norm: 0.0331 loss: 0.6868 +2023/06/01 00:28:48 - mmengine - INFO - Epoch(train) [5][1800/5758] lr: 9.0546e-04 eta: 21:28:17 time: 0.7610 data_time: 0.0016 memory: 28783 grad_norm: 0.0136 loss: 0.6868 +2023/06/01 00:30:10 - mmengine - INFO - Epoch(train) [5][1900/5758] lr: 9.0546e-04 eta: 21:26:39 time: 0.8194 data_time: 0.0024 memory: 28783 grad_norm: 0.0155 loss: 0.6852 +2023/06/01 00:31:05 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_20230531_183239 +2023/06/01 00:31:30 - mmengine - INFO - Epoch(train) [5][2000/5758] lr: 9.0546e-04 eta: 21:24:55 time: 0.8177 data_time: 0.0028 memory: 28783 grad_norm: 0.0147 loss: 0.6829 +2023/06/01 00:32:51 - mmengine - INFO - Epoch(train) [5][2100/5758] lr: 9.0546e-04 eta: 21:23:13 time: 0.8035 data_time: 0.0020 memory: 28783 grad_norm: 0.0182 loss: 0.6865 +2023/06/01 00:34:11 - mmengine - INFO - Epoch(train) [5][2200/5758] lr: 9.0546e-04 eta: 21:21:27 time: 0.8146 data_time: 0.0026 memory: 28783 grad_norm: 0.0114 loss: 0.6845 +2023/06/01 00:35:31 - mmengine - INFO - Epoch(train) [5][2300/5758] lr: 9.0546e-04 eta: 21:19:42 time: 0.7872 data_time: 0.0023 memory: 28783 grad_norm: 0.0119 loss: 0.6867 +2023/06/01 00:36:52 - mmengine - INFO - Epoch(train) [5][2400/5758] lr: 9.0546e-04 eta: 21:17:59 time: 0.7996 data_time: 0.0018 memory: 28783 grad_norm: 0.0168 loss: 0.6891 +2023/06/01 00:38:14 - mmengine - INFO - Epoch(train) [5][2500/5758] lr: 9.0546e-04 eta: 21:16:23 time: 0.8244 data_time: 0.0016 memory: 28783 grad_norm: 0.0122 loss: 0.6869 +2023/06/01 00:39:36 - mmengine - INFO - Epoch(train) [5][2600/5758] lr: 9.0546e-04 eta: 21:14:45 time: 0.8165 data_time: 0.0020 memory: 28783 grad_norm: 0.0183 loss: 0.6858 +2023/06/01 00:40:55 - mmengine - INFO - Epoch(train) [5][2700/5758] lr: 9.0546e-04 eta: 21:12:57 time: 0.7558 data_time: 0.0026 memory: 28783 grad_norm: 0.0152 loss: 0.6887 +2023/06/01 00:42:15 - mmengine - INFO - Epoch(train) [5][2800/5758] lr: 9.0546e-04 eta: 21:11:13 time: 0.7817 data_time: 0.0024 memory: 28783 grad_norm: 0.0258 loss: 0.6867 +2023/06/01 00:43:34 - mmengine - INFO - Epoch(train) [5][2900/5758] lr: 9.0546e-04 eta: 21:09:25 time: 0.8854 data_time: 0.0018 memory: 28783 grad_norm: 0.0186 loss: 0.6892 +2023/06/01 00:44:29 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_20230531_183239 +2023/06/01 00:44:54 - mmengine - INFO - Epoch(train) [5][3000/5758] lr: 9.0546e-04 eta: 21:07:40 time: 0.8518 data_time: 0.0017 memory: 28783 grad_norm: 0.0162 loss: 0.6867 +2023/06/01 00:46:13 - mmengine - INFO - Epoch(train) [5][3100/5758] lr: 9.0546e-04 eta: 21:05:56 time: 0.7822 data_time: 0.0024 memory: 28783 grad_norm: 0.0178 loss: 0.6883 +2023/06/01 00:47:34 - mmengine - INFO - Epoch(train) [5][3200/5758] lr: 9.0546e-04 eta: 21:04:13 time: 0.7932 data_time: 0.0025 memory: 28783 grad_norm: 0.0174 loss: 0.6872 +2023/06/01 00:48:54 - mmengine - INFO - Epoch(train) [5][3300/5758] lr: 9.0546e-04 eta: 21:02:32 time: 0.8034 data_time: 0.0031 memory: 28783 grad_norm: 0.0217 loss: 0.6892 +2023/06/01 00:50:14 - mmengine - INFO - Epoch(train) [5][3400/5758] lr: 9.0546e-04 eta: 21:00:48 time: 0.8362 data_time: 0.0025 memory: 28783 grad_norm: 0.0159 loss: 0.6844 +2023/06/01 00:51:35 - mmengine - INFO - Epoch(train) [5][3500/5758] lr: 9.0546e-04 eta: 20:59:09 time: 0.8367 data_time: 0.0019 memory: 28783 grad_norm: 0.0183 loss: 0.6848 +2023/06/01 00:52:54 - mmengine - INFO - Epoch(train) [5][3600/5758] lr: 9.0546e-04 eta: 20:57:24 time: 0.7896 data_time: 0.0019 memory: 28783 grad_norm: 0.0173 loss: 0.6856 +2023/06/01 00:54:14 - mmengine - INFO - Epoch(train) [5][3700/5758] lr: 9.0546e-04 eta: 20:55:42 time: 0.8051 data_time: 0.0018 memory: 28783 grad_norm: 0.0284 loss: 0.6843 +2023/06/01 00:55:36 - mmengine - INFO - Epoch(train) [5][3800/5758] lr: 9.0546e-04 eta: 20:54:05 time: 0.8402 data_time: 0.0025 memory: 28783 grad_norm: 0.0120 loss: 0.6865 +2023/06/01 00:56:57 - mmengine - INFO - Epoch(train) [5][3900/5758] lr: 9.0546e-04 eta: 20:52:26 time: 0.8217 data_time: 0.0018 memory: 28783 grad_norm: 0.0215 loss: 0.6900 +2023/06/01 00:57:52 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_20230531_183239 +2023/06/01 00:58:17 - mmengine - INFO - Epoch(train) [5][4000/5758] lr: 9.0546e-04 eta: 20:50:42 time: 0.7893 data_time: 0.0019 memory: 28783 grad_norm: 0.0199 loss: 0.6904 +2023/06/01 00:59:36 - mmengine - INFO - Epoch(train) [5][4100/5758] lr: 9.0546e-04 eta: 20:48:57 time: 0.7667 data_time: 0.0025 memory: 28783 grad_norm: 0.0220 loss: 0.6863 +2023/06/01 01:00:55 - mmengine - INFO - Epoch(train) [5][4200/5758] lr: 9.0546e-04 eta: 20:47:14 time: 0.8114 data_time: 0.0018 memory: 28783 grad_norm: 0.0095 loss: 0.6875 +2023/06/01 01:02:14 - mmengine - INFO - Epoch(train) [5][4300/5758] lr: 9.0546e-04 eta: 20:45:28 time: 0.7757 data_time: 0.0022 memory: 28783 grad_norm: 0.0181 loss: 0.6886 +2023/06/01 01:03:33 - mmengine - INFO - Epoch(train) [5][4400/5758] lr: 9.0546e-04 eta: 20:43:45 time: 0.8377 data_time: 0.0018 memory: 28783 grad_norm: 0.0148 loss: 0.6838 +2023/06/01 01:04:53 - mmengine - INFO - Epoch(train) [5][4500/5758] lr: 9.0546e-04 eta: 20:42:03 time: 0.7694 data_time: 0.0017 memory: 28783 grad_norm: 0.0198 loss: 0.6872 +2023/06/01 01:06:12 - mmengine - INFO - Epoch(train) [5][4600/5758] lr: 9.0546e-04 eta: 20:40:19 time: 0.7474 data_time: 0.0016 memory: 28783 grad_norm: 0.0140 loss: 0.6887 +2023/06/01 01:08:35 - mmengine - INFO - Epoch(train) [5][4700/5758] lr: 9.0546e-04 eta: 20:41:58 time: 0.7523 data_time: 0.0020 memory: 28783 grad_norm: 0.0146 loss: 0.6834 +2023/06/01 01:09:54 - mmengine - INFO - Epoch(train) [5][4800/5758] lr: 9.0546e-04 eta: 20:40:12 time: 0.7727 data_time: 0.0019 memory: 28783 grad_norm: 0.0182 loss: 0.6910 +2023/06/01 01:11:11 - mmengine - INFO - Epoch(train) [5][4900/5758] lr: 9.0546e-04 eta: 20:38:21 time: 0.7825 data_time: 0.0016 memory: 28783 grad_norm: 0.0188 loss: 0.6916 +2023/06/01 01:12:03 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_20230531_183239 +2023/06/01 01:12:29 - mmengine - INFO - Epoch(train) [5][5000/5758] lr: 9.0546e-04 eta: 20:36:32 time: 0.7599 data_time: 0.0024 memory: 28783 grad_norm: 0.0197 loss: 0.6834 +2023/06/01 01:13:49 - mmengine - INFO - Epoch(train) [5][5100/5758] lr: 9.0546e-04 eta: 20:34:51 time: 0.8114 data_time: 0.0019 memory: 28783 grad_norm: 0.0180 loss: 0.6863 +2023/06/01 01:15:08 - mmengine - INFO - Epoch(train) [5][5200/5758] lr: 9.0546e-04 eta: 20:33:07 time: 0.8071 data_time: 0.0018 memory: 28783 grad_norm: 0.0129 loss: 0.6861 +2023/06/01 01:16:27 - mmengine - INFO - Epoch(train) [5][5300/5758] lr: 9.0546e-04 eta: 20:31:24 time: 0.8299 data_time: 0.0018 memory: 28783 grad_norm: 0.0132 loss: 0.6844 +2023/06/01 01:17:46 - mmengine - INFO - Epoch(train) [5][5400/5758] lr: 9.0546e-04 eta: 20:29:41 time: 0.8194 data_time: 0.0016 memory: 28783 grad_norm: 0.0123 loss: 0.6874 +2023/06/01 01:19:08 - mmengine - INFO - Epoch(train) [5][5500/5758] lr: 9.0546e-04 eta: 20:28:05 time: 0.8270 data_time: 0.0018 memory: 28783 grad_norm: 0.0110 loss: 0.6894 +2023/06/01 01:20:26 - mmengine - INFO - Epoch(train) [5][5600/5758] lr: 9.0546e-04 eta: 20:26:21 time: 0.8103 data_time: 0.0019 memory: 28783 grad_norm: 0.0144 loss: 0.6906 +2023/06/01 01:21:48 - mmengine - INFO - Epoch(train) [5][5700/5758] lr: 9.0546e-04 eta: 20:24:47 time: 0.8874 data_time: 0.0018 memory: 28783 grad_norm: 0.0112 loss: 0.6875 +2023/06/01 01:22:35 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_20230531_183239 +2023/06/01 01:22:35 - mmengine - INFO - Saving checkpoint at 5 epochs +2023/06/01 01:22:53 - mmengine - INFO - Epoch(val) [5][16/16] accuracy/top1: 100.0000 single-label/precision_classwise: [100.0, 0.0] single-label/recall_classwise: [100.0, 0.0] single-label/f1-score_classwise: [100.0, 0.0] data_time: 0.2131 time: 0.6276 +2023/06/01 01:24:15 - mmengine - INFO - Epoch(train) [6][ 100/5758] lr: 8.5502e-04 eta: 20:22:17 time: 0.7649 data_time: 0.0385 memory: 28783 grad_norm: 0.0163 loss: 0.6850 +2023/06/01 01:25:35 - mmengine - INFO - Epoch(train) [6][ 200/5758] lr: 8.5502e-04 eta: 20:20:37 time: 0.8349 data_time: 0.0018 memory: 28783 grad_norm: 0.0180 loss: 0.6861 +2023/06/01 01:25:44 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_20230531_183239 +2023/06/01 01:26:57 - mmengine - INFO - Epoch(train) [6][ 300/5758] lr: 8.5502e-04 eta: 20:19:02 time: 0.8016 data_time: 0.0019 memory: 28783 grad_norm: 0.0201 loss: 0.6859 +2023/06/01 01:28:17 - mmengine - INFO - Epoch(train) [6][ 400/5758] lr: 8.5502e-04 eta: 20:17:22 time: 0.8147 data_time: 0.0018 memory: 28783 grad_norm: 0.0237 loss: 0.6893 +2023/06/01 01:29:37 - mmengine - INFO - Epoch(train) [6][ 500/5758] lr: 8.5502e-04 eta: 20:15:44 time: 0.7748 data_time: 0.0017 memory: 28783 grad_norm: 0.0167 loss: 0.6863 +2023/06/01 01:30:58 - mmengine - INFO - Epoch(train) [6][ 600/5758] lr: 8.5502e-04 eta: 20:14:07 time: 0.8421 data_time: 0.0017 memory: 28783 grad_norm: 0.0163 loss: 0.6843 +2023/06/01 01:32:16 - mmengine - INFO - Epoch(train) [6][ 700/5758] lr: 8.5502e-04 eta: 20:12:23 time: 0.7396 data_time: 0.0021 memory: 28783 grad_norm: 0.0095 loss: 0.6898 +2023/06/01 01:33:37 - mmengine - INFO - Epoch(train) [6][ 800/5758] lr: 8.5502e-04 eta: 20:10:46 time: 0.7769 data_time: 0.0017 memory: 28783 grad_norm: 0.0237 loss: 0.6884 +2023/06/01 01:34:57 - mmengine - INFO - Epoch(train) [6][ 900/5758] lr: 8.5502e-04 eta: 20:09:06 time: 0.7722 data_time: 0.0020 memory: 28783 grad_norm: 0.0248 loss: 0.6847 +2023/06/01 01:36:18 - mmengine - INFO - Epoch(train) [6][1000/5758] lr: 8.5502e-04 eta: 20:07:31 time: 0.8122 data_time: 0.0016 memory: 28783 grad_norm: 0.0202 loss: 0.6881 +2023/06/01 01:37:39 - mmengine - INFO - Epoch(train) [6][1100/5758] lr: 8.5502e-04 eta: 20:05:54 time: 0.7424 data_time: 0.0019 memory: 28783 grad_norm: 0.0149 loss: 0.6882 +2023/06/01 01:39:00 - mmengine - INFO - Epoch(train) [6][1200/5758] lr: 8.5502e-04 eta: 20:04:19 time: 0.8245 data_time: 0.0019 memory: 28783 grad_norm: 0.0161 loss: 0.6885 +2023/06/01 01:39:08 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_20230531_183239 +2023/06/01 01:40:21 - mmengine - INFO - Epoch(train) [6][1300/5758] lr: 8.5502e-04 eta: 20:02:43 time: 0.8262 data_time: 0.0018 memory: 28783 grad_norm: 0.0172 loss: 0.6886 +2023/06/01 01:41:43 - mmengine - INFO - Epoch(train) [6][1400/5758] lr: 8.5502e-04 eta: 20:01:11 time: 0.8255 data_time: 0.0023 memory: 28783 grad_norm: 0.0138 loss: 0.6885 +2023/06/01 01:43:05 - mmengine - INFO - Epoch(train) [6][1500/5758] lr: 8.5502e-04 eta: 19:59:36 time: 0.8229 data_time: 0.0022 memory: 28783 grad_norm: 0.0161 loss: 0.6848 +2023/06/01 01:44:26 - mmengine - INFO - Epoch(train) [6][1600/5758] lr: 8.5502e-04 eta: 19:58:01 time: 0.8320 data_time: 0.0018 memory: 28783 grad_norm: 0.0124 loss: 0.6858 +2023/06/01 01:45:47 - mmengine - INFO - Epoch(train) [6][1700/5758] lr: 8.5502e-04 eta: 19:56:25 time: 0.8024 data_time: 0.0018 memory: 28783 grad_norm: 0.0197 loss: 0.6861 +2023/06/01 01:47:06 - mmengine - INFO - Epoch(train) [6][1800/5758] lr: 8.5502e-04 eta: 19:54:47 time: 0.7735 data_time: 0.0016 memory: 28783 grad_norm: 0.0202 loss: 0.6902 +2023/06/01 01:48:27 - mmengine - INFO - Epoch(train) [6][1900/5758] lr: 8.5502e-04 eta: 19:53:11 time: 0.8267 data_time: 0.0021 memory: 28783 grad_norm: 0.0135 loss: 0.6865 +2023/06/01 01:49:49 - mmengine - INFO - Epoch(train) [6][2000/5758] lr: 8.5502e-04 eta: 19:51:39 time: 0.7772 data_time: 0.0020 memory: 28783 grad_norm: 0.0149 loss: 0.6875 +2023/06/01 01:51:09 - mmengine - INFO - Epoch(train) [6][2100/5758] lr: 8.5502e-04 eta: 19:50:00 time: 0.8421 data_time: 0.0022 memory: 28783 grad_norm: 0.0151 loss: 0.6888 +2023/06/01 01:52:31 - mmengine - INFO - Epoch(train) [6][2200/5758] lr: 8.5502e-04 eta: 19:48:29 time: 0.7947 data_time: 0.0017 memory: 28783 grad_norm: 0.0147 loss: 0.6870 +2023/06/01 01:52:39 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_20230531_183239 +2023/06/01 01:53:52 - mmengine - INFO - Epoch(train) [6][2300/5758] lr: 8.5502e-04 eta: 19:46:54 time: 0.7981 data_time: 0.0016 memory: 28783 grad_norm: 0.0202 loss: 0.6863 +2023/06/01 01:55:13 - mmengine - INFO - Epoch(train) [6][2400/5758] lr: 8.5502e-04 eta: 19:45:19 time: 0.8718 data_time: 0.0017 memory: 28783 grad_norm: 0.0114 loss: 0.6881 +2023/06/01 01:56:35 - mmengine - INFO - Epoch(train) [6][2500/5758] lr: 8.5502e-04 eta: 19:43:47 time: 0.8049 data_time: 0.0026 memory: 28783 grad_norm: 0.0235 loss: 0.6888 +2023/06/01 01:57:55 - mmengine - INFO - Epoch(train) [6][2600/5758] lr: 8.5502e-04 eta: 19:42:09 time: 0.8178 data_time: 0.0017 memory: 28783 grad_norm: 0.0185 loss: 0.6896 +2023/06/01 01:59:16 - mmengine - INFO - Epoch(train) [6][2700/5758] lr: 8.5502e-04 eta: 19:40:35 time: 0.7765 data_time: 0.0020 memory: 28783 grad_norm: 0.0157 loss: 0.6862 +2023/06/01 02:00:37 - mmengine - INFO - Epoch(train) [6][2800/5758] lr: 8.5502e-04 eta: 19:39:00 time: 0.8254 data_time: 0.0029 memory: 28783 grad_norm: 0.0159 loss: 0.6853 +2023/06/01 02:01:56 - mmengine - INFO - Epoch(train) [6][2900/5758] lr: 8.5502e-04 eta: 19:37:21 time: 0.8082 data_time: 0.0020 memory: 28783 grad_norm: 0.0151 loss: 0.6864 +2023/06/01 02:03:17 - mmengine - INFO - Epoch(train) [6][3000/5758] lr: 8.5502e-04 eta: 19:35:47 time: 0.7971 data_time: 0.0019 memory: 28783 grad_norm: 0.0200 loss: 0.6889 +2023/06/01 02:04:40 - mmengine - INFO - Epoch(train) [6][3100/5758] lr: 8.5502e-04 eta: 19:34:18 time: 0.7880 data_time: 0.0021 memory: 28783 grad_norm: 0.0094 loss: 0.6891 +2023/06/01 02:06:00 - mmengine - INFO - Epoch(train) [6][3200/5758] lr: 8.5502e-04 eta: 19:32:43 time: 0.8410 data_time: 0.0017 memory: 28783 grad_norm: 0.0184 loss: 0.6870 +2023/06/01 02:06:09 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_20230531_183239 +2023/06/01 02:07:23 - mmengine - INFO - Epoch(train) [6][3300/5758] lr: 8.5502e-04 eta: 19:31:14 time: 0.8130 data_time: 0.0020 memory: 28783 grad_norm: 0.0159 loss: 0.6843 +2023/06/01 02:08:42 - mmengine - INFO - Epoch(train) [6][3400/5758] lr: 8.5502e-04 eta: 19:29:35 time: 0.7733 data_time: 0.0017 memory: 28783 grad_norm: 0.0176 loss: 0.6857 +2023/06/01 02:10:04 - mmengine - INFO - Epoch(train) [6][3500/5758] lr: 8.5502e-04 eta: 19:28:02 time: 0.8349 data_time: 0.0027 memory: 28783 grad_norm: 0.0229 loss: 0.6888 +2023/06/01 02:11:24 - mmengine - INFO - Epoch(train) [6][3600/5758] lr: 8.5502e-04 eta: 19:26:26 time: 0.8194 data_time: 0.0018 memory: 28783 grad_norm: 0.0126 loss: 0.6830 +2023/06/01 02:12:44 - mmengine - INFO - Epoch(train) [6][3700/5758] lr: 8.5502e-04 eta: 19:24:50 time: 0.7715 data_time: 0.0023 memory: 28783 grad_norm: 0.0158 loss: 0.6846 +2023/06/01 02:14:05 - mmengine - INFO - Epoch(train) [6][3800/5758] lr: 8.5502e-04 eta: 19:23:17 time: 0.8338 data_time: 0.0019 memory: 28783 grad_norm: 0.0161 loss: 0.6892 +2023/06/01 02:15:27 - mmengine - INFO - Epoch(train) [6][3900/5758] lr: 8.5502e-04 eta: 19:21:45 time: 0.8558 data_time: 0.0019 memory: 28783 grad_norm: 0.0217 loss: 0.6891 +2023/06/01 02:16:47 - mmengine - INFO - Epoch(train) [6][4000/5758] lr: 8.5502e-04 eta: 19:20:10 time: 0.7855 data_time: 0.0017 memory: 28783 grad_norm: 0.0163 loss: 0.6873 +2023/06/01 02:18:09 - mmengine - INFO - Epoch(train) [6][4100/5758] lr: 8.5502e-04 eta: 19:18:40 time: 0.8178 data_time: 0.0028 memory: 28783 grad_norm: 0.0157 loss: 0.6905 +2023/06/01 02:19:32 - mmengine - INFO - Epoch(train) [6][4200/5758] lr: 8.5502e-04 eta: 19:17:10 time: 0.7925 data_time: 0.0022 memory: 28783 grad_norm: 0.0143 loss: 0.6870 +2023/06/01 02:19:40 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_20230531_183239 +2023/06/01 02:20:54 - mmengine - INFO - Epoch(train) [6][4300/5758] lr: 8.5502e-04 eta: 19:15:40 time: 0.8921 data_time: 0.0018 memory: 28783 grad_norm: 0.0133 loss: 0.6889 +2023/06/01 02:22:15 - mmengine - INFO - Epoch(train) [6][4400/5758] lr: 8.5502e-04 eta: 19:14:08 time: 0.8625 data_time: 0.0023 memory: 28783 grad_norm: 0.0185 loss: 0.6841 +2023/06/01 02:23:38 - mmengine - INFO - Epoch(train) [6][4500/5758] lr: 8.5502e-04 eta: 19:12:39 time: 0.7996 data_time: 0.0018 memory: 28783 grad_norm: 0.0157 loss: 0.6855 +2023/06/01 02:25:00 - mmengine - INFO - Epoch(train) [6][4600/5758] lr: 8.5502e-04 eta: 19:11:08 time: 0.8175 data_time: 0.0023 memory: 28783 grad_norm: 0.0181 loss: 0.6882 +2023/06/01 02:26:22 - mmengine - INFO - Epoch(train) [6][4700/5758] lr: 8.5502e-04 eta: 19:09:38 time: 0.8014 data_time: 0.0018 memory: 28783 grad_norm: 0.0139 loss: 0.6850 +2023/06/01 02:27:43 - mmengine - INFO - Epoch(train) [6][4800/5758] lr: 8.5502e-04 eta: 19:08:05 time: 0.7559 data_time: 0.0019 memory: 28783 grad_norm: 0.0145 loss: 0.6857 +2023/06/01 02:29:05 - mmengine - INFO - Epoch(train) [6][4900/5758] lr: 8.5502e-04 eta: 19:06:34 time: 0.8427 data_time: 0.0020 memory: 28783 grad_norm: 0.0384 loss: 0.6882 +2023/06/01 02:30:25 - mmengine - INFO - Epoch(train) [6][5000/5758] lr: 8.5502e-04 eta: 19:05:00 time: 0.8081 data_time: 0.0018 memory: 28783 grad_norm: 0.0334 loss: 0.6900 +2023/06/01 02:31:51 - mmengine - INFO - Epoch(train) [6][5100/5758] lr: 8.5502e-04 eta: 19:03:38 time: 0.8713 data_time: 0.0018 memory: 28783 grad_norm: 0.0335 loss: 0.6870 +2023/06/01 02:33:13 - mmengine - INFO - Epoch(train) [6][5200/5758] lr: 8.5502e-04 eta: 19:02:09 time: 0.8587 data_time: 0.0020 memory: 28783 grad_norm: 0.0239 loss: 0.6834 +2023/06/01 02:33:21 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_20230531_183239 +2023/06/01 02:34:33 - mmengine - INFO - Epoch(train) [6][5300/5758] lr: 8.5502e-04 eta: 19:00:35 time: 0.7655 data_time: 0.0015 memory: 28783 grad_norm: 0.0232 loss: 0.6893 +2023/06/01 02:35:54 - mmengine - INFO - Epoch(train) [6][5400/5758] lr: 8.5502e-04 eta: 18:59:01 time: 0.8653 data_time: 0.0018 memory: 28783 grad_norm: 0.0235 loss: 0.6864 +2023/06/01 02:37:15 - mmengine - INFO - Epoch(train) [6][5500/5758] lr: 8.5502e-04 eta: 18:57:29 time: 0.8010 data_time: 0.0029 memory: 28783 grad_norm: 0.0195 loss: 0.6834 +2023/06/01 02:38:37 - mmengine - INFO - Epoch(train) [6][5600/5758] lr: 8.5502e-04 eta: 18:55:59 time: 0.8344 data_time: 0.0024 memory: 28783 grad_norm: 0.0193 loss: 0.6857 +2023/06/01 02:39:57 - mmengine - INFO - Epoch(train) [6][5700/5758] lr: 8.5502e-04 eta: 18:54:25 time: 0.8585 data_time: 0.0028 memory: 28783 grad_norm: 0.0190 loss: 0.6903 +2023/06/01 02:40:46 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_20230531_183239 +2023/06/01 02:40:46 - mmengine - INFO - Saving checkpoint at 6 epochs +2023/06/01 02:41:03 - mmengine - INFO - Epoch(val) [6][16/16] accuracy/top1: 100.0000 single-label/precision_classwise: [100.0, 0.0] single-label/recall_classwise: [100.0, 0.0] single-label/f1-score_classwise: [100.0, 0.0] data_time: 0.2128 time: 0.6272 +2023/06/01 02:42:26 - mmengine - INFO - Epoch(train) [7][ 100/5758] lr: 7.9595e-04 eta: 18:52:07 time: 0.7517 data_time: 0.0054 memory: 28783 grad_norm: 0.0170 loss: 0.6851 +2023/06/01 02:43:47 - mmengine - INFO - Epoch(train) [7][ 200/5758] lr: 7.9595e-04 eta: 18:50:34 time: 0.8486 data_time: 0.0018 memory: 28783 grad_norm: 0.0212 loss: 0.6903 +2023/06/01 02:45:08 - mmengine - INFO - Epoch(train) [7][ 300/5758] lr: 7.9595e-04 eta: 18:49:02 time: 0.8561 data_time: 0.0017 memory: 28783 grad_norm: 0.0162 loss: 0.6866 +2023/06/01 02:46:30 - mmengine - INFO - Epoch(train) [7][ 400/5758] lr: 7.9595e-04 eta: 18:47:31 time: 0.8277 data_time: 0.0020 memory: 28783 grad_norm: 0.0212 loss: 0.6837 +2023/06/01 02:47:10 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_20230531_183239 +2023/06/01 02:47:50 - mmengine - INFO - Epoch(train) [7][ 500/5758] lr: 7.9595e-04 eta: 18:45:58 time: 0.8401 data_time: 0.0018 memory: 28783 grad_norm: 0.0233 loss: 0.6871 +2023/06/01 02:49:13 - mmengine - INFO - Epoch(train) [7][ 600/5758] lr: 7.9595e-04 eta: 18:44:30 time: 0.8155 data_time: 0.0016 memory: 28783 grad_norm: 0.0177 loss: 0.6821 +2023/06/01 02:50:33 - mmengine - INFO - Epoch(train) [7][ 700/5758] lr: 7.9595e-04 eta: 18:42:55 time: 0.8206 data_time: 0.0019 memory: 28783 grad_norm: 0.0168 loss: 0.6869 +2023/06/01 02:51:53 - mmengine - INFO - Epoch(train) [7][ 800/5758] lr: 7.9595e-04 eta: 18:41:23 time: 0.8040 data_time: 0.2923 memory: 28783 grad_norm: 0.0152 loss: 0.6861 +2023/06/01 02:53:15 - mmengine - INFO - Epoch(train) [7][ 900/5758] lr: 7.9595e-04 eta: 18:39:53 time: 0.8413 data_time: 0.3397 memory: 28783 grad_norm: 0.0198 loss: 0.6854 +2023/06/01 02:54:40 - mmengine - INFO - Epoch(train) [7][1000/5758] lr: 7.9595e-04 eta: 18:38:29 time: 0.8029 data_time: 0.2863 memory: 28783 grad_norm: 0.0221 loss: 0.6890 +2023/06/01 02:56:03 - mmengine - INFO - Epoch(train) [7][1100/5758] lr: 7.9595e-04 eta: 18:37:03 time: 0.8516 data_time: 0.3324 memory: 28783 grad_norm: 0.0188 loss: 0.6857 +2023/06/01 02:57:27 - mmengine - INFO - Epoch(train) [7][1200/5758] lr: 7.9595e-04 eta: 18:35:37 time: 0.8367 data_time: 0.3209 memory: 28783 grad_norm: 0.0179 loss: 0.6867 +2023/06/01 02:58:50 - mmengine - INFO - Epoch(train) [7][1300/5758] lr: 7.9595e-04 eta: 18:34:12 time: 0.8953 data_time: 0.3767 memory: 28783 grad_norm: 0.0158 loss: 0.6880 +2023/06/01 03:00:14 - mmengine - INFO - Epoch(train) [7][1400/5758] lr: 7.9595e-04 eta: 18:32:45 time: 0.8477 data_time: 0.3434 memory: 28783 grad_norm: 0.0223 loss: 0.6853 +2023/06/01 03:01:00 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_20230531_183239 +2023/06/01 03:01:37 - mmengine - INFO - Epoch(train) [7][1500/5758] lr: 7.9595e-04 eta: 18:31:19 time: 0.8567 data_time: 0.3418 memory: 28783 grad_norm: 0.0199 loss: 0.6861 +2023/06/01 03:02:59 - mmengine - INFO - Epoch(train) [7][1600/5758] lr: 7.9595e-04 eta: 18:29:50 time: 0.8189 data_time: 0.3156 memory: 28783 grad_norm: 0.0147 loss: 0.6850 +2023/06/01 03:04:22 - mmengine - INFO - Epoch(train) [7][1700/5758] lr: 7.9595e-04 eta: 18:28:22 time: 0.8381 data_time: 0.3218 memory: 28783 grad_norm: 0.0211 loss: 0.6843 +2023/06/01 03:05:45 - mmengine - INFO - Epoch(train) [7][1800/5758] lr: 7.9595e-04 eta: 18:26:55 time: 0.8505 data_time: 0.3440 memory: 28783 grad_norm: 0.0131 loss: 0.6876 +2023/06/01 03:07:08 - mmengine - INFO - Epoch(train) [7][1900/5758] lr: 7.9595e-04 eta: 18:25:27 time: 0.7997 data_time: 0.2885 memory: 28783 grad_norm: 0.0145 loss: 0.6893 +2023/06/01 03:08:30 - mmengine - INFO - Epoch(train) [7][2000/5758] lr: 7.9595e-04 eta: 18:23:58 time: 0.8252 data_time: 0.3208 memory: 28783 grad_norm: 0.0153 loss: 0.6876 +2023/06/01 03:09:50 - mmengine - INFO - Epoch(train) [7][2100/5758] lr: 7.9595e-04 eta: 18:22:25 time: 0.7752 data_time: 0.2729 memory: 28783 grad_norm: 0.0200 loss: 0.6905 +2023/06/01 03:11:12 - mmengine - INFO - Epoch(train) [7][2200/5758] lr: 7.9595e-04 eta: 18:20:56 time: 0.7908 data_time: 0.2874 memory: 28783 grad_norm: 0.0148 loss: 0.6878 +2023/06/01 03:12:33 - mmengine - INFO - Epoch(train) [7][2300/5758] lr: 7.9595e-04 eta: 18:19:25 time: 0.7963 data_time: 0.2927 memory: 28783 grad_norm: 0.0212 loss: 0.6867 +2023/06/01 03:13:52 - mmengine - INFO - Epoch(train) [7][2400/5758] lr: 7.9595e-04 eta: 18:17:51 time: 0.7543 data_time: 0.2499 memory: 28783 grad_norm: 0.0215 loss: 0.6857 +2023/06/01 03:14:35 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_20230531_183239 +2023/06/01 03:15:11 - mmengine - INFO - Epoch(train) [7][2500/5758] lr: 7.9595e-04 eta: 18:16:14 time: 0.7801 data_time: 0.2763 memory: 28783 grad_norm: 0.0192 loss: 0.6884 +2023/06/01 03:16:37 - mmengine - INFO - Epoch(train) [7][2600/5758] lr: 7.9595e-04 eta: 18:14:54 time: 0.8416 data_time: 0.3375 memory: 28783 grad_norm: 0.0208 loss: 0.6886 +2023/06/01 03:17:59 - mmengine - INFO - Epoch(train) [7][2700/5758] lr: 7.9595e-04 eta: 18:13:26 time: 0.8801 data_time: 0.3771 memory: 28783 grad_norm: 0.0188 loss: 0.6849 +2023/06/01 03:19:22 - mmengine - INFO - Epoch(train) [7][2800/5758] lr: 7.9595e-04 eta: 18:11:59 time: 0.8021 data_time: 0.2976 memory: 28783 grad_norm: 0.0181 loss: 0.6893 +2023/06/01 03:20:38 - mmengine - INFO - Epoch(train) [7][2900/5758] lr: 7.9595e-04 eta: 18:10:18 time: 0.7166 data_time: 0.2050 memory: 28783 grad_norm: 0.0186 loss: 0.6838 +2023/06/01 03:22:01 - mmengine - INFO - Epoch(train) [7][3000/5758] lr: 7.9595e-04 eta: 18:08:50 time: 0.7740 data_time: 0.0017 memory: 28783 grad_norm: 0.0136 loss: 0.6845 +2023/06/01 03:23:22 - mmengine - INFO - Epoch(train) [7][3100/5758] lr: 7.9595e-04 eta: 18:07:20 time: 0.8409 data_time: 0.0845 memory: 28783 grad_norm: 0.0133 loss: 0.6865 +2023/06/01 03:24:45 - mmengine - INFO - Epoch(train) [7][3200/5758] lr: 7.9595e-04 eta: 18:05:53 time: 0.8186 data_time: 0.0018 memory: 28783 grad_norm: 0.0188 loss: 0.6881 +2023/06/01 03:26:07 - mmengine - INFO - Epoch(train) [7][3300/5758] lr: 7.9595e-04 eta: 18:04:25 time: 0.8447 data_time: 0.0019 memory: 28783 grad_norm: 0.0250 loss: 0.6846 +2023/06/01 03:27:28 - mmengine - INFO - Epoch(train) [7][3400/5758] lr: 7.9595e-04 eta: 18:02:54 time: 0.8644 data_time: 0.0017 memory: 28783 grad_norm: 0.0154 loss: 0.6840 +2023/06/01 03:28:10 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_20230531_183239 +2023/06/01 03:28:50 - mmengine - INFO - Epoch(train) [7][3500/5758] lr: 7.9595e-04 eta: 18:01:27 time: 0.8465 data_time: 0.0024 memory: 28783 grad_norm: 0.0177 loss: 0.6878 +2023/06/01 03:30:12 - mmengine - INFO - Epoch(train) [7][3600/5758] lr: 7.9595e-04 eta: 17:59:57 time: 0.8780 data_time: 0.0019 memory: 28783 grad_norm: 0.0186 loss: 0.6886 +2023/06/01 03:31:35 - mmengine - INFO - Epoch(train) [7][3700/5758] lr: 7.9595e-04 eta: 17:58:31 time: 0.8227 data_time: 0.0017 memory: 28783 grad_norm: 0.0198 loss: 0.6856 +2023/06/01 03:32:58 - mmengine - INFO - Epoch(train) [7][3800/5758] lr: 7.9595e-04 eta: 17:57:04 time: 0.8511 data_time: 0.0019 memory: 28783 grad_norm: 0.0186 loss: 0.6852 +2023/06/01 03:34:19 - mmengine - INFO - Epoch(train) [7][3900/5758] lr: 7.9595e-04 eta: 17:55:35 time: 0.8082 data_time: 0.0018 memory: 28783 grad_norm: 0.0190 loss: 0.6851 +2023/06/01 03:35:42 - mmengine - INFO - Epoch(train) [7][4000/5758] lr: 7.9595e-04 eta: 17:54:07 time: 0.7962 data_time: 0.0017 memory: 28783 grad_norm: 0.0176 loss: 0.6859 +2023/06/01 03:37:03 - mmengine - INFO - Epoch(train) [7][4100/5758] lr: 7.9595e-04 eta: 17:52:38 time: 0.8690 data_time: 0.0020 memory: 28783 grad_norm: 0.0154 loss: 0.6879 +2023/06/01 03:38:26 - mmengine - INFO - Epoch(train) [7][4200/5758] lr: 7.9595e-04 eta: 17:51:10 time: 0.8068 data_time: 0.0022 memory: 28783 grad_norm: 0.8099 loss: 0.6889 +2023/06/01 03:39:47 - mmengine - INFO - Epoch(train) [7][4300/5758] lr: 7.9595e-04 eta: 17:49:40 time: 0.8488 data_time: 0.0021 memory: 28783 grad_norm: 0.0232 loss: 0.6846 +2023/06/01 03:41:08 - mmengine - INFO - Epoch(train) [7][4400/5758] lr: 7.9595e-04 eta: 17:48:10 time: 0.8235 data_time: 0.0018 memory: 28783 grad_norm: 0.0148 loss: 0.6849 +2023/06/01 03:41:49 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_20230531_183239 +2023/06/01 03:42:29 - mmengine - INFO - Epoch(train) [7][4500/5758] lr: 7.9595e-04 eta: 17:46:39 time: 0.8097 data_time: 0.0019 memory: 28783 grad_norm: 0.0112 loss: 0.6835 +2023/06/01 03:43:49 - mmengine - INFO - Epoch(train) [7][4600/5758] lr: 7.9595e-04 eta: 17:45:08 time: 0.8394 data_time: 0.0019 memory: 28783 grad_norm: 0.0177 loss: 0.6860 +2023/06/01 03:45:08 - mmengine - INFO - Epoch(train) [7][4700/5758] lr: 7.9595e-04 eta: 17:43:35 time: 0.8047 data_time: 0.0018 memory: 28783 grad_norm: 0.0138 loss: 0.6885 +2023/06/01 03:46:29 - mmengine - INFO - Epoch(train) [7][4800/5758] lr: 7.9595e-04 eta: 17:42:05 time: 0.8533 data_time: 0.0018 memory: 28783 grad_norm: 0.0170 loss: 0.6874 +2023/06/01 03:47:53 - mmengine - INFO - Epoch(train) [7][4900/5758] lr: 7.9595e-04 eta: 17:40:40 time: 0.8903 data_time: 0.0018 memory: 28783 grad_norm: 0.0193 loss: 0.6813 +2023/06/01 03:49:16 - mmengine - INFO - Epoch(train) [7][5000/5758] lr: 7.9595e-04 eta: 17:39:14 time: 0.8734 data_time: 0.0017 memory: 28783 grad_norm: 0.0180 loss: 0.6893 +2023/06/01 03:50:39 - mmengine - INFO - Epoch(train) [7][5100/5758] lr: 7.9595e-04 eta: 17:37:47 time: 0.8268 data_time: 0.0019 memory: 28783 grad_norm: 0.0178 loss: 0.6879 +2023/06/01 03:52:01 - mmengine - INFO - Epoch(train) [7][5200/5758] lr: 7.9595e-04 eta: 17:36:21 time: 0.9307 data_time: 0.0020 memory: 28783 grad_norm: 0.0141 loss: 0.6858 +2023/06/01 03:53:23 - mmengine - INFO - Epoch(train) [7][5300/5758] lr: 7.9595e-04 eta: 17:34:53 time: 0.7953 data_time: 0.0017 memory: 28783 grad_norm: 0.0093 loss: 0.6836 +2023/06/01 03:54:44 - mmengine - INFO - Epoch(train) [7][5400/5758] lr: 7.9595e-04 eta: 17:33:22 time: 0.8185 data_time: 0.0017 memory: 28783 grad_norm: 0.0202 loss: 0.6899 +2023/06/01 03:55:27 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_20230531_183239 +2023/06/01 03:56:06 - mmengine - INFO - Epoch(train) [7][5500/5758] lr: 7.9595e-04 eta: 17:31:54 time: 0.7978 data_time: 0.0017 memory: 28783 grad_norm: 0.0122 loss: 0.6886 +2023/06/01 03:57:29 - mmengine - INFO - Epoch(train) [7][5600/5758] lr: 7.9595e-04 eta: 17:30:28 time: 0.8521 data_time: 0.0017 memory: 28783 grad_norm: 0.0130 loss: 0.6856 +2023/06/01 03:58:49 - mmengine - INFO - Epoch(train) [7][5700/5758] lr: 7.9595e-04 eta: 17:28:58 time: 0.8296 data_time: 0.0017 memory: 28783 grad_norm: 0.0174 loss: 0.6840 +2023/06/01 03:59:36 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_20230531_183239 +2023/06/01 03:59:36 - mmengine - INFO - Saving checkpoint at 7 epochs +2023/06/01 03:59:54 - mmengine - INFO - Epoch(val) [7][16/16] accuracy/top1: 100.0000 single-label/precision_classwise: [100.0, 0.0] single-label/recall_classwise: [100.0, 0.0] single-label/f1-score_classwise: [100.0, 0.0] data_time: 0.2213 time: 0.6372 +2023/06/01 04:01:21 - mmengine - INFO - Epoch(train) [8][ 100/5758] lr: 7.2973e-04 eta: 17:26:45 time: 0.8509 data_time: 0.1767 memory: 28783 grad_norm: 0.0115 loss: 0.6881 +2023/06/01 04:02:40 - mmengine - INFO - Epoch(train) [8][ 200/5758] lr: 7.2973e-04 eta: 17:25:13 time: 0.8216 data_time: 0.2109 memory: 28783 grad_norm: 0.0157 loss: 0.6835 +2023/06/01 04:04:02 - mmengine - INFO - Epoch(train) [8][ 300/5758] lr: 7.2973e-04 eta: 17:23:45 time: 0.8163 data_time: 0.0020 memory: 28783 grad_norm: 0.0135 loss: 0.6881 +2023/06/01 04:05:23 - mmengine - INFO - Epoch(train) [8][ 400/5758] lr: 7.2973e-04 eta: 17:22:15 time: 0.7795 data_time: 0.0016 memory: 28783 grad_norm: 0.0198 loss: 0.6859 +2023/06/01 04:06:46 - mmengine - INFO - Epoch(train) [8][ 500/5758] lr: 7.2973e-04 eta: 17:20:49 time: 0.8328 data_time: 0.0016 memory: 28783 grad_norm: 0.0192 loss: 0.6904 +2023/06/01 04:08:09 - mmengine - INFO - Epoch(train) [8][ 600/5758] lr: 7.2973e-04 eta: 17:19:23 time: 0.8075 data_time: 0.0016 memory: 28783 grad_norm: 0.0217 loss: 0.6876 +2023/06/01 04:09:24 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_20230531_183239 +2023/06/01 04:09:30 - mmengine - INFO - Epoch(train) [8][ 700/5758] lr: 7.2973e-04 eta: 17:17:54 time: 0.8112 data_time: 0.0025 memory: 28783 grad_norm: 0.0118 loss: 0.6842 +2023/06/01 04:10:51 - mmengine - INFO - Epoch(train) [8][ 800/5758] lr: 7.2973e-04 eta: 17:16:25 time: 0.7763 data_time: 0.0020 memory: 28783 grad_norm: 0.0141 loss: 0.6851 +2023/06/01 04:12:12 - mmengine - INFO - Epoch(train) [8][ 900/5758] lr: 7.2973e-04 eta: 17:14:55 time: 0.8363 data_time: 0.0018 memory: 28783 grad_norm: 0.0162 loss: 0.6846 +2023/06/01 04:13:33 - mmengine - INFO - Epoch(train) [8][1000/5758] lr: 7.2973e-04 eta: 17:13:26 time: 0.8303 data_time: 0.0017 memory: 28783 grad_norm: 0.0206 loss: 0.6891 +2023/06/01 04:14:57 - mmengine - INFO - Epoch(train) [8][1100/5758] lr: 7.2973e-04 eta: 17:12:02 time: 0.7985 data_time: 0.0018 memory: 28783 grad_norm: 0.0202 loss: 0.6848 +2023/06/01 04:16:19 - mmengine - INFO - Epoch(train) [8][1200/5758] lr: 7.2973e-04 eta: 17:10:35 time: 0.8626 data_time: 0.3605 memory: 28783 grad_norm: 0.0205 loss: 0.6852 +2023/06/01 04:17:40 - mmengine - INFO - Epoch(train) [8][1300/5758] lr: 7.2973e-04 eta: 17:09:06 time: 0.7553 data_time: 0.2516 memory: 28783 grad_norm: 0.0194 loss: 0.6866 +2023/06/01 04:19:01 - mmengine - INFO - Epoch(train) [8][1400/5758] lr: 7.2973e-04 eta: 17:07:36 time: 0.7168 data_time: 0.2104 memory: 28783 grad_norm: 0.0113 loss: 0.6893 +2023/06/01 04:20:20 - mmengine - INFO - Epoch(train) [8][1500/5758] lr: 7.2973e-04 eta: 17:06:05 time: 0.7893 data_time: 0.1728 memory: 28783 grad_norm: 0.0205 loss: 0.6859 +2023/06/01 04:21:42 - mmengine - INFO - Epoch(train) [8][1600/5758] lr: 7.2973e-04 eta: 17:04:36 time: 0.8095 data_time: 0.0018 memory: 28783 grad_norm: 0.0159 loss: 0.6870 +2023/06/01 04:22:58 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_20230531_183239 +2023/06/01 04:23:04 - mmengine - INFO - Epoch(train) [8][1700/5758] lr: 7.2973e-04 eta: 17:03:09 time: 0.8007 data_time: 0.0020 memory: 28783 grad_norm: 0.0201 loss: 0.6876 +2023/06/01 04:24:26 - mmengine - INFO - Epoch(train) [8][1800/5758] lr: 7.2973e-04 eta: 17:01:43 time: 0.8972 data_time: 0.0020 memory: 28783 grad_norm: 0.0178 loss: 0.6895 +2023/06/01 04:25:49 - mmengine - INFO - Epoch(train) [8][1900/5758] lr: 7.2973e-04 eta: 17:00:16 time: 0.8005 data_time: 0.0025 memory: 28783 grad_norm: 0.0142 loss: 0.6844 +2023/06/01 04:27:10 - mmengine - INFO - Epoch(train) [8][2000/5758] lr: 7.2973e-04 eta: 16:58:47 time: 0.7823 data_time: 0.0016 memory: 28783 grad_norm: 0.0162 loss: 0.6884 +2023/06/01 04:28:31 - mmengine - INFO - Epoch(train) [8][2100/5758] lr: 7.2973e-04 eta: 16:57:19 time: 0.8572 data_time: 0.0020 memory: 28783 grad_norm: 0.0220 loss: 0.6889 +2023/06/01 04:29:54 - mmengine - INFO - Epoch(train) [8][2200/5758] lr: 7.2973e-04 eta: 16:55:54 time: 0.8040 data_time: 0.0033 memory: 28783 grad_norm: 0.0102 loss: 0.6837 +2023/06/01 04:31:15 - mmengine - INFO - Epoch(train) [8][2300/5758] lr: 7.2973e-04 eta: 16:54:25 time: 0.8095 data_time: 0.0017 memory: 28783 grad_norm: 0.0155 loss: 0.6842 +2023/06/01 04:32:37 - mmengine - INFO - Epoch(train) [8][2400/5758] lr: 7.2973e-04 eta: 16:52:57 time: 0.8551 data_time: 0.0018 memory: 28783 grad_norm: 0.0165 loss: 0.6859 +2023/06/01 04:33:58 - mmengine - INFO - Epoch(train) [8][2500/5758] lr: 7.2973e-04 eta: 16:51:28 time: 0.8096 data_time: 0.0021 memory: 28783 grad_norm: 0.0176 loss: 0.6891 +2023/06/01 04:35:19 - mmengine - INFO - Epoch(train) [8][2600/5758] lr: 7.2973e-04 eta: 16:50:00 time: 0.8430 data_time: 0.0017 memory: 28783 grad_norm: 0.0211 loss: 0.6902 +2023/06/01 04:36:35 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_20230531_183239 +2023/06/01 04:36:42 - mmengine - INFO - Epoch(train) [8][2700/5758] lr: 7.2973e-04 eta: 16:48:34 time: 0.8127 data_time: 0.0023 memory: 28783 grad_norm: 0.0190 loss: 0.6881 +2023/06/01 04:38:03 - mmengine - INFO - Epoch(train) [8][2800/5758] lr: 7.2973e-04 eta: 16:47:06 time: 0.7912 data_time: 0.0018 memory: 28783 grad_norm: 0.0096 loss: 0.6858 +2023/06/01 04:39:25 - mmengine - INFO - Epoch(train) [8][2900/5758] lr: 7.2973e-04 eta: 16:45:39 time: 0.8097 data_time: 0.0019 memory: 28783 grad_norm: 0.0201 loss: 0.6859 +2023/06/01 04:40:45 - mmengine - INFO - Epoch(train) [8][3000/5758] lr: 7.2973e-04 eta: 16:44:09 time: 0.7969 data_time: 0.0021 memory: 28783 grad_norm: 0.0180 loss: 0.6841 +2023/06/01 04:42:06 - mmengine - INFO - Epoch(train) [8][3100/5758] lr: 7.2973e-04 eta: 16:42:40 time: 0.8258 data_time: 0.0018 memory: 28783 grad_norm: 0.0181 loss: 0.6890 +2023/06/01 04:43:27 - mmengine - INFO - Epoch(train) [8][3200/5758] lr: 7.2973e-04 eta: 16:41:12 time: 0.8090 data_time: 0.0017 memory: 28783 grad_norm: 0.0145 loss: 0.6880 +2023/06/01 04:44:48 - mmengine - INFO - Epoch(train) [8][3300/5758] lr: 7.2973e-04 eta: 16:39:42 time: 0.7844 data_time: 0.0017 memory: 28783 grad_norm: 0.0148 loss: 0.6863 +2023/06/01 04:46:09 - mmengine - INFO - Epoch(train) [8][3400/5758] lr: 7.2973e-04 eta: 16:38:15 time: 0.8027 data_time: 0.0016 memory: 28783 grad_norm: 0.0123 loss: 0.6870 +2023/06/01 04:47:31 - mmengine - INFO - Epoch(train) [8][3500/5758] lr: 7.2973e-04 eta: 16:36:47 time: 0.8362 data_time: 0.0018 memory: 28783 grad_norm: 0.0162 loss: 0.6828 +2023/06/01 04:48:51 - mmengine - INFO - Epoch(train) [8][3600/5758] lr: 7.2973e-04 eta: 16:35:18 time: 0.8031 data_time: 0.0017 memory: 28783 grad_norm: 0.0211 loss: 0.6869 +2023/06/01 04:50:08 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_20230531_183239 +2023/06/01 04:50:14 - mmengine - INFO - Epoch(train) [8][3700/5758] lr: 7.2973e-04 eta: 16:33:52 time: 0.8144 data_time: 0.0024 memory: 28783 grad_norm: 0.0128 loss: 0.6844 +2023/06/01 04:51:34 - mmengine - INFO - Epoch(train) [8][3800/5758] lr: 7.2973e-04 eta: 16:32:23 time: 0.8152 data_time: 0.0018 memory: 28783 grad_norm: 0.0090 loss: 0.6857 +2023/06/01 04:52:56 - mmengine - INFO - Epoch(train) [8][3900/5758] lr: 7.2973e-04 eta: 16:30:55 time: 0.8057 data_time: 0.0026 memory: 28783 grad_norm: 0.0130 loss: 0.6858 +2023/06/01 04:54:18 - mmengine - INFO - Epoch(train) [8][4000/5758] lr: 7.2973e-04 eta: 16:29:29 time: 0.8713 data_time: 0.0022 memory: 28783 grad_norm: 0.0178 loss: 0.6838 +2023/06/01 04:55:39 - mmengine - INFO - Epoch(train) [8][4100/5758] lr: 7.2973e-04 eta: 16:28:02 time: 0.8257 data_time: 0.0021 memory: 28783 grad_norm: 0.0155 loss: 0.6873 +2023/06/01 04:57:00 - mmengine - INFO - Epoch(train) [8][4200/5758] lr: 7.2973e-04 eta: 16:26:33 time: 0.7685 data_time: 0.0025 memory: 28783 grad_norm: 0.0181 loss: 0.6862 +2023/06/01 04:58:21 - mmengine - INFO - Epoch(train) [8][4300/5758] lr: 7.2973e-04 eta: 16:25:05 time: 0.7909 data_time: 0.0019 memory: 28783 grad_norm: 0.0102 loss: 0.6883 +2023/06/01 04:59:42 - mmengine - INFO - Epoch(train) [8][4400/5758] lr: 7.2973e-04 eta: 16:23:36 time: 0.8168 data_time: 0.0026 memory: 28783 grad_norm: 0.0206 loss: 0.6878 +2023/06/01 05:01:03 - mmengine - INFO - Epoch(train) [8][4500/5758] lr: 7.2973e-04 eta: 16:22:08 time: 0.8485 data_time: 0.0029 memory: 28783 grad_norm: 0.0146 loss: 0.6859 +2023/06/01 05:02:25 - mmengine - INFO - Epoch(train) [8][4600/5758] lr: 7.2973e-04 eta: 16:20:42 time: 0.7753 data_time: 0.0025 memory: 28783 grad_norm: 0.0180 loss: 0.6885 +2023/06/01 05:03:41 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_20230531_183239 +2023/06/01 05:03:47 - mmengine - INFO - Epoch(train) [8][4700/5758] lr: 7.2973e-04 eta: 16:19:15 time: 0.8395 data_time: 0.0024 memory: 28783 grad_norm: 0.0148 loss: 0.6870 +2023/06/01 05:05:08 - mmengine - INFO - Epoch(train) [8][4800/5758] lr: 7.2973e-04 eta: 16:17:48 time: 0.8047 data_time: 0.0021 memory: 28783 grad_norm: 0.0098 loss: 0.6907 +2023/06/01 05:06:31 - mmengine - INFO - Epoch(train) [8][4900/5758] lr: 7.2973e-04 eta: 16:16:22 time: 0.8291 data_time: 0.0017 memory: 28783 grad_norm: 0.0154 loss: 0.6874 +2023/06/01 05:07:50 - mmengine - INFO - Epoch(train) [8][5000/5758] lr: 7.2973e-04 eta: 16:14:51 time: 0.8154 data_time: 0.0016 memory: 28783 grad_norm: 0.0188 loss: 0.6829 +2023/06/01 05:09:11 - mmengine - INFO - Epoch(train) [8][5100/5758] lr: 7.2973e-04 eta: 16:13:24 time: 0.7922 data_time: 0.0017 memory: 28783 grad_norm: 0.0227 loss: 0.6895 +2023/06/01 05:10:33 - mmengine - INFO - Epoch(train) [8][5200/5758] lr: 7.2973e-04 eta: 16:11:57 time: 0.7759 data_time: 0.0017 memory: 28783 grad_norm: 0.0178 loss: 0.6875 +2023/06/01 05:11:54 - mmengine - INFO - Epoch(train) [8][5300/5758] lr: 7.2973e-04 eta: 16:10:29 time: 0.7868 data_time: 0.0020 memory: 28783 grad_norm: 0.0165 loss: 0.6843 +2023/06/01 05:13:16 - mmengine - INFO - Epoch(train) [8][5400/5758] lr: 7.2973e-04 eta: 16:09:02 time: 0.7778 data_time: 0.0022 memory: 28783 grad_norm: 0.0177 loss: 0.6848 +2023/06/01 05:14:36 - mmengine - INFO - Epoch(train) [8][5500/5758] lr: 7.2973e-04 eta: 16:07:32 time: 0.7659 data_time: 0.0015 memory: 28783 grad_norm: 0.0141 loss: 0.6838 +2023/06/01 05:15:56 - mmengine - INFO - Epoch(train) [8][5600/5758] lr: 7.2973e-04 eta: 16:06:04 time: 0.7757 data_time: 0.0020 memory: 28783 grad_norm: 0.0150 loss: 0.6884 +2023/06/01 05:17:11 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_20230531_183239 +2023/06/01 05:17:17 - mmengine - INFO - Epoch(train) [8][5700/5758] lr: 7.2973e-04 eta: 16:04:36 time: 0.7966 data_time: 0.0022 memory: 28783 grad_norm: 0.0167 loss: 0.6846 +2023/06/01 05:18:03 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_20230531_183239 +2023/06/01 05:18:03 - mmengine - INFO - Saving checkpoint at 8 epochs +2023/06/01 05:18:22 - mmengine - INFO - Epoch(val) [8][16/16] accuracy/top1: 100.0000 single-label/precision_classwise: [100.0, 0.0] single-label/recall_classwise: [100.0, 0.0] single-label/f1-score_classwise: [100.0, 0.0] data_time: 0.2374 time: 0.6570 +2023/06/01 05:19:47 - mmengine - INFO - Epoch(train) [9][ 100/5758] lr: 6.5796e-04 eta: 16:02:22 time: 0.7795 data_time: 0.0024 memory: 28783 grad_norm: 0.0198 loss: 0.6865 +2023/06/01 05:21:07 - mmengine - INFO - Epoch(train) [9][ 200/5758] lr: 6.5796e-04 eta: 16:00:53 time: 0.8062 data_time: 0.0020 memory: 28783 grad_norm: 0.0200 loss: 0.6888 +2023/06/01 05:22:28 - mmengine - INFO - Epoch(train) [9][ 300/5758] lr: 6.5796e-04 eta: 15:59:25 time: 0.8260 data_time: 0.0023 memory: 28783 grad_norm: 0.0128 loss: 0.6854 +2023/06/01 05:23:47 - mmengine - INFO - Epoch(train) [9][ 400/5758] lr: 6.5796e-04 eta: 15:57:56 time: 0.7888 data_time: 0.0019 memory: 28783 grad_norm: 0.0174 loss: 0.6876 +2023/06/01 05:25:09 - mmengine - INFO - Epoch(train) [9][ 500/5758] lr: 6.5796e-04 eta: 15:56:29 time: 0.8541 data_time: 0.0029 memory: 28783 grad_norm: 0.0131 loss: 0.6862 +2023/06/01 05:26:31 - mmengine - INFO - Epoch(train) [9][ 600/5758] lr: 6.5796e-04 eta: 15:55:03 time: 0.7888 data_time: 0.0027 memory: 28783 grad_norm: 0.0178 loss: 0.6868 +2023/06/01 05:27:52 - mmengine - INFO - Epoch(train) [9][ 700/5758] lr: 6.5796e-04 eta: 15:53:34 time: 0.7875 data_time: 0.0031 memory: 28783 grad_norm: 0.0178 loss: 0.6855 +2023/06/01 05:29:11 - mmengine - INFO - Epoch(train) [9][ 800/5758] lr: 6.5796e-04 eta: 15:52:05 time: 0.7667 data_time: 0.0024 memory: 28783 grad_norm: 0.0126 loss: 0.6891 +2023/06/01 05:30:32 - mmengine - INFO - Epoch(train) [9][ 900/5758] lr: 6.5796e-04 eta: 15:50:37 time: 0.7551 data_time: 0.0022 memory: 28783 grad_norm: 0.0129 loss: 0.6859 +2023/06/01 05:30:59 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_20230531_183239 +2023/06/01 05:31:52 - mmengine - INFO - Epoch(train) [9][1000/5758] lr: 6.5796e-04 eta: 15:49:08 time: 0.7887 data_time: 0.0018 memory: 28783 grad_norm: 0.0125 loss: 0.6913 +2023/06/01 05:33:12 - mmengine - INFO - Epoch(train) [9][1100/5758] lr: 6.5796e-04 eta: 15:47:39 time: 0.7885 data_time: 0.0017 memory: 28783 grad_norm: 0.0163 loss: 0.6868 +2023/06/01 05:34:31 - mmengine - INFO - Epoch(train) [9][1200/5758] lr: 6.5796e-04 eta: 15:46:09 time: 0.8099 data_time: 0.0018 memory: 28783 grad_norm: 0.0178 loss: 0.6868 +2023/06/01 05:35:52 - mmengine - INFO - Epoch(train) [9][1300/5758] lr: 6.5796e-04 eta: 15:44:42 time: 0.8420 data_time: 0.0028 memory: 28783 grad_norm: 0.0167 loss: 0.6840 +2023/06/01 05:37:13 - mmengine - INFO - Epoch(train) [9][1400/5758] lr: 6.5796e-04 eta: 15:43:14 time: 0.7702 data_time: 0.0017 memory: 28783 grad_norm: 0.0163 loss: 0.6880 +2023/06/01 05:38:33 - mmengine - INFO - Epoch(train) [9][1500/5758] lr: 6.5796e-04 eta: 15:41:46 time: 0.8083 data_time: 0.0018 memory: 28783 grad_norm: 0.0145 loss: 0.6875 +2023/06/01 05:39:54 - mmengine - INFO - Epoch(train) [9][1600/5758] lr: 6.5796e-04 eta: 15:40:19 time: 0.8419 data_time: 0.0020 memory: 28783 grad_norm: 0.0154 loss: 0.6857 +2023/06/01 05:41:15 - mmengine - INFO - Epoch(train) [9][1700/5758] lr: 6.5796e-04 eta: 15:38:51 time: 0.8011 data_time: 0.0022 memory: 28783 grad_norm: 0.0066 loss: 0.6868 +2023/06/01 05:42:34 - mmengine - INFO - Epoch(train) [9][1800/5758] lr: 6.5796e-04 eta: 15:37:22 time: 0.7784 data_time: 0.0020 memory: 28783 grad_norm: 0.0146 loss: 0.6887 +2023/06/01 05:43:55 - mmengine - INFO - Epoch(train) [9][1900/5758] lr: 6.5796e-04 eta: 15:35:54 time: 0.8535 data_time: 0.0017 memory: 28783 grad_norm: 0.0155 loss: 0.6875 +2023/06/01 05:44:22 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_20230531_183239 +2023/06/01 05:45:15 - mmengine - INFO - Epoch(train) [9][2000/5758] lr: 6.5796e-04 eta: 15:34:26 time: 0.8277 data_time: 0.0017 memory: 28783 grad_norm: 0.0287 loss: 0.6899 +2023/06/01 05:46:37 - mmengine - INFO - Epoch(train) [9][2100/5758] lr: 6.5796e-04 eta: 15:32:59 time: 0.7595 data_time: 0.0017 memory: 28783 grad_norm: 0.0221 loss: 0.6857 +2023/06/01 05:47:58 - mmengine - INFO - Epoch(train) [9][2200/5758] lr: 6.5796e-04 eta: 15:31:32 time: 0.8263 data_time: 0.0019 memory: 28783 grad_norm: 0.0142 loss: 0.6850 +2023/06/01 05:49:19 - mmengine - INFO - Epoch(train) [9][2300/5758] lr: 6.5796e-04 eta: 15:30:05 time: 0.8381 data_time: 0.0017 memory: 28783 grad_norm: 0.0174 loss: 0.6890 +2023/06/01 05:50:40 - mmengine - INFO - Epoch(train) [9][2400/5758] lr: 6.5796e-04 eta: 15:28:38 time: 0.7937 data_time: 0.0025 memory: 28783 grad_norm: 0.0120 loss: 0.6880 +2023/06/01 05:52:02 - mmengine - INFO - Epoch(train) [9][2500/5758] lr: 6.5796e-04 eta: 15:27:13 time: 0.8436 data_time: 0.0018 memory: 28783 grad_norm: 0.0168 loss: 0.6870 +2023/06/01 05:53:22 - mmengine - INFO - Epoch(train) [9][2600/5758] lr: 6.5796e-04 eta: 15:25:45 time: 0.7876 data_time: 0.0018 memory: 28783 grad_norm: 0.0232 loss: 0.6854 +2023/06/01 05:54:43 - mmengine - INFO - Epoch(train) [9][2700/5758] lr: 6.5796e-04 eta: 15:24:18 time: 0.8110 data_time: 0.0017 memory: 28783 grad_norm: 0.0110 loss: 0.6875 +2023/06/01 05:56:04 - mmengine - INFO - Epoch(train) [9][2800/5758] lr: 6.5796e-04 eta: 15:22:50 time: 0.7865 data_time: 0.0018 memory: 28783 grad_norm: 0.0136 loss: 0.6873 +2023/06/01 05:57:27 - mmengine - INFO - Epoch(train) [9][2900/5758] lr: 6.5796e-04 eta: 15:21:27 time: 0.8124 data_time: 0.0020 memory: 28783 grad_norm: 0.0191 loss: 0.6865 +2023/06/01 05:57:55 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_20230531_183239 +2023/06/01 05:58:50 - mmengine - INFO - Epoch(train) [9][3000/5758] lr: 6.5796e-04 eta: 15:20:01 time: 0.8831 data_time: 0.0027 memory: 28783 grad_norm: 0.0219 loss: 0.6884 +2023/06/01 06:00:11 - mmengine - INFO - Epoch(train) [9][3100/5758] lr: 6.5796e-04 eta: 15:18:35 time: 0.8273 data_time: 0.0024 memory: 28783 grad_norm: 0.0149 loss: 0.6886 +2023/06/01 06:01:31 - mmengine - INFO - Epoch(train) [9][3200/5758] lr: 6.5796e-04 eta: 15:17:07 time: 0.7653 data_time: 0.0034 memory: 28783 grad_norm: 0.0229 loss: 0.6842 +2023/06/01 06:02:53 - mmengine - INFO - Epoch(train) [9][3300/5758] lr: 6.5796e-04 eta: 15:15:41 time: 0.7957 data_time: 0.0027 memory: 28783 grad_norm: 0.0183 loss: 0.6881 +2023/06/01 06:04:14 - mmengine - INFO - Epoch(train) [9][3400/5758] lr: 6.5796e-04 eta: 15:14:14 time: 0.7885 data_time: 0.0019 memory: 28783 grad_norm: 0.0180 loss: 0.6865 +2023/06/01 06:05:34 - mmengine - INFO - Epoch(train) [9][3500/5758] lr: 6.5796e-04 eta: 15:12:47 time: 0.7979 data_time: 0.0024 memory: 28783 grad_norm: 0.0187 loss: 0.6843 +2023/06/01 06:06:57 - mmengine - INFO - Epoch(train) [9][3600/5758] lr: 6.5796e-04 eta: 15:11:22 time: 0.8601 data_time: 0.0023 memory: 28783 grad_norm: 0.0204 loss: 0.6884 +2023/06/01 06:08:17 - mmengine - INFO - Epoch(train) [9][3700/5758] lr: 6.5796e-04 eta: 15:09:54 time: 0.7963 data_time: 0.0028 memory: 28783 grad_norm: 0.0109 loss: 0.6871 +2023/06/01 06:09:38 - mmengine - INFO - Epoch(train) [9][3800/5758] lr: 6.5796e-04 eta: 15:08:27 time: 0.7886 data_time: 0.0024 memory: 28783 grad_norm: 0.0145 loss: 0.6866 +2023/06/01 06:10:57 - mmengine - INFO - Epoch(train) [9][3900/5758] lr: 6.5796e-04 eta: 15:06:59 time: 0.8509 data_time: 0.0031 memory: 28783 grad_norm: 0.0197 loss: 0.6828 +2023/06/01 06:11:25 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_20230531_183239 +2023/06/01 06:12:19 - mmengine - INFO - Epoch(train) [9][4000/5758] lr: 6.5796e-04 eta: 15:05:32 time: 0.7776 data_time: 0.0021 memory: 28783 grad_norm: 0.0163 loss: 0.6884 +2023/06/01 06:13:38 - mmengine - INFO - Epoch(train) [9][4100/5758] lr: 6.5796e-04 eta: 15:04:04 time: 0.8472 data_time: 0.0027 memory: 28783 grad_norm: 0.0203 loss: 0.6885 +2023/06/01 06:14:59 - mmengine - INFO - Epoch(train) [9][4200/5758] lr: 6.5796e-04 eta: 15:02:37 time: 0.7698 data_time: 0.0028 memory: 28783 grad_norm: 0.0185 loss: 0.6864 +2023/06/01 06:16:20 - mmengine - INFO - Epoch(train) [9][4300/5758] lr: 6.5796e-04 eta: 15:01:10 time: 0.8138 data_time: 0.0025 memory: 28783 grad_norm: 0.0083 loss: 0.6900 +2023/06/01 06:17:42 - mmengine - INFO - Epoch(train) [9][4400/5758] lr: 6.5796e-04 eta: 14:59:45 time: 0.8480 data_time: 0.0024 memory: 28783 grad_norm: 0.0161 loss: 0.6874 +2023/06/01 06:19:03 - mmengine - INFO - Epoch(train) [9][4500/5758] lr: 6.5796e-04 eta: 14:58:18 time: 0.8253 data_time: 0.0028 memory: 28783 grad_norm: 0.0135 loss: 0.6885 +2023/06/01 06:20:25 - mmengine - INFO - Epoch(train) [9][4600/5758] lr: 6.5796e-04 eta: 14:56:53 time: 0.8205 data_time: 0.0026 memory: 28783 grad_norm: 0.0157 loss: 0.6883 +2023/06/01 06:21:45 - mmengine - INFO - Epoch(train) [9][4700/5758] lr: 6.5796e-04 eta: 14:55:25 time: 0.8077 data_time: 0.0029 memory: 28783 grad_norm: 0.0206 loss: 0.6856 +2023/06/01 06:23:08 - mmengine - INFO - Epoch(train) [9][4800/5758] lr: 6.5796e-04 eta: 14:54:01 time: 0.8818 data_time: 0.0027 memory: 28783 grad_norm: 0.0137 loss: 0.6902 +2023/06/01 06:24:28 - mmengine - INFO - Epoch(train) [9][4900/5758] lr: 6.5796e-04 eta: 14:52:34 time: 0.8431 data_time: 0.0025 memory: 28783 grad_norm: 0.0116 loss: 0.6891 +2023/06/01 06:24:56 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_20230531_183239 +2023/06/01 06:25:51 - mmengine - INFO - Epoch(train) [9][5000/5758] lr: 6.5796e-04 eta: 14:51:10 time: 0.8893 data_time: 0.0027 memory: 28783 grad_norm: 0.0262 loss: 0.6876 +2023/06/01 06:27:12 - mmengine - INFO - Epoch(train) [9][5100/5758] lr: 6.5796e-04 eta: 14:49:43 time: 0.7929 data_time: 0.0025 memory: 28783 grad_norm: 0.0182 loss: 0.6865 +2023/06/01 06:28:34 - mmengine - INFO - Epoch(train) [9][5200/5758] lr: 6.5796e-04 eta: 14:48:17 time: 0.8284 data_time: 0.0024 memory: 28783 grad_norm: 0.0163 loss: 0.6888 +2023/06/01 06:29:56 - mmengine - INFO - Epoch(train) [9][5300/5758] lr: 6.5796e-04 eta: 14:46:52 time: 0.8068 data_time: 0.0024 memory: 28783 grad_norm: 0.0270 loss: 0.6855 +2023/06/01 06:31:18 - mmengine - INFO - Epoch(train) [9][5400/5758] lr: 6.5796e-04 eta: 14:45:28 time: 0.8059 data_time: 0.0023 memory: 28783 grad_norm: 0.0113 loss: 0.6866 +2023/06/01 06:32:40 - mmengine - INFO - Epoch(train) [9][5500/5758] lr: 6.5796e-04 eta: 14:44:02 time: 0.8256 data_time: 0.0024 memory: 28783 grad_norm: 0.0179 loss: 0.6880 +2023/06/01 06:34:04 - mmengine - INFO - Epoch(train) [9][5600/5758] lr: 6.5796e-04 eta: 14:42:39 time: 0.8245 data_time: 0.0020 memory: 28783 grad_norm: 0.0185 loss: 0.6857 +2023/06/01 06:35:26 - mmengine - INFO - Epoch(train) [9][5700/5758] lr: 6.5796e-04 eta: 14:41:14 time: 0.8124 data_time: 0.0020 memory: 28783 grad_norm: 0.0181 loss: 0.6861 +2023/06/01 06:36:13 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_20230531_183239 +2023/06/01 06:36:13 - mmengine - INFO - Saving checkpoint at 9 epochs +2023/06/01 06:36:31 - mmengine - INFO - Epoch(val) [9][16/16] accuracy/top1: 100.0000 single-label/precision_classwise: [100.0, 0.0] single-label/recall_classwise: [100.0, 0.0] single-label/f1-score_classwise: [100.0, 0.0] data_time: 0.2267 time: 0.6419 +2023/06/01 06:37:56 - mmengine - INFO - Epoch(train) [10][ 100/5758] lr: 5.8244e-04 eta: 14:39:02 time: 0.7971 data_time: 0.1550 memory: 28783 grad_norm: 0.0098 loss: 0.6852 +2023/06/01 06:38:58 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_20230531_183239 +2023/06/01 06:39:16 - mmengine - INFO - Epoch(train) [10][ 200/5758] lr: 5.8244e-04 eta: 14:37:35 time: 0.8727 data_time: 0.0021 memory: 28783 grad_norm: 0.0157 loss: 0.6865 +2023/06/01 06:40:39 - mmengine - INFO - Epoch(train) [10][ 300/5758] lr: 5.8244e-04 eta: 14:36:11 time: 0.8344 data_time: 0.0018 memory: 28783 grad_norm: 0.0212 loss: 0.6866 +2023/06/01 06:42:02 - mmengine - INFO - Epoch(train) [10][ 400/5758] lr: 5.8244e-04 eta: 14:34:48 time: 0.8313 data_time: 0.0018 memory: 28783 grad_norm: 0.0151 loss: 0.6868 +2023/06/01 06:43:24 - mmengine - INFO - Epoch(train) [10][ 500/5758] lr: 5.8244e-04 eta: 14:33:22 time: 0.8613 data_time: 0.0019 memory: 28783 grad_norm: 0.0198 loss: 0.6846 +2023/06/01 06:44:46 - mmengine - INFO - Epoch(train) [10][ 600/5758] lr: 5.8244e-04 eta: 14:31:57 time: 0.7988 data_time: 0.0020 memory: 28783 grad_norm: 0.0168 loss: 0.6868 +2023/06/01 06:46:08 - mmengine - INFO - Epoch(train) [10][ 700/5758] lr: 5.8244e-04 eta: 14:30:32 time: 0.8495 data_time: 0.0021 memory: 28783 grad_norm: 0.0210 loss: 0.6888 +2023/06/01 06:47:31 - mmengine - INFO - Epoch(train) [10][ 800/5758] lr: 5.8244e-04 eta: 14:29:08 time: 0.8379 data_time: 0.0021 memory: 28783 grad_norm: 0.0130 loss: 0.6892 +2023/06/01 06:48:52 - mmengine - INFO - Epoch(train) [10][ 900/5758] lr: 5.8244e-04 eta: 14:27:43 time: 0.9098 data_time: 0.0020 memory: 28783 grad_norm: 0.0145 loss: 0.6886 +2023/06/01 06:50:14 - mmengine - INFO - Epoch(train) [10][1000/5758] lr: 5.8244e-04 eta: 14:26:17 time: 0.8264 data_time: 0.0021 memory: 28783 grad_norm: 0.0229 loss: 0.6875 +2023/06/01 06:51:35 - mmengine - INFO - Epoch(train) [10][1100/5758] lr: 5.8244e-04 eta: 14:24:52 time: 0.8170 data_time: 0.0019 memory: 28783 grad_norm: 0.0169 loss: 0.6871 +2023/06/01 06:52:40 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_20230531_183239 +2023/06/01 06:52:57 - mmengine - INFO - Epoch(train) [10][1200/5758] lr: 5.8244e-04 eta: 14:23:26 time: 0.8109 data_time: 0.0021 memory: 28783 grad_norm: 0.0180 loss: 0.6848 +2023/06/01 06:54:20 - mmengine - INFO - Epoch(train) [10][1300/5758] lr: 5.8244e-04 eta: 14:22:02 time: 0.8495 data_time: 0.0018 memory: 28783 grad_norm: 0.0227 loss: 0.6889 +2023/06/01 06:55:43 - mmengine - INFO - Epoch(train) [10][1400/5758] lr: 5.8244e-04 eta: 14:20:38 time: 0.8519 data_time: 0.0032 memory: 28783 grad_norm: 0.0184 loss: 0.6897 +2023/06/01 06:57:05 - mmengine - INFO - Epoch(train) [10][1500/5758] lr: 5.8244e-04 eta: 14:19:13 time: 0.7955 data_time: 0.0020 memory: 28783 grad_norm: 0.0180 loss: 0.6845 +2023/06/01 06:58:27 - mmengine - INFO - Epoch(train) [10][1600/5758] lr: 5.8244e-04 eta: 14:17:48 time: 0.8353 data_time: 0.0026 memory: 28783 grad_norm: 0.0106 loss: 0.6876 +2023/06/01 06:59:49 - mmengine - INFO - Epoch(train) [10][1700/5758] lr: 5.8244e-04 eta: 14:16:23 time: 0.7651 data_time: 0.0019 memory: 28783 grad_norm: 0.0196 loss: 0.6845 +2023/06/01 07:01:10 - mmengine - INFO - Epoch(train) [10][1800/5758] lr: 5.8244e-04 eta: 14:14:57 time: 0.8041 data_time: 0.0019 memory: 28783 grad_norm: 0.0157 loss: 0.6880 +2023/06/01 07:02:31 - mmengine - INFO - Epoch(train) [10][1900/5758] lr: 5.8244e-04 eta: 14:13:32 time: 0.7745 data_time: 0.0018 memory: 28783 grad_norm: 0.0239 loss: 0.6859 +2023/06/01 07:03:54 - mmengine - INFO - Epoch(train) [10][2000/5758] lr: 5.8244e-04 eta: 14:12:08 time: 0.7640 data_time: 0.0018 memory: 28783 grad_norm: 0.0119 loss: 0.6838 +2023/06/01 07:05:17 - mmengine - INFO - Epoch(train) [10][2100/5758] lr: 5.8244e-04 eta: 14:10:44 time: 0.8308 data_time: 0.0017 memory: 28783 grad_norm: 0.0177 loss: 0.6833 +2023/06/01 07:06:20 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_20230531_183239 +2023/06/01 07:06:38 - mmengine - INFO - Epoch(train) [10][2200/5758] lr: 5.8244e-04 eta: 14:09:17 time: 0.8155 data_time: 0.0032 memory: 28783 grad_norm: 0.0171 loss: 0.6826 +2023/06/01 07:08:00 - mmengine - INFO - Epoch(train) [10][2300/5758] lr: 5.8244e-04 eta: 14:07:53 time: 0.8524 data_time: 0.0021 memory: 28783 grad_norm: 0.0175 loss: 0.6864 +2023/06/01 07:09:22 - mmengine - INFO - Epoch(train) [10][2400/5758] lr: 5.8244e-04 eta: 14:06:28 time: 0.8137 data_time: 0.0027 memory: 28783 grad_norm: 0.0168 loss: 0.6876 +2023/06/01 07:10:42 - mmengine - INFO - Epoch(train) [10][2500/5758] lr: 5.8244e-04 eta: 14:05:01 time: 0.7913 data_time: 0.0023 memory: 28783 grad_norm: 0.0141 loss: 0.6849 +2023/06/01 07:12:05 - mmengine - INFO - Epoch(train) [10][2600/5758] lr: 5.8244e-04 eta: 14:03:37 time: 0.8285 data_time: 0.0018 memory: 28783 grad_norm: 0.0125 loss: 0.6896 +2023/06/01 07:13:26 - mmengine - INFO - Epoch(train) [10][2700/5758] lr: 5.8244e-04 eta: 14:02:11 time: 0.7923 data_time: 0.0018 memory: 28783 grad_norm: 0.0257 loss: 0.6871 +2023/06/01 07:14:49 - mmengine - INFO - Epoch(train) [10][2800/5758] lr: 5.8244e-04 eta: 14:00:47 time: 0.8054 data_time: 0.0020 memory: 28783 grad_norm: 0.0194 loss: 0.6867 +2023/06/01 07:16:10 - mmengine - INFO - Epoch(train) [10][2900/5758] lr: 5.8244e-04 eta: 13:59:21 time: 0.8190 data_time: 0.0019 memory: 28783 grad_norm: 0.0123 loss: 0.6860 +2023/06/01 07:17:31 - mmengine - INFO - Epoch(train) [10][3000/5758] lr: 5.8244e-04 eta: 13:57:55 time: 0.7882 data_time: 0.0022 memory: 28783 grad_norm: 0.0186 loss: 0.6885 +2023/06/01 07:18:52 - mmengine - INFO - Epoch(train) [10][3100/5758] lr: 5.8244e-04 eta: 13:56:29 time: 0.8364 data_time: 0.0030 memory: 28783 grad_norm: 0.0128 loss: 0.6892 +2023/06/01 07:19:57 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_20230531_183239 +2023/06/01 07:20:14 - mmengine - INFO - Epoch(train) [10][3200/5758] lr: 5.8244e-04 eta: 13:55:05 time: 0.8116 data_time: 0.0023 memory: 28783 grad_norm: 0.0175 loss: 0.6873 +2023/06/01 07:21:36 - mmengine - INFO - Epoch(train) [10][3300/5758] lr: 5.8244e-04 eta: 13:53:40 time: 0.7922 data_time: 0.0019 memory: 28783 grad_norm: 0.0150 loss: 0.6867 +2023/06/01 07:22:57 - mmengine - INFO - Epoch(train) [10][3400/5758] lr: 5.8244e-04 eta: 13:52:15 time: 0.8037 data_time: 0.0036 memory: 28783 grad_norm: 0.0129 loss: 0.6877 +2023/06/01 07:24:19 - mmengine - INFO - Epoch(train) [10][3500/5758] lr: 5.8244e-04 eta: 13:50:49 time: 0.8377 data_time: 0.0024 memory: 28783 grad_norm: 0.0186 loss: 0.6859 +2023/06/01 07:25:42 - mmengine - INFO - Epoch(train) [10][3600/5758] lr: 5.8244e-04 eta: 13:49:26 time: 0.8563 data_time: 0.0020 memory: 28783 grad_norm: 0.0148 loss: 0.6853 +2023/06/01 07:27:03 - mmengine - INFO - Epoch(train) [10][3700/5758] lr: 5.8244e-04 eta: 13:48:00 time: 0.8015 data_time: 0.0019 memory: 28783 grad_norm: 0.0204 loss: 0.6879 +2023/06/01 07:28:25 - mmengine - INFO - Epoch(train) [10][3800/5758] lr: 5.8244e-04 eta: 13:46:35 time: 0.8275 data_time: 0.0025 memory: 28783 grad_norm: 0.0180 loss: 0.6896 +2023/06/01 07:29:48 - mmengine - INFO - Epoch(train) [10][3900/5758] lr: 5.8244e-04 eta: 13:45:12 time: 0.7865 data_time: 0.0019 memory: 28783 grad_norm: 0.0162 loss: 0.6871 +2023/06/01 07:31:10 - mmengine - INFO - Epoch(train) [10][4000/5758] lr: 5.8244e-04 eta: 13:43:47 time: 0.8425 data_time: 0.0037 memory: 28783 grad_norm: 0.0161 loss: 0.6831 +2023/06/01 07:32:34 - mmengine - INFO - Epoch(train) [10][4100/5758] lr: 5.8244e-04 eta: 13:42:24 time: 0.8078 data_time: 0.0019 memory: 28783 grad_norm: 0.0135 loss: 0.6869 +2023/06/01 07:33:39 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_20230531_183239 +2023/06/01 07:33:57 - mmengine - INFO - Epoch(train) [10][4200/5758] lr: 5.8244e-04 eta: 13:41:01 time: 0.8467 data_time: 0.0018 memory: 28783 grad_norm: 0.0192 loss: 0.6884 +2023/06/01 07:35:19 - mmengine - INFO - Epoch(train) [10][4300/5758] lr: 5.8244e-04 eta: 13:39:36 time: 0.8091 data_time: 0.0019 memory: 28783 grad_norm: 0.0141 loss: 0.6869 +2023/06/01 07:36:41 - mmengine - INFO - Epoch(train) [10][4400/5758] lr: 5.8244e-04 eta: 13:38:11 time: 0.7851 data_time: 0.0019 memory: 28783 grad_norm: 0.0212 loss: 0.6903 +2023/06/01 07:38:03 - mmengine - INFO - Epoch(train) [10][4500/5758] lr: 5.8244e-04 eta: 13:36:47 time: 0.8148 data_time: 0.0018 memory: 28783 grad_norm: 0.0166 loss: 0.6876 +2023/06/01 07:39:28 - mmengine - INFO - Epoch(train) [10][4600/5758] lr: 5.8244e-04 eta: 13:35:25 time: 0.9233 data_time: 0.0027 memory: 28783 grad_norm: 0.0150 loss: 0.6889 +2023/06/01 07:40:51 - mmengine - INFO - Epoch(train) [10][4700/5758] lr: 5.8244e-04 eta: 13:34:02 time: 0.8824 data_time: 0.0024 memory: 28783 grad_norm: 0.0140 loss: 0.6855 +2023/06/01 07:42:13 - mmengine - INFO - Epoch(train) [10][4800/5758] lr: 5.8244e-04 eta: 13:32:37 time: 0.8272 data_time: 0.0024 memory: 28783 grad_norm: 0.0170 loss: 0.6868 +2023/06/01 07:43:35 - mmengine - INFO - Epoch(train) [10][4900/5758] lr: 5.8244e-04 eta: 13:31:13 time: 0.8407 data_time: 0.0026 memory: 28783 grad_norm: 0.0162 loss: 0.6861 +2023/06/01 07:44:57 - mmengine - INFO - Epoch(train) [10][5000/5758] lr: 5.8244e-04 eta: 13:29:48 time: 0.8939 data_time: 0.1457 memory: 28783 grad_norm: 0.0186 loss: 0.6876 +2023/06/01 07:46:18 - mmengine - INFO - Epoch(train) [10][5100/5758] lr: 5.8244e-04 eta: 13:28:23 time: 0.8384 data_time: 0.2385 memory: 28783 grad_norm: 0.0175 loss: 0.6852 +2023/06/01 07:47:24 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_20230531_183239 +2023/06/01 07:47:42 - mmengine - INFO - Epoch(train) [10][5200/5758] lr: 5.8244e-04 eta: 13:27:00 time: 0.8303 data_time: 0.3128 memory: 28783 grad_norm: 0.0214 loss: 0.6882 +2023/06/01 07:49:06 - mmengine - INFO - Epoch(train) [10][5300/5758] lr: 5.8244e-04 eta: 13:25:38 time: 0.8063 data_time: 0.2952 memory: 28783 grad_norm: 0.0170 loss: 0.6846 +2023/06/01 07:50:30 - mmengine - INFO - Epoch(train) [10][5400/5758] lr: 5.8244e-04 eta: 13:24:15 time: 0.7970 data_time: 0.2802 memory: 28783 grad_norm: 0.0218 loss: 0.6815 +2023/06/01 07:51:53 - mmengine - INFO - Epoch(train) [10][5500/5758] lr: 5.8244e-04 eta: 13:22:51 time: 0.7745 data_time: 0.2672 memory: 28783 grad_norm: 0.0182 loss: 0.6878 +2023/06/01 07:53:16 - mmengine - INFO - Epoch(train) [10][5600/5758] lr: 5.8244e-04 eta: 13:21:28 time: 0.7158 data_time: 0.2102 memory: 28783 grad_norm: 0.0231 loss: 0.6849 +2023/06/01 07:54:39 - mmengine - INFO - Epoch(train) [10][5700/5758] lr: 5.8244e-04 eta: 13:20:04 time: 0.8021 data_time: 0.2967 memory: 28783 grad_norm: 0.0196 loss: 0.6895 +2023/06/01 07:55:26 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_20230531_183239 +2023/06/01 07:55:26 - mmengine - INFO - Saving checkpoint at 10 epochs +2023/06/01 07:55:43 - mmengine - INFO - Epoch(val) [10][16/16] accuracy/top1: 100.0000 single-label/precision_classwise: [100.0, 0.0] single-label/recall_classwise: [100.0, 0.0] single-label/f1-score_classwise: [100.0, 0.0] data_time: 0.2220 time: 0.6373 +2023/06/01 07:57:16 - mmengine - INFO - Epoch(train) [11][ 100/5758] lr: 5.0500e-04 eta: 13:18:00 time: 0.8691 data_time: 0.3339 memory: 28783 grad_norm: 0.0142 loss: 0.6861 +2023/06/01 07:58:40 - mmengine - INFO - Epoch(train) [11][ 200/5758] lr: 5.0500e-04 eta: 13:16:37 time: 0.8362 data_time: 0.3303 memory: 28783 grad_norm: 0.0142 loss: 0.6861 +2023/06/01 08:00:05 - mmengine - INFO - Epoch(train) [11][ 300/5758] lr: 5.0500e-04 eta: 13:15:15 time: 0.9236 data_time: 0.3936 memory: 28783 grad_norm: 0.0223 loss: 0.6844 +2023/06/01 08:01:29 - mmengine - INFO - Epoch(train) [11][ 400/5758] lr: 5.0500e-04 eta: 13:13:53 time: 0.8368 data_time: 0.3332 memory: 28783 grad_norm: 0.0168 loss: 0.6895 +2023/06/01 08:01:46 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_20230531_183239 +2023/06/01 08:02:54 - mmengine - INFO - Epoch(train) [11][ 500/5758] lr: 5.0500e-04 eta: 13:12:31 time: 0.8754 data_time: 0.3460 memory: 28783 grad_norm: 0.0123 loss: 0.6850 +2023/06/01 08:04:18 - mmengine - INFO - Epoch(train) [11][ 600/5758] lr: 5.0500e-04 eta: 13:11:09 time: 0.8671 data_time: 0.3626 memory: 28783 grad_norm: 0.0174 loss: 0.6863 +2023/06/01 08:05:42 - mmengine - INFO - Epoch(train) [11][ 700/5758] lr: 5.0500e-04 eta: 13:09:46 time: 0.8383 data_time: 0.3340 memory: 28783 grad_norm: 0.0145 loss: 0.6846 +2023/06/01 08:07:07 - mmengine - INFO - Epoch(train) [11][ 800/5758] lr: 5.0500e-04 eta: 13:08:24 time: 0.8293 data_time: 0.3212 memory: 28783 grad_norm: 0.0156 loss: 0.6831 +2023/06/01 08:08:34 - mmengine - INFO - Epoch(train) [11][ 900/5758] lr: 5.0500e-04 eta: 13:07:04 time: 0.8493 data_time: 0.3451 memory: 28783 grad_norm: 0.0215 loss: 0.6872 +2023/06/01 08:09:59 - mmengine - INFO - Epoch(train) [11][1000/5758] lr: 5.0500e-04 eta: 13:05:42 time: 0.8386 data_time: 0.3325 memory: 28783 grad_norm: 0.0206 loss: 0.6864 +2023/06/01 08:11:23 - mmengine - INFO - Epoch(train) [11][1100/5758] lr: 5.0500e-04 eta: 13:04:20 time: 0.7914 data_time: 0.2867 memory: 28783 grad_norm: 0.0161 loss: 0.6870 +2023/06/01 08:12:48 - mmengine - INFO - Epoch(train) [11][1200/5758] lr: 5.0500e-04 eta: 13:02:59 time: 0.8200 data_time: 0.3149 memory: 28783 grad_norm: 0.0156 loss: 0.6877 +2023/06/01 08:14:13 - mmengine - INFO - Epoch(train) [11][1300/5758] lr: 5.0500e-04 eta: 13:01:37 time: 0.8530 data_time: 0.3462 memory: 28783 grad_norm: 0.0136 loss: 0.6892 +2023/06/01 08:15:38 - mmengine - INFO - Epoch(train) [11][1400/5758] lr: 5.0500e-04 eta: 13:00:15 time: 0.8674 data_time: 0.3644 memory: 28783 grad_norm: 0.0122 loss: 0.6892 +2023/06/01 08:15:55 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_20230531_183239 +2023/06/01 08:17:04 - mmengine - INFO - Epoch(train) [11][1500/5758] lr: 5.0500e-04 eta: 12:58:54 time: 0.8504 data_time: 0.3425 memory: 28783 grad_norm: 0.0164 loss: 0.6893 +2023/06/01 08:18:29 - mmengine - INFO - Epoch(train) [11][1600/5758] lr: 5.0500e-04 eta: 12:57:33 time: 0.8467 data_time: 0.3399 memory: 28783 grad_norm: 0.0152 loss: 0.6894 +2023/06/01 08:19:56 - mmengine - INFO - Epoch(train) [11][1700/5758] lr: 5.0500e-04 eta: 12:56:13 time: 0.8614 data_time: 0.3552 memory: 28783 grad_norm: 0.0179 loss: 0.6898 +2023/06/01 08:21:20 - mmengine - INFO - Epoch(train) [11][1800/5758] lr: 5.0500e-04 eta: 12:54:49 time: 0.8241 data_time: 0.3188 memory: 28783 grad_norm: 0.0197 loss: 0.6828 +2023/06/01 08:22:44 - mmengine - INFO - Epoch(train) [11][1900/5758] lr: 5.0500e-04 eta: 12:53:27 time: 0.8532 data_time: 0.3459 memory: 28783 grad_norm: 0.0159 loss: 0.6862 +2023/06/01 08:24:10 - mmengine - INFO - Epoch(train) [11][2000/5758] lr: 5.0500e-04 eta: 12:52:06 time: 0.8471 data_time: 0.3424 memory: 28783 grad_norm: 0.0152 loss: 0.6873 +2023/06/01 08:25:37 - mmengine - INFO - Epoch(train) [11][2100/5758] lr: 5.0500e-04 eta: 12:50:47 time: 0.8820 data_time: 0.3770 memory: 28783 grad_norm: 0.0225 loss: 0.6905 +2023/06/01 08:27:02 - mmengine - INFO - Epoch(train) [11][2200/5758] lr: 5.0500e-04 eta: 12:49:25 time: 0.8612 data_time: 0.3512 memory: 28783 grad_norm: 0.0119 loss: 0.6868 +2023/06/01 08:28:28 - mmengine - INFO - Epoch(train) [11][2300/5758] lr: 5.0500e-04 eta: 12:48:04 time: 0.8146 data_time: 0.3091 memory: 28783 grad_norm: 0.0257 loss: 0.6893 +2023/06/01 08:29:51 - mmengine - INFO - Epoch(train) [11][2400/5758] lr: 5.0500e-04 eta: 12:46:39 time: 0.8361 data_time: 0.3292 memory: 28783 grad_norm: 0.0218 loss: 0.6894 +2023/06/01 08:30:08 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_20230531_183239 +2023/06/01 08:31:13 - mmengine - INFO - Epoch(train) [11][2500/5758] lr: 5.0500e-04 eta: 12:45:15 time: 0.7857 data_time: 0.2831 memory: 28783 grad_norm: 0.0144 loss: 0.6883 +2023/06/01 08:32:33 - mmengine - INFO - Epoch(train) [11][2600/5758] lr: 5.0500e-04 eta: 12:43:49 time: 0.8101 data_time: 0.1891 memory: 28783 grad_norm: 0.0112 loss: 0.6866 +2023/06/01 08:33:56 - mmengine - INFO - Epoch(train) [11][2700/5758] lr: 5.0500e-04 eta: 12:42:25 time: 0.8152 data_time: 0.0020 memory: 28783 grad_norm: 0.0219 loss: 0.6850 +2023/06/01 08:35:21 - mmengine - INFO - Epoch(train) [11][2800/5758] lr: 5.0500e-04 eta: 12:41:03 time: 0.8374 data_time: 0.0021 memory: 28783 grad_norm: 0.0202 loss: 0.6889 +2023/06/01 08:36:45 - mmengine - INFO - Epoch(train) [11][2900/5758] lr: 5.0500e-04 eta: 12:39:40 time: 0.8345 data_time: 0.0022 memory: 28783 grad_norm: 0.0140 loss: 0.6859 +2023/06/01 08:38:10 - mmengine - INFO - Epoch(train) [11][3000/5758] lr: 5.0500e-04 eta: 12:38:18 time: 0.8692 data_time: 0.0028 memory: 28783 grad_norm: 0.0122 loss: 0.6867 +2023/06/01 08:39:34 - mmengine - INFO - Epoch(train) [11][3100/5758] lr: 5.0500e-04 eta: 12:36:56 time: 0.8334 data_time: 0.0019 memory: 28783 grad_norm: 0.0183 loss: 0.6900 +2023/06/01 08:40:59 - mmengine - INFO - Epoch(train) [11][3200/5758] lr: 5.0500e-04 eta: 12:35:34 time: 0.9126 data_time: 0.0021 memory: 28783 grad_norm: 0.0265 loss: 0.6906 +2023/06/01 08:42:24 - mmengine - INFO - Epoch(train) [11][3300/5758] lr: 5.0500e-04 eta: 12:34:12 time: 0.8225 data_time: 0.0017 memory: 28783 grad_norm: 0.0175 loss: 0.6887 +2023/06/01 08:43:49 - mmengine - INFO - Epoch(train) [11][3400/5758] lr: 5.0500e-04 eta: 12:32:50 time: 0.8365 data_time: 0.0019 memory: 28783 grad_norm: 0.0121 loss: 0.6879 +2023/06/01 08:44:06 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_20230531_183239 +2023/06/01 08:45:15 - mmengine - INFO - Epoch(train) [11][3500/5758] lr: 5.0500e-04 eta: 12:31:29 time: 0.8697 data_time: 0.0019 memory: 28783 grad_norm: 0.0129 loss: 0.6874 +2023/06/01 08:46:42 - mmengine - INFO - Epoch(train) [11][3600/5758] lr: 5.0500e-04 eta: 12:30:08 time: 0.9100 data_time: 0.0021 memory: 28783 grad_norm: 0.0201 loss: 0.6847 +2023/06/01 08:48:06 - mmengine - INFO - Epoch(train) [11][3700/5758] lr: 5.0500e-04 eta: 12:28:46 time: 0.8353 data_time: 0.0018 memory: 28783 grad_norm: 0.0140 loss: 0.6910 +2023/06/01 08:49:32 - mmengine - INFO - Epoch(train) [11][3800/5758] lr: 5.0500e-04 eta: 12:27:25 time: 0.8691 data_time: 0.0018 memory: 28783 grad_norm: 0.0090 loss: 0.6857 +2023/06/01 08:50:58 - mmengine - INFO - Epoch(train) [11][3900/5758] lr: 5.0500e-04 eta: 12:26:03 time: 0.8463 data_time: 0.0022 memory: 28783 grad_norm: 0.0162 loss: 0.6874 +2023/06/01 08:52:24 - mmengine - INFO - Epoch(train) [11][4000/5758] lr: 5.0500e-04 eta: 12:24:43 time: 0.8642 data_time: 0.0020 memory: 28783 grad_norm: 0.0164 loss: 0.6867 +2023/06/01 08:53:49 - mmengine - INFO - Epoch(train) [11][4100/5758] lr: 5.0500e-04 eta: 12:23:20 time: 0.8878 data_time: 0.0020 memory: 28783 grad_norm: 0.0157 loss: 0.6845 +2023/06/01 08:55:13 - mmengine - INFO - Epoch(train) [11][4200/5758] lr: 5.0500e-04 eta: 12:21:58 time: 0.8721 data_time: 0.0020 memory: 28783 grad_norm: 0.0133 loss: 0.6864 +2023/06/01 08:56:39 - mmengine - INFO - Epoch(train) [11][4300/5758] lr: 5.0500e-04 eta: 12:20:36 time: 0.8419 data_time: 0.0020 memory: 28783 grad_norm: 0.0087 loss: 0.6860 +2023/06/01 08:58:04 - mmengine - INFO - Epoch(train) [11][4400/5758] lr: 5.0500e-04 eta: 12:19:14 time: 0.8749 data_time: 0.0019 memory: 28783 grad_norm: 0.0202 loss: 0.6841 +2023/06/01 08:58:20 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_20230531_183239 +2023/06/01 08:59:27 - mmengine - INFO - Epoch(train) [11][4500/5758] lr: 5.0500e-04 eta: 12:17:51 time: 0.8234 data_time: 0.0018 memory: 28783 grad_norm: 0.0061 loss: 0.6893 +2023/06/01 09:00:51 - mmengine - INFO - Epoch(train) [11][4600/5758] lr: 5.0500e-04 eta: 12:16:28 time: 0.7978 data_time: 0.0019 memory: 28783 grad_norm: 0.0125 loss: 0.6862 +2023/06/01 09:02:15 - mmengine - INFO - Epoch(train) [11][4700/5758] lr: 5.0500e-04 eta: 12:15:05 time: 0.8113 data_time: 0.0019 memory: 28783 grad_norm: 0.0118 loss: 0.6867 +2023/06/01 09:03:40 - mmengine - INFO - Epoch(train) [11][4800/5758] lr: 5.0500e-04 eta: 12:13:42 time: 0.8190 data_time: 0.0019 memory: 28783 grad_norm: 0.0145 loss: 0.6854 +2023/06/01 09:05:05 - mmengine - INFO - Epoch(train) [11][4900/5758] lr: 5.0500e-04 eta: 12:12:21 time: 0.8347 data_time: 0.0020 memory: 28783 grad_norm: 0.0149 loss: 0.6880 +2023/06/01 09:06:32 - mmengine - INFO - Epoch(train) [11][5000/5758] lr: 5.0500e-04 eta: 12:11:00 time: 0.8008 data_time: 0.0017 memory: 28783 grad_norm: 0.0179 loss: 0.6860 +2023/06/01 09:07:58 - mmengine - INFO - Epoch(train) [11][5100/5758] lr: 5.0500e-04 eta: 12:09:39 time: 0.8852 data_time: 0.0019 memory: 28783 grad_norm: 0.0214 loss: 0.6883 +2023/06/01 09:09:22 - mmengine - INFO - Epoch(train) [11][5200/5758] lr: 5.0500e-04 eta: 12:08:16 time: 0.7993 data_time: 0.0020 memory: 28783 grad_norm: 0.0184 loss: 0.6874 +2023/06/01 09:10:45 - mmengine - INFO - Epoch(train) [11][5300/5758] lr: 5.0500e-04 eta: 12:06:52 time: 0.8223 data_time: 0.0024 memory: 28783 grad_norm: 0.0130 loss: 0.6872 +2023/06/01 09:12:13 - mmengine - INFO - Epoch(train) [11][5400/5758] lr: 5.0500e-04 eta: 12:05:32 time: 0.9990 data_time: 0.0021 memory: 28783 grad_norm: 0.0130 loss: 0.6860 +2023/06/01 09:12:34 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_20230531_183239 +2023/06/01 09:14:09 - mmengine - INFO - Epoch(train) [11][5500/5758] lr: 5.0500e-04 eta: 12:04:36 time: 0.8524 data_time: 0.0021 memory: 28783 grad_norm: 0.0146 loss: 0.6827 +2023/06/01 09:15:34 - mmengine - INFO - Epoch(train) [11][5600/5758] lr: 5.0500e-04 eta: 12:03:14 time: 0.8890 data_time: 0.0019 memory: 28783 grad_norm: 0.0178 loss: 0.6831 +2023/06/01 09:16:58 - mmengine - INFO - Epoch(train) [11][5700/5758] lr: 5.0500e-04 eta: 12:01:51 time: 0.8503 data_time: 0.0018 memory: 28783 grad_norm: 0.0120 loss: 0.6871 +2023/06/01 09:17:48 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_20230531_183239 +2023/06/01 09:17:48 - mmengine - INFO - Saving checkpoint at 11 epochs +2023/06/01 09:18:05 - mmengine - INFO - Epoch(val) [11][16/16] accuracy/top1: 100.0000 single-label/precision_classwise: [100.0, 0.0] single-label/recall_classwise: [100.0, 0.0] single-label/f1-score_classwise: [100.0, 0.0] data_time: 0.2185 time: 0.6325 +2023/06/01 09:19:33 - mmengine - INFO - Epoch(train) [12][ 100/5758] lr: 4.2756e-04 eta: 11:59:44 time: 0.8019 data_time: 0.0034 memory: 28783 grad_norm: 0.0123 loss: 0.6885 +2023/06/01 09:20:57 - mmengine - INFO - Epoch(train) [12][ 200/5758] lr: 4.2756e-04 eta: 11:58:20 time: 0.8427 data_time: 0.0027 memory: 28783 grad_norm: 0.0127 loss: 0.6879 +2023/06/01 09:22:24 - mmengine - INFO - Epoch(train) [12][ 300/5758] lr: 4.2756e-04 eta: 11:56:59 time: 0.8585 data_time: 0.0020 memory: 28783 grad_norm: 0.0123 loss: 0.6878 +2023/06/01 09:23:48 - mmengine - INFO - Epoch(train) [12][ 400/5758] lr: 4.2756e-04 eta: 11:55:36 time: 0.8231 data_time: 0.0016 memory: 28783 grad_norm: 0.0140 loss: 0.6866 +2023/06/01 09:25:11 - mmengine - INFO - Epoch(train) [12][ 500/5758] lr: 4.2756e-04 eta: 11:54:13 time: 0.8921 data_time: 0.0019 memory: 28783 grad_norm: 0.0119 loss: 0.6863 +2023/06/01 09:26:35 - mmengine - INFO - Epoch(train) [12][ 600/5758] lr: 4.2756e-04 eta: 11:52:50 time: 0.8552 data_time: 0.0022 memory: 28783 grad_norm: 0.0183 loss: 0.6882 +2023/06/01 09:27:27 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_20230531_183239 +2023/06/01 09:28:00 - mmengine - INFO - Epoch(train) [12][ 700/5758] lr: 4.2756e-04 eta: 11:51:27 time: 0.8214 data_time: 0.0023 memory: 28783 grad_norm: 0.0166 loss: 0.6889 +2023/06/01 09:29:23 - mmengine - INFO - Epoch(train) [12][ 800/5758] lr: 4.2756e-04 eta: 11:50:04 time: 0.8742 data_time: 0.0020 memory: 28783 grad_norm: 0.0135 loss: 0.6860 +2023/06/01 09:30:46 - mmengine - INFO - Epoch(train) [12][ 900/5758] lr: 4.2756e-04 eta: 11:48:40 time: 0.8400 data_time: 0.0019 memory: 28783 grad_norm: 0.0192 loss: 0.6848 +2023/06/01 09:32:12 - mmengine - INFO - Epoch(train) [12][1000/5758] lr: 4.2756e-04 eta: 11:47:18 time: 0.9529 data_time: 0.0019 memory: 28783 grad_norm: 0.0139 loss: 0.6886 +2023/06/01 09:33:50 - mmengine - INFO - Epoch(train) [12][1100/5758] lr: 4.2756e-04 eta: 11:46:05 time: 0.7990 data_time: 0.0021 memory: 28783 grad_norm: 0.0195 loss: 0.6856 +2023/06/01 09:35:14 - mmengine - INFO - Epoch(train) [12][1200/5758] lr: 4.2756e-04 eta: 11:44:43 time: 0.8602 data_time: 0.0017 memory: 28783 grad_norm: 0.0096 loss: 0.6870 +2023/06/01 09:36:40 - mmengine - INFO - Epoch(train) [12][1300/5758] lr: 4.2756e-04 eta: 11:43:21 time: 0.8852 data_time: 0.0018 memory: 28783 grad_norm: 0.0104 loss: 0.6858 +2023/06/01 09:38:04 - mmengine - INFO - Epoch(train) [12][1400/5758] lr: 4.2756e-04 eta: 11:41:58 time: 0.8484 data_time: 0.0022 memory: 28783 grad_norm: 0.0156 loss: 0.6859 +2023/06/01 09:39:30 - mmengine - INFO - Epoch(train) [12][1500/5758] lr: 4.2756e-04 eta: 11:40:36 time: 0.8932 data_time: 0.0019 memory: 28783 grad_norm: 0.0123 loss: 0.6875 +2023/06/01 09:40:56 - mmengine - INFO - Epoch(train) [12][1600/5758] lr: 4.2756e-04 eta: 11:39:15 time: 0.8371 data_time: 0.0019 memory: 28783 grad_norm: 0.0253 loss: 0.6850 +2023/06/01 09:41:48 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_20230531_183239 +2023/06/01 09:42:21 - mmengine - INFO - Epoch(train) [12][1700/5758] lr: 4.2756e-04 eta: 11:37:52 time: 0.8798 data_time: 0.0025 memory: 28783 grad_norm: 0.0174 loss: 0.6841 +2023/06/01 09:43:46 - mmengine - INFO - Epoch(train) [12][1800/5758] lr: 4.2756e-04 eta: 11:36:29 time: 0.8409 data_time: 0.0018 memory: 28783 grad_norm: 0.0182 loss: 0.6867 +2023/06/01 09:45:10 - mmengine - INFO - Epoch(train) [12][1900/5758] lr: 4.2756e-04 eta: 11:35:06 time: 0.8279 data_time: 0.0021 memory: 28783 grad_norm: 0.0157 loss: 0.6895 +2023/06/01 09:46:36 - mmengine - INFO - Epoch(train) [12][2000/5758] lr: 4.2756e-04 eta: 11:33:44 time: 0.8910 data_time: 0.0025 memory: 28783 grad_norm: 0.0226 loss: 0.6879 +2023/06/01 09:48:02 - mmengine - INFO - Epoch(train) [12][2100/5758] lr: 4.2756e-04 eta: 11:32:23 time: 0.8668 data_time: 0.0019 memory: 28783 grad_norm: 0.0195 loss: 0.6869 +2023/06/01 09:49:28 - mmengine - INFO - Epoch(train) [12][2200/5758] lr: 4.2756e-04 eta: 11:31:01 time: 0.8440 data_time: 0.0023 memory: 28783 grad_norm: 0.0166 loss: 0.6881 +2023/06/01 09:50:54 - mmengine - INFO - Epoch(train) [12][2300/5758] lr: 4.2756e-04 eta: 11:29:39 time: 0.8423 data_time: 0.0023 memory: 28783 grad_norm: 0.0221 loss: 0.6878 +2023/06/01 09:52:18 - mmengine - INFO - Epoch(train) [12][2400/5758] lr: 4.2756e-04 eta: 11:28:16 time: 0.8344 data_time: 0.0019 memory: 28783 grad_norm: 0.0151 loss: 0.6859 +2023/06/01 09:53:43 - mmengine - INFO - Epoch(train) [12][2500/5758] lr: 4.2756e-04 eta: 11:26:54 time: 0.8059 data_time: 0.0018 memory: 28783 grad_norm: 0.0171 loss: 0.6860 +2023/06/01 09:55:06 - mmengine - INFO - Epoch(train) [12][2600/5758] lr: 4.2756e-04 eta: 11:25:30 time: 0.8589 data_time: 0.0017 memory: 28783 grad_norm: 0.0101 loss: 0.6839 +2023/06/01 09:55:58 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_20230531_183239 +2023/06/01 09:56:32 - mmengine - INFO - Epoch(train) [12][2700/5758] lr: 4.2756e-04 eta: 11:24:08 time: 0.8400 data_time: 0.0016 memory: 28783 grad_norm: 0.0190 loss: 0.6875 +2023/06/01 09:57:56 - mmengine - INFO - Epoch(train) [12][2800/5758] lr: 4.2756e-04 eta: 11:22:45 time: 0.8393 data_time: 0.0019 memory: 28783 grad_norm: 0.0175 loss: 0.6852 +2023/06/01 09:59:20 - mmengine - INFO - Epoch(train) [12][2900/5758] lr: 4.2756e-04 eta: 11:21:22 time: 0.8399 data_time: 0.0018 memory: 28783 grad_norm: 0.0143 loss: 0.6830 +2023/06/01 10:00:46 - mmengine - INFO - Epoch(train) [12][3000/5758] lr: 4.2756e-04 eta: 11:20:00 time: 0.8521 data_time: 0.0019 memory: 28783 grad_norm: 0.0137 loss: 0.6875 +2023/06/01 10:02:12 - mmengine - INFO - Epoch(train) [12][3100/5758] lr: 4.2756e-04 eta: 11:18:38 time: 0.8310 data_time: 0.0023 memory: 28783 grad_norm: 0.0157 loss: 0.6871 +2023/06/01 10:03:38 - mmengine - INFO - Epoch(train) [12][3200/5758] lr: 4.2756e-04 eta: 11:17:17 time: 0.8852 data_time: 0.0019 memory: 28783 grad_norm: 0.0158 loss: 0.6893 +2023/06/01 10:05:03 - mmengine - INFO - Epoch(train) [12][3300/5758] lr: 4.2756e-04 eta: 11:15:54 time: 0.8477 data_time: 0.0021 memory: 28783 grad_norm: 0.0197 loss: 0.6863 +2023/06/01 10:06:27 - mmengine - INFO - Epoch(train) [12][3400/5758] lr: 4.2756e-04 eta: 11:14:30 time: 0.8687 data_time: 0.0028 memory: 28783 grad_norm: 0.0178 loss: 0.6864 +2023/06/01 10:07:50 - mmengine - INFO - Epoch(train) [12][3500/5758] lr: 4.2756e-04 eta: 11:13:06 time: 0.8157 data_time: 0.0020 memory: 28783 grad_norm: 0.0132 loss: 0.6862 +2023/06/01 10:09:16 - mmengine - INFO - Epoch(train) [12][3600/5758] lr: 4.2756e-04 eta: 11:11:45 time: 0.8469 data_time: 0.0026 memory: 28783 grad_norm: 0.0123 loss: 0.6868 +2023/06/01 10:10:09 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_20230531_183239 +2023/06/01 10:10:42 - mmengine - INFO - Epoch(train) [12][3700/5758] lr: 4.2756e-04 eta: 11:10:23 time: 0.8431 data_time: 0.0017 memory: 28783 grad_norm: 0.0189 loss: 0.6843 +2023/06/01 10:12:08 - mmengine - INFO - Epoch(train) [12][3800/5758] lr: 4.2756e-04 eta: 11:09:01 time: 0.8603 data_time: 0.0019 memory: 28783 grad_norm: 0.0168 loss: 0.6841 +2023/06/01 10:13:30 - mmengine - INFO - Epoch(train) [12][3900/5758] lr: 4.2756e-04 eta: 11:07:36 time: 0.8255 data_time: 0.0017 memory: 28783 grad_norm: 0.0179 loss: 0.6876 +2023/06/01 10:14:56 - mmengine - INFO - Epoch(train) [12][4000/5758] lr: 4.2756e-04 eta: 11:06:14 time: 0.9006 data_time: 0.0020 memory: 28783 grad_norm: 0.0168 loss: 0.6865 +2023/06/01 10:16:23 - mmengine - INFO - Epoch(train) [12][4100/5758] lr: 4.2756e-04 eta: 11:04:53 time: 0.8562 data_time: 0.0021 memory: 28783 grad_norm: 0.0189 loss: 0.6834 +2023/06/01 10:17:48 - mmengine - INFO - Epoch(train) [12][4200/5758] lr: 4.2756e-04 eta: 11:03:31 time: 0.8357 data_time: 0.0020 memory: 28783 grad_norm: 0.0227 loss: 0.6891 +2023/06/01 10:19:13 - mmengine - INFO - Epoch(train) [12][4300/5758] lr: 4.2756e-04 eta: 11:02:08 time: 0.8504 data_time: 0.0022 memory: 28783 grad_norm: 0.0136 loss: 0.6874 +2023/06/01 10:20:40 - mmengine - INFO - Epoch(train) [12][4400/5758] lr: 4.2756e-04 eta: 11:00:47 time: 0.8262 data_time: 0.0020 memory: 28783 grad_norm: 0.0183 loss: 0.6885 +2023/06/01 10:22:07 - mmengine - INFO - Epoch(train) [12][4500/5758] lr: 4.2756e-04 eta: 10:59:25 time: 0.8423 data_time: 0.0018 memory: 28783 grad_norm: 0.0166 loss: 0.6845 +2023/06/01 10:23:32 - mmengine - INFO - Epoch(train) [12][4600/5758] lr: 4.2756e-04 eta: 10:58:03 time: 0.8438 data_time: 0.0018 memory: 28783 grad_norm: 0.0189 loss: 0.6824 +2023/06/01 10:24:23 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_20230531_183239 +2023/06/01 10:24:56 - mmengine - INFO - Epoch(train) [12][4700/5758] lr: 4.2756e-04 eta: 10:56:39 time: 0.8112 data_time: 0.0019 memory: 28783 grad_norm: 0.0180 loss: 0.6894 +2023/06/01 10:26:21 - mmengine - INFO - Epoch(train) [12][4800/5758] lr: 4.2756e-04 eta: 10:55:17 time: 0.8221 data_time: 0.0028 memory: 28783 grad_norm: 0.0248 loss: 0.6874 +2023/06/01 10:27:46 - mmengine - INFO - Epoch(train) [12][4900/5758] lr: 4.2756e-04 eta: 10:53:54 time: 0.8419 data_time: 0.0020 memory: 28783 grad_norm: 0.0229 loss: 0.6850 +2023/06/01 10:29:12 - mmengine - INFO - Epoch(train) [12][5000/5758] lr: 4.2756e-04 eta: 10:52:32 time: 0.8444 data_time: 0.0019 memory: 28783 grad_norm: 0.0123 loss: 0.6879 +2023/06/01 10:30:39 - mmengine - INFO - Epoch(train) [12][5100/5758] lr: 4.2756e-04 eta: 10:51:11 time: 0.8816 data_time: 0.0024 memory: 28783 grad_norm: 0.0224 loss: 0.6852 +2023/06/01 10:32:04 - mmengine - INFO - Epoch(train) [12][5200/5758] lr: 4.2756e-04 eta: 10:49:48 time: 0.8663 data_time: 0.0025 memory: 28783 grad_norm: 0.0163 loss: 0.6869 +2023/06/01 10:33:30 - mmengine - INFO - Epoch(train) [12][5300/5758] lr: 4.2756e-04 eta: 10:48:26 time: 0.8967 data_time: 0.0018 memory: 28783 grad_norm: 0.0122 loss: 0.6906 +2023/06/01 10:34:57 - mmengine - INFO - Epoch(train) [12][5400/5758] lr: 4.2756e-04 eta: 10:47:04 time: 0.8886 data_time: 0.0022 memory: 28783 grad_norm: 0.0231 loss: 0.6870 +2023/06/01 10:36:22 - mmengine - INFO - Epoch(train) [12][5500/5758] lr: 4.2756e-04 eta: 10:45:42 time: 0.8600 data_time: 0.0018 memory: 28783 grad_norm: 0.0163 loss: 0.6875 +2023/06/01 10:37:48 - mmengine - INFO - Epoch(train) [12][5600/5758] lr: 4.2756e-04 eta: 10:44:20 time: 0.8383 data_time: 0.0022 memory: 28783 grad_norm: 0.0111 loss: 0.6883 +2023/06/01 10:38:41 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_20230531_183239 +2023/06/01 10:39:15 - mmengine - INFO - Epoch(train) [12][5700/5758] lr: 4.2756e-04 eta: 10:42:58 time: 0.8883 data_time: 0.0026 memory: 28783 grad_norm: 0.0184 loss: 0.6870 +2023/06/01 10:40:06 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_20230531_183239 +2023/06/01 10:40:06 - mmengine - INFO - Saving checkpoint at 12 epochs +2023/06/01 10:40:24 - mmengine - INFO - Epoch(val) [12][16/16] accuracy/top1: 100.0000 single-label/precision_classwise: [100.0, 0.0] single-label/recall_classwise: [100.0, 0.0] single-label/f1-score_classwise: [100.0, 0.0] data_time: 0.2063 time: 0.6214 +2023/06/01 10:41:52 - mmengine - INFO - Epoch(train) [13][ 100/5758] lr: 3.5204e-04 eta: 10:40:51 time: 0.8461 data_time: 0.2392 memory: 28783 grad_norm: 0.0225 loss: 0.6879 +2023/06/01 10:43:18 - mmengine - INFO - Epoch(train) [13][ 200/5758] lr: 3.5204e-04 eta: 10:39:29 time: 0.8045 data_time: 0.0018 memory: 28783 grad_norm: 0.0122 loss: 0.6877 +2023/06/01 10:44:43 - mmengine - INFO - Epoch(train) [13][ 300/5758] lr: 3.5204e-04 eta: 10:38:06 time: 0.8119 data_time: 0.0023 memory: 28783 grad_norm: 0.0194 loss: 0.6865 +2023/06/01 10:46:08 - mmengine - INFO - Epoch(train) [13][ 400/5758] lr: 3.5204e-04 eta: 10:36:43 time: 0.8681 data_time: 0.0031 memory: 28783 grad_norm: 0.0085 loss: 0.6869 +2023/06/01 10:47:33 - mmengine - INFO - Epoch(train) [13][ 500/5758] lr: 3.5204e-04 eta: 10:35:21 time: 0.8630 data_time: 0.0019 memory: 28783 grad_norm: 0.0150 loss: 0.6853 +2023/06/01 10:48:57 - mmengine - INFO - Epoch(train) [13][ 600/5758] lr: 3.5204e-04 eta: 10:33:57 time: 0.8287 data_time: 0.0025 memory: 28783 grad_norm: 0.0196 loss: 0.6859 +2023/06/01 10:50:22 - mmengine - INFO - Epoch(train) [13][ 700/5758] lr: 3.5204e-04 eta: 10:32:34 time: 0.8506 data_time: 0.0021 memory: 28783 grad_norm: 0.0183 loss: 0.6875 +2023/06/01 10:51:46 - mmengine - INFO - Epoch(train) [13][ 800/5758] lr: 3.5204e-04 eta: 10:31:11 time: 0.8006 data_time: 0.0018 memory: 28783 grad_norm: 0.0135 loss: 0.6845 +2023/06/01 10:53:13 - mmengine - INFO - Epoch(train) [13][ 900/5758] lr: 3.5204e-04 eta: 10:29:49 time: 0.8689 data_time: 0.0031 memory: 28783 grad_norm: 0.0173 loss: 0.6884 +2023/06/01 10:53:15 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_20230531_183239 +2023/06/01 10:54:40 - mmengine - INFO - Epoch(train) [13][1000/5758] lr: 3.5204e-04 eta: 10:28:28 time: 0.8936 data_time: 0.0028 memory: 28783 grad_norm: 0.0105 loss: 0.6861 +2023/06/01 10:56:08 - mmengine - INFO - Epoch(train) [13][1100/5758] lr: 3.5204e-04 eta: 10:27:07 time: 0.8652 data_time: 0.0026 memory: 28783 grad_norm: 0.0137 loss: 0.6871 +2023/06/01 10:57:35 - mmengine - INFO - Epoch(train) [13][1200/5758] lr: 3.5204e-04 eta: 10:25:45 time: 0.8800 data_time: 0.0025 memory: 28783 grad_norm: 0.0140 loss: 0.6889 +2023/06/01 10:59:01 - mmengine - INFO - Epoch(train) [13][1300/5758] lr: 3.5204e-04 eta: 10:24:23 time: 0.8606 data_time: 0.0018 memory: 28783 grad_norm: 0.0191 loss: 0.6874 +2023/06/01 11:00:26 - mmengine - INFO - Epoch(train) [13][1400/5758] lr: 3.5204e-04 eta: 10:23:00 time: 0.8789 data_time: 0.0019 memory: 28783 grad_norm: 0.0136 loss: 0.6865 +2023/06/01 11:01:53 - mmengine - INFO - Epoch(train) [13][1500/5758] lr: 3.5204e-04 eta: 10:21:38 time: 0.8228 data_time: 0.0028 memory: 28783 grad_norm: 0.0174 loss: 0.6879 +2023/06/01 11:03:17 - mmengine - INFO - Epoch(train) [13][1600/5758] lr: 3.5204e-04 eta: 10:20:15 time: 0.8581 data_time: 0.0019 memory: 28783 grad_norm: 0.0151 loss: 0.6872 +2023/06/01 11:04:44 - mmengine - INFO - Epoch(train) [13][1700/5758] lr: 3.5204e-04 eta: 10:18:54 time: 0.9177 data_time: 0.0028 memory: 28783 grad_norm: 0.0155 loss: 0.6888 +2023/06/01 11:06:10 - mmengine - INFO - Epoch(train) [13][1800/5758] lr: 3.5204e-04 eta: 10:17:31 time: 0.8287 data_time: 0.0026 memory: 28783 grad_norm: 0.0161 loss: 0.6872 +2023/06/01 11:07:35 - mmengine - INFO - Epoch(train) [13][1900/5758] lr: 3.5204e-04 eta: 10:16:08 time: 0.8193 data_time: 0.0019 memory: 28783 grad_norm: 0.0147 loss: 0.6879 +2023/06/01 11:07:37 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_20230531_183239 +2023/06/01 11:09:00 - mmengine - INFO - Epoch(train) [13][2000/5758] lr: 3.5204e-04 eta: 10:14:45 time: 0.8365 data_time: 0.0027 memory: 28783 grad_norm: 0.0101 loss: 0.6877 +2023/06/01 11:10:27 - mmengine - INFO - Epoch(train) [13][2100/5758] lr: 3.5204e-04 eta: 10:13:23 time: 0.7949 data_time: 0.0021 memory: 28783 grad_norm: 0.0139 loss: 0.6877 +2023/06/01 11:11:54 - mmengine - INFO - Epoch(train) [13][2200/5758] lr: 3.5204e-04 eta: 10:12:02 time: 0.8805 data_time: 0.0025 memory: 28783 grad_norm: 0.0235 loss: 0.6889 +2023/06/01 11:13:20 - mmengine - INFO - Epoch(train) [13][2300/5758] lr: 3.5204e-04 eta: 10:10:39 time: 0.9044 data_time: 0.0021 memory: 28783 grad_norm: 0.0186 loss: 0.6872 +2023/06/01 11:14:49 - mmengine - INFO - Epoch(train) [13][2400/5758] lr: 3.5204e-04 eta: 10:09:19 time: 0.8947 data_time: 0.0039 memory: 28783 grad_norm: 0.0159 loss: 0.6890 +2023/06/01 11:16:17 - mmengine - INFO - Epoch(train) [13][2500/5758] lr: 3.5204e-04 eta: 10:07:58 time: 0.8426 data_time: 0.0026 memory: 28783 grad_norm: 0.0153 loss: 0.6860 +2023/06/01 11:17:43 - mmengine - INFO - Epoch(train) [13][2600/5758] lr: 3.5204e-04 eta: 10:06:36 time: 0.8604 data_time: 0.0029 memory: 28783 grad_norm: 0.0235 loss: 0.6868 +2023/06/01 11:19:09 - mmengine - INFO - Epoch(train) [13][2700/5758] lr: 3.5204e-04 eta: 10:05:13 time: 0.8655 data_time: 0.0026 memory: 28783 grad_norm: 0.0274 loss: 0.6850 +2023/06/01 11:20:35 - mmengine - INFO - Epoch(train) [13][2800/5758] lr: 3.5204e-04 eta: 10:03:51 time: 0.8917 data_time: 0.0026 memory: 28783 grad_norm: 0.0155 loss: 0.6822 +2023/06/01 11:22:03 - mmengine - INFO - Epoch(train) [13][2900/5758] lr: 3.5204e-04 eta: 10:02:30 time: 0.8420 data_time: 0.0030 memory: 28783 grad_norm: 0.0099 loss: 0.6837 +2023/06/01 11:22:05 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_20230531_183239 +2023/06/01 11:23:32 - mmengine - INFO - Epoch(train) [13][3000/5758] lr: 3.5204e-04 eta: 10:01:09 time: 0.8914 data_time: 0.0019 memory: 28783 grad_norm: 0.0101 loss: 0.6852 +2023/06/01 11:25:00 - mmengine - INFO - Epoch(train) [13][3100/5758] lr: 3.5204e-04 eta: 9:59:48 time: 0.8656 data_time: 0.0022 memory: 28783 grad_norm: 0.0191 loss: 0.6895 +2023/06/01 11:26:28 - mmengine - INFO - Epoch(train) [13][3200/5758] lr: 3.5204e-04 eta: 9:58:27 time: 0.9712 data_time: 0.0021 memory: 28783 grad_norm: 0.0148 loss: 0.6904 +2023/06/01 11:27:54 - mmengine - INFO - Epoch(train) [13][3300/5758] lr: 3.5204e-04 eta: 9:57:04 time: 0.8268 data_time: 0.0023 memory: 28783 grad_norm: 0.0226 loss: 0.6879 +2023/06/01 11:29:20 - mmengine - INFO - Epoch(train) [13][3400/5758] lr: 3.5204e-04 eta: 9:55:41 time: 0.8800 data_time: 0.0023 memory: 28783 grad_norm: 0.0252 loss: 0.6871 +2023/06/01 11:30:45 - mmengine - INFO - Epoch(train) [13][3500/5758] lr: 3.5204e-04 eta: 9:54:18 time: 0.8518 data_time: 0.0030 memory: 28783 grad_norm: 0.0183 loss: 0.6880 +2023/06/01 11:32:10 - mmengine - INFO - Epoch(train) [13][3600/5758] lr: 3.5204e-04 eta: 9:52:56 time: 0.9025 data_time: 0.0033 memory: 28783 grad_norm: 0.0154 loss: 0.6872 +2023/06/01 11:33:38 - mmengine - INFO - Epoch(train) [13][3700/5758] lr: 3.5204e-04 eta: 9:51:34 time: 0.8562 data_time: 0.0024 memory: 28783 grad_norm: 0.0133 loss: 0.6877 +2023/06/01 11:35:05 - mmengine - INFO - Epoch(train) [13][3800/5758] lr: 3.5204e-04 eta: 9:50:12 time: 0.8512 data_time: 0.0019 memory: 28783 grad_norm: 0.0125 loss: 0.6886 +2023/06/01 11:36:32 - mmengine - INFO - Epoch(train) [13][3900/5758] lr: 3.5204e-04 eta: 9:48:50 time: 0.8504 data_time: 0.0024 memory: 28783 grad_norm: 0.0178 loss: 0.6906 +2023/06/01 11:36:34 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_20230531_183239 +2023/06/01 11:38:00 - mmengine - INFO - Epoch(train) [13][4000/5758] lr: 3.5204e-04 eta: 9:47:29 time: 0.8618 data_time: 0.0029 memory: 28783 grad_norm: 0.0221 loss: 0.6833 +2023/06/01 11:39:28 - mmengine - INFO - Epoch(train) [13][4100/5758] lr: 3.5204e-04 eta: 9:46:07 time: 0.8837 data_time: 0.0031 memory: 28783 grad_norm: 0.0144 loss: 0.6848 +2023/06/01 11:40:55 - mmengine - INFO - Epoch(train) [13][4200/5758] lr: 3.5204e-04 eta: 9:44:45 time: 0.8541 data_time: 0.0033 memory: 28783 grad_norm: 0.0137 loss: 0.6870 +2023/06/01 11:42:21 - mmengine - INFO - Epoch(train) [13][4300/5758] lr: 3.5204e-04 eta: 9:43:23 time: 0.8271 data_time: 0.0024 memory: 28783 grad_norm: 0.0125 loss: 0.6887 +2023/06/01 11:43:42 - mmengine - INFO - Epoch(train) [13][4400/5758] lr: 3.5204e-04 eta: 9:41:57 time: 0.8426 data_time: 0.0025 memory: 28783 grad_norm: 0.0124 loss: 0.6899 +2023/06/01 11:45:05 - mmengine - INFO - Epoch(train) [13][4500/5758] lr: 3.5204e-04 eta: 9:40:33 time: 0.8379 data_time: 0.0025 memory: 28783 grad_norm: 0.0182 loss: 0.6853 +2023/06/01 11:46:27 - mmengine - INFO - Epoch(train) [13][4600/5758] lr: 3.5204e-04 eta: 9:39:09 time: 0.7893 data_time: 0.0025 memory: 28783 grad_norm: 0.0123 loss: 0.6819 +2023/06/01 11:47:50 - mmengine - INFO - Epoch(train) [13][4700/5758] lr: 3.5204e-04 eta: 9:37:44 time: 0.8151 data_time: 0.0027 memory: 28783 grad_norm: 0.0234 loss: 0.6811 +2023/06/01 11:49:12 - mmengine - INFO - Epoch(train) [13][4800/5758] lr: 3.5204e-04 eta: 9:36:19 time: 0.8243 data_time: 0.0024 memory: 28783 grad_norm: 0.0126 loss: 0.6849 +2023/06/01 11:50:34 - mmengine - INFO - Epoch(train) [13][4900/5758] lr: 3.5204e-04 eta: 9:34:54 time: 0.8251 data_time: 0.0019 memory: 28783 grad_norm: 0.0122 loss: 0.6851 +2023/06/01 11:50:36 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_20230531_183239 +2023/06/01 11:51:58 - mmengine - INFO - Epoch(train) [13][5000/5758] lr: 3.5204e-04 eta: 9:33:31 time: 0.8068 data_time: 0.0021 memory: 28783 grad_norm: 0.0170 loss: 0.6880 +2023/06/01 11:53:21 - mmengine - INFO - Epoch(train) [13][5100/5758] lr: 3.5204e-04 eta: 9:32:06 time: 0.8281 data_time: 0.0027 memory: 28783 grad_norm: 0.0138 loss: 0.6903 +2023/06/01 11:54:43 - mmengine - INFO - Epoch(train) [13][5200/5758] lr: 3.5204e-04 eta: 9:30:42 time: 0.7985 data_time: 0.0025 memory: 28783 grad_norm: 0.0156 loss: 0.6899 +2023/06/01 11:56:04 - mmengine - INFO - Epoch(train) [13][5300/5758] lr: 3.5204e-04 eta: 9:29:17 time: 0.8126 data_time: 0.0021 memory: 28783 grad_norm: 0.0183 loss: 0.6861 +2023/06/01 11:57:27 - mmengine - INFO - Epoch(train) [13][5400/5758] lr: 3.5204e-04 eta: 9:27:52 time: 0.7871 data_time: 0.0021 memory: 28783 grad_norm: 0.0177 loss: 0.6882 +2023/06/01 11:58:47 - mmengine - INFO - Epoch(train) [13][5500/5758] lr: 3.5204e-04 eta: 9:26:26 time: 0.7920 data_time: 0.0023 memory: 28783 grad_norm: 0.0154 loss: 0.6873 +2023/06/01 12:00:09 - mmengine - INFO - Epoch(train) [13][5600/5758] lr: 3.5204e-04 eta: 9:25:01 time: 0.8064 data_time: 0.0027 memory: 28783 grad_norm: 0.0137 loss: 0.6894 +2023/06/01 12:01:28 - mmengine - INFO - Epoch(train) [13][5700/5758] lr: 3.5204e-04 eta: 9:23:35 time: 0.7716 data_time: 0.0025 memory: 28783 grad_norm: 0.0187 loss: 0.6883 +2023/06/01 12:02:17 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_20230531_183239 +2023/06/01 12:02:17 - mmengine - INFO - Saving checkpoint at 13 epochs +2023/06/01 12:02:35 - mmengine - INFO - Epoch(val) [13][16/16] accuracy/top1: 100.0000 single-label/precision_classwise: [100.0, 0.0] single-label/recall_classwise: [100.0, 0.0] single-label/f1-score_classwise: [100.0, 0.0] data_time: 0.2167 time: 0.6322 +2023/06/01 12:03:59 - mmengine - INFO - Epoch(train) [14][ 100/5758] lr: 2.8027e-04 eta: 9:21:23 time: 0.7668 data_time: 0.0031 memory: 28783 grad_norm: 0.0214 loss: 0.6845 +2023/06/01 12:04:38 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_20230531_183239 +2023/06/01 12:05:19 - mmengine - INFO - Epoch(train) [14][ 200/5758] lr: 2.8027e-04 eta: 9:19:57 time: 0.7944 data_time: 0.1835 memory: 28783 grad_norm: 0.0123 loss: 0.6864 +2023/06/01 12:06:41 - mmengine - INFO - Epoch(train) [14][ 300/5758] lr: 2.8027e-04 eta: 9:18:33 time: 0.8210 data_time: 0.0026 memory: 28783 grad_norm: 0.0120 loss: 0.6865 +2023/06/01 12:08:02 - mmengine - INFO - Epoch(train) [14][ 400/5758] lr: 2.8027e-04 eta: 9:17:07 time: 0.8174 data_time: 0.0021 memory: 28783 grad_norm: 0.0158 loss: 0.6852 +2023/06/01 12:09:24 - mmengine - INFO - Epoch(train) [14][ 500/5758] lr: 2.8027e-04 eta: 9:15:42 time: 0.9005 data_time: 0.0024 memory: 28783 grad_norm: 0.0256 loss: 0.6867 +2023/06/01 12:10:44 - mmengine - INFO - Epoch(train) [14][ 600/5758] lr: 2.8027e-04 eta: 9:14:17 time: 0.7948 data_time: 0.0029 memory: 28783 grad_norm: 0.0106 loss: 0.6882 +2023/06/01 12:12:05 - mmengine - INFO - Epoch(train) [14][ 700/5758] lr: 2.8027e-04 eta: 9:12:52 time: 0.8507 data_time: 0.0023 memory: 28783 grad_norm: 0.0136 loss: 0.6898 +2023/06/01 12:13:27 - mmengine - INFO - Epoch(train) [14][ 800/5758] lr: 2.8027e-04 eta: 9:11:27 time: 0.8167 data_time: 0.0029 memory: 28783 grad_norm: 0.0178 loss: 0.6882 +2023/06/01 12:14:49 - mmengine - INFO - Epoch(train) [14][ 900/5758] lr: 2.8027e-04 eta: 9:10:02 time: 0.7838 data_time: 0.0029 memory: 28783 grad_norm: 0.0155 loss: 0.6885 +2023/06/01 12:16:12 - mmengine - INFO - Epoch(train) [14][1000/5758] lr: 2.8027e-04 eta: 9:08:38 time: 0.8248 data_time: 0.0032 memory: 28783 grad_norm: 0.0147 loss: 0.6863 +2023/06/01 12:17:32 - mmengine - INFO - Epoch(train) [14][1100/5758] lr: 2.8027e-04 eta: 9:07:12 time: 0.7866 data_time: 0.0022 memory: 28783 grad_norm: 0.0144 loss: 0.6890 +2023/06/01 12:18:07 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_20230531_183239 +2023/06/01 12:18:54 - mmengine - INFO - Epoch(train) [14][1200/5758] lr: 2.8027e-04 eta: 9:05:48 time: 0.8043 data_time: 0.0028 memory: 28783 grad_norm: 0.0172 loss: 0.6866 +2023/06/01 12:20:18 - mmengine - INFO - Epoch(train) [14][1300/5758] lr: 2.8027e-04 eta: 9:04:24 time: 0.8166 data_time: 0.0024 memory: 28783 grad_norm: 0.0155 loss: 0.6835 +2023/06/01 12:21:43 - mmengine - INFO - Epoch(train) [14][1400/5758] lr: 2.8027e-04 eta: 9:03:01 time: 0.8313 data_time: 0.0026 memory: 28783 grad_norm: 0.0144 loss: 0.6863 +2023/06/01 12:23:05 - mmengine - INFO - Epoch(train) [14][1500/5758] lr: 2.8027e-04 eta: 9:01:36 time: 0.8546 data_time: 0.0026 memory: 28783 grad_norm: 0.0088 loss: 0.6860 +2023/06/01 12:24:29 - mmengine - INFO - Epoch(train) [14][1600/5758] lr: 2.8027e-04 eta: 9:00:13 time: 0.8058 data_time: 0.0028 memory: 28783 grad_norm: 0.0105 loss: 0.6870 +2023/06/01 12:25:50 - mmengine - INFO - Epoch(train) [14][1700/5758] lr: 2.8027e-04 eta: 8:58:47 time: 0.7848 data_time: 0.0023 memory: 28783 grad_norm: 0.0222 loss: 0.6882 +2023/06/01 12:27:12 - mmengine - INFO - Epoch(train) [14][1800/5758] lr: 2.8027e-04 eta: 8:57:23 time: 0.8897 data_time: 0.0024 memory: 28783 grad_norm: 0.0112 loss: 0.6867 +2023/06/01 12:28:35 - mmengine - INFO - Epoch(train) [14][1900/5758] lr: 2.8027e-04 eta: 8:55:59 time: 0.7789 data_time: 0.0029 memory: 28783 grad_norm: 0.0124 loss: 0.6870 +2023/06/01 12:29:55 - mmengine - INFO - Epoch(train) [14][2000/5758] lr: 2.8027e-04 eta: 8:54:33 time: 0.8136 data_time: 0.0033 memory: 28783 grad_norm: 0.0167 loss: 0.6875 +2023/06/01 12:31:17 - mmengine - INFO - Epoch(train) [14][2100/5758] lr: 2.8027e-04 eta: 8:53:08 time: 0.8345 data_time: 0.0030 memory: 28783 grad_norm: 0.0173 loss: 0.6875 +2023/06/01 12:31:52 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_20230531_183239 +2023/06/01 12:32:37 - mmengine - INFO - Epoch(train) [14][2200/5758] lr: 2.8027e-04 eta: 8:51:43 time: 0.8361 data_time: 0.0028 memory: 28783 grad_norm: 0.0144 loss: 0.6866 +2023/06/01 12:34:00 - mmengine - INFO - Epoch(train) [14][2300/5758] lr: 2.8027e-04 eta: 8:50:19 time: 0.7899 data_time: 0.0029 memory: 28783 grad_norm: 0.0133 loss: 0.6875 +2023/06/01 12:35:20 - mmengine - INFO - Epoch(train) [14][2400/5758] lr: 2.8027e-04 eta: 8:48:53 time: 0.7778 data_time: 0.0021 memory: 28783 grad_norm: 0.0206 loss: 0.6844 +2023/06/01 12:36:40 - mmengine - INFO - Epoch(train) [14][2500/5758] lr: 2.8027e-04 eta: 8:47:28 time: 0.8321 data_time: 0.0029 memory: 28783 grad_norm: 0.0199 loss: 0.6862 +2023/06/01 12:37:59 - mmengine - INFO - Epoch(train) [14][2600/5758] lr: 2.8027e-04 eta: 8:46:02 time: 0.7844 data_time: 0.0022 memory: 28783 grad_norm: 0.0172 loss: 0.6848 +2023/06/01 12:39:19 - mmengine - INFO - Epoch(train) [14][2700/5758] lr: 2.8027e-04 eta: 8:44:36 time: 0.8000 data_time: 0.0031 memory: 28783 grad_norm: 0.0128 loss: 0.6837 +2023/06/01 12:40:41 - mmengine - INFO - Epoch(train) [14][2800/5758] lr: 2.8027e-04 eta: 8:43:12 time: 0.7850 data_time: 0.0020 memory: 28783 grad_norm: 0.0080 loss: 0.6877 +2023/06/01 12:42:02 - mmengine - INFO - Epoch(train) [14][2900/5758] lr: 2.8027e-04 eta: 8:41:47 time: 0.8241 data_time: 0.0025 memory: 28783 grad_norm: 0.0185 loss: 0.6870 +2023/06/01 12:43:24 - mmengine - INFO - Epoch(train) [14][3000/5758] lr: 2.8027e-04 eta: 8:40:22 time: 0.8788 data_time: 0.0021 memory: 28783 grad_norm: 0.0144 loss: 0.6889 +2023/06/01 12:44:44 - mmengine - INFO - Epoch(train) [14][3100/5758] lr: 2.8027e-04 eta: 8:38:57 time: 0.8197 data_time: 0.0020 memory: 28783 grad_norm: 0.0212 loss: 0.6894 +2023/06/01 12:45:20 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_20230531_183239 +2023/06/01 12:46:06 - mmengine - INFO - Epoch(train) [14][3200/5758] lr: 2.8027e-04 eta: 8:37:32 time: 0.8351 data_time: 0.0024 memory: 28783 grad_norm: 0.0150 loss: 0.6881 +2023/06/01 12:47:27 - mmengine - INFO - Epoch(train) [14][3300/5758] lr: 2.8027e-04 eta: 8:36:08 time: 0.8262 data_time: 0.0022 memory: 28783 grad_norm: 0.0164 loss: 0.6865 +2023/06/01 12:48:49 - mmengine - INFO - Epoch(train) [14][3400/5758] lr: 2.8027e-04 eta: 8:34:43 time: 0.8359 data_time: 0.0018 memory: 28783 grad_norm: 0.0137 loss: 0.6877 +2023/06/01 12:50:12 - mmengine - INFO - Epoch(train) [14][3500/5758] lr: 2.8027e-04 eta: 8:33:19 time: 0.8375 data_time: 0.0022 memory: 28783 grad_norm: 0.0127 loss: 0.6852 +2023/06/01 12:51:33 - mmengine - INFO - Epoch(train) [14][3600/5758] lr: 2.8027e-04 eta: 8:31:54 time: 0.8217 data_time: 0.0019 memory: 28783 grad_norm: 0.0090 loss: 0.6861 +2023/06/01 12:52:57 - mmengine - INFO - Epoch(train) [14][3700/5758] lr: 2.8027e-04 eta: 8:30:31 time: 0.8031 data_time: 0.0020 memory: 28783 grad_norm: 0.0122 loss: 0.6854 +2023/06/01 12:54:18 - mmengine - INFO - Epoch(train) [14][3800/5758] lr: 2.8027e-04 eta: 8:29:06 time: 0.7952 data_time: 0.0026 memory: 28783 grad_norm: 0.0142 loss: 0.6872 +2023/06/01 12:55:39 - mmengine - INFO - Epoch(train) [14][3900/5758] lr: 2.8027e-04 eta: 8:27:41 time: 0.8175 data_time: 0.0027 memory: 28783 grad_norm: 0.0171 loss: 0.6861 +2023/06/01 12:57:02 - mmengine - INFO - Epoch(train) [14][4000/5758] lr: 2.8027e-04 eta: 8:26:16 time: 0.7848 data_time: 0.0017 memory: 28783 grad_norm: 0.0083 loss: 0.6911 +2023/06/01 12:58:24 - mmengine - INFO - Epoch(train) [14][4100/5758] lr: 2.8027e-04 eta: 8:24:52 time: 0.7926 data_time: 0.0019 memory: 28783 grad_norm: 0.0117 loss: 0.6882 +2023/06/01 12:58:59 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_20230531_183239 +2023/06/01 12:59:46 - mmengine - INFO - Epoch(train) [14][4200/5758] lr: 2.8027e-04 eta: 8:23:28 time: 0.8272 data_time: 0.0021 memory: 28783 grad_norm: 0.0084 loss: 0.6892 +2023/06/01 13:01:07 - mmengine - INFO - Epoch(train) [14][4300/5758] lr: 2.8027e-04 eta: 8:22:03 time: 0.8088 data_time: 0.0022 memory: 28783 grad_norm: 0.0247 loss: 0.6865 +2023/06/01 13:02:33 - mmengine - INFO - Epoch(train) [14][4400/5758] lr: 2.8027e-04 eta: 8:20:40 time: 0.8304 data_time: 0.0017 memory: 28783 grad_norm: 0.0161 loss: 0.6868 +2023/06/01 13:03:56 - mmengine - INFO - Epoch(train) [14][4500/5758] lr: 2.8027e-04 eta: 8:19:16 time: 0.8130 data_time: 0.0021 memory: 28783 grad_norm: 0.0124 loss: 0.6881 +2023/06/01 13:05:17 - mmengine - INFO - Epoch(train) [14][4600/5758] lr: 2.8027e-04 eta: 8:17:51 time: 0.8191 data_time: 0.0029 memory: 28783 grad_norm: 0.0203 loss: 0.6883 +2023/06/01 13:06:40 - mmengine - INFO - Epoch(train) [14][4700/5758] lr: 2.8027e-04 eta: 8:16:27 time: 0.8153 data_time: 0.0019 memory: 28783 grad_norm: 0.0263 loss: 0.6887 +2023/06/01 13:08:03 - mmengine - INFO - Epoch(train) [14][4800/5758] lr: 2.8027e-04 eta: 8:15:03 time: 0.7950 data_time: 0.0030 memory: 28783 grad_norm: 0.0153 loss: 0.6874 +2023/06/01 13:09:26 - mmengine - INFO - Epoch(train) [14][4900/5758] lr: 2.8027e-04 eta: 8:13:39 time: 0.8027 data_time: 0.0033 memory: 28783 grad_norm: 0.0147 loss: 0.6864 +2023/06/01 13:10:49 - mmengine - INFO - Epoch(train) [14][5000/5758] lr: 2.8027e-04 eta: 8:12:16 time: 0.8853 data_time: 0.0020 memory: 28783 grad_norm: 0.0160 loss: 0.6837 +2023/06/01 13:12:11 - mmengine - INFO - Epoch(train) [14][5100/5758] lr: 2.8027e-04 eta: 8:10:51 time: 0.8559 data_time: 0.0019 memory: 28783 grad_norm: 0.0169 loss: 0.6865 +2023/06/01 13:12:48 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_20230531_183239 +2023/06/01 13:13:34 - mmengine - INFO - Epoch(train) [14][5200/5758] lr: 2.8027e-04 eta: 8:09:27 time: 0.8053 data_time: 0.0029 memory: 28783 grad_norm: 0.0194 loss: 0.6888 +2023/06/01 13:14:59 - mmengine - INFO - Epoch(train) [14][5300/5758] lr: 2.8027e-04 eta: 8:08:04 time: 0.8283 data_time: 0.0022 memory: 28783 grad_norm: 0.0199 loss: 0.6870 +2023/06/01 13:16:23 - mmengine - INFO - Epoch(train) [14][5400/5758] lr: 2.8027e-04 eta: 8:06:41 time: 0.9100 data_time: 0.0021 memory: 28783 grad_norm: 0.0176 loss: 0.6845 +2023/06/01 13:17:58 - mmengine - INFO - Epoch(train) [14][5500/5758] lr: 2.8027e-04 eta: 8:05:22 time: 0.7959 data_time: 0.0019 memory: 28783 grad_norm: 0.0156 loss: 0.6870 +2023/06/01 13:19:21 - mmengine - INFO - Epoch(train) [14][5600/5758] lr: 2.8027e-04 eta: 8:03:58 time: 0.8478 data_time: 0.0030 memory: 28783 grad_norm: 0.0148 loss: 0.6898 +2023/06/01 13:20:44 - mmengine - INFO - Epoch(train) [14][5700/5758] lr: 2.8027e-04 eta: 8:02:34 time: 0.7931 data_time: 0.0036 memory: 28783 grad_norm: 0.0133 loss: 0.6880 +2023/06/01 13:21:32 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_20230531_183239 +2023/06/01 13:21:32 - mmengine - INFO - Saving checkpoint at 14 epochs +2023/06/01 13:21:50 - mmengine - INFO - Epoch(val) [14][16/16] accuracy/top1: 100.0000 single-label/precision_classwise: [100.0, 0.0] single-label/recall_classwise: [100.0, 0.0] single-label/f1-score_classwise: [100.0, 0.0] data_time: 0.2232 time: 0.6496 +2023/06/01 13:23:17 - mmengine - INFO - Epoch(train) [15][ 100/5758] lr: 2.1405e-04 eta: 8:00:23 time: 0.7656 data_time: 0.0026 memory: 28783 grad_norm: 0.0135 loss: 0.6872 +2023/06/01 13:24:38 - mmengine - INFO - Epoch(train) [15][ 200/5758] lr: 2.1405e-04 eta: 7:58:58 time: 0.7880 data_time: 0.0019 memory: 28783 grad_norm: 0.0176 loss: 0.6848 +2023/06/01 13:26:01 - mmengine - INFO - Epoch(train) [15][ 300/5758] lr: 2.1405e-04 eta: 7:57:34 time: 0.8316 data_time: 0.0020 memory: 28783 grad_norm: 0.0173 loss: 0.6914 +2023/06/01 13:27:19 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_20230531_183239 +2023/06/01 13:27:30 - mmengine - INFO - Epoch(train) [15][ 400/5758] lr: 2.1405e-04 eta: 7:56:12 time: 0.9419 data_time: 0.0022 memory: 28783 grad_norm: 0.0168 loss: 0.6873 +2023/06/01 13:28:53 - mmengine - INFO - Epoch(train) [15][ 500/5758] lr: 2.1405e-04 eta: 7:54:49 time: 0.8436 data_time: 0.0019 memory: 28783 grad_norm: 0.0156 loss: 0.6823 +2023/06/01 13:30:19 - mmengine - INFO - Epoch(train) [15][ 600/5758] lr: 2.1405e-04 eta: 7:53:26 time: 0.8085 data_time: 0.0026 memory: 28783 grad_norm: 0.0223 loss: 0.6870 +2023/06/01 13:31:43 - mmengine - INFO - Epoch(train) [15][ 700/5758] lr: 2.1405e-04 eta: 7:52:02 time: 0.8183 data_time: 0.0026 memory: 28783 grad_norm: 0.0192 loss: 0.6875 +2023/06/01 13:33:07 - mmengine - INFO - Epoch(train) [15][ 800/5758] lr: 2.1405e-04 eta: 7:50:39 time: 0.8187 data_time: 0.0023 memory: 28783 grad_norm: 0.0143 loss: 0.6901 +2023/06/01 13:34:31 - mmengine - INFO - Epoch(train) [15][ 900/5758] lr: 2.1405e-04 eta: 7:49:15 time: 0.8031 data_time: 0.0023 memory: 28783 grad_norm: 0.0185 loss: 0.6867 +2023/06/01 13:35:55 - mmengine - INFO - Epoch(train) [15][1000/5758] lr: 2.1405e-04 eta: 7:47:51 time: 0.8651 data_time: 0.0019 memory: 28783 grad_norm: 0.0134 loss: 0.6871 +2023/06/01 13:37:19 - mmengine - INFO - Epoch(train) [15][1100/5758] lr: 2.1405e-04 eta: 7:46:28 time: 0.8802 data_time: 0.0028 memory: 28783 grad_norm: 0.0235 loss: 0.6875 +2023/06/01 13:38:42 - mmengine - INFO - Epoch(train) [15][1200/5758] lr: 2.1405e-04 eta: 7:45:04 time: 0.8107 data_time: 0.0027 memory: 28783 grad_norm: 0.0168 loss: 0.6885 +2023/06/01 13:40:05 - mmengine - INFO - Epoch(train) [15][1300/5758] lr: 2.1405e-04 eta: 7:43:40 time: 0.8388 data_time: 0.0022 memory: 28783 grad_norm: 0.0138 loss: 0.6855 +2023/06/01 13:41:19 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_20230531_183239 +2023/06/01 13:41:29 - mmengine - INFO - Epoch(train) [15][1400/5758] lr: 2.1405e-04 eta: 7:42:16 time: 0.8324 data_time: 0.0020 memory: 28783 grad_norm: 0.0146 loss: 0.6831 +2023/06/01 13:42:51 - mmengine - INFO - Epoch(train) [15][1500/5758] lr: 2.1405e-04 eta: 7:40:52 time: 0.8172 data_time: 0.0022 memory: 28783 grad_norm: 0.0187 loss: 0.6869 +2023/06/01 13:44:15 - mmengine - INFO - Epoch(train) [15][1600/5758] lr: 2.1405e-04 eta: 7:39:29 time: 0.8382 data_time: 0.0020 memory: 28783 grad_norm: 0.0206 loss: 0.6866 +2023/06/01 13:45:37 - mmengine - INFO - Epoch(train) [15][1700/5758] lr: 2.1405e-04 eta: 7:38:04 time: 0.8150 data_time: 0.0027 memory: 28783 grad_norm: 0.0244 loss: 0.6868 +2023/06/01 13:47:01 - mmengine - INFO - Epoch(train) [15][1800/5758] lr: 2.1405e-04 eta: 7:36:41 time: 0.8041 data_time: 0.0021 memory: 28783 grad_norm: 0.0147 loss: 0.6865 +2023/06/01 13:48:25 - mmengine - INFO - Epoch(train) [15][1900/5758] lr: 2.1405e-04 eta: 7:35:17 time: 0.8241 data_time: 0.0026 memory: 28783 grad_norm: 0.0178 loss: 0.6849 +2023/06/01 13:49:48 - mmengine - INFO - Epoch(train) [15][2000/5758] lr: 2.1405e-04 eta: 7:33:53 time: 0.7964 data_time: 0.0019 memory: 28783 grad_norm: 0.0121 loss: 0.6855 +2023/06/01 13:51:12 - mmengine - INFO - Epoch(train) [15][2100/5758] lr: 2.1405e-04 eta: 7:32:30 time: 0.8318 data_time: 0.0030 memory: 28783 grad_norm: 0.0205 loss: 0.6836 +2023/06/01 13:52:34 - mmengine - INFO - Epoch(train) [15][2200/5758] lr: 2.1405e-04 eta: 7:31:06 time: 0.8170 data_time: 0.0024 memory: 28783 grad_norm: 0.0150 loss: 0.6870 +2023/06/01 13:53:57 - mmengine - INFO - Epoch(train) [15][2300/5758] lr: 2.1405e-04 eta: 7:29:42 time: 0.8418 data_time: 0.0020 memory: 28783 grad_norm: 0.0108 loss: 0.6867 +2023/06/01 13:55:11 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_20230531_183239 +2023/06/01 13:55:21 - mmengine - INFO - Epoch(train) [15][2400/5758] lr: 2.1405e-04 eta: 7:28:18 time: 0.8383 data_time: 0.0018 memory: 28783 grad_norm: 0.0105 loss: 0.6853 +2023/06/01 13:56:45 - mmengine - INFO - Epoch(train) [15][2500/5758] lr: 2.1405e-04 eta: 7:26:54 time: 0.8452 data_time: 0.0038 memory: 28783 grad_norm: 0.0200 loss: 0.6870 +2023/06/01 13:58:07 - mmengine - INFO - Epoch(train) [15][2600/5758] lr: 2.1405e-04 eta: 7:25:30 time: 0.8324 data_time: 0.0025 memory: 28783 grad_norm: 0.0214 loss: 0.6880 +2023/06/01 13:59:32 - mmengine - INFO - Epoch(train) [15][2700/5758] lr: 2.1405e-04 eta: 7:24:07 time: 0.8288 data_time: 0.0021 memory: 28783 grad_norm: 0.0176 loss: 0.6861 +2023/06/01 14:00:54 - mmengine - INFO - Epoch(train) [15][2800/5758] lr: 2.1405e-04 eta: 7:22:43 time: 0.7987 data_time: 0.0022 memory: 28783 grad_norm: 0.0094 loss: 0.6863 +2023/06/01 14:02:16 - mmengine - INFO - Epoch(train) [15][2900/5758] lr: 2.1405e-04 eta: 7:21:18 time: 0.7948 data_time: 0.0024 memory: 28783 grad_norm: 0.0146 loss: 0.6835 +2023/06/01 14:03:40 - mmengine - INFO - Epoch(train) [15][3000/5758] lr: 2.1405e-04 eta: 7:19:55 time: 0.8145 data_time: 0.0019 memory: 28783 grad_norm: 0.0144 loss: 0.6898 +2023/06/01 14:05:03 - mmengine - INFO - Epoch(train) [15][3100/5758] lr: 2.1405e-04 eta: 7:18:31 time: 0.7987 data_time: 0.0019 memory: 28783 grad_norm: 0.0164 loss: 0.6870 +2023/06/01 14:06:25 - mmengine - INFO - Epoch(train) [15][3200/5758] lr: 2.1405e-04 eta: 7:17:06 time: 0.8342 data_time: 0.0018 memory: 28783 grad_norm: 0.0117 loss: 0.6853 +2023/06/01 14:07:48 - mmengine - INFO - Epoch(train) [15][3300/5758] lr: 2.1405e-04 eta: 7:15:42 time: 0.8202 data_time: 0.0019 memory: 28783 grad_norm: 0.0202 loss: 0.6887 +2023/06/01 14:09:03 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_20230531_183239 +2023/06/01 14:09:12 - mmengine - INFO - Epoch(train) [15][3400/5758] lr: 2.1405e-04 eta: 7:14:19 time: 0.8667 data_time: 0.0032 memory: 28783 grad_norm: 0.0168 loss: 0.6890 +2023/06/01 14:10:35 - mmengine - INFO - Epoch(train) [15][3500/5758] lr: 2.1405e-04 eta: 7:12:55 time: 0.7882 data_time: 0.0020 memory: 28783 grad_norm: 0.0200 loss: 0.6905 +2023/06/01 14:11:57 - mmengine - INFO - Epoch(train) [15][3600/5758] lr: 2.1405e-04 eta: 7:11:31 time: 0.8020 data_time: 0.0019 memory: 28783 grad_norm: 0.0214 loss: 0.6856 +2023/06/01 14:13:21 - mmengine - INFO - Epoch(train) [15][3700/5758] lr: 2.1405e-04 eta: 7:10:07 time: 0.8266 data_time: 0.0021 memory: 28783 grad_norm: 0.0163 loss: 0.6845 +2023/06/01 14:14:44 - mmengine - INFO - Epoch(train) [15][3800/5758] lr: 2.1405e-04 eta: 7:08:44 time: 0.8584 data_time: 0.0023 memory: 28783 grad_norm: 0.0116 loss: 0.6859 +2023/06/01 14:16:07 - mmengine - INFO - Epoch(train) [15][3900/5758] lr: 2.1405e-04 eta: 7:07:20 time: 0.8427 data_time: 0.0030 memory: 28783 grad_norm: 0.0160 loss: 0.6879 +2023/06/01 14:17:30 - mmengine - INFO - Epoch(train) [15][4000/5758] lr: 2.1405e-04 eta: 7:05:56 time: 0.8642 data_time: 0.0028 memory: 28783 grad_norm: 0.0146 loss: 0.6870 +2023/06/01 14:18:54 - mmengine - INFO - Epoch(train) [15][4100/5758] lr: 2.1405e-04 eta: 7:04:32 time: 0.8754 data_time: 0.0032 memory: 28783 grad_norm: 0.0172 loss: 0.6882 +2023/06/01 14:20:17 - mmengine - INFO - Epoch(train) [15][4200/5758] lr: 2.1405e-04 eta: 7:03:08 time: 0.8871 data_time: 0.0019 memory: 28783 grad_norm: 0.0152 loss: 0.6874 +2023/06/01 14:21:43 - mmengine - INFO - Epoch(train) [15][4300/5758] lr: 2.1405e-04 eta: 7:01:45 time: 0.8672 data_time: 0.0020 memory: 28783 grad_norm: 0.0154 loss: 0.6843 +2023/06/01 14:23:00 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_20230531_183239 +2023/06/01 14:23:09 - mmengine - INFO - Epoch(train) [15][4400/5758] lr: 2.1405e-04 eta: 7:00:23 time: 0.8680 data_time: 0.0019 memory: 28783 grad_norm: 0.0269 loss: 0.6863 +2023/06/01 14:24:33 - mmengine - INFO - Epoch(train) [15][4500/5758] lr: 2.1405e-04 eta: 6:58:59 time: 0.8587 data_time: 0.0022 memory: 28783 grad_norm: 0.0151 loss: 0.6934 +2023/06/01 14:25:56 - mmengine - INFO - Epoch(train) [15][4600/5758] lr: 2.1405e-04 eta: 6:57:35 time: 0.7982 data_time: 0.0026 memory: 28783 grad_norm: 0.0171 loss: 0.6858 +2023/06/01 14:27:21 - mmengine - INFO - Epoch(train) [15][4700/5758] lr: 2.1405e-04 eta: 6:56:12 time: 0.8567 data_time: 0.0024 memory: 28783 grad_norm: 0.0291 loss: 0.6863 +2023/06/01 14:28:49 - mmengine - INFO - Epoch(train) [15][4800/5758] lr: 2.1405e-04 eta: 6:54:50 time: 0.8862 data_time: 0.0019 memory: 28783 grad_norm: 0.0172 loss: 0.6862 +2023/06/01 14:30:17 - mmengine - INFO - Epoch(train) [15][4900/5758] lr: 2.1405e-04 eta: 6:53:28 time: 0.9209 data_time: 0.0026 memory: 28783 grad_norm: 0.0137 loss: 0.6857 +2023/06/01 14:31:46 - mmengine - INFO - Epoch(train) [15][5000/5758] lr: 2.1405e-04 eta: 6:52:06 time: 0.8701 data_time: 0.0019 memory: 28783 grad_norm: 0.0134 loss: 0.6882 +2023/06/01 14:33:14 - mmengine - INFO - Epoch(train) [15][5100/5758] lr: 2.1405e-04 eta: 6:50:43 time: 0.9163 data_time: 0.0024 memory: 28783 grad_norm: 0.0113 loss: 0.6889 +2023/06/01 14:34:43 - mmengine - INFO - Epoch(train) [15][5200/5758] lr: 2.1405e-04 eta: 6:49:22 time: 0.9291 data_time: 0.0019 memory: 28783 grad_norm: 0.0216 loss: 0.6858 +2023/06/01 14:36:12 - mmengine - INFO - Epoch(train) [15][5300/5758] lr: 2.1405e-04 eta: 6:48:00 time: 0.8792 data_time: 0.0024 memory: 28783 grad_norm: 0.0095 loss: 0.6900 +2023/06/01 14:37:29 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_20230531_183239 +2023/06/01 14:37:38 - mmengine - INFO - Epoch(train) [15][5400/5758] lr: 2.1405e-04 eta: 6:46:37 time: 0.8106 data_time: 0.0020 memory: 28783 grad_norm: 0.0195 loss: 0.6853 +2023/06/01 14:39:03 - mmengine - INFO - Epoch(train) [15][5500/5758] lr: 2.1405e-04 eta: 6:45:13 time: 0.7689 data_time: 0.0025 memory: 28783 grad_norm: 0.0104 loss: 0.6898 +2023/06/01 14:40:28 - mmengine - INFO - Epoch(train) [15][5600/5758] lr: 2.1405e-04 eta: 6:43:50 time: 0.8461 data_time: 0.0023 memory: 28783 grad_norm: 0.0159 loss: 0.6871 +2023/06/01 14:41:52 - mmengine - INFO - Epoch(train) [15][5700/5758] lr: 2.1405e-04 eta: 6:42:27 time: 0.8235 data_time: 0.0020 memory: 28783 grad_norm: 0.0148 loss: 0.6844 +2023/06/01 14:42:43 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_20230531_183239 +2023/06/01 14:42:43 - mmengine - INFO - Saving checkpoint at 15 epochs +2023/06/01 14:43:01 - mmengine - INFO - Epoch(val) [15][16/16] accuracy/top1: 100.0000 single-label/precision_classwise: [100.0, 0.0] single-label/recall_classwise: [100.0, 0.0] single-label/f1-score_classwise: [100.0, 0.0] data_time: 0.2366 time: 0.6512 +2023/06/01 14:44:28 - mmengine - INFO - Epoch(train) [16][ 100/5758] lr: 1.5498e-04 eta: 6:40:16 time: 0.8516 data_time: 0.2266 memory: 28783 grad_norm: 0.0166 loss: 0.6876 +2023/06/01 14:45:49 - mmengine - INFO - Epoch(train) [16][ 200/5758] lr: 1.5498e-04 eta: 6:38:52 time: 0.8205 data_time: 0.0023 memory: 28783 grad_norm: 0.0129 loss: 0.6883 +2023/06/01 14:47:14 - mmengine - INFO - Epoch(train) [16][ 300/5758] lr: 1.5498e-04 eta: 6:37:28 time: 0.8687 data_time: 0.0022 memory: 28783 grad_norm: 0.0149 loss: 0.6885 +2023/06/01 14:48:36 - mmengine - INFO - Epoch(train) [16][ 400/5758] lr: 1.5498e-04 eta: 6:36:04 time: 0.8202 data_time: 0.0022 memory: 28783 grad_norm: 0.0197 loss: 0.6858 +2023/06/01 14:49:59 - mmengine - INFO - Epoch(train) [16][ 500/5758] lr: 1.5498e-04 eta: 6:34:40 time: 0.8295 data_time: 0.0020 memory: 28783 grad_norm: 0.0169 loss: 0.6844 +2023/06/01 14:51:20 - mmengine - INFO - Epoch(train) [16][ 600/5758] lr: 1.5498e-04 eta: 6:33:15 time: 0.7651 data_time: 0.0018 memory: 28783 grad_norm: 0.0198 loss: 0.6868 +2023/06/01 14:51:44 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_20230531_183239 +2023/06/01 14:52:41 - mmengine - INFO - Epoch(train) [16][ 700/5758] lr: 1.5498e-04 eta: 6:31:51 time: 0.8584 data_time: 0.0024 memory: 28783 grad_norm: 0.0135 loss: 0.6890 +2023/06/01 14:54:02 - mmengine - INFO - Epoch(train) [16][ 800/5758] lr: 1.5498e-04 eta: 6:30:26 time: 0.7579 data_time: 0.0032 memory: 28783 grad_norm: 0.0253 loss: 0.6882 +2023/06/01 14:55:25 - mmengine - INFO - Epoch(train) [16][ 900/5758] lr: 1.5498e-04 eta: 6:29:02 time: 0.8440 data_time: 0.0020 memory: 28783 grad_norm: 0.0124 loss: 0.6889 +2023/06/01 14:56:44 - mmengine - INFO - Epoch(train) [16][1000/5758] lr: 1.5498e-04 eta: 6:27:37 time: 0.7685 data_time: 0.0029 memory: 28783 grad_norm: 0.0151 loss: 0.6898 +2023/06/01 14:58:05 - mmengine - INFO - Epoch(train) [16][1100/5758] lr: 1.5498e-04 eta: 6:26:13 time: 0.7332 data_time: 0.0020 memory: 28783 grad_norm: 0.0178 loss: 0.6849 +2023/06/01 14:59:25 - mmengine - INFO - Epoch(train) [16][1200/5758] lr: 1.5498e-04 eta: 6:24:48 time: 0.7905 data_time: 0.0020 memory: 28783 grad_norm: 0.0114 loss: 0.6870 +2023/06/01 15:00:45 - mmengine - INFO - Epoch(train) [16][1300/5758] lr: 1.5498e-04 eta: 6:23:23 time: 0.7859 data_time: 0.0022 memory: 28783 grad_norm: 0.0137 loss: 0.6884 +2023/06/01 15:02:02 - mmengine - INFO - Epoch(train) [16][1400/5758] lr: 1.5498e-04 eta: 6:21:57 time: 0.7414 data_time: 0.0020 memory: 28783 grad_norm: 0.0231 loss: 0.6866 +2023/06/01 15:03:20 - mmengine - INFO - Epoch(train) [16][1500/5758] lr: 1.5498e-04 eta: 6:20:32 time: 0.7456 data_time: 0.0018 memory: 28783 grad_norm: 0.0206 loss: 0.6874 +2023/06/01 15:04:39 - mmengine - INFO - Epoch(train) [16][1600/5758] lr: 1.5498e-04 eta: 6:19:07 time: 0.8247 data_time: 0.0019 memory: 28783 grad_norm: 0.0165 loss: 0.6879 +2023/06/01 15:05:03 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_20230531_183239 +2023/06/01 15:05:59 - mmengine - INFO - Epoch(train) [16][1700/5758] lr: 1.5498e-04 eta: 6:17:42 time: 0.7488 data_time: 0.0019 memory: 28783 grad_norm: 0.0126 loss: 0.6881 +2023/06/01 15:07:19 - mmengine - INFO - Epoch(train) [16][1800/5758] lr: 1.5498e-04 eta: 6:16:17 time: 0.8109 data_time: 0.0038 memory: 28783 grad_norm: 0.0155 loss: 0.6866 +2023/06/01 15:08:40 - mmengine - INFO - Epoch(train) [16][1900/5758] lr: 1.5498e-04 eta: 6:14:53 time: 0.8090 data_time: 0.0018 memory: 28783 grad_norm: 0.0180 loss: 0.6867 diff --git a/swin_small_4xb256_1e-3lr_5m/20230531_183239/vis_data/20230531_183239.json b/swin_small_4xb256_1e-3lr_5m/20230531_183239/vis_data/20230531_183239.json new file mode 100644 index 0000000000000000000000000000000000000000..9138437d0403f42823af9db0b01021b12629110e --- /dev/null +++ b/swin_small_4xb256_1e-3lr_5m/20230531_183239/vis_data/20230531_183239.json @@ -0,0 +1,889 @@ +{"lr": 0.001, "data_time": 0.33590447902679443, "grad_norm": 1.187183529138565, "loss": 0.6847400009632111, "time": 0.8413348197937012, "epoch": 1, "memory": 28798, "step": 100} +{"lr": 0.001, "data_time": 0.21662020683288574, "grad_norm": 1.7217679768800735, "loss": 0.6908431828022004, "time": 0.8396015882492065, "epoch": 1, "memory": 28798, "step": 200} +{"lr": 0.001, "data_time": 0.002046799659729004, "grad_norm": 0.7791802749037743, "loss": 0.685979676246643, "time": 0.8650855302810669, "epoch": 1, "memory": 28798, "step": 300} +{"lr": 0.001, "data_time": 0.0023226499557495116, "grad_norm": 1.8430660009384154, "loss": 0.6948411226272583, "time": 0.9164520978927613, "epoch": 1, "memory": 28798, "step": 400} +{"lr": 0.001, "data_time": 0.0018687725067138671, "grad_norm": 0.6885813616216183, "loss": 0.6911965548992157, "time": 0.9577422380447388, "epoch": 1, "memory": 28798, "step": 500} +{"lr": 0.001, "data_time": 0.0020292520523071287, "grad_norm": 0.6634811967611313, "loss": 0.6864281952381134, "time": 0.8940370321273804, "epoch": 1, "memory": 28798, "step": 600} +{"lr": 0.001, "data_time": 0.0021838903427124023, "grad_norm": 0.6898281827569008, "loss": 0.6847228586673737, "time": 0.9111965894699097, "epoch": 1, "memory": 28798, "step": 700} +{"lr": 0.001, "data_time": 0.001630878448486328, "grad_norm": 1.3765226341784, "loss": 0.6887052834033967, "time": 0.943936800956726, "epoch": 1, "memory": 28798, "step": 800} +{"lr": 0.001, "data_time": 0.0016585111618041993, "grad_norm": 0.8592354223132134, "loss": 0.6897273421287536, "time": 0.9013750314712524, "epoch": 1, "memory": 28798, "step": 900} +{"lr": 0.001, "data_time": 0.0018284320831298828, "grad_norm": 0.518747566640377, "loss": 0.6888404786586761, "time": 1.1501312017440797, "epoch": 1, "memory": 28798, "step": 1000} +{"lr": 0.001, "data_time": 0.0016209840774536132, "grad_norm": 0.5395256511867046, "loss": 0.688236290216446, "time": 1.0908938884735107, "epoch": 1, "memory": 28798, "step": 1100} +{"lr": 0.001, "data_time": 0.0016809463500976562, "grad_norm": 0.9257483318448066, "loss": 0.6897540330886841, "time": 1.7967471361160279, "epoch": 1, "memory": 28798, "step": 1200} +{"lr": 0.001, "data_time": 0.001671147346496582, "grad_norm": 0.3737223416566849, "loss": 0.6879740357398987, "time": 0.9751446723937989, "epoch": 1, "memory": 28798, "step": 1300} +{"lr": 0.001, "data_time": 0.0016964435577392577, "grad_norm": 0.7635870240628719, "loss": 0.690438163280487, "time": 0.8699488162994384, "epoch": 1, "memory": 28798, "step": 1400} +{"lr": 0.001, "data_time": 0.0015594482421875, "grad_norm": 0.4468700304627419, "loss": 0.6883688926696777, "time": 0.9833091259002685, "epoch": 1, "memory": 28798, "step": 1500} +{"lr": 0.001, "data_time": 0.001657557487487793, "grad_norm": 0.21296481229364872, "loss": 0.6855560719966889, "time": 0.8429322481155396, "epoch": 1, "memory": 28798, "step": 1600} +{"lr": 0.001, "data_time": 0.001414346694946289, "grad_norm": 0.3191097557544708, "loss": 0.6899114906787872, "time": 0.9023676156997681, "epoch": 1, "memory": 28798, "step": 1700} +{"lr": 0.001, "data_time": 0.0015766620635986328, "grad_norm": 0.38395332768559454, "loss": 0.6843568980693817, "time": 0.9407166719436646, "epoch": 1, "memory": 28798, "step": 1800} +{"lr": 0.001, "data_time": 0.0015876293182373047, "grad_norm": 0.20893616713583468, "loss": 0.6874257266521454, "time": 0.903137493133545, "epoch": 1, "memory": 28798, "step": 1900} +{"lr": 0.001, "data_time": 0.0015334606170654297, "grad_norm": 0.29613580070436, "loss": 0.6882599294185638, "time": 1.0162176609039306, "epoch": 1, "memory": 28798, "step": 2000} +{"lr": 0.001, "data_time": 0.0019433021545410157, "grad_norm": 0.23353887908160686, "loss": 0.6890283942222595, "time": 0.967284870147705, "epoch": 1, "memory": 28798, "step": 2100} +{"lr": 0.001, "data_time": 0.0016798734664916991, "grad_norm": 0.27702537178993225, "loss": 0.6872062087059021, "time": 1.1158102035522461, "epoch": 1, "memory": 28798, "step": 2200} +{"lr": 0.001, "data_time": 0.0017933845520019531, "grad_norm": 0.16620225086808205, "loss": 0.6852505743503571, "time": 1.0250746726989746, "epoch": 1, "memory": 28798, "step": 2300} +{"lr": 0.001, "data_time": 0.0013074159622192382, "grad_norm": 0.16672469899058343, "loss": 0.6863704442977905, "time": 0.9309884548187256, "epoch": 1, "memory": 28798, "step": 2400} +{"lr": 0.001, "data_time": 0.0020125389099121095, "grad_norm": 0.28441546857357025, "loss": 0.6850416421890259, "time": 0.855058741569519, "epoch": 1, "memory": 28798, "step": 2500} +{"lr": 0.001, "data_time": 0.0016599655151367187, "grad_norm": 0.2358134739100933, "loss": 0.686743825674057, "time": 0.9001273393630982, "epoch": 1, "memory": 28798, "step": 2600} +{"lr": 0.001, "data_time": 0.0015526533126831055, "grad_norm": 0.26860551312565806, "loss": 0.6846187889575959, "time": 0.8460203647613526, "epoch": 1, "memory": 28798, "step": 2700} +{"lr": 0.001, "data_time": 0.0017335653305053712, "grad_norm": 0.18999040648341178, "loss": 0.6850415289402008, "time": 0.9114830255508423, "epoch": 1, "memory": 28798, "step": 2800} +{"lr": 0.001, "data_time": 0.001981043815612793, "grad_norm": 0.27065986320376395, "loss": 0.6882188379764557, "time": 0.9323644161224365, "epoch": 1, "memory": 28798, "step": 2900} +{"lr": 0.001, "data_time": 0.0016637325286865234, "grad_norm": 0.17620864287018775, "loss": 0.6837712287902832, "time": 0.8893620491027832, "epoch": 1, "memory": 28798, "step": 3000} +{"lr": 0.001, "data_time": 0.0017996549606323243, "grad_norm": 0.1975460097193718, "loss": 0.6860647797584534, "time": 0.997292423248291, "epoch": 1, "memory": 28798, "step": 3100} +{"lr": 0.001, "data_time": 0.0017212152481079102, "grad_norm": 0.18541229702532291, "loss": 0.6901660084724426, "time": 0.8814517498016358, "epoch": 1, "memory": 28798, "step": 3200} +{"lr": 0.001, "data_time": 0.001845979690551758, "grad_norm": 0.09229977168142796, "loss": 0.6833701193332672, "time": 0.9323680400848389, "epoch": 1, "memory": 28798, "step": 3300} +{"lr": 0.001, "data_time": 0.001603388786315918, "grad_norm": 0.19935780689120292, "loss": 0.6885235249996186, "time": 1.0112058401107789, "epoch": 1, "memory": 28798, "step": 3400} +{"lr": 0.001, "data_time": 0.001819014549255371, "grad_norm": 0.16067650243639947, "loss": 0.6889486908912659, "time": 0.895413875579834, "epoch": 1, "memory": 28798, "step": 3500} +{"lr": 0.001, "data_time": 0.002013564109802246, "grad_norm": 0.14938931372016667, "loss": 0.6931148529052734, "time": 0.9706980943679809, "epoch": 1, "memory": 28798, "step": 3600} +{"lr": 0.001, "data_time": 0.0018714666366577148, "grad_norm": 0.12052799835801124, "loss": 0.6882530331611634, "time": 0.9076480388641357, "epoch": 1, "memory": 28798, "step": 3700} +{"lr": 0.001, "data_time": 0.0018136739730834962, "grad_norm": 0.13805863428860904, "loss": 0.6835283875465393, "time": 0.7688174724578858, "epoch": 1, "memory": 28798, "step": 3800} +{"lr": 0.001, "data_time": 0.0017467737197875977, "grad_norm": 0.18656774386763572, "loss": 0.6844374060630798, "time": 0.8669900178909302, "epoch": 1, "memory": 28798, "step": 3900} +{"lr": 0.001, "data_time": 0.0016543149948120117, "grad_norm": 0.18251721002161503, "loss": 0.6876611471176147, "time": 0.8675621271133422, "epoch": 1, "memory": 28798, "step": 4000} +{"lr": 0.001, "data_time": 0.0021095514297485352, "grad_norm": 0.12997964397072792, "loss": 0.6865510106086731, "time": 0.7381432294845581, "epoch": 1, "memory": 28798, "step": 4100} +{"lr": 0.001, "data_time": 0.001603245735168457, "grad_norm": 0.13632102496922016, "loss": 0.6867913782596589, "time": 0.7909459352493287, "epoch": 1, "memory": 28798, "step": 4200} +{"lr": 0.001, "data_time": 0.0017872333526611328, "grad_norm": 0.13055530302226542, "loss": 0.6876545310020447, "time": 0.8414000749588013, "epoch": 1, "memory": 28798, "step": 4300} +{"lr": 0.001, "data_time": 0.001677727699279785, "grad_norm": 0.7684229619801044, "loss": 0.6813175678253174, "time": 0.8178308010101318, "epoch": 1, "memory": 28798, "step": 4400} +{"lr": 0.001, "data_time": 0.0015346765518188476, "grad_norm": 0.153719025850296, "loss": 0.6841858744621276, "time": 0.8224682569503784, "epoch": 1, "memory": 28798, "step": 4500} +{"lr": 0.001, "data_time": 0.0015207052230834961, "grad_norm": 0.10010228492319584, "loss": 0.6867010712623596, "time": 0.8346450090408325, "epoch": 1, "memory": 28798, "step": 4600} +{"lr": 0.001, "data_time": 0.0015330553054809571, "grad_norm": 0.09172356463968753, "loss": 0.6880784034729004, "time": 0.7683933496475219, "epoch": 1, "memory": 28798, "step": 4700} +{"lr": 0.001, "data_time": 0.0015619754791259765, "grad_norm": 0.09726331569254398, "loss": 0.6853234350681305, "time": 0.8191373586654663, "epoch": 1, "memory": 28798, "step": 4800} +{"lr": 0.001, "data_time": 0.0017874717712402343, "grad_norm": 0.10995372105389833, "loss": 0.6863053739070892, "time": 0.7955687046051025, "epoch": 1, "memory": 28798, "step": 4900} +{"lr": 0.001, "data_time": 0.0015886068344116212, "grad_norm": 0.16352200247347354, "loss": 0.6862053871154785, "time": 0.8013665676116943, "epoch": 1, "memory": 28798, "step": 5000} +{"lr": 0.001, "data_time": 0.0019288778305053711, "grad_norm": 0.12119524236768484, "loss": 0.6847285449504852, "time": 0.7832417964935303, "epoch": 1, "memory": 28798, "step": 5100} +{"lr": 0.001, "data_time": 0.0018793821334838867, "grad_norm": 0.1053498338907957, "loss": 0.6855199933052063, "time": 0.7722936630249023, "epoch": 1, "memory": 28798, "step": 5200} +{"lr": 0.001, "data_time": 0.001998138427734375, "grad_norm": 0.11447365432977677, "loss": 0.6891373336315155, "time": 0.839335298538208, "epoch": 1, "memory": 28798, "step": 5300} +{"lr": 0.001, "data_time": 0.002456045150756836, "grad_norm": 0.10331161096692085, "loss": 0.6884491860866546, "time": 0.8279985427856446, "epoch": 1, "memory": 28798, "step": 5400} +{"lr": 0.001, "data_time": 0.0018787860870361328, "grad_norm": 0.13169370582327247, "loss": 0.6850335359573364, "time": 0.8244468450546265, "epoch": 1, "memory": 28798, "step": 5500} +{"lr": 0.001, "data_time": 0.0016954421997070312, "grad_norm": 0.09026205968111753, "loss": 0.6849363386631012, "time": 0.8310160636901855, "epoch": 1, "memory": 28798, "step": 5600} +{"lr": 0.001, "data_time": 0.0016336441040039062, "grad_norm": 0.1151382902637124, "loss": 0.6874211132526398, "time": 0.8125093698501586, "epoch": 1, "memory": 28798, "step": 5700} +{"accuracy/top1": 100.0, "data_time": 0.2904546409845352, "time": 0.8415422290563583, "step": 1} +{"lr": 0.0009939057285945933, "data_time": 0.11489784717559814, "grad_norm": 0.1119145980104804, "loss": 0.6851836323738099, "time": 0.7385829448699951, "epoch": 2, "memory": 28798, "step": 5858} +{"lr": 0.0009939057285945933, "data_time": 0.23536550998687744, "grad_norm": 0.12526991479098798, "loss": 0.6892807424068451, "time": 0.7840444326400757, "epoch": 2, "memory": 28783, "step": 5958} +{"lr": 0.0009939057285945933, "data_time": 0.001851940155029297, "grad_norm": 0.09118716344237328, "loss": 0.6850068032741546, "time": 0.8325890302658081, "epoch": 2, "memory": 28783, "step": 6058} +{"lr": 0.0009939057285945933, "data_time": 0.0017272472381591798, "grad_norm": 0.08466822886839509, "loss": 0.6854560792446136, "time": 0.8943645715713501, "epoch": 2, "memory": 28783, "step": 6158} +{"lr": 0.0009939057285945933, "data_time": 0.0020479917526245116, "grad_norm": 0.09755303636193276, "loss": 0.689743059873581, "time": 0.8275129079818726, "epoch": 2, "memory": 28783, "step": 6258} +{"lr": 0.0009939057285945933, "data_time": 0.002652263641357422, "grad_norm": 0.06423695608973504, "loss": 0.6862825572490692, "time": 0.9148306846618652, "epoch": 2, "memory": 28783, "step": 6358} +{"lr": 0.0009939057285945933, "data_time": 0.0020377397537231444, "grad_norm": 0.09123450815677643, "loss": 0.6879670560359955, "time": 0.8873496294021607, "epoch": 2, "memory": 28783, "step": 6458} +{"lr": 0.0009939057285945933, "data_time": 0.00245974063873291, "grad_norm": 0.1289809938520193, "loss": 0.68823601603508, "time": 0.8818986654281616, "epoch": 2, "memory": 28783, "step": 6558} +{"lr": 0.0009939057285945933, "data_time": 0.032562541961669925, "grad_norm": 0.1374927183613181, "loss": 0.6854689180850982, "time": 0.858594012260437, "epoch": 2, "memory": 28783, "step": 6658} +{"lr": 0.0009939057285945933, "data_time": 0.0013387680053710937, "grad_norm": 0.09858463630080223, "loss": 0.6865781843662262, "time": 0.8951478958129883, "epoch": 2, "memory": 28783, "step": 6758} +{"lr": 0.0009939057285945933, "data_time": 0.0018634557723999023, "grad_norm": 0.1305597260594368, "loss": 0.6850895404815673, "time": 0.7789987325668335, "epoch": 2, "memory": 28783, "step": 6858} +{"lr": 0.0009939057285945933, "data_time": 0.0016655206680297851, "grad_norm": 0.06447077486664057, "loss": 0.68461953997612, "time": 0.9761712312698364, "epoch": 2, "memory": 28783, "step": 6958} +{"lr": 0.0009939057285945933, "data_time": 0.0025378704071044923, "grad_norm": 0.06561579406261445, "loss": 0.6870323240756988, "time": 0.8225229740142822, "epoch": 2, "memory": 28783, "step": 7058} +{"lr": 0.0009939057285945933, "data_time": 0.0016849040985107422, "grad_norm": 0.0958001771941781, "loss": 0.6843083381652832, "time": 0.8733738422393799, "epoch": 2, "memory": 28783, "step": 7158} +{"lr": 0.0009939057285945933, "data_time": 0.0023676395416259766, "grad_norm": 0.13181908074766396, "loss": 0.6872279942035675, "time": 0.8258849382400513, "epoch": 2, "memory": 28783, "step": 7258} +{"lr": 0.0009939057285945933, "data_time": 0.002006077766418457, "grad_norm": 0.0896741697564721, "loss": 0.6851314842700958, "time": 0.8508948564529419, "epoch": 2, "memory": 28783, "step": 7358} +{"lr": 0.0009939057285945933, "data_time": 0.002572154998779297, "grad_norm": 0.09456936344504356, "loss": 0.6865829825401306, "time": 0.8397979736328125, "epoch": 2, "memory": 28783, "step": 7458} +{"lr": 0.0009939057285945933, "data_time": 0.0017290353775024415, "grad_norm": 0.04374804254621267, "loss": 0.684164959192276, "time": 0.8577231884002685, "epoch": 2, "memory": 28783, "step": 7558} +{"lr": 0.0009939057285945933, "data_time": 0.0018528461456298827, "grad_norm": 0.10155281480401754, "loss": 0.688321304321289, "time": 0.9475823163986206, "epoch": 2, "memory": 28783, "step": 7658} +{"lr": 0.0009939057285945933, "data_time": 0.0017137765884399415, "grad_norm": 0.05001990171149373, "loss": 0.6900282025337219, "time": 0.851217246055603, "epoch": 2, "memory": 28783, "step": 7758} +{"lr": 0.0009939057285945933, "data_time": 0.0016271352767944336, "grad_norm": 0.10887768808752299, "loss": 0.6876216173171997, "time": 0.880221962928772, "epoch": 2, "memory": 28783, "step": 7858} +{"lr": 0.0009939057285945933, "data_time": 0.0015771150588989257, "grad_norm": 0.05973792253062129, "loss": 0.6859862327575683, "time": 0.8772577524185181, "epoch": 2, "memory": 28783, "step": 7958} +{"lr": 0.0009939057285945933, "data_time": 0.0016332387924194336, "grad_norm": 0.08887978848069907, "loss": 0.6890406787395478, "time": 0.8825002431869506, "epoch": 2, "memory": 28783, "step": 8058} +{"lr": 0.0009939057285945933, "data_time": 0.0020130634307861327, "grad_norm": 0.061076579988002776, "loss": 0.6835352778434753, "time": 0.8731219530105591, "epoch": 2, "memory": 28783, "step": 8158} +{"lr": 0.0009939057285945933, "data_time": 0.0018208742141723633, "grad_norm": 0.04753756057471037, "loss": 0.6854871273040771, "time": 0.8559173345565796, "epoch": 2, "memory": 28783, "step": 8258} +{"lr": 0.0009939057285945933, "data_time": 0.002297186851501465, "grad_norm": 0.029006470972672106, "loss": 0.6850190162658691, "time": 0.8677525997161866, "epoch": 2, "memory": 28783, "step": 8358} +{"lr": 0.0009939057285945933, "data_time": 0.0016402006149291992, "grad_norm": 0.06308088432997465, "loss": 0.6902522623538971, "time": 0.8611478328704834, "epoch": 2, "memory": 28783, "step": 8458} +{"lr": 0.0009939057285945933, "data_time": 0.0017788410186767578, "grad_norm": 0.05403971169143915, "loss": 0.6853444218635559, "time": 0.8659428119659424, "epoch": 2, "memory": 28783, "step": 8558} +{"lr": 0.0009939057285945933, "data_time": 0.0019292831420898438, "grad_norm": 0.06539376210421324, "loss": 0.6852969110012055, "time": 0.8211614608764648, "epoch": 2, "memory": 28783, "step": 8658} +{"lr": 0.0009939057285945933, "data_time": 0.0033303022384643553, "grad_norm": 0.05729785547591746, "loss": 0.6872235417366028, "time": 0.9224058628082276, "epoch": 2, "memory": 28783, "step": 8758} +{"lr": 0.0009939057285945933, "data_time": 0.0019618511199951173, "grad_norm": 0.07496860194951296, "loss": 0.6859241485595703, "time": 0.9416101694107055, "epoch": 2, "memory": 28783, "step": 8858} +{"lr": 0.0009939057285945933, "data_time": 0.0025458335876464844, "grad_norm": 0.05293703200295567, "loss": 0.6839104056358337, "time": 0.8826780080795288, "epoch": 2, "memory": 28783, "step": 8958} +{"lr": 0.0009939057285945933, "data_time": 0.001965618133544922, "grad_norm": 0.038466890202835204, "loss": 0.688208156824112, "time": 0.8703254699707031, "epoch": 2, "memory": 28783, "step": 9058} +{"lr": 0.0009939057285945933, "data_time": 0.0022713422775268556, "grad_norm": 0.04295786349102855, "loss": 0.6866271257400512, "time": 0.7981902122497558, "epoch": 2, "memory": 28783, "step": 9158} +{"lr": 0.0009939057285945933, "data_time": 0.0027085542678833008, "grad_norm": 0.0508892887737602, "loss": 0.6848026752471924, "time": 0.8314115762710571, "epoch": 2, "memory": 28783, "step": 9258} +{"lr": 0.0009939057285945933, "data_time": 0.0026367902755737305, "grad_norm": 0.055432308837771414, "loss": 0.6862557232379913, "time": 0.8449159145355225, "epoch": 2, "memory": 28783, "step": 9358} +{"lr": 0.0009939057285945933, "data_time": 0.001835036277770996, "grad_norm": 0.03311171652749181, "loss": 0.6850582480430603, "time": 0.8194102764129638, "epoch": 2, "memory": 28783, "step": 9458} +{"lr": 0.0009939057285945933, "data_time": 0.0022808074951171874, "grad_norm": 0.057787072472274305, "loss": 0.6863271236419678, "time": 0.8861421108245849, "epoch": 2, "memory": 28783, "step": 9558} +{"lr": 0.0009939057285945933, "data_time": 0.0018909215927124024, "grad_norm": 0.06150261862203479, "loss": 0.6836685121059418, "time": 0.895897912979126, "epoch": 2, "memory": 28783, "step": 9658} +{"lr": 0.0009939057285945933, "data_time": 0.0027924537658691405, "grad_norm": 0.044922967720776794, "loss": 0.6883501291275025, "time": 0.8687206745147705, "epoch": 2, "memory": 28783, "step": 9758} +{"lr": 0.0009939057285945933, "data_time": 0.002883601188659668, "grad_norm": 0.03776383614167571, "loss": 0.6856978595256805, "time": 0.8509725093841553, "epoch": 2, "memory": 28783, "step": 9858} +{"lr": 0.0009939057285945933, "data_time": 0.0017762184143066406, "grad_norm": 0.05196431670337916, "loss": 0.6858752787113189, "time": 0.8747856378555298, "epoch": 2, "memory": 28783, "step": 9958} +{"lr": 0.0009939057285945933, "data_time": 0.002619647979736328, "grad_norm": 0.038504788372665646, "loss": 0.6874200642108917, "time": 0.9334343671798706, "epoch": 2, "memory": 28783, "step": 10058} +{"lr": 0.0009939057285945933, "data_time": 0.002363920211791992, "grad_norm": 0.03802372813224793, "loss": 0.6889680445194244, "time": 0.8259455680847168, "epoch": 2, "memory": 28783, "step": 10158} +{"lr": 0.0009939057285945933, "data_time": 0.0020737409591674804, "grad_norm": 0.04958138414658606, "loss": 0.6843723356723785, "time": 0.8747926473617553, "epoch": 2, "memory": 28783, "step": 10258} +{"lr": 0.0009939057285945933, "data_time": 0.0017771005630493164, "grad_norm": 0.05166795584373176, "loss": 0.6877075135707855, "time": 0.943445348739624, "epoch": 2, "memory": 28783, "step": 10358} +{"lr": 0.0009939057285945933, "data_time": 0.0017478227615356444, "grad_norm": 0.027391258254647256, "loss": 0.6841959953308105, "time": 0.7990774869918823, "epoch": 2, "memory": 28783, "step": 10458} +{"lr": 0.0009939057285945933, "data_time": 0.002631092071533203, "grad_norm": 0.050929443538188936, "loss": 0.6882081031799316, "time": 0.8281407594680786, "epoch": 2, "memory": 28783, "step": 10558} +{"lr": 0.0009939057285945933, "data_time": 0.0023395538330078123, "grad_norm": 0.035794493090361354, "loss": 0.6882830798625946, "time": 0.8703291177749634, "epoch": 2, "memory": 28783, "step": 10658} +{"lr": 0.0009939057285945933, "data_time": 0.001960015296936035, "grad_norm": 0.03046689215116203, "loss": 0.689048558473587, "time": 0.838711929321289, "epoch": 2, "memory": 28783, "step": 10758} +{"lr": 0.0009939057285945933, "data_time": 0.0016475677490234374, "grad_norm": 0.03824340747669339, "loss": 0.6871283531188965, "time": 0.8687110662460327, "epoch": 2, "memory": 28783, "step": 10858} +{"lr": 0.0009939057285945933, "data_time": 0.0016471624374389648, "grad_norm": 0.0415898721665144, "loss": 0.6892750203609467, "time": 0.8054296731948852, "epoch": 2, "memory": 28783, "step": 10958} +{"lr": 0.0009939057285945933, "data_time": 0.0026760101318359375, "grad_norm": 0.0330808587372303, "loss": 0.6857958555221557, "time": 0.8644672632217407, "epoch": 2, "memory": 28783, "step": 11058} +{"lr": 0.0009939057285945933, "data_time": 0.0016078472137451172, "grad_norm": 0.04337030947208405, "loss": 0.6839715778827667, "time": 0.8387632846832276, "epoch": 2, "memory": 28783, "step": 11158} +{"lr": 0.0009939057285945933, "data_time": 0.0031455039978027345, "grad_norm": 0.05182581262197346, "loss": 0.6873214721679688, "time": 0.8523617506027221, "epoch": 2, "memory": 28783, "step": 11258} +{"lr": 0.0009939057285945933, "data_time": 0.002145862579345703, "grad_norm": 0.03979668482206762, "loss": 0.6871269643306732, "time": 0.8563596963882446, "epoch": 2, "memory": 28783, "step": 11358} +{"lr": 0.0009939057285945933, "data_time": 0.0023670196533203125, "grad_norm": 0.03013358060270548, "loss": 0.6840694546699524, "time": 0.8306557655334472, "epoch": 2, "memory": 28783, "step": 11458} +{"accuracy/top1": 100.0, "data_time": 0.23624329005970673, "time": 0.6698071255403406, "step": 2} +{"lr": 0.0009757729755661012, "data_time": 0.15603256225585938, "grad_norm": 0.0319600532297045, "loss": 0.6842123568058014, "time": 0.8801895380020142, "epoch": 3, "memory": 28783, "step": 11616} +{"lr": 0.0009757729755661012, "data_time": 0.0020602941513061523, "grad_norm": 0.034785063657909635, "loss": 0.6888893246650696, "time": 0.8316410779953003, "epoch": 3, "memory": 28783, "step": 11716} +{"lr": 0.0009757729755661012, "data_time": 0.002340388298034668, "grad_norm": 0.040128201432526114, "loss": 0.6884433448314666, "time": 0.8581866025924683, "epoch": 3, "memory": 28783, "step": 11816} +{"lr": 0.0009757729755661012, "data_time": 0.0014608144760131837, "grad_norm": 0.026125182490795852, "loss": 0.6869147956371308, "time": 0.8026888608932495, "epoch": 3, "memory": 28783, "step": 11916} +{"lr": 0.0009757729755661012, "data_time": 0.0018100738525390625, "grad_norm": 0.03515590270981193, "loss": 0.6843294858932495, "time": 0.8197768449783325, "epoch": 3, "memory": 28783, "step": 12016} +{"lr": 0.0009757729755661012, "data_time": 0.002685070037841797, "grad_norm": 0.03052263017743826, "loss": 0.6916319966316223, "time": 0.831379747390747, "epoch": 3, "memory": 28783, "step": 12116} +{"lr": 0.0009757729755661012, "data_time": 0.0021782875061035155, "grad_norm": 0.03311447571031749, "loss": 0.6845222771167755, "time": 0.8886878490447998, "epoch": 3, "memory": 28783, "step": 12216} +{"lr": 0.0009757729755661012, "data_time": 0.00205686092376709, "grad_norm": 0.03073037527501583, "loss": 0.6866430819034577, "time": 0.8934651613235474, "epoch": 3, "memory": 28783, "step": 12316} +{"lr": 0.0009757729755661012, "data_time": 0.002156639099121094, "grad_norm": 0.02904034472303465, "loss": 0.6870495080947876, "time": 0.8314910888671875, "epoch": 3, "memory": 28783, "step": 12416} +{"lr": 0.0009757729755661012, "data_time": 0.0015803337097167968, "grad_norm": 0.030422977358102798, "loss": 0.6876676201820373, "time": 0.8498664140701294, "epoch": 3, "memory": 28783, "step": 12516} +{"lr": 0.0009757729755661012, "data_time": 0.002436256408691406, "grad_norm": 0.03213986800983548, "loss": 0.6809326410293579, "time": 0.8727119207382202, "epoch": 3, "memory": 28783, "step": 12616} +{"lr": 0.0009757729755661012, "data_time": 0.0026295661926269533, "grad_norm": 0.039869609847664834, "loss": 0.6893934607505798, "time": 0.9144095182418823, "epoch": 3, "memory": 28783, "step": 12716} +{"lr": 0.0009757729755661012, "data_time": 0.0019165277481079102, "grad_norm": 0.030882916337577625, "loss": 0.6870858371257782, "time": 0.8783764600753784, "epoch": 3, "memory": 28783, "step": 12816} +{"lr": 0.0009757729755661012, "data_time": 0.0023956537246704102, "grad_norm": 0.025332201132550837, "loss": 0.6866398930549622, "time": 0.8041995286941528, "epoch": 3, "memory": 28783, "step": 12916} +{"lr": 0.0009757729755661012, "data_time": 0.0019859552383422853, "grad_norm": 0.03307523438706994, "loss": 0.6877510190010071, "time": 0.8169021844863892, "epoch": 3, "memory": 28783, "step": 13016} +{"lr": 0.0009757729755661012, "data_time": 0.0018195152282714845, "grad_norm": 0.01916305311024189, "loss": 0.6880592048168183, "time": 0.8306993246078491, "epoch": 3, "memory": 28783, "step": 13116} +{"lr": 0.0009757729755661012, "data_time": 0.0018338918685913085, "grad_norm": 0.023170626442879438, "loss": 0.6882040917873382, "time": 0.8646248817443848, "epoch": 3, "memory": 28783, "step": 13216} +{"lr": 0.0009757729755661012, "data_time": 0.00217585563659668, "grad_norm": 0.03558538598008454, "loss": 0.6882471084594727, "time": 0.9349054098129272, "epoch": 3, "memory": 28783, "step": 13316} +{"lr": 0.0009757729755661012, "data_time": 0.002730393409729004, "grad_norm": 0.03971612909808755, "loss": 0.6896975994110107, "time": 0.8742197751998901, "epoch": 3, "memory": 28783, "step": 13416} +{"lr": 0.0009757729755661012, "data_time": 0.002499890327453613, "grad_norm": 0.03385852263309062, "loss": 0.688189423084259, "time": 0.9291218757629395, "epoch": 3, "memory": 28783, "step": 13516} +{"lr": 0.0009757729755661012, "data_time": 0.0025034189224243165, "grad_norm": 0.03595380522310734, "loss": 0.6903407573699951, "time": 0.902942419052124, "epoch": 3, "memory": 28783, "step": 13616} +{"lr": 0.0009757729755661012, "data_time": 0.0033739328384399412, "grad_norm": 0.026730207540094853, "loss": 0.6836494386196137, "time": 0.8466433763504029, "epoch": 3, "memory": 28783, "step": 13716} +{"lr": 0.0009757729755661012, "data_time": 0.0024436235427856444, "grad_norm": 0.019502490386366846, "loss": 0.6842268526554107, "time": 0.8439517021179199, "epoch": 3, "memory": 28783, "step": 13816} +{"lr": 0.0009757729755661012, "data_time": 0.0015889883041381836, "grad_norm": 0.04170419373549521, "loss": 0.6888813734054565, "time": 0.8161389112472535, "epoch": 3, "memory": 28783, "step": 13916} +{"lr": 0.0009757729755661012, "data_time": 0.0018450260162353516, "grad_norm": 0.03205676367506385, "loss": 0.688663774728775, "time": 0.9335797071456909, "epoch": 3, "memory": 28783, "step": 14016} +{"lr": 0.0009757729755661012, "data_time": 0.002497696876525879, "grad_norm": 0.02901815567165613, "loss": 0.6877306699752808, "time": 0.890400743484497, "epoch": 3, "memory": 28783, "step": 14116} +{"lr": 0.0009757729755661012, "data_time": 0.0016369342803955079, "grad_norm": 0.028329021483659744, "loss": 0.6899175524711609, "time": 0.8838005781173706, "epoch": 3, "memory": 28783, "step": 14216} +{"lr": 0.0009757729755661012, "data_time": 0.0018515348434448241, "grad_norm": 0.01916082629468292, "loss": 0.6878528296947479, "time": 0.8419302463531494, "epoch": 3, "memory": 28783, "step": 14316} +{"lr": 0.0009757729755661012, "data_time": 0.0017276525497436524, "grad_norm": 0.029594281874597072, "loss": 0.6893251299858093, "time": 0.9079666614532471, "epoch": 3, "memory": 28783, "step": 14416} +{"lr": 0.0009757729755661012, "data_time": 0.0026272296905517577, "grad_norm": 0.029666680260561407, "loss": 0.6882790446281433, "time": 0.8397300958633422, "epoch": 3, "memory": 28783, "step": 14516} +{"lr": 0.0009757729755661012, "data_time": 0.0024530887603759766, "grad_norm": 0.019724297011271118, "loss": 0.6835509836673737, "time": 0.8690378665924072, "epoch": 3, "memory": 28783, "step": 14616} +{"lr": 0.0009757729755661012, "data_time": 0.0023832321166992188, "grad_norm": 0.02569525404833257, "loss": 0.6878702521324158, "time": 0.7856831789016724, "epoch": 3, "memory": 28783, "step": 14716} +{"lr": 0.0009757729755661012, "data_time": 0.0020241498947143554, "grad_norm": 0.03336631013080478, "loss": 0.6883101344108582, "time": 0.8269784688949585, "epoch": 3, "memory": 28783, "step": 14816} +{"lr": 0.0009757729755661012, "data_time": 0.0019346714019775391, "grad_norm": 0.019326245319098233, "loss": 0.6824221432209014, "time": 0.7631311893463135, "epoch": 3, "memory": 28783, "step": 14916} +{"lr": 0.0009757729755661012, "data_time": 0.002571845054626465, "grad_norm": 0.023271534778177737, "loss": 0.6884177565574646, "time": 0.8476747274398804, "epoch": 3, "memory": 28783, "step": 15016} +{"lr": 0.0009757729755661012, "data_time": 0.001752328872680664, "grad_norm": 0.021654575224965812, "loss": 0.6862824082374572, "time": 0.8012657165527344, "epoch": 3, "memory": 28783, "step": 15116} +{"lr": 0.0009757729755661012, "data_time": 0.0024129867553710936, "grad_norm": 0.024710742104798556, "loss": 0.6866450428962707, "time": 0.8590291738510132, "epoch": 3, "memory": 28783, "step": 15216} +{"lr": 0.0009757729755661012, "data_time": 0.0022533655166625975, "grad_norm": 0.01789157767780125, "loss": 0.6886017024517059, "time": 0.7916885375976562, "epoch": 3, "memory": 28783, "step": 15316} +{"lr": 0.0009757729755661012, "data_time": 0.0016936302185058595, "grad_norm": 0.02560994178056717, "loss": 0.6897801280021667, "time": 0.7781678915023804, "epoch": 3, "memory": 28783, "step": 15416} +{"lr": 0.0009757729755661012, "data_time": 0.002540302276611328, "grad_norm": 0.018297767313197256, "loss": 0.6882023394107819, "time": 0.861829137802124, "epoch": 3, "memory": 28783, "step": 15516} +{"lr": 0.0009757729755661012, "data_time": 0.0023749113082885743, "grad_norm": 0.024319959757849573, "loss": 0.6845597207546235, "time": 0.7990557670593261, "epoch": 3, "memory": 28783, "step": 15616} +{"lr": 0.0009757729755661012, "data_time": 0.0024063825607299806, "grad_norm": 0.017872722959145902, "loss": 0.6847500383853913, "time": 0.8461315155029296, "epoch": 3, "memory": 28783, "step": 15716} +{"lr": 0.0009757729755661012, "data_time": 0.0017181873321533204, "grad_norm": 0.02685516527853906, "loss": 0.6859213054180145, "time": 0.8860133409500122, "epoch": 3, "memory": 28783, "step": 15816} +{"lr": 0.0009757729755661012, "data_time": 0.002274155616760254, "grad_norm": 0.020420729462057353, "loss": 0.6882957756519318, "time": 0.7865877628326416, "epoch": 3, "memory": 28783, "step": 15916} +{"lr": 0.0009757729755661012, "data_time": 0.0025748014450073242, "grad_norm": 0.023899454763159154, "loss": 0.689612740278244, "time": 0.8770796537399292, "epoch": 3, "memory": 28783, "step": 16016} +{"lr": 0.0009757729755661012, "data_time": 0.0022995710372924806, "grad_norm": 0.04271694906055927, "loss": 0.6881500124931336, "time": 0.8408160448074341, "epoch": 3, "memory": 28783, "step": 16116} +{"lr": 0.0009757729755661012, "data_time": 0.0029107332229614258, "grad_norm": 0.04083000328391791, "loss": 0.6855602383613586, "time": 0.8630110025405884, "epoch": 3, "memory": 28783, "step": 16216} +{"lr": 0.0009757729755661012, "data_time": 0.002035045623779297, "grad_norm": 0.039553499594330785, "loss": 0.6860554337501525, "time": 0.7795724391937255, "epoch": 3, "memory": 28783, "step": 16316} +{"lr": 0.0009757729755661012, "data_time": 0.0019381046295166016, "grad_norm": 0.03328489847481251, "loss": 0.6891292452812194, "time": 0.8267752647399902, "epoch": 3, "memory": 28783, "step": 16416} +{"lr": 0.0009757729755661012, "data_time": 0.00224452018737793, "grad_norm": 0.027175406087189914, "loss": 0.6853455245494843, "time": 0.7782455444335937, "epoch": 3, "memory": 28783, "step": 16516} +{"lr": 0.0009757729755661012, "data_time": 0.0024422645568847657, "grad_norm": 0.020163445826619865, "loss": 0.6882229030132294, "time": 0.7930678844451904, "epoch": 3, "memory": 28783, "step": 16616} +{"lr": 0.0009757729755661012, "data_time": 0.002415776252746582, "grad_norm": 0.023008845327422024, "loss": 0.6872393250465393, "time": 0.9097555160522461, "epoch": 3, "memory": 28783, "step": 16716} +{"lr": 0.0009757729755661012, "data_time": 0.0026239395141601563, "grad_norm": 0.026061313063837587, "loss": 0.6863723039627075, "time": 0.8081557989120484, "epoch": 3, "memory": 28783, "step": 16816} +{"lr": 0.0009757729755661012, "data_time": 0.002093195915222168, "grad_norm": 0.02261505932547152, "loss": 0.6847315073013306, "time": 0.8320325374603271, "epoch": 3, "memory": 28783, "step": 16916} +{"lr": 0.0009757729755661012, "data_time": 0.002511000633239746, "grad_norm": 0.031159522105008365, "loss": 0.6862626016139984, "time": 0.8609819650650025, "epoch": 3, "memory": 28783, "step": 17016} +{"lr": 0.0009757729755661012, "data_time": 0.0020642518997192384, "grad_norm": 0.03190843695774674, "loss": 0.6833876013755799, "time": 0.8509474515914917, "epoch": 3, "memory": 28783, "step": 17116} +{"lr": 0.0009757729755661012, "data_time": 0.002220606803894043, "grad_norm": 0.03270708271302283, "loss": 0.6867267310619354, "time": 0.7736714601516723, "epoch": 3, "memory": 28783, "step": 17216} +{"accuracy/top1": 100.0, "data_time": 0.24135509659262264, "time": 0.6569869237787583, "step": 3} +{"lr": 0.0009460482294732422, "data_time": 0.27177393436431885, "grad_norm": 0.02520683826878667, "loss": 0.685707426071167, "time": 0.8160369873046875, "epoch": 4, "memory": 28783, "step": 17374} +{"lr": 0.0009460482294732422, "data_time": 0.1556546926498413, "grad_norm": 0.029967418173328042, "loss": 0.6872336387634277, "time": 0.7802465438842774, "epoch": 4, "memory": 28783, "step": 17474} +{"lr": 0.0009460482294732422, "data_time": 0.001715707778930664, "grad_norm": 0.02238327800296247, "loss": 0.68704394698143, "time": 0.7861227750778198, "epoch": 4, "memory": 28783, "step": 17574} +{"lr": 0.0009460482294732422, "data_time": 0.0022268056869506835, "grad_norm": 0.026502556120976805, "loss": 0.686622679233551, "time": 0.7784161806106568, "epoch": 4, "memory": 28783, "step": 17674} +{"lr": 0.0009460482294732422, "data_time": 0.002174186706542969, "grad_norm": 0.015318740205839277, "loss": 0.6857123255729676, "time": 0.6623723030090332, "epoch": 4, "memory": 28783, "step": 17774} +{"lr": 0.0009460482294732422, "data_time": 0.0024941682815551756, "grad_norm": 0.036090648267418146, "loss": 0.689178079366684, "time": 0.7903032779693604, "epoch": 4, "memory": 28783, "step": 17874} +{"lr": 0.0009460482294732422, "data_time": 0.0016437768936157227, "grad_norm": 0.020696333050727843, "loss": 0.6884391725063324, "time": 0.8287275552749633, "epoch": 4, "memory": 28783, "step": 17974} +{"lr": 0.0009460482294732422, "data_time": 0.0015989780426025391, "grad_norm": 0.029684445937164128, "loss": 0.6866159021854401, "time": 0.8033382654190063, "epoch": 4, "memory": 28783, "step": 18074} +{"lr": 0.0009460482294732422, "data_time": 0.0023195505142211913, "grad_norm": 0.016789997764863073, "loss": 0.6897854745388031, "time": 0.7449037551879882, "epoch": 4, "memory": 28783, "step": 18174} +{"lr": 0.0009460482294732422, "data_time": 0.0019205331802368164, "grad_norm": 0.023764927685260773, "loss": 0.6864448547363281, "time": 0.810158085823059, "epoch": 4, "memory": 28783, "step": 18274} +{"lr": 0.0009460482294732422, "data_time": 0.002965569496154785, "grad_norm": 0.027680600015446543, "loss": 0.6844488143920898, "time": 0.8246042251586914, "epoch": 4, "memory": 28783, "step": 18374} +{"lr": 0.0009460482294732422, "data_time": 0.002289414405822754, "grad_norm": 0.01356877707876265, "loss": 0.6841145396232605, "time": 0.8021170616149902, "epoch": 4, "memory": 28783, "step": 18474} +{"lr": 0.0009460482294732422, "data_time": 0.0021143674850463865, "grad_norm": 0.02218390116468072, "loss": 0.6860071241855621, "time": 0.8011329174041748, "epoch": 4, "memory": 28783, "step": 18574} +{"lr": 0.0009460482294732422, "data_time": 0.0019268989562988281, "grad_norm": 0.022346356231719256, "loss": 0.6873934447765351, "time": 0.8456741809844971, "epoch": 4, "memory": 28783, "step": 18674} +{"lr": 0.0009460482294732422, "data_time": 0.0017806529998779298, "grad_norm": 0.019611597340554, "loss": 0.6864718854427337, "time": 0.8185041666030883, "epoch": 4, "memory": 28783, "step": 18774} +{"lr": 0.0009460482294732422, "data_time": 0.0020433664321899414, "grad_norm": 0.016291472362354398, "loss": 0.6877755224704742, "time": 0.8482494592666626, "epoch": 4, "memory": 28783, "step": 18874} +{"lr": 0.0009460482294732422, "data_time": 0.00248415470123291, "grad_norm": 0.015391213377006352, "loss": 0.6861748933792114, "time": 0.7707361459732056, "epoch": 4, "memory": 28783, "step": 18974} +{"lr": 0.0009460482294732422, "data_time": 0.0016952276229858399, "grad_norm": 0.02560833408497274, "loss": 0.6846916735172272, "time": 0.7614190578460693, "epoch": 4, "memory": 28783, "step": 19074} +{"lr": 0.0009460482294732422, "data_time": 0.0017401218414306641, "grad_norm": 0.016099559073336423, "loss": 0.6865309596061706, "time": 0.7657003164291382, "epoch": 4, "memory": 28783, "step": 19174} +{"lr": 0.0009460482294732422, "data_time": 0.002272343635559082, "grad_norm": 0.022295292373746632, "loss": 0.6868140935897827, "time": 0.8766581058502197, "epoch": 4, "memory": 28783, "step": 19274} +{"lr": 0.0009460482294732422, "data_time": 0.002442622184753418, "grad_norm": 0.015944064315408468, "loss": 0.6904959321022034, "time": 0.8379930257797241, "epoch": 4, "memory": 28783, "step": 19374} +{"lr": 0.0009460482294732422, "data_time": 0.0019188165664672852, "grad_norm": 0.022994579700753092, "loss": 0.6865868210792542, "time": 0.9252864837646484, "epoch": 4, "memory": 28783, "step": 19474} +{"lr": 0.0009460482294732422, "data_time": 0.001770782470703125, "grad_norm": 0.02085417225025594, "loss": 0.6846494317054749, "time": 0.905491828918457, "epoch": 4, "memory": 28783, "step": 19574} +{"lr": 0.0009460482294732422, "data_time": 0.0021780729293823242, "grad_norm": 0.018619938916526736, "loss": 0.6884246349334717, "time": 0.7952443122863769, "epoch": 4, "memory": 28783, "step": 19674} +{"lr": 0.0009460482294732422, "data_time": 0.0019256591796875, "grad_norm": 0.017054049647413194, "loss": 0.6887535750865936, "time": 0.8280713081359863, "epoch": 4, "memory": 28783, "step": 19774} +{"lr": 0.0009460482294732422, "data_time": 0.002608919143676758, "grad_norm": 0.017060671676881613, "loss": 0.6889707565307617, "time": 0.908462381362915, "epoch": 4, "memory": 28783, "step": 19874} +{"lr": 0.0009460482294732422, "data_time": 0.0017066478729248046, "grad_norm": 0.022778198728337883, "loss": 0.6879903554916382, "time": 0.844331932067871, "epoch": 4, "memory": 28783, "step": 19974} +{"lr": 0.0009460482294732422, "data_time": 0.001827073097229004, "grad_norm": 0.015492981299757957, "loss": 0.6909884691238404, "time": 0.7941437244415284, "epoch": 4, "memory": 28783, "step": 20074} +{"lr": 0.0009460482294732422, "data_time": 0.0015969276428222656, "grad_norm": 0.011150317033752799, "loss": 0.687701278924942, "time": 0.7402042865753173, "epoch": 4, "memory": 28783, "step": 20174} +{"lr": 0.0009460482294732422, "data_time": 0.0017906665802001954, "grad_norm": 0.025648143514990805, "loss": 0.6873724579811096, "time": 0.7655733823776245, "epoch": 4, "memory": 28783, "step": 20274} +{"lr": 0.0009460482294732422, "data_time": 0.0015273809432983399, "grad_norm": 0.015162501740269363, "loss": 0.6887618899345398, "time": 0.8254800796508789, "epoch": 4, "memory": 28783, "step": 20374} +{"lr": 0.0009460482294732422, "data_time": 0.0017567873001098633, "grad_norm": 0.025566277944017203, "loss": 0.6850540101528168, "time": 0.8347216129302979, "epoch": 4, "memory": 28783, "step": 20474} +{"lr": 0.0009460482294732422, "data_time": 0.0018349409103393555, "grad_norm": 0.01583637185394764, "loss": 0.6897993505001068, "time": 0.8248376846313477, "epoch": 4, "memory": 28783, "step": 20574} +{"lr": 0.0009460482294732422, "data_time": 0.0020338773727416994, "grad_norm": 0.021004949510097504, "loss": 0.6828468859195709, "time": 0.7930845975875854, "epoch": 4, "memory": 28783, "step": 20674} +{"lr": 0.0009460482294732422, "data_time": 0.0020594120025634764, "grad_norm": 0.018910041358321906, "loss": 0.6877440989017487, "time": 0.8832167625427246, "epoch": 4, "memory": 28783, "step": 20774} +{"lr": 0.0009460482294732422, "data_time": 0.001921868324279785, "grad_norm": 0.02118273505475372, "loss": 0.6865809857845306, "time": 0.8105828285217285, "epoch": 4, "memory": 28783, "step": 20874} +{"lr": 0.0009460482294732422, "data_time": 0.001959085464477539, "grad_norm": 0.016271999990567564, "loss": 0.6873459756374359, "time": 0.7846592187881469, "epoch": 4, "memory": 28783, "step": 20974} +{"lr": 0.0009460482294732422, "data_time": 0.0017027616500854491, "grad_norm": 0.020117701822891833, "loss": 0.6867531418800354, "time": 0.8508880853652954, "epoch": 4, "memory": 28783, "step": 21074} +{"lr": 0.0009460482294732422, "data_time": 0.0017918825149536132, "grad_norm": 0.024550669733434914, "loss": 0.687385618686676, "time": 0.8273667812347412, "epoch": 4, "memory": 28783, "step": 21174} +{"lr": 0.0009460482294732422, "data_time": 0.0019077301025390626, "grad_norm": 0.01758038029074669, "loss": 0.6860480785369873, "time": 0.7815929412841797, "epoch": 4, "memory": 28783, "step": 21274} +{"lr": 0.0009460482294732422, "data_time": 0.0018862485885620117, "grad_norm": 0.015043219458311797, "loss": 0.6862777590751648, "time": 0.8440651893615723, "epoch": 4, "memory": 28783, "step": 21374} +{"lr": 0.0009460482294732422, "data_time": 0.0017036199569702148, "grad_norm": 0.016285820165649056, "loss": 0.689351886510849, "time": 0.7999744892120362, "epoch": 4, "memory": 28783, "step": 21474} +{"lr": 0.0009460482294732422, "data_time": 0.001898050308227539, "grad_norm": 0.01762022227048874, "loss": 0.6893795967102051, "time": 0.8926919221878051, "epoch": 4, "memory": 28783, "step": 21574} +{"lr": 0.0009460482294732422, "data_time": 0.052608704566955565, "grad_norm": 0.023961476050317287, "loss": 0.6867991864681244, "time": 0.8131070137023926, "epoch": 4, "memory": 28783, "step": 21674} +{"lr": 0.0009460482294732422, "data_time": 0.13437292575836182, "grad_norm": 0.008743662922643124, "loss": 0.6895983636379241, "time": 0.7482769966125489, "epoch": 4, "memory": 28783, "step": 21774} +{"lr": 0.0009460482294732422, "data_time": 0.36082210540771487, "grad_norm": 0.019423121633008123, "loss": 0.6859278976917267, "time": 0.8646171569824219, "epoch": 4, "memory": 28783, "step": 21874} +{"lr": 0.0009460482294732422, "data_time": 0.28841822147369384, "grad_norm": 0.01611707587726414, "loss": 0.6821094393730164, "time": 0.804296898841858, "epoch": 4, "memory": 28783, "step": 21974} +{"lr": 0.0009460482294732422, "data_time": 0.3223053216934204, "grad_norm": 0.015545494575053453, "loss": 0.6854293584823609, "time": 0.8259261131286622, "epoch": 4, "memory": 28783, "step": 22074} +{"lr": 0.0009460482294732422, "data_time": 0.320308518409729, "grad_norm": 0.016658765822649003, "loss": 0.6850722014904023, "time": 0.8289936780929565, "epoch": 4, "memory": 28783, "step": 22174} +{"lr": 0.0009460482294732422, "data_time": 0.3064169645309448, "grad_norm": 0.014246461691800504, "loss": 0.6861003637313843, "time": 0.8129561901092529, "epoch": 4, "memory": 28783, "step": 22274} +{"lr": 0.0009460482294732422, "data_time": 0.3194380044937134, "grad_norm": 0.0113477194448933, "loss": 0.6856185317039489, "time": 0.8231329202651978, "epoch": 4, "memory": 28783, "step": 22374} +{"lr": 0.0009460482294732422, "data_time": 0.3566581249237061, "grad_norm": 0.02150777440983802, "loss": 0.6859536170959473, "time": 0.861137080192566, "epoch": 4, "memory": 28783, "step": 22474} +{"lr": 0.0009460482294732422, "data_time": 0.30666141510009765, "grad_norm": 0.02201671889051795, "loss": 0.6879827201366424, "time": 0.8114376544952393, "epoch": 4, "memory": 28783, "step": 22574} +{"lr": 0.0009460482294732422, "data_time": 0.27706668376922605, "grad_norm": 0.01324184676632285, "loss": 0.686724579334259, "time": 0.7841790437698364, "epoch": 4, "memory": 28783, "step": 22674} +{"lr": 0.0009460482294732422, "data_time": 0.15586607456207274, "grad_norm": 0.019132132339291275, "loss": 0.6883108198642731, "time": 0.7770811319351196, "epoch": 4, "memory": 28783, "step": 22774} +{"lr": 0.0009460482294732422, "data_time": 0.0016731500625610351, "grad_norm": 0.015146529953926801, "loss": 0.6915353000164032, "time": 0.802668285369873, "epoch": 4, "memory": 28783, "step": 22874} +{"lr": 0.0009460482294732422, "data_time": 0.001768207550048828, "grad_norm": 0.02003692085854709, "loss": 0.6877279937267303, "time": 0.7683682203292846, "epoch": 4, "memory": 28783, "step": 22974} +{"accuracy/top1": 100.0, "data_time": 0.21070276989656336, "time": 0.6255511816810159, "step": 4} +{"lr": 0.0009054634122155991, "data_time": 0.0023488521575927733, "grad_norm": 0.01978237871080637, "loss": 0.6857580482959748, "time": 0.7440827131271363, "epoch": 5, "memory": 28783, "step": 23132} +{"lr": 0.0009054634122155991, "data_time": 0.001589345932006836, "grad_norm": 0.01621111078420654, "loss": 0.6885083854198456, "time": 0.8214529275894165, "epoch": 5, "memory": 28783, "step": 23232} +{"lr": 0.0009054634122155991, "data_time": 0.0017946004867553712, "grad_norm": 0.015389048820361495, "loss": 0.6877581059932709, "time": 0.7936827182769776, "epoch": 5, "memory": 28783, "step": 23332} +{"lr": 0.0009054634122155991, "data_time": 0.0019068956375122071, "grad_norm": 0.01602655553724617, "loss": 0.6847742557525635, "time": 0.7926679611206054, "epoch": 5, "memory": 28783, "step": 23432} +{"lr": 0.0009054634122155991, "data_time": 0.0016833066940307618, "grad_norm": 0.01653628007043153, "loss": 0.6900790333747864, "time": 0.8546588659286499, "epoch": 5, "memory": 28783, "step": 23532} +{"lr": 0.0009054634122155991, "data_time": 0.0014732837677001952, "grad_norm": 0.019846688024699688, "loss": 0.6886083960533143, "time": 0.8249326944351196, "epoch": 5, "memory": 28783, "step": 23632} +{"lr": 0.0009054634122155991, "data_time": 0.002274346351623535, "grad_norm": 0.019802834442816674, "loss": 0.6862328767776489, "time": 0.7773101329803467, "epoch": 5, "memory": 28783, "step": 23732} +{"lr": 0.0009054634122155991, "data_time": 0.001839900016784668, "grad_norm": 0.018447005655616522, "loss": 0.6862954020500183, "time": 0.8398216247558594, "epoch": 5, "memory": 28783, "step": 23832} +{"lr": 0.0009054634122155991, "data_time": 0.001974010467529297, "grad_norm": 0.024368872703053056, "loss": 0.6901835262775421, "time": 0.8179062843322754, "epoch": 5, "memory": 28783, "step": 23932} +{"lr": 0.0009054634122155991, "data_time": 0.0016718864440917968, "grad_norm": 0.019588706316426396, "loss": 0.6887371957302093, "time": 0.761960792541504, "epoch": 5, "memory": 28783, "step": 24032} +{"lr": 0.0009054634122155991, "data_time": 0.002263045310974121, "grad_norm": 0.01628736048005521, "loss": 0.6862403869628906, "time": 0.784402585029602, "epoch": 5, "memory": 28783, "step": 24132} +{"lr": 0.0009054634122155991, "data_time": 0.002751803398132324, "grad_norm": 0.013741782843135297, "loss": 0.6810522973537445, "time": 0.8334048748016357, "epoch": 5, "memory": 28783, "step": 24232} +{"lr": 0.0009054634122155991, "data_time": 0.002513623237609863, "grad_norm": 0.021936559700407088, "loss": 0.6858477830886841, "time": 0.8193792343139649, "epoch": 5, "memory": 28783, "step": 24332} +{"lr": 0.0009054634122155991, "data_time": 0.0017559528350830078, "grad_norm": 0.01746827968163416, "loss": 0.6847206890583039, "time": 0.7778014898300171, "epoch": 5, "memory": 28783, "step": 24432} +{"lr": 0.0009054634122155991, "data_time": 0.0017470121383666992, "grad_norm": 0.01922081718221307, "loss": 0.6864348530769349, "time": 0.8111227989196778, "epoch": 5, "memory": 28783, "step": 24532} +{"lr": 0.0009054634122155991, "data_time": 0.0018786907196044922, "grad_norm": 0.01580275730229914, "loss": 0.6858112275600433, "time": 0.8756220579147339, "epoch": 5, "memory": 28783, "step": 24632} +{"lr": 0.0009054634122155991, "data_time": 0.0016021728515625, "grad_norm": 0.033075568196363744, "loss": 0.6867753863334656, "time": 0.7886070251464844, "epoch": 5, "memory": 28783, "step": 24732} +{"lr": 0.0009054634122155991, "data_time": 0.0015631914138793945, "grad_norm": 0.013563393824733793, "loss": 0.6867523193359375, "time": 0.7609542608261108, "epoch": 5, "memory": 28783, "step": 24832} +{"lr": 0.0009054634122155991, "data_time": 0.0024301290512084963, "grad_norm": 0.015505463723093272, "loss": 0.6852036118507385, "time": 0.8193845272064209, "epoch": 5, "memory": 28783, "step": 24932} +{"lr": 0.0009054634122155991, "data_time": 0.002778482437133789, "grad_norm": 0.014682882349006832, "loss": 0.6829340577125549, "time": 0.8177146434783935, "epoch": 5, "memory": 28783, "step": 25032} +{"lr": 0.0009054634122155991, "data_time": 0.002022838592529297, "grad_norm": 0.01816959111019969, "loss": 0.6864844381809234, "time": 0.803546142578125, "epoch": 5, "memory": 28783, "step": 25132} +{"lr": 0.0009054634122155991, "data_time": 0.002598214149475098, "grad_norm": 0.011350920656695962, "loss": 0.6844705045223236, "time": 0.814617919921875, "epoch": 5, "memory": 28783, "step": 25232} +{"lr": 0.0009054634122155991, "data_time": 0.0023195743560791016, "grad_norm": 0.01194750884314999, "loss": 0.6867002427577973, "time": 0.787207818031311, "epoch": 5, "memory": 28783, "step": 25332} +{"lr": 0.0009054634122155991, "data_time": 0.0017548561096191405, "grad_norm": 0.016847985982894897, "loss": 0.6891045987606048, "time": 0.7996291160583496, "epoch": 5, "memory": 28783, "step": 25432} +{"lr": 0.0009054634122155991, "data_time": 0.001551198959350586, "grad_norm": 0.012166864122264087, "loss": 0.686947351694107, "time": 0.8243926048278809, "epoch": 5, "memory": 28783, "step": 25532} +{"lr": 0.0009054634122155991, "data_time": 0.0020097970962524416, "grad_norm": 0.018306543538346886, "loss": 0.6858312368392945, "time": 0.8164604902267456, "epoch": 5, "memory": 28783, "step": 25632} +{"lr": 0.0009054634122155991, "data_time": 0.0025929689407348635, "grad_norm": 0.015245183336082846, "loss": 0.6887175500392914, "time": 0.7557824373245239, "epoch": 5, "memory": 28783, "step": 25732} +{"lr": 0.0009054634122155991, "data_time": 0.002368521690368652, "grad_norm": 0.025837548170238735, "loss": 0.6867116332054138, "time": 0.7816663026809693, "epoch": 5, "memory": 28783, "step": 25832} +{"lr": 0.0009054634122155991, "data_time": 0.0017647266387939454, "grad_norm": 0.01864590812474489, "loss": 0.6891556560993195, "time": 0.8854470729827881, "epoch": 5, "memory": 28783, "step": 25932} +{"lr": 0.0009054634122155991, "data_time": 0.0017224788665771485, "grad_norm": 0.016243601916357875, "loss": 0.6866788446903229, "time": 0.8518370628356934, "epoch": 5, "memory": 28783, "step": 26032} +{"lr": 0.0009054634122155991, "data_time": 0.002374696731567383, "grad_norm": 0.017827885830774902, "loss": 0.6882756471633911, "time": 0.7822090148925781, "epoch": 5, "memory": 28783, "step": 26132} +{"lr": 0.0009054634122155991, "data_time": 0.002496147155761719, "grad_norm": 0.017350861546583475, "loss": 0.6872244894504547, "time": 0.7932394504547119, "epoch": 5, "memory": 28783, "step": 26232} +{"lr": 0.0009054634122155991, "data_time": 0.0031243562698364258, "grad_norm": 0.021743663883535193, "loss": 0.6892023324966431, "time": 0.8034002780914307, "epoch": 5, "memory": 28783, "step": 26332} +{"lr": 0.0009054634122155991, "data_time": 0.0024820804595947266, "grad_norm": 0.015899774129502477, "loss": 0.6843734443187713, "time": 0.8362406015396118, "epoch": 5, "memory": 28783, "step": 26432} +{"lr": 0.0009054634122155991, "data_time": 0.0019497394561767579, "grad_norm": 0.0183228648500517, "loss": 0.6847833752632141, "time": 0.8367327690124512, "epoch": 5, "memory": 28783, "step": 26532} +{"lr": 0.0009054634122155991, "data_time": 0.0018743276596069336, "grad_norm": 0.017327110562473536, "loss": 0.6856024205684662, "time": 0.7896200180053711, "epoch": 5, "memory": 28783, "step": 26632} +{"lr": 0.0009054634122155991, "data_time": 0.0017811059951782227, "grad_norm": 0.028434014599770307, "loss": 0.6843396186828613, "time": 0.8050744533538818, "epoch": 5, "memory": 28783, "step": 26732} +{"lr": 0.0009054634122155991, "data_time": 0.002515101432800293, "grad_norm": 0.012014444149099291, "loss": 0.6864933371543884, "time": 0.8401695251464844, "epoch": 5, "memory": 28783, "step": 26832} +{"lr": 0.0009054634122155991, "data_time": 0.0017719030380249023, "grad_norm": 0.021527543873526157, "loss": 0.6900273263454437, "time": 0.8216508150100708, "epoch": 5, "memory": 28783, "step": 26932} +{"lr": 0.0009054634122155991, "data_time": 0.0018821954727172852, "grad_norm": 0.019868181413039566, "loss": 0.6904372155666352, "time": 0.7892855644226074, "epoch": 5, "memory": 28783, "step": 27032} +{"lr": 0.0009054634122155991, "data_time": 0.002454662322998047, "grad_norm": 0.021997840027324855, "loss": 0.6862986445426941, "time": 0.7666745901107788, "epoch": 5, "memory": 28783, "step": 27132} +{"lr": 0.0009054634122155991, "data_time": 0.0017728567123413085, "grad_norm": 0.009503014292567969, "loss": 0.6874842941761017, "time": 0.8114336490631103, "epoch": 5, "memory": 28783, "step": 27232} +{"lr": 0.0009054634122155991, "data_time": 0.0021835088729858397, "grad_norm": 0.018126374180428683, "loss": 0.68863565325737, "time": 0.7756914854049682, "epoch": 5, "memory": 28783, "step": 27332} +{"lr": 0.0009054634122155991, "data_time": 0.0018012046813964844, "grad_norm": 0.014824231527745724, "loss": 0.6837986826896667, "time": 0.8376925468444825, "epoch": 5, "memory": 28783, "step": 27432} +{"lr": 0.0009054634122155991, "data_time": 0.001721501350402832, "grad_norm": 0.01978010511957109, "loss": 0.6871576905250549, "time": 0.76935293674469, "epoch": 5, "memory": 28783, "step": 27532} +{"lr": 0.0009054634122155991, "data_time": 0.0016092777252197266, "grad_norm": 0.013995923660695552, "loss": 0.6886760652065277, "time": 0.7473759174346923, "epoch": 5, "memory": 28783, "step": 27632} +{"lr": 0.0009054634122155991, "data_time": 0.001996016502380371, "grad_norm": 0.014643355918815359, "loss": 0.6834181725978852, "time": 0.7522551774978637, "epoch": 5, "memory": 28783, "step": 27732} +{"lr": 0.0009054634122155991, "data_time": 0.001869058609008789, "grad_norm": 0.018232701462693512, "loss": 0.6909704387187958, "time": 0.7726557493209839, "epoch": 5, "memory": 28783, "step": 27832} +{"lr": 0.0009054634122155991, "data_time": 0.0016345500946044922, "grad_norm": 0.01877805000403896, "loss": 0.6915616273880005, "time": 0.7824899196624756, "epoch": 5, "memory": 28783, "step": 27932} +{"lr": 0.0009054634122155991, "data_time": 0.002409958839416504, "grad_norm": 0.01969190021045506, "loss": 0.6833604395389556, "time": 0.7599211931228638, "epoch": 5, "memory": 28783, "step": 28032} +{"lr": 0.0009054634122155991, "data_time": 0.0018677473068237304, "grad_norm": 0.017953099100850523, "loss": 0.6863316774368287, "time": 0.8113799810409545, "epoch": 5, "memory": 28783, "step": 28132} +{"lr": 0.0009054634122155991, "data_time": 0.0017926931381225587, "grad_norm": 0.01289085887838155, "loss": 0.686074960231781, "time": 0.8070666790008545, "epoch": 5, "memory": 28783, "step": 28232} +{"lr": 0.0009054634122155991, "data_time": 0.0017676353454589844, "grad_norm": 0.013170716888271271, "loss": 0.6844303905963898, "time": 0.8298574924468994, "epoch": 5, "memory": 28783, "step": 28332} +{"lr": 0.0009054634122155991, "data_time": 0.0016425848007202148, "grad_norm": 0.012275812891311943, "loss": 0.6874430894851684, "time": 0.8194347620010376, "epoch": 5, "memory": 28783, "step": 28432} +{"lr": 0.0009054634122155991, "data_time": 0.0018143415451049804, "grad_norm": 0.011029295483604073, "loss": 0.6893566370010376, "time": 0.8269855737686157, "epoch": 5, "memory": 28783, "step": 28532} +{"lr": 0.0009054634122155991, "data_time": 0.0019134521484375, "grad_norm": 0.014390631462447345, "loss": 0.6906480729579926, "time": 0.8102564334869384, "epoch": 5, "memory": 28783, "step": 28632} +{"lr": 0.0009054634122155991, "data_time": 0.001816868782043457, "grad_norm": 0.011227173113729805, "loss": 0.68746337890625, "time": 0.8873985528945922, "epoch": 5, "memory": 28783, "step": 28732} +{"accuracy/top1": 100.0, "data_time": 0.2130567747003892, "time": 0.6275536032284007, "step": 5} +{"lr": 0.0008550178566873411, "data_time": 0.03847064971923828, "grad_norm": 0.016264502005651594, "loss": 0.6850412607192993, "time": 0.7648952722549438, "epoch": 6, "memory": 28783, "step": 28890} +{"lr": 0.0008550178566873411, "data_time": 0.0017691612243652343, "grad_norm": 0.01797385523095727, "loss": 0.6861240804195404, "time": 0.8349392652511597, "epoch": 6, "memory": 28783, "step": 28990} +{"lr": 0.0008550178566873411, "data_time": 0.0018528938293457032, "grad_norm": 0.02012402010150254, "loss": 0.6858736753463746, "time": 0.801614236831665, "epoch": 6, "memory": 28783, "step": 29090} +{"lr": 0.0008550178566873411, "data_time": 0.0017559289932250976, "grad_norm": 0.02366946768015623, "loss": 0.689288717508316, "time": 0.8147270917892456, "epoch": 6, "memory": 28783, "step": 29190} +{"lr": 0.0008550178566873411, "data_time": 0.0017263174057006836, "grad_norm": 0.01673456421121955, "loss": 0.6863142311573028, "time": 0.7748468160629273, "epoch": 6, "memory": 28783, "step": 29290} +{"lr": 0.0008550178566873411, "data_time": 0.0016908884048461915, "grad_norm": 0.016275232611224055, "loss": 0.6843490362167358, "time": 0.8420747280120849, "epoch": 6, "memory": 28783, "step": 29390} +{"lr": 0.0008550178566873411, "data_time": 0.002105450630187988, "grad_norm": 0.009479525731876493, "loss": 0.6898092031478882, "time": 0.7396365404129028, "epoch": 6, "memory": 28783, "step": 29490} +{"lr": 0.0008550178566873411, "data_time": 0.0017427921295166016, "grad_norm": 0.02365730090532452, "loss": 0.6884424567222596, "time": 0.7768778085708619, "epoch": 6, "memory": 28783, "step": 29590} +{"lr": 0.0008550178566873411, "data_time": 0.001991462707519531, "grad_norm": 0.024823940126225352, "loss": 0.684698885679245, "time": 0.772196102142334, "epoch": 6, "memory": 28783, "step": 29690} +{"lr": 0.0008550178566873411, "data_time": 0.0015755414962768555, "grad_norm": 0.020247644768096505, "loss": 0.6880524396896363, "time": 0.8122198820114136, "epoch": 6, "memory": 28783, "step": 29790} +{"lr": 0.0008550178566873411, "data_time": 0.0019313573837280273, "grad_norm": 0.01491468558087945, "loss": 0.6881711661815644, "time": 0.7423748016357422, "epoch": 6, "memory": 28783, "step": 29890} +{"lr": 0.0008550178566873411, "data_time": 0.0019288063049316406, "grad_norm": 0.016054106689989567, "loss": 0.6884989082813263, "time": 0.8245399475097657, "epoch": 6, "memory": 28783, "step": 29990} +{"lr": 0.0008550178566873411, "data_time": 0.0018289327621459962, "grad_norm": 0.01718117855489254, "loss": 0.688566142320633, "time": 0.8262383460998535, "epoch": 6, "memory": 28783, "step": 30090} +{"lr": 0.0008550178566873411, "data_time": 0.0023021459579467773, "grad_norm": 0.013830536813475191, "loss": 0.6884703040122986, "time": 0.825492000579834, "epoch": 6, "memory": 28783, "step": 30190} +{"lr": 0.0008550178566873411, "data_time": 0.0021821022033691405, "grad_norm": 0.016090997448191046, "loss": 0.6848032534122467, "time": 0.8229004383087158, "epoch": 6, "memory": 28783, "step": 30290} +{"lr": 0.0008550178566873411, "data_time": 0.0018199920654296876, "grad_norm": 0.012432183441706001, "loss": 0.6858310401439667, "time": 0.832021975517273, "epoch": 6, "memory": 28783, "step": 30390} +{"lr": 0.0008550178566873411, "data_time": 0.0018375396728515625, "grad_norm": 0.019704459933564067, "loss": 0.6861458837985992, "time": 0.8024144172668457, "epoch": 6, "memory": 28783, "step": 30490} +{"lr": 0.0008550178566873411, "data_time": 0.0015511751174926759, "grad_norm": 0.020204418664798142, "loss": 0.6902287602424622, "time": 0.7734831809997559, "epoch": 6, "memory": 28783, "step": 30590} +{"lr": 0.0008550178566873411, "data_time": 0.0020596981048583984, "grad_norm": 0.013450464070774614, "loss": 0.686540299654007, "time": 0.8267415761947632, "epoch": 6, "memory": 28783, "step": 30690} +{"lr": 0.0008550178566873411, "data_time": 0.0020363569259643556, "grad_norm": 0.014911745325662196, "loss": 0.6875357866287232, "time": 0.7771576404571533, "epoch": 6, "memory": 28783, "step": 30790} +{"lr": 0.0008550178566873411, "data_time": 0.0022123098373413087, "grad_norm": 0.015077015268616378, "loss": 0.6888209342956543, "time": 0.8421217918395996, "epoch": 6, "memory": 28783, "step": 30890} +{"lr": 0.0008550178566873411, "data_time": 0.0016525745391845702, "grad_norm": 0.014739341498352588, "loss": 0.6869848310947418, "time": 0.7946889162063598, "epoch": 6, "memory": 28783, "step": 30990} +{"lr": 0.0008550178566873411, "data_time": 0.0016237974166870117, "grad_norm": 0.020212713163346052, "loss": 0.6863488256931305, "time": 0.7981140613555908, "epoch": 6, "memory": 28783, "step": 31090} +{"lr": 0.0008550178566873411, "data_time": 0.001683497428894043, "grad_norm": 0.011441466386895627, "loss": 0.688124930858612, "time": 0.871791934967041, "epoch": 6, "memory": 28783, "step": 31190} +{"lr": 0.0008550178566873411, "data_time": 0.0025690317153930662, "grad_norm": 0.023499373346567154, "loss": 0.6888080537319183, "time": 0.8048616886138916, "epoch": 6, "memory": 28783, "step": 31290} +{"lr": 0.0008550178566873411, "data_time": 0.0016889095306396485, "grad_norm": 0.018465373641811313, "loss": 0.6895635426044464, "time": 0.8178466081619262, "epoch": 6, "memory": 28783, "step": 31390} +{"lr": 0.0008550178566873411, "data_time": 0.002007722854614258, "grad_norm": 0.015655048273038118, "loss": 0.6862462401390076, "time": 0.7765378952026367, "epoch": 6, "memory": 28783, "step": 31490} +{"lr": 0.0008550178566873411, "data_time": 0.002905130386352539, "grad_norm": 0.015905791777186096, "loss": 0.6853396892547607, "time": 0.8254060983657837, "epoch": 6, "memory": 28783, "step": 31590} +{"lr": 0.0008550178566873411, "data_time": 0.0020444393157958984, "grad_norm": 0.015115947742015124, "loss": 0.6864209949970246, "time": 0.8081553936004638, "epoch": 6, "memory": 28783, "step": 31690} +{"lr": 0.0008550178566873411, "data_time": 0.0019181013107299806, "grad_norm": 0.02002663635648787, "loss": 0.6888629078865052, "time": 0.7970602512359619, "epoch": 6, "memory": 28783, "step": 31790} +{"lr": 0.0008550178566873411, "data_time": 0.0021001338958740235, "grad_norm": 0.009449188085272908, "loss": 0.6891397953033447, "time": 0.7879553079605103, "epoch": 6, "memory": 28783, "step": 31890} +{"lr": 0.0008550178566873411, "data_time": 0.0017438411712646484, "grad_norm": 0.018369228683877737, "loss": 0.6870072305202484, "time": 0.8410459041595459, "epoch": 6, "memory": 28783, "step": 31990} +{"lr": 0.0008550178566873411, "data_time": 0.0019531011581420898, "grad_norm": 0.015933985286392272, "loss": 0.6843014359474182, "time": 0.813026237487793, "epoch": 6, "memory": 28783, "step": 32090} +{"lr": 0.0008550178566873411, "data_time": 0.001747274398803711, "grad_norm": 0.0175642151851207, "loss": 0.6857497990131378, "time": 0.7733380794525146, "epoch": 6, "memory": 28783, "step": 32190} +{"lr": 0.0008550178566873411, "data_time": 0.002689051628112793, "grad_norm": 0.02285115484846756, "loss": 0.6887994766235351, "time": 0.8348977088928222, "epoch": 6, "memory": 28783, "step": 32290} +{"lr": 0.0008550178566873411, "data_time": 0.001812124252319336, "grad_norm": 0.012550061126239598, "loss": 0.6830160737037658, "time": 0.8194183111190796, "epoch": 6, "memory": 28783, "step": 32390} +{"lr": 0.0008550178566873411, "data_time": 0.0022510528564453126, "grad_norm": 0.015787654975429177, "loss": 0.6845639824867249, "time": 0.7714890241622925, "epoch": 6, "memory": 28783, "step": 32490} +{"lr": 0.0008550178566873411, "data_time": 0.00185089111328125, "grad_norm": 0.016110290260985493, "loss": 0.6892039895057678, "time": 0.8338452577590942, "epoch": 6, "memory": 28783, "step": 32590} +{"lr": 0.0008550178566873411, "data_time": 0.001908588409423828, "grad_norm": 0.021656455099582674, "loss": 0.6890843749046326, "time": 0.8557557106018067, "epoch": 6, "memory": 28783, "step": 32690} +{"lr": 0.0008550178566873411, "data_time": 0.0017270803451538085, "grad_norm": 0.01627308176830411, "loss": 0.6873088657855988, "time": 0.7854939699172974, "epoch": 6, "memory": 28783, "step": 32790} +{"lr": 0.0008550178566873411, "data_time": 0.0027657270431518553, "grad_norm": 0.01565114166587591, "loss": 0.6905042350292205, "time": 0.8177520751953125, "epoch": 6, "memory": 28783, "step": 32890} +{"lr": 0.0008550178566873411, "data_time": 0.0022096872329711915, "grad_norm": 0.014277510601095856, "loss": 0.6869805037975312, "time": 0.7925125360488892, "epoch": 6, "memory": 28783, "step": 32990} +{"lr": 0.0008550178566873411, "data_time": 0.0018399953842163086, "grad_norm": 0.013290160486940294, "loss": 0.6889128386974335, "time": 0.8921339750289917, "epoch": 6, "memory": 28783, "step": 33090} +{"lr": 0.0008550178566873411, "data_time": 0.0022870302200317383, "grad_norm": 0.018545915349386634, "loss": 0.6841189384460449, "time": 0.8625333070755005, "epoch": 6, "memory": 28783, "step": 33190} +{"lr": 0.0008550178566873411, "data_time": 0.0018452167510986327, "grad_norm": 0.01571862050332129, "loss": 0.6855437934398652, "time": 0.7996485710144043, "epoch": 6, "memory": 28783, "step": 33290} +{"lr": 0.0008550178566873411, "data_time": 0.0023263216018676756, "grad_norm": 0.01806652401573956, "loss": 0.6881719887256622, "time": 0.8174720764160156, "epoch": 6, "memory": 28783, "step": 33390} +{"lr": 0.0008550178566873411, "data_time": 0.0018170595169067383, "grad_norm": 0.013850232562981546, "loss": 0.6849901378154755, "time": 0.801367974281311, "epoch": 6, "memory": 28783, "step": 33490} +{"lr": 0.0008550178566873411, "data_time": 0.0018849611282348634, "grad_norm": 0.014545754552818835, "loss": 0.6857048273086548, "time": 0.7559276342391967, "epoch": 6, "memory": 28783, "step": 33590} +{"lr": 0.0008550178566873411, "data_time": 0.002006220817565918, "grad_norm": 0.03836859054863453, "loss": 0.6882247507572175, "time": 0.8427144765853882, "epoch": 6, "memory": 28783, "step": 33690} +{"lr": 0.0008550178566873411, "data_time": 0.00178072452545166, "grad_norm": 0.03336802404373884, "loss": 0.6900188505649567, "time": 0.808061671257019, "epoch": 6, "memory": 28783, "step": 33790} +{"lr": 0.0008550178566873411, "data_time": 0.0018287420272827149, "grad_norm": 0.03354310598224401, "loss": 0.6870254516601563, "time": 0.8713032722473144, "epoch": 6, "memory": 28783, "step": 33890} +{"lr": 0.0008550178566873411, "data_time": 0.0020443201065063477, "grad_norm": 0.02385171614587307, "loss": 0.6834163248538971, "time": 0.8586906433105469, "epoch": 6, "memory": 28783, "step": 33990} +{"lr": 0.0008550178566873411, "data_time": 0.0015150070190429687, "grad_norm": 0.023183808848261832, "loss": 0.6892798781394959, "time": 0.765532660484314, "epoch": 6, "memory": 28783, "step": 34090} +{"lr": 0.0008550178566873411, "data_time": 0.001829814910888672, "grad_norm": 0.023499545454978944, "loss": 0.6863574862480164, "time": 0.8652956008911132, "epoch": 6, "memory": 28783, "step": 34190} +{"lr": 0.0008550178566873411, "data_time": 0.002898502349853516, "grad_norm": 0.019519153516739607, "loss": 0.683392482995987, "time": 0.8010478019714355, "epoch": 6, "memory": 28783, "step": 34290} +{"lr": 0.0008550178566873411, "data_time": 0.002373623847961426, "grad_norm": 0.01928608426824212, "loss": 0.6857008755207061, "time": 0.8344033718109131, "epoch": 6, "memory": 28783, "step": 34390} +{"lr": 0.0008550178566873411, "data_time": 0.002845001220703125, "grad_norm": 0.018983273347839714, "loss": 0.6903075754642487, "time": 0.8585350751876831, "epoch": 6, "memory": 28783, "step": 34490} +{"accuracy/top1": 100.0, "data_time": 0.21281940796796014, "time": 0.6271857794593362, "step": 6} +{"lr": 0.0007959536998847743, "data_time": 0.00540931224822998, "grad_norm": 0.01697277408093214, "loss": 0.6851157963275909, "time": 0.7517443418502807, "epoch": 7, "memory": 28783, "step": 34648} +{"lr": 0.0007959536998847743, "data_time": 0.0018120288848876953, "grad_norm": 0.02122509405016899, "loss": 0.6902824699878692, "time": 0.8486426115036011, "epoch": 7, "memory": 28783, "step": 34748} +{"lr": 0.0007959536998847743, "data_time": 0.0016950845718383789, "grad_norm": 0.01617599939927459, "loss": 0.6866198480129242, "time": 0.8561187982559204, "epoch": 7, "memory": 28783, "step": 34848} +{"lr": 0.0007959536998847743, "data_time": 0.002018284797668457, "grad_norm": 0.02124827723018825, "loss": 0.6836631298065186, "time": 0.8276988983154296, "epoch": 7, "memory": 28783, "step": 34948} +{"lr": 0.0007959536998847743, "data_time": 0.001775217056274414, "grad_norm": 0.023269442655146123, "loss": 0.6870947539806366, "time": 0.8401309490203858, "epoch": 7, "memory": 28783, "step": 35048} +{"lr": 0.0007959536998847743, "data_time": 0.0015749931335449219, "grad_norm": 0.017691097455099225, "loss": 0.682102108001709, "time": 0.8155062913894653, "epoch": 7, "memory": 28783, "step": 35148} +{"lr": 0.0007959536998847743, "data_time": 0.0019195795059204102, "grad_norm": 0.01677331989631057, "loss": 0.6869235813617707, "time": 0.8206398487091064, "epoch": 7, "memory": 28783, "step": 35248} +{"lr": 0.0007959536998847743, "data_time": 0.29232115745544435, "grad_norm": 0.015201316401362418, "loss": 0.6860824525356293, "time": 0.8040383577346801, "epoch": 7, "memory": 28783, "step": 35348} +{"lr": 0.0007959536998847743, "data_time": 0.33966894149780275, "grad_norm": 0.01976326904259622, "loss": 0.6853908598423004, "time": 0.841283130645752, "epoch": 7, "memory": 28783, "step": 35448} +{"lr": 0.0007959536998847743, "data_time": 0.28628833293914796, "grad_norm": 0.022069073142483832, "loss": 0.6889750838279725, "time": 0.8028847455978394, "epoch": 7, "memory": 28783, "step": 35548} +{"lr": 0.0007959536998847743, "data_time": 0.3324286937713623, "grad_norm": 0.018840861273929477, "loss": 0.6857481360435486, "time": 0.8515539407730103, "epoch": 7, "memory": 28783, "step": 35648} +{"lr": 0.0007959536998847743, "data_time": 0.32093422412872313, "grad_norm": 0.017893133498728276, "loss": 0.6867282271385193, "time": 0.8367080688476562, "epoch": 7, "memory": 28783, "step": 35748} +{"lr": 0.0007959536998847743, "data_time": 0.3767037630081177, "grad_norm": 0.015831386763602497, "loss": 0.6879664957523346, "time": 0.8953001260757446, "epoch": 7, "memory": 28783, "step": 35848} +{"lr": 0.0007959536998847743, "data_time": 0.34341838359832766, "grad_norm": 0.022307319147512317, "loss": 0.6852727890014648, "time": 0.8477460622787476, "epoch": 7, "memory": 28783, "step": 35948} +{"lr": 0.0007959536998847743, "data_time": 0.3418277263641357, "grad_norm": 0.019870053441263736, "loss": 0.686118996143341, "time": 0.8566797971725464, "epoch": 7, "memory": 28783, "step": 36048} +{"lr": 0.0007959536998847743, "data_time": 0.31561152935028075, "grad_norm": 0.014727455470710993, "loss": 0.6849813759326935, "time": 0.8189444780349732, "epoch": 7, "memory": 28783, "step": 36148} +{"lr": 0.0007959536998847743, "data_time": 0.32175674438476565, "grad_norm": 0.02106335312128067, "loss": 0.6842724800109863, "time": 0.8381194353103638, "epoch": 7, "memory": 28783, "step": 36248} +{"lr": 0.0007959536998847743, "data_time": 0.3440293550491333, "grad_norm": 0.013122439105063676, "loss": 0.6875522017478943, "time": 0.850522494316101, "epoch": 7, "memory": 28783, "step": 36348} +{"lr": 0.0007959536998847743, "data_time": 0.2884658336639404, "grad_norm": 0.014520787284709513, "loss": 0.6893287718296051, "time": 0.7996633291244507, "epoch": 7, "memory": 28783, "step": 36448} +{"lr": 0.0007959536998847743, "data_time": 0.3208257436752319, "grad_norm": 0.015263349004089832, "loss": 0.6875774025917053, "time": 0.8252098798751831, "epoch": 7, "memory": 28783, "step": 36548} +{"lr": 0.0007959536998847743, "data_time": 0.27292938232421876, "grad_norm": 0.020008544623851775, "loss": 0.6905016839504242, "time": 0.7751622676849366, "epoch": 7, "memory": 28783, "step": 36648} +{"lr": 0.0007959536998847743, "data_time": 0.28741676807403566, "grad_norm": 0.01480810260400176, "loss": 0.6877633213996888, "time": 0.7908438444137573, "epoch": 7, "memory": 28783, "step": 36748} +{"lr": 0.0007959536998847743, "data_time": 0.29274702072143555, "grad_norm": 0.021225845348089932, "loss": 0.6866928994655609, "time": 0.7962955713272095, "epoch": 7, "memory": 28783, "step": 36848} +{"lr": 0.0007959536998847743, "data_time": 0.24992833137512208, "grad_norm": 0.021492880955338477, "loss": 0.6857226133346558, "time": 0.7542682647705078, "epoch": 7, "memory": 28783, "step": 36948} +{"lr": 0.0007959536998847743, "data_time": 0.2763364315032959, "grad_norm": 0.01923343800008297, "loss": 0.6883521258831025, "time": 0.7801258563995361, "epoch": 7, "memory": 28783, "step": 37048} +{"lr": 0.0007959536998847743, "data_time": 0.3374753952026367, "grad_norm": 0.020849947445094587, "loss": 0.688578337430954, "time": 0.8415805339813233, "epoch": 7, "memory": 28783, "step": 37148} +{"lr": 0.0007959536998847743, "data_time": 0.3770790100097656, "grad_norm": 0.018818801792804152, "loss": 0.6848691463470459, "time": 0.8801254510879517, "epoch": 7, "memory": 28783, "step": 37248} +{"lr": 0.0007959536998847743, "data_time": 0.297573184967041, "grad_norm": 0.018131504766643047, "loss": 0.6892874658107757, "time": 0.8020789861679077, "epoch": 7, "memory": 28783, "step": 37348} +{"lr": 0.0007959536998847743, "data_time": 0.20495426654815674, "grad_norm": 0.018579118931666017, "loss": 0.6837514460086822, "time": 0.7166455507278442, "epoch": 7, "memory": 28783, "step": 37448} +{"lr": 0.0007959536998847743, "data_time": 0.0017408132553100586, "grad_norm": 0.013621792895719409, "loss": 0.684460973739624, "time": 0.7740114450454711, "epoch": 7, "memory": 28783, "step": 37548} +{"lr": 0.0007959536998847743, "data_time": 0.08445844650268555, "grad_norm": 0.013314396888017655, "loss": 0.6864557564258575, "time": 0.8408969640731812, "epoch": 7, "memory": 28783, "step": 37648} +{"lr": 0.0007959536998847743, "data_time": 0.0017871618270874023, "grad_norm": 0.018849158845841884, "loss": 0.6880821228027344, "time": 0.8186265468597412, "epoch": 7, "memory": 28783, "step": 37748} +{"lr": 0.0007959536998847743, "data_time": 0.0019169330596923828, "grad_norm": 0.024961274396628142, "loss": 0.6845693945884704, "time": 0.8447279214859009, "epoch": 7, "memory": 28783, "step": 37848} +{"lr": 0.0007959536998847743, "data_time": 0.0016588687896728516, "grad_norm": 0.01537142035085708, "loss": 0.6839743673801422, "time": 0.8644203424453736, "epoch": 7, "memory": 28783, "step": 37948} +{"lr": 0.0007959536998847743, "data_time": 0.0023919343948364258, "grad_norm": 0.01768775642849505, "loss": 0.6877807736396789, "time": 0.8464528799057007, "epoch": 7, "memory": 28783, "step": 38048} +{"lr": 0.0007959536998847743, "data_time": 0.0018649816513061524, "grad_norm": 0.018632559943944214, "loss": 0.6885584950447082, "time": 0.8780155420303345, "epoch": 7, "memory": 28783, "step": 38148} +{"lr": 0.0007959536998847743, "data_time": 0.001674962043762207, "grad_norm": 0.01983517063781619, "loss": 0.685562115907669, "time": 0.8226769447326661, "epoch": 7, "memory": 28783, "step": 38248} +{"lr": 0.0007959536998847743, "data_time": 0.0018870830535888672, "grad_norm": 0.018642416410148143, "loss": 0.6852209508419037, "time": 0.8511182069778442, "epoch": 7, "memory": 28783, "step": 38348} +{"lr": 0.0007959536998847743, "data_time": 0.0017599105834960938, "grad_norm": 0.019039638852700592, "loss": 0.6851335883140564, "time": 0.8081854581832886, "epoch": 7, "memory": 28783, "step": 38448} +{"lr": 0.0007959536998847743, "data_time": 0.0017230510711669922, "grad_norm": 0.017614476894959808, "loss": 0.6859308362007142, "time": 0.7961933135986328, "epoch": 7, "memory": 28783, "step": 38548} +{"lr": 0.0007959536998847743, "data_time": 0.001994490623474121, "grad_norm": 0.01535075893625617, "loss": 0.6879223108291626, "time": 0.8690444469451905, "epoch": 7, "memory": 28783, "step": 38648} +{"lr": 0.0007959536998847743, "data_time": 0.0021904468536376952, "grad_norm": 0.8099116334691644, "loss": 0.6889193534851075, "time": 0.8068270444869995, "epoch": 7, "memory": 28783, "step": 38748} +{"lr": 0.0007959536998847743, "data_time": 0.0021352052688598635, "grad_norm": 0.023233903013169764, "loss": 0.6845666885375976, "time": 0.8488040685653686, "epoch": 7, "memory": 28783, "step": 38848} +{"lr": 0.0007959536998847743, "data_time": 0.001798558235168457, "grad_norm": 0.014812073856592178, "loss": 0.6848809897899628, "time": 0.8234731197357178, "epoch": 7, "memory": 28783, "step": 38948} +{"lr": 0.0007959536998847743, "data_time": 0.0018880128860473632, "grad_norm": 0.011171391513198615, "loss": 0.6835210800170899, "time": 0.8097221374511718, "epoch": 7, "memory": 28783, "step": 39048} +{"lr": 0.0007959536998847743, "data_time": 0.0019295930862426758, "grad_norm": 0.0177352077094838, "loss": 0.6859819829463959, "time": 0.8393646001815795, "epoch": 7, "memory": 28783, "step": 39148} +{"lr": 0.0007959536998847743, "data_time": 0.0017655134201049806, "grad_norm": 0.01382433008402586, "loss": 0.6885336220264435, "time": 0.8047173023223877, "epoch": 7, "memory": 28783, "step": 39248} +{"lr": 0.0007959536998847743, "data_time": 0.0018420219421386719, "grad_norm": 0.01699437089264393, "loss": 0.6874267578125, "time": 0.8533265829086304, "epoch": 7, "memory": 28783, "step": 39348} +{"lr": 0.0007959536998847743, "data_time": 0.001817798614501953, "grad_norm": 0.019277336332015692, "loss": 0.681348568201065, "time": 0.8902977466583252, "epoch": 7, "memory": 28783, "step": 39448} +{"lr": 0.0007959536998847743, "data_time": 0.0017141342163085938, "grad_norm": 0.017996445181779563, "loss": 0.6892510950565338, "time": 0.8733532428741455, "epoch": 7, "memory": 28783, "step": 39548} +{"lr": 0.0007959536998847743, "data_time": 0.0018942594528198243, "grad_norm": 0.017822445137426257, "loss": 0.6878711998462677, "time": 0.8268157720565796, "epoch": 7, "memory": 28783, "step": 39648} +{"lr": 0.0007959536998847743, "data_time": 0.002043890953063965, "grad_norm": 0.014071428310126066, "loss": 0.6857958257198333, "time": 0.9307317733764648, "epoch": 7, "memory": 28783, "step": 39748} +{"lr": 0.0007959536998847743, "data_time": 0.0017424345016479493, "grad_norm": 0.009252369753085077, "loss": 0.6836173892021179, "time": 0.79527747631073, "epoch": 7, "memory": 28783, "step": 39848} +{"lr": 0.0007959536998847743, "data_time": 0.0017492294311523438, "grad_norm": 0.0202229589689523, "loss": 0.6898975431919098, "time": 0.8185217380523682, "epoch": 7, "memory": 28783, "step": 39948} +{"lr": 0.0007959536998847743, "data_time": 0.001741337776184082, "grad_norm": 0.012179149826988579, "loss": 0.688639760017395, "time": 0.7978439092636108, "epoch": 7, "memory": 28783, "step": 40048} +{"lr": 0.0007959536998847743, "data_time": 0.001706242561340332, "grad_norm": 0.01295425179414451, "loss": 0.6856117606163025, "time": 0.8520701408386231, "epoch": 7, "memory": 28783, "step": 40148} +{"lr": 0.0007959536998847743, "data_time": 0.0017007112503051758, "grad_norm": 0.01737766668666154, "loss": 0.6840229749679565, "time": 0.8295711040496826, "epoch": 7, "memory": 28783, "step": 40248} +{"accuracy/top1": 100.0, "data_time": 0.22127382895525763, "time": 0.6371942407944623, "step": 7} +{"lr": 0.0007297252973710758, "data_time": 0.1767284631729126, "grad_norm": 0.01150515889748931, "loss": 0.6880894839763642, "time": 0.8509277105331421, "epoch": 8, "memory": 28783, "step": 40406} +{"lr": 0.0007297252973710758, "data_time": 0.21086158752441406, "grad_norm": 0.015664160903543233, "loss": 0.6834785640239716, "time": 0.8216288805007934, "epoch": 8, "memory": 28783, "step": 40506} +{"lr": 0.0007297252973710758, "data_time": 0.0019620418548583984, "grad_norm": 0.013502050004899502, "loss": 0.6881194651126862, "time": 0.8162750720977783, "epoch": 8, "memory": 28783, "step": 40606} +{"lr": 0.0007297252973710758, "data_time": 0.0016234636306762696, "grad_norm": 0.019846187462098895, "loss": 0.6858789980411529, "time": 0.7795439481735229, "epoch": 8, "memory": 28783, "step": 40706} +{"lr": 0.0007297252973710758, "data_time": 0.0015878915786743165, "grad_norm": 0.019163573649711906, "loss": 0.6904479563236237, "time": 0.8327592849731446, "epoch": 8, "memory": 28783, "step": 40806} +{"lr": 0.0007297252973710758, "data_time": 0.001642751693725586, "grad_norm": 0.02173926420509815, "loss": 0.6876375555992127, "time": 0.807506513595581, "epoch": 8, "memory": 28783, "step": 40906} +{"lr": 0.0007297252973710758, "data_time": 0.0024679422378540037, "grad_norm": 0.011791924037970603, "loss": 0.684174531698227, "time": 0.8112164735794067, "epoch": 8, "memory": 28783, "step": 41006} +{"lr": 0.0007297252973710758, "data_time": 0.002004408836364746, "grad_norm": 0.01406180146150291, "loss": 0.6850703954696655, "time": 0.7763309240341186, "epoch": 8, "memory": 28783, "step": 41106} +{"lr": 0.0007297252973710758, "data_time": 0.001792311668395996, "grad_norm": 0.016189863812178372, "loss": 0.6845689654350281, "time": 0.8363158464431762, "epoch": 8, "memory": 28783, "step": 41206} +{"lr": 0.0007297252973710758, "data_time": 0.0016970634460449219, "grad_norm": 0.02064953404478729, "loss": 0.6891416013240814, "time": 0.830332612991333, "epoch": 8, "memory": 28783, "step": 41306} +{"lr": 0.0007297252973710758, "data_time": 0.0017680883407592773, "grad_norm": 0.02018622092436999, "loss": 0.6847874343395233, "time": 0.7984568357467652, "epoch": 8, "memory": 28783, "step": 41406} +{"lr": 0.0007297252973710758, "data_time": 0.36053225994110105, "grad_norm": 0.020458186184987424, "loss": 0.6852293372154236, "time": 0.8626003980636596, "epoch": 8, "memory": 28783, "step": 41506} +{"lr": 0.0007297252973710758, "data_time": 0.2516258001327515, "grad_norm": 0.01941661594901234, "loss": 0.6865932166576385, "time": 0.7552858114242553, "epoch": 8, "memory": 28783, "step": 41606} +{"lr": 0.0007297252973710758, "data_time": 0.2104280710220337, "grad_norm": 0.011324568837881088, "loss": 0.6893284142017364, "time": 0.7168056011199951, "epoch": 8, "memory": 28783, "step": 41706} +{"lr": 0.0007297252973710758, "data_time": 0.17278699874877929, "grad_norm": 0.020460043149068952, "loss": 0.685898506641388, "time": 0.7893294334411621, "epoch": 8, "memory": 28783, "step": 41806} +{"lr": 0.0007297252973710758, "data_time": 0.0018121004104614258, "grad_norm": 0.015894679352641106, "loss": 0.6870485007762909, "time": 0.8094709634780883, "epoch": 8, "memory": 28783, "step": 41906} +{"lr": 0.0007297252973710758, "data_time": 0.0020395517349243164, "grad_norm": 0.020082036941312255, "loss": 0.6875883340835571, "time": 0.800652003288269, "epoch": 8, "memory": 28783, "step": 42006} +{"lr": 0.0007297252973710758, "data_time": 0.0020413875579833986, "grad_norm": 0.017825177405029535, "loss": 0.6895458459854126, "time": 0.8972333908081055, "epoch": 8, "memory": 28783, "step": 42106} +{"lr": 0.0007297252973710758, "data_time": 0.0025058269500732424, "grad_norm": 0.014239225978963077, "loss": 0.684444534778595, "time": 0.8005325078964234, "epoch": 8, "memory": 28783, "step": 42206} +{"lr": 0.0007297252973710758, "data_time": 0.0016438722610473632, "grad_norm": 0.01622798021417111, "loss": 0.6883606851100922, "time": 0.782297945022583, "epoch": 8, "memory": 28783, "step": 42306} +{"lr": 0.0007297252973710758, "data_time": 0.0020476818084716798, "grad_norm": 0.022021299693733454, "loss": 0.688879132270813, "time": 0.8572196006774903, "epoch": 8, "memory": 28783, "step": 42406} +{"lr": 0.0007297252973710758, "data_time": 0.0032892227172851562, "grad_norm": 0.010213263262994588, "loss": 0.6837104856967926, "time": 0.8039742231369018, "epoch": 8, "memory": 28783, "step": 42506} +{"lr": 0.0007297252973710758, "data_time": 0.001682424545288086, "grad_norm": 0.015494284802116454, "loss": 0.6841726422309875, "time": 0.8095448255538941, "epoch": 8, "memory": 28783, "step": 42606} +{"lr": 0.0007297252973710758, "data_time": 0.001751542091369629, "grad_norm": 0.016526938928291202, "loss": 0.6859462976455688, "time": 0.8550708293914795, "epoch": 8, "memory": 28783, "step": 42706} +{"lr": 0.0007297252973710758, "data_time": 0.002066969871520996, "grad_norm": 0.017640549945645033, "loss": 0.6890587925910949, "time": 0.8095893144607544, "epoch": 8, "memory": 28783, "step": 42806} +{"lr": 0.0007297252973710758, "data_time": 0.0016910552978515625, "grad_norm": 0.021079135406762362, "loss": 0.6901887834072113, "time": 0.8429555177688599, "epoch": 8, "memory": 28783, "step": 42906} +{"lr": 0.0007297252973710758, "data_time": 0.002324557304382324, "grad_norm": 0.018978520389646293, "loss": 0.6881242215633392, "time": 0.8127419948577881, "epoch": 8, "memory": 28783, "step": 43006} +{"lr": 0.0007297252973710758, "data_time": 0.0017854928970336913, "grad_norm": 0.009593458962626755, "loss": 0.6858097434043884, "time": 0.7912485837936402, "epoch": 8, "memory": 28783, "step": 43106} +{"lr": 0.0007297252973710758, "data_time": 0.0018796205520629882, "grad_norm": 0.02009905909653753, "loss": 0.6858508288860321, "time": 0.8097288370132446, "epoch": 8, "memory": 28783, "step": 43206} +{"lr": 0.0007297252973710758, "data_time": 0.0020702838897705077, "grad_norm": 0.017961289966478944, "loss": 0.6840964615345001, "time": 0.7968644857406616, "epoch": 8, "memory": 28783, "step": 43306} +{"lr": 0.0007297252973710758, "data_time": 0.0017736196517944337, "grad_norm": 0.018074063712265342, "loss": 0.6889858484268189, "time": 0.8258228540420532, "epoch": 8, "memory": 28783, "step": 43406} +{"lr": 0.0007297252973710758, "data_time": 0.0016751766204833984, "grad_norm": 0.01450836684089154, "loss": 0.6880022525787354, "time": 0.8089641809463501, "epoch": 8, "memory": 28783, "step": 43506} +{"lr": 0.0007297252973710758, "data_time": 0.0016956567764282227, "grad_norm": 0.014779439964331686, "loss": 0.6862829804420472, "time": 0.7844366550445556, "epoch": 8, "memory": 28783, "step": 43606} +{"lr": 0.0007297252973710758, "data_time": 0.0016278505325317382, "grad_norm": 0.012287978845415637, "loss": 0.6869882702827453, "time": 0.8027417898178101, "epoch": 8, "memory": 28783, "step": 43706} +{"lr": 0.0007297252973710758, "data_time": 0.0017699718475341797, "grad_norm": 0.016230658342828976, "loss": 0.6827964782714844, "time": 0.8362174510955811, "epoch": 8, "memory": 28783, "step": 43806} +{"lr": 0.0007297252973710758, "data_time": 0.0016646385192871094, "grad_norm": 0.021101759106386453, "loss": 0.6868793904781342, "time": 0.8031327247619628, "epoch": 8, "memory": 28783, "step": 43906} +{"lr": 0.0007297252973710758, "data_time": 0.0024164438247680662, "grad_norm": 0.012828164966776967, "loss": 0.6843660533428192, "time": 0.8144444465637207, "epoch": 8, "memory": 28783, "step": 44006} +{"lr": 0.0007297252973710758, "data_time": 0.0018096923828125, "grad_norm": 0.008951877418439835, "loss": 0.6857014656066894, "time": 0.8152435541152954, "epoch": 8, "memory": 28783, "step": 44106} +{"lr": 0.0007297252973710758, "data_time": 0.0025657176971435546, "grad_norm": 0.013029518257826566, "loss": 0.685771006345749, "time": 0.8057149648666382, "epoch": 8, "memory": 28783, "step": 44206} +{"lr": 0.0007297252973710758, "data_time": 0.0022495508193969725, "grad_norm": 0.017793799098581076, "loss": 0.6838438749313355, "time": 0.8712785243988037, "epoch": 8, "memory": 28783, "step": 44306} +{"lr": 0.0007297252973710758, "data_time": 0.002106046676635742, "grad_norm": 0.015450227167457342, "loss": 0.687344777584076, "time": 0.8256576061248779, "epoch": 8, "memory": 28783, "step": 44406} +{"lr": 0.0007297252973710758, "data_time": 0.0024627685546875, "grad_norm": 0.018130247574299575, "loss": 0.6861902713775635, "time": 0.7684907197952271, "epoch": 8, "memory": 28783, "step": 44506} +{"lr": 0.0007297252973710758, "data_time": 0.0018656015396118163, "grad_norm": 0.010195440729148686, "loss": 0.6883004784584046, "time": 0.7909414291381835, "epoch": 8, "memory": 28783, "step": 44606} +{"lr": 0.0007297252973710758, "data_time": 0.0026448726654052734, "grad_norm": 0.0206029093824327, "loss": 0.6878214597702026, "time": 0.8168389081954956, "epoch": 8, "memory": 28783, "step": 44706} +{"lr": 0.0007297252973710758, "data_time": 0.0029001951217651365, "grad_norm": 0.014642440964234994, "loss": 0.6859261691570282, "time": 0.8484565258026123, "epoch": 8, "memory": 28783, "step": 44806} +{"lr": 0.0007297252973710758, "data_time": 0.0024739980697631838, "grad_norm": 0.017981468280777336, "loss": 0.6885385870933532, "time": 0.7753104209899903, "epoch": 8, "memory": 28783, "step": 44906} +{"lr": 0.0007297252973710758, "data_time": 0.0023864984512329103, "grad_norm": 0.014773014979436994, "loss": 0.6869674801826477, "time": 0.8395127534866333, "epoch": 8, "memory": 28783, "step": 45006} +{"lr": 0.0007297252973710758, "data_time": 0.002082347869873047, "grad_norm": 0.009750491514569148, "loss": 0.6907267987728118, "time": 0.8047058582305908, "epoch": 8, "memory": 28783, "step": 45106} +{"lr": 0.0007297252973710758, "data_time": 0.001661968231201172, "grad_norm": 0.015422008372843266, "loss": 0.6874495625495911, "time": 0.8290945053100586, "epoch": 8, "memory": 28783, "step": 45206} +{"lr": 0.0007297252973710758, "data_time": 0.001624298095703125, "grad_norm": 0.018805124121718107, "loss": 0.6828702390193939, "time": 0.815378975868225, "epoch": 8, "memory": 28783, "step": 45306} +{"lr": 0.0007297252973710758, "data_time": 0.001702880859375, "grad_norm": 0.022728974279016257, "loss": 0.6894953548908234, "time": 0.7922136306762695, "epoch": 8, "memory": 28783, "step": 45406} +{"lr": 0.0007297252973710758, "data_time": 0.00167391300201416, "grad_norm": 0.017786211939528583, "loss": 0.6875094890594482, "time": 0.775871729850769, "epoch": 8, "memory": 28783, "step": 45506} +{"lr": 0.0007297252973710758, "data_time": 0.0019589900970458985, "grad_norm": 0.016535743908025324, "loss": 0.6843287467956543, "time": 0.7867937803268432, "epoch": 8, "memory": 28783, "step": 45606} +{"lr": 0.0007297252973710758, "data_time": 0.0022088289260864258, "grad_norm": 0.017723800521343945, "loss": 0.6848374128341674, "time": 0.7778054475784302, "epoch": 8, "memory": 28783, "step": 45706} +{"lr": 0.0007297252973710758, "data_time": 0.001461338996887207, "grad_norm": 0.014086908928584307, "loss": 0.6838295698165894, "time": 0.7659364461898803, "epoch": 8, "memory": 28783, "step": 45806} +{"lr": 0.0007297252973710758, "data_time": 0.001958465576171875, "grad_norm": 0.015010241628624498, "loss": 0.6883653104305267, "time": 0.7757092237472534, "epoch": 8, "memory": 28783, "step": 45906} +{"lr": 0.0007297252973710758, "data_time": 0.0021871089935302734, "grad_norm": 0.01669732518494129, "loss": 0.6845536231994629, "time": 0.7965921878814697, "epoch": 8, "memory": 28783, "step": 46006} +{"accuracy/top1": 100.0, "data_time": 0.23743409268996296, "time": 0.6569828285890467, "step": 8} +{"lr": 0.0006579634122155991, "data_time": 0.0024076223373413084, "grad_norm": 0.019781443290412426, "loss": 0.6864916682243347, "time": 0.7794753074645996, "epoch": 9, "memory": 28783, "step": 46164} +{"lr": 0.0006579634122155991, "data_time": 0.0019613027572631834, "grad_norm": 0.019997008610516787, "loss": 0.6887592673301697, "time": 0.8062206029891967, "epoch": 9, "memory": 28783, "step": 46264} +{"lr": 0.0006579634122155991, "data_time": 0.00228729248046875, "grad_norm": 0.012811769964173436, "loss": 0.6853896021842957, "time": 0.8259541273117066, "epoch": 9, "memory": 28783, "step": 46364} +{"lr": 0.0006579634122155991, "data_time": 0.0019213199615478516, "grad_norm": 0.017396972794085742, "loss": 0.6875722289085389, "time": 0.7887672185897827, "epoch": 9, "memory": 28783, "step": 46464} +{"lr": 0.0006579634122155991, "data_time": 0.002908778190612793, "grad_norm": 0.01309593878686428, "loss": 0.6862093210220337, "time": 0.854130482673645, "epoch": 9, "memory": 28783, "step": 46564} +{"lr": 0.0006579634122155991, "data_time": 0.0026603460311889647, "grad_norm": 0.01784549541771412, "loss": 0.6867650210857391, "time": 0.7887613534927368, "epoch": 9, "memory": 28783, "step": 46664} +{"lr": 0.0006579634122155991, "data_time": 0.003061199188232422, "grad_norm": 0.0178492097184062, "loss": 0.6855368733406066, "time": 0.787494444847107, "epoch": 9, "memory": 28783, "step": 46764} +{"lr": 0.0006579634122155991, "data_time": 0.002389192581176758, "grad_norm": 0.012624053051695228, "loss": 0.6891029179096222, "time": 0.7666549682617188, "epoch": 9, "memory": 28783, "step": 46864} +{"lr": 0.0006579634122155991, "data_time": 0.0022454261779785156, "grad_norm": 0.012896141922101378, "loss": 0.685942804813385, "time": 0.7550723552703857, "epoch": 9, "memory": 28783, "step": 46964} +{"lr": 0.0006579634122155991, "data_time": 0.0018169641494750976, "grad_norm": 0.01250883205793798, "loss": 0.6913094282150268, "time": 0.788721227645874, "epoch": 9, "memory": 28783, "step": 47064} +{"lr": 0.0006579634122155991, "data_time": 0.0017420530319213867, "grad_norm": 0.016285714320838453, "loss": 0.6868237376213073, "time": 0.7884947538375855, "epoch": 9, "memory": 28783, "step": 47164} +{"lr": 0.0006579634122155991, "data_time": 0.0018303632736206056, "grad_norm": 0.017772531998343766, "loss": 0.6867738842964173, "time": 0.8098517656326294, "epoch": 9, "memory": 28783, "step": 47264} +{"lr": 0.0006579634122155991, "data_time": 0.002754521369934082, "grad_norm": 0.016665896121412515, "loss": 0.6840065836906433, "time": 0.8419631958007813, "epoch": 9, "memory": 28783, "step": 47364} +{"lr": 0.0006579634122155991, "data_time": 0.0017142772674560546, "grad_norm": 0.016265262814704327, "loss": 0.6880294740200043, "time": 0.7701505184173584, "epoch": 9, "memory": 28783, "step": 47464} +{"lr": 0.0006579634122155991, "data_time": 0.0018386125564575195, "grad_norm": 0.014546527637867257, "loss": 0.6875187039375306, "time": 0.8083031892776489, "epoch": 9, "memory": 28783, "step": 47564} +{"lr": 0.0006579634122155991, "data_time": 0.0020377397537231444, "grad_norm": 0.015437063807621599, "loss": 0.6856964707374573, "time": 0.8418564558029175, "epoch": 9, "memory": 28783, "step": 47664} +{"lr": 0.0006579634122155991, "data_time": 0.002225494384765625, "grad_norm": 0.0065986689180135725, "loss": 0.6867503643035888, "time": 0.8010595798492431, "epoch": 9, "memory": 28783, "step": 47764} +{"lr": 0.0006579634122155991, "data_time": 0.002016448974609375, "grad_norm": 0.014555867318995297, "loss": 0.6886663317680359, "time": 0.7784498691558838, "epoch": 9, "memory": 28783, "step": 47864} +{"lr": 0.0006579634122155991, "data_time": 0.0016640663146972657, "grad_norm": 0.015458536427468061, "loss": 0.6875035107135773, "time": 0.8535054683685303, "epoch": 9, "memory": 28783, "step": 47964} +{"lr": 0.0006579634122155991, "data_time": 0.0017481088638305665, "grad_norm": 0.028671895572915673, "loss": 0.6898967027664185, "time": 0.8276613235473633, "epoch": 9, "memory": 28783, "step": 48064} +{"lr": 0.0006579634122155991, "data_time": 0.0016767501831054688, "grad_norm": 0.022093181125819684, "loss": 0.6856845915317535, "time": 0.7595356225967407, "epoch": 9, "memory": 28783, "step": 48164} +{"lr": 0.0006579634122155991, "data_time": 0.0019210100173950196, "grad_norm": 0.014214502763934433, "loss": 0.6850096583366394, "time": 0.8262840747833252, "epoch": 9, "memory": 28783, "step": 48264} +{"lr": 0.0006579634122155991, "data_time": 0.0016521215438842773, "grad_norm": 0.017442161875078455, "loss": 0.6890328764915467, "time": 0.8380665302276611, "epoch": 9, "memory": 28783, "step": 48364} +{"lr": 0.0006579634122155991, "data_time": 0.0025158166885375977, "grad_norm": 0.011965154483914376, "loss": 0.687999427318573, "time": 0.7937122583389282, "epoch": 9, "memory": 28783, "step": 48464} +{"lr": 0.0006579634122155991, "data_time": 0.0017914533615112304, "grad_norm": 0.01684018108062446, "loss": 0.6870417118072509, "time": 0.8436041116714478, "epoch": 9, "memory": 28783, "step": 48564} +{"lr": 0.0006579634122155991, "data_time": 0.0017537593841552735, "grad_norm": 0.023170825937995687, "loss": 0.6853677570819855, "time": 0.7876108646392822, "epoch": 9, "memory": 28783, "step": 48664} +{"lr": 0.0006579634122155991, "data_time": 0.00168609619140625, "grad_norm": 0.011022717703599482, "loss": 0.6874555468559265, "time": 0.810965633392334, "epoch": 9, "memory": 28783, "step": 48764} +{"lr": 0.0006579634122155991, "data_time": 0.0018185853958129882, "grad_norm": 0.013551229005679488, "loss": 0.6873404622077942, "time": 0.786519455909729, "epoch": 9, "memory": 28783, "step": 48864} +{"lr": 0.0006579634122155991, "data_time": 0.002039384841918945, "grad_norm": 0.019113608356565236, "loss": 0.6864564061164856, "time": 0.812389326095581, "epoch": 9, "memory": 28783, "step": 48964} +{"lr": 0.0006579634122155991, "data_time": 0.002731513977050781, "grad_norm": 0.021905680443160234, "loss": 0.6883671879768372, "time": 0.8831088304519653, "epoch": 9, "memory": 28783, "step": 49064} +{"lr": 0.0006579634122155991, "data_time": 0.0024415016174316405, "grad_norm": 0.014867144648451357, "loss": 0.6886488735675812, "time": 0.8272521495819092, "epoch": 9, "memory": 28783, "step": 49164} +{"lr": 0.0006579634122155991, "data_time": 0.0033639192581176756, "grad_norm": 0.022948447009548546, "loss": 0.6842088460922241, "time": 0.7653190612792968, "epoch": 9, "memory": 28783, "step": 49264} +{"lr": 0.0006579634122155991, "data_time": 0.0027080774307250977, "grad_norm": 0.01826594895683229, "loss": 0.6880537033081054, "time": 0.7956825494766235, "epoch": 9, "memory": 28783, "step": 49364} +{"lr": 0.0006579634122155991, "data_time": 0.0019319772720336914, "grad_norm": 0.017960479762405158, "loss": 0.6865352869033814, "time": 0.788453483581543, "epoch": 9, "memory": 28783, "step": 49464} +{"lr": 0.0006579634122155991, "data_time": 0.0023500204086303713, "grad_norm": 0.018686064379289748, "loss": 0.6842729091644287, "time": 0.7978798389434815, "epoch": 9, "memory": 28783, "step": 49564} +{"lr": 0.0006579634122155991, "data_time": 0.00227503776550293, "grad_norm": 0.020437034708447755, "loss": 0.6884462833404541, "time": 0.8601483583450318, "epoch": 9, "memory": 28783, "step": 49664} +{"lr": 0.0006579634122155991, "data_time": 0.002756357192993164, "grad_norm": 0.010903456446249038, "loss": 0.687099039554596, "time": 0.7962905406951905, "epoch": 9, "memory": 28783, "step": 49764} +{"lr": 0.0006579634122155991, "data_time": 0.0024381160736083986, "grad_norm": 0.014546839147806167, "loss": 0.6866072475910187, "time": 0.7885930061340332, "epoch": 9, "memory": 28783, "step": 49864} +{"lr": 0.0006579634122155991, "data_time": 0.0030847787857055664, "grad_norm": 0.01970738952513784, "loss": 0.6827532708644867, "time": 0.8508641958236695, "epoch": 9, "memory": 28783, "step": 49964} +{"lr": 0.0006579634122155991, "data_time": 0.0020726919174194336, "grad_norm": 0.016316135716624557, "loss": 0.6884059429168701, "time": 0.7776422739028931, "epoch": 9, "memory": 28783, "step": 50064} +{"lr": 0.0006579634122155991, "data_time": 0.002725529670715332, "grad_norm": 0.020280647068284453, "loss": 0.6885368764400482, "time": 0.8471766710281372, "epoch": 9, "memory": 28783, "step": 50164} +{"lr": 0.0006579634122155991, "data_time": 0.00283968448638916, "grad_norm": 0.01854660043027252, "loss": 0.6863507449626922, "time": 0.7698218584060669, "epoch": 9, "memory": 28783, "step": 50264} +{"lr": 0.0006579634122155991, "data_time": 0.0025454044342041017, "grad_norm": 0.00827188534894958, "loss": 0.6900393664836884, "time": 0.8138433933258057, "epoch": 9, "memory": 28783, "step": 50364} +{"lr": 0.0006579634122155991, "data_time": 0.0023783206939697265, "grad_norm": 0.016054447554051877, "loss": 0.6873500943183899, "time": 0.848004150390625, "epoch": 9, "memory": 28783, "step": 50464} +{"lr": 0.0006579634122155991, "data_time": 0.0027923107147216795, "grad_norm": 0.013526805839501322, "loss": 0.6884520709514618, "time": 0.8253152370452881, "epoch": 9, "memory": 28783, "step": 50564} +{"lr": 0.0006579634122155991, "data_time": 0.0026437997817993163, "grad_norm": 0.01568017478566617, "loss": 0.6882683515548706, "time": 0.8205284357070923, "epoch": 9, "memory": 28783, "step": 50664} +{"lr": 0.0006579634122155991, "data_time": 0.002894091606140137, "grad_norm": 0.020560159953311084, "loss": 0.6856422841548919, "time": 0.8077261924743653, "epoch": 9, "memory": 28783, "step": 50764} +{"lr": 0.0006579634122155991, "data_time": 0.0027227878570556642, "grad_norm": 0.013660538289695979, "loss": 0.6901677668094635, "time": 0.8817827939987183, "epoch": 9, "memory": 28783, "step": 50864} +{"lr": 0.0006579634122155991, "data_time": 0.002523994445800781, "grad_norm": 0.011582893878221511, "loss": 0.6891176402568817, "time": 0.843101167678833, "epoch": 9, "memory": 28783, "step": 50964} +{"lr": 0.0006579634122155991, "data_time": 0.0027343034744262695, "grad_norm": 0.026205797819420694, "loss": 0.6876045644283295, "time": 0.8893002033233642, "epoch": 9, "memory": 28783, "step": 51064} +{"lr": 0.0006579634122155991, "data_time": 0.0024560213088989256, "grad_norm": 0.018182085431180894, "loss": 0.686473798751831, "time": 0.7928990840911865, "epoch": 9, "memory": 28783, "step": 51164} +{"lr": 0.0006579634122155991, "data_time": 0.0023867130279541016, "grad_norm": 0.016295361745869742, "loss": 0.6888134717941284, "time": 0.8284400939941406, "epoch": 9, "memory": 28783, "step": 51264} +{"lr": 0.0006579634122155991, "data_time": 0.0023748397827148436, "grad_norm": 0.027016353514045476, "loss": 0.6855190753936767, "time": 0.8068228483200073, "epoch": 9, "memory": 28783, "step": 51364} +{"lr": 0.0006579634122155991, "data_time": 0.0022649288177490233, "grad_norm": 0.01126832386944443, "loss": 0.68656405210495, "time": 0.8059137344360352, "epoch": 9, "memory": 28783, "step": 51464} +{"lr": 0.0006579634122155991, "data_time": 0.0024251937866210938, "grad_norm": 0.017916414327919482, "loss": 0.687969172000885, "time": 0.8256438255310059, "epoch": 9, "memory": 28783, "step": 51564} +{"lr": 0.0006579634122155991, "data_time": 0.0019690752029418944, "grad_norm": 0.01850375476060435, "loss": 0.6857397079467773, "time": 0.824545431137085, "epoch": 9, "memory": 28783, "step": 51664} +{"lr": 0.0006579634122155991, "data_time": 0.0020044326782226564, "grad_norm": 0.01808455412974581, "loss": 0.6860807836055756, "time": 0.8124317407608033, "epoch": 9, "memory": 28783, "step": 51764} +{"accuracy/top1": 100.0, "data_time": 0.22674721830031452, "time": 0.6419453620910645, "step": 9} +{"lr": 0.0005824350601949144, "data_time": 0.15495584011077881, "grad_norm": 0.00975538050988689, "loss": 0.6851868152618408, "time": 0.7970738410949707, "epoch": 10, "memory": 28783, "step": 51922} +{"lr": 0.0005824350601949144, "data_time": 0.002073216438293457, "grad_norm": 0.015719201776664703, "loss": 0.686484295129776, "time": 0.8727296352386474, "epoch": 10, "memory": 28783, "step": 52022} +{"lr": 0.0005824350601949144, "data_time": 0.0018186569213867188, "grad_norm": 0.021165580995148047, "loss": 0.6866058111190796, "time": 0.8343714475631714, "epoch": 10, "memory": 28783, "step": 52122} +{"lr": 0.0005824350601949144, "data_time": 0.0018466711044311523, "grad_norm": 0.015086830453947186, "loss": 0.6867835998535157, "time": 0.8312772035598754, "epoch": 10, "memory": 28783, "step": 52222} +{"lr": 0.0005824350601949144, "data_time": 0.0019497394561767579, "grad_norm": 0.0197906780987978, "loss": 0.6845935881137848, "time": 0.8613074541091919, "epoch": 10, "memory": 28783, "step": 52322} +{"lr": 0.0005824350601949144, "data_time": 0.0019980907440185548, "grad_norm": 0.016787458141334356, "loss": 0.6867964863777161, "time": 0.7988025426864624, "epoch": 10, "memory": 28783, "step": 52422} +{"lr": 0.0005824350601949144, "data_time": 0.0020562410354614258, "grad_norm": 0.020981702487915754, "loss": 0.6887876689434052, "time": 0.8495381355285645, "epoch": 10, "memory": 28783, "step": 52522} +{"lr": 0.0005824350601949144, "data_time": 0.0020780086517333983, "grad_norm": 0.013002254930324853, "loss": 0.6891759812831879, "time": 0.8378580331802368, "epoch": 10, "memory": 28783, "step": 52622} +{"lr": 0.0005824350601949144, "data_time": 0.001979827880859375, "grad_norm": 0.0145253571216017, "loss": 0.6885647475719452, "time": 0.9098420381546021, "epoch": 10, "memory": 28783, "step": 52722} +{"lr": 0.0005824350601949144, "data_time": 0.002054119110107422, "grad_norm": 0.02291962190065533, "loss": 0.687529319524765, "time": 0.8263602495193482, "epoch": 10, "memory": 28783, "step": 52822} +{"lr": 0.0005824350601949144, "data_time": 0.0019410371780395508, "grad_norm": 0.01692937088664621, "loss": 0.6871381521224975, "time": 0.8170312404632568, "epoch": 10, "memory": 28783, "step": 52922} +{"lr": 0.0005824350601949144, "data_time": 0.0020849943161010743, "grad_norm": 0.01803222324233502, "loss": 0.6848252892494202, "time": 0.8109024047851563, "epoch": 10, "memory": 28783, "step": 53022} +{"lr": 0.0005824350601949144, "data_time": 0.0018395185470581055, "grad_norm": 0.022690441505983473, "loss": 0.688908976316452, "time": 0.8494559288024902, "epoch": 10, "memory": 28783, "step": 53122} +{"lr": 0.0005824350601949144, "data_time": 0.003214311599731445, "grad_norm": 0.018362203426659108, "loss": 0.6897436201572418, "time": 0.8518648147583008, "epoch": 10, "memory": 28783, "step": 53222} +{"lr": 0.0005824350601949144, "data_time": 0.0020096302032470703, "grad_norm": 0.01799560021609068, "loss": 0.6845239341259003, "time": 0.7954816102981568, "epoch": 10, "memory": 28783, "step": 53322} +{"lr": 0.0005824350601949144, "data_time": 0.0026044368743896483, "grad_norm": 0.010587186249904335, "loss": 0.687562370300293, "time": 0.8352644920349122, "epoch": 10, "memory": 28783, "step": 53422} +{"lr": 0.0005824350601949144, "data_time": 0.0018564939498901367, "grad_norm": 0.019613031949847936, "loss": 0.6844630062580108, "time": 0.765064811706543, "epoch": 10, "memory": 28783, "step": 53522} +{"lr": 0.0005824350601949144, "data_time": 0.001860833168029785, "grad_norm": 0.01571825020946562, "loss": 0.6879616141319275, "time": 0.804143762588501, "epoch": 10, "memory": 28783, "step": 53622} +{"lr": 0.0005824350601949144, "data_time": 0.0017686128616333009, "grad_norm": 0.02392355175688863, "loss": 0.6858514904975891, "time": 0.7745452642440795, "epoch": 10, "memory": 28783, "step": 53722} +{"lr": 0.0005824350601949144, "data_time": 0.0018432855606079102, "grad_norm": 0.011880033416673541, "loss": 0.683768343925476, "time": 0.7639914751052856, "epoch": 10, "memory": 28783, "step": 53822} +{"lr": 0.0005824350601949144, "data_time": 0.0016751050949096679, "grad_norm": 0.017738129105418922, "loss": 0.6832839012145996, "time": 0.830788254737854, "epoch": 10, "memory": 28783, "step": 53922} +{"lr": 0.0005824350601949144, "data_time": 0.0031814813613891602, "grad_norm": 0.017122036882210524, "loss": 0.6826117038726807, "time": 0.8154723167419433, "epoch": 10, "memory": 28783, "step": 54022} +{"lr": 0.0005824350601949144, "data_time": 0.0021076202392578125, "grad_norm": 0.017524270294234156, "loss": 0.6864458441734314, "time": 0.852436876296997, "epoch": 10, "memory": 28783, "step": 54122} +{"lr": 0.0005824350601949144, "data_time": 0.0026784181594848634, "grad_norm": 0.016776928049512207, "loss": 0.687641030550003, "time": 0.8137255430221557, "epoch": 10, "memory": 28783, "step": 54222} +{"lr": 0.0005824350601949144, "data_time": 0.0022649288177490233, "grad_norm": 0.014092632965184749, "loss": 0.6849035382270813, "time": 0.7912559270858764, "epoch": 10, "memory": 28783, "step": 54322} +{"lr": 0.0005824350601949144, "data_time": 0.0018433094024658202, "grad_norm": 0.012496439449023455, "loss": 0.6896382927894592, "time": 0.8284731864929199, "epoch": 10, "memory": 28783, "step": 54422} +{"lr": 0.0005824350601949144, "data_time": 0.0017837285995483398, "grad_norm": 0.0257398568559438, "loss": 0.6870508193969727, "time": 0.7922945976257324, "epoch": 10, "memory": 28783, "step": 54522} +{"lr": 0.0005824350601949144, "data_time": 0.00201263427734375, "grad_norm": 0.019401557138189675, "loss": 0.6867332875728607, "time": 0.8053722381591797, "epoch": 10, "memory": 28783, "step": 54622} +{"lr": 0.0005824350601949144, "data_time": 0.0019307374954223634, "grad_norm": 0.01231341080274433, "loss": 0.6859995365142822, "time": 0.8189839124679565, "epoch": 10, "memory": 28783, "step": 54722} +{"lr": 0.0005824350601949144, "data_time": 0.0021706581115722655, "grad_norm": 0.01855875412002206, "loss": 0.6885318517684936, "time": 0.7882414340972901, "epoch": 10, "memory": 28783, "step": 54822} +{"lr": 0.0005824350601949144, "data_time": 0.0030200958251953127, "grad_norm": 0.012831724854186178, "loss": 0.6892087459564209, "time": 0.8363796234130859, "epoch": 10, "memory": 28783, "step": 54922} +{"lr": 0.0005824350601949144, "data_time": 0.0023002147674560545, "grad_norm": 0.017541662626899778, "loss": 0.6872913300991058, "time": 0.8115854501724243, "epoch": 10, "memory": 28783, "step": 55022} +{"lr": 0.0005824350601949144, "data_time": 0.0019317865371704102, "grad_norm": 0.01501223393715918, "loss": 0.6866838216781617, "time": 0.7921674251556396, "epoch": 10, "memory": 28783, "step": 55122} +{"lr": 0.0005824350601949144, "data_time": 0.0035938024520874023, "grad_norm": 0.012913845560979098, "loss": 0.6876675605773925, "time": 0.8037017107009887, "epoch": 10, "memory": 28783, "step": 55222} +{"lr": 0.0005824350601949144, "data_time": 0.0023813962936401365, "grad_norm": 0.018609405495226383, "loss": 0.6858887314796448, "time": 0.8376826524734498, "epoch": 10, "memory": 28783, "step": 55322} +{"lr": 0.0005824350601949144, "data_time": 0.0019999980926513673, "grad_norm": 0.014814723422750831, "loss": 0.6853049516677856, "time": 0.856293797492981, "epoch": 10, "memory": 28783, "step": 55422} +{"lr": 0.0005824350601949144, "data_time": 0.0018617630004882813, "grad_norm": 0.0204045329708606, "loss": 0.6878536224365235, "time": 0.80146164894104, "epoch": 10, "memory": 28783, "step": 55522} +{"lr": 0.0005824350601949144, "data_time": 0.002483510971069336, "grad_norm": 0.017967253969982267, "loss": 0.6896375954151154, "time": 0.8275414705276489, "epoch": 10, "memory": 28783, "step": 55622} +{"lr": 0.0005824350601949144, "data_time": 0.00185546875, "grad_norm": 0.016180197877110912, "loss": 0.687091201543808, "time": 0.7865345478057861, "epoch": 10, "memory": 28783, "step": 55722} +{"lr": 0.0005824350601949144, "data_time": 0.0036696910858154295, "grad_norm": 0.016136669309344143, "loss": 0.6830904126167298, "time": 0.8425149917602539, "epoch": 10, "memory": 28783, "step": 55822} +{"lr": 0.0005824350601949144, "data_time": 0.001925492286682129, "grad_norm": 0.013465837901458144, "loss": 0.686905813217163, "time": 0.8077531576156616, "epoch": 10, "memory": 28783, "step": 55922} +{"lr": 0.0005824350601949144, "data_time": 0.0017953634262084961, "grad_norm": 0.019176779454573988, "loss": 0.6883704662322998, "time": 0.8467078447341919, "epoch": 10, "memory": 28783, "step": 56022} +{"lr": 0.0005824350601949144, "data_time": 0.0018522262573242188, "grad_norm": 0.014131482504308224, "loss": 0.686920040845871, "time": 0.8090510606765747, "epoch": 10, "memory": 28783, "step": 56122} +{"lr": 0.0005824350601949144, "data_time": 0.001916980743408203, "grad_norm": 0.021248600841499866, "loss": 0.6902573704719543, "time": 0.7851123809814453, "epoch": 10, "memory": 28783, "step": 56222} +{"lr": 0.0005824350601949144, "data_time": 0.0018211841583251954, "grad_norm": 0.016638199565932155, "loss": 0.6876445651054383, "time": 0.8147528409957886, "epoch": 10, "memory": 28783, "step": 56322} +{"lr": 0.0005824350601949144, "data_time": 0.002654409408569336, "grad_norm": 0.014992894604802132, "loss": 0.6888919591903686, "time": 0.9232783794403077, "epoch": 10, "memory": 28783, "step": 56422} +{"lr": 0.0005824350601949144, "data_time": 0.0023769140243530273, "grad_norm": 0.013957629329524934, "loss": 0.6855066955089569, "time": 0.8823898553848266, "epoch": 10, "memory": 28783, "step": 56522} +{"lr": 0.0005824350601949144, "data_time": 0.0023835420608520506, "grad_norm": 0.016980611742474137, "loss": 0.6868089735507965, "time": 0.8272493839263916, "epoch": 10, "memory": 28783, "step": 56622} +{"lr": 0.0005824350601949144, "data_time": 0.002582216262817383, "grad_norm": 0.01621195066254586, "loss": 0.6860782980918885, "time": 0.840746283531189, "epoch": 10, "memory": 28783, "step": 56722} +{"lr": 0.0005824350601949144, "data_time": 0.14568696022033692, "grad_norm": 0.01863554958254099, "loss": 0.6876372575759888, "time": 0.8938787937164306, "epoch": 10, "memory": 28783, "step": 56822} +{"lr": 0.0005824350601949144, "data_time": 0.23851263523101807, "grad_norm": 0.017496292339637877, "loss": 0.6851976811885834, "time": 0.838429069519043, "epoch": 10, "memory": 28783, "step": 56922} +{"lr": 0.0005824350601949144, "data_time": 0.3127598285675049, "grad_norm": 0.02135424907319248, "loss": 0.6882346212863922, "time": 0.8302545309066772, "epoch": 10, "memory": 28783, "step": 57022} +{"lr": 0.0005824350601949144, "data_time": 0.2952101707458496, "grad_norm": 0.017044686170993372, "loss": 0.6846368968486786, "time": 0.8063157320022583, "epoch": 10, "memory": 28783, "step": 57122} +{"lr": 0.0005824350601949144, "data_time": 0.28017406463623046, "grad_norm": 0.02175892653176561, "loss": 0.6814761459827423, "time": 0.7970028877258301, "epoch": 10, "memory": 28783, "step": 57222} +{"lr": 0.0005824350601949144, "data_time": 0.267191743850708, "grad_norm": 0.018163497699424623, "loss": 0.6877678751945495, "time": 0.7744769811630249, "epoch": 10, "memory": 28783, "step": 57322} +{"lr": 0.0005824350601949144, "data_time": 0.21019601821899414, "grad_norm": 0.02312041134573519, "loss": 0.6849162101745605, "time": 0.7158305406570434, "epoch": 10, "memory": 28783, "step": 57422} +{"lr": 0.0005824350601949144, "data_time": 0.2966966390609741, "grad_norm": 0.019629032735247166, "loss": 0.6894556999206543, "time": 0.802053713798523, "epoch": 10, "memory": 28783, "step": 57522} +{"accuracy/top1": 100.0, "data_time": 0.22202599749845617, "time": 0.6372561595019173, "step": 10} +{"lr": 0.0005050000000000001, "data_time": 0.33390305042266843, "grad_norm": 0.014232703985180705, "loss": 0.6860788226127624, "time": 0.8691173076629639, "epoch": 11, "memory": 28783, "step": 57680} +{"lr": 0.0005050000000000001, "data_time": 0.330348539352417, "grad_norm": 0.014199422171805054, "loss": 0.6860699653625488, "time": 0.8362077236175537, "epoch": 11, "memory": 28783, "step": 57780} +{"lr": 0.0005050000000000001, "data_time": 0.39358646869659425, "grad_norm": 0.022275893250480294, "loss": 0.6843575775623322, "time": 0.9236335515975952, "epoch": 11, "memory": 28783, "step": 57880} +{"lr": 0.0005050000000000001, "data_time": 0.33318612575531004, "grad_norm": 0.016775678703561427, "loss": 0.6894590795040131, "time": 0.8368151426315308, "epoch": 11, "memory": 28783, "step": 57980} +{"lr": 0.0005050000000000001, "data_time": 0.3459513187408447, "grad_norm": 0.012328343483386562, "loss": 0.6849977731704712, "time": 0.8754051446914672, "epoch": 11, "memory": 28783, "step": 58080} +{"lr": 0.0005050000000000001, "data_time": 0.36264605522155763, "grad_norm": 0.01742911710171029, "loss": 0.6863394856452942, "time": 0.8671110153198243, "epoch": 11, "memory": 28783, "step": 58180} +{"lr": 0.0005050000000000001, "data_time": 0.3340389013290405, "grad_norm": 0.01449612674769014, "loss": 0.6846177458763123, "time": 0.8383068561553955, "epoch": 11, "memory": 28783, "step": 58280} +{"lr": 0.0005050000000000001, "data_time": 0.3212372064590454, "grad_norm": 0.015622683428227901, "loss": 0.6831147789955139, "time": 0.829296326637268, "epoch": 11, "memory": 28783, "step": 58380} +{"lr": 0.0005050000000000001, "data_time": 0.3451340675354004, "grad_norm": 0.021475805202499032, "loss": 0.6871937930583953, "time": 0.8493071317672729, "epoch": 11, "memory": 28783, "step": 58480} +{"lr": 0.0005050000000000001, "data_time": 0.3325185775756836, "grad_norm": 0.020584947615861892, "loss": 0.6863519251346588, "time": 0.838594651222229, "epoch": 11, "memory": 28783, "step": 58580} +{"lr": 0.0005050000000000001, "data_time": 0.28668100833892823, "grad_norm": 0.01612954700831324, "loss": 0.6870429158210755, "time": 0.7914071321487427, "epoch": 11, "memory": 28783, "step": 58680} +{"lr": 0.0005050000000000001, "data_time": 0.3148604154586792, "grad_norm": 0.015629613283090295, "loss": 0.6877296268939972, "time": 0.8199707984924316, "epoch": 11, "memory": 28783, "step": 58780} +{"lr": 0.0005050000000000001, "data_time": 0.34616312980651853, "grad_norm": 0.013550182757899164, "loss": 0.6892147719860077, "time": 0.8530388355255127, "epoch": 11, "memory": 28783, "step": 58880} +{"lr": 0.0005050000000000001, "data_time": 0.3644432783126831, "grad_norm": 0.012172738916706293, "loss": 0.6892185509204865, "time": 0.8674351930618286, "epoch": 11, "memory": 28783, "step": 58980} +{"lr": 0.0005050000000000001, "data_time": 0.3425153732299805, "grad_norm": 0.016395441099302844, "loss": 0.6892509460449219, "time": 0.8503780364990234, "epoch": 11, "memory": 28783, "step": 59080} +{"lr": 0.0005050000000000001, "data_time": 0.33985285758972167, "grad_norm": 0.015192930371267722, "loss": 0.6893633425235748, "time": 0.8467146635055542, "epoch": 11, "memory": 28783, "step": 59180} +{"lr": 0.0005050000000000001, "data_time": 0.35523161888122556, "grad_norm": 0.017875791247934102, "loss": 0.6898441016674042, "time": 0.8614187240600586, "epoch": 11, "memory": 28783, "step": 59280} +{"lr": 0.0005050000000000001, "data_time": 0.31883018016815184, "grad_norm": 0.01974796229042113, "loss": 0.6828283607959748, "time": 0.8241447448730469, "epoch": 11, "memory": 28783, "step": 59380} +{"lr": 0.0005050000000000001, "data_time": 0.3458577632904053, "grad_norm": 0.015877533543971367, "loss": 0.6861612796783447, "time": 0.8532354354858398, "epoch": 11, "memory": 28783, "step": 59480} +{"lr": 0.0005050000000000001, "data_time": 0.3423829317092896, "grad_norm": 0.015195346274413168, "loss": 0.687296348810196, "time": 0.847105073928833, "epoch": 11, "memory": 28783, "step": 59580} +{"lr": 0.0005050000000000001, "data_time": 0.3770418643951416, "grad_norm": 0.022456830926239492, "loss": 0.6905148088932037, "time": 0.881967306137085, "epoch": 11, "memory": 28783, "step": 59680} +{"lr": 0.0005050000000000001, "data_time": 0.35122835636138916, "grad_norm": 0.011916636599926278, "loss": 0.6868367969989777, "time": 0.8611812591552734, "epoch": 11, "memory": 28783, "step": 59780} +{"lr": 0.0005050000000000001, "data_time": 0.3091218709945679, "grad_norm": 0.02567956829443574, "loss": 0.6893227696418762, "time": 0.8145798444747925, "epoch": 11, "memory": 28783, "step": 59880} +{"lr": 0.0005050000000000001, "data_time": 0.3292386531829834, "grad_norm": 0.021753088338300586, "loss": 0.6893504381179809, "time": 0.8360827207565308, "epoch": 11, "memory": 28783, "step": 59980} +{"lr": 0.0005050000000000001, "data_time": 0.2831078767776489, "grad_norm": 0.014434442017227411, "loss": 0.6883375585079193, "time": 0.7857233762741089, "epoch": 11, "memory": 28783, "step": 60080} +{"lr": 0.0005050000000000001, "data_time": 0.18909857273101807, "grad_norm": 0.011160801281221211, "loss": 0.6866437137126923, "time": 0.8101491451263427, "epoch": 11, "memory": 28783, "step": 60180} +{"lr": 0.0005050000000000001, "data_time": 0.002041149139404297, "grad_norm": 0.021918557886965572, "loss": 0.6850089490413666, "time": 0.8151986837387085, "epoch": 11, "memory": 28783, "step": 60280} +{"lr": 0.0005050000000000001, "data_time": 0.0020836830139160157, "grad_norm": 0.02017070451984182, "loss": 0.6889029443264008, "time": 0.8374037504196167, "epoch": 11, "memory": 28783, "step": 60380} +{"lr": 0.0005050000000000001, "data_time": 0.0022475004196166994, "grad_norm": 0.014040674548596144, "loss": 0.685858428478241, "time": 0.8345201015472412, "epoch": 11, "memory": 28783, "step": 60480} +{"lr": 0.0005050000000000001, "data_time": 0.002798318862915039, "grad_norm": 0.01217449464602396, "loss": 0.6867355525493621, "time": 0.8692013025283813, "epoch": 11, "memory": 28783, "step": 60580} +{"lr": 0.0005050000000000001, "data_time": 0.0018689393997192382, "grad_norm": 0.018253942718729377, "loss": 0.6899765431880951, "time": 0.833388376235962, "epoch": 11, "memory": 28783, "step": 60680} +{"lr": 0.0005050000000000001, "data_time": 0.002090907096862793, "grad_norm": 0.026523547183023764, "loss": 0.6906438887119293, "time": 0.9125625133514405, "epoch": 11, "memory": 28783, "step": 60780} +{"lr": 0.0005050000000000001, "data_time": 0.0017291784286499023, "grad_norm": 0.017480567819438873, "loss": 0.6886890709400177, "time": 0.8224637031555175, "epoch": 11, "memory": 28783, "step": 60880} +{"lr": 0.0005050000000000001, "data_time": 0.0018859386444091796, "grad_norm": 0.012101862154668196, "loss": 0.6879057168960572, "time": 0.8365171432495118, "epoch": 11, "memory": 28783, "step": 60980} +{"lr": 0.0005050000000000001, "data_time": 0.0019419670104980468, "grad_norm": 0.012853042338974774, "loss": 0.6873636722564698, "time": 0.8697279930114746, "epoch": 11, "memory": 28783, "step": 61080} +{"lr": 0.0005050000000000001, "data_time": 0.002118682861328125, "grad_norm": 0.020067825028672815, "loss": 0.6847352743148803, "time": 0.909968638420105, "epoch": 11, "memory": 28783, "step": 61180} +{"lr": 0.0005050000000000001, "data_time": 0.0018488168716430664, "grad_norm": 0.013973988778889179, "loss": 0.6909814417362213, "time": 0.8352821350097657, "epoch": 11, "memory": 28783, "step": 61280} +{"lr": 0.0005050000000000001, "data_time": 0.0017964363098144532, "grad_norm": 0.00903941192664206, "loss": 0.6856515228748321, "time": 0.8690871715545654, "epoch": 11, "memory": 28783, "step": 61380} +{"lr": 0.0005050000000000001, "data_time": 0.002246546745300293, "grad_norm": 0.016161096398718655, "loss": 0.6874457240104676, "time": 0.8462628602981568, "epoch": 11, "memory": 28783, "step": 61480} +{"lr": 0.0005050000000000001, "data_time": 0.0019682168960571287, "grad_norm": 0.01643331847153604, "loss": 0.6866926908493042, "time": 0.8641710996627807, "epoch": 11, "memory": 28783, "step": 61580} +{"lr": 0.0005050000000000001, "data_time": 0.001953125, "grad_norm": 0.015661456622183323, "loss": 0.684465229511261, "time": 0.8878344297409058, "epoch": 11, "memory": 28783, "step": 61680} +{"lr": 0.0005050000000000001, "data_time": 0.00203557014465332, "grad_norm": 0.01332063355948776, "loss": 0.6864002048969269, "time": 0.8721402645111084, "epoch": 11, "memory": 28783, "step": 61780} +{"lr": 0.0005050000000000001, "data_time": 0.0020078420639038086, "grad_norm": 0.008680033922428266, "loss": 0.6860170960426331, "time": 0.8419245719909668, "epoch": 11, "memory": 28783, "step": 61880} +{"lr": 0.0005050000000000001, "data_time": 0.0018712997436523438, "grad_norm": 0.020153040206059812, "loss": 0.6840706110000611, "time": 0.8748828649520874, "epoch": 11, "memory": 28783, "step": 61980} +{"lr": 0.0005050000000000001, "data_time": 0.0018357515335083007, "grad_norm": 0.006081509188516065, "loss": 0.6893117904663086, "time": 0.823386836051941, "epoch": 11, "memory": 28783, "step": 62080} +{"lr": 0.0005050000000000001, "data_time": 0.0018768310546875, "grad_norm": 0.012463466078042985, "loss": 0.6862398684024811, "time": 0.7977720260620117, "epoch": 11, "memory": 28783, "step": 62180} +{"lr": 0.0005050000000000001, "data_time": 0.0018864870071411133, "grad_norm": 0.011752224061638117, "loss": 0.6866662263870239, "time": 0.8112723588943481, "epoch": 11, "memory": 28783, "step": 62280} +{"lr": 0.0005050000000000001, "data_time": 0.0018591165542602539, "grad_norm": 0.014484515134245157, "loss": 0.6853614449501038, "time": 0.81897132396698, "epoch": 11, "memory": 28783, "step": 62380} +{"lr": 0.0005050000000000001, "data_time": 0.0019700050354003905, "grad_norm": 0.014939905516803265, "loss": 0.6879810571670533, "time": 0.8347316265106202, "epoch": 11, "memory": 28783, "step": 62480} +{"lr": 0.0005050000000000001, "data_time": 0.0016882896423339843, "grad_norm": 0.017923609586432576, "loss": 0.6860221087932586, "time": 0.800758957862854, "epoch": 11, "memory": 28783, "step": 62580} +{"lr": 0.0005050000000000001, "data_time": 0.0018634557723999023, "grad_norm": 0.02136535132303834, "loss": 0.688265037536621, "time": 0.8851779699325562, "epoch": 11, "memory": 28783, "step": 62680} +{"lr": 0.0005050000000000001, "data_time": 0.002018284797668457, "grad_norm": 0.018396999686956406, "loss": 0.687425148487091, "time": 0.7992514610290528, "epoch": 11, "memory": 28783, "step": 62780} +{"lr": 0.0005050000000000001, "data_time": 0.0023943424224853516, "grad_norm": 0.012992601515725255, "loss": 0.6872276544570923, "time": 0.8223165512084961, "epoch": 11, "memory": 28783, "step": 62880} +{"lr": 0.0005050000000000001, "data_time": 0.002091670036315918, "grad_norm": 0.01296411354560405, "loss": 0.6859677195549011, "time": 0.9990283250808716, "epoch": 11, "memory": 28783, "step": 62980} +{"lr": 0.0005050000000000001, "data_time": 0.00213007926940918, "grad_norm": 0.014627908845432103, "loss": 0.6826976597309112, "time": 0.8523520469665528, "epoch": 11, "memory": 28783, "step": 63080} +{"lr": 0.0005050000000000001, "data_time": 0.0019238710403442383, "grad_norm": 0.017776364041492344, "loss": 0.6831452310085296, "time": 0.8890249729156494, "epoch": 11, "memory": 28783, "step": 63180} +{"lr": 0.0005050000000000001, "data_time": 0.001802206039428711, "grad_norm": 0.012028198200277984, "loss": 0.6871235847473145, "time": 0.8503433465957642, "epoch": 11, "memory": 28783, "step": 63280} +{"accuracy/top1": 100.0, "data_time": 0.21847601497874541, "time": 0.6325038741616642, "step": 11} +{"lr": 0.000427564939805086, "data_time": 0.003440999984741211, "grad_norm": 0.012254272896097974, "loss": 0.688452398777008, "time": 0.8018713235855103, "epoch": 12, "memory": 28783, "step": 63438} +{"lr": 0.000427564939805086, "data_time": 0.0027014970779418944, "grad_norm": 0.012671843508724124, "loss": 0.6879237651824951, "time": 0.8426581144332885, "epoch": 12, "memory": 28783, "step": 63538} +{"lr": 0.000427564939805086, "data_time": 0.0020052194595336914, "grad_norm": 0.012314799171872438, "loss": 0.6878165185451508, "time": 0.8584505319595337, "epoch": 12, "memory": 28783, "step": 63638} +{"lr": 0.000427564939805086, "data_time": 0.0015796899795532226, "grad_norm": 0.014028619066812098, "loss": 0.6865696728229522, "time": 0.8230879783630372, "epoch": 12, "memory": 28783, "step": 63738} +{"lr": 0.000427564939805086, "data_time": 0.0019037723541259766, "grad_norm": 0.011870158626697958, "loss": 0.6863193094730378, "time": 0.8921442031860352, "epoch": 12, "memory": 28783, "step": 63838} +{"lr": 0.000427564939805086, "data_time": 0.0021541357040405274, "grad_norm": 0.018259978853166103, "loss": 0.6882329642772674, "time": 0.8551560401916504, "epoch": 12, "memory": 28783, "step": 63938} +{"lr": 0.000427564939805086, "data_time": 0.0022708892822265623, "grad_norm": 0.016573956818319857, "loss": 0.6889025926589966, "time": 0.82143075466156, "epoch": 12, "memory": 28783, "step": 64038} +{"lr": 0.000427564939805086, "data_time": 0.0020264387130737305, "grad_norm": 0.013501499826088548, "loss": 0.6859869599342346, "time": 0.8742319107055664, "epoch": 12, "memory": 28783, "step": 64138} +{"lr": 0.000427564939805086, "data_time": 0.0019479036331176759, "grad_norm": 0.019222495844587684, "loss": 0.684797215461731, "time": 0.8399986982345581, "epoch": 12, "memory": 28783, "step": 64238} +{"lr": 0.000427564939805086, "data_time": 0.0019345760345458984, "grad_norm": 0.01394376807147637, "loss": 0.6886085152626038, "time": 0.9529498100280762, "epoch": 12, "memory": 28783, "step": 64338} +{"lr": 0.000427564939805086, "data_time": 0.0021494150161743162, "grad_norm": 0.019535945355892183, "loss": 0.6855676114559174, "time": 0.7990355014801025, "epoch": 12, "memory": 28783, "step": 64438} +{"lr": 0.000427564939805086, "data_time": 0.001722574234008789, "grad_norm": 0.00962800437118858, "loss": 0.6870296001434326, "time": 0.8601633548736572, "epoch": 12, "memory": 28783, "step": 64538} +{"lr": 0.000427564939805086, "data_time": 0.0017733335494995116, "grad_norm": 0.010389964492060245, "loss": 0.6857517063617706, "time": 0.8852148532867432, "epoch": 12, "memory": 28783, "step": 64638} +{"lr": 0.000427564939805086, "data_time": 0.0022205114364624023, "grad_norm": 0.015625704545527697, "loss": 0.6859215438365937, "time": 0.8484145641326905, "epoch": 12, "memory": 28783, "step": 64738} +{"lr": 0.000427564939805086, "data_time": 0.0019207477569580078, "grad_norm": 0.012254656321601942, "loss": 0.6875371158123016, "time": 0.8932466506958008, "epoch": 12, "memory": 28783, "step": 64838} +{"lr": 0.000427564939805086, "data_time": 0.001924896240234375, "grad_norm": 0.025284463725984096, "loss": 0.6850191652774811, "time": 0.8371482610702514, "epoch": 12, "memory": 28783, "step": 64938} +{"lr": 0.000427564939805086, "data_time": 0.0024759769439697266, "grad_norm": 0.01742397854104638, "loss": 0.6841047823429107, "time": 0.8798106908798218, "epoch": 12, "memory": 28783, "step": 65038} +{"lr": 0.000427564939805086, "data_time": 0.0018042325973510742, "grad_norm": 0.018168100458569824, "loss": 0.6866636395454406, "time": 0.840888237953186, "epoch": 12, "memory": 28783, "step": 65138} +{"lr": 0.000427564939805086, "data_time": 0.0020921707153320314, "grad_norm": 0.01569122385699302, "loss": 0.6895307660102844, "time": 0.8279127359390259, "epoch": 12, "memory": 28783, "step": 65238} +{"lr": 0.000427564939805086, "data_time": 0.0025150537490844726, "grad_norm": 0.02255082824267447, "loss": 0.6879225075244904, "time": 0.8909866333007812, "epoch": 12, "memory": 28783, "step": 65338} +{"lr": 0.000427564939805086, "data_time": 0.0018845796585083008, "grad_norm": 0.01949697150848806, "loss": 0.6868943929672241, "time": 0.8667906522750854, "epoch": 12, "memory": 28783, "step": 65438} +{"lr": 0.000427564939805086, "data_time": 0.002279186248779297, "grad_norm": 0.016585860983468592, "loss": 0.6881010830402374, "time": 0.8440192699432373, "epoch": 12, "memory": 28783, "step": 65538} +{"lr": 0.000427564939805086, "data_time": 0.0022717714309692383, "grad_norm": 0.022135975118726493, "loss": 0.6877662181854248, "time": 0.8422539234161377, "epoch": 12, "memory": 28783, "step": 65638} +{"lr": 0.000427564939805086, "data_time": 0.0018958330154418945, "grad_norm": 0.015082068298943341, "loss": 0.6859381020069122, "time": 0.8344122171401978, "epoch": 12, "memory": 28783, "step": 65738} +{"lr": 0.000427564939805086, "data_time": 0.001839280128479004, "grad_norm": 0.017079555289819838, "loss": 0.6860392153263092, "time": 0.8058897733688355, "epoch": 12, "memory": 28783, "step": 65838} +{"lr": 0.000427564939805086, "data_time": 0.0016717910766601562, "grad_norm": 0.010059173277113587, "loss": 0.6839419901371002, "time": 0.8589228868484498, "epoch": 12, "memory": 28783, "step": 65938} +{"lr": 0.000427564939805086, "data_time": 0.0015815973281860351, "grad_norm": 0.01895637606503442, "loss": 0.687496793270111, "time": 0.8400229454040528, "epoch": 12, "memory": 28783, "step": 66038} +{"lr": 0.000427564939805086, "data_time": 0.001855325698852539, "grad_norm": 0.017460882826708258, "loss": 0.6851868987083435, "time": 0.839257001876831, "epoch": 12, "memory": 28783, "step": 66138} +{"lr": 0.000427564939805086, "data_time": 0.0017693042755126953, "grad_norm": 0.0143052568892017, "loss": 0.6830495893955231, "time": 0.8398881673812866, "epoch": 12, "memory": 28783, "step": 66238} +{"lr": 0.000427564939805086, "data_time": 0.0019068956375122071, "grad_norm": 0.013714489690028132, "loss": 0.6875060558319092, "time": 0.8520808696746827, "epoch": 12, "memory": 28783, "step": 66338} +{"lr": 0.000427564939805086, "data_time": 0.0022751331329345704, "grad_norm": 0.015660790947731585, "loss": 0.6870501816272736, "time": 0.8310092687606812, "epoch": 12, "memory": 28783, "step": 66438} +{"lr": 0.000427564939805086, "data_time": 0.001861906051635742, "grad_norm": 0.015771122719161212, "loss": 0.6893124282360077, "time": 0.8851603507995606, "epoch": 12, "memory": 28783, "step": 66538} +{"lr": 0.000427564939805086, "data_time": 0.002104663848876953, "grad_norm": 0.01973067447543144, "loss": 0.6862702906131745, "time": 0.8477134704589844, "epoch": 12, "memory": 28783, "step": 66638} +{"lr": 0.000427564939805086, "data_time": 0.002760505676269531, "grad_norm": 0.017841457203030587, "loss": 0.6864173471927643, "time": 0.8686709880828858, "epoch": 12, "memory": 28783, "step": 66738} +{"lr": 0.000427564939805086, "data_time": 0.001983499526977539, "grad_norm": 0.013224598742090166, "loss": 0.6861699461936951, "time": 0.8156566143035888, "epoch": 12, "memory": 28783, "step": 66838} +{"lr": 0.000427564939805086, "data_time": 0.0026387929916381835, "grad_norm": 0.012288924073800445, "loss": 0.6868104100227356, "time": 0.846895170211792, "epoch": 12, "memory": 28783, "step": 66938} +{"lr": 0.000427564939805086, "data_time": 0.0017224550247192383, "grad_norm": 0.018949965585488827, "loss": 0.6843415439128876, "time": 0.8430649518966675, "epoch": 12, "memory": 28783, "step": 67038} +{"lr": 0.000427564939805086, "data_time": 0.0018741369247436523, "grad_norm": 0.016846169158816336, "loss": 0.6841327607631683, "time": 0.8603096246719361, "epoch": 12, "memory": 28783, "step": 67138} +{"lr": 0.000427564939805086, "data_time": 0.0016807317733764648, "grad_norm": 0.017878101440146565, "loss": 0.687576973438263, "time": 0.8255111455917359, "epoch": 12, "memory": 28783, "step": 67238} +{"lr": 0.000427564939805086, "data_time": 0.0020295381546020508, "grad_norm": 0.016764623462222517, "loss": 0.6865106284618377, "time": 0.9006153821945191, "epoch": 12, "memory": 28783, "step": 67338} +{"lr": 0.000427564939805086, "data_time": 0.0020602941513061523, "grad_norm": 0.0188767671585083, "loss": 0.6833829045295715, "time": 0.8561848878860474, "epoch": 12, "memory": 28783, "step": 67438} +{"lr": 0.000427564939805086, "data_time": 0.0020209789276123048, "grad_norm": 0.022730698401574045, "loss": 0.6891467928886413, "time": 0.8356692314147949, "epoch": 12, "memory": 28783, "step": 67538} +{"lr": 0.000427564939805086, "data_time": 0.0021651268005371095, "grad_norm": 0.013563672988675534, "loss": 0.6873798906803131, "time": 0.8504010677337647, "epoch": 12, "memory": 28783, "step": 67638} +{"lr": 0.000427564939805086, "data_time": 0.002049112319946289, "grad_norm": 0.018280728533864022, "loss": 0.6884930729866028, "time": 0.8261601686477661, "epoch": 12, "memory": 28783, "step": 67738} +{"lr": 0.000427564939805086, "data_time": 0.001821422576904297, "grad_norm": 0.016631564171984792, "loss": 0.6844547271728516, "time": 0.8422627210617065, "epoch": 12, "memory": 28783, "step": 67838} +{"lr": 0.000427564939805086, "data_time": 0.00184478759765625, "grad_norm": 0.018905861396342517, "loss": 0.6824126243591309, "time": 0.8438386201858521, "epoch": 12, "memory": 28783, "step": 67938} +{"lr": 0.000427564939805086, "data_time": 0.001901102066040039, "grad_norm": 0.01797306233784184, "loss": 0.6894153237342835, "time": 0.8111754179000854, "epoch": 12, "memory": 28783, "step": 68038} +{"lr": 0.000427564939805086, "data_time": 0.002787184715270996, "grad_norm": 0.02484312830492854, "loss": 0.6873823583126069, "time": 0.822121262550354, "epoch": 12, "memory": 28783, "step": 68138} +{"lr": 0.000427564939805086, "data_time": 0.0019865512847900392, "grad_norm": 0.02294129424262792, "loss": 0.6850340187549591, "time": 0.8418996334075928, "epoch": 12, "memory": 28783, "step": 68238} +{"lr": 0.000427564939805086, "data_time": 0.0019071578979492187, "grad_norm": 0.012277470075059683, "loss": 0.6879194438457489, "time": 0.8444482803344726, "epoch": 12, "memory": 28783, "step": 68338} +{"lr": 0.000427564939805086, "data_time": 0.002357935905456543, "grad_norm": 0.022407754976302385, "loss": 0.6852332472801208, "time": 0.8816360712051392, "epoch": 12, "memory": 28783, "step": 68438} +{"lr": 0.000427564939805086, "data_time": 0.002547550201416016, "grad_norm": 0.016252647247165443, "loss": 0.6869350075721741, "time": 0.8663325309753418, "epoch": 12, "memory": 28783, "step": 68538} +{"lr": 0.000427564939805086, "data_time": 0.0017745494842529297, "grad_norm": 0.012211003992706537, "loss": 0.6905552208423614, "time": 0.8967301368713378, "epoch": 12, "memory": 28783, "step": 68638} +{"lr": 0.000427564939805086, "data_time": 0.0021529197692871094, "grad_norm": 0.02307587442919612, "loss": 0.6870291054248809, "time": 0.8886263847351075, "epoch": 12, "memory": 28783, "step": 68738} +{"lr": 0.000427564939805086, "data_time": 0.0018274545669555663, "grad_norm": 0.01625542587134987, "loss": 0.6875106155872345, "time": 0.8600252628326416, "epoch": 12, "memory": 28783, "step": 68838} +{"lr": 0.000427564939805086, "data_time": 0.0021867275238037108, "grad_norm": 0.011149125744123013, "loss": 0.6883181154727935, "time": 0.8382859468460083, "epoch": 12, "memory": 28783, "step": 68938} +{"lr": 0.000427564939805086, "data_time": 0.0025548458099365233, "grad_norm": 0.018422122637275606, "loss": 0.6870428025722504, "time": 0.8882803678512573, "epoch": 12, "memory": 28783, "step": 69038} +{"accuracy/top1": 100.0, "data_time": 0.20626535135156968, "time": 0.6214081539827234, "step": 12} +{"lr": 0.0003520365877844012, "data_time": 0.2391516923904419, "grad_norm": 0.02247819406911731, "loss": 0.6879157781600952, "time": 0.8461340188980102, "epoch": 13, "memory": 28783, "step": 69196} +{"lr": 0.0003520365877844012, "data_time": 0.0018452882766723632, "grad_norm": 0.012166666146367789, "loss": 0.687698382139206, "time": 0.8045434713363647, "epoch": 13, "memory": 28783, "step": 69296} +{"lr": 0.0003520365877844012, "data_time": 0.0023088932037353517, "grad_norm": 0.019380286114756017, "loss": 0.6865039825439453, "time": 0.8118725061416626, "epoch": 13, "memory": 28783, "step": 69396} +{"lr": 0.0003520365877844012, "data_time": 0.003086972236633301, "grad_norm": 0.008478107361588627, "loss": 0.686888438463211, "time": 0.8681352376937866, "epoch": 13, "memory": 28783, "step": 69496} +{"lr": 0.0003520365877844012, "data_time": 0.0018507003784179687, "grad_norm": 0.015004638629034162, "loss": 0.685306864976883, "time": 0.863008451461792, "epoch": 13, "memory": 28783, "step": 69596} +{"lr": 0.0003520365877844012, "data_time": 0.0024507522583007814, "grad_norm": 0.019598237005993724, "loss": 0.6859407603740693, "time": 0.8286648988723755, "epoch": 13, "memory": 28783, "step": 69696} +{"lr": 0.0003520365877844012, "data_time": 0.0020800352096557615, "grad_norm": 0.018289242358878254, "loss": 0.6875308990478516, "time": 0.8506077527999878, "epoch": 13, "memory": 28783, "step": 69796} +{"lr": 0.0003520365877844012, "data_time": 0.0018233776092529297, "grad_norm": 0.013454599329270423, "loss": 0.684500253200531, "time": 0.8005702257156372, "epoch": 13, "memory": 28783, "step": 69896} +{"lr": 0.0003520365877844012, "data_time": 0.0031012535095214845, "grad_norm": 0.017343352455645798, "loss": 0.6883727550506592, "time": 0.8688923835754394, "epoch": 13, "memory": 28783, "step": 69996} +{"lr": 0.0003520365877844012, "data_time": 0.002783656120300293, "grad_norm": 0.010464380035409704, "loss": 0.6860755383968353, "time": 0.8936461687088013, "epoch": 13, "memory": 28783, "step": 70096} +{"lr": 0.0003520365877844012, "data_time": 0.002584958076477051, "grad_norm": 0.013679419201798737, "loss": 0.6870516419410706, "time": 0.865241527557373, "epoch": 13, "memory": 28783, "step": 70196} +{"lr": 0.0003520365877844012, "data_time": 0.0024837493896484376, "grad_norm": 0.01400958439335227, "loss": 0.6888535916805267, "time": 0.8799507141113281, "epoch": 13, "memory": 28783, "step": 70296} +{"lr": 0.0003520365877844012, "data_time": 0.0017689228057861327, "grad_norm": 0.019060352473752575, "loss": 0.6873691499233245, "time": 0.8606380939483642, "epoch": 13, "memory": 28783, "step": 70396} +{"lr": 0.0003520365877844012, "data_time": 0.001884794235229492, "grad_norm": 0.013557263580150903, "loss": 0.6864943146705628, "time": 0.8788732528686524, "epoch": 13, "memory": 28783, "step": 70496} +{"lr": 0.0003520365877844012, "data_time": 0.0027837038040161135, "grad_norm": 0.017442819301504642, "loss": 0.6879061222076416, "time": 0.8228353261947632, "epoch": 13, "memory": 28783, "step": 70596} +{"lr": 0.0003520365877844012, "data_time": 0.0019437313079833985, "grad_norm": 0.015117300814017654, "loss": 0.6872159004211426, "time": 0.8581367254257202, "epoch": 13, "memory": 28783, "step": 70696} +{"lr": 0.0003520365877844012, "data_time": 0.0028244733810424806, "grad_norm": 0.015494049922563136, "loss": 0.6887699007987976, "time": 0.9177424907684326, "epoch": 13, "memory": 28783, "step": 70796} +{"lr": 0.0003520365877844012, "data_time": 0.002637290954589844, "grad_norm": 0.016117434320040046, "loss": 0.6872215270996094, "time": 0.8286547660827637, "epoch": 13, "memory": 28783, "step": 70896} +{"lr": 0.0003520365877844012, "data_time": 0.0019123077392578125, "grad_norm": 0.01466503927949816, "loss": 0.6878536522388459, "time": 0.8192996263504029, "epoch": 13, "memory": 28783, "step": 70996} +{"lr": 0.0003520365877844012, "data_time": 0.002655959129333496, "grad_norm": 0.010116344201378524, "loss": 0.6876698732376099, "time": 0.8364684581756592, "epoch": 13, "memory": 28783, "step": 71096} +{"lr": 0.0003520365877844012, "data_time": 0.0020856142044067385, "grad_norm": 0.013932437263429165, "loss": 0.6877418458461761, "time": 0.7948732852935791, "epoch": 13, "memory": 28783, "step": 71196} +{"lr": 0.0003520365877844012, "data_time": 0.002506852149963379, "grad_norm": 0.02352831920143217, "loss": 0.6889092564582825, "time": 0.8805351734161377, "epoch": 13, "memory": 28783, "step": 71296} +{"lr": 0.0003520365877844012, "data_time": 0.0020704746246337892, "grad_norm": 0.018580973194912077, "loss": 0.6872140407562256, "time": 0.9044097661972046, "epoch": 13, "memory": 28783, "step": 71396} +{"lr": 0.0003520365877844012, "data_time": 0.003917074203491211, "grad_norm": 0.01591803561896086, "loss": 0.6890011310577393, "time": 0.8946885347366333, "epoch": 13, "memory": 28783, "step": 71496} +{"lr": 0.0003520365877844012, "data_time": 0.0026073694229125977, "grad_norm": 0.015283292019739747, "loss": 0.6860139489173889, "time": 0.8426134347915649, "epoch": 13, "memory": 28783, "step": 71596} +{"lr": 0.0003520365877844012, "data_time": 0.0028566360473632813, "grad_norm": 0.023452968802303076, "loss": 0.6867613911628723, "time": 0.8604297161102294, "epoch": 13, "memory": 28783, "step": 71696} +{"lr": 0.0003520365877844012, "data_time": 0.0025577306747436523, "grad_norm": 0.027375939395278694, "loss": 0.6850374102592468, "time": 0.8654968023300171, "epoch": 13, "memory": 28783, "step": 71796} +{"lr": 0.0003520365877844012, "data_time": 0.00260317325592041, "grad_norm": 0.01553009677445516, "loss": 0.6822042286396026, "time": 0.8916721820831299, "epoch": 13, "memory": 28783, "step": 71896} +{"lr": 0.0003520365877844012, "data_time": 0.0030301094055175783, "grad_norm": 0.009857999958330765, "loss": 0.683708256483078, "time": 0.841990613937378, "epoch": 13, "memory": 28783, "step": 71996} +{"lr": 0.0003520365877844012, "data_time": 0.001949000358581543, "grad_norm": 0.01007278913166374, "loss": 0.6851643741130828, "time": 0.8914402246475219, "epoch": 13, "memory": 28783, "step": 72096} +{"lr": 0.0003520365877844012, "data_time": 0.0021855831146240234, "grad_norm": 0.01911925037857145, "loss": 0.6894653737545013, "time": 0.8655587196350097, "epoch": 13, "memory": 28783, "step": 72196} +{"lr": 0.0003520365877844012, "data_time": 0.0020829439163208008, "grad_norm": 0.014841298246756196, "loss": 0.6904347121715546, "time": 0.9711669445037842, "epoch": 13, "memory": 28783, "step": 72296} +{"lr": 0.0003520365877844012, "data_time": 0.002265143394470215, "grad_norm": 0.02255669248988852, "loss": 0.6879143357276917, "time": 0.8268091201782226, "epoch": 13, "memory": 28783, "step": 72396} +{"lr": 0.0003520365877844012, "data_time": 0.002348661422729492, "grad_norm": 0.025222789077088238, "loss": 0.6871301174163819, "time": 0.8799536943435669, "epoch": 13, "memory": 28783, "step": 72496} +{"lr": 0.0003520365877844012, "data_time": 0.003020763397216797, "grad_norm": 0.018251256039366125, "loss": 0.6879877865314483, "time": 0.8517894983291626, "epoch": 13, "memory": 28783, "step": 72596} +{"lr": 0.0003520365877844012, "data_time": 0.0033486127853393556, "grad_norm": 0.015391803072998301, "loss": 0.6872004687786102, "time": 0.9024513006210327, "epoch": 13, "memory": 28783, "step": 72696} +{"lr": 0.0003520365877844012, "data_time": 0.002414250373840332, "grad_norm": 0.013337351003428921, "loss": 0.6876972556114197, "time": 0.8562380313873291, "epoch": 13, "memory": 28783, "step": 72796} +{"lr": 0.0003520365877844012, "data_time": 0.0018981456756591796, "grad_norm": 0.012450885854195803, "loss": 0.6886140763759613, "time": 0.851210069656372, "epoch": 13, "memory": 28783, "step": 72896} +{"lr": 0.0003520365877844012, "data_time": 0.002394366264343262, "grad_norm": 0.01784762986935675, "loss": 0.6905607402324676, "time": 0.8503563880920411, "epoch": 13, "memory": 28783, "step": 72996} +{"lr": 0.0003520365877844012, "data_time": 0.0029362916946411135, "grad_norm": 0.022130879014730452, "loss": 0.6833449304103851, "time": 0.8617544174194336, "epoch": 13, "memory": 28783, "step": 73096} +{"lr": 0.0003520365877844012, "data_time": 0.0031109094619750977, "grad_norm": 0.014414389850571752, "loss": 0.6847840309143066, "time": 0.8836791276931762, "epoch": 13, "memory": 28783, "step": 73196} +{"lr": 0.0003520365877844012, "data_time": 0.0033122539520263673, "grad_norm": 0.013657961692661046, "loss": 0.6869725584983826, "time": 0.8541029691696167, "epoch": 13, "memory": 28783, "step": 73296} +{"lr": 0.0003520365877844012, "data_time": 0.0024121284484863283, "grad_norm": 0.012511292798444628, "loss": 0.6887347280979157, "time": 0.8270941019058228, "epoch": 13, "memory": 28783, "step": 73396} +{"lr": 0.0003520365877844012, "data_time": 0.0024880170822143555, "grad_norm": 0.012370800855569542, "loss": 0.6899017512798309, "time": 0.8426279067993164, "epoch": 13, "memory": 28783, "step": 73496} +{"lr": 0.0003520365877844012, "data_time": 0.0025377273559570312, "grad_norm": 0.018190123606473207, "loss": 0.6852566301822662, "time": 0.8379014730453491, "epoch": 13, "memory": 28783, "step": 73596} +{"lr": 0.0003520365877844012, "data_time": 0.0025491952896118165, "grad_norm": 0.012276675552129745, "loss": 0.681931871175766, "time": 0.7893479585647583, "epoch": 13, "memory": 28783, "step": 73696} +{"lr": 0.0003520365877844012, "data_time": 0.0026648759841918944, "grad_norm": 0.023400061391294004, "loss": 0.6810989201068878, "time": 0.8150842189788818, "epoch": 13, "memory": 28783, "step": 73796} +{"lr": 0.0003520365877844012, "data_time": 0.0023877620697021484, "grad_norm": 0.012619656487368047, "loss": 0.6848697185516357, "time": 0.824254059791565, "epoch": 13, "memory": 28783, "step": 73896} +{"lr": 0.0003520365877844012, "data_time": 0.001892852783203125, "grad_norm": 0.01216881915461272, "loss": 0.6850947201251983, "time": 0.8250591278076171, "epoch": 13, "memory": 28783, "step": 73996} +{"lr": 0.0003520365877844012, "data_time": 0.0020848512649536133, "grad_norm": 0.0169877891195938, "loss": 0.6879559636116028, "time": 0.8067756414413452, "epoch": 13, "memory": 28783, "step": 74096} +{"lr": 0.0003520365877844012, "data_time": 0.0026565313339233397, "grad_norm": 0.013792618177831173, "loss": 0.6902757346630096, "time": 0.8280647039413452, "epoch": 13, "memory": 28783, "step": 74196} +{"lr": 0.0003520365877844012, "data_time": 0.0025008201599121095, "grad_norm": 0.015572900662664323, "loss": 0.6899118900299073, "time": 0.7984892368316651, "epoch": 13, "memory": 28783, "step": 74296} +{"lr": 0.0003520365877844012, "data_time": 0.0020769834518432617, "grad_norm": 0.018312862515449523, "loss": 0.6861032009124756, "time": 0.8125770092010498, "epoch": 13, "memory": 28783, "step": 74396} +{"lr": 0.0003520365877844012, "data_time": 0.0021264076232910155, "grad_norm": 0.01774419692810625, "loss": 0.6882382690906524, "time": 0.787130880355835, "epoch": 13, "memory": 28783, "step": 74496} +{"lr": 0.0003520365877844012, "data_time": 0.0022942304611206056, "grad_norm": 0.015399366314522923, "loss": 0.6872778475284577, "time": 0.7920415163040161, "epoch": 13, "memory": 28783, "step": 74596} +{"lr": 0.0003520365877844012, "data_time": 0.0026883602142333983, "grad_norm": 0.01367082130163908, "loss": 0.6894431829452514, "time": 0.806365180015564, "epoch": 13, "memory": 28783, "step": 74696} +{"lr": 0.0003520365877844012, "data_time": 0.0024976491928100585, "grad_norm": 0.01867692113155499, "loss": 0.6883374631404877, "time": 0.7716233491897583, "epoch": 13, "memory": 28783, "step": 74796} +{"accuracy/top1": 100.0, "data_time": 0.21666373926050522, "time": 0.6321520244373995, "step": 13} +{"lr": 0.00028027470262892445, "data_time": 0.0031208038330078126, "grad_norm": 0.021379628300201146, "loss": 0.6845173120498658, "time": 0.766830039024353, "epoch": 14, "memory": 28783, "step": 74954} +{"lr": 0.00028027470262892445, "data_time": 0.18351433277130128, "grad_norm": 0.012256648496259005, "loss": 0.6863982260227204, "time": 0.7944447278976441, "epoch": 14, "memory": 28783, "step": 75054} +{"lr": 0.00028027470262892445, "data_time": 0.0025827169418334963, "grad_norm": 0.011965835932642221, "loss": 0.6864938914775849, "time": 0.8210030555725097, "epoch": 14, "memory": 28783, "step": 75154} +{"lr": 0.00028027470262892445, "data_time": 0.0021210193634033205, "grad_norm": 0.015844987309537827, "loss": 0.6851883113384247, "time": 0.8173990964889526, "epoch": 14, "memory": 28783, "step": 75254} +{"lr": 0.00028027470262892445, "data_time": 0.002375888824462891, "grad_norm": 0.025583311403170227, "loss": 0.6866584420204163, "time": 0.9005350828170776, "epoch": 14, "memory": 28783, "step": 75354} +{"lr": 0.00028027470262892445, "data_time": 0.0028691768646240236, "grad_norm": 0.010602948523592204, "loss": 0.6881591320037842, "time": 0.7948050498962402, "epoch": 14, "memory": 28783, "step": 75454} +{"lr": 0.00028027470262892445, "data_time": 0.002345418930053711, "grad_norm": 0.013583542476408184, "loss": 0.6897511720657349, "time": 0.8507160902023315, "epoch": 14, "memory": 28783, "step": 75554} +{"lr": 0.00028027470262892445, "data_time": 0.0028682947158813477, "grad_norm": 0.01781910303980112, "loss": 0.6882295072078705, "time": 0.8167087316513062, "epoch": 14, "memory": 28783, "step": 75654} +{"lr": 0.00028027470262892445, "data_time": 0.002935957908630371, "grad_norm": 0.01549930670298636, "loss": 0.6884795784950256, "time": 0.7837977409362793, "epoch": 14, "memory": 28783, "step": 75754} +{"lr": 0.00028027470262892445, "data_time": 0.003167080879211426, "grad_norm": 0.014664636831730605, "loss": 0.6862571835517883, "time": 0.8247785568237305, "epoch": 14, "memory": 28783, "step": 75854} +{"lr": 0.00028027470262892445, "data_time": 0.0021590232849121095, "grad_norm": 0.014382861135527492, "loss": 0.6890288352966308, "time": 0.7866461992263794, "epoch": 14, "memory": 28783, "step": 75954} +{"lr": 0.00028027470262892445, "data_time": 0.002756071090698242, "grad_norm": 0.017164668021723627, "loss": 0.6865702033042907, "time": 0.8042606592178345, "epoch": 14, "memory": 28783, "step": 76054} +{"lr": 0.00028027470262892445, "data_time": 0.0023998737335205076, "grad_norm": 0.015464434633031487, "loss": 0.6834678173065185, "time": 0.8166307210922241, "epoch": 14, "memory": 28783, "step": 76154} +{"lr": 0.00028027470262892445, "data_time": 0.002602553367614746, "grad_norm": 0.014378064661286771, "loss": 0.6863162696361542, "time": 0.8312625408172607, "epoch": 14, "memory": 28783, "step": 76254} +{"lr": 0.00028027470262892445, "data_time": 0.0025655508041381838, "grad_norm": 0.00882388096069917, "loss": 0.6859928727149963, "time": 0.8546101331710816, "epoch": 14, "memory": 28783, "step": 76354} +{"lr": 0.00028027470262892445, "data_time": 0.0028191566467285155, "grad_norm": 0.010516448388807475, "loss": 0.687013590335846, "time": 0.8058027029037476, "epoch": 14, "memory": 28783, "step": 76454} +{"lr": 0.00028027470262892445, "data_time": 0.002307438850402832, "grad_norm": 0.022224119922611862, "loss": 0.6882230103015899, "time": 0.7847615718841553, "epoch": 14, "memory": 28783, "step": 76554} +{"lr": 0.00028027470262892445, "data_time": 0.0023659706115722657, "grad_norm": 0.011203650292009116, "loss": 0.6866700649261475, "time": 0.8896780729293823, "epoch": 14, "memory": 28783, "step": 76654} +{"lr": 0.00028027470262892445, "data_time": 0.002873349189758301, "grad_norm": 0.012446122476831079, "loss": 0.6870320022106171, "time": 0.7789308786392212, "epoch": 14, "memory": 28783, "step": 76754} +{"lr": 0.00028027470262892445, "data_time": 0.0033449888229370116, "grad_norm": 0.01665814840234816, "loss": 0.6875148057937622, "time": 0.8135859727859497, "epoch": 14, "memory": 28783, "step": 76854} +{"lr": 0.00028027470262892445, "data_time": 0.002987241744995117, "grad_norm": 0.017333590518683197, "loss": 0.6875333726406098, "time": 0.8344608306884765, "epoch": 14, "memory": 28783, "step": 76954} +{"lr": 0.00028027470262892445, "data_time": 0.002769279479980469, "grad_norm": 0.014409756287932396, "loss": 0.6865765690803528, "time": 0.8360893964767456, "epoch": 14, "memory": 28783, "step": 77054} +{"lr": 0.00028027470262892445, "data_time": 0.0028784513473510743, "grad_norm": 0.013263238349463791, "loss": 0.6874877333641052, "time": 0.7898921728134155, "epoch": 14, "memory": 28783, "step": 77154} +{"lr": 0.00028027470262892445, "data_time": 0.0021254777908325194, "grad_norm": 0.02061710082925856, "loss": 0.6844447910785675, "time": 0.777776575088501, "epoch": 14, "memory": 28783, "step": 77254} +{"lr": 0.00028027470262892445, "data_time": 0.0028898000717163088, "grad_norm": 0.019925792654976248, "loss": 0.6862314343452454, "time": 0.8320996999740601, "epoch": 14, "memory": 28783, "step": 77354} +{"lr": 0.00028027470262892445, "data_time": 0.0022061824798583984, "grad_norm": 0.01716822008602321, "loss": 0.684758996963501, "time": 0.784383225440979, "epoch": 14, "memory": 28783, "step": 77454} +{"lr": 0.00028027470262892445, "data_time": 0.0030763626098632814, "grad_norm": 0.01281963336514309, "loss": 0.683683580160141, "time": 0.7999552249908447, "epoch": 14, "memory": 28783, "step": 77554} +{"lr": 0.00028027470262892445, "data_time": 0.0020072460174560547, "grad_norm": 0.008046269562328234, "loss": 0.6877181529998779, "time": 0.7850032567977905, "epoch": 14, "memory": 28783, "step": 77654} +{"lr": 0.00028027470262892445, "data_time": 0.002471613883972168, "grad_norm": 0.018460680561838673, "loss": 0.6869895458221436, "time": 0.8241334915161133, "epoch": 14, "memory": 28783, "step": 77754} +{"lr": 0.00028027470262892445, "data_time": 0.002103853225708008, "grad_norm": 0.01441455683670938, "loss": 0.6888548135757446, "time": 0.878797173500061, "epoch": 14, "memory": 28783, "step": 77854} +{"lr": 0.00028027470262892445, "data_time": 0.0020147323608398437, "grad_norm": 0.021186025463975966, "loss": 0.6893752098083497, "time": 0.8197164535522461, "epoch": 14, "memory": 28783, "step": 77954} +{"lr": 0.00028027470262892445, "data_time": 0.002424931526184082, "grad_norm": 0.015030666696839034, "loss": 0.6881118655204773, "time": 0.8350815296173095, "epoch": 14, "memory": 28783, "step": 78054} +{"lr": 0.00028027470262892445, "data_time": 0.0022409439086914064, "grad_norm": 0.016399089235346765, "loss": 0.686488401889801, "time": 0.826201319694519, "epoch": 14, "memory": 28783, "step": 78154} +{"lr": 0.00028027470262892445, "data_time": 0.001822662353515625, "grad_norm": 0.013706484372960404, "loss": 0.6877138614654541, "time": 0.8358863592147827, "epoch": 14, "memory": 28783, "step": 78254} +{"lr": 0.00028027470262892445, "data_time": 0.0022010087966918947, "grad_norm": 0.012685577967204154, "loss": 0.6851814568042756, "time": 0.8375303268432617, "epoch": 14, "memory": 28783, "step": 78354} +{"lr": 0.00028027470262892445, "data_time": 0.001901698112487793, "grad_norm": 0.008980188029818236, "loss": 0.6860692381858826, "time": 0.8216911554336548, "epoch": 14, "memory": 28783, "step": 78454} +{"lr": 0.00028027470262892445, "data_time": 0.0020236968994140625, "grad_norm": 0.012190910009667277, "loss": 0.6853796422481537, "time": 0.8030661344528198, "epoch": 14, "memory": 28783, "step": 78554} +{"lr": 0.00028027470262892445, "data_time": 0.0025640249252319334, "grad_norm": 0.014230663341004401, "loss": 0.6871629476547241, "time": 0.7951892852783203, "epoch": 14, "memory": 28783, "step": 78654} +{"lr": 0.00028027470262892445, "data_time": 0.002682042121887207, "grad_norm": 0.017096467316150665, "loss": 0.6860884428024292, "time": 0.8175258636474609, "epoch": 14, "memory": 28783, "step": 78754} +{"lr": 0.00028027470262892445, "data_time": 0.0017179489135742188, "grad_norm": 0.00833444639283698, "loss": 0.6911397516727448, "time": 0.7847953557968139, "epoch": 14, "memory": 28783, "step": 78854} +{"lr": 0.00028027470262892445, "data_time": 0.0019270658493041992, "grad_norm": 0.011716116871684789, "loss": 0.688162910938263, "time": 0.7925777435302734, "epoch": 14, "memory": 28783, "step": 78954} +{"lr": 0.00028027470262892445, "data_time": 0.002113795280456543, "grad_norm": 0.008377769030630589, "loss": 0.6892408430576324, "time": 0.8271990060806275, "epoch": 14, "memory": 28783, "step": 79054} +{"lr": 0.00028027470262892445, "data_time": 0.0022131681442260744, "grad_norm": 0.02468175757676363, "loss": 0.6864872097969055, "time": 0.8087864875793457, "epoch": 14, "memory": 28783, "step": 79154} +{"lr": 0.00028027470262892445, "data_time": 0.001732206344604492, "grad_norm": 0.016062717139720916, "loss": 0.6868020832538605, "time": 0.8304159879684448, "epoch": 14, "memory": 28783, "step": 79254} +{"lr": 0.00028027470262892445, "data_time": 0.0020791053771972655, "grad_norm": 0.01238457476720214, "loss": 0.6881404280662536, "time": 0.8129947900772094, "epoch": 14, "memory": 28783, "step": 79354} +{"lr": 0.00028027470262892445, "data_time": 0.0028953790664672852, "grad_norm": 0.020267345802858473, "loss": 0.6883187830448151, "time": 0.8191087722778321, "epoch": 14, "memory": 28783, "step": 79454} +{"lr": 0.00028027470262892445, "data_time": 0.0019062042236328125, "grad_norm": 0.02627642937004566, "loss": 0.6887010395526886, "time": 0.8152916908264161, "epoch": 14, "memory": 28783, "step": 79554} +{"lr": 0.00028027470262892445, "data_time": 0.0030180692672729494, "grad_norm": 0.015312521567102522, "loss": 0.6873525321483612, "time": 0.7950273513793945, "epoch": 14, "memory": 28783, "step": 79654} +{"lr": 0.00028027470262892445, "data_time": 0.003254342079162598, "grad_norm": 0.01467336919158697, "loss": 0.68640296459198, "time": 0.8026775360107422, "epoch": 14, "memory": 28783, "step": 79754} +{"lr": 0.00028027470262892445, "data_time": 0.0019814014434814454, "grad_norm": 0.016030821396270767, "loss": 0.6836517930030823, "time": 0.8852525472640991, "epoch": 14, "memory": 28783, "step": 79854} +{"lr": 0.00028027470262892445, "data_time": 0.0019116163253784179, "grad_norm": 0.01687209411757067, "loss": 0.68647780418396, "time": 0.8558833599090576, "epoch": 14, "memory": 28783, "step": 79954} +{"lr": 0.00028027470262892445, "data_time": 0.0029361486434936524, "grad_norm": 0.019384603830985725, "loss": 0.6887512147426605, "time": 0.8053285598754882, "epoch": 14, "memory": 28783, "step": 80054} +{"lr": 0.00028027470262892445, "data_time": 0.0022010087966918947, "grad_norm": 0.019917760509997606, "loss": 0.686985844373703, "time": 0.8282535552978516, "epoch": 14, "memory": 28783, "step": 80154} +{"lr": 0.00028027470262892445, "data_time": 0.0020809412002563477, "grad_norm": 0.017564585572108625, "loss": 0.6845010399818421, "time": 0.9100208282470703, "epoch": 14, "memory": 28783, "step": 80254} +{"lr": 0.00028027470262892445, "data_time": 0.001949167251586914, "grad_norm": 0.015602843632223085, "loss": 0.6869618237018585, "time": 0.7959365844726562, "epoch": 14, "memory": 28783, "step": 80354} +{"lr": 0.00028027470262892445, "data_time": 0.0029601573944091795, "grad_norm": 0.014760750695131719, "loss": 0.6897834122180939, "time": 0.8478416681289673, "epoch": 14, "memory": 28783, "step": 80454} +{"lr": 0.00028027470262892445, "data_time": 0.003617119789123535, "grad_norm": 0.013280410110019148, "loss": 0.6880097806453704, "time": 0.793059229850769, "epoch": 14, "memory": 28783, "step": 80554} +{"accuracy/top1": 100.0, "data_time": 0.22324672867270076, "time": 0.6495649534113267, "step": 14} +{"lr": 0.0002140463001152259, "data_time": 0.002576136589050293, "grad_norm": 0.01352698476985097, "loss": 0.6871983647346497, "time": 0.765565824508667, "epoch": 15, "memory": 28783, "step": 80712} +{"lr": 0.0002140463001152259, "data_time": 0.0019073963165283203, "grad_norm": 0.017645997437648477, "loss": 0.6848306834697724, "time": 0.7879814386367798, "epoch": 15, "memory": 28783, "step": 80812} +{"lr": 0.0002140463001152259, "data_time": 0.001973462104797363, "grad_norm": 0.017264298698864877, "loss": 0.6913569450378418, "time": 0.831602144241333, "epoch": 15, "memory": 28783, "step": 80912} +{"lr": 0.0002140463001152259, "data_time": 0.0021908760070800783, "grad_norm": 0.016802679211832584, "loss": 0.6873147785663605, "time": 0.9419013500213623, "epoch": 15, "memory": 28783, "step": 81012} +{"lr": 0.0002140463001152259, "data_time": 0.0018620252609252929, "grad_norm": 0.015621217398438603, "loss": 0.6823112905025482, "time": 0.8435795783996582, "epoch": 15, "memory": 28783, "step": 81112} +{"lr": 0.0002140463001152259, "data_time": 0.002634286880493164, "grad_norm": 0.022311592614278196, "loss": 0.6869670987129212, "time": 0.8085411071777344, "epoch": 15, "memory": 28783, "step": 81212} +{"lr": 0.0002140463001152259, "data_time": 0.002593684196472168, "grad_norm": 0.019172173563856632, "loss": 0.6874598681926727, "time": 0.8183466911315918, "epoch": 15, "memory": 28783, "step": 81312} +{"lr": 0.0002140463001152259, "data_time": 0.0022911310195922853, "grad_norm": 0.01431084774667397, "loss": 0.6900889873504639, "time": 0.8187406063079834, "epoch": 15, "memory": 28783, "step": 81412} +{"lr": 0.0002140463001152259, "data_time": 0.002336883544921875, "grad_norm": 0.018483986612409353, "loss": 0.6866706728935241, "time": 0.8031017065048218, "epoch": 15, "memory": 28783, "step": 81512} +{"lr": 0.0002140463001152259, "data_time": 0.0019074201583862305, "grad_norm": 0.013430599425919353, "loss": 0.6870581150054932, "time": 0.8650548458099365, "epoch": 15, "memory": 28783, "step": 81612} +{"lr": 0.0002140463001152259, "data_time": 0.002767324447631836, "grad_norm": 0.023504098667763172, "loss": 0.6874674141407013, "time": 0.8802296876907348, "epoch": 15, "memory": 28783, "step": 81712} +{"lr": 0.0002140463001152259, "data_time": 0.002679324150085449, "grad_norm": 0.016823733667843045, "loss": 0.688474464416504, "time": 0.8106569290161133, "epoch": 15, "memory": 28783, "step": 81812} +{"lr": 0.0002140463001152259, "data_time": 0.002176046371459961, "grad_norm": 0.013821120385546237, "loss": 0.68550985455513, "time": 0.8387659788131714, "epoch": 15, "memory": 28783, "step": 81912} +{"lr": 0.0002140463001152259, "data_time": 0.002002310752868652, "grad_norm": 0.014627343835309147, "loss": 0.6831127405166626, "time": 0.8324304580688476, "epoch": 15, "memory": 28783, "step": 82012} +{"lr": 0.0002140463001152259, "data_time": 0.0021606922149658204, "grad_norm": 0.018662805599160493, "loss": 0.6869138181209564, "time": 0.8171991348266602, "epoch": 15, "memory": 28783, "step": 82112} +{"lr": 0.0002140463001152259, "data_time": 0.0020258188247680663, "grad_norm": 0.02056035230634734, "loss": 0.6865976035594941, "time": 0.8381967306137085, "epoch": 15, "memory": 28783, "step": 82212} +{"lr": 0.0002140463001152259, "data_time": 0.0026579380035400392, "grad_norm": 0.024374304851517082, "loss": 0.6868169367313385, "time": 0.8150118827819824, "epoch": 15, "memory": 28783, "step": 82312} +{"lr": 0.0002140463001152259, "data_time": 0.002124333381652832, "grad_norm": 0.014691986329853534, "loss": 0.6864885449409485, "time": 0.8040873050689697, "epoch": 15, "memory": 28783, "step": 82412} +{"lr": 0.0002140463001152259, "data_time": 0.0026203393936157227, "grad_norm": 0.017796302586793898, "loss": 0.6849068462848663, "time": 0.8241257190704345, "epoch": 15, "memory": 28783, "step": 82512} +{"lr": 0.0002140463001152259, "data_time": 0.001891016960144043, "grad_norm": 0.012056048377417028, "loss": 0.6855287492275238, "time": 0.7964436531066894, "epoch": 15, "memory": 28783, "step": 82612} +{"lr": 0.0002140463001152259, "data_time": 0.002974200248718262, "grad_norm": 0.020515192253515124, "loss": 0.6836224734783173, "time": 0.831844711303711, "epoch": 15, "memory": 28783, "step": 82712} +{"lr": 0.0002140463001152259, "data_time": 0.00243992805480957, "grad_norm": 0.015027505811303854, "loss": 0.6869742274284363, "time": 0.8170264959335327, "epoch": 15, "memory": 28783, "step": 82812} +{"lr": 0.0002140463001152259, "data_time": 0.0020360231399536135, "grad_norm": 0.010840017691953108, "loss": 0.6866565525531769, "time": 0.8417579650878906, "epoch": 15, "memory": 28783, "step": 82912} +{"lr": 0.0002140463001152259, "data_time": 0.00180509090423584, "grad_norm": 0.01052761425380595, "loss": 0.6853402435779572, "time": 0.8383221626281738, "epoch": 15, "memory": 28783, "step": 83012} +{"lr": 0.0002140463001152259, "data_time": 0.0038327455520629885, "grad_norm": 0.01995938392356038, "loss": 0.6869740664958954, "time": 0.8452216386795044, "epoch": 15, "memory": 28783, "step": 83112} +{"lr": 0.0002140463001152259, "data_time": 0.0024830102920532227, "grad_norm": 0.021384344017133117, "loss": 0.6879705846309662, "time": 0.8323727369308471, "epoch": 15, "memory": 28783, "step": 83212} +{"lr": 0.0002140463001152259, "data_time": 0.002127265930175781, "grad_norm": 0.017593488725833595, "loss": 0.6860734760761261, "time": 0.8288000345230102, "epoch": 15, "memory": 28783, "step": 83312} +{"lr": 0.0002140463001152259, "data_time": 0.00223696231842041, "grad_norm": 0.009376094792969525, "loss": 0.6862518072128296, "time": 0.7986794948577881, "epoch": 15, "memory": 28783, "step": 83412} +{"lr": 0.0002140463001152259, "data_time": 0.002404952049255371, "grad_norm": 0.014639827655628324, "loss": 0.6834733366966248, "time": 0.7947625637054443, "epoch": 15, "memory": 28783, "step": 83512} +{"lr": 0.0002140463001152259, "data_time": 0.001947021484375, "grad_norm": 0.014395470917224883, "loss": 0.6898046553134918, "time": 0.8145194530487061, "epoch": 15, "memory": 28783, "step": 83612} +{"lr": 0.0002140463001152259, "data_time": 0.0019422531127929687, "grad_norm": 0.016355888824909924, "loss": 0.6870259881019593, "time": 0.7987046480178833, "epoch": 15, "memory": 28783, "step": 83712} +{"lr": 0.0002140463001152259, "data_time": 0.0018370866775512696, "grad_norm": 0.011732210795162245, "loss": 0.6852873206138611, "time": 0.8341565608978272, "epoch": 15, "memory": 28783, "step": 83812} +{"lr": 0.0002140463001152259, "data_time": 0.0018714189529418946, "grad_norm": 0.02023541498929262, "loss": 0.6886974394321441, "time": 0.8202132463455201, "epoch": 15, "memory": 28783, "step": 83912} +{"lr": 0.0002140463001152259, "data_time": 0.003199911117553711, "grad_norm": 0.01677515732590109, "loss": 0.6890270471572876, "time": 0.866722846031189, "epoch": 15, "memory": 28783, "step": 84012} +{"lr": 0.0002140463001152259, "data_time": 0.001996612548828125, "grad_norm": 0.019989075209014117, "loss": 0.6904678165912628, "time": 0.7882216930389404, "epoch": 15, "memory": 28783, "step": 84112} +{"lr": 0.0002140463001152259, "data_time": 0.0018927335739135742, "grad_norm": 0.02138390401378274, "loss": 0.6856461286544799, "time": 0.802038311958313, "epoch": 15, "memory": 28783, "step": 84212} +{"lr": 0.0002140463001152259, "data_time": 0.002139759063720703, "grad_norm": 0.016336982301436365, "loss": 0.6845249235630035, "time": 0.8265966892242431, "epoch": 15, "memory": 28783, "step": 84312} +{"lr": 0.0002140463001152259, "data_time": 0.002336001396179199, "grad_norm": 0.011610458232462406, "loss": 0.6858574390411377, "time": 0.8583698511123657, "epoch": 15, "memory": 28783, "step": 84412} +{"lr": 0.0002140463001152259, "data_time": 0.0029568195343017576, "grad_norm": 0.015989073656965048, "loss": 0.6879020512104035, "time": 0.8427110195159913, "epoch": 15, "memory": 28783, "step": 84512} +{"lr": 0.0002140463001152259, "data_time": 0.0027734041213989258, "grad_norm": 0.01457352500874549, "loss": 0.6869529604911804, "time": 0.8642256259918213, "epoch": 15, "memory": 28783, "step": 84612} +{"lr": 0.0002140463001152259, "data_time": 0.003186631202697754, "grad_norm": 0.01722420649603009, "loss": 0.688180935382843, "time": 0.8754186630249023, "epoch": 15, "memory": 28783, "step": 84712} +{"lr": 0.0002140463001152259, "data_time": 0.0019004106521606444, "grad_norm": 0.015163661120459438, "loss": 0.687413090467453, "time": 0.8870687007904052, "epoch": 15, "memory": 28783, "step": 84812} +{"lr": 0.0002140463001152259, "data_time": 0.0019695043563842775, "grad_norm": 0.01540433221962303, "loss": 0.6843005776405334, "time": 0.8672104835510254, "epoch": 15, "memory": 28783, "step": 84912} +{"lr": 0.0002140463001152259, "data_time": 0.0018759489059448243, "grad_norm": 0.02686847122386098, "loss": 0.686324667930603, "time": 0.8680312871932984, "epoch": 15, "memory": 28783, "step": 85012} +{"lr": 0.0002140463001152259, "data_time": 0.0021885395050048827, "grad_norm": 0.015108896105084568, "loss": 0.6933867156505584, "time": 0.8587390899658203, "epoch": 15, "memory": 28783, "step": 85112} +{"lr": 0.0002140463001152259, "data_time": 0.0026365995407104494, "grad_norm": 0.017090910207480194, "loss": 0.6857710361480713, "time": 0.7982375144958496, "epoch": 15, "memory": 28783, "step": 85212} +{"lr": 0.0002140463001152259, "data_time": 0.002439737319946289, "grad_norm": 0.02905628685839474, "loss": 0.6862540483474732, "time": 0.8567112207412719, "epoch": 15, "memory": 28783, "step": 85312} +{"lr": 0.0002140463001152259, "data_time": 0.0019452571868896484, "grad_norm": 0.01718382416293025, "loss": 0.6862488448619842, "time": 0.8861680507659913, "epoch": 15, "memory": 28783, "step": 85412} +{"lr": 0.0002140463001152259, "data_time": 0.0026148319244384765, "grad_norm": 0.013659542216919363, "loss": 0.6856607317924499, "time": 0.920865511894226, "epoch": 15, "memory": 28783, "step": 85512} +{"lr": 0.0002140463001152259, "data_time": 0.0019452571868896484, "grad_norm": 0.013395216770004482, "loss": 0.6882145464420318, "time": 0.8700824499130249, "epoch": 15, "memory": 28783, "step": 85612} +{"lr": 0.0002140463001152259, "data_time": 0.0024453878402709963, "grad_norm": 0.011258493200875819, "loss": 0.6888556718826294, "time": 0.9163310050964355, "epoch": 15, "memory": 28783, "step": 85712} +{"lr": 0.0002140463001152259, "data_time": 0.0019266843795776368, "grad_norm": 0.021566955652087928, "loss": 0.6857811272144317, "time": 0.9290709495544434, "epoch": 15, "memory": 28783, "step": 85812} +{"lr": 0.0002140463001152259, "data_time": 0.0024388313293457033, "grad_norm": 0.009459848949336446, "loss": 0.6900098919868469, "time": 0.8792054414749145, "epoch": 15, "memory": 28783, "step": 85912} +{"lr": 0.0002140463001152259, "data_time": 0.0020087480545043944, "grad_norm": 0.01949253408238292, "loss": 0.6852916657924653, "time": 0.8105745553970337, "epoch": 15, "memory": 28783, "step": 86012} +{"lr": 0.0002140463001152259, "data_time": 0.00248410701751709, "grad_norm": 0.010425873077474534, "loss": 0.6898028314113617, "time": 0.7689057350158691, "epoch": 15, "memory": 28783, "step": 86112} +{"lr": 0.0002140463001152259, "data_time": 0.002334284782409668, "grad_norm": 0.01594096914632246, "loss": 0.6871099531650543, "time": 0.8460749626159668, "epoch": 15, "memory": 28783, "step": 86212} +{"lr": 0.0002140463001152259, "data_time": 0.002018570899963379, "grad_norm": 0.01481670760549605, "loss": 0.6843600749969483, "time": 0.8235304117202759, "epoch": 15, "memory": 28783, "step": 86312} +{"accuracy/top1": 100.0, "data_time": 0.23661115590263815, "time": 0.6512091580559226, "step": 15} +{"lr": 0.00015498214331265904, "data_time": 0.22658779621124267, "grad_norm": 0.016640614450443535, "loss": 0.6875964224338531, "time": 0.8516465902328492, "epoch": 16, "memory": 28783, "step": 86470} +{"lr": 0.00015498214331265904, "data_time": 0.0023043155670166016, "grad_norm": 0.01285829917760566, "loss": 0.6883208692073822, "time": 0.8204551219940186, "epoch": 16, "memory": 28783, "step": 86570} +{"lr": 0.00015498214331265904, "data_time": 0.002229595184326172, "grad_norm": 0.014896116405725478, "loss": 0.6885277032852173, "time": 0.8687238931655884, "epoch": 16, "memory": 28783, "step": 86670} +{"lr": 0.00015498214331265904, "data_time": 0.002166485786437988, "grad_norm": 0.01971497352933511, "loss": 0.6858066618442535, "time": 0.8202325105667114, "epoch": 16, "memory": 28783, "step": 86770} +{"lr": 0.00015498214331265904, "data_time": 0.0019934892654418944, "grad_norm": 0.016891338932327925, "loss": 0.6843840420246124, "time": 0.8294904708862305, "epoch": 16, "memory": 28783, "step": 86870} +{"lr": 0.00015498214331265904, "data_time": 0.0018459558486938477, "grad_norm": 0.019779964530607687, "loss": 0.6868058502674103, "time": 0.765126085281372, "epoch": 16, "memory": 28783, "step": 86970} +{"lr": 0.00015498214331265904, "data_time": 0.0023764848709106447, "grad_norm": 0.013541481946595013, "loss": 0.6889892041683197, "time": 0.8584175348281861, "epoch": 16, "memory": 28783, "step": 87070} +{"lr": 0.00015498214331265904, "data_time": 0.003218936920166016, "grad_norm": 0.025307168252766132, "loss": 0.6882204055786133, "time": 0.7579382181167602, "epoch": 16, "memory": 28783, "step": 87170} +{"lr": 0.00015498214331265904, "data_time": 0.00200347900390625, "grad_norm": 0.012366747169289738, "loss": 0.6888687908649445, "time": 0.8440059900283814, "epoch": 16, "memory": 28783, "step": 87270} +{"lr": 0.00015498214331265904, "data_time": 0.0028717041015625, "grad_norm": 0.015069715020945295, "loss": 0.6897814571857452, "time": 0.7684584140777588, "epoch": 16, "memory": 28783, "step": 87370} +{"lr": 0.00015498214331265904, "data_time": 0.0020277738571166993, "grad_norm": 0.01779211309039965, "loss": 0.6848843812942504, "time": 0.7331821918487549, "epoch": 16, "memory": 28783, "step": 87470} +{"lr": 0.00015498214331265904, "data_time": 0.0019935846328735353, "grad_norm": 0.011437329440377652, "loss": 0.6870453238487244, "time": 0.7904813528060913, "epoch": 16, "memory": 28783, "step": 87570} +{"lr": 0.00015498214331265904, "data_time": 0.0022403478622436525, "grad_norm": 0.013743928447365761, "loss": 0.6884121537208557, "time": 0.7858628511428833, "epoch": 16, "memory": 28783, "step": 87670} +{"lr": 0.00015498214331265904, "data_time": 0.0020228147506713865, "grad_norm": 0.02309629893861711, "loss": 0.6865698575973511, "time": 0.7413806200027466, "epoch": 16, "memory": 28783, "step": 87770} +{"lr": 0.00015498214331265904, "data_time": 0.001766061782836914, "grad_norm": 0.020602161018177866, "loss": 0.6873650968074798, "time": 0.7455927610397339, "epoch": 16, "memory": 28783, "step": 87870} +{"lr": 0.00015498214331265904, "data_time": 0.0018593311309814454, "grad_norm": 0.01645621273200959, "loss": 0.6879251003265381, "time": 0.8247413873672486, "epoch": 16, "memory": 28783, "step": 87970} +{"lr": 0.00015498214331265904, "data_time": 0.0019395828247070312, "grad_norm": 0.012563132867217063, "loss": 0.6880704820156097, "time": 0.7488123416900635, "epoch": 16, "memory": 28783, "step": 88070} +{"lr": 0.00015498214331265904, "data_time": 0.003758525848388672, "grad_norm": 0.015504688181681558, "loss": 0.6865750610828399, "time": 0.8109008312225342, "epoch": 16, "memory": 28783, "step": 88170} +{"lr": 0.00015498214331265904, "data_time": 0.0017815351486206055, "grad_norm": 0.017982418416067958, "loss": 0.6867218971252441, "time": 0.8089564800262451, "epoch": 16, "memory": 28783, "step": 88270} diff --git a/swin_small_4xb256_1e-3lr_5m/20230531_183239/vis_data/config.py b/swin_small_4xb256_1e-3lr_5m/20230531_183239/vis_data/config.py new file mode 100644 index 0000000000000000000000000000000000000000..96ad3949be10f7f7b050950edab63a0911fe12b4 --- /dev/null +++ b/swin_small_4xb256_1e-3lr_5m/20230531_183239/vis_data/config.py @@ -0,0 +1,175 @@ +optim_wrapper = dict( + optimizer=dict( + type='AdamW', + lr=0.001, + weight_decay=0.05, + eps=1e-08, + betas=(0.9, 0.999), + _scope_='mmpretrain'), + paramwise_cfg=dict( + norm_decay_mult=0.0, + bias_decay_mult=0.0, + flat_decay_mult=0.0, + custom_keys=dict({ + '.absolute_pos_embed': dict(decay_mult=0.0), + '.relative_position_bias_table': dict(decay_mult=0.0) + })), + type='AmpOptimWrapper', + dtype='bfloat16', + clip_grad=dict(max_norm=5.0)) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=True, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=20, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=1024) +model = dict( + type='ImageClassifier', + backbone=dict( + type='SwinTransformer', arch='small', img_size=224, + drop_path_rate=0.3), + neck=dict(type='GlobalAveragePooling'), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=768, + init_cfg=None, + loss=dict( + type='LabelSmoothLoss', label_smooth_val=0.1, mode='original'), + cal_acc=False), + init_cfg=[ + dict(type='TruncNormal', layer='Linear', std=0.02, bias=0.0), + dict(type='Constant', layer='LayerNorm', val=1.0, bias=0.0) + ], + train_cfg=None) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stablediffusionV1-5R2-dpmsolver-25-5m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/swin_small_4xb256_1e-3lr_5m' diff --git a/swin_small_4xb256_1e-3lr_5m/20230531_183239/vis_data/events.out.tfevents.1685529167.SH-IDC1-10-140-24-130.3627.0 b/swin_small_4xb256_1e-3lr_5m/20230531_183239/vis_data/events.out.tfevents.1685529167.SH-IDC1-10-140-24-130.3627.0 new file mode 100644 index 0000000000000000000000000000000000000000..bbd61229fa2229330b6a9b399418d7c5a8c6b520 --- /dev/null +++ b/swin_small_4xb256_1e-3lr_5m/20230531_183239/vis_data/events.out.tfevents.1685529167.SH-IDC1-10-140-24-130.3627.0 @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:6b982f8196a96ea5cdd550fc572bcc613f4ef4477ad8236fb08ea2881cdb6965 +size 1349161 diff --git a/swin_small_4xb256_1e-3lr_5m/20230531_183239/vis_data/scalars.json b/swin_small_4xb256_1e-3lr_5m/20230531_183239/vis_data/scalars.json new file mode 100644 index 0000000000000000000000000000000000000000..9138437d0403f42823af9db0b01021b12629110e --- /dev/null +++ b/swin_small_4xb256_1e-3lr_5m/20230531_183239/vis_data/scalars.json @@ -0,0 +1,889 @@ +{"lr": 0.001, "data_time": 0.33590447902679443, "grad_norm": 1.187183529138565, "loss": 0.6847400009632111, "time": 0.8413348197937012, "epoch": 1, "memory": 28798, "step": 100} +{"lr": 0.001, "data_time": 0.21662020683288574, "grad_norm": 1.7217679768800735, "loss": 0.6908431828022004, "time": 0.8396015882492065, "epoch": 1, "memory": 28798, "step": 200} +{"lr": 0.001, "data_time": 0.002046799659729004, "grad_norm": 0.7791802749037743, "loss": 0.685979676246643, "time": 0.8650855302810669, "epoch": 1, "memory": 28798, "step": 300} +{"lr": 0.001, "data_time": 0.0023226499557495116, "grad_norm": 1.8430660009384154, "loss": 0.6948411226272583, "time": 0.9164520978927613, "epoch": 1, "memory": 28798, "step": 400} +{"lr": 0.001, "data_time": 0.0018687725067138671, "grad_norm": 0.6885813616216183, "loss": 0.6911965548992157, "time": 0.9577422380447388, "epoch": 1, "memory": 28798, "step": 500} +{"lr": 0.001, "data_time": 0.0020292520523071287, "grad_norm": 0.6634811967611313, "loss": 0.6864281952381134, "time": 0.8940370321273804, "epoch": 1, "memory": 28798, "step": 600} +{"lr": 0.001, "data_time": 0.0021838903427124023, "grad_norm": 0.6898281827569008, "loss": 0.6847228586673737, "time": 0.9111965894699097, "epoch": 1, "memory": 28798, "step": 700} +{"lr": 0.001, "data_time": 0.001630878448486328, "grad_norm": 1.3765226341784, "loss": 0.6887052834033967, "time": 0.943936800956726, "epoch": 1, "memory": 28798, "step": 800} +{"lr": 0.001, "data_time": 0.0016585111618041993, "grad_norm": 0.8592354223132134, "loss": 0.6897273421287536, "time": 0.9013750314712524, "epoch": 1, "memory": 28798, "step": 900} +{"lr": 0.001, "data_time": 0.0018284320831298828, "grad_norm": 0.518747566640377, "loss": 0.6888404786586761, "time": 1.1501312017440797, "epoch": 1, "memory": 28798, "step": 1000} +{"lr": 0.001, "data_time": 0.0016209840774536132, "grad_norm": 0.5395256511867046, "loss": 0.688236290216446, "time": 1.0908938884735107, "epoch": 1, "memory": 28798, "step": 1100} +{"lr": 0.001, "data_time": 0.0016809463500976562, "grad_norm": 0.9257483318448066, "loss": 0.6897540330886841, "time": 1.7967471361160279, "epoch": 1, "memory": 28798, "step": 1200} +{"lr": 0.001, "data_time": 0.001671147346496582, "grad_norm": 0.3737223416566849, "loss": 0.6879740357398987, "time": 0.9751446723937989, "epoch": 1, "memory": 28798, "step": 1300} +{"lr": 0.001, "data_time": 0.0016964435577392577, "grad_norm": 0.7635870240628719, "loss": 0.690438163280487, "time": 0.8699488162994384, "epoch": 1, "memory": 28798, "step": 1400} +{"lr": 0.001, "data_time": 0.0015594482421875, "grad_norm": 0.4468700304627419, "loss": 0.6883688926696777, "time": 0.9833091259002685, "epoch": 1, "memory": 28798, "step": 1500} +{"lr": 0.001, "data_time": 0.001657557487487793, "grad_norm": 0.21296481229364872, "loss": 0.6855560719966889, "time": 0.8429322481155396, "epoch": 1, "memory": 28798, "step": 1600} +{"lr": 0.001, "data_time": 0.001414346694946289, "grad_norm": 0.3191097557544708, "loss": 0.6899114906787872, "time": 0.9023676156997681, "epoch": 1, "memory": 28798, "step": 1700} +{"lr": 0.001, "data_time": 0.0015766620635986328, "grad_norm": 0.38395332768559454, "loss": 0.6843568980693817, "time": 0.9407166719436646, "epoch": 1, "memory": 28798, "step": 1800} +{"lr": 0.001, "data_time": 0.0015876293182373047, "grad_norm": 0.20893616713583468, "loss": 0.6874257266521454, "time": 0.903137493133545, "epoch": 1, "memory": 28798, "step": 1900} +{"lr": 0.001, "data_time": 0.0015334606170654297, "grad_norm": 0.29613580070436, "loss": 0.6882599294185638, "time": 1.0162176609039306, "epoch": 1, "memory": 28798, "step": 2000} +{"lr": 0.001, "data_time": 0.0019433021545410157, "grad_norm": 0.23353887908160686, "loss": 0.6890283942222595, "time": 0.967284870147705, "epoch": 1, "memory": 28798, "step": 2100} +{"lr": 0.001, "data_time": 0.0016798734664916991, "grad_norm": 0.27702537178993225, "loss": 0.6872062087059021, "time": 1.1158102035522461, "epoch": 1, "memory": 28798, "step": 2200} +{"lr": 0.001, "data_time": 0.0017933845520019531, "grad_norm": 0.16620225086808205, "loss": 0.6852505743503571, "time": 1.0250746726989746, "epoch": 1, "memory": 28798, "step": 2300} +{"lr": 0.001, "data_time": 0.0013074159622192382, "grad_norm": 0.16672469899058343, "loss": 0.6863704442977905, "time": 0.9309884548187256, "epoch": 1, "memory": 28798, "step": 2400} +{"lr": 0.001, "data_time": 0.0020125389099121095, "grad_norm": 0.28441546857357025, "loss": 0.6850416421890259, "time": 0.855058741569519, "epoch": 1, "memory": 28798, "step": 2500} +{"lr": 0.001, "data_time": 0.0016599655151367187, "grad_norm": 0.2358134739100933, "loss": 0.686743825674057, "time": 0.9001273393630982, "epoch": 1, "memory": 28798, "step": 2600} +{"lr": 0.001, "data_time": 0.0015526533126831055, "grad_norm": 0.26860551312565806, "loss": 0.6846187889575959, "time": 0.8460203647613526, "epoch": 1, "memory": 28798, "step": 2700} +{"lr": 0.001, "data_time": 0.0017335653305053712, "grad_norm": 0.18999040648341178, "loss": 0.6850415289402008, "time": 0.9114830255508423, "epoch": 1, "memory": 28798, "step": 2800} +{"lr": 0.001, "data_time": 0.001981043815612793, "grad_norm": 0.27065986320376395, "loss": 0.6882188379764557, "time": 0.9323644161224365, "epoch": 1, "memory": 28798, "step": 2900} +{"lr": 0.001, "data_time": 0.0016637325286865234, "grad_norm": 0.17620864287018775, "loss": 0.6837712287902832, "time": 0.8893620491027832, "epoch": 1, "memory": 28798, "step": 3000} +{"lr": 0.001, "data_time": 0.0017996549606323243, "grad_norm": 0.1975460097193718, "loss": 0.6860647797584534, "time": 0.997292423248291, "epoch": 1, "memory": 28798, "step": 3100} +{"lr": 0.001, "data_time": 0.0017212152481079102, "grad_norm": 0.18541229702532291, "loss": 0.6901660084724426, "time": 0.8814517498016358, "epoch": 1, "memory": 28798, "step": 3200} +{"lr": 0.001, "data_time": 0.001845979690551758, "grad_norm": 0.09229977168142796, "loss": 0.6833701193332672, "time": 0.9323680400848389, "epoch": 1, "memory": 28798, "step": 3300} +{"lr": 0.001, "data_time": 0.001603388786315918, "grad_norm": 0.19935780689120292, "loss": 0.6885235249996186, "time": 1.0112058401107789, "epoch": 1, "memory": 28798, "step": 3400} +{"lr": 0.001, "data_time": 0.001819014549255371, "grad_norm": 0.16067650243639947, "loss": 0.6889486908912659, "time": 0.895413875579834, "epoch": 1, "memory": 28798, "step": 3500} +{"lr": 0.001, "data_time": 0.002013564109802246, "grad_norm": 0.14938931372016667, "loss": 0.6931148529052734, "time": 0.9706980943679809, "epoch": 1, "memory": 28798, "step": 3600} +{"lr": 0.001, "data_time": 0.0018714666366577148, "grad_norm": 0.12052799835801124, "loss": 0.6882530331611634, "time": 0.9076480388641357, "epoch": 1, "memory": 28798, "step": 3700} +{"lr": 0.001, "data_time": 0.0018136739730834962, "grad_norm": 0.13805863428860904, "loss": 0.6835283875465393, "time": 0.7688174724578858, "epoch": 1, "memory": 28798, "step": 3800} +{"lr": 0.001, "data_time": 0.0017467737197875977, "grad_norm": 0.18656774386763572, "loss": 0.6844374060630798, "time": 0.8669900178909302, "epoch": 1, "memory": 28798, "step": 3900} +{"lr": 0.001, "data_time": 0.0016543149948120117, "grad_norm": 0.18251721002161503, "loss": 0.6876611471176147, "time": 0.8675621271133422, "epoch": 1, "memory": 28798, "step": 4000} +{"lr": 0.001, "data_time": 0.0021095514297485352, "grad_norm": 0.12997964397072792, "loss": 0.6865510106086731, "time": 0.7381432294845581, "epoch": 1, "memory": 28798, "step": 4100} +{"lr": 0.001, "data_time": 0.001603245735168457, "grad_norm": 0.13632102496922016, "loss": 0.6867913782596589, "time": 0.7909459352493287, "epoch": 1, "memory": 28798, "step": 4200} +{"lr": 0.001, "data_time": 0.0017872333526611328, "grad_norm": 0.13055530302226542, "loss": 0.6876545310020447, "time": 0.8414000749588013, "epoch": 1, "memory": 28798, "step": 4300} +{"lr": 0.001, "data_time": 0.001677727699279785, "grad_norm": 0.7684229619801044, "loss": 0.6813175678253174, "time": 0.8178308010101318, "epoch": 1, "memory": 28798, "step": 4400} +{"lr": 0.001, "data_time": 0.0015346765518188476, "grad_norm": 0.153719025850296, "loss": 0.6841858744621276, "time": 0.8224682569503784, "epoch": 1, "memory": 28798, "step": 4500} +{"lr": 0.001, "data_time": 0.0015207052230834961, "grad_norm": 0.10010228492319584, "loss": 0.6867010712623596, "time": 0.8346450090408325, "epoch": 1, "memory": 28798, "step": 4600} +{"lr": 0.001, "data_time": 0.0015330553054809571, "grad_norm": 0.09172356463968753, "loss": 0.6880784034729004, "time": 0.7683933496475219, "epoch": 1, "memory": 28798, "step": 4700} +{"lr": 0.001, "data_time": 0.0015619754791259765, "grad_norm": 0.09726331569254398, "loss": 0.6853234350681305, "time": 0.8191373586654663, "epoch": 1, "memory": 28798, "step": 4800} +{"lr": 0.001, "data_time": 0.0017874717712402343, "grad_norm": 0.10995372105389833, "loss": 0.6863053739070892, "time": 0.7955687046051025, "epoch": 1, "memory": 28798, "step": 4900} +{"lr": 0.001, "data_time": 0.0015886068344116212, "grad_norm": 0.16352200247347354, "loss": 0.6862053871154785, "time": 0.8013665676116943, "epoch": 1, "memory": 28798, "step": 5000} +{"lr": 0.001, "data_time": 0.0019288778305053711, "grad_norm": 0.12119524236768484, "loss": 0.6847285449504852, "time": 0.7832417964935303, "epoch": 1, "memory": 28798, "step": 5100} +{"lr": 0.001, "data_time": 0.0018793821334838867, "grad_norm": 0.1053498338907957, "loss": 0.6855199933052063, "time": 0.7722936630249023, "epoch": 1, "memory": 28798, "step": 5200} +{"lr": 0.001, "data_time": 0.001998138427734375, "grad_norm": 0.11447365432977677, "loss": 0.6891373336315155, "time": 0.839335298538208, "epoch": 1, "memory": 28798, "step": 5300} +{"lr": 0.001, "data_time": 0.002456045150756836, "grad_norm": 0.10331161096692085, "loss": 0.6884491860866546, "time": 0.8279985427856446, "epoch": 1, "memory": 28798, "step": 5400} +{"lr": 0.001, "data_time": 0.0018787860870361328, "grad_norm": 0.13169370582327247, "loss": 0.6850335359573364, "time": 0.8244468450546265, "epoch": 1, "memory": 28798, "step": 5500} +{"lr": 0.001, "data_time": 0.0016954421997070312, "grad_norm": 0.09026205968111753, "loss": 0.6849363386631012, "time": 0.8310160636901855, "epoch": 1, "memory": 28798, "step": 5600} +{"lr": 0.001, "data_time": 0.0016336441040039062, "grad_norm": 0.1151382902637124, "loss": 0.6874211132526398, "time": 0.8125093698501586, "epoch": 1, "memory": 28798, "step": 5700} +{"accuracy/top1": 100.0, "data_time": 0.2904546409845352, "time": 0.8415422290563583, "step": 1} +{"lr": 0.0009939057285945933, "data_time": 0.11489784717559814, "grad_norm": 0.1119145980104804, "loss": 0.6851836323738099, "time": 0.7385829448699951, "epoch": 2, "memory": 28798, "step": 5858} +{"lr": 0.0009939057285945933, "data_time": 0.23536550998687744, "grad_norm": 0.12526991479098798, "loss": 0.6892807424068451, "time": 0.7840444326400757, "epoch": 2, "memory": 28783, "step": 5958} +{"lr": 0.0009939057285945933, "data_time": 0.001851940155029297, "grad_norm": 0.09118716344237328, "loss": 0.6850068032741546, "time": 0.8325890302658081, "epoch": 2, "memory": 28783, "step": 6058} +{"lr": 0.0009939057285945933, "data_time": 0.0017272472381591798, "grad_norm": 0.08466822886839509, "loss": 0.6854560792446136, "time": 0.8943645715713501, "epoch": 2, "memory": 28783, "step": 6158} +{"lr": 0.0009939057285945933, "data_time": 0.0020479917526245116, "grad_norm": 0.09755303636193276, "loss": 0.689743059873581, "time": 0.8275129079818726, "epoch": 2, "memory": 28783, "step": 6258} +{"lr": 0.0009939057285945933, "data_time": 0.002652263641357422, "grad_norm": 0.06423695608973504, "loss": 0.6862825572490692, "time": 0.9148306846618652, "epoch": 2, "memory": 28783, "step": 6358} +{"lr": 0.0009939057285945933, "data_time": 0.0020377397537231444, "grad_norm": 0.09123450815677643, "loss": 0.6879670560359955, "time": 0.8873496294021607, "epoch": 2, "memory": 28783, "step": 6458} +{"lr": 0.0009939057285945933, "data_time": 0.00245974063873291, "grad_norm": 0.1289809938520193, "loss": 0.68823601603508, "time": 0.8818986654281616, "epoch": 2, "memory": 28783, "step": 6558} +{"lr": 0.0009939057285945933, "data_time": 0.032562541961669925, "grad_norm": 0.1374927183613181, "loss": 0.6854689180850982, "time": 0.858594012260437, "epoch": 2, "memory": 28783, "step": 6658} +{"lr": 0.0009939057285945933, "data_time": 0.0013387680053710937, "grad_norm": 0.09858463630080223, "loss": 0.6865781843662262, "time": 0.8951478958129883, "epoch": 2, "memory": 28783, "step": 6758} +{"lr": 0.0009939057285945933, "data_time": 0.0018634557723999023, "grad_norm": 0.1305597260594368, "loss": 0.6850895404815673, "time": 0.7789987325668335, "epoch": 2, "memory": 28783, "step": 6858} +{"lr": 0.0009939057285945933, "data_time": 0.0016655206680297851, "grad_norm": 0.06447077486664057, "loss": 0.68461953997612, "time": 0.9761712312698364, "epoch": 2, "memory": 28783, "step": 6958} +{"lr": 0.0009939057285945933, "data_time": 0.0025378704071044923, "grad_norm": 0.06561579406261445, "loss": 0.6870323240756988, "time": 0.8225229740142822, "epoch": 2, "memory": 28783, "step": 7058} +{"lr": 0.0009939057285945933, "data_time": 0.0016849040985107422, "grad_norm": 0.0958001771941781, "loss": 0.6843083381652832, "time": 0.8733738422393799, "epoch": 2, "memory": 28783, "step": 7158} +{"lr": 0.0009939057285945933, "data_time": 0.0023676395416259766, "grad_norm": 0.13181908074766396, "loss": 0.6872279942035675, "time": 0.8258849382400513, "epoch": 2, "memory": 28783, "step": 7258} +{"lr": 0.0009939057285945933, "data_time": 0.002006077766418457, "grad_norm": 0.0896741697564721, "loss": 0.6851314842700958, "time": 0.8508948564529419, "epoch": 2, "memory": 28783, "step": 7358} +{"lr": 0.0009939057285945933, "data_time": 0.002572154998779297, "grad_norm": 0.09456936344504356, "loss": 0.6865829825401306, "time": 0.8397979736328125, "epoch": 2, "memory": 28783, "step": 7458} +{"lr": 0.0009939057285945933, "data_time": 0.0017290353775024415, "grad_norm": 0.04374804254621267, "loss": 0.684164959192276, "time": 0.8577231884002685, "epoch": 2, "memory": 28783, "step": 7558} +{"lr": 0.0009939057285945933, "data_time": 0.0018528461456298827, "grad_norm": 0.10155281480401754, "loss": 0.688321304321289, "time": 0.9475823163986206, "epoch": 2, "memory": 28783, "step": 7658} +{"lr": 0.0009939057285945933, "data_time": 0.0017137765884399415, "grad_norm": 0.05001990171149373, "loss": 0.6900282025337219, "time": 0.851217246055603, "epoch": 2, "memory": 28783, "step": 7758} +{"lr": 0.0009939057285945933, "data_time": 0.0016271352767944336, "grad_norm": 0.10887768808752299, "loss": 0.6876216173171997, "time": 0.880221962928772, "epoch": 2, "memory": 28783, "step": 7858} +{"lr": 0.0009939057285945933, "data_time": 0.0015771150588989257, "grad_norm": 0.05973792253062129, "loss": 0.6859862327575683, "time": 0.8772577524185181, "epoch": 2, "memory": 28783, "step": 7958} +{"lr": 0.0009939057285945933, "data_time": 0.0016332387924194336, "grad_norm": 0.08887978848069907, "loss": 0.6890406787395478, "time": 0.8825002431869506, "epoch": 2, "memory": 28783, "step": 8058} +{"lr": 0.0009939057285945933, "data_time": 0.0020130634307861327, "grad_norm": 0.061076579988002776, "loss": 0.6835352778434753, "time": 0.8731219530105591, "epoch": 2, "memory": 28783, "step": 8158} +{"lr": 0.0009939057285945933, "data_time": 0.0018208742141723633, "grad_norm": 0.04753756057471037, "loss": 0.6854871273040771, "time": 0.8559173345565796, "epoch": 2, "memory": 28783, "step": 8258} +{"lr": 0.0009939057285945933, "data_time": 0.002297186851501465, "grad_norm": 0.029006470972672106, "loss": 0.6850190162658691, "time": 0.8677525997161866, "epoch": 2, "memory": 28783, "step": 8358} +{"lr": 0.0009939057285945933, "data_time": 0.0016402006149291992, "grad_norm": 0.06308088432997465, "loss": 0.6902522623538971, "time": 0.8611478328704834, "epoch": 2, "memory": 28783, "step": 8458} +{"lr": 0.0009939057285945933, "data_time": 0.0017788410186767578, "grad_norm": 0.05403971169143915, "loss": 0.6853444218635559, "time": 0.8659428119659424, "epoch": 2, "memory": 28783, "step": 8558} +{"lr": 0.0009939057285945933, "data_time": 0.0019292831420898438, "grad_norm": 0.06539376210421324, "loss": 0.6852969110012055, "time": 0.8211614608764648, "epoch": 2, "memory": 28783, "step": 8658} +{"lr": 0.0009939057285945933, "data_time": 0.0033303022384643553, "grad_norm": 0.05729785547591746, "loss": 0.6872235417366028, "time": 0.9224058628082276, "epoch": 2, "memory": 28783, "step": 8758} +{"lr": 0.0009939057285945933, "data_time": 0.0019618511199951173, "grad_norm": 0.07496860194951296, "loss": 0.6859241485595703, "time": 0.9416101694107055, "epoch": 2, "memory": 28783, "step": 8858} +{"lr": 0.0009939057285945933, "data_time": 0.0025458335876464844, "grad_norm": 0.05293703200295567, "loss": 0.6839104056358337, "time": 0.8826780080795288, "epoch": 2, "memory": 28783, "step": 8958} +{"lr": 0.0009939057285945933, "data_time": 0.001965618133544922, "grad_norm": 0.038466890202835204, "loss": 0.688208156824112, "time": 0.8703254699707031, "epoch": 2, "memory": 28783, "step": 9058} +{"lr": 0.0009939057285945933, "data_time": 0.0022713422775268556, "grad_norm": 0.04295786349102855, "loss": 0.6866271257400512, "time": 0.7981902122497558, "epoch": 2, "memory": 28783, "step": 9158} +{"lr": 0.0009939057285945933, "data_time": 0.0027085542678833008, "grad_norm": 0.0508892887737602, "loss": 0.6848026752471924, "time": 0.8314115762710571, "epoch": 2, "memory": 28783, "step": 9258} +{"lr": 0.0009939057285945933, "data_time": 0.0026367902755737305, "grad_norm": 0.055432308837771414, "loss": 0.6862557232379913, "time": 0.8449159145355225, "epoch": 2, "memory": 28783, "step": 9358} +{"lr": 0.0009939057285945933, "data_time": 0.001835036277770996, "grad_norm": 0.03311171652749181, "loss": 0.6850582480430603, "time": 0.8194102764129638, "epoch": 2, "memory": 28783, "step": 9458} +{"lr": 0.0009939057285945933, "data_time": 0.0022808074951171874, "grad_norm": 0.057787072472274305, "loss": 0.6863271236419678, "time": 0.8861421108245849, "epoch": 2, "memory": 28783, "step": 9558} +{"lr": 0.0009939057285945933, "data_time": 0.0018909215927124024, "grad_norm": 0.06150261862203479, "loss": 0.6836685121059418, "time": 0.895897912979126, "epoch": 2, "memory": 28783, "step": 9658} +{"lr": 0.0009939057285945933, "data_time": 0.0027924537658691405, "grad_norm": 0.044922967720776794, "loss": 0.6883501291275025, "time": 0.8687206745147705, "epoch": 2, "memory": 28783, "step": 9758} +{"lr": 0.0009939057285945933, "data_time": 0.002883601188659668, "grad_norm": 0.03776383614167571, "loss": 0.6856978595256805, "time": 0.8509725093841553, "epoch": 2, "memory": 28783, "step": 9858} +{"lr": 0.0009939057285945933, "data_time": 0.0017762184143066406, "grad_norm": 0.05196431670337916, "loss": 0.6858752787113189, "time": 0.8747856378555298, "epoch": 2, "memory": 28783, "step": 9958} +{"lr": 0.0009939057285945933, "data_time": 0.002619647979736328, "grad_norm": 0.038504788372665646, "loss": 0.6874200642108917, "time": 0.9334343671798706, "epoch": 2, "memory": 28783, "step": 10058} +{"lr": 0.0009939057285945933, "data_time": 0.002363920211791992, "grad_norm": 0.03802372813224793, "loss": 0.6889680445194244, "time": 0.8259455680847168, "epoch": 2, "memory": 28783, "step": 10158} +{"lr": 0.0009939057285945933, "data_time": 0.0020737409591674804, "grad_norm": 0.04958138414658606, "loss": 0.6843723356723785, "time": 0.8747926473617553, "epoch": 2, "memory": 28783, "step": 10258} +{"lr": 0.0009939057285945933, "data_time": 0.0017771005630493164, "grad_norm": 0.05166795584373176, "loss": 0.6877075135707855, "time": 0.943445348739624, "epoch": 2, "memory": 28783, "step": 10358} +{"lr": 0.0009939057285945933, "data_time": 0.0017478227615356444, "grad_norm": 0.027391258254647256, "loss": 0.6841959953308105, "time": 0.7990774869918823, "epoch": 2, "memory": 28783, "step": 10458} +{"lr": 0.0009939057285945933, "data_time": 0.002631092071533203, "grad_norm": 0.050929443538188936, "loss": 0.6882081031799316, "time": 0.8281407594680786, "epoch": 2, "memory": 28783, "step": 10558} +{"lr": 0.0009939057285945933, "data_time": 0.0023395538330078123, "grad_norm": 0.035794493090361354, "loss": 0.6882830798625946, "time": 0.8703291177749634, "epoch": 2, "memory": 28783, "step": 10658} +{"lr": 0.0009939057285945933, "data_time": 0.001960015296936035, "grad_norm": 0.03046689215116203, "loss": 0.689048558473587, "time": 0.838711929321289, "epoch": 2, "memory": 28783, "step": 10758} +{"lr": 0.0009939057285945933, "data_time": 0.0016475677490234374, "grad_norm": 0.03824340747669339, "loss": 0.6871283531188965, "time": 0.8687110662460327, "epoch": 2, "memory": 28783, "step": 10858} +{"lr": 0.0009939057285945933, "data_time": 0.0016471624374389648, "grad_norm": 0.0415898721665144, "loss": 0.6892750203609467, "time": 0.8054296731948852, "epoch": 2, "memory": 28783, "step": 10958} +{"lr": 0.0009939057285945933, "data_time": 0.0026760101318359375, "grad_norm": 0.0330808587372303, "loss": 0.6857958555221557, "time": 0.8644672632217407, "epoch": 2, "memory": 28783, "step": 11058} +{"lr": 0.0009939057285945933, "data_time": 0.0016078472137451172, "grad_norm": 0.04337030947208405, "loss": 0.6839715778827667, "time": 0.8387632846832276, "epoch": 2, "memory": 28783, "step": 11158} +{"lr": 0.0009939057285945933, "data_time": 0.0031455039978027345, "grad_norm": 0.05182581262197346, "loss": 0.6873214721679688, "time": 0.8523617506027221, "epoch": 2, "memory": 28783, "step": 11258} +{"lr": 0.0009939057285945933, "data_time": 0.002145862579345703, "grad_norm": 0.03979668482206762, "loss": 0.6871269643306732, "time": 0.8563596963882446, "epoch": 2, "memory": 28783, "step": 11358} +{"lr": 0.0009939057285945933, "data_time": 0.0023670196533203125, "grad_norm": 0.03013358060270548, "loss": 0.6840694546699524, "time": 0.8306557655334472, "epoch": 2, "memory": 28783, "step": 11458} +{"accuracy/top1": 100.0, "data_time": 0.23624329005970673, "time": 0.6698071255403406, "step": 2} +{"lr": 0.0009757729755661012, "data_time": 0.15603256225585938, "grad_norm": 0.0319600532297045, "loss": 0.6842123568058014, "time": 0.8801895380020142, "epoch": 3, "memory": 28783, "step": 11616} +{"lr": 0.0009757729755661012, "data_time": 0.0020602941513061523, "grad_norm": 0.034785063657909635, "loss": 0.6888893246650696, "time": 0.8316410779953003, "epoch": 3, "memory": 28783, "step": 11716} +{"lr": 0.0009757729755661012, "data_time": 0.002340388298034668, "grad_norm": 0.040128201432526114, "loss": 0.6884433448314666, "time": 0.8581866025924683, "epoch": 3, "memory": 28783, "step": 11816} +{"lr": 0.0009757729755661012, "data_time": 0.0014608144760131837, "grad_norm": 0.026125182490795852, "loss": 0.6869147956371308, "time": 0.8026888608932495, "epoch": 3, "memory": 28783, "step": 11916} +{"lr": 0.0009757729755661012, "data_time": 0.0018100738525390625, "grad_norm": 0.03515590270981193, "loss": 0.6843294858932495, "time": 0.8197768449783325, "epoch": 3, "memory": 28783, "step": 12016} +{"lr": 0.0009757729755661012, "data_time": 0.002685070037841797, "grad_norm": 0.03052263017743826, "loss": 0.6916319966316223, "time": 0.831379747390747, "epoch": 3, "memory": 28783, "step": 12116} +{"lr": 0.0009757729755661012, "data_time": 0.0021782875061035155, "grad_norm": 0.03311447571031749, "loss": 0.6845222771167755, "time": 0.8886878490447998, "epoch": 3, "memory": 28783, "step": 12216} +{"lr": 0.0009757729755661012, "data_time": 0.00205686092376709, "grad_norm": 0.03073037527501583, "loss": 0.6866430819034577, "time": 0.8934651613235474, "epoch": 3, "memory": 28783, "step": 12316} +{"lr": 0.0009757729755661012, "data_time": 0.002156639099121094, "grad_norm": 0.02904034472303465, "loss": 0.6870495080947876, "time": 0.8314910888671875, "epoch": 3, "memory": 28783, "step": 12416} +{"lr": 0.0009757729755661012, "data_time": 0.0015803337097167968, "grad_norm": 0.030422977358102798, "loss": 0.6876676201820373, "time": 0.8498664140701294, "epoch": 3, "memory": 28783, "step": 12516} +{"lr": 0.0009757729755661012, "data_time": 0.002436256408691406, "grad_norm": 0.03213986800983548, "loss": 0.6809326410293579, "time": 0.8727119207382202, "epoch": 3, "memory": 28783, "step": 12616} +{"lr": 0.0009757729755661012, "data_time": 0.0026295661926269533, "grad_norm": 0.039869609847664834, "loss": 0.6893934607505798, "time": 0.9144095182418823, "epoch": 3, "memory": 28783, "step": 12716} +{"lr": 0.0009757729755661012, "data_time": 0.0019165277481079102, "grad_norm": 0.030882916337577625, "loss": 0.6870858371257782, "time": 0.8783764600753784, "epoch": 3, "memory": 28783, "step": 12816} +{"lr": 0.0009757729755661012, "data_time": 0.0023956537246704102, "grad_norm": 0.025332201132550837, "loss": 0.6866398930549622, "time": 0.8041995286941528, "epoch": 3, "memory": 28783, "step": 12916} +{"lr": 0.0009757729755661012, "data_time": 0.0019859552383422853, "grad_norm": 0.03307523438706994, "loss": 0.6877510190010071, "time": 0.8169021844863892, "epoch": 3, "memory": 28783, "step": 13016} +{"lr": 0.0009757729755661012, "data_time": 0.0018195152282714845, "grad_norm": 0.01916305311024189, "loss": 0.6880592048168183, "time": 0.8306993246078491, "epoch": 3, "memory": 28783, "step": 13116} +{"lr": 0.0009757729755661012, "data_time": 0.0018338918685913085, "grad_norm": 0.023170626442879438, "loss": 0.6882040917873382, "time": 0.8646248817443848, "epoch": 3, "memory": 28783, "step": 13216} +{"lr": 0.0009757729755661012, "data_time": 0.00217585563659668, "grad_norm": 0.03558538598008454, "loss": 0.6882471084594727, "time": 0.9349054098129272, "epoch": 3, "memory": 28783, "step": 13316} +{"lr": 0.0009757729755661012, "data_time": 0.002730393409729004, "grad_norm": 0.03971612909808755, "loss": 0.6896975994110107, "time": 0.8742197751998901, "epoch": 3, "memory": 28783, "step": 13416} +{"lr": 0.0009757729755661012, "data_time": 0.002499890327453613, "grad_norm": 0.03385852263309062, "loss": 0.688189423084259, "time": 0.9291218757629395, "epoch": 3, "memory": 28783, "step": 13516} +{"lr": 0.0009757729755661012, "data_time": 0.0025034189224243165, "grad_norm": 0.03595380522310734, "loss": 0.6903407573699951, "time": 0.902942419052124, "epoch": 3, "memory": 28783, "step": 13616} +{"lr": 0.0009757729755661012, "data_time": 0.0033739328384399412, "grad_norm": 0.026730207540094853, "loss": 0.6836494386196137, "time": 0.8466433763504029, "epoch": 3, "memory": 28783, "step": 13716} +{"lr": 0.0009757729755661012, "data_time": 0.0024436235427856444, "grad_norm": 0.019502490386366846, "loss": 0.6842268526554107, "time": 0.8439517021179199, "epoch": 3, "memory": 28783, "step": 13816} +{"lr": 0.0009757729755661012, "data_time": 0.0015889883041381836, "grad_norm": 0.04170419373549521, "loss": 0.6888813734054565, "time": 0.8161389112472535, "epoch": 3, "memory": 28783, "step": 13916} +{"lr": 0.0009757729755661012, "data_time": 0.0018450260162353516, "grad_norm": 0.03205676367506385, "loss": 0.688663774728775, "time": 0.9335797071456909, "epoch": 3, "memory": 28783, "step": 14016} +{"lr": 0.0009757729755661012, "data_time": 0.002497696876525879, "grad_norm": 0.02901815567165613, "loss": 0.6877306699752808, "time": 0.890400743484497, "epoch": 3, "memory": 28783, "step": 14116} +{"lr": 0.0009757729755661012, "data_time": 0.0016369342803955079, "grad_norm": 0.028329021483659744, "loss": 0.6899175524711609, "time": 0.8838005781173706, "epoch": 3, "memory": 28783, "step": 14216} +{"lr": 0.0009757729755661012, "data_time": 0.0018515348434448241, "grad_norm": 0.01916082629468292, "loss": 0.6878528296947479, "time": 0.8419302463531494, "epoch": 3, "memory": 28783, "step": 14316} +{"lr": 0.0009757729755661012, "data_time": 0.0017276525497436524, "grad_norm": 0.029594281874597072, "loss": 0.6893251299858093, "time": 0.9079666614532471, "epoch": 3, "memory": 28783, "step": 14416} +{"lr": 0.0009757729755661012, "data_time": 0.0026272296905517577, "grad_norm": 0.029666680260561407, "loss": 0.6882790446281433, "time": 0.8397300958633422, "epoch": 3, "memory": 28783, "step": 14516} +{"lr": 0.0009757729755661012, "data_time": 0.0024530887603759766, "grad_norm": 0.019724297011271118, "loss": 0.6835509836673737, "time": 0.8690378665924072, "epoch": 3, "memory": 28783, "step": 14616} +{"lr": 0.0009757729755661012, "data_time": 0.0023832321166992188, "grad_norm": 0.02569525404833257, "loss": 0.6878702521324158, "time": 0.7856831789016724, "epoch": 3, "memory": 28783, "step": 14716} +{"lr": 0.0009757729755661012, "data_time": 0.0020241498947143554, "grad_norm": 0.03336631013080478, "loss": 0.6883101344108582, "time": 0.8269784688949585, "epoch": 3, "memory": 28783, "step": 14816} +{"lr": 0.0009757729755661012, "data_time": 0.0019346714019775391, "grad_norm": 0.019326245319098233, "loss": 0.6824221432209014, "time": 0.7631311893463135, "epoch": 3, "memory": 28783, "step": 14916} +{"lr": 0.0009757729755661012, "data_time": 0.002571845054626465, "grad_norm": 0.023271534778177737, "loss": 0.6884177565574646, "time": 0.8476747274398804, "epoch": 3, "memory": 28783, "step": 15016} +{"lr": 0.0009757729755661012, "data_time": 0.001752328872680664, "grad_norm": 0.021654575224965812, "loss": 0.6862824082374572, "time": 0.8012657165527344, "epoch": 3, "memory": 28783, "step": 15116} +{"lr": 0.0009757729755661012, "data_time": 0.0024129867553710936, "grad_norm": 0.024710742104798556, "loss": 0.6866450428962707, "time": 0.8590291738510132, "epoch": 3, "memory": 28783, "step": 15216} +{"lr": 0.0009757729755661012, "data_time": 0.0022533655166625975, "grad_norm": 0.01789157767780125, "loss": 0.6886017024517059, "time": 0.7916885375976562, "epoch": 3, "memory": 28783, "step": 15316} +{"lr": 0.0009757729755661012, "data_time": 0.0016936302185058595, "grad_norm": 0.02560994178056717, "loss": 0.6897801280021667, "time": 0.7781678915023804, "epoch": 3, "memory": 28783, "step": 15416} +{"lr": 0.0009757729755661012, "data_time": 0.002540302276611328, "grad_norm": 0.018297767313197256, "loss": 0.6882023394107819, "time": 0.861829137802124, "epoch": 3, "memory": 28783, "step": 15516} +{"lr": 0.0009757729755661012, "data_time": 0.0023749113082885743, "grad_norm": 0.024319959757849573, "loss": 0.6845597207546235, "time": 0.7990557670593261, "epoch": 3, "memory": 28783, "step": 15616} +{"lr": 0.0009757729755661012, "data_time": 0.0024063825607299806, "grad_norm": 0.017872722959145902, "loss": 0.6847500383853913, "time": 0.8461315155029296, "epoch": 3, "memory": 28783, "step": 15716} +{"lr": 0.0009757729755661012, "data_time": 0.0017181873321533204, "grad_norm": 0.02685516527853906, "loss": 0.6859213054180145, "time": 0.8860133409500122, "epoch": 3, "memory": 28783, "step": 15816} +{"lr": 0.0009757729755661012, "data_time": 0.002274155616760254, "grad_norm": 0.020420729462057353, "loss": 0.6882957756519318, "time": 0.7865877628326416, "epoch": 3, "memory": 28783, "step": 15916} +{"lr": 0.0009757729755661012, "data_time": 0.0025748014450073242, "grad_norm": 0.023899454763159154, "loss": 0.689612740278244, "time": 0.8770796537399292, "epoch": 3, "memory": 28783, "step": 16016} +{"lr": 0.0009757729755661012, "data_time": 0.0022995710372924806, "grad_norm": 0.04271694906055927, "loss": 0.6881500124931336, "time": 0.8408160448074341, "epoch": 3, "memory": 28783, "step": 16116} +{"lr": 0.0009757729755661012, "data_time": 0.0029107332229614258, "grad_norm": 0.04083000328391791, "loss": 0.6855602383613586, "time": 0.8630110025405884, "epoch": 3, "memory": 28783, "step": 16216} +{"lr": 0.0009757729755661012, "data_time": 0.002035045623779297, "grad_norm": 0.039553499594330785, "loss": 0.6860554337501525, "time": 0.7795724391937255, "epoch": 3, "memory": 28783, "step": 16316} +{"lr": 0.0009757729755661012, "data_time": 0.0019381046295166016, "grad_norm": 0.03328489847481251, "loss": 0.6891292452812194, "time": 0.8267752647399902, "epoch": 3, "memory": 28783, "step": 16416} +{"lr": 0.0009757729755661012, "data_time": 0.00224452018737793, "grad_norm": 0.027175406087189914, "loss": 0.6853455245494843, "time": 0.7782455444335937, "epoch": 3, "memory": 28783, "step": 16516} +{"lr": 0.0009757729755661012, "data_time": 0.0024422645568847657, "grad_norm": 0.020163445826619865, "loss": 0.6882229030132294, "time": 0.7930678844451904, "epoch": 3, "memory": 28783, "step": 16616} +{"lr": 0.0009757729755661012, "data_time": 0.002415776252746582, "grad_norm": 0.023008845327422024, "loss": 0.6872393250465393, "time": 0.9097555160522461, "epoch": 3, "memory": 28783, "step": 16716} +{"lr": 0.0009757729755661012, "data_time": 0.0026239395141601563, "grad_norm": 0.026061313063837587, "loss": 0.6863723039627075, "time": 0.8081557989120484, "epoch": 3, "memory": 28783, "step": 16816} +{"lr": 0.0009757729755661012, "data_time": 0.002093195915222168, "grad_norm": 0.02261505932547152, "loss": 0.6847315073013306, "time": 0.8320325374603271, "epoch": 3, "memory": 28783, "step": 16916} +{"lr": 0.0009757729755661012, "data_time": 0.002511000633239746, "grad_norm": 0.031159522105008365, "loss": 0.6862626016139984, "time": 0.8609819650650025, "epoch": 3, "memory": 28783, "step": 17016} +{"lr": 0.0009757729755661012, "data_time": 0.0020642518997192384, "grad_norm": 0.03190843695774674, "loss": 0.6833876013755799, "time": 0.8509474515914917, "epoch": 3, "memory": 28783, "step": 17116} +{"lr": 0.0009757729755661012, "data_time": 0.002220606803894043, "grad_norm": 0.03270708271302283, "loss": 0.6867267310619354, "time": 0.7736714601516723, "epoch": 3, "memory": 28783, "step": 17216} +{"accuracy/top1": 100.0, "data_time": 0.24135509659262264, "time": 0.6569869237787583, "step": 3} +{"lr": 0.0009460482294732422, "data_time": 0.27177393436431885, "grad_norm": 0.02520683826878667, "loss": 0.685707426071167, "time": 0.8160369873046875, "epoch": 4, "memory": 28783, "step": 17374} +{"lr": 0.0009460482294732422, "data_time": 0.1556546926498413, "grad_norm": 0.029967418173328042, "loss": 0.6872336387634277, "time": 0.7802465438842774, "epoch": 4, "memory": 28783, "step": 17474} +{"lr": 0.0009460482294732422, "data_time": 0.001715707778930664, "grad_norm": 0.02238327800296247, "loss": 0.68704394698143, "time": 0.7861227750778198, "epoch": 4, "memory": 28783, "step": 17574} +{"lr": 0.0009460482294732422, "data_time": 0.0022268056869506835, "grad_norm": 0.026502556120976805, "loss": 0.686622679233551, "time": 0.7784161806106568, "epoch": 4, "memory": 28783, "step": 17674} +{"lr": 0.0009460482294732422, "data_time": 0.002174186706542969, "grad_norm": 0.015318740205839277, "loss": 0.6857123255729676, "time": 0.6623723030090332, "epoch": 4, "memory": 28783, "step": 17774} +{"lr": 0.0009460482294732422, "data_time": 0.0024941682815551756, "grad_norm": 0.036090648267418146, "loss": 0.689178079366684, "time": 0.7903032779693604, "epoch": 4, "memory": 28783, "step": 17874} +{"lr": 0.0009460482294732422, "data_time": 0.0016437768936157227, "grad_norm": 0.020696333050727843, "loss": 0.6884391725063324, "time": 0.8287275552749633, "epoch": 4, "memory": 28783, "step": 17974} +{"lr": 0.0009460482294732422, "data_time": 0.0015989780426025391, "grad_norm": 0.029684445937164128, "loss": 0.6866159021854401, "time": 0.8033382654190063, "epoch": 4, "memory": 28783, "step": 18074} +{"lr": 0.0009460482294732422, "data_time": 0.0023195505142211913, "grad_norm": 0.016789997764863073, "loss": 0.6897854745388031, "time": 0.7449037551879882, "epoch": 4, "memory": 28783, "step": 18174} +{"lr": 0.0009460482294732422, "data_time": 0.0019205331802368164, "grad_norm": 0.023764927685260773, "loss": 0.6864448547363281, "time": 0.810158085823059, "epoch": 4, "memory": 28783, "step": 18274} +{"lr": 0.0009460482294732422, "data_time": 0.002965569496154785, "grad_norm": 0.027680600015446543, "loss": 0.6844488143920898, "time": 0.8246042251586914, "epoch": 4, "memory": 28783, "step": 18374} +{"lr": 0.0009460482294732422, "data_time": 0.002289414405822754, "grad_norm": 0.01356877707876265, "loss": 0.6841145396232605, "time": 0.8021170616149902, "epoch": 4, "memory": 28783, "step": 18474} +{"lr": 0.0009460482294732422, "data_time": 0.0021143674850463865, "grad_norm": 0.02218390116468072, "loss": 0.6860071241855621, "time": 0.8011329174041748, "epoch": 4, "memory": 28783, "step": 18574} +{"lr": 0.0009460482294732422, "data_time": 0.0019268989562988281, "grad_norm": 0.022346356231719256, "loss": 0.6873934447765351, "time": 0.8456741809844971, "epoch": 4, "memory": 28783, "step": 18674} +{"lr": 0.0009460482294732422, "data_time": 0.0017806529998779298, "grad_norm": 0.019611597340554, "loss": 0.6864718854427337, "time": 0.8185041666030883, "epoch": 4, "memory": 28783, "step": 18774} +{"lr": 0.0009460482294732422, "data_time": 0.0020433664321899414, "grad_norm": 0.016291472362354398, "loss": 0.6877755224704742, "time": 0.8482494592666626, "epoch": 4, "memory": 28783, "step": 18874} +{"lr": 0.0009460482294732422, "data_time": 0.00248415470123291, "grad_norm": 0.015391213377006352, "loss": 0.6861748933792114, "time": 0.7707361459732056, "epoch": 4, "memory": 28783, "step": 18974} +{"lr": 0.0009460482294732422, "data_time": 0.0016952276229858399, "grad_norm": 0.02560833408497274, "loss": 0.6846916735172272, "time": 0.7614190578460693, "epoch": 4, "memory": 28783, "step": 19074} +{"lr": 0.0009460482294732422, "data_time": 0.0017401218414306641, "grad_norm": 0.016099559073336423, "loss": 0.6865309596061706, "time": 0.7657003164291382, "epoch": 4, "memory": 28783, "step": 19174} +{"lr": 0.0009460482294732422, "data_time": 0.002272343635559082, "grad_norm": 0.022295292373746632, "loss": 0.6868140935897827, "time": 0.8766581058502197, "epoch": 4, "memory": 28783, "step": 19274} +{"lr": 0.0009460482294732422, "data_time": 0.002442622184753418, "grad_norm": 0.015944064315408468, "loss": 0.6904959321022034, "time": 0.8379930257797241, "epoch": 4, "memory": 28783, "step": 19374} +{"lr": 0.0009460482294732422, "data_time": 0.0019188165664672852, "grad_norm": 0.022994579700753092, "loss": 0.6865868210792542, "time": 0.9252864837646484, "epoch": 4, "memory": 28783, "step": 19474} +{"lr": 0.0009460482294732422, "data_time": 0.001770782470703125, "grad_norm": 0.02085417225025594, "loss": 0.6846494317054749, "time": 0.905491828918457, "epoch": 4, "memory": 28783, "step": 19574} +{"lr": 0.0009460482294732422, "data_time": 0.0021780729293823242, "grad_norm": 0.018619938916526736, "loss": 0.6884246349334717, "time": 0.7952443122863769, "epoch": 4, "memory": 28783, "step": 19674} +{"lr": 0.0009460482294732422, "data_time": 0.0019256591796875, "grad_norm": 0.017054049647413194, "loss": 0.6887535750865936, "time": 0.8280713081359863, "epoch": 4, "memory": 28783, "step": 19774} +{"lr": 0.0009460482294732422, "data_time": 0.002608919143676758, "grad_norm": 0.017060671676881613, "loss": 0.6889707565307617, "time": 0.908462381362915, "epoch": 4, "memory": 28783, "step": 19874} +{"lr": 0.0009460482294732422, "data_time": 0.0017066478729248046, "grad_norm": 0.022778198728337883, "loss": 0.6879903554916382, "time": 0.844331932067871, "epoch": 4, "memory": 28783, "step": 19974} +{"lr": 0.0009460482294732422, "data_time": 0.001827073097229004, "grad_norm": 0.015492981299757957, "loss": 0.6909884691238404, "time": 0.7941437244415284, "epoch": 4, "memory": 28783, "step": 20074} +{"lr": 0.0009460482294732422, "data_time": 0.0015969276428222656, "grad_norm": 0.011150317033752799, "loss": 0.687701278924942, "time": 0.7402042865753173, "epoch": 4, "memory": 28783, "step": 20174} +{"lr": 0.0009460482294732422, "data_time": 0.0017906665802001954, "grad_norm": 0.025648143514990805, "loss": 0.6873724579811096, "time": 0.7655733823776245, "epoch": 4, "memory": 28783, "step": 20274} +{"lr": 0.0009460482294732422, "data_time": 0.0015273809432983399, "grad_norm": 0.015162501740269363, "loss": 0.6887618899345398, "time": 0.8254800796508789, "epoch": 4, "memory": 28783, "step": 20374} +{"lr": 0.0009460482294732422, "data_time": 0.0017567873001098633, "grad_norm": 0.025566277944017203, "loss": 0.6850540101528168, "time": 0.8347216129302979, "epoch": 4, "memory": 28783, "step": 20474} +{"lr": 0.0009460482294732422, "data_time": 0.0018349409103393555, "grad_norm": 0.01583637185394764, "loss": 0.6897993505001068, "time": 0.8248376846313477, "epoch": 4, "memory": 28783, "step": 20574} +{"lr": 0.0009460482294732422, "data_time": 0.0020338773727416994, "grad_norm": 0.021004949510097504, "loss": 0.6828468859195709, "time": 0.7930845975875854, "epoch": 4, "memory": 28783, "step": 20674} +{"lr": 0.0009460482294732422, "data_time": 0.0020594120025634764, "grad_norm": 0.018910041358321906, "loss": 0.6877440989017487, "time": 0.8832167625427246, "epoch": 4, "memory": 28783, "step": 20774} +{"lr": 0.0009460482294732422, "data_time": 0.001921868324279785, "grad_norm": 0.02118273505475372, "loss": 0.6865809857845306, "time": 0.8105828285217285, "epoch": 4, "memory": 28783, "step": 20874} +{"lr": 0.0009460482294732422, "data_time": 0.001959085464477539, "grad_norm": 0.016271999990567564, "loss": 0.6873459756374359, "time": 0.7846592187881469, "epoch": 4, "memory": 28783, "step": 20974} +{"lr": 0.0009460482294732422, "data_time": 0.0017027616500854491, "grad_norm": 0.020117701822891833, "loss": 0.6867531418800354, "time": 0.8508880853652954, "epoch": 4, "memory": 28783, "step": 21074} +{"lr": 0.0009460482294732422, "data_time": 0.0017918825149536132, "grad_norm": 0.024550669733434914, "loss": 0.687385618686676, "time": 0.8273667812347412, "epoch": 4, "memory": 28783, "step": 21174} +{"lr": 0.0009460482294732422, "data_time": 0.0019077301025390626, "grad_norm": 0.01758038029074669, "loss": 0.6860480785369873, "time": 0.7815929412841797, "epoch": 4, "memory": 28783, "step": 21274} +{"lr": 0.0009460482294732422, "data_time": 0.0018862485885620117, "grad_norm": 0.015043219458311797, "loss": 0.6862777590751648, "time": 0.8440651893615723, "epoch": 4, "memory": 28783, "step": 21374} +{"lr": 0.0009460482294732422, "data_time": 0.0017036199569702148, "grad_norm": 0.016285820165649056, "loss": 0.689351886510849, "time": 0.7999744892120362, "epoch": 4, "memory": 28783, "step": 21474} +{"lr": 0.0009460482294732422, "data_time": 0.001898050308227539, "grad_norm": 0.01762022227048874, "loss": 0.6893795967102051, "time": 0.8926919221878051, "epoch": 4, "memory": 28783, "step": 21574} +{"lr": 0.0009460482294732422, "data_time": 0.052608704566955565, "grad_norm": 0.023961476050317287, "loss": 0.6867991864681244, "time": 0.8131070137023926, "epoch": 4, "memory": 28783, "step": 21674} +{"lr": 0.0009460482294732422, "data_time": 0.13437292575836182, "grad_norm": 0.008743662922643124, "loss": 0.6895983636379241, "time": 0.7482769966125489, "epoch": 4, "memory": 28783, "step": 21774} +{"lr": 0.0009460482294732422, "data_time": 0.36082210540771487, "grad_norm": 0.019423121633008123, "loss": 0.6859278976917267, "time": 0.8646171569824219, "epoch": 4, "memory": 28783, "step": 21874} +{"lr": 0.0009460482294732422, "data_time": 0.28841822147369384, "grad_norm": 0.01611707587726414, "loss": 0.6821094393730164, "time": 0.804296898841858, "epoch": 4, "memory": 28783, "step": 21974} +{"lr": 0.0009460482294732422, "data_time": 0.3223053216934204, "grad_norm": 0.015545494575053453, "loss": 0.6854293584823609, "time": 0.8259261131286622, "epoch": 4, "memory": 28783, "step": 22074} +{"lr": 0.0009460482294732422, "data_time": 0.320308518409729, "grad_norm": 0.016658765822649003, "loss": 0.6850722014904023, "time": 0.8289936780929565, "epoch": 4, "memory": 28783, "step": 22174} +{"lr": 0.0009460482294732422, "data_time": 0.3064169645309448, "grad_norm": 0.014246461691800504, "loss": 0.6861003637313843, "time": 0.8129561901092529, "epoch": 4, "memory": 28783, "step": 22274} +{"lr": 0.0009460482294732422, "data_time": 0.3194380044937134, "grad_norm": 0.0113477194448933, "loss": 0.6856185317039489, "time": 0.8231329202651978, "epoch": 4, "memory": 28783, "step": 22374} +{"lr": 0.0009460482294732422, "data_time": 0.3566581249237061, "grad_norm": 0.02150777440983802, "loss": 0.6859536170959473, "time": 0.861137080192566, "epoch": 4, "memory": 28783, "step": 22474} +{"lr": 0.0009460482294732422, "data_time": 0.30666141510009765, "grad_norm": 0.02201671889051795, "loss": 0.6879827201366424, "time": 0.8114376544952393, "epoch": 4, "memory": 28783, "step": 22574} +{"lr": 0.0009460482294732422, "data_time": 0.27706668376922605, "grad_norm": 0.01324184676632285, "loss": 0.686724579334259, "time": 0.7841790437698364, "epoch": 4, "memory": 28783, "step": 22674} +{"lr": 0.0009460482294732422, "data_time": 0.15586607456207274, "grad_norm": 0.019132132339291275, "loss": 0.6883108198642731, "time": 0.7770811319351196, "epoch": 4, "memory": 28783, "step": 22774} +{"lr": 0.0009460482294732422, "data_time": 0.0016731500625610351, "grad_norm": 0.015146529953926801, "loss": 0.6915353000164032, "time": 0.802668285369873, "epoch": 4, "memory": 28783, "step": 22874} +{"lr": 0.0009460482294732422, "data_time": 0.001768207550048828, "grad_norm": 0.02003692085854709, "loss": 0.6877279937267303, "time": 0.7683682203292846, "epoch": 4, "memory": 28783, "step": 22974} +{"accuracy/top1": 100.0, "data_time": 0.21070276989656336, "time": 0.6255511816810159, "step": 4} +{"lr": 0.0009054634122155991, "data_time": 0.0023488521575927733, "grad_norm": 0.01978237871080637, "loss": 0.6857580482959748, "time": 0.7440827131271363, "epoch": 5, "memory": 28783, "step": 23132} +{"lr": 0.0009054634122155991, "data_time": 0.001589345932006836, "grad_norm": 0.01621111078420654, "loss": 0.6885083854198456, "time": 0.8214529275894165, "epoch": 5, "memory": 28783, "step": 23232} +{"lr": 0.0009054634122155991, "data_time": 0.0017946004867553712, "grad_norm": 0.015389048820361495, "loss": 0.6877581059932709, "time": 0.7936827182769776, "epoch": 5, "memory": 28783, "step": 23332} +{"lr": 0.0009054634122155991, "data_time": 0.0019068956375122071, "grad_norm": 0.01602655553724617, "loss": 0.6847742557525635, "time": 0.7926679611206054, "epoch": 5, "memory": 28783, "step": 23432} +{"lr": 0.0009054634122155991, "data_time": 0.0016833066940307618, "grad_norm": 0.01653628007043153, "loss": 0.6900790333747864, "time": 0.8546588659286499, "epoch": 5, "memory": 28783, "step": 23532} +{"lr": 0.0009054634122155991, "data_time": 0.0014732837677001952, "grad_norm": 0.019846688024699688, "loss": 0.6886083960533143, "time": 0.8249326944351196, "epoch": 5, "memory": 28783, "step": 23632} +{"lr": 0.0009054634122155991, "data_time": 0.002274346351623535, "grad_norm": 0.019802834442816674, "loss": 0.6862328767776489, "time": 0.7773101329803467, "epoch": 5, "memory": 28783, "step": 23732} +{"lr": 0.0009054634122155991, "data_time": 0.001839900016784668, "grad_norm": 0.018447005655616522, "loss": 0.6862954020500183, "time": 0.8398216247558594, "epoch": 5, "memory": 28783, "step": 23832} +{"lr": 0.0009054634122155991, "data_time": 0.001974010467529297, "grad_norm": 0.024368872703053056, "loss": 0.6901835262775421, "time": 0.8179062843322754, "epoch": 5, "memory": 28783, "step": 23932} +{"lr": 0.0009054634122155991, "data_time": 0.0016718864440917968, "grad_norm": 0.019588706316426396, "loss": 0.6887371957302093, "time": 0.761960792541504, "epoch": 5, "memory": 28783, "step": 24032} +{"lr": 0.0009054634122155991, "data_time": 0.002263045310974121, "grad_norm": 0.01628736048005521, "loss": 0.6862403869628906, "time": 0.784402585029602, "epoch": 5, "memory": 28783, "step": 24132} +{"lr": 0.0009054634122155991, "data_time": 0.002751803398132324, "grad_norm": 0.013741782843135297, "loss": 0.6810522973537445, "time": 0.8334048748016357, "epoch": 5, "memory": 28783, "step": 24232} +{"lr": 0.0009054634122155991, "data_time": 0.002513623237609863, "grad_norm": 0.021936559700407088, "loss": 0.6858477830886841, "time": 0.8193792343139649, "epoch": 5, "memory": 28783, "step": 24332} +{"lr": 0.0009054634122155991, "data_time": 0.0017559528350830078, "grad_norm": 0.01746827968163416, "loss": 0.6847206890583039, "time": 0.7778014898300171, "epoch": 5, "memory": 28783, "step": 24432} +{"lr": 0.0009054634122155991, "data_time": 0.0017470121383666992, "grad_norm": 0.01922081718221307, "loss": 0.6864348530769349, "time": 0.8111227989196778, "epoch": 5, "memory": 28783, "step": 24532} +{"lr": 0.0009054634122155991, "data_time": 0.0018786907196044922, "grad_norm": 0.01580275730229914, "loss": 0.6858112275600433, "time": 0.8756220579147339, "epoch": 5, "memory": 28783, "step": 24632} +{"lr": 0.0009054634122155991, "data_time": 0.0016021728515625, "grad_norm": 0.033075568196363744, "loss": 0.6867753863334656, "time": 0.7886070251464844, "epoch": 5, "memory": 28783, "step": 24732} +{"lr": 0.0009054634122155991, "data_time": 0.0015631914138793945, "grad_norm": 0.013563393824733793, "loss": 0.6867523193359375, "time": 0.7609542608261108, "epoch": 5, "memory": 28783, "step": 24832} +{"lr": 0.0009054634122155991, "data_time": 0.0024301290512084963, "grad_norm": 0.015505463723093272, "loss": 0.6852036118507385, "time": 0.8193845272064209, "epoch": 5, "memory": 28783, "step": 24932} +{"lr": 0.0009054634122155991, "data_time": 0.002778482437133789, "grad_norm": 0.014682882349006832, "loss": 0.6829340577125549, "time": 0.8177146434783935, "epoch": 5, "memory": 28783, "step": 25032} +{"lr": 0.0009054634122155991, "data_time": 0.002022838592529297, "grad_norm": 0.01816959111019969, "loss": 0.6864844381809234, "time": 0.803546142578125, "epoch": 5, "memory": 28783, "step": 25132} +{"lr": 0.0009054634122155991, "data_time": 0.002598214149475098, "grad_norm": 0.011350920656695962, "loss": 0.6844705045223236, "time": 0.814617919921875, "epoch": 5, "memory": 28783, "step": 25232} +{"lr": 0.0009054634122155991, "data_time": 0.0023195743560791016, "grad_norm": 0.01194750884314999, "loss": 0.6867002427577973, "time": 0.787207818031311, "epoch": 5, "memory": 28783, "step": 25332} +{"lr": 0.0009054634122155991, "data_time": 0.0017548561096191405, "grad_norm": 0.016847985982894897, "loss": 0.6891045987606048, "time": 0.7996291160583496, "epoch": 5, "memory": 28783, "step": 25432} +{"lr": 0.0009054634122155991, "data_time": 0.001551198959350586, "grad_norm": 0.012166864122264087, "loss": 0.686947351694107, "time": 0.8243926048278809, "epoch": 5, "memory": 28783, "step": 25532} +{"lr": 0.0009054634122155991, "data_time": 0.0020097970962524416, "grad_norm": 0.018306543538346886, "loss": 0.6858312368392945, "time": 0.8164604902267456, "epoch": 5, "memory": 28783, "step": 25632} +{"lr": 0.0009054634122155991, "data_time": 0.0025929689407348635, "grad_norm": 0.015245183336082846, "loss": 0.6887175500392914, "time": 0.7557824373245239, "epoch": 5, "memory": 28783, "step": 25732} +{"lr": 0.0009054634122155991, "data_time": 0.002368521690368652, "grad_norm": 0.025837548170238735, "loss": 0.6867116332054138, "time": 0.7816663026809693, "epoch": 5, "memory": 28783, "step": 25832} +{"lr": 0.0009054634122155991, "data_time": 0.0017647266387939454, "grad_norm": 0.01864590812474489, "loss": 0.6891556560993195, "time": 0.8854470729827881, "epoch": 5, "memory": 28783, "step": 25932} +{"lr": 0.0009054634122155991, "data_time": 0.0017224788665771485, "grad_norm": 0.016243601916357875, "loss": 0.6866788446903229, "time": 0.8518370628356934, "epoch": 5, "memory": 28783, "step": 26032} +{"lr": 0.0009054634122155991, "data_time": 0.002374696731567383, "grad_norm": 0.017827885830774902, "loss": 0.6882756471633911, "time": 0.7822090148925781, "epoch": 5, "memory": 28783, "step": 26132} +{"lr": 0.0009054634122155991, "data_time": 0.002496147155761719, "grad_norm": 0.017350861546583475, "loss": 0.6872244894504547, "time": 0.7932394504547119, "epoch": 5, "memory": 28783, "step": 26232} +{"lr": 0.0009054634122155991, "data_time": 0.0031243562698364258, "grad_norm": 0.021743663883535193, "loss": 0.6892023324966431, "time": 0.8034002780914307, "epoch": 5, "memory": 28783, "step": 26332} +{"lr": 0.0009054634122155991, "data_time": 0.0024820804595947266, "grad_norm": 0.015899774129502477, "loss": 0.6843734443187713, "time": 0.8362406015396118, "epoch": 5, "memory": 28783, "step": 26432} +{"lr": 0.0009054634122155991, "data_time": 0.0019497394561767579, "grad_norm": 0.0183228648500517, "loss": 0.6847833752632141, "time": 0.8367327690124512, "epoch": 5, "memory": 28783, "step": 26532} +{"lr": 0.0009054634122155991, "data_time": 0.0018743276596069336, "grad_norm": 0.017327110562473536, "loss": 0.6856024205684662, "time": 0.7896200180053711, "epoch": 5, "memory": 28783, "step": 26632} +{"lr": 0.0009054634122155991, "data_time": 0.0017811059951782227, "grad_norm": 0.028434014599770307, "loss": 0.6843396186828613, "time": 0.8050744533538818, "epoch": 5, "memory": 28783, "step": 26732} +{"lr": 0.0009054634122155991, "data_time": 0.002515101432800293, "grad_norm": 0.012014444149099291, "loss": 0.6864933371543884, "time": 0.8401695251464844, "epoch": 5, "memory": 28783, "step": 26832} +{"lr": 0.0009054634122155991, "data_time": 0.0017719030380249023, "grad_norm": 0.021527543873526157, "loss": 0.6900273263454437, "time": 0.8216508150100708, "epoch": 5, "memory": 28783, "step": 26932} +{"lr": 0.0009054634122155991, "data_time": 0.0018821954727172852, "grad_norm": 0.019868181413039566, "loss": 0.6904372155666352, "time": 0.7892855644226074, "epoch": 5, "memory": 28783, "step": 27032} +{"lr": 0.0009054634122155991, "data_time": 0.002454662322998047, "grad_norm": 0.021997840027324855, "loss": 0.6862986445426941, "time": 0.7666745901107788, "epoch": 5, "memory": 28783, "step": 27132} +{"lr": 0.0009054634122155991, "data_time": 0.0017728567123413085, "grad_norm": 0.009503014292567969, "loss": 0.6874842941761017, "time": 0.8114336490631103, "epoch": 5, "memory": 28783, "step": 27232} +{"lr": 0.0009054634122155991, "data_time": 0.0021835088729858397, "grad_norm": 0.018126374180428683, "loss": 0.68863565325737, "time": 0.7756914854049682, "epoch": 5, "memory": 28783, "step": 27332} +{"lr": 0.0009054634122155991, "data_time": 0.0018012046813964844, "grad_norm": 0.014824231527745724, "loss": 0.6837986826896667, "time": 0.8376925468444825, "epoch": 5, "memory": 28783, "step": 27432} +{"lr": 0.0009054634122155991, "data_time": 0.001721501350402832, "grad_norm": 0.01978010511957109, "loss": 0.6871576905250549, "time": 0.76935293674469, "epoch": 5, "memory": 28783, "step": 27532} +{"lr": 0.0009054634122155991, "data_time": 0.0016092777252197266, "grad_norm": 0.013995923660695552, "loss": 0.6886760652065277, "time": 0.7473759174346923, "epoch": 5, "memory": 28783, "step": 27632} +{"lr": 0.0009054634122155991, "data_time": 0.001996016502380371, "grad_norm": 0.014643355918815359, "loss": 0.6834181725978852, "time": 0.7522551774978637, "epoch": 5, "memory": 28783, "step": 27732} +{"lr": 0.0009054634122155991, "data_time": 0.001869058609008789, "grad_norm": 0.018232701462693512, "loss": 0.6909704387187958, "time": 0.7726557493209839, "epoch": 5, "memory": 28783, "step": 27832} +{"lr": 0.0009054634122155991, "data_time": 0.0016345500946044922, "grad_norm": 0.01877805000403896, "loss": 0.6915616273880005, "time": 0.7824899196624756, "epoch": 5, "memory": 28783, "step": 27932} +{"lr": 0.0009054634122155991, "data_time": 0.002409958839416504, "grad_norm": 0.01969190021045506, "loss": 0.6833604395389556, "time": 0.7599211931228638, "epoch": 5, "memory": 28783, "step": 28032} +{"lr": 0.0009054634122155991, "data_time": 0.0018677473068237304, "grad_norm": 0.017953099100850523, "loss": 0.6863316774368287, "time": 0.8113799810409545, "epoch": 5, "memory": 28783, "step": 28132} +{"lr": 0.0009054634122155991, "data_time": 0.0017926931381225587, "grad_norm": 0.01289085887838155, "loss": 0.686074960231781, "time": 0.8070666790008545, "epoch": 5, "memory": 28783, "step": 28232} +{"lr": 0.0009054634122155991, "data_time": 0.0017676353454589844, "grad_norm": 0.013170716888271271, "loss": 0.6844303905963898, "time": 0.8298574924468994, "epoch": 5, "memory": 28783, "step": 28332} +{"lr": 0.0009054634122155991, "data_time": 0.0016425848007202148, "grad_norm": 0.012275812891311943, "loss": 0.6874430894851684, "time": 0.8194347620010376, "epoch": 5, "memory": 28783, "step": 28432} +{"lr": 0.0009054634122155991, "data_time": 0.0018143415451049804, "grad_norm": 0.011029295483604073, "loss": 0.6893566370010376, "time": 0.8269855737686157, "epoch": 5, "memory": 28783, "step": 28532} +{"lr": 0.0009054634122155991, "data_time": 0.0019134521484375, "grad_norm": 0.014390631462447345, "loss": 0.6906480729579926, "time": 0.8102564334869384, "epoch": 5, "memory": 28783, "step": 28632} +{"lr": 0.0009054634122155991, "data_time": 0.001816868782043457, "grad_norm": 0.011227173113729805, "loss": 0.68746337890625, "time": 0.8873985528945922, "epoch": 5, "memory": 28783, "step": 28732} +{"accuracy/top1": 100.0, "data_time": 0.2130567747003892, "time": 0.6275536032284007, "step": 5} +{"lr": 0.0008550178566873411, "data_time": 0.03847064971923828, "grad_norm": 0.016264502005651594, "loss": 0.6850412607192993, "time": 0.7648952722549438, "epoch": 6, "memory": 28783, "step": 28890} +{"lr": 0.0008550178566873411, "data_time": 0.0017691612243652343, "grad_norm": 0.01797385523095727, "loss": 0.6861240804195404, "time": 0.8349392652511597, "epoch": 6, "memory": 28783, "step": 28990} +{"lr": 0.0008550178566873411, "data_time": 0.0018528938293457032, "grad_norm": 0.02012402010150254, "loss": 0.6858736753463746, "time": 0.801614236831665, "epoch": 6, "memory": 28783, "step": 29090} +{"lr": 0.0008550178566873411, "data_time": 0.0017559289932250976, "grad_norm": 0.02366946768015623, "loss": 0.689288717508316, "time": 0.8147270917892456, "epoch": 6, "memory": 28783, "step": 29190} +{"lr": 0.0008550178566873411, "data_time": 0.0017263174057006836, "grad_norm": 0.01673456421121955, "loss": 0.6863142311573028, "time": 0.7748468160629273, "epoch": 6, "memory": 28783, "step": 29290} +{"lr": 0.0008550178566873411, "data_time": 0.0016908884048461915, "grad_norm": 0.016275232611224055, "loss": 0.6843490362167358, "time": 0.8420747280120849, "epoch": 6, "memory": 28783, "step": 29390} +{"lr": 0.0008550178566873411, "data_time": 0.002105450630187988, "grad_norm": 0.009479525731876493, "loss": 0.6898092031478882, "time": 0.7396365404129028, "epoch": 6, "memory": 28783, "step": 29490} +{"lr": 0.0008550178566873411, "data_time": 0.0017427921295166016, "grad_norm": 0.02365730090532452, "loss": 0.6884424567222596, "time": 0.7768778085708619, "epoch": 6, "memory": 28783, "step": 29590} +{"lr": 0.0008550178566873411, "data_time": 0.001991462707519531, "grad_norm": 0.024823940126225352, "loss": 0.684698885679245, "time": 0.772196102142334, "epoch": 6, "memory": 28783, "step": 29690} +{"lr": 0.0008550178566873411, "data_time": 0.0015755414962768555, "grad_norm": 0.020247644768096505, "loss": 0.6880524396896363, "time": 0.8122198820114136, "epoch": 6, "memory": 28783, "step": 29790} +{"lr": 0.0008550178566873411, "data_time": 0.0019313573837280273, "grad_norm": 0.01491468558087945, "loss": 0.6881711661815644, "time": 0.7423748016357422, "epoch": 6, "memory": 28783, "step": 29890} +{"lr": 0.0008550178566873411, "data_time": 0.0019288063049316406, "grad_norm": 0.016054106689989567, "loss": 0.6884989082813263, "time": 0.8245399475097657, "epoch": 6, "memory": 28783, "step": 29990} +{"lr": 0.0008550178566873411, "data_time": 0.0018289327621459962, "grad_norm": 0.01718117855489254, "loss": 0.688566142320633, "time": 0.8262383460998535, "epoch": 6, "memory": 28783, "step": 30090} +{"lr": 0.0008550178566873411, "data_time": 0.0023021459579467773, "grad_norm": 0.013830536813475191, "loss": 0.6884703040122986, "time": 0.825492000579834, "epoch": 6, "memory": 28783, "step": 30190} +{"lr": 0.0008550178566873411, "data_time": 0.0021821022033691405, "grad_norm": 0.016090997448191046, "loss": 0.6848032534122467, "time": 0.8229004383087158, "epoch": 6, "memory": 28783, "step": 30290} +{"lr": 0.0008550178566873411, "data_time": 0.0018199920654296876, "grad_norm": 0.012432183441706001, "loss": 0.6858310401439667, "time": 0.832021975517273, "epoch": 6, "memory": 28783, "step": 30390} +{"lr": 0.0008550178566873411, "data_time": 0.0018375396728515625, "grad_norm": 0.019704459933564067, "loss": 0.6861458837985992, "time": 0.8024144172668457, "epoch": 6, "memory": 28783, "step": 30490} +{"lr": 0.0008550178566873411, "data_time": 0.0015511751174926759, "grad_norm": 0.020204418664798142, "loss": 0.6902287602424622, "time": 0.7734831809997559, "epoch": 6, "memory": 28783, "step": 30590} +{"lr": 0.0008550178566873411, "data_time": 0.0020596981048583984, "grad_norm": 0.013450464070774614, "loss": 0.686540299654007, "time": 0.8267415761947632, "epoch": 6, "memory": 28783, "step": 30690} +{"lr": 0.0008550178566873411, "data_time": 0.0020363569259643556, "grad_norm": 0.014911745325662196, "loss": 0.6875357866287232, "time": 0.7771576404571533, "epoch": 6, "memory": 28783, "step": 30790} +{"lr": 0.0008550178566873411, "data_time": 0.0022123098373413087, "grad_norm": 0.015077015268616378, "loss": 0.6888209342956543, "time": 0.8421217918395996, "epoch": 6, "memory": 28783, "step": 30890} +{"lr": 0.0008550178566873411, "data_time": 0.0016525745391845702, "grad_norm": 0.014739341498352588, "loss": 0.6869848310947418, "time": 0.7946889162063598, "epoch": 6, "memory": 28783, "step": 30990} +{"lr": 0.0008550178566873411, "data_time": 0.0016237974166870117, "grad_norm": 0.020212713163346052, "loss": 0.6863488256931305, "time": 0.7981140613555908, "epoch": 6, "memory": 28783, "step": 31090} +{"lr": 0.0008550178566873411, "data_time": 0.001683497428894043, "grad_norm": 0.011441466386895627, "loss": 0.688124930858612, "time": 0.871791934967041, "epoch": 6, "memory": 28783, "step": 31190} +{"lr": 0.0008550178566873411, "data_time": 0.0025690317153930662, "grad_norm": 0.023499373346567154, "loss": 0.6888080537319183, "time": 0.8048616886138916, "epoch": 6, "memory": 28783, "step": 31290} +{"lr": 0.0008550178566873411, "data_time": 0.0016889095306396485, "grad_norm": 0.018465373641811313, "loss": 0.6895635426044464, "time": 0.8178466081619262, "epoch": 6, "memory": 28783, "step": 31390} +{"lr": 0.0008550178566873411, "data_time": 0.002007722854614258, "grad_norm": 0.015655048273038118, "loss": 0.6862462401390076, "time": 0.7765378952026367, "epoch": 6, "memory": 28783, "step": 31490} +{"lr": 0.0008550178566873411, "data_time": 0.002905130386352539, "grad_norm": 0.015905791777186096, "loss": 0.6853396892547607, "time": 0.8254060983657837, "epoch": 6, "memory": 28783, "step": 31590} +{"lr": 0.0008550178566873411, "data_time": 0.0020444393157958984, "grad_norm": 0.015115947742015124, "loss": 0.6864209949970246, "time": 0.8081553936004638, "epoch": 6, "memory": 28783, "step": 31690} +{"lr": 0.0008550178566873411, "data_time": 0.0019181013107299806, "grad_norm": 0.02002663635648787, "loss": 0.6888629078865052, "time": 0.7970602512359619, "epoch": 6, "memory": 28783, "step": 31790} +{"lr": 0.0008550178566873411, "data_time": 0.0021001338958740235, "grad_norm": 0.009449188085272908, "loss": 0.6891397953033447, "time": 0.7879553079605103, "epoch": 6, "memory": 28783, "step": 31890} +{"lr": 0.0008550178566873411, "data_time": 0.0017438411712646484, "grad_norm": 0.018369228683877737, "loss": 0.6870072305202484, "time": 0.8410459041595459, "epoch": 6, "memory": 28783, "step": 31990} +{"lr": 0.0008550178566873411, "data_time": 0.0019531011581420898, "grad_norm": 0.015933985286392272, "loss": 0.6843014359474182, "time": 0.813026237487793, "epoch": 6, "memory": 28783, "step": 32090} +{"lr": 0.0008550178566873411, "data_time": 0.001747274398803711, "grad_norm": 0.0175642151851207, "loss": 0.6857497990131378, "time": 0.7733380794525146, "epoch": 6, "memory": 28783, "step": 32190} +{"lr": 0.0008550178566873411, "data_time": 0.002689051628112793, "grad_norm": 0.02285115484846756, "loss": 0.6887994766235351, "time": 0.8348977088928222, "epoch": 6, "memory": 28783, "step": 32290} +{"lr": 0.0008550178566873411, "data_time": 0.001812124252319336, "grad_norm": 0.012550061126239598, "loss": 0.6830160737037658, "time": 0.8194183111190796, "epoch": 6, "memory": 28783, "step": 32390} +{"lr": 0.0008550178566873411, "data_time": 0.0022510528564453126, "grad_norm": 0.015787654975429177, "loss": 0.6845639824867249, "time": 0.7714890241622925, "epoch": 6, "memory": 28783, "step": 32490} +{"lr": 0.0008550178566873411, "data_time": 0.00185089111328125, "grad_norm": 0.016110290260985493, "loss": 0.6892039895057678, "time": 0.8338452577590942, "epoch": 6, "memory": 28783, "step": 32590} +{"lr": 0.0008550178566873411, "data_time": 0.001908588409423828, "grad_norm": 0.021656455099582674, "loss": 0.6890843749046326, "time": 0.8557557106018067, "epoch": 6, "memory": 28783, "step": 32690} +{"lr": 0.0008550178566873411, "data_time": 0.0017270803451538085, "grad_norm": 0.01627308176830411, "loss": 0.6873088657855988, "time": 0.7854939699172974, "epoch": 6, "memory": 28783, "step": 32790} +{"lr": 0.0008550178566873411, "data_time": 0.0027657270431518553, "grad_norm": 0.01565114166587591, "loss": 0.6905042350292205, "time": 0.8177520751953125, "epoch": 6, "memory": 28783, "step": 32890} +{"lr": 0.0008550178566873411, "data_time": 0.0022096872329711915, "grad_norm": 0.014277510601095856, "loss": 0.6869805037975312, "time": 0.7925125360488892, "epoch": 6, "memory": 28783, "step": 32990} +{"lr": 0.0008550178566873411, "data_time": 0.0018399953842163086, "grad_norm": 0.013290160486940294, "loss": 0.6889128386974335, "time": 0.8921339750289917, "epoch": 6, "memory": 28783, "step": 33090} +{"lr": 0.0008550178566873411, "data_time": 0.0022870302200317383, "grad_norm": 0.018545915349386634, "loss": 0.6841189384460449, "time": 0.8625333070755005, "epoch": 6, "memory": 28783, "step": 33190} +{"lr": 0.0008550178566873411, "data_time": 0.0018452167510986327, "grad_norm": 0.01571862050332129, "loss": 0.6855437934398652, "time": 0.7996485710144043, "epoch": 6, "memory": 28783, "step": 33290} +{"lr": 0.0008550178566873411, "data_time": 0.0023263216018676756, "grad_norm": 0.01806652401573956, "loss": 0.6881719887256622, "time": 0.8174720764160156, "epoch": 6, "memory": 28783, "step": 33390} +{"lr": 0.0008550178566873411, "data_time": 0.0018170595169067383, "grad_norm": 0.013850232562981546, "loss": 0.6849901378154755, "time": 0.801367974281311, "epoch": 6, "memory": 28783, "step": 33490} +{"lr": 0.0008550178566873411, "data_time": 0.0018849611282348634, "grad_norm": 0.014545754552818835, "loss": 0.6857048273086548, "time": 0.7559276342391967, "epoch": 6, "memory": 28783, "step": 33590} +{"lr": 0.0008550178566873411, "data_time": 0.002006220817565918, "grad_norm": 0.03836859054863453, "loss": 0.6882247507572175, "time": 0.8427144765853882, "epoch": 6, "memory": 28783, "step": 33690} +{"lr": 0.0008550178566873411, "data_time": 0.00178072452545166, "grad_norm": 0.03336802404373884, "loss": 0.6900188505649567, "time": 0.808061671257019, "epoch": 6, "memory": 28783, "step": 33790} +{"lr": 0.0008550178566873411, "data_time": 0.0018287420272827149, "grad_norm": 0.03354310598224401, "loss": 0.6870254516601563, "time": 0.8713032722473144, "epoch": 6, "memory": 28783, "step": 33890} +{"lr": 0.0008550178566873411, "data_time": 0.0020443201065063477, "grad_norm": 0.02385171614587307, "loss": 0.6834163248538971, "time": 0.8586906433105469, "epoch": 6, "memory": 28783, "step": 33990} +{"lr": 0.0008550178566873411, "data_time": 0.0015150070190429687, "grad_norm": 0.023183808848261832, "loss": 0.6892798781394959, "time": 0.765532660484314, "epoch": 6, "memory": 28783, "step": 34090} +{"lr": 0.0008550178566873411, "data_time": 0.001829814910888672, "grad_norm": 0.023499545454978944, "loss": 0.6863574862480164, "time": 0.8652956008911132, "epoch": 6, "memory": 28783, "step": 34190} +{"lr": 0.0008550178566873411, "data_time": 0.002898502349853516, "grad_norm": 0.019519153516739607, "loss": 0.683392482995987, "time": 0.8010478019714355, "epoch": 6, "memory": 28783, "step": 34290} +{"lr": 0.0008550178566873411, "data_time": 0.002373623847961426, "grad_norm": 0.01928608426824212, "loss": 0.6857008755207061, "time": 0.8344033718109131, "epoch": 6, "memory": 28783, "step": 34390} +{"lr": 0.0008550178566873411, "data_time": 0.002845001220703125, "grad_norm": 0.018983273347839714, "loss": 0.6903075754642487, "time": 0.8585350751876831, "epoch": 6, "memory": 28783, "step": 34490} +{"accuracy/top1": 100.0, "data_time": 0.21281940796796014, "time": 0.6271857794593362, "step": 6} +{"lr": 0.0007959536998847743, "data_time": 0.00540931224822998, "grad_norm": 0.01697277408093214, "loss": 0.6851157963275909, "time": 0.7517443418502807, "epoch": 7, "memory": 28783, "step": 34648} +{"lr": 0.0007959536998847743, "data_time": 0.0018120288848876953, "grad_norm": 0.02122509405016899, "loss": 0.6902824699878692, "time": 0.8486426115036011, "epoch": 7, "memory": 28783, "step": 34748} +{"lr": 0.0007959536998847743, "data_time": 0.0016950845718383789, "grad_norm": 0.01617599939927459, "loss": 0.6866198480129242, "time": 0.8561187982559204, "epoch": 7, "memory": 28783, "step": 34848} +{"lr": 0.0007959536998847743, "data_time": 0.002018284797668457, "grad_norm": 0.02124827723018825, "loss": 0.6836631298065186, "time": 0.8276988983154296, "epoch": 7, "memory": 28783, "step": 34948} +{"lr": 0.0007959536998847743, "data_time": 0.001775217056274414, "grad_norm": 0.023269442655146123, "loss": 0.6870947539806366, "time": 0.8401309490203858, "epoch": 7, "memory": 28783, "step": 35048} +{"lr": 0.0007959536998847743, "data_time": 0.0015749931335449219, "grad_norm": 0.017691097455099225, "loss": 0.682102108001709, "time": 0.8155062913894653, "epoch": 7, "memory": 28783, "step": 35148} +{"lr": 0.0007959536998847743, "data_time": 0.0019195795059204102, "grad_norm": 0.01677331989631057, "loss": 0.6869235813617707, "time": 0.8206398487091064, "epoch": 7, "memory": 28783, "step": 35248} +{"lr": 0.0007959536998847743, "data_time": 0.29232115745544435, "grad_norm": 0.015201316401362418, "loss": 0.6860824525356293, "time": 0.8040383577346801, "epoch": 7, "memory": 28783, "step": 35348} +{"lr": 0.0007959536998847743, "data_time": 0.33966894149780275, "grad_norm": 0.01976326904259622, "loss": 0.6853908598423004, "time": 0.841283130645752, "epoch": 7, "memory": 28783, "step": 35448} +{"lr": 0.0007959536998847743, "data_time": 0.28628833293914796, "grad_norm": 0.022069073142483832, "loss": 0.6889750838279725, "time": 0.8028847455978394, "epoch": 7, "memory": 28783, "step": 35548} +{"lr": 0.0007959536998847743, "data_time": 0.3324286937713623, "grad_norm": 0.018840861273929477, "loss": 0.6857481360435486, "time": 0.8515539407730103, "epoch": 7, "memory": 28783, "step": 35648} +{"lr": 0.0007959536998847743, "data_time": 0.32093422412872313, "grad_norm": 0.017893133498728276, "loss": 0.6867282271385193, "time": 0.8367080688476562, "epoch": 7, "memory": 28783, "step": 35748} +{"lr": 0.0007959536998847743, "data_time": 0.3767037630081177, "grad_norm": 0.015831386763602497, "loss": 0.6879664957523346, "time": 0.8953001260757446, "epoch": 7, "memory": 28783, "step": 35848} +{"lr": 0.0007959536998847743, "data_time": 0.34341838359832766, "grad_norm": 0.022307319147512317, "loss": 0.6852727890014648, "time": 0.8477460622787476, "epoch": 7, "memory": 28783, "step": 35948} +{"lr": 0.0007959536998847743, "data_time": 0.3418277263641357, "grad_norm": 0.019870053441263736, "loss": 0.686118996143341, "time": 0.8566797971725464, "epoch": 7, "memory": 28783, "step": 36048} +{"lr": 0.0007959536998847743, "data_time": 0.31561152935028075, "grad_norm": 0.014727455470710993, "loss": 0.6849813759326935, "time": 0.8189444780349732, "epoch": 7, "memory": 28783, "step": 36148} +{"lr": 0.0007959536998847743, "data_time": 0.32175674438476565, "grad_norm": 0.02106335312128067, "loss": 0.6842724800109863, "time": 0.8381194353103638, "epoch": 7, "memory": 28783, "step": 36248} +{"lr": 0.0007959536998847743, "data_time": 0.3440293550491333, "grad_norm": 0.013122439105063676, "loss": 0.6875522017478943, "time": 0.850522494316101, "epoch": 7, "memory": 28783, "step": 36348} +{"lr": 0.0007959536998847743, "data_time": 0.2884658336639404, "grad_norm": 0.014520787284709513, "loss": 0.6893287718296051, "time": 0.7996633291244507, "epoch": 7, "memory": 28783, "step": 36448} +{"lr": 0.0007959536998847743, "data_time": 0.3208257436752319, "grad_norm": 0.015263349004089832, "loss": 0.6875774025917053, "time": 0.8252098798751831, "epoch": 7, "memory": 28783, "step": 36548} +{"lr": 0.0007959536998847743, "data_time": 0.27292938232421876, "grad_norm": 0.020008544623851775, "loss": 0.6905016839504242, "time": 0.7751622676849366, "epoch": 7, "memory": 28783, "step": 36648} +{"lr": 0.0007959536998847743, "data_time": 0.28741676807403566, "grad_norm": 0.01480810260400176, "loss": 0.6877633213996888, "time": 0.7908438444137573, "epoch": 7, "memory": 28783, "step": 36748} +{"lr": 0.0007959536998847743, "data_time": 0.29274702072143555, "grad_norm": 0.021225845348089932, "loss": 0.6866928994655609, "time": 0.7962955713272095, "epoch": 7, "memory": 28783, "step": 36848} +{"lr": 0.0007959536998847743, "data_time": 0.24992833137512208, "grad_norm": 0.021492880955338477, "loss": 0.6857226133346558, "time": 0.7542682647705078, "epoch": 7, "memory": 28783, "step": 36948} +{"lr": 0.0007959536998847743, "data_time": 0.2763364315032959, "grad_norm": 0.01923343800008297, "loss": 0.6883521258831025, "time": 0.7801258563995361, "epoch": 7, "memory": 28783, "step": 37048} +{"lr": 0.0007959536998847743, "data_time": 0.3374753952026367, "grad_norm": 0.020849947445094587, "loss": 0.688578337430954, "time": 0.8415805339813233, "epoch": 7, "memory": 28783, "step": 37148} +{"lr": 0.0007959536998847743, "data_time": 0.3770790100097656, "grad_norm": 0.018818801792804152, "loss": 0.6848691463470459, "time": 0.8801254510879517, "epoch": 7, "memory": 28783, "step": 37248} +{"lr": 0.0007959536998847743, "data_time": 0.297573184967041, "grad_norm": 0.018131504766643047, "loss": 0.6892874658107757, "time": 0.8020789861679077, "epoch": 7, "memory": 28783, "step": 37348} +{"lr": 0.0007959536998847743, "data_time": 0.20495426654815674, "grad_norm": 0.018579118931666017, "loss": 0.6837514460086822, "time": 0.7166455507278442, "epoch": 7, "memory": 28783, "step": 37448} +{"lr": 0.0007959536998847743, "data_time": 0.0017408132553100586, "grad_norm": 0.013621792895719409, "loss": 0.684460973739624, "time": 0.7740114450454711, "epoch": 7, "memory": 28783, "step": 37548} +{"lr": 0.0007959536998847743, "data_time": 0.08445844650268555, "grad_norm": 0.013314396888017655, "loss": 0.6864557564258575, "time": 0.8408969640731812, "epoch": 7, "memory": 28783, "step": 37648} +{"lr": 0.0007959536998847743, "data_time": 0.0017871618270874023, "grad_norm": 0.018849158845841884, "loss": 0.6880821228027344, "time": 0.8186265468597412, "epoch": 7, "memory": 28783, "step": 37748} +{"lr": 0.0007959536998847743, "data_time": 0.0019169330596923828, "grad_norm": 0.024961274396628142, "loss": 0.6845693945884704, "time": 0.8447279214859009, "epoch": 7, "memory": 28783, "step": 37848} +{"lr": 0.0007959536998847743, "data_time": 0.0016588687896728516, "grad_norm": 0.01537142035085708, "loss": 0.6839743673801422, "time": 0.8644203424453736, "epoch": 7, "memory": 28783, "step": 37948} +{"lr": 0.0007959536998847743, "data_time": 0.0023919343948364258, "grad_norm": 0.01768775642849505, "loss": 0.6877807736396789, "time": 0.8464528799057007, "epoch": 7, "memory": 28783, "step": 38048} +{"lr": 0.0007959536998847743, "data_time": 0.0018649816513061524, "grad_norm": 0.018632559943944214, "loss": 0.6885584950447082, "time": 0.8780155420303345, "epoch": 7, "memory": 28783, "step": 38148} +{"lr": 0.0007959536998847743, "data_time": 0.001674962043762207, "grad_norm": 0.01983517063781619, "loss": 0.685562115907669, "time": 0.8226769447326661, "epoch": 7, "memory": 28783, "step": 38248} +{"lr": 0.0007959536998847743, "data_time": 0.0018870830535888672, "grad_norm": 0.018642416410148143, "loss": 0.6852209508419037, "time": 0.8511182069778442, "epoch": 7, "memory": 28783, "step": 38348} +{"lr": 0.0007959536998847743, "data_time": 0.0017599105834960938, "grad_norm": 0.019039638852700592, "loss": 0.6851335883140564, "time": 0.8081854581832886, "epoch": 7, "memory": 28783, "step": 38448} +{"lr": 0.0007959536998847743, "data_time": 0.0017230510711669922, "grad_norm": 0.017614476894959808, "loss": 0.6859308362007142, "time": 0.7961933135986328, "epoch": 7, "memory": 28783, "step": 38548} +{"lr": 0.0007959536998847743, "data_time": 0.001994490623474121, "grad_norm": 0.01535075893625617, "loss": 0.6879223108291626, "time": 0.8690444469451905, "epoch": 7, "memory": 28783, "step": 38648} +{"lr": 0.0007959536998847743, "data_time": 0.0021904468536376952, "grad_norm": 0.8099116334691644, "loss": 0.6889193534851075, "time": 0.8068270444869995, "epoch": 7, "memory": 28783, "step": 38748} +{"lr": 0.0007959536998847743, "data_time": 0.0021352052688598635, "grad_norm": 0.023233903013169764, "loss": 0.6845666885375976, "time": 0.8488040685653686, "epoch": 7, "memory": 28783, "step": 38848} +{"lr": 0.0007959536998847743, "data_time": 0.001798558235168457, "grad_norm": 0.014812073856592178, "loss": 0.6848809897899628, "time": 0.8234731197357178, "epoch": 7, "memory": 28783, "step": 38948} +{"lr": 0.0007959536998847743, "data_time": 0.0018880128860473632, "grad_norm": 0.011171391513198615, "loss": 0.6835210800170899, "time": 0.8097221374511718, "epoch": 7, "memory": 28783, "step": 39048} +{"lr": 0.0007959536998847743, "data_time": 0.0019295930862426758, "grad_norm": 0.0177352077094838, "loss": 0.6859819829463959, "time": 0.8393646001815795, "epoch": 7, "memory": 28783, "step": 39148} +{"lr": 0.0007959536998847743, "data_time": 0.0017655134201049806, "grad_norm": 0.01382433008402586, "loss": 0.6885336220264435, "time": 0.8047173023223877, "epoch": 7, "memory": 28783, "step": 39248} +{"lr": 0.0007959536998847743, "data_time": 0.0018420219421386719, "grad_norm": 0.01699437089264393, "loss": 0.6874267578125, "time": 0.8533265829086304, "epoch": 7, "memory": 28783, "step": 39348} +{"lr": 0.0007959536998847743, "data_time": 0.001817798614501953, "grad_norm": 0.019277336332015692, "loss": 0.681348568201065, "time": 0.8902977466583252, "epoch": 7, "memory": 28783, "step": 39448} +{"lr": 0.0007959536998847743, "data_time": 0.0017141342163085938, "grad_norm": 0.017996445181779563, "loss": 0.6892510950565338, "time": 0.8733532428741455, "epoch": 7, "memory": 28783, "step": 39548} +{"lr": 0.0007959536998847743, "data_time": 0.0018942594528198243, "grad_norm": 0.017822445137426257, "loss": 0.6878711998462677, "time": 0.8268157720565796, "epoch": 7, "memory": 28783, "step": 39648} +{"lr": 0.0007959536998847743, "data_time": 0.002043890953063965, "grad_norm": 0.014071428310126066, "loss": 0.6857958257198333, "time": 0.9307317733764648, "epoch": 7, "memory": 28783, "step": 39748} +{"lr": 0.0007959536998847743, "data_time": 0.0017424345016479493, "grad_norm": 0.009252369753085077, "loss": 0.6836173892021179, "time": 0.79527747631073, "epoch": 7, "memory": 28783, "step": 39848} +{"lr": 0.0007959536998847743, "data_time": 0.0017492294311523438, "grad_norm": 0.0202229589689523, "loss": 0.6898975431919098, "time": 0.8185217380523682, "epoch": 7, "memory": 28783, "step": 39948} +{"lr": 0.0007959536998847743, "data_time": 0.001741337776184082, "grad_norm": 0.012179149826988579, "loss": 0.688639760017395, "time": 0.7978439092636108, "epoch": 7, "memory": 28783, "step": 40048} +{"lr": 0.0007959536998847743, "data_time": 0.001706242561340332, "grad_norm": 0.01295425179414451, "loss": 0.6856117606163025, "time": 0.8520701408386231, "epoch": 7, "memory": 28783, "step": 40148} +{"lr": 0.0007959536998847743, "data_time": 0.0017007112503051758, "grad_norm": 0.01737766668666154, "loss": 0.6840229749679565, "time": 0.8295711040496826, "epoch": 7, "memory": 28783, "step": 40248} +{"accuracy/top1": 100.0, "data_time": 0.22127382895525763, "time": 0.6371942407944623, "step": 7} +{"lr": 0.0007297252973710758, "data_time": 0.1767284631729126, "grad_norm": 0.01150515889748931, "loss": 0.6880894839763642, "time": 0.8509277105331421, "epoch": 8, "memory": 28783, "step": 40406} +{"lr": 0.0007297252973710758, "data_time": 0.21086158752441406, "grad_norm": 0.015664160903543233, "loss": 0.6834785640239716, "time": 0.8216288805007934, "epoch": 8, "memory": 28783, "step": 40506} +{"lr": 0.0007297252973710758, "data_time": 0.0019620418548583984, "grad_norm": 0.013502050004899502, "loss": 0.6881194651126862, "time": 0.8162750720977783, "epoch": 8, "memory": 28783, "step": 40606} +{"lr": 0.0007297252973710758, "data_time": 0.0016234636306762696, "grad_norm": 0.019846187462098895, "loss": 0.6858789980411529, "time": 0.7795439481735229, "epoch": 8, "memory": 28783, "step": 40706} +{"lr": 0.0007297252973710758, "data_time": 0.0015878915786743165, "grad_norm": 0.019163573649711906, "loss": 0.6904479563236237, "time": 0.8327592849731446, "epoch": 8, "memory": 28783, "step": 40806} +{"lr": 0.0007297252973710758, "data_time": 0.001642751693725586, "grad_norm": 0.02173926420509815, "loss": 0.6876375555992127, "time": 0.807506513595581, "epoch": 8, "memory": 28783, "step": 40906} +{"lr": 0.0007297252973710758, "data_time": 0.0024679422378540037, "grad_norm": 0.011791924037970603, "loss": 0.684174531698227, "time": 0.8112164735794067, "epoch": 8, "memory": 28783, "step": 41006} +{"lr": 0.0007297252973710758, "data_time": 0.002004408836364746, "grad_norm": 0.01406180146150291, "loss": 0.6850703954696655, "time": 0.7763309240341186, "epoch": 8, "memory": 28783, "step": 41106} +{"lr": 0.0007297252973710758, "data_time": 0.001792311668395996, "grad_norm": 0.016189863812178372, "loss": 0.6845689654350281, "time": 0.8363158464431762, "epoch": 8, "memory": 28783, "step": 41206} +{"lr": 0.0007297252973710758, "data_time": 0.0016970634460449219, "grad_norm": 0.02064953404478729, "loss": 0.6891416013240814, "time": 0.830332612991333, "epoch": 8, "memory": 28783, "step": 41306} +{"lr": 0.0007297252973710758, "data_time": 0.0017680883407592773, "grad_norm": 0.02018622092436999, "loss": 0.6847874343395233, "time": 0.7984568357467652, "epoch": 8, "memory": 28783, "step": 41406} +{"lr": 0.0007297252973710758, "data_time": 0.36053225994110105, "grad_norm": 0.020458186184987424, "loss": 0.6852293372154236, "time": 0.8626003980636596, "epoch": 8, "memory": 28783, "step": 41506} +{"lr": 0.0007297252973710758, "data_time": 0.2516258001327515, "grad_norm": 0.01941661594901234, "loss": 0.6865932166576385, "time": 0.7552858114242553, "epoch": 8, "memory": 28783, "step": 41606} +{"lr": 0.0007297252973710758, "data_time": 0.2104280710220337, "grad_norm": 0.011324568837881088, "loss": 0.6893284142017364, "time": 0.7168056011199951, "epoch": 8, "memory": 28783, "step": 41706} +{"lr": 0.0007297252973710758, "data_time": 0.17278699874877929, "grad_norm": 0.020460043149068952, "loss": 0.685898506641388, "time": 0.7893294334411621, "epoch": 8, "memory": 28783, "step": 41806} +{"lr": 0.0007297252973710758, "data_time": 0.0018121004104614258, "grad_norm": 0.015894679352641106, "loss": 0.6870485007762909, "time": 0.8094709634780883, "epoch": 8, "memory": 28783, "step": 41906} +{"lr": 0.0007297252973710758, "data_time": 0.0020395517349243164, "grad_norm": 0.020082036941312255, "loss": 0.6875883340835571, "time": 0.800652003288269, "epoch": 8, "memory": 28783, "step": 42006} +{"lr": 0.0007297252973710758, "data_time": 0.0020413875579833986, "grad_norm": 0.017825177405029535, "loss": 0.6895458459854126, "time": 0.8972333908081055, "epoch": 8, "memory": 28783, "step": 42106} +{"lr": 0.0007297252973710758, "data_time": 0.0025058269500732424, "grad_norm": 0.014239225978963077, "loss": 0.684444534778595, "time": 0.8005325078964234, "epoch": 8, "memory": 28783, "step": 42206} +{"lr": 0.0007297252973710758, "data_time": 0.0016438722610473632, "grad_norm": 0.01622798021417111, "loss": 0.6883606851100922, "time": 0.782297945022583, "epoch": 8, "memory": 28783, "step": 42306} +{"lr": 0.0007297252973710758, "data_time": 0.0020476818084716798, "grad_norm": 0.022021299693733454, "loss": 0.688879132270813, "time": 0.8572196006774903, "epoch": 8, "memory": 28783, "step": 42406} +{"lr": 0.0007297252973710758, "data_time": 0.0032892227172851562, "grad_norm": 0.010213263262994588, "loss": 0.6837104856967926, "time": 0.8039742231369018, "epoch": 8, "memory": 28783, "step": 42506} +{"lr": 0.0007297252973710758, "data_time": 0.001682424545288086, "grad_norm": 0.015494284802116454, "loss": 0.6841726422309875, "time": 0.8095448255538941, "epoch": 8, "memory": 28783, "step": 42606} +{"lr": 0.0007297252973710758, "data_time": 0.001751542091369629, "grad_norm": 0.016526938928291202, "loss": 0.6859462976455688, "time": 0.8550708293914795, "epoch": 8, "memory": 28783, "step": 42706} +{"lr": 0.0007297252973710758, "data_time": 0.002066969871520996, "grad_norm": 0.017640549945645033, "loss": 0.6890587925910949, "time": 0.8095893144607544, "epoch": 8, "memory": 28783, "step": 42806} +{"lr": 0.0007297252973710758, "data_time": 0.0016910552978515625, "grad_norm": 0.021079135406762362, "loss": 0.6901887834072113, "time": 0.8429555177688599, "epoch": 8, "memory": 28783, "step": 42906} +{"lr": 0.0007297252973710758, "data_time": 0.002324557304382324, "grad_norm": 0.018978520389646293, "loss": 0.6881242215633392, "time": 0.8127419948577881, "epoch": 8, "memory": 28783, "step": 43006} +{"lr": 0.0007297252973710758, "data_time": 0.0017854928970336913, "grad_norm": 0.009593458962626755, "loss": 0.6858097434043884, "time": 0.7912485837936402, "epoch": 8, "memory": 28783, "step": 43106} +{"lr": 0.0007297252973710758, "data_time": 0.0018796205520629882, "grad_norm": 0.02009905909653753, "loss": 0.6858508288860321, "time": 0.8097288370132446, "epoch": 8, "memory": 28783, "step": 43206} +{"lr": 0.0007297252973710758, "data_time": 0.0020702838897705077, "grad_norm": 0.017961289966478944, "loss": 0.6840964615345001, "time": 0.7968644857406616, "epoch": 8, "memory": 28783, "step": 43306} +{"lr": 0.0007297252973710758, "data_time": 0.0017736196517944337, "grad_norm": 0.018074063712265342, "loss": 0.6889858484268189, "time": 0.8258228540420532, "epoch": 8, "memory": 28783, "step": 43406} +{"lr": 0.0007297252973710758, "data_time": 0.0016751766204833984, "grad_norm": 0.01450836684089154, "loss": 0.6880022525787354, "time": 0.8089641809463501, "epoch": 8, "memory": 28783, "step": 43506} +{"lr": 0.0007297252973710758, "data_time": 0.0016956567764282227, "grad_norm": 0.014779439964331686, "loss": 0.6862829804420472, "time": 0.7844366550445556, "epoch": 8, "memory": 28783, "step": 43606} +{"lr": 0.0007297252973710758, "data_time": 0.0016278505325317382, "grad_norm": 0.012287978845415637, "loss": 0.6869882702827453, "time": 0.8027417898178101, "epoch": 8, "memory": 28783, "step": 43706} +{"lr": 0.0007297252973710758, "data_time": 0.0017699718475341797, "grad_norm": 0.016230658342828976, "loss": 0.6827964782714844, "time": 0.8362174510955811, "epoch": 8, "memory": 28783, "step": 43806} +{"lr": 0.0007297252973710758, "data_time": 0.0016646385192871094, "grad_norm": 0.021101759106386453, "loss": 0.6868793904781342, "time": 0.8031327247619628, "epoch": 8, "memory": 28783, "step": 43906} +{"lr": 0.0007297252973710758, "data_time": 0.0024164438247680662, "grad_norm": 0.012828164966776967, "loss": 0.6843660533428192, "time": 0.8144444465637207, "epoch": 8, "memory": 28783, "step": 44006} +{"lr": 0.0007297252973710758, "data_time": 0.0018096923828125, "grad_norm": 0.008951877418439835, "loss": 0.6857014656066894, "time": 0.8152435541152954, "epoch": 8, "memory": 28783, "step": 44106} +{"lr": 0.0007297252973710758, "data_time": 0.0025657176971435546, "grad_norm": 0.013029518257826566, "loss": 0.685771006345749, "time": 0.8057149648666382, "epoch": 8, "memory": 28783, "step": 44206} +{"lr": 0.0007297252973710758, "data_time": 0.0022495508193969725, "grad_norm": 0.017793799098581076, "loss": 0.6838438749313355, "time": 0.8712785243988037, "epoch": 8, "memory": 28783, "step": 44306} +{"lr": 0.0007297252973710758, "data_time": 0.002106046676635742, "grad_norm": 0.015450227167457342, "loss": 0.687344777584076, "time": 0.8256576061248779, "epoch": 8, "memory": 28783, "step": 44406} +{"lr": 0.0007297252973710758, "data_time": 0.0024627685546875, "grad_norm": 0.018130247574299575, "loss": 0.6861902713775635, "time": 0.7684907197952271, "epoch": 8, "memory": 28783, "step": 44506} +{"lr": 0.0007297252973710758, "data_time": 0.0018656015396118163, "grad_norm": 0.010195440729148686, "loss": 0.6883004784584046, "time": 0.7909414291381835, "epoch": 8, "memory": 28783, "step": 44606} +{"lr": 0.0007297252973710758, "data_time": 0.0026448726654052734, "grad_norm": 0.0206029093824327, "loss": 0.6878214597702026, "time": 0.8168389081954956, "epoch": 8, "memory": 28783, "step": 44706} +{"lr": 0.0007297252973710758, "data_time": 0.0029001951217651365, "grad_norm": 0.014642440964234994, "loss": 0.6859261691570282, "time": 0.8484565258026123, "epoch": 8, "memory": 28783, "step": 44806} +{"lr": 0.0007297252973710758, "data_time": 0.0024739980697631838, "grad_norm": 0.017981468280777336, "loss": 0.6885385870933532, "time": 0.7753104209899903, "epoch": 8, "memory": 28783, "step": 44906} +{"lr": 0.0007297252973710758, "data_time": 0.0023864984512329103, "grad_norm": 0.014773014979436994, "loss": 0.6869674801826477, "time": 0.8395127534866333, "epoch": 8, "memory": 28783, "step": 45006} +{"lr": 0.0007297252973710758, "data_time": 0.002082347869873047, "grad_norm": 0.009750491514569148, "loss": 0.6907267987728118, "time": 0.8047058582305908, "epoch": 8, "memory": 28783, "step": 45106} +{"lr": 0.0007297252973710758, "data_time": 0.001661968231201172, "grad_norm": 0.015422008372843266, "loss": 0.6874495625495911, "time": 0.8290945053100586, "epoch": 8, "memory": 28783, "step": 45206} +{"lr": 0.0007297252973710758, "data_time": 0.001624298095703125, "grad_norm": 0.018805124121718107, "loss": 0.6828702390193939, "time": 0.815378975868225, "epoch": 8, "memory": 28783, "step": 45306} +{"lr": 0.0007297252973710758, "data_time": 0.001702880859375, "grad_norm": 0.022728974279016257, "loss": 0.6894953548908234, "time": 0.7922136306762695, "epoch": 8, "memory": 28783, "step": 45406} +{"lr": 0.0007297252973710758, "data_time": 0.00167391300201416, "grad_norm": 0.017786211939528583, "loss": 0.6875094890594482, "time": 0.775871729850769, "epoch": 8, "memory": 28783, "step": 45506} +{"lr": 0.0007297252973710758, "data_time": 0.0019589900970458985, "grad_norm": 0.016535743908025324, "loss": 0.6843287467956543, "time": 0.7867937803268432, "epoch": 8, "memory": 28783, "step": 45606} +{"lr": 0.0007297252973710758, "data_time": 0.0022088289260864258, "grad_norm": 0.017723800521343945, "loss": 0.6848374128341674, "time": 0.7778054475784302, "epoch": 8, "memory": 28783, "step": 45706} +{"lr": 0.0007297252973710758, "data_time": 0.001461338996887207, "grad_norm": 0.014086908928584307, "loss": 0.6838295698165894, "time": 0.7659364461898803, "epoch": 8, "memory": 28783, "step": 45806} +{"lr": 0.0007297252973710758, "data_time": 0.001958465576171875, "grad_norm": 0.015010241628624498, "loss": 0.6883653104305267, "time": 0.7757092237472534, "epoch": 8, "memory": 28783, "step": 45906} +{"lr": 0.0007297252973710758, "data_time": 0.0021871089935302734, "grad_norm": 0.01669732518494129, "loss": 0.6845536231994629, "time": 0.7965921878814697, "epoch": 8, "memory": 28783, "step": 46006} +{"accuracy/top1": 100.0, "data_time": 0.23743409268996296, "time": 0.6569828285890467, "step": 8} +{"lr": 0.0006579634122155991, "data_time": 0.0024076223373413084, "grad_norm": 0.019781443290412426, "loss": 0.6864916682243347, "time": 0.7794753074645996, "epoch": 9, "memory": 28783, "step": 46164} +{"lr": 0.0006579634122155991, "data_time": 0.0019613027572631834, "grad_norm": 0.019997008610516787, "loss": 0.6887592673301697, "time": 0.8062206029891967, "epoch": 9, "memory": 28783, "step": 46264} +{"lr": 0.0006579634122155991, "data_time": 0.00228729248046875, "grad_norm": 0.012811769964173436, "loss": 0.6853896021842957, "time": 0.8259541273117066, "epoch": 9, "memory": 28783, "step": 46364} +{"lr": 0.0006579634122155991, "data_time": 0.0019213199615478516, "grad_norm": 0.017396972794085742, "loss": 0.6875722289085389, "time": 0.7887672185897827, "epoch": 9, "memory": 28783, "step": 46464} +{"lr": 0.0006579634122155991, "data_time": 0.002908778190612793, "grad_norm": 0.01309593878686428, "loss": 0.6862093210220337, "time": 0.854130482673645, "epoch": 9, "memory": 28783, "step": 46564} +{"lr": 0.0006579634122155991, "data_time": 0.0026603460311889647, "grad_norm": 0.01784549541771412, "loss": 0.6867650210857391, "time": 0.7887613534927368, "epoch": 9, "memory": 28783, "step": 46664} +{"lr": 0.0006579634122155991, "data_time": 0.003061199188232422, "grad_norm": 0.0178492097184062, "loss": 0.6855368733406066, "time": 0.787494444847107, "epoch": 9, "memory": 28783, "step": 46764} +{"lr": 0.0006579634122155991, "data_time": 0.002389192581176758, "grad_norm": 0.012624053051695228, "loss": 0.6891029179096222, "time": 0.7666549682617188, "epoch": 9, "memory": 28783, "step": 46864} +{"lr": 0.0006579634122155991, "data_time": 0.0022454261779785156, "grad_norm": 0.012896141922101378, "loss": 0.685942804813385, "time": 0.7550723552703857, "epoch": 9, "memory": 28783, "step": 46964} +{"lr": 0.0006579634122155991, "data_time": 0.0018169641494750976, "grad_norm": 0.01250883205793798, "loss": 0.6913094282150268, "time": 0.788721227645874, "epoch": 9, "memory": 28783, "step": 47064} +{"lr": 0.0006579634122155991, "data_time": 0.0017420530319213867, "grad_norm": 0.016285714320838453, "loss": 0.6868237376213073, "time": 0.7884947538375855, "epoch": 9, "memory": 28783, "step": 47164} +{"lr": 0.0006579634122155991, "data_time": 0.0018303632736206056, "grad_norm": 0.017772531998343766, "loss": 0.6867738842964173, "time": 0.8098517656326294, "epoch": 9, "memory": 28783, "step": 47264} +{"lr": 0.0006579634122155991, "data_time": 0.002754521369934082, "grad_norm": 0.016665896121412515, "loss": 0.6840065836906433, "time": 0.8419631958007813, "epoch": 9, "memory": 28783, "step": 47364} +{"lr": 0.0006579634122155991, "data_time": 0.0017142772674560546, "grad_norm": 0.016265262814704327, "loss": 0.6880294740200043, "time": 0.7701505184173584, "epoch": 9, "memory": 28783, "step": 47464} +{"lr": 0.0006579634122155991, "data_time": 0.0018386125564575195, "grad_norm": 0.014546527637867257, "loss": 0.6875187039375306, "time": 0.8083031892776489, "epoch": 9, "memory": 28783, "step": 47564} +{"lr": 0.0006579634122155991, "data_time": 0.0020377397537231444, "grad_norm": 0.015437063807621599, "loss": 0.6856964707374573, "time": 0.8418564558029175, "epoch": 9, "memory": 28783, "step": 47664} +{"lr": 0.0006579634122155991, "data_time": 0.002225494384765625, "grad_norm": 0.0065986689180135725, "loss": 0.6867503643035888, "time": 0.8010595798492431, "epoch": 9, "memory": 28783, "step": 47764} +{"lr": 0.0006579634122155991, "data_time": 0.002016448974609375, "grad_norm": 0.014555867318995297, "loss": 0.6886663317680359, "time": 0.7784498691558838, "epoch": 9, "memory": 28783, "step": 47864} +{"lr": 0.0006579634122155991, "data_time": 0.0016640663146972657, "grad_norm": 0.015458536427468061, "loss": 0.6875035107135773, "time": 0.8535054683685303, "epoch": 9, "memory": 28783, "step": 47964} +{"lr": 0.0006579634122155991, "data_time": 0.0017481088638305665, "grad_norm": 0.028671895572915673, "loss": 0.6898967027664185, "time": 0.8276613235473633, "epoch": 9, "memory": 28783, "step": 48064} +{"lr": 0.0006579634122155991, "data_time": 0.0016767501831054688, "grad_norm": 0.022093181125819684, "loss": 0.6856845915317535, "time": 0.7595356225967407, "epoch": 9, "memory": 28783, "step": 48164} +{"lr": 0.0006579634122155991, "data_time": 0.0019210100173950196, "grad_norm": 0.014214502763934433, "loss": 0.6850096583366394, "time": 0.8262840747833252, "epoch": 9, "memory": 28783, "step": 48264} +{"lr": 0.0006579634122155991, "data_time": 0.0016521215438842773, "grad_norm": 0.017442161875078455, "loss": 0.6890328764915467, "time": 0.8380665302276611, "epoch": 9, "memory": 28783, "step": 48364} +{"lr": 0.0006579634122155991, "data_time": 0.0025158166885375977, "grad_norm": 0.011965154483914376, "loss": 0.687999427318573, "time": 0.7937122583389282, "epoch": 9, "memory": 28783, "step": 48464} +{"lr": 0.0006579634122155991, "data_time": 0.0017914533615112304, "grad_norm": 0.01684018108062446, "loss": 0.6870417118072509, "time": 0.8436041116714478, "epoch": 9, "memory": 28783, "step": 48564} +{"lr": 0.0006579634122155991, "data_time": 0.0017537593841552735, "grad_norm": 0.023170825937995687, "loss": 0.6853677570819855, "time": 0.7876108646392822, "epoch": 9, "memory": 28783, "step": 48664} +{"lr": 0.0006579634122155991, "data_time": 0.00168609619140625, "grad_norm": 0.011022717703599482, "loss": 0.6874555468559265, "time": 0.810965633392334, "epoch": 9, "memory": 28783, "step": 48764} +{"lr": 0.0006579634122155991, "data_time": 0.0018185853958129882, "grad_norm": 0.013551229005679488, "loss": 0.6873404622077942, "time": 0.786519455909729, "epoch": 9, "memory": 28783, "step": 48864} +{"lr": 0.0006579634122155991, "data_time": 0.002039384841918945, "grad_norm": 0.019113608356565236, "loss": 0.6864564061164856, "time": 0.812389326095581, "epoch": 9, "memory": 28783, "step": 48964} +{"lr": 0.0006579634122155991, "data_time": 0.002731513977050781, "grad_norm": 0.021905680443160234, "loss": 0.6883671879768372, "time": 0.8831088304519653, "epoch": 9, "memory": 28783, "step": 49064} +{"lr": 0.0006579634122155991, "data_time": 0.0024415016174316405, "grad_norm": 0.014867144648451357, "loss": 0.6886488735675812, "time": 0.8272521495819092, "epoch": 9, "memory": 28783, "step": 49164} +{"lr": 0.0006579634122155991, "data_time": 0.0033639192581176756, "grad_norm": 0.022948447009548546, "loss": 0.6842088460922241, "time": 0.7653190612792968, "epoch": 9, "memory": 28783, "step": 49264} +{"lr": 0.0006579634122155991, "data_time": 0.0027080774307250977, "grad_norm": 0.01826594895683229, "loss": 0.6880537033081054, "time": 0.7956825494766235, "epoch": 9, "memory": 28783, "step": 49364} +{"lr": 0.0006579634122155991, "data_time": 0.0019319772720336914, "grad_norm": 0.017960479762405158, "loss": 0.6865352869033814, "time": 0.788453483581543, "epoch": 9, "memory": 28783, "step": 49464} +{"lr": 0.0006579634122155991, "data_time": 0.0023500204086303713, "grad_norm": 0.018686064379289748, "loss": 0.6842729091644287, "time": 0.7978798389434815, "epoch": 9, "memory": 28783, "step": 49564} +{"lr": 0.0006579634122155991, "data_time": 0.00227503776550293, "grad_norm": 0.020437034708447755, "loss": 0.6884462833404541, "time": 0.8601483583450318, "epoch": 9, "memory": 28783, "step": 49664} +{"lr": 0.0006579634122155991, "data_time": 0.002756357192993164, "grad_norm": 0.010903456446249038, "loss": 0.687099039554596, "time": 0.7962905406951905, "epoch": 9, "memory": 28783, "step": 49764} +{"lr": 0.0006579634122155991, "data_time": 0.0024381160736083986, "grad_norm": 0.014546839147806167, "loss": 0.6866072475910187, "time": 0.7885930061340332, "epoch": 9, "memory": 28783, "step": 49864} +{"lr": 0.0006579634122155991, "data_time": 0.0030847787857055664, "grad_norm": 0.01970738952513784, "loss": 0.6827532708644867, "time": 0.8508641958236695, "epoch": 9, "memory": 28783, "step": 49964} +{"lr": 0.0006579634122155991, "data_time": 0.0020726919174194336, "grad_norm": 0.016316135716624557, "loss": 0.6884059429168701, "time": 0.7776422739028931, "epoch": 9, "memory": 28783, "step": 50064} +{"lr": 0.0006579634122155991, "data_time": 0.002725529670715332, "grad_norm": 0.020280647068284453, "loss": 0.6885368764400482, "time": 0.8471766710281372, "epoch": 9, "memory": 28783, "step": 50164} +{"lr": 0.0006579634122155991, "data_time": 0.00283968448638916, "grad_norm": 0.01854660043027252, "loss": 0.6863507449626922, "time": 0.7698218584060669, "epoch": 9, "memory": 28783, "step": 50264} +{"lr": 0.0006579634122155991, "data_time": 0.0025454044342041017, "grad_norm": 0.00827188534894958, "loss": 0.6900393664836884, "time": 0.8138433933258057, "epoch": 9, "memory": 28783, "step": 50364} +{"lr": 0.0006579634122155991, "data_time": 0.0023783206939697265, "grad_norm": 0.016054447554051877, "loss": 0.6873500943183899, "time": 0.848004150390625, "epoch": 9, "memory": 28783, "step": 50464} +{"lr": 0.0006579634122155991, "data_time": 0.0027923107147216795, "grad_norm": 0.013526805839501322, "loss": 0.6884520709514618, "time": 0.8253152370452881, "epoch": 9, "memory": 28783, "step": 50564} +{"lr": 0.0006579634122155991, "data_time": 0.0026437997817993163, "grad_norm": 0.01568017478566617, "loss": 0.6882683515548706, "time": 0.8205284357070923, "epoch": 9, "memory": 28783, "step": 50664} +{"lr": 0.0006579634122155991, "data_time": 0.002894091606140137, "grad_norm": 0.020560159953311084, "loss": 0.6856422841548919, "time": 0.8077261924743653, "epoch": 9, "memory": 28783, "step": 50764} +{"lr": 0.0006579634122155991, "data_time": 0.0027227878570556642, "grad_norm": 0.013660538289695979, "loss": 0.6901677668094635, "time": 0.8817827939987183, "epoch": 9, "memory": 28783, "step": 50864} +{"lr": 0.0006579634122155991, "data_time": 0.002523994445800781, "grad_norm": 0.011582893878221511, "loss": 0.6891176402568817, "time": 0.843101167678833, "epoch": 9, "memory": 28783, "step": 50964} +{"lr": 0.0006579634122155991, "data_time": 0.0027343034744262695, "grad_norm": 0.026205797819420694, "loss": 0.6876045644283295, "time": 0.8893002033233642, "epoch": 9, "memory": 28783, "step": 51064} +{"lr": 0.0006579634122155991, "data_time": 0.0024560213088989256, "grad_norm": 0.018182085431180894, "loss": 0.686473798751831, "time": 0.7928990840911865, "epoch": 9, "memory": 28783, "step": 51164} +{"lr": 0.0006579634122155991, "data_time": 0.0023867130279541016, "grad_norm": 0.016295361745869742, "loss": 0.6888134717941284, "time": 0.8284400939941406, "epoch": 9, "memory": 28783, "step": 51264} +{"lr": 0.0006579634122155991, "data_time": 0.0023748397827148436, "grad_norm": 0.027016353514045476, "loss": 0.6855190753936767, "time": 0.8068228483200073, "epoch": 9, "memory": 28783, "step": 51364} +{"lr": 0.0006579634122155991, "data_time": 0.0022649288177490233, "grad_norm": 0.01126832386944443, "loss": 0.68656405210495, "time": 0.8059137344360352, "epoch": 9, "memory": 28783, "step": 51464} +{"lr": 0.0006579634122155991, "data_time": 0.0024251937866210938, "grad_norm": 0.017916414327919482, "loss": 0.687969172000885, "time": 0.8256438255310059, "epoch": 9, "memory": 28783, "step": 51564} +{"lr": 0.0006579634122155991, "data_time": 0.0019690752029418944, "grad_norm": 0.01850375476060435, "loss": 0.6857397079467773, "time": 0.824545431137085, "epoch": 9, "memory": 28783, "step": 51664} +{"lr": 0.0006579634122155991, "data_time": 0.0020044326782226564, "grad_norm": 0.01808455412974581, "loss": 0.6860807836055756, "time": 0.8124317407608033, "epoch": 9, "memory": 28783, "step": 51764} +{"accuracy/top1": 100.0, "data_time": 0.22674721830031452, "time": 0.6419453620910645, "step": 9} +{"lr": 0.0005824350601949144, "data_time": 0.15495584011077881, "grad_norm": 0.00975538050988689, "loss": 0.6851868152618408, "time": 0.7970738410949707, "epoch": 10, "memory": 28783, "step": 51922} +{"lr": 0.0005824350601949144, "data_time": 0.002073216438293457, "grad_norm": 0.015719201776664703, "loss": 0.686484295129776, "time": 0.8727296352386474, "epoch": 10, "memory": 28783, "step": 52022} +{"lr": 0.0005824350601949144, "data_time": 0.0018186569213867188, "grad_norm": 0.021165580995148047, "loss": 0.6866058111190796, "time": 0.8343714475631714, "epoch": 10, "memory": 28783, "step": 52122} +{"lr": 0.0005824350601949144, "data_time": 0.0018466711044311523, "grad_norm": 0.015086830453947186, "loss": 0.6867835998535157, "time": 0.8312772035598754, "epoch": 10, "memory": 28783, "step": 52222} +{"lr": 0.0005824350601949144, "data_time": 0.0019497394561767579, "grad_norm": 0.0197906780987978, "loss": 0.6845935881137848, "time": 0.8613074541091919, "epoch": 10, "memory": 28783, "step": 52322} +{"lr": 0.0005824350601949144, "data_time": 0.0019980907440185548, "grad_norm": 0.016787458141334356, "loss": 0.6867964863777161, "time": 0.7988025426864624, "epoch": 10, "memory": 28783, "step": 52422} +{"lr": 0.0005824350601949144, "data_time": 0.0020562410354614258, "grad_norm": 0.020981702487915754, "loss": 0.6887876689434052, "time": 0.8495381355285645, "epoch": 10, "memory": 28783, "step": 52522} +{"lr": 0.0005824350601949144, "data_time": 0.0020780086517333983, "grad_norm": 0.013002254930324853, "loss": 0.6891759812831879, "time": 0.8378580331802368, "epoch": 10, "memory": 28783, "step": 52622} +{"lr": 0.0005824350601949144, "data_time": 0.001979827880859375, "grad_norm": 0.0145253571216017, "loss": 0.6885647475719452, "time": 0.9098420381546021, "epoch": 10, "memory": 28783, "step": 52722} +{"lr": 0.0005824350601949144, "data_time": 0.002054119110107422, "grad_norm": 0.02291962190065533, "loss": 0.687529319524765, "time": 0.8263602495193482, "epoch": 10, "memory": 28783, "step": 52822} +{"lr": 0.0005824350601949144, "data_time": 0.0019410371780395508, "grad_norm": 0.01692937088664621, "loss": 0.6871381521224975, "time": 0.8170312404632568, "epoch": 10, "memory": 28783, "step": 52922} +{"lr": 0.0005824350601949144, "data_time": 0.0020849943161010743, "grad_norm": 0.01803222324233502, "loss": 0.6848252892494202, "time": 0.8109024047851563, "epoch": 10, "memory": 28783, "step": 53022} +{"lr": 0.0005824350601949144, "data_time": 0.0018395185470581055, "grad_norm": 0.022690441505983473, "loss": 0.688908976316452, "time": 0.8494559288024902, "epoch": 10, "memory": 28783, "step": 53122} +{"lr": 0.0005824350601949144, "data_time": 0.003214311599731445, "grad_norm": 0.018362203426659108, "loss": 0.6897436201572418, "time": 0.8518648147583008, "epoch": 10, "memory": 28783, "step": 53222} +{"lr": 0.0005824350601949144, "data_time": 0.0020096302032470703, "grad_norm": 0.01799560021609068, "loss": 0.6845239341259003, "time": 0.7954816102981568, "epoch": 10, "memory": 28783, "step": 53322} +{"lr": 0.0005824350601949144, "data_time": 0.0026044368743896483, "grad_norm": 0.010587186249904335, "loss": 0.687562370300293, "time": 0.8352644920349122, "epoch": 10, "memory": 28783, "step": 53422} +{"lr": 0.0005824350601949144, "data_time": 0.0018564939498901367, "grad_norm": 0.019613031949847936, "loss": 0.6844630062580108, "time": 0.765064811706543, "epoch": 10, "memory": 28783, "step": 53522} +{"lr": 0.0005824350601949144, "data_time": 0.001860833168029785, "grad_norm": 0.01571825020946562, "loss": 0.6879616141319275, "time": 0.804143762588501, "epoch": 10, "memory": 28783, "step": 53622} +{"lr": 0.0005824350601949144, "data_time": 0.0017686128616333009, "grad_norm": 0.02392355175688863, "loss": 0.6858514904975891, "time": 0.7745452642440795, "epoch": 10, "memory": 28783, "step": 53722} +{"lr": 0.0005824350601949144, "data_time": 0.0018432855606079102, "grad_norm": 0.011880033416673541, "loss": 0.683768343925476, "time": 0.7639914751052856, "epoch": 10, "memory": 28783, "step": 53822} +{"lr": 0.0005824350601949144, "data_time": 0.0016751050949096679, "grad_norm": 0.017738129105418922, "loss": 0.6832839012145996, "time": 0.830788254737854, "epoch": 10, "memory": 28783, "step": 53922} +{"lr": 0.0005824350601949144, "data_time": 0.0031814813613891602, "grad_norm": 0.017122036882210524, "loss": 0.6826117038726807, "time": 0.8154723167419433, "epoch": 10, "memory": 28783, "step": 54022} +{"lr": 0.0005824350601949144, "data_time": 0.0021076202392578125, "grad_norm": 0.017524270294234156, "loss": 0.6864458441734314, "time": 0.852436876296997, "epoch": 10, "memory": 28783, "step": 54122} +{"lr": 0.0005824350601949144, "data_time": 0.0026784181594848634, "grad_norm": 0.016776928049512207, "loss": 0.687641030550003, "time": 0.8137255430221557, "epoch": 10, "memory": 28783, "step": 54222} +{"lr": 0.0005824350601949144, "data_time": 0.0022649288177490233, "grad_norm": 0.014092632965184749, "loss": 0.6849035382270813, "time": 0.7912559270858764, "epoch": 10, "memory": 28783, "step": 54322} +{"lr": 0.0005824350601949144, "data_time": 0.0018433094024658202, "grad_norm": 0.012496439449023455, "loss": 0.6896382927894592, "time": 0.8284731864929199, "epoch": 10, "memory": 28783, "step": 54422} +{"lr": 0.0005824350601949144, "data_time": 0.0017837285995483398, "grad_norm": 0.0257398568559438, "loss": 0.6870508193969727, "time": 0.7922945976257324, "epoch": 10, "memory": 28783, "step": 54522} +{"lr": 0.0005824350601949144, "data_time": 0.00201263427734375, "grad_norm": 0.019401557138189675, "loss": 0.6867332875728607, "time": 0.8053722381591797, "epoch": 10, "memory": 28783, "step": 54622} +{"lr": 0.0005824350601949144, "data_time": 0.0019307374954223634, "grad_norm": 0.01231341080274433, "loss": 0.6859995365142822, "time": 0.8189839124679565, "epoch": 10, "memory": 28783, "step": 54722} +{"lr": 0.0005824350601949144, "data_time": 0.0021706581115722655, "grad_norm": 0.01855875412002206, "loss": 0.6885318517684936, "time": 0.7882414340972901, "epoch": 10, "memory": 28783, "step": 54822} +{"lr": 0.0005824350601949144, "data_time": 0.0030200958251953127, "grad_norm": 0.012831724854186178, "loss": 0.6892087459564209, "time": 0.8363796234130859, "epoch": 10, "memory": 28783, "step": 54922} +{"lr": 0.0005824350601949144, "data_time": 0.0023002147674560545, "grad_norm": 0.017541662626899778, "loss": 0.6872913300991058, "time": 0.8115854501724243, "epoch": 10, "memory": 28783, "step": 55022} +{"lr": 0.0005824350601949144, "data_time": 0.0019317865371704102, "grad_norm": 0.01501223393715918, "loss": 0.6866838216781617, "time": 0.7921674251556396, "epoch": 10, "memory": 28783, "step": 55122} +{"lr": 0.0005824350601949144, "data_time": 0.0035938024520874023, "grad_norm": 0.012913845560979098, "loss": 0.6876675605773925, "time": 0.8037017107009887, "epoch": 10, "memory": 28783, "step": 55222} +{"lr": 0.0005824350601949144, "data_time": 0.0023813962936401365, "grad_norm": 0.018609405495226383, "loss": 0.6858887314796448, "time": 0.8376826524734498, "epoch": 10, "memory": 28783, "step": 55322} +{"lr": 0.0005824350601949144, "data_time": 0.0019999980926513673, "grad_norm": 0.014814723422750831, "loss": 0.6853049516677856, "time": 0.856293797492981, "epoch": 10, "memory": 28783, "step": 55422} +{"lr": 0.0005824350601949144, "data_time": 0.0018617630004882813, "grad_norm": 0.0204045329708606, "loss": 0.6878536224365235, "time": 0.80146164894104, "epoch": 10, "memory": 28783, "step": 55522} +{"lr": 0.0005824350601949144, "data_time": 0.002483510971069336, "grad_norm": 0.017967253969982267, "loss": 0.6896375954151154, "time": 0.8275414705276489, "epoch": 10, "memory": 28783, "step": 55622} +{"lr": 0.0005824350601949144, "data_time": 0.00185546875, "grad_norm": 0.016180197877110912, "loss": 0.687091201543808, "time": 0.7865345478057861, "epoch": 10, "memory": 28783, "step": 55722} +{"lr": 0.0005824350601949144, "data_time": 0.0036696910858154295, "grad_norm": 0.016136669309344143, "loss": 0.6830904126167298, "time": 0.8425149917602539, "epoch": 10, "memory": 28783, "step": 55822} +{"lr": 0.0005824350601949144, "data_time": 0.001925492286682129, "grad_norm": 0.013465837901458144, "loss": 0.686905813217163, "time": 0.8077531576156616, "epoch": 10, "memory": 28783, "step": 55922} +{"lr": 0.0005824350601949144, "data_time": 0.0017953634262084961, "grad_norm": 0.019176779454573988, "loss": 0.6883704662322998, "time": 0.8467078447341919, "epoch": 10, "memory": 28783, "step": 56022} +{"lr": 0.0005824350601949144, "data_time": 0.0018522262573242188, "grad_norm": 0.014131482504308224, "loss": 0.686920040845871, "time": 0.8090510606765747, "epoch": 10, "memory": 28783, "step": 56122} +{"lr": 0.0005824350601949144, "data_time": 0.001916980743408203, "grad_norm": 0.021248600841499866, "loss": 0.6902573704719543, "time": 0.7851123809814453, "epoch": 10, "memory": 28783, "step": 56222} +{"lr": 0.0005824350601949144, "data_time": 0.0018211841583251954, "grad_norm": 0.016638199565932155, "loss": 0.6876445651054383, "time": 0.8147528409957886, "epoch": 10, "memory": 28783, "step": 56322} +{"lr": 0.0005824350601949144, "data_time": 0.002654409408569336, "grad_norm": 0.014992894604802132, "loss": 0.6888919591903686, "time": 0.9232783794403077, "epoch": 10, "memory": 28783, "step": 56422} +{"lr": 0.0005824350601949144, "data_time": 0.0023769140243530273, "grad_norm": 0.013957629329524934, "loss": 0.6855066955089569, "time": 0.8823898553848266, "epoch": 10, "memory": 28783, "step": 56522} +{"lr": 0.0005824350601949144, "data_time": 0.0023835420608520506, "grad_norm": 0.016980611742474137, "loss": 0.6868089735507965, "time": 0.8272493839263916, "epoch": 10, "memory": 28783, "step": 56622} +{"lr": 0.0005824350601949144, "data_time": 0.002582216262817383, "grad_norm": 0.01621195066254586, "loss": 0.6860782980918885, "time": 0.840746283531189, "epoch": 10, "memory": 28783, "step": 56722} +{"lr": 0.0005824350601949144, "data_time": 0.14568696022033692, "grad_norm": 0.01863554958254099, "loss": 0.6876372575759888, "time": 0.8938787937164306, "epoch": 10, "memory": 28783, "step": 56822} +{"lr": 0.0005824350601949144, "data_time": 0.23851263523101807, "grad_norm": 0.017496292339637877, "loss": 0.6851976811885834, "time": 0.838429069519043, "epoch": 10, "memory": 28783, "step": 56922} +{"lr": 0.0005824350601949144, "data_time": 0.3127598285675049, "grad_norm": 0.02135424907319248, "loss": 0.6882346212863922, "time": 0.8302545309066772, "epoch": 10, "memory": 28783, "step": 57022} +{"lr": 0.0005824350601949144, "data_time": 0.2952101707458496, "grad_norm": 0.017044686170993372, "loss": 0.6846368968486786, "time": 0.8063157320022583, "epoch": 10, "memory": 28783, "step": 57122} +{"lr": 0.0005824350601949144, "data_time": 0.28017406463623046, "grad_norm": 0.02175892653176561, "loss": 0.6814761459827423, "time": 0.7970028877258301, "epoch": 10, "memory": 28783, "step": 57222} +{"lr": 0.0005824350601949144, "data_time": 0.267191743850708, "grad_norm": 0.018163497699424623, "loss": 0.6877678751945495, "time": 0.7744769811630249, "epoch": 10, "memory": 28783, "step": 57322} +{"lr": 0.0005824350601949144, "data_time": 0.21019601821899414, "grad_norm": 0.02312041134573519, "loss": 0.6849162101745605, "time": 0.7158305406570434, "epoch": 10, "memory": 28783, "step": 57422} +{"lr": 0.0005824350601949144, "data_time": 0.2966966390609741, "grad_norm": 0.019629032735247166, "loss": 0.6894556999206543, "time": 0.802053713798523, "epoch": 10, "memory": 28783, "step": 57522} +{"accuracy/top1": 100.0, "data_time": 0.22202599749845617, "time": 0.6372561595019173, "step": 10} +{"lr": 0.0005050000000000001, "data_time": 0.33390305042266843, "grad_norm": 0.014232703985180705, "loss": 0.6860788226127624, "time": 0.8691173076629639, "epoch": 11, "memory": 28783, "step": 57680} +{"lr": 0.0005050000000000001, "data_time": 0.330348539352417, "grad_norm": 0.014199422171805054, "loss": 0.6860699653625488, "time": 0.8362077236175537, "epoch": 11, "memory": 28783, "step": 57780} +{"lr": 0.0005050000000000001, "data_time": 0.39358646869659425, "grad_norm": 0.022275893250480294, "loss": 0.6843575775623322, "time": 0.9236335515975952, "epoch": 11, "memory": 28783, "step": 57880} +{"lr": 0.0005050000000000001, "data_time": 0.33318612575531004, "grad_norm": 0.016775678703561427, "loss": 0.6894590795040131, "time": 0.8368151426315308, "epoch": 11, "memory": 28783, "step": 57980} +{"lr": 0.0005050000000000001, "data_time": 0.3459513187408447, "grad_norm": 0.012328343483386562, "loss": 0.6849977731704712, "time": 0.8754051446914672, "epoch": 11, "memory": 28783, "step": 58080} +{"lr": 0.0005050000000000001, "data_time": 0.36264605522155763, "grad_norm": 0.01742911710171029, "loss": 0.6863394856452942, "time": 0.8671110153198243, "epoch": 11, "memory": 28783, "step": 58180} +{"lr": 0.0005050000000000001, "data_time": 0.3340389013290405, "grad_norm": 0.01449612674769014, "loss": 0.6846177458763123, "time": 0.8383068561553955, "epoch": 11, "memory": 28783, "step": 58280} +{"lr": 0.0005050000000000001, "data_time": 0.3212372064590454, "grad_norm": 0.015622683428227901, "loss": 0.6831147789955139, "time": 0.829296326637268, "epoch": 11, "memory": 28783, "step": 58380} +{"lr": 0.0005050000000000001, "data_time": 0.3451340675354004, "grad_norm": 0.021475805202499032, "loss": 0.6871937930583953, "time": 0.8493071317672729, "epoch": 11, "memory": 28783, "step": 58480} +{"lr": 0.0005050000000000001, "data_time": 0.3325185775756836, "grad_norm": 0.020584947615861892, "loss": 0.6863519251346588, "time": 0.838594651222229, "epoch": 11, "memory": 28783, "step": 58580} +{"lr": 0.0005050000000000001, "data_time": 0.28668100833892823, "grad_norm": 0.01612954700831324, "loss": 0.6870429158210755, "time": 0.7914071321487427, "epoch": 11, "memory": 28783, "step": 58680} +{"lr": 0.0005050000000000001, "data_time": 0.3148604154586792, "grad_norm": 0.015629613283090295, "loss": 0.6877296268939972, "time": 0.8199707984924316, "epoch": 11, "memory": 28783, "step": 58780} +{"lr": 0.0005050000000000001, "data_time": 0.34616312980651853, "grad_norm": 0.013550182757899164, "loss": 0.6892147719860077, "time": 0.8530388355255127, "epoch": 11, "memory": 28783, "step": 58880} +{"lr": 0.0005050000000000001, "data_time": 0.3644432783126831, "grad_norm": 0.012172738916706293, "loss": 0.6892185509204865, "time": 0.8674351930618286, "epoch": 11, "memory": 28783, "step": 58980} +{"lr": 0.0005050000000000001, "data_time": 0.3425153732299805, "grad_norm": 0.016395441099302844, "loss": 0.6892509460449219, "time": 0.8503780364990234, "epoch": 11, "memory": 28783, "step": 59080} +{"lr": 0.0005050000000000001, "data_time": 0.33985285758972167, "grad_norm": 0.015192930371267722, "loss": 0.6893633425235748, "time": 0.8467146635055542, "epoch": 11, "memory": 28783, "step": 59180} +{"lr": 0.0005050000000000001, "data_time": 0.35523161888122556, "grad_norm": 0.017875791247934102, "loss": 0.6898441016674042, "time": 0.8614187240600586, "epoch": 11, "memory": 28783, "step": 59280} +{"lr": 0.0005050000000000001, "data_time": 0.31883018016815184, "grad_norm": 0.01974796229042113, "loss": 0.6828283607959748, "time": 0.8241447448730469, "epoch": 11, "memory": 28783, "step": 59380} +{"lr": 0.0005050000000000001, "data_time": 0.3458577632904053, "grad_norm": 0.015877533543971367, "loss": 0.6861612796783447, "time": 0.8532354354858398, "epoch": 11, "memory": 28783, "step": 59480} +{"lr": 0.0005050000000000001, "data_time": 0.3423829317092896, "grad_norm": 0.015195346274413168, "loss": 0.687296348810196, "time": 0.847105073928833, "epoch": 11, "memory": 28783, "step": 59580} +{"lr": 0.0005050000000000001, "data_time": 0.3770418643951416, "grad_norm": 0.022456830926239492, "loss": 0.6905148088932037, "time": 0.881967306137085, "epoch": 11, "memory": 28783, "step": 59680} +{"lr": 0.0005050000000000001, "data_time": 0.35122835636138916, "grad_norm": 0.011916636599926278, "loss": 0.6868367969989777, "time": 0.8611812591552734, "epoch": 11, "memory": 28783, "step": 59780} +{"lr": 0.0005050000000000001, "data_time": 0.3091218709945679, "grad_norm": 0.02567956829443574, "loss": 0.6893227696418762, "time": 0.8145798444747925, "epoch": 11, "memory": 28783, "step": 59880} +{"lr": 0.0005050000000000001, "data_time": 0.3292386531829834, "grad_norm": 0.021753088338300586, "loss": 0.6893504381179809, "time": 0.8360827207565308, "epoch": 11, "memory": 28783, "step": 59980} +{"lr": 0.0005050000000000001, "data_time": 0.2831078767776489, "grad_norm": 0.014434442017227411, "loss": 0.6883375585079193, "time": 0.7857233762741089, "epoch": 11, "memory": 28783, "step": 60080} +{"lr": 0.0005050000000000001, "data_time": 0.18909857273101807, "grad_norm": 0.011160801281221211, "loss": 0.6866437137126923, "time": 0.8101491451263427, "epoch": 11, "memory": 28783, "step": 60180} +{"lr": 0.0005050000000000001, "data_time": 0.002041149139404297, "grad_norm": 0.021918557886965572, "loss": 0.6850089490413666, "time": 0.8151986837387085, "epoch": 11, "memory": 28783, "step": 60280} +{"lr": 0.0005050000000000001, "data_time": 0.0020836830139160157, "grad_norm": 0.02017070451984182, "loss": 0.6889029443264008, "time": 0.8374037504196167, "epoch": 11, "memory": 28783, "step": 60380} +{"lr": 0.0005050000000000001, "data_time": 0.0022475004196166994, "grad_norm": 0.014040674548596144, "loss": 0.685858428478241, "time": 0.8345201015472412, "epoch": 11, "memory": 28783, "step": 60480} +{"lr": 0.0005050000000000001, "data_time": 0.002798318862915039, "grad_norm": 0.01217449464602396, "loss": 0.6867355525493621, "time": 0.8692013025283813, "epoch": 11, "memory": 28783, "step": 60580} +{"lr": 0.0005050000000000001, "data_time": 0.0018689393997192382, "grad_norm": 0.018253942718729377, "loss": 0.6899765431880951, "time": 0.833388376235962, "epoch": 11, "memory": 28783, "step": 60680} +{"lr": 0.0005050000000000001, "data_time": 0.002090907096862793, "grad_norm": 0.026523547183023764, "loss": 0.6906438887119293, "time": 0.9125625133514405, "epoch": 11, "memory": 28783, "step": 60780} +{"lr": 0.0005050000000000001, "data_time": 0.0017291784286499023, "grad_norm": 0.017480567819438873, "loss": 0.6886890709400177, "time": 0.8224637031555175, "epoch": 11, "memory": 28783, "step": 60880} +{"lr": 0.0005050000000000001, "data_time": 0.0018859386444091796, "grad_norm": 0.012101862154668196, "loss": 0.6879057168960572, "time": 0.8365171432495118, "epoch": 11, "memory": 28783, "step": 60980} +{"lr": 0.0005050000000000001, "data_time": 0.0019419670104980468, "grad_norm": 0.012853042338974774, "loss": 0.6873636722564698, "time": 0.8697279930114746, "epoch": 11, "memory": 28783, "step": 61080} +{"lr": 0.0005050000000000001, "data_time": 0.002118682861328125, "grad_norm": 0.020067825028672815, "loss": 0.6847352743148803, "time": 0.909968638420105, "epoch": 11, "memory": 28783, "step": 61180} +{"lr": 0.0005050000000000001, "data_time": 0.0018488168716430664, "grad_norm": 0.013973988778889179, "loss": 0.6909814417362213, "time": 0.8352821350097657, "epoch": 11, "memory": 28783, "step": 61280} +{"lr": 0.0005050000000000001, "data_time": 0.0017964363098144532, "grad_norm": 0.00903941192664206, "loss": 0.6856515228748321, "time": 0.8690871715545654, "epoch": 11, "memory": 28783, "step": 61380} +{"lr": 0.0005050000000000001, "data_time": 0.002246546745300293, "grad_norm": 0.016161096398718655, "loss": 0.6874457240104676, "time": 0.8462628602981568, "epoch": 11, "memory": 28783, "step": 61480} +{"lr": 0.0005050000000000001, "data_time": 0.0019682168960571287, "grad_norm": 0.01643331847153604, "loss": 0.6866926908493042, "time": 0.8641710996627807, "epoch": 11, "memory": 28783, "step": 61580} +{"lr": 0.0005050000000000001, "data_time": 0.001953125, "grad_norm": 0.015661456622183323, "loss": 0.684465229511261, "time": 0.8878344297409058, "epoch": 11, "memory": 28783, "step": 61680} +{"lr": 0.0005050000000000001, "data_time": 0.00203557014465332, "grad_norm": 0.01332063355948776, "loss": 0.6864002048969269, "time": 0.8721402645111084, "epoch": 11, "memory": 28783, "step": 61780} +{"lr": 0.0005050000000000001, "data_time": 0.0020078420639038086, "grad_norm": 0.008680033922428266, "loss": 0.6860170960426331, "time": 0.8419245719909668, "epoch": 11, "memory": 28783, "step": 61880} +{"lr": 0.0005050000000000001, "data_time": 0.0018712997436523438, "grad_norm": 0.020153040206059812, "loss": 0.6840706110000611, "time": 0.8748828649520874, "epoch": 11, "memory": 28783, "step": 61980} +{"lr": 0.0005050000000000001, "data_time": 0.0018357515335083007, "grad_norm": 0.006081509188516065, "loss": 0.6893117904663086, "time": 0.823386836051941, "epoch": 11, "memory": 28783, "step": 62080} +{"lr": 0.0005050000000000001, "data_time": 0.0018768310546875, "grad_norm": 0.012463466078042985, "loss": 0.6862398684024811, "time": 0.7977720260620117, "epoch": 11, "memory": 28783, "step": 62180} +{"lr": 0.0005050000000000001, "data_time": 0.0018864870071411133, "grad_norm": 0.011752224061638117, "loss": 0.6866662263870239, "time": 0.8112723588943481, "epoch": 11, "memory": 28783, "step": 62280} +{"lr": 0.0005050000000000001, "data_time": 0.0018591165542602539, "grad_norm": 0.014484515134245157, "loss": 0.6853614449501038, "time": 0.81897132396698, "epoch": 11, "memory": 28783, "step": 62380} +{"lr": 0.0005050000000000001, "data_time": 0.0019700050354003905, "grad_norm": 0.014939905516803265, "loss": 0.6879810571670533, "time": 0.8347316265106202, "epoch": 11, "memory": 28783, "step": 62480} +{"lr": 0.0005050000000000001, "data_time": 0.0016882896423339843, "grad_norm": 0.017923609586432576, "loss": 0.6860221087932586, "time": 0.800758957862854, "epoch": 11, "memory": 28783, "step": 62580} +{"lr": 0.0005050000000000001, "data_time": 0.0018634557723999023, "grad_norm": 0.02136535132303834, "loss": 0.688265037536621, "time": 0.8851779699325562, "epoch": 11, "memory": 28783, "step": 62680} +{"lr": 0.0005050000000000001, "data_time": 0.002018284797668457, "grad_norm": 0.018396999686956406, "loss": 0.687425148487091, "time": 0.7992514610290528, "epoch": 11, "memory": 28783, "step": 62780} +{"lr": 0.0005050000000000001, "data_time": 0.0023943424224853516, "grad_norm": 0.012992601515725255, "loss": 0.6872276544570923, "time": 0.8223165512084961, "epoch": 11, "memory": 28783, "step": 62880} +{"lr": 0.0005050000000000001, "data_time": 0.002091670036315918, "grad_norm": 0.01296411354560405, "loss": 0.6859677195549011, "time": 0.9990283250808716, "epoch": 11, "memory": 28783, "step": 62980} +{"lr": 0.0005050000000000001, "data_time": 0.00213007926940918, "grad_norm": 0.014627908845432103, "loss": 0.6826976597309112, "time": 0.8523520469665528, "epoch": 11, "memory": 28783, "step": 63080} +{"lr": 0.0005050000000000001, "data_time": 0.0019238710403442383, "grad_norm": 0.017776364041492344, "loss": 0.6831452310085296, "time": 0.8890249729156494, "epoch": 11, "memory": 28783, "step": 63180} +{"lr": 0.0005050000000000001, "data_time": 0.001802206039428711, "grad_norm": 0.012028198200277984, "loss": 0.6871235847473145, "time": 0.8503433465957642, "epoch": 11, "memory": 28783, "step": 63280} +{"accuracy/top1": 100.0, "data_time": 0.21847601497874541, "time": 0.6325038741616642, "step": 11} +{"lr": 0.000427564939805086, "data_time": 0.003440999984741211, "grad_norm": 0.012254272896097974, "loss": 0.688452398777008, "time": 0.8018713235855103, "epoch": 12, "memory": 28783, "step": 63438} +{"lr": 0.000427564939805086, "data_time": 0.0027014970779418944, "grad_norm": 0.012671843508724124, "loss": 0.6879237651824951, "time": 0.8426581144332885, "epoch": 12, "memory": 28783, "step": 63538} +{"lr": 0.000427564939805086, "data_time": 0.0020052194595336914, "grad_norm": 0.012314799171872438, "loss": 0.6878165185451508, "time": 0.8584505319595337, "epoch": 12, "memory": 28783, "step": 63638} +{"lr": 0.000427564939805086, "data_time": 0.0015796899795532226, "grad_norm": 0.014028619066812098, "loss": 0.6865696728229522, "time": 0.8230879783630372, "epoch": 12, "memory": 28783, "step": 63738} +{"lr": 0.000427564939805086, "data_time": 0.0019037723541259766, "grad_norm": 0.011870158626697958, "loss": 0.6863193094730378, "time": 0.8921442031860352, "epoch": 12, "memory": 28783, "step": 63838} +{"lr": 0.000427564939805086, "data_time": 0.0021541357040405274, "grad_norm": 0.018259978853166103, "loss": 0.6882329642772674, "time": 0.8551560401916504, "epoch": 12, "memory": 28783, "step": 63938} +{"lr": 0.000427564939805086, "data_time": 0.0022708892822265623, "grad_norm": 0.016573956818319857, "loss": 0.6889025926589966, "time": 0.82143075466156, "epoch": 12, "memory": 28783, "step": 64038} +{"lr": 0.000427564939805086, "data_time": 0.0020264387130737305, "grad_norm": 0.013501499826088548, "loss": 0.6859869599342346, "time": 0.8742319107055664, "epoch": 12, "memory": 28783, "step": 64138} +{"lr": 0.000427564939805086, "data_time": 0.0019479036331176759, "grad_norm": 0.019222495844587684, "loss": 0.684797215461731, "time": 0.8399986982345581, "epoch": 12, "memory": 28783, "step": 64238} +{"lr": 0.000427564939805086, "data_time": 0.0019345760345458984, "grad_norm": 0.01394376807147637, "loss": 0.6886085152626038, "time": 0.9529498100280762, "epoch": 12, "memory": 28783, "step": 64338} +{"lr": 0.000427564939805086, "data_time": 0.0021494150161743162, "grad_norm": 0.019535945355892183, "loss": 0.6855676114559174, "time": 0.7990355014801025, "epoch": 12, "memory": 28783, "step": 64438} +{"lr": 0.000427564939805086, "data_time": 0.001722574234008789, "grad_norm": 0.00962800437118858, "loss": 0.6870296001434326, "time": 0.8601633548736572, "epoch": 12, "memory": 28783, "step": 64538} +{"lr": 0.000427564939805086, "data_time": 0.0017733335494995116, "grad_norm": 0.010389964492060245, "loss": 0.6857517063617706, "time": 0.8852148532867432, "epoch": 12, "memory": 28783, "step": 64638} +{"lr": 0.000427564939805086, "data_time": 0.0022205114364624023, "grad_norm": 0.015625704545527697, "loss": 0.6859215438365937, "time": 0.8484145641326905, "epoch": 12, "memory": 28783, "step": 64738} +{"lr": 0.000427564939805086, "data_time": 0.0019207477569580078, "grad_norm": 0.012254656321601942, "loss": 0.6875371158123016, "time": 0.8932466506958008, "epoch": 12, "memory": 28783, "step": 64838} +{"lr": 0.000427564939805086, "data_time": 0.001924896240234375, "grad_norm": 0.025284463725984096, "loss": 0.6850191652774811, "time": 0.8371482610702514, "epoch": 12, "memory": 28783, "step": 64938} +{"lr": 0.000427564939805086, "data_time": 0.0024759769439697266, "grad_norm": 0.01742397854104638, "loss": 0.6841047823429107, "time": 0.8798106908798218, "epoch": 12, "memory": 28783, "step": 65038} +{"lr": 0.000427564939805086, "data_time": 0.0018042325973510742, "grad_norm": 0.018168100458569824, "loss": 0.6866636395454406, "time": 0.840888237953186, "epoch": 12, "memory": 28783, "step": 65138} +{"lr": 0.000427564939805086, "data_time": 0.0020921707153320314, "grad_norm": 0.01569122385699302, "loss": 0.6895307660102844, "time": 0.8279127359390259, "epoch": 12, "memory": 28783, "step": 65238} +{"lr": 0.000427564939805086, "data_time": 0.0025150537490844726, "grad_norm": 0.02255082824267447, "loss": 0.6879225075244904, "time": 0.8909866333007812, "epoch": 12, "memory": 28783, "step": 65338} +{"lr": 0.000427564939805086, "data_time": 0.0018845796585083008, "grad_norm": 0.01949697150848806, "loss": 0.6868943929672241, "time": 0.8667906522750854, "epoch": 12, "memory": 28783, "step": 65438} +{"lr": 0.000427564939805086, "data_time": 0.002279186248779297, "grad_norm": 0.016585860983468592, "loss": 0.6881010830402374, "time": 0.8440192699432373, "epoch": 12, "memory": 28783, "step": 65538} +{"lr": 0.000427564939805086, "data_time": 0.0022717714309692383, "grad_norm": 0.022135975118726493, "loss": 0.6877662181854248, "time": 0.8422539234161377, "epoch": 12, "memory": 28783, "step": 65638} +{"lr": 0.000427564939805086, "data_time": 0.0018958330154418945, "grad_norm": 0.015082068298943341, "loss": 0.6859381020069122, "time": 0.8344122171401978, "epoch": 12, "memory": 28783, "step": 65738} +{"lr": 0.000427564939805086, "data_time": 0.001839280128479004, "grad_norm": 0.017079555289819838, "loss": 0.6860392153263092, "time": 0.8058897733688355, "epoch": 12, "memory": 28783, "step": 65838} +{"lr": 0.000427564939805086, "data_time": 0.0016717910766601562, "grad_norm": 0.010059173277113587, "loss": 0.6839419901371002, "time": 0.8589228868484498, "epoch": 12, "memory": 28783, "step": 65938} +{"lr": 0.000427564939805086, "data_time": 0.0015815973281860351, "grad_norm": 0.01895637606503442, "loss": 0.687496793270111, "time": 0.8400229454040528, "epoch": 12, "memory": 28783, "step": 66038} +{"lr": 0.000427564939805086, "data_time": 0.001855325698852539, "grad_norm": 0.017460882826708258, "loss": 0.6851868987083435, "time": 0.839257001876831, "epoch": 12, "memory": 28783, "step": 66138} +{"lr": 0.000427564939805086, "data_time": 0.0017693042755126953, "grad_norm": 0.0143052568892017, "loss": 0.6830495893955231, "time": 0.8398881673812866, "epoch": 12, "memory": 28783, "step": 66238} +{"lr": 0.000427564939805086, "data_time": 0.0019068956375122071, "grad_norm": 0.013714489690028132, "loss": 0.6875060558319092, "time": 0.8520808696746827, "epoch": 12, "memory": 28783, "step": 66338} +{"lr": 0.000427564939805086, "data_time": 0.0022751331329345704, "grad_norm": 0.015660790947731585, "loss": 0.6870501816272736, "time": 0.8310092687606812, "epoch": 12, "memory": 28783, "step": 66438} +{"lr": 0.000427564939805086, "data_time": 0.001861906051635742, "grad_norm": 0.015771122719161212, "loss": 0.6893124282360077, "time": 0.8851603507995606, "epoch": 12, "memory": 28783, "step": 66538} +{"lr": 0.000427564939805086, "data_time": 0.002104663848876953, "grad_norm": 0.01973067447543144, "loss": 0.6862702906131745, "time": 0.8477134704589844, "epoch": 12, "memory": 28783, "step": 66638} +{"lr": 0.000427564939805086, "data_time": 0.002760505676269531, "grad_norm": 0.017841457203030587, "loss": 0.6864173471927643, "time": 0.8686709880828858, "epoch": 12, "memory": 28783, "step": 66738} +{"lr": 0.000427564939805086, "data_time": 0.001983499526977539, "grad_norm": 0.013224598742090166, "loss": 0.6861699461936951, "time": 0.8156566143035888, "epoch": 12, "memory": 28783, "step": 66838} +{"lr": 0.000427564939805086, "data_time": 0.0026387929916381835, "grad_norm": 0.012288924073800445, "loss": 0.6868104100227356, "time": 0.846895170211792, "epoch": 12, "memory": 28783, "step": 66938} +{"lr": 0.000427564939805086, "data_time": 0.0017224550247192383, "grad_norm": 0.018949965585488827, "loss": 0.6843415439128876, "time": 0.8430649518966675, "epoch": 12, "memory": 28783, "step": 67038} +{"lr": 0.000427564939805086, "data_time": 0.0018741369247436523, "grad_norm": 0.016846169158816336, "loss": 0.6841327607631683, "time": 0.8603096246719361, "epoch": 12, "memory": 28783, "step": 67138} +{"lr": 0.000427564939805086, "data_time": 0.0016807317733764648, "grad_norm": 0.017878101440146565, "loss": 0.687576973438263, "time": 0.8255111455917359, "epoch": 12, "memory": 28783, "step": 67238} +{"lr": 0.000427564939805086, "data_time": 0.0020295381546020508, "grad_norm": 0.016764623462222517, "loss": 0.6865106284618377, "time": 0.9006153821945191, "epoch": 12, "memory": 28783, "step": 67338} +{"lr": 0.000427564939805086, "data_time": 0.0020602941513061523, "grad_norm": 0.0188767671585083, "loss": 0.6833829045295715, "time": 0.8561848878860474, "epoch": 12, "memory": 28783, "step": 67438} +{"lr": 0.000427564939805086, "data_time": 0.0020209789276123048, "grad_norm": 0.022730698401574045, "loss": 0.6891467928886413, "time": 0.8356692314147949, "epoch": 12, "memory": 28783, "step": 67538} +{"lr": 0.000427564939805086, "data_time": 0.0021651268005371095, "grad_norm": 0.013563672988675534, "loss": 0.6873798906803131, "time": 0.8504010677337647, "epoch": 12, "memory": 28783, "step": 67638} +{"lr": 0.000427564939805086, "data_time": 0.002049112319946289, "grad_norm": 0.018280728533864022, "loss": 0.6884930729866028, "time": 0.8261601686477661, "epoch": 12, "memory": 28783, "step": 67738} +{"lr": 0.000427564939805086, "data_time": 0.001821422576904297, "grad_norm": 0.016631564171984792, "loss": 0.6844547271728516, "time": 0.8422627210617065, "epoch": 12, "memory": 28783, "step": 67838} +{"lr": 0.000427564939805086, "data_time": 0.00184478759765625, "grad_norm": 0.018905861396342517, "loss": 0.6824126243591309, "time": 0.8438386201858521, "epoch": 12, "memory": 28783, "step": 67938} +{"lr": 0.000427564939805086, "data_time": 0.001901102066040039, "grad_norm": 0.01797306233784184, "loss": 0.6894153237342835, "time": 0.8111754179000854, "epoch": 12, "memory": 28783, "step": 68038} +{"lr": 0.000427564939805086, "data_time": 0.002787184715270996, "grad_norm": 0.02484312830492854, "loss": 0.6873823583126069, "time": 0.822121262550354, "epoch": 12, "memory": 28783, "step": 68138} +{"lr": 0.000427564939805086, "data_time": 0.0019865512847900392, "grad_norm": 0.02294129424262792, "loss": 0.6850340187549591, "time": 0.8418996334075928, "epoch": 12, "memory": 28783, "step": 68238} +{"lr": 0.000427564939805086, "data_time": 0.0019071578979492187, "grad_norm": 0.012277470075059683, "loss": 0.6879194438457489, "time": 0.8444482803344726, "epoch": 12, "memory": 28783, "step": 68338} +{"lr": 0.000427564939805086, "data_time": 0.002357935905456543, "grad_norm": 0.022407754976302385, "loss": 0.6852332472801208, "time": 0.8816360712051392, "epoch": 12, "memory": 28783, "step": 68438} +{"lr": 0.000427564939805086, "data_time": 0.002547550201416016, "grad_norm": 0.016252647247165443, "loss": 0.6869350075721741, "time": 0.8663325309753418, "epoch": 12, "memory": 28783, "step": 68538} +{"lr": 0.000427564939805086, "data_time": 0.0017745494842529297, "grad_norm": 0.012211003992706537, "loss": 0.6905552208423614, "time": 0.8967301368713378, "epoch": 12, "memory": 28783, "step": 68638} +{"lr": 0.000427564939805086, "data_time": 0.0021529197692871094, "grad_norm": 0.02307587442919612, "loss": 0.6870291054248809, "time": 0.8886263847351075, "epoch": 12, "memory": 28783, "step": 68738} +{"lr": 0.000427564939805086, "data_time": 0.0018274545669555663, "grad_norm": 0.01625542587134987, "loss": 0.6875106155872345, "time": 0.8600252628326416, "epoch": 12, "memory": 28783, "step": 68838} +{"lr": 0.000427564939805086, "data_time": 0.0021867275238037108, "grad_norm": 0.011149125744123013, "loss": 0.6883181154727935, "time": 0.8382859468460083, "epoch": 12, "memory": 28783, "step": 68938} +{"lr": 0.000427564939805086, "data_time": 0.0025548458099365233, "grad_norm": 0.018422122637275606, "loss": 0.6870428025722504, "time": 0.8882803678512573, "epoch": 12, "memory": 28783, "step": 69038} +{"accuracy/top1": 100.0, "data_time": 0.20626535135156968, "time": 0.6214081539827234, "step": 12} +{"lr": 0.0003520365877844012, "data_time": 0.2391516923904419, "grad_norm": 0.02247819406911731, "loss": 0.6879157781600952, "time": 0.8461340188980102, "epoch": 13, "memory": 28783, "step": 69196} +{"lr": 0.0003520365877844012, "data_time": 0.0018452882766723632, "grad_norm": 0.012166666146367789, "loss": 0.687698382139206, "time": 0.8045434713363647, "epoch": 13, "memory": 28783, "step": 69296} +{"lr": 0.0003520365877844012, "data_time": 0.0023088932037353517, "grad_norm": 0.019380286114756017, "loss": 0.6865039825439453, "time": 0.8118725061416626, "epoch": 13, "memory": 28783, "step": 69396} +{"lr": 0.0003520365877844012, "data_time": 0.003086972236633301, "grad_norm": 0.008478107361588627, "loss": 0.686888438463211, "time": 0.8681352376937866, "epoch": 13, "memory": 28783, "step": 69496} +{"lr": 0.0003520365877844012, "data_time": 0.0018507003784179687, "grad_norm": 0.015004638629034162, "loss": 0.685306864976883, "time": 0.863008451461792, "epoch": 13, "memory": 28783, "step": 69596} +{"lr": 0.0003520365877844012, "data_time": 0.0024507522583007814, "grad_norm": 0.019598237005993724, "loss": 0.6859407603740693, "time": 0.8286648988723755, "epoch": 13, "memory": 28783, "step": 69696} +{"lr": 0.0003520365877844012, "data_time": 0.0020800352096557615, "grad_norm": 0.018289242358878254, "loss": 0.6875308990478516, "time": 0.8506077527999878, "epoch": 13, "memory": 28783, "step": 69796} +{"lr": 0.0003520365877844012, "data_time": 0.0018233776092529297, "grad_norm": 0.013454599329270423, "loss": 0.684500253200531, "time": 0.8005702257156372, "epoch": 13, "memory": 28783, "step": 69896} +{"lr": 0.0003520365877844012, "data_time": 0.0031012535095214845, "grad_norm": 0.017343352455645798, "loss": 0.6883727550506592, "time": 0.8688923835754394, "epoch": 13, "memory": 28783, "step": 69996} +{"lr": 0.0003520365877844012, "data_time": 0.002783656120300293, "grad_norm": 0.010464380035409704, "loss": 0.6860755383968353, "time": 0.8936461687088013, "epoch": 13, "memory": 28783, "step": 70096} +{"lr": 0.0003520365877844012, "data_time": 0.002584958076477051, "grad_norm": 0.013679419201798737, "loss": 0.6870516419410706, "time": 0.865241527557373, "epoch": 13, "memory": 28783, "step": 70196} +{"lr": 0.0003520365877844012, "data_time": 0.0024837493896484376, "grad_norm": 0.01400958439335227, "loss": 0.6888535916805267, "time": 0.8799507141113281, "epoch": 13, "memory": 28783, "step": 70296} +{"lr": 0.0003520365877844012, "data_time": 0.0017689228057861327, "grad_norm": 0.019060352473752575, "loss": 0.6873691499233245, "time": 0.8606380939483642, "epoch": 13, "memory": 28783, "step": 70396} +{"lr": 0.0003520365877844012, "data_time": 0.001884794235229492, "grad_norm": 0.013557263580150903, "loss": 0.6864943146705628, "time": 0.8788732528686524, "epoch": 13, "memory": 28783, "step": 70496} +{"lr": 0.0003520365877844012, "data_time": 0.0027837038040161135, "grad_norm": 0.017442819301504642, "loss": 0.6879061222076416, "time": 0.8228353261947632, "epoch": 13, "memory": 28783, "step": 70596} +{"lr": 0.0003520365877844012, "data_time": 0.0019437313079833985, "grad_norm": 0.015117300814017654, "loss": 0.6872159004211426, "time": 0.8581367254257202, "epoch": 13, "memory": 28783, "step": 70696} +{"lr": 0.0003520365877844012, "data_time": 0.0028244733810424806, "grad_norm": 0.015494049922563136, "loss": 0.6887699007987976, "time": 0.9177424907684326, "epoch": 13, "memory": 28783, "step": 70796} +{"lr": 0.0003520365877844012, "data_time": 0.002637290954589844, "grad_norm": 0.016117434320040046, "loss": 0.6872215270996094, "time": 0.8286547660827637, "epoch": 13, "memory": 28783, "step": 70896} +{"lr": 0.0003520365877844012, "data_time": 0.0019123077392578125, "grad_norm": 0.01466503927949816, "loss": 0.6878536522388459, "time": 0.8192996263504029, "epoch": 13, "memory": 28783, "step": 70996} +{"lr": 0.0003520365877844012, "data_time": 0.002655959129333496, "grad_norm": 0.010116344201378524, "loss": 0.6876698732376099, "time": 0.8364684581756592, "epoch": 13, "memory": 28783, "step": 71096} +{"lr": 0.0003520365877844012, "data_time": 0.0020856142044067385, "grad_norm": 0.013932437263429165, "loss": 0.6877418458461761, "time": 0.7948732852935791, "epoch": 13, "memory": 28783, "step": 71196} +{"lr": 0.0003520365877844012, "data_time": 0.002506852149963379, "grad_norm": 0.02352831920143217, "loss": 0.6889092564582825, "time": 0.8805351734161377, "epoch": 13, "memory": 28783, "step": 71296} +{"lr": 0.0003520365877844012, "data_time": 0.0020704746246337892, "grad_norm": 0.018580973194912077, "loss": 0.6872140407562256, "time": 0.9044097661972046, "epoch": 13, "memory": 28783, "step": 71396} +{"lr": 0.0003520365877844012, "data_time": 0.003917074203491211, "grad_norm": 0.01591803561896086, "loss": 0.6890011310577393, "time": 0.8946885347366333, "epoch": 13, "memory": 28783, "step": 71496} +{"lr": 0.0003520365877844012, "data_time": 0.0026073694229125977, "grad_norm": 0.015283292019739747, "loss": 0.6860139489173889, "time": 0.8426134347915649, "epoch": 13, "memory": 28783, "step": 71596} +{"lr": 0.0003520365877844012, "data_time": 0.0028566360473632813, "grad_norm": 0.023452968802303076, "loss": 0.6867613911628723, "time": 0.8604297161102294, "epoch": 13, "memory": 28783, "step": 71696} +{"lr": 0.0003520365877844012, "data_time": 0.0025577306747436523, "grad_norm": 0.027375939395278694, "loss": 0.6850374102592468, "time": 0.8654968023300171, "epoch": 13, "memory": 28783, "step": 71796} +{"lr": 0.0003520365877844012, "data_time": 0.00260317325592041, "grad_norm": 0.01553009677445516, "loss": 0.6822042286396026, "time": 0.8916721820831299, "epoch": 13, "memory": 28783, "step": 71896} +{"lr": 0.0003520365877844012, "data_time": 0.0030301094055175783, "grad_norm": 0.009857999958330765, "loss": 0.683708256483078, "time": 0.841990613937378, "epoch": 13, "memory": 28783, "step": 71996} +{"lr": 0.0003520365877844012, "data_time": 0.001949000358581543, "grad_norm": 0.01007278913166374, "loss": 0.6851643741130828, "time": 0.8914402246475219, "epoch": 13, "memory": 28783, "step": 72096} +{"lr": 0.0003520365877844012, "data_time": 0.0021855831146240234, "grad_norm": 0.01911925037857145, "loss": 0.6894653737545013, "time": 0.8655587196350097, "epoch": 13, "memory": 28783, "step": 72196} +{"lr": 0.0003520365877844012, "data_time": 0.0020829439163208008, "grad_norm": 0.014841298246756196, "loss": 0.6904347121715546, "time": 0.9711669445037842, "epoch": 13, "memory": 28783, "step": 72296} +{"lr": 0.0003520365877844012, "data_time": 0.002265143394470215, "grad_norm": 0.02255669248988852, "loss": 0.6879143357276917, "time": 0.8268091201782226, "epoch": 13, "memory": 28783, "step": 72396} +{"lr": 0.0003520365877844012, "data_time": 0.002348661422729492, "grad_norm": 0.025222789077088238, "loss": 0.6871301174163819, "time": 0.8799536943435669, "epoch": 13, "memory": 28783, "step": 72496} +{"lr": 0.0003520365877844012, "data_time": 0.003020763397216797, "grad_norm": 0.018251256039366125, "loss": 0.6879877865314483, "time": 0.8517894983291626, "epoch": 13, "memory": 28783, "step": 72596} +{"lr": 0.0003520365877844012, "data_time": 0.0033486127853393556, "grad_norm": 0.015391803072998301, "loss": 0.6872004687786102, "time": 0.9024513006210327, "epoch": 13, "memory": 28783, "step": 72696} +{"lr": 0.0003520365877844012, "data_time": 0.002414250373840332, "grad_norm": 0.013337351003428921, "loss": 0.6876972556114197, "time": 0.8562380313873291, "epoch": 13, "memory": 28783, "step": 72796} +{"lr": 0.0003520365877844012, "data_time": 0.0018981456756591796, "grad_norm": 0.012450885854195803, "loss": 0.6886140763759613, "time": 0.851210069656372, "epoch": 13, "memory": 28783, "step": 72896} +{"lr": 0.0003520365877844012, "data_time": 0.002394366264343262, "grad_norm": 0.01784762986935675, "loss": 0.6905607402324676, "time": 0.8503563880920411, "epoch": 13, "memory": 28783, "step": 72996} +{"lr": 0.0003520365877844012, "data_time": 0.0029362916946411135, "grad_norm": 0.022130879014730452, "loss": 0.6833449304103851, "time": 0.8617544174194336, "epoch": 13, "memory": 28783, "step": 73096} +{"lr": 0.0003520365877844012, "data_time": 0.0031109094619750977, "grad_norm": 0.014414389850571752, "loss": 0.6847840309143066, "time": 0.8836791276931762, "epoch": 13, "memory": 28783, "step": 73196} +{"lr": 0.0003520365877844012, "data_time": 0.0033122539520263673, "grad_norm": 0.013657961692661046, "loss": 0.6869725584983826, "time": 0.8541029691696167, "epoch": 13, "memory": 28783, "step": 73296} +{"lr": 0.0003520365877844012, "data_time": 0.0024121284484863283, "grad_norm": 0.012511292798444628, "loss": 0.6887347280979157, "time": 0.8270941019058228, "epoch": 13, "memory": 28783, "step": 73396} +{"lr": 0.0003520365877844012, "data_time": 0.0024880170822143555, "grad_norm": 0.012370800855569542, "loss": 0.6899017512798309, "time": 0.8426279067993164, "epoch": 13, "memory": 28783, "step": 73496} +{"lr": 0.0003520365877844012, "data_time": 0.0025377273559570312, "grad_norm": 0.018190123606473207, "loss": 0.6852566301822662, "time": 0.8379014730453491, "epoch": 13, "memory": 28783, "step": 73596} +{"lr": 0.0003520365877844012, "data_time": 0.0025491952896118165, "grad_norm": 0.012276675552129745, "loss": 0.681931871175766, "time": 0.7893479585647583, "epoch": 13, "memory": 28783, "step": 73696} +{"lr": 0.0003520365877844012, "data_time": 0.0026648759841918944, "grad_norm": 0.023400061391294004, "loss": 0.6810989201068878, "time": 0.8150842189788818, "epoch": 13, "memory": 28783, "step": 73796} +{"lr": 0.0003520365877844012, "data_time": 0.0023877620697021484, "grad_norm": 0.012619656487368047, "loss": 0.6848697185516357, "time": 0.824254059791565, "epoch": 13, "memory": 28783, "step": 73896} +{"lr": 0.0003520365877844012, "data_time": 0.001892852783203125, "grad_norm": 0.01216881915461272, "loss": 0.6850947201251983, "time": 0.8250591278076171, "epoch": 13, "memory": 28783, "step": 73996} +{"lr": 0.0003520365877844012, "data_time": 0.0020848512649536133, "grad_norm": 0.0169877891195938, "loss": 0.6879559636116028, "time": 0.8067756414413452, "epoch": 13, "memory": 28783, "step": 74096} +{"lr": 0.0003520365877844012, "data_time": 0.0026565313339233397, "grad_norm": 0.013792618177831173, "loss": 0.6902757346630096, "time": 0.8280647039413452, "epoch": 13, "memory": 28783, "step": 74196} +{"lr": 0.0003520365877844012, "data_time": 0.0025008201599121095, "grad_norm": 0.015572900662664323, "loss": 0.6899118900299073, "time": 0.7984892368316651, "epoch": 13, "memory": 28783, "step": 74296} +{"lr": 0.0003520365877844012, "data_time": 0.0020769834518432617, "grad_norm": 0.018312862515449523, "loss": 0.6861032009124756, "time": 0.8125770092010498, "epoch": 13, "memory": 28783, "step": 74396} +{"lr": 0.0003520365877844012, "data_time": 0.0021264076232910155, "grad_norm": 0.01774419692810625, "loss": 0.6882382690906524, "time": 0.787130880355835, "epoch": 13, "memory": 28783, "step": 74496} +{"lr": 0.0003520365877844012, "data_time": 0.0022942304611206056, "grad_norm": 0.015399366314522923, "loss": 0.6872778475284577, "time": 0.7920415163040161, "epoch": 13, "memory": 28783, "step": 74596} +{"lr": 0.0003520365877844012, "data_time": 0.0026883602142333983, "grad_norm": 0.01367082130163908, "loss": 0.6894431829452514, "time": 0.806365180015564, "epoch": 13, "memory": 28783, "step": 74696} +{"lr": 0.0003520365877844012, "data_time": 0.0024976491928100585, "grad_norm": 0.01867692113155499, "loss": 0.6883374631404877, "time": 0.7716233491897583, "epoch": 13, "memory": 28783, "step": 74796} +{"accuracy/top1": 100.0, "data_time": 0.21666373926050522, "time": 0.6321520244373995, "step": 13} +{"lr": 0.00028027470262892445, "data_time": 0.0031208038330078126, "grad_norm": 0.021379628300201146, "loss": 0.6845173120498658, "time": 0.766830039024353, "epoch": 14, "memory": 28783, "step": 74954} +{"lr": 0.00028027470262892445, "data_time": 0.18351433277130128, "grad_norm": 0.012256648496259005, "loss": 0.6863982260227204, "time": 0.7944447278976441, "epoch": 14, "memory": 28783, "step": 75054} +{"lr": 0.00028027470262892445, "data_time": 0.0025827169418334963, "grad_norm": 0.011965835932642221, "loss": 0.6864938914775849, "time": 0.8210030555725097, "epoch": 14, "memory": 28783, "step": 75154} +{"lr": 0.00028027470262892445, "data_time": 0.0021210193634033205, "grad_norm": 0.015844987309537827, "loss": 0.6851883113384247, "time": 0.8173990964889526, "epoch": 14, "memory": 28783, "step": 75254} +{"lr": 0.00028027470262892445, "data_time": 0.002375888824462891, "grad_norm": 0.025583311403170227, "loss": 0.6866584420204163, "time": 0.9005350828170776, "epoch": 14, "memory": 28783, "step": 75354} +{"lr": 0.00028027470262892445, "data_time": 0.0028691768646240236, "grad_norm": 0.010602948523592204, "loss": 0.6881591320037842, "time": 0.7948050498962402, "epoch": 14, "memory": 28783, "step": 75454} +{"lr": 0.00028027470262892445, "data_time": 0.002345418930053711, "grad_norm": 0.013583542476408184, "loss": 0.6897511720657349, "time": 0.8507160902023315, "epoch": 14, "memory": 28783, "step": 75554} +{"lr": 0.00028027470262892445, "data_time": 0.0028682947158813477, "grad_norm": 0.01781910303980112, "loss": 0.6882295072078705, "time": 0.8167087316513062, "epoch": 14, "memory": 28783, "step": 75654} +{"lr": 0.00028027470262892445, "data_time": 0.002935957908630371, "grad_norm": 0.01549930670298636, "loss": 0.6884795784950256, "time": 0.7837977409362793, "epoch": 14, "memory": 28783, "step": 75754} +{"lr": 0.00028027470262892445, "data_time": 0.003167080879211426, "grad_norm": 0.014664636831730605, "loss": 0.6862571835517883, "time": 0.8247785568237305, "epoch": 14, "memory": 28783, "step": 75854} +{"lr": 0.00028027470262892445, "data_time": 0.0021590232849121095, "grad_norm": 0.014382861135527492, "loss": 0.6890288352966308, "time": 0.7866461992263794, "epoch": 14, "memory": 28783, "step": 75954} +{"lr": 0.00028027470262892445, "data_time": 0.002756071090698242, "grad_norm": 0.017164668021723627, "loss": 0.6865702033042907, "time": 0.8042606592178345, "epoch": 14, "memory": 28783, "step": 76054} +{"lr": 0.00028027470262892445, "data_time": 0.0023998737335205076, "grad_norm": 0.015464434633031487, "loss": 0.6834678173065185, "time": 0.8166307210922241, "epoch": 14, "memory": 28783, "step": 76154} +{"lr": 0.00028027470262892445, "data_time": 0.002602553367614746, "grad_norm": 0.014378064661286771, "loss": 0.6863162696361542, "time": 0.8312625408172607, "epoch": 14, "memory": 28783, "step": 76254} +{"lr": 0.00028027470262892445, "data_time": 0.0025655508041381838, "grad_norm": 0.00882388096069917, "loss": 0.6859928727149963, "time": 0.8546101331710816, "epoch": 14, "memory": 28783, "step": 76354} +{"lr": 0.00028027470262892445, "data_time": 0.0028191566467285155, "grad_norm": 0.010516448388807475, "loss": 0.687013590335846, "time": 0.8058027029037476, "epoch": 14, "memory": 28783, "step": 76454} +{"lr": 0.00028027470262892445, "data_time": 0.002307438850402832, "grad_norm": 0.022224119922611862, "loss": 0.6882230103015899, "time": 0.7847615718841553, "epoch": 14, "memory": 28783, "step": 76554} +{"lr": 0.00028027470262892445, "data_time": 0.0023659706115722657, "grad_norm": 0.011203650292009116, "loss": 0.6866700649261475, "time": 0.8896780729293823, "epoch": 14, "memory": 28783, "step": 76654} +{"lr": 0.00028027470262892445, "data_time": 0.002873349189758301, "grad_norm": 0.012446122476831079, "loss": 0.6870320022106171, "time": 0.7789308786392212, "epoch": 14, "memory": 28783, "step": 76754} +{"lr": 0.00028027470262892445, "data_time": 0.0033449888229370116, "grad_norm": 0.01665814840234816, "loss": 0.6875148057937622, "time": 0.8135859727859497, "epoch": 14, "memory": 28783, "step": 76854} +{"lr": 0.00028027470262892445, "data_time": 0.002987241744995117, "grad_norm": 0.017333590518683197, "loss": 0.6875333726406098, "time": 0.8344608306884765, "epoch": 14, "memory": 28783, "step": 76954} +{"lr": 0.00028027470262892445, "data_time": 0.002769279479980469, "grad_norm": 0.014409756287932396, "loss": 0.6865765690803528, "time": 0.8360893964767456, "epoch": 14, "memory": 28783, "step": 77054} +{"lr": 0.00028027470262892445, "data_time": 0.0028784513473510743, "grad_norm": 0.013263238349463791, "loss": 0.6874877333641052, "time": 0.7898921728134155, "epoch": 14, "memory": 28783, "step": 77154} +{"lr": 0.00028027470262892445, "data_time": 0.0021254777908325194, "grad_norm": 0.02061710082925856, "loss": 0.6844447910785675, "time": 0.777776575088501, "epoch": 14, "memory": 28783, "step": 77254} +{"lr": 0.00028027470262892445, "data_time": 0.0028898000717163088, "grad_norm": 0.019925792654976248, "loss": 0.6862314343452454, "time": 0.8320996999740601, "epoch": 14, "memory": 28783, "step": 77354} +{"lr": 0.00028027470262892445, "data_time": 0.0022061824798583984, "grad_norm": 0.01716822008602321, "loss": 0.684758996963501, "time": 0.784383225440979, "epoch": 14, "memory": 28783, "step": 77454} +{"lr": 0.00028027470262892445, "data_time": 0.0030763626098632814, "grad_norm": 0.01281963336514309, "loss": 0.683683580160141, "time": 0.7999552249908447, "epoch": 14, "memory": 28783, "step": 77554} +{"lr": 0.00028027470262892445, "data_time": 0.0020072460174560547, "grad_norm": 0.008046269562328234, "loss": 0.6877181529998779, "time": 0.7850032567977905, "epoch": 14, "memory": 28783, "step": 77654} +{"lr": 0.00028027470262892445, "data_time": 0.002471613883972168, "grad_norm": 0.018460680561838673, "loss": 0.6869895458221436, "time": 0.8241334915161133, "epoch": 14, "memory": 28783, "step": 77754} +{"lr": 0.00028027470262892445, "data_time": 0.002103853225708008, "grad_norm": 0.01441455683670938, "loss": 0.6888548135757446, "time": 0.878797173500061, "epoch": 14, "memory": 28783, "step": 77854} +{"lr": 0.00028027470262892445, "data_time": 0.0020147323608398437, "grad_norm": 0.021186025463975966, "loss": 0.6893752098083497, "time": 0.8197164535522461, "epoch": 14, "memory": 28783, "step": 77954} +{"lr": 0.00028027470262892445, "data_time": 0.002424931526184082, "grad_norm": 0.015030666696839034, "loss": 0.6881118655204773, "time": 0.8350815296173095, "epoch": 14, "memory": 28783, "step": 78054} +{"lr": 0.00028027470262892445, "data_time": 0.0022409439086914064, "grad_norm": 0.016399089235346765, "loss": 0.686488401889801, "time": 0.826201319694519, "epoch": 14, "memory": 28783, "step": 78154} +{"lr": 0.00028027470262892445, "data_time": 0.001822662353515625, "grad_norm": 0.013706484372960404, "loss": 0.6877138614654541, "time": 0.8358863592147827, "epoch": 14, "memory": 28783, "step": 78254} +{"lr": 0.00028027470262892445, "data_time": 0.0022010087966918947, "grad_norm": 0.012685577967204154, "loss": 0.6851814568042756, "time": 0.8375303268432617, "epoch": 14, "memory": 28783, "step": 78354} +{"lr": 0.00028027470262892445, "data_time": 0.001901698112487793, "grad_norm": 0.008980188029818236, "loss": 0.6860692381858826, "time": 0.8216911554336548, "epoch": 14, "memory": 28783, "step": 78454} +{"lr": 0.00028027470262892445, "data_time": 0.0020236968994140625, "grad_norm": 0.012190910009667277, "loss": 0.6853796422481537, "time": 0.8030661344528198, "epoch": 14, "memory": 28783, "step": 78554} +{"lr": 0.00028027470262892445, "data_time": 0.0025640249252319334, "grad_norm": 0.014230663341004401, "loss": 0.6871629476547241, "time": 0.7951892852783203, "epoch": 14, "memory": 28783, "step": 78654} +{"lr": 0.00028027470262892445, "data_time": 0.002682042121887207, "grad_norm": 0.017096467316150665, "loss": 0.6860884428024292, "time": 0.8175258636474609, "epoch": 14, "memory": 28783, "step": 78754} +{"lr": 0.00028027470262892445, "data_time": 0.0017179489135742188, "grad_norm": 0.00833444639283698, "loss": 0.6911397516727448, "time": 0.7847953557968139, "epoch": 14, "memory": 28783, "step": 78854} +{"lr": 0.00028027470262892445, "data_time": 0.0019270658493041992, "grad_norm": 0.011716116871684789, "loss": 0.688162910938263, "time": 0.7925777435302734, "epoch": 14, "memory": 28783, "step": 78954} +{"lr": 0.00028027470262892445, "data_time": 0.002113795280456543, "grad_norm": 0.008377769030630589, "loss": 0.6892408430576324, "time": 0.8271990060806275, "epoch": 14, "memory": 28783, "step": 79054} +{"lr": 0.00028027470262892445, "data_time": 0.0022131681442260744, "grad_norm": 0.02468175757676363, "loss": 0.6864872097969055, "time": 0.8087864875793457, "epoch": 14, "memory": 28783, "step": 79154} +{"lr": 0.00028027470262892445, "data_time": 0.001732206344604492, "grad_norm": 0.016062717139720916, "loss": 0.6868020832538605, "time": 0.8304159879684448, "epoch": 14, "memory": 28783, "step": 79254} +{"lr": 0.00028027470262892445, "data_time": 0.0020791053771972655, "grad_norm": 0.01238457476720214, "loss": 0.6881404280662536, "time": 0.8129947900772094, "epoch": 14, "memory": 28783, "step": 79354} +{"lr": 0.00028027470262892445, "data_time": 0.0028953790664672852, "grad_norm": 0.020267345802858473, "loss": 0.6883187830448151, "time": 0.8191087722778321, "epoch": 14, "memory": 28783, "step": 79454} +{"lr": 0.00028027470262892445, "data_time": 0.0019062042236328125, "grad_norm": 0.02627642937004566, "loss": 0.6887010395526886, "time": 0.8152916908264161, "epoch": 14, "memory": 28783, "step": 79554} +{"lr": 0.00028027470262892445, "data_time": 0.0030180692672729494, "grad_norm": 0.015312521567102522, "loss": 0.6873525321483612, "time": 0.7950273513793945, "epoch": 14, "memory": 28783, "step": 79654} +{"lr": 0.00028027470262892445, "data_time": 0.003254342079162598, "grad_norm": 0.01467336919158697, "loss": 0.68640296459198, "time": 0.8026775360107422, "epoch": 14, "memory": 28783, "step": 79754} +{"lr": 0.00028027470262892445, "data_time": 0.0019814014434814454, "grad_norm": 0.016030821396270767, "loss": 0.6836517930030823, "time": 0.8852525472640991, "epoch": 14, "memory": 28783, "step": 79854} +{"lr": 0.00028027470262892445, "data_time": 0.0019116163253784179, "grad_norm": 0.01687209411757067, "loss": 0.68647780418396, "time": 0.8558833599090576, "epoch": 14, "memory": 28783, "step": 79954} +{"lr": 0.00028027470262892445, "data_time": 0.0029361486434936524, "grad_norm": 0.019384603830985725, "loss": 0.6887512147426605, "time": 0.8053285598754882, "epoch": 14, "memory": 28783, "step": 80054} +{"lr": 0.00028027470262892445, "data_time": 0.0022010087966918947, "grad_norm": 0.019917760509997606, "loss": 0.686985844373703, "time": 0.8282535552978516, "epoch": 14, "memory": 28783, "step": 80154} +{"lr": 0.00028027470262892445, "data_time": 0.0020809412002563477, "grad_norm": 0.017564585572108625, "loss": 0.6845010399818421, "time": 0.9100208282470703, "epoch": 14, "memory": 28783, "step": 80254} +{"lr": 0.00028027470262892445, "data_time": 0.001949167251586914, "grad_norm": 0.015602843632223085, "loss": 0.6869618237018585, "time": 0.7959365844726562, "epoch": 14, "memory": 28783, "step": 80354} +{"lr": 0.00028027470262892445, "data_time": 0.0029601573944091795, "grad_norm": 0.014760750695131719, "loss": 0.6897834122180939, "time": 0.8478416681289673, "epoch": 14, "memory": 28783, "step": 80454} +{"lr": 0.00028027470262892445, "data_time": 0.003617119789123535, "grad_norm": 0.013280410110019148, "loss": 0.6880097806453704, "time": 0.793059229850769, "epoch": 14, "memory": 28783, "step": 80554} +{"accuracy/top1": 100.0, "data_time": 0.22324672867270076, "time": 0.6495649534113267, "step": 14} +{"lr": 0.0002140463001152259, "data_time": 0.002576136589050293, "grad_norm": 0.01352698476985097, "loss": 0.6871983647346497, "time": 0.765565824508667, "epoch": 15, "memory": 28783, "step": 80712} +{"lr": 0.0002140463001152259, "data_time": 0.0019073963165283203, "grad_norm": 0.017645997437648477, "loss": 0.6848306834697724, "time": 0.7879814386367798, "epoch": 15, "memory": 28783, "step": 80812} +{"lr": 0.0002140463001152259, "data_time": 0.001973462104797363, "grad_norm": 0.017264298698864877, "loss": 0.6913569450378418, "time": 0.831602144241333, "epoch": 15, "memory": 28783, "step": 80912} +{"lr": 0.0002140463001152259, "data_time": 0.0021908760070800783, "grad_norm": 0.016802679211832584, "loss": 0.6873147785663605, "time": 0.9419013500213623, "epoch": 15, "memory": 28783, "step": 81012} +{"lr": 0.0002140463001152259, "data_time": 0.0018620252609252929, "grad_norm": 0.015621217398438603, "loss": 0.6823112905025482, "time": 0.8435795783996582, "epoch": 15, "memory": 28783, "step": 81112} +{"lr": 0.0002140463001152259, "data_time": 0.002634286880493164, "grad_norm": 0.022311592614278196, "loss": 0.6869670987129212, "time": 0.8085411071777344, "epoch": 15, "memory": 28783, "step": 81212} +{"lr": 0.0002140463001152259, "data_time": 0.002593684196472168, "grad_norm": 0.019172173563856632, "loss": 0.6874598681926727, "time": 0.8183466911315918, "epoch": 15, "memory": 28783, "step": 81312} +{"lr": 0.0002140463001152259, "data_time": 0.0022911310195922853, "grad_norm": 0.01431084774667397, "loss": 0.6900889873504639, "time": 0.8187406063079834, "epoch": 15, "memory": 28783, "step": 81412} +{"lr": 0.0002140463001152259, "data_time": 0.002336883544921875, "grad_norm": 0.018483986612409353, "loss": 0.6866706728935241, "time": 0.8031017065048218, "epoch": 15, "memory": 28783, "step": 81512} +{"lr": 0.0002140463001152259, "data_time": 0.0019074201583862305, "grad_norm": 0.013430599425919353, "loss": 0.6870581150054932, "time": 0.8650548458099365, "epoch": 15, "memory": 28783, "step": 81612} +{"lr": 0.0002140463001152259, "data_time": 0.002767324447631836, "grad_norm": 0.023504098667763172, "loss": 0.6874674141407013, "time": 0.8802296876907348, "epoch": 15, "memory": 28783, "step": 81712} +{"lr": 0.0002140463001152259, "data_time": 0.002679324150085449, "grad_norm": 0.016823733667843045, "loss": 0.688474464416504, "time": 0.8106569290161133, "epoch": 15, "memory": 28783, "step": 81812} +{"lr": 0.0002140463001152259, "data_time": 0.002176046371459961, "grad_norm": 0.013821120385546237, "loss": 0.68550985455513, "time": 0.8387659788131714, "epoch": 15, "memory": 28783, "step": 81912} +{"lr": 0.0002140463001152259, "data_time": 0.002002310752868652, "grad_norm": 0.014627343835309147, "loss": 0.6831127405166626, "time": 0.8324304580688476, "epoch": 15, "memory": 28783, "step": 82012} +{"lr": 0.0002140463001152259, "data_time": 0.0021606922149658204, "grad_norm": 0.018662805599160493, "loss": 0.6869138181209564, "time": 0.8171991348266602, "epoch": 15, "memory": 28783, "step": 82112} +{"lr": 0.0002140463001152259, "data_time": 0.0020258188247680663, "grad_norm": 0.02056035230634734, "loss": 0.6865976035594941, "time": 0.8381967306137085, "epoch": 15, "memory": 28783, "step": 82212} +{"lr": 0.0002140463001152259, "data_time": 0.0026579380035400392, "grad_norm": 0.024374304851517082, "loss": 0.6868169367313385, "time": 0.8150118827819824, "epoch": 15, "memory": 28783, "step": 82312} +{"lr": 0.0002140463001152259, "data_time": 0.002124333381652832, "grad_norm": 0.014691986329853534, "loss": 0.6864885449409485, "time": 0.8040873050689697, "epoch": 15, "memory": 28783, "step": 82412} +{"lr": 0.0002140463001152259, "data_time": 0.0026203393936157227, "grad_norm": 0.017796302586793898, "loss": 0.6849068462848663, "time": 0.8241257190704345, "epoch": 15, "memory": 28783, "step": 82512} +{"lr": 0.0002140463001152259, "data_time": 0.001891016960144043, "grad_norm": 0.012056048377417028, "loss": 0.6855287492275238, "time": 0.7964436531066894, "epoch": 15, "memory": 28783, "step": 82612} +{"lr": 0.0002140463001152259, "data_time": 0.002974200248718262, "grad_norm": 0.020515192253515124, "loss": 0.6836224734783173, "time": 0.831844711303711, "epoch": 15, "memory": 28783, "step": 82712} +{"lr": 0.0002140463001152259, "data_time": 0.00243992805480957, "grad_norm": 0.015027505811303854, "loss": 0.6869742274284363, "time": 0.8170264959335327, "epoch": 15, "memory": 28783, "step": 82812} +{"lr": 0.0002140463001152259, "data_time": 0.0020360231399536135, "grad_norm": 0.010840017691953108, "loss": 0.6866565525531769, "time": 0.8417579650878906, "epoch": 15, "memory": 28783, "step": 82912} +{"lr": 0.0002140463001152259, "data_time": 0.00180509090423584, "grad_norm": 0.01052761425380595, "loss": 0.6853402435779572, "time": 0.8383221626281738, "epoch": 15, "memory": 28783, "step": 83012} +{"lr": 0.0002140463001152259, "data_time": 0.0038327455520629885, "grad_norm": 0.01995938392356038, "loss": 0.6869740664958954, "time": 0.8452216386795044, "epoch": 15, "memory": 28783, "step": 83112} +{"lr": 0.0002140463001152259, "data_time": 0.0024830102920532227, "grad_norm": 0.021384344017133117, "loss": 0.6879705846309662, "time": 0.8323727369308471, "epoch": 15, "memory": 28783, "step": 83212} +{"lr": 0.0002140463001152259, "data_time": 0.002127265930175781, "grad_norm": 0.017593488725833595, "loss": 0.6860734760761261, "time": 0.8288000345230102, "epoch": 15, "memory": 28783, "step": 83312} +{"lr": 0.0002140463001152259, "data_time": 0.00223696231842041, "grad_norm": 0.009376094792969525, "loss": 0.6862518072128296, "time": 0.7986794948577881, "epoch": 15, "memory": 28783, "step": 83412} +{"lr": 0.0002140463001152259, "data_time": 0.002404952049255371, "grad_norm": 0.014639827655628324, "loss": 0.6834733366966248, "time": 0.7947625637054443, "epoch": 15, "memory": 28783, "step": 83512} +{"lr": 0.0002140463001152259, "data_time": 0.001947021484375, "grad_norm": 0.014395470917224883, "loss": 0.6898046553134918, "time": 0.8145194530487061, "epoch": 15, "memory": 28783, "step": 83612} +{"lr": 0.0002140463001152259, "data_time": 0.0019422531127929687, "grad_norm": 0.016355888824909924, "loss": 0.6870259881019593, "time": 0.7987046480178833, "epoch": 15, "memory": 28783, "step": 83712} +{"lr": 0.0002140463001152259, "data_time": 0.0018370866775512696, "grad_norm": 0.011732210795162245, "loss": 0.6852873206138611, "time": 0.8341565608978272, "epoch": 15, "memory": 28783, "step": 83812} +{"lr": 0.0002140463001152259, "data_time": 0.0018714189529418946, "grad_norm": 0.02023541498929262, "loss": 0.6886974394321441, "time": 0.8202132463455201, "epoch": 15, "memory": 28783, "step": 83912} +{"lr": 0.0002140463001152259, "data_time": 0.003199911117553711, "grad_norm": 0.01677515732590109, "loss": 0.6890270471572876, "time": 0.866722846031189, "epoch": 15, "memory": 28783, "step": 84012} +{"lr": 0.0002140463001152259, "data_time": 0.001996612548828125, "grad_norm": 0.019989075209014117, "loss": 0.6904678165912628, "time": 0.7882216930389404, "epoch": 15, "memory": 28783, "step": 84112} +{"lr": 0.0002140463001152259, "data_time": 0.0018927335739135742, "grad_norm": 0.02138390401378274, "loss": 0.6856461286544799, "time": 0.802038311958313, "epoch": 15, "memory": 28783, "step": 84212} +{"lr": 0.0002140463001152259, "data_time": 0.002139759063720703, "grad_norm": 0.016336982301436365, "loss": 0.6845249235630035, "time": 0.8265966892242431, "epoch": 15, "memory": 28783, "step": 84312} +{"lr": 0.0002140463001152259, "data_time": 0.002336001396179199, "grad_norm": 0.011610458232462406, "loss": 0.6858574390411377, "time": 0.8583698511123657, "epoch": 15, "memory": 28783, "step": 84412} +{"lr": 0.0002140463001152259, "data_time": 0.0029568195343017576, "grad_norm": 0.015989073656965048, "loss": 0.6879020512104035, "time": 0.8427110195159913, "epoch": 15, "memory": 28783, "step": 84512} +{"lr": 0.0002140463001152259, "data_time": 0.0027734041213989258, "grad_norm": 0.01457352500874549, "loss": 0.6869529604911804, "time": 0.8642256259918213, "epoch": 15, "memory": 28783, "step": 84612} +{"lr": 0.0002140463001152259, "data_time": 0.003186631202697754, "grad_norm": 0.01722420649603009, "loss": 0.688180935382843, "time": 0.8754186630249023, "epoch": 15, "memory": 28783, "step": 84712} +{"lr": 0.0002140463001152259, "data_time": 0.0019004106521606444, "grad_norm": 0.015163661120459438, "loss": 0.687413090467453, "time": 0.8870687007904052, "epoch": 15, "memory": 28783, "step": 84812} +{"lr": 0.0002140463001152259, "data_time": 0.0019695043563842775, "grad_norm": 0.01540433221962303, "loss": 0.6843005776405334, "time": 0.8672104835510254, "epoch": 15, "memory": 28783, "step": 84912} +{"lr": 0.0002140463001152259, "data_time": 0.0018759489059448243, "grad_norm": 0.02686847122386098, "loss": 0.686324667930603, "time": 0.8680312871932984, "epoch": 15, "memory": 28783, "step": 85012} +{"lr": 0.0002140463001152259, "data_time": 0.0021885395050048827, "grad_norm": 0.015108896105084568, "loss": 0.6933867156505584, "time": 0.8587390899658203, "epoch": 15, "memory": 28783, "step": 85112} +{"lr": 0.0002140463001152259, "data_time": 0.0026365995407104494, "grad_norm": 0.017090910207480194, "loss": 0.6857710361480713, "time": 0.7982375144958496, "epoch": 15, "memory": 28783, "step": 85212} +{"lr": 0.0002140463001152259, "data_time": 0.002439737319946289, "grad_norm": 0.02905628685839474, "loss": 0.6862540483474732, "time": 0.8567112207412719, "epoch": 15, "memory": 28783, "step": 85312} +{"lr": 0.0002140463001152259, "data_time": 0.0019452571868896484, "grad_norm": 0.01718382416293025, "loss": 0.6862488448619842, "time": 0.8861680507659913, "epoch": 15, "memory": 28783, "step": 85412} +{"lr": 0.0002140463001152259, "data_time": 0.0026148319244384765, "grad_norm": 0.013659542216919363, "loss": 0.6856607317924499, "time": 0.920865511894226, "epoch": 15, "memory": 28783, "step": 85512} +{"lr": 0.0002140463001152259, "data_time": 0.0019452571868896484, "grad_norm": 0.013395216770004482, "loss": 0.6882145464420318, "time": 0.8700824499130249, "epoch": 15, "memory": 28783, "step": 85612} +{"lr": 0.0002140463001152259, "data_time": 0.0024453878402709963, "grad_norm": 0.011258493200875819, "loss": 0.6888556718826294, "time": 0.9163310050964355, "epoch": 15, "memory": 28783, "step": 85712} +{"lr": 0.0002140463001152259, "data_time": 0.0019266843795776368, "grad_norm": 0.021566955652087928, "loss": 0.6857811272144317, "time": 0.9290709495544434, "epoch": 15, "memory": 28783, "step": 85812} +{"lr": 0.0002140463001152259, "data_time": 0.0024388313293457033, "grad_norm": 0.009459848949336446, "loss": 0.6900098919868469, "time": 0.8792054414749145, "epoch": 15, "memory": 28783, "step": 85912} +{"lr": 0.0002140463001152259, "data_time": 0.0020087480545043944, "grad_norm": 0.01949253408238292, "loss": 0.6852916657924653, "time": 0.8105745553970337, "epoch": 15, "memory": 28783, "step": 86012} +{"lr": 0.0002140463001152259, "data_time": 0.00248410701751709, "grad_norm": 0.010425873077474534, "loss": 0.6898028314113617, "time": 0.7689057350158691, "epoch": 15, "memory": 28783, "step": 86112} +{"lr": 0.0002140463001152259, "data_time": 0.002334284782409668, "grad_norm": 0.01594096914632246, "loss": 0.6871099531650543, "time": 0.8460749626159668, "epoch": 15, "memory": 28783, "step": 86212} +{"lr": 0.0002140463001152259, "data_time": 0.002018570899963379, "grad_norm": 0.01481670760549605, "loss": 0.6843600749969483, "time": 0.8235304117202759, "epoch": 15, "memory": 28783, "step": 86312} +{"accuracy/top1": 100.0, "data_time": 0.23661115590263815, "time": 0.6512091580559226, "step": 15} +{"lr": 0.00015498214331265904, "data_time": 0.22658779621124267, "grad_norm": 0.016640614450443535, "loss": 0.6875964224338531, "time": 0.8516465902328492, "epoch": 16, "memory": 28783, "step": 86470} +{"lr": 0.00015498214331265904, "data_time": 0.0023043155670166016, "grad_norm": 0.01285829917760566, "loss": 0.6883208692073822, "time": 0.8204551219940186, "epoch": 16, "memory": 28783, "step": 86570} +{"lr": 0.00015498214331265904, "data_time": 0.002229595184326172, "grad_norm": 0.014896116405725478, "loss": 0.6885277032852173, "time": 0.8687238931655884, "epoch": 16, "memory": 28783, "step": 86670} +{"lr": 0.00015498214331265904, "data_time": 0.002166485786437988, "grad_norm": 0.01971497352933511, "loss": 0.6858066618442535, "time": 0.8202325105667114, "epoch": 16, "memory": 28783, "step": 86770} +{"lr": 0.00015498214331265904, "data_time": 0.0019934892654418944, "grad_norm": 0.016891338932327925, "loss": 0.6843840420246124, "time": 0.8294904708862305, "epoch": 16, "memory": 28783, "step": 86870} +{"lr": 0.00015498214331265904, "data_time": 0.0018459558486938477, "grad_norm": 0.019779964530607687, "loss": 0.6868058502674103, "time": 0.765126085281372, "epoch": 16, "memory": 28783, "step": 86970} +{"lr": 0.00015498214331265904, "data_time": 0.0023764848709106447, "grad_norm": 0.013541481946595013, "loss": 0.6889892041683197, "time": 0.8584175348281861, "epoch": 16, "memory": 28783, "step": 87070} +{"lr": 0.00015498214331265904, "data_time": 0.003218936920166016, "grad_norm": 0.025307168252766132, "loss": 0.6882204055786133, "time": 0.7579382181167602, "epoch": 16, "memory": 28783, "step": 87170} +{"lr": 0.00015498214331265904, "data_time": 0.00200347900390625, "grad_norm": 0.012366747169289738, "loss": 0.6888687908649445, "time": 0.8440059900283814, "epoch": 16, "memory": 28783, "step": 87270} +{"lr": 0.00015498214331265904, "data_time": 0.0028717041015625, "grad_norm": 0.015069715020945295, "loss": 0.6897814571857452, "time": 0.7684584140777588, "epoch": 16, "memory": 28783, "step": 87370} +{"lr": 0.00015498214331265904, "data_time": 0.0020277738571166993, "grad_norm": 0.01779211309039965, "loss": 0.6848843812942504, "time": 0.7331821918487549, "epoch": 16, "memory": 28783, "step": 87470} +{"lr": 0.00015498214331265904, "data_time": 0.0019935846328735353, "grad_norm": 0.011437329440377652, "loss": 0.6870453238487244, "time": 0.7904813528060913, "epoch": 16, "memory": 28783, "step": 87570} +{"lr": 0.00015498214331265904, "data_time": 0.0022403478622436525, "grad_norm": 0.013743928447365761, "loss": 0.6884121537208557, "time": 0.7858628511428833, "epoch": 16, "memory": 28783, "step": 87670} +{"lr": 0.00015498214331265904, "data_time": 0.0020228147506713865, "grad_norm": 0.02309629893861711, "loss": 0.6865698575973511, "time": 0.7413806200027466, "epoch": 16, "memory": 28783, "step": 87770} +{"lr": 0.00015498214331265904, "data_time": 0.001766061782836914, "grad_norm": 0.020602161018177866, "loss": 0.6873650968074798, "time": 0.7455927610397339, "epoch": 16, "memory": 28783, "step": 87870} +{"lr": 0.00015498214331265904, "data_time": 0.0018593311309814454, "grad_norm": 0.01645621273200959, "loss": 0.6879251003265381, "time": 0.8247413873672486, "epoch": 16, "memory": 28783, "step": 87970} +{"lr": 0.00015498214331265904, "data_time": 0.0019395828247070312, "grad_norm": 0.012563132867217063, "loss": 0.6880704820156097, "time": 0.7488123416900635, "epoch": 16, "memory": 28783, "step": 88070} +{"lr": 0.00015498214331265904, "data_time": 0.003758525848388672, "grad_norm": 0.015504688181681558, "loss": 0.6865750610828399, "time": 0.8109008312225342, "epoch": 16, "memory": 28783, "step": 88170} +{"lr": 0.00015498214331265904, "data_time": 0.0017815351486206055, "grad_norm": 0.017982418416067958, "loss": 0.6867218971252441, "time": 0.8089564800262451, "epoch": 16, "memory": 28783, "step": 88270} diff --git a/swin_small_4xb256_1e-3lr_5m/20230531_183239/vis_data/vis_image/0000000.png_1.png b/swin_small_4xb256_1e-3lr_5m/20230531_183239/vis_data/vis_image/0000000.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..934769c1f2bb2402694533d16138c8dbc1ce4a97 Binary files /dev/null and b/swin_small_4xb256_1e-3lr_5m/20230531_183239/vis_data/vis_image/0000000.png_1.png differ diff --git a/swin_small_4xb256_1e-3lr_5m/20230531_183239/vis_data/vis_image/0000000.png_10.png b/swin_small_4xb256_1e-3lr_5m/20230531_183239/vis_data/vis_image/0000000.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..b1d5db9b91db71ddcbbba2e61273dd6141d70f3d Binary files /dev/null and b/swin_small_4xb256_1e-3lr_5m/20230531_183239/vis_data/vis_image/0000000.png_10.png differ diff --git a/swin_small_4xb256_1e-3lr_5m/20230531_183239/vis_data/vis_image/0000000.png_11.png b/swin_small_4xb256_1e-3lr_5m/20230531_183239/vis_data/vis_image/0000000.png_11.png new file mode 100644 index 0000000000000000000000000000000000000000..934769c1f2bb2402694533d16138c8dbc1ce4a97 Binary files /dev/null and b/swin_small_4xb256_1e-3lr_5m/20230531_183239/vis_data/vis_image/0000000.png_11.png differ diff --git a/swin_small_4xb256_1e-3lr_5m/20230531_183239/vis_data/vis_image/0000000.png_12.png b/swin_small_4xb256_1e-3lr_5m/20230531_183239/vis_data/vis_image/0000000.png_12.png new file mode 100644 index 0000000000000000000000000000000000000000..934769c1f2bb2402694533d16138c8dbc1ce4a97 Binary files /dev/null and b/swin_small_4xb256_1e-3lr_5m/20230531_183239/vis_data/vis_image/0000000.png_12.png differ diff --git a/swin_small_4xb256_1e-3lr_5m/20230531_183239/vis_data/vis_image/0000000.png_13.png b/swin_small_4xb256_1e-3lr_5m/20230531_183239/vis_data/vis_image/0000000.png_13.png new file mode 100644 index 0000000000000000000000000000000000000000..934769c1f2bb2402694533d16138c8dbc1ce4a97 Binary files /dev/null and b/swin_small_4xb256_1e-3lr_5m/20230531_183239/vis_data/vis_image/0000000.png_13.png differ diff --git a/swin_small_4xb256_1e-3lr_5m/20230531_183239/vis_data/vis_image/0000000.png_14.png b/swin_small_4xb256_1e-3lr_5m/20230531_183239/vis_data/vis_image/0000000.png_14.png new file mode 100644 index 0000000000000000000000000000000000000000..934769c1f2bb2402694533d16138c8dbc1ce4a97 Binary files /dev/null and b/swin_small_4xb256_1e-3lr_5m/20230531_183239/vis_data/vis_image/0000000.png_14.png differ diff --git a/swin_small_4xb256_1e-3lr_5m/20230531_183239/vis_data/vis_image/0000000.png_15.png b/swin_small_4xb256_1e-3lr_5m/20230531_183239/vis_data/vis_image/0000000.png_15.png new file mode 100644 index 0000000000000000000000000000000000000000..934769c1f2bb2402694533d16138c8dbc1ce4a97 Binary files /dev/null and b/swin_small_4xb256_1e-3lr_5m/20230531_183239/vis_data/vis_image/0000000.png_15.png differ diff --git a/swin_small_4xb256_1e-3lr_5m/20230531_183239/vis_data/vis_image/0000000.png_2.png b/swin_small_4xb256_1e-3lr_5m/20230531_183239/vis_data/vis_image/0000000.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..b1d5db9b91db71ddcbbba2e61273dd6141d70f3d Binary files /dev/null and b/swin_small_4xb256_1e-3lr_5m/20230531_183239/vis_data/vis_image/0000000.png_2.png differ diff --git a/swin_small_4xb256_1e-3lr_5m/20230531_183239/vis_data/vis_image/0000000.png_3.png b/swin_small_4xb256_1e-3lr_5m/20230531_183239/vis_data/vis_image/0000000.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..934769c1f2bb2402694533d16138c8dbc1ce4a97 Binary files /dev/null and b/swin_small_4xb256_1e-3lr_5m/20230531_183239/vis_data/vis_image/0000000.png_3.png differ diff --git a/swin_small_4xb256_1e-3lr_5m/20230531_183239/vis_data/vis_image/0000000.png_4.png b/swin_small_4xb256_1e-3lr_5m/20230531_183239/vis_data/vis_image/0000000.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..934769c1f2bb2402694533d16138c8dbc1ce4a97 Binary files /dev/null and b/swin_small_4xb256_1e-3lr_5m/20230531_183239/vis_data/vis_image/0000000.png_4.png differ diff --git a/swin_small_4xb256_1e-3lr_5m/20230531_183239/vis_data/vis_image/0000000.png_5.png b/swin_small_4xb256_1e-3lr_5m/20230531_183239/vis_data/vis_image/0000000.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..934769c1f2bb2402694533d16138c8dbc1ce4a97 Binary files /dev/null and b/swin_small_4xb256_1e-3lr_5m/20230531_183239/vis_data/vis_image/0000000.png_5.png differ diff --git a/swin_small_4xb256_1e-3lr_5m/20230531_183239/vis_data/vis_image/0000000.png_6.png b/swin_small_4xb256_1e-3lr_5m/20230531_183239/vis_data/vis_image/0000000.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..934769c1f2bb2402694533d16138c8dbc1ce4a97 Binary files /dev/null and b/swin_small_4xb256_1e-3lr_5m/20230531_183239/vis_data/vis_image/0000000.png_6.png differ diff --git a/swin_small_4xb256_1e-3lr_5m/20230531_183239/vis_data/vis_image/0000000.png_7.png b/swin_small_4xb256_1e-3lr_5m/20230531_183239/vis_data/vis_image/0000000.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..934769c1f2bb2402694533d16138c8dbc1ce4a97 Binary files /dev/null and b/swin_small_4xb256_1e-3lr_5m/20230531_183239/vis_data/vis_image/0000000.png_7.png differ diff --git a/swin_small_4xb256_1e-3lr_5m/20230531_183239/vis_data/vis_image/0000000.png_8.png b/swin_small_4xb256_1e-3lr_5m/20230531_183239/vis_data/vis_image/0000000.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..934769c1f2bb2402694533d16138c8dbc1ce4a97 Binary files /dev/null and b/swin_small_4xb256_1e-3lr_5m/20230531_183239/vis_data/vis_image/0000000.png_8.png differ diff --git a/swin_small_4xb256_1e-3lr_5m/20230531_183239/vis_data/vis_image/0000000.png_9.png b/swin_small_4xb256_1e-3lr_5m/20230531_183239/vis_data/vis_image/0000000.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..b1d5db9b91db71ddcbbba2e61273dd6141d70f3d Binary files /dev/null and b/swin_small_4xb256_1e-3lr_5m/20230531_183239/vis_data/vis_image/0000000.png_9.png differ diff --git a/swin_small_4xb256_1e-3lr_5m/epoch_1.pth b/swin_small_4xb256_1e-3lr_5m/epoch_1.pth new file mode 100644 index 0000000000000000000000000000000000000000..9560fae064d2db4fe21f358248cfe8e25118b77a --- /dev/null +++ b/swin_small_4xb256_1e-3lr_5m/epoch_1.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:1dafeb4ecaefdb0628e2ebb511d9fb48e922b9b12d58b34304a43aaf98b01b43 +size 587498333 diff --git a/swin_small_4xb256_1e-3lr_5m/epoch_10.pth b/swin_small_4xb256_1e-3lr_5m/epoch_10.pth new file mode 100644 index 0000000000000000000000000000000000000000..72a050cbf024900d199687c8ab068c5c05dbcd12 --- /dev/null +++ b/swin_small_4xb256_1e-3lr_5m/epoch_10.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:91517097313b11e0f91958ba83b9596ff801e9321b5fae14af77840b83572b38 +size 592202461 diff --git a/swin_small_4xb256_1e-3lr_5m/epoch_11.pth b/swin_small_4xb256_1e-3lr_5m/epoch_11.pth new file mode 100644 index 0000000000000000000000000000000000000000..13c55deba1709838153d793b564d2e5ad96e42b1 --- /dev/null +++ b/swin_small_4xb256_1e-3lr_5m/epoch_11.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:f13228c19c72372355b96ef2b4f8b45b01b4322422001ba0efb64d00ce7e3433 +size 592718045 diff --git a/swin_small_4xb256_1e-3lr_5m/epoch_12.pth b/swin_small_4xb256_1e-3lr_5m/epoch_12.pth new file mode 100644 index 0000000000000000000000000000000000000000..5b7c391fd8a48ca8e08921570c5885ed4fa8257a --- /dev/null +++ b/swin_small_4xb256_1e-3lr_5m/epoch_12.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:a479687e2537e049171a9e6a4a3179c6b767b86ceda44241260f7225537b9b8f +size 593239325 diff --git a/swin_small_4xb256_1e-3lr_5m/epoch_13.pth b/swin_small_4xb256_1e-3lr_5m/epoch_13.pth new file mode 100644 index 0000000000000000000000000000000000000000..774e4a7ed744ba28d7b3023ee701d15f724ff6bb --- /dev/null +++ b/swin_small_4xb256_1e-3lr_5m/epoch_13.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:897c5614463928da684d205b0bcfc103c766dcf9ef9f6125d8c233bfa6080502 +size 593767197 diff --git a/swin_small_4xb256_1e-3lr_5m/epoch_14.pth b/swin_small_4xb256_1e-3lr_5m/epoch_14.pth new file mode 100644 index 0000000000000000000000000000000000000000..d0a423a055d708753c94eb7a13a861f29a9c57b8 --- /dev/null +++ b/swin_small_4xb256_1e-3lr_5m/epoch_14.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:660b02d7d43f330b74dabf6fb9ed373636e55c2ea6c343ecd6ebad0877d145b9 +size 594283549 diff --git a/swin_small_4xb256_1e-3lr_5m/epoch_15.pth b/swin_small_4xb256_1e-3lr_5m/epoch_15.pth new file mode 100644 index 0000000000000000000000000000000000000000..0c4c28e9c7f229352f357976665b467f12a5573d --- /dev/null +++ b/swin_small_4xb256_1e-3lr_5m/epoch_15.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:d80ef2a264b12215d2256c49dd54d8841b7bd6f9e426d5fbeb1e5969425eb0ea +size 594811037 diff --git a/swin_small_4xb256_1e-3lr_5m/epoch_2.pth b/swin_small_4xb256_1e-3lr_5m/epoch_2.pth new file mode 100644 index 0000000000000000000000000000000000000000..feaa192647cebd63ab9ac74c91ea167cf87f6732 --- /dev/null +++ b/swin_small_4xb256_1e-3lr_5m/epoch_2.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:dbf465e847b682b8d10ecf2477eb804d53ce89bd4a991f88c12de2f382f63516 +size 588011037 diff --git a/swin_small_4xb256_1e-3lr_5m/epoch_3.pth b/swin_small_4xb256_1e-3lr_5m/epoch_3.pth new file mode 100644 index 0000000000000000000000000000000000000000..be7ac1b999c824cd5c3b4c9113eeb69fe169f429 --- /dev/null +++ b/swin_small_4xb256_1e-3lr_5m/epoch_3.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:1d33d5d5194728b812a4eababac7bedaf2c9e0957d55ba698400e2c6dac66c1a +size 588534109 diff --git a/swin_small_4xb256_1e-3lr_5m/epoch_4.pth b/swin_small_4xb256_1e-3lr_5m/epoch_4.pth new file mode 100644 index 0000000000000000000000000000000000000000..aa7e8c4d03a682cce21f54eca6c5f5e2727632a3 --- /dev/null +++ b/swin_small_4xb256_1e-3lr_5m/epoch_4.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:aa9041511a0119c095b74cd2319396b67e5d2d6b3792df644a4094577b8d4c57 +size 589044253 diff --git a/swin_small_4xb256_1e-3lr_5m/epoch_5.pth b/swin_small_4xb256_1e-3lr_5m/epoch_5.pth new file mode 100644 index 0000000000000000000000000000000000000000..fd1202a30381a3ae759369a11bd28446cd4c5eb4 --- /dev/null +++ b/swin_small_4xb256_1e-3lr_5m/epoch_5.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:9113cc8e3629b522b288c1c1354a25f957ad3f0d55314d1380b7b648153f8b92 +size 589571293 diff --git a/swin_small_4xb256_1e-3lr_5m/epoch_6.pth b/swin_small_4xb256_1e-3lr_5m/epoch_6.pth new file mode 100644 index 0000000000000000000000000000000000000000..e3f22f20b6e3c70d8f4e07b802da2d1313e215ec --- /dev/null +++ b/swin_small_4xb256_1e-3lr_5m/epoch_6.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:7e539222f89c9ac65743d808b3f4bf7423806c2e0bd815740525cc862a148fa2 +size 590093021 diff --git a/swin_small_4xb256_1e-3lr_5m/epoch_7.pth b/swin_small_4xb256_1e-3lr_5m/epoch_7.pth new file mode 100644 index 0000000000000000000000000000000000000000..1cdbdee116fbeb50089bdaf572bc5cb261c4735b --- /dev/null +++ b/swin_small_4xb256_1e-3lr_5m/epoch_7.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:492840d75a86aa60344d3abb6db4b2b82e8e06c47d7df36984b04b3ab90e4144 +size 590620509 diff --git a/swin_small_4xb256_1e-3lr_5m/epoch_8.pth b/swin_small_4xb256_1e-3lr_5m/epoch_8.pth new file mode 100644 index 0000000000000000000000000000000000000000..92b41f12146be7371ab01f1f89d327eb3fa0174b --- /dev/null +++ b/swin_small_4xb256_1e-3lr_5m/epoch_8.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c51e34cdc3944fcc9ceec727add526c3ca3c39f832d2e94d8ec4b871e85948a2 +size 591141661 diff --git a/swin_small_4xb256_1e-3lr_5m/epoch_9.pth b/swin_small_4xb256_1e-3lr_5m/epoch_9.pth new file mode 100644 index 0000000000000000000000000000000000000000..3930240cdd646795187075b4b044ed7c794f44de --- /dev/null +++ b/swin_small_4xb256_1e-3lr_5m/epoch_9.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:0043c88d453ed65b45b44daea0a3d69ea8dbfa268122d7836c1c1332594260f5 +size 591669341 diff --git a/swin_small_4xb256_1e-3lr_5m/last_checkpoint b/swin_small_4xb256_1e-3lr_5m/last_checkpoint new file mode 100644 index 0000000000000000000000000000000000000000..258d8700b828d1d9711657728bdcb5ac555c25e9 --- /dev/null +++ b/swin_small_4xb256_1e-3lr_5m/last_checkpoint @@ -0,0 +1 @@ +/mnt/petrelfs/luzeyu/workspace/fakebench/mmpretrain/workdir/swin_small_4xb256_1e-3lr_5m/epoch_15.pth \ No newline at end of file diff --git a/swin_small_4xb256_1e-3lr_5m/swin_small_4xb256_fake5m.py b/swin_small_4xb256_1e-3lr_5m/swin_small_4xb256_fake5m.py new file mode 100644 index 0000000000000000000000000000000000000000..96ad3949be10f7f7b050950edab63a0911fe12b4 --- /dev/null +++ b/swin_small_4xb256_1e-3lr_5m/swin_small_4xb256_fake5m.py @@ -0,0 +1,175 @@ +optim_wrapper = dict( + optimizer=dict( + type='AdamW', + lr=0.001, + weight_decay=0.05, + eps=1e-08, + betas=(0.9, 0.999), + _scope_='mmpretrain'), + paramwise_cfg=dict( + norm_decay_mult=0.0, + bias_decay_mult=0.0, + flat_decay_mult=0.0, + custom_keys=dict({ + '.absolute_pos_embed': dict(decay_mult=0.0), + '.relative_position_bias_table': dict(decay_mult=0.0) + })), + type='AmpOptimWrapper', + dtype='bfloat16', + clip_grad=dict(max_norm=5.0)) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=True, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=20, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=1024) +model = dict( + type='ImageClassifier', + backbone=dict( + type='SwinTransformer', arch='small', img_size=224, + drop_path_rate=0.3), + neck=dict(type='GlobalAveragePooling'), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=768, + init_cfg=None, + loss=dict( + type='LabelSmoothLoss', label_smooth_val=0.1, mode='original'), + cal_acc=False), + init_cfg=[ + dict(type='TruncNormal', layer='Linear', std=0.02, bias=0.0), + dict(type='Constant', layer='LayerNorm', val=1.0, bias=0.0) + ], + train_cfg=None) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stablediffusionV1-5R2-dpmsolver-25-5m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/swin_small_4xb256_1e-3lr_5m' diff --git a/swin_small_4xb256_fake5m_lr1e-4_10epoch/20230601_145758/20230601_145758.log b/swin_small_4xb256_fake5m_lr1e-4_10epoch/20230601_145758/20230601_145758.log new file mode 100644 index 0000000000000000000000000000000000000000..fc32e11a2ce7f4d0fa73406260af286245c6f9d9 --- /dev/null +++ b/swin_small_4xb256_fake5m_lr1e-4_10epoch/20230601_145758/20230601_145758.log @@ -0,0 +1,1604 @@ +2023/06/01 14:58:02 - mmengine - INFO - +------------------------------------------------------------ +System environment: + sys.platform: linux + Python: 3.10.9 (main, Mar 8 2023, 10:47:38) [GCC 11.2.0] + CUDA available: True + numpy_random_seed: 886372997 + GPU 0,1,2,3: NVIDIA A100-SXM4-80GB + CUDA_HOME: /mnt/petrelfs/share/cuda-11.6 + NVCC: Cuda compilation tools, release 11.6, V11.6.124 + GCC: gcc (GCC) 4.8.5 20150623 (Red Hat 4.8.5-44) + PyTorch: 1.13.1 + PyTorch compiling details: PyTorch built with: + - GCC 9.3 + - C++ Version: 201402 + - Intel(R) oneAPI Math Kernel Library Version 2021.4-Product Build 20210904 for Intel(R) 64 architecture applications + - Intel(R) MKL-DNN v2.6.0 (Git Hash 52b5f107dd9cf10910aaa19cb47f3abf9b349815) + - OpenMP 201511 (a.k.a. OpenMP 4.5) + - LAPACK is enabled (usually provided by MKL) + - NNPACK is enabled + - CPU capability usage: AVX2 + - CUDA Runtime 11.6 + - NVCC architecture flags: -gencode;arch=compute_37,code=sm_37;-gencode;arch=compute_50,code=sm_50;-gencode;arch=compute_60,code=sm_60;-gencode;arch=compute_61,code=sm_61;-gencode;arch=compute_70,code=sm_70;-gencode;arch=compute_75,code=sm_75;-gencode;arch=compute_80,code=sm_80;-gencode;arch=compute_86,code=sm_86;-gencode;arch=compute_37,code=compute_37 + - CuDNN 8.3.2 (built against CUDA 11.5) + - Magma 2.6.1 + - Build settings: BLAS_INFO=mkl, BUILD_TYPE=Release, CUDA_VERSION=11.6, CUDNN_VERSION=8.3.2, CXX_COMPILER=/opt/rh/devtoolset-9/root/usr/bin/c++, CXX_FLAGS= -fabi-version=11 -Wno-deprecated -fvisibility-inlines-hidden -DUSE_PTHREADPOOL -fopenmp -DNDEBUG -DUSE_KINETO -DUSE_FBGEMM -DUSE_QNNPACK -DUSE_PYTORCH_QNNPACK -DUSE_XNNPACK -DSYMBOLICATE_MOBILE_DEBUG_HANDLE -DEDGE_PROFILER_USE_KINETO -O2 -fPIC -Wno-narrowing -Wall -Wextra -Werror=return-type -Werror=non-virtual-dtor -Wno-missing-field-initializers -Wno-type-limits -Wno-array-bounds -Wno-unknown-pragmas -Wunused-local-typedefs -Wno-unused-parameter -Wno-unused-function -Wno-unused-result -Wno-strict-overflow -Wno-strict-aliasing -Wno-error=deprecated-declarations -Wno-stringop-overflow -Wno-psabi -Wno-error=pedantic -Wno-error=redundant-decls -Wno-error=old-style-cast -fdiagnostics-color=always -faligned-new -Wno-unused-but-set-variable -Wno-maybe-uninitialized -fno-math-errno -fno-trapping-math -Werror=format -Werror=cast-function-type -Wno-stringop-overflow, LAPACK_INFO=mkl, PERF_WITH_AVX=1, PERF_WITH_AVX2=1, PERF_WITH_AVX512=1, TORCH_VERSION=1.13.1, USE_CUDA=ON, USE_CUDNN=ON, USE_EXCEPTION_PTR=1, USE_GFLAGS=OFF, USE_GLOG=OFF, USE_MKL=ON, USE_MKLDNN=ON, USE_MPI=OFF, USE_NCCL=ON, USE_NNPACK=ON, USE_OPENMP=ON, USE_ROCM=OFF, + + TorchVision: 0.14.1 + OpenCV: 4.7.0 + MMEngine: 0.7.3 + +Runtime environment: + cudnn_benchmark: True + mp_cfg: {'mp_start_method': 'fork', 'opencv_num_threads': 0} + dist_cfg: {'backend': 'nccl'} + seed: None + deterministic: False + Distributed launcher: slurm + Distributed training: True + GPU number: 4 +------------------------------------------------------------ + +2023/06/01 14:58:03 - mmengine - INFO - Config: +optim_wrapper = dict( + optimizer=dict( + type='AdamW', + lr=0.0001, + weight_decay=0.05, + eps=1e-08, + betas=(0.9, 0.999), + _scope_='mmpretrain'), + paramwise_cfg=dict( + norm_decay_mult=0.0, + bias_decay_mult=0.0, + flat_decay_mult=0.0, + custom_keys=dict({ + '.absolute_pos_embed': dict(decay_mult=0.0), + '.relative_position_bias_table': dict(decay_mult=0.0) + })), + type='AmpOptimWrapper', + dtype='bfloat16', + clip_grad=dict(max_norm=5.0)) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=True, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=1024) +model = dict( + type='ImageClassifier', + backbone=dict( + type='SwinTransformer', arch='small', img_size=224, + drop_path_rate=0.3), + neck=dict(type='GlobalAveragePooling'), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=768, + init_cfg=None, + loss=dict( + type='LabelSmoothLoss', label_smooth_val=0.1, mode='original'), + cal_acc=False), + init_cfg=[ + dict(type='TruncNormal', layer='Linear', std=0.02, bias=0.0), + dict(type='Constant', layer='LayerNorm', val=1.0, bias=0.0) + ], + train_cfg=None) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stablediffusionV1-5R2-dpmsolver-25-5m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/swin_small_4xb256_fake5m_lr1e-4_10epoch' + +2023/06/01 14:58:07 - mmengine - INFO - Hooks will be executed in the following order: +before_run: +(VERY_HIGH ) RuntimeInfoHook +(BELOW_NORMAL) LoggerHook + -------------------- +before_train: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_train_epoch: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(NORMAL ) DistSamplerSeedHook + -------------------- +before_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook + -------------------- +after_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +after_train_epoch: +(NORMAL ) IterTimerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_val_epoch: +(NORMAL ) IterTimerHook + -------------------- +before_val_iter: +(NORMAL ) IterTimerHook + -------------------- +after_val_iter: +(NORMAL ) IterTimerHook +(NORMAL ) VisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_val_epoch: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +after_train: +(VERY_LOW ) CheckpointHook + -------------------- +before_test_epoch: +(NORMAL ) IterTimerHook + -------------------- +before_test_iter: +(NORMAL ) IterTimerHook + -------------------- +after_test_iter: +(NORMAL ) IterTimerHook +(NORMAL ) VisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_test_epoch: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_run: +(BELOW_NORMAL) LoggerHook + -------------------- +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.patch_embed.projection.bias:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.patch_embed.norm.weight:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.patch_embed.norm.bias:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.0.blocks.0.norm1.weight:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.0.blocks.0.norm1.bias:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.0.blocks.0.attn.w_msa.relative_position_bias_table:lr=0.0001 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.0.blocks.0.attn.w_msa.relative_position_bias_table:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.0.blocks.0.attn.w_msa.relative_position_bias_table:decay_mult=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.0.blocks.0.attn.w_msa.qkv.bias:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.0.blocks.0.attn.w_msa.proj.bias:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.0.blocks.0.norm2.weight:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.0.blocks.0.norm2.bias:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.0.blocks.0.ffn.layers.0.0.bias:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.0.blocks.0.ffn.layers.1.bias:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.0.blocks.1.norm1.weight:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.0.blocks.1.norm1.bias:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.0.blocks.1.attn.w_msa.relative_position_bias_table:lr=0.0001 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.0.blocks.1.attn.w_msa.relative_position_bias_table:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.0.blocks.1.attn.w_msa.relative_position_bias_table:decay_mult=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.0.blocks.1.attn.w_msa.qkv.bias:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.0.blocks.1.attn.w_msa.proj.bias:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.0.blocks.1.norm2.weight:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.0.blocks.1.norm2.bias:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.0.blocks.1.ffn.layers.0.0.bias:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.0.blocks.1.ffn.layers.1.bias:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.0.downsample.norm.weight:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.0.downsample.norm.bias:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.1.blocks.0.norm1.weight:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.1.blocks.0.norm1.bias:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.1.blocks.0.attn.w_msa.relative_position_bias_table:lr=0.0001 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.1.blocks.0.attn.w_msa.relative_position_bias_table:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.1.blocks.0.attn.w_msa.relative_position_bias_table:decay_mult=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.1.blocks.0.attn.w_msa.qkv.bias:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.1.blocks.0.attn.w_msa.proj.bias:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.1.blocks.0.norm2.weight:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.1.blocks.0.norm2.bias:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.1.blocks.0.ffn.layers.0.0.bias:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.1.blocks.0.ffn.layers.1.bias:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.1.blocks.1.norm1.weight:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.1.blocks.1.norm1.bias:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.1.blocks.1.attn.w_msa.relative_position_bias_table:lr=0.0001 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.1.blocks.1.attn.w_msa.relative_position_bias_table:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.1.blocks.1.attn.w_msa.relative_position_bias_table:decay_mult=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.1.blocks.1.attn.w_msa.qkv.bias:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.1.blocks.1.attn.w_msa.proj.bias:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.1.blocks.1.norm2.weight:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.1.blocks.1.norm2.bias:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.1.blocks.1.ffn.layers.0.0.bias:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.1.blocks.1.ffn.layers.1.bias:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.1.downsample.norm.weight:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.1.downsample.norm.bias:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.0.norm1.weight:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.0.norm1.bias:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.0.attn.w_msa.relative_position_bias_table:lr=0.0001 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.0.attn.w_msa.relative_position_bias_table:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.0.attn.w_msa.relative_position_bias_table:decay_mult=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.0.attn.w_msa.qkv.bias:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.0.attn.w_msa.proj.bias:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.0.norm2.weight:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.0.norm2.bias:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.0.ffn.layers.0.0.bias:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.0.ffn.layers.1.bias:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.1.norm1.weight:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.1.norm1.bias:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.1.attn.w_msa.relative_position_bias_table:lr=0.0001 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.1.attn.w_msa.relative_position_bias_table:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.1.attn.w_msa.relative_position_bias_table:decay_mult=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.1.attn.w_msa.qkv.bias:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.1.attn.w_msa.proj.bias:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.1.norm2.weight:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.1.norm2.bias:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.1.ffn.layers.0.0.bias:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.1.ffn.layers.1.bias:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.2.norm1.weight:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.2.norm1.bias:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.2.attn.w_msa.relative_position_bias_table:lr=0.0001 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.2.attn.w_msa.relative_position_bias_table:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.2.attn.w_msa.relative_position_bias_table:decay_mult=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.2.attn.w_msa.qkv.bias:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.2.attn.w_msa.proj.bias:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.2.norm2.weight:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.2.norm2.bias:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.2.ffn.layers.0.0.bias:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.2.ffn.layers.1.bias:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.3.norm1.weight:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.3.norm1.bias:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.3.attn.w_msa.relative_position_bias_table:lr=0.0001 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.3.attn.w_msa.relative_position_bias_table:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.3.attn.w_msa.relative_position_bias_table:decay_mult=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.3.attn.w_msa.qkv.bias:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.3.attn.w_msa.proj.bias:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.3.norm2.weight:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.3.norm2.bias:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.3.ffn.layers.0.0.bias:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.3.ffn.layers.1.bias:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.4.norm1.weight:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.4.norm1.bias:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.4.attn.w_msa.relative_position_bias_table:lr=0.0001 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.4.attn.w_msa.relative_position_bias_table:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.4.attn.w_msa.relative_position_bias_table:decay_mult=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.4.attn.w_msa.qkv.bias:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.4.attn.w_msa.proj.bias:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.4.norm2.weight:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.4.norm2.bias:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.4.ffn.layers.0.0.bias:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.4.ffn.layers.1.bias:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.5.norm1.weight:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.5.norm1.bias:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.5.attn.w_msa.relative_position_bias_table:lr=0.0001 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.5.attn.w_msa.relative_position_bias_table:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.5.attn.w_msa.relative_position_bias_table:decay_mult=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.5.attn.w_msa.qkv.bias:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.5.attn.w_msa.proj.bias:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.5.norm2.weight:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.5.norm2.bias:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.5.ffn.layers.0.0.bias:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.5.ffn.layers.1.bias:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.6.norm1.weight:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.6.norm1.bias:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.6.attn.w_msa.relative_position_bias_table:lr=0.0001 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.6.attn.w_msa.relative_position_bias_table:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.6.attn.w_msa.relative_position_bias_table:decay_mult=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.6.attn.w_msa.qkv.bias:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.6.attn.w_msa.proj.bias:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.6.norm2.weight:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.6.norm2.bias:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.6.ffn.layers.0.0.bias:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.6.ffn.layers.1.bias:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.7.norm1.weight:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.7.norm1.bias:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.7.attn.w_msa.relative_position_bias_table:lr=0.0001 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.7.attn.w_msa.relative_position_bias_table:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.7.attn.w_msa.relative_position_bias_table:decay_mult=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.7.attn.w_msa.qkv.bias:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.7.attn.w_msa.proj.bias:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.7.norm2.weight:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.7.norm2.bias:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.7.ffn.layers.0.0.bias:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.7.ffn.layers.1.bias:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.8.norm1.weight:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.8.norm1.bias:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.8.attn.w_msa.relative_position_bias_table:lr=0.0001 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.8.attn.w_msa.relative_position_bias_table:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.8.attn.w_msa.relative_position_bias_table:decay_mult=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.8.attn.w_msa.qkv.bias:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.8.attn.w_msa.proj.bias:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.8.norm2.weight:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.8.norm2.bias:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.8.ffn.layers.0.0.bias:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.8.ffn.layers.1.bias:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.9.norm1.weight:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.9.norm1.bias:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.9.attn.w_msa.relative_position_bias_table:lr=0.0001 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.9.attn.w_msa.relative_position_bias_table:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.9.attn.w_msa.relative_position_bias_table:decay_mult=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.9.attn.w_msa.qkv.bias:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.9.attn.w_msa.proj.bias:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.9.norm2.weight:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.9.norm2.bias:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.9.ffn.layers.0.0.bias:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.9.ffn.layers.1.bias:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.10.norm1.weight:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.10.norm1.bias:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.10.attn.w_msa.relative_position_bias_table:lr=0.0001 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.10.attn.w_msa.relative_position_bias_table:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.10.attn.w_msa.relative_position_bias_table:decay_mult=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.10.attn.w_msa.qkv.bias:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.10.attn.w_msa.proj.bias:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.10.norm2.weight:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.10.norm2.bias:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.10.ffn.layers.0.0.bias:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.10.ffn.layers.1.bias:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.11.norm1.weight:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.11.norm1.bias:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.11.attn.w_msa.relative_position_bias_table:lr=0.0001 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.11.attn.w_msa.relative_position_bias_table:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.11.attn.w_msa.relative_position_bias_table:decay_mult=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.11.attn.w_msa.qkv.bias:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.11.attn.w_msa.proj.bias:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.11.norm2.weight:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.11.norm2.bias:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.11.ffn.layers.0.0.bias:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.11.ffn.layers.1.bias:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.12.norm1.weight:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.12.norm1.bias:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.12.attn.w_msa.relative_position_bias_table:lr=0.0001 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.12.attn.w_msa.relative_position_bias_table:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.12.attn.w_msa.relative_position_bias_table:decay_mult=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.12.attn.w_msa.qkv.bias:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.12.attn.w_msa.proj.bias:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.12.norm2.weight:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.12.norm2.bias:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.12.ffn.layers.0.0.bias:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.12.ffn.layers.1.bias:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.13.norm1.weight:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.13.norm1.bias:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.13.attn.w_msa.relative_position_bias_table:lr=0.0001 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.13.attn.w_msa.relative_position_bias_table:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.13.attn.w_msa.relative_position_bias_table:decay_mult=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.13.attn.w_msa.qkv.bias:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.13.attn.w_msa.proj.bias:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.13.norm2.weight:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.13.norm2.bias:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.13.ffn.layers.0.0.bias:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.13.ffn.layers.1.bias:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.14.norm1.weight:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.14.norm1.bias:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.14.attn.w_msa.relative_position_bias_table:lr=0.0001 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.14.attn.w_msa.relative_position_bias_table:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.14.attn.w_msa.relative_position_bias_table:decay_mult=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.14.attn.w_msa.qkv.bias:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.14.attn.w_msa.proj.bias:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.14.norm2.weight:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.14.norm2.bias:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.14.ffn.layers.0.0.bias:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.14.ffn.layers.1.bias:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.15.norm1.weight:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.15.norm1.bias:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.15.attn.w_msa.relative_position_bias_table:lr=0.0001 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.15.attn.w_msa.relative_position_bias_table:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.15.attn.w_msa.relative_position_bias_table:decay_mult=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.15.attn.w_msa.qkv.bias:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.15.attn.w_msa.proj.bias:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.15.norm2.weight:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.15.norm2.bias:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.15.ffn.layers.0.0.bias:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.15.ffn.layers.1.bias:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.16.norm1.weight:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.16.norm1.bias:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.16.attn.w_msa.relative_position_bias_table:lr=0.0001 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.16.attn.w_msa.relative_position_bias_table:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.16.attn.w_msa.relative_position_bias_table:decay_mult=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.16.attn.w_msa.qkv.bias:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.16.attn.w_msa.proj.bias:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.16.norm2.weight:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.16.norm2.bias:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.16.ffn.layers.0.0.bias:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.16.ffn.layers.1.bias:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.17.norm1.weight:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.17.norm1.bias:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.17.attn.w_msa.relative_position_bias_table:lr=0.0001 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.17.attn.w_msa.relative_position_bias_table:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.17.attn.w_msa.relative_position_bias_table:decay_mult=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.17.attn.w_msa.qkv.bias:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.17.attn.w_msa.proj.bias:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.17.norm2.weight:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.17.norm2.bias:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.17.ffn.layers.0.0.bias:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.17.ffn.layers.1.bias:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.downsample.norm.weight:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.2.downsample.norm.bias:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.3.blocks.0.norm1.weight:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.3.blocks.0.norm1.bias:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.3.blocks.0.attn.w_msa.relative_position_bias_table:lr=0.0001 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.3.blocks.0.attn.w_msa.relative_position_bias_table:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.3.blocks.0.attn.w_msa.relative_position_bias_table:decay_mult=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.3.blocks.0.attn.w_msa.qkv.bias:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.3.blocks.0.attn.w_msa.proj.bias:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.3.blocks.0.norm2.weight:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.3.blocks.0.norm2.bias:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.3.blocks.0.ffn.layers.0.0.bias:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.3.blocks.0.ffn.layers.1.bias:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.3.blocks.1.norm1.weight:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.3.blocks.1.norm1.bias:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.3.blocks.1.attn.w_msa.relative_position_bias_table:lr=0.0001 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.3.blocks.1.attn.w_msa.relative_position_bias_table:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.3.blocks.1.attn.w_msa.relative_position_bias_table:decay_mult=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.3.blocks.1.attn.w_msa.qkv.bias:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.3.blocks.1.attn.w_msa.proj.bias:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.3.blocks.1.norm2.weight:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.3.blocks.1.norm2.bias:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.3.blocks.1.ffn.layers.0.0.bias:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.stages.3.blocks.1.ffn.layers.1.bias:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.norm3.weight:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- backbone.norm3.bias:weight_decay=0.0 +2023/06/01 14:58:34 - mmengine - INFO - paramwise_options -- head.fc.bias:weight_decay=0.0 +Name of parameter - Initialization information + +backbone.patch_embed.projection.weight - torch.Size([96, 3, 4, 4]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.patch_embed.projection.bias - torch.Size([96]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.patch_embed.norm.weight - torch.Size([96]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.patch_embed.norm.bias - torch.Size([96]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.0.blocks.0.norm1.weight - torch.Size([96]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.0.blocks.0.norm1.bias - torch.Size([96]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.0.blocks.0.attn.w_msa.relative_position_bias_table - torch.Size([169, 3]): +Initialized by user-defined `init_weights` in WindowMSA + +backbone.stages.0.blocks.0.attn.w_msa.qkv.weight - torch.Size([288, 96]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.blocks.0.attn.w_msa.qkv.bias - torch.Size([288]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.blocks.0.attn.w_msa.proj.weight - torch.Size([96, 96]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.blocks.0.attn.w_msa.proj.bias - torch.Size([96]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.blocks.0.norm2.weight - torch.Size([96]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.0.blocks.0.norm2.bias - torch.Size([96]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.0.blocks.0.ffn.layers.0.0.weight - torch.Size([384, 96]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.blocks.0.ffn.layers.0.0.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.blocks.0.ffn.layers.1.weight - torch.Size([96, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.blocks.0.ffn.layers.1.bias - torch.Size([96]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.blocks.1.norm1.weight - torch.Size([96]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.0.blocks.1.norm1.bias - torch.Size([96]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.0.blocks.1.attn.w_msa.relative_position_bias_table - torch.Size([169, 3]): +Initialized by user-defined `init_weights` in WindowMSA + +backbone.stages.0.blocks.1.attn.w_msa.qkv.weight - torch.Size([288, 96]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.blocks.1.attn.w_msa.qkv.bias - torch.Size([288]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.blocks.1.attn.w_msa.proj.weight - torch.Size([96, 96]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.blocks.1.attn.w_msa.proj.bias - torch.Size([96]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.blocks.1.norm2.weight - torch.Size([96]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.0.blocks.1.norm2.bias - torch.Size([96]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.0.blocks.1.ffn.layers.0.0.weight - torch.Size([384, 96]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.blocks.1.ffn.layers.0.0.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.blocks.1.ffn.layers.1.weight - torch.Size([96, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.blocks.1.ffn.layers.1.bias - torch.Size([96]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.downsample.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.0.downsample.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.0.downsample.reduction.weight - torch.Size([192, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.blocks.0.norm1.weight - torch.Size([192]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.1.blocks.0.norm1.bias - torch.Size([192]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.1.blocks.0.attn.w_msa.relative_position_bias_table - torch.Size([169, 6]): +Initialized by user-defined `init_weights` in WindowMSA + +backbone.stages.1.blocks.0.attn.w_msa.qkv.weight - torch.Size([576, 192]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.blocks.0.attn.w_msa.qkv.bias - torch.Size([576]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.blocks.0.attn.w_msa.proj.weight - torch.Size([192, 192]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.blocks.0.attn.w_msa.proj.bias - torch.Size([192]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.blocks.0.norm2.weight - torch.Size([192]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.1.blocks.0.norm2.bias - torch.Size([192]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.1.blocks.0.ffn.layers.0.0.weight - torch.Size([768, 192]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.blocks.0.ffn.layers.0.0.bias - torch.Size([768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.blocks.0.ffn.layers.1.weight - torch.Size([192, 768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.blocks.0.ffn.layers.1.bias - torch.Size([192]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.blocks.1.norm1.weight - torch.Size([192]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.1.blocks.1.norm1.bias - torch.Size([192]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.1.blocks.1.attn.w_msa.relative_position_bias_table - torch.Size([169, 6]): +Initialized by user-defined `init_weights` in WindowMSA + +backbone.stages.1.blocks.1.attn.w_msa.qkv.weight - torch.Size([576, 192]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.blocks.1.attn.w_msa.qkv.bias - torch.Size([576]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.blocks.1.attn.w_msa.proj.weight - torch.Size([192, 192]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.blocks.1.attn.w_msa.proj.bias - torch.Size([192]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.blocks.1.norm2.weight - torch.Size([192]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.1.blocks.1.norm2.bias - torch.Size([192]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.1.blocks.1.ffn.layers.0.0.weight - torch.Size([768, 192]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.blocks.1.ffn.layers.0.0.bias - torch.Size([768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.blocks.1.ffn.layers.1.weight - torch.Size([192, 768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.blocks.1.ffn.layers.1.bias - torch.Size([192]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.downsample.norm.weight - torch.Size([768]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.1.downsample.norm.bias - torch.Size([768]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.1.downsample.reduction.weight - torch.Size([384, 768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.0.norm1.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.0.norm1.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.0.attn.w_msa.relative_position_bias_table - torch.Size([169, 12]): +Initialized by user-defined `init_weights` in WindowMSA + +backbone.stages.2.blocks.0.attn.w_msa.qkv.weight - torch.Size([1152, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.0.attn.w_msa.qkv.bias - torch.Size([1152]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.0.attn.w_msa.proj.weight - torch.Size([384, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.0.attn.w_msa.proj.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.0.norm2.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.0.norm2.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.0.ffn.layers.0.0.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.0.ffn.layers.0.0.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.0.ffn.layers.1.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.0.ffn.layers.1.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.1.norm1.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.1.norm1.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.1.attn.w_msa.relative_position_bias_table - torch.Size([169, 12]): +Initialized by user-defined `init_weights` in WindowMSA + +backbone.stages.2.blocks.1.attn.w_msa.qkv.weight - torch.Size([1152, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.1.attn.w_msa.qkv.bias - torch.Size([1152]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.1.attn.w_msa.proj.weight - torch.Size([384, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.1.attn.w_msa.proj.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.1.norm2.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.1.norm2.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.1.ffn.layers.0.0.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.1.ffn.layers.0.0.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.1.ffn.layers.1.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.1.ffn.layers.1.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.2.norm1.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.2.norm1.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.2.attn.w_msa.relative_position_bias_table - torch.Size([169, 12]): +Initialized by user-defined `init_weights` in WindowMSA + +backbone.stages.2.blocks.2.attn.w_msa.qkv.weight - torch.Size([1152, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.2.attn.w_msa.qkv.bias - torch.Size([1152]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.2.attn.w_msa.proj.weight - torch.Size([384, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.2.attn.w_msa.proj.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.2.norm2.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.2.norm2.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.2.ffn.layers.0.0.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.2.ffn.layers.0.0.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.2.ffn.layers.1.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.2.ffn.layers.1.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.3.norm1.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.3.norm1.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.3.attn.w_msa.relative_position_bias_table - torch.Size([169, 12]): +Initialized by user-defined `init_weights` in WindowMSA + +backbone.stages.2.blocks.3.attn.w_msa.qkv.weight - torch.Size([1152, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.3.attn.w_msa.qkv.bias - torch.Size([1152]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.3.attn.w_msa.proj.weight - torch.Size([384, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.3.attn.w_msa.proj.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.3.norm2.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.3.norm2.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.3.ffn.layers.0.0.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.3.ffn.layers.0.0.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.3.ffn.layers.1.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.3.ffn.layers.1.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.4.norm1.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.4.norm1.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.4.attn.w_msa.relative_position_bias_table - torch.Size([169, 12]): +Initialized by user-defined `init_weights` in WindowMSA + +backbone.stages.2.blocks.4.attn.w_msa.qkv.weight - torch.Size([1152, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.4.attn.w_msa.qkv.bias - torch.Size([1152]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.4.attn.w_msa.proj.weight - torch.Size([384, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.4.attn.w_msa.proj.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.4.norm2.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.4.norm2.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.4.ffn.layers.0.0.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.4.ffn.layers.0.0.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.4.ffn.layers.1.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.4.ffn.layers.1.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.5.norm1.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.5.norm1.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.5.attn.w_msa.relative_position_bias_table - torch.Size([169, 12]): +Initialized by user-defined `init_weights` in WindowMSA + +backbone.stages.2.blocks.5.attn.w_msa.qkv.weight - torch.Size([1152, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.5.attn.w_msa.qkv.bias - torch.Size([1152]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.5.attn.w_msa.proj.weight - torch.Size([384, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.5.attn.w_msa.proj.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.5.norm2.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.5.norm2.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.5.ffn.layers.0.0.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.5.ffn.layers.0.0.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.5.ffn.layers.1.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.5.ffn.layers.1.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.6.norm1.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.6.norm1.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.6.attn.w_msa.relative_position_bias_table - torch.Size([169, 12]): +Initialized by user-defined `init_weights` in WindowMSA + +backbone.stages.2.blocks.6.attn.w_msa.qkv.weight - torch.Size([1152, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.6.attn.w_msa.qkv.bias - torch.Size([1152]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.6.attn.w_msa.proj.weight - torch.Size([384, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.6.attn.w_msa.proj.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.6.norm2.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.6.norm2.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.6.ffn.layers.0.0.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.6.ffn.layers.0.0.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.6.ffn.layers.1.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.6.ffn.layers.1.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.7.norm1.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.7.norm1.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.7.attn.w_msa.relative_position_bias_table - torch.Size([169, 12]): +Initialized by user-defined `init_weights` in WindowMSA + +backbone.stages.2.blocks.7.attn.w_msa.qkv.weight - torch.Size([1152, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.7.attn.w_msa.qkv.bias - torch.Size([1152]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.7.attn.w_msa.proj.weight - torch.Size([384, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.7.attn.w_msa.proj.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.7.norm2.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.7.norm2.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.7.ffn.layers.0.0.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.7.ffn.layers.0.0.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.7.ffn.layers.1.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.7.ffn.layers.1.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.8.norm1.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.8.norm1.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.8.attn.w_msa.relative_position_bias_table - torch.Size([169, 12]): +Initialized by user-defined `init_weights` in WindowMSA + +backbone.stages.2.blocks.8.attn.w_msa.qkv.weight - torch.Size([1152, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.8.attn.w_msa.qkv.bias - torch.Size([1152]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.8.attn.w_msa.proj.weight - torch.Size([384, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.8.attn.w_msa.proj.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.8.norm2.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.8.norm2.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.8.ffn.layers.0.0.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.8.ffn.layers.0.0.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.8.ffn.layers.1.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.8.ffn.layers.1.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.9.norm1.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.9.norm1.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.9.attn.w_msa.relative_position_bias_table - torch.Size([169, 12]): +Initialized by user-defined `init_weights` in WindowMSA + +backbone.stages.2.blocks.9.attn.w_msa.qkv.weight - torch.Size([1152, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.9.attn.w_msa.qkv.bias - torch.Size([1152]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.9.attn.w_msa.proj.weight - torch.Size([384, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.9.attn.w_msa.proj.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.9.norm2.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.9.norm2.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.9.ffn.layers.0.0.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.9.ffn.layers.0.0.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.9.ffn.layers.1.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.9.ffn.layers.1.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.10.norm1.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.10.norm1.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.10.attn.w_msa.relative_position_bias_table - torch.Size([169, 12]): +Initialized by user-defined `init_weights` in WindowMSA + +backbone.stages.2.blocks.10.attn.w_msa.qkv.weight - torch.Size([1152, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.10.attn.w_msa.qkv.bias - torch.Size([1152]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.10.attn.w_msa.proj.weight - torch.Size([384, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.10.attn.w_msa.proj.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.10.norm2.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.10.norm2.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.10.ffn.layers.0.0.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.10.ffn.layers.0.0.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.10.ffn.layers.1.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.10.ffn.layers.1.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.11.norm1.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.11.norm1.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.11.attn.w_msa.relative_position_bias_table - torch.Size([169, 12]): +Initialized by user-defined `init_weights` in WindowMSA + +backbone.stages.2.blocks.11.attn.w_msa.qkv.weight - torch.Size([1152, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.11.attn.w_msa.qkv.bias - torch.Size([1152]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.11.attn.w_msa.proj.weight - torch.Size([384, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.11.attn.w_msa.proj.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.11.norm2.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.11.norm2.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.11.ffn.layers.0.0.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.11.ffn.layers.0.0.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.11.ffn.layers.1.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.11.ffn.layers.1.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.12.norm1.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.12.norm1.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.12.attn.w_msa.relative_position_bias_table - torch.Size([169, 12]): +Initialized by user-defined `init_weights` in WindowMSA + +backbone.stages.2.blocks.12.attn.w_msa.qkv.weight - torch.Size([1152, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.12.attn.w_msa.qkv.bias - torch.Size([1152]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.12.attn.w_msa.proj.weight - torch.Size([384, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.12.attn.w_msa.proj.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.12.norm2.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.12.norm2.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.12.ffn.layers.0.0.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.12.ffn.layers.0.0.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.12.ffn.layers.1.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.12.ffn.layers.1.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.13.norm1.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.13.norm1.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.13.attn.w_msa.relative_position_bias_table - torch.Size([169, 12]): +Initialized by user-defined `init_weights` in WindowMSA + +backbone.stages.2.blocks.13.attn.w_msa.qkv.weight - torch.Size([1152, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.13.attn.w_msa.qkv.bias - torch.Size([1152]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.13.attn.w_msa.proj.weight - torch.Size([384, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.13.attn.w_msa.proj.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.13.norm2.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.13.norm2.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.13.ffn.layers.0.0.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.13.ffn.layers.0.0.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.13.ffn.layers.1.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.13.ffn.layers.1.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.14.norm1.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.14.norm1.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.14.attn.w_msa.relative_position_bias_table - torch.Size([169, 12]): +Initialized by user-defined `init_weights` in WindowMSA + +backbone.stages.2.blocks.14.attn.w_msa.qkv.weight - torch.Size([1152, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.14.attn.w_msa.qkv.bias - torch.Size([1152]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.14.attn.w_msa.proj.weight - torch.Size([384, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.14.attn.w_msa.proj.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.14.norm2.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.14.norm2.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.14.ffn.layers.0.0.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.14.ffn.layers.0.0.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.14.ffn.layers.1.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.14.ffn.layers.1.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.15.norm1.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.15.norm1.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.15.attn.w_msa.relative_position_bias_table - torch.Size([169, 12]): +Initialized by user-defined `init_weights` in WindowMSA + +backbone.stages.2.blocks.15.attn.w_msa.qkv.weight - torch.Size([1152, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.15.attn.w_msa.qkv.bias - torch.Size([1152]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.15.attn.w_msa.proj.weight - torch.Size([384, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.15.attn.w_msa.proj.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.15.norm2.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.15.norm2.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.15.ffn.layers.0.0.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.15.ffn.layers.0.0.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.15.ffn.layers.1.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.15.ffn.layers.1.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.16.norm1.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.16.norm1.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.16.attn.w_msa.relative_position_bias_table - torch.Size([169, 12]): +Initialized by user-defined `init_weights` in WindowMSA + +backbone.stages.2.blocks.16.attn.w_msa.qkv.weight - torch.Size([1152, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.16.attn.w_msa.qkv.bias - torch.Size([1152]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.16.attn.w_msa.proj.weight - torch.Size([384, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.16.attn.w_msa.proj.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.16.norm2.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.16.norm2.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.16.ffn.layers.0.0.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.16.ffn.layers.0.0.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.16.ffn.layers.1.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.16.ffn.layers.1.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.17.norm1.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.17.norm1.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.17.attn.w_msa.relative_position_bias_table - torch.Size([169, 12]): +Initialized by user-defined `init_weights` in WindowMSA + +backbone.stages.2.blocks.17.attn.w_msa.qkv.weight - torch.Size([1152, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.17.attn.w_msa.qkv.bias - torch.Size([1152]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.17.attn.w_msa.proj.weight - torch.Size([384, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.17.attn.w_msa.proj.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.17.norm2.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.17.norm2.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.17.ffn.layers.0.0.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.17.ffn.layers.0.0.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.17.ffn.layers.1.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.17.ffn.layers.1.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.downsample.norm.weight - torch.Size([1536]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.downsample.norm.bias - torch.Size([1536]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.downsample.reduction.weight - torch.Size([768, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.blocks.0.norm1.weight - torch.Size([768]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.3.blocks.0.norm1.bias - torch.Size([768]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.3.blocks.0.attn.w_msa.relative_position_bias_table - torch.Size([169, 24]): +Initialized by user-defined `init_weights` in WindowMSA + +backbone.stages.3.blocks.0.attn.w_msa.qkv.weight - torch.Size([2304, 768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.blocks.0.attn.w_msa.qkv.bias - torch.Size([2304]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.blocks.0.attn.w_msa.proj.weight - torch.Size([768, 768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.blocks.0.attn.w_msa.proj.bias - torch.Size([768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.blocks.0.norm2.weight - torch.Size([768]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.3.blocks.0.norm2.bias - torch.Size([768]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.3.blocks.0.ffn.layers.0.0.weight - torch.Size([3072, 768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.blocks.0.ffn.layers.0.0.bias - torch.Size([3072]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.blocks.0.ffn.layers.1.weight - torch.Size([768, 3072]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.blocks.0.ffn.layers.1.bias - torch.Size([768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.blocks.1.norm1.weight - torch.Size([768]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.3.blocks.1.norm1.bias - torch.Size([768]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.3.blocks.1.attn.w_msa.relative_position_bias_table - torch.Size([169, 24]): +Initialized by user-defined `init_weights` in WindowMSA + +backbone.stages.3.blocks.1.attn.w_msa.qkv.weight - torch.Size([2304, 768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.blocks.1.attn.w_msa.qkv.bias - torch.Size([2304]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.blocks.1.attn.w_msa.proj.weight - torch.Size([768, 768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.blocks.1.attn.w_msa.proj.bias - torch.Size([768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.blocks.1.norm2.weight - torch.Size([768]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.3.blocks.1.norm2.bias - torch.Size([768]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.3.blocks.1.ffn.layers.0.0.weight - torch.Size([3072, 768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.blocks.1.ffn.layers.0.0.bias - torch.Size([3072]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.blocks.1.ffn.layers.1.weight - torch.Size([768, 3072]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.blocks.1.ffn.layers.1.bias - torch.Size([768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.norm3.weight - torch.Size([768]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.norm3.bias - torch.Size([768]): +The value is the same before and after calling `init_weights` of ImageClassifier + +head.fc.weight - torch.Size([2, 768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +head.fc.bias - torch.Size([2]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 +2023/06/01 14:58:35 - mmengine - WARNING - "FileClient" will be deprecated in future. Please use io functions in https://mmengine.readthedocs.io/en/latest/api/fileio.html#file-io +2023/06/01 14:58:35 - mmengine - WARNING - "HardDiskBackend" is the alias of "LocalBackend" and the former will be deprecated in future. +2023/06/01 14:58:35 - mmengine - INFO - Checkpoints will be saved to /mnt/petrelfs/luzeyu/workspace/fakebench/mmpretrain/workdir/swin_small_4xb256_fake5m_lr1e-4_10epoch. +2023/06/01 15:00:03 - mmengine - INFO - Epoch(train) [1][ 100/5758] lr: 1.0000e-04 eta: 14:01:11 time: 0.7763 data_time: 0.0021 memory: 28798 grad_norm: 0.6705 loss: 0.6062 +2023/06/01 15:01:18 - mmengine - INFO - Epoch(train) [1][ 200/5758] lr: 1.0000e-04 eta: 12:57:43 time: 0.6906 data_time: 0.0017 memory: 28798 grad_norm: 1.1626 loss: 0.5714 +2023/06/01 15:02:32 - mmengine - INFO - Epoch(train) [1][ 300/5758] lr: 1.0000e-04 eta: 12:34:41 time: 0.7719 data_time: 0.0015 memory: 28798 grad_norm: 1.5613 loss: 0.5301 +2023/06/01 15:03:49 - mmengine - INFO - Epoch(train) [1][ 400/5758] lr: 1.0000e-04 eta: 12:27:33 time: 0.7413 data_time: 0.0019 memory: 28798 grad_norm: 2.1140 loss: 0.5166 +2023/06/01 15:05:05 - mmengine - INFO - Epoch(train) [1][ 500/5758] lr: 1.0000e-04 eta: 12:21:39 time: 0.7377 data_time: 0.0018 memory: 28798 grad_norm: 3.2186 loss: 0.5097 +2023/06/01 15:06:23 - mmengine - INFO - Epoch(train) [1][ 600/5758] lr: 1.0000e-04 eta: 12:20:29 time: 0.7309 data_time: 0.0021 memory: 28798 grad_norm: 2.1621 loss: 0.4747 +2023/06/01 15:07:36 - mmengine - INFO - Epoch(train) [1][ 700/5758] lr: 1.0000e-04 eta: 12:12:14 time: 0.7392 data_time: 0.0174 memory: 28798 grad_norm: 3.3911 loss: 0.4516 +2023/06/01 15:08:50 - mmengine - INFO - Epoch(train) [1][ 800/5758] lr: 1.0000e-04 eta: 12:07:55 time: 0.7468 data_time: 0.1882 memory: 28798 grad_norm: 5.2211 loss: 0.4579 +2023/06/01 15:10:08 - mmengine - INFO - Epoch(train) [1][ 900/5758] lr: 1.0000e-04 eta: 12:07:00 time: 0.7830 data_time: 0.1557 memory: 28798 grad_norm: 4.6969 loss: 0.4452 +2023/06/01 15:11:29 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_lr1e-4_10epoch_20230601_145758 +2023/06/01 15:11:29 - mmengine - INFO - Epoch(train) [1][1000/5758] lr: 1.0000e-04 eta: 12:09:53 time: 0.8109 data_time: 0.0020 memory: 28798 grad_norm: 5.2248 loss: 0.4158 +2023/06/01 15:12:48 - mmengine - INFO - Epoch(train) [1][1100/5758] lr: 1.0000e-04 eta: 12:09:54 time: 0.8031 data_time: 0.0785 memory: 28798 grad_norm: 5.6849 loss: 0.4143 +2023/06/01 15:14:09 - mmengine - INFO - Epoch(train) [1][1200/5758] lr: 1.0000e-04 eta: 12:11:16 time: 0.8169 data_time: 0.1713 memory: 28798 grad_norm: 6.5455 loss: 0.4374 +2023/06/01 15:15:29 - mmengine - INFO - Epoch(train) [1][1300/5758] lr: 1.0000e-04 eta: 12:11:32 time: 0.8478 data_time: 0.1482 memory: 28798 grad_norm: 5.5206 loss: 0.4201 +2023/06/01 15:16:50 - mmengine - INFO - Epoch(train) [1][1400/5758] lr: 1.0000e-04 eta: 12:12:36 time: 0.8419 data_time: 0.0529 memory: 28798 grad_norm: 4.8836 loss: 0.4078 +2023/06/01 15:18:13 - mmengine - INFO - Epoch(train) [1][1500/5758] lr: 1.0000e-04 eta: 12:14:02 time: 0.7989 data_time: 0.2666 memory: 28798 grad_norm: 7.7918 loss: 0.3932 +2023/06/01 15:19:35 - mmengine - INFO - Epoch(train) [1][1600/5758] lr: 1.0000e-04 eta: 12:14:58 time: 0.8332 data_time: 0.2499 memory: 28798 grad_norm: 11.6814 loss: 0.4387 +2023/06/01 15:20:55 - mmengine - INFO - Epoch(train) [1][1700/5758] lr: 1.0000e-04 eta: 12:14:14 time: 0.7855 data_time: 0.1454 memory: 28798 grad_norm: 6.8590 loss: 0.3929 +2023/06/01 15:22:19 - mmengine - INFO - Epoch(train) [1][1800/5758] lr: 1.0000e-04 eta: 12:15:18 time: 0.8223 data_time: 0.3192 memory: 28798 grad_norm: 5.1236 loss: 0.3864 +2023/06/01 15:23:44 - mmengine - INFO - Epoch(train) [1][1900/5758] lr: 1.0000e-04 eta: 12:17:09 time: 0.8928 data_time: 0.3837 memory: 28798 grad_norm: 8.2443 loss: 0.3742 +2023/06/01 15:25:06 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_lr1e-4_10epoch_20230601_145758 +2023/06/01 15:25:06 - mmengine - INFO - Epoch(train) [1][2000/5758] lr: 1.0000e-04 eta: 12:16:39 time: 0.8035 data_time: 0.2841 memory: 28798 grad_norm: 7.0601 loss: 0.3939 +2023/06/01 15:26:25 - mmengine - INFO - Epoch(train) [1][2100/5758] lr: 1.0000e-04 eta: 12:15:10 time: 0.7488 data_time: 0.2436 memory: 28798 grad_norm: 16.2040 loss: 0.4162 +2023/06/01 15:27:45 - mmengine - INFO - Epoch(train) [1][2200/5758] lr: 1.0000e-04 eta: 12:14:17 time: 0.7401 data_time: 0.2210 memory: 28798 grad_norm: 6.9176 loss: 0.3634 +2023/06/01 15:29:10 - mmengine - INFO - Epoch(train) [1][2300/5758] lr: 1.0000e-04 eta: 12:14:53 time: 0.8813 data_time: 0.3714 memory: 28798 grad_norm: 8.0170 loss: 0.3624 +2023/06/01 15:30:33 - mmengine - INFO - Epoch(train) [1][2400/5758] lr: 1.0000e-04 eta: 12:14:58 time: 0.7959 data_time: 0.2890 memory: 28798 grad_norm: 8.5125 loss: 0.3735 +2023/06/01 15:31:57 - mmengine - INFO - Epoch(train) [1][2500/5758] lr: 1.0000e-04 eta: 12:15:03 time: 0.8811 data_time: 0.3608 memory: 28798 grad_norm: 7.5858 loss: 0.3519 +2023/06/01 15:33:21 - mmengine - INFO - Epoch(train) [1][2600/5758] lr: 1.0000e-04 eta: 12:14:59 time: 0.8774 data_time: 0.3699 memory: 28798 grad_norm: 16.2857 loss: 0.3764 +2023/06/01 15:34:40 - mmengine - INFO - Epoch(train) [1][2700/5758] lr: 1.0000e-04 eta: 12:13:27 time: 0.8199 data_time: 0.3132 memory: 28798 grad_norm: 7.6101 loss: 0.3433 +2023/06/01 15:36:01 - mmengine - INFO - Epoch(train) [1][2800/5758] lr: 1.0000e-04 eta: 12:12:13 time: 0.8349 data_time: 0.3233 memory: 28798 grad_norm: 7.7804 loss: 0.3408 +2023/06/01 15:37:27 - mmengine - INFO - Epoch(train) [1][2900/5758] lr: 1.0000e-04 eta: 12:12:38 time: 0.7902 data_time: 0.2815 memory: 28798 grad_norm: 13.4135 loss: 0.3541 +2023/06/01 15:38:45 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_lr1e-4_10epoch_20230601_145758 +2023/06/01 15:38:45 - mmengine - INFO - Epoch(train) [1][3000/5758] lr: 1.0000e-04 eta: 12:10:44 time: 0.7797 data_time: 0.2675 memory: 28798 grad_norm: 13.4851 loss: 0.3433 +2023/06/01 15:40:05 - mmengine - INFO - Epoch(train) [1][3100/5758] lr: 1.0000e-04 eta: 12:09:11 time: 0.7560 data_time: 0.1362 memory: 28798 grad_norm: 5.6695 loss: 0.3430 +2023/06/01 15:41:27 - mmengine - INFO - Epoch(train) [1][3200/5758] lr: 1.0000e-04 eta: 12:08:21 time: 0.7884 data_time: 0.0025 memory: 28798 grad_norm: 23.5131 loss: 0.3792 +2023/06/01 15:42:47 - mmengine - INFO - Epoch(train) [1][3300/5758] lr: 1.0000e-04 eta: 12:07:06 time: 0.8394 data_time: 0.0030 memory: 28798 grad_norm: 6.3269 loss: 0.3306 +2023/06/01 15:44:06 - mmengine - INFO - Epoch(train) [1][3400/5758] lr: 1.0000e-04 eta: 12:05:13 time: 0.7954 data_time: 0.0021 memory: 28798 grad_norm: 20.5074 loss: 0.3468 +2023/06/01 15:45:26 - mmengine - INFO - Epoch(train) [1][3500/5758] lr: 1.0000e-04 eta: 12:03:47 time: 0.7939 data_time: 0.0023 memory: 28798 grad_norm: 9.2070 loss: 0.3290 +2023/06/01 15:46:47 - mmengine - INFO - Epoch(train) [1][3600/5758] lr: 1.0000e-04 eta: 12:02:38 time: 0.8422 data_time: 0.0021 memory: 28798 grad_norm: 8.3803 loss: 0.3227 +2023/06/01 15:48:05 - mmengine - INFO - Epoch(train) [1][3700/5758] lr: 1.0000e-04 eta: 12:00:50 time: 0.7844 data_time: 0.0018 memory: 28798 grad_norm: 15.2620 loss: 0.3457 +2023/06/01 15:49:25 - mmengine - INFO - Epoch(train) [1][3800/5758] lr: 1.0000e-04 eta: 11:59:19 time: 0.8164 data_time: 0.0019 memory: 28798 grad_norm: 7.0374 loss: 0.3087 +2023/06/01 15:50:44 - mmengine - INFO - Epoch(train) [1][3900/5758] lr: 1.0000e-04 eta: 11:57:47 time: 0.8042 data_time: 0.0027 memory: 28798 grad_norm: 16.2761 loss: 0.3059 +2023/06/01 15:51:59 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_lr1e-4_10epoch_20230601_145758 +2023/06/01 15:51:59 - mmengine - INFO - Epoch(train) [1][4000/5758] lr: 1.0000e-04 eta: 11:55:18 time: 0.7415 data_time: 0.0019 memory: 28798 grad_norm: 11.9100 loss: 0.3202 +2023/06/01 15:53:17 - mmengine - INFO - Epoch(train) [1][4100/5758] lr: 1.0000e-04 eta: 11:53:31 time: 0.8138 data_time: 0.0020 memory: 28798 grad_norm: 12.1173 loss: 0.3216 +2023/06/01 15:54:35 - mmengine - INFO - Epoch(train) [1][4200/5758] lr: 1.0000e-04 eta: 11:51:44 time: 0.7554 data_time: 0.0019 memory: 28798 grad_norm: 11.9865 loss: 0.3188 +2023/06/01 15:55:53 - mmengine - INFO - Epoch(train) [1][4300/5758] lr: 1.0000e-04 eta: 11:50:02 time: 0.7607 data_time: 0.0019 memory: 28798 grad_norm: 24.5623 loss: 0.3640 diff --git a/swin_small_4xb256_fake5m_lr1e-4_10epoch/20230601_145758/vis_data/20230601_145758.json b/swin_small_4xb256_fake5m_lr1e-4_10epoch/20230601_145758/vis_data/20230601_145758.json new file mode 100644 index 0000000000000000000000000000000000000000..949cd8b7676f7558b712b2ebfea04d7dde1454a4 --- /dev/null +++ b/swin_small_4xb256_fake5m_lr1e-4_10epoch/20230601_145758/vis_data/20230601_145758.json @@ -0,0 +1,43 @@ +{"lr": 0.0001, "data_time": 0.002135610580444336, "grad_norm": 0.6704509288072587, "loss": 0.6062085151672363, "time": 0.7762885808944702, "epoch": 1, "memory": 28798, "step": 100} +{"lr": 0.0001, "data_time": 0.0017275571823120116, "grad_norm": 1.1625975847244263, "loss": 0.5713563799858093, "time": 0.6906431674957275, "epoch": 1, "memory": 28798, "step": 200} +{"lr": 0.0001, "data_time": 0.0014902591705322266, "grad_norm": 1.561329984664917, "loss": 0.5300649046897888, "time": 0.7719131469726562, "epoch": 1, "memory": 28798, "step": 300} +{"lr": 0.0001, "data_time": 0.0018556356430053712, "grad_norm": 2.1139844059944153, "loss": 0.5165584325790405, "time": 0.7412705183029175, "epoch": 1, "memory": 28798, "step": 400} +{"lr": 0.0001, "data_time": 0.0018137454986572265, "grad_norm": 3.2185787320137025, "loss": 0.5097117364406586, "time": 0.7377352714538574, "epoch": 1, "memory": 28798, "step": 500} +{"lr": 0.0001, "data_time": 0.002111053466796875, "grad_norm": 2.162058413028717, "loss": 0.474714520573616, "time": 0.7309098482131958, "epoch": 1, "memory": 28798, "step": 600} +{"lr": 0.0001, "data_time": 0.01737983226776123, "grad_norm": 3.3910586714744566, "loss": 0.45162872076034544, "time": 0.7392367601394654, "epoch": 1, "memory": 28798, "step": 700} +{"lr": 0.0001, "data_time": 0.1882148027420044, "grad_norm": 5.221115052700043, "loss": 0.45787692070007324, "time": 0.746848726272583, "epoch": 1, "memory": 28798, "step": 800} +{"lr": 0.0001, "data_time": 0.15573256015777587, "grad_norm": 4.696902573108673, "loss": 0.44515292942523954, "time": 0.7830199241638184, "epoch": 1, "memory": 28798, "step": 900} +{"lr": 0.0001, "data_time": 0.001980447769165039, "grad_norm": 5.224834561347961, "loss": 0.41580674350261687, "time": 0.8109351396560669, "epoch": 1, "memory": 28798, "step": 1000} +{"lr": 0.0001, "data_time": 0.07847323417663574, "grad_norm": 5.684916365146637, "loss": 0.4143247872591019, "time": 0.8031377553939819, "epoch": 1, "memory": 28798, "step": 1100} +{"lr": 0.0001, "data_time": 0.17131073474884034, "grad_norm": 6.545488166809082, "loss": 0.4373924136161804, "time": 0.8168726205825806, "epoch": 1, "memory": 28798, "step": 1200} +{"lr": 0.0001, "data_time": 0.148163366317749, "grad_norm": 5.520623433589935, "loss": 0.42006475627422335, "time": 0.8478182792663574, "epoch": 1, "memory": 28798, "step": 1300} +{"lr": 0.0001, "data_time": 0.052912020683288576, "grad_norm": 4.88360276222229, "loss": 0.4077721267938614, "time": 0.8419150590896607, "epoch": 1, "memory": 28798, "step": 1400} +{"lr": 0.0001, "data_time": 0.2666065454483032, "grad_norm": 7.791791701316834, "loss": 0.393210631608963, "time": 0.7989365577697753, "epoch": 1, "memory": 28798, "step": 1500} +{"lr": 0.0001, "data_time": 0.24985790252685547, "grad_norm": 11.681429147720337, "loss": 0.43874755799770354, "time": 0.8331817150115967, "epoch": 1, "memory": 28798, "step": 1600} +{"lr": 0.0001, "data_time": 0.14540016651153564, "grad_norm": 6.859038209915161, "loss": 0.3928732484579086, "time": 0.7855009317398072, "epoch": 1, "memory": 28798, "step": 1700} +{"lr": 0.0001, "data_time": 0.31921954154968263, "grad_norm": 5.1235524892807005, "loss": 0.38636680841445925, "time": 0.8223020315170289, "epoch": 1, "memory": 28798, "step": 1800} +{"lr": 0.0001, "data_time": 0.3836942672729492, "grad_norm": 8.244262409210204, "loss": 0.37420834600925446, "time": 0.892802619934082, "epoch": 1, "memory": 28798, "step": 1900} +{"lr": 0.0001, "data_time": 0.2840893268585205, "grad_norm": 7.060136818885804, "loss": 0.393887659907341, "time": 0.8034720659255982, "epoch": 1, "memory": 28798, "step": 2000} +{"lr": 0.0001, "data_time": 0.24361073970794678, "grad_norm": 16.203964400291444, "loss": 0.4162459999322891, "time": 0.748774790763855, "epoch": 1, "memory": 28798, "step": 2100} +{"lr": 0.0001, "data_time": 0.22101283073425293, "grad_norm": 6.91764805316925, "loss": 0.36337478160858155, "time": 0.7401018619537354, "epoch": 1, "memory": 28798, "step": 2200} +{"lr": 0.0001, "data_time": 0.37143445014953613, "grad_norm": 8.017024970054626, "loss": 0.36236098408699036, "time": 0.8812745094299317, "epoch": 1, "memory": 28798, "step": 2300} +{"lr": 0.0001, "data_time": 0.28904483318328855, "grad_norm": 8.512488317489623, "loss": 0.3735008090734482, "time": 0.7959485054016113, "epoch": 1, "memory": 28798, "step": 2400} +{"lr": 0.0001, "data_time": 0.3608128070831299, "grad_norm": 7.58581326007843, "loss": 0.35188140869140627, "time": 0.8811181306838989, "epoch": 1, "memory": 28798, "step": 2500} +{"lr": 0.0001, "data_time": 0.3699070453643799, "grad_norm": 16.285736203193665, "loss": 0.3764201730489731, "time": 0.8774486780166626, "epoch": 1, "memory": 28798, "step": 2600} +{"lr": 0.0001, "data_time": 0.3131943941116333, "grad_norm": 7.610110890865326, "loss": 0.3433153510093689, "time": 0.8198633193969727, "epoch": 1, "memory": 28798, "step": 2700} +{"lr": 0.0001, "data_time": 0.32328999042510986, "grad_norm": 7.780367851257324, "loss": 0.3407628029584885, "time": 0.834944486618042, "epoch": 1, "memory": 28798, "step": 2800} +{"lr": 0.0001, "data_time": 0.28146862983703613, "grad_norm": 13.41350474357605, "loss": 0.35408104956150055, "time": 0.7901511430740357, "epoch": 1, "memory": 28798, "step": 2900} +{"lr": 0.0001, "data_time": 0.26748154163360593, "grad_norm": 13.485143852233886, "loss": 0.34326085448265076, "time": 0.7796675682067871, "epoch": 1, "memory": 28798, "step": 3000} +{"lr": 0.0001, "data_time": 0.13624439239501954, "grad_norm": 5.669486069679261, "loss": 0.34299163818359374, "time": 0.756018590927124, "epoch": 1, "memory": 28798, "step": 3100} +{"lr": 0.0001, "data_time": 0.0024565696716308595, "grad_norm": 23.513097500801088, "loss": 0.37924374639987946, "time": 0.788448691368103, "epoch": 1, "memory": 28798, "step": 3200} +{"lr": 0.0001, "data_time": 0.0030142545700073244, "grad_norm": 6.326855874061584, "loss": 0.3306167900562286, "time": 0.8394367694854736, "epoch": 1, "memory": 28798, "step": 3300} +{"lr": 0.0001, "data_time": 0.002055859565734863, "grad_norm": 20.507438039779665, "loss": 0.34681891798973086, "time": 0.7953815937042237, "epoch": 1, "memory": 28798, "step": 3400} +{"lr": 0.0001, "data_time": 0.0022616863250732424, "grad_norm": 9.20704482793808, "loss": 0.32901549637317656, "time": 0.7938506603240967, "epoch": 1, "memory": 28798, "step": 3500} +{"lr": 0.0001, "data_time": 0.0021247625350952148, "grad_norm": 8.380255317687988, "loss": 0.3226520329713821, "time": 0.842173433303833, "epoch": 1, "memory": 28798, "step": 3600} +{"lr": 0.0001, "data_time": 0.0018197059631347655, "grad_norm": 15.262038326263427, "loss": 0.3456566572189331, "time": 0.7844484567642211, "epoch": 1, "memory": 28798, "step": 3700} +{"lr": 0.0001, "data_time": 0.0018650293350219727, "grad_norm": 7.037445616722107, "loss": 0.3086604744195938, "time": 0.8164170503616333, "epoch": 1, "memory": 28798, "step": 3800} +{"lr": 0.0001, "data_time": 0.002668619155883789, "grad_norm": 16.276074934005738, "loss": 0.3058915615081787, "time": 0.8042102575302124, "epoch": 1, "memory": 28798, "step": 3900} +{"lr": 0.0001, "data_time": 0.0019179821014404298, "grad_norm": 11.909996354579926, "loss": 0.3201518923044205, "time": 0.7415214776992798, "epoch": 1, "memory": 28798, "step": 4000} +{"lr": 0.0001, "data_time": 0.0019809246063232423, "grad_norm": 12.117259097099303, "loss": 0.32160766124725343, "time": 0.8137803077697754, "epoch": 1, "memory": 28798, "step": 4100} +{"lr": 0.0001, "data_time": 0.001900959014892578, "grad_norm": 11.986530435085296, "loss": 0.31881475150585176, "time": 0.7554236888885498, "epoch": 1, "memory": 28798, "step": 4200} +{"lr": 0.0001, "data_time": 0.001946234703063965, "grad_norm": 24.562296509742737, "loss": 0.3639715313911438, "time": 0.7606828451156616, "epoch": 1, "memory": 28798, "step": 4300} diff --git a/swin_small_4xb256_fake5m_lr1e-4_10epoch/20230601_145758/vis_data/config.py b/swin_small_4xb256_fake5m_lr1e-4_10epoch/20230601_145758/vis_data/config.py new file mode 100644 index 0000000000000000000000000000000000000000..07c9f4bb104a3c4f894e7a05987a8354fd463e1f --- /dev/null +++ b/swin_small_4xb256_fake5m_lr1e-4_10epoch/20230601_145758/vis_data/config.py @@ -0,0 +1,175 @@ +optim_wrapper = dict( + optimizer=dict( + type='AdamW', + lr=0.0001, + weight_decay=0.05, + eps=1e-08, + betas=(0.9, 0.999), + _scope_='mmpretrain'), + paramwise_cfg=dict( + norm_decay_mult=0.0, + bias_decay_mult=0.0, + flat_decay_mult=0.0, + custom_keys=dict({ + '.absolute_pos_embed': dict(decay_mult=0.0), + '.relative_position_bias_table': dict(decay_mult=0.0) + })), + type='AmpOptimWrapper', + dtype='bfloat16', + clip_grad=dict(max_norm=5.0)) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=True, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=1024) +model = dict( + type='ImageClassifier', + backbone=dict( + type='SwinTransformer', arch='small', img_size=224, + drop_path_rate=0.3), + neck=dict(type='GlobalAveragePooling'), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=768, + init_cfg=None, + loss=dict( + type='LabelSmoothLoss', label_smooth_val=0.1, mode='original'), + cal_acc=False), + init_cfg=[ + dict(type='TruncNormal', layer='Linear', std=0.02, bias=0.0), + dict(type='Constant', layer='LayerNorm', val=1.0, bias=0.0) + ], + train_cfg=None) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stablediffusionV1-5R2-dpmsolver-25-5m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/swin_small_4xb256_fake5m_lr1e-4_10epoch' diff --git a/swin_small_4xb256_fake5m_lr1e-4_10epoch/20230601_145758/vis_data/events.out.tfevents.1685602685.SH-IDC1-10-140-24-19.180503.0 b/swin_small_4xb256_fake5m_lr1e-4_10epoch/20230601_145758/vis_data/events.out.tfevents.1685602685.SH-IDC1-10-140-24-19.180503.0 new file mode 100644 index 0000000000000000000000000000000000000000..54b0fe3ecf95a61c26746f44184821fd2c8ab62a --- /dev/null +++ b/swin_small_4xb256_fake5m_lr1e-4_10epoch/20230601_145758/vis_data/events.out.tfevents.1685602685.SH-IDC1-10-140-24-19.180503.0 @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:cb723609e928898c8d116a651a98cf389ceac760a73296345ecb5fd942339a3e +size 19213 diff --git a/swin_small_4xb256_fake5m_lr1e-4_10epoch/20230601_145758/vis_data/scalars.json b/swin_small_4xb256_fake5m_lr1e-4_10epoch/20230601_145758/vis_data/scalars.json new file mode 100644 index 0000000000000000000000000000000000000000..949cd8b7676f7558b712b2ebfea04d7dde1454a4 --- /dev/null +++ b/swin_small_4xb256_fake5m_lr1e-4_10epoch/20230601_145758/vis_data/scalars.json @@ -0,0 +1,43 @@ +{"lr": 0.0001, "data_time": 0.002135610580444336, "grad_norm": 0.6704509288072587, "loss": 0.6062085151672363, "time": 0.7762885808944702, "epoch": 1, "memory": 28798, "step": 100} +{"lr": 0.0001, "data_time": 0.0017275571823120116, "grad_norm": 1.1625975847244263, "loss": 0.5713563799858093, "time": 0.6906431674957275, "epoch": 1, "memory": 28798, "step": 200} +{"lr": 0.0001, "data_time": 0.0014902591705322266, "grad_norm": 1.561329984664917, "loss": 0.5300649046897888, "time": 0.7719131469726562, "epoch": 1, "memory": 28798, "step": 300} +{"lr": 0.0001, "data_time": 0.0018556356430053712, "grad_norm": 2.1139844059944153, "loss": 0.5165584325790405, "time": 0.7412705183029175, "epoch": 1, "memory": 28798, "step": 400} +{"lr": 0.0001, "data_time": 0.0018137454986572265, "grad_norm": 3.2185787320137025, "loss": 0.5097117364406586, "time": 0.7377352714538574, "epoch": 1, "memory": 28798, "step": 500} +{"lr": 0.0001, "data_time": 0.002111053466796875, "grad_norm": 2.162058413028717, "loss": 0.474714520573616, "time": 0.7309098482131958, "epoch": 1, "memory": 28798, "step": 600} +{"lr": 0.0001, "data_time": 0.01737983226776123, "grad_norm": 3.3910586714744566, "loss": 0.45162872076034544, "time": 0.7392367601394654, "epoch": 1, "memory": 28798, "step": 700} +{"lr": 0.0001, "data_time": 0.1882148027420044, "grad_norm": 5.221115052700043, "loss": 0.45787692070007324, "time": 0.746848726272583, "epoch": 1, "memory": 28798, "step": 800} +{"lr": 0.0001, "data_time": 0.15573256015777587, "grad_norm": 4.696902573108673, "loss": 0.44515292942523954, "time": 0.7830199241638184, "epoch": 1, "memory": 28798, "step": 900} +{"lr": 0.0001, "data_time": 0.001980447769165039, "grad_norm": 5.224834561347961, "loss": 0.41580674350261687, "time": 0.8109351396560669, "epoch": 1, "memory": 28798, "step": 1000} +{"lr": 0.0001, "data_time": 0.07847323417663574, "grad_norm": 5.684916365146637, "loss": 0.4143247872591019, "time": 0.8031377553939819, "epoch": 1, "memory": 28798, "step": 1100} +{"lr": 0.0001, "data_time": 0.17131073474884034, "grad_norm": 6.545488166809082, "loss": 0.4373924136161804, "time": 0.8168726205825806, "epoch": 1, "memory": 28798, "step": 1200} +{"lr": 0.0001, "data_time": 0.148163366317749, "grad_norm": 5.520623433589935, "loss": 0.42006475627422335, "time": 0.8478182792663574, "epoch": 1, "memory": 28798, "step": 1300} +{"lr": 0.0001, "data_time": 0.052912020683288576, "grad_norm": 4.88360276222229, "loss": 0.4077721267938614, "time": 0.8419150590896607, "epoch": 1, "memory": 28798, "step": 1400} +{"lr": 0.0001, "data_time": 0.2666065454483032, "grad_norm": 7.791791701316834, "loss": 0.393210631608963, "time": 0.7989365577697753, "epoch": 1, "memory": 28798, "step": 1500} +{"lr": 0.0001, "data_time": 0.24985790252685547, "grad_norm": 11.681429147720337, "loss": 0.43874755799770354, "time": 0.8331817150115967, "epoch": 1, "memory": 28798, "step": 1600} +{"lr": 0.0001, "data_time": 0.14540016651153564, "grad_norm": 6.859038209915161, "loss": 0.3928732484579086, "time": 0.7855009317398072, "epoch": 1, "memory": 28798, "step": 1700} +{"lr": 0.0001, "data_time": 0.31921954154968263, "grad_norm": 5.1235524892807005, "loss": 0.38636680841445925, "time": 0.8223020315170289, "epoch": 1, "memory": 28798, "step": 1800} +{"lr": 0.0001, "data_time": 0.3836942672729492, "grad_norm": 8.244262409210204, "loss": 0.37420834600925446, "time": 0.892802619934082, "epoch": 1, "memory": 28798, "step": 1900} +{"lr": 0.0001, "data_time": 0.2840893268585205, "grad_norm": 7.060136818885804, "loss": 0.393887659907341, "time": 0.8034720659255982, "epoch": 1, "memory": 28798, "step": 2000} +{"lr": 0.0001, "data_time": 0.24361073970794678, "grad_norm": 16.203964400291444, "loss": 0.4162459999322891, "time": 0.748774790763855, "epoch": 1, "memory": 28798, "step": 2100} +{"lr": 0.0001, "data_time": 0.22101283073425293, "grad_norm": 6.91764805316925, "loss": 0.36337478160858155, "time": 0.7401018619537354, "epoch": 1, "memory": 28798, "step": 2200} +{"lr": 0.0001, "data_time": 0.37143445014953613, "grad_norm": 8.017024970054626, "loss": 0.36236098408699036, "time": 0.8812745094299317, "epoch": 1, "memory": 28798, "step": 2300} +{"lr": 0.0001, "data_time": 0.28904483318328855, "grad_norm": 8.512488317489623, "loss": 0.3735008090734482, "time": 0.7959485054016113, "epoch": 1, "memory": 28798, "step": 2400} +{"lr": 0.0001, "data_time": 0.3608128070831299, "grad_norm": 7.58581326007843, "loss": 0.35188140869140627, "time": 0.8811181306838989, "epoch": 1, "memory": 28798, "step": 2500} +{"lr": 0.0001, "data_time": 0.3699070453643799, "grad_norm": 16.285736203193665, "loss": 0.3764201730489731, "time": 0.8774486780166626, "epoch": 1, "memory": 28798, "step": 2600} +{"lr": 0.0001, "data_time": 0.3131943941116333, "grad_norm": 7.610110890865326, "loss": 0.3433153510093689, "time": 0.8198633193969727, "epoch": 1, "memory": 28798, "step": 2700} +{"lr": 0.0001, "data_time": 0.32328999042510986, "grad_norm": 7.780367851257324, "loss": 0.3407628029584885, "time": 0.834944486618042, "epoch": 1, "memory": 28798, "step": 2800} +{"lr": 0.0001, "data_time": 0.28146862983703613, "grad_norm": 13.41350474357605, "loss": 0.35408104956150055, "time": 0.7901511430740357, "epoch": 1, "memory": 28798, "step": 2900} +{"lr": 0.0001, "data_time": 0.26748154163360593, "grad_norm": 13.485143852233886, "loss": 0.34326085448265076, "time": 0.7796675682067871, "epoch": 1, "memory": 28798, "step": 3000} +{"lr": 0.0001, "data_time": 0.13624439239501954, "grad_norm": 5.669486069679261, "loss": 0.34299163818359374, "time": 0.756018590927124, "epoch": 1, "memory": 28798, "step": 3100} +{"lr": 0.0001, "data_time": 0.0024565696716308595, "grad_norm": 23.513097500801088, "loss": 0.37924374639987946, "time": 0.788448691368103, "epoch": 1, "memory": 28798, "step": 3200} +{"lr": 0.0001, "data_time": 0.0030142545700073244, "grad_norm": 6.326855874061584, "loss": 0.3306167900562286, "time": 0.8394367694854736, "epoch": 1, "memory": 28798, "step": 3300} +{"lr": 0.0001, "data_time": 0.002055859565734863, "grad_norm": 20.507438039779665, "loss": 0.34681891798973086, "time": 0.7953815937042237, "epoch": 1, "memory": 28798, "step": 3400} +{"lr": 0.0001, "data_time": 0.0022616863250732424, "grad_norm": 9.20704482793808, "loss": 0.32901549637317656, "time": 0.7938506603240967, "epoch": 1, "memory": 28798, "step": 3500} +{"lr": 0.0001, "data_time": 0.0021247625350952148, "grad_norm": 8.380255317687988, "loss": 0.3226520329713821, "time": 0.842173433303833, "epoch": 1, "memory": 28798, "step": 3600} +{"lr": 0.0001, "data_time": 0.0018197059631347655, "grad_norm": 15.262038326263427, "loss": 0.3456566572189331, "time": 0.7844484567642211, "epoch": 1, "memory": 28798, "step": 3700} +{"lr": 0.0001, "data_time": 0.0018650293350219727, "grad_norm": 7.037445616722107, "loss": 0.3086604744195938, "time": 0.8164170503616333, "epoch": 1, "memory": 28798, "step": 3800} +{"lr": 0.0001, "data_time": 0.002668619155883789, "grad_norm": 16.276074934005738, "loss": 0.3058915615081787, "time": 0.8042102575302124, "epoch": 1, "memory": 28798, "step": 3900} +{"lr": 0.0001, "data_time": 0.0019179821014404298, "grad_norm": 11.909996354579926, "loss": 0.3201518923044205, "time": 0.7415214776992798, "epoch": 1, "memory": 28798, "step": 4000} +{"lr": 0.0001, "data_time": 0.0019809246063232423, "grad_norm": 12.117259097099303, "loss": 0.32160766124725343, "time": 0.8137803077697754, "epoch": 1, "memory": 28798, "step": 4100} +{"lr": 0.0001, "data_time": 0.001900959014892578, "grad_norm": 11.986530435085296, "loss": 0.31881475150585176, "time": 0.7554236888885498, "epoch": 1, "memory": 28798, "step": 4200} +{"lr": 0.0001, "data_time": 0.001946234703063965, "grad_norm": 24.562296509742737, "loss": 0.3639715313911438, "time": 0.7606828451156616, "epoch": 1, "memory": 28798, "step": 4300} diff --git a/swin_small_4xb256_fake5m_lr1e-4_10epoch/20230601_155932/20230601_155932.log b/swin_small_4xb256_fake5m_lr1e-4_10epoch/20230601_155932/20230601_155932.log new file mode 100644 index 0000000000000000000000000000000000000000..f0fb5a81eba2d6f2a436bbc1f20207dc1aec4149 --- /dev/null +++ b/swin_small_4xb256_fake5m_lr1e-4_10epoch/20230601_155932/20230601_155932.log @@ -0,0 +1,1601 @@ +2023/06/01 15:59:37 - mmengine - INFO - +------------------------------------------------------------ +System environment: + sys.platform: linux + Python: 3.10.9 (main, Mar 8 2023, 10:47:38) [GCC 11.2.0] + CUDA available: True + numpy_random_seed: 1756306841 + GPU 0,1,2,3: NVIDIA A100-SXM4-80GB + CUDA_HOME: /mnt/petrelfs/share/cuda-11.6 + NVCC: Cuda compilation tools, release 11.6, V11.6.124 + GCC: gcc (GCC) 4.8.5 20150623 (Red Hat 4.8.5-44) + PyTorch: 1.13.1 + PyTorch compiling details: PyTorch built with: + - GCC 9.3 + - C++ Version: 201402 + - Intel(R) oneAPI Math Kernel Library Version 2021.4-Product Build 20210904 for Intel(R) 64 architecture applications + - Intel(R) MKL-DNN v2.6.0 (Git Hash 52b5f107dd9cf10910aaa19cb47f3abf9b349815) + - OpenMP 201511 (a.k.a. OpenMP 4.5) + - LAPACK is enabled (usually provided by MKL) + - NNPACK is enabled + - CPU capability usage: AVX2 + - CUDA Runtime 11.6 + - NVCC architecture flags: -gencode;arch=compute_37,code=sm_37;-gencode;arch=compute_50,code=sm_50;-gencode;arch=compute_60,code=sm_60;-gencode;arch=compute_61,code=sm_61;-gencode;arch=compute_70,code=sm_70;-gencode;arch=compute_75,code=sm_75;-gencode;arch=compute_80,code=sm_80;-gencode;arch=compute_86,code=sm_86;-gencode;arch=compute_37,code=compute_37 + - CuDNN 8.3.2 (built against CUDA 11.5) + - Magma 2.6.1 + - Build settings: BLAS_INFO=mkl, BUILD_TYPE=Release, CUDA_VERSION=11.6, CUDNN_VERSION=8.3.2, CXX_COMPILER=/opt/rh/devtoolset-9/root/usr/bin/c++, CXX_FLAGS= -fabi-version=11 -Wno-deprecated -fvisibility-inlines-hidden -DUSE_PTHREADPOOL -fopenmp -DNDEBUG -DUSE_KINETO -DUSE_FBGEMM -DUSE_QNNPACK -DUSE_PYTORCH_QNNPACK -DUSE_XNNPACK -DSYMBOLICATE_MOBILE_DEBUG_HANDLE -DEDGE_PROFILER_USE_KINETO -O2 -fPIC -Wno-narrowing -Wall -Wextra -Werror=return-type -Werror=non-virtual-dtor -Wno-missing-field-initializers -Wno-type-limits -Wno-array-bounds -Wno-unknown-pragmas -Wunused-local-typedefs -Wno-unused-parameter -Wno-unused-function -Wno-unused-result -Wno-strict-overflow -Wno-strict-aliasing -Wno-error=deprecated-declarations -Wno-stringop-overflow -Wno-psabi -Wno-error=pedantic -Wno-error=redundant-decls -Wno-error=old-style-cast -fdiagnostics-color=always -faligned-new -Wno-unused-but-set-variable -Wno-maybe-uninitialized -fno-math-errno -fno-trapping-math -Werror=format -Werror=cast-function-type -Wno-stringop-overflow, LAPACK_INFO=mkl, PERF_WITH_AVX=1, PERF_WITH_AVX2=1, PERF_WITH_AVX512=1, TORCH_VERSION=1.13.1, USE_CUDA=ON, USE_CUDNN=ON, USE_EXCEPTION_PTR=1, USE_GFLAGS=OFF, USE_GLOG=OFF, USE_MKL=ON, USE_MKLDNN=ON, USE_MPI=OFF, USE_NCCL=ON, USE_NNPACK=ON, USE_OPENMP=ON, USE_ROCM=OFF, + + TorchVision: 0.14.1 + OpenCV: 4.7.0 + MMEngine: 0.7.3 + +Runtime environment: + cudnn_benchmark: True + mp_cfg: {'mp_start_method': 'fork', 'opencv_num_threads': 0} + dist_cfg: {'backend': 'nccl'} + seed: None + deterministic: False + Distributed launcher: slurm + Distributed training: True + GPU number: 4 +------------------------------------------------------------ + +2023/06/01 15:59:37 - mmengine - INFO - Config: +optim_wrapper = dict( + optimizer=dict( + type='AdamW', + lr=0.0001, + weight_decay=0.05, + eps=1e-08, + betas=(0.9, 0.999), + _scope_='mmpretrain'), + paramwise_cfg=dict( + norm_decay_mult=0.0, + bias_decay_mult=0.0, + flat_decay_mult=0.0, + custom_keys=dict({ + '.absolute_pos_embed': dict(decay_mult=0.0), + '.relative_position_bias_table': dict(decay_mult=0.0) + })), + type='AmpOptimWrapper', + dtype='bfloat16', + clip_grad=dict(max_norm=5.0)) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=True, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=1024) +model = dict( + type='ImageClassifier', + backbone=dict( + type='SwinTransformer', arch='small', img_size=224, + drop_path_rate=0.3), + neck=dict(type='GlobalAveragePooling'), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=768, + init_cfg=None, + loss=dict( + type='LabelSmoothLoss', label_smooth_val=0.1, mode='original'), + cal_acc=False), + init_cfg=[ + dict(type='TruncNormal', layer='Linear', std=0.02, bias=0.0), + dict(type='Constant', layer='LayerNorm', val=1.0, bias=0.0) + ], + train_cfg=None) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stablediffusionV1-5R2-dpmsolver-25-5m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/swin_small_4xb256_fake5m_lr1e-4_10epoch' + +2023/06/01 15:59:41 - mmengine - INFO - Hooks will be executed in the following order: +before_run: +(VERY_HIGH ) RuntimeInfoHook +(BELOW_NORMAL) LoggerHook + -------------------- +before_train: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_train_epoch: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(NORMAL ) DistSamplerSeedHook + -------------------- +before_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook + -------------------- +after_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +after_train_epoch: +(NORMAL ) IterTimerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_val_epoch: +(NORMAL ) IterTimerHook + -------------------- +before_val_iter: +(NORMAL ) IterTimerHook + -------------------- +after_val_iter: +(NORMAL ) IterTimerHook +(NORMAL ) VisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_val_epoch: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +after_train: +(VERY_LOW ) CheckpointHook + -------------------- +before_test_epoch: +(NORMAL ) IterTimerHook + -------------------- +before_test_iter: +(NORMAL ) IterTimerHook + -------------------- +after_test_iter: +(NORMAL ) IterTimerHook +(NORMAL ) VisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_test_epoch: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_run: +(BELOW_NORMAL) LoggerHook + -------------------- +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.patch_embed.projection.bias:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.patch_embed.norm.weight:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.patch_embed.norm.bias:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.0.blocks.0.norm1.weight:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.0.blocks.0.norm1.bias:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.0.blocks.0.attn.w_msa.relative_position_bias_table:lr=0.0001 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.0.blocks.0.attn.w_msa.relative_position_bias_table:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.0.blocks.0.attn.w_msa.relative_position_bias_table:decay_mult=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.0.blocks.0.attn.w_msa.qkv.bias:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.0.blocks.0.attn.w_msa.proj.bias:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.0.blocks.0.norm2.weight:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.0.blocks.0.norm2.bias:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.0.blocks.0.ffn.layers.0.0.bias:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.0.blocks.0.ffn.layers.1.bias:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.0.blocks.1.norm1.weight:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.0.blocks.1.norm1.bias:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.0.blocks.1.attn.w_msa.relative_position_bias_table:lr=0.0001 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.0.blocks.1.attn.w_msa.relative_position_bias_table:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.0.blocks.1.attn.w_msa.relative_position_bias_table:decay_mult=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.0.blocks.1.attn.w_msa.qkv.bias:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.0.blocks.1.attn.w_msa.proj.bias:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.0.blocks.1.norm2.weight:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.0.blocks.1.norm2.bias:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.0.blocks.1.ffn.layers.0.0.bias:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.0.blocks.1.ffn.layers.1.bias:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.0.downsample.norm.weight:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.0.downsample.norm.bias:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.1.blocks.0.norm1.weight:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.1.blocks.0.norm1.bias:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.1.blocks.0.attn.w_msa.relative_position_bias_table:lr=0.0001 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.1.blocks.0.attn.w_msa.relative_position_bias_table:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.1.blocks.0.attn.w_msa.relative_position_bias_table:decay_mult=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.1.blocks.0.attn.w_msa.qkv.bias:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.1.blocks.0.attn.w_msa.proj.bias:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.1.blocks.0.norm2.weight:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.1.blocks.0.norm2.bias:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.1.blocks.0.ffn.layers.0.0.bias:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.1.blocks.0.ffn.layers.1.bias:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.1.blocks.1.norm1.weight:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.1.blocks.1.norm1.bias:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.1.blocks.1.attn.w_msa.relative_position_bias_table:lr=0.0001 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.1.blocks.1.attn.w_msa.relative_position_bias_table:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.1.blocks.1.attn.w_msa.relative_position_bias_table:decay_mult=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.1.blocks.1.attn.w_msa.qkv.bias:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.1.blocks.1.attn.w_msa.proj.bias:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.1.blocks.1.norm2.weight:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.1.blocks.1.norm2.bias:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.1.blocks.1.ffn.layers.0.0.bias:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.1.blocks.1.ffn.layers.1.bias:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.1.downsample.norm.weight:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.1.downsample.norm.bias:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.0.norm1.weight:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.0.norm1.bias:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.0.attn.w_msa.relative_position_bias_table:lr=0.0001 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.0.attn.w_msa.relative_position_bias_table:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.0.attn.w_msa.relative_position_bias_table:decay_mult=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.0.attn.w_msa.qkv.bias:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.0.attn.w_msa.proj.bias:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.0.norm2.weight:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.0.norm2.bias:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.0.ffn.layers.0.0.bias:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.0.ffn.layers.1.bias:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.1.norm1.weight:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.1.norm1.bias:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.1.attn.w_msa.relative_position_bias_table:lr=0.0001 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.1.attn.w_msa.relative_position_bias_table:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.1.attn.w_msa.relative_position_bias_table:decay_mult=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.1.attn.w_msa.qkv.bias:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.1.attn.w_msa.proj.bias:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.1.norm2.weight:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.1.norm2.bias:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.1.ffn.layers.0.0.bias:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.1.ffn.layers.1.bias:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.2.norm1.weight:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.2.norm1.bias:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.2.attn.w_msa.relative_position_bias_table:lr=0.0001 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.2.attn.w_msa.relative_position_bias_table:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.2.attn.w_msa.relative_position_bias_table:decay_mult=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.2.attn.w_msa.qkv.bias:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.2.attn.w_msa.proj.bias:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.2.norm2.weight:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.2.norm2.bias:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.2.ffn.layers.0.0.bias:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.2.ffn.layers.1.bias:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.3.norm1.weight:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.3.norm1.bias:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.3.attn.w_msa.relative_position_bias_table:lr=0.0001 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.3.attn.w_msa.relative_position_bias_table:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.3.attn.w_msa.relative_position_bias_table:decay_mult=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.3.attn.w_msa.qkv.bias:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.3.attn.w_msa.proj.bias:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.3.norm2.weight:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.3.norm2.bias:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.3.ffn.layers.0.0.bias:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.3.ffn.layers.1.bias:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.4.norm1.weight:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.4.norm1.bias:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.4.attn.w_msa.relative_position_bias_table:lr=0.0001 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.4.attn.w_msa.relative_position_bias_table:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.4.attn.w_msa.relative_position_bias_table:decay_mult=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.4.attn.w_msa.qkv.bias:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.4.attn.w_msa.proj.bias:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.4.norm2.weight:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.4.norm2.bias:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.4.ffn.layers.0.0.bias:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.4.ffn.layers.1.bias:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.5.norm1.weight:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.5.norm1.bias:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.5.attn.w_msa.relative_position_bias_table:lr=0.0001 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.5.attn.w_msa.relative_position_bias_table:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.5.attn.w_msa.relative_position_bias_table:decay_mult=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.5.attn.w_msa.qkv.bias:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.5.attn.w_msa.proj.bias:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.5.norm2.weight:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.5.norm2.bias:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.5.ffn.layers.0.0.bias:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.5.ffn.layers.1.bias:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.6.norm1.weight:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.6.norm1.bias:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.6.attn.w_msa.relative_position_bias_table:lr=0.0001 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.6.attn.w_msa.relative_position_bias_table:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.6.attn.w_msa.relative_position_bias_table:decay_mult=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.6.attn.w_msa.qkv.bias:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.6.attn.w_msa.proj.bias:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.6.norm2.weight:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.6.norm2.bias:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.6.ffn.layers.0.0.bias:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.6.ffn.layers.1.bias:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.7.norm1.weight:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.7.norm1.bias:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.7.attn.w_msa.relative_position_bias_table:lr=0.0001 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.7.attn.w_msa.relative_position_bias_table:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.7.attn.w_msa.relative_position_bias_table:decay_mult=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.7.attn.w_msa.qkv.bias:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.7.attn.w_msa.proj.bias:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.7.norm2.weight:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.7.norm2.bias:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.7.ffn.layers.0.0.bias:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.7.ffn.layers.1.bias:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.8.norm1.weight:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.8.norm1.bias:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.8.attn.w_msa.relative_position_bias_table:lr=0.0001 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.8.attn.w_msa.relative_position_bias_table:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.8.attn.w_msa.relative_position_bias_table:decay_mult=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.8.attn.w_msa.qkv.bias:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.8.attn.w_msa.proj.bias:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.8.norm2.weight:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.8.norm2.bias:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.8.ffn.layers.0.0.bias:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.8.ffn.layers.1.bias:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.9.norm1.weight:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.9.norm1.bias:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.9.attn.w_msa.relative_position_bias_table:lr=0.0001 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.9.attn.w_msa.relative_position_bias_table:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.9.attn.w_msa.relative_position_bias_table:decay_mult=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.9.attn.w_msa.qkv.bias:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.9.attn.w_msa.proj.bias:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.9.norm2.weight:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.9.norm2.bias:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.9.ffn.layers.0.0.bias:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.9.ffn.layers.1.bias:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.10.norm1.weight:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.10.norm1.bias:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.10.attn.w_msa.relative_position_bias_table:lr=0.0001 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.10.attn.w_msa.relative_position_bias_table:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.10.attn.w_msa.relative_position_bias_table:decay_mult=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.10.attn.w_msa.qkv.bias:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.10.attn.w_msa.proj.bias:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.10.norm2.weight:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.10.norm2.bias:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.10.ffn.layers.0.0.bias:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.10.ffn.layers.1.bias:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.11.norm1.weight:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.11.norm1.bias:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.11.attn.w_msa.relative_position_bias_table:lr=0.0001 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.11.attn.w_msa.relative_position_bias_table:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.11.attn.w_msa.relative_position_bias_table:decay_mult=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.11.attn.w_msa.qkv.bias:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.11.attn.w_msa.proj.bias:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.11.norm2.weight:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.11.norm2.bias:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.11.ffn.layers.0.0.bias:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.11.ffn.layers.1.bias:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.12.norm1.weight:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.12.norm1.bias:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.12.attn.w_msa.relative_position_bias_table:lr=0.0001 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.12.attn.w_msa.relative_position_bias_table:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.12.attn.w_msa.relative_position_bias_table:decay_mult=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.12.attn.w_msa.qkv.bias:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.12.attn.w_msa.proj.bias:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.12.norm2.weight:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.12.norm2.bias:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.12.ffn.layers.0.0.bias:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.12.ffn.layers.1.bias:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.13.norm1.weight:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.13.norm1.bias:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.13.attn.w_msa.relative_position_bias_table:lr=0.0001 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.13.attn.w_msa.relative_position_bias_table:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.13.attn.w_msa.relative_position_bias_table:decay_mult=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.13.attn.w_msa.qkv.bias:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.13.attn.w_msa.proj.bias:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.13.norm2.weight:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.13.norm2.bias:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.13.ffn.layers.0.0.bias:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.13.ffn.layers.1.bias:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.14.norm1.weight:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.14.norm1.bias:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.14.attn.w_msa.relative_position_bias_table:lr=0.0001 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.14.attn.w_msa.relative_position_bias_table:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.14.attn.w_msa.relative_position_bias_table:decay_mult=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.14.attn.w_msa.qkv.bias:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.14.attn.w_msa.proj.bias:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.14.norm2.weight:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.14.norm2.bias:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.14.ffn.layers.0.0.bias:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.14.ffn.layers.1.bias:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.15.norm1.weight:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.15.norm1.bias:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.15.attn.w_msa.relative_position_bias_table:lr=0.0001 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.15.attn.w_msa.relative_position_bias_table:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.15.attn.w_msa.relative_position_bias_table:decay_mult=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.15.attn.w_msa.qkv.bias:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.15.attn.w_msa.proj.bias:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.15.norm2.weight:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.15.norm2.bias:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.15.ffn.layers.0.0.bias:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.15.ffn.layers.1.bias:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.16.norm1.weight:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.16.norm1.bias:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.16.attn.w_msa.relative_position_bias_table:lr=0.0001 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.16.attn.w_msa.relative_position_bias_table:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.16.attn.w_msa.relative_position_bias_table:decay_mult=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.16.attn.w_msa.qkv.bias:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.16.attn.w_msa.proj.bias:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.16.norm2.weight:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.16.norm2.bias:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.16.ffn.layers.0.0.bias:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.16.ffn.layers.1.bias:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.17.norm1.weight:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.17.norm1.bias:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.17.attn.w_msa.relative_position_bias_table:lr=0.0001 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.17.attn.w_msa.relative_position_bias_table:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.17.attn.w_msa.relative_position_bias_table:decay_mult=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.17.attn.w_msa.qkv.bias:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.17.attn.w_msa.proj.bias:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.17.norm2.weight:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.17.norm2.bias:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.17.ffn.layers.0.0.bias:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.17.ffn.layers.1.bias:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.downsample.norm.weight:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.2.downsample.norm.bias:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.3.blocks.0.norm1.weight:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.3.blocks.0.norm1.bias:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.3.blocks.0.attn.w_msa.relative_position_bias_table:lr=0.0001 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.3.blocks.0.attn.w_msa.relative_position_bias_table:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.3.blocks.0.attn.w_msa.relative_position_bias_table:decay_mult=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.3.blocks.0.attn.w_msa.qkv.bias:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.3.blocks.0.attn.w_msa.proj.bias:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.3.blocks.0.norm2.weight:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.3.blocks.0.norm2.bias:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.3.blocks.0.ffn.layers.0.0.bias:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.3.blocks.0.ffn.layers.1.bias:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.3.blocks.1.norm1.weight:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.3.blocks.1.norm1.bias:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.3.blocks.1.attn.w_msa.relative_position_bias_table:lr=0.0001 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.3.blocks.1.attn.w_msa.relative_position_bias_table:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.3.blocks.1.attn.w_msa.relative_position_bias_table:decay_mult=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.3.blocks.1.attn.w_msa.qkv.bias:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.3.blocks.1.attn.w_msa.proj.bias:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.3.blocks.1.norm2.weight:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.3.blocks.1.norm2.bias:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.3.blocks.1.ffn.layers.0.0.bias:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.stages.3.blocks.1.ffn.layers.1.bias:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.norm3.weight:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- backbone.norm3.bias:weight_decay=0.0 +2023/06/01 16:00:15 - mmengine - INFO - paramwise_options -- head.fc.bias:weight_decay=0.0 +Name of parameter - Initialization information + +backbone.patch_embed.projection.weight - torch.Size([96, 3, 4, 4]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.patch_embed.projection.bias - torch.Size([96]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.patch_embed.norm.weight - torch.Size([96]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.patch_embed.norm.bias - torch.Size([96]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.0.blocks.0.norm1.weight - torch.Size([96]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.0.blocks.0.norm1.bias - torch.Size([96]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.0.blocks.0.attn.w_msa.relative_position_bias_table - torch.Size([169, 3]): +Initialized by user-defined `init_weights` in WindowMSA + +backbone.stages.0.blocks.0.attn.w_msa.qkv.weight - torch.Size([288, 96]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.blocks.0.attn.w_msa.qkv.bias - torch.Size([288]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.blocks.0.attn.w_msa.proj.weight - torch.Size([96, 96]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.blocks.0.attn.w_msa.proj.bias - torch.Size([96]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.blocks.0.norm2.weight - torch.Size([96]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.0.blocks.0.norm2.bias - torch.Size([96]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.0.blocks.0.ffn.layers.0.0.weight - torch.Size([384, 96]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.blocks.0.ffn.layers.0.0.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.blocks.0.ffn.layers.1.weight - torch.Size([96, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.blocks.0.ffn.layers.1.bias - torch.Size([96]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.blocks.1.norm1.weight - torch.Size([96]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.0.blocks.1.norm1.bias - torch.Size([96]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.0.blocks.1.attn.w_msa.relative_position_bias_table - torch.Size([169, 3]): +Initialized by user-defined `init_weights` in WindowMSA + +backbone.stages.0.blocks.1.attn.w_msa.qkv.weight - torch.Size([288, 96]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.blocks.1.attn.w_msa.qkv.bias - torch.Size([288]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.blocks.1.attn.w_msa.proj.weight - torch.Size([96, 96]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.blocks.1.attn.w_msa.proj.bias - torch.Size([96]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.blocks.1.norm2.weight - torch.Size([96]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.0.blocks.1.norm2.bias - torch.Size([96]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.0.blocks.1.ffn.layers.0.0.weight - torch.Size([384, 96]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.blocks.1.ffn.layers.0.0.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.blocks.1.ffn.layers.1.weight - torch.Size([96, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.blocks.1.ffn.layers.1.bias - torch.Size([96]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.downsample.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.0.downsample.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.0.downsample.reduction.weight - torch.Size([192, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.blocks.0.norm1.weight - torch.Size([192]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.1.blocks.0.norm1.bias - torch.Size([192]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.1.blocks.0.attn.w_msa.relative_position_bias_table - torch.Size([169, 6]): +Initialized by user-defined `init_weights` in WindowMSA + +backbone.stages.1.blocks.0.attn.w_msa.qkv.weight - torch.Size([576, 192]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.blocks.0.attn.w_msa.qkv.bias - torch.Size([576]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.blocks.0.attn.w_msa.proj.weight - torch.Size([192, 192]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.blocks.0.attn.w_msa.proj.bias - torch.Size([192]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.blocks.0.norm2.weight - torch.Size([192]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.1.blocks.0.norm2.bias - torch.Size([192]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.1.blocks.0.ffn.layers.0.0.weight - torch.Size([768, 192]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.blocks.0.ffn.layers.0.0.bias - torch.Size([768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.blocks.0.ffn.layers.1.weight - torch.Size([192, 768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.blocks.0.ffn.layers.1.bias - torch.Size([192]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.blocks.1.norm1.weight - torch.Size([192]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.1.blocks.1.norm1.bias - torch.Size([192]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.1.blocks.1.attn.w_msa.relative_position_bias_table - torch.Size([169, 6]): +Initialized by user-defined `init_weights` in WindowMSA + +backbone.stages.1.blocks.1.attn.w_msa.qkv.weight - torch.Size([576, 192]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.blocks.1.attn.w_msa.qkv.bias - torch.Size([576]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.blocks.1.attn.w_msa.proj.weight - torch.Size([192, 192]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.blocks.1.attn.w_msa.proj.bias - torch.Size([192]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.blocks.1.norm2.weight - torch.Size([192]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.1.blocks.1.norm2.bias - torch.Size([192]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.1.blocks.1.ffn.layers.0.0.weight - torch.Size([768, 192]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.blocks.1.ffn.layers.0.0.bias - torch.Size([768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.blocks.1.ffn.layers.1.weight - torch.Size([192, 768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.blocks.1.ffn.layers.1.bias - torch.Size([192]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.downsample.norm.weight - torch.Size([768]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.1.downsample.norm.bias - torch.Size([768]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.1.downsample.reduction.weight - torch.Size([384, 768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.0.norm1.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.0.norm1.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.0.attn.w_msa.relative_position_bias_table - torch.Size([169, 12]): +Initialized by user-defined `init_weights` in WindowMSA + +backbone.stages.2.blocks.0.attn.w_msa.qkv.weight - torch.Size([1152, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.0.attn.w_msa.qkv.bias - torch.Size([1152]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.0.attn.w_msa.proj.weight - torch.Size([384, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.0.attn.w_msa.proj.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.0.norm2.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.0.norm2.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.0.ffn.layers.0.0.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.0.ffn.layers.0.0.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.0.ffn.layers.1.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.0.ffn.layers.1.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.1.norm1.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.1.norm1.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.1.attn.w_msa.relative_position_bias_table - torch.Size([169, 12]): +Initialized by user-defined `init_weights` in WindowMSA + +backbone.stages.2.blocks.1.attn.w_msa.qkv.weight - torch.Size([1152, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.1.attn.w_msa.qkv.bias - torch.Size([1152]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.1.attn.w_msa.proj.weight - torch.Size([384, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.1.attn.w_msa.proj.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.1.norm2.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.1.norm2.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.1.ffn.layers.0.0.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.1.ffn.layers.0.0.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.1.ffn.layers.1.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.1.ffn.layers.1.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.2.norm1.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.2.norm1.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.2.attn.w_msa.relative_position_bias_table - torch.Size([169, 12]): +Initialized by user-defined `init_weights` in WindowMSA + +backbone.stages.2.blocks.2.attn.w_msa.qkv.weight - torch.Size([1152, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.2.attn.w_msa.qkv.bias - torch.Size([1152]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.2.attn.w_msa.proj.weight - torch.Size([384, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.2.attn.w_msa.proj.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.2.norm2.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.2.norm2.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.2.ffn.layers.0.0.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.2.ffn.layers.0.0.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.2.ffn.layers.1.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.2.ffn.layers.1.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.3.norm1.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.3.norm1.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.3.attn.w_msa.relative_position_bias_table - torch.Size([169, 12]): +Initialized by user-defined `init_weights` in WindowMSA + +backbone.stages.2.blocks.3.attn.w_msa.qkv.weight - torch.Size([1152, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.3.attn.w_msa.qkv.bias - torch.Size([1152]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.3.attn.w_msa.proj.weight - torch.Size([384, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.3.attn.w_msa.proj.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.3.norm2.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.3.norm2.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.3.ffn.layers.0.0.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.3.ffn.layers.0.0.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.3.ffn.layers.1.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.3.ffn.layers.1.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.4.norm1.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.4.norm1.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.4.attn.w_msa.relative_position_bias_table - torch.Size([169, 12]): +Initialized by user-defined `init_weights` in WindowMSA + +backbone.stages.2.blocks.4.attn.w_msa.qkv.weight - torch.Size([1152, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.4.attn.w_msa.qkv.bias - torch.Size([1152]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.4.attn.w_msa.proj.weight - torch.Size([384, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.4.attn.w_msa.proj.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.4.norm2.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.4.norm2.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.4.ffn.layers.0.0.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.4.ffn.layers.0.0.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.4.ffn.layers.1.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.4.ffn.layers.1.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.5.norm1.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.5.norm1.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.5.attn.w_msa.relative_position_bias_table - torch.Size([169, 12]): +Initialized by user-defined `init_weights` in WindowMSA + +backbone.stages.2.blocks.5.attn.w_msa.qkv.weight - torch.Size([1152, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.5.attn.w_msa.qkv.bias - torch.Size([1152]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.5.attn.w_msa.proj.weight - torch.Size([384, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.5.attn.w_msa.proj.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.5.norm2.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.5.norm2.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.5.ffn.layers.0.0.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.5.ffn.layers.0.0.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.5.ffn.layers.1.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.5.ffn.layers.1.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.6.norm1.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.6.norm1.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.6.attn.w_msa.relative_position_bias_table - torch.Size([169, 12]): +Initialized by user-defined `init_weights` in WindowMSA + +backbone.stages.2.blocks.6.attn.w_msa.qkv.weight - torch.Size([1152, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.6.attn.w_msa.qkv.bias - torch.Size([1152]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.6.attn.w_msa.proj.weight - torch.Size([384, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.6.attn.w_msa.proj.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.6.norm2.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.6.norm2.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.6.ffn.layers.0.0.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.6.ffn.layers.0.0.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.6.ffn.layers.1.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.6.ffn.layers.1.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.7.norm1.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.7.norm1.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.7.attn.w_msa.relative_position_bias_table - torch.Size([169, 12]): +Initialized by user-defined `init_weights` in WindowMSA + +backbone.stages.2.blocks.7.attn.w_msa.qkv.weight - torch.Size([1152, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.7.attn.w_msa.qkv.bias - torch.Size([1152]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.7.attn.w_msa.proj.weight - torch.Size([384, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.7.attn.w_msa.proj.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.7.norm2.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.7.norm2.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.7.ffn.layers.0.0.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.7.ffn.layers.0.0.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.7.ffn.layers.1.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.7.ffn.layers.1.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.8.norm1.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.8.norm1.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.8.attn.w_msa.relative_position_bias_table - torch.Size([169, 12]): +Initialized by user-defined `init_weights` in WindowMSA + +backbone.stages.2.blocks.8.attn.w_msa.qkv.weight - torch.Size([1152, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.8.attn.w_msa.qkv.bias - torch.Size([1152]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.8.attn.w_msa.proj.weight - torch.Size([384, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.8.attn.w_msa.proj.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.8.norm2.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.8.norm2.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.8.ffn.layers.0.0.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.8.ffn.layers.0.0.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.8.ffn.layers.1.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.8.ffn.layers.1.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.9.norm1.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.9.norm1.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.9.attn.w_msa.relative_position_bias_table - torch.Size([169, 12]): +Initialized by user-defined `init_weights` in WindowMSA + +backbone.stages.2.blocks.9.attn.w_msa.qkv.weight - torch.Size([1152, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.9.attn.w_msa.qkv.bias - torch.Size([1152]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.9.attn.w_msa.proj.weight - torch.Size([384, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.9.attn.w_msa.proj.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.9.norm2.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.9.norm2.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.9.ffn.layers.0.0.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.9.ffn.layers.0.0.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.9.ffn.layers.1.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.9.ffn.layers.1.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.10.norm1.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.10.norm1.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.10.attn.w_msa.relative_position_bias_table - torch.Size([169, 12]): +Initialized by user-defined `init_weights` in WindowMSA + +backbone.stages.2.blocks.10.attn.w_msa.qkv.weight - torch.Size([1152, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.10.attn.w_msa.qkv.bias - torch.Size([1152]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.10.attn.w_msa.proj.weight - torch.Size([384, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.10.attn.w_msa.proj.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.10.norm2.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.10.norm2.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.10.ffn.layers.0.0.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.10.ffn.layers.0.0.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.10.ffn.layers.1.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.10.ffn.layers.1.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.11.norm1.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.11.norm1.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.11.attn.w_msa.relative_position_bias_table - torch.Size([169, 12]): +Initialized by user-defined `init_weights` in WindowMSA + +backbone.stages.2.blocks.11.attn.w_msa.qkv.weight - torch.Size([1152, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.11.attn.w_msa.qkv.bias - torch.Size([1152]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.11.attn.w_msa.proj.weight - torch.Size([384, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.11.attn.w_msa.proj.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.11.norm2.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.11.norm2.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.11.ffn.layers.0.0.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.11.ffn.layers.0.0.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.11.ffn.layers.1.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.11.ffn.layers.1.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.12.norm1.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.12.norm1.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.12.attn.w_msa.relative_position_bias_table - torch.Size([169, 12]): +Initialized by user-defined `init_weights` in WindowMSA + +backbone.stages.2.blocks.12.attn.w_msa.qkv.weight - torch.Size([1152, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.12.attn.w_msa.qkv.bias - torch.Size([1152]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.12.attn.w_msa.proj.weight - torch.Size([384, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.12.attn.w_msa.proj.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.12.norm2.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.12.norm2.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.12.ffn.layers.0.0.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.12.ffn.layers.0.0.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.12.ffn.layers.1.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.12.ffn.layers.1.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.13.norm1.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.13.norm1.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.13.attn.w_msa.relative_position_bias_table - torch.Size([169, 12]): +Initialized by user-defined `init_weights` in WindowMSA + +backbone.stages.2.blocks.13.attn.w_msa.qkv.weight - torch.Size([1152, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.13.attn.w_msa.qkv.bias - torch.Size([1152]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.13.attn.w_msa.proj.weight - torch.Size([384, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.13.attn.w_msa.proj.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.13.norm2.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.13.norm2.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.13.ffn.layers.0.0.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.13.ffn.layers.0.0.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.13.ffn.layers.1.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.13.ffn.layers.1.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.14.norm1.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.14.norm1.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.14.attn.w_msa.relative_position_bias_table - torch.Size([169, 12]): +Initialized by user-defined `init_weights` in WindowMSA + +backbone.stages.2.blocks.14.attn.w_msa.qkv.weight - torch.Size([1152, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.14.attn.w_msa.qkv.bias - torch.Size([1152]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.14.attn.w_msa.proj.weight - torch.Size([384, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.14.attn.w_msa.proj.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.14.norm2.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.14.norm2.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.14.ffn.layers.0.0.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.14.ffn.layers.0.0.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.14.ffn.layers.1.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.14.ffn.layers.1.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.15.norm1.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.15.norm1.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.15.attn.w_msa.relative_position_bias_table - torch.Size([169, 12]): +Initialized by user-defined `init_weights` in WindowMSA + +backbone.stages.2.blocks.15.attn.w_msa.qkv.weight - torch.Size([1152, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.15.attn.w_msa.qkv.bias - torch.Size([1152]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.15.attn.w_msa.proj.weight - torch.Size([384, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.15.attn.w_msa.proj.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.15.norm2.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.15.norm2.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.15.ffn.layers.0.0.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.15.ffn.layers.0.0.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.15.ffn.layers.1.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.15.ffn.layers.1.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.16.norm1.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.16.norm1.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.16.attn.w_msa.relative_position_bias_table - torch.Size([169, 12]): +Initialized by user-defined `init_weights` in WindowMSA + +backbone.stages.2.blocks.16.attn.w_msa.qkv.weight - torch.Size([1152, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.16.attn.w_msa.qkv.bias - torch.Size([1152]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.16.attn.w_msa.proj.weight - torch.Size([384, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.16.attn.w_msa.proj.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.16.norm2.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.16.norm2.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.16.ffn.layers.0.0.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.16.ffn.layers.0.0.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.16.ffn.layers.1.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.16.ffn.layers.1.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.17.norm1.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.17.norm1.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.17.attn.w_msa.relative_position_bias_table - torch.Size([169, 12]): +Initialized by user-defined `init_weights` in WindowMSA + +backbone.stages.2.blocks.17.attn.w_msa.qkv.weight - torch.Size([1152, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.17.attn.w_msa.qkv.bias - torch.Size([1152]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.17.attn.w_msa.proj.weight - torch.Size([384, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.17.attn.w_msa.proj.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.17.norm2.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.17.norm2.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.17.ffn.layers.0.0.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.17.ffn.layers.0.0.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.17.ffn.layers.1.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.17.ffn.layers.1.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.downsample.norm.weight - torch.Size([1536]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.downsample.norm.bias - torch.Size([1536]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.downsample.reduction.weight - torch.Size([768, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.blocks.0.norm1.weight - torch.Size([768]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.3.blocks.0.norm1.bias - torch.Size([768]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.3.blocks.0.attn.w_msa.relative_position_bias_table - torch.Size([169, 24]): +Initialized by user-defined `init_weights` in WindowMSA + +backbone.stages.3.blocks.0.attn.w_msa.qkv.weight - torch.Size([2304, 768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.blocks.0.attn.w_msa.qkv.bias - torch.Size([2304]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.blocks.0.attn.w_msa.proj.weight - torch.Size([768, 768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.blocks.0.attn.w_msa.proj.bias - torch.Size([768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.blocks.0.norm2.weight - torch.Size([768]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.3.blocks.0.norm2.bias - torch.Size([768]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.3.blocks.0.ffn.layers.0.0.weight - torch.Size([3072, 768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.blocks.0.ffn.layers.0.0.bias - torch.Size([3072]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.blocks.0.ffn.layers.1.weight - torch.Size([768, 3072]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.blocks.0.ffn.layers.1.bias - torch.Size([768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.blocks.1.norm1.weight - torch.Size([768]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.3.blocks.1.norm1.bias - torch.Size([768]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.3.blocks.1.attn.w_msa.relative_position_bias_table - torch.Size([169, 24]): +Initialized by user-defined `init_weights` in WindowMSA + +backbone.stages.3.blocks.1.attn.w_msa.qkv.weight - torch.Size([2304, 768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.blocks.1.attn.w_msa.qkv.bias - torch.Size([2304]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.blocks.1.attn.w_msa.proj.weight - torch.Size([768, 768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.blocks.1.attn.w_msa.proj.bias - torch.Size([768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.blocks.1.norm2.weight - torch.Size([768]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.3.blocks.1.norm2.bias - torch.Size([768]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.3.blocks.1.ffn.layers.0.0.weight - torch.Size([3072, 768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.blocks.1.ffn.layers.0.0.bias - torch.Size([3072]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.blocks.1.ffn.layers.1.weight - torch.Size([768, 3072]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.blocks.1.ffn.layers.1.bias - torch.Size([768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.norm3.weight - torch.Size([768]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.norm3.bias - torch.Size([768]): +The value is the same before and after calling `init_weights` of ImageClassifier + +head.fc.weight - torch.Size([2, 768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +head.fc.bias - torch.Size([2]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 +2023/06/01 16:00:16 - mmengine - WARNING - "FileClient" will be deprecated in future. Please use io functions in https://mmengine.readthedocs.io/en/latest/api/fileio.html#file-io +2023/06/01 16:00:16 - mmengine - WARNING - "HardDiskBackend" is the alias of "LocalBackend" and the former will be deprecated in future. +2023/06/01 16:00:16 - mmengine - INFO - Checkpoints will be saved to /mnt/petrelfs/luzeyu/workspace/fakebench/mmpretrain/workdir/swin_small_4xb256_fake5m_lr1e-4_10epoch. +2023/06/01 16:01:44 - mmengine - INFO - Epoch(train) [1][ 100/5758] lr: 1.0000e-04 eta: 14:10:22 time: 0.8011 data_time: 0.0016 memory: 28798 grad_norm: 0.6827 loss: 0.5936 +2023/06/01 16:03:06 - mmengine - INFO - Epoch(train) [1][ 200/5758] lr: 1.0000e-04 eta: 13:36:32 time: 0.8667 data_time: 0.0017 memory: 28798 grad_norm: 1.2591 loss: 0.5711 +2023/06/01 16:04:29 - mmengine - INFO - Epoch(train) [1][ 300/5758] lr: 1.0000e-04 eta: 13:24:52 time: 0.8052 data_time: 0.0016 memory: 28798 grad_norm: 1.3795 loss: 0.5229 +2023/06/01 16:05:50 - mmengine - INFO - Epoch(train) [1][ 400/5758] lr: 1.0000e-04 eta: 13:16:45 time: 0.8453 data_time: 0.0017 memory: 28798 grad_norm: 1.8317 loss: 0.5045 +2023/06/01 16:07:13 - mmengine - INFO - Epoch(train) [1][ 500/5758] lr: 1.0000e-04 eta: 13:14:34 time: 0.9132 data_time: 0.0017 memory: 28798 grad_norm: 1.8738 loss: 0.4915 +2023/06/01 16:08:37 - mmengine - INFO - Epoch(train) [1][ 600/5758] lr: 1.0000e-04 eta: 13:12:52 time: 0.8061 data_time: 0.0017 memory: 28798 grad_norm: 4.0890 loss: 0.4914 +2023/06/01 16:10:00 - mmengine - INFO - Epoch(train) [1][ 700/5758] lr: 1.0000e-04 eta: 13:10:46 time: 0.8019 data_time: 0.0025 memory: 28798 grad_norm: 2.1137 loss: 0.4548 +2023/06/01 16:11:22 - mmengine - INFO - Epoch(train) [1][ 800/5758] lr: 1.0000e-04 eta: 13:08:20 time: 0.8234 data_time: 0.0021 memory: 28798 grad_norm: 3.2558 loss: 0.4419 +2023/06/01 16:12:44 - mmengine - INFO - Epoch(train) [1][ 900/5758] lr: 1.0000e-04 eta: 13:05:44 time: 0.8352 data_time: 0.0018 memory: 28798 grad_norm: 6.5980 loss: 0.4632 +2023/06/01 16:14:06 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_lr1e-4_10epoch_20230601_155932 +2023/06/01 16:14:06 - mmengine - INFO - Epoch(train) [1][1000/5758] lr: 1.0000e-04 eta: 13:03:27 time: 0.7668 data_time: 0.0018 memory: 28798 grad_norm: 5.2922 loss: 0.4398 +2023/06/01 16:15:31 - mmengine - INFO - Epoch(train) [1][1100/5758] lr: 1.0000e-04 eta: 13:03:37 time: 0.8853 data_time: 0.0016 memory: 28798 grad_norm: 9.7139 loss: 0.4221 +2023/06/01 16:16:52 - mmengine - INFO - Epoch(train) [1][1200/5758] lr: 1.0000e-04 eta: 12:59:50 time: 0.8224 data_time: 0.0018 memory: 28798 grad_norm: 4.9466 loss: 0.4151 +2023/06/01 16:18:13 - mmengine - INFO - Epoch(train) [1][1300/5758] lr: 1.0000e-04 eta: 12:57:16 time: 0.8189 data_time: 0.0017 memory: 28798 grad_norm: 4.4423 loss: 0.4262 +2023/06/01 16:19:35 - mmengine - INFO - Epoch(train) [1][1400/5758] lr: 1.0000e-04 eta: 12:55:12 time: 0.8028 data_time: 0.0021 memory: 28798 grad_norm: 9.7794 loss: 0.4240 +2023/06/01 16:20:58 - mmengine - INFO - Epoch(train) [1][1500/5758] lr: 1.0000e-04 eta: 12:53:49 time: 0.8397 data_time: 0.0016 memory: 28798 grad_norm: 9.0040 loss: 0.4143 +2023/06/01 16:22:18 - mmengine - INFO - Epoch(train) [1][1600/5758] lr: 1.0000e-04 eta: 12:51:06 time: 0.8200 data_time: 0.0017 memory: 28798 grad_norm: 6.9932 loss: 0.4104 +2023/06/01 16:23:43 - mmengine - INFO - Epoch(train) [1][1700/5758] lr: 1.0000e-04 eta: 12:51:07 time: 0.8770 data_time: 0.0027 memory: 28798 grad_norm: 9.5250 loss: 0.4239 +2023/06/01 16:25:05 - mmengine - INFO - Epoch(train) [1][1800/5758] lr: 1.0000e-04 eta: 12:49:00 time: 0.7856 data_time: 0.0019 memory: 28798 grad_norm: 4.8318 loss: 0.3933 +2023/06/01 16:26:28 - mmengine - INFO - Epoch(train) [1][1900/5758] lr: 1.0000e-04 eta: 12:48:02 time: 0.8747 data_time: 0.0017 memory: 28798 grad_norm: 5.7949 loss: 0.3888 +2023/06/01 16:27:52 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_lr1e-4_10epoch_20230601_155932 +2023/06/01 16:27:52 - mmengine - INFO - Epoch(train) [1][2000/5758] lr: 1.0000e-04 eta: 12:47:14 time: 0.8897 data_time: 0.0021 memory: 28798 grad_norm: 5.9448 loss: 0.3760 +2023/06/01 16:29:17 - mmengine - INFO - Epoch(train) [1][2100/5758] lr: 1.0000e-04 eta: 12:46:35 time: 0.8468 data_time: 0.0020 memory: 28798 grad_norm: 12.6300 loss: 0.3872 +2023/06/01 16:30:42 - mmengine - INFO - Epoch(train) [1][2200/5758] lr: 1.0000e-04 eta: 12:46:01 time: 0.8513 data_time: 0.0017 memory: 28798 grad_norm: 6.4090 loss: 0.3575 +2023/06/01 16:32:03 - mmengine - INFO - Epoch(train) [1][2300/5758] lr: 1.0000e-04 eta: 12:44:11 time: 0.8478 data_time: 0.0020 memory: 28798 grad_norm: 14.5297 loss: 0.3792 +2023/06/01 16:33:26 - mmengine - INFO - Epoch(train) [1][2400/5758] lr: 1.0000e-04 eta: 12:42:51 time: 0.8486 data_time: 0.0020 memory: 28798 grad_norm: 12.9950 loss: 0.3565 +2023/06/01 16:34:50 - mmengine - INFO - Epoch(train) [1][2500/5758] lr: 1.0000e-04 eta: 12:41:41 time: 0.7891 data_time: 0.0016 memory: 28798 grad_norm: 12.4348 loss: 0.3879 +2023/06/01 16:36:11 - mmengine - INFO - Epoch(train) [1][2600/5758] lr: 1.0000e-04 eta: 12:39:40 time: 0.7783 data_time: 0.0016 memory: 28798 grad_norm: 7.6296 loss: 0.3430 +2023/06/01 16:37:32 - mmengine - INFO - Epoch(train) [1][2700/5758] lr: 1.0000e-04 eta: 12:37:28 time: 0.7944 data_time: 0.0018 memory: 28798 grad_norm: 16.5892 loss: 0.3662 +2023/06/01 16:38:54 - mmengine - INFO - Epoch(train) [1][2800/5758] lr: 1.0000e-04 eta: 12:35:50 time: 0.8246 data_time: 0.0020 memory: 28798 grad_norm: 14.6269 loss: 0.3673 +2023/06/01 16:40:19 - mmengine - INFO - Epoch(train) [1][2900/5758] lr: 1.0000e-04 eta: 12:35:18 time: 0.8674 data_time: 0.0017 memory: 28798 grad_norm: 11.5140 loss: 0.3996 +2023/06/01 16:41:41 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_lr1e-4_10epoch_20230601_155932 +2023/06/01 16:41:41 - mmengine - INFO - Epoch(train) [1][3000/5758] lr: 1.0000e-04 eta: 12:33:37 time: 0.8155 data_time: 0.0026 memory: 28798 grad_norm: 6.6478 loss: 0.3243 +2023/06/01 16:43:04 - mmengine - INFO - Epoch(train) [1][3100/5758] lr: 1.0000e-04 eta: 12:32:21 time: 0.8269 data_time: 0.0019 memory: 28798 grad_norm: 6.8281 loss: 0.3577 +2023/06/01 16:44:25 - mmengine - INFO - Epoch(train) [1][3200/5758] lr: 1.0000e-04 eta: 12:30:21 time: 0.7980 data_time: 0.0017 memory: 28798 grad_norm: 9.3759 loss: 0.3282 +2023/06/01 16:45:47 - mmengine - INFO - Epoch(train) [1][3300/5758] lr: 1.0000e-04 eta: 12:28:48 time: 0.8014 data_time: 0.0022 memory: 28798 grad_norm: 23.4084 loss: 0.4111 +2023/06/01 16:47:08 - mmengine - INFO - Epoch(train) [1][3400/5758] lr: 1.0000e-04 eta: 12:26:57 time: 0.7767 data_time: 0.0017 memory: 28798 grad_norm: 13.5821 loss: 0.3274 +2023/06/01 16:48:30 - mmengine - INFO - Epoch(train) [1][3500/5758] lr: 1.0000e-04 eta: 12:25:26 time: 0.8272 data_time: 0.0021 memory: 28798 grad_norm: 4.6718 loss: 0.3217 +2023/06/01 16:49:50 - mmengine - INFO - Epoch(train) [1][3600/5758] lr: 1.0000e-04 eta: 12:23:23 time: 0.8015 data_time: 0.0019 memory: 28798 grad_norm: 4.4943 loss: 0.3060 +2023/06/01 16:51:13 - mmengine - INFO - Epoch(train) [1][3700/5758] lr: 1.0000e-04 eta: 12:22:05 time: 0.7870 data_time: 0.0015 memory: 28798 grad_norm: 4.7809 loss: 0.3073 +2023/06/01 16:52:36 - mmengine - INFO - Epoch(train) [1][3800/5758] lr: 1.0000e-04 eta: 12:20:39 time: 0.8185 data_time: 0.0018 memory: 28798 grad_norm: 9.9156 loss: 0.3181 +2023/06/01 16:53:57 - mmengine - INFO - Epoch(train) [1][3900/5758] lr: 1.0000e-04 eta: 12:19:03 time: 0.8365 data_time: 0.0019 memory: 28798 grad_norm: 7.7289 loss: 0.3020 +2023/06/01 16:55:21 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_lr1e-4_10epoch_20230601_155932 +2023/06/01 16:55:21 - mmengine - INFO - Epoch(train) [1][4000/5758] lr: 1.0000e-04 eta: 12:17:51 time: 0.7956 data_time: 0.0017 memory: 28798 grad_norm: 8.2281 loss: 0.3029 diff --git a/swin_small_4xb256_fake5m_lr1e-4_10epoch/20230601_155932/vis_data/20230601_155932.json b/swin_small_4xb256_fake5m_lr1e-4_10epoch/20230601_155932/vis_data/20230601_155932.json new file mode 100644 index 0000000000000000000000000000000000000000..7647cd746a11a2b7893d6f153da6770a8ebf15dd --- /dev/null +++ b/swin_small_4xb256_fake5m_lr1e-4_10epoch/20230601_155932/vis_data/20230601_155932.json @@ -0,0 +1,40 @@ +{"lr": 0.0001, "data_time": 0.0016358852386474609, "grad_norm": 0.6826717019081116, "loss": 0.5935846924781799, "time": 0.8011452198028565, "epoch": 1, "memory": 28798, "step": 100} +{"lr": 0.0001, "data_time": 0.0017016410827636718, "grad_norm": 1.2591084003448487, "loss": 0.5710584998130799, "time": 0.8666714429855347, "epoch": 1, "memory": 28798, "step": 200} +{"lr": 0.0001, "data_time": 0.001643061637878418, "grad_norm": 1.3794927954673768, "loss": 0.5229363799095154, "time": 0.8052175283432007, "epoch": 1, "memory": 28798, "step": 300} +{"lr": 0.0001, "data_time": 0.0016553878784179687, "grad_norm": 1.8317444324493408, "loss": 0.5045214802026748, "time": 0.8453119039535523, "epoch": 1, "memory": 28798, "step": 400} +{"lr": 0.0001, "data_time": 0.0017046451568603516, "grad_norm": 1.8737725734710693, "loss": 0.4915499031543732, "time": 0.9132485389709473, "epoch": 1, "memory": 28798, "step": 500} +{"lr": 0.0001, "data_time": 0.00168917179107666, "grad_norm": 4.088968598842621, "loss": 0.491387265920639, "time": 0.8060833215713501, "epoch": 1, "memory": 28798, "step": 600} +{"lr": 0.0001, "data_time": 0.002523636817932129, "grad_norm": 2.1137129604816436, "loss": 0.45475015342235564, "time": 0.8018754959106446, "epoch": 1, "memory": 28798, "step": 700} +{"lr": 0.0001, "data_time": 0.0021129846572875977, "grad_norm": 3.255834722518921, "loss": 0.44188469648361206, "time": 0.8233593463897705, "epoch": 1, "memory": 28798, "step": 800} +{"lr": 0.0001, "data_time": 0.0018212318420410157, "grad_norm": 6.59802017211914, "loss": 0.4631790190935135, "time": 0.8352148056030273, "epoch": 1, "memory": 28798, "step": 900} +{"lr": 0.0001, "data_time": 0.0018018722534179688, "grad_norm": 5.292197728157044, "loss": 0.43976627588272094, "time": 0.7668165922164917, "epoch": 1, "memory": 28798, "step": 1000} +{"lr": 0.0001, "data_time": 0.0016475439071655274, "grad_norm": 9.713938117027283, "loss": 0.42211116552352906, "time": 0.8853485107421875, "epoch": 1, "memory": 28798, "step": 1100} +{"lr": 0.0001, "data_time": 0.0017646551132202148, "grad_norm": 4.946593928337097, "loss": 0.4150597840547562, "time": 0.8224147081375122, "epoch": 1, "memory": 28798, "step": 1200} +{"lr": 0.0001, "data_time": 0.0017141580581665039, "grad_norm": 4.442331027984619, "loss": 0.42618680000305176, "time": 0.8189225435256958, "epoch": 1, "memory": 28798, "step": 1300} +{"lr": 0.0001, "data_time": 0.0020910978317260744, "grad_norm": 9.779367160797118, "loss": 0.4239910125732422, "time": 0.8027831077575683, "epoch": 1, "memory": 28798, "step": 1400} +{"lr": 0.0001, "data_time": 0.0016209602355957032, "grad_norm": 9.00402204990387, "loss": 0.4142557501792908, "time": 0.8396742343902588, "epoch": 1, "memory": 28798, "step": 1500} +{"lr": 0.0001, "data_time": 0.0017014265060424805, "grad_norm": 6.99322247505188, "loss": 0.4104282259941101, "time": 0.8199834108352662, "epoch": 1, "memory": 28798, "step": 1600} +{"lr": 0.0001, "data_time": 0.002666807174682617, "grad_norm": 9.525004410743714, "loss": 0.42388567328453064, "time": 0.8770117282867431, "epoch": 1, "memory": 28798, "step": 1700} +{"lr": 0.0001, "data_time": 0.0018558025360107422, "grad_norm": 4.831781053543091, "loss": 0.39327278435230256, "time": 0.7856359720230103, "epoch": 1, "memory": 28798, "step": 1800} +{"lr": 0.0001, "data_time": 0.0017184734344482422, "grad_norm": 5.7949342966079715, "loss": 0.38883269429206846, "time": 0.8746841430664063, "epoch": 1, "memory": 28798, "step": 1900} +{"lr": 0.0001, "data_time": 0.002102231979370117, "grad_norm": 5.9448317527771, "loss": 0.3760419011116028, "time": 0.8897496223449707, "epoch": 1, "memory": 28798, "step": 2000} +{"lr": 0.0001, "data_time": 0.001977872848510742, "grad_norm": 12.630048084259034, "loss": 0.3871970385313034, "time": 0.846810245513916, "epoch": 1, "memory": 28798, "step": 2100} +{"lr": 0.0001, "data_time": 0.0016800642013549804, "grad_norm": 6.408954977989197, "loss": 0.35754445791244505, "time": 0.8513432025909424, "epoch": 1, "memory": 28798, "step": 2200} +{"lr": 0.0001, "data_time": 0.0020012617111206054, "grad_norm": 14.529697275161743, "loss": 0.3792106330394745, "time": 0.8477707147598267, "epoch": 1, "memory": 28798, "step": 2300} +{"lr": 0.0001, "data_time": 0.002001023292541504, "grad_norm": 12.994986748695373, "loss": 0.3565497905015945, "time": 0.8485530138015747, "epoch": 1, "memory": 28798, "step": 2400} +{"lr": 0.0001, "data_time": 0.0016268253326416015, "grad_norm": 12.43475399017334, "loss": 0.3879263669252396, "time": 0.7890827894210816, "epoch": 1, "memory": 28798, "step": 2500} +{"lr": 0.0001, "data_time": 0.0016362905502319337, "grad_norm": 7.629641771316528, "loss": 0.34301136136054994, "time": 0.7783349037170411, "epoch": 1, "memory": 28798, "step": 2600} +{"lr": 0.0001, "data_time": 0.0017586469650268555, "grad_norm": 16.589220929145814, "loss": 0.36621828079223634, "time": 0.7944331884384155, "epoch": 1, "memory": 28798, "step": 2700} +{"lr": 0.0001, "data_time": 0.002016139030456543, "grad_norm": 14.626896715164184, "loss": 0.36732604205608366, "time": 0.8246288299560547, "epoch": 1, "memory": 28798, "step": 2800} +{"lr": 0.0001, "data_time": 0.0016806840896606446, "grad_norm": 11.514037919044494, "loss": 0.3996082961559296, "time": 0.8673771858215332, "epoch": 1, "memory": 28798, "step": 2900} +{"lr": 0.0001, "data_time": 0.002578449249267578, "grad_norm": 6.6478078126907345, "loss": 0.3242823928594589, "time": 0.8155043125152588, "epoch": 1, "memory": 28798, "step": 3000} +{"lr": 0.0001, "data_time": 0.0018572568893432616, "grad_norm": 6.828146457672119, "loss": 0.3577097594738007, "time": 0.8268700361251831, "epoch": 1, "memory": 28798, "step": 3100} +{"lr": 0.0001, "data_time": 0.0017281055450439453, "grad_norm": 9.37593960762024, "loss": 0.32819118797779084, "time": 0.7979895830154419, "epoch": 1, "memory": 28798, "step": 3200} +{"lr": 0.0001, "data_time": 0.0021569490432739257, "grad_norm": 23.40840873718262, "loss": 0.4111290514469147, "time": 0.8013572692871094, "epoch": 1, "memory": 28798, "step": 3300} +{"lr": 0.0001, "data_time": 0.0016727924346923828, "grad_norm": 13.582075595855713, "loss": 0.3273530423641205, "time": 0.7767271041870117, "epoch": 1, "memory": 28798, "step": 3400} +{"lr": 0.0001, "data_time": 0.002120232582092285, "grad_norm": 4.67175931930542, "loss": 0.3217101663351059, "time": 0.8272306680679321, "epoch": 1, "memory": 28798, "step": 3500} +{"lr": 0.0001, "data_time": 0.0019310712814331055, "grad_norm": 4.494315493106842, "loss": 0.3059556096792221, "time": 0.801482892036438, "epoch": 1, "memory": 28798, "step": 3600} +{"lr": 0.0001, "data_time": 0.0015377521514892579, "grad_norm": 4.780852317810059, "loss": 0.3073240280151367, "time": 0.7870206832885742, "epoch": 1, "memory": 28798, "step": 3700} +{"lr": 0.0001, "data_time": 0.0018393278121948242, "grad_norm": 9.915586733818055, "loss": 0.31806694269180297, "time": 0.8184965372085571, "epoch": 1, "memory": 28798, "step": 3800} +{"lr": 0.0001, "data_time": 0.001886272430419922, "grad_norm": 7.728871440887451, "loss": 0.30202472805976865, "time": 0.8365265846252441, "epoch": 1, "memory": 28798, "step": 3900} +{"lr": 0.0001, "data_time": 0.0017246723175048829, "grad_norm": 8.228133702278138, "loss": 0.3029386967420578, "time": 0.7956446170806885, "epoch": 1, "memory": 28798, "step": 4000} diff --git a/swin_small_4xb256_fake5m_lr1e-4_10epoch/20230601_155932/vis_data/config.py b/swin_small_4xb256_fake5m_lr1e-4_10epoch/20230601_155932/vis_data/config.py new file mode 100644 index 0000000000000000000000000000000000000000..07c9f4bb104a3c4f894e7a05987a8354fd463e1f --- /dev/null +++ b/swin_small_4xb256_fake5m_lr1e-4_10epoch/20230601_155932/vis_data/config.py @@ -0,0 +1,175 @@ +optim_wrapper = dict( + optimizer=dict( + type='AdamW', + lr=0.0001, + weight_decay=0.05, + eps=1e-08, + betas=(0.9, 0.999), + _scope_='mmpretrain'), + paramwise_cfg=dict( + norm_decay_mult=0.0, + bias_decay_mult=0.0, + flat_decay_mult=0.0, + custom_keys=dict({ + '.absolute_pos_embed': dict(decay_mult=0.0), + '.relative_position_bias_table': dict(decay_mult=0.0) + })), + type='AmpOptimWrapper', + dtype='bfloat16', + clip_grad=dict(max_norm=5.0)) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=True, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=1024) +model = dict( + type='ImageClassifier', + backbone=dict( + type='SwinTransformer', arch='small', img_size=224, + drop_path_rate=0.3), + neck=dict(type='GlobalAveragePooling'), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=768, + init_cfg=None, + loss=dict( + type='LabelSmoothLoss', label_smooth_val=0.1, mode='original'), + cal_acc=False), + init_cfg=[ + dict(type='TruncNormal', layer='Linear', std=0.02, bias=0.0), + dict(type='Constant', layer='LayerNorm', val=1.0, bias=0.0) + ], + train_cfg=None) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stablediffusionV1-5R2-dpmsolver-25-5m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/swin_small_4xb256_fake5m_lr1e-4_10epoch' diff --git a/swin_small_4xb256_fake5m_lr1e-4_10epoch/20230601_155932/vis_data/events.out.tfevents.1685606380.SH-IDC1-10-140-24-80.260194.0 b/swin_small_4xb256_fake5m_lr1e-4_10epoch/20230601_155932/vis_data/events.out.tfevents.1685606380.SH-IDC1-10-140-24-80.260194.0 new file mode 100644 index 0000000000000000000000000000000000000000..9310887fb938740476f7a1276aad73ac805deab7 --- /dev/null +++ b/swin_small_4xb256_fake5m_lr1e-4_10epoch/20230601_155932/vis_data/events.out.tfevents.1685606380.SH-IDC1-10-140-24-80.260194.0 @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ef1a398f37c134a8ab76e6ffff5c4abd0fbb5eed554f51d3d3207c08f6ad3ebb +size 18277 diff --git a/swin_small_4xb256_fake5m_lr1e-4_10epoch/20230601_155932/vis_data/scalars.json b/swin_small_4xb256_fake5m_lr1e-4_10epoch/20230601_155932/vis_data/scalars.json new file mode 100644 index 0000000000000000000000000000000000000000..7647cd746a11a2b7893d6f153da6770a8ebf15dd --- /dev/null +++ b/swin_small_4xb256_fake5m_lr1e-4_10epoch/20230601_155932/vis_data/scalars.json @@ -0,0 +1,40 @@ +{"lr": 0.0001, "data_time": 0.0016358852386474609, "grad_norm": 0.6826717019081116, "loss": 0.5935846924781799, "time": 0.8011452198028565, "epoch": 1, "memory": 28798, "step": 100} +{"lr": 0.0001, "data_time": 0.0017016410827636718, "grad_norm": 1.2591084003448487, "loss": 0.5710584998130799, "time": 0.8666714429855347, "epoch": 1, "memory": 28798, "step": 200} +{"lr": 0.0001, "data_time": 0.001643061637878418, "grad_norm": 1.3794927954673768, "loss": 0.5229363799095154, "time": 0.8052175283432007, "epoch": 1, "memory": 28798, "step": 300} +{"lr": 0.0001, "data_time": 0.0016553878784179687, "grad_norm": 1.8317444324493408, "loss": 0.5045214802026748, "time": 0.8453119039535523, "epoch": 1, "memory": 28798, "step": 400} +{"lr": 0.0001, "data_time": 0.0017046451568603516, "grad_norm": 1.8737725734710693, "loss": 0.4915499031543732, "time": 0.9132485389709473, "epoch": 1, "memory": 28798, "step": 500} +{"lr": 0.0001, "data_time": 0.00168917179107666, "grad_norm": 4.088968598842621, "loss": 0.491387265920639, "time": 0.8060833215713501, "epoch": 1, "memory": 28798, "step": 600} +{"lr": 0.0001, "data_time": 0.002523636817932129, "grad_norm": 2.1137129604816436, "loss": 0.45475015342235564, "time": 0.8018754959106446, "epoch": 1, "memory": 28798, "step": 700} +{"lr": 0.0001, "data_time": 0.0021129846572875977, "grad_norm": 3.255834722518921, "loss": 0.44188469648361206, "time": 0.8233593463897705, "epoch": 1, "memory": 28798, "step": 800} +{"lr": 0.0001, "data_time": 0.0018212318420410157, "grad_norm": 6.59802017211914, "loss": 0.4631790190935135, "time": 0.8352148056030273, "epoch": 1, "memory": 28798, "step": 900} +{"lr": 0.0001, "data_time": 0.0018018722534179688, "grad_norm": 5.292197728157044, "loss": 0.43976627588272094, "time": 0.7668165922164917, "epoch": 1, "memory": 28798, "step": 1000} +{"lr": 0.0001, "data_time": 0.0016475439071655274, "grad_norm": 9.713938117027283, "loss": 0.42211116552352906, "time": 0.8853485107421875, "epoch": 1, "memory": 28798, "step": 1100} +{"lr": 0.0001, "data_time": 0.0017646551132202148, "grad_norm": 4.946593928337097, "loss": 0.4150597840547562, "time": 0.8224147081375122, "epoch": 1, "memory": 28798, "step": 1200} +{"lr": 0.0001, "data_time": 0.0017141580581665039, "grad_norm": 4.442331027984619, "loss": 0.42618680000305176, "time": 0.8189225435256958, "epoch": 1, "memory": 28798, "step": 1300} +{"lr": 0.0001, "data_time": 0.0020910978317260744, "grad_norm": 9.779367160797118, "loss": 0.4239910125732422, "time": 0.8027831077575683, "epoch": 1, "memory": 28798, "step": 1400} +{"lr": 0.0001, "data_time": 0.0016209602355957032, "grad_norm": 9.00402204990387, "loss": 0.4142557501792908, "time": 0.8396742343902588, "epoch": 1, "memory": 28798, "step": 1500} +{"lr": 0.0001, "data_time": 0.0017014265060424805, "grad_norm": 6.99322247505188, "loss": 0.4104282259941101, "time": 0.8199834108352662, "epoch": 1, "memory": 28798, "step": 1600} +{"lr": 0.0001, "data_time": 0.002666807174682617, "grad_norm": 9.525004410743714, "loss": 0.42388567328453064, "time": 0.8770117282867431, "epoch": 1, "memory": 28798, "step": 1700} +{"lr": 0.0001, "data_time": 0.0018558025360107422, "grad_norm": 4.831781053543091, "loss": 0.39327278435230256, "time": 0.7856359720230103, "epoch": 1, "memory": 28798, "step": 1800} +{"lr": 0.0001, "data_time": 0.0017184734344482422, "grad_norm": 5.7949342966079715, "loss": 0.38883269429206846, "time": 0.8746841430664063, "epoch": 1, "memory": 28798, "step": 1900} +{"lr": 0.0001, "data_time": 0.002102231979370117, "grad_norm": 5.9448317527771, "loss": 0.3760419011116028, "time": 0.8897496223449707, "epoch": 1, "memory": 28798, "step": 2000} +{"lr": 0.0001, "data_time": 0.001977872848510742, "grad_norm": 12.630048084259034, "loss": 0.3871970385313034, "time": 0.846810245513916, "epoch": 1, "memory": 28798, "step": 2100} +{"lr": 0.0001, "data_time": 0.0016800642013549804, "grad_norm": 6.408954977989197, "loss": 0.35754445791244505, "time": 0.8513432025909424, "epoch": 1, "memory": 28798, "step": 2200} +{"lr": 0.0001, "data_time": 0.0020012617111206054, "grad_norm": 14.529697275161743, "loss": 0.3792106330394745, "time": 0.8477707147598267, "epoch": 1, "memory": 28798, "step": 2300} +{"lr": 0.0001, "data_time": 0.002001023292541504, "grad_norm": 12.994986748695373, "loss": 0.3565497905015945, "time": 0.8485530138015747, "epoch": 1, "memory": 28798, "step": 2400} +{"lr": 0.0001, "data_time": 0.0016268253326416015, "grad_norm": 12.43475399017334, "loss": 0.3879263669252396, "time": 0.7890827894210816, "epoch": 1, "memory": 28798, "step": 2500} +{"lr": 0.0001, "data_time": 0.0016362905502319337, "grad_norm": 7.629641771316528, "loss": 0.34301136136054994, "time": 0.7783349037170411, "epoch": 1, "memory": 28798, "step": 2600} +{"lr": 0.0001, "data_time": 0.0017586469650268555, "grad_norm": 16.589220929145814, "loss": 0.36621828079223634, "time": 0.7944331884384155, "epoch": 1, "memory": 28798, "step": 2700} +{"lr": 0.0001, "data_time": 0.002016139030456543, "grad_norm": 14.626896715164184, "loss": 0.36732604205608366, "time": 0.8246288299560547, "epoch": 1, "memory": 28798, "step": 2800} +{"lr": 0.0001, "data_time": 0.0016806840896606446, "grad_norm": 11.514037919044494, "loss": 0.3996082961559296, "time": 0.8673771858215332, "epoch": 1, "memory": 28798, "step": 2900} +{"lr": 0.0001, "data_time": 0.002578449249267578, "grad_norm": 6.6478078126907345, "loss": 0.3242823928594589, "time": 0.8155043125152588, "epoch": 1, "memory": 28798, "step": 3000} +{"lr": 0.0001, "data_time": 0.0018572568893432616, "grad_norm": 6.828146457672119, "loss": 0.3577097594738007, "time": 0.8268700361251831, "epoch": 1, "memory": 28798, "step": 3100} +{"lr": 0.0001, "data_time": 0.0017281055450439453, "grad_norm": 9.37593960762024, "loss": 0.32819118797779084, "time": 0.7979895830154419, "epoch": 1, "memory": 28798, "step": 3200} +{"lr": 0.0001, "data_time": 0.0021569490432739257, "grad_norm": 23.40840873718262, "loss": 0.4111290514469147, "time": 0.8013572692871094, "epoch": 1, "memory": 28798, "step": 3300} +{"lr": 0.0001, "data_time": 0.0016727924346923828, "grad_norm": 13.582075595855713, "loss": 0.3273530423641205, "time": 0.7767271041870117, "epoch": 1, "memory": 28798, "step": 3400} +{"lr": 0.0001, "data_time": 0.002120232582092285, "grad_norm": 4.67175931930542, "loss": 0.3217101663351059, "time": 0.8272306680679321, "epoch": 1, "memory": 28798, "step": 3500} +{"lr": 0.0001, "data_time": 0.0019310712814331055, "grad_norm": 4.494315493106842, "loss": 0.3059556096792221, "time": 0.801482892036438, "epoch": 1, "memory": 28798, "step": 3600} +{"lr": 0.0001, "data_time": 0.0015377521514892579, "grad_norm": 4.780852317810059, "loss": 0.3073240280151367, "time": 0.7870206832885742, "epoch": 1, "memory": 28798, "step": 3700} +{"lr": 0.0001, "data_time": 0.0018393278121948242, "grad_norm": 9.915586733818055, "loss": 0.31806694269180297, "time": 0.8184965372085571, "epoch": 1, "memory": 28798, "step": 3800} +{"lr": 0.0001, "data_time": 0.001886272430419922, "grad_norm": 7.728871440887451, "loss": 0.30202472805976865, "time": 0.8365265846252441, "epoch": 1, "memory": 28798, "step": 3900} +{"lr": 0.0001, "data_time": 0.0017246723175048829, "grad_norm": 8.228133702278138, "loss": 0.3029386967420578, "time": 0.7956446170806885, "epoch": 1, "memory": 28798, "step": 4000} diff --git a/swin_small_4xb256_fake5m_lr1e-4_10epoch/20230601_165803/20230601_165803.log b/swin_small_4xb256_fake5m_lr1e-4_10epoch/20230601_165803/20230601_165803.log new file mode 100644 index 0000000000000000000000000000000000000000..9a9befa677aeeee407a2f52471f7834f0319fd98 --- /dev/null +++ b/swin_small_4xb256_fake5m_lr1e-4_10epoch/20230601_165803/20230601_165803.log @@ -0,0 +1,1779 @@ +2023/06/01 16:58:07 - mmengine - INFO - +------------------------------------------------------------ +System environment: + sys.platform: linux + Python: 3.10.9 (main, Mar 8 2023, 10:47:38) [GCC 11.2.0] + CUDA available: True + numpy_random_seed: 394707077 + GPU 0,1,2,3: NVIDIA A100-SXM4-80GB + CUDA_HOME: /mnt/petrelfs/share/cuda-11.6 + NVCC: Cuda compilation tools, release 11.6, V11.6.124 + GCC: gcc (GCC) 4.8.5 20150623 (Red Hat 4.8.5-44) + PyTorch: 1.13.1 + PyTorch compiling details: PyTorch built with: + - GCC 9.3 + - C++ Version: 201402 + - Intel(R) oneAPI Math Kernel Library Version 2021.4-Product Build 20210904 for Intel(R) 64 architecture applications + - Intel(R) MKL-DNN v2.6.0 (Git Hash 52b5f107dd9cf10910aaa19cb47f3abf9b349815) + - OpenMP 201511 (a.k.a. OpenMP 4.5) + - LAPACK is enabled (usually provided by MKL) + - NNPACK is enabled + - CPU capability usage: AVX2 + - CUDA Runtime 11.6 + - NVCC architecture flags: -gencode;arch=compute_37,code=sm_37;-gencode;arch=compute_50,code=sm_50;-gencode;arch=compute_60,code=sm_60;-gencode;arch=compute_61,code=sm_61;-gencode;arch=compute_70,code=sm_70;-gencode;arch=compute_75,code=sm_75;-gencode;arch=compute_80,code=sm_80;-gencode;arch=compute_86,code=sm_86;-gencode;arch=compute_37,code=compute_37 + - CuDNN 8.3.2 (built against CUDA 11.5) + - Magma 2.6.1 + - Build settings: BLAS_INFO=mkl, BUILD_TYPE=Release, CUDA_VERSION=11.6, CUDNN_VERSION=8.3.2, CXX_COMPILER=/opt/rh/devtoolset-9/root/usr/bin/c++, CXX_FLAGS= -fabi-version=11 -Wno-deprecated -fvisibility-inlines-hidden -DUSE_PTHREADPOOL -fopenmp -DNDEBUG -DUSE_KINETO -DUSE_FBGEMM -DUSE_QNNPACK -DUSE_PYTORCH_QNNPACK -DUSE_XNNPACK -DSYMBOLICATE_MOBILE_DEBUG_HANDLE -DEDGE_PROFILER_USE_KINETO -O2 -fPIC -Wno-narrowing -Wall -Wextra -Werror=return-type -Werror=non-virtual-dtor -Wno-missing-field-initializers -Wno-type-limits -Wno-array-bounds -Wno-unknown-pragmas -Wunused-local-typedefs -Wno-unused-parameter -Wno-unused-function -Wno-unused-result -Wno-strict-overflow -Wno-strict-aliasing -Wno-error=deprecated-declarations -Wno-stringop-overflow -Wno-psabi -Wno-error=pedantic -Wno-error=redundant-decls -Wno-error=old-style-cast -fdiagnostics-color=always -faligned-new -Wno-unused-but-set-variable -Wno-maybe-uninitialized -fno-math-errno -fno-trapping-math -Werror=format -Werror=cast-function-type -Wno-stringop-overflow, LAPACK_INFO=mkl, PERF_WITH_AVX=1, PERF_WITH_AVX2=1, PERF_WITH_AVX512=1, TORCH_VERSION=1.13.1, USE_CUDA=ON, USE_CUDNN=ON, USE_EXCEPTION_PTR=1, USE_GFLAGS=OFF, USE_GLOG=OFF, USE_MKL=ON, USE_MKLDNN=ON, USE_MPI=OFF, USE_NCCL=ON, USE_NNPACK=ON, USE_OPENMP=ON, USE_ROCM=OFF, + + TorchVision: 0.14.1 + OpenCV: 4.7.0 + MMEngine: 0.7.3 + +Runtime environment: + cudnn_benchmark: True + mp_cfg: {'mp_start_method': 'fork', 'opencv_num_threads': 0} + dist_cfg: {'backend': 'nccl'} + seed: None + deterministic: False + Distributed launcher: slurm + Distributed training: True + GPU number: 4 +------------------------------------------------------------ + +2023/06/01 16:58:08 - mmengine - INFO - Config: +optim_wrapper = dict( + optimizer=dict( + type='AdamW', + lr=0.0001, + weight_decay=0.05, + eps=1e-08, + betas=(0.9, 0.999), + _scope_='mmpretrain'), + paramwise_cfg=dict( + norm_decay_mult=0.0, + bias_decay_mult=0.0, + flat_decay_mult=0.0, + custom_keys=dict({ + '.absolute_pos_embed': dict(decay_mult=0.0), + '.relative_position_bias_table': dict(decay_mult=0.0) + })), + type='AmpOptimWrapper', + dtype='bfloat16', + clip_grad=dict(max_norm=5.0)) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=True, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=1024) +model = dict( + type='ImageClassifier', + backbone=dict( + type='SwinTransformer', arch='small', img_size=224, + drop_path_rate=0.3), + neck=dict(type='GlobalAveragePooling'), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=768, + init_cfg=None, + loss=dict( + type='LabelSmoothLoss', label_smooth_val=0.1, mode='original'), + cal_acc=False), + init_cfg=[ + dict(type='TruncNormal', layer='Linear', std=0.02, bias=0.0), + dict(type='Constant', layer='LayerNorm', val=1.0, bias=0.0) + ], + train_cfg=None) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stablediffusionV1-5R2-dpmsolver-25-5m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/swin_small_4xb256_fake5m_lr1e-4_10epoch' + +2023/06/01 16:58:12 - mmengine - INFO - Hooks will be executed in the following order: +before_run: +(VERY_HIGH ) RuntimeInfoHook +(BELOW_NORMAL) LoggerHook + -------------------- +before_train: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_train_epoch: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(NORMAL ) DistSamplerSeedHook + -------------------- +before_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook + -------------------- +after_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +after_train_epoch: +(NORMAL ) IterTimerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_val_epoch: +(NORMAL ) IterTimerHook + -------------------- +before_val_iter: +(NORMAL ) IterTimerHook + -------------------- +after_val_iter: +(NORMAL ) IterTimerHook +(NORMAL ) VisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_val_epoch: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +after_train: +(VERY_LOW ) CheckpointHook + -------------------- +before_test_epoch: +(NORMAL ) IterTimerHook + -------------------- +before_test_iter: +(NORMAL ) IterTimerHook + -------------------- +after_test_iter: +(NORMAL ) IterTimerHook +(NORMAL ) VisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_test_epoch: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_run: +(BELOW_NORMAL) LoggerHook + -------------------- +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.patch_embed.projection.bias:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.patch_embed.norm.weight:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.patch_embed.norm.bias:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.0.blocks.0.norm1.weight:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.0.blocks.0.norm1.bias:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.0.blocks.0.attn.w_msa.relative_position_bias_table:lr=0.0001 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.0.blocks.0.attn.w_msa.relative_position_bias_table:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.0.blocks.0.attn.w_msa.relative_position_bias_table:decay_mult=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.0.blocks.0.attn.w_msa.qkv.bias:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.0.blocks.0.attn.w_msa.proj.bias:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.0.blocks.0.norm2.weight:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.0.blocks.0.norm2.bias:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.0.blocks.0.ffn.layers.0.0.bias:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.0.blocks.0.ffn.layers.1.bias:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.0.blocks.1.norm1.weight:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.0.blocks.1.norm1.bias:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.0.blocks.1.attn.w_msa.relative_position_bias_table:lr=0.0001 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.0.blocks.1.attn.w_msa.relative_position_bias_table:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.0.blocks.1.attn.w_msa.relative_position_bias_table:decay_mult=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.0.blocks.1.attn.w_msa.qkv.bias:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.0.blocks.1.attn.w_msa.proj.bias:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.0.blocks.1.norm2.weight:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.0.blocks.1.norm2.bias:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.0.blocks.1.ffn.layers.0.0.bias:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.0.blocks.1.ffn.layers.1.bias:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.0.downsample.norm.weight:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.0.downsample.norm.bias:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.1.blocks.0.norm1.weight:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.1.blocks.0.norm1.bias:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.1.blocks.0.attn.w_msa.relative_position_bias_table:lr=0.0001 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.1.blocks.0.attn.w_msa.relative_position_bias_table:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.1.blocks.0.attn.w_msa.relative_position_bias_table:decay_mult=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.1.blocks.0.attn.w_msa.qkv.bias:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.1.blocks.0.attn.w_msa.proj.bias:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.1.blocks.0.norm2.weight:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.1.blocks.0.norm2.bias:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.1.blocks.0.ffn.layers.0.0.bias:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.1.blocks.0.ffn.layers.1.bias:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.1.blocks.1.norm1.weight:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.1.blocks.1.norm1.bias:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.1.blocks.1.attn.w_msa.relative_position_bias_table:lr=0.0001 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.1.blocks.1.attn.w_msa.relative_position_bias_table:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.1.blocks.1.attn.w_msa.relative_position_bias_table:decay_mult=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.1.blocks.1.attn.w_msa.qkv.bias:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.1.blocks.1.attn.w_msa.proj.bias:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.1.blocks.1.norm2.weight:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.1.blocks.1.norm2.bias:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.1.blocks.1.ffn.layers.0.0.bias:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.1.blocks.1.ffn.layers.1.bias:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.1.downsample.norm.weight:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.1.downsample.norm.bias:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.0.norm1.weight:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.0.norm1.bias:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.0.attn.w_msa.relative_position_bias_table:lr=0.0001 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.0.attn.w_msa.relative_position_bias_table:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.0.attn.w_msa.relative_position_bias_table:decay_mult=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.0.attn.w_msa.qkv.bias:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.0.attn.w_msa.proj.bias:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.0.norm2.weight:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.0.norm2.bias:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.0.ffn.layers.0.0.bias:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.0.ffn.layers.1.bias:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.1.norm1.weight:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.1.norm1.bias:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.1.attn.w_msa.relative_position_bias_table:lr=0.0001 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.1.attn.w_msa.relative_position_bias_table:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.1.attn.w_msa.relative_position_bias_table:decay_mult=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.1.attn.w_msa.qkv.bias:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.1.attn.w_msa.proj.bias:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.1.norm2.weight:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.1.norm2.bias:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.1.ffn.layers.0.0.bias:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.1.ffn.layers.1.bias:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.2.norm1.weight:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.2.norm1.bias:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.2.attn.w_msa.relative_position_bias_table:lr=0.0001 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.2.attn.w_msa.relative_position_bias_table:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.2.attn.w_msa.relative_position_bias_table:decay_mult=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.2.attn.w_msa.qkv.bias:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.2.attn.w_msa.proj.bias:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.2.norm2.weight:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.2.norm2.bias:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.2.ffn.layers.0.0.bias:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.2.ffn.layers.1.bias:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.3.norm1.weight:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.3.norm1.bias:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.3.attn.w_msa.relative_position_bias_table:lr=0.0001 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.3.attn.w_msa.relative_position_bias_table:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.3.attn.w_msa.relative_position_bias_table:decay_mult=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.3.attn.w_msa.qkv.bias:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.3.attn.w_msa.proj.bias:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.3.norm2.weight:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.3.norm2.bias:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.3.ffn.layers.0.0.bias:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.3.ffn.layers.1.bias:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.4.norm1.weight:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.4.norm1.bias:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.4.attn.w_msa.relative_position_bias_table:lr=0.0001 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.4.attn.w_msa.relative_position_bias_table:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.4.attn.w_msa.relative_position_bias_table:decay_mult=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.4.attn.w_msa.qkv.bias:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.4.attn.w_msa.proj.bias:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.4.norm2.weight:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.4.norm2.bias:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.4.ffn.layers.0.0.bias:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.4.ffn.layers.1.bias:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.5.norm1.weight:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.5.norm1.bias:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.5.attn.w_msa.relative_position_bias_table:lr=0.0001 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.5.attn.w_msa.relative_position_bias_table:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.5.attn.w_msa.relative_position_bias_table:decay_mult=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.5.attn.w_msa.qkv.bias:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.5.attn.w_msa.proj.bias:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.5.norm2.weight:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.5.norm2.bias:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.5.ffn.layers.0.0.bias:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.5.ffn.layers.1.bias:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.6.norm1.weight:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.6.norm1.bias:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.6.attn.w_msa.relative_position_bias_table:lr=0.0001 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.6.attn.w_msa.relative_position_bias_table:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.6.attn.w_msa.relative_position_bias_table:decay_mult=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.6.attn.w_msa.qkv.bias:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.6.attn.w_msa.proj.bias:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.6.norm2.weight:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.6.norm2.bias:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.6.ffn.layers.0.0.bias:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.6.ffn.layers.1.bias:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.7.norm1.weight:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.7.norm1.bias:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.7.attn.w_msa.relative_position_bias_table:lr=0.0001 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.7.attn.w_msa.relative_position_bias_table:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.7.attn.w_msa.relative_position_bias_table:decay_mult=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.7.attn.w_msa.qkv.bias:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.7.attn.w_msa.proj.bias:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.7.norm2.weight:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.7.norm2.bias:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.7.ffn.layers.0.0.bias:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.7.ffn.layers.1.bias:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.8.norm1.weight:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.8.norm1.bias:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.8.attn.w_msa.relative_position_bias_table:lr=0.0001 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.8.attn.w_msa.relative_position_bias_table:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.8.attn.w_msa.relative_position_bias_table:decay_mult=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.8.attn.w_msa.qkv.bias:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.8.attn.w_msa.proj.bias:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.8.norm2.weight:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.8.norm2.bias:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.8.ffn.layers.0.0.bias:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.8.ffn.layers.1.bias:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.9.norm1.weight:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.9.norm1.bias:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.9.attn.w_msa.relative_position_bias_table:lr=0.0001 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.9.attn.w_msa.relative_position_bias_table:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.9.attn.w_msa.relative_position_bias_table:decay_mult=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.9.attn.w_msa.qkv.bias:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.9.attn.w_msa.proj.bias:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.9.norm2.weight:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.9.norm2.bias:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.9.ffn.layers.0.0.bias:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.9.ffn.layers.1.bias:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.10.norm1.weight:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.10.norm1.bias:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.10.attn.w_msa.relative_position_bias_table:lr=0.0001 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.10.attn.w_msa.relative_position_bias_table:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.10.attn.w_msa.relative_position_bias_table:decay_mult=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.10.attn.w_msa.qkv.bias:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.10.attn.w_msa.proj.bias:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.10.norm2.weight:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.10.norm2.bias:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.10.ffn.layers.0.0.bias:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.10.ffn.layers.1.bias:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.11.norm1.weight:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.11.norm1.bias:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.11.attn.w_msa.relative_position_bias_table:lr=0.0001 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.11.attn.w_msa.relative_position_bias_table:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.11.attn.w_msa.relative_position_bias_table:decay_mult=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.11.attn.w_msa.qkv.bias:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.11.attn.w_msa.proj.bias:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.11.norm2.weight:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.11.norm2.bias:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.11.ffn.layers.0.0.bias:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.11.ffn.layers.1.bias:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.12.norm1.weight:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.12.norm1.bias:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.12.attn.w_msa.relative_position_bias_table:lr=0.0001 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.12.attn.w_msa.relative_position_bias_table:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.12.attn.w_msa.relative_position_bias_table:decay_mult=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.12.attn.w_msa.qkv.bias:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.12.attn.w_msa.proj.bias:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.12.norm2.weight:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.12.norm2.bias:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.12.ffn.layers.0.0.bias:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.12.ffn.layers.1.bias:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.13.norm1.weight:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.13.norm1.bias:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.13.attn.w_msa.relative_position_bias_table:lr=0.0001 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.13.attn.w_msa.relative_position_bias_table:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.13.attn.w_msa.relative_position_bias_table:decay_mult=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.13.attn.w_msa.qkv.bias:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.13.attn.w_msa.proj.bias:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.13.norm2.weight:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.13.norm2.bias:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.13.ffn.layers.0.0.bias:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.13.ffn.layers.1.bias:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.14.norm1.weight:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.14.norm1.bias:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.14.attn.w_msa.relative_position_bias_table:lr=0.0001 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.14.attn.w_msa.relative_position_bias_table:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.14.attn.w_msa.relative_position_bias_table:decay_mult=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.14.attn.w_msa.qkv.bias:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.14.attn.w_msa.proj.bias:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.14.norm2.weight:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.14.norm2.bias:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.14.ffn.layers.0.0.bias:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.14.ffn.layers.1.bias:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.15.norm1.weight:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.15.norm1.bias:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.15.attn.w_msa.relative_position_bias_table:lr=0.0001 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.15.attn.w_msa.relative_position_bias_table:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.15.attn.w_msa.relative_position_bias_table:decay_mult=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.15.attn.w_msa.qkv.bias:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.15.attn.w_msa.proj.bias:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.15.norm2.weight:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.15.norm2.bias:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.15.ffn.layers.0.0.bias:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.15.ffn.layers.1.bias:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.16.norm1.weight:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.16.norm1.bias:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.16.attn.w_msa.relative_position_bias_table:lr=0.0001 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.16.attn.w_msa.relative_position_bias_table:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.16.attn.w_msa.relative_position_bias_table:decay_mult=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.16.attn.w_msa.qkv.bias:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.16.attn.w_msa.proj.bias:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.16.norm2.weight:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.16.norm2.bias:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.16.ffn.layers.0.0.bias:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.16.ffn.layers.1.bias:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.17.norm1.weight:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.17.norm1.bias:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.17.attn.w_msa.relative_position_bias_table:lr=0.0001 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.17.attn.w_msa.relative_position_bias_table:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.17.attn.w_msa.relative_position_bias_table:decay_mult=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.17.attn.w_msa.qkv.bias:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.17.attn.w_msa.proj.bias:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.17.norm2.weight:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.17.norm2.bias:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.17.ffn.layers.0.0.bias:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.17.ffn.layers.1.bias:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.downsample.norm.weight:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.2.downsample.norm.bias:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.3.blocks.0.norm1.weight:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.3.blocks.0.norm1.bias:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.3.blocks.0.attn.w_msa.relative_position_bias_table:lr=0.0001 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.3.blocks.0.attn.w_msa.relative_position_bias_table:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.3.blocks.0.attn.w_msa.relative_position_bias_table:decay_mult=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.3.blocks.0.attn.w_msa.qkv.bias:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.3.blocks.0.attn.w_msa.proj.bias:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.3.blocks.0.norm2.weight:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.3.blocks.0.norm2.bias:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.3.blocks.0.ffn.layers.0.0.bias:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.3.blocks.0.ffn.layers.1.bias:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.3.blocks.1.norm1.weight:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.3.blocks.1.norm1.bias:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.3.blocks.1.attn.w_msa.relative_position_bias_table:lr=0.0001 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.3.blocks.1.attn.w_msa.relative_position_bias_table:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.3.blocks.1.attn.w_msa.relative_position_bias_table:decay_mult=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.3.blocks.1.attn.w_msa.qkv.bias:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.3.blocks.1.attn.w_msa.proj.bias:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.3.blocks.1.norm2.weight:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.3.blocks.1.norm2.bias:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.3.blocks.1.ffn.layers.0.0.bias:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.stages.3.blocks.1.ffn.layers.1.bias:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.norm3.weight:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- backbone.norm3.bias:weight_decay=0.0 +2023/06/01 16:58:38 - mmengine - INFO - paramwise_options -- head.fc.bias:weight_decay=0.0 +Name of parameter - Initialization information + +backbone.patch_embed.projection.weight - torch.Size([96, 3, 4, 4]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.patch_embed.projection.bias - torch.Size([96]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.patch_embed.norm.weight - torch.Size([96]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.patch_embed.norm.bias - torch.Size([96]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.0.blocks.0.norm1.weight - torch.Size([96]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.0.blocks.0.norm1.bias - torch.Size([96]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.0.blocks.0.attn.w_msa.relative_position_bias_table - torch.Size([169, 3]): +Initialized by user-defined `init_weights` in WindowMSA + +backbone.stages.0.blocks.0.attn.w_msa.qkv.weight - torch.Size([288, 96]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.blocks.0.attn.w_msa.qkv.bias - torch.Size([288]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.blocks.0.attn.w_msa.proj.weight - torch.Size([96, 96]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.blocks.0.attn.w_msa.proj.bias - torch.Size([96]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.blocks.0.norm2.weight - torch.Size([96]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.0.blocks.0.norm2.bias - torch.Size([96]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.0.blocks.0.ffn.layers.0.0.weight - torch.Size([384, 96]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.blocks.0.ffn.layers.0.0.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.blocks.0.ffn.layers.1.weight - torch.Size([96, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.blocks.0.ffn.layers.1.bias - torch.Size([96]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.blocks.1.norm1.weight - torch.Size([96]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.0.blocks.1.norm1.bias - torch.Size([96]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.0.blocks.1.attn.w_msa.relative_position_bias_table - torch.Size([169, 3]): +Initialized by user-defined `init_weights` in WindowMSA + +backbone.stages.0.blocks.1.attn.w_msa.qkv.weight - torch.Size([288, 96]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.blocks.1.attn.w_msa.qkv.bias - torch.Size([288]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.blocks.1.attn.w_msa.proj.weight - torch.Size([96, 96]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.blocks.1.attn.w_msa.proj.bias - torch.Size([96]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.blocks.1.norm2.weight - torch.Size([96]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.0.blocks.1.norm2.bias - torch.Size([96]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.0.blocks.1.ffn.layers.0.0.weight - torch.Size([384, 96]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.blocks.1.ffn.layers.0.0.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.blocks.1.ffn.layers.1.weight - torch.Size([96, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.blocks.1.ffn.layers.1.bias - torch.Size([96]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.downsample.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.0.downsample.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.0.downsample.reduction.weight - torch.Size([192, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.blocks.0.norm1.weight - torch.Size([192]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.1.blocks.0.norm1.bias - torch.Size([192]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.1.blocks.0.attn.w_msa.relative_position_bias_table - torch.Size([169, 6]): +Initialized by user-defined `init_weights` in WindowMSA + +backbone.stages.1.blocks.0.attn.w_msa.qkv.weight - torch.Size([576, 192]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.blocks.0.attn.w_msa.qkv.bias - torch.Size([576]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.blocks.0.attn.w_msa.proj.weight - torch.Size([192, 192]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.blocks.0.attn.w_msa.proj.bias - torch.Size([192]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.blocks.0.norm2.weight - torch.Size([192]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.1.blocks.0.norm2.bias - torch.Size([192]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.1.blocks.0.ffn.layers.0.0.weight - torch.Size([768, 192]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.blocks.0.ffn.layers.0.0.bias - torch.Size([768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.blocks.0.ffn.layers.1.weight - torch.Size([192, 768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.blocks.0.ffn.layers.1.bias - torch.Size([192]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.blocks.1.norm1.weight - torch.Size([192]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.1.blocks.1.norm1.bias - torch.Size([192]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.1.blocks.1.attn.w_msa.relative_position_bias_table - torch.Size([169, 6]): +Initialized by user-defined `init_weights` in WindowMSA + +backbone.stages.1.blocks.1.attn.w_msa.qkv.weight - torch.Size([576, 192]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.blocks.1.attn.w_msa.qkv.bias - torch.Size([576]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.blocks.1.attn.w_msa.proj.weight - torch.Size([192, 192]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.blocks.1.attn.w_msa.proj.bias - torch.Size([192]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.blocks.1.norm2.weight - torch.Size([192]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.1.blocks.1.norm2.bias - torch.Size([192]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.1.blocks.1.ffn.layers.0.0.weight - torch.Size([768, 192]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.blocks.1.ffn.layers.0.0.bias - torch.Size([768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.blocks.1.ffn.layers.1.weight - torch.Size([192, 768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.blocks.1.ffn.layers.1.bias - torch.Size([192]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.downsample.norm.weight - torch.Size([768]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.1.downsample.norm.bias - torch.Size([768]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.1.downsample.reduction.weight - torch.Size([384, 768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.0.norm1.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.0.norm1.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.0.attn.w_msa.relative_position_bias_table - torch.Size([169, 12]): +Initialized by user-defined `init_weights` in WindowMSA + +backbone.stages.2.blocks.0.attn.w_msa.qkv.weight - torch.Size([1152, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.0.attn.w_msa.qkv.bias - torch.Size([1152]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.0.attn.w_msa.proj.weight - torch.Size([384, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.0.attn.w_msa.proj.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.0.norm2.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.0.norm2.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.0.ffn.layers.0.0.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.0.ffn.layers.0.0.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.0.ffn.layers.1.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.0.ffn.layers.1.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.1.norm1.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.1.norm1.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.1.attn.w_msa.relative_position_bias_table - torch.Size([169, 12]): +Initialized by user-defined `init_weights` in WindowMSA + +backbone.stages.2.blocks.1.attn.w_msa.qkv.weight - torch.Size([1152, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.1.attn.w_msa.qkv.bias - torch.Size([1152]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.1.attn.w_msa.proj.weight - torch.Size([384, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.1.attn.w_msa.proj.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.1.norm2.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.1.norm2.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.1.ffn.layers.0.0.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.1.ffn.layers.0.0.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.1.ffn.layers.1.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.1.ffn.layers.1.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.2.norm1.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.2.norm1.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.2.attn.w_msa.relative_position_bias_table - torch.Size([169, 12]): +Initialized by user-defined `init_weights` in WindowMSA + +backbone.stages.2.blocks.2.attn.w_msa.qkv.weight - torch.Size([1152, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.2.attn.w_msa.qkv.bias - torch.Size([1152]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.2.attn.w_msa.proj.weight - torch.Size([384, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.2.attn.w_msa.proj.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.2.norm2.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.2.norm2.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.2.ffn.layers.0.0.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.2.ffn.layers.0.0.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.2.ffn.layers.1.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.2.ffn.layers.1.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.3.norm1.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.3.norm1.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.3.attn.w_msa.relative_position_bias_table - torch.Size([169, 12]): +Initialized by user-defined `init_weights` in WindowMSA + +backbone.stages.2.blocks.3.attn.w_msa.qkv.weight - torch.Size([1152, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.3.attn.w_msa.qkv.bias - torch.Size([1152]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.3.attn.w_msa.proj.weight - torch.Size([384, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.3.attn.w_msa.proj.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.3.norm2.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.3.norm2.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.3.ffn.layers.0.0.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.3.ffn.layers.0.0.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.3.ffn.layers.1.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.3.ffn.layers.1.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.4.norm1.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.4.norm1.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.4.attn.w_msa.relative_position_bias_table - torch.Size([169, 12]): +Initialized by user-defined `init_weights` in WindowMSA + +backbone.stages.2.blocks.4.attn.w_msa.qkv.weight - torch.Size([1152, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.4.attn.w_msa.qkv.bias - torch.Size([1152]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.4.attn.w_msa.proj.weight - torch.Size([384, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.4.attn.w_msa.proj.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.4.norm2.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.4.norm2.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.4.ffn.layers.0.0.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.4.ffn.layers.0.0.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.4.ffn.layers.1.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.4.ffn.layers.1.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.5.norm1.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.5.norm1.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.5.attn.w_msa.relative_position_bias_table - torch.Size([169, 12]): +Initialized by user-defined `init_weights` in WindowMSA + +backbone.stages.2.blocks.5.attn.w_msa.qkv.weight - torch.Size([1152, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.5.attn.w_msa.qkv.bias - torch.Size([1152]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.5.attn.w_msa.proj.weight - torch.Size([384, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.5.attn.w_msa.proj.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.5.norm2.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.5.norm2.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.5.ffn.layers.0.0.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.5.ffn.layers.0.0.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.5.ffn.layers.1.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.5.ffn.layers.1.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.6.norm1.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.6.norm1.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.6.attn.w_msa.relative_position_bias_table - torch.Size([169, 12]): +Initialized by user-defined `init_weights` in WindowMSA + +backbone.stages.2.blocks.6.attn.w_msa.qkv.weight - torch.Size([1152, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.6.attn.w_msa.qkv.bias - torch.Size([1152]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.6.attn.w_msa.proj.weight - torch.Size([384, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.6.attn.w_msa.proj.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.6.norm2.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.6.norm2.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.6.ffn.layers.0.0.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.6.ffn.layers.0.0.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.6.ffn.layers.1.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.6.ffn.layers.1.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.7.norm1.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.7.norm1.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.7.attn.w_msa.relative_position_bias_table - torch.Size([169, 12]): +Initialized by user-defined `init_weights` in WindowMSA + +backbone.stages.2.blocks.7.attn.w_msa.qkv.weight - torch.Size([1152, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.7.attn.w_msa.qkv.bias - torch.Size([1152]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.7.attn.w_msa.proj.weight - torch.Size([384, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.7.attn.w_msa.proj.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.7.norm2.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.7.norm2.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.7.ffn.layers.0.0.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.7.ffn.layers.0.0.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.7.ffn.layers.1.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.7.ffn.layers.1.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.8.norm1.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.8.norm1.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.8.attn.w_msa.relative_position_bias_table - torch.Size([169, 12]): +Initialized by user-defined `init_weights` in WindowMSA + +backbone.stages.2.blocks.8.attn.w_msa.qkv.weight - torch.Size([1152, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.8.attn.w_msa.qkv.bias - torch.Size([1152]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.8.attn.w_msa.proj.weight - torch.Size([384, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.8.attn.w_msa.proj.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.8.norm2.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.8.norm2.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.8.ffn.layers.0.0.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.8.ffn.layers.0.0.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.8.ffn.layers.1.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.8.ffn.layers.1.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.9.norm1.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.9.norm1.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.9.attn.w_msa.relative_position_bias_table - torch.Size([169, 12]): +Initialized by user-defined `init_weights` in WindowMSA + +backbone.stages.2.blocks.9.attn.w_msa.qkv.weight - torch.Size([1152, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.9.attn.w_msa.qkv.bias - torch.Size([1152]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.9.attn.w_msa.proj.weight - torch.Size([384, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.9.attn.w_msa.proj.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.9.norm2.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.9.norm2.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.9.ffn.layers.0.0.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.9.ffn.layers.0.0.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.9.ffn.layers.1.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.9.ffn.layers.1.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.10.norm1.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.10.norm1.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.10.attn.w_msa.relative_position_bias_table - torch.Size([169, 12]): +Initialized by user-defined `init_weights` in WindowMSA + +backbone.stages.2.blocks.10.attn.w_msa.qkv.weight - torch.Size([1152, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.10.attn.w_msa.qkv.bias - torch.Size([1152]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.10.attn.w_msa.proj.weight - torch.Size([384, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.10.attn.w_msa.proj.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.10.norm2.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.10.norm2.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.10.ffn.layers.0.0.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.10.ffn.layers.0.0.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.10.ffn.layers.1.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.10.ffn.layers.1.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.11.norm1.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.11.norm1.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.11.attn.w_msa.relative_position_bias_table - torch.Size([169, 12]): +Initialized by user-defined `init_weights` in WindowMSA + +backbone.stages.2.blocks.11.attn.w_msa.qkv.weight - torch.Size([1152, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.11.attn.w_msa.qkv.bias - torch.Size([1152]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.11.attn.w_msa.proj.weight - torch.Size([384, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.11.attn.w_msa.proj.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.11.norm2.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.11.norm2.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.11.ffn.layers.0.0.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.11.ffn.layers.0.0.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.11.ffn.layers.1.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.11.ffn.layers.1.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.12.norm1.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.12.norm1.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.12.attn.w_msa.relative_position_bias_table - torch.Size([169, 12]): +Initialized by user-defined `init_weights` in WindowMSA + +backbone.stages.2.blocks.12.attn.w_msa.qkv.weight - torch.Size([1152, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.12.attn.w_msa.qkv.bias - torch.Size([1152]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.12.attn.w_msa.proj.weight - torch.Size([384, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.12.attn.w_msa.proj.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.12.norm2.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.12.norm2.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.12.ffn.layers.0.0.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.12.ffn.layers.0.0.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.12.ffn.layers.1.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.12.ffn.layers.1.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.13.norm1.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.13.norm1.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.13.attn.w_msa.relative_position_bias_table - torch.Size([169, 12]): +Initialized by user-defined `init_weights` in WindowMSA + +backbone.stages.2.blocks.13.attn.w_msa.qkv.weight - torch.Size([1152, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.13.attn.w_msa.qkv.bias - torch.Size([1152]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.13.attn.w_msa.proj.weight - torch.Size([384, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.13.attn.w_msa.proj.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.13.norm2.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.13.norm2.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.13.ffn.layers.0.0.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.13.ffn.layers.0.0.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.13.ffn.layers.1.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.13.ffn.layers.1.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.14.norm1.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.14.norm1.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.14.attn.w_msa.relative_position_bias_table - torch.Size([169, 12]): +Initialized by user-defined `init_weights` in WindowMSA + +backbone.stages.2.blocks.14.attn.w_msa.qkv.weight - torch.Size([1152, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.14.attn.w_msa.qkv.bias - torch.Size([1152]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.14.attn.w_msa.proj.weight - torch.Size([384, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.14.attn.w_msa.proj.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.14.norm2.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.14.norm2.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.14.ffn.layers.0.0.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.14.ffn.layers.0.0.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.14.ffn.layers.1.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.14.ffn.layers.1.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.15.norm1.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.15.norm1.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.15.attn.w_msa.relative_position_bias_table - torch.Size([169, 12]): +Initialized by user-defined `init_weights` in WindowMSA + +backbone.stages.2.blocks.15.attn.w_msa.qkv.weight - torch.Size([1152, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.15.attn.w_msa.qkv.bias - torch.Size([1152]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.15.attn.w_msa.proj.weight - torch.Size([384, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.15.attn.w_msa.proj.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.15.norm2.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.15.norm2.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.15.ffn.layers.0.0.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.15.ffn.layers.0.0.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.15.ffn.layers.1.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.15.ffn.layers.1.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.16.norm1.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.16.norm1.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.16.attn.w_msa.relative_position_bias_table - torch.Size([169, 12]): +Initialized by user-defined `init_weights` in WindowMSA + +backbone.stages.2.blocks.16.attn.w_msa.qkv.weight - torch.Size([1152, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.16.attn.w_msa.qkv.bias - torch.Size([1152]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.16.attn.w_msa.proj.weight - torch.Size([384, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.16.attn.w_msa.proj.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.16.norm2.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.16.norm2.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.16.ffn.layers.0.0.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.16.ffn.layers.0.0.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.16.ffn.layers.1.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.16.ffn.layers.1.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.17.norm1.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.17.norm1.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.17.attn.w_msa.relative_position_bias_table - torch.Size([169, 12]): +Initialized by user-defined `init_weights` in WindowMSA + +backbone.stages.2.blocks.17.attn.w_msa.qkv.weight - torch.Size([1152, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.17.attn.w_msa.qkv.bias - torch.Size([1152]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.17.attn.w_msa.proj.weight - torch.Size([384, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.17.attn.w_msa.proj.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.17.norm2.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.17.norm2.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.17.ffn.layers.0.0.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.17.ffn.layers.0.0.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.17.ffn.layers.1.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.17.ffn.layers.1.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.downsample.norm.weight - torch.Size([1536]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.downsample.norm.bias - torch.Size([1536]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.downsample.reduction.weight - torch.Size([768, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.blocks.0.norm1.weight - torch.Size([768]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.3.blocks.0.norm1.bias - torch.Size([768]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.3.blocks.0.attn.w_msa.relative_position_bias_table - torch.Size([169, 24]): +Initialized by user-defined `init_weights` in WindowMSA + +backbone.stages.3.blocks.0.attn.w_msa.qkv.weight - torch.Size([2304, 768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.blocks.0.attn.w_msa.qkv.bias - torch.Size([2304]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.blocks.0.attn.w_msa.proj.weight - torch.Size([768, 768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.blocks.0.attn.w_msa.proj.bias - torch.Size([768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.blocks.0.norm2.weight - torch.Size([768]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.3.blocks.0.norm2.bias - torch.Size([768]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.3.blocks.0.ffn.layers.0.0.weight - torch.Size([3072, 768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.blocks.0.ffn.layers.0.0.bias - torch.Size([3072]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.blocks.0.ffn.layers.1.weight - torch.Size([768, 3072]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.blocks.0.ffn.layers.1.bias - torch.Size([768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.blocks.1.norm1.weight - torch.Size([768]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.3.blocks.1.norm1.bias - torch.Size([768]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.3.blocks.1.attn.w_msa.relative_position_bias_table - torch.Size([169, 24]): +Initialized by user-defined `init_weights` in WindowMSA + +backbone.stages.3.blocks.1.attn.w_msa.qkv.weight - torch.Size([2304, 768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.blocks.1.attn.w_msa.qkv.bias - torch.Size([2304]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.blocks.1.attn.w_msa.proj.weight - torch.Size([768, 768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.blocks.1.attn.w_msa.proj.bias - torch.Size([768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.blocks.1.norm2.weight - torch.Size([768]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.3.blocks.1.norm2.bias - torch.Size([768]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.3.blocks.1.ffn.layers.0.0.weight - torch.Size([3072, 768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.blocks.1.ffn.layers.0.0.bias - torch.Size([3072]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.blocks.1.ffn.layers.1.weight - torch.Size([768, 3072]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.blocks.1.ffn.layers.1.bias - torch.Size([768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.norm3.weight - torch.Size([768]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.norm3.bias - torch.Size([768]): +The value is the same before and after calling `init_weights` of ImageClassifier + +head.fc.weight - torch.Size([2, 768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +head.fc.bias - torch.Size([2]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 +2023/06/01 16:58:39 - mmengine - WARNING - "FileClient" will be deprecated in future. Please use io functions in https://mmengine.readthedocs.io/en/latest/api/fileio.html#file-io +2023/06/01 16:58:39 - mmengine - WARNING - "HardDiskBackend" is the alias of "LocalBackend" and the former will be deprecated in future. +2023/06/01 16:58:39 - mmengine - INFO - Checkpoints will be saved to /mnt/petrelfs/luzeyu/workspace/fakebench/mmpretrain/workdir/swin_small_4xb256_fake5m_lr1e-4_10epoch. +2023/06/01 17:00:06 - mmengine - INFO - Epoch(train) [1][ 100/5758] lr: 1.0000e-04 eta: 13:52:14 time: 0.7638 data_time: 0.2612 memory: 28798 grad_norm: 0.9949 loss: 0.5972 +2023/06/01 17:01:23 - mmengine - INFO - Epoch(train) [1][ 200/5758] lr: 1.0000e-04 eta: 13:05:39 time: 0.7551 data_time: 0.1275 memory: 28798 grad_norm: 1.0334 loss: 0.5732 +2023/06/01 17:02:43 - mmengine - INFO - Epoch(train) [1][ 300/5758] lr: 1.0000e-04 eta: 12:56:28 time: 0.7133 data_time: 0.0017 memory: 28798 grad_norm: 2.0384 loss: 0.5475 +2023/06/01 17:04:05 - mmengine - INFO - Epoch(train) [1][ 400/5758] lr: 1.0000e-04 eta: 12:55:52 time: 0.9228 data_time: 0.0038 memory: 28798 grad_norm: 2.4527 loss: 0.5087 +2023/06/01 17:05:24 - mmengine - INFO - Epoch(train) [1][ 500/5758] lr: 1.0000e-04 eta: 12:50:53 time: 0.8123 data_time: 0.0016 memory: 28798 grad_norm: 2.6107 loss: 0.5025 +2023/06/01 17:06:40 - mmengine - INFO - Epoch(train) [1][ 600/5758] lr: 1.0000e-04 eta: 12:41:35 time: 0.7984 data_time: 0.0016 memory: 28798 grad_norm: 3.8179 loss: 0.4701 +2023/06/01 17:07:59 - mmengine - INFO - Epoch(train) [1][ 700/5758] lr: 1.0000e-04 eta: 12:37:40 time: 0.7848 data_time: 0.0016 memory: 28798 grad_norm: 6.1250 loss: 0.4786 +2023/06/01 17:09:19 - mmengine - INFO - Epoch(train) [1][ 800/5758] lr: 1.0000e-04 eta: 12:36:23 time: 0.8083 data_time: 0.0020 memory: 28798 grad_norm: 5.4930 loss: 0.4538 +2023/06/01 17:10:37 - mmengine - INFO - Epoch(train) [1][ 900/5758] lr: 1.0000e-04 eta: 12:33:25 time: 0.7563 data_time: 0.0019 memory: 28798 grad_norm: 3.6155 loss: 0.4503 +2023/06/01 17:11:56 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_lr1e-4_10epoch_20230601_165803 +2023/06/01 17:11:56 - mmengine - INFO - Epoch(train) [1][1000/5758] lr: 1.0000e-04 eta: 12:31:16 time: 0.7647 data_time: 0.0014 memory: 28798 grad_norm: 5.3183 loss: 0.4268 +2023/06/01 17:13:14 - mmengine - INFO - Epoch(train) [1][1100/5758] lr: 1.0000e-04 eta: 12:28:14 time: 0.7644 data_time: 0.0020 memory: 28798 grad_norm: 7.1229 loss: 0.4310 +2023/06/01 17:14:34 - mmengine - INFO - Epoch(train) [1][1200/5758] lr: 1.0000e-04 eta: 12:27:47 time: 0.7858 data_time: 0.0016 memory: 28798 grad_norm: 8.1877 loss: 0.4295 +2023/06/01 17:15:54 - mmengine - INFO - Epoch(train) [1][1300/5758] lr: 1.0000e-04 eta: 12:26:33 time: 0.7782 data_time: 0.0018 memory: 28798 grad_norm: 6.0580 loss: 0.4054 +2023/06/01 17:17:14 - mmengine - INFO - Epoch(train) [1][1400/5758] lr: 1.0000e-04 eta: 12:25:37 time: 0.8470 data_time: 0.0020 memory: 28798 grad_norm: 7.2441 loss: 0.4202 +2023/06/01 17:18:32 - mmengine - INFO - Epoch(train) [1][1500/5758] lr: 1.0000e-04 eta: 12:23:10 time: 0.7499 data_time: 0.0016 memory: 28798 grad_norm: 9.4152 loss: 0.4133 +2023/06/01 17:19:49 - mmengine - INFO - Epoch(train) [1][1600/5758] lr: 1.0000e-04 eta: 12:20:43 time: 0.8087 data_time: 0.0019 memory: 28798 grad_norm: 10.4071 loss: 0.4110 +2023/06/01 17:21:09 - mmengine - INFO - Epoch(train) [1][1700/5758] lr: 1.0000e-04 eta: 12:19:30 time: 0.8089 data_time: 0.0024 memory: 28798 grad_norm: 8.9750 loss: 0.3980 +2023/06/01 17:22:27 - mmengine - INFO - Epoch(train) [1][1800/5758] lr: 1.0000e-04 eta: 12:17:17 time: 0.7795 data_time: 0.0021 memory: 28798 grad_norm: 8.6084 loss: 0.3878 +2023/06/01 17:23:45 - mmengine - INFO - Epoch(train) [1][1900/5758] lr: 1.0000e-04 eta: 12:15:26 time: 0.7407 data_time: 0.0022 memory: 28798 grad_norm: 7.9959 loss: 0.3900 +2023/06/01 17:25:06 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_lr1e-4_10epoch_20230601_165803 +2023/06/01 17:25:06 - mmengine - INFO - Epoch(train) [1][2000/5758] lr: 1.0000e-04 eta: 12:15:02 time: 0.7820 data_time: 0.0020 memory: 28798 grad_norm: 12.1932 loss: 0.4058 +2023/06/01 17:26:24 - mmengine - INFO - Epoch(train) [1][2100/5758] lr: 1.0000e-04 eta: 12:13:05 time: 0.7462 data_time: 0.0017 memory: 28798 grad_norm: 8.7543 loss: 0.3798 +2023/06/01 17:27:44 - mmengine - INFO - Epoch(train) [1][2200/5758] lr: 1.0000e-04 eta: 12:11:51 time: 0.7928 data_time: 0.0016 memory: 28798 grad_norm: 12.1650 loss: 0.3818 +2023/06/01 17:29:03 - mmengine - INFO - Epoch(train) [1][2300/5758] lr: 1.0000e-04 eta: 12:10:45 time: 0.8495 data_time: 0.0016 memory: 28798 grad_norm: 7.2061 loss: 0.3837 +2023/06/01 17:30:23 - mmengine - INFO - Epoch(train) [1][2400/5758] lr: 1.0000e-04 eta: 12:09:38 time: 0.7786 data_time: 0.0014 memory: 28798 grad_norm: 8.7690 loss: 0.3693 +2023/06/01 17:31:42 - mmengine - INFO - Epoch(train) [1][2500/5758] lr: 1.0000e-04 eta: 12:08:00 time: 0.8010 data_time: 0.0015 memory: 28798 grad_norm: 9.2640 loss: 0.3618 +2023/06/01 17:33:01 - mmengine - INFO - Epoch(train) [1][2600/5758] lr: 1.0000e-04 eta: 12:06:35 time: 0.7718 data_time: 0.0015 memory: 28798 grad_norm: 9.5435 loss: 0.4085 +2023/06/01 17:34:21 - mmengine - INFO - Epoch(train) [1][2700/5758] lr: 1.0000e-04 eta: 12:05:27 time: 0.7746 data_time: 0.0016 memory: 28798 grad_norm: 7.5530 loss: 0.3530 +2023/06/01 17:35:57 - mmengine - INFO - Epoch(train) [1][2800/5758] lr: 1.0000e-04 eta: 12:09:48 time: 0.8354 data_time: 0.2005 memory: 28798 grad_norm: 8.7739 loss: 0.3353 +2023/06/01 17:37:18 - mmengine - INFO - Epoch(train) [1][2900/5758] lr: 1.0000e-04 eta: 12:08:44 time: 0.7989 data_time: 0.0019 memory: 28798 grad_norm: 10.1943 loss: 0.3400 +2023/06/01 17:38:37 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_lr1e-4_10epoch_20230601_165803 +2023/06/01 17:38:37 - mmengine - INFO - Epoch(train) [1][3000/5758] lr: 1.0000e-04 eta: 12:06:55 time: 0.7651 data_time: 0.0018 memory: 28798 grad_norm: 10.7361 loss: 0.3538 +2023/06/01 17:39:55 - mmengine - INFO - Epoch(train) [1][3100/5758] lr: 1.0000e-04 eta: 12:05:14 time: 0.7593 data_time: 0.0020 memory: 28798 grad_norm: 8.4274 loss: 0.3421 +2023/06/01 17:41:16 - mmengine - INFO - Epoch(train) [1][3200/5758] lr: 1.0000e-04 eta: 12:04:10 time: 0.8447 data_time: 0.0016 memory: 28798 grad_norm: 9.4681 loss: 0.3274 +2023/06/01 17:42:37 - mmengine - INFO - Epoch(train) [1][3300/5758] lr: 1.0000e-04 eta: 12:03:06 time: 0.8461 data_time: 0.0017 memory: 28798 grad_norm: 11.8420 loss: 0.3391 +2023/06/01 17:43:57 - mmengine - INFO - Epoch(train) [1][3400/5758] lr: 1.0000e-04 eta: 12:01:53 time: 0.7893 data_time: 0.0020 memory: 28798 grad_norm: 14.0577 loss: 0.3493 +2023/06/01 17:45:19 - mmengine - INFO - Epoch(train) [1][3500/5758] lr: 1.0000e-04 eta: 12:00:52 time: 0.8102 data_time: 0.0016 memory: 28798 grad_norm: 11.6613 loss: 0.3422 +2023/06/01 17:46:38 - mmengine - INFO - Epoch(train) [1][3600/5758] lr: 1.0000e-04 eta: 11:59:29 time: 0.7679 data_time: 0.0016 memory: 28798 grad_norm: 12.7395 loss: 0.3535 +2023/06/01 17:48:00 - mmengine - INFO - Epoch(train) [1][3700/5758] lr: 1.0000e-04 eta: 11:58:37 time: 0.7919 data_time: 0.0019 memory: 28798 grad_norm: 14.6559 loss: 0.3317 +2023/06/01 17:49:20 - mmengine - INFO - Epoch(train) [1][3800/5758] lr: 1.0000e-04 eta: 11:57:20 time: 0.7980 data_time: 0.0017 memory: 28798 grad_norm: 8.0193 loss: 0.3204 +2023/06/01 17:50:41 - mmengine - INFO - Epoch(train) [1][3900/5758] lr: 1.0000e-04 eta: 11:56:07 time: 0.7871 data_time: 0.0018 memory: 28798 grad_norm: 16.5991 loss: 0.3389 +2023/06/01 17:52:04 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_lr1e-4_10epoch_20230601_165803 +2023/06/01 17:52:04 - mmengine - INFO - Epoch(train) [1][4000/5758] lr: 1.0000e-04 eta: 11:55:26 time: 0.9098 data_time: 0.0019 memory: 28798 grad_norm: 11.1319 loss: 0.3105 +2023/06/01 17:53:25 - mmengine - INFO - Epoch(train) [1][4100/5758] lr: 1.0000e-04 eta: 11:54:15 time: 0.8589 data_time: 0.0019 memory: 28798 grad_norm: 7.1359 loss: 0.3261 +2023/06/01 17:54:45 - mmengine - INFO - Epoch(train) [1][4200/5758] lr: 1.0000e-04 eta: 11:52:53 time: 0.7752 data_time: 0.0020 memory: 28798 grad_norm: 11.2859 loss: 0.3347 +2023/06/01 17:56:07 - mmengine - INFO - Epoch(train) [1][4300/5758] lr: 1.0000e-04 eta: 11:51:58 time: 0.7565 data_time: 0.0017 memory: 28798 grad_norm: 12.6451 loss: 0.3033 +2023/06/01 17:57:26 - mmengine - INFO - Epoch(train) [1][4400/5758] lr: 1.0000e-04 eta: 11:50:28 time: 0.8102 data_time: 0.0018 memory: 28798 grad_norm: 14.8535 loss: 0.3321 +2023/06/01 17:58:47 - mmengine - INFO - Epoch(train) [1][4500/5758] lr: 1.0000e-04 eta: 11:49:10 time: 0.8232 data_time: 0.0017 memory: 28798 grad_norm: 13.1495 loss: 0.3288 +2023/06/01 18:00:06 - mmengine - INFO - Epoch(train) [1][4600/5758] lr: 1.0000e-04 eta: 11:47:41 time: 0.7856 data_time: 0.0017 memory: 28798 grad_norm: 7.7415 loss: 0.2992 +2023/06/01 18:01:24 - mmengine - INFO - Epoch(train) [1][4700/5758] lr: 1.0000e-04 eta: 11:45:59 time: 0.7725 data_time: 0.0018 memory: 28798 grad_norm: 7.6865 loss: 0.2818 +2023/06/01 18:02:45 - mmengine - INFO - Epoch(train) [1][4800/5758] lr: 1.0000e-04 eta: 11:44:43 time: 0.7991 data_time: 0.0014 memory: 28798 grad_norm: 7.7781 loss: 0.2862 +2023/06/01 18:04:03 - mmengine - INFO - Epoch(train) [1][4900/5758] lr: 1.0000e-04 eta: 11:43:07 time: 0.8177 data_time: 0.0016 memory: 28798 grad_norm: 7.9361 loss: 0.2836 +2023/06/01 18:05:20 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_lr1e-4_10epoch_20230601_165803 +2023/06/01 18:05:20 - mmengine - INFO - Epoch(train) [1][5000/5758] lr: 1.0000e-04 eta: 11:41:06 time: 0.7379 data_time: 0.0017 memory: 28798 grad_norm: 8.3670 loss: 0.2897 +2023/06/01 18:06:38 - mmengine - INFO - Epoch(train) [1][5100/5758] lr: 1.0000e-04 eta: 11:39:33 time: 0.7638 data_time: 0.0015 memory: 28798 grad_norm: 7.8815 loss: 0.2753 +2023/06/01 18:07:54 - mmengine - INFO - Epoch(train) [1][5200/5758] lr: 1.0000e-04 eta: 11:37:35 time: 0.7932 data_time: 0.0016 memory: 28798 grad_norm: 5.6267 loss: 0.2834 +2023/06/01 18:09:13 - mmengine - INFO - Epoch(train) [1][5300/5758] lr: 1.0000e-04 eta: 11:36:01 time: 0.7624 data_time: 0.0014 memory: 28798 grad_norm: 17.1866 loss: 0.3126 +2023/06/01 18:10:31 - mmengine - INFO - Epoch(train) [1][5400/5758] lr: 1.0000e-04 eta: 11:34:26 time: 0.7655 data_time: 0.0016 memory: 28798 grad_norm: 7.6025 loss: 0.2828 +2023/06/01 18:11:47 - mmengine - INFO - Epoch(train) [1][5500/5758] lr: 1.0000e-04 eta: 11:32:28 time: 0.7662 data_time: 0.0016 memory: 28798 grad_norm: 14.8327 loss: 0.2980 +2023/06/01 18:13:04 - mmengine - INFO - Epoch(train) [1][5600/5758] lr: 1.0000e-04 eta: 11:30:43 time: 0.7696 data_time: 0.0016 memory: 28798 grad_norm: 9.5760 loss: 0.2863 +2023/06/01 18:14:20 - mmengine - INFO - Epoch(train) [1][5700/5758] lr: 1.0000e-04 eta: 11:28:50 time: 0.7377 data_time: 0.0017 memory: 28798 grad_norm: 14.9669 loss: 0.3145 +2023/06/01 18:15:05 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_lr1e-4_10epoch_20230601_165803 +2023/06/01 18:15:05 - mmengine - INFO - Saving checkpoint at 1 epochs +2023/06/01 18:15:26 - mmengine - INFO - Epoch(val) [1][16/16] accuracy/top1: 98.6364 single-label/precision_classwise: [100.0, 0.0] single-label/recall_classwise: [98.6364517211914, 0.0] single-label/f1-score_classwise: [99.31355285644531, 0.0] data_time: 0.2643 time: 0.8043 +2023/06/01 18:16:52 - mmengine - INFO - Epoch(train) [2][ 100/5758] lr: 9.7798e-05 eta: 11:27:29 time: 0.7853 data_time: 0.2253 memory: 28798 grad_norm: 4.7247 loss: 0.2758 +2023/06/01 18:18:07 - mmengine - INFO - Epoch(train) [2][ 200/5758] lr: 9.7798e-05 eta: 11:25:27 time: 0.7541 data_time: 0.1354 memory: 28783 grad_norm: 8.8290 loss: 0.2815 +2023/06/01 18:18:38 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_lr1e-4_10epoch_20230601_165803 +2023/06/01 18:19:22 - mmengine - INFO - Epoch(train) [2][ 300/5758] lr: 9.7798e-05 eta: 11:23:29 time: 0.7702 data_time: 0.2454 memory: 28783 grad_norm: 4.8542 loss: 0.2632 +2023/06/01 18:20:40 - mmengine - INFO - Epoch(train) [2][ 400/5758] lr: 9.7798e-05 eta: 11:21:57 time: 0.8204 data_time: 0.1613 memory: 28783 grad_norm: 6.4906 loss: 0.2696 +2023/06/01 18:21:56 - mmengine - INFO - Epoch(train) [2][ 500/5758] lr: 9.7798e-05 eta: 11:20:13 time: 0.7683 data_time: 0.0019 memory: 28783 grad_norm: 7.4814 loss: 0.2600 +2023/06/01 18:23:14 - mmengine - INFO - Epoch(train) [2][ 600/5758] lr: 9.7798e-05 eta: 11:18:40 time: 0.7667 data_time: 0.0017 memory: 28783 grad_norm: 3.6977 loss: 0.2583 +2023/06/01 18:24:34 - mmengine - INFO - Epoch(train) [2][ 700/5758] lr: 9.7798e-05 eta: 11:17:26 time: 0.8665 data_time: 0.0017 memory: 28783 grad_norm: 5.7077 loss: 0.2515 +2023/06/01 18:25:52 - mmengine - INFO - Epoch(train) [2][ 800/5758] lr: 9.7798e-05 eta: 11:15:52 time: 0.7268 data_time: 0.0015 memory: 28783 grad_norm: 4.4349 loss: 0.2566 +2023/06/01 18:27:10 - mmengine - INFO - Epoch(train) [2][ 900/5758] lr: 9.7798e-05 eta: 11:14:22 time: 0.7821 data_time: 0.0016 memory: 28783 grad_norm: 8.4204 loss: 0.2468 +2023/06/01 18:28:29 - mmengine - INFO - Epoch(train) [2][1000/5758] lr: 9.7798e-05 eta: 11:13:00 time: 0.8128 data_time: 0.0018 memory: 28783 grad_norm: 5.1028 loss: 0.2470 +2023/06/01 18:29:47 - mmengine - INFO - Epoch(train) [2][1100/5758] lr: 9.7798e-05 eta: 11:11:23 time: 0.7818 data_time: 0.0015 memory: 28783 grad_norm: 7.1841 loss: 0.2739 +2023/06/01 18:31:03 - mmengine - INFO - Epoch(train) [2][1200/5758] lr: 9.7798e-05 eta: 11:09:43 time: 0.7374 data_time: 0.0016 memory: 28783 grad_norm: 9.5210 loss: 0.2547 +2023/06/01 18:31:35 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_lr1e-4_10epoch_20230601_165803 +2023/06/01 18:32:21 - mmengine - INFO - Epoch(train) [2][1300/5758] lr: 9.7798e-05 eta: 11:08:14 time: 0.7829 data_time: 0.0020 memory: 28783 grad_norm: 7.7907 loss: 0.2447 +2023/06/01 18:33:39 - mmengine - INFO - Epoch(train) [2][1400/5758] lr: 9.7798e-05 eta: 11:06:46 time: 0.8394 data_time: 0.0019 memory: 28783 grad_norm: 6.5480 loss: 0.2479 +2023/06/01 18:34:55 - mmengine - INFO - Epoch(train) [2][1500/5758] lr: 9.7798e-05 eta: 11:05:05 time: 0.7773 data_time: 0.0016 memory: 28783 grad_norm: 4.4015 loss: 0.2535 +2023/06/01 18:36:14 - mmengine - INFO - Epoch(train) [2][1600/5758] lr: 9.7798e-05 eta: 11:03:38 time: 0.9168 data_time: 0.0024 memory: 28783 grad_norm: 3.6738 loss: 0.2360 +2023/06/01 18:37:32 - mmengine - INFO - Epoch(train) [2][1700/5758] lr: 9.7798e-05 eta: 11:02:14 time: 0.7193 data_time: 0.0017 memory: 28783 grad_norm: 11.9527 loss: 0.2416 +2023/06/01 18:38:47 - mmengine - INFO - Epoch(train) [2][1800/5758] lr: 9.7798e-05 eta: 11:00:27 time: 0.7895 data_time: 0.0018 memory: 28783 grad_norm: 1.9830 loss: 0.2413 +2023/06/01 18:40:04 - mmengine - INFO - Epoch(train) [2][1900/5758] lr: 9.7798e-05 eta: 10:58:52 time: 0.7554 data_time: 0.0017 memory: 28783 grad_norm: 5.4274 loss: 0.2391 +2023/06/01 18:41:20 - mmengine - INFO - Epoch(train) [2][2000/5758] lr: 9.7798e-05 eta: 10:57:12 time: 0.7589 data_time: 0.0018 memory: 28783 grad_norm: 4.1846 loss: 0.2352 +2023/06/01 18:42:33 - mmengine - INFO - Epoch(train) [2][2100/5758] lr: 9.7798e-05 eta: 10:55:12 time: 0.7662 data_time: 0.0023 memory: 28783 grad_norm: 4.5129 loss: 0.2320 +2023/06/01 18:43:46 - mmengine - INFO - Epoch(train) [2][2200/5758] lr: 9.7798e-05 eta: 10:53:14 time: 0.7455 data_time: 0.0017 memory: 28783 grad_norm: 5.2786 loss: 0.2440 +2023/06/01 18:44:17 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_lr1e-4_10epoch_20230601_165803 +2023/06/01 18:45:00 - mmengine - INFO - Epoch(train) [2][2300/5758] lr: 9.7798e-05 eta: 10:51:24 time: 0.7241 data_time: 0.0018 memory: 28783 grad_norm: 4.6993 loss: 0.2369 +2023/06/01 18:46:12 - mmengine - INFO - Epoch(train) [2][2400/5758] lr: 9.7798e-05 eta: 10:49:26 time: 0.7128 data_time: 0.0016 memory: 28783 grad_norm: 5.6312 loss: 0.2428 +2023/06/01 18:47:53 - mmengine - INFO - Epoch(train) [2][2500/5758] lr: 9.7798e-05 eta: 10:50:18 time: 0.7258 data_time: 0.0022 memory: 28783 grad_norm: 3.8033 loss: 0.2254 +2023/06/01 18:49:08 - mmengine - INFO - Epoch(train) [2][2600/5758] lr: 9.7798e-05 eta: 10:48:38 time: 0.7557 data_time: 0.0017 memory: 28783 grad_norm: 5.9559 loss: 0.2376 +2023/06/01 18:50:21 - mmengine - INFO - Epoch(train) [2][2700/5758] lr: 9.7798e-05 eta: 10:46:42 time: 0.7473 data_time: 0.0017 memory: 28783 grad_norm: 4.8319 loss: 0.2331 +2023/06/01 18:51:35 - mmengine - INFO - Epoch(train) [2][2800/5758] lr: 9.7798e-05 eta: 10:44:52 time: 0.8053 data_time: 0.0016 memory: 28783 grad_norm: 3.1565 loss: 0.2284 +2023/06/01 18:52:50 - mmengine - INFO - Epoch(train) [2][2900/5758] lr: 9.7798e-05 eta: 10:43:13 time: 0.7156 data_time: 0.0016 memory: 28783 grad_norm: 1.9715 loss: 0.2233 +2023/06/01 18:54:03 - mmengine - INFO - Epoch(train) [2][3000/5758] lr: 9.7798e-05 eta: 10:41:22 time: 0.6969 data_time: 0.0017 memory: 28783 grad_norm: 3.2040 loss: 0.2288 +2023/06/01 18:55:18 - mmengine - INFO - Epoch(train) [2][3100/5758] lr: 9.7798e-05 eta: 10:39:43 time: 0.7188 data_time: 0.0017 memory: 28783 grad_norm: 1.5660 loss: 0.2182 +2023/06/01 18:56:33 - mmengine - INFO - Epoch(train) [2][3200/5758] lr: 9.7798e-05 eta: 10:38:02 time: 0.8174 data_time: 0.0023 memory: 28783 grad_norm: 2.7557 loss: 0.2235 +2023/06/01 18:57:07 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_lr1e-4_10epoch_20230601_165803 +2023/06/01 18:57:49 - mmengine - INFO - Epoch(train) [2][3300/5758] lr: 9.7798e-05 eta: 10:36:31 time: 0.7033 data_time: 0.0015 memory: 28783 grad_norm: 2.5499 loss: 0.2221 +2023/06/01 18:59:03 - mmengine - INFO - Epoch(train) [2][3400/5758] lr: 9.7798e-05 eta: 10:34:44 time: 0.7431 data_time: 0.0016 memory: 28783 grad_norm: 3.8313 loss: 0.2244 +2023/06/01 19:00:15 - mmengine - INFO - Epoch(train) [2][3500/5758] lr: 9.7798e-05 eta: 10:32:53 time: 0.7462 data_time: 0.0020 memory: 28783 grad_norm: 5.1581 loss: 0.2317 +2023/06/01 19:01:29 - mmengine - INFO - Epoch(train) [2][3600/5758] lr: 9.7798e-05 eta: 10:31:09 time: 0.7830 data_time: 0.0016 memory: 28783 grad_norm: 3.8463 loss: 0.2214 +2023/06/01 19:02:42 - mmengine - INFO - Epoch(train) [2][3700/5758] lr: 9.7798e-05 eta: 10:29:22 time: 0.7665 data_time: 0.0016 memory: 28783 grad_norm: 3.1762 loss: 0.2285 +2023/06/01 19:03:57 - mmengine - INFO - Epoch(train) [2][3800/5758] lr: 9.7798e-05 eta: 10:27:49 time: 0.7468 data_time: 0.0017 memory: 28783 grad_norm: 7.0664 loss: 0.2182 +2023/06/01 19:05:13 - mmengine - INFO - Epoch(train) [2][3900/5758] lr: 9.7798e-05 eta: 10:26:16 time: 0.7534 data_time: 0.0016 memory: 28783 grad_norm: 2.3539 loss: 0.2267 +2023/06/01 19:06:28 - mmengine - INFO - Epoch(train) [2][4000/5758] lr: 9.7798e-05 eta: 10:24:43 time: 0.7146 data_time: 0.0017 memory: 28783 grad_norm: 2.2450 loss: 0.2216 +2023/06/01 19:07:44 - mmengine - INFO - Epoch(train) [2][4100/5758] lr: 9.7798e-05 eta: 10:23:14 time: 0.7444 data_time: 0.0020 memory: 28783 grad_norm: 3.1566 loss: 0.2208 +2023/06/01 19:09:00 - mmengine - INFO - Epoch(train) [2][4200/5758] lr: 9.7798e-05 eta: 10:21:40 time: 0.7958 data_time: 0.0016 memory: 28783 grad_norm: 4.5439 loss: 0.2200 +2023/06/01 19:09:30 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_lr1e-4_10epoch_20230601_165803 +2023/06/01 19:10:15 - mmengine - INFO - Epoch(train) [2][4300/5758] lr: 9.7798e-05 eta: 10:20:10 time: 0.8050 data_time: 0.0016 memory: 28783 grad_norm: 2.7435 loss: 0.2216 +2023/06/01 19:11:32 - mmengine - INFO - Epoch(train) [2][4400/5758] lr: 9.7798e-05 eta: 10:18:45 time: 0.7350 data_time: 0.0026 memory: 28783 grad_norm: 3.7366 loss: 0.2291 +2023/06/01 19:12:47 - mmengine - INFO - Epoch(train) [2][4500/5758] lr: 9.7798e-05 eta: 10:17:10 time: 0.7108 data_time: 0.0014 memory: 28783 grad_norm: 4.9828 loss: 0.2223 +2023/06/01 19:14:01 - mmengine - INFO - Epoch(train) [2][4600/5758] lr: 9.7798e-05 eta: 10:15:34 time: 0.7626 data_time: 0.0020 memory: 28783 grad_norm: 1.8968 loss: 0.2178 +2023/06/01 19:15:16 - mmengine - INFO - Epoch(train) [2][4700/5758] lr: 9.7798e-05 eta: 10:14:00 time: 0.7143 data_time: 0.0020 memory: 28783 grad_norm: 1.9122 loss: 0.2100 +2023/06/01 19:16:31 - mmengine - INFO - Epoch(train) [2][4800/5758] lr: 9.7798e-05 eta: 10:12:25 time: 0.8343 data_time: 0.0018 memory: 28783 grad_norm: 2.5711 loss: 0.2190 +2023/06/01 19:18:04 - mmengine - INFO - Epoch(train) [2][4900/5758] lr: 9.7798e-05 eta: 10:12:13 time: 2.6420 data_time: 1.8293 memory: 28783 grad_norm: 2.8795 loss: 0.2232 +2023/06/01 19:19:21 - mmengine - INFO - Epoch(train) [2][5000/5758] lr: 9.7798e-05 eta: 10:10:52 time: 0.7352 data_time: 0.2317 memory: 28783 grad_norm: 1.2968 loss: 0.2141 +2023/06/01 19:20:37 - mmengine - INFO - Epoch(train) [2][5100/5758] lr: 9.7798e-05 eta: 10:09:24 time: 0.7945 data_time: 0.2769 memory: 28783 grad_norm: 3.4218 loss: 0.2206 +2023/06/01 19:21:50 - mmengine - INFO - Epoch(train) [2][5200/5758] lr: 9.7798e-05 eta: 10:07:43 time: 0.6702 data_time: 0.1600 memory: 28783 grad_norm: 1.7300 loss: 0.2163 +2023/06/01 19:22:21 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_lr1e-4_10epoch_20230601_165803 +2023/06/01 19:23:01 - mmengine - INFO - Epoch(train) [2][5300/5758] lr: 9.7798e-05 eta: 10:05:53 time: 0.6916 data_time: 0.1896 memory: 28783 grad_norm: 1.4095 loss: 0.2242 +2023/06/01 19:24:12 - mmengine - INFO - Epoch(train) [2][5400/5758] lr: 9.7798e-05 eta: 10:04:06 time: 0.6839 data_time: 0.1011 memory: 28783 grad_norm: 2.3308 loss: 0.2176 +2023/06/01 19:25:25 - mmengine - INFO - Epoch(train) [2][5500/5758] lr: 9.7798e-05 eta: 10:02:28 time: 0.7198 data_time: 0.0015 memory: 28783 grad_norm: 1.9042 loss: 0.2168 +2023/06/01 19:26:40 - mmengine - INFO - Epoch(train) [2][5600/5758] lr: 9.7798e-05 eta: 10:00:55 time: 0.7231 data_time: 0.0018 memory: 28783 grad_norm: 2.2148 loss: 0.2167 +2023/06/01 19:27:54 - mmengine - INFO - Epoch(train) [2][5700/5758] lr: 9.7798e-05 eta: 9:59:22 time: 0.7298 data_time: 0.0018 memory: 28783 grad_norm: 1.5354 loss: 0.2215 +2023/06/01 19:28:39 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_lr1e-4_10epoch_20230601_165803 +2023/06/01 19:28:39 - mmengine - INFO - Saving checkpoint at 2 epochs +2023/06/01 19:28:57 - mmengine - INFO - Epoch(val) [2][16/16] accuracy/top1: 99.5770 single-label/precision_classwise: [100.0, 0.0] single-label/recall_classwise: [99.57704162597656, 0.0] single-label/f1-score_classwise: [99.78807067871094, 0.0] data_time: 0.2030 time: 0.6183 +2023/06/01 19:30:18 - mmengine - INFO - Epoch(train) [3][ 100/5758] lr: 9.1406e-05 eta: 9:57:30 time: 0.7363 data_time: 0.0017 memory: 28783 grad_norm: 2.5473 loss: 0.2204 +2023/06/01 19:31:33 - mmengine - INFO - Epoch(train) [3][ 200/5758] lr: 9.1406e-05 eta: 9:55:59 time: 0.7760 data_time: 0.0017 memory: 28783 grad_norm: 1.9136 loss: 0.2122 +2023/06/01 19:32:47 - mmengine - INFO - Epoch(train) [3][ 300/5758] lr: 9.1406e-05 eta: 9:54:27 time: 0.7625 data_time: 0.0021 memory: 28783 grad_norm: 1.8601 loss: 0.2152 +2023/06/01 19:34:02 - mmengine - INFO - Epoch(train) [3][ 400/5758] lr: 9.1406e-05 eta: 9:52:56 time: 0.7394 data_time: 0.0018 memory: 28783 grad_norm: 1.5550 loss: 0.2136 +2023/06/01 19:35:05 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_lr1e-4_10epoch_20230601_165803 +2023/06/01 19:35:16 - mmengine - INFO - Epoch(train) [3][ 500/5758] lr: 9.1406e-05 eta: 9:51:24 time: 0.7282 data_time: 0.0018 memory: 28783 grad_norm: 2.9241 loss: 0.2126 +2023/06/01 19:36:33 - mmengine - INFO - Epoch(train) [3][ 600/5758] lr: 9.1406e-05 eta: 9:50:02 time: 0.8083 data_time: 0.0017 memory: 28783 grad_norm: 2.0709 loss: 0.2125 +2023/06/01 19:37:49 - mmengine - INFO - Epoch(train) [3][ 700/5758] lr: 9.1406e-05 eta: 9:48:37 time: 0.7609 data_time: 0.0017 memory: 28783 grad_norm: 2.1444 loss: 0.2145 +2023/06/01 19:39:11 - mmengine - INFO - Epoch(train) [3][ 800/5758] lr: 9.1406e-05 eta: 9:47:36 time: 0.8115 data_time: 0.0017 memory: 28783 grad_norm: 1.3800 loss: 0.2140 +2023/06/01 19:40:28 - mmengine - INFO - Epoch(train) [3][ 900/5758] lr: 9.1406e-05 eta: 9:46:16 time: 0.7176 data_time: 0.0015 memory: 28783 grad_norm: 1.5817 loss: 0.2098 +2023/06/01 19:41:47 - mmengine - INFO - Epoch(train) [3][1000/5758] lr: 9.1406e-05 eta: 9:45:02 time: 0.8914 data_time: 0.0018 memory: 28783 grad_norm: 1.1911 loss: 0.2122 +2023/06/01 19:43:07 - mmengine - INFO - Epoch(train) [3][1100/5758] lr: 9.1406e-05 eta: 9:43:50 time: 0.7900 data_time: 0.0018 memory: 28783 grad_norm: 1.8294 loss: 0.2123 +2023/06/01 19:44:25 - mmengine - INFO - Epoch(train) [3][1200/5758] lr: 9.1406e-05 eta: 9:42:34 time: 0.8613 data_time: 0.1020 memory: 28783 grad_norm: 1.2541 loss: 0.2087 +2023/06/01 19:45:47 - mmengine - INFO - Epoch(train) [3][1300/5758] lr: 9.1406e-05 eta: 9:41:28 time: 0.8129 data_time: 0.0017 memory: 28783 grad_norm: 1.3484 loss: 0.2111 +2023/06/01 19:47:05 - mmengine - INFO - Epoch(train) [3][1400/5758] lr: 9.1406e-05 eta: 9:40:13 time: 0.8086 data_time: 0.0018 memory: 28783 grad_norm: 1.1029 loss: 0.2110 +2023/06/01 19:48:16 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_lr1e-4_10epoch_20230601_165803 +2023/06/01 19:48:27 - mmengine - INFO - Epoch(train) [3][1500/5758] lr: 9.1406e-05 eta: 9:39:06 time: 0.7960 data_time: 0.0019 memory: 28783 grad_norm: 1.2602 loss: 0.2155 +2023/06/01 19:49:48 - mmengine - INFO - Epoch(train) [3][1600/5758] lr: 9.1406e-05 eta: 9:38:02 time: 0.7560 data_time: 0.0019 memory: 28783 grad_norm: 1.3525 loss: 0.2087 +2023/06/01 19:51:07 - mmengine - INFO - Epoch(train) [3][1700/5758] lr: 9.1406e-05 eta: 9:36:44 time: 0.7298 data_time: 0.0016 memory: 28783 grad_norm: 1.2084 loss: 0.2128 +2023/06/01 19:52:28 - mmengine - INFO - Epoch(train) [3][1800/5758] lr: 9.1406e-05 eta: 9:35:38 time: 0.8350 data_time: 0.0015 memory: 28783 grad_norm: 1.3942 loss: 0.2107 +2023/06/01 19:53:47 - mmengine - INFO - Epoch(train) [3][1900/5758] lr: 9.1406e-05 eta: 9:34:22 time: 0.7560 data_time: 0.0023 memory: 28783 grad_norm: 2.2864 loss: 0.2123 +2023/06/01 19:55:04 - mmengine - INFO - Epoch(train) [3][2000/5758] lr: 9.1406e-05 eta: 9:33:03 time: 0.7641 data_time: 0.0018 memory: 28783 grad_norm: 1.2045 loss: 0.2095 +2023/06/01 19:56:20 - mmengine - INFO - Epoch(train) [3][2100/5758] lr: 9.1406e-05 eta: 9:31:36 time: 0.7750 data_time: 0.0019 memory: 28783 grad_norm: 1.4417 loss: 0.2154 +2023/06/01 19:57:35 - mmengine - INFO - Epoch(train) [3][2200/5758] lr: 9.1406e-05 eta: 9:30:08 time: 0.7901 data_time: 0.0956 memory: 28783 grad_norm: 1.5441 loss: 0.2142 +2023/06/01 19:58:50 - mmengine - INFO - Epoch(train) [3][2300/5758] lr: 9.1406e-05 eta: 9:28:43 time: 0.7652 data_time: 0.0605 memory: 28783 grad_norm: 2.2110 loss: 0.2095 +2023/06/01 20:00:11 - mmengine - INFO - Epoch(train) [3][2400/5758] lr: 9.1406e-05 eta: 9:27:34 time: 0.8278 data_time: 0.0014 memory: 28783 grad_norm: 1.0306 loss: 0.2092 +2023/06/01 20:01:16 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_lr1e-4_10epoch_20230601_165803 +2023/06/01 20:01:28 - mmengine - INFO - Epoch(train) [3][2500/5758] lr: 9.1406e-05 eta: 9:26:12 time: 0.7532 data_time: 0.1077 memory: 28783 grad_norm: 2.0192 loss: 0.2223 +2023/06/01 20:02:49 - mmengine - INFO - Epoch(train) [3][2600/5758] lr: 9.1406e-05 eta: 9:25:03 time: 0.7804 data_time: 0.2761 memory: 28783 grad_norm: 0.9989 loss: 0.2160 +2023/06/01 20:04:07 - mmengine - INFO - Epoch(train) [3][2700/5758] lr: 9.1406e-05 eta: 9:23:46 time: 0.7172 data_time: 0.2132 memory: 28783 grad_norm: 1.3192 loss: 0.2080 +2023/06/01 20:05:23 - mmengine - INFO - Epoch(train) [3][2800/5758] lr: 9.1406e-05 eta: 9:22:23 time: 0.7647 data_time: 0.2608 memory: 28783 grad_norm: 1.3700 loss: 0.2111 +2023/06/01 20:06:38 - mmengine - INFO - Epoch(train) [3][2900/5758] lr: 9.1406e-05 eta: 9:20:53 time: 0.7368 data_time: 0.1925 memory: 28783 grad_norm: 1.3877 loss: 0.2132 +2023/06/01 20:07:54 - mmengine - INFO - Epoch(train) [3][3000/5758] lr: 9.1406e-05 eta: 9:19:32 time: 0.7891 data_time: 0.0015 memory: 28783 grad_norm: 1.2400 loss: 0.2062 +2023/06/01 20:09:08 - mmengine - INFO - Epoch(train) [3][3100/5758] lr: 9.1406e-05 eta: 9:18:00 time: 0.7504 data_time: 0.0019 memory: 28783 grad_norm: 2.0698 loss: 0.2121 +2023/06/01 20:10:27 - mmengine - INFO - Epoch(train) [3][3200/5758] lr: 9.1406e-05 eta: 9:16:46 time: 0.7860 data_time: 0.0018 memory: 28783 grad_norm: 2.0235 loss: 0.2098 +2023/06/01 20:11:41 - mmengine - INFO - Epoch(train) [3][3300/5758] lr: 9.1406e-05 eta: 9:15:18 time: 0.7156 data_time: 0.0015 memory: 28783 grad_norm: 1.0350 loss: 0.2101 +2023/06/01 20:12:58 - mmengine - INFO - Epoch(train) [3][3400/5758] lr: 9.1406e-05 eta: 9:13:56 time: 0.7649 data_time: 0.0016 memory: 28783 grad_norm: 1.1589 loss: 0.2069 +2023/06/01 20:13:59 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_lr1e-4_10epoch_20230601_165803 +2023/06/01 20:14:12 - mmengine - INFO - Epoch(train) [3][3500/5758] lr: 9.1406e-05 eta: 9:12:28 time: 0.7189 data_time: 0.0016 memory: 28783 grad_norm: 1.2026 loss: 0.2093 +2023/06/01 20:15:28 - mmengine - INFO - Epoch(train) [3][3600/5758] lr: 9.1406e-05 eta: 9:11:04 time: 0.7618 data_time: 0.0016 memory: 28783 grad_norm: 0.9875 loss: 0.2076 +2023/06/01 20:16:45 - mmengine - INFO - Epoch(train) [3][3700/5758] lr: 9.1406e-05 eta: 9:09:44 time: 0.8212 data_time: 0.0016 memory: 28783 grad_norm: 1.5115 loss: 0.2080 +2023/06/01 20:18:01 - mmengine - INFO - Epoch(train) [3][3800/5758] lr: 9.1406e-05 eta: 9:08:22 time: 0.7386 data_time: 0.0016 memory: 28783 grad_norm: 1.2106 loss: 0.2109 +2023/06/01 20:19:16 - mmengine - INFO - Epoch(train) [3][3900/5758] lr: 9.1406e-05 eta: 9:06:54 time: 0.7756 data_time: 0.0018 memory: 28783 grad_norm: 0.9896 loss: 0.2101 +2023/06/01 20:20:35 - mmengine - INFO - Epoch(train) [3][4000/5758] lr: 9.1406e-05 eta: 9:05:41 time: 0.7412 data_time: 0.0489 memory: 28783 grad_norm: 2.0264 loss: 0.2136 +2023/06/01 20:21:54 - mmengine - INFO - Epoch(train) [3][4100/5758] lr: 9.1406e-05 eta: 9:04:25 time: 0.7665 data_time: 0.0016 memory: 28783 grad_norm: 1.3295 loss: 0.2107 +2023/06/01 20:23:12 - mmengine - INFO - Epoch(train) [3][4200/5758] lr: 9.1406e-05 eta: 9:03:08 time: 0.8191 data_time: 0.0015 memory: 28783 grad_norm: 1.3675 loss: 0.2056 +2023/06/01 20:24:31 - mmengine - INFO - Epoch(train) [3][4300/5758] lr: 9.1406e-05 eta: 9:01:53 time: 0.8034 data_time: 0.0019 memory: 28783 grad_norm: 1.2479 loss: 0.2056 +2023/06/01 20:25:48 - mmengine - INFO - Epoch(train) [3][4400/5758] lr: 9.1406e-05 eta: 9:00:34 time: 0.7314 data_time: 0.0015 memory: 28783 grad_norm: 1.1334 loss: 0.2112 +2023/06/01 20:26:54 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_lr1e-4_10epoch_20230601_165803 +2023/06/01 20:27:06 - mmengine - INFO - Epoch(train) [3][4500/5758] lr: 9.1406e-05 eta: 8:59:16 time: 0.7656 data_time: 0.0015 memory: 28783 grad_norm: 1.0096 loss: 0.2104 +2023/06/01 20:28:24 - mmengine - INFO - Epoch(train) [3][4600/5758] lr: 9.1406e-05 eta: 8:57:59 time: 0.7799 data_time: 0.0017 memory: 28783 grad_norm: 0.7067 loss: 0.2060 +2023/06/01 20:29:43 - mmengine - INFO - Epoch(train) [3][4700/5758] lr: 9.1406e-05 eta: 8:56:43 time: 0.8069 data_time: 0.0015 memory: 28783 grad_norm: 0.7893 loss: 0.2095 +2023/06/01 20:31:59 - mmengine - INFO - Epoch(train) [3][4800/5758] lr: 9.1406e-05 eta: 8:57:52 time: 0.9077 data_time: 0.0016 memory: 28783 grad_norm: 1.3125 loss: 0.2024 +2023/06/01 20:33:14 - mmengine - INFO - Epoch(train) [3][4900/5758] lr: 9.1406e-05 eta: 8:56:27 time: 0.7250 data_time: 0.0015 memory: 28783 grad_norm: 1.2686 loss: 0.2104 +2023/06/01 20:34:30 - mmengine - INFO - Epoch(train) [3][5000/5758] lr: 9.1406e-05 eta: 8:55:02 time: 0.7545 data_time: 0.0018 memory: 28783 grad_norm: 0.6721 loss: 0.2051 +2023/06/01 20:35:48 - mmengine - INFO - Epoch(train) [3][5100/5758] lr: 9.1406e-05 eta: 8:53:44 time: 0.7651 data_time: 0.0019 memory: 28783 grad_norm: 1.4056 loss: 0.2102 +2023/06/01 20:37:04 - mmengine - INFO - Epoch(train) [3][5200/5758] lr: 9.1406e-05 eta: 8:52:20 time: 0.6973 data_time: 0.0018 memory: 28783 grad_norm: 1.3470 loss: 0.2093 +2023/06/01 20:38:21 - mmengine - INFO - Epoch(train) [3][5300/5758] lr: 9.1406e-05 eta: 8:50:59 time: 0.7734 data_time: 0.0018 memory: 28783 grad_norm: 1.3098 loss: 0.2061 +2023/06/01 20:39:39 - mmengine - INFO - Epoch(train) [3][5400/5758] lr: 9.1406e-05 eta: 8:49:40 time: 0.8132 data_time: 0.0017 memory: 28783 grad_norm: 0.8919 loss: 0.2075 +2023/06/01 20:40:45 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_lr1e-4_10epoch_20230601_165803 +2023/06/01 20:40:59 - mmengine - INFO - Epoch(train) [3][5500/5758] lr: 9.1406e-05 eta: 8:48:27 time: 0.7260 data_time: 0.0015 memory: 28783 grad_norm: 0.9888 loss: 0.2055 +2023/06/01 20:42:16 - mmengine - INFO - Epoch(train) [3][5600/5758] lr: 9.1406e-05 eta: 8:47:07 time: 0.8159 data_time: 0.0016 memory: 28783 grad_norm: 0.9398 loss: 0.2074 +2023/06/01 20:43:32 - mmengine - INFO - Epoch(train) [3][5700/5758] lr: 9.1406e-05 eta: 8:45:43 time: 0.7101 data_time: 0.0016 memory: 28783 grad_norm: 0.9887 loss: 0.2099 +2023/06/01 20:44:16 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_lr1e-4_10epoch_20230601_165803 +2023/06/01 20:44:16 - mmengine - INFO - Saving checkpoint at 3 epochs +2023/06/01 20:44:34 - mmengine - INFO - Epoch(val) [3][16/16] accuracy/top1: 99.4255 single-label/precision_classwise: [100.0, 0.0] single-label/recall_classwise: [99.42554473876953, 0.0] single-label/f1-score_classwise: [99.71195220947266, 0.0] data_time: 0.2030 time: 0.6163 +2023/06/01 20:45:56 - mmengine - INFO - Epoch(train) [4][ 100/5758] lr: 8.1450e-05 eta: 8:43:46 time: 0.7449 data_time: 0.0713 memory: 28783 grad_norm: 0.7140 loss: 0.2045 +2023/06/01 20:47:11 - mmengine - INFO - Epoch(train) [4][ 200/5758] lr: 8.1450e-05 eta: 8:42:21 time: 0.7120 data_time: 0.0017 memory: 28783 grad_norm: 0.8423 loss: 0.2116 +2023/06/01 20:48:29 - mmengine - INFO - Epoch(train) [4][ 300/5758] lr: 8.1450e-05 eta: 8:41:02 time: 0.7980 data_time: 0.0016 memory: 28783 grad_norm: 0.9320 loss: 0.2066 +2023/06/01 20:49:42 - mmengine - INFO - Epoch(train) [4][ 400/5758] lr: 8.1450e-05 eta: 8:39:32 time: 0.7049 data_time: 0.0018 memory: 28783 grad_norm: 0.9375 loss: 0.2076 +2023/06/01 20:51:05 - mmengine - INFO - Epoch(train) [4][ 500/5758] lr: 8.1450e-05 eta: 8:38:26 time: 0.8853 data_time: 0.0018 memory: 28783 grad_norm: 1.1117 loss: 0.2070 +2023/06/01 20:52:39 - mmengine - INFO - Epoch(train) [4][ 600/5758] lr: 8.1450e-05 eta: 8:37:42 time: 0.7465 data_time: 0.0017 memory: 28783 grad_norm: 1.1212 loss: 0.2076 +2023/06/01 20:54:15 - mmengine - INFO - Epoch(train) [4][ 700/5758] lr: 8.1450e-05 eta: 8:37:03 time: 0.9494 data_time: 0.0016 memory: 28783 grad_norm: 0.8347 loss: 0.2067 +2023/06/01 20:54:44 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_lr1e-4_10epoch_20230601_165803 +2023/06/01 20:55:55 - mmengine - INFO - Epoch(train) [4][ 800/5758] lr: 8.1450e-05 eta: 8:36:33 time: 1.0260 data_time: 0.0015 memory: 28783 grad_norm: 1.3261 loss: 0.2065 +2023/06/01 20:57:36 - mmengine - INFO - Epoch(train) [4][ 900/5758] lr: 8.1450e-05 eta: 8:36:03 time: 1.0383 data_time: 0.0687 memory: 28783 grad_norm: 0.8589 loss: 0.2091 +2023/06/01 20:59:08 - mmengine - INFO - Epoch(train) [4][1000/5758] lr: 8.1450e-05 eta: 8:35:14 time: 0.7482 data_time: 0.0016 memory: 28783 grad_norm: 0.9770 loss: 0.2069 +2023/06/01 21:00:28 - mmengine - INFO - Epoch(train) [4][1100/5758] lr: 8.1450e-05 eta: 8:33:58 time: 0.7960 data_time: 0.0017 memory: 28783 grad_norm: 0.8574 loss: 0.2040 +2023/06/01 21:01:54 - mmengine - INFO - Epoch(train) [4][1200/5758] lr: 8.1450e-05 eta: 8:32:54 time: 0.8132 data_time: 0.0017 memory: 28783 grad_norm: 1.1316 loss: 0.2094 +2023/06/01 21:03:20 - mmengine - INFO - Epoch(train) [4][1300/5758] lr: 8.1450e-05 eta: 8:31:51 time: 0.8543 data_time: 0.0015 memory: 28783 grad_norm: 1.2810 loss: 0.2054 +2023/06/01 21:04:43 - mmengine - INFO - Epoch(train) [4][1400/5758] lr: 8.1450e-05 eta: 8:30:41 time: 0.8720 data_time: 0.0017 memory: 28783 grad_norm: 0.9205 loss: 0.2044 +2023/06/01 21:06:23 - mmengine - INFO - Epoch(train) [4][1500/5758] lr: 8.1450e-05 eta: 8:30:06 time: 1.4627 data_time: 0.0016 memory: 28783 grad_norm: 1.2260 loss: 0.2043 +2023/06/01 21:08:10 - mmengine - INFO - Epoch(train) [4][1600/5758] lr: 8.1450e-05 eta: 8:29:46 time: 1.0631 data_time: 0.0016 memory: 28783 grad_norm: 0.8021 loss: 0.2101 +2023/06/01 21:09:56 - mmengine - INFO - Epoch(train) [4][1700/5758] lr: 8.1450e-05 eta: 8:29:22 time: 0.9851 data_time: 0.0018 memory: 28783 grad_norm: 1.4101 loss: 0.2081 +2023/06/01 21:10:26 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_lr1e-4_10epoch_20230601_165803 +2023/06/01 21:11:53 - mmengine - INFO - Epoch(train) [4][1800/5758] lr: 8.1450e-05 eta: 8:29:17 time: 1.1624 data_time: 0.0014 memory: 28783 grad_norm: 1.8440 loss: 0.2100 +2023/06/01 21:13:53 - mmengine - INFO - Epoch(train) [4][1900/5758] lr: 8.1450e-05 eta: 8:29:20 time: 1.0555 data_time: 0.4064 memory: 28783 grad_norm: 0.7899 loss: 0.2046 +2023/06/01 21:15:42 - mmengine - INFO - Epoch(train) [4][2000/5758] lr: 8.1450e-05 eta: 8:29:00 time: 1.1866 data_time: 0.6836 memory: 28783 grad_norm: 0.8104 loss: 0.2088 +2023/06/01 21:17:13 - mmengine - INFO - Epoch(train) [4][2100/5758] lr: 8.1450e-05 eta: 8:28:01 time: 0.7386 data_time: 0.2251 memory: 28783 grad_norm: 0.7875 loss: 0.2059 +2023/06/01 21:18:24 - mmengine - INFO - Epoch(train) [4][2200/5758] lr: 8.1450e-05 eta: 8:26:25 time: 0.7393 data_time: 0.2411 memory: 28783 grad_norm: 1.3534 loss: 0.2081 +2023/06/01 21:19:36 - mmengine - INFO - Epoch(train) [4][2300/5758] lr: 8.1450e-05 eta: 8:24:49 time: 0.7056 data_time: 0.1747 memory: 28783 grad_norm: 0.9966 loss: 0.2029 diff --git a/swin_small_4xb256_fake5m_lr1e-4_10epoch/20230601_165803/vis_data/20230601_165803.json b/swin_small_4xb256_fake5m_lr1e-4_10epoch/20230601_165803/vis_data/20230601_165803.json new file mode 100644 index 0000000000000000000000000000000000000000..4e6765d99e3dadc5047b8ab4363a2e5902d577b9 --- /dev/null +++ b/swin_small_4xb256_fake5m_lr1e-4_10epoch/20230601_165803/vis_data/20230601_165803.json @@ -0,0 +1,197 @@ +{"lr": 0.0001, "data_time": 0.2611548662185669, "grad_norm": 0.9948760330677032, "loss": 0.5971919178962708, "time": 0.763830041885376, "epoch": 1, "memory": 28798, "step": 100} +{"lr": 0.0001, "data_time": 0.12748069763183595, "grad_norm": 1.033381861448288, "loss": 0.5732303202152252, "time": 0.7551178455352783, "epoch": 1, "memory": 28798, "step": 200} +{"lr": 0.0001, "data_time": 0.001685333251953125, "grad_norm": 2.0383641600608824, "loss": 0.5474633693695068, "time": 0.7133275032043457, "epoch": 1, "memory": 28798, "step": 300} +{"lr": 0.0001, "data_time": 0.0037578344345092773, "grad_norm": 2.4526761531829835, "loss": 0.5087024360895157, "time": 0.9227797508239746, "epoch": 1, "memory": 28798, "step": 400} +{"lr": 0.0001, "data_time": 0.0015799760818481444, "grad_norm": 2.6106780767440796, "loss": 0.502456533908844, "time": 0.8122771739959717, "epoch": 1, "memory": 28798, "step": 500} +{"lr": 0.0001, "data_time": 0.0016286849975585937, "grad_norm": 3.8179392337799074, "loss": 0.4700884073972702, "time": 0.798379921913147, "epoch": 1, "memory": 28798, "step": 600} +{"lr": 0.0001, "data_time": 0.0016109704971313476, "grad_norm": 6.124994063377381, "loss": 0.4786356151103973, "time": 0.7847991943359375, "epoch": 1, "memory": 28798, "step": 700} +{"lr": 0.0001, "data_time": 0.002045273780822754, "grad_norm": 5.492973017692566, "loss": 0.45380268096923826, "time": 0.8083441019058227, "epoch": 1, "memory": 28798, "step": 800} +{"lr": 0.0001, "data_time": 0.0018804311752319337, "grad_norm": 3.615518593788147, "loss": 0.450284481048584, "time": 0.7562673330307007, "epoch": 1, "memory": 28798, "step": 900} +{"lr": 0.0001, "data_time": 0.0014334917068481445, "grad_norm": 5.318289971351623, "loss": 0.42677096128463743, "time": 0.764702558517456, "epoch": 1, "memory": 28798, "step": 1000} +{"lr": 0.0001, "data_time": 0.001995277404785156, "grad_norm": 7.122946190834045, "loss": 0.431024095416069, "time": 0.7643571853637695, "epoch": 1, "memory": 28798, "step": 1100} +{"lr": 0.0001, "data_time": 0.0016049385070800782, "grad_norm": 8.187716698646545, "loss": 0.4295435696840286, "time": 0.7858440637588501, "epoch": 1, "memory": 28798, "step": 1200} +{"lr": 0.0001, "data_time": 0.0017634391784667968, "grad_norm": 6.058029818534851, "loss": 0.40536400079727175, "time": 0.7782463550567627, "epoch": 1, "memory": 28798, "step": 1300} +{"lr": 0.0001, "data_time": 0.0020489931106567384, "grad_norm": 7.24406931400299, "loss": 0.42019582688808443, "time": 0.8470067262649537, "epoch": 1, "memory": 28798, "step": 1400} +{"lr": 0.0001, "data_time": 0.0015645980834960937, "grad_norm": 9.415212643146514, "loss": 0.41330052018165586, "time": 0.749920916557312, "epoch": 1, "memory": 28798, "step": 1500} +{"lr": 0.0001, "data_time": 0.0018555402755737304, "grad_norm": 10.407112336158752, "loss": 0.41097092926502227, "time": 0.8086743831634522, "epoch": 1, "memory": 28798, "step": 1600} +{"lr": 0.0001, "data_time": 0.002411293983459473, "grad_norm": 8.974961280822754, "loss": 0.39800150096416476, "time": 0.8088978290557861, "epoch": 1, "memory": 28798, "step": 1700} +{"lr": 0.0001, "data_time": 0.002052760124206543, "grad_norm": 8.608369994163514, "loss": 0.38782241344451907, "time": 0.7794685125350952, "epoch": 1, "memory": 28798, "step": 1800} +{"lr": 0.0001, "data_time": 0.002239561080932617, "grad_norm": 7.995923852920532, "loss": 0.3900362730026245, "time": 0.7406694650650024, "epoch": 1, "memory": 28798, "step": 1900} +{"lr": 0.0001, "data_time": 0.0019824028015136717, "grad_norm": 12.193249082565307, "loss": 0.4058050960302353, "time": 0.7820487976074219, "epoch": 1, "memory": 28798, "step": 2000} +{"lr": 0.0001, "data_time": 0.0016951560974121094, "grad_norm": 8.75432903766632, "loss": 0.37981541752815245, "time": 0.7462271213531494, "epoch": 1, "memory": 28798, "step": 2100} +{"lr": 0.0001, "data_time": 0.0015626192092895509, "grad_norm": 12.1649915933609, "loss": 0.381779682636261, "time": 0.7927675485610962, "epoch": 1, "memory": 28798, "step": 2200} +{"lr": 0.0001, "data_time": 0.0016306161880493164, "grad_norm": 7.206118392944336, "loss": 0.3836579114198685, "time": 0.8494669437408447, "epoch": 1, "memory": 28798, "step": 2300} +{"lr": 0.0001, "data_time": 0.0014415025711059571, "grad_norm": 8.76898214817047, "loss": 0.36932373046875, "time": 0.7785675287246704, "epoch": 1, "memory": 28798, "step": 2400} +{"lr": 0.0001, "data_time": 0.0015406608581542969, "grad_norm": 9.264030408859252, "loss": 0.361787348985672, "time": 0.8009753942489624, "epoch": 1, "memory": 28798, "step": 2500} +{"lr": 0.0001, "data_time": 0.0015421152114868165, "grad_norm": 9.543509221076965, "loss": 0.4084623873233795, "time": 0.7717762231826782, "epoch": 1, "memory": 28798, "step": 2600} +{"lr": 0.0001, "data_time": 0.0016265392303466796, "grad_norm": 7.552957081794739, "loss": 0.3529518723487854, "time": 0.7745697021484375, "epoch": 1, "memory": 28798, "step": 2700} +{"lr": 0.0001, "data_time": 0.20046417713165282, "grad_norm": 8.773928093910218, "loss": 0.3352687120437622, "time": 0.8354354619979858, "epoch": 1, "memory": 28798, "step": 2800} +{"lr": 0.0001, "data_time": 0.0019224166870117187, "grad_norm": 10.194337511062622, "loss": 0.33998887836933134, "time": 0.7989093542099, "epoch": 1, "memory": 28798, "step": 2900} +{"lr": 0.0001, "data_time": 0.0017558574676513673, "grad_norm": 10.736096847057343, "loss": 0.35384186506271365, "time": 0.7650700807571411, "epoch": 1, "memory": 28798, "step": 3000} +{"lr": 0.0001, "data_time": 0.0019510269165039062, "grad_norm": 8.427406597137452, "loss": 0.3420874625444412, "time": 0.75934317111969, "epoch": 1, "memory": 28798, "step": 3100} +{"lr": 0.0001, "data_time": 0.0016048192977905274, "grad_norm": 9.468148756027222, "loss": 0.3274377554655075, "time": 0.8447348833084106, "epoch": 1, "memory": 28798, "step": 3200} +{"lr": 0.0001, "data_time": 0.0016980409622192384, "grad_norm": 11.842045307159424, "loss": 0.33908976018428805, "time": 0.8461315393447876, "epoch": 1, "memory": 28798, "step": 3300} +{"lr": 0.0001, "data_time": 0.0019853591918945314, "grad_norm": 14.057745933532715, "loss": 0.34925416111946106, "time": 0.7893282413482666, "epoch": 1, "memory": 28798, "step": 3400} +{"lr": 0.0001, "data_time": 0.0016464471817016601, "grad_norm": 11.661341524124145, "loss": 0.34215838015079497, "time": 0.8101903438568115, "epoch": 1, "memory": 28798, "step": 3500} +{"lr": 0.0001, "data_time": 0.0015691518783569336, "grad_norm": 12.739499568939209, "loss": 0.35349526107311247, "time": 0.7678930521011352, "epoch": 1, "memory": 28798, "step": 3600} +{"lr": 0.0001, "data_time": 0.0018642902374267577, "grad_norm": 14.655861473083496, "loss": 0.3317358762025833, "time": 0.7918614625930787, "epoch": 1, "memory": 28798, "step": 3700} +{"lr": 0.0001, "data_time": 0.001734614372253418, "grad_norm": 8.019277739524842, "loss": 0.3204199582338333, "time": 0.7980497121810913, "epoch": 1, "memory": 28798, "step": 3800} +{"lr": 0.0001, "data_time": 0.0017982244491577149, "grad_norm": 16.59911365509033, "loss": 0.3388776183128357, "time": 0.7871023178100586, "epoch": 1, "memory": 28798, "step": 3900} +{"lr": 0.0001, "data_time": 0.001947331428527832, "grad_norm": 11.131891167163849, "loss": 0.31053749918937684, "time": 0.9097763538360596, "epoch": 1, "memory": 28798, "step": 4000} +{"lr": 0.0001, "data_time": 0.0018900632858276367, "grad_norm": 7.135937070846557, "loss": 0.32608037292957304, "time": 0.8588865041732788, "epoch": 1, "memory": 28798, "step": 4100} +{"lr": 0.0001, "data_time": 0.002013373374938965, "grad_norm": 11.285934841632843, "loss": 0.33473441302776336, "time": 0.7751770257949829, "epoch": 1, "memory": 28798, "step": 4200} +{"lr": 0.0001, "data_time": 0.0017449378967285157, "grad_norm": 12.645131719112396, "loss": 0.3032625824213028, "time": 0.7565198183059693, "epoch": 1, "memory": 28798, "step": 4300} +{"lr": 0.0001, "data_time": 0.001753830909729004, "grad_norm": 14.853507995605469, "loss": 0.33213984668254853, "time": 0.8101833820343017, "epoch": 1, "memory": 28798, "step": 4400} +{"lr": 0.0001, "data_time": 0.0016500234603881836, "grad_norm": 13.149504327774048, "loss": 0.3288163751363754, "time": 0.8231769800186157, "epoch": 1, "memory": 28798, "step": 4500} +{"lr": 0.0001, "data_time": 0.001666688919067383, "grad_norm": 7.7415166854858395, "loss": 0.2992019057273865, "time": 0.7856241941452027, "epoch": 1, "memory": 28798, "step": 4600} +{"lr": 0.0001, "data_time": 0.001813220977783203, "grad_norm": 7.6865216255187985, "loss": 0.2817671522498131, "time": 0.7724915027618409, "epoch": 1, "memory": 28798, "step": 4700} +{"lr": 0.0001, "data_time": 0.0014267206192016602, "grad_norm": 7.778097343444824, "loss": 0.2861554190516472, "time": 0.7990787744522094, "epoch": 1, "memory": 28798, "step": 4800} +{"lr": 0.0001, "data_time": 0.0016474246978759766, "grad_norm": 7.936123585700988, "loss": 0.2836196631193161, "time": 0.8176966190338135, "epoch": 1, "memory": 28798, "step": 4900} +{"lr": 0.0001, "data_time": 0.0016502857208251954, "grad_norm": 8.367023301124572, "loss": 0.2896931916475296, "time": 0.7379039764404297, "epoch": 1, "memory": 28798, "step": 5000} +{"lr": 0.0001, "data_time": 0.0015475749969482422, "grad_norm": 7.881516706943512, "loss": 0.27526378631591797, "time": 0.7637993812561035, "epoch": 1, "memory": 28798, "step": 5100} +{"lr": 0.0001, "data_time": 0.0016379356384277344, "grad_norm": 5.626676559448242, "loss": 0.2834407016634941, "time": 0.793156623840332, "epoch": 1, "memory": 28798, "step": 5200} +{"lr": 0.0001, "data_time": 0.0014394044876098633, "grad_norm": 17.186562919616698, "loss": 0.3125844717025757, "time": 0.7624467611312866, "epoch": 1, "memory": 28798, "step": 5300} +{"lr": 0.0001, "data_time": 0.0015750169754028321, "grad_norm": 7.602498078346253, "loss": 0.2827661603689194, "time": 0.7655160188674927, "epoch": 1, "memory": 28798, "step": 5400} +{"lr": 0.0001, "data_time": 0.0016482353210449218, "grad_norm": 14.832658922672271, "loss": 0.29803421497344973, "time": 0.7661600589752198, "epoch": 1, "memory": 28798, "step": 5500} +{"lr": 0.0001, "data_time": 0.0015974044799804688, "grad_norm": 9.575963139533997, "loss": 0.28630885630846026, "time": 0.7695990562438965, "epoch": 1, "memory": 28798, "step": 5600} +{"lr": 0.0001, "data_time": 0.0017279386520385742, "grad_norm": 14.966920685768127, "loss": 0.3144871652126312, "time": 0.7377325057983398, "epoch": 1, "memory": 28798, "step": 5700} +{"accuracy/top1": 98.63644409179688, "data_time": 0.2643384635448456, "time": 0.8042525202035904, "step": 1} +{"lr": 9.779754323328192e-05, "data_time": 0.2253279685974121, "grad_norm": 4.724720764160156, "loss": 0.275802007317543, "time": 0.7853192567825318, "epoch": 2, "memory": 28798, "step": 5858} +{"lr": 9.779754323328192e-05, "data_time": 0.13539741039276124, "grad_norm": 8.829039835929871, "loss": 0.281463822722435, "time": 0.7541395902633667, "epoch": 2, "memory": 28783, "step": 5958} +{"lr": 9.779754323328192e-05, "data_time": 0.24535365104675294, "grad_norm": 4.85420960187912, "loss": 0.2631674766540527, "time": 0.7702489137649536, "epoch": 2, "memory": 28783, "step": 6058} +{"lr": 9.779754323328192e-05, "data_time": 0.16128273010253907, "grad_norm": 6.4906369090080265, "loss": 0.26960329711437225, "time": 0.8204433679580688, "epoch": 2, "memory": 28783, "step": 6158} +{"lr": 9.779754323328192e-05, "data_time": 0.0019009828567504883, "grad_norm": 7.4814086079597475, "loss": 0.26002991646528245, "time": 0.7683497428894043, "epoch": 2, "memory": 28783, "step": 6258} +{"lr": 9.779754323328192e-05, "data_time": 0.0017068624496459962, "grad_norm": 3.6976542353630064, "loss": 0.2582589194178581, "time": 0.7667133808135986, "epoch": 2, "memory": 28783, "step": 6358} +{"lr": 9.779754323328192e-05, "data_time": 0.001719522476196289, "grad_norm": 5.707702016830444, "loss": 0.2515236109495163, "time": 0.8664642095565795, "epoch": 2, "memory": 28783, "step": 6458} +{"lr": 9.779754323328192e-05, "data_time": 0.0014872312545776366, "grad_norm": 4.434868597984314, "loss": 0.25663683712482455, "time": 0.7268047094345093, "epoch": 2, "memory": 28783, "step": 6558} +{"lr": 9.779754323328192e-05, "data_time": 0.0016342878341674804, "grad_norm": 8.42042521238327, "loss": 0.24680227935314178, "time": 0.7821118116378785, "epoch": 2, "memory": 28783, "step": 6658} +{"lr": 9.779754323328192e-05, "data_time": 0.0018155574798583984, "grad_norm": 5.102838933467865, "loss": 0.2469940662384033, "time": 0.8128172159194946, "epoch": 2, "memory": 28783, "step": 6758} +{"lr": 9.779754323328192e-05, "data_time": 0.0015127420425415038, "grad_norm": 7.1840561628341675, "loss": 0.2738844335079193, "time": 0.7818080663681031, "epoch": 2, "memory": 28783, "step": 6858} +{"lr": 9.779754323328192e-05, "data_time": 0.001641225814819336, "grad_norm": 9.520950174331665, "loss": 0.2546579748392105, "time": 0.7373530149459839, "epoch": 2, "memory": 28783, "step": 6958} +{"lr": 9.779754323328192e-05, "data_time": 0.001997566223144531, "grad_norm": 7.790714287757874, "loss": 0.2447388768196106, "time": 0.7828716993331909, "epoch": 2, "memory": 28783, "step": 7058} +{"lr": 9.779754323328192e-05, "data_time": 0.0018570899963378906, "grad_norm": 6.547972404956818, "loss": 0.24792872965335847, "time": 0.8393717527389526, "epoch": 2, "memory": 28783, "step": 7158} +{"lr": 9.779754323328192e-05, "data_time": 0.0016109704971313476, "grad_norm": 4.401484966278076, "loss": 0.2534740626811981, "time": 0.7773492574691773, "epoch": 2, "memory": 28783, "step": 7258} +{"lr": 9.779754323328192e-05, "data_time": 0.0023687124252319337, "grad_norm": 3.6738228678703306, "loss": 0.2359653428196907, "time": 0.916810417175293, "epoch": 2, "memory": 28783, "step": 7358} +{"lr": 9.779754323328192e-05, "data_time": 0.0017107009887695312, "grad_norm": 11.95268611907959, "loss": 0.24162269234657288, "time": 0.7193113327026367, "epoch": 2, "memory": 28783, "step": 7458} +{"lr": 9.779754323328192e-05, "data_time": 0.001832890510559082, "grad_norm": 1.9830315947532653, "loss": 0.24130344837903978, "time": 0.7895451784133911, "epoch": 2, "memory": 28783, "step": 7558} +{"lr": 9.779754323328192e-05, "data_time": 0.0016902923583984376, "grad_norm": 5.427372705936432, "loss": 0.2390815481543541, "time": 0.7553702116012573, "epoch": 2, "memory": 28783, "step": 7658} +{"lr": 9.779754323328192e-05, "data_time": 0.0018090486526489257, "grad_norm": 4.1846026420593265, "loss": 0.23522198796272278, "time": 0.7588916540145874, "epoch": 2, "memory": 28783, "step": 7758} +{"lr": 9.779754323328192e-05, "data_time": 0.0022718429565429686, "grad_norm": 4.5128955960273744, "loss": 0.2320399358868599, "time": 0.7661934852600097, "epoch": 2, "memory": 28783, "step": 7858} +{"lr": 9.779754323328192e-05, "data_time": 0.0017469406127929687, "grad_norm": 5.278597688674926, "loss": 0.2440346896648407, "time": 0.7454902172088623, "epoch": 2, "memory": 28783, "step": 7958} +{"lr": 9.779754323328192e-05, "data_time": 0.0017522573471069336, "grad_norm": 4.699299311637878, "loss": 0.2368510529398918, "time": 0.7240928888320923, "epoch": 2, "memory": 28783, "step": 8058} +{"lr": 9.779754323328192e-05, "data_time": 0.0015929460525512696, "grad_norm": 5.631152653694153, "loss": 0.24275548309087752, "time": 0.7127774000167847, "epoch": 2, "memory": 28783, "step": 8158} +{"lr": 9.779754323328192e-05, "data_time": 0.0022130489349365236, "grad_norm": 3.8032525062561033, "loss": 0.22538014203310014, "time": 0.7257648468017578, "epoch": 2, "memory": 28783, "step": 8258} +{"lr": 9.779754323328192e-05, "data_time": 0.0017465829849243164, "grad_norm": 5.955878686904907, "loss": 0.2376284494996071, "time": 0.7556620836257935, "epoch": 2, "memory": 28783, "step": 8358} +{"lr": 9.779754323328192e-05, "data_time": 0.0017087459564208984, "grad_norm": 4.831857204437256, "loss": 0.23309049010276794, "time": 0.7472589492797852, "epoch": 2, "memory": 28783, "step": 8458} +{"lr": 9.779754323328192e-05, "data_time": 0.0015890836715698243, "grad_norm": 3.156527227163315, "loss": 0.22835725992918016, "time": 0.8053215265274047, "epoch": 2, "memory": 28783, "step": 8558} +{"lr": 9.779754323328192e-05, "data_time": 0.001584339141845703, "grad_norm": 1.9715153217315673, "loss": 0.22326327562332154, "time": 0.7155921936035157, "epoch": 2, "memory": 28783, "step": 8658} +{"lr": 9.779754323328192e-05, "data_time": 0.0017405271530151368, "grad_norm": 3.204046630859375, "loss": 0.22883425951004027, "time": 0.6969146490097046, "epoch": 2, "memory": 28783, "step": 8758} +{"lr": 9.779754323328192e-05, "data_time": 0.0016877174377441407, "grad_norm": 1.5660253405570983, "loss": 0.21822546571493148, "time": 0.7188021898269653, "epoch": 2, "memory": 28783, "step": 8858} +{"lr": 9.779754323328192e-05, "data_time": 0.0022782325744628907, "grad_norm": 2.75571203827858, "loss": 0.22351438105106353, "time": 0.8173815011978149, "epoch": 2, "memory": 28783, "step": 8958} +{"lr": 9.779754323328192e-05, "data_time": 0.001535797119140625, "grad_norm": 2.5499474883079527, "loss": 0.22208157926797867, "time": 0.7032848834991455, "epoch": 2, "memory": 28783, "step": 9058} +{"lr": 9.779754323328192e-05, "data_time": 0.0016241788864135743, "grad_norm": 3.831286382675171, "loss": 0.22439639270305634, "time": 0.7431313276290894, "epoch": 2, "memory": 28783, "step": 9158} +{"lr": 9.779754323328192e-05, "data_time": 0.002005267143249512, "grad_norm": 5.158092880249024, "loss": 0.2316832423210144, "time": 0.7462419509887696, "epoch": 2, "memory": 28783, "step": 9258} +{"lr": 9.779754323328192e-05, "data_time": 0.0016271352767944336, "grad_norm": 3.846285009384155, "loss": 0.22138564735651017, "time": 0.7830148458480835, "epoch": 2, "memory": 28783, "step": 9358} +{"lr": 9.779754323328192e-05, "data_time": 0.001618647575378418, "grad_norm": 3.1761711955070497, "loss": 0.22847191393375396, "time": 0.7664671659469604, "epoch": 2, "memory": 28783, "step": 9458} +{"lr": 9.779754323328192e-05, "data_time": 0.001716923713684082, "grad_norm": 7.066403663158416, "loss": 0.21818615198135377, "time": 0.7468266248703003, "epoch": 2, "memory": 28783, "step": 9558} +{"lr": 9.779754323328192e-05, "data_time": 0.0015748977661132813, "grad_norm": 2.3538535118103026, "loss": 0.22665224373340606, "time": 0.7534285545349121, "epoch": 2, "memory": 28783, "step": 9658} +{"lr": 9.779754323328192e-05, "data_time": 0.001693415641784668, "grad_norm": 2.245039939880371, "loss": 0.22155413776636124, "time": 0.7146346807479859, "epoch": 2, "memory": 28783, "step": 9758} +{"lr": 9.779754323328192e-05, "data_time": 0.0020279645919799804, "grad_norm": 3.1565908253192902, "loss": 0.22076133638620377, "time": 0.7443537473678589, "epoch": 2, "memory": 28783, "step": 9858} +{"lr": 9.779754323328192e-05, "data_time": 0.001599287986755371, "grad_norm": 4.543850445747376, "loss": 0.22001899033784866, "time": 0.7957648038864136, "epoch": 2, "memory": 28783, "step": 9958} +{"lr": 9.779754323328192e-05, "data_time": 0.0015918254852294923, "grad_norm": 2.7435157179832457, "loss": 0.22163452655076982, "time": 0.8049781560897827, "epoch": 2, "memory": 28783, "step": 10058} +{"lr": 9.779754323328192e-05, "data_time": 0.0025667905807495116, "grad_norm": 3.7365896940231322, "loss": 0.22913289219141006, "time": 0.7350122690200805, "epoch": 2, "memory": 28783, "step": 10158} +{"lr": 9.779754323328192e-05, "data_time": 0.0014409542083740235, "grad_norm": 4.982752060890197, "loss": 0.22232090830802917, "time": 0.7107707738876343, "epoch": 2, "memory": 28783, "step": 10258} +{"lr": 9.779754323328192e-05, "data_time": 0.0020370721817016602, "grad_norm": 1.8967858254909515, "loss": 0.2177594393491745, "time": 0.7625745058059692, "epoch": 2, "memory": 28783, "step": 10358} +{"lr": 9.779754323328192e-05, "data_time": 0.0019598484039306642, "grad_norm": 1.9121993958950043, "loss": 0.20995037108659745, "time": 0.7142647266387939, "epoch": 2, "memory": 28783, "step": 10458} +{"lr": 9.779754323328192e-05, "data_time": 0.0018325328826904296, "grad_norm": 2.5711088299751284, "loss": 0.21899385899305343, "time": 0.8342780828475952, "epoch": 2, "memory": 28783, "step": 10558} +{"lr": 9.779754323328192e-05, "data_time": 1.8292505979537963, "grad_norm": 2.879544985294342, "loss": 0.22317344844341278, "time": 2.6419941425323485, "epoch": 2, "memory": 28783, "step": 10658} +{"lr": 9.779754323328192e-05, "data_time": 0.2317431926727295, "grad_norm": 1.296843558549881, "loss": 0.21406361907720567, "time": 0.7352376699447631, "epoch": 2, "memory": 28783, "step": 10758} +{"lr": 9.779754323328192e-05, "data_time": 0.2768770456314087, "grad_norm": 3.4218323826789856, "loss": 0.22064824104309083, "time": 0.7944846868515014, "epoch": 2, "memory": 28783, "step": 10858} +{"lr": 9.779754323328192e-05, "data_time": 0.1599552869796753, "grad_norm": 1.7299805223941802, "loss": 0.21627172380685805, "time": 0.6702358722686768, "epoch": 2, "memory": 28783, "step": 10958} +{"lr": 9.779754323328192e-05, "data_time": 0.18956103324890136, "grad_norm": 1.409479123353958, "loss": 0.22424711883068085, "time": 0.691569185256958, "epoch": 2, "memory": 28783, "step": 11058} +{"lr": 9.779754323328192e-05, "data_time": 0.10107715129852295, "grad_norm": 2.3308180451393126, "loss": 0.2175878569483757, "time": 0.6839389324188232, "epoch": 2, "memory": 28783, "step": 11158} +{"lr": 9.779754323328192e-05, "data_time": 0.0015223026275634766, "grad_norm": 1.9041896104812621, "loss": 0.21678828150033952, "time": 0.7197707891464233, "epoch": 2, "memory": 28783, "step": 11258} +{"lr": 9.779754323328192e-05, "data_time": 0.0017743349075317384, "grad_norm": 2.214759635925293, "loss": 0.21667275279760362, "time": 0.7230534315109253, "epoch": 2, "memory": 28783, "step": 11358} +{"lr": 9.779754323328192e-05, "data_time": 0.0017892122268676758, "grad_norm": 1.5354084491729736, "loss": 0.22148067355155945, "time": 0.7298334836959839, "epoch": 2, "memory": 28783, "step": 11458} +{"accuracy/top1": 99.57704162597656, "data_time": 0.2029575179604923, "time": 0.6182985586278579, "step": 2} +{"lr": 9.140576474687264e-05, "data_time": 0.0017037391662597656, "grad_norm": 2.547348564863205, "loss": 0.22042838335037232, "time": 0.7363026142120361, "epoch": 3, "memory": 28783, "step": 11616} +{"lr": 9.140576474687264e-05, "data_time": 0.0016712188720703126, "grad_norm": 1.9135839641094208, "loss": 0.21216972321271896, "time": 0.7759701251983643, "epoch": 3, "memory": 28783, "step": 11716} +{"lr": 9.140576474687264e-05, "data_time": 0.002136373519897461, "grad_norm": 1.8601315200328827, "loss": 0.21522459536790847, "time": 0.7624718904495239, "epoch": 3, "memory": 28783, "step": 11816} +{"lr": 9.140576474687264e-05, "data_time": 0.0017839670181274414, "grad_norm": 1.554959350824356, "loss": 0.21358174085617065, "time": 0.7393719673156738, "epoch": 3, "memory": 28783, "step": 11916} +{"lr": 9.140576474687264e-05, "data_time": 0.001764988899230957, "grad_norm": 2.924062669277191, "loss": 0.21263508796691893, "time": 0.7282135248184204, "epoch": 3, "memory": 28783, "step": 12016} +{"lr": 9.140576474687264e-05, "data_time": 0.0017034053802490235, "grad_norm": 2.0709295213222503, "loss": 0.21246255189180374, "time": 0.8083432435989379, "epoch": 3, "memory": 28783, "step": 12116} +{"lr": 9.140576474687264e-05, "data_time": 0.0017165660858154297, "grad_norm": 2.144436377286911, "loss": 0.21445005834102632, "time": 0.7608980178833008, "epoch": 3, "memory": 28783, "step": 12216} +{"lr": 9.140576474687264e-05, "data_time": 0.0017304658889770509, "grad_norm": 1.3800033688545228, "loss": 0.214017653465271, "time": 0.8114840269088746, "epoch": 3, "memory": 28783, "step": 12316} +{"lr": 9.140576474687264e-05, "data_time": 0.0015224456787109376, "grad_norm": 1.5817308306694031, "loss": 0.20982833504676818, "time": 0.7176219463348389, "epoch": 3, "memory": 28783, "step": 12416} +{"lr": 9.140576474687264e-05, "data_time": 0.0017634391784667968, "grad_norm": 1.1910838425159453, "loss": 0.2121584177017212, "time": 0.89136221408844, "epoch": 3, "memory": 28783, "step": 12516} +{"lr": 9.140576474687264e-05, "data_time": 0.0018435239791870118, "grad_norm": 1.829367995262146, "loss": 0.21225885003805162, "time": 0.7900148868560791, "epoch": 3, "memory": 28783, "step": 12616} +{"lr": 9.140576474687264e-05, "data_time": 0.10203487873077392, "grad_norm": 1.2540997505187987, "loss": 0.2087197095155716, "time": 0.8613012075424195, "epoch": 3, "memory": 28783, "step": 12716} +{"lr": 9.140576474687264e-05, "data_time": 0.0017075538635253906, "grad_norm": 1.3484358072280884, "loss": 0.2111063003540039, "time": 0.8129325628280639, "epoch": 3, "memory": 28783, "step": 12816} +{"lr": 9.140576474687264e-05, "data_time": 0.0018035650253295898, "grad_norm": 1.1028524696826936, "loss": 0.21101672053337098, "time": 0.8085900545120239, "epoch": 3, "memory": 28783, "step": 12916} +{"lr": 9.140576474687264e-05, "data_time": 0.001885843276977539, "grad_norm": 1.2602094292640686, "loss": 0.21546115577220917, "time": 0.79597008228302, "epoch": 3, "memory": 28783, "step": 13016} +{"lr": 9.140576474687264e-05, "data_time": 0.0019144773483276366, "grad_norm": 1.3525001168251038, "loss": 0.20870959013700485, "time": 0.7559643983840942, "epoch": 3, "memory": 28783, "step": 13116} +{"lr": 9.140576474687264e-05, "data_time": 0.001648259162902832, "grad_norm": 1.2084201335906983, "loss": 0.21275661140680313, "time": 0.7298112869262695, "epoch": 3, "memory": 28783, "step": 13216} +{"lr": 9.140576474687264e-05, "data_time": 0.0015499591827392578, "grad_norm": 1.3942279934883117, "loss": 0.21074751168489456, "time": 0.8349781990051269, "epoch": 3, "memory": 28783, "step": 13316} +{"lr": 9.140576474687264e-05, "data_time": 0.002311038970947266, "grad_norm": 2.2864261150360106, "loss": 0.21232101023197175, "time": 0.7560360908508301, "epoch": 3, "memory": 28783, "step": 13416} +{"lr": 9.140576474687264e-05, "data_time": 0.0017961025238037109, "grad_norm": 1.2045234262943267, "loss": 0.20949556529521943, "time": 0.7641310214996337, "epoch": 3, "memory": 28783, "step": 13516} +{"lr": 9.140576474687264e-05, "data_time": 0.0019094705581665038, "grad_norm": 1.4417009353637695, "loss": 0.2153782069683075, "time": 0.7750011682510376, "epoch": 3, "memory": 28783, "step": 13616} +{"lr": 9.140576474687264e-05, "data_time": 0.09557888507843018, "grad_norm": 1.544138091802597, "loss": 0.21417417228221894, "time": 0.7901232004165649, "epoch": 3, "memory": 28783, "step": 13716} +{"lr": 9.140576474687264e-05, "data_time": 0.06048867702484131, "grad_norm": 2.211031836271286, "loss": 0.2095340371131897, "time": 0.7652333498001098, "epoch": 3, "memory": 28783, "step": 13816} +{"lr": 9.140576474687264e-05, "data_time": 0.0013819217681884765, "grad_norm": 1.0305872529745101, "loss": 0.20918484181165695, "time": 0.8278063058853149, "epoch": 3, "memory": 28783, "step": 13916} +{"lr": 9.140576474687264e-05, "data_time": 0.10765769481658935, "grad_norm": 2.019196939468384, "loss": 0.22233359515666962, "time": 0.7532198429107666, "epoch": 3, "memory": 28783, "step": 14016} +{"lr": 9.140576474687264e-05, "data_time": 0.27614758014678953, "grad_norm": 0.9989032059907913, "loss": 0.2159930184483528, "time": 0.7803572416305542, "epoch": 3, "memory": 28783, "step": 14116} +{"lr": 9.140576474687264e-05, "data_time": 0.21318349838256836, "grad_norm": 1.3191742479801178, "loss": 0.20799032896757125, "time": 0.7172081232070923, "epoch": 3, "memory": 28783, "step": 14216} +{"lr": 9.140576474687264e-05, "data_time": 0.260835862159729, "grad_norm": 1.3699993193149567, "loss": 0.21112867444753647, "time": 0.7647407293319702, "epoch": 3, "memory": 28783, "step": 14316} +{"lr": 9.140576474687264e-05, "data_time": 0.1925347089767456, "grad_norm": 1.387695175409317, "loss": 0.21318716406822205, "time": 0.7368336915969849, "epoch": 3, "memory": 28783, "step": 14416} +{"lr": 9.140576474687264e-05, "data_time": 0.0015162944793701172, "grad_norm": 1.239967554807663, "loss": 0.2061662420630455, "time": 0.7891481876373291, "epoch": 3, "memory": 28783, "step": 14516} +{"lr": 9.140576474687264e-05, "data_time": 0.0018846750259399413, "grad_norm": 2.0698291540145872, "loss": 0.21210735589265822, "time": 0.7504413366317749, "epoch": 3, "memory": 28783, "step": 14616} +{"lr": 9.140576474687264e-05, "data_time": 0.0018059730529785157, "grad_norm": 2.023476779460907, "loss": 0.2098442181944847, "time": 0.7859753608703614, "epoch": 3, "memory": 28783, "step": 14716} +{"lr": 9.140576474687264e-05, "data_time": 0.0015339612960815429, "grad_norm": 1.0350459814071655, "loss": 0.2101355627179146, "time": 0.7155972480773926, "epoch": 3, "memory": 28783, "step": 14816} +{"lr": 9.140576474687264e-05, "data_time": 0.0015639543533325195, "grad_norm": 1.1589035451412202, "loss": 0.2069369599223137, "time": 0.7649103879928589, "epoch": 3, "memory": 28783, "step": 14916} +{"lr": 9.140576474687264e-05, "data_time": 0.00160672664642334, "grad_norm": 1.2026117026805878, "loss": 0.20926864445209503, "time": 0.7189409732818604, "epoch": 3, "memory": 28783, "step": 15016} +{"lr": 9.140576474687264e-05, "data_time": 0.001593303680419922, "grad_norm": 0.9875022858381272, "loss": 0.20764354169368743, "time": 0.7618232488632202, "epoch": 3, "memory": 28783, "step": 15116} +{"lr": 9.140576474687264e-05, "data_time": 0.0016371250152587891, "grad_norm": 1.5114789605140686, "loss": 0.20797639191150666, "time": 0.8211654901504517, "epoch": 3, "memory": 28783, "step": 15216} +{"lr": 9.140576474687264e-05, "data_time": 0.0016020774841308595, "grad_norm": 1.2105599641799927, "loss": 0.21090261340141297, "time": 0.7385528564453125, "epoch": 3, "memory": 28783, "step": 15316} +{"lr": 9.140576474687264e-05, "data_time": 0.0018048286437988281, "grad_norm": 0.9896108508110046, "loss": 0.210080286860466, "time": 0.7755727052688599, "epoch": 3, "memory": 28783, "step": 15416} +{"lr": 9.140576474687264e-05, "data_time": 0.048856568336486814, "grad_norm": 2.026436024904251, "loss": 0.21362337619066238, "time": 0.7411954164505005, "epoch": 3, "memory": 28783, "step": 15516} +{"lr": 9.140576474687264e-05, "data_time": 0.0016203880310058593, "grad_norm": 1.3295194804668427, "loss": 0.21074522137641907, "time": 0.7665291547775268, "epoch": 3, "memory": 28783, "step": 15616} +{"lr": 9.140576474687264e-05, "data_time": 0.0014732837677001952, "grad_norm": 1.3675204813480377, "loss": 0.20556535720825195, "time": 0.8190885543823242, "epoch": 3, "memory": 28783, "step": 15716} +{"lr": 9.140576474687264e-05, "data_time": 0.0018591165542602539, "grad_norm": 1.2479036509990693, "loss": 0.20555228590965272, "time": 0.8033952474594116, "epoch": 3, "memory": 28783, "step": 15816} +{"lr": 9.140576474687264e-05, "data_time": 0.0015214920043945313, "grad_norm": 1.1333971738815307, "loss": 0.2112056851387024, "time": 0.7313814163208008, "epoch": 3, "memory": 28783, "step": 15916} +{"lr": 9.140576474687264e-05, "data_time": 0.001503896713256836, "grad_norm": 1.0096331983804703, "loss": 0.2103559747338295, "time": 0.7655946254730225, "epoch": 3, "memory": 28783, "step": 16016} +{"lr": 9.140576474687264e-05, "data_time": 0.001732039451599121, "grad_norm": 0.7067256242036819, "loss": 0.20596436262130738, "time": 0.7798765659332275, "epoch": 3, "memory": 28783, "step": 16116} +{"lr": 9.140576474687264e-05, "data_time": 0.0014956235885620118, "grad_norm": 0.7893205374479294, "loss": 0.20947739481925964, "time": 0.8068555355072021, "epoch": 3, "memory": 28783, "step": 16216} +{"lr": 9.140576474687264e-05, "data_time": 0.0016243696212768556, "grad_norm": 1.3125432908535004, "loss": 0.20244491696357728, "time": 0.9077004432678223, "epoch": 3, "memory": 28783, "step": 16316} +{"lr": 9.140576474687264e-05, "data_time": 0.0015056133270263672, "grad_norm": 1.268625909090042, "loss": 0.2104177340865135, "time": 0.7249560117721557, "epoch": 3, "memory": 28783, "step": 16416} +{"lr": 9.140576474687264e-05, "data_time": 0.0017547607421875, "grad_norm": 0.6721316695213317, "loss": 0.20510686337947845, "time": 0.7545232772827148, "epoch": 3, "memory": 28783, "step": 16516} +{"lr": 9.140576474687264e-05, "data_time": 0.0018735408782958984, "grad_norm": 1.4055960297584533, "loss": 0.21015082597732543, "time": 0.765139126777649, "epoch": 3, "memory": 28783, "step": 16616} +{"lr": 9.140576474687264e-05, "data_time": 0.001816701889038086, "grad_norm": 1.3470157921314239, "loss": 0.20931606888771057, "time": 0.6972701787948609, "epoch": 3, "memory": 28783, "step": 16716} +{"lr": 9.140576474687264e-05, "data_time": 0.0018264293670654298, "grad_norm": 1.309765875339508, "loss": 0.2060650423169136, "time": 0.7734025239944458, "epoch": 3, "memory": 28783, "step": 16816} +{"lr": 9.140576474687264e-05, "data_time": 0.0017119884490966798, "grad_norm": 0.8919267356395721, "loss": 0.20747613459825515, "time": 0.8131618738174439, "epoch": 3, "memory": 28783, "step": 16916} +{"lr": 9.140576474687264e-05, "data_time": 0.0015355348587036133, "grad_norm": 0.9887870609760284, "loss": 0.2054794982075691, "time": 0.7259827375411987, "epoch": 3, "memory": 28783, "step": 17016} +{"lr": 9.140576474687264e-05, "data_time": 0.0016350269317626954, "grad_norm": 0.9397801756858826, "loss": 0.2074136883020401, "time": 0.8159034252166748, "epoch": 3, "memory": 28783, "step": 17116} +{"lr": 9.140576474687264e-05, "data_time": 0.0016236305236816406, "grad_norm": 0.9886764645576477, "loss": 0.20989191085100173, "time": 0.7101340532302857, "epoch": 3, "memory": 28783, "step": 17216} +{"accuracy/top1": 99.425537109375, "data_time": 0.20295374533709357, "time": 0.6162600657519173, "step": 3} +{"lr": 8.14503363531613e-05, "data_time": 0.07129447460174561, "grad_norm": 0.7139622837305069, "loss": 0.20446813106536865, "time": 0.7448784112930298, "epoch": 4, "memory": 28783, "step": 17374} +{"lr": 8.14503363531613e-05, "data_time": 0.0016505718231201172, "grad_norm": 0.8423473298549652, "loss": 0.2116495981812477, "time": 0.7120396614074707, "epoch": 4, "memory": 28783, "step": 17474} +{"lr": 8.14503363531613e-05, "data_time": 0.001586008071899414, "grad_norm": 0.9319919884204865, "loss": 0.20657327473163606, "time": 0.7979679584503174, "epoch": 4, "memory": 28783, "step": 17574} +{"lr": 8.14503363531613e-05, "data_time": 0.001841568946838379, "grad_norm": 0.9374666512012482, "loss": 0.20756369084119797, "time": 0.7048688173294068, "epoch": 4, "memory": 28783, "step": 17674} +{"lr": 8.14503363531613e-05, "data_time": 0.0018159866333007813, "grad_norm": 1.1117026597261428, "loss": 0.20703705102205278, "time": 0.8853085994720459, "epoch": 4, "memory": 28783, "step": 17774} +{"lr": 8.14503363531613e-05, "data_time": 0.0017165660858154297, "grad_norm": 1.121205472946167, "loss": 0.20755206942558288, "time": 0.7465423107147217, "epoch": 4, "memory": 28783, "step": 17874} +{"lr": 8.14503363531613e-05, "data_time": 0.0016196966171264648, "grad_norm": 0.8346521109342575, "loss": 0.20673878639936447, "time": 0.9494416952133179, "epoch": 4, "memory": 28783, "step": 17974} +{"lr": 8.14503363531613e-05, "data_time": 0.001536107063293457, "grad_norm": 1.3260571867227555, "loss": 0.20651856809854507, "time": 1.0259686946868896, "epoch": 4, "memory": 28783, "step": 18074} +{"lr": 8.14503363531613e-05, "data_time": 0.06866841316223145, "grad_norm": 0.858877283334732, "loss": 0.2090943679213524, "time": 1.0382978439331054, "epoch": 4, "memory": 28783, "step": 18174} +{"lr": 8.14503363531613e-05, "data_time": 0.001637101173400879, "grad_norm": 0.977012836933136, "loss": 0.20690066516399383, "time": 0.7482481479644776, "epoch": 4, "memory": 28783, "step": 18274} +{"lr": 8.14503363531613e-05, "data_time": 0.001680302619934082, "grad_norm": 0.8573934376239777, "loss": 0.20402872264385224, "time": 0.796010947227478, "epoch": 4, "memory": 28783, "step": 18374} +{"lr": 8.14503363531613e-05, "data_time": 0.001658010482788086, "grad_norm": 1.1316018223762512, "loss": 0.20936628729104995, "time": 0.8131815671920777, "epoch": 4, "memory": 28783, "step": 18474} +{"lr": 8.14503363531613e-05, "data_time": 0.0015490531921386718, "grad_norm": 1.2810150742530824, "loss": 0.20540878623723985, "time": 0.854335618019104, "epoch": 4, "memory": 28783, "step": 18574} +{"lr": 8.14503363531613e-05, "data_time": 0.001682138442993164, "grad_norm": 0.9204590931534767, "loss": 0.20437522530555724, "time": 0.8720471858978271, "epoch": 4, "memory": 28783, "step": 18674} +{"lr": 8.14503363531613e-05, "data_time": 0.0015542984008789062, "grad_norm": 1.2260378628969193, "loss": 0.20431203842163087, "time": 1.462713623046875, "epoch": 4, "memory": 28783, "step": 18774} +{"lr": 8.14503363531613e-05, "data_time": 0.0015694379806518554, "grad_norm": 0.802131500840187, "loss": 0.21011782586574554, "time": 1.0631263732910157, "epoch": 4, "memory": 28783, "step": 18874} +{"lr": 8.14503363531613e-05, "data_time": 0.0018329381942749023, "grad_norm": 1.4100565791130066, "loss": 0.20812755972146987, "time": 0.9850741386413574, "epoch": 4, "memory": 28783, "step": 18974} +{"lr": 8.14503363531613e-05, "data_time": 0.0013610601425170898, "grad_norm": 1.8439662277698516, "loss": 0.21002061516046525, "time": 1.1623518466949463, "epoch": 4, "memory": 28783, "step": 19074} +{"lr": 8.14503363531613e-05, "data_time": 0.4064238786697388, "grad_norm": 0.7899254560470581, "loss": 0.2045873299241066, "time": 1.0554860353469848, "epoch": 4, "memory": 28783, "step": 19174} +{"lr": 8.14503363531613e-05, "data_time": 0.6836257696151733, "grad_norm": 0.8103839159011841, "loss": 0.20875571221113204, "time": 1.1865941047668458, "epoch": 4, "memory": 28783, "step": 19274} +{"lr": 8.14503363531613e-05, "data_time": 0.2250920295715332, "grad_norm": 0.7874846875667572, "loss": 0.20594585537910462, "time": 0.7385987520217896, "epoch": 4, "memory": 28783, "step": 19374} +{"lr": 8.14503363531613e-05, "data_time": 0.24105098247528076, "grad_norm": 1.353374284505844, "loss": 0.20811700969934463, "time": 0.7392971515655518, "epoch": 4, "memory": 28783, "step": 19474} +{"lr": 8.14503363531613e-05, "data_time": 0.1747352361679077, "grad_norm": 0.9965912759304046, "loss": 0.20290562957525254, "time": 0.7055529117584228, "epoch": 4, "memory": 28783, "step": 19574} diff --git a/swin_small_4xb256_fake5m_lr1e-4_10epoch/20230601_165803/vis_data/config.py b/swin_small_4xb256_fake5m_lr1e-4_10epoch/20230601_165803/vis_data/config.py new file mode 100644 index 0000000000000000000000000000000000000000..07c9f4bb104a3c4f894e7a05987a8354fd463e1f --- /dev/null +++ b/swin_small_4xb256_fake5m_lr1e-4_10epoch/20230601_165803/vis_data/config.py @@ -0,0 +1,175 @@ +optim_wrapper = dict( + optimizer=dict( + type='AdamW', + lr=0.0001, + weight_decay=0.05, + eps=1e-08, + betas=(0.9, 0.999), + _scope_='mmpretrain'), + paramwise_cfg=dict( + norm_decay_mult=0.0, + bias_decay_mult=0.0, + flat_decay_mult=0.0, + custom_keys=dict({ + '.absolute_pos_embed': dict(decay_mult=0.0), + '.relative_position_bias_table': dict(decay_mult=0.0) + })), + type='AmpOptimWrapper', + dtype='bfloat16', + clip_grad=dict(max_norm=5.0)) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=True, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=1024) +model = dict( + type='ImageClassifier', + backbone=dict( + type='SwinTransformer', arch='small', img_size=224, + drop_path_rate=0.3), + neck=dict(type='GlobalAveragePooling'), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=768, + init_cfg=None, + loss=dict( + type='LabelSmoothLoss', label_smooth_val=0.1, mode='original'), + cal_acc=False), + init_cfg=[ + dict(type='TruncNormal', layer='Linear', std=0.02, bias=0.0), + dict(type='Constant', layer='LayerNorm', val=1.0, bias=0.0) + ], + train_cfg=None) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stablediffusionV1-5R2-dpmsolver-25-5m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/swin_small_4xb256_fake5m_lr1e-4_10epoch' diff --git a/swin_small_4xb256_fake5m_lr1e-4_10epoch/20230601_165803/vis_data/events.out.tfevents.1685609890.SH-IDC1-10-140-24-106.72230.0 b/swin_small_4xb256_fake5m_lr1e-4_10epoch/20230601_165803/vis_data/events.out.tfevents.1685609890.SH-IDC1-10-140-24-106.72230.0 new file mode 100644 index 0000000000000000000000000000000000000000..acdf9b8c715abcf5e120fd9a62141d671b45510c --- /dev/null +++ b/swin_small_4xb256_fake5m_lr1e-4_10epoch/20230601_165803/vis_data/events.out.tfevents.1685609890.SH-IDC1-10-140-24-106.72230.0 @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:a410c35711be7af2770377fad0b835c31158f85c2decc31c14bfa568b94c9d31 +size 279661 diff --git a/swin_small_4xb256_fake5m_lr1e-4_10epoch/20230601_165803/vis_data/scalars.json b/swin_small_4xb256_fake5m_lr1e-4_10epoch/20230601_165803/vis_data/scalars.json new file mode 100644 index 0000000000000000000000000000000000000000..4e6765d99e3dadc5047b8ab4363a2e5902d577b9 --- /dev/null +++ b/swin_small_4xb256_fake5m_lr1e-4_10epoch/20230601_165803/vis_data/scalars.json @@ -0,0 +1,197 @@ +{"lr": 0.0001, "data_time": 0.2611548662185669, "grad_norm": 0.9948760330677032, "loss": 0.5971919178962708, "time": 0.763830041885376, "epoch": 1, "memory": 28798, "step": 100} +{"lr": 0.0001, "data_time": 0.12748069763183595, "grad_norm": 1.033381861448288, "loss": 0.5732303202152252, "time": 0.7551178455352783, "epoch": 1, "memory": 28798, "step": 200} +{"lr": 0.0001, "data_time": 0.001685333251953125, "grad_norm": 2.0383641600608824, "loss": 0.5474633693695068, "time": 0.7133275032043457, "epoch": 1, "memory": 28798, "step": 300} +{"lr": 0.0001, "data_time": 0.0037578344345092773, "grad_norm": 2.4526761531829835, "loss": 0.5087024360895157, "time": 0.9227797508239746, "epoch": 1, "memory": 28798, "step": 400} +{"lr": 0.0001, "data_time": 0.0015799760818481444, "grad_norm": 2.6106780767440796, "loss": 0.502456533908844, "time": 0.8122771739959717, "epoch": 1, "memory": 28798, "step": 500} +{"lr": 0.0001, "data_time": 0.0016286849975585937, "grad_norm": 3.8179392337799074, "loss": 0.4700884073972702, "time": 0.798379921913147, "epoch": 1, "memory": 28798, "step": 600} +{"lr": 0.0001, "data_time": 0.0016109704971313476, "grad_norm": 6.124994063377381, "loss": 0.4786356151103973, "time": 0.7847991943359375, "epoch": 1, "memory": 28798, "step": 700} +{"lr": 0.0001, "data_time": 0.002045273780822754, "grad_norm": 5.492973017692566, "loss": 0.45380268096923826, "time": 0.8083441019058227, "epoch": 1, "memory": 28798, "step": 800} +{"lr": 0.0001, "data_time": 0.0018804311752319337, "grad_norm": 3.615518593788147, "loss": 0.450284481048584, "time": 0.7562673330307007, "epoch": 1, "memory": 28798, "step": 900} +{"lr": 0.0001, "data_time": 0.0014334917068481445, "grad_norm": 5.318289971351623, "loss": 0.42677096128463743, "time": 0.764702558517456, "epoch": 1, "memory": 28798, "step": 1000} +{"lr": 0.0001, "data_time": 0.001995277404785156, "grad_norm": 7.122946190834045, "loss": 0.431024095416069, "time": 0.7643571853637695, "epoch": 1, "memory": 28798, "step": 1100} +{"lr": 0.0001, "data_time": 0.0016049385070800782, "grad_norm": 8.187716698646545, "loss": 0.4295435696840286, "time": 0.7858440637588501, "epoch": 1, "memory": 28798, "step": 1200} +{"lr": 0.0001, "data_time": 0.0017634391784667968, "grad_norm": 6.058029818534851, "loss": 0.40536400079727175, "time": 0.7782463550567627, "epoch": 1, "memory": 28798, "step": 1300} +{"lr": 0.0001, "data_time": 0.0020489931106567384, "grad_norm": 7.24406931400299, "loss": 0.42019582688808443, "time": 0.8470067262649537, "epoch": 1, "memory": 28798, "step": 1400} +{"lr": 0.0001, "data_time": 0.0015645980834960937, "grad_norm": 9.415212643146514, "loss": 0.41330052018165586, "time": 0.749920916557312, "epoch": 1, "memory": 28798, "step": 1500} +{"lr": 0.0001, "data_time": 0.0018555402755737304, "grad_norm": 10.407112336158752, "loss": 0.41097092926502227, "time": 0.8086743831634522, "epoch": 1, "memory": 28798, "step": 1600} +{"lr": 0.0001, "data_time": 0.002411293983459473, "grad_norm": 8.974961280822754, "loss": 0.39800150096416476, "time": 0.8088978290557861, "epoch": 1, "memory": 28798, "step": 1700} +{"lr": 0.0001, "data_time": 0.002052760124206543, "grad_norm": 8.608369994163514, "loss": 0.38782241344451907, "time": 0.7794685125350952, "epoch": 1, "memory": 28798, "step": 1800} +{"lr": 0.0001, "data_time": 0.002239561080932617, "grad_norm": 7.995923852920532, "loss": 0.3900362730026245, "time": 0.7406694650650024, "epoch": 1, "memory": 28798, "step": 1900} +{"lr": 0.0001, "data_time": 0.0019824028015136717, "grad_norm": 12.193249082565307, "loss": 0.4058050960302353, "time": 0.7820487976074219, "epoch": 1, "memory": 28798, "step": 2000} +{"lr": 0.0001, "data_time": 0.0016951560974121094, "grad_norm": 8.75432903766632, "loss": 0.37981541752815245, "time": 0.7462271213531494, "epoch": 1, "memory": 28798, "step": 2100} +{"lr": 0.0001, "data_time": 0.0015626192092895509, "grad_norm": 12.1649915933609, "loss": 0.381779682636261, "time": 0.7927675485610962, "epoch": 1, "memory": 28798, "step": 2200} +{"lr": 0.0001, "data_time": 0.0016306161880493164, "grad_norm": 7.206118392944336, "loss": 0.3836579114198685, "time": 0.8494669437408447, "epoch": 1, "memory": 28798, "step": 2300} +{"lr": 0.0001, "data_time": 0.0014415025711059571, "grad_norm": 8.76898214817047, "loss": 0.36932373046875, "time": 0.7785675287246704, "epoch": 1, "memory": 28798, "step": 2400} +{"lr": 0.0001, "data_time": 0.0015406608581542969, "grad_norm": 9.264030408859252, "loss": 0.361787348985672, "time": 0.8009753942489624, "epoch": 1, "memory": 28798, "step": 2500} +{"lr": 0.0001, "data_time": 0.0015421152114868165, "grad_norm": 9.543509221076965, "loss": 0.4084623873233795, "time": 0.7717762231826782, "epoch": 1, "memory": 28798, "step": 2600} +{"lr": 0.0001, "data_time": 0.0016265392303466796, "grad_norm": 7.552957081794739, "loss": 0.3529518723487854, "time": 0.7745697021484375, "epoch": 1, "memory": 28798, "step": 2700} +{"lr": 0.0001, "data_time": 0.20046417713165282, "grad_norm": 8.773928093910218, "loss": 0.3352687120437622, "time": 0.8354354619979858, "epoch": 1, "memory": 28798, "step": 2800} +{"lr": 0.0001, "data_time": 0.0019224166870117187, "grad_norm": 10.194337511062622, "loss": 0.33998887836933134, "time": 0.7989093542099, "epoch": 1, "memory": 28798, "step": 2900} +{"lr": 0.0001, "data_time": 0.0017558574676513673, "grad_norm": 10.736096847057343, "loss": 0.35384186506271365, "time": 0.7650700807571411, "epoch": 1, "memory": 28798, "step": 3000} +{"lr": 0.0001, "data_time": 0.0019510269165039062, "grad_norm": 8.427406597137452, "loss": 0.3420874625444412, "time": 0.75934317111969, "epoch": 1, "memory": 28798, "step": 3100} +{"lr": 0.0001, "data_time": 0.0016048192977905274, "grad_norm": 9.468148756027222, "loss": 0.3274377554655075, "time": 0.8447348833084106, "epoch": 1, "memory": 28798, "step": 3200} +{"lr": 0.0001, "data_time": 0.0016980409622192384, "grad_norm": 11.842045307159424, "loss": 0.33908976018428805, "time": 0.8461315393447876, "epoch": 1, "memory": 28798, "step": 3300} +{"lr": 0.0001, "data_time": 0.0019853591918945314, "grad_norm": 14.057745933532715, "loss": 0.34925416111946106, "time": 0.7893282413482666, "epoch": 1, "memory": 28798, "step": 3400} +{"lr": 0.0001, "data_time": 0.0016464471817016601, "grad_norm": 11.661341524124145, "loss": 0.34215838015079497, "time": 0.8101903438568115, "epoch": 1, "memory": 28798, "step": 3500} +{"lr": 0.0001, "data_time": 0.0015691518783569336, "grad_norm": 12.739499568939209, "loss": 0.35349526107311247, "time": 0.7678930521011352, "epoch": 1, "memory": 28798, "step": 3600} +{"lr": 0.0001, "data_time": 0.0018642902374267577, "grad_norm": 14.655861473083496, "loss": 0.3317358762025833, "time": 0.7918614625930787, "epoch": 1, "memory": 28798, "step": 3700} +{"lr": 0.0001, "data_time": 0.001734614372253418, "grad_norm": 8.019277739524842, "loss": 0.3204199582338333, "time": 0.7980497121810913, "epoch": 1, "memory": 28798, "step": 3800} +{"lr": 0.0001, "data_time": 0.0017982244491577149, "grad_norm": 16.59911365509033, "loss": 0.3388776183128357, "time": 0.7871023178100586, "epoch": 1, "memory": 28798, "step": 3900} +{"lr": 0.0001, "data_time": 0.001947331428527832, "grad_norm": 11.131891167163849, "loss": 0.31053749918937684, "time": 0.9097763538360596, "epoch": 1, "memory": 28798, "step": 4000} +{"lr": 0.0001, "data_time": 0.0018900632858276367, "grad_norm": 7.135937070846557, "loss": 0.32608037292957304, "time": 0.8588865041732788, "epoch": 1, "memory": 28798, "step": 4100} +{"lr": 0.0001, "data_time": 0.002013373374938965, "grad_norm": 11.285934841632843, "loss": 0.33473441302776336, "time": 0.7751770257949829, "epoch": 1, "memory": 28798, "step": 4200} +{"lr": 0.0001, "data_time": 0.0017449378967285157, "grad_norm": 12.645131719112396, "loss": 0.3032625824213028, "time": 0.7565198183059693, "epoch": 1, "memory": 28798, "step": 4300} +{"lr": 0.0001, "data_time": 0.001753830909729004, "grad_norm": 14.853507995605469, "loss": 0.33213984668254853, "time": 0.8101833820343017, "epoch": 1, "memory": 28798, "step": 4400} +{"lr": 0.0001, "data_time": 0.0016500234603881836, "grad_norm": 13.149504327774048, "loss": 0.3288163751363754, "time": 0.8231769800186157, "epoch": 1, "memory": 28798, "step": 4500} +{"lr": 0.0001, "data_time": 0.001666688919067383, "grad_norm": 7.7415166854858395, "loss": 0.2992019057273865, "time": 0.7856241941452027, "epoch": 1, "memory": 28798, "step": 4600} +{"lr": 0.0001, "data_time": 0.001813220977783203, "grad_norm": 7.6865216255187985, "loss": 0.2817671522498131, "time": 0.7724915027618409, "epoch": 1, "memory": 28798, "step": 4700} +{"lr": 0.0001, "data_time": 0.0014267206192016602, "grad_norm": 7.778097343444824, "loss": 0.2861554190516472, "time": 0.7990787744522094, "epoch": 1, "memory": 28798, "step": 4800} +{"lr": 0.0001, "data_time": 0.0016474246978759766, "grad_norm": 7.936123585700988, "loss": 0.2836196631193161, "time": 0.8176966190338135, "epoch": 1, "memory": 28798, "step": 4900} +{"lr": 0.0001, "data_time": 0.0016502857208251954, "grad_norm": 8.367023301124572, "loss": 0.2896931916475296, "time": 0.7379039764404297, "epoch": 1, "memory": 28798, "step": 5000} +{"lr": 0.0001, "data_time": 0.0015475749969482422, "grad_norm": 7.881516706943512, "loss": 0.27526378631591797, "time": 0.7637993812561035, "epoch": 1, "memory": 28798, "step": 5100} +{"lr": 0.0001, "data_time": 0.0016379356384277344, "grad_norm": 5.626676559448242, "loss": 0.2834407016634941, "time": 0.793156623840332, "epoch": 1, "memory": 28798, "step": 5200} +{"lr": 0.0001, "data_time": 0.0014394044876098633, "grad_norm": 17.186562919616698, "loss": 0.3125844717025757, "time": 0.7624467611312866, "epoch": 1, "memory": 28798, "step": 5300} +{"lr": 0.0001, "data_time": 0.0015750169754028321, "grad_norm": 7.602498078346253, "loss": 0.2827661603689194, "time": 0.7655160188674927, "epoch": 1, "memory": 28798, "step": 5400} +{"lr": 0.0001, "data_time": 0.0016482353210449218, "grad_norm": 14.832658922672271, "loss": 0.29803421497344973, "time": 0.7661600589752198, "epoch": 1, "memory": 28798, "step": 5500} +{"lr": 0.0001, "data_time": 0.0015974044799804688, "grad_norm": 9.575963139533997, "loss": 0.28630885630846026, "time": 0.7695990562438965, "epoch": 1, "memory": 28798, "step": 5600} +{"lr": 0.0001, "data_time": 0.0017279386520385742, "grad_norm": 14.966920685768127, "loss": 0.3144871652126312, "time": 0.7377325057983398, "epoch": 1, "memory": 28798, "step": 5700} +{"accuracy/top1": 98.63644409179688, "data_time": 0.2643384635448456, "time": 0.8042525202035904, "step": 1} +{"lr": 9.779754323328192e-05, "data_time": 0.2253279685974121, "grad_norm": 4.724720764160156, "loss": 0.275802007317543, "time": 0.7853192567825318, "epoch": 2, "memory": 28798, "step": 5858} +{"lr": 9.779754323328192e-05, "data_time": 0.13539741039276124, "grad_norm": 8.829039835929871, "loss": 0.281463822722435, "time": 0.7541395902633667, "epoch": 2, "memory": 28783, "step": 5958} +{"lr": 9.779754323328192e-05, "data_time": 0.24535365104675294, "grad_norm": 4.85420960187912, "loss": 0.2631674766540527, "time": 0.7702489137649536, "epoch": 2, "memory": 28783, "step": 6058} +{"lr": 9.779754323328192e-05, "data_time": 0.16128273010253907, "grad_norm": 6.4906369090080265, "loss": 0.26960329711437225, "time": 0.8204433679580688, "epoch": 2, "memory": 28783, "step": 6158} +{"lr": 9.779754323328192e-05, "data_time": 0.0019009828567504883, "grad_norm": 7.4814086079597475, "loss": 0.26002991646528245, "time": 0.7683497428894043, "epoch": 2, "memory": 28783, "step": 6258} +{"lr": 9.779754323328192e-05, "data_time": 0.0017068624496459962, "grad_norm": 3.6976542353630064, "loss": 0.2582589194178581, "time": 0.7667133808135986, "epoch": 2, "memory": 28783, "step": 6358} +{"lr": 9.779754323328192e-05, "data_time": 0.001719522476196289, "grad_norm": 5.707702016830444, "loss": 0.2515236109495163, "time": 0.8664642095565795, "epoch": 2, "memory": 28783, "step": 6458} +{"lr": 9.779754323328192e-05, "data_time": 0.0014872312545776366, "grad_norm": 4.434868597984314, "loss": 0.25663683712482455, "time": 0.7268047094345093, "epoch": 2, "memory": 28783, "step": 6558} +{"lr": 9.779754323328192e-05, "data_time": 0.0016342878341674804, "grad_norm": 8.42042521238327, "loss": 0.24680227935314178, "time": 0.7821118116378785, "epoch": 2, "memory": 28783, "step": 6658} +{"lr": 9.779754323328192e-05, "data_time": 0.0018155574798583984, "grad_norm": 5.102838933467865, "loss": 0.2469940662384033, "time": 0.8128172159194946, "epoch": 2, "memory": 28783, "step": 6758} +{"lr": 9.779754323328192e-05, "data_time": 0.0015127420425415038, "grad_norm": 7.1840561628341675, "loss": 0.2738844335079193, "time": 0.7818080663681031, "epoch": 2, "memory": 28783, "step": 6858} +{"lr": 9.779754323328192e-05, "data_time": 0.001641225814819336, "grad_norm": 9.520950174331665, "loss": 0.2546579748392105, "time": 0.7373530149459839, "epoch": 2, "memory": 28783, "step": 6958} +{"lr": 9.779754323328192e-05, "data_time": 0.001997566223144531, "grad_norm": 7.790714287757874, "loss": 0.2447388768196106, "time": 0.7828716993331909, "epoch": 2, "memory": 28783, "step": 7058} +{"lr": 9.779754323328192e-05, "data_time": 0.0018570899963378906, "grad_norm": 6.547972404956818, "loss": 0.24792872965335847, "time": 0.8393717527389526, "epoch": 2, "memory": 28783, "step": 7158} +{"lr": 9.779754323328192e-05, "data_time": 0.0016109704971313476, "grad_norm": 4.401484966278076, "loss": 0.2534740626811981, "time": 0.7773492574691773, "epoch": 2, "memory": 28783, "step": 7258} +{"lr": 9.779754323328192e-05, "data_time": 0.0023687124252319337, "grad_norm": 3.6738228678703306, "loss": 0.2359653428196907, "time": 0.916810417175293, "epoch": 2, "memory": 28783, "step": 7358} +{"lr": 9.779754323328192e-05, "data_time": 0.0017107009887695312, "grad_norm": 11.95268611907959, "loss": 0.24162269234657288, "time": 0.7193113327026367, "epoch": 2, "memory": 28783, "step": 7458} +{"lr": 9.779754323328192e-05, "data_time": 0.001832890510559082, "grad_norm": 1.9830315947532653, "loss": 0.24130344837903978, "time": 0.7895451784133911, "epoch": 2, "memory": 28783, "step": 7558} +{"lr": 9.779754323328192e-05, "data_time": 0.0016902923583984376, "grad_norm": 5.427372705936432, "loss": 0.2390815481543541, "time": 0.7553702116012573, "epoch": 2, "memory": 28783, "step": 7658} +{"lr": 9.779754323328192e-05, "data_time": 0.0018090486526489257, "grad_norm": 4.1846026420593265, "loss": 0.23522198796272278, "time": 0.7588916540145874, "epoch": 2, "memory": 28783, "step": 7758} +{"lr": 9.779754323328192e-05, "data_time": 0.0022718429565429686, "grad_norm": 4.5128955960273744, "loss": 0.2320399358868599, "time": 0.7661934852600097, "epoch": 2, "memory": 28783, "step": 7858} +{"lr": 9.779754323328192e-05, "data_time": 0.0017469406127929687, "grad_norm": 5.278597688674926, "loss": 0.2440346896648407, "time": 0.7454902172088623, "epoch": 2, "memory": 28783, "step": 7958} +{"lr": 9.779754323328192e-05, "data_time": 0.0017522573471069336, "grad_norm": 4.699299311637878, "loss": 0.2368510529398918, "time": 0.7240928888320923, "epoch": 2, "memory": 28783, "step": 8058} +{"lr": 9.779754323328192e-05, "data_time": 0.0015929460525512696, "grad_norm": 5.631152653694153, "loss": 0.24275548309087752, "time": 0.7127774000167847, "epoch": 2, "memory": 28783, "step": 8158} +{"lr": 9.779754323328192e-05, "data_time": 0.0022130489349365236, "grad_norm": 3.8032525062561033, "loss": 0.22538014203310014, "time": 0.7257648468017578, "epoch": 2, "memory": 28783, "step": 8258} +{"lr": 9.779754323328192e-05, "data_time": 0.0017465829849243164, "grad_norm": 5.955878686904907, "loss": 0.2376284494996071, "time": 0.7556620836257935, "epoch": 2, "memory": 28783, "step": 8358} +{"lr": 9.779754323328192e-05, "data_time": 0.0017087459564208984, "grad_norm": 4.831857204437256, "loss": 0.23309049010276794, "time": 0.7472589492797852, "epoch": 2, "memory": 28783, "step": 8458} +{"lr": 9.779754323328192e-05, "data_time": 0.0015890836715698243, "grad_norm": 3.156527227163315, "loss": 0.22835725992918016, "time": 0.8053215265274047, "epoch": 2, "memory": 28783, "step": 8558} +{"lr": 9.779754323328192e-05, "data_time": 0.001584339141845703, "grad_norm": 1.9715153217315673, "loss": 0.22326327562332154, "time": 0.7155921936035157, "epoch": 2, "memory": 28783, "step": 8658} +{"lr": 9.779754323328192e-05, "data_time": 0.0017405271530151368, "grad_norm": 3.204046630859375, "loss": 0.22883425951004027, "time": 0.6969146490097046, "epoch": 2, "memory": 28783, "step": 8758} +{"lr": 9.779754323328192e-05, "data_time": 0.0016877174377441407, "grad_norm": 1.5660253405570983, "loss": 0.21822546571493148, "time": 0.7188021898269653, "epoch": 2, "memory": 28783, "step": 8858} +{"lr": 9.779754323328192e-05, "data_time": 0.0022782325744628907, "grad_norm": 2.75571203827858, "loss": 0.22351438105106353, "time": 0.8173815011978149, "epoch": 2, "memory": 28783, "step": 8958} +{"lr": 9.779754323328192e-05, "data_time": 0.001535797119140625, "grad_norm": 2.5499474883079527, "loss": 0.22208157926797867, "time": 0.7032848834991455, "epoch": 2, "memory": 28783, "step": 9058} +{"lr": 9.779754323328192e-05, "data_time": 0.0016241788864135743, "grad_norm": 3.831286382675171, "loss": 0.22439639270305634, "time": 0.7431313276290894, "epoch": 2, "memory": 28783, "step": 9158} +{"lr": 9.779754323328192e-05, "data_time": 0.002005267143249512, "grad_norm": 5.158092880249024, "loss": 0.2316832423210144, "time": 0.7462419509887696, "epoch": 2, "memory": 28783, "step": 9258} +{"lr": 9.779754323328192e-05, "data_time": 0.0016271352767944336, "grad_norm": 3.846285009384155, "loss": 0.22138564735651017, "time": 0.7830148458480835, "epoch": 2, "memory": 28783, "step": 9358} +{"lr": 9.779754323328192e-05, "data_time": 0.001618647575378418, "grad_norm": 3.1761711955070497, "loss": 0.22847191393375396, "time": 0.7664671659469604, "epoch": 2, "memory": 28783, "step": 9458} +{"lr": 9.779754323328192e-05, "data_time": 0.001716923713684082, "grad_norm": 7.066403663158416, "loss": 0.21818615198135377, "time": 0.7468266248703003, "epoch": 2, "memory": 28783, "step": 9558} +{"lr": 9.779754323328192e-05, "data_time": 0.0015748977661132813, "grad_norm": 2.3538535118103026, "loss": 0.22665224373340606, "time": 0.7534285545349121, "epoch": 2, "memory": 28783, "step": 9658} +{"lr": 9.779754323328192e-05, "data_time": 0.001693415641784668, "grad_norm": 2.245039939880371, "loss": 0.22155413776636124, "time": 0.7146346807479859, "epoch": 2, "memory": 28783, "step": 9758} +{"lr": 9.779754323328192e-05, "data_time": 0.0020279645919799804, "grad_norm": 3.1565908253192902, "loss": 0.22076133638620377, "time": 0.7443537473678589, "epoch": 2, "memory": 28783, "step": 9858} +{"lr": 9.779754323328192e-05, "data_time": 0.001599287986755371, "grad_norm": 4.543850445747376, "loss": 0.22001899033784866, "time": 0.7957648038864136, "epoch": 2, "memory": 28783, "step": 9958} +{"lr": 9.779754323328192e-05, "data_time": 0.0015918254852294923, "grad_norm": 2.7435157179832457, "loss": 0.22163452655076982, "time": 0.8049781560897827, "epoch": 2, "memory": 28783, "step": 10058} +{"lr": 9.779754323328192e-05, "data_time": 0.0025667905807495116, "grad_norm": 3.7365896940231322, "loss": 0.22913289219141006, "time": 0.7350122690200805, "epoch": 2, "memory": 28783, "step": 10158} +{"lr": 9.779754323328192e-05, "data_time": 0.0014409542083740235, "grad_norm": 4.982752060890197, "loss": 0.22232090830802917, "time": 0.7107707738876343, "epoch": 2, "memory": 28783, "step": 10258} +{"lr": 9.779754323328192e-05, "data_time": 0.0020370721817016602, "grad_norm": 1.8967858254909515, "loss": 0.2177594393491745, "time": 0.7625745058059692, "epoch": 2, "memory": 28783, "step": 10358} +{"lr": 9.779754323328192e-05, "data_time": 0.0019598484039306642, "grad_norm": 1.9121993958950043, "loss": 0.20995037108659745, "time": 0.7142647266387939, "epoch": 2, "memory": 28783, "step": 10458} +{"lr": 9.779754323328192e-05, "data_time": 0.0018325328826904296, "grad_norm": 2.5711088299751284, "loss": 0.21899385899305343, "time": 0.8342780828475952, "epoch": 2, "memory": 28783, "step": 10558} +{"lr": 9.779754323328192e-05, "data_time": 1.8292505979537963, "grad_norm": 2.879544985294342, "loss": 0.22317344844341278, "time": 2.6419941425323485, "epoch": 2, "memory": 28783, "step": 10658} +{"lr": 9.779754323328192e-05, "data_time": 0.2317431926727295, "grad_norm": 1.296843558549881, "loss": 0.21406361907720567, "time": 0.7352376699447631, "epoch": 2, "memory": 28783, "step": 10758} +{"lr": 9.779754323328192e-05, "data_time": 0.2768770456314087, "grad_norm": 3.4218323826789856, "loss": 0.22064824104309083, "time": 0.7944846868515014, "epoch": 2, "memory": 28783, "step": 10858} +{"lr": 9.779754323328192e-05, "data_time": 0.1599552869796753, "grad_norm": 1.7299805223941802, "loss": 0.21627172380685805, "time": 0.6702358722686768, "epoch": 2, "memory": 28783, "step": 10958} +{"lr": 9.779754323328192e-05, "data_time": 0.18956103324890136, "grad_norm": 1.409479123353958, "loss": 0.22424711883068085, "time": 0.691569185256958, "epoch": 2, "memory": 28783, "step": 11058} +{"lr": 9.779754323328192e-05, "data_time": 0.10107715129852295, "grad_norm": 2.3308180451393126, "loss": 0.2175878569483757, "time": 0.6839389324188232, "epoch": 2, "memory": 28783, "step": 11158} +{"lr": 9.779754323328192e-05, "data_time": 0.0015223026275634766, "grad_norm": 1.9041896104812621, "loss": 0.21678828150033952, "time": 0.7197707891464233, "epoch": 2, "memory": 28783, "step": 11258} +{"lr": 9.779754323328192e-05, "data_time": 0.0017743349075317384, "grad_norm": 2.214759635925293, "loss": 0.21667275279760362, "time": 0.7230534315109253, "epoch": 2, "memory": 28783, "step": 11358} +{"lr": 9.779754323328192e-05, "data_time": 0.0017892122268676758, "grad_norm": 1.5354084491729736, "loss": 0.22148067355155945, "time": 0.7298334836959839, "epoch": 2, "memory": 28783, "step": 11458} +{"accuracy/top1": 99.57704162597656, "data_time": 0.2029575179604923, "time": 0.6182985586278579, "step": 2} +{"lr": 9.140576474687264e-05, "data_time": 0.0017037391662597656, "grad_norm": 2.547348564863205, "loss": 0.22042838335037232, "time": 0.7363026142120361, "epoch": 3, "memory": 28783, "step": 11616} +{"lr": 9.140576474687264e-05, "data_time": 0.0016712188720703126, "grad_norm": 1.9135839641094208, "loss": 0.21216972321271896, "time": 0.7759701251983643, "epoch": 3, "memory": 28783, "step": 11716} +{"lr": 9.140576474687264e-05, "data_time": 0.002136373519897461, "grad_norm": 1.8601315200328827, "loss": 0.21522459536790847, "time": 0.7624718904495239, "epoch": 3, "memory": 28783, "step": 11816} +{"lr": 9.140576474687264e-05, "data_time": 0.0017839670181274414, "grad_norm": 1.554959350824356, "loss": 0.21358174085617065, "time": 0.7393719673156738, "epoch": 3, "memory": 28783, "step": 11916} +{"lr": 9.140576474687264e-05, "data_time": 0.001764988899230957, "grad_norm": 2.924062669277191, "loss": 0.21263508796691893, "time": 0.7282135248184204, "epoch": 3, "memory": 28783, "step": 12016} +{"lr": 9.140576474687264e-05, "data_time": 0.0017034053802490235, "grad_norm": 2.0709295213222503, "loss": 0.21246255189180374, "time": 0.8083432435989379, "epoch": 3, "memory": 28783, "step": 12116} +{"lr": 9.140576474687264e-05, "data_time": 0.0017165660858154297, "grad_norm": 2.144436377286911, "loss": 0.21445005834102632, "time": 0.7608980178833008, "epoch": 3, "memory": 28783, "step": 12216} +{"lr": 9.140576474687264e-05, "data_time": 0.0017304658889770509, "grad_norm": 1.3800033688545228, "loss": 0.214017653465271, "time": 0.8114840269088746, "epoch": 3, "memory": 28783, "step": 12316} +{"lr": 9.140576474687264e-05, "data_time": 0.0015224456787109376, "grad_norm": 1.5817308306694031, "loss": 0.20982833504676818, "time": 0.7176219463348389, "epoch": 3, "memory": 28783, "step": 12416} +{"lr": 9.140576474687264e-05, "data_time": 0.0017634391784667968, "grad_norm": 1.1910838425159453, "loss": 0.2121584177017212, "time": 0.89136221408844, "epoch": 3, "memory": 28783, "step": 12516} +{"lr": 9.140576474687264e-05, "data_time": 0.0018435239791870118, "grad_norm": 1.829367995262146, "loss": 0.21225885003805162, "time": 0.7900148868560791, "epoch": 3, "memory": 28783, "step": 12616} +{"lr": 9.140576474687264e-05, "data_time": 0.10203487873077392, "grad_norm": 1.2540997505187987, "loss": 0.2087197095155716, "time": 0.8613012075424195, "epoch": 3, "memory": 28783, "step": 12716} +{"lr": 9.140576474687264e-05, "data_time": 0.0017075538635253906, "grad_norm": 1.3484358072280884, "loss": 0.2111063003540039, "time": 0.8129325628280639, "epoch": 3, "memory": 28783, "step": 12816} +{"lr": 9.140576474687264e-05, "data_time": 0.0018035650253295898, "grad_norm": 1.1028524696826936, "loss": 0.21101672053337098, "time": 0.8085900545120239, "epoch": 3, "memory": 28783, "step": 12916} +{"lr": 9.140576474687264e-05, "data_time": 0.001885843276977539, "grad_norm": 1.2602094292640686, "loss": 0.21546115577220917, "time": 0.79597008228302, "epoch": 3, "memory": 28783, "step": 13016} +{"lr": 9.140576474687264e-05, "data_time": 0.0019144773483276366, "grad_norm": 1.3525001168251038, "loss": 0.20870959013700485, "time": 0.7559643983840942, "epoch": 3, "memory": 28783, "step": 13116} +{"lr": 9.140576474687264e-05, "data_time": 0.001648259162902832, "grad_norm": 1.2084201335906983, "loss": 0.21275661140680313, "time": 0.7298112869262695, "epoch": 3, "memory": 28783, "step": 13216} +{"lr": 9.140576474687264e-05, "data_time": 0.0015499591827392578, "grad_norm": 1.3942279934883117, "loss": 0.21074751168489456, "time": 0.8349781990051269, "epoch": 3, "memory": 28783, "step": 13316} +{"lr": 9.140576474687264e-05, "data_time": 0.002311038970947266, "grad_norm": 2.2864261150360106, "loss": 0.21232101023197175, "time": 0.7560360908508301, "epoch": 3, "memory": 28783, "step": 13416} +{"lr": 9.140576474687264e-05, "data_time": 0.0017961025238037109, "grad_norm": 1.2045234262943267, "loss": 0.20949556529521943, "time": 0.7641310214996337, "epoch": 3, "memory": 28783, "step": 13516} +{"lr": 9.140576474687264e-05, "data_time": 0.0019094705581665038, "grad_norm": 1.4417009353637695, "loss": 0.2153782069683075, "time": 0.7750011682510376, "epoch": 3, "memory": 28783, "step": 13616} +{"lr": 9.140576474687264e-05, "data_time": 0.09557888507843018, "grad_norm": 1.544138091802597, "loss": 0.21417417228221894, "time": 0.7901232004165649, "epoch": 3, "memory": 28783, "step": 13716} +{"lr": 9.140576474687264e-05, "data_time": 0.06048867702484131, "grad_norm": 2.211031836271286, "loss": 0.2095340371131897, "time": 0.7652333498001098, "epoch": 3, "memory": 28783, "step": 13816} +{"lr": 9.140576474687264e-05, "data_time": 0.0013819217681884765, "grad_norm": 1.0305872529745101, "loss": 0.20918484181165695, "time": 0.8278063058853149, "epoch": 3, "memory": 28783, "step": 13916} +{"lr": 9.140576474687264e-05, "data_time": 0.10765769481658935, "grad_norm": 2.019196939468384, "loss": 0.22233359515666962, "time": 0.7532198429107666, "epoch": 3, "memory": 28783, "step": 14016} +{"lr": 9.140576474687264e-05, "data_time": 0.27614758014678953, "grad_norm": 0.9989032059907913, "loss": 0.2159930184483528, "time": 0.7803572416305542, "epoch": 3, "memory": 28783, "step": 14116} +{"lr": 9.140576474687264e-05, "data_time": 0.21318349838256836, "grad_norm": 1.3191742479801178, "loss": 0.20799032896757125, "time": 0.7172081232070923, "epoch": 3, "memory": 28783, "step": 14216} +{"lr": 9.140576474687264e-05, "data_time": 0.260835862159729, "grad_norm": 1.3699993193149567, "loss": 0.21112867444753647, "time": 0.7647407293319702, "epoch": 3, "memory": 28783, "step": 14316} +{"lr": 9.140576474687264e-05, "data_time": 0.1925347089767456, "grad_norm": 1.387695175409317, "loss": 0.21318716406822205, "time": 0.7368336915969849, "epoch": 3, "memory": 28783, "step": 14416} +{"lr": 9.140576474687264e-05, "data_time": 0.0015162944793701172, "grad_norm": 1.239967554807663, "loss": 0.2061662420630455, "time": 0.7891481876373291, "epoch": 3, "memory": 28783, "step": 14516} +{"lr": 9.140576474687264e-05, "data_time": 0.0018846750259399413, "grad_norm": 2.0698291540145872, "loss": 0.21210735589265822, "time": 0.7504413366317749, "epoch": 3, "memory": 28783, "step": 14616} +{"lr": 9.140576474687264e-05, "data_time": 0.0018059730529785157, "grad_norm": 2.023476779460907, "loss": 0.2098442181944847, "time": 0.7859753608703614, "epoch": 3, "memory": 28783, "step": 14716} +{"lr": 9.140576474687264e-05, "data_time": 0.0015339612960815429, "grad_norm": 1.0350459814071655, "loss": 0.2101355627179146, "time": 0.7155972480773926, "epoch": 3, "memory": 28783, "step": 14816} +{"lr": 9.140576474687264e-05, "data_time": 0.0015639543533325195, "grad_norm": 1.1589035451412202, "loss": 0.2069369599223137, "time": 0.7649103879928589, "epoch": 3, "memory": 28783, "step": 14916} +{"lr": 9.140576474687264e-05, "data_time": 0.00160672664642334, "grad_norm": 1.2026117026805878, "loss": 0.20926864445209503, "time": 0.7189409732818604, "epoch": 3, "memory": 28783, "step": 15016} +{"lr": 9.140576474687264e-05, "data_time": 0.001593303680419922, "grad_norm": 0.9875022858381272, "loss": 0.20764354169368743, "time": 0.7618232488632202, "epoch": 3, "memory": 28783, "step": 15116} +{"lr": 9.140576474687264e-05, "data_time": 0.0016371250152587891, "grad_norm": 1.5114789605140686, "loss": 0.20797639191150666, "time": 0.8211654901504517, "epoch": 3, "memory": 28783, "step": 15216} +{"lr": 9.140576474687264e-05, "data_time": 0.0016020774841308595, "grad_norm": 1.2105599641799927, "loss": 0.21090261340141297, "time": 0.7385528564453125, "epoch": 3, "memory": 28783, "step": 15316} +{"lr": 9.140576474687264e-05, "data_time": 0.0018048286437988281, "grad_norm": 0.9896108508110046, "loss": 0.210080286860466, "time": 0.7755727052688599, "epoch": 3, "memory": 28783, "step": 15416} +{"lr": 9.140576474687264e-05, "data_time": 0.048856568336486814, "grad_norm": 2.026436024904251, "loss": 0.21362337619066238, "time": 0.7411954164505005, "epoch": 3, "memory": 28783, "step": 15516} +{"lr": 9.140576474687264e-05, "data_time": 0.0016203880310058593, "grad_norm": 1.3295194804668427, "loss": 0.21074522137641907, "time": 0.7665291547775268, "epoch": 3, "memory": 28783, "step": 15616} +{"lr": 9.140576474687264e-05, "data_time": 0.0014732837677001952, "grad_norm": 1.3675204813480377, "loss": 0.20556535720825195, "time": 0.8190885543823242, "epoch": 3, "memory": 28783, "step": 15716} +{"lr": 9.140576474687264e-05, "data_time": 0.0018591165542602539, "grad_norm": 1.2479036509990693, "loss": 0.20555228590965272, "time": 0.8033952474594116, "epoch": 3, "memory": 28783, "step": 15816} +{"lr": 9.140576474687264e-05, "data_time": 0.0015214920043945313, "grad_norm": 1.1333971738815307, "loss": 0.2112056851387024, "time": 0.7313814163208008, "epoch": 3, "memory": 28783, "step": 15916} +{"lr": 9.140576474687264e-05, "data_time": 0.001503896713256836, "grad_norm": 1.0096331983804703, "loss": 0.2103559747338295, "time": 0.7655946254730225, "epoch": 3, "memory": 28783, "step": 16016} +{"lr": 9.140576474687264e-05, "data_time": 0.001732039451599121, "grad_norm": 0.7067256242036819, "loss": 0.20596436262130738, "time": 0.7798765659332275, "epoch": 3, "memory": 28783, "step": 16116} +{"lr": 9.140576474687264e-05, "data_time": 0.0014956235885620118, "grad_norm": 0.7893205374479294, "loss": 0.20947739481925964, "time": 0.8068555355072021, "epoch": 3, "memory": 28783, "step": 16216} +{"lr": 9.140576474687264e-05, "data_time": 0.0016243696212768556, "grad_norm": 1.3125432908535004, "loss": 0.20244491696357728, "time": 0.9077004432678223, "epoch": 3, "memory": 28783, "step": 16316} +{"lr": 9.140576474687264e-05, "data_time": 0.0015056133270263672, "grad_norm": 1.268625909090042, "loss": 0.2104177340865135, "time": 0.7249560117721557, "epoch": 3, "memory": 28783, "step": 16416} +{"lr": 9.140576474687264e-05, "data_time": 0.0017547607421875, "grad_norm": 0.6721316695213317, "loss": 0.20510686337947845, "time": 0.7545232772827148, "epoch": 3, "memory": 28783, "step": 16516} +{"lr": 9.140576474687264e-05, "data_time": 0.0018735408782958984, "grad_norm": 1.4055960297584533, "loss": 0.21015082597732543, "time": 0.765139126777649, "epoch": 3, "memory": 28783, "step": 16616} +{"lr": 9.140576474687264e-05, "data_time": 0.001816701889038086, "grad_norm": 1.3470157921314239, "loss": 0.20931606888771057, "time": 0.6972701787948609, "epoch": 3, "memory": 28783, "step": 16716} +{"lr": 9.140576474687264e-05, "data_time": 0.0018264293670654298, "grad_norm": 1.309765875339508, "loss": 0.2060650423169136, "time": 0.7734025239944458, "epoch": 3, "memory": 28783, "step": 16816} +{"lr": 9.140576474687264e-05, "data_time": 0.0017119884490966798, "grad_norm": 0.8919267356395721, "loss": 0.20747613459825515, "time": 0.8131618738174439, "epoch": 3, "memory": 28783, "step": 16916} +{"lr": 9.140576474687264e-05, "data_time": 0.0015355348587036133, "grad_norm": 0.9887870609760284, "loss": 0.2054794982075691, "time": 0.7259827375411987, "epoch": 3, "memory": 28783, "step": 17016} +{"lr": 9.140576474687264e-05, "data_time": 0.0016350269317626954, "grad_norm": 0.9397801756858826, "loss": 0.2074136883020401, "time": 0.8159034252166748, "epoch": 3, "memory": 28783, "step": 17116} +{"lr": 9.140576474687264e-05, "data_time": 0.0016236305236816406, "grad_norm": 0.9886764645576477, "loss": 0.20989191085100173, "time": 0.7101340532302857, "epoch": 3, "memory": 28783, "step": 17216} +{"accuracy/top1": 99.425537109375, "data_time": 0.20295374533709357, "time": 0.6162600657519173, "step": 3} +{"lr": 8.14503363531613e-05, "data_time": 0.07129447460174561, "grad_norm": 0.7139622837305069, "loss": 0.20446813106536865, "time": 0.7448784112930298, "epoch": 4, "memory": 28783, "step": 17374} +{"lr": 8.14503363531613e-05, "data_time": 0.0016505718231201172, "grad_norm": 0.8423473298549652, "loss": 0.2116495981812477, "time": 0.7120396614074707, "epoch": 4, "memory": 28783, "step": 17474} +{"lr": 8.14503363531613e-05, "data_time": 0.001586008071899414, "grad_norm": 0.9319919884204865, "loss": 0.20657327473163606, "time": 0.7979679584503174, "epoch": 4, "memory": 28783, "step": 17574} +{"lr": 8.14503363531613e-05, "data_time": 0.001841568946838379, "grad_norm": 0.9374666512012482, "loss": 0.20756369084119797, "time": 0.7048688173294068, "epoch": 4, "memory": 28783, "step": 17674} +{"lr": 8.14503363531613e-05, "data_time": 0.0018159866333007813, "grad_norm": 1.1117026597261428, "loss": 0.20703705102205278, "time": 0.8853085994720459, "epoch": 4, "memory": 28783, "step": 17774} +{"lr": 8.14503363531613e-05, "data_time": 0.0017165660858154297, "grad_norm": 1.121205472946167, "loss": 0.20755206942558288, "time": 0.7465423107147217, "epoch": 4, "memory": 28783, "step": 17874} +{"lr": 8.14503363531613e-05, "data_time": 0.0016196966171264648, "grad_norm": 0.8346521109342575, "loss": 0.20673878639936447, "time": 0.9494416952133179, "epoch": 4, "memory": 28783, "step": 17974} +{"lr": 8.14503363531613e-05, "data_time": 0.001536107063293457, "grad_norm": 1.3260571867227555, "loss": 0.20651856809854507, "time": 1.0259686946868896, "epoch": 4, "memory": 28783, "step": 18074} +{"lr": 8.14503363531613e-05, "data_time": 0.06866841316223145, "grad_norm": 0.858877283334732, "loss": 0.2090943679213524, "time": 1.0382978439331054, "epoch": 4, "memory": 28783, "step": 18174} +{"lr": 8.14503363531613e-05, "data_time": 0.001637101173400879, "grad_norm": 0.977012836933136, "loss": 0.20690066516399383, "time": 0.7482481479644776, "epoch": 4, "memory": 28783, "step": 18274} +{"lr": 8.14503363531613e-05, "data_time": 0.001680302619934082, "grad_norm": 0.8573934376239777, "loss": 0.20402872264385224, "time": 0.796010947227478, "epoch": 4, "memory": 28783, "step": 18374} +{"lr": 8.14503363531613e-05, "data_time": 0.001658010482788086, "grad_norm": 1.1316018223762512, "loss": 0.20936628729104995, "time": 0.8131815671920777, "epoch": 4, "memory": 28783, "step": 18474} +{"lr": 8.14503363531613e-05, "data_time": 0.0015490531921386718, "grad_norm": 1.2810150742530824, "loss": 0.20540878623723985, "time": 0.854335618019104, "epoch": 4, "memory": 28783, "step": 18574} +{"lr": 8.14503363531613e-05, "data_time": 0.001682138442993164, "grad_norm": 0.9204590931534767, "loss": 0.20437522530555724, "time": 0.8720471858978271, "epoch": 4, "memory": 28783, "step": 18674} +{"lr": 8.14503363531613e-05, "data_time": 0.0015542984008789062, "grad_norm": 1.2260378628969193, "loss": 0.20431203842163087, "time": 1.462713623046875, "epoch": 4, "memory": 28783, "step": 18774} +{"lr": 8.14503363531613e-05, "data_time": 0.0015694379806518554, "grad_norm": 0.802131500840187, "loss": 0.21011782586574554, "time": 1.0631263732910157, "epoch": 4, "memory": 28783, "step": 18874} +{"lr": 8.14503363531613e-05, "data_time": 0.0018329381942749023, "grad_norm": 1.4100565791130066, "loss": 0.20812755972146987, "time": 0.9850741386413574, "epoch": 4, "memory": 28783, "step": 18974} +{"lr": 8.14503363531613e-05, "data_time": 0.0013610601425170898, "grad_norm": 1.8439662277698516, "loss": 0.21002061516046525, "time": 1.1623518466949463, "epoch": 4, "memory": 28783, "step": 19074} +{"lr": 8.14503363531613e-05, "data_time": 0.4064238786697388, "grad_norm": 0.7899254560470581, "loss": 0.2045873299241066, "time": 1.0554860353469848, "epoch": 4, "memory": 28783, "step": 19174} +{"lr": 8.14503363531613e-05, "data_time": 0.6836257696151733, "grad_norm": 0.8103839159011841, "loss": 0.20875571221113204, "time": 1.1865941047668458, "epoch": 4, "memory": 28783, "step": 19274} +{"lr": 8.14503363531613e-05, "data_time": 0.2250920295715332, "grad_norm": 0.7874846875667572, "loss": 0.20594585537910462, "time": 0.7385987520217896, "epoch": 4, "memory": 28783, "step": 19374} +{"lr": 8.14503363531613e-05, "data_time": 0.24105098247528076, "grad_norm": 1.353374284505844, "loss": 0.20811700969934463, "time": 0.7392971515655518, "epoch": 4, "memory": 28783, "step": 19474} +{"lr": 8.14503363531613e-05, "data_time": 0.1747352361679077, "grad_norm": 0.9965912759304046, "loss": 0.20290562957525254, "time": 0.7055529117584228, "epoch": 4, "memory": 28783, "step": 19574} diff --git a/swin_small_4xb256_fake5m_lr1e-4_10epoch/20230601_165803/vis_data/vis_image/0000000.png_1.png b/swin_small_4xb256_fake5m_lr1e-4_10epoch/20230601_165803/vis_data/vis_image/0000000.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..b44147a5c0411cb50267bbc3e54b286eff6d6bc7 Binary files /dev/null and b/swin_small_4xb256_fake5m_lr1e-4_10epoch/20230601_165803/vis_data/vis_image/0000000.png_1.png differ diff --git a/swin_small_4xb256_fake5m_lr1e-4_10epoch/20230601_165803/vis_data/vis_image/0000000.png_2.png b/swin_small_4xb256_fake5m_lr1e-4_10epoch/20230601_165803/vis_data/vis_image/0000000.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..b2aa1aaaf9f97ce8b19f3bbce6b7024e76a7ba4c Binary files /dev/null and b/swin_small_4xb256_fake5m_lr1e-4_10epoch/20230601_165803/vis_data/vis_image/0000000.png_2.png differ diff --git a/swin_small_4xb256_fake5m_lr1e-4_10epoch/20230601_165803/vis_data/vis_image/0000000.png_3.png b/swin_small_4xb256_fake5m_lr1e-4_10epoch/20230601_165803/vis_data/vis_image/0000000.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..b44147a5c0411cb50267bbc3e54b286eff6d6bc7 Binary files /dev/null and b/swin_small_4xb256_fake5m_lr1e-4_10epoch/20230601_165803/vis_data/vis_image/0000000.png_3.png differ diff --git a/swin_small_4xb256_fake5m_lr1e-4_10epoch/20230601_215742/20230601_215742.log b/swin_small_4xb256_fake5m_lr1e-4_10epoch/20230601_215742/20230601_215742.log new file mode 100644 index 0000000000000000000000000000000000000000..1c71c4f4ff5686c7de0d4adb7ce9fabbbc3d0573 --- /dev/null +++ b/swin_small_4xb256_fake5m_lr1e-4_10epoch/20230601_215742/20230601_215742.log @@ -0,0 +1,2214 @@ +2023/06/01 21:57:46 - mmengine - INFO - +------------------------------------------------------------ +System environment: + sys.platform: linux + Python: 3.10.9 (main, Mar 8 2023, 10:47:38) [GCC 11.2.0] + CUDA available: True + numpy_random_seed: 267371071 + GPU 0,1,2,3: NVIDIA A100-SXM4-80GB + CUDA_HOME: /mnt/petrelfs/share/cuda-11.6 + NVCC: Cuda compilation tools, release 11.6, V11.6.124 + GCC: gcc (GCC) 4.8.5 20150623 (Red Hat 4.8.5-44) + PyTorch: 1.13.1 + PyTorch compiling details: PyTorch built with: + - GCC 9.3 + - C++ Version: 201402 + - Intel(R) oneAPI Math Kernel Library Version 2021.4-Product Build 20210904 for Intel(R) 64 architecture applications + - Intel(R) MKL-DNN v2.6.0 (Git Hash 52b5f107dd9cf10910aaa19cb47f3abf9b349815) + - OpenMP 201511 (a.k.a. OpenMP 4.5) + - LAPACK is enabled (usually provided by MKL) + - NNPACK is enabled + - CPU capability usage: AVX2 + - CUDA Runtime 11.6 + - NVCC architecture flags: -gencode;arch=compute_37,code=sm_37;-gencode;arch=compute_50,code=sm_50;-gencode;arch=compute_60,code=sm_60;-gencode;arch=compute_61,code=sm_61;-gencode;arch=compute_70,code=sm_70;-gencode;arch=compute_75,code=sm_75;-gencode;arch=compute_80,code=sm_80;-gencode;arch=compute_86,code=sm_86;-gencode;arch=compute_37,code=compute_37 + - CuDNN 8.3.2 (built against CUDA 11.5) + - Magma 2.6.1 + - Build settings: BLAS_INFO=mkl, BUILD_TYPE=Release, CUDA_VERSION=11.6, CUDNN_VERSION=8.3.2, CXX_COMPILER=/opt/rh/devtoolset-9/root/usr/bin/c++, CXX_FLAGS= -fabi-version=11 -Wno-deprecated -fvisibility-inlines-hidden -DUSE_PTHREADPOOL -fopenmp -DNDEBUG -DUSE_KINETO -DUSE_FBGEMM -DUSE_QNNPACK -DUSE_PYTORCH_QNNPACK -DUSE_XNNPACK -DSYMBOLICATE_MOBILE_DEBUG_HANDLE -DEDGE_PROFILER_USE_KINETO -O2 -fPIC -Wno-narrowing -Wall -Wextra -Werror=return-type -Werror=non-virtual-dtor -Wno-missing-field-initializers -Wno-type-limits -Wno-array-bounds -Wno-unknown-pragmas -Wunused-local-typedefs -Wno-unused-parameter -Wno-unused-function -Wno-unused-result -Wno-strict-overflow -Wno-strict-aliasing -Wno-error=deprecated-declarations -Wno-stringop-overflow -Wno-psabi -Wno-error=pedantic -Wno-error=redundant-decls -Wno-error=old-style-cast -fdiagnostics-color=always -faligned-new -Wno-unused-but-set-variable -Wno-maybe-uninitialized -fno-math-errno -fno-trapping-math -Werror=format -Werror=cast-function-type -Wno-stringop-overflow, LAPACK_INFO=mkl, PERF_WITH_AVX=1, PERF_WITH_AVX2=1, PERF_WITH_AVX512=1, TORCH_VERSION=1.13.1, USE_CUDA=ON, USE_CUDNN=ON, USE_EXCEPTION_PTR=1, USE_GFLAGS=OFF, USE_GLOG=OFF, USE_MKL=ON, USE_MKLDNN=ON, USE_MPI=OFF, USE_NCCL=ON, USE_NNPACK=ON, USE_OPENMP=ON, USE_ROCM=OFF, + + TorchVision: 0.14.1 + OpenCV: 4.7.0 + MMEngine: 0.7.3 + +Runtime environment: + cudnn_benchmark: True + mp_cfg: {'mp_start_method': 'fork', 'opencv_num_threads': 0} + dist_cfg: {'backend': 'nccl'} + seed: None + deterministic: False + Distributed launcher: slurm + Distributed training: True + GPU number: 4 +------------------------------------------------------------ + +2023/06/01 21:57:47 - mmengine - INFO - Config: +optim_wrapper = dict( + optimizer=dict( + type='AdamW', + lr=0.0001, + weight_decay=0.05, + eps=1e-08, + betas=(0.9, 0.999), + _scope_='mmpretrain'), + paramwise_cfg=dict( + norm_decay_mult=0.0, + bias_decay_mult=0.0, + flat_decay_mult=0.0, + custom_keys=dict({ + '.absolute_pos_embed': dict(decay_mult=0.0), + '.relative_position_bias_table': dict(decay_mult=0.0) + })), + type='AmpOptimWrapper', + dtype='bfloat16', + clip_grad=dict(max_norm=5.0)) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=True, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=1024) +model = dict( + type='ImageClassifier', + backbone=dict( + type='SwinTransformer', arch='small', img_size=224, + drop_path_rate=0.3), + neck=dict(type='GlobalAveragePooling'), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=768, + init_cfg=None, + loss=dict( + type='LabelSmoothLoss', label_smooth_val=0.1, mode='original'), + cal_acc=False), + init_cfg=[ + dict(type='TruncNormal', layer='Linear', std=0.02, bias=0.0), + dict(type='Constant', layer='LayerNorm', val=1.0, bias=0.0) + ], + train_cfg=None) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stablediffusionV1-5R2-dpmsolver-25-5m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/swin_small_4xb256_fake5m_lr1e-4_10epoch' + +2023/06/01 21:57:51 - mmengine - INFO - Hooks will be executed in the following order: +before_run: +(VERY_HIGH ) RuntimeInfoHook +(BELOW_NORMAL) LoggerHook + -------------------- +before_train: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_train_epoch: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(NORMAL ) DistSamplerSeedHook + -------------------- +before_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook + -------------------- +after_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +after_train_epoch: +(NORMAL ) IterTimerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_val_epoch: +(NORMAL ) IterTimerHook + -------------------- +before_val_iter: +(NORMAL ) IterTimerHook + -------------------- +after_val_iter: +(NORMAL ) IterTimerHook +(NORMAL ) VisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_val_epoch: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +after_train: +(VERY_LOW ) CheckpointHook + -------------------- +before_test_epoch: +(NORMAL ) IterTimerHook + -------------------- +before_test_iter: +(NORMAL ) IterTimerHook + -------------------- +after_test_iter: +(NORMAL ) IterTimerHook +(NORMAL ) VisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_test_epoch: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_run: +(BELOW_NORMAL) LoggerHook + -------------------- +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.patch_embed.projection.bias:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.patch_embed.norm.weight:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.patch_embed.norm.bias:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.0.blocks.0.norm1.weight:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.0.blocks.0.norm1.bias:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.0.blocks.0.attn.w_msa.relative_position_bias_table:lr=0.0001 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.0.blocks.0.attn.w_msa.relative_position_bias_table:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.0.blocks.0.attn.w_msa.relative_position_bias_table:decay_mult=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.0.blocks.0.attn.w_msa.qkv.bias:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.0.blocks.0.attn.w_msa.proj.bias:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.0.blocks.0.norm2.weight:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.0.blocks.0.norm2.bias:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.0.blocks.0.ffn.layers.0.0.bias:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.0.blocks.0.ffn.layers.1.bias:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.0.blocks.1.norm1.weight:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.0.blocks.1.norm1.bias:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.0.blocks.1.attn.w_msa.relative_position_bias_table:lr=0.0001 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.0.blocks.1.attn.w_msa.relative_position_bias_table:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.0.blocks.1.attn.w_msa.relative_position_bias_table:decay_mult=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.0.blocks.1.attn.w_msa.qkv.bias:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.0.blocks.1.attn.w_msa.proj.bias:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.0.blocks.1.norm2.weight:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.0.blocks.1.norm2.bias:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.0.blocks.1.ffn.layers.0.0.bias:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.0.blocks.1.ffn.layers.1.bias:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.0.downsample.norm.weight:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.0.downsample.norm.bias:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.1.blocks.0.norm1.weight:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.1.blocks.0.norm1.bias:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.1.blocks.0.attn.w_msa.relative_position_bias_table:lr=0.0001 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.1.blocks.0.attn.w_msa.relative_position_bias_table:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.1.blocks.0.attn.w_msa.relative_position_bias_table:decay_mult=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.1.blocks.0.attn.w_msa.qkv.bias:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.1.blocks.0.attn.w_msa.proj.bias:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.1.blocks.0.norm2.weight:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.1.blocks.0.norm2.bias:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.1.blocks.0.ffn.layers.0.0.bias:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.1.blocks.0.ffn.layers.1.bias:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.1.blocks.1.norm1.weight:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.1.blocks.1.norm1.bias:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.1.blocks.1.attn.w_msa.relative_position_bias_table:lr=0.0001 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.1.blocks.1.attn.w_msa.relative_position_bias_table:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.1.blocks.1.attn.w_msa.relative_position_bias_table:decay_mult=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.1.blocks.1.attn.w_msa.qkv.bias:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.1.blocks.1.attn.w_msa.proj.bias:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.1.blocks.1.norm2.weight:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.1.blocks.1.norm2.bias:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.1.blocks.1.ffn.layers.0.0.bias:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.1.blocks.1.ffn.layers.1.bias:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.1.downsample.norm.weight:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.1.downsample.norm.bias:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.0.norm1.weight:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.0.norm1.bias:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.0.attn.w_msa.relative_position_bias_table:lr=0.0001 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.0.attn.w_msa.relative_position_bias_table:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.0.attn.w_msa.relative_position_bias_table:decay_mult=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.0.attn.w_msa.qkv.bias:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.0.attn.w_msa.proj.bias:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.0.norm2.weight:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.0.norm2.bias:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.0.ffn.layers.0.0.bias:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.0.ffn.layers.1.bias:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.1.norm1.weight:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.1.norm1.bias:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.1.attn.w_msa.relative_position_bias_table:lr=0.0001 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.1.attn.w_msa.relative_position_bias_table:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.1.attn.w_msa.relative_position_bias_table:decay_mult=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.1.attn.w_msa.qkv.bias:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.1.attn.w_msa.proj.bias:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.1.norm2.weight:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.1.norm2.bias:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.1.ffn.layers.0.0.bias:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.1.ffn.layers.1.bias:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.2.norm1.weight:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.2.norm1.bias:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.2.attn.w_msa.relative_position_bias_table:lr=0.0001 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.2.attn.w_msa.relative_position_bias_table:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.2.attn.w_msa.relative_position_bias_table:decay_mult=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.2.attn.w_msa.qkv.bias:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.2.attn.w_msa.proj.bias:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.2.norm2.weight:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.2.norm2.bias:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.2.ffn.layers.0.0.bias:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.2.ffn.layers.1.bias:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.3.norm1.weight:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.3.norm1.bias:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.3.attn.w_msa.relative_position_bias_table:lr=0.0001 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.3.attn.w_msa.relative_position_bias_table:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.3.attn.w_msa.relative_position_bias_table:decay_mult=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.3.attn.w_msa.qkv.bias:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.3.attn.w_msa.proj.bias:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.3.norm2.weight:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.3.norm2.bias:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.3.ffn.layers.0.0.bias:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.3.ffn.layers.1.bias:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.4.norm1.weight:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.4.norm1.bias:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.4.attn.w_msa.relative_position_bias_table:lr=0.0001 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.4.attn.w_msa.relative_position_bias_table:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.4.attn.w_msa.relative_position_bias_table:decay_mult=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.4.attn.w_msa.qkv.bias:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.4.attn.w_msa.proj.bias:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.4.norm2.weight:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.4.norm2.bias:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.4.ffn.layers.0.0.bias:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.4.ffn.layers.1.bias:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.5.norm1.weight:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.5.norm1.bias:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.5.attn.w_msa.relative_position_bias_table:lr=0.0001 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.5.attn.w_msa.relative_position_bias_table:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.5.attn.w_msa.relative_position_bias_table:decay_mult=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.5.attn.w_msa.qkv.bias:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.5.attn.w_msa.proj.bias:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.5.norm2.weight:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.5.norm2.bias:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.5.ffn.layers.0.0.bias:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.5.ffn.layers.1.bias:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.6.norm1.weight:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.6.norm1.bias:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.6.attn.w_msa.relative_position_bias_table:lr=0.0001 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.6.attn.w_msa.relative_position_bias_table:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.6.attn.w_msa.relative_position_bias_table:decay_mult=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.6.attn.w_msa.qkv.bias:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.6.attn.w_msa.proj.bias:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.6.norm2.weight:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.6.norm2.bias:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.6.ffn.layers.0.0.bias:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.6.ffn.layers.1.bias:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.7.norm1.weight:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.7.norm1.bias:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.7.attn.w_msa.relative_position_bias_table:lr=0.0001 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.7.attn.w_msa.relative_position_bias_table:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.7.attn.w_msa.relative_position_bias_table:decay_mult=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.7.attn.w_msa.qkv.bias:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.7.attn.w_msa.proj.bias:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.7.norm2.weight:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.7.norm2.bias:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.7.ffn.layers.0.0.bias:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.7.ffn.layers.1.bias:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.8.norm1.weight:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.8.norm1.bias:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.8.attn.w_msa.relative_position_bias_table:lr=0.0001 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.8.attn.w_msa.relative_position_bias_table:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.8.attn.w_msa.relative_position_bias_table:decay_mult=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.8.attn.w_msa.qkv.bias:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.8.attn.w_msa.proj.bias:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.8.norm2.weight:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.8.norm2.bias:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.8.ffn.layers.0.0.bias:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.8.ffn.layers.1.bias:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.9.norm1.weight:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.9.norm1.bias:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.9.attn.w_msa.relative_position_bias_table:lr=0.0001 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.9.attn.w_msa.relative_position_bias_table:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.9.attn.w_msa.relative_position_bias_table:decay_mult=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.9.attn.w_msa.qkv.bias:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.9.attn.w_msa.proj.bias:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.9.norm2.weight:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.9.norm2.bias:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.9.ffn.layers.0.0.bias:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.9.ffn.layers.1.bias:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.10.norm1.weight:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.10.norm1.bias:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.10.attn.w_msa.relative_position_bias_table:lr=0.0001 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.10.attn.w_msa.relative_position_bias_table:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.10.attn.w_msa.relative_position_bias_table:decay_mult=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.10.attn.w_msa.qkv.bias:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.10.attn.w_msa.proj.bias:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.10.norm2.weight:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.10.norm2.bias:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.10.ffn.layers.0.0.bias:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.10.ffn.layers.1.bias:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.11.norm1.weight:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.11.norm1.bias:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.11.attn.w_msa.relative_position_bias_table:lr=0.0001 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.11.attn.w_msa.relative_position_bias_table:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.11.attn.w_msa.relative_position_bias_table:decay_mult=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.11.attn.w_msa.qkv.bias:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.11.attn.w_msa.proj.bias:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.11.norm2.weight:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.11.norm2.bias:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.11.ffn.layers.0.0.bias:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.11.ffn.layers.1.bias:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.12.norm1.weight:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.12.norm1.bias:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.12.attn.w_msa.relative_position_bias_table:lr=0.0001 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.12.attn.w_msa.relative_position_bias_table:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.12.attn.w_msa.relative_position_bias_table:decay_mult=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.12.attn.w_msa.qkv.bias:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.12.attn.w_msa.proj.bias:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.12.norm2.weight:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.12.norm2.bias:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.12.ffn.layers.0.0.bias:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.12.ffn.layers.1.bias:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.13.norm1.weight:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.13.norm1.bias:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.13.attn.w_msa.relative_position_bias_table:lr=0.0001 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.13.attn.w_msa.relative_position_bias_table:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.13.attn.w_msa.relative_position_bias_table:decay_mult=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.13.attn.w_msa.qkv.bias:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.13.attn.w_msa.proj.bias:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.13.norm2.weight:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.13.norm2.bias:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.13.ffn.layers.0.0.bias:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.13.ffn.layers.1.bias:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.14.norm1.weight:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.14.norm1.bias:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.14.attn.w_msa.relative_position_bias_table:lr=0.0001 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.14.attn.w_msa.relative_position_bias_table:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.14.attn.w_msa.relative_position_bias_table:decay_mult=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.14.attn.w_msa.qkv.bias:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.14.attn.w_msa.proj.bias:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.14.norm2.weight:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.14.norm2.bias:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.14.ffn.layers.0.0.bias:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.14.ffn.layers.1.bias:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.15.norm1.weight:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.15.norm1.bias:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.15.attn.w_msa.relative_position_bias_table:lr=0.0001 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.15.attn.w_msa.relative_position_bias_table:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.15.attn.w_msa.relative_position_bias_table:decay_mult=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.15.attn.w_msa.qkv.bias:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.15.attn.w_msa.proj.bias:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.15.norm2.weight:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.15.norm2.bias:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.15.ffn.layers.0.0.bias:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.15.ffn.layers.1.bias:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.16.norm1.weight:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.16.norm1.bias:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.16.attn.w_msa.relative_position_bias_table:lr=0.0001 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.16.attn.w_msa.relative_position_bias_table:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.16.attn.w_msa.relative_position_bias_table:decay_mult=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.16.attn.w_msa.qkv.bias:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.16.attn.w_msa.proj.bias:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.16.norm2.weight:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.16.norm2.bias:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.16.ffn.layers.0.0.bias:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.16.ffn.layers.1.bias:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.17.norm1.weight:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.17.norm1.bias:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.17.attn.w_msa.relative_position_bias_table:lr=0.0001 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.17.attn.w_msa.relative_position_bias_table:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.17.attn.w_msa.relative_position_bias_table:decay_mult=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.17.attn.w_msa.qkv.bias:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.17.attn.w_msa.proj.bias:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.17.norm2.weight:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.17.norm2.bias:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.17.ffn.layers.0.0.bias:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.17.ffn.layers.1.bias:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.downsample.norm.weight:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.2.downsample.norm.bias:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.3.blocks.0.norm1.weight:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.3.blocks.0.norm1.bias:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.3.blocks.0.attn.w_msa.relative_position_bias_table:lr=0.0001 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.3.blocks.0.attn.w_msa.relative_position_bias_table:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.3.blocks.0.attn.w_msa.relative_position_bias_table:decay_mult=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.3.blocks.0.attn.w_msa.qkv.bias:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.3.blocks.0.attn.w_msa.proj.bias:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.3.blocks.0.norm2.weight:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.3.blocks.0.norm2.bias:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.3.blocks.0.ffn.layers.0.0.bias:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.3.blocks.0.ffn.layers.1.bias:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.3.blocks.1.norm1.weight:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.3.blocks.1.norm1.bias:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.3.blocks.1.attn.w_msa.relative_position_bias_table:lr=0.0001 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.3.blocks.1.attn.w_msa.relative_position_bias_table:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.3.blocks.1.attn.w_msa.relative_position_bias_table:decay_mult=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.3.blocks.1.attn.w_msa.qkv.bias:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.3.blocks.1.attn.w_msa.proj.bias:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.3.blocks.1.norm2.weight:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.3.blocks.1.norm2.bias:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.3.blocks.1.ffn.layers.0.0.bias:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.stages.3.blocks.1.ffn.layers.1.bias:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.norm3.weight:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- backbone.norm3.bias:weight_decay=0.0 +2023/06/01 21:58:28 - mmengine - INFO - paramwise_options -- head.fc.bias:weight_decay=0.0 +Name of parameter - Initialization information + +backbone.patch_embed.projection.weight - torch.Size([96, 3, 4, 4]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.patch_embed.projection.bias - torch.Size([96]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.patch_embed.norm.weight - torch.Size([96]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.patch_embed.norm.bias - torch.Size([96]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.0.blocks.0.norm1.weight - torch.Size([96]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.0.blocks.0.norm1.bias - torch.Size([96]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.0.blocks.0.attn.w_msa.relative_position_bias_table - torch.Size([169, 3]): +Initialized by user-defined `init_weights` in WindowMSA + +backbone.stages.0.blocks.0.attn.w_msa.qkv.weight - torch.Size([288, 96]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.blocks.0.attn.w_msa.qkv.bias - torch.Size([288]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.blocks.0.attn.w_msa.proj.weight - torch.Size([96, 96]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.blocks.0.attn.w_msa.proj.bias - torch.Size([96]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.blocks.0.norm2.weight - torch.Size([96]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.0.blocks.0.norm2.bias - torch.Size([96]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.0.blocks.0.ffn.layers.0.0.weight - torch.Size([384, 96]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.blocks.0.ffn.layers.0.0.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.blocks.0.ffn.layers.1.weight - torch.Size([96, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.blocks.0.ffn.layers.1.bias - torch.Size([96]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.blocks.1.norm1.weight - torch.Size([96]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.0.blocks.1.norm1.bias - torch.Size([96]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.0.blocks.1.attn.w_msa.relative_position_bias_table - torch.Size([169, 3]): +Initialized by user-defined `init_weights` in WindowMSA + +backbone.stages.0.blocks.1.attn.w_msa.qkv.weight - torch.Size([288, 96]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.blocks.1.attn.w_msa.qkv.bias - torch.Size([288]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.blocks.1.attn.w_msa.proj.weight - torch.Size([96, 96]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.blocks.1.attn.w_msa.proj.bias - torch.Size([96]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.blocks.1.norm2.weight - torch.Size([96]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.0.blocks.1.norm2.bias - torch.Size([96]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.0.blocks.1.ffn.layers.0.0.weight - torch.Size([384, 96]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.blocks.1.ffn.layers.0.0.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.blocks.1.ffn.layers.1.weight - torch.Size([96, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.blocks.1.ffn.layers.1.bias - torch.Size([96]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.downsample.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.0.downsample.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.0.downsample.reduction.weight - torch.Size([192, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.blocks.0.norm1.weight - torch.Size([192]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.1.blocks.0.norm1.bias - torch.Size([192]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.1.blocks.0.attn.w_msa.relative_position_bias_table - torch.Size([169, 6]): +Initialized by user-defined `init_weights` in WindowMSA + +backbone.stages.1.blocks.0.attn.w_msa.qkv.weight - torch.Size([576, 192]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.blocks.0.attn.w_msa.qkv.bias - torch.Size([576]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.blocks.0.attn.w_msa.proj.weight - torch.Size([192, 192]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.blocks.0.attn.w_msa.proj.bias - torch.Size([192]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.blocks.0.norm2.weight - torch.Size([192]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.1.blocks.0.norm2.bias - torch.Size([192]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.1.blocks.0.ffn.layers.0.0.weight - torch.Size([768, 192]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.blocks.0.ffn.layers.0.0.bias - torch.Size([768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.blocks.0.ffn.layers.1.weight - torch.Size([192, 768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.blocks.0.ffn.layers.1.bias - torch.Size([192]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.blocks.1.norm1.weight - torch.Size([192]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.1.blocks.1.norm1.bias - torch.Size([192]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.1.blocks.1.attn.w_msa.relative_position_bias_table - torch.Size([169, 6]): +Initialized by user-defined `init_weights` in WindowMSA + +backbone.stages.1.blocks.1.attn.w_msa.qkv.weight - torch.Size([576, 192]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.blocks.1.attn.w_msa.qkv.bias - torch.Size([576]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.blocks.1.attn.w_msa.proj.weight - torch.Size([192, 192]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.blocks.1.attn.w_msa.proj.bias - torch.Size([192]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.blocks.1.norm2.weight - torch.Size([192]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.1.blocks.1.norm2.bias - torch.Size([192]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.1.blocks.1.ffn.layers.0.0.weight - torch.Size([768, 192]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.blocks.1.ffn.layers.0.0.bias - torch.Size([768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.blocks.1.ffn.layers.1.weight - torch.Size([192, 768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.blocks.1.ffn.layers.1.bias - torch.Size([192]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.downsample.norm.weight - torch.Size([768]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.1.downsample.norm.bias - torch.Size([768]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.1.downsample.reduction.weight - torch.Size([384, 768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.0.norm1.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.0.norm1.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.0.attn.w_msa.relative_position_bias_table - torch.Size([169, 12]): +Initialized by user-defined `init_weights` in WindowMSA + +backbone.stages.2.blocks.0.attn.w_msa.qkv.weight - torch.Size([1152, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.0.attn.w_msa.qkv.bias - torch.Size([1152]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.0.attn.w_msa.proj.weight - torch.Size([384, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.0.attn.w_msa.proj.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.0.norm2.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.0.norm2.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.0.ffn.layers.0.0.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.0.ffn.layers.0.0.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.0.ffn.layers.1.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.0.ffn.layers.1.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.1.norm1.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.1.norm1.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.1.attn.w_msa.relative_position_bias_table - torch.Size([169, 12]): +Initialized by user-defined `init_weights` in WindowMSA + +backbone.stages.2.blocks.1.attn.w_msa.qkv.weight - torch.Size([1152, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.1.attn.w_msa.qkv.bias - torch.Size([1152]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.1.attn.w_msa.proj.weight - torch.Size([384, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.1.attn.w_msa.proj.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.1.norm2.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.1.norm2.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.1.ffn.layers.0.0.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.1.ffn.layers.0.0.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.1.ffn.layers.1.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.1.ffn.layers.1.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.2.norm1.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.2.norm1.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.2.attn.w_msa.relative_position_bias_table - torch.Size([169, 12]): +Initialized by user-defined `init_weights` in WindowMSA + +backbone.stages.2.blocks.2.attn.w_msa.qkv.weight - torch.Size([1152, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.2.attn.w_msa.qkv.bias - torch.Size([1152]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.2.attn.w_msa.proj.weight - torch.Size([384, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.2.attn.w_msa.proj.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.2.norm2.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.2.norm2.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.2.ffn.layers.0.0.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.2.ffn.layers.0.0.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.2.ffn.layers.1.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.2.ffn.layers.1.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.3.norm1.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.3.norm1.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.3.attn.w_msa.relative_position_bias_table - torch.Size([169, 12]): +Initialized by user-defined `init_weights` in WindowMSA + +backbone.stages.2.blocks.3.attn.w_msa.qkv.weight - torch.Size([1152, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.3.attn.w_msa.qkv.bias - torch.Size([1152]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.3.attn.w_msa.proj.weight - torch.Size([384, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.3.attn.w_msa.proj.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.3.norm2.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.3.norm2.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.3.ffn.layers.0.0.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.3.ffn.layers.0.0.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.3.ffn.layers.1.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.3.ffn.layers.1.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.4.norm1.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.4.norm1.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.4.attn.w_msa.relative_position_bias_table - torch.Size([169, 12]): +Initialized by user-defined `init_weights` in WindowMSA + +backbone.stages.2.blocks.4.attn.w_msa.qkv.weight - torch.Size([1152, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.4.attn.w_msa.qkv.bias - torch.Size([1152]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.4.attn.w_msa.proj.weight - torch.Size([384, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.4.attn.w_msa.proj.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.4.norm2.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.4.norm2.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.4.ffn.layers.0.0.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.4.ffn.layers.0.0.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.4.ffn.layers.1.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.4.ffn.layers.1.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.5.norm1.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.5.norm1.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.5.attn.w_msa.relative_position_bias_table - torch.Size([169, 12]): +Initialized by user-defined `init_weights` in WindowMSA + +backbone.stages.2.blocks.5.attn.w_msa.qkv.weight - torch.Size([1152, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.5.attn.w_msa.qkv.bias - torch.Size([1152]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.5.attn.w_msa.proj.weight - torch.Size([384, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.5.attn.w_msa.proj.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.5.norm2.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.5.norm2.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.5.ffn.layers.0.0.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.5.ffn.layers.0.0.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.5.ffn.layers.1.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.5.ffn.layers.1.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.6.norm1.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.6.norm1.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.6.attn.w_msa.relative_position_bias_table - torch.Size([169, 12]): +Initialized by user-defined `init_weights` in WindowMSA + +backbone.stages.2.blocks.6.attn.w_msa.qkv.weight - torch.Size([1152, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.6.attn.w_msa.qkv.bias - torch.Size([1152]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.6.attn.w_msa.proj.weight - torch.Size([384, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.6.attn.w_msa.proj.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.6.norm2.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.6.norm2.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.6.ffn.layers.0.0.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.6.ffn.layers.0.0.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.6.ffn.layers.1.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.6.ffn.layers.1.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.7.norm1.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.7.norm1.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.7.attn.w_msa.relative_position_bias_table - torch.Size([169, 12]): +Initialized by user-defined `init_weights` in WindowMSA + +backbone.stages.2.blocks.7.attn.w_msa.qkv.weight - torch.Size([1152, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.7.attn.w_msa.qkv.bias - torch.Size([1152]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.7.attn.w_msa.proj.weight - torch.Size([384, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.7.attn.w_msa.proj.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.7.norm2.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.7.norm2.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.7.ffn.layers.0.0.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.7.ffn.layers.0.0.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.7.ffn.layers.1.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.7.ffn.layers.1.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.8.norm1.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.8.norm1.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.8.attn.w_msa.relative_position_bias_table - torch.Size([169, 12]): +Initialized by user-defined `init_weights` in WindowMSA + +backbone.stages.2.blocks.8.attn.w_msa.qkv.weight - torch.Size([1152, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.8.attn.w_msa.qkv.bias - torch.Size([1152]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.8.attn.w_msa.proj.weight - torch.Size([384, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.8.attn.w_msa.proj.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.8.norm2.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.8.norm2.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.8.ffn.layers.0.0.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.8.ffn.layers.0.0.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.8.ffn.layers.1.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.8.ffn.layers.1.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.9.norm1.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.9.norm1.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.9.attn.w_msa.relative_position_bias_table - torch.Size([169, 12]): +Initialized by user-defined `init_weights` in WindowMSA + +backbone.stages.2.blocks.9.attn.w_msa.qkv.weight - torch.Size([1152, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.9.attn.w_msa.qkv.bias - torch.Size([1152]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.9.attn.w_msa.proj.weight - torch.Size([384, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.9.attn.w_msa.proj.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.9.norm2.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.9.norm2.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.9.ffn.layers.0.0.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.9.ffn.layers.0.0.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.9.ffn.layers.1.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.9.ffn.layers.1.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.10.norm1.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.10.norm1.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.10.attn.w_msa.relative_position_bias_table - torch.Size([169, 12]): +Initialized by user-defined `init_weights` in WindowMSA + +backbone.stages.2.blocks.10.attn.w_msa.qkv.weight - torch.Size([1152, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.10.attn.w_msa.qkv.bias - torch.Size([1152]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.10.attn.w_msa.proj.weight - torch.Size([384, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.10.attn.w_msa.proj.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.10.norm2.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.10.norm2.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.10.ffn.layers.0.0.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.10.ffn.layers.0.0.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.10.ffn.layers.1.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.10.ffn.layers.1.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.11.norm1.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.11.norm1.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.11.attn.w_msa.relative_position_bias_table - torch.Size([169, 12]): +Initialized by user-defined `init_weights` in WindowMSA + +backbone.stages.2.blocks.11.attn.w_msa.qkv.weight - torch.Size([1152, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.11.attn.w_msa.qkv.bias - torch.Size([1152]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.11.attn.w_msa.proj.weight - torch.Size([384, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.11.attn.w_msa.proj.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.11.norm2.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.11.norm2.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.11.ffn.layers.0.0.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.11.ffn.layers.0.0.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.11.ffn.layers.1.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.11.ffn.layers.1.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.12.norm1.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.12.norm1.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.12.attn.w_msa.relative_position_bias_table - torch.Size([169, 12]): +Initialized by user-defined `init_weights` in WindowMSA + +backbone.stages.2.blocks.12.attn.w_msa.qkv.weight - torch.Size([1152, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.12.attn.w_msa.qkv.bias - torch.Size([1152]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.12.attn.w_msa.proj.weight - torch.Size([384, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.12.attn.w_msa.proj.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.12.norm2.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.12.norm2.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.12.ffn.layers.0.0.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.12.ffn.layers.0.0.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.12.ffn.layers.1.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.12.ffn.layers.1.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.13.norm1.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.13.norm1.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.13.attn.w_msa.relative_position_bias_table - torch.Size([169, 12]): +Initialized by user-defined `init_weights` in WindowMSA + +backbone.stages.2.blocks.13.attn.w_msa.qkv.weight - torch.Size([1152, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.13.attn.w_msa.qkv.bias - torch.Size([1152]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.13.attn.w_msa.proj.weight - torch.Size([384, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.13.attn.w_msa.proj.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.13.norm2.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.13.norm2.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.13.ffn.layers.0.0.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.13.ffn.layers.0.0.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.13.ffn.layers.1.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.13.ffn.layers.1.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.14.norm1.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.14.norm1.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.14.attn.w_msa.relative_position_bias_table - torch.Size([169, 12]): +Initialized by user-defined `init_weights` in WindowMSA + +backbone.stages.2.blocks.14.attn.w_msa.qkv.weight - torch.Size([1152, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.14.attn.w_msa.qkv.bias - torch.Size([1152]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.14.attn.w_msa.proj.weight - torch.Size([384, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.14.attn.w_msa.proj.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.14.norm2.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.14.norm2.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.14.ffn.layers.0.0.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.14.ffn.layers.0.0.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.14.ffn.layers.1.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.14.ffn.layers.1.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.15.norm1.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.15.norm1.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.15.attn.w_msa.relative_position_bias_table - torch.Size([169, 12]): +Initialized by user-defined `init_weights` in WindowMSA + +backbone.stages.2.blocks.15.attn.w_msa.qkv.weight - torch.Size([1152, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.15.attn.w_msa.qkv.bias - torch.Size([1152]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.15.attn.w_msa.proj.weight - torch.Size([384, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.15.attn.w_msa.proj.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.15.norm2.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.15.norm2.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.15.ffn.layers.0.0.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.15.ffn.layers.0.0.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.15.ffn.layers.1.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.15.ffn.layers.1.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.16.norm1.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.16.norm1.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.16.attn.w_msa.relative_position_bias_table - torch.Size([169, 12]): +Initialized by user-defined `init_weights` in WindowMSA + +backbone.stages.2.blocks.16.attn.w_msa.qkv.weight - torch.Size([1152, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.16.attn.w_msa.qkv.bias - torch.Size([1152]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.16.attn.w_msa.proj.weight - torch.Size([384, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.16.attn.w_msa.proj.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.16.norm2.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.16.norm2.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.16.ffn.layers.0.0.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.16.ffn.layers.0.0.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.16.ffn.layers.1.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.16.ffn.layers.1.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.17.norm1.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.17.norm1.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.17.attn.w_msa.relative_position_bias_table - torch.Size([169, 12]): +Initialized by user-defined `init_weights` in WindowMSA + +backbone.stages.2.blocks.17.attn.w_msa.qkv.weight - torch.Size([1152, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.17.attn.w_msa.qkv.bias - torch.Size([1152]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.17.attn.w_msa.proj.weight - torch.Size([384, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.17.attn.w_msa.proj.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.17.norm2.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.17.norm2.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.17.ffn.layers.0.0.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.17.ffn.layers.0.0.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.17.ffn.layers.1.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.17.ffn.layers.1.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.downsample.norm.weight - torch.Size([1536]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.downsample.norm.bias - torch.Size([1536]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.downsample.reduction.weight - torch.Size([768, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.blocks.0.norm1.weight - torch.Size([768]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.3.blocks.0.norm1.bias - torch.Size([768]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.3.blocks.0.attn.w_msa.relative_position_bias_table - torch.Size([169, 24]): +Initialized by user-defined `init_weights` in WindowMSA + +backbone.stages.3.blocks.0.attn.w_msa.qkv.weight - torch.Size([2304, 768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.blocks.0.attn.w_msa.qkv.bias - torch.Size([2304]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.blocks.0.attn.w_msa.proj.weight - torch.Size([768, 768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.blocks.0.attn.w_msa.proj.bias - torch.Size([768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.blocks.0.norm2.weight - torch.Size([768]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.3.blocks.0.norm2.bias - torch.Size([768]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.3.blocks.0.ffn.layers.0.0.weight - torch.Size([3072, 768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.blocks.0.ffn.layers.0.0.bias - torch.Size([3072]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.blocks.0.ffn.layers.1.weight - torch.Size([768, 3072]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.blocks.0.ffn.layers.1.bias - torch.Size([768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.blocks.1.norm1.weight - torch.Size([768]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.3.blocks.1.norm1.bias - torch.Size([768]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.3.blocks.1.attn.w_msa.relative_position_bias_table - torch.Size([169, 24]): +Initialized by user-defined `init_weights` in WindowMSA + +backbone.stages.3.blocks.1.attn.w_msa.qkv.weight - torch.Size([2304, 768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.blocks.1.attn.w_msa.qkv.bias - torch.Size([2304]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.blocks.1.attn.w_msa.proj.weight - torch.Size([768, 768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.blocks.1.attn.w_msa.proj.bias - torch.Size([768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.blocks.1.norm2.weight - torch.Size([768]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.3.blocks.1.norm2.bias - torch.Size([768]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.3.blocks.1.ffn.layers.0.0.weight - torch.Size([3072, 768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.blocks.1.ffn.layers.0.0.bias - torch.Size([3072]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.blocks.1.ffn.layers.1.weight - torch.Size([768, 3072]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.blocks.1.ffn.layers.1.bias - torch.Size([768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.norm3.weight - torch.Size([768]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.norm3.bias - torch.Size([768]): +The value is the same before and after calling `init_weights` of ImageClassifier + +head.fc.weight - torch.Size([2, 768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +head.fc.bias - torch.Size([2]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 +2023/06/01 21:58:29 - mmengine - WARNING - "FileClient" will be deprecated in future. Please use io functions in https://mmengine.readthedocs.io/en/latest/api/fileio.html#file-io +2023/06/01 21:58:29 - mmengine - WARNING - "HardDiskBackend" is the alias of "LocalBackend" and the former will be deprecated in future. +2023/06/01 21:58:29 - mmengine - INFO - Checkpoints will be saved to /mnt/petrelfs/luzeyu/workspace/fakebench/mmpretrain/workdir/swin_small_4xb256_fake5m_lr1e-4_10epoch. +2023/06/01 22:00:03 - mmengine - INFO - Epoch(train) [1][ 100/5758] lr: 1.0000e-04 eta: 15:00:45 time: 0.8802 data_time: 0.2837 memory: 28798 grad_norm: 1.1496 loss: 0.6059 +2023/06/01 22:01:31 - mmengine - INFO - Epoch(train) [1][ 200/5758] lr: 1.0000e-04 eta: 14:28:49 time: 1.0053 data_time: 0.4572 memory: 28798 grad_norm: 1.5306 loss: 0.5631 +2023/06/01 22:02:57 - mmengine - INFO - Epoch(train) [1][ 300/5758] lr: 1.0000e-04 eta: 14:11:10 time: 0.9259 data_time: 0.2215 memory: 28798 grad_norm: 2.0284 loss: 0.5297 +2023/06/01 22:04:22 - mmengine - INFO - Epoch(train) [1][ 400/5758] lr: 1.0000e-04 eta: 14:00:06 time: 0.9341 data_time: 0.2911 memory: 28798 grad_norm: 2.5006 loss: 0.5156 +2023/06/01 22:05:46 - mmengine - INFO - Epoch(train) [1][ 500/5758] lr: 1.0000e-04 eta: 13:51:15 time: 0.8412 data_time: 0.1230 memory: 28798 grad_norm: 4.3186 loss: 0.5011 +2023/06/01 22:07:13 - mmengine - INFO - Epoch(train) [1][ 600/5758] lr: 1.0000e-04 eta: 13:48:42 time: 0.9766 data_time: 0.1355 memory: 28798 grad_norm: 4.5865 loss: 0.4731 +2023/06/01 22:08:37 - mmengine - INFO - Epoch(train) [1][ 700/5758] lr: 1.0000e-04 eta: 13:42:41 time: 0.8893 data_time: 0.1647 memory: 28798 grad_norm: 3.8184 loss: 0.4545 +2023/06/01 22:09:59 - mmengine - INFO - Epoch(train) [1][ 800/5758] lr: 1.0000e-04 eta: 13:36:00 time: 0.8825 data_time: 0.1264 memory: 28798 grad_norm: 6.1841 loss: 0.4635 +2023/06/01 22:11:27 - mmengine - INFO - Epoch(train) [1][ 900/5758] lr: 1.0000e-04 eta: 13:36:07 time: 0.7175 data_time: 0.0019 memory: 28798 grad_norm: 3.7891 loss: 0.4483 +2023/06/01 22:12:58 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_lr1e-4_10epoch_20230601_215742 +2023/06/01 22:12:58 - mmengine - INFO - Epoch(train) [1][1000/5758] lr: 1.0000e-04 eta: 13:39:05 time: 0.8958 data_time: 0.0022 memory: 28798 grad_norm: 6.0323 loss: 0.4484 +2023/06/01 22:14:22 - mmengine - INFO - Epoch(train) [1][1100/5758] lr: 1.0000e-04 eta: 13:35:13 time: 0.7581 data_time: 0.0016 memory: 28798 grad_norm: 3.7644 loss: 0.4270 +2023/06/01 22:15:50 - mmengine - INFO - Epoch(train) [1][1200/5758] lr: 1.0000e-04 eta: 13:35:09 time: 0.7184 data_time: 0.0357 memory: 28798 grad_norm: 8.6618 loss: 0.4228 +2023/06/01 22:17:16 - mmengine - INFO - Epoch(train) [1][1300/5758] lr: 1.0000e-04 eta: 13:33:05 time: 0.7568 data_time: 0.1960 memory: 28798 grad_norm: 12.0725 loss: 0.4322 +2023/06/01 22:18:40 - mmengine - INFO - Epoch(train) [1][1400/5758] lr: 1.0000e-04 eta: 13:30:04 time: 0.9501 data_time: 0.1318 memory: 28798 grad_norm: 9.7848 loss: 0.4166 +2023/06/01 22:20:07 - mmengine - INFO - Epoch(train) [1][1500/5758] lr: 1.0000e-04 eta: 13:28:48 time: 0.7976 data_time: 0.0021 memory: 28798 grad_norm: 8.5640 loss: 0.4024 +2023/06/01 22:21:33 - mmengine - INFO - Epoch(train) [1][1600/5758] lr: 1.0000e-04 eta: 13:27:17 time: 0.8243 data_time: 0.0023 memory: 28798 grad_norm: 10.9330 loss: 0.4379 +2023/06/01 22:23:04 - mmengine - INFO - Epoch(train) [1][1700/5758] lr: 1.0000e-04 eta: 13:28:13 time: 0.7808 data_time: 0.1610 memory: 28798 grad_norm: 14.2624 loss: 0.4118 +2023/06/01 22:24:32 - mmengine - INFO - Epoch(train) [1][1800/5758] lr: 1.0000e-04 eta: 13:27:08 time: 0.9717 data_time: 0.3968 memory: 28798 grad_norm: 6.8426 loss: 0.4044 +2023/06/01 22:26:03 - mmengine - INFO - Epoch(train) [1][1900/5758] lr: 1.0000e-04 eta: 13:27:35 time: 0.8954 data_time: 0.2469 memory: 28798 grad_norm: 6.0332 loss: 0.3813 +2023/06/01 22:27:31 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_lr1e-4_10epoch_20230601_215742 +2023/06/01 22:27:31 - mmengine - INFO - Epoch(train) [1][2000/5758] lr: 1.0000e-04 eta: 13:26:53 time: 0.8790 data_time: 0.3589 memory: 28798 grad_norm: 11.1640 loss: 0.3767 +2023/06/01 22:29:01 - mmengine - INFO - Epoch(train) [1][2100/5758] lr: 1.0000e-04 eta: 13:26:36 time: 0.8404 data_time: 0.3277 memory: 28798 grad_norm: 6.9214 loss: 0.3695 +2023/06/01 22:30:27 - mmengine - INFO - Epoch(train) [1][2200/5758] lr: 1.0000e-04 eta: 13:24:42 time: 0.9089 data_time: 0.3895 memory: 28798 grad_norm: 16.0706 loss: 0.3762 +2023/06/01 22:31:56 - mmengine - INFO - Epoch(train) [1][2300/5758] lr: 1.0000e-04 eta: 13:23:53 time: 0.8222 data_time: 0.3172 memory: 28798 grad_norm: 10.9651 loss: 0.3938 +2023/06/01 22:33:23 - mmengine - INFO - Epoch(train) [1][2400/5758] lr: 1.0000e-04 eta: 13:22:20 time: 0.9547 data_time: 0.4321 memory: 28798 grad_norm: 11.4193 loss: 0.3853 +2023/06/01 22:34:52 - mmengine - INFO - Epoch(train) [1][2500/5758] lr: 1.0000e-04 eta: 13:21:34 time: 0.9849 data_time: 0.4447 memory: 28798 grad_norm: 20.4749 loss: 0.3967 +2023/06/01 22:36:21 - mmengine - INFO - Epoch(train) [1][2600/5758] lr: 1.0000e-04 eta: 13:20:32 time: 0.8514 data_time: 0.3377 memory: 28798 grad_norm: 8.1619 loss: 0.3465 +2023/06/01 22:37:51 - mmengine - INFO - Epoch(train) [1][2700/5758] lr: 1.0000e-04 eta: 13:20:14 time: 1.0519 data_time: 0.4989 memory: 28798 grad_norm: 4.8205 loss: 0.3565 +2023/06/01 22:39:19 - mmengine - INFO - Epoch(train) [1][2800/5758] lr: 1.0000e-04 eta: 13:18:55 time: 0.8059 data_time: 0.2992 memory: 28798 grad_norm: 17.9652 loss: 0.3608 +2023/06/01 22:40:45 - mmengine - INFO - Epoch(train) [1][2900/5758] lr: 1.0000e-04 eta: 13:16:57 time: 0.8675 data_time: 0.3399 memory: 28798 grad_norm: 14.2151 loss: 0.3633 +2023/06/01 22:42:17 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_lr1e-4_10epoch_20230601_215742 +2023/06/01 22:42:17 - mmengine - INFO - Epoch(train) [1][3000/5758] lr: 1.0000e-04 eta: 13:16:43 time: 0.9157 data_time: 0.3960 memory: 28798 grad_norm: 7.2075 loss: 0.3426 +2023/06/01 22:43:49 - mmengine - INFO - Epoch(train) [1][3100/5758] lr: 1.0000e-04 eta: 13:16:31 time: 0.9411 data_time: 0.4100 memory: 28798 grad_norm: 11.3211 loss: 0.3522 +2023/06/01 22:45:19 - mmengine - INFO - Epoch(train) [1][3200/5758] lr: 1.0000e-04 eta: 13:15:53 time: 0.7645 data_time: 0.2562 memory: 28798 grad_norm: 14.3808 loss: 0.3566 +2023/06/01 22:46:51 - mmengine - INFO - Epoch(train) [1][3300/5758] lr: 1.0000e-04 eta: 13:15:33 time: 0.8485 data_time: 0.3422 memory: 28798 grad_norm: 13.3459 loss: 0.3288 +2023/06/01 22:48:21 - mmengine - INFO - Epoch(train) [1][3400/5758] lr: 1.0000e-04 eta: 13:14:39 time: 0.8322 data_time: 0.3237 memory: 28798 grad_norm: 8.8210 loss: 0.3175 +2023/06/01 22:49:51 - mmengine - INFO - Epoch(train) [1][3500/5758] lr: 1.0000e-04 eta: 13:13:43 time: 1.0275 data_time: 0.4927 memory: 28798 grad_norm: 6.5424 loss: 0.3355 +2023/06/01 22:51:21 - mmengine - INFO - Epoch(train) [1][3600/5758] lr: 1.0000e-04 eta: 13:12:47 time: 0.8759 data_time: 0.2771 memory: 28798 grad_norm: 11.5644 loss: 0.3205 +2023/06/01 22:52:53 - mmengine - INFO - Epoch(train) [1][3700/5758] lr: 1.0000e-04 eta: 13:12:16 time: 0.8442 data_time: 0.0951 memory: 28798 grad_norm: 5.5171 loss: 0.3161 +2023/06/01 22:54:27 - mmengine - INFO - Epoch(train) [1][3800/5758] lr: 1.0000e-04 eta: 13:12:04 time: 0.8697 data_time: 0.0028 memory: 28798 grad_norm: 25.5704 loss: 0.4094 +2023/06/01 22:55:54 - mmengine - INFO - Epoch(train) [1][3900/5758] lr: 1.0000e-04 eta: 13:10:14 time: 0.7843 data_time: 0.0017 memory: 28798 grad_norm: 14.4033 loss: 0.3184 +2023/06/01 22:57:22 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_lr1e-4_10epoch_20230601_215742 +2023/06/01 22:57:22 - mmengine - INFO - Epoch(train) [1][4000/5758] lr: 1.0000e-04 eta: 13:08:45 time: 0.8678 data_time: 0.0028 memory: 28798 grad_norm: 13.1149 loss: 0.3070 +2023/06/01 22:58:52 - mmengine - INFO - Epoch(train) [1][4100/5758] lr: 1.0000e-04 eta: 13:07:33 time: 0.9147 data_time: 0.0029 memory: 28798 grad_norm: 13.9967 loss: 0.3208 +2023/06/01 23:00:25 - mmengine - INFO - Epoch(train) [1][4200/5758] lr: 1.0000e-04 eta: 13:07:01 time: 1.0080 data_time: 0.0026 memory: 28798 grad_norm: 4.9921 loss: 0.3064 +2023/06/01 23:01:52 - mmengine - INFO - Epoch(train) [1][4300/5758] lr: 1.0000e-04 eta: 13:05:18 time: 0.8624 data_time: 0.0022 memory: 28798 grad_norm: 17.3184 loss: 0.3293 +2023/06/01 23:03:23 - mmengine - INFO - Epoch(train) [1][4400/5758] lr: 1.0000e-04 eta: 13:04:25 time: 0.8715 data_time: 0.0031 memory: 28798 grad_norm: 7.4838 loss: 0.3053 +2023/06/01 23:04:54 - mmengine - INFO - Epoch(train) [1][4500/5758] lr: 1.0000e-04 eta: 13:03:20 time: 0.9511 data_time: 0.0025 memory: 28798 grad_norm: 17.1346 loss: 0.3356 +2023/06/01 23:06:22 - mmengine - INFO - Epoch(train) [1][4600/5758] lr: 1.0000e-04 eta: 13:01:49 time: 0.8073 data_time: 0.0016 memory: 28798 grad_norm: 7.5547 loss: 0.3055 +2023/06/01 23:07:55 - mmengine - INFO - Epoch(train) [1][4700/5758] lr: 1.0000e-04 eta: 13:01:10 time: 0.8453 data_time: 0.0018 memory: 28798 grad_norm: 12.2834 loss: 0.3419 +2023/06/01 23:09:23 - mmengine - INFO - Epoch(train) [1][4800/5758] lr: 1.0000e-04 eta: 12:59:32 time: 0.8357 data_time: 0.0052 memory: 28798 grad_norm: 11.8146 loss: 0.3139 +2023/06/01 23:10:57 - mmengine - INFO - Epoch(train) [1][4900/5758] lr: 1.0000e-04 eta: 12:59:05 time: 0.9410 data_time: 0.0026 memory: 28798 grad_norm: 19.7075 loss: 0.3382 +2023/06/01 23:12:26 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_lr1e-4_10epoch_20230601_215742 +2023/06/01 23:12:26 - mmengine - INFO - Epoch(train) [1][5000/5758] lr: 1.0000e-04 eta: 12:57:42 time: 0.8171 data_time: 0.0018 memory: 28798 grad_norm: 15.5692 loss: 0.3450 +2023/06/01 23:13:53 - mmengine - INFO - Epoch(train) [1][5100/5758] lr: 1.0000e-04 eta: 12:55:47 time: 0.8111 data_time: 0.0025 memory: 28798 grad_norm: 21.3685 loss: 0.3238 +2023/06/01 23:15:22 - mmengine - INFO - Epoch(train) [1][5200/5758] lr: 1.0000e-04 eta: 12:54:24 time: 0.8335 data_time: 0.0018 memory: 28798 grad_norm: 10.3695 loss: 0.2823 +2023/06/01 23:16:51 - mmengine - INFO - Epoch(train) [1][5300/5758] lr: 1.0000e-04 eta: 12:52:58 time: 0.9791 data_time: 0.0018 memory: 28798 grad_norm: 16.0708 loss: 0.3089 +2023/06/01 23:18:18 - mmengine - INFO - Epoch(train) [1][5400/5758] lr: 1.0000e-04 eta: 12:51:18 time: 0.9507 data_time: 0.0031 memory: 28798 grad_norm: 8.5688 loss: 0.2873 +2023/06/01 23:19:50 - mmengine - INFO - Epoch(train) [1][5500/5758] lr: 1.0000e-04 eta: 12:50:14 time: 0.9241 data_time: 0.0032 memory: 28798 grad_norm: 5.6643 loss: 0.2761 +2023/06/01 23:21:21 - mmengine - INFO - Epoch(train) [1][5600/5758] lr: 1.0000e-04 eta: 12:49:10 time: 1.0534 data_time: 0.0066 memory: 28798 grad_norm: 6.4081 loss: 0.2866 +2023/06/01 23:22:49 - mmengine - INFO - Epoch(train) [1][5700/5758] lr: 1.0000e-04 eta: 12:47:33 time: 0.8970 data_time: 0.0033 memory: 28798 grad_norm: 6.4947 loss: 0.2806 +2023/06/01 23:23:34 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_lr1e-4_10epoch_20230601_215742 +2023/06/01 23:23:34 - mmengine - INFO - Saving checkpoint at 1 epochs +2023/06/01 23:24:02 - mmengine - INFO - Epoch(val) [1][16/16] accuracy/top1: 98.8132 single-label/precision_classwise: [100.0, 0.0] single-label/recall_classwise: [98.8132095336914, 0.0] single-label/f1-score_classwise: [99.40306854248047, 0.0] data_time: 0.3786 time: 1.0039 +2023/06/01 23:25:41 - mmengine - INFO - Epoch(train) [2][ 100/5758] lr: 9.7798e-05 eta: 12:45:42 time: 1.0490 data_time: 0.3840 memory: 28798 grad_norm: 4.1619 loss: 0.2769 +2023/06/01 23:27:07 - mmengine - INFO - Epoch(train) [2][ 200/5758] lr: 9.7798e-05 eta: 12:43:54 time: 0.8727 data_time: 0.1486 memory: 28783 grad_norm: 4.1896 loss: 0.2639 +2023/06/01 23:27:47 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_lr1e-4_10epoch_20230601_215742 +2023/06/01 23:28:38 - mmengine - INFO - Epoch(train) [2][ 300/5758] lr: 9.7798e-05 eta: 12:42:44 time: 0.9440 data_time: 0.0028 memory: 28783 grad_norm: 5.8567 loss: 0.2715 +2023/06/01 23:30:04 - mmengine - INFO - Epoch(train) [2][ 400/5758] lr: 9.7798e-05 eta: 12:40:52 time: 0.9153 data_time: 0.1457 memory: 28783 grad_norm: 12.2244 loss: 0.2872 +2023/06/01 23:31:35 - mmengine - INFO - Epoch(train) [2][ 500/5758] lr: 9.7798e-05 eta: 12:39:40 time: 0.8883 data_time: 0.3520 memory: 28783 grad_norm: 6.2108 loss: 0.2665 +2023/06/01 23:33:06 - mmengine - INFO - Epoch(train) [2][ 600/5758] lr: 9.7798e-05 eta: 12:38:29 time: 0.8984 data_time: 0.3772 memory: 28783 grad_norm: 3.9215 loss: 0.2590 +2023/06/01 23:34:37 - mmengine - INFO - Epoch(train) [2][ 700/5758] lr: 9.7798e-05 eta: 12:37:13 time: 0.7520 data_time: 0.2410 memory: 28783 grad_norm: 9.5356 loss: 0.2719 +2023/06/01 23:36:08 - mmengine - INFO - Epoch(train) [2][ 800/5758] lr: 9.7798e-05 eta: 12:36:05 time: 0.8914 data_time: 0.3728 memory: 28783 grad_norm: 18.7965 loss: 0.2840 +2023/06/01 23:37:36 - mmengine - INFO - Epoch(train) [2][ 900/5758] lr: 9.7798e-05 eta: 12:34:28 time: 0.9225 data_time: 0.4152 memory: 28783 grad_norm: 5.0771 loss: 0.2609 +2023/06/01 23:39:06 - mmengine - INFO - Epoch(train) [2][1000/5758] lr: 9.7798e-05 eta: 12:33:10 time: 0.9759 data_time: 0.4156 memory: 28783 grad_norm: 15.9126 loss: 0.2787 +2023/06/01 23:40:37 - mmengine - INFO - Epoch(train) [2][1100/5758] lr: 9.7798e-05 eta: 12:31:52 time: 0.9700 data_time: 0.4600 memory: 28783 grad_norm: 3.7417 loss: 0.2555 +2023/06/01 23:42:09 - mmengine - INFO - Epoch(train) [2][1200/5758] lr: 9.7798e-05 eta: 12:30:50 time: 0.9404 data_time: 0.4134 memory: 28783 grad_norm: 3.5301 loss: 0.2491 +2023/06/01 23:42:52 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_lr1e-4_10epoch_20230601_215742 +2023/06/01 23:43:43 - mmengine - INFO - Epoch(train) [2][1300/5758] lr: 9.7798e-05 eta: 12:29:54 time: 0.9979 data_time: 0.4789 memory: 28783 grad_norm: 3.1146 loss: 0.2537 +2023/06/01 23:45:12 - mmengine - INFO - Epoch(train) [2][1400/5758] lr: 9.7798e-05 eta: 12:28:26 time: 0.8927 data_time: 0.3671 memory: 28783 grad_norm: 9.0718 loss: 0.2690 +2023/06/01 23:46:47 - mmengine - INFO - Epoch(train) [2][1500/5758] lr: 9.7798e-05 eta: 12:27:37 time: 0.9015 data_time: 0.3772 memory: 28783 grad_norm: 5.8936 loss: 0.2456 +2023/06/01 23:48:18 - mmengine - INFO - Epoch(train) [2][1600/5758] lr: 9.7798e-05 eta: 12:26:22 time: 0.9221 data_time: 0.3809 memory: 28783 grad_norm: 5.3704 loss: 0.2492 +2023/06/01 23:49:53 - mmengine - INFO - Epoch(train) [2][1700/5758] lr: 9.7798e-05 eta: 12:25:32 time: 0.9938 data_time: 0.4558 memory: 28783 grad_norm: 8.5210 loss: 0.2679 +2023/06/01 23:51:26 - mmengine - INFO - Epoch(train) [2][1800/5758] lr: 9.7798e-05 eta: 12:24:29 time: 0.8921 data_time: 0.3625 memory: 28783 grad_norm: 2.5579 loss: 0.2437 +2023/06/01 23:52:59 - mmengine - INFO - Epoch(train) [2][1900/5758] lr: 9.7798e-05 eta: 12:23:22 time: 0.9113 data_time: 0.3899 memory: 28783 grad_norm: 4.3726 loss: 0.2428 +2023/06/01 23:54:31 - mmengine - INFO - Epoch(train) [2][2000/5758] lr: 9.7798e-05 eta: 12:22:08 time: 1.0189 data_time: 0.5068 memory: 28783 grad_norm: 4.5386 loss: 0.2475 +2023/06/01 23:56:01 - mmengine - INFO - Epoch(train) [2][2100/5758] lr: 9.7798e-05 eta: 12:20:43 time: 0.8366 data_time: 0.3233 memory: 28783 grad_norm: 5.3834 loss: 0.2443 +2023/06/01 23:57:34 - mmengine - INFO - Epoch(train) [2][2200/5758] lr: 9.7798e-05 eta: 12:19:39 time: 0.8473 data_time: 0.3371 memory: 28783 grad_norm: 8.6770 loss: 0.2449 +2023/06/01 23:58:17 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_lr1e-4_10epoch_20230601_215742 +2023/06/01 23:59:07 - mmengine - INFO - Epoch(train) [2][2300/5758] lr: 9.7798e-05 eta: 12:18:30 time: 0.9851 data_time: 0.4483 memory: 28783 grad_norm: 8.4887 loss: 0.2498 +2023/06/02 00:00:38 - mmengine - INFO - Epoch(train) [2][2400/5758] lr: 9.7798e-05 eta: 12:17:06 time: 0.8089 data_time: 0.2924 memory: 28783 grad_norm: 2.4638 loss: 0.2431 +2023/06/02 00:02:10 - mmengine - INFO - Epoch(train) [2][2500/5758] lr: 9.7798e-05 eta: 12:15:52 time: 0.9934 data_time: 0.4742 memory: 28783 grad_norm: 4.8978 loss: 0.2456 +2023/06/02 00:03:40 - mmengine - INFO - Epoch(train) [2][2600/5758] lr: 9.7798e-05 eta: 12:14:29 time: 0.8580 data_time: 0.3147 memory: 28783 grad_norm: 4.0632 loss: 0.2378 +2023/06/02 00:05:08 - mmengine - INFO - Epoch(train) [2][2700/5758] lr: 9.7798e-05 eta: 12:12:50 time: 0.7924 data_time: 0.2870 memory: 28783 grad_norm: 3.8485 loss: 0.2327 +2023/06/02 00:06:44 - mmengine - INFO - Epoch(train) [2][2800/5758] lr: 9.7798e-05 eta: 12:11:55 time: 0.9835 data_time: 0.4635 memory: 28783 grad_norm: 2.7544 loss: 0.2309 +2023/06/02 00:08:14 - mmengine - INFO - Epoch(train) [2][2900/5758] lr: 9.7798e-05 eta: 12:10:28 time: 0.9933 data_time: 0.4630 memory: 28783 grad_norm: 2.4494 loss: 0.2383 +2023/06/02 00:09:47 - mmengine - INFO - Epoch(train) [2][3000/5758] lr: 9.7798e-05 eta: 12:09:17 time: 0.9370 data_time: 0.4174 memory: 28783 grad_norm: 8.0012 loss: 0.2366 +2023/06/02 00:11:18 - mmengine - INFO - Epoch(train) [2][3100/5758] lr: 9.7798e-05 eta: 12:07:57 time: 0.9992 data_time: 0.4664 memory: 28783 grad_norm: 2.1215 loss: 0.2264 +2023/06/02 00:12:44 - mmengine - INFO - Epoch(train) [2][3200/5758] lr: 9.7798e-05 eta: 12:06:08 time: 0.8182 data_time: 0.3096 memory: 28783 grad_norm: 9.5066 loss: 0.2283 +2023/06/02 00:13:27 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_lr1e-4_10epoch_20230601_215742 +2023/06/02 00:14:15 - mmengine - INFO - Epoch(train) [2][3300/5758] lr: 9.7798e-05 eta: 12:04:46 time: 0.9207 data_time: 0.3951 memory: 28783 grad_norm: 7.0511 loss: 0.2564 +2023/06/02 00:15:49 - mmengine - INFO - Epoch(train) [2][3400/5758] lr: 9.7798e-05 eta: 12:03:39 time: 0.8559 data_time: 0.3377 memory: 28783 grad_norm: 2.7573 loss: 0.2325 +2023/06/02 00:17:23 - mmengine - INFO - Epoch(train) [2][3500/5758] lr: 9.7798e-05 eta: 12:02:30 time: 1.0598 data_time: 0.4870 memory: 28783 grad_norm: 2.7590 loss: 0.2360 +2023/06/02 00:18:54 - mmengine - INFO - Epoch(train) [2][3600/5758] lr: 9.7798e-05 eta: 12:01:08 time: 0.9436 data_time: 0.4315 memory: 28783 grad_norm: 2.9784 loss: 0.2242 +2023/06/02 00:20:26 - mmengine - INFO - Epoch(train) [2][3700/5758] lr: 9.7798e-05 eta: 11:59:49 time: 0.8966 data_time: 0.3649 memory: 28783 grad_norm: 3.7143 loss: 0.2301 +2023/06/02 00:21:57 - mmengine - INFO - Epoch(train) [2][3800/5758] lr: 9.7798e-05 eta: 11:58:28 time: 0.8507 data_time: 0.3485 memory: 28783 grad_norm: 4.1654 loss: 0.2289 +2023/06/02 00:23:28 - mmengine - INFO - Epoch(train) [2][3900/5758] lr: 9.7798e-05 eta: 11:57:04 time: 0.9176 data_time: 0.3992 memory: 28783 grad_norm: 4.1011 loss: 0.2301 +2023/06/02 00:25:00 - mmengine - INFO - Epoch(train) [2][4000/5758] lr: 9.7798e-05 eta: 11:55:45 time: 0.8501 data_time: 0.3455 memory: 28783 grad_norm: 4.7220 loss: 0.2237 +2023/06/02 00:26:33 - mmengine - INFO - Epoch(train) [2][4100/5758] lr: 9.7798e-05 eta: 11:54:28 time: 0.9194 data_time: 0.4002 memory: 28783 grad_norm: 3.0715 loss: 0.2343 +2023/06/02 00:28:02 - mmengine - INFO - Epoch(train) [2][4200/5758] lr: 9.7798e-05 eta: 11:52:56 time: 0.8182 data_time: 0.3120 memory: 28783 grad_norm: 1.6475 loss: 0.2224 +2023/06/02 00:28:46 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_lr1e-4_10epoch_20230601_215742 +2023/06/02 00:29:34 - mmengine - INFO - Epoch(train) [2][4300/5758] lr: 9.7798e-05 eta: 11:51:36 time: 0.8919 data_time: 0.3542 memory: 28783 grad_norm: 2.7323 loss: 0.2206 +2023/06/02 00:31:06 - mmengine - INFO - Epoch(train) [2][4400/5758] lr: 9.7798e-05 eta: 11:50:18 time: 0.9701 data_time: 0.4390 memory: 28783 grad_norm: 2.8713 loss: 0.2190 +2023/06/02 00:32:39 - mmengine - INFO - Epoch(train) [2][4500/5758] lr: 9.7798e-05 eta: 11:48:59 time: 0.8785 data_time: 0.3462 memory: 28783 grad_norm: 1.8940 loss: 0.2245 +2023/06/02 00:34:09 - mmengine - INFO - Epoch(train) [2][4600/5758] lr: 9.7798e-05 eta: 11:47:32 time: 0.8904 data_time: 0.3694 memory: 28783 grad_norm: 2.0799 loss: 0.2208 +2023/06/02 00:35:41 - mmengine - INFO - Epoch(train) [2][4700/5758] lr: 9.7798e-05 eta: 11:46:11 time: 0.9836 data_time: 0.4367 memory: 28783 grad_norm: 3.5731 loss: 0.2306 +2023/06/02 00:37:11 - mmengine - INFO - Epoch(train) [2][4800/5758] lr: 9.7798e-05 eta: 11:44:42 time: 0.8601 data_time: 0.3524 memory: 28783 grad_norm: 1.9234 loss: 0.2200 +2023/06/02 00:38:46 - mmengine - INFO - Epoch(train) [2][4900/5758] lr: 9.7798e-05 eta: 11:43:36 time: 0.9549 data_time: 0.4357 memory: 28783 grad_norm: 1.6966 loss: 0.2197 +2023/06/02 00:40:15 - mmengine - INFO - Epoch(train) [2][5000/5758] lr: 9.7798e-05 eta: 11:42:01 time: 0.7813 data_time: 0.2779 memory: 28783 grad_norm: 2.0280 loss: 0.2189 +2023/06/02 00:41:45 - mmengine - INFO - Epoch(train) [2][5100/5758] lr: 9.7798e-05 eta: 11:40:30 time: 0.8396 data_time: 0.3048 memory: 28783 grad_norm: 2.4688 loss: 0.2178 +2023/06/02 00:43:18 - mmengine - INFO - Epoch(train) [2][5200/5758] lr: 9.7798e-05 eta: 11:39:13 time: 0.9062 data_time: 0.3942 memory: 28783 grad_norm: 1.7279 loss: 0.2191 +2023/06/02 00:44:00 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_lr1e-4_10epoch_20230601_215742 +2023/06/02 00:44:48 - mmengine - INFO - Epoch(train) [2][5300/5758] lr: 9.7798e-05 eta: 11:37:43 time: 0.8706 data_time: 0.3328 memory: 28783 grad_norm: 3.3575 loss: 0.2198 +2023/06/02 00:46:21 - mmengine - INFO - Epoch(train) [2][5400/5758] lr: 9.7798e-05 eta: 11:36:24 time: 0.9027 data_time: 0.3829 memory: 28783 grad_norm: 2.4025 loss: 0.2191 +2023/06/02 00:47:54 - mmengine - INFO - Epoch(train) [2][5500/5758] lr: 9.7798e-05 eta: 11:35:06 time: 0.9741 data_time: 0.4512 memory: 28783 grad_norm: 1.9819 loss: 0.2182 +2023/06/02 00:49:26 - mmengine - INFO - Epoch(train) [2][5600/5758] lr: 9.7798e-05 eta: 11:33:44 time: 0.8589 data_time: 0.3433 memory: 28783 grad_norm: 1.8669 loss: 0.2199 +2023/06/02 00:50:53 - mmengine - INFO - Epoch(train) [2][5700/5758] lr: 9.7798e-05 eta: 11:32:05 time: 0.7965 data_time: 0.2904 memory: 28783 grad_norm: 1.7138 loss: 0.2168 +2023/06/02 00:51:46 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_lr1e-4_10epoch_20230601_215742 +2023/06/02 00:51:46 - mmengine - INFO - Saving checkpoint at 2 epochs +2023/06/02 00:52:08 - mmengine - INFO - Epoch(val) [2][16/16] accuracy/top1: 99.5076 single-label/precision_classwise: [100.0, 0.0] single-label/recall_classwise: [99.50760650634766, 0.0] single-label/f1-score_classwise: [99.7531967163086, 0.0] data_time: 0.2443 time: 0.6612 +2023/06/02 00:53:44 - mmengine - INFO - Epoch(train) [3][ 100/5758] lr: 9.1406e-05 eta: 11:30:11 time: 0.9204 data_time: 0.3417 memory: 28783 grad_norm: 4.9919 loss: 0.2289 +2023/06/02 00:55:19 - mmengine - INFO - Epoch(train) [3][ 200/5758] lr: 9.1406e-05 eta: 11:28:59 time: 0.9393 data_time: 0.3465 memory: 28783 grad_norm: 1.4809 loss: 0.2140 +2023/06/02 00:56:47 - mmengine - INFO - Epoch(train) [3][ 300/5758] lr: 9.1406e-05 eta: 11:27:21 time: 0.9688 data_time: 0.4533 memory: 28783 grad_norm: 3.4678 loss: 0.2252 +2023/06/02 00:58:20 - mmengine - INFO - Epoch(train) [3][ 400/5758] lr: 9.1406e-05 eta: 11:26:03 time: 1.0829 data_time: 0.5469 memory: 28783 grad_norm: 1.8249 loss: 0.2154 +2023/06/02 00:59:38 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_lr1e-4_10epoch_20230601_215742 +2023/06/02 00:59:50 - mmengine - INFO - Epoch(train) [3][ 500/5758] lr: 9.1406e-05 eta: 11:24:33 time: 0.9390 data_time: 0.3975 memory: 28783 grad_norm: 1.5757 loss: 0.2162 +2023/06/02 01:01:23 - mmengine - INFO - Epoch(train) [3][ 600/5758] lr: 9.1406e-05 eta: 11:23:10 time: 0.9061 data_time: 0.3890 memory: 28783 grad_norm: 3.1360 loss: 0.2171 +2023/06/02 01:02:56 - mmengine - INFO - Epoch(train) [3][ 700/5758] lr: 9.1406e-05 eta: 11:21:51 time: 0.9092 data_time: 0.3880 memory: 28783 grad_norm: 2.3387 loss: 0.2131 +2023/06/02 01:04:25 - mmengine - INFO - Epoch(train) [3][ 800/5758] lr: 9.1406e-05 eta: 11:20:19 time: 0.9058 data_time: 0.4014 memory: 28783 grad_norm: 2.5021 loss: 0.2186 +2023/06/02 01:05:56 - mmengine - INFO - Epoch(train) [3][ 900/5758] lr: 9.1406e-05 eta: 11:18:51 time: 0.9913 data_time: 0.4402 memory: 28783 grad_norm: 1.5621 loss: 0.2100 +2023/06/02 01:07:28 - mmengine - INFO - Epoch(train) [3][1000/5758] lr: 9.1406e-05 eta: 11:17:26 time: 0.8778 data_time: 0.3723 memory: 28783 grad_norm: 2.1560 loss: 0.2116 +2023/06/02 01:09:01 - mmengine - INFO - Epoch(train) [3][1100/5758] lr: 9.1406e-05 eta: 11:16:08 time: 0.9526 data_time: 0.4227 memory: 28783 grad_norm: 3.1885 loss: 0.2241 +2023/06/02 01:10:34 - mmengine - INFO - Epoch(train) [3][1200/5758] lr: 9.1406e-05 eta: 11:14:46 time: 0.9364 data_time: 0.4059 memory: 28783 grad_norm: 1.6028 loss: 0.2146 +2023/06/02 01:12:06 - mmengine - INFO - Epoch(train) [3][1300/5758] lr: 9.1406e-05 eta: 11:13:21 time: 0.8682 data_time: 0.3477 memory: 28783 grad_norm: 1.6502 loss: 0.2128 +2023/06/02 01:13:41 - mmengine - INFO - Epoch(train) [3][1400/5758] lr: 9.1406e-05 eta: 11:12:08 time: 0.9288 data_time: 0.4140 memory: 28783 grad_norm: 2.0179 loss: 0.2152 +2023/06/02 01:14:58 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_lr1e-4_10epoch_20230601_215742 +2023/06/02 01:15:11 - mmengine - INFO - Epoch(train) [3][1500/5758] lr: 9.1406e-05 eta: 11:10:37 time: 0.9187 data_time: 0.3942 memory: 28783 grad_norm: 1.2643 loss: 0.2143 +2023/06/02 01:16:43 - mmengine - INFO - Epoch(train) [3][1600/5758] lr: 9.1406e-05 eta: 11:09:14 time: 0.9469 data_time: 0.4297 memory: 28783 grad_norm: 2.9025 loss: 0.2158 +2023/06/02 01:18:15 - mmengine - INFO - Epoch(train) [3][1700/5758] lr: 9.1406e-05 eta: 11:07:48 time: 0.9417 data_time: 0.4209 memory: 28783 grad_norm: 1.6698 loss: 0.2111 +2023/06/02 01:19:44 - mmengine - INFO - Epoch(train) [3][1800/5758] lr: 9.1406e-05 eta: 11:06:15 time: 0.9461 data_time: 0.4146 memory: 28783 grad_norm: 1.8726 loss: 0.2084 +2023/06/02 01:21:17 - mmengine - INFO - Epoch(train) [3][1900/5758] lr: 9.1406e-05 eta: 11:04:50 time: 0.8851 data_time: 0.3677 memory: 28783 grad_norm: 2.3280 loss: 0.2135 +2023/06/02 01:22:47 - mmengine - INFO - Epoch(train) [3][2000/5758] lr: 9.1406e-05 eta: 11:03:21 time: 1.0227 data_time: 0.4987 memory: 28783 grad_norm: 2.1439 loss: 0.2129 +2023/06/02 01:24:16 - mmengine - INFO - Epoch(train) [3][2100/5758] lr: 9.1406e-05 eta: 11:01:45 time: 0.9612 data_time: 0.4322 memory: 28783 grad_norm: 1.7115 loss: 0.2114 +2023/06/02 01:25:45 - mmengine - INFO - Epoch(train) [3][2200/5758] lr: 9.1406e-05 eta: 11:00:12 time: 0.8291 data_time: 0.3274 memory: 28783 grad_norm: 1.5761 loss: 0.2112 +2023/06/02 01:27:15 - mmengine - INFO - Epoch(train) [3][2300/5758] lr: 9.1406e-05 eta: 10:58:41 time: 0.8635 data_time: 0.3388 memory: 28783 grad_norm: 1.2157 loss: 0.2136 +2023/06/02 01:28:41 - mmengine - INFO - Epoch(train) [3][2400/5758] lr: 9.1406e-05 eta: 10:56:57 time: 0.8584 data_time: 0.3208 memory: 28783 grad_norm: 1.9318 loss: 0.2114 +2023/06/02 01:30:00 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_lr1e-4_10epoch_20230601_215742 +2023/06/02 01:30:13 - mmengine - INFO - Epoch(train) [3][2500/5758] lr: 9.1406e-05 eta: 10:55:31 time: 0.9521 data_time: 0.4241 memory: 28783 grad_norm: 2.0981 loss: 0.2145 +2023/06/02 01:31:44 - mmengine - INFO - Epoch(train) [3][2600/5758] lr: 9.1406e-05 eta: 10:54:02 time: 0.9630 data_time: 0.4383 memory: 28783 grad_norm: 1.5923 loss: 0.2129 +2023/06/02 01:33:14 - mmengine - INFO - Epoch(train) [3][2700/5758] lr: 9.1406e-05 eta: 10:52:32 time: 1.0036 data_time: 0.4507 memory: 28783 grad_norm: 1.4237 loss: 0.2050 +2023/06/02 01:34:41 - mmengine - INFO - Epoch(train) [3][2800/5758] lr: 9.1406e-05 eta: 10:50:53 time: 0.9463 data_time: 0.4187 memory: 28783 grad_norm: 1.3269 loss: 0.2091 +2023/06/02 01:36:11 - mmengine - INFO - Epoch(train) [3][2900/5758] lr: 9.1406e-05 eta: 10:49:19 time: 0.7959 data_time: 0.2802 memory: 28783 grad_norm: 1.6407 loss: 0.2094 +2023/06/02 01:37:36 - mmengine - INFO - Epoch(train) [3][3000/5758] lr: 9.1406e-05 eta: 10:47:36 time: 0.8174 data_time: 0.3092 memory: 28783 grad_norm: 1.1379 loss: 0.2077 +2023/06/02 01:39:05 - mmengine - INFO - Epoch(train) [3][3100/5758] lr: 9.1406e-05 eta: 10:46:02 time: 0.8540 data_time: 0.3260 memory: 28783 grad_norm: 1.2323 loss: 0.2101 +2023/06/02 01:40:40 - mmengine - INFO - Epoch(train) [3][3200/5758] lr: 9.1406e-05 eta: 10:44:43 time: 0.8760 data_time: 0.3706 memory: 28783 grad_norm: 1.0656 loss: 0.2103 +2023/06/02 01:42:11 - mmengine - INFO - Epoch(train) [3][3300/5758] lr: 9.1406e-05 eta: 10:43:18 time: 0.8942 data_time: 0.3760 memory: 28783 grad_norm: 1.8808 loss: 0.2105 +2023/06/02 01:43:43 - mmengine - INFO - Epoch(train) [3][3400/5758] lr: 9.1406e-05 eta: 10:41:52 time: 1.0165 data_time: 0.4818 memory: 28783 grad_norm: 1.8256 loss: 0.2113 +2023/06/02 01:45:05 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_lr1e-4_10epoch_20230601_215742 +2023/06/02 01:45:18 - mmengine - INFO - Epoch(train) [3][3500/5758] lr: 9.1406e-05 eta: 10:40:33 time: 0.9386 data_time: 0.4129 memory: 28783 grad_norm: 1.2121 loss: 0.2127 +2023/06/02 01:46:49 - mmengine - INFO - Epoch(train) [3][3600/5758] lr: 9.1406e-05 eta: 10:39:06 time: 0.9260 data_time: 0.4051 memory: 28783 grad_norm: 2.5419 loss: 0.2151 +2023/06/02 01:48:25 - mmengine - INFO - Epoch(train) [3][3700/5758] lr: 9.1406e-05 eta: 10:37:51 time: 1.0452 data_time: 0.5196 memory: 28783 grad_norm: 2.4269 loss: 0.2131 +2023/06/02 01:49:57 - mmengine - INFO - Epoch(train) [3][3800/5758] lr: 9.1406e-05 eta: 10:36:26 time: 0.9554 data_time: 0.4267 memory: 28783 grad_norm: 1.5699 loss: 0.2147 +2023/06/02 01:51:27 - mmengine - INFO - Epoch(train) [3][3900/5758] lr: 9.1406e-05 eta: 10:34:54 time: 0.9724 data_time: 0.4424 memory: 28783 grad_norm: 1.3582 loss: 0.2093 +2023/06/02 01:53:00 - mmengine - INFO - Epoch(train) [3][4000/5758] lr: 9.1406e-05 eta: 10:33:31 time: 0.9872 data_time: 0.4571 memory: 28783 grad_norm: 1.8174 loss: 0.2117 +2023/06/02 01:54:27 - mmengine - INFO - Epoch(train) [3][4100/5758] lr: 9.1406e-05 eta: 10:31:53 time: 0.8520 data_time: 0.3400 memory: 28783 grad_norm: 1.3211 loss: 0.2090 +2023/06/02 01:55:54 - mmengine - INFO - Epoch(train) [3][4200/5758] lr: 9.1406e-05 eta: 10:30:13 time: 0.9213 data_time: 0.4010 memory: 28783 grad_norm: 1.5703 loss: 0.2115 +2023/06/02 01:57:14 - mmengine - INFO - Epoch(train) [3][4300/5758] lr: 9.1406e-05 eta: 10:28:16 time: 0.7282 data_time: 0.2236 memory: 28783 grad_norm: 1.5953 loss: 0.2068 +2023/06/02 01:58:38 - mmengine - INFO - Epoch(train) [3][4400/5758] lr: 9.1406e-05 eta: 10:26:27 time: 0.8022 data_time: 0.2912 memory: 28783 grad_norm: 0.9083 loss: 0.2078 +2023/06/02 01:59:51 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_lr1e-4_10epoch_20230601_215742 +2023/06/02 02:00:03 - mmengine - INFO - Epoch(train) [3][4500/5758] lr: 9.1406e-05 eta: 10:24:44 time: 0.8808 data_time: 0.3436 memory: 28783 grad_norm: 1.3006 loss: 0.2083 +2023/06/02 02:01:27 - mmengine - INFO - Epoch(train) [3][4600/5758] lr: 9.1406e-05 eta: 10:22:57 time: 0.8163 data_time: 0.3092 memory: 28783 grad_norm: 1.7438 loss: 0.2115 +2023/06/02 02:02:50 - mmengine - INFO - Epoch(train) [3][4700/5758] lr: 9.1406e-05 eta: 10:21:09 time: 0.8309 data_time: 0.2674 memory: 28783 grad_norm: 1.4219 loss: 0.2073 +2023/06/02 02:04:12 - mmengine - INFO - Epoch(train) [3][4800/5758] lr: 9.1406e-05 eta: 10:19:18 time: 0.7453 data_time: 0.1598 memory: 28783 grad_norm: 1.2673 loss: 0.2094 +2023/06/02 02:05:36 - mmengine - INFO - Epoch(train) [3][4900/5758] lr: 9.1406e-05 eta: 10:17:35 time: 0.8646 data_time: 0.0837 memory: 28783 grad_norm: 1.2609 loss: 0.2133 +2023/06/02 02:07:01 - mmengine - INFO - Epoch(train) [3][5000/5758] lr: 9.1406e-05 eta: 10:15:53 time: 0.7910 data_time: 0.0128 memory: 28783 grad_norm: 1.0886 loss: 0.2061 +2023/06/02 02:08:25 - mmengine - INFO - Epoch(train) [3][5100/5758] lr: 9.1406e-05 eta: 10:14:07 time: 0.7758 data_time: 0.0018 memory: 28783 grad_norm: 0.8696 loss: 0.2041 +2023/06/02 02:09:51 - mmengine - INFO - Epoch(train) [3][5200/5758] lr: 9.1406e-05 eta: 10:12:27 time: 0.9271 data_time: 0.0023 memory: 28783 grad_norm: 1.2496 loss: 0.2047 +2023/06/02 02:11:10 - mmengine - INFO - Epoch(train) [3][5300/5758] lr: 9.1406e-05 eta: 10:10:31 time: 0.8782 data_time: 0.0021 memory: 28783 grad_norm: 1.0303 loss: 0.2092 +2023/06/02 02:12:35 - mmengine - INFO - Epoch(train) [3][5400/5758] lr: 9.1406e-05 eta: 10:08:49 time: 0.7746 data_time: 0.0014 memory: 28783 grad_norm: 2.0796 loss: 0.2156 +2023/06/02 02:13:48 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_lr1e-4_10epoch_20230601_215742 +2023/06/02 02:13:59 - mmengine - INFO - Epoch(train) [3][5500/5758] lr: 9.1406e-05 eta: 10:07:06 time: 0.7694 data_time: 0.0019 memory: 28783 grad_norm: 1.2639 loss: 0.2094 +2023/06/02 02:15:25 - mmengine - INFO - Epoch(train) [3][5600/5758] lr: 9.1406e-05 eta: 10:05:26 time: 0.9989 data_time: 0.0026 memory: 28783 grad_norm: 1.0399 loss: 0.2065 +2023/06/02 02:16:50 - mmengine - INFO - Epoch(train) [3][5700/5758] lr: 9.1406e-05 eta: 10:03:45 time: 0.8722 data_time: 0.0025 memory: 28783 grad_norm: 1.0689 loss: 0.2101 +2023/06/02 02:17:41 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_lr1e-4_10epoch_20230601_215742 +2023/06/02 02:17:41 - mmengine - INFO - Saving checkpoint at 3 epochs +2023/06/02 02:18:01 - mmengine - INFO - Epoch(val) [3][16/16] accuracy/top1: 99.8359 single-label/precision_classwise: [100.0, 0.0] single-label/recall_classwise: [99.83586883544922, 0.0] single-label/f1-score_classwise: [99.91786193847656, 0.0] data_time: 0.2443 time: 0.6611 +2023/06/02 02:19:34 - mmengine - INFO - Epoch(train) [4][ 100/5758] lr: 8.1450e-05 eta: 10:01:28 time: 0.9271 data_time: 0.1584 memory: 28783 grad_norm: 0.9486 loss: 0.2062 +2023/06/02 02:21:00 - mmengine - INFO - Epoch(train) [4][ 200/5758] lr: 8.1450e-05 eta: 9:59:50 time: 0.9239 data_time: 0.1749 memory: 28783 grad_norm: 1.2736 loss: 0.2056 +2023/06/02 02:22:25 - mmengine - INFO - Epoch(train) [4][ 300/5758] lr: 8.1450e-05 eta: 9:58:09 time: 0.8133 data_time: 0.0969 memory: 28783 grad_norm: 0.9428 loss: 0.2072 +2023/06/02 02:23:53 - mmengine - INFO - Epoch(train) [4][ 400/5758] lr: 8.1450e-05 eta: 9:56:36 time: 0.8681 data_time: 0.0017 memory: 28783 grad_norm: 0.7737 loss: 0.2065 +2023/06/02 02:25:19 - mmengine - INFO - Epoch(train) [4][ 500/5758] lr: 8.1450e-05 eta: 9:54:57 time: 0.9601 data_time: 0.0027 memory: 28783 grad_norm: 1.2201 loss: 0.2086 +2023/06/02 02:26:44 - mmengine - INFO - Epoch(train) [4][ 600/5758] lr: 8.1450e-05 eta: 9:53:17 time: 0.9038 data_time: 0.0118 memory: 28783 grad_norm: 1.3274 loss: 0.2047 +2023/06/02 02:28:09 - mmengine - INFO - Epoch(train) [4][ 700/5758] lr: 8.1450e-05 eta: 9:51:38 time: 0.8714 data_time: 0.0021 memory: 28783 grad_norm: 0.8235 loss: 0.2061 +2023/06/02 02:28:31 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_lr1e-4_10epoch_20230601_215742 +2023/06/02 02:29:33 - mmengine - INFO - Epoch(train) [4][ 800/5758] lr: 8.1450e-05 eta: 9:49:55 time: 0.9287 data_time: 0.0024 memory: 28783 grad_norm: 1.2387 loss: 0.2086 +2023/06/02 02:30:54 - mmengine - INFO - Epoch(train) [4][ 900/5758] lr: 8.1450e-05 eta: 9:48:08 time: 0.7168 data_time: 0.0016 memory: 28783 grad_norm: 0.9570 loss: 0.2059 +2023/06/02 02:32:20 - mmengine - INFO - Epoch(train) [4][1000/5758] lr: 8.1450e-05 eta: 9:46:30 time: 0.8189 data_time: 0.0016 memory: 28783 grad_norm: 1.2040 loss: 0.2065 +2023/06/02 02:33:44 - mmengine - INFO - Epoch(train) [4][1100/5758] lr: 8.1450e-05 eta: 9:44:50 time: 0.9110 data_time: 0.0024 memory: 28783 grad_norm: 0.8021 loss: 0.2066 +2023/06/02 02:35:09 - mmengine - INFO - Epoch(train) [4][1200/5758] lr: 8.1450e-05 eta: 9:43:09 time: 0.7593 data_time: 0.0022 memory: 28783 grad_norm: 1.0723 loss: 0.2107 +2023/06/02 02:36:33 - mmengine - INFO - Epoch(train) [4][1300/5758] lr: 8.1450e-05 eta: 9:41:28 time: 0.9461 data_time: 0.0025 memory: 28783 grad_norm: 1.1991 loss: 0.2059 +2023/06/02 02:37:55 - mmengine - INFO - Epoch(train) [4][1400/5758] lr: 8.1450e-05 eta: 9:39:45 time: 0.7589 data_time: 0.0029 memory: 28783 grad_norm: 0.6547 loss: 0.2095 +2023/06/02 02:39:19 - mmengine - INFO - Epoch(train) [4][1500/5758] lr: 8.1450e-05 eta: 9:38:04 time: 0.9276 data_time: 0.0039 memory: 28783 grad_norm: 0.9341 loss: 0.2064 +2023/06/02 02:40:43 - mmengine - INFO - Epoch(train) [4][1600/5758] lr: 8.1450e-05 eta: 9:36:23 time: 0.7641 data_time: 0.0017 memory: 28783 grad_norm: 1.3940 loss: 0.2076 +2023/06/02 02:42:08 - mmengine - INFO - Epoch(train) [4][1700/5758] lr: 8.1450e-05 eta: 9:34:45 time: 0.8952 data_time: 0.0025 memory: 28783 grad_norm: 1.4498 loss: 0.2028 +2023/06/02 02:42:30 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_lr1e-4_10epoch_20230601_215742 +2023/06/02 02:43:31 - mmengine - INFO - Epoch(train) [4][1800/5758] lr: 8.1450e-05 eta: 9:33:03 time: 0.8861 data_time: 0.0024 memory: 28783 grad_norm: 1.2632 loss: 0.2109 +2023/06/02 02:44:55 - mmengine - INFO - Epoch(train) [4][1900/5758] lr: 8.1450e-05 eta: 9:31:23 time: 0.7154 data_time: 0.0015 memory: 28783 grad_norm: 0.8616 loss: 0.2092 +2023/06/02 02:46:20 - mmengine - INFO - Epoch(train) [4][2000/5758] lr: 8.1450e-05 eta: 9:29:45 time: 0.8167 data_time: 0.0019 memory: 28783 grad_norm: 0.7398 loss: 0.2033 +2023/06/02 02:47:41 - mmengine - INFO - Epoch(train) [4][2100/5758] lr: 8.1450e-05 eta: 9:28:00 time: 0.7740 data_time: 0.0018 memory: 28783 grad_norm: 0.7307 loss: 0.2065 +2023/06/02 02:49:02 - mmengine - INFO - Epoch(train) [4][2200/5758] lr: 8.1450e-05 eta: 9:26:15 time: 0.8627 data_time: 0.0023 memory: 28783 grad_norm: 0.8455 loss: 0.2050 +2023/06/02 02:50:27 - mmengine - INFO - Epoch(train) [4][2300/5758] lr: 8.1450e-05 eta: 9:24:38 time: 0.8437 data_time: 0.0017 memory: 28783 grad_norm: 1.1941 loss: 0.2076 +2023/06/02 02:51:52 - mmengine - INFO - Epoch(train) [4][2400/5758] lr: 8.1450e-05 eta: 9:22:59 time: 0.9071 data_time: 0.0051 memory: 28783 grad_norm: 0.8481 loss: 0.2040 +2023/06/02 02:53:16 - mmengine - INFO - Epoch(train) [4][2500/5758] lr: 8.1450e-05 eta: 9:21:21 time: 0.8056 data_time: 0.0017 memory: 28783 grad_norm: 1.4306 loss: 0.2102 +2023/06/02 02:54:42 - mmengine - INFO - Epoch(train) [4][2600/5758] lr: 8.1450e-05 eta: 9:19:47 time: 0.8275 data_time: 0.0020 memory: 28783 grad_norm: 0.8969 loss: 0.2049 +2023/06/02 02:56:07 - mmengine - INFO - Epoch(train) [4][2700/5758] lr: 8.1450e-05 eta: 9:18:10 time: 0.9778 data_time: 0.0036 memory: 28783 grad_norm: 0.9320 loss: 0.2046 +2023/06/02 02:56:30 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_lr1e-4_10epoch_20230601_215742 +2023/06/02 02:57:32 - mmengine - INFO - Epoch(train) [4][2800/5758] lr: 8.1450e-05 eta: 9:16:32 time: 0.8735 data_time: 0.0021 memory: 28783 grad_norm: 0.9420 loss: 0.2048 +2023/06/02 02:58:58 - mmengine - INFO - Epoch(train) [4][2900/5758] lr: 8.1450e-05 eta: 9:14:57 time: 0.8794 data_time: 0.0019 memory: 28783 grad_norm: 0.8713 loss: 0.2034 +2023/06/02 03:00:25 - mmengine - INFO - Epoch(train) [4][3000/5758] lr: 8.1450e-05 eta: 9:13:26 time: 0.9041 data_time: 0.0030 memory: 28783 grad_norm: 1.0520 loss: 0.2066 +2023/06/02 03:01:51 - mmengine - INFO - Epoch(train) [4][3100/5758] lr: 8.1450e-05 eta: 9:11:50 time: 0.8119 data_time: 0.0018 memory: 28783 grad_norm: 0.7072 loss: 0.2051 +2023/06/02 03:03:20 - mmengine - INFO - Epoch(train) [4][3200/5758] lr: 8.1450e-05 eta: 9:10:21 time: 0.8202 data_time: 0.0020 memory: 28783 grad_norm: 0.7490 loss: 0.2031 +2023/06/02 03:04:45 - mmengine - INFO - Epoch(train) [4][3300/5758] lr: 8.1450e-05 eta: 9:08:45 time: 0.7922 data_time: 0.0017 memory: 28783 grad_norm: 1.0777 loss: 0.2085 +2023/06/02 03:06:10 - mmengine - INFO - Epoch(train) [4][3400/5758] lr: 8.1450e-05 eta: 9:07:10 time: 0.8532 data_time: 0.0021 memory: 28783 grad_norm: 1.1145 loss: 0.2055 +2023/06/02 03:07:30 - mmengine - INFO - Epoch(train) [4][3500/5758] lr: 8.1450e-05 eta: 9:05:25 time: 0.8022 data_time: 0.0020 memory: 28783 grad_norm: 0.8493 loss: 0.2043 +2023/06/02 03:08:53 - mmengine - INFO - Epoch(train) [4][3600/5758] lr: 8.1450e-05 eta: 9:03:46 time: 0.8174 data_time: 0.0023 memory: 28783 grad_norm: 0.8483 loss: 0.2037 +2023/06/02 03:10:15 - mmengine - INFO - Epoch(train) [4][3700/5758] lr: 8.1450e-05 eta: 9:02:05 time: 0.8144 data_time: 0.0020 memory: 28783 grad_norm: 0.7161 loss: 0.2067 +2023/06/02 03:10:39 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_lr1e-4_10epoch_20230601_215742 +2023/06/02 03:11:42 - mmengine - INFO - Epoch(train) [4][3800/5758] lr: 8.1450e-05 eta: 9:00:32 time: 0.8935 data_time: 0.0025 memory: 28783 grad_norm: 0.9564 loss: 0.2062 +2023/06/02 03:13:06 - mmengine - INFO - Epoch(train) [4][3900/5758] lr: 8.1450e-05 eta: 8:58:56 time: 0.8126 data_time: 0.0018 memory: 28783 grad_norm: 0.8982 loss: 0.2072 +2023/06/02 03:14:31 - mmengine - INFO - Epoch(train) [4][4000/5758] lr: 8.1450e-05 eta: 8:57:20 time: 0.8360 data_time: 0.0017 memory: 28783 grad_norm: 0.7342 loss: 0.2076 +2023/06/02 03:15:58 - mmengine - INFO - Epoch(train) [4][4100/5758] lr: 8.1450e-05 eta: 8:55:47 time: 0.9660 data_time: 0.0023 memory: 28783 grad_norm: 0.9482 loss: 0.2064 +2023/06/02 03:17:22 - mmengine - INFO - Epoch(train) [4][4200/5758] lr: 8.1450e-05 eta: 8:54:12 time: 0.9556 data_time: 0.0034 memory: 28783 grad_norm: 0.8121 loss: 0.2057 +2023/06/02 03:18:44 - mmengine - INFO - Epoch(train) [4][4300/5758] lr: 8.1450e-05 eta: 8:52:32 time: 0.7833 data_time: 0.0018 memory: 28783 grad_norm: 0.8101 loss: 0.2068 +2023/06/02 03:20:06 - mmengine - INFO - Epoch(train) [4][4400/5758] lr: 8.1450e-05 eta: 8:50:51 time: 0.7720 data_time: 0.0030 memory: 28783 grad_norm: 1.1346 loss: 0.2075 +2023/06/02 03:21:28 - mmengine - INFO - Epoch(train) [4][4500/5758] lr: 8.1450e-05 eta: 8:49:11 time: 0.7661 data_time: 0.0017 memory: 28783 grad_norm: 0.6462 loss: 0.2032 +2023/06/02 03:22:48 - mmengine - INFO - Epoch(train) [4][4600/5758] lr: 8.1450e-05 eta: 8:47:28 time: 0.7498 data_time: 0.0018 memory: 28783 grad_norm: 0.6698 loss: 0.2014 +2023/06/02 03:24:11 - mmengine - INFO - Epoch(train) [4][4700/5758] lr: 8.1450e-05 eta: 8:45:51 time: 0.8809 data_time: 0.0023 memory: 28783 grad_norm: 1.2094 loss: 0.2062 +2023/06/02 03:24:33 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_lr1e-4_10epoch_20230601_215742 +2023/06/02 03:25:33 - mmengine - INFO - Epoch(train) [4][4800/5758] lr: 8.1450e-05 eta: 8:44:12 time: 0.7701 data_time: 0.0020 memory: 28783 grad_norm: 0.5823 loss: 0.2048 +2023/06/02 03:26:56 - mmengine - INFO - Epoch(train) [4][4900/5758] lr: 8.1450e-05 eta: 8:42:34 time: 0.9246 data_time: 0.0024 memory: 28783 grad_norm: 0.9493 loss: 0.2050 +2023/06/02 03:28:18 - mmengine - INFO - Epoch(train) [4][5000/5758] lr: 8.1450e-05 eta: 8:40:55 time: 0.8140 data_time: 0.0020 memory: 28783 grad_norm: 0.9327 loss: 0.2024 +2023/06/02 03:29:47 - mmengine - INFO - Epoch(train) [4][5100/5758] lr: 8.1450e-05 eta: 8:39:27 time: 0.9900 data_time: 0.0026 memory: 28783 grad_norm: 1.0285 loss: 0.2035 +2023/06/02 03:31:11 - mmengine - INFO - Epoch(train) [4][5200/5758] lr: 8.1450e-05 eta: 8:37:52 time: 0.8355 data_time: 0.0022 memory: 28783 grad_norm: 0.8105 loss: 0.2051 +2023/06/02 03:32:36 - mmengine - INFO - Epoch(train) [4][5300/5758] lr: 8.1450e-05 eta: 8:36:17 time: 0.9092 data_time: 0.0022 memory: 28783 grad_norm: 0.7004 loss: 0.2050 +2023/06/02 03:34:00 - mmengine - INFO - Epoch(train) [4][5400/5758] lr: 8.1450e-05 eta: 8:34:42 time: 0.8773 data_time: 0.0016 memory: 28783 grad_norm: 0.8098 loss: 0.2062 +2023/06/02 03:35:25 - mmengine - INFO - Epoch(train) [4][5500/5758] lr: 8.1450e-05 eta: 8:33:08 time: 0.8351 data_time: 0.0024 memory: 28783 grad_norm: 0.7755 loss: 0.2043 +2023/06/02 03:36:52 - mmengine - INFO - Epoch(train) [4][5600/5758] lr: 8.1450e-05 eta: 8:31:38 time: 0.8411 data_time: 0.0016 memory: 28783 grad_norm: 1.7324 loss: 0.2049 +2023/06/02 03:38:16 - mmengine - INFO - Epoch(train) [4][5700/5758] lr: 8.1450e-05 eta: 8:30:03 time: 0.8160 data_time: 0.0019 memory: 28783 grad_norm: 1.1104 loss: 0.2081 +2023/06/02 03:38:40 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_lr1e-4_10epoch_20230601_215742 +2023/06/02 03:39:07 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_lr1e-4_10epoch_20230601_215742 +2023/06/02 03:39:07 - mmengine - INFO - Saving checkpoint at 4 epochs +2023/06/02 03:39:26 - mmengine - INFO - Epoch(val) [4][16/16] accuracy/top1: 99.8990 single-label/precision_classwise: [100.0, 0.0] single-label/recall_classwise: [99.89899444580078, 0.0] single-label/f1-score_classwise: [99.94947814941406, 0.0] data_time: 0.2242 time: 0.6452 +2023/06/02 03:40:59 - mmengine - INFO - Epoch(train) [5][ 100/5758] lr: 6.8906e-05 eta: 8:27:50 time: 0.8387 data_time: 0.3319 memory: 28783 grad_norm: 0.5819 loss: 0.2046 +2023/06/02 03:42:26 - mmengine - INFO - Epoch(train) [5][ 200/5758] lr: 6.8906e-05 eta: 8:26:20 time: 0.7941 data_time: 0.2808 memory: 28783 grad_norm: 0.7127 loss: 0.2044 +2023/06/02 03:43:50 - mmengine - INFO - Epoch(train) [5][ 300/5758] lr: 6.8906e-05 eta: 8:24:44 time: 0.8685 data_time: 0.0027 memory: 28783 grad_norm: 0.9764 loss: 0.2047 +2023/06/02 03:45:15 - mmengine - INFO - Epoch(train) [5][ 400/5758] lr: 6.8906e-05 eta: 8:23:10 time: 0.8259 data_time: 0.0020 memory: 28783 grad_norm: 0.7020 loss: 0.2028 +2023/06/02 03:46:37 - mmengine - INFO - Epoch(train) [5][ 500/5758] lr: 6.8906e-05 eta: 8:21:32 time: 0.8996 data_time: 0.0025 memory: 28783 grad_norm: 0.7867 loss: 0.2028 +2023/06/02 03:48:00 - mmengine - INFO - Epoch(train) [5][ 600/5758] lr: 6.8906e-05 eta: 8:19:57 time: 0.9003 data_time: 0.0022 memory: 28783 grad_norm: 0.6545 loss: 0.2040 +2023/06/02 03:49:22 - mmengine - INFO - Epoch(train) [5][ 700/5758] lr: 6.8906e-05 eta: 8:18:19 time: 0.8825 data_time: 0.0031 memory: 28783 grad_norm: 0.7304 loss: 0.2015 +2023/06/02 03:50:44 - mmengine - INFO - Epoch(train) [5][ 800/5758] lr: 6.8906e-05 eta: 8:16:42 time: 0.7452 data_time: 0.0020 memory: 28783 grad_norm: 0.7447 loss: 0.2047 +2023/06/02 03:52:11 - mmengine - INFO - Epoch(train) [5][ 900/5758] lr: 6.8906e-05 eta: 8:15:12 time: 0.8373 data_time: 0.0016 memory: 28783 grad_norm: 1.0320 loss: 0.2068 +2023/06/02 03:53:08 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_lr1e-4_10epoch_20230601_215742 +2023/06/02 03:53:35 - mmengine - INFO - Epoch(train) [5][1000/5758] lr: 6.8906e-05 eta: 8:13:38 time: 0.9185 data_time: 0.0022 memory: 28783 grad_norm: 1.0076 loss: 0.2043 +2023/06/02 03:54:55 - mmengine - INFO - Epoch(train) [5][1100/5758] lr: 6.8906e-05 eta: 8:11:58 time: 0.8009 data_time: 0.0022 memory: 28783 grad_norm: 0.5286 loss: 0.2033 +2023/06/02 03:56:18 - mmengine - INFO - Epoch(train) [5][1200/5758] lr: 6.8906e-05 eta: 8:10:23 time: 0.8633 data_time: 0.0035 memory: 28783 grad_norm: 0.6689 loss: 0.2045 +2023/06/02 03:57:39 - mmengine - INFO - Epoch(train) [5][1300/5758] lr: 6.8906e-05 eta: 8:08:44 time: 0.7413 data_time: 0.0017 memory: 28783 grad_norm: 0.7286 loss: 0.2021 +2023/06/02 03:59:03 - mmengine - INFO - Epoch(train) [5][1400/5758] lr: 6.8906e-05 eta: 8:07:10 time: 0.8425 data_time: 0.0017 memory: 28783 grad_norm: 0.6850 loss: 0.2082 +2023/06/02 04:00:31 - mmengine - INFO - Epoch(train) [5][1500/5758] lr: 6.8906e-05 eta: 8:05:42 time: 0.8036 data_time: 0.0020 memory: 28783 grad_norm: 0.6095 loss: 0.2022 +2023/06/02 04:01:57 - mmengine - INFO - Epoch(train) [5][1600/5758] lr: 6.8906e-05 eta: 8:04:11 time: 0.7969 data_time: 0.0020 memory: 28783 grad_norm: 1.5473 loss: 0.2061 +2023/06/02 04:03:22 - mmengine - INFO - Epoch(train) [5][1700/5758] lr: 6.8906e-05 eta: 8:02:39 time: 0.9417 data_time: 0.0038 memory: 28783 grad_norm: 0.4208 loss: 0.2012 +2023/06/02 04:04:48 - mmengine - INFO - Epoch(train) [5][1800/5758] lr: 6.8906e-05 eta: 8:01:07 time: 0.8561 data_time: 0.0021 memory: 28783 grad_norm: 0.7695 loss: 0.2038 +2023/06/02 04:06:14 - mmengine - INFO - Epoch(train) [5][1900/5758] lr: 6.8906e-05 eta: 7:59:37 time: 0.9322 data_time: 0.0020 memory: 28783 grad_norm: 0.5742 loss: 0.2010 +2023/06/02 04:07:12 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_lr1e-4_10epoch_20230601_215742 +2023/06/02 04:07:40 - mmengine - INFO - Epoch(train) [5][2000/5758] lr: 6.8906e-05 eta: 7:58:05 time: 0.9280 data_time: 0.0025 memory: 28783 grad_norm: 0.7578 loss: 0.2017 +2023/06/02 04:09:03 - mmengine - INFO - Epoch(train) [5][2100/5758] lr: 6.8906e-05 eta: 7:56:31 time: 0.7954 data_time: 0.0021 memory: 28783 grad_norm: 0.7197 loss: 0.2020 +2023/06/02 04:10:32 - mmengine - INFO - Epoch(train) [5][2200/5758] lr: 6.8906e-05 eta: 7:55:03 time: 0.8733 data_time: 0.0017 memory: 28783 grad_norm: 0.4823 loss: 0.2026 +2023/06/02 04:12:01 - mmengine - INFO - Epoch(train) [5][2300/5758] lr: 6.8906e-05 eta: 7:53:37 time: 0.8383 data_time: 0.0020 memory: 28783 grad_norm: 0.9549 loss: 0.2040 +2023/06/02 04:13:29 - mmengine - INFO - Epoch(train) [5][2400/5758] lr: 6.8906e-05 eta: 7:52:08 time: 0.9125 data_time: 0.0024 memory: 28783 grad_norm: 0.8259 loss: 0.2060 +2023/06/02 04:14:54 - mmengine - INFO - Epoch(train) [5][2500/5758] lr: 6.8906e-05 eta: 7:50:36 time: 0.8700 data_time: 0.0019 memory: 28783 grad_norm: 0.5937 loss: 0.2035 +2023/06/02 04:16:21 - mmengine - INFO - Epoch(train) [5][2600/5758] lr: 6.8906e-05 eta: 7:49:07 time: 0.7872 data_time: 0.0019 memory: 28783 grad_norm: 0.7664 loss: 0.2034 +2023/06/02 04:17:49 - mmengine - INFO - Epoch(train) [5][2700/5758] lr: 6.8906e-05 eta: 7:47:39 time: 0.9325 data_time: 0.0025 memory: 28783 grad_norm: 0.9104 loss: 0.2006 +2023/06/02 04:19:13 - mmengine - INFO - Epoch(train) [5][2800/5758] lr: 6.8906e-05 eta: 7:46:06 time: 0.8318 data_time: 0.0019 memory: 28783 grad_norm: 0.6789 loss: 0.2045 +2023/06/02 04:20:38 - mmengine - INFO - Epoch(train) [5][2900/5758] lr: 6.8906e-05 eta: 7:44:33 time: 0.8249 data_time: 0.0020 memory: 28783 grad_norm: 0.8591 loss: 0.2021 +2023/06/02 04:21:33 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_lr1e-4_10epoch_20230601_215742 +2023/06/02 04:22:00 - mmengine - INFO - Epoch(train) [5][3000/5758] lr: 6.8906e-05 eta: 7:42:58 time: 0.9176 data_time: 0.1705 memory: 28783 grad_norm: 0.7385 loss: 0.2050 +2023/06/02 04:23:24 - mmengine - INFO - Epoch(train) [5][3100/5758] lr: 6.8906e-05 eta: 7:41:25 time: 0.8747 data_time: 0.1663 memory: 28783 grad_norm: 0.8209 loss: 0.2068 +2023/06/02 04:24:50 - mmengine - INFO - Epoch(train) [5][3200/5758] lr: 6.8906e-05 eta: 7:39:55 time: 0.9614 data_time: 0.0036 memory: 28783 grad_norm: 0.7996 loss: 0.2040 +2023/06/02 04:26:10 - mmengine - INFO - Epoch(train) [5][3300/5758] lr: 6.8906e-05 eta: 7:38:18 time: 0.9425 data_time: 0.0031 memory: 28783 grad_norm: 0.8042 loss: 0.2005 +2023/06/02 04:27:36 - mmengine - INFO - Epoch(train) [5][3400/5758] lr: 6.8906e-05 eta: 7:36:47 time: 0.8653 data_time: 0.0017 memory: 28783 grad_norm: 0.5531 loss: 0.2041 +2023/06/02 04:29:00 - mmengine - INFO - Epoch(train) [5][3500/5758] lr: 6.8906e-05 eta: 7:35:15 time: 0.8249 data_time: 0.0022 memory: 28783 grad_norm: 0.5613 loss: 0.2022 +2023/06/02 04:30:23 - mmengine - INFO - Epoch(train) [5][3600/5758] lr: 6.8906e-05 eta: 7:33:40 time: 0.7633 data_time: 0.0016 memory: 28783 grad_norm: 0.8798 loss: 0.2034 +2023/06/02 04:31:45 - mmengine - INFO - Epoch(train) [5][3700/5758] lr: 6.8906e-05 eta: 7:32:06 time: 0.8931 data_time: 0.0037 memory: 28783 grad_norm: 1.0159 loss: 0.2023 +2023/06/02 04:33:09 - mmengine - INFO - Epoch(train) [5][3800/5758] lr: 6.8906e-05 eta: 7:30:33 time: 0.7910 data_time: 0.0019 memory: 28783 grad_norm: 0.6107 loss: 0.2039 +2023/06/02 04:34:33 - mmengine - INFO - Epoch(train) [5][3900/5758] lr: 6.8906e-05 eta: 7:29:01 time: 0.9180 data_time: 0.0027 memory: 28783 grad_norm: 0.8399 loss: 0.2019 +2023/06/02 04:35:31 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_lr1e-4_10epoch_20230601_215742 +2023/06/02 04:35:57 - mmengine - INFO - Epoch(train) [5][4000/5758] lr: 6.8906e-05 eta: 7:27:28 time: 0.8257 data_time: 0.0018 memory: 28783 grad_norm: 0.6113 loss: 0.2032 +2023/06/02 04:37:23 - mmengine - INFO - Epoch(train) [5][4100/5758] lr: 6.8906e-05 eta: 7:25:59 time: 0.8880 data_time: 0.0027 memory: 28783 grad_norm: 0.6875 loss: 0.2024 +2023/06/02 04:38:48 - mmengine - INFO - Epoch(train) [5][4200/5758] lr: 6.8906e-05 eta: 7:24:27 time: 0.8801 data_time: 0.0017 memory: 28783 grad_norm: 0.6870 loss: 0.2024 +2023/06/02 04:40:12 - mmengine - INFO - Epoch(train) [5][4300/5758] lr: 6.8906e-05 eta: 7:22:55 time: 0.8372 data_time: 0.0035 memory: 28783 grad_norm: 0.6702 loss: 0.2014 +2023/06/02 04:41:36 - mmengine - INFO - Epoch(train) [5][4400/5758] lr: 6.8906e-05 eta: 7:21:23 time: 0.9502 data_time: 0.0029 memory: 28783 grad_norm: 0.5011 loss: 0.2011 +2023/06/02 04:43:00 - mmengine - INFO - Epoch(train) [5][4500/5758] lr: 6.8906e-05 eta: 7:19:51 time: 0.8282 data_time: 0.0019 memory: 28783 grad_norm: 0.9513 loss: 0.2081 +2023/06/02 04:44:24 - mmengine - INFO - Epoch(train) [5][4600/5758] lr: 6.8906e-05 eta: 7:18:19 time: 0.8092 data_time: 0.0019 memory: 28783 grad_norm: 0.7864 loss: 0.2039 +2023/06/02 04:45:49 - mmengine - INFO - Epoch(train) [5][4700/5758] lr: 6.8906e-05 eta: 7:16:48 time: 0.8482 data_time: 0.0022 memory: 28783 grad_norm: 1.2227 loss: 0.2027 +2023/06/02 04:47:16 - mmengine - INFO - Epoch(train) [5][4800/5758] lr: 6.8906e-05 eta: 7:15:20 time: 0.8230 data_time: 0.0021 memory: 28783 grad_norm: 0.5874 loss: 0.2017 +2023/06/02 04:48:43 - mmengine - INFO - Epoch(train) [5][4900/5758] lr: 6.8906e-05 eta: 7:13:51 time: 0.8157 data_time: 0.0017 memory: 28783 grad_norm: 0.3926 loss: 0.2026 +2023/06/02 04:49:41 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_lr1e-4_10epoch_20230601_215742 +2023/06/02 04:50:10 - mmengine - INFO - Epoch(train) [5][5000/5758] lr: 6.8906e-05 eta: 7:12:22 time: 0.9590 data_time: 0.0027 memory: 28783 grad_norm: 0.6538 loss: 0.2029 +2023/06/02 04:51:34 - mmengine - INFO - Epoch(train) [5][5100/5758] lr: 6.8906e-05 eta: 7:10:50 time: 0.7456 data_time: 0.0014 memory: 28783 grad_norm: 0.9114 loss: 0.2015 +2023/06/02 04:52:57 - mmengine - INFO - Epoch(train) [5][5200/5758] lr: 6.8906e-05 eta: 7:09:18 time: 0.8601 data_time: 0.0017 memory: 28783 grad_norm: 0.9279 loss: 0.2036 +2023/06/02 04:54:21 - mmengine - INFO - Epoch(train) [5][5300/5758] lr: 6.8906e-05 eta: 7:07:46 time: 0.9092 data_time: 0.0018 memory: 28783 grad_norm: 0.5491 loss: 0.2018 +2023/06/02 04:55:48 - mmengine - INFO - Epoch(train) [5][5400/5758] lr: 6.8906e-05 eta: 7:06:17 time: 0.8719 data_time: 0.0016 memory: 28783 grad_norm: 0.6158 loss: 0.2026 +2023/06/02 04:57:10 - mmengine - INFO - Epoch(train) [5][5500/5758] lr: 6.8906e-05 eta: 7:04:44 time: 0.8094 data_time: 0.0020 memory: 28783 grad_norm: 0.6290 loss: 0.2043 +2023/06/02 04:58:35 - mmengine - INFO - Epoch(train) [5][5600/5758] lr: 6.8906e-05 eta: 7:03:13 time: 0.8368 data_time: 0.0025 memory: 28783 grad_norm: 0.5193 loss: 0.2011 +2023/06/02 04:59:56 - mmengine - INFO - Epoch(train) [5][5700/5758] lr: 6.8906e-05 eta: 7:01:39 time: 0.8017 data_time: 0.0018 memory: 28783 grad_norm: 0.8042 loss: 0.2045 +2023/06/02 05:00:45 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_lr1e-4_10epoch_20230601_215742 +2023/06/02 05:00:45 - mmengine - INFO - Saving checkpoint at 5 epochs +2023/06/02 05:01:06 - mmengine - INFO - Epoch(val) [5][16/16] accuracy/top1: 99.7917 single-label/precision_classwise: [100.0, 0.0] single-label/recall_classwise: [99.79167938232422, 0.0] single-label/f1-score_classwise: [99.8957290649414, 0.0] data_time: 0.2498 time: 0.6703 +2023/06/02 05:02:35 - mmengine - INFO - Epoch(train) [6][ 100/5758] lr: 5.5000e-05 eta: 6:59:20 time: 0.7802 data_time: 0.2367 memory: 28783 grad_norm: 0.6579 loss: 0.2037 +2023/06/02 05:03:56 - mmengine - INFO - Epoch(train) [6][ 200/5758] lr: 5.5000e-05 eta: 6:57:45 time: 0.7921 data_time: 0.2342 memory: 28783 grad_norm: 0.4249 loss: 0.2006 +2023/06/02 05:04:04 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_lr1e-4_10epoch_20230601_215742 +2023/06/02 05:05:21 - mmengine - INFO - Epoch(train) [6][ 300/5758] lr: 5.5000e-05 eta: 6:56:15 time: 0.8250 data_time: 0.1533 memory: 28783 grad_norm: 0.5292 loss: 0.2023 +2023/06/02 05:06:42 - mmengine - INFO - Epoch(train) [6][ 400/5758] lr: 5.5000e-05 eta: 6:54:41 time: 0.8863 data_time: 0.3480 memory: 28783 grad_norm: 0.4441 loss: 0.2009 +2023/06/02 05:08:06 - mmengine - INFO - Epoch(train) [6][ 500/5758] lr: 5.5000e-05 eta: 6:53:10 time: 0.8637 data_time: 0.3591 memory: 28783 grad_norm: 0.5437 loss: 0.2016 +2023/06/02 05:09:29 - mmengine - INFO - Epoch(train) [6][ 600/5758] lr: 5.5000e-05 eta: 6:51:37 time: 0.8652 data_time: 0.3502 memory: 28783 grad_norm: 0.6725 loss: 0.2018 +2023/06/02 05:10:51 - mmengine - INFO - Epoch(train) [6][ 700/5758] lr: 5.5000e-05 eta: 6:50:05 time: 0.7572 data_time: 0.2341 memory: 28783 grad_norm: 0.6596 loss: 0.2032 +2023/06/02 05:12:13 - mmengine - INFO - Epoch(train) [6][ 800/5758] lr: 5.5000e-05 eta: 6:48:32 time: 0.7345 data_time: 0.2281 memory: 28783 grad_norm: 0.7802 loss: 0.2068 +2023/06/02 05:13:34 - mmengine - INFO - Epoch(train) [6][ 900/5758] lr: 5.5000e-05 eta: 6:46:58 time: 0.7100 data_time: 0.1906 memory: 28783 grad_norm: 0.4643 loss: 0.2019 +2023/06/02 05:14:56 - mmengine - INFO - Epoch(train) [6][1000/5758] lr: 5.5000e-05 eta: 6:45:25 time: 0.8915 data_time: 0.3219 memory: 28783 grad_norm: 0.7992 loss: 0.2019 +2023/06/02 05:16:17 - mmengine - INFO - Epoch(train) [6][1100/5758] lr: 5.5000e-05 eta: 6:43:52 time: 0.7961 data_time: 0.1744 memory: 28783 grad_norm: 0.6372 loss: 0.2015 +2023/06/02 05:17:39 - mmengine - INFO - Epoch(train) [6][1200/5758] lr: 5.5000e-05 eta: 6:42:19 time: 0.7871 data_time: 0.0851 memory: 28783 grad_norm: 0.4634 loss: 0.2030 +2023/06/02 05:17:47 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_lr1e-4_10epoch_20230601_215742 +2023/06/02 05:19:04 - mmengine - INFO - Epoch(train) [6][1300/5758] lr: 5.5000e-05 eta: 6:40:49 time: 0.9774 data_time: 0.0025 memory: 28783 grad_norm: 0.5535 loss: 0.2016 +2023/06/02 05:20:27 - mmengine - INFO - Epoch(train) [6][1400/5758] lr: 5.5000e-05 eta: 6:39:18 time: 0.7677 data_time: 0.0022 memory: 28783 grad_norm: 0.6899 loss: 0.2015 +2023/06/02 05:21:48 - mmengine - INFO - Epoch(train) [6][1500/5758] lr: 5.5000e-05 eta: 6:37:44 time: 0.7385 data_time: 0.0018 memory: 28783 grad_norm: 0.7161 loss: 0.2032 +2023/06/02 05:23:12 - mmengine - INFO - Epoch(train) [6][1600/5758] lr: 5.5000e-05 eta: 6:36:14 time: 0.8557 data_time: 0.0021 memory: 28783 grad_norm: 0.5423 loss: 0.1994 +2023/06/02 05:24:31 - mmengine - INFO - Epoch(train) [6][1700/5758] lr: 5.5000e-05 eta: 6:34:39 time: 0.7191 data_time: 0.0019 memory: 28783 grad_norm: 0.4187 loss: 0.2011 +2023/06/02 05:25:52 - mmengine - INFO - Epoch(train) [6][1800/5758] lr: 5.5000e-05 eta: 6:33:06 time: 0.8347 data_time: 0.0019 memory: 28783 grad_norm: 0.5010 loss: 0.2029 +2023/06/02 05:27:15 - mmengine - INFO - Epoch(train) [6][1900/5758] lr: 5.5000e-05 eta: 6:31:35 time: 0.9516 data_time: 0.0056 memory: 28783 grad_norm: 0.5716 loss: 0.2019 +2023/06/02 05:28:33 - mmengine - INFO - Epoch(train) [6][2000/5758] lr: 5.5000e-05 eta: 6:29:59 time: 0.8388 data_time: 0.0019 memory: 28783 grad_norm: 0.4385 loss: 0.1999 +2023/06/02 05:29:55 - mmengine - INFO - Epoch(train) [6][2100/5758] lr: 5.5000e-05 eta: 6:28:27 time: 0.8487 data_time: 0.0027 memory: 28783 grad_norm: 0.4062 loss: 0.2020 +2023/06/02 05:31:15 - mmengine - INFO - Epoch(train) [6][2200/5758] lr: 5.5000e-05 eta: 6:26:54 time: 0.7736 data_time: 0.0021 memory: 28783 grad_norm: 0.6403 loss: 0.2013 +2023/06/02 05:31:23 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_lr1e-4_10epoch_20230601_215742 +2023/06/02 05:32:37 - mmengine - INFO - Epoch(train) [6][2300/5758] lr: 5.5000e-05 eta: 6:25:22 time: 0.8640 data_time: 0.0025 memory: 28783 grad_norm: 0.6221 loss: 0.2036 +2023/06/02 05:34:02 - mmengine - INFO - Epoch(train) [6][2400/5758] lr: 5.5000e-05 eta: 6:23:53 time: 0.8603 data_time: 0.0023 memory: 28783 grad_norm: 0.7112 loss: 0.2028 +2023/06/02 05:35:28 - mmengine - INFO - Epoch(train) [6][2500/5758] lr: 5.5000e-05 eta: 6:22:24 time: 1.0009 data_time: 0.0031 memory: 28783 grad_norm: 0.6034 loss: 0.2050 +2023/06/02 05:36:51 - mmengine - INFO - Epoch(train) [6][2600/5758] lr: 5.5000e-05 eta: 6:20:53 time: 0.7816 data_time: 0.0017 memory: 28783 grad_norm: 0.5462 loss: 0.2021 +2023/06/02 05:38:16 - mmengine - INFO - Epoch(train) [6][2700/5758] lr: 5.5000e-05 eta: 6:19:24 time: 0.8270 data_time: 0.0017 memory: 28783 grad_norm: 0.4831 loss: 0.2015 +2023/06/02 05:39:43 - mmengine - INFO - Epoch(train) [6][2800/5758] lr: 5.5000e-05 eta: 6:17:57 time: 0.8349 data_time: 0.0022 memory: 28783 grad_norm: 0.5505 loss: 0.2005 +2023/06/02 05:41:06 - mmengine - INFO - Epoch(train) [6][2900/5758] lr: 5.5000e-05 eta: 6:16:26 time: 0.8211 data_time: 0.0018 memory: 28783 grad_norm: 0.6400 loss: 0.2014 +2023/06/02 05:42:28 - mmengine - INFO - Epoch(train) [6][3000/5758] lr: 5.5000e-05 eta: 6:14:55 time: 0.7560 data_time: 0.0017 memory: 28783 grad_norm: 0.8682 loss: 0.2027 +2023/06/02 05:43:53 - mmengine - INFO - Epoch(train) [6][3100/5758] lr: 5.5000e-05 eta: 6:13:25 time: 0.7682 data_time: 0.0022 memory: 28783 grad_norm: 0.3733 loss: 0.2024 +2023/06/02 05:45:14 - mmengine - INFO - Epoch(train) [6][3200/5758] lr: 5.5000e-05 eta: 6:11:54 time: 0.8148 data_time: 0.0019 memory: 28783 grad_norm: 0.4945 loss: 0.2040 +2023/06/02 05:45:23 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_lr1e-4_10epoch_20230601_215742 +2023/06/02 05:46:41 - mmengine - INFO - Epoch(train) [6][3300/5758] lr: 5.5000e-05 eta: 6:10:25 time: 0.8359 data_time: 0.0015 memory: 28783 grad_norm: 0.5048 loss: 0.2020 +2023/06/02 05:48:07 - mmengine - INFO - Epoch(train) [6][3400/5758] lr: 5.5000e-05 eta: 6:08:57 time: 0.7922 data_time: 0.0016 memory: 28783 grad_norm: 0.6258 loss: 0.2005 +2023/06/02 05:49:27 - mmengine - INFO - Epoch(train) [6][3500/5758] lr: 5.5000e-05 eta: 6:07:25 time: 0.7749 data_time: 0.0018 memory: 28783 grad_norm: 0.5815 loss: 0.2016 +2023/06/02 05:50:51 - mmengine - INFO - Epoch(train) [6][3600/5758] lr: 5.5000e-05 eta: 6:05:56 time: 0.7863 data_time: 0.0018 memory: 28783 grad_norm: 0.6044 loss: 0.2008 +2023/06/02 05:52:12 - mmengine - INFO - Epoch(train) [6][3700/5758] lr: 5.5000e-05 eta: 6:04:24 time: 0.7167 data_time: 0.0019 memory: 28783 grad_norm: 0.7972 loss: 0.2037 +2023/06/02 05:53:34 - mmengine - INFO - Epoch(train) [6][3800/5758] lr: 5.5000e-05 eta: 6:02:52 time: 0.7763 data_time: 0.0021 memory: 28783 grad_norm: 0.4211 loss: 0.2002 +2023/06/02 05:54:59 - mmengine - INFO - Epoch(train) [6][3900/5758] lr: 5.5000e-05 eta: 6:01:23 time: 0.8502 data_time: 0.0018 memory: 28783 grad_norm: 0.5495 loss: 0.2040 +2023/06/02 05:56:22 - mmengine - INFO - Epoch(train) [6][4000/5758] lr: 5.5000e-05 eta: 5:59:53 time: 0.8198 data_time: 0.0033 memory: 28783 grad_norm: 0.6393 loss: 0.2041 +2023/06/02 05:57:48 - mmengine - INFO - Epoch(train) [6][4100/5758] lr: 5.5000e-05 eta: 5:58:25 time: 0.7753 data_time: 0.0019 memory: 28783 grad_norm: 0.5554 loss: 0.2019 +2023/06/02 05:59:14 - mmengine - INFO - Epoch(train) [6][4200/5758] lr: 5.5000e-05 eta: 5:56:57 time: 0.9092 data_time: 0.0032 memory: 28783 grad_norm: 0.3656 loss: 0.1995 +2023/06/02 05:59:23 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_lr1e-4_10epoch_20230601_215742 +2023/06/02 06:00:39 - mmengine - INFO - Epoch(train) [6][4300/5758] lr: 5.5000e-05 eta: 5:55:28 time: 0.7857 data_time: 0.0016 memory: 28783 grad_norm: 0.5958 loss: 0.2008 +2023/06/02 06:02:03 - mmengine - INFO - Epoch(train) [6][4400/5758] lr: 5.5000e-05 eta: 5:53:59 time: 0.7801 data_time: 0.0016 memory: 28783 grad_norm: 0.7053 loss: 0.2020 +2023/06/02 06:03:25 - mmengine - INFO - Epoch(train) [6][4500/5758] lr: 5.5000e-05 eta: 5:52:28 time: 0.9589 data_time: 0.0024 memory: 28783 grad_norm: 0.4131 loss: 0.2018 +2023/06/02 06:04:51 - mmengine - INFO - Epoch(train) [6][4600/5758] lr: 5.5000e-05 eta: 5:51:00 time: 0.9160 data_time: 0.0024 memory: 28783 grad_norm: 0.8198 loss: 0.2016 +2023/06/02 06:06:14 - mmengine - INFO - Epoch(train) [6][4700/5758] lr: 5.5000e-05 eta: 5:49:31 time: 0.8429 data_time: 0.0017 memory: 28783 grad_norm: 0.5498 loss: 0.2022 +2023/06/02 06:07:37 - mmengine - INFO - Epoch(train) [6][4800/5758] lr: 5.5000e-05 eta: 5:48:01 time: 0.8233 data_time: 0.0019 memory: 28783 grad_norm: 0.9303 loss: 0.2028 +2023/06/02 06:09:00 - mmengine - INFO - Epoch(train) [6][4900/5758] lr: 5.5000e-05 eta: 5:46:31 time: 0.8907 data_time: 0.0023 memory: 28783 grad_norm: 0.5280 loss: 0.2047 +2023/06/02 06:10:25 - mmengine - INFO - Epoch(train) [6][5000/5758] lr: 5.5000e-05 eta: 5:45:02 time: 0.7626 data_time: 0.0016 memory: 28783 grad_norm: 0.3647 loss: 0.2016 +2023/06/02 06:11:49 - mmengine - INFO - Epoch(train) [6][5100/5758] lr: 5.5000e-05 eta: 5:43:33 time: 0.8343 data_time: 0.0019 memory: 28783 grad_norm: 0.3085 loss: 0.2019 +2023/06/02 06:13:18 - mmengine - INFO - Epoch(train) [6][5200/5758] lr: 5.5000e-05 eta: 5:42:08 time: 0.9068 data_time: 0.0020 memory: 28783 grad_norm: 0.6019 loss: 0.2012 +2023/06/02 06:13:27 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_lr1e-4_10epoch_20230601_215742 +2023/06/02 06:14:48 - mmengine - INFO - Epoch(train) [6][5300/5758] lr: 5.5000e-05 eta: 5:40:42 time: 0.9851 data_time: 0.0025 memory: 28783 grad_norm: 0.5749 loss: 0.2032 +2023/06/02 06:16:19 - mmengine - INFO - Epoch(train) [6][5400/5758] lr: 5.5000e-05 eta: 5:39:18 time: 0.9787 data_time: 0.0029 memory: 28783 grad_norm: 0.4084 loss: 0.2027 +2023/06/02 06:17:45 - mmengine - INFO - Epoch(train) [6][5500/5758] lr: 5.5000e-05 eta: 5:37:50 time: 0.8776 data_time: 0.0018 memory: 28783 grad_norm: 0.4705 loss: 0.2019 +2023/06/02 06:19:15 - mmengine - INFO - Epoch(train) [6][5600/5758] lr: 5.5000e-05 eta: 5:36:25 time: 1.0334 data_time: 0.0032 memory: 28783 grad_norm: 1.0642 loss: 0.2076 +2023/06/02 06:20:41 - mmengine - INFO - Epoch(train) [6][5700/5758] lr: 5.5000e-05 eta: 5:34:58 time: 0.7545 data_time: 0.0016 memory: 28783 grad_norm: 0.5163 loss: 0.2011 +2023/06/02 06:21:30 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_lr1e-4_10epoch_20230601_215742 +2023/06/02 06:21:30 - mmengine - INFO - Saving checkpoint at 6 epochs +2023/06/02 06:21:50 - mmengine - INFO - Epoch(val) [6][16/16] accuracy/top1: 99.5455 single-label/precision_classwise: [100.0, 0.0] single-label/recall_classwise: [99.54548645019531, 0.0] single-label/f1-score_classwise: [99.77222442626953, 0.0] data_time: 0.2649 time: 0.6808 +2023/06/02 06:23:26 - mmengine - INFO - Epoch(train) [7][ 100/5758] lr: 4.1094e-05 eta: 5:32:45 time: 0.8140 data_time: 0.2570 memory: 28783 grad_norm: 0.5305 loss: 0.2038 +2023/06/02 06:24:54 - mmengine - INFO - Epoch(train) [7][ 200/5758] lr: 4.1094e-05 eta: 5:31:18 time: 0.9949 data_time: 0.3320 memory: 28783 grad_norm: 0.6574 loss: 0.2009 +2023/06/02 06:26:20 - mmengine - INFO - Epoch(train) [7][ 300/5758] lr: 4.1094e-05 eta: 5:29:51 time: 0.8849 data_time: 0.0026 memory: 28783 grad_norm: 0.6975 loss: 0.2028 +2023/06/02 06:27:47 - mmengine - INFO - Epoch(train) [7][ 400/5758] lr: 4.1094e-05 eta: 5:28:23 time: 0.8067 data_time: 0.0019 memory: 28783 grad_norm: 0.7316 loss: 0.2006 +2023/06/02 06:28:31 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_lr1e-4_10epoch_20230601_215742 +2023/06/02 06:29:13 - mmengine - INFO - Epoch(train) [7][ 500/5758] lr: 4.1094e-05 eta: 5:26:56 time: 0.8141 data_time: 0.0017 memory: 28783 grad_norm: 0.3904 loss: 0.2000 +2023/06/02 06:30:34 - mmengine - INFO - Epoch(train) [7][ 600/5758] lr: 4.1094e-05 eta: 5:25:24 time: 0.7880 data_time: 0.0019 memory: 28783 grad_norm: 0.7536 loss: 0.2011 +2023/06/02 06:31:58 - mmengine - INFO - Epoch(train) [7][ 700/5758] lr: 4.1094e-05 eta: 5:23:56 time: 0.9132 data_time: 0.0023 memory: 28783 grad_norm: 0.6121 loss: 0.2021 +2023/06/02 06:33:21 - mmengine - INFO - Epoch(train) [7][ 800/5758] lr: 4.1094e-05 eta: 5:22:26 time: 0.8503 data_time: 0.0035 memory: 28783 grad_norm: 0.5684 loss: 0.2025 +2023/06/02 06:34:42 - mmengine - INFO - Epoch(train) [7][ 900/5758] lr: 4.1094e-05 eta: 5:20:55 time: 0.8818 data_time: 0.0025 memory: 28783 grad_norm: 0.3548 loss: 0.2009 +2023/06/02 06:36:04 - mmengine - INFO - Epoch(train) [7][1000/5758] lr: 4.1094e-05 eta: 5:19:25 time: 0.7270 data_time: 0.0016 memory: 28783 grad_norm: 0.5783 loss: 0.2024 +2023/06/02 06:37:28 - mmengine - INFO - Epoch(train) [7][1100/5758] lr: 4.1094e-05 eta: 5:17:56 time: 0.8768 data_time: 0.0766 memory: 28783 grad_norm: 0.4349 loss: 0.2009 +2023/06/02 06:38:49 - mmengine - INFO - Epoch(train) [7][1200/5758] lr: 4.1094e-05 eta: 5:16:26 time: 0.7718 data_time: 0.0018 memory: 28783 grad_norm: 0.3603 loss: 0.2000 +2023/06/02 06:40:13 - mmengine - INFO - Epoch(train) [7][1300/5758] lr: 4.1094e-05 eta: 5:14:57 time: 0.8286 data_time: 0.0016 memory: 28783 grad_norm: 0.6901 loss: 0.2033 +2023/06/02 06:41:38 - mmengine - INFO - Epoch(train) [7][1400/5758] lr: 4.1094e-05 eta: 5:13:29 time: 0.7771 data_time: 0.0026 memory: 28783 grad_norm: 0.5387 loss: 0.2012 +2023/06/02 06:42:22 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_lr1e-4_10epoch_20230601_215742 +2023/06/02 06:43:02 - mmengine - INFO - Epoch(train) [7][1500/5758] lr: 4.1094e-05 eta: 5:12:00 time: 0.8777 data_time: 0.0018 memory: 28783 grad_norm: 0.5511 loss: 0.2002 +2023/06/02 06:44:28 - mmengine - INFO - Epoch(train) [7][1600/5758] lr: 4.1094e-05 eta: 5:10:33 time: 0.8440 data_time: 0.0018 memory: 28783 grad_norm: 0.4498 loss: 0.2009 +2023/06/02 06:45:50 - mmengine - INFO - Epoch(train) [7][1700/5758] lr: 4.1094e-05 eta: 5:09:03 time: 0.8718 data_time: 0.0027 memory: 28783 grad_norm: 0.3717 loss: 0.1993 +2023/06/02 06:47:13 - mmengine - INFO - Epoch(train) [7][1800/5758] lr: 4.1094e-05 eta: 5:07:33 time: 0.7626 data_time: 0.0017 memory: 28783 grad_norm: 0.5331 loss: 0.2012 +2023/06/02 06:48:39 - mmengine - INFO - Epoch(train) [7][1900/5758] lr: 4.1094e-05 eta: 5:06:06 time: 0.9333 data_time: 0.0026 memory: 28783 grad_norm: 0.2660 loss: 0.1999 +2023/06/02 06:50:05 - mmengine - INFO - Epoch(train) [7][2000/5758] lr: 4.1094e-05 eta: 5:04:39 time: 0.8786 data_time: 0.0017 memory: 28783 grad_norm: 0.3881 loss: 0.2011 +2023/06/02 06:51:30 - mmengine - INFO - Epoch(train) [7][2100/5758] lr: 4.1094e-05 eta: 5:03:11 time: 0.8532 data_time: 0.0018 memory: 28783 grad_norm: 0.4851 loss: 0.1994 +2023/06/02 06:52:58 - mmengine - INFO - Epoch(train) [7][2200/5758] lr: 4.1094e-05 eta: 5:01:44 time: 0.9557 data_time: 0.0027 memory: 28783 grad_norm: 0.8457 loss: 0.2006 +2023/06/02 06:54:20 - mmengine - INFO - Epoch(train) [7][2300/5758] lr: 4.1094e-05 eta: 5:00:15 time: 0.7914 data_time: 0.0020 memory: 28783 grad_norm: 0.5865 loss: 0.2002 +2023/06/02 06:55:50 - mmengine - INFO - Epoch(train) [7][2400/5758] lr: 4.1094e-05 eta: 4:58:49 time: 0.7526 data_time: 0.0017 memory: 28783 grad_norm: 0.4370 loss: 0.2012 +2023/06/02 06:56:34 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_lr1e-4_10epoch_20230601_215742 +2023/06/02 06:57:17 - mmengine - INFO - Epoch(train) [7][2500/5758] lr: 4.1094e-05 eta: 4:57:22 time: 0.9600 data_time: 0.0035 memory: 28783 grad_norm: 0.4532 loss: 0.1995 +2023/06/02 06:58:41 - mmengine - INFO - Epoch(train) [7][2600/5758] lr: 4.1094e-05 eta: 4:55:54 time: 0.8015 data_time: 0.0020 memory: 28783 grad_norm: 0.5822 loss: 0.1996 +2023/06/02 07:00:08 - mmengine - INFO - Epoch(train) [7][2700/5758] lr: 4.1094e-05 eta: 4:54:27 time: 0.8951 data_time: 0.0027 memory: 28783 grad_norm: 0.3273 loss: 0.2002 +2023/06/02 07:01:34 - mmengine - INFO - Epoch(train) [7][2800/5758] lr: 4.1094e-05 eta: 4:53:00 time: 0.7437 data_time: 0.0020 memory: 28783 grad_norm: 0.3746 loss: 0.2001 +2023/06/02 07:02:58 - mmengine - INFO - Epoch(train) [7][2900/5758] lr: 4.1094e-05 eta: 4:51:31 time: 0.9008 data_time: 0.0032 memory: 28783 grad_norm: 0.4760 loss: 0.2000 +2023/06/02 07:04:23 - mmengine - INFO - Epoch(train) [7][3000/5758] lr: 4.1094e-05 eta: 4:50:04 time: 0.7836 data_time: 0.0018 memory: 28783 grad_norm: 0.5593 loss: 0.1994 +2023/06/02 07:05:46 - mmengine - INFO - Epoch(train) [7][3100/5758] lr: 4.1094e-05 eta: 4:48:34 time: 0.8986 data_time: 0.0028 memory: 28783 grad_norm: 0.4368 loss: 0.2001 +2023/06/02 07:07:08 - mmengine - INFO - Epoch(train) [7][3200/5758] lr: 4.1094e-05 eta: 4:47:05 time: 0.8397 data_time: 0.0020 memory: 28783 grad_norm: 0.3960 loss: 0.1997 +2023/06/02 07:08:33 - mmengine - INFO - Epoch(train) [7][3300/5758] lr: 4.1094e-05 eta: 4:45:37 time: 0.8755 data_time: 0.0031 memory: 28783 grad_norm: 0.4780 loss: 0.2010 +2023/06/02 07:09:56 - mmengine - INFO - Epoch(train) [7][3400/5758] lr: 4.1094e-05 eta: 4:44:09 time: 0.9036 data_time: 0.0029 memory: 28783 grad_norm: 0.4984 loss: 0.1993 +2023/06/02 07:10:39 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_lr1e-4_10epoch_20230601_215742 +2023/06/02 07:11:18 - mmengine - INFO - Epoch(train) [7][3500/5758] lr: 4.1094e-05 eta: 4:42:39 time: 0.7434 data_time: 0.0017 memory: 28783 grad_norm: 0.5670 loss: 0.2009 +2023/06/02 07:12:43 - mmengine - INFO - Epoch(train) [7][3600/5758] lr: 4.1094e-05 eta: 4:41:11 time: 0.8600 data_time: 0.0022 memory: 28783 grad_norm: 0.4579 loss: 0.2013 +2023/06/02 07:14:06 - mmengine - INFO - Epoch(train) [7][3700/5758] lr: 4.1094e-05 eta: 4:39:42 time: 0.7751 data_time: 0.0016 memory: 28783 grad_norm: 0.4868 loss: 0.1997 +2023/06/02 07:15:31 - mmengine - INFO - Epoch(train) [7][3800/5758] lr: 4.1094e-05 eta: 4:38:15 time: 0.9615 data_time: 0.0042 memory: 28783 grad_norm: 0.4641 loss: 0.2013 +2023/06/02 07:16:54 - mmengine - INFO - Epoch(train) [7][3900/5758] lr: 4.1094e-05 eta: 4:36:46 time: 0.9268 data_time: 0.0021 memory: 28783 grad_norm: 0.3765 loss: 0.1999 +2023/06/02 07:18:16 - mmengine - INFO - Epoch(train) [7][4000/5758] lr: 4.1094e-05 eta: 4:35:17 time: 0.7984 data_time: 0.0020 memory: 28783 grad_norm: 0.4711 loss: 0.1991 +2023/06/02 07:19:40 - mmengine - INFO - Epoch(train) [7][4100/5758] lr: 4.1094e-05 eta: 4:33:49 time: 0.7945 data_time: 0.0018 memory: 28783 grad_norm: 0.3720 loss: 0.1995 +2023/06/02 07:21:08 - mmengine - INFO - Epoch(train) [7][4200/5758] lr: 4.1094e-05 eta: 4:32:22 time: 0.8917 data_time: 0.0017 memory: 28783 grad_norm: 0.6079 loss: 0.2025 +2023/06/02 07:22:31 - mmengine - INFO - Epoch(train) [7][4300/5758] lr: 4.1094e-05 eta: 4:30:54 time: 0.9038 data_time: 0.0035 memory: 28783 grad_norm: 0.4597 loss: 0.2032 +2023/06/02 07:23:57 - mmengine - INFO - Epoch(train) [7][4400/5758] lr: 4.1094e-05 eta: 4:29:27 time: 0.8065 data_time: 0.0019 memory: 28783 grad_norm: 0.6150 loss: 0.2004 +2023/06/02 07:24:42 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_lr1e-4_10epoch_20230601_215742 +2023/06/02 07:25:22 - mmengine - INFO - Epoch(train) [7][4500/5758] lr: 4.1094e-05 eta: 4:27:59 time: 0.9513 data_time: 0.0019 memory: 28783 grad_norm: 0.2860 loss: 0.2002 +2023/06/02 07:26:48 - mmengine - INFO - Epoch(train) [7][4600/5758] lr: 4.1094e-05 eta: 4:26:32 time: 0.8299 data_time: 0.0016 memory: 28783 grad_norm: 0.4855 loss: 0.1995 +2023/06/02 07:28:13 - mmengine - INFO - Epoch(train) [7][4700/5758] lr: 4.1094e-05 eta: 4:25:05 time: 0.7714 data_time: 0.0017 memory: 28783 grad_norm: 0.3842 loss: 0.2023 +2023/06/02 07:29:37 - mmengine - INFO - Epoch(train) [7][4800/5758] lr: 4.1094e-05 eta: 4:23:36 time: 0.8272 data_time: 0.0017 memory: 28783 grad_norm: 0.3703 loss: 0.2003 +2023/06/02 07:31:03 - mmengine - INFO - Epoch(train) [7][4900/5758] lr: 4.1094e-05 eta: 4:22:09 time: 0.7821 data_time: 0.0018 memory: 28783 grad_norm: 0.4352 loss: 0.2002 +2023/06/02 07:32:29 - mmengine - INFO - Epoch(train) [7][5000/5758] lr: 4.1094e-05 eta: 4:20:42 time: 0.8237 data_time: 0.0019 memory: 28783 grad_norm: 0.3398 loss: 0.2003 +2023/06/02 07:33:54 - mmengine - INFO - Epoch(train) [7][5100/5758] lr: 4.1094e-05 eta: 4:19:15 time: 0.8437 data_time: 0.0016 memory: 28783 grad_norm: 0.4106 loss: 0.1994 +2023/06/02 07:35:21 - mmengine - INFO - Epoch(train) [7][5200/5758] lr: 4.1094e-05 eta: 4:17:48 time: 0.7963 data_time: 0.0017 memory: 28783 grad_norm: 0.4387 loss: 0.2037 +2023/06/02 07:36:44 - mmengine - INFO - Epoch(train) [7][5300/5758] lr: 4.1094e-05 eta: 4:16:20 time: 0.7394 data_time: 0.0017 memory: 28783 grad_norm: 0.4938 loss: 0.1998 +2023/06/02 07:38:08 - mmengine - INFO - Epoch(train) [7][5400/5758] lr: 4.1094e-05 eta: 4:14:52 time: 0.7432 data_time: 0.0015 memory: 28783 grad_norm: 0.4962 loss: 0.1997 +2023/06/02 07:38:49 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_lr1e-4_10epoch_20230601_215742 +2023/06/02 07:39:29 - mmengine - INFO - Epoch(train) [7][5500/5758] lr: 4.1094e-05 eta: 4:13:22 time: 0.9282 data_time: 0.0893 memory: 28783 grad_norm: 0.3904 loss: 0.2006 +2023/06/02 07:41:37 - mmengine - INFO - Epoch(train) [7][5600/5758] lr: 4.1094e-05 eta: 4:12:14 time: 0.7588 data_time: 0.0018 memory: 28783 grad_norm: 0.9225 loss: 0.2010 +2023/06/02 07:42:57 - mmengine - INFO - Epoch(train) [7][5700/5758] lr: 4.1094e-05 eta: 4:10:44 time: 0.8363 data_time: 0.0047 memory: 28783 grad_norm: 0.5344 loss: 0.1992 +2023/06/02 07:43:46 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_lr1e-4_10epoch_20230601_215742 +2023/06/02 07:43:46 - mmengine - INFO - Saving checkpoint at 7 epochs +2023/06/02 07:44:07 - mmengine - INFO - Epoch(val) [7][16/16] accuracy/top1: 99.9558 single-label/precision_classwise: [100.0, 0.0] single-label/recall_classwise: [99.955810546875, 0.0] single-label/f1-score_classwise: [99.97789764404297, 0.0] data_time: 0.3266 time: 0.7415 +2023/06/02 07:45:36 - mmengine - INFO - Epoch(train) [8][ 100/5758] lr: 2.8550e-05 eta: 4:08:27 time: 0.8124 data_time: 0.3024 memory: 28783 grad_norm: 0.3496 loss: 0.2011 +2023/06/02 07:46:58 - mmengine - INFO - Epoch(train) [8][ 200/5758] lr: 2.8550e-05 eta: 4:06:58 time: 0.8971 data_time: 0.2267 memory: 28783 grad_norm: 0.4127 loss: 0.2005 +2023/06/02 07:48:18 - mmengine - INFO - Epoch(train) [8][ 300/5758] lr: 2.8550e-05 eta: 4:05:29 time: 0.7763 data_time: 0.2134 memory: 28783 grad_norm: 0.5613 loss: 0.2004 +2023/06/02 07:49:40 - mmengine - INFO - Epoch(train) [8][ 400/5758] lr: 2.8550e-05 eta: 4:04:00 time: 0.7559 data_time: 0.1615 memory: 28783 grad_norm: 0.3882 loss: 0.2011 +2023/06/02 07:51:05 - mmengine - INFO - Epoch(train) [8][ 500/5758] lr: 2.8550e-05 eta: 4:02:32 time: 0.7757 data_time: 0.1616 memory: 28783 grad_norm: 0.4840 loss: 0.2000 +2023/06/02 07:52:27 - mmengine - INFO - Epoch(train) [8][ 600/5758] lr: 2.8550e-05 eta: 4:01:04 time: 0.8355 data_time: 0.2551 memory: 28783 grad_norm: 0.4395 loss: 0.1998 +2023/06/02 07:53:46 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_lr1e-4_10epoch_20230601_215742 +2023/06/02 07:53:49 - mmengine - INFO - Epoch(train) [8][ 700/5758] lr: 2.8550e-05 eta: 3:59:35 time: 0.7693 data_time: 0.2436 memory: 28783 grad_norm: 0.4508 loss: 0.1998 +2023/06/02 07:55:11 - mmengine - INFO - Epoch(train) [8][ 800/5758] lr: 2.8550e-05 eta: 3:58:06 time: 0.9608 data_time: 0.2095 memory: 28783 grad_norm: 0.2767 loss: 0.2020 +2023/06/02 07:56:33 - mmengine - INFO - Epoch(train) [8][ 900/5758] lr: 2.8550e-05 eta: 3:56:38 time: 0.9432 data_time: 0.1693 memory: 28783 grad_norm: 0.3794 loss: 0.1999 +2023/06/02 07:57:58 - mmengine - INFO - Epoch(train) [8][1000/5758] lr: 2.8550e-05 eta: 3:55:10 time: 0.8092 data_time: 0.0021 memory: 28783 grad_norm: 0.3468 loss: 0.2001 +2023/06/02 07:59:20 - mmengine - INFO - Epoch(train) [8][1100/5758] lr: 2.8550e-05 eta: 3:53:42 time: 0.7599 data_time: 0.0016 memory: 28783 grad_norm: 0.5459 loss: 0.1991 +2023/06/02 08:00:44 - mmengine - INFO - Epoch(train) [8][1200/5758] lr: 2.8550e-05 eta: 3:52:14 time: 0.8348 data_time: 0.0016 memory: 28783 grad_norm: 0.8896 loss: 0.2009 +2023/06/02 08:02:11 - mmengine - INFO - Epoch(train) [8][1300/5758] lr: 2.8550e-05 eta: 3:50:47 time: 0.8940 data_time: 0.0023 memory: 28783 grad_norm: 0.4814 loss: 0.2000 +2023/06/02 08:03:38 - mmengine - INFO - Epoch(train) [8][1400/5758] lr: 2.8550e-05 eta: 3:49:21 time: 0.8363 data_time: 0.0019 memory: 28783 grad_norm: 0.3747 loss: 0.2003 +2023/06/02 08:05:03 - mmengine - INFO - Epoch(train) [8][1500/5758] lr: 2.8550e-05 eta: 3:47:54 time: 0.8487 data_time: 0.0020 memory: 28783 grad_norm: 0.2931 loss: 0.1990 +2023/06/02 08:06:27 - mmengine - INFO - Epoch(train) [8][1600/5758] lr: 2.8550e-05 eta: 3:46:26 time: 0.9286 data_time: 0.0050 memory: 28783 grad_norm: 0.3378 loss: 0.1997 +2023/06/02 08:07:48 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_lr1e-4_10epoch_20230601_215742 +2023/06/02 08:07:51 - mmengine - INFO - Epoch(train) [8][1700/5758] lr: 2.8550e-05 eta: 3:44:58 time: 0.8040 data_time: 0.0381 memory: 28783 grad_norm: 0.3319 loss: 0.1988 +2023/06/02 08:09:14 - mmengine - INFO - Epoch(train) [8][1800/5758] lr: 2.8550e-05 eta: 3:43:30 time: 0.8586 data_time: 0.1508 memory: 28783 grad_norm: 0.5127 loss: 0.1998 +2023/06/02 08:10:41 - mmengine - INFO - Epoch(train) [8][1900/5758] lr: 2.8550e-05 eta: 3:42:04 time: 0.8017 data_time: 0.0483 memory: 28783 grad_norm: 0.1921 loss: 0.1987 +2023/06/02 08:12:04 - mmengine - INFO - Epoch(train) [8][2000/5758] lr: 2.8550e-05 eta: 3:40:36 time: 0.8734 data_time: 0.2078 memory: 28783 grad_norm: 0.6226 loss: 0.2015 +2023/06/02 08:13:32 - mmengine - INFO - Epoch(train) [8][2100/5758] lr: 2.8550e-05 eta: 3:39:09 time: 0.7948 data_time: 0.0017 memory: 28783 grad_norm: 0.4400 loss: 0.2020 +2023/06/02 08:14:56 - mmengine - INFO - Epoch(train) [8][2200/5758] lr: 2.8550e-05 eta: 3:37:42 time: 0.7514 data_time: 0.0345 memory: 28783 grad_norm: 0.3869 loss: 0.2000 +2023/06/02 08:16:17 - mmengine - INFO - Epoch(train) [8][2300/5758] lr: 2.8550e-05 eta: 3:36:13 time: 0.8793 data_time: 0.1083 memory: 28783 grad_norm: 0.2929 loss: 0.1990 +2023/06/02 08:17:40 - mmengine - INFO - Epoch(train) [8][2400/5758] lr: 2.8550e-05 eta: 3:34:45 time: 0.7908 data_time: 0.1361 memory: 28783 grad_norm: 0.2890 loss: 0.1998 +2023/06/02 08:19:04 - mmengine - INFO - Epoch(train) [8][2500/5758] lr: 2.8550e-05 eta: 3:33:18 time: 0.7439 data_time: 0.2393 memory: 28783 grad_norm: 0.3636 loss: 0.1997 +2023/06/02 08:20:30 - mmengine - INFO - Epoch(train) [8][2600/5758] lr: 2.8550e-05 eta: 3:31:51 time: 0.9269 data_time: 0.1720 memory: 28783 grad_norm: 0.2769 loss: 0.1988 +2023/06/02 08:21:46 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_lr1e-4_10epoch_20230601_215742 +2023/06/02 08:21:51 - mmengine - INFO - Epoch(train) [8][2700/5758] lr: 2.8550e-05 eta: 3:30:22 time: 0.7904 data_time: 0.0617 memory: 28783 grad_norm: 0.7011 loss: 0.2011 +2023/06/02 08:23:13 - mmengine - INFO - Epoch(train) [8][2800/5758] lr: 2.8550e-05 eta: 3:28:54 time: 0.8595 data_time: 0.0731 memory: 28783 grad_norm: 0.4067 loss: 0.2009 +2023/06/02 08:24:36 - mmengine - INFO - Epoch(train) [8][2900/5758] lr: 2.8550e-05 eta: 3:27:26 time: 0.7759 data_time: 0.2648 memory: 28783 grad_norm: 0.3903 loss: 0.2017 +2023/06/02 08:26:01 - mmengine - INFO - Epoch(train) [8][3000/5758] lr: 2.8550e-05 eta: 3:25:59 time: 0.8795 data_time: 0.0253 memory: 28783 grad_norm: 0.4094 loss: 0.1995 +2023/06/02 08:27:27 - mmengine - INFO - Epoch(train) [8][3100/5758] lr: 2.8550e-05 eta: 3:24:32 time: 0.7434 data_time: 0.0016 memory: 28783 grad_norm: 0.5903 loss: 0.2014 +2023/06/02 08:28:48 - mmengine - INFO - Epoch(train) [8][3200/5758] lr: 2.8550e-05 eta: 3:23:04 time: 0.8079 data_time: 0.0021 memory: 28783 grad_norm: 0.3647 loss: 0.1989 +2023/06/02 08:30:12 - mmengine - INFO - Epoch(train) [8][3300/5758] lr: 2.8550e-05 eta: 3:21:37 time: 0.9307 data_time: 0.0032 memory: 28783 grad_norm: 0.4206 loss: 0.2004 +2023/06/02 08:31:35 - mmengine - INFO - Epoch(train) [8][3400/5758] lr: 2.8550e-05 eta: 3:20:09 time: 0.9138 data_time: 0.0018 memory: 28783 grad_norm: 0.1869 loss: 0.1999 +2023/06/02 08:32:55 - mmengine - INFO - Epoch(train) [8][3500/5758] lr: 2.8550e-05 eta: 3:18:40 time: 0.8124 data_time: 0.0016 memory: 28783 grad_norm: 0.4697 loss: 0.1998 +2023/06/02 08:34:17 - mmengine - INFO - Epoch(train) [8][3600/5758] lr: 2.8550e-05 eta: 3:17:12 time: 0.8867 data_time: 0.0023 memory: 28783 grad_norm: 0.3456 loss: 0.2001 +2023/06/02 08:35:38 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_lr1e-4_10epoch_20230601_215742 +2023/06/02 08:35:44 - mmengine - INFO - Epoch(train) [8][3700/5758] lr: 2.8550e-05 eta: 3:15:46 time: 0.7741 data_time: 0.0020 memory: 28783 grad_norm: 0.3255 loss: 0.2007 +2023/06/02 08:37:10 - mmengine - INFO - Epoch(train) [8][3800/5758] lr: 2.8550e-05 eta: 3:14:19 time: 0.8418 data_time: 0.0020 memory: 28783 grad_norm: 0.3425 loss: 0.1990 +2023/06/02 08:38:34 - mmengine - INFO - Epoch(train) [8][3900/5758] lr: 2.8550e-05 eta: 3:12:52 time: 0.7932 data_time: 0.0021 memory: 28783 grad_norm: 0.3316 loss: 0.1991 +2023/06/02 08:39:59 - mmengine - INFO - Epoch(train) [8][4000/5758] lr: 2.8550e-05 eta: 3:11:25 time: 0.7859 data_time: 0.0018 memory: 28783 grad_norm: 0.4273 loss: 0.2015 +2023/06/02 08:41:25 - mmengine - INFO - Epoch(train) [8][4100/5758] lr: 2.8550e-05 eta: 3:09:58 time: 0.8281 data_time: 0.0018 memory: 28783 grad_norm: 0.3826 loss: 0.2006 +2023/06/02 08:42:52 - mmengine - INFO - Epoch(train) [8][4200/5758] lr: 2.8550e-05 eta: 3:08:32 time: 0.8384 data_time: 0.0017 memory: 28783 grad_norm: 0.5383 loss: 0.2021 +2023/06/02 08:44:17 - mmengine - INFO - Epoch(train) [8][4300/5758] lr: 2.8550e-05 eta: 3:07:05 time: 0.8010 data_time: 0.0019 memory: 28783 grad_norm: 0.4321 loss: 0.2005 +2023/06/02 08:45:45 - mmengine - INFO - Epoch(train) [8][4400/5758] lr: 2.8550e-05 eta: 3:05:39 time: 0.8617 data_time: 0.0019 memory: 28783 grad_norm: 0.3800 loss: 0.2002 +2023/06/02 08:47:14 - mmengine - INFO - Epoch(train) [8][4500/5758] lr: 2.8550e-05 eta: 3:04:13 time: 0.8359 data_time: 0.0020 memory: 28783 grad_norm: 0.6177 loss: 0.1999 +2023/06/02 08:48:44 - mmengine - INFO - Epoch(train) [8][4600/5758] lr: 2.8550e-05 eta: 3:02:47 time: 0.8957 data_time: 0.0018 memory: 28783 grad_norm: 0.2700 loss: 0.1994 +2023/06/02 08:50:03 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_lr1e-4_10epoch_20230601_215742 +2023/06/02 08:50:10 - mmengine - INFO - Epoch(train) [8][4700/5758] lr: 2.8550e-05 eta: 3:01:21 time: 0.9269 data_time: 0.0021 memory: 28783 grad_norm: 0.4735 loss: 0.2005 +2023/06/02 08:51:38 - mmengine - INFO - Epoch(train) [8][4800/5758] lr: 2.8550e-05 eta: 2:59:55 time: 0.8555 data_time: 0.0016 memory: 28783 grad_norm: 0.3516 loss: 0.1995 +2023/06/02 08:53:03 - mmengine - INFO - Epoch(train) [8][4900/5758] lr: 2.8550e-05 eta: 2:58:28 time: 0.7707 data_time: 0.0018 memory: 28783 grad_norm: 0.3170 loss: 0.1990 +2023/06/02 08:54:27 - mmengine - INFO - Epoch(train) [8][5000/5758] lr: 2.8550e-05 eta: 2:57:00 time: 0.9431 data_time: 0.0047 memory: 28783 grad_norm: 0.5063 loss: 0.1999 +2023/06/02 08:55:53 - mmengine - INFO - Epoch(train) [8][5100/5758] lr: 2.8550e-05 eta: 2:55:34 time: 0.8423 data_time: 0.0023 memory: 28783 grad_norm: 0.1525 loss: 0.1997 +2023/06/02 08:57:19 - mmengine - INFO - Epoch(train) [8][5200/5758] lr: 2.8550e-05 eta: 2:54:07 time: 0.8008 data_time: 0.0019 memory: 28783 grad_norm: 0.4671 loss: 0.1993 +2023/06/02 08:58:45 - mmengine - INFO - Epoch(train) [8][5300/5758] lr: 2.8550e-05 eta: 2:52:40 time: 0.9225 data_time: 0.0059 memory: 28783 grad_norm: 0.3141 loss: 0.1992 +2023/06/02 09:00:12 - mmengine - INFO - Epoch(train) [8][5400/5758] lr: 2.8550e-05 eta: 2:51:14 time: 0.8995 data_time: 0.0023 memory: 28783 grad_norm: 0.9853 loss: 0.2018 +2023/06/02 09:01:35 - mmengine - INFO - Epoch(train) [8][5500/5758] lr: 2.8550e-05 eta: 2:49:46 time: 0.8120 data_time: 0.0020 memory: 28783 grad_norm: 0.3619 loss: 0.2012 +2023/06/02 09:03:03 - mmengine - INFO - Epoch(train) [8][5600/5758] lr: 2.8550e-05 eta: 2:48:20 time: 0.8677 data_time: 0.0021 memory: 28783 grad_norm: 0.2626 loss: 0.1991 +2023/06/02 09:04:19 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_lr1e-4_10epoch_20230601_215742 +2023/06/02 09:04:26 - mmengine - INFO - Epoch(train) [8][5700/5758] lr: 2.8550e-05 eta: 2:46:53 time: 0.8837 data_time: 0.0028 memory: 28783 grad_norm: 0.4918 loss: 0.1997 +2023/06/02 09:05:14 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_lr1e-4_10epoch_20230601_215742 +2023/06/02 09:05:14 - mmengine - INFO - Saving checkpoint at 8 epochs +2023/06/02 09:05:33 - mmengine - INFO - Epoch(val) [8][16/16] accuracy/top1: 99.9242 single-label/precision_classwise: [100.0, 0.0] single-label/recall_classwise: [99.92424774169922, 0.0] single-label/f1-score_classwise: [99.96210479736328, 0.0] data_time: 0.2054 time: 0.6206 +2023/06/02 09:07:04 - mmengine - INFO - Epoch(train) [9][ 100/5758] lr: 1.8594e-05 eta: 2:44:37 time: 0.8225 data_time: 0.3114 memory: 28783 grad_norm: 0.5146 loss: 0.2001 +2023/06/02 09:08:31 - mmengine - INFO - Epoch(train) [9][ 200/5758] lr: 1.8594e-05 eta: 2:43:10 time: 0.8247 data_time: 0.2124 memory: 28783 grad_norm: 0.3040 loss: 0.2007 +2023/06/02 09:09:56 - mmengine - INFO - Epoch(train) [9][ 300/5758] lr: 1.8594e-05 eta: 2:41:44 time: 0.7791 data_time: 0.2733 memory: 28783 grad_norm: 0.4373 loss: 0.1992 +2023/06/02 09:11:18 - mmengine - INFO - Epoch(train) [9][ 400/5758] lr: 1.8594e-05 eta: 2:40:16 time: 0.8539 data_time: 0.3246 memory: 28783 grad_norm: 0.2531 loss: 0.1988 +2023/06/02 09:12:43 - mmengine - INFO - Epoch(train) [9][ 500/5758] lr: 1.8594e-05 eta: 2:38:49 time: 0.8362 data_time: 0.3200 memory: 28783 grad_norm: 0.2349 loss: 0.1998 +2023/06/02 09:14:10 - mmengine - INFO - Epoch(train) [9][ 600/5758] lr: 1.8594e-05 eta: 2:37:23 time: 0.8733 data_time: 0.3531 memory: 28783 grad_norm: 0.2370 loss: 0.2000 +2023/06/02 09:15:35 - mmengine - INFO - Epoch(train) [9][ 700/5758] lr: 1.8594e-05 eta: 2:35:56 time: 0.8480 data_time: 0.2327 memory: 28783 grad_norm: 0.3532 loss: 0.1987 +2023/06/02 09:17:03 - mmengine - INFO - Epoch(train) [9][ 800/5758] lr: 1.8594e-05 eta: 2:34:30 time: 0.7945 data_time: 0.2804 memory: 28783 grad_norm: 0.3165 loss: 0.1993 +2023/06/02 09:18:31 - mmengine - INFO - Epoch(train) [9][ 900/5758] lr: 1.8594e-05 eta: 2:33:04 time: 0.9263 data_time: 0.2589 memory: 28783 grad_norm: 0.2437 loss: 0.1995 +2023/06/02 09:19:01 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_lr1e-4_10epoch_20230601_215742 +2023/06/02 09:19:59 - mmengine - INFO - Epoch(train) [9][1000/5758] lr: 1.8594e-05 eta: 2:31:37 time: 0.9424 data_time: 0.0035 memory: 28783 grad_norm: 0.1810 loss: 0.1996 +2023/06/02 09:21:24 - mmengine - INFO - Epoch(train) [9][1100/5758] lr: 1.8594e-05 eta: 2:30:11 time: 0.8317 data_time: 0.0020 memory: 28783 grad_norm: 0.3583 loss: 0.1991 +2023/06/02 09:22:52 - mmengine - INFO - Epoch(train) [9][1200/5758] lr: 1.8594e-05 eta: 2:28:44 time: 0.8387 data_time: 0.0061 memory: 28783 grad_norm: 0.2616 loss: 0.1991 +2023/06/02 09:24:18 - mmengine - INFO - Epoch(train) [9][1300/5758] lr: 1.8594e-05 eta: 2:27:18 time: 0.7974 data_time: 0.0019 memory: 28783 grad_norm: 0.2789 loss: 0.2013 +2023/06/02 09:25:45 - mmengine - INFO - Epoch(train) [9][1400/5758] lr: 1.8594e-05 eta: 2:25:51 time: 0.8213 data_time: 0.0020 memory: 28783 grad_norm: 0.2620 loss: 0.1987 +2023/06/02 09:27:07 - mmengine - INFO - Epoch(train) [9][1500/5758] lr: 1.8594e-05 eta: 2:24:24 time: 0.7681 data_time: 0.0017 memory: 28783 grad_norm: 0.3868 loss: 0.1995 +2023/06/02 09:28:31 - mmengine - INFO - Epoch(train) [9][1600/5758] lr: 1.8594e-05 eta: 2:22:57 time: 0.9743 data_time: 0.1263 memory: 28783 grad_norm: 0.4179 loss: 0.1998 +2023/06/02 09:29:58 - mmengine - INFO - Epoch(train) [9][1700/5758] lr: 1.8594e-05 eta: 2:21:30 time: 0.8499 data_time: 0.0022 memory: 28783 grad_norm: 0.3808 loss: 0.1996 +2023/06/02 09:31:23 - mmengine - INFO - Epoch(train) [9][1800/5758] lr: 1.8594e-05 eta: 2:20:04 time: 0.8195 data_time: 0.0022 memory: 28783 grad_norm: 0.2973 loss: 0.1990 +2023/06/02 09:32:46 - mmengine - INFO - Epoch(train) [9][1900/5758] lr: 1.8594e-05 eta: 2:18:36 time: 0.7543 data_time: 0.0053 memory: 28783 grad_norm: 0.3173 loss: 0.1991 +2023/06/02 09:33:14 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_lr1e-4_10epoch_20230601_215742 +2023/06/02 09:34:09 - mmengine - INFO - Epoch(train) [9][2000/5758] lr: 1.8594e-05 eta: 2:17:09 time: 0.8292 data_time: 0.0044 memory: 28783 grad_norm: 0.2702 loss: 0.1989 +2023/06/02 09:35:34 - mmengine - INFO - Epoch(train) [9][2100/5758] lr: 1.8594e-05 eta: 2:15:43 time: 0.8059 data_time: 0.0019 memory: 28783 grad_norm: 0.4502 loss: 0.1992 +2023/06/02 09:37:00 - mmengine - INFO - Epoch(train) [9][2200/5758] lr: 1.8594e-05 eta: 2:14:16 time: 0.7517 data_time: 0.0020 memory: 28783 grad_norm: 0.3632 loss: 0.1999 +2023/06/02 09:38:26 - mmengine - INFO - Epoch(train) [9][2300/5758] lr: 1.8594e-05 eta: 2:12:49 time: 0.8640 data_time: 0.0019 memory: 28783 grad_norm: 0.2055 loss: 0.1989 +2023/06/02 09:39:48 - mmengine - INFO - Epoch(train) [9][2400/5758] lr: 1.8594e-05 eta: 2:11:22 time: 0.7844 data_time: 0.0024 memory: 28783 grad_norm: 0.2602 loss: 0.1987 +2023/06/02 09:41:15 - mmengine - INFO - Epoch(train) [9][2500/5758] lr: 1.8594e-05 eta: 2:09:56 time: 0.9115 data_time: 0.0027 memory: 28783 grad_norm: 0.3142 loss: 0.1994 +2023/06/02 09:42:42 - mmengine - INFO - Epoch(train) [9][2600/5758] lr: 1.8594e-05 eta: 2:08:29 time: 0.8528 data_time: 0.0022 memory: 28783 grad_norm: 0.2406 loss: 0.1986 +2023/06/02 09:44:05 - mmengine - INFO - Epoch(train) [9][2700/5758] lr: 1.8594e-05 eta: 2:07:02 time: 0.8380 data_time: 0.0020 memory: 28783 grad_norm: 0.4137 loss: 0.2027 +2023/06/02 09:45:29 - mmengine - INFO - Epoch(train) [9][2800/5758] lr: 1.8594e-05 eta: 2:05:35 time: 0.7925 data_time: 0.0017 memory: 28783 grad_norm: 0.5483 loss: 0.1991 +2023/06/02 09:46:54 - mmengine - INFO - Epoch(train) [9][2900/5758] lr: 1.8594e-05 eta: 2:04:09 time: 0.8821 data_time: 0.0027 memory: 28783 grad_norm: 0.2809 loss: 0.1992 +2023/06/02 09:47:23 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_lr1e-4_10epoch_20230601_215742 +2023/06/02 09:48:20 - mmengine - INFO - Epoch(train) [9][3000/5758] lr: 1.8594e-05 eta: 2:02:42 time: 0.9235 data_time: 0.0029 memory: 28783 grad_norm: 0.3208 loss: 0.1993 +2023/06/02 09:49:46 - mmengine - INFO - Epoch(train) [9][3100/5758] lr: 1.8594e-05 eta: 2:01:15 time: 0.8595 data_time: 0.0020 memory: 28783 grad_norm: 0.3346 loss: 0.1996 +2023/06/02 09:51:13 - mmengine - INFO - Epoch(train) [9][3200/5758] lr: 1.8594e-05 eta: 1:59:49 time: 0.7800 data_time: 0.0018 memory: 28783 grad_norm: 0.1843 loss: 0.1989 +2023/06/02 09:52:39 - mmengine - INFO - Epoch(train) [9][3300/5758] lr: 1.8594e-05 eta: 1:58:23 time: 0.9234 data_time: 0.0033 memory: 28783 grad_norm: 0.2937 loss: 0.1986 +2023/06/02 09:54:04 - mmengine - INFO - Epoch(train) [9][3400/5758] lr: 1.8594e-05 eta: 1:56:56 time: 0.9004 data_time: 0.0025 memory: 28783 grad_norm: 0.3456 loss: 0.1999 +2023/06/02 09:55:26 - mmengine - INFO - Epoch(train) [9][3500/5758] lr: 1.8594e-05 eta: 1:55:29 time: 0.7805 data_time: 0.0020 memory: 28783 grad_norm: 0.2446 loss: 0.2001 +2023/06/02 09:56:49 - mmengine - INFO - Epoch(train) [9][3600/5758] lr: 1.8594e-05 eta: 1:54:02 time: 0.7839 data_time: 0.0017 memory: 28783 grad_norm: 0.4213 loss: 0.2001 +2023/06/02 09:58:16 - mmengine - INFO - Epoch(train) [9][3700/5758] lr: 1.8594e-05 eta: 1:52:35 time: 0.7988 data_time: 0.0025 memory: 28783 grad_norm: 0.3484 loss: 0.2000 +2023/06/02 09:59:39 - mmengine - INFO - Epoch(train) [9][3800/5758] lr: 1.8594e-05 eta: 1:51:08 time: 0.8218 data_time: 0.0023 memory: 28783 grad_norm: 0.3710 loss: 0.1986 +2023/06/02 10:01:03 - mmengine - INFO - Epoch(train) [9][3900/5758] lr: 1.8594e-05 eta: 1:49:42 time: 0.8231 data_time: 0.0017 memory: 28783 grad_norm: 0.1877 loss: 0.1987 +2023/06/02 10:01:33 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_lr1e-4_10epoch_20230601_215742 +2023/06/02 10:02:29 - mmengine - INFO - Epoch(train) [9][4000/5758] lr: 1.8594e-05 eta: 1:48:15 time: 0.7972 data_time: 0.0022 memory: 28783 grad_norm: 0.4158 loss: 0.2005 +2023/06/02 10:03:53 - mmengine - INFO - Epoch(train) [9][4100/5758] lr: 1.8594e-05 eta: 1:46:48 time: 0.8016 data_time: 0.0022 memory: 28783 grad_norm: 0.1843 loss: 0.1995 +2023/06/02 10:05:20 - mmengine - INFO - Epoch(train) [9][4200/5758] lr: 1.8594e-05 eta: 1:45:22 time: 0.7558 data_time: 0.0020 memory: 28783 grad_norm: 0.2811 loss: 0.1990 +2023/06/02 10:06:46 - mmengine - INFO - Epoch(train) [9][4300/5758] lr: 1.8594e-05 eta: 1:43:55 time: 0.8563 data_time: 0.0018 memory: 28783 grad_norm: 0.3857 loss: 0.1989 +2023/06/02 10:08:08 - mmengine - INFO - Epoch(train) [9][4400/5758] lr: 1.8594e-05 eta: 1:42:28 time: 0.8917 data_time: 0.0021 memory: 28783 grad_norm: 0.2892 loss: 0.2000 +2023/06/02 10:09:30 - mmengine - INFO - Epoch(train) [9][4500/5758] lr: 1.8594e-05 eta: 1:41:01 time: 0.9341 data_time: 0.0023 memory: 28783 grad_norm: 0.1743 loss: 0.1986 +2023/06/02 10:10:56 - mmengine - INFO - Epoch(train) [9][4600/5758] lr: 1.8594e-05 eta: 1:39:35 time: 0.8189 data_time: 0.0020 memory: 28783 grad_norm: 0.3208 loss: 0.2001 +2023/06/02 10:12:20 - mmengine - INFO - Epoch(train) [9][4700/5758] lr: 1.8594e-05 eta: 1:38:08 time: 0.7660 data_time: 0.0016 memory: 28783 grad_norm: 0.3781 loss: 0.1996 +2023/06/02 10:13:46 - mmengine - INFO - Epoch(train) [9][4800/5758] lr: 1.8594e-05 eta: 1:36:42 time: 0.8132 data_time: 0.0017 memory: 28783 grad_norm: 0.3290 loss: 0.2000 +2023/06/02 10:15:14 - mmengine - INFO - Epoch(train) [9][4900/5758] lr: 1.8594e-05 eta: 1:35:16 time: 0.8445 data_time: 0.0017 memory: 28783 grad_norm: 0.1926 loss: 0.1991 +2023/06/02 10:15:44 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_lr1e-4_10epoch_20230601_215742 +2023/06/02 10:16:43 - mmengine - INFO - Epoch(train) [9][5000/5758] lr: 1.8594e-05 eta: 1:33:49 time: 0.8967 data_time: 0.0018 memory: 28783 grad_norm: 0.2376 loss: 0.1991 +2023/06/02 10:18:11 - mmengine - INFO - Epoch(train) [9][5100/5758] lr: 1.8594e-05 eta: 1:32:23 time: 0.8451 data_time: 0.0018 memory: 28783 grad_norm: 0.4425 loss: 0.1992 +2023/06/02 10:19:39 - mmengine - INFO - Epoch(train) [9][5200/5758] lr: 1.8594e-05 eta: 1:30:57 time: 0.8396 data_time: 0.0016 memory: 28783 grad_norm: 0.3727 loss: 0.1998 +2023/06/02 10:21:06 - mmengine - INFO - Epoch(train) [9][5300/5758] lr: 1.8594e-05 eta: 1:29:31 time: 0.8631 data_time: 0.0029 memory: 28783 grad_norm: 0.2957 loss: 0.1991 +2023/06/02 10:22:31 - mmengine - INFO - Epoch(train) [9][5400/5758] lr: 1.8594e-05 eta: 1:28:04 time: 0.8605 data_time: 0.0017 memory: 28783 grad_norm: 0.1703 loss: 0.1986 +2023/06/02 10:23:57 - mmengine - INFO - Epoch(train) [9][5500/5758] lr: 1.8594e-05 eta: 1:26:38 time: 0.9337 data_time: 0.0043 memory: 28783 grad_norm: 0.3017 loss: 0.1988 +2023/06/02 10:25:23 - mmengine - INFO - Epoch(train) [9][5600/5758] lr: 1.8594e-05 eta: 1:25:11 time: 0.9466 data_time: 0.0037 memory: 28783 grad_norm: 0.2117 loss: 0.1992 +2023/06/02 10:26:50 - mmengine - INFO - Epoch(train) [9][5700/5758] lr: 1.8594e-05 eta: 1:23:45 time: 0.8132 data_time: 0.0018 memory: 28783 grad_norm: 0.3543 loss: 0.1998 +2023/06/02 10:27:38 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_lr1e-4_10epoch_20230601_215742 +2023/06/02 10:27:38 - mmengine - INFO - Saving checkpoint at 9 epochs +2023/06/02 10:27:58 - mmengine - INFO - Epoch(val) [9][16/16] accuracy/top1: 99.9684 single-label/precision_classwise: [100.0, 0.0] single-label/recall_classwise: [99.96842956542969, 0.0] single-label/f1-score_classwise: [99.98420715332031, 0.0] data_time: 0.2079 time: 0.6231 +2023/06/02 10:29:29 - mmengine - INFO - Epoch(train) [10][ 100/5758] lr: 1.2202e-05 eta: 1:21:29 time: 0.9107 data_time: 0.1353 memory: 28783 grad_norm: 0.1930 loss: 0.1992 +2023/06/02 10:30:33 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_lr1e-4_10epoch_20230601_215742 +2023/06/02 10:30:53 - mmengine - INFO - Epoch(train) [10][ 200/5758] lr: 1.2202e-05 eta: 1:20:02 time: 0.8882 data_time: 0.2233 memory: 28783 grad_norm: 0.3241 loss: 0.1994 +2023/06/02 10:32:17 - mmengine - INFO - Epoch(train) [10][ 300/5758] lr: 1.2202e-05 eta: 1:18:35 time: 0.8835 data_time: 0.2135 memory: 28783 grad_norm: 0.3562 loss: 0.1992 +2023/06/02 10:33:41 - mmengine - INFO - Epoch(train) [10][ 400/5758] lr: 1.2202e-05 eta: 1:17:09 time: 0.8346 data_time: 0.1142 memory: 28783 grad_norm: 0.0928 loss: 0.1985 +2023/06/02 10:35:03 - mmengine - INFO - Epoch(train) [10][ 500/5758] lr: 1.2202e-05 eta: 1:15:42 time: 0.7938 data_time: 0.0577 memory: 28783 grad_norm: 0.2211 loss: 0.1987 +2023/06/02 10:36:28 - mmengine - INFO - Epoch(train) [10][ 600/5758] lr: 1.2202e-05 eta: 1:14:15 time: 0.8490 data_time: 0.0017 memory: 28783 grad_norm: 0.3011 loss: 0.2010 +2023/06/02 10:37:52 - mmengine - INFO - Epoch(train) [10][ 700/5758] lr: 1.2202e-05 eta: 1:12:49 time: 0.7836 data_time: 0.2268 memory: 28783 grad_norm: 0.2437 loss: 0.1990 +2023/06/02 10:39:18 - mmengine - INFO - Epoch(train) [10][ 800/5758] lr: 1.2202e-05 eta: 1:11:22 time: 0.8662 data_time: 0.1106 memory: 28783 grad_norm: 0.2070 loss: 0.1986 +2023/06/02 10:40:45 - mmengine - INFO - Epoch(train) [10][ 900/5758] lr: 1.2202e-05 eta: 1:09:56 time: 0.8858 data_time: 0.0120 memory: 28783 grad_norm: 0.2800 loss: 0.2002 +2023/06/02 10:42:11 - mmengine - INFO - Epoch(train) [10][1000/5758] lr: 1.2202e-05 eta: 1:08:30 time: 0.9430 data_time: 0.0035 memory: 28783 grad_norm: 0.2120 loss: 0.1994 +2023/06/02 10:43:39 - mmengine - INFO - Epoch(train) [10][1100/5758] lr: 1.2202e-05 eta: 1:07:03 time: 0.8428 data_time: 0.0018 memory: 28783 grad_norm: 0.2766 loss: 0.1991 +2023/06/02 10:44:51 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_lr1e-4_10epoch_20230601_215742 +2023/06/02 10:45:09 - mmengine - INFO - Epoch(train) [10][1200/5758] lr: 1.2202e-05 eta: 1:05:37 time: 0.8509 data_time: 0.0020 memory: 28783 grad_norm: 0.3117 loss: 0.1991 +2023/06/02 10:46:40 - mmengine - INFO - Epoch(train) [10][1300/5758] lr: 1.2202e-05 eta: 1:04:11 time: 0.8970 data_time: 0.0016 memory: 28783 grad_norm: 0.2789 loss: 0.2002 +2023/06/02 10:48:07 - mmengine - INFO - Epoch(train) [10][1400/5758] lr: 1.2202e-05 eta: 1:02:45 time: 0.9008 data_time: 0.0025 memory: 28783 grad_norm: 0.1610 loss: 0.1986 +2023/06/02 10:49:38 - mmengine - INFO - Epoch(train) [10][1500/5758] lr: 1.2202e-05 eta: 1:01:19 time: 0.9430 data_time: 0.0023 memory: 28783 grad_norm: 0.0997 loss: 0.1989 +2023/06/02 10:51:06 - mmengine - INFO - Epoch(train) [10][1600/5758] lr: 1.2202e-05 eta: 0:59:53 time: 0.8756 data_time: 0.0021 memory: 28783 grad_norm: 0.2919 loss: 0.1995 +2023/06/02 10:52:36 - mmengine - INFO - Epoch(train) [10][1700/5758] lr: 1.2202e-05 eta: 0:58:26 time: 0.9553 data_time: 0.0031 memory: 28783 grad_norm: 0.3731 loss: 0.1995 +2023/06/02 10:54:04 - mmengine - INFO - Epoch(train) [10][1800/5758] lr: 1.2202e-05 eta: 0:57:00 time: 0.9699 data_time: 0.0022 memory: 28783 grad_norm: 0.5374 loss: 0.1994 +2023/06/02 10:55:32 - mmengine - INFO - Epoch(train) [10][1900/5758] lr: 1.2202e-05 eta: 0:55:34 time: 0.9012 data_time: 0.0023 memory: 28783 grad_norm: 0.2043 loss: 0.1987 +2023/06/02 10:56:58 - mmengine - INFO - Epoch(train) [10][2000/5758] lr: 1.2202e-05 eta: 0:54:07 time: 0.7622 data_time: 0.0019 memory: 28783 grad_norm: 0.1514 loss: 0.1991 +2023/06/02 10:58:25 - mmengine - INFO - Epoch(train) [10][2100/5758] lr: 1.2202e-05 eta: 0:52:41 time: 0.7885 data_time: 0.0018 memory: 28783 grad_norm: 0.1600 loss: 0.1986 +2023/06/02 10:59:33 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_lr1e-4_10epoch_20230601_215742 +2023/06/02 10:59:54 - mmengine - INFO - Epoch(train) [10][2200/5758] lr: 1.2202e-05 eta: 0:51:15 time: 1.0244 data_time: 0.0035 memory: 28783 grad_norm: 0.1601 loss: 0.1993 +2023/06/02 11:01:19 - mmengine - INFO - Epoch(train) [10][2300/5758] lr: 1.2202e-05 eta: 0:49:48 time: 0.9077 data_time: 0.0034 memory: 28783 grad_norm: 0.2291 loss: 0.1990 +2023/06/02 11:02:45 - mmengine - INFO - Epoch(train) [10][2400/5758] lr: 1.2202e-05 eta: 0:48:22 time: 0.8655 data_time: 0.0017 memory: 28783 grad_norm: 0.4976 loss: 0.1997 +2023/06/02 11:04:11 - mmengine - INFO - Epoch(train) [10][2500/5758] lr: 1.2202e-05 eta: 0:46:55 time: 0.8174 data_time: 0.0016 memory: 28783 grad_norm: 0.2486 loss: 0.1993 +2023/06/02 11:05:37 - mmengine - INFO - Epoch(train) [10][2600/5758] lr: 1.2202e-05 eta: 0:45:29 time: 0.9116 data_time: 0.0031 memory: 28783 grad_norm: 0.2542 loss: 0.1987 +2023/06/02 11:07:03 - mmengine - INFO - Epoch(train) [10][2700/5758] lr: 1.2202e-05 eta: 0:44:02 time: 0.8773 data_time: 0.0023 memory: 28783 grad_norm: 0.3610 loss: 0.2008 +2023/06/02 11:08:28 - mmengine - INFO - Epoch(train) [10][2800/5758] lr: 1.2202e-05 eta: 0:42:36 time: 0.8673 data_time: 0.0024 memory: 28783 grad_norm: 0.2083 loss: 0.1993 +2023/06/02 11:09:54 - mmengine - INFO - Epoch(train) [10][2900/5758] lr: 1.2202e-05 eta: 0:41:09 time: 0.7648 data_time: 0.0018 memory: 28783 grad_norm: 0.6618 loss: 0.1996 +2023/06/02 11:11:19 - mmengine - INFO - Epoch(train) [10][3000/5758] lr: 1.2202e-05 eta: 0:39:43 time: 0.7911 data_time: 0.0027 memory: 28783 grad_norm: 0.2356 loss: 0.1987 +2023/06/02 11:12:43 - mmengine - INFO - Epoch(train) [10][3100/5758] lr: 1.2202e-05 eta: 0:38:16 time: 0.8555 data_time: 0.0024 memory: 28783 grad_norm: 0.3155 loss: 0.1997 +2023/06/02 11:13:53 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_lr1e-4_10epoch_20230601_215742 +2023/06/02 11:14:10 - mmengine - INFO - Epoch(train) [10][3200/5758] lr: 1.2202e-05 eta: 0:36:50 time: 0.8116 data_time: 0.0089 memory: 28783 grad_norm: 0.1727 loss: 0.1994 +2023/06/02 11:15:33 - mmengine - INFO - Epoch(train) [10][3300/5758] lr: 1.2202e-05 eta: 0:35:23 time: 0.8221 data_time: 0.1582 memory: 28783 grad_norm: 0.3541 loss: 0.1995 +2023/06/02 11:16:58 - mmengine - INFO - Epoch(train) [10][3400/5758] lr: 1.2202e-05 eta: 0:33:57 time: 0.8965 data_time: 0.1481 memory: 28783 grad_norm: 0.2612 loss: 0.1986 +2023/06/02 11:18:24 - mmengine - INFO - Epoch(train) [10][3500/5758] lr: 1.2202e-05 eta: 0:32:31 time: 0.9045 data_time: 0.0674 memory: 28783 grad_norm: 0.3018 loss: 0.1995 +2023/06/02 11:19:52 - mmengine - INFO - Epoch(train) [10][3600/5758] lr: 1.2202e-05 eta: 0:31:04 time: 0.8526 data_time: 0.0192 memory: 28783 grad_norm: 0.2660 loss: 0.1992 +2023/06/02 11:21:21 - mmengine - INFO - Epoch(train) [10][3700/5758] lr: 1.2202e-05 eta: 0:29:38 time: 0.8376 data_time: 0.0018 memory: 28783 grad_norm: 0.4114 loss: 0.1989 +2023/06/02 11:22:48 - mmengine - INFO - Epoch(train) [10][3800/5758] lr: 1.2202e-05 eta: 0:28:12 time: 0.8873 data_time: 0.0023 memory: 28783 grad_norm: 0.3615 loss: 0.1990 +2023/06/02 11:24:16 - mmengine - INFO - Epoch(train) [10][3900/5758] lr: 1.2202e-05 eta: 0:26:45 time: 0.8065 data_time: 0.0026 memory: 28783 grad_norm: 0.1956 loss: 0.1998 +2023/06/02 11:25:44 - mmengine - INFO - Epoch(train) [10][4000/5758] lr: 1.2202e-05 eta: 0:25:19 time: 0.9117 data_time: 0.0027 memory: 28783 grad_norm: 0.1707 loss: 0.1987 +2023/06/02 11:27:12 - mmengine - INFO - Epoch(train) [10][4100/5758] lr: 1.2202e-05 eta: 0:23:52 time: 0.8401 data_time: 0.0025 memory: 28783 grad_norm: 0.1427 loss: 0.1986 +2023/06/02 11:28:21 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_lr1e-4_10epoch_20230601_215742 +2023/06/02 11:28:38 - mmengine - INFO - Epoch(train) [10][4200/5758] lr: 1.2202e-05 eta: 0:22:26 time: 0.8213 data_time: 0.0020 memory: 28783 grad_norm: 0.2872 loss: 0.1994 +2023/06/02 11:30:04 - mmengine - INFO - Epoch(train) [10][4300/5758] lr: 1.2202e-05 eta: 0:21:00 time: 0.9557 data_time: 0.0023 memory: 28783 grad_norm: 0.1579 loss: 0.2002 +2023/06/02 11:31:30 - mmengine - INFO - Epoch(train) [10][4400/5758] lr: 1.2202e-05 eta: 0:19:33 time: 0.9123 data_time: 0.0042 memory: 28783 grad_norm: 0.1674 loss: 0.1990 +2023/06/02 11:32:57 - mmengine - INFO - Epoch(train) [10][4500/5758] lr: 1.2202e-05 eta: 0:18:07 time: 0.8098 data_time: 0.0018 memory: 28783 grad_norm: 0.2765 loss: 0.1990 +2023/06/02 11:34:24 - mmengine - INFO - Epoch(train) [10][4600/5758] lr: 1.2202e-05 eta: 0:16:40 time: 0.8895 data_time: 0.0020 memory: 28783 grad_norm: 0.2394 loss: 0.1996 +2023/06/02 11:35:50 - mmengine - INFO - Epoch(train) [10][4700/5758] lr: 1.2202e-05 eta: 0:15:14 time: 0.8621 data_time: 0.0022 memory: 28783 grad_norm: 0.3644 loss: 0.2002 +2023/06/02 11:37:13 - mmengine - INFO - Epoch(train) [10][4800/5758] lr: 1.2202e-05 eta: 0:13:47 time: 0.8395 data_time: 0.0022 memory: 28783 grad_norm: 0.2618 loss: 0.1988 +2023/06/02 11:38:37 - mmengine - INFO - Epoch(train) [10][4900/5758] lr: 1.2202e-05 eta: 0:12:21 time: 0.7629 data_time: 0.0018 memory: 28783 grad_norm: 0.3552 loss: 0.1996 +2023/06/02 11:40:03 - mmengine - INFO - Epoch(train) [10][5000/5758] lr: 1.2202e-05 eta: 0:10:55 time: 0.7984 data_time: 0.0017 memory: 28783 grad_norm: 0.3564 loss: 0.1992 +2023/06/02 11:41:29 - mmengine - INFO - Epoch(train) [10][5100/5758] lr: 1.2202e-05 eta: 0:09:28 time: 0.7876 data_time: 0.0097 memory: 28783 grad_norm: 0.4211 loss: 0.2000 +2023/06/02 11:42:37 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_lr1e-4_10epoch_20230601_215742 +2023/06/02 11:42:58 - mmengine - INFO - Epoch(train) [10][5200/5758] lr: 1.2202e-05 eta: 0:08:02 time: 0.9507 data_time: 0.2023 memory: 28783 grad_norm: 0.2265 loss: 0.1990 +2023/06/02 11:44:23 - mmengine - INFO - Epoch(train) [10][5300/5758] lr: 1.2202e-05 eta: 0:06:35 time: 0.8301 data_time: 0.1407 memory: 28783 grad_norm: 0.2880 loss: 0.1987 +2023/06/02 11:45:48 - mmengine - INFO - Epoch(train) [10][5400/5758] lr: 1.2202e-05 eta: 0:05:09 time: 0.8176 data_time: 0.0640 memory: 28783 grad_norm: 0.1465 loss: 0.1988 +2023/06/02 11:47:18 - mmengine - INFO - Epoch(train) [10][5500/5758] lr: 1.2202e-05 eta: 0:03:42 time: 0.8183 data_time: 0.1102 memory: 28783 grad_norm: 0.5016 loss: 0.1998 +2023/06/02 11:48:43 - mmengine - INFO - Epoch(train) [10][5600/5758] lr: 1.2202e-05 eta: 0:02:16 time: 0.9866 data_time: 0.0150 memory: 28783 grad_norm: 0.2388 loss: 0.2005 +2023/06/02 11:50:11 - mmengine - INFO - Epoch(train) [10][5700/5758] lr: 1.2202e-05 eta: 0:00:50 time: 0.9692 data_time: 0.0037 memory: 28783 grad_norm: 1.2485 loss: 0.1987 +2023/06/02 11:51:04 - mmengine - INFO - Exp name: swin_small_4xb256_fake5m_lr1e-4_10epoch_20230601_215742 +2023/06/02 11:51:04 - mmengine - INFO - Saving checkpoint at 10 epochs +2023/06/02 11:51:24 - mmengine - INFO - Epoch(val) [10][16/16] accuracy/top1: 99.9684 single-label/precision_classwise: [100.0, 0.0] single-label/recall_classwise: [99.96842956542969, 0.0] single-label/f1-score_classwise: [99.98420715332031, 0.0] data_time: 0.2039 time: 0.6186 diff --git a/swin_small_4xb256_fake5m_lr1e-4_10epoch/20230601_215742/vis_data/20230601_215742.json b/swin_small_4xb256_fake5m_lr1e-4_10epoch/20230601_215742/vis_data/20230601_215742.json new file mode 100644 index 0000000000000000000000000000000000000000..a357100b511907b74eea150685e3e0b2eb47f1a1 --- /dev/null +++ b/swin_small_4xb256_fake5m_lr1e-4_10epoch/20230601_215742/vis_data/20230601_215742.json @@ -0,0 +1,580 @@ +{"lr": 0.0001, "data_time": 0.2836922168731689, "grad_norm": 1.1496466875076294, "loss": 0.6058561384677887, "time": 0.8801626920700073, "epoch": 1, "memory": 28798, "step": 100} +{"lr": 0.0001, "data_time": 0.4571942567825317, "grad_norm": 1.530609768629074, "loss": 0.5631434917449951, "time": 1.005289363861084, "epoch": 1, "memory": 28798, "step": 200} +{"lr": 0.0001, "data_time": 0.22151150703430175, "grad_norm": 2.028353178501129, "loss": 0.5296812951564789, "time": 0.9259072303771972, "epoch": 1, "memory": 28798, "step": 300} +{"lr": 0.0001, "data_time": 0.291133189201355, "grad_norm": 2.500625509023666, "loss": 0.5156413376331329, "time": 0.9341419219970704, "epoch": 1, "memory": 28798, "step": 400} +{"lr": 0.0001, "data_time": 0.12301592826843262, "grad_norm": 4.318602168560028, "loss": 0.5010578453540802, "time": 0.8412275791168213, "epoch": 1, "memory": 28798, "step": 500} +{"lr": 0.0001, "data_time": 0.13549425601959228, "grad_norm": 4.586491513252258, "loss": 0.4730592280626297, "time": 0.9765780925750732, "epoch": 1, "memory": 28798, "step": 600} +{"lr": 0.0001, "data_time": 0.16471946239471436, "grad_norm": 3.8183645486831663, "loss": 0.45454353988170626, "time": 0.8893230438232422, "epoch": 1, "memory": 28798, "step": 700} +{"lr": 0.0001, "data_time": 0.12637732028961182, "grad_norm": 6.184123384952545, "loss": 0.4634770840406418, "time": 0.8824618577957153, "epoch": 1, "memory": 28798, "step": 800} +{"lr": 0.0001, "data_time": 0.001864790916442871, "grad_norm": 3.7891004204750063, "loss": 0.44833930730819704, "time": 0.7174569368362427, "epoch": 1, "memory": 28798, "step": 900} +{"lr": 0.0001, "data_time": 0.0021742820739746095, "grad_norm": 6.032311713695526, "loss": 0.44841069281101226, "time": 0.8957549571990967, "epoch": 1, "memory": 28798, "step": 1000} +{"lr": 0.0001, "data_time": 0.0016405105590820313, "grad_norm": 3.7643929719924927, "loss": 0.42695329487323763, "time": 0.7580570459365845, "epoch": 1, "memory": 28798, "step": 1100} +{"lr": 0.0001, "data_time": 0.03567514419555664, "grad_norm": 8.661775887012482, "loss": 0.4228070259094238, "time": 0.7184351682662964, "epoch": 1, "memory": 28798, "step": 1200} +{"lr": 0.0001, "data_time": 0.1960144519805908, "grad_norm": 12.072546911239623, "loss": 0.43221528828144073, "time": 0.7568394422531128, "epoch": 1, "memory": 28798, "step": 1300} +{"lr": 0.0001, "data_time": 0.13183763027191162, "grad_norm": 9.784806501865386, "loss": 0.4166201651096344, "time": 0.9501391172409057, "epoch": 1, "memory": 28798, "step": 1400} +{"lr": 0.0001, "data_time": 0.0021010637283325195, "grad_norm": 8.564002323150635, "loss": 0.4024180382490158, "time": 0.7975595951080322, "epoch": 1, "memory": 28798, "step": 1500} +{"lr": 0.0001, "data_time": 0.002309083938598633, "grad_norm": 10.933034062385559, "loss": 0.43787465393543246, "time": 0.824269437789917, "epoch": 1, "memory": 28798, "step": 1600} +{"lr": 0.0001, "data_time": 0.16104977130889891, "grad_norm": 14.262428903579712, "loss": 0.4117710143327713, "time": 0.7807987689971924, "epoch": 1, "memory": 28798, "step": 1700} +{"lr": 0.0001, "data_time": 0.3967538118362427, "grad_norm": 6.842606687545777, "loss": 0.40439617037773135, "time": 0.971693754196167, "epoch": 1, "memory": 28798, "step": 1800} +{"lr": 0.0001, "data_time": 0.2468998908996582, "grad_norm": 6.033211255073548, "loss": 0.3812806487083435, "time": 0.8953802108764648, "epoch": 1, "memory": 28798, "step": 1900} +{"lr": 0.0001, "data_time": 0.3589166164398193, "grad_norm": 11.163988208770752, "loss": 0.3766529768705368, "time": 0.8789925098419189, "epoch": 1, "memory": 28798, "step": 2000} +{"lr": 0.0001, "data_time": 0.32769558429718015, "grad_norm": 6.921413576602935, "loss": 0.369454437494278, "time": 0.8403992176055908, "epoch": 1, "memory": 28798, "step": 2100} +{"lr": 0.0001, "data_time": 0.3894771337509155, "grad_norm": 16.07062668800354, "loss": 0.3762036055326462, "time": 0.9089213132858276, "epoch": 1, "memory": 28798, "step": 2200} +{"lr": 0.0001, "data_time": 0.31716940402984617, "grad_norm": 10.965106916427612, "loss": 0.3938374280929565, "time": 0.8222328186035156, "epoch": 1, "memory": 28798, "step": 2300} +{"lr": 0.0001, "data_time": 0.4321016788482666, "grad_norm": 11.419286108016967, "loss": 0.38529468774795533, "time": 0.9546619892120362, "epoch": 1, "memory": 28798, "step": 2400} +{"lr": 0.0001, "data_time": 0.4447009801864624, "grad_norm": 20.4749276638031, "loss": 0.39668129980564115, "time": 0.984907865524292, "epoch": 1, "memory": 28798, "step": 2500} +{"lr": 0.0001, "data_time": 0.3377225399017334, "grad_norm": 8.161865782737731, "loss": 0.34652950465679166, "time": 0.8513520002365113, "epoch": 1, "memory": 28798, "step": 2600} +{"lr": 0.0001, "data_time": 0.4989269018173218, "grad_norm": 4.820529353618622, "loss": 0.35654892921447756, "time": 1.0519294500350953, "epoch": 1, "memory": 28798, "step": 2700} +{"lr": 0.0001, "data_time": 0.2992296934127808, "grad_norm": 17.96520357131958, "loss": 0.36084577441215515, "time": 0.8058756589889526, "epoch": 1, "memory": 28798, "step": 2800} +{"lr": 0.0001, "data_time": 0.33991820812225343, "grad_norm": 14.215149307250977, "loss": 0.3633415728807449, "time": 0.8674774169921875, "epoch": 1, "memory": 28798, "step": 2900} +{"lr": 0.0001, "data_time": 0.3959709882736206, "grad_norm": 7.207492160797119, "loss": 0.3425798386335373, "time": 0.9157244920730591, "epoch": 1, "memory": 28798, "step": 3000} +{"lr": 0.0001, "data_time": 0.4099679946899414, "grad_norm": 11.321051740646363, "loss": 0.3522310942411423, "time": 0.9410527467727661, "epoch": 1, "memory": 28798, "step": 3100} +{"lr": 0.0001, "data_time": 0.2561708211898804, "grad_norm": 14.380841886997223, "loss": 0.3565657764673233, "time": 0.7644975662231446, "epoch": 1, "memory": 28798, "step": 3200} +{"lr": 0.0001, "data_time": 0.34224073886871337, "grad_norm": 13.345923590660096, "loss": 0.3287695676088333, "time": 0.8484790325164795, "epoch": 1, "memory": 28798, "step": 3300} +{"lr": 0.0001, "data_time": 0.3237328052520752, "grad_norm": 8.821040737628937, "loss": 0.31754212975502016, "time": 0.8322324275970459, "epoch": 1, "memory": 28798, "step": 3400} +{"lr": 0.0001, "data_time": 0.4927107572555542, "grad_norm": 6.542399203777313, "loss": 0.33546442091464995, "time": 1.027521538734436, "epoch": 1, "memory": 28798, "step": 3500} +{"lr": 0.0001, "data_time": 0.2770772218704224, "grad_norm": 11.564435482025146, "loss": 0.32049880623817445, "time": 0.8759184122085572, "epoch": 1, "memory": 28798, "step": 3600} +{"lr": 0.0001, "data_time": 0.09506022930145264, "grad_norm": 5.517071306705475, "loss": 0.3160597801208496, "time": 0.8441982984542846, "epoch": 1, "memory": 28798, "step": 3700} +{"lr": 0.0001, "data_time": 0.002781963348388672, "grad_norm": 25.570412254333498, "loss": 0.40939798653125764, "time": 0.8697173118591308, "epoch": 1, "memory": 28798, "step": 3800} +{"lr": 0.0001, "data_time": 0.0017022132873535157, "grad_norm": 14.403342628479004, "loss": 0.3184275537729263, "time": 0.7843376874923706, "epoch": 1, "memory": 28798, "step": 3900} +{"lr": 0.0001, "data_time": 0.002773571014404297, "grad_norm": 13.11492998600006, "loss": 0.3069653451442719, "time": 0.867843747138977, "epoch": 1, "memory": 28798, "step": 4000} +{"lr": 0.0001, "data_time": 0.002892303466796875, "grad_norm": 13.996675157546997, "loss": 0.32083846628665924, "time": 0.9147140979766846, "epoch": 1, "memory": 28798, "step": 4100} +{"lr": 0.0001, "data_time": 0.002616620063781738, "grad_norm": 4.992052376270294, "loss": 0.30638874471187594, "time": 1.0079552412033081, "epoch": 1, "memory": 28798, "step": 4200} +{"lr": 0.0001, "data_time": 0.0022130727767944334, "grad_norm": 17.318383026123048, "loss": 0.3292596787214279, "time": 0.8624425172805786, "epoch": 1, "memory": 28798, "step": 4300} +{"lr": 0.0001, "data_time": 0.003058719635009766, "grad_norm": 7.483843278884888, "loss": 0.3052514672279358, "time": 0.87148118019104, "epoch": 1, "memory": 28798, "step": 4400} +{"lr": 0.0001, "data_time": 0.0025171279907226563, "grad_norm": 17.134615588188172, "loss": 0.3356024086475372, "time": 0.9511116027832032, "epoch": 1, "memory": 28798, "step": 4500} +{"lr": 0.0001, "data_time": 0.001605391502380371, "grad_norm": 7.5547421932220455, "loss": 0.30554085671901704, "time": 0.8072681665420532, "epoch": 1, "memory": 28798, "step": 4600} +{"lr": 0.0001, "data_time": 0.0018180608749389648, "grad_norm": 12.283394455909729, "loss": 0.3418678015470505, "time": 0.8453145265579224, "epoch": 1, "memory": 28798, "step": 4700} +{"lr": 0.0001, "data_time": 0.005225729942321777, "grad_norm": 11.814648818969726, "loss": 0.3138525366783142, "time": 0.8357168197631836, "epoch": 1, "memory": 28798, "step": 4800} +{"lr": 0.0001, "data_time": 0.0026343822479248046, "grad_norm": 19.7075412273407, "loss": 0.33815790712833405, "time": 0.9410369157791137, "epoch": 1, "memory": 28798, "step": 4900} +{"lr": 0.0001, "data_time": 0.0018437862396240234, "grad_norm": 15.569157981872559, "loss": 0.34495705366134644, "time": 0.8171157360076904, "epoch": 1, "memory": 28798, "step": 5000} +{"lr": 0.0001, "data_time": 0.002482438087463379, "grad_norm": 21.3685173869133, "loss": 0.323750838637352, "time": 0.8111368656158447, "epoch": 1, "memory": 28798, "step": 5100} +{"lr": 0.0001, "data_time": 0.001826310157775879, "grad_norm": 10.369468212127686, "loss": 0.2822732239961624, "time": 0.8334531307220459, "epoch": 1, "memory": 28798, "step": 5200} +{"lr": 0.0001, "data_time": 0.0018332242965698243, "grad_norm": 16.070834732055665, "loss": 0.3088524341583252, "time": 0.979086446762085, "epoch": 1, "memory": 28798, "step": 5300} +{"lr": 0.0001, "data_time": 0.0031427860260009764, "grad_norm": 8.568836116790772, "loss": 0.28733194768428805, "time": 0.9507247924804687, "epoch": 1, "memory": 28798, "step": 5400} +{"lr": 0.0001, "data_time": 0.0032306671142578124, "grad_norm": 5.664257299900055, "loss": 0.27612653076648713, "time": 0.9241308450698853, "epoch": 1, "memory": 28798, "step": 5500} +{"lr": 0.0001, "data_time": 0.006642651557922363, "grad_norm": 6.408056902885437, "loss": 0.2865743860602379, "time": 1.0534418821334839, "epoch": 1, "memory": 28798, "step": 5600} +{"lr": 0.0001, "data_time": 0.0032753705978393554, "grad_norm": 6.494708776473999, "loss": 0.2806232213973999, "time": 0.8970215559005738, "epoch": 1, "memory": 28798, "step": 5700} +{"accuracy/top1": 98.81320190429688, "data_time": 0.37856800854206085, "time": 1.0039068758487701, "step": 1} +{"lr": 9.779754323328192e-05, "data_time": 0.38404159545898436, "grad_norm": 4.161892879009247, "loss": 0.27686722576618195, "time": 1.0489650011062621, "epoch": 2, "memory": 28798, "step": 5858} +{"lr": 9.779754323328192e-05, "data_time": 0.1485985517501831, "grad_norm": 4.189602971076965, "loss": 0.2639403223991394, "time": 0.8727232456207276, "epoch": 2, "memory": 28783, "step": 5958} +{"lr": 9.779754323328192e-05, "data_time": 0.0027503252029418947, "grad_norm": 5.856673264503479, "loss": 0.2715042933821678, "time": 0.944049620628357, "epoch": 2, "memory": 28783, "step": 6058} +{"lr": 9.779754323328192e-05, "data_time": 0.14574897289276123, "grad_norm": 12.224393248558044, "loss": 0.28722442239522933, "time": 0.9152533769607544, "epoch": 2, "memory": 28783, "step": 6158} +{"lr": 9.779754323328192e-05, "data_time": 0.3519598960876465, "grad_norm": 6.210837030410767, "loss": 0.26650336682796477, "time": 0.8882764101028442, "epoch": 2, "memory": 28783, "step": 6258} +{"lr": 9.779754323328192e-05, "data_time": 0.3771564722061157, "grad_norm": 3.9214691162109374, "loss": 0.2589657470583916, "time": 0.8983821630477905, "epoch": 2, "memory": 28783, "step": 6358} +{"lr": 9.779754323328192e-05, "data_time": 0.24103910923004152, "grad_norm": 9.535576891899108, "loss": 0.27185446619987486, "time": 0.7520332098007202, "epoch": 2, "memory": 28783, "step": 6458} +{"lr": 9.779754323328192e-05, "data_time": 0.3728133201599121, "grad_norm": 18.79650083780289, "loss": 0.284046670794487, "time": 0.8914100170135498, "epoch": 2, "memory": 28783, "step": 6558} +{"lr": 9.779754323328192e-05, "data_time": 0.4151688814163208, "grad_norm": 5.077081775665283, "loss": 0.2608530417084694, "time": 0.9224797010421752, "epoch": 2, "memory": 28783, "step": 6658} +{"lr": 9.779754323328192e-05, "data_time": 0.4156410455703735, "grad_norm": 15.912617897987365, "loss": 0.2787394493818283, "time": 0.9759310960769654, "epoch": 2, "memory": 28783, "step": 6758} +{"lr": 9.779754323328192e-05, "data_time": 0.4599510669708252, "grad_norm": 3.741653895378113, "loss": 0.2555323660373688, "time": 0.9699604988098145, "epoch": 2, "memory": 28783, "step": 6858} +{"lr": 9.779754323328192e-05, "data_time": 0.41335225105285645, "grad_norm": 3.530123460292816, "loss": 0.24913543313741685, "time": 0.9404096364974975, "epoch": 2, "memory": 28783, "step": 6958} +{"lr": 9.779754323328192e-05, "data_time": 0.47891817092895506, "grad_norm": 3.114583659172058, "loss": 0.2537370786070824, "time": 0.9979316711425781, "epoch": 2, "memory": 28783, "step": 7058} +{"lr": 9.779754323328192e-05, "data_time": 0.3670587301254272, "grad_norm": 9.071798992156982, "loss": 0.2690409183502197, "time": 0.8926671028137207, "epoch": 2, "memory": 28783, "step": 7158} +{"lr": 9.779754323328192e-05, "data_time": 0.37719457149505614, "grad_norm": 5.893570971488953, "loss": 0.24558966606855392, "time": 0.9014822721481324, "epoch": 2, "memory": 28783, "step": 7258} +{"lr": 9.779754323328192e-05, "data_time": 0.3808925628662109, "grad_norm": 5.37038791179657, "loss": 0.24915929287672042, "time": 0.9220894336700439, "epoch": 2, "memory": 28783, "step": 7358} +{"lr": 9.779754323328192e-05, "data_time": 0.4557778358459473, "grad_norm": 8.520963859558105, "loss": 0.2678984612226486, "time": 0.9937985897064209, "epoch": 2, "memory": 28783, "step": 7458} +{"lr": 9.779754323328192e-05, "data_time": 0.3625136137008667, "grad_norm": 2.5578632593154906, "loss": 0.24371749758720399, "time": 0.8921395301818847, "epoch": 2, "memory": 28783, "step": 7558} +{"lr": 9.779754323328192e-05, "data_time": 0.38992717266082766, "grad_norm": 4.372581660747528, "loss": 0.2427796557545662, "time": 0.9112889051437378, "epoch": 2, "memory": 28783, "step": 7658} +{"lr": 9.779754323328192e-05, "data_time": 0.5067874431610108, "grad_norm": 4.538609194755554, "loss": 0.24747199416160584, "time": 1.0188977479934693, "epoch": 2, "memory": 28783, "step": 7758} +{"lr": 9.779754323328192e-05, "data_time": 0.32326455116271974, "grad_norm": 5.38339084982872, "loss": 0.24430325925350188, "time": 0.8366075754165649, "epoch": 2, "memory": 28783, "step": 7858} +{"lr": 9.779754323328192e-05, "data_time": 0.33705978393554686, "grad_norm": 8.676961934566497, "loss": 0.2449193611741066, "time": 0.847307276725769, "epoch": 2, "memory": 28783, "step": 7958} +{"lr": 9.779754323328192e-05, "data_time": 0.4482605218887329, "grad_norm": 8.488741207122803, "loss": 0.24980992674827576, "time": 0.9851110219955445, "epoch": 2, "memory": 28783, "step": 8058} +{"lr": 9.779754323328192e-05, "data_time": 0.29235076904296875, "grad_norm": 2.4637938380241393, "loss": 0.24306034594774245, "time": 0.8088602066040039, "epoch": 2, "memory": 28783, "step": 8158} +{"lr": 9.779754323328192e-05, "data_time": 0.47416865825653076, "grad_norm": 4.897786259651184, "loss": 0.24559639841318132, "time": 0.9933578968048096, "epoch": 2, "memory": 28783, "step": 8258} +{"lr": 9.779754323328192e-05, "data_time": 0.31468679904937746, "grad_norm": 4.063173985481262, "loss": 0.23778026849031447, "time": 0.858015775680542, "epoch": 2, "memory": 28783, "step": 8358} +{"lr": 9.779754323328192e-05, "data_time": 0.2870336055755615, "grad_norm": 3.8485030055046083, "loss": 0.23274193704128265, "time": 0.7924270629882812, "epoch": 2, "memory": 28783, "step": 8458} +{"lr": 9.779754323328192e-05, "data_time": 0.46350076198577883, "grad_norm": 2.754397398233414, "loss": 0.2309154450893402, "time": 0.9834944486618042, "epoch": 2, "memory": 28783, "step": 8558} +{"lr": 9.779754323328192e-05, "data_time": 0.4629937171936035, "grad_norm": 2.449435019493103, "loss": 0.2383464351296425, "time": 0.9932717084884644, "epoch": 2, "memory": 28783, "step": 8658} +{"lr": 9.779754323328192e-05, "data_time": 0.41739070415496826, "grad_norm": 8.001243257522583, "loss": 0.23662171363830567, "time": 0.9370226383209228, "epoch": 2, "memory": 28783, "step": 8758} +{"lr": 9.779754323328192e-05, "data_time": 0.4663546562194824, "grad_norm": 2.121528911590576, "loss": 0.22635733634233474, "time": 0.9992094993591308, "epoch": 2, "memory": 28783, "step": 8858} +{"lr": 9.779754323328192e-05, "data_time": 0.309641695022583, "grad_norm": 9.506640934944153, "loss": 0.22828049659729005, "time": 0.8181777000427246, "epoch": 2, "memory": 28783, "step": 8958} +{"lr": 9.779754323328192e-05, "data_time": 0.39505982398986816, "grad_norm": 7.051111888885498, "loss": 0.2564105600118637, "time": 0.9207224369049072, "epoch": 2, "memory": 28783, "step": 9058} +{"lr": 9.779754323328192e-05, "data_time": 0.33772108554840086, "grad_norm": 2.7572768568992614, "loss": 0.23249821215867997, "time": 0.8559291362762451, "epoch": 2, "memory": 28783, "step": 9158} +{"lr": 9.779754323328192e-05, "data_time": 0.48703229427337646, "grad_norm": 2.758994436264038, "loss": 0.23604095578193665, "time": 1.0598036527633667, "epoch": 2, "memory": 28783, "step": 9258} +{"lr": 9.779754323328192e-05, "data_time": 0.4315407991409302, "grad_norm": 2.9783724665641786, "loss": 0.22422760277986525, "time": 0.9436061143875122, "epoch": 2, "memory": 28783, "step": 9358} +{"lr": 9.779754323328192e-05, "data_time": 0.3649235486984253, "grad_norm": 3.7142862677574158, "loss": 0.2300599217414856, "time": 0.8965791940689087, "epoch": 2, "memory": 28783, "step": 9458} +{"lr": 9.779754323328192e-05, "data_time": 0.3484864473342896, "grad_norm": 4.165369379520416, "loss": 0.22893126308918, "time": 0.850684642791748, "epoch": 2, "memory": 28783, "step": 9558} +{"lr": 9.779754323328192e-05, "data_time": 0.399152946472168, "grad_norm": 4.1010830998420715, "loss": 0.23013428449630738, "time": 0.9175976276397705, "epoch": 2, "memory": 28783, "step": 9658} +{"lr": 9.779754323328192e-05, "data_time": 0.3455408811569214, "grad_norm": 4.721993350982666, "loss": 0.22365710586309434, "time": 0.8500763654708863, "epoch": 2, "memory": 28783, "step": 9758} +{"lr": 9.779754323328192e-05, "data_time": 0.4002149343490601, "grad_norm": 3.0715201020240785, "loss": 0.23425911366939545, "time": 0.9193668842315674, "epoch": 2, "memory": 28783, "step": 9858} +{"lr": 9.779754323328192e-05, "data_time": 0.31200993061065674, "grad_norm": 1.6474700093269348, "loss": 0.22240086942911147, "time": 0.8182117462158203, "epoch": 2, "memory": 28783, "step": 9958} +{"lr": 9.779754323328192e-05, "data_time": 0.3542064905166626, "grad_norm": 2.732310104370117, "loss": 0.220636785030365, "time": 0.8919193983078003, "epoch": 2, "memory": 28783, "step": 10058} +{"lr": 9.779754323328192e-05, "data_time": 0.43896524906158446, "grad_norm": 2.871346414089203, "loss": 0.21896641999483107, "time": 0.9701106786727905, "epoch": 2, "memory": 28783, "step": 10158} +{"lr": 9.779754323328192e-05, "data_time": 0.3461632251739502, "grad_norm": 1.8939796268939972, "loss": 0.22449996918439866, "time": 0.8784794330596923, "epoch": 2, "memory": 28783, "step": 10258} +{"lr": 9.779754323328192e-05, "data_time": 0.36941554546356203, "grad_norm": 2.0798502802848815, "loss": 0.22079380601644516, "time": 0.8904143095016479, "epoch": 2, "memory": 28783, "step": 10358} +{"lr": 9.779754323328192e-05, "data_time": 0.436677360534668, "grad_norm": 3.5731325268745424, "loss": 0.23058802634477615, "time": 0.9835659503936768, "epoch": 2, "memory": 28783, "step": 10458} +{"lr": 9.779754323328192e-05, "data_time": 0.35236248970031736, "grad_norm": 1.923442804813385, "loss": 0.22004744857549668, "time": 0.8601249694824219, "epoch": 2, "memory": 28783, "step": 10558} +{"lr": 9.779754323328192e-05, "data_time": 0.4356717109680176, "grad_norm": 1.6966484546661378, "loss": 0.2197301283478737, "time": 0.9548774719238281, "epoch": 2, "memory": 28783, "step": 10658} +{"lr": 9.779754323328192e-05, "data_time": 0.2778660297393799, "grad_norm": 2.028006798028946, "loss": 0.2189297616481781, "time": 0.7813084602355957, "epoch": 2, "memory": 28783, "step": 10758} +{"lr": 9.779754323328192e-05, "data_time": 0.3047685384750366, "grad_norm": 2.468812119960785, "loss": 0.2178144261240959, "time": 0.8396297216415405, "epoch": 2, "memory": 28783, "step": 10858} +{"lr": 9.779754323328192e-05, "data_time": 0.3942467212677002, "grad_norm": 1.7278743147850038, "loss": 0.21907654255628586, "time": 0.9061642408370971, "epoch": 2, "memory": 28783, "step": 10958} +{"lr": 9.779754323328192e-05, "data_time": 0.33275651931762695, "grad_norm": 3.3575263261795043, "loss": 0.21982483863830565, "time": 0.8706469535827637, "epoch": 2, "memory": 28783, "step": 11058} +{"lr": 9.779754323328192e-05, "data_time": 0.38285465240478517, "grad_norm": 2.402519243955612, "loss": 0.21914853155612946, "time": 0.9026891231536865, "epoch": 2, "memory": 28783, "step": 11158} +{"lr": 9.779754323328192e-05, "data_time": 0.45121476650238035, "grad_norm": 1.9819308876991273, "loss": 0.21823106110095977, "time": 0.9740733623504638, "epoch": 2, "memory": 28783, "step": 11258} +{"lr": 9.779754323328192e-05, "data_time": 0.3432604789733887, "grad_norm": 1.8669257402420043, "loss": 0.21990397274494172, "time": 0.858852219581604, "epoch": 2, "memory": 28783, "step": 11358} +{"lr": 9.779754323328192e-05, "data_time": 0.2904154062271118, "grad_norm": 1.7137621462345123, "loss": 0.21675174683332443, "time": 0.7964584827423096, "epoch": 2, "memory": 28783, "step": 11458} +{"accuracy/top1": 99.50760650634766, "data_time": 0.24432659149169922, "time": 0.6612234957077924, "step": 2} +{"lr": 9.140576474687264e-05, "data_time": 0.34173638820648194, "grad_norm": 4.991871964931488, "loss": 0.22893251180648805, "time": 0.9203919887542724, "epoch": 3, "memory": 28783, "step": 11616} +{"lr": 9.140576474687264e-05, "data_time": 0.3464978218078613, "grad_norm": 1.480863666534424, "loss": 0.2140223801136017, "time": 0.9392688751220704, "epoch": 3, "memory": 28783, "step": 11716} +{"lr": 9.140576474687264e-05, "data_time": 0.4532526254653931, "grad_norm": 3.467792344093323, "loss": 0.2252451717853546, "time": 0.9688373327255249, "epoch": 3, "memory": 28783, "step": 11816} +{"lr": 9.140576474687264e-05, "data_time": 0.5468847990036011, "grad_norm": 1.8249448239803314, "loss": 0.21538495272397995, "time": 1.082854652404785, "epoch": 3, "memory": 28783, "step": 11916} +{"lr": 9.140576474687264e-05, "data_time": 0.39746015071868895, "grad_norm": 1.5756512761116028, "loss": 0.21620576083660126, "time": 0.9389734983444213, "epoch": 3, "memory": 28783, "step": 12016} +{"lr": 9.140576474687264e-05, "data_time": 0.3889593601226807, "grad_norm": 3.1359798312187195, "loss": 0.217052561044693, "time": 0.9061183214187623, "epoch": 3, "memory": 28783, "step": 12116} +{"lr": 9.140576474687264e-05, "data_time": 0.3880429983139038, "grad_norm": 2.338692659139633, "loss": 0.2130947843194008, "time": 0.9091630220413208, "epoch": 3, "memory": 28783, "step": 12216} +{"lr": 9.140576474687264e-05, "data_time": 0.4014167070388794, "grad_norm": 2.50209698677063, "loss": 0.21859853118658065, "time": 0.9058146238327026, "epoch": 3, "memory": 28783, "step": 12316} +{"lr": 9.140576474687264e-05, "data_time": 0.4401841163635254, "grad_norm": 1.5620913565158845, "loss": 0.20996534675359727, "time": 0.991347336769104, "epoch": 3, "memory": 28783, "step": 12416} +{"lr": 9.140576474687264e-05, "data_time": 0.3722501039505005, "grad_norm": 2.1559678494930266, "loss": 0.2115705728530884, "time": 0.8778233051300048, "epoch": 3, "memory": 28783, "step": 12516} +{"lr": 9.140576474687264e-05, "data_time": 0.42272875308990476, "grad_norm": 3.18854238986969, "loss": 0.2241248294711113, "time": 0.9526077508926392, "epoch": 3, "memory": 28783, "step": 12616} +{"lr": 9.140576474687264e-05, "data_time": 0.40594820976257323, "grad_norm": 1.6028058171272277, "loss": 0.2145908147096634, "time": 0.9364082336425781, "epoch": 3, "memory": 28783, "step": 12716} +{"lr": 9.140576474687264e-05, "data_time": 0.34771692752838135, "grad_norm": 1.650160163640976, "loss": 0.21283124387264252, "time": 0.8681785583496093, "epoch": 3, "memory": 28783, "step": 12816} +{"lr": 9.140576474687264e-05, "data_time": 0.41398632526397705, "grad_norm": 2.017889451980591, "loss": 0.21516158878803254, "time": 0.928774070739746, "epoch": 3, "memory": 28783, "step": 12916} +{"lr": 9.140576474687264e-05, "data_time": 0.39421517848968507, "grad_norm": 1.2642747581005096, "loss": 0.21430819779634475, "time": 0.9186938285827637, "epoch": 3, "memory": 28783, "step": 13016} +{"lr": 9.140576474687264e-05, "data_time": 0.4297230958938599, "grad_norm": 2.902486139535904, "loss": 0.21576231420040132, "time": 0.9468563795089722, "epoch": 3, "memory": 28783, "step": 13116} +{"lr": 9.140576474687264e-05, "data_time": 0.42093353271484374, "grad_norm": 1.669806843996048, "loss": 0.21110720038414002, "time": 0.9416931629180908, "epoch": 3, "memory": 28783, "step": 13216} +{"lr": 9.140576474687264e-05, "data_time": 0.41459462642669676, "grad_norm": 1.8726254403591156, "loss": 0.2083861544728279, "time": 0.9460616111755371, "epoch": 3, "memory": 28783, "step": 13316} +{"lr": 9.140576474687264e-05, "data_time": 0.36772024631500244, "grad_norm": 2.3280391097068787, "loss": 0.21353710889816285, "time": 0.8850645542144775, "epoch": 3, "memory": 28783, "step": 13416} +{"lr": 9.140576474687264e-05, "data_time": 0.4986832618713379, "grad_norm": 2.1439351558685305, "loss": 0.212885282933712, "time": 1.0227462291717528, "epoch": 3, "memory": 28783, "step": 13516} +{"lr": 9.140576474687264e-05, "data_time": 0.4322330474853516, "grad_norm": 1.7114878594875336, "loss": 0.21144051998853683, "time": 0.9611533403396606, "epoch": 3, "memory": 28783, "step": 13616} +{"lr": 9.140576474687264e-05, "data_time": 0.327402663230896, "grad_norm": 1.5761158168315887, "loss": 0.21120853275060653, "time": 0.8291275024414062, "epoch": 3, "memory": 28783, "step": 13716} +{"lr": 9.140576474687264e-05, "data_time": 0.33883545398712156, "grad_norm": 1.2156826674938201, "loss": 0.21361772418022157, "time": 0.863519811630249, "epoch": 3, "memory": 28783, "step": 13816} +{"lr": 9.140576474687264e-05, "data_time": 0.3207821846008301, "grad_norm": 1.9318087100982666, "loss": 0.21138177514076234, "time": 0.8583831071853638, "epoch": 3, "memory": 28783, "step": 13916} +{"lr": 9.140576474687264e-05, "data_time": 0.4241142749786377, "grad_norm": 2.098145824670792, "loss": 0.21452728807926177, "time": 0.9520506858825684, "epoch": 3, "memory": 28783, "step": 14016} +{"lr": 9.140576474687264e-05, "data_time": 0.43827433586120607, "grad_norm": 1.5923111021518708, "loss": 0.21286115199327468, "time": 0.9630373954772949, "epoch": 3, "memory": 28783, "step": 14116} +{"lr": 9.140576474687264e-05, "data_time": 0.450687050819397, "grad_norm": 1.4236567795276642, "loss": 0.20501015186309815, "time": 1.00359787940979, "epoch": 3, "memory": 28783, "step": 14216} +{"lr": 9.140576474687264e-05, "data_time": 0.41872968673706057, "grad_norm": 1.32694154381752, "loss": 0.2091246575117111, "time": 0.9462849855422973, "epoch": 3, "memory": 28783, "step": 14316} +{"lr": 9.140576474687264e-05, "data_time": 0.2802180528640747, "grad_norm": 1.640733528137207, "loss": 0.20935942977666855, "time": 0.7958567619323731, "epoch": 3, "memory": 28783, "step": 14416} +{"lr": 9.140576474687264e-05, "data_time": 0.3091710090637207, "grad_norm": 1.1378621339797974, "loss": 0.20767311453819276, "time": 0.8174018383026123, "epoch": 3, "memory": 28783, "step": 14516} +{"lr": 9.140576474687264e-05, "data_time": 0.32596402168273925, "grad_norm": 1.2323434114456178, "loss": 0.21011255532503129, "time": 0.8540189743041993, "epoch": 3, "memory": 28783, "step": 14616} +{"lr": 9.140576474687264e-05, "data_time": 0.37062528133392336, "grad_norm": 1.0656420469284058, "loss": 0.21032456159591675, "time": 0.8760164499282836, "epoch": 3, "memory": 28783, "step": 14716} +{"lr": 9.140576474687264e-05, "data_time": 0.37595129013061523, "grad_norm": 1.8807616651058197, "loss": 0.21050831228494643, "time": 0.8942030668258667, "epoch": 3, "memory": 28783, "step": 14816} +{"lr": 9.140576474687264e-05, "data_time": 0.4818445682525635, "grad_norm": 1.8256277203559876, "loss": 0.21127363294363022, "time": 1.016457486152649, "epoch": 3, "memory": 28783, "step": 14916} +{"lr": 9.140576474687264e-05, "data_time": 0.41287271976470946, "grad_norm": 1.2120871603488923, "loss": 0.21269291490316392, "time": 0.9386276006698608, "epoch": 3, "memory": 28783, "step": 15016} +{"lr": 9.140576474687264e-05, "data_time": 0.4050600051879883, "grad_norm": 2.5419181048870088, "loss": 0.21512603759765625, "time": 0.9259554386138916, "epoch": 3, "memory": 28783, "step": 15116} +{"lr": 9.140576474687264e-05, "data_time": 0.51962730884552, "grad_norm": 2.4268842816352842, "loss": 0.21307655572891235, "time": 1.0452313661575316, "epoch": 3, "memory": 28783, "step": 15216} +{"lr": 9.140576474687264e-05, "data_time": 0.42670629024505613, "grad_norm": 1.5698530793190002, "loss": 0.21473020911216736, "time": 0.9553998231887817, "epoch": 3, "memory": 28783, "step": 15316} +{"lr": 9.140576474687264e-05, "data_time": 0.44242675304412843, "grad_norm": 1.3581580519676208, "loss": 0.20926800817251207, "time": 0.9724407911300659, "epoch": 3, "memory": 28783, "step": 15416} +{"lr": 9.140576474687264e-05, "data_time": 0.4571129083633423, "grad_norm": 1.8173506081104278, "loss": 0.21171002835035324, "time": 0.987190580368042, "epoch": 3, "memory": 28783, "step": 15516} +{"lr": 9.140576474687264e-05, "data_time": 0.340023398399353, "grad_norm": 1.3211472660303116, "loss": 0.20899121910333635, "time": 0.8520113468170166, "epoch": 3, "memory": 28783, "step": 15616} +{"lr": 9.140576474687264e-05, "data_time": 0.40095632076263427, "grad_norm": 1.5702925741672515, "loss": 0.2115289032459259, "time": 0.9212728977203369, "epoch": 3, "memory": 28783, "step": 15716} +{"lr": 9.140576474687264e-05, "data_time": 0.22360050678253174, "grad_norm": 1.5953031480312347, "loss": 0.20678865164518356, "time": 0.7282232046127319, "epoch": 3, "memory": 28783, "step": 15816} +{"lr": 9.140576474687264e-05, "data_time": 0.2912003993988037, "grad_norm": 0.9083232283592224, "loss": 0.20775903910398483, "time": 0.8021667957305908, "epoch": 3, "memory": 28783, "step": 15916} +{"lr": 9.140576474687264e-05, "data_time": 0.3436312198638916, "grad_norm": 1.3005921363830566, "loss": 0.20833759009838104, "time": 0.8808267831802368, "epoch": 3, "memory": 28783, "step": 16016} +{"lr": 9.140576474687264e-05, "data_time": 0.30919647216796875, "grad_norm": 1.743823230266571, "loss": 0.21149243265390397, "time": 0.8162663221359253, "epoch": 3, "memory": 28783, "step": 16116} +{"lr": 9.140576474687264e-05, "data_time": 0.26744005680084226, "grad_norm": 1.4218769907951354, "loss": 0.20726995170116425, "time": 0.8309375762939453, "epoch": 3, "memory": 28783, "step": 16216} +{"lr": 9.140576474687264e-05, "data_time": 0.1598140001296997, "grad_norm": 1.2673326671123504, "loss": 0.209407040476799, "time": 0.7453443527221679, "epoch": 3, "memory": 28783, "step": 16316} +{"lr": 9.140576474687264e-05, "data_time": 0.08368620872497559, "grad_norm": 1.2609032690525055, "loss": 0.21325397491455078, "time": 0.8645741462707519, "epoch": 3, "memory": 28783, "step": 16416} +{"lr": 9.140576474687264e-05, "data_time": 0.012835836410522461, "grad_norm": 1.088646912574768, "loss": 0.2061121016740799, "time": 0.7909500122070312, "epoch": 3, "memory": 28783, "step": 16516} +{"lr": 9.140576474687264e-05, "data_time": 0.0017733335494995116, "grad_norm": 0.8696327984333039, "loss": 0.20412105768918992, "time": 0.7758328199386597, "epoch": 3, "memory": 28783, "step": 16616} +{"lr": 9.140576474687264e-05, "data_time": 0.0023193597793579102, "grad_norm": 1.2496171474456788, "loss": 0.20469192266464234, "time": 0.9271472215652465, "epoch": 3, "memory": 28783, "step": 16716} +{"lr": 9.140576474687264e-05, "data_time": 0.002138352394104004, "grad_norm": 1.030261254310608, "loss": 0.20918433517217636, "time": 0.8781866788864136, "epoch": 3, "memory": 28783, "step": 16816} +{"lr": 9.140576474687264e-05, "data_time": 0.0014387845993041992, "grad_norm": 2.07956263422966, "loss": 0.21559395939111708, "time": 0.7746203184127808, "epoch": 3, "memory": 28783, "step": 16916} +{"lr": 9.140576474687264e-05, "data_time": 0.0018655776977539063, "grad_norm": 1.2639388561248779, "loss": 0.20937535911798477, "time": 0.7693906307220459, "epoch": 3, "memory": 28783, "step": 17016} +{"lr": 9.140576474687264e-05, "data_time": 0.002580690383911133, "grad_norm": 1.0399180829524994, "loss": 0.2064996540546417, "time": 0.9988693237304688, "epoch": 3, "memory": 28783, "step": 17116} +{"lr": 9.140576474687264e-05, "data_time": 0.002512550354003906, "grad_norm": 1.0688959777355194, "loss": 0.21010167598724366, "time": 0.8722483396530152, "epoch": 3, "memory": 28783, "step": 17216} +{"accuracy/top1": 99.83586883544922, "data_time": 0.2442783327663646, "time": 0.6610640357522404, "step": 3} +{"lr": 8.14503363531613e-05, "data_time": 0.15842914581298828, "grad_norm": 0.9486164212226867, "loss": 0.20616671591997146, "time": 0.9271233320236206, "epoch": 4, "memory": 28783, "step": 17374} +{"lr": 8.14503363531613e-05, "data_time": 0.1748908281326294, "grad_norm": 1.2735835433006286, "loss": 0.2055667370557785, "time": 0.923888874053955, "epoch": 4, "memory": 28783, "step": 17474} +{"lr": 8.14503363531613e-05, "data_time": 0.09691615104675293, "grad_norm": 0.9428302496671677, "loss": 0.2072424590587616, "time": 0.8132968664169311, "epoch": 4, "memory": 28783, "step": 17574} +{"lr": 8.14503363531613e-05, "data_time": 0.001745295524597168, "grad_norm": 0.7736806273460388, "loss": 0.20651400983333587, "time": 0.8681283950805664, "epoch": 4, "memory": 28783, "step": 17674} +{"lr": 8.14503363531613e-05, "data_time": 0.002722954750061035, "grad_norm": 1.2201237201690673, "loss": 0.20861655175685884, "time": 0.9600642204284668, "epoch": 4, "memory": 28783, "step": 17774} +{"lr": 8.14503363531613e-05, "data_time": 0.011842823028564453, "grad_norm": 1.3274381279945373, "loss": 0.2047136589884758, "time": 0.90377197265625, "epoch": 4, "memory": 28783, "step": 17874} +{"lr": 8.14503363531613e-05, "data_time": 0.0020856142044067385, "grad_norm": 0.8235398411750794, "loss": 0.20607851147651673, "time": 0.871361780166626, "epoch": 4, "memory": 28783, "step": 17974} +{"lr": 8.14503363531613e-05, "data_time": 0.002403712272644043, "grad_norm": 1.2387137442827225, "loss": 0.20855880081653594, "time": 0.9287378549575805, "epoch": 4, "memory": 28783, "step": 18074} +{"lr": 8.14503363531613e-05, "data_time": 0.0015957117080688476, "grad_norm": 0.9570003181695939, "loss": 0.20592181980609894, "time": 0.7168264150619507, "epoch": 4, "memory": 28783, "step": 18174} +{"lr": 8.14503363531613e-05, "data_time": 0.0016207933425903321, "grad_norm": 1.2040386378765107, "loss": 0.20649221688508987, "time": 0.8188527584075928, "epoch": 4, "memory": 28783, "step": 18274} +{"lr": 8.14503363531613e-05, "data_time": 0.00239717960357666, "grad_norm": 0.8021222949028015, "loss": 0.2065666452050209, "time": 0.9109915256500244, "epoch": 4, "memory": 28783, "step": 18374} +{"lr": 8.14503363531613e-05, "data_time": 0.002166438102722168, "grad_norm": 1.0723474770784378, "loss": 0.2106533721089363, "time": 0.7592983245849609, "epoch": 4, "memory": 28783, "step": 18474} +{"lr": 8.14503363531613e-05, "data_time": 0.0025032520294189452, "grad_norm": 1.1991454631090164, "loss": 0.20588413774967193, "time": 0.9460506200790405, "epoch": 4, "memory": 28783, "step": 18574} +{"lr": 8.14503363531613e-05, "data_time": 0.0029126644134521485, "grad_norm": 0.6547329306602478, "loss": 0.20953926295042039, "time": 0.7588682651519776, "epoch": 4, "memory": 28783, "step": 18674} +{"lr": 8.14503363531613e-05, "data_time": 0.003932929039001465, "grad_norm": 0.9340867400169373, "loss": 0.20635195970535278, "time": 0.9276320695877075, "epoch": 4, "memory": 28783, "step": 18774} +{"lr": 8.14503363531613e-05, "data_time": 0.0017466306686401366, "grad_norm": 1.3939781427383422, "loss": 0.2075723096728325, "time": 0.7641024351119995, "epoch": 4, "memory": 28783, "step": 18874} +{"lr": 8.14503363531613e-05, "data_time": 0.0025475263595581056, "grad_norm": 1.449819341301918, "loss": 0.20281902849674224, "time": 0.8952036142349243, "epoch": 4, "memory": 28783, "step": 18974} +{"lr": 8.14503363531613e-05, "data_time": 0.002364826202392578, "grad_norm": 1.2631963044404984, "loss": 0.2108501061797142, "time": 0.8860666990280152, "epoch": 4, "memory": 28783, "step": 19074} +{"lr": 8.14503363531613e-05, "data_time": 0.001476883888244629, "grad_norm": 0.8616365134716034, "loss": 0.20916038304567336, "time": 0.7153547048568726, "epoch": 4, "memory": 28783, "step": 19174} +{"lr": 8.14503363531613e-05, "data_time": 0.0018721818923950195, "grad_norm": 0.7398032575845719, "loss": 0.2032613754272461, "time": 0.8166702270507813, "epoch": 4, "memory": 28783, "step": 19274} +{"lr": 8.14503363531613e-05, "data_time": 0.0017771482467651366, "grad_norm": 0.7306557953357696, "loss": 0.2064534217119217, "time": 0.7739821910858155, "epoch": 4, "memory": 28783, "step": 19374} +{"lr": 8.14503363531613e-05, "data_time": 0.0023369073867797853, "grad_norm": 0.8454796463251114, "loss": 0.2049805149435997, "time": 0.8626639366149902, "epoch": 4, "memory": 28783, "step": 19474} +{"lr": 8.14503363531613e-05, "data_time": 0.0016691923141479493, "grad_norm": 1.1940773099660873, "loss": 0.20755469053983688, "time": 0.8436608076095581, "epoch": 4, "memory": 28783, "step": 19574} +{"lr": 8.14503363531613e-05, "data_time": 0.0051238775253295895, "grad_norm": 0.8481484055519104, "loss": 0.2040230005979538, "time": 0.9070647954940796, "epoch": 4, "memory": 28783, "step": 19674} +{"lr": 8.14503363531613e-05, "data_time": 0.0016668319702148437, "grad_norm": 1.4306426346302032, "loss": 0.21022323817014693, "time": 0.8055631160736084, "epoch": 4, "memory": 28783, "step": 19774} +{"lr": 8.14503363531613e-05, "data_time": 0.0020183324813842773, "grad_norm": 0.8969280123710632, "loss": 0.20493603646755218, "time": 0.8274871349334717, "epoch": 4, "memory": 28783, "step": 19874} +{"lr": 8.14503363531613e-05, "data_time": 0.0035917997360229493, "grad_norm": 0.9320100218057632, "loss": 0.20461425185203552, "time": 0.9778493881225586, "epoch": 4, "memory": 28783, "step": 19974} +{"lr": 8.14503363531613e-05, "data_time": 0.002080059051513672, "grad_norm": 0.9420254468917847, "loss": 0.2047571897506714, "time": 0.8735392570495606, "epoch": 4, "memory": 28783, "step": 20074} +{"lr": 8.14503363531613e-05, "data_time": 0.0018732070922851563, "grad_norm": 0.8712501719594001, "loss": 0.20343103408813476, "time": 0.8793980598449707, "epoch": 4, "memory": 28783, "step": 20174} +{"lr": 8.14503363531613e-05, "data_time": 0.003032684326171875, "grad_norm": 1.051969164609909, "loss": 0.2065539091825485, "time": 0.9040723085403443, "epoch": 4, "memory": 28783, "step": 20274} +{"lr": 8.14503363531613e-05, "data_time": 0.001804351806640625, "grad_norm": 0.7072114735841751, "loss": 0.20508118271827697, "time": 0.8119118213653564, "epoch": 4, "memory": 28783, "step": 20374} +{"lr": 8.14503363531613e-05, "data_time": 0.0019825220108032225, "grad_norm": 0.7490186542272568, "loss": 0.2030818998813629, "time": 0.8202088832855224, "epoch": 4, "memory": 28783, "step": 20474} +{"lr": 8.14503363531613e-05, "data_time": 0.0017294883728027344, "grad_norm": 1.0776535362005233, "loss": 0.2084824562072754, "time": 0.7922296524047852, "epoch": 4, "memory": 28783, "step": 20574} +{"lr": 8.14503363531613e-05, "data_time": 0.0021042823791503906, "grad_norm": 1.1144632935523986, "loss": 0.20546067804098128, "time": 0.8531909227371216, "epoch": 4, "memory": 28783, "step": 20674} +{"lr": 8.14503363531613e-05, "data_time": 0.001963210105895996, "grad_norm": 0.8492861807346344, "loss": 0.20431856513023378, "time": 0.8022182941436767, "epoch": 4, "memory": 28783, "step": 20774} +{"lr": 8.14503363531613e-05, "data_time": 0.0023144006729125975, "grad_norm": 0.8482524693012238, "loss": 0.20370493680238724, "time": 0.8173549175262451, "epoch": 4, "memory": 28783, "step": 20874} +{"lr": 8.14503363531613e-05, "data_time": 0.0019672155380249024, "grad_norm": 0.7161398589611053, "loss": 0.2066895216703415, "time": 0.8143734931945801, "epoch": 4, "memory": 28783, "step": 20974} +{"lr": 8.14503363531613e-05, "data_time": 0.0025456428527832033, "grad_norm": 0.9564333558082581, "loss": 0.20615943223237992, "time": 0.8935346364974975, "epoch": 4, "memory": 28783, "step": 21074} +{"lr": 8.14503363531613e-05, "data_time": 0.0018461227416992187, "grad_norm": 0.8982215970754623, "loss": 0.20722531378269196, "time": 0.812614107131958, "epoch": 4, "memory": 28783, "step": 21174} +{"lr": 8.14503363531613e-05, "data_time": 0.0016650676727294922, "grad_norm": 0.7342030823230743, "loss": 0.20762034952640535, "time": 0.8359655380249024, "epoch": 4, "memory": 28783, "step": 21274} +{"lr": 8.14503363531613e-05, "data_time": 0.0023229837417602537, "grad_norm": 0.9481951981782913, "loss": 0.20643476843833924, "time": 0.9659729719161987, "epoch": 4, "memory": 28783, "step": 21374} +{"lr": 8.14503363531613e-05, "data_time": 0.0034432411193847656, "grad_norm": 0.8121060967445374, "loss": 0.2057229295372963, "time": 0.9556218147277832, "epoch": 4, "memory": 28783, "step": 21474} +{"lr": 8.14503363531613e-05, "data_time": 0.0018439769744873046, "grad_norm": 0.8101279348134994, "loss": 0.20679338723421098, "time": 0.7833482980728149, "epoch": 4, "memory": 28783, "step": 21574} +{"lr": 8.14503363531613e-05, "data_time": 0.002988457679748535, "grad_norm": 1.1345936894416808, "loss": 0.20748918503522873, "time": 0.7720442771911621, "epoch": 4, "memory": 28783, "step": 21674} +{"lr": 8.14503363531613e-05, "data_time": 0.0017228126525878906, "grad_norm": 0.6462352246046066, "loss": 0.20321632772684098, "time": 0.7660947799682617, "epoch": 4, "memory": 28783, "step": 21774} +{"lr": 8.14503363531613e-05, "data_time": 0.0017563581466674804, "grad_norm": 0.6697968736290931, "loss": 0.2014237642288208, "time": 0.7498160600662231, "epoch": 4, "memory": 28783, "step": 21874} +{"lr": 8.14503363531613e-05, "data_time": 0.002251148223876953, "grad_norm": 1.2094328701496124, "loss": 0.2062023714184761, "time": 0.8809028148651123, "epoch": 4, "memory": 28783, "step": 21974} +{"lr": 8.14503363531613e-05, "data_time": 0.002000856399536133, "grad_norm": 0.5823024600744248, "loss": 0.20479606240987777, "time": 0.7701223611831665, "epoch": 4, "memory": 28783, "step": 22074} +{"lr": 8.14503363531613e-05, "data_time": 0.0024200439453125, "grad_norm": 0.9492810845375061, "loss": 0.20501076728105544, "time": 0.9245847225189209, "epoch": 4, "memory": 28783, "step": 22174} +{"lr": 8.14503363531613e-05, "data_time": 0.0019545555114746094, "grad_norm": 0.9327262550592422, "loss": 0.20240710377693177, "time": 0.8140376806259155, "epoch": 4, "memory": 28783, "step": 22274} +{"lr": 8.14503363531613e-05, "data_time": 0.002622032165527344, "grad_norm": 1.0285163223743439, "loss": 0.20348014086484909, "time": 0.9900183200836181, "epoch": 4, "memory": 28783, "step": 22374} +{"lr": 8.14503363531613e-05, "data_time": 0.002169895172119141, "grad_norm": 0.8105373054742813, "loss": 0.20507268905639647, "time": 0.8354963302612305, "epoch": 4, "memory": 28783, "step": 22474} +{"lr": 8.14503363531613e-05, "data_time": 0.0021898508071899413, "grad_norm": 0.7004149116575717, "loss": 0.2050428107380867, "time": 0.9091761350631714, "epoch": 4, "memory": 28783, "step": 22574} +{"lr": 8.14503363531613e-05, "data_time": 0.0016341447830200196, "grad_norm": 0.8098372787237167, "loss": 0.2061666041612625, "time": 0.8773354530334473, "epoch": 4, "memory": 28783, "step": 22674} +{"lr": 8.14503363531613e-05, "data_time": 0.0023780584335327147, "grad_norm": 0.7755119383335114, "loss": 0.20429756194353105, "time": 0.8351455688476562, "epoch": 4, "memory": 28783, "step": 22774} +{"lr": 8.14503363531613e-05, "data_time": 0.0016416788101196289, "grad_norm": 1.7323849946260452, "loss": 0.20490384846925735, "time": 0.8411281108856201, "epoch": 4, "memory": 28783, "step": 22874} +{"lr": 8.14503363531613e-05, "data_time": 0.0018893718719482423, "grad_norm": 1.1103879570961, "loss": 0.20811793357133865, "time": 0.81603102684021, "epoch": 4, "memory": 28783, "step": 22974} +{"accuracy/top1": 99.89899444580078, "data_time": 0.22424493116490982, "time": 0.6452209388508516, "step": 4} +{"lr": 6.890576474687264e-05, "data_time": 0.33192808628082277, "grad_norm": 0.5818910449743271, "loss": 0.20459140837192535, "time": 0.8387040615081787, "epoch": 5, "memory": 28783, "step": 23132} +{"lr": 6.890576474687264e-05, "data_time": 0.2808361053466797, "grad_norm": 0.7127429842948914, "loss": 0.20435028672218322, "time": 0.7941132545471191, "epoch": 5, "memory": 28783, "step": 23232} +{"lr": 6.890576474687264e-05, "data_time": 0.0026577234268188475, "grad_norm": 0.9764065265655517, "loss": 0.20468062907457352, "time": 0.8684720754623413, "epoch": 5, "memory": 28783, "step": 23332} +{"lr": 6.890576474687264e-05, "data_time": 0.0019591331481933595, "grad_norm": 0.7020221054553986, "loss": 0.20276259034872054, "time": 0.8258968353271484, "epoch": 5, "memory": 28783, "step": 23432} +{"lr": 6.890576474687264e-05, "data_time": 0.002506256103515625, "grad_norm": 0.7867343083024025, "loss": 0.20279908627271653, "time": 0.8995970010757446, "epoch": 5, "memory": 28783, "step": 23532} +{"lr": 6.890576474687264e-05, "data_time": 0.0021756649017333984, "grad_norm": 0.6544739693403244, "loss": 0.20398417264223098, "time": 0.9003485679626465, "epoch": 5, "memory": 28783, "step": 23632} +{"lr": 6.890576474687264e-05, "data_time": 0.0031285762786865236, "grad_norm": 0.730390340089798, "loss": 0.20151670575141906, "time": 0.882508373260498, "epoch": 5, "memory": 28783, "step": 23732} +{"lr": 6.890576474687264e-05, "data_time": 0.0019951343536376955, "grad_norm": 0.7446844309568406, "loss": 0.20467254519462585, "time": 0.7451502799987793, "epoch": 5, "memory": 28783, "step": 23832} +{"lr": 6.890576474687264e-05, "data_time": 0.001630878448486328, "grad_norm": 1.0319697707891464, "loss": 0.20675748139619826, "time": 0.8372640371322632, "epoch": 5, "memory": 28783, "step": 23932} +{"lr": 6.890576474687264e-05, "data_time": 0.0022076845169067385, "grad_norm": 1.007606989145279, "loss": 0.20434479117393495, "time": 0.918543791770935, "epoch": 5, "memory": 28783, "step": 24032} +{"lr": 6.890576474687264e-05, "data_time": 0.0021894216537475587, "grad_norm": 0.528626212477684, "loss": 0.20331996381282808, "time": 0.8008938550949096, "epoch": 5, "memory": 28783, "step": 24132} +{"lr": 6.890576474687264e-05, "data_time": 0.003458523750305176, "grad_norm": 0.6689302936196327, "loss": 0.2045394226908684, "time": 0.8632954359054565, "epoch": 5, "memory": 28783, "step": 24232} +{"lr": 6.890576474687264e-05, "data_time": 0.0016916513442993164, "grad_norm": 0.7286306858062744, "loss": 0.20205857902765273, "time": 0.7412585496902466, "epoch": 5, "memory": 28783, "step": 24332} +{"lr": 6.890576474687264e-05, "data_time": 0.001738905906677246, "grad_norm": 0.685021185874939, "loss": 0.20819469392299653, "time": 0.8424660682678222, "epoch": 5, "memory": 28783, "step": 24432} +{"lr": 6.890576474687264e-05, "data_time": 0.002035975456237793, "grad_norm": 0.6094655677676201, "loss": 0.20216030329465867, "time": 0.8036423206329346, "epoch": 5, "memory": 28783, "step": 24532} +{"lr": 6.890576474687264e-05, "data_time": 0.0019531965255737303, "grad_norm": 1.5473129957914353, "loss": 0.20610381066799163, "time": 0.7969301223754883, "epoch": 5, "memory": 28783, "step": 24632} +{"lr": 6.890576474687264e-05, "data_time": 0.003784966468811035, "grad_norm": 0.42079600393772126, "loss": 0.20122822225093842, "time": 0.941700553894043, "epoch": 5, "memory": 28783, "step": 24732} +{"lr": 6.890576474687264e-05, "data_time": 0.0020723342895507812, "grad_norm": 0.7694544076919556, "loss": 0.20376111567020416, "time": 0.8561123132705688, "epoch": 5, "memory": 28783, "step": 24832} +{"lr": 6.890576474687264e-05, "data_time": 0.001956009864807129, "grad_norm": 0.5741630673408509, "loss": 0.20102161020040513, "time": 0.9322062730789185, "epoch": 5, "memory": 28783, "step": 24932} +{"lr": 6.890576474687264e-05, "data_time": 0.002545619010925293, "grad_norm": 0.7578251883387566, "loss": 0.20165254771709443, "time": 0.9280125141143799, "epoch": 5, "memory": 28783, "step": 25032} +{"lr": 6.890576474687264e-05, "data_time": 0.0021206140518188477, "grad_norm": 0.719692063331604, "loss": 0.20198911279439927, "time": 0.7953967094421387, "epoch": 5, "memory": 28783, "step": 25132} +{"lr": 6.890576474687264e-05, "data_time": 0.0016574859619140625, "grad_norm": 0.4823285847902298, "loss": 0.20262670367956162, "time": 0.873339319229126, "epoch": 5, "memory": 28783, "step": 25232} +{"lr": 6.890576474687264e-05, "data_time": 0.002013969421386719, "grad_norm": 0.9548934191465378, "loss": 0.2040390118956566, "time": 0.8383258819580078, "epoch": 5, "memory": 28783, "step": 25332} +{"lr": 6.890576474687264e-05, "data_time": 0.0024251699447631835, "grad_norm": 0.8258676022291184, "loss": 0.20598920434713364, "time": 0.9125054121017456, "epoch": 5, "memory": 28783, "step": 25432} +{"lr": 6.890576474687264e-05, "data_time": 0.0018953800201416016, "grad_norm": 0.5936780512332916, "loss": 0.2034745842218399, "time": 0.8699544191360473, "epoch": 5, "memory": 28783, "step": 25532} +{"lr": 6.890576474687264e-05, "data_time": 0.001897907257080078, "grad_norm": 0.7664115726947784, "loss": 0.2033932939171791, "time": 0.7872421264648437, "epoch": 5, "memory": 28783, "step": 25632} +{"lr": 6.890576474687264e-05, "data_time": 0.002547621726989746, "grad_norm": 0.9103539705276489, "loss": 0.20057430416345595, "time": 0.9325079202651978, "epoch": 5, "memory": 28783, "step": 25732} +{"lr": 6.890576474687264e-05, "data_time": 0.0018838405609130859, "grad_norm": 0.6789240419864655, "loss": 0.2045421615242958, "time": 0.8318154335021972, "epoch": 5, "memory": 28783, "step": 25832} +{"lr": 6.890576474687264e-05, "data_time": 0.001972484588623047, "grad_norm": 0.8590569019317627, "loss": 0.20210847407579421, "time": 0.824900484085083, "epoch": 5, "memory": 28783, "step": 25932} +{"lr": 6.890576474687264e-05, "data_time": 0.17052371501922609, "grad_norm": 0.7384711638092994, "loss": 0.20500597208738328, "time": 0.9175719499588013, "epoch": 5, "memory": 28783, "step": 26032} +{"lr": 6.890576474687264e-05, "data_time": 0.1663076639175415, "grad_norm": 0.820941099524498, "loss": 0.20676555782556533, "time": 0.8747415781021118, "epoch": 5, "memory": 28783, "step": 26132} +{"lr": 6.890576474687264e-05, "data_time": 0.003639030456542969, "grad_norm": 0.7995735704898834, "loss": 0.2040101855993271, "time": 0.9614399194717407, "epoch": 5, "memory": 28783, "step": 26232} +{"lr": 6.890576474687264e-05, "data_time": 0.003133678436279297, "grad_norm": 0.8042361319065094, "loss": 0.20054792016744613, "time": 0.9424630403518677, "epoch": 5, "memory": 28783, "step": 26332} +{"lr": 6.890576474687264e-05, "data_time": 0.0017218828201293946, "grad_norm": 0.5530852839350701, "loss": 0.2040867730975151, "time": 0.8652605533599853, "epoch": 5, "memory": 28783, "step": 26432} +{"lr": 6.890576474687264e-05, "data_time": 0.002207183837890625, "grad_norm": 0.5613486021757126, "loss": 0.20223472714424134, "time": 0.8249388933181763, "epoch": 5, "memory": 28783, "step": 26532} +{"lr": 6.890576474687264e-05, "data_time": 0.0016343832015991212, "grad_norm": 0.8797535479068757, "loss": 0.20342277735471725, "time": 0.7633192062377929, "epoch": 5, "memory": 28783, "step": 26632} +{"lr": 6.890576474687264e-05, "data_time": 0.003683805465698242, "grad_norm": 1.0159285381436347, "loss": 0.2023201122879982, "time": 0.8931429386138916, "epoch": 5, "memory": 28783, "step": 26732} +{"lr": 6.890576474687264e-05, "data_time": 0.0019260406494140624, "grad_norm": 0.6106607258319855, "loss": 0.20391769409179689, "time": 0.790984058380127, "epoch": 5, "memory": 28783, "step": 26832} +{"lr": 6.890576474687264e-05, "data_time": 0.002742624282836914, "grad_norm": 0.839930745959282, "loss": 0.20187632143497466, "time": 0.9179551362991333, "epoch": 5, "memory": 28783, "step": 26932} +{"lr": 6.890576474687264e-05, "data_time": 0.0017934083938598634, "grad_norm": 0.6112815082073212, "loss": 0.2032255232334137, "time": 0.8257436990737915, "epoch": 5, "memory": 28783, "step": 27032} +{"lr": 6.890576474687264e-05, "data_time": 0.00274960994720459, "grad_norm": 0.6874577879905701, "loss": 0.20240055918693542, "time": 0.8879693746566772, "epoch": 5, "memory": 28783, "step": 27132} +{"lr": 6.890576474687264e-05, "data_time": 0.0017387866973876953, "grad_norm": 0.6869521290063858, "loss": 0.20243849903345107, "time": 0.880066180229187, "epoch": 5, "memory": 28783, "step": 27232} +{"lr": 6.890576474687264e-05, "data_time": 0.0035073041915893556, "grad_norm": 0.6701879560947418, "loss": 0.2014155089855194, "time": 0.8371518850326538, "epoch": 5, "memory": 28783, "step": 27332} +{"lr": 6.890576474687264e-05, "data_time": 0.0028847694396972657, "grad_norm": 0.5011413045227527, "loss": 0.20108413994312285, "time": 0.9502418518066407, "epoch": 5, "memory": 28783, "step": 27432} +{"lr": 6.890576474687264e-05, "data_time": 0.0019443988800048827, "grad_norm": 0.9512621909379959, "loss": 0.20810155868530272, "time": 0.8282215118408203, "epoch": 5, "memory": 28783, "step": 27532} +{"lr": 6.890576474687264e-05, "data_time": 0.001908278465270996, "grad_norm": 0.7863615319132805, "loss": 0.20385856926441193, "time": 0.809152626991272, "epoch": 5, "memory": 28783, "step": 27632} +{"lr": 6.890576474687264e-05, "data_time": 0.0021807193756103516, "grad_norm": 1.2226685851812362, "loss": 0.20267819464206696, "time": 0.8482498407363892, "epoch": 5, "memory": 28783, "step": 27732} +{"lr": 6.890576474687264e-05, "data_time": 0.0021015167236328124, "grad_norm": 0.5874336376786232, "loss": 0.2017254501581192, "time": 0.8229806900024415, "epoch": 5, "memory": 28783, "step": 27832} +{"lr": 6.890576474687264e-05, "data_time": 0.001706385612487793, "grad_norm": 0.39259429574012755, "loss": 0.2026368945837021, "time": 0.8157365322113037, "epoch": 5, "memory": 28783, "step": 27932} +{"lr": 6.890576474687264e-05, "data_time": 0.002717757225036621, "grad_norm": 0.6538422375917434, "loss": 0.20286358445882796, "time": 0.9589800834655762, "epoch": 5, "memory": 28783, "step": 28032} +{"lr": 6.890576474687264e-05, "data_time": 0.0013824224472045899, "grad_norm": 0.9113956540822983, "loss": 0.20145105719566345, "time": 0.7455509185791016, "epoch": 5, "memory": 28783, "step": 28132} +{"lr": 6.890576474687264e-05, "data_time": 0.0017171144485473634, "grad_norm": 0.927887174487114, "loss": 0.20362524390220643, "time": 0.8601174592971802, "epoch": 5, "memory": 28783, "step": 28232} +{"lr": 6.890576474687264e-05, "data_time": 0.0018078804016113282, "grad_norm": 0.5490619979798794, "loss": 0.20178521126508714, "time": 0.9091894626617432, "epoch": 5, "memory": 28783, "step": 28332} +{"lr": 6.890576474687264e-05, "data_time": 0.0015626668930053711, "grad_norm": 0.6157525017857551, "loss": 0.2025834009051323, "time": 0.8718742847442627, "epoch": 5, "memory": 28783, "step": 28432} +{"lr": 6.890576474687264e-05, "data_time": 0.0019608259201049803, "grad_norm": 0.6289623349905014, "loss": 0.2043049991130829, "time": 0.8093969821929932, "epoch": 5, "memory": 28783, "step": 28532} +{"lr": 6.890576474687264e-05, "data_time": 0.0024734258651733397, "grad_norm": 0.5192620635032654, "loss": 0.20114141702651978, "time": 0.8368217945098877, "epoch": 5, "memory": 28783, "step": 28632} +{"lr": 6.890576474687264e-05, "data_time": 0.001771402359008789, "grad_norm": 0.8042371943593025, "loss": 0.20445415675640105, "time": 0.801706600189209, "epoch": 5, "memory": 28783, "step": 28732} +{"accuracy/top1": 99.79167938232422, "data_time": 0.24976594307843378, "time": 0.6703277756186092, "step": 5} +{"lr": 5.500000000000001e-05, "data_time": 0.2367393732070923, "grad_norm": 0.6578721165657043, "loss": 0.20373403280973434, "time": 0.7802339553833008, "epoch": 6, "memory": 28783, "step": 28890} +{"lr": 5.500000000000001e-05, "data_time": 0.23415262699127198, "grad_norm": 0.42493289709091187, "loss": 0.2005619525909424, "time": 0.7920568466186524, "epoch": 6, "memory": 28783, "step": 28990} +{"lr": 5.500000000000001e-05, "data_time": 0.15327730178833007, "grad_norm": 0.5292289935052394, "loss": 0.2023264840245247, "time": 0.8249850749969483, "epoch": 6, "memory": 28783, "step": 29090} +{"lr": 5.500000000000001e-05, "data_time": 0.34800536632537843, "grad_norm": 0.4441455349326134, "loss": 0.20092955380678176, "time": 0.8863354921340942, "epoch": 6, "memory": 28783, "step": 29190} +{"lr": 5.500000000000001e-05, "data_time": 0.35909485816955566, "grad_norm": 0.5437352448701859, "loss": 0.20161206722259523, "time": 0.8636725187301636, "epoch": 6, "memory": 28783, "step": 29290} +{"lr": 5.500000000000001e-05, "data_time": 0.35016751289367676, "grad_norm": 0.6724790245294571, "loss": 0.2017546609044075, "time": 0.865167212486267, "epoch": 6, "memory": 28783, "step": 29390} +{"lr": 5.500000000000001e-05, "data_time": 0.23414967060089112, "grad_norm": 0.6596096843481064, "loss": 0.20320844203233718, "time": 0.7571777105331421, "epoch": 6, "memory": 28783, "step": 29490} +{"lr": 5.500000000000001e-05, "data_time": 0.22813107967376708, "grad_norm": 0.78020164296031, "loss": 0.20676292479038239, "time": 0.7345282554626464, "epoch": 6, "memory": 28783, "step": 29590} +{"lr": 5.500000000000001e-05, "data_time": 0.19055211544036865, "grad_norm": 0.4643255159258842, "loss": 0.20193393528461456, "time": 0.7099626779556274, "epoch": 6, "memory": 28783, "step": 29690} +{"lr": 5.500000000000001e-05, "data_time": 0.32185938358306887, "grad_norm": 0.799211610853672, "loss": 0.20188792794942856, "time": 0.891472315788269, "epoch": 6, "memory": 28783, "step": 29790} +{"lr": 5.500000000000001e-05, "data_time": 0.1744075059890747, "grad_norm": 0.6371877282857895, "loss": 0.2015355110168457, "time": 0.7960968255996704, "epoch": 6, "memory": 28783, "step": 29890} +{"lr": 5.500000000000001e-05, "data_time": 0.08508853912353516, "grad_norm": 0.46341052651405334, "loss": 0.2030060276389122, "time": 0.7870536088943482, "epoch": 6, "memory": 28783, "step": 29990} +{"lr": 5.500000000000001e-05, "data_time": 0.0024773120880126954, "grad_norm": 0.5534893929958343, "loss": 0.20161805301904678, "time": 0.9774402856826783, "epoch": 6, "memory": 28783, "step": 30090} +{"lr": 5.500000000000001e-05, "data_time": 0.0022177696228027344, "grad_norm": 0.6899497509002686, "loss": 0.2014864131808281, "time": 0.767720103263855, "epoch": 6, "memory": 28783, "step": 30190} +{"lr": 5.500000000000001e-05, "data_time": 0.001803112030029297, "grad_norm": 0.716052895784378, "loss": 0.20321467220783235, "time": 0.7384618759155274, "epoch": 6, "memory": 28783, "step": 30290} +{"lr": 5.500000000000001e-05, "data_time": 0.0020949602127075194, "grad_norm": 0.5422618538141251, "loss": 0.199382983148098, "time": 0.8557018995285034, "epoch": 6, "memory": 28783, "step": 30390} +{"lr": 5.500000000000001e-05, "data_time": 0.001886296272277832, "grad_norm": 0.4186623305082321, "loss": 0.2010822594165802, "time": 0.7190708875656128, "epoch": 6, "memory": 28783, "step": 30490} +{"lr": 5.500000000000001e-05, "data_time": 0.0019061803817749024, "grad_norm": 0.5010376758873463, "loss": 0.2028880938887596, "time": 0.8347115278244018, "epoch": 6, "memory": 28783, "step": 30590} +{"lr": 5.500000000000001e-05, "data_time": 0.005642509460449219, "grad_norm": 0.5715576559305191, "loss": 0.2018888548016548, "time": 0.9515722990036011, "epoch": 6, "memory": 28783, "step": 30690} +{"lr": 5.500000000000001e-05, "data_time": 0.001876521110534668, "grad_norm": 0.43848167508840563, "loss": 0.19990312159061432, "time": 0.8387673377990723, "epoch": 6, "memory": 28783, "step": 30790} +{"lr": 5.500000000000001e-05, "data_time": 0.0026645898818969727, "grad_norm": 0.4061574816703796, "loss": 0.20202040821313857, "time": 0.8487327337265015, "epoch": 6, "memory": 28783, "step": 30890} +{"lr": 5.500000000000001e-05, "data_time": 0.0021139144897460937, "grad_norm": 0.6403361782431602, "loss": 0.20125248432159423, "time": 0.7736462593078614, "epoch": 6, "memory": 28783, "step": 30990} +{"lr": 5.500000000000001e-05, "data_time": 0.0024642467498779295, "grad_norm": 0.6221163600683213, "loss": 0.203619085252285, "time": 0.8640312194824219, "epoch": 6, "memory": 28783, "step": 31090} +{"lr": 5.500000000000001e-05, "data_time": 0.0023496866226196287, "grad_norm": 0.7111595347523689, "loss": 0.20276792645454406, "time": 0.8602776765823364, "epoch": 6, "memory": 28783, "step": 31190} +{"lr": 5.500000000000001e-05, "data_time": 0.003142380714416504, "grad_norm": 0.603440237045288, "loss": 0.20502182692289353, "time": 1.0008724212646485, "epoch": 6, "memory": 28783, "step": 31290} +{"lr": 5.500000000000001e-05, "data_time": 0.0017239809036254882, "grad_norm": 0.5461910426616668, "loss": 0.20207417011260986, "time": 0.7815804958343506, "epoch": 6, "memory": 28783, "step": 31390} +{"lr": 5.500000000000001e-05, "data_time": 0.0016937017440795898, "grad_norm": 0.483082864433527, "loss": 0.20152270644903184, "time": 0.8269901514053345, "epoch": 6, "memory": 28783, "step": 31490} +{"lr": 5.500000000000001e-05, "data_time": 0.002161860466003418, "grad_norm": 0.5504985824227333, "loss": 0.20051645040512084, "time": 0.8348847389221191, "epoch": 6, "memory": 28783, "step": 31590} +{"lr": 5.500000000000001e-05, "data_time": 0.0017740726470947266, "grad_norm": 0.6400432795286178, "loss": 0.20143334418535233, "time": 0.8210888147354126, "epoch": 6, "memory": 28783, "step": 31690} +{"lr": 5.500000000000001e-05, "data_time": 0.0016729354858398438, "grad_norm": 0.8681560799479484, "loss": 0.20269544422626495, "time": 0.755955982208252, "epoch": 6, "memory": 28783, "step": 31790} +{"lr": 5.500000000000001e-05, "data_time": 0.002167534828186035, "grad_norm": 0.37329685539007185, "loss": 0.2024100810289383, "time": 0.7681767463684082, "epoch": 6, "memory": 28783, "step": 31890} +{"lr": 5.500000000000001e-05, "data_time": 0.0019175052642822266, "grad_norm": 0.49450180530548093, "loss": 0.2039802834391594, "time": 0.8147980213165283, "epoch": 6, "memory": 28783, "step": 31990} +{"lr": 5.500000000000001e-05, "data_time": 0.0015493392944335937, "grad_norm": 0.5048115089535713, "loss": 0.20200861096382142, "time": 0.8358577251434326, "epoch": 6, "memory": 28783, "step": 32090} +{"lr": 5.500000000000001e-05, "data_time": 0.0016155004501342773, "grad_norm": 0.6258242383599282, "loss": 0.20049929022789, "time": 0.7922482967376709, "epoch": 6, "memory": 28783, "step": 32190} +{"lr": 5.500000000000001e-05, "data_time": 0.0017861604690551757, "grad_norm": 0.5814644038677216, "loss": 0.20161413997411728, "time": 0.7749020099639893, "epoch": 6, "memory": 28783, "step": 32290} +{"lr": 5.500000000000001e-05, "data_time": 0.0017628908157348634, "grad_norm": 0.604364612698555, "loss": 0.20077844858169555, "time": 0.7862651109695434, "epoch": 6, "memory": 28783, "step": 32390} +{"lr": 5.500000000000001e-05, "data_time": 0.0019086360931396484, "grad_norm": 0.7971808582544326, "loss": 0.20373074412345887, "time": 0.716651177406311, "epoch": 6, "memory": 28783, "step": 32490} +{"lr": 5.500000000000001e-05, "data_time": 0.0020807504653930662, "grad_norm": 0.4211017806082964, "loss": 0.20024603456258774, "time": 0.7763030052185058, "epoch": 6, "memory": 28783, "step": 32590} +{"lr": 5.500000000000001e-05, "data_time": 0.0017891883850097655, "grad_norm": 0.5494905471801758, "loss": 0.20403071641921997, "time": 0.8501994848251343, "epoch": 6, "memory": 28783, "step": 32690} +{"lr": 5.500000000000001e-05, "data_time": 0.0032614469528198242, "grad_norm": 0.6392549939453602, "loss": 0.20406720638275147, "time": 0.8198360919952392, "epoch": 6, "memory": 28783, "step": 32790} +{"lr": 5.500000000000001e-05, "data_time": 0.0018544197082519531, "grad_norm": 0.5553544420748949, "loss": 0.20192965269088745, "time": 0.7752921104431152, "epoch": 6, "memory": 28783, "step": 32890} +{"lr": 5.500000000000001e-05, "data_time": 0.0031661510467529295, "grad_norm": 0.3655755288898945, "loss": 0.19952612221240998, "time": 0.9092059373855591, "epoch": 6, "memory": 28783, "step": 32990} +{"lr": 5.500000000000001e-05, "data_time": 0.0016002655029296875, "grad_norm": 0.5958395235240459, "loss": 0.2008439153432846, "time": 0.7857270956039428, "epoch": 6, "memory": 28783, "step": 33090} +{"lr": 5.500000000000001e-05, "data_time": 0.0016041994094848633, "grad_norm": 0.7052890866994858, "loss": 0.2020125523209572, "time": 0.7801207304000854, "epoch": 6, "memory": 28783, "step": 33190} +{"lr": 5.500000000000001e-05, "data_time": 0.0023583412170410157, "grad_norm": 0.41311595141887664, "loss": 0.20182453691959382, "time": 0.958920931816101, "epoch": 6, "memory": 28783, "step": 33290} +{"lr": 5.500000000000001e-05, "data_time": 0.0024319887161254883, "grad_norm": 0.8197527393698693, "loss": 0.20162223875522614, "time": 0.915973424911499, "epoch": 6, "memory": 28783, "step": 33390} +{"lr": 5.500000000000001e-05, "data_time": 0.0016688823699951172, "grad_norm": 0.549808344244957, "loss": 0.20218923240900039, "time": 0.8429347515106201, "epoch": 6, "memory": 28783, "step": 33490} +{"lr": 5.500000000000001e-05, "data_time": 0.0018612861633300782, "grad_norm": 0.9302523612976075, "loss": 0.20278658717870712, "time": 0.8232696056365967, "epoch": 6, "memory": 28783, "step": 33590} +{"lr": 5.500000000000001e-05, "data_time": 0.0022600412368774412, "grad_norm": 0.5280377715826035, "loss": 0.20467469990253448, "time": 0.8906569242477417, "epoch": 6, "memory": 28783, "step": 33690} +{"lr": 5.500000000000001e-05, "data_time": 0.0016196489334106446, "grad_norm": 0.3647256713360548, "loss": 0.20158257931470872, "time": 0.7626033306121827, "epoch": 6, "memory": 28783, "step": 33790} +{"lr": 5.500000000000001e-05, "data_time": 0.001937270164489746, "grad_norm": 0.3085356891155243, "loss": 0.20190771520137787, "time": 0.8342926979064942, "epoch": 6, "memory": 28783, "step": 33890} +{"lr": 5.500000000000001e-05, "data_time": 0.0019712209701538085, "grad_norm": 0.6018663965165615, "loss": 0.20116854310035706, "time": 0.9067827701568604, "epoch": 6, "memory": 28783, "step": 33990} +{"lr": 5.500000000000001e-05, "data_time": 0.002509808540344238, "grad_norm": 0.5749215565621852, "loss": 0.2031683385372162, "time": 0.9851001739501953, "epoch": 6, "memory": 28783, "step": 34090} +{"lr": 5.500000000000001e-05, "data_time": 0.002916765213012695, "grad_norm": 0.4083619400858879, "loss": 0.20266218483448029, "time": 0.9786816835403442, "epoch": 6, "memory": 28783, "step": 34190} +{"lr": 5.500000000000001e-05, "data_time": 0.0017733335494995116, "grad_norm": 0.4704970747232437, "loss": 0.20194581300020217, "time": 0.8775602340698242, "epoch": 6, "memory": 28783, "step": 34290} +{"lr": 5.500000000000001e-05, "data_time": 0.0032355070114135744, "grad_norm": 1.0642066419124603, "loss": 0.20757760852575302, "time": 1.0334163188934327, "epoch": 6, "memory": 28783, "step": 34390} +{"lr": 5.500000000000001e-05, "data_time": 0.0015826225280761719, "grad_norm": 0.5162757977843284, "loss": 0.20109372586011887, "time": 0.7544770956039428, "epoch": 6, "memory": 28783, "step": 34490} +{"accuracy/top1": 99.54547882080078, "data_time": 0.2649371203254251, "time": 0.6807534133686739, "step": 6} +{"lr": 4.109423525312737e-05, "data_time": 0.2570410013198853, "grad_norm": 0.5304535984992981, "loss": 0.20379682183265685, "time": 0.8140489339828492, "epoch": 7, "memory": 28783, "step": 34648} +{"lr": 4.109423525312737e-05, "data_time": 0.3319602727890015, "grad_norm": 0.6574077785015107, "loss": 0.2008529469370842, "time": 0.9948984861373902, "epoch": 7, "memory": 28783, "step": 34748} +{"lr": 4.109423525312737e-05, "data_time": 0.002550840377807617, "grad_norm": 0.6974850043654441, "loss": 0.2028369501233101, "time": 0.8849415302276611, "epoch": 7, "memory": 28783, "step": 34848} +{"lr": 4.109423525312737e-05, "data_time": 0.0018515586853027344, "grad_norm": 0.731633386015892, "loss": 0.20060226917266846, "time": 0.8067174196243286, "epoch": 7, "memory": 28783, "step": 34948} +{"lr": 4.109423525312737e-05, "data_time": 0.0017177581787109375, "grad_norm": 0.39038731157779694, "loss": 0.19997669905424117, "time": 0.8141330242156982, "epoch": 7, "memory": 28783, "step": 35048} +{"lr": 4.109423525312737e-05, "data_time": 0.001930999755859375, "grad_norm": 0.7535681523382664, "loss": 0.2011270508170128, "time": 0.7879596710205078, "epoch": 7, "memory": 28783, "step": 35148} +{"lr": 4.109423525312737e-05, "data_time": 0.0022539854049682616, "grad_norm": 0.6121316776610911, "loss": 0.20207958221435546, "time": 0.9131685733795166, "epoch": 7, "memory": 28783, "step": 35248} +{"lr": 4.109423525312737e-05, "data_time": 0.003457784652709961, "grad_norm": 0.5684374149888753, "loss": 0.2024962857365608, "time": 0.8502788305282593, "epoch": 7, "memory": 28783, "step": 35348} +{"lr": 4.109423525312737e-05, "data_time": 0.0024862051010131835, "grad_norm": 0.35475060492753985, "loss": 0.20085809677839278, "time": 0.8818483352661133, "epoch": 7, "memory": 28783, "step": 35448} +{"lr": 4.109423525312737e-05, "data_time": 0.0016196966171264648, "grad_norm": 0.5782546050846576, "loss": 0.20242090672254562, "time": 0.7270224571228028, "epoch": 7, "memory": 28783, "step": 35548} +{"lr": 4.109423525312737e-05, "data_time": 0.07658209800720214, "grad_norm": 0.43485084772109983, "loss": 0.20089922547340394, "time": 0.8768445491790772, "epoch": 7, "memory": 28783, "step": 35648} +{"lr": 4.109423525312737e-05, "data_time": 0.0017717838287353515, "grad_norm": 0.3603450506925583, "loss": 0.19999355375766753, "time": 0.7718201637268066, "epoch": 7, "memory": 28783, "step": 35748} +{"lr": 4.109423525312737e-05, "data_time": 0.0015961408615112304, "grad_norm": 0.6901052325963974, "loss": 0.2033108502626419, "time": 0.8285606145858765, "epoch": 7, "memory": 28783, "step": 35848} +{"lr": 4.109423525312737e-05, "data_time": 0.0026313304901123048, "grad_norm": 0.5386508405208588, "loss": 0.20118250995874404, "time": 0.7770572662353515, "epoch": 7, "memory": 28783, "step": 35948} +{"lr": 4.109423525312737e-05, "data_time": 0.0018102884292602538, "grad_norm": 0.5510954394936561, "loss": 0.2002228692173958, "time": 0.8777055740356445, "epoch": 7, "memory": 28783, "step": 36048} +{"lr": 4.109423525312737e-05, "data_time": 0.0017973661422729492, "grad_norm": 0.44983562976121905, "loss": 0.20088137984275817, "time": 0.8439640045166016, "epoch": 7, "memory": 28783, "step": 36148} +{"lr": 4.109423525312737e-05, "data_time": 0.0027127742767333986, "grad_norm": 0.37173020765185355, "loss": 0.19929925799369813, "time": 0.8717893600463867, "epoch": 7, "memory": 28783, "step": 36248} +{"lr": 4.109423525312737e-05, "data_time": 0.001730060577392578, "grad_norm": 0.5331165784969926, "loss": 0.201166507601738, "time": 0.7625855207443237, "epoch": 7, "memory": 28783, "step": 36348} +{"lr": 4.109423525312737e-05, "data_time": 0.002619028091430664, "grad_norm": 0.2659651841968298, "loss": 0.1999305322766304, "time": 0.9332836389541626, "epoch": 7, "memory": 28783, "step": 36448} +{"lr": 4.109423525312737e-05, "data_time": 0.001663970947265625, "grad_norm": 0.38814663216471673, "loss": 0.20105892717838286, "time": 0.8786244630813599, "epoch": 7, "memory": 28783, "step": 36548} +{"lr": 4.109423525312737e-05, "data_time": 0.001772594451904297, "grad_norm": 0.48508835919201376, "loss": 0.19943130314350127, "time": 0.853243613243103, "epoch": 7, "memory": 28783, "step": 36648} +{"lr": 4.109423525312737e-05, "data_time": 0.002699422836303711, "grad_norm": 0.8456787571310997, "loss": 0.20061280727386474, "time": 0.9556742668151855, "epoch": 7, "memory": 28783, "step": 36748} +{"lr": 4.109423525312737e-05, "data_time": 0.002043747901916504, "grad_norm": 0.5865349501371384, "loss": 0.20019570738077164, "time": 0.7913581371307373, "epoch": 7, "memory": 28783, "step": 36848} +{"lr": 4.109423525312737e-05, "data_time": 0.001745319366455078, "grad_norm": 0.4370046705007553, "loss": 0.20116384476423263, "time": 0.7525956630706787, "epoch": 7, "memory": 28783, "step": 36948} +{"lr": 4.109423525312737e-05, "data_time": 0.0035280704498291014, "grad_norm": 0.4532147437334061, "loss": 0.1995263069868088, "time": 0.9600348472595215, "epoch": 7, "memory": 28783, "step": 37048} +{"lr": 4.109423525312737e-05, "data_time": 0.002026486396789551, "grad_norm": 0.5822193264961243, "loss": 0.1996021643280983, "time": 0.8014884233474732, "epoch": 7, "memory": 28783, "step": 37148} +{"lr": 4.109423525312737e-05, "data_time": 0.0026641845703125, "grad_norm": 0.32727089077234267, "loss": 0.20017112493515016, "time": 0.8950970649719239, "epoch": 7, "memory": 28783, "step": 37248} +{"lr": 4.109423525312737e-05, "data_time": 0.002028346061706543, "grad_norm": 0.37457580752670766, "loss": 0.2001060962677002, "time": 0.7436938524246216, "epoch": 7, "memory": 28783, "step": 37348} +{"lr": 4.109423525312737e-05, "data_time": 0.0031909704208374023, "grad_norm": 0.476037173718214, "loss": 0.20002157092094422, "time": 0.9007952213287354, "epoch": 7, "memory": 28783, "step": 37448} +{"lr": 4.109423525312737e-05, "data_time": 0.0017638444900512696, "grad_norm": 0.559312067180872, "loss": 0.19941863864660264, "time": 0.7836038827896118, "epoch": 7, "memory": 28783, "step": 37548} +{"lr": 4.109423525312737e-05, "data_time": 0.0027809858322143553, "grad_norm": 0.43681189604103565, "loss": 0.20012189894914628, "time": 0.898586630821228, "epoch": 7, "memory": 28783, "step": 37648} +{"lr": 4.109423525312737e-05, "data_time": 0.0019791841506958006, "grad_norm": 0.39596344493329527, "loss": 0.19973721355199814, "time": 0.8397162437438965, "epoch": 7, "memory": 28783, "step": 37748} +{"lr": 4.109423525312737e-05, "data_time": 0.0030755519866943358, "grad_norm": 0.4780492804944515, "loss": 0.20098764449357986, "time": 0.8755083084106445, "epoch": 7, "memory": 28783, "step": 37848} +{"lr": 4.109423525312737e-05, "data_time": 0.0029319047927856444, "grad_norm": 0.49840219151228665, "loss": 0.19930738508701323, "time": 0.9035987377166748, "epoch": 7, "memory": 28783, "step": 37948} +{"lr": 4.109423525312737e-05, "data_time": 0.0016916990280151367, "grad_norm": 0.5670174971222878, "loss": 0.20086415261030197, "time": 0.7433984518051148, "epoch": 7, "memory": 28783, "step": 38048} +{"lr": 4.109423525312737e-05, "data_time": 0.00224001407623291, "grad_norm": 0.4579376712441444, "loss": 0.20129559189081192, "time": 0.859979772567749, "epoch": 7, "memory": 28783, "step": 38148} +{"lr": 4.109423525312737e-05, "data_time": 0.0016407489776611329, "grad_norm": 0.48678094148635864, "loss": 0.19966082721948625, "time": 0.7750612735748291, "epoch": 7, "memory": 28783, "step": 38248} +{"lr": 4.109423525312737e-05, "data_time": 0.004172062873840332, "grad_norm": 0.46413626074790953, "loss": 0.2013307586312294, "time": 0.9615135669708252, "epoch": 7, "memory": 28783, "step": 38348} +{"lr": 4.109423525312737e-05, "data_time": 0.0021019220352172852, "grad_norm": 0.376536026597023, "loss": 0.19990138560533524, "time": 0.9268024682998657, "epoch": 7, "memory": 28783, "step": 38448} +{"lr": 4.109423525312737e-05, "data_time": 0.0019810676574707033, "grad_norm": 0.4711449682712555, "loss": 0.19912588149309157, "time": 0.798414945602417, "epoch": 7, "memory": 28783, "step": 38548} +{"lr": 4.109423525312737e-05, "data_time": 0.0017978429794311523, "grad_norm": 0.37200071290135384, "loss": 0.1994968444108963, "time": 0.7944779396057129, "epoch": 7, "memory": 28783, "step": 38648} +{"lr": 4.109423525312737e-05, "data_time": 0.0017348527908325195, "grad_norm": 0.6079226233065128, "loss": 0.20247834026813508, "time": 0.8917044878005982, "epoch": 7, "memory": 28783, "step": 38748} +{"lr": 4.109423525312737e-05, "data_time": 0.003507566452026367, "grad_norm": 0.4597327888011932, "loss": 0.2032423883676529, "time": 0.9037577629089355, "epoch": 7, "memory": 28783, "step": 38848} +{"lr": 4.109423525312737e-05, "data_time": 0.0018663883209228515, "grad_norm": 0.614964597299695, "loss": 0.2004171773791313, "time": 0.8065264225006104, "epoch": 7, "memory": 28783, "step": 38948} +{"lr": 4.109423525312737e-05, "data_time": 0.0018650054931640624, "grad_norm": 0.2859881963580847, "loss": 0.20019351243972777, "time": 0.9513052701950073, "epoch": 7, "memory": 28783, "step": 39048} +{"lr": 4.109423525312737e-05, "data_time": 0.0016150236129760741, "grad_norm": 0.4854913234710693, "loss": 0.19948391169309615, "time": 0.8299093008041382, "epoch": 7, "memory": 28783, "step": 39148} +{"lr": 4.109423525312737e-05, "data_time": 0.001726984977722168, "grad_norm": 0.3842244539409876, "loss": 0.20226747393608094, "time": 0.771368432044983, "epoch": 7, "memory": 28783, "step": 39248} +{"lr": 4.109423525312737e-05, "data_time": 0.0016712427139282226, "grad_norm": 0.3702883360907435, "loss": 0.2003416895866394, "time": 0.8271692037582398, "epoch": 7, "memory": 28783, "step": 39348} +{"lr": 4.109423525312737e-05, "data_time": 0.0017933130264282226, "grad_norm": 0.43516001403331755, "loss": 0.20015901774168016, "time": 0.7821183204650879, "epoch": 7, "memory": 28783, "step": 39448} +{"lr": 4.109423525312737e-05, "data_time": 0.0018541574478149413, "grad_norm": 0.3398057587444782, "loss": 0.20029745399951934, "time": 0.8236588478088379, "epoch": 7, "memory": 28783, "step": 39548} +{"lr": 4.109423525312737e-05, "data_time": 0.001649951934814453, "grad_norm": 0.4105547945946455, "loss": 0.19939403980970383, "time": 0.8436854124069214, "epoch": 7, "memory": 28783, "step": 39648} +{"lr": 4.109423525312737e-05, "data_time": 0.0016717195510864257, "grad_norm": 0.43874640837311746, "loss": 0.20365124493837355, "time": 0.7962920904159546, "epoch": 7, "memory": 28783, "step": 39748} +{"lr": 4.109423525312737e-05, "data_time": 0.0017191410064697266, "grad_norm": 0.49381513670086863, "loss": 0.1998294323682785, "time": 0.7393632173538208, "epoch": 7, "memory": 28783, "step": 39848} +{"lr": 4.109423525312737e-05, "data_time": 0.0015441179275512695, "grad_norm": 0.4961566381156445, "loss": 0.19969123601913452, "time": 0.7431836128234863, "epoch": 7, "memory": 28783, "step": 39948} +{"lr": 4.109423525312737e-05, "data_time": 0.08929152488708496, "grad_norm": 0.3904197491705418, "loss": 0.2005852296948433, "time": 0.9282039642333985, "epoch": 7, "memory": 28783, "step": 40048} +{"lr": 4.109423525312737e-05, "data_time": 0.0017680644989013672, "grad_norm": 0.9225305676460266, "loss": 0.2009689301252365, "time": 0.7587546825408935, "epoch": 7, "memory": 28783, "step": 40148} +{"lr": 4.109423525312737e-05, "data_time": 0.004726099967956543, "grad_norm": 0.534372840076685, "loss": 0.19917624592781066, "time": 0.8363152027130127, "epoch": 7, "memory": 28783, "step": 40248} +{"accuracy/top1": 99.955810546875, "data_time": 0.32658713004168344, "time": 0.7414570555967444, "step": 7} +{"lr": 2.8549663646838717e-05, "data_time": 0.3024162292480469, "grad_norm": 0.34962107576429846, "loss": 0.2010873794555664, "time": 0.8124499082565307, "epoch": 8, "memory": 28783, "step": 40406} +{"lr": 2.8549663646838717e-05, "data_time": 0.22667150497436522, "grad_norm": 0.41273797005414964, "loss": 0.20046880096197128, "time": 0.8970856666564941, "epoch": 8, "memory": 28783, "step": 40506} +{"lr": 2.8549663646838717e-05, "data_time": 0.2134321451187134, "grad_norm": 0.561300278455019, "loss": 0.20044224411249162, "time": 0.7762515068054199, "epoch": 8, "memory": 28783, "step": 40606} +{"lr": 2.8549663646838717e-05, "data_time": 0.16150379180908203, "grad_norm": 0.38815378621220586, "loss": 0.20110579282045365, "time": 0.7558535814285279, "epoch": 8, "memory": 28783, "step": 40706} +{"lr": 2.8549663646838717e-05, "data_time": 0.16156353950500488, "grad_norm": 0.4839564699679613, "loss": 0.20004840642213823, "time": 0.7756923913955689, "epoch": 8, "memory": 28783, "step": 40806} +{"lr": 2.8549663646838717e-05, "data_time": 0.2550986289978027, "grad_norm": 0.43951306864619255, "loss": 0.19977218955755233, "time": 0.8354975938796997, "epoch": 8, "memory": 28783, "step": 40906} +{"lr": 2.8549663646838717e-05, "data_time": 0.24363205432891846, "grad_norm": 0.45079954117536547, "loss": 0.19976240992546082, "time": 0.769334363937378, "epoch": 8, "memory": 28783, "step": 41006} +{"lr": 2.8549663646838717e-05, "data_time": 0.2094966411590576, "grad_norm": 0.27666075006127355, "loss": 0.20195276737213136, "time": 0.9607676029205322, "epoch": 8, "memory": 28783, "step": 41106} +{"lr": 2.8549663646838717e-05, "data_time": 0.1692953586578369, "grad_norm": 0.37942475695163014, "loss": 0.19994042366743087, "time": 0.9432299613952637, "epoch": 8, "memory": 28783, "step": 41206} +{"lr": 2.8549663646838717e-05, "data_time": 0.002052116394042969, "grad_norm": 0.3468345979228616, "loss": 0.20011111795902253, "time": 0.8091646671295166, "epoch": 8, "memory": 28783, "step": 41306} +{"lr": 2.8549663646838717e-05, "data_time": 0.0016150474548339844, "grad_norm": 0.5458853382617235, "loss": 0.1991449177265167, "time": 0.7599417209625244, "epoch": 8, "memory": 28783, "step": 41406} +{"lr": 2.8549663646838717e-05, "data_time": 0.0015526294708251953, "grad_norm": 0.8896437801420689, "loss": 0.20091405212879182, "time": 0.8348368167877197, "epoch": 8, "memory": 28783, "step": 41506} +{"lr": 2.8549663646838717e-05, "data_time": 0.0023251771926879883, "grad_norm": 0.48138552606105806, "loss": 0.2000388041138649, "time": 0.8940004348754883, "epoch": 8, "memory": 28783, "step": 41606} +{"lr": 2.8549663646838717e-05, "data_time": 0.0019300460815429687, "grad_norm": 0.3746541578322649, "loss": 0.200286203622818, "time": 0.8362886190414429, "epoch": 8, "memory": 28783, "step": 41706} +{"lr": 2.8549663646838717e-05, "data_time": 0.0020024776458740234, "grad_norm": 0.29308299366384744, "loss": 0.19902280122041702, "time": 0.8487228870391845, "epoch": 8, "memory": 28783, "step": 41806} +{"lr": 2.8549663646838717e-05, "data_time": 0.005043601989746094, "grad_norm": 0.337833072245121, "loss": 0.19973892867565154, "time": 0.9286252498626709, "epoch": 8, "memory": 28783, "step": 41906} +{"lr": 2.8549663646838717e-05, "data_time": 0.03813190460205078, "grad_norm": 0.33194583766162394, "loss": 0.19875675886869432, "time": 0.8040109157562256, "epoch": 8, "memory": 28783, "step": 42006} +{"lr": 2.8549663646838717e-05, "data_time": 0.1507725715637207, "grad_norm": 0.5127151221036911, "loss": 0.1998363807797432, "time": 0.8586250305175781, "epoch": 8, "memory": 28783, "step": 42106} +{"lr": 2.8549663646838717e-05, "data_time": 0.04830701351165771, "grad_norm": 0.19206122159957886, "loss": 0.19865394681692122, "time": 0.8016723155975342, "epoch": 8, "memory": 28783, "step": 42206} +{"lr": 2.8549663646838717e-05, "data_time": 0.20780677795410157, "grad_norm": 0.6225557684898376, "loss": 0.2015370950102806, "time": 0.873365068435669, "epoch": 8, "memory": 28783, "step": 42306} +{"lr": 2.8549663646838717e-05, "data_time": 0.0017276525497436524, "grad_norm": 0.4400020346045494, "loss": 0.20197580009698868, "time": 0.7947998762130737, "epoch": 8, "memory": 28783, "step": 42406} +{"lr": 2.8549663646838717e-05, "data_time": 0.0344501256942749, "grad_norm": 0.3868662118911743, "loss": 0.19997474551200867, "time": 0.7513847589492798, "epoch": 8, "memory": 28783, "step": 42506} +{"lr": 2.8549663646838717e-05, "data_time": 0.10830955505371094, "grad_norm": 0.2928969316184521, "loss": 0.1989822193980217, "time": 0.8792859792709351, "epoch": 8, "memory": 28783, "step": 42606} +{"lr": 2.8549663646838717e-05, "data_time": 0.13611235618591308, "grad_norm": 0.28904529884457586, "loss": 0.199774332344532, "time": 0.7907635927200317, "epoch": 8, "memory": 28783, "step": 42706} +{"lr": 2.8549663646838717e-05, "data_time": 0.23929970264434813, "grad_norm": 0.3635988987982273, "loss": 0.1997097462415695, "time": 0.7439430952072144, "epoch": 8, "memory": 28783, "step": 42806} +{"lr": 2.8549663646838717e-05, "data_time": 0.17196111679077147, "grad_norm": 0.276888757199049, "loss": 0.198770934343338, "time": 0.9269291877746582, "epoch": 8, "memory": 28783, "step": 42906} +{"lr": 2.8549663646838717e-05, "data_time": 0.061671996116638185, "grad_norm": 0.7010551903396844, "loss": 0.20110541582107544, "time": 0.7903759241104126, "epoch": 8, "memory": 28783, "step": 43006} +{"lr": 2.8549663646838717e-05, "data_time": 0.07312889099121093, "grad_norm": 0.40671457201242445, "loss": 0.20088512301445008, "time": 0.8595164299011231, "epoch": 8, "memory": 28783, "step": 43106} +{"lr": 2.8549663646838717e-05, "data_time": 0.2648433446884155, "grad_norm": 0.39031699299812317, "loss": 0.20173955261707305, "time": 0.7758603811264038, "epoch": 8, "memory": 28783, "step": 43206} +{"lr": 2.8549663646838717e-05, "data_time": 0.025339365005493164, "grad_norm": 0.4093537891283631, "loss": 0.1994659811258316, "time": 0.8794605255126953, "epoch": 8, "memory": 28783, "step": 43306} +{"lr": 2.8549663646838717e-05, "data_time": 0.0016026735305786134, "grad_norm": 0.5903012270107866, "loss": 0.20138724148273468, "time": 0.7433746814727783, "epoch": 8, "memory": 28783, "step": 43406} +{"lr": 2.8549663646838717e-05, "data_time": 0.0021006107330322266, "grad_norm": 0.3646797526627779, "loss": 0.19887280464172363, "time": 0.8078948497772217, "epoch": 8, "memory": 28783, "step": 43506} +{"lr": 2.8549663646838717e-05, "data_time": 0.0032089948654174805, "grad_norm": 0.42064387165009975, "loss": 0.20039766728878022, "time": 0.9307423114776612, "epoch": 8, "memory": 28783, "step": 43606} +{"lr": 2.8549663646838717e-05, "data_time": 0.0017543077468872071, "grad_norm": 0.18693483658134938, "loss": 0.19990795403718947, "time": 0.9137980222702027, "epoch": 8, "memory": 28783, "step": 43706} +{"lr": 2.8549663646838717e-05, "data_time": 0.0015865325927734374, "grad_norm": 0.4696742117404938, "loss": 0.19978507459163666, "time": 0.8123665571212768, "epoch": 8, "memory": 28783, "step": 43806} +{"lr": 2.8549663646838717e-05, "data_time": 0.0023093938827514647, "grad_norm": 0.34556571766734123, "loss": 0.20007470697164537, "time": 0.8867485761642456, "epoch": 8, "memory": 28783, "step": 43906} +{"lr": 2.8549663646838717e-05, "data_time": 0.001970839500427246, "grad_norm": 0.32554522417485715, "loss": 0.20074764490127564, "time": 0.77413969039917, "epoch": 8, "memory": 28783, "step": 44006} +{"lr": 2.8549663646838717e-05, "data_time": 0.001967763900756836, "grad_norm": 0.342456965893507, "loss": 0.19901065081357955, "time": 0.8418396949768067, "epoch": 8, "memory": 28783, "step": 44106} +{"lr": 2.8549663646838717e-05, "data_time": 0.002103900909423828, "grad_norm": 0.33164366427809, "loss": 0.1990720212459564, "time": 0.7931753873825074, "epoch": 8, "memory": 28783, "step": 44206} +{"lr": 2.8549663646838717e-05, "data_time": 0.0017536640167236327, "grad_norm": 0.42727058231830595, "loss": 0.2015230730175972, "time": 0.7858794927597046, "epoch": 8, "memory": 28783, "step": 44306} +{"lr": 2.8549663646838717e-05, "data_time": 0.001806187629699707, "grad_norm": 0.3826434537768364, "loss": 0.20064708441495896, "time": 0.8281020164489746, "epoch": 8, "memory": 28783, "step": 44406} +{"lr": 2.8549663646838717e-05, "data_time": 0.0016503095626831054, "grad_norm": 0.5383035026490688, "loss": 0.20213476270437242, "time": 0.8384078741073608, "epoch": 8, "memory": 28783, "step": 44506} +{"lr": 2.8549663646838717e-05, "data_time": 0.0019080162048339844, "grad_norm": 0.4320871841162443, "loss": 0.2004919692873955, "time": 0.8009705543518066, "epoch": 8, "memory": 28783, "step": 44606} +{"lr": 2.8549663646838717e-05, "data_time": 0.0018613576889038087, "grad_norm": 0.3800024766474962, "loss": 0.2002116397023201, "time": 0.8616921901702881, "epoch": 8, "memory": 28783, "step": 44706} +{"lr": 2.8549663646838717e-05, "data_time": 0.0020083189010620117, "grad_norm": 0.6177150852978229, "loss": 0.19994091540575026, "time": 0.835884690284729, "epoch": 8, "memory": 28783, "step": 44806} +{"lr": 2.8549663646838717e-05, "data_time": 0.0017868280410766602, "grad_norm": 0.27003599554300306, "loss": 0.19936210960149764, "time": 0.8957467555999756, "epoch": 8, "memory": 28783, "step": 44906} +{"lr": 2.8549663646838717e-05, "data_time": 0.0021057605743408205, "grad_norm": 0.47350957840681074, "loss": 0.2004600703716278, "time": 0.9268666744232178, "epoch": 8, "memory": 28783, "step": 45006} +{"lr": 2.8549663646838717e-05, "data_time": 0.0016373395919799805, "grad_norm": 0.3515616409480572, "loss": 0.19945160746574403, "time": 0.8554864645004272, "epoch": 8, "memory": 28783, "step": 45106} +{"lr": 2.8549663646838717e-05, "data_time": 0.0017644643783569336, "grad_norm": 0.31704646423459054, "loss": 0.19900186806917192, "time": 0.7706925868988037, "epoch": 8, "memory": 28783, "step": 45206} +{"lr": 2.8549663646838717e-05, "data_time": 0.004744291305541992, "grad_norm": 0.5062963657081128, "loss": 0.1999295622110367, "time": 0.9430924177169799, "epoch": 8, "memory": 28783, "step": 45306} +{"lr": 2.8549663646838717e-05, "data_time": 0.00233762264251709, "grad_norm": 0.15249796607531607, "loss": 0.19973793625831604, "time": 0.8423353433609009, "epoch": 8, "memory": 28783, "step": 45406} +{"lr": 2.8549663646838717e-05, "data_time": 0.0018955469131469727, "grad_norm": 0.46713529601693154, "loss": 0.1992884933948517, "time": 0.8008431911468505, "epoch": 8, "memory": 28783, "step": 45506} +{"lr": 2.8549663646838717e-05, "data_time": 0.005891561508178711, "grad_norm": 0.3140792388468981, "loss": 0.19922010749578475, "time": 0.9224577665328979, "epoch": 8, "memory": 28783, "step": 45606} +{"lr": 2.8549663646838717e-05, "data_time": 0.0022818326950073244, "grad_norm": 0.9852781891822815, "loss": 0.20175724625587463, "time": 0.8994702816009521, "epoch": 8, "memory": 28783, "step": 45706} +{"lr": 2.8549663646838717e-05, "data_time": 0.0019786596298217774, "grad_norm": 0.36189441829919816, "loss": 0.20120375454425812, "time": 0.8120423555374146, "epoch": 8, "memory": 28783, "step": 45806} +{"lr": 2.8549663646838717e-05, "data_time": 0.002092885971069336, "grad_norm": 0.2625554073601961, "loss": 0.19908541440963745, "time": 0.8677398204803467, "epoch": 8, "memory": 28783, "step": 45906} +{"lr": 2.8549663646838717e-05, "data_time": 0.0027587890625, "grad_norm": 0.4917635917663574, "loss": 0.19970357567071914, "time": 0.8837395429611206, "epoch": 8, "memory": 28783, "step": 46006} +{"accuracy/top1": 99.92424774169922, "data_time": 0.20535842110128963, "time": 0.6206446254954618, "step": 8} +{"lr": 1.8594235253127375e-05, "data_time": 0.31137473583221437, "grad_norm": 0.5146086907014251, "loss": 0.20010972768068314, "time": 0.822504711151123, "epoch": 9, "memory": 28783, "step": 46164} +{"lr": 1.8594235253127375e-05, "data_time": 0.21240315437316895, "grad_norm": 0.30397630073130133, "loss": 0.20071698427200318, "time": 0.8247225999832153, "epoch": 9, "memory": 28783, "step": 46264} +{"lr": 1.8594235253127375e-05, "data_time": 0.2732653856277466, "grad_norm": 0.43730893582105634, "loss": 0.19915433824062348, "time": 0.7790990591049194, "epoch": 9, "memory": 28783, "step": 46364} +{"lr": 1.8594235253127375e-05, "data_time": 0.32458369731903075, "grad_norm": 0.25308656990528106, "loss": 0.19879508912563323, "time": 0.8539206504821777, "epoch": 9, "memory": 28783, "step": 46464} +{"lr": 1.8594235253127375e-05, "data_time": 0.3200049877166748, "grad_norm": 0.23486879272386432, "loss": 0.19984957277774812, "time": 0.8361745119094849, "epoch": 9, "memory": 28783, "step": 46564} +{"lr": 1.8594235253127375e-05, "data_time": 0.35312695503234864, "grad_norm": 0.23702208884060383, "loss": 0.20000497996807098, "time": 0.8733285903930664, "epoch": 9, "memory": 28783, "step": 46664} +{"lr": 1.8594235253127375e-05, "data_time": 0.23268754482269288, "grad_norm": 0.3531709173694253, "loss": 0.1987038642168045, "time": 0.8479858636856079, "epoch": 9, "memory": 28783, "step": 46764} +{"lr": 1.8594235253127375e-05, "data_time": 0.2803735971450806, "grad_norm": 0.31653384109959004, "loss": 0.19929299056529998, "time": 0.794532322883606, "epoch": 9, "memory": 28783, "step": 46864} +{"lr": 1.8594235253127375e-05, "data_time": 0.25891790390014646, "grad_norm": 0.24372444190084935, "loss": 0.19953625351190568, "time": 0.9263168573379517, "epoch": 9, "memory": 28783, "step": 46964} +{"lr": 1.8594235253127375e-05, "data_time": 0.0035434246063232424, "grad_norm": 0.181035035289824, "loss": 0.19964278638362884, "time": 0.94240300655365, "epoch": 9, "memory": 28783, "step": 47064} +{"lr": 1.8594235253127375e-05, "data_time": 0.0020153999328613283, "grad_norm": 0.35834435671567916, "loss": 0.199106827378273, "time": 0.8317167282104492, "epoch": 9, "memory": 28783, "step": 47164} +{"lr": 1.8594235253127375e-05, "data_time": 0.0061484575271606445, "grad_norm": 0.26160954162478445, "loss": 0.19912890940904618, "time": 0.8386593818664551, "epoch": 9, "memory": 28783, "step": 47264} +{"lr": 1.8594235253127375e-05, "data_time": 0.0019340276718139648, "grad_norm": 0.27888133358210326, "loss": 0.20130230635404586, "time": 0.797444200515747, "epoch": 9, "memory": 28783, "step": 47364} +{"lr": 1.8594235253127375e-05, "data_time": 0.0020103693008422852, "grad_norm": 0.2620167948305607, "loss": 0.19866974800825118, "time": 0.8212775945663452, "epoch": 9, "memory": 28783, "step": 47464} +{"lr": 1.8594235253127375e-05, "data_time": 0.0017138957977294923, "grad_norm": 0.38680356815457345, "loss": 0.19949098229408263, "time": 0.7680622816085816, "epoch": 9, "memory": 28783, "step": 47564} +{"lr": 1.8594235253127375e-05, "data_time": 0.12626335620880128, "grad_norm": 0.4178945438936353, "loss": 0.19984035342931747, "time": 0.9743272542953492, "epoch": 9, "memory": 28783, "step": 47664} +{"lr": 1.8594235253127375e-05, "data_time": 0.0022294044494628905, "grad_norm": 0.38078233897686004, "loss": 0.19955205768346787, "time": 0.8499480009078979, "epoch": 9, "memory": 28783, "step": 47764} +{"lr": 1.8594235253127375e-05, "data_time": 0.002201485633850098, "grad_norm": 0.29727129861712454, "loss": 0.19900396168231965, "time": 0.819485855102539, "epoch": 9, "memory": 28783, "step": 47864} +{"lr": 1.8594235253127375e-05, "data_time": 0.0053002357482910155, "grad_norm": 0.3172834202647209, "loss": 0.19905504882335662, "time": 0.7542827606201172, "epoch": 9, "memory": 28783, "step": 47964} +{"lr": 1.8594235253127375e-05, "data_time": 0.004354524612426758, "grad_norm": 0.2702242273837328, "loss": 0.19889454692602157, "time": 0.8291808843612671, "epoch": 9, "memory": 28783, "step": 48064} +{"lr": 1.8594235253127375e-05, "data_time": 0.001863551139831543, "grad_norm": 0.4502132132649422, "loss": 0.19917707443237304, "time": 0.8059434413909912, "epoch": 9, "memory": 28783, "step": 48164} +{"lr": 1.8594235253127375e-05, "data_time": 0.002024984359741211, "grad_norm": 0.36323790764436126, "loss": 0.19987213015556335, "time": 0.7517117023468017, "epoch": 9, "memory": 28783, "step": 48264} +{"lr": 1.8594235253127375e-05, "data_time": 0.0018815994262695312, "grad_norm": 0.20545513220131398, "loss": 0.19894106984138488, "time": 0.8639640808105469, "epoch": 9, "memory": 28783, "step": 48364} +{"lr": 1.8594235253127375e-05, "data_time": 0.002356886863708496, "grad_norm": 0.26018533618189393, "loss": 0.19868551343679428, "time": 0.7843921184539795, "epoch": 9, "memory": 28783, "step": 48464} +{"lr": 1.8594235253127375e-05, "data_time": 0.002684283256530762, "grad_norm": 0.31424256768077613, "loss": 0.19942981451749803, "time": 0.911528468132019, "epoch": 9, "memory": 28783, "step": 48564} +{"lr": 1.8594235253127375e-05, "data_time": 0.0021970272064208984, "grad_norm": 0.24057991933077574, "loss": 0.19856883138418197, "time": 0.8527655601501465, "epoch": 9, "memory": 28783, "step": 48664} +{"lr": 1.8594235253127375e-05, "data_time": 0.0019884824752807615, "grad_norm": 0.41368495635688307, "loss": 0.20269487351179122, "time": 0.8380220651626586, "epoch": 9, "memory": 28783, "step": 48764} +{"lr": 1.8594235253127375e-05, "data_time": 0.0016928672790527343, "grad_norm": 0.5483454247005284, "loss": 0.19914476871490477, "time": 0.7924807786941528, "epoch": 9, "memory": 28783, "step": 48864} +{"lr": 1.8594235253127375e-05, "data_time": 0.002683591842651367, "grad_norm": 0.280898093059659, "loss": 0.19919253140687943, "time": 0.8820698022842407, "epoch": 9, "memory": 28783, "step": 48964} +{"lr": 1.8594235253127375e-05, "data_time": 0.0029338836669921876, "grad_norm": 0.3208234779536724, "loss": 0.19925323724746705, "time": 0.9235302686691285, "epoch": 9, "memory": 28783, "step": 49064} +{"lr": 1.8594235253127375e-05, "data_time": 0.002023649215698242, "grad_norm": 0.3346285748295486, "loss": 0.19955916851758956, "time": 0.8595220804214477, "epoch": 9, "memory": 28783, "step": 49164} +{"lr": 1.8594235253127375e-05, "data_time": 0.0018409490585327148, "grad_norm": 0.18432259727269412, "loss": 0.19887392073869706, "time": 0.7800418376922608, "epoch": 9, "memory": 28783, "step": 49264} +{"lr": 1.8594235253127375e-05, "data_time": 0.0033204078674316405, "grad_norm": 0.2936939208768308, "loss": 0.1985804796218872, "time": 0.9233827829360962, "epoch": 9, "memory": 28783, "step": 49364} +{"lr": 1.8594235253127375e-05, "data_time": 0.0024909734725952148, "grad_norm": 0.34559784568846225, "loss": 0.19987995326519012, "time": 0.9003695249557495, "epoch": 9, "memory": 28783, "step": 49464} +{"lr": 1.8594235253127375e-05, "data_time": 0.0019880056381225584, "grad_norm": 0.24455010928213597, "loss": 0.20014021694660186, "time": 0.7804906845092774, "epoch": 9, "memory": 28783, "step": 49564} +{"lr": 1.8594235253127375e-05, "data_time": 0.001706385612487793, "grad_norm": 0.42133839689195157, "loss": 0.20006576925516129, "time": 0.783941125869751, "epoch": 9, "memory": 28783, "step": 49664} +{"lr": 1.8594235253127375e-05, "data_time": 0.0025417327880859373, "grad_norm": 0.3483779177069664, "loss": 0.20002958178520203, "time": 0.7987715244293213, "epoch": 9, "memory": 28783, "step": 49764} +{"lr": 1.8594235253127375e-05, "data_time": 0.0023015499114990234, "grad_norm": 0.37101004449650643, "loss": 0.1986461713910103, "time": 0.8217520952224732, "epoch": 9, "memory": 28783, "step": 49864} +{"lr": 1.8594235253127375e-05, "data_time": 0.0017209768295288086, "grad_norm": 0.18766246289014815, "loss": 0.19873131364583968, "time": 0.8231493473052979, "epoch": 9, "memory": 28783, "step": 49964} +{"lr": 1.8594235253127375e-05, "data_time": 0.002186441421508789, "grad_norm": 0.4158086239360273, "loss": 0.20050179213285446, "time": 0.7971986055374145, "epoch": 9, "memory": 28783, "step": 50064} +{"lr": 1.8594235253127375e-05, "data_time": 0.0021574020385742186, "grad_norm": 0.18429773338139058, "loss": 0.19953390210866928, "time": 0.8015501022338867, "epoch": 9, "memory": 28783, "step": 50164} +{"lr": 1.8594235253127375e-05, "data_time": 0.002024388313293457, "grad_norm": 0.28110020104795697, "loss": 0.1989580750465393, "time": 0.7558328151702881, "epoch": 9, "memory": 28783, "step": 50264} +{"lr": 1.8594235253127375e-05, "data_time": 0.0017935276031494141, "grad_norm": 0.3857104729861021, "loss": 0.1988803744316101, "time": 0.8563242673873901, "epoch": 9, "memory": 28783, "step": 50364} +{"lr": 1.8594235253127375e-05, "data_time": 0.0021255970001220702, "grad_norm": 0.2891829600557685, "loss": 0.2000194400548935, "time": 0.8917397022247314, "epoch": 9, "memory": 28783, "step": 50464} +{"lr": 1.8594235253127375e-05, "data_time": 0.0022556543350219726, "grad_norm": 0.17433524392545224, "loss": 0.1986156314611435, "time": 0.9340741634368896, "epoch": 9, "memory": 28783, "step": 50564} +{"lr": 1.8594235253127375e-05, "data_time": 0.0020201683044433595, "grad_norm": 0.3208316644653678, "loss": 0.20012563169002534, "time": 0.8188866376876831, "epoch": 9, "memory": 28783, "step": 50664} +{"lr": 1.8594235253127375e-05, "data_time": 0.0015877962112426757, "grad_norm": 0.3780954498797655, "loss": 0.19963158220052718, "time": 0.7659719705581665, "epoch": 9, "memory": 28783, "step": 50764} +{"lr": 1.8594235253127375e-05, "data_time": 0.0017437934875488281, "grad_norm": 0.32902325987815856, "loss": 0.19996160864830018, "time": 0.813164496421814, "epoch": 9, "memory": 28783, "step": 50864} +{"lr": 1.8594235253127375e-05, "data_time": 0.0017119646072387695, "grad_norm": 0.1926439231261611, "loss": 0.19912930130958556, "time": 0.8445338249206543, "epoch": 9, "memory": 28783, "step": 50964} +{"lr": 1.8594235253127375e-05, "data_time": 0.001751255989074707, "grad_norm": 0.2375560365850106, "loss": 0.1990738958120346, "time": 0.8966788530349732, "epoch": 9, "memory": 28783, "step": 51064} +{"lr": 1.8594235253127375e-05, "data_time": 0.0017923593521118164, "grad_norm": 0.4424662021920085, "loss": 0.19921157658100128, "time": 0.8450961351394654, "epoch": 9, "memory": 28783, "step": 51164} +{"lr": 1.8594235253127375e-05, "data_time": 0.0016463994979858398, "grad_norm": 0.37272116243839265, "loss": 0.19975984692573548, "time": 0.8395662069320678, "epoch": 9, "memory": 28783, "step": 51264} +{"lr": 1.8594235253127375e-05, "data_time": 0.0029436588287353516, "grad_norm": 0.2957117594778538, "loss": 0.19911667406558992, "time": 0.8630541086196899, "epoch": 9, "memory": 28783, "step": 51364} +{"lr": 1.8594235253127375e-05, "data_time": 0.0017013311386108398, "grad_norm": 0.17027930729091167, "loss": 0.19863860458135604, "time": 0.8604765892028808, "epoch": 9, "memory": 28783, "step": 51464} +{"lr": 1.8594235253127375e-05, "data_time": 0.004299187660217285, "grad_norm": 0.30173583431169393, "loss": 0.1988319382071495, "time": 0.9336992263793945, "epoch": 9, "memory": 28783, "step": 51564} +{"lr": 1.8594235253127375e-05, "data_time": 0.0036576032638549806, "grad_norm": 0.2116669312119484, "loss": 0.19918847531080247, "time": 0.946616244316101, "epoch": 9, "memory": 28783, "step": 51664} +{"lr": 1.8594235253127375e-05, "data_time": 0.0017846584320068359, "grad_norm": 0.354281132761389, "loss": 0.19975424855947493, "time": 0.8132057428359986, "epoch": 9, "memory": 28783, "step": 51764} +{"accuracy/top1": 99.96842956542969, "data_time": 0.20790562910192154, "time": 0.6230652612798354, "step": 9} +{"lr": 1.2202456766718093e-05, "data_time": 0.13533883094787597, "grad_norm": 0.19301287070848047, "loss": 0.19917882829904557, "time": 0.9107456684112549, "epoch": 10, "memory": 28783, "step": 51922} +{"lr": 1.2202456766718093e-05, "data_time": 0.22331948280334474, "grad_norm": 0.32406466929242017, "loss": 0.19940428137779237, "time": 0.8882270336151123, "epoch": 10, "memory": 28783, "step": 52022} +{"lr": 1.2202456766718093e-05, "data_time": 0.21347033977508545, "grad_norm": 0.35621108375489713, "loss": 0.19919017255306243, "time": 0.8834541320800782, "epoch": 10, "memory": 28783, "step": 52122} +{"lr": 1.2202456766718093e-05, "data_time": 0.11416516304016114, "grad_norm": 0.09284001146443188, "loss": 0.19853756725788116, "time": 0.8346410512924194, "epoch": 10, "memory": 28783, "step": 52222} +{"lr": 1.2202456766718093e-05, "data_time": 0.057669997215270996, "grad_norm": 0.22106292573735117, "loss": 0.19868463277816772, "time": 0.7937537908554078, "epoch": 10, "memory": 28783, "step": 52322} +{"lr": 1.2202456766718093e-05, "data_time": 0.0016902685165405273, "grad_norm": 0.3011444129049778, "loss": 0.20097961723804475, "time": 0.8490314483642578, "epoch": 10, "memory": 28783, "step": 52422} +{"lr": 1.2202456766718093e-05, "data_time": 0.22680144309997557, "grad_norm": 0.24370603486895562, "loss": 0.1990018293261528, "time": 0.7836069822311401, "epoch": 10, "memory": 28783, "step": 52522} +{"lr": 1.2202456766718093e-05, "data_time": 0.11063516139984131, "grad_norm": 0.20700901304371655, "loss": 0.19861238002777098, "time": 0.8661819934844971, "epoch": 10, "memory": 28783, "step": 52622} +{"lr": 1.2202456766718093e-05, "data_time": 0.012012052536010741, "grad_norm": 0.2800087394192815, "loss": 0.2002096265554428, "time": 0.8858421325683594, "epoch": 10, "memory": 28783, "step": 52722} +{"lr": 1.2202456766718093e-05, "data_time": 0.0035236835479736327, "grad_norm": 0.21200039803516119, "loss": 0.19939270615577698, "time": 0.942995572090149, "epoch": 10, "memory": 28783, "step": 52822} +{"lr": 1.2202456766718093e-05, "data_time": 0.0017570257186889648, "grad_norm": 0.2766184563515708, "loss": 0.19905015975236892, "time": 0.8427855253219605, "epoch": 10, "memory": 28783, "step": 52922} +{"lr": 1.2202456766718093e-05, "data_time": 0.0020080327987670897, "grad_norm": 0.3116778463125229, "loss": 0.19909278601408004, "time": 0.8508692741394043, "epoch": 10, "memory": 28783, "step": 53022} +{"lr": 1.2202456766718093e-05, "data_time": 0.0016348600387573243, "grad_norm": 0.27893140707165004, "loss": 0.200196073949337, "time": 0.8970194339752198, "epoch": 10, "memory": 28783, "step": 53122} +{"lr": 1.2202456766718093e-05, "data_time": 0.002549886703491211, "grad_norm": 0.16098835570737718, "loss": 0.19857225716114044, "time": 0.9008079767227173, "epoch": 10, "memory": 28783, "step": 53222} +{"lr": 1.2202456766718093e-05, "data_time": 0.002266192436218262, "grad_norm": 0.09969056223053485, "loss": 0.1988989979028702, "time": 0.943049144744873, "epoch": 10, "memory": 28783, "step": 53322} +{"lr": 1.2202456766718093e-05, "data_time": 0.0021249055862426758, "grad_norm": 0.29190522104036065, "loss": 0.19949427545070647, "time": 0.8756329774856567, "epoch": 10, "memory": 28783, "step": 53422} +{"lr": 1.2202456766718093e-05, "data_time": 0.0030753374099731444, "grad_norm": 0.3730745716020465, "loss": 0.19954436421394348, "time": 0.9552780151367187, "epoch": 10, "memory": 28783, "step": 53522} +{"lr": 1.2202456766718093e-05, "data_time": 0.0022222518920898436, "grad_norm": 0.537413590401411, "loss": 0.1994130328297615, "time": 0.9699157476425171, "epoch": 10, "memory": 28783, "step": 53622} +{"lr": 1.2202456766718093e-05, "data_time": 0.0023249387741088867, "grad_norm": 0.20427514445036649, "loss": 0.1986580327153206, "time": 0.9012442111968995, "epoch": 10, "memory": 28783, "step": 53722} +{"lr": 1.2202456766718093e-05, "data_time": 0.0018619775772094726, "grad_norm": 0.15136250564828516, "loss": 0.19909613877534865, "time": 0.7621897697448731, "epoch": 10, "memory": 28783, "step": 53822} +{"lr": 1.2202456766718093e-05, "data_time": 0.0017845630645751953, "grad_norm": 0.1600125606637448, "loss": 0.19864918887615204, "time": 0.7884865045547486, "epoch": 10, "memory": 28783, "step": 53922} +{"lr": 1.2202456766718093e-05, "data_time": 0.0034561634063720705, "grad_norm": 0.16006032302975653, "loss": 0.1993183746933937, "time": 1.0244053602218628, "epoch": 10, "memory": 28783, "step": 54022} +{"lr": 1.2202456766718093e-05, "data_time": 0.0033855676651000977, "grad_norm": 0.22914999574422837, "loss": 0.19899467825889589, "time": 0.9077329635620117, "epoch": 10, "memory": 28783, "step": 54122} +{"lr": 1.2202456766718093e-05, "data_time": 0.0017466306686401366, "grad_norm": 0.49763806434348223, "loss": 0.19974515587091446, "time": 0.8655405282974243, "epoch": 10, "memory": 28783, "step": 54222} +{"lr": 1.2202456766718093e-05, "data_time": 0.0015778541564941406, "grad_norm": 0.24855962507426738, "loss": 0.1993138983845711, "time": 0.8173866510391236, "epoch": 10, "memory": 28783, "step": 54322} +{"lr": 1.2202456766718093e-05, "data_time": 0.003122901916503906, "grad_norm": 0.2541944248601794, "loss": 0.19873177111148835, "time": 0.911636209487915, "epoch": 10, "memory": 28783, "step": 54422} +{"lr": 1.2202456766718093e-05, "data_time": 0.0023128271102905275, "grad_norm": 0.36098565105348823, "loss": 0.20084340125322342, "time": 0.8772715330123901, "epoch": 10, "memory": 28783, "step": 54522} +{"lr": 1.2202456766718093e-05, "data_time": 0.0024342775344848634, "grad_norm": 0.2083258915692568, "loss": 0.1992798313498497, "time": 0.8672708034515381, "epoch": 10, "memory": 28783, "step": 54622} +{"lr": 1.2202456766718093e-05, "data_time": 0.0017805099487304688, "grad_norm": 0.6617534391582012, "loss": 0.19961759597063064, "time": 0.7648499011993408, "epoch": 10, "memory": 28783, "step": 54722} +{"lr": 1.2202456766718093e-05, "data_time": 0.0027037620544433593, "grad_norm": 0.23564027417451144, "loss": 0.19873437136411667, "time": 0.7910962581634522, "epoch": 10, "memory": 28783, "step": 54822} +{"lr": 1.2202456766718093e-05, "data_time": 0.0023592472076416015, "grad_norm": 0.3154816431924701, "loss": 0.19969159811735154, "time": 0.8554802894592285, "epoch": 10, "memory": 28783, "step": 54922} +{"lr": 1.2202456766718093e-05, "data_time": 0.00893406867980957, "grad_norm": 0.17273001032881438, "loss": 0.19944452345371247, "time": 0.8116443157196045, "epoch": 10, "memory": 28783, "step": 55022} +{"lr": 1.2202456766718093e-05, "data_time": 0.1581697940826416, "grad_norm": 0.35414931140840056, "loss": 0.19946963340044022, "time": 0.8220585107803344, "epoch": 10, "memory": 28783, "step": 55122} +{"lr": 1.2202456766718093e-05, "data_time": 0.14808144569396972, "grad_norm": 0.26122904345393183, "loss": 0.19855034351348877, "time": 0.8965449094772339, "epoch": 10, "memory": 28783, "step": 55222} +{"lr": 1.2202456766718093e-05, "data_time": 0.06742925643920898, "grad_norm": 0.3018125146627426, "loss": 0.19951617270708083, "time": 0.9045409440994263, "epoch": 10, "memory": 28783, "step": 55322} +{"lr": 1.2202456766718093e-05, "data_time": 0.0192230224609375, "grad_norm": 0.26598530132323506, "loss": 0.19922469407320023, "time": 0.8526290655136108, "epoch": 10, "memory": 28783, "step": 55422} +{"lr": 1.2202456766718093e-05, "data_time": 0.0017828702926635741, "grad_norm": 0.41142064253799615, "loss": 0.1989261493086815, "time": 0.8375653982162475, "epoch": 10, "memory": 28783, "step": 55522} +{"lr": 1.2202456766718093e-05, "data_time": 0.0022647857666015627, "grad_norm": 0.3614894837141037, "loss": 0.19897508025169372, "time": 0.8872854709625244, "epoch": 10, "memory": 28783, "step": 55622} +{"lr": 1.2202456766718093e-05, "data_time": 0.0026006460189819335, "grad_norm": 0.19555093450471758, "loss": 0.199823234975338, "time": 0.8065153598785401, "epoch": 10, "memory": 28783, "step": 55722} +{"lr": 1.2202456766718093e-05, "data_time": 0.0026729345321655274, "grad_norm": 0.1707129318267107, "loss": 0.19865113347768784, "time": 0.9116674184799194, "epoch": 10, "memory": 28783, "step": 55822} +{"lr": 1.2202456766718093e-05, "data_time": 0.0024576902389526365, "grad_norm": 0.1427366454154253, "loss": 0.19858566522598267, "time": 0.8400616645812988, "epoch": 10, "memory": 28783, "step": 55922} +{"lr": 1.2202456766718093e-05, "data_time": 0.001971149444580078, "grad_norm": 0.28716955725103616, "loss": 0.19943950772285463, "time": 0.8212677478790283, "epoch": 10, "memory": 28783, "step": 56022} +{"lr": 1.2202456766718093e-05, "data_time": 0.0023377180099487306, "grad_norm": 0.15794655294157564, "loss": 0.20017153024673462, "time": 0.9557317972183228, "epoch": 10, "memory": 28783, "step": 56122} +{"lr": 1.2202456766718093e-05, "data_time": 0.004235410690307617, "grad_norm": 0.16744602820836008, "loss": 0.19898456782102586, "time": 0.9123448133468628, "epoch": 10, "memory": 28783, "step": 56222} +{"lr": 1.2202456766718093e-05, "data_time": 0.0017568826675415038, "grad_norm": 0.27650420600548387, "loss": 0.19901863038539885, "time": 0.8098387479782104, "epoch": 10, "memory": 28783, "step": 56322} +{"lr": 1.2202456766718093e-05, "data_time": 0.0019932985305786133, "grad_norm": 0.23942797668278218, "loss": 0.19960306584835052, "time": 0.8894573450088501, "epoch": 10, "memory": 28783, "step": 56422} +{"lr": 1.2202456766718093e-05, "data_time": 0.00220944881439209, "grad_norm": 0.36436382234096526, "loss": 0.20020554661750795, "time": 0.8621434211730957, "epoch": 10, "memory": 28783, "step": 56522} +{"lr": 1.2202456766718093e-05, "data_time": 0.0021554231643676758, "grad_norm": 0.2618440832477063, "loss": 0.19875305145978928, "time": 0.8395252227783203, "epoch": 10, "memory": 28783, "step": 56622} +{"lr": 1.2202456766718093e-05, "data_time": 0.0017965316772460937, "grad_norm": 0.35517810806632044, "loss": 0.1996025949716568, "time": 0.7628729104995727, "epoch": 10, "memory": 28783, "step": 56722} +{"lr": 1.2202456766718093e-05, "data_time": 0.0017314434051513671, "grad_norm": 0.3564167030155659, "loss": 0.19922177642583846, "time": 0.798421835899353, "epoch": 10, "memory": 28783, "step": 56822} +{"lr": 1.2202456766718093e-05, "data_time": 0.009748721122741699, "grad_norm": 0.42114703767001627, "loss": 0.1999643102288246, "time": 0.7876276731491089, "epoch": 10, "memory": 28783, "step": 56922} +{"lr": 1.2202456766718093e-05, "data_time": 0.2023008108139038, "grad_norm": 0.22653089836239815, "loss": 0.1989861860871315, "time": 0.9507332086563111, "epoch": 10, "memory": 28783, "step": 57022} +{"lr": 1.2202456766718093e-05, "data_time": 0.14067411422729492, "grad_norm": 0.28799118380993605, "loss": 0.1986668199300766, "time": 0.8300880670547486, "epoch": 10, "memory": 28783, "step": 57122} +{"lr": 1.2202456766718093e-05, "data_time": 0.06402592658996582, "grad_norm": 0.1465356174390763, "loss": 0.198752561211586, "time": 0.8175986051559448, "epoch": 10, "memory": 28783, "step": 57222} +{"lr": 1.2202456766718093e-05, "data_time": 0.11023647785186767, "grad_norm": 0.5015743039548397, "loss": 0.19983679354190825, "time": 0.8183055400848389, "epoch": 10, "memory": 28783, "step": 57322} +{"lr": 1.2202456766718093e-05, "data_time": 0.015048861503601074, "grad_norm": 0.2388107413891703, "loss": 0.20051872879266738, "time": 0.9866007804870606, "epoch": 10, "memory": 28783, "step": 57422} +{"lr": 1.2202456766718093e-05, "data_time": 0.0037036418914794924, "grad_norm": 1.2485242391005158, "loss": 0.19872440844774247, "time": 0.969190239906311, "epoch": 10, "memory": 28783, "step": 57522} +{"accuracy/top1": 99.96842956542969, "data_time": 0.20385313034057617, "time": 0.6185694161583396, "step": 10} diff --git a/swin_small_4xb256_fake5m_lr1e-4_10epoch/20230601_215742/vis_data/config.py b/swin_small_4xb256_fake5m_lr1e-4_10epoch/20230601_215742/vis_data/config.py new file mode 100644 index 0000000000000000000000000000000000000000..07c9f4bb104a3c4f894e7a05987a8354fd463e1f --- /dev/null +++ b/swin_small_4xb256_fake5m_lr1e-4_10epoch/20230601_215742/vis_data/config.py @@ -0,0 +1,175 @@ +optim_wrapper = dict( + optimizer=dict( + type='AdamW', + lr=0.0001, + weight_decay=0.05, + eps=1e-08, + betas=(0.9, 0.999), + _scope_='mmpretrain'), + paramwise_cfg=dict( + norm_decay_mult=0.0, + bias_decay_mult=0.0, + flat_decay_mult=0.0, + custom_keys=dict({ + '.absolute_pos_embed': dict(decay_mult=0.0), + '.relative_position_bias_table': dict(decay_mult=0.0) + })), + type='AmpOptimWrapper', + dtype='bfloat16', + clip_grad=dict(max_norm=5.0)) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=True, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=1024) +model = dict( + type='ImageClassifier', + backbone=dict( + type='SwinTransformer', arch='small', img_size=224, + drop_path_rate=0.3), + neck=dict(type='GlobalAveragePooling'), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=768, + init_cfg=None, + loss=dict( + type='LabelSmoothLoss', label_smooth_val=0.1, mode='original'), + cal_acc=False), + init_cfg=[ + dict(type='TruncNormal', layer='Linear', std=0.02, bias=0.0), + dict(type='Constant', layer='LayerNorm', val=1.0, bias=0.0) + ], + train_cfg=None) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stablediffusionV1-5R2-dpmsolver-25-5m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/swin_small_4xb256_fake5m_lr1e-4_10epoch' diff --git a/swin_small_4xb256_fake5m_lr1e-4_10epoch/20230601_215742/vis_data/events.out.tfevents.1685627870.SH-IDC1-10-140-24-125.179316.0 b/swin_small_4xb256_fake5m_lr1e-4_10epoch/20230601_215742/vis_data/events.out.tfevents.1685627870.SH-IDC1-10-140-24-125.179316.0 new file mode 100644 index 0000000000000000000000000000000000000000..faf44af908a58d28f389ae594b1017a119c9c6fc --- /dev/null +++ b/swin_small_4xb256_fake5m_lr1e-4_10epoch/20230601_215742/vis_data/events.out.tfevents.1685627870.SH-IDC1-10-140-24-125.179316.0 @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:20de2e9525f973bbdd1fe128464d961c32e719edd30edadbb32c0c31cc059a5f +size 896817 diff --git a/swin_small_4xb256_fake5m_lr1e-4_10epoch/20230601_215742/vis_data/scalars.json b/swin_small_4xb256_fake5m_lr1e-4_10epoch/20230601_215742/vis_data/scalars.json new file mode 100644 index 0000000000000000000000000000000000000000..a357100b511907b74eea150685e3e0b2eb47f1a1 --- /dev/null +++ b/swin_small_4xb256_fake5m_lr1e-4_10epoch/20230601_215742/vis_data/scalars.json @@ -0,0 +1,580 @@ +{"lr": 0.0001, "data_time": 0.2836922168731689, "grad_norm": 1.1496466875076294, "loss": 0.6058561384677887, "time": 0.8801626920700073, "epoch": 1, "memory": 28798, "step": 100} +{"lr": 0.0001, "data_time": 0.4571942567825317, "grad_norm": 1.530609768629074, "loss": 0.5631434917449951, "time": 1.005289363861084, "epoch": 1, "memory": 28798, "step": 200} +{"lr": 0.0001, "data_time": 0.22151150703430175, "grad_norm": 2.028353178501129, "loss": 0.5296812951564789, "time": 0.9259072303771972, "epoch": 1, "memory": 28798, "step": 300} +{"lr": 0.0001, "data_time": 0.291133189201355, "grad_norm": 2.500625509023666, "loss": 0.5156413376331329, "time": 0.9341419219970704, "epoch": 1, "memory": 28798, "step": 400} +{"lr": 0.0001, "data_time": 0.12301592826843262, "grad_norm": 4.318602168560028, "loss": 0.5010578453540802, "time": 0.8412275791168213, "epoch": 1, "memory": 28798, "step": 500} +{"lr": 0.0001, "data_time": 0.13549425601959228, "grad_norm": 4.586491513252258, "loss": 0.4730592280626297, "time": 0.9765780925750732, "epoch": 1, "memory": 28798, "step": 600} +{"lr": 0.0001, "data_time": 0.16471946239471436, "grad_norm": 3.8183645486831663, "loss": 0.45454353988170626, "time": 0.8893230438232422, "epoch": 1, "memory": 28798, "step": 700} +{"lr": 0.0001, "data_time": 0.12637732028961182, "grad_norm": 6.184123384952545, "loss": 0.4634770840406418, "time": 0.8824618577957153, "epoch": 1, "memory": 28798, "step": 800} +{"lr": 0.0001, "data_time": 0.001864790916442871, "grad_norm": 3.7891004204750063, "loss": 0.44833930730819704, "time": 0.7174569368362427, "epoch": 1, "memory": 28798, "step": 900} +{"lr": 0.0001, "data_time": 0.0021742820739746095, "grad_norm": 6.032311713695526, "loss": 0.44841069281101226, "time": 0.8957549571990967, "epoch": 1, "memory": 28798, "step": 1000} +{"lr": 0.0001, "data_time": 0.0016405105590820313, "grad_norm": 3.7643929719924927, "loss": 0.42695329487323763, "time": 0.7580570459365845, "epoch": 1, "memory": 28798, "step": 1100} +{"lr": 0.0001, "data_time": 0.03567514419555664, "grad_norm": 8.661775887012482, "loss": 0.4228070259094238, "time": 0.7184351682662964, "epoch": 1, "memory": 28798, "step": 1200} +{"lr": 0.0001, "data_time": 0.1960144519805908, "grad_norm": 12.072546911239623, "loss": 0.43221528828144073, "time": 0.7568394422531128, "epoch": 1, "memory": 28798, "step": 1300} +{"lr": 0.0001, "data_time": 0.13183763027191162, "grad_norm": 9.784806501865386, "loss": 0.4166201651096344, "time": 0.9501391172409057, "epoch": 1, "memory": 28798, "step": 1400} +{"lr": 0.0001, "data_time": 0.0021010637283325195, "grad_norm": 8.564002323150635, "loss": 0.4024180382490158, "time": 0.7975595951080322, "epoch": 1, "memory": 28798, "step": 1500} +{"lr": 0.0001, "data_time": 0.002309083938598633, "grad_norm": 10.933034062385559, "loss": 0.43787465393543246, "time": 0.824269437789917, "epoch": 1, "memory": 28798, "step": 1600} +{"lr": 0.0001, "data_time": 0.16104977130889891, "grad_norm": 14.262428903579712, "loss": 0.4117710143327713, "time": 0.7807987689971924, "epoch": 1, "memory": 28798, "step": 1700} +{"lr": 0.0001, "data_time": 0.3967538118362427, "grad_norm": 6.842606687545777, "loss": 0.40439617037773135, "time": 0.971693754196167, "epoch": 1, "memory": 28798, "step": 1800} +{"lr": 0.0001, "data_time": 0.2468998908996582, "grad_norm": 6.033211255073548, "loss": 0.3812806487083435, "time": 0.8953802108764648, "epoch": 1, "memory": 28798, "step": 1900} +{"lr": 0.0001, "data_time": 0.3589166164398193, "grad_norm": 11.163988208770752, "loss": 0.3766529768705368, "time": 0.8789925098419189, "epoch": 1, "memory": 28798, "step": 2000} +{"lr": 0.0001, "data_time": 0.32769558429718015, "grad_norm": 6.921413576602935, "loss": 0.369454437494278, "time": 0.8403992176055908, "epoch": 1, "memory": 28798, "step": 2100} +{"lr": 0.0001, "data_time": 0.3894771337509155, "grad_norm": 16.07062668800354, "loss": 0.3762036055326462, "time": 0.9089213132858276, "epoch": 1, "memory": 28798, "step": 2200} +{"lr": 0.0001, "data_time": 0.31716940402984617, "grad_norm": 10.965106916427612, "loss": 0.3938374280929565, "time": 0.8222328186035156, "epoch": 1, "memory": 28798, "step": 2300} +{"lr": 0.0001, "data_time": 0.4321016788482666, "grad_norm": 11.419286108016967, "loss": 0.38529468774795533, "time": 0.9546619892120362, "epoch": 1, "memory": 28798, "step": 2400} +{"lr": 0.0001, "data_time": 0.4447009801864624, "grad_norm": 20.4749276638031, "loss": 0.39668129980564115, "time": 0.984907865524292, "epoch": 1, "memory": 28798, "step": 2500} +{"lr": 0.0001, "data_time": 0.3377225399017334, "grad_norm": 8.161865782737731, "loss": 0.34652950465679166, "time": 0.8513520002365113, "epoch": 1, "memory": 28798, "step": 2600} +{"lr": 0.0001, "data_time": 0.4989269018173218, "grad_norm": 4.820529353618622, "loss": 0.35654892921447756, "time": 1.0519294500350953, "epoch": 1, "memory": 28798, "step": 2700} +{"lr": 0.0001, "data_time": 0.2992296934127808, "grad_norm": 17.96520357131958, "loss": 0.36084577441215515, "time": 0.8058756589889526, "epoch": 1, "memory": 28798, "step": 2800} +{"lr": 0.0001, "data_time": 0.33991820812225343, "grad_norm": 14.215149307250977, "loss": 0.3633415728807449, "time": 0.8674774169921875, "epoch": 1, "memory": 28798, "step": 2900} +{"lr": 0.0001, "data_time": 0.3959709882736206, "grad_norm": 7.207492160797119, "loss": 0.3425798386335373, "time": 0.9157244920730591, "epoch": 1, "memory": 28798, "step": 3000} +{"lr": 0.0001, "data_time": 0.4099679946899414, "grad_norm": 11.321051740646363, "loss": 0.3522310942411423, "time": 0.9410527467727661, "epoch": 1, "memory": 28798, "step": 3100} +{"lr": 0.0001, "data_time": 0.2561708211898804, "grad_norm": 14.380841886997223, "loss": 0.3565657764673233, "time": 0.7644975662231446, "epoch": 1, "memory": 28798, "step": 3200} +{"lr": 0.0001, "data_time": 0.34224073886871337, "grad_norm": 13.345923590660096, "loss": 0.3287695676088333, "time": 0.8484790325164795, "epoch": 1, "memory": 28798, "step": 3300} +{"lr": 0.0001, "data_time": 0.3237328052520752, "grad_norm": 8.821040737628937, "loss": 0.31754212975502016, "time": 0.8322324275970459, "epoch": 1, "memory": 28798, "step": 3400} +{"lr": 0.0001, "data_time": 0.4927107572555542, "grad_norm": 6.542399203777313, "loss": 0.33546442091464995, "time": 1.027521538734436, "epoch": 1, "memory": 28798, "step": 3500} +{"lr": 0.0001, "data_time": 0.2770772218704224, "grad_norm": 11.564435482025146, "loss": 0.32049880623817445, "time": 0.8759184122085572, "epoch": 1, "memory": 28798, "step": 3600} +{"lr": 0.0001, "data_time": 0.09506022930145264, "grad_norm": 5.517071306705475, "loss": 0.3160597801208496, "time": 0.8441982984542846, "epoch": 1, "memory": 28798, "step": 3700} +{"lr": 0.0001, "data_time": 0.002781963348388672, "grad_norm": 25.570412254333498, "loss": 0.40939798653125764, "time": 0.8697173118591308, "epoch": 1, "memory": 28798, "step": 3800} +{"lr": 0.0001, "data_time": 0.0017022132873535157, "grad_norm": 14.403342628479004, "loss": 0.3184275537729263, "time": 0.7843376874923706, "epoch": 1, "memory": 28798, "step": 3900} +{"lr": 0.0001, "data_time": 0.002773571014404297, "grad_norm": 13.11492998600006, "loss": 0.3069653451442719, "time": 0.867843747138977, "epoch": 1, "memory": 28798, "step": 4000} +{"lr": 0.0001, "data_time": 0.002892303466796875, "grad_norm": 13.996675157546997, "loss": 0.32083846628665924, "time": 0.9147140979766846, "epoch": 1, "memory": 28798, "step": 4100} +{"lr": 0.0001, "data_time": 0.002616620063781738, "grad_norm": 4.992052376270294, "loss": 0.30638874471187594, "time": 1.0079552412033081, "epoch": 1, "memory": 28798, "step": 4200} +{"lr": 0.0001, "data_time": 0.0022130727767944334, "grad_norm": 17.318383026123048, "loss": 0.3292596787214279, "time": 0.8624425172805786, "epoch": 1, "memory": 28798, "step": 4300} +{"lr": 0.0001, "data_time": 0.003058719635009766, "grad_norm": 7.483843278884888, "loss": 0.3052514672279358, "time": 0.87148118019104, "epoch": 1, "memory": 28798, "step": 4400} +{"lr": 0.0001, "data_time": 0.0025171279907226563, "grad_norm": 17.134615588188172, "loss": 0.3356024086475372, "time": 0.9511116027832032, "epoch": 1, "memory": 28798, "step": 4500} +{"lr": 0.0001, "data_time": 0.001605391502380371, "grad_norm": 7.5547421932220455, "loss": 0.30554085671901704, "time": 0.8072681665420532, "epoch": 1, "memory": 28798, "step": 4600} +{"lr": 0.0001, "data_time": 0.0018180608749389648, "grad_norm": 12.283394455909729, "loss": 0.3418678015470505, "time": 0.8453145265579224, "epoch": 1, "memory": 28798, "step": 4700} +{"lr": 0.0001, "data_time": 0.005225729942321777, "grad_norm": 11.814648818969726, "loss": 0.3138525366783142, "time": 0.8357168197631836, "epoch": 1, "memory": 28798, "step": 4800} +{"lr": 0.0001, "data_time": 0.0026343822479248046, "grad_norm": 19.7075412273407, "loss": 0.33815790712833405, "time": 0.9410369157791137, "epoch": 1, "memory": 28798, "step": 4900} +{"lr": 0.0001, "data_time": 0.0018437862396240234, "grad_norm": 15.569157981872559, "loss": 0.34495705366134644, "time": 0.8171157360076904, "epoch": 1, "memory": 28798, "step": 5000} +{"lr": 0.0001, "data_time": 0.002482438087463379, "grad_norm": 21.3685173869133, "loss": 0.323750838637352, "time": 0.8111368656158447, "epoch": 1, "memory": 28798, "step": 5100} +{"lr": 0.0001, "data_time": 0.001826310157775879, "grad_norm": 10.369468212127686, "loss": 0.2822732239961624, "time": 0.8334531307220459, "epoch": 1, "memory": 28798, "step": 5200} +{"lr": 0.0001, "data_time": 0.0018332242965698243, "grad_norm": 16.070834732055665, "loss": 0.3088524341583252, "time": 0.979086446762085, "epoch": 1, "memory": 28798, "step": 5300} +{"lr": 0.0001, "data_time": 0.0031427860260009764, "grad_norm": 8.568836116790772, "loss": 0.28733194768428805, "time": 0.9507247924804687, "epoch": 1, "memory": 28798, "step": 5400} +{"lr": 0.0001, "data_time": 0.0032306671142578124, "grad_norm": 5.664257299900055, "loss": 0.27612653076648713, "time": 0.9241308450698853, "epoch": 1, "memory": 28798, "step": 5500} +{"lr": 0.0001, "data_time": 0.006642651557922363, "grad_norm": 6.408056902885437, "loss": 0.2865743860602379, "time": 1.0534418821334839, "epoch": 1, "memory": 28798, "step": 5600} +{"lr": 0.0001, "data_time": 0.0032753705978393554, "grad_norm": 6.494708776473999, "loss": 0.2806232213973999, "time": 0.8970215559005738, "epoch": 1, "memory": 28798, "step": 5700} +{"accuracy/top1": 98.81320190429688, "data_time": 0.37856800854206085, "time": 1.0039068758487701, "step": 1} +{"lr": 9.779754323328192e-05, "data_time": 0.38404159545898436, "grad_norm": 4.161892879009247, "loss": 0.27686722576618195, "time": 1.0489650011062621, "epoch": 2, "memory": 28798, "step": 5858} +{"lr": 9.779754323328192e-05, "data_time": 0.1485985517501831, "grad_norm": 4.189602971076965, "loss": 0.2639403223991394, "time": 0.8727232456207276, "epoch": 2, "memory": 28783, "step": 5958} +{"lr": 9.779754323328192e-05, "data_time": 0.0027503252029418947, "grad_norm": 5.856673264503479, "loss": 0.2715042933821678, "time": 0.944049620628357, "epoch": 2, "memory": 28783, "step": 6058} +{"lr": 9.779754323328192e-05, "data_time": 0.14574897289276123, "grad_norm": 12.224393248558044, "loss": 0.28722442239522933, "time": 0.9152533769607544, "epoch": 2, "memory": 28783, "step": 6158} +{"lr": 9.779754323328192e-05, "data_time": 0.3519598960876465, "grad_norm": 6.210837030410767, "loss": 0.26650336682796477, "time": 0.8882764101028442, "epoch": 2, "memory": 28783, "step": 6258} +{"lr": 9.779754323328192e-05, "data_time": 0.3771564722061157, "grad_norm": 3.9214691162109374, "loss": 0.2589657470583916, "time": 0.8983821630477905, "epoch": 2, "memory": 28783, "step": 6358} +{"lr": 9.779754323328192e-05, "data_time": 0.24103910923004152, "grad_norm": 9.535576891899108, "loss": 0.27185446619987486, "time": 0.7520332098007202, "epoch": 2, "memory": 28783, "step": 6458} +{"lr": 9.779754323328192e-05, "data_time": 0.3728133201599121, "grad_norm": 18.79650083780289, "loss": 0.284046670794487, "time": 0.8914100170135498, "epoch": 2, "memory": 28783, "step": 6558} +{"lr": 9.779754323328192e-05, "data_time": 0.4151688814163208, "grad_norm": 5.077081775665283, "loss": 0.2608530417084694, "time": 0.9224797010421752, "epoch": 2, "memory": 28783, "step": 6658} +{"lr": 9.779754323328192e-05, "data_time": 0.4156410455703735, "grad_norm": 15.912617897987365, "loss": 0.2787394493818283, "time": 0.9759310960769654, "epoch": 2, "memory": 28783, "step": 6758} +{"lr": 9.779754323328192e-05, "data_time": 0.4599510669708252, "grad_norm": 3.741653895378113, "loss": 0.2555323660373688, "time": 0.9699604988098145, "epoch": 2, "memory": 28783, "step": 6858} +{"lr": 9.779754323328192e-05, "data_time": 0.41335225105285645, "grad_norm": 3.530123460292816, "loss": 0.24913543313741685, "time": 0.9404096364974975, "epoch": 2, "memory": 28783, "step": 6958} +{"lr": 9.779754323328192e-05, "data_time": 0.47891817092895506, "grad_norm": 3.114583659172058, "loss": 0.2537370786070824, "time": 0.9979316711425781, "epoch": 2, "memory": 28783, "step": 7058} +{"lr": 9.779754323328192e-05, "data_time": 0.3670587301254272, "grad_norm": 9.071798992156982, "loss": 0.2690409183502197, "time": 0.8926671028137207, "epoch": 2, "memory": 28783, "step": 7158} +{"lr": 9.779754323328192e-05, "data_time": 0.37719457149505614, "grad_norm": 5.893570971488953, "loss": 0.24558966606855392, "time": 0.9014822721481324, "epoch": 2, "memory": 28783, "step": 7258} +{"lr": 9.779754323328192e-05, "data_time": 0.3808925628662109, "grad_norm": 5.37038791179657, "loss": 0.24915929287672042, "time": 0.9220894336700439, "epoch": 2, "memory": 28783, "step": 7358} +{"lr": 9.779754323328192e-05, "data_time": 0.4557778358459473, "grad_norm": 8.520963859558105, "loss": 0.2678984612226486, "time": 0.9937985897064209, "epoch": 2, "memory": 28783, "step": 7458} +{"lr": 9.779754323328192e-05, "data_time": 0.3625136137008667, "grad_norm": 2.5578632593154906, "loss": 0.24371749758720399, "time": 0.8921395301818847, "epoch": 2, "memory": 28783, "step": 7558} +{"lr": 9.779754323328192e-05, "data_time": 0.38992717266082766, "grad_norm": 4.372581660747528, "loss": 0.2427796557545662, "time": 0.9112889051437378, "epoch": 2, "memory": 28783, "step": 7658} +{"lr": 9.779754323328192e-05, "data_time": 0.5067874431610108, "grad_norm": 4.538609194755554, "loss": 0.24747199416160584, "time": 1.0188977479934693, "epoch": 2, "memory": 28783, "step": 7758} +{"lr": 9.779754323328192e-05, "data_time": 0.32326455116271974, "grad_norm": 5.38339084982872, "loss": 0.24430325925350188, "time": 0.8366075754165649, "epoch": 2, "memory": 28783, "step": 7858} +{"lr": 9.779754323328192e-05, "data_time": 0.33705978393554686, "grad_norm": 8.676961934566497, "loss": 0.2449193611741066, "time": 0.847307276725769, "epoch": 2, "memory": 28783, "step": 7958} +{"lr": 9.779754323328192e-05, "data_time": 0.4482605218887329, "grad_norm": 8.488741207122803, "loss": 0.24980992674827576, "time": 0.9851110219955445, "epoch": 2, "memory": 28783, "step": 8058} +{"lr": 9.779754323328192e-05, "data_time": 0.29235076904296875, "grad_norm": 2.4637938380241393, "loss": 0.24306034594774245, "time": 0.8088602066040039, "epoch": 2, "memory": 28783, "step": 8158} +{"lr": 9.779754323328192e-05, "data_time": 0.47416865825653076, "grad_norm": 4.897786259651184, "loss": 0.24559639841318132, "time": 0.9933578968048096, "epoch": 2, "memory": 28783, "step": 8258} +{"lr": 9.779754323328192e-05, "data_time": 0.31468679904937746, "grad_norm": 4.063173985481262, "loss": 0.23778026849031447, "time": 0.858015775680542, "epoch": 2, "memory": 28783, "step": 8358} +{"lr": 9.779754323328192e-05, "data_time": 0.2870336055755615, "grad_norm": 3.8485030055046083, "loss": 0.23274193704128265, "time": 0.7924270629882812, "epoch": 2, "memory": 28783, "step": 8458} +{"lr": 9.779754323328192e-05, "data_time": 0.46350076198577883, "grad_norm": 2.754397398233414, "loss": 0.2309154450893402, "time": 0.9834944486618042, "epoch": 2, "memory": 28783, "step": 8558} +{"lr": 9.779754323328192e-05, "data_time": 0.4629937171936035, "grad_norm": 2.449435019493103, "loss": 0.2383464351296425, "time": 0.9932717084884644, "epoch": 2, "memory": 28783, "step": 8658} +{"lr": 9.779754323328192e-05, "data_time": 0.41739070415496826, "grad_norm": 8.001243257522583, "loss": 0.23662171363830567, "time": 0.9370226383209228, "epoch": 2, "memory": 28783, "step": 8758} +{"lr": 9.779754323328192e-05, "data_time": 0.4663546562194824, "grad_norm": 2.121528911590576, "loss": 0.22635733634233474, "time": 0.9992094993591308, "epoch": 2, "memory": 28783, "step": 8858} +{"lr": 9.779754323328192e-05, "data_time": 0.309641695022583, "grad_norm": 9.506640934944153, "loss": 0.22828049659729005, "time": 0.8181777000427246, "epoch": 2, "memory": 28783, "step": 8958} +{"lr": 9.779754323328192e-05, "data_time": 0.39505982398986816, "grad_norm": 7.051111888885498, "loss": 0.2564105600118637, "time": 0.9207224369049072, "epoch": 2, "memory": 28783, "step": 9058} +{"lr": 9.779754323328192e-05, "data_time": 0.33772108554840086, "grad_norm": 2.7572768568992614, "loss": 0.23249821215867997, "time": 0.8559291362762451, "epoch": 2, "memory": 28783, "step": 9158} +{"lr": 9.779754323328192e-05, "data_time": 0.48703229427337646, "grad_norm": 2.758994436264038, "loss": 0.23604095578193665, "time": 1.0598036527633667, "epoch": 2, "memory": 28783, "step": 9258} +{"lr": 9.779754323328192e-05, "data_time": 0.4315407991409302, "grad_norm": 2.9783724665641786, "loss": 0.22422760277986525, "time": 0.9436061143875122, "epoch": 2, "memory": 28783, "step": 9358} +{"lr": 9.779754323328192e-05, "data_time": 0.3649235486984253, "grad_norm": 3.7142862677574158, "loss": 0.2300599217414856, "time": 0.8965791940689087, "epoch": 2, "memory": 28783, "step": 9458} +{"lr": 9.779754323328192e-05, "data_time": 0.3484864473342896, "grad_norm": 4.165369379520416, "loss": 0.22893126308918, "time": 0.850684642791748, "epoch": 2, "memory": 28783, "step": 9558} +{"lr": 9.779754323328192e-05, "data_time": 0.399152946472168, "grad_norm": 4.1010830998420715, "loss": 0.23013428449630738, "time": 0.9175976276397705, "epoch": 2, "memory": 28783, "step": 9658} +{"lr": 9.779754323328192e-05, "data_time": 0.3455408811569214, "grad_norm": 4.721993350982666, "loss": 0.22365710586309434, "time": 0.8500763654708863, "epoch": 2, "memory": 28783, "step": 9758} +{"lr": 9.779754323328192e-05, "data_time": 0.4002149343490601, "grad_norm": 3.0715201020240785, "loss": 0.23425911366939545, "time": 0.9193668842315674, "epoch": 2, "memory": 28783, "step": 9858} +{"lr": 9.779754323328192e-05, "data_time": 0.31200993061065674, "grad_norm": 1.6474700093269348, "loss": 0.22240086942911147, "time": 0.8182117462158203, "epoch": 2, "memory": 28783, "step": 9958} +{"lr": 9.779754323328192e-05, "data_time": 0.3542064905166626, "grad_norm": 2.732310104370117, "loss": 0.220636785030365, "time": 0.8919193983078003, "epoch": 2, "memory": 28783, "step": 10058} +{"lr": 9.779754323328192e-05, "data_time": 0.43896524906158446, "grad_norm": 2.871346414089203, "loss": 0.21896641999483107, "time": 0.9701106786727905, "epoch": 2, "memory": 28783, "step": 10158} +{"lr": 9.779754323328192e-05, "data_time": 0.3461632251739502, "grad_norm": 1.8939796268939972, "loss": 0.22449996918439866, "time": 0.8784794330596923, "epoch": 2, "memory": 28783, "step": 10258} +{"lr": 9.779754323328192e-05, "data_time": 0.36941554546356203, "grad_norm": 2.0798502802848815, "loss": 0.22079380601644516, "time": 0.8904143095016479, "epoch": 2, "memory": 28783, "step": 10358} +{"lr": 9.779754323328192e-05, "data_time": 0.436677360534668, "grad_norm": 3.5731325268745424, "loss": 0.23058802634477615, "time": 0.9835659503936768, "epoch": 2, "memory": 28783, "step": 10458} +{"lr": 9.779754323328192e-05, "data_time": 0.35236248970031736, "grad_norm": 1.923442804813385, "loss": 0.22004744857549668, "time": 0.8601249694824219, "epoch": 2, "memory": 28783, "step": 10558} +{"lr": 9.779754323328192e-05, "data_time": 0.4356717109680176, "grad_norm": 1.6966484546661378, "loss": 0.2197301283478737, "time": 0.9548774719238281, "epoch": 2, "memory": 28783, "step": 10658} +{"lr": 9.779754323328192e-05, "data_time": 0.2778660297393799, "grad_norm": 2.028006798028946, "loss": 0.2189297616481781, "time": 0.7813084602355957, "epoch": 2, "memory": 28783, "step": 10758} +{"lr": 9.779754323328192e-05, "data_time": 0.3047685384750366, "grad_norm": 2.468812119960785, "loss": 0.2178144261240959, "time": 0.8396297216415405, "epoch": 2, "memory": 28783, "step": 10858} +{"lr": 9.779754323328192e-05, "data_time": 0.3942467212677002, "grad_norm": 1.7278743147850038, "loss": 0.21907654255628586, "time": 0.9061642408370971, "epoch": 2, "memory": 28783, "step": 10958} +{"lr": 9.779754323328192e-05, "data_time": 0.33275651931762695, "grad_norm": 3.3575263261795043, "loss": 0.21982483863830565, "time": 0.8706469535827637, "epoch": 2, "memory": 28783, "step": 11058} +{"lr": 9.779754323328192e-05, "data_time": 0.38285465240478517, "grad_norm": 2.402519243955612, "loss": 0.21914853155612946, "time": 0.9026891231536865, "epoch": 2, "memory": 28783, "step": 11158} +{"lr": 9.779754323328192e-05, "data_time": 0.45121476650238035, "grad_norm": 1.9819308876991273, "loss": 0.21823106110095977, "time": 0.9740733623504638, "epoch": 2, "memory": 28783, "step": 11258} +{"lr": 9.779754323328192e-05, "data_time": 0.3432604789733887, "grad_norm": 1.8669257402420043, "loss": 0.21990397274494172, "time": 0.858852219581604, "epoch": 2, "memory": 28783, "step": 11358} +{"lr": 9.779754323328192e-05, "data_time": 0.2904154062271118, "grad_norm": 1.7137621462345123, "loss": 0.21675174683332443, "time": 0.7964584827423096, "epoch": 2, "memory": 28783, "step": 11458} +{"accuracy/top1": 99.50760650634766, "data_time": 0.24432659149169922, "time": 0.6612234957077924, "step": 2} +{"lr": 9.140576474687264e-05, "data_time": 0.34173638820648194, "grad_norm": 4.991871964931488, "loss": 0.22893251180648805, "time": 0.9203919887542724, "epoch": 3, "memory": 28783, "step": 11616} +{"lr": 9.140576474687264e-05, "data_time": 0.3464978218078613, "grad_norm": 1.480863666534424, "loss": 0.2140223801136017, "time": 0.9392688751220704, "epoch": 3, "memory": 28783, "step": 11716} +{"lr": 9.140576474687264e-05, "data_time": 0.4532526254653931, "grad_norm": 3.467792344093323, "loss": 0.2252451717853546, "time": 0.9688373327255249, "epoch": 3, "memory": 28783, "step": 11816} +{"lr": 9.140576474687264e-05, "data_time": 0.5468847990036011, "grad_norm": 1.8249448239803314, "loss": 0.21538495272397995, "time": 1.082854652404785, "epoch": 3, "memory": 28783, "step": 11916} +{"lr": 9.140576474687264e-05, "data_time": 0.39746015071868895, "grad_norm": 1.5756512761116028, "loss": 0.21620576083660126, "time": 0.9389734983444213, "epoch": 3, "memory": 28783, "step": 12016} +{"lr": 9.140576474687264e-05, "data_time": 0.3889593601226807, "grad_norm": 3.1359798312187195, "loss": 0.217052561044693, "time": 0.9061183214187623, "epoch": 3, "memory": 28783, "step": 12116} +{"lr": 9.140576474687264e-05, "data_time": 0.3880429983139038, "grad_norm": 2.338692659139633, "loss": 0.2130947843194008, "time": 0.9091630220413208, "epoch": 3, "memory": 28783, "step": 12216} +{"lr": 9.140576474687264e-05, "data_time": 0.4014167070388794, "grad_norm": 2.50209698677063, "loss": 0.21859853118658065, "time": 0.9058146238327026, "epoch": 3, "memory": 28783, "step": 12316} +{"lr": 9.140576474687264e-05, "data_time": 0.4401841163635254, "grad_norm": 1.5620913565158845, "loss": 0.20996534675359727, "time": 0.991347336769104, "epoch": 3, "memory": 28783, "step": 12416} +{"lr": 9.140576474687264e-05, "data_time": 0.3722501039505005, "grad_norm": 2.1559678494930266, "loss": 0.2115705728530884, "time": 0.8778233051300048, "epoch": 3, "memory": 28783, "step": 12516} +{"lr": 9.140576474687264e-05, "data_time": 0.42272875308990476, "grad_norm": 3.18854238986969, "loss": 0.2241248294711113, "time": 0.9526077508926392, "epoch": 3, "memory": 28783, "step": 12616} +{"lr": 9.140576474687264e-05, "data_time": 0.40594820976257323, "grad_norm": 1.6028058171272277, "loss": 0.2145908147096634, "time": 0.9364082336425781, "epoch": 3, "memory": 28783, "step": 12716} +{"lr": 9.140576474687264e-05, "data_time": 0.34771692752838135, "grad_norm": 1.650160163640976, "loss": 0.21283124387264252, "time": 0.8681785583496093, "epoch": 3, "memory": 28783, "step": 12816} +{"lr": 9.140576474687264e-05, "data_time": 0.41398632526397705, "grad_norm": 2.017889451980591, "loss": 0.21516158878803254, "time": 0.928774070739746, "epoch": 3, "memory": 28783, "step": 12916} +{"lr": 9.140576474687264e-05, "data_time": 0.39421517848968507, "grad_norm": 1.2642747581005096, "loss": 0.21430819779634475, "time": 0.9186938285827637, "epoch": 3, "memory": 28783, "step": 13016} +{"lr": 9.140576474687264e-05, "data_time": 0.4297230958938599, "grad_norm": 2.902486139535904, "loss": 0.21576231420040132, "time": 0.9468563795089722, "epoch": 3, "memory": 28783, "step": 13116} +{"lr": 9.140576474687264e-05, "data_time": 0.42093353271484374, "grad_norm": 1.669806843996048, "loss": 0.21110720038414002, "time": 0.9416931629180908, "epoch": 3, "memory": 28783, "step": 13216} +{"lr": 9.140576474687264e-05, "data_time": 0.41459462642669676, "grad_norm": 1.8726254403591156, "loss": 0.2083861544728279, "time": 0.9460616111755371, "epoch": 3, "memory": 28783, "step": 13316} +{"lr": 9.140576474687264e-05, "data_time": 0.36772024631500244, "grad_norm": 2.3280391097068787, "loss": 0.21353710889816285, "time": 0.8850645542144775, "epoch": 3, "memory": 28783, "step": 13416} +{"lr": 9.140576474687264e-05, "data_time": 0.4986832618713379, "grad_norm": 2.1439351558685305, "loss": 0.212885282933712, "time": 1.0227462291717528, "epoch": 3, "memory": 28783, "step": 13516} +{"lr": 9.140576474687264e-05, "data_time": 0.4322330474853516, "grad_norm": 1.7114878594875336, "loss": 0.21144051998853683, "time": 0.9611533403396606, "epoch": 3, "memory": 28783, "step": 13616} +{"lr": 9.140576474687264e-05, "data_time": 0.327402663230896, "grad_norm": 1.5761158168315887, "loss": 0.21120853275060653, "time": 0.8291275024414062, "epoch": 3, "memory": 28783, "step": 13716} +{"lr": 9.140576474687264e-05, "data_time": 0.33883545398712156, "grad_norm": 1.2156826674938201, "loss": 0.21361772418022157, "time": 0.863519811630249, "epoch": 3, "memory": 28783, "step": 13816} +{"lr": 9.140576474687264e-05, "data_time": 0.3207821846008301, "grad_norm": 1.9318087100982666, "loss": 0.21138177514076234, "time": 0.8583831071853638, "epoch": 3, "memory": 28783, "step": 13916} +{"lr": 9.140576474687264e-05, "data_time": 0.4241142749786377, "grad_norm": 2.098145824670792, "loss": 0.21452728807926177, "time": 0.9520506858825684, "epoch": 3, "memory": 28783, "step": 14016} +{"lr": 9.140576474687264e-05, "data_time": 0.43827433586120607, "grad_norm": 1.5923111021518708, "loss": 0.21286115199327468, "time": 0.9630373954772949, "epoch": 3, "memory": 28783, "step": 14116} +{"lr": 9.140576474687264e-05, "data_time": 0.450687050819397, "grad_norm": 1.4236567795276642, "loss": 0.20501015186309815, "time": 1.00359787940979, "epoch": 3, "memory": 28783, "step": 14216} +{"lr": 9.140576474687264e-05, "data_time": 0.41872968673706057, "grad_norm": 1.32694154381752, "loss": 0.2091246575117111, "time": 0.9462849855422973, "epoch": 3, "memory": 28783, "step": 14316} +{"lr": 9.140576474687264e-05, "data_time": 0.2802180528640747, "grad_norm": 1.640733528137207, "loss": 0.20935942977666855, "time": 0.7958567619323731, "epoch": 3, "memory": 28783, "step": 14416} +{"lr": 9.140576474687264e-05, "data_time": 0.3091710090637207, "grad_norm": 1.1378621339797974, "loss": 0.20767311453819276, "time": 0.8174018383026123, "epoch": 3, "memory": 28783, "step": 14516} +{"lr": 9.140576474687264e-05, "data_time": 0.32596402168273925, "grad_norm": 1.2323434114456178, "loss": 0.21011255532503129, "time": 0.8540189743041993, "epoch": 3, "memory": 28783, "step": 14616} +{"lr": 9.140576474687264e-05, "data_time": 0.37062528133392336, "grad_norm": 1.0656420469284058, "loss": 0.21032456159591675, "time": 0.8760164499282836, "epoch": 3, "memory": 28783, "step": 14716} +{"lr": 9.140576474687264e-05, "data_time": 0.37595129013061523, "grad_norm": 1.8807616651058197, "loss": 0.21050831228494643, "time": 0.8942030668258667, "epoch": 3, "memory": 28783, "step": 14816} +{"lr": 9.140576474687264e-05, "data_time": 0.4818445682525635, "grad_norm": 1.8256277203559876, "loss": 0.21127363294363022, "time": 1.016457486152649, "epoch": 3, "memory": 28783, "step": 14916} +{"lr": 9.140576474687264e-05, "data_time": 0.41287271976470946, "grad_norm": 1.2120871603488923, "loss": 0.21269291490316392, "time": 0.9386276006698608, "epoch": 3, "memory": 28783, "step": 15016} +{"lr": 9.140576474687264e-05, "data_time": 0.4050600051879883, "grad_norm": 2.5419181048870088, "loss": 0.21512603759765625, "time": 0.9259554386138916, "epoch": 3, "memory": 28783, "step": 15116} +{"lr": 9.140576474687264e-05, "data_time": 0.51962730884552, "grad_norm": 2.4268842816352842, "loss": 0.21307655572891235, "time": 1.0452313661575316, "epoch": 3, "memory": 28783, "step": 15216} +{"lr": 9.140576474687264e-05, "data_time": 0.42670629024505613, "grad_norm": 1.5698530793190002, "loss": 0.21473020911216736, "time": 0.9553998231887817, "epoch": 3, "memory": 28783, "step": 15316} +{"lr": 9.140576474687264e-05, "data_time": 0.44242675304412843, "grad_norm": 1.3581580519676208, "loss": 0.20926800817251207, "time": 0.9724407911300659, "epoch": 3, "memory": 28783, "step": 15416} +{"lr": 9.140576474687264e-05, "data_time": 0.4571129083633423, "grad_norm": 1.8173506081104278, "loss": 0.21171002835035324, "time": 0.987190580368042, "epoch": 3, "memory": 28783, "step": 15516} +{"lr": 9.140576474687264e-05, "data_time": 0.340023398399353, "grad_norm": 1.3211472660303116, "loss": 0.20899121910333635, "time": 0.8520113468170166, "epoch": 3, "memory": 28783, "step": 15616} +{"lr": 9.140576474687264e-05, "data_time": 0.40095632076263427, "grad_norm": 1.5702925741672515, "loss": 0.2115289032459259, "time": 0.9212728977203369, "epoch": 3, "memory": 28783, "step": 15716} +{"lr": 9.140576474687264e-05, "data_time": 0.22360050678253174, "grad_norm": 1.5953031480312347, "loss": 0.20678865164518356, "time": 0.7282232046127319, "epoch": 3, "memory": 28783, "step": 15816} +{"lr": 9.140576474687264e-05, "data_time": 0.2912003993988037, "grad_norm": 0.9083232283592224, "loss": 0.20775903910398483, "time": 0.8021667957305908, "epoch": 3, "memory": 28783, "step": 15916} +{"lr": 9.140576474687264e-05, "data_time": 0.3436312198638916, "grad_norm": 1.3005921363830566, "loss": 0.20833759009838104, "time": 0.8808267831802368, "epoch": 3, "memory": 28783, "step": 16016} +{"lr": 9.140576474687264e-05, "data_time": 0.30919647216796875, "grad_norm": 1.743823230266571, "loss": 0.21149243265390397, "time": 0.8162663221359253, "epoch": 3, "memory": 28783, "step": 16116} +{"lr": 9.140576474687264e-05, "data_time": 0.26744005680084226, "grad_norm": 1.4218769907951354, "loss": 0.20726995170116425, "time": 0.8309375762939453, "epoch": 3, "memory": 28783, "step": 16216} +{"lr": 9.140576474687264e-05, "data_time": 0.1598140001296997, "grad_norm": 1.2673326671123504, "loss": 0.209407040476799, "time": 0.7453443527221679, "epoch": 3, "memory": 28783, "step": 16316} +{"lr": 9.140576474687264e-05, "data_time": 0.08368620872497559, "grad_norm": 1.2609032690525055, "loss": 0.21325397491455078, "time": 0.8645741462707519, "epoch": 3, "memory": 28783, "step": 16416} +{"lr": 9.140576474687264e-05, "data_time": 0.012835836410522461, "grad_norm": 1.088646912574768, "loss": 0.2061121016740799, "time": 0.7909500122070312, "epoch": 3, "memory": 28783, "step": 16516} +{"lr": 9.140576474687264e-05, "data_time": 0.0017733335494995116, "grad_norm": 0.8696327984333039, "loss": 0.20412105768918992, "time": 0.7758328199386597, "epoch": 3, "memory": 28783, "step": 16616} +{"lr": 9.140576474687264e-05, "data_time": 0.0023193597793579102, "grad_norm": 1.2496171474456788, "loss": 0.20469192266464234, "time": 0.9271472215652465, "epoch": 3, "memory": 28783, "step": 16716} +{"lr": 9.140576474687264e-05, "data_time": 0.002138352394104004, "grad_norm": 1.030261254310608, "loss": 0.20918433517217636, "time": 0.8781866788864136, "epoch": 3, "memory": 28783, "step": 16816} +{"lr": 9.140576474687264e-05, "data_time": 0.0014387845993041992, "grad_norm": 2.07956263422966, "loss": 0.21559395939111708, "time": 0.7746203184127808, "epoch": 3, "memory": 28783, "step": 16916} +{"lr": 9.140576474687264e-05, "data_time": 0.0018655776977539063, "grad_norm": 1.2639388561248779, "loss": 0.20937535911798477, "time": 0.7693906307220459, "epoch": 3, "memory": 28783, "step": 17016} +{"lr": 9.140576474687264e-05, "data_time": 0.002580690383911133, "grad_norm": 1.0399180829524994, "loss": 0.2064996540546417, "time": 0.9988693237304688, "epoch": 3, "memory": 28783, "step": 17116} +{"lr": 9.140576474687264e-05, "data_time": 0.002512550354003906, "grad_norm": 1.0688959777355194, "loss": 0.21010167598724366, "time": 0.8722483396530152, "epoch": 3, "memory": 28783, "step": 17216} +{"accuracy/top1": 99.83586883544922, "data_time": 0.2442783327663646, "time": 0.6610640357522404, "step": 3} +{"lr": 8.14503363531613e-05, "data_time": 0.15842914581298828, "grad_norm": 0.9486164212226867, "loss": 0.20616671591997146, "time": 0.9271233320236206, "epoch": 4, "memory": 28783, "step": 17374} +{"lr": 8.14503363531613e-05, "data_time": 0.1748908281326294, "grad_norm": 1.2735835433006286, "loss": 0.2055667370557785, "time": 0.923888874053955, "epoch": 4, "memory": 28783, "step": 17474} +{"lr": 8.14503363531613e-05, "data_time": 0.09691615104675293, "grad_norm": 0.9428302496671677, "loss": 0.2072424590587616, "time": 0.8132968664169311, "epoch": 4, "memory": 28783, "step": 17574} +{"lr": 8.14503363531613e-05, "data_time": 0.001745295524597168, "grad_norm": 0.7736806273460388, "loss": 0.20651400983333587, "time": 0.8681283950805664, "epoch": 4, "memory": 28783, "step": 17674} +{"lr": 8.14503363531613e-05, "data_time": 0.002722954750061035, "grad_norm": 1.2201237201690673, "loss": 0.20861655175685884, "time": 0.9600642204284668, "epoch": 4, "memory": 28783, "step": 17774} +{"lr": 8.14503363531613e-05, "data_time": 0.011842823028564453, "grad_norm": 1.3274381279945373, "loss": 0.2047136589884758, "time": 0.90377197265625, "epoch": 4, "memory": 28783, "step": 17874} +{"lr": 8.14503363531613e-05, "data_time": 0.0020856142044067385, "grad_norm": 0.8235398411750794, "loss": 0.20607851147651673, "time": 0.871361780166626, "epoch": 4, "memory": 28783, "step": 17974} +{"lr": 8.14503363531613e-05, "data_time": 0.002403712272644043, "grad_norm": 1.2387137442827225, "loss": 0.20855880081653594, "time": 0.9287378549575805, "epoch": 4, "memory": 28783, "step": 18074} +{"lr": 8.14503363531613e-05, "data_time": 0.0015957117080688476, "grad_norm": 0.9570003181695939, "loss": 0.20592181980609894, "time": 0.7168264150619507, "epoch": 4, "memory": 28783, "step": 18174} +{"lr": 8.14503363531613e-05, "data_time": 0.0016207933425903321, "grad_norm": 1.2040386378765107, "loss": 0.20649221688508987, "time": 0.8188527584075928, "epoch": 4, "memory": 28783, "step": 18274} +{"lr": 8.14503363531613e-05, "data_time": 0.00239717960357666, "grad_norm": 0.8021222949028015, "loss": 0.2065666452050209, "time": 0.9109915256500244, "epoch": 4, "memory": 28783, "step": 18374} +{"lr": 8.14503363531613e-05, "data_time": 0.002166438102722168, "grad_norm": 1.0723474770784378, "loss": 0.2106533721089363, "time": 0.7592983245849609, "epoch": 4, "memory": 28783, "step": 18474} +{"lr": 8.14503363531613e-05, "data_time": 0.0025032520294189452, "grad_norm": 1.1991454631090164, "loss": 0.20588413774967193, "time": 0.9460506200790405, "epoch": 4, "memory": 28783, "step": 18574} +{"lr": 8.14503363531613e-05, "data_time": 0.0029126644134521485, "grad_norm": 0.6547329306602478, "loss": 0.20953926295042039, "time": 0.7588682651519776, "epoch": 4, "memory": 28783, "step": 18674} +{"lr": 8.14503363531613e-05, "data_time": 0.003932929039001465, "grad_norm": 0.9340867400169373, "loss": 0.20635195970535278, "time": 0.9276320695877075, "epoch": 4, "memory": 28783, "step": 18774} +{"lr": 8.14503363531613e-05, "data_time": 0.0017466306686401366, "grad_norm": 1.3939781427383422, "loss": 0.2075723096728325, "time": 0.7641024351119995, "epoch": 4, "memory": 28783, "step": 18874} +{"lr": 8.14503363531613e-05, "data_time": 0.0025475263595581056, "grad_norm": 1.449819341301918, "loss": 0.20281902849674224, "time": 0.8952036142349243, "epoch": 4, "memory": 28783, "step": 18974} +{"lr": 8.14503363531613e-05, "data_time": 0.002364826202392578, "grad_norm": 1.2631963044404984, "loss": 0.2108501061797142, "time": 0.8860666990280152, "epoch": 4, "memory": 28783, "step": 19074} +{"lr": 8.14503363531613e-05, "data_time": 0.001476883888244629, "grad_norm": 0.8616365134716034, "loss": 0.20916038304567336, "time": 0.7153547048568726, "epoch": 4, "memory": 28783, "step": 19174} +{"lr": 8.14503363531613e-05, "data_time": 0.0018721818923950195, "grad_norm": 0.7398032575845719, "loss": 0.2032613754272461, "time": 0.8166702270507813, "epoch": 4, "memory": 28783, "step": 19274} +{"lr": 8.14503363531613e-05, "data_time": 0.0017771482467651366, "grad_norm": 0.7306557953357696, "loss": 0.2064534217119217, "time": 0.7739821910858155, "epoch": 4, "memory": 28783, "step": 19374} +{"lr": 8.14503363531613e-05, "data_time": 0.0023369073867797853, "grad_norm": 0.8454796463251114, "loss": 0.2049805149435997, "time": 0.8626639366149902, "epoch": 4, "memory": 28783, "step": 19474} +{"lr": 8.14503363531613e-05, "data_time": 0.0016691923141479493, "grad_norm": 1.1940773099660873, "loss": 0.20755469053983688, "time": 0.8436608076095581, "epoch": 4, "memory": 28783, "step": 19574} +{"lr": 8.14503363531613e-05, "data_time": 0.0051238775253295895, "grad_norm": 0.8481484055519104, "loss": 0.2040230005979538, "time": 0.9070647954940796, "epoch": 4, "memory": 28783, "step": 19674} +{"lr": 8.14503363531613e-05, "data_time": 0.0016668319702148437, "grad_norm": 1.4306426346302032, "loss": 0.21022323817014693, "time": 0.8055631160736084, "epoch": 4, "memory": 28783, "step": 19774} +{"lr": 8.14503363531613e-05, "data_time": 0.0020183324813842773, "grad_norm": 0.8969280123710632, "loss": 0.20493603646755218, "time": 0.8274871349334717, "epoch": 4, "memory": 28783, "step": 19874} +{"lr": 8.14503363531613e-05, "data_time": 0.0035917997360229493, "grad_norm": 0.9320100218057632, "loss": 0.20461425185203552, "time": 0.9778493881225586, "epoch": 4, "memory": 28783, "step": 19974} +{"lr": 8.14503363531613e-05, "data_time": 0.002080059051513672, "grad_norm": 0.9420254468917847, "loss": 0.2047571897506714, "time": 0.8735392570495606, "epoch": 4, "memory": 28783, "step": 20074} +{"lr": 8.14503363531613e-05, "data_time": 0.0018732070922851563, "grad_norm": 0.8712501719594001, "loss": 0.20343103408813476, "time": 0.8793980598449707, "epoch": 4, "memory": 28783, "step": 20174} +{"lr": 8.14503363531613e-05, "data_time": 0.003032684326171875, "grad_norm": 1.051969164609909, "loss": 0.2065539091825485, "time": 0.9040723085403443, "epoch": 4, "memory": 28783, "step": 20274} +{"lr": 8.14503363531613e-05, "data_time": 0.001804351806640625, "grad_norm": 0.7072114735841751, "loss": 0.20508118271827697, "time": 0.8119118213653564, "epoch": 4, "memory": 28783, "step": 20374} +{"lr": 8.14503363531613e-05, "data_time": 0.0019825220108032225, "grad_norm": 0.7490186542272568, "loss": 0.2030818998813629, "time": 0.8202088832855224, "epoch": 4, "memory": 28783, "step": 20474} +{"lr": 8.14503363531613e-05, "data_time": 0.0017294883728027344, "grad_norm": 1.0776535362005233, "loss": 0.2084824562072754, "time": 0.7922296524047852, "epoch": 4, "memory": 28783, "step": 20574} +{"lr": 8.14503363531613e-05, "data_time": 0.0021042823791503906, "grad_norm": 1.1144632935523986, "loss": 0.20546067804098128, "time": 0.8531909227371216, "epoch": 4, "memory": 28783, "step": 20674} +{"lr": 8.14503363531613e-05, "data_time": 0.001963210105895996, "grad_norm": 0.8492861807346344, "loss": 0.20431856513023378, "time": 0.8022182941436767, "epoch": 4, "memory": 28783, "step": 20774} +{"lr": 8.14503363531613e-05, "data_time": 0.0023144006729125975, "grad_norm": 0.8482524693012238, "loss": 0.20370493680238724, "time": 0.8173549175262451, "epoch": 4, "memory": 28783, "step": 20874} +{"lr": 8.14503363531613e-05, "data_time": 0.0019672155380249024, "grad_norm": 0.7161398589611053, "loss": 0.2066895216703415, "time": 0.8143734931945801, "epoch": 4, "memory": 28783, "step": 20974} +{"lr": 8.14503363531613e-05, "data_time": 0.0025456428527832033, "grad_norm": 0.9564333558082581, "loss": 0.20615943223237992, "time": 0.8935346364974975, "epoch": 4, "memory": 28783, "step": 21074} +{"lr": 8.14503363531613e-05, "data_time": 0.0018461227416992187, "grad_norm": 0.8982215970754623, "loss": 0.20722531378269196, "time": 0.812614107131958, "epoch": 4, "memory": 28783, "step": 21174} +{"lr": 8.14503363531613e-05, "data_time": 0.0016650676727294922, "grad_norm": 0.7342030823230743, "loss": 0.20762034952640535, "time": 0.8359655380249024, "epoch": 4, "memory": 28783, "step": 21274} +{"lr": 8.14503363531613e-05, "data_time": 0.0023229837417602537, "grad_norm": 0.9481951981782913, "loss": 0.20643476843833924, "time": 0.9659729719161987, "epoch": 4, "memory": 28783, "step": 21374} +{"lr": 8.14503363531613e-05, "data_time": 0.0034432411193847656, "grad_norm": 0.8121060967445374, "loss": 0.2057229295372963, "time": 0.9556218147277832, "epoch": 4, "memory": 28783, "step": 21474} +{"lr": 8.14503363531613e-05, "data_time": 0.0018439769744873046, "grad_norm": 0.8101279348134994, "loss": 0.20679338723421098, "time": 0.7833482980728149, "epoch": 4, "memory": 28783, "step": 21574} +{"lr": 8.14503363531613e-05, "data_time": 0.002988457679748535, "grad_norm": 1.1345936894416808, "loss": 0.20748918503522873, "time": 0.7720442771911621, "epoch": 4, "memory": 28783, "step": 21674} +{"lr": 8.14503363531613e-05, "data_time": 0.0017228126525878906, "grad_norm": 0.6462352246046066, "loss": 0.20321632772684098, "time": 0.7660947799682617, "epoch": 4, "memory": 28783, "step": 21774} +{"lr": 8.14503363531613e-05, "data_time": 0.0017563581466674804, "grad_norm": 0.6697968736290931, "loss": 0.2014237642288208, "time": 0.7498160600662231, "epoch": 4, "memory": 28783, "step": 21874} +{"lr": 8.14503363531613e-05, "data_time": 0.002251148223876953, "grad_norm": 1.2094328701496124, "loss": 0.2062023714184761, "time": 0.8809028148651123, "epoch": 4, "memory": 28783, "step": 21974} +{"lr": 8.14503363531613e-05, "data_time": 0.002000856399536133, "grad_norm": 0.5823024600744248, "loss": 0.20479606240987777, "time": 0.7701223611831665, "epoch": 4, "memory": 28783, "step": 22074} +{"lr": 8.14503363531613e-05, "data_time": 0.0024200439453125, "grad_norm": 0.9492810845375061, "loss": 0.20501076728105544, "time": 0.9245847225189209, "epoch": 4, "memory": 28783, "step": 22174} +{"lr": 8.14503363531613e-05, "data_time": 0.0019545555114746094, "grad_norm": 0.9327262550592422, "loss": 0.20240710377693177, "time": 0.8140376806259155, "epoch": 4, "memory": 28783, "step": 22274} +{"lr": 8.14503363531613e-05, "data_time": 0.002622032165527344, "grad_norm": 1.0285163223743439, "loss": 0.20348014086484909, "time": 0.9900183200836181, "epoch": 4, "memory": 28783, "step": 22374} +{"lr": 8.14503363531613e-05, "data_time": 0.002169895172119141, "grad_norm": 0.8105373054742813, "loss": 0.20507268905639647, "time": 0.8354963302612305, "epoch": 4, "memory": 28783, "step": 22474} +{"lr": 8.14503363531613e-05, "data_time": 0.0021898508071899413, "grad_norm": 0.7004149116575717, "loss": 0.2050428107380867, "time": 0.9091761350631714, "epoch": 4, "memory": 28783, "step": 22574} +{"lr": 8.14503363531613e-05, "data_time": 0.0016341447830200196, "grad_norm": 0.8098372787237167, "loss": 0.2061666041612625, "time": 0.8773354530334473, "epoch": 4, "memory": 28783, "step": 22674} +{"lr": 8.14503363531613e-05, "data_time": 0.0023780584335327147, "grad_norm": 0.7755119383335114, "loss": 0.20429756194353105, "time": 0.8351455688476562, "epoch": 4, "memory": 28783, "step": 22774} +{"lr": 8.14503363531613e-05, "data_time": 0.0016416788101196289, "grad_norm": 1.7323849946260452, "loss": 0.20490384846925735, "time": 0.8411281108856201, "epoch": 4, "memory": 28783, "step": 22874} +{"lr": 8.14503363531613e-05, "data_time": 0.0018893718719482423, "grad_norm": 1.1103879570961, "loss": 0.20811793357133865, "time": 0.81603102684021, "epoch": 4, "memory": 28783, "step": 22974} +{"accuracy/top1": 99.89899444580078, "data_time": 0.22424493116490982, "time": 0.6452209388508516, "step": 4} +{"lr": 6.890576474687264e-05, "data_time": 0.33192808628082277, "grad_norm": 0.5818910449743271, "loss": 0.20459140837192535, "time": 0.8387040615081787, "epoch": 5, "memory": 28783, "step": 23132} +{"lr": 6.890576474687264e-05, "data_time": 0.2808361053466797, "grad_norm": 0.7127429842948914, "loss": 0.20435028672218322, "time": 0.7941132545471191, "epoch": 5, "memory": 28783, "step": 23232} +{"lr": 6.890576474687264e-05, "data_time": 0.0026577234268188475, "grad_norm": 0.9764065265655517, "loss": 0.20468062907457352, "time": 0.8684720754623413, "epoch": 5, "memory": 28783, "step": 23332} +{"lr": 6.890576474687264e-05, "data_time": 0.0019591331481933595, "grad_norm": 0.7020221054553986, "loss": 0.20276259034872054, "time": 0.8258968353271484, "epoch": 5, "memory": 28783, "step": 23432} +{"lr": 6.890576474687264e-05, "data_time": 0.002506256103515625, "grad_norm": 0.7867343083024025, "loss": 0.20279908627271653, "time": 0.8995970010757446, "epoch": 5, "memory": 28783, "step": 23532} +{"lr": 6.890576474687264e-05, "data_time": 0.0021756649017333984, "grad_norm": 0.6544739693403244, "loss": 0.20398417264223098, "time": 0.9003485679626465, "epoch": 5, "memory": 28783, "step": 23632} +{"lr": 6.890576474687264e-05, "data_time": 0.0031285762786865236, "grad_norm": 0.730390340089798, "loss": 0.20151670575141906, "time": 0.882508373260498, "epoch": 5, "memory": 28783, "step": 23732} +{"lr": 6.890576474687264e-05, "data_time": 0.0019951343536376955, "grad_norm": 0.7446844309568406, "loss": 0.20467254519462585, "time": 0.7451502799987793, "epoch": 5, "memory": 28783, "step": 23832} +{"lr": 6.890576474687264e-05, "data_time": 0.001630878448486328, "grad_norm": 1.0319697707891464, "loss": 0.20675748139619826, "time": 0.8372640371322632, "epoch": 5, "memory": 28783, "step": 23932} +{"lr": 6.890576474687264e-05, "data_time": 0.0022076845169067385, "grad_norm": 1.007606989145279, "loss": 0.20434479117393495, "time": 0.918543791770935, "epoch": 5, "memory": 28783, "step": 24032} +{"lr": 6.890576474687264e-05, "data_time": 0.0021894216537475587, "grad_norm": 0.528626212477684, "loss": 0.20331996381282808, "time": 0.8008938550949096, "epoch": 5, "memory": 28783, "step": 24132} +{"lr": 6.890576474687264e-05, "data_time": 0.003458523750305176, "grad_norm": 0.6689302936196327, "loss": 0.2045394226908684, "time": 0.8632954359054565, "epoch": 5, "memory": 28783, "step": 24232} +{"lr": 6.890576474687264e-05, "data_time": 0.0016916513442993164, "grad_norm": 0.7286306858062744, "loss": 0.20205857902765273, "time": 0.7412585496902466, "epoch": 5, "memory": 28783, "step": 24332} +{"lr": 6.890576474687264e-05, "data_time": 0.001738905906677246, "grad_norm": 0.685021185874939, "loss": 0.20819469392299653, "time": 0.8424660682678222, "epoch": 5, "memory": 28783, "step": 24432} +{"lr": 6.890576474687264e-05, "data_time": 0.002035975456237793, "grad_norm": 0.6094655677676201, "loss": 0.20216030329465867, "time": 0.8036423206329346, "epoch": 5, "memory": 28783, "step": 24532} +{"lr": 6.890576474687264e-05, "data_time": 0.0019531965255737303, "grad_norm": 1.5473129957914353, "loss": 0.20610381066799163, "time": 0.7969301223754883, "epoch": 5, "memory": 28783, "step": 24632} +{"lr": 6.890576474687264e-05, "data_time": 0.003784966468811035, "grad_norm": 0.42079600393772126, "loss": 0.20122822225093842, "time": 0.941700553894043, "epoch": 5, "memory": 28783, "step": 24732} +{"lr": 6.890576474687264e-05, "data_time": 0.0020723342895507812, "grad_norm": 0.7694544076919556, "loss": 0.20376111567020416, "time": 0.8561123132705688, "epoch": 5, "memory": 28783, "step": 24832} +{"lr": 6.890576474687264e-05, "data_time": 0.001956009864807129, "grad_norm": 0.5741630673408509, "loss": 0.20102161020040513, "time": 0.9322062730789185, "epoch": 5, "memory": 28783, "step": 24932} +{"lr": 6.890576474687264e-05, "data_time": 0.002545619010925293, "grad_norm": 0.7578251883387566, "loss": 0.20165254771709443, "time": 0.9280125141143799, "epoch": 5, "memory": 28783, "step": 25032} +{"lr": 6.890576474687264e-05, "data_time": 0.0021206140518188477, "grad_norm": 0.719692063331604, "loss": 0.20198911279439927, "time": 0.7953967094421387, "epoch": 5, "memory": 28783, "step": 25132} +{"lr": 6.890576474687264e-05, "data_time": 0.0016574859619140625, "grad_norm": 0.4823285847902298, "loss": 0.20262670367956162, "time": 0.873339319229126, "epoch": 5, "memory": 28783, "step": 25232} +{"lr": 6.890576474687264e-05, "data_time": 0.002013969421386719, "grad_norm": 0.9548934191465378, "loss": 0.2040390118956566, "time": 0.8383258819580078, "epoch": 5, "memory": 28783, "step": 25332} +{"lr": 6.890576474687264e-05, "data_time": 0.0024251699447631835, "grad_norm": 0.8258676022291184, "loss": 0.20598920434713364, "time": 0.9125054121017456, "epoch": 5, "memory": 28783, "step": 25432} +{"lr": 6.890576474687264e-05, "data_time": 0.0018953800201416016, "grad_norm": 0.5936780512332916, "loss": 0.2034745842218399, "time": 0.8699544191360473, "epoch": 5, "memory": 28783, "step": 25532} +{"lr": 6.890576474687264e-05, "data_time": 0.001897907257080078, "grad_norm": 0.7664115726947784, "loss": 0.2033932939171791, "time": 0.7872421264648437, "epoch": 5, "memory": 28783, "step": 25632} +{"lr": 6.890576474687264e-05, "data_time": 0.002547621726989746, "grad_norm": 0.9103539705276489, "loss": 0.20057430416345595, "time": 0.9325079202651978, "epoch": 5, "memory": 28783, "step": 25732} +{"lr": 6.890576474687264e-05, "data_time": 0.0018838405609130859, "grad_norm": 0.6789240419864655, "loss": 0.2045421615242958, "time": 0.8318154335021972, "epoch": 5, "memory": 28783, "step": 25832} +{"lr": 6.890576474687264e-05, "data_time": 0.001972484588623047, "grad_norm": 0.8590569019317627, "loss": 0.20210847407579421, "time": 0.824900484085083, "epoch": 5, "memory": 28783, "step": 25932} +{"lr": 6.890576474687264e-05, "data_time": 0.17052371501922609, "grad_norm": 0.7384711638092994, "loss": 0.20500597208738328, "time": 0.9175719499588013, "epoch": 5, "memory": 28783, "step": 26032} +{"lr": 6.890576474687264e-05, "data_time": 0.1663076639175415, "grad_norm": 0.820941099524498, "loss": 0.20676555782556533, "time": 0.8747415781021118, "epoch": 5, "memory": 28783, "step": 26132} +{"lr": 6.890576474687264e-05, "data_time": 0.003639030456542969, "grad_norm": 0.7995735704898834, "loss": 0.2040101855993271, "time": 0.9614399194717407, "epoch": 5, "memory": 28783, "step": 26232} +{"lr": 6.890576474687264e-05, "data_time": 0.003133678436279297, "grad_norm": 0.8042361319065094, "loss": 0.20054792016744613, "time": 0.9424630403518677, "epoch": 5, "memory": 28783, "step": 26332} +{"lr": 6.890576474687264e-05, "data_time": 0.0017218828201293946, "grad_norm": 0.5530852839350701, "loss": 0.2040867730975151, "time": 0.8652605533599853, "epoch": 5, "memory": 28783, "step": 26432} +{"lr": 6.890576474687264e-05, "data_time": 0.002207183837890625, "grad_norm": 0.5613486021757126, "loss": 0.20223472714424134, "time": 0.8249388933181763, "epoch": 5, "memory": 28783, "step": 26532} +{"lr": 6.890576474687264e-05, "data_time": 0.0016343832015991212, "grad_norm": 0.8797535479068757, "loss": 0.20342277735471725, "time": 0.7633192062377929, "epoch": 5, "memory": 28783, "step": 26632} +{"lr": 6.890576474687264e-05, "data_time": 0.003683805465698242, "grad_norm": 1.0159285381436347, "loss": 0.2023201122879982, "time": 0.8931429386138916, "epoch": 5, "memory": 28783, "step": 26732} +{"lr": 6.890576474687264e-05, "data_time": 0.0019260406494140624, "grad_norm": 0.6106607258319855, "loss": 0.20391769409179689, "time": 0.790984058380127, "epoch": 5, "memory": 28783, "step": 26832} +{"lr": 6.890576474687264e-05, "data_time": 0.002742624282836914, "grad_norm": 0.839930745959282, "loss": 0.20187632143497466, "time": 0.9179551362991333, "epoch": 5, "memory": 28783, "step": 26932} +{"lr": 6.890576474687264e-05, "data_time": 0.0017934083938598634, "grad_norm": 0.6112815082073212, "loss": 0.2032255232334137, "time": 0.8257436990737915, "epoch": 5, "memory": 28783, "step": 27032} +{"lr": 6.890576474687264e-05, "data_time": 0.00274960994720459, "grad_norm": 0.6874577879905701, "loss": 0.20240055918693542, "time": 0.8879693746566772, "epoch": 5, "memory": 28783, "step": 27132} +{"lr": 6.890576474687264e-05, "data_time": 0.0017387866973876953, "grad_norm": 0.6869521290063858, "loss": 0.20243849903345107, "time": 0.880066180229187, "epoch": 5, "memory": 28783, "step": 27232} +{"lr": 6.890576474687264e-05, "data_time": 0.0035073041915893556, "grad_norm": 0.6701879560947418, "loss": 0.2014155089855194, "time": 0.8371518850326538, "epoch": 5, "memory": 28783, "step": 27332} +{"lr": 6.890576474687264e-05, "data_time": 0.0028847694396972657, "grad_norm": 0.5011413045227527, "loss": 0.20108413994312285, "time": 0.9502418518066407, "epoch": 5, "memory": 28783, "step": 27432} +{"lr": 6.890576474687264e-05, "data_time": 0.0019443988800048827, "grad_norm": 0.9512621909379959, "loss": 0.20810155868530272, "time": 0.8282215118408203, "epoch": 5, "memory": 28783, "step": 27532} +{"lr": 6.890576474687264e-05, "data_time": 0.001908278465270996, "grad_norm": 0.7863615319132805, "loss": 0.20385856926441193, "time": 0.809152626991272, "epoch": 5, "memory": 28783, "step": 27632} +{"lr": 6.890576474687264e-05, "data_time": 0.0021807193756103516, "grad_norm": 1.2226685851812362, "loss": 0.20267819464206696, "time": 0.8482498407363892, "epoch": 5, "memory": 28783, "step": 27732} +{"lr": 6.890576474687264e-05, "data_time": 0.0021015167236328124, "grad_norm": 0.5874336376786232, "loss": 0.2017254501581192, "time": 0.8229806900024415, "epoch": 5, "memory": 28783, "step": 27832} +{"lr": 6.890576474687264e-05, "data_time": 0.001706385612487793, "grad_norm": 0.39259429574012755, "loss": 0.2026368945837021, "time": 0.8157365322113037, "epoch": 5, "memory": 28783, "step": 27932} +{"lr": 6.890576474687264e-05, "data_time": 0.002717757225036621, "grad_norm": 0.6538422375917434, "loss": 0.20286358445882796, "time": 0.9589800834655762, "epoch": 5, "memory": 28783, "step": 28032} +{"lr": 6.890576474687264e-05, "data_time": 0.0013824224472045899, "grad_norm": 0.9113956540822983, "loss": 0.20145105719566345, "time": 0.7455509185791016, "epoch": 5, "memory": 28783, "step": 28132} +{"lr": 6.890576474687264e-05, "data_time": 0.0017171144485473634, "grad_norm": 0.927887174487114, "loss": 0.20362524390220643, "time": 0.8601174592971802, "epoch": 5, "memory": 28783, "step": 28232} +{"lr": 6.890576474687264e-05, "data_time": 0.0018078804016113282, "grad_norm": 0.5490619979798794, "loss": 0.20178521126508714, "time": 0.9091894626617432, "epoch": 5, "memory": 28783, "step": 28332} +{"lr": 6.890576474687264e-05, "data_time": 0.0015626668930053711, "grad_norm": 0.6157525017857551, "loss": 0.2025834009051323, "time": 0.8718742847442627, "epoch": 5, "memory": 28783, "step": 28432} +{"lr": 6.890576474687264e-05, "data_time": 0.0019608259201049803, "grad_norm": 0.6289623349905014, "loss": 0.2043049991130829, "time": 0.8093969821929932, "epoch": 5, "memory": 28783, "step": 28532} +{"lr": 6.890576474687264e-05, "data_time": 0.0024734258651733397, "grad_norm": 0.5192620635032654, "loss": 0.20114141702651978, "time": 0.8368217945098877, "epoch": 5, "memory": 28783, "step": 28632} +{"lr": 6.890576474687264e-05, "data_time": 0.001771402359008789, "grad_norm": 0.8042371943593025, "loss": 0.20445415675640105, "time": 0.801706600189209, "epoch": 5, "memory": 28783, "step": 28732} +{"accuracy/top1": 99.79167938232422, "data_time": 0.24976594307843378, "time": 0.6703277756186092, "step": 5} +{"lr": 5.500000000000001e-05, "data_time": 0.2367393732070923, "grad_norm": 0.6578721165657043, "loss": 0.20373403280973434, "time": 0.7802339553833008, "epoch": 6, "memory": 28783, "step": 28890} +{"lr": 5.500000000000001e-05, "data_time": 0.23415262699127198, "grad_norm": 0.42493289709091187, "loss": 0.2005619525909424, "time": 0.7920568466186524, "epoch": 6, "memory": 28783, "step": 28990} +{"lr": 5.500000000000001e-05, "data_time": 0.15327730178833007, "grad_norm": 0.5292289935052394, "loss": 0.2023264840245247, "time": 0.8249850749969483, "epoch": 6, "memory": 28783, "step": 29090} +{"lr": 5.500000000000001e-05, "data_time": 0.34800536632537843, "grad_norm": 0.4441455349326134, "loss": 0.20092955380678176, "time": 0.8863354921340942, "epoch": 6, "memory": 28783, "step": 29190} +{"lr": 5.500000000000001e-05, "data_time": 0.35909485816955566, "grad_norm": 0.5437352448701859, "loss": 0.20161206722259523, "time": 0.8636725187301636, "epoch": 6, "memory": 28783, "step": 29290} +{"lr": 5.500000000000001e-05, "data_time": 0.35016751289367676, "grad_norm": 0.6724790245294571, "loss": 0.2017546609044075, "time": 0.865167212486267, "epoch": 6, "memory": 28783, "step": 29390} +{"lr": 5.500000000000001e-05, "data_time": 0.23414967060089112, "grad_norm": 0.6596096843481064, "loss": 0.20320844203233718, "time": 0.7571777105331421, "epoch": 6, "memory": 28783, "step": 29490} +{"lr": 5.500000000000001e-05, "data_time": 0.22813107967376708, "grad_norm": 0.78020164296031, "loss": 0.20676292479038239, "time": 0.7345282554626464, "epoch": 6, "memory": 28783, "step": 29590} +{"lr": 5.500000000000001e-05, "data_time": 0.19055211544036865, "grad_norm": 0.4643255159258842, "loss": 0.20193393528461456, "time": 0.7099626779556274, "epoch": 6, "memory": 28783, "step": 29690} +{"lr": 5.500000000000001e-05, "data_time": 0.32185938358306887, "grad_norm": 0.799211610853672, "loss": 0.20188792794942856, "time": 0.891472315788269, "epoch": 6, "memory": 28783, "step": 29790} +{"lr": 5.500000000000001e-05, "data_time": 0.1744075059890747, "grad_norm": 0.6371877282857895, "loss": 0.2015355110168457, "time": 0.7960968255996704, "epoch": 6, "memory": 28783, "step": 29890} +{"lr": 5.500000000000001e-05, "data_time": 0.08508853912353516, "grad_norm": 0.46341052651405334, "loss": 0.2030060276389122, "time": 0.7870536088943482, "epoch": 6, "memory": 28783, "step": 29990} +{"lr": 5.500000000000001e-05, "data_time": 0.0024773120880126954, "grad_norm": 0.5534893929958343, "loss": 0.20161805301904678, "time": 0.9774402856826783, "epoch": 6, "memory": 28783, "step": 30090} +{"lr": 5.500000000000001e-05, "data_time": 0.0022177696228027344, "grad_norm": 0.6899497509002686, "loss": 0.2014864131808281, "time": 0.767720103263855, "epoch": 6, "memory": 28783, "step": 30190} +{"lr": 5.500000000000001e-05, "data_time": 0.001803112030029297, "grad_norm": 0.716052895784378, "loss": 0.20321467220783235, "time": 0.7384618759155274, "epoch": 6, "memory": 28783, "step": 30290} +{"lr": 5.500000000000001e-05, "data_time": 0.0020949602127075194, "grad_norm": 0.5422618538141251, "loss": 0.199382983148098, "time": 0.8557018995285034, "epoch": 6, "memory": 28783, "step": 30390} +{"lr": 5.500000000000001e-05, "data_time": 0.001886296272277832, "grad_norm": 0.4186623305082321, "loss": 0.2010822594165802, "time": 0.7190708875656128, "epoch": 6, "memory": 28783, "step": 30490} +{"lr": 5.500000000000001e-05, "data_time": 0.0019061803817749024, "grad_norm": 0.5010376758873463, "loss": 0.2028880938887596, "time": 0.8347115278244018, "epoch": 6, "memory": 28783, "step": 30590} +{"lr": 5.500000000000001e-05, "data_time": 0.005642509460449219, "grad_norm": 0.5715576559305191, "loss": 0.2018888548016548, "time": 0.9515722990036011, "epoch": 6, "memory": 28783, "step": 30690} +{"lr": 5.500000000000001e-05, "data_time": 0.001876521110534668, "grad_norm": 0.43848167508840563, "loss": 0.19990312159061432, "time": 0.8387673377990723, "epoch": 6, "memory": 28783, "step": 30790} +{"lr": 5.500000000000001e-05, "data_time": 0.0026645898818969727, "grad_norm": 0.4061574816703796, "loss": 0.20202040821313857, "time": 0.8487327337265015, "epoch": 6, "memory": 28783, "step": 30890} +{"lr": 5.500000000000001e-05, "data_time": 0.0021139144897460937, "grad_norm": 0.6403361782431602, "loss": 0.20125248432159423, "time": 0.7736462593078614, "epoch": 6, "memory": 28783, "step": 30990} +{"lr": 5.500000000000001e-05, "data_time": 0.0024642467498779295, "grad_norm": 0.6221163600683213, "loss": 0.203619085252285, "time": 0.8640312194824219, "epoch": 6, "memory": 28783, "step": 31090} +{"lr": 5.500000000000001e-05, "data_time": 0.0023496866226196287, "grad_norm": 0.7111595347523689, "loss": 0.20276792645454406, "time": 0.8602776765823364, "epoch": 6, "memory": 28783, "step": 31190} +{"lr": 5.500000000000001e-05, "data_time": 0.003142380714416504, "grad_norm": 0.603440237045288, "loss": 0.20502182692289353, "time": 1.0008724212646485, "epoch": 6, "memory": 28783, "step": 31290} +{"lr": 5.500000000000001e-05, "data_time": 0.0017239809036254882, "grad_norm": 0.5461910426616668, "loss": 0.20207417011260986, "time": 0.7815804958343506, "epoch": 6, "memory": 28783, "step": 31390} +{"lr": 5.500000000000001e-05, "data_time": 0.0016937017440795898, "grad_norm": 0.483082864433527, "loss": 0.20152270644903184, "time": 0.8269901514053345, "epoch": 6, "memory": 28783, "step": 31490} +{"lr": 5.500000000000001e-05, "data_time": 0.002161860466003418, "grad_norm": 0.5504985824227333, "loss": 0.20051645040512084, "time": 0.8348847389221191, "epoch": 6, "memory": 28783, "step": 31590} +{"lr": 5.500000000000001e-05, "data_time": 0.0017740726470947266, "grad_norm": 0.6400432795286178, "loss": 0.20143334418535233, "time": 0.8210888147354126, "epoch": 6, "memory": 28783, "step": 31690} +{"lr": 5.500000000000001e-05, "data_time": 0.0016729354858398438, "grad_norm": 0.8681560799479484, "loss": 0.20269544422626495, "time": 0.755955982208252, "epoch": 6, "memory": 28783, "step": 31790} +{"lr": 5.500000000000001e-05, "data_time": 0.002167534828186035, "grad_norm": 0.37329685539007185, "loss": 0.2024100810289383, "time": 0.7681767463684082, "epoch": 6, "memory": 28783, "step": 31890} +{"lr": 5.500000000000001e-05, "data_time": 0.0019175052642822266, "grad_norm": 0.49450180530548093, "loss": 0.2039802834391594, "time": 0.8147980213165283, "epoch": 6, "memory": 28783, "step": 31990} +{"lr": 5.500000000000001e-05, "data_time": 0.0015493392944335937, "grad_norm": 0.5048115089535713, "loss": 0.20200861096382142, "time": 0.8358577251434326, "epoch": 6, "memory": 28783, "step": 32090} +{"lr": 5.500000000000001e-05, "data_time": 0.0016155004501342773, "grad_norm": 0.6258242383599282, "loss": 0.20049929022789, "time": 0.7922482967376709, "epoch": 6, "memory": 28783, "step": 32190} +{"lr": 5.500000000000001e-05, "data_time": 0.0017861604690551757, "grad_norm": 0.5814644038677216, "loss": 0.20161413997411728, "time": 0.7749020099639893, "epoch": 6, "memory": 28783, "step": 32290} +{"lr": 5.500000000000001e-05, "data_time": 0.0017628908157348634, "grad_norm": 0.604364612698555, "loss": 0.20077844858169555, "time": 0.7862651109695434, "epoch": 6, "memory": 28783, "step": 32390} +{"lr": 5.500000000000001e-05, "data_time": 0.0019086360931396484, "grad_norm": 0.7971808582544326, "loss": 0.20373074412345887, "time": 0.716651177406311, "epoch": 6, "memory": 28783, "step": 32490} +{"lr": 5.500000000000001e-05, "data_time": 0.0020807504653930662, "grad_norm": 0.4211017806082964, "loss": 0.20024603456258774, "time": 0.7763030052185058, "epoch": 6, "memory": 28783, "step": 32590} +{"lr": 5.500000000000001e-05, "data_time": 0.0017891883850097655, "grad_norm": 0.5494905471801758, "loss": 0.20403071641921997, "time": 0.8501994848251343, "epoch": 6, "memory": 28783, "step": 32690} +{"lr": 5.500000000000001e-05, "data_time": 0.0032614469528198242, "grad_norm": 0.6392549939453602, "loss": 0.20406720638275147, "time": 0.8198360919952392, "epoch": 6, "memory": 28783, "step": 32790} +{"lr": 5.500000000000001e-05, "data_time": 0.0018544197082519531, "grad_norm": 0.5553544420748949, "loss": 0.20192965269088745, "time": 0.7752921104431152, "epoch": 6, "memory": 28783, "step": 32890} +{"lr": 5.500000000000001e-05, "data_time": 0.0031661510467529295, "grad_norm": 0.3655755288898945, "loss": 0.19952612221240998, "time": 0.9092059373855591, "epoch": 6, "memory": 28783, "step": 32990} +{"lr": 5.500000000000001e-05, "data_time": 0.0016002655029296875, "grad_norm": 0.5958395235240459, "loss": 0.2008439153432846, "time": 0.7857270956039428, "epoch": 6, "memory": 28783, "step": 33090} +{"lr": 5.500000000000001e-05, "data_time": 0.0016041994094848633, "grad_norm": 0.7052890866994858, "loss": 0.2020125523209572, "time": 0.7801207304000854, "epoch": 6, "memory": 28783, "step": 33190} +{"lr": 5.500000000000001e-05, "data_time": 0.0023583412170410157, "grad_norm": 0.41311595141887664, "loss": 0.20182453691959382, "time": 0.958920931816101, "epoch": 6, "memory": 28783, "step": 33290} +{"lr": 5.500000000000001e-05, "data_time": 0.0024319887161254883, "grad_norm": 0.8197527393698693, "loss": 0.20162223875522614, "time": 0.915973424911499, "epoch": 6, "memory": 28783, "step": 33390} +{"lr": 5.500000000000001e-05, "data_time": 0.0016688823699951172, "grad_norm": 0.549808344244957, "loss": 0.20218923240900039, "time": 0.8429347515106201, "epoch": 6, "memory": 28783, "step": 33490} +{"lr": 5.500000000000001e-05, "data_time": 0.0018612861633300782, "grad_norm": 0.9302523612976075, "loss": 0.20278658717870712, "time": 0.8232696056365967, "epoch": 6, "memory": 28783, "step": 33590} +{"lr": 5.500000000000001e-05, "data_time": 0.0022600412368774412, "grad_norm": 0.5280377715826035, "loss": 0.20467469990253448, "time": 0.8906569242477417, "epoch": 6, "memory": 28783, "step": 33690} +{"lr": 5.500000000000001e-05, "data_time": 0.0016196489334106446, "grad_norm": 0.3647256713360548, "loss": 0.20158257931470872, "time": 0.7626033306121827, "epoch": 6, "memory": 28783, "step": 33790} +{"lr": 5.500000000000001e-05, "data_time": 0.001937270164489746, "grad_norm": 0.3085356891155243, "loss": 0.20190771520137787, "time": 0.8342926979064942, "epoch": 6, "memory": 28783, "step": 33890} +{"lr": 5.500000000000001e-05, "data_time": 0.0019712209701538085, "grad_norm": 0.6018663965165615, "loss": 0.20116854310035706, "time": 0.9067827701568604, "epoch": 6, "memory": 28783, "step": 33990} +{"lr": 5.500000000000001e-05, "data_time": 0.002509808540344238, "grad_norm": 0.5749215565621852, "loss": 0.2031683385372162, "time": 0.9851001739501953, "epoch": 6, "memory": 28783, "step": 34090} +{"lr": 5.500000000000001e-05, "data_time": 0.002916765213012695, "grad_norm": 0.4083619400858879, "loss": 0.20266218483448029, "time": 0.9786816835403442, "epoch": 6, "memory": 28783, "step": 34190} +{"lr": 5.500000000000001e-05, "data_time": 0.0017733335494995116, "grad_norm": 0.4704970747232437, "loss": 0.20194581300020217, "time": 0.8775602340698242, "epoch": 6, "memory": 28783, "step": 34290} +{"lr": 5.500000000000001e-05, "data_time": 0.0032355070114135744, "grad_norm": 1.0642066419124603, "loss": 0.20757760852575302, "time": 1.0334163188934327, "epoch": 6, "memory": 28783, "step": 34390} +{"lr": 5.500000000000001e-05, "data_time": 0.0015826225280761719, "grad_norm": 0.5162757977843284, "loss": 0.20109372586011887, "time": 0.7544770956039428, "epoch": 6, "memory": 28783, "step": 34490} +{"accuracy/top1": 99.54547882080078, "data_time": 0.2649371203254251, "time": 0.6807534133686739, "step": 6} +{"lr": 4.109423525312737e-05, "data_time": 0.2570410013198853, "grad_norm": 0.5304535984992981, "loss": 0.20379682183265685, "time": 0.8140489339828492, "epoch": 7, "memory": 28783, "step": 34648} +{"lr": 4.109423525312737e-05, "data_time": 0.3319602727890015, "grad_norm": 0.6574077785015107, "loss": 0.2008529469370842, "time": 0.9948984861373902, "epoch": 7, "memory": 28783, "step": 34748} +{"lr": 4.109423525312737e-05, "data_time": 0.002550840377807617, "grad_norm": 0.6974850043654441, "loss": 0.2028369501233101, "time": 0.8849415302276611, "epoch": 7, "memory": 28783, "step": 34848} +{"lr": 4.109423525312737e-05, "data_time": 0.0018515586853027344, "grad_norm": 0.731633386015892, "loss": 0.20060226917266846, "time": 0.8067174196243286, "epoch": 7, "memory": 28783, "step": 34948} +{"lr": 4.109423525312737e-05, "data_time": 0.0017177581787109375, "grad_norm": 0.39038731157779694, "loss": 0.19997669905424117, "time": 0.8141330242156982, "epoch": 7, "memory": 28783, "step": 35048} +{"lr": 4.109423525312737e-05, "data_time": 0.001930999755859375, "grad_norm": 0.7535681523382664, "loss": 0.2011270508170128, "time": 0.7879596710205078, "epoch": 7, "memory": 28783, "step": 35148} +{"lr": 4.109423525312737e-05, "data_time": 0.0022539854049682616, "grad_norm": 0.6121316776610911, "loss": 0.20207958221435546, "time": 0.9131685733795166, "epoch": 7, "memory": 28783, "step": 35248} +{"lr": 4.109423525312737e-05, "data_time": 0.003457784652709961, "grad_norm": 0.5684374149888753, "loss": 0.2024962857365608, "time": 0.8502788305282593, "epoch": 7, "memory": 28783, "step": 35348} +{"lr": 4.109423525312737e-05, "data_time": 0.0024862051010131835, "grad_norm": 0.35475060492753985, "loss": 0.20085809677839278, "time": 0.8818483352661133, "epoch": 7, "memory": 28783, "step": 35448} +{"lr": 4.109423525312737e-05, "data_time": 0.0016196966171264648, "grad_norm": 0.5782546050846576, "loss": 0.20242090672254562, "time": 0.7270224571228028, "epoch": 7, "memory": 28783, "step": 35548} +{"lr": 4.109423525312737e-05, "data_time": 0.07658209800720214, "grad_norm": 0.43485084772109983, "loss": 0.20089922547340394, "time": 0.8768445491790772, "epoch": 7, "memory": 28783, "step": 35648} +{"lr": 4.109423525312737e-05, "data_time": 0.0017717838287353515, "grad_norm": 0.3603450506925583, "loss": 0.19999355375766753, "time": 0.7718201637268066, "epoch": 7, "memory": 28783, "step": 35748} +{"lr": 4.109423525312737e-05, "data_time": 0.0015961408615112304, "grad_norm": 0.6901052325963974, "loss": 0.2033108502626419, "time": 0.8285606145858765, "epoch": 7, "memory": 28783, "step": 35848} +{"lr": 4.109423525312737e-05, "data_time": 0.0026313304901123048, "grad_norm": 0.5386508405208588, "loss": 0.20118250995874404, "time": 0.7770572662353515, "epoch": 7, "memory": 28783, "step": 35948} +{"lr": 4.109423525312737e-05, "data_time": 0.0018102884292602538, "grad_norm": 0.5510954394936561, "loss": 0.2002228692173958, "time": 0.8777055740356445, "epoch": 7, "memory": 28783, "step": 36048} +{"lr": 4.109423525312737e-05, "data_time": 0.0017973661422729492, "grad_norm": 0.44983562976121905, "loss": 0.20088137984275817, "time": 0.8439640045166016, "epoch": 7, "memory": 28783, "step": 36148} +{"lr": 4.109423525312737e-05, "data_time": 0.0027127742767333986, "grad_norm": 0.37173020765185355, "loss": 0.19929925799369813, "time": 0.8717893600463867, "epoch": 7, "memory": 28783, "step": 36248} +{"lr": 4.109423525312737e-05, "data_time": 0.001730060577392578, "grad_norm": 0.5331165784969926, "loss": 0.201166507601738, "time": 0.7625855207443237, "epoch": 7, "memory": 28783, "step": 36348} +{"lr": 4.109423525312737e-05, "data_time": 0.002619028091430664, "grad_norm": 0.2659651841968298, "loss": 0.1999305322766304, "time": 0.9332836389541626, "epoch": 7, "memory": 28783, "step": 36448} +{"lr": 4.109423525312737e-05, "data_time": 0.001663970947265625, "grad_norm": 0.38814663216471673, "loss": 0.20105892717838286, "time": 0.8786244630813599, "epoch": 7, "memory": 28783, "step": 36548} +{"lr": 4.109423525312737e-05, "data_time": 0.001772594451904297, "grad_norm": 0.48508835919201376, "loss": 0.19943130314350127, "time": 0.853243613243103, "epoch": 7, "memory": 28783, "step": 36648} +{"lr": 4.109423525312737e-05, "data_time": 0.002699422836303711, "grad_norm": 0.8456787571310997, "loss": 0.20061280727386474, "time": 0.9556742668151855, "epoch": 7, "memory": 28783, "step": 36748} +{"lr": 4.109423525312737e-05, "data_time": 0.002043747901916504, "grad_norm": 0.5865349501371384, "loss": 0.20019570738077164, "time": 0.7913581371307373, "epoch": 7, "memory": 28783, "step": 36848} +{"lr": 4.109423525312737e-05, "data_time": 0.001745319366455078, "grad_norm": 0.4370046705007553, "loss": 0.20116384476423263, "time": 0.7525956630706787, "epoch": 7, "memory": 28783, "step": 36948} +{"lr": 4.109423525312737e-05, "data_time": 0.0035280704498291014, "grad_norm": 0.4532147437334061, "loss": 0.1995263069868088, "time": 0.9600348472595215, "epoch": 7, "memory": 28783, "step": 37048} +{"lr": 4.109423525312737e-05, "data_time": 0.002026486396789551, "grad_norm": 0.5822193264961243, "loss": 0.1996021643280983, "time": 0.8014884233474732, "epoch": 7, "memory": 28783, "step": 37148} +{"lr": 4.109423525312737e-05, "data_time": 0.0026641845703125, "grad_norm": 0.32727089077234267, "loss": 0.20017112493515016, "time": 0.8950970649719239, "epoch": 7, "memory": 28783, "step": 37248} +{"lr": 4.109423525312737e-05, "data_time": 0.002028346061706543, "grad_norm": 0.37457580752670766, "loss": 0.2001060962677002, "time": 0.7436938524246216, "epoch": 7, "memory": 28783, "step": 37348} +{"lr": 4.109423525312737e-05, "data_time": 0.0031909704208374023, "grad_norm": 0.476037173718214, "loss": 0.20002157092094422, "time": 0.9007952213287354, "epoch": 7, "memory": 28783, "step": 37448} +{"lr": 4.109423525312737e-05, "data_time": 0.0017638444900512696, "grad_norm": 0.559312067180872, "loss": 0.19941863864660264, "time": 0.7836038827896118, "epoch": 7, "memory": 28783, "step": 37548} +{"lr": 4.109423525312737e-05, "data_time": 0.0027809858322143553, "grad_norm": 0.43681189604103565, "loss": 0.20012189894914628, "time": 0.898586630821228, "epoch": 7, "memory": 28783, "step": 37648} +{"lr": 4.109423525312737e-05, "data_time": 0.0019791841506958006, "grad_norm": 0.39596344493329527, "loss": 0.19973721355199814, "time": 0.8397162437438965, "epoch": 7, "memory": 28783, "step": 37748} +{"lr": 4.109423525312737e-05, "data_time": 0.0030755519866943358, "grad_norm": 0.4780492804944515, "loss": 0.20098764449357986, "time": 0.8755083084106445, "epoch": 7, "memory": 28783, "step": 37848} +{"lr": 4.109423525312737e-05, "data_time": 0.0029319047927856444, "grad_norm": 0.49840219151228665, "loss": 0.19930738508701323, "time": 0.9035987377166748, "epoch": 7, "memory": 28783, "step": 37948} +{"lr": 4.109423525312737e-05, "data_time": 0.0016916990280151367, "grad_norm": 0.5670174971222878, "loss": 0.20086415261030197, "time": 0.7433984518051148, "epoch": 7, "memory": 28783, "step": 38048} +{"lr": 4.109423525312737e-05, "data_time": 0.00224001407623291, "grad_norm": 0.4579376712441444, "loss": 0.20129559189081192, "time": 0.859979772567749, "epoch": 7, "memory": 28783, "step": 38148} +{"lr": 4.109423525312737e-05, "data_time": 0.0016407489776611329, "grad_norm": 0.48678094148635864, "loss": 0.19966082721948625, "time": 0.7750612735748291, "epoch": 7, "memory": 28783, "step": 38248} +{"lr": 4.109423525312737e-05, "data_time": 0.004172062873840332, "grad_norm": 0.46413626074790953, "loss": 0.2013307586312294, "time": 0.9615135669708252, "epoch": 7, "memory": 28783, "step": 38348} +{"lr": 4.109423525312737e-05, "data_time": 0.0021019220352172852, "grad_norm": 0.376536026597023, "loss": 0.19990138560533524, "time": 0.9268024682998657, "epoch": 7, "memory": 28783, "step": 38448} +{"lr": 4.109423525312737e-05, "data_time": 0.0019810676574707033, "grad_norm": 0.4711449682712555, "loss": 0.19912588149309157, "time": 0.798414945602417, "epoch": 7, "memory": 28783, "step": 38548} +{"lr": 4.109423525312737e-05, "data_time": 0.0017978429794311523, "grad_norm": 0.37200071290135384, "loss": 0.1994968444108963, "time": 0.7944779396057129, "epoch": 7, "memory": 28783, "step": 38648} +{"lr": 4.109423525312737e-05, "data_time": 0.0017348527908325195, "grad_norm": 0.6079226233065128, "loss": 0.20247834026813508, "time": 0.8917044878005982, "epoch": 7, "memory": 28783, "step": 38748} +{"lr": 4.109423525312737e-05, "data_time": 0.003507566452026367, "grad_norm": 0.4597327888011932, "loss": 0.2032423883676529, "time": 0.9037577629089355, "epoch": 7, "memory": 28783, "step": 38848} +{"lr": 4.109423525312737e-05, "data_time": 0.0018663883209228515, "grad_norm": 0.614964597299695, "loss": 0.2004171773791313, "time": 0.8065264225006104, "epoch": 7, "memory": 28783, "step": 38948} +{"lr": 4.109423525312737e-05, "data_time": 0.0018650054931640624, "grad_norm": 0.2859881963580847, "loss": 0.20019351243972777, "time": 0.9513052701950073, "epoch": 7, "memory": 28783, "step": 39048} +{"lr": 4.109423525312737e-05, "data_time": 0.0016150236129760741, "grad_norm": 0.4854913234710693, "loss": 0.19948391169309615, "time": 0.8299093008041382, "epoch": 7, "memory": 28783, "step": 39148} +{"lr": 4.109423525312737e-05, "data_time": 0.001726984977722168, "grad_norm": 0.3842244539409876, "loss": 0.20226747393608094, "time": 0.771368432044983, "epoch": 7, "memory": 28783, "step": 39248} +{"lr": 4.109423525312737e-05, "data_time": 0.0016712427139282226, "grad_norm": 0.3702883360907435, "loss": 0.2003416895866394, "time": 0.8271692037582398, "epoch": 7, "memory": 28783, "step": 39348} +{"lr": 4.109423525312737e-05, "data_time": 0.0017933130264282226, "grad_norm": 0.43516001403331755, "loss": 0.20015901774168016, "time": 0.7821183204650879, "epoch": 7, "memory": 28783, "step": 39448} +{"lr": 4.109423525312737e-05, "data_time": 0.0018541574478149413, "grad_norm": 0.3398057587444782, "loss": 0.20029745399951934, "time": 0.8236588478088379, "epoch": 7, "memory": 28783, "step": 39548} +{"lr": 4.109423525312737e-05, "data_time": 0.001649951934814453, "grad_norm": 0.4105547945946455, "loss": 0.19939403980970383, "time": 0.8436854124069214, "epoch": 7, "memory": 28783, "step": 39648} +{"lr": 4.109423525312737e-05, "data_time": 0.0016717195510864257, "grad_norm": 0.43874640837311746, "loss": 0.20365124493837355, "time": 0.7962920904159546, "epoch": 7, "memory": 28783, "step": 39748} +{"lr": 4.109423525312737e-05, "data_time": 0.0017191410064697266, "grad_norm": 0.49381513670086863, "loss": 0.1998294323682785, "time": 0.7393632173538208, "epoch": 7, "memory": 28783, "step": 39848} +{"lr": 4.109423525312737e-05, "data_time": 0.0015441179275512695, "grad_norm": 0.4961566381156445, "loss": 0.19969123601913452, "time": 0.7431836128234863, "epoch": 7, "memory": 28783, "step": 39948} +{"lr": 4.109423525312737e-05, "data_time": 0.08929152488708496, "grad_norm": 0.3904197491705418, "loss": 0.2005852296948433, "time": 0.9282039642333985, "epoch": 7, "memory": 28783, "step": 40048} +{"lr": 4.109423525312737e-05, "data_time": 0.0017680644989013672, "grad_norm": 0.9225305676460266, "loss": 0.2009689301252365, "time": 0.7587546825408935, "epoch": 7, "memory": 28783, "step": 40148} +{"lr": 4.109423525312737e-05, "data_time": 0.004726099967956543, "grad_norm": 0.534372840076685, "loss": 0.19917624592781066, "time": 0.8363152027130127, "epoch": 7, "memory": 28783, "step": 40248} +{"accuracy/top1": 99.955810546875, "data_time": 0.32658713004168344, "time": 0.7414570555967444, "step": 7} +{"lr": 2.8549663646838717e-05, "data_time": 0.3024162292480469, "grad_norm": 0.34962107576429846, "loss": 0.2010873794555664, "time": 0.8124499082565307, "epoch": 8, "memory": 28783, "step": 40406} +{"lr": 2.8549663646838717e-05, "data_time": 0.22667150497436522, "grad_norm": 0.41273797005414964, "loss": 0.20046880096197128, "time": 0.8970856666564941, "epoch": 8, "memory": 28783, "step": 40506} +{"lr": 2.8549663646838717e-05, "data_time": 0.2134321451187134, "grad_norm": 0.561300278455019, "loss": 0.20044224411249162, "time": 0.7762515068054199, "epoch": 8, "memory": 28783, "step": 40606} +{"lr": 2.8549663646838717e-05, "data_time": 0.16150379180908203, "grad_norm": 0.38815378621220586, "loss": 0.20110579282045365, "time": 0.7558535814285279, "epoch": 8, "memory": 28783, "step": 40706} +{"lr": 2.8549663646838717e-05, "data_time": 0.16156353950500488, "grad_norm": 0.4839564699679613, "loss": 0.20004840642213823, "time": 0.7756923913955689, "epoch": 8, "memory": 28783, "step": 40806} +{"lr": 2.8549663646838717e-05, "data_time": 0.2550986289978027, "grad_norm": 0.43951306864619255, "loss": 0.19977218955755233, "time": 0.8354975938796997, "epoch": 8, "memory": 28783, "step": 40906} +{"lr": 2.8549663646838717e-05, "data_time": 0.24363205432891846, "grad_norm": 0.45079954117536547, "loss": 0.19976240992546082, "time": 0.769334363937378, "epoch": 8, "memory": 28783, "step": 41006} +{"lr": 2.8549663646838717e-05, "data_time": 0.2094966411590576, "grad_norm": 0.27666075006127355, "loss": 0.20195276737213136, "time": 0.9607676029205322, "epoch": 8, "memory": 28783, "step": 41106} +{"lr": 2.8549663646838717e-05, "data_time": 0.1692953586578369, "grad_norm": 0.37942475695163014, "loss": 0.19994042366743087, "time": 0.9432299613952637, "epoch": 8, "memory": 28783, "step": 41206} +{"lr": 2.8549663646838717e-05, "data_time": 0.002052116394042969, "grad_norm": 0.3468345979228616, "loss": 0.20011111795902253, "time": 0.8091646671295166, "epoch": 8, "memory": 28783, "step": 41306} +{"lr": 2.8549663646838717e-05, "data_time": 0.0016150474548339844, "grad_norm": 0.5458853382617235, "loss": 0.1991449177265167, "time": 0.7599417209625244, "epoch": 8, "memory": 28783, "step": 41406} +{"lr": 2.8549663646838717e-05, "data_time": 0.0015526294708251953, "grad_norm": 0.8896437801420689, "loss": 0.20091405212879182, "time": 0.8348368167877197, "epoch": 8, "memory": 28783, "step": 41506} +{"lr": 2.8549663646838717e-05, "data_time": 0.0023251771926879883, "grad_norm": 0.48138552606105806, "loss": 0.2000388041138649, "time": 0.8940004348754883, "epoch": 8, "memory": 28783, "step": 41606} +{"lr": 2.8549663646838717e-05, "data_time": 0.0019300460815429687, "grad_norm": 0.3746541578322649, "loss": 0.200286203622818, "time": 0.8362886190414429, "epoch": 8, "memory": 28783, "step": 41706} +{"lr": 2.8549663646838717e-05, "data_time": 0.0020024776458740234, "grad_norm": 0.29308299366384744, "loss": 0.19902280122041702, "time": 0.8487228870391845, "epoch": 8, "memory": 28783, "step": 41806} +{"lr": 2.8549663646838717e-05, "data_time": 0.005043601989746094, "grad_norm": 0.337833072245121, "loss": 0.19973892867565154, "time": 0.9286252498626709, "epoch": 8, "memory": 28783, "step": 41906} +{"lr": 2.8549663646838717e-05, "data_time": 0.03813190460205078, "grad_norm": 0.33194583766162394, "loss": 0.19875675886869432, "time": 0.8040109157562256, "epoch": 8, "memory": 28783, "step": 42006} +{"lr": 2.8549663646838717e-05, "data_time": 0.1507725715637207, "grad_norm": 0.5127151221036911, "loss": 0.1998363807797432, "time": 0.8586250305175781, "epoch": 8, "memory": 28783, "step": 42106} +{"lr": 2.8549663646838717e-05, "data_time": 0.04830701351165771, "grad_norm": 0.19206122159957886, "loss": 0.19865394681692122, "time": 0.8016723155975342, "epoch": 8, "memory": 28783, "step": 42206} +{"lr": 2.8549663646838717e-05, "data_time": 0.20780677795410157, "grad_norm": 0.6225557684898376, "loss": 0.2015370950102806, "time": 0.873365068435669, "epoch": 8, "memory": 28783, "step": 42306} +{"lr": 2.8549663646838717e-05, "data_time": 0.0017276525497436524, "grad_norm": 0.4400020346045494, "loss": 0.20197580009698868, "time": 0.7947998762130737, "epoch": 8, "memory": 28783, "step": 42406} +{"lr": 2.8549663646838717e-05, "data_time": 0.0344501256942749, "grad_norm": 0.3868662118911743, "loss": 0.19997474551200867, "time": 0.7513847589492798, "epoch": 8, "memory": 28783, "step": 42506} +{"lr": 2.8549663646838717e-05, "data_time": 0.10830955505371094, "grad_norm": 0.2928969316184521, "loss": 0.1989822193980217, "time": 0.8792859792709351, "epoch": 8, "memory": 28783, "step": 42606} +{"lr": 2.8549663646838717e-05, "data_time": 0.13611235618591308, "grad_norm": 0.28904529884457586, "loss": 0.199774332344532, "time": 0.7907635927200317, "epoch": 8, "memory": 28783, "step": 42706} +{"lr": 2.8549663646838717e-05, "data_time": 0.23929970264434813, "grad_norm": 0.3635988987982273, "loss": 0.1997097462415695, "time": 0.7439430952072144, "epoch": 8, "memory": 28783, "step": 42806} +{"lr": 2.8549663646838717e-05, "data_time": 0.17196111679077147, "grad_norm": 0.276888757199049, "loss": 0.198770934343338, "time": 0.9269291877746582, "epoch": 8, "memory": 28783, "step": 42906} +{"lr": 2.8549663646838717e-05, "data_time": 0.061671996116638185, "grad_norm": 0.7010551903396844, "loss": 0.20110541582107544, "time": 0.7903759241104126, "epoch": 8, "memory": 28783, "step": 43006} +{"lr": 2.8549663646838717e-05, "data_time": 0.07312889099121093, "grad_norm": 0.40671457201242445, "loss": 0.20088512301445008, "time": 0.8595164299011231, "epoch": 8, "memory": 28783, "step": 43106} +{"lr": 2.8549663646838717e-05, "data_time": 0.2648433446884155, "grad_norm": 0.39031699299812317, "loss": 0.20173955261707305, "time": 0.7758603811264038, "epoch": 8, "memory": 28783, "step": 43206} +{"lr": 2.8549663646838717e-05, "data_time": 0.025339365005493164, "grad_norm": 0.4093537891283631, "loss": 0.1994659811258316, "time": 0.8794605255126953, "epoch": 8, "memory": 28783, "step": 43306} +{"lr": 2.8549663646838717e-05, "data_time": 0.0016026735305786134, "grad_norm": 0.5903012270107866, "loss": 0.20138724148273468, "time": 0.7433746814727783, "epoch": 8, "memory": 28783, "step": 43406} +{"lr": 2.8549663646838717e-05, "data_time": 0.0021006107330322266, "grad_norm": 0.3646797526627779, "loss": 0.19887280464172363, "time": 0.8078948497772217, "epoch": 8, "memory": 28783, "step": 43506} +{"lr": 2.8549663646838717e-05, "data_time": 0.0032089948654174805, "grad_norm": 0.42064387165009975, "loss": 0.20039766728878022, "time": 0.9307423114776612, "epoch": 8, "memory": 28783, "step": 43606} +{"lr": 2.8549663646838717e-05, "data_time": 0.0017543077468872071, "grad_norm": 0.18693483658134938, "loss": 0.19990795403718947, "time": 0.9137980222702027, "epoch": 8, "memory": 28783, "step": 43706} +{"lr": 2.8549663646838717e-05, "data_time": 0.0015865325927734374, "grad_norm": 0.4696742117404938, "loss": 0.19978507459163666, "time": 0.8123665571212768, "epoch": 8, "memory": 28783, "step": 43806} +{"lr": 2.8549663646838717e-05, "data_time": 0.0023093938827514647, "grad_norm": 0.34556571766734123, "loss": 0.20007470697164537, "time": 0.8867485761642456, "epoch": 8, "memory": 28783, "step": 43906} +{"lr": 2.8549663646838717e-05, "data_time": 0.001970839500427246, "grad_norm": 0.32554522417485715, "loss": 0.20074764490127564, "time": 0.77413969039917, "epoch": 8, "memory": 28783, "step": 44006} +{"lr": 2.8549663646838717e-05, "data_time": 0.001967763900756836, "grad_norm": 0.342456965893507, "loss": 0.19901065081357955, "time": 0.8418396949768067, "epoch": 8, "memory": 28783, "step": 44106} +{"lr": 2.8549663646838717e-05, "data_time": 0.002103900909423828, "grad_norm": 0.33164366427809, "loss": 0.1990720212459564, "time": 0.7931753873825074, "epoch": 8, "memory": 28783, "step": 44206} +{"lr": 2.8549663646838717e-05, "data_time": 0.0017536640167236327, "grad_norm": 0.42727058231830595, "loss": 0.2015230730175972, "time": 0.7858794927597046, "epoch": 8, "memory": 28783, "step": 44306} +{"lr": 2.8549663646838717e-05, "data_time": 0.001806187629699707, "grad_norm": 0.3826434537768364, "loss": 0.20064708441495896, "time": 0.8281020164489746, "epoch": 8, "memory": 28783, "step": 44406} +{"lr": 2.8549663646838717e-05, "data_time": 0.0016503095626831054, "grad_norm": 0.5383035026490688, "loss": 0.20213476270437242, "time": 0.8384078741073608, "epoch": 8, "memory": 28783, "step": 44506} +{"lr": 2.8549663646838717e-05, "data_time": 0.0019080162048339844, "grad_norm": 0.4320871841162443, "loss": 0.2004919692873955, "time": 0.8009705543518066, "epoch": 8, "memory": 28783, "step": 44606} +{"lr": 2.8549663646838717e-05, "data_time": 0.0018613576889038087, "grad_norm": 0.3800024766474962, "loss": 0.2002116397023201, "time": 0.8616921901702881, "epoch": 8, "memory": 28783, "step": 44706} +{"lr": 2.8549663646838717e-05, "data_time": 0.0020083189010620117, "grad_norm": 0.6177150852978229, "loss": 0.19994091540575026, "time": 0.835884690284729, "epoch": 8, "memory": 28783, "step": 44806} +{"lr": 2.8549663646838717e-05, "data_time": 0.0017868280410766602, "grad_norm": 0.27003599554300306, "loss": 0.19936210960149764, "time": 0.8957467555999756, "epoch": 8, "memory": 28783, "step": 44906} +{"lr": 2.8549663646838717e-05, "data_time": 0.0021057605743408205, "grad_norm": 0.47350957840681074, "loss": 0.2004600703716278, "time": 0.9268666744232178, "epoch": 8, "memory": 28783, "step": 45006} +{"lr": 2.8549663646838717e-05, "data_time": 0.0016373395919799805, "grad_norm": 0.3515616409480572, "loss": 0.19945160746574403, "time": 0.8554864645004272, "epoch": 8, "memory": 28783, "step": 45106} +{"lr": 2.8549663646838717e-05, "data_time": 0.0017644643783569336, "grad_norm": 0.31704646423459054, "loss": 0.19900186806917192, "time": 0.7706925868988037, "epoch": 8, "memory": 28783, "step": 45206} +{"lr": 2.8549663646838717e-05, "data_time": 0.004744291305541992, "grad_norm": 0.5062963657081128, "loss": 0.1999295622110367, "time": 0.9430924177169799, "epoch": 8, "memory": 28783, "step": 45306} +{"lr": 2.8549663646838717e-05, "data_time": 0.00233762264251709, "grad_norm": 0.15249796607531607, "loss": 0.19973793625831604, "time": 0.8423353433609009, "epoch": 8, "memory": 28783, "step": 45406} +{"lr": 2.8549663646838717e-05, "data_time": 0.0018955469131469727, "grad_norm": 0.46713529601693154, "loss": 0.1992884933948517, "time": 0.8008431911468505, "epoch": 8, "memory": 28783, "step": 45506} +{"lr": 2.8549663646838717e-05, "data_time": 0.005891561508178711, "grad_norm": 0.3140792388468981, "loss": 0.19922010749578475, "time": 0.9224577665328979, "epoch": 8, "memory": 28783, "step": 45606} +{"lr": 2.8549663646838717e-05, "data_time": 0.0022818326950073244, "grad_norm": 0.9852781891822815, "loss": 0.20175724625587463, "time": 0.8994702816009521, "epoch": 8, "memory": 28783, "step": 45706} +{"lr": 2.8549663646838717e-05, "data_time": 0.0019786596298217774, "grad_norm": 0.36189441829919816, "loss": 0.20120375454425812, "time": 0.8120423555374146, "epoch": 8, "memory": 28783, "step": 45806} +{"lr": 2.8549663646838717e-05, "data_time": 0.002092885971069336, "grad_norm": 0.2625554073601961, "loss": 0.19908541440963745, "time": 0.8677398204803467, "epoch": 8, "memory": 28783, "step": 45906} +{"lr": 2.8549663646838717e-05, "data_time": 0.0027587890625, "grad_norm": 0.4917635917663574, "loss": 0.19970357567071914, "time": 0.8837395429611206, "epoch": 8, "memory": 28783, "step": 46006} +{"accuracy/top1": 99.92424774169922, "data_time": 0.20535842110128963, "time": 0.6206446254954618, "step": 8} +{"lr": 1.8594235253127375e-05, "data_time": 0.31137473583221437, "grad_norm": 0.5146086907014251, "loss": 0.20010972768068314, "time": 0.822504711151123, "epoch": 9, "memory": 28783, "step": 46164} +{"lr": 1.8594235253127375e-05, "data_time": 0.21240315437316895, "grad_norm": 0.30397630073130133, "loss": 0.20071698427200318, "time": 0.8247225999832153, "epoch": 9, "memory": 28783, "step": 46264} +{"lr": 1.8594235253127375e-05, "data_time": 0.2732653856277466, "grad_norm": 0.43730893582105634, "loss": 0.19915433824062348, "time": 0.7790990591049194, "epoch": 9, "memory": 28783, "step": 46364} +{"lr": 1.8594235253127375e-05, "data_time": 0.32458369731903075, "grad_norm": 0.25308656990528106, "loss": 0.19879508912563323, "time": 0.8539206504821777, "epoch": 9, "memory": 28783, "step": 46464} +{"lr": 1.8594235253127375e-05, "data_time": 0.3200049877166748, "grad_norm": 0.23486879272386432, "loss": 0.19984957277774812, "time": 0.8361745119094849, "epoch": 9, "memory": 28783, "step": 46564} +{"lr": 1.8594235253127375e-05, "data_time": 0.35312695503234864, "grad_norm": 0.23702208884060383, "loss": 0.20000497996807098, "time": 0.8733285903930664, "epoch": 9, "memory": 28783, "step": 46664} +{"lr": 1.8594235253127375e-05, "data_time": 0.23268754482269288, "grad_norm": 0.3531709173694253, "loss": 0.1987038642168045, "time": 0.8479858636856079, "epoch": 9, "memory": 28783, "step": 46764} +{"lr": 1.8594235253127375e-05, "data_time": 0.2803735971450806, "grad_norm": 0.31653384109959004, "loss": 0.19929299056529998, "time": 0.794532322883606, "epoch": 9, "memory": 28783, "step": 46864} +{"lr": 1.8594235253127375e-05, "data_time": 0.25891790390014646, "grad_norm": 0.24372444190084935, "loss": 0.19953625351190568, "time": 0.9263168573379517, "epoch": 9, "memory": 28783, "step": 46964} +{"lr": 1.8594235253127375e-05, "data_time": 0.0035434246063232424, "grad_norm": 0.181035035289824, "loss": 0.19964278638362884, "time": 0.94240300655365, "epoch": 9, "memory": 28783, "step": 47064} +{"lr": 1.8594235253127375e-05, "data_time": 0.0020153999328613283, "grad_norm": 0.35834435671567916, "loss": 0.199106827378273, "time": 0.8317167282104492, "epoch": 9, "memory": 28783, "step": 47164} +{"lr": 1.8594235253127375e-05, "data_time": 0.0061484575271606445, "grad_norm": 0.26160954162478445, "loss": 0.19912890940904618, "time": 0.8386593818664551, "epoch": 9, "memory": 28783, "step": 47264} +{"lr": 1.8594235253127375e-05, "data_time": 0.0019340276718139648, "grad_norm": 0.27888133358210326, "loss": 0.20130230635404586, "time": 0.797444200515747, "epoch": 9, "memory": 28783, "step": 47364} +{"lr": 1.8594235253127375e-05, "data_time": 0.0020103693008422852, "grad_norm": 0.2620167948305607, "loss": 0.19866974800825118, "time": 0.8212775945663452, "epoch": 9, "memory": 28783, "step": 47464} +{"lr": 1.8594235253127375e-05, "data_time": 0.0017138957977294923, "grad_norm": 0.38680356815457345, "loss": 0.19949098229408263, "time": 0.7680622816085816, "epoch": 9, "memory": 28783, "step": 47564} +{"lr": 1.8594235253127375e-05, "data_time": 0.12626335620880128, "grad_norm": 0.4178945438936353, "loss": 0.19984035342931747, "time": 0.9743272542953492, "epoch": 9, "memory": 28783, "step": 47664} +{"lr": 1.8594235253127375e-05, "data_time": 0.0022294044494628905, "grad_norm": 0.38078233897686004, "loss": 0.19955205768346787, "time": 0.8499480009078979, "epoch": 9, "memory": 28783, "step": 47764} +{"lr": 1.8594235253127375e-05, "data_time": 0.002201485633850098, "grad_norm": 0.29727129861712454, "loss": 0.19900396168231965, "time": 0.819485855102539, "epoch": 9, "memory": 28783, "step": 47864} +{"lr": 1.8594235253127375e-05, "data_time": 0.0053002357482910155, "grad_norm": 0.3172834202647209, "loss": 0.19905504882335662, "time": 0.7542827606201172, "epoch": 9, "memory": 28783, "step": 47964} +{"lr": 1.8594235253127375e-05, "data_time": 0.004354524612426758, "grad_norm": 0.2702242273837328, "loss": 0.19889454692602157, "time": 0.8291808843612671, "epoch": 9, "memory": 28783, "step": 48064} +{"lr": 1.8594235253127375e-05, "data_time": 0.001863551139831543, "grad_norm": 0.4502132132649422, "loss": 0.19917707443237304, "time": 0.8059434413909912, "epoch": 9, "memory": 28783, "step": 48164} +{"lr": 1.8594235253127375e-05, "data_time": 0.002024984359741211, "grad_norm": 0.36323790764436126, "loss": 0.19987213015556335, "time": 0.7517117023468017, "epoch": 9, "memory": 28783, "step": 48264} +{"lr": 1.8594235253127375e-05, "data_time": 0.0018815994262695312, "grad_norm": 0.20545513220131398, "loss": 0.19894106984138488, "time": 0.8639640808105469, "epoch": 9, "memory": 28783, "step": 48364} +{"lr": 1.8594235253127375e-05, "data_time": 0.002356886863708496, "grad_norm": 0.26018533618189393, "loss": 0.19868551343679428, "time": 0.7843921184539795, "epoch": 9, "memory": 28783, "step": 48464} +{"lr": 1.8594235253127375e-05, "data_time": 0.002684283256530762, "grad_norm": 0.31424256768077613, "loss": 0.19942981451749803, "time": 0.911528468132019, "epoch": 9, "memory": 28783, "step": 48564} +{"lr": 1.8594235253127375e-05, "data_time": 0.0021970272064208984, "grad_norm": 0.24057991933077574, "loss": 0.19856883138418197, "time": 0.8527655601501465, "epoch": 9, "memory": 28783, "step": 48664} +{"lr": 1.8594235253127375e-05, "data_time": 0.0019884824752807615, "grad_norm": 0.41368495635688307, "loss": 0.20269487351179122, "time": 0.8380220651626586, "epoch": 9, "memory": 28783, "step": 48764} +{"lr": 1.8594235253127375e-05, "data_time": 0.0016928672790527343, "grad_norm": 0.5483454247005284, "loss": 0.19914476871490477, "time": 0.7924807786941528, "epoch": 9, "memory": 28783, "step": 48864} +{"lr": 1.8594235253127375e-05, "data_time": 0.002683591842651367, "grad_norm": 0.280898093059659, "loss": 0.19919253140687943, "time": 0.8820698022842407, "epoch": 9, "memory": 28783, "step": 48964} +{"lr": 1.8594235253127375e-05, "data_time": 0.0029338836669921876, "grad_norm": 0.3208234779536724, "loss": 0.19925323724746705, "time": 0.9235302686691285, "epoch": 9, "memory": 28783, "step": 49064} +{"lr": 1.8594235253127375e-05, "data_time": 0.002023649215698242, "grad_norm": 0.3346285748295486, "loss": 0.19955916851758956, "time": 0.8595220804214477, "epoch": 9, "memory": 28783, "step": 49164} +{"lr": 1.8594235253127375e-05, "data_time": 0.0018409490585327148, "grad_norm": 0.18432259727269412, "loss": 0.19887392073869706, "time": 0.7800418376922608, "epoch": 9, "memory": 28783, "step": 49264} +{"lr": 1.8594235253127375e-05, "data_time": 0.0033204078674316405, "grad_norm": 0.2936939208768308, "loss": 0.1985804796218872, "time": 0.9233827829360962, "epoch": 9, "memory": 28783, "step": 49364} +{"lr": 1.8594235253127375e-05, "data_time": 0.0024909734725952148, "grad_norm": 0.34559784568846225, "loss": 0.19987995326519012, "time": 0.9003695249557495, "epoch": 9, "memory": 28783, "step": 49464} +{"lr": 1.8594235253127375e-05, "data_time": 0.0019880056381225584, "grad_norm": 0.24455010928213597, "loss": 0.20014021694660186, "time": 0.7804906845092774, "epoch": 9, "memory": 28783, "step": 49564} +{"lr": 1.8594235253127375e-05, "data_time": 0.001706385612487793, "grad_norm": 0.42133839689195157, "loss": 0.20006576925516129, "time": 0.783941125869751, "epoch": 9, "memory": 28783, "step": 49664} +{"lr": 1.8594235253127375e-05, "data_time": 0.0025417327880859373, "grad_norm": 0.3483779177069664, "loss": 0.20002958178520203, "time": 0.7987715244293213, "epoch": 9, "memory": 28783, "step": 49764} +{"lr": 1.8594235253127375e-05, "data_time": 0.0023015499114990234, "grad_norm": 0.37101004449650643, "loss": 0.1986461713910103, "time": 0.8217520952224732, "epoch": 9, "memory": 28783, "step": 49864} +{"lr": 1.8594235253127375e-05, "data_time": 0.0017209768295288086, "grad_norm": 0.18766246289014815, "loss": 0.19873131364583968, "time": 0.8231493473052979, "epoch": 9, "memory": 28783, "step": 49964} +{"lr": 1.8594235253127375e-05, "data_time": 0.002186441421508789, "grad_norm": 0.4158086239360273, "loss": 0.20050179213285446, "time": 0.7971986055374145, "epoch": 9, "memory": 28783, "step": 50064} +{"lr": 1.8594235253127375e-05, "data_time": 0.0021574020385742186, "grad_norm": 0.18429773338139058, "loss": 0.19953390210866928, "time": 0.8015501022338867, "epoch": 9, "memory": 28783, "step": 50164} +{"lr": 1.8594235253127375e-05, "data_time": 0.002024388313293457, "grad_norm": 0.28110020104795697, "loss": 0.1989580750465393, "time": 0.7558328151702881, "epoch": 9, "memory": 28783, "step": 50264} +{"lr": 1.8594235253127375e-05, "data_time": 0.0017935276031494141, "grad_norm": 0.3857104729861021, "loss": 0.1988803744316101, "time": 0.8563242673873901, "epoch": 9, "memory": 28783, "step": 50364} +{"lr": 1.8594235253127375e-05, "data_time": 0.0021255970001220702, "grad_norm": 0.2891829600557685, "loss": 0.2000194400548935, "time": 0.8917397022247314, "epoch": 9, "memory": 28783, "step": 50464} +{"lr": 1.8594235253127375e-05, "data_time": 0.0022556543350219726, "grad_norm": 0.17433524392545224, "loss": 0.1986156314611435, "time": 0.9340741634368896, "epoch": 9, "memory": 28783, "step": 50564} +{"lr": 1.8594235253127375e-05, "data_time": 0.0020201683044433595, "grad_norm": 0.3208316644653678, "loss": 0.20012563169002534, "time": 0.8188866376876831, "epoch": 9, "memory": 28783, "step": 50664} +{"lr": 1.8594235253127375e-05, "data_time": 0.0015877962112426757, "grad_norm": 0.3780954498797655, "loss": 0.19963158220052718, "time": 0.7659719705581665, "epoch": 9, "memory": 28783, "step": 50764} +{"lr": 1.8594235253127375e-05, "data_time": 0.0017437934875488281, "grad_norm": 0.32902325987815856, "loss": 0.19996160864830018, "time": 0.813164496421814, "epoch": 9, "memory": 28783, "step": 50864} +{"lr": 1.8594235253127375e-05, "data_time": 0.0017119646072387695, "grad_norm": 0.1926439231261611, "loss": 0.19912930130958556, "time": 0.8445338249206543, "epoch": 9, "memory": 28783, "step": 50964} +{"lr": 1.8594235253127375e-05, "data_time": 0.001751255989074707, "grad_norm": 0.2375560365850106, "loss": 0.1990738958120346, "time": 0.8966788530349732, "epoch": 9, "memory": 28783, "step": 51064} +{"lr": 1.8594235253127375e-05, "data_time": 0.0017923593521118164, "grad_norm": 0.4424662021920085, "loss": 0.19921157658100128, "time": 0.8450961351394654, "epoch": 9, "memory": 28783, "step": 51164} +{"lr": 1.8594235253127375e-05, "data_time": 0.0016463994979858398, "grad_norm": 0.37272116243839265, "loss": 0.19975984692573548, "time": 0.8395662069320678, "epoch": 9, "memory": 28783, "step": 51264} +{"lr": 1.8594235253127375e-05, "data_time": 0.0029436588287353516, "grad_norm": 0.2957117594778538, "loss": 0.19911667406558992, "time": 0.8630541086196899, "epoch": 9, "memory": 28783, "step": 51364} +{"lr": 1.8594235253127375e-05, "data_time": 0.0017013311386108398, "grad_norm": 0.17027930729091167, "loss": 0.19863860458135604, "time": 0.8604765892028808, "epoch": 9, "memory": 28783, "step": 51464} +{"lr": 1.8594235253127375e-05, "data_time": 0.004299187660217285, "grad_norm": 0.30173583431169393, "loss": 0.1988319382071495, "time": 0.9336992263793945, "epoch": 9, "memory": 28783, "step": 51564} +{"lr": 1.8594235253127375e-05, "data_time": 0.0036576032638549806, "grad_norm": 0.2116669312119484, "loss": 0.19918847531080247, "time": 0.946616244316101, "epoch": 9, "memory": 28783, "step": 51664} +{"lr": 1.8594235253127375e-05, "data_time": 0.0017846584320068359, "grad_norm": 0.354281132761389, "loss": 0.19975424855947493, "time": 0.8132057428359986, "epoch": 9, "memory": 28783, "step": 51764} +{"accuracy/top1": 99.96842956542969, "data_time": 0.20790562910192154, "time": 0.6230652612798354, "step": 9} +{"lr": 1.2202456766718093e-05, "data_time": 0.13533883094787597, "grad_norm": 0.19301287070848047, "loss": 0.19917882829904557, "time": 0.9107456684112549, "epoch": 10, "memory": 28783, "step": 51922} +{"lr": 1.2202456766718093e-05, "data_time": 0.22331948280334474, "grad_norm": 0.32406466929242017, "loss": 0.19940428137779237, "time": 0.8882270336151123, "epoch": 10, "memory": 28783, "step": 52022} +{"lr": 1.2202456766718093e-05, "data_time": 0.21347033977508545, "grad_norm": 0.35621108375489713, "loss": 0.19919017255306243, "time": 0.8834541320800782, "epoch": 10, "memory": 28783, "step": 52122} +{"lr": 1.2202456766718093e-05, "data_time": 0.11416516304016114, "grad_norm": 0.09284001146443188, "loss": 0.19853756725788116, "time": 0.8346410512924194, "epoch": 10, "memory": 28783, "step": 52222} +{"lr": 1.2202456766718093e-05, "data_time": 0.057669997215270996, "grad_norm": 0.22106292573735117, "loss": 0.19868463277816772, "time": 0.7937537908554078, "epoch": 10, "memory": 28783, "step": 52322} +{"lr": 1.2202456766718093e-05, "data_time": 0.0016902685165405273, "grad_norm": 0.3011444129049778, "loss": 0.20097961723804475, "time": 0.8490314483642578, "epoch": 10, "memory": 28783, "step": 52422} +{"lr": 1.2202456766718093e-05, "data_time": 0.22680144309997557, "grad_norm": 0.24370603486895562, "loss": 0.1990018293261528, "time": 0.7836069822311401, "epoch": 10, "memory": 28783, "step": 52522} +{"lr": 1.2202456766718093e-05, "data_time": 0.11063516139984131, "grad_norm": 0.20700901304371655, "loss": 0.19861238002777098, "time": 0.8661819934844971, "epoch": 10, "memory": 28783, "step": 52622} +{"lr": 1.2202456766718093e-05, "data_time": 0.012012052536010741, "grad_norm": 0.2800087394192815, "loss": 0.2002096265554428, "time": 0.8858421325683594, "epoch": 10, "memory": 28783, "step": 52722} +{"lr": 1.2202456766718093e-05, "data_time": 0.0035236835479736327, "grad_norm": 0.21200039803516119, "loss": 0.19939270615577698, "time": 0.942995572090149, "epoch": 10, "memory": 28783, "step": 52822} +{"lr": 1.2202456766718093e-05, "data_time": 0.0017570257186889648, "grad_norm": 0.2766184563515708, "loss": 0.19905015975236892, "time": 0.8427855253219605, "epoch": 10, "memory": 28783, "step": 52922} +{"lr": 1.2202456766718093e-05, "data_time": 0.0020080327987670897, "grad_norm": 0.3116778463125229, "loss": 0.19909278601408004, "time": 0.8508692741394043, "epoch": 10, "memory": 28783, "step": 53022} +{"lr": 1.2202456766718093e-05, "data_time": 0.0016348600387573243, "grad_norm": 0.27893140707165004, "loss": 0.200196073949337, "time": 0.8970194339752198, "epoch": 10, "memory": 28783, "step": 53122} +{"lr": 1.2202456766718093e-05, "data_time": 0.002549886703491211, "grad_norm": 0.16098835570737718, "loss": 0.19857225716114044, "time": 0.9008079767227173, "epoch": 10, "memory": 28783, "step": 53222} +{"lr": 1.2202456766718093e-05, "data_time": 0.002266192436218262, "grad_norm": 0.09969056223053485, "loss": 0.1988989979028702, "time": 0.943049144744873, "epoch": 10, "memory": 28783, "step": 53322} +{"lr": 1.2202456766718093e-05, "data_time": 0.0021249055862426758, "grad_norm": 0.29190522104036065, "loss": 0.19949427545070647, "time": 0.8756329774856567, "epoch": 10, "memory": 28783, "step": 53422} +{"lr": 1.2202456766718093e-05, "data_time": 0.0030753374099731444, "grad_norm": 0.3730745716020465, "loss": 0.19954436421394348, "time": 0.9552780151367187, "epoch": 10, "memory": 28783, "step": 53522} +{"lr": 1.2202456766718093e-05, "data_time": 0.0022222518920898436, "grad_norm": 0.537413590401411, "loss": 0.1994130328297615, "time": 0.9699157476425171, "epoch": 10, "memory": 28783, "step": 53622} +{"lr": 1.2202456766718093e-05, "data_time": 0.0023249387741088867, "grad_norm": 0.20427514445036649, "loss": 0.1986580327153206, "time": 0.9012442111968995, "epoch": 10, "memory": 28783, "step": 53722} +{"lr": 1.2202456766718093e-05, "data_time": 0.0018619775772094726, "grad_norm": 0.15136250564828516, "loss": 0.19909613877534865, "time": 0.7621897697448731, "epoch": 10, "memory": 28783, "step": 53822} +{"lr": 1.2202456766718093e-05, "data_time": 0.0017845630645751953, "grad_norm": 0.1600125606637448, "loss": 0.19864918887615204, "time": 0.7884865045547486, "epoch": 10, "memory": 28783, "step": 53922} +{"lr": 1.2202456766718093e-05, "data_time": 0.0034561634063720705, "grad_norm": 0.16006032302975653, "loss": 0.1993183746933937, "time": 1.0244053602218628, "epoch": 10, "memory": 28783, "step": 54022} +{"lr": 1.2202456766718093e-05, "data_time": 0.0033855676651000977, "grad_norm": 0.22914999574422837, "loss": 0.19899467825889589, "time": 0.9077329635620117, "epoch": 10, "memory": 28783, "step": 54122} +{"lr": 1.2202456766718093e-05, "data_time": 0.0017466306686401366, "grad_norm": 0.49763806434348223, "loss": 0.19974515587091446, "time": 0.8655405282974243, "epoch": 10, "memory": 28783, "step": 54222} +{"lr": 1.2202456766718093e-05, "data_time": 0.0015778541564941406, "grad_norm": 0.24855962507426738, "loss": 0.1993138983845711, "time": 0.8173866510391236, "epoch": 10, "memory": 28783, "step": 54322} +{"lr": 1.2202456766718093e-05, "data_time": 0.003122901916503906, "grad_norm": 0.2541944248601794, "loss": 0.19873177111148835, "time": 0.911636209487915, "epoch": 10, "memory": 28783, "step": 54422} +{"lr": 1.2202456766718093e-05, "data_time": 0.0023128271102905275, "grad_norm": 0.36098565105348823, "loss": 0.20084340125322342, "time": 0.8772715330123901, "epoch": 10, "memory": 28783, "step": 54522} +{"lr": 1.2202456766718093e-05, "data_time": 0.0024342775344848634, "grad_norm": 0.2083258915692568, "loss": 0.1992798313498497, "time": 0.8672708034515381, "epoch": 10, "memory": 28783, "step": 54622} +{"lr": 1.2202456766718093e-05, "data_time": 0.0017805099487304688, "grad_norm": 0.6617534391582012, "loss": 0.19961759597063064, "time": 0.7648499011993408, "epoch": 10, "memory": 28783, "step": 54722} +{"lr": 1.2202456766718093e-05, "data_time": 0.0027037620544433593, "grad_norm": 0.23564027417451144, "loss": 0.19873437136411667, "time": 0.7910962581634522, "epoch": 10, "memory": 28783, "step": 54822} +{"lr": 1.2202456766718093e-05, "data_time": 0.0023592472076416015, "grad_norm": 0.3154816431924701, "loss": 0.19969159811735154, "time": 0.8554802894592285, "epoch": 10, "memory": 28783, "step": 54922} +{"lr": 1.2202456766718093e-05, "data_time": 0.00893406867980957, "grad_norm": 0.17273001032881438, "loss": 0.19944452345371247, "time": 0.8116443157196045, "epoch": 10, "memory": 28783, "step": 55022} +{"lr": 1.2202456766718093e-05, "data_time": 0.1581697940826416, "grad_norm": 0.35414931140840056, "loss": 0.19946963340044022, "time": 0.8220585107803344, "epoch": 10, "memory": 28783, "step": 55122} +{"lr": 1.2202456766718093e-05, "data_time": 0.14808144569396972, "grad_norm": 0.26122904345393183, "loss": 0.19855034351348877, "time": 0.8965449094772339, "epoch": 10, "memory": 28783, "step": 55222} +{"lr": 1.2202456766718093e-05, "data_time": 0.06742925643920898, "grad_norm": 0.3018125146627426, "loss": 0.19951617270708083, "time": 0.9045409440994263, "epoch": 10, "memory": 28783, "step": 55322} +{"lr": 1.2202456766718093e-05, "data_time": 0.0192230224609375, "grad_norm": 0.26598530132323506, "loss": 0.19922469407320023, "time": 0.8526290655136108, "epoch": 10, "memory": 28783, "step": 55422} +{"lr": 1.2202456766718093e-05, "data_time": 0.0017828702926635741, "grad_norm": 0.41142064253799615, "loss": 0.1989261493086815, "time": 0.8375653982162475, "epoch": 10, "memory": 28783, "step": 55522} +{"lr": 1.2202456766718093e-05, "data_time": 0.0022647857666015627, "grad_norm": 0.3614894837141037, "loss": 0.19897508025169372, "time": 0.8872854709625244, "epoch": 10, "memory": 28783, "step": 55622} +{"lr": 1.2202456766718093e-05, "data_time": 0.0026006460189819335, "grad_norm": 0.19555093450471758, "loss": 0.199823234975338, "time": 0.8065153598785401, "epoch": 10, "memory": 28783, "step": 55722} +{"lr": 1.2202456766718093e-05, "data_time": 0.0026729345321655274, "grad_norm": 0.1707129318267107, "loss": 0.19865113347768784, "time": 0.9116674184799194, "epoch": 10, "memory": 28783, "step": 55822} +{"lr": 1.2202456766718093e-05, "data_time": 0.0024576902389526365, "grad_norm": 0.1427366454154253, "loss": 0.19858566522598267, "time": 0.8400616645812988, "epoch": 10, "memory": 28783, "step": 55922} +{"lr": 1.2202456766718093e-05, "data_time": 0.001971149444580078, "grad_norm": 0.28716955725103616, "loss": 0.19943950772285463, "time": 0.8212677478790283, "epoch": 10, "memory": 28783, "step": 56022} +{"lr": 1.2202456766718093e-05, "data_time": 0.0023377180099487306, "grad_norm": 0.15794655294157564, "loss": 0.20017153024673462, "time": 0.9557317972183228, "epoch": 10, "memory": 28783, "step": 56122} +{"lr": 1.2202456766718093e-05, "data_time": 0.004235410690307617, "grad_norm": 0.16744602820836008, "loss": 0.19898456782102586, "time": 0.9123448133468628, "epoch": 10, "memory": 28783, "step": 56222} +{"lr": 1.2202456766718093e-05, "data_time": 0.0017568826675415038, "grad_norm": 0.27650420600548387, "loss": 0.19901863038539885, "time": 0.8098387479782104, "epoch": 10, "memory": 28783, "step": 56322} +{"lr": 1.2202456766718093e-05, "data_time": 0.0019932985305786133, "grad_norm": 0.23942797668278218, "loss": 0.19960306584835052, "time": 0.8894573450088501, "epoch": 10, "memory": 28783, "step": 56422} +{"lr": 1.2202456766718093e-05, "data_time": 0.00220944881439209, "grad_norm": 0.36436382234096526, "loss": 0.20020554661750795, "time": 0.8621434211730957, "epoch": 10, "memory": 28783, "step": 56522} +{"lr": 1.2202456766718093e-05, "data_time": 0.0021554231643676758, "grad_norm": 0.2618440832477063, "loss": 0.19875305145978928, "time": 0.8395252227783203, "epoch": 10, "memory": 28783, "step": 56622} +{"lr": 1.2202456766718093e-05, "data_time": 0.0017965316772460937, "grad_norm": 0.35517810806632044, "loss": 0.1996025949716568, "time": 0.7628729104995727, "epoch": 10, "memory": 28783, "step": 56722} +{"lr": 1.2202456766718093e-05, "data_time": 0.0017314434051513671, "grad_norm": 0.3564167030155659, "loss": 0.19922177642583846, "time": 0.798421835899353, "epoch": 10, "memory": 28783, "step": 56822} +{"lr": 1.2202456766718093e-05, "data_time": 0.009748721122741699, "grad_norm": 0.42114703767001627, "loss": 0.1999643102288246, "time": 0.7876276731491089, "epoch": 10, "memory": 28783, "step": 56922} +{"lr": 1.2202456766718093e-05, "data_time": 0.2023008108139038, "grad_norm": 0.22653089836239815, "loss": 0.1989861860871315, "time": 0.9507332086563111, "epoch": 10, "memory": 28783, "step": 57022} +{"lr": 1.2202456766718093e-05, "data_time": 0.14067411422729492, "grad_norm": 0.28799118380993605, "loss": 0.1986668199300766, "time": 0.8300880670547486, "epoch": 10, "memory": 28783, "step": 57122} +{"lr": 1.2202456766718093e-05, "data_time": 0.06402592658996582, "grad_norm": 0.1465356174390763, "loss": 0.198752561211586, "time": 0.8175986051559448, "epoch": 10, "memory": 28783, "step": 57222} +{"lr": 1.2202456766718093e-05, "data_time": 0.11023647785186767, "grad_norm": 0.5015743039548397, "loss": 0.19983679354190825, "time": 0.8183055400848389, "epoch": 10, "memory": 28783, "step": 57322} +{"lr": 1.2202456766718093e-05, "data_time": 0.015048861503601074, "grad_norm": 0.2388107413891703, "loss": 0.20051872879266738, "time": 0.9866007804870606, "epoch": 10, "memory": 28783, "step": 57422} +{"lr": 1.2202456766718093e-05, "data_time": 0.0037036418914794924, "grad_norm": 1.2485242391005158, "loss": 0.19872440844774247, "time": 0.969190239906311, "epoch": 10, "memory": 28783, "step": 57522} +{"accuracy/top1": 99.96842956542969, "data_time": 0.20385313034057617, "time": 0.6185694161583396, "step": 10} diff --git a/swin_small_4xb256_fake5m_lr1e-4_10epoch/20230601_215742/vis_data/vis_image/0000000.png_1.png b/swin_small_4xb256_fake5m_lr1e-4_10epoch/20230601_215742/vis_data/vis_image/0000000.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..b2aa1aaaf9f97ce8b19f3bbce6b7024e76a7ba4c Binary files /dev/null and b/swin_small_4xb256_fake5m_lr1e-4_10epoch/20230601_215742/vis_data/vis_image/0000000.png_1.png differ diff --git a/swin_small_4xb256_fake5m_lr1e-4_10epoch/20230601_215742/vis_data/vis_image/0000000.png_10.png b/swin_small_4xb256_fake5m_lr1e-4_10epoch/20230601_215742/vis_data/vis_image/0000000.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..b44147a5c0411cb50267bbc3e54b286eff6d6bc7 Binary files /dev/null and b/swin_small_4xb256_fake5m_lr1e-4_10epoch/20230601_215742/vis_data/vis_image/0000000.png_10.png differ diff --git a/swin_small_4xb256_fake5m_lr1e-4_10epoch/20230601_215742/vis_data/vis_image/0000000.png_2.png b/swin_small_4xb256_fake5m_lr1e-4_10epoch/20230601_215742/vis_data/vis_image/0000000.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..b44147a5c0411cb50267bbc3e54b286eff6d6bc7 Binary files /dev/null and b/swin_small_4xb256_fake5m_lr1e-4_10epoch/20230601_215742/vis_data/vis_image/0000000.png_2.png differ diff --git a/swin_small_4xb256_fake5m_lr1e-4_10epoch/20230601_215742/vis_data/vis_image/0000000.png_3.png b/swin_small_4xb256_fake5m_lr1e-4_10epoch/20230601_215742/vis_data/vis_image/0000000.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..b2aa1aaaf9f97ce8b19f3bbce6b7024e76a7ba4c Binary files /dev/null and b/swin_small_4xb256_fake5m_lr1e-4_10epoch/20230601_215742/vis_data/vis_image/0000000.png_3.png differ diff --git a/swin_small_4xb256_fake5m_lr1e-4_10epoch/20230601_215742/vis_data/vis_image/0000000.png_4.png b/swin_small_4xb256_fake5m_lr1e-4_10epoch/20230601_215742/vis_data/vis_image/0000000.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..b44147a5c0411cb50267bbc3e54b286eff6d6bc7 Binary files /dev/null and b/swin_small_4xb256_fake5m_lr1e-4_10epoch/20230601_215742/vis_data/vis_image/0000000.png_4.png differ diff --git a/swin_small_4xb256_fake5m_lr1e-4_10epoch/20230601_215742/vis_data/vis_image/0000000.png_5.png b/swin_small_4xb256_fake5m_lr1e-4_10epoch/20230601_215742/vis_data/vis_image/0000000.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..b44147a5c0411cb50267bbc3e54b286eff6d6bc7 Binary files /dev/null and b/swin_small_4xb256_fake5m_lr1e-4_10epoch/20230601_215742/vis_data/vis_image/0000000.png_5.png differ diff --git a/swin_small_4xb256_fake5m_lr1e-4_10epoch/20230601_215742/vis_data/vis_image/0000000.png_6.png b/swin_small_4xb256_fake5m_lr1e-4_10epoch/20230601_215742/vis_data/vis_image/0000000.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..b44147a5c0411cb50267bbc3e54b286eff6d6bc7 Binary files /dev/null and b/swin_small_4xb256_fake5m_lr1e-4_10epoch/20230601_215742/vis_data/vis_image/0000000.png_6.png differ diff --git a/swin_small_4xb256_fake5m_lr1e-4_10epoch/20230601_215742/vis_data/vis_image/0000000.png_7.png b/swin_small_4xb256_fake5m_lr1e-4_10epoch/20230601_215742/vis_data/vis_image/0000000.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..b44147a5c0411cb50267bbc3e54b286eff6d6bc7 Binary files /dev/null and b/swin_small_4xb256_fake5m_lr1e-4_10epoch/20230601_215742/vis_data/vis_image/0000000.png_7.png differ diff --git a/swin_small_4xb256_fake5m_lr1e-4_10epoch/20230601_215742/vis_data/vis_image/0000000.png_8.png b/swin_small_4xb256_fake5m_lr1e-4_10epoch/20230601_215742/vis_data/vis_image/0000000.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..b44147a5c0411cb50267bbc3e54b286eff6d6bc7 Binary files /dev/null and b/swin_small_4xb256_fake5m_lr1e-4_10epoch/20230601_215742/vis_data/vis_image/0000000.png_8.png differ diff --git a/swin_small_4xb256_fake5m_lr1e-4_10epoch/20230601_215742/vis_data/vis_image/0000000.png_9.png b/swin_small_4xb256_fake5m_lr1e-4_10epoch/20230601_215742/vis_data/vis_image/0000000.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..b44147a5c0411cb50267bbc3e54b286eff6d6bc7 Binary files /dev/null and b/swin_small_4xb256_fake5m_lr1e-4_10epoch/20230601_215742/vis_data/vis_image/0000000.png_9.png differ diff --git a/swin_small_4xb256_fake5m_lr1e-4_10epoch/epoch_1.pth b/swin_small_4xb256_fake5m_lr1e-4_10epoch/epoch_1.pth new file mode 100644 index 0000000000000000000000000000000000000000..5dbef0df9e9123cd3530c8f9b7cf6ab6dcd3e50b --- /dev/null +++ b/swin_small_4xb256_fake5m_lr1e-4_10epoch/epoch_1.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:37a05b676497e607b7c59b5d93257598c0dc78d72ed3f5562deed7c2758fac0b +size 587485405 diff --git a/swin_small_4xb256_fake5m_lr1e-4_10epoch/epoch_10.pth b/swin_small_4xb256_fake5m_lr1e-4_10epoch/epoch_10.pth new file mode 100644 index 0000000000000000000000000000000000000000..16e0889f2e80dca9bc0671520ae1aea4096c2ce5 --- /dev/null +++ b/swin_small_4xb256_fake5m_lr1e-4_10epoch/epoch_10.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:b9f155d29c8eec4e8871778f5c4ba08f00cc30839e4dc6bd81c36d91864b4e4d +size 592192157 diff --git a/swin_small_4xb256_fake5m_lr1e-4_10epoch/epoch_2.pth b/swin_small_4xb256_fake5m_lr1e-4_10epoch/epoch_2.pth new file mode 100644 index 0000000000000000000000000000000000000000..dfad22a9ae635819063adf3deae7f0c8cb60a6d7 --- /dev/null +++ b/swin_small_4xb256_fake5m_lr1e-4_10epoch/epoch_2.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:515ab7deb7062c0e4ab1361b2b38b71d66c3c2a6821da3fc0afda8215780648c +size 588013597 diff --git a/swin_small_4xb256_fake5m_lr1e-4_10epoch/epoch_3.pth b/swin_small_4xb256_fake5m_lr1e-4_10epoch/epoch_3.pth new file mode 100644 index 0000000000000000000000000000000000000000..ee31cbdc69b8ec2267ec4309dbf1dde99aa21228 --- /dev/null +++ b/swin_small_4xb256_fake5m_lr1e-4_10epoch/epoch_3.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:d3b750a64cef8cfb7e1c208eb2325eda93b4b5fc2047078ae30aa4680fcd2a1c +size 588537373 diff --git a/swin_small_4xb256_fake5m_lr1e-4_10epoch/epoch_4.pth b/swin_small_4xb256_fake5m_lr1e-4_10epoch/epoch_4.pth new file mode 100644 index 0000000000000000000000000000000000000000..0b6c0b4c410da56e451c1f349277d44bc9f0f124 --- /dev/null +++ b/swin_small_4xb256_fake5m_lr1e-4_10epoch/epoch_4.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:2e8db116d32b0a51a548e44e213eebb7e23286a8c783c8136017eba03c5317d6 +size 589049821 diff --git a/swin_small_4xb256_fake5m_lr1e-4_10epoch/epoch_5.pth b/swin_small_4xb256_fake5m_lr1e-4_10epoch/epoch_5.pth new file mode 100644 index 0000000000000000000000000000000000000000..96d18d820ef2cc9b9743aeebbf31f8493d7f2928 --- /dev/null +++ b/swin_small_4xb256_fake5m_lr1e-4_10epoch/epoch_5.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:0adf877dd19753ff349e99c27bcbfc3fdd01781151b08fcd8538db7ca80b4a10 +size 589556573 diff --git a/swin_small_4xb256_fake5m_lr1e-4_10epoch/epoch_6.pth b/swin_small_4xb256_fake5m_lr1e-4_10epoch/epoch_6.pth new file mode 100644 index 0000000000000000000000000000000000000000..b1ec459c202ed2860d8c80d04ad6265e86392a30 --- /dev/null +++ b/swin_small_4xb256_fake5m_lr1e-4_10epoch/epoch_6.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:537d5865a3f68512c7768c759c633d52e54bd7d651c3a1648a3c0f4196843ea2 +size 590086109 diff --git a/swin_small_4xb256_fake5m_lr1e-4_10epoch/epoch_7.pth b/swin_small_4xb256_fake5m_lr1e-4_10epoch/epoch_7.pth new file mode 100644 index 0000000000000000000000000000000000000000..cb5ee9344baecbfd37799e349e68fc07aaeadcc2 --- /dev/null +++ b/swin_small_4xb256_fake5m_lr1e-4_10epoch/epoch_7.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:5d3dc92b63b347b420471c191b01819cc9845880ec351bdbf7313737e474af7c +size 590620701 diff --git a/swin_small_4xb256_fake5m_lr1e-4_10epoch/epoch_8.pth b/swin_small_4xb256_fake5m_lr1e-4_10epoch/epoch_8.pth new file mode 100644 index 0000000000000000000000000000000000000000..da8659657ca2df0d22191a475fe50e74cb8c1086 --- /dev/null +++ b/swin_small_4xb256_fake5m_lr1e-4_10epoch/epoch_8.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:1fcd0b0b04f9f3ff6f98fc9d5a510d9ffc722f6590a77b8bf565ce8c3655b17a +size 591154845 diff --git a/swin_small_4xb256_fake5m_lr1e-4_10epoch/epoch_9.pth b/swin_small_4xb256_fake5m_lr1e-4_10epoch/epoch_9.pth new file mode 100644 index 0000000000000000000000000000000000000000..1befd6e4bf2ca9d6f7ec96e2385c398e6b6cafcd --- /dev/null +++ b/swin_small_4xb256_fake5m_lr1e-4_10epoch/epoch_9.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:bd85e9733bc8c31f8a6ef8f104f75a3f79275809a91f4dae7edf505109007dc2 +size 591671005 diff --git a/swin_small_4xb256_fake5m_lr1e-4_10epoch/last_checkpoint b/swin_small_4xb256_fake5m_lr1e-4_10epoch/last_checkpoint new file mode 100644 index 0000000000000000000000000000000000000000..e1d32b49aea35e9b2f717635c19e06771156b974 --- /dev/null +++ b/swin_small_4xb256_fake5m_lr1e-4_10epoch/last_checkpoint @@ -0,0 +1 @@ +/mnt/petrelfs/luzeyu/workspace/fakebench/mmpretrain/workdir/swin_small_4xb256_fake5m_lr1e-4_10epoch/epoch_10.pth \ No newline at end of file diff --git a/swin_small_4xb256_fake5m_lr1e-4_10epoch/swin_small_4xb256_fake5m_lr1e-4_10epoch.py b/swin_small_4xb256_fake5m_lr1e-4_10epoch/swin_small_4xb256_fake5m_lr1e-4_10epoch.py new file mode 100644 index 0000000000000000000000000000000000000000..07c9f4bb104a3c4f894e7a05987a8354fd463e1f --- /dev/null +++ b/swin_small_4xb256_fake5m_lr1e-4_10epoch/swin_small_4xb256_fake5m_lr1e-4_10epoch.py @@ -0,0 +1,175 @@ +optim_wrapper = dict( + optimizer=dict( + type='AdamW', + lr=0.0001, + weight_decay=0.05, + eps=1e-08, + betas=(0.9, 0.999), + _scope_='mmpretrain'), + paramwise_cfg=dict( + norm_decay_mult=0.0, + bias_decay_mult=0.0, + flat_decay_mult=0.0, + custom_keys=dict({ + '.absolute_pos_embed': dict(decay_mult=0.0), + '.relative_position_bias_table': dict(decay_mult=0.0) + })), + type='AmpOptimWrapper', + dtype='bfloat16', + clip_grad=dict(max_norm=5.0)) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=True, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=10, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=1024) +model = dict( + type='ImageClassifier', + backbone=dict( + type='SwinTransformer', arch='small', img_size=224, + drop_path_rate=0.3), + neck=dict(type='GlobalAveragePooling'), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=768, + init_cfg=None, + loss=dict( + type='LabelSmoothLoss', label_smooth_val=0.1, mode='original'), + cal_acc=False), + init_cfg=[ + dict(type='TruncNormal', layer='Linear', std=0.02, bias=0.0), + dict(type='Constant', layer='LayerNorm', val=1.0, bias=0.0) + ], + train_cfg=None) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stablediffusionV1-5R2-dpmsolver-25-5m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/swin_small_4xb256_fake5m_lr1e-4_10epoch' diff --git a/swin_tiny_4xb256_1e-3lr_5m/20230531_183311/20230531_183311.log b/swin_tiny_4xb256_1e-3lr_5m/20230531_183311/20230531_183311.log new file mode 100644 index 0000000000000000000000000000000000000000..25386592b9ba13ee0909e1a83a84cc7f6b79d1fa --- /dev/null +++ b/swin_tiny_4xb256_1e-3lr_5m/20230531_183311/20230531_183311.log @@ -0,0 +1,2939 @@ +2023/05/31 18:33:15 - mmengine - INFO - +------------------------------------------------------------ +System environment: + sys.platform: linux + Python: 3.10.9 (main, Mar 8 2023, 10:47:38) [GCC 11.2.0] + CUDA available: True + numpy_random_seed: 172268957 + GPU 0,1,2,3: NVIDIA A100-SXM4-80GB + CUDA_HOME: /mnt/petrelfs/share/cuda-11.6 + NVCC: Cuda compilation tools, release 11.6, V11.6.124 + GCC: gcc (GCC) 4.8.5 20150623 (Red Hat 4.8.5-44) + PyTorch: 1.13.1 + PyTorch compiling details: PyTorch built with: + - GCC 9.3 + - C++ Version: 201402 + - Intel(R) oneAPI Math Kernel Library Version 2021.4-Product Build 20210904 for Intel(R) 64 architecture applications + - Intel(R) MKL-DNN v2.6.0 (Git Hash 52b5f107dd9cf10910aaa19cb47f3abf9b349815) + - OpenMP 201511 (a.k.a. OpenMP 4.5) + - LAPACK is enabled (usually provided by MKL) + - NNPACK is enabled + - CPU capability usage: AVX2 + - CUDA Runtime 11.6 + - NVCC architecture flags: -gencode;arch=compute_37,code=sm_37;-gencode;arch=compute_50,code=sm_50;-gencode;arch=compute_60,code=sm_60;-gencode;arch=compute_61,code=sm_61;-gencode;arch=compute_70,code=sm_70;-gencode;arch=compute_75,code=sm_75;-gencode;arch=compute_80,code=sm_80;-gencode;arch=compute_86,code=sm_86;-gencode;arch=compute_37,code=compute_37 + - CuDNN 8.3.2 (built against CUDA 11.5) + - Magma 2.6.1 + - Build settings: BLAS_INFO=mkl, BUILD_TYPE=Release, CUDA_VERSION=11.6, CUDNN_VERSION=8.3.2, CXX_COMPILER=/opt/rh/devtoolset-9/root/usr/bin/c++, CXX_FLAGS= -fabi-version=11 -Wno-deprecated -fvisibility-inlines-hidden -DUSE_PTHREADPOOL -fopenmp -DNDEBUG -DUSE_KINETO -DUSE_FBGEMM -DUSE_QNNPACK -DUSE_PYTORCH_QNNPACK -DUSE_XNNPACK -DSYMBOLICATE_MOBILE_DEBUG_HANDLE -DEDGE_PROFILER_USE_KINETO -O2 -fPIC -Wno-narrowing -Wall -Wextra -Werror=return-type -Werror=non-virtual-dtor -Wno-missing-field-initializers -Wno-type-limits -Wno-array-bounds -Wno-unknown-pragmas -Wunused-local-typedefs -Wno-unused-parameter -Wno-unused-function -Wno-unused-result -Wno-strict-overflow -Wno-strict-aliasing -Wno-error=deprecated-declarations -Wno-stringop-overflow -Wno-psabi -Wno-error=pedantic -Wno-error=redundant-decls -Wno-error=old-style-cast -fdiagnostics-color=always -faligned-new -Wno-unused-but-set-variable -Wno-maybe-uninitialized -fno-math-errno -fno-trapping-math -Werror=format -Werror=cast-function-type -Wno-stringop-overflow, LAPACK_INFO=mkl, PERF_WITH_AVX=1, PERF_WITH_AVX2=1, PERF_WITH_AVX512=1, TORCH_VERSION=1.13.1, USE_CUDA=ON, USE_CUDNN=ON, USE_EXCEPTION_PTR=1, USE_GFLAGS=OFF, USE_GLOG=OFF, USE_MKL=ON, USE_MKLDNN=ON, USE_MPI=OFF, USE_NCCL=ON, USE_NNPACK=ON, USE_OPENMP=ON, USE_ROCM=OFF, + + TorchVision: 0.14.1 + OpenCV: 4.7.0 + MMEngine: 0.7.3 + +Runtime environment: + cudnn_benchmark: True + mp_cfg: {'mp_start_method': 'fork', 'opencv_num_threads': 0} + dist_cfg: {'backend': 'nccl'} + seed: None + deterministic: False + Distributed launcher: slurm + Distributed training: True + GPU number: 4 +------------------------------------------------------------ + +2023/05/31 18:33:16 - mmengine - INFO - Config: +optim_wrapper = dict( + optimizer=dict( + type='AdamW', + lr=0.001, + weight_decay=0.05, + eps=1e-08, + betas=(0.9, 0.999), + _scope_='mmpretrain'), + paramwise_cfg=dict( + norm_decay_mult=0.0, + bias_decay_mult=0.0, + flat_decay_mult=0.0, + custom_keys=dict({ + '.absolute_pos_embed': dict(decay_mult=0.0), + '.relative_position_bias_table': dict(decay_mult=0.0) + })), + type='AmpOptimWrapper', + dtype='bfloat16', + clip_grad=dict(max_norm=5.0)) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=True, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=20, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=1024) +model = dict( + type='ImageClassifier', + backbone=dict( + type='SwinTransformer', arch='tiny', img_size=224, drop_path_rate=0.2), + neck=dict(type='GlobalAveragePooling'), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=768, + init_cfg=None, + loss=dict( + type='LabelSmoothLoss', label_smooth_val=0.1, mode='original'), + cal_acc=False), + init_cfg=[ + dict(type='TruncNormal', layer='Linear', std=0.02, bias=0.0), + dict(type='Constant', layer='LayerNorm', val=1.0, bias=0.0) + ], + train_cfg=None) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=128, + num_workers=10, + dataset=dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stablediffusionV1-5R2-dpmsolver-25-5m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/swin_tiny_4xb256_1e-3lr_5m' + +2023/05/31 18:33:19 - mmengine - INFO - Hooks will be executed in the following order: +before_run: +(VERY_HIGH ) RuntimeInfoHook +(BELOW_NORMAL) LoggerHook + -------------------- +before_train: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_train_epoch: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(NORMAL ) DistSamplerSeedHook + -------------------- +before_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook + -------------------- +after_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +after_train_epoch: +(NORMAL ) IterTimerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_val_epoch: +(NORMAL ) IterTimerHook + -------------------- +before_val_iter: +(NORMAL ) IterTimerHook + -------------------- +after_val_iter: +(NORMAL ) IterTimerHook +(NORMAL ) VisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_val_epoch: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +after_train: +(VERY_LOW ) CheckpointHook + -------------------- +before_test_epoch: +(NORMAL ) IterTimerHook + -------------------- +before_test_iter: +(NORMAL ) IterTimerHook + -------------------- +after_test_iter: +(NORMAL ) IterTimerHook +(NORMAL ) VisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_test_epoch: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_run: +(BELOW_NORMAL) LoggerHook + -------------------- +2023/05/31 18:33:48 - mmengine - INFO - paramwise_options -- backbone.patch_embed.projection.bias:weight_decay=0.0 +2023/05/31 18:33:48 - mmengine - INFO - paramwise_options -- backbone.patch_embed.norm.weight:weight_decay=0.0 +2023/05/31 18:33:48 - mmengine - INFO - paramwise_options -- backbone.patch_embed.norm.bias:weight_decay=0.0 +2023/05/31 18:33:48 - mmengine - INFO - paramwise_options -- backbone.stages.0.blocks.0.norm1.weight:weight_decay=0.0 +2023/05/31 18:33:48 - mmengine - INFO - paramwise_options -- backbone.stages.0.blocks.0.norm1.bias:weight_decay=0.0 +2023/05/31 18:33:48 - mmengine - INFO - paramwise_options -- backbone.stages.0.blocks.0.attn.w_msa.relative_position_bias_table:lr=0.001 +2023/05/31 18:33:48 - mmengine - INFO - paramwise_options -- backbone.stages.0.blocks.0.attn.w_msa.relative_position_bias_table:weight_decay=0.0 +2023/05/31 18:33:48 - mmengine - INFO - paramwise_options -- backbone.stages.0.blocks.0.attn.w_msa.relative_position_bias_table:decay_mult=0.0 +2023/05/31 18:33:48 - mmengine - INFO - paramwise_options -- backbone.stages.0.blocks.0.attn.w_msa.qkv.bias:weight_decay=0.0 +2023/05/31 18:33:48 - mmengine - INFO - paramwise_options -- backbone.stages.0.blocks.0.attn.w_msa.proj.bias:weight_decay=0.0 +2023/05/31 18:33:48 - mmengine - INFO - paramwise_options -- backbone.stages.0.blocks.0.norm2.weight:weight_decay=0.0 +2023/05/31 18:33:48 - mmengine - INFO - paramwise_options -- backbone.stages.0.blocks.0.norm2.bias:weight_decay=0.0 +2023/05/31 18:33:48 - mmengine - INFO - paramwise_options -- backbone.stages.0.blocks.0.ffn.layers.0.0.bias:weight_decay=0.0 +2023/05/31 18:33:48 - mmengine - INFO - paramwise_options -- backbone.stages.0.blocks.0.ffn.layers.1.bias:weight_decay=0.0 +2023/05/31 18:33:48 - mmengine - INFO - paramwise_options -- backbone.stages.0.blocks.1.norm1.weight:weight_decay=0.0 +2023/05/31 18:33:48 - mmengine - INFO - paramwise_options -- backbone.stages.0.blocks.1.norm1.bias:weight_decay=0.0 +2023/05/31 18:33:48 - mmengine - INFO - paramwise_options -- backbone.stages.0.blocks.1.attn.w_msa.relative_position_bias_table:lr=0.001 +2023/05/31 18:33:48 - mmengine - INFO - paramwise_options -- backbone.stages.0.blocks.1.attn.w_msa.relative_position_bias_table:weight_decay=0.0 +2023/05/31 18:33:48 - mmengine - INFO - paramwise_options -- backbone.stages.0.blocks.1.attn.w_msa.relative_position_bias_table:decay_mult=0.0 +2023/05/31 18:33:48 - mmengine - INFO - paramwise_options -- backbone.stages.0.blocks.1.attn.w_msa.qkv.bias:weight_decay=0.0 +2023/05/31 18:33:48 - mmengine - INFO - paramwise_options -- backbone.stages.0.blocks.1.attn.w_msa.proj.bias:weight_decay=0.0 +2023/05/31 18:33:48 - mmengine - INFO - paramwise_options -- backbone.stages.0.blocks.1.norm2.weight:weight_decay=0.0 +2023/05/31 18:33:48 - mmengine - INFO - paramwise_options -- backbone.stages.0.blocks.1.norm2.bias:weight_decay=0.0 +2023/05/31 18:33:48 - mmengine - INFO - paramwise_options -- backbone.stages.0.blocks.1.ffn.layers.0.0.bias:weight_decay=0.0 +2023/05/31 18:33:48 - mmengine - INFO - paramwise_options -- backbone.stages.0.blocks.1.ffn.layers.1.bias:weight_decay=0.0 +2023/05/31 18:33:48 - mmengine - INFO - paramwise_options -- backbone.stages.0.downsample.norm.weight:weight_decay=0.0 +2023/05/31 18:33:48 - mmengine - INFO - paramwise_options -- backbone.stages.0.downsample.norm.bias:weight_decay=0.0 +2023/05/31 18:33:48 - mmengine - INFO - paramwise_options -- backbone.stages.1.blocks.0.norm1.weight:weight_decay=0.0 +2023/05/31 18:33:48 - mmengine - INFO - paramwise_options -- backbone.stages.1.blocks.0.norm1.bias:weight_decay=0.0 +2023/05/31 18:33:48 - mmengine - INFO - paramwise_options -- backbone.stages.1.blocks.0.attn.w_msa.relative_position_bias_table:lr=0.001 +2023/05/31 18:33:48 - mmengine - INFO - paramwise_options -- backbone.stages.1.blocks.0.attn.w_msa.relative_position_bias_table:weight_decay=0.0 +2023/05/31 18:33:48 - mmengine - INFO - paramwise_options -- backbone.stages.1.blocks.0.attn.w_msa.relative_position_bias_table:decay_mult=0.0 +2023/05/31 18:33:48 - mmengine - INFO - paramwise_options -- backbone.stages.1.blocks.0.attn.w_msa.qkv.bias:weight_decay=0.0 +2023/05/31 18:33:48 - mmengine - INFO - paramwise_options -- backbone.stages.1.blocks.0.attn.w_msa.proj.bias:weight_decay=0.0 +2023/05/31 18:33:48 - mmengine - INFO - paramwise_options -- backbone.stages.1.blocks.0.norm2.weight:weight_decay=0.0 +2023/05/31 18:33:48 - mmengine - INFO - paramwise_options -- backbone.stages.1.blocks.0.norm2.bias:weight_decay=0.0 +2023/05/31 18:33:48 - mmengine - INFO - paramwise_options -- backbone.stages.1.blocks.0.ffn.layers.0.0.bias:weight_decay=0.0 +2023/05/31 18:33:48 - mmengine - INFO - paramwise_options -- backbone.stages.1.blocks.0.ffn.layers.1.bias:weight_decay=0.0 +2023/05/31 18:33:48 - mmengine - INFO - paramwise_options -- backbone.stages.1.blocks.1.norm1.weight:weight_decay=0.0 +2023/05/31 18:33:48 - mmengine - INFO - paramwise_options -- backbone.stages.1.blocks.1.norm1.bias:weight_decay=0.0 +2023/05/31 18:33:48 - mmengine - INFO - paramwise_options -- backbone.stages.1.blocks.1.attn.w_msa.relative_position_bias_table:lr=0.001 +2023/05/31 18:33:48 - mmengine - INFO - paramwise_options -- backbone.stages.1.blocks.1.attn.w_msa.relative_position_bias_table:weight_decay=0.0 +2023/05/31 18:33:48 - mmengine - INFO - paramwise_options -- backbone.stages.1.blocks.1.attn.w_msa.relative_position_bias_table:decay_mult=0.0 +2023/05/31 18:33:48 - mmengine - INFO - paramwise_options -- backbone.stages.1.blocks.1.attn.w_msa.qkv.bias:weight_decay=0.0 +2023/05/31 18:33:48 - mmengine - INFO - paramwise_options -- backbone.stages.1.blocks.1.attn.w_msa.proj.bias:weight_decay=0.0 +2023/05/31 18:33:48 - mmengine - INFO - paramwise_options -- backbone.stages.1.blocks.1.norm2.weight:weight_decay=0.0 +2023/05/31 18:33:48 - mmengine - INFO - paramwise_options -- backbone.stages.1.blocks.1.norm2.bias:weight_decay=0.0 +2023/05/31 18:33:48 - mmengine - INFO - paramwise_options -- backbone.stages.1.blocks.1.ffn.layers.0.0.bias:weight_decay=0.0 +2023/05/31 18:33:48 - mmengine - INFO - paramwise_options -- backbone.stages.1.blocks.1.ffn.layers.1.bias:weight_decay=0.0 +2023/05/31 18:33:48 - mmengine - INFO - paramwise_options -- backbone.stages.1.downsample.norm.weight:weight_decay=0.0 +2023/05/31 18:33:48 - mmengine - INFO - paramwise_options -- backbone.stages.1.downsample.norm.bias:weight_decay=0.0 +2023/05/31 18:33:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.0.norm1.weight:weight_decay=0.0 +2023/05/31 18:33:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.0.norm1.bias:weight_decay=0.0 +2023/05/31 18:33:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.0.attn.w_msa.relative_position_bias_table:lr=0.001 +2023/05/31 18:33:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.0.attn.w_msa.relative_position_bias_table:weight_decay=0.0 +2023/05/31 18:33:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.0.attn.w_msa.relative_position_bias_table:decay_mult=0.0 +2023/05/31 18:33:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.0.attn.w_msa.qkv.bias:weight_decay=0.0 +2023/05/31 18:33:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.0.attn.w_msa.proj.bias:weight_decay=0.0 +2023/05/31 18:33:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.0.norm2.weight:weight_decay=0.0 +2023/05/31 18:33:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.0.norm2.bias:weight_decay=0.0 +2023/05/31 18:33:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.0.ffn.layers.0.0.bias:weight_decay=0.0 +2023/05/31 18:33:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.0.ffn.layers.1.bias:weight_decay=0.0 +2023/05/31 18:33:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.1.norm1.weight:weight_decay=0.0 +2023/05/31 18:33:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.1.norm1.bias:weight_decay=0.0 +2023/05/31 18:33:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.1.attn.w_msa.relative_position_bias_table:lr=0.001 +2023/05/31 18:33:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.1.attn.w_msa.relative_position_bias_table:weight_decay=0.0 +2023/05/31 18:33:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.1.attn.w_msa.relative_position_bias_table:decay_mult=0.0 +2023/05/31 18:33:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.1.attn.w_msa.qkv.bias:weight_decay=0.0 +2023/05/31 18:33:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.1.attn.w_msa.proj.bias:weight_decay=0.0 +2023/05/31 18:33:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.1.norm2.weight:weight_decay=0.0 +2023/05/31 18:33:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.1.norm2.bias:weight_decay=0.0 +2023/05/31 18:33:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.1.ffn.layers.0.0.bias:weight_decay=0.0 +2023/05/31 18:33:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.1.ffn.layers.1.bias:weight_decay=0.0 +2023/05/31 18:33:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.2.norm1.weight:weight_decay=0.0 +2023/05/31 18:33:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.2.norm1.bias:weight_decay=0.0 +2023/05/31 18:33:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.2.attn.w_msa.relative_position_bias_table:lr=0.001 +2023/05/31 18:33:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.2.attn.w_msa.relative_position_bias_table:weight_decay=0.0 +2023/05/31 18:33:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.2.attn.w_msa.relative_position_bias_table:decay_mult=0.0 +2023/05/31 18:33:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.2.attn.w_msa.qkv.bias:weight_decay=0.0 +2023/05/31 18:33:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.2.attn.w_msa.proj.bias:weight_decay=0.0 +2023/05/31 18:33:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.2.norm2.weight:weight_decay=0.0 +2023/05/31 18:33:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.2.norm2.bias:weight_decay=0.0 +2023/05/31 18:33:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.2.ffn.layers.0.0.bias:weight_decay=0.0 +2023/05/31 18:33:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.2.ffn.layers.1.bias:weight_decay=0.0 +2023/05/31 18:33:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.3.norm1.weight:weight_decay=0.0 +2023/05/31 18:33:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.3.norm1.bias:weight_decay=0.0 +2023/05/31 18:33:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.3.attn.w_msa.relative_position_bias_table:lr=0.001 +2023/05/31 18:33:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.3.attn.w_msa.relative_position_bias_table:weight_decay=0.0 +2023/05/31 18:33:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.3.attn.w_msa.relative_position_bias_table:decay_mult=0.0 +2023/05/31 18:33:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.3.attn.w_msa.qkv.bias:weight_decay=0.0 +2023/05/31 18:33:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.3.attn.w_msa.proj.bias:weight_decay=0.0 +2023/05/31 18:33:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.3.norm2.weight:weight_decay=0.0 +2023/05/31 18:33:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.3.norm2.bias:weight_decay=0.0 +2023/05/31 18:33:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.3.ffn.layers.0.0.bias:weight_decay=0.0 +2023/05/31 18:33:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.3.ffn.layers.1.bias:weight_decay=0.0 +2023/05/31 18:33:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.4.norm1.weight:weight_decay=0.0 +2023/05/31 18:33:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.4.norm1.bias:weight_decay=0.0 +2023/05/31 18:33:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.4.attn.w_msa.relative_position_bias_table:lr=0.001 +2023/05/31 18:33:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.4.attn.w_msa.relative_position_bias_table:weight_decay=0.0 +2023/05/31 18:33:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.4.attn.w_msa.relative_position_bias_table:decay_mult=0.0 +2023/05/31 18:33:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.4.attn.w_msa.qkv.bias:weight_decay=0.0 +2023/05/31 18:33:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.4.attn.w_msa.proj.bias:weight_decay=0.0 +2023/05/31 18:33:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.4.norm2.weight:weight_decay=0.0 +2023/05/31 18:33:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.4.norm2.bias:weight_decay=0.0 +2023/05/31 18:33:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.4.ffn.layers.0.0.bias:weight_decay=0.0 +2023/05/31 18:33:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.4.ffn.layers.1.bias:weight_decay=0.0 +2023/05/31 18:33:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.5.norm1.weight:weight_decay=0.0 +2023/05/31 18:33:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.5.norm1.bias:weight_decay=0.0 +2023/05/31 18:33:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.5.attn.w_msa.relative_position_bias_table:lr=0.001 +2023/05/31 18:33:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.5.attn.w_msa.relative_position_bias_table:weight_decay=0.0 +2023/05/31 18:33:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.5.attn.w_msa.relative_position_bias_table:decay_mult=0.0 +2023/05/31 18:33:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.5.attn.w_msa.qkv.bias:weight_decay=0.0 +2023/05/31 18:33:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.5.attn.w_msa.proj.bias:weight_decay=0.0 +2023/05/31 18:33:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.5.norm2.weight:weight_decay=0.0 +2023/05/31 18:33:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.5.norm2.bias:weight_decay=0.0 +2023/05/31 18:33:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.5.ffn.layers.0.0.bias:weight_decay=0.0 +2023/05/31 18:33:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.blocks.5.ffn.layers.1.bias:weight_decay=0.0 +2023/05/31 18:33:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.downsample.norm.weight:weight_decay=0.0 +2023/05/31 18:33:48 - mmengine - INFO - paramwise_options -- backbone.stages.2.downsample.norm.bias:weight_decay=0.0 +2023/05/31 18:33:48 - mmengine - INFO - paramwise_options -- backbone.stages.3.blocks.0.norm1.weight:weight_decay=0.0 +2023/05/31 18:33:48 - mmengine - INFO - paramwise_options -- backbone.stages.3.blocks.0.norm1.bias:weight_decay=0.0 +2023/05/31 18:33:48 - mmengine - INFO - paramwise_options -- backbone.stages.3.blocks.0.attn.w_msa.relative_position_bias_table:lr=0.001 +2023/05/31 18:33:48 - mmengine - INFO - paramwise_options -- backbone.stages.3.blocks.0.attn.w_msa.relative_position_bias_table:weight_decay=0.0 +2023/05/31 18:33:48 - mmengine - INFO - paramwise_options -- backbone.stages.3.blocks.0.attn.w_msa.relative_position_bias_table:decay_mult=0.0 +2023/05/31 18:33:48 - mmengine - INFO - paramwise_options -- backbone.stages.3.blocks.0.attn.w_msa.qkv.bias:weight_decay=0.0 +2023/05/31 18:33:48 - mmengine - INFO - paramwise_options -- backbone.stages.3.blocks.0.attn.w_msa.proj.bias:weight_decay=0.0 +2023/05/31 18:33:48 - mmengine - INFO - paramwise_options -- backbone.stages.3.blocks.0.norm2.weight:weight_decay=0.0 +2023/05/31 18:33:48 - mmengine - INFO - paramwise_options -- backbone.stages.3.blocks.0.norm2.bias:weight_decay=0.0 +2023/05/31 18:33:48 - mmengine - INFO - paramwise_options -- backbone.stages.3.blocks.0.ffn.layers.0.0.bias:weight_decay=0.0 +2023/05/31 18:33:48 - mmengine - INFO - paramwise_options -- backbone.stages.3.blocks.0.ffn.layers.1.bias:weight_decay=0.0 +2023/05/31 18:33:48 - mmengine - INFO - paramwise_options -- backbone.stages.3.blocks.1.norm1.weight:weight_decay=0.0 +2023/05/31 18:33:48 - mmengine - INFO - paramwise_options -- backbone.stages.3.blocks.1.norm1.bias:weight_decay=0.0 +2023/05/31 18:33:48 - mmengine - INFO - paramwise_options -- backbone.stages.3.blocks.1.attn.w_msa.relative_position_bias_table:lr=0.001 +2023/05/31 18:33:48 - mmengine - INFO - paramwise_options -- backbone.stages.3.blocks.1.attn.w_msa.relative_position_bias_table:weight_decay=0.0 +2023/05/31 18:33:48 - mmengine - INFO - paramwise_options -- backbone.stages.3.blocks.1.attn.w_msa.relative_position_bias_table:decay_mult=0.0 +2023/05/31 18:33:48 - mmengine - INFO - paramwise_options -- backbone.stages.3.blocks.1.attn.w_msa.qkv.bias:weight_decay=0.0 +2023/05/31 18:33:48 - mmengine - INFO - paramwise_options -- backbone.stages.3.blocks.1.attn.w_msa.proj.bias:weight_decay=0.0 +2023/05/31 18:33:48 - mmengine - INFO - paramwise_options -- backbone.stages.3.blocks.1.norm2.weight:weight_decay=0.0 +2023/05/31 18:33:48 - mmengine - INFO - paramwise_options -- backbone.stages.3.blocks.1.norm2.bias:weight_decay=0.0 +2023/05/31 18:33:48 - mmengine - INFO - paramwise_options -- backbone.stages.3.blocks.1.ffn.layers.0.0.bias:weight_decay=0.0 +2023/05/31 18:33:48 - mmengine - INFO - paramwise_options -- backbone.stages.3.blocks.1.ffn.layers.1.bias:weight_decay=0.0 +2023/05/31 18:33:48 - mmengine - INFO - paramwise_options -- backbone.norm3.weight:weight_decay=0.0 +2023/05/31 18:33:48 - mmengine - INFO - paramwise_options -- backbone.norm3.bias:weight_decay=0.0 +2023/05/31 18:33:48 - mmengine - INFO - paramwise_options -- head.fc.bias:weight_decay=0.0 +Name of parameter - Initialization information + +backbone.patch_embed.projection.weight - torch.Size([96, 3, 4, 4]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.patch_embed.projection.bias - torch.Size([96]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.patch_embed.norm.weight - torch.Size([96]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.patch_embed.norm.bias - torch.Size([96]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.0.blocks.0.norm1.weight - torch.Size([96]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.0.blocks.0.norm1.bias - torch.Size([96]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.0.blocks.0.attn.w_msa.relative_position_bias_table - torch.Size([169, 3]): +Initialized by user-defined `init_weights` in WindowMSA + +backbone.stages.0.blocks.0.attn.w_msa.qkv.weight - torch.Size([288, 96]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.blocks.0.attn.w_msa.qkv.bias - torch.Size([288]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.blocks.0.attn.w_msa.proj.weight - torch.Size([96, 96]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.blocks.0.attn.w_msa.proj.bias - torch.Size([96]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.blocks.0.norm2.weight - torch.Size([96]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.0.blocks.0.norm2.bias - torch.Size([96]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.0.blocks.0.ffn.layers.0.0.weight - torch.Size([384, 96]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.blocks.0.ffn.layers.0.0.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.blocks.0.ffn.layers.1.weight - torch.Size([96, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.blocks.0.ffn.layers.1.bias - torch.Size([96]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.blocks.1.norm1.weight - torch.Size([96]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.0.blocks.1.norm1.bias - torch.Size([96]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.0.blocks.1.attn.w_msa.relative_position_bias_table - torch.Size([169, 3]): +Initialized by user-defined `init_weights` in WindowMSA + +backbone.stages.0.blocks.1.attn.w_msa.qkv.weight - torch.Size([288, 96]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.blocks.1.attn.w_msa.qkv.bias - torch.Size([288]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.blocks.1.attn.w_msa.proj.weight - torch.Size([96, 96]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.blocks.1.attn.w_msa.proj.bias - torch.Size([96]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.blocks.1.norm2.weight - torch.Size([96]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.0.blocks.1.norm2.bias - torch.Size([96]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.0.blocks.1.ffn.layers.0.0.weight - torch.Size([384, 96]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.blocks.1.ffn.layers.0.0.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.blocks.1.ffn.layers.1.weight - torch.Size([96, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.blocks.1.ffn.layers.1.bias - torch.Size([96]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.0.downsample.norm.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.0.downsample.norm.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.0.downsample.reduction.weight - torch.Size([192, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.blocks.0.norm1.weight - torch.Size([192]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.1.blocks.0.norm1.bias - torch.Size([192]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.1.blocks.0.attn.w_msa.relative_position_bias_table - torch.Size([169, 6]): +Initialized by user-defined `init_weights` in WindowMSA + +backbone.stages.1.blocks.0.attn.w_msa.qkv.weight - torch.Size([576, 192]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.blocks.0.attn.w_msa.qkv.bias - torch.Size([576]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.blocks.0.attn.w_msa.proj.weight - torch.Size([192, 192]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.blocks.0.attn.w_msa.proj.bias - torch.Size([192]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.blocks.0.norm2.weight - torch.Size([192]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.1.blocks.0.norm2.bias - torch.Size([192]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.1.blocks.0.ffn.layers.0.0.weight - torch.Size([768, 192]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.blocks.0.ffn.layers.0.0.bias - torch.Size([768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.blocks.0.ffn.layers.1.weight - torch.Size([192, 768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.blocks.0.ffn.layers.1.bias - torch.Size([192]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.blocks.1.norm1.weight - torch.Size([192]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.1.blocks.1.norm1.bias - torch.Size([192]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.1.blocks.1.attn.w_msa.relative_position_bias_table - torch.Size([169, 6]): +Initialized by user-defined `init_weights` in WindowMSA + +backbone.stages.1.blocks.1.attn.w_msa.qkv.weight - torch.Size([576, 192]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.blocks.1.attn.w_msa.qkv.bias - torch.Size([576]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.blocks.1.attn.w_msa.proj.weight - torch.Size([192, 192]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.blocks.1.attn.w_msa.proj.bias - torch.Size([192]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.blocks.1.norm2.weight - torch.Size([192]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.1.blocks.1.norm2.bias - torch.Size([192]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.1.blocks.1.ffn.layers.0.0.weight - torch.Size([768, 192]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.blocks.1.ffn.layers.0.0.bias - torch.Size([768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.blocks.1.ffn.layers.1.weight - torch.Size([192, 768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.blocks.1.ffn.layers.1.bias - torch.Size([192]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.1.downsample.norm.weight - torch.Size([768]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.1.downsample.norm.bias - torch.Size([768]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.1.downsample.reduction.weight - torch.Size([384, 768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.0.norm1.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.0.norm1.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.0.attn.w_msa.relative_position_bias_table - torch.Size([169, 12]): +Initialized by user-defined `init_weights` in WindowMSA + +backbone.stages.2.blocks.0.attn.w_msa.qkv.weight - torch.Size([1152, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.0.attn.w_msa.qkv.bias - torch.Size([1152]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.0.attn.w_msa.proj.weight - torch.Size([384, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.0.attn.w_msa.proj.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.0.norm2.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.0.norm2.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.0.ffn.layers.0.0.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.0.ffn.layers.0.0.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.0.ffn.layers.1.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.0.ffn.layers.1.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.1.norm1.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.1.norm1.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.1.attn.w_msa.relative_position_bias_table - torch.Size([169, 12]): +Initialized by user-defined `init_weights` in WindowMSA + +backbone.stages.2.blocks.1.attn.w_msa.qkv.weight - torch.Size([1152, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.1.attn.w_msa.qkv.bias - torch.Size([1152]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.1.attn.w_msa.proj.weight - torch.Size([384, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.1.attn.w_msa.proj.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.1.norm2.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.1.norm2.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.1.ffn.layers.0.0.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.1.ffn.layers.0.0.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.1.ffn.layers.1.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.1.ffn.layers.1.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.2.norm1.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.2.norm1.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.2.attn.w_msa.relative_position_bias_table - torch.Size([169, 12]): +Initialized by user-defined `init_weights` in WindowMSA + +backbone.stages.2.blocks.2.attn.w_msa.qkv.weight - torch.Size([1152, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.2.attn.w_msa.qkv.bias - torch.Size([1152]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.2.attn.w_msa.proj.weight - torch.Size([384, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.2.attn.w_msa.proj.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.2.norm2.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.2.norm2.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.2.ffn.layers.0.0.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.2.ffn.layers.0.0.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.2.ffn.layers.1.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.2.ffn.layers.1.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.3.norm1.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.3.norm1.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.3.attn.w_msa.relative_position_bias_table - torch.Size([169, 12]): +Initialized by user-defined `init_weights` in WindowMSA + +backbone.stages.2.blocks.3.attn.w_msa.qkv.weight - torch.Size([1152, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.3.attn.w_msa.qkv.bias - torch.Size([1152]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.3.attn.w_msa.proj.weight - torch.Size([384, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.3.attn.w_msa.proj.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.3.norm2.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.3.norm2.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.3.ffn.layers.0.0.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.3.ffn.layers.0.0.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.3.ffn.layers.1.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.3.ffn.layers.1.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.4.norm1.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.4.norm1.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.4.attn.w_msa.relative_position_bias_table - torch.Size([169, 12]): +Initialized by user-defined `init_weights` in WindowMSA + +backbone.stages.2.blocks.4.attn.w_msa.qkv.weight - torch.Size([1152, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.4.attn.w_msa.qkv.bias - torch.Size([1152]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.4.attn.w_msa.proj.weight - torch.Size([384, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.4.attn.w_msa.proj.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.4.norm2.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.4.norm2.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.4.ffn.layers.0.0.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.4.ffn.layers.0.0.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.4.ffn.layers.1.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.4.ffn.layers.1.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.5.norm1.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.5.norm1.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.5.attn.w_msa.relative_position_bias_table - torch.Size([169, 12]): +Initialized by user-defined `init_weights` in WindowMSA + +backbone.stages.2.blocks.5.attn.w_msa.qkv.weight - torch.Size([1152, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.5.attn.w_msa.qkv.bias - torch.Size([1152]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.5.attn.w_msa.proj.weight - torch.Size([384, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.5.attn.w_msa.proj.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.5.norm2.weight - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.5.norm2.bias - torch.Size([384]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.blocks.5.ffn.layers.0.0.weight - torch.Size([1536, 384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.5.ffn.layers.0.0.bias - torch.Size([1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.5.ffn.layers.1.weight - torch.Size([384, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.blocks.5.ffn.layers.1.bias - torch.Size([384]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.2.downsample.norm.weight - torch.Size([1536]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.downsample.norm.bias - torch.Size([1536]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.2.downsample.reduction.weight - torch.Size([768, 1536]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.blocks.0.norm1.weight - torch.Size([768]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.3.blocks.0.norm1.bias - torch.Size([768]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.3.blocks.0.attn.w_msa.relative_position_bias_table - torch.Size([169, 24]): +Initialized by user-defined `init_weights` in WindowMSA + +backbone.stages.3.blocks.0.attn.w_msa.qkv.weight - torch.Size([2304, 768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.blocks.0.attn.w_msa.qkv.bias - torch.Size([2304]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.blocks.0.attn.w_msa.proj.weight - torch.Size([768, 768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.blocks.0.attn.w_msa.proj.bias - torch.Size([768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.blocks.0.norm2.weight - torch.Size([768]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.3.blocks.0.norm2.bias - torch.Size([768]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.3.blocks.0.ffn.layers.0.0.weight - torch.Size([3072, 768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.blocks.0.ffn.layers.0.0.bias - torch.Size([3072]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.blocks.0.ffn.layers.1.weight - torch.Size([768, 3072]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.blocks.0.ffn.layers.1.bias - torch.Size([768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.blocks.1.norm1.weight - torch.Size([768]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.3.blocks.1.norm1.bias - torch.Size([768]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.3.blocks.1.attn.w_msa.relative_position_bias_table - torch.Size([169, 24]): +Initialized by user-defined `init_weights` in WindowMSA + +backbone.stages.3.blocks.1.attn.w_msa.qkv.weight - torch.Size([2304, 768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.blocks.1.attn.w_msa.qkv.bias - torch.Size([2304]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.blocks.1.attn.w_msa.proj.weight - torch.Size([768, 768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.blocks.1.attn.w_msa.proj.bias - torch.Size([768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.blocks.1.norm2.weight - torch.Size([768]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.3.blocks.1.norm2.bias - torch.Size([768]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.stages.3.blocks.1.ffn.layers.0.0.weight - torch.Size([3072, 768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.blocks.1.ffn.layers.0.0.bias - torch.Size([3072]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.blocks.1.ffn.layers.1.weight - torch.Size([768, 3072]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.stages.3.blocks.1.ffn.layers.1.bias - torch.Size([768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +backbone.norm3.weight - torch.Size([768]): +The value is the same before and after calling `init_weights` of ImageClassifier + +backbone.norm3.bias - torch.Size([768]): +The value is the same before and after calling `init_weights` of ImageClassifier + +head.fc.weight - torch.Size([2, 768]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 + +head.fc.bias - torch.Size([2]): +TruncNormalInit: a=-2, b=2, mean=0, std=0.02, bias=0.0 +2023/05/31 18:33:48 - mmengine - WARNING - "FileClient" will be deprecated in future. Please use io functions in https://mmengine.readthedocs.io/en/latest/api/fileio.html#file-io +2023/05/31 18:33:48 - mmengine - WARNING - "HardDiskBackend" is the alias of "LocalBackend" and the former will be deprecated in future. +2023/05/31 18:33:48 - mmengine - INFO - Checkpoints will be saved to /mnt/petrelfs/luzeyu/workspace/fakebench/mmpretrain/workdir/swin_tiny_4xb256_1e-3lr_5m. +2023/05/31 18:34:40 - mmengine - INFO - Epoch(train) [1][ 100/11516] lr: 1.0000e-03 eta: 1 day, 8:50:58 time: 0.4269 data_time: 0.1916 memory: 9474 grad_norm: 0.8052 loss: 0.6850 +2023/05/31 18:35:25 - mmengine - INFO - Epoch(train) [1][ 200/11516] lr: 1.0000e-03 eta: 1 day, 6:43:41 time: 0.4384 data_time: 0.0010 memory: 9474 grad_norm: 1.2080 loss: 0.6879 +2023/05/31 18:36:14 - mmengine - INFO - Epoch(train) [1][ 300/11516] lr: 1.0000e-03 eta: 1 day, 6:57:21 time: 0.4772 data_time: 0.0011 memory: 9474 grad_norm: 1.2378 loss: 0.6797 +2023/05/31 18:37:00 - mmengine - INFO - Epoch(train) [1][ 400/11516] lr: 1.0000e-03 eta: 1 day, 6:36:00 time: 0.4897 data_time: 0.0011 memory: 9474 grad_norm: 0.7536 loss: 0.6640 +2023/05/31 18:37:46 - mmengine - INFO - Epoch(train) [1][ 500/11516] lr: 1.0000e-03 eta: 1 day, 6:17:03 time: 0.4923 data_time: 0.0011 memory: 9474 grad_norm: 2.4277 loss: 0.6726 +2023/05/31 18:38:31 - mmengine - INFO - Epoch(train) [1][ 600/11516] lr: 1.0000e-03 eta: 1 day, 6:03:27 time: 0.5169 data_time: 0.0010 memory: 9474 grad_norm: 0.8478 loss: 0.6811 +2023/05/31 18:39:19 - mmengine - INFO - Epoch(train) [1][ 700/11516] lr: 1.0000e-03 eta: 1 day, 6:08:08 time: 0.4830 data_time: 0.0012 memory: 9474 grad_norm: 0.7502 loss: 0.6688 +2023/05/31 18:40:06 - mmengine - INFO - Epoch(train) [1][ 800/11516] lr: 1.0000e-03 eta: 1 day, 6:04:18 time: 0.4418 data_time: 0.0011 memory: 9474 grad_norm: 0.6410 loss: 0.6691 +2023/05/31 18:40:51 - mmengine - INFO - Epoch(train) [1][ 900/11516] lr: 1.0000e-03 eta: 1 day, 5:56:18 time: 0.4214 data_time: 0.0010 memory: 9474 grad_norm: 1.9923 loss: 0.6726 +2023/05/31 18:41:37 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/05/31 18:41:37 - mmengine - INFO - Epoch(train) [1][ 1000/11516] lr: 1.0000e-03 eta: 1 day, 5:50:59 time: 0.4638 data_time: 0.0013 memory: 9474 grad_norm: 0.9893 loss: 0.6661 +2023/05/31 18:42:23 - mmengine - INFO - Epoch(train) [1][ 1100/11516] lr: 1.0000e-03 eta: 1 day, 5:45:59 time: 0.4614 data_time: 0.0010 memory: 9474 grad_norm: 8.3054 loss: 0.6420 +2023/05/31 18:43:08 - mmengine - INFO - Epoch(train) [1][ 1200/11516] lr: 1.0000e-03 eta: 1 day, 5:41:09 time: 0.4384 data_time: 0.0011 memory: 9474 grad_norm: 0.7923 loss: 0.6670 +2023/05/31 18:43:53 - mmengine - INFO - Epoch(train) [1][ 1300/11516] lr: 1.0000e-03 eta: 1 day, 5:34:16 time: 0.3878 data_time: 0.0010 memory: 9474 grad_norm: 0.6782 loss: 0.6481 +2023/05/31 18:44:40 - mmengine - INFO - Epoch(train) [1][ 1400/11516] lr: 1.0000e-03 eta: 1 day, 5:34:43 time: 0.4624 data_time: 0.0010 memory: 9474 grad_norm: 0.5881 loss: 0.6777 +2023/05/31 18:45:25 - mmengine - INFO - Epoch(train) [1][ 1500/11516] lr: 1.0000e-03 eta: 1 day, 5:30:54 time: 0.4059 data_time: 0.0010 memory: 9474 grad_norm: 0.5484 loss: 0.6642 +2023/05/31 18:46:11 - mmengine - INFO - Epoch(train) [1][ 1600/11516] lr: 1.0000e-03 eta: 1 day, 5:29:13 time: 0.4653 data_time: 0.0010 memory: 9474 grad_norm: 0.5723 loss: 0.6534 +2023/05/31 18:46:55 - mmengine - INFO - Epoch(train) [1][ 1700/11516] lr: 1.0000e-03 eta: 1 day, 5:23:32 time: 0.4096 data_time: 0.0010 memory: 9474 grad_norm: 1.8854 loss: 0.6610 +2023/05/31 18:47:44 - mmengine - INFO - Epoch(train) [1][ 1800/11516] lr: 1.0000e-03 eta: 1 day, 5:27:53 time: 0.5168 data_time: 0.0011 memory: 9474 grad_norm: 3.3008 loss: 0.6572 +2023/05/31 18:48:35 - mmengine - INFO - Epoch(train) [1][ 1900/11516] lr: 1.0000e-03 eta: 1 day, 5:36:26 time: 0.4467 data_time: 0.0010 memory: 9474 grad_norm: 3.2456 loss: 0.6460 +2023/05/31 18:49:27 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/05/31 18:49:27 - mmengine - INFO - Epoch(train) [1][ 2000/11516] lr: 1.0000e-03 eta: 1 day, 5:45:18 time: 0.4887 data_time: 0.0010 memory: 9474 grad_norm: 2.9178 loss: 0.6701 +2023/05/31 18:50:17 - mmengine - INFO - Epoch(train) [1][ 2100/11516] lr: 1.0000e-03 eta: 1 day, 5:51:24 time: 0.6039 data_time: 0.0011 memory: 9474 grad_norm: 24.6493 loss: 0.6704 +2023/05/31 18:51:13 - mmengine - INFO - Epoch(train) [1][ 2200/11516] lr: 1.0000e-03 eta: 1 day, 6:04:35 time: 0.5378 data_time: 0.0010 memory: 9474 grad_norm: 132.8763 loss: 0.6619 +2023/05/31 18:52:07 - mmengine - INFO - Epoch(train) [1][ 2300/11516] lr: 1.0000e-03 eta: 1 day, 6:15:59 time: 0.4891 data_time: 0.0009 memory: 9474 grad_norm: 0.5663 loss: 0.6899 +2023/05/31 18:53:03 - mmengine - INFO - Epoch(train) [1][ 2400/11516] lr: 1.0000e-03 eta: 1 day, 6:27:18 time: 0.5455 data_time: 0.0010 memory: 9474 grad_norm: 0.4827 loss: 0.6909 +2023/05/31 18:53:56 - mmengine - INFO - Epoch(train) [1][ 2500/11516] lr: 1.0000e-03 eta: 1 day, 6:34:16 time: 0.4413 data_time: 0.0009 memory: 9474 grad_norm: 0.2273 loss: 0.6805 +2023/05/31 18:54:42 - mmengine - INFO - Epoch(train) [1][ 2600/11516] lr: 1.0000e-03 eta: 1 day, 6:30:26 time: 0.4517 data_time: 0.0010 memory: 9474 grad_norm: 0.4640 loss: 0.6886 +2023/05/31 18:55:30 - mmengine - INFO - Epoch(train) [1][ 2700/11516] lr: 1.0000e-03 eta: 1 day, 6:28:37 time: 0.4247 data_time: 0.0010 memory: 9474 grad_norm: 0.4279 loss: 0.6880 +2023/05/31 18:56:19 - mmengine - INFO - Epoch(train) [1][ 2800/11516] lr: 1.0000e-03 eta: 1 day, 6:28:29 time: 0.4853 data_time: 0.0010 memory: 9474 grad_norm: 0.5727 loss: 0.6858 +2023/05/31 18:57:08 - mmengine - INFO - Epoch(train) [1][ 2900/11516] lr: 1.0000e-03 eta: 1 day, 6:29:32 time: 0.5215 data_time: 0.0009 memory: 9474 grad_norm: 0.4035 loss: 0.6894 +2023/05/31 18:57:58 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/05/31 18:57:58 - mmengine - INFO - Epoch(train) [1][ 3000/11516] lr: 1.0000e-03 eta: 1 day, 6:31:02 time: 0.4506 data_time: 0.0012 memory: 9474 grad_norm: 0.1603 loss: 0.6836 +2023/05/31 18:58:47 - mmengine - INFO - Epoch(train) [1][ 3100/11516] lr: 1.0000e-03 eta: 1 day, 6:30:55 time: 0.5652 data_time: 0.0008 memory: 9474 grad_norm: 0.2446 loss: 0.6849 +2023/05/31 18:59:38 - mmengine - INFO - Epoch(train) [1][ 3200/11516] lr: 1.0000e-03 eta: 1 day, 6:32:53 time: 0.4899 data_time: 0.0010 memory: 9474 grad_norm: 0.4427 loss: 0.6896 +2023/05/31 19:00:27 - mmengine - INFO - Epoch(train) [1][ 3300/11516] lr: 1.0000e-03 eta: 1 day, 6:32:47 time: 0.4792 data_time: 0.0010 memory: 9474 grad_norm: 0.2442 loss: 0.6918 +2023/05/31 19:01:13 - mmengine - INFO - Epoch(train) [1][ 3400/11516] lr: 1.0000e-03 eta: 1 day, 6:29:34 time: 0.5593 data_time: 0.0009 memory: 9474 grad_norm: 0.2904 loss: 0.6890 +2023/05/31 19:02:00 - mmengine - INFO - Epoch(train) [1][ 3500/11516] lr: 1.0000e-03 eta: 1 day, 6:27:17 time: 0.4398 data_time: 0.0010 memory: 9474 grad_norm: 0.2884 loss: 0.6909 +2023/05/31 19:02:47 - mmengine - INFO - Epoch(train) [1][ 3600/11516] lr: 1.0000e-03 eta: 1 day, 6:24:39 time: 0.4101 data_time: 0.0010 memory: 9474 grad_norm: 0.3513 loss: 0.6869 +2023/05/31 19:03:34 - mmengine - INFO - Epoch(train) [1][ 3700/11516] lr: 1.0000e-03 eta: 1 day, 6:22:29 time: 0.4872 data_time: 0.0011 memory: 9474 grad_norm: 0.4672 loss: 0.6890 +2023/05/31 19:04:20 - mmengine - INFO - Epoch(train) [1][ 3800/11516] lr: 1.0000e-03 eta: 1 day, 6:19:27 time: 0.4606 data_time: 0.0010 memory: 9474 grad_norm: 0.2243 loss: 0.6924 +2023/05/31 19:05:06 - mmengine - INFO - Epoch(train) [1][ 3900/11516] lr: 1.0000e-03 eta: 1 day, 6:16:31 time: 0.4327 data_time: 0.0009 memory: 9474 grad_norm: 0.2459 loss: 0.6896 +2023/05/31 19:05:55 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/05/31 19:05:55 - mmengine - INFO - Epoch(train) [1][ 4000/11516] lr: 1.0000e-03 eta: 1 day, 6:16:30 time: 0.5555 data_time: 0.0009 memory: 9474 grad_norm: 0.1898 loss: 0.6918 +2023/05/31 19:06:46 - mmengine - INFO - Epoch(train) [1][ 4100/11516] lr: 1.0000e-03 eta: 1 day, 6:18:07 time: 0.4993 data_time: 0.0009 memory: 9474 grad_norm: 0.2338 loss: 0.6852 +2023/05/31 19:07:37 - mmengine - INFO - Epoch(train) [1][ 4200/11516] lr: 1.0000e-03 eta: 1 day, 6:20:38 time: 0.4505 data_time: 0.0013 memory: 9474 grad_norm: 0.3909 loss: 0.6907 +2023/05/31 19:08:33 - mmengine - INFO - Epoch(train) [1][ 4300/11516] lr: 1.0000e-03 eta: 1 day, 6:25:46 time: 0.4613 data_time: 0.0009 memory: 9474 grad_norm: 0.2585 loss: 0.6854 +2023/05/31 19:09:26 - mmengine - INFO - Epoch(train) [1][ 4400/11516] lr: 1.0000e-03 eta: 1 day, 6:29:18 time: 0.4619 data_time: 0.0011 memory: 9474 grad_norm: 0.2268 loss: 0.6889 +2023/05/31 19:10:20 - mmengine - INFO - Epoch(train) [1][ 4500/11516] lr: 1.0000e-03 eta: 1 day, 6:32:33 time: 0.5613 data_time: 0.0013 memory: 9474 grad_norm: 0.1960 loss: 0.6884 +2023/05/31 19:11:13 - mmengine - INFO - Epoch(train) [1][ 4600/11516] lr: 1.0000e-03 eta: 1 day, 6:35:36 time: 0.4140 data_time: 0.0011 memory: 9474 grad_norm: 0.2073 loss: 0.6850 +2023/05/31 19:12:03 - mmengine - INFO - Epoch(train) [1][ 4700/11516] lr: 1.0000e-03 eta: 1 day, 6:35:27 time: 0.4663 data_time: 0.0012 memory: 9474 grad_norm: 0.2313 loss: 0.6821 +2023/05/31 19:12:49 - mmengine - INFO - Epoch(train) [1][ 4800/11516] lr: 1.0000e-03 eta: 1 day, 6:32:52 time: 0.4760 data_time: 0.0010 memory: 9474 grad_norm: 0.1822 loss: 0.6858 +2023/05/31 19:13:38 - mmengine - INFO - Epoch(train) [1][ 4900/11516] lr: 1.0000e-03 eta: 1 day, 6:31:48 time: 0.5322 data_time: 0.0010 memory: 9474 grad_norm: 0.1703 loss: 0.6810 +2023/05/31 19:14:24 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/05/31 19:14:24 - mmengine - INFO - Epoch(train) [1][ 5000/11516] lr: 1.0000e-03 eta: 1 day, 6:29:01 time: 0.4481 data_time: 0.0010 memory: 9474 grad_norm: 0.1946 loss: 0.6892 +2023/05/31 19:15:09 - mmengine - INFO - Epoch(train) [1][ 5100/11516] lr: 1.0000e-03 eta: 1 day, 6:25:33 time: 0.5343 data_time: 0.0010 memory: 9474 grad_norm: 0.2027 loss: 0.6933 +2023/05/31 19:15:55 - mmengine - INFO - Epoch(train) [1][ 5200/11516] lr: 1.0000e-03 eta: 1 day, 6:22:47 time: 0.5068 data_time: 0.0010 memory: 9474 grad_norm: 0.1961 loss: 0.6837 +2023/05/31 19:16:41 - mmengine - INFO - Epoch(train) [1][ 5300/11516] lr: 1.0000e-03 eta: 1 day, 6:20:36 time: 0.5657 data_time: 0.0010 memory: 9474 grad_norm: 0.2148 loss: 0.6859 +2023/05/31 19:17:29 - mmengine - INFO - Epoch(train) [1][ 5400/11516] lr: 1.0000e-03 eta: 1 day, 6:19:04 time: 0.4616 data_time: 0.0010 memory: 9474 grad_norm: 0.1685 loss: 0.6887 +2023/05/31 19:18:15 - mmengine - INFO - Epoch(train) [1][ 5500/11516] lr: 1.0000e-03 eta: 1 day, 6:16:53 time: 0.4233 data_time: 0.0010 memory: 9474 grad_norm: 0.1633 loss: 0.6856 +2023/05/31 19:19:01 - mmengine - INFO - Epoch(train) [1][ 5600/11516] lr: 1.0000e-03 eta: 1 day, 6:14:03 time: 0.4761 data_time: 0.0010 memory: 9474 grad_norm: 0.1990 loss: 0.6885 +2023/05/31 19:19:47 - mmengine - INFO - Epoch(train) [1][ 5700/11516] lr: 1.0000e-03 eta: 1 day, 6:11:50 time: 0.5139 data_time: 0.0010 memory: 9474 grad_norm: 0.2088 loss: 0.6890 +2023/05/31 19:20:33 - mmengine - INFO - Epoch(train) [1][ 5800/11516] lr: 1.0000e-03 eta: 1 day, 6:09:24 time: 0.5121 data_time: 0.0010 memory: 9474 grad_norm: 0.1276 loss: 0.6907 +2023/05/31 19:21:18 - mmengine - INFO - Epoch(train) [1][ 5900/11516] lr: 1.0000e-03 eta: 1 day, 6:06:11 time: 0.4585 data_time: 0.0010 memory: 9474 grad_norm: 0.2262 loss: 0.6937 +2023/05/31 19:22:03 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/05/31 19:22:03 - mmengine - INFO - Epoch(train) [1][ 6000/11516] lr: 1.0000e-03 eta: 1 day, 6:03:22 time: 0.4318 data_time: 0.0010 memory: 9474 grad_norm: 0.2533 loss: 0.6887 +2023/05/31 19:22:48 - mmengine - INFO - Epoch(train) [1][ 6100/11516] lr: 1.0000e-03 eta: 1 day, 6:00:36 time: 0.4150 data_time: 0.0010 memory: 9474 grad_norm: 0.2085 loss: 0.6801 +2023/05/31 19:23:32 - mmengine - INFO - Epoch(train) [1][ 6200/11516] lr: 1.0000e-03 eta: 1 day, 5:57:36 time: 0.4432 data_time: 0.0010 memory: 9474 grad_norm: 0.2214 loss: 0.6856 +2023/05/31 19:24:19 - mmengine - INFO - Epoch(train) [1][ 6300/11516] lr: 1.0000e-03 eta: 1 day, 5:56:16 time: 0.4504 data_time: 0.0014 memory: 9474 grad_norm: 0.1444 loss: 0.6891 +2023/05/31 19:25:07 - mmengine - INFO - Epoch(train) [1][ 6400/11516] lr: 1.0000e-03 eta: 1 day, 5:54:51 time: 0.4679 data_time: 0.0010 memory: 9474 grad_norm: 0.1146 loss: 0.6907 +2023/05/31 19:25:52 - mmengine - INFO - Epoch(train) [1][ 6500/11516] lr: 1.0000e-03 eta: 1 day, 5:52:34 time: 0.4362 data_time: 0.0009 memory: 9474 grad_norm: 0.1845 loss: 0.6838 +2023/05/31 19:26:40 - mmengine - INFO - Epoch(train) [1][ 6600/11516] lr: 1.0000e-03 eta: 1 day, 5:51:33 time: 0.4376 data_time: 0.0010 memory: 9474 grad_norm: 0.1626 loss: 0.6911 +2023/05/31 19:27:26 - mmengine - INFO - Epoch(train) [1][ 6700/11516] lr: 1.0000e-03 eta: 1 day, 5:49:49 time: 0.4575 data_time: 0.0010 memory: 9474 grad_norm: 0.1635 loss: 0.6810 +2023/05/31 19:28:14 - mmengine - INFO - Epoch(train) [1][ 6800/11516] lr: 1.0000e-03 eta: 1 day, 5:49:08 time: 0.4316 data_time: 0.0011 memory: 9474 grad_norm: 0.1678 loss: 0.6851 +2023/05/31 19:29:01 - mmengine - INFO - Epoch(train) [1][ 6900/11516] lr: 1.0000e-03 eta: 1 day, 5:47:34 time: 0.4458 data_time: 0.0011 memory: 9474 grad_norm: 0.1885 loss: 0.6768 +2023/05/31 19:29:48 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/05/31 19:29:48 - mmengine - INFO - Epoch(train) [1][ 7000/11516] lr: 1.0000e-03 eta: 1 day, 5:46:27 time: 0.4218 data_time: 0.0013 memory: 9474 grad_norm: 0.1760 loss: 0.6886 +2023/05/31 19:30:34 - mmengine - INFO - Epoch(train) [1][ 7100/11516] lr: 1.0000e-03 eta: 1 day, 5:44:24 time: 0.5111 data_time: 0.0011 memory: 9474 grad_norm: 0.1821 loss: 0.6890 +2023/05/31 19:31:19 - mmengine - INFO - Epoch(train) [1][ 7200/11516] lr: 1.0000e-03 eta: 1 day, 5:42:03 time: 0.4269 data_time: 0.0013 memory: 9474 grad_norm: 0.1257 loss: 0.6871 +2023/05/31 19:32:04 - mmengine - INFO - Epoch(train) [1][ 7300/11516] lr: 1.0000e-03 eta: 1 day, 5:40:02 time: 0.4327 data_time: 0.0011 memory: 9474 grad_norm: 0.1673 loss: 0.6883 +2023/05/31 19:32:50 - mmengine - INFO - Epoch(train) [1][ 7400/11516] lr: 1.0000e-03 eta: 1 day, 5:37:53 time: 0.4496 data_time: 0.0011 memory: 9474 grad_norm: 0.1992 loss: 0.6864 +2023/05/31 19:33:34 - mmengine - INFO - Epoch(train) [1][ 7500/11516] lr: 1.0000e-03 eta: 1 day, 5:35:30 time: 0.4295 data_time: 0.0011 memory: 9474 grad_norm: 0.2271 loss: 0.6883 +2023/05/31 19:34:17 - mmengine - INFO - Epoch(train) [1][ 7600/11516] lr: 1.0000e-03 eta: 1 day, 5:32:18 time: 0.4675 data_time: 0.0009 memory: 9474 grad_norm: 0.1085 loss: 0.6850 +2023/05/31 19:35:04 - mmengine - INFO - Epoch(train) [1][ 7700/11516] lr: 1.0000e-03 eta: 1 day, 5:31:07 time: 0.4138 data_time: 0.0011 memory: 9474 grad_norm: 0.2098 loss: 0.6876 +2023/05/31 19:35:47 - mmengine - INFO - Epoch(train) [1][ 7800/11516] lr: 1.0000e-03 eta: 1 day, 5:28:13 time: 0.4554 data_time: 0.0010 memory: 9474 grad_norm: 0.1084 loss: 0.6887 +2023/05/31 19:36:30 - mmengine - INFO - Epoch(train) [1][ 7900/11516] lr: 1.0000e-03 eta: 1 day, 5:25:02 time: 0.5044 data_time: 0.0009 memory: 9474 grad_norm: 0.1093 loss: 0.6847 +2023/05/31 19:37:14 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/05/31 19:37:14 - mmengine - INFO - Epoch(train) [1][ 8000/11516] lr: 1.0000e-03 eta: 1 day, 5:22:21 time: 0.4144 data_time: 0.0010 memory: 9474 grad_norm: 0.1007 loss: 0.6876 +2023/05/31 19:37:56 - mmengine - INFO - Epoch(train) [1][ 8100/11516] lr: 1.0000e-03 eta: 1 day, 5:19:15 time: 0.3926 data_time: 0.0010 memory: 9474 grad_norm: 0.1331 loss: 0.6844 +2023/05/31 19:38:37 - mmengine - INFO - Epoch(train) [1][ 8200/11516] lr: 1.0000e-03 eta: 1 day, 5:15:36 time: 0.4042 data_time: 0.0011 memory: 9474 grad_norm: 0.1349 loss: 0.6889 +2023/05/31 19:39:20 - mmengine - INFO - Epoch(train) [1][ 8300/11516] lr: 1.0000e-03 eta: 1 day, 5:12:31 time: 0.3975 data_time: 0.0010 memory: 9474 grad_norm: 0.1128 loss: 0.6876 +2023/05/31 19:40:02 - mmengine - INFO - Epoch(train) [1][ 8400/11516] lr: 1.0000e-03 eta: 1 day, 5:09:34 time: 0.4118 data_time: 0.0010 memory: 9474 grad_norm: 0.0988 loss: 0.6861 +2023/05/31 19:40:44 - mmengine - INFO - Epoch(train) [1][ 8500/11516] lr: 1.0000e-03 eta: 1 day, 5:06:26 time: 0.4339 data_time: 0.0010 memory: 9474 grad_norm: 0.1379 loss: 0.6873 +2023/05/31 19:41:26 - mmengine - INFO - Epoch(train) [1][ 8600/11516] lr: 1.0000e-03 eta: 1 day, 5:03:32 time: 0.4291 data_time: 0.0009 memory: 9474 grad_norm: 0.1052 loss: 0.6859 +2023/05/31 19:42:09 - mmengine - INFO - Epoch(train) [1][ 8700/11516] lr: 1.0000e-03 eta: 1 day, 5:00:52 time: 0.4569 data_time: 0.0010 memory: 9474 grad_norm: 0.1037 loss: 0.6860 +2023/05/31 19:42:52 - mmengine - INFO - Epoch(train) [1][ 8800/11516] lr: 1.0000e-03 eta: 1 day, 4:58:12 time: 0.3763 data_time: 0.0014 memory: 9474 grad_norm: 0.0864 loss: 0.6815 +2023/05/31 19:43:35 - mmengine - INFO - Epoch(train) [1][ 8900/11516] lr: 1.0000e-03 eta: 1 day, 4:55:56 time: 0.4261 data_time: 0.0011 memory: 9474 grad_norm: 0.0829 loss: 0.6876 +2023/05/31 19:44:18 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/05/31 19:44:18 - mmengine - INFO - Epoch(train) [1][ 9000/11516] lr: 1.0000e-03 eta: 1 day, 4:53:18 time: 0.4016 data_time: 0.0013 memory: 9474 grad_norm: 0.1154 loss: 0.6890 +2023/05/31 19:45:00 - mmengine - INFO - Epoch(train) [1][ 9100/11516] lr: 1.0000e-03 eta: 1 day, 4:50:38 time: 0.4114 data_time: 0.0011 memory: 9474 grad_norm: 0.0936 loss: 0.6811 +2023/05/31 19:45:44 - mmengine - INFO - Epoch(train) [1][ 9200/11516] lr: 1.0000e-03 eta: 1 day, 4:48:44 time: 0.3920 data_time: 0.0010 memory: 9474 grad_norm: 0.0876 loss: 0.6842 +2023/05/31 19:46:25 - mmengine - INFO - Epoch(train) [1][ 9300/11516] lr: 1.0000e-03 eta: 1 day, 4:45:24 time: 0.4353 data_time: 0.0011 memory: 9474 grad_norm: 0.0783 loss: 0.6842 +2023/05/31 19:47:05 - mmengine - INFO - Epoch(train) [1][ 9400/11516] lr: 1.0000e-03 eta: 1 day, 4:41:58 time: 0.4073 data_time: 0.0011 memory: 9474 grad_norm: 0.0740 loss: 0.6865 +2023/05/31 19:47:44 - mmengine - INFO - Epoch(train) [1][ 9500/11516] lr: 1.0000e-03 eta: 1 day, 4:38:16 time: 0.4142 data_time: 0.0011 memory: 9474 grad_norm: 0.0864 loss: 0.6852 +2023/05/31 19:48:24 - mmengine - INFO - Epoch(train) [1][ 9600/11516] lr: 1.0000e-03 eta: 1 day, 4:34:59 time: 0.3751 data_time: 0.0011 memory: 9474 grad_norm: 0.0889 loss: 0.6820 +2023/05/31 19:49:04 - mmengine - INFO - Epoch(train) [1][ 9700/11516] lr: 1.0000e-03 eta: 1 day, 4:31:40 time: 0.4095 data_time: 0.0010 memory: 9474 grad_norm: 0.0714 loss: 0.6932 +2023/05/31 19:49:44 - mmengine - INFO - Epoch(train) [1][ 9800/11516] lr: 1.0000e-03 eta: 1 day, 4:28:15 time: 0.4012 data_time: 0.0011 memory: 9474 grad_norm: 0.0976 loss: 0.6862 +2023/05/31 19:50:24 - mmengine - INFO - Epoch(train) [1][ 9900/11516] lr: 1.0000e-03 eta: 1 day, 4:25:19 time: 0.4107 data_time: 0.0011 memory: 9474 grad_norm: 0.1007 loss: 0.6831 +2023/05/31 19:51:04 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/05/31 19:51:04 - mmengine - INFO - Epoch(train) [1][10000/11516] lr: 1.0000e-03 eta: 1 day, 4:22:16 time: 0.4604 data_time: 0.0010 memory: 9474 grad_norm: 0.0664 loss: 0.6833 +2023/05/31 19:51:45 - mmengine - INFO - Epoch(train) [1][10100/11516] lr: 1.0000e-03 eta: 1 day, 4:19:14 time: 0.3841 data_time: 0.0009 memory: 9474 grad_norm: 0.1276 loss: 0.6819 +2023/05/31 19:52:25 - mmengine - INFO - Epoch(train) [1][10200/11516] lr: 1.0000e-03 eta: 1 day, 4:16:23 time: 0.4100 data_time: 0.0011 memory: 9474 grad_norm: 0.0643 loss: 0.6876 +2023/05/31 19:53:06 - mmengine - INFO - Epoch(train) [1][10300/11516] lr: 1.0000e-03 eta: 1 day, 4:13:35 time: 0.3771 data_time: 0.0010 memory: 9474 grad_norm: 0.0872 loss: 0.6833 +2023/05/31 19:53:46 - mmengine - INFO - Epoch(train) [1][10400/11516] lr: 1.0000e-03 eta: 1 day, 4:10:42 time: 0.3995 data_time: 0.0009 memory: 9474 grad_norm: 0.1113 loss: 0.6888 +2023/05/31 19:54:26 - mmengine - INFO - Epoch(train) [1][10500/11516] lr: 1.0000e-03 eta: 1 day, 4:08:01 time: 0.3903 data_time: 0.0010 memory: 9474 grad_norm: 0.0840 loss: 0.6866 +2023/05/31 19:55:08 - mmengine - INFO - Epoch(train) [1][10600/11516] lr: 1.0000e-03 eta: 1 day, 4:05:35 time: 0.4022 data_time: 0.0013 memory: 9474 grad_norm: 0.0685 loss: 0.6827 +2023/05/31 19:55:47 - mmengine - INFO - Epoch(train) [1][10700/11516] lr: 1.0000e-03 eta: 1 day, 4:02:39 time: 0.3921 data_time: 0.0010 memory: 9474 grad_norm: 0.0583 loss: 0.6870 +2023/05/31 19:56:29 - mmengine - INFO - Epoch(train) [1][10800/11516] lr: 1.0000e-03 eta: 1 day, 4:00:31 time: 0.4209 data_time: 0.0010 memory: 9474 grad_norm: 0.0780 loss: 0.6848 +2023/05/31 19:57:10 - mmengine - INFO - Epoch(train) [1][10900/11516] lr: 1.0000e-03 eta: 1 day, 3:58:01 time: 0.3921 data_time: 0.0011 memory: 9474 grad_norm: 0.0630 loss: 0.6852 +2023/05/31 19:57:51 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/05/31 19:57:51 - mmengine - INFO - Epoch(train) [1][11000/11516] lr: 1.0000e-03 eta: 1 day, 3:55:27 time: 0.4039 data_time: 0.0010 memory: 9474 grad_norm: 0.0776 loss: 0.6900 +2023/05/31 19:58:32 - mmengine - INFO - Epoch(train) [1][11100/11516] lr: 1.0000e-03 eta: 1 day, 3:53:07 time: 0.3837 data_time: 0.0011 memory: 9474 grad_norm: 0.0776 loss: 0.6858 +2023/05/31 19:59:14 - mmengine - INFO - Epoch(train) [1][11200/11516] lr: 1.0000e-03 eta: 1 day, 3:51:06 time: 0.3940 data_time: 0.0011 memory: 9474 grad_norm: 0.0722 loss: 0.6898 +2023/05/31 19:59:56 - mmengine - INFO - Epoch(train) [1][11300/11516] lr: 1.0000e-03 eta: 1 day, 3:49:13 time: 0.4238 data_time: 0.0011 memory: 9474 grad_norm: 0.0628 loss: 0.6920 +2023/05/31 20:00:36 - mmengine - INFO - Epoch(train) [1][11400/11516] lr: 1.0000e-03 eta: 1 day, 3:46:47 time: 0.3736 data_time: 0.0015 memory: 9474 grad_norm: 0.0886 loss: 0.6866 +2023/05/31 20:01:17 - mmengine - INFO - Epoch(train) [1][11500/11516] lr: 1.0000e-03 eta: 1 day, 3:44:21 time: 0.4357 data_time: 0.0083 memory: 9474 grad_norm: 0.0653 loss: 0.6869 +2023/05/31 20:01:22 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/05/31 20:01:22 - mmengine - INFO - Saving checkpoint at 1 epochs +2023/05/31 20:01:40 - mmengine - INFO - Epoch(val) [1][16/16] accuracy/top1: 100.0000 single-label/precision_classwise: [100.0, 0.0] single-label/recall_classwise: [100.0, 0.0] single-label/f1-score_classwise: [100.0, 0.0] data_time: 0.3099 time: 0.6909 +2023/05/31 20:02:24 - mmengine - INFO - Epoch(train) [2][ 100/11516] lr: 9.9391e-04 eta: 1 day, 3:42:09 time: 0.4641 data_time: 0.0708 memory: 9474 grad_norm: 0.1100 loss: 0.6879 +2023/05/31 20:03:07 - mmengine - INFO - Epoch(train) [2][ 200/11516] lr: 9.9391e-04 eta: 1 day, 3:40:31 time: 0.4323 data_time: 0.0010 memory: 9466 grad_norm: 0.0643 loss: 0.6863 +2023/05/31 20:03:48 - mmengine - INFO - Epoch(train) [2][ 300/11516] lr: 9.9391e-04 eta: 1 day, 3:38:31 time: 0.4900 data_time: 0.0012 memory: 9466 grad_norm: 0.0751 loss: 0.6849 +2023/05/31 20:04:28 - mmengine - INFO - Epoch(train) [2][ 400/11516] lr: 9.9391e-04 eta: 1 day, 3:36:07 time: 0.4028 data_time: 0.0010 memory: 9466 grad_norm: 0.0693 loss: 0.6876 +2023/05/31 20:05:02 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/05/31 20:05:10 - mmengine - INFO - Epoch(train) [2][ 500/11516] lr: 9.9391e-04 eta: 1 day, 3:34:11 time: 0.4004 data_time: 0.0011 memory: 9466 grad_norm: 0.0694 loss: 0.6846 +2023/05/31 20:05:50 - mmengine - INFO - Epoch(train) [2][ 600/11516] lr: 9.9391e-04 eta: 1 day, 3:31:52 time: 0.4354 data_time: 0.0010 memory: 9466 grad_norm: 0.0625 loss: 0.6886 +2023/05/31 20:06:34 - mmengine - INFO - Epoch(train) [2][ 700/11516] lr: 9.9391e-04 eta: 1 day, 3:30:33 time: 0.5043 data_time: 0.0013 memory: 9466 grad_norm: 0.0581 loss: 0.6871 +2023/05/31 20:07:19 - mmengine - INFO - Epoch(train) [2][ 800/11516] lr: 9.9391e-04 eta: 1 day, 3:29:46 time: 0.3949 data_time: 0.0012 memory: 9466 grad_norm: 0.0625 loss: 0.6895 +2023/05/31 20:08:04 - mmengine - INFO - Epoch(train) [2][ 900/11516] lr: 9.9391e-04 eta: 1 day, 3:28:56 time: 0.4600 data_time: 0.0011 memory: 9466 grad_norm: 0.0790 loss: 0.6860 +2023/05/31 20:08:48 - mmengine - INFO - Epoch(train) [2][ 1000/11516] lr: 9.9391e-04 eta: 1 day, 3:27:46 time: 0.4622 data_time: 0.0011 memory: 9466 grad_norm: 0.0657 loss: 0.6879 +2023/05/31 20:09:34 - mmengine - INFO - Epoch(train) [2][ 1100/11516] lr: 9.9391e-04 eta: 1 day, 3:27:03 time: 0.4736 data_time: 0.0011 memory: 9466 grad_norm: 0.0604 loss: 0.6913 +2023/05/31 20:10:18 - mmengine - INFO - Epoch(train) [2][ 1200/11516] lr: 9.9391e-04 eta: 1 day, 3:25:55 time: 0.4740 data_time: 0.0010 memory: 9466 grad_norm: 0.0592 loss: 0.6904 +2023/05/31 20:11:02 - mmengine - INFO - Epoch(train) [2][ 1300/11516] lr: 9.9391e-04 eta: 1 day, 3:24:45 time: 0.4029 data_time: 0.0012 memory: 9466 grad_norm: 0.0596 loss: 0.6871 +2023/05/31 20:11:45 - mmengine - INFO - Epoch(train) [2][ 1400/11516] lr: 9.9391e-04 eta: 1 day, 3:23:21 time: 0.4679 data_time: 0.0011 memory: 9466 grad_norm: 0.0650 loss: 0.6827 +2023/05/31 20:12:23 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/05/31 20:12:31 - mmengine - INFO - Epoch(train) [2][ 1500/11516] lr: 9.9391e-04 eta: 1 day, 3:22:44 time: 0.3952 data_time: 0.0010 memory: 9466 grad_norm: 0.0438 loss: 0.6886 +2023/05/31 20:13:17 - mmengine - INFO - Epoch(train) [2][ 1600/11516] lr: 9.9391e-04 eta: 1 day, 3:22:08 time: 0.5633 data_time: 0.0010 memory: 9466 grad_norm: 0.0383 loss: 0.6880 +2023/05/31 20:14:02 - mmengine - INFO - Epoch(train) [2][ 1700/11516] lr: 9.9391e-04 eta: 1 day, 3:21:24 time: 0.4600 data_time: 0.0011 memory: 9466 grad_norm: 0.0658 loss: 0.6826 +2023/05/31 20:14:47 - mmengine - INFO - Epoch(train) [2][ 1800/11516] lr: 9.9391e-04 eta: 1 day, 3:20:30 time: 0.4204 data_time: 0.0011 memory: 9466 grad_norm: 0.0838 loss: 0.6833 +2023/05/31 20:15:32 - mmengine - INFO - Epoch(train) [2][ 1900/11516] lr: 9.9391e-04 eta: 1 day, 3:19:36 time: 0.4352 data_time: 0.0011 memory: 9466 grad_norm: 0.0482 loss: 0.6858 +2023/05/31 20:16:16 - mmengine - INFO - Epoch(train) [2][ 2000/11516] lr: 9.9391e-04 eta: 1 day, 3:18:24 time: 0.4297 data_time: 0.0010 memory: 9466 grad_norm: 0.0503 loss: 0.6835 +2023/05/31 20:17:00 - mmengine - INFO - Epoch(train) [2][ 2100/11516] lr: 9.9391e-04 eta: 1 day, 3:17:23 time: 0.3988 data_time: 0.0011 memory: 9466 grad_norm: 0.0481 loss: 0.6889 +2023/05/31 20:17:44 - mmengine - INFO - Epoch(train) [2][ 2200/11516] lr: 9.9391e-04 eta: 1 day, 3:16:17 time: 0.4223 data_time: 0.0011 memory: 9466 grad_norm: 0.0480 loss: 0.6915 +2023/05/31 20:18:29 - mmengine - INFO - Epoch(train) [2][ 2300/11516] lr: 9.9391e-04 eta: 1 day, 3:15:30 time: 0.4085 data_time: 0.0011 memory: 9466 grad_norm: 0.0386 loss: 0.6875 +2023/05/31 20:19:13 - mmengine - INFO - Epoch(train) [2][ 2400/11516] lr: 9.9391e-04 eta: 1 day, 3:14:29 time: 0.4898 data_time: 0.0012 memory: 9466 grad_norm: 0.0378 loss: 0.6899 +2023/05/31 20:19:50 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/05/31 20:19:58 - mmengine - INFO - Epoch(train) [2][ 2500/11516] lr: 9.9391e-04 eta: 1 day, 3:13:25 time: 0.4233 data_time: 0.0011 memory: 9466 grad_norm: 0.0630 loss: 0.6826 +2023/05/31 20:20:44 - mmengine - INFO - Epoch(train) [2][ 2600/11516] lr: 9.9391e-04 eta: 1 day, 3:12:56 time: 0.4419 data_time: 0.0011 memory: 9466 grad_norm: 0.0396 loss: 0.6864 +2023/05/31 20:21:28 - mmengine - INFO - Epoch(train) [2][ 2700/11516] lr: 9.9391e-04 eta: 1 day, 3:11:58 time: 0.4348 data_time: 0.0011 memory: 9466 grad_norm: 0.0582 loss: 0.6863 +2023/05/31 20:22:13 - mmengine - INFO - Epoch(train) [2][ 2800/11516] lr: 9.9391e-04 eta: 1 day, 3:10:59 time: 0.4726 data_time: 0.0011 memory: 9466 grad_norm: 0.0383 loss: 0.6906 +2023/05/31 20:22:58 - mmengine - INFO - Epoch(train) [2][ 2900/11516] lr: 9.9391e-04 eta: 1 day, 3:10:09 time: 0.4505 data_time: 0.0011 memory: 9466 grad_norm: 0.0525 loss: 0.6814 +2023/05/31 20:23:42 - mmengine - INFO - Epoch(train) [2][ 3000/11516] lr: 9.9391e-04 eta: 1 day, 3:09:10 time: 0.4128 data_time: 0.0010 memory: 9466 grad_norm: 0.0462 loss: 0.6880 +2023/05/31 20:24:25 - mmengine - INFO - Epoch(train) [2][ 3100/11516] lr: 9.9391e-04 eta: 1 day, 3:07:51 time: 0.4170 data_time: 0.0010 memory: 9466 grad_norm: 0.0523 loss: 0.6854 +2023/05/31 20:25:09 - mmengine - INFO - Epoch(train) [2][ 3200/11516] lr: 9.9391e-04 eta: 1 day, 3:06:49 time: 0.4091 data_time: 0.0010 memory: 9466 grad_norm: 0.0594 loss: 0.6881 +2023/05/31 20:25:55 - mmengine - INFO - Epoch(train) [2][ 3300/11516] lr: 9.9391e-04 eta: 1 day, 3:06:12 time: 0.5492 data_time: 0.0011 memory: 9466 grad_norm: 0.0391 loss: 0.6858 +2023/05/31 20:26:37 - mmengine - INFO - Epoch(train) [2][ 3400/11516] lr: 9.9391e-04 eta: 1 day, 3:04:41 time: 0.4130 data_time: 0.0011 memory: 9466 grad_norm: 0.0438 loss: 0.6866 +2023/05/31 20:27:15 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/05/31 20:27:23 - mmengine - INFO - Epoch(train) [2][ 3500/11516] lr: 9.9391e-04 eta: 1 day, 3:04:07 time: 0.4556 data_time: 0.0012 memory: 9466 grad_norm: 0.0514 loss: 0.6854 +2023/05/31 20:28:07 - mmengine - INFO - Epoch(train) [2][ 3600/11516] lr: 9.9391e-04 eta: 1 day, 3:02:54 time: 0.4236 data_time: 0.0012 memory: 9466 grad_norm: 0.0266 loss: 0.6853 +2023/05/31 20:28:52 - mmengine - INFO - Epoch(train) [2][ 3700/11516] lr: 9.9391e-04 eta: 1 day, 3:02:04 time: 0.4460 data_time: 0.0011 memory: 9466 grad_norm: 0.0481 loss: 0.6889 +2023/05/31 20:29:35 - mmengine - INFO - Epoch(train) [2][ 3800/11516] lr: 9.9391e-04 eta: 1 day, 3:00:54 time: 0.4266 data_time: 0.0010 memory: 9466 grad_norm: 0.0630 loss: 0.6901 +2023/05/31 20:30:18 - mmengine - INFO - Epoch(train) [2][ 3900/11516] lr: 9.9391e-04 eta: 1 day, 2:59:38 time: 0.4837 data_time: 0.0010 memory: 9466 grad_norm: 0.0324 loss: 0.6861 +2023/05/31 20:31:02 - mmengine - INFO - Epoch(train) [2][ 4000/11516] lr: 9.9391e-04 eta: 1 day, 2:58:40 time: 0.4131 data_time: 0.0011 memory: 9466 grad_norm: 0.0530 loss: 0.6845 +2023/05/31 20:31:46 - mmengine - INFO - Epoch(train) [2][ 4100/11516] lr: 9.9391e-04 eta: 1 day, 2:57:28 time: 0.4349 data_time: 0.0010 memory: 9466 grad_norm: 0.0471 loss: 0.6833 +2023/05/31 20:32:30 - mmengine - INFO - Epoch(train) [2][ 4200/11516] lr: 9.9391e-04 eta: 1 day, 2:56:27 time: 0.4572 data_time: 0.0010 memory: 9466 grad_norm: 0.0325 loss: 0.6896 +2023/05/31 20:33:14 - mmengine - INFO - Epoch(train) [2][ 4300/11516] lr: 9.9391e-04 eta: 1 day, 2:55:28 time: 0.4206 data_time: 0.0011 memory: 9466 grad_norm: 0.0590 loss: 0.6904 +2023/05/31 20:34:00 - mmengine - INFO - Epoch(train) [2][ 4400/11516] lr: 9.9391e-04 eta: 1 day, 2:54:56 time: 0.5405 data_time: 0.0010 memory: 9466 grad_norm: 0.0413 loss: 0.6915 +2023/05/31 20:34:37 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/05/31 20:34:45 - mmengine - INFO - Epoch(train) [2][ 4500/11516] lr: 9.9391e-04 eta: 1 day, 2:54:10 time: 0.5038 data_time: 0.0010 memory: 9466 grad_norm: 0.0405 loss: 0.6893 +2023/05/31 20:35:30 - mmengine - INFO - Epoch(train) [2][ 4600/11516] lr: 9.9391e-04 eta: 1 day, 2:53:21 time: 0.4745 data_time: 0.0010 memory: 9466 grad_norm: 0.0344 loss: 0.6910 +2023/05/31 20:36:14 - mmengine - INFO - Epoch(train) [2][ 4700/11516] lr: 9.9391e-04 eta: 1 day, 2:52:19 time: 0.4528 data_time: 0.0010 memory: 9466 grad_norm: 0.0467 loss: 0.6837 +2023/05/31 20:36:59 - mmengine - INFO - Epoch(train) [2][ 4800/11516] lr: 9.9391e-04 eta: 1 day, 2:51:27 time: 0.4154 data_time: 0.0010 memory: 9466 grad_norm: 0.0514 loss: 0.6850 +2023/05/31 20:37:43 - mmengine - INFO - Epoch(train) [2][ 4900/11516] lr: 9.9391e-04 eta: 1 day, 2:50:36 time: 0.4346 data_time: 0.0010 memory: 9466 grad_norm: 0.0520 loss: 0.6895 +2023/05/31 20:38:26 - mmengine - INFO - Epoch(train) [2][ 5000/11516] lr: 9.9391e-04 eta: 1 day, 2:49:13 time: 0.4143 data_time: 0.0010 memory: 9466 grad_norm: 0.0348 loss: 0.6847 +2023/05/31 20:39:10 - mmengine - INFO - Epoch(train) [2][ 5100/11516] lr: 9.9391e-04 eta: 1 day, 2:48:17 time: 0.4122 data_time: 0.0011 memory: 9466 grad_norm: 0.0353 loss: 0.6843 +2023/05/31 20:39:53 - mmengine - INFO - Epoch(train) [2][ 5200/11516] lr: 9.9391e-04 eta: 1 day, 2:47:10 time: 0.4130 data_time: 0.0010 memory: 9466 grad_norm: 0.0337 loss: 0.6873 +2023/05/31 20:40:38 - mmengine - INFO - Epoch(train) [2][ 5300/11516] lr: 9.9391e-04 eta: 1 day, 2:46:11 time: 0.4561 data_time: 0.0011 memory: 9466 grad_norm: 0.0595 loss: 0.6903 +2023/05/31 20:41:22 - mmengine - INFO - Epoch(train) [2][ 5400/11516] lr: 9.9391e-04 eta: 1 day, 2:45:22 time: 0.4121 data_time: 0.0011 memory: 9466 grad_norm: 0.0330 loss: 0.6896 +2023/05/31 20:41:59 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/05/31 20:42:07 - mmengine - INFO - Epoch(train) [2][ 5500/11516] lr: 9.9391e-04 eta: 1 day, 2:44:25 time: 0.4512 data_time: 0.0011 memory: 9466 grad_norm: 0.0475 loss: 0.6829 +2023/05/31 20:42:49 - mmengine - INFO - Epoch(train) [2][ 5600/11516] lr: 9.9391e-04 eta: 1 day, 2:43:10 time: 0.4276 data_time: 0.0011 memory: 9466 grad_norm: 0.0522 loss: 0.6843 +2023/05/31 20:43:34 - mmengine - INFO - Epoch(train) [2][ 5700/11516] lr: 9.9391e-04 eta: 1 day, 2:42:19 time: 0.4242 data_time: 0.0011 memory: 9466 grad_norm: 0.0687 loss: 0.6868 +2023/05/31 20:44:17 - mmengine - INFO - Epoch(train) [2][ 5800/11516] lr: 9.9391e-04 eta: 1 day, 2:41:04 time: 0.4299 data_time: 0.0012 memory: 9466 grad_norm: 0.0221 loss: 0.6896 +2023/05/31 20:45:00 - mmengine - INFO - Epoch(train) [2][ 5900/11516] lr: 9.9391e-04 eta: 1 day, 2:40:02 time: 0.4183 data_time: 0.0011 memory: 9466 grad_norm: 0.0353 loss: 0.6865 +2023/05/31 20:45:46 - mmengine - INFO - Epoch(train) [2][ 6000/11516] lr: 9.9391e-04 eta: 1 day, 2:39:23 time: 0.4258 data_time: 0.0012 memory: 9466 grad_norm: 0.0373 loss: 0.6877 +2023/05/31 20:46:31 - mmengine - INFO - Epoch(train) [2][ 6100/11516] lr: 9.9391e-04 eta: 1 day, 2:38:34 time: 0.4774 data_time: 0.0010 memory: 9466 grad_norm: 0.0467 loss: 0.6802 +2023/05/31 20:47:15 - mmengine - INFO - Epoch(train) [2][ 6200/11516] lr: 9.9391e-04 eta: 1 day, 2:37:42 time: 0.4868 data_time: 0.0010 memory: 9466 grad_norm: 0.0363 loss: 0.6818 +2023/05/31 20:47:58 - mmengine - INFO - Epoch(train) [2][ 6300/11516] lr: 9.9391e-04 eta: 1 day, 2:36:25 time: 0.4498 data_time: 0.0011 memory: 9466 grad_norm: 0.0315 loss: 0.6831 +2023/05/31 20:48:41 - mmengine - INFO - Epoch(train) [2][ 6400/11516] lr: 9.9391e-04 eta: 1 day, 2:35:23 time: 0.4413 data_time: 0.0010 memory: 9466 grad_norm: 0.0234 loss: 0.6895 +2023/05/31 20:49:17 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/05/31 20:49:25 - mmengine - INFO - Epoch(train) [2][ 6500/11516] lr: 9.9391e-04 eta: 1 day, 2:34:25 time: 0.4274 data_time: 0.0011 memory: 9466 grad_norm: 0.0515 loss: 0.6876 +2023/05/31 20:50:09 - mmengine - INFO - Epoch(train) [2][ 6600/11516] lr: 9.9391e-04 eta: 1 day, 2:33:23 time: 0.4172 data_time: 0.0010 memory: 9466 grad_norm: 0.0303 loss: 0.6843 +2023/05/31 20:50:52 - mmengine - INFO - Epoch(train) [2][ 6700/11516] lr: 9.9391e-04 eta: 1 day, 2:32:15 time: 0.4024 data_time: 0.0010 memory: 9466 grad_norm: 0.0248 loss: 0.6800 +2023/05/31 20:51:34 - mmengine - INFO - Epoch(train) [2][ 6800/11516] lr: 9.9391e-04 eta: 1 day, 2:30:57 time: 0.4804 data_time: 0.0010 memory: 9466 grad_norm: 0.0337 loss: 0.6904 +2023/05/31 20:52:19 - mmengine - INFO - Epoch(train) [2][ 6900/11516] lr: 9.9391e-04 eta: 1 day, 2:30:10 time: 0.4307 data_time: 0.0010 memory: 9466 grad_norm: 0.0358 loss: 0.6825 +2023/05/31 20:53:04 - mmengine - INFO - Epoch(train) [2][ 7000/11516] lr: 9.9391e-04 eta: 1 day, 2:29:27 time: 0.4387 data_time: 0.0015 memory: 9466 grad_norm: 0.0284 loss: 0.6894 +2023/05/31 20:53:48 - mmengine - INFO - Epoch(train) [2][ 7100/11516] lr: 9.9391e-04 eta: 1 day, 2:28:23 time: 0.4826 data_time: 0.0011 memory: 9466 grad_norm: 0.0393 loss: 0.6891 +2023/05/31 20:54:30 - mmengine - INFO - Epoch(train) [2][ 7200/11516] lr: 9.9391e-04 eta: 1 day, 2:27:12 time: 0.3986 data_time: 0.0011 memory: 9466 grad_norm: 0.0265 loss: 0.6855 +2023/05/31 20:55:14 - mmengine - INFO - Epoch(train) [2][ 7300/11516] lr: 9.9391e-04 eta: 1 day, 2:26:07 time: 0.4109 data_time: 0.0012 memory: 9466 grad_norm: 0.0403 loss: 0.6886 +2023/05/31 20:55:57 - mmengine - INFO - Epoch(train) [2][ 7400/11516] lr: 9.9391e-04 eta: 1 day, 2:25:09 time: 0.4611 data_time: 0.0011 memory: 9466 grad_norm: 0.0393 loss: 0.6831 +2023/05/31 20:56:32 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/05/31 20:56:40 - mmengine - INFO - Epoch(train) [2][ 7500/11516] lr: 9.9391e-04 eta: 1 day, 2:23:55 time: 0.4583 data_time: 0.0010 memory: 9466 grad_norm: 0.0216 loss: 0.6859 +2023/05/31 20:57:24 - mmengine - INFO - Epoch(train) [2][ 7600/11516] lr: 9.9391e-04 eta: 1 day, 2:23:04 time: 0.4418 data_time: 0.0011 memory: 9466 grad_norm: 0.0254 loss: 0.6922 +2023/05/31 20:58:07 - mmengine - INFO - Epoch(train) [2][ 7700/11516] lr: 9.9391e-04 eta: 1 day, 2:21:58 time: 0.4440 data_time: 0.0011 memory: 9466 grad_norm: 0.0278 loss: 0.6827 +2023/05/31 20:58:52 - mmengine - INFO - Epoch(train) [2][ 7800/11516] lr: 9.9391e-04 eta: 1 day, 2:21:08 time: 0.4593 data_time: 0.0012 memory: 9466 grad_norm: 0.0216 loss: 0.6884 +2023/05/31 20:59:35 - mmengine - INFO - Epoch(train) [2][ 7900/11516] lr: 9.9391e-04 eta: 1 day, 2:20:08 time: 0.4202 data_time: 0.0012 memory: 9466 grad_norm: 0.0291 loss: 0.6824 +2023/05/31 21:00:19 - mmengine - INFO - Epoch(train) [2][ 8000/11516] lr: 9.9391e-04 eta: 1 day, 2:19:11 time: 0.4508 data_time: 0.0011 memory: 9466 grad_norm: 0.0320 loss: 0.6855 +2023/05/31 21:01:02 - mmengine - INFO - Epoch(train) [2][ 8100/11516] lr: 9.9391e-04 eta: 1 day, 2:18:02 time: 0.4024 data_time: 0.0010 memory: 9466 grad_norm: 0.0335 loss: 0.6862 +2023/05/31 21:01:44 - mmengine - INFO - Epoch(train) [2][ 8200/11516] lr: 9.9391e-04 eta: 1 day, 2:16:45 time: 0.4327 data_time: 0.0011 memory: 9466 grad_norm: 0.0254 loss: 0.6823 +2023/05/31 21:02:26 - mmengine - INFO - Epoch(train) [2][ 8300/11516] lr: 9.9391e-04 eta: 1 day, 2:15:33 time: 0.3986 data_time: 0.0013 memory: 9466 grad_norm: 0.0359 loss: 0.6887 +2023/05/31 21:03:09 - mmengine - INFO - Epoch(train) [2][ 8400/11516] lr: 9.9391e-04 eta: 1 day, 2:14:22 time: 0.4348 data_time: 0.0012 memory: 9466 grad_norm: 0.0259 loss: 0.6873 +2023/05/31 21:03:43 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/05/31 21:03:51 - mmengine - INFO - Epoch(train) [2][ 8500/11516] lr: 9.9391e-04 eta: 1 day, 2:13:08 time: 0.4368 data_time: 0.0011 memory: 9466 grad_norm: 0.0265 loss: 0.6900 +2023/05/31 21:04:35 - mmengine - INFO - Epoch(train) [2][ 8600/11516] lr: 9.9391e-04 eta: 1 day, 2:12:18 time: 0.4175 data_time: 0.0011 memory: 9466 grad_norm: 0.0188 loss: 0.6840 +2023/05/31 21:05:18 - mmengine - INFO - Epoch(train) [2][ 8700/11516] lr: 9.9391e-04 eta: 1 day, 2:11:10 time: 0.4102 data_time: 0.0011 memory: 9466 grad_norm: 0.0304 loss: 0.6832 +2023/05/31 21:05:59 - mmengine - INFO - Epoch(train) [2][ 8800/11516] lr: 9.9391e-04 eta: 1 day, 2:09:49 time: 0.3942 data_time: 0.0011 memory: 9466 grad_norm: 0.0347 loss: 0.6820 +2023/05/31 21:06:41 - mmengine - INFO - Epoch(train) [2][ 8900/11516] lr: 9.9391e-04 eta: 1 day, 2:08:34 time: 0.4059 data_time: 0.0014 memory: 9466 grad_norm: 0.0323 loss: 0.6807 +2023/05/31 21:07:24 - mmengine - INFO - Epoch(train) [2][ 9000/11516] lr: 9.9391e-04 eta: 1 day, 2:07:33 time: 0.4128 data_time: 0.0011 memory: 9466 grad_norm: 0.0248 loss: 0.6879 +2023/05/31 21:08:07 - mmengine - INFO - Epoch(train) [2][ 9100/11516] lr: 9.9391e-04 eta: 1 day, 2:06:28 time: 0.4093 data_time: 0.0012 memory: 9466 grad_norm: 0.0231 loss: 0.6897 +2023/05/31 21:08:50 - mmengine - INFO - Epoch(train) [2][ 9200/11516] lr: 9.9391e-04 eta: 1 day, 2:05:21 time: 0.4601 data_time: 0.0011 memory: 9466 grad_norm: 0.0286 loss: 0.6881 +2023/05/31 21:09:33 - mmengine - INFO - Epoch(train) [2][ 9300/11516] lr: 9.9391e-04 eta: 1 day, 2:04:23 time: 0.4470 data_time: 0.0013 memory: 9466 grad_norm: 0.0397 loss: 0.6876 +2023/05/31 21:10:17 - mmengine - INFO - Epoch(train) [2][ 9400/11516] lr: 9.9391e-04 eta: 1 day, 2:03:25 time: 0.4388 data_time: 0.0012 memory: 9466 grad_norm: 0.0199 loss: 0.6846 +2023/05/31 21:10:53 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/05/31 21:11:01 - mmengine - INFO - Epoch(train) [2][ 9500/11516] lr: 9.9391e-04 eta: 1 day, 2:02:32 time: 0.4471 data_time: 0.0011 memory: 9466 grad_norm: 0.0281 loss: 0.6877 +2023/05/31 21:11:44 - mmengine - INFO - Epoch(train) [2][ 9600/11516] lr: 9.9391e-04 eta: 1 day, 2:01:29 time: 0.4225 data_time: 0.0013 memory: 9466 grad_norm: 0.0282 loss: 0.6905 +2023/05/31 21:12:28 - mmengine - INFO - Epoch(train) [2][ 9700/11516] lr: 9.9391e-04 eta: 1 day, 2:00:40 time: 0.4495 data_time: 0.0012 memory: 9466 grad_norm: 0.0358 loss: 0.6840 +2023/05/31 21:13:12 - mmengine - INFO - Epoch(train) [2][ 9800/11516] lr: 9.9391e-04 eta: 1 day, 1:59:41 time: 0.4285 data_time: 0.0011 memory: 9466 grad_norm: 0.0214 loss: 0.6872 +2023/05/31 21:13:54 - mmengine - INFO - Epoch(train) [2][ 9900/11516] lr: 9.9391e-04 eta: 1 day, 1:58:38 time: 0.4396 data_time: 0.0011 memory: 9466 grad_norm: 0.0251 loss: 0.6862 +2023/05/31 21:14:37 - mmengine - INFO - Epoch(train) [2][10000/11516] lr: 9.9391e-04 eta: 1 day, 1:57:32 time: 0.4404 data_time: 0.0013 memory: 9466 grad_norm: 0.0310 loss: 0.6841 +2023/05/31 21:15:20 - mmengine - INFO - Epoch(train) [2][10100/11516] lr: 9.9391e-04 eta: 1 day, 1:56:31 time: 0.4924 data_time: 0.0010 memory: 9466 grad_norm: 0.0231 loss: 0.6890 +2023/05/31 21:16:03 - mmengine - INFO - Epoch(train) [2][10200/11516] lr: 9.9391e-04 eta: 1 day, 1:55:32 time: 0.4428 data_time: 0.0011 memory: 9466 grad_norm: 0.0172 loss: 0.6831 +2023/05/31 21:16:47 - mmengine - INFO - Epoch(train) [2][10300/11516] lr: 9.9391e-04 eta: 1 day, 1:54:35 time: 0.4359 data_time: 0.0012 memory: 9466 grad_norm: 0.0352 loss: 0.6863 +2023/05/31 21:17:29 - mmengine - INFO - Epoch(train) [2][10400/11516] lr: 9.9391e-04 eta: 1 day, 1:53:26 time: 0.4163 data_time: 0.0014 memory: 9466 grad_norm: 0.0269 loss: 0.6851 +2023/05/31 21:18:04 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/05/31 21:18:12 - mmengine - INFO - Epoch(train) [2][10500/11516] lr: 9.9391e-04 eta: 1 day, 1:52:22 time: 0.4047 data_time: 0.0010 memory: 9466 grad_norm: 0.0252 loss: 0.6891 +2023/05/31 21:18:54 - mmengine - INFO - Epoch(train) [2][10600/11516] lr: 9.9391e-04 eta: 1 day, 1:51:13 time: 0.4211 data_time: 0.0012 memory: 9466 grad_norm: 0.0152 loss: 0.6889 +2023/05/31 21:19:36 - mmengine - INFO - Epoch(train) [2][10700/11516] lr: 9.9391e-04 eta: 1 day, 1:50:04 time: 0.3980 data_time: 0.0013 memory: 9466 grad_norm: 0.0125 loss: 0.6819 +2023/05/31 21:20:18 - mmengine - INFO - Epoch(train) [2][10800/11516] lr: 9.9391e-04 eta: 1 day, 1:48:56 time: 0.4656 data_time: 0.0011 memory: 9466 grad_norm: 0.0245 loss: 0.6845 +2023/05/31 21:21:01 - mmengine - INFO - Epoch(train) [2][10900/11516] lr: 9.9391e-04 eta: 1 day, 1:47:53 time: 0.4046 data_time: 0.0012 memory: 9466 grad_norm: 0.0249 loss: 0.6825 +2023/05/31 21:21:45 - mmengine - INFO - Epoch(train) [2][11000/11516] lr: 9.9391e-04 eta: 1 day, 1:47:03 time: 0.3737 data_time: 0.0011 memory: 9466 grad_norm: 0.0183 loss: 0.6876 +2023/05/31 21:22:29 - mmengine - INFO - Epoch(train) [2][11100/11516] lr: 9.9391e-04 eta: 1 day, 1:46:11 time: 0.4647 data_time: 0.0011 memory: 9466 grad_norm: 0.0223 loss: 0.6854 +2023/05/31 21:23:13 - mmengine - INFO - Epoch(train) [2][11200/11516] lr: 9.9391e-04 eta: 1 day, 1:45:19 time: 0.4137 data_time: 0.0011 memory: 9466 grad_norm: 0.0254 loss: 0.6868 +2023/05/31 21:23:58 - mmengine - INFO - Epoch(train) [2][11300/11516] lr: 9.9391e-04 eta: 1 day, 1:44:42 time: 0.4218 data_time: 0.0013 memory: 9466 grad_norm: 0.0249 loss: 0.6839 +2023/05/31 21:24:40 - mmengine - INFO - Epoch(train) [2][11400/11516] lr: 9.9391e-04 eta: 1 day, 1:43:32 time: 0.3979 data_time: 0.0014 memory: 9466 grad_norm: 0.0314 loss: 0.6872 +2023/05/31 21:25:15 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/05/31 21:25:23 - mmengine - INFO - Epoch(train) [2][11500/11516] lr: 9.9391e-04 eta: 1 day, 1:42:35 time: 0.4825 data_time: 0.0046 memory: 9466 grad_norm: 0.0209 loss: 0.6901 +2023/05/31 21:25:29 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/05/31 21:25:29 - mmengine - INFO - Saving checkpoint at 2 epochs +2023/05/31 21:25:45 - mmengine - INFO - Epoch(val) [2][16/16] accuracy/top1: 100.0000 single-label/precision_classwise: [100.0, 0.0] single-label/recall_classwise: [100.0, 0.0] single-label/f1-score_classwise: [100.0, 0.0] data_time: 0.3063 time: 0.5602 +2023/05/31 21:26:30 - mmengine - INFO - Epoch(train) [3][ 100/11516] lr: 9.7577e-04 eta: 1 day, 1:41:32 time: 0.4401 data_time: 0.0010 memory: 9466 grad_norm: 0.0156 loss: 0.6836 +2023/05/31 21:27:15 - mmengine - INFO - Epoch(train) [3][ 200/11516] lr: 9.7577e-04 eta: 1 day, 1:40:44 time: 0.4418 data_time: 0.0014 memory: 9465 grad_norm: 0.0233 loss: 0.6926 +2023/05/31 21:27:57 - mmengine - INFO - Epoch(train) [3][ 300/11516] lr: 9.7577e-04 eta: 1 day, 1:39:41 time: 0.4060 data_time: 0.0828 memory: 9465 grad_norm: 0.0197 loss: 0.6878 +2023/05/31 21:28:41 - mmengine - INFO - Epoch(train) [3][ 400/11516] lr: 9.7577e-04 eta: 1 day, 1:38:52 time: 0.4709 data_time: 0.0010 memory: 9465 grad_norm: 0.0184 loss: 0.6872 +2023/05/31 21:29:26 - mmengine - INFO - Epoch(train) [3][ 500/11516] lr: 9.7577e-04 eta: 1 day, 1:38:05 time: 0.4315 data_time: 0.0017 memory: 9465 grad_norm: 0.0222 loss: 0.6896 +2023/05/31 21:30:10 - mmengine - INFO - Epoch(train) [3][ 600/11516] lr: 9.7577e-04 eta: 1 day, 1:37:18 time: 0.4765 data_time: 0.0018 memory: 9465 grad_norm: 0.0193 loss: 0.6906 +2023/05/31 21:30:54 - mmengine - INFO - Epoch(train) [3][ 700/11516] lr: 9.7577e-04 eta: 1 day, 1:36:32 time: 0.4802 data_time: 0.0015 memory: 9465 grad_norm: 0.0140 loss: 0.6879 +2023/05/31 21:31:37 - mmengine - INFO - Epoch(train) [3][ 800/11516] lr: 9.7577e-04 eta: 1 day, 1:35:30 time: 0.4002 data_time: 0.0014 memory: 9465 grad_norm: 0.0221 loss: 0.6856 +2023/05/31 21:32:20 - mmengine - INFO - Epoch(train) [3][ 900/11516] lr: 9.7577e-04 eta: 1 day, 1:34:31 time: 0.4503 data_time: 0.0012 memory: 9465 grad_norm: 0.0132 loss: 0.6847 +2023/05/31 21:32:47 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/05/31 21:33:04 - mmengine - INFO - Epoch(train) [3][ 1000/11516] lr: 9.7577e-04 eta: 1 day, 1:33:39 time: 0.4728 data_time: 0.0011 memory: 9465 grad_norm: 0.0265 loss: 0.6887 +2023/05/31 21:33:48 - mmengine - INFO - Epoch(train) [3][ 1100/11516] lr: 9.7577e-04 eta: 1 day, 1:32:54 time: 0.4385 data_time: 0.0010 memory: 9465 grad_norm: 0.0244 loss: 0.6888 +2023/05/31 21:34:30 - mmengine - INFO - Epoch(train) [3][ 1200/11516] lr: 9.7577e-04 eta: 1 day, 1:31:44 time: 0.3953 data_time: 0.0021 memory: 9465 grad_norm: 0.0207 loss: 0.6849 +2023/05/31 21:35:14 - mmengine - INFO - Epoch(train) [3][ 1300/11516] lr: 9.7577e-04 eta: 1 day, 1:30:55 time: 0.4517 data_time: 0.0012 memory: 9465 grad_norm: 0.0227 loss: 0.6876 +2023/05/31 21:35:59 - mmengine - INFO - Epoch(train) [3][ 1400/11516] lr: 9.7577e-04 eta: 1 day, 1:30:14 time: 0.4769 data_time: 0.0020 memory: 9465 grad_norm: 0.0231 loss: 0.6854 +2023/05/31 21:36:41 - mmengine - INFO - Epoch(train) [3][ 1500/11516] lr: 9.7577e-04 eta: 1 day, 1:29:09 time: 0.4407 data_time: 0.0013 memory: 9465 grad_norm: 0.0646 loss: 0.6834 +2023/05/31 21:37:24 - mmengine - INFO - Epoch(train) [3][ 1600/11516] lr: 9.7577e-04 eta: 1 day, 1:28:10 time: 0.3936 data_time: 0.0011 memory: 9465 grad_norm: 0.0408 loss: 0.6834 +2023/05/31 21:38:08 - mmengine - INFO - Epoch(train) [3][ 1700/11516] lr: 9.7577e-04 eta: 1 day, 1:27:20 time: 0.4511 data_time: 0.0011 memory: 9465 grad_norm: 0.0252 loss: 0.6851 +2023/05/31 21:38:52 - mmengine - INFO - Epoch(train) [3][ 1800/11516] lr: 9.7577e-04 eta: 1 day, 1:26:32 time: 0.4055 data_time: 0.0015 memory: 9465 grad_norm: 0.0257 loss: 0.6857 +2023/05/31 21:39:36 - mmengine - INFO - Epoch(train) [3][ 1900/11516] lr: 9.7577e-04 eta: 1 day, 1:25:38 time: 0.4019 data_time: 0.0012 memory: 9465 grad_norm: 0.0276 loss: 0.6887 +2023/05/31 21:40:03 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/05/31 21:40:18 - mmengine - INFO - Epoch(train) [3][ 2000/11516] lr: 9.7577e-04 eta: 1 day, 1:24:38 time: 0.4309 data_time: 0.0018 memory: 9465 grad_norm: 0.0215 loss: 0.6858 +2023/05/31 21:41:03 - mmengine - INFO - Epoch(train) [3][ 2100/11516] lr: 9.7577e-04 eta: 1 day, 1:23:52 time: 0.4016 data_time: 0.0012 memory: 9465 grad_norm: 0.0350 loss: 0.6927 +2023/05/31 21:41:46 - mmengine - INFO - Epoch(train) [3][ 2200/11516] lr: 9.7577e-04 eta: 1 day, 1:22:55 time: 0.4333 data_time: 0.0014 memory: 9465 grad_norm: 0.0287 loss: 0.6882 +2023/05/31 21:42:29 - mmengine - INFO - Epoch(train) [3][ 2300/11516] lr: 9.7577e-04 eta: 1 day, 1:21:57 time: 0.4458 data_time: 0.0015 memory: 9465 grad_norm: 0.0374 loss: 0.6857 +2023/05/31 21:43:13 - mmengine - INFO - Epoch(train) [3][ 2400/11516] lr: 9.7577e-04 eta: 1 day, 1:21:12 time: 0.4265 data_time: 0.0014 memory: 9465 grad_norm: 0.0255 loss: 0.6898 +2023/05/31 21:43:57 - mmengine - INFO - Epoch(train) [3][ 2500/11516] lr: 9.7577e-04 eta: 1 day, 1:20:18 time: 0.4279 data_time: 0.0011 memory: 9465 grad_norm: 0.0317 loss: 0.6847 +2023/05/31 21:44:39 - mmengine - INFO - Epoch(train) [3][ 2600/11516] lr: 9.7577e-04 eta: 1 day, 1:19:15 time: 0.4249 data_time: 0.0011 memory: 9465 grad_norm: 0.0335 loss: 0.6860 +2023/05/31 21:45:22 - mmengine - INFO - Epoch(train) [3][ 2700/11516] lr: 9.7577e-04 eta: 1 day, 1:18:19 time: 0.4043 data_time: 0.0011 memory: 9465 grad_norm: 0.0273 loss: 0.6842 +2023/05/31 21:46:03 - mmengine - INFO - Epoch(train) [3][ 2800/11516] lr: 9.7577e-04 eta: 1 day, 1:17:07 time: 0.4139 data_time: 0.0013 memory: 9465 grad_norm: 0.0275 loss: 0.6905 +2023/05/31 21:46:45 - mmengine - INFO - Epoch(train) [3][ 2900/11516] lr: 9.7577e-04 eta: 1 day, 1:16:01 time: 0.3769 data_time: 0.0014 memory: 9465 grad_norm: 0.0255 loss: 0.6837 +2023/05/31 21:47:13 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/05/31 21:47:29 - mmengine - INFO - Epoch(train) [3][ 3000/11516] lr: 9.7577e-04 eta: 1 day, 1:15:19 time: 0.4408 data_time: 0.0014 memory: 9465 grad_norm: 0.0243 loss: 0.6848 +2023/05/31 21:48:13 - mmengine - INFO - Epoch(train) [3][ 3100/11516] lr: 9.7577e-04 eta: 1 day, 1:14:27 time: 0.4898 data_time: 0.0013 memory: 9465 grad_norm: 0.0231 loss: 0.6848 +2023/05/31 21:48:56 - mmengine - INFO - Epoch(train) [3][ 3200/11516] lr: 9.7577e-04 eta: 1 day, 1:13:28 time: 0.3919 data_time: 0.0021 memory: 9465 grad_norm: 0.0246 loss: 0.6875 +2023/05/31 21:49:43 - mmengine - INFO - Epoch(train) [3][ 3300/11516] lr: 9.7577e-04 eta: 1 day, 1:13:02 time: 0.4524 data_time: 0.0017 memory: 9465 grad_norm: 0.0246 loss: 0.6875 +2023/05/31 21:50:25 - mmengine - INFO - Epoch(train) [3][ 3400/11516] lr: 9.7577e-04 eta: 1 day, 1:11:59 time: 0.3781 data_time: 0.0011 memory: 9465 grad_norm: 0.0195 loss: 0.6878 +2023/05/31 21:51:11 - mmengine - INFO - Epoch(train) [3][ 3500/11516] lr: 9.7577e-04 eta: 1 day, 1:11:26 time: 0.5775 data_time: 0.0011 memory: 9465 grad_norm: 0.0298 loss: 0.6883 +2023/05/31 21:51:54 - mmengine - INFO - Epoch(train) [3][ 3600/11516] lr: 9.7577e-04 eta: 1 day, 1:10:33 time: 0.4970 data_time: 0.0011 memory: 9465 grad_norm: 0.0256 loss: 0.6868 +2023/05/31 21:52:38 - mmengine - INFO - Epoch(train) [3][ 3700/11516] lr: 9.7577e-04 eta: 1 day, 1:09:45 time: 0.4040 data_time: 0.0012 memory: 9465 grad_norm: 0.0260 loss: 0.6870 +2023/05/31 21:53:23 - mmengine - INFO - Epoch(train) [3][ 3800/11516] lr: 9.7577e-04 eta: 1 day, 1:09:02 time: 0.4867 data_time: 0.0012 memory: 9465 grad_norm: 0.0167 loss: 0.6842 +2023/05/31 21:54:08 - mmengine - INFO - Epoch(train) [3][ 3900/11516] lr: 9.7577e-04 eta: 1 day, 1:08:20 time: 0.4553 data_time: 0.0011 memory: 9465 grad_norm: 0.0268 loss: 0.6836 +2023/05/31 21:54:35 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/05/31 21:54:49 - mmengine - INFO - Epoch(train) [3][ 4000/11516] lr: 9.7577e-04 eta: 1 day, 1:07:14 time: 0.4208 data_time: 0.0012 memory: 9465 grad_norm: 0.0151 loss: 0.6878 +2023/05/31 21:55:33 - mmengine - INFO - Epoch(train) [3][ 4100/11516] lr: 9.7577e-04 eta: 1 day, 1:06:26 time: 0.4653 data_time: 0.0012 memory: 9465 grad_norm: 0.0283 loss: 0.6842 +2023/05/31 21:56:16 - mmengine - INFO - Epoch(train) [3][ 4200/11516] lr: 9.7577e-04 eta: 1 day, 1:05:29 time: 0.4014 data_time: 0.0010 memory: 9465 grad_norm: 0.0172 loss: 0.6849 +2023/05/31 21:57:00 - mmengine - INFO - Epoch(train) [3][ 4300/11516] lr: 9.7577e-04 eta: 1 day, 1:04:37 time: 0.4507 data_time: 0.0012 memory: 9465 grad_norm: 0.0299 loss: 0.6846 +2023/05/31 21:57:43 - mmengine - INFO - Epoch(train) [3][ 4400/11516] lr: 9.7577e-04 eta: 1 day, 1:03:44 time: 0.3993 data_time: 0.0011 memory: 9465 grad_norm: 0.0222 loss: 0.6884 +2023/05/31 21:58:27 - mmengine - INFO - Epoch(train) [3][ 4500/11516] lr: 9.7577e-04 eta: 1 day, 1:02:54 time: 0.4295 data_time: 0.0011 memory: 9465 grad_norm: 0.0297 loss: 0.6873 +2023/05/31 21:59:09 - mmengine - INFO - Epoch(train) [3][ 4600/11516] lr: 9.7577e-04 eta: 1 day, 1:01:52 time: 0.4159 data_time: 0.0012 memory: 9465 grad_norm: 0.0249 loss: 0.6867 +2023/05/31 21:59:52 - mmengine - INFO - Epoch(train) [3][ 4700/11516] lr: 9.7577e-04 eta: 1 day, 1:01:01 time: 0.4141 data_time: 0.0012 memory: 9465 grad_norm: 0.0222 loss: 0.6866 +2023/05/31 22:00:36 - mmengine - INFO - Epoch(train) [3][ 4800/11516] lr: 9.7577e-04 eta: 1 day, 1:00:09 time: 0.4324 data_time: 0.0013 memory: 9465 grad_norm: 0.0222 loss: 0.6859 +2023/05/31 22:01:19 - mmengine - INFO - Epoch(train) [3][ 4900/11516] lr: 9.7577e-04 eta: 1 day, 0:59:18 time: 0.4775 data_time: 0.0011 memory: 9465 grad_norm: 0.0211 loss: 0.6843 +2023/05/31 22:01:50 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/05/31 22:02:06 - mmengine - INFO - Epoch(train) [3][ 5000/11516] lr: 9.7577e-04 eta: 1 day, 0:58:53 time: 0.4861 data_time: 0.0011 memory: 9465 grad_norm: 0.0187 loss: 0.6915 +2023/05/31 22:02:51 - mmengine - INFO - Epoch(train) [3][ 5100/11516] lr: 9.7577e-04 eta: 1 day, 0:58:07 time: 0.3929 data_time: 0.0012 memory: 9465 grad_norm: 0.0216 loss: 0.6879 +2023/05/31 22:03:37 - mmengine - INFO - Epoch(train) [3][ 5200/11516] lr: 9.7577e-04 eta: 1 day, 0:57:33 time: 0.4096 data_time: 0.0011 memory: 9465 grad_norm: 0.0215 loss: 0.6859 +2023/05/31 22:04:22 - mmengine - INFO - Epoch(train) [3][ 5300/11516] lr: 9.7577e-04 eta: 1 day, 0:56:55 time: 0.4643 data_time: 0.0011 memory: 9465 grad_norm: 0.0200 loss: 0.6840 +2023/05/31 22:05:06 - mmengine - INFO - Epoch(train) [3][ 5400/11516] lr: 9.7577e-04 eta: 1 day, 0:56:07 time: 0.4720 data_time: 0.0012 memory: 9465 grad_norm: 0.0291 loss: 0.6802 +2023/05/31 22:05:51 - mmengine - INFO - Epoch(train) [3][ 5500/11516] lr: 9.7577e-04 eta: 1 day, 0:55:26 time: 0.4400 data_time: 0.0012 memory: 9465 grad_norm: 0.0264 loss: 0.6879 +2023/05/31 22:06:35 - mmengine - INFO - Epoch(train) [3][ 5600/11516] lr: 9.7577e-04 eta: 1 day, 0:54:38 time: 0.4093 data_time: 0.0012 memory: 9465 grad_norm: 0.0225 loss: 0.6909 +2023/05/31 22:07:16 - mmengine - INFO - Epoch(train) [3][ 5700/11516] lr: 9.7577e-04 eta: 1 day, 0:53:33 time: 0.4298 data_time: 0.0010 memory: 9465 grad_norm: 0.0315 loss: 0.6901 +2023/05/31 22:07:59 - mmengine - INFO - Epoch(train) [3][ 5800/11516] lr: 9.7577e-04 eta: 1 day, 0:52:32 time: 0.4676 data_time: 0.0011 memory: 9465 grad_norm: 0.0240 loss: 0.6860 +2023/05/31 22:08:39 - mmengine - INFO - Epoch(train) [3][ 5900/11516] lr: 9.7577e-04 eta: 1 day, 0:51:21 time: 0.3716 data_time: 0.0013 memory: 9465 grad_norm: 0.0354 loss: 0.6848 +2023/05/31 22:09:06 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/05/31 22:09:21 - mmengine - INFO - Epoch(train) [3][ 6000/11516] lr: 9.7577e-04 eta: 1 day, 0:50:20 time: 0.4040 data_time: 0.0012 memory: 9465 grad_norm: 0.0227 loss: 0.6884 +2023/05/31 22:10:01 - mmengine - INFO - Epoch(train) [3][ 6100/11516] lr: 9.7577e-04 eta: 1 day, 0:49:05 time: 0.3791 data_time: 0.0011 memory: 9465 grad_norm: 0.0294 loss: 0.6878 +2023/05/31 22:10:43 - mmengine - INFO - Epoch(train) [3][ 6200/11516] lr: 9.7577e-04 eta: 1 day, 0:48:02 time: 0.4256 data_time: 0.0011 memory: 9465 grad_norm: 0.0240 loss: 0.6835 +2023/05/31 22:11:24 - mmengine - INFO - Epoch(train) [3][ 6300/11516] lr: 9.7577e-04 eta: 1 day, 0:46:57 time: 0.4725 data_time: 0.0011 memory: 9465 grad_norm: 0.0188 loss: 0.6891 +2023/05/31 22:12:07 - mmengine - INFO - Epoch(train) [3][ 6400/11516] lr: 9.7577e-04 eta: 1 day, 0:46:02 time: 0.3840 data_time: 0.0013 memory: 9465 grad_norm: 0.0155 loss: 0.6822 +2023/05/31 22:12:49 - mmengine - INFO - Epoch(train) [3][ 6500/11516] lr: 9.7577e-04 eta: 1 day, 0:44:58 time: 0.3995 data_time: 0.0014 memory: 9465 grad_norm: 0.0272 loss: 0.6855 +2023/05/31 22:13:31 - mmengine - INFO - Epoch(train) [3][ 6600/11516] lr: 9.7577e-04 eta: 1 day, 0:43:59 time: 0.4032 data_time: 0.0012 memory: 9465 grad_norm: 0.0158 loss: 0.6835 +2023/05/31 22:14:14 - mmengine - INFO - Epoch(train) [3][ 6700/11516] lr: 9.7577e-04 eta: 1 day, 0:43:04 time: 0.4222 data_time: 0.0013 memory: 9465 grad_norm: 0.0219 loss: 0.6898 +2023/05/31 22:14:55 - mmengine - INFO - Epoch(train) [3][ 6800/11516] lr: 9.7577e-04 eta: 1 day, 0:42:02 time: 0.3972 data_time: 0.0011 memory: 9465 grad_norm: 0.0226 loss: 0.6876 +2023/05/31 22:15:39 - mmengine - INFO - Epoch(train) [3][ 6900/11516] lr: 9.7577e-04 eta: 1 day, 0:41:10 time: 0.4479 data_time: 0.0010 memory: 9465 grad_norm: 0.0229 loss: 0.6816 +2023/05/31 22:16:06 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/05/31 22:16:21 - mmengine - INFO - Epoch(train) [3][ 7000/11516] lr: 9.7577e-04 eta: 1 day, 0:40:09 time: 0.3996 data_time: 0.0011 memory: 9465 grad_norm: 0.0316 loss: 0.6903 +2023/05/31 22:17:01 - mmengine - INFO - Epoch(train) [3][ 7100/11516] lr: 9.7577e-04 eta: 1 day, 0:39:02 time: 0.3982 data_time: 0.0010 memory: 9465 grad_norm: 0.0290 loss: 0.6860 +2023/05/31 22:17:43 - mmengine - INFO - Epoch(train) [3][ 7200/11516] lr: 9.7577e-04 eta: 1 day, 0:37:57 time: 0.3853 data_time: 0.0012 memory: 9465 grad_norm: 0.0270 loss: 0.6889 +2023/05/31 22:18:23 - mmengine - INFO - Epoch(train) [3][ 7300/11516] lr: 9.7577e-04 eta: 1 day, 0:36:44 time: 0.3844 data_time: 0.0012 memory: 9465 grad_norm: 0.0385 loss: 0.6865 +2023/05/31 22:19:02 - mmengine - INFO - Epoch(train) [3][ 7400/11516] lr: 9.7577e-04 eta: 1 day, 0:35:29 time: 0.3911 data_time: 0.0012 memory: 9465 grad_norm: 0.0251 loss: 0.6896 +2023/05/31 22:19:44 - mmengine - INFO - Epoch(train) [3][ 7500/11516] lr: 9.7577e-04 eta: 1 day, 0:34:28 time: 0.4421 data_time: 0.0015 memory: 9465 grad_norm: 0.0316 loss: 0.6935 +2023/05/31 22:20:25 - mmengine - INFO - Epoch(train) [3][ 7600/11516] lr: 9.7577e-04 eta: 1 day, 0:33:23 time: 0.3897 data_time: 0.0013 memory: 9465 grad_norm: 0.0241 loss: 0.6928 +2023/05/31 22:21:06 - mmengine - INFO - Epoch(train) [3][ 7700/11516] lr: 9.7577e-04 eta: 1 day, 0:32:19 time: 0.4438 data_time: 0.0011 memory: 9465 grad_norm: 0.0207 loss: 0.6865 +2023/05/31 22:21:49 - mmengine - INFO - Epoch(train) [3][ 7800/11516] lr: 9.7577e-04 eta: 1 day, 0:31:24 time: 0.4161 data_time: 0.0018 memory: 9465 grad_norm: 0.0226 loss: 0.6830 +2023/05/31 22:22:30 - mmengine - INFO - Epoch(train) [3][ 7900/11516] lr: 9.7577e-04 eta: 1 day, 0:30:22 time: 0.3708 data_time: 0.0014 memory: 9465 grad_norm: 0.0236 loss: 0.6823 +2023/05/31 22:22:56 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/05/31 22:23:10 - mmengine - INFO - Epoch(train) [3][ 8000/11516] lr: 9.7577e-04 eta: 1 day, 0:29:07 time: 0.3916 data_time: 0.0011 memory: 9465 grad_norm: 0.0300 loss: 0.6843 +2023/05/31 22:23:51 - mmengine - INFO - Epoch(train) [3][ 8100/11516] lr: 9.7577e-04 eta: 1 day, 0:28:01 time: 0.4167 data_time: 0.0014 memory: 9465 grad_norm: 0.0211 loss: 0.6829 +2023/05/31 22:24:32 - mmengine - INFO - Epoch(train) [3][ 8200/11516] lr: 9.7577e-04 eta: 1 day, 0:26:59 time: 0.4249 data_time: 0.0010 memory: 9465 grad_norm: 0.0178 loss: 0.6875 +2023/05/31 22:25:13 - mmengine - INFO - Epoch(train) [3][ 8300/11516] lr: 9.7577e-04 eta: 1 day, 0:25:54 time: 0.4745 data_time: 0.0011 memory: 9465 grad_norm: 0.0205 loss: 0.6857 +2023/05/31 22:25:56 - mmengine - INFO - Epoch(train) [3][ 8400/11516] lr: 9.7577e-04 eta: 1 day, 0:24:59 time: 0.4194 data_time: 0.0011 memory: 9465 grad_norm: 0.0170 loss: 0.6914 +2023/05/31 22:26:36 - mmengine - INFO - Epoch(train) [3][ 8500/11516] lr: 9.7577e-04 eta: 1 day, 0:23:52 time: 0.4020 data_time: 0.0010 memory: 9465 grad_norm: 0.0265 loss: 0.6877 +2023/05/31 22:27:16 - mmengine - INFO - Epoch(train) [3][ 8600/11516] lr: 9.7577e-04 eta: 1 day, 0:22:43 time: 0.3809 data_time: 0.0011 memory: 9465 grad_norm: 0.0268 loss: 0.6821 +2023/05/31 22:27:59 - mmengine - INFO - Epoch(train) [3][ 8700/11516] lr: 9.7577e-04 eta: 1 day, 0:21:46 time: 0.4136 data_time: 0.0014 memory: 9465 grad_norm: 0.0212 loss: 0.6859 +2023/05/31 22:28:39 - mmengine - INFO - Epoch(train) [3][ 8800/11516] lr: 9.7577e-04 eta: 1 day, 0:20:41 time: 0.4465 data_time: 0.0013 memory: 9465 grad_norm: 0.0159 loss: 0.6898 +2023/05/31 22:29:20 - mmengine - INFO - Epoch(train) [3][ 8900/11516] lr: 9.7577e-04 eta: 1 day, 0:19:33 time: 0.4039 data_time: 0.0011 memory: 9465 grad_norm: 0.0326 loss: 0.6879 +2023/05/31 22:29:46 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/05/31 22:30:00 - mmengine - INFO - Epoch(train) [3][ 9000/11516] lr: 9.7577e-04 eta: 1 day, 0:18:25 time: 0.4214 data_time: 0.0010 memory: 9465 grad_norm: 0.0262 loss: 0.6840 +2023/05/31 22:30:42 - mmengine - INFO - Epoch(train) [3][ 9100/11516] lr: 9.7577e-04 eta: 1 day, 0:17:26 time: 0.4218 data_time: 0.0011 memory: 9465 grad_norm: 0.0273 loss: 0.6802 +2023/05/31 22:31:22 - mmengine - INFO - Epoch(train) [3][ 9200/11516] lr: 9.7577e-04 eta: 1 day, 0:16:21 time: 0.4318 data_time: 0.0011 memory: 9465 grad_norm: 0.0192 loss: 0.6841 +2023/05/31 22:32:03 - mmengine - INFO - Epoch(train) [3][ 9300/11516] lr: 9.7577e-04 eta: 1 day, 0:15:14 time: 0.4215 data_time: 0.0011 memory: 9465 grad_norm: 0.0219 loss: 0.6837 +2023/05/31 22:32:46 - mmengine - INFO - Epoch(train) [3][ 9400/11516] lr: 9.7577e-04 eta: 1 day, 0:14:21 time: 0.4585 data_time: 0.0011 memory: 9465 grad_norm: 0.0269 loss: 0.6887 +2023/05/31 22:33:26 - mmengine - INFO - Epoch(train) [3][ 9500/11516] lr: 9.7577e-04 eta: 1 day, 0:13:16 time: 0.4570 data_time: 0.0011 memory: 9465 grad_norm: 0.0263 loss: 0.6821 +2023/05/31 22:34:09 - mmengine - INFO - Epoch(train) [3][ 9600/11516] lr: 9.7577e-04 eta: 1 day, 0:12:25 time: 0.4684 data_time: 0.0010 memory: 9465 grad_norm: 0.0242 loss: 0.6896 +2023/05/31 22:34:52 - mmengine - INFO - Epoch(train) [3][ 9700/11516] lr: 9.7577e-04 eta: 1 day, 0:11:32 time: 0.4366 data_time: 0.0010 memory: 9465 grad_norm: 0.0199 loss: 0.6848 +2023/05/31 22:35:32 - mmengine - INFO - Epoch(train) [3][ 9800/11516] lr: 9.7577e-04 eta: 1 day, 0:10:27 time: 0.4182 data_time: 0.0011 memory: 9465 grad_norm: 0.0256 loss: 0.6919 +2023/05/31 22:36:13 - mmengine - INFO - Epoch(train) [3][ 9900/11516] lr: 9.7577e-04 eta: 1 day, 0:09:23 time: 0.4447 data_time: 0.0011 memory: 9465 grad_norm: 0.0158 loss: 0.6893 +2023/05/31 22:36:39 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/05/31 22:36:54 - mmengine - INFO - Epoch(train) [3][10000/11516] lr: 9.7577e-04 eta: 1 day, 0:08:19 time: 0.4020 data_time: 0.0012 memory: 9465 grad_norm: 0.0247 loss: 0.6856 +2023/05/31 22:37:35 - mmengine - INFO - Epoch(train) [3][10100/11516] lr: 9.7577e-04 eta: 1 day, 0:07:19 time: 0.4286 data_time: 0.0013 memory: 9465 grad_norm: 0.0223 loss: 0.6881 +2023/05/31 22:38:16 - mmengine - INFO - Epoch(train) [3][10200/11516] lr: 9.7577e-04 eta: 1 day, 0:06:18 time: 0.4505 data_time: 0.0012 memory: 9465 grad_norm: 0.0165 loss: 0.6924 +2023/05/31 22:38:59 - mmengine - INFO - Epoch(train) [3][10300/11516] lr: 9.7577e-04 eta: 1 day, 0:05:26 time: 0.4597 data_time: 0.0010 memory: 9465 grad_norm: 0.0309 loss: 0.6893 +2023/05/31 22:39:40 - mmengine - INFO - Epoch(train) [3][10400/11516] lr: 9.7577e-04 eta: 1 day, 0:04:23 time: 0.4274 data_time: 0.0011 memory: 9465 grad_norm: 0.0253 loss: 0.6842 +2023/05/31 22:40:20 - mmengine - INFO - Epoch(train) [3][10500/11516] lr: 9.7577e-04 eta: 1 day, 0:03:14 time: 0.3994 data_time: 0.0011 memory: 9465 grad_norm: 0.0224 loss: 0.6887 +2023/05/31 22:41:00 - mmengine - INFO - Epoch(train) [3][10600/11516] lr: 9.7577e-04 eta: 1 day, 0:02:09 time: 0.4220 data_time: 0.0017 memory: 9465 grad_norm: 0.0256 loss: 0.6867 +2023/05/31 22:41:40 - mmengine - INFO - Epoch(train) [3][10700/11516] lr: 9.7577e-04 eta: 1 day, 0:01:00 time: 0.4258 data_time: 0.0011 memory: 9465 grad_norm: 0.0155 loss: 0.6840 +2023/05/31 22:42:19 - mmengine - INFO - Epoch(train) [3][10800/11516] lr: 9.7577e-04 eta: 23:59:46 time: 0.4159 data_time: 0.0013 memory: 9465 grad_norm: 0.0236 loss: 0.6917 +2023/05/31 22:42:58 - mmengine - INFO - Epoch(train) [3][10900/11516] lr: 9.7577e-04 eta: 23:58:38 time: 0.3897 data_time: 0.0011 memory: 9465 grad_norm: 0.0301 loss: 0.6870 +2023/05/31 22:43:24 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/05/31 22:43:41 - mmengine - INFO - Epoch(train) [3][11000/11516] lr: 9.7577e-04 eta: 23:57:44 time: 0.4457 data_time: 0.0013 memory: 9465 grad_norm: 0.0252 loss: 0.6834 +2023/05/31 22:44:23 - mmengine - INFO - Epoch(train) [3][11100/11516] lr: 9.7577e-04 eta: 23:56:51 time: 0.4223 data_time: 0.0012 memory: 9465 grad_norm: 0.0224 loss: 0.6860 +2023/05/31 22:45:05 - mmengine - INFO - Epoch(train) [3][11200/11516] lr: 9.7577e-04 eta: 23:55:54 time: 0.4390 data_time: 0.2247 memory: 9465 grad_norm: 0.0296 loss: 0.6807 +2023/05/31 22:45:47 - mmengine - INFO - Epoch(train) [3][11300/11516] lr: 9.7577e-04 eta: 23:55:02 time: 0.3956 data_time: 0.1984 memory: 9465 grad_norm: 0.0218 loss: 0.6878 +2023/05/31 22:46:28 - mmengine - INFO - Epoch(train) [3][11400/11516] lr: 9.7577e-04 eta: 23:53:58 time: 0.3767 data_time: 0.1896 memory: 9465 grad_norm: 0.0288 loss: 0.6858 +2023/05/31 22:47:13 - mmengine - INFO - Epoch(train) [3][11500/11516] lr: 9.7577e-04 eta: 23:53:20 time: 0.3810 data_time: 0.0151 memory: 9465 grad_norm: 0.0243 loss: 0.6859 +2023/05/31 22:47:18 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/05/31 22:47:18 - mmengine - INFO - Saving checkpoint at 3 epochs +2023/05/31 22:47:35 - mmengine - INFO - Epoch(val) [3][16/16] accuracy/top1: 100.0000 single-label/precision_classwise: [100.0, 0.0] single-label/recall_classwise: [100.0, 0.0] single-label/f1-score_classwise: [100.0, 0.0] data_time: 0.3211 time: 0.5737 +2023/05/31 22:48:19 - mmengine - INFO - Epoch(train) [4][ 100/11516] lr: 9.4605e-04 eta: 23:52:23 time: 0.4066 data_time: 0.0018 memory: 9465 grad_norm: 0.0348 loss: 0.6879 +2023/05/31 22:48:59 - mmengine - INFO - Epoch(train) [4][ 200/11516] lr: 9.4605e-04 eta: 23:51:13 time: 0.4006 data_time: 0.0016 memory: 9465 grad_norm: 0.0343 loss: 0.6897 +2023/05/31 22:49:41 - mmengine - INFO - Epoch(train) [4][ 300/11516] lr: 9.4605e-04 eta: 23:50:18 time: 0.4655 data_time: 0.0186 memory: 9465 grad_norm: 0.0148 loss: 0.6880 +2023/05/31 22:50:23 - mmengine - INFO - Epoch(train) [4][ 400/11516] lr: 9.4605e-04 eta: 23:49:28 time: 0.4726 data_time: 0.0017 memory: 9465 grad_norm: 0.0156 loss: 0.6876 +2023/05/31 22:50:46 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/05/31 22:51:05 - mmengine - INFO - Epoch(train) [4][ 500/11516] lr: 9.4605e-04 eta: 23:48:31 time: 0.4212 data_time: 0.0012 memory: 9465 grad_norm: 0.0265 loss: 0.6781 +2023/05/31 22:51:47 - mmengine - INFO - Epoch(train) [4][ 600/11516] lr: 9.4605e-04 eta: 23:47:35 time: 0.4139 data_time: 0.0013 memory: 9465 grad_norm: 0.0204 loss: 0.6843 +2023/05/31 22:52:34 - mmengine - INFO - Epoch(train) [4][ 700/11516] lr: 9.4605e-04 eta: 23:47:11 time: 0.7197 data_time: 0.5352 memory: 9465 grad_norm: 0.0288 loss: 0.6834 +2023/05/31 22:53:30 - mmengine - INFO - Epoch(train) [4][ 800/11516] lr: 9.4605e-04 eta: 23:47:34 time: 0.4068 data_time: 0.2202 memory: 9465 grad_norm: 0.0267 loss: 0.6872 +2023/05/31 22:54:08 - mmengine - INFO - Epoch(train) [4][ 900/11516] lr: 9.4605e-04 eta: 23:46:19 time: 0.3568 data_time: 0.1738 memory: 9465 grad_norm: 0.0232 loss: 0.6831 +2023/05/31 22:54:46 - mmengine - INFO - Epoch(train) [4][ 1000/11516] lr: 9.4605e-04 eta: 23:45:01 time: 0.3715 data_time: 0.0783 memory: 9465 grad_norm: 0.0187 loss: 0.6833 +2023/05/31 22:55:27 - mmengine - INFO - Epoch(train) [4][ 1100/11516] lr: 9.4605e-04 eta: 23:44:00 time: 0.4032 data_time: 0.0011 memory: 9465 grad_norm: 0.0199 loss: 0.6869 +2023/05/31 22:56:08 - mmengine - INFO - Epoch(train) [4][ 1200/11516] lr: 9.4605e-04 eta: 23:43:00 time: 0.4363 data_time: 0.0011 memory: 9465 grad_norm: 0.0181 loss: 0.6897 +2023/05/31 22:56:48 - mmengine - INFO - Epoch(train) [4][ 1300/11516] lr: 9.4605e-04 eta: 23:41:54 time: 0.4167 data_time: 0.0011 memory: 9465 grad_norm: 0.0264 loss: 0.6827 +2023/05/31 22:57:28 - mmengine - INFO - Epoch(train) [4][ 1400/11516] lr: 9.4605e-04 eta: 23:40:53 time: 0.4157 data_time: 0.0011 memory: 9465 grad_norm: 0.0160 loss: 0.6827 +2023/05/31 22:57:48 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/05/31 22:58:07 - mmengine - INFO - Epoch(train) [4][ 1500/11516] lr: 9.4605e-04 eta: 23:39:41 time: 0.3799 data_time: 0.0011 memory: 9465 grad_norm: 0.0329 loss: 0.6906 +2023/05/31 22:58:47 - mmengine - INFO - Epoch(train) [4][ 1600/11516] lr: 9.4605e-04 eta: 23:38:35 time: 0.4036 data_time: 0.0011 memory: 9465 grad_norm: 0.0128 loss: 0.6866 +2023/05/31 22:59:27 - mmengine - INFO - Epoch(train) [4][ 1700/11516] lr: 9.4605e-04 eta: 23:37:30 time: 0.3948 data_time: 0.0011 memory: 9465 grad_norm: 0.0343 loss: 0.6886 +2023/05/31 23:00:08 - mmengine - INFO - Epoch(train) [4][ 1800/11516] lr: 9.4605e-04 eta: 23:36:34 time: 0.4118 data_time: 0.0011 memory: 9465 grad_norm: 0.0376 loss: 0.6902 +2023/05/31 23:00:50 - mmengine - INFO - Epoch(train) [4][ 1900/11516] lr: 9.4605e-04 eta: 23:35:40 time: 0.3553 data_time: 0.0011 memory: 9465 grad_norm: 0.0238 loss: 0.6882 +2023/05/31 23:01:33 - mmengine - INFO - Epoch(train) [4][ 2000/11516] lr: 9.4605e-04 eta: 23:34:50 time: 0.4216 data_time: 0.0011 memory: 9465 grad_norm: 0.0185 loss: 0.6900 +2023/05/31 23:02:13 - mmengine - INFO - Epoch(train) [4][ 2100/11516] lr: 9.4605e-04 eta: 23:33:47 time: 0.4165 data_time: 0.0011 memory: 9465 grad_norm: 0.0245 loss: 0.6846 +2023/05/31 23:02:54 - mmengine - INFO - Epoch(train) [4][ 2200/11516] lr: 9.4605e-04 eta: 23:32:47 time: 0.3853 data_time: 0.0012 memory: 9465 grad_norm: 0.0181 loss: 0.6884 +2023/05/31 23:03:34 - mmengine - INFO - Epoch(train) [4][ 2300/11516] lr: 9.4605e-04 eta: 23:31:46 time: 0.3810 data_time: 0.0011 memory: 9465 grad_norm: 0.0275 loss: 0.6903 +2023/05/31 23:04:15 - mmengine - INFO - Epoch(train) [4][ 2400/11516] lr: 9.4605e-04 eta: 23:30:47 time: 0.3779 data_time: 0.0013 memory: 9465 grad_norm: 0.0242 loss: 0.6884 +2023/05/31 23:04:35 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/05/31 23:04:55 - mmengine - INFO - Epoch(train) [4][ 2500/11516] lr: 9.4605e-04 eta: 23:29:45 time: 0.4320 data_time: 0.0013 memory: 9465 grad_norm: 0.0231 loss: 0.6838 +2023/05/31 23:05:37 - mmengine - INFO - Epoch(train) [4][ 2600/11516] lr: 9.4605e-04 eta: 23:28:49 time: 0.4059 data_time: 0.0011 memory: 9465 grad_norm: 0.0357 loss: 0.6903 +2023/05/31 23:06:17 - mmengine - INFO - Epoch(train) [4][ 2700/11516] lr: 9.4605e-04 eta: 23:27:48 time: 0.3588 data_time: 0.0011 memory: 9465 grad_norm: 0.0215 loss: 0.6827 +2023/05/31 23:06:57 - mmengine - INFO - Epoch(train) [4][ 2800/11516] lr: 9.4605e-04 eta: 23:26:46 time: 0.3968 data_time: 0.0011 memory: 9465 grad_norm: 0.0270 loss: 0.6894 +2023/05/31 23:07:38 - mmengine - INFO - Epoch(train) [4][ 2900/11516] lr: 9.4605e-04 eta: 23:25:47 time: 0.4284 data_time: 0.0011 memory: 9465 grad_norm: 0.0215 loss: 0.6862 +2023/05/31 23:08:18 - mmengine - INFO - Epoch(train) [4][ 3000/11516] lr: 9.4605e-04 eta: 23:24:41 time: 0.3692 data_time: 0.0011 memory: 9465 grad_norm: 0.0157 loss: 0.6852 +2023/05/31 23:08:59 - mmengine - INFO - Epoch(train) [4][ 3100/11516] lr: 9.4605e-04 eta: 23:23:45 time: 0.4189 data_time: 0.0010 memory: 9465 grad_norm: 0.0205 loss: 0.6844 +2023/05/31 23:09:40 - mmengine - INFO - Epoch(train) [4][ 3200/11516] lr: 9.4605e-04 eta: 23:22:48 time: 0.4475 data_time: 0.0011 memory: 9465 grad_norm: 0.0239 loss: 0.6859 +2023/05/31 23:10:21 - mmengine - INFO - Epoch(train) [4][ 3300/11516] lr: 9.4605e-04 eta: 23:21:52 time: 0.4393 data_time: 0.0012 memory: 9465 grad_norm: 0.0172 loss: 0.6845 +2023/05/31 23:11:02 - mmengine - INFO - Epoch(train) [4][ 3400/11516] lr: 9.4605e-04 eta: 23:20:52 time: 0.3671 data_time: 0.0012 memory: 9465 grad_norm: 0.0228 loss: 0.6860 +2023/05/31 23:11:24 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/05/31 23:11:44 - mmengine - INFO - Epoch(train) [4][ 3500/11516] lr: 9.4605e-04 eta: 23:20:01 time: 0.3956 data_time: 0.0012 memory: 9465 grad_norm: 0.0155 loss: 0.6862 +2023/05/31 23:12:23 - mmengine - INFO - Epoch(train) [4][ 3600/11516] lr: 9.4605e-04 eta: 23:18:52 time: 0.3952 data_time: 0.0011 memory: 9465 grad_norm: 0.0259 loss: 0.6884 +2023/05/31 23:13:05 - mmengine - INFO - Epoch(train) [4][ 3700/11516] lr: 9.4605e-04 eta: 23:18:01 time: 0.4146 data_time: 0.0012 memory: 9465 grad_norm: 0.0213 loss: 0.6879 +2023/05/31 23:13:47 - mmengine - INFO - Epoch(train) [4][ 3800/11516] lr: 9.4605e-04 eta: 23:17:10 time: 0.4229 data_time: 0.0011 memory: 9465 grad_norm: 0.0210 loss: 0.6901 +2023/05/31 23:14:29 - mmengine - INFO - Epoch(train) [4][ 3900/11516] lr: 9.4605e-04 eta: 23:16:15 time: 0.4033 data_time: 0.0012 memory: 9465 grad_norm: 0.0204 loss: 0.6883 +2023/05/31 23:15:10 - mmengine - INFO - Epoch(train) [4][ 4000/11516] lr: 9.4605e-04 eta: 23:15:21 time: 0.3834 data_time: 0.0011 memory: 9465 grad_norm: 0.0238 loss: 0.6869 +2023/05/31 23:15:50 - mmengine - INFO - Epoch(train) [4][ 4100/11516] lr: 9.4605e-04 eta: 23:14:18 time: 0.4029 data_time: 0.0010 memory: 9465 grad_norm: 0.0271 loss: 0.6860 +2023/05/31 23:16:31 - mmengine - INFO - Epoch(train) [4][ 4200/11516] lr: 9.4605e-04 eta: 23:13:22 time: 0.3666 data_time: 0.0011 memory: 9465 grad_norm: 0.0260 loss: 0.6888 +2023/05/31 23:17:12 - mmengine - INFO - Epoch(train) [4][ 4300/11516] lr: 9.4605e-04 eta: 23:12:22 time: 0.3954 data_time: 0.0012 memory: 9465 grad_norm: 0.0268 loss: 0.6836 +2023/05/31 23:17:53 - mmengine - INFO - Epoch(train) [4][ 4400/11516] lr: 9.4605e-04 eta: 23:11:28 time: 0.4549 data_time: 0.0013 memory: 9465 grad_norm: 0.0133 loss: 0.6878 +2023/05/31 23:18:15 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/05/31 23:18:36 - mmengine - INFO - Epoch(train) [4][ 4500/11516] lr: 9.4605e-04 eta: 23:10:42 time: 0.4607 data_time: 0.0011 memory: 9465 grad_norm: 0.0171 loss: 0.6893 +2023/05/31 23:19:16 - mmengine - INFO - Epoch(train) [4][ 4600/11516] lr: 9.4605e-04 eta: 23:09:41 time: 0.3970 data_time: 0.0012 memory: 9465 grad_norm: 0.0367 loss: 0.6870 +2023/05/31 23:19:57 - mmengine - INFO - Epoch(train) [4][ 4700/11516] lr: 9.4605e-04 eta: 23:08:41 time: 0.3837 data_time: 0.0013 memory: 9465 grad_norm: 0.0275 loss: 0.6825 +2023/05/31 23:20:40 - mmengine - INFO - Epoch(train) [4][ 4800/11516] lr: 9.4605e-04 eta: 23:07:59 time: 0.4444 data_time: 0.0012 memory: 9465 grad_norm: 0.0251 loss: 0.6823 +2023/05/31 23:21:20 - mmengine - INFO - Epoch(train) [4][ 4900/11516] lr: 9.4605e-04 eta: 23:06:56 time: 0.4316 data_time: 0.0011 memory: 9465 grad_norm: 0.0175 loss: 0.6841 +2023/05/31 23:21:59 - mmengine - INFO - Epoch(train) [4][ 5000/11516] lr: 9.4605e-04 eta: 23:05:52 time: 0.4154 data_time: 0.0012 memory: 9465 grad_norm: 0.0160 loss: 0.6873 +2023/05/31 23:22:40 - mmengine - INFO - Epoch(train) [4][ 5100/11516] lr: 9.4605e-04 eta: 23:04:55 time: 0.3903 data_time: 0.0012 memory: 9465 grad_norm: 0.0176 loss: 0.6892 +2023/05/31 23:23:20 - mmengine - INFO - Epoch(train) [4][ 5200/11516] lr: 9.4605e-04 eta: 23:03:53 time: 0.4172 data_time: 0.0012 memory: 9465 grad_norm: 0.0218 loss: 0.6907 +2023/05/31 23:24:02 - mmengine - INFO - Epoch(train) [4][ 5300/11516] lr: 9.4605e-04 eta: 23:03:04 time: 0.4402 data_time: 0.0013 memory: 9465 grad_norm: 0.0161 loss: 0.6854 +2023/05/31 23:24:42 - mmengine - INFO - Epoch(train) [4][ 5400/11516] lr: 9.4605e-04 eta: 23:02:03 time: 0.3724 data_time: 0.0011 memory: 9465 grad_norm: 0.0234 loss: 0.6845 +2023/05/31 23:25:03 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/05/31 23:25:23 - mmengine - INFO - Epoch(train) [4][ 5500/11516] lr: 9.4605e-04 eta: 23:01:05 time: 0.4076 data_time: 0.0011 memory: 9465 grad_norm: 0.0187 loss: 0.6819 +2023/05/31 23:26:03 - mmengine - INFO - Epoch(train) [4][ 5600/11516] lr: 9.4605e-04 eta: 23:00:07 time: 0.4174 data_time: 0.0011 memory: 9465 grad_norm: 0.0138 loss: 0.6879 +2023/05/31 23:26:44 - mmengine - INFO - Epoch(train) [4][ 5700/11516] lr: 9.4605e-04 eta: 22:59:10 time: 0.4069 data_time: 0.0012 memory: 9465 grad_norm: 0.0209 loss: 0.6850 +2023/05/31 23:27:23 - mmengine - INFO - Epoch(train) [4][ 5800/11516] lr: 9.4605e-04 eta: 22:58:06 time: 0.3831 data_time: 0.0013 memory: 9465 grad_norm: 0.0171 loss: 0.6840 +2023/05/31 23:28:03 - mmengine - INFO - Epoch(train) [4][ 5900/11516] lr: 9.4605e-04 eta: 22:57:05 time: 0.4001 data_time: 0.0011 memory: 9465 grad_norm: 0.0318 loss: 0.6905 +2023/05/31 23:28:43 - mmengine - INFO - Epoch(train) [4][ 6000/11516] lr: 9.4605e-04 eta: 22:56:06 time: 0.4375 data_time: 0.0011 memory: 9465 grad_norm: 0.0137 loss: 0.6867 +2023/05/31 23:29:24 - mmengine - INFO - Epoch(train) [4][ 6100/11516] lr: 9.4605e-04 eta: 22:55:07 time: 0.4556 data_time: 0.0011 memory: 9465 grad_norm: 0.0285 loss: 0.6843 +2023/05/31 23:30:04 - mmengine - INFO - Epoch(train) [4][ 6200/11516] lr: 9.4605e-04 eta: 22:54:12 time: 0.4214 data_time: 0.0010 memory: 9465 grad_norm: 0.0132 loss: 0.6893 +2023/05/31 23:30:44 - mmengine - INFO - Epoch(train) [4][ 6300/11516] lr: 9.4605e-04 eta: 22:53:10 time: 0.3892 data_time: 0.0011 memory: 9465 grad_norm: 0.0297 loss: 0.6814 +2023/05/31 23:31:23 - mmengine - INFO - Epoch(train) [4][ 6400/11516] lr: 9.4605e-04 eta: 22:52:07 time: 0.3738 data_time: 0.0011 memory: 9465 grad_norm: 0.0159 loss: 0.6881 +2023/05/31 23:31:45 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/05/31 23:32:06 - mmengine - INFO - Epoch(train) [4][ 6500/11516] lr: 9.4605e-04 eta: 22:51:20 time: 0.4458 data_time: 0.0011 memory: 9465 grad_norm: 0.0194 loss: 0.6848 +2023/05/31 23:32:48 - mmengine - INFO - Epoch(train) [4][ 6600/11516] lr: 9.4605e-04 eta: 22:50:28 time: 0.4216 data_time: 0.0010 memory: 9465 grad_norm: 0.0309 loss: 0.6798 +2023/05/31 23:33:28 - mmengine - INFO - Epoch(train) [4][ 6700/11516] lr: 9.4605e-04 eta: 22:49:31 time: 0.3929 data_time: 0.0011 memory: 9465 grad_norm: 0.0276 loss: 0.6880 +2023/05/31 23:34:08 - mmengine - INFO - Epoch(train) [4][ 6800/11516] lr: 9.4605e-04 eta: 22:48:30 time: 0.4202 data_time: 0.0012 memory: 9465 grad_norm: 0.0240 loss: 0.6902 +2023/05/31 23:34:49 - mmengine - INFO - Epoch(train) [4][ 6900/11516] lr: 9.4605e-04 eta: 22:47:34 time: 0.4312 data_time: 0.0015 memory: 9465 grad_norm: 0.0306 loss: 0.6885 +2023/05/31 23:35:29 - mmengine - INFO - Epoch(train) [4][ 7000/11516] lr: 9.4605e-04 eta: 22:46:36 time: 0.4038 data_time: 0.0011 memory: 9465 grad_norm: 0.0283 loss: 0.6854 +2023/05/31 23:36:09 - mmengine - INFO - Epoch(train) [4][ 7100/11516] lr: 9.4605e-04 eta: 22:45:36 time: 0.4282 data_time: 0.0011 memory: 9465 grad_norm: 0.0255 loss: 0.6832 +2023/05/31 23:36:49 - mmengine - INFO - Epoch(train) [4][ 7200/11516] lr: 9.4605e-04 eta: 22:44:37 time: 0.4317 data_time: 0.0011 memory: 9465 grad_norm: 0.0271 loss: 0.6852 +2023/05/31 23:37:28 - mmengine - INFO - Epoch(train) [4][ 7300/11516] lr: 9.4605e-04 eta: 22:43:38 time: 0.3934 data_time: 0.0011 memory: 9465 grad_norm: 0.0196 loss: 0.6874 +2023/05/31 23:38:09 - mmengine - INFO - Epoch(train) [4][ 7400/11516] lr: 9.4605e-04 eta: 22:42:43 time: 0.4015 data_time: 0.0011 memory: 9465 grad_norm: 0.0234 loss: 0.6857 +2023/05/31 23:38:31 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/05/31 23:38:50 - mmengine - INFO - Epoch(train) [4][ 7500/11516] lr: 9.4605e-04 eta: 22:41:47 time: 0.4143 data_time: 0.0011 memory: 9465 grad_norm: 0.0228 loss: 0.6842 +2023/05/31 23:39:31 - mmengine - INFO - Epoch(train) [4][ 7600/11516] lr: 9.4605e-04 eta: 22:40:54 time: 0.4790 data_time: 0.0011 memory: 9465 grad_norm: 0.0263 loss: 0.6877 +2023/05/31 23:40:11 - mmengine - INFO - Epoch(train) [4][ 7700/11516] lr: 9.4605e-04 eta: 22:39:54 time: 0.4056 data_time: 0.0011 memory: 9465 grad_norm: 0.0308 loss: 0.6877 +2023/05/31 23:40:52 - mmengine - INFO - Epoch(train) [4][ 7800/11516] lr: 9.4605e-04 eta: 22:38:59 time: 0.4452 data_time: 0.0011 memory: 9465 grad_norm: 0.0210 loss: 0.6931 +2023/05/31 23:41:32 - mmengine - INFO - Epoch(train) [4][ 7900/11516] lr: 9.4605e-04 eta: 22:38:04 time: 0.4099 data_time: 0.0011 memory: 9465 grad_norm: 0.0263 loss: 0.6856 +2023/05/31 23:42:12 - mmengine - INFO - Epoch(train) [4][ 8000/11516] lr: 9.4605e-04 eta: 22:37:06 time: 0.4446 data_time: 0.0011 memory: 9465 grad_norm: 0.0288 loss: 0.6888 +2023/05/31 23:42:54 - mmengine - INFO - Epoch(train) [4][ 8100/11516] lr: 9.4605e-04 eta: 22:36:15 time: 0.4420 data_time: 0.0011 memory: 9465 grad_norm: 0.0340 loss: 0.6887 +2023/05/31 23:43:36 - mmengine - INFO - Epoch(train) [4][ 8200/11516] lr: 9.4605e-04 eta: 22:35:25 time: 0.4010 data_time: 0.0013 memory: 9465 grad_norm: 0.0219 loss: 0.6858 +2023/05/31 23:44:16 - mmengine - INFO - Epoch(train) [4][ 8300/11516] lr: 9.4605e-04 eta: 22:34:29 time: 0.4137 data_time: 0.0013 memory: 9465 grad_norm: 0.0289 loss: 0.6886 +2023/05/31 23:45:00 - mmengine - INFO - Epoch(train) [4][ 8400/11516] lr: 9.4605e-04 eta: 22:33:45 time: 0.4143 data_time: 0.0014 memory: 9465 grad_norm: 0.0334 loss: 0.6899 +2023/05/31 23:45:22 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/05/31 23:45:41 - mmengine - INFO - Epoch(train) [4][ 8500/11516] lr: 9.4605e-04 eta: 22:32:54 time: 0.4109 data_time: 0.0011 memory: 9465 grad_norm: 0.0251 loss: 0.6843 +2023/05/31 23:46:22 - mmengine - INFO - Epoch(train) [4][ 8600/11516] lr: 9.4605e-04 eta: 22:32:00 time: 0.4023 data_time: 0.0012 memory: 9465 grad_norm: 0.3745 loss: 0.6884 +2023/05/31 23:47:03 - mmengine - INFO - Epoch(train) [4][ 8700/11516] lr: 9.4605e-04 eta: 22:31:04 time: 0.3909 data_time: 0.0012 memory: 9465 grad_norm: 0.0376 loss: 0.6868 +2023/05/31 23:47:42 - mmengine - INFO - Epoch(train) [4][ 8800/11516] lr: 9.4605e-04 eta: 22:30:03 time: 0.3657 data_time: 0.0011 memory: 9465 grad_norm: 0.0275 loss: 0.6899 +2023/05/31 23:48:23 - mmengine - INFO - Epoch(train) [4][ 8900/11516] lr: 9.4605e-04 eta: 22:29:10 time: 0.4374 data_time: 0.0012 memory: 9465 grad_norm: 0.0273 loss: 0.6862 +2023/05/31 23:49:04 - mmengine - INFO - Epoch(train) [4][ 9000/11516] lr: 9.4605e-04 eta: 22:28:19 time: 0.4056 data_time: 0.0011 memory: 9465 grad_norm: 0.0184 loss: 0.6917 +2023/05/31 23:49:45 - mmengine - INFO - Epoch(train) [4][ 9100/11516] lr: 9.4605e-04 eta: 22:27:25 time: 0.3888 data_time: 0.0010 memory: 9465 grad_norm: 0.0187 loss: 0.6841 +2023/05/31 23:50:27 - mmengine - INFO - Epoch(train) [4][ 9200/11516] lr: 9.4605e-04 eta: 22:26:34 time: 0.4558 data_time: 0.0011 memory: 9465 grad_norm: 0.0305 loss: 0.6902 +2023/05/31 23:51:07 - mmengine - INFO - Epoch(train) [4][ 9300/11516] lr: 9.4605e-04 eta: 22:25:37 time: 0.3683 data_time: 0.0011 memory: 9465 grad_norm: 0.0221 loss: 0.6826 +2023/05/31 23:51:46 - mmengine - INFO - Epoch(train) [4][ 9400/11516] lr: 9.4605e-04 eta: 22:24:38 time: 0.3993 data_time: 0.0011 memory: 9465 grad_norm: 0.0221 loss: 0.6860 +2023/05/31 23:52:08 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/05/31 23:52:27 - mmengine - INFO - Epoch(train) [4][ 9500/11516] lr: 9.4605e-04 eta: 22:23:43 time: 0.3810 data_time: 0.0013 memory: 9465 grad_norm: 0.0206 loss: 0.6875 +2023/05/31 23:53:06 - mmengine - INFO - Epoch(train) [4][ 9600/11516] lr: 9.4605e-04 eta: 22:22:43 time: 0.3802 data_time: 0.0013 memory: 9465 grad_norm: 0.0171 loss: 0.6902 +2023/05/31 23:53:46 - mmengine - INFO - Epoch(train) [4][ 9700/11516] lr: 9.4605e-04 eta: 22:21:45 time: 0.3496 data_time: 0.0012 memory: 9465 grad_norm: 0.0154 loss: 0.6871 +2023/05/31 23:54:26 - mmengine - INFO - Epoch(train) [4][ 9800/11516] lr: 9.4605e-04 eta: 22:20:48 time: 0.4466 data_time: 0.0011 memory: 9465 grad_norm: 0.0319 loss: 0.6849 +2023/05/31 23:55:05 - mmengine - INFO - Epoch(train) [4][ 9900/11516] lr: 9.4605e-04 eta: 22:19:49 time: 0.3779 data_time: 0.0011 memory: 9465 grad_norm: 0.0191 loss: 0.6856 +2023/05/31 23:55:45 - mmengine - INFO - Epoch(train) [4][10000/11516] lr: 9.4605e-04 eta: 22:18:52 time: 0.3735 data_time: 0.0012 memory: 9465 grad_norm: 0.0246 loss: 0.6909 +2023/05/31 23:56:26 - mmengine - INFO - Epoch(train) [4][10100/11516] lr: 9.4605e-04 eta: 22:17:57 time: 0.4328 data_time: 0.0013 memory: 9465 grad_norm: 0.0155 loss: 0.6853 +2023/05/31 23:57:06 - mmengine - INFO - Epoch(train) [4][10200/11516] lr: 9.4605e-04 eta: 22:17:01 time: 0.4147 data_time: 0.0013 memory: 9465 grad_norm: 0.0220 loss: 0.6852 +2023/05/31 23:57:45 - mmengine - INFO - Epoch(train) [4][10300/11516] lr: 9.4605e-04 eta: 22:16:01 time: 0.3835 data_time: 0.0014 memory: 9465 grad_norm: 0.0283 loss: 0.6891 +2023/05/31 23:58:25 - mmengine - INFO - Epoch(train) [4][10400/11516] lr: 9.4605e-04 eta: 22:15:05 time: 0.4219 data_time: 0.0010 memory: 9465 grad_norm: 0.0200 loss: 0.6843 +2023/05/31 23:58:46 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/05/31 23:59:06 - mmengine - INFO - Epoch(train) [4][10500/11516] lr: 9.4605e-04 eta: 22:14:10 time: 0.3833 data_time: 0.0011 memory: 9465 grad_norm: 0.0197 loss: 0.6835 +2023/05/31 23:59:47 - mmengine - INFO - Epoch(train) [4][10600/11516] lr: 9.4605e-04 eta: 22:13:18 time: 0.4207 data_time: 0.0010 memory: 9465 grad_norm: 0.0218 loss: 0.6866 +2023/06/01 00:00:27 - mmengine - INFO - Epoch(train) [4][10700/11516] lr: 9.4605e-04 eta: 22:12:23 time: 0.3922 data_time: 0.0012 memory: 9465 grad_norm: 0.0221 loss: 0.6841 +2023/06/01 00:01:07 - mmengine - INFO - Epoch(train) [4][10800/11516] lr: 9.4605e-04 eta: 22:11:26 time: 0.4150 data_time: 0.0015 memory: 9465 grad_norm: 0.0142 loss: 0.6886 +2023/06/01 00:01:47 - mmengine - INFO - Epoch(train) [4][10900/11516] lr: 9.4605e-04 eta: 22:10:31 time: 0.3833 data_time: 0.0013 memory: 9465 grad_norm: 0.0245 loss: 0.6844 +2023/06/01 00:02:27 - mmengine - INFO - Epoch(train) [4][11000/11516] lr: 9.4605e-04 eta: 22:09:34 time: 0.3860 data_time: 0.0011 memory: 9465 grad_norm: 0.0301 loss: 0.6890 +2023/06/01 00:03:07 - mmengine - INFO - Epoch(train) [4][11100/11516] lr: 9.4605e-04 eta: 22:08:39 time: 0.3658 data_time: 0.0011 memory: 9465 grad_norm: 0.0424 loss: 0.6898 +2023/06/01 00:03:48 - mmengine - INFO - Epoch(train) [4][11200/11516] lr: 9.4605e-04 eta: 22:07:45 time: 0.3858 data_time: 0.0012 memory: 9465 grad_norm: 0.0176 loss: 0.6829 +2023/06/01 00:04:28 - mmengine - INFO - Epoch(train) [4][11300/11516] lr: 9.4605e-04 eta: 22:06:49 time: 0.3862 data_time: 0.0018 memory: 9465 grad_norm: 0.0250 loss: 0.6883 +2023/06/01 00:05:04 - mmengine - INFO - Epoch(train) [4][11400/11516] lr: 9.4605e-04 eta: 22:05:39 time: 0.4019 data_time: 0.0014 memory: 9465 grad_norm: 0.0292 loss: 0.6861 +2023/06/01 00:05:26 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/06/01 00:05:45 - mmengine - INFO - Epoch(train) [4][11500/11516] lr: 9.4605e-04 eta: 22:04:46 time: 0.3984 data_time: 0.0074 memory: 9465 grad_norm: 0.0248 loss: 0.6878 +2023/06/01 00:05:50 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/06/01 00:05:50 - mmengine - INFO - Saving checkpoint at 4 epochs +2023/06/01 00:06:06 - mmengine - INFO - Epoch(val) [4][16/16] accuracy/top1: 100.0000 single-label/precision_classwise: [100.0, 0.0] single-label/recall_classwise: [100.0, 0.0] single-label/f1-score_classwise: [100.0, 0.0] data_time: 0.2998 time: 0.5582 +2023/06/01 00:06:48 - mmengine - INFO - Epoch(train) [5][ 100/11516] lr: 9.0546e-04 eta: 22:03:43 time: 0.4162 data_time: 0.0012 memory: 9465 grad_norm: 0.0266 loss: 0.6881 +2023/06/01 00:07:28 - mmengine - INFO - Epoch(train) [5][ 200/11516] lr: 9.0546e-04 eta: 22:02:47 time: 0.3711 data_time: 0.0014 memory: 9465 grad_norm: 0.0296 loss: 0.6812 +2023/06/01 00:08:07 - mmengine - INFO - Epoch(train) [5][ 300/11516] lr: 9.0546e-04 eta: 22:01:50 time: 0.3952 data_time: 0.0013 memory: 9465 grad_norm: 0.0092 loss: 0.6855 +2023/06/01 00:08:48 - mmengine - INFO - Epoch(train) [5][ 400/11516] lr: 9.0546e-04 eta: 22:00:56 time: 0.4123 data_time: 0.0013 memory: 9465 grad_norm: 0.0259 loss: 0.6860 +2023/06/01 00:09:28 - mmengine - INFO - Epoch(train) [5][ 500/11516] lr: 9.0546e-04 eta: 22:00:00 time: 0.3992 data_time: 0.0011 memory: 9465 grad_norm: 0.0297 loss: 0.6889 +2023/06/01 00:10:09 - mmengine - INFO - Epoch(train) [5][ 600/11516] lr: 9.0546e-04 eta: 21:59:11 time: 0.4416 data_time: 0.0011 memory: 9465 grad_norm: 0.0278 loss: 0.6881 +2023/06/01 00:10:49 - mmengine - INFO - Epoch(train) [5][ 700/11516] lr: 9.0546e-04 eta: 21:58:15 time: 0.3963 data_time: 0.0011 memory: 9465 grad_norm: 0.0183 loss: 0.6887 +2023/06/01 00:11:30 - mmengine - INFO - Epoch(train) [5][ 800/11516] lr: 9.0546e-04 eta: 21:57:23 time: 0.3821 data_time: 0.0012 memory: 9465 grad_norm: 0.0118 loss: 0.6876 +2023/06/01 00:12:09 - mmengine - INFO - Epoch(train) [5][ 900/11516] lr: 9.0546e-04 eta: 21:56:25 time: 0.3979 data_time: 0.0012 memory: 9465 grad_norm: 0.0199 loss: 0.6813 +2023/06/01 00:12:22 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/06/01 00:12:49 - mmengine - INFO - Epoch(train) [5][ 1000/11516] lr: 9.0546e-04 eta: 21:55:29 time: 0.3741 data_time: 0.0011 memory: 9465 grad_norm: 0.0263 loss: 0.6859 +2023/06/01 00:13:29 - mmengine - INFO - Epoch(train) [5][ 1100/11516] lr: 9.0546e-04 eta: 21:54:35 time: 0.3792 data_time: 0.0011 memory: 9465 grad_norm: 0.0356 loss: 0.6877 +2023/06/01 00:14:09 - mmengine - INFO - Epoch(train) [5][ 1200/11516] lr: 9.0546e-04 eta: 21:53:40 time: 0.4017 data_time: 0.0010 memory: 9465 grad_norm: 0.0204 loss: 0.6863 +2023/06/01 00:14:49 - mmengine - INFO - Epoch(train) [5][ 1300/11516] lr: 9.0546e-04 eta: 21:52:45 time: 0.3819 data_time: 0.0011 memory: 9465 grad_norm: 0.0150 loss: 0.6832 +2023/06/01 00:15:30 - mmengine - INFO - Epoch(train) [5][ 1400/11516] lr: 9.0546e-04 eta: 21:51:52 time: 0.3966 data_time: 0.0011 memory: 9465 grad_norm: 0.0152 loss: 0.6857 +2023/06/01 00:16:10 - mmengine - INFO - Epoch(train) [5][ 1500/11516] lr: 9.0546e-04 eta: 21:50:59 time: 0.4375 data_time: 0.0011 memory: 9465 grad_norm: 0.0241 loss: 0.6896 +2023/06/01 00:16:50 - mmengine - INFO - Epoch(train) [5][ 1600/11516] lr: 9.0546e-04 eta: 21:50:04 time: 0.4044 data_time: 0.0011 memory: 9465 grad_norm: 0.0178 loss: 0.6913 +2023/06/01 00:17:31 - mmengine - INFO - Epoch(train) [5][ 1700/11516] lr: 9.0546e-04 eta: 21:49:12 time: 0.4084 data_time: 0.0010 memory: 9465 grad_norm: 0.0256 loss: 0.6877 +2023/06/01 00:18:10 - mmengine - INFO - Epoch(train) [5][ 1800/11516] lr: 9.0546e-04 eta: 21:48:16 time: 0.3811 data_time: 0.0011 memory: 9465 grad_norm: 0.0268 loss: 0.6869 +2023/06/01 00:18:51 - mmengine - INFO - Epoch(train) [5][ 1900/11516] lr: 9.0546e-04 eta: 21:47:23 time: 0.3894 data_time: 0.0011 memory: 9465 grad_norm: 0.0205 loss: 0.6815 +2023/06/01 00:19:04 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/06/01 00:19:31 - mmengine - INFO - Epoch(train) [5][ 2000/11516] lr: 9.0546e-04 eta: 21:46:28 time: 0.3879 data_time: 0.0011 memory: 9465 grad_norm: 0.0315 loss: 0.6889 +2023/06/01 00:20:12 - mmengine - INFO - Epoch(train) [5][ 2100/11516] lr: 9.0546e-04 eta: 21:45:39 time: 0.4121 data_time: 0.0011 memory: 9465 grad_norm: 0.0230 loss: 0.6834 +2023/06/01 00:20:52 - mmengine - INFO - Epoch(train) [5][ 2200/11516] lr: 9.0546e-04 eta: 21:44:45 time: 0.3934 data_time: 0.0012 memory: 9465 grad_norm: 0.0318 loss: 0.6875 +2023/06/01 00:21:31 - mmengine - INFO - Epoch(train) [5][ 2300/11516] lr: 9.0546e-04 eta: 21:43:47 time: 0.3972 data_time: 0.0011 memory: 9465 grad_norm: 0.0213 loss: 0.6879 +2023/06/01 00:22:12 - mmengine - INFO - Epoch(train) [5][ 2400/11516] lr: 9.0546e-04 eta: 21:42:54 time: 0.3985 data_time: 0.0013 memory: 9465 grad_norm: 0.0224 loss: 0.6829 +2023/06/01 00:22:52 - mmengine - INFO - Epoch(train) [5][ 2500/11516] lr: 9.0546e-04 eta: 21:42:01 time: 0.3717 data_time: 0.0012 memory: 9465 grad_norm: 0.0270 loss: 0.6909 +2023/06/01 00:23:32 - mmengine - INFO - Epoch(train) [5][ 2600/11516] lr: 9.0546e-04 eta: 21:41:07 time: 0.4479 data_time: 0.0011 memory: 9465 grad_norm: 0.0288 loss: 0.6876 +2023/06/01 00:24:13 - mmengine - INFO - Epoch(train) [5][ 2700/11516] lr: 9.0546e-04 eta: 21:40:16 time: 0.4325 data_time: 0.0012 memory: 9465 grad_norm: 0.0202 loss: 0.6805 +2023/06/01 00:24:52 - mmengine - INFO - Epoch(train) [5][ 2800/11516] lr: 9.0546e-04 eta: 21:39:18 time: 0.3839 data_time: 0.0011 memory: 9465 grad_norm: 0.0268 loss: 0.6873 +2023/06/01 00:25:32 - mmengine - INFO - Epoch(train) [5][ 2900/11516] lr: 9.0546e-04 eta: 21:38:27 time: 0.4129 data_time: 0.0011 memory: 9465 grad_norm: 0.0319 loss: 0.6838 +2023/06/01 00:25:45 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/06/01 00:26:11 - mmengine - INFO - Epoch(train) [5][ 3000/11516] lr: 9.0546e-04 eta: 21:37:29 time: 0.3984 data_time: 0.0011 memory: 9465 grad_norm: 0.0175 loss: 0.6862 +2023/06/01 00:26:52 - mmengine - INFO - Epoch(train) [5][ 3100/11516] lr: 9.0546e-04 eta: 21:36:37 time: 0.4750 data_time: 0.0015 memory: 9465 grad_norm: 0.0101 loss: 0.6816 +2023/06/01 00:27:33 - mmengine - INFO - Epoch(train) [5][ 3200/11516] lr: 9.0546e-04 eta: 21:35:46 time: 0.3976 data_time: 0.0013 memory: 9465 grad_norm: 0.0232 loss: 0.6871 +2023/06/01 00:28:12 - mmengine - INFO - Epoch(train) [5][ 3300/11516] lr: 9.0546e-04 eta: 21:34:51 time: 0.3688 data_time: 0.0011 memory: 9465 grad_norm: 0.0288 loss: 0.6840 +2023/06/01 00:28:53 - mmengine - INFO - Epoch(train) [5][ 3400/11516] lr: 9.0546e-04 eta: 21:33:58 time: 0.4345 data_time: 0.0012 memory: 9465 grad_norm: 0.0259 loss: 0.6897 +2023/06/01 00:29:33 - mmengine - INFO - Epoch(train) [5][ 3500/11516] lr: 9.0546e-04 eta: 21:33:06 time: 0.4100 data_time: 0.0012 memory: 9465 grad_norm: 0.0308 loss: 0.6824 +2023/06/01 00:30:13 - mmengine - INFO - Epoch(train) [5][ 3600/11516] lr: 9.0546e-04 eta: 21:32:12 time: 0.3828 data_time: 0.0012 memory: 9465 grad_norm: 0.0270 loss: 0.6890 +2023/06/01 00:30:52 - mmengine - INFO - Epoch(train) [5][ 3700/11516] lr: 9.0546e-04 eta: 21:31:17 time: 0.3668 data_time: 0.0011 memory: 9465 grad_norm: 0.0292 loss: 0.6889 +2023/06/01 00:31:34 - mmengine - INFO - Epoch(train) [5][ 3800/11516] lr: 9.0546e-04 eta: 21:30:28 time: 0.4311 data_time: 0.0011 memory: 9465 grad_norm: 0.0232 loss: 0.6853 +2023/06/01 00:32:12 - mmengine - INFO - Epoch(train) [5][ 3900/11516] lr: 9.0546e-04 eta: 21:29:29 time: 0.3900 data_time: 0.0013 memory: 9465 grad_norm: 0.0233 loss: 0.6893 +2023/06/01 00:32:26 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/06/01 00:32:53 - mmengine - INFO - Epoch(train) [5][ 4000/11516] lr: 9.0546e-04 eta: 21:28:40 time: 0.4112 data_time: 0.0011 memory: 9465 grad_norm: 0.0286 loss: 0.6857 +2023/06/01 00:33:33 - mmengine - INFO - Epoch(train) [5][ 4100/11516] lr: 9.0546e-04 eta: 21:27:45 time: 0.3804 data_time: 0.0011 memory: 9465 grad_norm: 0.0304 loss: 0.6874 +2023/06/01 00:34:12 - mmengine - INFO - Epoch(train) [5][ 4200/11516] lr: 9.0546e-04 eta: 21:26:49 time: 0.4008 data_time: 0.0012 memory: 9465 grad_norm: 0.0265 loss: 0.6893 +2023/06/01 00:34:52 - mmengine - INFO - Epoch(train) [5][ 4300/11516] lr: 9.0546e-04 eta: 21:25:55 time: 0.4039 data_time: 0.0012 memory: 9465 grad_norm: 0.0153 loss: 0.6802 +2023/06/01 00:35:32 - mmengine - INFO - Epoch(train) [5][ 4400/11516] lr: 9.0546e-04 eta: 21:25:01 time: 0.3926 data_time: 0.0012 memory: 9465 grad_norm: 0.0266 loss: 0.6873 +2023/06/01 00:36:12 - mmengine - INFO - Epoch(train) [5][ 4500/11516] lr: 9.0546e-04 eta: 21:24:08 time: 0.4179 data_time: 0.0013 memory: 9465 grad_norm: 0.0181 loss: 0.6870 +2023/06/01 00:36:50 - mmengine - INFO - Epoch(train) [5][ 4600/11516] lr: 9.0546e-04 eta: 21:23:10 time: 0.3686 data_time: 0.0011 memory: 9465 grad_norm: 0.0181 loss: 0.6925 +2023/06/01 00:37:30 - mmengine - INFO - Epoch(train) [5][ 4700/11516] lr: 9.0546e-04 eta: 21:22:16 time: 0.3922 data_time: 0.0012 memory: 9465 grad_norm: 0.0239 loss: 0.6910 +2023/06/01 00:38:09 - mmengine - INFO - Epoch(train) [5][ 4800/11516] lr: 9.0546e-04 eta: 21:21:18 time: 0.3558 data_time: 0.0013 memory: 9465 grad_norm: 0.0306 loss: 0.6896 +2023/06/01 00:38:49 - mmengine - INFO - Epoch(train) [5][ 4900/11516] lr: 9.0546e-04 eta: 21:20:25 time: 0.3989 data_time: 0.0011 memory: 9465 grad_norm: 0.0223 loss: 0.6903 +2023/06/01 00:39:01 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/06/01 00:39:28 - mmengine - INFO - Epoch(train) [5][ 5000/11516] lr: 9.0546e-04 eta: 21:19:30 time: 0.4035 data_time: 0.0015 memory: 9465 grad_norm: 0.0267 loss: 0.6855 +2023/06/01 00:40:07 - mmengine - INFO - Epoch(train) [5][ 5100/11516] lr: 9.0546e-04 eta: 21:18:36 time: 0.3790 data_time: 0.0018 memory: 9465 grad_norm: 0.0255 loss: 0.6854 +2023/06/01 00:40:47 - mmengine - INFO - Epoch(train) [5][ 5200/11516] lr: 9.0546e-04 eta: 21:17:43 time: 0.4187 data_time: 0.0013 memory: 9465 grad_norm: 0.0350 loss: 0.6819 +2023/06/01 00:41:27 - mmengine - INFO - Epoch(train) [5][ 5300/11516] lr: 9.0546e-04 eta: 21:16:48 time: 0.3874 data_time: 0.0014 memory: 9465 grad_norm: 0.0175 loss: 0.6885 +2023/06/01 00:42:07 - mmengine - INFO - Epoch(train) [5][ 5400/11516] lr: 9.0546e-04 eta: 21:15:56 time: 0.4116 data_time: 0.0013 memory: 9465 grad_norm: 0.0296 loss: 0.6863 +2023/06/01 00:42:48 - mmengine - INFO - Epoch(train) [5][ 5500/11516] lr: 9.0546e-04 eta: 21:15:07 time: 0.4139 data_time: 0.0011 memory: 9465 grad_norm: 0.0261 loss: 0.6865 +2023/06/01 00:43:29 - mmengine - INFO - Epoch(train) [5][ 5600/11516] lr: 9.0546e-04 eta: 21:14:17 time: 0.3921 data_time: 0.0010 memory: 9465 grad_norm: 0.0317 loss: 0.6904 +2023/06/01 00:44:09 - mmengine - INFO - Epoch(train) [5][ 5700/11516] lr: 9.0546e-04 eta: 21:13:25 time: 0.3885 data_time: 0.0012 memory: 9465 grad_norm: 0.0160 loss: 0.6852 +2023/06/01 00:44:49 - mmengine - INFO - Epoch(train) [5][ 5800/11516] lr: 9.0546e-04 eta: 21:12:34 time: 0.3833 data_time: 0.0011 memory: 9465 grad_norm: 0.0179 loss: 0.6870 +2023/06/01 00:45:29 - mmengine - INFO - Epoch(train) [5][ 5900/11516] lr: 9.0546e-04 eta: 21:11:40 time: 0.4002 data_time: 0.0012 memory: 9465 grad_norm: 0.0225 loss: 0.6871 +2023/06/01 00:45:42 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/06/01 00:46:10 - mmengine - INFO - Epoch(train) [5][ 6000/11516] lr: 9.0546e-04 eta: 21:10:51 time: 0.4081 data_time: 0.0010 memory: 9465 grad_norm: 0.0250 loss: 0.6886 +2023/06/01 00:46:49 - mmengine - INFO - Epoch(train) [5][ 6100/11516] lr: 9.0546e-04 eta: 21:09:55 time: 0.4007 data_time: 0.0011 memory: 9465 grad_norm: 0.0236 loss: 0.6872 +2023/06/01 00:47:29 - mmengine - INFO - Epoch(train) [5][ 6200/11516] lr: 9.0546e-04 eta: 21:09:05 time: 0.4007 data_time: 0.0012 memory: 9465 grad_norm: 0.0240 loss: 0.6904 +2023/06/01 00:48:09 - mmengine - INFO - Epoch(train) [5][ 6300/11516] lr: 9.0546e-04 eta: 21:08:11 time: 0.4023 data_time: 0.0012 memory: 9465 grad_norm: 0.0277 loss: 0.6899 +2023/06/01 00:48:49 - mmengine - INFO - Epoch(train) [5][ 6400/11516] lr: 9.0546e-04 eta: 21:07:20 time: 0.4304 data_time: 0.0012 memory: 9465 grad_norm: 0.0224 loss: 0.6849 +2023/06/01 00:49:29 - mmengine - INFO - Epoch(train) [5][ 6500/11516] lr: 9.0546e-04 eta: 21:06:28 time: 0.3841 data_time: 0.0012 memory: 9465 grad_norm: 0.0251 loss: 0.6867 +2023/06/01 00:50:10 - mmengine - INFO - Epoch(train) [5][ 6600/11516] lr: 9.0546e-04 eta: 21:05:38 time: 0.3799 data_time: 0.0013 memory: 9465 grad_norm: 0.0203 loss: 0.6934 +2023/06/01 00:50:49 - mmengine - INFO - Epoch(train) [5][ 6700/11516] lr: 9.0546e-04 eta: 21:04:44 time: 0.3992 data_time: 0.0010 memory: 9465 grad_norm: 0.0363 loss: 0.6915 +2023/06/01 00:51:30 - mmengine - INFO - Epoch(train) [5][ 6800/11516] lr: 9.0546e-04 eta: 21:03:54 time: 0.3976 data_time: 0.0013 memory: 9465 grad_norm: 0.0250 loss: 0.6864 +2023/06/01 00:52:10 - mmengine - INFO - Epoch(train) [5][ 6900/11516] lr: 9.0546e-04 eta: 21:03:01 time: 0.3810 data_time: 0.0012 memory: 9465 grad_norm: 0.0178 loss: 0.6871 +2023/06/01 00:52:23 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/06/01 00:52:50 - mmengine - INFO - Epoch(train) [5][ 7000/11516] lr: 9.0546e-04 eta: 21:02:10 time: 0.4036 data_time: 0.0010 memory: 9465 grad_norm: 0.0261 loss: 0.6859 +2023/06/01 00:53:30 - mmengine - INFO - Epoch(train) [5][ 7100/11516] lr: 9.0546e-04 eta: 21:01:18 time: 0.4022 data_time: 0.0011 memory: 9465 grad_norm: 0.0263 loss: 0.6870 +2023/06/01 00:54:09 - mmengine - INFO - Epoch(train) [5][ 7200/11516] lr: 9.0546e-04 eta: 21:00:26 time: 0.3938 data_time: 0.0012 memory: 9465 grad_norm: 0.0180 loss: 0.6889 +2023/06/01 00:54:50 - mmengine - INFO - Epoch(train) [5][ 7300/11516] lr: 9.0546e-04 eta: 20:59:36 time: 0.4256 data_time: 0.0011 memory: 9465 grad_norm: 0.0278 loss: 0.6899 +2023/06/01 00:55:29 - mmengine - INFO - Epoch(train) [5][ 7400/11516] lr: 9.0546e-04 eta: 20:58:40 time: 0.3600 data_time: 0.0011 memory: 9465 grad_norm: 0.0270 loss: 0.6827 +2023/06/01 00:56:09 - mmengine - INFO - Epoch(train) [5][ 7500/11516] lr: 9.0546e-04 eta: 20:57:48 time: 0.3842 data_time: 0.0011 memory: 9465 grad_norm: 0.0248 loss: 0.6896 +2023/06/01 00:56:49 - mmengine - INFO - Epoch(train) [5][ 7600/11516] lr: 9.0546e-04 eta: 20:56:58 time: 0.3979 data_time: 0.0012 memory: 9465 grad_norm: 0.0267 loss: 0.6885 +2023/06/01 00:57:30 - mmengine - INFO - Epoch(train) [5][ 7700/11516] lr: 9.0546e-04 eta: 20:56:09 time: 0.4238 data_time: 0.0011 memory: 9465 grad_norm: 0.0215 loss: 0.6848 +2023/06/01 00:58:11 - mmengine - INFO - Epoch(train) [5][ 7800/11516] lr: 9.0546e-04 eta: 20:55:20 time: 0.3655 data_time: 0.0011 memory: 9465 grad_norm: 0.0173 loss: 0.6870 +2023/06/01 00:58:50 - mmengine - INFO - Epoch(train) [5][ 7900/11516] lr: 9.0546e-04 eta: 20:54:27 time: 0.3865 data_time: 0.0011 memory: 9465 grad_norm: 0.0192 loss: 0.6887 +2023/06/01 00:59:03 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/06/01 00:59:30 - mmengine - INFO - Epoch(train) [5][ 8000/11516] lr: 9.0546e-04 eta: 20:53:36 time: 0.3936 data_time: 0.0012 memory: 9465 grad_norm: 0.0193 loss: 0.6880 +2023/06/01 01:00:10 - mmengine - INFO - Epoch(train) [5][ 8100/11516] lr: 9.0546e-04 eta: 20:52:45 time: 0.3610 data_time: 0.0012 memory: 9465 grad_norm: 0.0205 loss: 0.6911 +2023/06/01 01:00:50 - mmengine - INFO - Epoch(train) [5][ 8200/11516] lr: 9.0546e-04 eta: 20:51:51 time: 0.3869 data_time: 0.0012 memory: 9465 grad_norm: 0.0230 loss: 0.6905 +2023/06/01 01:01:30 - mmengine - INFO - Epoch(train) [5][ 8300/11516] lr: 9.0546e-04 eta: 20:51:00 time: 0.3878 data_time: 0.0012 memory: 9465 grad_norm: 0.0165 loss: 0.6883 +2023/06/01 01:02:09 - mmengine - INFO - Epoch(train) [5][ 8400/11516] lr: 9.0546e-04 eta: 20:50:07 time: 0.3749 data_time: 0.0011 memory: 9465 grad_norm: 0.0187 loss: 0.6865 +2023/06/01 01:02:50 - mmengine - INFO - Epoch(train) [5][ 8500/11516] lr: 9.0546e-04 eta: 20:49:17 time: 0.3984 data_time: 0.0011 memory: 9465 grad_norm: 0.0297 loss: 0.6901 +2023/06/01 01:03:29 - mmengine - INFO - Epoch(train) [5][ 8600/11516] lr: 9.0546e-04 eta: 20:48:25 time: 0.3941 data_time: 0.0011 memory: 9465 grad_norm: 0.0147 loss: 0.6826 +2023/06/01 01:04:09 - mmengine - INFO - Epoch(train) [5][ 8700/11516] lr: 9.0546e-04 eta: 20:47:33 time: 0.3965 data_time: 0.0011 memory: 9465 grad_norm: 0.0200 loss: 0.6886 +2023/06/01 01:04:49 - mmengine - INFO - Epoch(train) [5][ 8800/11516] lr: 9.0546e-04 eta: 20:46:41 time: 0.4018 data_time: 0.0011 memory: 9465 grad_norm: 0.0184 loss: 0.6849 +2023/06/01 01:05:30 - mmengine - INFO - Epoch(train) [5][ 8900/11516] lr: 9.0546e-04 eta: 20:45:52 time: 0.4081 data_time: 0.0013 memory: 9465 grad_norm: 0.0309 loss: 0.6882 +2023/06/01 01:05:43 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/06/01 01:06:08 - mmengine - INFO - Epoch(train) [5][ 9000/11516] lr: 9.0546e-04 eta: 20:44:57 time: 0.4052 data_time: 0.0010 memory: 9465 grad_norm: 0.0282 loss: 0.6849 +2023/06/01 01:06:50 - mmengine - INFO - Epoch(train) [5][ 9100/11516] lr: 9.0546e-04 eta: 20:44:10 time: 0.4347 data_time: 0.0011 memory: 9465 grad_norm: 0.0230 loss: 0.6810 +2023/06/01 01:08:35 - mmengine - INFO - Epoch(train) [5][ 9200/11516] lr: 9.0546e-04 eta: 20:46:46 time: 0.3958 data_time: 0.1967 memory: 9465 grad_norm: 0.0276 loss: 0.6841 +2023/06/01 01:09:14 - mmengine - INFO - Epoch(train) [5][ 9300/11516] lr: 9.0546e-04 eta: 20:45:51 time: 0.4309 data_time: 0.2462 memory: 9465 grad_norm: 0.0207 loss: 0.6868 +2023/06/01 01:09:52 - mmengine - INFO - Epoch(train) [5][ 9400/11516] lr: 9.0546e-04 eta: 20:44:54 time: 0.3400 data_time: 0.1575 memory: 9465 grad_norm: 0.0179 loss: 0.6869 +2023/06/01 01:10:31 - mmengine - INFO - Epoch(train) [5][ 9500/11516] lr: 9.0546e-04 eta: 20:44:01 time: 0.3926 data_time: 0.0013 memory: 9465 grad_norm: 0.0279 loss: 0.6849 +2023/06/01 01:11:12 - mmengine - INFO - Epoch(train) [5][ 9600/11516] lr: 9.0546e-04 eta: 20:43:11 time: 0.3732 data_time: 0.0012 memory: 9465 grad_norm: 0.0112 loss: 0.6929 +2023/06/01 01:11:51 - mmengine - INFO - Epoch(train) [5][ 9700/11516] lr: 9.0546e-04 eta: 20:42:18 time: 0.3847 data_time: 0.0013 memory: 9465 grad_norm: 0.0322 loss: 0.6909 +2023/06/01 01:12:32 - mmengine - INFO - Epoch(train) [5][ 9800/11516] lr: 9.0546e-04 eta: 20:41:28 time: 0.4013 data_time: 0.0015 memory: 9465 grad_norm: 0.0177 loss: 0.6873 +2023/06/01 01:13:11 - mmengine - INFO - Epoch(train) [5][ 9900/11516] lr: 9.0546e-04 eta: 20:40:35 time: 0.3882 data_time: 0.0017 memory: 9465 grad_norm: 0.0249 loss: 0.6895 +2023/06/01 01:13:24 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/06/01 01:13:52 - mmengine - INFO - Epoch(train) [5][10000/11516] lr: 9.0546e-04 eta: 20:39:46 time: 0.3908 data_time: 0.0012 memory: 9465 grad_norm: 0.0290 loss: 0.6850 +2023/06/01 01:14:31 - mmengine - INFO - Epoch(train) [5][10100/11516] lr: 9.0546e-04 eta: 20:38:53 time: 0.3827 data_time: 0.0011 memory: 9465 grad_norm: 0.0240 loss: 0.6875 +2023/06/01 01:15:11 - mmengine - INFO - Epoch(train) [5][10200/11516] lr: 9.0546e-04 eta: 20:38:02 time: 0.4017 data_time: 0.0013 memory: 9465 grad_norm: 0.0261 loss: 0.6829 +2023/06/01 01:15:50 - mmengine - INFO - Epoch(train) [5][10300/11516] lr: 9.0546e-04 eta: 20:37:09 time: 0.4014 data_time: 0.0012 memory: 9465 grad_norm: 0.0261 loss: 0.6879 +2023/06/01 01:16:31 - mmengine - INFO - Epoch(train) [5][10400/11516] lr: 9.0546e-04 eta: 20:36:21 time: 0.3909 data_time: 0.0011 memory: 9465 grad_norm: 0.0249 loss: 0.6895 +2023/06/01 01:17:10 - mmengine - INFO - Epoch(train) [5][10500/11516] lr: 9.0546e-04 eta: 20:35:27 time: 0.3915 data_time: 0.0011 memory: 9465 grad_norm: 0.0254 loss: 0.6872 +2023/06/01 01:17:50 - mmengine - INFO - Epoch(train) [5][10600/11516] lr: 9.0546e-04 eta: 20:34:35 time: 0.3909 data_time: 0.0013 memory: 9465 grad_norm: 0.0230 loss: 0.6882 +2023/06/01 01:18:31 - mmengine - INFO - Epoch(train) [5][10700/11516] lr: 9.0546e-04 eta: 20:33:46 time: 0.3998 data_time: 0.0010 memory: 9465 grad_norm: 0.0246 loss: 0.6884 +2023/06/01 01:19:11 - mmengine - INFO - Epoch(train) [5][10800/11516] lr: 9.0546e-04 eta: 20:32:57 time: 0.4460 data_time: 0.0011 memory: 9465 grad_norm: 0.0267 loss: 0.6889 +2023/06/01 01:19:52 - mmengine - INFO - Epoch(train) [5][10900/11516] lr: 9.0546e-04 eta: 20:32:08 time: 0.4158 data_time: 0.0011 memory: 9465 grad_norm: 0.0261 loss: 0.6899 +2023/06/01 01:20:05 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/06/01 01:20:31 - mmengine - INFO - Epoch(train) [5][11000/11516] lr: 9.0546e-04 eta: 20:31:15 time: 0.3817 data_time: 0.0013 memory: 9465 grad_norm: 0.0241 loss: 0.6927 +2023/06/01 01:21:11 - mmengine - INFO - Epoch(train) [5][11100/11516] lr: 9.0546e-04 eta: 20:30:25 time: 0.4084 data_time: 0.0012 memory: 9465 grad_norm: 0.0283 loss: 0.6903 +2023/06/01 01:21:52 - mmengine - INFO - Epoch(train) [5][11200/11516] lr: 9.0546e-04 eta: 20:29:37 time: 0.3946 data_time: 0.0012 memory: 9465 grad_norm: 0.0248 loss: 0.6855 +2023/06/01 01:22:32 - mmengine - INFO - Epoch(train) [5][11300/11516] lr: 9.0546e-04 eta: 20:28:47 time: 0.3747 data_time: 0.0012 memory: 9465 grad_norm: 0.0234 loss: 0.6881 +2023/06/01 01:23:11 - mmengine - INFO - Epoch(train) [5][11400/11516] lr: 9.0546e-04 eta: 20:27:53 time: 0.4330 data_time: 0.0019 memory: 9465 grad_norm: 0.0180 loss: 0.6855 +2023/06/01 01:23:51 - mmengine - INFO - Epoch(train) [5][11500/11516] lr: 9.0546e-04 eta: 20:27:02 time: 0.3930 data_time: 0.0072 memory: 9465 grad_norm: 0.0161 loss: 0.6846 +2023/06/01 01:23:56 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/06/01 01:23:56 - mmengine - INFO - Saving checkpoint at 5 epochs +2023/06/01 01:24:12 - mmengine - INFO - Epoch(val) [5][16/16] accuracy/top1: 100.0000 single-label/precision_classwise: [100.0, 0.0] single-label/recall_classwise: [100.0, 0.0] single-label/f1-score_classwise: [100.0, 0.0] data_time: 0.3097 time: 0.5619 +2023/06/01 01:24:53 - mmengine - INFO - Epoch(train) [6][ 100/11516] lr: 8.5502e-04 eta: 20:26:01 time: 0.3676 data_time: 0.1031 memory: 9465 grad_norm: 0.0261 loss: 0.6888 +2023/06/01 01:25:32 - mmengine - INFO - Epoch(train) [6][ 200/11516] lr: 8.5502e-04 eta: 20:25:09 time: 0.3881 data_time: 0.0012 memory: 9465 grad_norm: 0.0164 loss: 0.6853 +2023/06/01 01:26:12 - mmengine - INFO - Epoch(train) [6][ 300/11516] lr: 8.5502e-04 eta: 20:24:17 time: 0.3818 data_time: 0.0011 memory: 9465 grad_norm: 0.0266 loss: 0.6839 +2023/06/01 01:26:53 - mmengine - INFO - Epoch(train) [6][ 400/11516] lr: 8.5502e-04 eta: 20:23:30 time: 0.3932 data_time: 0.0011 memory: 9465 grad_norm: 0.0248 loss: 0.6888 +2023/06/01 01:27:01 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/06/01 01:27:33 - mmengine - INFO - Epoch(train) [6][ 500/11516] lr: 8.5502e-04 eta: 20:22:41 time: 0.4020 data_time: 0.0011 memory: 9465 grad_norm: 0.0233 loss: 0.6845 +2023/06/01 01:28:13 - mmengine - INFO - Epoch(train) [6][ 600/11516] lr: 8.5502e-04 eta: 20:21:51 time: 0.3959 data_time: 0.0012 memory: 9465 grad_norm: 0.0150 loss: 0.6902 +2023/06/01 01:28:53 - mmengine - INFO - Epoch(train) [6][ 700/11516] lr: 8.5502e-04 eta: 20:20:59 time: 0.3639 data_time: 0.0011 memory: 9465 grad_norm: 0.0192 loss: 0.6878 +2023/06/01 01:29:34 - mmengine - INFO - Epoch(train) [6][ 800/11516] lr: 8.5502e-04 eta: 20:20:12 time: 0.3892 data_time: 0.0011 memory: 9465 grad_norm: 0.0094 loss: 0.6884 +2023/06/01 01:30:14 - mmengine - INFO - Epoch(train) [6][ 900/11516] lr: 8.5502e-04 eta: 20:19:23 time: 0.4119 data_time: 0.0011 memory: 9465 grad_norm: 0.0350 loss: 0.6903 +2023/06/01 01:30:54 - mmengine - INFO - Epoch(train) [6][ 1000/11516] lr: 8.5502e-04 eta: 20:18:32 time: 0.4099 data_time: 0.0012 memory: 9465 grad_norm: 0.0302 loss: 0.6892 +2023/06/01 01:31:33 - mmengine - INFO - Epoch(train) [6][ 1100/11516] lr: 8.5502e-04 eta: 20:17:38 time: 0.3778 data_time: 0.0011 memory: 9465 grad_norm: 0.0214 loss: 0.6876 +2023/06/01 01:32:11 - mmengine - INFO - Epoch(train) [6][ 1200/11516] lr: 8.5502e-04 eta: 20:16:44 time: 0.3939 data_time: 0.0011 memory: 9465 grad_norm: 0.0237 loss: 0.6921 +2023/06/01 01:32:53 - mmengine - INFO - Epoch(train) [6][ 1300/11516] lr: 8.5502e-04 eta: 20:15:58 time: 0.4462 data_time: 0.0012 memory: 9465 grad_norm: 0.0114 loss: 0.6873 +2023/06/01 01:33:31 - mmengine - INFO - Epoch(train) [6][ 1400/11516] lr: 8.5502e-04 eta: 20:15:02 time: 0.3830 data_time: 0.0011 memory: 9465 grad_norm: 0.0308 loss: 0.6823 +2023/06/01 01:33:40 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/06/01 01:34:13 - mmengine - INFO - Epoch(train) [6][ 1500/11516] lr: 8.5502e-04 eta: 20:14:19 time: 0.4046 data_time: 0.0011 memory: 9465 grad_norm: 0.0201 loss: 0.6852 +2023/06/01 01:34:53 - mmengine - INFO - Epoch(train) [6][ 1600/11516] lr: 8.5502e-04 eta: 20:13:29 time: 0.3738 data_time: 0.0027 memory: 9465 grad_norm: 0.0164 loss: 0.6864 +2023/06/01 01:35:34 - mmengine - INFO - Epoch(train) [6][ 1700/11516] lr: 8.5502e-04 eta: 20:12:41 time: 0.4363 data_time: 0.0012 memory: 9465 grad_norm: 0.0222 loss: 0.6865 +2023/06/01 01:36:15 - mmengine - INFO - Epoch(train) [6][ 1800/11516] lr: 8.5502e-04 eta: 20:11:55 time: 0.4015 data_time: 0.0010 memory: 9465 grad_norm: 0.0266 loss: 0.6915 +2023/06/01 01:36:56 - mmengine - INFO - Epoch(train) [6][ 1900/11516] lr: 8.5502e-04 eta: 20:11:07 time: 0.4184 data_time: 0.0011 memory: 9465 grad_norm: 0.0274 loss: 0.6868 +2023/06/01 01:37:35 - mmengine - INFO - Epoch(train) [6][ 2000/11516] lr: 8.5502e-04 eta: 20:10:16 time: 0.4199 data_time: 0.0011 memory: 9465 grad_norm: 0.0215 loss: 0.6934 +2023/06/01 01:38:15 - mmengine - INFO - Epoch(train) [6][ 2100/11516] lr: 8.5502e-04 eta: 20:09:26 time: 0.3899 data_time: 0.0010 memory: 9465 grad_norm: 0.0283 loss: 0.6866 +2023/06/01 01:38:55 - mmengine - INFO - Epoch(train) [6][ 2200/11516] lr: 8.5502e-04 eta: 20:08:37 time: 0.4043 data_time: 0.0012 memory: 9465 grad_norm: 0.0163 loss: 0.6849 +2023/06/01 01:39:36 - mmengine - INFO - Epoch(train) [6][ 2300/11516] lr: 8.5502e-04 eta: 20:07:50 time: 0.4310 data_time: 0.0011 memory: 9465 grad_norm: 0.0212 loss: 0.6871 +2023/06/01 01:40:16 - mmengine - INFO - Epoch(train) [6][ 2400/11516] lr: 8.5502e-04 eta: 20:07:00 time: 0.3818 data_time: 0.0011 memory: 9465 grad_norm: 0.0213 loss: 0.6841 +2023/06/01 01:40:25 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/06/01 01:40:56 - mmengine - INFO - Epoch(train) [6][ 2500/11516] lr: 8.5502e-04 eta: 20:06:10 time: 0.4078 data_time: 0.0011 memory: 9465 grad_norm: 0.0209 loss: 0.6857 +2023/06/01 01:41:36 - mmengine - INFO - Epoch(train) [6][ 2600/11516] lr: 8.5502e-04 eta: 20:05:20 time: 0.4549 data_time: 0.0014 memory: 9465 grad_norm: 0.0256 loss: 0.6941 +2023/06/01 01:42:18 - mmengine - INFO - Epoch(train) [6][ 2700/11516] lr: 8.5502e-04 eta: 20:04:34 time: 0.3902 data_time: 0.0011 memory: 9465 grad_norm: 0.0175 loss: 0.6862 +2023/06/01 01:42:58 - mmengine - INFO - Epoch(train) [6][ 2800/11516] lr: 8.5502e-04 eta: 20:03:45 time: 0.3981 data_time: 0.0012 memory: 9465 grad_norm: 0.0153 loss: 0.6832 +2023/06/01 01:43:38 - mmengine - INFO - Epoch(train) [6][ 2900/11516] lr: 8.5502e-04 eta: 20:02:57 time: 0.4164 data_time: 0.0011 memory: 9465 grad_norm: 0.0192 loss: 0.6912 +2023/06/01 01:44:20 - mmengine - INFO - Epoch(train) [6][ 3000/11516] lr: 8.5502e-04 eta: 20:02:12 time: 0.4330 data_time: 0.0011 memory: 9465 grad_norm: 0.0260 loss: 0.6903 +2023/06/01 01:45:01 - mmengine - INFO - Epoch(train) [6][ 3100/11516] lr: 8.5502e-04 eta: 20:01:25 time: 0.4027 data_time: 0.0011 memory: 9465 grad_norm: 0.0208 loss: 0.6834 +2023/06/01 01:45:41 - mmengine - INFO - Epoch(train) [6][ 3200/11516] lr: 8.5502e-04 eta: 20:00:36 time: 0.3965 data_time: 0.0012 memory: 9465 grad_norm: 0.0148 loss: 0.6909 +2023/06/01 01:46:21 - mmengine - INFO - Epoch(train) [6][ 3300/11516] lr: 8.5502e-04 eta: 19:59:48 time: 0.3891 data_time: 0.0011 memory: 9465 grad_norm: 0.0268 loss: 0.6863 +2023/06/01 01:47:02 - mmengine - INFO - Epoch(train) [6][ 3400/11516] lr: 8.5502e-04 eta: 19:59:01 time: 0.4083 data_time: 0.0011 memory: 9465 grad_norm: 0.0324 loss: 0.6916 +2023/06/01 01:47:11 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/06/01 01:47:44 - mmengine - INFO - Epoch(train) [6][ 3500/11516] lr: 8.5502e-04 eta: 19:58:17 time: 0.4270 data_time: 0.0011 memory: 9465 grad_norm: 0.0301 loss: 0.6843 +2023/06/01 01:48:26 - mmengine - INFO - Epoch(train) [6][ 3600/11516] lr: 8.5502e-04 eta: 19:57:33 time: 0.4436 data_time: 0.0012 memory: 9465 grad_norm: 0.0289 loss: 0.6860 +2023/06/01 01:49:07 - mmengine - INFO - Epoch(train) [6][ 3700/11516] lr: 8.5502e-04 eta: 19:56:46 time: 0.4158 data_time: 0.0013 memory: 9465 grad_norm: 0.0286 loss: 0.6838 +2023/06/01 01:49:47 - mmengine - INFO - Epoch(train) [6][ 3800/11516] lr: 8.5502e-04 eta: 19:55:55 time: 0.4052 data_time: 0.0011 memory: 9465 grad_norm: 0.0239 loss: 0.6811 +2023/06/01 01:50:29 - mmengine - INFO - Epoch(train) [6][ 3900/11516] lr: 8.5502e-04 eta: 19:55:13 time: 0.4475 data_time: 0.0014 memory: 9465 grad_norm: 0.0259 loss: 0.6897 +2023/06/01 01:51:10 - mmengine - INFO - Epoch(train) [6][ 4000/11516] lr: 8.5502e-04 eta: 19:54:25 time: 0.3874 data_time: 0.0011 memory: 9465 grad_norm: 0.0116 loss: 0.6843 +2023/06/01 01:51:50 - mmengine - INFO - Epoch(train) [6][ 4100/11516] lr: 8.5502e-04 eta: 19:53:37 time: 0.3699 data_time: 0.0011 memory: 9465 grad_norm: 0.0201 loss: 0.6913 +2023/06/01 01:52:32 - mmengine - INFO - Epoch(train) [6][ 4200/11516] lr: 8.5502e-04 eta: 19:52:54 time: 0.4076 data_time: 0.0011 memory: 9465 grad_norm: 0.0162 loss: 0.6849 +2023/06/01 01:53:13 - mmengine - INFO - Epoch(train) [6][ 4300/11516] lr: 8.5502e-04 eta: 19:52:06 time: 0.4347 data_time: 0.0012 memory: 9465 grad_norm: 0.0269 loss: 0.6850 +2023/06/01 01:53:54 - mmengine - INFO - Epoch(train) [6][ 4400/11516] lr: 8.5502e-04 eta: 19:51:19 time: 0.4216 data_time: 0.0011 memory: 9465 grad_norm: 0.0241 loss: 0.6830 +2023/06/01 01:54:01 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/06/01 01:54:34 - mmengine - INFO - Epoch(train) [6][ 4500/11516] lr: 8.5502e-04 eta: 19:50:31 time: 0.4641 data_time: 0.0011 memory: 9465 grad_norm: 0.0160 loss: 0.6874 +2023/06/01 01:55:15 - mmengine - INFO - Epoch(train) [6][ 4600/11516] lr: 8.5502e-04 eta: 19:49:44 time: 0.3876 data_time: 0.0014 memory: 9465 grad_norm: 0.0211 loss: 0.6865 +2023/06/01 01:55:55 - mmengine - INFO - Epoch(train) [6][ 4700/11516] lr: 8.5502e-04 eta: 19:48:55 time: 0.3865 data_time: 0.0011 memory: 9465 grad_norm: 0.0253 loss: 0.6892 +2023/06/01 01:56:35 - mmengine - INFO - Epoch(train) [6][ 4800/11516] lr: 8.5502e-04 eta: 19:48:08 time: 0.4205 data_time: 0.0012 memory: 9465 grad_norm: 0.0183 loss: 0.6870 +2023/06/01 01:57:16 - mmengine - INFO - Epoch(train) [6][ 4900/11516] lr: 8.5502e-04 eta: 19:47:21 time: 0.4113 data_time: 0.0012 memory: 9465 grad_norm: 0.0213 loss: 0.6879 +2023/06/01 01:57:57 - mmengine - INFO - Epoch(train) [6][ 5000/11516] lr: 8.5502e-04 eta: 19:46:34 time: 0.4213 data_time: 0.0013 memory: 9465 grad_norm: 0.0271 loss: 0.6856 +2023/06/01 01:58:38 - mmengine - INFO - Epoch(train) [6][ 5100/11516] lr: 8.5502e-04 eta: 19:45:47 time: 0.4197 data_time: 0.0014 memory: 9465 grad_norm: 0.0309 loss: 0.6903 +2023/06/01 01:59:19 - mmengine - INFO - Epoch(train) [6][ 5200/11516] lr: 8.5502e-04 eta: 19:45:01 time: 0.3900 data_time: 0.0012 memory: 9465 grad_norm: 0.0172 loss: 0.6841 +2023/06/01 01:59:59 - mmengine - INFO - Epoch(train) [6][ 5300/11516] lr: 8.5502e-04 eta: 19:44:12 time: 0.4087 data_time: 0.0011 memory: 9465 grad_norm: 0.0337 loss: 0.6852 +2023/06/01 02:00:41 - mmengine - INFO - Epoch(train) [6][ 5400/11516] lr: 8.5502e-04 eta: 19:43:28 time: 0.4005 data_time: 0.0011 memory: 9465 grad_norm: 0.0223 loss: 0.6884 +2023/06/01 02:00:49 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/06/01 02:01:21 - mmengine - INFO - Epoch(train) [6][ 5500/11516] lr: 8.5502e-04 eta: 19:42:41 time: 0.4137 data_time: 0.0012 memory: 9465 grad_norm: 0.0180 loss: 0.6883 +2023/06/01 02:02:01 - mmengine - INFO - Epoch(train) [6][ 5600/11516] lr: 8.5502e-04 eta: 19:41:51 time: 0.3946 data_time: 0.0011 memory: 9465 grad_norm: 0.0161 loss: 0.6900 +2023/06/01 02:02:41 - mmengine - INFO - Epoch(train) [6][ 5700/11516] lr: 8.5502e-04 eta: 19:41:02 time: 0.4165 data_time: 0.0012 memory: 9465 grad_norm: 0.0219 loss: 0.6898 +2023/06/01 02:03:22 - mmengine - INFO - Epoch(train) [6][ 5800/11516] lr: 8.5502e-04 eta: 19:40:16 time: 0.4099 data_time: 0.0012 memory: 9465 grad_norm: 0.0249 loss: 0.6884 +2023/06/01 02:04:03 - mmengine - INFO - Epoch(train) [6][ 5900/11516] lr: 8.5502e-04 eta: 19:39:29 time: 0.3999 data_time: 0.0011 memory: 9465 grad_norm: 0.0274 loss: 0.6885 +2023/06/01 02:04:43 - mmengine - INFO - Epoch(train) [6][ 6000/11516] lr: 8.5502e-04 eta: 19:38:42 time: 0.4092 data_time: 0.0012 memory: 9465 grad_norm: 0.0261 loss: 0.6823 +2023/06/01 02:05:25 - mmengine - INFO - Epoch(train) [6][ 6100/11516] lr: 8.5502e-04 eta: 19:37:58 time: 0.4456 data_time: 0.0011 memory: 9465 grad_norm: 0.0231 loss: 0.6835 +2023/06/01 02:06:06 - mmengine - INFO - Epoch(train) [6][ 6200/11516] lr: 8.5502e-04 eta: 19:37:11 time: 0.3936 data_time: 0.0012 memory: 9465 grad_norm: 0.0183 loss: 0.6867 +2023/06/01 02:06:48 - mmengine - INFO - Epoch(train) [6][ 6300/11516] lr: 8.5502e-04 eta: 19:36:28 time: 0.4847 data_time: 0.0011 memory: 9465 grad_norm: 0.0250 loss: 0.6824 +2023/06/01 02:07:30 - mmengine - INFO - Epoch(train) [6][ 6400/11516] lr: 8.5502e-04 eta: 19:35:43 time: 0.3961 data_time: 0.0011 memory: 9465 grad_norm: 0.0181 loss: 0.6876 +2023/06/01 02:07:38 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/06/01 02:08:12 - mmengine - INFO - Epoch(train) [6][ 6500/11516] lr: 8.5502e-04 eta: 19:35:00 time: 0.4246 data_time: 0.0011 memory: 9465 grad_norm: 0.0189 loss: 0.6884 +2023/06/01 02:08:52 - mmengine - INFO - Epoch(train) [6][ 6600/11516] lr: 8.5502e-04 eta: 19:34:13 time: 0.3710 data_time: 0.0011 memory: 9465 grad_norm: 0.0259 loss: 0.6852 +2023/06/01 02:09:33 - mmengine - INFO - Epoch(train) [6][ 6700/11516] lr: 8.5502e-04 eta: 19:33:26 time: 0.3934 data_time: 0.0012 memory: 9465 grad_norm: 0.0291 loss: 0.6852 +2023/06/01 02:10:14 - mmengine - INFO - Epoch(train) [6][ 6800/11516] lr: 8.5502e-04 eta: 19:32:40 time: 0.3818 data_time: 0.0011 memory: 9465 grad_norm: 0.0154 loss: 0.6896 +2023/06/01 02:10:57 - mmengine - INFO - Epoch(train) [6][ 6900/11516] lr: 8.5502e-04 eta: 19:31:59 time: 0.4220 data_time: 0.0011 memory: 9465 grad_norm: 0.0233 loss: 0.6893 +2023/06/01 02:11:38 - mmengine - INFO - Epoch(train) [6][ 7000/11516] lr: 8.5502e-04 eta: 19:31:14 time: 0.4356 data_time: 0.0011 memory: 9465 grad_norm: 0.0118 loss: 0.6862 +2023/06/01 02:12:19 - mmengine - INFO - Epoch(train) [6][ 7100/11516] lr: 8.5502e-04 eta: 19:30:28 time: 0.4023 data_time: 0.0015 memory: 9465 grad_norm: 0.0242 loss: 0.6829 +2023/06/01 02:13:00 - mmengine - INFO - Epoch(train) [6][ 7200/11516] lr: 8.5502e-04 eta: 19:29:42 time: 0.4315 data_time: 0.0012 memory: 9465 grad_norm: 0.0335 loss: 0.6884 +2023/06/01 02:13:43 - mmengine - INFO - Epoch(train) [6][ 7300/11516] lr: 8.5502e-04 eta: 19:28:59 time: 0.4185 data_time: 0.0012 memory: 9465 grad_norm: 0.0258 loss: 0.6862 +2023/06/01 02:14:23 - mmengine - INFO - Epoch(train) [6][ 7400/11516] lr: 8.5502e-04 eta: 19:28:11 time: 0.3934 data_time: 0.0012 memory: 9465 grad_norm: 0.0214 loss: 0.6873 +2023/06/01 02:14:31 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/06/01 02:15:04 - mmengine - INFO - Epoch(train) [6][ 7500/11516] lr: 8.5502e-04 eta: 19:27:25 time: 0.4054 data_time: 0.0011 memory: 9465 grad_norm: 0.0216 loss: 0.6842 +2023/06/01 02:15:43 - mmengine - INFO - Epoch(train) [6][ 7600/11516] lr: 8.5502e-04 eta: 19:26:35 time: 0.3870 data_time: 0.0011 memory: 9465 grad_norm: 0.0254 loss: 0.6899 +2023/06/01 02:16:23 - mmengine - INFO - Epoch(train) [6][ 7700/11516] lr: 8.5502e-04 eta: 19:25:45 time: 0.4116 data_time: 0.0012 memory: 9465 grad_norm: 0.0205 loss: 0.6837 +2023/06/01 02:17:05 - mmengine - INFO - Epoch(train) [6][ 7800/11516] lr: 8.5502e-04 eta: 19:25:03 time: 0.4677 data_time: 0.0011 memory: 9465 grad_norm: 0.0211 loss: 0.6861 +2023/06/01 02:17:45 - mmengine - INFO - Epoch(train) [6][ 7900/11516] lr: 8.5502e-04 eta: 19:24:15 time: 0.4064 data_time: 0.0012 memory: 9465 grad_norm: 0.0180 loss: 0.6894 +2023/06/01 02:18:25 - mmengine - INFO - Epoch(train) [6][ 8000/11516] lr: 8.5502e-04 eta: 19:23:27 time: 0.3806 data_time: 0.0011 memory: 9465 grad_norm: 0.0271 loss: 0.6889 +2023/06/01 02:19:06 - mmengine - INFO - Epoch(train) [6][ 8100/11516] lr: 8.5502e-04 eta: 19:22:41 time: 0.3879 data_time: 0.0012 memory: 9465 grad_norm: 0.0214 loss: 0.6910 +2023/06/01 02:19:48 - mmengine - INFO - Epoch(train) [6][ 8200/11516] lr: 8.5502e-04 eta: 19:21:56 time: 0.3795 data_time: 0.0012 memory: 9465 grad_norm: 0.0163 loss: 0.6805 +2023/06/01 02:20:29 - mmengine - INFO - Epoch(train) [6][ 8300/11516] lr: 8.5502e-04 eta: 19:21:12 time: 0.3948 data_time: 0.0010 memory: 9465 grad_norm: 0.0244 loss: 0.6871 +2023/06/01 02:21:10 - mmengine - INFO - Epoch(train) [6][ 8400/11516] lr: 8.5502e-04 eta: 19:20:26 time: 0.4532 data_time: 0.0012 memory: 9465 grad_norm: 0.0177 loss: 0.6872 +2023/06/01 02:21:18 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/06/01 02:21:53 - mmengine - INFO - Epoch(train) [6][ 8500/11516] lr: 8.5502e-04 eta: 19:19:44 time: 0.4154 data_time: 0.0011 memory: 9465 grad_norm: 0.0195 loss: 0.6906 +2023/06/01 02:22:33 - mmengine - INFO - Epoch(train) [6][ 8600/11516] lr: 8.5502e-04 eta: 19:18:57 time: 0.4014 data_time: 0.0012 memory: 9465 grad_norm: 0.0350 loss: 0.6900 +2023/06/01 02:23:15 - mmengine - INFO - Epoch(train) [6][ 8700/11516] lr: 8.5502e-04 eta: 19:18:12 time: 0.4042 data_time: 0.0011 memory: 9465 grad_norm: 0.0230 loss: 0.6868 +2023/06/01 02:23:55 - mmengine - INFO - Epoch(train) [6][ 8800/11516] lr: 8.5502e-04 eta: 19:17:25 time: 0.4069 data_time: 0.0012 memory: 9465 grad_norm: 0.0254 loss: 0.6918 +2023/06/01 02:24:35 - mmengine - INFO - Epoch(train) [6][ 8900/11516] lr: 8.5502e-04 eta: 19:16:38 time: 0.4094 data_time: 0.0012 memory: 9465 grad_norm: 0.0252 loss: 0.6869 +2023/06/01 02:25:16 - mmengine - INFO - Epoch(train) [6][ 9000/11516] lr: 8.5502e-04 eta: 19:15:51 time: 0.4043 data_time: 0.0013 memory: 9465 grad_norm: 0.0277 loss: 0.6838 +2023/06/01 02:25:58 - mmengine - INFO - Epoch(train) [6][ 9100/11516] lr: 8.5502e-04 eta: 19:15:07 time: 0.4100 data_time: 0.0012 memory: 9465 grad_norm: 0.0296 loss: 0.6818 +2023/06/01 02:26:38 - mmengine - INFO - Epoch(train) [6][ 9200/11516] lr: 8.5502e-04 eta: 19:14:20 time: 0.4110 data_time: 0.0012 memory: 9465 grad_norm: 0.0169 loss: 0.6904 +2023/06/01 02:27:19 - mmengine - INFO - Epoch(train) [6][ 9300/11516] lr: 8.5502e-04 eta: 19:13:34 time: 0.3954 data_time: 0.0013 memory: 9465 grad_norm: 0.0236 loss: 0.6841 +2023/06/01 02:28:00 - mmengine - INFO - Epoch(train) [6][ 9400/11516] lr: 8.5502e-04 eta: 19:12:48 time: 0.4060 data_time: 0.0012 memory: 9465 grad_norm: 0.0191 loss: 0.6901 +2023/06/01 02:28:08 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/06/01 02:28:41 - mmengine - INFO - Epoch(train) [6][ 9500/11516] lr: 8.5502e-04 eta: 19:12:03 time: 0.4300 data_time: 0.0012 memory: 9465 grad_norm: 0.0202 loss: 0.6868 +2023/06/01 02:29:22 - mmengine - INFO - Epoch(train) [6][ 9600/11516] lr: 8.5502e-04 eta: 19:11:16 time: 0.3859 data_time: 0.0012 memory: 9465 grad_norm: 0.0234 loss: 0.6802 +2023/06/01 02:30:02 - mmengine - INFO - Epoch(train) [6][ 9700/11516] lr: 8.5502e-04 eta: 19:10:29 time: 0.4053 data_time: 0.0012 memory: 9465 grad_norm: 0.0212 loss: 0.6821 +2023/06/01 02:30:44 - mmengine - INFO - Epoch(train) [6][ 9800/11516] lr: 8.5502e-04 eta: 19:09:45 time: 0.3934 data_time: 0.0012 memory: 9465 grad_norm: 0.0197 loss: 0.6860 +2023/06/01 02:31:24 - mmengine - INFO - Epoch(train) [6][ 9900/11516] lr: 8.5502e-04 eta: 19:08:57 time: 0.4117 data_time: 0.0011 memory: 9465 grad_norm: 0.0240 loss: 0.6860 +2023/06/01 02:32:05 - mmengine - INFO - Epoch(train) [6][10000/11516] lr: 8.5502e-04 eta: 19:08:11 time: 0.3942 data_time: 0.0011 memory: 9465 grad_norm: 0.0238 loss: 0.6898 +2023/06/01 02:32:46 - mmengine - INFO - Epoch(train) [6][10100/11516] lr: 8.5502e-04 eta: 19:07:27 time: 0.4362 data_time: 0.0012 memory: 9465 grad_norm: 0.0149 loss: 0.6895 +2023/06/01 02:33:28 - mmengine - INFO - Epoch(train) [6][10200/11516] lr: 8.5502e-04 eta: 19:06:43 time: 0.4204 data_time: 0.0013 memory: 9465 grad_norm: 0.0228 loss: 0.6915 +2023/06/01 02:34:08 - mmengine - INFO - Epoch(train) [6][10300/11516] lr: 8.5502e-04 eta: 19:05:55 time: 0.3963 data_time: 0.0012 memory: 9465 grad_norm: 0.0167 loss: 0.6868 +2023/06/01 02:34:49 - mmengine - INFO - Epoch(train) [6][10400/11516] lr: 8.5502e-04 eta: 19:05:09 time: 0.4070 data_time: 0.0012 memory: 9465 grad_norm: 0.0158 loss: 0.6871 +2023/06/01 02:34:57 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/06/01 02:35:29 - mmengine - INFO - Epoch(train) [6][10500/11516] lr: 8.5502e-04 eta: 19:04:23 time: 0.4242 data_time: 0.0013 memory: 9465 grad_norm: 0.0170 loss: 0.6894 +2023/06/01 02:36:09 - mmengine - INFO - Epoch(train) [6][10600/11516] lr: 8.5502e-04 eta: 19:03:35 time: 0.4323 data_time: 0.0013 memory: 9465 grad_norm: 0.0222 loss: 0.6860 +2023/06/01 02:36:50 - mmengine - INFO - Epoch(train) [6][10700/11516] lr: 8.5502e-04 eta: 19:02:49 time: 0.3881 data_time: 0.0012 memory: 9465 grad_norm: 0.0153 loss: 0.6899 +2023/06/01 02:37:30 - mmengine - INFO - Epoch(train) [6][10800/11516] lr: 8.5502e-04 eta: 19:02:00 time: 0.4153 data_time: 0.0012 memory: 9465 grad_norm: 0.0352 loss: 0.6839 +2023/06/01 02:38:10 - mmengine - INFO - Epoch(train) [6][10900/11516] lr: 8.5502e-04 eta: 19:01:12 time: 0.3775 data_time: 0.0012 memory: 9465 grad_norm: 0.0261 loss: 0.6918 +2023/06/01 02:38:50 - mmengine - INFO - Epoch(train) [6][11000/11516] lr: 8.5502e-04 eta: 19:00:25 time: 0.4147 data_time: 0.0012 memory: 9465 grad_norm: 0.0180 loss: 0.6895 +2023/06/01 02:39:32 - mmengine - INFO - Epoch(train) [6][11100/11516] lr: 8.5502e-04 eta: 18:59:42 time: 0.4142 data_time: 0.0016 memory: 9465 grad_norm: 0.0175 loss: 0.6869 +2023/06/01 02:40:13 - mmengine - INFO - Epoch(train) [6][11200/11516] lr: 8.5502e-04 eta: 18:58:56 time: 0.3898 data_time: 0.0012 memory: 9465 grad_norm: 0.0282 loss: 0.6860 +2023/06/01 02:40:51 - mmengine - INFO - Epoch(train) [6][11300/11516] lr: 8.5502e-04 eta: 18:58:04 time: 0.3601 data_time: 0.0013 memory: 9465 grad_norm: 0.0337 loss: 0.6845 +2023/06/01 02:41:30 - mmengine - INFO - Epoch(train) [6][11400/11516] lr: 8.5502e-04 eta: 18:57:13 time: 0.4357 data_time: 0.0012 memory: 9465 grad_norm: 0.0192 loss: 0.6847 +2023/06/01 02:41:38 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/06/01 02:42:11 - mmengine - INFO - Epoch(train) [6][11500/11516] lr: 8.5502e-04 eta: 18:56:28 time: 0.3996 data_time: 0.0073 memory: 9465 grad_norm: 0.0191 loss: 0.6824 +2023/06/01 02:42:16 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/06/01 02:42:16 - mmengine - INFO - Saving checkpoint at 6 epochs +2023/06/01 02:42:33 - mmengine - INFO - Epoch(val) [6][16/16] accuracy/top1: 100.0000 single-label/precision_classwise: [100.0, 0.0] single-label/recall_classwise: [100.0, 0.0] single-label/f1-score_classwise: [100.0, 0.0] data_time: 0.3055 time: 0.5641 +2023/06/01 02:43:16 - mmengine - INFO - Epoch(train) [7][ 100/11516] lr: 7.9595e-04 eta: 18:55:39 time: 0.4473 data_time: 0.0226 memory: 9465 grad_norm: 0.0132 loss: 0.6878 +2023/06/01 02:43:59 - mmengine - INFO - Epoch(train) [7][ 200/11516] lr: 7.9595e-04 eta: 18:54:57 time: 0.4476 data_time: 0.0011 memory: 9465 grad_norm: 0.0285 loss: 0.6901 +2023/06/01 02:44:39 - mmengine - INFO - Epoch(train) [7][ 300/11516] lr: 7.9595e-04 eta: 18:54:10 time: 0.4229 data_time: 0.0013 memory: 9465 grad_norm: 0.0232 loss: 0.6806 +2023/06/01 02:45:19 - mmengine - INFO - Epoch(train) [7][ 400/11516] lr: 7.9595e-04 eta: 18:53:24 time: 0.4265 data_time: 0.0015 memory: 9465 grad_norm: 0.0261 loss: 0.6891 +2023/06/01 02:46:00 - mmengine - INFO - Epoch(train) [7][ 500/11516] lr: 7.9595e-04 eta: 18:52:37 time: 0.3576 data_time: 0.0011 memory: 9465 grad_norm: 0.0209 loss: 0.6852 +2023/06/01 02:46:40 - mmengine - INFO - Epoch(train) [7][ 600/11516] lr: 7.9595e-04 eta: 18:51:51 time: 0.4385 data_time: 0.0012 memory: 9465 grad_norm: 0.0220 loss: 0.6836 +2023/06/01 02:47:21 - mmengine - INFO - Epoch(train) [7][ 700/11516] lr: 7.9595e-04 eta: 18:51:05 time: 0.4129 data_time: 0.0011 memory: 9465 grad_norm: 0.0229 loss: 0.6840 +2023/06/01 02:48:02 - mmengine - INFO - Epoch(train) [7][ 800/11516] lr: 7.9595e-04 eta: 18:50:19 time: 0.4388 data_time: 0.0013 memory: 9465 grad_norm: 0.0137 loss: 0.6849 +2023/06/01 02:48:43 - mmengine - INFO - Epoch(train) [7][ 900/11516] lr: 7.9595e-04 eta: 18:49:34 time: 0.3797 data_time: 0.0012 memory: 9465 grad_norm: 0.0144 loss: 0.6857 +2023/06/01 02:48:43 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/06/01 02:49:26 - mmengine - INFO - Epoch(train) [7][ 1000/11516] lr: 7.9595e-04 eta: 18:48:54 time: 0.4348 data_time: 0.0026 memory: 9465 grad_norm: 0.0233 loss: 0.6845 +2023/06/01 02:50:06 - mmengine - INFO - Epoch(train) [7][ 1100/11516] lr: 7.9595e-04 eta: 18:48:07 time: 0.3985 data_time: 0.0011 memory: 9465 grad_norm: 0.0232 loss: 0.6841 +2023/06/01 02:50:48 - mmengine - INFO - Epoch(train) [7][ 1200/11516] lr: 7.9595e-04 eta: 18:47:24 time: 0.4709 data_time: 0.0012 memory: 9465 grad_norm: 0.0176 loss: 0.6880 +2023/06/01 02:51:30 - mmengine - INFO - Epoch(train) [7][ 1300/11516] lr: 7.9595e-04 eta: 18:46:41 time: 0.3998 data_time: 0.0012 memory: 9465 grad_norm: 0.0237 loss: 0.6894 +2023/06/01 02:52:10 - mmengine - INFO - Epoch(train) [7][ 1400/11516] lr: 7.9595e-04 eta: 18:45:54 time: 0.4117 data_time: 0.0011 memory: 9465 grad_norm: 0.0263 loss: 0.6901 +2023/06/01 02:52:50 - mmengine - INFO - Epoch(train) [7][ 1500/11516] lr: 7.9595e-04 eta: 18:45:07 time: 0.3990 data_time: 0.0013 memory: 9465 grad_norm: 0.0275 loss: 0.6841 +2023/06/01 02:53:31 - mmengine - INFO - Epoch(train) [7][ 1600/11516] lr: 7.9595e-04 eta: 18:44:22 time: 0.4035 data_time: 0.0011 memory: 9465 grad_norm: 0.0319 loss: 0.6849 +2023/06/01 02:54:13 - mmengine - INFO - Epoch(train) [7][ 1700/11516] lr: 7.9595e-04 eta: 18:43:39 time: 0.4155 data_time: 0.0028 memory: 9465 grad_norm: 0.0255 loss: 0.6865 +2023/06/01 02:54:52 - mmengine - INFO - Epoch(train) [7][ 1800/11516] lr: 7.9595e-04 eta: 18:42:49 time: 0.3875 data_time: 0.0011 memory: 9465 grad_norm: 0.0144 loss: 0.6866 +2023/06/01 02:55:34 - mmengine - INFO - Epoch(train) [7][ 1900/11516] lr: 7.9595e-04 eta: 18:42:05 time: 0.4038 data_time: 0.0014 memory: 9465 grad_norm: 0.0346 loss: 0.6866 +2023/06/01 02:55:35 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/06/01 02:56:14 - mmengine - INFO - Epoch(train) [7][ 2000/11516] lr: 7.9595e-04 eta: 18:41:18 time: 0.3991 data_time: 0.0012 memory: 9465 grad_norm: 0.0184 loss: 0.6870 +2023/06/01 02:56:54 - mmengine - INFO - Epoch(train) [7][ 2100/11516] lr: 7.9595e-04 eta: 18:40:31 time: 0.3925 data_time: 0.0011 memory: 9465 grad_norm: 0.0326 loss: 0.6811 +2023/06/01 02:57:33 - mmengine - INFO - Epoch(train) [7][ 2200/11516] lr: 7.9595e-04 eta: 18:39:42 time: 0.3703 data_time: 0.0011 memory: 9465 grad_norm: 0.0241 loss: 0.6864 +2023/06/01 02:58:13 - mmengine - INFO - Epoch(train) [7][ 2300/11516] lr: 7.9595e-04 eta: 18:38:55 time: 0.3916 data_time: 0.0012 memory: 9465 grad_norm: 0.0285 loss: 0.6946 +2023/06/01 02:58:55 - mmengine - INFO - Epoch(train) [7][ 2400/11516] lr: 7.9595e-04 eta: 18:38:11 time: 0.4179 data_time: 0.0012 memory: 9465 grad_norm: 0.0190 loss: 0.6891 +2023/06/01 02:59:36 - mmengine - INFO - Epoch(train) [7][ 2500/11516] lr: 7.9595e-04 eta: 18:37:26 time: 0.4112 data_time: 0.0011 memory: 9465 grad_norm: 0.0310 loss: 0.6839 +2023/06/01 03:00:18 - mmengine - INFO - Epoch(train) [7][ 2600/11516] lr: 7.9595e-04 eta: 18:36:42 time: 0.4152 data_time: 0.0012 memory: 9465 grad_norm: 0.0216 loss: 0.6927 +2023/06/01 03:00:57 - mmengine - INFO - Epoch(train) [7][ 2700/11516] lr: 7.9595e-04 eta: 18:35:54 time: 0.3625 data_time: 0.0012 memory: 9465 grad_norm: 0.0287 loss: 0.6846 +2023/06/01 03:01:37 - mmengine - INFO - Epoch(train) [7][ 2800/11516] lr: 7.9595e-04 eta: 18:35:08 time: 0.3874 data_time: 0.0012 memory: 9465 grad_norm: 0.0214 loss: 0.6866 +2023/06/01 03:02:18 - mmengine - INFO - Epoch(train) [7][ 2900/11516] lr: 7.9595e-04 eta: 18:34:22 time: 0.4071 data_time: 0.0013 memory: 9465 grad_norm: 0.0213 loss: 0.6894 +2023/06/01 03:02:19 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/06/01 03:03:01 - mmengine - INFO - Epoch(train) [7][ 3000/11516] lr: 7.9595e-04 eta: 18:33:40 time: 0.4422 data_time: 0.0012 memory: 9465 grad_norm: 0.0287 loss: 0.6893 +2023/06/01 03:03:41 - mmengine - INFO - Epoch(train) [7][ 3100/11516] lr: 7.9595e-04 eta: 18:32:55 time: 0.3787 data_time: 0.0011 memory: 9465 grad_norm: 0.0214 loss: 0.6883 +2023/06/01 03:04:22 - mmengine - INFO - Epoch(train) [7][ 3200/11516] lr: 7.9595e-04 eta: 18:32:10 time: 0.4094 data_time: 0.0012 memory: 9465 grad_norm: 0.0249 loss: 0.6873 +2023/06/01 03:05:03 - mmengine - INFO - Epoch(train) [7][ 3300/11516] lr: 7.9595e-04 eta: 18:31:23 time: 0.4166 data_time: 0.0011 memory: 9465 grad_norm: 0.0168 loss: 0.6896 +2023/06/01 03:05:43 - mmengine - INFO - Epoch(train) [7][ 3400/11516] lr: 7.9595e-04 eta: 18:30:38 time: 0.3982 data_time: 0.0012 memory: 9465 grad_norm: 0.0167 loss: 0.6863 +2023/06/01 03:06:24 - mmengine - INFO - Epoch(train) [7][ 3500/11516] lr: 7.9595e-04 eta: 18:29:51 time: 0.3673 data_time: 0.0011 memory: 9465 grad_norm: 0.0255 loss: 0.6828 +2023/06/01 03:07:05 - mmengine - INFO - Epoch(train) [7][ 3600/11516] lr: 7.9595e-04 eta: 18:29:07 time: 0.3901 data_time: 0.0011 memory: 9465 grad_norm: 0.0231 loss: 0.6871 +2023/06/01 03:07:45 - mmengine - INFO - Epoch(train) [7][ 3700/11516] lr: 7.9595e-04 eta: 18:28:21 time: 0.3825 data_time: 0.0012 memory: 9465 grad_norm: 0.0193 loss: 0.6848 +2023/06/01 03:08:27 - mmengine - INFO - Epoch(train) [7][ 3800/11516] lr: 7.9595e-04 eta: 18:27:38 time: 0.4376 data_time: 0.0011 memory: 9465 grad_norm: 0.0223 loss: 0.6839 +2023/06/01 03:09:09 - mmengine - INFO - Epoch(train) [7][ 3900/11516] lr: 7.9595e-04 eta: 18:26:54 time: 0.4053 data_time: 0.0011 memory: 9465 grad_norm: 0.0242 loss: 0.6877 +2023/06/01 03:09:09 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/06/01 03:09:50 - mmengine - INFO - Epoch(train) [7][ 4000/11516] lr: 7.9595e-04 eta: 18:26:10 time: 0.3898 data_time: 0.0012 memory: 9465 grad_norm: 0.0248 loss: 0.6862 +2023/06/01 03:10:30 - mmengine - INFO - Epoch(train) [7][ 4100/11516] lr: 7.9595e-04 eta: 18:25:23 time: 0.3895 data_time: 0.0013 memory: 9465 grad_norm: 0.0289 loss: 0.6918 +2023/06/01 03:11:11 - mmengine - INFO - Epoch(train) [7][ 4200/11516] lr: 7.9595e-04 eta: 18:24:38 time: 0.3962 data_time: 0.0013 memory: 9465 grad_norm: 0.0178 loss: 0.6895 +2023/06/01 03:11:51 - mmengine - INFO - Epoch(train) [7][ 4300/11516] lr: 7.9595e-04 eta: 18:23:50 time: 0.3996 data_time: 0.0012 memory: 9465 grad_norm: 0.0178 loss: 0.6870 +2023/06/01 03:12:32 - mmengine - INFO - Epoch(train) [7][ 4400/11516] lr: 7.9595e-04 eta: 18:23:06 time: 0.4225 data_time: 0.0012 memory: 9465 grad_norm: 0.0121 loss: 0.6853 +2023/06/01 03:13:12 - mmengine - INFO - Epoch(train) [7][ 4500/11516] lr: 7.9595e-04 eta: 18:22:20 time: 0.3969 data_time: 0.0011 memory: 9465 grad_norm: 0.0269 loss: 0.6847 +2023/06/01 03:13:52 - mmengine - INFO - Epoch(train) [7][ 4600/11516] lr: 7.9595e-04 eta: 18:21:34 time: 0.3934 data_time: 0.0012 memory: 9465 grad_norm: 0.0374 loss: 0.6894 +2023/06/01 03:14:34 - mmengine - INFO - Epoch(train) [7][ 4700/11516] lr: 7.9595e-04 eta: 18:20:50 time: 0.4324 data_time: 0.0012 memory: 9465 grad_norm: 0.0132 loss: 0.6871 +2023/06/01 03:15:15 - mmengine - INFO - Epoch(train) [7][ 4800/11516] lr: 7.9595e-04 eta: 18:20:05 time: 0.4220 data_time: 0.0011 memory: 9465 grad_norm: 0.0094 loss: 0.6890 +2023/06/01 03:15:56 - mmengine - INFO - Epoch(train) [7][ 4900/11516] lr: 7.9595e-04 eta: 18:19:21 time: 0.3974 data_time: 0.0012 memory: 9465 grad_norm: 0.0209 loss: 0.6817 +2023/06/01 03:15:57 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/06/01 03:16:36 - mmengine - INFO - Epoch(train) [7][ 5000/11516] lr: 7.9595e-04 eta: 18:18:35 time: 0.3994 data_time: 0.0012 memory: 9465 grad_norm: 0.0179 loss: 0.6810 +2023/06/01 03:17:17 - mmengine - INFO - Epoch(train) [7][ 5100/11516] lr: 7.9595e-04 eta: 18:17:49 time: 0.4127 data_time: 0.0012 memory: 9465 grad_norm: 0.0141 loss: 0.6843 +2023/06/01 03:17:57 - mmengine - INFO - Epoch(train) [7][ 5200/11516] lr: 7.9595e-04 eta: 18:17:03 time: 0.4044 data_time: 0.0012 memory: 9465 grad_norm: 0.0247 loss: 0.6854 +2023/06/01 03:18:37 - mmengine - INFO - Epoch(train) [7][ 5300/11516] lr: 7.9595e-04 eta: 18:16:16 time: 0.3998 data_time: 0.0011 memory: 9465 grad_norm: 0.0216 loss: 0.6904 +2023/06/01 03:19:19 - mmengine - INFO - Epoch(train) [7][ 5400/11516] lr: 7.9595e-04 eta: 18:15:32 time: 0.3843 data_time: 0.0013 memory: 9465 grad_norm: 0.0222 loss: 0.6855 +2023/06/01 03:19:59 - mmengine - INFO - Epoch(train) [7][ 5500/11516] lr: 7.9595e-04 eta: 18:14:47 time: 0.4551 data_time: 0.0011 memory: 9465 grad_norm: 0.0316 loss: 0.6926 +2023/06/01 03:20:41 - mmengine - INFO - Epoch(train) [7][ 5600/11516] lr: 7.9595e-04 eta: 18:14:03 time: 0.4486 data_time: 0.0013 memory: 9465 grad_norm: 0.0208 loss: 0.6894 +2023/06/01 03:21:21 - mmengine - INFO - Epoch(train) [7][ 5700/11516] lr: 7.9595e-04 eta: 18:13:17 time: 0.4434 data_time: 0.0011 memory: 9465 grad_norm: 0.0255 loss: 0.6861 +2023/06/01 03:22:02 - mmengine - INFO - Epoch(train) [7][ 5800/11516] lr: 7.9595e-04 eta: 18:12:32 time: 0.4089 data_time: 0.0011 memory: 9465 grad_norm: 0.0274 loss: 0.6913 +2023/06/01 03:22:43 - mmengine - INFO - Epoch(train) [7][ 5900/11516] lr: 7.9595e-04 eta: 18:11:48 time: 0.4130 data_time: 0.0012 memory: 9465 grad_norm: 0.0165 loss: 0.6849 +2023/06/01 03:22:44 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/06/01 03:23:25 - mmengine - INFO - Epoch(train) [7][ 6000/11516] lr: 7.9595e-04 eta: 18:11:04 time: 0.3944 data_time: 0.0013 memory: 9465 grad_norm: 0.0163 loss: 0.6885 +2023/06/01 03:24:05 - mmengine - INFO - Epoch(train) [7][ 6100/11516] lr: 7.9595e-04 eta: 18:10:18 time: 0.3833 data_time: 0.0011 memory: 9465 grad_norm: 0.0184 loss: 0.6828 +2023/06/01 03:24:45 - mmengine - INFO - Epoch(train) [7][ 6200/11516] lr: 7.9595e-04 eta: 18:09:31 time: 0.4102 data_time: 0.0012 memory: 9465 grad_norm: 0.0277 loss: 0.6827 +2023/06/01 03:25:25 - mmengine - INFO - Epoch(train) [7][ 6300/11516] lr: 7.9595e-04 eta: 18:08:45 time: 0.3762 data_time: 0.0012 memory: 9465 grad_norm: 0.0284 loss: 0.6861 +2023/06/01 03:26:05 - mmengine - INFO - Epoch(train) [7][ 6400/11516] lr: 7.9595e-04 eta: 18:07:58 time: 0.3971 data_time: 0.0013 memory: 9465 grad_norm: 0.0263 loss: 0.6839 +2023/06/01 03:26:46 - mmengine - INFO - Epoch(train) [7][ 6500/11516] lr: 7.9595e-04 eta: 18:07:13 time: 0.3830 data_time: 0.0014 memory: 9465 grad_norm: 0.0298 loss: 0.6906 +2023/06/01 03:27:27 - mmengine - INFO - Epoch(train) [7][ 6600/11516] lr: 7.9595e-04 eta: 18:06:28 time: 0.3759 data_time: 0.0012 memory: 9465 grad_norm: 0.0180 loss: 0.6807 +2023/06/01 03:28:08 - mmengine - INFO - Epoch(train) [7][ 6700/11516] lr: 7.9595e-04 eta: 18:05:45 time: 0.4705 data_time: 0.0011 memory: 9465 grad_norm: 0.0249 loss: 0.6850 +2023/06/01 03:28:50 - mmengine - INFO - Epoch(train) [7][ 6800/11516] lr: 7.9595e-04 eta: 18:05:02 time: 0.4366 data_time: 0.0013 memory: 9465 grad_norm: 0.0313 loss: 0.6838 +2023/06/01 03:29:31 - mmengine - INFO - Epoch(train) [7][ 6900/11516] lr: 7.9595e-04 eta: 18:04:18 time: 0.3788 data_time: 0.0012 memory: 9465 grad_norm: 0.0249 loss: 0.6888 +2023/06/01 03:29:32 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/06/01 03:30:12 - mmengine - INFO - Epoch(train) [7][ 7000/11516] lr: 7.9595e-04 eta: 18:03:32 time: 0.3895 data_time: 0.0012 memory: 9465 grad_norm: 0.0186 loss: 0.6865 +2023/06/01 03:30:52 - mmengine - INFO - Epoch(train) [7][ 7100/11516] lr: 7.9595e-04 eta: 18:02:47 time: 0.3791 data_time: 0.0012 memory: 9465 grad_norm: 0.0137 loss: 0.6841 +2023/06/01 03:31:31 - mmengine - INFO - Epoch(train) [7][ 7200/11516] lr: 7.9595e-04 eta: 18:01:58 time: 0.3779 data_time: 0.0012 memory: 9465 grad_norm: 0.0305 loss: 0.6863 +2023/06/01 03:32:13 - mmengine - INFO - Epoch(train) [7][ 7300/11516] lr: 7.9595e-04 eta: 18:01:15 time: 0.4389 data_time: 0.0012 memory: 9465 grad_norm: 0.0240 loss: 0.6926 +2023/06/01 03:32:53 - mmengine - INFO - Epoch(train) [7][ 7400/11516] lr: 7.9595e-04 eta: 18:00:30 time: 0.3803 data_time: 0.0012 memory: 9465 grad_norm: 0.0218 loss: 0.6871 +2023/06/01 03:33:33 - mmengine - INFO - Epoch(train) [7][ 7500/11516] lr: 7.9595e-04 eta: 17:59:43 time: 0.3713 data_time: 0.0012 memory: 9465 grad_norm: 0.0294 loss: 0.6822 +2023/06/01 03:34:13 - mmengine - INFO - Epoch(train) [7][ 7600/11516] lr: 7.9595e-04 eta: 17:58:56 time: 0.4177 data_time: 0.0012 memory: 9465 grad_norm: 0.0185 loss: 0.6826 +2023/06/01 03:34:54 - mmengine - INFO - Epoch(train) [7][ 7700/11516] lr: 7.9595e-04 eta: 17:58:12 time: 0.4107 data_time: 0.0012 memory: 9465 grad_norm: 0.0229 loss: 0.6876 +2023/06/01 03:35:35 - mmengine - INFO - Epoch(train) [7][ 7800/11516] lr: 7.9595e-04 eta: 17:57:28 time: 0.4024 data_time: 0.0012 memory: 9465 grad_norm: 0.0231 loss: 0.6914 +2023/06/01 03:36:17 - mmengine - INFO - Epoch(train) [7][ 7900/11516] lr: 7.9595e-04 eta: 17:56:45 time: 0.4121 data_time: 0.0014 memory: 9465 grad_norm: 0.0318 loss: 0.6898 +2023/06/01 03:36:18 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/06/01 03:36:58 - mmengine - INFO - Epoch(train) [7][ 8000/11516] lr: 7.9595e-04 eta: 17:56:00 time: 0.4420 data_time: 0.0011 memory: 9465 grad_norm: 0.0194 loss: 0.6842 +2023/06/01 03:37:40 - mmengine - INFO - Epoch(train) [7][ 8100/11516] lr: 7.9595e-04 eta: 17:55:17 time: 0.4409 data_time: 0.0012 memory: 9465 grad_norm: 0.0211 loss: 0.6831 +2023/06/01 03:38:20 - mmengine - INFO - Epoch(train) [7][ 8200/11516] lr: 7.9595e-04 eta: 17:54:32 time: 0.3879 data_time: 0.0016 memory: 9465 grad_norm: 0.0228 loss: 0.6855 +2023/06/01 03:39:02 - mmengine - INFO - Epoch(train) [7][ 8300/11516] lr: 7.9595e-04 eta: 17:53:49 time: 0.4229 data_time: 0.0012 memory: 9465 grad_norm: 0.0171 loss: 0.6846 +2023/06/01 03:39:44 - mmengine - INFO - Epoch(train) [7][ 8400/11516] lr: 7.9595e-04 eta: 17:53:06 time: 0.3941 data_time: 0.0012 memory: 9465 grad_norm: 0.0352 loss: 0.6838 +2023/06/01 03:40:24 - mmengine - INFO - Epoch(train) [7][ 8500/11516] lr: 7.9595e-04 eta: 17:52:21 time: 0.4434 data_time: 0.0014 memory: 9465 grad_norm: 0.0238 loss: 0.6801 +2023/06/01 03:41:05 - mmengine - INFO - Epoch(train) [7][ 8600/11516] lr: 7.9595e-04 eta: 17:51:36 time: 0.4075 data_time: 0.0012 memory: 9465 grad_norm: 0.0268 loss: 0.6902 +2023/06/01 03:41:45 - mmengine - INFO - Epoch(train) [7][ 8700/11516] lr: 7.9595e-04 eta: 17:50:49 time: 0.4115 data_time: 0.0013 memory: 9465 grad_norm: 0.0251 loss: 0.6854 +2023/06/01 03:42:26 - mmengine - INFO - Epoch(train) [7][ 8800/11516] lr: 7.9595e-04 eta: 17:50:06 time: 0.4011 data_time: 0.0012 memory: 9465 grad_norm: 0.0304 loss: 0.6884 +2023/06/01 03:43:08 - mmengine - INFO - Epoch(train) [7][ 8900/11516] lr: 7.9595e-04 eta: 17:49:23 time: 0.4028 data_time: 0.0012 memory: 9465 grad_norm: 0.0214 loss: 0.6868 +2023/06/01 03:43:09 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/06/01 03:43:50 - mmengine - INFO - Epoch(train) [7][ 9000/11516] lr: 7.9595e-04 eta: 17:48:40 time: 0.4239 data_time: 0.0013 memory: 9465 grad_norm: 0.0251 loss: 0.6887 +2023/06/01 03:44:31 - mmengine - INFO - Epoch(train) [7][ 9100/11516] lr: 7.9595e-04 eta: 17:47:57 time: 0.4038 data_time: 0.0013 memory: 9465 grad_norm: 0.0206 loss: 0.6861 +2023/06/01 03:45:12 - mmengine - INFO - Epoch(train) [7][ 9200/11516] lr: 7.9595e-04 eta: 17:47:12 time: 0.4030 data_time: 0.0012 memory: 9465 grad_norm: 0.0219 loss: 0.6868 +2023/06/01 03:45:52 - mmengine - INFO - Epoch(train) [7][ 9300/11516] lr: 7.9595e-04 eta: 17:46:26 time: 0.3923 data_time: 0.0012 memory: 9465 grad_norm: 0.0276 loss: 0.6871 +2023/06/01 03:46:32 - mmengine - INFO - Epoch(train) [7][ 9400/11516] lr: 7.9595e-04 eta: 17:45:40 time: 0.3998 data_time: 0.0014 memory: 9465 grad_norm: 0.0243 loss: 0.6896 +2023/06/01 03:47:14 - mmengine - INFO - Epoch(train) [7][ 9500/11516] lr: 7.9595e-04 eta: 17:44:57 time: 0.4036 data_time: 0.0011 memory: 9465 grad_norm: 0.0231 loss: 0.6867 +2023/06/01 03:47:54 - mmengine - INFO - Epoch(train) [7][ 9600/11516] lr: 7.9595e-04 eta: 17:44:12 time: 0.4252 data_time: 0.0013 memory: 9465 grad_norm: 0.0313 loss: 0.6873 +2023/06/01 03:48:34 - mmengine - INFO - Epoch(train) [7][ 9700/11516] lr: 7.9595e-04 eta: 17:43:25 time: 0.3892 data_time: 0.0012 memory: 9465 grad_norm: 0.0285 loss: 0.6913 +2023/06/01 03:49:15 - mmengine - INFO - Epoch(train) [7][ 9800/11516] lr: 7.9595e-04 eta: 17:42:41 time: 0.4075 data_time: 0.0014 memory: 9465 grad_norm: 0.0200 loss: 0.6845 +2023/06/01 03:49:56 - mmengine - INFO - Epoch(train) [7][ 9900/11516] lr: 7.9595e-04 eta: 17:41:56 time: 0.3920 data_time: 0.0013 memory: 9465 grad_norm: 0.0257 loss: 0.6842 +2023/06/01 03:49:57 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/06/01 03:50:38 - mmengine - INFO - Epoch(train) [7][10000/11516] lr: 7.9595e-04 eta: 17:41:14 time: 0.3859 data_time: 0.0013 memory: 9465 grad_norm: 0.0256 loss: 0.6894 +2023/06/01 03:51:20 - mmengine - INFO - Epoch(train) [7][10100/11516] lr: 7.9595e-04 eta: 17:40:31 time: 0.4403 data_time: 0.0012 memory: 9465 grad_norm: 0.0228 loss: 0.6859 +2023/06/01 03:51:59 - mmengine - INFO - Epoch(train) [7][10200/11516] lr: 7.9595e-04 eta: 17:39:44 time: 0.4115 data_time: 0.0012 memory: 9465 grad_norm: 0.0263 loss: 0.6829 +2023/06/01 03:52:41 - mmengine - INFO - Epoch(train) [7][10300/11516] lr: 7.9595e-04 eta: 17:39:01 time: 0.4091 data_time: 0.0015 memory: 9465 grad_norm: 0.0189 loss: 0.6861 +2023/06/01 03:53:22 - mmengine - INFO - Epoch(train) [7][10400/11516] lr: 7.9595e-04 eta: 17:38:17 time: 0.4074 data_time: 0.0012 memory: 9465 grad_norm: 0.0323 loss: 0.6906 +2023/06/01 03:54:02 - mmengine - INFO - Epoch(train) [7][10500/11516] lr: 7.9595e-04 eta: 17:37:31 time: 0.4111 data_time: 0.0014 memory: 9465 grad_norm: 0.0205 loss: 0.6851 +2023/06/01 03:54:43 - mmengine - INFO - Epoch(train) [7][10600/11516] lr: 7.9595e-04 eta: 17:36:46 time: 0.4257 data_time: 0.0012 memory: 9465 grad_norm: 0.0254 loss: 0.6858 +2023/06/01 03:55:24 - mmengine - INFO - Epoch(train) [7][10700/11516] lr: 7.9595e-04 eta: 17:36:03 time: 0.3962 data_time: 0.0012 memory: 9465 grad_norm: 0.0257 loss: 0.6889 +2023/06/01 03:56:04 - mmengine - INFO - Epoch(train) [7][10800/11516] lr: 7.9595e-04 eta: 17:35:17 time: 0.4198 data_time: 0.0012 memory: 9465 grad_norm: 0.0272 loss: 0.6870 +2023/06/01 03:56:45 - mmengine - INFO - Epoch(train) [7][10900/11516] lr: 7.9595e-04 eta: 17:34:32 time: 0.3938 data_time: 0.0012 memory: 9465 grad_norm: 0.0297 loss: 0.6844 +2023/06/01 03:56:45 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/06/01 03:57:25 - mmengine - INFO - Epoch(train) [7][11000/11516] lr: 7.9595e-04 eta: 17:33:47 time: 0.3853 data_time: 0.0014 memory: 9465 grad_norm: 0.0205 loss: 0.6875 +2023/06/01 03:58:07 - mmengine - INFO - Epoch(train) [7][11100/11516] lr: 7.9595e-04 eta: 17:33:03 time: 0.4090 data_time: 0.0012 memory: 9465 grad_norm: 0.0189 loss: 0.6915 +2023/06/01 03:58:48 - mmengine - INFO - Epoch(train) [7][11200/11516] lr: 7.9595e-04 eta: 17:32:19 time: 0.4055 data_time: 0.0013 memory: 9465 grad_norm: 0.0284 loss: 0.6907 +2023/06/01 03:59:27 - mmengine - INFO - Epoch(train) [7][11300/11516] lr: 7.9595e-04 eta: 17:31:33 time: 0.3897 data_time: 0.0012 memory: 9465 grad_norm: 0.0237 loss: 0.6890 +2023/06/01 04:00:06 - mmengine - INFO - Epoch(train) [7][11400/11516] lr: 7.9595e-04 eta: 17:30:45 time: 0.4697 data_time: 0.0018 memory: 9465 grad_norm: 0.0171 loss: 0.6891 +2023/06/01 04:00:48 - mmengine - INFO - Epoch(train) [7][11500/11516] lr: 7.9595e-04 eta: 17:30:02 time: 0.3889 data_time: 0.0072 memory: 9465 grad_norm: 0.0179 loss: 0.6851 +2023/06/01 04:00:53 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/06/01 04:00:53 - mmengine - INFO - Saving checkpoint at 7 epochs +2023/06/01 04:01:09 - mmengine - INFO - Epoch(val) [7][16/16] accuracy/top1: 100.0000 single-label/precision_classwise: [100.0, 0.0] single-label/recall_classwise: [100.0, 0.0] single-label/f1-score_classwise: [100.0, 0.0] data_time: 0.3063 time: 0.5598 +2023/06/01 04:01:51 - mmengine - INFO - Epoch(train) [8][ 100/11516] lr: 7.2973e-04 eta: 17:29:09 time: 0.3870 data_time: 0.0845 memory: 9465 grad_norm: 0.0233 loss: 0.6918 +2023/06/01 04:02:33 - mmengine - INFO - Epoch(train) [8][ 200/11516] lr: 7.2973e-04 eta: 17:28:26 time: 0.3595 data_time: 0.0013 memory: 9465 grad_norm: 0.0290 loss: 0.6855 +2023/06/01 04:03:13 - mmengine - INFO - Epoch(train) [8][ 300/11516] lr: 7.2973e-04 eta: 17:27:41 time: 0.3827 data_time: 0.0012 memory: 9465 grad_norm: 0.0122 loss: 0.6885 +2023/06/01 04:03:47 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/06/01 04:03:53 - mmengine - INFO - Epoch(train) [8][ 400/11516] lr: 7.2973e-04 eta: 17:26:55 time: 0.3924 data_time: 0.0012 memory: 9465 grad_norm: 0.0247 loss: 0.6870 +2023/06/01 04:04:35 - mmengine - INFO - Epoch(train) [8][ 500/11516] lr: 7.2973e-04 eta: 17:26:13 time: 0.3920 data_time: 0.0012 memory: 9465 grad_norm: 0.0210 loss: 0.6873 +2023/06/01 04:05:17 - mmengine - INFO - Epoch(train) [8][ 600/11516] lr: 7.2973e-04 eta: 17:25:30 time: 0.4291 data_time: 0.0015 memory: 9465 grad_norm: 0.0242 loss: 0.6885 +2023/06/01 04:05:58 - mmengine - INFO - Epoch(train) [8][ 700/11516] lr: 7.2973e-04 eta: 17:24:46 time: 0.3991 data_time: 0.0011 memory: 9465 grad_norm: 0.0210 loss: 0.6871 +2023/06/01 04:06:39 - mmengine - INFO - Epoch(train) [8][ 800/11516] lr: 7.2973e-04 eta: 17:24:02 time: 0.4479 data_time: 0.0012 memory: 9465 grad_norm: 0.0214 loss: 0.6907 +2023/06/01 04:07:21 - mmengine - INFO - Epoch(train) [8][ 900/11516] lr: 7.2973e-04 eta: 17:23:20 time: 0.4333 data_time: 0.0012 memory: 9465 grad_norm: 0.0255 loss: 0.6890 +2023/06/01 04:08:03 - mmengine - INFO - Epoch(train) [8][ 1000/11516] lr: 7.2973e-04 eta: 17:22:37 time: 0.4597 data_time: 0.0012 memory: 9465 grad_norm: 0.0284 loss: 0.6854 +2023/06/01 04:08:44 - mmengine - INFO - Epoch(train) [8][ 1100/11516] lr: 7.2973e-04 eta: 17:21:54 time: 0.3773 data_time: 0.0012 memory: 9465 grad_norm: 0.0265 loss: 0.6852 +2023/06/01 04:09:25 - mmengine - INFO - Epoch(train) [8][ 1200/11516] lr: 7.2973e-04 eta: 17:21:10 time: 0.4045 data_time: 0.0012 memory: 9465 grad_norm: 0.0151 loss: 0.6829 +2023/06/01 04:10:06 - mmengine - INFO - Epoch(train) [8][ 1300/11516] lr: 7.2973e-04 eta: 17:20:26 time: 0.4122 data_time: 0.0013 memory: 9465 grad_norm: 0.0201 loss: 0.6826 +2023/06/01 04:10:41 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/06/01 04:10:47 - mmengine - INFO - Epoch(train) [8][ 1400/11516] lr: 7.2973e-04 eta: 17:19:42 time: 0.3867 data_time: 0.0012 memory: 9465 grad_norm: 0.0374 loss: 0.6820 +2023/06/01 04:11:28 - mmengine - INFO - Epoch(train) [8][ 1500/11516] lr: 7.2973e-04 eta: 17:18:59 time: 0.4171 data_time: 0.0012 memory: 9465 grad_norm: 0.0265 loss: 0.6865 +2023/06/01 04:12:09 - mmengine - INFO - Epoch(train) [8][ 1600/11516] lr: 7.2973e-04 eta: 17:18:14 time: 0.3936 data_time: 0.0012 memory: 9465 grad_norm: 0.0207 loss: 0.6857 +2023/06/01 04:12:51 - mmengine - INFO - Epoch(train) [8][ 1700/11516] lr: 7.2973e-04 eta: 17:17:32 time: 0.3955 data_time: 0.0013 memory: 9465 grad_norm: 0.0252 loss: 0.6887 +2023/06/01 04:13:32 - mmengine - INFO - Epoch(train) [8][ 1800/11516] lr: 7.2973e-04 eta: 17:16:47 time: 0.4180 data_time: 0.0013 memory: 9465 grad_norm: 0.0138 loss: 0.6886 +2023/06/01 04:14:13 - mmengine - INFO - Epoch(train) [8][ 1900/11516] lr: 7.2973e-04 eta: 17:16:04 time: 0.4146 data_time: 0.0015 memory: 9465 grad_norm: 0.0321 loss: 0.6853 +2023/06/01 04:14:54 - mmengine - INFO - Epoch(train) [8][ 2000/11516] lr: 7.2973e-04 eta: 17:15:20 time: 0.4193 data_time: 0.0011 memory: 9465 grad_norm: 0.0218 loss: 0.6866 +2023/06/01 04:15:36 - mmengine - INFO - Epoch(train) [8][ 2100/11516] lr: 7.2973e-04 eta: 17:14:37 time: 0.4084 data_time: 0.0011 memory: 9465 grad_norm: 0.0203 loss: 0.6887 +2023/06/01 04:16:17 - mmengine - INFO - Epoch(train) [8][ 2200/11516] lr: 7.2973e-04 eta: 17:13:54 time: 0.3995 data_time: 0.0012 memory: 9465 grad_norm: 0.0288 loss: 0.6823 +2023/06/01 04:16:58 - mmengine - INFO - Epoch(train) [8][ 2300/11516] lr: 7.2973e-04 eta: 17:13:09 time: 0.3795 data_time: 0.0012 memory: 9465 grad_norm: 0.0160 loss: 0.6875 +2023/06/01 04:17:32 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/06/01 04:17:38 - mmengine - INFO - Epoch(train) [8][ 2400/11516] lr: 7.2973e-04 eta: 17:12:24 time: 0.3933 data_time: 0.0011 memory: 9465 grad_norm: 0.0227 loss: 0.6855 +2023/06/01 04:18:18 - mmengine - INFO - Epoch(train) [8][ 2500/11516] lr: 7.2973e-04 eta: 17:11:39 time: 0.3599 data_time: 0.0013 memory: 9465 grad_norm: 0.0309 loss: 0.6859 +2023/06/01 04:19:00 - mmengine - INFO - Epoch(train) [8][ 2600/11516] lr: 7.2973e-04 eta: 17:10:56 time: 0.3889 data_time: 0.0011 memory: 9465 grad_norm: 0.0210 loss: 0.6811 +2023/06/01 04:19:41 - mmengine - INFO - Epoch(train) [8][ 2700/11516] lr: 7.2973e-04 eta: 17:10:11 time: 0.4009 data_time: 0.0011 memory: 9465 grad_norm: 0.0171 loss: 0.6818 +2023/06/01 04:20:21 - mmengine - INFO - Epoch(train) [8][ 2800/11516] lr: 7.2973e-04 eta: 17:09:26 time: 0.4141 data_time: 0.0013 memory: 9465 grad_norm: 0.0236 loss: 0.6817 +2023/06/01 04:21:03 - mmengine - INFO - Epoch(train) [8][ 2900/11516] lr: 7.2973e-04 eta: 17:08:44 time: 0.4140 data_time: 0.0010 memory: 9465 grad_norm: 0.0208 loss: 0.6875 +2023/06/01 04:21:43 - mmengine - INFO - Epoch(train) [8][ 3000/11516] lr: 7.2973e-04 eta: 17:07:59 time: 0.4012 data_time: 0.0012 memory: 9465 grad_norm: 0.0233 loss: 0.6937 +2023/06/01 04:22:23 - mmengine - INFO - Epoch(train) [8][ 3100/11516] lr: 7.2973e-04 eta: 17:07:13 time: 0.3826 data_time: 0.0012 memory: 9465 grad_norm: 0.0345 loss: 0.6824 +2023/06/01 04:23:05 - mmengine - INFO - Epoch(train) [8][ 3200/11516] lr: 7.2973e-04 eta: 17:06:30 time: 0.4235 data_time: 0.0012 memory: 9465 grad_norm: 0.0266 loss: 0.6879 +2023/06/01 04:23:46 - mmengine - INFO - Epoch(train) [8][ 3300/11516] lr: 7.2973e-04 eta: 17:05:46 time: 0.4350 data_time: 0.0013 memory: 9465 grad_norm: 0.0231 loss: 0.6865 +2023/06/01 04:24:21 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/06/01 04:24:28 - mmengine - INFO - Epoch(train) [8][ 3400/11516] lr: 7.2973e-04 eta: 17:05:04 time: 0.4128 data_time: 0.0012 memory: 9465 grad_norm: 0.0248 loss: 0.6884 +2023/06/01 04:25:07 - mmengine - INFO - Epoch(train) [8][ 3500/11516] lr: 7.2973e-04 eta: 17:04:18 time: 0.4116 data_time: 0.0013 memory: 9465 grad_norm: 0.0226 loss: 0.6872 +2023/06/01 04:25:48 - mmengine - INFO - Epoch(train) [8][ 3600/11516] lr: 7.2973e-04 eta: 17:03:33 time: 0.4063 data_time: 0.0014 memory: 9465 grad_norm: 0.0209 loss: 0.6860 +2023/06/01 04:26:29 - mmengine - INFO - Epoch(train) [8][ 3700/11516] lr: 7.2973e-04 eta: 17:02:49 time: 0.3881 data_time: 0.0011 memory: 9465 grad_norm: 0.0261 loss: 0.6917 +2023/06/01 04:27:10 - mmengine - INFO - Epoch(train) [8][ 3800/11516] lr: 7.2973e-04 eta: 17:02:06 time: 0.4128 data_time: 0.0011 memory: 9465 grad_norm: 0.0155 loss: 0.6921 +2023/06/01 04:27:51 - mmengine - INFO - Epoch(train) [8][ 3900/11516] lr: 7.2973e-04 eta: 17:01:22 time: 0.4275 data_time: 0.0011 memory: 9465 grad_norm: 0.0135 loss: 0.6865 +2023/06/01 04:28:31 - mmengine - INFO - Epoch(train) [8][ 4000/11516] lr: 7.2973e-04 eta: 17:00:37 time: 0.3853 data_time: 0.0011 memory: 9465 grad_norm: 0.0270 loss: 0.6890 +2023/06/01 04:29:16 - mmengine - INFO - Epoch(train) [8][ 4100/11516] lr: 7.2973e-04 eta: 17:00:00 time: 0.6343 data_time: 0.4488 memory: 9465 grad_norm: 0.0182 loss: 0.6857 +2023/06/01 04:29:54 - mmengine - INFO - Epoch(train) [8][ 4200/11516] lr: 7.2973e-04 eta: 16:59:11 time: 0.3698 data_time: 0.1860 memory: 9465 grad_norm: 0.0235 loss: 0.6857 +2023/06/01 04:30:35 - mmengine - INFO - Epoch(train) [8][ 4300/11516] lr: 7.2973e-04 eta: 16:58:26 time: 0.4290 data_time: 0.1949 memory: 9465 grad_norm: 0.0301 loss: 0.6878 +2023/06/01 04:31:09 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/06/01 04:31:16 - mmengine - INFO - Epoch(train) [8][ 4400/11516] lr: 7.2973e-04 eta: 16:57:43 time: 0.4285 data_time: 0.0013 memory: 9465 grad_norm: 0.0161 loss: 0.6813 +2023/06/01 04:31:56 - mmengine - INFO - Epoch(train) [8][ 4500/11516] lr: 7.2973e-04 eta: 16:56:57 time: 0.4063 data_time: 0.0014 memory: 9465 grad_norm: 0.0197 loss: 0.6805 +2023/06/01 04:32:37 - mmengine - INFO - Epoch(train) [8][ 4600/11516] lr: 7.2973e-04 eta: 16:56:13 time: 0.4608 data_time: 0.0012 memory: 9465 grad_norm: 0.0198 loss: 0.6873 +2023/06/01 04:33:18 - mmengine - INFO - Epoch(train) [8][ 4700/11516] lr: 7.2973e-04 eta: 16:55:29 time: 0.4043 data_time: 0.0013 memory: 9465 grad_norm: 0.0357 loss: 0.6897 +2023/06/01 04:34:00 - mmengine - INFO - Epoch(train) [8][ 4800/11516] lr: 7.2973e-04 eta: 16:54:47 time: 0.4417 data_time: 0.0012 memory: 9465 grad_norm: 0.0232 loss: 0.6886 +2023/06/01 04:34:40 - mmengine - INFO - Epoch(train) [8][ 4900/11516] lr: 7.2973e-04 eta: 16:54:03 time: 0.3721 data_time: 0.0014 memory: 9465 grad_norm: 0.0222 loss: 0.6852 +2023/06/01 04:35:21 - mmengine - INFO - Epoch(train) [8][ 5000/11516] lr: 7.2973e-04 eta: 16:53:18 time: 0.4131 data_time: 0.0013 memory: 9465 grad_norm: 0.0182 loss: 0.6862 +2023/06/01 04:36:02 - mmengine - INFO - Epoch(train) [8][ 5100/11516] lr: 7.2973e-04 eta: 16:52:34 time: 0.4241 data_time: 0.0014 memory: 9465 grad_norm: 0.0283 loss: 0.6855 +2023/06/01 04:36:42 - mmengine - INFO - Epoch(train) [8][ 5200/11516] lr: 7.2973e-04 eta: 16:51:49 time: 0.3932 data_time: 0.0013 memory: 9465 grad_norm: 0.0166 loss: 0.6876 +2023/06/01 04:37:23 - mmengine - INFO - Epoch(train) [8][ 5300/11516] lr: 7.2973e-04 eta: 16:51:05 time: 0.4089 data_time: 0.0011 memory: 9465 grad_norm: 0.0142 loss: 0.6865 +2023/06/01 04:37:56 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/06/01 04:38:03 - mmengine - INFO - Epoch(train) [8][ 5400/11516] lr: 7.2973e-04 eta: 16:50:20 time: 0.3914 data_time: 0.0012 memory: 9465 grad_norm: 0.0099 loss: 0.6822 +2023/06/01 04:38:44 - mmengine - INFO - Epoch(train) [8][ 5500/11516] lr: 7.2973e-04 eta: 16:49:37 time: 0.4271 data_time: 0.0013 memory: 9465 grad_norm: 0.0268 loss: 0.6847 +2023/06/01 04:39:25 - mmengine - INFO - Epoch(train) [8][ 5600/11516] lr: 7.2973e-04 eta: 16:48:53 time: 0.3956 data_time: 0.0013 memory: 9465 grad_norm: 0.0181 loss: 0.6871 +2023/06/01 04:40:07 - mmengine - INFO - Epoch(train) [8][ 5700/11516] lr: 7.2973e-04 eta: 16:48:10 time: 0.4086 data_time: 0.0012 memory: 9465 grad_norm: 0.0166 loss: 0.6895 +2023/06/01 04:40:47 - mmengine - INFO - Epoch(train) [8][ 5800/11516] lr: 7.2973e-04 eta: 16:47:26 time: 0.4141 data_time: 0.0012 memory: 9465 grad_norm: 0.0244 loss: 0.6828 +2023/06/01 04:41:29 - mmengine - INFO - Epoch(train) [8][ 5900/11516] lr: 7.2973e-04 eta: 16:46:43 time: 0.4052 data_time: 0.0012 memory: 9465 grad_norm: 0.0396 loss: 0.6897 +2023/06/01 04:42:11 - mmengine - INFO - Epoch(train) [8][ 6000/11516] lr: 7.2973e-04 eta: 16:46:01 time: 0.4100 data_time: 0.0011 memory: 9465 grad_norm: 0.0226 loss: 0.6872 +2023/06/01 04:42:51 - mmengine - INFO - Epoch(train) [8][ 6100/11516] lr: 7.2973e-04 eta: 16:45:16 time: 0.3995 data_time: 0.0012 memory: 9465 grad_norm: 0.0158 loss: 0.6898 +2023/06/01 04:43:32 - mmengine - INFO - Epoch(train) [8][ 6200/11516] lr: 7.2973e-04 eta: 16:44:33 time: 0.3922 data_time: 0.0012 memory: 9465 grad_norm: 0.0297 loss: 0.6860 +2023/06/01 04:44:13 - mmengine - INFO - Epoch(train) [8][ 6300/11516] lr: 7.2973e-04 eta: 16:43:48 time: 0.4284 data_time: 0.0013 memory: 9465 grad_norm: 0.0186 loss: 0.6941 +2023/06/01 04:44:48 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/06/01 04:44:54 - mmengine - INFO - Epoch(train) [8][ 6400/11516] lr: 7.2973e-04 eta: 16:43:05 time: 0.4027 data_time: 0.0012 memory: 9465 grad_norm: 0.0223 loss: 0.6856 +2023/06/01 04:45:38 - mmengine - INFO - Epoch(train) [8][ 6500/11516] lr: 7.2973e-04 eta: 16:42:26 time: 0.4627 data_time: 0.0011 memory: 9465 grad_norm: 0.0217 loss: 0.6878 +2023/06/01 04:46:19 - mmengine - INFO - Epoch(train) [8][ 6600/11516] lr: 7.2973e-04 eta: 16:41:43 time: 0.4116 data_time: 0.0012 memory: 9465 grad_norm: 0.0143 loss: 0.6901 +2023/06/01 04:46:59 - mmengine - INFO - Epoch(train) [8][ 6700/11516] lr: 7.2973e-04 eta: 16:40:58 time: 0.4256 data_time: 0.0014 memory: 9465 grad_norm: 0.0180 loss: 0.6926 +2023/06/01 04:47:40 - mmengine - INFO - Epoch(train) [8][ 6800/11516] lr: 7.2973e-04 eta: 16:40:13 time: 0.4029 data_time: 0.0017 memory: 9465 grad_norm: 0.0198 loss: 0.6851 +2023/06/01 04:48:20 - mmengine - INFO - Epoch(train) [8][ 6900/11516] lr: 7.2973e-04 eta: 16:39:29 time: 0.3996 data_time: 0.0017 memory: 9465 grad_norm: 0.0232 loss: 0.6840 +2023/06/01 04:49:01 - mmengine - INFO - Epoch(train) [8][ 7000/11516] lr: 7.2973e-04 eta: 16:38:46 time: 0.4121 data_time: 0.0012 memory: 9465 grad_norm: 0.0282 loss: 0.6847 +2023/06/01 04:49:43 - mmengine - INFO - Epoch(train) [8][ 7100/11516] lr: 7.2973e-04 eta: 16:38:02 time: 0.4120 data_time: 0.0019 memory: 9465 grad_norm: 0.0278 loss: 0.6824 +2023/06/01 04:50:23 - mmengine - INFO - Epoch(train) [8][ 7200/11516] lr: 7.2973e-04 eta: 16:37:17 time: 0.4199 data_time: 0.0013 memory: 9465 grad_norm: 0.0189 loss: 0.6841 +2023/06/01 04:51:03 - mmengine - INFO - Epoch(train) [8][ 7300/11516] lr: 7.2973e-04 eta: 16:36:32 time: 0.4103 data_time: 0.0013 memory: 9465 grad_norm: 0.0200 loss: 0.6858 +2023/06/01 04:51:38 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/06/01 04:51:45 - mmengine - INFO - Epoch(train) [8][ 7400/11516] lr: 7.2973e-04 eta: 16:35:50 time: 0.3752 data_time: 0.0011 memory: 9465 grad_norm: 0.0152 loss: 0.6901 +2023/06/01 04:52:25 - mmengine - INFO - Epoch(train) [8][ 7500/11516] lr: 7.2973e-04 eta: 16:35:05 time: 0.3809 data_time: 0.0012 memory: 9465 grad_norm: 0.0313 loss: 0.6882 +2023/06/01 04:53:06 - mmengine - INFO - Epoch(train) [8][ 7600/11516] lr: 7.2973e-04 eta: 16:34:21 time: 0.3982 data_time: 0.0011 memory: 9465 grad_norm: 0.0177 loss: 0.6841 +2023/06/01 04:53:48 - mmengine - INFO - Epoch(train) [8][ 7700/11516] lr: 7.2973e-04 eta: 16:33:40 time: 0.4844 data_time: 0.0013 memory: 9465 grad_norm: 0.0260 loss: 0.6875 +2023/06/01 04:54:28 - mmengine - INFO - Epoch(train) [8][ 7800/11516] lr: 7.2973e-04 eta: 16:32:54 time: 0.4259 data_time: 0.0012 memory: 9465 grad_norm: 0.0208 loss: 0.6871 +2023/06/01 04:55:09 - mmengine - INFO - Epoch(train) [8][ 7900/11516] lr: 7.2973e-04 eta: 16:32:11 time: 0.4015 data_time: 0.0012 memory: 9465 grad_norm: 0.0180 loss: 0.6881 +2023/06/01 04:55:51 - mmengine - INFO - Epoch(train) [8][ 8000/11516] lr: 7.2973e-04 eta: 16:31:28 time: 0.3879 data_time: 0.0014 memory: 9465 grad_norm: 0.0217 loss: 0.6839 +2023/06/01 04:56:31 - mmengine - INFO - Epoch(train) [8][ 8100/11516] lr: 7.2973e-04 eta: 16:30:44 time: 0.4129 data_time: 0.0012 memory: 9465 grad_norm: 0.0225 loss: 0.6892 +2023/06/01 04:57:11 - mmengine - INFO - Epoch(train) [8][ 8200/11516] lr: 7.2973e-04 eta: 16:29:59 time: 0.4014 data_time: 0.0012 memory: 9465 grad_norm: 0.0312 loss: 0.6895 +2023/06/01 04:57:51 - mmengine - INFO - Epoch(train) [8][ 8300/11516] lr: 7.2973e-04 eta: 16:29:14 time: 0.3792 data_time: 0.0011 memory: 9465 grad_norm: 0.0183 loss: 0.6843 +2023/06/01 04:58:25 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/06/01 04:58:32 - mmengine - INFO - Epoch(train) [8][ 8400/11516] lr: 7.2973e-04 eta: 16:28:30 time: 0.4048 data_time: 0.0012 memory: 9465 grad_norm: 0.0159 loss: 0.6828 +2023/06/01 04:59:13 - mmengine - INFO - Epoch(train) [8][ 8500/11516] lr: 7.2973e-04 eta: 16:27:47 time: 0.4081 data_time: 0.0012 memory: 9465 grad_norm: 0.0180 loss: 0.6859 +2023/06/01 04:59:55 - mmengine - INFO - Epoch(train) [8][ 8600/11516] lr: 7.2973e-04 eta: 16:27:05 time: 0.4496 data_time: 0.0012 memory: 9465 grad_norm: 0.0305 loss: 0.6924 +2023/06/01 05:00:36 - mmengine - INFO - Epoch(train) [8][ 8700/11516] lr: 7.2973e-04 eta: 16:26:22 time: 0.4264 data_time: 0.0013 memory: 9465 grad_norm: 0.0161 loss: 0.6884 +2023/06/01 05:01:16 - mmengine - INFO - Epoch(train) [8][ 8800/11516] lr: 7.2973e-04 eta: 16:25:36 time: 0.3953 data_time: 0.0012 memory: 9465 grad_norm: 0.0277 loss: 0.6868 +2023/06/01 05:01:56 - mmengine - INFO - Epoch(train) [8][ 8900/11516] lr: 7.2973e-04 eta: 16:24:51 time: 0.4063 data_time: 0.0012 memory: 9465 grad_norm: 0.0267 loss: 0.6879 +2023/06/01 05:02:37 - mmengine - INFO - Epoch(train) [8][ 9000/11516] lr: 7.2973e-04 eta: 16:24:08 time: 0.4143 data_time: 0.0011 memory: 9465 grad_norm: 0.0254 loss: 0.6886 +2023/06/01 05:03:18 - mmengine - INFO - Epoch(train) [8][ 9100/11516] lr: 7.2973e-04 eta: 16:23:24 time: 0.4152 data_time: 0.0012 memory: 9465 grad_norm: 0.0316 loss: 0.6853 +2023/06/01 05:03:58 - mmengine - INFO - Epoch(train) [8][ 9200/11516] lr: 7.2973e-04 eta: 16:22:39 time: 0.4193 data_time: 0.0012 memory: 9465 grad_norm: 0.0195 loss: 0.6892 +2023/06/01 05:04:38 - mmengine - INFO - Epoch(train) [8][ 9300/11516] lr: 7.2973e-04 eta: 16:21:54 time: 0.3981 data_time: 0.0012 memory: 9465 grad_norm: 0.0303 loss: 0.6798 +2023/06/01 05:05:12 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/06/01 05:05:19 - mmengine - INFO - Epoch(train) [8][ 9400/11516] lr: 7.2973e-04 eta: 16:21:10 time: 0.3964 data_time: 0.0013 memory: 9465 grad_norm: 0.0346 loss: 0.6841 +2023/06/01 05:06:00 - mmengine - INFO - Epoch(train) [8][ 9500/11516] lr: 7.2973e-04 eta: 16:20:26 time: 0.4103 data_time: 0.0012 memory: 9465 grad_norm: 0.0173 loss: 0.6854 +2023/06/01 05:06:40 - mmengine - INFO - Epoch(train) [8][ 9600/11516] lr: 7.2973e-04 eta: 16:19:42 time: 0.4005 data_time: 0.0011 memory: 9465 grad_norm: 0.0176 loss: 0.6928 +2023/06/01 05:07:21 - mmengine - INFO - Epoch(train) [8][ 9700/11516] lr: 7.2973e-04 eta: 16:18:58 time: 0.3981 data_time: 0.0011 memory: 9465 grad_norm: 0.0203 loss: 0.6875 +2023/06/01 05:08:01 - mmengine - INFO - Epoch(train) [8][ 9800/11516] lr: 7.2973e-04 eta: 16:18:14 time: 0.3789 data_time: 0.0011 memory: 9465 grad_norm: 0.0169 loss: 0.6832 +2023/06/01 05:08:41 - mmengine - INFO - Epoch(train) [8][ 9900/11516] lr: 7.2973e-04 eta: 16:17:29 time: 0.4212 data_time: 0.0012 memory: 9465 grad_norm: 0.0162 loss: 0.6864 +2023/06/01 05:09:21 - mmengine - INFO - Epoch(train) [8][10000/11516] lr: 7.2973e-04 eta: 16:16:44 time: 0.3782 data_time: 0.0012 memory: 9465 grad_norm: 0.0235 loss: 0.6911 +2023/06/01 05:10:02 - mmengine - INFO - Epoch(train) [8][10100/11516] lr: 7.2973e-04 eta: 16:16:00 time: 0.4124 data_time: 0.0011 memory: 9465 grad_norm: 0.0285 loss: 0.6844 +2023/06/01 05:10:43 - mmengine - INFO - Epoch(train) [8][10200/11516] lr: 7.2973e-04 eta: 16:15:16 time: 0.3892 data_time: 0.0012 memory: 9465 grad_norm: 0.0150 loss: 0.6882 +2023/06/01 05:11:24 - mmengine - INFO - Epoch(train) [8][10300/11516] lr: 7.2973e-04 eta: 16:14:33 time: 0.4145 data_time: 0.0013 memory: 9465 grad_norm: 0.0251 loss: 0.6864 +2023/06/01 05:11:59 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/06/01 05:12:05 - mmengine - INFO - Epoch(train) [8][10400/11516] lr: 7.2973e-04 eta: 16:13:50 time: 0.4223 data_time: 0.0014 memory: 9465 grad_norm: 0.0302 loss: 0.6929 +2023/06/01 05:12:46 - mmengine - INFO - Epoch(train) [8][10500/11516] lr: 7.2973e-04 eta: 16:13:06 time: 0.3963 data_time: 0.0015 memory: 9465 grad_norm: 0.0142 loss: 0.6901 +2023/06/01 05:13:26 - mmengine - INFO - Epoch(train) [8][10600/11516] lr: 7.2973e-04 eta: 16:12:21 time: 0.4212 data_time: 0.0022 memory: 9465 grad_norm: 0.0264 loss: 0.6893 +2023/06/01 05:14:06 - mmengine - INFO - Epoch(train) [8][10700/11516] lr: 7.2973e-04 eta: 16:11:37 time: 0.4123 data_time: 0.0012 memory: 9465 grad_norm: 0.0265 loss: 0.6854 +2023/06/01 05:14:48 - mmengine - INFO - Epoch(train) [8][10800/11516] lr: 7.2973e-04 eta: 16:10:54 time: 0.4057 data_time: 0.0011 memory: 9465 grad_norm: 0.0315 loss: 0.6852 +2023/06/01 05:15:29 - mmengine - INFO - Epoch(train) [8][10900/11516] lr: 7.2973e-04 eta: 16:10:11 time: 0.4032 data_time: 0.0012 memory: 9465 grad_norm: 0.0286 loss: 0.6866 +2023/06/01 05:16:10 - mmengine - INFO - Epoch(train) [8][11000/11516] lr: 7.2973e-04 eta: 16:09:28 time: 0.4312 data_time: 0.0011 memory: 9465 grad_norm: 0.0253 loss: 0.6860 +2023/06/01 05:16:51 - mmengine - INFO - Epoch(train) [8][11100/11516] lr: 7.2973e-04 eta: 16:08:44 time: 0.4220 data_time: 0.0013 memory: 9465 grad_norm: 0.0341 loss: 0.6895 +2023/06/01 05:17:30 - mmengine - INFO - Epoch(train) [8][11200/11516] lr: 7.2973e-04 eta: 16:07:59 time: 0.3939 data_time: 0.0013 memory: 9465 grad_norm: 0.0298 loss: 0.6916 +2023/06/01 05:18:10 - mmengine - INFO - Epoch(train) [8][11300/11516] lr: 7.2973e-04 eta: 16:07:13 time: 0.3675 data_time: 0.0012 memory: 9465 grad_norm: 0.0297 loss: 0.6858 +2023/06/01 05:18:43 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/06/01 05:18:50 - mmengine - INFO - Epoch(train) [8][11400/11516] lr: 7.2973e-04 eta: 16:06:29 time: 0.4439 data_time: 0.0012 memory: 9465 grad_norm: 0.0240 loss: 0.6879 +2023/06/01 05:19:31 - mmengine - INFO - Epoch(train) [8][11500/11516] lr: 7.2973e-04 eta: 16:05:45 time: 0.4058 data_time: 0.0073 memory: 9465 grad_norm: 0.0166 loss: 0.6868 +2023/06/01 05:19:36 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/06/01 05:19:36 - mmengine - INFO - Saving checkpoint at 8 epochs +2023/06/01 05:19:53 - mmengine - INFO - Epoch(val) [8][16/16] accuracy/top1: 100.0000 single-label/precision_classwise: [100.0, 0.0] single-label/recall_classwise: [100.0, 0.0] single-label/f1-score_classwise: [100.0, 0.0] data_time: 0.2821 time: 0.5355 +2023/06/01 05:20:36 - mmengine - INFO - Epoch(train) [9][ 100/11516] lr: 6.5796e-04 eta: 16:04:55 time: 0.3961 data_time: 0.2132 memory: 9465 grad_norm: 0.0234 loss: 0.6916 +2023/06/01 05:21:15 - mmengine - INFO - Epoch(train) [9][ 200/11516] lr: 6.5796e-04 eta: 16:04:09 time: 0.4303 data_time: 0.0476 memory: 9465 grad_norm: 0.0153 loss: 0.6891 +2023/06/01 05:21:55 - mmengine - INFO - Epoch(train) [9][ 300/11516] lr: 6.5796e-04 eta: 16:03:25 time: 0.4091 data_time: 0.0012 memory: 9465 grad_norm: 0.0108 loss: 0.6848 +2023/06/01 05:22:37 - mmengine - INFO - Epoch(train) [9][ 400/11516] lr: 6.5796e-04 eta: 16:02:42 time: 0.4342 data_time: 0.0012 memory: 9465 grad_norm: 0.0240 loss: 0.6876 +2023/06/01 05:23:18 - mmengine - INFO - Epoch(train) [9][ 500/11516] lr: 6.5796e-04 eta: 16:01:59 time: 0.4946 data_time: 0.0011 memory: 9465 grad_norm: 0.0241 loss: 0.6896 +2023/06/01 05:23:59 - mmengine - INFO - Epoch(train) [9][ 600/11516] lr: 6.5796e-04 eta: 16:01:17 time: 0.4017 data_time: 0.0015 memory: 9465 grad_norm: 0.0189 loss: 0.6846 +2023/06/01 05:24:40 - mmengine - INFO - Epoch(train) [9][ 700/11516] lr: 6.5796e-04 eta: 16:00:33 time: 0.4144 data_time: 0.0014 memory: 9465 grad_norm: 0.0228 loss: 0.6879 +2023/06/01 05:25:20 - mmengine - INFO - Epoch(train) [9][ 800/11516] lr: 6.5796e-04 eta: 15:59:48 time: 0.4170 data_time: 0.0012 memory: 9465 grad_norm: 0.0238 loss: 0.6876 +2023/06/01 05:25:49 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/06/01 05:26:00 - mmengine - INFO - Epoch(train) [9][ 900/11516] lr: 6.5796e-04 eta: 15:59:04 time: 0.4090 data_time: 0.0013 memory: 9465 grad_norm: 0.0164 loss: 0.6862 +2023/06/01 05:26:41 - mmengine - INFO - Epoch(train) [9][ 1000/11516] lr: 6.5796e-04 eta: 15:58:20 time: 0.4129 data_time: 0.0013 memory: 9465 grad_norm: 0.0160 loss: 0.6853 +2023/06/01 05:27:21 - mmengine - INFO - Epoch(train) [9][ 1100/11516] lr: 6.5796e-04 eta: 15:57:36 time: 0.4026 data_time: 0.0012 memory: 9465 grad_norm: 0.0216 loss: 0.6862 +2023/06/01 05:28:02 - mmengine - INFO - Epoch(train) [9][ 1200/11516] lr: 6.5796e-04 eta: 15:56:51 time: 0.3990 data_time: 0.0012 memory: 9465 grad_norm: 0.0301 loss: 0.6870 +2023/06/01 05:28:41 - mmengine - INFO - Epoch(train) [9][ 1300/11516] lr: 6.5796e-04 eta: 15:56:06 time: 0.3883 data_time: 0.0012 memory: 9465 grad_norm: 0.0190 loss: 0.6873 +2023/06/01 05:29:22 - mmengine - INFO - Epoch(train) [9][ 1400/11516] lr: 6.5796e-04 eta: 15:55:23 time: 0.4238 data_time: 0.0011 memory: 9465 grad_norm: 0.0219 loss: 0.6882 +2023/06/01 05:30:02 - mmengine - INFO - Epoch(train) [9][ 1500/11516] lr: 6.5796e-04 eta: 15:54:37 time: 0.3880 data_time: 0.0012 memory: 9465 grad_norm: 0.0223 loss: 0.6851 +2023/06/01 05:30:42 - mmengine - INFO - Epoch(train) [9][ 1600/11516] lr: 6.5796e-04 eta: 15:53:53 time: 0.4056 data_time: 0.0010 memory: 9465 grad_norm: 0.0204 loss: 0.6865 +2023/06/01 05:31:23 - mmengine - INFO - Epoch(train) [9][ 1700/11516] lr: 6.5796e-04 eta: 15:53:09 time: 0.4523 data_time: 0.0013 memory: 9465 grad_norm: 0.0263 loss: 0.6856 +2023/06/01 05:32:04 - mmengine - INFO - Epoch(train) [9][ 1800/11516] lr: 6.5796e-04 eta: 15:52:26 time: 0.4379 data_time: 0.0016 memory: 9465 grad_norm: 0.0202 loss: 0.6885 +2023/06/01 05:32:32 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/06/01 05:32:43 - mmengine - INFO - Epoch(train) [9][ 1900/11516] lr: 6.5796e-04 eta: 15:51:41 time: 0.3716 data_time: 0.0013 memory: 9465 grad_norm: 0.0217 loss: 0.6896 +2023/06/01 05:33:25 - mmengine - INFO - Epoch(train) [9][ 2000/11516] lr: 6.5796e-04 eta: 15:50:59 time: 0.3944 data_time: 0.0012 memory: 9465 grad_norm: 0.0220 loss: 0.6843 +2023/06/01 05:34:05 - mmengine - INFO - Epoch(train) [9][ 2100/11516] lr: 6.5796e-04 eta: 15:50:14 time: 0.3850 data_time: 0.0012 memory: 9465 grad_norm: 0.0223 loss: 0.6893 +2023/06/01 05:34:46 - mmengine - INFO - Epoch(train) [9][ 2200/11516] lr: 6.5796e-04 eta: 15:49:30 time: 0.4002 data_time: 0.0013 memory: 9465 grad_norm: 0.0202 loss: 0.6863 +2023/06/01 05:35:28 - mmengine - INFO - Epoch(train) [9][ 2300/11516] lr: 6.5796e-04 eta: 15:48:49 time: 0.4249 data_time: 0.0014 memory: 9465 grad_norm: 0.0282 loss: 0.6930 +2023/06/01 05:36:08 - mmengine - INFO - Epoch(train) [9][ 2400/11516] lr: 6.5796e-04 eta: 15:48:04 time: 0.3842 data_time: 0.0013 memory: 9465 grad_norm: 0.0195 loss: 0.6851 +2023/06/01 05:36:49 - mmengine - INFO - Epoch(train) [9][ 2500/11516] lr: 6.5796e-04 eta: 15:47:21 time: 0.4317 data_time: 0.0013 memory: 9465 grad_norm: 0.0087 loss: 0.6844 +2023/06/01 05:37:29 - mmengine - INFO - Epoch(train) [9][ 2600/11516] lr: 6.5796e-04 eta: 15:46:36 time: 0.4316 data_time: 0.0013 memory: 9465 grad_norm: 0.0207 loss: 0.6878 +2023/06/01 05:38:09 - mmengine - INFO - Epoch(train) [9][ 2700/11516] lr: 6.5796e-04 eta: 15:45:53 time: 0.3958 data_time: 0.0013 memory: 9465 grad_norm: 0.0166 loss: 0.6863 +2023/06/01 05:38:50 - mmengine - INFO - Epoch(train) [9][ 2800/11516] lr: 6.5796e-04 eta: 15:45:09 time: 0.4397 data_time: 0.0013 memory: 9465 grad_norm: 0.0333 loss: 0.6900 +2023/06/01 05:39:19 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/06/01 05:39:31 - mmengine - INFO - Epoch(train) [9][ 2900/11516] lr: 6.5796e-04 eta: 15:44:25 time: 0.4259 data_time: 0.0013 memory: 9465 grad_norm: 0.0306 loss: 0.6875 +2023/06/01 05:40:10 - mmengine - INFO - Epoch(train) [9][ 3000/11516] lr: 6.5796e-04 eta: 15:43:40 time: 0.4169 data_time: 0.0013 memory: 9465 grad_norm: 0.0234 loss: 0.6892 +2023/06/01 05:40:52 - mmengine - INFO - Epoch(train) [9][ 3100/11516] lr: 6.5796e-04 eta: 15:42:58 time: 0.4266 data_time: 0.0011 memory: 9465 grad_norm: 0.0167 loss: 0.6889 +2023/06/01 05:41:33 - mmengine - INFO - Epoch(train) [9][ 3200/11516] lr: 6.5796e-04 eta: 15:42:14 time: 0.4270 data_time: 0.0013 memory: 9465 grad_norm: 0.0185 loss: 0.6867 +2023/06/01 05:42:14 - mmengine - INFO - Epoch(train) [9][ 3300/11516] lr: 6.5796e-04 eta: 15:41:32 time: 0.4428 data_time: 0.0012 memory: 9465 grad_norm: 0.0182 loss: 0.6862 +2023/06/01 05:42:55 - mmengine - INFO - Epoch(train) [9][ 3400/11516] lr: 6.5796e-04 eta: 15:40:48 time: 0.4208 data_time: 0.0017 memory: 9465 grad_norm: 0.0300 loss: 0.6792 +2023/06/01 05:43:36 - mmengine - INFO - Epoch(train) [9][ 3500/11516] lr: 6.5796e-04 eta: 15:40:06 time: 0.3938 data_time: 0.0013 memory: 9465 grad_norm: 0.0242 loss: 0.6876 +2023/06/01 05:44:17 - mmengine - INFO - Epoch(train) [9][ 3600/11516] lr: 6.5796e-04 eta: 15:39:23 time: 0.4133 data_time: 0.0012 memory: 9465 grad_norm: 0.0173 loss: 0.6875 +2023/06/01 05:44:57 - mmengine - INFO - Epoch(train) [9][ 3700/11516] lr: 6.5796e-04 eta: 15:38:38 time: 0.3906 data_time: 0.0012 memory: 9465 grad_norm: 0.0187 loss: 0.6856 +2023/06/01 05:45:37 - mmengine - INFO - Epoch(train) [9][ 3800/11516] lr: 6.5796e-04 eta: 15:37:54 time: 0.4010 data_time: 0.0013 memory: 9465 grad_norm: 0.0269 loss: 0.6896 +2023/06/01 05:46:06 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/06/01 05:46:18 - mmengine - INFO - Epoch(train) [9][ 3900/11516] lr: 6.5796e-04 eta: 15:37:10 time: 0.4021 data_time: 0.0012 memory: 9465 grad_norm: 0.0265 loss: 0.6890 +2023/06/01 05:46:59 - mmengine - INFO - Epoch(train) [9][ 4000/11516] lr: 6.5796e-04 eta: 15:36:27 time: 0.4227 data_time: 0.0011 memory: 9465 grad_norm: 0.0174 loss: 0.6824 +2023/06/01 05:47:38 - mmengine - INFO - Epoch(train) [9][ 4100/11516] lr: 6.5796e-04 eta: 15:35:42 time: 0.3781 data_time: 0.0011 memory: 9465 grad_norm: 0.0290 loss: 0.6830 +2023/06/01 05:48:19 - mmengine - INFO - Epoch(train) [9][ 4200/11516] lr: 6.5796e-04 eta: 15:34:58 time: 0.4219 data_time: 0.0012 memory: 9465 grad_norm: 0.0235 loss: 0.6858 +2023/06/01 05:49:00 - mmengine - INFO - Epoch(train) [9][ 4300/11516] lr: 6.5796e-04 eta: 15:34:15 time: 0.4188 data_time: 0.0012 memory: 9465 grad_norm: 0.0167 loss: 0.6858 +2023/06/01 05:49:41 - mmengine - INFO - Epoch(train) [9][ 4400/11516] lr: 6.5796e-04 eta: 15:33:32 time: 0.3722 data_time: 0.0012 memory: 9465 grad_norm: 0.0164 loss: 0.6834 +2023/06/01 05:50:22 - mmengine - INFO - Epoch(train) [9][ 4500/11516] lr: 6.5796e-04 eta: 15:32:49 time: 0.3775 data_time: 0.0014 memory: 9465 grad_norm: 0.0168 loss: 0.6890 +2023/06/01 05:51:02 - mmengine - INFO - Epoch(train) [9][ 4600/11516] lr: 6.5796e-04 eta: 15:32:04 time: 0.4062 data_time: 0.0012 memory: 9465 grad_norm: 0.0279 loss: 0.6881 +2023/06/01 05:51:43 - mmengine - INFO - Epoch(train) [9][ 4700/11516] lr: 6.5796e-04 eta: 15:31:21 time: 0.4085 data_time: 0.0012 memory: 9465 grad_norm: 0.0150 loss: 0.6884 +2023/06/01 05:52:24 - mmengine - INFO - Epoch(train) [9][ 4800/11516] lr: 6.5796e-04 eta: 15:30:38 time: 0.4333 data_time: 0.0013 memory: 9465 grad_norm: 0.0253 loss: 0.6820 +2023/06/01 05:52:53 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/06/01 05:53:04 - mmengine - INFO - Epoch(train) [9][ 4900/11516] lr: 6.5796e-04 eta: 15:29:53 time: 0.3761 data_time: 0.0013 memory: 9465 grad_norm: 0.0193 loss: 0.6868 +2023/06/01 05:53:44 - mmengine - INFO - Epoch(train) [9][ 5000/11516] lr: 6.5796e-04 eta: 15:29:10 time: 0.3789 data_time: 0.0013 memory: 9465 grad_norm: 0.0207 loss: 0.6885 +2023/06/01 05:54:26 - mmengine - INFO - Epoch(train) [9][ 5100/11516] lr: 6.5796e-04 eta: 15:28:27 time: 0.4024 data_time: 0.0013 memory: 9465 grad_norm: 0.0242 loss: 0.6868 +2023/06/01 05:55:07 - mmengine - INFO - Epoch(train) [9][ 5200/11516] lr: 6.5796e-04 eta: 15:27:44 time: 0.4225 data_time: 0.0013 memory: 9465 grad_norm: 0.0199 loss: 0.6854 +2023/06/01 05:55:47 - mmengine - INFO - Epoch(train) [9][ 5300/11516] lr: 6.5796e-04 eta: 15:27:00 time: 0.3833 data_time: 0.0012 memory: 9465 grad_norm: 0.0227 loss: 0.6880 +2023/06/01 05:56:27 - mmengine - INFO - Epoch(train) [9][ 5400/11516] lr: 6.5796e-04 eta: 15:26:16 time: 0.3927 data_time: 0.0013 memory: 9465 grad_norm: 0.0269 loss: 0.6822 +2023/06/01 05:57:09 - mmengine - INFO - Epoch(train) [9][ 5500/11516] lr: 6.5796e-04 eta: 15:25:34 time: 0.4256 data_time: 0.0017 memory: 9465 grad_norm: 0.0272 loss: 0.6898 +2023/06/01 05:57:51 - mmengine - INFO - Epoch(train) [9][ 5600/11516] lr: 6.5796e-04 eta: 15:24:52 time: 0.4974 data_time: 0.0016 memory: 9465 grad_norm: 0.0168 loss: 0.6855 +2023/06/01 05:58:32 - mmengine - INFO - Epoch(train) [9][ 5700/11516] lr: 6.5796e-04 eta: 15:24:10 time: 0.4128 data_time: 0.0013 memory: 9465 grad_norm: 0.0153 loss: 0.6905 +2023/06/01 05:59:12 - mmengine - INFO - Epoch(train) [9][ 5800/11516] lr: 6.5796e-04 eta: 15:23:26 time: 0.4070 data_time: 0.0013 memory: 9465 grad_norm: 0.0141 loss: 0.6879 +2023/06/01 05:59:41 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/06/01 05:59:53 - mmengine - INFO - Epoch(train) [9][ 5900/11516] lr: 6.5796e-04 eta: 15:22:42 time: 0.4430 data_time: 0.0012 memory: 9465 grad_norm: 0.0299 loss: 0.6872 +2023/06/01 06:00:35 - mmengine - INFO - Epoch(train) [9][ 6000/11516] lr: 6.5796e-04 eta: 15:22:00 time: 0.3957 data_time: 0.0014 memory: 9465 grad_norm: 0.0236 loss: 0.6860 +2023/06/01 06:01:15 - mmengine - INFO - Epoch(train) [9][ 6100/11516] lr: 6.5796e-04 eta: 15:21:16 time: 0.3941 data_time: 0.0013 memory: 9465 grad_norm: 0.0187 loss: 0.6859 +2023/06/01 06:01:56 - mmengine - INFO - Epoch(train) [9][ 6200/11516] lr: 6.5796e-04 eta: 15:20:33 time: 0.4224 data_time: 0.0014 memory: 9465 grad_norm: 0.0169 loss: 0.6880 +2023/06/01 06:02:36 - mmengine - INFO - Epoch(train) [9][ 6300/11516] lr: 6.5796e-04 eta: 15:19:49 time: 0.3892 data_time: 0.0013 memory: 9465 grad_norm: 0.0228 loss: 0.6867 +2023/06/01 06:03:17 - mmengine - INFO - Epoch(train) [9][ 6400/11516] lr: 6.5796e-04 eta: 15:19:06 time: 0.3846 data_time: 0.0013 memory: 9465 grad_norm: 0.0259 loss: 0.6896 +2023/06/01 06:03:57 - mmengine - INFO - Epoch(train) [9][ 6500/11516] lr: 6.5796e-04 eta: 15:18:21 time: 0.4137 data_time: 0.0012 memory: 9465 grad_norm: 0.0117 loss: 0.6870 +2023/06/01 06:04:38 - mmengine - INFO - Epoch(train) [9][ 6600/11516] lr: 6.5796e-04 eta: 15:17:38 time: 0.3777 data_time: 0.0012 memory: 9465 grad_norm: 0.0307 loss: 0.6873 +2023/06/01 06:05:19 - mmengine - INFO - Epoch(train) [9][ 6700/11516] lr: 6.5796e-04 eta: 15:16:56 time: 0.4071 data_time: 0.0012 memory: 9465 grad_norm: 0.0127 loss: 0.6883 +2023/06/01 06:06:00 - mmengine - INFO - Epoch(train) [9][ 6800/11516] lr: 6.5796e-04 eta: 15:16:13 time: 0.3950 data_time: 0.0013 memory: 9465 grad_norm: 0.0318 loss: 0.6826 +2023/06/01 06:06:30 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/06/01 06:06:41 - mmengine - INFO - Epoch(train) [9][ 6900/11516] lr: 6.5796e-04 eta: 15:15:29 time: 0.3722 data_time: 0.0012 memory: 9465 grad_norm: 0.0194 loss: 0.6884 +2023/06/01 06:07:21 - mmengine - INFO - Epoch(train) [9][ 7000/11516] lr: 6.5796e-04 eta: 15:14:45 time: 0.3881 data_time: 0.0013 memory: 9465 grad_norm: 0.0276 loss: 0.6807 +2023/06/01 06:08:02 - mmengine - INFO - Epoch(train) [9][ 7100/11516] lr: 6.5796e-04 eta: 15:14:02 time: 0.4049 data_time: 0.0012 memory: 9465 grad_norm: 0.0268 loss: 0.6877 +2023/06/01 06:08:43 - mmengine - INFO - Epoch(train) [9][ 7200/11516] lr: 6.5796e-04 eta: 15:13:19 time: 0.3817 data_time: 0.0012 memory: 9465 grad_norm: 0.0302 loss: 0.6888 +2023/06/01 06:09:24 - mmengine - INFO - Epoch(train) [9][ 7300/11516] lr: 6.5796e-04 eta: 15:12:36 time: 0.4117 data_time: 0.0011 memory: 9465 grad_norm: 0.0280 loss: 0.6888 +2023/06/01 06:10:05 - mmengine - INFO - Epoch(train) [9][ 7400/11516] lr: 6.5796e-04 eta: 15:11:54 time: 0.4159 data_time: 0.0012 memory: 9465 grad_norm: 0.0155 loss: 0.6900 +2023/06/01 06:10:48 - mmengine - INFO - Epoch(train) [9][ 7500/11516] lr: 6.5796e-04 eta: 15:11:14 time: 0.4747 data_time: 0.0012 memory: 9465 grad_norm: 0.0304 loss: 0.6921 +2023/06/01 06:11:30 - mmengine - INFO - Epoch(train) [9][ 7600/11516] lr: 6.5796e-04 eta: 15:10:31 time: 0.4400 data_time: 0.0012 memory: 9465 grad_norm: 0.0298 loss: 0.6888 +2023/06/01 06:12:10 - mmengine - INFO - Epoch(train) [9][ 7700/11516] lr: 6.5796e-04 eta: 15:09:48 time: 0.3785 data_time: 0.0012 memory: 9465 grad_norm: 0.0252 loss: 0.6870 +2023/06/01 06:12:52 - mmengine - INFO - Epoch(train) [9][ 7800/11516] lr: 6.5796e-04 eta: 15:09:05 time: 0.3821 data_time: 0.0013 memory: 9465 grad_norm: 0.0252 loss: 0.6825 +2023/06/01 06:13:21 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/06/01 06:13:33 - mmengine - INFO - Epoch(train) [9][ 7900/11516] lr: 6.5796e-04 eta: 15:08:23 time: 0.4208 data_time: 0.0012 memory: 9465 grad_norm: 0.0237 loss: 0.6865 +2023/06/01 06:14:14 - mmengine - INFO - Epoch(train) [9][ 8000/11516] lr: 6.5796e-04 eta: 15:07:40 time: 0.3837 data_time: 0.0012 memory: 9465 grad_norm: 0.0224 loss: 0.6822 +2023/06/01 06:14:54 - mmengine - INFO - Epoch(train) [9][ 8100/11516] lr: 6.5796e-04 eta: 15:06:56 time: 0.3988 data_time: 0.0013 memory: 9465 grad_norm: 0.0201 loss: 0.6835 +2023/06/01 06:15:35 - mmengine - INFO - Epoch(train) [9][ 8200/11516] lr: 6.5796e-04 eta: 15:06:12 time: 0.4340 data_time: 0.0012 memory: 9465 grad_norm: 0.0261 loss: 0.6859 +2023/06/01 06:16:15 - mmengine - INFO - Epoch(train) [9][ 8300/11516] lr: 6.5796e-04 eta: 15:05:29 time: 0.4059 data_time: 0.0012 memory: 9465 grad_norm: 0.0315 loss: 0.6863 +2023/06/01 06:16:57 - mmengine - INFO - Epoch(train) [9][ 8400/11516] lr: 6.5796e-04 eta: 15:04:47 time: 0.3979 data_time: 0.0012 memory: 9465 grad_norm: 0.0339 loss: 0.6856 +2023/06/01 06:17:38 - mmengine - INFO - Epoch(train) [9][ 8500/11516] lr: 6.5796e-04 eta: 15:04:04 time: 0.4204 data_time: 0.0012 memory: 9465 grad_norm: 0.0194 loss: 0.6882 +2023/06/01 06:18:19 - mmengine - INFO - Epoch(train) [9][ 8600/11516] lr: 6.5796e-04 eta: 15:03:21 time: 0.4341 data_time: 0.0012 memory: 9465 grad_norm: 0.0209 loss: 0.6846 +2023/06/01 06:18:59 - mmengine - INFO - Epoch(train) [9][ 8700/11516] lr: 6.5796e-04 eta: 15:02:37 time: 0.4252 data_time: 0.0012 memory: 9465 grad_norm: 0.0217 loss: 0.6866 +2023/06/01 06:19:42 - mmengine - INFO - Epoch(train) [9][ 8800/11516] lr: 6.5796e-04 eta: 15:01:57 time: 0.4315 data_time: 0.0012 memory: 9465 grad_norm: 0.0184 loss: 0.6897 +2023/06/01 06:20:11 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/06/01 06:20:23 - mmengine - INFO - Epoch(train) [9][ 8900/11516] lr: 6.5796e-04 eta: 15:01:14 time: 0.4266 data_time: 0.0012 memory: 9465 grad_norm: 0.0205 loss: 0.6878 +2023/06/01 06:21:03 - mmengine - INFO - Epoch(train) [9][ 9000/11516] lr: 6.5796e-04 eta: 15:00:29 time: 0.3764 data_time: 0.0012 memory: 9465 grad_norm: 0.0221 loss: 0.6875 +2023/06/01 06:21:44 - mmengine - INFO - Epoch(train) [9][ 9100/11516] lr: 6.5796e-04 eta: 14:59:47 time: 0.4267 data_time: 0.0013 memory: 9465 grad_norm: 0.0315 loss: 0.6911 +2023/06/01 06:22:26 - mmengine - INFO - Epoch(train) [9][ 9200/11516] lr: 6.5796e-04 eta: 14:59:04 time: 0.4185 data_time: 0.0014 memory: 9465 grad_norm: 0.0203 loss: 0.6822 +2023/06/01 06:23:07 - mmengine - INFO - Epoch(train) [9][ 9300/11516] lr: 6.5796e-04 eta: 14:58:22 time: 0.4001 data_time: 0.0013 memory: 9465 grad_norm: 0.0213 loss: 0.6890 +2023/06/01 06:23:48 - mmengine - INFO - Epoch(train) [9][ 9400/11516] lr: 6.5796e-04 eta: 14:57:39 time: 0.4378 data_time: 0.0013 memory: 9465 grad_norm: 0.0207 loss: 0.6890 +2023/06/01 06:24:29 - mmengine - INFO - Epoch(train) [9][ 9500/11516] lr: 6.5796e-04 eta: 14:56:57 time: 0.4190 data_time: 0.0012 memory: 9465 grad_norm: 0.0304 loss: 0.6890 +2023/06/01 06:25:11 - mmengine - INFO - Epoch(train) [9][ 9600/11516] lr: 6.5796e-04 eta: 14:56:15 time: 0.3891 data_time: 0.0012 memory: 9465 grad_norm: 0.0284 loss: 0.6843 +2023/06/01 06:25:52 - mmengine - INFO - Epoch(train) [9][ 9700/11516] lr: 6.5796e-04 eta: 14:55:32 time: 0.4348 data_time: 0.0013 memory: 9465 grad_norm: 0.0263 loss: 0.6874 +2023/06/01 06:26:35 - mmengine - INFO - Epoch(train) [9][ 9800/11516] lr: 6.5796e-04 eta: 14:54:51 time: 0.4044 data_time: 0.0011 memory: 9465 grad_norm: 0.0253 loss: 0.6882 +2023/06/01 06:27:04 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/06/01 06:27:16 - mmengine - INFO - Epoch(train) [9][ 9900/11516] lr: 6.5796e-04 eta: 14:54:08 time: 0.4123 data_time: 0.0014 memory: 9465 grad_norm: 0.0330 loss: 0.6894 +2023/06/01 06:27:57 - mmengine - INFO - Epoch(train) [9][10000/11516] lr: 6.5796e-04 eta: 14:53:26 time: 0.4069 data_time: 0.0012 memory: 9465 grad_norm: 0.0155 loss: 0.6841 +2023/06/01 06:28:38 - mmengine - INFO - Epoch(train) [9][10100/11516] lr: 6.5796e-04 eta: 14:52:43 time: 0.3965 data_time: 0.0013 memory: 9465 grad_norm: 0.0185 loss: 0.6879 +2023/06/01 06:29:20 - mmengine - INFO - Epoch(train) [9][10200/11516] lr: 6.5796e-04 eta: 14:52:01 time: 0.4238 data_time: 0.0012 memory: 9465 grad_norm: 0.0242 loss: 0.6832 +2023/06/01 06:30:01 - mmengine - INFO - Epoch(train) [9][10300/11516] lr: 6.5796e-04 eta: 14:51:19 time: 0.3995 data_time: 0.0013 memory: 9465 grad_norm: 0.0251 loss: 0.6910 +2023/06/01 06:30:42 - mmengine - INFO - Epoch(train) [9][10400/11516] lr: 6.5796e-04 eta: 14:50:36 time: 0.3964 data_time: 0.0014 memory: 9465 grad_norm: 0.0189 loss: 0.6875 +2023/06/01 06:31:23 - mmengine - INFO - Epoch(train) [9][10500/11516] lr: 6.5796e-04 eta: 14:49:53 time: 0.4161 data_time: 0.0015 memory: 9465 grad_norm: 0.0214 loss: 0.6877 +2023/06/01 06:32:04 - mmengine - INFO - Epoch(train) [9][10600/11516] lr: 6.5796e-04 eta: 14:49:10 time: 0.4047 data_time: 0.0012 memory: 9465 grad_norm: 0.0178 loss: 0.6852 +2023/06/01 06:32:46 - mmengine - INFO - Epoch(train) [9][10700/11516] lr: 6.5796e-04 eta: 14:48:28 time: 0.4169 data_time: 0.0011 memory: 9465 grad_norm: 0.0249 loss: 0.6884 +2023/06/01 06:33:27 - mmengine - INFO - Epoch(train) [9][10800/11516] lr: 6.5796e-04 eta: 14:47:45 time: 0.3887 data_time: 0.0012 memory: 9465 grad_norm: 0.0182 loss: 0.6901 +2023/06/01 06:33:56 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/06/01 06:34:09 - mmengine - INFO - Epoch(train) [9][10900/11516] lr: 6.5796e-04 eta: 14:47:03 time: 0.4027 data_time: 0.0012 memory: 9465 grad_norm: 0.0205 loss: 0.6794 +2023/06/01 06:34:51 - mmengine - INFO - Epoch(train) [9][11000/11516] lr: 6.5796e-04 eta: 14:46:22 time: 0.4594 data_time: 0.0012 memory: 9465 grad_norm: 0.0197 loss: 0.6906 +2023/06/01 06:35:32 - mmengine - INFO - Epoch(train) [9][11100/11516] lr: 6.5796e-04 eta: 14:45:40 time: 0.4001 data_time: 0.0012 memory: 9465 grad_norm: 0.0145 loss: 0.6874 +2023/06/01 06:36:13 - mmengine - INFO - Epoch(train) [9][11200/11516] lr: 6.5796e-04 eta: 14:44:56 time: 0.4112 data_time: 0.0024 memory: 9465 grad_norm: 0.0364 loss: 0.6868 +2023/06/01 06:36:52 - mmengine - INFO - Epoch(train) [9][11300/11516] lr: 6.5796e-04 eta: 14:44:11 time: 0.4030 data_time: 0.0013 memory: 9465 grad_norm: 0.0240 loss: 0.6868 +2023/06/01 06:37:32 - mmengine - INFO - Epoch(train) [9][11400/11516] lr: 6.5796e-04 eta: 14:43:27 time: 0.4160 data_time: 0.0011 memory: 9465 grad_norm: 0.0287 loss: 0.6859 +2023/06/01 06:38:12 - mmengine - INFO - Epoch(train) [9][11500/11516] lr: 6.5796e-04 eta: 14:42:43 time: 0.3970 data_time: 0.0079 memory: 9465 grad_norm: 0.0214 loss: 0.6927 +2023/06/01 06:38:17 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/06/01 06:38:17 - mmengine - INFO - Saving checkpoint at 9 epochs +2023/06/01 06:38:33 - mmengine - INFO - Epoch(val) [9][16/16] accuracy/top1: 100.0000 single-label/precision_classwise: [100.0, 0.0] single-label/recall_classwise: [100.0, 0.0] single-label/f1-score_classwise: [100.0, 0.0] data_time: 0.2937 time: 0.5468 +2023/06/01 06:39:16 - mmengine - INFO - Epoch(train) [10][ 100/11516] lr: 5.8244e-04 eta: 14:41:54 time: 0.3872 data_time: 0.0804 memory: 9465 grad_norm: 0.0139 loss: 0.6850 +2023/06/01 06:39:56 - mmengine - INFO - Epoch(train) [10][ 200/11516] lr: 5.8244e-04 eta: 14:41:10 time: 0.3745 data_time: 0.1246 memory: 9465 grad_norm: 0.0186 loss: 0.6929 +2023/06/01 06:40:37 - mmengine - INFO - Epoch(train) [10][ 300/11516] lr: 5.8244e-04 eta: 14:40:27 time: 0.3911 data_time: 0.0012 memory: 9465 grad_norm: 0.0193 loss: 0.6920 +2023/06/01 06:40:59 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/06/01 06:41:18 - mmengine - INFO - Epoch(train) [10][ 400/11516] lr: 5.8244e-04 eta: 14:39:45 time: 0.4162 data_time: 0.0014 memory: 9465 grad_norm: 0.0205 loss: 0.6909 +2023/06/01 06:41:59 - mmengine - INFO - Epoch(train) [10][ 500/11516] lr: 5.8244e-04 eta: 14:39:02 time: 0.3931 data_time: 0.0014 memory: 9465 grad_norm: 0.0223 loss: 0.6895 +2023/06/01 06:42:40 - mmengine - INFO - Epoch(train) [10][ 600/11516] lr: 5.8244e-04 eta: 14:38:19 time: 0.4322 data_time: 0.0014 memory: 9465 grad_norm: 0.0236 loss: 0.6878 +2023/06/01 06:43:21 - mmengine - INFO - Epoch(train) [10][ 700/11516] lr: 5.8244e-04 eta: 14:37:36 time: 0.4269 data_time: 0.0012 memory: 9465 grad_norm: 0.0197 loss: 0.6821 +2023/06/01 06:44:03 - mmengine - INFO - Epoch(train) [10][ 800/11516] lr: 5.8244e-04 eta: 14:36:54 time: 0.4212 data_time: 0.0012 memory: 9465 grad_norm: 0.0161 loss: 0.6871 +2023/06/01 06:44:43 - mmengine - INFO - Epoch(train) [10][ 900/11516] lr: 5.8244e-04 eta: 14:36:11 time: 0.3942 data_time: 0.0012 memory: 9465 grad_norm: 0.0189 loss: 0.6876 +2023/06/01 06:45:25 - mmengine - INFO - Epoch(train) [10][ 1000/11516] lr: 5.8244e-04 eta: 14:35:29 time: 0.4117 data_time: 0.0014 memory: 9465 grad_norm: 0.0283 loss: 0.6881 +2023/06/01 06:46:07 - mmengine - INFO - Epoch(train) [10][ 1100/11516] lr: 5.8244e-04 eta: 14:34:47 time: 0.4208 data_time: 0.0014 memory: 9465 grad_norm: 0.0255 loss: 0.6845 +2023/06/01 06:46:49 - mmengine - INFO - Epoch(train) [10][ 1200/11516] lr: 5.8244e-04 eta: 14:34:06 time: 0.4270 data_time: 0.0015 memory: 9465 grad_norm: 0.0332 loss: 0.6921 +2023/06/01 06:47:30 - mmengine - INFO - Epoch(train) [10][ 1300/11516] lr: 5.8244e-04 eta: 14:33:22 time: 0.4025 data_time: 0.0016 memory: 9465 grad_norm: 0.0215 loss: 0.6887 +2023/06/01 06:47:51 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/06/01 06:48:10 - mmengine - INFO - Epoch(train) [10][ 1400/11516] lr: 5.8244e-04 eta: 14:32:39 time: 0.4074 data_time: 0.0013 memory: 9465 grad_norm: 0.0193 loss: 0.6901 +2023/06/01 06:48:51 - mmengine - INFO - Epoch(train) [10][ 1500/11516] lr: 5.8244e-04 eta: 14:31:57 time: 0.4227 data_time: 0.0014 memory: 9465 grad_norm: 0.0210 loss: 0.6887 +2023/06/01 06:49:34 - mmengine - INFO - Epoch(train) [10][ 1600/11516] lr: 5.8244e-04 eta: 14:31:15 time: 0.4081 data_time: 0.0013 memory: 9465 grad_norm: 0.0112 loss: 0.6935 +2023/06/01 06:50:16 - mmengine - INFO - Epoch(train) [10][ 1700/11516] lr: 5.8244e-04 eta: 14:30:34 time: 0.4532 data_time: 0.0013 memory: 9465 grad_norm: 0.0166 loss: 0.6876 +2023/06/01 06:50:57 - mmengine - INFO - Epoch(train) [10][ 1800/11516] lr: 5.8244e-04 eta: 14:29:52 time: 0.4051 data_time: 0.0013 memory: 9465 grad_norm: 0.0367 loss: 0.6848 +2023/06/01 06:51:37 - mmengine - INFO - Epoch(train) [10][ 1900/11516] lr: 5.8244e-04 eta: 14:29:08 time: 0.4207 data_time: 0.0013 memory: 9465 grad_norm: 0.0151 loss: 0.6832 +2023/06/01 06:52:18 - mmengine - INFO - Epoch(train) [10][ 2000/11516] lr: 5.8244e-04 eta: 14:28:25 time: 0.3905 data_time: 0.0013 memory: 9465 grad_norm: 0.0193 loss: 0.6855 +2023/06/01 06:53:00 - mmengine - INFO - Epoch(train) [10][ 2100/11516] lr: 5.8244e-04 eta: 14:27:43 time: 0.4225 data_time: 0.0014 memory: 9465 grad_norm: 0.0193 loss: 0.6879 +2023/06/01 06:53:42 - mmengine - INFO - Epoch(train) [10][ 2200/11516] lr: 5.8244e-04 eta: 14:27:02 time: 0.4184 data_time: 0.0014 memory: 9465 grad_norm: 0.0159 loss: 0.6879 +2023/06/01 06:54:23 - mmengine - INFO - Epoch(train) [10][ 2300/11516] lr: 5.8244e-04 eta: 14:26:19 time: 0.4241 data_time: 0.0013 memory: 9465 grad_norm: 0.0154 loss: 0.6887 +2023/06/01 06:54:43 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/06/01 06:55:03 - mmengine - INFO - Epoch(train) [10][ 2400/11516] lr: 5.8244e-04 eta: 14:25:35 time: 0.4306 data_time: 0.0014 memory: 9465 grad_norm: 0.0271 loss: 0.6842 +2023/06/01 06:55:44 - mmengine - INFO - Epoch(train) [10][ 2500/11516] lr: 5.8244e-04 eta: 14:24:52 time: 0.4136 data_time: 0.0012 memory: 9465 grad_norm: 0.0264 loss: 0.6878 +2023/06/01 06:56:25 - mmengine - INFO - Epoch(train) [10][ 2600/11516] lr: 5.8244e-04 eta: 14:24:10 time: 0.4164 data_time: 0.0012 memory: 9465 grad_norm: 0.0158 loss: 0.6897 +2023/06/01 06:57:06 - mmengine - INFO - Epoch(train) [10][ 2700/11516] lr: 5.8244e-04 eta: 14:23:27 time: 0.4280 data_time: 0.0013 memory: 9465 grad_norm: 0.0196 loss: 0.6850 +2023/06/01 06:57:47 - mmengine - INFO - Epoch(train) [10][ 2800/11516] lr: 5.8244e-04 eta: 14:22:44 time: 0.3970 data_time: 0.0012 memory: 9465 grad_norm: 0.0244 loss: 0.6859 +2023/06/01 06:58:28 - mmengine - INFO - Epoch(train) [10][ 2900/11516] lr: 5.8244e-04 eta: 14:22:01 time: 0.3918 data_time: 0.0012 memory: 9465 grad_norm: 0.0193 loss: 0.6839 +2023/06/01 06:59:09 - mmengine - INFO - Epoch(train) [10][ 3000/11516] lr: 5.8244e-04 eta: 14:21:19 time: 0.4468 data_time: 0.0012 memory: 9465 grad_norm: 0.0310 loss: 0.6849 +2023/06/01 06:59:50 - mmengine - INFO - Epoch(train) [10][ 3100/11516] lr: 5.8244e-04 eta: 14:20:36 time: 0.4294 data_time: 0.0014 memory: 9465 grad_norm: 0.0302 loss: 0.6901 +2023/06/01 07:00:31 - mmengine - INFO - Epoch(train) [10][ 3200/11516] lr: 5.8244e-04 eta: 14:19:54 time: 0.4044 data_time: 0.0013 memory: 9465 grad_norm: 0.0200 loss: 0.6830 +2023/06/01 07:01:11 - mmengine - INFO - Epoch(train) [10][ 3300/11516] lr: 5.8244e-04 eta: 14:19:10 time: 0.3855 data_time: 0.0012 memory: 9465 grad_norm: 0.0208 loss: 0.6902 +2023/06/01 07:01:33 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/06/01 07:01:52 - mmengine - INFO - Epoch(train) [10][ 3400/11516] lr: 5.8244e-04 eta: 14:18:27 time: 0.3942 data_time: 0.0011 memory: 9465 grad_norm: 0.0223 loss: 0.6845 +2023/06/01 07:02:33 - mmengine - INFO - Epoch(train) [10][ 3500/11516] lr: 5.8244e-04 eta: 14:17:44 time: 0.4170 data_time: 0.0012 memory: 9465 grad_norm: 0.0157 loss: 0.6870 +2023/06/01 07:03:15 - mmengine - INFO - Epoch(train) [10][ 3600/11516] lr: 5.8244e-04 eta: 14:17:02 time: 0.4121 data_time: 0.0013 memory: 9465 grad_norm: 0.0184 loss: 0.6825 +2023/06/01 07:03:55 - mmengine - INFO - Epoch(train) [10][ 3700/11516] lr: 5.8244e-04 eta: 14:16:19 time: 0.4075 data_time: 0.0012 memory: 9465 grad_norm: 0.0202 loss: 0.6805 +2023/06/01 07:04:37 - mmengine - INFO - Epoch(train) [10][ 3800/11516] lr: 5.8244e-04 eta: 14:15:37 time: 0.4158 data_time: 0.0013 memory: 9465 grad_norm: 0.0199 loss: 0.6839 +2023/06/01 07:05:18 - mmengine - INFO - Epoch(train) [10][ 3900/11516] lr: 5.8244e-04 eta: 14:14:54 time: 0.4090 data_time: 0.0012 memory: 9465 grad_norm: 0.0236 loss: 0.6887 +2023/06/01 07:05:59 - mmengine - INFO - Epoch(train) [10][ 4000/11516] lr: 5.8244e-04 eta: 14:14:12 time: 0.3921 data_time: 0.0014 memory: 9465 grad_norm: 0.0080 loss: 0.6865 +2023/06/01 07:06:39 - mmengine - INFO - Epoch(train) [10][ 4100/11516] lr: 5.8244e-04 eta: 14:13:28 time: 0.4177 data_time: 0.0013 memory: 9465 grad_norm: 0.0199 loss: 0.6856 +2023/06/01 07:07:20 - mmengine - INFO - Epoch(train) [10][ 4200/11516] lr: 5.8244e-04 eta: 14:12:45 time: 0.4240 data_time: 0.0012 memory: 9465 grad_norm: 0.0159 loss: 0.6819 +2023/06/01 07:08:00 - mmengine - INFO - Epoch(train) [10][ 4300/11516] lr: 5.8244e-04 eta: 14:12:02 time: 0.4113 data_time: 0.0013 memory: 9465 grad_norm: 0.0164 loss: 0.6855 +2023/06/01 07:08:22 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/06/01 07:08:41 - mmengine - INFO - Epoch(train) [10][ 4400/11516] lr: 5.8244e-04 eta: 14:11:19 time: 0.4022 data_time: 0.0014 memory: 9465 grad_norm: 0.0193 loss: 0.6876 +2023/06/01 07:09:24 - mmengine - INFO - Epoch(train) [10][ 4500/11516] lr: 5.8244e-04 eta: 14:10:39 time: 0.4513 data_time: 0.0012 memory: 9465 grad_norm: 0.0187 loss: 0.6879 +2023/06/01 07:10:06 - mmengine - INFO - Epoch(train) [10][ 4600/11516] lr: 5.8244e-04 eta: 14:09:57 time: 0.3951 data_time: 0.0015 memory: 9465 grad_norm: 0.0312 loss: 0.6895 +2023/06/01 07:10:46 - mmengine - INFO - Epoch(train) [10][ 4700/11516] lr: 5.8244e-04 eta: 14:09:14 time: 0.3847 data_time: 0.0012 memory: 9465 grad_norm: 0.0194 loss: 0.6873 +2023/06/01 07:11:27 - mmengine - INFO - Epoch(train) [10][ 4800/11516] lr: 5.8244e-04 eta: 14:08:31 time: 0.4149 data_time: 0.0014 memory: 9465 grad_norm: 0.0161 loss: 0.6881 +2023/06/01 07:12:08 - mmengine - INFO - Epoch(train) [10][ 4900/11516] lr: 5.8244e-04 eta: 14:07:48 time: 0.3929 data_time: 0.0012 memory: 9465 grad_norm: 0.0219 loss: 0.6920 +2023/06/01 07:12:49 - mmengine - INFO - Epoch(train) [10][ 5000/11516] lr: 5.8244e-04 eta: 14:07:05 time: 0.3955 data_time: 0.0012 memory: 9465 grad_norm: 0.0373 loss: 0.6892 +2023/06/01 07:13:30 - mmengine - INFO - Epoch(train) [10][ 5100/11516] lr: 5.8244e-04 eta: 14:06:23 time: 0.4266 data_time: 0.0013 memory: 9465 grad_norm: 0.0367 loss: 0.6849 +2023/06/01 07:14:12 - mmengine - INFO - Epoch(train) [10][ 5200/11516] lr: 5.8244e-04 eta: 14:05:41 time: 0.3953 data_time: 0.0012 memory: 9465 grad_norm: 0.0305 loss: 0.6896 +2023/06/01 07:14:52 - mmengine - INFO - Epoch(train) [10][ 5300/11516] lr: 5.8244e-04 eta: 14:04:57 time: 0.4078 data_time: 0.0013 memory: 9465 grad_norm: 0.0174 loss: 0.6846 +2023/06/01 07:15:14 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/06/01 07:15:34 - mmengine - INFO - Epoch(train) [10][ 5400/11516] lr: 5.8244e-04 eta: 14:04:16 time: 0.4132 data_time: 0.0013 memory: 9465 grad_norm: 0.0276 loss: 0.6843 +2023/06/01 07:16:15 - mmengine - INFO - Epoch(train) [10][ 5500/11516] lr: 5.8244e-04 eta: 14:03:33 time: 0.3907 data_time: 0.0012 memory: 9465 grad_norm: 0.0176 loss: 0.6845 +2023/06/01 07:16:56 - mmengine - INFO - Epoch(train) [10][ 5600/11516] lr: 5.8244e-04 eta: 14:02:50 time: 0.4174 data_time: 0.0012 memory: 9465 grad_norm: 0.0133 loss: 0.6875 +2023/06/01 07:17:37 - mmengine - INFO - Epoch(train) [10][ 5700/11516] lr: 5.8244e-04 eta: 14:02:08 time: 0.4175 data_time: 0.0012 memory: 9465 grad_norm: 0.0169 loss: 0.6865 +2023/06/01 07:18:19 - mmengine - INFO - Epoch(train) [10][ 5800/11516] lr: 5.8244e-04 eta: 14:01:26 time: 0.4122 data_time: 0.0012 memory: 9465 grad_norm: 0.0220 loss: 0.6885 +2023/06/01 07:19:00 - mmengine - INFO - Epoch(train) [10][ 5900/11516] lr: 5.8244e-04 eta: 14:00:43 time: 0.3972 data_time: 0.0013 memory: 9465 grad_norm: 0.0241 loss: 0.6845 +2023/06/01 07:19:41 - mmengine - INFO - Epoch(train) [10][ 6000/11516] lr: 5.8244e-04 eta: 14:00:01 time: 0.4002 data_time: 0.0012 memory: 9465 grad_norm: 0.0192 loss: 0.6887 +2023/06/01 07:20:22 - mmengine - INFO - Epoch(train) [10][ 6100/11516] lr: 5.8244e-04 eta: 13:59:18 time: 0.4490 data_time: 0.0013 memory: 9465 grad_norm: 0.0200 loss: 0.6904 +2023/06/01 07:21:03 - mmengine - INFO - Epoch(train) [10][ 6200/11516] lr: 5.8244e-04 eta: 13:58:36 time: 0.4038 data_time: 0.0012 memory: 9465 grad_norm: 0.0253 loss: 0.6889 +2023/06/01 07:21:44 - mmengine - INFO - Epoch(train) [10][ 6300/11516] lr: 5.8244e-04 eta: 13:57:53 time: 0.4258 data_time: 0.0012 memory: 9465 grad_norm: 0.0256 loss: 0.6867 +2023/06/01 07:22:06 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/06/01 07:22:25 - mmengine - INFO - Epoch(train) [10][ 6400/11516] lr: 5.8244e-04 eta: 13:57:11 time: 0.4148 data_time: 0.0013 memory: 9465 grad_norm: 0.0300 loss: 0.6862 +2023/06/01 07:23:07 - mmengine - INFO - Epoch(train) [10][ 6500/11516] lr: 5.8244e-04 eta: 13:56:29 time: 0.4091 data_time: 0.0012 memory: 9465 grad_norm: 0.0222 loss: 0.6808 +2023/06/01 07:23:48 - mmengine - INFO - Epoch(train) [10][ 6600/11516] lr: 5.8244e-04 eta: 13:55:46 time: 0.3775 data_time: 0.0012 memory: 9465 grad_norm: 0.0264 loss: 0.6916 +2023/06/01 07:24:30 - mmengine - INFO - Epoch(train) [10][ 6700/11516] lr: 5.8244e-04 eta: 13:55:05 time: 0.3986 data_time: 0.0012 memory: 9465 grad_norm: 0.0301 loss: 0.6924 +2023/06/01 07:25:12 - mmengine - INFO - Epoch(train) [10][ 6800/11516] lr: 5.8244e-04 eta: 13:54:24 time: 0.4083 data_time: 0.0013 memory: 9465 grad_norm: 0.0233 loss: 0.6916 +2023/06/01 07:25:54 - mmengine - INFO - Epoch(train) [10][ 6900/11516] lr: 5.8244e-04 eta: 13:53:41 time: 0.4076 data_time: 0.0012 memory: 9465 grad_norm: 0.0146 loss: 0.6903 +2023/06/01 07:26:35 - mmengine - INFO - Epoch(train) [10][ 7000/11516] lr: 5.8244e-04 eta: 13:53:00 time: 0.4589 data_time: 0.0015 memory: 9465 grad_norm: 0.0195 loss: 0.6895 +2023/06/01 07:27:17 - mmengine - INFO - Epoch(train) [10][ 7100/11516] lr: 5.8244e-04 eta: 13:52:18 time: 0.4069 data_time: 0.0013 memory: 9465 grad_norm: 0.0169 loss: 0.6862 +2023/06/01 07:27:57 - mmengine - INFO - Epoch(train) [10][ 7200/11516] lr: 5.8244e-04 eta: 13:51:34 time: 0.3966 data_time: 0.0012 memory: 9465 grad_norm: 0.0240 loss: 0.6865 +2023/06/01 07:28:39 - mmengine - INFO - Epoch(train) [10][ 7300/11516] lr: 5.8244e-04 eta: 13:50:52 time: 0.4124 data_time: 0.0013 memory: 9465 grad_norm: 0.0192 loss: 0.6847 +2023/06/01 07:29:01 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/06/01 07:29:21 - mmengine - INFO - Epoch(train) [10][ 7400/11516] lr: 5.8244e-04 eta: 13:50:11 time: 0.4366 data_time: 0.0014 memory: 9465 grad_norm: 0.0308 loss: 0.6867 +2023/06/01 07:30:02 - mmengine - INFO - Epoch(train) [10][ 7500/11516] lr: 5.8244e-04 eta: 13:49:28 time: 0.4074 data_time: 0.0012 memory: 9465 grad_norm: 0.0252 loss: 0.6830 +2023/06/01 07:30:42 - mmengine - INFO - Epoch(train) [10][ 7600/11516] lr: 5.8244e-04 eta: 13:48:45 time: 0.4237 data_time: 0.0012 memory: 9465 grad_norm: 0.0160 loss: 0.6824 +2023/06/01 07:31:22 - mmengine - INFO - Epoch(train) [10][ 7700/11516] lr: 5.8244e-04 eta: 13:48:02 time: 0.3967 data_time: 0.0012 memory: 9465 grad_norm: 0.0237 loss: 0.6827 +2023/06/01 07:32:04 - mmengine - INFO - Epoch(train) [10][ 7800/11516] lr: 5.8244e-04 eta: 13:47:19 time: 0.4308 data_time: 0.0013 memory: 9465 grad_norm: 0.0198 loss: 0.6808 +2023/06/01 07:32:45 - mmengine - INFO - Epoch(train) [10][ 7900/11516] lr: 5.8244e-04 eta: 13:46:37 time: 0.4087 data_time: 0.0013 memory: 9465 grad_norm: 0.0222 loss: 0.6856 +2023/06/01 07:33:26 - mmengine - INFO - Epoch(train) [10][ 8000/11516] lr: 5.8244e-04 eta: 13:45:54 time: 0.4392 data_time: 0.0013 memory: 9465 grad_norm: 0.0193 loss: 0.6886 +2023/06/01 07:34:07 - mmengine - INFO - Epoch(train) [10][ 8100/11516] lr: 5.8244e-04 eta: 13:45:12 time: 0.4058 data_time: 0.0014 memory: 9465 grad_norm: 0.0190 loss: 0.6863 +2023/06/01 07:34:49 - mmengine - INFO - Epoch(train) [10][ 8200/11516] lr: 5.8244e-04 eta: 13:44:30 time: 0.4242 data_time: 0.0012 memory: 9465 grad_norm: 0.0154 loss: 0.6854 +2023/06/01 07:35:30 - mmengine - INFO - Epoch(train) [10][ 8300/11516] lr: 5.8244e-04 eta: 13:43:48 time: 0.4222 data_time: 0.0013 memory: 9465 grad_norm: 0.0166 loss: 0.6842 +2023/06/01 07:35:52 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/06/01 07:36:11 - mmengine - INFO - Epoch(train) [10][ 8400/11516] lr: 5.8244e-04 eta: 13:43:05 time: 0.4089 data_time: 0.0013 memory: 9465 grad_norm: 0.0206 loss: 0.6900 +2023/06/01 07:36:53 - mmengine - INFO - Epoch(train) [10][ 8500/11516] lr: 5.8244e-04 eta: 13:42:24 time: 0.4102 data_time: 0.0012 memory: 9465 grad_norm: 0.0180 loss: 0.6862 +2023/06/01 07:37:34 - mmengine - INFO - Epoch(train) [10][ 8600/11516] lr: 5.8244e-04 eta: 13:41:41 time: 0.3894 data_time: 0.0013 memory: 9465 grad_norm: 0.0208 loss: 0.6860 +2023/06/01 07:38:15 - mmengine - INFO - Epoch(train) [10][ 8700/11516] lr: 5.8244e-04 eta: 13:40:58 time: 0.4587 data_time: 0.0013 memory: 9465 grad_norm: 0.0166 loss: 0.6894 +2023/06/01 07:38:58 - mmengine - INFO - Epoch(train) [10][ 8800/11516] lr: 5.8244e-04 eta: 13:40:18 time: 0.4222 data_time: 0.0013 memory: 9465 grad_norm: 0.0242 loss: 0.6857 +2023/06/01 07:39:38 - mmengine - INFO - Epoch(train) [10][ 8900/11516] lr: 5.8244e-04 eta: 13:39:34 time: 0.4044 data_time: 0.0013 memory: 9465 grad_norm: 0.0244 loss: 0.6880 +2023/06/01 07:40:18 - mmengine - INFO - Epoch(train) [10][ 9000/11516] lr: 5.8244e-04 eta: 13:38:51 time: 0.4102 data_time: 0.0013 memory: 9465 grad_norm: 0.0194 loss: 0.6884 +2023/06/01 07:40:58 - mmengine - INFO - Epoch(train) [10][ 9100/11516] lr: 5.8244e-04 eta: 13:38:07 time: 0.4124 data_time: 0.0013 memory: 9465 grad_norm: 0.0239 loss: 0.6902 +2023/06/01 07:41:39 - mmengine - INFO - Epoch(train) [10][ 9200/11516] lr: 5.8244e-04 eta: 13:37:25 time: 0.4187 data_time: 0.0012 memory: 9465 grad_norm: 0.0302 loss: 0.6843 +2023/06/01 07:42:20 - mmengine - INFO - Epoch(train) [10][ 9300/11516] lr: 5.8244e-04 eta: 13:36:42 time: 0.3987 data_time: 0.0013 memory: 9465 grad_norm: 0.0350 loss: 0.6869 +2023/06/01 07:42:41 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/06/01 07:43:02 - mmengine - INFO - Epoch(train) [10][ 9400/11516] lr: 5.8244e-04 eta: 13:36:01 time: 0.4907 data_time: 0.0012 memory: 9465 grad_norm: 0.0250 loss: 0.6926 +2023/06/01 07:43:45 - mmengine - INFO - Epoch(train) [10][ 9500/11516] lr: 5.8244e-04 eta: 13:35:20 time: 0.4229 data_time: 0.0013 memory: 9465 grad_norm: 0.0292 loss: 0.6851 +2023/06/01 07:44:25 - mmengine - INFO - Epoch(train) [10][ 9600/11516] lr: 5.8244e-04 eta: 13:34:37 time: 0.4044 data_time: 0.0012 memory: 9465 grad_norm: 0.0186 loss: 0.6833 +2023/06/01 07:45:05 - mmengine - INFO - Epoch(train) [10][ 9700/11516] lr: 5.8244e-04 eta: 13:33:53 time: 0.3872 data_time: 0.0012 memory: 9465 grad_norm: 0.0173 loss: 0.6860 +2023/06/01 07:45:47 - mmengine - INFO - Epoch(train) [10][ 9800/11516] lr: 5.8244e-04 eta: 13:33:11 time: 0.3953 data_time: 0.0013 memory: 9465 grad_norm: 0.0217 loss: 0.6915 +2023/06/01 07:46:27 - mmengine - INFO - Epoch(train) [10][ 9900/11516] lr: 5.8244e-04 eta: 13:32:28 time: 0.4210 data_time: 0.0013 memory: 9465 grad_norm: 0.0162 loss: 0.6846 +2023/06/01 07:47:09 - mmengine - INFO - Epoch(train) [10][10000/11516] lr: 5.8244e-04 eta: 13:31:47 time: 0.4064 data_time: 0.0013 memory: 9465 grad_norm: 0.0201 loss: 0.6853 +2023/06/01 07:47:51 - mmengine - INFO - Epoch(train) [10][10100/11516] lr: 5.8244e-04 eta: 13:31:05 time: 0.4414 data_time: 0.0015 memory: 9465 grad_norm: 0.0291 loss: 0.6879 +2023/06/01 07:48:33 - mmengine - INFO - Epoch(train) [10][10200/11516] lr: 5.8244e-04 eta: 13:30:24 time: 0.4268 data_time: 0.0011 memory: 9465 grad_norm: 0.0200 loss: 0.6871 +2023/06/01 07:49:14 - mmengine - INFO - Epoch(train) [10][10300/11516] lr: 5.8244e-04 eta: 13:29:41 time: 0.4045 data_time: 0.0012 memory: 9465 grad_norm: 0.0157 loss: 0.6869 +2023/06/01 07:49:37 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/06/01 07:49:56 - mmengine - INFO - Epoch(train) [10][10400/11516] lr: 5.8244e-04 eta: 13:29:00 time: 0.4240 data_time: 0.0013 memory: 9465 grad_norm: 0.0188 loss: 0.6870 +2023/06/01 07:50:37 - mmengine - INFO - Epoch(train) [10][10500/11516] lr: 5.8244e-04 eta: 13:28:17 time: 0.4174 data_time: 0.0012 memory: 9465 grad_norm: 0.0207 loss: 0.6857 +2023/06/01 07:51:18 - mmengine - INFO - Epoch(train) [10][10600/11516] lr: 5.8244e-04 eta: 13:27:34 time: 0.4093 data_time: 0.0012 memory: 9465 grad_norm: 0.0271 loss: 0.6837 +2023/06/01 07:51:58 - mmengine - INFO - Epoch(train) [10][10700/11516] lr: 5.8244e-04 eta: 13:26:51 time: 0.4248 data_time: 0.0013 memory: 9465 grad_norm: 0.0222 loss: 0.6899 +2023/06/01 07:52:38 - mmengine - INFO - Epoch(train) [10][10800/11516] lr: 5.8244e-04 eta: 13:26:07 time: 0.4004 data_time: 0.0013 memory: 9465 grad_norm: 0.0276 loss: 0.6881 +2023/06/01 07:53:19 - mmengine - INFO - Epoch(train) [10][10900/11516] lr: 5.8244e-04 eta: 13:25:25 time: 0.4081 data_time: 0.0012 memory: 9465 grad_norm: 0.0203 loss: 0.6833 +2023/06/01 07:53:59 - mmengine - INFO - Epoch(train) [10][11000/11516] lr: 5.8244e-04 eta: 13:24:41 time: 0.4081 data_time: 0.0013 memory: 9465 grad_norm: 0.0195 loss: 0.6837 +2023/06/01 07:54:39 - mmengine - INFO - Epoch(train) [10][11100/11516] lr: 5.8244e-04 eta: 13:23:58 time: 0.4273 data_time: 0.0012 memory: 9465 grad_norm: 0.0178 loss: 0.6832 +2023/06/01 07:55:19 - mmengine - INFO - Epoch(train) [10][11200/11516] lr: 5.8244e-04 eta: 13:23:14 time: 0.3691 data_time: 0.0012 memory: 9465 grad_norm: 0.0172 loss: 0.6894 +2023/06/01 07:55:58 - mmengine - INFO - Epoch(train) [10][11300/11516] lr: 5.8244e-04 eta: 13:22:30 time: 0.4567 data_time: 0.0021 memory: 9465 grad_norm: 0.0314 loss: 0.6835 +2023/06/01 07:56:20 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/06/01 07:56:40 - mmengine - INFO - Epoch(train) [10][11400/11516] lr: 5.8244e-04 eta: 13:21:48 time: 0.4413 data_time: 0.0016 memory: 9465 grad_norm: 0.0173 loss: 0.6818 +2023/06/01 07:57:22 - mmengine - INFO - Epoch(train) [10][11500/11516] lr: 5.8244e-04 eta: 13:21:07 time: 0.4231 data_time: 0.0074 memory: 9465 grad_norm: 0.0154 loss: 0.6815 +2023/06/01 07:57:27 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/06/01 07:57:27 - mmengine - INFO - Saving checkpoint at 10 epochs +2023/06/01 07:57:44 - mmengine - INFO - Epoch(val) [10][16/16] accuracy/top1: 100.0000 single-label/precision_classwise: [100.0, 0.0] single-label/recall_classwise: [100.0, 0.0] single-label/f1-score_classwise: [100.0, 0.0] data_time: 0.3131 time: 0.5659 +2023/06/01 07:58:26 - mmengine - INFO - Epoch(train) [11][ 100/11516] lr: 5.0500e-04 eta: 13:20:17 time: 0.3846 data_time: 0.1416 memory: 9465 grad_norm: 0.0181 loss: 0.6851 +2023/06/01 07:59:08 - mmengine - INFO - Epoch(train) [11][ 200/11516] lr: 5.0500e-04 eta: 13:19:35 time: 0.4263 data_time: 0.0012 memory: 9465 grad_norm: 0.0303 loss: 0.6859 +2023/06/01 07:59:50 - mmengine - INFO - Epoch(train) [11][ 300/11516] lr: 5.0500e-04 eta: 13:18:54 time: 0.4278 data_time: 0.0015 memory: 9465 grad_norm: 0.0266 loss: 0.6922 +2023/06/01 08:00:31 - mmengine - INFO - Epoch(train) [11][ 400/11516] lr: 5.0500e-04 eta: 13:18:12 time: 0.4226 data_time: 0.0012 memory: 9465 grad_norm: 0.0252 loss: 0.6911 +2023/06/01 08:01:13 - mmengine - INFO - Epoch(train) [11][ 500/11516] lr: 5.0500e-04 eta: 13:17:31 time: 0.4010 data_time: 0.0012 memory: 9465 grad_norm: 0.0259 loss: 0.6853 +2023/06/01 08:01:55 - mmengine - INFO - Epoch(train) [11][ 600/11516] lr: 5.0500e-04 eta: 13:16:49 time: 0.4170 data_time: 0.0012 memory: 9465 grad_norm: 0.0196 loss: 0.6902 +2023/06/01 08:02:37 - mmengine - INFO - Epoch(train) [11][ 700/11516] lr: 5.0500e-04 eta: 13:16:08 time: 0.4154 data_time: 0.0015 memory: 9465 grad_norm: 0.0166 loss: 0.6906 +2023/06/01 08:03:19 - mmengine - INFO - Epoch(train) [11][ 800/11516] lr: 5.0500e-04 eta: 13:15:26 time: 0.4090 data_time: 0.0013 memory: 9465 grad_norm: 0.0185 loss: 0.6844 +2023/06/01 08:03:36 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/06/01 08:04:01 - mmengine - INFO - Epoch(train) [11][ 900/11516] lr: 5.0500e-04 eta: 13:14:44 time: 0.4196 data_time: 0.0014 memory: 9465 grad_norm: 0.0192 loss: 0.6852 +2023/06/01 08:04:42 - mmengine - INFO - Epoch(train) [11][ 1000/11516] lr: 5.0500e-04 eta: 13:14:02 time: 0.4054 data_time: 0.0017 memory: 9465 grad_norm: 0.0253 loss: 0.6915 +2023/06/01 08:05:23 - mmengine - INFO - Epoch(train) [11][ 1100/11516] lr: 5.0500e-04 eta: 13:13:20 time: 0.4082 data_time: 0.0014 memory: 9465 grad_norm: 0.0344 loss: 0.6859 +2023/06/01 08:06:06 - mmengine - INFO - Epoch(train) [11][ 1200/11516] lr: 5.0500e-04 eta: 13:12:39 time: 0.4037 data_time: 0.0012 memory: 9465 grad_norm: 0.0240 loss: 0.6888 +2023/06/01 08:06:48 - mmengine - INFO - Epoch(train) [11][ 1300/11516] lr: 5.0500e-04 eta: 13:11:58 time: 0.4144 data_time: 0.0013 memory: 9465 grad_norm: 0.0293 loss: 0.6877 +2023/06/01 08:07:30 - mmengine - INFO - Epoch(train) [11][ 1400/11516] lr: 5.0500e-04 eta: 13:11:16 time: 0.4203 data_time: 0.0013 memory: 9465 grad_norm: 0.0221 loss: 0.6870 +2023/06/01 08:08:12 - mmengine - INFO - Epoch(train) [11][ 1500/11516] lr: 5.0500e-04 eta: 13:10:35 time: 0.4192 data_time: 0.0013 memory: 9465 grad_norm: 0.0146 loss: 0.6854 +2023/06/01 08:08:55 - mmengine - INFO - Epoch(train) [11][ 1600/11516] lr: 5.0500e-04 eta: 13:09:54 time: 0.4244 data_time: 0.0013 memory: 9465 grad_norm: 0.0250 loss: 0.6874 +2023/06/01 08:09:38 - mmengine - INFO - Epoch(train) [11][ 1700/11516] lr: 5.0500e-04 eta: 13:09:13 time: 0.4669 data_time: 0.0012 memory: 9465 grad_norm: 0.0127 loss: 0.6921 +2023/06/01 08:10:20 - mmengine - INFO - Epoch(train) [11][ 1800/11516] lr: 5.0500e-04 eta: 13:08:32 time: 0.4248 data_time: 0.0013 memory: 9465 grad_norm: 0.0126 loss: 0.6829 +2023/06/01 08:10:36 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/06/01 08:11:01 - mmengine - INFO - Epoch(train) [11][ 1900/11516] lr: 5.0500e-04 eta: 13:07:50 time: 0.4110 data_time: 0.0012 memory: 9465 grad_norm: 0.0265 loss: 0.6873 +2023/06/01 08:11:43 - mmengine - INFO - Epoch(train) [11][ 2000/11516] lr: 5.0500e-04 eta: 13:07:08 time: 0.4185 data_time: 0.0015 memory: 9465 grad_norm: 0.0225 loss: 0.6939 +2023/06/01 08:12:27 - mmengine - INFO - Epoch(train) [11][ 2100/11516] lr: 5.0500e-04 eta: 13:06:28 time: 0.4270 data_time: 0.0014 memory: 9465 grad_norm: 0.0159 loss: 0.6896 +2023/06/01 08:13:09 - mmengine - INFO - Epoch(train) [11][ 2200/11516] lr: 5.0500e-04 eta: 13:05:47 time: 0.4168 data_time: 0.0012 memory: 9465 grad_norm: 0.0222 loss: 0.6885 +2023/06/01 08:13:52 - mmengine - INFO - Epoch(train) [11][ 2300/11516] lr: 5.0500e-04 eta: 13:05:06 time: 0.4212 data_time: 0.0014 memory: 9465 grad_norm: 0.0255 loss: 0.6830 +2023/06/01 08:14:34 - mmengine - INFO - Epoch(train) [11][ 2400/11516] lr: 5.0500e-04 eta: 13:04:25 time: 0.4176 data_time: 0.0012 memory: 9465 grad_norm: 0.0300 loss: 0.6895 +2023/06/01 08:15:16 - mmengine - INFO - Epoch(train) [11][ 2500/11516] lr: 5.0500e-04 eta: 13:03:43 time: 0.4122 data_time: 0.0012 memory: 9465 grad_norm: 0.0238 loss: 0.6853 +2023/06/01 08:15:58 - mmengine - INFO - Epoch(train) [11][ 2600/11516] lr: 5.0500e-04 eta: 13:03:02 time: 0.4201 data_time: 0.0013 memory: 9465 grad_norm: 0.0317 loss: 0.6830 +2023/06/01 08:16:40 - mmengine - INFO - Epoch(train) [11][ 2700/11516] lr: 5.0500e-04 eta: 13:02:21 time: 0.4197 data_time: 0.0012 memory: 9465 grad_norm: 0.0181 loss: 0.6852 +2023/06/01 08:17:22 - mmengine - INFO - Epoch(train) [11][ 2800/11516] lr: 5.0500e-04 eta: 13:01:39 time: 0.4110 data_time: 0.0013 memory: 9465 grad_norm: 0.0217 loss: 0.6853 +2023/06/01 08:17:39 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/06/01 08:18:05 - mmengine - INFO - Epoch(train) [11][ 2900/11516] lr: 5.0500e-04 eta: 13:00:58 time: 0.4284 data_time: 0.0014 memory: 9465 grad_norm: 0.0259 loss: 0.6886 +2023/06/01 08:18:48 - mmengine - INFO - Epoch(train) [11][ 3000/11516] lr: 5.0500e-04 eta: 13:00:17 time: 0.4139 data_time: 0.0014 memory: 9465 grad_norm: 0.0187 loss: 0.6858 +2023/06/01 08:19:30 - mmengine - INFO - Epoch(train) [11][ 3100/11516] lr: 5.0500e-04 eta: 12:59:36 time: 0.4514 data_time: 0.0014 memory: 9465 grad_norm: 0.0202 loss: 0.6904 +2023/06/01 08:20:12 - mmengine - INFO - Epoch(train) [11][ 3200/11516] lr: 5.0500e-04 eta: 12:58:55 time: 0.4073 data_time: 0.0012 memory: 9465 grad_norm: 0.0346 loss: 0.6901 +2023/06/01 08:20:54 - mmengine - INFO - Epoch(train) [11][ 3300/11516] lr: 5.0500e-04 eta: 12:58:13 time: 0.4055 data_time: 0.0012 memory: 9465 grad_norm: 0.0197 loss: 0.6868 +2023/06/01 08:21:37 - mmengine - INFO - Epoch(train) [11][ 3400/11516] lr: 5.0500e-04 eta: 12:57:32 time: 0.4370 data_time: 0.0013 memory: 9465 grad_norm: 0.0291 loss: 0.6934 +2023/06/01 08:22:19 - mmengine - INFO - Epoch(train) [11][ 3500/11516] lr: 5.0500e-04 eta: 12:56:51 time: 0.4469 data_time: 0.0013 memory: 9465 grad_norm: 0.0303 loss: 0.6889 +2023/06/01 08:23:02 - mmengine - INFO - Epoch(train) [11][ 3600/11516] lr: 5.0500e-04 eta: 12:56:10 time: 0.4124 data_time: 0.0013 memory: 9465 grad_norm: 0.0331 loss: 0.6867 +2023/06/01 08:23:43 - mmengine - INFO - Epoch(train) [11][ 3700/11516] lr: 5.0500e-04 eta: 12:55:28 time: 0.4180 data_time: 0.0013 memory: 9465 grad_norm: 0.0185 loss: 0.6873 +2023/06/01 08:24:24 - mmengine - INFO - Epoch(train) [11][ 3800/11516] lr: 5.0500e-04 eta: 12:54:46 time: 0.4034 data_time: 0.0014 memory: 9465 grad_norm: 0.0223 loss: 0.6839 +2023/06/01 08:24:41 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/06/01 08:25:07 - mmengine - INFO - Epoch(train) [11][ 3900/11516] lr: 5.0500e-04 eta: 12:54:05 time: 0.4400 data_time: 0.0014 memory: 9465 grad_norm: 0.0494 loss: 0.6855 +2023/06/01 08:25:49 - mmengine - INFO - Epoch(train) [11][ 4000/11516] lr: 5.0500e-04 eta: 12:53:24 time: 0.4342 data_time: 0.0014 memory: 9465 grad_norm: 0.1126 loss: 0.6847 +2023/06/01 08:26:31 - mmengine - INFO - Epoch(train) [11][ 4100/11516] lr: 5.0500e-04 eta: 12:52:42 time: 0.4318 data_time: 0.0012 memory: 9465 grad_norm: 0.0472 loss: 0.6833 +2023/06/01 08:27:13 - mmengine - INFO - Epoch(train) [11][ 4200/11516] lr: 5.0500e-04 eta: 12:52:00 time: 0.4099 data_time: 0.0014 memory: 9465 grad_norm: 0.2207 loss: 0.6839 +2023/06/01 08:27:55 - mmengine - INFO - Epoch(train) [11][ 4300/11516] lr: 5.0500e-04 eta: 12:51:19 time: 0.4447 data_time: 0.0013 memory: 9465 grad_norm: 5.8065 loss: 0.6772 +2023/06/01 08:28:37 - mmengine - INFO - Epoch(train) [11][ 4400/11516] lr: 5.0500e-04 eta: 12:50:37 time: 0.4193 data_time: 0.0013 memory: 9465 grad_norm: 2.1082 loss: 0.6880 +2023/06/01 08:29:18 - mmengine - INFO - Epoch(train) [11][ 4500/11516] lr: 5.0500e-04 eta: 12:49:55 time: 0.4145 data_time: 0.0013 memory: 9465 grad_norm: 11.0263 loss: 0.6785 +2023/06/01 08:30:00 - mmengine - INFO - Epoch(train) [11][ 4600/11516] lr: 5.0500e-04 eta: 12:49:14 time: 0.4069 data_time: 0.0013 memory: 9465 grad_norm: 1.8233 loss: 0.6820 +2023/06/01 08:30:43 - mmengine - INFO - Epoch(train) [11][ 4700/11516] lr: 5.0500e-04 eta: 12:48:33 time: 0.4289 data_time: 0.0013 memory: 9465 grad_norm: 1.4509 loss: 0.6800 +2023/06/01 08:31:26 - mmengine - INFO - Epoch(train) [11][ 4800/11516] lr: 5.0500e-04 eta: 12:47:52 time: 0.4281 data_time: 0.0013 memory: 9465 grad_norm: 1.1201 loss: 0.6786 +2023/06/01 08:31:42 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/06/01 08:32:08 - mmengine - INFO - Epoch(train) [11][ 4900/11516] lr: 5.0500e-04 eta: 12:47:11 time: 0.4543 data_time: 0.0013 memory: 9465 grad_norm: 0.0860 loss: 0.6788 +2023/06/01 08:32:50 - mmengine - INFO - Epoch(train) [11][ 5000/11516] lr: 5.0500e-04 eta: 12:46:29 time: 0.4336 data_time: 0.0012 memory: 9465 grad_norm: 0.3386 loss: 0.6814 +2023/06/01 08:33:33 - mmengine - INFO - Epoch(train) [11][ 5100/11516] lr: 5.0500e-04 eta: 12:45:48 time: 0.4031 data_time: 0.0012 memory: 9465 grad_norm: 0.1002 loss: 0.6799 +2023/06/01 08:34:16 - mmengine - INFO - Epoch(train) [11][ 5200/11516] lr: 5.0500e-04 eta: 12:45:08 time: 0.4436 data_time: 0.0013 memory: 9465 grad_norm: 0.1479 loss: 0.6739 +2023/06/01 08:34:58 - mmengine - INFO - Epoch(train) [11][ 5300/11516] lr: 5.0500e-04 eta: 12:44:26 time: 0.3911 data_time: 0.0014 memory: 9465 grad_norm: 1.6711 loss: 0.6736 +2023/06/01 08:35:41 - mmengine - INFO - Epoch(train) [11][ 5400/11516] lr: 5.0500e-04 eta: 12:43:46 time: 0.4329 data_time: 0.0013 memory: 9465 grad_norm: 0.0840 loss: 0.6805 +2023/06/01 08:36:23 - mmengine - INFO - Epoch(train) [11][ 5500/11516] lr: 5.0500e-04 eta: 12:43:05 time: 0.4129 data_time: 0.0014 memory: 9465 grad_norm: 9.8499 loss: 0.6692 +2023/06/01 08:37:06 - mmengine - INFO - Epoch(train) [11][ 5600/11516] lr: 5.0500e-04 eta: 12:42:23 time: 0.4331 data_time: 0.0014 memory: 9465 grad_norm: 0.6834 loss: 0.6872 +2023/06/01 08:37:48 - mmengine - INFO - Epoch(train) [11][ 5700/11516] lr: 5.0500e-04 eta: 12:41:42 time: 0.4234 data_time: 0.0012 memory: 9465 grad_norm: 47.2562 loss: 0.6793 +2023/06/01 08:38:30 - mmengine - INFO - Epoch(train) [11][ 5800/11516] lr: 5.0500e-04 eta: 12:41:00 time: 0.4043 data_time: 0.0012 memory: 9465 grad_norm: 0.5495 loss: 0.6733 +2023/06/01 08:38:47 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/06/01 08:39:12 - mmengine - INFO - Epoch(train) [11][ 5900/11516] lr: 5.0500e-04 eta: 12:40:19 time: 0.4123 data_time: 0.0014 memory: 9465 grad_norm: 0.3460 loss: 0.6815 +2023/06/01 08:39:55 - mmengine - INFO - Epoch(train) [11][ 6000/11516] lr: 5.0500e-04 eta: 12:39:38 time: 0.4166 data_time: 0.0013 memory: 9465 grad_norm: 1.5020 loss: 0.6745 +2023/06/01 08:40:36 - mmengine - INFO - Epoch(train) [11][ 6100/11516] lr: 5.0500e-04 eta: 12:38:56 time: 0.3968 data_time: 0.0013 memory: 9465 grad_norm: 0.9124 loss: 0.6776 +2023/06/01 08:41:17 - mmengine - INFO - Epoch(train) [11][ 6200/11516] lr: 5.0500e-04 eta: 12:38:13 time: 0.3895 data_time: 0.0014 memory: 9465 grad_norm: 3.3543 loss: 0.6728 +2023/06/01 08:41:59 - mmengine - INFO - Epoch(train) [11][ 6300/11516] lr: 5.0500e-04 eta: 12:37:32 time: 0.3976 data_time: 0.0014 memory: 9465 grad_norm: 15.7907 loss: 0.6835 +2023/06/01 08:42:41 - mmengine - INFO - Epoch(train) [11][ 6400/11516] lr: 5.0500e-04 eta: 12:36:50 time: 0.4169 data_time: 0.0014 memory: 9465 grad_norm: 11.3138 loss: 0.6737 +2023/06/01 08:43:22 - mmengine - INFO - Epoch(train) [11][ 6500/11516] lr: 5.0500e-04 eta: 12:36:08 time: 0.4300 data_time: 0.0014 memory: 9465 grad_norm: 3.2814 loss: 0.6774 +2023/06/01 08:44:05 - mmengine - INFO - Epoch(train) [11][ 6600/11516] lr: 5.0500e-04 eta: 12:35:27 time: 0.4272 data_time: 0.0015 memory: 9465 grad_norm: 1.0304 loss: 0.6823 +2023/06/01 08:44:46 - mmengine - INFO - Epoch(train) [11][ 6700/11516] lr: 5.0500e-04 eta: 12:34:45 time: 0.4061 data_time: 0.0014 memory: 9465 grad_norm: 0.8497 loss: 0.6811 +2023/06/01 08:45:29 - mmengine - INFO - Epoch(train) [11][ 6800/11516] lr: 5.0500e-04 eta: 12:34:04 time: 0.4266 data_time: 0.0015 memory: 9465 grad_norm: 0.5275 loss: 0.6731 +2023/06/01 08:45:46 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/06/01 08:46:11 - mmengine - INFO - Epoch(train) [11][ 6900/11516] lr: 5.0500e-04 eta: 12:33:23 time: 0.4262 data_time: 0.0014 memory: 9465 grad_norm: 2.6347 loss: 0.6795 +2023/06/01 08:46:54 - mmengine - INFO - Epoch(train) [11][ 7000/11516] lr: 5.0500e-04 eta: 12:32:42 time: 0.4264 data_time: 0.0016 memory: 9465 grad_norm: 1.9452 loss: 0.6715 +2023/06/01 08:47:36 - mmengine - INFO - Epoch(train) [11][ 7100/11516] lr: 5.0500e-04 eta: 12:32:00 time: 0.4091 data_time: 0.0013 memory: 9465 grad_norm: 8.5382 loss: 0.6889 +2023/06/01 08:48:18 - mmengine - INFO - Epoch(train) [11][ 7200/11516] lr: 5.0500e-04 eta: 12:31:19 time: 0.4019 data_time: 0.0012 memory: 9465 grad_norm: 2.1022 loss: 0.6825 +2023/06/01 08:49:01 - mmengine - INFO - Epoch(train) [11][ 7300/11516] lr: 5.0500e-04 eta: 12:30:38 time: 0.4226 data_time: 0.0013 memory: 9465 grad_norm: 6.8676 loss: 0.6739 +2023/06/01 08:49:44 - mmengine - INFO - Epoch(train) [11][ 7400/11516] lr: 5.0500e-04 eta: 12:29:57 time: 0.4366 data_time: 0.0012 memory: 9465 grad_norm: 83.4390 loss: 0.6784 +2023/06/01 08:50:25 - mmengine - INFO - Epoch(train) [11][ 7500/11516] lr: 5.0500e-04 eta: 12:29:15 time: 0.3971 data_time: 0.0014 memory: 9465 grad_norm: 2.2073 loss: 0.6637 +2023/06/01 08:51:07 - mmengine - INFO - Epoch(train) [11][ 7600/11516] lr: 5.0500e-04 eta: 12:28:33 time: 0.4012 data_time: 0.0014 memory: 9465 grad_norm: 1.3129 loss: 0.6778 +2023/06/01 08:51:49 - mmengine - INFO - Epoch(train) [11][ 7700/11516] lr: 5.0500e-04 eta: 12:27:52 time: 0.3979 data_time: 0.0014 memory: 9465 grad_norm: 0.9810 loss: 0.6686 +2023/06/01 08:52:32 - mmengine - INFO - Epoch(train) [11][ 7800/11516] lr: 5.0500e-04 eta: 12:27:11 time: 0.4578 data_time: 0.0013 memory: 9465 grad_norm: 2.4079 loss: 0.6691 +2023/06/01 08:52:49 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/06/01 08:53:13 - mmengine - INFO - Epoch(train) [11][ 7900/11516] lr: 5.0500e-04 eta: 12:26:29 time: 0.4324 data_time: 0.0012 memory: 9465 grad_norm: 22.9348 loss: 0.6794 +2023/06/01 08:53:56 - mmengine - INFO - Epoch(train) [11][ 8000/11516] lr: 5.0500e-04 eta: 12:25:48 time: 0.4006 data_time: 0.0015 memory: 9465 grad_norm: 21.4627 loss: 0.6670 +2023/06/01 08:54:38 - mmengine - INFO - Epoch(train) [11][ 8100/11516] lr: 5.0500e-04 eta: 12:25:06 time: 0.4076 data_time: 0.0013 memory: 9465 grad_norm: 7.6485 loss: 0.6770 +2023/06/01 08:55:21 - mmengine - INFO - Epoch(train) [11][ 8200/11516] lr: 5.0500e-04 eta: 12:24:26 time: 0.4010 data_time: 0.0014 memory: 9465 grad_norm: 2.5653 loss: 0.6730 +2023/06/01 08:56:03 - mmengine - INFO - Epoch(train) [11][ 8300/11516] lr: 5.0500e-04 eta: 12:23:45 time: 0.4118 data_time: 0.0013 memory: 9465 grad_norm: 4.9043 loss: 0.6876 +2023/06/01 08:56:45 - mmengine - INFO - Epoch(train) [11][ 8400/11516] lr: 5.0500e-04 eta: 12:23:03 time: 0.4129 data_time: 0.0014 memory: 9465 grad_norm: 1.3246 loss: 0.6804 +2023/06/01 08:57:27 - mmengine - INFO - Epoch(train) [11][ 8500/11516] lr: 5.0500e-04 eta: 12:22:21 time: 0.3892 data_time: 0.0014 memory: 9465 grad_norm: 2.6230 loss: 0.6782 +2023/06/01 08:58:09 - mmengine - INFO - Epoch(train) [11][ 8600/11516] lr: 5.0500e-04 eta: 12:21:39 time: 0.4101 data_time: 0.0014 memory: 9465 grad_norm: 6.4751 loss: 0.6696 +2023/06/01 08:58:51 - mmengine - INFO - Epoch(train) [11][ 8700/11516] lr: 5.0500e-04 eta: 12:20:58 time: 0.4204 data_time: 0.0014 memory: 9465 grad_norm: 111.0918 loss: 0.6710 +2023/06/01 08:59:33 - mmengine - INFO - Epoch(train) [11][ 8800/11516] lr: 5.0500e-04 eta: 12:20:16 time: 0.3943 data_time: 0.0014 memory: 9465 grad_norm: 43.5777 loss: 0.6812 +2023/06/01 08:59:50 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/06/01 09:00:15 - mmengine - INFO - Epoch(train) [11][ 8900/11516] lr: 5.0500e-04 eta: 12:19:35 time: 0.4630 data_time: 0.0013 memory: 9465 grad_norm: 183.5683 loss: 0.6774 +2023/06/01 09:00:57 - mmengine - INFO - Epoch(train) [11][ 9000/11516] lr: 5.0500e-04 eta: 12:18:54 time: 0.4452 data_time: 0.0012 memory: 9465 grad_norm: 11.1078 loss: 0.6797 +2023/06/01 09:01:39 - mmengine - INFO - Epoch(train) [11][ 9100/11516] lr: 5.0500e-04 eta: 12:18:11 time: 0.4193 data_time: 0.0013 memory: 9465 grad_norm: 2.8670 loss: 0.6846 +2023/06/01 09:02:22 - mmengine - INFO - Epoch(train) [11][ 9200/11516] lr: 5.0500e-04 eta: 12:17:31 time: 0.4495 data_time: 0.0014 memory: 9465 grad_norm: 11.6784 loss: 0.6808 +2023/06/01 09:03:04 - mmengine - INFO - Epoch(train) [11][ 9300/11516] lr: 5.0500e-04 eta: 12:16:49 time: 0.4242 data_time: 0.0013 memory: 9465 grad_norm: 13.7742 loss: 0.6752 +2023/06/01 09:03:47 - mmengine - INFO - Epoch(train) [11][ 9400/11516] lr: 5.0500e-04 eta: 12:16:09 time: 0.4398 data_time: 0.0013 memory: 9465 grad_norm: 39.0020 loss: 0.6818 +2023/06/01 09:04:29 - mmengine - INFO - Epoch(train) [11][ 9500/11516] lr: 5.0500e-04 eta: 12:15:27 time: 0.4272 data_time: 0.0013 memory: 9465 grad_norm: 26.9332 loss: 0.6731 +2023/06/01 09:05:13 - mmengine - INFO - Epoch(train) [11][ 9600/11516] lr: 5.0500e-04 eta: 12:14:47 time: 0.4462 data_time: 0.0013 memory: 9465 grad_norm: 7.1107 loss: 0.6828 +2023/06/01 09:05:55 - mmengine - INFO - Epoch(train) [11][ 9700/11516] lr: 5.0500e-04 eta: 12:14:06 time: 0.4126 data_time: 0.0014 memory: 9465 grad_norm: 116.4794 loss: 0.6885 +2023/06/01 09:06:37 - mmengine - INFO - Epoch(train) [11][ 9800/11516] lr: 5.0500e-04 eta: 12:13:24 time: 0.4091 data_time: 0.0013 memory: 9465 grad_norm: 36.2547 loss: 0.6842 +2023/06/01 09:06:53 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/06/01 09:07:19 - mmengine - INFO - Epoch(train) [11][ 9900/11516] lr: 5.0500e-04 eta: 12:12:43 time: 0.4380 data_time: 0.0012 memory: 9465 grad_norm: 10.0476 loss: 0.6899 +2023/06/01 09:08:01 - mmengine - INFO - Epoch(train) [11][10000/11516] lr: 5.0500e-04 eta: 12:12:01 time: 0.4383 data_time: 0.0013 memory: 9465 grad_norm: 21.4826 loss: 0.6871 +2023/06/01 09:08:44 - mmengine - INFO - Epoch(train) [11][10100/11516] lr: 5.0500e-04 eta: 12:11:20 time: 0.4230 data_time: 0.0013 memory: 9465 grad_norm: 13.8483 loss: 0.6852 +2023/06/01 09:09:28 - mmengine - INFO - Epoch(train) [11][10200/11516] lr: 5.0500e-04 eta: 12:10:40 time: 0.4771 data_time: 0.0012 memory: 9465 grad_norm: 3.9802 loss: 0.6885 +2023/06/01 09:10:10 - mmengine - INFO - Epoch(train) [11][10300/11516] lr: 5.0500e-04 eta: 12:09:59 time: 0.4209 data_time: 0.0013 memory: 9465 grad_norm: 7.6920 loss: 0.6889 +2023/06/01 09:10:52 - mmengine - INFO - Epoch(train) [11][10400/11516] lr: 5.0500e-04 eta: 12:09:17 time: 0.4102 data_time: 0.0012 memory: 9465 grad_norm: 27.0678 loss: 0.6847 +2023/06/01 09:11:34 - mmengine - INFO - Epoch(train) [11][10500/11516] lr: 5.0500e-04 eta: 12:08:36 time: 0.4051 data_time: 0.0013 memory: 9465 grad_norm: 3.8574 loss: 0.6835 +2023/06/01 09:12:20 - mmengine - INFO - Epoch(train) [11][10600/11516] lr: 5.0500e-04 eta: 12:07:57 time: 0.4652 data_time: 0.0015 memory: 9465 grad_norm: 22.4843 loss: 0.6853 +2023/06/01 09:13:26 - mmengine - INFO - Epoch(train) [11][10700/11516] lr: 5.0500e-04 eta: 12:07:35 time: 0.6327 data_time: 0.0013 memory: 9465 grad_norm: 68.1242 loss: 0.6865 +2023/06/01 09:14:13 - mmengine - INFO - Epoch(train) [11][10800/11516] lr: 5.0500e-04 eta: 12:06:58 time: 0.3989 data_time: 0.0013 memory: 9465 grad_norm: 51.8566 loss: 0.6801 +2023/06/01 09:14:30 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/06/01 09:14:55 - mmengine - INFO - Epoch(train) [11][10900/11516] lr: 5.0500e-04 eta: 12:06:16 time: 0.4326 data_time: 0.0013 memory: 9465 grad_norm: 22.4680 loss: 0.6816 +2023/06/01 09:15:37 - mmengine - INFO - Epoch(train) [11][11000/11516] lr: 5.0500e-04 eta: 12:05:34 time: 0.4336 data_time: 0.0013 memory: 9465 grad_norm: 28.2552 loss: 0.6863 +2023/06/01 09:16:18 - mmengine - INFO - Epoch(train) [11][11100/11516] lr: 5.0500e-04 eta: 12:04:53 time: 0.4049 data_time: 0.0012 memory: 9465 grad_norm: 69.6691 loss: 0.6870 +2023/06/01 09:17:00 - mmengine - INFO - Epoch(train) [11][11200/11516] lr: 5.0500e-04 eta: 12:04:11 time: 0.4018 data_time: 0.0012 memory: 9465 grad_norm: 31.0741 loss: 0.6885 +2023/06/01 09:17:44 - mmengine - INFO - Epoch(train) [11][11300/11516] lr: 5.0500e-04 eta: 12:03:30 time: 0.4456 data_time: 0.0012 memory: 9465 grad_norm: 167.5287 loss: 0.6872 +2023/06/01 09:18:23 - mmengine - INFO - Epoch(train) [11][11400/11516] lr: 5.0500e-04 eta: 12:02:46 time: 0.4013 data_time: 0.0012 memory: 9465 grad_norm: 29.9626 loss: 0.6873 +2023/06/01 09:19:06 - mmengine - INFO - Epoch(train) [11][11500/11516] lr: 5.0500e-04 eta: 12:02:06 time: 0.4139 data_time: 0.0076 memory: 9465 grad_norm: 8.4570 loss: 0.6900 +2023/06/01 09:19:11 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/06/01 09:19:11 - mmengine - INFO - Saving checkpoint at 11 epochs +2023/06/01 09:19:28 - mmengine - INFO - Epoch(val) [11][16/16] accuracy/top1: 100.0000 single-label/precision_classwise: [100.0, 0.0] single-label/recall_classwise: [100.0, 0.0] single-label/f1-score_classwise: [100.0, 0.0] data_time: 0.3107 time: 0.5700 +2023/06/01 09:20:11 - mmengine - INFO - Epoch(train) [12][ 100/11516] lr: 4.2756e-04 eta: 12:01:18 time: 0.4361 data_time: 0.0767 memory: 9465 grad_norm: 4.3487 loss: 0.6898 +2023/06/01 09:20:53 - mmengine - INFO - Epoch(train) [12][ 200/11516] lr: 4.2756e-04 eta: 12:00:36 time: 0.4186 data_time: 0.0013 memory: 9465 grad_norm: 13.4253 loss: 0.6843 +2023/06/01 09:21:34 - mmengine - INFO - Epoch(train) [12][ 300/11516] lr: 4.2756e-04 eta: 11:59:53 time: 0.4027 data_time: 0.0014 memory: 9465 grad_norm: 9.8479 loss: 0.6790 +2023/06/01 09:21:43 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/06/01 09:22:17 - mmengine - INFO - Epoch(train) [12][ 400/11516] lr: 4.2756e-04 eta: 11:59:13 time: 0.5000 data_time: 0.0012 memory: 9465 grad_norm: 18.2187 loss: 0.6833 +2023/06/01 09:22:59 - mmengine - INFO - Epoch(train) [12][ 500/11516] lr: 4.2756e-04 eta: 11:58:31 time: 0.4102 data_time: 0.0013 memory: 9465 grad_norm: 13.6887 loss: 0.6831 +2023/06/01 09:23:40 - mmengine - INFO - Epoch(train) [12][ 600/11516] lr: 4.2756e-04 eta: 11:57:48 time: 0.4043 data_time: 0.0013 memory: 9465 grad_norm: 74.4274 loss: 0.6810 +2023/06/01 09:24:23 - mmengine - INFO - Epoch(train) [12][ 700/11516] lr: 4.2756e-04 eta: 11:57:08 time: 0.4117 data_time: 0.0013 memory: 9465 grad_norm: 188.3025 loss: 0.6754 +2023/06/01 09:25:06 - mmengine - INFO - Epoch(train) [12][ 800/11516] lr: 4.2756e-04 eta: 11:56:27 time: 0.4134 data_time: 0.0013 memory: 9465 grad_norm: 11.5787 loss: 0.6837 +2023/06/01 09:25:48 - mmengine - INFO - Epoch(train) [12][ 900/11516] lr: 4.2756e-04 eta: 11:55:45 time: 0.4208 data_time: 0.0014 memory: 9465 grad_norm: 1.4723 loss: 0.6792 +2023/06/01 09:26:30 - mmengine - INFO - Epoch(train) [12][ 1000/11516] lr: 4.2756e-04 eta: 11:55:03 time: 0.4394 data_time: 0.0013 memory: 9465 grad_norm: 3.5020 loss: 0.6809 +2023/06/01 09:27:12 - mmengine - INFO - Epoch(train) [12][ 1100/11516] lr: 4.2756e-04 eta: 11:54:22 time: 0.4618 data_time: 0.0015 memory: 9465 grad_norm: 10.5512 loss: 0.6819 +2023/06/01 09:27:55 - mmengine - INFO - Epoch(train) [12][ 1200/11516] lr: 4.2756e-04 eta: 11:53:41 time: 0.4523 data_time: 0.0012 memory: 9465 grad_norm: 20.9451 loss: 0.6804 +2023/06/01 09:28:37 - mmengine - INFO - Epoch(train) [12][ 1300/11516] lr: 4.2756e-04 eta: 11:52:59 time: 0.4211 data_time: 0.0013 memory: 9465 grad_norm: 14.1791 loss: 0.6758 +2023/06/01 09:28:47 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/06/01 09:29:21 - mmengine - INFO - Epoch(train) [12][ 1400/11516] lr: 4.2756e-04 eta: 11:52:19 time: 0.4092 data_time: 0.0013 memory: 9465 grad_norm: 3.3984 loss: 0.6759 +2023/06/01 09:30:02 - mmengine - INFO - Epoch(train) [12][ 1500/11516] lr: 4.2756e-04 eta: 11:51:37 time: 0.4041 data_time: 0.0013 memory: 9465 grad_norm: 0.8378 loss: 0.6752 +2023/06/01 09:30:44 - mmengine - INFO - Epoch(train) [12][ 1600/11516] lr: 4.2756e-04 eta: 11:50:55 time: 0.4167 data_time: 0.0014 memory: 9465 grad_norm: 2.9319 loss: 0.6733 +2023/06/01 09:31:26 - mmengine - INFO - Epoch(train) [12][ 1700/11516] lr: 4.2756e-04 eta: 11:50:13 time: 0.4201 data_time: 0.0013 memory: 9465 grad_norm: 3.4713 loss: 0.6816 +2023/06/01 09:32:08 - mmengine - INFO - Epoch(train) [12][ 1800/11516] lr: 4.2756e-04 eta: 11:49:31 time: 0.4277 data_time: 0.0012 memory: 9465 grad_norm: 196.7182 loss: 0.6683 +2023/06/01 09:32:58 - mmengine - INFO - Epoch(train) [12][ 1900/11516] lr: 4.2756e-04 eta: 11:48:56 time: 0.5494 data_time: 0.0014 memory: 9465 grad_norm: 15.3144 loss: 0.6784 +2023/06/01 09:33:45 - mmengine - INFO - Epoch(train) [12][ 2000/11516] lr: 4.2756e-04 eta: 11:48:19 time: 0.3984 data_time: 0.0014 memory: 9465 grad_norm: 1.7640 loss: 0.6709 +2023/06/01 09:34:27 - mmengine - INFO - Epoch(train) [12][ 2100/11516] lr: 4.2756e-04 eta: 11:47:37 time: 0.4447 data_time: 0.0013 memory: 9465 grad_norm: 3.4312 loss: 0.6763 +2023/06/01 09:35:09 - mmengine - INFO - Epoch(train) [12][ 2200/11516] lr: 4.2756e-04 eta: 11:46:55 time: 0.4263 data_time: 0.0014 memory: 9465 grad_norm: 2.7703 loss: 0.6715 +2023/06/01 09:35:51 - mmengine - INFO - Epoch(train) [12][ 2300/11516] lr: 4.2756e-04 eta: 11:46:14 time: 0.4114 data_time: 0.0015 memory: 9465 grad_norm: 3.4703 loss: 0.6754 +2023/06/01 09:36:00 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/06/01 09:36:34 - mmengine - INFO - Epoch(train) [12][ 2400/11516] lr: 4.2756e-04 eta: 11:45:32 time: 0.4061 data_time: 0.0015 memory: 9465 grad_norm: 1.2597 loss: 0.6692 +2023/06/01 09:37:16 - mmengine - INFO - Epoch(train) [12][ 2500/11516] lr: 4.2756e-04 eta: 11:44:51 time: 0.4293 data_time: 0.0014 memory: 9465 grad_norm: 2.1850 loss: 0.6867 +2023/06/01 09:37:57 - mmengine - INFO - Epoch(train) [12][ 2600/11516] lr: 4.2756e-04 eta: 11:44:09 time: 0.4091 data_time: 0.0013 memory: 9465 grad_norm: 0.8963 loss: 0.6872 +2023/06/01 09:38:39 - mmengine - INFO - Epoch(train) [12][ 2700/11516] lr: 4.2756e-04 eta: 11:43:27 time: 0.4141 data_time: 0.0013 memory: 9465 grad_norm: 2.5361 loss: 0.6819 +2023/06/01 09:39:21 - mmengine - INFO - Epoch(train) [12][ 2800/11516] lr: 4.2756e-04 eta: 11:42:45 time: 0.4410 data_time: 0.0014 memory: 9465 grad_norm: 0.0305 loss: 0.6869 +2023/06/01 09:40:03 - mmengine - INFO - Epoch(train) [12][ 2900/11516] lr: 4.2756e-04 eta: 11:42:03 time: 0.4109 data_time: 0.0013 memory: 9465 grad_norm: 0.0365 loss: 0.6903 +2023/06/01 09:40:45 - mmengine - INFO - Epoch(train) [12][ 3000/11516] lr: 4.2756e-04 eta: 11:41:21 time: 0.4688 data_time: 0.0014 memory: 9465 grad_norm: 0.0251 loss: 0.6882 +2023/06/01 09:41:26 - mmengine - INFO - Epoch(train) [12][ 3100/11516] lr: 4.2756e-04 eta: 11:40:40 time: 0.4340 data_time: 0.0014 memory: 9465 grad_norm: 0.0181 loss: 0.6847 +2023/06/01 09:42:09 - mmengine - INFO - Epoch(train) [12][ 3200/11516] lr: 4.2756e-04 eta: 11:39:58 time: 0.4438 data_time: 0.0013 memory: 9465 grad_norm: 0.0364 loss: 0.6900 +2023/06/01 09:42:51 - mmengine - INFO - Epoch(train) [12][ 3300/11516] lr: 4.2756e-04 eta: 11:39:17 time: 0.4202 data_time: 0.0015 memory: 9465 grad_norm: 0.0239 loss: 0.6896 +2023/06/01 09:43:00 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/06/01 09:43:34 - mmengine - INFO - Epoch(train) [12][ 3400/11516] lr: 4.2756e-04 eta: 11:38:35 time: 0.4413 data_time: 0.0014 memory: 9465 grad_norm: 0.0201 loss: 0.6873 +2023/06/01 09:44:15 - mmengine - INFO - Epoch(train) [12][ 3500/11516] lr: 4.2756e-04 eta: 11:37:53 time: 0.4054 data_time: 0.0013 memory: 9465 grad_norm: 0.0215 loss: 0.6870 +2023/06/01 09:44:57 - mmengine - INFO - Epoch(train) [12][ 3600/11516] lr: 4.2756e-04 eta: 11:37:12 time: 0.4108 data_time: 0.0014 memory: 9465 grad_norm: 0.0577 loss: 0.6855 +2023/06/01 09:45:39 - mmengine - INFO - Epoch(train) [12][ 3700/11516] lr: 4.2756e-04 eta: 11:36:30 time: 0.4158 data_time: 0.0013 memory: 9465 grad_norm: 0.0222 loss: 0.6871 +2023/06/01 09:46:21 - mmengine - INFO - Epoch(train) [12][ 3800/11516] lr: 4.2756e-04 eta: 11:35:48 time: 0.4546 data_time: 0.0013 memory: 9465 grad_norm: 0.0369 loss: 0.6891 +2023/06/01 09:47:03 - mmengine - INFO - Epoch(train) [12][ 3900/11516] lr: 4.2756e-04 eta: 11:35:06 time: 0.4138 data_time: 0.0013 memory: 9465 grad_norm: 0.0271 loss: 0.6815 +2023/06/01 09:47:44 - mmengine - INFO - Epoch(train) [12][ 4000/11516] lr: 4.2756e-04 eta: 11:34:24 time: 0.3864 data_time: 0.0013 memory: 9465 grad_norm: 0.0271 loss: 0.6850 +2023/06/01 09:48:27 - mmengine - INFO - Epoch(train) [12][ 4100/11516] lr: 4.2756e-04 eta: 11:33:43 time: 0.4003 data_time: 0.0017 memory: 9465 grad_norm: 0.0320 loss: 0.6858 +2023/06/01 09:49:09 - mmengine - INFO - Epoch(train) [12][ 4200/11516] lr: 4.2756e-04 eta: 11:33:01 time: 0.4527 data_time: 0.0015 memory: 9465 grad_norm: 0.0259 loss: 0.6855 +2023/06/01 09:49:51 - mmengine - INFO - Epoch(train) [12][ 4300/11516] lr: 4.2756e-04 eta: 11:32:20 time: 0.4470 data_time: 0.0013 memory: 9465 grad_norm: 0.0213 loss: 0.6853 +2023/06/01 09:50:00 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/06/01 09:50:33 - mmengine - INFO - Epoch(train) [12][ 4400/11516] lr: 4.2756e-04 eta: 11:31:38 time: 0.4411 data_time: 0.0015 memory: 9465 grad_norm: 0.0218 loss: 0.6891 +2023/06/01 09:51:15 - mmengine - INFO - Epoch(train) [12][ 4500/11516] lr: 4.2756e-04 eta: 11:30:56 time: 0.4067 data_time: 0.0015 memory: 9465 grad_norm: 0.0193 loss: 0.6872 +2023/06/01 09:51:57 - mmengine - INFO - Epoch(train) [12][ 4600/11516] lr: 4.2756e-04 eta: 11:30:15 time: 0.4118 data_time: 0.0015 memory: 9465 grad_norm: 0.0245 loss: 0.6823 +2023/06/01 09:52:38 - mmengine - INFO - Epoch(train) [12][ 4700/11516] lr: 4.2756e-04 eta: 11:29:33 time: 0.3782 data_time: 0.0014 memory: 9465 grad_norm: 0.0247 loss: 0.6912 +2023/06/01 09:53:20 - mmengine - INFO - Epoch(train) [12][ 4800/11516] lr: 4.2756e-04 eta: 11:28:51 time: 0.4211 data_time: 0.0014 memory: 9465 grad_norm: 0.0231 loss: 0.6890 +2023/06/01 09:54:04 - mmengine - INFO - Epoch(train) [12][ 4900/11516] lr: 4.2756e-04 eta: 11:28:11 time: 0.4405 data_time: 0.0014 memory: 9465 grad_norm: 0.0197 loss: 0.6872 +2023/06/01 09:54:46 - mmengine - INFO - Epoch(train) [12][ 5000/11516] lr: 4.2756e-04 eta: 11:27:29 time: 0.4131 data_time: 0.0013 memory: 9465 grad_norm: 0.0220 loss: 0.6893 +2023/06/01 09:55:29 - mmengine - INFO - Epoch(train) [12][ 5100/11516] lr: 4.2756e-04 eta: 11:26:48 time: 0.4508 data_time: 0.0014 memory: 9465 grad_norm: 0.0172 loss: 0.6846 +2023/06/01 09:56:10 - mmengine - INFO - Epoch(train) [12][ 5200/11516] lr: 4.2756e-04 eta: 11:26:06 time: 0.4022 data_time: 0.0013 memory: 9465 grad_norm: 0.0187 loss: 0.6892 +2023/06/01 09:56:52 - mmengine - INFO - Epoch(train) [12][ 5300/11516] lr: 4.2756e-04 eta: 11:25:24 time: 0.3866 data_time: 0.0014 memory: 9465 grad_norm: 0.0254 loss: 0.6908 +2023/06/01 09:57:02 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/06/01 09:57:35 - mmengine - INFO - Epoch(train) [12][ 5400/11516] lr: 4.2756e-04 eta: 11:24:43 time: 0.4221 data_time: 0.0014 memory: 9465 grad_norm: 0.0184 loss: 0.6832 +2023/06/01 09:58:18 - mmengine - INFO - Epoch(train) [12][ 5500/11516] lr: 4.2756e-04 eta: 11:24:02 time: 0.4193 data_time: 0.0017 memory: 9465 grad_norm: 0.0193 loss: 0.6892 +2023/06/01 09:59:03 - mmengine - INFO - Epoch(train) [12][ 5600/11516] lr: 4.2756e-04 eta: 11:23:22 time: 0.4040 data_time: 0.0015 memory: 9465 grad_norm: 0.0219 loss: 0.6881 +2023/06/01 09:59:45 - mmengine - INFO - Epoch(train) [12][ 5700/11516] lr: 4.2756e-04 eta: 11:22:41 time: 0.4176 data_time: 0.0015 memory: 9465 grad_norm: 0.0205 loss: 0.6850 +2023/06/01 10:00:27 - mmengine - INFO - Epoch(train) [12][ 5800/11516] lr: 4.2756e-04 eta: 11:21:59 time: 0.4125 data_time: 0.0013 memory: 9465 grad_norm: 0.0196 loss: 0.6870 +2023/06/01 10:01:10 - mmengine - INFO - Epoch(train) [12][ 5900/11516] lr: 4.2756e-04 eta: 11:21:18 time: 0.4463 data_time: 0.0014 memory: 9465 grad_norm: 0.0192 loss: 0.6884 +2023/06/01 10:01:53 - mmengine - INFO - Epoch(train) [12][ 6000/11516] lr: 4.2756e-04 eta: 11:20:37 time: 0.4384 data_time: 0.0014 memory: 9465 grad_norm: 0.0251 loss: 0.6884 +2023/06/01 10:02:36 - mmengine - INFO - Epoch(train) [12][ 6100/11516] lr: 4.2756e-04 eta: 11:19:56 time: 0.4235 data_time: 0.0011 memory: 9465 grad_norm: 0.0164 loss: 0.6856 +2023/06/01 10:03:19 - mmengine - INFO - Epoch(train) [12][ 6200/11516] lr: 4.2756e-04 eta: 11:19:15 time: 0.4073 data_time: 0.0014 memory: 9465 grad_norm: 0.0235 loss: 0.6875 +2023/06/01 10:04:01 - mmengine - INFO - Epoch(train) [12][ 6300/11516] lr: 4.2756e-04 eta: 11:18:33 time: 0.4270 data_time: 0.0013 memory: 9465 grad_norm: 0.0233 loss: 0.6820 +2023/06/01 10:04:10 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/06/01 10:04:43 - mmengine - INFO - Epoch(train) [12][ 6400/11516] lr: 4.2756e-04 eta: 11:17:52 time: 0.4213 data_time: 0.0012 memory: 9465 grad_norm: 0.0248 loss: 0.6920 +2023/06/01 10:05:26 - mmengine - INFO - Epoch(train) [12][ 6500/11516] lr: 4.2756e-04 eta: 11:17:11 time: 0.4296 data_time: 0.0013 memory: 9465 grad_norm: 0.0338 loss: 0.6895 +2023/06/01 10:06:08 - mmengine - INFO - Epoch(train) [12][ 6600/11516] lr: 4.2756e-04 eta: 11:16:29 time: 0.4017 data_time: 0.0013 memory: 9465 grad_norm: 0.0191 loss: 0.6906 +2023/06/01 10:06:50 - mmengine - INFO - Epoch(train) [12][ 6700/11516] lr: 4.2756e-04 eta: 11:15:47 time: 0.4289 data_time: 0.0017 memory: 9465 grad_norm: 0.0199 loss: 0.6894 +2023/06/01 10:07:33 - mmengine - INFO - Epoch(train) [12][ 6800/11516] lr: 4.2756e-04 eta: 11:15:06 time: 0.4464 data_time: 0.0014 memory: 9465 grad_norm: 0.0243 loss: 0.6894 +2023/06/01 10:08:17 - mmengine - INFO - Epoch(train) [12][ 6900/11516] lr: 4.2756e-04 eta: 11:14:26 time: 0.4564 data_time: 0.0014 memory: 9465 grad_norm: 0.0132 loss: 0.6843 +2023/06/01 10:08:59 - mmengine - INFO - Epoch(train) [12][ 7000/11516] lr: 4.2756e-04 eta: 11:13:44 time: 0.4240 data_time: 0.0013 memory: 9465 grad_norm: 0.0226 loss: 0.6804 +2023/06/01 10:09:43 - mmengine - INFO - Epoch(train) [12][ 7100/11516] lr: 4.2756e-04 eta: 11:13:04 time: 0.4044 data_time: 0.0014 memory: 9465 grad_norm: 0.0220 loss: 0.6857 +2023/06/01 10:10:26 - mmengine - INFO - Epoch(train) [12][ 7200/11516] lr: 4.2756e-04 eta: 11:12:23 time: 0.4111 data_time: 0.0016 memory: 9465 grad_norm: 0.0184 loss: 0.6870 +2023/06/01 10:11:10 - mmengine - INFO - Epoch(train) [12][ 7300/11516] lr: 4.2756e-04 eta: 11:11:42 time: 0.4070 data_time: 0.0013 memory: 9465 grad_norm: 0.0303 loss: 0.6862 +2023/06/01 10:11:19 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/06/01 10:11:51 - mmengine - INFO - Epoch(train) [12][ 7400/11516] lr: 4.2756e-04 eta: 11:11:00 time: 0.3955 data_time: 0.0012 memory: 9465 grad_norm: 0.0246 loss: 0.6904 +2023/06/01 10:12:34 - mmengine - INFO - Epoch(train) [12][ 7500/11516] lr: 4.2756e-04 eta: 11:10:19 time: 0.4045 data_time: 0.0013 memory: 9465 grad_norm: 0.0290 loss: 0.6837 +2023/06/01 10:13:17 - mmengine - INFO - Epoch(train) [12][ 7600/11516] lr: 4.2756e-04 eta: 11:09:38 time: 0.4229 data_time: 0.0013 memory: 9465 grad_norm: 0.0201 loss: 0.6868 +2023/06/01 10:14:00 - mmengine - INFO - Epoch(train) [12][ 7700/11516] lr: 4.2756e-04 eta: 11:08:57 time: 0.4216 data_time: 0.0014 memory: 9465 grad_norm: 0.0248 loss: 0.6843 +2023/06/01 10:14:43 - mmengine - INFO - Epoch(train) [12][ 7800/11516] lr: 4.2756e-04 eta: 11:08:16 time: 0.4240 data_time: 0.0013 memory: 9465 grad_norm: 0.0151 loss: 0.6871 +2023/06/01 10:15:25 - mmengine - INFO - Epoch(train) [12][ 7900/11516] lr: 4.2756e-04 eta: 11:07:35 time: 0.4838 data_time: 0.0015 memory: 9465 grad_norm: 0.0228 loss: 0.6919 +2023/06/01 10:16:08 - mmengine - INFO - Epoch(train) [12][ 8000/11516] lr: 4.2756e-04 eta: 11:06:53 time: 0.4338 data_time: 0.0013 memory: 9465 grad_norm: 0.0297 loss: 0.6816 +2023/06/01 10:16:50 - mmengine - INFO - Epoch(train) [12][ 8100/11516] lr: 4.2756e-04 eta: 11:06:12 time: 0.4061 data_time: 0.0013 memory: 9465 grad_norm: 0.0199 loss: 0.6860 +2023/06/01 10:17:33 - mmengine - INFO - Epoch(train) [12][ 8200/11516] lr: 4.2756e-04 eta: 11:05:31 time: 0.4025 data_time: 0.0013 memory: 9465 grad_norm: 0.0192 loss: 0.6852 +2023/06/01 10:18:17 - mmengine - INFO - Epoch(train) [12][ 8300/11516] lr: 4.2756e-04 eta: 11:04:50 time: 0.4202 data_time: 0.0014 memory: 9465 grad_norm: 0.0240 loss: 0.6818 +2023/06/01 10:18:26 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/06/01 10:18:59 - mmengine - INFO - Epoch(train) [12][ 8400/11516] lr: 4.2756e-04 eta: 11:04:08 time: 0.4085 data_time: 0.0014 memory: 9465 grad_norm: 0.0286 loss: 0.6868 +2023/06/01 10:19:41 - mmengine - INFO - Epoch(train) [12][ 8500/11516] lr: 4.2756e-04 eta: 11:03:27 time: 0.4077 data_time: 0.0015 memory: 9465 grad_norm: 0.0167 loss: 0.6867 +2023/06/01 10:20:23 - mmengine - INFO - Epoch(train) [12][ 8600/11516] lr: 4.2756e-04 eta: 11:02:45 time: 0.4268 data_time: 0.0015 memory: 9465 grad_norm: 0.0158 loss: 0.6860 +2023/06/01 10:21:06 - mmengine - INFO - Epoch(train) [12][ 8700/11516] lr: 4.2756e-04 eta: 11:02:04 time: 0.4038 data_time: 0.0013 memory: 9465 grad_norm: 0.0292 loss: 0.6862 +2023/06/01 10:21:48 - mmengine - INFO - Epoch(train) [12][ 8800/11516] lr: 4.2756e-04 eta: 11:01:23 time: 0.4109 data_time: 0.0014 memory: 9465 grad_norm: 0.0203 loss: 0.6876 +2023/06/01 10:22:30 - mmengine - INFO - Epoch(train) [12][ 8900/11516] lr: 4.2756e-04 eta: 11:00:41 time: 0.4106 data_time: 0.0014 memory: 9465 grad_norm: 0.0234 loss: 0.6904 +2023/06/01 10:23:11 - mmengine - INFO - Epoch(train) [12][ 9000/11516] lr: 4.2756e-04 eta: 10:59:58 time: 0.4114 data_time: 0.0015 memory: 9465 grad_norm: 0.0180 loss: 0.6851 +2023/06/01 10:23:54 - mmengine - INFO - Epoch(train) [12][ 9100/11516] lr: 4.2756e-04 eta: 10:59:17 time: 0.4287 data_time: 0.0015 memory: 9465 grad_norm: 0.0263 loss: 0.6874 +2023/06/01 10:24:37 - mmengine - INFO - Epoch(train) [12][ 9200/11516] lr: 4.2756e-04 eta: 10:58:36 time: 0.4189 data_time: 0.0013 memory: 9465 grad_norm: 0.0122 loss: 0.6862 +2023/06/01 10:25:20 - mmengine - INFO - Epoch(train) [12][ 9300/11516] lr: 4.2756e-04 eta: 10:57:55 time: 0.4316 data_time: 0.0012 memory: 9465 grad_norm: 0.0264 loss: 0.6858 +2023/06/01 10:25:29 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/06/01 10:26:02 - mmengine - INFO - Epoch(train) [12][ 9400/11516] lr: 4.2756e-04 eta: 10:57:14 time: 0.4313 data_time: 0.0012 memory: 9465 grad_norm: 0.0246 loss: 0.6861 +2023/06/01 10:26:46 - mmengine - INFO - Epoch(train) [12][ 9500/11516] lr: 4.2756e-04 eta: 10:56:33 time: 0.4810 data_time: 0.0015 memory: 9465 grad_norm: 0.0183 loss: 0.6849 +2023/06/01 10:27:28 - mmengine - INFO - Epoch(train) [12][ 9600/11516] lr: 4.2756e-04 eta: 10:55:51 time: 0.4267 data_time: 0.0013 memory: 9465 grad_norm: 0.0192 loss: 0.6901 +2023/06/01 10:28:11 - mmengine - INFO - Epoch(train) [12][ 9700/11516] lr: 4.2756e-04 eta: 10:55:10 time: 0.4434 data_time: 0.0017 memory: 9465 grad_norm: 0.0174 loss: 0.6901 +2023/06/01 10:28:54 - mmengine - INFO - Epoch(train) [12][ 9800/11516] lr: 4.2756e-04 eta: 10:54:29 time: 0.4468 data_time: 0.0014 memory: 9465 grad_norm: 0.0205 loss: 0.6884 +2023/06/01 10:29:36 - mmengine - INFO - Epoch(train) [12][ 9900/11516] lr: 4.2756e-04 eta: 10:53:48 time: 0.3911 data_time: 0.0013 memory: 9465 grad_norm: 0.0183 loss: 0.6854 +2023/06/01 10:30:19 - mmengine - INFO - Epoch(train) [12][10000/11516] lr: 4.2756e-04 eta: 10:53:06 time: 0.4592 data_time: 0.0014 memory: 9465 grad_norm: 0.0123 loss: 0.6868 +2023/06/01 10:31:02 - mmengine - INFO - Epoch(train) [12][10100/11516] lr: 4.2756e-04 eta: 10:52:25 time: 0.4549 data_time: 0.0013 memory: 9465 grad_norm: 0.0220 loss: 0.6875 +2023/06/01 10:31:44 - mmengine - INFO - Epoch(train) [12][10200/11516] lr: 4.2756e-04 eta: 10:51:44 time: 0.4544 data_time: 0.0013 memory: 9465 grad_norm: 0.0200 loss: 0.6920 +2023/06/01 10:32:27 - mmengine - INFO - Epoch(train) [12][10300/11516] lr: 4.2756e-04 eta: 10:51:02 time: 0.4361 data_time: 0.0014 memory: 9465 grad_norm: 0.0225 loss: 0.6889 +2023/06/01 10:32:37 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/06/01 10:33:10 - mmengine - INFO - Epoch(train) [12][10400/11516] lr: 4.2756e-04 eta: 10:50:21 time: 0.4119 data_time: 0.0013 memory: 9465 grad_norm: 0.0249 loss: 0.6891 +2023/06/01 10:33:53 - mmengine - INFO - Epoch(train) [12][10500/11516] lr: 4.2756e-04 eta: 10:49:40 time: 0.4000 data_time: 0.0020 memory: 9465 grad_norm: 0.0205 loss: 0.6907 +2023/06/01 10:34:36 - mmengine - INFO - Epoch(train) [12][10600/11516] lr: 4.2756e-04 eta: 10:48:59 time: 0.4357 data_time: 0.0014 memory: 9465 grad_norm: 0.0216 loss: 0.6905 +2023/06/01 10:35:19 - mmengine - INFO - Epoch(train) [12][10700/11516] lr: 4.2756e-04 eta: 10:48:18 time: 0.4319 data_time: 0.0013 memory: 9465 grad_norm: 0.0227 loss: 0.6857 +2023/06/01 10:36:01 - mmengine - INFO - Epoch(train) [12][10800/11516] lr: 4.2756e-04 eta: 10:47:36 time: 0.4310 data_time: 0.0014 memory: 9465 grad_norm: 0.0257 loss: 0.6850 +2023/06/01 10:36:44 - mmengine - INFO - Epoch(train) [12][10900/11516] lr: 4.2756e-04 eta: 10:46:55 time: 0.4399 data_time: 0.0013 memory: 9465 grad_norm: 0.0264 loss: 0.6828 +2023/06/01 10:37:26 - mmengine - INFO - Epoch(train) [12][11000/11516] lr: 4.2756e-04 eta: 10:46:13 time: 0.4175 data_time: 0.0013 memory: 9465 grad_norm: 0.0208 loss: 0.6891 +2023/06/01 10:38:10 - mmengine - INFO - Epoch(train) [12][11100/11516] lr: 4.2756e-04 eta: 10:45:33 time: 0.4109 data_time: 0.0013 memory: 9465 grad_norm: 0.0298 loss: 0.6874 +2023/06/01 10:38:53 - mmengine - INFO - Epoch(train) [12][11200/11516] lr: 4.2756e-04 eta: 10:44:52 time: 0.4285 data_time: 0.0014 memory: 9465 grad_norm: 0.0206 loss: 0.6829 +2023/06/01 10:39:35 - mmengine - INFO - Epoch(train) [12][11300/11516] lr: 4.2756e-04 eta: 10:44:10 time: 0.4436 data_time: 0.0012 memory: 9465 grad_norm: 0.0320 loss: 0.6877 +2023/06/01 10:39:44 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/06/01 10:40:15 - mmengine - INFO - Epoch(train) [12][11400/11516] lr: 4.2756e-04 eta: 10:43:27 time: 0.3867 data_time: 0.0023 memory: 9465 grad_norm: 0.0375 loss: 0.6899 +2023/06/01 10:40:58 - mmengine - INFO - Epoch(train) [12][11500/11516] lr: 4.2756e-04 eta: 10:42:46 time: 0.4042 data_time: 0.0079 memory: 9465 grad_norm: 0.0237 loss: 0.6854 +2023/06/01 10:41:03 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/06/01 10:41:03 - mmengine - INFO - Saving checkpoint at 12 epochs +2023/06/01 10:41:19 - mmengine - INFO - Epoch(val) [12][16/16] accuracy/top1: 100.0000 single-label/precision_classwise: [100.0, 0.0] single-label/recall_classwise: [100.0, 0.0] single-label/f1-score_classwise: [100.0, 0.0] data_time: 0.2980 time: 0.5517 +2023/06/01 10:42:05 - mmengine - INFO - Epoch(train) [13][ 100/11516] lr: 3.5204e-04 eta: 10:41:58 time: 0.4686 data_time: 0.1873 memory: 9465 grad_norm: 0.0175 loss: 0.6881 +2023/06/01 10:42:46 - mmengine - INFO - Epoch(train) [13][ 200/11516] lr: 3.5204e-04 eta: 10:41:16 time: 0.4445 data_time: 0.0012 memory: 9465 grad_norm: 0.0267 loss: 0.6868 +2023/06/01 10:43:28 - mmengine - INFO - Epoch(train) [13][ 300/11516] lr: 3.5204e-04 eta: 10:40:35 time: 0.4294 data_time: 0.0012 memory: 9465 grad_norm: 0.0276 loss: 0.6891 +2023/06/01 10:44:11 - mmengine - INFO - Epoch(train) [13][ 400/11516] lr: 3.5204e-04 eta: 10:39:53 time: 0.4326 data_time: 0.0013 memory: 9465 grad_norm: 0.0366 loss: 0.6862 +2023/06/01 10:44:54 - mmengine - INFO - Epoch(train) [13][ 500/11516] lr: 3.5204e-04 eta: 10:39:12 time: 0.4582 data_time: 0.0012 memory: 9465 grad_norm: 0.0257 loss: 0.6926 +2023/06/01 10:45:36 - mmengine - INFO - Epoch(train) [13][ 600/11516] lr: 3.5204e-04 eta: 10:38:30 time: 0.4244 data_time: 0.0014 memory: 9465 grad_norm: 0.0226 loss: 0.6889 +2023/06/01 10:46:18 - mmengine - INFO - Epoch(train) [13][ 700/11516] lr: 3.5204e-04 eta: 10:37:49 time: 0.4073 data_time: 0.0013 memory: 9465 grad_norm: 0.0218 loss: 0.6898 +2023/06/01 10:46:59 - mmengine - INFO - Epoch(train) [13][ 800/11516] lr: 3.5204e-04 eta: 10:37:07 time: 0.4200 data_time: 0.0014 memory: 9465 grad_norm: 0.0138 loss: 0.6899 +2023/06/01 10:47:01 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/06/01 10:47:44 - mmengine - INFO - Epoch(train) [13][ 900/11516] lr: 3.5204e-04 eta: 10:36:26 time: 0.4337 data_time: 0.0015 memory: 9465 grad_norm: 0.0177 loss: 0.6862 +2023/06/01 10:48:26 - mmengine - INFO - Epoch(train) [13][ 1000/11516] lr: 3.5204e-04 eta: 10:35:45 time: 0.4152 data_time: 0.0012 memory: 9465 grad_norm: 0.0239 loss: 0.6860 +2023/06/01 10:49:10 - mmengine - INFO - Epoch(train) [13][ 1100/11516] lr: 3.5204e-04 eta: 10:35:04 time: 0.4187 data_time: 0.0012 memory: 9465 grad_norm: 0.0190 loss: 0.6848 +2023/06/01 10:49:53 - mmengine - INFO - Epoch(train) [13][ 1200/11516] lr: 3.5204e-04 eta: 10:34:23 time: 0.4419 data_time: 0.0013 memory: 9465 grad_norm: 0.0168 loss: 0.6843 +2023/06/01 10:50:35 - mmengine - INFO - Epoch(train) [13][ 1300/11516] lr: 3.5204e-04 eta: 10:33:41 time: 0.4503 data_time: 0.0013 memory: 9465 grad_norm: 0.0158 loss: 0.6855 +2023/06/01 10:51:18 - mmengine - INFO - Epoch(train) [13][ 1400/11516] lr: 3.5204e-04 eta: 10:33:00 time: 0.3866 data_time: 0.0014 memory: 9465 grad_norm: 0.0308 loss: 0.6858 +2023/06/01 10:52:01 - mmengine - INFO - Epoch(train) [13][ 1500/11516] lr: 3.5204e-04 eta: 10:32:19 time: 0.4436 data_time: 0.0014 memory: 9465 grad_norm: 0.0132 loss: 0.6867 +2023/06/01 10:52:43 - mmengine - INFO - Epoch(train) [13][ 1600/11516] lr: 3.5204e-04 eta: 10:31:38 time: 0.4172 data_time: 0.0012 memory: 9465 grad_norm: 0.0160 loss: 0.6891 +2023/06/01 10:53:27 - mmengine - INFO - Epoch(train) [13][ 1700/11516] lr: 3.5204e-04 eta: 10:30:57 time: 0.4244 data_time: 0.0013 memory: 9465 grad_norm: 0.0223 loss: 0.6881 +2023/06/01 10:54:09 - mmengine - INFO - Epoch(train) [13][ 1800/11516] lr: 3.5204e-04 eta: 10:30:15 time: 0.4387 data_time: 0.0013 memory: 9465 grad_norm: 0.0323 loss: 0.6855 +2023/06/01 10:54:11 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/06/01 10:54:53 - mmengine - INFO - Epoch(train) [13][ 1900/11516] lr: 3.5204e-04 eta: 10:29:34 time: 0.4207 data_time: 0.0013 memory: 9465 grad_norm: 0.0273 loss: 0.6826 +2023/06/01 10:55:35 - mmengine - INFO - Epoch(train) [13][ 2000/11516] lr: 3.5204e-04 eta: 10:28:53 time: 0.4150 data_time: 0.0014 memory: 9465 grad_norm: 0.0221 loss: 0.6878 +2023/06/01 10:56:17 - mmengine - INFO - Epoch(train) [13][ 2100/11516] lr: 3.5204e-04 eta: 10:28:11 time: 0.3960 data_time: 0.0014 memory: 9465 grad_norm: 0.0296 loss: 0.6839 +2023/06/01 10:57:01 - mmengine - INFO - Epoch(train) [13][ 2200/11516] lr: 3.5204e-04 eta: 10:27:30 time: 0.4117 data_time: 0.0013 memory: 9465 grad_norm: 0.0265 loss: 0.6792 +2023/06/01 10:57:44 - mmengine - INFO - Epoch(train) [13][ 2300/11516] lr: 3.5204e-04 eta: 10:26:49 time: 0.4378 data_time: 0.0015 memory: 9465 grad_norm: 0.0209 loss: 0.6864 +2023/06/01 10:58:27 - mmengine - INFO - Epoch(train) [13][ 2400/11516] lr: 3.5204e-04 eta: 10:26:08 time: 0.4376 data_time: 0.0014 memory: 9465 grad_norm: 0.0342 loss: 0.6850 +2023/06/01 10:59:10 - mmengine - INFO - Epoch(train) [13][ 2500/11516] lr: 3.5204e-04 eta: 10:25:27 time: 0.4034 data_time: 0.0013 memory: 9465 grad_norm: 0.0185 loss: 0.6865 +2023/06/01 10:59:53 - mmengine - INFO - Epoch(train) [13][ 2600/11516] lr: 3.5204e-04 eta: 10:24:45 time: 0.4225 data_time: 0.0014 memory: 9465 grad_norm: 0.0232 loss: 0.6868 +2023/06/01 11:00:36 - mmengine - INFO - Epoch(train) [13][ 2700/11516] lr: 3.5204e-04 eta: 10:24:04 time: 0.4352 data_time: 0.0014 memory: 9465 grad_norm: 0.0200 loss: 0.6862 +2023/06/01 11:01:18 - mmengine - INFO - Epoch(train) [13][ 2800/11516] lr: 3.5204e-04 eta: 10:23:23 time: 0.4319 data_time: 0.0014 memory: 9465 grad_norm: 0.0229 loss: 0.6920 +2023/06/01 11:01:20 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/06/01 11:02:01 - mmengine - INFO - Epoch(train) [13][ 2900/11516] lr: 3.5204e-04 eta: 10:22:41 time: 0.4230 data_time: 0.0016 memory: 9465 grad_norm: 0.0240 loss: 0.6870 +2023/06/01 11:02:43 - mmengine - INFO - Epoch(train) [13][ 3000/11516] lr: 3.5204e-04 eta: 10:22:00 time: 0.4229 data_time: 0.0013 memory: 9465 grad_norm: 0.0246 loss: 0.6892 +2023/06/01 11:03:26 - mmengine - INFO - Epoch(train) [13][ 3100/11516] lr: 3.5204e-04 eta: 10:21:19 time: 0.4393 data_time: 0.0014 memory: 9465 grad_norm: 0.0233 loss: 0.6876 +2023/06/01 11:04:09 - mmengine - INFO - Epoch(train) [13][ 3200/11516] lr: 3.5204e-04 eta: 10:20:37 time: 0.4276 data_time: 0.0014 memory: 9465 grad_norm: 0.0126 loss: 0.6855 +2023/06/01 11:04:52 - mmengine - INFO - Epoch(train) [13][ 3300/11516] lr: 3.5204e-04 eta: 10:19:56 time: 0.4132 data_time: 0.0015 memory: 9465 grad_norm: 0.0153 loss: 0.6881 +2023/06/01 11:05:35 - mmengine - INFO - Epoch(train) [13][ 3400/11516] lr: 3.5204e-04 eta: 10:19:15 time: 0.3968 data_time: 0.0013 memory: 9465 grad_norm: 0.0193 loss: 0.6860 +2023/06/01 11:06:17 - mmengine - INFO - Epoch(train) [13][ 3500/11516] lr: 3.5204e-04 eta: 10:18:33 time: 0.4250 data_time: 0.0015 memory: 9465 grad_norm: 0.0181 loss: 0.6886 +2023/06/01 11:06:59 - mmengine - INFO - Epoch(train) [13][ 3600/11516] lr: 3.5204e-04 eta: 10:17:52 time: 0.4183 data_time: 0.0013 memory: 9465 grad_norm: 0.0222 loss: 0.6866 +2023/06/01 11:07:42 - mmengine - INFO - Epoch(train) [13][ 3700/11516] lr: 3.5204e-04 eta: 10:17:10 time: 0.4504 data_time: 0.0013 memory: 9465 grad_norm: 0.0224 loss: 0.6873 +2023/06/01 11:08:25 - mmengine - INFO - Epoch(train) [13][ 3800/11516] lr: 3.5204e-04 eta: 10:16:29 time: 0.4324 data_time: 0.0013 memory: 9465 grad_norm: 0.0194 loss: 0.6879 +2023/06/01 11:08:27 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/06/01 11:09:09 - mmengine - INFO - Epoch(train) [13][ 3900/11516] lr: 3.5204e-04 eta: 10:15:48 time: 0.4312 data_time: 0.0013 memory: 9465 grad_norm: 0.0169 loss: 0.6885 +2023/06/01 11:09:52 - mmengine - INFO - Epoch(train) [13][ 4000/11516] lr: 3.5204e-04 eta: 10:15:08 time: 0.4292 data_time: 0.0014 memory: 9465 grad_norm: 0.0196 loss: 0.6895 +2023/06/01 11:10:37 - mmengine - INFO - Epoch(train) [13][ 4100/11516] lr: 3.5204e-04 eta: 10:14:28 time: 0.4558 data_time: 0.0013 memory: 9465 grad_norm: 0.0220 loss: 0.6823 +2023/06/01 11:11:21 - mmengine - INFO - Epoch(train) [13][ 4200/11516] lr: 3.5204e-04 eta: 10:13:47 time: 0.4304 data_time: 0.0014 memory: 9465 grad_norm: 0.0288 loss: 0.6839 +2023/06/01 11:12:05 - mmengine - INFO - Epoch(train) [13][ 4300/11516] lr: 3.5204e-04 eta: 10:13:06 time: 0.4542 data_time: 0.0013 memory: 9465 grad_norm: 0.0306 loss: 0.6900 +2023/06/01 11:12:48 - mmengine - INFO - Epoch(train) [13][ 4400/11516] lr: 3.5204e-04 eta: 10:12:25 time: 0.4223 data_time: 0.0013 memory: 9465 grad_norm: 0.0179 loss: 0.6853 +2023/06/01 11:13:31 - mmengine - INFO - Epoch(train) [13][ 4500/11516] lr: 3.5204e-04 eta: 10:11:44 time: 0.4351 data_time: 0.0014 memory: 9465 grad_norm: 0.0099 loss: 0.6896 +2023/06/01 11:14:16 - mmengine - INFO - Epoch(train) [13][ 4600/11516] lr: 3.5204e-04 eta: 10:11:04 time: 0.4540 data_time: 0.0013 memory: 9465 grad_norm: 0.0337 loss: 0.6935 +2023/06/01 11:15:02 - mmengine - INFO - Epoch(train) [13][ 4700/11516] lr: 3.5204e-04 eta: 10:10:25 time: 0.4531 data_time: 0.0013 memory: 9465 grad_norm: 0.0271 loss: 0.6876 +2023/06/01 11:15:46 - mmengine - INFO - Epoch(train) [13][ 4800/11516] lr: 3.5204e-04 eta: 10:09:44 time: 0.4386 data_time: 0.0015 memory: 9465 grad_norm: 0.0199 loss: 0.6860 +2023/06/01 11:15:48 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/06/01 11:16:30 - mmengine - INFO - Epoch(train) [13][ 4900/11516] lr: 3.5204e-04 eta: 10:09:03 time: 0.4342 data_time: 0.0012 memory: 9465 grad_norm: 0.0337 loss: 0.6895 +2023/06/01 11:17:14 - mmengine - INFO - Epoch(train) [13][ 5000/11516] lr: 3.5204e-04 eta: 10:08:22 time: 0.4190 data_time: 0.0013 memory: 9465 grad_norm: 0.0267 loss: 0.6910 +2023/06/01 11:17:57 - mmengine - INFO - Epoch(train) [13][ 5100/11516] lr: 3.5204e-04 eta: 10:07:41 time: 0.4054 data_time: 0.0013 memory: 9465 grad_norm: 0.0163 loss: 0.6833 +2023/06/01 11:18:40 - mmengine - INFO - Epoch(train) [13][ 5200/11516] lr: 3.5204e-04 eta: 10:07:00 time: 0.4397 data_time: 0.0014 memory: 9465 grad_norm: 0.0261 loss: 0.6857 +2023/06/01 11:19:24 - mmengine - INFO - Epoch(train) [13][ 5300/11516] lr: 3.5204e-04 eta: 10:06:19 time: 0.4256 data_time: 0.0013 memory: 9465 grad_norm: 0.0283 loss: 0.6879 +2023/06/01 11:20:06 - mmengine - INFO - Epoch(train) [13][ 5400/11516] lr: 3.5204e-04 eta: 10:05:37 time: 0.4452 data_time: 0.0013 memory: 9465 grad_norm: 0.0195 loss: 0.6839 +2023/06/01 11:20:48 - mmengine - INFO - Epoch(train) [13][ 5500/11516] lr: 3.5204e-04 eta: 10:04:56 time: 0.4636 data_time: 0.0012 memory: 9465 grad_norm: 0.0212 loss: 0.6844 +2023/06/01 11:21:32 - mmengine - INFO - Epoch(train) [13][ 5600/11516] lr: 3.5204e-04 eta: 10:04:15 time: 0.4363 data_time: 0.0012 memory: 9465 grad_norm: 0.0236 loss: 0.6870 +2023/06/01 11:22:16 - mmengine - INFO - Epoch(train) [13][ 5700/11516] lr: 3.5204e-04 eta: 10:03:34 time: 0.4167 data_time: 0.0013 memory: 9465 grad_norm: 0.0281 loss: 0.6863 +2023/06/01 11:22:59 - mmengine - INFO - Epoch(train) [13][ 5800/11516] lr: 3.5204e-04 eta: 10:02:53 time: 0.4213 data_time: 0.0015 memory: 9465 grad_norm: 0.0173 loss: 0.6914 +2023/06/01 11:23:01 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/06/01 11:23:43 - mmengine - INFO - Epoch(train) [13][ 5900/11516] lr: 3.5204e-04 eta: 10:02:13 time: 0.4568 data_time: 0.0024 memory: 9465 grad_norm: 0.0164 loss: 0.6848 +2023/06/01 11:24:27 - mmengine - INFO - Epoch(train) [13][ 6000/11516] lr: 3.5204e-04 eta: 10:01:32 time: 0.4473 data_time: 0.0015 memory: 9465 grad_norm: 0.0210 loss: 0.6862 +2023/06/01 11:25:11 - mmengine - INFO - Epoch(train) [13][ 6100/11516] lr: 3.5204e-04 eta: 10:00:51 time: 0.4448 data_time: 0.0013 memory: 9465 grad_norm: 0.0231 loss: 0.6883 +2023/06/01 11:25:55 - mmengine - INFO - Epoch(train) [13][ 6200/11516] lr: 3.5204e-04 eta: 10:00:11 time: 0.5097 data_time: 0.0012 memory: 9465 grad_norm: 0.0228 loss: 0.6848 +2023/06/01 11:26:38 - mmengine - INFO - Epoch(train) [13][ 6300/11516] lr: 3.5204e-04 eta: 9:59:29 time: 0.4158 data_time: 0.0013 memory: 9465 grad_norm: 0.0173 loss: 0.6886 +2023/06/01 11:27:21 - mmengine - INFO - Epoch(train) [13][ 6400/11516] lr: 3.5204e-04 eta: 9:58:48 time: 0.4319 data_time: 0.0013 memory: 9465 grad_norm: 0.0171 loss: 0.6874 +2023/06/01 11:28:03 - mmengine - INFO - Epoch(train) [13][ 6500/11516] lr: 3.5204e-04 eta: 9:58:06 time: 0.4290 data_time: 0.0013 memory: 9465 grad_norm: 0.0283 loss: 0.6830 +2023/06/01 11:28:46 - mmengine - INFO - Epoch(train) [13][ 6600/11516] lr: 3.5204e-04 eta: 9:57:25 time: 0.4300 data_time: 0.0014 memory: 9465 grad_norm: 0.0181 loss: 0.6855 +2023/06/01 11:29:28 - mmengine - INFO - Epoch(train) [13][ 6700/11516] lr: 3.5204e-04 eta: 9:56:43 time: 0.4172 data_time: 0.0013 memory: 9465 grad_norm: 0.0286 loss: 0.6886 +2023/06/01 11:30:09 - mmengine - INFO - Epoch(train) [13][ 6800/11516] lr: 3.5204e-04 eta: 9:56:01 time: 0.4307 data_time: 0.0013 memory: 9465 grad_norm: 0.0223 loss: 0.6896 +2023/06/01 11:30:11 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/06/01 11:30:53 - mmengine - INFO - Epoch(train) [13][ 6900/11516] lr: 3.5204e-04 eta: 9:55:20 time: 0.4040 data_time: 0.0014 memory: 9465 grad_norm: 0.0165 loss: 0.6855 +2023/06/01 11:31:36 - mmengine - INFO - Epoch(train) [13][ 7000/11516] lr: 3.5204e-04 eta: 9:54:39 time: 0.4270 data_time: 0.0013 memory: 9465 grad_norm: 0.0158 loss: 0.6888 +2023/06/01 11:32:19 - mmengine - INFO - Epoch(train) [13][ 7100/11516] lr: 3.5204e-04 eta: 9:53:57 time: 0.4738 data_time: 0.0014 memory: 9465 grad_norm: 0.0295 loss: 0.6885 +2023/06/01 11:33:03 - mmengine - INFO - Epoch(train) [13][ 7200/11516] lr: 3.5204e-04 eta: 9:53:16 time: 0.4256 data_time: 0.0013 memory: 9465 grad_norm: 0.0168 loss: 0.6827 +2023/06/01 11:33:48 - mmengine - INFO - Epoch(train) [13][ 7300/11516] lr: 3.5204e-04 eta: 9:52:36 time: 0.4465 data_time: 0.0014 memory: 9465 grad_norm: 0.0180 loss: 0.6885 +2023/06/01 11:34:32 - mmengine - INFO - Epoch(train) [13][ 7400/11516] lr: 3.5204e-04 eta: 9:51:56 time: 0.4355 data_time: 0.0013 memory: 9465 grad_norm: 0.0252 loss: 0.6857 +2023/06/01 11:35:17 - mmengine - INFO - Epoch(train) [13][ 7500/11516] lr: 3.5204e-04 eta: 9:51:15 time: 0.4229 data_time: 0.0016 memory: 9465 grad_norm: 0.0161 loss: 0.6883 +2023/06/01 11:36:00 - mmengine - INFO - Epoch(train) [13][ 7600/11516] lr: 3.5204e-04 eta: 9:50:35 time: 0.4555 data_time: 0.0014 memory: 9465 grad_norm: 0.0258 loss: 0.6870 +2023/06/01 11:36:45 - mmengine - INFO - Epoch(train) [13][ 7700/11516] lr: 3.5204e-04 eta: 9:49:54 time: 0.4758 data_time: 0.0014 memory: 9465 grad_norm: 0.0247 loss: 0.6902 +2023/06/01 11:37:30 - mmengine - INFO - Epoch(train) [13][ 7800/11516] lr: 3.5204e-04 eta: 9:49:14 time: 0.4306 data_time: 0.0013 memory: 9465 grad_norm: 0.0187 loss: 0.6887 +2023/06/01 11:37:32 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/06/01 11:38:14 - mmengine - INFO - Epoch(train) [13][ 7900/11516] lr: 3.5204e-04 eta: 9:48:33 time: 0.4317 data_time: 0.0015 memory: 9465 grad_norm: 0.0170 loss: 0.6870 +2023/06/01 11:38:58 - mmengine - INFO - Epoch(train) [13][ 8000/11516] lr: 3.5204e-04 eta: 9:47:52 time: 0.4517 data_time: 0.0014 memory: 9465 grad_norm: 0.0169 loss: 0.6879 +2023/06/01 11:39:42 - mmengine - INFO - Epoch(train) [13][ 8100/11516] lr: 3.5204e-04 eta: 9:47:12 time: 0.4266 data_time: 0.0015 memory: 9465 grad_norm: 0.0191 loss: 0.6873 +2023/06/01 11:40:25 - mmengine - INFO - Epoch(train) [13][ 8200/11516] lr: 3.5204e-04 eta: 9:46:30 time: 0.4159 data_time: 0.0013 memory: 9465 grad_norm: 0.0321 loss: 0.6874 +2023/06/01 11:41:07 - mmengine - INFO - Epoch(train) [13][ 8300/11516] lr: 3.5204e-04 eta: 9:45:49 time: 0.4226 data_time: 0.0013 memory: 9465 grad_norm: 0.0281 loss: 0.6881 +2023/06/01 11:41:51 - mmengine - INFO - Epoch(train) [13][ 8400/11516] lr: 3.5204e-04 eta: 9:45:08 time: 0.4211 data_time: 0.0013 memory: 9465 grad_norm: 0.0196 loss: 0.6883 +2023/06/01 11:42:34 - mmengine - INFO - Epoch(train) [13][ 8500/11516] lr: 3.5204e-04 eta: 9:44:26 time: 0.4369 data_time: 0.0013 memory: 9465 grad_norm: 0.0241 loss: 0.6909 +2023/06/01 11:43:15 - mmengine - INFO - Epoch(train) [13][ 8600/11516] lr: 3.5204e-04 eta: 9:43:44 time: 0.3904 data_time: 0.0013 memory: 9465 grad_norm: 0.0265 loss: 0.6873 +2023/06/01 11:43:56 - mmengine - INFO - Epoch(train) [13][ 8700/11516] lr: 3.5204e-04 eta: 9:43:01 time: 0.4043 data_time: 0.0014 memory: 9465 grad_norm: 0.0180 loss: 0.6805 +2023/06/01 11:44:38 - mmengine - INFO - Epoch(train) [13][ 8800/11516] lr: 3.5204e-04 eta: 9:42:19 time: 0.4326 data_time: 0.0014 memory: 9465 grad_norm: 0.0171 loss: 0.6896 +2023/06/01 11:44:39 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/06/01 11:45:20 - mmengine - INFO - Epoch(train) [13][ 8900/11516] lr: 3.5204e-04 eta: 9:41:38 time: 0.4009 data_time: 0.0013 memory: 9465 grad_norm: 0.0227 loss: 0.6880 +2023/06/01 11:46:02 - mmengine - INFO - Epoch(train) [13][ 9000/11516] lr: 3.5204e-04 eta: 9:40:56 time: 0.3997 data_time: 0.0013 memory: 9465 grad_norm: 0.0168 loss: 0.6839 +2023/06/01 11:46:43 - mmengine - INFO - Epoch(train) [13][ 9100/11516] lr: 3.5204e-04 eta: 9:40:13 time: 0.4051 data_time: 0.0013 memory: 9465 grad_norm: 0.0212 loss: 0.6870 +2023/06/01 11:47:25 - mmengine - INFO - Epoch(train) [13][ 9200/11516] lr: 3.5204e-04 eta: 9:39:31 time: 0.4454 data_time: 0.0016 memory: 9465 grad_norm: 0.0173 loss: 0.6870 +2023/06/01 11:48:05 - mmengine - INFO - Epoch(train) [13][ 9300/11516] lr: 3.5204e-04 eta: 9:38:49 time: 0.4150 data_time: 0.0014 memory: 9465 grad_norm: 0.0279 loss: 0.6824 +2023/06/01 11:48:48 - mmengine - INFO - Epoch(train) [13][ 9400/11516] lr: 3.5204e-04 eta: 9:38:07 time: 0.4445 data_time: 0.0013 memory: 9465 grad_norm: 0.0183 loss: 0.6849 +2023/06/01 11:49:30 - mmengine - INFO - Epoch(train) [13][ 9500/11516] lr: 3.5204e-04 eta: 9:37:25 time: 0.3852 data_time: 0.0013 memory: 9465 grad_norm: 0.0235 loss: 0.6894 +2023/06/01 11:50:12 - mmengine - INFO - Epoch(train) [13][ 9600/11516] lr: 3.5204e-04 eta: 9:36:43 time: 0.4274 data_time: 0.0013 memory: 9465 grad_norm: 0.0157 loss: 0.6868 +2023/06/01 11:50:53 - mmengine - INFO - Epoch(train) [13][ 9700/11516] lr: 3.5204e-04 eta: 9:36:01 time: 0.4033 data_time: 0.0013 memory: 9465 grad_norm: 0.0166 loss: 0.6851 +2023/06/01 11:51:35 - mmengine - INFO - Epoch(train) [13][ 9800/11516] lr: 3.5204e-04 eta: 9:35:19 time: 0.4567 data_time: 0.0013 memory: 9465 grad_norm: 0.0266 loss: 0.6867 +2023/06/01 11:51:37 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/06/01 11:52:17 - mmengine - INFO - Epoch(train) [13][ 9900/11516] lr: 3.5204e-04 eta: 9:34:37 time: 0.3970 data_time: 0.0014 memory: 9465 grad_norm: 0.0184 loss: 0.6862 +2023/06/01 11:52:59 - mmengine - INFO - Epoch(train) [13][10000/11516] lr: 3.5204e-04 eta: 9:33:55 time: 0.4022 data_time: 0.0013 memory: 9465 grad_norm: 0.0130 loss: 0.6824 +2023/06/01 11:53:42 - mmengine - INFO - Epoch(train) [13][10100/11516] lr: 3.5204e-04 eta: 9:33:14 time: 0.3920 data_time: 0.0014 memory: 9465 grad_norm: 0.0178 loss: 0.6860 +2023/06/01 11:54:24 - mmengine - INFO - Epoch(train) [13][10200/11516] lr: 3.5204e-04 eta: 9:32:32 time: 0.3878 data_time: 0.0014 memory: 9465 grad_norm: 0.0219 loss: 0.6832 +2023/06/01 11:55:04 - mmengine - INFO - Epoch(train) [13][10300/11516] lr: 3.5204e-04 eta: 9:31:49 time: 0.4030 data_time: 0.0013 memory: 9465 grad_norm: 0.0296 loss: 0.6863 +2023/06/01 11:55:46 - mmengine - INFO - Epoch(train) [13][10400/11516] lr: 3.5204e-04 eta: 9:31:07 time: 0.4425 data_time: 0.0013 memory: 9465 grad_norm: 0.0217 loss: 0.6841 +2023/06/01 11:56:27 - mmengine - INFO - Epoch(train) [13][10500/11516] lr: 3.5204e-04 eta: 9:30:24 time: 0.4498 data_time: 0.0014 memory: 9465 grad_norm: 0.0258 loss: 0.6886 +2023/06/01 11:57:07 - mmengine - INFO - Epoch(train) [13][10600/11516] lr: 3.5204e-04 eta: 9:29:42 time: 0.3939 data_time: 0.0013 memory: 9465 grad_norm: 0.0245 loss: 0.6842 +2023/06/01 11:57:48 - mmengine - INFO - Epoch(train) [13][10700/11516] lr: 3.5204e-04 eta: 9:28:59 time: 0.4220 data_time: 0.0013 memory: 9465 grad_norm: 0.0173 loss: 0.6848 +2023/06/01 11:58:29 - mmengine - INFO - Epoch(train) [13][10800/11516] lr: 3.5204e-04 eta: 9:28:17 time: 0.4177 data_time: 0.0013 memory: 9465 grad_norm: 0.0232 loss: 0.6845 +2023/06/01 11:58:30 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/06/01 11:59:11 - mmengine - INFO - Epoch(train) [13][10900/11516] lr: 3.5204e-04 eta: 9:27:34 time: 0.4121 data_time: 0.0016 memory: 9465 grad_norm: 0.0256 loss: 0.6857 +2023/06/01 11:59:51 - mmengine - INFO - Epoch(train) [13][11000/11516] lr: 3.5204e-04 eta: 9:26:52 time: 0.3965 data_time: 0.0013 memory: 9465 grad_norm: 0.0201 loss: 0.6857 +2023/06/01 12:00:32 - mmengine - INFO - Epoch(train) [13][11100/11516] lr: 3.5204e-04 eta: 9:26:09 time: 0.4047 data_time: 0.0014 memory: 9465 grad_norm: 0.0327 loss: 0.6920 +2023/06/01 12:01:13 - mmengine - INFO - Epoch(train) [13][11200/11516] lr: 3.5204e-04 eta: 9:25:27 time: 0.4113 data_time: 0.0015 memory: 9465 grad_norm: 0.0312 loss: 0.6884 +2023/06/01 12:01:53 - mmengine - INFO - Epoch(train) [13][11300/11516] lr: 3.5204e-04 eta: 9:24:44 time: 0.3924 data_time: 0.0013 memory: 9465 grad_norm: 0.0297 loss: 0.6896 +2023/06/01 12:02:32 - mmengine - INFO - Epoch(train) [13][11400/11516] lr: 3.5204e-04 eta: 9:24:00 time: 0.4069 data_time: 0.0020 memory: 9465 grad_norm: 0.0224 loss: 0.6844 +2023/06/01 12:03:12 - mmengine - INFO - Epoch(train) [13][11500/11516] lr: 3.5204e-04 eta: 9:23:17 time: 0.3860 data_time: 0.0114 memory: 9465 grad_norm: 0.0245 loss: 0.6846 +2023/06/01 12:03:17 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/06/01 12:03:17 - mmengine - INFO - Saving checkpoint at 13 epochs +2023/06/01 12:03:33 - mmengine - INFO - Epoch(val) [13][16/16] accuracy/top1: 100.0000 single-label/precision_classwise: [100.0, 0.0] single-label/recall_classwise: [100.0, 0.0] single-label/f1-score_classwise: [100.0, 0.0] data_time: 0.3074 time: 0.5612 +2023/06/01 12:04:17 - mmengine - INFO - Epoch(train) [14][ 100/11516] lr: 2.8027e-04 eta: 9:22:28 time: 0.4496 data_time: 0.0406 memory: 9465 grad_norm: 0.0236 loss: 0.6857 +2023/06/01 12:04:57 - mmengine - INFO - Epoch(train) [14][ 200/11516] lr: 2.8027e-04 eta: 9:21:46 time: 0.4567 data_time: 0.0014 memory: 9465 grad_norm: 0.0370 loss: 0.6922 +2023/06/01 12:05:34 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/06/01 12:05:38 - mmengine - INFO - Epoch(train) [14][ 300/11516] lr: 2.8027e-04 eta: 9:21:03 time: 0.3936 data_time: 0.0016 memory: 9465 grad_norm: 0.0250 loss: 0.6892 +2023/06/01 12:06:19 - mmengine - INFO - Epoch(train) [14][ 400/11516] lr: 2.8027e-04 eta: 9:20:21 time: 0.4038 data_time: 0.0014 memory: 9465 grad_norm: 0.0280 loss: 0.6854 +2023/06/01 12:07:00 - mmengine - INFO - Epoch(train) [14][ 500/11516] lr: 2.8027e-04 eta: 9:19:38 time: 0.4048 data_time: 0.0016 memory: 9465 grad_norm: 0.0217 loss: 0.6890 +2023/06/01 12:07:40 - mmengine - INFO - Epoch(train) [14][ 600/11516] lr: 2.8027e-04 eta: 9:18:56 time: 0.3884 data_time: 0.0016 memory: 9465 grad_norm: 0.0187 loss: 0.6893 +2023/06/01 12:08:21 - mmengine - INFO - Epoch(train) [14][ 700/11516] lr: 2.8027e-04 eta: 9:18:13 time: 0.3860 data_time: 0.0014 memory: 9465 grad_norm: 0.0293 loss: 0.6798 +2023/06/01 12:09:02 - mmengine - INFO - Epoch(train) [14][ 800/11516] lr: 2.8027e-04 eta: 9:17:31 time: 0.4175 data_time: 0.0015 memory: 9465 grad_norm: 0.0168 loss: 0.6895 +2023/06/01 12:09:43 - mmengine - INFO - Epoch(train) [14][ 900/11516] lr: 2.8027e-04 eta: 9:16:48 time: 0.4312 data_time: 0.0014 memory: 9465 grad_norm: 0.0183 loss: 0.6832 +2023/06/01 12:10:24 - mmengine - INFO - Epoch(train) [14][ 1000/11516] lr: 2.8027e-04 eta: 9:16:06 time: 0.3948 data_time: 0.0023 memory: 9465 grad_norm: 0.0176 loss: 0.6860 +2023/06/01 12:11:05 - mmengine - INFO - Epoch(train) [14][ 1100/11516] lr: 2.8027e-04 eta: 9:15:23 time: 0.4168 data_time: 0.0015 memory: 9465 grad_norm: 0.0249 loss: 0.6871 +2023/06/01 12:11:46 - mmengine - INFO - Epoch(train) [14][ 1200/11516] lr: 2.8027e-04 eta: 9:14:41 time: 0.4189 data_time: 0.0013 memory: 9465 grad_norm: 0.0300 loss: 0.6806 +2023/06/01 12:12:23 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/06/01 12:12:26 - mmengine - INFO - Epoch(train) [14][ 1300/11516] lr: 2.8027e-04 eta: 9:13:58 time: 0.3802 data_time: 0.0016 memory: 9465 grad_norm: 0.0186 loss: 0.6834 +2023/06/01 12:13:08 - mmengine - INFO - Epoch(train) [14][ 1400/11516] lr: 2.8027e-04 eta: 9:13:16 time: 0.4099 data_time: 0.0014 memory: 9465 grad_norm: 0.0196 loss: 0.6849 +2023/06/01 12:13:48 - mmengine - INFO - Epoch(train) [14][ 1500/11516] lr: 2.8027e-04 eta: 9:12:33 time: 0.4103 data_time: 0.0013 memory: 9465 grad_norm: 0.0201 loss: 0.6882 +2023/06/01 12:14:29 - mmengine - INFO - Epoch(train) [14][ 1600/11516] lr: 2.8027e-04 eta: 9:11:51 time: 0.4223 data_time: 0.0013 memory: 9465 grad_norm: 0.0238 loss: 0.6888 +2023/06/01 12:15:09 - mmengine - INFO - Epoch(train) [14][ 1700/11516] lr: 2.8027e-04 eta: 9:11:08 time: 0.3873 data_time: 0.0012 memory: 9465 grad_norm: 0.0281 loss: 0.6867 +2023/06/01 12:15:50 - mmengine - INFO - Epoch(train) [14][ 1800/11516] lr: 2.8027e-04 eta: 9:10:25 time: 0.3862 data_time: 0.0013 memory: 9465 grad_norm: 0.0163 loss: 0.6875 +2023/06/01 12:16:29 - mmengine - INFO - Epoch(train) [14][ 1900/11516] lr: 2.8027e-04 eta: 9:09:43 time: 0.3970 data_time: 0.0013 memory: 9465 grad_norm: 0.0181 loss: 0.6835 +2023/06/01 12:17:10 - mmengine - INFO - Epoch(train) [14][ 2000/11516] lr: 2.8027e-04 eta: 9:09:00 time: 0.3931 data_time: 0.0014 memory: 9465 grad_norm: 0.0185 loss: 0.6914 +2023/06/01 12:17:51 - mmengine - INFO - Epoch(train) [14][ 2100/11516] lr: 2.8027e-04 eta: 9:08:18 time: 0.4001 data_time: 0.0013 memory: 9465 grad_norm: 0.0135 loss: 0.6799 +2023/06/01 12:18:32 - mmengine - INFO - Epoch(train) [14][ 2200/11516] lr: 2.8027e-04 eta: 9:07:35 time: 0.4364 data_time: 0.0013 memory: 9465 grad_norm: 0.0318 loss: 0.6827 +2023/06/01 12:19:11 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/06/01 12:19:14 - mmengine - INFO - Epoch(train) [14][ 2300/11516] lr: 2.8027e-04 eta: 9:06:53 time: 0.4121 data_time: 0.0017 memory: 9465 grad_norm: 0.0309 loss: 0.6900 +2023/06/01 12:19:58 - mmengine - INFO - Epoch(train) [14][ 2400/11516] lr: 2.8027e-04 eta: 9:06:12 time: 0.4283 data_time: 0.0014 memory: 9465 grad_norm: 0.0241 loss: 0.6892 +2023/06/01 12:20:39 - mmengine - INFO - Epoch(train) [14][ 2500/11516] lr: 2.8027e-04 eta: 9:05:30 time: 0.4028 data_time: 0.0014 memory: 9465 grad_norm: 0.0236 loss: 0.6899 +2023/06/01 12:21:22 - mmengine - INFO - Epoch(train) [14][ 2600/11516] lr: 2.8027e-04 eta: 9:04:48 time: 0.4194 data_time: 0.0014 memory: 9465 grad_norm: 0.0289 loss: 0.6872 +2023/06/01 12:22:04 - mmengine - INFO - Epoch(train) [14][ 2700/11516] lr: 2.8027e-04 eta: 9:04:07 time: 0.3882 data_time: 0.0017 memory: 9465 grad_norm: 0.0324 loss: 0.6836 +2023/06/01 12:22:45 - mmengine - INFO - Epoch(train) [14][ 2800/11516] lr: 2.8027e-04 eta: 9:03:24 time: 0.4036 data_time: 0.0015 memory: 9465 grad_norm: 0.0217 loss: 0.6844 +2023/06/01 12:23:28 - mmengine - INFO - Epoch(train) [14][ 2900/11516] lr: 2.8027e-04 eta: 9:02:43 time: 0.4421 data_time: 0.0015 memory: 9465 grad_norm: 0.0183 loss: 0.6859 +2023/06/01 12:24:08 - mmengine - INFO - Epoch(train) [14][ 3000/11516] lr: 2.8027e-04 eta: 9:02:00 time: 0.4098 data_time: 0.0016 memory: 9465 grad_norm: 0.0192 loss: 0.6874 +2023/06/01 12:24:48 - mmengine - INFO - Epoch(train) [14][ 3100/11516] lr: 2.8027e-04 eta: 9:01:18 time: 0.3967 data_time: 0.0016 memory: 9465 grad_norm: 0.0153 loss: 0.6865 +2023/06/01 12:25:29 - mmengine - INFO - Epoch(train) [14][ 3200/11516] lr: 2.8027e-04 eta: 9:00:35 time: 0.4105 data_time: 0.0013 memory: 9465 grad_norm: 0.0197 loss: 0.6829 +2023/06/01 12:26:05 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/06/01 12:26:09 - mmengine - INFO - Epoch(train) [14][ 3300/11516] lr: 2.8027e-04 eta: 8:59:52 time: 0.3875 data_time: 0.0014 memory: 9465 grad_norm: 0.0212 loss: 0.6763 +2023/06/01 12:26:50 - mmengine - INFO - Epoch(train) [14][ 3400/11516] lr: 2.8027e-04 eta: 8:59:10 time: 0.4622 data_time: 0.0017 memory: 9465 grad_norm: 0.0156 loss: 0.6831 +2023/06/01 12:27:31 - mmengine - INFO - Epoch(train) [14][ 3500/11516] lr: 2.8027e-04 eta: 8:58:27 time: 0.4225 data_time: 0.0016 memory: 9465 grad_norm: 0.0189 loss: 0.6830 +2023/06/01 12:28:12 - mmengine - INFO - Epoch(train) [14][ 3600/11516] lr: 2.8027e-04 eta: 8:57:45 time: 0.4188 data_time: 0.0014 memory: 9465 grad_norm: 0.0202 loss: 0.6846 +2023/06/01 12:28:51 - mmengine - INFO - Epoch(train) [14][ 3700/11516] lr: 2.8027e-04 eta: 8:57:02 time: 0.3797 data_time: 0.0013 memory: 9465 grad_norm: 0.0167 loss: 0.6841 +2023/06/01 12:29:32 - mmengine - INFO - Epoch(train) [14][ 3800/11516] lr: 2.8027e-04 eta: 8:56:19 time: 0.4168 data_time: 0.0019 memory: 9465 grad_norm: 0.0215 loss: 0.6869 +2023/06/01 12:30:13 - mmengine - INFO - Epoch(train) [14][ 3900/11516] lr: 2.8027e-04 eta: 8:55:37 time: 0.4406 data_time: 0.0014 memory: 9465 grad_norm: 0.0210 loss: 0.6868 +2023/06/01 12:30:53 - mmengine - INFO - Epoch(train) [14][ 4000/11516] lr: 2.8027e-04 eta: 8:54:54 time: 0.4353 data_time: 0.0014 memory: 9465 grad_norm: 0.0179 loss: 0.6830 +2023/06/01 12:31:34 - mmengine - INFO - Epoch(train) [14][ 4100/11516] lr: 2.8027e-04 eta: 8:54:12 time: 0.3883 data_time: 0.0015 memory: 9465 grad_norm: 0.0309 loss: 0.6855 +2023/06/01 12:32:15 - mmengine - INFO - Epoch(train) [14][ 4200/11516] lr: 2.8027e-04 eta: 8:53:30 time: 0.4297 data_time: 0.0014 memory: 9465 grad_norm: 0.0368 loss: 0.6864 +2023/06/01 12:32:54 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/06/01 12:32:58 - mmengine - INFO - Epoch(train) [14][ 4300/11516] lr: 2.8027e-04 eta: 8:52:48 time: 0.4448 data_time: 0.0015 memory: 9465 grad_norm: 0.0116 loss: 0.6860 +2023/06/01 12:33:40 - mmengine - INFO - Epoch(train) [14][ 4400/11516] lr: 2.8027e-04 eta: 8:52:06 time: 0.3970 data_time: 0.0014 memory: 9465 grad_norm: 0.0189 loss: 0.6898 +2023/06/01 12:34:20 - mmengine - INFO - Epoch(train) [14][ 4500/11516] lr: 2.8027e-04 eta: 8:51:23 time: 0.3953 data_time: 0.0014 memory: 9465 grad_norm: 0.0163 loss: 0.6871 +2023/06/01 12:35:00 - mmengine - INFO - Epoch(train) [14][ 4600/11516] lr: 2.8027e-04 eta: 8:50:41 time: 0.3828 data_time: 0.0016 memory: 9465 grad_norm: 0.0244 loss: 0.6839 +2023/06/01 12:35:42 - mmengine - INFO - Epoch(train) [14][ 4700/11516] lr: 2.8027e-04 eta: 8:49:59 time: 0.4069 data_time: 0.0014 memory: 9465 grad_norm: 0.0511 loss: 0.6827 +2023/06/01 12:36:23 - mmengine - INFO - Epoch(train) [14][ 4800/11516] lr: 2.8027e-04 eta: 8:49:16 time: 0.4227 data_time: 0.0014 memory: 9465 grad_norm: 0.0148 loss: 0.6876 +2023/06/01 12:37:04 - mmengine - INFO - Epoch(train) [14][ 4900/11516] lr: 2.8027e-04 eta: 8:48:34 time: 0.4065 data_time: 0.0014 memory: 9465 grad_norm: 0.0421 loss: 0.6855 +2023/06/01 12:37:45 - mmengine - INFO - Epoch(train) [14][ 5000/11516] lr: 2.8027e-04 eta: 8:47:52 time: 0.4217 data_time: 0.0015 memory: 9465 grad_norm: 0.0528 loss: 0.6866 +2023/06/01 12:38:26 - mmengine - INFO - Epoch(train) [14][ 5100/11516] lr: 2.8027e-04 eta: 8:47:09 time: 0.3991 data_time: 0.0013 memory: 9465 grad_norm: 0.0413 loss: 0.6906 +2023/06/01 12:39:06 - mmengine - INFO - Epoch(train) [14][ 5200/11516] lr: 2.8027e-04 eta: 8:46:27 time: 0.4468 data_time: 0.0013 memory: 9465 grad_norm: 0.0483 loss: 0.6867 +2023/06/01 12:39:43 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/06/01 12:39:47 - mmengine - INFO - Epoch(train) [14][ 5300/11516] lr: 2.8027e-04 eta: 8:45:44 time: 0.4100 data_time: 0.0016 memory: 9465 grad_norm: 0.1907 loss: 0.6856 +2023/06/01 12:40:28 - mmengine - INFO - Epoch(train) [14][ 5400/11516] lr: 2.8027e-04 eta: 8:45:02 time: 0.3975 data_time: 0.0020 memory: 9465 grad_norm: 0.0274 loss: 0.6825 +2023/06/01 12:41:10 - mmengine - INFO - Epoch(train) [14][ 5500/11516] lr: 2.8027e-04 eta: 8:44:20 time: 0.4042 data_time: 0.0021 memory: 9465 grad_norm: 0.0362 loss: 0.6836 +2023/06/01 12:41:51 - mmengine - INFO - Epoch(train) [14][ 5600/11516] lr: 2.8027e-04 eta: 8:43:38 time: 0.3902 data_time: 0.0022 memory: 9465 grad_norm: 0.0384 loss: 0.6850 +2023/06/01 12:42:32 - mmengine - INFO - Epoch(train) [14][ 5700/11516] lr: 2.8027e-04 eta: 8:42:55 time: 0.3999 data_time: 0.0026 memory: 9465 grad_norm: 0.0264 loss: 0.6855 +2023/06/01 12:43:13 - mmengine - INFO - Epoch(train) [14][ 5800/11516] lr: 2.8027e-04 eta: 8:42:13 time: 0.4012 data_time: 0.0015 memory: 9465 grad_norm: 0.0713 loss: 0.6866 +2023/06/01 12:43:54 - mmengine - INFO - Epoch(train) [14][ 5900/11516] lr: 2.8027e-04 eta: 8:41:31 time: 0.4033 data_time: 0.0020 memory: 9465 grad_norm: 0.0428 loss: 0.6910 +2023/06/01 12:44:37 - mmengine - INFO - Epoch(train) [14][ 6000/11516] lr: 2.8027e-04 eta: 8:40:49 time: 0.4134 data_time: 0.0018 memory: 9465 grad_norm: 0.0232 loss: 0.6863 +2023/06/01 12:45:18 - mmengine - INFO - Epoch(train) [14][ 6100/11516] lr: 2.8027e-04 eta: 8:40:07 time: 0.4257 data_time: 0.0020 memory: 9465 grad_norm: 0.0806 loss: 0.6866 +2023/06/01 12:45:57 - mmengine - INFO - Epoch(train) [14][ 6200/11516] lr: 2.8027e-04 eta: 8:39:24 time: 0.3798 data_time: 0.0026 memory: 9465 grad_norm: 0.0233 loss: 0.6900 +2023/06/01 12:46:34 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/06/01 12:46:39 - mmengine - INFO - Epoch(train) [14][ 6300/11516] lr: 2.8027e-04 eta: 8:38:42 time: 0.4714 data_time: 0.0025 memory: 9465 grad_norm: 0.0318 loss: 0.6919 +2023/06/01 12:47:19 - mmengine - INFO - Epoch(train) [14][ 6400/11516] lr: 2.8027e-04 eta: 8:37:59 time: 0.4028 data_time: 0.0023 memory: 9465 grad_norm: 0.0904 loss: 0.6870 +2023/06/01 12:47:59 - mmengine - INFO - Epoch(train) [14][ 6500/11516] lr: 2.8027e-04 eta: 8:37:16 time: 0.3939 data_time: 0.0019 memory: 9465 grad_norm: 0.0381 loss: 0.6843 +2023/06/01 12:48:39 - mmengine - INFO - Epoch(train) [14][ 6600/11516] lr: 2.8027e-04 eta: 8:36:34 time: 0.3899 data_time: 0.0019 memory: 9465 grad_norm: 0.1432 loss: 0.6866 +2023/06/01 12:49:20 - mmengine - INFO - Epoch(train) [14][ 6700/11516] lr: 2.8027e-04 eta: 8:35:51 time: 0.4034 data_time: 0.0022 memory: 9465 grad_norm: 0.0315 loss: 0.6880 +2023/06/01 12:50:01 - mmengine - INFO - Epoch(train) [14][ 6800/11516] lr: 2.8027e-04 eta: 8:35:09 time: 0.4083 data_time: 0.0024 memory: 9465 grad_norm: 0.0277 loss: 0.6814 +2023/06/01 12:50:44 - mmengine - INFO - Epoch(train) [14][ 6900/11516] lr: 2.8027e-04 eta: 8:34:27 time: 0.4278 data_time: 0.0020 memory: 9465 grad_norm: 0.0285 loss: 0.6839 +2023/06/01 12:51:24 - mmengine - INFO - Epoch(train) [14][ 7000/11516] lr: 2.8027e-04 eta: 8:33:45 time: 0.4227 data_time: 0.0020 memory: 9465 grad_norm: 0.0245 loss: 0.6873 +2023/06/01 12:52:05 - mmengine - INFO - Epoch(train) [14][ 7100/11516] lr: 2.8027e-04 eta: 8:33:02 time: 0.4373 data_time: 0.0019 memory: 9465 grad_norm: 0.0313 loss: 0.6884 +2023/06/01 12:52:45 - mmengine - INFO - Epoch(train) [14][ 7200/11516] lr: 2.8027e-04 eta: 8:32:20 time: 0.4303 data_time: 0.0019 memory: 9465 grad_norm: 0.0222 loss: 0.6841 +2023/06/01 12:53:22 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/06/01 12:53:25 - mmengine - INFO - Epoch(train) [14][ 7300/11516] lr: 2.8027e-04 eta: 8:31:37 time: 0.3942 data_time: 0.0021 memory: 9465 grad_norm: 0.0342 loss: 0.6835 +2023/06/01 12:54:06 - mmengine - INFO - Epoch(train) [14][ 7400/11516] lr: 2.8027e-04 eta: 8:30:55 time: 0.4314 data_time: 0.0021 memory: 9465 grad_norm: 0.0135 loss: 0.6881 +2023/06/01 12:54:47 - mmengine - INFO - Epoch(train) [14][ 7500/11516] lr: 2.8027e-04 eta: 8:30:12 time: 0.4013 data_time: 0.0017 memory: 9465 grad_norm: 0.0291 loss: 0.6895 +2023/06/01 12:55:28 - mmengine - INFO - Epoch(train) [14][ 7600/11516] lr: 2.8027e-04 eta: 8:29:30 time: 0.3946 data_time: 0.0017 memory: 9465 grad_norm: 0.0239 loss: 0.6865 +2023/06/01 12:56:08 - mmengine - INFO - Epoch(train) [14][ 7700/11516] lr: 2.8027e-04 eta: 8:28:48 time: 0.3993 data_time: 0.0023 memory: 9465 grad_norm: 0.0270 loss: 0.6864 +2023/06/01 12:56:48 - mmengine - INFO - Epoch(train) [14][ 7800/11516] lr: 2.8027e-04 eta: 8:28:05 time: 0.3973 data_time: 0.0022 memory: 9465 grad_norm: 0.0868 loss: 0.6799 +2023/06/01 12:57:30 - mmengine - INFO - Epoch(train) [14][ 7900/11516] lr: 2.8027e-04 eta: 8:27:23 time: 0.4209 data_time: 0.0022 memory: 9465 grad_norm: 0.0280 loss: 0.6861 +2023/06/01 12:58:10 - mmengine - INFO - Epoch(train) [14][ 8000/11516] lr: 2.8027e-04 eta: 8:26:40 time: 0.3966 data_time: 0.0019 memory: 9465 grad_norm: 0.0297 loss: 0.6827 +2023/06/01 12:58:52 - mmengine - INFO - Epoch(train) [14][ 8100/11516] lr: 2.8027e-04 eta: 8:25:58 time: 0.4014 data_time: 0.0017 memory: 9465 grad_norm: 0.0330 loss: 0.6838 +2023/06/01 12:59:33 - mmengine - INFO - Epoch(train) [14][ 8200/11516] lr: 2.8027e-04 eta: 8:25:16 time: 0.3966 data_time: 0.0026 memory: 9465 grad_norm: 0.0246 loss: 0.6842 +2023/06/01 13:00:11 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/06/01 13:00:15 - mmengine - INFO - Epoch(train) [14][ 8300/11516] lr: 2.8027e-04 eta: 8:24:34 time: 0.4106 data_time: 0.0018 memory: 9465 grad_norm: 0.0293 loss: 0.6883 +2023/06/01 13:00:56 - mmengine - INFO - Epoch(train) [14][ 8400/11516] lr: 2.8027e-04 eta: 8:23:52 time: 0.4161 data_time: 0.0019 memory: 9465 grad_norm: 0.0216 loss: 0.6903 +2023/06/01 13:01:36 - mmengine - INFO - Epoch(train) [14][ 8500/11516] lr: 2.8027e-04 eta: 8:23:09 time: 0.4352 data_time: 0.0018 memory: 9465 grad_norm: 0.0624 loss: 0.6862 +2023/06/01 13:02:17 - mmengine - INFO - Epoch(train) [14][ 8600/11516] lr: 2.8027e-04 eta: 8:22:27 time: 0.4010 data_time: 0.0018 memory: 9465 grad_norm: 0.0371 loss: 0.6847 +2023/06/01 13:02:59 - mmengine - INFO - Epoch(train) [14][ 8700/11516] lr: 2.8027e-04 eta: 8:21:45 time: 0.3933 data_time: 0.0018 memory: 9465 grad_norm: 0.0195 loss: 0.6915 +2023/06/01 13:03:41 - mmengine - INFO - Epoch(train) [14][ 8800/11516] lr: 2.8027e-04 eta: 8:21:03 time: 0.4825 data_time: 0.0021 memory: 9465 grad_norm: 0.0430 loss: 0.6870 +2023/06/01 13:04:22 - mmengine - INFO - Epoch(train) [14][ 8900/11516] lr: 2.8027e-04 eta: 8:20:21 time: 0.4044 data_time: 0.0019 memory: 9465 grad_norm: 0.0835 loss: 0.6912 +2023/06/01 13:05:04 - mmengine - INFO - Epoch(train) [14][ 9000/11516] lr: 2.8027e-04 eta: 8:19:39 time: 0.4237 data_time: 0.0015 memory: 9465 grad_norm: 0.0355 loss: 0.6826 +2023/06/01 13:05:45 - mmengine - INFO - Epoch(train) [14][ 9100/11516] lr: 2.8027e-04 eta: 8:18:57 time: 0.4070 data_time: 0.0023 memory: 9465 grad_norm: 0.1217 loss: 0.6852 +2023/06/01 13:06:27 - mmengine - INFO - Epoch(train) [14][ 9200/11516] lr: 2.8027e-04 eta: 8:18:15 time: 0.4108 data_time: 0.0020 memory: 9465 grad_norm: 0.0286 loss: 0.6916 +2023/06/01 13:07:05 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/06/01 13:07:08 - mmengine - INFO - Epoch(train) [14][ 9300/11516] lr: 2.8027e-04 eta: 8:17:33 time: 0.4042 data_time: 0.0022 memory: 9465 grad_norm: 0.0199 loss: 0.6862 +2023/06/01 13:07:49 - mmengine - INFO - Epoch(train) [14][ 9400/11516] lr: 2.8027e-04 eta: 8:16:50 time: 0.4085 data_time: 0.0021 memory: 9465 grad_norm: 0.0156 loss: 0.6872 +2023/06/01 13:08:30 - mmengine - INFO - Epoch(train) [14][ 9500/11516] lr: 2.8027e-04 eta: 8:16:08 time: 0.4155 data_time: 0.0014 memory: 9465 grad_norm: 0.0554 loss: 0.6954 +2023/06/01 13:09:12 - mmengine - INFO - Epoch(train) [14][ 9600/11516] lr: 2.8027e-04 eta: 8:15:26 time: 0.4119 data_time: 0.0013 memory: 9465 grad_norm: 0.0289 loss: 0.6839 +2023/06/01 13:09:54 - mmengine - INFO - Epoch(train) [14][ 9700/11516] lr: 2.8027e-04 eta: 8:14:45 time: 0.4218 data_time: 0.0014 memory: 9465 grad_norm: 0.0395 loss: 0.6871 +2023/06/01 13:10:36 - mmengine - INFO - Epoch(train) [14][ 9800/11516] lr: 2.8027e-04 eta: 8:14:03 time: 0.4491 data_time: 0.0016 memory: 9465 grad_norm: 0.0557 loss: 0.6911 +2023/06/01 13:11:17 - mmengine - INFO - Epoch(train) [14][ 9900/11516] lr: 2.8027e-04 eta: 8:13:21 time: 0.4095 data_time: 0.0018 memory: 9465 grad_norm: 0.1232 loss: 0.6852 +2023/06/01 13:11:59 - mmengine - INFO - Epoch(train) [14][10000/11516] lr: 2.8027e-04 eta: 8:12:39 time: 0.4360 data_time: 0.0017 memory: 9465 grad_norm: 0.0302 loss: 0.6819 +2023/06/01 13:12:41 - mmengine - INFO - Epoch(train) [14][10100/11516] lr: 2.8027e-04 eta: 8:11:57 time: 0.4068 data_time: 0.0024 memory: 9465 grad_norm: 0.0263 loss: 0.6900 +2023/06/01 13:13:22 - mmengine - INFO - Epoch(train) [14][10200/11516] lr: 2.8027e-04 eta: 8:11:15 time: 0.4057 data_time: 0.0016 memory: 9465 grad_norm: 0.0232 loss: 0.6842 +2023/06/01 13:14:00 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/06/01 13:14:04 - mmengine - INFO - Epoch(train) [14][10300/11516] lr: 2.8027e-04 eta: 8:10:33 time: 0.3958 data_time: 0.0015 memory: 9465 grad_norm: 0.0242 loss: 0.6827 +2023/06/01 13:14:47 - mmengine - INFO - Epoch(train) [14][10400/11516] lr: 2.8027e-04 eta: 8:09:51 time: 0.4216 data_time: 0.0014 memory: 9465 grad_norm: 0.0196 loss: 0.6908 +2023/06/01 13:15:29 - mmengine - INFO - Epoch(train) [14][10500/11516] lr: 2.8027e-04 eta: 8:09:10 time: 0.4625 data_time: 0.0014 memory: 9465 grad_norm: 0.0396 loss: 0.6838 +2023/06/01 13:16:12 - mmengine - INFO - Epoch(train) [14][10600/11516] lr: 2.8027e-04 eta: 8:08:28 time: 0.4821 data_time: 0.0020 memory: 9465 grad_norm: 0.0226 loss: 0.6872 +2023/06/01 13:17:05 - mmengine - INFO - Epoch(train) [14][10700/11516] lr: 2.8027e-04 eta: 8:07:51 time: 0.4752 data_time: 0.0018 memory: 9465 grad_norm: 0.0295 loss: 0.6851 +2023/06/01 13:17:49 - mmengine - INFO - Epoch(train) [14][10800/11516] lr: 2.8027e-04 eta: 8:07:10 time: 0.4238 data_time: 0.0021 memory: 9465 grad_norm: 0.0200 loss: 0.6894 +2023/06/01 13:18:31 - mmengine - INFO - Epoch(train) [14][10900/11516] lr: 2.8027e-04 eta: 8:06:28 time: 0.3976 data_time: 0.0015 memory: 9465 grad_norm: 0.0288 loss: 0.6833 +2023/06/01 13:19:12 - mmengine - INFO - Epoch(train) [14][11000/11516] lr: 2.8027e-04 eta: 8:05:46 time: 0.4083 data_time: 0.0017 memory: 9465 grad_norm: 0.0313 loss: 0.6815 +2023/06/01 13:19:54 - mmengine - INFO - Epoch(train) [14][11100/11516] lr: 2.8027e-04 eta: 8:05:05 time: 0.4357 data_time: 0.0015 memory: 9465 grad_norm: 0.0195 loss: 0.6827 +2023/06/01 13:20:36 - mmengine - INFO - Epoch(train) [14][11200/11516] lr: 2.8027e-04 eta: 8:04:23 time: 0.4341 data_time: 0.0020 memory: 9465 grad_norm: 0.0229 loss: 0.6837 +2023/06/01 13:21:14 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/06/01 13:21:19 - mmengine - INFO - Epoch(train) [14][11300/11516] lr: 2.8027e-04 eta: 8:03:41 time: 0.4702 data_time: 0.0014 memory: 9465 grad_norm: 0.0194 loss: 0.6835 +2023/06/01 13:21:58 - mmengine - INFO - Epoch(train) [14][11400/11516] lr: 2.8027e-04 eta: 8:02:58 time: 0.4000 data_time: 0.0014 memory: 9465 grad_norm: 0.0241 loss: 0.6846 +2023/06/01 13:22:40 - mmengine - INFO - Epoch(train) [14][11500/11516] lr: 2.8027e-04 eta: 8:02:16 time: 0.4068 data_time: 0.0060 memory: 9465 grad_norm: 0.0276 loss: 0.6893 +2023/06/01 13:22:45 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/06/01 13:22:45 - mmengine - INFO - Saving checkpoint at 14 epochs +2023/06/01 13:23:02 - mmengine - INFO - Epoch(val) [14][16/16] accuracy/top1: 100.0000 single-label/precision_classwise: [100.0, 0.0] single-label/recall_classwise: [100.0, 0.0] single-label/f1-score_classwise: [100.0, 0.0] data_time: 0.2853 time: 0.5454 +2023/06/01 13:23:45 - mmengine - INFO - Epoch(train) [15][ 100/11516] lr: 2.1405e-04 eta: 8:01:28 time: 0.4227 data_time: 0.0812 memory: 9465 grad_norm: 0.0187 loss: 0.6884 +2023/06/01 13:24:27 - mmengine - INFO - Epoch(train) [15][ 200/11516] lr: 2.1405e-04 eta: 8:00:46 time: 0.4066 data_time: 0.0015 memory: 9465 grad_norm: 0.0172 loss: 0.6914 +2023/06/01 13:25:08 - mmengine - INFO - Epoch(train) [15][ 300/11516] lr: 2.1405e-04 eta: 8:00:04 time: 0.3904 data_time: 0.0015 memory: 9465 grad_norm: 0.0135 loss: 0.6875 +2023/06/01 13:25:50 - mmengine - INFO - Epoch(train) [15][ 400/11516] lr: 2.1405e-04 eta: 7:59:22 time: 0.4141 data_time: 0.0013 memory: 9465 grad_norm: 0.0163 loss: 0.6861 +2023/06/01 13:26:34 - mmengine - INFO - Epoch(train) [15][ 500/11516] lr: 2.1405e-04 eta: 7:58:41 time: 0.4387 data_time: 0.0013 memory: 9465 grad_norm: 0.0192 loss: 0.6880 +2023/06/01 13:27:19 - mmengine - INFO - Epoch(train) [15][ 600/11516] lr: 2.1405e-04 eta: 7:58:00 time: 0.4543 data_time: 0.0013 memory: 9465 grad_norm: 0.0184 loss: 0.6838 +2023/06/01 13:28:04 - mmengine - INFO - Epoch(train) [15][ 700/11516] lr: 2.1405e-04 eta: 7:57:20 time: 0.4828 data_time: 0.0022 memory: 9465 grad_norm: 0.0233 loss: 0.6863 +2023/06/01 13:28:36 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/06/01 13:28:48 - mmengine - INFO - Epoch(train) [15][ 800/11516] lr: 2.1405e-04 eta: 7:56:39 time: 0.4705 data_time: 0.0016 memory: 9465 grad_norm: 0.0274 loss: 0.6828 +2023/06/01 13:29:31 - mmengine - INFO - Epoch(train) [15][ 900/11516] lr: 2.1405e-04 eta: 7:55:57 time: 0.4060 data_time: 0.0014 memory: 9465 grad_norm: 0.0257 loss: 0.6865 +2023/06/01 13:30:12 - mmengine - INFO - Epoch(train) [15][ 1000/11516] lr: 2.1405e-04 eta: 7:55:15 time: 0.4622 data_time: 0.0013 memory: 9465 grad_norm: 0.0957 loss: 0.6915 +2023/06/01 13:30:55 - mmengine - INFO - Epoch(train) [15][ 1100/11516] lr: 2.1405e-04 eta: 7:54:34 time: 0.4240 data_time: 0.0014 memory: 9465 grad_norm: 0.0311 loss: 0.6913 +2023/06/01 13:31:37 - mmengine - INFO - Epoch(train) [15][ 1200/11516] lr: 2.1405e-04 eta: 7:53:52 time: 0.4177 data_time: 0.0014 memory: 9465 grad_norm: 0.0222 loss: 0.6889 +2023/06/01 13:32:18 - mmengine - INFO - Epoch(train) [15][ 1300/11516] lr: 2.1405e-04 eta: 7:53:10 time: 0.4286 data_time: 0.0013 memory: 9465 grad_norm: 0.0187 loss: 0.6862 +2023/06/01 13:33:00 - mmengine - INFO - Epoch(train) [15][ 1400/11516] lr: 2.1405e-04 eta: 7:52:28 time: 0.4225 data_time: 0.0013 memory: 9465 grad_norm: 0.0235 loss: 0.6847 +2023/06/01 13:33:41 - mmengine - INFO - Epoch(train) [15][ 1500/11516] lr: 2.1405e-04 eta: 7:51:45 time: 0.4112 data_time: 0.0014 memory: 9465 grad_norm: 0.0170 loss: 0.6876 +2023/06/01 13:34:24 - mmengine - INFO - Epoch(train) [15][ 1600/11516] lr: 2.1405e-04 eta: 7:51:04 time: 0.4500 data_time: 0.0017 memory: 9465 grad_norm: 0.0215 loss: 0.6873 +2023/06/01 13:35:04 - mmengine - INFO - Epoch(train) [15][ 1700/11516] lr: 2.1405e-04 eta: 7:50:21 time: 0.3973 data_time: 0.0019 memory: 9465 grad_norm: 0.0124 loss: 0.6895 +2023/06/01 13:35:33 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/06/01 13:35:45 - mmengine - INFO - Epoch(train) [15][ 1800/11516] lr: 2.1405e-04 eta: 7:49:39 time: 0.4235 data_time: 0.0021 memory: 9465 grad_norm: 0.0242 loss: 0.6879 +2023/06/01 13:36:27 - mmengine - INFO - Epoch(train) [15][ 1900/11516] lr: 2.1405e-04 eta: 7:48:57 time: 0.4033 data_time: 0.0014 memory: 9465 grad_norm: 0.0235 loss: 0.6884 +2023/06/01 13:37:08 - mmengine - INFO - Epoch(train) [15][ 2000/11516] lr: 2.1405e-04 eta: 7:48:15 time: 0.4234 data_time: 0.0014 memory: 9465 grad_norm: 0.0198 loss: 0.6834 +2023/06/01 13:37:51 - mmengine - INFO - Epoch(train) [15][ 2100/11516] lr: 2.1405e-04 eta: 7:47:34 time: 0.4088 data_time: 0.0013 memory: 9465 grad_norm: 0.0114 loss: 0.6868 +2023/06/01 13:38:32 - mmengine - INFO - Epoch(train) [15][ 2200/11516] lr: 2.1405e-04 eta: 7:46:52 time: 0.4160 data_time: 0.0018 memory: 9465 grad_norm: 0.0289 loss: 0.6871 +2023/06/01 13:39:15 - mmengine - INFO - Epoch(train) [15][ 2300/11516] lr: 2.1405e-04 eta: 7:46:10 time: 0.4321 data_time: 0.0016 memory: 9465 grad_norm: 0.0160 loss: 0.6868 +2023/06/01 13:39:57 - mmengine - INFO - Epoch(train) [15][ 2400/11516] lr: 2.1405e-04 eta: 7:45:28 time: 0.4425 data_time: 0.0014 memory: 9465 grad_norm: 0.0183 loss: 0.6878 +2023/06/01 13:40:39 - mmengine - INFO - Epoch(train) [15][ 2500/11516] lr: 2.1405e-04 eta: 7:44:46 time: 0.4179 data_time: 0.0013 memory: 9465 grad_norm: 0.0155 loss: 0.6906 +2023/06/01 13:41:20 - mmengine - INFO - Epoch(train) [15][ 2600/11516] lr: 2.1405e-04 eta: 7:44:04 time: 0.4104 data_time: 0.0013 memory: 9465 grad_norm: 0.0242 loss: 0.6788 +2023/06/01 13:42:01 - mmengine - INFO - Epoch(train) [15][ 2700/11516] lr: 2.1405e-04 eta: 7:43:22 time: 0.4289 data_time: 0.0013 memory: 9465 grad_norm: 0.0264 loss: 0.6880 +2023/06/01 13:42:31 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/06/01 13:42:43 - mmengine - INFO - Epoch(train) [15][ 2800/11516] lr: 2.1405e-04 eta: 7:42:40 time: 0.4370 data_time: 0.0013 memory: 9465 grad_norm: 0.0223 loss: 0.6822 +2023/06/01 13:43:25 - mmengine - INFO - Epoch(train) [15][ 2900/11516] lr: 2.1405e-04 eta: 7:41:58 time: 0.4107 data_time: 0.0014 memory: 9465 grad_norm: 0.0222 loss: 0.6826 +2023/06/01 13:44:06 - mmengine - INFO - Epoch(train) [15][ 3000/11516] lr: 2.1405e-04 eta: 7:41:16 time: 0.3929 data_time: 0.0014 memory: 9465 grad_norm: 0.0207 loss: 0.6900 +2023/06/01 13:44:48 - mmengine - INFO - Epoch(train) [15][ 3100/11516] lr: 2.1405e-04 eta: 7:40:34 time: 0.4345 data_time: 0.0013 memory: 9465 grad_norm: 0.0249 loss: 0.6860 +2023/06/01 13:45:30 - mmengine - INFO - Epoch(train) [15][ 3200/11516] lr: 2.1405e-04 eta: 7:39:52 time: 0.4063 data_time: 0.0027 memory: 9465 grad_norm: 0.0157 loss: 0.6840 +2023/06/01 13:46:12 - mmengine - INFO - Epoch(train) [15][ 3300/11516] lr: 2.1405e-04 eta: 7:39:10 time: 0.4249 data_time: 0.0015 memory: 9465 grad_norm: 0.0174 loss: 0.6879 +2023/06/01 13:46:54 - mmengine - INFO - Epoch(train) [15][ 3400/11516] lr: 2.1405e-04 eta: 7:38:29 time: 0.3997 data_time: 0.0030 memory: 9465 grad_norm: 0.0175 loss: 0.6863 +2023/06/01 13:47:36 - mmengine - INFO - Epoch(train) [15][ 3500/11516] lr: 2.1405e-04 eta: 7:37:47 time: 0.4287 data_time: 0.0013 memory: 9465 grad_norm: 0.0328 loss: 0.6849 +2023/06/01 13:48:18 - mmengine - INFO - Epoch(train) [15][ 3600/11516] lr: 2.1405e-04 eta: 7:37:05 time: 0.4084 data_time: 0.0014 memory: 9465 grad_norm: 0.0215 loss: 0.6866 +2023/06/01 13:49:00 - mmengine - INFO - Epoch(train) [15][ 3700/11516] lr: 2.1405e-04 eta: 7:36:23 time: 0.4555 data_time: 0.0013 memory: 9465 grad_norm: 0.0152 loss: 0.6840 +2023/06/01 13:49:31 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/06/01 13:49:42 - mmengine - INFO - Epoch(train) [15][ 3800/11516] lr: 2.1405e-04 eta: 7:35:41 time: 0.3958 data_time: 0.0022 memory: 9465 grad_norm: 0.0238 loss: 0.6797 +2023/06/01 13:50:23 - mmengine - INFO - Epoch(train) [15][ 3900/11516] lr: 2.1405e-04 eta: 7:34:59 time: 0.4110 data_time: 0.0013 memory: 9465 grad_norm: 0.0276 loss: 0.6873 +2023/06/01 13:51:03 - mmengine - INFO - Epoch(train) [15][ 4000/11516] lr: 2.1405e-04 eta: 7:34:16 time: 0.3845 data_time: 0.0012 memory: 9465 grad_norm: 0.0200 loss: 0.6842 +2023/06/01 13:51:45 - mmengine - INFO - Epoch(train) [15][ 4100/11516] lr: 2.1405e-04 eta: 7:33:35 time: 0.4241 data_time: 0.0013 memory: 9465 grad_norm: 0.0262 loss: 0.6882 +2023/06/01 13:52:26 - mmengine - INFO - Epoch(train) [15][ 4200/11516] lr: 2.1405e-04 eta: 7:32:53 time: 0.4012 data_time: 0.0013 memory: 9465 grad_norm: 0.0274 loss: 0.6869 +2023/06/01 13:53:08 - mmengine - INFO - Epoch(train) [15][ 4300/11516] lr: 2.1405e-04 eta: 7:32:10 time: 0.4263 data_time: 0.0013 memory: 9465 grad_norm: 0.0189 loss: 0.6868 +2023/06/01 13:53:49 - mmengine - INFO - Epoch(train) [15][ 4400/11516] lr: 2.1405e-04 eta: 7:31:28 time: 0.4086 data_time: 0.0013 memory: 9465 grad_norm: 0.0196 loss: 0.6909 +2023/06/01 13:54:30 - mmengine - INFO - Epoch(train) [15][ 4500/11516] lr: 2.1405e-04 eta: 7:30:46 time: 0.4184 data_time: 0.0015 memory: 9465 grad_norm: 0.0218 loss: 0.6859 +2023/06/01 13:55:12 - mmengine - INFO - Epoch(train) [15][ 4600/11516] lr: 2.1405e-04 eta: 7:30:04 time: 0.3847 data_time: 0.0018 memory: 9465 grad_norm: 0.0290 loss: 0.6882 +2023/06/01 13:55:54 - mmengine - INFO - Epoch(train) [15][ 4700/11516] lr: 2.1405e-04 eta: 7:29:22 time: 0.4169 data_time: 0.0019 memory: 9465 grad_norm: 0.0227 loss: 0.6879 +2023/06/01 13:56:24 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/06/01 13:56:36 - mmengine - INFO - Epoch(train) [15][ 4800/11516] lr: 2.1405e-04 eta: 7:28:40 time: 0.4234 data_time: 0.0013 memory: 9465 grad_norm: 0.0174 loss: 0.6829 +2023/06/01 13:57:16 - mmengine - INFO - Epoch(train) [15][ 4900/11516] lr: 2.1405e-04 eta: 7:27:58 time: 0.4094 data_time: 0.0013 memory: 9465 grad_norm: 0.0230 loss: 0.6796 +2023/06/01 13:57:58 - mmengine - INFO - Epoch(train) [15][ 5000/11516] lr: 2.1405e-04 eta: 7:27:16 time: 0.4009 data_time: 0.0012 memory: 9465 grad_norm: 0.0231 loss: 0.6875 +2023/06/01 13:58:40 - mmengine - INFO - Epoch(train) [15][ 5100/11516] lr: 2.1405e-04 eta: 7:26:34 time: 0.4467 data_time: 0.0021 memory: 9465 grad_norm: 0.0139 loss: 0.6854 +2023/06/01 13:59:21 - mmengine - INFO - Epoch(train) [15][ 5200/11516] lr: 2.1405e-04 eta: 7:25:52 time: 0.4396 data_time: 0.0015 memory: 9465 grad_norm: 0.0192 loss: 0.6847 +2023/06/01 14:00:03 - mmengine - INFO - Epoch(train) [15][ 5300/11516] lr: 2.1405e-04 eta: 7:25:10 time: 0.4142 data_time: 0.0018 memory: 9465 grad_norm: 0.0177 loss: 0.6889 +2023/06/01 14:00:44 - mmengine - INFO - Epoch(train) [15][ 5400/11516] lr: 2.1405e-04 eta: 7:24:28 time: 0.4275 data_time: 0.0025 memory: 9465 grad_norm: 0.0176 loss: 0.6884 +2023/06/01 14:01:24 - mmengine - INFO - Epoch(train) [15][ 5500/11516] lr: 2.1405e-04 eta: 7:23:46 time: 0.4032 data_time: 0.0016 memory: 9465 grad_norm: 0.0312 loss: 0.6915 +2023/06/01 14:02:05 - mmengine - INFO - Epoch(train) [15][ 5600/11516] lr: 2.1405e-04 eta: 7:23:03 time: 0.3975 data_time: 0.0013 memory: 9465 grad_norm: 0.0211 loss: 0.6834 +2023/06/01 14:02:46 - mmengine - INFO - Epoch(train) [15][ 5700/11516] lr: 2.1405e-04 eta: 7:22:21 time: 0.3933 data_time: 0.0018 memory: 9465 grad_norm: 0.0133 loss: 0.6885 +2023/06/01 14:03:16 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/06/01 14:03:28 - mmengine - INFO - Epoch(train) [15][ 5800/11516] lr: 2.1405e-04 eta: 7:21:39 time: 0.4238 data_time: 0.0013 memory: 9465 grad_norm: 0.0254 loss: 0.6860 +2023/06/01 14:04:09 - mmengine - INFO - Epoch(train) [15][ 5900/11516] lr: 2.1405e-04 eta: 7:20:57 time: 0.4130 data_time: 0.0013 memory: 9465 grad_norm: 0.0206 loss: 0.6824 +2023/06/01 14:04:50 - mmengine - INFO - Epoch(train) [15][ 6000/11516] lr: 2.1405e-04 eta: 7:20:15 time: 0.4086 data_time: 0.0016 memory: 9465 grad_norm: 0.0253 loss: 0.6865 +2023/06/01 14:05:31 - mmengine - INFO - Epoch(train) [15][ 6100/11516] lr: 2.1405e-04 eta: 7:19:33 time: 0.3938 data_time: 0.0013 memory: 9465 grad_norm: 0.0127 loss: 0.6860 +2023/06/01 14:06:14 - mmengine - INFO - Epoch(train) [15][ 6200/11516] lr: 2.1405e-04 eta: 7:18:51 time: 0.4315 data_time: 0.0013 memory: 9465 grad_norm: 0.0169 loss: 0.6786 +2023/06/01 14:06:56 - mmengine - INFO - Epoch(train) [15][ 6300/11516] lr: 2.1405e-04 eta: 7:18:09 time: 0.4063 data_time: 0.0014 memory: 9465 grad_norm: 0.0223 loss: 0.6852 +2023/06/01 14:07:38 - mmengine - INFO - Epoch(train) [15][ 6400/11516] lr: 2.1405e-04 eta: 7:17:28 time: 0.4136 data_time: 0.0026 memory: 9465 grad_norm: 0.0270 loss: 0.6816 +2023/06/01 14:08:19 - mmengine - INFO - Epoch(train) [15][ 6500/11516] lr: 2.1405e-04 eta: 7:16:45 time: 0.3838 data_time: 0.0025 memory: 9465 grad_norm: 0.0223 loss: 0.6875 +2023/06/01 14:09:01 - mmengine - INFO - Epoch(train) [15][ 6600/11516] lr: 2.1405e-04 eta: 7:16:04 time: 0.4113 data_time: 0.0017 memory: 9465 grad_norm: 0.0246 loss: 0.6911 +2023/06/01 14:09:44 - mmengine - INFO - Epoch(train) [15][ 6700/11516] lr: 2.1405e-04 eta: 7:15:22 time: 0.4557 data_time: 0.0015 memory: 9465 grad_norm: 0.0207 loss: 0.6877 +2023/06/01 14:10:14 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/06/01 14:10:26 - mmengine - INFO - Epoch(train) [15][ 6800/11516] lr: 2.1405e-04 eta: 7:14:40 time: 0.4201 data_time: 0.0016 memory: 9465 grad_norm: 0.0124 loss: 0.6887 +2023/06/01 14:11:07 - mmengine - INFO - Epoch(train) [15][ 6900/11516] lr: 2.1405e-04 eta: 7:13:58 time: 0.4085 data_time: 0.0027 memory: 9465 grad_norm: 0.0254 loss: 0.6862 +2023/06/01 14:11:49 - mmengine - INFO - Epoch(train) [15][ 7000/11516] lr: 2.1405e-04 eta: 7:13:16 time: 0.4133 data_time: 0.0017 memory: 9465 grad_norm: 0.0239 loss: 0.6874 +2023/06/01 14:12:30 - mmengine - INFO - Epoch(train) [15][ 7100/11516] lr: 2.1405e-04 eta: 7:12:34 time: 0.3964 data_time: 0.0021 memory: 9465 grad_norm: 0.0230 loss: 0.6873 +2023/06/01 14:13:12 - mmengine - INFO - Epoch(train) [15][ 7200/11516] lr: 2.1405e-04 eta: 7:11:52 time: 0.4126 data_time: 0.0013 memory: 9465 grad_norm: 0.0283 loss: 0.6857 +2023/06/01 14:13:54 - mmengine - INFO - Epoch(train) [15][ 7300/11516] lr: 2.1405e-04 eta: 7:11:10 time: 0.4129 data_time: 0.0013 memory: 9465 grad_norm: 0.0303 loss: 0.6909 +2023/06/01 14:14:37 - mmengine - INFO - Epoch(train) [15][ 7400/11516] lr: 2.1405e-04 eta: 7:10:29 time: 0.4128 data_time: 0.0018 memory: 9465 grad_norm: 0.0265 loss: 0.6908 +2023/06/01 14:15:20 - mmengine - INFO - Epoch(train) [15][ 7500/11516] lr: 2.1405e-04 eta: 7:09:47 time: 0.4256 data_time: 0.0012 memory: 9465 grad_norm: 0.0222 loss: 0.6894 +2023/06/01 14:16:01 - mmengine - INFO - Epoch(train) [15][ 7600/11516] lr: 2.1405e-04 eta: 7:09:05 time: 0.4206 data_time: 0.0013 memory: 9465 grad_norm: 0.0252 loss: 0.6874 +2023/06/01 14:16:43 - mmengine - INFO - Epoch(train) [15][ 7700/11516] lr: 2.1405e-04 eta: 7:08:24 time: 0.4089 data_time: 0.0014 memory: 9465 grad_norm: 0.0272 loss: 0.6858 +2023/06/01 14:17:14 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/06/01 14:17:26 - mmengine - INFO - Epoch(train) [15][ 7800/11516] lr: 2.1405e-04 eta: 7:07:42 time: 0.4542 data_time: 0.0013 memory: 9465 grad_norm: 0.0161 loss: 0.6830 +2023/06/01 14:18:08 - mmengine - INFO - Epoch(train) [15][ 7900/11516] lr: 2.1405e-04 eta: 7:07:00 time: 0.4459 data_time: 0.0019 memory: 9465 grad_norm: 0.0160 loss: 0.6846 +2023/06/01 14:18:51 - mmengine - INFO - Epoch(train) [15][ 8000/11516] lr: 2.1405e-04 eta: 7:06:19 time: 0.4340 data_time: 0.0013 memory: 9465 grad_norm: 0.0207 loss: 0.6892 +2023/06/01 14:19:33 - mmengine - INFO - Epoch(train) [15][ 8100/11516] lr: 2.1405e-04 eta: 7:05:37 time: 0.3913 data_time: 0.0014 memory: 9465 grad_norm: 0.0146 loss: 0.6913 +2023/06/01 14:20:15 - mmengine - INFO - Epoch(train) [15][ 8200/11516] lr: 2.1405e-04 eta: 7:04:55 time: 0.4066 data_time: 0.0014 memory: 9465 grad_norm: 0.0234 loss: 0.6863 +2023/06/01 14:20:58 - mmengine - INFO - Epoch(train) [15][ 8300/11516] lr: 2.1405e-04 eta: 7:04:14 time: 0.4317 data_time: 0.0015 memory: 9465 grad_norm: 0.0188 loss: 0.6831 +2023/06/01 14:21:40 - mmengine - INFO - Epoch(train) [15][ 8400/11516] lr: 2.1405e-04 eta: 7:03:32 time: 0.4507 data_time: 0.0013 memory: 9465 grad_norm: 0.0321 loss: 0.6876 +2023/06/01 14:22:24 - mmengine - INFO - Epoch(train) [15][ 8500/11516] lr: 2.1405e-04 eta: 7:02:51 time: 0.4193 data_time: 0.0015 memory: 9465 grad_norm: 0.0247 loss: 0.6878 +2023/06/01 14:23:07 - mmengine - INFO - Epoch(train) [15][ 8600/11516] lr: 2.1405e-04 eta: 7:02:09 time: 0.4202 data_time: 0.0014 memory: 9465 grad_norm: 0.0213 loss: 0.6850 +2023/06/01 14:23:48 - mmengine - INFO - Epoch(train) [15][ 8700/11516] lr: 2.1405e-04 eta: 7:01:27 time: 0.4181 data_time: 0.0016 memory: 9465 grad_norm: 0.0270 loss: 0.6882 +2023/06/01 14:24:19 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/06/01 14:24:30 - mmengine - INFO - Epoch(train) [15][ 8800/11516] lr: 2.1405e-04 eta: 7:00:45 time: 0.4283 data_time: 0.0013 memory: 9465 grad_norm: 0.0223 loss: 0.6908 +2023/06/01 14:25:12 - mmengine - INFO - Epoch(train) [15][ 8900/11516] lr: 2.1405e-04 eta: 7:00:03 time: 0.4130 data_time: 0.0013 memory: 9465 grad_norm: 0.0217 loss: 0.6860 +2023/06/01 14:25:53 - mmengine - INFO - Epoch(train) [15][ 9000/11516] lr: 2.1405e-04 eta: 6:59:21 time: 0.4097 data_time: 0.0014 memory: 9465 grad_norm: 0.0248 loss: 0.6855 +2023/06/01 14:26:36 - mmengine - INFO - Epoch(train) [15][ 9100/11516] lr: 2.1405e-04 eta: 6:58:39 time: 0.4275 data_time: 0.0013 memory: 9465 grad_norm: 0.0244 loss: 0.6865 +2023/06/01 14:27:18 - mmengine - INFO - Epoch(train) [15][ 9200/11516] lr: 2.1405e-04 eta: 6:57:58 time: 0.4240 data_time: 0.0014 memory: 9465 grad_norm: 0.0246 loss: 0.6870 +2023/06/01 14:28:01 - mmengine - INFO - Epoch(train) [15][ 9300/11516] lr: 2.1405e-04 eta: 6:57:16 time: 0.4308 data_time: 0.0019 memory: 9465 grad_norm: 0.0119 loss: 0.6788 +2023/06/01 14:28:45 - mmengine - INFO - Epoch(train) [15][ 9400/11516] lr: 2.1405e-04 eta: 6:56:35 time: 0.4181 data_time: 0.0013 memory: 9465 grad_norm: 0.0249 loss: 0.6833 +2023/06/01 14:29:29 - mmengine - INFO - Epoch(train) [15][ 9500/11516] lr: 2.1405e-04 eta: 6:55:54 time: 0.4272 data_time: 0.0014 memory: 9465 grad_norm: 0.0111 loss: 0.6902 +2023/06/01 14:30:14 - mmengine - INFO - Epoch(train) [15][ 9600/11516] lr: 2.1405e-04 eta: 6:55:13 time: 0.4708 data_time: 0.0016 memory: 9465 grad_norm: 0.0204 loss: 0.6868 +2023/06/01 14:30:58 - mmengine - INFO - Epoch(train) [15][ 9700/11516] lr: 2.1405e-04 eta: 6:54:32 time: 0.4309 data_time: 0.0018 memory: 9465 grad_norm: 0.0280 loss: 0.6947 +2023/06/01 14:31:30 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/06/01 14:31:42 - mmengine - INFO - Epoch(train) [15][ 9800/11516] lr: 2.1405e-04 eta: 6:53:51 time: 0.4315 data_time: 0.0018 memory: 9465 grad_norm: 0.0215 loss: 0.6858 +2023/06/01 14:32:26 - mmengine - INFO - Epoch(train) [15][ 9900/11516] lr: 2.1405e-04 eta: 6:53:10 time: 0.4322 data_time: 0.0014 memory: 9465 grad_norm: 0.0195 loss: 0.6880 +2023/06/01 14:33:10 - mmengine - INFO - Epoch(train) [15][10000/11516] lr: 2.1405e-04 eta: 6:52:29 time: 0.4360 data_time: 0.0013 memory: 9465 grad_norm: 0.0181 loss: 0.6844 +2023/06/01 14:33:55 - mmengine - INFO - Epoch(train) [15][10100/11516] lr: 2.1405e-04 eta: 6:51:48 time: 0.4730 data_time: 0.0020 memory: 9465 grad_norm: 0.0179 loss: 0.6865 +2023/06/01 14:34:39 - mmengine - INFO - Epoch(train) [15][10200/11516] lr: 2.1405e-04 eta: 6:51:07 time: 0.4528 data_time: 0.0012 memory: 9465 grad_norm: 0.0222 loss: 0.6863 +2023/06/01 14:35:25 - mmengine - INFO - Epoch(train) [15][10300/11516] lr: 2.1405e-04 eta: 6:50:26 time: 0.4340 data_time: 0.0014 memory: 9465 grad_norm: 0.0166 loss: 0.6862 +2023/06/01 14:36:09 - mmengine - INFO - Epoch(train) [15][10400/11516] lr: 2.1405e-04 eta: 6:49:45 time: 0.4403 data_time: 0.0017 memory: 9465 grad_norm: 0.0224 loss: 0.6857 +2023/06/01 14:36:53 - mmengine - INFO - Epoch(train) [15][10500/11516] lr: 2.1405e-04 eta: 6:49:04 time: 0.4194 data_time: 0.0013 memory: 9465 grad_norm: 0.0188 loss: 0.6884 +2023/06/01 14:37:35 - mmengine - INFO - Epoch(train) [15][10600/11516] lr: 2.1405e-04 eta: 6:48:22 time: 0.4083 data_time: 0.0012 memory: 9465 grad_norm: 0.0212 loss: 0.6873 +2023/06/01 14:38:17 - mmengine - INFO - Epoch(train) [15][10700/11516] lr: 2.1405e-04 eta: 6:47:40 time: 0.4567 data_time: 0.0014 memory: 9465 grad_norm: 0.0246 loss: 0.6844 +2023/06/01 14:38:47 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/06/01 14:38:58 - mmengine - INFO - Epoch(train) [15][10800/11516] lr: 2.1405e-04 eta: 6:46:58 time: 0.4276 data_time: 0.0013 memory: 9465 grad_norm: 0.0172 loss: 0.6858 +2023/06/01 14:39:41 - mmengine - INFO - Epoch(train) [15][10900/11516] lr: 2.1405e-04 eta: 6:46:16 time: 0.4261 data_time: 0.0012 memory: 9465 grad_norm: 0.0190 loss: 0.6892 +2023/06/01 14:40:23 - mmengine - INFO - Epoch(train) [15][11000/11516] lr: 2.1405e-04 eta: 6:45:35 time: 0.4097 data_time: 0.0015 memory: 9465 grad_norm: 0.0264 loss: 0.6917 +2023/06/01 14:41:05 - mmengine - INFO - Epoch(train) [15][11100/11516] lr: 2.1405e-04 eta: 6:44:53 time: 0.4211 data_time: 0.0014 memory: 9465 grad_norm: 0.0215 loss: 0.6943 +2023/06/01 14:41:48 - mmengine - INFO - Epoch(train) [15][11200/11516] lr: 2.1405e-04 eta: 6:44:11 time: 0.4232 data_time: 0.0013 memory: 9465 grad_norm: 0.0186 loss: 0.6830 +2023/06/01 14:42:29 - mmengine - INFO - Epoch(train) [15][11300/11516] lr: 2.1405e-04 eta: 6:43:29 time: 0.3965 data_time: 0.0012 memory: 9465 grad_norm: 0.0251 loss: 0.6889 +2023/06/01 14:43:10 - mmengine - INFO - Epoch(train) [15][11400/11516] lr: 2.1405e-04 eta: 6:42:47 time: 0.4225 data_time: 0.0014 memory: 9465 grad_norm: 0.0318 loss: 0.6835 +2023/06/01 14:43:52 - mmengine - INFO - Epoch(train) [15][11500/11516] lr: 2.1405e-04 eta: 6:42:05 time: 0.4219 data_time: 0.0051 memory: 9465 grad_norm: 0.0249 loss: 0.6878 +2023/06/01 14:43:58 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/06/01 14:43:58 - mmengine - INFO - Saving checkpoint at 15 epochs +2023/06/01 14:44:15 - mmengine - INFO - Epoch(val) [15][16/16] accuracy/top1: 100.0000 single-label/precision_classwise: [100.0, 0.0] single-label/recall_classwise: [100.0, 0.0] single-label/f1-score_classwise: [100.0, 0.0] data_time: 0.2774 time: 0.5408 +2023/06/01 14:44:57 - mmengine - INFO - Epoch(train) [16][ 100/11516] lr: 1.5498e-04 eta: 6:41:16 time: 0.4284 data_time: 0.0480 memory: 9465 grad_norm: 0.0124 loss: 0.6874 +2023/06/01 14:45:38 - mmengine - INFO - Epoch(train) [16][ 200/11516] lr: 1.5498e-04 eta: 6:40:34 time: 0.4147 data_time: 0.0014 memory: 9465 grad_norm: 0.0265 loss: 0.6849 +2023/06/01 14:46:03 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/06/01 14:46:20 - mmengine - INFO - Epoch(train) [16][ 300/11516] lr: 1.5498e-04 eta: 6:39:52 time: 0.4386 data_time: 0.0013 memory: 9465 grad_norm: 0.0213 loss: 0.6919 +2023/06/01 14:47:02 - mmengine - INFO - Epoch(train) [16][ 400/11516] lr: 1.5498e-04 eta: 6:39:10 time: 0.4209 data_time: 0.0023 memory: 9465 grad_norm: 0.0168 loss: 0.6860 +2023/06/01 14:47:43 - mmengine - INFO - Epoch(train) [16][ 500/11516] lr: 1.5498e-04 eta: 6:38:28 time: 0.4088 data_time: 0.0016 memory: 9465 grad_norm: 0.0209 loss: 0.6868 +2023/06/01 14:48:25 - mmengine - INFO - Epoch(train) [16][ 600/11516] lr: 1.5498e-04 eta: 6:37:46 time: 0.4329 data_time: 0.0014 memory: 9465 grad_norm: 0.0318 loss: 0.6909 +2023/06/01 14:49:08 - mmengine - INFO - Epoch(train) [16][ 700/11516] lr: 1.5498e-04 eta: 6:37:04 time: 0.4691 data_time: 0.0016 memory: 9465 grad_norm: 0.0319 loss: 0.6856 +2023/06/01 14:49:50 - mmengine - INFO - Epoch(train) [16][ 800/11516] lr: 1.5498e-04 eta: 6:36:23 time: 0.4403 data_time: 0.0013 memory: 9465 grad_norm: 0.0258 loss: 0.6870 +2023/06/01 14:50:31 - mmengine - INFO - Epoch(train) [16][ 900/11516] lr: 1.5498e-04 eta: 6:35:41 time: 0.4154 data_time: 0.0020 memory: 9465 grad_norm: 0.0185 loss: 0.6882 +2023/06/01 14:51:12 - mmengine - INFO - Epoch(train) [16][ 1000/11516] lr: 1.5498e-04 eta: 6:34:58 time: 0.3884 data_time: 0.0016 memory: 9465 grad_norm: 0.0294 loss: 0.6844 +2023/06/01 14:51:51 - mmengine - INFO - Epoch(train) [16][ 1100/11516] lr: 1.5498e-04 eta: 6:34:16 time: 0.3579 data_time: 0.0013 memory: 9465 grad_norm: 0.0243 loss: 0.6870 +2023/06/01 14:52:31 - mmengine - INFO - Epoch(train) [16][ 1200/11516] lr: 1.5498e-04 eta: 6:33:33 time: 0.3797 data_time: 0.0015 memory: 9465 grad_norm: 0.0254 loss: 0.6884 +2023/06/01 14:52:56 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/06/01 14:53:11 - mmengine - INFO - Epoch(train) [16][ 1300/11516] lr: 1.5498e-04 eta: 6:32:50 time: 0.3753 data_time: 0.0012 memory: 9465 grad_norm: 0.0183 loss: 0.6886 +2023/06/01 14:53:51 - mmengine - INFO - Epoch(train) [16][ 1400/11516] lr: 1.5498e-04 eta: 6:32:08 time: 0.4132 data_time: 0.0011 memory: 9465 grad_norm: 0.0161 loss: 0.6888 +2023/06/01 14:54:30 - mmengine - INFO - Epoch(train) [16][ 1500/11516] lr: 1.5498e-04 eta: 6:31:25 time: 0.3690 data_time: 0.0015 memory: 9465 grad_norm: 0.0228 loss: 0.6871 +2023/06/01 14:55:10 - mmengine - INFO - Epoch(train) [16][ 1600/11516] lr: 1.5498e-04 eta: 6:30:43 time: 0.4115 data_time: 0.0019 memory: 9465 grad_norm: 0.0191 loss: 0.6839 +2023/06/01 14:55:51 - mmengine - INFO - Epoch(train) [16][ 1700/11516] lr: 1.5498e-04 eta: 6:30:00 time: 0.3962 data_time: 0.0015 memory: 9465 grad_norm: 0.0154 loss: 0.6843 +2023/06/01 14:56:30 - mmengine - INFO - Epoch(train) [16][ 1800/11516] lr: 1.5498e-04 eta: 6:29:18 time: 0.4263 data_time: 0.0015 memory: 9465 grad_norm: 0.0259 loss: 0.6854 +2023/06/01 14:57:10 - mmengine - INFO - Epoch(train) [16][ 1900/11516] lr: 1.5498e-04 eta: 6:28:35 time: 0.3768 data_time: 0.0013 memory: 9465 grad_norm: 0.0165 loss: 0.6849 +2023/06/01 14:57:50 - mmengine - INFO - Epoch(train) [16][ 2000/11516] lr: 1.5498e-04 eta: 6:27:53 time: 0.3816 data_time: 0.0014 memory: 9465 grad_norm: 0.0192 loss: 0.6876 +2023/06/01 14:58:31 - mmengine - INFO - Epoch(train) [16][ 2100/11516] lr: 1.5498e-04 eta: 6:27:11 time: 0.4103 data_time: 0.0012 memory: 9465 grad_norm: 0.0197 loss: 0.6892 +2023/06/01 14:59:11 - mmengine - INFO - Epoch(train) [16][ 2200/11516] lr: 1.5498e-04 eta: 6:26:28 time: 0.3904 data_time: 0.0015 memory: 9465 grad_norm: 0.0321 loss: 0.6810 +2023/06/01 14:59:35 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/06/01 14:59:52 - mmengine - INFO - Epoch(train) [16][ 2300/11516] lr: 1.5498e-04 eta: 6:25:46 time: 0.4653 data_time: 0.0020 memory: 9465 grad_norm: 0.0235 loss: 0.6893 +2023/06/01 15:00:33 - mmengine - INFO - Epoch(train) [16][ 2400/11516] lr: 1.5498e-04 eta: 6:25:04 time: 0.3933 data_time: 0.0019 memory: 9465 grad_norm: 0.0256 loss: 0.6807 +2023/06/01 15:01:13 - mmengine - INFO - Epoch(train) [16][ 2500/11516] lr: 1.5498e-04 eta: 6:24:21 time: 0.4022 data_time: 0.0022 memory: 9465 grad_norm: 0.0191 loss: 0.6858 +2023/06/01 15:01:52 - mmengine - INFO - Epoch(train) [16][ 2600/11516] lr: 1.5498e-04 eta: 6:23:39 time: 0.3855 data_time: 0.0021 memory: 9465 grad_norm: 0.0184 loss: 0.6855 +2023/06/01 15:02:31 - mmengine - INFO - Epoch(train) [16][ 2700/11516] lr: 1.5498e-04 eta: 6:22:56 time: 0.3946 data_time: 0.0019 memory: 9465 grad_norm: 0.0212 loss: 0.6815 +2023/06/01 15:03:10 - mmengine - INFO - Epoch(train) [16][ 2800/11516] lr: 1.5498e-04 eta: 6:22:13 time: 0.3887 data_time: 0.0016 memory: 9465 grad_norm: 0.0229 loss: 0.6902 +2023/06/01 15:03:49 - mmengine - INFO - Epoch(train) [16][ 2900/11516] lr: 1.5498e-04 eta: 6:21:30 time: 0.3687 data_time: 0.0020 memory: 9465 grad_norm: 0.0240 loss: 0.6901 +2023/06/01 15:04:28 - mmengine - INFO - Epoch(train) [16][ 3000/11516] lr: 1.5498e-04 eta: 6:20:48 time: 0.3842 data_time: 0.0019 memory: 9465 grad_norm: 0.0249 loss: 0.6886 +2023/06/01 15:05:09 - mmengine - INFO - Epoch(train) [16][ 3100/11516] lr: 1.5498e-04 eta: 6:20:05 time: 0.4017 data_time: 0.0014 memory: 9465 grad_norm: 0.0305 loss: 0.6847 +2023/06/01 15:05:49 - mmengine - INFO - Epoch(train) [16][ 3200/11516] lr: 1.5498e-04 eta: 6:19:23 time: 0.4245 data_time: 0.0013 memory: 9465 grad_norm: 0.0307 loss: 0.6837 +2023/06/01 15:06:13 - mmengine - INFO - Exp name: swin_tiny_4xb256_fake5m_20230531_183311 +2023/06/01 15:06:29 - mmengine - INFO - Epoch(train) [16][ 3300/11516] lr: 1.5498e-04 eta: 6:18:40 time: 0.3820 data_time: 0.0016 memory: 9465 grad_norm: 0.0215 loss: 0.6821 +2023/06/01 15:07:09 - mmengine - INFO - Epoch(train) [16][ 3400/11516] lr: 1.5498e-04 eta: 6:17:58 time: 0.3769 data_time: 0.0023 memory: 9465 grad_norm: 0.0270 loss: 0.6868 +2023/06/01 15:07:48 - mmengine - INFO - Epoch(train) [16][ 3500/11516] lr: 1.5498e-04 eta: 6:17:15 time: 0.3977 data_time: 0.0013 memory: 9465 grad_norm: 0.0186 loss: 0.6899 +2023/06/01 15:08:27 - mmengine - INFO - Epoch(train) [16][ 3600/11516] lr: 1.5498e-04 eta: 6:16:32 time: 0.3873 data_time: 0.0014 memory: 9465 grad_norm: 0.0210 loss: 0.6854 +2023/06/01 15:09:05 - mmengine - INFO - Epoch(train) [16][ 3700/11516] lr: 1.5498e-04 eta: 6:15:50 time: 0.3739 data_time: 0.0014 memory: 9465 grad_norm: 0.0268 loss: 0.6820 diff --git a/swin_tiny_4xb256_1e-3lr_5m/20230531_183311/vis_data/20230531_183311.json b/swin_tiny_4xb256_1e-3lr_5m/20230531_183311/vis_data/20230531_183311.json new file mode 100644 index 0000000000000000000000000000000000000000..6d13eee61c9532cba47a6e8dc19473529cca92a2 --- /dev/null +++ b/swin_tiny_4xb256_1e-3lr_5m/20230531_183311/vis_data/20230531_183311.json @@ -0,0 +1,1777 @@ +{"lr": 0.001, "data_time": 0.19160523414611816, "grad_norm": 0.8052142232656478, "loss": 0.6849520623683929, "time": 0.42688555717468263, "epoch": 1, "memory": 9474, "step": 100} +{"lr": 0.001, "data_time": 0.0009674787521362305, "grad_norm": 1.2080376029014588, "loss": 0.6878873109817505, "time": 0.4383848190307617, "epoch": 1, "memory": 9474, "step": 200} +{"lr": 0.001, "data_time": 0.001109480857849121, "grad_norm": 1.2378409653902054, "loss": 0.6797099351882935, "time": 0.4772242307662964, "epoch": 1, "memory": 9474, "step": 300} +{"lr": 0.001, "data_time": 0.001096630096435547, "grad_norm": 0.7535933956503869, "loss": 0.6639681220054626, "time": 0.4896991729736328, "epoch": 1, "memory": 9474, "step": 400} +{"lr": 0.001, "data_time": 0.0010877132415771484, "grad_norm": 2.427699086070061, "loss": 0.672640472650528, "time": 0.49225332736968996, "epoch": 1, "memory": 9474, "step": 500} +{"lr": 0.001, "data_time": 0.0009778976440429688, "grad_norm": 0.8477599889039993, "loss": 0.6810718059539795, "time": 0.516904878616333, "epoch": 1, "memory": 9474, "step": 600} +{"lr": 0.001, "data_time": 0.001172924041748047, "grad_norm": 0.7502188369631767, "loss": 0.6687590122222901, "time": 0.48300533294677733, "epoch": 1, "memory": 9474, "step": 700} +{"lr": 0.001, "data_time": 0.0010997772216796875, "grad_norm": 0.6409901082515717, "loss": 0.6691092848777771, "time": 0.4418308973312378, "epoch": 1, "memory": 9474, "step": 800} +{"lr": 0.001, "data_time": 0.0010256052017211914, "grad_norm": 1.9922522366046906, "loss": 0.6726115763187408, "time": 0.42144126892089845, "epoch": 1, "memory": 9474, "step": 900} +{"lr": 0.001, "data_time": 0.001267242431640625, "grad_norm": 0.9892641425132751, "loss": 0.6661223530769348, "time": 0.4637609958648682, "epoch": 1, "memory": 9474, "step": 1000} +{"lr": 0.001, "data_time": 0.0010303974151611329, "grad_norm": 8.305355420708656, "loss": 0.6419571578502655, "time": 0.46142849922180174, "epoch": 1, "memory": 9474, "step": 1100} +{"lr": 0.001, "data_time": 0.0010545492172241212, "grad_norm": 0.792270827293396, "loss": 0.6670249342918396, "time": 0.43836815357208253, "epoch": 1, "memory": 9474, "step": 1200} +{"lr": 0.001, "data_time": 0.001044297218322754, "grad_norm": 0.6781573936343193, "loss": 0.6480621933937073, "time": 0.38781630992889404, "epoch": 1, "memory": 9474, "step": 1300} +{"lr": 0.001, "data_time": 0.0010371446609497071, "grad_norm": 0.5881157577037811, "loss": 0.6777416348457337, "time": 0.46243746280670167, "epoch": 1, "memory": 9474, "step": 1400} +{"lr": 0.001, "data_time": 0.0010274648666381836, "grad_norm": 0.5483857661485672, "loss": 0.6642019152641296, "time": 0.4059020519256592, "epoch": 1, "memory": 9474, "step": 1500} +{"lr": 0.001, "data_time": 0.0010054826736450196, "grad_norm": 0.5723021060228348, "loss": 0.6534375131130219, "time": 0.4653234243392944, "epoch": 1, "memory": 9474, "step": 1600} +{"lr": 0.001, "data_time": 0.0009948253631591798, "grad_norm": 1.8854305177927018, "loss": 0.6609666228294373, "time": 0.4095946788787842, "epoch": 1, "memory": 9474, "step": 1700} +{"lr": 0.001, "data_time": 0.001050853729248047, "grad_norm": 3.300791251659393, "loss": 0.6572105407714843, "time": 0.516826057434082, "epoch": 1, "memory": 9474, "step": 1800} +{"lr": 0.001, "data_time": 0.0009797334671020508, "grad_norm": 3.2456099927425384, "loss": 0.6459945976734162, "time": 0.44674875736236574, "epoch": 1, "memory": 9474, "step": 1900} +{"lr": 0.001, "data_time": 0.0009960412979125976, "grad_norm": 2.9178244560956954, "loss": 0.6701214969158172, "time": 0.48870298862457273, "epoch": 1, "memory": 9474, "step": 2000} +{"lr": 0.001, "data_time": 0.0010712385177612306, "grad_norm": 24.649349164962768, "loss": 0.6704214990139008, "time": 0.6039324760437011, "epoch": 1, "memory": 9474, "step": 2100} +{"lr": 0.001, "data_time": 0.0009820938110351562, "grad_norm": 132.87631206512452, "loss": 0.6618600070476532, "time": 0.5378271341323853, "epoch": 1, "memory": 9474, "step": 2200} +{"lr": 0.001, "data_time": 0.0009410858154296875, "grad_norm": 0.5663121178746223, "loss": 0.6898861110210419, "time": 0.4891343355178833, "epoch": 1, "memory": 9474, "step": 2300} +{"lr": 0.001, "data_time": 0.0009967803955078125, "grad_norm": 0.48270824924111366, "loss": 0.6908625543117524, "time": 0.5455003023147583, "epoch": 1, "memory": 9474, "step": 2400} +{"lr": 0.001, "data_time": 0.0009343624114990234, "grad_norm": 0.22726595103740693, "loss": 0.6804853200912475, "time": 0.4412868022918701, "epoch": 1, "memory": 9474, "step": 2500} +{"lr": 0.001, "data_time": 0.0009561061859130859, "grad_norm": 0.4640265181660652, "loss": 0.6885727345943451, "time": 0.45170691013336184, "epoch": 1, "memory": 9474, "step": 2600} +{"lr": 0.001, "data_time": 0.0010427713394165039, "grad_norm": 0.42788181230425837, "loss": 0.6879625499248505, "time": 0.42471895217895506, "epoch": 1, "memory": 9474, "step": 2700} +{"lr": 0.001, "data_time": 0.0010392904281616212, "grad_norm": 0.5726554363965988, "loss": 0.6858113765716553, "time": 0.48532853126525877, "epoch": 1, "memory": 9474, "step": 2800} +{"lr": 0.001, "data_time": 0.0009303569793701171, "grad_norm": 0.40351647436618804, "loss": 0.6893806636333466, "time": 0.521523904800415, "epoch": 1, "memory": 9474, "step": 2900} +{"lr": 0.001, "data_time": 0.0011853218078613282, "grad_norm": 0.16025364845991136, "loss": 0.6835512280464172, "time": 0.4506367683410645, "epoch": 1, "memory": 9474, "step": 3000} +{"lr": 0.001, "data_time": 0.0008364439010620118, "grad_norm": 0.24458550326526166, "loss": 0.6849212646484375, "time": 0.5652475595474243, "epoch": 1, "memory": 9474, "step": 3100} +{"lr": 0.001, "data_time": 0.0009660482406616211, "grad_norm": 0.44269598126411436, "loss": 0.6895747065544129, "time": 0.48994646072387693, "epoch": 1, "memory": 9474, "step": 3200} +{"lr": 0.001, "data_time": 0.0009510278701782227, "grad_norm": 0.24424223899841307, "loss": 0.6918225228786469, "time": 0.47918052673339845, "epoch": 1, "memory": 9474, "step": 3300} +{"lr": 0.001, "data_time": 0.0009103775024414063, "grad_norm": 0.29042947441339495, "loss": 0.6890306532382965, "time": 0.5593473196029664, "epoch": 1, "memory": 9474, "step": 3400} +{"lr": 0.001, "data_time": 0.0010324478149414062, "grad_norm": 0.2884215623140335, "loss": 0.6909060895442962, "time": 0.4397998809814453, "epoch": 1, "memory": 9474, "step": 3500} +{"lr": 0.001, "data_time": 0.000988173484802246, "grad_norm": 0.3512984409928322, "loss": 0.6868723154067993, "time": 0.41014585494995115, "epoch": 1, "memory": 9474, "step": 3600} +{"lr": 0.001, "data_time": 0.001079702377319336, "grad_norm": 0.46717650145292283, "loss": 0.6889779210090637, "time": 0.48719546794891355, "epoch": 1, "memory": 9474, "step": 3700} +{"lr": 0.001, "data_time": 0.0009716510772705078, "grad_norm": 0.2242715172469616, "loss": 0.692361181974411, "time": 0.46064176559448244, "epoch": 1, "memory": 9474, "step": 3800} +{"lr": 0.001, "data_time": 0.0009479999542236328, "grad_norm": 0.24586505629122257, "loss": 0.6896000802516937, "time": 0.4327003717422485, "epoch": 1, "memory": 9474, "step": 3900} +{"lr": 0.001, "data_time": 0.0009173393249511719, "grad_norm": 0.18979598507285117, "loss": 0.6918147563934326, "time": 0.5554824829101562, "epoch": 1, "memory": 9474, "step": 4000} +{"lr": 0.001, "data_time": 0.0009491443634033203, "grad_norm": 0.23377782851457596, "loss": 0.6852491140365601, "time": 0.499322247505188, "epoch": 1, "memory": 9474, "step": 4100} +{"lr": 0.001, "data_time": 0.001265573501586914, "grad_norm": 0.39092123657464983, "loss": 0.6906617701053619, "time": 0.45052011013031007, "epoch": 1, "memory": 9474, "step": 4200} +{"lr": 0.001, "data_time": 0.0009168148040771485, "grad_norm": 0.2585465915501118, "loss": 0.6853789627552033, "time": 0.4613243818283081, "epoch": 1, "memory": 9474, "step": 4300} +{"lr": 0.001, "data_time": 0.0011272668838500977, "grad_norm": 0.2267835909500718, "loss": 0.6889314770698547, "time": 0.46185941696166993, "epoch": 1, "memory": 9474, "step": 4400} +{"lr": 0.001, "data_time": 0.001326441764831543, "grad_norm": 0.19596254006028174, "loss": 0.6883785903453827, "time": 0.5613390207290649, "epoch": 1, "memory": 9474, "step": 4500} +{"lr": 0.001, "data_time": 0.00107266902923584, "grad_norm": 0.20727041438221933, "loss": 0.6849764704704284, "time": 0.4139907121658325, "epoch": 1, "memory": 9474, "step": 4600} +{"lr": 0.001, "data_time": 0.0011577367782592773, "grad_norm": 0.23133864887058736, "loss": 0.6821418702602386, "time": 0.4663443326950073, "epoch": 1, "memory": 9474, "step": 4700} +{"lr": 0.001, "data_time": 0.0009682178497314453, "grad_norm": 0.18215838652104138, "loss": 0.6857521176338196, "time": 0.47600295543670657, "epoch": 1, "memory": 9474, "step": 4800} +{"lr": 0.001, "data_time": 0.0009597539901733398, "grad_norm": 0.17026583850383759, "loss": 0.6810456097126008, "time": 0.5322046995162963, "epoch": 1, "memory": 9474, "step": 4900} +{"lr": 0.001, "data_time": 0.0009952545166015624, "grad_norm": 0.19457727447152137, "loss": 0.6891527593135833, "time": 0.4480944871902466, "epoch": 1, "memory": 9474, "step": 5000} +{"lr": 0.001, "data_time": 0.0010361194610595704, "grad_norm": 0.2026959676295519, "loss": 0.6932987153530121, "time": 0.5342537641525269, "epoch": 1, "memory": 9474, "step": 5100} +{"lr": 0.001, "data_time": 0.0010344743728637694, "grad_norm": 0.19610126949846746, "loss": 0.6837357580661774, "time": 0.5068094491958618, "epoch": 1, "memory": 9474, "step": 5200} +{"lr": 0.001, "data_time": 0.001011061668395996, "grad_norm": 0.21482629626989364, "loss": 0.6858660340309143, "time": 0.565664005279541, "epoch": 1, "memory": 9474, "step": 5300} +{"lr": 0.001, "data_time": 0.0009800195693969727, "grad_norm": 0.16851673275232315, "loss": 0.6887180030345916, "time": 0.4616303682327271, "epoch": 1, "memory": 9474, "step": 5400} +{"lr": 0.001, "data_time": 0.0010013818740844727, "grad_norm": 0.16333863157778977, "loss": 0.6855832993984222, "time": 0.42328202724456787, "epoch": 1, "memory": 9474, "step": 5500} +{"lr": 0.001, "data_time": 0.0010354280471801757, "grad_norm": 0.19897820930927992, "loss": 0.6884875357151031, "time": 0.47611050605773925, "epoch": 1, "memory": 9474, "step": 5600} +{"lr": 0.001, "data_time": 0.0010261058807373047, "grad_norm": 0.208826738409698, "loss": 0.6890342056751251, "time": 0.5138506174087525, "epoch": 1, "memory": 9474, "step": 5700} +{"lr": 0.001, "data_time": 0.0010219573974609374, "grad_norm": 0.12756846696138383, "loss": 0.6907239317893982, "time": 0.512112307548523, "epoch": 1, "memory": 9474, "step": 5800} +{"lr": 0.001, "data_time": 0.001003599166870117, "grad_norm": 0.22615315057337285, "loss": 0.693650609254837, "time": 0.4584911584854126, "epoch": 1, "memory": 9474, "step": 5900} +{"lr": 0.001, "data_time": 0.0009990692138671874, "grad_norm": 0.2532569320872426, "loss": 0.6887039959430694, "time": 0.43177931308746337, "epoch": 1, "memory": 9474, "step": 6000} +{"lr": 0.001, "data_time": 0.0009796857833862306, "grad_norm": 0.20852719470858574, "loss": 0.6801067411899566, "time": 0.415038013458252, "epoch": 1, "memory": 9474, "step": 6100} +{"lr": 0.001, "data_time": 0.0010406017303466798, "grad_norm": 0.22137072421610354, "loss": 0.6856032311916351, "time": 0.44321963787078855, "epoch": 1, "memory": 9474, "step": 6200} +{"lr": 0.001, "data_time": 0.0013756275177001954, "grad_norm": 0.14437412656843662, "loss": 0.6890648424625396, "time": 0.45041627883911134, "epoch": 1, "memory": 9474, "step": 6300} +{"lr": 0.001, "data_time": 0.0009638309478759765, "grad_norm": 0.11462613651528955, "loss": 0.6906584203243256, "time": 0.4679267406463623, "epoch": 1, "memory": 9474, "step": 6400} +{"lr": 0.001, "data_time": 0.0009244203567504883, "grad_norm": 0.18450075276196004, "loss": 0.683798748254776, "time": 0.4362278938293457, "epoch": 1, "memory": 9474, "step": 6500} +{"lr": 0.001, "data_time": 0.0010439395904541016, "grad_norm": 0.16256289929151535, "loss": 0.6911483645439148, "time": 0.4376434564590454, "epoch": 1, "memory": 9474, "step": 6600} +{"lr": 0.001, "data_time": 0.000995326042175293, "grad_norm": 0.16351005053147674, "loss": 0.6809832274913787, "time": 0.45746004581451416, "epoch": 1, "memory": 9474, "step": 6700} +{"lr": 0.001, "data_time": 0.0010692119598388673, "grad_norm": 0.16776065863668918, "loss": 0.6851230680942535, "time": 0.4315709352493286, "epoch": 1, "memory": 9474, "step": 6800} +{"lr": 0.001, "data_time": 0.0011238336563110351, "grad_norm": 0.18845577947795392, "loss": 0.6767730951309204, "time": 0.4457768678665161, "epoch": 1, "memory": 9474, "step": 6900} +{"lr": 0.001, "data_time": 0.001305699348449707, "grad_norm": 0.1759693343192339, "loss": 0.6886209487915039, "time": 0.4217818021774292, "epoch": 1, "memory": 9474, "step": 7000} +{"lr": 0.001, "data_time": 0.0010958433151245118, "grad_norm": 0.18210300225764514, "loss": 0.6890046715736389, "time": 0.5111133098602295, "epoch": 1, "memory": 9474, "step": 7100} +{"lr": 0.001, "data_time": 0.0012515783309936523, "grad_norm": 0.12574998512864113, "loss": 0.6871379852294922, "time": 0.4268744945526123, "epoch": 1, "memory": 9474, "step": 7200} +{"lr": 0.001, "data_time": 0.001128220558166504, "grad_norm": 0.16726765520870684, "loss": 0.6882547914981842, "time": 0.4326565027236938, "epoch": 1, "memory": 9474, "step": 7300} +{"lr": 0.001, "data_time": 0.0010713100433349609, "grad_norm": 0.199175663292408, "loss": 0.6863829076290131, "time": 0.4495775938034058, "epoch": 1, "memory": 9474, "step": 7400} +{"lr": 0.001, "data_time": 0.0010512113571166993, "grad_norm": 0.2270539177581668, "loss": 0.6883134067058563, "time": 0.42945754528045654, "epoch": 1, "memory": 9474, "step": 7500} +{"lr": 0.001, "data_time": 0.0009206295013427735, "grad_norm": 0.1085498021915555, "loss": 0.6849838256835937, "time": 0.4674868583679199, "epoch": 1, "memory": 9474, "step": 7600} +{"lr": 0.001, "data_time": 0.0010736703872680664, "grad_norm": 0.20984667725861073, "loss": 0.6875992119312286, "time": 0.41382741928100586, "epoch": 1, "memory": 9474, "step": 7700} +{"lr": 0.001, "data_time": 0.0010170459747314454, "grad_norm": 0.10837054271250963, "loss": 0.6887068629264832, "time": 0.45543999671936036, "epoch": 1, "memory": 9474, "step": 7800} +{"lr": 0.001, "data_time": 0.0009465694427490234, "grad_norm": 0.10927009116858244, "loss": 0.6847162783145905, "time": 0.5044364929199219, "epoch": 1, "memory": 9474, "step": 7900} +{"lr": 0.001, "data_time": 0.0009895086288452149, "grad_norm": 0.10070986961945891, "loss": 0.6876133859157563, "time": 0.4144383192062378, "epoch": 1, "memory": 9474, "step": 8000} +{"lr": 0.001, "data_time": 0.0009637355804443359, "grad_norm": 0.1331441055983305, "loss": 0.6843811333179474, "time": 0.392572808265686, "epoch": 1, "memory": 9474, "step": 8100} +{"lr": 0.001, "data_time": 0.0010759592056274413, "grad_norm": 0.13488234179094433, "loss": 0.6888718068599701, "time": 0.40422210693359373, "epoch": 1, "memory": 9474, "step": 8200} +{"lr": 0.001, "data_time": 0.001038360595703125, "grad_norm": 0.11280787996947765, "loss": 0.6876307129859924, "time": 0.39751129150390624, "epoch": 1, "memory": 9474, "step": 8300} +{"lr": 0.001, "data_time": 0.0009946107864379882, "grad_norm": 0.09878557398915291, "loss": 0.6860785484313965, "time": 0.4117953062057495, "epoch": 1, "memory": 9474, "step": 8400} +{"lr": 0.001, "data_time": 0.0009869575500488282, "grad_norm": 0.1379073467105627, "loss": 0.6873394250869751, "time": 0.43392760753631593, "epoch": 1, "memory": 9474, "step": 8500} +{"lr": 0.001, "data_time": 0.0008696079254150391, "grad_norm": 0.10519644804298878, "loss": 0.6859173238277435, "time": 0.4291120767593384, "epoch": 1, "memory": 9474, "step": 8600} +{"lr": 0.001, "data_time": 0.0010144472122192382, "grad_norm": 0.10368053894490004, "loss": 0.68602956533432, "time": 0.45687916278839114, "epoch": 1, "memory": 9474, "step": 8700} +{"lr": 0.001, "data_time": 0.0014067888259887695, "grad_norm": 0.08638579975813628, "loss": 0.6815131843090058, "time": 0.3762549877166748, "epoch": 1, "memory": 9474, "step": 8800} +{"lr": 0.001, "data_time": 0.0011425256729125977, "grad_norm": 0.08293162267655134, "loss": 0.6876282215118408, "time": 0.42606263160705565, "epoch": 1, "memory": 9474, "step": 8900} +{"lr": 0.001, "data_time": 0.001256704330444336, "grad_norm": 0.11535952351987362, "loss": 0.6890344023704529, "time": 0.4015597581863403, "epoch": 1, "memory": 9474, "step": 9000} +{"lr": 0.001, "data_time": 0.0011426925659179688, "grad_norm": 0.09358172118663788, "loss": 0.681056809425354, "time": 0.4113957643508911, "epoch": 1, "memory": 9474, "step": 9100} +{"lr": 0.001, "data_time": 0.0009740829467773437, "grad_norm": 0.08762898705899716, "loss": 0.6841725528240203, "time": 0.3920086145401001, "epoch": 1, "memory": 9474, "step": 9200} +{"lr": 0.001, "data_time": 0.001075434684753418, "grad_norm": 0.07830161657184362, "loss": 0.6841848134994507, "time": 0.43527817726135254, "epoch": 1, "memory": 9474, "step": 9300} +{"lr": 0.001, "data_time": 0.0010875463485717773, "grad_norm": 0.074044300429523, "loss": 0.6865354955196381, "time": 0.40725576877593994, "epoch": 1, "memory": 9474, "step": 9400} +{"lr": 0.001, "data_time": 0.0010560989379882813, "grad_norm": 0.08644570319447667, "loss": 0.6852303683757782, "time": 0.4142030715942383, "epoch": 1, "memory": 9474, "step": 9500} +{"lr": 0.001, "data_time": 0.0010829687118530274, "grad_norm": 0.08891921145841479, "loss": 0.6819977521896362, "time": 0.37514657974243165, "epoch": 1, "memory": 9474, "step": 9600} +{"lr": 0.001, "data_time": 0.001034379005432129, "grad_norm": 0.07139426893554628, "loss": 0.6931855022907257, "time": 0.409479546546936, "epoch": 1, "memory": 9474, "step": 9700} +{"lr": 0.001, "data_time": 0.0010714054107666016, "grad_norm": 0.09761939700692893, "loss": 0.6861530482769013, "time": 0.4011998176574707, "epoch": 1, "memory": 9474, "step": 9800} +{"lr": 0.001, "data_time": 0.0011259078979492187, "grad_norm": 0.10074464427307248, "loss": 0.6831479549407959, "time": 0.41072404384613037, "epoch": 1, "memory": 9474, "step": 9900} +{"lr": 0.001, "data_time": 0.0010134220123291016, "grad_norm": 0.06637590676546097, "loss": 0.6832959115505218, "time": 0.46042556762695314, "epoch": 1, "memory": 9474, "step": 10000} +{"lr": 0.001, "data_time": 0.0009308338165283203, "grad_norm": 0.1275673158466816, "loss": 0.6818589746952057, "time": 0.38405086994171145, "epoch": 1, "memory": 9474, "step": 10100} +{"lr": 0.001, "data_time": 0.0011119365692138672, "grad_norm": 0.06425417019054294, "loss": 0.6876189649105072, "time": 0.40998854637146, "epoch": 1, "memory": 9474, "step": 10200} +{"lr": 0.001, "data_time": 0.0010421037673950194, "grad_norm": 0.08715715687721967, "loss": 0.6832851946353913, "time": 0.3770972967147827, "epoch": 1, "memory": 9474, "step": 10300} +{"lr": 0.001, "data_time": 0.000909876823425293, "grad_norm": 0.11128245387226343, "loss": 0.6887672781944275, "time": 0.3994786262512207, "epoch": 1, "memory": 9474, "step": 10400} +{"lr": 0.001, "data_time": 0.000986933708190918, "grad_norm": 0.08399312794208527, "loss": 0.6866066873073577, "time": 0.39032630920410155, "epoch": 1, "memory": 9474, "step": 10500} +{"lr": 0.001, "data_time": 0.0012809038162231445, "grad_norm": 0.06851767767220736, "loss": 0.6826740264892578, "time": 0.40223581790924073, "epoch": 1, "memory": 9474, "step": 10600} +{"lr": 0.001, "data_time": 0.0009658575057983398, "grad_norm": 0.05832801419310272, "loss": 0.6870149672031403, "time": 0.39209136962890623, "epoch": 1, "memory": 9474, "step": 10700} +{"lr": 0.001, "data_time": 0.0009809494018554687, "grad_norm": 0.07795922216027976, "loss": 0.6847951591014863, "time": 0.4209475040435791, "epoch": 1, "memory": 9474, "step": 10800} +{"lr": 0.001, "data_time": 0.001141667366027832, "grad_norm": 0.06304550403729081, "loss": 0.685225784778595, "time": 0.3920806646347046, "epoch": 1, "memory": 9474, "step": 10900} +{"lr": 0.001, "data_time": 0.0009903907775878906, "grad_norm": 0.07760525858029724, "loss": 0.6900427520275116, "time": 0.4039088487625122, "epoch": 1, "memory": 9474, "step": 11000} +{"lr": 0.001, "data_time": 0.0010686635971069336, "grad_norm": 0.07758190110325813, "loss": 0.685785299539566, "time": 0.3837052583694458, "epoch": 1, "memory": 9474, "step": 11100} +{"lr": 0.001, "data_time": 0.001104593276977539, "grad_norm": 0.0721831951290369, "loss": 0.6898037552833557, "time": 0.39397780895233153, "epoch": 1, "memory": 9474, "step": 11200} +{"lr": 0.001, "data_time": 0.0010874032974243163, "grad_norm": 0.06284296195954084, "loss": 0.6920159876346588, "time": 0.42384178638458253, "epoch": 1, "memory": 9474, "step": 11300} +{"lr": 0.001, "data_time": 0.001458287239074707, "grad_norm": 0.08862904305569827, "loss": 0.6865630865097045, "time": 0.37363195419311523, "epoch": 1, "memory": 9474, "step": 11400} +{"lr": 0.001, "data_time": 0.008296585083007813, "grad_norm": 0.0653378666844219, "loss": 0.6869338035583497, "time": 0.4357378244400024, "epoch": 1, "memory": 9474, "step": 11500} +{"accuracy/top1": 100.0, "data_time": 0.309885635972023, "time": 0.6909352540969849, "step": 1} +{"lr": 0.0009939057285945933, "data_time": 0.07083356380462646, "grad_norm": 0.1099855212494731, "loss": 0.6879038155078888, "time": 0.46412322521209715, "epoch": 2, "memory": 9474, "step": 11616} +{"lr": 0.0009939057285945933, "data_time": 0.001030731201171875, "grad_norm": 0.06427400205284357, "loss": 0.6862953543663025, "time": 0.4322756290435791, "epoch": 2, "memory": 9466, "step": 11716} +{"lr": 0.0009939057285945933, "data_time": 0.0011673927307128905, "grad_norm": 0.0751484197564423, "loss": 0.6848814904689788, "time": 0.48997957706451417, "epoch": 2, "memory": 9466, "step": 11816} +{"lr": 0.0009939057285945933, "data_time": 0.0010225772857666016, "grad_norm": 0.06925361389294267, "loss": 0.6876432597637177, "time": 0.4027824878692627, "epoch": 2, "memory": 9466, "step": 11916} +{"lr": 0.0009939057285945933, "data_time": 0.0010874032974243163, "grad_norm": 0.06935103619471192, "loss": 0.684624993801117, "time": 0.40043728351593016, "epoch": 2, "memory": 9466, "step": 12016} +{"lr": 0.0009939057285945933, "data_time": 0.0010295867919921874, "grad_norm": 0.06250485554337501, "loss": 0.6886254191398621, "time": 0.43543667793273927, "epoch": 2, "memory": 9466, "step": 12116} +{"lr": 0.0009939057285945933, "data_time": 0.0012871026992797852, "grad_norm": 0.058108051493763926, "loss": 0.6871399939060211, "time": 0.504337215423584, "epoch": 2, "memory": 9466, "step": 12216} +{"lr": 0.0009939057285945933, "data_time": 0.0012480258941650391, "grad_norm": 0.06246477738022804, "loss": 0.689530348777771, "time": 0.39492509365081785, "epoch": 2, "memory": 9466, "step": 12316} +{"lr": 0.0009939057285945933, "data_time": 0.0010716438293457032, "grad_norm": 0.07895727716386318, "loss": 0.6859519243240356, "time": 0.46003384590148927, "epoch": 2, "memory": 9466, "step": 12416} +{"lr": 0.0009939057285945933, "data_time": 0.001078963279724121, "grad_norm": 0.06565370513126254, "loss": 0.6878838241100311, "time": 0.46222763061523436, "epoch": 2, "memory": 9466, "step": 12516} +{"lr": 0.0009939057285945933, "data_time": 0.0011182785034179687, "grad_norm": 0.06038331417366862, "loss": 0.6912600874900818, "time": 0.473629093170166, "epoch": 2, "memory": 9466, "step": 12616} +{"lr": 0.0009939057285945933, "data_time": 0.0009978771209716796, "grad_norm": 0.05918410774320364, "loss": 0.690394276380539, "time": 0.47400593757629395, "epoch": 2, "memory": 9466, "step": 12716} +{"lr": 0.0009939057285945933, "data_time": 0.0011910438537597657, "grad_norm": 0.05964112002402544, "loss": 0.6870898425579071, "time": 0.4028794765472412, "epoch": 2, "memory": 9466, "step": 12816} +{"lr": 0.0009939057285945933, "data_time": 0.001068282127380371, "grad_norm": 0.0650144257582724, "loss": 0.6826631903648377, "time": 0.46787664890289304, "epoch": 2, "memory": 9466, "step": 12916} +{"lr": 0.0009939057285945933, "data_time": 0.0010412216186523437, "grad_norm": 0.043797826487571, "loss": 0.688552850484848, "time": 0.3952311992645264, "epoch": 2, "memory": 9466, "step": 13016} +{"lr": 0.0009939057285945933, "data_time": 0.001025557518005371, "grad_norm": 0.03829173250123859, "loss": 0.6879911184310913, "time": 0.5633004903793335, "epoch": 2, "memory": 9466, "step": 13116} +{"lr": 0.0009939057285945933, "data_time": 0.0010823249816894532, "grad_norm": 0.06581303123384714, "loss": 0.6825634062290191, "time": 0.45997314453125, "epoch": 2, "memory": 9466, "step": 13216} +{"lr": 0.0009939057285945933, "data_time": 0.0011197805404663086, "grad_norm": 0.08384905718266963, "loss": 0.6832798004150391, "time": 0.4203648090362549, "epoch": 2, "memory": 9466, "step": 13316} +{"lr": 0.0009939057285945933, "data_time": 0.001103353500366211, "grad_norm": 0.04816665789112449, "loss": 0.6857930243015289, "time": 0.43519017696380613, "epoch": 2, "memory": 9466, "step": 13416} +{"lr": 0.0009939057285945933, "data_time": 0.001039886474609375, "grad_norm": 0.05029902392998338, "loss": 0.6834621250629425, "time": 0.42972192764282224, "epoch": 2, "memory": 9466, "step": 13516} +{"lr": 0.0009939057285945933, "data_time": 0.0011239051818847656, "grad_norm": 0.04806776299374178, "loss": 0.6888626396656037, "time": 0.39881505966186526, "epoch": 2, "memory": 9466, "step": 13616} +{"lr": 0.0009939057285945933, "data_time": 0.001100301742553711, "grad_norm": 0.047950968914665285, "loss": 0.6914717495441437, "time": 0.4222588539123535, "epoch": 2, "memory": 9466, "step": 13716} +{"lr": 0.0009939057285945933, "data_time": 0.0010953187942504884, "grad_norm": 0.03864505849778652, "loss": 0.687501072883606, "time": 0.40851192474365233, "epoch": 2, "memory": 9466, "step": 13816} +{"lr": 0.0009939057285945933, "data_time": 0.0012400388717651368, "grad_norm": 0.03776288768276572, "loss": 0.6899020135402679, "time": 0.48978049755096437, "epoch": 2, "memory": 9466, "step": 13916} +{"lr": 0.0009939057285945933, "data_time": 0.0010712862014770508, "grad_norm": 0.06297209057956935, "loss": 0.6826109707355499, "time": 0.42327046394348145, "epoch": 2, "memory": 9466, "step": 14016} +{"lr": 0.0009939057285945933, "data_time": 0.0010754823684692382, "grad_norm": 0.0396063226275146, "loss": 0.6863962829113006, "time": 0.4419059753417969, "epoch": 2, "memory": 9466, "step": 14116} +{"lr": 0.0009939057285945933, "data_time": 0.0011355876922607422, "grad_norm": 0.05816199849359691, "loss": 0.686329436302185, "time": 0.43482882976531984, "epoch": 2, "memory": 9466, "step": 14216} +{"lr": 0.0009939057285945933, "data_time": 0.0011022090911865234, "grad_norm": 0.03832084927707911, "loss": 0.690645182132721, "time": 0.47258481979370115, "epoch": 2, "memory": 9466, "step": 14316} +{"lr": 0.0009939057285945933, "data_time": 0.0010602235794067382, "grad_norm": 0.052494379319250585, "loss": 0.6814186632633209, "time": 0.45048458576202394, "epoch": 2, "memory": 9466, "step": 14416} +{"lr": 0.0009939057285945933, "data_time": 0.0009874343872070313, "grad_norm": 0.0462041768245399, "loss": 0.6880272209644318, "time": 0.4128320932388306, "epoch": 2, "memory": 9466, "step": 14516} +{"lr": 0.0009939057285945933, "data_time": 0.001046442985534668, "grad_norm": 0.0523037008009851, "loss": 0.685356217622757, "time": 0.4169517993927002, "epoch": 2, "memory": 9466, "step": 14616} +{"lr": 0.0009939057285945933, "data_time": 0.0010082006454467773, "grad_norm": 0.05938305910676718, "loss": 0.6881160974502564, "time": 0.409075403213501, "epoch": 2, "memory": 9466, "step": 14716} +{"lr": 0.0009939057285945933, "data_time": 0.0010674715042114258, "grad_norm": 0.03911679950542748, "loss": 0.6858436703681946, "time": 0.549166464805603, "epoch": 2, "memory": 9466, "step": 14816} +{"lr": 0.0009939057285945933, "data_time": 0.001062321662902832, "grad_norm": 0.04382241545245051, "loss": 0.6866205334663391, "time": 0.41299474239349365, "epoch": 2, "memory": 9466, "step": 14916} +{"lr": 0.0009939057285945933, "data_time": 0.0011612653732299804, "grad_norm": 0.05139650125056505, "loss": 0.6853583335876465, "time": 0.45559077262878417, "epoch": 2, "memory": 9466, "step": 15016} +{"lr": 0.0009939057285945933, "data_time": 0.0011552810668945313, "grad_norm": 0.026551441475749015, "loss": 0.6852880954742432, "time": 0.4236022710800171, "epoch": 2, "memory": 9466, "step": 15116} +{"lr": 0.0009939057285945933, "data_time": 0.0011253595352172852, "grad_norm": 0.048103848658502105, "loss": 0.6889339566230774, "time": 0.4459666728973389, "epoch": 2, "memory": 9466, "step": 15216} +{"lr": 0.0009939057285945933, "data_time": 0.001021409034729004, "grad_norm": 0.06299279164522886, "loss": 0.6900611698627472, "time": 0.42662069797515867, "epoch": 2, "memory": 9466, "step": 15316} +{"lr": 0.0009939057285945933, "data_time": 0.0010375022888183595, "grad_norm": 0.03240831485018134, "loss": 0.6861496269702911, "time": 0.4836645364761353, "epoch": 2, "memory": 9466, "step": 15416} +{"lr": 0.0009939057285945933, "data_time": 0.0011087179183959961, "grad_norm": 0.0530106445774436, "loss": 0.6845498561859131, "time": 0.4130805492401123, "epoch": 2, "memory": 9466, "step": 15516} +{"lr": 0.0009939057285945933, "data_time": 0.0009972810745239257, "grad_norm": 0.047135541401803496, "loss": 0.6832665026187896, "time": 0.4348788022994995, "epoch": 2, "memory": 9466, "step": 15616} +{"lr": 0.0009939057285945933, "data_time": 0.0010476112365722656, "grad_norm": 0.032505690841935574, "loss": 0.6895552337169647, "time": 0.45715720653533937, "epoch": 2, "memory": 9466, "step": 15716} +{"lr": 0.0009939057285945933, "data_time": 0.0010787248611450195, "grad_norm": 0.05903021469712257, "loss": 0.6903643667697906, "time": 0.4205663204193115, "epoch": 2, "memory": 9466, "step": 15816} +{"lr": 0.0009939057285945933, "data_time": 0.0010478973388671875, "grad_norm": 0.04131353124976158, "loss": 0.691485857963562, "time": 0.5404685735702515, "epoch": 2, "memory": 9466, "step": 15916} +{"lr": 0.0009939057285945933, "data_time": 0.0010309457778930665, "grad_norm": 0.040479103871621194, "loss": 0.6893355786800385, "time": 0.5037957906723023, "epoch": 2, "memory": 9466, "step": 16016} +{"lr": 0.0009939057285945933, "data_time": 0.001025390625, "grad_norm": 0.03441180996596813, "loss": 0.6909900546073914, "time": 0.4745124101638794, "epoch": 2, "memory": 9466, "step": 16116} +{"lr": 0.0009939057285945933, "data_time": 0.0010228157043457031, "grad_norm": 0.046719822310842575, "loss": 0.6836608350276947, "time": 0.4528194904327393, "epoch": 2, "memory": 9466, "step": 16216} +{"lr": 0.0009939057285945933, "data_time": 0.001015782356262207, "grad_norm": 0.05144949820823967, "loss": 0.6850019097328186, "time": 0.415437126159668, "epoch": 2, "memory": 9466, "step": 16316} +{"lr": 0.0009939057285945933, "data_time": 0.0010032415390014648, "grad_norm": 0.05196168276015669, "loss": 0.6894936501979828, "time": 0.4345522403717041, "epoch": 2, "memory": 9466, "step": 16416} +{"lr": 0.0009939057285945933, "data_time": 0.0009929418563842773, "grad_norm": 0.034834797796793285, "loss": 0.6846549689769745, "time": 0.4143122673034668, "epoch": 2, "memory": 9466, "step": 16516} +{"lr": 0.0009939057285945933, "data_time": 0.0011261224746704102, "grad_norm": 0.03526101897004992, "loss": 0.6842919528484345, "time": 0.41216745376586916, "epoch": 2, "memory": 9466, "step": 16616} +{"lr": 0.0009939057285945933, "data_time": 0.001013970375061035, "grad_norm": 0.03366449736058712, "loss": 0.6872649133205414, "time": 0.4130429983139038, "epoch": 2, "memory": 9466, "step": 16716} +{"lr": 0.0009939057285945933, "data_time": 0.001103043556213379, "grad_norm": 0.05951396320015192, "loss": 0.690333080291748, "time": 0.456061863899231, "epoch": 2, "memory": 9466, "step": 16816} +{"lr": 0.0009939057285945933, "data_time": 0.0010926961898803712, "grad_norm": 0.032994561549276116, "loss": 0.6895784199237823, "time": 0.4121453285217285, "epoch": 2, "memory": 9466, "step": 16916} +{"lr": 0.0009939057285945933, "data_time": 0.0010883808135986328, "grad_norm": 0.04752643899992108, "loss": 0.6828528463840484, "time": 0.4512415647506714, "epoch": 2, "memory": 9466, "step": 17016} +{"lr": 0.0009939057285945933, "data_time": 0.001089167594909668, "grad_norm": 0.0521976382471621, "loss": 0.6842605650424958, "time": 0.42761313915252686, "epoch": 2, "memory": 9466, "step": 17116} +{"lr": 0.0009939057285945933, "data_time": 0.001126265525817871, "grad_norm": 0.06867835689336062, "loss": 0.6868312180042266, "time": 0.4242203712463379, "epoch": 2, "memory": 9466, "step": 17216} +{"lr": 0.0009939057285945933, "data_time": 0.0011965513229370117, "grad_norm": 0.022139145992696285, "loss": 0.6896021842956543, "time": 0.4299457550048828, "epoch": 2, "memory": 9466, "step": 17316} +{"lr": 0.0009939057285945933, "data_time": 0.001110672950744629, "grad_norm": 0.03525620233267546, "loss": 0.6865286350250244, "time": 0.41831867694854735, "epoch": 2, "memory": 9466, "step": 17416} +{"lr": 0.0009939057285945933, "data_time": 0.001197481155395508, "grad_norm": 0.03726433843839914, "loss": 0.6877101123332977, "time": 0.4258462905883789, "epoch": 2, "memory": 9466, "step": 17516} +{"lr": 0.0009939057285945933, "data_time": 0.0010362386703491211, "grad_norm": 0.04673717836849391, "loss": 0.6802491545677185, "time": 0.47735061645507815, "epoch": 2, "memory": 9466, "step": 17616} +{"lr": 0.0009939057285945933, "data_time": 0.0010113239288330079, "grad_norm": 0.03628114508464932, "loss": 0.6817511022090912, "time": 0.4868102312088013, "epoch": 2, "memory": 9466, "step": 17716} +{"lr": 0.0009939057285945933, "data_time": 0.001108860969543457, "grad_norm": 0.03147273710928857, "loss": 0.6830770611763001, "time": 0.44975972175598145, "epoch": 2, "memory": 9466, "step": 17816} +{"lr": 0.0009939057285945933, "data_time": 0.001001596450805664, "grad_norm": 0.023424675222486258, "loss": 0.6894601941108703, "time": 0.4413499116897583, "epoch": 2, "memory": 9466, "step": 17916} +{"lr": 0.0009939057285945933, "data_time": 0.0010518074035644532, "grad_norm": 0.05152399064972997, "loss": 0.6875954151153565, "time": 0.4273714780807495, "epoch": 2, "memory": 9466, "step": 18016} +{"lr": 0.0009939057285945933, "data_time": 0.0010159015655517578, "grad_norm": 0.030310118291527032, "loss": 0.6842800438404083, "time": 0.4172009229660034, "epoch": 2, "memory": 9466, "step": 18116} +{"lr": 0.0009939057285945933, "data_time": 0.0010413169860839845, "grad_norm": 0.024765848577953876, "loss": 0.6799925565719604, "time": 0.40242578983306887, "epoch": 2, "memory": 9466, "step": 18216} +{"lr": 0.0009939057285945933, "data_time": 0.0009528636932373047, "grad_norm": 0.03369774019811302, "loss": 0.6903659999370575, "time": 0.48039536476135253, "epoch": 2, "memory": 9466, "step": 18316} +{"lr": 0.0009939057285945933, "data_time": 0.001019287109375, "grad_norm": 0.035824154131114484, "loss": 0.6824732899665833, "time": 0.4307317972183228, "epoch": 2, "memory": 9466, "step": 18416} +{"lr": 0.0009939057285945933, "data_time": 0.0014621734619140625, "grad_norm": 0.028447221009992064, "loss": 0.689435213804245, "time": 0.4386845350265503, "epoch": 2, "memory": 9466, "step": 18516} +{"lr": 0.0009939057285945933, "data_time": 0.0011492252349853515, "grad_norm": 0.039267354225739834, "loss": 0.6890923380851746, "time": 0.4825651407241821, "epoch": 2, "memory": 9466, "step": 18616} +{"lr": 0.0009939057285945933, "data_time": 0.001076340675354004, "grad_norm": 0.0264707637950778, "loss": 0.6854984283447265, "time": 0.3985506772994995, "epoch": 2, "memory": 9466, "step": 18716} +{"lr": 0.0009939057285945933, "data_time": 0.0011612653732299804, "grad_norm": 0.04025698853656649, "loss": 0.6885736048221588, "time": 0.41092500686645506, "epoch": 2, "memory": 9466, "step": 18816} +{"lr": 0.0009939057285945933, "data_time": 0.0010822057723999024, "grad_norm": 0.03930663289502263, "loss": 0.6831455588340759, "time": 0.46106586456298826, "epoch": 2, "memory": 9466, "step": 18916} +{"lr": 0.0009939057285945933, "data_time": 0.0010393381118774414, "grad_norm": 0.021630261489190162, "loss": 0.6858853161334991, "time": 0.45829143524169924, "epoch": 2, "memory": 9466, "step": 19016} +{"lr": 0.0009939057285945933, "data_time": 0.0011249542236328124, "grad_norm": 0.025389747810550034, "loss": 0.6922311604022979, "time": 0.4417997360229492, "epoch": 2, "memory": 9466, "step": 19116} +{"lr": 0.0009939057285945933, "data_time": 0.0011394739151000977, "grad_norm": 0.027805525809526443, "loss": 0.6826779186725617, "time": 0.44397075176239015, "epoch": 2, "memory": 9466, "step": 19216} +{"lr": 0.0009939057285945933, "data_time": 0.001194906234741211, "grad_norm": 0.021581029053777456, "loss": 0.6884427964687347, "time": 0.45931086540222166, "epoch": 2, "memory": 9466, "step": 19316} +{"lr": 0.0009939057285945933, "data_time": 0.0011829614639282226, "grad_norm": 0.029122319514863194, "loss": 0.6823606014251709, "time": 0.42016050815582273, "epoch": 2, "memory": 9466, "step": 19416} +{"lr": 0.0009939057285945933, "data_time": 0.0010602235794067382, "grad_norm": 0.032002720597665756, "loss": 0.6855169177055359, "time": 0.450806999206543, "epoch": 2, "memory": 9466, "step": 19516} +{"lr": 0.0009939057285945933, "data_time": 0.0010131120681762696, "grad_norm": 0.03347331110853702, "loss": 0.6862307786941528, "time": 0.40237624645233155, "epoch": 2, "memory": 9466, "step": 19616} +{"lr": 0.0009939057285945933, "data_time": 0.0010608434677124023, "grad_norm": 0.025364167243242263, "loss": 0.682321161031723, "time": 0.43271775245666505, "epoch": 2, "memory": 9466, "step": 19716} +{"lr": 0.0009939057285945933, "data_time": 0.0012502670288085938, "grad_norm": 0.03594321897253394, "loss": 0.6886745691299438, "time": 0.3986459493637085, "epoch": 2, "memory": 9466, "step": 19816} +{"lr": 0.0009939057285945933, "data_time": 0.0012074947357177735, "grad_norm": 0.025908530689775944, "loss": 0.6872937381267548, "time": 0.4347603559494019, "epoch": 2, "memory": 9466, "step": 19916} +{"lr": 0.0009939057285945933, "data_time": 0.001089000701904297, "grad_norm": 0.026527242409065364, "loss": 0.6900459051132202, "time": 0.4368203401565552, "epoch": 2, "memory": 9466, "step": 20016} +{"lr": 0.0009939057285945933, "data_time": 0.001088404655456543, "grad_norm": 0.01877260854234919, "loss": 0.6839976489543915, "time": 0.41747846603393557, "epoch": 2, "memory": 9466, "step": 20116} +{"lr": 0.0009939057285945933, "data_time": 0.0010925054550170899, "grad_norm": 0.030362183065153658, "loss": 0.683194762468338, "time": 0.41016592979431155, "epoch": 2, "memory": 9466, "step": 20216} +{"lr": 0.0009939057285945933, "data_time": 0.0010634660720825195, "grad_norm": 0.03470392026938498, "loss": 0.6819584190845489, "time": 0.39416506290435793, "epoch": 2, "memory": 9466, "step": 20316} +{"lr": 0.0009939057285945933, "data_time": 0.0013753652572631836, "grad_norm": 0.03229015273973346, "loss": 0.6807324111461639, "time": 0.40594961643218996, "epoch": 2, "memory": 9466, "step": 20416} +{"lr": 0.0009939057285945933, "data_time": 0.0011197328567504883, "grad_norm": 0.024766219686716794, "loss": 0.6878861069679261, "time": 0.41277735233306884, "epoch": 2, "memory": 9466, "step": 20516} +{"lr": 0.0009939057285945933, "data_time": 0.0011850357055664062, "grad_norm": 0.023120308574289084, "loss": 0.6896874368190765, "time": 0.40929038524627687, "epoch": 2, "memory": 9466, "step": 20616} +{"lr": 0.0009939057285945933, "data_time": 0.0011122941970825196, "grad_norm": 0.028626284468919037, "loss": 0.6880630731582642, "time": 0.46014392375946045, "epoch": 2, "memory": 9466, "step": 20716} +{"lr": 0.0009939057285945933, "data_time": 0.0012867212295532226, "grad_norm": 0.03971298469696194, "loss": 0.6875614821910858, "time": 0.44701123237609863, "epoch": 2, "memory": 9466, "step": 20816} +{"lr": 0.0009939057285945933, "data_time": 0.0011586904525756835, "grad_norm": 0.0199462678283453, "loss": 0.6845956027507782, "time": 0.4387508869171143, "epoch": 2, "memory": 9466, "step": 20916} +{"lr": 0.0009939057285945933, "data_time": 0.0011193275451660157, "grad_norm": 0.02812248272821307, "loss": 0.6877283215522766, "time": 0.44707579612731935, "epoch": 2, "memory": 9466, "step": 21016} +{"lr": 0.0009939057285945933, "data_time": 0.0012781143188476563, "grad_norm": 0.028240500867832453, "loss": 0.690543282032013, "time": 0.4225125551223755, "epoch": 2, "memory": 9466, "step": 21116} +{"lr": 0.0009939057285945933, "data_time": 0.0011524438858032226, "grad_norm": 0.03582395128905773, "loss": 0.6840147912502289, "time": 0.44947845935821534, "epoch": 2, "memory": 9466, "step": 21216} +{"lr": 0.0009939057285945933, "data_time": 0.0010776281356811523, "grad_norm": 0.021379622141830625, "loss": 0.687153959274292, "time": 0.4285252094268799, "epoch": 2, "memory": 9466, "step": 21316} +{"lr": 0.0009939057285945933, "data_time": 0.001127934455871582, "grad_norm": 0.025071394885890184, "loss": 0.6861791789531708, "time": 0.4396047592163086, "epoch": 2, "memory": 9466, "step": 21416} +{"lr": 0.0009939057285945933, "data_time": 0.001260828971862793, "grad_norm": 0.030992146208882333, "loss": 0.684092378616333, "time": 0.4403985500335693, "epoch": 2, "memory": 9466, "step": 21516} +{"lr": 0.0009939057285945933, "data_time": 0.0010259628295898437, "grad_norm": 0.023117261379957198, "loss": 0.6890429496765137, "time": 0.4923547744750977, "epoch": 2, "memory": 9466, "step": 21616} +{"lr": 0.0009939057285945933, "data_time": 0.0011034965515136718, "grad_norm": 0.017198899178765714, "loss": 0.6831491529941559, "time": 0.44282629489898684, "epoch": 2, "memory": 9466, "step": 21716} +{"lr": 0.0009939057285945933, "data_time": 0.0011950492858886718, "grad_norm": 0.03520979601889849, "loss": 0.6863300621509552, "time": 0.4358607053756714, "epoch": 2, "memory": 9466, "step": 21816} +{"lr": 0.0009939057285945933, "data_time": 0.0013605833053588867, "grad_norm": 0.026939908997155725, "loss": 0.6850939512252807, "time": 0.4162754535675049, "epoch": 2, "memory": 9466, "step": 21916} +{"lr": 0.0009939057285945933, "data_time": 0.0010367631912231445, "grad_norm": 0.025196888181380928, "loss": 0.6890845596790314, "time": 0.4047020435333252, "epoch": 2, "memory": 9466, "step": 22016} +{"lr": 0.0009939057285945933, "data_time": 0.001160883903503418, "grad_norm": 0.015232666418887675, "loss": 0.6888692080974579, "time": 0.42114341259002686, "epoch": 2, "memory": 9466, "step": 22116} +{"lr": 0.0009939057285945933, "data_time": 0.0012646913528442383, "grad_norm": 0.012506846012547613, "loss": 0.6819145679473877, "time": 0.39800872802734377, "epoch": 2, "memory": 9466, "step": 22216} +{"lr": 0.0009939057285945933, "data_time": 0.0011089801788330077, "grad_norm": 0.024451967608183623, "loss": 0.6845054745674133, "time": 0.46559360027313235, "epoch": 2, "memory": 9466, "step": 22316} +{"lr": 0.0009939057285945933, "data_time": 0.0011581897735595704, "grad_norm": 0.024922854756005107, "loss": 0.6825241982936859, "time": 0.40462474822998046, "epoch": 2, "memory": 9466, "step": 22416} +{"lr": 0.0009939057285945933, "data_time": 0.001146554946899414, "grad_norm": 0.018348787701688705, "loss": 0.6876084089279175, "time": 0.37368662357330323, "epoch": 2, "memory": 9466, "step": 22516} +{"lr": 0.0009939057285945933, "data_time": 0.0010806083679199218, "grad_norm": 0.022323320084251464, "loss": 0.6853807032108307, "time": 0.46470203399658205, "epoch": 2, "memory": 9466, "step": 22616} +{"lr": 0.0009939057285945933, "data_time": 0.0011414766311645507, "grad_norm": 0.025385763216763734, "loss": 0.6867986142635345, "time": 0.4136732339859009, "epoch": 2, "memory": 9466, "step": 22716} +{"lr": 0.0009939057285945933, "data_time": 0.0012858390808105468, "grad_norm": 0.024933330668136477, "loss": 0.6839031398296356, "time": 0.421768856048584, "epoch": 2, "memory": 9466, "step": 22816} +{"lr": 0.0009939057285945933, "data_time": 0.0013912200927734375, "grad_norm": 0.031420634919777514, "loss": 0.6872365832328796, "time": 0.3978781223297119, "epoch": 2, "memory": 9466, "step": 22916} +{"lr": 0.0009939057285945933, "data_time": 0.004609894752502441, "grad_norm": 0.02087268108734861, "loss": 0.6900867044925689, "time": 0.4825006008148193, "epoch": 2, "memory": 9466, "step": 23016} +{"accuracy/top1": 100.0, "data_time": 0.30627135669483857, "time": 0.5601891910328585, "step": 2} +{"lr": 0.0009757729755661012, "data_time": 0.000995039939880371, "grad_norm": 0.015585147636011243, "loss": 0.6836028575897217, "time": 0.44008607864379884, "epoch": 3, "memory": 9466, "step": 23132} +{"lr": 0.0009757729755661012, "data_time": 0.001377701759338379, "grad_norm": 0.023318680422380566, "loss": 0.6925673604011535, "time": 0.4418300151824951, "epoch": 3, "memory": 9465, "step": 23232} +{"lr": 0.0009757729755661012, "data_time": 0.0827981948852539, "grad_norm": 0.019727103505283596, "loss": 0.6877822697162628, "time": 0.4060245990753174, "epoch": 3, "memory": 9465, "step": 23332} +{"lr": 0.0009757729755661012, "data_time": 0.0009917736053466797, "grad_norm": 0.018417895305901766, "loss": 0.6871657907962799, "time": 0.4709378719329834, "epoch": 3, "memory": 9465, "step": 23432} +{"lr": 0.0009757729755661012, "data_time": 0.0017249345779418944, "grad_norm": 0.02222048370167613, "loss": 0.6895881295204163, "time": 0.431545090675354, "epoch": 3, "memory": 9465, "step": 23532} +{"lr": 0.0009757729755661012, "data_time": 0.0017734527587890624, "grad_norm": 0.019312548195011915, "loss": 0.6906211972236633, "time": 0.4764890432357788, "epoch": 3, "memory": 9465, "step": 23632} +{"lr": 0.0009757729755661012, "data_time": 0.0014914512634277345, "grad_norm": 0.014033994544297457, "loss": 0.6879104256629944, "time": 0.48021764755249025, "epoch": 3, "memory": 9465, "step": 23732} +{"lr": 0.0009757729755661012, "data_time": 0.0014476299285888672, "grad_norm": 0.02209121969062835, "loss": 0.6855984449386596, "time": 0.40016770362854004, "epoch": 3, "memory": 9465, "step": 23832} +{"lr": 0.0009757729755661012, "data_time": 0.0012039899826049804, "grad_norm": 0.013217152189463377, "loss": 0.6847003161907196, "time": 0.45034902095794677, "epoch": 3, "memory": 9465, "step": 23932} +{"lr": 0.0009757729755661012, "data_time": 0.0011378049850463868, "grad_norm": 0.026478892378509044, "loss": 0.6886955976486206, "time": 0.47276129722595217, "epoch": 3, "memory": 9465, "step": 24032} +{"lr": 0.0009757729755661012, "data_time": 0.0010313987731933594, "grad_norm": 0.024408833286724985, "loss": 0.6888004958629608, "time": 0.43849332332611085, "epoch": 3, "memory": 9465, "step": 24132} +{"lr": 0.0009757729755661012, "data_time": 0.0020800113677978517, "grad_norm": 0.020728322584182023, "loss": 0.6848903357982635, "time": 0.39533162117004395, "epoch": 3, "memory": 9465, "step": 24232} +{"lr": 0.0009757729755661012, "data_time": 0.0012228727340698243, "grad_norm": 0.02266580640571192, "loss": 0.687611174583435, "time": 0.4516587495803833, "epoch": 3, "memory": 9465, "step": 24332} +{"lr": 0.0009757729755661012, "data_time": 0.0020180463790893553, "grad_norm": 0.02308184818830341, "loss": 0.685395109653473, "time": 0.4768563508987427, "epoch": 3, "memory": 9465, "step": 24432} +{"lr": 0.0009757729755661012, "data_time": 0.0013065099716186523, "grad_norm": 0.06457884428091347, "loss": 0.6833694756031037, "time": 0.4407362461090088, "epoch": 3, "memory": 9465, "step": 24532} +{"lr": 0.0009757729755661012, "data_time": 0.0011156320571899414, "grad_norm": 0.040838798554614184, "loss": 0.6834480226039886, "time": 0.3935695171356201, "epoch": 3, "memory": 9465, "step": 24632} +{"lr": 0.0009757729755661012, "data_time": 0.0010981798171997071, "grad_norm": 0.025170611450448632, "loss": 0.6850519716739655, "time": 0.4510664463043213, "epoch": 3, "memory": 9465, "step": 24732} +{"lr": 0.0009757729755661012, "data_time": 0.0014521360397338867, "grad_norm": 0.025674613169394435, "loss": 0.6856638967990876, "time": 0.40546751022338867, "epoch": 3, "memory": 9465, "step": 24832} +{"lr": 0.0009757729755661012, "data_time": 0.0012404680252075194, "grad_norm": 0.027631144598126413, "loss": 0.6887315809726715, "time": 0.4018693208694458, "epoch": 3, "memory": 9465, "step": 24932} +{"lr": 0.0009757729755661012, "data_time": 0.0018202304840087891, "grad_norm": 0.021495186537504197, "loss": 0.6858207285404205, "time": 0.4309407949447632, "epoch": 3, "memory": 9465, "step": 25032} +{"lr": 0.0009757729755661012, "data_time": 0.001217961311340332, "grad_norm": 0.03502406873740256, "loss": 0.6926734924316407, "time": 0.40155487060546874, "epoch": 3, "memory": 9465, "step": 25132} +{"lr": 0.0009757729755661012, "data_time": 0.0013715982437133788, "grad_norm": 0.02867203941568732, "loss": 0.6881754457950592, "time": 0.4332838773727417, "epoch": 3, "memory": 9465, "step": 25232} +{"lr": 0.0009757729755661012, "data_time": 0.001523447036743164, "grad_norm": 0.03737427806481719, "loss": 0.6857482075691224, "time": 0.44575021266937254, "epoch": 3, "memory": 9465, "step": 25332} +{"lr": 0.0009757729755661012, "data_time": 0.0013747453689575196, "grad_norm": 0.02552356361411512, "loss": 0.6898065567016601, "time": 0.4265328884124756, "epoch": 3, "memory": 9465, "step": 25432} +{"lr": 0.0009757729755661012, "data_time": 0.0010834455490112305, "grad_norm": 0.03172828210517764, "loss": 0.6846775054931641, "time": 0.4278853893280029, "epoch": 3, "memory": 9465, "step": 25532} +{"lr": 0.0009757729755661012, "data_time": 0.00110933780670166, "grad_norm": 0.033474443340674044, "loss": 0.6859627842903138, "time": 0.4249113082885742, "epoch": 3, "memory": 9465, "step": 25632} +{"lr": 0.0009757729755661012, "data_time": 0.0010965347290039062, "grad_norm": 0.02725795153528452, "loss": 0.6842111229896546, "time": 0.40430710315704343, "epoch": 3, "memory": 9465, "step": 25732} +{"lr": 0.0009757729755661012, "data_time": 0.0013139009475708007, "grad_norm": 0.027517162123695017, "loss": 0.6904976963996887, "time": 0.41391568183898925, "epoch": 3, "memory": 9465, "step": 25832} +{"lr": 0.0009757729755661012, "data_time": 0.0013836145401000977, "grad_norm": 0.025473095057532192, "loss": 0.6836946249008179, "time": 0.37690105438232424, "epoch": 3, "memory": 9465, "step": 25932} +{"lr": 0.0009757729755661012, "data_time": 0.0013869762420654296, "grad_norm": 0.024343622149899602, "loss": 0.6847755968570709, "time": 0.44075562953948977, "epoch": 3, "memory": 9465, "step": 26032} +{"lr": 0.0009757729755661012, "data_time": 0.0013386011123657227, "grad_norm": 0.02313335072249174, "loss": 0.6847593605518341, "time": 0.4898258924484253, "epoch": 3, "memory": 9465, "step": 26132} +{"lr": 0.0009757729755661012, "data_time": 0.0021242856979370116, "grad_norm": 0.024646509112790226, "loss": 0.6875008940696716, "time": 0.3919433832168579, "epoch": 3, "memory": 9465, "step": 26232} +{"lr": 0.0009757729755661012, "data_time": 0.0016807317733764648, "grad_norm": 0.0246181502006948, "loss": 0.6875292062759399, "time": 0.4523853540420532, "epoch": 3, "memory": 9465, "step": 26332} +{"lr": 0.0009757729755661012, "data_time": 0.0010595083236694335, "grad_norm": 0.019470269698649643, "loss": 0.6877839267253876, "time": 0.3780526876449585, "epoch": 3, "memory": 9465, "step": 26432} +{"lr": 0.0009757729755661012, "data_time": 0.0010975122451782227, "grad_norm": 0.02979953051544726, "loss": 0.6883388817310333, "time": 0.5774923324584961, "epoch": 3, "memory": 9465, "step": 26532} +{"lr": 0.0009757729755661012, "data_time": 0.001065373420715332, "grad_norm": 0.02564713526517153, "loss": 0.6867683947086334, "time": 0.49703762531280515, "epoch": 3, "memory": 9465, "step": 26632} +{"lr": 0.0009757729755661012, "data_time": 0.001167893409729004, "grad_norm": 0.026032890332862733, "loss": 0.6869809746742248, "time": 0.4039982795715332, "epoch": 3, "memory": 9465, "step": 26732} +{"lr": 0.0009757729755661012, "data_time": 0.0011603116989135742, "grad_norm": 0.01670405617915094, "loss": 0.6842011153697968, "time": 0.48672292232513426, "epoch": 3, "memory": 9465, "step": 26832} +{"lr": 0.0009757729755661012, "data_time": 0.0011061906814575195, "grad_norm": 0.026842055190354586, "loss": 0.6836141109466553, "time": 0.45530383586883544, "epoch": 3, "memory": 9465, "step": 26932} +{"lr": 0.0009757729755661012, "data_time": 0.0011821985244750977, "grad_norm": 0.015099308732897044, "loss": 0.6877972960472107, "time": 0.42082149982452394, "epoch": 3, "memory": 9465, "step": 27032} +{"lr": 0.0009757729755661012, "data_time": 0.0011876821517944336, "grad_norm": 0.02831619172357023, "loss": 0.6841572225093842, "time": 0.46534578800201415, "epoch": 3, "memory": 9465, "step": 27132} +{"lr": 0.0009757729755661012, "data_time": 0.0010373353958129882, "grad_norm": 0.017154495231807232, "loss": 0.6849205136299134, "time": 0.40144083499908445, "epoch": 3, "memory": 9465, "step": 27232} +{"lr": 0.0009757729755661012, "data_time": 0.0012067317962646484, "grad_norm": 0.0299342380836606, "loss": 0.684631985425949, "time": 0.4506873607635498, "epoch": 3, "memory": 9465, "step": 27332} +{"lr": 0.0009757729755661012, "data_time": 0.0011445999145507813, "grad_norm": 0.0222253386862576, "loss": 0.6883554875850677, "time": 0.3992605447769165, "epoch": 3, "memory": 9465, "step": 27432} +{"lr": 0.0009757729755661012, "data_time": 0.001079702377319336, "grad_norm": 0.02965725902467966, "loss": 0.687291544675827, "time": 0.42951183319091796, "epoch": 3, "memory": 9465, "step": 27532} +{"lr": 0.0009757729755661012, "data_time": 0.0011515140533447266, "grad_norm": 0.024875061493366955, "loss": 0.6867052674293518, "time": 0.41588122844696046, "epoch": 3, "memory": 9465, "step": 27632} +{"lr": 0.0009757729755661012, "data_time": 0.0012460470199584961, "grad_norm": 0.02218859880231321, "loss": 0.68658407330513, "time": 0.41407184600830077, "epoch": 3, "memory": 9465, "step": 27732} +{"lr": 0.0009757729755661012, "data_time": 0.0012710809707641602, "grad_norm": 0.02219894644804299, "loss": 0.6858759343624115, "time": 0.4324398756027222, "epoch": 3, "memory": 9465, "step": 27832} +{"lr": 0.0009757729755661012, "data_time": 0.001050424575805664, "grad_norm": 0.021095025166869164, "loss": 0.6843375205993653, "time": 0.47747251987457273, "epoch": 3, "memory": 9465, "step": 27932} +{"lr": 0.0009757729755661012, "data_time": 0.0011342287063598633, "grad_norm": 0.018732304126024245, "loss": 0.6914817154407501, "time": 0.48612802028656005, "epoch": 3, "memory": 9465, "step": 28032} +{"lr": 0.0009757729755661012, "data_time": 0.001188182830810547, "grad_norm": 0.02164520784281194, "loss": 0.6878631353378296, "time": 0.3929041147232056, "epoch": 3, "memory": 9465, "step": 28132} +{"lr": 0.0009757729755661012, "data_time": 0.0011054515838623048, "grad_norm": 0.021526194317266344, "loss": 0.685868912935257, "time": 0.40958170890808104, "epoch": 3, "memory": 9465, "step": 28232} +{"lr": 0.0009757729755661012, "data_time": 0.0011272192001342773, "grad_norm": 0.020032352395355703, "loss": 0.6839922964572906, "time": 0.46430985927581786, "epoch": 3, "memory": 9465, "step": 28332} +{"lr": 0.0009757729755661012, "data_time": 0.0011687278747558594, "grad_norm": 0.029056368954479696, "loss": 0.6802111506462097, "time": 0.4719816207885742, "epoch": 3, "memory": 9465, "step": 28432} +{"lr": 0.0009757729755661012, "data_time": 0.0012011051177978516, "grad_norm": 0.026354875555261968, "loss": 0.6879319608211517, "time": 0.4400376558303833, "epoch": 3, "memory": 9465, "step": 28532} +{"lr": 0.0009757729755661012, "data_time": 0.0011624574661254882, "grad_norm": 0.02252791910432279, "loss": 0.6908713400363922, "time": 0.40926480293273926, "epoch": 3, "memory": 9465, "step": 28632} +{"lr": 0.0009757729755661012, "data_time": 0.0010194063186645507, "grad_norm": 0.03153423047624528, "loss": 0.6901094853878021, "time": 0.4298314809799194, "epoch": 3, "memory": 9465, "step": 28732} +{"lr": 0.0009757729755661012, "data_time": 0.001091742515563965, "grad_norm": 0.02403234178200364, "loss": 0.6859693348407745, "time": 0.46756978034973146, "epoch": 3, "memory": 9465, "step": 28832} +{"lr": 0.0009757729755661012, "data_time": 0.0013289213180541991, "grad_norm": 0.0354332847520709, "loss": 0.684811782836914, "time": 0.3715671539306641, "epoch": 3, "memory": 9465, "step": 28932} +{"lr": 0.0009757729755661012, "data_time": 0.0012017011642456055, "grad_norm": 0.022738070320338012, "loss": 0.6883825361728668, "time": 0.4039838075637817, "epoch": 3, "memory": 9465, "step": 29032} +{"lr": 0.0009757729755661012, "data_time": 0.0011409759521484376, "grad_norm": 0.029429469164460896, "loss": 0.6878215253353119, "time": 0.37913217544555666, "epoch": 3, "memory": 9465, "step": 29132} +{"lr": 0.0009757729755661012, "data_time": 0.0011286735534667969, "grad_norm": 0.02399361697025597, "loss": 0.6835292994976043, "time": 0.4256344556808472, "epoch": 3, "memory": 9465, "step": 29232} +{"lr": 0.0009757729755661012, "data_time": 0.0011286258697509766, "grad_norm": 0.01877906555309892, "loss": 0.6891405820846558, "time": 0.4724832534790039, "epoch": 3, "memory": 9465, "step": 29332} +{"lr": 0.0009757729755661012, "data_time": 0.0012565851211547852, "grad_norm": 0.01546756005845964, "loss": 0.682167261838913, "time": 0.3839565753936768, "epoch": 3, "memory": 9465, "step": 29432} +{"lr": 0.0009757729755661012, "data_time": 0.001350259780883789, "grad_norm": 0.02717026323080063, "loss": 0.6854710876941681, "time": 0.3995219230651855, "epoch": 3, "memory": 9465, "step": 29532} +{"lr": 0.0009757729755661012, "data_time": 0.0012035608291625977, "grad_norm": 0.015795053634792566, "loss": 0.6834986269474029, "time": 0.4032184839248657, "epoch": 3, "memory": 9465, "step": 29632} +{"lr": 0.0009757729755661012, "data_time": 0.0013095855712890625, "grad_norm": 0.02186451982706785, "loss": 0.6897809982299805, "time": 0.4221600294113159, "epoch": 3, "memory": 9465, "step": 29732} +{"lr": 0.0009757729755661012, "data_time": 0.0011242389678955077, "grad_norm": 0.022603030130267144, "loss": 0.6875552356243133, "time": 0.3972029447555542, "epoch": 3, "memory": 9465, "step": 29832} +{"lr": 0.0009757729755661012, "data_time": 0.0010331630706787109, "grad_norm": 0.02292302092537284, "loss": 0.6815848410129547, "time": 0.4479137182235718, "epoch": 3, "memory": 9465, "step": 29932} +{"lr": 0.0009757729755661012, "data_time": 0.001129317283630371, "grad_norm": 0.03163904398679733, "loss": 0.6903496265411377, "time": 0.399628734588623, "epoch": 3, "memory": 9465, "step": 30032} +{"lr": 0.0009757729755661012, "data_time": 0.001031637191772461, "grad_norm": 0.029005956277251242, "loss": 0.6859833240509033, "time": 0.398192024230957, "epoch": 3, "memory": 9465, "step": 30132} +{"lr": 0.0009757729755661012, "data_time": 0.0012382984161376954, "grad_norm": 0.026983386278152464, "loss": 0.6888517439365387, "time": 0.3853166103363037, "epoch": 3, "memory": 9465, "step": 30232} +{"lr": 0.0009757729755661012, "data_time": 0.0012030601501464844, "grad_norm": 0.038520692754536866, "loss": 0.686540961265564, "time": 0.38440546989440916, "epoch": 3, "memory": 9465, "step": 30332} +{"lr": 0.0009757729755661012, "data_time": 0.001174759864807129, "grad_norm": 0.02511755982413888, "loss": 0.6896385133266449, "time": 0.39105055332183836, "epoch": 3, "memory": 9465, "step": 30432} +{"lr": 0.0009757729755661012, "data_time": 0.0014658689498901368, "grad_norm": 0.031639125244691965, "loss": 0.6934526026248932, "time": 0.44205350875854493, "epoch": 3, "memory": 9465, "step": 30532} +{"lr": 0.0009757729755661012, "data_time": 0.0012601375579833984, "grad_norm": 0.02406288618221879, "loss": 0.6928059995174408, "time": 0.38973872661590575, "epoch": 3, "memory": 9465, "step": 30632} +{"lr": 0.0009757729755661012, "data_time": 0.0011130332946777343, "grad_norm": 0.020718642440624536, "loss": 0.6865417540073395, "time": 0.44378411769866943, "epoch": 3, "memory": 9465, "step": 30732} +{"lr": 0.0009757729755661012, "data_time": 0.001770782470703125, "grad_norm": 0.022590391407720744, "loss": 0.6830469191074371, "time": 0.41610231399536135, "epoch": 3, "memory": 9465, "step": 30832} +{"lr": 0.0009757729755661012, "data_time": 0.0013607978820800782, "grad_norm": 0.02363686291500926, "loss": 0.6823496639728546, "time": 0.3708021879196167, "epoch": 3, "memory": 9465, "step": 30932} +{"lr": 0.0009757729755661012, "data_time": 0.0010550260543823243, "grad_norm": 0.03002280886285007, "loss": 0.6842589616775513, "time": 0.3915779829025269, "epoch": 3, "memory": 9465, "step": 31032} +{"lr": 0.0009757729755661012, "data_time": 0.0014104366302490235, "grad_norm": 0.021123229106888176, "loss": 0.682874721288681, "time": 0.41668713092803955, "epoch": 3, "memory": 9465, "step": 31132} +{"lr": 0.0009757729755661012, "data_time": 0.001031780242919922, "grad_norm": 0.017836342193186284, "loss": 0.6875498414039611, "time": 0.4248785495758057, "epoch": 3, "memory": 9465, "step": 31232} +{"lr": 0.0009757729755661012, "data_time": 0.0010653972625732423, "grad_norm": 0.02051288477377966, "loss": 0.6856572270393372, "time": 0.47451279163360593, "epoch": 3, "memory": 9465, "step": 31332} +{"lr": 0.0009757729755661012, "data_time": 0.001134777069091797, "grad_norm": 0.01699108590837568, "loss": 0.6913679778575897, "time": 0.41942386627197265, "epoch": 3, "memory": 9465, "step": 31432} +{"lr": 0.0009757729755661012, "data_time": 0.001046442985534668, "grad_norm": 0.026454728469252585, "loss": 0.687749183177948, "time": 0.40201330184936523, "epoch": 3, "memory": 9465, "step": 31532} +{"lr": 0.0009757729755661012, "data_time": 0.0010889291763305664, "grad_norm": 0.026782857766374946, "loss": 0.6820603430271148, "time": 0.38089675903320314, "epoch": 3, "memory": 9465, "step": 31632} +{"lr": 0.0009757729755661012, "data_time": 0.0013752460479736328, "grad_norm": 0.021188395144417883, "loss": 0.6859297692775727, "time": 0.41359283924102785, "epoch": 3, "memory": 9465, "step": 31732} +{"lr": 0.0009757729755661012, "data_time": 0.0012529611587524414, "grad_norm": 0.015946542425081135, "loss": 0.6897715508937836, "time": 0.44652998447418213, "epoch": 3, "memory": 9465, "step": 31832} +{"lr": 0.0009757729755661012, "data_time": 0.001079893112182617, "grad_norm": 0.032581039471551776, "loss": 0.6879302859306335, "time": 0.4039323329925537, "epoch": 3, "memory": 9465, "step": 31932} +{"lr": 0.0009757729755661012, "data_time": 0.0009735107421875, "grad_norm": 0.02621848154813051, "loss": 0.684019786119461, "time": 0.4213979959487915, "epoch": 3, "memory": 9465, "step": 32032} +{"lr": 0.0009757729755661012, "data_time": 0.0010878324508666992, "grad_norm": 0.027288965182378888, "loss": 0.680187976360321, "time": 0.4218135833740234, "epoch": 3, "memory": 9465, "step": 32132} +{"lr": 0.0009757729755661012, "data_time": 0.0010739803314208985, "grad_norm": 0.019247781950980425, "loss": 0.6840679943561554, "time": 0.43177151679992676, "epoch": 3, "memory": 9465, "step": 32232} +{"lr": 0.0009757729755661012, "data_time": 0.0011451005935668944, "grad_norm": 0.021946579916402698, "loss": 0.683692067861557, "time": 0.42153475284576414, "epoch": 3, "memory": 9465, "step": 32332} +{"lr": 0.0009757729755661012, "data_time": 0.0010820627212524414, "grad_norm": 0.026870765211060643, "loss": 0.6886960387229919, "time": 0.4584962844848633, "epoch": 3, "memory": 9465, "step": 32432} +{"lr": 0.0009757729755661012, "data_time": 0.0010807037353515625, "grad_norm": 0.02631870238110423, "loss": 0.682082062959671, "time": 0.4570074319839478, "epoch": 3, "memory": 9465, "step": 32532} +{"lr": 0.0009757729755661012, "data_time": 0.0009947776794433593, "grad_norm": 0.024161409866064788, "loss": 0.6895573616027832, "time": 0.4683802604675293, "epoch": 3, "memory": 9465, "step": 32632} +{"lr": 0.0009757729755661012, "data_time": 0.0010188102722167968, "grad_norm": 0.01994178690947592, "loss": 0.6848426759243011, "time": 0.4365750551223755, "epoch": 3, "memory": 9465, "step": 32732} +{"lr": 0.0009757729755661012, "data_time": 0.0010864019393920898, "grad_norm": 0.025550302490592, "loss": 0.6919171154499054, "time": 0.41815476417541503, "epoch": 3, "memory": 9465, "step": 32832} +{"lr": 0.0009757729755661012, "data_time": 0.0011384248733520507, "grad_norm": 0.01579303703038022, "loss": 0.6893027245998382, "time": 0.4446742534637451, "epoch": 3, "memory": 9465, "step": 32932} +{"lr": 0.0009757729755661012, "data_time": 0.0012124300003051758, "grad_norm": 0.02473419322632253, "loss": 0.6855925440788269, "time": 0.40201406478881835, "epoch": 3, "memory": 9465, "step": 33032} +{"lr": 0.0009757729755661012, "data_time": 0.0013197898864746095, "grad_norm": 0.02229913806077093, "loss": 0.6881005585193634, "time": 0.42860519886016846, "epoch": 3, "memory": 9465, "step": 33132} +{"lr": 0.0009757729755661012, "data_time": 0.001178145408630371, "grad_norm": 0.016469798039179296, "loss": 0.692356276512146, "time": 0.45051984786987304, "epoch": 3, "memory": 9465, "step": 33232} +{"lr": 0.0009757729755661012, "data_time": 0.0009961843490600586, "grad_norm": 0.03090803502127528, "loss": 0.689253854751587, "time": 0.4596554279327393, "epoch": 3, "memory": 9465, "step": 33332} +{"lr": 0.0009757729755661012, "data_time": 0.0011402130126953124, "grad_norm": 0.025316892098635434, "loss": 0.6842014372348786, "time": 0.42744667530059816, "epoch": 3, "memory": 9465, "step": 33432} +{"lr": 0.0009757729755661012, "data_time": 0.0010706424713134766, "grad_norm": 0.022424074914306404, "loss": 0.6886593043804169, "time": 0.39935548305511476, "epoch": 3, "memory": 9465, "step": 33532} +{"lr": 0.0009757729755661012, "data_time": 0.00169680118560791, "grad_norm": 0.02563798618502915, "loss": 0.6867034196853637, "time": 0.4220255374908447, "epoch": 3, "memory": 9465, "step": 33632} +{"lr": 0.0009757729755661012, "data_time": 0.0010977745056152343, "grad_norm": 0.015483600366860629, "loss": 0.6839949071407319, "time": 0.4257720708847046, "epoch": 3, "memory": 9465, "step": 33732} +{"lr": 0.0009757729755661012, "data_time": 0.0013451337814331054, "grad_norm": 0.023556360881775616, "loss": 0.691718477010727, "time": 0.41590681076049807, "epoch": 3, "memory": 9465, "step": 33832} +{"lr": 0.0009757729755661012, "data_time": 0.0010607719421386718, "grad_norm": 0.030128264846280218, "loss": 0.6869981110095977, "time": 0.38972184658050535, "epoch": 3, "memory": 9465, "step": 33932} +{"lr": 0.0009757729755661012, "data_time": 0.001278853416442871, "grad_norm": 0.025171750318259002, "loss": 0.6834398865699768, "time": 0.4456507444381714, "epoch": 3, "memory": 9465, "step": 34032} +{"lr": 0.0009757729755661012, "data_time": 0.0011839628219604491, "grad_norm": 0.02239755503833294, "loss": 0.6860175549983978, "time": 0.4222816228866577, "epoch": 3, "memory": 9465, "step": 34132} +{"lr": 0.0009757729755661012, "data_time": 0.2247065544128418, "grad_norm": 0.029609837243333458, "loss": 0.6806548655033111, "time": 0.4390366554260254, "epoch": 3, "memory": 9465, "step": 34232} +{"lr": 0.0009757729755661012, "data_time": 0.19836778640747071, "grad_norm": 0.021756906434893607, "loss": 0.6878196001052856, "time": 0.3956012487411499, "epoch": 3, "memory": 9465, "step": 34332} +{"lr": 0.0009757729755661012, "data_time": 0.18957347869873048, "grad_norm": 0.028835076210089027, "loss": 0.6858284890651702, "time": 0.37671647071838377, "epoch": 3, "memory": 9465, "step": 34432} +{"lr": 0.0009757729755661012, "data_time": 0.015145373344421387, "grad_norm": 0.024327561492100358, "loss": 0.6859240829944611, "time": 0.3810101509094238, "epoch": 3, "memory": 9465, "step": 34532} +{"accuracy/top1": 100.0, "data_time": 0.32107078327852134, "time": 0.5737238631528967, "step": 3} +{"lr": 0.0009460482294732422, "data_time": 0.0017899036407470702, "grad_norm": 0.03476463651750237, "loss": 0.6878685235977173, "time": 0.40660455226898196, "epoch": 4, "memory": 9465, "step": 34648} +{"lr": 0.0009460482294732422, "data_time": 0.0016024351119995118, "grad_norm": 0.034308757586404684, "loss": 0.6896538257598877, "time": 0.40057930946350095, "epoch": 4, "memory": 9465, "step": 34748} +{"lr": 0.0009460482294732422, "data_time": 0.018561601638793945, "grad_norm": 0.014790708804503084, "loss": 0.6880465269088745, "time": 0.4655344247817993, "epoch": 4, "memory": 9465, "step": 34848} +{"lr": 0.0009460482294732422, "data_time": 0.0016839027404785157, "grad_norm": 0.01557667445158586, "loss": 0.6876308202743531, "time": 0.47255942821502683, "epoch": 4, "memory": 9465, "step": 34948} +{"lr": 0.0009460482294732422, "data_time": 0.00119781494140625, "grad_norm": 0.02648501314688474, "loss": 0.678122740983963, "time": 0.4211986780166626, "epoch": 4, "memory": 9465, "step": 35048} +{"lr": 0.0009460482294732422, "data_time": 0.0012935400009155273, "grad_norm": 0.020421169931069016, "loss": 0.6842995524406433, "time": 0.4139136075973511, "epoch": 4, "memory": 9465, "step": 35148} +{"lr": 0.0009460482294732422, "data_time": 0.5352178335189819, "grad_norm": 0.028799089195672423, "loss": 0.6834197521209717, "time": 0.7197421073913575, "epoch": 4, "memory": 9465, "step": 35248} +{"lr": 0.0009460482294732422, "data_time": 0.22021760940551757, "grad_norm": 0.02672121240757406, "loss": 0.68721644282341, "time": 0.40677647590637206, "epoch": 4, "memory": 9465, "step": 35348} +{"lr": 0.0009460482294732422, "data_time": 0.17384004592895508, "grad_norm": 0.023233586567221208, "loss": 0.6831177651882172, "time": 0.3568373918533325, "epoch": 4, "memory": 9465, "step": 35448} +{"lr": 0.0009460482294732422, "data_time": 0.0783167839050293, "grad_norm": 0.01866559034679085, "loss": 0.6832765281200409, "time": 0.3714780569076538, "epoch": 4, "memory": 9465, "step": 35548} +{"lr": 0.0009460482294732422, "data_time": 0.001073312759399414, "grad_norm": 0.019859836203977466, "loss": 0.6869368433952332, "time": 0.403188419342041, "epoch": 4, "memory": 9465, "step": 35648} +{"lr": 0.0009460482294732422, "data_time": 0.001113748550415039, "grad_norm": 0.018139785109087824, "loss": 0.6897226393222808, "time": 0.43632261753082274, "epoch": 4, "memory": 9465, "step": 35748} +{"lr": 0.0009460482294732422, "data_time": 0.0010857343673706054, "grad_norm": 0.026441440032795072, "loss": 0.6826783895492554, "time": 0.41672937870025634, "epoch": 4, "memory": 9465, "step": 35848} +{"lr": 0.0009460482294732422, "data_time": 0.001102590560913086, "grad_norm": 0.01604182834853418, "loss": 0.6827332258224488, "time": 0.4156961917877197, "epoch": 4, "memory": 9465, "step": 35948} +{"lr": 0.0009460482294732422, "data_time": 0.0011310815811157227, "grad_norm": 0.03293159557506442, "loss": 0.6906485855579376, "time": 0.3799347162246704, "epoch": 4, "memory": 9465, "step": 36048} +{"lr": 0.0009460482294732422, "data_time": 0.001097869873046875, "grad_norm": 0.012823704117909073, "loss": 0.6866336822509765, "time": 0.4036120891571045, "epoch": 4, "memory": 9465, "step": 36148} +{"lr": 0.0009460482294732422, "data_time": 0.0010981082916259766, "grad_norm": 0.03432498765178025, "loss": 0.6885882139205932, "time": 0.3947927951812744, "epoch": 4, "memory": 9465, "step": 36248} +{"lr": 0.0009460482294732422, "data_time": 0.0011486768722534179, "grad_norm": 0.037587924674153325, "loss": 0.6901862502098084, "time": 0.4117638111114502, "epoch": 4, "memory": 9465, "step": 36348} +{"lr": 0.0009460482294732422, "data_time": 0.0011402368545532227, "grad_norm": 0.02379846046678722, "loss": 0.6881829559803009, "time": 0.35527381896972654, "epoch": 4, "memory": 9465, "step": 36448} +{"lr": 0.0009460482294732422, "data_time": 0.001129889488220215, "grad_norm": 0.018451213405933233, "loss": 0.6899665355682373, "time": 0.4215824842453003, "epoch": 4, "memory": 9465, "step": 36548} +{"lr": 0.0009460482294732422, "data_time": 0.0011128902435302735, "grad_norm": 0.02452612491324544, "loss": 0.6846395015716553, "time": 0.41646690368652345, "epoch": 4, "memory": 9465, "step": 36648} +{"lr": 0.0009460482294732422, "data_time": 0.0012495040893554688, "grad_norm": 0.018059630505740644, "loss": 0.6883924603462219, "time": 0.3853238582611084, "epoch": 4, "memory": 9465, "step": 36748} +{"lr": 0.0009460482294732422, "data_time": 0.0011029958724975586, "grad_norm": 0.027530170959653332, "loss": 0.6902863144874573, "time": 0.3810310363769531, "epoch": 4, "memory": 9465, "step": 36848} +{"lr": 0.0009460482294732422, "data_time": 0.0012690305709838867, "grad_norm": 0.024206779804080723, "loss": 0.6883936524391174, "time": 0.377858829498291, "epoch": 4, "memory": 9465, "step": 36948} +{"lr": 0.0009460482294732422, "data_time": 0.0012712955474853515, "grad_norm": 0.023071225523017345, "loss": 0.6838258504867554, "time": 0.4320306062698364, "epoch": 4, "memory": 9465, "step": 37048} +{"lr": 0.0009460482294732422, "data_time": 0.0011139869689941405, "grad_norm": 0.03565419996157289, "loss": 0.6902921795845032, "time": 0.4058772802352905, "epoch": 4, "memory": 9465, "step": 37148} +{"lr": 0.0009460482294732422, "data_time": 0.0011394023895263672, "grad_norm": 0.02153901937417686, "loss": 0.6827071964740753, "time": 0.35878307819366456, "epoch": 4, "memory": 9465, "step": 37248} +{"lr": 0.0009460482294732422, "data_time": 0.001119542121887207, "grad_norm": 0.026963997120037674, "loss": 0.6893606722354889, "time": 0.3967958927154541, "epoch": 4, "memory": 9465, "step": 37348} +{"lr": 0.0009460482294732422, "data_time": 0.0011075496673583984, "grad_norm": 0.02151414748514071, "loss": 0.6861848771572113, "time": 0.4284475803375244, "epoch": 4, "memory": 9465, "step": 37448} +{"lr": 0.0009460482294732422, "data_time": 0.0010763883590698242, "grad_norm": 0.015718328609364108, "loss": 0.6851791739463806, "time": 0.3692357301712036, "epoch": 4, "memory": 9465, "step": 37548} +{"lr": 0.0009460482294732422, "data_time": 0.0010373353958129882, "grad_norm": 0.02049769986188039, "loss": 0.6843901336193084, "time": 0.41892502307891843, "epoch": 4, "memory": 9465, "step": 37648} +{"lr": 0.0009460482294732422, "data_time": 0.0010868310928344727, "grad_norm": 0.02394439820200205, "loss": 0.6859405159950256, "time": 0.44752552509307864, "epoch": 4, "memory": 9465, "step": 37748} +{"lr": 0.0009460482294732422, "data_time": 0.001164698600769043, "grad_norm": 0.017176639661192893, "loss": 0.6845137417316437, "time": 0.43931078910827637, "epoch": 4, "memory": 9465, "step": 37848} +{"lr": 0.0009460482294732422, "data_time": 0.0011846780776977538, "grad_norm": 0.02280675176298246, "loss": 0.6860374450683594, "time": 0.36711075305938723, "epoch": 4, "memory": 9465, "step": 37948} +{"lr": 0.0009460482294732422, "data_time": 0.0012263059616088867, "grad_norm": 0.0154698476777412, "loss": 0.6861841261386872, "time": 0.39556992053985596, "epoch": 4, "memory": 9465, "step": 38048} +{"lr": 0.0009460482294732422, "data_time": 0.0010687589645385742, "grad_norm": 0.02592717092484236, "loss": 0.6884170293807983, "time": 0.39520721435546874, "epoch": 4, "memory": 9465, "step": 38148} +{"lr": 0.0009460482294732422, "data_time": 0.0011953115463256836, "grad_norm": 0.02128877486102283, "loss": 0.6879209697246551, "time": 0.4146375894546509, "epoch": 4, "memory": 9465, "step": 38248} +{"lr": 0.0009460482294732422, "data_time": 0.0011084794998168946, "grad_norm": 0.02099588655983098, "loss": 0.6900817453861237, "time": 0.4229058504104614, "epoch": 4, "memory": 9465, "step": 38348} +{"lr": 0.0009460482294732422, "data_time": 0.001208925247192383, "grad_norm": 0.020350764226168393, "loss": 0.6883242309093476, "time": 0.4033116102218628, "epoch": 4, "memory": 9465, "step": 38448} +{"lr": 0.0009460482294732422, "data_time": 0.0010816335678100586, "grad_norm": 0.02382119637914002, "loss": 0.6869382739067078, "time": 0.3833896636962891, "epoch": 4, "memory": 9465, "step": 38548} +{"lr": 0.0009460482294732422, "data_time": 0.0010337591171264648, "grad_norm": 0.027084255684167148, "loss": 0.6859582245349884, "time": 0.40285491943359375, "epoch": 4, "memory": 9465, "step": 38648} +{"lr": 0.0009460482294732422, "data_time": 0.0010890960693359375, "grad_norm": 0.025976664526388048, "loss": 0.6888171792030334, "time": 0.3666045904159546, "epoch": 4, "memory": 9465, "step": 38748} +{"lr": 0.0009460482294732422, "data_time": 0.001171875, "grad_norm": 0.026826502894982694, "loss": 0.6836021900177002, "time": 0.3953967332839966, "epoch": 4, "memory": 9465, "step": 38848} +{"lr": 0.0009460482294732422, "data_time": 0.0012588024139404297, "grad_norm": 0.01330707217566669, "loss": 0.6877638936042786, "time": 0.4548689842224121, "epoch": 4, "memory": 9465, "step": 38948} +{"lr": 0.0009460482294732422, "data_time": 0.0010664939880371093, "grad_norm": 0.01710113591980189, "loss": 0.6892962992191315, "time": 0.4607393741607666, "epoch": 4, "memory": 9465, "step": 39048} +{"lr": 0.0009460482294732422, "data_time": 0.0012426376342773438, "grad_norm": 0.03672229354269803, "loss": 0.686964362859726, "time": 0.39701979160308837, "epoch": 4, "memory": 9465, "step": 39148} +{"lr": 0.0009460482294732422, "data_time": 0.0012912750244140625, "grad_norm": 0.02750856992788613, "loss": 0.6825249552726745, "time": 0.3837315082550049, "epoch": 4, "memory": 9465, "step": 39248} +{"lr": 0.0009460482294732422, "data_time": 0.0011709928512573242, "grad_norm": 0.02512131310068071, "loss": 0.6823361039161682, "time": 0.4443613290786743, "epoch": 4, "memory": 9465, "step": 39348} +{"lr": 0.0009460482294732422, "data_time": 0.0011461257934570312, "grad_norm": 0.017460952047258615, "loss": 0.6841272652149201, "time": 0.43163836002349854, "epoch": 4, "memory": 9465, "step": 39448} +{"lr": 0.0009460482294732422, "data_time": 0.0012474775314331055, "grad_norm": 0.015957727911882102, "loss": 0.6873302578926086, "time": 0.4154085874557495, "epoch": 4, "memory": 9465, "step": 39548} +{"lr": 0.0009460482294732422, "data_time": 0.001211071014404297, "grad_norm": 0.01755048946943134, "loss": 0.6892419338226319, "time": 0.3902950048446655, "epoch": 4, "memory": 9465, "step": 39648} +{"lr": 0.0009460482294732422, "data_time": 0.0011631965637207032, "grad_norm": 0.02180348220281303, "loss": 0.6907113075256348, "time": 0.4171720027923584, "epoch": 4, "memory": 9465, "step": 39748} +{"lr": 0.0009460482294732422, "data_time": 0.0012620449066162109, "grad_norm": 0.016116890986450017, "loss": 0.6853848397731781, "time": 0.4402278184890747, "epoch": 4, "memory": 9465, "step": 39848} +{"lr": 0.0009460482294732422, "data_time": 0.0011441469192504882, "grad_norm": 0.023416503658518197, "loss": 0.6844816923141479, "time": 0.3723562717437744, "epoch": 4, "memory": 9465, "step": 39948} +{"lr": 0.0009460482294732422, "data_time": 0.0010658740997314454, "grad_norm": 0.01867930297739804, "loss": 0.6819418549537659, "time": 0.4076195478439331, "epoch": 4, "memory": 9465, "step": 40048} +{"lr": 0.0009460482294732422, "data_time": 0.0011111974716186523, "grad_norm": 0.013838802732061594, "loss": 0.6878926634788514, "time": 0.41741237640380857, "epoch": 4, "memory": 9465, "step": 40148} +{"lr": 0.0009460482294732422, "data_time": 0.0011829376220703126, "grad_norm": 0.02090329332277179, "loss": 0.6850333929061889, "time": 0.4069153070449829, "epoch": 4, "memory": 9465, "step": 40248} +{"lr": 0.0009460482294732422, "data_time": 0.0013393640518188476, "grad_norm": 0.017084522580262274, "loss": 0.683957862854004, "time": 0.38308372497558596, "epoch": 4, "memory": 9465, "step": 40348} +{"lr": 0.0009460482294732422, "data_time": 0.0011396169662475585, "grad_norm": 0.031805541133508086, "loss": 0.6905127882957458, "time": 0.4001450777053833, "epoch": 4, "memory": 9465, "step": 40448} +{"lr": 0.0009460482294732422, "data_time": 0.0011447668075561523, "grad_norm": 0.013711705664172768, "loss": 0.6867374300956726, "time": 0.43746190071105956, "epoch": 4, "memory": 9465, "step": 40548} +{"lr": 0.0009460482294732422, "data_time": 0.0010962963104248046, "grad_norm": 0.028451330936513843, "loss": 0.68432577252388, "time": 0.4556013822555542, "epoch": 4, "memory": 9465, "step": 40648} +{"lr": 0.0009460482294732422, "data_time": 0.001048898696899414, "grad_norm": 0.01319067805306986, "loss": 0.6893169283866882, "time": 0.421393084526062, "epoch": 4, "memory": 9465, "step": 40748} +{"lr": 0.0009460482294732422, "data_time": 0.0011313676834106446, "grad_norm": 0.029743977077305316, "loss": 0.6813984453678131, "time": 0.3892336130142212, "epoch": 4, "memory": 9465, "step": 40848} +{"lr": 0.0009460482294732422, "data_time": 0.0010990142822265626, "grad_norm": 0.015915067866444588, "loss": 0.6880536556243897, "time": 0.3737616777420044, "epoch": 4, "memory": 9465, "step": 40948} +{"lr": 0.0009460482294732422, "data_time": 0.0011123180389404296, "grad_norm": 0.019405145803466438, "loss": 0.6847852945327759, "time": 0.44578964710235597, "epoch": 4, "memory": 9465, "step": 41048} +{"lr": 0.0009460482294732422, "data_time": 0.0009724617004394532, "grad_norm": 0.03085338226519525, "loss": 0.6798136293888092, "time": 0.4215927839279175, "epoch": 4, "memory": 9465, "step": 41148} +{"lr": 0.0009460482294732422, "data_time": 0.0011057138442993164, "grad_norm": 0.02757742926478386, "loss": 0.6880335152149201, "time": 0.3928692102432251, "epoch": 4, "memory": 9465, "step": 41248} +{"lr": 0.0009460482294732422, "data_time": 0.0012017011642456055, "grad_norm": 0.024022822733968496, "loss": 0.6902175903320312, "time": 0.4202180147171021, "epoch": 4, "memory": 9465, "step": 41348} +{"lr": 0.0009460482294732422, "data_time": 0.001481342315673828, "grad_norm": 0.030624729616101833, "loss": 0.6884732663631439, "time": 0.43123784065246584, "epoch": 4, "memory": 9465, "step": 41448} +{"lr": 0.0009460482294732422, "data_time": 0.0011002779006958007, "grad_norm": 0.028252162295393644, "loss": 0.685443776845932, "time": 0.40383124351501465, "epoch": 4, "memory": 9465, "step": 41548} +{"lr": 0.0009460482294732422, "data_time": 0.0011414051055908202, "grad_norm": 0.025519554503262043, "loss": 0.6831911206245422, "time": 0.42822425365447997, "epoch": 4, "memory": 9465, "step": 41648} +{"lr": 0.0009460482294732422, "data_time": 0.0011412382125854492, "grad_norm": 0.02709073149599135, "loss": 0.6852254331111908, "time": 0.43169431686401366, "epoch": 4, "memory": 9465, "step": 41748} +{"lr": 0.0009460482294732422, "data_time": 0.001149606704711914, "grad_norm": 0.019600513426121326, "loss": 0.6874465703964233, "time": 0.3934406042098999, "epoch": 4, "memory": 9465, "step": 41848} +{"lr": 0.0009460482294732422, "data_time": 0.0011408567428588868, "grad_norm": 0.02341573985759169, "loss": 0.6857075035572052, "time": 0.4015033483505249, "epoch": 4, "memory": 9465, "step": 41948} +{"lr": 0.0009460482294732422, "data_time": 0.001144099235534668, "grad_norm": 0.022814054088667036, "loss": 0.684168541431427, "time": 0.41434321403503416, "epoch": 4, "memory": 9465, "step": 42048} +{"lr": 0.0009460482294732422, "data_time": 0.001092243194580078, "grad_norm": 0.026281535206362606, "loss": 0.6876562416553498, "time": 0.4789563179016113, "epoch": 4, "memory": 9465, "step": 42148} +{"lr": 0.0009460482294732422, "data_time": 0.0011058092117309571, "grad_norm": 0.030756097449921072, "loss": 0.6876628339290619, "time": 0.40562398433685304, "epoch": 4, "memory": 9465, "step": 42248} +{"lr": 0.0009460482294732422, "data_time": 0.0010761499404907226, "grad_norm": 0.020979050220921634, "loss": 0.6930504322052002, "time": 0.44523024559020996, "epoch": 4, "memory": 9465, "step": 42348} +{"lr": 0.0009460482294732422, "data_time": 0.001087021827697754, "grad_norm": 0.02626004386693239, "loss": 0.6856265604496002, "time": 0.40987200736999513, "epoch": 4, "memory": 9465, "step": 42448} +{"lr": 0.0009460482294732422, "data_time": 0.001085042953491211, "grad_norm": 0.028808306576684116, "loss": 0.6887922346591949, "time": 0.4446295499801636, "epoch": 4, "memory": 9465, "step": 42548} +{"lr": 0.0009460482294732422, "data_time": 0.0010744810104370117, "grad_norm": 0.03397733778692782, "loss": 0.688660454750061, "time": 0.44204392433166506, "epoch": 4, "memory": 9465, "step": 42648} +{"lr": 0.0009460482294732422, "data_time": 0.0013291358947753907, "grad_norm": 0.021859666192904116, "loss": 0.6858417630195618, "time": 0.40098745822906495, "epoch": 4, "memory": 9465, "step": 42748} +{"lr": 0.0009460482294732422, "data_time": 0.0013178586959838867, "grad_norm": 0.028944857395254077, "loss": 0.6886160969734192, "time": 0.4137450933456421, "epoch": 4, "memory": 9465, "step": 42848} +{"lr": 0.0009460482294732422, "data_time": 0.0013623237609863281, "grad_norm": 0.033438387070782484, "loss": 0.6898542106151581, "time": 0.4142767429351807, "epoch": 4, "memory": 9465, "step": 42948} +{"lr": 0.0009460482294732422, "data_time": 0.00107574462890625, "grad_norm": 0.02505057826638222, "loss": 0.6842765569686889, "time": 0.4109280824661255, "epoch": 4, "memory": 9465, "step": 43048} +{"lr": 0.0009460482294732422, "data_time": 0.0012395858764648437, "grad_norm": 0.37447545626200734, "loss": 0.6884382665157318, "time": 0.40226519107818604, "epoch": 4, "memory": 9465, "step": 43148} +{"lr": 0.0009460482294732422, "data_time": 0.0012097835540771484, "grad_norm": 0.03758322820067406, "loss": 0.6867580950260163, "time": 0.39088571071624756, "epoch": 4, "memory": 9465, "step": 43248} +{"lr": 0.0009460482294732422, "data_time": 0.0010784387588500977, "grad_norm": 0.02749922494404018, "loss": 0.689893227815628, "time": 0.36574792861938477, "epoch": 4, "memory": 9465, "step": 43348} +{"lr": 0.0009460482294732422, "data_time": 0.0011917829513549804, "grad_norm": 0.027267923951148985, "loss": 0.6861893355846405, "time": 0.4374396324157715, "epoch": 4, "memory": 9465, "step": 43448} +{"lr": 0.0009460482294732422, "data_time": 0.0010781049728393554, "grad_norm": 0.018396165641024708, "loss": 0.6917386293411255, "time": 0.40561809539794924, "epoch": 4, "memory": 9465, "step": 43548} +{"lr": 0.0009460482294732422, "data_time": 0.001032567024230957, "grad_norm": 0.018677343521267174, "loss": 0.6840654134750366, "time": 0.3888076305389404, "epoch": 4, "memory": 9465, "step": 43648} +{"lr": 0.0009460482294732422, "data_time": 0.0011232852935791015, "grad_norm": 0.03051765011623502, "loss": 0.6902405619621277, "time": 0.45583975315093994, "epoch": 4, "memory": 9465, "step": 43748} +{"lr": 0.0009460482294732422, "data_time": 0.001131129264831543, "grad_norm": 0.022079009166918694, "loss": 0.6826335966587067, "time": 0.3682593107223511, "epoch": 4, "memory": 9465, "step": 43848} +{"lr": 0.0009460482294732422, "data_time": 0.0011067867279052734, "grad_norm": 0.0220866784802638, "loss": 0.6860295534133911, "time": 0.3993252754211426, "epoch": 4, "memory": 9465, "step": 43948} +{"lr": 0.0009460482294732422, "data_time": 0.0013279199600219726, "grad_norm": 0.020557160209864377, "loss": 0.6874822616577149, "time": 0.3809523344039917, "epoch": 4, "memory": 9465, "step": 44048} +{"lr": 0.0009460482294732422, "data_time": 0.0012639760971069336, "grad_norm": 0.017060778685845435, "loss": 0.6902158260345459, "time": 0.3801664113998413, "epoch": 4, "memory": 9465, "step": 44148} +{"lr": 0.0009460482294732422, "data_time": 0.0011796236038208007, "grad_norm": 0.015369787137024105, "loss": 0.6870765745639801, "time": 0.34963796138763426, "epoch": 4, "memory": 9465, "step": 44248} +{"lr": 0.0009460482294732422, "data_time": 0.0011382102966308594, "grad_norm": 0.031882503814995286, "loss": 0.6848923325538635, "time": 0.4465653657913208, "epoch": 4, "memory": 9465, "step": 44348} +{"lr": 0.0009460482294732422, "data_time": 0.0011010408401489259, "grad_norm": 0.019127347599714996, "loss": 0.6856461822986603, "time": 0.3779271364212036, "epoch": 4, "memory": 9465, "step": 44448} +{"lr": 0.0009460482294732422, "data_time": 0.0012026786804199218, "grad_norm": 0.024599857884459196, "loss": 0.6908772110939025, "time": 0.373524808883667, "epoch": 4, "memory": 9465, "step": 44548} +{"lr": 0.0009460482294732422, "data_time": 0.0012852907180786132, "grad_norm": 0.015454904607031495, "loss": 0.6853369414806366, "time": 0.4327890157699585, "epoch": 4, "memory": 9465, "step": 44648} +{"lr": 0.0009460482294732422, "data_time": 0.0012766599655151367, "grad_norm": 0.02203433304093778, "loss": 0.6851997554302216, "time": 0.41467597484588625, "epoch": 4, "memory": 9465, "step": 44748} +{"lr": 0.0009460482294732422, "data_time": 0.0013851642608642579, "grad_norm": 0.02830750960856676, "loss": 0.6890788674354553, "time": 0.38348381519317626, "epoch": 4, "memory": 9465, "step": 44848} +{"lr": 0.0009460482294732422, "data_time": 0.0010076999664306641, "grad_norm": 0.01999299090821296, "loss": 0.6843239605426789, "time": 0.4219272375106812, "epoch": 4, "memory": 9465, "step": 44948} +{"lr": 0.0009460482294732422, "data_time": 0.0011414527893066407, "grad_norm": 0.019654296827502547, "loss": 0.6835480213165284, "time": 0.3832509517669678, "epoch": 4, "memory": 9465, "step": 45048} +{"lr": 0.0009460482294732422, "data_time": 0.001023721694946289, "grad_norm": 0.021794243599288166, "loss": 0.686592161655426, "time": 0.4207120180130005, "epoch": 4, "memory": 9465, "step": 45148} +{"lr": 0.0009460482294732422, "data_time": 0.0012047767639160156, "grad_norm": 0.022075356566347182, "loss": 0.6841388285160065, "time": 0.3922042608261108, "epoch": 4, "memory": 9465, "step": 45248} +{"lr": 0.0009460482294732422, "data_time": 0.001503300666809082, "grad_norm": 0.014208039059303701, "loss": 0.6885696828365326, "time": 0.41503353118896485, "epoch": 4, "memory": 9465, "step": 45348} +{"lr": 0.0009460482294732422, "data_time": 0.0013016700744628907, "grad_norm": 0.024549993546679617, "loss": 0.6844257116317749, "time": 0.38327693939208984, "epoch": 4, "memory": 9465, "step": 45448} +{"lr": 0.0009460482294732422, "data_time": 0.0010725021362304687, "grad_norm": 0.030095902178436516, "loss": 0.6890177369117737, "time": 0.38603711128234863, "epoch": 4, "memory": 9465, "step": 45548} +{"lr": 0.0009460482294732422, "data_time": 0.001148390769958496, "grad_norm": 0.04237807409372181, "loss": 0.6897955179214478, "time": 0.3657516956329346, "epoch": 4, "memory": 9465, "step": 45648} +{"lr": 0.0009460482294732422, "data_time": 0.0011873483657836915, "grad_norm": 0.017647675471380353, "loss": 0.6829340755939484, "time": 0.3857662916183472, "epoch": 4, "memory": 9465, "step": 45748} +{"lr": 0.0009460482294732422, "data_time": 0.0018066167831420898, "grad_norm": 0.02500118047464639, "loss": 0.6883044838905334, "time": 0.3862330675125122, "epoch": 4, "memory": 9465, "step": 45848} +{"lr": 0.0009460482294732422, "data_time": 0.001381826400756836, "grad_norm": 0.029156508552841843, "loss": 0.6860745310783386, "time": 0.4018967628479004, "epoch": 4, "memory": 9465, "step": 45948} +{"lr": 0.0009460482294732422, "data_time": 0.007388973236083984, "grad_norm": 0.024827901367098094, "loss": 0.6878408789634705, "time": 0.3983702898025513, "epoch": 4, "memory": 9465, "step": 46048} +{"accuracy/top1": 100.0, "data_time": 0.2998020087971407, "time": 0.5581597861121682, "step": 4} +{"lr": 0.0009054634122155991, "data_time": 0.0012440919876098634, "grad_norm": 0.026587527780793607, "loss": 0.6880975723266601, "time": 0.41617484092712403, "epoch": 5, "memory": 9465, "step": 46164} +{"lr": 0.0009054634122155991, "data_time": 0.001371908187866211, "grad_norm": 0.029598460206761955, "loss": 0.6811716258525848, "time": 0.3711184024810791, "epoch": 5, "memory": 9465, "step": 46264} +{"lr": 0.0009054634122155991, "data_time": 0.0013388872146606445, "grad_norm": 0.00921646914212033, "loss": 0.6854901432991027, "time": 0.3951786756515503, "epoch": 5, "memory": 9465, "step": 46364} +{"lr": 0.0009054634122155991, "data_time": 0.0012554168701171876, "grad_norm": 0.02585256015881896, "loss": 0.6859726309776306, "time": 0.4122645139694214, "epoch": 5, "memory": 9465, "step": 46464} +{"lr": 0.0009054634122155991, "data_time": 0.0010745763778686524, "grad_norm": 0.029718164424411952, "loss": 0.6889049768447876, "time": 0.3991572856903076, "epoch": 5, "memory": 9465, "step": 46564} +{"lr": 0.0009054634122155991, "data_time": 0.0011084794998168946, "grad_norm": 0.027763966005295515, "loss": 0.6881490051746368, "time": 0.44162285327911377, "epoch": 5, "memory": 9465, "step": 46664} +{"lr": 0.0009054634122155991, "data_time": 0.0011112213134765625, "grad_norm": 0.018291403772309422, "loss": 0.6887192845344543, "time": 0.3962730407714844, "epoch": 5, "memory": 9465, "step": 46764} +{"lr": 0.0009054634122155991, "data_time": 0.0011777162551879882, "grad_norm": 0.01177821836899966, "loss": 0.6876420617103577, "time": 0.38205268383026125, "epoch": 5, "memory": 9465, "step": 46864} +{"lr": 0.0009054634122155991, "data_time": 0.0012171506881713868, "grad_norm": 0.01986559834331274, "loss": 0.6813242316246033, "time": 0.397896933555603, "epoch": 5, "memory": 9465, "step": 46964} +{"lr": 0.0009054634122155991, "data_time": 0.0010857582092285156, "grad_norm": 0.02631110083311796, "loss": 0.6859330415725708, "time": 0.37413601875305175, "epoch": 5, "memory": 9465, "step": 47064} +{"lr": 0.0009054634122155991, "data_time": 0.0010780811309814454, "grad_norm": 0.035640973143745217, "loss": 0.6877168595790863, "time": 0.3792332887649536, "epoch": 5, "memory": 9465, "step": 47164} +{"lr": 0.0009054634122155991, "data_time": 0.0010349273681640625, "grad_norm": 0.02035343919415027, "loss": 0.6863429069519043, "time": 0.40172040462493896, "epoch": 5, "memory": 9465, "step": 47264} +{"lr": 0.0009054634122155991, "data_time": 0.001057267189025879, "grad_norm": 0.01496747275814414, "loss": 0.6831871330738067, "time": 0.38187637329101565, "epoch": 5, "memory": 9465, "step": 47364} +{"lr": 0.0009054634122155991, "data_time": 0.0010881900787353515, "grad_norm": 0.015223153587430715, "loss": 0.68567755818367, "time": 0.3965941905975342, "epoch": 5, "memory": 9465, "step": 47464} +{"lr": 0.0009054634122155991, "data_time": 0.0010868310928344727, "grad_norm": 0.024140493012964725, "loss": 0.6896021783351898, "time": 0.4375352144241333, "epoch": 5, "memory": 9465, "step": 47564} +{"lr": 0.0009054634122155991, "data_time": 0.0011009931564331054, "grad_norm": 0.017783907242119312, "loss": 0.6912515997886658, "time": 0.4043778657913208, "epoch": 5, "memory": 9465, "step": 47664} +{"lr": 0.0009054634122155991, "data_time": 0.0010442495346069335, "grad_norm": 0.0255945828743279, "loss": 0.6877342224121094, "time": 0.40841400623321533, "epoch": 5, "memory": 9465, "step": 47764} +{"lr": 0.0009054634122155991, "data_time": 0.0011137723922729492, "grad_norm": 0.02683678106404841, "loss": 0.6868508338928223, "time": 0.381062650680542, "epoch": 5, "memory": 9465, "step": 47864} +{"lr": 0.0009054634122155991, "data_time": 0.001055598258972168, "grad_norm": 0.020497567194979638, "loss": 0.6814903676509857, "time": 0.3894315004348755, "epoch": 5, "memory": 9465, "step": 47964} +{"lr": 0.0009054634122155991, "data_time": 0.0011021852493286132, "grad_norm": 0.031537147238850594, "loss": 0.6888923287391663, "time": 0.3879029035568237, "epoch": 5, "memory": 9465, "step": 48064} +{"lr": 0.0009054634122155991, "data_time": 0.0010887861251831054, "grad_norm": 0.022996595688164236, "loss": 0.6833690822124481, "time": 0.41205859184265137, "epoch": 5, "memory": 9465, "step": 48164} +{"lr": 0.0009054634122155991, "data_time": 0.0011677503585815429, "grad_norm": 0.03175321174785495, "loss": 0.6875175476074219, "time": 0.3933622121810913, "epoch": 5, "memory": 9465, "step": 48264} +{"lr": 0.0009054634122155991, "data_time": 0.001057910919189453, "grad_norm": 0.021280095819383858, "loss": 0.687904554605484, "time": 0.3971649408340454, "epoch": 5, "memory": 9465, "step": 48364} +{"lr": 0.0009054634122155991, "data_time": 0.001323866844177246, "grad_norm": 0.022431167401373388, "loss": 0.6828836560249328, "time": 0.39847261905670167, "epoch": 5, "memory": 9465, "step": 48464} +{"lr": 0.0009054634122155991, "data_time": 0.0011553049087524414, "grad_norm": 0.026961484737694264, "loss": 0.6908922612667083, "time": 0.37170188426971434, "epoch": 5, "memory": 9465, "step": 48564} +{"lr": 0.0009054634122155991, "data_time": 0.001134490966796875, "grad_norm": 0.028834646474570035, "loss": 0.6875865757465363, "time": 0.4479071855545044, "epoch": 5, "memory": 9465, "step": 48664} +{"lr": 0.0009054634122155991, "data_time": 0.001175093650817871, "grad_norm": 0.020229836599901317, "loss": 0.6805278599262238, "time": 0.432504677772522, "epoch": 5, "memory": 9465, "step": 48764} +{"lr": 0.0009054634122155991, "data_time": 0.0011219501495361329, "grad_norm": 0.02679470703005791, "loss": 0.6873334586620331, "time": 0.383939790725708, "epoch": 5, "memory": 9465, "step": 48864} +{"lr": 0.0009054634122155991, "data_time": 0.0011063814163208008, "grad_norm": 0.03192573469132185, "loss": 0.6838356375694274, "time": 0.41294288635253906, "epoch": 5, "memory": 9465, "step": 48964} +{"lr": 0.0009054634122155991, "data_time": 0.0010793447494506837, "grad_norm": 0.0175002969102934, "loss": 0.6861941516399384, "time": 0.39840657711029054, "epoch": 5, "memory": 9465, "step": 49064} +{"lr": 0.0009054634122155991, "data_time": 0.0015295982360839845, "grad_norm": 0.010096137854270637, "loss": 0.6816293716430664, "time": 0.47495198249816895, "epoch": 5, "memory": 9465, "step": 49164} +{"lr": 0.0009054634122155991, "data_time": 0.0012670278549194336, "grad_norm": 0.02322162096388638, "loss": 0.6870709776878356, "time": 0.3975664615631104, "epoch": 5, "memory": 9465, "step": 49264} +{"lr": 0.0009054634122155991, "data_time": 0.0010991334915161134, "grad_norm": 0.028841956635005772, "loss": 0.6840360999107361, "time": 0.36883208751678465, "epoch": 5, "memory": 9465, "step": 49364} +{"lr": 0.0009054634122155991, "data_time": 0.0011717557907104492, "grad_norm": 0.025854251813143492, "loss": 0.6896846652030945, "time": 0.4344851732254028, "epoch": 5, "memory": 9465, "step": 49464} +{"lr": 0.0009054634122155991, "data_time": 0.0012343645095825196, "grad_norm": 0.03077122033573687, "loss": 0.6824256181716919, "time": 0.4099503755569458, "epoch": 5, "memory": 9465, "step": 49564} +{"lr": 0.0009054634122155991, "data_time": 0.0012484073638916015, "grad_norm": 0.02699918523430824, "loss": 0.6889729678630829, "time": 0.3828179359436035, "epoch": 5, "memory": 9465, "step": 49664} +{"lr": 0.0009054634122155991, "data_time": 0.001108837127685547, "grad_norm": 0.029219097923487425, "loss": 0.6889478445053101, "time": 0.36678876876831057, "epoch": 5, "memory": 9465, "step": 49764} +{"lr": 0.0009054634122155991, "data_time": 0.0010905265808105469, "grad_norm": 0.02324791522696614, "loss": 0.6852601051330567, "time": 0.43106679916381835, "epoch": 5, "memory": 9465, "step": 49864} +{"lr": 0.0009054634122155991, "data_time": 0.001279759407043457, "grad_norm": 0.02332518082112074, "loss": 0.6892894983291626, "time": 0.390034556388855, "epoch": 5, "memory": 9465, "step": 49964} +{"lr": 0.0009054634122155991, "data_time": 0.0011096477508544921, "grad_norm": 0.028612617636099458, "loss": 0.685692286491394, "time": 0.4111783981323242, "epoch": 5, "memory": 9465, "step": 50064} +{"lr": 0.0009054634122155991, "data_time": 0.0011332273483276368, "grad_norm": 0.030447939690202475, "loss": 0.6874166846275329, "time": 0.3804218053817749, "epoch": 5, "memory": 9465, "step": 50164} +{"lr": 0.0009054634122155991, "data_time": 0.0011574983596801757, "grad_norm": 0.02650841875001788, "loss": 0.6892519533634186, "time": 0.40076098442077634, "epoch": 5, "memory": 9465, "step": 50264} +{"lr": 0.0009054634122155991, "data_time": 0.001217055320739746, "grad_norm": 0.015318753337487579, "loss": 0.680166357755661, "time": 0.40390989780426023, "epoch": 5, "memory": 9465, "step": 50364} +{"lr": 0.0009054634122155991, "data_time": 0.0012067079544067383, "grad_norm": 0.026635066093876957, "loss": 0.6873058319091797, "time": 0.39262354373931885, "epoch": 5, "memory": 9465, "step": 50464} +{"lr": 0.0009054634122155991, "data_time": 0.0012506961822509766, "grad_norm": 0.0180527399701532, "loss": 0.6869944036006927, "time": 0.41787846088409425, "epoch": 5, "memory": 9465, "step": 50564} +{"lr": 0.0009054634122155991, "data_time": 0.0010959863662719726, "grad_norm": 0.01809955690987408, "loss": 0.6925374031066894, "time": 0.3686471939086914, "epoch": 5, "memory": 9465, "step": 50664} +{"lr": 0.0009054634122155991, "data_time": 0.0012143611907958984, "grad_norm": 0.023882671422325075, "loss": 0.6909743010997772, "time": 0.39217796325683596, "epoch": 5, "memory": 9465, "step": 50764} +{"lr": 0.0009054634122155991, "data_time": 0.0012526273727416991, "grad_norm": 0.03057125969789922, "loss": 0.6895804286003113, "time": 0.3557945013046265, "epoch": 5, "memory": 9465, "step": 50864} +{"lr": 0.0009054634122155991, "data_time": 0.0011016607284545898, "grad_norm": 0.022335830773226917, "loss": 0.6903132677078248, "time": 0.39890298843383787, "epoch": 5, "memory": 9465, "step": 50964} +{"lr": 0.0009054634122155991, "data_time": 0.0014529943466186524, "grad_norm": 0.026697930041700603, "loss": 0.6855130016803741, "time": 0.4034978151321411, "epoch": 5, "memory": 9465, "step": 51064} +{"lr": 0.0009054634122155991, "data_time": 0.0017690181732177735, "grad_norm": 0.025505088176578282, "loss": 0.6853675186634064, "time": 0.37899184226989746, "epoch": 5, "memory": 9465, "step": 51164} +{"lr": 0.0009054634122155991, "data_time": 0.0012786149978637694, "grad_norm": 0.034993641055189074, "loss": 0.681854385137558, "time": 0.41874496936798095, "epoch": 5, "memory": 9465, "step": 51264} +{"lr": 0.0009054634122155991, "data_time": 0.0013869285583496093, "grad_norm": 0.01751966478768736, "loss": 0.6884867131710053, "time": 0.38738670349121096, "epoch": 5, "memory": 9465, "step": 51364} +{"lr": 0.0009054634122155991, "data_time": 0.0012630462646484376, "grad_norm": 0.029562947619706393, "loss": 0.6862551987171173, "time": 0.4115957736968994, "epoch": 5, "memory": 9465, "step": 51464} +{"lr": 0.0009054634122155991, "data_time": 0.001076078414916992, "grad_norm": 0.02607720545493066, "loss": 0.6865394115447998, "time": 0.41392245292663576, "epoch": 5, "memory": 9465, "step": 51564} +{"lr": 0.0009054634122155991, "data_time": 0.0010185956954956055, "grad_norm": 0.031732980720698835, "loss": 0.6904223203659058, "time": 0.3920707941055298, "epoch": 5, "memory": 9465, "step": 51664} +{"lr": 0.0009054634122155991, "data_time": 0.0011585712432861327, "grad_norm": 0.016031731851398945, "loss": 0.6851561784744262, "time": 0.38847401142120364, "epoch": 5, "memory": 9465, "step": 51764} +{"lr": 0.0009054634122155991, "data_time": 0.0011475086212158203, "grad_norm": 0.01787509429268539, "loss": 0.6870312750339508, "time": 0.3833015441894531, "epoch": 5, "memory": 9465, "step": 51864} +{"lr": 0.0009054634122155991, "data_time": 0.0011997222900390625, "grad_norm": 0.022489652619697154, "loss": 0.6870958864688873, "time": 0.40016496181488037, "epoch": 5, "memory": 9465, "step": 51964} +{"lr": 0.0009054634122155991, "data_time": 0.0010323524475097656, "grad_norm": 0.025026615266688167, "loss": 0.6886358857154846, "time": 0.40814259052276614, "epoch": 5, "memory": 9465, "step": 52064} +{"lr": 0.0009054634122155991, "data_time": 0.0011286497116088866, "grad_norm": 0.0235617560101673, "loss": 0.6871602952480316, "time": 0.4007275104522705, "epoch": 5, "memory": 9465, "step": 52164} +{"lr": 0.0009054634122155991, "data_time": 0.001175069808959961, "grad_norm": 0.023988850094610825, "loss": 0.6903709650039673, "time": 0.40066406726837156, "epoch": 5, "memory": 9465, "step": 52264} +{"lr": 0.0009054634122155991, "data_time": 0.0011972904205322266, "grad_norm": 0.02772388606099412, "loss": 0.6898591458797455, "time": 0.4022976875305176, "epoch": 5, "memory": 9465, "step": 52364} +{"lr": 0.0009054634122155991, "data_time": 0.0011831283569335937, "grad_norm": 0.02243322432041168, "loss": 0.6849162459373475, "time": 0.4304262638092041, "epoch": 5, "memory": 9465, "step": 52464} +{"lr": 0.0009054634122155991, "data_time": 0.0012135744094848634, "grad_norm": 0.025075363041833044, "loss": 0.686655443906784, "time": 0.38414065837860106, "epoch": 5, "memory": 9465, "step": 52564} +{"lr": 0.0009054634122155991, "data_time": 0.0013298988342285156, "grad_norm": 0.020284536166582257, "loss": 0.6933638453483582, "time": 0.3798767328262329, "epoch": 5, "memory": 9465, "step": 52664} +{"lr": 0.0009054634122155991, "data_time": 0.0010272979736328125, "grad_norm": 0.036314313393086196, "loss": 0.6914688766002655, "time": 0.39915931224823, "epoch": 5, "memory": 9465, "step": 52764} +{"lr": 0.0009054634122155991, "data_time": 0.0012793779373168946, "grad_norm": 0.024953949404880403, "loss": 0.6863766312599182, "time": 0.39757044315338136, "epoch": 5, "memory": 9465, "step": 52864} +{"lr": 0.0009054634122155991, "data_time": 0.0011752843856811523, "grad_norm": 0.01782432678155601, "loss": 0.6870884716510772, "time": 0.38098416328430174, "epoch": 5, "memory": 9465, "step": 52964} +{"lr": 0.0009054634122155991, "data_time": 0.0010037899017333984, "grad_norm": 0.026098742755129932, "loss": 0.6859421968460083, "time": 0.4036207914352417, "epoch": 5, "memory": 9465, "step": 53064} +{"lr": 0.0009054634122155991, "data_time": 0.0011294126510620118, "grad_norm": 0.026348127191886304, "loss": 0.6870138049125671, "time": 0.40224056243896483, "epoch": 5, "memory": 9465, "step": 53164} +{"lr": 0.0009054634122155991, "data_time": 0.0011500835418701172, "grad_norm": 0.018034082953818142, "loss": 0.6889199793338776, "time": 0.39376053810119627, "epoch": 5, "memory": 9465, "step": 53264} +{"lr": 0.0009054634122155991, "data_time": 0.0010768413543701173, "grad_norm": 0.027831611060537397, "loss": 0.6899236142635345, "time": 0.4256305456161499, "epoch": 5, "memory": 9465, "step": 53364} +{"lr": 0.0009054634122155991, "data_time": 0.001101374626159668, "grad_norm": 0.026979637006297706, "loss": 0.6827156960964202, "time": 0.35997519493103025, "epoch": 5, "memory": 9465, "step": 53464} +{"lr": 0.0009054634122155991, "data_time": 0.001135849952697754, "grad_norm": 0.02476250040344894, "loss": 0.6896296381950379, "time": 0.38420493602752687, "epoch": 5, "memory": 9465, "step": 53564} +{"lr": 0.0009054634122155991, "data_time": 0.0011581897735595704, "grad_norm": 0.02668171711266041, "loss": 0.6884955704212189, "time": 0.3978543281555176, "epoch": 5, "memory": 9465, "step": 53664} +{"lr": 0.0009054634122155991, "data_time": 0.0011167526245117188, "grad_norm": 0.021480609849095344, "loss": 0.6848095834255219, "time": 0.42375526428222654, "epoch": 5, "memory": 9465, "step": 53764} +{"lr": 0.0009054634122155991, "data_time": 0.0011123418807983398, "grad_norm": 0.017277513444423676, "loss": 0.6869917929172515, "time": 0.3655188798904419, "epoch": 5, "memory": 9465, "step": 53864} +{"lr": 0.0009054634122155991, "data_time": 0.001119375228881836, "grad_norm": 0.019226023933151738, "loss": 0.6887372136116028, "time": 0.38648648262023927, "epoch": 5, "memory": 9465, "step": 53964} +{"lr": 0.0009054634122155991, "data_time": 0.0011758089065551757, "grad_norm": 0.019273080118000507, "loss": 0.6879786312580108, "time": 0.39363129138946534, "epoch": 5, "memory": 9465, "step": 54064} +{"lr": 0.0009054634122155991, "data_time": 0.0011520147323608398, "grad_norm": 0.02052807351574302, "loss": 0.6910687863826752, "time": 0.3610013723373413, "epoch": 5, "memory": 9465, "step": 54164} +{"lr": 0.0009054634122155991, "data_time": 0.0012211084365844726, "grad_norm": 0.022979992744512855, "loss": 0.6904963552951813, "time": 0.38689513206481935, "epoch": 5, "memory": 9465, "step": 54264} +{"lr": 0.0009054634122155991, "data_time": 0.001228165626525879, "grad_norm": 0.016480088559910656, "loss": 0.688277804851532, "time": 0.3878389120101929, "epoch": 5, "memory": 9465, "step": 54364} +{"lr": 0.0009054634122155991, "data_time": 0.0011260271072387694, "grad_norm": 0.018702167994342746, "loss": 0.6865172922611237, "time": 0.3749183177947998, "epoch": 5, "memory": 9465, "step": 54464} +{"lr": 0.0009054634122155991, "data_time": 0.0011040449142456054, "grad_norm": 0.029707753891125323, "loss": 0.6900529026985168, "time": 0.39838101863861086, "epoch": 5, "memory": 9465, "step": 54564} +{"lr": 0.0009054634122155991, "data_time": 0.001137709617614746, "grad_norm": 0.014718397427350283, "loss": 0.6826073586940765, "time": 0.39407494068145754, "epoch": 5, "memory": 9465, "step": 54664} +{"lr": 0.0009054634122155991, "data_time": 0.0011273622512817383, "grad_norm": 0.02002229744102806, "loss": 0.6886222124099731, "time": 0.39647731781005857, "epoch": 5, "memory": 9465, "step": 54764} +{"lr": 0.0009054634122155991, "data_time": 0.0011443614959716797, "grad_norm": 0.018369539640843868, "loss": 0.6849116742610931, "time": 0.4018464326858521, "epoch": 5, "memory": 9465, "step": 54864} +{"lr": 0.0009054634122155991, "data_time": 0.001343512535095215, "grad_norm": 0.030916664004325866, "loss": 0.6881608724594116, "time": 0.4080906867980957, "epoch": 5, "memory": 9465, "step": 54964} +{"lr": 0.0009054634122155991, "data_time": 0.001047801971435547, "grad_norm": 0.02816597940400243, "loss": 0.6849474430084228, "time": 0.4051995038986206, "epoch": 5, "memory": 9465, "step": 55064} +{"lr": 0.0009054634122155991, "data_time": 0.0010867834091186524, "grad_norm": 0.022959691612049936, "loss": 0.6809873819351197, "time": 0.4346762657165527, "epoch": 5, "memory": 9465, "step": 55164} +{"lr": 0.0009054634122155991, "data_time": 0.1966933488845825, "grad_norm": 0.027581992093473673, "loss": 0.6840690612792969, "time": 0.3958333730697632, "epoch": 5, "memory": 9465, "step": 55264} +{"lr": 0.0009054634122155991, "data_time": 0.2462209939956665, "grad_norm": 0.02065160626079887, "loss": 0.6868190467357635, "time": 0.4308686971664429, "epoch": 5, "memory": 9465, "step": 55364} +{"lr": 0.0009054634122155991, "data_time": 0.1575378179550171, "grad_norm": 0.017900220002047718, "loss": 0.6868641614913941, "time": 0.3400084972381592, "epoch": 5, "memory": 9465, "step": 55464} +{"lr": 0.0009054634122155991, "data_time": 0.0013090133666992187, "grad_norm": 0.027901506843045355, "loss": 0.6848702013492585, "time": 0.3926393032073975, "epoch": 5, "memory": 9465, "step": 55564} +{"lr": 0.0009054634122155991, "data_time": 0.0011811971664428711, "grad_norm": 0.011158675141632556, "loss": 0.6929126560688019, "time": 0.37322607040405276, "epoch": 5, "memory": 9465, "step": 55664} +{"lr": 0.0009054634122155991, "data_time": 0.001319718360900879, "grad_norm": 0.032238740008324385, "loss": 0.6909414887428283, "time": 0.38470749855041503, "epoch": 5, "memory": 9465, "step": 55764} +{"lr": 0.0009054634122155991, "data_time": 0.0014949798583984374, "grad_norm": 0.017692160699516533, "loss": 0.6872556626796722, "time": 0.40132732391357423, "epoch": 5, "memory": 9465, "step": 55864} +{"lr": 0.0009054634122155991, "data_time": 0.0016646623611450196, "grad_norm": 0.024927551858127117, "loss": 0.6894688308238983, "time": 0.388171648979187, "epoch": 5, "memory": 9465, "step": 55964} +{"lr": 0.0009054634122155991, "data_time": 0.001196146011352539, "grad_norm": 0.028992805175948887, "loss": 0.6850293695926666, "time": 0.3908120632171631, "epoch": 5, "memory": 9465, "step": 56064} +{"lr": 0.0009054634122155991, "data_time": 0.0011138200759887695, "grad_norm": 0.02397423905786127, "loss": 0.687543261051178, "time": 0.3826998949050903, "epoch": 5, "memory": 9465, "step": 56164} +{"lr": 0.0009054634122155991, "data_time": 0.001337432861328125, "grad_norm": 0.026055848505347968, "loss": 0.6829025328159333, "time": 0.4017136335372925, "epoch": 5, "memory": 9465, "step": 56264} +{"lr": 0.0009054634122155991, "data_time": 0.0011916875839233399, "grad_norm": 0.02607398647814989, "loss": 0.6879303753376007, "time": 0.40138583183288573, "epoch": 5, "memory": 9465, "step": 56364} +{"lr": 0.0009054634122155991, "data_time": 0.0011496305465698241, "grad_norm": 0.024865913321264088, "loss": 0.6895306766033172, "time": 0.39088757038116456, "epoch": 5, "memory": 9465, "step": 56464} +{"lr": 0.0009054634122155991, "data_time": 0.0011301040649414062, "grad_norm": 0.02537779277190566, "loss": 0.687163770198822, "time": 0.39146401882171633, "epoch": 5, "memory": 9465, "step": 56564} +{"lr": 0.0009054634122155991, "data_time": 0.0012575626373291016, "grad_norm": 0.02295676473295316, "loss": 0.6881832003593444, "time": 0.3909043550491333, "epoch": 5, "memory": 9465, "step": 56664} +{"lr": 0.0009054634122155991, "data_time": 0.001027393341064453, "grad_norm": 0.024595651170238854, "loss": 0.6884438097476959, "time": 0.39977173805236815, "epoch": 5, "memory": 9465, "step": 56764} +{"lr": 0.0009054634122155991, "data_time": 0.0011107921600341797, "grad_norm": 0.026656482857652007, "loss": 0.6888610482215881, "time": 0.44601051807403563, "epoch": 5, "memory": 9465, "step": 56864} +{"lr": 0.0009054634122155991, "data_time": 0.0010900020599365235, "grad_norm": 0.0261293675750494, "loss": 0.6898661613464355, "time": 0.4157944917678833, "epoch": 5, "memory": 9465, "step": 56964} +{"lr": 0.0009054634122155991, "data_time": 0.0012702226638793945, "grad_norm": 0.024139224365353585, "loss": 0.6927463173866272, "time": 0.381671667098999, "epoch": 5, "memory": 9465, "step": 57064} +{"lr": 0.0009054634122155991, "data_time": 0.001151442527770996, "grad_norm": 0.028337081242352725, "loss": 0.6903347849845887, "time": 0.40841495990753174, "epoch": 5, "memory": 9465, "step": 57164} +{"lr": 0.0009054634122155991, "data_time": 0.001214742660522461, "grad_norm": 0.024786919355392456, "loss": 0.6854541897773743, "time": 0.3946031332015991, "epoch": 5, "memory": 9465, "step": 57264} +{"lr": 0.0009054634122155991, "data_time": 0.0011946439743041992, "grad_norm": 0.023413435183465482, "loss": 0.6880646169185638, "time": 0.37474076747894286, "epoch": 5, "memory": 9465, "step": 57364} +{"lr": 0.0009054634122155991, "data_time": 0.0018743515014648438, "grad_norm": 0.018026468594325707, "loss": 0.6855308651924134, "time": 0.43295631408691404, "epoch": 5, "memory": 9465, "step": 57464} +{"lr": 0.0009054634122155991, "data_time": 0.007212376594543457, "grad_norm": 0.016114775411551818, "loss": 0.6845888435840607, "time": 0.39299988746643066, "epoch": 5, "memory": 9465, "step": 57564} +{"accuracy/top1": 100.0, "data_time": 0.3096703641554889, "time": 0.5618760445538689, "step": 5} +{"lr": 0.0008550178566873411, "data_time": 0.10306141376495362, "grad_norm": 0.02607856991235167, "loss": 0.6887886345386505, "time": 0.3675510883331299, "epoch": 6, "memory": 9465, "step": 57680} +{"lr": 0.0008550178566873411, "data_time": 0.0011992931365966796, "grad_norm": 0.016421881050337105, "loss": 0.6853217840194702, "time": 0.3881377696990967, "epoch": 6, "memory": 9465, "step": 57780} +{"lr": 0.0008550178566873411, "data_time": 0.001097393035888672, "grad_norm": 0.02655152587685734, "loss": 0.6838648736476898, "time": 0.3818311929702759, "epoch": 6, "memory": 9465, "step": 57880} +{"lr": 0.0008550178566873411, "data_time": 0.0011381149291992188, "grad_norm": 0.024827568465843795, "loss": 0.6888383448123931, "time": 0.39320361614227295, "epoch": 6, "memory": 9465, "step": 57980} +{"lr": 0.0008550178566873411, "data_time": 0.001102280616760254, "grad_norm": 0.023271420691162347, "loss": 0.6845191359519959, "time": 0.40198745727539065, "epoch": 6, "memory": 9465, "step": 58080} +{"lr": 0.0008550178566873411, "data_time": 0.0012479782104492187, "grad_norm": 0.014999636285938322, "loss": 0.6901805341243744, "time": 0.39585835933685304, "epoch": 6, "memory": 9465, "step": 58180} +{"lr": 0.0008550178566873411, "data_time": 0.001110243797302246, "grad_norm": 0.019241804839111865, "loss": 0.6877995491027832, "time": 0.36391875743865965, "epoch": 6, "memory": 9465, "step": 58280} +{"lr": 0.0008550178566873411, "data_time": 0.0010864019393920898, "grad_norm": 0.009412611136212945, "loss": 0.6884370386600495, "time": 0.3892103433609009, "epoch": 6, "memory": 9465, "step": 58380} +{"lr": 0.0008550178566873411, "data_time": 0.0011346101760864257, "grad_norm": 0.03503970931051299, "loss": 0.6902829885482789, "time": 0.41189630031585694, "epoch": 6, "memory": 9465, "step": 58480} +{"lr": 0.0008550178566873411, "data_time": 0.0011670827865600587, "grad_norm": 0.03020229073590599, "loss": 0.6892358183860778, "time": 0.4098546028137207, "epoch": 6, "memory": 9465, "step": 58580} +{"lr": 0.0008550178566873411, "data_time": 0.001118946075439453, "grad_norm": 0.021369007194880395, "loss": 0.6875794708728791, "time": 0.37779951095581055, "epoch": 6, "memory": 9465, "step": 58680} +{"lr": 0.0008550178566873411, "data_time": 0.0010996103286743165, "grad_norm": 0.02368443019222468, "loss": 0.6920821368694305, "time": 0.3939052581787109, "epoch": 6, "memory": 9465, "step": 58780} +{"lr": 0.0008550178566873411, "data_time": 0.0011996746063232422, "grad_norm": 0.011431300081312656, "loss": 0.687268203496933, "time": 0.4462315559387207, "epoch": 6, "memory": 9465, "step": 58880} +{"lr": 0.0008550178566873411, "data_time": 0.0011156082153320312, "grad_norm": 0.03080502925440669, "loss": 0.6822718024253845, "time": 0.38296535015106203, "epoch": 6, "memory": 9465, "step": 58980} +{"lr": 0.0008550178566873411, "data_time": 0.0011321067810058593, "grad_norm": 0.020110561919864267, "loss": 0.6852346539497376, "time": 0.40460069179534913, "epoch": 6, "memory": 9465, "step": 59080} +{"lr": 0.0008550178566873411, "data_time": 0.002657508850097656, "grad_norm": 0.01642611682182178, "loss": 0.6863704025745392, "time": 0.3737973213195801, "epoch": 6, "memory": 9465, "step": 59180} +{"lr": 0.0008550178566873411, "data_time": 0.0011547327041625977, "grad_norm": 0.022235743375495075, "loss": 0.6865163683891297, "time": 0.4362701654434204, "epoch": 6, "memory": 9465, "step": 59280} +{"lr": 0.0008550178566873411, "data_time": 0.001006746292114258, "grad_norm": 0.026628897758200765, "loss": 0.6915045559406281, "time": 0.4014540672302246, "epoch": 6, "memory": 9465, "step": 59380} +{"lr": 0.0008550178566873411, "data_time": 0.0010770320892333984, "grad_norm": 0.02739627774572, "loss": 0.686798894405365, "time": 0.4184353113174438, "epoch": 6, "memory": 9465, "step": 59480} +{"lr": 0.0008550178566873411, "data_time": 0.0010686874389648437, "grad_norm": 0.021532252011820675, "loss": 0.6933930337429046, "time": 0.41991171836853025, "epoch": 6, "memory": 9465, "step": 59580} +{"lr": 0.0008550178566873411, "data_time": 0.001038193702697754, "grad_norm": 0.0282734930049628, "loss": 0.6866487681865692, "time": 0.38993492126464846, "epoch": 6, "memory": 9465, "step": 59680} +{"lr": 0.0008550178566873411, "data_time": 0.0011579275131225586, "grad_norm": 0.0163154071662575, "loss": 0.6848962366580963, "time": 0.4043025732040405, "epoch": 6, "memory": 9465, "step": 59780} +{"lr": 0.0008550178566873411, "data_time": 0.0010789871215820313, "grad_norm": 0.02124893106520176, "loss": 0.6870622158050537, "time": 0.4309748888015747, "epoch": 6, "memory": 9465, "step": 59880} +{"lr": 0.0008550178566873411, "data_time": 0.0010882854461669923, "grad_norm": 0.021265176229644567, "loss": 0.6840592145919799, "time": 0.38179304599761965, "epoch": 6, "memory": 9465, "step": 59980} +{"lr": 0.0008550178566873411, "data_time": 0.00113983154296875, "grad_norm": 0.020948754576966166, "loss": 0.6856973528861999, "time": 0.40779705047607423, "epoch": 6, "memory": 9465, "step": 60080} +{"lr": 0.0008550178566873411, "data_time": 0.001367497444152832, "grad_norm": 0.025594734493643044, "loss": 0.6940591335296631, "time": 0.4549035310745239, "epoch": 6, "memory": 9465, "step": 60180} +{"lr": 0.0008550178566873411, "data_time": 0.0010826587677001953, "grad_norm": 0.017471459379885347, "loss": 0.686150062084198, "time": 0.3902132511138916, "epoch": 6, "memory": 9465, "step": 60280} +{"lr": 0.0008550178566873411, "data_time": 0.0012021541595458984, "grad_norm": 0.015292374766431749, "loss": 0.683225828409195, "time": 0.3980573177337646, "epoch": 6, "memory": 9465, "step": 60380} +{"lr": 0.0008550178566873411, "data_time": 0.0011262178421020507, "grad_norm": 0.019223866751417518, "loss": 0.6911708950996399, "time": 0.41641333103179934, "epoch": 6, "memory": 9465, "step": 60480} +{"lr": 0.0008550178566873411, "data_time": 0.0010864019393920898, "grad_norm": 0.026042074616998433, "loss": 0.6903434634208679, "time": 0.43301615715026853, "epoch": 6, "memory": 9465, "step": 60580} +{"lr": 0.0008550178566873411, "data_time": 0.001121211051940918, "grad_norm": 0.020801701271557248, "loss": 0.6834124207496644, "time": 0.40266146659851076, "epoch": 6, "memory": 9465, "step": 60680} +{"lr": 0.0008550178566873411, "data_time": 0.0011501073837280273, "grad_norm": 0.014840489346534013, "loss": 0.6909108400344849, "time": 0.3964644432067871, "epoch": 6, "memory": 9465, "step": 60780} +{"lr": 0.0008550178566873411, "data_time": 0.001073169708251953, "grad_norm": 0.026812200294807555, "loss": 0.6863328039646148, "time": 0.3890755414962769, "epoch": 6, "memory": 9465, "step": 60880} +{"lr": 0.0008550178566873411, "data_time": 0.0010693073272705078, "grad_norm": 0.03237656396813691, "loss": 0.6915999114513397, "time": 0.4082698583602905, "epoch": 6, "memory": 9465, "step": 60980} +{"lr": 0.0008550178566873411, "data_time": 0.0010993480682373047, "grad_norm": 0.03013187083415687, "loss": 0.6842646121978759, "time": 0.42697293758392335, "epoch": 6, "memory": 9465, "step": 61080} +{"lr": 0.0008550178566873411, "data_time": 0.001182413101196289, "grad_norm": 0.028919320087879896, "loss": 0.6859990298748017, "time": 0.4435647249221802, "epoch": 6, "memory": 9465, "step": 61180} +{"lr": 0.0008550178566873411, "data_time": 0.0012660264968872071, "grad_norm": 0.028606146585661917, "loss": 0.6837523519992829, "time": 0.41578099727630613, "epoch": 6, "memory": 9465, "step": 61280} +{"lr": 0.0008550178566873411, "data_time": 0.0010733604431152344, "grad_norm": 0.023944671917706727, "loss": 0.6810744404792786, "time": 0.4051700830459595, "epoch": 6, "memory": 9465, "step": 61380} +{"lr": 0.0008550178566873411, "data_time": 0.0013903379440307617, "grad_norm": 0.025947124883532523, "loss": 0.6896799206733704, "time": 0.44748764038085936, "epoch": 6, "memory": 9465, "step": 61480} +{"lr": 0.0008550178566873411, "data_time": 0.0010825634002685548, "grad_norm": 0.011630263109691442, "loss": 0.6843417644500732, "time": 0.3873655557632446, "epoch": 6, "memory": 9465, "step": 61580} +{"lr": 0.0008550178566873411, "data_time": 0.0010849237442016602, "grad_norm": 0.020101899770088492, "loss": 0.6912844240665436, "time": 0.3698854684829712, "epoch": 6, "memory": 9465, "step": 61680} +{"lr": 0.0008550178566873411, "data_time": 0.0011237144470214843, "grad_norm": 0.016183562227524817, "loss": 0.6849010467529297, "time": 0.4075817108154297, "epoch": 6, "memory": 9465, "step": 61780} +{"lr": 0.0008550178566873411, "data_time": 0.001196885108947754, "grad_norm": 0.02694909244310111, "loss": 0.6850013375282288, "time": 0.4347084522247314, "epoch": 6, "memory": 9465, "step": 61880} +{"lr": 0.0008550178566873411, "data_time": 0.0010751724243164063, "grad_norm": 0.02409428098471835, "loss": 0.6830321192741394, "time": 0.42159261703491213, "epoch": 6, "memory": 9465, "step": 61980} +{"lr": 0.0008550178566873411, "data_time": 0.0011061429977416992, "grad_norm": 0.015985024417750536, "loss": 0.6874011397361756, "time": 0.4641463041305542, "epoch": 6, "memory": 9465, "step": 62080} +{"lr": 0.0008550178566873411, "data_time": 0.0013673067092895507, "grad_norm": 0.02110781449591741, "loss": 0.6865115404129029, "time": 0.38755435943603517, "epoch": 6, "memory": 9465, "step": 62180} +{"lr": 0.0008550178566873411, "data_time": 0.0010976552963256835, "grad_norm": 0.025282340170815588, "loss": 0.6891542911529541, "time": 0.386476469039917, "epoch": 6, "memory": 9465, "step": 62280} +{"lr": 0.0008550178566873411, "data_time": 0.0011896610260009766, "grad_norm": 0.018300859036389738, "loss": 0.6869811832904815, "time": 0.4204688787460327, "epoch": 6, "memory": 9465, "step": 62380} +{"lr": 0.0008550178566873411, "data_time": 0.0012111902236938477, "grad_norm": 0.021349584963172675, "loss": 0.6878549575805664, "time": 0.411251425743103, "epoch": 6, "memory": 9465, "step": 62480} +{"lr": 0.0008550178566873411, "data_time": 0.0013198375701904297, "grad_norm": 0.02706974893808365, "loss": 0.6856260359287262, "time": 0.42128643989562986, "epoch": 6, "memory": 9465, "step": 62580} +{"lr": 0.0008550178566873411, "data_time": 0.0014317035675048828, "grad_norm": 0.030903576174750924, "loss": 0.6902890741825104, "time": 0.41969847679138184, "epoch": 6, "memory": 9465, "step": 62680} +{"lr": 0.0008550178566873411, "data_time": 0.001162409782409668, "grad_norm": 0.017232837807387113, "loss": 0.6841163575649262, "time": 0.39000930786132815, "epoch": 6, "memory": 9465, "step": 62780} +{"lr": 0.0008550178566873411, "data_time": 0.0011013269424438477, "grad_norm": 0.03369742791401222, "loss": 0.6851688861846924, "time": 0.40869951248168945, "epoch": 6, "memory": 9465, "step": 62880} +{"lr": 0.0008550178566873411, "data_time": 0.0011281728744506835, "grad_norm": 0.02228668605675921, "loss": 0.6884321808815003, "time": 0.4005002498626709, "epoch": 6, "memory": 9465, "step": 62980} +{"lr": 0.0008550178566873411, "data_time": 0.001163172721862793, "grad_norm": 0.017968063976150007, "loss": 0.6882640779018402, "time": 0.4137157440185547, "epoch": 6, "memory": 9465, "step": 63080} +{"lr": 0.0008550178566873411, "data_time": 0.0011319398880004882, "grad_norm": 0.01606730920029804, "loss": 0.6900041937828064, "time": 0.39462196826934814, "epoch": 6, "memory": 9465, "step": 63180} +{"lr": 0.0008550178566873411, "data_time": 0.0011784791946411132, "grad_norm": 0.021908795461058617, "loss": 0.6898101925849914, "time": 0.41650776863098143, "epoch": 6, "memory": 9465, "step": 63280} +{"lr": 0.0008550178566873411, "data_time": 0.0012079477310180664, "grad_norm": 0.024879211466759442, "loss": 0.688402795791626, "time": 0.40994791984558104, "epoch": 6, "memory": 9465, "step": 63380} +{"lr": 0.0008550178566873411, "data_time": 0.001107168197631836, "grad_norm": 0.027438949700444935, "loss": 0.6885316550731659, "time": 0.3998621225357056, "epoch": 6, "memory": 9465, "step": 63480} +{"lr": 0.0008550178566873411, "data_time": 0.001163315773010254, "grad_norm": 0.026064815791323782, "loss": 0.6823230147361755, "time": 0.4091726064682007, "epoch": 6, "memory": 9465, "step": 63580} +{"lr": 0.0008550178566873411, "data_time": 0.0011488676071166992, "grad_norm": 0.02314729627687484, "loss": 0.683469420671463, "time": 0.4455825090408325, "epoch": 6, "memory": 9465, "step": 63680} +{"lr": 0.0008550178566873411, "data_time": 0.0012412309646606446, "grad_norm": 0.018279211735352872, "loss": 0.6866757035255432, "time": 0.3936056137084961, "epoch": 6, "memory": 9465, "step": 63780} +{"lr": 0.0008550178566873411, "data_time": 0.0010842084884643555, "grad_norm": 0.025038791983388365, "loss": 0.6824251770973205, "time": 0.48471248149871826, "epoch": 6, "memory": 9465, "step": 63880} +{"lr": 0.0008550178566873411, "data_time": 0.0010651588439941407, "grad_norm": 0.01808282525744289, "loss": 0.6876177608966827, "time": 0.39613330364227295, "epoch": 6, "memory": 9465, "step": 63980} +{"lr": 0.0008550178566873411, "data_time": 0.0010905504226684571, "grad_norm": 0.018851933989208192, "loss": 0.6884236335754395, "time": 0.4246366500854492, "epoch": 6, "memory": 9465, "step": 64080} +{"lr": 0.0008550178566873411, "data_time": 0.0011005163192749023, "grad_norm": 0.0259113168111071, "loss": 0.6852250516414642, "time": 0.3710044860839844, "epoch": 6, "memory": 9465, "step": 64180} +{"lr": 0.0008550178566873411, "data_time": 0.001170516014099121, "grad_norm": 0.029111536033451558, "loss": 0.6851958274841309, "time": 0.3933532476425171, "epoch": 6, "memory": 9465, "step": 64280} +{"lr": 0.0008550178566873411, "data_time": 0.0011178970336914063, "grad_norm": 0.015426943911006675, "loss": 0.6896191537380219, "time": 0.3818197250366211, "epoch": 6, "memory": 9465, "step": 64380} +{"lr": 0.0008550178566873411, "data_time": 0.0010658979415893554, "grad_norm": 0.023294601123780012, "loss": 0.6892889618873597, "time": 0.4220273494720459, "epoch": 6, "memory": 9465, "step": 64480} +{"lr": 0.0008550178566873411, "data_time": 0.0011204719543457032, "grad_norm": 0.011773345549590886, "loss": 0.6862027585506439, "time": 0.43556506633758546, "epoch": 6, "memory": 9465, "step": 64580} +{"lr": 0.0008550178566873411, "data_time": 0.001467728614807129, "grad_norm": 0.02419381303479895, "loss": 0.6829163193702698, "time": 0.40226283073425295, "epoch": 6, "memory": 9465, "step": 64680} +{"lr": 0.0008550178566873411, "data_time": 0.0012456655502319336, "grad_norm": 0.03351171747781336, "loss": 0.6883672893047332, "time": 0.43147013187408445, "epoch": 6, "memory": 9465, "step": 64780} +{"lr": 0.0008550178566873411, "data_time": 0.0011522769927978516, "grad_norm": 0.025805607502115892, "loss": 0.6861656308174133, "time": 0.4185208320617676, "epoch": 6, "memory": 9465, "step": 64880} +{"lr": 0.0008550178566873411, "data_time": 0.001204204559326172, "grad_norm": 0.02143499543890357, "loss": 0.687349659204483, "time": 0.39340035915374755, "epoch": 6, "memory": 9465, "step": 64980} +{"lr": 0.0008550178566873411, "data_time": 0.0011176347732543945, "grad_norm": 0.021567908296128736, "loss": 0.6841687262058258, "time": 0.40543975830078127, "epoch": 6, "memory": 9465, "step": 65080} +{"lr": 0.0008550178566873411, "data_time": 0.0011295318603515626, "grad_norm": 0.025428109848871828, "loss": 0.6899242162704468, "time": 0.387009334564209, "epoch": 6, "memory": 9465, "step": 65180} +{"lr": 0.0008550178566873411, "data_time": 0.0011513233184814453, "grad_norm": 0.020488849189132453, "loss": 0.6837246060371399, "time": 0.4116119623184204, "epoch": 6, "memory": 9465, "step": 65280} +{"lr": 0.0008550178566873411, "data_time": 0.0011474609375, "grad_norm": 0.02112674816744402, "loss": 0.6861176073551178, "time": 0.46766388416290283, "epoch": 6, "memory": 9465, "step": 65380} +{"lr": 0.0008550178566873411, "data_time": 0.0012309551239013672, "grad_norm": 0.01799351954832673, "loss": 0.6893830060958862, "time": 0.4063719749450684, "epoch": 6, "memory": 9465, "step": 65480} +{"lr": 0.0008550178566873411, "data_time": 0.001080632209777832, "grad_norm": 0.027082108333706854, "loss": 0.688882440328598, "time": 0.3806299209594727, "epoch": 6, "memory": 9465, "step": 65580} +{"lr": 0.0008550178566873411, "data_time": 0.0011981010437011718, "grad_norm": 0.0213951176032424, "loss": 0.6909701764583588, "time": 0.38790957927703856, "epoch": 6, "memory": 9465, "step": 65680} +{"lr": 0.0008550178566873411, "data_time": 0.0011966705322265625, "grad_norm": 0.016349894972518087, "loss": 0.6805295407772064, "time": 0.3795259952545166, "epoch": 6, "memory": 9465, "step": 65780} +{"lr": 0.0008550178566873411, "data_time": 0.0010399103164672851, "grad_norm": 0.0243533281609416, "loss": 0.6870722115039826, "time": 0.3947573184967041, "epoch": 6, "memory": 9465, "step": 65880} +{"lr": 0.0008550178566873411, "data_time": 0.0011670827865600587, "grad_norm": 0.017661346250679343, "loss": 0.687247508764267, "time": 0.4532417058944702, "epoch": 6, "memory": 9465, "step": 65980} +{"lr": 0.0008550178566873411, "data_time": 0.001065516471862793, "grad_norm": 0.019467962806811557, "loss": 0.6906374752521515, "time": 0.4153579235076904, "epoch": 6, "memory": 9465, "step": 66080} +{"lr": 0.0008550178566873411, "data_time": 0.0011503219604492188, "grad_norm": 0.03497491907328367, "loss": 0.6900132894515991, "time": 0.40140585899353026, "epoch": 6, "memory": 9465, "step": 66180} +{"lr": 0.0008550178566873411, "data_time": 0.0011035442352294923, "grad_norm": 0.02296763132326305, "loss": 0.686825406551361, "time": 0.40420947074890134, "epoch": 6, "memory": 9465, "step": 66280} +{"lr": 0.0008550178566873411, "data_time": 0.0011605024337768555, "grad_norm": 0.025388493668287992, "loss": 0.6918170928955079, "time": 0.4069295644760132, "epoch": 6, "memory": 9465, "step": 66380} +{"lr": 0.0008550178566873411, "data_time": 0.0012253761291503907, "grad_norm": 0.025179001805372535, "loss": 0.6868872821331025, "time": 0.4094263792037964, "epoch": 6, "memory": 9465, "step": 66480} +{"lr": 0.0008550178566873411, "data_time": 0.0013419866561889648, "grad_norm": 0.027657877001911402, "loss": 0.6838114738464356, "time": 0.4042673587799072, "epoch": 6, "memory": 9465, "step": 66580} +{"lr": 0.0008550178566873411, "data_time": 0.001176595687866211, "grad_norm": 0.029566391464322807, "loss": 0.6818347692489624, "time": 0.409966254234314, "epoch": 6, "memory": 9465, "step": 66680} +{"lr": 0.0008550178566873411, "data_time": 0.0012098073959350586, "grad_norm": 0.01688590687699616, "loss": 0.6903729736804962, "time": 0.4109964847564697, "epoch": 6, "memory": 9465, "step": 66780} +{"lr": 0.0008550178566873411, "data_time": 0.001275348663330078, "grad_norm": 0.023581396555528045, "loss": 0.6841236174106597, "time": 0.3953559160232544, "epoch": 6, "memory": 9465, "step": 66880} +{"lr": 0.0008550178566873411, "data_time": 0.0011750221252441405, "grad_norm": 0.019107314106076956, "loss": 0.6900860786437988, "time": 0.4060039043426514, "epoch": 6, "memory": 9465, "step": 66980} +{"lr": 0.0008550178566873411, "data_time": 0.0012302875518798828, "grad_norm": 0.02024886240833439, "loss": 0.686821186542511, "time": 0.42997496128082274, "epoch": 6, "memory": 9465, "step": 67080} +{"lr": 0.0008550178566873411, "data_time": 0.0011691808700561523, "grad_norm": 0.023429017094895244, "loss": 0.680237877368927, "time": 0.38594486713409426, "epoch": 6, "memory": 9465, "step": 67180} +{"lr": 0.0008550178566873411, "data_time": 0.001174330711364746, "grad_norm": 0.021199182444252073, "loss": 0.6820933282375335, "time": 0.40532851219177246, "epoch": 6, "memory": 9465, "step": 67280} +{"lr": 0.0008550178566873411, "data_time": 0.0011693239212036133, "grad_norm": 0.01970489202067256, "loss": 0.6860453009605407, "time": 0.39337117671966554, "epoch": 6, "memory": 9465, "step": 67380} +{"lr": 0.0008550178566873411, "data_time": 0.0011412858963012694, "grad_norm": 0.024003434274345637, "loss": 0.6859983444213867, "time": 0.4117098569869995, "epoch": 6, "memory": 9465, "step": 67480} +{"lr": 0.0008550178566873411, "data_time": 0.0011181592941284179, "grad_norm": 0.023815575637854636, "loss": 0.6898360013961792, "time": 0.39422993659973143, "epoch": 6, "memory": 9465, "step": 67580} +{"lr": 0.0008550178566873411, "data_time": 0.0012259244918823241, "grad_norm": 0.01485637214500457, "loss": 0.6895285248756409, "time": 0.43624048233032225, "epoch": 6, "memory": 9465, "step": 67680} +{"lr": 0.0008550178566873411, "data_time": 0.0012720584869384765, "grad_norm": 0.022814353555440904, "loss": 0.691504716873169, "time": 0.4203951835632324, "epoch": 6, "memory": 9465, "step": 67780} +{"lr": 0.0008550178566873411, "data_time": 0.0012421369552612304, "grad_norm": 0.016687876591458917, "loss": 0.6868099987506866, "time": 0.3962958335876465, "epoch": 6, "memory": 9465, "step": 67880} +{"lr": 0.0008550178566873411, "data_time": 0.0012056827545166016, "grad_norm": 0.015783934318460525, "loss": 0.6871471047401428, "time": 0.40698909759521484, "epoch": 6, "memory": 9465, "step": 67980} +{"lr": 0.0008550178566873411, "data_time": 0.001257467269897461, "grad_norm": 0.017015948914922772, "loss": 0.6893740594387054, "time": 0.4241846799850464, "epoch": 6, "memory": 9465, "step": 68080} +{"lr": 0.0008550178566873411, "data_time": 0.0012694597244262695, "grad_norm": 0.02218850781209767, "loss": 0.6860420525074005, "time": 0.4323299407958984, "epoch": 6, "memory": 9465, "step": 68180} +{"lr": 0.0008550178566873411, "data_time": 0.0012068748474121094, "grad_norm": 0.015337636042386294, "loss": 0.6899284422397614, "time": 0.38808631896972656, "epoch": 6, "memory": 9465, "step": 68280} +{"lr": 0.0008550178566873411, "data_time": 0.0012377023696899414, "grad_norm": 0.03523201327770949, "loss": 0.6839330554008484, "time": 0.41525607109069823, "epoch": 6, "memory": 9465, "step": 68380} +{"lr": 0.0008550178566873411, "data_time": 0.0011533737182617188, "grad_norm": 0.02610473702661693, "loss": 0.6918040812015533, "time": 0.37747156620025635, "epoch": 6, "memory": 9465, "step": 68480} +{"lr": 0.0008550178566873411, "data_time": 0.001188945770263672, "grad_norm": 0.018019030266441406, "loss": 0.6894958972930908, "time": 0.41466484069824217, "epoch": 6, "memory": 9465, "step": 68580} +{"lr": 0.0008550178566873411, "data_time": 0.0015630245208740235, "grad_norm": 0.01750457133166492, "loss": 0.6869166016578674, "time": 0.4141963720321655, "epoch": 6, "memory": 9465, "step": 68680} +{"lr": 0.0008550178566873411, "data_time": 0.0012135982513427734, "grad_norm": 0.028229457233101128, "loss": 0.6859838128089905, "time": 0.3897592782974243, "epoch": 6, "memory": 9465, "step": 68780} +{"lr": 0.0008550178566873411, "data_time": 0.0012891054153442382, "grad_norm": 0.033662870479747656, "loss": 0.6845110654830933, "time": 0.3601423740386963, "epoch": 6, "memory": 9465, "step": 68880} +{"lr": 0.0008550178566873411, "data_time": 0.0012295961380004883, "grad_norm": 0.019249517982825638, "loss": 0.6847482442855835, "time": 0.43565006256103517, "epoch": 6, "memory": 9465, "step": 68980} +{"lr": 0.0008550178566873411, "data_time": 0.007334566116333008, "grad_norm": 0.0190709559712559, "loss": 0.6824127018451691, "time": 0.3995782136917114, "epoch": 6, "memory": 9465, "step": 69080} +{"accuracy/top1": 100.0, "data_time": 0.3054828924291274, "time": 0.5640647832085105, "step": 6} +{"lr": 0.0007959536998847743, "data_time": 0.022605299949645996, "grad_norm": 0.01316798001062125, "loss": 0.6878198325634003, "time": 0.44731831550598145, "epoch": 7, "memory": 9465, "step": 69196} +{"lr": 0.0007959536998847743, "data_time": 0.0010965585708618164, "grad_norm": 0.028479007072746754, "loss": 0.6900545597076416, "time": 0.44763081073760985, "epoch": 7, "memory": 9465, "step": 69296} +{"lr": 0.0007959536998847743, "data_time": 0.00127718448638916, "grad_norm": 0.02315240743337199, "loss": 0.6806198239326477, "time": 0.4228838920593262, "epoch": 7, "memory": 9465, "step": 69396} +{"lr": 0.0007959536998847743, "data_time": 0.0015025854110717774, "grad_norm": 0.026145265670493245, "loss": 0.6891481876373291, "time": 0.4264572858810425, "epoch": 7, "memory": 9465, "step": 69496} +{"lr": 0.0007959536998847743, "data_time": 0.0010885953903198241, "grad_norm": 0.020853764610365033, "loss": 0.6852071940898895, "time": 0.35759451389312746, "epoch": 7, "memory": 9465, "step": 69596} +{"lr": 0.0007959536998847743, "data_time": 0.0011881113052368164, "grad_norm": 0.02196980845183134, "loss": 0.6836467266082764, "time": 0.43849189281463624, "epoch": 7, "memory": 9465, "step": 69696} +{"lr": 0.0007959536998847743, "data_time": 0.0010839462280273437, "grad_norm": 0.022946777241304516, "loss": 0.6840446650981903, "time": 0.4129193782806396, "epoch": 7, "memory": 9465, "step": 69796} +{"lr": 0.0007959536998847743, "data_time": 0.0012560606002807618, "grad_norm": 0.013690093078184872, "loss": 0.6849134087562561, "time": 0.4388380765914917, "epoch": 7, "memory": 9465, "step": 69896} +{"lr": 0.0007959536998847743, "data_time": 0.0011666059494018556, "grad_norm": 0.014410020341165364, "loss": 0.6856664478778839, "time": 0.3797380208969116, "epoch": 7, "memory": 9465, "step": 69996} +{"lr": 0.0007959536998847743, "data_time": 0.002634882926940918, "grad_norm": 0.023347340803593398, "loss": 0.6844936788082123, "time": 0.4347724914550781, "epoch": 7, "memory": 9465, "step": 70096} +{"lr": 0.0007959536998847743, "data_time": 0.0010866880416870116, "grad_norm": 0.02320546139962971, "loss": 0.6840683877468109, "time": 0.3985123872756958, "epoch": 7, "memory": 9465, "step": 70196} +{"lr": 0.0007959536998847743, "data_time": 0.0012055397033691405, "grad_norm": 0.01761086212936789, "loss": 0.6879605293273926, "time": 0.4709017276763916, "epoch": 7, "memory": 9465, "step": 70296} +{"lr": 0.0007959536998847743, "data_time": 0.0012353897094726563, "grad_norm": 0.02368117372971028, "loss": 0.6894306123256684, "time": 0.39980330467224123, "epoch": 7, "memory": 9465, "step": 70396} +{"lr": 0.0007959536998847743, "data_time": 0.0011112451553344726, "grad_norm": 0.026349100284278393, "loss": 0.6901382505893707, "time": 0.4116779327392578, "epoch": 7, "memory": 9465, "step": 70496} +{"lr": 0.0007959536998847743, "data_time": 0.001292872428894043, "grad_norm": 0.02749940538778901, "loss": 0.6840810477733612, "time": 0.3990472316741943, "epoch": 7, "memory": 9465, "step": 70596} +{"lr": 0.0007959536998847743, "data_time": 0.0010713100433349609, "grad_norm": 0.031892047519795595, "loss": 0.6849338054656983, "time": 0.40349245071411133, "epoch": 7, "memory": 9465, "step": 70696} +{"lr": 0.0007959536998847743, "data_time": 0.0027619123458862303, "grad_norm": 0.02551441374234855, "loss": 0.6865213751792908, "time": 0.4154572725296021, "epoch": 7, "memory": 9465, "step": 70796} +{"lr": 0.0007959536998847743, "data_time": 0.0011179447174072266, "grad_norm": 0.014401370845735072, "loss": 0.6865893602371216, "time": 0.38748998641967775, "epoch": 7, "memory": 9465, "step": 70896} +{"lr": 0.0007959536998847743, "data_time": 0.0013726711273193359, "grad_norm": 0.03455458893440664, "loss": 0.6865870893001557, "time": 0.4037840127944946, "epoch": 7, "memory": 9465, "step": 70996} +{"lr": 0.0007959536998847743, "data_time": 0.0011995077133178712, "grad_norm": 0.018426940822973848, "loss": 0.6870003044605255, "time": 0.3991276502609253, "epoch": 7, "memory": 9465, "step": 71096} +{"lr": 0.0007959536998847743, "data_time": 0.0011152982711791991, "grad_norm": 0.032567341532558206, "loss": 0.6811158239841462, "time": 0.39254446029663087, "epoch": 7, "memory": 9465, "step": 71196} +{"lr": 0.0007959536998847743, "data_time": 0.0010578393936157226, "grad_norm": 0.024077949032653125, "loss": 0.686410653591156, "time": 0.3703174352645874, "epoch": 7, "memory": 9465, "step": 71296} +{"lr": 0.0007959536998847743, "data_time": 0.0011644363403320312, "grad_norm": 0.028478661738336086, "loss": 0.6945944964885712, "time": 0.39158904552459717, "epoch": 7, "memory": 9465, "step": 71396} +{"lr": 0.0007959536998847743, "data_time": 0.0011600732803344726, "grad_norm": 0.018969375151209535, "loss": 0.689129626750946, "time": 0.4178628921508789, "epoch": 7, "memory": 9465, "step": 71496} +{"lr": 0.0007959536998847743, "data_time": 0.0010562658309936524, "grad_norm": 0.030955977737903595, "loss": 0.683920043706894, "time": 0.411151123046875, "epoch": 7, "memory": 9465, "step": 71596} +{"lr": 0.0007959536998847743, "data_time": 0.0012195110321044922, "grad_norm": 0.021646203519776464, "loss": 0.692710280418396, "time": 0.4151927947998047, "epoch": 7, "memory": 9465, "step": 71696} +{"lr": 0.0007959536998847743, "data_time": 0.0012052059173583984, "grad_norm": 0.028709866828285156, "loss": 0.6846365094184875, "time": 0.36250553131103513, "epoch": 7, "memory": 9465, "step": 71796} +{"lr": 0.0007959536998847743, "data_time": 0.0012458086013793946, "grad_norm": 0.02136602159589529, "loss": 0.6866275489330291, "time": 0.3874361038208008, "epoch": 7, "memory": 9465, "step": 71896} +{"lr": 0.0007959536998847743, "data_time": 0.0013145685195922851, "grad_norm": 0.02129147609230131, "loss": 0.6894185721874238, "time": 0.407142186164856, "epoch": 7, "memory": 9465, "step": 71996} +{"lr": 0.0007959536998847743, "data_time": 0.0011846303939819336, "grad_norm": 0.028740060911513866, "loss": 0.6892673373222351, "time": 0.4421504020690918, "epoch": 7, "memory": 9465, "step": 72096} +{"lr": 0.0007959536998847743, "data_time": 0.0011426210403442383, "grad_norm": 0.02137885382398963, "loss": 0.6883216917514801, "time": 0.37873086929321287, "epoch": 7, "memory": 9465, "step": 72196} +{"lr": 0.0007959536998847743, "data_time": 0.001166391372680664, "grad_norm": 0.0248980559874326, "loss": 0.6873321831226349, "time": 0.4093770027160645, "epoch": 7, "memory": 9465, "step": 72296} +{"lr": 0.0007959536998847743, "data_time": 0.0011440277099609374, "grad_norm": 0.016781652311328797, "loss": 0.689619517326355, "time": 0.4165563821792603, "epoch": 7, "memory": 9465, "step": 72396} +{"lr": 0.0007959536998847743, "data_time": 0.0011880636215209962, "grad_norm": 0.01665127892047167, "loss": 0.6863309919834137, "time": 0.3981989860534668, "epoch": 7, "memory": 9465, "step": 72496} +{"lr": 0.0007959536998847743, "data_time": 0.001066732406616211, "grad_norm": 0.025460918783210218, "loss": 0.6827969551086426, "time": 0.3673178911209106, "epoch": 7, "memory": 9465, "step": 72596} +{"lr": 0.0007959536998847743, "data_time": 0.0010986804962158203, "grad_norm": 0.02312437416985631, "loss": 0.6871290862560272, "time": 0.39009847640991213, "epoch": 7, "memory": 9465, "step": 72696} +{"lr": 0.0007959536998847743, "data_time": 0.0011986494064331055, "grad_norm": 0.019325556652620434, "loss": 0.684772652387619, "time": 0.38254680633544924, "epoch": 7, "memory": 9465, "step": 72796} +{"lr": 0.0007959536998847743, "data_time": 0.0011121511459350586, "grad_norm": 0.02225032856222242, "loss": 0.6838533043861389, "time": 0.4375830888748169, "epoch": 7, "memory": 9465, "step": 72896} +{"lr": 0.0007959536998847743, "data_time": 0.0010936498641967774, "grad_norm": 0.024222818808630107, "loss": 0.6877457797527313, "time": 0.4053443670272827, "epoch": 7, "memory": 9465, "step": 72996} +{"lr": 0.0007959536998847743, "data_time": 0.0012072324752807617, "grad_norm": 0.02482799154240638, "loss": 0.6861601173877716, "time": 0.3897785902023315, "epoch": 7, "memory": 9465, "step": 73096} +{"lr": 0.0007959536998847743, "data_time": 0.00133209228515625, "grad_norm": 0.028896976122632624, "loss": 0.6917809903621673, "time": 0.3895012617111206, "epoch": 7, "memory": 9465, "step": 73196} +{"lr": 0.0007959536998847743, "data_time": 0.0013009071350097655, "grad_norm": 0.017806610558182, "loss": 0.6895358264446259, "time": 0.39615821838378906, "epoch": 7, "memory": 9465, "step": 73296} +{"lr": 0.0007959536998847743, "data_time": 0.001237964630126953, "grad_norm": 0.01780034021940082, "loss": 0.6869791269302368, "time": 0.3995818138122559, "epoch": 7, "memory": 9465, "step": 73396} +{"lr": 0.0007959536998847743, "data_time": 0.0011965274810791016, "grad_norm": 0.012079532991629094, "loss": 0.6852951526641846, "time": 0.4225271463394165, "epoch": 7, "memory": 9465, "step": 73496} +{"lr": 0.0007959536998847743, "data_time": 0.0011064529418945313, "grad_norm": 0.026877340488135816, "loss": 0.6846539080142975, "time": 0.3968620300292969, "epoch": 7, "memory": 9465, "step": 73596} +{"lr": 0.0007959536998847743, "data_time": 0.0011870861053466797, "grad_norm": 0.03740522600710392, "loss": 0.689351600408554, "time": 0.39341351985931394, "epoch": 7, "memory": 9465, "step": 73696} +{"lr": 0.0007959536998847743, "data_time": 0.0011953353881835938, "grad_norm": 0.013193433929700404, "loss": 0.6871483862400055, "time": 0.4323933839797974, "epoch": 7, "memory": 9465, "step": 73796} +{"lr": 0.0007959536998847743, "data_time": 0.0011172771453857421, "grad_norm": 0.009352261293679475, "loss": 0.6889617741107941, "time": 0.42199344635009767, "epoch": 7, "memory": 9465, "step": 73896} +{"lr": 0.0007959536998847743, "data_time": 0.001228928565979004, "grad_norm": 0.020918296393938364, "loss": 0.6816730439662934, "time": 0.3974435567855835, "epoch": 7, "memory": 9465, "step": 73996} +{"lr": 0.0007959536998847743, "data_time": 0.0012335062026977539, "grad_norm": 0.017875133221969007, "loss": 0.6810477554798127, "time": 0.39940171241760253, "epoch": 7, "memory": 9465, "step": 74096} +{"lr": 0.0007959536998847743, "data_time": 0.0011986255645751952, "grad_norm": 0.014076375891454518, "loss": 0.6842746794223785, "time": 0.4127380132675171, "epoch": 7, "memory": 9465, "step": 74196} +{"lr": 0.0007959536998847743, "data_time": 0.0012447357177734375, "grad_norm": 0.02469316883943975, "loss": 0.6853709757328034, "time": 0.40438406467437743, "epoch": 7, "memory": 9465, "step": 74296} +{"lr": 0.0007959536998847743, "data_time": 0.0011493206024169923, "grad_norm": 0.021600378945004196, "loss": 0.6904186487197876, "time": 0.3997747182846069, "epoch": 7, "memory": 9465, "step": 74396} +{"lr": 0.0007959536998847743, "data_time": 0.001335000991821289, "grad_norm": 0.022198598738759755, "loss": 0.6854990541934967, "time": 0.3842605113983154, "epoch": 7, "memory": 9465, "step": 74496} +{"lr": 0.0007959536998847743, "data_time": 0.0011192798614501954, "grad_norm": 0.03163915043696761, "loss": 0.6925682902336121, "time": 0.45507111549377444, "epoch": 7, "memory": 9465, "step": 74596} +{"lr": 0.0007959536998847743, "data_time": 0.001257038116455078, "grad_norm": 0.020809374330565334, "loss": 0.6893840789794922, "time": 0.4486205816268921, "epoch": 7, "memory": 9465, "step": 74696} +{"lr": 0.0007959536998847743, "data_time": 0.0011239767074584961, "grad_norm": 0.025458967802114783, "loss": 0.686081075668335, "time": 0.4434143781661987, "epoch": 7, "memory": 9465, "step": 74796} +{"lr": 0.0007959536998847743, "data_time": 0.0011315107345581054, "grad_norm": 0.027446240209974347, "loss": 0.691266405582428, "time": 0.40890026092529297, "epoch": 7, "memory": 9465, "step": 74896} +{"lr": 0.0007959536998847743, "data_time": 0.0011792421340942384, "grad_norm": 0.01645224934909493, "loss": 0.6849417626857758, "time": 0.4130192518234253, "epoch": 7, "memory": 9465, "step": 74996} +{"lr": 0.0007959536998847743, "data_time": 0.001281905174255371, "grad_norm": 0.016274695470929145, "loss": 0.6884645402431488, "time": 0.3943858861923218, "epoch": 7, "memory": 9465, "step": 75096} +{"lr": 0.0007959536998847743, "data_time": 0.001122283935546875, "grad_norm": 0.0183821105863899, "loss": 0.6827846050262452, "time": 0.38330254554748533, "epoch": 7, "memory": 9465, "step": 75196} +{"lr": 0.0007959536998847743, "data_time": 0.0011628389358520508, "grad_norm": 0.02766981702297926, "loss": 0.6827305257320404, "time": 0.41019539833068847, "epoch": 7, "memory": 9465, "step": 75296} +{"lr": 0.0007959536998847743, "data_time": 0.0011681079864501952, "grad_norm": 0.028430684260092674, "loss": 0.6861222445964813, "time": 0.37620279788970945, "epoch": 7, "memory": 9465, "step": 75396} +{"lr": 0.0007959536998847743, "data_time": 0.0012502431869506835, "grad_norm": 0.026339181466028094, "loss": 0.6839255213737487, "time": 0.39713211059570314, "epoch": 7, "memory": 9465, "step": 75496} +{"lr": 0.0007959536998847743, "data_time": 0.0013794422149658204, "grad_norm": 0.029753775848075748, "loss": 0.6906422078609467, "time": 0.3829894304275513, "epoch": 7, "memory": 9465, "step": 75596} +{"lr": 0.0007959536998847743, "data_time": 0.0012267827987670898, "grad_norm": 0.017975526629015803, "loss": 0.6806785941123963, "time": 0.3759139537811279, "epoch": 7, "memory": 9465, "step": 75696} +{"lr": 0.0007959536998847743, "data_time": 0.0010826826095581056, "grad_norm": 0.024884248990565537, "loss": 0.6849643230438233, "time": 0.4704716682434082, "epoch": 7, "memory": 9465, "step": 75796} +{"lr": 0.0007959536998847743, "data_time": 0.0012604951858520507, "grad_norm": 0.03132741232402623, "loss": 0.683762663602829, "time": 0.43655588626861574, "epoch": 7, "memory": 9465, "step": 75896} +{"lr": 0.0007959536998847743, "data_time": 0.001195073127746582, "grad_norm": 0.024875835841521622, "loss": 0.6888093113899231, "time": 0.37883970737457273, "epoch": 7, "memory": 9465, "step": 75996} +{"lr": 0.0007959536998847743, "data_time": 0.0012192487716674804, "grad_norm": 0.018560318695381283, "loss": 0.6864881336688995, "time": 0.38946144580841063, "epoch": 7, "memory": 9465, "step": 76096} +{"lr": 0.0007959536998847743, "data_time": 0.001201462745666504, "grad_norm": 0.013664885348407551, "loss": 0.6840902686119079, "time": 0.37909176349639895, "epoch": 7, "memory": 9465, "step": 76196} +{"lr": 0.0007959536998847743, "data_time": 0.0011518239974975587, "grad_norm": 0.030464647081680597, "loss": 0.6863432645797729, "time": 0.37787656784057616, "epoch": 7, "memory": 9465, "step": 76296} +{"lr": 0.0007959536998847743, "data_time": 0.0012253522872924805, "grad_norm": 0.023973657190799712, "loss": 0.6926301717758179, "time": 0.4389045238494873, "epoch": 7, "memory": 9465, "step": 76396} +{"lr": 0.0007959536998847743, "data_time": 0.001152777671813965, "grad_norm": 0.02176537667401135, "loss": 0.6871478378772735, "time": 0.38025965690612795, "epoch": 7, "memory": 9465, "step": 76496} +{"lr": 0.0007959536998847743, "data_time": 0.0011547327041625977, "grad_norm": 0.029439127212390302, "loss": 0.6821521103382111, "time": 0.3712587833404541, "epoch": 7, "memory": 9465, "step": 76596} +{"lr": 0.0007959536998847743, "data_time": 0.0011970996856689453, "grad_norm": 0.018512492533773184, "loss": 0.6826198101043701, "time": 0.41774275302886965, "epoch": 7, "memory": 9465, "step": 76696} +{"lr": 0.0007959536998847743, "data_time": 0.0011801958084106446, "grad_norm": 0.02286639059893787, "loss": 0.6876256883144378, "time": 0.4106889009475708, "epoch": 7, "memory": 9465, "step": 76796} +{"lr": 0.0007959536998847743, "data_time": 0.0011584043502807617, "grad_norm": 0.023090194049291312, "loss": 0.6913845479488373, "time": 0.40238194465637206, "epoch": 7, "memory": 9465, "step": 76896} +{"lr": 0.0007959536998847743, "data_time": 0.0014173030853271485, "grad_norm": 0.03176913349889219, "loss": 0.6898376643657684, "time": 0.41205744743347167, "epoch": 7, "memory": 9465, "step": 76996} +{"lr": 0.0007959536998847743, "data_time": 0.0011304140090942383, "grad_norm": 0.01942989700473845, "loss": 0.6842361092567444, "time": 0.4420263051986694, "epoch": 7, "memory": 9465, "step": 77096} +{"lr": 0.0007959536998847743, "data_time": 0.0011557340621948242, "grad_norm": 0.02109438246116042, "loss": 0.6830998361110687, "time": 0.4409440279006958, "epoch": 7, "memory": 9465, "step": 77196} +{"lr": 0.0007959536998847743, "data_time": 0.0016119241714477538, "grad_norm": 0.02279579215683043, "loss": 0.6854933440685272, "time": 0.3879295825958252, "epoch": 7, "memory": 9465, "step": 77296} +{"lr": 0.0007959536998847743, "data_time": 0.0011524915695190429, "grad_norm": 0.01711699743755162, "loss": 0.6846040070056916, "time": 0.42294399738311766, "epoch": 7, "memory": 9465, "step": 77396} +{"lr": 0.0007959536998847743, "data_time": 0.0011591196060180664, "grad_norm": 0.03519861763343215, "loss": 0.6838362872600555, "time": 0.394149112701416, "epoch": 7, "memory": 9465, "step": 77496} +{"lr": 0.0007959536998847743, "data_time": 0.0013530254364013672, "grad_norm": 0.023785974842030554, "loss": 0.6800795793533325, "time": 0.4433858633041382, "epoch": 7, "memory": 9465, "step": 77596} +{"lr": 0.0007959536998847743, "data_time": 0.0012445449829101562, "grad_norm": 0.026843556854873895, "loss": 0.6902290403842926, "time": 0.40749545097351075, "epoch": 7, "memory": 9465, "step": 77696} +{"lr": 0.0007959536998847743, "data_time": 0.0012624502182006837, "grad_norm": 0.02514321736525744, "loss": 0.6854418158531189, "time": 0.4115041971206665, "epoch": 7, "memory": 9465, "step": 77796} +{"lr": 0.0007959536998847743, "data_time": 0.0012320995330810547, "grad_norm": 0.03037467645481229, "loss": 0.6883927881717682, "time": 0.4010903358459473, "epoch": 7, "memory": 9465, "step": 77896} +{"lr": 0.0007959536998847743, "data_time": 0.0012339115142822265, "grad_norm": 0.021365776285529138, "loss": 0.6868184626102447, "time": 0.4028171062469482, "epoch": 7, "memory": 9465, "step": 77996} +{"lr": 0.0007959536998847743, "data_time": 0.0013242483139038086, "grad_norm": 0.025139387452509255, "loss": 0.6887413263320923, "time": 0.423887825012207, "epoch": 7, "memory": 9465, "step": 78096} +{"lr": 0.0007959536998847743, "data_time": 0.0012716293334960938, "grad_norm": 0.020559466490522028, "loss": 0.6861476004123688, "time": 0.40378665924072266, "epoch": 7, "memory": 9465, "step": 78196} +{"lr": 0.0007959536998847743, "data_time": 0.0012105226516723633, "grad_norm": 0.021940127667039632, "loss": 0.6868208050727844, "time": 0.40303223133087157, "epoch": 7, "memory": 9465, "step": 78296} +{"lr": 0.0007959536998847743, "data_time": 0.0012447834014892578, "grad_norm": 0.027616849076002835, "loss": 0.6870582818984985, "time": 0.39228756427764894, "epoch": 7, "memory": 9465, "step": 78396} +{"lr": 0.0007959536998847743, "data_time": 0.0013557195663452149, "grad_norm": 0.02429812690243125, "loss": 0.6896453738212586, "time": 0.3997992753982544, "epoch": 7, "memory": 9465, "step": 78496} +{"lr": 0.0007959536998847743, "data_time": 0.0011014223098754882, "grad_norm": 0.02309339742641896, "loss": 0.6866641640663147, "time": 0.4035929203033447, "epoch": 7, "memory": 9465, "step": 78596} +{"lr": 0.0007959536998847743, "data_time": 0.0012946605682373046, "grad_norm": 0.03126331111416221, "loss": 0.6872867465019226, "time": 0.4252474308013916, "epoch": 7, "memory": 9465, "step": 78696} +{"lr": 0.0007959536998847743, "data_time": 0.0011641740798950194, "grad_norm": 0.02848511151969433, "loss": 0.6913005232810974, "time": 0.38922045230865476, "epoch": 7, "memory": 9465, "step": 78796} +{"lr": 0.0007959536998847743, "data_time": 0.0013628482818603515, "grad_norm": 0.020038897823542356, "loss": 0.6844573140144348, "time": 0.40750024318695066, "epoch": 7, "memory": 9465, "step": 78896} +{"lr": 0.0007959536998847743, "data_time": 0.0012709617614746094, "grad_norm": 0.02573901703581214, "loss": 0.6842101216316223, "time": 0.3919532775878906, "epoch": 7, "memory": 9465, "step": 78996} +{"lr": 0.0007959536998847743, "data_time": 0.0013043403625488282, "grad_norm": 0.02561226738616824, "loss": 0.6893555283546448, "time": 0.3859339714050293, "epoch": 7, "memory": 9465, "step": 79096} +{"lr": 0.0007959536998847743, "data_time": 0.0011783361434936524, "grad_norm": 0.02280616246862337, "loss": 0.6859048366546631, "time": 0.4402990102767944, "epoch": 7, "memory": 9465, "step": 79196} +{"lr": 0.0007959536998847743, "data_time": 0.0012436389923095703, "grad_norm": 0.02633576896041632, "loss": 0.6829159498214722, "time": 0.41145374774932864, "epoch": 7, "memory": 9465, "step": 79296} +{"lr": 0.0007959536998847743, "data_time": 0.0014607906341552734, "grad_norm": 0.018908580765128136, "loss": 0.6861242353916168, "time": 0.40912585258483886, "epoch": 7, "memory": 9465, "step": 79396} +{"lr": 0.0007959536998847743, "data_time": 0.001175689697265625, "grad_norm": 0.032280557323247194, "loss": 0.6906120359897614, "time": 0.40739758014678956, "epoch": 7, "memory": 9465, "step": 79496} +{"lr": 0.0007959536998847743, "data_time": 0.0014308691024780273, "grad_norm": 0.02054695161059499, "loss": 0.6850894331932068, "time": 0.41107728481292727, "epoch": 7, "memory": 9465, "step": 79596} +{"lr": 0.0007959536998847743, "data_time": 0.0011907339096069336, "grad_norm": 0.025354425748810173, "loss": 0.6858440995216369, "time": 0.425725531578064, "epoch": 7, "memory": 9465, "step": 79696} +{"lr": 0.0007959536998847743, "data_time": 0.0011603355407714844, "grad_norm": 0.025696104858070612, "loss": 0.688878870010376, "time": 0.3961838722229004, "epoch": 7, "memory": 9465, "step": 79796} +{"lr": 0.0007959536998847743, "data_time": 0.0012267827987670898, "grad_norm": 0.027191118663176896, "loss": 0.6869683980941772, "time": 0.4198110103607178, "epoch": 7, "memory": 9465, "step": 79896} +{"lr": 0.0007959536998847743, "data_time": 0.0011564254760742187, "grad_norm": 0.0296568168210797, "loss": 0.6843812048435212, "time": 0.39384415149688723, "epoch": 7, "memory": 9465, "step": 79996} +{"lr": 0.0007959536998847743, "data_time": 0.0013707876205444336, "grad_norm": 0.020537405088543893, "loss": 0.6874856650829315, "time": 0.3853396654129028, "epoch": 7, "memory": 9465, "step": 80096} +{"lr": 0.0007959536998847743, "data_time": 0.0012266397476196288, "grad_norm": 0.018940770369954407, "loss": 0.6914855420589447, "time": 0.4090343713760376, "epoch": 7, "memory": 9465, "step": 80196} +{"lr": 0.0007959536998847743, "data_time": 0.0013055086135864257, "grad_norm": 0.028440984967164695, "loss": 0.6907103359699249, "time": 0.405469274520874, "epoch": 7, "memory": 9465, "step": 80296} +{"lr": 0.0007959536998847743, "data_time": 0.0012084484100341798, "grad_norm": 0.02373149050399661, "loss": 0.6889601469039917, "time": 0.3896803617477417, "epoch": 7, "memory": 9465, "step": 80396} +{"lr": 0.0007959536998847743, "data_time": 0.0017619371414184571, "grad_norm": 0.01712862554122694, "loss": 0.6891317307949066, "time": 0.46973021030426027, "epoch": 7, "memory": 9465, "step": 80496} +{"lr": 0.0007959536998847743, "data_time": 0.007248044013977051, "grad_norm": 0.01794057325460017, "loss": 0.6850728094577789, "time": 0.38886318206787107, "epoch": 7, "memory": 9465, "step": 80596} +{"accuracy/top1": 100.0, "data_time": 0.30626318034003763, "time": 0.5597720987656537, "step": 7} +{"lr": 0.0007297252973710758, "data_time": 0.08447706699371338, "grad_norm": 0.023277034144848587, "loss": 0.6917500734329224, "time": 0.38696401119232177, "epoch": 8, "memory": 9465, "step": 80712} +{"lr": 0.0007297252973710758, "data_time": 0.0013402700424194336, "grad_norm": 0.028954599611461164, "loss": 0.6854966700077056, "time": 0.3594782829284668, "epoch": 8, "memory": 9465, "step": 80812} +{"lr": 0.0007297252973710758, "data_time": 0.001197481155395508, "grad_norm": 0.012177791181602515, "loss": 0.6884900450706481, "time": 0.3827489376068115, "epoch": 8, "memory": 9465, "step": 80912} +{"lr": 0.0007297252973710758, "data_time": 0.0012011289596557616, "grad_norm": 0.02472599729662761, "loss": 0.6870001137256623, "time": 0.3923820734024048, "epoch": 8, "memory": 9465, "step": 81012} +{"lr": 0.0007297252973710758, "data_time": 0.001185464859008789, "grad_norm": 0.02104019853286445, "loss": 0.6873000979423523, "time": 0.39201998710632324, "epoch": 8, "memory": 9465, "step": 81112} +{"lr": 0.0007297252973710758, "data_time": 0.0014771223068237305, "grad_norm": 0.024233171017840505, "loss": 0.6884700119495392, "time": 0.42905073165893554, "epoch": 8, "memory": 9465, "step": 81212} +{"lr": 0.0007297252973710758, "data_time": 0.0010958433151245118, "grad_norm": 0.020955532940570266, "loss": 0.6871427893638611, "time": 0.3991347312927246, "epoch": 8, "memory": 9465, "step": 81312} +{"lr": 0.0007297252973710758, "data_time": 0.0012176990509033202, "grad_norm": 0.021414365619421005, "loss": 0.690663194656372, "time": 0.44790043830871584, "epoch": 8, "memory": 9465, "step": 81412} +{"lr": 0.0007297252973710758, "data_time": 0.0012326240539550781, "grad_norm": 0.02547398323658854, "loss": 0.6890263557434082, "time": 0.4333322763442993, "epoch": 8, "memory": 9465, "step": 81512} +{"lr": 0.0007297252973710758, "data_time": 0.001192164421081543, "grad_norm": 0.028446821169927715, "loss": 0.6854319989681243, "time": 0.45966885089874265, "epoch": 8, "memory": 9465, "step": 81612} +{"lr": 0.0007297252973710758, "data_time": 0.0011616230010986328, "grad_norm": 0.02645565108396113, "loss": 0.6852212011814117, "time": 0.37725861072540284, "epoch": 8, "memory": 9465, "step": 81712} +{"lr": 0.0007297252973710758, "data_time": 0.0011505603790283204, "grad_norm": 0.015086224023252725, "loss": 0.6828705668449402, "time": 0.40449302196502684, "epoch": 8, "memory": 9465, "step": 81812} +{"lr": 0.0007297252973710758, "data_time": 0.0012871265411376954, "grad_norm": 0.02012719577178359, "loss": 0.6825924575328827, "time": 0.4122276782989502, "epoch": 8, "memory": 9465, "step": 81912} +{"lr": 0.0007297252973710758, "data_time": 0.0011530876159667968, "grad_norm": 0.037408497929573056, "loss": 0.6819666743278503, "time": 0.3867088794708252, "epoch": 8, "memory": 9465, "step": 82012} +{"lr": 0.0007297252973710758, "data_time": 0.0012230396270751954, "grad_norm": 0.026454673893749715, "loss": 0.686492258310318, "time": 0.41706888675689696, "epoch": 8, "memory": 9465, "step": 82112} +{"lr": 0.0007297252973710758, "data_time": 0.0011826038360595703, "grad_norm": 0.020667389349546282, "loss": 0.6856553554534912, "time": 0.3936256170272827, "epoch": 8, "memory": 9465, "step": 82212} +{"lr": 0.0007297252973710758, "data_time": 0.0012821912765502929, "grad_norm": 0.025168696604669093, "loss": 0.6886514246463775, "time": 0.3955011129379272, "epoch": 8, "memory": 9465, "step": 82312} +{"lr": 0.0007297252973710758, "data_time": 0.0012822389602661134, "grad_norm": 0.013831566064618528, "loss": 0.688584178686142, "time": 0.41800537109375, "epoch": 8, "memory": 9465, "step": 82412} +{"lr": 0.0007297252973710758, "data_time": 0.0014978170394897461, "grad_norm": 0.03208147487603128, "loss": 0.6853248953819275, "time": 0.41456100940704343, "epoch": 8, "memory": 9465, "step": 82512} +{"lr": 0.0007297252973710758, "data_time": 0.0011444091796875, "grad_norm": 0.021822835318744183, "loss": 0.6866038501262665, "time": 0.4193004369735718, "epoch": 8, "memory": 9465, "step": 82612} +{"lr": 0.0007297252973710758, "data_time": 0.0011280298233032227, "grad_norm": 0.020326456637121737, "loss": 0.6886625587940216, "time": 0.40841264724731446, "epoch": 8, "memory": 9465, "step": 82712} +{"lr": 0.0007297252973710758, "data_time": 0.001205778121948242, "grad_norm": 0.02877616004552692, "loss": 0.6822507977485657, "time": 0.39947516918182374, "epoch": 8, "memory": 9465, "step": 82812} +{"lr": 0.0007297252973710758, "data_time": 0.001206183433532715, "grad_norm": 0.01602583437343128, "loss": 0.6874861299991608, "time": 0.3794590711593628, "epoch": 8, "memory": 9465, "step": 82912} +{"lr": 0.0007297252973710758, "data_time": 0.0011326789855957032, "grad_norm": 0.022695856541395186, "loss": 0.6854865074157714, "time": 0.3932990789413452, "epoch": 8, "memory": 9465, "step": 83012} +{"lr": 0.0007297252973710758, "data_time": 0.001281428337097168, "grad_norm": 0.03093284945935011, "loss": 0.6858721256256104, "time": 0.35994648933410645, "epoch": 8, "memory": 9465, "step": 83112} +{"lr": 0.0007297252973710758, "data_time": 0.0011367321014404297, "grad_norm": 0.02100286316126585, "loss": 0.6811021983623504, "time": 0.3889488935470581, "epoch": 8, "memory": 9465, "step": 83212} +{"lr": 0.0007297252973710758, "data_time": 0.0010964632034301757, "grad_norm": 0.017064707742247266, "loss": 0.6817765295505523, "time": 0.4008512496948242, "epoch": 8, "memory": 9465, "step": 83312} +{"lr": 0.0007297252973710758, "data_time": 0.0013264656066894532, "grad_norm": 0.02359428247436881, "loss": 0.6816625952720642, "time": 0.4141338348388672, "epoch": 8, "memory": 9465, "step": 83412} +{"lr": 0.0007297252973710758, "data_time": 0.001012706756591797, "grad_norm": 0.020844429964199662, "loss": 0.687481015920639, "time": 0.41402957439422605, "epoch": 8, "memory": 9465, "step": 83512} +{"lr": 0.0007297252973710758, "data_time": 0.001160430908203125, "grad_norm": 0.023272443003952502, "loss": 0.6937455773353577, "time": 0.4011988639831543, "epoch": 8, "memory": 9465, "step": 83612} +{"lr": 0.0007297252973710758, "data_time": 0.001248311996459961, "grad_norm": 0.03451326321810484, "loss": 0.6823717772960662, "time": 0.38255836963653567, "epoch": 8, "memory": 9465, "step": 83712} +{"lr": 0.0007297252973710758, "data_time": 0.001235032081604004, "grad_norm": 0.026614805823192, "loss": 0.6878908574581146, "time": 0.4235095739364624, "epoch": 8, "memory": 9465, "step": 83812} +{"lr": 0.0007297252973710758, "data_time": 0.001331329345703125, "grad_norm": 0.023098617256619037, "loss": 0.686500358581543, "time": 0.43498597145080564, "epoch": 8, "memory": 9465, "step": 83912} +{"lr": 0.0007297252973710758, "data_time": 0.001229715347290039, "grad_norm": 0.024808541499078275, "loss": 0.6883776068687439, "time": 0.41284153461456297, "epoch": 8, "memory": 9465, "step": 84012} +{"lr": 0.0007297252973710758, "data_time": 0.001268458366394043, "grad_norm": 0.0225776806473732, "loss": 0.6872128546237946, "time": 0.41157534122467043, "epoch": 8, "memory": 9465, "step": 84112} +{"lr": 0.0007297252973710758, "data_time": 0.001364898681640625, "grad_norm": 0.020946176210418343, "loss": 0.6859925150871277, "time": 0.40628390312194823, "epoch": 8, "memory": 9465, "step": 84212} +{"lr": 0.0007297252973710758, "data_time": 0.0010748863220214843, "grad_norm": 0.026104954653419554, "loss": 0.6917192161083221, "time": 0.38813974857330324, "epoch": 8, "memory": 9465, "step": 84312} +{"lr": 0.0007297252973710758, "data_time": 0.0010761260986328126, "grad_norm": 0.015459264512173831, "loss": 0.6921374261379242, "time": 0.4128370523452759, "epoch": 8, "memory": 9465, "step": 84412} +{"lr": 0.0007297252973710758, "data_time": 0.0010637283325195313, "grad_norm": 0.013546761852921918, "loss": 0.6864924967288971, "time": 0.42748820781707764, "epoch": 8, "memory": 9465, "step": 84512} +{"lr": 0.0007297252973710758, "data_time": 0.0011240243911743164, "grad_norm": 0.027023932128213347, "loss": 0.689000940322876, "time": 0.3852929353713989, "epoch": 8, "memory": 9465, "step": 84612} +{"lr": 0.0007297252973710758, "data_time": 0.4488420248031616, "grad_norm": 0.01824718089774251, "loss": 0.6857051968574523, "time": 0.6342859745025635, "epoch": 8, "memory": 9465, "step": 84712} +{"lr": 0.0007297252973710758, "data_time": 0.18600850105285643, "grad_norm": 0.023536592070013285, "loss": 0.6856965780258178, "time": 0.3697806358337402, "epoch": 8, "memory": 9465, "step": 84812} +{"lr": 0.0007297252973710758, "data_time": 0.19487955570220947, "grad_norm": 0.030082438513636588, "loss": 0.6878475427627564, "time": 0.42903966903686525, "epoch": 8, "memory": 9465, "step": 84912} +{"lr": 0.0007297252973710758, "data_time": 0.0012598514556884765, "grad_norm": 0.016145851742476226, "loss": 0.6812762320041656, "time": 0.42850422859191895, "epoch": 8, "memory": 9465, "step": 85012} +{"lr": 0.0007297252973710758, "data_time": 0.0013853549957275391, "grad_norm": 0.019658190233167262, "loss": 0.6805278420448303, "time": 0.4062574148178101, "epoch": 8, "memory": 9465, "step": 85112} +{"lr": 0.0007297252973710758, "data_time": 0.001171278953552246, "grad_norm": 0.019844309845939278, "loss": 0.6873342275619507, "time": 0.4607765436172485, "epoch": 8, "memory": 9465, "step": 85212} +{"lr": 0.0007297252973710758, "data_time": 0.001273798942565918, "grad_norm": 0.03570436269510537, "loss": 0.6896699666976929, "time": 0.40429470539093015, "epoch": 8, "memory": 9465, "step": 85312} +{"lr": 0.0007297252973710758, "data_time": 0.0012304544448852538, "grad_norm": 0.02322594877332449, "loss": 0.6885689616203308, "time": 0.4417370080947876, "epoch": 8, "memory": 9465, "step": 85412} +{"lr": 0.0007297252973710758, "data_time": 0.0013516902923583984, "grad_norm": 0.022229703702032567, "loss": 0.6852067291736603, "time": 0.37205698490142824, "epoch": 8, "memory": 9465, "step": 85512} +{"lr": 0.0007297252973710758, "data_time": 0.0012696266174316406, "grad_norm": 0.01820289217866957, "loss": 0.6862127542495727, "time": 0.41309685707092286, "epoch": 8, "memory": 9465, "step": 85612} +{"lr": 0.0007297252973710758, "data_time": 0.0014190673828125, "grad_norm": 0.028342454880475997, "loss": 0.6854941487312317, "time": 0.4241366147994995, "epoch": 8, "memory": 9465, "step": 85712} +{"lr": 0.0007297252973710758, "data_time": 0.0012758255004882812, "grad_norm": 0.016617366089485587, "loss": 0.6876291632652283, "time": 0.39320549964904783, "epoch": 8, "memory": 9465, "step": 85812} +{"lr": 0.0007297252973710758, "data_time": 0.0011086225509643554, "grad_norm": 0.014217626006575302, "loss": 0.6864918768405914, "time": 0.4088895797729492, "epoch": 8, "memory": 9465, "step": 85912} +{"lr": 0.0007297252973710758, "data_time": 0.0012398958206176758, "grad_norm": 0.009858486987650395, "loss": 0.6821779072284698, "time": 0.39141838550567626, "epoch": 8, "memory": 9465, "step": 86012} +{"lr": 0.0007297252973710758, "data_time": 0.0012630224227905273, "grad_norm": 0.026772038033232092, "loss": 0.684651905298233, "time": 0.42710540294647215, "epoch": 8, "memory": 9465, "step": 86112} +{"lr": 0.0007297252973710758, "data_time": 0.0012676239013671875, "grad_norm": 0.018086789874359965, "loss": 0.687130618095398, "time": 0.3956389665603638, "epoch": 8, "memory": 9465, "step": 86212} +{"lr": 0.0007297252973710758, "data_time": 0.001221632957458496, "grad_norm": 0.016600300755817442, "loss": 0.689531821012497, "time": 0.40857982635498047, "epoch": 8, "memory": 9465, "step": 86312} +{"lr": 0.0007297252973710758, "data_time": 0.0011564731597900391, "grad_norm": 0.02437058964278549, "loss": 0.6827565252780914, "time": 0.414105749130249, "epoch": 8, "memory": 9465, "step": 86412} +{"lr": 0.0007297252973710758, "data_time": 0.0011553525924682616, "grad_norm": 0.03960214058170095, "loss": 0.689712381362915, "time": 0.4052400588989258, "epoch": 8, "memory": 9465, "step": 86512} +{"lr": 0.0007297252973710758, "data_time": 0.0011344432830810546, "grad_norm": 0.022618011478334664, "loss": 0.6871628165245056, "time": 0.41004009246826173, "epoch": 8, "memory": 9465, "step": 86612} +{"lr": 0.0007297252973710758, "data_time": 0.0011563777923583984, "grad_norm": 0.015828104829415678, "loss": 0.6897797107696533, "time": 0.399530029296875, "epoch": 8, "memory": 9465, "step": 86712} +{"lr": 0.0007297252973710758, "data_time": 0.001150655746459961, "grad_norm": 0.029650754760950804, "loss": 0.6859541833400726, "time": 0.3921663284301758, "epoch": 8, "memory": 9465, "step": 86812} +{"lr": 0.0007297252973710758, "data_time": 0.0012651443481445312, "grad_norm": 0.01859331934247166, "loss": 0.694051080942154, "time": 0.42839818000793456, "epoch": 8, "memory": 9465, "step": 86912} +{"lr": 0.0007297252973710758, "data_time": 0.001208019256591797, "grad_norm": 0.022289685532450677, "loss": 0.6856022715568543, "time": 0.40266454219818115, "epoch": 8, "memory": 9465, "step": 87012} +{"lr": 0.0007297252973710758, "data_time": 0.0011482954025268555, "grad_norm": 0.02167035867460072, "loss": 0.6878075122833252, "time": 0.46272203922271726, "epoch": 8, "memory": 9465, "step": 87112} +{"lr": 0.0007297252973710758, "data_time": 0.0011661052703857422, "grad_norm": 0.014263918832875788, "loss": 0.6901412069797516, "time": 0.4115774393081665, "epoch": 8, "memory": 9465, "step": 87212} +{"lr": 0.0007297252973710758, "data_time": 0.0014473915100097657, "grad_norm": 0.01804503982421011, "loss": 0.6925581812858581, "time": 0.42563605308532715, "epoch": 8, "memory": 9465, "step": 87312} +{"lr": 0.0007297252973710758, "data_time": 0.0016886234283447266, "grad_norm": 0.01981075443327427, "loss": 0.6851196885108948, "time": 0.40288805961608887, "epoch": 8, "memory": 9465, "step": 87412} +{"lr": 0.0007297252973710758, "data_time": 0.0017270326614379882, "grad_norm": 0.023243179032579064, "loss": 0.6839932262897491, "time": 0.3996387004852295, "epoch": 8, "memory": 9465, "step": 87512} +{"lr": 0.0007297252973710758, "data_time": 0.0011975526809692382, "grad_norm": 0.02818984305486083, "loss": 0.6847031474113464, "time": 0.4121211767196655, "epoch": 8, "memory": 9465, "step": 87612} +{"lr": 0.0007297252973710758, "data_time": 0.0018535375595092774, "grad_norm": 0.02783660786226392, "loss": 0.6824337482452393, "time": 0.41201837062835694, "epoch": 8, "memory": 9465, "step": 87712} +{"lr": 0.0007297252973710758, "data_time": 0.001318216323852539, "grad_norm": 0.01894610975869, "loss": 0.6840990722179413, "time": 0.41992831230163574, "epoch": 8, "memory": 9465, "step": 87812} +{"lr": 0.0007297252973710758, "data_time": 0.0012621164321899414, "grad_norm": 0.020018459553830324, "loss": 0.6858372092247009, "time": 0.4103346824645996, "epoch": 8, "memory": 9465, "step": 87912} +{"lr": 0.0007297252973710758, "data_time": 0.0011466026306152343, "grad_norm": 0.01521154697984457, "loss": 0.6900941610336304, "time": 0.3752453327178955, "epoch": 8, "memory": 9465, "step": 88012} +{"lr": 0.0007297252973710758, "data_time": 0.0011850833892822266, "grad_norm": 0.03133570360951125, "loss": 0.6882062613964081, "time": 0.38090322017669676, "epoch": 8, "memory": 9465, "step": 88112} +{"lr": 0.0007297252973710758, "data_time": 0.0010639190673828124, "grad_norm": 0.01772117118816823, "loss": 0.6841363072395324, "time": 0.39819214344024656, "epoch": 8, "memory": 9465, "step": 88212} +{"lr": 0.0007297252973710758, "data_time": 0.0013394355773925781, "grad_norm": 0.026048915507271887, "loss": 0.6875195622444152, "time": 0.48444066047668455, "epoch": 8, "memory": 9465, "step": 88312} +{"lr": 0.0007297252973710758, "data_time": 0.001224374771118164, "grad_norm": 0.02083610587287694, "loss": 0.6871038615703583, "time": 0.4258877754211426, "epoch": 8, "memory": 9465, "step": 88412} +{"lr": 0.0007297252973710758, "data_time": 0.0012317895889282227, "grad_norm": 0.018010238697752357, "loss": 0.6880809366703033, "time": 0.4014564037322998, "epoch": 8, "memory": 9465, "step": 88512} +{"lr": 0.0007297252973710758, "data_time": 0.0014194965362548829, "grad_norm": 0.021679312142077835, "loss": 0.6838792622089386, "time": 0.38787331581115725, "epoch": 8, "memory": 9465, "step": 88612} +{"lr": 0.0007297252973710758, "data_time": 0.001169109344482422, "grad_norm": 0.022499702125787734, "loss": 0.6892295241355896, "time": 0.41290483474731443, "epoch": 8, "memory": 9465, "step": 88712} +{"lr": 0.0007297252973710758, "data_time": 0.0011868000030517579, "grad_norm": 0.03122631898149848, "loss": 0.689541357755661, "time": 0.40138251781463624, "epoch": 8, "memory": 9465, "step": 88812} +{"lr": 0.0007297252973710758, "data_time": 0.0011133670806884766, "grad_norm": 0.018278383743017913, "loss": 0.6843248307704926, "time": 0.37915620803833006, "epoch": 8, "memory": 9465, "step": 88912} +{"lr": 0.0007297252973710758, "data_time": 0.0012299537658691407, "grad_norm": 0.01585726662306115, "loss": 0.682845950126648, "time": 0.4047560453414917, "epoch": 8, "memory": 9465, "step": 89012} +{"lr": 0.0007297252973710758, "data_time": 0.0012470483779907227, "grad_norm": 0.01803396185860038, "loss": 0.6858809888362885, "time": 0.40805821418762206, "epoch": 8, "memory": 9465, "step": 89112} +{"lr": 0.0007297252973710758, "data_time": 0.0011643409729003907, "grad_norm": 0.03050779558252543, "loss": 0.6923555791378021, "time": 0.44962224960327146, "epoch": 8, "memory": 9465, "step": 89212} +{"lr": 0.0007297252973710758, "data_time": 0.001296687126159668, "grad_norm": 0.016095905905240217, "loss": 0.6883684039115906, "time": 0.4264099359512329, "epoch": 8, "memory": 9465, "step": 89312} +{"lr": 0.0007297252973710758, "data_time": 0.0011588096618652343, "grad_norm": 0.02768868338316679, "loss": 0.6868324279785156, "time": 0.39534337520599366, "epoch": 8, "memory": 9465, "step": 89412} +{"lr": 0.0007297252973710758, "data_time": 0.0011719226837158202, "grad_norm": 0.026650596118997782, "loss": 0.687915700674057, "time": 0.4062569856643677, "epoch": 8, "memory": 9465, "step": 89512} +{"lr": 0.0007297252973710758, "data_time": 0.0011327266693115234, "grad_norm": 0.025433100806549192, "loss": 0.6886302828788757, "time": 0.41426811218261717, "epoch": 8, "memory": 9465, "step": 89612} +{"lr": 0.0007297252973710758, "data_time": 0.0012078523635864259, "grad_norm": 0.03159934119321406, "loss": 0.6852601170539856, "time": 0.4151745796203613, "epoch": 8, "memory": 9465, "step": 89712} +{"lr": 0.0007297252973710758, "data_time": 0.0011881589889526367, "grad_norm": 0.01946532188449055, "loss": 0.689168268442154, "time": 0.4193460941314697, "epoch": 8, "memory": 9465, "step": 89812} +{"lr": 0.0007297252973710758, "data_time": 0.001197981834411621, "grad_norm": 0.030266016721725464, "loss": 0.6798470675945282, "time": 0.39806716442108153, "epoch": 8, "memory": 9465, "step": 89912} +{"lr": 0.0007297252973710758, "data_time": 0.0012630462646484376, "grad_norm": 0.03459731894545257, "loss": 0.6840537667274476, "time": 0.3963681697845459, "epoch": 8, "memory": 9465, "step": 90012} +{"lr": 0.0007297252973710758, "data_time": 0.001177668571472168, "grad_norm": 0.01730092572979629, "loss": 0.6854472517967224, "time": 0.4102996587753296, "epoch": 8, "memory": 9465, "step": 90112} +{"lr": 0.0007297252973710758, "data_time": 0.0011160135269165038, "grad_norm": 0.017559571348829194, "loss": 0.6928240776062011, "time": 0.40048491954803467, "epoch": 8, "memory": 9465, "step": 90212} +{"lr": 0.0007297252973710758, "data_time": 0.001104259490966797, "grad_norm": 0.020261542464140803, "loss": 0.6874666154384613, "time": 0.39814741611480714, "epoch": 8, "memory": 9465, "step": 90312} +{"lr": 0.0007297252973710758, "data_time": 0.0010899066925048827, "grad_norm": 0.016924621985526755, "loss": 0.683213222026825, "time": 0.37892324924468995, "epoch": 8, "memory": 9465, "step": 90412} +{"lr": 0.0007297252973710758, "data_time": 0.001235222816467285, "grad_norm": 0.016168435849249364, "loss": 0.6863664031028748, "time": 0.4211923360824585, "epoch": 8, "memory": 9465, "step": 90512} +{"lr": 0.0007297252973710758, "data_time": 0.0011926651000976562, "grad_norm": 0.023549871100112797, "loss": 0.6911492764949798, "time": 0.378183650970459, "epoch": 8, "memory": 9465, "step": 90612} +{"lr": 0.0007297252973710758, "data_time": 0.001127314567565918, "grad_norm": 0.02852336352225393, "loss": 0.684430330991745, "time": 0.41238906383514407, "epoch": 8, "memory": 9465, "step": 90712} +{"lr": 0.0007297252973710758, "data_time": 0.001247715950012207, "grad_norm": 0.014994151052087545, "loss": 0.6882424712181091, "time": 0.389222526550293, "epoch": 8, "memory": 9465, "step": 90812} +{"lr": 0.0007297252973710758, "data_time": 0.0012914896011352538, "grad_norm": 0.025133726629428567, "loss": 0.6864078521728516, "time": 0.41447546482086184, "epoch": 8, "memory": 9465, "step": 90912} +{"lr": 0.0007297252973710758, "data_time": 0.0013945579528808593, "grad_norm": 0.030232203379273414, "loss": 0.6928736448287964, "time": 0.42232742309570315, "epoch": 8, "memory": 9465, "step": 91012} +{"lr": 0.0007297252973710758, "data_time": 0.0015008926391601562, "grad_norm": 0.014172716904431582, "loss": 0.6900639533996582, "time": 0.3962843656539917, "epoch": 8, "memory": 9465, "step": 91112} +{"lr": 0.0007297252973710758, "data_time": 0.00219118595123291, "grad_norm": 0.026410426991060377, "loss": 0.6892644107341767, "time": 0.42124485969543457, "epoch": 8, "memory": 9465, "step": 91212} +{"lr": 0.0007297252973710758, "data_time": 0.001233220100402832, "grad_norm": 0.02648461067583412, "loss": 0.6854311227798462, "time": 0.41231694221496584, "epoch": 8, "memory": 9465, "step": 91312} +{"lr": 0.0007297252973710758, "data_time": 0.0011304378509521484, "grad_norm": 0.03151758648455143, "loss": 0.6852314591407775, "time": 0.40566048622131345, "epoch": 8, "memory": 9465, "step": 91412} +{"lr": 0.0007297252973710758, "data_time": 0.0012223958969116212, "grad_norm": 0.02860237369313836, "loss": 0.686645257472992, "time": 0.40322227478027345, "epoch": 8, "memory": 9465, "step": 91512} +{"lr": 0.0007297252973710758, "data_time": 0.0011346578598022462, "grad_norm": 0.025324149383232, "loss": 0.6860270082950592, "time": 0.43115417957305907, "epoch": 8, "memory": 9465, "step": 91612} +{"lr": 0.0007297252973710758, "data_time": 0.0013410806655883788, "grad_norm": 0.03411727251950651, "loss": 0.6894769132137298, "time": 0.42202248573303225, "epoch": 8, "memory": 9465, "step": 91712} +{"lr": 0.0007297252973710758, "data_time": 0.0012592554092407226, "grad_norm": 0.029838087595999242, "loss": 0.6915993332862854, "time": 0.3938627004623413, "epoch": 8, "memory": 9465, "step": 91812} +{"lr": 0.0007297252973710758, "data_time": 0.001236724853515625, "grad_norm": 0.02972464361228049, "loss": 0.6858161389827728, "time": 0.3675376415252686, "epoch": 8, "memory": 9465, "step": 91912} +{"lr": 0.0007297252973710758, "data_time": 0.001224827766418457, "grad_norm": 0.02403484135866165, "loss": 0.6879396319389344, "time": 0.44392731189727785, "epoch": 8, "memory": 9465, "step": 92012} +{"lr": 0.0007297252973710758, "data_time": 0.007342123985290527, "grad_norm": 0.016623965930193663, "loss": 0.6868308842182159, "time": 0.40576231479644775, "epoch": 8, "memory": 9465, "step": 92112} +{"accuracy/top1": 100.0, "data_time": 0.2820587438695571, "time": 0.5355185200186336, "step": 8} +{"lr": 0.0006579634122155991, "data_time": 0.21316945552825928, "grad_norm": 0.0234228708781302, "loss": 0.6916179835796357, "time": 0.396101975440979, "epoch": 9, "memory": 9465, "step": 92228} +{"lr": 0.0006579634122155991, "data_time": 0.04762403964996338, "grad_norm": 0.015285061160102486, "loss": 0.6891472935676575, "time": 0.43029587268829345, "epoch": 9, "memory": 9465, "step": 92328} +{"lr": 0.0006579634122155991, "data_time": 0.0012217283248901368, "grad_norm": 0.010836223780643195, "loss": 0.6847985863685608, "time": 0.4090731143951416, "epoch": 9, "memory": 9465, "step": 92428} +{"lr": 0.0006579634122155991, "data_time": 0.001170206069946289, "grad_norm": 0.023985810473095625, "loss": 0.6876272141933442, "time": 0.4341857671737671, "epoch": 9, "memory": 9465, "step": 92528} +{"lr": 0.0006579634122155991, "data_time": 0.0011333227157592773, "grad_norm": 0.024069071374833585, "loss": 0.6895691573619842, "time": 0.494635009765625, "epoch": 9, "memory": 9465, "step": 92628} +{"lr": 0.0006579634122155991, "data_time": 0.0014793872833251953, "grad_norm": 0.01886992247309536, "loss": 0.684550428390503, "time": 0.401670503616333, "epoch": 9, "memory": 9465, "step": 92728} +{"lr": 0.0006579634122155991, "data_time": 0.0013691902160644532, "grad_norm": 0.022801924776285887, "loss": 0.6878998517990113, "time": 0.41439669132232665, "epoch": 9, "memory": 9465, "step": 92828} +{"lr": 0.0006579634122155991, "data_time": 0.0012319564819335937, "grad_norm": 0.02375576599733904, "loss": 0.6876055479049683, "time": 0.4170017957687378, "epoch": 9, "memory": 9465, "step": 92928} +{"lr": 0.0006579634122155991, "data_time": 0.0012707471847534179, "grad_norm": 0.016375646647065877, "loss": 0.6861737668514252, "time": 0.40897631645202637, "epoch": 9, "memory": 9465, "step": 93028} +{"lr": 0.0006579634122155991, "data_time": 0.0012650728225708009, "grad_norm": 0.0159662498626858, "loss": 0.6853348195552826, "time": 0.4129276990890503, "epoch": 9, "memory": 9465, "step": 93128} +{"lr": 0.0006579634122155991, "data_time": 0.0012438297271728516, "grad_norm": 0.021633118903264404, "loss": 0.6862115025520324, "time": 0.40263330936431885, "epoch": 9, "memory": 9465, "step": 93228} +{"lr": 0.0006579634122155991, "data_time": 0.001230168342590332, "grad_norm": 0.030090645235031844, "loss": 0.6870250225067138, "time": 0.39903361797332765, "epoch": 9, "memory": 9465, "step": 93328} +{"lr": 0.0006579634122155991, "data_time": 0.0012187480926513672, "grad_norm": 0.019000594364479186, "loss": 0.6873220205307007, "time": 0.388306713104248, "epoch": 9, "memory": 9465, "step": 93428} +{"lr": 0.0006579634122155991, "data_time": 0.0011335134506225586, "grad_norm": 0.021870032860897483, "loss": 0.6882164001464843, "time": 0.42376468181610105, "epoch": 9, "memory": 9465, "step": 93528} +{"lr": 0.0006579634122155991, "data_time": 0.001211690902709961, "grad_norm": 0.02227505382616073, "loss": 0.685147899389267, "time": 0.3879836559295654, "epoch": 9, "memory": 9465, "step": 93628} +{"lr": 0.0006579634122155991, "data_time": 0.001041102409362793, "grad_norm": 0.020397668541409075, "loss": 0.686493045091629, "time": 0.4055838108062744, "epoch": 9, "memory": 9465, "step": 93728} +{"lr": 0.0006579634122155991, "data_time": 0.001279306411743164, "grad_norm": 0.026303742500022053, "loss": 0.6855940282344818, "time": 0.4523406267166138, "epoch": 9, "memory": 9465, "step": 93828} +{"lr": 0.0006579634122155991, "data_time": 0.0016120433807373046, "grad_norm": 0.020197241753339767, "loss": 0.688467937707901, "time": 0.437934947013855, "epoch": 9, "memory": 9465, "step": 93928} +{"lr": 0.0006579634122155991, "data_time": 0.001268315315246582, "grad_norm": 0.02168433303013444, "loss": 0.6895956993103027, "time": 0.3716425895690918, "epoch": 9, "memory": 9465, "step": 94028} +{"lr": 0.0006579634122155991, "data_time": 0.001171255111694336, "grad_norm": 0.021969118062406777, "loss": 0.6842774868011474, "time": 0.394378662109375, "epoch": 9, "memory": 9465, "step": 94128} +{"lr": 0.0006579634122155991, "data_time": 0.0011736154556274414, "grad_norm": 0.022297876980155708, "loss": 0.6893316030502319, "time": 0.38500878810882566, "epoch": 9, "memory": 9465, "step": 94228} +{"lr": 0.0006579634122155991, "data_time": 0.0012616634368896485, "grad_norm": 0.02019666782580316, "loss": 0.6862617373466492, "time": 0.4002028226852417, "epoch": 9, "memory": 9465, "step": 94328} +{"lr": 0.0006579634122155991, "data_time": 0.0013898849487304688, "grad_norm": 0.02815713370218873, "loss": 0.6930375337600708, "time": 0.42486631870269775, "epoch": 9, "memory": 9465, "step": 94428} +{"lr": 0.0006579634122155991, "data_time": 0.001305699348449707, "grad_norm": 0.01949597888160497, "loss": 0.6851170063018799, "time": 0.38422245979309083, "epoch": 9, "memory": 9465, "step": 94528} +{"lr": 0.0006579634122155991, "data_time": 0.001342177391052246, "grad_norm": 0.008654957532417028, "loss": 0.6843866109848022, "time": 0.43166699409484866, "epoch": 9, "memory": 9465, "step": 94628} +{"lr": 0.0006579634122155991, "data_time": 0.0012539863586425782, "grad_norm": 0.020748585771070794, "loss": 0.6877948224544526, "time": 0.4316052198410034, "epoch": 9, "memory": 9465, "step": 94728} +{"lr": 0.0006579634122155991, "data_time": 0.001287221908569336, "grad_norm": 0.016579051362350584, "loss": 0.6863248229026795, "time": 0.3958492994308472, "epoch": 9, "memory": 9465, "step": 94828} +{"lr": 0.0006579634122155991, "data_time": 0.0012607574462890625, "grad_norm": 0.0332856229506433, "loss": 0.6900288879871368, "time": 0.439675498008728, "epoch": 9, "memory": 9465, "step": 94928} +{"lr": 0.0006579634122155991, "data_time": 0.0012688636779785156, "grad_norm": 0.030556600773707032, "loss": 0.6874858319759369, "time": 0.4259216785430908, "epoch": 9, "memory": 9465, "step": 95028} +{"lr": 0.0006579634122155991, "data_time": 0.0012953042984008788, "grad_norm": 0.023428255063481628, "loss": 0.6891760647296905, "time": 0.41685597896575927, "epoch": 9, "memory": 9465, "step": 95128} +{"lr": 0.0006579634122155991, "data_time": 0.0010758161544799805, "grad_norm": 0.016678719443734735, "loss": 0.6888595283031463, "time": 0.4266186237335205, "epoch": 9, "memory": 9465, "step": 95228} +{"lr": 0.0006579634122155991, "data_time": 0.0013385772705078124, "grad_norm": 0.01847218030306976, "loss": 0.6866507172584534, "time": 0.42702975273132326, "epoch": 9, "memory": 9465, "step": 95328} +{"lr": 0.0006579634122155991, "data_time": 0.001246500015258789, "grad_norm": 0.018197174969827756, "loss": 0.6861519396305085, "time": 0.4428264856338501, "epoch": 9, "memory": 9465, "step": 95428} +{"lr": 0.0006579634122155991, "data_time": 0.0017017602920532226, "grad_norm": 0.03002235693857074, "loss": 0.6791818618774415, "time": 0.4208193302154541, "epoch": 9, "memory": 9465, "step": 95528} +{"lr": 0.0006579634122155991, "data_time": 0.001302480697631836, "grad_norm": 0.024182516941800715, "loss": 0.687585073709488, "time": 0.3937672138214111, "epoch": 9, "memory": 9465, "step": 95628} +{"lr": 0.0006579634122155991, "data_time": 0.0012251853942871094, "grad_norm": 0.017337709735147654, "loss": 0.6875366985797882, "time": 0.4132816314697266, "epoch": 9, "memory": 9465, "step": 95728} +{"lr": 0.0006579634122155991, "data_time": 0.0011914968490600586, "grad_norm": 0.01870363319758326, "loss": 0.685635793209076, "time": 0.3906074285507202, "epoch": 9, "memory": 9465, "step": 95828} +{"lr": 0.0006579634122155991, "data_time": 0.0012731313705444335, "grad_norm": 0.026897142501547934, "loss": 0.6896338284015655, "time": 0.40096268653869627, "epoch": 9, "memory": 9465, "step": 95928} +{"lr": 0.0006579634122155991, "data_time": 0.0011874675750732423, "grad_norm": 0.026509001932572572, "loss": 0.6889773905277252, "time": 0.4020853519439697, "epoch": 9, "memory": 9465, "step": 96028} +{"lr": 0.0006579634122155991, "data_time": 0.0011425018310546875, "grad_norm": 0.017402482125908135, "loss": 0.6823527455329895, "time": 0.42270355224609374, "epoch": 9, "memory": 9465, "step": 96128} +{"lr": 0.0006579634122155991, "data_time": 0.0011400461196899414, "grad_norm": 0.02895842413417995, "loss": 0.6830333232879638, "time": 0.37814233303070066, "epoch": 9, "memory": 9465, "step": 96228} +{"lr": 0.0006579634122155991, "data_time": 0.0011759757995605468, "grad_norm": 0.023471585963852702, "loss": 0.6858464479446411, "time": 0.42189993858337405, "epoch": 9, "memory": 9465, "step": 96328} +{"lr": 0.0006579634122155991, "data_time": 0.001221466064453125, "grad_norm": 0.0167266639880836, "loss": 0.6858098208904266, "time": 0.41883187294006347, "epoch": 9, "memory": 9465, "step": 96428} +{"lr": 0.0006579634122155991, "data_time": 0.0012425899505615235, "grad_norm": 0.01642593522556126, "loss": 0.6833726942539216, "time": 0.3721702814102173, "epoch": 9, "memory": 9465, "step": 96528} +{"lr": 0.0006579634122155991, "data_time": 0.0013547182083129884, "grad_norm": 0.01680624104337767, "loss": 0.6890249371528625, "time": 0.3774850845336914, "epoch": 9, "memory": 9465, "step": 96628} +{"lr": 0.0006579634122155991, "data_time": 0.0012394905090332032, "grad_norm": 0.0279288194142282, "loss": 0.6881324768066406, "time": 0.40616984367370607, "epoch": 9, "memory": 9465, "step": 96728} +{"lr": 0.0006579634122155991, "data_time": 0.0012462854385375977, "grad_norm": 0.01503880093805492, "loss": 0.6883620500564576, "time": 0.40850157737731935, "epoch": 9, "memory": 9465, "step": 96828} +{"lr": 0.0006579634122155991, "data_time": 0.001309037208557129, "grad_norm": 0.025298327021300793, "loss": 0.6819562196731568, "time": 0.4332940101623535, "epoch": 9, "memory": 9465, "step": 96928} +{"lr": 0.0006579634122155991, "data_time": 0.0012633800506591797, "grad_norm": 0.0193327292567119, "loss": 0.686792254447937, "time": 0.37613890171051023, "epoch": 9, "memory": 9465, "step": 97028} +{"lr": 0.0006579634122155991, "data_time": 0.0012845039367675782, "grad_norm": 0.020719549898058177, "loss": 0.6884935736656189, "time": 0.3788888931274414, "epoch": 9, "memory": 9465, "step": 97128} +{"lr": 0.0006579634122155991, "data_time": 0.0013127088546752929, "grad_norm": 0.02421436319127679, "loss": 0.6868078708648682, "time": 0.4024457216262817, "epoch": 9, "memory": 9465, "step": 97228} +{"lr": 0.0006579634122155991, "data_time": 0.001302814483642578, "grad_norm": 0.01988020937424153, "loss": 0.6854013919830322, "time": 0.42250816822052, "epoch": 9, "memory": 9465, "step": 97328} +{"lr": 0.0006579634122155991, "data_time": 0.0011916160583496094, "grad_norm": 0.022660642373375596, "loss": 0.6879563987255096, "time": 0.38332290649414064, "epoch": 9, "memory": 9465, "step": 97428} +{"lr": 0.0006579634122155991, "data_time": 0.0013241767883300781, "grad_norm": 0.02686635326826945, "loss": 0.6822048485279083, "time": 0.39268765449523924, "epoch": 9, "memory": 9465, "step": 97528} +{"lr": 0.0006579634122155991, "data_time": 0.0016896724700927734, "grad_norm": 0.027195728803053498, "loss": 0.6898121058940887, "time": 0.4255571126937866, "epoch": 9, "memory": 9465, "step": 97628} +{"lr": 0.0006579634122155991, "data_time": 0.0016141176223754884, "grad_norm": 0.0168335084192222, "loss": 0.6855267524719239, "time": 0.4973564863204956, "epoch": 9, "memory": 9465, "step": 97728} +{"lr": 0.0006579634122155991, "data_time": 0.0012742280960083008, "grad_norm": 0.015347224660217762, "loss": 0.6904765784740448, "time": 0.412762975692749, "epoch": 9, "memory": 9465, "step": 97828} +{"lr": 0.0006579634122155991, "data_time": 0.001308298110961914, "grad_norm": 0.014064839849015697, "loss": 0.6878889083862305, "time": 0.40703887939453126, "epoch": 9, "memory": 9465, "step": 97928} +{"lr": 0.0006579634122155991, "data_time": 0.0012140274047851562, "grad_norm": 0.029925155732780696, "loss": 0.6871650576591491, "time": 0.44304039478302004, "epoch": 9, "memory": 9465, "step": 98028} +{"lr": 0.0006579634122155991, "data_time": 0.0013524770736694336, "grad_norm": 0.023618994350545108, "loss": 0.6859748244285584, "time": 0.39569027423858644, "epoch": 9, "memory": 9465, "step": 98128} +{"lr": 0.0006579634122155991, "data_time": 0.0012902021408081055, "grad_norm": 0.018692607013508677, "loss": 0.6859188854694367, "time": 0.3941347122192383, "epoch": 9, "memory": 9465, "step": 98228} +{"lr": 0.0006579634122155991, "data_time": 0.0013632535934448241, "grad_norm": 0.01690459387609735, "loss": 0.6879775702953339, "time": 0.4223912000656128, "epoch": 9, "memory": 9465, "step": 98328} +{"lr": 0.0006579634122155991, "data_time": 0.001253199577331543, "grad_norm": 0.02279731973540038, "loss": 0.6867093026638031, "time": 0.38918366432189944, "epoch": 9, "memory": 9465, "step": 98428} +{"lr": 0.0006579634122155991, "data_time": 0.001302337646484375, "grad_norm": 0.025851780222728847, "loss": 0.6896377682685852, "time": 0.38460984230041506, "epoch": 9, "memory": 9465, "step": 98528} +{"lr": 0.0006579634122155991, "data_time": 0.0011951923370361328, "grad_norm": 0.011671100463718175, "loss": 0.6869745969772338, "time": 0.41370439529418945, "epoch": 9, "memory": 9465, "step": 98628} +{"lr": 0.0006579634122155991, "data_time": 0.0012279510498046874, "grad_norm": 0.030667106620967388, "loss": 0.6872926354408264, "time": 0.37771522998809814, "epoch": 9, "memory": 9465, "step": 98728} +{"lr": 0.0006579634122155991, "data_time": 0.0011768102645874023, "grad_norm": 0.012709329824429006, "loss": 0.688293707370758, "time": 0.40707762241363527, "epoch": 9, "memory": 9465, "step": 98828} +{"lr": 0.0006579634122155991, "data_time": 0.0012590408325195313, "grad_norm": 0.03183954511769116, "loss": 0.6826155662536622, "time": 0.39499711990356445, "epoch": 9, "memory": 9465, "step": 98928} +{"lr": 0.0006579634122155991, "data_time": 0.0011528730392456055, "grad_norm": 0.019440938270417974, "loss": 0.6883721292018891, "time": 0.3722320318222046, "epoch": 9, "memory": 9465, "step": 99028} +{"lr": 0.0006579634122155991, "data_time": 0.0012531518936157227, "grad_norm": 0.027565144328400494, "loss": 0.6806558012962342, "time": 0.38812177181243895, "epoch": 9, "memory": 9465, "step": 99128} +{"lr": 0.0006579634122155991, "data_time": 0.0011704921722412109, "grad_norm": 0.02675180253572762, "loss": 0.6876543939113617, "time": 0.40490987300872805, "epoch": 9, "memory": 9465, "step": 99228} +{"lr": 0.0006579634122155991, "data_time": 0.0012486696243286133, "grad_norm": 0.030168826109729707, "loss": 0.6888048231601716, "time": 0.381657338142395, "epoch": 9, "memory": 9465, "step": 99328} +{"lr": 0.0006579634122155991, "data_time": 0.0011351585388183593, "grad_norm": 0.028028684156015515, "loss": 0.6887628912925721, "time": 0.4116715431213379, "epoch": 9, "memory": 9465, "step": 99428} +{"lr": 0.0006579634122155991, "data_time": 0.001234722137451172, "grad_norm": 0.015527657582424581, "loss": 0.6899739027023315, "time": 0.41590731143951415, "epoch": 9, "memory": 9465, "step": 99528} +{"lr": 0.0006579634122155991, "data_time": 0.0012464523315429688, "grad_norm": 0.030353108141571283, "loss": 0.6921488285064697, "time": 0.47470993995666505, "epoch": 9, "memory": 9465, "step": 99628} +{"lr": 0.0006579634122155991, "data_time": 0.0012114286422729493, "grad_norm": 0.029835300310514866, "loss": 0.6887951672077179, "time": 0.4400243520736694, "epoch": 9, "memory": 9465, "step": 99728} +{"lr": 0.0006579634122155991, "data_time": 0.0012079715728759766, "grad_norm": 0.025192872807383537, "loss": 0.6869944512844086, "time": 0.3784618377685547, "epoch": 9, "memory": 9465, "step": 99828} +{"lr": 0.0006579634122155991, "data_time": 0.0012845754623413085, "grad_norm": 0.02521377563243732, "loss": 0.6825485050678253, "time": 0.3821455478668213, "epoch": 9, "memory": 9465, "step": 99928} +{"lr": 0.0006579634122155991, "data_time": 0.0012262344360351562, "grad_norm": 0.023734575789421796, "loss": 0.686549437046051, "time": 0.42076358795166013, "epoch": 9, "memory": 9465, "step": 100028} +{"lr": 0.0006579634122155991, "data_time": 0.001176929473876953, "grad_norm": 0.022418857598677278, "loss": 0.6821747839450836, "time": 0.3837489366531372, "epoch": 9, "memory": 9465, "step": 100128} +{"lr": 0.0006579634122155991, "data_time": 0.0012503862380981445, "grad_norm": 0.02005963319679722, "loss": 0.6834537267684937, "time": 0.398844575881958, "epoch": 9, "memory": 9465, "step": 100228} +{"lr": 0.0006579634122155991, "data_time": 0.001214003562927246, "grad_norm": 0.02611571641173214, "loss": 0.6858814537525177, "time": 0.4339579105377197, "epoch": 9, "memory": 9465, "step": 100328} +{"lr": 0.0006579634122155991, "data_time": 0.0012241601943969727, "grad_norm": 0.03148531918413937, "loss": 0.6862568199634552, "time": 0.40589926242828367, "epoch": 9, "memory": 9465, "step": 100428} +{"lr": 0.0006579634122155991, "data_time": 0.0012091636657714845, "grad_norm": 0.033940835297107695, "loss": 0.6856343150138855, "time": 0.3979132890701294, "epoch": 9, "memory": 9465, "step": 100528} +{"lr": 0.0006579634122155991, "data_time": 0.001158738136291504, "grad_norm": 0.019394355139229446, "loss": 0.6881734967231751, "time": 0.4203500270843506, "epoch": 9, "memory": 9465, "step": 100628} +{"lr": 0.0006579634122155991, "data_time": 0.0012041568756103516, "grad_norm": 0.020852927234955132, "loss": 0.6846345901489258, "time": 0.4341021776199341, "epoch": 9, "memory": 9465, "step": 100728} +{"lr": 0.0006579634122155991, "data_time": 0.0011731863021850586, "grad_norm": 0.021724076382815836, "loss": 0.6865647315979004, "time": 0.4252492666244507, "epoch": 9, "memory": 9465, "step": 100828} +{"lr": 0.0006579634122155991, "data_time": 0.00117340087890625, "grad_norm": 0.01835655963514, "loss": 0.6896529734134674, "time": 0.4314539670944214, "epoch": 9, "memory": 9465, "step": 100928} +{"lr": 0.0006579634122155991, "data_time": 0.0012068510055541991, "grad_norm": 0.02051096213981509, "loss": 0.6877733886241912, "time": 0.42658677101135256, "epoch": 9, "memory": 9465, "step": 101028} +{"lr": 0.0006579634122155991, "data_time": 0.0012018442153930663, "grad_norm": 0.022103709960356356, "loss": 0.687483298778534, "time": 0.3764130353927612, "epoch": 9, "memory": 9465, "step": 101128} +{"lr": 0.0006579634122155991, "data_time": 0.0013461589813232421, "grad_norm": 0.031499871611595155, "loss": 0.6910863101482392, "time": 0.4267105579376221, "epoch": 9, "memory": 9465, "step": 101228} +{"lr": 0.0006579634122155991, "data_time": 0.0013976573944091796, "grad_norm": 0.020280804665526377, "loss": 0.6822407126426697, "time": 0.418511962890625, "epoch": 9, "memory": 9465, "step": 101328} +{"lr": 0.0006579634122155991, "data_time": 0.0012742996215820313, "grad_norm": 0.021270034462213518, "loss": 0.6890386521816254, "time": 0.4001427173614502, "epoch": 9, "memory": 9465, "step": 101428} +{"lr": 0.0006579634122155991, "data_time": 0.0012665987014770508, "grad_norm": 0.020746080833487213, "loss": 0.6890262246131897, "time": 0.43777031898498536, "epoch": 9, "memory": 9465, "step": 101528} +{"lr": 0.0006579634122155991, "data_time": 0.00123443603515625, "grad_norm": 0.030383997317403554, "loss": 0.6889639258384704, "time": 0.41897132396698, "epoch": 9, "memory": 9465, "step": 101628} +{"lr": 0.0006579634122155991, "data_time": 0.0012430429458618164, "grad_norm": 0.028350890969159083, "loss": 0.6842820227146149, "time": 0.3890629768371582, "epoch": 9, "memory": 9465, "step": 101728} +{"lr": 0.0006579634122155991, "data_time": 0.001310443878173828, "grad_norm": 0.026324432389810682, "loss": 0.6874460697174072, "time": 0.4347853183746338, "epoch": 9, "memory": 9465, "step": 101828} +{"lr": 0.0006579634122155991, "data_time": 0.0011256933212280273, "grad_norm": 0.02528970562852919, "loss": 0.688225245475769, "time": 0.4044374465942383, "epoch": 9, "memory": 9465, "step": 101928} +{"lr": 0.0006579634122155991, "data_time": 0.0013589859008789062, "grad_norm": 0.03300562538206577, "loss": 0.6893951058387756, "time": 0.41233081817626954, "epoch": 9, "memory": 9465, "step": 102028} +{"lr": 0.0006579634122155991, "data_time": 0.0012313604354858398, "grad_norm": 0.01551717626862228, "loss": 0.6841417729854584, "time": 0.40691328048706055, "epoch": 9, "memory": 9465, "step": 102128} +{"lr": 0.0006579634122155991, "data_time": 0.0013275861740112305, "grad_norm": 0.018457791511900724, "loss": 0.6879209458827973, "time": 0.396535587310791, "epoch": 9, "memory": 9465, "step": 102228} +{"lr": 0.0006579634122155991, "data_time": 0.0011706590652465821, "grad_norm": 0.02424745513126254, "loss": 0.683158141374588, "time": 0.4237743616104126, "epoch": 9, "memory": 9465, "step": 102328} +{"lr": 0.0006579634122155991, "data_time": 0.001270437240600586, "grad_norm": 0.025121981231495737, "loss": 0.6909631550312042, "time": 0.39954843521118166, "epoch": 9, "memory": 9465, "step": 102428} +{"lr": 0.0006579634122155991, "data_time": 0.001394367218017578, "grad_norm": 0.01886954139918089, "loss": 0.6874514400959015, "time": 0.3963741064071655, "epoch": 9, "memory": 9465, "step": 102528} +{"lr": 0.0006579634122155991, "data_time": 0.0015153884887695312, "grad_norm": 0.021362853090977296, "loss": 0.6876825988292694, "time": 0.41606242656707765, "epoch": 9, "memory": 9465, "step": 102628} +{"lr": 0.0006579634122155991, "data_time": 0.0012106180191040039, "grad_norm": 0.017824929393827914, "loss": 0.6851708590984344, "time": 0.40473825931549073, "epoch": 9, "memory": 9465, "step": 102728} +{"lr": 0.0006579634122155991, "data_time": 0.0011362075805664063, "grad_norm": 0.024924999999348073, "loss": 0.6884391784667969, "time": 0.41689631938934324, "epoch": 9, "memory": 9465, "step": 102828} +{"lr": 0.0006579634122155991, "data_time": 0.0012011051177978516, "grad_norm": 0.01819471991620958, "loss": 0.6900894582271576, "time": 0.38872756958007815, "epoch": 9, "memory": 9465, "step": 102928} +{"lr": 0.0006579634122155991, "data_time": 0.0011909723281860352, "grad_norm": 0.02048467976273969, "loss": 0.6794112265110016, "time": 0.40266592502593995, "epoch": 9, "memory": 9465, "step": 103028} +{"lr": 0.0006579634122155991, "data_time": 0.0012024402618408202, "grad_norm": 0.019741402816725893, "loss": 0.6905640184879303, "time": 0.4593635559082031, "epoch": 9, "memory": 9465, "step": 103128} +{"lr": 0.0006579634122155991, "data_time": 0.001197671890258789, "grad_norm": 0.014451659796759487, "loss": 0.687445068359375, "time": 0.4000712871551514, "epoch": 9, "memory": 9465, "step": 103228} +{"lr": 0.0006579634122155991, "data_time": 0.002363777160644531, "grad_norm": 0.03637495692819357, "loss": 0.6868040025234222, "time": 0.4112098217010498, "epoch": 9, "memory": 9465, "step": 103328} +{"lr": 0.0006579634122155991, "data_time": 0.0013206005096435547, "grad_norm": 0.024048620974645017, "loss": 0.6868383586406708, "time": 0.40303263664245603, "epoch": 9, "memory": 9465, "step": 103428} +{"lr": 0.0006579634122155991, "data_time": 0.0011281490325927735, "grad_norm": 0.028736840840429066, "loss": 0.6859465658664703, "time": 0.4159888029098511, "epoch": 9, "memory": 9465, "step": 103528} +{"lr": 0.0006579634122155991, "data_time": 0.007851600646972656, "grad_norm": 0.0213921959977597, "loss": 0.6927263677120209, "time": 0.39697990417480467, "epoch": 9, "memory": 9465, "step": 103628} +{"accuracy/top1": 100.0, "data_time": 0.29366599812227134, "time": 0.546827610801248, "step": 9} +{"lr": 0.0005824350601949144, "data_time": 0.0804297924041748, "grad_norm": 0.013860852117068135, "loss": 0.684959477186203, "time": 0.38721024990081787, "epoch": 10, "memory": 9465, "step": 103744} +{"lr": 0.0005824350601949144, "data_time": 0.1245685338973999, "grad_norm": 0.018632193864323198, "loss": 0.6929050743579864, "time": 0.37447028160095214, "epoch": 10, "memory": 9465, "step": 103844} +{"lr": 0.0005824350601949144, "data_time": 0.0011754274368286134, "grad_norm": 0.019292819849215447, "loss": 0.6920138597488403, "time": 0.3911352872848511, "epoch": 10, "memory": 9465, "step": 103944} +{"lr": 0.0005824350601949144, "data_time": 0.0014087915420532226, "grad_norm": 0.020488617662340402, "loss": 0.6909263849258422, "time": 0.41617798805236816, "epoch": 10, "memory": 9465, "step": 104044} +{"lr": 0.0005824350601949144, "data_time": 0.0013932228088378907, "grad_norm": 0.02226473849732429, "loss": 0.6895103275775909, "time": 0.39312267303466797, "epoch": 10, "memory": 9465, "step": 104144} +{"lr": 0.0005824350601949144, "data_time": 0.0014477014541625977, "grad_norm": 0.023597622476518153, "loss": 0.6877914369106293, "time": 0.4322096347808838, "epoch": 10, "memory": 9465, "step": 104244} +{"lr": 0.0005824350601949144, "data_time": 0.001198554039001465, "grad_norm": 0.01968251629732549, "loss": 0.6820769429206848, "time": 0.4268588304519653, "epoch": 10, "memory": 9465, "step": 104344} +{"lr": 0.0005824350601949144, "data_time": 0.001171422004699707, "grad_norm": 0.016102762633818202, "loss": 0.6871347963809967, "time": 0.42116336822509765, "epoch": 10, "memory": 9465, "step": 104444} +{"lr": 0.0005824350601949144, "data_time": 0.001184391975402832, "grad_norm": 0.018892600177787246, "loss": 0.6875913798809051, "time": 0.39415688514709474, "epoch": 10, "memory": 9465, "step": 104544} +{"lr": 0.0005824350601949144, "data_time": 0.0013638973236083985, "grad_norm": 0.028315074648708104, "loss": 0.6880623936653137, "time": 0.41167688369750977, "epoch": 10, "memory": 9465, "step": 104644} +{"lr": 0.0005824350601949144, "data_time": 0.0013931512832641602, "grad_norm": 0.025507490057498215, "loss": 0.6845067024230957, "time": 0.4208183526992798, "epoch": 10, "memory": 9465, "step": 104744} +{"lr": 0.0005824350601949144, "data_time": 0.0014745235443115235, "grad_norm": 0.03318627830594778, "loss": 0.6920928955078125, "time": 0.42701256275177, "epoch": 10, "memory": 9465, "step": 104844} +{"lr": 0.0005824350601949144, "data_time": 0.001563429832458496, "grad_norm": 0.021531057625543325, "loss": 0.6887095749378205, "time": 0.40250022411346437, "epoch": 10, "memory": 9465, "step": 104944} +{"lr": 0.0005824350601949144, "data_time": 0.0012716531753540039, "grad_norm": 0.019334487069863827, "loss": 0.6901447772979736, "time": 0.4074194669723511, "epoch": 10, "memory": 9465, "step": 105044} +{"lr": 0.0005824350601949144, "data_time": 0.0013767242431640624, "grad_norm": 0.020994788268581033, "loss": 0.6887441456317902, "time": 0.4227423191070557, "epoch": 10, "memory": 9465, "step": 105144} +{"lr": 0.0005824350601949144, "data_time": 0.0013175487518310546, "grad_norm": 0.011245237581897527, "loss": 0.6934567332267761, "time": 0.40805351734161377, "epoch": 10, "memory": 9465, "step": 105244} +{"lr": 0.0005824350601949144, "data_time": 0.0012956619262695312, "grad_norm": 0.016551787732169032, "loss": 0.6876155853271484, "time": 0.4531639337539673, "epoch": 10, "memory": 9465, "step": 105344} +{"lr": 0.0005824350601949144, "data_time": 0.0013253927230834962, "grad_norm": 0.03667248806450516, "loss": 0.6847810208797455, "time": 0.4050638675689697, "epoch": 10, "memory": 9465, "step": 105444} +{"lr": 0.0005824350601949144, "data_time": 0.0012734651565551758, "grad_norm": 0.015060140064451843, "loss": 0.6832494020462037, "time": 0.42069971561431885, "epoch": 10, "memory": 9465, "step": 105544} +{"lr": 0.0005824350601949144, "data_time": 0.0013261079788208009, "grad_norm": 0.01931798756122589, "loss": 0.6855398416519165, "time": 0.3904897928237915, "epoch": 10, "memory": 9465, "step": 105644} +{"lr": 0.0005824350601949144, "data_time": 0.0013689517974853516, "grad_norm": 0.019257644531899133, "loss": 0.6879183769226074, "time": 0.42248609066009524, "epoch": 10, "memory": 9465, "step": 105744} +{"lr": 0.0005824350601949144, "data_time": 0.001352548599243164, "grad_norm": 0.015942090796306728, "loss": 0.6878875076770783, "time": 0.41836502552032473, "epoch": 10, "memory": 9465, "step": 105844} +{"lr": 0.0005824350601949144, "data_time": 0.0013040781021118164, "grad_norm": 0.015442746109329164, "loss": 0.6886744022369384, "time": 0.42407824993133547, "epoch": 10, "memory": 9465, "step": 105944} +{"lr": 0.0005824350601949144, "data_time": 0.0014090538024902344, "grad_norm": 0.027122497325763106, "loss": 0.6842085003852845, "time": 0.4306140661239624, "epoch": 10, "memory": 9465, "step": 106044} +{"lr": 0.0005824350601949144, "data_time": 0.0012214183807373047, "grad_norm": 0.02640653122216463, "loss": 0.6877634763717652, "time": 0.41361632347106936, "epoch": 10, "memory": 9465, "step": 106144} +{"lr": 0.0005824350601949144, "data_time": 0.0012167930603027345, "grad_norm": 0.015804155403748153, "loss": 0.6896946609020234, "time": 0.4164297342300415, "epoch": 10, "memory": 9465, "step": 106244} +{"lr": 0.0005824350601949144, "data_time": 0.0012772560119628906, "grad_norm": 0.019613543688319624, "loss": 0.6849848628044128, "time": 0.4279727220535278, "epoch": 10, "memory": 9465, "step": 106344} +{"lr": 0.0005824350601949144, "data_time": 0.0011636734008789063, "grad_norm": 0.02436817828565836, "loss": 0.6858685910701752, "time": 0.3970036029815674, "epoch": 10, "memory": 9465, "step": 106444} +{"lr": 0.0005824350601949144, "data_time": 0.0012277126312255859, "grad_norm": 0.019343935884535313, "loss": 0.6839489638805389, "time": 0.3918203592300415, "epoch": 10, "memory": 9465, "step": 106544} +{"lr": 0.0005824350601949144, "data_time": 0.0012360572814941405, "grad_norm": 0.031003579404205082, "loss": 0.6848509788513184, "time": 0.4468141794204712, "epoch": 10, "memory": 9465, "step": 106644} +{"lr": 0.0005824350601949144, "data_time": 0.0013895988464355468, "grad_norm": 0.030244832765311, "loss": 0.6901010274887085, "time": 0.42935519218444823, "epoch": 10, "memory": 9465, "step": 106744} +{"lr": 0.0005824350601949144, "data_time": 0.0013350248336791992, "grad_norm": 0.019989809626713394, "loss": 0.6829847097396851, "time": 0.40437328815460205, "epoch": 10, "memory": 9465, "step": 106844} +{"lr": 0.0005824350601949144, "data_time": 0.0012146472930908204, "grad_norm": 0.020788033190183342, "loss": 0.6902220666408538, "time": 0.3854987382888794, "epoch": 10, "memory": 9465, "step": 106944} +{"lr": 0.0005824350601949144, "data_time": 0.0011486530303955079, "grad_norm": 0.02227186084492132, "loss": 0.6845167636871338, "time": 0.39415597915649414, "epoch": 10, "memory": 9465, "step": 107044} +{"lr": 0.0005824350601949144, "data_time": 0.0011939287185668945, "grad_norm": 0.015719819790683688, "loss": 0.6869750022888184, "time": 0.41704909801483153, "epoch": 10, "memory": 9465, "step": 107144} +{"lr": 0.0005824350601949144, "data_time": 0.0012527942657470704, "grad_norm": 0.018355125561356544, "loss": 0.6825352430343627, "time": 0.41206822395324705, "epoch": 10, "memory": 9465, "step": 107244} +{"lr": 0.0005824350601949144, "data_time": 0.0012099266052246094, "grad_norm": 0.020194839173927902, "loss": 0.6805288314819335, "time": 0.40749800205230713, "epoch": 10, "memory": 9465, "step": 107344} +{"lr": 0.0005824350601949144, "data_time": 0.0013065338134765625, "grad_norm": 0.019903663831064476, "loss": 0.6839028298854828, "time": 0.41579749584198, "epoch": 10, "memory": 9465, "step": 107444} +{"lr": 0.0005824350601949144, "data_time": 0.0012295007705688476, "grad_norm": 0.023555016424506903, "loss": 0.6887306034564972, "time": 0.40896925926208494, "epoch": 10, "memory": 9465, "step": 107544} +{"lr": 0.0005824350601949144, "data_time": 0.0013578414916992187, "grad_norm": 0.008026356843765825, "loss": 0.6864909052848815, "time": 0.39210598468780516, "epoch": 10, "memory": 9465, "step": 107644} +{"lr": 0.0005824350601949144, "data_time": 0.0013225793838500977, "grad_norm": 0.01994687117403373, "loss": 0.6855593800544739, "time": 0.41773104667663574, "epoch": 10, "memory": 9465, "step": 107744} +{"lr": 0.0005824350601949144, "data_time": 0.0011875391006469726, "grad_norm": 0.015936049411538988, "loss": 0.6819383978843689, "time": 0.4239983558654785, "epoch": 10, "memory": 9465, "step": 107844} +{"lr": 0.0005824350601949144, "data_time": 0.0012995481491088867, "grad_norm": 0.0163878261228092, "loss": 0.6855361342430115, "time": 0.4112903833389282, "epoch": 10, "memory": 9465, "step": 107944} +{"lr": 0.0005824350601949144, "data_time": 0.001398181915283203, "grad_norm": 0.019322561647277324, "loss": 0.6876338481903076, "time": 0.4021707057952881, "epoch": 10, "memory": 9465, "step": 108044} +{"lr": 0.0005824350601949144, "data_time": 0.0012164831161499024, "grad_norm": 0.01873928731074557, "loss": 0.6879354000091553, "time": 0.4513095378875732, "epoch": 10, "memory": 9465, "step": 108144} +{"lr": 0.0005824350601949144, "data_time": 0.0015116214752197265, "grad_norm": 0.031186088593676685, "loss": 0.6894765496253967, "time": 0.39508953094482424, "epoch": 10, "memory": 9465, "step": 108244} +{"lr": 0.0005824350601949144, "data_time": 0.00115659236907959, "grad_norm": 0.019406280026305466, "loss": 0.6872951865196228, "time": 0.38469705581665037, "epoch": 10, "memory": 9465, "step": 108344} +{"lr": 0.0005824350601949144, "data_time": 0.0013974905014038086, "grad_norm": 0.016117651236709208, "loss": 0.6880839169025421, "time": 0.414851975440979, "epoch": 10, "memory": 9465, "step": 108444} +{"lr": 0.0005824350601949144, "data_time": 0.0012469530105590821, "grad_norm": 0.021941594569943844, "loss": 0.691986083984375, "time": 0.3929115056991577, "epoch": 10, "memory": 9465, "step": 108544} +{"lr": 0.0005824350601949144, "data_time": 0.001242852210998535, "grad_norm": 0.03727470549638383, "loss": 0.689212304353714, "time": 0.3955352306365967, "epoch": 10, "memory": 9465, "step": 108644} +{"lr": 0.0005824350601949144, "data_time": 0.0013436079025268555, "grad_norm": 0.036667583021335305, "loss": 0.6849142134189605, "time": 0.4265798330307007, "epoch": 10, "memory": 9465, "step": 108744} +{"lr": 0.0005824350601949144, "data_time": 0.0012444496154785157, "grad_norm": 0.030465218820609152, "loss": 0.6896410048007965, "time": 0.3953246116638184, "epoch": 10, "memory": 9465, "step": 108844} +{"lr": 0.0005824350601949144, "data_time": 0.001269221305847168, "grad_norm": 0.017400019080378117, "loss": 0.6845678567886353, "time": 0.4077514886856079, "epoch": 10, "memory": 9465, "step": 108944} +{"lr": 0.0005824350601949144, "data_time": 0.0013150691986083985, "grad_norm": 0.02755537023767829, "loss": 0.684292984008789, "time": 0.4132340431213379, "epoch": 10, "memory": 9465, "step": 109044} +{"lr": 0.0005824350601949144, "data_time": 0.0011596202850341797, "grad_norm": 0.017595816362882034, "loss": 0.6844670414924622, "time": 0.39069271087646484, "epoch": 10, "memory": 9465, "step": 109144} +{"lr": 0.0005824350601949144, "data_time": 0.0012371301651000976, "grad_norm": 0.013313841167837381, "loss": 0.6874563574790955, "time": 0.41736831665039065, "epoch": 10, "memory": 9465, "step": 109244} +{"lr": 0.0005824350601949144, "data_time": 0.001233220100402832, "grad_norm": 0.016935050557367503, "loss": 0.6864929080009461, "time": 0.417470645904541, "epoch": 10, "memory": 9465, "step": 109344} +{"lr": 0.0005824350601949144, "data_time": 0.001202535629272461, "grad_norm": 0.022035099868662655, "loss": 0.6885490894317627, "time": 0.41223371028900146, "epoch": 10, "memory": 9465, "step": 109444} +{"lr": 0.0005824350601949144, "data_time": 0.0012761354446411133, "grad_norm": 0.02410687633091584, "loss": 0.6844773352146148, "time": 0.39715070724487306, "epoch": 10, "memory": 9465, "step": 109544} +{"lr": 0.0005824350601949144, "data_time": 0.001157855987548828, "grad_norm": 0.01917083191219717, "loss": 0.688711941242218, "time": 0.40023744106292725, "epoch": 10, "memory": 9465, "step": 109644} +{"lr": 0.0005824350601949144, "data_time": 0.0013338088989257812, "grad_norm": 0.019987792504252866, "loss": 0.6903825759887695, "time": 0.4490066051483154, "epoch": 10, "memory": 9465, "step": 109744} +{"lr": 0.0005824350601949144, "data_time": 0.0012418031692504883, "grad_norm": 0.025325045277713797, "loss": 0.6888796389102936, "time": 0.40377490520477294, "epoch": 10, "memory": 9465, "step": 109844} +{"lr": 0.0005824350601949144, "data_time": 0.0012151479721069336, "grad_norm": 0.025578478910028936, "loss": 0.6866759479045867, "time": 0.42577595710754396, "epoch": 10, "memory": 9465, "step": 109944} +{"lr": 0.0005824350601949144, "data_time": 0.001265406608581543, "grad_norm": 0.030046832282096148, "loss": 0.6862091302871705, "time": 0.41481242179870603, "epoch": 10, "memory": 9465, "step": 110044} +{"lr": 0.0005824350601949144, "data_time": 0.0011924505233764648, "grad_norm": 0.022230990463867784, "loss": 0.680774348974228, "time": 0.4091353416442871, "epoch": 10, "memory": 9465, "step": 110144} +{"lr": 0.0005824350601949144, "data_time": 0.0012224674224853515, "grad_norm": 0.02642576047219336, "loss": 0.6916180908679962, "time": 0.3774830102920532, "epoch": 10, "memory": 9465, "step": 110244} +{"lr": 0.0005824350601949144, "data_time": 0.0012408018112182618, "grad_norm": 0.030100225936621428, "loss": 0.692443722486496, "time": 0.39855546951293946, "epoch": 10, "memory": 9465, "step": 110344} +{"lr": 0.0005824350601949144, "data_time": 0.0013021230697631836, "grad_norm": 0.02331925416365266, "loss": 0.6915897607803345, "time": 0.4083374261856079, "epoch": 10, "memory": 9465, "step": 110444} +{"lr": 0.0005824350601949144, "data_time": 0.0012148380279541015, "grad_norm": 0.01458752986509353, "loss": 0.6902616620063782, "time": 0.4076382637023926, "epoch": 10, "memory": 9465, "step": 110544} +{"lr": 0.0005824350601949144, "data_time": 0.001472330093383789, "grad_norm": 0.019471430499106646, "loss": 0.689478051662445, "time": 0.4589089393615723, "epoch": 10, "memory": 9465, "step": 110644} +{"lr": 0.0005824350601949144, "data_time": 0.0012511014938354492, "grad_norm": 0.016889481525868177, "loss": 0.6862181067466736, "time": 0.4069258213043213, "epoch": 10, "memory": 9465, "step": 110744} +{"lr": 0.0005824350601949144, "data_time": 0.0012067079544067383, "grad_norm": 0.024049390805885196, "loss": 0.6864958047866822, "time": 0.3966191053390503, "epoch": 10, "memory": 9465, "step": 110844} +{"lr": 0.0005824350601949144, "data_time": 0.0013444185256958007, "grad_norm": 0.019203606760129333, "loss": 0.6846551954746246, "time": 0.4123916864395142, "epoch": 10, "memory": 9465, "step": 110944} +{"lr": 0.0005824350601949144, "data_time": 0.0014084815979003907, "grad_norm": 0.030780664936173707, "loss": 0.6866586089134217, "time": 0.4366321086883545, "epoch": 10, "memory": 9465, "step": 111044} +{"lr": 0.0005824350601949144, "data_time": 0.0012423276901245117, "grad_norm": 0.02523041081149131, "loss": 0.682997876405716, "time": 0.4074322462081909, "epoch": 10, "memory": 9465, "step": 111144} +{"lr": 0.0005824350601949144, "data_time": 0.001174616813659668, "grad_norm": 0.015973761153873056, "loss": 0.682374769449234, "time": 0.42370193004608153, "epoch": 10, "memory": 9465, "step": 111244} +{"lr": 0.0005824350601949144, "data_time": 0.0011987686157226562, "grad_norm": 0.023700334306340665, "loss": 0.6827379882335662, "time": 0.3967373609542847, "epoch": 10, "memory": 9465, "step": 111344} +{"lr": 0.0005824350601949144, "data_time": 0.0012578964233398438, "grad_norm": 0.019831335172057152, "loss": 0.6807875454425811, "time": 0.4308000087738037, "epoch": 10, "memory": 9465, "step": 111444} +{"lr": 0.0005824350601949144, "data_time": 0.0013262033462524414, "grad_norm": 0.02222868036478758, "loss": 0.6855636715888977, "time": 0.40865292549133303, "epoch": 10, "memory": 9465, "step": 111544} +{"lr": 0.0005824350601949144, "data_time": 0.0013063907623291015, "grad_norm": 0.01934937541373074, "loss": 0.6886425614356995, "time": 0.4392297029495239, "epoch": 10, "memory": 9465, "step": 111644} +{"lr": 0.0005824350601949144, "data_time": 0.0014129638671875, "grad_norm": 0.018973741790978237, "loss": 0.6863362193107605, "time": 0.4058452367782593, "epoch": 10, "memory": 9465, "step": 111744} +{"lr": 0.0005824350601949144, "data_time": 0.0012498617172241211, "grad_norm": 0.01542835826985538, "loss": 0.6853533446788788, "time": 0.42419798374176027, "epoch": 10, "memory": 9465, "step": 111844} +{"lr": 0.0005824350601949144, "data_time": 0.0012610197067260743, "grad_norm": 0.01657525389455259, "loss": 0.6841616094112396, "time": 0.4221523284912109, "epoch": 10, "memory": 9465, "step": 111944} +{"lr": 0.0005824350601949144, "data_time": 0.0013408422470092773, "grad_norm": 0.020618134015239774, "loss": 0.6899962604045868, "time": 0.4089498519897461, "epoch": 10, "memory": 9465, "step": 112044} +{"lr": 0.0005824350601949144, "data_time": 0.001222085952758789, "grad_norm": 0.017995651811361313, "loss": 0.6861677825450897, "time": 0.41017196178436277, "epoch": 10, "memory": 9465, "step": 112144} +{"lr": 0.0005824350601949144, "data_time": 0.0013261556625366211, "grad_norm": 0.020788966538384558, "loss": 0.6859902501106262, "time": 0.38940489292144775, "epoch": 10, "memory": 9465, "step": 112244} +{"lr": 0.0005824350601949144, "data_time": 0.0013193845748901366, "grad_norm": 0.016596147720701993, "loss": 0.68940749168396, "time": 0.45871844291687014, "epoch": 10, "memory": 9465, "step": 112344} +{"lr": 0.0005824350601949144, "data_time": 0.0012585639953613282, "grad_norm": 0.024194601358613, "loss": 0.6857317388057709, "time": 0.4222301721572876, "epoch": 10, "memory": 9465, "step": 112444} +{"lr": 0.0005824350601949144, "data_time": 0.0013432025909423829, "grad_norm": 0.024401912372559308, "loss": 0.6879629909992218, "time": 0.4044027328491211, "epoch": 10, "memory": 9465, "step": 112544} +{"lr": 0.0005824350601949144, "data_time": 0.00132293701171875, "grad_norm": 0.01939155168365687, "loss": 0.6883826732635498, "time": 0.41016521453857424, "epoch": 10, "memory": 9465, "step": 112644} +{"lr": 0.0005824350601949144, "data_time": 0.0012619733810424806, "grad_norm": 0.023890880751423538, "loss": 0.690188205242157, "time": 0.412406325340271, "epoch": 10, "memory": 9465, "step": 112744} +{"lr": 0.0005824350601949144, "data_time": 0.0012366294860839844, "grad_norm": 0.03024088656529784, "loss": 0.6842540144920349, "time": 0.4187090635299683, "epoch": 10, "memory": 9465, "step": 112844} +{"lr": 0.0005824350601949144, "data_time": 0.0013144254684448243, "grad_norm": 0.03496296582743526, "loss": 0.6869054853916168, "time": 0.3987264156341553, "epoch": 10, "memory": 9465, "step": 112944} +{"lr": 0.0005824350601949144, "data_time": 0.0012226104736328125, "grad_norm": 0.02497257823124528, "loss": 0.6926319003105164, "time": 0.4907251834869385, "epoch": 10, "memory": 9465, "step": 113044} +{"lr": 0.0005824350601949144, "data_time": 0.0012903213500976562, "grad_norm": 0.029246483603492378, "loss": 0.6850897789001464, "time": 0.42286856174468995, "epoch": 10, "memory": 9465, "step": 113144} +{"lr": 0.0005824350601949144, "data_time": 0.0011801004409790039, "grad_norm": 0.0186012509977445, "loss": 0.6832999348640442, "time": 0.4044408082962036, "epoch": 10, "memory": 9465, "step": 113244} +{"lr": 0.0005824350601949144, "data_time": 0.0012021064758300781, "grad_norm": 0.01730568166822195, "loss": 0.686027979850769, "time": 0.3872162103652954, "epoch": 10, "memory": 9465, "step": 113344} +{"lr": 0.0005824350601949144, "data_time": 0.0012649059295654296, "grad_norm": 0.021720893494784833, "loss": 0.691539216041565, "time": 0.3952643394470215, "epoch": 10, "memory": 9465, "step": 113444} +{"lr": 0.0005824350601949144, "data_time": 0.0013449668884277343, "grad_norm": 0.016197738028131424, "loss": 0.684565806388855, "time": 0.42101283073425294, "epoch": 10, "memory": 9465, "step": 113544} +{"lr": 0.0005824350601949144, "data_time": 0.0013080120086669921, "grad_norm": 0.020098356902599333, "loss": 0.6852782845497132, "time": 0.40638060569763185, "epoch": 10, "memory": 9465, "step": 113644} +{"lr": 0.0005824350601949144, "data_time": 0.0014593124389648438, "grad_norm": 0.029148967657238244, "loss": 0.6879068195819855, "time": 0.44136064052581786, "epoch": 10, "memory": 9465, "step": 113744} +{"lr": 0.0005824350601949144, "data_time": 0.00114288330078125, "grad_norm": 0.02002816878957674, "loss": 0.6871368885040283, "time": 0.42678251266479494, "epoch": 10, "memory": 9465, "step": 113844} +{"lr": 0.0005824350601949144, "data_time": 0.0012193679809570312, "grad_norm": 0.01567211588844657, "loss": 0.6868526756763458, "time": 0.404461932182312, "epoch": 10, "memory": 9465, "step": 113944} +{"lr": 0.0005824350601949144, "data_time": 0.001265573501586914, "grad_norm": 0.018838391359895467, "loss": 0.6869959533214569, "time": 0.42400307655334474, "epoch": 10, "memory": 9465, "step": 114044} +{"lr": 0.0005824350601949144, "data_time": 0.0012116432189941406, "grad_norm": 0.020720690640155227, "loss": 0.6856711804866791, "time": 0.41737139225006104, "epoch": 10, "memory": 9465, "step": 114144} +{"lr": 0.0005824350601949144, "data_time": 0.001208019256591797, "grad_norm": 0.02710183516610414, "loss": 0.6836977303028107, "time": 0.4092968463897705, "epoch": 10, "memory": 9465, "step": 114244} +{"lr": 0.0005824350601949144, "data_time": 0.0013273954391479492, "grad_norm": 0.022203076060395687, "loss": 0.6899046838283539, "time": 0.4248371124267578, "epoch": 10, "memory": 9465, "step": 114344} +{"lr": 0.0005824350601949144, "data_time": 0.0013229846954345703, "grad_norm": 0.02756560589186847, "loss": 0.688097459077835, "time": 0.40036327838897706, "epoch": 10, "memory": 9465, "step": 114444} +{"lr": 0.0005824350601949144, "data_time": 0.001192760467529297, "grad_norm": 0.020312445575837047, "loss": 0.6833064258098602, "time": 0.4081276893615723, "epoch": 10, "memory": 9465, "step": 114544} +{"lr": 0.0005824350601949144, "data_time": 0.001324915885925293, "grad_norm": 0.019513128011021764, "loss": 0.6836528837680816, "time": 0.4081394195556641, "epoch": 10, "memory": 9465, "step": 114644} +{"lr": 0.0005824350601949144, "data_time": 0.0012111425399780273, "grad_norm": 0.01780185862444341, "loss": 0.6832261919975281, "time": 0.427313494682312, "epoch": 10, "memory": 9465, "step": 114744} +{"lr": 0.0005824350601949144, "data_time": 0.0012250900268554687, "grad_norm": 0.017171705327928066, "loss": 0.6893916189670563, "time": 0.3691226959228516, "epoch": 10, "memory": 9465, "step": 114844} +{"lr": 0.0005824350601949144, "data_time": 0.002109813690185547, "grad_norm": 0.031381679512560365, "loss": 0.6834503531455993, "time": 0.4566788911819458, "epoch": 10, "memory": 9465, "step": 114944} +{"lr": 0.0005824350601949144, "data_time": 0.0016053199768066406, "grad_norm": 0.01726784014608711, "loss": 0.6817797839641571, "time": 0.44130597114562986, "epoch": 10, "memory": 9465, "step": 115044} +{"lr": 0.0005824350601949144, "data_time": 0.0073616266250610355, "grad_norm": 0.01543681266484782, "loss": 0.6815003514289856, "time": 0.42305099964141846, "epoch": 10, "memory": 9465, "step": 115144} +{"accuracy/top1": 100.0, "data_time": 0.31313453001134534, "time": 0.5658689386704389, "step": 10} +{"lr": 0.0005050000000000001, "data_time": 0.14164562225341798, "grad_norm": 0.018051203107461333, "loss": 0.6851199448108674, "time": 0.38462414741516116, "epoch": 11, "memory": 9465, "step": 115260} +{"lr": 0.0005050000000000001, "data_time": 0.0012343883514404296, "grad_norm": 0.03034874270670116, "loss": 0.6858681082725525, "time": 0.4263297080993652, "epoch": 11, "memory": 9465, "step": 115360} +{"lr": 0.0005050000000000001, "data_time": 0.001535511016845703, "grad_norm": 0.026627645036205648, "loss": 0.6922080814838409, "time": 0.4277806758880615, "epoch": 11, "memory": 9465, "step": 115460} +{"lr": 0.0005050000000000001, "data_time": 0.0012439966201782226, "grad_norm": 0.025177362328395247, "loss": 0.6910534083843232, "time": 0.4225890636444092, "epoch": 11, "memory": 9465, "step": 115560} +{"lr": 0.0005050000000000001, "data_time": 0.0012293577194213868, "grad_norm": 0.025857007037848233, "loss": 0.6852569341659546, "time": 0.4010313987731934, "epoch": 11, "memory": 9465, "step": 115660} +{"lr": 0.0005050000000000001, "data_time": 0.0012104034423828125, "grad_norm": 0.019561112439259887, "loss": 0.690194845199585, "time": 0.41695983409881593, "epoch": 11, "memory": 9465, "step": 115760} +{"lr": 0.0005050000000000001, "data_time": 0.0014673948287963867, "grad_norm": 0.01657569374947343, "loss": 0.6906159043312072, "time": 0.4154002904891968, "epoch": 11, "memory": 9465, "step": 115860} +{"lr": 0.0005050000000000001, "data_time": 0.0013001680374145508, "grad_norm": 0.018498407537117602, "loss": 0.6843622982501983, "time": 0.4089759349822998, "epoch": 11, "memory": 9465, "step": 115960} +{"lr": 0.0005050000000000001, "data_time": 0.0014367341995239259, "grad_norm": 0.019207592034945264, "loss": 0.6852038204669952, "time": 0.41964030265808105, "epoch": 11, "memory": 9465, "step": 116060} +{"lr": 0.0005050000000000001, "data_time": 0.0016860723495483398, "grad_norm": 0.025294320657849312, "loss": 0.6915205597877503, "time": 0.40537395477294924, "epoch": 11, "memory": 9465, "step": 116160} +{"lr": 0.0005050000000000001, "data_time": 0.0013617992401123047, "grad_norm": 0.03438049773685634, "loss": 0.6858792901039124, "time": 0.4081716775894165, "epoch": 11, "memory": 9465, "step": 116260} +{"lr": 0.0005050000000000001, "data_time": 0.0012148618698120117, "grad_norm": 0.024008320609573273, "loss": 0.6888195455074311, "time": 0.40371079444885255, "epoch": 11, "memory": 9465, "step": 116360} +{"lr": 0.0005050000000000001, "data_time": 0.0012859582901000976, "grad_norm": 0.02932122233323753, "loss": 0.6876915872097016, "time": 0.41438817977905273, "epoch": 11, "memory": 9465, "step": 116460} +{"lr": 0.0005050000000000001, "data_time": 0.0012621164321899414, "grad_norm": 0.022084969142451883, "loss": 0.6869653582572937, "time": 0.4203095197677612, "epoch": 11, "memory": 9465, "step": 116560} +{"lr": 0.0005050000000000001, "data_time": 0.0012619972229003906, "grad_norm": 0.0145648067293223, "loss": 0.6853629887104035, "time": 0.41915919780731203, "epoch": 11, "memory": 9465, "step": 116660} +{"lr": 0.0005050000000000001, "data_time": 0.0013493776321411132, "grad_norm": 0.02500706957653165, "loss": 0.6874424040317535, "time": 0.42435810565948484, "epoch": 11, "memory": 9465, "step": 116760} +{"lr": 0.0005050000000000001, "data_time": 0.001235651969909668, "grad_norm": 0.012749405700014904, "loss": 0.6920623362064362, "time": 0.4669149160385132, "epoch": 11, "memory": 9465, "step": 116860} +{"lr": 0.0005050000000000001, "data_time": 0.001263904571533203, "grad_norm": 0.012591069098562002, "loss": 0.6829416632652283, "time": 0.4248375654220581, "epoch": 11, "memory": 9465, "step": 116960} +{"lr": 0.0005050000000000001, "data_time": 0.0012073278427124023, "grad_norm": 0.02648661967832595, "loss": 0.6873231291770935, "time": 0.4109781265258789, "epoch": 11, "memory": 9465, "step": 117060} +{"lr": 0.0005050000000000001, "data_time": 0.0014919757843017579, "grad_norm": 0.022504648682661353, "loss": 0.6939443469047546, "time": 0.4185105562210083, "epoch": 11, "memory": 9465, "step": 117160} +{"lr": 0.0005050000000000001, "data_time": 0.0014093399047851562, "grad_norm": 0.01594353067339398, "loss": 0.6896181166172027, "time": 0.42703282833099365, "epoch": 11, "memory": 9465, "step": 117260} +{"lr": 0.0005050000000000001, "data_time": 0.0012426137924194335, "grad_norm": 0.02223492683842778, "loss": 0.688520348072052, "time": 0.4168489933013916, "epoch": 11, "memory": 9465, "step": 117360} +{"lr": 0.0005050000000000001, "data_time": 0.0013597726821899414, "grad_norm": 0.02550687571638264, "loss": 0.6830237984657288, "time": 0.4211557626724243, "epoch": 11, "memory": 9465, "step": 117460} +{"lr": 0.0005050000000000001, "data_time": 0.0012285232543945313, "grad_norm": 0.029983170243212953, "loss": 0.6894727110862732, "time": 0.4175532102584839, "epoch": 11, "memory": 9465, "step": 117560} +{"lr": 0.0005050000000000001, "data_time": 0.001228642463684082, "grad_norm": 0.023841824021656067, "loss": 0.685344797372818, "time": 0.4121598243713379, "epoch": 11, "memory": 9465, "step": 117660} +{"lr": 0.0005050000000000001, "data_time": 0.0013395786285400391, "grad_norm": 0.03169720433652401, "loss": 0.6829832375049592, "time": 0.4200542211532593, "epoch": 11, "memory": 9465, "step": 117760} +{"lr": 0.0005050000000000001, "data_time": 0.0012280702590942382, "grad_norm": 0.018102559680119156, "loss": 0.6852307736873626, "time": 0.4196688175201416, "epoch": 11, "memory": 9465, "step": 117860} +{"lr": 0.0005050000000000001, "data_time": 0.0013463258743286132, "grad_norm": 0.021662126760929823, "loss": 0.6852931618690491, "time": 0.4109614610671997, "epoch": 11, "memory": 9465, "step": 117960} +{"lr": 0.0005050000000000001, "data_time": 0.0013517141342163086, "grad_norm": 0.025886794889811427, "loss": 0.6886232554912567, "time": 0.42843639850616455, "epoch": 11, "memory": 9465, "step": 118060} +{"lr": 0.0005050000000000001, "data_time": 0.001398921012878418, "grad_norm": 0.018659650138579308, "loss": 0.6858470857143402, "time": 0.4139073848724365, "epoch": 11, "memory": 9465, "step": 118160} +{"lr": 0.0005050000000000001, "data_time": 0.0013772964477539063, "grad_norm": 0.02023630291223526, "loss": 0.690393716096878, "time": 0.4513564109802246, "epoch": 11, "memory": 9465, "step": 118260} +{"lr": 0.0005050000000000001, "data_time": 0.0011944055557250976, "grad_norm": 0.03460154184140265, "loss": 0.690068107843399, "time": 0.4073303699493408, "epoch": 11, "memory": 9465, "step": 118360} +{"lr": 0.0005050000000000001, "data_time": 0.001187276840209961, "grad_norm": 0.019733951438684016, "loss": 0.6868141531944275, "time": 0.4054840326309204, "epoch": 11, "memory": 9465, "step": 118460} +{"lr": 0.0005050000000000001, "data_time": 0.0012950897216796875, "grad_norm": 0.029082222480792553, "loss": 0.6933544933795929, "time": 0.436969518661499, "epoch": 11, "memory": 9465, "step": 118560} +{"lr": 0.0005050000000000001, "data_time": 0.0012650489807128906, "grad_norm": 0.030345451540779323, "loss": 0.6888700425624847, "time": 0.44688436985015867, "epoch": 11, "memory": 9465, "step": 118660} +{"lr": 0.0005050000000000001, "data_time": 0.0013356685638427734, "grad_norm": 0.033061045221984386, "loss": 0.6866863250732422, "time": 0.4123854398727417, "epoch": 11, "memory": 9465, "step": 118760} +{"lr": 0.0005050000000000001, "data_time": 0.0012506484985351563, "grad_norm": 0.018534427601844074, "loss": 0.6873053550720215, "time": 0.41796696186065674, "epoch": 11, "memory": 9465, "step": 118860} +{"lr": 0.0005050000000000001, "data_time": 0.0013590335845947265, "grad_norm": 0.022318332106806338, "loss": 0.6839488089084625, "time": 0.4034369468688965, "epoch": 11, "memory": 9465, "step": 118960} +{"lr": 0.0005050000000000001, "data_time": 0.0014347076416015626, "grad_norm": 0.04936443231999874, "loss": 0.6854557871818543, "time": 0.44001576900482176, "epoch": 11, "memory": 9465, "step": 119060} +{"lr": 0.0005050000000000001, "data_time": 0.0013712167739868164, "grad_norm": 0.11262964736670256, "loss": 0.6847430109977722, "time": 0.4342339515686035, "epoch": 11, "memory": 9465, "step": 119160} +{"lr": 0.0005050000000000001, "data_time": 0.0012398719787597655, "grad_norm": 0.047233869507908824, "loss": 0.6832587540149688, "time": 0.43180623054504397, "epoch": 11, "memory": 9465, "step": 119260} +{"lr": 0.0005050000000000001, "data_time": 0.00135495662689209, "grad_norm": 0.22071321560069918, "loss": 0.6838791906833649, "time": 0.40990076065063474, "epoch": 11, "memory": 9465, "step": 119360} +{"lr": 0.0005050000000000001, "data_time": 0.0012654304504394532, "grad_norm": 5.8065368115901945, "loss": 0.6771953344345093, "time": 0.4447260618209839, "epoch": 11, "memory": 9465, "step": 119460} +{"lr": 0.0005050000000000001, "data_time": 0.0013187646865844727, "grad_norm": 2.108214722573757, "loss": 0.6880443215370178, "time": 0.41931159496307374, "epoch": 11, "memory": 9465, "step": 119560} +{"lr": 0.0005050000000000001, "data_time": 0.0013099193572998046, "grad_norm": 11.026284086704255, "loss": 0.6785147964954377, "time": 0.4144579887390137, "epoch": 11, "memory": 9465, "step": 119660} +{"lr": 0.0005050000000000001, "data_time": 0.00125577449798584, "grad_norm": 1.82330731600523, "loss": 0.6819824814796448, "time": 0.4068878173828125, "epoch": 11, "memory": 9465, "step": 119760} +{"lr": 0.0005050000000000001, "data_time": 0.0012647390365600586, "grad_norm": 1.4509143367409707, "loss": 0.67999626994133, "time": 0.4289198398590088, "epoch": 11, "memory": 9465, "step": 119860} +{"lr": 0.0005050000000000001, "data_time": 0.0012695074081420898, "grad_norm": 1.1200944758951663, "loss": 0.678550523519516, "time": 0.4281005859375, "epoch": 11, "memory": 9465, "step": 119960} +{"lr": 0.0005050000000000001, "data_time": 0.0012899875640869141, "grad_norm": 0.08603005451150239, "loss": 0.6788290202617645, "time": 0.4542787790298462, "epoch": 11, "memory": 9465, "step": 120060} +{"lr": 0.0005050000000000001, "data_time": 0.0012372970581054688, "grad_norm": 0.3386359246447682, "loss": 0.6814377546310425, "time": 0.43357300758361816, "epoch": 11, "memory": 9465, "step": 120160} +{"lr": 0.0005050000000000001, "data_time": 0.0012021780014038086, "grad_norm": 0.10021833311766386, "loss": 0.6798589050769805, "time": 0.40308120250701907, "epoch": 11, "memory": 9465, "step": 120260} +{"lr": 0.0005050000000000001, "data_time": 0.0012942314147949218, "grad_norm": 0.14789856560528278, "loss": 0.6738684296607971, "time": 0.4436042785644531, "epoch": 11, "memory": 9465, "step": 120360} +{"lr": 0.0005050000000000001, "data_time": 0.0013613224029541016, "grad_norm": 1.6710926983505487, "loss": 0.6735816538333893, "time": 0.3911003828048706, "epoch": 11, "memory": 9465, "step": 120460} +{"lr": 0.0005050000000000001, "data_time": 0.0012817859649658203, "grad_norm": 0.08404935821890831, "loss": 0.6804642856121064, "time": 0.43289999961853026, "epoch": 11, "memory": 9465, "step": 120560} +{"lr": 0.0005050000000000001, "data_time": 0.0013806343078613282, "grad_norm": 9.849910619854928, "loss": 0.6692216038703919, "time": 0.4129424333572388, "epoch": 11, "memory": 9465, "step": 120660} +{"lr": 0.0005050000000000001, "data_time": 0.0013935327529907226, "grad_norm": 0.683423176407814, "loss": 0.6871620059013367, "time": 0.43312318325042726, "epoch": 11, "memory": 9465, "step": 120760} +{"lr": 0.0005050000000000001, "data_time": 0.001224350929260254, "grad_norm": 47.25621848106384, "loss": 0.6792611658573151, "time": 0.4233675003051758, "epoch": 11, "memory": 9465, "step": 120860} +{"lr": 0.0005050000000000001, "data_time": 0.001247739791870117, "grad_norm": 0.5494678448885679, "loss": 0.673319411277771, "time": 0.40434112548828127, "epoch": 11, "memory": 9465, "step": 120960} +{"lr": 0.0005050000000000001, "data_time": 0.0014000177383422852, "grad_norm": 0.34597770050168036, "loss": 0.6815334379673004, "time": 0.4122968912124634, "epoch": 11, "memory": 9465, "step": 121060} +{"lr": 0.0005050000000000001, "data_time": 0.0013393402099609376, "grad_norm": 1.5020484685897828, "loss": 0.6745238959789276, "time": 0.4165686845779419, "epoch": 11, "memory": 9465, "step": 121160} +{"lr": 0.0005050000000000001, "data_time": 0.001291036605834961, "grad_norm": 0.9124003171920776, "loss": 0.6776463925838471, "time": 0.3968178749084473, "epoch": 11, "memory": 9465, "step": 121260} +{"lr": 0.0005050000000000001, "data_time": 0.001394796371459961, "grad_norm": 3.354268269240856, "loss": 0.6727799534797668, "time": 0.38947367668151855, "epoch": 11, "memory": 9465, "step": 121360} +{"lr": 0.0005050000000000001, "data_time": 0.0014145612716674806, "grad_norm": 15.790675684064627, "loss": 0.6834679365158081, "time": 0.39763631820678713, "epoch": 11, "memory": 9465, "step": 121460} +{"lr": 0.0005050000000000001, "data_time": 0.001433110237121582, "grad_norm": 11.313840195536613, "loss": 0.673699015378952, "time": 0.4168685436248779, "epoch": 11, "memory": 9465, "step": 121560} +{"lr": 0.0005050000000000001, "data_time": 0.0013829469680786133, "grad_norm": 3.2813807114958764, "loss": 0.677393114566803, "time": 0.42999911308288574, "epoch": 11, "memory": 9465, "step": 121660} +{"lr": 0.0005050000000000001, "data_time": 0.0014870405197143556, "grad_norm": 1.0303614147007465, "loss": 0.682282030582428, "time": 0.4271831035614014, "epoch": 11, "memory": 9465, "step": 121760} +{"lr": 0.0005050000000000001, "data_time": 0.0013890981674194336, "grad_norm": 0.8496859695762395, "loss": 0.6810908675193786, "time": 0.40605978965759276, "epoch": 11, "memory": 9465, "step": 121860} +{"lr": 0.0005050000000000001, "data_time": 0.001494431495666504, "grad_norm": 0.5275362681597471, "loss": 0.6731388330459595, "time": 0.4266343593597412, "epoch": 11, "memory": 9465, "step": 121960} +{"lr": 0.0005050000000000001, "data_time": 0.001422286033630371, "grad_norm": 2.6346925273537636, "loss": 0.6795392811298371, "time": 0.4261953592300415, "epoch": 11, "memory": 9465, "step": 122060} +{"lr": 0.0005050000000000001, "data_time": 0.0016381978988647462, "grad_norm": 1.9452454678714275, "loss": 0.671459972858429, "time": 0.42636497020721437, "epoch": 11, "memory": 9465, "step": 122160} +{"lr": 0.0005050000000000001, "data_time": 0.0013174772262573241, "grad_norm": 8.538171362876891, "loss": 0.6888985455036163, "time": 0.4091381072998047, "epoch": 11, "memory": 9465, "step": 122260} +{"lr": 0.0005050000000000001, "data_time": 0.0012390851974487306, "grad_norm": 2.1021955788135527, "loss": 0.6825085937976837, "time": 0.40187487602233884, "epoch": 11, "memory": 9465, "step": 122360} +{"lr": 0.0005050000000000001, "data_time": 0.0013252019882202149, "grad_norm": 6.867604106664658, "loss": 0.6739128708839417, "time": 0.42259793281555175, "epoch": 11, "memory": 9465, "step": 122460} +{"lr": 0.0005050000000000001, "data_time": 0.0012115955352783204, "grad_norm": 83.43901827335358, "loss": 0.678394615650177, "time": 0.4365677833557129, "epoch": 11, "memory": 9465, "step": 122560} +{"lr": 0.0005050000000000001, "data_time": 0.0013599634170532227, "grad_norm": 2.20732289403677, "loss": 0.6636655867099762, "time": 0.3971044301986694, "epoch": 11, "memory": 9465, "step": 122660} +{"lr": 0.0005050000000000001, "data_time": 0.0013832330703735351, "grad_norm": 1.3128623604774474, "loss": 0.6777714014053344, "time": 0.40115063190460204, "epoch": 11, "memory": 9465, "step": 122760} +{"lr": 0.0005050000000000001, "data_time": 0.0013721704483032227, "grad_norm": 0.9810119360685349, "loss": 0.6685629665851593, "time": 0.3978790283203125, "epoch": 11, "memory": 9465, "step": 122860} +{"lr": 0.0005050000000000001, "data_time": 0.0012880802154541016, "grad_norm": 2.4078777730464935, "loss": 0.6691409051418304, "time": 0.45783884525299073, "epoch": 11, "memory": 9465, "step": 122960} +{"lr": 0.0005050000000000001, "data_time": 0.001217484474182129, "grad_norm": 22.93484022319317, "loss": 0.6794389367103577, "time": 0.43236308097839354, "epoch": 11, "memory": 9465, "step": 123060} +{"lr": 0.0005050000000000001, "data_time": 0.0015253782272338866, "grad_norm": 21.462710642814635, "loss": 0.6670138657093048, "time": 0.40058207511901855, "epoch": 11, "memory": 9465, "step": 123160} +{"lr": 0.0005050000000000001, "data_time": 0.0013308286666870116, "grad_norm": 7.648467391729355, "loss": 0.6769792556762695, "time": 0.40759427547454835, "epoch": 11, "memory": 9465, "step": 123260} +{"lr": 0.0005050000000000001, "data_time": 0.0013679981231689454, "grad_norm": 2.5652765333652496, "loss": 0.673036915063858, "time": 0.40096383094787597, "epoch": 11, "memory": 9465, "step": 123360} +{"lr": 0.0005050000000000001, "data_time": 0.0013195276260375977, "grad_norm": 4.90426222383976, "loss": 0.6876177847385406, "time": 0.41182947158813477, "epoch": 11, "memory": 9465, "step": 123460} +{"lr": 0.0005050000000000001, "data_time": 0.0013573884963989258, "grad_norm": 1.3246337741613388, "loss": 0.680418211221695, "time": 0.4129197120666504, "epoch": 11, "memory": 9465, "step": 123560} +{"lr": 0.0005050000000000001, "data_time": 0.0013721227645874024, "grad_norm": 2.6230124294757844, "loss": 0.6782394289970398, "time": 0.3892005443572998, "epoch": 11, "memory": 9465, "step": 123660} +{"lr": 0.0005050000000000001, "data_time": 0.0014016866683959962, "grad_norm": 6.475113850831986, "loss": 0.6695801615715027, "time": 0.4101029634475708, "epoch": 11, "memory": 9465, "step": 123760} +{"lr": 0.0005050000000000001, "data_time": 0.0013608694076538085, "grad_norm": 111.09181776046753, "loss": 0.6710107207298279, "time": 0.42042667865753175, "epoch": 11, "memory": 9465, "step": 123860} +{"lr": 0.0005050000000000001, "data_time": 0.0013858318328857423, "grad_norm": 43.57771167755127, "loss": 0.6811672568321228, "time": 0.394344162940979, "epoch": 11, "memory": 9465, "step": 123960} +{"lr": 0.0005050000000000001, "data_time": 0.0013201475143432618, "grad_norm": 183.56825037002562, "loss": 0.6773841500282287, "time": 0.46300063133239744, "epoch": 11, "memory": 9465, "step": 124060} +{"lr": 0.0005050000000000001, "data_time": 0.0012443780899047852, "grad_norm": 11.107847380638123, "loss": 0.679725068807602, "time": 0.4452333450317383, "epoch": 11, "memory": 9465, "step": 124160} +{"lr": 0.0005050000000000001, "data_time": 0.0012809514999389648, "grad_norm": 2.8669633507728576, "loss": 0.684642207622528, "time": 0.41928672790527344, "epoch": 11, "memory": 9465, "step": 124260} +{"lr": 0.0005050000000000001, "data_time": 0.0013806819915771484, "grad_norm": 11.678421294689178, "loss": 0.6807779788970947, "time": 0.4495292901992798, "epoch": 11, "memory": 9465, "step": 124360} +{"lr": 0.0005050000000000001, "data_time": 0.0012891769409179687, "grad_norm": 13.774215412139892, "loss": 0.6751906752586365, "time": 0.4242283821105957, "epoch": 11, "memory": 9465, "step": 124460} +{"lr": 0.0005050000000000001, "data_time": 0.00128326416015625, "grad_norm": 39.00203651189804, "loss": 0.6818271160125733, "time": 0.43977179527282717, "epoch": 11, "memory": 9465, "step": 124560} +{"lr": 0.0005050000000000001, "data_time": 0.0013326644897460938, "grad_norm": 26.933155679702757, "loss": 0.6730850815773011, "time": 0.42717363834381106, "epoch": 11, "memory": 9465, "step": 124660} +{"lr": 0.0005050000000000001, "data_time": 0.0012716054916381836, "grad_norm": 7.110695087909699, "loss": 0.6827796816825866, "time": 0.4461843252182007, "epoch": 11, "memory": 9465, "step": 124760} +{"lr": 0.0005050000000000001, "data_time": 0.0013957023620605469, "grad_norm": 116.47941331863403, "loss": 0.688453197479248, "time": 0.41263723373413086, "epoch": 11, "memory": 9465, "step": 124860} +{"lr": 0.0005050000000000001, "data_time": 0.0013117074966430664, "grad_norm": 36.254668053984645, "loss": 0.6842398405075073, "time": 0.4091363430023193, "epoch": 11, "memory": 9465, "step": 124960} +{"lr": 0.0005050000000000001, "data_time": 0.0012436628341674805, "grad_norm": 10.047633263468743, "loss": 0.6898526310920715, "time": 0.4380378246307373, "epoch": 11, "memory": 9465, "step": 125060} +{"lr": 0.0005050000000000001, "data_time": 0.001349020004272461, "grad_norm": 21.482556068897246, "loss": 0.687119847536087, "time": 0.4383219242095947, "epoch": 11, "memory": 9465, "step": 125160} +{"lr": 0.0005050000000000001, "data_time": 0.0012610197067260743, "grad_norm": 13.848260939121246, "loss": 0.6851519286632538, "time": 0.4230274438858032, "epoch": 11, "memory": 9465, "step": 125260} +{"lr": 0.0005050000000000001, "data_time": 0.0011785268783569337, "grad_norm": 3.980233770608902, "loss": 0.6885164201259613, "time": 0.4771219253540039, "epoch": 11, "memory": 9465, "step": 125360} +{"lr": 0.0005050000000000001, "data_time": 0.0012760639190673828, "grad_norm": 7.69204626083374, "loss": 0.6889068841934204, "time": 0.42093820571899415, "epoch": 11, "memory": 9465, "step": 125460} +{"lr": 0.0005050000000000001, "data_time": 0.001223134994506836, "grad_norm": 27.067822051048278, "loss": 0.6846552133560181, "time": 0.4102396726608276, "epoch": 11, "memory": 9465, "step": 125560} +{"lr": 0.0005050000000000001, "data_time": 0.0012709617614746094, "grad_norm": 3.8573687791824343, "loss": 0.6834540724754333, "time": 0.4050873279571533, "epoch": 11, "memory": 9465, "step": 125660} +{"lr": 0.0005050000000000001, "data_time": 0.0014838933944702148, "grad_norm": 22.484287905693055, "loss": 0.685315066576004, "time": 0.4652240753173828, "epoch": 11, "memory": 9465, "step": 125760} +{"lr": 0.0005050000000000001, "data_time": 0.0013320446014404297, "grad_norm": 68.12415418624877, "loss": 0.6865013837814331, "time": 0.6326909303665161, "epoch": 11, "memory": 9465, "step": 125860} +{"lr": 0.0005050000000000001, "data_time": 0.0013475656509399415, "grad_norm": 51.85662202835083, "loss": 0.6800556898117065, "time": 0.3989089488983154, "epoch": 11, "memory": 9465, "step": 125960} +{"lr": 0.0005050000000000001, "data_time": 0.0012872934341430664, "grad_norm": 22.468032431602477, "loss": 0.6815946578979493, "time": 0.43263342380523684, "epoch": 11, "memory": 9465, "step": 126060} +{"lr": 0.0005050000000000001, "data_time": 0.0012725591659545898, "grad_norm": 28.255167770385743, "loss": 0.6863328099250794, "time": 0.43357722759246825, "epoch": 11, "memory": 9465, "step": 126160} +{"lr": 0.0005050000000000001, "data_time": 0.0012325525283813476, "grad_norm": 69.66912631988525, "loss": 0.6869555890560151, "time": 0.404861307144165, "epoch": 11, "memory": 9465, "step": 126260} +{"lr": 0.0005050000000000001, "data_time": 0.0012387514114379882, "grad_norm": 31.074062204360963, "loss": 0.6884749054908752, "time": 0.4018389701843262, "epoch": 11, "memory": 9465, "step": 126360} +{"lr": 0.0005050000000000001, "data_time": 0.001231551170349121, "grad_norm": 167.52872245311738, "loss": 0.6871707201004028, "time": 0.4456325054168701, "epoch": 11, "memory": 9465, "step": 126460} +{"lr": 0.0005050000000000001, "data_time": 0.0012233257293701172, "grad_norm": 29.962598490715028, "loss": 0.6872786402702331, "time": 0.4013463258743286, "epoch": 11, "memory": 9465, "step": 126560} +{"lr": 0.0005050000000000001, "data_time": 0.007564949989318848, "grad_norm": 8.457032573223113, "loss": 0.6900364756584167, "time": 0.413921594619751, "epoch": 11, "memory": 9465, "step": 126660} +{"accuracy/top1": 100.0, "data_time": 0.31066016589894013, "time": 0.570034013074987, "step": 11} +{"lr": 0.000427564939805086, "data_time": 0.076735520362854, "grad_norm": 4.348672594130039, "loss": 0.6898119688034058, "time": 0.4361224412918091, "epoch": 12, "memory": 9465, "step": 126776} +{"lr": 0.000427564939805086, "data_time": 0.0012937068939208984, "grad_norm": 13.425345730781554, "loss": 0.6842809200286866, "time": 0.41864862442016604, "epoch": 12, "memory": 9465, "step": 126876} +{"lr": 0.000427564939805086, "data_time": 0.0013746976852416991, "grad_norm": 9.847851753234863, "loss": 0.6790269792079926, "time": 0.40267632007598875, "epoch": 12, "memory": 9465, "step": 126976} +{"lr": 0.000427564939805086, "data_time": 0.0011780977249145508, "grad_norm": 18.218738269805907, "loss": 0.683313512802124, "time": 0.4999894142150879, "epoch": 12, "memory": 9465, "step": 127076} +{"lr": 0.000427564939805086, "data_time": 0.0012818098068237305, "grad_norm": 13.688666260242462, "loss": 0.6830900192260743, "time": 0.41017515659332277, "epoch": 12, "memory": 9465, "step": 127176} +{"lr": 0.000427564939805086, "data_time": 0.0013007402420043945, "grad_norm": 74.42742584943771, "loss": 0.6809566378593445, "time": 0.4042999267578125, "epoch": 12, "memory": 9465, "step": 127276} +{"lr": 0.000427564939805086, "data_time": 0.0012687444686889648, "grad_norm": 188.30246211886407, "loss": 0.6753818213939666, "time": 0.4117432594299316, "epoch": 12, "memory": 9465, "step": 127376} +{"lr": 0.000427564939805086, "data_time": 0.0013460874557495116, "grad_norm": 11.578718554973602, "loss": 0.6836562275886535, "time": 0.4134324550628662, "epoch": 12, "memory": 9465, "step": 127476} +{"lr": 0.000427564939805086, "data_time": 0.0013631820678710938, "grad_norm": 1.4723381996154785, "loss": 0.6792226493358612, "time": 0.42078337669372556, "epoch": 12, "memory": 9465, "step": 127576} +{"lr": 0.000427564939805086, "data_time": 0.0012538909912109375, "grad_norm": 3.5019516080617903, "loss": 0.6809235095977784, "time": 0.43942258358001707, "epoch": 12, "memory": 9465, "step": 127676} +{"lr": 0.000427564939805086, "data_time": 0.0015204429626464843, "grad_norm": 10.551207971572875, "loss": 0.6818658530712127, "time": 0.46178693771362306, "epoch": 12, "memory": 9465, "step": 127776} +{"lr": 0.000427564939805086, "data_time": 0.001210331916809082, "grad_norm": 20.945073568820952, "loss": 0.6804120361804962, "time": 0.4522751808166504, "epoch": 12, "memory": 9465, "step": 127876} +{"lr": 0.000427564939805086, "data_time": 0.0012592315673828126, "grad_norm": 14.179097974300385, "loss": 0.6758145213127136, "time": 0.421085786819458, "epoch": 12, "memory": 9465, "step": 127976} +{"lr": 0.000427564939805086, "data_time": 0.0012780189514160155, "grad_norm": 3.3984122291207313, "loss": 0.6759363532066345, "time": 0.4091540813446045, "epoch": 12, "memory": 9465, "step": 128076} +{"lr": 0.000427564939805086, "data_time": 0.0012897968292236329, "grad_norm": 0.837803053855896, "loss": 0.6752262115478516, "time": 0.40413990020751955, "epoch": 12, "memory": 9465, "step": 128176} +{"lr": 0.000427564939805086, "data_time": 0.0013520956039428712, "grad_norm": 2.931898007541895, "loss": 0.6732583999633789, "time": 0.4167243242263794, "epoch": 12, "memory": 9465, "step": 128276} +{"lr": 0.000427564939805086, "data_time": 0.001275491714477539, "grad_norm": 3.4713230073451995, "loss": 0.6816255033016205, "time": 0.42007038593292234, "epoch": 12, "memory": 9465, "step": 128376} +{"lr": 0.000427564939805086, "data_time": 0.0012035369873046875, "grad_norm": 196.71823449134826, "loss": 0.6683463215827942, "time": 0.42772114276885986, "epoch": 12, "memory": 9465, "step": 128476} +{"lr": 0.000427564939805086, "data_time": 0.0013960838317871095, "grad_norm": 15.31438370347023, "loss": 0.6783772885799408, "time": 0.5493594169616699, "epoch": 12, "memory": 9465, "step": 128576} +{"lr": 0.000427564939805086, "data_time": 0.001378178596496582, "grad_norm": 1.7639539506286384, "loss": 0.6709182560443878, "time": 0.3983979940414429, "epoch": 12, "memory": 9465, "step": 128676} +{"lr": 0.000427564939805086, "data_time": 0.0012935400009155273, "grad_norm": 3.431198874115944, "loss": 0.6762523114681244, "time": 0.4447017669677734, "epoch": 12, "memory": 9465, "step": 128776} +{"lr": 0.000427564939805086, "data_time": 0.001427602767944336, "grad_norm": 2.770304651558399, "loss": 0.6715142607688904, "time": 0.4263080358505249, "epoch": 12, "memory": 9465, "step": 128876} +{"lr": 0.000427564939805086, "data_time": 0.0014818668365478515, "grad_norm": 3.470287262648344, "loss": 0.6753794312477112, "time": 0.41139366626739504, "epoch": 12, "memory": 9465, "step": 128976} +{"lr": 0.000427564939805086, "data_time": 0.0015148162841796876, "grad_norm": 1.2597437039017678, "loss": 0.6691596806049347, "time": 0.40613598823547364, "epoch": 12, "memory": 9465, "step": 129076} +{"lr": 0.000427564939805086, "data_time": 0.001400899887084961, "grad_norm": 2.1850260317325594, "loss": 0.6866812586784363, "time": 0.42925822734832764, "epoch": 12, "memory": 9465, "step": 129176} +{"lr": 0.000427564939805086, "data_time": 0.0013475418090820312, "grad_norm": 0.8963435543701053, "loss": 0.6871735692024231, "time": 0.4090800523757935, "epoch": 12, "memory": 9465, "step": 129276} +{"lr": 0.000427564939805086, "data_time": 0.001323866844177246, "grad_norm": 2.536113853752613, "loss": 0.6818703353404999, "time": 0.41405088901519777, "epoch": 12, "memory": 9465, "step": 129376} +{"lr": 0.000427564939805086, "data_time": 0.0014016866683959962, "grad_norm": 0.030517227482050657, "loss": 0.6869018912315369, "time": 0.44098801612854005, "epoch": 12, "memory": 9465, "step": 129476} +{"lr": 0.000427564939805086, "data_time": 0.0012927532196044921, "grad_norm": 0.036532994173467156, "loss": 0.6903063833713532, "time": 0.41088435649871824, "epoch": 12, "memory": 9465, "step": 129576} +{"lr": 0.000427564939805086, "data_time": 0.0013560056686401367, "grad_norm": 0.0251450982876122, "loss": 0.6881662428379058, "time": 0.4688152074813843, "epoch": 12, "memory": 9465, "step": 129676} +{"lr": 0.000427564939805086, "data_time": 0.0014296770095825195, "grad_norm": 0.01814848983194679, "loss": 0.6847158551216126, "time": 0.43398659229278563, "epoch": 12, "memory": 9465, "step": 129776} +{"lr": 0.000427564939805086, "data_time": 0.0012911319732666015, "grad_norm": 0.03640967914834618, "loss": 0.6900094389915467, "time": 0.4438145637512207, "epoch": 12, "memory": 9465, "step": 129876} +{"lr": 0.000427564939805086, "data_time": 0.0015025138854980469, "grad_norm": 0.023893103143200277, "loss": 0.6896107614040374, "time": 0.4201754808425903, "epoch": 12, "memory": 9465, "step": 129976} +{"lr": 0.000427564939805086, "data_time": 0.0013779640197753907, "grad_norm": 0.020082084462046623, "loss": 0.6873001635074616, "time": 0.4413077116012573, "epoch": 12, "memory": 9465, "step": 130076} +{"lr": 0.000427564939805086, "data_time": 0.0012671947479248047, "grad_norm": 0.021519444510340692, "loss": 0.687029379606247, "time": 0.4054450750350952, "epoch": 12, "memory": 9465, "step": 130176} +{"lr": 0.000427564939805086, "data_time": 0.0014216184616088867, "grad_norm": 0.05771431168541312, "loss": 0.6854559421539307, "time": 0.41077234745025637, "epoch": 12, "memory": 9465, "step": 130276} +{"lr": 0.000427564939805086, "data_time": 0.0012619733810424806, "grad_norm": 0.022173736547119915, "loss": 0.6871199786663056, "time": 0.4157752752304077, "epoch": 12, "memory": 9465, "step": 130376} +{"lr": 0.000427564939805086, "data_time": 0.0013138771057128907, "grad_norm": 0.036936514638364316, "loss": 0.6890743017196655, "time": 0.45462851524353026, "epoch": 12, "memory": 9465, "step": 130476} +{"lr": 0.000427564939805086, "data_time": 0.0013151168823242188, "grad_norm": 0.027146380953490735, "loss": 0.6815440595149994, "time": 0.4137685775756836, "epoch": 12, "memory": 9465, "step": 130576} +{"lr": 0.000427564939805086, "data_time": 0.0012878179550170898, "grad_norm": 0.027129650907590987, "loss": 0.684997022151947, "time": 0.3863675594329834, "epoch": 12, "memory": 9465, "step": 130676} +{"lr": 0.000427564939805086, "data_time": 0.0016545772552490235, "grad_norm": 0.03196599064394832, "loss": 0.6857732832431793, "time": 0.4002965450286865, "epoch": 12, "memory": 9465, "step": 130776} +{"lr": 0.000427564939805086, "data_time": 0.0015331506729125977, "grad_norm": 0.025930631812661885, "loss": 0.6855455875396729, "time": 0.452677583694458, "epoch": 12, "memory": 9465, "step": 130876} +{"lr": 0.000427564939805086, "data_time": 0.001256251335144043, "grad_norm": 0.02131669588852674, "loss": 0.6853181421756744, "time": 0.44700634479522705, "epoch": 12, "memory": 9465, "step": 130976} +{"lr": 0.000427564939805086, "data_time": 0.001496577262878418, "grad_norm": 0.021841109939850868, "loss": 0.6890956461429596, "time": 0.44107742309570314, "epoch": 12, "memory": 9465, "step": 131076} +{"lr": 0.000427564939805086, "data_time": 0.0015111923217773437, "grad_norm": 0.01930496752029285, "loss": 0.6871892094612122, "time": 0.40665972232818604, "epoch": 12, "memory": 9465, "step": 131176} +{"lr": 0.000427564939805086, "data_time": 0.0014603376388549806, "grad_norm": 0.02447275447193533, "loss": 0.6822593510150909, "time": 0.4118276834487915, "epoch": 12, "memory": 9465, "step": 131276} +{"lr": 0.000427564939805086, "data_time": 0.0013532400131225585, "grad_norm": 0.02468832125887275, "loss": 0.6912485063076019, "time": 0.37818121910095215, "epoch": 12, "memory": 9465, "step": 131376} +{"lr": 0.000427564939805086, "data_time": 0.0013502836227416992, "grad_norm": 0.023085615155287086, "loss": 0.6889755249023437, "time": 0.4210802555084229, "epoch": 12, "memory": 9465, "step": 131476} +{"lr": 0.000427564939805086, "data_time": 0.0013721466064453125, "grad_norm": 0.019682963425293565, "loss": 0.6871603846549987, "time": 0.44051265716552734, "epoch": 12, "memory": 9465, "step": 131576} +{"lr": 0.000427564939805086, "data_time": 0.0013153076171875, "grad_norm": 0.022028203285299242, "loss": 0.6892701923847199, "time": 0.4130614995956421, "epoch": 12, "memory": 9465, "step": 131676} +{"lr": 0.000427564939805086, "data_time": 0.001412653923034668, "grad_norm": 0.017201270908117294, "loss": 0.6846320748329162, "time": 0.45083136558532716, "epoch": 12, "memory": 9465, "step": 131776} +{"lr": 0.000427564939805086, "data_time": 0.0012716531753540039, "grad_norm": 0.018730456568300725, "loss": 0.6891506969928741, "time": 0.4022256851196289, "epoch": 12, "memory": 9465, "step": 131876} +{"lr": 0.000427564939805086, "data_time": 0.001374363899230957, "grad_norm": 0.025354755041189492, "loss": 0.6907829225063324, "time": 0.3865982532501221, "epoch": 12, "memory": 9465, "step": 131976} +{"lr": 0.000427564939805086, "data_time": 0.0013847589492797852, "grad_norm": 0.01840975587256253, "loss": 0.6831769108772278, "time": 0.42206737995147703, "epoch": 12, "memory": 9465, "step": 132076} +{"lr": 0.000427564939805086, "data_time": 0.0016530275344848633, "grad_norm": 0.019254678534343837, "loss": 0.689174884557724, "time": 0.41932222843170164, "epoch": 12, "memory": 9465, "step": 132176} +{"lr": 0.000427564939805086, "data_time": 0.001480698585510254, "grad_norm": 0.021869329328183085, "loss": 0.6880807399749755, "time": 0.4040341854095459, "epoch": 12, "memory": 9465, "step": 132276} +{"lr": 0.000427564939805086, "data_time": 0.0014929771423339844, "grad_norm": 0.020475224265828727, "loss": 0.6849862813949585, "time": 0.41760499477386476, "epoch": 12, "memory": 9465, "step": 132376} +{"lr": 0.000427564939805086, "data_time": 0.0012695789337158203, "grad_norm": 0.01955119805643335, "loss": 0.6870149850845337, "time": 0.41249804496765136, "epoch": 12, "memory": 9465, "step": 132476} +{"lr": 0.000427564939805086, "data_time": 0.0013772010803222655, "grad_norm": 0.019221801293315365, "loss": 0.6884264945983887, "time": 0.44633028507232664, "epoch": 12, "memory": 9465, "step": 132576} +{"lr": 0.000427564939805086, "data_time": 0.0013963699340820313, "grad_norm": 0.02509285109117627, "loss": 0.6884343028068542, "time": 0.4383856296539307, "epoch": 12, "memory": 9465, "step": 132676} +{"lr": 0.000427564939805086, "data_time": 0.0011419057846069336, "grad_norm": 0.01643194416537881, "loss": 0.6855869948863983, "time": 0.42352981567382814, "epoch": 12, "memory": 9465, "step": 132776} +{"lr": 0.000427564939805086, "data_time": 0.0014334678649902343, "grad_norm": 0.02347443662583828, "loss": 0.6875027775764465, "time": 0.40734381675720216, "epoch": 12, "memory": 9465, "step": 132876} +{"lr": 0.000427564939805086, "data_time": 0.0013040781021118164, "grad_norm": 0.02326798252761364, "loss": 0.6820044338703155, "time": 0.4270260572433472, "epoch": 12, "memory": 9465, "step": 132976} +{"lr": 0.000427564939805086, "data_time": 0.0012149810791015625, "grad_norm": 0.024808078981004657, "loss": 0.6920408725738525, "time": 0.4212906837463379, "epoch": 12, "memory": 9465, "step": 133076} +{"lr": 0.000427564939805086, "data_time": 0.001304292678833008, "grad_norm": 0.03381731836125255, "loss": 0.6895127177238465, "time": 0.42964370250701905, "epoch": 12, "memory": 9465, "step": 133176} +{"lr": 0.000427564939805086, "data_time": 0.001300954818725586, "grad_norm": 0.019100240874104203, "loss": 0.6905601620674133, "time": 0.4016817569732666, "epoch": 12, "memory": 9465, "step": 133276} +{"lr": 0.000427564939805086, "data_time": 0.001690506935119629, "grad_norm": 0.019882703525945546, "loss": 0.6893523037433624, "time": 0.4288515567779541, "epoch": 12, "memory": 9465, "step": 133376} +{"lr": 0.000427564939805086, "data_time": 0.0013581037521362305, "grad_norm": 0.024326645722612738, "loss": 0.6894127905368805, "time": 0.4464055299758911, "epoch": 12, "memory": 9465, "step": 133476} +{"lr": 0.000427564939805086, "data_time": 0.0013773202896118163, "grad_norm": 0.013203220115974546, "loss": 0.6842819094657898, "time": 0.45641722679138186, "epoch": 12, "memory": 9465, "step": 133576} +{"lr": 0.000427564939805086, "data_time": 0.00134124755859375, "grad_norm": 0.022629460087046027, "loss": 0.6804419755935669, "time": 0.42402608394622804, "epoch": 12, "memory": 9465, "step": 133676} +{"lr": 0.000427564939805086, "data_time": 0.0013596773147583007, "grad_norm": 0.022043393761850893, "loss": 0.685667097568512, "time": 0.40441606044769285, "epoch": 12, "memory": 9465, "step": 133776} +{"lr": 0.000427564939805086, "data_time": 0.0015870332717895508, "grad_norm": 0.018440839857794343, "loss": 0.686989140510559, "time": 0.41110920906066895, "epoch": 12, "memory": 9465, "step": 133876} +{"lr": 0.000427564939805086, "data_time": 0.0013441801071166991, "grad_norm": 0.030286093521863223, "loss": 0.6861549079418182, "time": 0.40697195529937746, "epoch": 12, "memory": 9465, "step": 133976} +{"lr": 0.000427564939805086, "data_time": 0.001190185546875, "grad_norm": 0.024573275772854685, "loss": 0.6904351890087128, "time": 0.3955228328704834, "epoch": 12, "memory": 9465, "step": 134076} +{"lr": 0.000427564939805086, "data_time": 0.001268482208251953, "grad_norm": 0.02895478003192693, "loss": 0.6836984992027283, "time": 0.40450875759124755, "epoch": 12, "memory": 9465, "step": 134176} +{"lr": 0.000427564939805086, "data_time": 0.0013086318969726563, "grad_norm": 0.020092893857508898, "loss": 0.6867953479290009, "time": 0.4229351758956909, "epoch": 12, "memory": 9465, "step": 134276} +{"lr": 0.000427564939805086, "data_time": 0.0014414548873901366, "grad_norm": 0.02477827437687665, "loss": 0.6843395233154297, "time": 0.4216409921646118, "epoch": 12, "memory": 9465, "step": 134376} +{"lr": 0.000427564939805086, "data_time": 0.0013153553009033203, "grad_norm": 0.01508900262415409, "loss": 0.68714160323143, "time": 0.4240150213241577, "epoch": 12, "memory": 9465, "step": 134476} +{"lr": 0.000427564939805086, "data_time": 0.001452493667602539, "grad_norm": 0.02276517413556576, "loss": 0.6918884932994842, "time": 0.4838341951370239, "epoch": 12, "memory": 9465, "step": 134576} +{"lr": 0.000427564939805086, "data_time": 0.0012554645538330079, "grad_norm": 0.029659863817505537, "loss": 0.6815708875656128, "time": 0.433766770362854, "epoch": 12, "memory": 9465, "step": 134676} +{"lr": 0.000427564939805086, "data_time": 0.0012537240982055664, "grad_norm": 0.019865823443979025, "loss": 0.6860040843486785, "time": 0.40606448650360105, "epoch": 12, "memory": 9465, "step": 134776} +{"lr": 0.000427564939805086, "data_time": 0.0012784957885742187, "grad_norm": 0.019153623550664634, "loss": 0.6851597964763642, "time": 0.4024749040603638, "epoch": 12, "memory": 9465, "step": 134876} +{"lr": 0.000427564939805086, "data_time": 0.001424694061279297, "grad_norm": 0.02400211044587195, "loss": 0.6818427979946137, "time": 0.4202421188354492, "epoch": 12, "memory": 9465, "step": 134976} +{"lr": 0.000427564939805086, "data_time": 0.0014413118362426758, "grad_norm": 0.028626030031591653, "loss": 0.6868133068084716, "time": 0.4084906816482544, "epoch": 12, "memory": 9465, "step": 135076} +{"lr": 0.000427564939805086, "data_time": 0.001470041275024414, "grad_norm": 0.01666400389513001, "loss": 0.6866598844528198, "time": 0.40768468379974365, "epoch": 12, "memory": 9465, "step": 135176} +{"lr": 0.000427564939805086, "data_time": 0.0014740705490112304, "grad_norm": 0.015819929726421832, "loss": 0.6859961926937104, "time": 0.4268024444580078, "epoch": 12, "memory": 9465, "step": 135276} +{"lr": 0.000427564939805086, "data_time": 0.0012934446334838868, "grad_norm": 0.02918328596279025, "loss": 0.6861585319042206, "time": 0.40381383895874023, "epoch": 12, "memory": 9465, "step": 135376} +{"lr": 0.000427564939805086, "data_time": 0.0014419317245483398, "grad_norm": 0.020316384651232512, "loss": 0.6876485645771027, "time": 0.41089928150177, "epoch": 12, "memory": 9465, "step": 135476} +{"lr": 0.000427564939805086, "data_time": 0.0013978719711303712, "grad_norm": 0.023399069788865745, "loss": 0.6904264211654663, "time": 0.4105861186981201, "epoch": 12, "memory": 9465, "step": 135576} +{"lr": 0.000427564939805086, "data_time": 0.0014662027359008789, "grad_norm": 0.018001660704612732, "loss": 0.6850806176662445, "time": 0.4113790512084961, "epoch": 12, "memory": 9465, "step": 135676} +{"lr": 0.000427564939805086, "data_time": 0.001482105255126953, "grad_norm": 0.0262545530218631, "loss": 0.6874183356761933, "time": 0.42870097160339354, "epoch": 12, "memory": 9465, "step": 135776} +{"lr": 0.000427564939805086, "data_time": 0.001321554183959961, "grad_norm": 0.01224442480597645, "loss": 0.6862155795097351, "time": 0.4189307689666748, "epoch": 12, "memory": 9465, "step": 135876} +{"lr": 0.000427564939805086, "data_time": 0.0011930227279663085, "grad_norm": 0.026414019777439533, "loss": 0.6858326196670532, "time": 0.4315566301345825, "epoch": 12, "memory": 9465, "step": 135976} +{"lr": 0.000427564939805086, "data_time": 0.001241016387939453, "grad_norm": 0.024612182425335048, "loss": 0.6861414194107056, "time": 0.43127634525299074, "epoch": 12, "memory": 9465, "step": 136076} +{"lr": 0.000427564939805086, "data_time": 0.0014866113662719727, "grad_norm": 0.018336525186896324, "loss": 0.6848597586154938, "time": 0.4810450792312622, "epoch": 12, "memory": 9465, "step": 136176} +{"lr": 0.000427564939805086, "data_time": 0.0012953758239746093, "grad_norm": 0.019203244987875224, "loss": 0.690115088224411, "time": 0.4266946792602539, "epoch": 12, "memory": 9465, "step": 136276} +{"lr": 0.000427564939805086, "data_time": 0.0016733407974243164, "grad_norm": 0.0174126991070807, "loss": 0.69014573097229, "time": 0.4433808088302612, "epoch": 12, "memory": 9465, "step": 136376} +{"lr": 0.000427564939805086, "data_time": 0.0014173269271850586, "grad_norm": 0.020531594473868608, "loss": 0.6884080231189728, "time": 0.44682042598724364, "epoch": 12, "memory": 9465, "step": 136476} +{"lr": 0.000427564939805086, "data_time": 0.00130157470703125, "grad_norm": 0.018273185286670924, "loss": 0.6853847205638885, "time": 0.39111766815185545, "epoch": 12, "memory": 9465, "step": 136576} +{"lr": 0.000427564939805086, "data_time": 0.0013633966445922852, "grad_norm": 0.012315124575980007, "loss": 0.6867692649364472, "time": 0.4591775894165039, "epoch": 12, "memory": 9465, "step": 136676} +{"lr": 0.000427564939805086, "data_time": 0.001344442367553711, "grad_norm": 0.021995778544805944, "loss": 0.6875159919261933, "time": 0.4549398899078369, "epoch": 12, "memory": 9465, "step": 136776} +{"lr": 0.000427564939805086, "data_time": 0.001265406608581543, "grad_norm": 0.020000772573985158, "loss": 0.6919745922088623, "time": 0.45438556671142577, "epoch": 12, "memory": 9465, "step": 136876} +{"lr": 0.000427564939805086, "data_time": 0.001387333869934082, "grad_norm": 0.022508371085859836, "loss": 0.6888543665409088, "time": 0.43606531620025635, "epoch": 12, "memory": 9465, "step": 136976} +{"lr": 0.000427564939805086, "data_time": 0.0012754440307617188, "grad_norm": 0.024935787287540735, "loss": 0.6890964150428772, "time": 0.41189501285552976, "epoch": 12, "memory": 9465, "step": 137076} +{"lr": 0.000427564939805086, "data_time": 0.0020142078399658205, "grad_norm": 0.020476224087178706, "loss": 0.6907122969627381, "time": 0.3999741792678833, "epoch": 12, "memory": 9465, "step": 137176} +{"lr": 0.000427564939805086, "data_time": 0.0014447450637817382, "grad_norm": 0.02158180233091116, "loss": 0.6905366480350494, "time": 0.43572728633880614, "epoch": 12, "memory": 9465, "step": 137276} +{"lr": 0.000427564939805086, "data_time": 0.001335906982421875, "grad_norm": 0.022731278091669083, "loss": 0.6856934368610382, "time": 0.43190479278564453, "epoch": 12, "memory": 9465, "step": 137376} +{"lr": 0.000427564939805086, "data_time": 0.0013525724411010743, "grad_norm": 0.025684100296348333, "loss": 0.6849841117858887, "time": 0.43099782466888426, "epoch": 12, "memory": 9465, "step": 137476} +{"lr": 0.000427564939805086, "data_time": 0.0012893199920654297, "grad_norm": 0.026394939993042498, "loss": 0.6828485250473022, "time": 0.43994388580322263, "epoch": 12, "memory": 9465, "step": 137576} +{"lr": 0.000427564939805086, "data_time": 0.001317000389099121, "grad_norm": 0.020767204923322423, "loss": 0.6891070485115052, "time": 0.41753444671630857, "epoch": 12, "memory": 9465, "step": 137676} +{"lr": 0.000427564939805086, "data_time": 0.0013175249099731446, "grad_norm": 0.02978737003868446, "loss": 0.6874478399753571, "time": 0.41090338230133056, "epoch": 12, "memory": 9465, "step": 137776} +{"lr": 0.000427564939805086, "data_time": 0.0013688087463378906, "grad_norm": 0.020555397635325788, "loss": 0.6828809797763824, "time": 0.4285306215286255, "epoch": 12, "memory": 9465, "step": 137876} +{"lr": 0.000427564939805086, "data_time": 0.0012490272521972657, "grad_norm": 0.03204272212460637, "loss": 0.6877374231815339, "time": 0.4435636758804321, "epoch": 12, "memory": 9465, "step": 137976} +{"lr": 0.000427564939805086, "data_time": 0.0023162841796875, "grad_norm": 0.037516922818031165, "loss": 0.6898782253265381, "time": 0.3867438077926636, "epoch": 12, "memory": 9465, "step": 138076} +{"lr": 0.000427564939805086, "data_time": 0.00787358283996582, "grad_norm": 0.02366722277365625, "loss": 0.6853773534297943, "time": 0.4042431116104126, "epoch": 12, "memory": 9465, "step": 138176} +{"accuracy/top1": 100.0, "data_time": 0.29801522984224205, "time": 0.5517384725458482, "step": 12} +{"lr": 0.0003520365877844012, "data_time": 0.1872790575027466, "grad_norm": 0.017510948225390167, "loss": 0.6881148517131805, "time": 0.468621039390564, "epoch": 13, "memory": 9465, "step": 138292} +{"lr": 0.0003520365877844012, "data_time": 0.0012444019317626952, "grad_norm": 0.026656911056488753, "loss": 0.6868264317512512, "time": 0.4445474863052368, "epoch": 13, "memory": 9465, "step": 138392} +{"lr": 0.0003520365877844012, "data_time": 0.0011973381042480469, "grad_norm": 0.027554817497730255, "loss": 0.6890848696231842, "time": 0.4294422626495361, "epoch": 13, "memory": 9465, "step": 138492} +{"lr": 0.0003520365877844012, "data_time": 0.0013140439987182617, "grad_norm": 0.03655003334861249, "loss": 0.6861842572689056, "time": 0.43264293670654297, "epoch": 13, "memory": 9465, "step": 138592} +{"lr": 0.0003520365877844012, "data_time": 0.0012440443038940429, "grad_norm": 0.02566453628242016, "loss": 0.6926171720027924, "time": 0.45820322036743166, "epoch": 13, "memory": 9465, "step": 138692} +{"lr": 0.0003520365877844012, "data_time": 0.00138397216796875, "grad_norm": 0.022582877962850034, "loss": 0.6888538777828217, "time": 0.4243925094604492, "epoch": 13, "memory": 9465, "step": 138792} +{"lr": 0.0003520365877844012, "data_time": 0.0013495683670043945, "grad_norm": 0.021802978264167903, "loss": 0.6897751808166503, "time": 0.40727231502532957, "epoch": 13, "memory": 9465, "step": 138892} +{"lr": 0.0003520365877844012, "data_time": 0.0013855695724487305, "grad_norm": 0.013806198944803328, "loss": 0.6899027168750763, "time": 0.41996567249298095, "epoch": 13, "memory": 9465, "step": 138992} +{"lr": 0.0003520365877844012, "data_time": 0.0014684677124023437, "grad_norm": 0.01774993594735861, "loss": 0.686185497045517, "time": 0.43367202281951905, "epoch": 13, "memory": 9465, "step": 139092} +{"lr": 0.0003520365877844012, "data_time": 0.0012041568756103516, "grad_norm": 0.023941261286381632, "loss": 0.6860288798809051, "time": 0.41524133682250974, "epoch": 13, "memory": 9465, "step": 139192} +{"lr": 0.0003520365877844012, "data_time": 0.0012471675872802734, "grad_norm": 0.018963639996945857, "loss": 0.6848430573940277, "time": 0.41869263648986815, "epoch": 13, "memory": 9465, "step": 139292} +{"lr": 0.0003520365877844012, "data_time": 0.0013098001480102539, "grad_norm": 0.016764594428241252, "loss": 0.6842615306377411, "time": 0.44188990592956545, "epoch": 13, "memory": 9465, "step": 139392} +{"lr": 0.0003520365877844012, "data_time": 0.0012956619262695312, "grad_norm": 0.0158146571891848, "loss": 0.6855414569377899, "time": 0.45032515525817873, "epoch": 13, "memory": 9465, "step": 139492} +{"lr": 0.0003520365877844012, "data_time": 0.0013699293136596679, "grad_norm": 0.030753080965951085, "loss": 0.6857570528984069, "time": 0.38656091690063477, "epoch": 13, "memory": 9465, "step": 139592} +{"lr": 0.0003520365877844012, "data_time": 0.0014206171035766602, "grad_norm": 0.013240466848947108, "loss": 0.686662209033966, "time": 0.44356157779693606, "epoch": 13, "memory": 9465, "step": 139692} +{"lr": 0.0003520365877844012, "data_time": 0.0012423038482666016, "grad_norm": 0.016023906273767353, "loss": 0.6891343235969544, "time": 0.4172191143035889, "epoch": 13, "memory": 9465, "step": 139792} +{"lr": 0.0003520365877844012, "data_time": 0.0013096332550048828, "grad_norm": 0.022254918655380606, "loss": 0.688138771057129, "time": 0.424369215965271, "epoch": 13, "memory": 9465, "step": 139892} +{"lr": 0.0003520365877844012, "data_time": 0.0012717962265014649, "grad_norm": 0.032299743383191526, "loss": 0.6855127274990082, "time": 0.4386639356613159, "epoch": 13, "memory": 9465, "step": 139992} +{"lr": 0.0003520365877844012, "data_time": 0.0013203382492065429, "grad_norm": 0.02728343503549695, "loss": 0.6826135277748108, "time": 0.42070319652557375, "epoch": 13, "memory": 9465, "step": 140092} +{"lr": 0.0003520365877844012, "data_time": 0.001439213752746582, "grad_norm": 0.022068626224063338, "loss": 0.6877608120441436, "time": 0.4150163412094116, "epoch": 13, "memory": 9465, "step": 140192} +{"lr": 0.0003520365877844012, "data_time": 0.0013849496841430663, "grad_norm": 0.029572939220815896, "loss": 0.683891487121582, "time": 0.39600937366485595, "epoch": 13, "memory": 9465, "step": 140292} +{"lr": 0.0003520365877844012, "data_time": 0.0013171911239624023, "grad_norm": 0.026504780864343047, "loss": 0.6791876435279847, "time": 0.4116565704345703, "epoch": 13, "memory": 9465, "step": 140392} +{"lr": 0.0003520365877844012, "data_time": 0.0014576911926269531, "grad_norm": 0.02092731527518481, "loss": 0.686355185508728, "time": 0.43780813217163084, "epoch": 13, "memory": 9465, "step": 140492} +{"lr": 0.0003520365877844012, "data_time": 0.0013976812362670899, "grad_norm": 0.034247031714767215, "loss": 0.6849726617336274, "time": 0.43756694793701173, "epoch": 13, "memory": 9465, "step": 140592} +{"lr": 0.0003520365877844012, "data_time": 0.0013062715530395507, "grad_norm": 0.018480710778385402, "loss": 0.6864704310894012, "time": 0.40336544513702394, "epoch": 13, "memory": 9465, "step": 140692} +{"lr": 0.0003520365877844012, "data_time": 0.001402902603149414, "grad_norm": 0.023231533565558495, "loss": 0.6867999255657196, "time": 0.42250733375549315, "epoch": 13, "memory": 9465, "step": 140792} +{"lr": 0.0003520365877844012, "data_time": 0.0013544321060180663, "grad_norm": 0.019997408241033555, "loss": 0.6861769199371338, "time": 0.43517193794250486, "epoch": 13, "memory": 9465, "step": 140892} +{"lr": 0.0003520365877844012, "data_time": 0.0013731718063354492, "grad_norm": 0.022900594654493032, "loss": 0.6920109510421752, "time": 0.43185617923736574, "epoch": 13, "memory": 9465, "step": 140992} +{"lr": 0.0003520365877844012, "data_time": 0.0015815258026123046, "grad_norm": 0.024012094689533116, "loss": 0.6870336174964905, "time": 0.4229752063751221, "epoch": 13, "memory": 9465, "step": 141092} +{"lr": 0.0003520365877844012, "data_time": 0.0013350725173950195, "grad_norm": 0.02456324272789061, "loss": 0.6892096519470214, "time": 0.4228933334350586, "epoch": 13, "memory": 9465, "step": 141192} +{"lr": 0.0003520365877844012, "data_time": 0.0013721227645874024, "grad_norm": 0.023345749638974667, "loss": 0.6876180946826935, "time": 0.43931851387023924, "epoch": 13, "memory": 9465, "step": 141292} +{"lr": 0.0003520365877844012, "data_time": 0.001420307159423828, "grad_norm": 0.012605708825867623, "loss": 0.6855099916458129, "time": 0.42755842208862305, "epoch": 13, "memory": 9465, "step": 141392} +{"lr": 0.0003520365877844012, "data_time": 0.0015364885330200195, "grad_norm": 0.01526969433762133, "loss": 0.6880516052246094, "time": 0.41323699951171877, "epoch": 13, "memory": 9465, "step": 141492} +{"lr": 0.0003520365877844012, "data_time": 0.001304292678833008, "grad_norm": 0.01929947747848928, "loss": 0.6860296785831451, "time": 0.3967778444290161, "epoch": 13, "memory": 9465, "step": 141592} +{"lr": 0.0003520365877844012, "data_time": 0.0014894247055053712, "grad_norm": 0.018084618775174022, "loss": 0.688594925403595, "time": 0.4250460147857666, "epoch": 13, "memory": 9465, "step": 141692} +{"lr": 0.0003520365877844012, "data_time": 0.0013441801071166991, "grad_norm": 0.022172150015830994, "loss": 0.6866418898105622, "time": 0.4182627201080322, "epoch": 13, "memory": 9465, "step": 141792} +{"lr": 0.0003520365877844012, "data_time": 0.0013438940048217773, "grad_norm": 0.02240907186642289, "loss": 0.6872819364070892, "time": 0.4504395008087158, "epoch": 13, "memory": 9465, "step": 141892} +{"lr": 0.0003520365877844012, "data_time": 0.001279449462890625, "grad_norm": 0.01939082032768056, "loss": 0.6878681242465973, "time": 0.4323746919631958, "epoch": 13, "memory": 9465, "step": 141992} +{"lr": 0.0003520365877844012, "data_time": 0.001287698745727539, "grad_norm": 0.016904144175350667, "loss": 0.6885225951671601, "time": 0.4312117338180542, "epoch": 13, "memory": 9465, "step": 142092} +{"lr": 0.0003520365877844012, "data_time": 0.0014436483383178712, "grad_norm": 0.019593195826746523, "loss": 0.6895209610462188, "time": 0.4291876554489136, "epoch": 13, "memory": 9465, "step": 142192} +{"lr": 0.0003520365877844012, "data_time": 0.0013062715530395507, "grad_norm": 0.022013190877623857, "loss": 0.6822513163089752, "time": 0.45579018592834475, "epoch": 13, "memory": 9465, "step": 142292} +{"lr": 0.0003520365877844012, "data_time": 0.0014356851577758789, "grad_norm": 0.028816881170496345, "loss": 0.68393874168396, "time": 0.4303570032119751, "epoch": 13, "memory": 9465, "step": 142392} +{"lr": 0.0003520365877844012, "data_time": 0.0013211488723754883, "grad_norm": 0.030636927112936974, "loss": 0.6899877965450287, "time": 0.45420079231262206, "epoch": 13, "memory": 9465, "step": 142492} +{"lr": 0.0003520365877844012, "data_time": 0.0012770652770996093, "grad_norm": 0.017876827262807638, "loss": 0.6853049278259278, "time": 0.42228124141693113, "epoch": 13, "memory": 9465, "step": 142592} +{"lr": 0.0003520365877844012, "data_time": 0.0013705730438232423, "grad_norm": 0.00992819966049865, "loss": 0.6896498501300812, "time": 0.43505353927612306, "epoch": 13, "memory": 9465, "step": 142692} +{"lr": 0.0003520365877844012, "data_time": 0.0013044357299804687, "grad_norm": 0.033733045449480414, "loss": 0.6935043334960938, "time": 0.45402557849884034, "epoch": 13, "memory": 9465, "step": 142792} +{"lr": 0.0003520365877844012, "data_time": 0.001292109489440918, "grad_norm": 0.027146546076983212, "loss": 0.6876082599163056, "time": 0.4531333208084106, "epoch": 13, "memory": 9465, "step": 142892} +{"lr": 0.0003520365877844012, "data_time": 0.001516413688659668, "grad_norm": 0.01991870717611164, "loss": 0.6859760463237763, "time": 0.4385573625564575, "epoch": 13, "memory": 9465, "step": 142992} +{"lr": 0.0003520365877844012, "data_time": 0.0012458324432373046, "grad_norm": 0.0336504980456084, "loss": 0.6895107924938202, "time": 0.43419690132141114, "epoch": 13, "memory": 9465, "step": 143092} +{"lr": 0.0003520365877844012, "data_time": 0.0012791872024536133, "grad_norm": 0.0266926902346313, "loss": 0.6909848272800445, "time": 0.4190091133117676, "epoch": 13, "memory": 9465, "step": 143192} +{"lr": 0.0003520365877844012, "data_time": 0.0013389825820922852, "grad_norm": 0.01625212614890188, "loss": 0.683299720287323, "time": 0.4054360866546631, "epoch": 13, "memory": 9465, "step": 143292} +{"lr": 0.0003520365877844012, "data_time": 0.0014005184173583984, "grad_norm": 0.02609210801310837, "loss": 0.6856689631938935, "time": 0.4396503925323486, "epoch": 13, "memory": 9465, "step": 143392} +{"lr": 0.0003520365877844012, "data_time": 0.0013449907302856446, "grad_norm": 0.02831558436155319, "loss": 0.687919282913208, "time": 0.425563645362854, "epoch": 13, "memory": 9465, "step": 143492} +{"lr": 0.0003520365877844012, "data_time": 0.0013115882873535156, "grad_norm": 0.019528388057369738, "loss": 0.683935409784317, "time": 0.44516539573669434, "epoch": 13, "memory": 9465, "step": 143592} +{"lr": 0.0003520365877844012, "data_time": 0.001217031478881836, "grad_norm": 0.02119381525553763, "loss": 0.684372067451477, "time": 0.4636390686035156, "epoch": 13, "memory": 9465, "step": 143692} +{"lr": 0.0003520365877844012, "data_time": 0.0012293815612792968, "grad_norm": 0.02362422493752092, "loss": 0.6869829535484314, "time": 0.43630223274230956, "epoch": 13, "memory": 9465, "step": 143792} +{"lr": 0.0003520365877844012, "data_time": 0.0013032674789428711, "grad_norm": 0.028147122485097497, "loss": 0.6863200783729553, "time": 0.41666560173034667, "epoch": 13, "memory": 9465, "step": 143892} +{"lr": 0.0003520365877844012, "data_time": 0.0014888525009155273, "grad_norm": 0.017308444855734705, "loss": 0.6914076685905457, "time": 0.42134904861450195, "epoch": 13, "memory": 9465, "step": 143992} +{"lr": 0.0003520365877844012, "data_time": 0.002397298812866211, "grad_norm": 0.016407575085759164, "loss": 0.6847853183746337, "time": 0.4567969083786011, "epoch": 13, "memory": 9465, "step": 144092} +{"lr": 0.0003520365877844012, "data_time": 0.001467442512512207, "grad_norm": 0.021011793659999965, "loss": 0.6861694097518921, "time": 0.4472799301147461, "epoch": 13, "memory": 9465, "step": 144192} +{"lr": 0.0003520365877844012, "data_time": 0.0013340473175048827, "grad_norm": 0.023131793458014728, "loss": 0.6882535576820373, "time": 0.4448213577270508, "epoch": 13, "memory": 9465, "step": 144292} +{"lr": 0.0003520365877844012, "data_time": 0.0012208938598632813, "grad_norm": 0.022750809043645858, "loss": 0.6848455965518951, "time": 0.5096841096878052, "epoch": 13, "memory": 9465, "step": 144392} +{"lr": 0.0003520365877844012, "data_time": 0.0012856006622314453, "grad_norm": 0.01734750246978365, "loss": 0.6886095404624939, "time": 0.4157839298248291, "epoch": 13, "memory": 9465, "step": 144492} +{"lr": 0.0003520365877844012, "data_time": 0.0012665033340454102, "grad_norm": 0.01712261061184108, "loss": 0.6874357998371124, "time": 0.431915283203125, "epoch": 13, "memory": 9465, "step": 144592} +{"lr": 0.0003520365877844012, "data_time": 0.0013406038284301757, "grad_norm": 0.028326851688325404, "loss": 0.6830387890338898, "time": 0.42901763916015623, "epoch": 13, "memory": 9465, "step": 144692} +{"lr": 0.0003520365877844012, "data_time": 0.001374363899230957, "grad_norm": 0.018115551909431815, "loss": 0.6854941308498382, "time": 0.43001062870025636, "epoch": 13, "memory": 9465, "step": 144792} +{"lr": 0.0003520365877844012, "data_time": 0.001308274269104004, "grad_norm": 0.028630966693162917, "loss": 0.6886234223842621, "time": 0.41717722415924074, "epoch": 13, "memory": 9465, "step": 144892} +{"lr": 0.0003520365877844012, "data_time": 0.0013165235519409179, "grad_norm": 0.022329334617825224, "loss": 0.6895581483840942, "time": 0.43071537017822265, "epoch": 13, "memory": 9465, "step": 144992} +{"lr": 0.0003520365877844012, "data_time": 0.0013569831848144532, "grad_norm": 0.01647830307483673, "loss": 0.6855230152606964, "time": 0.40398895740509033, "epoch": 13, "memory": 9465, "step": 145092} +{"lr": 0.0003520365877844012, "data_time": 0.0012874841690063477, "grad_norm": 0.01579254070529714, "loss": 0.6887718439102173, "time": 0.42697365283966066, "epoch": 13, "memory": 9465, "step": 145192} +{"lr": 0.0003520365877844012, "data_time": 0.0013854026794433594, "grad_norm": 0.029465958569198848, "loss": 0.6884518384933471, "time": 0.4738156795501709, "epoch": 13, "memory": 9465, "step": 145292} +{"lr": 0.0003520365877844012, "data_time": 0.0013043403625488282, "grad_norm": 0.01683385097421706, "loss": 0.6826500654220581, "time": 0.4256121635437012, "epoch": 13, "memory": 9465, "step": 145392} +{"lr": 0.0003520365877844012, "data_time": 0.0013903379440307617, "grad_norm": 0.01804582504555583, "loss": 0.6884677290916443, "time": 0.44648089408874514, "epoch": 13, "memory": 9465, "step": 145492} +{"lr": 0.0003520365877844012, "data_time": 0.001339888572692871, "grad_norm": 0.025155717320740224, "loss": 0.6856862604618073, "time": 0.4355248212814331, "epoch": 13, "memory": 9465, "step": 145592} +{"lr": 0.0003520365877844012, "data_time": 0.0015591859817504882, "grad_norm": 0.016057035746052863, "loss": 0.6883014917373658, "time": 0.42293860912323, "epoch": 13, "memory": 9465, "step": 145692} +{"lr": 0.0003520365877844012, "data_time": 0.0013610601425170898, "grad_norm": 0.02582974242977798, "loss": 0.6869897067546844, "time": 0.4554993391036987, "epoch": 13, "memory": 9465, "step": 145792} +{"lr": 0.0003520365877844012, "data_time": 0.0014297008514404298, "grad_norm": 0.024726646230556072, "loss": 0.69018075466156, "time": 0.4758347988128662, "epoch": 13, "memory": 9465, "step": 145892} +{"lr": 0.0003520365877844012, "data_time": 0.0013002872467041016, "grad_norm": 0.018723942223004997, "loss": 0.6887227237224579, "time": 0.43060152530670165, "epoch": 13, "memory": 9465, "step": 145992} +{"lr": 0.0003520365877844012, "data_time": 0.0014956474304199218, "grad_norm": 0.017027101363055408, "loss": 0.6869645178318023, "time": 0.4316537380218506, "epoch": 13, "memory": 9465, "step": 146092} +{"lr": 0.0003520365877844012, "data_time": 0.001397562026977539, "grad_norm": 0.016929687187075614, "loss": 0.6878800630569458, "time": 0.4517225980758667, "epoch": 13, "memory": 9465, "step": 146192} +{"lr": 0.0003520365877844012, "data_time": 0.0014849424362182618, "grad_norm": 0.019114992558024825, "loss": 0.6873187065124512, "time": 0.42660088539123536, "epoch": 13, "memory": 9465, "step": 146292} +{"lr": 0.0003520365877844012, "data_time": 0.0012789726257324218, "grad_norm": 0.03214428598294035, "loss": 0.6874493896961212, "time": 0.4159404993057251, "epoch": 13, "memory": 9465, "step": 146392} +{"lr": 0.0003520365877844012, "data_time": 0.0012720823287963867, "grad_norm": 0.028108444390818478, "loss": 0.6880874156951904, "time": 0.422563362121582, "epoch": 13, "memory": 9465, "step": 146492} +{"lr": 0.0003520365877844012, "data_time": 0.001325702667236328, "grad_norm": 0.019555490429047494, "loss": 0.6882882297039032, "time": 0.4210850238800049, "epoch": 13, "memory": 9465, "step": 146592} +{"lr": 0.0003520365877844012, "data_time": 0.0012747764587402344, "grad_norm": 0.02407969650812447, "loss": 0.690867680311203, "time": 0.4369359016418457, "epoch": 13, "memory": 9465, "step": 146692} +{"lr": 0.0003520365877844012, "data_time": 0.0013010501861572266, "grad_norm": 0.02648129309527576, "loss": 0.687272697687149, "time": 0.39036743640899657, "epoch": 13, "memory": 9465, "step": 146792} +{"lr": 0.0003520365877844012, "data_time": 0.0013555526733398438, "grad_norm": 0.018041092460043727, "loss": 0.680459588766098, "time": 0.40427727699279786, "epoch": 13, "memory": 9465, "step": 146892} +{"lr": 0.0003520365877844012, "data_time": 0.0014187097549438477, "grad_norm": 0.017114895593840627, "loss": 0.6896293878555297, "time": 0.432601261138916, "epoch": 13, "memory": 9465, "step": 146992} +{"lr": 0.0003520365877844012, "data_time": 0.0012846708297729493, "grad_norm": 0.02265499666100368, "loss": 0.6880480766296386, "time": 0.40092661380767824, "epoch": 13, "memory": 9465, "step": 147092} +{"lr": 0.0003520365877844012, "data_time": 0.001348400115966797, "grad_norm": 0.016791835159529, "loss": 0.6838982820510864, "time": 0.39970180988311765, "epoch": 13, "memory": 9465, "step": 147192} +{"lr": 0.0003520365877844012, "data_time": 0.0013466835021972655, "grad_norm": 0.021194009529426693, "loss": 0.6869896233081818, "time": 0.40505883693695066, "epoch": 13, "memory": 9465, "step": 147292} +{"lr": 0.0003520365877844012, "data_time": 0.0016285896301269532, "grad_norm": 0.017275077872909606, "loss": 0.6869803667068481, "time": 0.44537665843963625, "epoch": 13, "memory": 9465, "step": 147392} +{"lr": 0.0003520365877844012, "data_time": 0.0014423847198486329, "grad_norm": 0.027895302092656494, "loss": 0.6824289977550506, "time": 0.41504294872283937, "epoch": 13, "memory": 9465, "step": 147492} +{"lr": 0.0003520365877844012, "data_time": 0.001257467269897461, "grad_norm": 0.018320670281536878, "loss": 0.684909564256668, "time": 0.44445552825927737, "epoch": 13, "memory": 9465, "step": 147592} +{"lr": 0.0003520365877844012, "data_time": 0.0013446331024169922, "grad_norm": 0.02354107324499637, "loss": 0.6893847823143006, "time": 0.38520045280456544, "epoch": 13, "memory": 9465, "step": 147692} +{"lr": 0.0003520365877844012, "data_time": 0.001256561279296875, "grad_norm": 0.01573030628496781, "loss": 0.686847734451294, "time": 0.42736985683441164, "epoch": 13, "memory": 9465, "step": 147792} +{"lr": 0.0003520365877844012, "data_time": 0.0013084650039672852, "grad_norm": 0.016619264078326524, "loss": 0.6850524246692657, "time": 0.40325417518615725, "epoch": 13, "memory": 9465, "step": 147892} +{"lr": 0.0003520365877844012, "data_time": 0.0013421058654785156, "grad_norm": 0.026577353337779643, "loss": 0.6866621017456055, "time": 0.45670936107635496, "epoch": 13, "memory": 9465, "step": 147992} +{"lr": 0.0003520365877844012, "data_time": 0.0013792991638183593, "grad_norm": 0.01836807201616466, "loss": 0.6861838102340698, "time": 0.3969640493392944, "epoch": 13, "memory": 9465, "step": 148092} +{"lr": 0.0003520365877844012, "data_time": 0.0013062000274658204, "grad_norm": 0.013027273537591099, "loss": 0.6824309349060058, "time": 0.402159309387207, "epoch": 13, "memory": 9465, "step": 148192} +{"lr": 0.0003520365877844012, "data_time": 0.0013721466064453125, "grad_norm": 0.017777485214173792, "loss": 0.6860322594642639, "time": 0.39197554588317873, "epoch": 13, "memory": 9465, "step": 148292} +{"lr": 0.0003520365877844012, "data_time": 0.0014073610305786132, "grad_norm": 0.02188335722312331, "loss": 0.6832489430904388, "time": 0.38784499168395997, "epoch": 13, "memory": 9465, "step": 148392} +{"lr": 0.0003520365877844012, "data_time": 0.0013094663619995118, "grad_norm": 0.029621819080784918, "loss": 0.6863331019878387, "time": 0.40296661853790283, "epoch": 13, "memory": 9465, "step": 148492} +{"lr": 0.0003520365877844012, "data_time": 0.0013120651245117187, "grad_norm": 0.021716117835603656, "loss": 0.6840667843818664, "time": 0.44245638847351076, "epoch": 13, "memory": 9465, "step": 148592} +{"lr": 0.0003520365877844012, "data_time": 0.0014384031295776368, "grad_norm": 0.025756726786494256, "loss": 0.6886322021484375, "time": 0.4498127937316895, "epoch": 13, "memory": 9465, "step": 148692} +{"lr": 0.0003520365877844012, "data_time": 0.0012794017791748046, "grad_norm": 0.02449752176180482, "loss": 0.6841812789440155, "time": 0.39390561580657957, "epoch": 13, "memory": 9465, "step": 148792} +{"lr": 0.0003520365877844012, "data_time": 0.0013397455215454102, "grad_norm": 0.017251241515623406, "loss": 0.6848391532897949, "time": 0.4220003366470337, "epoch": 13, "memory": 9465, "step": 148892} +{"lr": 0.0003520365877844012, "data_time": 0.0012875080108642578, "grad_norm": 0.02317849623505026, "loss": 0.6844637930393219, "time": 0.4177465200424194, "epoch": 13, "memory": 9465, "step": 148992} +{"lr": 0.0003520365877844012, "data_time": 0.0016066551208496094, "grad_norm": 0.025580171146430075, "loss": 0.6857383906841278, "time": 0.4121209383010864, "epoch": 13, "memory": 9465, "step": 149092} +{"lr": 0.0003520365877844012, "data_time": 0.0013405323028564454, "grad_norm": 0.02009988371282816, "loss": 0.6857160985469818, "time": 0.3964763879776001, "epoch": 13, "memory": 9465, "step": 149192} +{"lr": 0.0003520365877844012, "data_time": 0.001444387435913086, "grad_norm": 0.03267081966623664, "loss": 0.6920430958271027, "time": 0.40468361377716067, "epoch": 13, "memory": 9465, "step": 149292} +{"lr": 0.0003520365877844012, "data_time": 0.001538228988647461, "grad_norm": 0.031247841753065587, "loss": 0.6883592486381531, "time": 0.4112871170043945, "epoch": 13, "memory": 9465, "step": 149392} +{"lr": 0.0003520365877844012, "data_time": 0.0013257503509521485, "grad_norm": 0.029722631350159644, "loss": 0.6896104454994202, "time": 0.3924055814743042, "epoch": 13, "memory": 9465, "step": 149492} +{"lr": 0.0003520365877844012, "data_time": 0.0020203590393066406, "grad_norm": 0.022426737286150456, "loss": 0.6843945145606994, "time": 0.406878399848938, "epoch": 13, "memory": 9465, "step": 149592} +{"lr": 0.0003520365877844012, "data_time": 0.011359167098999024, "grad_norm": 0.024466575216501953, "loss": 0.6845511078834534, "time": 0.38599724769592286, "epoch": 13, "memory": 9465, "step": 149692} +{"accuracy/top1": 100.0, "data_time": 0.30741747687844667, "time": 0.5612477695240694, "step": 13} +{"lr": 0.00028027470262892445, "data_time": 0.040578031539916994, "grad_norm": 0.02363255023956299, "loss": 0.6856961190700531, "time": 0.44960076808929444, "epoch": 14, "memory": 9465, "step": 149808} +{"lr": 0.00028027470262892445, "data_time": 0.0013695716857910155, "grad_norm": 0.03702503265812993, "loss": 0.6921934664249421, "time": 0.4566654682159424, "epoch": 14, "memory": 9465, "step": 149908} +{"lr": 0.00028027470262892445, "data_time": 0.001612401008605957, "grad_norm": 0.025036737532354893, "loss": 0.6891972303390503, "time": 0.3936192512512207, "epoch": 14, "memory": 9465, "step": 150008} +{"lr": 0.00028027470262892445, "data_time": 0.0014136075973510743, "grad_norm": 0.028032448887825013, "loss": 0.6853726267814636, "time": 0.40378150939941404, "epoch": 14, "memory": 9465, "step": 150108} +{"lr": 0.00028027470262892445, "data_time": 0.0016391277313232422, "grad_norm": 0.02167459358461201, "loss": 0.6890131831169128, "time": 0.4048474311828613, "epoch": 14, "memory": 9465, "step": 150208} +{"lr": 0.00028027470262892445, "data_time": 0.0016252994537353516, "grad_norm": 0.01866477259900421, "loss": 0.6893049836158752, "time": 0.3884351015090942, "epoch": 14, "memory": 9465, "step": 150308} +{"lr": 0.00028027470262892445, "data_time": 0.0014384984970092773, "grad_norm": 0.029272266384214162, "loss": 0.6797670483589172, "time": 0.38600687980651854, "epoch": 14, "memory": 9465, "step": 150408} +{"lr": 0.00028027470262892445, "data_time": 0.001467585563659668, "grad_norm": 0.016809142468264326, "loss": 0.6895006775856019, "time": 0.41752817630767824, "epoch": 14, "memory": 9465, "step": 150508} +{"lr": 0.00028027470262892445, "data_time": 0.0013631820678710938, "grad_norm": 0.0183136819396168, "loss": 0.6832292556762696, "time": 0.43120110034942627, "epoch": 14, "memory": 9465, "step": 150608} +{"lr": 0.00028027470262892445, "data_time": 0.0022603511810302735, "grad_norm": 0.01756924302317202, "loss": 0.6859959423542022, "time": 0.3948364734649658, "epoch": 14, "memory": 9465, "step": 150708} +{"lr": 0.00028027470262892445, "data_time": 0.001507735252380371, "grad_norm": 0.024924750439822674, "loss": 0.6871319591999054, "time": 0.4168450355529785, "epoch": 14, "memory": 9465, "step": 150808} +{"lr": 0.00028027470262892445, "data_time": 0.0013359546661376952, "grad_norm": 0.029977887263521553, "loss": 0.6806355714797974, "time": 0.4189192533493042, "epoch": 14, "memory": 9465, "step": 150908} +{"lr": 0.00028027470262892445, "data_time": 0.0015645742416381836, "grad_norm": 0.018552739499136807, "loss": 0.6834400951862335, "time": 0.3801810503005981, "epoch": 14, "memory": 9465, "step": 151008} +{"lr": 0.00028027470262892445, "data_time": 0.0013875722885131835, "grad_norm": 0.019572274759411812, "loss": 0.6849079072475434, "time": 0.4099190950393677, "epoch": 14, "memory": 9465, "step": 151108} +{"lr": 0.00028027470262892445, "data_time": 0.0013000965118408203, "grad_norm": 0.020054957189131528, "loss": 0.688249260187149, "time": 0.41030781269073485, "epoch": 14, "memory": 9465, "step": 151208} +{"lr": 0.00028027470262892445, "data_time": 0.0013312339782714845, "grad_norm": 0.023754696641117335, "loss": 0.6887979507446289, "time": 0.42234656810760496, "epoch": 14, "memory": 9465, "step": 151308} +{"lr": 0.00028027470262892445, "data_time": 0.0012336015701293946, "grad_norm": 0.02806314714252949, "loss": 0.6866526246070862, "time": 0.3873327970504761, "epoch": 14, "memory": 9465, "step": 151408} +{"lr": 0.00028027470262892445, "data_time": 0.001320052146911621, "grad_norm": 0.016271448781481013, "loss": 0.6874719798564911, "time": 0.3861680030822754, "epoch": 14, "memory": 9465, "step": 151508} +{"lr": 0.00028027470262892445, "data_time": 0.0012776374816894532, "grad_norm": 0.01814198181964457, "loss": 0.6835083842277527, "time": 0.3970287561416626, "epoch": 14, "memory": 9465, "step": 151608} +{"lr": 0.00028027470262892445, "data_time": 0.0013757705688476562, "grad_norm": 0.01849473762558773, "loss": 0.6913939416408539, "time": 0.393136191368103, "epoch": 14, "memory": 9465, "step": 151708} +{"lr": 0.00028027470262892445, "data_time": 0.0013037443161010743, "grad_norm": 0.013537014625035227, "loss": 0.6798896551132202, "time": 0.40006446838378906, "epoch": 14, "memory": 9465, "step": 151808} +{"lr": 0.00028027470262892445, "data_time": 0.0012566089630126954, "grad_norm": 0.031790386396460237, "loss": 0.682720524072647, "time": 0.4363923311233521, "epoch": 14, "memory": 9465, "step": 151908} +{"lr": 0.00028027470262892445, "data_time": 0.001699542999267578, "grad_norm": 0.030915590189397334, "loss": 0.6899689972400666, "time": 0.4120911121368408, "epoch": 14, "memory": 9465, "step": 152008} +{"lr": 0.00028027470262892445, "data_time": 0.0014336585998535156, "grad_norm": 0.024079860048368575, "loss": 0.6891722679138184, "time": 0.4283134460449219, "epoch": 14, "memory": 9465, "step": 152108} +{"lr": 0.00028027470262892445, "data_time": 0.0013783693313598633, "grad_norm": 0.02360130213201046, "loss": 0.6899493277072907, "time": 0.40278360843658445, "epoch": 14, "memory": 9465, "step": 152208} +{"lr": 0.00028027470262892445, "data_time": 0.0014038324356079102, "grad_norm": 0.028875906398752704, "loss": 0.6871864378452301, "time": 0.41941835880279543, "epoch": 14, "memory": 9465, "step": 152308} +{"lr": 0.00028027470262892445, "data_time": 0.0016535520553588867, "grad_norm": 0.0323748073540628, "loss": 0.6836329460144043, "time": 0.38824920654296874, "epoch": 14, "memory": 9465, "step": 152408} +{"lr": 0.00028027470262892445, "data_time": 0.0015223026275634766, "grad_norm": 0.02168747780378908, "loss": 0.6844379901885986, "time": 0.40359928607940676, "epoch": 14, "memory": 9465, "step": 152508} +{"lr": 0.00028027470262892445, "data_time": 0.0014693260192871094, "grad_norm": 0.0183307379251346, "loss": 0.6859334170818329, "time": 0.44213550090789794, "epoch": 14, "memory": 9465, "step": 152608} +{"lr": 0.00028027470262892445, "data_time": 0.0016148805618286133, "grad_norm": 0.01920490147313103, "loss": 0.6874366044998169, "time": 0.40976390838623045, "epoch": 14, "memory": 9465, "step": 152708} +{"lr": 0.00028027470262892445, "data_time": 0.0016313314437866211, "grad_norm": 0.015319457871373742, "loss": 0.686477530002594, "time": 0.39671027660369873, "epoch": 14, "memory": 9465, "step": 152808} +{"lr": 0.00028027470262892445, "data_time": 0.0012980222702026368, "grad_norm": 0.01969348026905209, "loss": 0.6829213261604309, "time": 0.4105373859405518, "epoch": 14, "memory": 9465, "step": 152908} +{"lr": 0.00028027470262892445, "data_time": 0.0014499902725219726, "grad_norm": 0.021235161134973167, "loss": 0.6763473749160767, "time": 0.38748793601989745, "epoch": 14, "memory": 9465, "step": 153008} +{"lr": 0.00028027470262892445, "data_time": 0.00168154239654541, "grad_norm": 0.015591676131589339, "loss": 0.6831153154373169, "time": 0.4621685266494751, "epoch": 14, "memory": 9465, "step": 153108} +{"lr": 0.00028027470262892445, "data_time": 0.001639246940612793, "grad_norm": 0.018944092630408704, "loss": 0.6830090641975403, "time": 0.42252817153930666, "epoch": 14, "memory": 9465, "step": 153208} +{"lr": 0.00028027470262892445, "data_time": 0.0013518095016479491, "grad_norm": 0.02016517990268767, "loss": 0.6845646619796752, "time": 0.4187780380249023, "epoch": 14, "memory": 9465, "step": 153308} +{"lr": 0.00028027470262892445, "data_time": 0.0013185501098632812, "grad_norm": 0.01672271793941036, "loss": 0.6840503096580506, "time": 0.37966482639312743, "epoch": 14, "memory": 9465, "step": 153408} +{"lr": 0.00028027470262892445, "data_time": 0.0018891096115112305, "grad_norm": 0.021451255236752332, "loss": 0.68685182929039, "time": 0.41676404476165774, "epoch": 14, "memory": 9465, "step": 153508} +{"lr": 0.00028027470262892445, "data_time": 0.0013552427291870118, "grad_norm": 0.02101659409236163, "loss": 0.6868496894836426, "time": 0.44055216312408446, "epoch": 14, "memory": 9465, "step": 153608} +{"lr": 0.00028027470262892445, "data_time": 0.0014045000076293944, "grad_norm": 0.017874143016524614, "loss": 0.683040851354599, "time": 0.43531954288482666, "epoch": 14, "memory": 9465, "step": 153708} +{"lr": 0.00028027470262892445, "data_time": 0.0014707565307617188, "grad_norm": 0.0309216583147645, "loss": 0.685529601573944, "time": 0.3882774591445923, "epoch": 14, "memory": 9465, "step": 153808} +{"lr": 0.00028027470262892445, "data_time": 0.001372051239013672, "grad_norm": 0.036785055231302974, "loss": 0.6863934159278869, "time": 0.42968709468841554, "epoch": 14, "memory": 9465, "step": 153908} +{"lr": 0.00028027470262892445, "data_time": 0.0014786481857299804, "grad_norm": 0.01155533161945641, "loss": 0.686000519990921, "time": 0.4448122262954712, "epoch": 14, "memory": 9465, "step": 154008} +{"lr": 0.00028027470262892445, "data_time": 0.0014128923416137696, "grad_norm": 0.018938945885747672, "loss": 0.6897554159164428, "time": 0.3970362186431885, "epoch": 14, "memory": 9465, "step": 154108} +{"lr": 0.00028027470262892445, "data_time": 0.0013833045959472656, "grad_norm": 0.01627225533593446, "loss": 0.6871192872524261, "time": 0.3953340291976929, "epoch": 14, "memory": 9465, "step": 154208} +{"lr": 0.00028027470262892445, "data_time": 0.0015748500823974609, "grad_norm": 0.024380440218374134, "loss": 0.683938843011856, "time": 0.38283524513244627, "epoch": 14, "memory": 9465, "step": 154308} +{"lr": 0.00028027470262892445, "data_time": 0.0014348268508911134, "grad_norm": 0.05108619704842567, "loss": 0.6826554536819458, "time": 0.40686051845550536, "epoch": 14, "memory": 9465, "step": 154408} +{"lr": 0.00028027470262892445, "data_time": 0.0014477968215942383, "grad_norm": 0.01483023869805038, "loss": 0.6876264333724975, "time": 0.42267732620239257, "epoch": 14, "memory": 9465, "step": 154508} +{"lr": 0.00028027470262892445, "data_time": 0.0013699293136596679, "grad_norm": 0.042110367212444544, "loss": 0.6855251073837281, "time": 0.40648598670959474, "epoch": 14, "memory": 9465, "step": 154608} +{"lr": 0.00028027470262892445, "data_time": 0.00148317813873291, "grad_norm": 0.05280132433399558, "loss": 0.686598539352417, "time": 0.42174410820007324, "epoch": 14, "memory": 9465, "step": 154708} +{"lr": 0.00028027470262892445, "data_time": 0.0012965202331542969, "grad_norm": 0.041294691059738396, "loss": 0.6906365215778351, "time": 0.3991217851638794, "epoch": 14, "memory": 9465, "step": 154808} +{"lr": 0.00028027470262892445, "data_time": 0.0013070106506347656, "grad_norm": 0.0483148371335119, "loss": 0.6866521239280701, "time": 0.4468418598175049, "epoch": 14, "memory": 9465, "step": 154908} +{"lr": 0.00028027470262892445, "data_time": 0.0015760421752929687, "grad_norm": 0.19067063089460135, "loss": 0.6856415331363678, "time": 0.4100227355957031, "epoch": 14, "memory": 9465, "step": 155008} +{"lr": 0.00028027470262892445, "data_time": 0.002025866508483887, "grad_norm": 0.02737265024334192, "loss": 0.6825403451919556, "time": 0.3975337266921997, "epoch": 14, "memory": 9465, "step": 155108} +{"lr": 0.00028027470262892445, "data_time": 0.002056574821472168, "grad_norm": 0.03622059300541878, "loss": 0.6835957884788513, "time": 0.4041880130767822, "epoch": 14, "memory": 9465, "step": 155208} +{"lr": 0.00028027470262892445, "data_time": 0.0022015810012817384, "grad_norm": 0.0383616394829005, "loss": 0.684986925125122, "time": 0.39022395610809324, "epoch": 14, "memory": 9465, "step": 155308} +{"lr": 0.00028027470262892445, "data_time": 0.002637624740600586, "grad_norm": 0.026423221081495286, "loss": 0.6854630827903747, "time": 0.3999086856842041, "epoch": 14, "memory": 9465, "step": 155408} +{"lr": 0.00028027470262892445, "data_time": 0.0014970779418945312, "grad_norm": 0.07129687988199293, "loss": 0.6865840375423431, "time": 0.4012380361557007, "epoch": 14, "memory": 9465, "step": 155508} +{"lr": 0.00028027470262892445, "data_time": 0.0019565105438232424, "grad_norm": 0.042816708935424684, "loss": 0.6909674525260925, "time": 0.40330042839050295, "epoch": 14, "memory": 9465, "step": 155608} +{"lr": 0.00028027470262892445, "data_time": 0.001831960678100586, "grad_norm": 0.02322618179023266, "loss": 0.6862935781478882, "time": 0.4133880376815796, "epoch": 14, "memory": 9465, "step": 155708} +{"lr": 0.00028027470262892445, "data_time": 0.001990056037902832, "grad_norm": 0.08056890348671004, "loss": 0.6866194248199463, "time": 0.42566213607788084, "epoch": 14, "memory": 9465, "step": 155808} +{"lr": 0.00028027470262892445, "data_time": 0.0026386499404907225, "grad_norm": 0.02329719658009708, "loss": 0.6899753332138061, "time": 0.37984864711761473, "epoch": 14, "memory": 9465, "step": 155908} +{"lr": 0.00028027470262892445, "data_time": 0.002470850944519043, "grad_norm": 0.031849815510213375, "loss": 0.6918782651424408, "time": 0.4714346170425415, "epoch": 14, "memory": 9465, "step": 156008} +{"lr": 0.00028027470262892445, "data_time": 0.0022544622421264648, "grad_norm": 0.09037298895418644, "loss": 0.6869585394859314, "time": 0.402772331237793, "epoch": 14, "memory": 9465, "step": 156108} +{"lr": 0.00028027470262892445, "data_time": 0.001898646354675293, "grad_norm": 0.038070216029882434, "loss": 0.6842526078224183, "time": 0.39388854503631593, "epoch": 14, "memory": 9465, "step": 156208} +{"lr": 0.00028027470262892445, "data_time": 0.001858663558959961, "grad_norm": 0.14319400303065777, "loss": 0.6865509212017059, "time": 0.3899399757385254, "epoch": 14, "memory": 9465, "step": 156308} +{"lr": 0.00028027470262892445, "data_time": 0.002242755889892578, "grad_norm": 0.031476462911814454, "loss": 0.6879517734050751, "time": 0.4034332513809204, "epoch": 14, "memory": 9465, "step": 156408} +{"lr": 0.00028027470262892445, "data_time": 0.0023975133895874023, "grad_norm": 0.027712808176875114, "loss": 0.681372219324112, "time": 0.4083216667175293, "epoch": 14, "memory": 9465, "step": 156508} +{"lr": 0.00028027470262892445, "data_time": 0.00198667049407959, "grad_norm": 0.028507962450385095, "loss": 0.6839473962783813, "time": 0.4277604818344116, "epoch": 14, "memory": 9465, "step": 156608} +{"lr": 0.00028027470262892445, "data_time": 0.0019521236419677735, "grad_norm": 0.02448674967745319, "loss": 0.6873429596424103, "time": 0.4227177858352661, "epoch": 14, "memory": 9465, "step": 156708} +{"lr": 0.00028027470262892445, "data_time": 0.001888108253479004, "grad_norm": 0.0313364093657583, "loss": 0.6884092569351197, "time": 0.4373295545578003, "epoch": 14, "memory": 9465, "step": 156808} +{"lr": 0.00028027470262892445, "data_time": 0.0019359111785888672, "grad_norm": 0.02219557317439467, "loss": 0.6841471076011658, "time": 0.4302870988845825, "epoch": 14, "memory": 9465, "step": 156908} +{"lr": 0.00028027470262892445, "data_time": 0.002111482620239258, "grad_norm": 0.03421001746319234, "loss": 0.6835174262523651, "time": 0.39418256282806396, "epoch": 14, "memory": 9465, "step": 157008} +{"lr": 0.00028027470262892445, "data_time": 0.0020578622817993162, "grad_norm": 0.013493796996772289, "loss": 0.6880625247955322, "time": 0.43144266605377196, "epoch": 14, "memory": 9465, "step": 157108} +{"lr": 0.00028027470262892445, "data_time": 0.0016768693923950196, "grad_norm": 0.029052113997749984, "loss": 0.689518803358078, "time": 0.40128026008605955, "epoch": 14, "memory": 9465, "step": 157208} +{"lr": 0.00028027470262892445, "data_time": 0.0016528606414794923, "grad_norm": 0.023886117665097118, "loss": 0.6865138351917267, "time": 0.3945567846298218, "epoch": 14, "memory": 9465, "step": 157308} +{"lr": 0.00028027470262892445, "data_time": 0.002325916290283203, "grad_norm": 0.027004258264787494, "loss": 0.6864407360553741, "time": 0.39932076930999755, "epoch": 14, "memory": 9465, "step": 157408} +{"lr": 0.00028027470262892445, "data_time": 0.002185344696044922, "grad_norm": 0.08682013629004359, "loss": 0.6798774540424347, "time": 0.39725723266601565, "epoch": 14, "memory": 9465, "step": 157508} +{"lr": 0.00028027470262892445, "data_time": 0.0021752119064331055, "grad_norm": 0.028024967899546026, "loss": 0.6860984563827515, "time": 0.4208559989929199, "epoch": 14, "memory": 9465, "step": 157608} +{"lr": 0.00028027470262892445, "data_time": 0.0019136667251586914, "grad_norm": 0.029734809999354184, "loss": 0.6827096998691559, "time": 0.3966014623641968, "epoch": 14, "memory": 9465, "step": 157708} +{"lr": 0.00028027470262892445, "data_time": 0.0017208337783813476, "grad_norm": 0.03303380589932203, "loss": 0.6838254630565643, "time": 0.40142972469329835, "epoch": 14, "memory": 9465, "step": 157808} +{"lr": 0.00028027470262892445, "data_time": 0.0026412248611450196, "grad_norm": 0.02461668774485588, "loss": 0.6841819941997528, "time": 0.39656853675842285, "epoch": 14, "memory": 9465, "step": 157908} +{"lr": 0.00028027470262892445, "data_time": 0.0018293142318725586, "grad_norm": 0.0292770110652782, "loss": 0.6882569372653962, "time": 0.41058878898620604, "epoch": 14, "memory": 9465, "step": 158008} +{"lr": 0.00028027470262892445, "data_time": 0.0019101142883300782, "grad_norm": 0.021645177667960525, "loss": 0.6902525365352631, "time": 0.41611638069152834, "epoch": 14, "memory": 9465, "step": 158108} +{"lr": 0.00028027470262892445, "data_time": 0.001793646812438965, "grad_norm": 0.06236221301369369, "loss": 0.6861551165580749, "time": 0.4351755857467651, "epoch": 14, "memory": 9465, "step": 158208} +{"lr": 0.00028027470262892445, "data_time": 0.001773834228515625, "grad_norm": 0.037126504629850385, "loss": 0.6846661806106568, "time": 0.40097382068634035, "epoch": 14, "memory": 9465, "step": 158308} +{"lr": 0.00028027470262892445, "data_time": 0.0017854690551757813, "grad_norm": 0.01951689086854458, "loss": 0.6915440082550048, "time": 0.3932562589645386, "epoch": 14, "memory": 9465, "step": 158408} +{"lr": 0.00028027470262892445, "data_time": 0.0020975589752197264, "grad_norm": 0.043020909558981656, "loss": 0.6870308697223664, "time": 0.4824603796005249, "epoch": 14, "memory": 9465, "step": 158508} +{"lr": 0.00028027470262892445, "data_time": 0.0019032716751098632, "grad_norm": 0.08353183420840651, "loss": 0.6912239372730256, "time": 0.4043946504592896, "epoch": 14, "memory": 9465, "step": 158608} +{"lr": 0.00028027470262892445, "data_time": 0.001536989212036133, "grad_norm": 0.035521398112177846, "loss": 0.6825567483901978, "time": 0.4237253427505493, "epoch": 14, "memory": 9465, "step": 158708} +{"lr": 0.00028027470262892445, "data_time": 0.002265143394470215, "grad_norm": 0.12171778678894044, "loss": 0.6851697981357574, "time": 0.40695106983184814, "epoch": 14, "memory": 9465, "step": 158808} +{"lr": 0.00028027470262892445, "data_time": 0.0019767045974731444, "grad_norm": 0.028613679856061936, "loss": 0.6916392803192138, "time": 0.41077594757080077, "epoch": 14, "memory": 9465, "step": 158908} +{"lr": 0.00028027470262892445, "data_time": 0.002177524566650391, "grad_norm": 0.019880210445262492, "loss": 0.686233127117157, "time": 0.40418262481689454, "epoch": 14, "memory": 9465, "step": 159008} +{"lr": 0.00028027470262892445, "data_time": 0.002059483528137207, "grad_norm": 0.015642314485739918, "loss": 0.6872299492359162, "time": 0.40851283073425293, "epoch": 14, "memory": 9465, "step": 159108} +{"lr": 0.00028027470262892445, "data_time": 0.0013713836669921875, "grad_norm": 0.05535896783694625, "loss": 0.695352977514267, "time": 0.41545073986053466, "epoch": 14, "memory": 9465, "step": 159208} +{"lr": 0.00028027470262892445, "data_time": 0.0013283014297485352, "grad_norm": 0.028885880671441554, "loss": 0.683948028087616, "time": 0.41185173988342283, "epoch": 14, "memory": 9465, "step": 159308} +{"lr": 0.00028027470262892445, "data_time": 0.0013547182083129884, "grad_norm": 0.03945363105740398, "loss": 0.6870556950569153, "time": 0.42182073593139646, "epoch": 14, "memory": 9465, "step": 159408} +{"lr": 0.00028027470262892445, "data_time": 0.001593613624572754, "grad_norm": 0.05568387564271689, "loss": 0.6910943686962128, "time": 0.44905292987823486, "epoch": 14, "memory": 9465, "step": 159508} +{"lr": 0.00028027470262892445, "data_time": 0.0017818927764892579, "grad_norm": 0.12322857351973653, "loss": 0.6851725339889526, "time": 0.4095428943634033, "epoch": 14, "memory": 9465, "step": 159608} +{"lr": 0.00028027470262892445, "data_time": 0.0017316102981567382, "grad_norm": 0.0301972683519125, "loss": 0.68194180727005, "time": 0.436036491394043, "epoch": 14, "memory": 9465, "step": 159708} +{"lr": 0.00028027470262892445, "data_time": 0.0023550987243652344, "grad_norm": 0.02634110040962696, "loss": 0.6899545848369598, "time": 0.40677814483642577, "epoch": 14, "memory": 9465, "step": 159808} +{"lr": 0.00028027470262892445, "data_time": 0.0015550374984741211, "grad_norm": 0.0231823165435344, "loss": 0.6842324733734131, "time": 0.4056877851486206, "epoch": 14, "memory": 9465, "step": 159908} +{"lr": 0.00028027470262892445, "data_time": 0.0014517068862915038, "grad_norm": 0.024236548971384764, "loss": 0.6827456235885621, "time": 0.39584548473358155, "epoch": 14, "memory": 9465, "step": 160008} +{"lr": 0.00028027470262892445, "data_time": 0.0014163970947265625, "grad_norm": 0.019567030342295767, "loss": 0.6907675683498382, "time": 0.42159667015075686, "epoch": 14, "memory": 9465, "step": 160108} +{"lr": 0.00028027470262892445, "data_time": 0.0014188766479492187, "grad_norm": 0.03959140651859343, "loss": 0.6838397264480591, "time": 0.4625308752059937, "epoch": 14, "memory": 9465, "step": 160208} +{"lr": 0.00028027470262892445, "data_time": 0.002048826217651367, "grad_norm": 0.022645322373136878, "loss": 0.6872423529624939, "time": 0.4821256875991821, "epoch": 14, "memory": 9465, "step": 160308} +{"lr": 0.00028027470262892445, "data_time": 0.001827239990234375, "grad_norm": 0.029500603955239058, "loss": 0.6850732922554016, "time": 0.4751749992370605, "epoch": 14, "memory": 9465, "step": 160408} +{"lr": 0.00028027470262892445, "data_time": 0.002065300941467285, "grad_norm": 0.01995218815281987, "loss": 0.6893764853477478, "time": 0.42381384372711184, "epoch": 14, "memory": 9465, "step": 160508} +{"lr": 0.00028027470262892445, "data_time": 0.0014534950256347655, "grad_norm": 0.0288451686501503, "loss": 0.6832719922065735, "time": 0.39760754108428953, "epoch": 14, "memory": 9465, "step": 160608} +{"lr": 0.00028027470262892445, "data_time": 0.0017058849334716797, "grad_norm": 0.03133925325237215, "loss": 0.6815317809581757, "time": 0.40827987194061277, "epoch": 14, "memory": 9465, "step": 160708} +{"lr": 0.00028027470262892445, "data_time": 0.0015241384506225586, "grad_norm": 0.019520680559799074, "loss": 0.6827360630035401, "time": 0.435684871673584, "epoch": 14, "memory": 9465, "step": 160808} +{"lr": 0.00028027470262892445, "data_time": 0.0020036697387695312, "grad_norm": 0.022928539512213318, "loss": 0.6836845993995666, "time": 0.4341479778289795, "epoch": 14, "memory": 9465, "step": 160908} +{"lr": 0.00028027470262892445, "data_time": 0.0014281749725341796, "grad_norm": 0.019421445624902844, "loss": 0.6835116386413574, "time": 0.47015607357025146, "epoch": 14, "memory": 9465, "step": 161008} +{"lr": 0.00028027470262892445, "data_time": 0.0014080047607421876, "grad_norm": 0.024089456733781843, "loss": 0.6846429884433747, "time": 0.39995200634002687, "epoch": 14, "memory": 9465, "step": 161108} +{"lr": 0.00028027470262892445, "data_time": 0.00597069263458252, "grad_norm": 0.027609765203669666, "loss": 0.6893363296985626, "time": 0.40682644844055177, "epoch": 14, "memory": 9465, "step": 161208} +{"accuracy/top1": 100.0, "data_time": 0.2853405054877786, "time": 0.5453655439264634, "step": 14} +{"lr": 0.0002140463001152259, "data_time": 0.08123340606689453, "grad_norm": 0.01871814049081877, "loss": 0.6883578598499298, "time": 0.4226930856704712, "epoch": 15, "memory": 9465, "step": 161324} +{"lr": 0.0002140463001152259, "data_time": 0.0014652729034423829, "grad_norm": 0.017157578165642916, "loss": 0.6913799524307251, "time": 0.40656795501708987, "epoch": 15, "memory": 9465, "step": 161424} +{"lr": 0.0002140463001152259, "data_time": 0.0014858722686767578, "grad_norm": 0.013531868599238805, "loss": 0.6874639749526977, "time": 0.3904269695281982, "epoch": 15, "memory": 9465, "step": 161524} +{"lr": 0.0002140463001152259, "data_time": 0.0012707233428955079, "grad_norm": 0.016254200006369502, "loss": 0.6860540568828583, "time": 0.41412451267242434, "epoch": 15, "memory": 9465, "step": 161624} +{"lr": 0.0002140463001152259, "data_time": 0.0013164758682250976, "grad_norm": 0.01915219735528808, "loss": 0.6880498588085174, "time": 0.4387094974517822, "epoch": 15, "memory": 9465, "step": 161724} +{"lr": 0.0002140463001152259, "data_time": 0.00133209228515625, "grad_norm": 0.01842809095978737, "loss": 0.6838313162326812, "time": 0.454335355758667, "epoch": 15, "memory": 9465, "step": 161824} +{"lr": 0.0002140463001152259, "data_time": 0.0021929264068603514, "grad_norm": 0.023325519729405643, "loss": 0.6863415241241455, "time": 0.4828128099441528, "epoch": 15, "memory": 9465, "step": 161924} +{"lr": 0.0002140463001152259, "data_time": 0.0015517950057983398, "grad_norm": 0.02742227101698518, "loss": 0.6828200161457062, "time": 0.4705327272415161, "epoch": 15, "memory": 9465, "step": 162024} +{"lr": 0.0002140463001152259, "data_time": 0.0014075756072998047, "grad_norm": 0.02574599109357223, "loss": 0.6864741265773773, "time": 0.4059830904006958, "epoch": 15, "memory": 9465, "step": 162124} +{"lr": 0.0002140463001152259, "data_time": 0.0012774944305419921, "grad_norm": 0.09568184425588697, "loss": 0.691490238904953, "time": 0.46218457221984866, "epoch": 15, "memory": 9465, "step": 162224} +{"lr": 0.0002140463001152259, "data_time": 0.0013579130172729492, "grad_norm": 0.031108124926686288, "loss": 0.6912797629833222, "time": 0.4240349292755127, "epoch": 15, "memory": 9465, "step": 162324} +{"lr": 0.0002140463001152259, "data_time": 0.0013774871826171876, "grad_norm": 0.022236520587466658, "loss": 0.6888606786727905, "time": 0.41772751808166503, "epoch": 15, "memory": 9465, "step": 162424} +{"lr": 0.0002140463001152259, "data_time": 0.0013132333755493165, "grad_norm": 0.01871593832038343, "loss": 0.6861822009086609, "time": 0.42862424850463865, "epoch": 15, "memory": 9465, "step": 162524} +{"lr": 0.0002140463001152259, "data_time": 0.0012571334838867188, "grad_norm": 0.02354892613366246, "loss": 0.6847135305404664, "time": 0.42249486446380613, "epoch": 15, "memory": 9465, "step": 162624} +{"lr": 0.0002140463001152259, "data_time": 0.0013778209686279297, "grad_norm": 0.017027739610057324, "loss": 0.6876058578491211, "time": 0.4112085819244385, "epoch": 15, "memory": 9465, "step": 162724} +{"lr": 0.0002140463001152259, "data_time": 0.0016553640365600587, "grad_norm": 0.021537176356650888, "loss": 0.6872940182685852, "time": 0.44995846748352053, "epoch": 15, "memory": 9465, "step": 162824} +{"lr": 0.0002140463001152259, "data_time": 0.0018907546997070312, "grad_norm": 0.0123816970793996, "loss": 0.6894800066947937, "time": 0.3972836256027222, "epoch": 15, "memory": 9465, "step": 162924} +{"lr": 0.0002140463001152259, "data_time": 0.0020686626434326173, "grad_norm": 0.024162659666035324, "loss": 0.6879245042800903, "time": 0.42353081703186035, "epoch": 15, "memory": 9465, "step": 163024} +{"lr": 0.0002140463001152259, "data_time": 0.001407027244567871, "grad_norm": 0.023543273913674057, "loss": 0.6884262084960937, "time": 0.40325064659118653, "epoch": 15, "memory": 9465, "step": 163124} +{"lr": 0.0002140463001152259, "data_time": 0.0013965606689453126, "grad_norm": 0.0197966959036421, "loss": 0.6834262669086456, "time": 0.4234490394592285, "epoch": 15, "memory": 9465, "step": 163224} +{"lr": 0.0002140463001152259, "data_time": 0.0013370275497436523, "grad_norm": 0.011397153418511153, "loss": 0.6868057489395142, "time": 0.4087916135787964, "epoch": 15, "memory": 9465, "step": 163324} +{"lr": 0.0002140463001152259, "data_time": 0.0018401861190795899, "grad_norm": 0.028872019331902265, "loss": 0.6871312022209167, "time": 0.416003680229187, "epoch": 15, "memory": 9465, "step": 163424} +{"lr": 0.0002140463001152259, "data_time": 0.0015646696090698242, "grad_norm": 0.01595398981589824, "loss": 0.6868103325366974, "time": 0.43205115795135496, "epoch": 15, "memory": 9465, "step": 163524} +{"lr": 0.0002140463001152259, "data_time": 0.0014375209808349609, "grad_norm": 0.01829653845052235, "loss": 0.6877792179584503, "time": 0.44249711036682127, "epoch": 15, "memory": 9465, "step": 163624} +{"lr": 0.0002140463001152259, "data_time": 0.0012975692749023437, "grad_norm": 0.01549928366439417, "loss": 0.6906494081020356, "time": 0.4179201364517212, "epoch": 15, "memory": 9465, "step": 163724} +{"lr": 0.0002140463001152259, "data_time": 0.0012684345245361328, "grad_norm": 0.02421788952779025, "loss": 0.6788218557834625, "time": 0.4103742837905884, "epoch": 15, "memory": 9465, "step": 163824} +{"lr": 0.0002140463001152259, "data_time": 0.0012929677963256837, "grad_norm": 0.026374685857445002, "loss": 0.6880486249923706, "time": 0.42886929512023925, "epoch": 15, "memory": 9465, "step": 163924} +{"lr": 0.0002140463001152259, "data_time": 0.0013208389282226562, "grad_norm": 0.022315075946971775, "loss": 0.6821879923343659, "time": 0.43700494766235354, "epoch": 15, "memory": 9465, "step": 164024} +{"lr": 0.0002140463001152259, "data_time": 0.001446366310119629, "grad_norm": 0.02220885206479579, "loss": 0.6825691878795623, "time": 0.41073644161224365, "epoch": 15, "memory": 9465, "step": 164124} +{"lr": 0.0002140463001152259, "data_time": 0.0013746976852416991, "grad_norm": 0.02066676509566605, "loss": 0.6900253236293793, "time": 0.3928713798522949, "epoch": 15, "memory": 9465, "step": 164224} +{"lr": 0.0002140463001152259, "data_time": 0.0013336658477783204, "grad_norm": 0.024910126929171383, "loss": 0.6860114872455597, "time": 0.43446879386901854, "epoch": 15, "memory": 9465, "step": 164324} +{"lr": 0.0002140463001152259, "data_time": 0.0027292728424072265, "grad_norm": 0.01571215141448192, "loss": 0.6840395569801331, "time": 0.4062842130661011, "epoch": 15, "memory": 9465, "step": 164424} +{"lr": 0.0002140463001152259, "data_time": 0.0014891862869262696, "grad_norm": 0.017370659043081105, "loss": 0.6879391372203827, "time": 0.42487130165100095, "epoch": 15, "memory": 9465, "step": 164524} +{"lr": 0.0002140463001152259, "data_time": 0.0029613494873046873, "grad_norm": 0.017539329291321338, "loss": 0.6863195359706878, "time": 0.39965410232543946, "epoch": 15, "memory": 9465, "step": 164624} +{"lr": 0.0002140463001152259, "data_time": 0.001259589195251465, "grad_norm": 0.032818767335265876, "loss": 0.6848550081253052, "time": 0.42869665622711184, "epoch": 15, "memory": 9465, "step": 164724} +{"lr": 0.0002140463001152259, "data_time": 0.0014023780822753906, "grad_norm": 0.021529263840056954, "loss": 0.6866447985172272, "time": 0.40840222835540774, "epoch": 15, "memory": 9465, "step": 164824} +{"lr": 0.0002140463001152259, "data_time": 0.0013232231140136719, "grad_norm": 0.015167720953468234, "loss": 0.6840157330036163, "time": 0.45548672676086427, "epoch": 15, "memory": 9465, "step": 164924} +{"lr": 0.0002140463001152259, "data_time": 0.002181410789489746, "grad_norm": 0.02375341553706676, "loss": 0.6797486484050751, "time": 0.39582879543304444, "epoch": 15, "memory": 9465, "step": 165024} +{"lr": 0.0002140463001152259, "data_time": 0.0013155460357666016, "grad_norm": 0.027573473216034473, "loss": 0.6872960150241851, "time": 0.41104028224945066, "epoch": 15, "memory": 9465, "step": 165124} +{"lr": 0.0002140463001152259, "data_time": 0.0011765480041503907, "grad_norm": 0.01996021153172478, "loss": 0.6842414379119873, "time": 0.38449184894561766, "epoch": 15, "memory": 9465, "step": 165224} +{"lr": 0.0002140463001152259, "data_time": 0.0012840747833251954, "grad_norm": 0.026236360101029275, "loss": 0.6881677746772766, "time": 0.4240997552871704, "epoch": 15, "memory": 9465, "step": 165324} +{"lr": 0.0002140463001152259, "data_time": 0.001308298110961914, "grad_norm": 0.027417116984724998, "loss": 0.6869114696979522, "time": 0.4011900186538696, "epoch": 15, "memory": 9465, "step": 165424} +{"lr": 0.0002140463001152259, "data_time": 0.0013127326965332031, "grad_norm": 0.018860459560528398, "loss": 0.6868092477321625, "time": 0.4262763261795044, "epoch": 15, "memory": 9465, "step": 165524} +{"lr": 0.0002140463001152259, "data_time": 0.0012764692306518554, "grad_norm": 0.01959541466785595, "loss": 0.6908905982971192, "time": 0.4086129903793335, "epoch": 15, "memory": 9465, "step": 165624} +{"lr": 0.0002140463001152259, "data_time": 0.00145721435546875, "grad_norm": 0.021767177432775498, "loss": 0.6858685433864593, "time": 0.41839349269866943, "epoch": 15, "memory": 9465, "step": 165724} +{"lr": 0.0002140463001152259, "data_time": 0.0018210172653198241, "grad_norm": 0.028958979714661836, "loss": 0.688239985704422, "time": 0.38466658592224123, "epoch": 15, "memory": 9465, "step": 165824} +{"lr": 0.0002140463001152259, "data_time": 0.001928997039794922, "grad_norm": 0.022747311904095114, "loss": 0.6879335582256317, "time": 0.4168570518493652, "epoch": 15, "memory": 9465, "step": 165924} +{"lr": 0.0002140463001152259, "data_time": 0.0012707233428955079, "grad_norm": 0.017367703886702657, "loss": 0.6828688263893128, "time": 0.42338855266571046, "epoch": 15, "memory": 9465, "step": 166024} +{"lr": 0.0002140463001152259, "data_time": 0.0013082027435302734, "grad_norm": 0.022956931637600063, "loss": 0.6796342492103576, "time": 0.4093561887741089, "epoch": 15, "memory": 9465, "step": 166124} +{"lr": 0.0002140463001152259, "data_time": 0.0011998414993286133, "grad_norm": 0.023125998489558697, "loss": 0.6874800384044647, "time": 0.40087432861328126, "epoch": 15, "memory": 9465, "step": 166224} +{"lr": 0.0002140463001152259, "data_time": 0.0021225690841674806, "grad_norm": 0.013915580738103017, "loss": 0.6853613078594207, "time": 0.4467417478561401, "epoch": 15, "memory": 9465, "step": 166324} +{"lr": 0.0002140463001152259, "data_time": 0.0014504432678222657, "grad_norm": 0.01924175429157913, "loss": 0.6847457170486451, "time": 0.4396371841430664, "epoch": 15, "memory": 9465, "step": 166424} +{"lr": 0.0002140463001152259, "data_time": 0.001787424087524414, "grad_norm": 0.01769362259947229, "loss": 0.6888521194458008, "time": 0.4142194986343384, "epoch": 15, "memory": 9465, "step": 166524} +{"lr": 0.0002140463001152259, "data_time": 0.0025406122207641602, "grad_norm": 0.017566913715563716, "loss": 0.6883926689624786, "time": 0.4275252342224121, "epoch": 15, "memory": 9465, "step": 166624} +{"lr": 0.0002140463001152259, "data_time": 0.0015596628189086914, "grad_norm": 0.031176075522671452, "loss": 0.6915314674377442, "time": 0.4031976699829102, "epoch": 15, "memory": 9465, "step": 166724} +{"lr": 0.0002140463001152259, "data_time": 0.0013432025909423829, "grad_norm": 0.02114136191084981, "loss": 0.6833634734153747, "time": 0.3974787473678589, "epoch": 15, "memory": 9465, "step": 166824} +{"lr": 0.0002140463001152259, "data_time": 0.0018229961395263671, "grad_norm": 0.013314515765523538, "loss": 0.6885230422019959, "time": 0.39331295490264895, "epoch": 15, "memory": 9465, "step": 166924} +{"lr": 0.0002140463001152259, "data_time": 0.0013289451599121094, "grad_norm": 0.025387100083753465, "loss": 0.6860304713249207, "time": 0.42376022338867186, "epoch": 15, "memory": 9465, "step": 167024} +{"lr": 0.0002140463001152259, "data_time": 0.0013043403625488282, "grad_norm": 0.020605178590631112, "loss": 0.6824390769004822, "time": 0.41295361518859863, "epoch": 15, "memory": 9465, "step": 167124} +{"lr": 0.0002140463001152259, "data_time": 0.0016310691833496093, "grad_norm": 0.025282455259002745, "loss": 0.686529690027237, "time": 0.4085672855377197, "epoch": 15, "memory": 9465, "step": 167224} +{"lr": 0.0002140463001152259, "data_time": 0.001294088363647461, "grad_norm": 0.012749508069828153, "loss": 0.6860068261623382, "time": 0.39381725788116456, "epoch": 15, "memory": 9465, "step": 167324} +{"lr": 0.0002140463001152259, "data_time": 0.0013279914855957031, "grad_norm": 0.01689598455559462, "loss": 0.67860706448555, "time": 0.43148863315582275, "epoch": 15, "memory": 9465, "step": 167424} +{"lr": 0.0002140463001152259, "data_time": 0.0013501167297363282, "grad_norm": 0.02227825871377718, "loss": 0.6852222859859467, "time": 0.40629117488861083, "epoch": 15, "memory": 9465, "step": 167524} +{"lr": 0.0002140463001152259, "data_time": 0.002580714225769043, "grad_norm": 0.027043653256259858, "loss": 0.6815916717052459, "time": 0.41361634731292723, "epoch": 15, "memory": 9465, "step": 167624} +{"lr": 0.0002140463001152259, "data_time": 0.0024840593338012694, "grad_norm": 0.022307089995592833, "loss": 0.6874937653541565, "time": 0.38384356498718264, "epoch": 15, "memory": 9465, "step": 167724} +{"lr": 0.0002140463001152259, "data_time": 0.0017174482345581055, "grad_norm": 0.02461586700519547, "loss": 0.6910554945468903, "time": 0.4112581729888916, "epoch": 15, "memory": 9465, "step": 167824} +{"lr": 0.0002140463001152259, "data_time": 0.0015367031097412109, "grad_norm": 0.020715971733443438, "loss": 0.6877492725849151, "time": 0.4556956052780151, "epoch": 15, "memory": 9465, "step": 167924} +{"lr": 0.0002140463001152259, "data_time": 0.0015857934951782227, "grad_norm": 0.01240457467501983, "loss": 0.6887055158615112, "time": 0.4200615406036377, "epoch": 15, "memory": 9465, "step": 168024} +{"lr": 0.0002140463001152259, "data_time": 0.0027471303939819334, "grad_norm": 0.025396501854993404, "loss": 0.686190128326416, "time": 0.4084672451019287, "epoch": 15, "memory": 9465, "step": 168124} +{"lr": 0.0002140463001152259, "data_time": 0.001683354377746582, "grad_norm": 0.023855562787503004, "loss": 0.6874362707138062, "time": 0.4132983207702637, "epoch": 15, "memory": 9465, "step": 168224} +{"lr": 0.0002140463001152259, "data_time": 0.0020859479904174806, "grad_norm": 0.022987095126882195, "loss": 0.6872956335544587, "time": 0.3963721990585327, "epoch": 15, "memory": 9465, "step": 168324} +{"lr": 0.0002140463001152259, "data_time": 0.0013257503509521485, "grad_norm": 0.028298271214589478, "loss": 0.6857425749301911, "time": 0.41259739398956297, "epoch": 15, "memory": 9465, "step": 168424} +{"lr": 0.0002140463001152259, "data_time": 0.0013125419616699218, "grad_norm": 0.030274981260299684, "loss": 0.6908557057380676, "time": 0.41288840770721436, "epoch": 15, "memory": 9465, "step": 168524} +{"lr": 0.0002140463001152259, "data_time": 0.0017639636993408204, "grad_norm": 0.02653438232373446, "loss": 0.6908407628536224, "time": 0.41278464794158937, "epoch": 15, "memory": 9465, "step": 168624} +{"lr": 0.0002140463001152259, "data_time": 0.0012291669845581055, "grad_norm": 0.022173001198098063, "loss": 0.6893778920173645, "time": 0.42564690113067627, "epoch": 15, "memory": 9465, "step": 168724} +{"lr": 0.0002140463001152259, "data_time": 0.001319289207458496, "grad_norm": 0.02520542065612972, "loss": 0.6874444842338562, "time": 0.42061102390289307, "epoch": 15, "memory": 9465, "step": 168824} +{"lr": 0.0002140463001152259, "data_time": 0.0014426231384277344, "grad_norm": 0.02720923819579184, "loss": 0.685785847902298, "time": 0.408914852142334, "epoch": 15, "memory": 9465, "step": 168924} +{"lr": 0.0002140463001152259, "data_time": 0.0013175010681152344, "grad_norm": 0.016094760701525955, "loss": 0.6829691350460052, "time": 0.454237961769104, "epoch": 15, "memory": 9465, "step": 169024} +{"lr": 0.0002140463001152259, "data_time": 0.0018984317779541016, "grad_norm": 0.01600051177665591, "loss": 0.6846305847167968, "time": 0.445867919921875, "epoch": 15, "memory": 9465, "step": 169124} +{"lr": 0.0002140463001152259, "data_time": 0.0013016939163208007, "grad_norm": 0.020676075041410513, "loss": 0.6891857981681824, "time": 0.43398048877716067, "epoch": 15, "memory": 9465, "step": 169224} +{"lr": 0.0002140463001152259, "data_time": 0.0014240503311157226, "grad_norm": 0.014566304977051914, "loss": 0.6912540793418884, "time": 0.39125206470489504, "epoch": 15, "memory": 9465, "step": 169324} +{"lr": 0.0002140463001152259, "data_time": 0.0013763904571533203, "grad_norm": 0.023430117953103034, "loss": 0.6863320827484131, "time": 0.4065514087677002, "epoch": 15, "memory": 9465, "step": 169424} +{"lr": 0.0002140463001152259, "data_time": 0.0014593124389648438, "grad_norm": 0.018793767696479335, "loss": 0.6831027686595916, "time": 0.4316640615463257, "epoch": 15, "memory": 9465, "step": 169524} +{"lr": 0.0002140463001152259, "data_time": 0.0013422250747680664, "grad_norm": 0.03210939399432391, "loss": 0.6876102387905121, "time": 0.45074079036712644, "epoch": 15, "memory": 9465, "step": 169624} +{"lr": 0.0002140463001152259, "data_time": 0.0014653682708740234, "grad_norm": 0.024660276668146252, "loss": 0.6877603650093078, "time": 0.4192806243896484, "epoch": 15, "memory": 9465, "step": 169724} +{"lr": 0.0002140463001152259, "data_time": 0.0013684511184692382, "grad_norm": 0.021273887343704702, "loss": 0.6850216567516327, "time": 0.4201963901519775, "epoch": 15, "memory": 9465, "step": 169824} +{"lr": 0.0002140463001152259, "data_time": 0.0016198158264160156, "grad_norm": 0.027001613425090908, "loss": 0.6882172286510467, "time": 0.41813514232635496, "epoch": 15, "memory": 9465, "step": 169924} +{"lr": 0.0002140463001152259, "data_time": 0.0012517929077148437, "grad_norm": 0.022277281119022517, "loss": 0.6907564401626587, "time": 0.4282623291015625, "epoch": 15, "memory": 9465, "step": 170024} +{"lr": 0.0002140463001152259, "data_time": 0.0013345956802368164, "grad_norm": 0.021735110017471016, "loss": 0.686005711555481, "time": 0.4130054235458374, "epoch": 15, "memory": 9465, "step": 170124} +{"lr": 0.0002140463001152259, "data_time": 0.001416921615600586, "grad_norm": 0.024807228142162786, "loss": 0.685529911518097, "time": 0.40968310832977295, "epoch": 15, "memory": 9465, "step": 170224} +{"lr": 0.0002140463001152259, "data_time": 0.0012783050537109376, "grad_norm": 0.024404172063805164, "loss": 0.6864898681640625, "time": 0.4274731159210205, "epoch": 15, "memory": 9465, "step": 170324} +{"lr": 0.0002140463001152259, "data_time": 0.001424098014831543, "grad_norm": 0.02463686892297119, "loss": 0.6869766891002655, "time": 0.4240144968032837, "epoch": 15, "memory": 9465, "step": 170424} +{"lr": 0.0002140463001152259, "data_time": 0.0018577814102172852, "grad_norm": 0.011882529768627137, "loss": 0.67880619764328, "time": 0.4308481216430664, "epoch": 15, "memory": 9465, "step": 170524} +{"lr": 0.0002140463001152259, "data_time": 0.001288437843322754, "grad_norm": 0.02491335610393435, "loss": 0.68327676653862, "time": 0.4180778741836548, "epoch": 15, "memory": 9465, "step": 170624} +{"lr": 0.0002140463001152259, "data_time": 0.0013918161392211914, "grad_norm": 0.011091677920194342, "loss": 0.6901532351970673, "time": 0.4272172212600708, "epoch": 15, "memory": 9465, "step": 170724} +{"lr": 0.0002140463001152259, "data_time": 0.0015508890151977538, "grad_norm": 0.020376018714159727, "loss": 0.6868217885494232, "time": 0.4707669496536255, "epoch": 15, "memory": 9465, "step": 170824} +{"lr": 0.0002140463001152259, "data_time": 0.0017530679702758788, "grad_norm": 0.027968662232160567, "loss": 0.6947459578514099, "time": 0.4309210300445557, "epoch": 15, "memory": 9465, "step": 170924} +{"lr": 0.0002140463001152259, "data_time": 0.0017899513244628907, "grad_norm": 0.021527378889732062, "loss": 0.6858453929424286, "time": 0.4314690828323364, "epoch": 15, "memory": 9465, "step": 171024} +{"lr": 0.0002140463001152259, "data_time": 0.0014004468917846679, "grad_norm": 0.019454246468376367, "loss": 0.687979382276535, "time": 0.43224637508392333, "epoch": 15, "memory": 9465, "step": 171124} +{"lr": 0.0002140463001152259, "data_time": 0.001273179054260254, "grad_norm": 0.018106868490576744, "loss": 0.6844498813152313, "time": 0.4359687328338623, "epoch": 15, "memory": 9465, "step": 171224} +{"lr": 0.0002140463001152259, "data_time": 0.0019911527633666992, "grad_norm": 0.017930368497036397, "loss": 0.6864727973937989, "time": 0.4729830503463745, "epoch": 15, "memory": 9465, "step": 171324} +{"lr": 0.0002140463001152259, "data_time": 0.001212763786315918, "grad_norm": 0.022242502425797284, "loss": 0.6863435804843903, "time": 0.45276756286621095, "epoch": 15, "memory": 9465, "step": 171424} +{"lr": 0.0002140463001152259, "data_time": 0.0013869047164916993, "grad_norm": 0.016562653821893036, "loss": 0.6861638486385345, "time": 0.43400847911834717, "epoch": 15, "memory": 9465, "step": 171524} +{"lr": 0.0002140463001152259, "data_time": 0.0017216920852661133, "grad_norm": 0.022390550188720226, "loss": 0.6856852114200592, "time": 0.4402602195739746, "epoch": 15, "memory": 9465, "step": 171624} +{"lr": 0.0002140463001152259, "data_time": 0.0013005971908569337, "grad_norm": 0.018840618315152824, "loss": 0.6884231984615325, "time": 0.4193726539611816, "epoch": 15, "memory": 9465, "step": 171724} +{"lr": 0.0002140463001152259, "data_time": 0.0012496471405029296, "grad_norm": 0.021178130363114177, "loss": 0.6872885525226593, "time": 0.40828404426574705, "epoch": 15, "memory": 9465, "step": 171824} +{"lr": 0.0002140463001152259, "data_time": 0.0014006614685058594, "grad_norm": 0.024584313973900863, "loss": 0.6844365000724792, "time": 0.45669419765472413, "epoch": 15, "memory": 9465, "step": 171924} +{"lr": 0.0002140463001152259, "data_time": 0.001261591911315918, "grad_norm": 0.017186284391209485, "loss": 0.6858316779136657, "time": 0.42757914066314695, "epoch": 15, "memory": 9465, "step": 172024} +{"lr": 0.0002140463001152259, "data_time": 0.0011881351470947265, "grad_norm": 0.01898417028714903, "loss": 0.6891905903816223, "time": 0.42606825828552247, "epoch": 15, "memory": 9465, "step": 172124} +{"lr": 0.0002140463001152259, "data_time": 0.0014975786209106446, "grad_norm": 0.026393118500709533, "loss": 0.6917441785335541, "time": 0.40974161624908445, "epoch": 15, "memory": 9465, "step": 172224} +{"lr": 0.0002140463001152259, "data_time": 0.0013780593872070312, "grad_norm": 0.0215091391466558, "loss": 0.6942771732807159, "time": 0.42108609676361086, "epoch": 15, "memory": 9465, "step": 172324} +{"lr": 0.0002140463001152259, "data_time": 0.00132443904876709, "grad_norm": 0.018597495416179298, "loss": 0.6829773545265198, "time": 0.42320957183837893, "epoch": 15, "memory": 9465, "step": 172424} +{"lr": 0.0002140463001152259, "data_time": 0.0012312889099121093, "grad_norm": 0.02508988493354991, "loss": 0.6888750612735748, "time": 0.39652304649353026, "epoch": 15, "memory": 9465, "step": 172524} +{"lr": 0.0002140463001152259, "data_time": 0.00141451358795166, "grad_norm": 0.03179497700184584, "loss": 0.6835142970085144, "time": 0.4224919557571411, "epoch": 15, "memory": 9465, "step": 172624} +{"lr": 0.0002140463001152259, "data_time": 0.005132699012756347, "grad_norm": 0.024911126727238296, "loss": 0.6877717256546021, "time": 0.42191898822784424, "epoch": 15, "memory": 9465, "step": 172724} +{"accuracy/top1": 100.0, "data_time": 0.2773989509133732, "time": 0.540760404923383, "step": 15} +{"lr": 0.00015498214331265904, "data_time": 0.04796533584594727, "grad_norm": 0.012402570992708205, "loss": 0.6874396622180938, "time": 0.42837932109832766, "epoch": 16, "memory": 9465, "step": 172840} +{"lr": 0.00015498214331265904, "data_time": 0.0013519287109375, "grad_norm": 0.026468004425987603, "loss": 0.6849124431610107, "time": 0.4147411584854126, "epoch": 16, "memory": 9465, "step": 172940} +{"lr": 0.00015498214331265904, "data_time": 0.0013030290603637696, "grad_norm": 0.021316367224790154, "loss": 0.6918719947338104, "time": 0.43860588073730467, "epoch": 16, "memory": 9465, "step": 173040} +{"lr": 0.00015498214331265904, "data_time": 0.0022736787796020508, "grad_norm": 0.016826165071688593, "loss": 0.6860176026821136, "time": 0.42089903354644775, "epoch": 16, "memory": 9465, "step": 173140} +{"lr": 0.00015498214331265904, "data_time": 0.0016296625137329102, "grad_norm": 0.020871743821771815, "loss": 0.6868066191673279, "time": 0.40884759426116946, "epoch": 16, "memory": 9465, "step": 173240} +{"lr": 0.00015498214331265904, "data_time": 0.0013618230819702148, "grad_norm": 0.031826614792225885, "loss": 0.6908757865428925, "time": 0.43290183544158933, "epoch": 16, "memory": 9465, "step": 173340} +{"lr": 0.00015498214331265904, "data_time": 0.0016141414642333984, "grad_norm": 0.03186066672205925, "loss": 0.6856403648853302, "time": 0.46909973621368406, "epoch": 16, "memory": 9465, "step": 173440} +{"lr": 0.00015498214331265904, "data_time": 0.0013225793838500977, "grad_norm": 0.025826570627395996, "loss": 0.6869646489620209, "time": 0.44034247398376464, "epoch": 16, "memory": 9465, "step": 173540} +{"lr": 0.00015498214331265904, "data_time": 0.0019741058349609375, "grad_norm": 0.01851236880174838, "loss": 0.6882412612438202, "time": 0.41535918712615966, "epoch": 16, "memory": 9465, "step": 173640} +{"lr": 0.00015498214331265904, "data_time": 0.0015933990478515625, "grad_norm": 0.029405278526246546, "loss": 0.684426361322403, "time": 0.3884117841720581, "epoch": 16, "memory": 9465, "step": 173740} +{"lr": 0.00015498214331265904, "data_time": 0.0013489246368408204, "grad_norm": 0.024283888889476656, "loss": 0.6869985878467559, "time": 0.3578831672668457, "epoch": 16, "memory": 9465, "step": 173840} +{"lr": 0.00015498214331265904, "data_time": 0.0014805555343627929, "grad_norm": 0.02539905053563416, "loss": 0.6883923649787903, "time": 0.3797483444213867, "epoch": 16, "memory": 9465, "step": 173940} +{"lr": 0.00015498214331265904, "data_time": 0.0012373924255371094, "grad_norm": 0.01833083345554769, "loss": 0.6885931849479675, "time": 0.37533481121063234, "epoch": 16, "memory": 9465, "step": 174040} +{"lr": 0.00015498214331265904, "data_time": 0.0011394739151000977, "grad_norm": 0.016121975146234035, "loss": 0.6888462603092194, "time": 0.4132434368133545, "epoch": 16, "memory": 9465, "step": 174140} +{"lr": 0.00015498214331265904, "data_time": 0.0015220403671264648, "grad_norm": 0.02277687704190612, "loss": 0.6871201574802399, "time": 0.36899352073669434, "epoch": 16, "memory": 9465, "step": 174240} +{"lr": 0.00015498214331265904, "data_time": 0.0019240856170654296, "grad_norm": 0.01913366199587472, "loss": 0.6838581979274749, "time": 0.411507248878479, "epoch": 16, "memory": 9465, "step": 174340} +{"lr": 0.00015498214331265904, "data_time": 0.0014511585235595704, "grad_norm": 0.015426935977302492, "loss": 0.6843148469924927, "time": 0.3962219476699829, "epoch": 16, "memory": 9465, "step": 174440} +{"lr": 0.00015498214331265904, "data_time": 0.0014513731002807617, "grad_norm": 0.025891577964648603, "loss": 0.6853933334350586, "time": 0.4263045310974121, "epoch": 16, "memory": 9465, "step": 174540} +{"lr": 0.00015498214331265904, "data_time": 0.0012848854064941406, "grad_norm": 0.016526716179214418, "loss": 0.6849218487739563, "time": 0.3768187999725342, "epoch": 16, "memory": 9465, "step": 174640} +{"lr": 0.00015498214331265904, "data_time": 0.001394510269165039, "grad_norm": 0.019196415692567824, "loss": 0.6875948488712311, "time": 0.38160154819488523, "epoch": 16, "memory": 9465, "step": 174740} +{"lr": 0.00015498214331265904, "data_time": 0.0012467145919799806, "grad_norm": 0.019742982322350146, "loss": 0.68916255235672, "time": 0.41026127338409424, "epoch": 16, "memory": 9465, "step": 174840} +{"lr": 0.00015498214331265904, "data_time": 0.001456284523010254, "grad_norm": 0.032116131024667996, "loss": 0.6809656322002411, "time": 0.3904338598251343, "epoch": 16, "memory": 9465, "step": 174940} +{"lr": 0.00015498214331265904, "data_time": 0.0020202875137329103, "grad_norm": 0.023475692491047084, "loss": 0.6892821431159973, "time": 0.4652515411376953, "epoch": 16, "memory": 9465, "step": 175040} +{"lr": 0.00015498214331265904, "data_time": 0.001911473274230957, "grad_norm": 0.025645952252671123, "loss": 0.6807098090648651, "time": 0.39325723648071287, "epoch": 16, "memory": 9465, "step": 175140} +{"lr": 0.00015498214331265904, "data_time": 0.0021973848342895508, "grad_norm": 0.019104114378569646, "loss": 0.6858468115329742, "time": 0.4022464036941528, "epoch": 16, "memory": 9465, "step": 175240} +{"lr": 0.00015498214331265904, "data_time": 0.002050304412841797, "grad_norm": 0.018378050858154894, "loss": 0.6855072259902955, "time": 0.38553872108459475, "epoch": 16, "memory": 9465, "step": 175340} +{"lr": 0.00015498214331265904, "data_time": 0.001916360855102539, "grad_norm": 0.02121393126435578, "loss": 0.6814954698085784, "time": 0.3946491479873657, "epoch": 16, "memory": 9465, "step": 175440} +{"lr": 0.00015498214331265904, "data_time": 0.0015674114227294921, "grad_norm": 0.022905727766919882, "loss": 0.6902084887027741, "time": 0.3887176036834717, "epoch": 16, "memory": 9465, "step": 175540} +{"lr": 0.00015498214331265904, "data_time": 0.0019622564315795897, "grad_norm": 0.02395902179996483, "loss": 0.6900645136833191, "time": 0.3686830997467041, "epoch": 16, "memory": 9465, "step": 175640} +{"lr": 0.00015498214331265904, "data_time": 0.0019403457641601562, "grad_norm": 0.0249115219572559, "loss": 0.6885612308979034, "time": 0.38416781425476076, "epoch": 16, "memory": 9465, "step": 175740} +{"lr": 0.00015498214331265904, "data_time": 0.0014475345611572265, "grad_norm": 0.0305089369736379, "loss": 0.6847354888916015, "time": 0.40173022747039794, "epoch": 16, "memory": 9465, "step": 175840} +{"lr": 0.00015498214331265904, "data_time": 0.0013242483139038086, "grad_norm": 0.030720880068838597, "loss": 0.6836639523506165, "time": 0.4245003700256348, "epoch": 16, "memory": 9465, "step": 175940} +{"lr": 0.00015498214331265904, "data_time": 0.0015934944152832032, "grad_norm": 0.021535947010852395, "loss": 0.6820919990539551, "time": 0.3820359230041504, "epoch": 16, "memory": 9465, "step": 176040} +{"lr": 0.00015498214331265904, "data_time": 0.002273297309875488, "grad_norm": 0.027006560820154846, "loss": 0.6867980062961578, "time": 0.37694535255432127, "epoch": 16, "memory": 9465, "step": 176140} +{"lr": 0.00015498214331265904, "data_time": 0.0013098716735839844, "grad_norm": 0.01856318542268127, "loss": 0.689912486076355, "time": 0.397737455368042, "epoch": 16, "memory": 9465, "step": 176240} +{"lr": 0.00015498214331265904, "data_time": 0.0013541936874389648, "grad_norm": 0.020981867541559042, "loss": 0.6854382276535034, "time": 0.3872718095779419, "epoch": 16, "memory": 9465, "step": 176340} +{"lr": 0.00015498214331265904, "data_time": 0.0013765335083007813, "grad_norm": 0.026773387455614285, "loss": 0.6820409655570984, "time": 0.37392315864562986, "epoch": 16, "memory": 9465, "step": 176440} diff --git a/swin_tiny_4xb256_1e-3lr_5m/20230531_183311/vis_data/config.py b/swin_tiny_4xb256_1e-3lr_5m/20230531_183311/vis_data/config.py new file mode 100644 index 0000000000000000000000000000000000000000..2f2d04453ec2c0ed6c5c62055a8b360ce00c32d5 --- /dev/null +++ b/swin_tiny_4xb256_1e-3lr_5m/20230531_183311/vis_data/config.py @@ -0,0 +1,174 @@ +optim_wrapper = dict( + optimizer=dict( + type='AdamW', + lr=0.001, + weight_decay=0.05, + eps=1e-08, + betas=(0.9, 0.999), + _scope_='mmpretrain'), + paramwise_cfg=dict( + norm_decay_mult=0.0, + bias_decay_mult=0.0, + flat_decay_mult=0.0, + custom_keys=dict({ + '.absolute_pos_embed': dict(decay_mult=0.0), + '.relative_position_bias_table': dict(decay_mult=0.0) + })), + type='AmpOptimWrapper', + dtype='bfloat16', + clip_grad=dict(max_norm=5.0)) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=True, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=20, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=1024) +model = dict( + type='ImageClassifier', + backbone=dict( + type='SwinTransformer', arch='tiny', img_size=224, drop_path_rate=0.2), + neck=dict(type='GlobalAveragePooling'), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=768, + init_cfg=None, + loss=dict( + type='LabelSmoothLoss', label_smooth_val=0.1, mode='original'), + cal_acc=False), + init_cfg=[ + dict(type='TruncNormal', layer='Linear', std=0.02, bias=0.0), + dict(type='Constant', layer='LayerNorm', val=1.0, bias=0.0) + ], + train_cfg=None) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=128, + num_workers=10, + dataset=dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stablediffusionV1-5R2-dpmsolver-25-5m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/swin_tiny_4xb256_1e-3lr_5m' diff --git a/swin_tiny_4xb256_1e-3lr_5m/20230531_183311/vis_data/events.out.tfevents.1685529198.SH-IDC1-10-140-24-130.4097.0 b/swin_tiny_4xb256_1e-3lr_5m/20230531_183311/vis_data/events.out.tfevents.1685529198.SH-IDC1-10-140-24-130.4097.0 new file mode 100644 index 0000000000000000000000000000000000000000..cbb2bde26d53fa580fd717b34e5d0d797b5db7e9 --- /dev/null +++ b/swin_tiny_4xb256_1e-3lr_5m/20230531_183311/vis_data/events.out.tfevents.1685529198.SH-IDC1-10-140-24-130.4097.0 @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:f56e9289a83e4d1bbb57f3555b7852d958937066aeafb91922267385fffb12cc +size 1632377 diff --git a/swin_tiny_4xb256_1e-3lr_5m/20230531_183311/vis_data/scalars.json b/swin_tiny_4xb256_1e-3lr_5m/20230531_183311/vis_data/scalars.json new file mode 100644 index 0000000000000000000000000000000000000000..6d13eee61c9532cba47a6e8dc19473529cca92a2 --- /dev/null +++ b/swin_tiny_4xb256_1e-3lr_5m/20230531_183311/vis_data/scalars.json @@ -0,0 +1,1777 @@ +{"lr": 0.001, "data_time": 0.19160523414611816, "grad_norm": 0.8052142232656478, "loss": 0.6849520623683929, "time": 0.42688555717468263, "epoch": 1, "memory": 9474, "step": 100} +{"lr": 0.001, "data_time": 0.0009674787521362305, "grad_norm": 1.2080376029014588, "loss": 0.6878873109817505, "time": 0.4383848190307617, "epoch": 1, "memory": 9474, "step": 200} +{"lr": 0.001, "data_time": 0.001109480857849121, "grad_norm": 1.2378409653902054, "loss": 0.6797099351882935, "time": 0.4772242307662964, "epoch": 1, "memory": 9474, "step": 300} +{"lr": 0.001, "data_time": 0.001096630096435547, "grad_norm": 0.7535933956503869, "loss": 0.6639681220054626, "time": 0.4896991729736328, "epoch": 1, "memory": 9474, "step": 400} +{"lr": 0.001, "data_time": 0.0010877132415771484, "grad_norm": 2.427699086070061, "loss": 0.672640472650528, "time": 0.49225332736968996, "epoch": 1, "memory": 9474, "step": 500} +{"lr": 0.001, "data_time": 0.0009778976440429688, "grad_norm": 0.8477599889039993, "loss": 0.6810718059539795, "time": 0.516904878616333, "epoch": 1, "memory": 9474, "step": 600} +{"lr": 0.001, "data_time": 0.001172924041748047, "grad_norm": 0.7502188369631767, "loss": 0.6687590122222901, "time": 0.48300533294677733, "epoch": 1, "memory": 9474, "step": 700} +{"lr": 0.001, "data_time": 0.0010997772216796875, "grad_norm": 0.6409901082515717, "loss": 0.6691092848777771, "time": 0.4418308973312378, "epoch": 1, "memory": 9474, "step": 800} +{"lr": 0.001, "data_time": 0.0010256052017211914, "grad_norm": 1.9922522366046906, "loss": 0.6726115763187408, "time": 0.42144126892089845, "epoch": 1, "memory": 9474, "step": 900} +{"lr": 0.001, "data_time": 0.001267242431640625, "grad_norm": 0.9892641425132751, "loss": 0.6661223530769348, "time": 0.4637609958648682, "epoch": 1, "memory": 9474, "step": 1000} +{"lr": 0.001, "data_time": 0.0010303974151611329, "grad_norm": 8.305355420708656, "loss": 0.6419571578502655, "time": 0.46142849922180174, "epoch": 1, "memory": 9474, "step": 1100} +{"lr": 0.001, "data_time": 0.0010545492172241212, "grad_norm": 0.792270827293396, "loss": 0.6670249342918396, "time": 0.43836815357208253, "epoch": 1, "memory": 9474, "step": 1200} +{"lr": 0.001, "data_time": 0.001044297218322754, "grad_norm": 0.6781573936343193, "loss": 0.6480621933937073, "time": 0.38781630992889404, "epoch": 1, "memory": 9474, "step": 1300} +{"lr": 0.001, "data_time": 0.0010371446609497071, "grad_norm": 0.5881157577037811, "loss": 0.6777416348457337, "time": 0.46243746280670167, "epoch": 1, "memory": 9474, "step": 1400} +{"lr": 0.001, "data_time": 0.0010274648666381836, "grad_norm": 0.5483857661485672, "loss": 0.6642019152641296, "time": 0.4059020519256592, "epoch": 1, "memory": 9474, "step": 1500} +{"lr": 0.001, "data_time": 0.0010054826736450196, "grad_norm": 0.5723021060228348, "loss": 0.6534375131130219, "time": 0.4653234243392944, "epoch": 1, "memory": 9474, "step": 1600} +{"lr": 0.001, "data_time": 0.0009948253631591798, "grad_norm": 1.8854305177927018, "loss": 0.6609666228294373, "time": 0.4095946788787842, "epoch": 1, "memory": 9474, "step": 1700} +{"lr": 0.001, "data_time": 0.001050853729248047, "grad_norm": 3.300791251659393, "loss": 0.6572105407714843, "time": 0.516826057434082, "epoch": 1, "memory": 9474, "step": 1800} +{"lr": 0.001, "data_time": 0.0009797334671020508, "grad_norm": 3.2456099927425384, "loss": 0.6459945976734162, "time": 0.44674875736236574, "epoch": 1, "memory": 9474, "step": 1900} +{"lr": 0.001, "data_time": 0.0009960412979125976, "grad_norm": 2.9178244560956954, "loss": 0.6701214969158172, "time": 0.48870298862457273, "epoch": 1, "memory": 9474, "step": 2000} +{"lr": 0.001, "data_time": 0.0010712385177612306, "grad_norm": 24.649349164962768, "loss": 0.6704214990139008, "time": 0.6039324760437011, "epoch": 1, "memory": 9474, "step": 2100} +{"lr": 0.001, "data_time": 0.0009820938110351562, "grad_norm": 132.87631206512452, "loss": 0.6618600070476532, "time": 0.5378271341323853, "epoch": 1, "memory": 9474, "step": 2200} +{"lr": 0.001, "data_time": 0.0009410858154296875, "grad_norm": 0.5663121178746223, "loss": 0.6898861110210419, "time": 0.4891343355178833, "epoch": 1, "memory": 9474, "step": 2300} +{"lr": 0.001, "data_time": 0.0009967803955078125, "grad_norm": 0.48270824924111366, "loss": 0.6908625543117524, "time": 0.5455003023147583, "epoch": 1, "memory": 9474, "step": 2400} +{"lr": 0.001, "data_time": 0.0009343624114990234, "grad_norm": 0.22726595103740693, "loss": 0.6804853200912475, "time": 0.4412868022918701, "epoch": 1, "memory": 9474, "step": 2500} +{"lr": 0.001, "data_time": 0.0009561061859130859, "grad_norm": 0.4640265181660652, "loss": 0.6885727345943451, "time": 0.45170691013336184, "epoch": 1, "memory": 9474, "step": 2600} +{"lr": 0.001, "data_time": 0.0010427713394165039, "grad_norm": 0.42788181230425837, "loss": 0.6879625499248505, "time": 0.42471895217895506, "epoch": 1, "memory": 9474, "step": 2700} +{"lr": 0.001, "data_time": 0.0010392904281616212, "grad_norm": 0.5726554363965988, "loss": 0.6858113765716553, "time": 0.48532853126525877, "epoch": 1, "memory": 9474, "step": 2800} +{"lr": 0.001, "data_time": 0.0009303569793701171, "grad_norm": 0.40351647436618804, "loss": 0.6893806636333466, "time": 0.521523904800415, "epoch": 1, "memory": 9474, "step": 2900} +{"lr": 0.001, "data_time": 0.0011853218078613282, "grad_norm": 0.16025364845991136, "loss": 0.6835512280464172, "time": 0.4506367683410645, "epoch": 1, "memory": 9474, "step": 3000} +{"lr": 0.001, "data_time": 0.0008364439010620118, "grad_norm": 0.24458550326526166, "loss": 0.6849212646484375, "time": 0.5652475595474243, "epoch": 1, "memory": 9474, "step": 3100} +{"lr": 0.001, "data_time": 0.0009660482406616211, "grad_norm": 0.44269598126411436, "loss": 0.6895747065544129, "time": 0.48994646072387693, "epoch": 1, "memory": 9474, "step": 3200} +{"lr": 0.001, "data_time": 0.0009510278701782227, "grad_norm": 0.24424223899841307, "loss": 0.6918225228786469, "time": 0.47918052673339845, "epoch": 1, "memory": 9474, "step": 3300} +{"lr": 0.001, "data_time": 0.0009103775024414063, "grad_norm": 0.29042947441339495, "loss": 0.6890306532382965, "time": 0.5593473196029664, "epoch": 1, "memory": 9474, "step": 3400} +{"lr": 0.001, "data_time": 0.0010324478149414062, "grad_norm": 0.2884215623140335, "loss": 0.6909060895442962, "time": 0.4397998809814453, "epoch": 1, "memory": 9474, "step": 3500} +{"lr": 0.001, "data_time": 0.000988173484802246, "grad_norm": 0.3512984409928322, "loss": 0.6868723154067993, "time": 0.41014585494995115, "epoch": 1, "memory": 9474, "step": 3600} +{"lr": 0.001, "data_time": 0.001079702377319336, "grad_norm": 0.46717650145292283, "loss": 0.6889779210090637, "time": 0.48719546794891355, "epoch": 1, "memory": 9474, "step": 3700} +{"lr": 0.001, "data_time": 0.0009716510772705078, "grad_norm": 0.2242715172469616, "loss": 0.692361181974411, "time": 0.46064176559448244, "epoch": 1, "memory": 9474, "step": 3800} +{"lr": 0.001, "data_time": 0.0009479999542236328, "grad_norm": 0.24586505629122257, "loss": 0.6896000802516937, "time": 0.4327003717422485, "epoch": 1, "memory": 9474, "step": 3900} +{"lr": 0.001, "data_time": 0.0009173393249511719, "grad_norm": 0.18979598507285117, "loss": 0.6918147563934326, "time": 0.5554824829101562, "epoch": 1, "memory": 9474, "step": 4000} +{"lr": 0.001, "data_time": 0.0009491443634033203, "grad_norm": 0.23377782851457596, "loss": 0.6852491140365601, "time": 0.499322247505188, "epoch": 1, "memory": 9474, "step": 4100} +{"lr": 0.001, "data_time": 0.001265573501586914, "grad_norm": 0.39092123657464983, "loss": 0.6906617701053619, "time": 0.45052011013031007, "epoch": 1, "memory": 9474, "step": 4200} +{"lr": 0.001, "data_time": 0.0009168148040771485, "grad_norm": 0.2585465915501118, "loss": 0.6853789627552033, "time": 0.4613243818283081, "epoch": 1, "memory": 9474, "step": 4300} +{"lr": 0.001, "data_time": 0.0011272668838500977, "grad_norm": 0.2267835909500718, "loss": 0.6889314770698547, "time": 0.46185941696166993, "epoch": 1, "memory": 9474, "step": 4400} +{"lr": 0.001, "data_time": 0.001326441764831543, "grad_norm": 0.19596254006028174, "loss": 0.6883785903453827, "time": 0.5613390207290649, "epoch": 1, "memory": 9474, "step": 4500} +{"lr": 0.001, "data_time": 0.00107266902923584, "grad_norm": 0.20727041438221933, "loss": 0.6849764704704284, "time": 0.4139907121658325, "epoch": 1, "memory": 9474, "step": 4600} +{"lr": 0.001, "data_time": 0.0011577367782592773, "grad_norm": 0.23133864887058736, "loss": 0.6821418702602386, "time": 0.4663443326950073, "epoch": 1, "memory": 9474, "step": 4700} +{"lr": 0.001, "data_time": 0.0009682178497314453, "grad_norm": 0.18215838652104138, "loss": 0.6857521176338196, "time": 0.47600295543670657, "epoch": 1, "memory": 9474, "step": 4800} +{"lr": 0.001, "data_time": 0.0009597539901733398, "grad_norm": 0.17026583850383759, "loss": 0.6810456097126008, "time": 0.5322046995162963, "epoch": 1, "memory": 9474, "step": 4900} +{"lr": 0.001, "data_time": 0.0009952545166015624, "grad_norm": 0.19457727447152137, "loss": 0.6891527593135833, "time": 0.4480944871902466, "epoch": 1, "memory": 9474, "step": 5000} +{"lr": 0.001, "data_time": 0.0010361194610595704, "grad_norm": 0.2026959676295519, "loss": 0.6932987153530121, "time": 0.5342537641525269, "epoch": 1, "memory": 9474, "step": 5100} +{"lr": 0.001, "data_time": 0.0010344743728637694, "grad_norm": 0.19610126949846746, "loss": 0.6837357580661774, "time": 0.5068094491958618, "epoch": 1, "memory": 9474, "step": 5200} +{"lr": 0.001, "data_time": 0.001011061668395996, "grad_norm": 0.21482629626989364, "loss": 0.6858660340309143, "time": 0.565664005279541, "epoch": 1, "memory": 9474, "step": 5300} +{"lr": 0.001, "data_time": 0.0009800195693969727, "grad_norm": 0.16851673275232315, "loss": 0.6887180030345916, "time": 0.4616303682327271, "epoch": 1, "memory": 9474, "step": 5400} +{"lr": 0.001, "data_time": 0.0010013818740844727, "grad_norm": 0.16333863157778977, "loss": 0.6855832993984222, "time": 0.42328202724456787, "epoch": 1, "memory": 9474, "step": 5500} +{"lr": 0.001, "data_time": 0.0010354280471801757, "grad_norm": 0.19897820930927992, "loss": 0.6884875357151031, "time": 0.47611050605773925, "epoch": 1, "memory": 9474, "step": 5600} +{"lr": 0.001, "data_time": 0.0010261058807373047, "grad_norm": 0.208826738409698, "loss": 0.6890342056751251, "time": 0.5138506174087525, "epoch": 1, "memory": 9474, "step": 5700} +{"lr": 0.001, "data_time": 0.0010219573974609374, "grad_norm": 0.12756846696138383, "loss": 0.6907239317893982, "time": 0.512112307548523, "epoch": 1, "memory": 9474, "step": 5800} +{"lr": 0.001, "data_time": 0.001003599166870117, "grad_norm": 0.22615315057337285, "loss": 0.693650609254837, "time": 0.4584911584854126, "epoch": 1, "memory": 9474, "step": 5900} +{"lr": 0.001, "data_time": 0.0009990692138671874, "grad_norm": 0.2532569320872426, "loss": 0.6887039959430694, "time": 0.43177931308746337, "epoch": 1, "memory": 9474, "step": 6000} +{"lr": 0.001, "data_time": 0.0009796857833862306, "grad_norm": 0.20852719470858574, "loss": 0.6801067411899566, "time": 0.415038013458252, "epoch": 1, "memory": 9474, "step": 6100} +{"lr": 0.001, "data_time": 0.0010406017303466798, "grad_norm": 0.22137072421610354, "loss": 0.6856032311916351, "time": 0.44321963787078855, "epoch": 1, "memory": 9474, "step": 6200} +{"lr": 0.001, "data_time": 0.0013756275177001954, "grad_norm": 0.14437412656843662, "loss": 0.6890648424625396, "time": 0.45041627883911134, "epoch": 1, "memory": 9474, "step": 6300} +{"lr": 0.001, "data_time": 0.0009638309478759765, "grad_norm": 0.11462613651528955, "loss": 0.6906584203243256, "time": 0.4679267406463623, "epoch": 1, "memory": 9474, "step": 6400} +{"lr": 0.001, "data_time": 0.0009244203567504883, "grad_norm": 0.18450075276196004, "loss": 0.683798748254776, "time": 0.4362278938293457, "epoch": 1, "memory": 9474, "step": 6500} +{"lr": 0.001, "data_time": 0.0010439395904541016, "grad_norm": 0.16256289929151535, "loss": 0.6911483645439148, "time": 0.4376434564590454, "epoch": 1, "memory": 9474, "step": 6600} +{"lr": 0.001, "data_time": 0.000995326042175293, "grad_norm": 0.16351005053147674, "loss": 0.6809832274913787, "time": 0.45746004581451416, "epoch": 1, "memory": 9474, "step": 6700} +{"lr": 0.001, "data_time": 0.0010692119598388673, "grad_norm": 0.16776065863668918, "loss": 0.6851230680942535, "time": 0.4315709352493286, "epoch": 1, "memory": 9474, "step": 6800} +{"lr": 0.001, "data_time": 0.0011238336563110351, "grad_norm": 0.18845577947795392, "loss": 0.6767730951309204, "time": 0.4457768678665161, "epoch": 1, "memory": 9474, "step": 6900} +{"lr": 0.001, "data_time": 0.001305699348449707, "grad_norm": 0.1759693343192339, "loss": 0.6886209487915039, "time": 0.4217818021774292, "epoch": 1, "memory": 9474, "step": 7000} +{"lr": 0.001, "data_time": 0.0010958433151245118, "grad_norm": 0.18210300225764514, "loss": 0.6890046715736389, "time": 0.5111133098602295, "epoch": 1, "memory": 9474, "step": 7100} +{"lr": 0.001, "data_time": 0.0012515783309936523, "grad_norm": 0.12574998512864113, "loss": 0.6871379852294922, "time": 0.4268744945526123, "epoch": 1, "memory": 9474, "step": 7200} +{"lr": 0.001, "data_time": 0.001128220558166504, "grad_norm": 0.16726765520870684, "loss": 0.6882547914981842, "time": 0.4326565027236938, "epoch": 1, "memory": 9474, "step": 7300} +{"lr": 0.001, "data_time": 0.0010713100433349609, "grad_norm": 0.199175663292408, "loss": 0.6863829076290131, "time": 0.4495775938034058, "epoch": 1, "memory": 9474, "step": 7400} +{"lr": 0.001, "data_time": 0.0010512113571166993, "grad_norm": 0.2270539177581668, "loss": 0.6883134067058563, "time": 0.42945754528045654, "epoch": 1, "memory": 9474, "step": 7500} +{"lr": 0.001, "data_time": 0.0009206295013427735, "grad_norm": 0.1085498021915555, "loss": 0.6849838256835937, "time": 0.4674868583679199, "epoch": 1, "memory": 9474, "step": 7600} +{"lr": 0.001, "data_time": 0.0010736703872680664, "grad_norm": 0.20984667725861073, "loss": 0.6875992119312286, "time": 0.41382741928100586, "epoch": 1, "memory": 9474, "step": 7700} +{"lr": 0.001, "data_time": 0.0010170459747314454, "grad_norm": 0.10837054271250963, "loss": 0.6887068629264832, "time": 0.45543999671936036, "epoch": 1, "memory": 9474, "step": 7800} +{"lr": 0.001, "data_time": 0.0009465694427490234, "grad_norm": 0.10927009116858244, "loss": 0.6847162783145905, "time": 0.5044364929199219, "epoch": 1, "memory": 9474, "step": 7900} +{"lr": 0.001, "data_time": 0.0009895086288452149, "grad_norm": 0.10070986961945891, "loss": 0.6876133859157563, "time": 0.4144383192062378, "epoch": 1, "memory": 9474, "step": 8000} +{"lr": 0.001, "data_time": 0.0009637355804443359, "grad_norm": 0.1331441055983305, "loss": 0.6843811333179474, "time": 0.392572808265686, "epoch": 1, "memory": 9474, "step": 8100} +{"lr": 0.001, "data_time": 0.0010759592056274413, "grad_norm": 0.13488234179094433, "loss": 0.6888718068599701, "time": 0.40422210693359373, "epoch": 1, "memory": 9474, "step": 8200} +{"lr": 0.001, "data_time": 0.001038360595703125, "grad_norm": 0.11280787996947765, "loss": 0.6876307129859924, "time": 0.39751129150390624, "epoch": 1, "memory": 9474, "step": 8300} +{"lr": 0.001, "data_time": 0.0009946107864379882, "grad_norm": 0.09878557398915291, "loss": 0.6860785484313965, "time": 0.4117953062057495, "epoch": 1, "memory": 9474, "step": 8400} +{"lr": 0.001, "data_time": 0.0009869575500488282, "grad_norm": 0.1379073467105627, "loss": 0.6873394250869751, "time": 0.43392760753631593, "epoch": 1, "memory": 9474, "step": 8500} +{"lr": 0.001, "data_time": 0.0008696079254150391, "grad_norm": 0.10519644804298878, "loss": 0.6859173238277435, "time": 0.4291120767593384, "epoch": 1, "memory": 9474, "step": 8600} +{"lr": 0.001, "data_time": 0.0010144472122192382, "grad_norm": 0.10368053894490004, "loss": 0.68602956533432, "time": 0.45687916278839114, "epoch": 1, "memory": 9474, "step": 8700} +{"lr": 0.001, "data_time": 0.0014067888259887695, "grad_norm": 0.08638579975813628, "loss": 0.6815131843090058, "time": 0.3762549877166748, "epoch": 1, "memory": 9474, "step": 8800} +{"lr": 0.001, "data_time": 0.0011425256729125977, "grad_norm": 0.08293162267655134, "loss": 0.6876282215118408, "time": 0.42606263160705565, "epoch": 1, "memory": 9474, "step": 8900} +{"lr": 0.001, "data_time": 0.001256704330444336, "grad_norm": 0.11535952351987362, "loss": 0.6890344023704529, "time": 0.4015597581863403, "epoch": 1, "memory": 9474, "step": 9000} +{"lr": 0.001, "data_time": 0.0011426925659179688, "grad_norm": 0.09358172118663788, "loss": 0.681056809425354, "time": 0.4113957643508911, "epoch": 1, "memory": 9474, "step": 9100} +{"lr": 0.001, "data_time": 0.0009740829467773437, "grad_norm": 0.08762898705899716, "loss": 0.6841725528240203, "time": 0.3920086145401001, "epoch": 1, "memory": 9474, "step": 9200} +{"lr": 0.001, "data_time": 0.001075434684753418, "grad_norm": 0.07830161657184362, "loss": 0.6841848134994507, "time": 0.43527817726135254, "epoch": 1, "memory": 9474, "step": 9300} +{"lr": 0.001, "data_time": 0.0010875463485717773, "grad_norm": 0.074044300429523, "loss": 0.6865354955196381, "time": 0.40725576877593994, "epoch": 1, "memory": 9474, "step": 9400} +{"lr": 0.001, "data_time": 0.0010560989379882813, "grad_norm": 0.08644570319447667, "loss": 0.6852303683757782, "time": 0.4142030715942383, "epoch": 1, "memory": 9474, "step": 9500} +{"lr": 0.001, "data_time": 0.0010829687118530274, "grad_norm": 0.08891921145841479, "loss": 0.6819977521896362, "time": 0.37514657974243165, "epoch": 1, "memory": 9474, "step": 9600} +{"lr": 0.001, "data_time": 0.001034379005432129, "grad_norm": 0.07139426893554628, "loss": 0.6931855022907257, "time": 0.409479546546936, "epoch": 1, "memory": 9474, "step": 9700} +{"lr": 0.001, "data_time": 0.0010714054107666016, "grad_norm": 0.09761939700692893, "loss": 0.6861530482769013, "time": 0.4011998176574707, "epoch": 1, "memory": 9474, "step": 9800} +{"lr": 0.001, "data_time": 0.0011259078979492187, "grad_norm": 0.10074464427307248, "loss": 0.6831479549407959, "time": 0.41072404384613037, "epoch": 1, "memory": 9474, "step": 9900} +{"lr": 0.001, "data_time": 0.0010134220123291016, "grad_norm": 0.06637590676546097, "loss": 0.6832959115505218, "time": 0.46042556762695314, "epoch": 1, "memory": 9474, "step": 10000} +{"lr": 0.001, "data_time": 0.0009308338165283203, "grad_norm": 0.1275673158466816, "loss": 0.6818589746952057, "time": 0.38405086994171145, "epoch": 1, "memory": 9474, "step": 10100} +{"lr": 0.001, "data_time": 0.0011119365692138672, "grad_norm": 0.06425417019054294, "loss": 0.6876189649105072, "time": 0.40998854637146, "epoch": 1, "memory": 9474, "step": 10200} +{"lr": 0.001, "data_time": 0.0010421037673950194, "grad_norm": 0.08715715687721967, "loss": 0.6832851946353913, "time": 0.3770972967147827, "epoch": 1, "memory": 9474, "step": 10300} +{"lr": 0.001, "data_time": 0.000909876823425293, "grad_norm": 0.11128245387226343, "loss": 0.6887672781944275, "time": 0.3994786262512207, "epoch": 1, "memory": 9474, "step": 10400} +{"lr": 0.001, "data_time": 0.000986933708190918, "grad_norm": 0.08399312794208527, "loss": 0.6866066873073577, "time": 0.39032630920410155, "epoch": 1, "memory": 9474, "step": 10500} +{"lr": 0.001, "data_time": 0.0012809038162231445, "grad_norm": 0.06851767767220736, "loss": 0.6826740264892578, "time": 0.40223581790924073, "epoch": 1, "memory": 9474, "step": 10600} +{"lr": 0.001, "data_time": 0.0009658575057983398, "grad_norm": 0.05832801419310272, "loss": 0.6870149672031403, "time": 0.39209136962890623, "epoch": 1, "memory": 9474, "step": 10700} +{"lr": 0.001, "data_time": 0.0009809494018554687, "grad_norm": 0.07795922216027976, "loss": 0.6847951591014863, "time": 0.4209475040435791, "epoch": 1, "memory": 9474, "step": 10800} +{"lr": 0.001, "data_time": 0.001141667366027832, "grad_norm": 0.06304550403729081, "loss": 0.685225784778595, "time": 0.3920806646347046, "epoch": 1, "memory": 9474, "step": 10900} +{"lr": 0.001, "data_time": 0.0009903907775878906, "grad_norm": 0.07760525858029724, "loss": 0.6900427520275116, "time": 0.4039088487625122, "epoch": 1, "memory": 9474, "step": 11000} +{"lr": 0.001, "data_time": 0.0010686635971069336, "grad_norm": 0.07758190110325813, "loss": 0.685785299539566, "time": 0.3837052583694458, "epoch": 1, "memory": 9474, "step": 11100} +{"lr": 0.001, "data_time": 0.001104593276977539, "grad_norm": 0.0721831951290369, "loss": 0.6898037552833557, "time": 0.39397780895233153, "epoch": 1, "memory": 9474, "step": 11200} +{"lr": 0.001, "data_time": 0.0010874032974243163, "grad_norm": 0.06284296195954084, "loss": 0.6920159876346588, "time": 0.42384178638458253, "epoch": 1, "memory": 9474, "step": 11300} +{"lr": 0.001, "data_time": 0.001458287239074707, "grad_norm": 0.08862904305569827, "loss": 0.6865630865097045, "time": 0.37363195419311523, "epoch": 1, "memory": 9474, "step": 11400} +{"lr": 0.001, "data_time": 0.008296585083007813, "grad_norm": 0.0653378666844219, "loss": 0.6869338035583497, "time": 0.4357378244400024, "epoch": 1, "memory": 9474, "step": 11500} +{"accuracy/top1": 100.0, "data_time": 0.309885635972023, "time": 0.6909352540969849, "step": 1} +{"lr": 0.0009939057285945933, "data_time": 0.07083356380462646, "grad_norm": 0.1099855212494731, "loss": 0.6879038155078888, "time": 0.46412322521209715, "epoch": 2, "memory": 9474, "step": 11616} +{"lr": 0.0009939057285945933, "data_time": 0.001030731201171875, "grad_norm": 0.06427400205284357, "loss": 0.6862953543663025, "time": 0.4322756290435791, "epoch": 2, "memory": 9466, "step": 11716} +{"lr": 0.0009939057285945933, "data_time": 0.0011673927307128905, "grad_norm": 0.0751484197564423, "loss": 0.6848814904689788, "time": 0.48997957706451417, "epoch": 2, "memory": 9466, "step": 11816} +{"lr": 0.0009939057285945933, "data_time": 0.0010225772857666016, "grad_norm": 0.06925361389294267, "loss": 0.6876432597637177, "time": 0.4027824878692627, "epoch": 2, "memory": 9466, "step": 11916} +{"lr": 0.0009939057285945933, "data_time": 0.0010874032974243163, "grad_norm": 0.06935103619471192, "loss": 0.684624993801117, "time": 0.40043728351593016, "epoch": 2, "memory": 9466, "step": 12016} +{"lr": 0.0009939057285945933, "data_time": 0.0010295867919921874, "grad_norm": 0.06250485554337501, "loss": 0.6886254191398621, "time": 0.43543667793273927, "epoch": 2, "memory": 9466, "step": 12116} +{"lr": 0.0009939057285945933, "data_time": 0.0012871026992797852, "grad_norm": 0.058108051493763926, "loss": 0.6871399939060211, "time": 0.504337215423584, "epoch": 2, "memory": 9466, "step": 12216} +{"lr": 0.0009939057285945933, "data_time": 0.0012480258941650391, "grad_norm": 0.06246477738022804, "loss": 0.689530348777771, "time": 0.39492509365081785, "epoch": 2, "memory": 9466, "step": 12316} +{"lr": 0.0009939057285945933, "data_time": 0.0010716438293457032, "grad_norm": 0.07895727716386318, "loss": 0.6859519243240356, "time": 0.46003384590148927, "epoch": 2, "memory": 9466, "step": 12416} +{"lr": 0.0009939057285945933, "data_time": 0.001078963279724121, "grad_norm": 0.06565370513126254, "loss": 0.6878838241100311, "time": 0.46222763061523436, "epoch": 2, "memory": 9466, "step": 12516} +{"lr": 0.0009939057285945933, "data_time": 0.0011182785034179687, "grad_norm": 0.06038331417366862, "loss": 0.6912600874900818, "time": 0.473629093170166, "epoch": 2, "memory": 9466, "step": 12616} +{"lr": 0.0009939057285945933, "data_time": 0.0009978771209716796, "grad_norm": 0.05918410774320364, "loss": 0.690394276380539, "time": 0.47400593757629395, "epoch": 2, "memory": 9466, "step": 12716} +{"lr": 0.0009939057285945933, "data_time": 0.0011910438537597657, "grad_norm": 0.05964112002402544, "loss": 0.6870898425579071, "time": 0.4028794765472412, "epoch": 2, "memory": 9466, "step": 12816} +{"lr": 0.0009939057285945933, "data_time": 0.001068282127380371, "grad_norm": 0.0650144257582724, "loss": 0.6826631903648377, "time": 0.46787664890289304, "epoch": 2, "memory": 9466, "step": 12916} +{"lr": 0.0009939057285945933, "data_time": 0.0010412216186523437, "grad_norm": 0.043797826487571, "loss": 0.688552850484848, "time": 0.3952311992645264, "epoch": 2, "memory": 9466, "step": 13016} +{"lr": 0.0009939057285945933, "data_time": 0.001025557518005371, "grad_norm": 0.03829173250123859, "loss": 0.6879911184310913, "time": 0.5633004903793335, "epoch": 2, "memory": 9466, "step": 13116} +{"lr": 0.0009939057285945933, "data_time": 0.0010823249816894532, "grad_norm": 0.06581303123384714, "loss": 0.6825634062290191, "time": 0.45997314453125, "epoch": 2, "memory": 9466, "step": 13216} +{"lr": 0.0009939057285945933, "data_time": 0.0011197805404663086, "grad_norm": 0.08384905718266963, "loss": 0.6832798004150391, "time": 0.4203648090362549, "epoch": 2, "memory": 9466, "step": 13316} +{"lr": 0.0009939057285945933, "data_time": 0.001103353500366211, "grad_norm": 0.04816665789112449, "loss": 0.6857930243015289, "time": 0.43519017696380613, "epoch": 2, "memory": 9466, "step": 13416} +{"lr": 0.0009939057285945933, "data_time": 0.001039886474609375, "grad_norm": 0.05029902392998338, "loss": 0.6834621250629425, "time": 0.42972192764282224, "epoch": 2, "memory": 9466, "step": 13516} +{"lr": 0.0009939057285945933, "data_time": 0.0011239051818847656, "grad_norm": 0.04806776299374178, "loss": 0.6888626396656037, "time": 0.39881505966186526, "epoch": 2, "memory": 9466, "step": 13616} +{"lr": 0.0009939057285945933, "data_time": 0.001100301742553711, "grad_norm": 0.047950968914665285, "loss": 0.6914717495441437, "time": 0.4222588539123535, "epoch": 2, "memory": 9466, "step": 13716} +{"lr": 0.0009939057285945933, "data_time": 0.0010953187942504884, "grad_norm": 0.03864505849778652, "loss": 0.687501072883606, "time": 0.40851192474365233, "epoch": 2, "memory": 9466, "step": 13816} +{"lr": 0.0009939057285945933, "data_time": 0.0012400388717651368, "grad_norm": 0.03776288768276572, "loss": 0.6899020135402679, "time": 0.48978049755096437, "epoch": 2, "memory": 9466, "step": 13916} +{"lr": 0.0009939057285945933, "data_time": 0.0010712862014770508, "grad_norm": 0.06297209057956935, "loss": 0.6826109707355499, "time": 0.42327046394348145, "epoch": 2, "memory": 9466, "step": 14016} +{"lr": 0.0009939057285945933, "data_time": 0.0010754823684692382, "grad_norm": 0.0396063226275146, "loss": 0.6863962829113006, "time": 0.4419059753417969, "epoch": 2, "memory": 9466, "step": 14116} +{"lr": 0.0009939057285945933, "data_time": 0.0011355876922607422, "grad_norm": 0.05816199849359691, "loss": 0.686329436302185, "time": 0.43482882976531984, "epoch": 2, "memory": 9466, "step": 14216} +{"lr": 0.0009939057285945933, "data_time": 0.0011022090911865234, "grad_norm": 0.03832084927707911, "loss": 0.690645182132721, "time": 0.47258481979370115, "epoch": 2, "memory": 9466, "step": 14316} +{"lr": 0.0009939057285945933, "data_time": 0.0010602235794067382, "grad_norm": 0.052494379319250585, "loss": 0.6814186632633209, "time": 0.45048458576202394, "epoch": 2, "memory": 9466, "step": 14416} +{"lr": 0.0009939057285945933, "data_time": 0.0009874343872070313, "grad_norm": 0.0462041768245399, "loss": 0.6880272209644318, "time": 0.4128320932388306, "epoch": 2, "memory": 9466, "step": 14516} +{"lr": 0.0009939057285945933, "data_time": 0.001046442985534668, "grad_norm": 0.0523037008009851, "loss": 0.685356217622757, "time": 0.4169517993927002, "epoch": 2, "memory": 9466, "step": 14616} +{"lr": 0.0009939057285945933, "data_time": 0.0010082006454467773, "grad_norm": 0.05938305910676718, "loss": 0.6881160974502564, "time": 0.409075403213501, "epoch": 2, "memory": 9466, "step": 14716} +{"lr": 0.0009939057285945933, "data_time": 0.0010674715042114258, "grad_norm": 0.03911679950542748, "loss": 0.6858436703681946, "time": 0.549166464805603, "epoch": 2, "memory": 9466, "step": 14816} +{"lr": 0.0009939057285945933, "data_time": 0.001062321662902832, "grad_norm": 0.04382241545245051, "loss": 0.6866205334663391, "time": 0.41299474239349365, "epoch": 2, "memory": 9466, "step": 14916} +{"lr": 0.0009939057285945933, "data_time": 0.0011612653732299804, "grad_norm": 0.05139650125056505, "loss": 0.6853583335876465, "time": 0.45559077262878417, "epoch": 2, "memory": 9466, "step": 15016} +{"lr": 0.0009939057285945933, "data_time": 0.0011552810668945313, "grad_norm": 0.026551441475749015, "loss": 0.6852880954742432, "time": 0.4236022710800171, "epoch": 2, "memory": 9466, "step": 15116} +{"lr": 0.0009939057285945933, "data_time": 0.0011253595352172852, "grad_norm": 0.048103848658502105, "loss": 0.6889339566230774, "time": 0.4459666728973389, "epoch": 2, "memory": 9466, "step": 15216} +{"lr": 0.0009939057285945933, "data_time": 0.001021409034729004, "grad_norm": 0.06299279164522886, "loss": 0.6900611698627472, "time": 0.42662069797515867, "epoch": 2, "memory": 9466, "step": 15316} +{"lr": 0.0009939057285945933, "data_time": 0.0010375022888183595, "grad_norm": 0.03240831485018134, "loss": 0.6861496269702911, "time": 0.4836645364761353, "epoch": 2, "memory": 9466, "step": 15416} +{"lr": 0.0009939057285945933, "data_time": 0.0011087179183959961, "grad_norm": 0.0530106445774436, "loss": 0.6845498561859131, "time": 0.4130805492401123, "epoch": 2, "memory": 9466, "step": 15516} +{"lr": 0.0009939057285945933, "data_time": 0.0009972810745239257, "grad_norm": 0.047135541401803496, "loss": 0.6832665026187896, "time": 0.4348788022994995, "epoch": 2, "memory": 9466, "step": 15616} +{"lr": 0.0009939057285945933, "data_time": 0.0010476112365722656, "grad_norm": 0.032505690841935574, "loss": 0.6895552337169647, "time": 0.45715720653533937, "epoch": 2, "memory": 9466, "step": 15716} +{"lr": 0.0009939057285945933, "data_time": 0.0010787248611450195, "grad_norm": 0.05903021469712257, "loss": 0.6903643667697906, "time": 0.4205663204193115, "epoch": 2, "memory": 9466, "step": 15816} +{"lr": 0.0009939057285945933, "data_time": 0.0010478973388671875, "grad_norm": 0.04131353124976158, "loss": 0.691485857963562, "time": 0.5404685735702515, "epoch": 2, "memory": 9466, "step": 15916} +{"lr": 0.0009939057285945933, "data_time": 0.0010309457778930665, "grad_norm": 0.040479103871621194, "loss": 0.6893355786800385, "time": 0.5037957906723023, "epoch": 2, "memory": 9466, "step": 16016} +{"lr": 0.0009939057285945933, "data_time": 0.001025390625, "grad_norm": 0.03441180996596813, "loss": 0.6909900546073914, "time": 0.4745124101638794, "epoch": 2, "memory": 9466, "step": 16116} +{"lr": 0.0009939057285945933, "data_time": 0.0010228157043457031, "grad_norm": 0.046719822310842575, "loss": 0.6836608350276947, "time": 0.4528194904327393, "epoch": 2, "memory": 9466, "step": 16216} +{"lr": 0.0009939057285945933, "data_time": 0.001015782356262207, "grad_norm": 0.05144949820823967, "loss": 0.6850019097328186, "time": 0.415437126159668, "epoch": 2, "memory": 9466, "step": 16316} +{"lr": 0.0009939057285945933, "data_time": 0.0010032415390014648, "grad_norm": 0.05196168276015669, "loss": 0.6894936501979828, "time": 0.4345522403717041, "epoch": 2, "memory": 9466, "step": 16416} +{"lr": 0.0009939057285945933, "data_time": 0.0009929418563842773, "grad_norm": 0.034834797796793285, "loss": 0.6846549689769745, "time": 0.4143122673034668, "epoch": 2, "memory": 9466, "step": 16516} +{"lr": 0.0009939057285945933, "data_time": 0.0011261224746704102, "grad_norm": 0.03526101897004992, "loss": 0.6842919528484345, "time": 0.41216745376586916, "epoch": 2, "memory": 9466, "step": 16616} +{"lr": 0.0009939057285945933, "data_time": 0.001013970375061035, "grad_norm": 0.03366449736058712, "loss": 0.6872649133205414, "time": 0.4130429983139038, "epoch": 2, "memory": 9466, "step": 16716} +{"lr": 0.0009939057285945933, "data_time": 0.001103043556213379, "grad_norm": 0.05951396320015192, "loss": 0.690333080291748, "time": 0.456061863899231, "epoch": 2, "memory": 9466, "step": 16816} +{"lr": 0.0009939057285945933, "data_time": 0.0010926961898803712, "grad_norm": 0.032994561549276116, "loss": 0.6895784199237823, "time": 0.4121453285217285, "epoch": 2, "memory": 9466, "step": 16916} +{"lr": 0.0009939057285945933, "data_time": 0.0010883808135986328, "grad_norm": 0.04752643899992108, "loss": 0.6828528463840484, "time": 0.4512415647506714, "epoch": 2, "memory": 9466, "step": 17016} +{"lr": 0.0009939057285945933, "data_time": 0.001089167594909668, "grad_norm": 0.0521976382471621, "loss": 0.6842605650424958, "time": 0.42761313915252686, "epoch": 2, "memory": 9466, "step": 17116} +{"lr": 0.0009939057285945933, "data_time": 0.001126265525817871, "grad_norm": 0.06867835689336062, "loss": 0.6868312180042266, "time": 0.4242203712463379, "epoch": 2, "memory": 9466, "step": 17216} +{"lr": 0.0009939057285945933, "data_time": 0.0011965513229370117, "grad_norm": 0.022139145992696285, "loss": 0.6896021842956543, "time": 0.4299457550048828, "epoch": 2, "memory": 9466, "step": 17316} +{"lr": 0.0009939057285945933, "data_time": 0.001110672950744629, "grad_norm": 0.03525620233267546, "loss": 0.6865286350250244, "time": 0.41831867694854735, "epoch": 2, "memory": 9466, "step": 17416} +{"lr": 0.0009939057285945933, "data_time": 0.001197481155395508, "grad_norm": 0.03726433843839914, "loss": 0.6877101123332977, "time": 0.4258462905883789, "epoch": 2, "memory": 9466, "step": 17516} +{"lr": 0.0009939057285945933, "data_time": 0.0010362386703491211, "grad_norm": 0.04673717836849391, "loss": 0.6802491545677185, "time": 0.47735061645507815, "epoch": 2, "memory": 9466, "step": 17616} +{"lr": 0.0009939057285945933, "data_time": 0.0010113239288330079, "grad_norm": 0.03628114508464932, "loss": 0.6817511022090912, "time": 0.4868102312088013, "epoch": 2, "memory": 9466, "step": 17716} +{"lr": 0.0009939057285945933, "data_time": 0.001108860969543457, "grad_norm": 0.03147273710928857, "loss": 0.6830770611763001, "time": 0.44975972175598145, "epoch": 2, "memory": 9466, "step": 17816} +{"lr": 0.0009939057285945933, "data_time": 0.001001596450805664, "grad_norm": 0.023424675222486258, "loss": 0.6894601941108703, "time": 0.4413499116897583, "epoch": 2, "memory": 9466, "step": 17916} +{"lr": 0.0009939057285945933, "data_time": 0.0010518074035644532, "grad_norm": 0.05152399064972997, "loss": 0.6875954151153565, "time": 0.4273714780807495, "epoch": 2, "memory": 9466, "step": 18016} +{"lr": 0.0009939057285945933, "data_time": 0.0010159015655517578, "grad_norm": 0.030310118291527032, "loss": 0.6842800438404083, "time": 0.4172009229660034, "epoch": 2, "memory": 9466, "step": 18116} +{"lr": 0.0009939057285945933, "data_time": 0.0010413169860839845, "grad_norm": 0.024765848577953876, "loss": 0.6799925565719604, "time": 0.40242578983306887, "epoch": 2, "memory": 9466, "step": 18216} +{"lr": 0.0009939057285945933, "data_time": 0.0009528636932373047, "grad_norm": 0.03369774019811302, "loss": 0.6903659999370575, "time": 0.48039536476135253, "epoch": 2, "memory": 9466, "step": 18316} +{"lr": 0.0009939057285945933, "data_time": 0.001019287109375, "grad_norm": 0.035824154131114484, "loss": 0.6824732899665833, "time": 0.4307317972183228, "epoch": 2, "memory": 9466, "step": 18416} +{"lr": 0.0009939057285945933, "data_time": 0.0014621734619140625, "grad_norm": 0.028447221009992064, "loss": 0.689435213804245, "time": 0.4386845350265503, "epoch": 2, "memory": 9466, "step": 18516} +{"lr": 0.0009939057285945933, "data_time": 0.0011492252349853515, "grad_norm": 0.039267354225739834, "loss": 0.6890923380851746, "time": 0.4825651407241821, "epoch": 2, "memory": 9466, "step": 18616} +{"lr": 0.0009939057285945933, "data_time": 0.001076340675354004, "grad_norm": 0.0264707637950778, "loss": 0.6854984283447265, "time": 0.3985506772994995, "epoch": 2, "memory": 9466, "step": 18716} +{"lr": 0.0009939057285945933, "data_time": 0.0011612653732299804, "grad_norm": 0.04025698853656649, "loss": 0.6885736048221588, "time": 0.41092500686645506, "epoch": 2, "memory": 9466, "step": 18816} +{"lr": 0.0009939057285945933, "data_time": 0.0010822057723999024, "grad_norm": 0.03930663289502263, "loss": 0.6831455588340759, "time": 0.46106586456298826, "epoch": 2, "memory": 9466, "step": 18916} +{"lr": 0.0009939057285945933, "data_time": 0.0010393381118774414, "grad_norm": 0.021630261489190162, "loss": 0.6858853161334991, "time": 0.45829143524169924, "epoch": 2, "memory": 9466, "step": 19016} +{"lr": 0.0009939057285945933, "data_time": 0.0011249542236328124, "grad_norm": 0.025389747810550034, "loss": 0.6922311604022979, "time": 0.4417997360229492, "epoch": 2, "memory": 9466, "step": 19116} +{"lr": 0.0009939057285945933, "data_time": 0.0011394739151000977, "grad_norm": 0.027805525809526443, "loss": 0.6826779186725617, "time": 0.44397075176239015, "epoch": 2, "memory": 9466, "step": 19216} +{"lr": 0.0009939057285945933, "data_time": 0.001194906234741211, "grad_norm": 0.021581029053777456, "loss": 0.6884427964687347, "time": 0.45931086540222166, "epoch": 2, "memory": 9466, "step": 19316} +{"lr": 0.0009939057285945933, "data_time": 0.0011829614639282226, "grad_norm": 0.029122319514863194, "loss": 0.6823606014251709, "time": 0.42016050815582273, "epoch": 2, "memory": 9466, "step": 19416} +{"lr": 0.0009939057285945933, "data_time": 0.0010602235794067382, "grad_norm": 0.032002720597665756, "loss": 0.6855169177055359, "time": 0.450806999206543, "epoch": 2, "memory": 9466, "step": 19516} +{"lr": 0.0009939057285945933, "data_time": 0.0010131120681762696, "grad_norm": 0.03347331110853702, "loss": 0.6862307786941528, "time": 0.40237624645233155, "epoch": 2, "memory": 9466, "step": 19616} +{"lr": 0.0009939057285945933, "data_time": 0.0010608434677124023, "grad_norm": 0.025364167243242263, "loss": 0.682321161031723, "time": 0.43271775245666505, "epoch": 2, "memory": 9466, "step": 19716} +{"lr": 0.0009939057285945933, "data_time": 0.0012502670288085938, "grad_norm": 0.03594321897253394, "loss": 0.6886745691299438, "time": 0.3986459493637085, "epoch": 2, "memory": 9466, "step": 19816} +{"lr": 0.0009939057285945933, "data_time": 0.0012074947357177735, "grad_norm": 0.025908530689775944, "loss": 0.6872937381267548, "time": 0.4347603559494019, "epoch": 2, "memory": 9466, "step": 19916} +{"lr": 0.0009939057285945933, "data_time": 0.001089000701904297, "grad_norm": 0.026527242409065364, "loss": 0.6900459051132202, "time": 0.4368203401565552, "epoch": 2, "memory": 9466, "step": 20016} +{"lr": 0.0009939057285945933, "data_time": 0.001088404655456543, "grad_norm": 0.01877260854234919, "loss": 0.6839976489543915, "time": 0.41747846603393557, "epoch": 2, "memory": 9466, "step": 20116} +{"lr": 0.0009939057285945933, "data_time": 0.0010925054550170899, "grad_norm": 0.030362183065153658, "loss": 0.683194762468338, "time": 0.41016592979431155, "epoch": 2, "memory": 9466, "step": 20216} +{"lr": 0.0009939057285945933, "data_time": 0.0010634660720825195, "grad_norm": 0.03470392026938498, "loss": 0.6819584190845489, "time": 0.39416506290435793, "epoch": 2, "memory": 9466, "step": 20316} +{"lr": 0.0009939057285945933, "data_time": 0.0013753652572631836, "grad_norm": 0.03229015273973346, "loss": 0.6807324111461639, "time": 0.40594961643218996, "epoch": 2, "memory": 9466, "step": 20416} +{"lr": 0.0009939057285945933, "data_time": 0.0011197328567504883, "grad_norm": 0.024766219686716794, "loss": 0.6878861069679261, "time": 0.41277735233306884, "epoch": 2, "memory": 9466, "step": 20516} +{"lr": 0.0009939057285945933, "data_time": 0.0011850357055664062, "grad_norm": 0.023120308574289084, "loss": 0.6896874368190765, "time": 0.40929038524627687, "epoch": 2, "memory": 9466, "step": 20616} +{"lr": 0.0009939057285945933, "data_time": 0.0011122941970825196, "grad_norm": 0.028626284468919037, "loss": 0.6880630731582642, "time": 0.46014392375946045, "epoch": 2, "memory": 9466, "step": 20716} +{"lr": 0.0009939057285945933, "data_time": 0.0012867212295532226, "grad_norm": 0.03971298469696194, "loss": 0.6875614821910858, "time": 0.44701123237609863, "epoch": 2, "memory": 9466, "step": 20816} +{"lr": 0.0009939057285945933, "data_time": 0.0011586904525756835, "grad_norm": 0.0199462678283453, "loss": 0.6845956027507782, "time": 0.4387508869171143, "epoch": 2, "memory": 9466, "step": 20916} +{"lr": 0.0009939057285945933, "data_time": 0.0011193275451660157, "grad_norm": 0.02812248272821307, "loss": 0.6877283215522766, "time": 0.44707579612731935, "epoch": 2, "memory": 9466, "step": 21016} +{"lr": 0.0009939057285945933, "data_time": 0.0012781143188476563, "grad_norm": 0.028240500867832453, "loss": 0.690543282032013, "time": 0.4225125551223755, "epoch": 2, "memory": 9466, "step": 21116} +{"lr": 0.0009939057285945933, "data_time": 0.0011524438858032226, "grad_norm": 0.03582395128905773, "loss": 0.6840147912502289, "time": 0.44947845935821534, "epoch": 2, "memory": 9466, "step": 21216} +{"lr": 0.0009939057285945933, "data_time": 0.0010776281356811523, "grad_norm": 0.021379622141830625, "loss": 0.687153959274292, "time": 0.4285252094268799, "epoch": 2, "memory": 9466, "step": 21316} +{"lr": 0.0009939057285945933, "data_time": 0.001127934455871582, "grad_norm": 0.025071394885890184, "loss": 0.6861791789531708, "time": 0.4396047592163086, "epoch": 2, "memory": 9466, "step": 21416} +{"lr": 0.0009939057285945933, "data_time": 0.001260828971862793, "grad_norm": 0.030992146208882333, "loss": 0.684092378616333, "time": 0.4403985500335693, "epoch": 2, "memory": 9466, "step": 21516} +{"lr": 0.0009939057285945933, "data_time": 0.0010259628295898437, "grad_norm": 0.023117261379957198, "loss": 0.6890429496765137, "time": 0.4923547744750977, "epoch": 2, "memory": 9466, "step": 21616} +{"lr": 0.0009939057285945933, "data_time": 0.0011034965515136718, "grad_norm": 0.017198899178765714, "loss": 0.6831491529941559, "time": 0.44282629489898684, "epoch": 2, "memory": 9466, "step": 21716} +{"lr": 0.0009939057285945933, "data_time": 0.0011950492858886718, "grad_norm": 0.03520979601889849, "loss": 0.6863300621509552, "time": 0.4358607053756714, "epoch": 2, "memory": 9466, "step": 21816} +{"lr": 0.0009939057285945933, "data_time": 0.0013605833053588867, "grad_norm": 0.026939908997155725, "loss": 0.6850939512252807, "time": 0.4162754535675049, "epoch": 2, "memory": 9466, "step": 21916} +{"lr": 0.0009939057285945933, "data_time": 0.0010367631912231445, "grad_norm": 0.025196888181380928, "loss": 0.6890845596790314, "time": 0.4047020435333252, "epoch": 2, "memory": 9466, "step": 22016} +{"lr": 0.0009939057285945933, "data_time": 0.001160883903503418, "grad_norm": 0.015232666418887675, "loss": 0.6888692080974579, "time": 0.42114341259002686, "epoch": 2, "memory": 9466, "step": 22116} +{"lr": 0.0009939057285945933, "data_time": 0.0012646913528442383, "grad_norm": 0.012506846012547613, "loss": 0.6819145679473877, "time": 0.39800872802734377, "epoch": 2, "memory": 9466, "step": 22216} +{"lr": 0.0009939057285945933, "data_time": 0.0011089801788330077, "grad_norm": 0.024451967608183623, "loss": 0.6845054745674133, "time": 0.46559360027313235, "epoch": 2, "memory": 9466, "step": 22316} +{"lr": 0.0009939057285945933, "data_time": 0.0011581897735595704, "grad_norm": 0.024922854756005107, "loss": 0.6825241982936859, "time": 0.40462474822998046, "epoch": 2, "memory": 9466, "step": 22416} +{"lr": 0.0009939057285945933, "data_time": 0.001146554946899414, "grad_norm": 0.018348787701688705, "loss": 0.6876084089279175, "time": 0.37368662357330323, "epoch": 2, "memory": 9466, "step": 22516} +{"lr": 0.0009939057285945933, "data_time": 0.0010806083679199218, "grad_norm": 0.022323320084251464, "loss": 0.6853807032108307, "time": 0.46470203399658205, "epoch": 2, "memory": 9466, "step": 22616} +{"lr": 0.0009939057285945933, "data_time": 0.0011414766311645507, "grad_norm": 0.025385763216763734, "loss": 0.6867986142635345, "time": 0.4136732339859009, "epoch": 2, "memory": 9466, "step": 22716} +{"lr": 0.0009939057285945933, "data_time": 0.0012858390808105468, "grad_norm": 0.024933330668136477, "loss": 0.6839031398296356, "time": 0.421768856048584, "epoch": 2, "memory": 9466, "step": 22816} +{"lr": 0.0009939057285945933, "data_time": 0.0013912200927734375, "grad_norm": 0.031420634919777514, "loss": 0.6872365832328796, "time": 0.3978781223297119, "epoch": 2, "memory": 9466, "step": 22916} +{"lr": 0.0009939057285945933, "data_time": 0.004609894752502441, "grad_norm": 0.02087268108734861, "loss": 0.6900867044925689, "time": 0.4825006008148193, "epoch": 2, "memory": 9466, "step": 23016} +{"accuracy/top1": 100.0, "data_time": 0.30627135669483857, "time": 0.5601891910328585, "step": 2} +{"lr": 0.0009757729755661012, "data_time": 0.000995039939880371, "grad_norm": 0.015585147636011243, "loss": 0.6836028575897217, "time": 0.44008607864379884, "epoch": 3, "memory": 9466, "step": 23132} +{"lr": 0.0009757729755661012, "data_time": 0.001377701759338379, "grad_norm": 0.023318680422380566, "loss": 0.6925673604011535, "time": 0.4418300151824951, "epoch": 3, "memory": 9465, "step": 23232} +{"lr": 0.0009757729755661012, "data_time": 0.0827981948852539, "grad_norm": 0.019727103505283596, "loss": 0.6877822697162628, "time": 0.4060245990753174, "epoch": 3, "memory": 9465, "step": 23332} +{"lr": 0.0009757729755661012, "data_time": 0.0009917736053466797, "grad_norm": 0.018417895305901766, "loss": 0.6871657907962799, "time": 0.4709378719329834, "epoch": 3, "memory": 9465, "step": 23432} +{"lr": 0.0009757729755661012, "data_time": 0.0017249345779418944, "grad_norm": 0.02222048370167613, "loss": 0.6895881295204163, "time": 0.431545090675354, "epoch": 3, "memory": 9465, "step": 23532} +{"lr": 0.0009757729755661012, "data_time": 0.0017734527587890624, "grad_norm": 0.019312548195011915, "loss": 0.6906211972236633, "time": 0.4764890432357788, "epoch": 3, "memory": 9465, "step": 23632} +{"lr": 0.0009757729755661012, "data_time": 0.0014914512634277345, "grad_norm": 0.014033994544297457, "loss": 0.6879104256629944, "time": 0.48021764755249025, "epoch": 3, "memory": 9465, "step": 23732} +{"lr": 0.0009757729755661012, "data_time": 0.0014476299285888672, "grad_norm": 0.02209121969062835, "loss": 0.6855984449386596, "time": 0.40016770362854004, "epoch": 3, "memory": 9465, "step": 23832} +{"lr": 0.0009757729755661012, "data_time": 0.0012039899826049804, "grad_norm": 0.013217152189463377, "loss": 0.6847003161907196, "time": 0.45034902095794677, "epoch": 3, "memory": 9465, "step": 23932} +{"lr": 0.0009757729755661012, "data_time": 0.0011378049850463868, "grad_norm": 0.026478892378509044, "loss": 0.6886955976486206, "time": 0.47276129722595217, "epoch": 3, "memory": 9465, "step": 24032} +{"lr": 0.0009757729755661012, "data_time": 0.0010313987731933594, "grad_norm": 0.024408833286724985, "loss": 0.6888004958629608, "time": 0.43849332332611085, "epoch": 3, "memory": 9465, "step": 24132} +{"lr": 0.0009757729755661012, "data_time": 0.0020800113677978517, "grad_norm": 0.020728322584182023, "loss": 0.6848903357982635, "time": 0.39533162117004395, "epoch": 3, "memory": 9465, "step": 24232} +{"lr": 0.0009757729755661012, "data_time": 0.0012228727340698243, "grad_norm": 0.02266580640571192, "loss": 0.687611174583435, "time": 0.4516587495803833, "epoch": 3, "memory": 9465, "step": 24332} +{"lr": 0.0009757729755661012, "data_time": 0.0020180463790893553, "grad_norm": 0.02308184818830341, "loss": 0.685395109653473, "time": 0.4768563508987427, "epoch": 3, "memory": 9465, "step": 24432} +{"lr": 0.0009757729755661012, "data_time": 0.0013065099716186523, "grad_norm": 0.06457884428091347, "loss": 0.6833694756031037, "time": 0.4407362461090088, "epoch": 3, "memory": 9465, "step": 24532} +{"lr": 0.0009757729755661012, "data_time": 0.0011156320571899414, "grad_norm": 0.040838798554614184, "loss": 0.6834480226039886, "time": 0.3935695171356201, "epoch": 3, "memory": 9465, "step": 24632} +{"lr": 0.0009757729755661012, "data_time": 0.0010981798171997071, "grad_norm": 0.025170611450448632, "loss": 0.6850519716739655, "time": 0.4510664463043213, "epoch": 3, "memory": 9465, "step": 24732} +{"lr": 0.0009757729755661012, "data_time": 0.0014521360397338867, "grad_norm": 0.025674613169394435, "loss": 0.6856638967990876, "time": 0.40546751022338867, "epoch": 3, "memory": 9465, "step": 24832} +{"lr": 0.0009757729755661012, "data_time": 0.0012404680252075194, "grad_norm": 0.027631144598126413, "loss": 0.6887315809726715, "time": 0.4018693208694458, "epoch": 3, "memory": 9465, "step": 24932} +{"lr": 0.0009757729755661012, "data_time": 0.0018202304840087891, "grad_norm": 0.021495186537504197, "loss": 0.6858207285404205, "time": 0.4309407949447632, "epoch": 3, "memory": 9465, "step": 25032} +{"lr": 0.0009757729755661012, "data_time": 0.001217961311340332, "grad_norm": 0.03502406873740256, "loss": 0.6926734924316407, "time": 0.40155487060546874, "epoch": 3, "memory": 9465, "step": 25132} +{"lr": 0.0009757729755661012, "data_time": 0.0013715982437133788, "grad_norm": 0.02867203941568732, "loss": 0.6881754457950592, "time": 0.4332838773727417, "epoch": 3, "memory": 9465, "step": 25232} +{"lr": 0.0009757729755661012, "data_time": 0.001523447036743164, "grad_norm": 0.03737427806481719, "loss": 0.6857482075691224, "time": 0.44575021266937254, "epoch": 3, "memory": 9465, "step": 25332} +{"lr": 0.0009757729755661012, "data_time": 0.0013747453689575196, "grad_norm": 0.02552356361411512, "loss": 0.6898065567016601, "time": 0.4265328884124756, "epoch": 3, "memory": 9465, "step": 25432} +{"lr": 0.0009757729755661012, "data_time": 0.0010834455490112305, "grad_norm": 0.03172828210517764, "loss": 0.6846775054931641, "time": 0.4278853893280029, "epoch": 3, "memory": 9465, "step": 25532} +{"lr": 0.0009757729755661012, "data_time": 0.00110933780670166, "grad_norm": 0.033474443340674044, "loss": 0.6859627842903138, "time": 0.4249113082885742, "epoch": 3, "memory": 9465, "step": 25632} +{"lr": 0.0009757729755661012, "data_time": 0.0010965347290039062, "grad_norm": 0.02725795153528452, "loss": 0.6842111229896546, "time": 0.40430710315704343, "epoch": 3, "memory": 9465, "step": 25732} +{"lr": 0.0009757729755661012, "data_time": 0.0013139009475708007, "grad_norm": 0.027517162123695017, "loss": 0.6904976963996887, "time": 0.41391568183898925, "epoch": 3, "memory": 9465, "step": 25832} +{"lr": 0.0009757729755661012, "data_time": 0.0013836145401000977, "grad_norm": 0.025473095057532192, "loss": 0.6836946249008179, "time": 0.37690105438232424, "epoch": 3, "memory": 9465, "step": 25932} +{"lr": 0.0009757729755661012, "data_time": 0.0013869762420654296, "grad_norm": 0.024343622149899602, "loss": 0.6847755968570709, "time": 0.44075562953948977, "epoch": 3, "memory": 9465, "step": 26032} +{"lr": 0.0009757729755661012, "data_time": 0.0013386011123657227, "grad_norm": 0.02313335072249174, "loss": 0.6847593605518341, "time": 0.4898258924484253, "epoch": 3, "memory": 9465, "step": 26132} +{"lr": 0.0009757729755661012, "data_time": 0.0021242856979370116, "grad_norm": 0.024646509112790226, "loss": 0.6875008940696716, "time": 0.3919433832168579, "epoch": 3, "memory": 9465, "step": 26232} +{"lr": 0.0009757729755661012, "data_time": 0.0016807317733764648, "grad_norm": 0.0246181502006948, "loss": 0.6875292062759399, "time": 0.4523853540420532, "epoch": 3, "memory": 9465, "step": 26332} +{"lr": 0.0009757729755661012, "data_time": 0.0010595083236694335, "grad_norm": 0.019470269698649643, "loss": 0.6877839267253876, "time": 0.3780526876449585, "epoch": 3, "memory": 9465, "step": 26432} +{"lr": 0.0009757729755661012, "data_time": 0.0010975122451782227, "grad_norm": 0.02979953051544726, "loss": 0.6883388817310333, "time": 0.5774923324584961, "epoch": 3, "memory": 9465, "step": 26532} +{"lr": 0.0009757729755661012, "data_time": 0.001065373420715332, "grad_norm": 0.02564713526517153, "loss": 0.6867683947086334, "time": 0.49703762531280515, "epoch": 3, "memory": 9465, "step": 26632} +{"lr": 0.0009757729755661012, "data_time": 0.001167893409729004, "grad_norm": 0.026032890332862733, "loss": 0.6869809746742248, "time": 0.4039982795715332, "epoch": 3, "memory": 9465, "step": 26732} +{"lr": 0.0009757729755661012, "data_time": 0.0011603116989135742, "grad_norm": 0.01670405617915094, "loss": 0.6842011153697968, "time": 0.48672292232513426, "epoch": 3, "memory": 9465, "step": 26832} +{"lr": 0.0009757729755661012, "data_time": 0.0011061906814575195, "grad_norm": 0.026842055190354586, "loss": 0.6836141109466553, "time": 0.45530383586883544, "epoch": 3, "memory": 9465, "step": 26932} +{"lr": 0.0009757729755661012, "data_time": 0.0011821985244750977, "grad_norm": 0.015099308732897044, "loss": 0.6877972960472107, "time": 0.42082149982452394, "epoch": 3, "memory": 9465, "step": 27032} +{"lr": 0.0009757729755661012, "data_time": 0.0011876821517944336, "grad_norm": 0.02831619172357023, "loss": 0.6841572225093842, "time": 0.46534578800201415, "epoch": 3, "memory": 9465, "step": 27132} +{"lr": 0.0009757729755661012, "data_time": 0.0010373353958129882, "grad_norm": 0.017154495231807232, "loss": 0.6849205136299134, "time": 0.40144083499908445, "epoch": 3, "memory": 9465, "step": 27232} +{"lr": 0.0009757729755661012, "data_time": 0.0012067317962646484, "grad_norm": 0.0299342380836606, "loss": 0.684631985425949, "time": 0.4506873607635498, "epoch": 3, "memory": 9465, "step": 27332} +{"lr": 0.0009757729755661012, "data_time": 0.0011445999145507813, "grad_norm": 0.0222253386862576, "loss": 0.6883554875850677, "time": 0.3992605447769165, "epoch": 3, "memory": 9465, "step": 27432} +{"lr": 0.0009757729755661012, "data_time": 0.001079702377319336, "grad_norm": 0.02965725902467966, "loss": 0.687291544675827, "time": 0.42951183319091796, "epoch": 3, "memory": 9465, "step": 27532} +{"lr": 0.0009757729755661012, "data_time": 0.0011515140533447266, "grad_norm": 0.024875061493366955, "loss": 0.6867052674293518, "time": 0.41588122844696046, "epoch": 3, "memory": 9465, "step": 27632} +{"lr": 0.0009757729755661012, "data_time": 0.0012460470199584961, "grad_norm": 0.02218859880231321, "loss": 0.68658407330513, "time": 0.41407184600830077, "epoch": 3, "memory": 9465, "step": 27732} +{"lr": 0.0009757729755661012, "data_time": 0.0012710809707641602, "grad_norm": 0.02219894644804299, "loss": 0.6858759343624115, "time": 0.4324398756027222, "epoch": 3, "memory": 9465, "step": 27832} +{"lr": 0.0009757729755661012, "data_time": 0.001050424575805664, "grad_norm": 0.021095025166869164, "loss": 0.6843375205993653, "time": 0.47747251987457273, "epoch": 3, "memory": 9465, "step": 27932} +{"lr": 0.0009757729755661012, "data_time": 0.0011342287063598633, "grad_norm": 0.018732304126024245, "loss": 0.6914817154407501, "time": 0.48612802028656005, "epoch": 3, "memory": 9465, "step": 28032} +{"lr": 0.0009757729755661012, "data_time": 0.001188182830810547, "grad_norm": 0.02164520784281194, "loss": 0.6878631353378296, "time": 0.3929041147232056, "epoch": 3, "memory": 9465, "step": 28132} +{"lr": 0.0009757729755661012, "data_time": 0.0011054515838623048, "grad_norm": 0.021526194317266344, "loss": 0.685868912935257, "time": 0.40958170890808104, "epoch": 3, "memory": 9465, "step": 28232} +{"lr": 0.0009757729755661012, "data_time": 0.0011272192001342773, "grad_norm": 0.020032352395355703, "loss": 0.6839922964572906, "time": 0.46430985927581786, "epoch": 3, "memory": 9465, "step": 28332} +{"lr": 0.0009757729755661012, "data_time": 0.0011687278747558594, "grad_norm": 0.029056368954479696, "loss": 0.6802111506462097, "time": 0.4719816207885742, "epoch": 3, "memory": 9465, "step": 28432} +{"lr": 0.0009757729755661012, "data_time": 0.0012011051177978516, "grad_norm": 0.026354875555261968, "loss": 0.6879319608211517, "time": 0.4400376558303833, "epoch": 3, "memory": 9465, "step": 28532} +{"lr": 0.0009757729755661012, "data_time": 0.0011624574661254882, "grad_norm": 0.02252791910432279, "loss": 0.6908713400363922, "time": 0.40926480293273926, "epoch": 3, "memory": 9465, "step": 28632} +{"lr": 0.0009757729755661012, "data_time": 0.0010194063186645507, "grad_norm": 0.03153423047624528, "loss": 0.6901094853878021, "time": 0.4298314809799194, "epoch": 3, "memory": 9465, "step": 28732} +{"lr": 0.0009757729755661012, "data_time": 0.001091742515563965, "grad_norm": 0.02403234178200364, "loss": 0.6859693348407745, "time": 0.46756978034973146, "epoch": 3, "memory": 9465, "step": 28832} +{"lr": 0.0009757729755661012, "data_time": 0.0013289213180541991, "grad_norm": 0.0354332847520709, "loss": 0.684811782836914, "time": 0.3715671539306641, "epoch": 3, "memory": 9465, "step": 28932} +{"lr": 0.0009757729755661012, "data_time": 0.0012017011642456055, "grad_norm": 0.022738070320338012, "loss": 0.6883825361728668, "time": 0.4039838075637817, "epoch": 3, "memory": 9465, "step": 29032} +{"lr": 0.0009757729755661012, "data_time": 0.0011409759521484376, "grad_norm": 0.029429469164460896, "loss": 0.6878215253353119, "time": 0.37913217544555666, "epoch": 3, "memory": 9465, "step": 29132} +{"lr": 0.0009757729755661012, "data_time": 0.0011286735534667969, "grad_norm": 0.02399361697025597, "loss": 0.6835292994976043, "time": 0.4256344556808472, "epoch": 3, "memory": 9465, "step": 29232} +{"lr": 0.0009757729755661012, "data_time": 0.0011286258697509766, "grad_norm": 0.01877906555309892, "loss": 0.6891405820846558, "time": 0.4724832534790039, "epoch": 3, "memory": 9465, "step": 29332} +{"lr": 0.0009757729755661012, "data_time": 0.0012565851211547852, "grad_norm": 0.01546756005845964, "loss": 0.682167261838913, "time": 0.3839565753936768, "epoch": 3, "memory": 9465, "step": 29432} +{"lr": 0.0009757729755661012, "data_time": 0.001350259780883789, "grad_norm": 0.02717026323080063, "loss": 0.6854710876941681, "time": 0.3995219230651855, "epoch": 3, "memory": 9465, "step": 29532} +{"lr": 0.0009757729755661012, "data_time": 0.0012035608291625977, "grad_norm": 0.015795053634792566, "loss": 0.6834986269474029, "time": 0.4032184839248657, "epoch": 3, "memory": 9465, "step": 29632} +{"lr": 0.0009757729755661012, "data_time": 0.0013095855712890625, "grad_norm": 0.02186451982706785, "loss": 0.6897809982299805, "time": 0.4221600294113159, "epoch": 3, "memory": 9465, "step": 29732} +{"lr": 0.0009757729755661012, "data_time": 0.0011242389678955077, "grad_norm": 0.022603030130267144, "loss": 0.6875552356243133, "time": 0.3972029447555542, "epoch": 3, "memory": 9465, "step": 29832} +{"lr": 0.0009757729755661012, "data_time": 0.0010331630706787109, "grad_norm": 0.02292302092537284, "loss": 0.6815848410129547, "time": 0.4479137182235718, "epoch": 3, "memory": 9465, "step": 29932} +{"lr": 0.0009757729755661012, "data_time": 0.001129317283630371, "grad_norm": 0.03163904398679733, "loss": 0.6903496265411377, "time": 0.399628734588623, "epoch": 3, "memory": 9465, "step": 30032} +{"lr": 0.0009757729755661012, "data_time": 0.001031637191772461, "grad_norm": 0.029005956277251242, "loss": 0.6859833240509033, "time": 0.398192024230957, "epoch": 3, "memory": 9465, "step": 30132} +{"lr": 0.0009757729755661012, "data_time": 0.0012382984161376954, "grad_norm": 0.026983386278152464, "loss": 0.6888517439365387, "time": 0.3853166103363037, "epoch": 3, "memory": 9465, "step": 30232} +{"lr": 0.0009757729755661012, "data_time": 0.0012030601501464844, "grad_norm": 0.038520692754536866, "loss": 0.686540961265564, "time": 0.38440546989440916, "epoch": 3, "memory": 9465, "step": 30332} +{"lr": 0.0009757729755661012, "data_time": 0.001174759864807129, "grad_norm": 0.02511755982413888, "loss": 0.6896385133266449, "time": 0.39105055332183836, "epoch": 3, "memory": 9465, "step": 30432} +{"lr": 0.0009757729755661012, "data_time": 0.0014658689498901368, "grad_norm": 0.031639125244691965, "loss": 0.6934526026248932, "time": 0.44205350875854493, "epoch": 3, "memory": 9465, "step": 30532} +{"lr": 0.0009757729755661012, "data_time": 0.0012601375579833984, "grad_norm": 0.02406288618221879, "loss": 0.6928059995174408, "time": 0.38973872661590575, "epoch": 3, "memory": 9465, "step": 30632} +{"lr": 0.0009757729755661012, "data_time": 0.0011130332946777343, "grad_norm": 0.020718642440624536, "loss": 0.6865417540073395, "time": 0.44378411769866943, "epoch": 3, "memory": 9465, "step": 30732} +{"lr": 0.0009757729755661012, "data_time": 0.001770782470703125, "grad_norm": 0.022590391407720744, "loss": 0.6830469191074371, "time": 0.41610231399536135, "epoch": 3, "memory": 9465, "step": 30832} +{"lr": 0.0009757729755661012, "data_time": 0.0013607978820800782, "grad_norm": 0.02363686291500926, "loss": 0.6823496639728546, "time": 0.3708021879196167, "epoch": 3, "memory": 9465, "step": 30932} +{"lr": 0.0009757729755661012, "data_time": 0.0010550260543823243, "grad_norm": 0.03002280886285007, "loss": 0.6842589616775513, "time": 0.3915779829025269, "epoch": 3, "memory": 9465, "step": 31032} +{"lr": 0.0009757729755661012, "data_time": 0.0014104366302490235, "grad_norm": 0.021123229106888176, "loss": 0.682874721288681, "time": 0.41668713092803955, "epoch": 3, "memory": 9465, "step": 31132} +{"lr": 0.0009757729755661012, "data_time": 0.001031780242919922, "grad_norm": 0.017836342193186284, "loss": 0.6875498414039611, "time": 0.4248785495758057, "epoch": 3, "memory": 9465, "step": 31232} +{"lr": 0.0009757729755661012, "data_time": 0.0010653972625732423, "grad_norm": 0.02051288477377966, "loss": 0.6856572270393372, "time": 0.47451279163360593, "epoch": 3, "memory": 9465, "step": 31332} +{"lr": 0.0009757729755661012, "data_time": 0.001134777069091797, "grad_norm": 0.01699108590837568, "loss": 0.6913679778575897, "time": 0.41942386627197265, "epoch": 3, "memory": 9465, "step": 31432} +{"lr": 0.0009757729755661012, "data_time": 0.001046442985534668, "grad_norm": 0.026454728469252585, "loss": 0.687749183177948, "time": 0.40201330184936523, "epoch": 3, "memory": 9465, "step": 31532} +{"lr": 0.0009757729755661012, "data_time": 0.0010889291763305664, "grad_norm": 0.026782857766374946, "loss": 0.6820603430271148, "time": 0.38089675903320314, "epoch": 3, "memory": 9465, "step": 31632} +{"lr": 0.0009757729755661012, "data_time": 0.0013752460479736328, "grad_norm": 0.021188395144417883, "loss": 0.6859297692775727, "time": 0.41359283924102785, "epoch": 3, "memory": 9465, "step": 31732} +{"lr": 0.0009757729755661012, "data_time": 0.0012529611587524414, "grad_norm": 0.015946542425081135, "loss": 0.6897715508937836, "time": 0.44652998447418213, "epoch": 3, "memory": 9465, "step": 31832} +{"lr": 0.0009757729755661012, "data_time": 0.001079893112182617, "grad_norm": 0.032581039471551776, "loss": 0.6879302859306335, "time": 0.4039323329925537, "epoch": 3, "memory": 9465, "step": 31932} +{"lr": 0.0009757729755661012, "data_time": 0.0009735107421875, "grad_norm": 0.02621848154813051, "loss": 0.684019786119461, "time": 0.4213979959487915, "epoch": 3, "memory": 9465, "step": 32032} +{"lr": 0.0009757729755661012, "data_time": 0.0010878324508666992, "grad_norm": 0.027288965182378888, "loss": 0.680187976360321, "time": 0.4218135833740234, "epoch": 3, "memory": 9465, "step": 32132} +{"lr": 0.0009757729755661012, "data_time": 0.0010739803314208985, "grad_norm": 0.019247781950980425, "loss": 0.6840679943561554, "time": 0.43177151679992676, "epoch": 3, "memory": 9465, "step": 32232} +{"lr": 0.0009757729755661012, "data_time": 0.0011451005935668944, "grad_norm": 0.021946579916402698, "loss": 0.683692067861557, "time": 0.42153475284576414, "epoch": 3, "memory": 9465, "step": 32332} +{"lr": 0.0009757729755661012, "data_time": 0.0010820627212524414, "grad_norm": 0.026870765211060643, "loss": 0.6886960387229919, "time": 0.4584962844848633, "epoch": 3, "memory": 9465, "step": 32432} +{"lr": 0.0009757729755661012, "data_time": 0.0010807037353515625, "grad_norm": 0.02631870238110423, "loss": 0.682082062959671, "time": 0.4570074319839478, "epoch": 3, "memory": 9465, "step": 32532} +{"lr": 0.0009757729755661012, "data_time": 0.0009947776794433593, "grad_norm": 0.024161409866064788, "loss": 0.6895573616027832, "time": 0.4683802604675293, "epoch": 3, "memory": 9465, "step": 32632} +{"lr": 0.0009757729755661012, "data_time": 0.0010188102722167968, "grad_norm": 0.01994178690947592, "loss": 0.6848426759243011, "time": 0.4365750551223755, "epoch": 3, "memory": 9465, "step": 32732} +{"lr": 0.0009757729755661012, "data_time": 0.0010864019393920898, "grad_norm": 0.025550302490592, "loss": 0.6919171154499054, "time": 0.41815476417541503, "epoch": 3, "memory": 9465, "step": 32832} +{"lr": 0.0009757729755661012, "data_time": 0.0011384248733520507, "grad_norm": 0.01579303703038022, "loss": 0.6893027245998382, "time": 0.4446742534637451, "epoch": 3, "memory": 9465, "step": 32932} +{"lr": 0.0009757729755661012, "data_time": 0.0012124300003051758, "grad_norm": 0.02473419322632253, "loss": 0.6855925440788269, "time": 0.40201406478881835, "epoch": 3, "memory": 9465, "step": 33032} +{"lr": 0.0009757729755661012, "data_time": 0.0013197898864746095, "grad_norm": 0.02229913806077093, "loss": 0.6881005585193634, "time": 0.42860519886016846, "epoch": 3, "memory": 9465, "step": 33132} +{"lr": 0.0009757729755661012, "data_time": 0.001178145408630371, "grad_norm": 0.016469798039179296, "loss": 0.692356276512146, "time": 0.45051984786987304, "epoch": 3, "memory": 9465, "step": 33232} +{"lr": 0.0009757729755661012, "data_time": 0.0009961843490600586, "grad_norm": 0.03090803502127528, "loss": 0.689253854751587, "time": 0.4596554279327393, "epoch": 3, "memory": 9465, "step": 33332} +{"lr": 0.0009757729755661012, "data_time": 0.0011402130126953124, "grad_norm": 0.025316892098635434, "loss": 0.6842014372348786, "time": 0.42744667530059816, "epoch": 3, "memory": 9465, "step": 33432} +{"lr": 0.0009757729755661012, "data_time": 0.0010706424713134766, "grad_norm": 0.022424074914306404, "loss": 0.6886593043804169, "time": 0.39935548305511476, "epoch": 3, "memory": 9465, "step": 33532} +{"lr": 0.0009757729755661012, "data_time": 0.00169680118560791, "grad_norm": 0.02563798618502915, "loss": 0.6867034196853637, "time": 0.4220255374908447, "epoch": 3, "memory": 9465, "step": 33632} +{"lr": 0.0009757729755661012, "data_time": 0.0010977745056152343, "grad_norm": 0.015483600366860629, "loss": 0.6839949071407319, "time": 0.4257720708847046, "epoch": 3, "memory": 9465, "step": 33732} +{"lr": 0.0009757729755661012, "data_time": 0.0013451337814331054, "grad_norm": 0.023556360881775616, "loss": 0.691718477010727, "time": 0.41590681076049807, "epoch": 3, "memory": 9465, "step": 33832} +{"lr": 0.0009757729755661012, "data_time": 0.0010607719421386718, "grad_norm": 0.030128264846280218, "loss": 0.6869981110095977, "time": 0.38972184658050535, "epoch": 3, "memory": 9465, "step": 33932} +{"lr": 0.0009757729755661012, "data_time": 0.001278853416442871, "grad_norm": 0.025171750318259002, "loss": 0.6834398865699768, "time": 0.4456507444381714, "epoch": 3, "memory": 9465, "step": 34032} +{"lr": 0.0009757729755661012, "data_time": 0.0011839628219604491, "grad_norm": 0.02239755503833294, "loss": 0.6860175549983978, "time": 0.4222816228866577, "epoch": 3, "memory": 9465, "step": 34132} +{"lr": 0.0009757729755661012, "data_time": 0.2247065544128418, "grad_norm": 0.029609837243333458, "loss": 0.6806548655033111, "time": 0.4390366554260254, "epoch": 3, "memory": 9465, "step": 34232} +{"lr": 0.0009757729755661012, "data_time": 0.19836778640747071, "grad_norm": 0.021756906434893607, "loss": 0.6878196001052856, "time": 0.3956012487411499, "epoch": 3, "memory": 9465, "step": 34332} +{"lr": 0.0009757729755661012, "data_time": 0.18957347869873048, "grad_norm": 0.028835076210089027, "loss": 0.6858284890651702, "time": 0.37671647071838377, "epoch": 3, "memory": 9465, "step": 34432} +{"lr": 0.0009757729755661012, "data_time": 0.015145373344421387, "grad_norm": 0.024327561492100358, "loss": 0.6859240829944611, "time": 0.3810101509094238, "epoch": 3, "memory": 9465, "step": 34532} +{"accuracy/top1": 100.0, "data_time": 0.32107078327852134, "time": 0.5737238631528967, "step": 3} +{"lr": 0.0009460482294732422, "data_time": 0.0017899036407470702, "grad_norm": 0.03476463651750237, "loss": 0.6878685235977173, "time": 0.40660455226898196, "epoch": 4, "memory": 9465, "step": 34648} +{"lr": 0.0009460482294732422, "data_time": 0.0016024351119995118, "grad_norm": 0.034308757586404684, "loss": 0.6896538257598877, "time": 0.40057930946350095, "epoch": 4, "memory": 9465, "step": 34748} +{"lr": 0.0009460482294732422, "data_time": 0.018561601638793945, "grad_norm": 0.014790708804503084, "loss": 0.6880465269088745, "time": 0.4655344247817993, "epoch": 4, "memory": 9465, "step": 34848} +{"lr": 0.0009460482294732422, "data_time": 0.0016839027404785157, "grad_norm": 0.01557667445158586, "loss": 0.6876308202743531, "time": 0.47255942821502683, "epoch": 4, "memory": 9465, "step": 34948} +{"lr": 0.0009460482294732422, "data_time": 0.00119781494140625, "grad_norm": 0.02648501314688474, "loss": 0.678122740983963, "time": 0.4211986780166626, "epoch": 4, "memory": 9465, "step": 35048} +{"lr": 0.0009460482294732422, "data_time": 0.0012935400009155273, "grad_norm": 0.020421169931069016, "loss": 0.6842995524406433, "time": 0.4139136075973511, "epoch": 4, "memory": 9465, "step": 35148} +{"lr": 0.0009460482294732422, "data_time": 0.5352178335189819, "grad_norm": 0.028799089195672423, "loss": 0.6834197521209717, "time": 0.7197421073913575, "epoch": 4, "memory": 9465, "step": 35248} +{"lr": 0.0009460482294732422, "data_time": 0.22021760940551757, "grad_norm": 0.02672121240757406, "loss": 0.68721644282341, "time": 0.40677647590637206, "epoch": 4, "memory": 9465, "step": 35348} +{"lr": 0.0009460482294732422, "data_time": 0.17384004592895508, "grad_norm": 0.023233586567221208, "loss": 0.6831177651882172, "time": 0.3568373918533325, "epoch": 4, "memory": 9465, "step": 35448} +{"lr": 0.0009460482294732422, "data_time": 0.0783167839050293, "grad_norm": 0.01866559034679085, "loss": 0.6832765281200409, "time": 0.3714780569076538, "epoch": 4, "memory": 9465, "step": 35548} +{"lr": 0.0009460482294732422, "data_time": 0.001073312759399414, "grad_norm": 0.019859836203977466, "loss": 0.6869368433952332, "time": 0.403188419342041, "epoch": 4, "memory": 9465, "step": 35648} +{"lr": 0.0009460482294732422, "data_time": 0.001113748550415039, "grad_norm": 0.018139785109087824, "loss": 0.6897226393222808, "time": 0.43632261753082274, "epoch": 4, "memory": 9465, "step": 35748} +{"lr": 0.0009460482294732422, "data_time": 0.0010857343673706054, "grad_norm": 0.026441440032795072, "loss": 0.6826783895492554, "time": 0.41672937870025634, "epoch": 4, "memory": 9465, "step": 35848} +{"lr": 0.0009460482294732422, "data_time": 0.001102590560913086, "grad_norm": 0.01604182834853418, "loss": 0.6827332258224488, "time": 0.4156961917877197, "epoch": 4, "memory": 9465, "step": 35948} +{"lr": 0.0009460482294732422, "data_time": 0.0011310815811157227, "grad_norm": 0.03293159557506442, "loss": 0.6906485855579376, "time": 0.3799347162246704, "epoch": 4, "memory": 9465, "step": 36048} +{"lr": 0.0009460482294732422, "data_time": 0.001097869873046875, "grad_norm": 0.012823704117909073, "loss": 0.6866336822509765, "time": 0.4036120891571045, "epoch": 4, "memory": 9465, "step": 36148} +{"lr": 0.0009460482294732422, "data_time": 0.0010981082916259766, "grad_norm": 0.03432498765178025, "loss": 0.6885882139205932, "time": 0.3947927951812744, "epoch": 4, "memory": 9465, "step": 36248} +{"lr": 0.0009460482294732422, "data_time": 0.0011486768722534179, "grad_norm": 0.037587924674153325, "loss": 0.6901862502098084, "time": 0.4117638111114502, "epoch": 4, "memory": 9465, "step": 36348} +{"lr": 0.0009460482294732422, "data_time": 0.0011402368545532227, "grad_norm": 0.02379846046678722, "loss": 0.6881829559803009, "time": 0.35527381896972654, "epoch": 4, "memory": 9465, "step": 36448} +{"lr": 0.0009460482294732422, "data_time": 0.001129889488220215, "grad_norm": 0.018451213405933233, "loss": 0.6899665355682373, "time": 0.4215824842453003, "epoch": 4, "memory": 9465, "step": 36548} +{"lr": 0.0009460482294732422, "data_time": 0.0011128902435302735, "grad_norm": 0.02452612491324544, "loss": 0.6846395015716553, "time": 0.41646690368652345, "epoch": 4, "memory": 9465, "step": 36648} +{"lr": 0.0009460482294732422, "data_time": 0.0012495040893554688, "grad_norm": 0.018059630505740644, "loss": 0.6883924603462219, "time": 0.3853238582611084, "epoch": 4, "memory": 9465, "step": 36748} +{"lr": 0.0009460482294732422, "data_time": 0.0011029958724975586, "grad_norm": 0.027530170959653332, "loss": 0.6902863144874573, "time": 0.3810310363769531, "epoch": 4, "memory": 9465, "step": 36848} +{"lr": 0.0009460482294732422, "data_time": 0.0012690305709838867, "grad_norm": 0.024206779804080723, "loss": 0.6883936524391174, "time": 0.377858829498291, "epoch": 4, "memory": 9465, "step": 36948} +{"lr": 0.0009460482294732422, "data_time": 0.0012712955474853515, "grad_norm": 0.023071225523017345, "loss": 0.6838258504867554, "time": 0.4320306062698364, "epoch": 4, "memory": 9465, "step": 37048} +{"lr": 0.0009460482294732422, "data_time": 0.0011139869689941405, "grad_norm": 0.03565419996157289, "loss": 0.6902921795845032, "time": 0.4058772802352905, "epoch": 4, "memory": 9465, "step": 37148} +{"lr": 0.0009460482294732422, "data_time": 0.0011394023895263672, "grad_norm": 0.02153901937417686, "loss": 0.6827071964740753, "time": 0.35878307819366456, "epoch": 4, "memory": 9465, "step": 37248} +{"lr": 0.0009460482294732422, "data_time": 0.001119542121887207, "grad_norm": 0.026963997120037674, "loss": 0.6893606722354889, "time": 0.3967958927154541, "epoch": 4, "memory": 9465, "step": 37348} +{"lr": 0.0009460482294732422, "data_time": 0.0011075496673583984, "grad_norm": 0.02151414748514071, "loss": 0.6861848771572113, "time": 0.4284475803375244, "epoch": 4, "memory": 9465, "step": 37448} +{"lr": 0.0009460482294732422, "data_time": 0.0010763883590698242, "grad_norm": 0.015718328609364108, "loss": 0.6851791739463806, "time": 0.3692357301712036, "epoch": 4, "memory": 9465, "step": 37548} +{"lr": 0.0009460482294732422, "data_time": 0.0010373353958129882, "grad_norm": 0.02049769986188039, "loss": 0.6843901336193084, "time": 0.41892502307891843, "epoch": 4, "memory": 9465, "step": 37648} +{"lr": 0.0009460482294732422, "data_time": 0.0010868310928344727, "grad_norm": 0.02394439820200205, "loss": 0.6859405159950256, "time": 0.44752552509307864, "epoch": 4, "memory": 9465, "step": 37748} +{"lr": 0.0009460482294732422, "data_time": 0.001164698600769043, "grad_norm": 0.017176639661192893, "loss": 0.6845137417316437, "time": 0.43931078910827637, "epoch": 4, "memory": 9465, "step": 37848} +{"lr": 0.0009460482294732422, "data_time": 0.0011846780776977538, "grad_norm": 0.02280675176298246, "loss": 0.6860374450683594, "time": 0.36711075305938723, "epoch": 4, "memory": 9465, "step": 37948} +{"lr": 0.0009460482294732422, "data_time": 0.0012263059616088867, "grad_norm": 0.0154698476777412, "loss": 0.6861841261386872, "time": 0.39556992053985596, "epoch": 4, "memory": 9465, "step": 38048} +{"lr": 0.0009460482294732422, "data_time": 0.0010687589645385742, "grad_norm": 0.02592717092484236, "loss": 0.6884170293807983, "time": 0.39520721435546874, "epoch": 4, "memory": 9465, "step": 38148} +{"lr": 0.0009460482294732422, "data_time": 0.0011953115463256836, "grad_norm": 0.02128877486102283, "loss": 0.6879209697246551, "time": 0.4146375894546509, "epoch": 4, "memory": 9465, "step": 38248} +{"lr": 0.0009460482294732422, "data_time": 0.0011084794998168946, "grad_norm": 0.02099588655983098, "loss": 0.6900817453861237, "time": 0.4229058504104614, "epoch": 4, "memory": 9465, "step": 38348} +{"lr": 0.0009460482294732422, "data_time": 0.001208925247192383, "grad_norm": 0.020350764226168393, "loss": 0.6883242309093476, "time": 0.4033116102218628, "epoch": 4, "memory": 9465, "step": 38448} +{"lr": 0.0009460482294732422, "data_time": 0.0010816335678100586, "grad_norm": 0.02382119637914002, "loss": 0.6869382739067078, "time": 0.3833896636962891, "epoch": 4, "memory": 9465, "step": 38548} +{"lr": 0.0009460482294732422, "data_time": 0.0010337591171264648, "grad_norm": 0.027084255684167148, "loss": 0.6859582245349884, "time": 0.40285491943359375, "epoch": 4, "memory": 9465, "step": 38648} +{"lr": 0.0009460482294732422, "data_time": 0.0010890960693359375, "grad_norm": 0.025976664526388048, "loss": 0.6888171792030334, "time": 0.3666045904159546, "epoch": 4, "memory": 9465, "step": 38748} +{"lr": 0.0009460482294732422, "data_time": 0.001171875, "grad_norm": 0.026826502894982694, "loss": 0.6836021900177002, "time": 0.3953967332839966, "epoch": 4, "memory": 9465, "step": 38848} +{"lr": 0.0009460482294732422, "data_time": 0.0012588024139404297, "grad_norm": 0.01330707217566669, "loss": 0.6877638936042786, "time": 0.4548689842224121, "epoch": 4, "memory": 9465, "step": 38948} +{"lr": 0.0009460482294732422, "data_time": 0.0010664939880371093, "grad_norm": 0.01710113591980189, "loss": 0.6892962992191315, "time": 0.4607393741607666, "epoch": 4, "memory": 9465, "step": 39048} +{"lr": 0.0009460482294732422, "data_time": 0.0012426376342773438, "grad_norm": 0.03672229354269803, "loss": 0.686964362859726, "time": 0.39701979160308837, "epoch": 4, "memory": 9465, "step": 39148} +{"lr": 0.0009460482294732422, "data_time": 0.0012912750244140625, "grad_norm": 0.02750856992788613, "loss": 0.6825249552726745, "time": 0.3837315082550049, "epoch": 4, "memory": 9465, "step": 39248} +{"lr": 0.0009460482294732422, "data_time": 0.0011709928512573242, "grad_norm": 0.02512131310068071, "loss": 0.6823361039161682, "time": 0.4443613290786743, "epoch": 4, "memory": 9465, "step": 39348} +{"lr": 0.0009460482294732422, "data_time": 0.0011461257934570312, "grad_norm": 0.017460952047258615, "loss": 0.6841272652149201, "time": 0.43163836002349854, "epoch": 4, "memory": 9465, "step": 39448} +{"lr": 0.0009460482294732422, "data_time": 0.0012474775314331055, "grad_norm": 0.015957727911882102, "loss": 0.6873302578926086, "time": 0.4154085874557495, "epoch": 4, "memory": 9465, "step": 39548} +{"lr": 0.0009460482294732422, "data_time": 0.001211071014404297, "grad_norm": 0.01755048946943134, "loss": 0.6892419338226319, "time": 0.3902950048446655, "epoch": 4, "memory": 9465, "step": 39648} +{"lr": 0.0009460482294732422, "data_time": 0.0011631965637207032, "grad_norm": 0.02180348220281303, "loss": 0.6907113075256348, "time": 0.4171720027923584, "epoch": 4, "memory": 9465, "step": 39748} +{"lr": 0.0009460482294732422, "data_time": 0.0012620449066162109, "grad_norm": 0.016116890986450017, "loss": 0.6853848397731781, "time": 0.4402278184890747, "epoch": 4, "memory": 9465, "step": 39848} +{"lr": 0.0009460482294732422, "data_time": 0.0011441469192504882, "grad_norm": 0.023416503658518197, "loss": 0.6844816923141479, "time": 0.3723562717437744, "epoch": 4, "memory": 9465, "step": 39948} +{"lr": 0.0009460482294732422, "data_time": 0.0010658740997314454, "grad_norm": 0.01867930297739804, "loss": 0.6819418549537659, "time": 0.4076195478439331, "epoch": 4, "memory": 9465, "step": 40048} +{"lr": 0.0009460482294732422, "data_time": 0.0011111974716186523, "grad_norm": 0.013838802732061594, "loss": 0.6878926634788514, "time": 0.41741237640380857, "epoch": 4, "memory": 9465, "step": 40148} +{"lr": 0.0009460482294732422, "data_time": 0.0011829376220703126, "grad_norm": 0.02090329332277179, "loss": 0.6850333929061889, "time": 0.4069153070449829, "epoch": 4, "memory": 9465, "step": 40248} +{"lr": 0.0009460482294732422, "data_time": 0.0013393640518188476, "grad_norm": 0.017084522580262274, "loss": 0.683957862854004, "time": 0.38308372497558596, "epoch": 4, "memory": 9465, "step": 40348} +{"lr": 0.0009460482294732422, "data_time": 0.0011396169662475585, "grad_norm": 0.031805541133508086, "loss": 0.6905127882957458, "time": 0.4001450777053833, "epoch": 4, "memory": 9465, "step": 40448} +{"lr": 0.0009460482294732422, "data_time": 0.0011447668075561523, "grad_norm": 0.013711705664172768, "loss": 0.6867374300956726, "time": 0.43746190071105956, "epoch": 4, "memory": 9465, "step": 40548} +{"lr": 0.0009460482294732422, "data_time": 0.0010962963104248046, "grad_norm": 0.028451330936513843, "loss": 0.68432577252388, "time": 0.4556013822555542, "epoch": 4, "memory": 9465, "step": 40648} +{"lr": 0.0009460482294732422, "data_time": 0.001048898696899414, "grad_norm": 0.01319067805306986, "loss": 0.6893169283866882, "time": 0.421393084526062, "epoch": 4, "memory": 9465, "step": 40748} +{"lr": 0.0009460482294732422, "data_time": 0.0011313676834106446, "grad_norm": 0.029743977077305316, "loss": 0.6813984453678131, "time": 0.3892336130142212, "epoch": 4, "memory": 9465, "step": 40848} +{"lr": 0.0009460482294732422, "data_time": 0.0010990142822265626, "grad_norm": 0.015915067866444588, "loss": 0.6880536556243897, "time": 0.3737616777420044, "epoch": 4, "memory": 9465, "step": 40948} +{"lr": 0.0009460482294732422, "data_time": 0.0011123180389404296, "grad_norm": 0.019405145803466438, "loss": 0.6847852945327759, "time": 0.44578964710235597, "epoch": 4, "memory": 9465, "step": 41048} +{"lr": 0.0009460482294732422, "data_time": 0.0009724617004394532, "grad_norm": 0.03085338226519525, "loss": 0.6798136293888092, "time": 0.4215927839279175, "epoch": 4, "memory": 9465, "step": 41148} +{"lr": 0.0009460482294732422, "data_time": 0.0011057138442993164, "grad_norm": 0.02757742926478386, "loss": 0.6880335152149201, "time": 0.3928692102432251, "epoch": 4, "memory": 9465, "step": 41248} +{"lr": 0.0009460482294732422, "data_time": 0.0012017011642456055, "grad_norm": 0.024022822733968496, "loss": 0.6902175903320312, "time": 0.4202180147171021, "epoch": 4, "memory": 9465, "step": 41348} +{"lr": 0.0009460482294732422, "data_time": 0.001481342315673828, "grad_norm": 0.030624729616101833, "loss": 0.6884732663631439, "time": 0.43123784065246584, "epoch": 4, "memory": 9465, "step": 41448} +{"lr": 0.0009460482294732422, "data_time": 0.0011002779006958007, "grad_norm": 0.028252162295393644, "loss": 0.685443776845932, "time": 0.40383124351501465, "epoch": 4, "memory": 9465, "step": 41548} +{"lr": 0.0009460482294732422, "data_time": 0.0011414051055908202, "grad_norm": 0.025519554503262043, "loss": 0.6831911206245422, "time": 0.42822425365447997, "epoch": 4, "memory": 9465, "step": 41648} +{"lr": 0.0009460482294732422, "data_time": 0.0011412382125854492, "grad_norm": 0.02709073149599135, "loss": 0.6852254331111908, "time": 0.43169431686401366, "epoch": 4, "memory": 9465, "step": 41748} +{"lr": 0.0009460482294732422, "data_time": 0.001149606704711914, "grad_norm": 0.019600513426121326, "loss": 0.6874465703964233, "time": 0.3934406042098999, "epoch": 4, "memory": 9465, "step": 41848} +{"lr": 0.0009460482294732422, "data_time": 0.0011408567428588868, "grad_norm": 0.02341573985759169, "loss": 0.6857075035572052, "time": 0.4015033483505249, "epoch": 4, "memory": 9465, "step": 41948} +{"lr": 0.0009460482294732422, "data_time": 0.001144099235534668, "grad_norm": 0.022814054088667036, "loss": 0.684168541431427, "time": 0.41434321403503416, "epoch": 4, "memory": 9465, "step": 42048} +{"lr": 0.0009460482294732422, "data_time": 0.001092243194580078, "grad_norm": 0.026281535206362606, "loss": 0.6876562416553498, "time": 0.4789563179016113, "epoch": 4, "memory": 9465, "step": 42148} +{"lr": 0.0009460482294732422, "data_time": 0.0011058092117309571, "grad_norm": 0.030756097449921072, "loss": 0.6876628339290619, "time": 0.40562398433685304, "epoch": 4, "memory": 9465, "step": 42248} +{"lr": 0.0009460482294732422, "data_time": 0.0010761499404907226, "grad_norm": 0.020979050220921634, "loss": 0.6930504322052002, "time": 0.44523024559020996, "epoch": 4, "memory": 9465, "step": 42348} +{"lr": 0.0009460482294732422, "data_time": 0.001087021827697754, "grad_norm": 0.02626004386693239, "loss": 0.6856265604496002, "time": 0.40987200736999513, "epoch": 4, "memory": 9465, "step": 42448} +{"lr": 0.0009460482294732422, "data_time": 0.001085042953491211, "grad_norm": 0.028808306576684116, "loss": 0.6887922346591949, "time": 0.4446295499801636, "epoch": 4, "memory": 9465, "step": 42548} +{"lr": 0.0009460482294732422, "data_time": 0.0010744810104370117, "grad_norm": 0.03397733778692782, "loss": 0.688660454750061, "time": 0.44204392433166506, "epoch": 4, "memory": 9465, "step": 42648} +{"lr": 0.0009460482294732422, "data_time": 0.0013291358947753907, "grad_norm": 0.021859666192904116, "loss": 0.6858417630195618, "time": 0.40098745822906495, "epoch": 4, "memory": 9465, "step": 42748} +{"lr": 0.0009460482294732422, "data_time": 0.0013178586959838867, "grad_norm": 0.028944857395254077, "loss": 0.6886160969734192, "time": 0.4137450933456421, "epoch": 4, "memory": 9465, "step": 42848} +{"lr": 0.0009460482294732422, "data_time": 0.0013623237609863281, "grad_norm": 0.033438387070782484, "loss": 0.6898542106151581, "time": 0.4142767429351807, "epoch": 4, "memory": 9465, "step": 42948} +{"lr": 0.0009460482294732422, "data_time": 0.00107574462890625, "grad_norm": 0.02505057826638222, "loss": 0.6842765569686889, "time": 0.4109280824661255, "epoch": 4, "memory": 9465, "step": 43048} +{"lr": 0.0009460482294732422, "data_time": 0.0012395858764648437, "grad_norm": 0.37447545626200734, "loss": 0.6884382665157318, "time": 0.40226519107818604, "epoch": 4, "memory": 9465, "step": 43148} +{"lr": 0.0009460482294732422, "data_time": 0.0012097835540771484, "grad_norm": 0.03758322820067406, "loss": 0.6867580950260163, "time": 0.39088571071624756, "epoch": 4, "memory": 9465, "step": 43248} +{"lr": 0.0009460482294732422, "data_time": 0.0010784387588500977, "grad_norm": 0.02749922494404018, "loss": 0.689893227815628, "time": 0.36574792861938477, "epoch": 4, "memory": 9465, "step": 43348} +{"lr": 0.0009460482294732422, "data_time": 0.0011917829513549804, "grad_norm": 0.027267923951148985, "loss": 0.6861893355846405, "time": 0.4374396324157715, "epoch": 4, "memory": 9465, "step": 43448} +{"lr": 0.0009460482294732422, "data_time": 0.0010781049728393554, "grad_norm": 0.018396165641024708, "loss": 0.6917386293411255, "time": 0.40561809539794924, "epoch": 4, "memory": 9465, "step": 43548} +{"lr": 0.0009460482294732422, "data_time": 0.001032567024230957, "grad_norm": 0.018677343521267174, "loss": 0.6840654134750366, "time": 0.3888076305389404, "epoch": 4, "memory": 9465, "step": 43648} +{"lr": 0.0009460482294732422, "data_time": 0.0011232852935791015, "grad_norm": 0.03051765011623502, "loss": 0.6902405619621277, "time": 0.45583975315093994, "epoch": 4, "memory": 9465, "step": 43748} +{"lr": 0.0009460482294732422, "data_time": 0.001131129264831543, "grad_norm": 0.022079009166918694, "loss": 0.6826335966587067, "time": 0.3682593107223511, "epoch": 4, "memory": 9465, "step": 43848} +{"lr": 0.0009460482294732422, "data_time": 0.0011067867279052734, "grad_norm": 0.0220866784802638, "loss": 0.6860295534133911, "time": 0.3993252754211426, "epoch": 4, "memory": 9465, "step": 43948} +{"lr": 0.0009460482294732422, "data_time": 0.0013279199600219726, "grad_norm": 0.020557160209864377, "loss": 0.6874822616577149, "time": 0.3809523344039917, "epoch": 4, "memory": 9465, "step": 44048} +{"lr": 0.0009460482294732422, "data_time": 0.0012639760971069336, "grad_norm": 0.017060778685845435, "loss": 0.6902158260345459, "time": 0.3801664113998413, "epoch": 4, "memory": 9465, "step": 44148} +{"lr": 0.0009460482294732422, "data_time": 0.0011796236038208007, "grad_norm": 0.015369787137024105, "loss": 0.6870765745639801, "time": 0.34963796138763426, "epoch": 4, "memory": 9465, "step": 44248} +{"lr": 0.0009460482294732422, "data_time": 0.0011382102966308594, "grad_norm": 0.031882503814995286, "loss": 0.6848923325538635, "time": 0.4465653657913208, "epoch": 4, "memory": 9465, "step": 44348} +{"lr": 0.0009460482294732422, "data_time": 0.0011010408401489259, "grad_norm": 0.019127347599714996, "loss": 0.6856461822986603, "time": 0.3779271364212036, "epoch": 4, "memory": 9465, "step": 44448} +{"lr": 0.0009460482294732422, "data_time": 0.0012026786804199218, "grad_norm": 0.024599857884459196, "loss": 0.6908772110939025, "time": 0.373524808883667, "epoch": 4, "memory": 9465, "step": 44548} +{"lr": 0.0009460482294732422, "data_time": 0.0012852907180786132, "grad_norm": 0.015454904607031495, "loss": 0.6853369414806366, "time": 0.4327890157699585, "epoch": 4, "memory": 9465, "step": 44648} +{"lr": 0.0009460482294732422, "data_time": 0.0012766599655151367, "grad_norm": 0.02203433304093778, "loss": 0.6851997554302216, "time": 0.41467597484588625, "epoch": 4, "memory": 9465, "step": 44748} +{"lr": 0.0009460482294732422, "data_time": 0.0013851642608642579, "grad_norm": 0.02830750960856676, "loss": 0.6890788674354553, "time": 0.38348381519317626, "epoch": 4, "memory": 9465, "step": 44848} +{"lr": 0.0009460482294732422, "data_time": 0.0010076999664306641, "grad_norm": 0.01999299090821296, "loss": 0.6843239605426789, "time": 0.4219272375106812, "epoch": 4, "memory": 9465, "step": 44948} +{"lr": 0.0009460482294732422, "data_time": 0.0011414527893066407, "grad_norm": 0.019654296827502547, "loss": 0.6835480213165284, "time": 0.3832509517669678, "epoch": 4, "memory": 9465, "step": 45048} +{"lr": 0.0009460482294732422, "data_time": 0.001023721694946289, "grad_norm": 0.021794243599288166, "loss": 0.686592161655426, "time": 0.4207120180130005, "epoch": 4, "memory": 9465, "step": 45148} +{"lr": 0.0009460482294732422, "data_time": 0.0012047767639160156, "grad_norm": 0.022075356566347182, "loss": 0.6841388285160065, "time": 0.3922042608261108, "epoch": 4, "memory": 9465, "step": 45248} +{"lr": 0.0009460482294732422, "data_time": 0.001503300666809082, "grad_norm": 0.014208039059303701, "loss": 0.6885696828365326, "time": 0.41503353118896485, "epoch": 4, "memory": 9465, "step": 45348} +{"lr": 0.0009460482294732422, "data_time": 0.0013016700744628907, "grad_norm": 0.024549993546679617, "loss": 0.6844257116317749, "time": 0.38327693939208984, "epoch": 4, "memory": 9465, "step": 45448} +{"lr": 0.0009460482294732422, "data_time": 0.0010725021362304687, "grad_norm": 0.030095902178436516, "loss": 0.6890177369117737, "time": 0.38603711128234863, "epoch": 4, "memory": 9465, "step": 45548} +{"lr": 0.0009460482294732422, "data_time": 0.001148390769958496, "grad_norm": 0.04237807409372181, "loss": 0.6897955179214478, "time": 0.3657516956329346, "epoch": 4, "memory": 9465, "step": 45648} +{"lr": 0.0009460482294732422, "data_time": 0.0011873483657836915, "grad_norm": 0.017647675471380353, "loss": 0.6829340755939484, "time": 0.3857662916183472, "epoch": 4, "memory": 9465, "step": 45748} +{"lr": 0.0009460482294732422, "data_time": 0.0018066167831420898, "grad_norm": 0.02500118047464639, "loss": 0.6883044838905334, "time": 0.3862330675125122, "epoch": 4, "memory": 9465, "step": 45848} +{"lr": 0.0009460482294732422, "data_time": 0.001381826400756836, "grad_norm": 0.029156508552841843, "loss": 0.6860745310783386, "time": 0.4018967628479004, "epoch": 4, "memory": 9465, "step": 45948} +{"lr": 0.0009460482294732422, "data_time": 0.007388973236083984, "grad_norm": 0.024827901367098094, "loss": 0.6878408789634705, "time": 0.3983702898025513, "epoch": 4, "memory": 9465, "step": 46048} +{"accuracy/top1": 100.0, "data_time": 0.2998020087971407, "time": 0.5581597861121682, "step": 4} +{"lr": 0.0009054634122155991, "data_time": 0.0012440919876098634, "grad_norm": 0.026587527780793607, "loss": 0.6880975723266601, "time": 0.41617484092712403, "epoch": 5, "memory": 9465, "step": 46164} +{"lr": 0.0009054634122155991, "data_time": 0.001371908187866211, "grad_norm": 0.029598460206761955, "loss": 0.6811716258525848, "time": 0.3711184024810791, "epoch": 5, "memory": 9465, "step": 46264} +{"lr": 0.0009054634122155991, "data_time": 0.0013388872146606445, "grad_norm": 0.00921646914212033, "loss": 0.6854901432991027, "time": 0.3951786756515503, "epoch": 5, "memory": 9465, "step": 46364} +{"lr": 0.0009054634122155991, "data_time": 0.0012554168701171876, "grad_norm": 0.02585256015881896, "loss": 0.6859726309776306, "time": 0.4122645139694214, "epoch": 5, "memory": 9465, "step": 46464} +{"lr": 0.0009054634122155991, "data_time": 0.0010745763778686524, "grad_norm": 0.029718164424411952, "loss": 0.6889049768447876, "time": 0.3991572856903076, "epoch": 5, "memory": 9465, "step": 46564} +{"lr": 0.0009054634122155991, "data_time": 0.0011084794998168946, "grad_norm": 0.027763966005295515, "loss": 0.6881490051746368, "time": 0.44162285327911377, "epoch": 5, "memory": 9465, "step": 46664} +{"lr": 0.0009054634122155991, "data_time": 0.0011112213134765625, "grad_norm": 0.018291403772309422, "loss": 0.6887192845344543, "time": 0.3962730407714844, "epoch": 5, "memory": 9465, "step": 46764} +{"lr": 0.0009054634122155991, "data_time": 0.0011777162551879882, "grad_norm": 0.01177821836899966, "loss": 0.6876420617103577, "time": 0.38205268383026125, "epoch": 5, "memory": 9465, "step": 46864} +{"lr": 0.0009054634122155991, "data_time": 0.0012171506881713868, "grad_norm": 0.01986559834331274, "loss": 0.6813242316246033, "time": 0.397896933555603, "epoch": 5, "memory": 9465, "step": 46964} +{"lr": 0.0009054634122155991, "data_time": 0.0010857582092285156, "grad_norm": 0.02631110083311796, "loss": 0.6859330415725708, "time": 0.37413601875305175, "epoch": 5, "memory": 9465, "step": 47064} +{"lr": 0.0009054634122155991, "data_time": 0.0010780811309814454, "grad_norm": 0.035640973143745217, "loss": 0.6877168595790863, "time": 0.3792332887649536, "epoch": 5, "memory": 9465, "step": 47164} +{"lr": 0.0009054634122155991, "data_time": 0.0010349273681640625, "grad_norm": 0.02035343919415027, "loss": 0.6863429069519043, "time": 0.40172040462493896, "epoch": 5, "memory": 9465, "step": 47264} +{"lr": 0.0009054634122155991, "data_time": 0.001057267189025879, "grad_norm": 0.01496747275814414, "loss": 0.6831871330738067, "time": 0.38187637329101565, "epoch": 5, "memory": 9465, "step": 47364} +{"lr": 0.0009054634122155991, "data_time": 0.0010881900787353515, "grad_norm": 0.015223153587430715, "loss": 0.68567755818367, "time": 0.3965941905975342, "epoch": 5, "memory": 9465, "step": 47464} +{"lr": 0.0009054634122155991, "data_time": 0.0010868310928344727, "grad_norm": 0.024140493012964725, "loss": 0.6896021783351898, "time": 0.4375352144241333, "epoch": 5, "memory": 9465, "step": 47564} +{"lr": 0.0009054634122155991, "data_time": 0.0011009931564331054, "grad_norm": 0.017783907242119312, "loss": 0.6912515997886658, "time": 0.4043778657913208, "epoch": 5, "memory": 9465, "step": 47664} +{"lr": 0.0009054634122155991, "data_time": 0.0010442495346069335, "grad_norm": 0.0255945828743279, "loss": 0.6877342224121094, "time": 0.40841400623321533, "epoch": 5, "memory": 9465, "step": 47764} +{"lr": 0.0009054634122155991, "data_time": 0.0011137723922729492, "grad_norm": 0.02683678106404841, "loss": 0.6868508338928223, "time": 0.381062650680542, "epoch": 5, "memory": 9465, "step": 47864} +{"lr": 0.0009054634122155991, "data_time": 0.001055598258972168, "grad_norm": 0.020497567194979638, "loss": 0.6814903676509857, "time": 0.3894315004348755, "epoch": 5, "memory": 9465, "step": 47964} +{"lr": 0.0009054634122155991, "data_time": 0.0011021852493286132, "grad_norm": 0.031537147238850594, "loss": 0.6888923287391663, "time": 0.3879029035568237, "epoch": 5, "memory": 9465, "step": 48064} +{"lr": 0.0009054634122155991, "data_time": 0.0010887861251831054, "grad_norm": 0.022996595688164236, "loss": 0.6833690822124481, "time": 0.41205859184265137, "epoch": 5, "memory": 9465, "step": 48164} +{"lr": 0.0009054634122155991, "data_time": 0.0011677503585815429, "grad_norm": 0.03175321174785495, "loss": 0.6875175476074219, "time": 0.3933622121810913, "epoch": 5, "memory": 9465, "step": 48264} +{"lr": 0.0009054634122155991, "data_time": 0.001057910919189453, "grad_norm": 0.021280095819383858, "loss": 0.687904554605484, "time": 0.3971649408340454, "epoch": 5, "memory": 9465, "step": 48364} +{"lr": 0.0009054634122155991, "data_time": 0.001323866844177246, "grad_norm": 0.022431167401373388, "loss": 0.6828836560249328, "time": 0.39847261905670167, "epoch": 5, "memory": 9465, "step": 48464} +{"lr": 0.0009054634122155991, "data_time": 0.0011553049087524414, "grad_norm": 0.026961484737694264, "loss": 0.6908922612667083, "time": 0.37170188426971434, "epoch": 5, "memory": 9465, "step": 48564} +{"lr": 0.0009054634122155991, "data_time": 0.001134490966796875, "grad_norm": 0.028834646474570035, "loss": 0.6875865757465363, "time": 0.4479071855545044, "epoch": 5, "memory": 9465, "step": 48664} +{"lr": 0.0009054634122155991, "data_time": 0.001175093650817871, "grad_norm": 0.020229836599901317, "loss": 0.6805278599262238, "time": 0.432504677772522, "epoch": 5, "memory": 9465, "step": 48764} +{"lr": 0.0009054634122155991, "data_time": 0.0011219501495361329, "grad_norm": 0.02679470703005791, "loss": 0.6873334586620331, "time": 0.383939790725708, "epoch": 5, "memory": 9465, "step": 48864} +{"lr": 0.0009054634122155991, "data_time": 0.0011063814163208008, "grad_norm": 0.03192573469132185, "loss": 0.6838356375694274, "time": 0.41294288635253906, "epoch": 5, "memory": 9465, "step": 48964} +{"lr": 0.0009054634122155991, "data_time": 0.0010793447494506837, "grad_norm": 0.0175002969102934, "loss": 0.6861941516399384, "time": 0.39840657711029054, "epoch": 5, "memory": 9465, "step": 49064} +{"lr": 0.0009054634122155991, "data_time": 0.0015295982360839845, "grad_norm": 0.010096137854270637, "loss": 0.6816293716430664, "time": 0.47495198249816895, "epoch": 5, "memory": 9465, "step": 49164} +{"lr": 0.0009054634122155991, "data_time": 0.0012670278549194336, "grad_norm": 0.02322162096388638, "loss": 0.6870709776878356, "time": 0.3975664615631104, "epoch": 5, "memory": 9465, "step": 49264} +{"lr": 0.0009054634122155991, "data_time": 0.0010991334915161134, "grad_norm": 0.028841956635005772, "loss": 0.6840360999107361, "time": 0.36883208751678465, "epoch": 5, "memory": 9465, "step": 49364} +{"lr": 0.0009054634122155991, "data_time": 0.0011717557907104492, "grad_norm": 0.025854251813143492, "loss": 0.6896846652030945, "time": 0.4344851732254028, "epoch": 5, "memory": 9465, "step": 49464} +{"lr": 0.0009054634122155991, "data_time": 0.0012343645095825196, "grad_norm": 0.03077122033573687, "loss": 0.6824256181716919, "time": 0.4099503755569458, "epoch": 5, "memory": 9465, "step": 49564} +{"lr": 0.0009054634122155991, "data_time": 0.0012484073638916015, "grad_norm": 0.02699918523430824, "loss": 0.6889729678630829, "time": 0.3828179359436035, "epoch": 5, "memory": 9465, "step": 49664} +{"lr": 0.0009054634122155991, "data_time": 0.001108837127685547, "grad_norm": 0.029219097923487425, "loss": 0.6889478445053101, "time": 0.36678876876831057, "epoch": 5, "memory": 9465, "step": 49764} +{"lr": 0.0009054634122155991, "data_time": 0.0010905265808105469, "grad_norm": 0.02324791522696614, "loss": 0.6852601051330567, "time": 0.43106679916381835, "epoch": 5, "memory": 9465, "step": 49864} +{"lr": 0.0009054634122155991, "data_time": 0.001279759407043457, "grad_norm": 0.02332518082112074, "loss": 0.6892894983291626, "time": 0.390034556388855, "epoch": 5, "memory": 9465, "step": 49964} +{"lr": 0.0009054634122155991, "data_time": 0.0011096477508544921, "grad_norm": 0.028612617636099458, "loss": 0.685692286491394, "time": 0.4111783981323242, "epoch": 5, "memory": 9465, "step": 50064} +{"lr": 0.0009054634122155991, "data_time": 0.0011332273483276368, "grad_norm": 0.030447939690202475, "loss": 0.6874166846275329, "time": 0.3804218053817749, "epoch": 5, "memory": 9465, "step": 50164} +{"lr": 0.0009054634122155991, "data_time": 0.0011574983596801757, "grad_norm": 0.02650841875001788, "loss": 0.6892519533634186, "time": 0.40076098442077634, "epoch": 5, "memory": 9465, "step": 50264} +{"lr": 0.0009054634122155991, "data_time": 0.001217055320739746, "grad_norm": 0.015318753337487579, "loss": 0.680166357755661, "time": 0.40390989780426023, "epoch": 5, "memory": 9465, "step": 50364} +{"lr": 0.0009054634122155991, "data_time": 0.0012067079544067383, "grad_norm": 0.026635066093876957, "loss": 0.6873058319091797, "time": 0.39262354373931885, "epoch": 5, "memory": 9465, "step": 50464} +{"lr": 0.0009054634122155991, "data_time": 0.0012506961822509766, "grad_norm": 0.0180527399701532, "loss": 0.6869944036006927, "time": 0.41787846088409425, "epoch": 5, "memory": 9465, "step": 50564} +{"lr": 0.0009054634122155991, "data_time": 0.0010959863662719726, "grad_norm": 0.01809955690987408, "loss": 0.6925374031066894, "time": 0.3686471939086914, "epoch": 5, "memory": 9465, "step": 50664} +{"lr": 0.0009054634122155991, "data_time": 0.0012143611907958984, "grad_norm": 0.023882671422325075, "loss": 0.6909743010997772, "time": 0.39217796325683596, "epoch": 5, "memory": 9465, "step": 50764} +{"lr": 0.0009054634122155991, "data_time": 0.0012526273727416991, "grad_norm": 0.03057125969789922, "loss": 0.6895804286003113, "time": 0.3557945013046265, "epoch": 5, "memory": 9465, "step": 50864} +{"lr": 0.0009054634122155991, "data_time": 0.0011016607284545898, "grad_norm": 0.022335830773226917, "loss": 0.6903132677078248, "time": 0.39890298843383787, "epoch": 5, "memory": 9465, "step": 50964} +{"lr": 0.0009054634122155991, "data_time": 0.0014529943466186524, "grad_norm": 0.026697930041700603, "loss": 0.6855130016803741, "time": 0.4034978151321411, "epoch": 5, "memory": 9465, "step": 51064} +{"lr": 0.0009054634122155991, "data_time": 0.0017690181732177735, "grad_norm": 0.025505088176578282, "loss": 0.6853675186634064, "time": 0.37899184226989746, "epoch": 5, "memory": 9465, "step": 51164} +{"lr": 0.0009054634122155991, "data_time": 0.0012786149978637694, "grad_norm": 0.034993641055189074, "loss": 0.681854385137558, "time": 0.41874496936798095, "epoch": 5, "memory": 9465, "step": 51264} +{"lr": 0.0009054634122155991, "data_time": 0.0013869285583496093, "grad_norm": 0.01751966478768736, "loss": 0.6884867131710053, "time": 0.38738670349121096, "epoch": 5, "memory": 9465, "step": 51364} +{"lr": 0.0009054634122155991, "data_time": 0.0012630462646484376, "grad_norm": 0.029562947619706393, "loss": 0.6862551987171173, "time": 0.4115957736968994, "epoch": 5, "memory": 9465, "step": 51464} +{"lr": 0.0009054634122155991, "data_time": 0.001076078414916992, "grad_norm": 0.02607720545493066, "loss": 0.6865394115447998, "time": 0.41392245292663576, "epoch": 5, "memory": 9465, "step": 51564} +{"lr": 0.0009054634122155991, "data_time": 0.0010185956954956055, "grad_norm": 0.031732980720698835, "loss": 0.6904223203659058, "time": 0.3920707941055298, "epoch": 5, "memory": 9465, "step": 51664} +{"lr": 0.0009054634122155991, "data_time": 0.0011585712432861327, "grad_norm": 0.016031731851398945, "loss": 0.6851561784744262, "time": 0.38847401142120364, "epoch": 5, "memory": 9465, "step": 51764} +{"lr": 0.0009054634122155991, "data_time": 0.0011475086212158203, "grad_norm": 0.01787509429268539, "loss": 0.6870312750339508, "time": 0.3833015441894531, "epoch": 5, "memory": 9465, "step": 51864} +{"lr": 0.0009054634122155991, "data_time": 0.0011997222900390625, "grad_norm": 0.022489652619697154, "loss": 0.6870958864688873, "time": 0.40016496181488037, "epoch": 5, "memory": 9465, "step": 51964} +{"lr": 0.0009054634122155991, "data_time": 0.0010323524475097656, "grad_norm": 0.025026615266688167, "loss": 0.6886358857154846, "time": 0.40814259052276614, "epoch": 5, "memory": 9465, "step": 52064} +{"lr": 0.0009054634122155991, "data_time": 0.0011286497116088866, "grad_norm": 0.0235617560101673, "loss": 0.6871602952480316, "time": 0.4007275104522705, "epoch": 5, "memory": 9465, "step": 52164} +{"lr": 0.0009054634122155991, "data_time": 0.001175069808959961, "grad_norm": 0.023988850094610825, "loss": 0.6903709650039673, "time": 0.40066406726837156, "epoch": 5, "memory": 9465, "step": 52264} +{"lr": 0.0009054634122155991, "data_time": 0.0011972904205322266, "grad_norm": 0.02772388606099412, "loss": 0.6898591458797455, "time": 0.4022976875305176, "epoch": 5, "memory": 9465, "step": 52364} +{"lr": 0.0009054634122155991, "data_time": 0.0011831283569335937, "grad_norm": 0.02243322432041168, "loss": 0.6849162459373475, "time": 0.4304262638092041, "epoch": 5, "memory": 9465, "step": 52464} +{"lr": 0.0009054634122155991, "data_time": 0.0012135744094848634, "grad_norm": 0.025075363041833044, "loss": 0.686655443906784, "time": 0.38414065837860106, "epoch": 5, "memory": 9465, "step": 52564} +{"lr": 0.0009054634122155991, "data_time": 0.0013298988342285156, "grad_norm": 0.020284536166582257, "loss": 0.6933638453483582, "time": 0.3798767328262329, "epoch": 5, "memory": 9465, "step": 52664} +{"lr": 0.0009054634122155991, "data_time": 0.0010272979736328125, "grad_norm": 0.036314313393086196, "loss": 0.6914688766002655, "time": 0.39915931224823, "epoch": 5, "memory": 9465, "step": 52764} +{"lr": 0.0009054634122155991, "data_time": 0.0012793779373168946, "grad_norm": 0.024953949404880403, "loss": 0.6863766312599182, "time": 0.39757044315338136, "epoch": 5, "memory": 9465, "step": 52864} +{"lr": 0.0009054634122155991, "data_time": 0.0011752843856811523, "grad_norm": 0.01782432678155601, "loss": 0.6870884716510772, "time": 0.38098416328430174, "epoch": 5, "memory": 9465, "step": 52964} +{"lr": 0.0009054634122155991, "data_time": 0.0010037899017333984, "grad_norm": 0.026098742755129932, "loss": 0.6859421968460083, "time": 0.4036207914352417, "epoch": 5, "memory": 9465, "step": 53064} +{"lr": 0.0009054634122155991, "data_time": 0.0011294126510620118, "grad_norm": 0.026348127191886304, "loss": 0.6870138049125671, "time": 0.40224056243896483, "epoch": 5, "memory": 9465, "step": 53164} +{"lr": 0.0009054634122155991, "data_time": 0.0011500835418701172, "grad_norm": 0.018034082953818142, "loss": 0.6889199793338776, "time": 0.39376053810119627, "epoch": 5, "memory": 9465, "step": 53264} +{"lr": 0.0009054634122155991, "data_time": 0.0010768413543701173, "grad_norm": 0.027831611060537397, "loss": 0.6899236142635345, "time": 0.4256305456161499, "epoch": 5, "memory": 9465, "step": 53364} +{"lr": 0.0009054634122155991, "data_time": 0.001101374626159668, "grad_norm": 0.026979637006297706, "loss": 0.6827156960964202, "time": 0.35997519493103025, "epoch": 5, "memory": 9465, "step": 53464} +{"lr": 0.0009054634122155991, "data_time": 0.001135849952697754, "grad_norm": 0.02476250040344894, "loss": 0.6896296381950379, "time": 0.38420493602752687, "epoch": 5, "memory": 9465, "step": 53564} +{"lr": 0.0009054634122155991, "data_time": 0.0011581897735595704, "grad_norm": 0.02668171711266041, "loss": 0.6884955704212189, "time": 0.3978543281555176, "epoch": 5, "memory": 9465, "step": 53664} +{"lr": 0.0009054634122155991, "data_time": 0.0011167526245117188, "grad_norm": 0.021480609849095344, "loss": 0.6848095834255219, "time": 0.42375526428222654, "epoch": 5, "memory": 9465, "step": 53764} +{"lr": 0.0009054634122155991, "data_time": 0.0011123418807983398, "grad_norm": 0.017277513444423676, "loss": 0.6869917929172515, "time": 0.3655188798904419, "epoch": 5, "memory": 9465, "step": 53864} +{"lr": 0.0009054634122155991, "data_time": 0.001119375228881836, "grad_norm": 0.019226023933151738, "loss": 0.6887372136116028, "time": 0.38648648262023927, "epoch": 5, "memory": 9465, "step": 53964} +{"lr": 0.0009054634122155991, "data_time": 0.0011758089065551757, "grad_norm": 0.019273080118000507, "loss": 0.6879786312580108, "time": 0.39363129138946534, "epoch": 5, "memory": 9465, "step": 54064} +{"lr": 0.0009054634122155991, "data_time": 0.0011520147323608398, "grad_norm": 0.02052807351574302, "loss": 0.6910687863826752, "time": 0.3610013723373413, "epoch": 5, "memory": 9465, "step": 54164} +{"lr": 0.0009054634122155991, "data_time": 0.0012211084365844726, "grad_norm": 0.022979992744512855, "loss": 0.6904963552951813, "time": 0.38689513206481935, "epoch": 5, "memory": 9465, "step": 54264} +{"lr": 0.0009054634122155991, "data_time": 0.001228165626525879, "grad_norm": 0.016480088559910656, "loss": 0.688277804851532, "time": 0.3878389120101929, "epoch": 5, "memory": 9465, "step": 54364} +{"lr": 0.0009054634122155991, "data_time": 0.0011260271072387694, "grad_norm": 0.018702167994342746, "loss": 0.6865172922611237, "time": 0.3749183177947998, "epoch": 5, "memory": 9465, "step": 54464} +{"lr": 0.0009054634122155991, "data_time": 0.0011040449142456054, "grad_norm": 0.029707753891125323, "loss": 0.6900529026985168, "time": 0.39838101863861086, "epoch": 5, "memory": 9465, "step": 54564} +{"lr": 0.0009054634122155991, "data_time": 0.001137709617614746, "grad_norm": 0.014718397427350283, "loss": 0.6826073586940765, "time": 0.39407494068145754, "epoch": 5, "memory": 9465, "step": 54664} +{"lr": 0.0009054634122155991, "data_time": 0.0011273622512817383, "grad_norm": 0.02002229744102806, "loss": 0.6886222124099731, "time": 0.39647731781005857, "epoch": 5, "memory": 9465, "step": 54764} +{"lr": 0.0009054634122155991, "data_time": 0.0011443614959716797, "grad_norm": 0.018369539640843868, "loss": 0.6849116742610931, "time": 0.4018464326858521, "epoch": 5, "memory": 9465, "step": 54864} +{"lr": 0.0009054634122155991, "data_time": 0.001343512535095215, "grad_norm": 0.030916664004325866, "loss": 0.6881608724594116, "time": 0.4080906867980957, "epoch": 5, "memory": 9465, "step": 54964} +{"lr": 0.0009054634122155991, "data_time": 0.001047801971435547, "grad_norm": 0.02816597940400243, "loss": 0.6849474430084228, "time": 0.4051995038986206, "epoch": 5, "memory": 9465, "step": 55064} +{"lr": 0.0009054634122155991, "data_time": 0.0010867834091186524, "grad_norm": 0.022959691612049936, "loss": 0.6809873819351197, "time": 0.4346762657165527, "epoch": 5, "memory": 9465, "step": 55164} +{"lr": 0.0009054634122155991, "data_time": 0.1966933488845825, "grad_norm": 0.027581992093473673, "loss": 0.6840690612792969, "time": 0.3958333730697632, "epoch": 5, "memory": 9465, "step": 55264} +{"lr": 0.0009054634122155991, "data_time": 0.2462209939956665, "grad_norm": 0.02065160626079887, "loss": 0.6868190467357635, "time": 0.4308686971664429, "epoch": 5, "memory": 9465, "step": 55364} +{"lr": 0.0009054634122155991, "data_time": 0.1575378179550171, "grad_norm": 0.017900220002047718, "loss": 0.6868641614913941, "time": 0.3400084972381592, "epoch": 5, "memory": 9465, "step": 55464} +{"lr": 0.0009054634122155991, "data_time": 0.0013090133666992187, "grad_norm": 0.027901506843045355, "loss": 0.6848702013492585, "time": 0.3926393032073975, "epoch": 5, "memory": 9465, "step": 55564} +{"lr": 0.0009054634122155991, "data_time": 0.0011811971664428711, "grad_norm": 0.011158675141632556, "loss": 0.6929126560688019, "time": 0.37322607040405276, "epoch": 5, "memory": 9465, "step": 55664} +{"lr": 0.0009054634122155991, "data_time": 0.001319718360900879, "grad_norm": 0.032238740008324385, "loss": 0.6909414887428283, "time": 0.38470749855041503, "epoch": 5, "memory": 9465, "step": 55764} +{"lr": 0.0009054634122155991, "data_time": 0.0014949798583984374, "grad_norm": 0.017692160699516533, "loss": 0.6872556626796722, "time": 0.40132732391357423, "epoch": 5, "memory": 9465, "step": 55864} +{"lr": 0.0009054634122155991, "data_time": 0.0016646623611450196, "grad_norm": 0.024927551858127117, "loss": 0.6894688308238983, "time": 0.388171648979187, "epoch": 5, "memory": 9465, "step": 55964} +{"lr": 0.0009054634122155991, "data_time": 0.001196146011352539, "grad_norm": 0.028992805175948887, "loss": 0.6850293695926666, "time": 0.3908120632171631, "epoch": 5, "memory": 9465, "step": 56064} +{"lr": 0.0009054634122155991, "data_time": 0.0011138200759887695, "grad_norm": 0.02397423905786127, "loss": 0.687543261051178, "time": 0.3826998949050903, "epoch": 5, "memory": 9465, "step": 56164} +{"lr": 0.0009054634122155991, "data_time": 0.001337432861328125, "grad_norm": 0.026055848505347968, "loss": 0.6829025328159333, "time": 0.4017136335372925, "epoch": 5, "memory": 9465, "step": 56264} +{"lr": 0.0009054634122155991, "data_time": 0.0011916875839233399, "grad_norm": 0.02607398647814989, "loss": 0.6879303753376007, "time": 0.40138583183288573, "epoch": 5, "memory": 9465, "step": 56364} +{"lr": 0.0009054634122155991, "data_time": 0.0011496305465698241, "grad_norm": 0.024865913321264088, "loss": 0.6895306766033172, "time": 0.39088757038116456, "epoch": 5, "memory": 9465, "step": 56464} +{"lr": 0.0009054634122155991, "data_time": 0.0011301040649414062, "grad_norm": 0.02537779277190566, "loss": 0.687163770198822, "time": 0.39146401882171633, "epoch": 5, "memory": 9465, "step": 56564} +{"lr": 0.0009054634122155991, "data_time": 0.0012575626373291016, "grad_norm": 0.02295676473295316, "loss": 0.6881832003593444, "time": 0.3909043550491333, "epoch": 5, "memory": 9465, "step": 56664} +{"lr": 0.0009054634122155991, "data_time": 0.001027393341064453, "grad_norm": 0.024595651170238854, "loss": 0.6884438097476959, "time": 0.39977173805236815, "epoch": 5, "memory": 9465, "step": 56764} +{"lr": 0.0009054634122155991, "data_time": 0.0011107921600341797, "grad_norm": 0.026656482857652007, "loss": 0.6888610482215881, "time": 0.44601051807403563, "epoch": 5, "memory": 9465, "step": 56864} +{"lr": 0.0009054634122155991, "data_time": 0.0010900020599365235, "grad_norm": 0.0261293675750494, "loss": 0.6898661613464355, "time": 0.4157944917678833, "epoch": 5, "memory": 9465, "step": 56964} +{"lr": 0.0009054634122155991, "data_time": 0.0012702226638793945, "grad_norm": 0.024139224365353585, "loss": 0.6927463173866272, "time": 0.381671667098999, "epoch": 5, "memory": 9465, "step": 57064} +{"lr": 0.0009054634122155991, "data_time": 0.001151442527770996, "grad_norm": 0.028337081242352725, "loss": 0.6903347849845887, "time": 0.40841495990753174, "epoch": 5, "memory": 9465, "step": 57164} +{"lr": 0.0009054634122155991, "data_time": 0.001214742660522461, "grad_norm": 0.024786919355392456, "loss": 0.6854541897773743, "time": 0.3946031332015991, "epoch": 5, "memory": 9465, "step": 57264} +{"lr": 0.0009054634122155991, "data_time": 0.0011946439743041992, "grad_norm": 0.023413435183465482, "loss": 0.6880646169185638, "time": 0.37474076747894286, "epoch": 5, "memory": 9465, "step": 57364} +{"lr": 0.0009054634122155991, "data_time": 0.0018743515014648438, "grad_norm": 0.018026468594325707, "loss": 0.6855308651924134, "time": 0.43295631408691404, "epoch": 5, "memory": 9465, "step": 57464} +{"lr": 0.0009054634122155991, "data_time": 0.007212376594543457, "grad_norm": 0.016114775411551818, "loss": 0.6845888435840607, "time": 0.39299988746643066, "epoch": 5, "memory": 9465, "step": 57564} +{"accuracy/top1": 100.0, "data_time": 0.3096703641554889, "time": 0.5618760445538689, "step": 5} +{"lr": 0.0008550178566873411, "data_time": 0.10306141376495362, "grad_norm": 0.02607856991235167, "loss": 0.6887886345386505, "time": 0.3675510883331299, "epoch": 6, "memory": 9465, "step": 57680} +{"lr": 0.0008550178566873411, "data_time": 0.0011992931365966796, "grad_norm": 0.016421881050337105, "loss": 0.6853217840194702, "time": 0.3881377696990967, "epoch": 6, "memory": 9465, "step": 57780} +{"lr": 0.0008550178566873411, "data_time": 0.001097393035888672, "grad_norm": 0.02655152587685734, "loss": 0.6838648736476898, "time": 0.3818311929702759, "epoch": 6, "memory": 9465, "step": 57880} +{"lr": 0.0008550178566873411, "data_time": 0.0011381149291992188, "grad_norm": 0.024827568465843795, "loss": 0.6888383448123931, "time": 0.39320361614227295, "epoch": 6, "memory": 9465, "step": 57980} +{"lr": 0.0008550178566873411, "data_time": 0.001102280616760254, "grad_norm": 0.023271420691162347, "loss": 0.6845191359519959, "time": 0.40198745727539065, "epoch": 6, "memory": 9465, "step": 58080} +{"lr": 0.0008550178566873411, "data_time": 0.0012479782104492187, "grad_norm": 0.014999636285938322, "loss": 0.6901805341243744, "time": 0.39585835933685304, "epoch": 6, "memory": 9465, "step": 58180} +{"lr": 0.0008550178566873411, "data_time": 0.001110243797302246, "grad_norm": 0.019241804839111865, "loss": 0.6877995491027832, "time": 0.36391875743865965, "epoch": 6, "memory": 9465, "step": 58280} +{"lr": 0.0008550178566873411, "data_time": 0.0010864019393920898, "grad_norm": 0.009412611136212945, "loss": 0.6884370386600495, "time": 0.3892103433609009, "epoch": 6, "memory": 9465, "step": 58380} +{"lr": 0.0008550178566873411, "data_time": 0.0011346101760864257, "grad_norm": 0.03503970931051299, "loss": 0.6902829885482789, "time": 0.41189630031585694, "epoch": 6, "memory": 9465, "step": 58480} +{"lr": 0.0008550178566873411, "data_time": 0.0011670827865600587, "grad_norm": 0.03020229073590599, "loss": 0.6892358183860778, "time": 0.4098546028137207, "epoch": 6, "memory": 9465, "step": 58580} +{"lr": 0.0008550178566873411, "data_time": 0.001118946075439453, "grad_norm": 0.021369007194880395, "loss": 0.6875794708728791, "time": 0.37779951095581055, "epoch": 6, "memory": 9465, "step": 58680} +{"lr": 0.0008550178566873411, "data_time": 0.0010996103286743165, "grad_norm": 0.02368443019222468, "loss": 0.6920821368694305, "time": 0.3939052581787109, "epoch": 6, "memory": 9465, "step": 58780} +{"lr": 0.0008550178566873411, "data_time": 0.0011996746063232422, "grad_norm": 0.011431300081312656, "loss": 0.687268203496933, "time": 0.4462315559387207, "epoch": 6, "memory": 9465, "step": 58880} +{"lr": 0.0008550178566873411, "data_time": 0.0011156082153320312, "grad_norm": 0.03080502925440669, "loss": 0.6822718024253845, "time": 0.38296535015106203, "epoch": 6, "memory": 9465, "step": 58980} +{"lr": 0.0008550178566873411, "data_time": 0.0011321067810058593, "grad_norm": 0.020110561919864267, "loss": 0.6852346539497376, "time": 0.40460069179534913, "epoch": 6, "memory": 9465, "step": 59080} +{"lr": 0.0008550178566873411, "data_time": 0.002657508850097656, "grad_norm": 0.01642611682182178, "loss": 0.6863704025745392, "time": 0.3737973213195801, "epoch": 6, "memory": 9465, "step": 59180} +{"lr": 0.0008550178566873411, "data_time": 0.0011547327041625977, "grad_norm": 0.022235743375495075, "loss": 0.6865163683891297, "time": 0.4362701654434204, "epoch": 6, "memory": 9465, "step": 59280} +{"lr": 0.0008550178566873411, "data_time": 0.001006746292114258, "grad_norm": 0.026628897758200765, "loss": 0.6915045559406281, "time": 0.4014540672302246, "epoch": 6, "memory": 9465, "step": 59380} +{"lr": 0.0008550178566873411, "data_time": 0.0010770320892333984, "grad_norm": 0.02739627774572, "loss": 0.686798894405365, "time": 0.4184353113174438, "epoch": 6, "memory": 9465, "step": 59480} +{"lr": 0.0008550178566873411, "data_time": 0.0010686874389648437, "grad_norm": 0.021532252011820675, "loss": 0.6933930337429046, "time": 0.41991171836853025, "epoch": 6, "memory": 9465, "step": 59580} +{"lr": 0.0008550178566873411, "data_time": 0.001038193702697754, "grad_norm": 0.0282734930049628, "loss": 0.6866487681865692, "time": 0.38993492126464846, "epoch": 6, "memory": 9465, "step": 59680} +{"lr": 0.0008550178566873411, "data_time": 0.0011579275131225586, "grad_norm": 0.0163154071662575, "loss": 0.6848962366580963, "time": 0.4043025732040405, "epoch": 6, "memory": 9465, "step": 59780} +{"lr": 0.0008550178566873411, "data_time": 0.0010789871215820313, "grad_norm": 0.02124893106520176, "loss": 0.6870622158050537, "time": 0.4309748888015747, "epoch": 6, "memory": 9465, "step": 59880} +{"lr": 0.0008550178566873411, "data_time": 0.0010882854461669923, "grad_norm": 0.021265176229644567, "loss": 0.6840592145919799, "time": 0.38179304599761965, "epoch": 6, "memory": 9465, "step": 59980} +{"lr": 0.0008550178566873411, "data_time": 0.00113983154296875, "grad_norm": 0.020948754576966166, "loss": 0.6856973528861999, "time": 0.40779705047607423, "epoch": 6, "memory": 9465, "step": 60080} +{"lr": 0.0008550178566873411, "data_time": 0.001367497444152832, "grad_norm": 0.025594734493643044, "loss": 0.6940591335296631, "time": 0.4549035310745239, "epoch": 6, "memory": 9465, "step": 60180} +{"lr": 0.0008550178566873411, "data_time": 0.0010826587677001953, "grad_norm": 0.017471459379885347, "loss": 0.686150062084198, "time": 0.3902132511138916, "epoch": 6, "memory": 9465, "step": 60280} +{"lr": 0.0008550178566873411, "data_time": 0.0012021541595458984, "grad_norm": 0.015292374766431749, "loss": 0.683225828409195, "time": 0.3980573177337646, "epoch": 6, "memory": 9465, "step": 60380} +{"lr": 0.0008550178566873411, "data_time": 0.0011262178421020507, "grad_norm": 0.019223866751417518, "loss": 0.6911708950996399, "time": 0.41641333103179934, "epoch": 6, "memory": 9465, "step": 60480} +{"lr": 0.0008550178566873411, "data_time": 0.0010864019393920898, "grad_norm": 0.026042074616998433, "loss": 0.6903434634208679, "time": 0.43301615715026853, "epoch": 6, "memory": 9465, "step": 60580} +{"lr": 0.0008550178566873411, "data_time": 0.001121211051940918, "grad_norm": 0.020801701271557248, "loss": 0.6834124207496644, "time": 0.40266146659851076, "epoch": 6, "memory": 9465, "step": 60680} +{"lr": 0.0008550178566873411, "data_time": 0.0011501073837280273, "grad_norm": 0.014840489346534013, "loss": 0.6909108400344849, "time": 0.3964644432067871, "epoch": 6, "memory": 9465, "step": 60780} +{"lr": 0.0008550178566873411, "data_time": 0.001073169708251953, "grad_norm": 0.026812200294807555, "loss": 0.6863328039646148, "time": 0.3890755414962769, "epoch": 6, "memory": 9465, "step": 60880} +{"lr": 0.0008550178566873411, "data_time": 0.0010693073272705078, "grad_norm": 0.03237656396813691, "loss": 0.6915999114513397, "time": 0.4082698583602905, "epoch": 6, "memory": 9465, "step": 60980} +{"lr": 0.0008550178566873411, "data_time": 0.0010993480682373047, "grad_norm": 0.03013187083415687, "loss": 0.6842646121978759, "time": 0.42697293758392335, "epoch": 6, "memory": 9465, "step": 61080} +{"lr": 0.0008550178566873411, "data_time": 0.001182413101196289, "grad_norm": 0.028919320087879896, "loss": 0.6859990298748017, "time": 0.4435647249221802, "epoch": 6, "memory": 9465, "step": 61180} +{"lr": 0.0008550178566873411, "data_time": 0.0012660264968872071, "grad_norm": 0.028606146585661917, "loss": 0.6837523519992829, "time": 0.41578099727630613, "epoch": 6, "memory": 9465, "step": 61280} +{"lr": 0.0008550178566873411, "data_time": 0.0010733604431152344, "grad_norm": 0.023944671917706727, "loss": 0.6810744404792786, "time": 0.4051700830459595, "epoch": 6, "memory": 9465, "step": 61380} +{"lr": 0.0008550178566873411, "data_time": 0.0013903379440307617, "grad_norm": 0.025947124883532523, "loss": 0.6896799206733704, "time": 0.44748764038085936, "epoch": 6, "memory": 9465, "step": 61480} +{"lr": 0.0008550178566873411, "data_time": 0.0010825634002685548, "grad_norm": 0.011630263109691442, "loss": 0.6843417644500732, "time": 0.3873655557632446, "epoch": 6, "memory": 9465, "step": 61580} +{"lr": 0.0008550178566873411, "data_time": 0.0010849237442016602, "grad_norm": 0.020101899770088492, "loss": 0.6912844240665436, "time": 0.3698854684829712, "epoch": 6, "memory": 9465, "step": 61680} +{"lr": 0.0008550178566873411, "data_time": 0.0011237144470214843, "grad_norm": 0.016183562227524817, "loss": 0.6849010467529297, "time": 0.4075817108154297, "epoch": 6, "memory": 9465, "step": 61780} +{"lr": 0.0008550178566873411, "data_time": 0.001196885108947754, "grad_norm": 0.02694909244310111, "loss": 0.6850013375282288, "time": 0.4347084522247314, "epoch": 6, "memory": 9465, "step": 61880} +{"lr": 0.0008550178566873411, "data_time": 0.0010751724243164063, "grad_norm": 0.02409428098471835, "loss": 0.6830321192741394, "time": 0.42159261703491213, "epoch": 6, "memory": 9465, "step": 61980} +{"lr": 0.0008550178566873411, "data_time": 0.0011061429977416992, "grad_norm": 0.015985024417750536, "loss": 0.6874011397361756, "time": 0.4641463041305542, "epoch": 6, "memory": 9465, "step": 62080} +{"lr": 0.0008550178566873411, "data_time": 0.0013673067092895507, "grad_norm": 0.02110781449591741, "loss": 0.6865115404129029, "time": 0.38755435943603517, "epoch": 6, "memory": 9465, "step": 62180} +{"lr": 0.0008550178566873411, "data_time": 0.0010976552963256835, "grad_norm": 0.025282340170815588, "loss": 0.6891542911529541, "time": 0.386476469039917, "epoch": 6, "memory": 9465, "step": 62280} +{"lr": 0.0008550178566873411, "data_time": 0.0011896610260009766, "grad_norm": 0.018300859036389738, "loss": 0.6869811832904815, "time": 0.4204688787460327, "epoch": 6, "memory": 9465, "step": 62380} +{"lr": 0.0008550178566873411, "data_time": 0.0012111902236938477, "grad_norm": 0.021349584963172675, "loss": 0.6878549575805664, "time": 0.411251425743103, "epoch": 6, "memory": 9465, "step": 62480} +{"lr": 0.0008550178566873411, "data_time": 0.0013198375701904297, "grad_norm": 0.02706974893808365, "loss": 0.6856260359287262, "time": 0.42128643989562986, "epoch": 6, "memory": 9465, "step": 62580} +{"lr": 0.0008550178566873411, "data_time": 0.0014317035675048828, "grad_norm": 0.030903576174750924, "loss": 0.6902890741825104, "time": 0.41969847679138184, "epoch": 6, "memory": 9465, "step": 62680} +{"lr": 0.0008550178566873411, "data_time": 0.001162409782409668, "grad_norm": 0.017232837807387113, "loss": 0.6841163575649262, "time": 0.39000930786132815, "epoch": 6, "memory": 9465, "step": 62780} +{"lr": 0.0008550178566873411, "data_time": 0.0011013269424438477, "grad_norm": 0.03369742791401222, "loss": 0.6851688861846924, "time": 0.40869951248168945, "epoch": 6, "memory": 9465, "step": 62880} +{"lr": 0.0008550178566873411, "data_time": 0.0011281728744506835, "grad_norm": 0.02228668605675921, "loss": 0.6884321808815003, "time": 0.4005002498626709, "epoch": 6, "memory": 9465, "step": 62980} +{"lr": 0.0008550178566873411, "data_time": 0.001163172721862793, "grad_norm": 0.017968063976150007, "loss": 0.6882640779018402, "time": 0.4137157440185547, "epoch": 6, "memory": 9465, "step": 63080} +{"lr": 0.0008550178566873411, "data_time": 0.0011319398880004882, "grad_norm": 0.01606730920029804, "loss": 0.6900041937828064, "time": 0.39462196826934814, "epoch": 6, "memory": 9465, "step": 63180} +{"lr": 0.0008550178566873411, "data_time": 0.0011784791946411132, "grad_norm": 0.021908795461058617, "loss": 0.6898101925849914, "time": 0.41650776863098143, "epoch": 6, "memory": 9465, "step": 63280} +{"lr": 0.0008550178566873411, "data_time": 0.0012079477310180664, "grad_norm": 0.024879211466759442, "loss": 0.688402795791626, "time": 0.40994791984558104, "epoch": 6, "memory": 9465, "step": 63380} +{"lr": 0.0008550178566873411, "data_time": 0.001107168197631836, "grad_norm": 0.027438949700444935, "loss": 0.6885316550731659, "time": 0.3998621225357056, "epoch": 6, "memory": 9465, "step": 63480} +{"lr": 0.0008550178566873411, "data_time": 0.001163315773010254, "grad_norm": 0.026064815791323782, "loss": 0.6823230147361755, "time": 0.4091726064682007, "epoch": 6, "memory": 9465, "step": 63580} +{"lr": 0.0008550178566873411, "data_time": 0.0011488676071166992, "grad_norm": 0.02314729627687484, "loss": 0.683469420671463, "time": 0.4455825090408325, "epoch": 6, "memory": 9465, "step": 63680} +{"lr": 0.0008550178566873411, "data_time": 0.0012412309646606446, "grad_norm": 0.018279211735352872, "loss": 0.6866757035255432, "time": 0.3936056137084961, "epoch": 6, "memory": 9465, "step": 63780} +{"lr": 0.0008550178566873411, "data_time": 0.0010842084884643555, "grad_norm": 0.025038791983388365, "loss": 0.6824251770973205, "time": 0.48471248149871826, "epoch": 6, "memory": 9465, "step": 63880} +{"lr": 0.0008550178566873411, "data_time": 0.0010651588439941407, "grad_norm": 0.01808282525744289, "loss": 0.6876177608966827, "time": 0.39613330364227295, "epoch": 6, "memory": 9465, "step": 63980} +{"lr": 0.0008550178566873411, "data_time": 0.0010905504226684571, "grad_norm": 0.018851933989208192, "loss": 0.6884236335754395, "time": 0.4246366500854492, "epoch": 6, "memory": 9465, "step": 64080} +{"lr": 0.0008550178566873411, "data_time": 0.0011005163192749023, "grad_norm": 0.0259113168111071, "loss": 0.6852250516414642, "time": 0.3710044860839844, "epoch": 6, "memory": 9465, "step": 64180} +{"lr": 0.0008550178566873411, "data_time": 0.001170516014099121, "grad_norm": 0.029111536033451558, "loss": 0.6851958274841309, "time": 0.3933532476425171, "epoch": 6, "memory": 9465, "step": 64280} +{"lr": 0.0008550178566873411, "data_time": 0.0011178970336914063, "grad_norm": 0.015426943911006675, "loss": 0.6896191537380219, "time": 0.3818197250366211, "epoch": 6, "memory": 9465, "step": 64380} +{"lr": 0.0008550178566873411, "data_time": 0.0010658979415893554, "grad_norm": 0.023294601123780012, "loss": 0.6892889618873597, "time": 0.4220273494720459, "epoch": 6, "memory": 9465, "step": 64480} +{"lr": 0.0008550178566873411, "data_time": 0.0011204719543457032, "grad_norm": 0.011773345549590886, "loss": 0.6862027585506439, "time": 0.43556506633758546, "epoch": 6, "memory": 9465, "step": 64580} +{"lr": 0.0008550178566873411, "data_time": 0.001467728614807129, "grad_norm": 0.02419381303479895, "loss": 0.6829163193702698, "time": 0.40226283073425295, "epoch": 6, "memory": 9465, "step": 64680} +{"lr": 0.0008550178566873411, "data_time": 0.0012456655502319336, "grad_norm": 0.03351171747781336, "loss": 0.6883672893047332, "time": 0.43147013187408445, "epoch": 6, "memory": 9465, "step": 64780} +{"lr": 0.0008550178566873411, "data_time": 0.0011522769927978516, "grad_norm": 0.025805607502115892, "loss": 0.6861656308174133, "time": 0.4185208320617676, "epoch": 6, "memory": 9465, "step": 64880} +{"lr": 0.0008550178566873411, "data_time": 0.001204204559326172, "grad_norm": 0.02143499543890357, "loss": 0.687349659204483, "time": 0.39340035915374755, "epoch": 6, "memory": 9465, "step": 64980} +{"lr": 0.0008550178566873411, "data_time": 0.0011176347732543945, "grad_norm": 0.021567908296128736, "loss": 0.6841687262058258, "time": 0.40543975830078127, "epoch": 6, "memory": 9465, "step": 65080} +{"lr": 0.0008550178566873411, "data_time": 0.0011295318603515626, "grad_norm": 0.025428109848871828, "loss": 0.6899242162704468, "time": 0.387009334564209, "epoch": 6, "memory": 9465, "step": 65180} +{"lr": 0.0008550178566873411, "data_time": 0.0011513233184814453, "grad_norm": 0.020488849189132453, "loss": 0.6837246060371399, "time": 0.4116119623184204, "epoch": 6, "memory": 9465, "step": 65280} +{"lr": 0.0008550178566873411, "data_time": 0.0011474609375, "grad_norm": 0.02112674816744402, "loss": 0.6861176073551178, "time": 0.46766388416290283, "epoch": 6, "memory": 9465, "step": 65380} +{"lr": 0.0008550178566873411, "data_time": 0.0012309551239013672, "grad_norm": 0.01799351954832673, "loss": 0.6893830060958862, "time": 0.4063719749450684, "epoch": 6, "memory": 9465, "step": 65480} +{"lr": 0.0008550178566873411, "data_time": 0.001080632209777832, "grad_norm": 0.027082108333706854, "loss": 0.688882440328598, "time": 0.3806299209594727, "epoch": 6, "memory": 9465, "step": 65580} +{"lr": 0.0008550178566873411, "data_time": 0.0011981010437011718, "grad_norm": 0.0213951176032424, "loss": 0.6909701764583588, "time": 0.38790957927703856, "epoch": 6, "memory": 9465, "step": 65680} +{"lr": 0.0008550178566873411, "data_time": 0.0011966705322265625, "grad_norm": 0.016349894972518087, "loss": 0.6805295407772064, "time": 0.3795259952545166, "epoch": 6, "memory": 9465, "step": 65780} +{"lr": 0.0008550178566873411, "data_time": 0.0010399103164672851, "grad_norm": 0.0243533281609416, "loss": 0.6870722115039826, "time": 0.3947573184967041, "epoch": 6, "memory": 9465, "step": 65880} +{"lr": 0.0008550178566873411, "data_time": 0.0011670827865600587, "grad_norm": 0.017661346250679343, "loss": 0.687247508764267, "time": 0.4532417058944702, "epoch": 6, "memory": 9465, "step": 65980} +{"lr": 0.0008550178566873411, "data_time": 0.001065516471862793, "grad_norm": 0.019467962806811557, "loss": 0.6906374752521515, "time": 0.4153579235076904, "epoch": 6, "memory": 9465, "step": 66080} +{"lr": 0.0008550178566873411, "data_time": 0.0011503219604492188, "grad_norm": 0.03497491907328367, "loss": 0.6900132894515991, "time": 0.40140585899353026, "epoch": 6, "memory": 9465, "step": 66180} +{"lr": 0.0008550178566873411, "data_time": 0.0011035442352294923, "grad_norm": 0.02296763132326305, "loss": 0.686825406551361, "time": 0.40420947074890134, "epoch": 6, "memory": 9465, "step": 66280} +{"lr": 0.0008550178566873411, "data_time": 0.0011605024337768555, "grad_norm": 0.025388493668287992, "loss": 0.6918170928955079, "time": 0.4069295644760132, "epoch": 6, "memory": 9465, "step": 66380} +{"lr": 0.0008550178566873411, "data_time": 0.0012253761291503907, "grad_norm": 0.025179001805372535, "loss": 0.6868872821331025, "time": 0.4094263792037964, "epoch": 6, "memory": 9465, "step": 66480} +{"lr": 0.0008550178566873411, "data_time": 0.0013419866561889648, "grad_norm": 0.027657877001911402, "loss": 0.6838114738464356, "time": 0.4042673587799072, "epoch": 6, "memory": 9465, "step": 66580} +{"lr": 0.0008550178566873411, "data_time": 0.001176595687866211, "grad_norm": 0.029566391464322807, "loss": 0.6818347692489624, "time": 0.409966254234314, "epoch": 6, "memory": 9465, "step": 66680} +{"lr": 0.0008550178566873411, "data_time": 0.0012098073959350586, "grad_norm": 0.01688590687699616, "loss": 0.6903729736804962, "time": 0.4109964847564697, "epoch": 6, "memory": 9465, "step": 66780} +{"lr": 0.0008550178566873411, "data_time": 0.001275348663330078, "grad_norm": 0.023581396555528045, "loss": 0.6841236174106597, "time": 0.3953559160232544, "epoch": 6, "memory": 9465, "step": 66880} +{"lr": 0.0008550178566873411, "data_time": 0.0011750221252441405, "grad_norm": 0.019107314106076956, "loss": 0.6900860786437988, "time": 0.4060039043426514, "epoch": 6, "memory": 9465, "step": 66980} +{"lr": 0.0008550178566873411, "data_time": 0.0012302875518798828, "grad_norm": 0.02024886240833439, "loss": 0.686821186542511, "time": 0.42997496128082274, "epoch": 6, "memory": 9465, "step": 67080} +{"lr": 0.0008550178566873411, "data_time": 0.0011691808700561523, "grad_norm": 0.023429017094895244, "loss": 0.680237877368927, "time": 0.38594486713409426, "epoch": 6, "memory": 9465, "step": 67180} +{"lr": 0.0008550178566873411, "data_time": 0.001174330711364746, "grad_norm": 0.021199182444252073, "loss": 0.6820933282375335, "time": 0.40532851219177246, "epoch": 6, "memory": 9465, "step": 67280} +{"lr": 0.0008550178566873411, "data_time": 0.0011693239212036133, "grad_norm": 0.01970489202067256, "loss": 0.6860453009605407, "time": 0.39337117671966554, "epoch": 6, "memory": 9465, "step": 67380} +{"lr": 0.0008550178566873411, "data_time": 0.0011412858963012694, "grad_norm": 0.024003434274345637, "loss": 0.6859983444213867, "time": 0.4117098569869995, "epoch": 6, "memory": 9465, "step": 67480} +{"lr": 0.0008550178566873411, "data_time": 0.0011181592941284179, "grad_norm": 0.023815575637854636, "loss": 0.6898360013961792, "time": 0.39422993659973143, "epoch": 6, "memory": 9465, "step": 67580} +{"lr": 0.0008550178566873411, "data_time": 0.0012259244918823241, "grad_norm": 0.01485637214500457, "loss": 0.6895285248756409, "time": 0.43624048233032225, "epoch": 6, "memory": 9465, "step": 67680} +{"lr": 0.0008550178566873411, "data_time": 0.0012720584869384765, "grad_norm": 0.022814353555440904, "loss": 0.691504716873169, "time": 0.4203951835632324, "epoch": 6, "memory": 9465, "step": 67780} +{"lr": 0.0008550178566873411, "data_time": 0.0012421369552612304, "grad_norm": 0.016687876591458917, "loss": 0.6868099987506866, "time": 0.3962958335876465, "epoch": 6, "memory": 9465, "step": 67880} +{"lr": 0.0008550178566873411, "data_time": 0.0012056827545166016, "grad_norm": 0.015783934318460525, "loss": 0.6871471047401428, "time": 0.40698909759521484, "epoch": 6, "memory": 9465, "step": 67980} +{"lr": 0.0008550178566873411, "data_time": 0.001257467269897461, "grad_norm": 0.017015948914922772, "loss": 0.6893740594387054, "time": 0.4241846799850464, "epoch": 6, "memory": 9465, "step": 68080} +{"lr": 0.0008550178566873411, "data_time": 0.0012694597244262695, "grad_norm": 0.02218850781209767, "loss": 0.6860420525074005, "time": 0.4323299407958984, "epoch": 6, "memory": 9465, "step": 68180} +{"lr": 0.0008550178566873411, "data_time": 0.0012068748474121094, "grad_norm": 0.015337636042386294, "loss": 0.6899284422397614, "time": 0.38808631896972656, "epoch": 6, "memory": 9465, "step": 68280} +{"lr": 0.0008550178566873411, "data_time": 0.0012377023696899414, "grad_norm": 0.03523201327770949, "loss": 0.6839330554008484, "time": 0.41525607109069823, "epoch": 6, "memory": 9465, "step": 68380} +{"lr": 0.0008550178566873411, "data_time": 0.0011533737182617188, "grad_norm": 0.02610473702661693, "loss": 0.6918040812015533, "time": 0.37747156620025635, "epoch": 6, "memory": 9465, "step": 68480} +{"lr": 0.0008550178566873411, "data_time": 0.001188945770263672, "grad_norm": 0.018019030266441406, "loss": 0.6894958972930908, "time": 0.41466484069824217, "epoch": 6, "memory": 9465, "step": 68580} +{"lr": 0.0008550178566873411, "data_time": 0.0015630245208740235, "grad_norm": 0.01750457133166492, "loss": 0.6869166016578674, "time": 0.4141963720321655, "epoch": 6, "memory": 9465, "step": 68680} +{"lr": 0.0008550178566873411, "data_time": 0.0012135982513427734, "grad_norm": 0.028229457233101128, "loss": 0.6859838128089905, "time": 0.3897592782974243, "epoch": 6, "memory": 9465, "step": 68780} +{"lr": 0.0008550178566873411, "data_time": 0.0012891054153442382, "grad_norm": 0.033662870479747656, "loss": 0.6845110654830933, "time": 0.3601423740386963, "epoch": 6, "memory": 9465, "step": 68880} +{"lr": 0.0008550178566873411, "data_time": 0.0012295961380004883, "grad_norm": 0.019249517982825638, "loss": 0.6847482442855835, "time": 0.43565006256103517, "epoch": 6, "memory": 9465, "step": 68980} +{"lr": 0.0008550178566873411, "data_time": 0.007334566116333008, "grad_norm": 0.0190709559712559, "loss": 0.6824127018451691, "time": 0.3995782136917114, "epoch": 6, "memory": 9465, "step": 69080} +{"accuracy/top1": 100.0, "data_time": 0.3054828924291274, "time": 0.5640647832085105, "step": 6} +{"lr": 0.0007959536998847743, "data_time": 0.022605299949645996, "grad_norm": 0.01316798001062125, "loss": 0.6878198325634003, "time": 0.44731831550598145, "epoch": 7, "memory": 9465, "step": 69196} +{"lr": 0.0007959536998847743, "data_time": 0.0010965585708618164, "grad_norm": 0.028479007072746754, "loss": 0.6900545597076416, "time": 0.44763081073760985, "epoch": 7, "memory": 9465, "step": 69296} +{"lr": 0.0007959536998847743, "data_time": 0.00127718448638916, "grad_norm": 0.02315240743337199, "loss": 0.6806198239326477, "time": 0.4228838920593262, "epoch": 7, "memory": 9465, "step": 69396} +{"lr": 0.0007959536998847743, "data_time": 0.0015025854110717774, "grad_norm": 0.026145265670493245, "loss": 0.6891481876373291, "time": 0.4264572858810425, "epoch": 7, "memory": 9465, "step": 69496} +{"lr": 0.0007959536998847743, "data_time": 0.0010885953903198241, "grad_norm": 0.020853764610365033, "loss": 0.6852071940898895, "time": 0.35759451389312746, "epoch": 7, "memory": 9465, "step": 69596} +{"lr": 0.0007959536998847743, "data_time": 0.0011881113052368164, "grad_norm": 0.02196980845183134, "loss": 0.6836467266082764, "time": 0.43849189281463624, "epoch": 7, "memory": 9465, "step": 69696} +{"lr": 0.0007959536998847743, "data_time": 0.0010839462280273437, "grad_norm": 0.022946777241304516, "loss": 0.6840446650981903, "time": 0.4129193782806396, "epoch": 7, "memory": 9465, "step": 69796} +{"lr": 0.0007959536998847743, "data_time": 0.0012560606002807618, "grad_norm": 0.013690093078184872, "loss": 0.6849134087562561, "time": 0.4388380765914917, "epoch": 7, "memory": 9465, "step": 69896} +{"lr": 0.0007959536998847743, "data_time": 0.0011666059494018556, "grad_norm": 0.014410020341165364, "loss": 0.6856664478778839, "time": 0.3797380208969116, "epoch": 7, "memory": 9465, "step": 69996} +{"lr": 0.0007959536998847743, "data_time": 0.002634882926940918, "grad_norm": 0.023347340803593398, "loss": 0.6844936788082123, "time": 0.4347724914550781, "epoch": 7, "memory": 9465, "step": 70096} +{"lr": 0.0007959536998847743, "data_time": 0.0010866880416870116, "grad_norm": 0.02320546139962971, "loss": 0.6840683877468109, "time": 0.3985123872756958, "epoch": 7, "memory": 9465, "step": 70196} +{"lr": 0.0007959536998847743, "data_time": 0.0012055397033691405, "grad_norm": 0.01761086212936789, "loss": 0.6879605293273926, "time": 0.4709017276763916, "epoch": 7, "memory": 9465, "step": 70296} +{"lr": 0.0007959536998847743, "data_time": 0.0012353897094726563, "grad_norm": 0.02368117372971028, "loss": 0.6894306123256684, "time": 0.39980330467224123, "epoch": 7, "memory": 9465, "step": 70396} +{"lr": 0.0007959536998847743, "data_time": 0.0011112451553344726, "grad_norm": 0.026349100284278393, "loss": 0.6901382505893707, "time": 0.4116779327392578, "epoch": 7, "memory": 9465, "step": 70496} +{"lr": 0.0007959536998847743, "data_time": 0.001292872428894043, "grad_norm": 0.02749940538778901, "loss": 0.6840810477733612, "time": 0.3990472316741943, "epoch": 7, "memory": 9465, "step": 70596} +{"lr": 0.0007959536998847743, "data_time": 0.0010713100433349609, "grad_norm": 0.031892047519795595, "loss": 0.6849338054656983, "time": 0.40349245071411133, "epoch": 7, "memory": 9465, "step": 70696} +{"lr": 0.0007959536998847743, "data_time": 0.0027619123458862303, "grad_norm": 0.02551441374234855, "loss": 0.6865213751792908, "time": 0.4154572725296021, "epoch": 7, "memory": 9465, "step": 70796} +{"lr": 0.0007959536998847743, "data_time": 0.0011179447174072266, "grad_norm": 0.014401370845735072, "loss": 0.6865893602371216, "time": 0.38748998641967775, "epoch": 7, "memory": 9465, "step": 70896} +{"lr": 0.0007959536998847743, "data_time": 0.0013726711273193359, "grad_norm": 0.03455458893440664, "loss": 0.6865870893001557, "time": 0.4037840127944946, "epoch": 7, "memory": 9465, "step": 70996} +{"lr": 0.0007959536998847743, "data_time": 0.0011995077133178712, "grad_norm": 0.018426940822973848, "loss": 0.6870003044605255, "time": 0.3991276502609253, "epoch": 7, "memory": 9465, "step": 71096} +{"lr": 0.0007959536998847743, "data_time": 0.0011152982711791991, "grad_norm": 0.032567341532558206, "loss": 0.6811158239841462, "time": 0.39254446029663087, "epoch": 7, "memory": 9465, "step": 71196} +{"lr": 0.0007959536998847743, "data_time": 0.0010578393936157226, "grad_norm": 0.024077949032653125, "loss": 0.686410653591156, "time": 0.3703174352645874, "epoch": 7, "memory": 9465, "step": 71296} +{"lr": 0.0007959536998847743, "data_time": 0.0011644363403320312, "grad_norm": 0.028478661738336086, "loss": 0.6945944964885712, "time": 0.39158904552459717, "epoch": 7, "memory": 9465, "step": 71396} +{"lr": 0.0007959536998847743, "data_time": 0.0011600732803344726, "grad_norm": 0.018969375151209535, "loss": 0.689129626750946, "time": 0.4178628921508789, "epoch": 7, "memory": 9465, "step": 71496} +{"lr": 0.0007959536998847743, "data_time": 0.0010562658309936524, "grad_norm": 0.030955977737903595, "loss": 0.683920043706894, "time": 0.411151123046875, "epoch": 7, "memory": 9465, "step": 71596} +{"lr": 0.0007959536998847743, "data_time": 0.0012195110321044922, "grad_norm": 0.021646203519776464, "loss": 0.692710280418396, "time": 0.4151927947998047, "epoch": 7, "memory": 9465, "step": 71696} +{"lr": 0.0007959536998847743, "data_time": 0.0012052059173583984, "grad_norm": 0.028709866828285156, "loss": 0.6846365094184875, "time": 0.36250553131103513, "epoch": 7, "memory": 9465, "step": 71796} +{"lr": 0.0007959536998847743, "data_time": 0.0012458086013793946, "grad_norm": 0.02136602159589529, "loss": 0.6866275489330291, "time": 0.3874361038208008, "epoch": 7, "memory": 9465, "step": 71896} +{"lr": 0.0007959536998847743, "data_time": 0.0013145685195922851, "grad_norm": 0.02129147609230131, "loss": 0.6894185721874238, "time": 0.407142186164856, "epoch": 7, "memory": 9465, "step": 71996} +{"lr": 0.0007959536998847743, "data_time": 0.0011846303939819336, "grad_norm": 0.028740060911513866, "loss": 0.6892673373222351, "time": 0.4421504020690918, "epoch": 7, "memory": 9465, "step": 72096} +{"lr": 0.0007959536998847743, "data_time": 0.0011426210403442383, "grad_norm": 0.02137885382398963, "loss": 0.6883216917514801, "time": 0.37873086929321287, "epoch": 7, "memory": 9465, "step": 72196} +{"lr": 0.0007959536998847743, "data_time": 0.001166391372680664, "grad_norm": 0.0248980559874326, "loss": 0.6873321831226349, "time": 0.4093770027160645, "epoch": 7, "memory": 9465, "step": 72296} +{"lr": 0.0007959536998847743, "data_time": 0.0011440277099609374, "grad_norm": 0.016781652311328797, "loss": 0.689619517326355, "time": 0.4165563821792603, "epoch": 7, "memory": 9465, "step": 72396} +{"lr": 0.0007959536998847743, "data_time": 0.0011880636215209962, "grad_norm": 0.01665127892047167, "loss": 0.6863309919834137, "time": 0.3981989860534668, "epoch": 7, "memory": 9465, "step": 72496} +{"lr": 0.0007959536998847743, "data_time": 0.001066732406616211, "grad_norm": 0.025460918783210218, "loss": 0.6827969551086426, "time": 0.3673178911209106, "epoch": 7, "memory": 9465, "step": 72596} +{"lr": 0.0007959536998847743, "data_time": 0.0010986804962158203, "grad_norm": 0.02312437416985631, "loss": 0.6871290862560272, "time": 0.39009847640991213, "epoch": 7, "memory": 9465, "step": 72696} +{"lr": 0.0007959536998847743, "data_time": 0.0011986494064331055, "grad_norm": 0.019325556652620434, "loss": 0.684772652387619, "time": 0.38254680633544924, "epoch": 7, "memory": 9465, "step": 72796} +{"lr": 0.0007959536998847743, "data_time": 0.0011121511459350586, "grad_norm": 0.02225032856222242, "loss": 0.6838533043861389, "time": 0.4375830888748169, "epoch": 7, "memory": 9465, "step": 72896} +{"lr": 0.0007959536998847743, "data_time": 0.0010936498641967774, "grad_norm": 0.024222818808630107, "loss": 0.6877457797527313, "time": 0.4053443670272827, "epoch": 7, "memory": 9465, "step": 72996} +{"lr": 0.0007959536998847743, "data_time": 0.0012072324752807617, "grad_norm": 0.02482799154240638, "loss": 0.6861601173877716, "time": 0.3897785902023315, "epoch": 7, "memory": 9465, "step": 73096} +{"lr": 0.0007959536998847743, "data_time": 0.00133209228515625, "grad_norm": 0.028896976122632624, "loss": 0.6917809903621673, "time": 0.3895012617111206, "epoch": 7, "memory": 9465, "step": 73196} +{"lr": 0.0007959536998847743, "data_time": 0.0013009071350097655, "grad_norm": 0.017806610558182, "loss": 0.6895358264446259, "time": 0.39615821838378906, "epoch": 7, "memory": 9465, "step": 73296} +{"lr": 0.0007959536998847743, "data_time": 0.001237964630126953, "grad_norm": 0.01780034021940082, "loss": 0.6869791269302368, "time": 0.3995818138122559, "epoch": 7, "memory": 9465, "step": 73396} +{"lr": 0.0007959536998847743, "data_time": 0.0011965274810791016, "grad_norm": 0.012079532991629094, "loss": 0.6852951526641846, "time": 0.4225271463394165, "epoch": 7, "memory": 9465, "step": 73496} +{"lr": 0.0007959536998847743, "data_time": 0.0011064529418945313, "grad_norm": 0.026877340488135816, "loss": 0.6846539080142975, "time": 0.3968620300292969, "epoch": 7, "memory": 9465, "step": 73596} +{"lr": 0.0007959536998847743, "data_time": 0.0011870861053466797, "grad_norm": 0.03740522600710392, "loss": 0.689351600408554, "time": 0.39341351985931394, "epoch": 7, "memory": 9465, "step": 73696} +{"lr": 0.0007959536998847743, "data_time": 0.0011953353881835938, "grad_norm": 0.013193433929700404, "loss": 0.6871483862400055, "time": 0.4323933839797974, "epoch": 7, "memory": 9465, "step": 73796} +{"lr": 0.0007959536998847743, "data_time": 0.0011172771453857421, "grad_norm": 0.009352261293679475, "loss": 0.6889617741107941, "time": 0.42199344635009767, "epoch": 7, "memory": 9465, "step": 73896} +{"lr": 0.0007959536998847743, "data_time": 0.001228928565979004, "grad_norm": 0.020918296393938364, "loss": 0.6816730439662934, "time": 0.3974435567855835, "epoch": 7, "memory": 9465, "step": 73996} +{"lr": 0.0007959536998847743, "data_time": 0.0012335062026977539, "grad_norm": 0.017875133221969007, "loss": 0.6810477554798127, "time": 0.39940171241760253, "epoch": 7, "memory": 9465, "step": 74096} +{"lr": 0.0007959536998847743, "data_time": 0.0011986255645751952, "grad_norm": 0.014076375891454518, "loss": 0.6842746794223785, "time": 0.4127380132675171, "epoch": 7, "memory": 9465, "step": 74196} +{"lr": 0.0007959536998847743, "data_time": 0.0012447357177734375, "grad_norm": 0.02469316883943975, "loss": 0.6853709757328034, "time": 0.40438406467437743, "epoch": 7, "memory": 9465, "step": 74296} +{"lr": 0.0007959536998847743, "data_time": 0.0011493206024169923, "grad_norm": 0.021600378945004196, "loss": 0.6904186487197876, "time": 0.3997747182846069, "epoch": 7, "memory": 9465, "step": 74396} +{"lr": 0.0007959536998847743, "data_time": 0.001335000991821289, "grad_norm": 0.022198598738759755, "loss": 0.6854990541934967, "time": 0.3842605113983154, "epoch": 7, "memory": 9465, "step": 74496} +{"lr": 0.0007959536998847743, "data_time": 0.0011192798614501954, "grad_norm": 0.03163915043696761, "loss": 0.6925682902336121, "time": 0.45507111549377444, "epoch": 7, "memory": 9465, "step": 74596} +{"lr": 0.0007959536998847743, "data_time": 0.001257038116455078, "grad_norm": 0.020809374330565334, "loss": 0.6893840789794922, "time": 0.4486205816268921, "epoch": 7, "memory": 9465, "step": 74696} +{"lr": 0.0007959536998847743, "data_time": 0.0011239767074584961, "grad_norm": 0.025458967802114783, "loss": 0.686081075668335, "time": 0.4434143781661987, "epoch": 7, "memory": 9465, "step": 74796} +{"lr": 0.0007959536998847743, "data_time": 0.0011315107345581054, "grad_norm": 0.027446240209974347, "loss": 0.691266405582428, "time": 0.40890026092529297, "epoch": 7, "memory": 9465, "step": 74896} +{"lr": 0.0007959536998847743, "data_time": 0.0011792421340942384, "grad_norm": 0.01645224934909493, "loss": 0.6849417626857758, "time": 0.4130192518234253, "epoch": 7, "memory": 9465, "step": 74996} +{"lr": 0.0007959536998847743, "data_time": 0.001281905174255371, "grad_norm": 0.016274695470929145, "loss": 0.6884645402431488, "time": 0.3943858861923218, "epoch": 7, "memory": 9465, "step": 75096} +{"lr": 0.0007959536998847743, "data_time": 0.001122283935546875, "grad_norm": 0.0183821105863899, "loss": 0.6827846050262452, "time": 0.38330254554748533, "epoch": 7, "memory": 9465, "step": 75196} +{"lr": 0.0007959536998847743, "data_time": 0.0011628389358520508, "grad_norm": 0.02766981702297926, "loss": 0.6827305257320404, "time": 0.41019539833068847, "epoch": 7, "memory": 9465, "step": 75296} +{"lr": 0.0007959536998847743, "data_time": 0.0011681079864501952, "grad_norm": 0.028430684260092674, "loss": 0.6861222445964813, "time": 0.37620279788970945, "epoch": 7, "memory": 9465, "step": 75396} +{"lr": 0.0007959536998847743, "data_time": 0.0012502431869506835, "grad_norm": 0.026339181466028094, "loss": 0.6839255213737487, "time": 0.39713211059570314, "epoch": 7, "memory": 9465, "step": 75496} +{"lr": 0.0007959536998847743, "data_time": 0.0013794422149658204, "grad_norm": 0.029753775848075748, "loss": 0.6906422078609467, "time": 0.3829894304275513, "epoch": 7, "memory": 9465, "step": 75596} +{"lr": 0.0007959536998847743, "data_time": 0.0012267827987670898, "grad_norm": 0.017975526629015803, "loss": 0.6806785941123963, "time": 0.3759139537811279, "epoch": 7, "memory": 9465, "step": 75696} +{"lr": 0.0007959536998847743, "data_time": 0.0010826826095581056, "grad_norm": 0.024884248990565537, "loss": 0.6849643230438233, "time": 0.4704716682434082, "epoch": 7, "memory": 9465, "step": 75796} +{"lr": 0.0007959536998847743, "data_time": 0.0012604951858520507, "grad_norm": 0.03132741232402623, "loss": 0.683762663602829, "time": 0.43655588626861574, "epoch": 7, "memory": 9465, "step": 75896} +{"lr": 0.0007959536998847743, "data_time": 0.001195073127746582, "grad_norm": 0.024875835841521622, "loss": 0.6888093113899231, "time": 0.37883970737457273, "epoch": 7, "memory": 9465, "step": 75996} +{"lr": 0.0007959536998847743, "data_time": 0.0012192487716674804, "grad_norm": 0.018560318695381283, "loss": 0.6864881336688995, "time": 0.38946144580841063, "epoch": 7, "memory": 9465, "step": 76096} +{"lr": 0.0007959536998847743, "data_time": 0.001201462745666504, "grad_norm": 0.013664885348407551, "loss": 0.6840902686119079, "time": 0.37909176349639895, "epoch": 7, "memory": 9465, "step": 76196} +{"lr": 0.0007959536998847743, "data_time": 0.0011518239974975587, "grad_norm": 0.030464647081680597, "loss": 0.6863432645797729, "time": 0.37787656784057616, "epoch": 7, "memory": 9465, "step": 76296} +{"lr": 0.0007959536998847743, "data_time": 0.0012253522872924805, "grad_norm": 0.023973657190799712, "loss": 0.6926301717758179, "time": 0.4389045238494873, "epoch": 7, "memory": 9465, "step": 76396} +{"lr": 0.0007959536998847743, "data_time": 0.001152777671813965, "grad_norm": 0.02176537667401135, "loss": 0.6871478378772735, "time": 0.38025965690612795, "epoch": 7, "memory": 9465, "step": 76496} +{"lr": 0.0007959536998847743, "data_time": 0.0011547327041625977, "grad_norm": 0.029439127212390302, "loss": 0.6821521103382111, "time": 0.3712587833404541, "epoch": 7, "memory": 9465, "step": 76596} +{"lr": 0.0007959536998847743, "data_time": 0.0011970996856689453, "grad_norm": 0.018512492533773184, "loss": 0.6826198101043701, "time": 0.41774275302886965, "epoch": 7, "memory": 9465, "step": 76696} +{"lr": 0.0007959536998847743, "data_time": 0.0011801958084106446, "grad_norm": 0.02286639059893787, "loss": 0.6876256883144378, "time": 0.4106889009475708, "epoch": 7, "memory": 9465, "step": 76796} +{"lr": 0.0007959536998847743, "data_time": 0.0011584043502807617, "grad_norm": 0.023090194049291312, "loss": 0.6913845479488373, "time": 0.40238194465637206, "epoch": 7, "memory": 9465, "step": 76896} +{"lr": 0.0007959536998847743, "data_time": 0.0014173030853271485, "grad_norm": 0.03176913349889219, "loss": 0.6898376643657684, "time": 0.41205744743347167, "epoch": 7, "memory": 9465, "step": 76996} +{"lr": 0.0007959536998847743, "data_time": 0.0011304140090942383, "grad_norm": 0.01942989700473845, "loss": 0.6842361092567444, "time": 0.4420263051986694, "epoch": 7, "memory": 9465, "step": 77096} +{"lr": 0.0007959536998847743, "data_time": 0.0011557340621948242, "grad_norm": 0.02109438246116042, "loss": 0.6830998361110687, "time": 0.4409440279006958, "epoch": 7, "memory": 9465, "step": 77196} +{"lr": 0.0007959536998847743, "data_time": 0.0016119241714477538, "grad_norm": 0.02279579215683043, "loss": 0.6854933440685272, "time": 0.3879295825958252, "epoch": 7, "memory": 9465, "step": 77296} +{"lr": 0.0007959536998847743, "data_time": 0.0011524915695190429, "grad_norm": 0.01711699743755162, "loss": 0.6846040070056916, "time": 0.42294399738311766, "epoch": 7, "memory": 9465, "step": 77396} +{"lr": 0.0007959536998847743, "data_time": 0.0011591196060180664, "grad_norm": 0.03519861763343215, "loss": 0.6838362872600555, "time": 0.394149112701416, "epoch": 7, "memory": 9465, "step": 77496} +{"lr": 0.0007959536998847743, "data_time": 0.0013530254364013672, "grad_norm": 0.023785974842030554, "loss": 0.6800795793533325, "time": 0.4433858633041382, "epoch": 7, "memory": 9465, "step": 77596} +{"lr": 0.0007959536998847743, "data_time": 0.0012445449829101562, "grad_norm": 0.026843556854873895, "loss": 0.6902290403842926, "time": 0.40749545097351075, "epoch": 7, "memory": 9465, "step": 77696} +{"lr": 0.0007959536998847743, "data_time": 0.0012624502182006837, "grad_norm": 0.02514321736525744, "loss": 0.6854418158531189, "time": 0.4115041971206665, "epoch": 7, "memory": 9465, "step": 77796} +{"lr": 0.0007959536998847743, "data_time": 0.0012320995330810547, "grad_norm": 0.03037467645481229, "loss": 0.6883927881717682, "time": 0.4010903358459473, "epoch": 7, "memory": 9465, "step": 77896} +{"lr": 0.0007959536998847743, "data_time": 0.0012339115142822265, "grad_norm": 0.021365776285529138, "loss": 0.6868184626102447, "time": 0.4028171062469482, "epoch": 7, "memory": 9465, "step": 77996} +{"lr": 0.0007959536998847743, "data_time": 0.0013242483139038086, "grad_norm": 0.025139387452509255, "loss": 0.6887413263320923, "time": 0.423887825012207, "epoch": 7, "memory": 9465, "step": 78096} +{"lr": 0.0007959536998847743, "data_time": 0.0012716293334960938, "grad_norm": 0.020559466490522028, "loss": 0.6861476004123688, "time": 0.40378665924072266, "epoch": 7, "memory": 9465, "step": 78196} +{"lr": 0.0007959536998847743, "data_time": 0.0012105226516723633, "grad_norm": 0.021940127667039632, "loss": 0.6868208050727844, "time": 0.40303223133087157, "epoch": 7, "memory": 9465, "step": 78296} +{"lr": 0.0007959536998847743, "data_time": 0.0012447834014892578, "grad_norm": 0.027616849076002835, "loss": 0.6870582818984985, "time": 0.39228756427764894, "epoch": 7, "memory": 9465, "step": 78396} +{"lr": 0.0007959536998847743, "data_time": 0.0013557195663452149, "grad_norm": 0.02429812690243125, "loss": 0.6896453738212586, "time": 0.3997992753982544, "epoch": 7, "memory": 9465, "step": 78496} +{"lr": 0.0007959536998847743, "data_time": 0.0011014223098754882, "grad_norm": 0.02309339742641896, "loss": 0.6866641640663147, "time": 0.4035929203033447, "epoch": 7, "memory": 9465, "step": 78596} +{"lr": 0.0007959536998847743, "data_time": 0.0012946605682373046, "grad_norm": 0.03126331111416221, "loss": 0.6872867465019226, "time": 0.4252474308013916, "epoch": 7, "memory": 9465, "step": 78696} +{"lr": 0.0007959536998847743, "data_time": 0.0011641740798950194, "grad_norm": 0.02848511151969433, "loss": 0.6913005232810974, "time": 0.38922045230865476, "epoch": 7, "memory": 9465, "step": 78796} +{"lr": 0.0007959536998847743, "data_time": 0.0013628482818603515, "grad_norm": 0.020038897823542356, "loss": 0.6844573140144348, "time": 0.40750024318695066, "epoch": 7, "memory": 9465, "step": 78896} +{"lr": 0.0007959536998847743, "data_time": 0.0012709617614746094, "grad_norm": 0.02573901703581214, "loss": 0.6842101216316223, "time": 0.3919532775878906, "epoch": 7, "memory": 9465, "step": 78996} +{"lr": 0.0007959536998847743, "data_time": 0.0013043403625488282, "grad_norm": 0.02561226738616824, "loss": 0.6893555283546448, "time": 0.3859339714050293, "epoch": 7, "memory": 9465, "step": 79096} +{"lr": 0.0007959536998847743, "data_time": 0.0011783361434936524, "grad_norm": 0.02280616246862337, "loss": 0.6859048366546631, "time": 0.4402990102767944, "epoch": 7, "memory": 9465, "step": 79196} +{"lr": 0.0007959536998847743, "data_time": 0.0012436389923095703, "grad_norm": 0.02633576896041632, "loss": 0.6829159498214722, "time": 0.41145374774932864, "epoch": 7, "memory": 9465, "step": 79296} +{"lr": 0.0007959536998847743, "data_time": 0.0014607906341552734, "grad_norm": 0.018908580765128136, "loss": 0.6861242353916168, "time": 0.40912585258483886, "epoch": 7, "memory": 9465, "step": 79396} +{"lr": 0.0007959536998847743, "data_time": 0.001175689697265625, "grad_norm": 0.032280557323247194, "loss": 0.6906120359897614, "time": 0.40739758014678956, "epoch": 7, "memory": 9465, "step": 79496} +{"lr": 0.0007959536998847743, "data_time": 0.0014308691024780273, "grad_norm": 0.02054695161059499, "loss": 0.6850894331932068, "time": 0.41107728481292727, "epoch": 7, "memory": 9465, "step": 79596} +{"lr": 0.0007959536998847743, "data_time": 0.0011907339096069336, "grad_norm": 0.025354425748810173, "loss": 0.6858440995216369, "time": 0.425725531578064, "epoch": 7, "memory": 9465, "step": 79696} +{"lr": 0.0007959536998847743, "data_time": 0.0011603355407714844, "grad_norm": 0.025696104858070612, "loss": 0.688878870010376, "time": 0.3961838722229004, "epoch": 7, "memory": 9465, "step": 79796} +{"lr": 0.0007959536998847743, "data_time": 0.0012267827987670898, "grad_norm": 0.027191118663176896, "loss": 0.6869683980941772, "time": 0.4198110103607178, "epoch": 7, "memory": 9465, "step": 79896} +{"lr": 0.0007959536998847743, "data_time": 0.0011564254760742187, "grad_norm": 0.0296568168210797, "loss": 0.6843812048435212, "time": 0.39384415149688723, "epoch": 7, "memory": 9465, "step": 79996} +{"lr": 0.0007959536998847743, "data_time": 0.0013707876205444336, "grad_norm": 0.020537405088543893, "loss": 0.6874856650829315, "time": 0.3853396654129028, "epoch": 7, "memory": 9465, "step": 80096} +{"lr": 0.0007959536998847743, "data_time": 0.0012266397476196288, "grad_norm": 0.018940770369954407, "loss": 0.6914855420589447, "time": 0.4090343713760376, "epoch": 7, "memory": 9465, "step": 80196} +{"lr": 0.0007959536998847743, "data_time": 0.0013055086135864257, "grad_norm": 0.028440984967164695, "loss": 0.6907103359699249, "time": 0.405469274520874, "epoch": 7, "memory": 9465, "step": 80296} +{"lr": 0.0007959536998847743, "data_time": 0.0012084484100341798, "grad_norm": 0.02373149050399661, "loss": 0.6889601469039917, "time": 0.3896803617477417, "epoch": 7, "memory": 9465, "step": 80396} +{"lr": 0.0007959536998847743, "data_time": 0.0017619371414184571, "grad_norm": 0.01712862554122694, "loss": 0.6891317307949066, "time": 0.46973021030426027, "epoch": 7, "memory": 9465, "step": 80496} +{"lr": 0.0007959536998847743, "data_time": 0.007248044013977051, "grad_norm": 0.01794057325460017, "loss": 0.6850728094577789, "time": 0.38886318206787107, "epoch": 7, "memory": 9465, "step": 80596} +{"accuracy/top1": 100.0, "data_time": 0.30626318034003763, "time": 0.5597720987656537, "step": 7} +{"lr": 0.0007297252973710758, "data_time": 0.08447706699371338, "grad_norm": 0.023277034144848587, "loss": 0.6917500734329224, "time": 0.38696401119232177, "epoch": 8, "memory": 9465, "step": 80712} +{"lr": 0.0007297252973710758, "data_time": 0.0013402700424194336, "grad_norm": 0.028954599611461164, "loss": 0.6854966700077056, "time": 0.3594782829284668, "epoch": 8, "memory": 9465, "step": 80812} +{"lr": 0.0007297252973710758, "data_time": 0.001197481155395508, "grad_norm": 0.012177791181602515, "loss": 0.6884900450706481, "time": 0.3827489376068115, "epoch": 8, "memory": 9465, "step": 80912} +{"lr": 0.0007297252973710758, "data_time": 0.0012011289596557616, "grad_norm": 0.02472599729662761, "loss": 0.6870001137256623, "time": 0.3923820734024048, "epoch": 8, "memory": 9465, "step": 81012} +{"lr": 0.0007297252973710758, "data_time": 0.001185464859008789, "grad_norm": 0.02104019853286445, "loss": 0.6873000979423523, "time": 0.39201998710632324, "epoch": 8, "memory": 9465, "step": 81112} +{"lr": 0.0007297252973710758, "data_time": 0.0014771223068237305, "grad_norm": 0.024233171017840505, "loss": 0.6884700119495392, "time": 0.42905073165893554, "epoch": 8, "memory": 9465, "step": 81212} +{"lr": 0.0007297252973710758, "data_time": 0.0010958433151245118, "grad_norm": 0.020955532940570266, "loss": 0.6871427893638611, "time": 0.3991347312927246, "epoch": 8, "memory": 9465, "step": 81312} +{"lr": 0.0007297252973710758, "data_time": 0.0012176990509033202, "grad_norm": 0.021414365619421005, "loss": 0.690663194656372, "time": 0.44790043830871584, "epoch": 8, "memory": 9465, "step": 81412} +{"lr": 0.0007297252973710758, "data_time": 0.0012326240539550781, "grad_norm": 0.02547398323658854, "loss": 0.6890263557434082, "time": 0.4333322763442993, "epoch": 8, "memory": 9465, "step": 81512} +{"lr": 0.0007297252973710758, "data_time": 0.001192164421081543, "grad_norm": 0.028446821169927715, "loss": 0.6854319989681243, "time": 0.45966885089874265, "epoch": 8, "memory": 9465, "step": 81612} +{"lr": 0.0007297252973710758, "data_time": 0.0011616230010986328, "grad_norm": 0.02645565108396113, "loss": 0.6852212011814117, "time": 0.37725861072540284, "epoch": 8, "memory": 9465, "step": 81712} +{"lr": 0.0007297252973710758, "data_time": 0.0011505603790283204, "grad_norm": 0.015086224023252725, "loss": 0.6828705668449402, "time": 0.40449302196502684, "epoch": 8, "memory": 9465, "step": 81812} +{"lr": 0.0007297252973710758, "data_time": 0.0012871265411376954, "grad_norm": 0.02012719577178359, "loss": 0.6825924575328827, "time": 0.4122276782989502, "epoch": 8, "memory": 9465, "step": 81912} +{"lr": 0.0007297252973710758, "data_time": 0.0011530876159667968, "grad_norm": 0.037408497929573056, "loss": 0.6819666743278503, "time": 0.3867088794708252, "epoch": 8, "memory": 9465, "step": 82012} +{"lr": 0.0007297252973710758, "data_time": 0.0012230396270751954, "grad_norm": 0.026454673893749715, "loss": 0.686492258310318, "time": 0.41706888675689696, "epoch": 8, "memory": 9465, "step": 82112} +{"lr": 0.0007297252973710758, "data_time": 0.0011826038360595703, "grad_norm": 0.020667389349546282, "loss": 0.6856553554534912, "time": 0.3936256170272827, "epoch": 8, "memory": 9465, "step": 82212} +{"lr": 0.0007297252973710758, "data_time": 0.0012821912765502929, "grad_norm": 0.025168696604669093, "loss": 0.6886514246463775, "time": 0.3955011129379272, "epoch": 8, "memory": 9465, "step": 82312} +{"lr": 0.0007297252973710758, "data_time": 0.0012822389602661134, "grad_norm": 0.013831566064618528, "loss": 0.688584178686142, "time": 0.41800537109375, "epoch": 8, "memory": 9465, "step": 82412} +{"lr": 0.0007297252973710758, "data_time": 0.0014978170394897461, "grad_norm": 0.03208147487603128, "loss": 0.6853248953819275, "time": 0.41456100940704343, "epoch": 8, "memory": 9465, "step": 82512} +{"lr": 0.0007297252973710758, "data_time": 0.0011444091796875, "grad_norm": 0.021822835318744183, "loss": 0.6866038501262665, "time": 0.4193004369735718, "epoch": 8, "memory": 9465, "step": 82612} +{"lr": 0.0007297252973710758, "data_time": 0.0011280298233032227, "grad_norm": 0.020326456637121737, "loss": 0.6886625587940216, "time": 0.40841264724731446, "epoch": 8, "memory": 9465, "step": 82712} +{"lr": 0.0007297252973710758, "data_time": 0.001205778121948242, "grad_norm": 0.02877616004552692, "loss": 0.6822507977485657, "time": 0.39947516918182374, "epoch": 8, "memory": 9465, "step": 82812} +{"lr": 0.0007297252973710758, "data_time": 0.001206183433532715, "grad_norm": 0.01602583437343128, "loss": 0.6874861299991608, "time": 0.3794590711593628, "epoch": 8, "memory": 9465, "step": 82912} +{"lr": 0.0007297252973710758, "data_time": 0.0011326789855957032, "grad_norm": 0.022695856541395186, "loss": 0.6854865074157714, "time": 0.3932990789413452, "epoch": 8, "memory": 9465, "step": 83012} +{"lr": 0.0007297252973710758, "data_time": 0.001281428337097168, "grad_norm": 0.03093284945935011, "loss": 0.6858721256256104, "time": 0.35994648933410645, "epoch": 8, "memory": 9465, "step": 83112} +{"lr": 0.0007297252973710758, "data_time": 0.0011367321014404297, "grad_norm": 0.02100286316126585, "loss": 0.6811021983623504, "time": 0.3889488935470581, "epoch": 8, "memory": 9465, "step": 83212} +{"lr": 0.0007297252973710758, "data_time": 0.0010964632034301757, "grad_norm": 0.017064707742247266, "loss": 0.6817765295505523, "time": 0.4008512496948242, "epoch": 8, "memory": 9465, "step": 83312} +{"lr": 0.0007297252973710758, "data_time": 0.0013264656066894532, "grad_norm": 0.02359428247436881, "loss": 0.6816625952720642, "time": 0.4141338348388672, "epoch": 8, "memory": 9465, "step": 83412} +{"lr": 0.0007297252973710758, "data_time": 0.001012706756591797, "grad_norm": 0.020844429964199662, "loss": 0.687481015920639, "time": 0.41402957439422605, "epoch": 8, "memory": 9465, "step": 83512} +{"lr": 0.0007297252973710758, "data_time": 0.001160430908203125, "grad_norm": 0.023272443003952502, "loss": 0.6937455773353577, "time": 0.4011988639831543, "epoch": 8, "memory": 9465, "step": 83612} +{"lr": 0.0007297252973710758, "data_time": 0.001248311996459961, "grad_norm": 0.03451326321810484, "loss": 0.6823717772960662, "time": 0.38255836963653567, "epoch": 8, "memory": 9465, "step": 83712} +{"lr": 0.0007297252973710758, "data_time": 0.001235032081604004, "grad_norm": 0.026614805823192, "loss": 0.6878908574581146, "time": 0.4235095739364624, "epoch": 8, "memory": 9465, "step": 83812} +{"lr": 0.0007297252973710758, "data_time": 0.001331329345703125, "grad_norm": 0.023098617256619037, "loss": 0.686500358581543, "time": 0.43498597145080564, "epoch": 8, "memory": 9465, "step": 83912} +{"lr": 0.0007297252973710758, "data_time": 0.001229715347290039, "grad_norm": 0.024808541499078275, "loss": 0.6883776068687439, "time": 0.41284153461456297, "epoch": 8, "memory": 9465, "step": 84012} +{"lr": 0.0007297252973710758, "data_time": 0.001268458366394043, "grad_norm": 0.0225776806473732, "loss": 0.6872128546237946, "time": 0.41157534122467043, "epoch": 8, "memory": 9465, "step": 84112} +{"lr": 0.0007297252973710758, "data_time": 0.001364898681640625, "grad_norm": 0.020946176210418343, "loss": 0.6859925150871277, "time": 0.40628390312194823, "epoch": 8, "memory": 9465, "step": 84212} +{"lr": 0.0007297252973710758, "data_time": 0.0010748863220214843, "grad_norm": 0.026104954653419554, "loss": 0.6917192161083221, "time": 0.38813974857330324, "epoch": 8, "memory": 9465, "step": 84312} +{"lr": 0.0007297252973710758, "data_time": 0.0010761260986328126, "grad_norm": 0.015459264512173831, "loss": 0.6921374261379242, "time": 0.4128370523452759, "epoch": 8, "memory": 9465, "step": 84412} +{"lr": 0.0007297252973710758, "data_time": 0.0010637283325195313, "grad_norm": 0.013546761852921918, "loss": 0.6864924967288971, "time": 0.42748820781707764, "epoch": 8, "memory": 9465, "step": 84512} +{"lr": 0.0007297252973710758, "data_time": 0.0011240243911743164, "grad_norm": 0.027023932128213347, "loss": 0.689000940322876, "time": 0.3852929353713989, "epoch": 8, "memory": 9465, "step": 84612} +{"lr": 0.0007297252973710758, "data_time": 0.4488420248031616, "grad_norm": 0.01824718089774251, "loss": 0.6857051968574523, "time": 0.6342859745025635, "epoch": 8, "memory": 9465, "step": 84712} +{"lr": 0.0007297252973710758, "data_time": 0.18600850105285643, "grad_norm": 0.023536592070013285, "loss": 0.6856965780258178, "time": 0.3697806358337402, "epoch": 8, "memory": 9465, "step": 84812} +{"lr": 0.0007297252973710758, "data_time": 0.19487955570220947, "grad_norm": 0.030082438513636588, "loss": 0.6878475427627564, "time": 0.42903966903686525, "epoch": 8, "memory": 9465, "step": 84912} +{"lr": 0.0007297252973710758, "data_time": 0.0012598514556884765, "grad_norm": 0.016145851742476226, "loss": 0.6812762320041656, "time": 0.42850422859191895, "epoch": 8, "memory": 9465, "step": 85012} +{"lr": 0.0007297252973710758, "data_time": 0.0013853549957275391, "grad_norm": 0.019658190233167262, "loss": 0.6805278420448303, "time": 0.4062574148178101, "epoch": 8, "memory": 9465, "step": 85112} +{"lr": 0.0007297252973710758, "data_time": 0.001171278953552246, "grad_norm": 0.019844309845939278, "loss": 0.6873342275619507, "time": 0.4607765436172485, "epoch": 8, "memory": 9465, "step": 85212} +{"lr": 0.0007297252973710758, "data_time": 0.001273798942565918, "grad_norm": 0.03570436269510537, "loss": 0.6896699666976929, "time": 0.40429470539093015, "epoch": 8, "memory": 9465, "step": 85312} +{"lr": 0.0007297252973710758, "data_time": 0.0012304544448852538, "grad_norm": 0.02322594877332449, "loss": 0.6885689616203308, "time": 0.4417370080947876, "epoch": 8, "memory": 9465, "step": 85412} +{"lr": 0.0007297252973710758, "data_time": 0.0013516902923583984, "grad_norm": 0.022229703702032567, "loss": 0.6852067291736603, "time": 0.37205698490142824, "epoch": 8, "memory": 9465, "step": 85512} +{"lr": 0.0007297252973710758, "data_time": 0.0012696266174316406, "grad_norm": 0.01820289217866957, "loss": 0.6862127542495727, "time": 0.41309685707092286, "epoch": 8, "memory": 9465, "step": 85612} +{"lr": 0.0007297252973710758, "data_time": 0.0014190673828125, "grad_norm": 0.028342454880475997, "loss": 0.6854941487312317, "time": 0.4241366147994995, "epoch": 8, "memory": 9465, "step": 85712} +{"lr": 0.0007297252973710758, "data_time": 0.0012758255004882812, "grad_norm": 0.016617366089485587, "loss": 0.6876291632652283, "time": 0.39320549964904783, "epoch": 8, "memory": 9465, "step": 85812} +{"lr": 0.0007297252973710758, "data_time": 0.0011086225509643554, "grad_norm": 0.014217626006575302, "loss": 0.6864918768405914, "time": 0.4088895797729492, "epoch": 8, "memory": 9465, "step": 85912} +{"lr": 0.0007297252973710758, "data_time": 0.0012398958206176758, "grad_norm": 0.009858486987650395, "loss": 0.6821779072284698, "time": 0.39141838550567626, "epoch": 8, "memory": 9465, "step": 86012} +{"lr": 0.0007297252973710758, "data_time": 0.0012630224227905273, "grad_norm": 0.026772038033232092, "loss": 0.684651905298233, "time": 0.42710540294647215, "epoch": 8, "memory": 9465, "step": 86112} +{"lr": 0.0007297252973710758, "data_time": 0.0012676239013671875, "grad_norm": 0.018086789874359965, "loss": 0.687130618095398, "time": 0.3956389665603638, "epoch": 8, "memory": 9465, "step": 86212} +{"lr": 0.0007297252973710758, "data_time": 0.001221632957458496, "grad_norm": 0.016600300755817442, "loss": 0.689531821012497, "time": 0.40857982635498047, "epoch": 8, "memory": 9465, "step": 86312} +{"lr": 0.0007297252973710758, "data_time": 0.0011564731597900391, "grad_norm": 0.02437058964278549, "loss": 0.6827565252780914, "time": 0.414105749130249, "epoch": 8, "memory": 9465, "step": 86412} +{"lr": 0.0007297252973710758, "data_time": 0.0011553525924682616, "grad_norm": 0.03960214058170095, "loss": 0.689712381362915, "time": 0.4052400588989258, "epoch": 8, "memory": 9465, "step": 86512} +{"lr": 0.0007297252973710758, "data_time": 0.0011344432830810546, "grad_norm": 0.022618011478334664, "loss": 0.6871628165245056, "time": 0.41004009246826173, "epoch": 8, "memory": 9465, "step": 86612} +{"lr": 0.0007297252973710758, "data_time": 0.0011563777923583984, "grad_norm": 0.015828104829415678, "loss": 0.6897797107696533, "time": 0.399530029296875, "epoch": 8, "memory": 9465, "step": 86712} +{"lr": 0.0007297252973710758, "data_time": 0.001150655746459961, "grad_norm": 0.029650754760950804, "loss": 0.6859541833400726, "time": 0.3921663284301758, "epoch": 8, "memory": 9465, "step": 86812} +{"lr": 0.0007297252973710758, "data_time": 0.0012651443481445312, "grad_norm": 0.01859331934247166, "loss": 0.694051080942154, "time": 0.42839818000793456, "epoch": 8, "memory": 9465, "step": 86912} +{"lr": 0.0007297252973710758, "data_time": 0.001208019256591797, "grad_norm": 0.022289685532450677, "loss": 0.6856022715568543, "time": 0.40266454219818115, "epoch": 8, "memory": 9465, "step": 87012} +{"lr": 0.0007297252973710758, "data_time": 0.0011482954025268555, "grad_norm": 0.02167035867460072, "loss": 0.6878075122833252, "time": 0.46272203922271726, "epoch": 8, "memory": 9465, "step": 87112} +{"lr": 0.0007297252973710758, "data_time": 0.0011661052703857422, "grad_norm": 0.014263918832875788, "loss": 0.6901412069797516, "time": 0.4115774393081665, "epoch": 8, "memory": 9465, "step": 87212} +{"lr": 0.0007297252973710758, "data_time": 0.0014473915100097657, "grad_norm": 0.01804503982421011, "loss": 0.6925581812858581, "time": 0.42563605308532715, "epoch": 8, "memory": 9465, "step": 87312} +{"lr": 0.0007297252973710758, "data_time": 0.0016886234283447266, "grad_norm": 0.01981075443327427, "loss": 0.6851196885108948, "time": 0.40288805961608887, "epoch": 8, "memory": 9465, "step": 87412} +{"lr": 0.0007297252973710758, "data_time": 0.0017270326614379882, "grad_norm": 0.023243179032579064, "loss": 0.6839932262897491, "time": 0.3996387004852295, "epoch": 8, "memory": 9465, "step": 87512} +{"lr": 0.0007297252973710758, "data_time": 0.0011975526809692382, "grad_norm": 0.02818984305486083, "loss": 0.6847031474113464, "time": 0.4121211767196655, "epoch": 8, "memory": 9465, "step": 87612} +{"lr": 0.0007297252973710758, "data_time": 0.0018535375595092774, "grad_norm": 0.02783660786226392, "loss": 0.6824337482452393, "time": 0.41201837062835694, "epoch": 8, "memory": 9465, "step": 87712} +{"lr": 0.0007297252973710758, "data_time": 0.001318216323852539, "grad_norm": 0.01894610975869, "loss": 0.6840990722179413, "time": 0.41992831230163574, "epoch": 8, "memory": 9465, "step": 87812} +{"lr": 0.0007297252973710758, "data_time": 0.0012621164321899414, "grad_norm": 0.020018459553830324, "loss": 0.6858372092247009, "time": 0.4103346824645996, "epoch": 8, "memory": 9465, "step": 87912} +{"lr": 0.0007297252973710758, "data_time": 0.0011466026306152343, "grad_norm": 0.01521154697984457, "loss": 0.6900941610336304, "time": 0.3752453327178955, "epoch": 8, "memory": 9465, "step": 88012} +{"lr": 0.0007297252973710758, "data_time": 0.0011850833892822266, "grad_norm": 0.03133570360951125, "loss": 0.6882062613964081, "time": 0.38090322017669676, "epoch": 8, "memory": 9465, "step": 88112} +{"lr": 0.0007297252973710758, "data_time": 0.0010639190673828124, "grad_norm": 0.01772117118816823, "loss": 0.6841363072395324, "time": 0.39819214344024656, "epoch": 8, "memory": 9465, "step": 88212} +{"lr": 0.0007297252973710758, "data_time": 0.0013394355773925781, "grad_norm": 0.026048915507271887, "loss": 0.6875195622444152, "time": 0.48444066047668455, "epoch": 8, "memory": 9465, "step": 88312} +{"lr": 0.0007297252973710758, "data_time": 0.001224374771118164, "grad_norm": 0.02083610587287694, "loss": 0.6871038615703583, "time": 0.4258877754211426, "epoch": 8, "memory": 9465, "step": 88412} +{"lr": 0.0007297252973710758, "data_time": 0.0012317895889282227, "grad_norm": 0.018010238697752357, "loss": 0.6880809366703033, "time": 0.4014564037322998, "epoch": 8, "memory": 9465, "step": 88512} +{"lr": 0.0007297252973710758, "data_time": 0.0014194965362548829, "grad_norm": 0.021679312142077835, "loss": 0.6838792622089386, "time": 0.38787331581115725, "epoch": 8, "memory": 9465, "step": 88612} +{"lr": 0.0007297252973710758, "data_time": 0.001169109344482422, "grad_norm": 0.022499702125787734, "loss": 0.6892295241355896, "time": 0.41290483474731443, "epoch": 8, "memory": 9465, "step": 88712} +{"lr": 0.0007297252973710758, "data_time": 0.0011868000030517579, "grad_norm": 0.03122631898149848, "loss": 0.689541357755661, "time": 0.40138251781463624, "epoch": 8, "memory": 9465, "step": 88812} +{"lr": 0.0007297252973710758, "data_time": 0.0011133670806884766, "grad_norm": 0.018278383743017913, "loss": 0.6843248307704926, "time": 0.37915620803833006, "epoch": 8, "memory": 9465, "step": 88912} +{"lr": 0.0007297252973710758, "data_time": 0.0012299537658691407, "grad_norm": 0.01585726662306115, "loss": 0.682845950126648, "time": 0.4047560453414917, "epoch": 8, "memory": 9465, "step": 89012} +{"lr": 0.0007297252973710758, "data_time": 0.0012470483779907227, "grad_norm": 0.01803396185860038, "loss": 0.6858809888362885, "time": 0.40805821418762206, "epoch": 8, "memory": 9465, "step": 89112} +{"lr": 0.0007297252973710758, "data_time": 0.0011643409729003907, "grad_norm": 0.03050779558252543, "loss": 0.6923555791378021, "time": 0.44962224960327146, "epoch": 8, "memory": 9465, "step": 89212} +{"lr": 0.0007297252973710758, "data_time": 0.001296687126159668, "grad_norm": 0.016095905905240217, "loss": 0.6883684039115906, "time": 0.4264099359512329, "epoch": 8, "memory": 9465, "step": 89312} +{"lr": 0.0007297252973710758, "data_time": 0.0011588096618652343, "grad_norm": 0.02768868338316679, "loss": 0.6868324279785156, "time": 0.39534337520599366, "epoch": 8, "memory": 9465, "step": 89412} +{"lr": 0.0007297252973710758, "data_time": 0.0011719226837158202, "grad_norm": 0.026650596118997782, "loss": 0.687915700674057, "time": 0.4062569856643677, "epoch": 8, "memory": 9465, "step": 89512} +{"lr": 0.0007297252973710758, "data_time": 0.0011327266693115234, "grad_norm": 0.025433100806549192, "loss": 0.6886302828788757, "time": 0.41426811218261717, "epoch": 8, "memory": 9465, "step": 89612} +{"lr": 0.0007297252973710758, "data_time": 0.0012078523635864259, "grad_norm": 0.03159934119321406, "loss": 0.6852601170539856, "time": 0.4151745796203613, "epoch": 8, "memory": 9465, "step": 89712} +{"lr": 0.0007297252973710758, "data_time": 0.0011881589889526367, "grad_norm": 0.01946532188449055, "loss": 0.689168268442154, "time": 0.4193460941314697, "epoch": 8, "memory": 9465, "step": 89812} +{"lr": 0.0007297252973710758, "data_time": 0.001197981834411621, "grad_norm": 0.030266016721725464, "loss": 0.6798470675945282, "time": 0.39806716442108153, "epoch": 8, "memory": 9465, "step": 89912} +{"lr": 0.0007297252973710758, "data_time": 0.0012630462646484376, "grad_norm": 0.03459731894545257, "loss": 0.6840537667274476, "time": 0.3963681697845459, "epoch": 8, "memory": 9465, "step": 90012} +{"lr": 0.0007297252973710758, "data_time": 0.001177668571472168, "grad_norm": 0.01730092572979629, "loss": 0.6854472517967224, "time": 0.4102996587753296, "epoch": 8, "memory": 9465, "step": 90112} +{"lr": 0.0007297252973710758, "data_time": 0.0011160135269165038, "grad_norm": 0.017559571348829194, "loss": 0.6928240776062011, "time": 0.40048491954803467, "epoch": 8, "memory": 9465, "step": 90212} +{"lr": 0.0007297252973710758, "data_time": 0.001104259490966797, "grad_norm": 0.020261542464140803, "loss": 0.6874666154384613, "time": 0.39814741611480714, "epoch": 8, "memory": 9465, "step": 90312} +{"lr": 0.0007297252973710758, "data_time": 0.0010899066925048827, "grad_norm": 0.016924621985526755, "loss": 0.683213222026825, "time": 0.37892324924468995, "epoch": 8, "memory": 9465, "step": 90412} +{"lr": 0.0007297252973710758, "data_time": 0.001235222816467285, "grad_norm": 0.016168435849249364, "loss": 0.6863664031028748, "time": 0.4211923360824585, "epoch": 8, "memory": 9465, "step": 90512} +{"lr": 0.0007297252973710758, "data_time": 0.0011926651000976562, "grad_norm": 0.023549871100112797, "loss": 0.6911492764949798, "time": 0.378183650970459, "epoch": 8, "memory": 9465, "step": 90612} +{"lr": 0.0007297252973710758, "data_time": 0.001127314567565918, "grad_norm": 0.02852336352225393, "loss": 0.684430330991745, "time": 0.41238906383514407, "epoch": 8, "memory": 9465, "step": 90712} +{"lr": 0.0007297252973710758, "data_time": 0.001247715950012207, "grad_norm": 0.014994151052087545, "loss": 0.6882424712181091, "time": 0.389222526550293, "epoch": 8, "memory": 9465, "step": 90812} +{"lr": 0.0007297252973710758, "data_time": 0.0012914896011352538, "grad_norm": 0.025133726629428567, "loss": 0.6864078521728516, "time": 0.41447546482086184, "epoch": 8, "memory": 9465, "step": 90912} +{"lr": 0.0007297252973710758, "data_time": 0.0013945579528808593, "grad_norm": 0.030232203379273414, "loss": 0.6928736448287964, "time": 0.42232742309570315, "epoch": 8, "memory": 9465, "step": 91012} +{"lr": 0.0007297252973710758, "data_time": 0.0015008926391601562, "grad_norm": 0.014172716904431582, "loss": 0.6900639533996582, "time": 0.3962843656539917, "epoch": 8, "memory": 9465, "step": 91112} +{"lr": 0.0007297252973710758, "data_time": 0.00219118595123291, "grad_norm": 0.026410426991060377, "loss": 0.6892644107341767, "time": 0.42124485969543457, "epoch": 8, "memory": 9465, "step": 91212} +{"lr": 0.0007297252973710758, "data_time": 0.001233220100402832, "grad_norm": 0.02648461067583412, "loss": 0.6854311227798462, "time": 0.41231694221496584, "epoch": 8, "memory": 9465, "step": 91312} +{"lr": 0.0007297252973710758, "data_time": 0.0011304378509521484, "grad_norm": 0.03151758648455143, "loss": 0.6852314591407775, "time": 0.40566048622131345, "epoch": 8, "memory": 9465, "step": 91412} +{"lr": 0.0007297252973710758, "data_time": 0.0012223958969116212, "grad_norm": 0.02860237369313836, "loss": 0.686645257472992, "time": 0.40322227478027345, "epoch": 8, "memory": 9465, "step": 91512} +{"lr": 0.0007297252973710758, "data_time": 0.0011346578598022462, "grad_norm": 0.025324149383232, "loss": 0.6860270082950592, "time": 0.43115417957305907, "epoch": 8, "memory": 9465, "step": 91612} +{"lr": 0.0007297252973710758, "data_time": 0.0013410806655883788, "grad_norm": 0.03411727251950651, "loss": 0.6894769132137298, "time": 0.42202248573303225, "epoch": 8, "memory": 9465, "step": 91712} +{"lr": 0.0007297252973710758, "data_time": 0.0012592554092407226, "grad_norm": 0.029838087595999242, "loss": 0.6915993332862854, "time": 0.3938627004623413, "epoch": 8, "memory": 9465, "step": 91812} +{"lr": 0.0007297252973710758, "data_time": 0.001236724853515625, "grad_norm": 0.02972464361228049, "loss": 0.6858161389827728, "time": 0.3675376415252686, "epoch": 8, "memory": 9465, "step": 91912} +{"lr": 0.0007297252973710758, "data_time": 0.001224827766418457, "grad_norm": 0.02403484135866165, "loss": 0.6879396319389344, "time": 0.44392731189727785, "epoch": 8, "memory": 9465, "step": 92012} +{"lr": 0.0007297252973710758, "data_time": 0.007342123985290527, "grad_norm": 0.016623965930193663, "loss": 0.6868308842182159, "time": 0.40576231479644775, "epoch": 8, "memory": 9465, "step": 92112} +{"accuracy/top1": 100.0, "data_time": 0.2820587438695571, "time": 0.5355185200186336, "step": 8} +{"lr": 0.0006579634122155991, "data_time": 0.21316945552825928, "grad_norm": 0.0234228708781302, "loss": 0.6916179835796357, "time": 0.396101975440979, "epoch": 9, "memory": 9465, "step": 92228} +{"lr": 0.0006579634122155991, "data_time": 0.04762403964996338, "grad_norm": 0.015285061160102486, "loss": 0.6891472935676575, "time": 0.43029587268829345, "epoch": 9, "memory": 9465, "step": 92328} +{"lr": 0.0006579634122155991, "data_time": 0.0012217283248901368, "grad_norm": 0.010836223780643195, "loss": 0.6847985863685608, "time": 0.4090731143951416, "epoch": 9, "memory": 9465, "step": 92428} +{"lr": 0.0006579634122155991, "data_time": 0.001170206069946289, "grad_norm": 0.023985810473095625, "loss": 0.6876272141933442, "time": 0.4341857671737671, "epoch": 9, "memory": 9465, "step": 92528} +{"lr": 0.0006579634122155991, "data_time": 0.0011333227157592773, "grad_norm": 0.024069071374833585, "loss": 0.6895691573619842, "time": 0.494635009765625, "epoch": 9, "memory": 9465, "step": 92628} +{"lr": 0.0006579634122155991, "data_time": 0.0014793872833251953, "grad_norm": 0.01886992247309536, "loss": 0.684550428390503, "time": 0.401670503616333, "epoch": 9, "memory": 9465, "step": 92728} +{"lr": 0.0006579634122155991, "data_time": 0.0013691902160644532, "grad_norm": 0.022801924776285887, "loss": 0.6878998517990113, "time": 0.41439669132232665, "epoch": 9, "memory": 9465, "step": 92828} +{"lr": 0.0006579634122155991, "data_time": 0.0012319564819335937, "grad_norm": 0.02375576599733904, "loss": 0.6876055479049683, "time": 0.4170017957687378, "epoch": 9, "memory": 9465, "step": 92928} +{"lr": 0.0006579634122155991, "data_time": 0.0012707471847534179, "grad_norm": 0.016375646647065877, "loss": 0.6861737668514252, "time": 0.40897631645202637, "epoch": 9, "memory": 9465, "step": 93028} +{"lr": 0.0006579634122155991, "data_time": 0.0012650728225708009, "grad_norm": 0.0159662498626858, "loss": 0.6853348195552826, "time": 0.4129276990890503, "epoch": 9, "memory": 9465, "step": 93128} +{"lr": 0.0006579634122155991, "data_time": 0.0012438297271728516, "grad_norm": 0.021633118903264404, "loss": 0.6862115025520324, "time": 0.40263330936431885, "epoch": 9, "memory": 9465, "step": 93228} +{"lr": 0.0006579634122155991, "data_time": 0.001230168342590332, "grad_norm": 0.030090645235031844, "loss": 0.6870250225067138, "time": 0.39903361797332765, "epoch": 9, "memory": 9465, "step": 93328} +{"lr": 0.0006579634122155991, "data_time": 0.0012187480926513672, "grad_norm": 0.019000594364479186, "loss": 0.6873220205307007, "time": 0.388306713104248, "epoch": 9, "memory": 9465, "step": 93428} +{"lr": 0.0006579634122155991, "data_time": 0.0011335134506225586, "grad_norm": 0.021870032860897483, "loss": 0.6882164001464843, "time": 0.42376468181610105, "epoch": 9, "memory": 9465, "step": 93528} +{"lr": 0.0006579634122155991, "data_time": 0.001211690902709961, "grad_norm": 0.02227505382616073, "loss": 0.685147899389267, "time": 0.3879836559295654, "epoch": 9, "memory": 9465, "step": 93628} +{"lr": 0.0006579634122155991, "data_time": 0.001041102409362793, "grad_norm": 0.020397668541409075, "loss": 0.686493045091629, "time": 0.4055838108062744, "epoch": 9, "memory": 9465, "step": 93728} +{"lr": 0.0006579634122155991, "data_time": 0.001279306411743164, "grad_norm": 0.026303742500022053, "loss": 0.6855940282344818, "time": 0.4523406267166138, "epoch": 9, "memory": 9465, "step": 93828} +{"lr": 0.0006579634122155991, "data_time": 0.0016120433807373046, "grad_norm": 0.020197241753339767, "loss": 0.688467937707901, "time": 0.437934947013855, "epoch": 9, "memory": 9465, "step": 93928} +{"lr": 0.0006579634122155991, "data_time": 0.001268315315246582, "grad_norm": 0.02168433303013444, "loss": 0.6895956993103027, "time": 0.3716425895690918, "epoch": 9, "memory": 9465, "step": 94028} +{"lr": 0.0006579634122155991, "data_time": 0.001171255111694336, "grad_norm": 0.021969118062406777, "loss": 0.6842774868011474, "time": 0.394378662109375, "epoch": 9, "memory": 9465, "step": 94128} +{"lr": 0.0006579634122155991, "data_time": 0.0011736154556274414, "grad_norm": 0.022297876980155708, "loss": 0.6893316030502319, "time": 0.38500878810882566, "epoch": 9, "memory": 9465, "step": 94228} +{"lr": 0.0006579634122155991, "data_time": 0.0012616634368896485, "grad_norm": 0.02019666782580316, "loss": 0.6862617373466492, "time": 0.4002028226852417, "epoch": 9, "memory": 9465, "step": 94328} +{"lr": 0.0006579634122155991, "data_time": 0.0013898849487304688, "grad_norm": 0.02815713370218873, "loss": 0.6930375337600708, "time": 0.42486631870269775, "epoch": 9, "memory": 9465, "step": 94428} +{"lr": 0.0006579634122155991, "data_time": 0.001305699348449707, "grad_norm": 0.01949597888160497, "loss": 0.6851170063018799, "time": 0.38422245979309083, "epoch": 9, "memory": 9465, "step": 94528} +{"lr": 0.0006579634122155991, "data_time": 0.001342177391052246, "grad_norm": 0.008654957532417028, "loss": 0.6843866109848022, "time": 0.43166699409484866, "epoch": 9, "memory": 9465, "step": 94628} +{"lr": 0.0006579634122155991, "data_time": 0.0012539863586425782, "grad_norm": 0.020748585771070794, "loss": 0.6877948224544526, "time": 0.4316052198410034, "epoch": 9, "memory": 9465, "step": 94728} +{"lr": 0.0006579634122155991, "data_time": 0.001287221908569336, "grad_norm": 0.016579051362350584, "loss": 0.6863248229026795, "time": 0.3958492994308472, "epoch": 9, "memory": 9465, "step": 94828} +{"lr": 0.0006579634122155991, "data_time": 0.0012607574462890625, "grad_norm": 0.0332856229506433, "loss": 0.6900288879871368, "time": 0.439675498008728, "epoch": 9, "memory": 9465, "step": 94928} +{"lr": 0.0006579634122155991, "data_time": 0.0012688636779785156, "grad_norm": 0.030556600773707032, "loss": 0.6874858319759369, "time": 0.4259216785430908, "epoch": 9, "memory": 9465, "step": 95028} +{"lr": 0.0006579634122155991, "data_time": 0.0012953042984008788, "grad_norm": 0.023428255063481628, "loss": 0.6891760647296905, "time": 0.41685597896575927, "epoch": 9, "memory": 9465, "step": 95128} +{"lr": 0.0006579634122155991, "data_time": 0.0010758161544799805, "grad_norm": 0.016678719443734735, "loss": 0.6888595283031463, "time": 0.4266186237335205, "epoch": 9, "memory": 9465, "step": 95228} +{"lr": 0.0006579634122155991, "data_time": 0.0013385772705078124, "grad_norm": 0.01847218030306976, "loss": 0.6866507172584534, "time": 0.42702975273132326, "epoch": 9, "memory": 9465, "step": 95328} +{"lr": 0.0006579634122155991, "data_time": 0.001246500015258789, "grad_norm": 0.018197174969827756, "loss": 0.6861519396305085, "time": 0.4428264856338501, "epoch": 9, "memory": 9465, "step": 95428} +{"lr": 0.0006579634122155991, "data_time": 0.0017017602920532226, "grad_norm": 0.03002235693857074, "loss": 0.6791818618774415, "time": 0.4208193302154541, "epoch": 9, "memory": 9465, "step": 95528} +{"lr": 0.0006579634122155991, "data_time": 0.001302480697631836, "grad_norm": 0.024182516941800715, "loss": 0.687585073709488, "time": 0.3937672138214111, "epoch": 9, "memory": 9465, "step": 95628} +{"lr": 0.0006579634122155991, "data_time": 0.0012251853942871094, "grad_norm": 0.017337709735147654, "loss": 0.6875366985797882, "time": 0.4132816314697266, "epoch": 9, "memory": 9465, "step": 95728} +{"lr": 0.0006579634122155991, "data_time": 0.0011914968490600586, "grad_norm": 0.01870363319758326, "loss": 0.685635793209076, "time": 0.3906074285507202, "epoch": 9, "memory": 9465, "step": 95828} +{"lr": 0.0006579634122155991, "data_time": 0.0012731313705444335, "grad_norm": 0.026897142501547934, "loss": 0.6896338284015655, "time": 0.40096268653869627, "epoch": 9, "memory": 9465, "step": 95928} +{"lr": 0.0006579634122155991, "data_time": 0.0011874675750732423, "grad_norm": 0.026509001932572572, "loss": 0.6889773905277252, "time": 0.4020853519439697, "epoch": 9, "memory": 9465, "step": 96028} +{"lr": 0.0006579634122155991, "data_time": 0.0011425018310546875, "grad_norm": 0.017402482125908135, "loss": 0.6823527455329895, "time": 0.42270355224609374, "epoch": 9, "memory": 9465, "step": 96128} +{"lr": 0.0006579634122155991, "data_time": 0.0011400461196899414, "grad_norm": 0.02895842413417995, "loss": 0.6830333232879638, "time": 0.37814233303070066, "epoch": 9, "memory": 9465, "step": 96228} +{"lr": 0.0006579634122155991, "data_time": 0.0011759757995605468, "grad_norm": 0.023471585963852702, "loss": 0.6858464479446411, "time": 0.42189993858337405, "epoch": 9, "memory": 9465, "step": 96328} +{"lr": 0.0006579634122155991, "data_time": 0.001221466064453125, "grad_norm": 0.0167266639880836, "loss": 0.6858098208904266, "time": 0.41883187294006347, "epoch": 9, "memory": 9465, "step": 96428} +{"lr": 0.0006579634122155991, "data_time": 0.0012425899505615235, "grad_norm": 0.01642593522556126, "loss": 0.6833726942539216, "time": 0.3721702814102173, "epoch": 9, "memory": 9465, "step": 96528} +{"lr": 0.0006579634122155991, "data_time": 0.0013547182083129884, "grad_norm": 0.01680624104337767, "loss": 0.6890249371528625, "time": 0.3774850845336914, "epoch": 9, "memory": 9465, "step": 96628} +{"lr": 0.0006579634122155991, "data_time": 0.0012394905090332032, "grad_norm": 0.0279288194142282, "loss": 0.6881324768066406, "time": 0.40616984367370607, "epoch": 9, "memory": 9465, "step": 96728} +{"lr": 0.0006579634122155991, "data_time": 0.0012462854385375977, "grad_norm": 0.01503880093805492, "loss": 0.6883620500564576, "time": 0.40850157737731935, "epoch": 9, "memory": 9465, "step": 96828} +{"lr": 0.0006579634122155991, "data_time": 0.001309037208557129, "grad_norm": 0.025298327021300793, "loss": 0.6819562196731568, "time": 0.4332940101623535, "epoch": 9, "memory": 9465, "step": 96928} +{"lr": 0.0006579634122155991, "data_time": 0.0012633800506591797, "grad_norm": 0.0193327292567119, "loss": 0.686792254447937, "time": 0.37613890171051023, "epoch": 9, "memory": 9465, "step": 97028} +{"lr": 0.0006579634122155991, "data_time": 0.0012845039367675782, "grad_norm": 0.020719549898058177, "loss": 0.6884935736656189, "time": 0.3788888931274414, "epoch": 9, "memory": 9465, "step": 97128} +{"lr": 0.0006579634122155991, "data_time": 0.0013127088546752929, "grad_norm": 0.02421436319127679, "loss": 0.6868078708648682, "time": 0.4024457216262817, "epoch": 9, "memory": 9465, "step": 97228} +{"lr": 0.0006579634122155991, "data_time": 0.001302814483642578, "grad_norm": 0.01988020937424153, "loss": 0.6854013919830322, "time": 0.42250816822052, "epoch": 9, "memory": 9465, "step": 97328} +{"lr": 0.0006579634122155991, "data_time": 0.0011916160583496094, "grad_norm": 0.022660642373375596, "loss": 0.6879563987255096, "time": 0.38332290649414064, "epoch": 9, "memory": 9465, "step": 97428} +{"lr": 0.0006579634122155991, "data_time": 0.0013241767883300781, "grad_norm": 0.02686635326826945, "loss": 0.6822048485279083, "time": 0.39268765449523924, "epoch": 9, "memory": 9465, "step": 97528} +{"lr": 0.0006579634122155991, "data_time": 0.0016896724700927734, "grad_norm": 0.027195728803053498, "loss": 0.6898121058940887, "time": 0.4255571126937866, "epoch": 9, "memory": 9465, "step": 97628} +{"lr": 0.0006579634122155991, "data_time": 0.0016141176223754884, "grad_norm": 0.0168335084192222, "loss": 0.6855267524719239, "time": 0.4973564863204956, "epoch": 9, "memory": 9465, "step": 97728} +{"lr": 0.0006579634122155991, "data_time": 0.0012742280960083008, "grad_norm": 0.015347224660217762, "loss": 0.6904765784740448, "time": 0.412762975692749, "epoch": 9, "memory": 9465, "step": 97828} +{"lr": 0.0006579634122155991, "data_time": 0.001308298110961914, "grad_norm": 0.014064839849015697, "loss": 0.6878889083862305, "time": 0.40703887939453126, "epoch": 9, "memory": 9465, "step": 97928} +{"lr": 0.0006579634122155991, "data_time": 0.0012140274047851562, "grad_norm": 0.029925155732780696, "loss": 0.6871650576591491, "time": 0.44304039478302004, "epoch": 9, "memory": 9465, "step": 98028} +{"lr": 0.0006579634122155991, "data_time": 0.0013524770736694336, "grad_norm": 0.023618994350545108, "loss": 0.6859748244285584, "time": 0.39569027423858644, "epoch": 9, "memory": 9465, "step": 98128} +{"lr": 0.0006579634122155991, "data_time": 0.0012902021408081055, "grad_norm": 0.018692607013508677, "loss": 0.6859188854694367, "time": 0.3941347122192383, "epoch": 9, "memory": 9465, "step": 98228} +{"lr": 0.0006579634122155991, "data_time": 0.0013632535934448241, "grad_norm": 0.01690459387609735, "loss": 0.6879775702953339, "time": 0.4223912000656128, "epoch": 9, "memory": 9465, "step": 98328} +{"lr": 0.0006579634122155991, "data_time": 0.001253199577331543, "grad_norm": 0.02279731973540038, "loss": 0.6867093026638031, "time": 0.38918366432189944, "epoch": 9, "memory": 9465, "step": 98428} +{"lr": 0.0006579634122155991, "data_time": 0.001302337646484375, "grad_norm": 0.025851780222728847, "loss": 0.6896377682685852, "time": 0.38460984230041506, "epoch": 9, "memory": 9465, "step": 98528} +{"lr": 0.0006579634122155991, "data_time": 0.0011951923370361328, "grad_norm": 0.011671100463718175, "loss": 0.6869745969772338, "time": 0.41370439529418945, "epoch": 9, "memory": 9465, "step": 98628} +{"lr": 0.0006579634122155991, "data_time": 0.0012279510498046874, "grad_norm": 0.030667106620967388, "loss": 0.6872926354408264, "time": 0.37771522998809814, "epoch": 9, "memory": 9465, "step": 98728} +{"lr": 0.0006579634122155991, "data_time": 0.0011768102645874023, "grad_norm": 0.012709329824429006, "loss": 0.688293707370758, "time": 0.40707762241363527, "epoch": 9, "memory": 9465, "step": 98828} +{"lr": 0.0006579634122155991, "data_time": 0.0012590408325195313, "grad_norm": 0.03183954511769116, "loss": 0.6826155662536622, "time": 0.39499711990356445, "epoch": 9, "memory": 9465, "step": 98928} +{"lr": 0.0006579634122155991, "data_time": 0.0011528730392456055, "grad_norm": 0.019440938270417974, "loss": 0.6883721292018891, "time": 0.3722320318222046, "epoch": 9, "memory": 9465, "step": 99028} +{"lr": 0.0006579634122155991, "data_time": 0.0012531518936157227, "grad_norm": 0.027565144328400494, "loss": 0.6806558012962342, "time": 0.38812177181243895, "epoch": 9, "memory": 9465, "step": 99128} +{"lr": 0.0006579634122155991, "data_time": 0.0011704921722412109, "grad_norm": 0.02675180253572762, "loss": 0.6876543939113617, "time": 0.40490987300872805, "epoch": 9, "memory": 9465, "step": 99228} +{"lr": 0.0006579634122155991, "data_time": 0.0012486696243286133, "grad_norm": 0.030168826109729707, "loss": 0.6888048231601716, "time": 0.381657338142395, "epoch": 9, "memory": 9465, "step": 99328} +{"lr": 0.0006579634122155991, "data_time": 0.0011351585388183593, "grad_norm": 0.028028684156015515, "loss": 0.6887628912925721, "time": 0.4116715431213379, "epoch": 9, "memory": 9465, "step": 99428} +{"lr": 0.0006579634122155991, "data_time": 0.001234722137451172, "grad_norm": 0.015527657582424581, "loss": 0.6899739027023315, "time": 0.41590731143951415, "epoch": 9, "memory": 9465, "step": 99528} +{"lr": 0.0006579634122155991, "data_time": 0.0012464523315429688, "grad_norm": 0.030353108141571283, "loss": 0.6921488285064697, "time": 0.47470993995666505, "epoch": 9, "memory": 9465, "step": 99628} +{"lr": 0.0006579634122155991, "data_time": 0.0012114286422729493, "grad_norm": 0.029835300310514866, "loss": 0.6887951672077179, "time": 0.4400243520736694, "epoch": 9, "memory": 9465, "step": 99728} +{"lr": 0.0006579634122155991, "data_time": 0.0012079715728759766, "grad_norm": 0.025192872807383537, "loss": 0.6869944512844086, "time": 0.3784618377685547, "epoch": 9, "memory": 9465, "step": 99828} +{"lr": 0.0006579634122155991, "data_time": 0.0012845754623413085, "grad_norm": 0.02521377563243732, "loss": 0.6825485050678253, "time": 0.3821455478668213, "epoch": 9, "memory": 9465, "step": 99928} +{"lr": 0.0006579634122155991, "data_time": 0.0012262344360351562, "grad_norm": 0.023734575789421796, "loss": 0.686549437046051, "time": 0.42076358795166013, "epoch": 9, "memory": 9465, "step": 100028} +{"lr": 0.0006579634122155991, "data_time": 0.001176929473876953, "grad_norm": 0.022418857598677278, "loss": 0.6821747839450836, "time": 0.3837489366531372, "epoch": 9, "memory": 9465, "step": 100128} +{"lr": 0.0006579634122155991, "data_time": 0.0012503862380981445, "grad_norm": 0.02005963319679722, "loss": 0.6834537267684937, "time": 0.398844575881958, "epoch": 9, "memory": 9465, "step": 100228} +{"lr": 0.0006579634122155991, "data_time": 0.001214003562927246, "grad_norm": 0.02611571641173214, "loss": 0.6858814537525177, "time": 0.4339579105377197, "epoch": 9, "memory": 9465, "step": 100328} +{"lr": 0.0006579634122155991, "data_time": 0.0012241601943969727, "grad_norm": 0.03148531918413937, "loss": 0.6862568199634552, "time": 0.40589926242828367, "epoch": 9, "memory": 9465, "step": 100428} +{"lr": 0.0006579634122155991, "data_time": 0.0012091636657714845, "grad_norm": 0.033940835297107695, "loss": 0.6856343150138855, "time": 0.3979132890701294, "epoch": 9, "memory": 9465, "step": 100528} +{"lr": 0.0006579634122155991, "data_time": 0.001158738136291504, "grad_norm": 0.019394355139229446, "loss": 0.6881734967231751, "time": 0.4203500270843506, "epoch": 9, "memory": 9465, "step": 100628} +{"lr": 0.0006579634122155991, "data_time": 0.0012041568756103516, "grad_norm": 0.020852927234955132, "loss": 0.6846345901489258, "time": 0.4341021776199341, "epoch": 9, "memory": 9465, "step": 100728} +{"lr": 0.0006579634122155991, "data_time": 0.0011731863021850586, "grad_norm": 0.021724076382815836, "loss": 0.6865647315979004, "time": 0.4252492666244507, "epoch": 9, "memory": 9465, "step": 100828} +{"lr": 0.0006579634122155991, "data_time": 0.00117340087890625, "grad_norm": 0.01835655963514, "loss": 0.6896529734134674, "time": 0.4314539670944214, "epoch": 9, "memory": 9465, "step": 100928} +{"lr": 0.0006579634122155991, "data_time": 0.0012068510055541991, "grad_norm": 0.02051096213981509, "loss": 0.6877733886241912, "time": 0.42658677101135256, "epoch": 9, "memory": 9465, "step": 101028} +{"lr": 0.0006579634122155991, "data_time": 0.0012018442153930663, "grad_norm": 0.022103709960356356, "loss": 0.687483298778534, "time": 0.3764130353927612, "epoch": 9, "memory": 9465, "step": 101128} +{"lr": 0.0006579634122155991, "data_time": 0.0013461589813232421, "grad_norm": 0.031499871611595155, "loss": 0.6910863101482392, "time": 0.4267105579376221, "epoch": 9, "memory": 9465, "step": 101228} +{"lr": 0.0006579634122155991, "data_time": 0.0013976573944091796, "grad_norm": 0.020280804665526377, "loss": 0.6822407126426697, "time": 0.418511962890625, "epoch": 9, "memory": 9465, "step": 101328} +{"lr": 0.0006579634122155991, "data_time": 0.0012742996215820313, "grad_norm": 0.021270034462213518, "loss": 0.6890386521816254, "time": 0.4001427173614502, "epoch": 9, "memory": 9465, "step": 101428} +{"lr": 0.0006579634122155991, "data_time": 0.0012665987014770508, "grad_norm": 0.020746080833487213, "loss": 0.6890262246131897, "time": 0.43777031898498536, "epoch": 9, "memory": 9465, "step": 101528} +{"lr": 0.0006579634122155991, "data_time": 0.00123443603515625, "grad_norm": 0.030383997317403554, "loss": 0.6889639258384704, "time": 0.41897132396698, "epoch": 9, "memory": 9465, "step": 101628} +{"lr": 0.0006579634122155991, "data_time": 0.0012430429458618164, "grad_norm": 0.028350890969159083, "loss": 0.6842820227146149, "time": 0.3890629768371582, "epoch": 9, "memory": 9465, "step": 101728} +{"lr": 0.0006579634122155991, "data_time": 0.001310443878173828, "grad_norm": 0.026324432389810682, "loss": 0.6874460697174072, "time": 0.4347853183746338, "epoch": 9, "memory": 9465, "step": 101828} +{"lr": 0.0006579634122155991, "data_time": 0.0011256933212280273, "grad_norm": 0.02528970562852919, "loss": 0.688225245475769, "time": 0.4044374465942383, "epoch": 9, "memory": 9465, "step": 101928} +{"lr": 0.0006579634122155991, "data_time": 0.0013589859008789062, "grad_norm": 0.03300562538206577, "loss": 0.6893951058387756, "time": 0.41233081817626954, "epoch": 9, "memory": 9465, "step": 102028} +{"lr": 0.0006579634122155991, "data_time": 0.0012313604354858398, "grad_norm": 0.01551717626862228, "loss": 0.6841417729854584, "time": 0.40691328048706055, "epoch": 9, "memory": 9465, "step": 102128} +{"lr": 0.0006579634122155991, "data_time": 0.0013275861740112305, "grad_norm": 0.018457791511900724, "loss": 0.6879209458827973, "time": 0.396535587310791, "epoch": 9, "memory": 9465, "step": 102228} +{"lr": 0.0006579634122155991, "data_time": 0.0011706590652465821, "grad_norm": 0.02424745513126254, "loss": 0.683158141374588, "time": 0.4237743616104126, "epoch": 9, "memory": 9465, "step": 102328} +{"lr": 0.0006579634122155991, "data_time": 0.001270437240600586, "grad_norm": 0.025121981231495737, "loss": 0.6909631550312042, "time": 0.39954843521118166, "epoch": 9, "memory": 9465, "step": 102428} +{"lr": 0.0006579634122155991, "data_time": 0.001394367218017578, "grad_norm": 0.01886954139918089, "loss": 0.6874514400959015, "time": 0.3963741064071655, "epoch": 9, "memory": 9465, "step": 102528} +{"lr": 0.0006579634122155991, "data_time": 0.0015153884887695312, "grad_norm": 0.021362853090977296, "loss": 0.6876825988292694, "time": 0.41606242656707765, "epoch": 9, "memory": 9465, "step": 102628} +{"lr": 0.0006579634122155991, "data_time": 0.0012106180191040039, "grad_norm": 0.017824929393827914, "loss": 0.6851708590984344, "time": 0.40473825931549073, "epoch": 9, "memory": 9465, "step": 102728} +{"lr": 0.0006579634122155991, "data_time": 0.0011362075805664063, "grad_norm": 0.024924999999348073, "loss": 0.6884391784667969, "time": 0.41689631938934324, "epoch": 9, "memory": 9465, "step": 102828} +{"lr": 0.0006579634122155991, "data_time": 0.0012011051177978516, "grad_norm": 0.01819471991620958, "loss": 0.6900894582271576, "time": 0.38872756958007815, "epoch": 9, "memory": 9465, "step": 102928} +{"lr": 0.0006579634122155991, "data_time": 0.0011909723281860352, "grad_norm": 0.02048467976273969, "loss": 0.6794112265110016, "time": 0.40266592502593995, "epoch": 9, "memory": 9465, "step": 103028} +{"lr": 0.0006579634122155991, "data_time": 0.0012024402618408202, "grad_norm": 0.019741402816725893, "loss": 0.6905640184879303, "time": 0.4593635559082031, "epoch": 9, "memory": 9465, "step": 103128} +{"lr": 0.0006579634122155991, "data_time": 0.001197671890258789, "grad_norm": 0.014451659796759487, "loss": 0.687445068359375, "time": 0.4000712871551514, "epoch": 9, "memory": 9465, "step": 103228} +{"lr": 0.0006579634122155991, "data_time": 0.002363777160644531, "grad_norm": 0.03637495692819357, "loss": 0.6868040025234222, "time": 0.4112098217010498, "epoch": 9, "memory": 9465, "step": 103328} +{"lr": 0.0006579634122155991, "data_time": 0.0013206005096435547, "grad_norm": 0.024048620974645017, "loss": 0.6868383586406708, "time": 0.40303263664245603, "epoch": 9, "memory": 9465, "step": 103428} +{"lr": 0.0006579634122155991, "data_time": 0.0011281490325927735, "grad_norm": 0.028736840840429066, "loss": 0.6859465658664703, "time": 0.4159888029098511, "epoch": 9, "memory": 9465, "step": 103528} +{"lr": 0.0006579634122155991, "data_time": 0.007851600646972656, "grad_norm": 0.0213921959977597, "loss": 0.6927263677120209, "time": 0.39697990417480467, "epoch": 9, "memory": 9465, "step": 103628} +{"accuracy/top1": 100.0, "data_time": 0.29366599812227134, "time": 0.546827610801248, "step": 9} +{"lr": 0.0005824350601949144, "data_time": 0.0804297924041748, "grad_norm": 0.013860852117068135, "loss": 0.684959477186203, "time": 0.38721024990081787, "epoch": 10, "memory": 9465, "step": 103744} +{"lr": 0.0005824350601949144, "data_time": 0.1245685338973999, "grad_norm": 0.018632193864323198, "loss": 0.6929050743579864, "time": 0.37447028160095214, "epoch": 10, "memory": 9465, "step": 103844} +{"lr": 0.0005824350601949144, "data_time": 0.0011754274368286134, "grad_norm": 0.019292819849215447, "loss": 0.6920138597488403, "time": 0.3911352872848511, "epoch": 10, "memory": 9465, "step": 103944} +{"lr": 0.0005824350601949144, "data_time": 0.0014087915420532226, "grad_norm": 0.020488617662340402, "loss": 0.6909263849258422, "time": 0.41617798805236816, "epoch": 10, "memory": 9465, "step": 104044} +{"lr": 0.0005824350601949144, "data_time": 0.0013932228088378907, "grad_norm": 0.02226473849732429, "loss": 0.6895103275775909, "time": 0.39312267303466797, "epoch": 10, "memory": 9465, "step": 104144} +{"lr": 0.0005824350601949144, "data_time": 0.0014477014541625977, "grad_norm": 0.023597622476518153, "loss": 0.6877914369106293, "time": 0.4322096347808838, "epoch": 10, "memory": 9465, "step": 104244} +{"lr": 0.0005824350601949144, "data_time": 0.001198554039001465, "grad_norm": 0.01968251629732549, "loss": 0.6820769429206848, "time": 0.4268588304519653, "epoch": 10, "memory": 9465, "step": 104344} +{"lr": 0.0005824350601949144, "data_time": 0.001171422004699707, "grad_norm": 0.016102762633818202, "loss": 0.6871347963809967, "time": 0.42116336822509765, "epoch": 10, "memory": 9465, "step": 104444} +{"lr": 0.0005824350601949144, "data_time": 0.001184391975402832, "grad_norm": 0.018892600177787246, "loss": 0.6875913798809051, "time": 0.39415688514709474, "epoch": 10, "memory": 9465, "step": 104544} +{"lr": 0.0005824350601949144, "data_time": 0.0013638973236083985, "grad_norm": 0.028315074648708104, "loss": 0.6880623936653137, "time": 0.41167688369750977, "epoch": 10, "memory": 9465, "step": 104644} +{"lr": 0.0005824350601949144, "data_time": 0.0013931512832641602, "grad_norm": 0.025507490057498215, "loss": 0.6845067024230957, "time": 0.4208183526992798, "epoch": 10, "memory": 9465, "step": 104744} +{"lr": 0.0005824350601949144, "data_time": 0.0014745235443115235, "grad_norm": 0.03318627830594778, "loss": 0.6920928955078125, "time": 0.42701256275177, "epoch": 10, "memory": 9465, "step": 104844} +{"lr": 0.0005824350601949144, "data_time": 0.001563429832458496, "grad_norm": 0.021531057625543325, "loss": 0.6887095749378205, "time": 0.40250022411346437, "epoch": 10, "memory": 9465, "step": 104944} +{"lr": 0.0005824350601949144, "data_time": 0.0012716531753540039, "grad_norm": 0.019334487069863827, "loss": 0.6901447772979736, "time": 0.4074194669723511, "epoch": 10, "memory": 9465, "step": 105044} +{"lr": 0.0005824350601949144, "data_time": 0.0013767242431640624, "grad_norm": 0.020994788268581033, "loss": 0.6887441456317902, "time": 0.4227423191070557, "epoch": 10, "memory": 9465, "step": 105144} +{"lr": 0.0005824350601949144, "data_time": 0.0013175487518310546, "grad_norm": 0.011245237581897527, "loss": 0.6934567332267761, "time": 0.40805351734161377, "epoch": 10, "memory": 9465, "step": 105244} +{"lr": 0.0005824350601949144, "data_time": 0.0012956619262695312, "grad_norm": 0.016551787732169032, "loss": 0.6876155853271484, "time": 0.4531639337539673, "epoch": 10, "memory": 9465, "step": 105344} +{"lr": 0.0005824350601949144, "data_time": 0.0013253927230834962, "grad_norm": 0.03667248806450516, "loss": 0.6847810208797455, "time": 0.4050638675689697, "epoch": 10, "memory": 9465, "step": 105444} +{"lr": 0.0005824350601949144, "data_time": 0.0012734651565551758, "grad_norm": 0.015060140064451843, "loss": 0.6832494020462037, "time": 0.42069971561431885, "epoch": 10, "memory": 9465, "step": 105544} +{"lr": 0.0005824350601949144, "data_time": 0.0013261079788208009, "grad_norm": 0.01931798756122589, "loss": 0.6855398416519165, "time": 0.3904897928237915, "epoch": 10, "memory": 9465, "step": 105644} +{"lr": 0.0005824350601949144, "data_time": 0.0013689517974853516, "grad_norm": 0.019257644531899133, "loss": 0.6879183769226074, "time": 0.42248609066009524, "epoch": 10, "memory": 9465, "step": 105744} +{"lr": 0.0005824350601949144, "data_time": 0.001352548599243164, "grad_norm": 0.015942090796306728, "loss": 0.6878875076770783, "time": 0.41836502552032473, "epoch": 10, "memory": 9465, "step": 105844} +{"lr": 0.0005824350601949144, "data_time": 0.0013040781021118164, "grad_norm": 0.015442746109329164, "loss": 0.6886744022369384, "time": 0.42407824993133547, "epoch": 10, "memory": 9465, "step": 105944} +{"lr": 0.0005824350601949144, "data_time": 0.0014090538024902344, "grad_norm": 0.027122497325763106, "loss": 0.6842085003852845, "time": 0.4306140661239624, "epoch": 10, "memory": 9465, "step": 106044} +{"lr": 0.0005824350601949144, "data_time": 0.0012214183807373047, "grad_norm": 0.02640653122216463, "loss": 0.6877634763717652, "time": 0.41361632347106936, "epoch": 10, "memory": 9465, "step": 106144} +{"lr": 0.0005824350601949144, "data_time": 0.0012167930603027345, "grad_norm": 0.015804155403748153, "loss": 0.6896946609020234, "time": 0.4164297342300415, "epoch": 10, "memory": 9465, "step": 106244} +{"lr": 0.0005824350601949144, "data_time": 0.0012772560119628906, "grad_norm": 0.019613543688319624, "loss": 0.6849848628044128, "time": 0.4279727220535278, "epoch": 10, "memory": 9465, "step": 106344} +{"lr": 0.0005824350601949144, "data_time": 0.0011636734008789063, "grad_norm": 0.02436817828565836, "loss": 0.6858685910701752, "time": 0.3970036029815674, "epoch": 10, "memory": 9465, "step": 106444} +{"lr": 0.0005824350601949144, "data_time": 0.0012277126312255859, "grad_norm": 0.019343935884535313, "loss": 0.6839489638805389, "time": 0.3918203592300415, "epoch": 10, "memory": 9465, "step": 106544} +{"lr": 0.0005824350601949144, "data_time": 0.0012360572814941405, "grad_norm": 0.031003579404205082, "loss": 0.6848509788513184, "time": 0.4468141794204712, "epoch": 10, "memory": 9465, "step": 106644} +{"lr": 0.0005824350601949144, "data_time": 0.0013895988464355468, "grad_norm": 0.030244832765311, "loss": 0.6901010274887085, "time": 0.42935519218444823, "epoch": 10, "memory": 9465, "step": 106744} +{"lr": 0.0005824350601949144, "data_time": 0.0013350248336791992, "grad_norm": 0.019989809626713394, "loss": 0.6829847097396851, "time": 0.40437328815460205, "epoch": 10, "memory": 9465, "step": 106844} +{"lr": 0.0005824350601949144, "data_time": 0.0012146472930908204, "grad_norm": 0.020788033190183342, "loss": 0.6902220666408538, "time": 0.3854987382888794, "epoch": 10, "memory": 9465, "step": 106944} +{"lr": 0.0005824350601949144, "data_time": 0.0011486530303955079, "grad_norm": 0.02227186084492132, "loss": 0.6845167636871338, "time": 0.39415597915649414, "epoch": 10, "memory": 9465, "step": 107044} +{"lr": 0.0005824350601949144, "data_time": 0.0011939287185668945, "grad_norm": 0.015719819790683688, "loss": 0.6869750022888184, "time": 0.41704909801483153, "epoch": 10, "memory": 9465, "step": 107144} +{"lr": 0.0005824350601949144, "data_time": 0.0012527942657470704, "grad_norm": 0.018355125561356544, "loss": 0.6825352430343627, "time": 0.41206822395324705, "epoch": 10, "memory": 9465, "step": 107244} +{"lr": 0.0005824350601949144, "data_time": 0.0012099266052246094, "grad_norm": 0.020194839173927902, "loss": 0.6805288314819335, "time": 0.40749800205230713, "epoch": 10, "memory": 9465, "step": 107344} +{"lr": 0.0005824350601949144, "data_time": 0.0013065338134765625, "grad_norm": 0.019903663831064476, "loss": 0.6839028298854828, "time": 0.41579749584198, "epoch": 10, "memory": 9465, "step": 107444} +{"lr": 0.0005824350601949144, "data_time": 0.0012295007705688476, "grad_norm": 0.023555016424506903, "loss": 0.6887306034564972, "time": 0.40896925926208494, "epoch": 10, "memory": 9465, "step": 107544} +{"lr": 0.0005824350601949144, "data_time": 0.0013578414916992187, "grad_norm": 0.008026356843765825, "loss": 0.6864909052848815, "time": 0.39210598468780516, "epoch": 10, "memory": 9465, "step": 107644} +{"lr": 0.0005824350601949144, "data_time": 0.0013225793838500977, "grad_norm": 0.01994687117403373, "loss": 0.6855593800544739, "time": 0.41773104667663574, "epoch": 10, "memory": 9465, "step": 107744} +{"lr": 0.0005824350601949144, "data_time": 0.0011875391006469726, "grad_norm": 0.015936049411538988, "loss": 0.6819383978843689, "time": 0.4239983558654785, "epoch": 10, "memory": 9465, "step": 107844} +{"lr": 0.0005824350601949144, "data_time": 0.0012995481491088867, "grad_norm": 0.0163878261228092, "loss": 0.6855361342430115, "time": 0.4112903833389282, "epoch": 10, "memory": 9465, "step": 107944} +{"lr": 0.0005824350601949144, "data_time": 0.001398181915283203, "grad_norm": 0.019322561647277324, "loss": 0.6876338481903076, "time": 0.4021707057952881, "epoch": 10, "memory": 9465, "step": 108044} +{"lr": 0.0005824350601949144, "data_time": 0.0012164831161499024, "grad_norm": 0.01873928731074557, "loss": 0.6879354000091553, "time": 0.4513095378875732, "epoch": 10, "memory": 9465, "step": 108144} +{"lr": 0.0005824350601949144, "data_time": 0.0015116214752197265, "grad_norm": 0.031186088593676685, "loss": 0.6894765496253967, "time": 0.39508953094482424, "epoch": 10, "memory": 9465, "step": 108244} +{"lr": 0.0005824350601949144, "data_time": 0.00115659236907959, "grad_norm": 0.019406280026305466, "loss": 0.6872951865196228, "time": 0.38469705581665037, "epoch": 10, "memory": 9465, "step": 108344} +{"lr": 0.0005824350601949144, "data_time": 0.0013974905014038086, "grad_norm": 0.016117651236709208, "loss": 0.6880839169025421, "time": 0.414851975440979, "epoch": 10, "memory": 9465, "step": 108444} +{"lr": 0.0005824350601949144, "data_time": 0.0012469530105590821, "grad_norm": 0.021941594569943844, "loss": 0.691986083984375, "time": 0.3929115056991577, "epoch": 10, "memory": 9465, "step": 108544} +{"lr": 0.0005824350601949144, "data_time": 0.001242852210998535, "grad_norm": 0.03727470549638383, "loss": 0.689212304353714, "time": 0.3955352306365967, "epoch": 10, "memory": 9465, "step": 108644} +{"lr": 0.0005824350601949144, "data_time": 0.0013436079025268555, "grad_norm": 0.036667583021335305, "loss": 0.6849142134189605, "time": 0.4265798330307007, "epoch": 10, "memory": 9465, "step": 108744} +{"lr": 0.0005824350601949144, "data_time": 0.0012444496154785157, "grad_norm": 0.030465218820609152, "loss": 0.6896410048007965, "time": 0.3953246116638184, "epoch": 10, "memory": 9465, "step": 108844} +{"lr": 0.0005824350601949144, "data_time": 0.001269221305847168, "grad_norm": 0.017400019080378117, "loss": 0.6845678567886353, "time": 0.4077514886856079, "epoch": 10, "memory": 9465, "step": 108944} +{"lr": 0.0005824350601949144, "data_time": 0.0013150691986083985, "grad_norm": 0.02755537023767829, "loss": 0.684292984008789, "time": 0.4132340431213379, "epoch": 10, "memory": 9465, "step": 109044} +{"lr": 0.0005824350601949144, "data_time": 0.0011596202850341797, "grad_norm": 0.017595816362882034, "loss": 0.6844670414924622, "time": 0.39069271087646484, "epoch": 10, "memory": 9465, "step": 109144} +{"lr": 0.0005824350601949144, "data_time": 0.0012371301651000976, "grad_norm": 0.013313841167837381, "loss": 0.6874563574790955, "time": 0.41736831665039065, "epoch": 10, "memory": 9465, "step": 109244} +{"lr": 0.0005824350601949144, "data_time": 0.001233220100402832, "grad_norm": 0.016935050557367503, "loss": 0.6864929080009461, "time": 0.417470645904541, "epoch": 10, "memory": 9465, "step": 109344} +{"lr": 0.0005824350601949144, "data_time": 0.001202535629272461, "grad_norm": 0.022035099868662655, "loss": 0.6885490894317627, "time": 0.41223371028900146, "epoch": 10, "memory": 9465, "step": 109444} +{"lr": 0.0005824350601949144, "data_time": 0.0012761354446411133, "grad_norm": 0.02410687633091584, "loss": 0.6844773352146148, "time": 0.39715070724487306, "epoch": 10, "memory": 9465, "step": 109544} +{"lr": 0.0005824350601949144, "data_time": 0.001157855987548828, "grad_norm": 0.01917083191219717, "loss": 0.688711941242218, "time": 0.40023744106292725, "epoch": 10, "memory": 9465, "step": 109644} +{"lr": 0.0005824350601949144, "data_time": 0.0013338088989257812, "grad_norm": 0.019987792504252866, "loss": 0.6903825759887695, "time": 0.4490066051483154, "epoch": 10, "memory": 9465, "step": 109744} +{"lr": 0.0005824350601949144, "data_time": 0.0012418031692504883, "grad_norm": 0.025325045277713797, "loss": 0.6888796389102936, "time": 0.40377490520477294, "epoch": 10, "memory": 9465, "step": 109844} +{"lr": 0.0005824350601949144, "data_time": 0.0012151479721069336, "grad_norm": 0.025578478910028936, "loss": 0.6866759479045867, "time": 0.42577595710754396, "epoch": 10, "memory": 9465, "step": 109944} +{"lr": 0.0005824350601949144, "data_time": 0.001265406608581543, "grad_norm": 0.030046832282096148, "loss": 0.6862091302871705, "time": 0.41481242179870603, "epoch": 10, "memory": 9465, "step": 110044} +{"lr": 0.0005824350601949144, "data_time": 0.0011924505233764648, "grad_norm": 0.022230990463867784, "loss": 0.680774348974228, "time": 0.4091353416442871, "epoch": 10, "memory": 9465, "step": 110144} +{"lr": 0.0005824350601949144, "data_time": 0.0012224674224853515, "grad_norm": 0.02642576047219336, "loss": 0.6916180908679962, "time": 0.3774830102920532, "epoch": 10, "memory": 9465, "step": 110244} +{"lr": 0.0005824350601949144, "data_time": 0.0012408018112182618, "grad_norm": 0.030100225936621428, "loss": 0.692443722486496, "time": 0.39855546951293946, "epoch": 10, "memory": 9465, "step": 110344} +{"lr": 0.0005824350601949144, "data_time": 0.0013021230697631836, "grad_norm": 0.02331925416365266, "loss": 0.6915897607803345, "time": 0.4083374261856079, "epoch": 10, "memory": 9465, "step": 110444} +{"lr": 0.0005824350601949144, "data_time": 0.0012148380279541015, "grad_norm": 0.01458752986509353, "loss": 0.6902616620063782, "time": 0.4076382637023926, "epoch": 10, "memory": 9465, "step": 110544} +{"lr": 0.0005824350601949144, "data_time": 0.001472330093383789, "grad_norm": 0.019471430499106646, "loss": 0.689478051662445, "time": 0.4589089393615723, "epoch": 10, "memory": 9465, "step": 110644} +{"lr": 0.0005824350601949144, "data_time": 0.0012511014938354492, "grad_norm": 0.016889481525868177, "loss": 0.6862181067466736, "time": 0.4069258213043213, "epoch": 10, "memory": 9465, "step": 110744} +{"lr": 0.0005824350601949144, "data_time": 0.0012067079544067383, "grad_norm": 0.024049390805885196, "loss": 0.6864958047866822, "time": 0.3966191053390503, "epoch": 10, "memory": 9465, "step": 110844} +{"lr": 0.0005824350601949144, "data_time": 0.0013444185256958007, "grad_norm": 0.019203606760129333, "loss": 0.6846551954746246, "time": 0.4123916864395142, "epoch": 10, "memory": 9465, "step": 110944} +{"lr": 0.0005824350601949144, "data_time": 0.0014084815979003907, "grad_norm": 0.030780664936173707, "loss": 0.6866586089134217, "time": 0.4366321086883545, "epoch": 10, "memory": 9465, "step": 111044} +{"lr": 0.0005824350601949144, "data_time": 0.0012423276901245117, "grad_norm": 0.02523041081149131, "loss": 0.682997876405716, "time": 0.4074322462081909, "epoch": 10, "memory": 9465, "step": 111144} +{"lr": 0.0005824350601949144, "data_time": 0.001174616813659668, "grad_norm": 0.015973761153873056, "loss": 0.682374769449234, "time": 0.42370193004608153, "epoch": 10, "memory": 9465, "step": 111244} +{"lr": 0.0005824350601949144, "data_time": 0.0011987686157226562, "grad_norm": 0.023700334306340665, "loss": 0.6827379882335662, "time": 0.3967373609542847, "epoch": 10, "memory": 9465, "step": 111344} +{"lr": 0.0005824350601949144, "data_time": 0.0012578964233398438, "grad_norm": 0.019831335172057152, "loss": 0.6807875454425811, "time": 0.4308000087738037, "epoch": 10, "memory": 9465, "step": 111444} +{"lr": 0.0005824350601949144, "data_time": 0.0013262033462524414, "grad_norm": 0.02222868036478758, "loss": 0.6855636715888977, "time": 0.40865292549133303, "epoch": 10, "memory": 9465, "step": 111544} +{"lr": 0.0005824350601949144, "data_time": 0.0013063907623291015, "grad_norm": 0.01934937541373074, "loss": 0.6886425614356995, "time": 0.4392297029495239, "epoch": 10, "memory": 9465, "step": 111644} +{"lr": 0.0005824350601949144, "data_time": 0.0014129638671875, "grad_norm": 0.018973741790978237, "loss": 0.6863362193107605, "time": 0.4058452367782593, "epoch": 10, "memory": 9465, "step": 111744} +{"lr": 0.0005824350601949144, "data_time": 0.0012498617172241211, "grad_norm": 0.01542835826985538, "loss": 0.6853533446788788, "time": 0.42419798374176027, "epoch": 10, "memory": 9465, "step": 111844} +{"lr": 0.0005824350601949144, "data_time": 0.0012610197067260743, "grad_norm": 0.01657525389455259, "loss": 0.6841616094112396, "time": 0.4221523284912109, "epoch": 10, "memory": 9465, "step": 111944} +{"lr": 0.0005824350601949144, "data_time": 0.0013408422470092773, "grad_norm": 0.020618134015239774, "loss": 0.6899962604045868, "time": 0.4089498519897461, "epoch": 10, "memory": 9465, "step": 112044} +{"lr": 0.0005824350601949144, "data_time": 0.001222085952758789, "grad_norm": 0.017995651811361313, "loss": 0.6861677825450897, "time": 0.41017196178436277, "epoch": 10, "memory": 9465, "step": 112144} +{"lr": 0.0005824350601949144, "data_time": 0.0013261556625366211, "grad_norm": 0.020788966538384558, "loss": 0.6859902501106262, "time": 0.38940489292144775, "epoch": 10, "memory": 9465, "step": 112244} +{"lr": 0.0005824350601949144, "data_time": 0.0013193845748901366, "grad_norm": 0.016596147720701993, "loss": 0.68940749168396, "time": 0.45871844291687014, "epoch": 10, "memory": 9465, "step": 112344} +{"lr": 0.0005824350601949144, "data_time": 0.0012585639953613282, "grad_norm": 0.024194601358613, "loss": 0.6857317388057709, "time": 0.4222301721572876, "epoch": 10, "memory": 9465, "step": 112444} +{"lr": 0.0005824350601949144, "data_time": 0.0013432025909423829, "grad_norm": 0.024401912372559308, "loss": 0.6879629909992218, "time": 0.4044027328491211, "epoch": 10, "memory": 9465, "step": 112544} +{"lr": 0.0005824350601949144, "data_time": 0.00132293701171875, "grad_norm": 0.01939155168365687, "loss": 0.6883826732635498, "time": 0.41016521453857424, "epoch": 10, "memory": 9465, "step": 112644} +{"lr": 0.0005824350601949144, "data_time": 0.0012619733810424806, "grad_norm": 0.023890880751423538, "loss": 0.690188205242157, "time": 0.412406325340271, "epoch": 10, "memory": 9465, "step": 112744} +{"lr": 0.0005824350601949144, "data_time": 0.0012366294860839844, "grad_norm": 0.03024088656529784, "loss": 0.6842540144920349, "time": 0.4187090635299683, "epoch": 10, "memory": 9465, "step": 112844} +{"lr": 0.0005824350601949144, "data_time": 0.0013144254684448243, "grad_norm": 0.03496296582743526, "loss": 0.6869054853916168, "time": 0.3987264156341553, "epoch": 10, "memory": 9465, "step": 112944} +{"lr": 0.0005824350601949144, "data_time": 0.0012226104736328125, "grad_norm": 0.02497257823124528, "loss": 0.6926319003105164, "time": 0.4907251834869385, "epoch": 10, "memory": 9465, "step": 113044} +{"lr": 0.0005824350601949144, "data_time": 0.0012903213500976562, "grad_norm": 0.029246483603492378, "loss": 0.6850897789001464, "time": 0.42286856174468995, "epoch": 10, "memory": 9465, "step": 113144} +{"lr": 0.0005824350601949144, "data_time": 0.0011801004409790039, "grad_norm": 0.0186012509977445, "loss": 0.6832999348640442, "time": 0.4044408082962036, "epoch": 10, "memory": 9465, "step": 113244} +{"lr": 0.0005824350601949144, "data_time": 0.0012021064758300781, "grad_norm": 0.01730568166822195, "loss": 0.686027979850769, "time": 0.3872162103652954, "epoch": 10, "memory": 9465, "step": 113344} +{"lr": 0.0005824350601949144, "data_time": 0.0012649059295654296, "grad_norm": 0.021720893494784833, "loss": 0.691539216041565, "time": 0.3952643394470215, "epoch": 10, "memory": 9465, "step": 113444} +{"lr": 0.0005824350601949144, "data_time": 0.0013449668884277343, "grad_norm": 0.016197738028131424, "loss": 0.684565806388855, "time": 0.42101283073425294, "epoch": 10, "memory": 9465, "step": 113544} +{"lr": 0.0005824350601949144, "data_time": 0.0013080120086669921, "grad_norm": 0.020098356902599333, "loss": 0.6852782845497132, "time": 0.40638060569763185, "epoch": 10, "memory": 9465, "step": 113644} +{"lr": 0.0005824350601949144, "data_time": 0.0014593124389648438, "grad_norm": 0.029148967657238244, "loss": 0.6879068195819855, "time": 0.44136064052581786, "epoch": 10, "memory": 9465, "step": 113744} +{"lr": 0.0005824350601949144, "data_time": 0.00114288330078125, "grad_norm": 0.02002816878957674, "loss": 0.6871368885040283, "time": 0.42678251266479494, "epoch": 10, "memory": 9465, "step": 113844} +{"lr": 0.0005824350601949144, "data_time": 0.0012193679809570312, "grad_norm": 0.01567211588844657, "loss": 0.6868526756763458, "time": 0.404461932182312, "epoch": 10, "memory": 9465, "step": 113944} +{"lr": 0.0005824350601949144, "data_time": 0.001265573501586914, "grad_norm": 0.018838391359895467, "loss": 0.6869959533214569, "time": 0.42400307655334474, "epoch": 10, "memory": 9465, "step": 114044} +{"lr": 0.0005824350601949144, "data_time": 0.0012116432189941406, "grad_norm": 0.020720690640155227, "loss": 0.6856711804866791, "time": 0.41737139225006104, "epoch": 10, "memory": 9465, "step": 114144} +{"lr": 0.0005824350601949144, "data_time": 0.001208019256591797, "grad_norm": 0.02710183516610414, "loss": 0.6836977303028107, "time": 0.4092968463897705, "epoch": 10, "memory": 9465, "step": 114244} +{"lr": 0.0005824350601949144, "data_time": 0.0013273954391479492, "grad_norm": 0.022203076060395687, "loss": 0.6899046838283539, "time": 0.4248371124267578, "epoch": 10, "memory": 9465, "step": 114344} +{"lr": 0.0005824350601949144, "data_time": 0.0013229846954345703, "grad_norm": 0.02756560589186847, "loss": 0.688097459077835, "time": 0.40036327838897706, "epoch": 10, "memory": 9465, "step": 114444} +{"lr": 0.0005824350601949144, "data_time": 0.001192760467529297, "grad_norm": 0.020312445575837047, "loss": 0.6833064258098602, "time": 0.4081276893615723, "epoch": 10, "memory": 9465, "step": 114544} +{"lr": 0.0005824350601949144, "data_time": 0.001324915885925293, "grad_norm": 0.019513128011021764, "loss": 0.6836528837680816, "time": 0.4081394195556641, "epoch": 10, "memory": 9465, "step": 114644} +{"lr": 0.0005824350601949144, "data_time": 0.0012111425399780273, "grad_norm": 0.01780185862444341, "loss": 0.6832261919975281, "time": 0.427313494682312, "epoch": 10, "memory": 9465, "step": 114744} +{"lr": 0.0005824350601949144, "data_time": 0.0012250900268554687, "grad_norm": 0.017171705327928066, "loss": 0.6893916189670563, "time": 0.3691226959228516, "epoch": 10, "memory": 9465, "step": 114844} +{"lr": 0.0005824350601949144, "data_time": 0.002109813690185547, "grad_norm": 0.031381679512560365, "loss": 0.6834503531455993, "time": 0.4566788911819458, "epoch": 10, "memory": 9465, "step": 114944} +{"lr": 0.0005824350601949144, "data_time": 0.0016053199768066406, "grad_norm": 0.01726784014608711, "loss": 0.6817797839641571, "time": 0.44130597114562986, "epoch": 10, "memory": 9465, "step": 115044} +{"lr": 0.0005824350601949144, "data_time": 0.0073616266250610355, "grad_norm": 0.01543681266484782, "loss": 0.6815003514289856, "time": 0.42305099964141846, "epoch": 10, "memory": 9465, "step": 115144} +{"accuracy/top1": 100.0, "data_time": 0.31313453001134534, "time": 0.5658689386704389, "step": 10} +{"lr": 0.0005050000000000001, "data_time": 0.14164562225341798, "grad_norm": 0.018051203107461333, "loss": 0.6851199448108674, "time": 0.38462414741516116, "epoch": 11, "memory": 9465, "step": 115260} +{"lr": 0.0005050000000000001, "data_time": 0.0012343883514404296, "grad_norm": 0.03034874270670116, "loss": 0.6858681082725525, "time": 0.4263297080993652, "epoch": 11, "memory": 9465, "step": 115360} +{"lr": 0.0005050000000000001, "data_time": 0.001535511016845703, "grad_norm": 0.026627645036205648, "loss": 0.6922080814838409, "time": 0.4277806758880615, "epoch": 11, "memory": 9465, "step": 115460} +{"lr": 0.0005050000000000001, "data_time": 0.0012439966201782226, "grad_norm": 0.025177362328395247, "loss": 0.6910534083843232, "time": 0.4225890636444092, "epoch": 11, "memory": 9465, "step": 115560} +{"lr": 0.0005050000000000001, "data_time": 0.0012293577194213868, "grad_norm": 0.025857007037848233, "loss": 0.6852569341659546, "time": 0.4010313987731934, "epoch": 11, "memory": 9465, "step": 115660} +{"lr": 0.0005050000000000001, "data_time": 0.0012104034423828125, "grad_norm": 0.019561112439259887, "loss": 0.690194845199585, "time": 0.41695983409881593, "epoch": 11, "memory": 9465, "step": 115760} +{"lr": 0.0005050000000000001, "data_time": 0.0014673948287963867, "grad_norm": 0.01657569374947343, "loss": 0.6906159043312072, "time": 0.4154002904891968, "epoch": 11, "memory": 9465, "step": 115860} +{"lr": 0.0005050000000000001, "data_time": 0.0013001680374145508, "grad_norm": 0.018498407537117602, "loss": 0.6843622982501983, "time": 0.4089759349822998, "epoch": 11, "memory": 9465, "step": 115960} +{"lr": 0.0005050000000000001, "data_time": 0.0014367341995239259, "grad_norm": 0.019207592034945264, "loss": 0.6852038204669952, "time": 0.41964030265808105, "epoch": 11, "memory": 9465, "step": 116060} +{"lr": 0.0005050000000000001, "data_time": 0.0016860723495483398, "grad_norm": 0.025294320657849312, "loss": 0.6915205597877503, "time": 0.40537395477294924, "epoch": 11, "memory": 9465, "step": 116160} +{"lr": 0.0005050000000000001, "data_time": 0.0013617992401123047, "grad_norm": 0.03438049773685634, "loss": 0.6858792901039124, "time": 0.4081716775894165, "epoch": 11, "memory": 9465, "step": 116260} +{"lr": 0.0005050000000000001, "data_time": 0.0012148618698120117, "grad_norm": 0.024008320609573273, "loss": 0.6888195455074311, "time": 0.40371079444885255, "epoch": 11, "memory": 9465, "step": 116360} +{"lr": 0.0005050000000000001, "data_time": 0.0012859582901000976, "grad_norm": 0.02932122233323753, "loss": 0.6876915872097016, "time": 0.41438817977905273, "epoch": 11, "memory": 9465, "step": 116460} +{"lr": 0.0005050000000000001, "data_time": 0.0012621164321899414, "grad_norm": 0.022084969142451883, "loss": 0.6869653582572937, "time": 0.4203095197677612, "epoch": 11, "memory": 9465, "step": 116560} +{"lr": 0.0005050000000000001, "data_time": 0.0012619972229003906, "grad_norm": 0.0145648067293223, "loss": 0.6853629887104035, "time": 0.41915919780731203, "epoch": 11, "memory": 9465, "step": 116660} +{"lr": 0.0005050000000000001, "data_time": 0.0013493776321411132, "grad_norm": 0.02500706957653165, "loss": 0.6874424040317535, "time": 0.42435810565948484, "epoch": 11, "memory": 9465, "step": 116760} +{"lr": 0.0005050000000000001, "data_time": 0.001235651969909668, "grad_norm": 0.012749405700014904, "loss": 0.6920623362064362, "time": 0.4669149160385132, "epoch": 11, "memory": 9465, "step": 116860} +{"lr": 0.0005050000000000001, "data_time": 0.001263904571533203, "grad_norm": 0.012591069098562002, "loss": 0.6829416632652283, "time": 0.4248375654220581, "epoch": 11, "memory": 9465, "step": 116960} +{"lr": 0.0005050000000000001, "data_time": 0.0012073278427124023, "grad_norm": 0.02648661967832595, "loss": 0.6873231291770935, "time": 0.4109781265258789, "epoch": 11, "memory": 9465, "step": 117060} +{"lr": 0.0005050000000000001, "data_time": 0.0014919757843017579, "grad_norm": 0.022504648682661353, "loss": 0.6939443469047546, "time": 0.4185105562210083, "epoch": 11, "memory": 9465, "step": 117160} +{"lr": 0.0005050000000000001, "data_time": 0.0014093399047851562, "grad_norm": 0.01594353067339398, "loss": 0.6896181166172027, "time": 0.42703282833099365, "epoch": 11, "memory": 9465, "step": 117260} +{"lr": 0.0005050000000000001, "data_time": 0.0012426137924194335, "grad_norm": 0.02223492683842778, "loss": 0.688520348072052, "time": 0.4168489933013916, "epoch": 11, "memory": 9465, "step": 117360} +{"lr": 0.0005050000000000001, "data_time": 0.0013597726821899414, "grad_norm": 0.02550687571638264, "loss": 0.6830237984657288, "time": 0.4211557626724243, "epoch": 11, "memory": 9465, "step": 117460} +{"lr": 0.0005050000000000001, "data_time": 0.0012285232543945313, "grad_norm": 0.029983170243212953, "loss": 0.6894727110862732, "time": 0.4175532102584839, "epoch": 11, "memory": 9465, "step": 117560} +{"lr": 0.0005050000000000001, "data_time": 0.001228642463684082, "grad_norm": 0.023841824021656067, "loss": 0.685344797372818, "time": 0.4121598243713379, "epoch": 11, "memory": 9465, "step": 117660} +{"lr": 0.0005050000000000001, "data_time": 0.0013395786285400391, "grad_norm": 0.03169720433652401, "loss": 0.6829832375049592, "time": 0.4200542211532593, "epoch": 11, "memory": 9465, "step": 117760} +{"lr": 0.0005050000000000001, "data_time": 0.0012280702590942382, "grad_norm": 0.018102559680119156, "loss": 0.6852307736873626, "time": 0.4196688175201416, "epoch": 11, "memory": 9465, "step": 117860} +{"lr": 0.0005050000000000001, "data_time": 0.0013463258743286132, "grad_norm": 0.021662126760929823, "loss": 0.6852931618690491, "time": 0.4109614610671997, "epoch": 11, "memory": 9465, "step": 117960} +{"lr": 0.0005050000000000001, "data_time": 0.0013517141342163086, "grad_norm": 0.025886794889811427, "loss": 0.6886232554912567, "time": 0.42843639850616455, "epoch": 11, "memory": 9465, "step": 118060} +{"lr": 0.0005050000000000001, "data_time": 0.001398921012878418, "grad_norm": 0.018659650138579308, "loss": 0.6858470857143402, "time": 0.4139073848724365, "epoch": 11, "memory": 9465, "step": 118160} +{"lr": 0.0005050000000000001, "data_time": 0.0013772964477539063, "grad_norm": 0.02023630291223526, "loss": 0.690393716096878, "time": 0.4513564109802246, "epoch": 11, "memory": 9465, "step": 118260} +{"lr": 0.0005050000000000001, "data_time": 0.0011944055557250976, "grad_norm": 0.03460154184140265, "loss": 0.690068107843399, "time": 0.4073303699493408, "epoch": 11, "memory": 9465, "step": 118360} +{"lr": 0.0005050000000000001, "data_time": 0.001187276840209961, "grad_norm": 0.019733951438684016, "loss": 0.6868141531944275, "time": 0.4054840326309204, "epoch": 11, "memory": 9465, "step": 118460} +{"lr": 0.0005050000000000001, "data_time": 0.0012950897216796875, "grad_norm": 0.029082222480792553, "loss": 0.6933544933795929, "time": 0.436969518661499, "epoch": 11, "memory": 9465, "step": 118560} +{"lr": 0.0005050000000000001, "data_time": 0.0012650489807128906, "grad_norm": 0.030345451540779323, "loss": 0.6888700425624847, "time": 0.44688436985015867, "epoch": 11, "memory": 9465, "step": 118660} +{"lr": 0.0005050000000000001, "data_time": 0.0013356685638427734, "grad_norm": 0.033061045221984386, "loss": 0.6866863250732422, "time": 0.4123854398727417, "epoch": 11, "memory": 9465, "step": 118760} +{"lr": 0.0005050000000000001, "data_time": 0.0012506484985351563, "grad_norm": 0.018534427601844074, "loss": 0.6873053550720215, "time": 0.41796696186065674, "epoch": 11, "memory": 9465, "step": 118860} +{"lr": 0.0005050000000000001, "data_time": 0.0013590335845947265, "grad_norm": 0.022318332106806338, "loss": 0.6839488089084625, "time": 0.4034369468688965, "epoch": 11, "memory": 9465, "step": 118960} +{"lr": 0.0005050000000000001, "data_time": 0.0014347076416015626, "grad_norm": 0.04936443231999874, "loss": 0.6854557871818543, "time": 0.44001576900482176, "epoch": 11, "memory": 9465, "step": 119060} +{"lr": 0.0005050000000000001, "data_time": 0.0013712167739868164, "grad_norm": 0.11262964736670256, "loss": 0.6847430109977722, "time": 0.4342339515686035, "epoch": 11, "memory": 9465, "step": 119160} +{"lr": 0.0005050000000000001, "data_time": 0.0012398719787597655, "grad_norm": 0.047233869507908824, "loss": 0.6832587540149688, "time": 0.43180623054504397, "epoch": 11, "memory": 9465, "step": 119260} +{"lr": 0.0005050000000000001, "data_time": 0.00135495662689209, "grad_norm": 0.22071321560069918, "loss": 0.6838791906833649, "time": 0.40990076065063474, "epoch": 11, "memory": 9465, "step": 119360} +{"lr": 0.0005050000000000001, "data_time": 0.0012654304504394532, "grad_norm": 5.8065368115901945, "loss": 0.6771953344345093, "time": 0.4447260618209839, "epoch": 11, "memory": 9465, "step": 119460} +{"lr": 0.0005050000000000001, "data_time": 0.0013187646865844727, "grad_norm": 2.108214722573757, "loss": 0.6880443215370178, "time": 0.41931159496307374, "epoch": 11, "memory": 9465, "step": 119560} +{"lr": 0.0005050000000000001, "data_time": 0.0013099193572998046, "grad_norm": 11.026284086704255, "loss": 0.6785147964954377, "time": 0.4144579887390137, "epoch": 11, "memory": 9465, "step": 119660} +{"lr": 0.0005050000000000001, "data_time": 0.00125577449798584, "grad_norm": 1.82330731600523, "loss": 0.6819824814796448, "time": 0.4068878173828125, "epoch": 11, "memory": 9465, "step": 119760} +{"lr": 0.0005050000000000001, "data_time": 0.0012647390365600586, "grad_norm": 1.4509143367409707, "loss": 0.67999626994133, "time": 0.4289198398590088, "epoch": 11, "memory": 9465, "step": 119860} +{"lr": 0.0005050000000000001, "data_time": 0.0012695074081420898, "grad_norm": 1.1200944758951663, "loss": 0.678550523519516, "time": 0.4281005859375, "epoch": 11, "memory": 9465, "step": 119960} +{"lr": 0.0005050000000000001, "data_time": 0.0012899875640869141, "grad_norm": 0.08603005451150239, "loss": 0.6788290202617645, "time": 0.4542787790298462, "epoch": 11, "memory": 9465, "step": 120060} +{"lr": 0.0005050000000000001, "data_time": 0.0012372970581054688, "grad_norm": 0.3386359246447682, "loss": 0.6814377546310425, "time": 0.43357300758361816, "epoch": 11, "memory": 9465, "step": 120160} +{"lr": 0.0005050000000000001, "data_time": 0.0012021780014038086, "grad_norm": 0.10021833311766386, "loss": 0.6798589050769805, "time": 0.40308120250701907, "epoch": 11, "memory": 9465, "step": 120260} +{"lr": 0.0005050000000000001, "data_time": 0.0012942314147949218, "grad_norm": 0.14789856560528278, "loss": 0.6738684296607971, "time": 0.4436042785644531, "epoch": 11, "memory": 9465, "step": 120360} +{"lr": 0.0005050000000000001, "data_time": 0.0013613224029541016, "grad_norm": 1.6710926983505487, "loss": 0.6735816538333893, "time": 0.3911003828048706, "epoch": 11, "memory": 9465, "step": 120460} +{"lr": 0.0005050000000000001, "data_time": 0.0012817859649658203, "grad_norm": 0.08404935821890831, "loss": 0.6804642856121064, "time": 0.43289999961853026, "epoch": 11, "memory": 9465, "step": 120560} +{"lr": 0.0005050000000000001, "data_time": 0.0013806343078613282, "grad_norm": 9.849910619854928, "loss": 0.6692216038703919, "time": 0.4129424333572388, "epoch": 11, "memory": 9465, "step": 120660} +{"lr": 0.0005050000000000001, "data_time": 0.0013935327529907226, "grad_norm": 0.683423176407814, "loss": 0.6871620059013367, "time": 0.43312318325042726, "epoch": 11, "memory": 9465, "step": 120760} +{"lr": 0.0005050000000000001, "data_time": 0.001224350929260254, "grad_norm": 47.25621848106384, "loss": 0.6792611658573151, "time": 0.4233675003051758, "epoch": 11, "memory": 9465, "step": 120860} +{"lr": 0.0005050000000000001, "data_time": 0.001247739791870117, "grad_norm": 0.5494678448885679, "loss": 0.673319411277771, "time": 0.40434112548828127, "epoch": 11, "memory": 9465, "step": 120960} +{"lr": 0.0005050000000000001, "data_time": 0.0014000177383422852, "grad_norm": 0.34597770050168036, "loss": 0.6815334379673004, "time": 0.4122968912124634, "epoch": 11, "memory": 9465, "step": 121060} +{"lr": 0.0005050000000000001, "data_time": 0.0013393402099609376, "grad_norm": 1.5020484685897828, "loss": 0.6745238959789276, "time": 0.4165686845779419, "epoch": 11, "memory": 9465, "step": 121160} +{"lr": 0.0005050000000000001, "data_time": 0.001291036605834961, "grad_norm": 0.9124003171920776, "loss": 0.6776463925838471, "time": 0.3968178749084473, "epoch": 11, "memory": 9465, "step": 121260} +{"lr": 0.0005050000000000001, "data_time": 0.001394796371459961, "grad_norm": 3.354268269240856, "loss": 0.6727799534797668, "time": 0.38947367668151855, "epoch": 11, "memory": 9465, "step": 121360} +{"lr": 0.0005050000000000001, "data_time": 0.0014145612716674806, "grad_norm": 15.790675684064627, "loss": 0.6834679365158081, "time": 0.39763631820678713, "epoch": 11, "memory": 9465, "step": 121460} +{"lr": 0.0005050000000000001, "data_time": 0.001433110237121582, "grad_norm": 11.313840195536613, "loss": 0.673699015378952, "time": 0.4168685436248779, "epoch": 11, "memory": 9465, "step": 121560} +{"lr": 0.0005050000000000001, "data_time": 0.0013829469680786133, "grad_norm": 3.2813807114958764, "loss": 0.677393114566803, "time": 0.42999911308288574, "epoch": 11, "memory": 9465, "step": 121660} +{"lr": 0.0005050000000000001, "data_time": 0.0014870405197143556, "grad_norm": 1.0303614147007465, "loss": 0.682282030582428, "time": 0.4271831035614014, "epoch": 11, "memory": 9465, "step": 121760} +{"lr": 0.0005050000000000001, "data_time": 0.0013890981674194336, "grad_norm": 0.8496859695762395, "loss": 0.6810908675193786, "time": 0.40605978965759276, "epoch": 11, "memory": 9465, "step": 121860} +{"lr": 0.0005050000000000001, "data_time": 0.001494431495666504, "grad_norm": 0.5275362681597471, "loss": 0.6731388330459595, "time": 0.4266343593597412, "epoch": 11, "memory": 9465, "step": 121960} +{"lr": 0.0005050000000000001, "data_time": 0.001422286033630371, "grad_norm": 2.6346925273537636, "loss": 0.6795392811298371, "time": 0.4261953592300415, "epoch": 11, "memory": 9465, "step": 122060} +{"lr": 0.0005050000000000001, "data_time": 0.0016381978988647462, "grad_norm": 1.9452454678714275, "loss": 0.671459972858429, "time": 0.42636497020721437, "epoch": 11, "memory": 9465, "step": 122160} +{"lr": 0.0005050000000000001, "data_time": 0.0013174772262573241, "grad_norm": 8.538171362876891, "loss": 0.6888985455036163, "time": 0.4091381072998047, "epoch": 11, "memory": 9465, "step": 122260} +{"lr": 0.0005050000000000001, "data_time": 0.0012390851974487306, "grad_norm": 2.1021955788135527, "loss": 0.6825085937976837, "time": 0.40187487602233884, "epoch": 11, "memory": 9465, "step": 122360} +{"lr": 0.0005050000000000001, "data_time": 0.0013252019882202149, "grad_norm": 6.867604106664658, "loss": 0.6739128708839417, "time": 0.42259793281555175, "epoch": 11, "memory": 9465, "step": 122460} +{"lr": 0.0005050000000000001, "data_time": 0.0012115955352783204, "grad_norm": 83.43901827335358, "loss": 0.678394615650177, "time": 0.4365677833557129, "epoch": 11, "memory": 9465, "step": 122560} +{"lr": 0.0005050000000000001, "data_time": 0.0013599634170532227, "grad_norm": 2.20732289403677, "loss": 0.6636655867099762, "time": 0.3971044301986694, "epoch": 11, "memory": 9465, "step": 122660} +{"lr": 0.0005050000000000001, "data_time": 0.0013832330703735351, "grad_norm": 1.3128623604774474, "loss": 0.6777714014053344, "time": 0.40115063190460204, "epoch": 11, "memory": 9465, "step": 122760} +{"lr": 0.0005050000000000001, "data_time": 0.0013721704483032227, "grad_norm": 0.9810119360685349, "loss": 0.6685629665851593, "time": 0.3978790283203125, "epoch": 11, "memory": 9465, "step": 122860} +{"lr": 0.0005050000000000001, "data_time": 0.0012880802154541016, "grad_norm": 2.4078777730464935, "loss": 0.6691409051418304, "time": 0.45783884525299073, "epoch": 11, "memory": 9465, "step": 122960} +{"lr": 0.0005050000000000001, "data_time": 0.001217484474182129, "grad_norm": 22.93484022319317, "loss": 0.6794389367103577, "time": 0.43236308097839354, "epoch": 11, "memory": 9465, "step": 123060} +{"lr": 0.0005050000000000001, "data_time": 0.0015253782272338866, "grad_norm": 21.462710642814635, "loss": 0.6670138657093048, "time": 0.40058207511901855, "epoch": 11, "memory": 9465, "step": 123160} +{"lr": 0.0005050000000000001, "data_time": 0.0013308286666870116, "grad_norm": 7.648467391729355, "loss": 0.6769792556762695, "time": 0.40759427547454835, "epoch": 11, "memory": 9465, "step": 123260} +{"lr": 0.0005050000000000001, "data_time": 0.0013679981231689454, "grad_norm": 2.5652765333652496, "loss": 0.673036915063858, "time": 0.40096383094787597, "epoch": 11, "memory": 9465, "step": 123360} +{"lr": 0.0005050000000000001, "data_time": 0.0013195276260375977, "grad_norm": 4.90426222383976, "loss": 0.6876177847385406, "time": 0.41182947158813477, "epoch": 11, "memory": 9465, "step": 123460} +{"lr": 0.0005050000000000001, "data_time": 0.0013573884963989258, "grad_norm": 1.3246337741613388, "loss": 0.680418211221695, "time": 0.4129197120666504, "epoch": 11, "memory": 9465, "step": 123560} +{"lr": 0.0005050000000000001, "data_time": 0.0013721227645874024, "grad_norm": 2.6230124294757844, "loss": 0.6782394289970398, "time": 0.3892005443572998, "epoch": 11, "memory": 9465, "step": 123660} +{"lr": 0.0005050000000000001, "data_time": 0.0014016866683959962, "grad_norm": 6.475113850831986, "loss": 0.6695801615715027, "time": 0.4101029634475708, "epoch": 11, "memory": 9465, "step": 123760} +{"lr": 0.0005050000000000001, "data_time": 0.0013608694076538085, "grad_norm": 111.09181776046753, "loss": 0.6710107207298279, "time": 0.42042667865753175, "epoch": 11, "memory": 9465, "step": 123860} +{"lr": 0.0005050000000000001, "data_time": 0.0013858318328857423, "grad_norm": 43.57771167755127, "loss": 0.6811672568321228, "time": 0.394344162940979, "epoch": 11, "memory": 9465, "step": 123960} +{"lr": 0.0005050000000000001, "data_time": 0.0013201475143432618, "grad_norm": 183.56825037002562, "loss": 0.6773841500282287, "time": 0.46300063133239744, "epoch": 11, "memory": 9465, "step": 124060} +{"lr": 0.0005050000000000001, "data_time": 0.0012443780899047852, "grad_norm": 11.107847380638123, "loss": 0.679725068807602, "time": 0.4452333450317383, "epoch": 11, "memory": 9465, "step": 124160} +{"lr": 0.0005050000000000001, "data_time": 0.0012809514999389648, "grad_norm": 2.8669633507728576, "loss": 0.684642207622528, "time": 0.41928672790527344, "epoch": 11, "memory": 9465, "step": 124260} +{"lr": 0.0005050000000000001, "data_time": 0.0013806819915771484, "grad_norm": 11.678421294689178, "loss": 0.6807779788970947, "time": 0.4495292901992798, "epoch": 11, "memory": 9465, "step": 124360} +{"lr": 0.0005050000000000001, "data_time": 0.0012891769409179687, "grad_norm": 13.774215412139892, "loss": 0.6751906752586365, "time": 0.4242283821105957, "epoch": 11, "memory": 9465, "step": 124460} +{"lr": 0.0005050000000000001, "data_time": 0.00128326416015625, "grad_norm": 39.00203651189804, "loss": 0.6818271160125733, "time": 0.43977179527282717, "epoch": 11, "memory": 9465, "step": 124560} +{"lr": 0.0005050000000000001, "data_time": 0.0013326644897460938, "grad_norm": 26.933155679702757, "loss": 0.6730850815773011, "time": 0.42717363834381106, "epoch": 11, "memory": 9465, "step": 124660} +{"lr": 0.0005050000000000001, "data_time": 0.0012716054916381836, "grad_norm": 7.110695087909699, "loss": 0.6827796816825866, "time": 0.4461843252182007, "epoch": 11, "memory": 9465, "step": 124760} +{"lr": 0.0005050000000000001, "data_time": 0.0013957023620605469, "grad_norm": 116.47941331863403, "loss": 0.688453197479248, "time": 0.41263723373413086, "epoch": 11, "memory": 9465, "step": 124860} +{"lr": 0.0005050000000000001, "data_time": 0.0013117074966430664, "grad_norm": 36.254668053984645, "loss": 0.6842398405075073, "time": 0.4091363430023193, "epoch": 11, "memory": 9465, "step": 124960} +{"lr": 0.0005050000000000001, "data_time": 0.0012436628341674805, "grad_norm": 10.047633263468743, "loss": 0.6898526310920715, "time": 0.4380378246307373, "epoch": 11, "memory": 9465, "step": 125060} +{"lr": 0.0005050000000000001, "data_time": 0.001349020004272461, "grad_norm": 21.482556068897246, "loss": 0.687119847536087, "time": 0.4383219242095947, "epoch": 11, "memory": 9465, "step": 125160} +{"lr": 0.0005050000000000001, "data_time": 0.0012610197067260743, "grad_norm": 13.848260939121246, "loss": 0.6851519286632538, "time": 0.4230274438858032, "epoch": 11, "memory": 9465, "step": 125260} +{"lr": 0.0005050000000000001, "data_time": 0.0011785268783569337, "grad_norm": 3.980233770608902, "loss": 0.6885164201259613, "time": 0.4771219253540039, "epoch": 11, "memory": 9465, "step": 125360} +{"lr": 0.0005050000000000001, "data_time": 0.0012760639190673828, "grad_norm": 7.69204626083374, "loss": 0.6889068841934204, "time": 0.42093820571899415, "epoch": 11, "memory": 9465, "step": 125460} +{"lr": 0.0005050000000000001, "data_time": 0.001223134994506836, "grad_norm": 27.067822051048278, "loss": 0.6846552133560181, "time": 0.4102396726608276, "epoch": 11, "memory": 9465, "step": 125560} +{"lr": 0.0005050000000000001, "data_time": 0.0012709617614746094, "grad_norm": 3.8573687791824343, "loss": 0.6834540724754333, "time": 0.4050873279571533, "epoch": 11, "memory": 9465, "step": 125660} +{"lr": 0.0005050000000000001, "data_time": 0.0014838933944702148, "grad_norm": 22.484287905693055, "loss": 0.685315066576004, "time": 0.4652240753173828, "epoch": 11, "memory": 9465, "step": 125760} +{"lr": 0.0005050000000000001, "data_time": 0.0013320446014404297, "grad_norm": 68.12415418624877, "loss": 0.6865013837814331, "time": 0.6326909303665161, "epoch": 11, "memory": 9465, "step": 125860} +{"lr": 0.0005050000000000001, "data_time": 0.0013475656509399415, "grad_norm": 51.85662202835083, "loss": 0.6800556898117065, "time": 0.3989089488983154, "epoch": 11, "memory": 9465, "step": 125960} +{"lr": 0.0005050000000000001, "data_time": 0.0012872934341430664, "grad_norm": 22.468032431602477, "loss": 0.6815946578979493, "time": 0.43263342380523684, "epoch": 11, "memory": 9465, "step": 126060} +{"lr": 0.0005050000000000001, "data_time": 0.0012725591659545898, "grad_norm": 28.255167770385743, "loss": 0.6863328099250794, "time": 0.43357722759246825, "epoch": 11, "memory": 9465, "step": 126160} +{"lr": 0.0005050000000000001, "data_time": 0.0012325525283813476, "grad_norm": 69.66912631988525, "loss": 0.6869555890560151, "time": 0.404861307144165, "epoch": 11, "memory": 9465, "step": 126260} +{"lr": 0.0005050000000000001, "data_time": 0.0012387514114379882, "grad_norm": 31.074062204360963, "loss": 0.6884749054908752, "time": 0.4018389701843262, "epoch": 11, "memory": 9465, "step": 126360} +{"lr": 0.0005050000000000001, "data_time": 0.001231551170349121, "grad_norm": 167.52872245311738, "loss": 0.6871707201004028, "time": 0.4456325054168701, "epoch": 11, "memory": 9465, "step": 126460} +{"lr": 0.0005050000000000001, "data_time": 0.0012233257293701172, "grad_norm": 29.962598490715028, "loss": 0.6872786402702331, "time": 0.4013463258743286, "epoch": 11, "memory": 9465, "step": 126560} +{"lr": 0.0005050000000000001, "data_time": 0.007564949989318848, "grad_norm": 8.457032573223113, "loss": 0.6900364756584167, "time": 0.413921594619751, "epoch": 11, "memory": 9465, "step": 126660} +{"accuracy/top1": 100.0, "data_time": 0.31066016589894013, "time": 0.570034013074987, "step": 11} +{"lr": 0.000427564939805086, "data_time": 0.076735520362854, "grad_norm": 4.348672594130039, "loss": 0.6898119688034058, "time": 0.4361224412918091, "epoch": 12, "memory": 9465, "step": 126776} +{"lr": 0.000427564939805086, "data_time": 0.0012937068939208984, "grad_norm": 13.425345730781554, "loss": 0.6842809200286866, "time": 0.41864862442016604, "epoch": 12, "memory": 9465, "step": 126876} +{"lr": 0.000427564939805086, "data_time": 0.0013746976852416991, "grad_norm": 9.847851753234863, "loss": 0.6790269792079926, "time": 0.40267632007598875, "epoch": 12, "memory": 9465, "step": 126976} +{"lr": 0.000427564939805086, "data_time": 0.0011780977249145508, "grad_norm": 18.218738269805907, "loss": 0.683313512802124, "time": 0.4999894142150879, "epoch": 12, "memory": 9465, "step": 127076} +{"lr": 0.000427564939805086, "data_time": 0.0012818098068237305, "grad_norm": 13.688666260242462, "loss": 0.6830900192260743, "time": 0.41017515659332277, "epoch": 12, "memory": 9465, "step": 127176} +{"lr": 0.000427564939805086, "data_time": 0.0013007402420043945, "grad_norm": 74.42742584943771, "loss": 0.6809566378593445, "time": 0.4042999267578125, "epoch": 12, "memory": 9465, "step": 127276} +{"lr": 0.000427564939805086, "data_time": 0.0012687444686889648, "grad_norm": 188.30246211886407, "loss": 0.6753818213939666, "time": 0.4117432594299316, "epoch": 12, "memory": 9465, "step": 127376} +{"lr": 0.000427564939805086, "data_time": 0.0013460874557495116, "grad_norm": 11.578718554973602, "loss": 0.6836562275886535, "time": 0.4134324550628662, "epoch": 12, "memory": 9465, "step": 127476} +{"lr": 0.000427564939805086, "data_time": 0.0013631820678710938, "grad_norm": 1.4723381996154785, "loss": 0.6792226493358612, "time": 0.42078337669372556, "epoch": 12, "memory": 9465, "step": 127576} +{"lr": 0.000427564939805086, "data_time": 0.0012538909912109375, "grad_norm": 3.5019516080617903, "loss": 0.6809235095977784, "time": 0.43942258358001707, "epoch": 12, "memory": 9465, "step": 127676} +{"lr": 0.000427564939805086, "data_time": 0.0015204429626464843, "grad_norm": 10.551207971572875, "loss": 0.6818658530712127, "time": 0.46178693771362306, "epoch": 12, "memory": 9465, "step": 127776} +{"lr": 0.000427564939805086, "data_time": 0.001210331916809082, "grad_norm": 20.945073568820952, "loss": 0.6804120361804962, "time": 0.4522751808166504, "epoch": 12, "memory": 9465, "step": 127876} +{"lr": 0.000427564939805086, "data_time": 0.0012592315673828126, "grad_norm": 14.179097974300385, "loss": 0.6758145213127136, "time": 0.421085786819458, "epoch": 12, "memory": 9465, "step": 127976} +{"lr": 0.000427564939805086, "data_time": 0.0012780189514160155, "grad_norm": 3.3984122291207313, "loss": 0.6759363532066345, "time": 0.4091540813446045, "epoch": 12, "memory": 9465, "step": 128076} +{"lr": 0.000427564939805086, "data_time": 0.0012897968292236329, "grad_norm": 0.837803053855896, "loss": 0.6752262115478516, "time": 0.40413990020751955, "epoch": 12, "memory": 9465, "step": 128176} +{"lr": 0.000427564939805086, "data_time": 0.0013520956039428712, "grad_norm": 2.931898007541895, "loss": 0.6732583999633789, "time": 0.4167243242263794, "epoch": 12, "memory": 9465, "step": 128276} +{"lr": 0.000427564939805086, "data_time": 0.001275491714477539, "grad_norm": 3.4713230073451995, "loss": 0.6816255033016205, "time": 0.42007038593292234, "epoch": 12, "memory": 9465, "step": 128376} +{"lr": 0.000427564939805086, "data_time": 0.0012035369873046875, "grad_norm": 196.71823449134826, "loss": 0.6683463215827942, "time": 0.42772114276885986, "epoch": 12, "memory": 9465, "step": 128476} +{"lr": 0.000427564939805086, "data_time": 0.0013960838317871095, "grad_norm": 15.31438370347023, "loss": 0.6783772885799408, "time": 0.5493594169616699, "epoch": 12, "memory": 9465, "step": 128576} +{"lr": 0.000427564939805086, "data_time": 0.001378178596496582, "grad_norm": 1.7639539506286384, "loss": 0.6709182560443878, "time": 0.3983979940414429, "epoch": 12, "memory": 9465, "step": 128676} +{"lr": 0.000427564939805086, "data_time": 0.0012935400009155273, "grad_norm": 3.431198874115944, "loss": 0.6762523114681244, "time": 0.4447017669677734, "epoch": 12, "memory": 9465, "step": 128776} +{"lr": 0.000427564939805086, "data_time": 0.001427602767944336, "grad_norm": 2.770304651558399, "loss": 0.6715142607688904, "time": 0.4263080358505249, "epoch": 12, "memory": 9465, "step": 128876} +{"lr": 0.000427564939805086, "data_time": 0.0014818668365478515, "grad_norm": 3.470287262648344, "loss": 0.6753794312477112, "time": 0.41139366626739504, "epoch": 12, "memory": 9465, "step": 128976} +{"lr": 0.000427564939805086, "data_time": 0.0015148162841796876, "grad_norm": 1.2597437039017678, "loss": 0.6691596806049347, "time": 0.40613598823547364, "epoch": 12, "memory": 9465, "step": 129076} +{"lr": 0.000427564939805086, "data_time": 0.001400899887084961, "grad_norm": 2.1850260317325594, "loss": 0.6866812586784363, "time": 0.42925822734832764, "epoch": 12, "memory": 9465, "step": 129176} +{"lr": 0.000427564939805086, "data_time": 0.0013475418090820312, "grad_norm": 0.8963435543701053, "loss": 0.6871735692024231, "time": 0.4090800523757935, "epoch": 12, "memory": 9465, "step": 129276} +{"lr": 0.000427564939805086, "data_time": 0.001323866844177246, "grad_norm": 2.536113853752613, "loss": 0.6818703353404999, "time": 0.41405088901519777, "epoch": 12, "memory": 9465, "step": 129376} +{"lr": 0.000427564939805086, "data_time": 0.0014016866683959962, "grad_norm": 0.030517227482050657, "loss": 0.6869018912315369, "time": 0.44098801612854005, "epoch": 12, "memory": 9465, "step": 129476} +{"lr": 0.000427564939805086, "data_time": 0.0012927532196044921, "grad_norm": 0.036532994173467156, "loss": 0.6903063833713532, "time": 0.41088435649871824, "epoch": 12, "memory": 9465, "step": 129576} +{"lr": 0.000427564939805086, "data_time": 0.0013560056686401367, "grad_norm": 0.0251450982876122, "loss": 0.6881662428379058, "time": 0.4688152074813843, "epoch": 12, "memory": 9465, "step": 129676} +{"lr": 0.000427564939805086, "data_time": 0.0014296770095825195, "grad_norm": 0.01814848983194679, "loss": 0.6847158551216126, "time": 0.43398659229278563, "epoch": 12, "memory": 9465, "step": 129776} +{"lr": 0.000427564939805086, "data_time": 0.0012911319732666015, "grad_norm": 0.03640967914834618, "loss": 0.6900094389915467, "time": 0.4438145637512207, "epoch": 12, "memory": 9465, "step": 129876} +{"lr": 0.000427564939805086, "data_time": 0.0015025138854980469, "grad_norm": 0.023893103143200277, "loss": 0.6896107614040374, "time": 0.4201754808425903, "epoch": 12, "memory": 9465, "step": 129976} +{"lr": 0.000427564939805086, "data_time": 0.0013779640197753907, "grad_norm": 0.020082084462046623, "loss": 0.6873001635074616, "time": 0.4413077116012573, "epoch": 12, "memory": 9465, "step": 130076} +{"lr": 0.000427564939805086, "data_time": 0.0012671947479248047, "grad_norm": 0.021519444510340692, "loss": 0.687029379606247, "time": 0.4054450750350952, "epoch": 12, "memory": 9465, "step": 130176} +{"lr": 0.000427564939805086, "data_time": 0.0014216184616088867, "grad_norm": 0.05771431168541312, "loss": 0.6854559421539307, "time": 0.41077234745025637, "epoch": 12, "memory": 9465, "step": 130276} +{"lr": 0.000427564939805086, "data_time": 0.0012619733810424806, "grad_norm": 0.022173736547119915, "loss": 0.6871199786663056, "time": 0.4157752752304077, "epoch": 12, "memory": 9465, "step": 130376} +{"lr": 0.000427564939805086, "data_time": 0.0013138771057128907, "grad_norm": 0.036936514638364316, "loss": 0.6890743017196655, "time": 0.45462851524353026, "epoch": 12, "memory": 9465, "step": 130476} +{"lr": 0.000427564939805086, "data_time": 0.0013151168823242188, "grad_norm": 0.027146380953490735, "loss": 0.6815440595149994, "time": 0.4137685775756836, "epoch": 12, "memory": 9465, "step": 130576} +{"lr": 0.000427564939805086, "data_time": 0.0012878179550170898, "grad_norm": 0.027129650907590987, "loss": 0.684997022151947, "time": 0.3863675594329834, "epoch": 12, "memory": 9465, "step": 130676} +{"lr": 0.000427564939805086, "data_time": 0.0016545772552490235, "grad_norm": 0.03196599064394832, "loss": 0.6857732832431793, "time": 0.4002965450286865, "epoch": 12, "memory": 9465, "step": 130776} +{"lr": 0.000427564939805086, "data_time": 0.0015331506729125977, "grad_norm": 0.025930631812661885, "loss": 0.6855455875396729, "time": 0.452677583694458, "epoch": 12, "memory": 9465, "step": 130876} +{"lr": 0.000427564939805086, "data_time": 0.001256251335144043, "grad_norm": 0.02131669588852674, "loss": 0.6853181421756744, "time": 0.44700634479522705, "epoch": 12, "memory": 9465, "step": 130976} +{"lr": 0.000427564939805086, "data_time": 0.001496577262878418, "grad_norm": 0.021841109939850868, "loss": 0.6890956461429596, "time": 0.44107742309570314, "epoch": 12, "memory": 9465, "step": 131076} +{"lr": 0.000427564939805086, "data_time": 0.0015111923217773437, "grad_norm": 0.01930496752029285, "loss": 0.6871892094612122, "time": 0.40665972232818604, "epoch": 12, "memory": 9465, "step": 131176} +{"lr": 0.000427564939805086, "data_time": 0.0014603376388549806, "grad_norm": 0.02447275447193533, "loss": 0.6822593510150909, "time": 0.4118276834487915, "epoch": 12, "memory": 9465, "step": 131276} +{"lr": 0.000427564939805086, "data_time": 0.0013532400131225585, "grad_norm": 0.02468832125887275, "loss": 0.6912485063076019, "time": 0.37818121910095215, "epoch": 12, "memory": 9465, "step": 131376} +{"lr": 0.000427564939805086, "data_time": 0.0013502836227416992, "grad_norm": 0.023085615155287086, "loss": 0.6889755249023437, "time": 0.4210802555084229, "epoch": 12, "memory": 9465, "step": 131476} +{"lr": 0.000427564939805086, "data_time": 0.0013721466064453125, "grad_norm": 0.019682963425293565, "loss": 0.6871603846549987, "time": 0.44051265716552734, "epoch": 12, "memory": 9465, "step": 131576} +{"lr": 0.000427564939805086, "data_time": 0.0013153076171875, "grad_norm": 0.022028203285299242, "loss": 0.6892701923847199, "time": 0.4130614995956421, "epoch": 12, "memory": 9465, "step": 131676} +{"lr": 0.000427564939805086, "data_time": 0.001412653923034668, "grad_norm": 0.017201270908117294, "loss": 0.6846320748329162, "time": 0.45083136558532716, "epoch": 12, "memory": 9465, "step": 131776} +{"lr": 0.000427564939805086, "data_time": 0.0012716531753540039, "grad_norm": 0.018730456568300725, "loss": 0.6891506969928741, "time": 0.4022256851196289, "epoch": 12, "memory": 9465, "step": 131876} +{"lr": 0.000427564939805086, "data_time": 0.001374363899230957, "grad_norm": 0.025354755041189492, "loss": 0.6907829225063324, "time": 0.3865982532501221, "epoch": 12, "memory": 9465, "step": 131976} +{"lr": 0.000427564939805086, "data_time": 0.0013847589492797852, "grad_norm": 0.01840975587256253, "loss": 0.6831769108772278, "time": 0.42206737995147703, "epoch": 12, "memory": 9465, "step": 132076} +{"lr": 0.000427564939805086, "data_time": 0.0016530275344848633, "grad_norm": 0.019254678534343837, "loss": 0.689174884557724, "time": 0.41932222843170164, "epoch": 12, "memory": 9465, "step": 132176} +{"lr": 0.000427564939805086, "data_time": 0.001480698585510254, "grad_norm": 0.021869329328183085, "loss": 0.6880807399749755, "time": 0.4040341854095459, "epoch": 12, "memory": 9465, "step": 132276} +{"lr": 0.000427564939805086, "data_time": 0.0014929771423339844, "grad_norm": 0.020475224265828727, "loss": 0.6849862813949585, "time": 0.41760499477386476, "epoch": 12, "memory": 9465, "step": 132376} +{"lr": 0.000427564939805086, "data_time": 0.0012695789337158203, "grad_norm": 0.01955119805643335, "loss": 0.6870149850845337, "time": 0.41249804496765136, "epoch": 12, "memory": 9465, "step": 132476} +{"lr": 0.000427564939805086, "data_time": 0.0013772010803222655, "grad_norm": 0.019221801293315365, "loss": 0.6884264945983887, "time": 0.44633028507232664, "epoch": 12, "memory": 9465, "step": 132576} +{"lr": 0.000427564939805086, "data_time": 0.0013963699340820313, "grad_norm": 0.02509285109117627, "loss": 0.6884343028068542, "time": 0.4383856296539307, "epoch": 12, "memory": 9465, "step": 132676} +{"lr": 0.000427564939805086, "data_time": 0.0011419057846069336, "grad_norm": 0.01643194416537881, "loss": 0.6855869948863983, "time": 0.42352981567382814, "epoch": 12, "memory": 9465, "step": 132776} +{"lr": 0.000427564939805086, "data_time": 0.0014334678649902343, "grad_norm": 0.02347443662583828, "loss": 0.6875027775764465, "time": 0.40734381675720216, "epoch": 12, "memory": 9465, "step": 132876} +{"lr": 0.000427564939805086, "data_time": 0.0013040781021118164, "grad_norm": 0.02326798252761364, "loss": 0.6820044338703155, "time": 0.4270260572433472, "epoch": 12, "memory": 9465, "step": 132976} +{"lr": 0.000427564939805086, "data_time": 0.0012149810791015625, "grad_norm": 0.024808078981004657, "loss": 0.6920408725738525, "time": 0.4212906837463379, "epoch": 12, "memory": 9465, "step": 133076} +{"lr": 0.000427564939805086, "data_time": 0.001304292678833008, "grad_norm": 0.03381731836125255, "loss": 0.6895127177238465, "time": 0.42964370250701905, "epoch": 12, "memory": 9465, "step": 133176} +{"lr": 0.000427564939805086, "data_time": 0.001300954818725586, "grad_norm": 0.019100240874104203, "loss": 0.6905601620674133, "time": 0.4016817569732666, "epoch": 12, "memory": 9465, "step": 133276} +{"lr": 0.000427564939805086, "data_time": 0.001690506935119629, "grad_norm": 0.019882703525945546, "loss": 0.6893523037433624, "time": 0.4288515567779541, "epoch": 12, "memory": 9465, "step": 133376} +{"lr": 0.000427564939805086, "data_time": 0.0013581037521362305, "grad_norm": 0.024326645722612738, "loss": 0.6894127905368805, "time": 0.4464055299758911, "epoch": 12, "memory": 9465, "step": 133476} +{"lr": 0.000427564939805086, "data_time": 0.0013773202896118163, "grad_norm": 0.013203220115974546, "loss": 0.6842819094657898, "time": 0.45641722679138186, "epoch": 12, "memory": 9465, "step": 133576} +{"lr": 0.000427564939805086, "data_time": 0.00134124755859375, "grad_norm": 0.022629460087046027, "loss": 0.6804419755935669, "time": 0.42402608394622804, "epoch": 12, "memory": 9465, "step": 133676} +{"lr": 0.000427564939805086, "data_time": 0.0013596773147583007, "grad_norm": 0.022043393761850893, "loss": 0.685667097568512, "time": 0.40441606044769285, "epoch": 12, "memory": 9465, "step": 133776} +{"lr": 0.000427564939805086, "data_time": 0.0015870332717895508, "grad_norm": 0.018440839857794343, "loss": 0.686989140510559, "time": 0.41110920906066895, "epoch": 12, "memory": 9465, "step": 133876} +{"lr": 0.000427564939805086, "data_time": 0.0013441801071166991, "grad_norm": 0.030286093521863223, "loss": 0.6861549079418182, "time": 0.40697195529937746, "epoch": 12, "memory": 9465, "step": 133976} +{"lr": 0.000427564939805086, "data_time": 0.001190185546875, "grad_norm": 0.024573275772854685, "loss": 0.6904351890087128, "time": 0.3955228328704834, "epoch": 12, "memory": 9465, "step": 134076} +{"lr": 0.000427564939805086, "data_time": 0.001268482208251953, "grad_norm": 0.02895478003192693, "loss": 0.6836984992027283, "time": 0.40450875759124755, "epoch": 12, "memory": 9465, "step": 134176} +{"lr": 0.000427564939805086, "data_time": 0.0013086318969726563, "grad_norm": 0.020092893857508898, "loss": 0.6867953479290009, "time": 0.4229351758956909, "epoch": 12, "memory": 9465, "step": 134276} +{"lr": 0.000427564939805086, "data_time": 0.0014414548873901366, "grad_norm": 0.02477827437687665, "loss": 0.6843395233154297, "time": 0.4216409921646118, "epoch": 12, "memory": 9465, "step": 134376} +{"lr": 0.000427564939805086, "data_time": 0.0013153553009033203, "grad_norm": 0.01508900262415409, "loss": 0.68714160323143, "time": 0.4240150213241577, "epoch": 12, "memory": 9465, "step": 134476} +{"lr": 0.000427564939805086, "data_time": 0.001452493667602539, "grad_norm": 0.02276517413556576, "loss": 0.6918884932994842, "time": 0.4838341951370239, "epoch": 12, "memory": 9465, "step": 134576} +{"lr": 0.000427564939805086, "data_time": 0.0012554645538330079, "grad_norm": 0.029659863817505537, "loss": 0.6815708875656128, "time": 0.433766770362854, "epoch": 12, "memory": 9465, "step": 134676} +{"lr": 0.000427564939805086, "data_time": 0.0012537240982055664, "grad_norm": 0.019865823443979025, "loss": 0.6860040843486785, "time": 0.40606448650360105, "epoch": 12, "memory": 9465, "step": 134776} +{"lr": 0.000427564939805086, "data_time": 0.0012784957885742187, "grad_norm": 0.019153623550664634, "loss": 0.6851597964763642, "time": 0.4024749040603638, "epoch": 12, "memory": 9465, "step": 134876} +{"lr": 0.000427564939805086, "data_time": 0.001424694061279297, "grad_norm": 0.02400211044587195, "loss": 0.6818427979946137, "time": 0.4202421188354492, "epoch": 12, "memory": 9465, "step": 134976} +{"lr": 0.000427564939805086, "data_time": 0.0014413118362426758, "grad_norm": 0.028626030031591653, "loss": 0.6868133068084716, "time": 0.4084906816482544, "epoch": 12, "memory": 9465, "step": 135076} +{"lr": 0.000427564939805086, "data_time": 0.001470041275024414, "grad_norm": 0.01666400389513001, "loss": 0.6866598844528198, "time": 0.40768468379974365, "epoch": 12, "memory": 9465, "step": 135176} +{"lr": 0.000427564939805086, "data_time": 0.0014740705490112304, "grad_norm": 0.015819929726421832, "loss": 0.6859961926937104, "time": 0.4268024444580078, "epoch": 12, "memory": 9465, "step": 135276} +{"lr": 0.000427564939805086, "data_time": 0.0012934446334838868, "grad_norm": 0.02918328596279025, "loss": 0.6861585319042206, "time": 0.40381383895874023, "epoch": 12, "memory": 9465, "step": 135376} +{"lr": 0.000427564939805086, "data_time": 0.0014419317245483398, "grad_norm": 0.020316384651232512, "loss": 0.6876485645771027, "time": 0.41089928150177, "epoch": 12, "memory": 9465, "step": 135476} +{"lr": 0.000427564939805086, "data_time": 0.0013978719711303712, "grad_norm": 0.023399069788865745, "loss": 0.6904264211654663, "time": 0.4105861186981201, "epoch": 12, "memory": 9465, "step": 135576} +{"lr": 0.000427564939805086, "data_time": 0.0014662027359008789, "grad_norm": 0.018001660704612732, "loss": 0.6850806176662445, "time": 0.4113790512084961, "epoch": 12, "memory": 9465, "step": 135676} +{"lr": 0.000427564939805086, "data_time": 0.001482105255126953, "grad_norm": 0.0262545530218631, "loss": 0.6874183356761933, "time": 0.42870097160339354, "epoch": 12, "memory": 9465, "step": 135776} +{"lr": 0.000427564939805086, "data_time": 0.001321554183959961, "grad_norm": 0.01224442480597645, "loss": 0.6862155795097351, "time": 0.4189307689666748, "epoch": 12, "memory": 9465, "step": 135876} +{"lr": 0.000427564939805086, "data_time": 0.0011930227279663085, "grad_norm": 0.026414019777439533, "loss": 0.6858326196670532, "time": 0.4315566301345825, "epoch": 12, "memory": 9465, "step": 135976} +{"lr": 0.000427564939805086, "data_time": 0.001241016387939453, "grad_norm": 0.024612182425335048, "loss": 0.6861414194107056, "time": 0.43127634525299074, "epoch": 12, "memory": 9465, "step": 136076} +{"lr": 0.000427564939805086, "data_time": 0.0014866113662719727, "grad_norm": 0.018336525186896324, "loss": 0.6848597586154938, "time": 0.4810450792312622, "epoch": 12, "memory": 9465, "step": 136176} +{"lr": 0.000427564939805086, "data_time": 0.0012953758239746093, "grad_norm": 0.019203244987875224, "loss": 0.690115088224411, "time": 0.4266946792602539, "epoch": 12, "memory": 9465, "step": 136276} +{"lr": 0.000427564939805086, "data_time": 0.0016733407974243164, "grad_norm": 0.0174126991070807, "loss": 0.69014573097229, "time": 0.4433808088302612, "epoch": 12, "memory": 9465, "step": 136376} +{"lr": 0.000427564939805086, "data_time": 0.0014173269271850586, "grad_norm": 0.020531594473868608, "loss": 0.6884080231189728, "time": 0.44682042598724364, "epoch": 12, "memory": 9465, "step": 136476} +{"lr": 0.000427564939805086, "data_time": 0.00130157470703125, "grad_norm": 0.018273185286670924, "loss": 0.6853847205638885, "time": 0.39111766815185545, "epoch": 12, "memory": 9465, "step": 136576} +{"lr": 0.000427564939805086, "data_time": 0.0013633966445922852, "grad_norm": 0.012315124575980007, "loss": 0.6867692649364472, "time": 0.4591775894165039, "epoch": 12, "memory": 9465, "step": 136676} +{"lr": 0.000427564939805086, "data_time": 0.001344442367553711, "grad_norm": 0.021995778544805944, "loss": 0.6875159919261933, "time": 0.4549398899078369, "epoch": 12, "memory": 9465, "step": 136776} +{"lr": 0.000427564939805086, "data_time": 0.001265406608581543, "grad_norm": 0.020000772573985158, "loss": 0.6919745922088623, "time": 0.45438556671142577, "epoch": 12, "memory": 9465, "step": 136876} +{"lr": 0.000427564939805086, "data_time": 0.001387333869934082, "grad_norm": 0.022508371085859836, "loss": 0.6888543665409088, "time": 0.43606531620025635, "epoch": 12, "memory": 9465, "step": 136976} +{"lr": 0.000427564939805086, "data_time": 0.0012754440307617188, "grad_norm": 0.024935787287540735, "loss": 0.6890964150428772, "time": 0.41189501285552976, "epoch": 12, "memory": 9465, "step": 137076} +{"lr": 0.000427564939805086, "data_time": 0.0020142078399658205, "grad_norm": 0.020476224087178706, "loss": 0.6907122969627381, "time": 0.3999741792678833, "epoch": 12, "memory": 9465, "step": 137176} +{"lr": 0.000427564939805086, "data_time": 0.0014447450637817382, "grad_norm": 0.02158180233091116, "loss": 0.6905366480350494, "time": 0.43572728633880614, "epoch": 12, "memory": 9465, "step": 137276} +{"lr": 0.000427564939805086, "data_time": 0.001335906982421875, "grad_norm": 0.022731278091669083, "loss": 0.6856934368610382, "time": 0.43190479278564453, "epoch": 12, "memory": 9465, "step": 137376} +{"lr": 0.000427564939805086, "data_time": 0.0013525724411010743, "grad_norm": 0.025684100296348333, "loss": 0.6849841117858887, "time": 0.43099782466888426, "epoch": 12, "memory": 9465, "step": 137476} +{"lr": 0.000427564939805086, "data_time": 0.0012893199920654297, "grad_norm": 0.026394939993042498, "loss": 0.6828485250473022, "time": 0.43994388580322263, "epoch": 12, "memory": 9465, "step": 137576} +{"lr": 0.000427564939805086, "data_time": 0.001317000389099121, "grad_norm": 0.020767204923322423, "loss": 0.6891070485115052, "time": 0.41753444671630857, "epoch": 12, "memory": 9465, "step": 137676} +{"lr": 0.000427564939805086, "data_time": 0.0013175249099731446, "grad_norm": 0.02978737003868446, "loss": 0.6874478399753571, "time": 0.41090338230133056, "epoch": 12, "memory": 9465, "step": 137776} +{"lr": 0.000427564939805086, "data_time": 0.0013688087463378906, "grad_norm": 0.020555397635325788, "loss": 0.6828809797763824, "time": 0.4285306215286255, "epoch": 12, "memory": 9465, "step": 137876} +{"lr": 0.000427564939805086, "data_time": 0.0012490272521972657, "grad_norm": 0.03204272212460637, "loss": 0.6877374231815339, "time": 0.4435636758804321, "epoch": 12, "memory": 9465, "step": 137976} +{"lr": 0.000427564939805086, "data_time": 0.0023162841796875, "grad_norm": 0.037516922818031165, "loss": 0.6898782253265381, "time": 0.3867438077926636, "epoch": 12, "memory": 9465, "step": 138076} +{"lr": 0.000427564939805086, "data_time": 0.00787358283996582, "grad_norm": 0.02366722277365625, "loss": 0.6853773534297943, "time": 0.4042431116104126, "epoch": 12, "memory": 9465, "step": 138176} +{"accuracy/top1": 100.0, "data_time": 0.29801522984224205, "time": 0.5517384725458482, "step": 12} +{"lr": 0.0003520365877844012, "data_time": 0.1872790575027466, "grad_norm": 0.017510948225390167, "loss": 0.6881148517131805, "time": 0.468621039390564, "epoch": 13, "memory": 9465, "step": 138292} +{"lr": 0.0003520365877844012, "data_time": 0.0012444019317626952, "grad_norm": 0.026656911056488753, "loss": 0.6868264317512512, "time": 0.4445474863052368, "epoch": 13, "memory": 9465, "step": 138392} +{"lr": 0.0003520365877844012, "data_time": 0.0011973381042480469, "grad_norm": 0.027554817497730255, "loss": 0.6890848696231842, "time": 0.4294422626495361, "epoch": 13, "memory": 9465, "step": 138492} +{"lr": 0.0003520365877844012, "data_time": 0.0013140439987182617, "grad_norm": 0.03655003334861249, "loss": 0.6861842572689056, "time": 0.43264293670654297, "epoch": 13, "memory": 9465, "step": 138592} +{"lr": 0.0003520365877844012, "data_time": 0.0012440443038940429, "grad_norm": 0.02566453628242016, "loss": 0.6926171720027924, "time": 0.45820322036743166, "epoch": 13, "memory": 9465, "step": 138692} +{"lr": 0.0003520365877844012, "data_time": 0.00138397216796875, "grad_norm": 0.022582877962850034, "loss": 0.6888538777828217, "time": 0.4243925094604492, "epoch": 13, "memory": 9465, "step": 138792} +{"lr": 0.0003520365877844012, "data_time": 0.0013495683670043945, "grad_norm": 0.021802978264167903, "loss": 0.6897751808166503, "time": 0.40727231502532957, "epoch": 13, "memory": 9465, "step": 138892} +{"lr": 0.0003520365877844012, "data_time": 0.0013855695724487305, "grad_norm": 0.013806198944803328, "loss": 0.6899027168750763, "time": 0.41996567249298095, "epoch": 13, "memory": 9465, "step": 138992} +{"lr": 0.0003520365877844012, "data_time": 0.0014684677124023437, "grad_norm": 0.01774993594735861, "loss": 0.686185497045517, "time": 0.43367202281951905, "epoch": 13, "memory": 9465, "step": 139092} +{"lr": 0.0003520365877844012, "data_time": 0.0012041568756103516, "grad_norm": 0.023941261286381632, "loss": 0.6860288798809051, "time": 0.41524133682250974, "epoch": 13, "memory": 9465, "step": 139192} +{"lr": 0.0003520365877844012, "data_time": 0.0012471675872802734, "grad_norm": 0.018963639996945857, "loss": 0.6848430573940277, "time": 0.41869263648986815, "epoch": 13, "memory": 9465, "step": 139292} +{"lr": 0.0003520365877844012, "data_time": 0.0013098001480102539, "grad_norm": 0.016764594428241252, "loss": 0.6842615306377411, "time": 0.44188990592956545, "epoch": 13, "memory": 9465, "step": 139392} +{"lr": 0.0003520365877844012, "data_time": 0.0012956619262695312, "grad_norm": 0.0158146571891848, "loss": 0.6855414569377899, "time": 0.45032515525817873, "epoch": 13, "memory": 9465, "step": 139492} +{"lr": 0.0003520365877844012, "data_time": 0.0013699293136596679, "grad_norm": 0.030753080965951085, "loss": 0.6857570528984069, "time": 0.38656091690063477, "epoch": 13, "memory": 9465, "step": 139592} +{"lr": 0.0003520365877844012, "data_time": 0.0014206171035766602, "grad_norm": 0.013240466848947108, "loss": 0.686662209033966, "time": 0.44356157779693606, "epoch": 13, "memory": 9465, "step": 139692} +{"lr": 0.0003520365877844012, "data_time": 0.0012423038482666016, "grad_norm": 0.016023906273767353, "loss": 0.6891343235969544, "time": 0.4172191143035889, "epoch": 13, "memory": 9465, "step": 139792} +{"lr": 0.0003520365877844012, "data_time": 0.0013096332550048828, "grad_norm": 0.022254918655380606, "loss": 0.688138771057129, "time": 0.424369215965271, "epoch": 13, "memory": 9465, "step": 139892} +{"lr": 0.0003520365877844012, "data_time": 0.0012717962265014649, "grad_norm": 0.032299743383191526, "loss": 0.6855127274990082, "time": 0.4386639356613159, "epoch": 13, "memory": 9465, "step": 139992} +{"lr": 0.0003520365877844012, "data_time": 0.0013203382492065429, "grad_norm": 0.02728343503549695, "loss": 0.6826135277748108, "time": 0.42070319652557375, "epoch": 13, "memory": 9465, "step": 140092} +{"lr": 0.0003520365877844012, "data_time": 0.001439213752746582, "grad_norm": 0.022068626224063338, "loss": 0.6877608120441436, "time": 0.4150163412094116, "epoch": 13, "memory": 9465, "step": 140192} +{"lr": 0.0003520365877844012, "data_time": 0.0013849496841430663, "grad_norm": 0.029572939220815896, "loss": 0.683891487121582, "time": 0.39600937366485595, "epoch": 13, "memory": 9465, "step": 140292} +{"lr": 0.0003520365877844012, "data_time": 0.0013171911239624023, "grad_norm": 0.026504780864343047, "loss": 0.6791876435279847, "time": 0.4116565704345703, "epoch": 13, "memory": 9465, "step": 140392} +{"lr": 0.0003520365877844012, "data_time": 0.0014576911926269531, "grad_norm": 0.02092731527518481, "loss": 0.686355185508728, "time": 0.43780813217163084, "epoch": 13, "memory": 9465, "step": 140492} +{"lr": 0.0003520365877844012, "data_time": 0.0013976812362670899, "grad_norm": 0.034247031714767215, "loss": 0.6849726617336274, "time": 0.43756694793701173, "epoch": 13, "memory": 9465, "step": 140592} +{"lr": 0.0003520365877844012, "data_time": 0.0013062715530395507, "grad_norm": 0.018480710778385402, "loss": 0.6864704310894012, "time": 0.40336544513702394, "epoch": 13, "memory": 9465, "step": 140692} +{"lr": 0.0003520365877844012, "data_time": 0.001402902603149414, "grad_norm": 0.023231533565558495, "loss": 0.6867999255657196, "time": 0.42250733375549315, "epoch": 13, "memory": 9465, "step": 140792} +{"lr": 0.0003520365877844012, "data_time": 0.0013544321060180663, "grad_norm": 0.019997408241033555, "loss": 0.6861769199371338, "time": 0.43517193794250486, "epoch": 13, "memory": 9465, "step": 140892} +{"lr": 0.0003520365877844012, "data_time": 0.0013731718063354492, "grad_norm": 0.022900594654493032, "loss": 0.6920109510421752, "time": 0.43185617923736574, "epoch": 13, "memory": 9465, "step": 140992} +{"lr": 0.0003520365877844012, "data_time": 0.0015815258026123046, "grad_norm": 0.024012094689533116, "loss": 0.6870336174964905, "time": 0.4229752063751221, "epoch": 13, "memory": 9465, "step": 141092} +{"lr": 0.0003520365877844012, "data_time": 0.0013350725173950195, "grad_norm": 0.02456324272789061, "loss": 0.6892096519470214, "time": 0.4228933334350586, "epoch": 13, "memory": 9465, "step": 141192} +{"lr": 0.0003520365877844012, "data_time": 0.0013721227645874024, "grad_norm": 0.023345749638974667, "loss": 0.6876180946826935, "time": 0.43931851387023924, "epoch": 13, "memory": 9465, "step": 141292} +{"lr": 0.0003520365877844012, "data_time": 0.001420307159423828, "grad_norm": 0.012605708825867623, "loss": 0.6855099916458129, "time": 0.42755842208862305, "epoch": 13, "memory": 9465, "step": 141392} +{"lr": 0.0003520365877844012, "data_time": 0.0015364885330200195, "grad_norm": 0.01526969433762133, "loss": 0.6880516052246094, "time": 0.41323699951171877, "epoch": 13, "memory": 9465, "step": 141492} +{"lr": 0.0003520365877844012, "data_time": 0.001304292678833008, "grad_norm": 0.01929947747848928, "loss": 0.6860296785831451, "time": 0.3967778444290161, "epoch": 13, "memory": 9465, "step": 141592} +{"lr": 0.0003520365877844012, "data_time": 0.0014894247055053712, "grad_norm": 0.018084618775174022, "loss": 0.688594925403595, "time": 0.4250460147857666, "epoch": 13, "memory": 9465, "step": 141692} +{"lr": 0.0003520365877844012, "data_time": 0.0013441801071166991, "grad_norm": 0.022172150015830994, "loss": 0.6866418898105622, "time": 0.4182627201080322, "epoch": 13, "memory": 9465, "step": 141792} +{"lr": 0.0003520365877844012, "data_time": 0.0013438940048217773, "grad_norm": 0.02240907186642289, "loss": 0.6872819364070892, "time": 0.4504395008087158, "epoch": 13, "memory": 9465, "step": 141892} +{"lr": 0.0003520365877844012, "data_time": 0.001279449462890625, "grad_norm": 0.01939082032768056, "loss": 0.6878681242465973, "time": 0.4323746919631958, "epoch": 13, "memory": 9465, "step": 141992} +{"lr": 0.0003520365877844012, "data_time": 0.001287698745727539, "grad_norm": 0.016904144175350667, "loss": 0.6885225951671601, "time": 0.4312117338180542, "epoch": 13, "memory": 9465, "step": 142092} +{"lr": 0.0003520365877844012, "data_time": 0.0014436483383178712, "grad_norm": 0.019593195826746523, "loss": 0.6895209610462188, "time": 0.4291876554489136, "epoch": 13, "memory": 9465, "step": 142192} +{"lr": 0.0003520365877844012, "data_time": 0.0013062715530395507, "grad_norm": 0.022013190877623857, "loss": 0.6822513163089752, "time": 0.45579018592834475, "epoch": 13, "memory": 9465, "step": 142292} +{"lr": 0.0003520365877844012, "data_time": 0.0014356851577758789, "grad_norm": 0.028816881170496345, "loss": 0.68393874168396, "time": 0.4303570032119751, "epoch": 13, "memory": 9465, "step": 142392} +{"lr": 0.0003520365877844012, "data_time": 0.0013211488723754883, "grad_norm": 0.030636927112936974, "loss": 0.6899877965450287, "time": 0.45420079231262206, "epoch": 13, "memory": 9465, "step": 142492} +{"lr": 0.0003520365877844012, "data_time": 0.0012770652770996093, "grad_norm": 0.017876827262807638, "loss": 0.6853049278259278, "time": 0.42228124141693113, "epoch": 13, "memory": 9465, "step": 142592} +{"lr": 0.0003520365877844012, "data_time": 0.0013705730438232423, "grad_norm": 0.00992819966049865, "loss": 0.6896498501300812, "time": 0.43505353927612306, "epoch": 13, "memory": 9465, "step": 142692} +{"lr": 0.0003520365877844012, "data_time": 0.0013044357299804687, "grad_norm": 0.033733045449480414, "loss": 0.6935043334960938, "time": 0.45402557849884034, "epoch": 13, "memory": 9465, "step": 142792} +{"lr": 0.0003520365877844012, "data_time": 0.001292109489440918, "grad_norm": 0.027146546076983212, "loss": 0.6876082599163056, "time": 0.4531333208084106, "epoch": 13, "memory": 9465, "step": 142892} +{"lr": 0.0003520365877844012, "data_time": 0.001516413688659668, "grad_norm": 0.01991870717611164, "loss": 0.6859760463237763, "time": 0.4385573625564575, "epoch": 13, "memory": 9465, "step": 142992} +{"lr": 0.0003520365877844012, "data_time": 0.0012458324432373046, "grad_norm": 0.0336504980456084, "loss": 0.6895107924938202, "time": 0.43419690132141114, "epoch": 13, "memory": 9465, "step": 143092} +{"lr": 0.0003520365877844012, "data_time": 0.0012791872024536133, "grad_norm": 0.0266926902346313, "loss": 0.6909848272800445, "time": 0.4190091133117676, "epoch": 13, "memory": 9465, "step": 143192} +{"lr": 0.0003520365877844012, "data_time": 0.0013389825820922852, "grad_norm": 0.01625212614890188, "loss": 0.683299720287323, "time": 0.4054360866546631, "epoch": 13, "memory": 9465, "step": 143292} +{"lr": 0.0003520365877844012, "data_time": 0.0014005184173583984, "grad_norm": 0.02609210801310837, "loss": 0.6856689631938935, "time": 0.4396503925323486, "epoch": 13, "memory": 9465, "step": 143392} +{"lr": 0.0003520365877844012, "data_time": 0.0013449907302856446, "grad_norm": 0.02831558436155319, "loss": 0.687919282913208, "time": 0.425563645362854, "epoch": 13, "memory": 9465, "step": 143492} +{"lr": 0.0003520365877844012, "data_time": 0.0013115882873535156, "grad_norm": 0.019528388057369738, "loss": 0.683935409784317, "time": 0.44516539573669434, "epoch": 13, "memory": 9465, "step": 143592} +{"lr": 0.0003520365877844012, "data_time": 0.001217031478881836, "grad_norm": 0.02119381525553763, "loss": 0.684372067451477, "time": 0.4636390686035156, "epoch": 13, "memory": 9465, "step": 143692} +{"lr": 0.0003520365877844012, "data_time": 0.0012293815612792968, "grad_norm": 0.02362422493752092, "loss": 0.6869829535484314, "time": 0.43630223274230956, "epoch": 13, "memory": 9465, "step": 143792} +{"lr": 0.0003520365877844012, "data_time": 0.0013032674789428711, "grad_norm": 0.028147122485097497, "loss": 0.6863200783729553, "time": 0.41666560173034667, "epoch": 13, "memory": 9465, "step": 143892} +{"lr": 0.0003520365877844012, "data_time": 0.0014888525009155273, "grad_norm": 0.017308444855734705, "loss": 0.6914076685905457, "time": 0.42134904861450195, "epoch": 13, "memory": 9465, "step": 143992} +{"lr": 0.0003520365877844012, "data_time": 0.002397298812866211, "grad_norm": 0.016407575085759164, "loss": 0.6847853183746337, "time": 0.4567969083786011, "epoch": 13, "memory": 9465, "step": 144092} +{"lr": 0.0003520365877844012, "data_time": 0.001467442512512207, "grad_norm": 0.021011793659999965, "loss": 0.6861694097518921, "time": 0.4472799301147461, "epoch": 13, "memory": 9465, "step": 144192} +{"lr": 0.0003520365877844012, "data_time": 0.0013340473175048827, "grad_norm": 0.023131793458014728, "loss": 0.6882535576820373, "time": 0.4448213577270508, "epoch": 13, "memory": 9465, "step": 144292} +{"lr": 0.0003520365877844012, "data_time": 0.0012208938598632813, "grad_norm": 0.022750809043645858, "loss": 0.6848455965518951, "time": 0.5096841096878052, "epoch": 13, "memory": 9465, "step": 144392} +{"lr": 0.0003520365877844012, "data_time": 0.0012856006622314453, "grad_norm": 0.01734750246978365, "loss": 0.6886095404624939, "time": 0.4157839298248291, "epoch": 13, "memory": 9465, "step": 144492} +{"lr": 0.0003520365877844012, "data_time": 0.0012665033340454102, "grad_norm": 0.01712261061184108, "loss": 0.6874357998371124, "time": 0.431915283203125, "epoch": 13, "memory": 9465, "step": 144592} +{"lr": 0.0003520365877844012, "data_time": 0.0013406038284301757, "grad_norm": 0.028326851688325404, "loss": 0.6830387890338898, "time": 0.42901763916015623, "epoch": 13, "memory": 9465, "step": 144692} +{"lr": 0.0003520365877844012, "data_time": 0.001374363899230957, "grad_norm": 0.018115551909431815, "loss": 0.6854941308498382, "time": 0.43001062870025636, "epoch": 13, "memory": 9465, "step": 144792} +{"lr": 0.0003520365877844012, "data_time": 0.001308274269104004, "grad_norm": 0.028630966693162917, "loss": 0.6886234223842621, "time": 0.41717722415924074, "epoch": 13, "memory": 9465, "step": 144892} +{"lr": 0.0003520365877844012, "data_time": 0.0013165235519409179, "grad_norm": 0.022329334617825224, "loss": 0.6895581483840942, "time": 0.43071537017822265, "epoch": 13, "memory": 9465, "step": 144992} +{"lr": 0.0003520365877844012, "data_time": 0.0013569831848144532, "grad_norm": 0.01647830307483673, "loss": 0.6855230152606964, "time": 0.40398895740509033, "epoch": 13, "memory": 9465, "step": 145092} +{"lr": 0.0003520365877844012, "data_time": 0.0012874841690063477, "grad_norm": 0.01579254070529714, "loss": 0.6887718439102173, "time": 0.42697365283966066, "epoch": 13, "memory": 9465, "step": 145192} +{"lr": 0.0003520365877844012, "data_time": 0.0013854026794433594, "grad_norm": 0.029465958569198848, "loss": 0.6884518384933471, "time": 0.4738156795501709, "epoch": 13, "memory": 9465, "step": 145292} +{"lr": 0.0003520365877844012, "data_time": 0.0013043403625488282, "grad_norm": 0.01683385097421706, "loss": 0.6826500654220581, "time": 0.4256121635437012, "epoch": 13, "memory": 9465, "step": 145392} +{"lr": 0.0003520365877844012, "data_time": 0.0013903379440307617, "grad_norm": 0.01804582504555583, "loss": 0.6884677290916443, "time": 0.44648089408874514, "epoch": 13, "memory": 9465, "step": 145492} +{"lr": 0.0003520365877844012, "data_time": 0.001339888572692871, "grad_norm": 0.025155717320740224, "loss": 0.6856862604618073, "time": 0.4355248212814331, "epoch": 13, "memory": 9465, "step": 145592} +{"lr": 0.0003520365877844012, "data_time": 0.0015591859817504882, "grad_norm": 0.016057035746052863, "loss": 0.6883014917373658, "time": 0.42293860912323, "epoch": 13, "memory": 9465, "step": 145692} +{"lr": 0.0003520365877844012, "data_time": 0.0013610601425170898, "grad_norm": 0.02582974242977798, "loss": 0.6869897067546844, "time": 0.4554993391036987, "epoch": 13, "memory": 9465, "step": 145792} +{"lr": 0.0003520365877844012, "data_time": 0.0014297008514404298, "grad_norm": 0.024726646230556072, "loss": 0.69018075466156, "time": 0.4758347988128662, "epoch": 13, "memory": 9465, "step": 145892} +{"lr": 0.0003520365877844012, "data_time": 0.0013002872467041016, "grad_norm": 0.018723942223004997, "loss": 0.6887227237224579, "time": 0.43060152530670165, "epoch": 13, "memory": 9465, "step": 145992} +{"lr": 0.0003520365877844012, "data_time": 0.0014956474304199218, "grad_norm": 0.017027101363055408, "loss": 0.6869645178318023, "time": 0.4316537380218506, "epoch": 13, "memory": 9465, "step": 146092} +{"lr": 0.0003520365877844012, "data_time": 0.001397562026977539, "grad_norm": 0.016929687187075614, "loss": 0.6878800630569458, "time": 0.4517225980758667, "epoch": 13, "memory": 9465, "step": 146192} +{"lr": 0.0003520365877844012, "data_time": 0.0014849424362182618, "grad_norm": 0.019114992558024825, "loss": 0.6873187065124512, "time": 0.42660088539123536, "epoch": 13, "memory": 9465, "step": 146292} +{"lr": 0.0003520365877844012, "data_time": 0.0012789726257324218, "grad_norm": 0.03214428598294035, "loss": 0.6874493896961212, "time": 0.4159404993057251, "epoch": 13, "memory": 9465, "step": 146392} +{"lr": 0.0003520365877844012, "data_time": 0.0012720823287963867, "grad_norm": 0.028108444390818478, "loss": 0.6880874156951904, "time": 0.422563362121582, "epoch": 13, "memory": 9465, "step": 146492} +{"lr": 0.0003520365877844012, "data_time": 0.001325702667236328, "grad_norm": 0.019555490429047494, "loss": 0.6882882297039032, "time": 0.4210850238800049, "epoch": 13, "memory": 9465, "step": 146592} +{"lr": 0.0003520365877844012, "data_time": 0.0012747764587402344, "grad_norm": 0.02407969650812447, "loss": 0.690867680311203, "time": 0.4369359016418457, "epoch": 13, "memory": 9465, "step": 146692} +{"lr": 0.0003520365877844012, "data_time": 0.0013010501861572266, "grad_norm": 0.02648129309527576, "loss": 0.687272697687149, "time": 0.39036743640899657, "epoch": 13, "memory": 9465, "step": 146792} +{"lr": 0.0003520365877844012, "data_time": 0.0013555526733398438, "grad_norm": 0.018041092460043727, "loss": 0.680459588766098, "time": 0.40427727699279786, "epoch": 13, "memory": 9465, "step": 146892} +{"lr": 0.0003520365877844012, "data_time": 0.0014187097549438477, "grad_norm": 0.017114895593840627, "loss": 0.6896293878555297, "time": 0.432601261138916, "epoch": 13, "memory": 9465, "step": 146992} +{"lr": 0.0003520365877844012, "data_time": 0.0012846708297729493, "grad_norm": 0.02265499666100368, "loss": 0.6880480766296386, "time": 0.40092661380767824, "epoch": 13, "memory": 9465, "step": 147092} +{"lr": 0.0003520365877844012, "data_time": 0.001348400115966797, "grad_norm": 0.016791835159529, "loss": 0.6838982820510864, "time": 0.39970180988311765, "epoch": 13, "memory": 9465, "step": 147192} +{"lr": 0.0003520365877844012, "data_time": 0.0013466835021972655, "grad_norm": 0.021194009529426693, "loss": 0.6869896233081818, "time": 0.40505883693695066, "epoch": 13, "memory": 9465, "step": 147292} +{"lr": 0.0003520365877844012, "data_time": 0.0016285896301269532, "grad_norm": 0.017275077872909606, "loss": 0.6869803667068481, "time": 0.44537665843963625, "epoch": 13, "memory": 9465, "step": 147392} +{"lr": 0.0003520365877844012, "data_time": 0.0014423847198486329, "grad_norm": 0.027895302092656494, "loss": 0.6824289977550506, "time": 0.41504294872283937, "epoch": 13, "memory": 9465, "step": 147492} +{"lr": 0.0003520365877844012, "data_time": 0.001257467269897461, "grad_norm": 0.018320670281536878, "loss": 0.684909564256668, "time": 0.44445552825927737, "epoch": 13, "memory": 9465, "step": 147592} +{"lr": 0.0003520365877844012, "data_time": 0.0013446331024169922, "grad_norm": 0.02354107324499637, "loss": 0.6893847823143006, "time": 0.38520045280456544, "epoch": 13, "memory": 9465, "step": 147692} +{"lr": 0.0003520365877844012, "data_time": 0.001256561279296875, "grad_norm": 0.01573030628496781, "loss": 0.686847734451294, "time": 0.42736985683441164, "epoch": 13, "memory": 9465, "step": 147792} +{"lr": 0.0003520365877844012, "data_time": 0.0013084650039672852, "grad_norm": 0.016619264078326524, "loss": 0.6850524246692657, "time": 0.40325417518615725, "epoch": 13, "memory": 9465, "step": 147892} +{"lr": 0.0003520365877844012, "data_time": 0.0013421058654785156, "grad_norm": 0.026577353337779643, "loss": 0.6866621017456055, "time": 0.45670936107635496, "epoch": 13, "memory": 9465, "step": 147992} +{"lr": 0.0003520365877844012, "data_time": 0.0013792991638183593, "grad_norm": 0.01836807201616466, "loss": 0.6861838102340698, "time": 0.3969640493392944, "epoch": 13, "memory": 9465, "step": 148092} +{"lr": 0.0003520365877844012, "data_time": 0.0013062000274658204, "grad_norm": 0.013027273537591099, "loss": 0.6824309349060058, "time": 0.402159309387207, "epoch": 13, "memory": 9465, "step": 148192} +{"lr": 0.0003520365877844012, "data_time": 0.0013721466064453125, "grad_norm": 0.017777485214173792, "loss": 0.6860322594642639, "time": 0.39197554588317873, "epoch": 13, "memory": 9465, "step": 148292} +{"lr": 0.0003520365877844012, "data_time": 0.0014073610305786132, "grad_norm": 0.02188335722312331, "loss": 0.6832489430904388, "time": 0.38784499168395997, "epoch": 13, "memory": 9465, "step": 148392} +{"lr": 0.0003520365877844012, "data_time": 0.0013094663619995118, "grad_norm": 0.029621819080784918, "loss": 0.6863331019878387, "time": 0.40296661853790283, "epoch": 13, "memory": 9465, "step": 148492} +{"lr": 0.0003520365877844012, "data_time": 0.0013120651245117187, "grad_norm": 0.021716117835603656, "loss": 0.6840667843818664, "time": 0.44245638847351076, "epoch": 13, "memory": 9465, "step": 148592} +{"lr": 0.0003520365877844012, "data_time": 0.0014384031295776368, "grad_norm": 0.025756726786494256, "loss": 0.6886322021484375, "time": 0.4498127937316895, "epoch": 13, "memory": 9465, "step": 148692} +{"lr": 0.0003520365877844012, "data_time": 0.0012794017791748046, "grad_norm": 0.02449752176180482, "loss": 0.6841812789440155, "time": 0.39390561580657957, "epoch": 13, "memory": 9465, "step": 148792} +{"lr": 0.0003520365877844012, "data_time": 0.0013397455215454102, "grad_norm": 0.017251241515623406, "loss": 0.6848391532897949, "time": 0.4220003366470337, "epoch": 13, "memory": 9465, "step": 148892} +{"lr": 0.0003520365877844012, "data_time": 0.0012875080108642578, "grad_norm": 0.02317849623505026, "loss": 0.6844637930393219, "time": 0.4177465200424194, "epoch": 13, "memory": 9465, "step": 148992} +{"lr": 0.0003520365877844012, "data_time": 0.0016066551208496094, "grad_norm": 0.025580171146430075, "loss": 0.6857383906841278, "time": 0.4121209383010864, "epoch": 13, "memory": 9465, "step": 149092} +{"lr": 0.0003520365877844012, "data_time": 0.0013405323028564454, "grad_norm": 0.02009988371282816, "loss": 0.6857160985469818, "time": 0.3964763879776001, "epoch": 13, "memory": 9465, "step": 149192} +{"lr": 0.0003520365877844012, "data_time": 0.001444387435913086, "grad_norm": 0.03267081966623664, "loss": 0.6920430958271027, "time": 0.40468361377716067, "epoch": 13, "memory": 9465, "step": 149292} +{"lr": 0.0003520365877844012, "data_time": 0.001538228988647461, "grad_norm": 0.031247841753065587, "loss": 0.6883592486381531, "time": 0.4112871170043945, "epoch": 13, "memory": 9465, "step": 149392} +{"lr": 0.0003520365877844012, "data_time": 0.0013257503509521485, "grad_norm": 0.029722631350159644, "loss": 0.6896104454994202, "time": 0.3924055814743042, "epoch": 13, "memory": 9465, "step": 149492} +{"lr": 0.0003520365877844012, "data_time": 0.0020203590393066406, "grad_norm": 0.022426737286150456, "loss": 0.6843945145606994, "time": 0.406878399848938, "epoch": 13, "memory": 9465, "step": 149592} +{"lr": 0.0003520365877844012, "data_time": 0.011359167098999024, "grad_norm": 0.024466575216501953, "loss": 0.6845511078834534, "time": 0.38599724769592286, "epoch": 13, "memory": 9465, "step": 149692} +{"accuracy/top1": 100.0, "data_time": 0.30741747687844667, "time": 0.5612477695240694, "step": 13} +{"lr": 0.00028027470262892445, "data_time": 0.040578031539916994, "grad_norm": 0.02363255023956299, "loss": 0.6856961190700531, "time": 0.44960076808929444, "epoch": 14, "memory": 9465, "step": 149808} +{"lr": 0.00028027470262892445, "data_time": 0.0013695716857910155, "grad_norm": 0.03702503265812993, "loss": 0.6921934664249421, "time": 0.4566654682159424, "epoch": 14, "memory": 9465, "step": 149908} +{"lr": 0.00028027470262892445, "data_time": 0.001612401008605957, "grad_norm": 0.025036737532354893, "loss": 0.6891972303390503, "time": 0.3936192512512207, "epoch": 14, "memory": 9465, "step": 150008} +{"lr": 0.00028027470262892445, "data_time": 0.0014136075973510743, "grad_norm": 0.028032448887825013, "loss": 0.6853726267814636, "time": 0.40378150939941404, "epoch": 14, "memory": 9465, "step": 150108} +{"lr": 0.00028027470262892445, "data_time": 0.0016391277313232422, "grad_norm": 0.02167459358461201, "loss": 0.6890131831169128, "time": 0.4048474311828613, "epoch": 14, "memory": 9465, "step": 150208} +{"lr": 0.00028027470262892445, "data_time": 0.0016252994537353516, "grad_norm": 0.01866477259900421, "loss": 0.6893049836158752, "time": 0.3884351015090942, "epoch": 14, "memory": 9465, "step": 150308} +{"lr": 0.00028027470262892445, "data_time": 0.0014384984970092773, "grad_norm": 0.029272266384214162, "loss": 0.6797670483589172, "time": 0.38600687980651854, "epoch": 14, "memory": 9465, "step": 150408} +{"lr": 0.00028027470262892445, "data_time": 0.001467585563659668, "grad_norm": 0.016809142468264326, "loss": 0.6895006775856019, "time": 0.41752817630767824, "epoch": 14, "memory": 9465, "step": 150508} +{"lr": 0.00028027470262892445, "data_time": 0.0013631820678710938, "grad_norm": 0.0183136819396168, "loss": 0.6832292556762696, "time": 0.43120110034942627, "epoch": 14, "memory": 9465, "step": 150608} +{"lr": 0.00028027470262892445, "data_time": 0.0022603511810302735, "grad_norm": 0.01756924302317202, "loss": 0.6859959423542022, "time": 0.3948364734649658, "epoch": 14, "memory": 9465, "step": 150708} +{"lr": 0.00028027470262892445, "data_time": 0.001507735252380371, "grad_norm": 0.024924750439822674, "loss": 0.6871319591999054, "time": 0.4168450355529785, "epoch": 14, "memory": 9465, "step": 150808} +{"lr": 0.00028027470262892445, "data_time": 0.0013359546661376952, "grad_norm": 0.029977887263521553, "loss": 0.6806355714797974, "time": 0.4189192533493042, "epoch": 14, "memory": 9465, "step": 150908} +{"lr": 0.00028027470262892445, "data_time": 0.0015645742416381836, "grad_norm": 0.018552739499136807, "loss": 0.6834400951862335, "time": 0.3801810503005981, "epoch": 14, "memory": 9465, "step": 151008} +{"lr": 0.00028027470262892445, "data_time": 0.0013875722885131835, "grad_norm": 0.019572274759411812, "loss": 0.6849079072475434, "time": 0.4099190950393677, "epoch": 14, "memory": 9465, "step": 151108} +{"lr": 0.00028027470262892445, "data_time": 0.0013000965118408203, "grad_norm": 0.020054957189131528, "loss": 0.688249260187149, "time": 0.41030781269073485, "epoch": 14, "memory": 9465, "step": 151208} +{"lr": 0.00028027470262892445, "data_time": 0.0013312339782714845, "grad_norm": 0.023754696641117335, "loss": 0.6887979507446289, "time": 0.42234656810760496, "epoch": 14, "memory": 9465, "step": 151308} +{"lr": 0.00028027470262892445, "data_time": 0.0012336015701293946, "grad_norm": 0.02806314714252949, "loss": 0.6866526246070862, "time": 0.3873327970504761, "epoch": 14, "memory": 9465, "step": 151408} +{"lr": 0.00028027470262892445, "data_time": 0.001320052146911621, "grad_norm": 0.016271448781481013, "loss": 0.6874719798564911, "time": 0.3861680030822754, "epoch": 14, "memory": 9465, "step": 151508} +{"lr": 0.00028027470262892445, "data_time": 0.0012776374816894532, "grad_norm": 0.01814198181964457, "loss": 0.6835083842277527, "time": 0.3970287561416626, "epoch": 14, "memory": 9465, "step": 151608} +{"lr": 0.00028027470262892445, "data_time": 0.0013757705688476562, "grad_norm": 0.01849473762558773, "loss": 0.6913939416408539, "time": 0.393136191368103, "epoch": 14, "memory": 9465, "step": 151708} +{"lr": 0.00028027470262892445, "data_time": 0.0013037443161010743, "grad_norm": 0.013537014625035227, "loss": 0.6798896551132202, "time": 0.40006446838378906, "epoch": 14, "memory": 9465, "step": 151808} +{"lr": 0.00028027470262892445, "data_time": 0.0012566089630126954, "grad_norm": 0.031790386396460237, "loss": 0.682720524072647, "time": 0.4363923311233521, "epoch": 14, "memory": 9465, "step": 151908} +{"lr": 0.00028027470262892445, "data_time": 0.001699542999267578, "grad_norm": 0.030915590189397334, "loss": 0.6899689972400666, "time": 0.4120911121368408, "epoch": 14, "memory": 9465, "step": 152008} +{"lr": 0.00028027470262892445, "data_time": 0.0014336585998535156, "grad_norm": 0.024079860048368575, "loss": 0.6891722679138184, "time": 0.4283134460449219, "epoch": 14, "memory": 9465, "step": 152108} +{"lr": 0.00028027470262892445, "data_time": 0.0013783693313598633, "grad_norm": 0.02360130213201046, "loss": 0.6899493277072907, "time": 0.40278360843658445, "epoch": 14, "memory": 9465, "step": 152208} +{"lr": 0.00028027470262892445, "data_time": 0.0014038324356079102, "grad_norm": 0.028875906398752704, "loss": 0.6871864378452301, "time": 0.41941835880279543, "epoch": 14, "memory": 9465, "step": 152308} +{"lr": 0.00028027470262892445, "data_time": 0.0016535520553588867, "grad_norm": 0.0323748073540628, "loss": 0.6836329460144043, "time": 0.38824920654296874, "epoch": 14, "memory": 9465, "step": 152408} +{"lr": 0.00028027470262892445, "data_time": 0.0015223026275634766, "grad_norm": 0.02168747780378908, "loss": 0.6844379901885986, "time": 0.40359928607940676, "epoch": 14, "memory": 9465, "step": 152508} +{"lr": 0.00028027470262892445, "data_time": 0.0014693260192871094, "grad_norm": 0.0183307379251346, "loss": 0.6859334170818329, "time": 0.44213550090789794, "epoch": 14, "memory": 9465, "step": 152608} +{"lr": 0.00028027470262892445, "data_time": 0.0016148805618286133, "grad_norm": 0.01920490147313103, "loss": 0.6874366044998169, "time": 0.40976390838623045, "epoch": 14, "memory": 9465, "step": 152708} +{"lr": 0.00028027470262892445, "data_time": 0.0016313314437866211, "grad_norm": 0.015319457871373742, "loss": 0.686477530002594, "time": 0.39671027660369873, "epoch": 14, "memory": 9465, "step": 152808} +{"lr": 0.00028027470262892445, "data_time": 0.0012980222702026368, "grad_norm": 0.01969348026905209, "loss": 0.6829213261604309, "time": 0.4105373859405518, "epoch": 14, "memory": 9465, "step": 152908} +{"lr": 0.00028027470262892445, "data_time": 0.0014499902725219726, "grad_norm": 0.021235161134973167, "loss": 0.6763473749160767, "time": 0.38748793601989745, "epoch": 14, "memory": 9465, "step": 153008} +{"lr": 0.00028027470262892445, "data_time": 0.00168154239654541, "grad_norm": 0.015591676131589339, "loss": 0.6831153154373169, "time": 0.4621685266494751, "epoch": 14, "memory": 9465, "step": 153108} +{"lr": 0.00028027470262892445, "data_time": 0.001639246940612793, "grad_norm": 0.018944092630408704, "loss": 0.6830090641975403, "time": 0.42252817153930666, "epoch": 14, "memory": 9465, "step": 153208} +{"lr": 0.00028027470262892445, "data_time": 0.0013518095016479491, "grad_norm": 0.02016517990268767, "loss": 0.6845646619796752, "time": 0.4187780380249023, "epoch": 14, "memory": 9465, "step": 153308} +{"lr": 0.00028027470262892445, "data_time": 0.0013185501098632812, "grad_norm": 0.01672271793941036, "loss": 0.6840503096580506, "time": 0.37966482639312743, "epoch": 14, "memory": 9465, "step": 153408} +{"lr": 0.00028027470262892445, "data_time": 0.0018891096115112305, "grad_norm": 0.021451255236752332, "loss": 0.68685182929039, "time": 0.41676404476165774, "epoch": 14, "memory": 9465, "step": 153508} +{"lr": 0.00028027470262892445, "data_time": 0.0013552427291870118, "grad_norm": 0.02101659409236163, "loss": 0.6868496894836426, "time": 0.44055216312408446, "epoch": 14, "memory": 9465, "step": 153608} +{"lr": 0.00028027470262892445, "data_time": 0.0014045000076293944, "grad_norm": 0.017874143016524614, "loss": 0.683040851354599, "time": 0.43531954288482666, "epoch": 14, "memory": 9465, "step": 153708} +{"lr": 0.00028027470262892445, "data_time": 0.0014707565307617188, "grad_norm": 0.0309216583147645, "loss": 0.685529601573944, "time": 0.3882774591445923, "epoch": 14, "memory": 9465, "step": 153808} +{"lr": 0.00028027470262892445, "data_time": 0.001372051239013672, "grad_norm": 0.036785055231302974, "loss": 0.6863934159278869, "time": 0.42968709468841554, "epoch": 14, "memory": 9465, "step": 153908} +{"lr": 0.00028027470262892445, "data_time": 0.0014786481857299804, "grad_norm": 0.01155533161945641, "loss": 0.686000519990921, "time": 0.4448122262954712, "epoch": 14, "memory": 9465, "step": 154008} +{"lr": 0.00028027470262892445, "data_time": 0.0014128923416137696, "grad_norm": 0.018938945885747672, "loss": 0.6897554159164428, "time": 0.3970362186431885, "epoch": 14, "memory": 9465, "step": 154108} +{"lr": 0.00028027470262892445, "data_time": 0.0013833045959472656, "grad_norm": 0.01627225533593446, "loss": 0.6871192872524261, "time": 0.3953340291976929, "epoch": 14, "memory": 9465, "step": 154208} +{"lr": 0.00028027470262892445, "data_time": 0.0015748500823974609, "grad_norm": 0.024380440218374134, "loss": 0.683938843011856, "time": 0.38283524513244627, "epoch": 14, "memory": 9465, "step": 154308} +{"lr": 0.00028027470262892445, "data_time": 0.0014348268508911134, "grad_norm": 0.05108619704842567, "loss": 0.6826554536819458, "time": 0.40686051845550536, "epoch": 14, "memory": 9465, "step": 154408} +{"lr": 0.00028027470262892445, "data_time": 0.0014477968215942383, "grad_norm": 0.01483023869805038, "loss": 0.6876264333724975, "time": 0.42267732620239257, "epoch": 14, "memory": 9465, "step": 154508} +{"lr": 0.00028027470262892445, "data_time": 0.0013699293136596679, "grad_norm": 0.042110367212444544, "loss": 0.6855251073837281, "time": 0.40648598670959474, "epoch": 14, "memory": 9465, "step": 154608} +{"lr": 0.00028027470262892445, "data_time": 0.00148317813873291, "grad_norm": 0.05280132433399558, "loss": 0.686598539352417, "time": 0.42174410820007324, "epoch": 14, "memory": 9465, "step": 154708} +{"lr": 0.00028027470262892445, "data_time": 0.0012965202331542969, "grad_norm": 0.041294691059738396, "loss": 0.6906365215778351, "time": 0.3991217851638794, "epoch": 14, "memory": 9465, "step": 154808} +{"lr": 0.00028027470262892445, "data_time": 0.0013070106506347656, "grad_norm": 0.0483148371335119, "loss": 0.6866521239280701, "time": 0.4468418598175049, "epoch": 14, "memory": 9465, "step": 154908} +{"lr": 0.00028027470262892445, "data_time": 0.0015760421752929687, "grad_norm": 0.19067063089460135, "loss": 0.6856415331363678, "time": 0.4100227355957031, "epoch": 14, "memory": 9465, "step": 155008} +{"lr": 0.00028027470262892445, "data_time": 0.002025866508483887, "grad_norm": 0.02737265024334192, "loss": 0.6825403451919556, "time": 0.3975337266921997, "epoch": 14, "memory": 9465, "step": 155108} +{"lr": 0.00028027470262892445, "data_time": 0.002056574821472168, "grad_norm": 0.03622059300541878, "loss": 0.6835957884788513, "time": 0.4041880130767822, "epoch": 14, "memory": 9465, "step": 155208} +{"lr": 0.00028027470262892445, "data_time": 0.0022015810012817384, "grad_norm": 0.0383616394829005, "loss": 0.684986925125122, "time": 0.39022395610809324, "epoch": 14, "memory": 9465, "step": 155308} +{"lr": 0.00028027470262892445, "data_time": 0.002637624740600586, "grad_norm": 0.026423221081495286, "loss": 0.6854630827903747, "time": 0.3999086856842041, "epoch": 14, "memory": 9465, "step": 155408} +{"lr": 0.00028027470262892445, "data_time": 0.0014970779418945312, "grad_norm": 0.07129687988199293, "loss": 0.6865840375423431, "time": 0.4012380361557007, "epoch": 14, "memory": 9465, "step": 155508} +{"lr": 0.00028027470262892445, "data_time": 0.0019565105438232424, "grad_norm": 0.042816708935424684, "loss": 0.6909674525260925, "time": 0.40330042839050295, "epoch": 14, "memory": 9465, "step": 155608} +{"lr": 0.00028027470262892445, "data_time": 0.001831960678100586, "grad_norm": 0.02322618179023266, "loss": 0.6862935781478882, "time": 0.4133880376815796, "epoch": 14, "memory": 9465, "step": 155708} +{"lr": 0.00028027470262892445, "data_time": 0.001990056037902832, "grad_norm": 0.08056890348671004, "loss": 0.6866194248199463, "time": 0.42566213607788084, "epoch": 14, "memory": 9465, "step": 155808} +{"lr": 0.00028027470262892445, "data_time": 0.0026386499404907225, "grad_norm": 0.02329719658009708, "loss": 0.6899753332138061, "time": 0.37984864711761473, "epoch": 14, "memory": 9465, "step": 155908} +{"lr": 0.00028027470262892445, "data_time": 0.002470850944519043, "grad_norm": 0.031849815510213375, "loss": 0.6918782651424408, "time": 0.4714346170425415, "epoch": 14, "memory": 9465, "step": 156008} +{"lr": 0.00028027470262892445, "data_time": 0.0022544622421264648, "grad_norm": 0.09037298895418644, "loss": 0.6869585394859314, "time": 0.402772331237793, "epoch": 14, "memory": 9465, "step": 156108} +{"lr": 0.00028027470262892445, "data_time": 0.001898646354675293, "grad_norm": 0.038070216029882434, "loss": 0.6842526078224183, "time": 0.39388854503631593, "epoch": 14, "memory": 9465, "step": 156208} +{"lr": 0.00028027470262892445, "data_time": 0.001858663558959961, "grad_norm": 0.14319400303065777, "loss": 0.6865509212017059, "time": 0.3899399757385254, "epoch": 14, "memory": 9465, "step": 156308} +{"lr": 0.00028027470262892445, "data_time": 0.002242755889892578, "grad_norm": 0.031476462911814454, "loss": 0.6879517734050751, "time": 0.4034332513809204, "epoch": 14, "memory": 9465, "step": 156408} +{"lr": 0.00028027470262892445, "data_time": 0.0023975133895874023, "grad_norm": 0.027712808176875114, "loss": 0.681372219324112, "time": 0.4083216667175293, "epoch": 14, "memory": 9465, "step": 156508} +{"lr": 0.00028027470262892445, "data_time": 0.00198667049407959, "grad_norm": 0.028507962450385095, "loss": 0.6839473962783813, "time": 0.4277604818344116, "epoch": 14, "memory": 9465, "step": 156608} +{"lr": 0.00028027470262892445, "data_time": 0.0019521236419677735, "grad_norm": 0.02448674967745319, "loss": 0.6873429596424103, "time": 0.4227177858352661, "epoch": 14, "memory": 9465, "step": 156708} +{"lr": 0.00028027470262892445, "data_time": 0.001888108253479004, "grad_norm": 0.0313364093657583, "loss": 0.6884092569351197, "time": 0.4373295545578003, "epoch": 14, "memory": 9465, "step": 156808} +{"lr": 0.00028027470262892445, "data_time": 0.0019359111785888672, "grad_norm": 0.02219557317439467, "loss": 0.6841471076011658, "time": 0.4302870988845825, "epoch": 14, "memory": 9465, "step": 156908} +{"lr": 0.00028027470262892445, "data_time": 0.002111482620239258, "grad_norm": 0.03421001746319234, "loss": 0.6835174262523651, "time": 0.39418256282806396, "epoch": 14, "memory": 9465, "step": 157008} +{"lr": 0.00028027470262892445, "data_time": 0.0020578622817993162, "grad_norm": 0.013493796996772289, "loss": 0.6880625247955322, "time": 0.43144266605377196, "epoch": 14, "memory": 9465, "step": 157108} +{"lr": 0.00028027470262892445, "data_time": 0.0016768693923950196, "grad_norm": 0.029052113997749984, "loss": 0.689518803358078, "time": 0.40128026008605955, "epoch": 14, "memory": 9465, "step": 157208} +{"lr": 0.00028027470262892445, "data_time": 0.0016528606414794923, "grad_norm": 0.023886117665097118, "loss": 0.6865138351917267, "time": 0.3945567846298218, "epoch": 14, "memory": 9465, "step": 157308} +{"lr": 0.00028027470262892445, "data_time": 0.002325916290283203, "grad_norm": 0.027004258264787494, "loss": 0.6864407360553741, "time": 0.39932076930999755, "epoch": 14, "memory": 9465, "step": 157408} +{"lr": 0.00028027470262892445, "data_time": 0.002185344696044922, "grad_norm": 0.08682013629004359, "loss": 0.6798774540424347, "time": 0.39725723266601565, "epoch": 14, "memory": 9465, "step": 157508} +{"lr": 0.00028027470262892445, "data_time": 0.0021752119064331055, "grad_norm": 0.028024967899546026, "loss": 0.6860984563827515, "time": 0.4208559989929199, "epoch": 14, "memory": 9465, "step": 157608} +{"lr": 0.00028027470262892445, "data_time": 0.0019136667251586914, "grad_norm": 0.029734809999354184, "loss": 0.6827096998691559, "time": 0.3966014623641968, "epoch": 14, "memory": 9465, "step": 157708} +{"lr": 0.00028027470262892445, "data_time": 0.0017208337783813476, "grad_norm": 0.03303380589932203, "loss": 0.6838254630565643, "time": 0.40142972469329835, "epoch": 14, "memory": 9465, "step": 157808} +{"lr": 0.00028027470262892445, "data_time": 0.0026412248611450196, "grad_norm": 0.02461668774485588, "loss": 0.6841819941997528, "time": 0.39656853675842285, "epoch": 14, "memory": 9465, "step": 157908} +{"lr": 0.00028027470262892445, "data_time": 0.0018293142318725586, "grad_norm": 0.0292770110652782, "loss": 0.6882569372653962, "time": 0.41058878898620604, "epoch": 14, "memory": 9465, "step": 158008} +{"lr": 0.00028027470262892445, "data_time": 0.0019101142883300782, "grad_norm": 0.021645177667960525, "loss": 0.6902525365352631, "time": 0.41611638069152834, "epoch": 14, "memory": 9465, "step": 158108} +{"lr": 0.00028027470262892445, "data_time": 0.001793646812438965, "grad_norm": 0.06236221301369369, "loss": 0.6861551165580749, "time": 0.4351755857467651, "epoch": 14, "memory": 9465, "step": 158208} +{"lr": 0.00028027470262892445, "data_time": 0.001773834228515625, "grad_norm": 0.037126504629850385, "loss": 0.6846661806106568, "time": 0.40097382068634035, "epoch": 14, "memory": 9465, "step": 158308} +{"lr": 0.00028027470262892445, "data_time": 0.0017854690551757813, "grad_norm": 0.01951689086854458, "loss": 0.6915440082550048, "time": 0.3932562589645386, "epoch": 14, "memory": 9465, "step": 158408} +{"lr": 0.00028027470262892445, "data_time": 0.0020975589752197264, "grad_norm": 0.043020909558981656, "loss": 0.6870308697223664, "time": 0.4824603796005249, "epoch": 14, "memory": 9465, "step": 158508} +{"lr": 0.00028027470262892445, "data_time": 0.0019032716751098632, "grad_norm": 0.08353183420840651, "loss": 0.6912239372730256, "time": 0.4043946504592896, "epoch": 14, "memory": 9465, "step": 158608} +{"lr": 0.00028027470262892445, "data_time": 0.001536989212036133, "grad_norm": 0.035521398112177846, "loss": 0.6825567483901978, "time": 0.4237253427505493, "epoch": 14, "memory": 9465, "step": 158708} +{"lr": 0.00028027470262892445, "data_time": 0.002265143394470215, "grad_norm": 0.12171778678894044, "loss": 0.6851697981357574, "time": 0.40695106983184814, "epoch": 14, "memory": 9465, "step": 158808} +{"lr": 0.00028027470262892445, "data_time": 0.0019767045974731444, "grad_norm": 0.028613679856061936, "loss": 0.6916392803192138, "time": 0.41077594757080077, "epoch": 14, "memory": 9465, "step": 158908} +{"lr": 0.00028027470262892445, "data_time": 0.002177524566650391, "grad_norm": 0.019880210445262492, "loss": 0.686233127117157, "time": 0.40418262481689454, "epoch": 14, "memory": 9465, "step": 159008} +{"lr": 0.00028027470262892445, "data_time": 0.002059483528137207, "grad_norm": 0.015642314485739918, "loss": 0.6872299492359162, "time": 0.40851283073425293, "epoch": 14, "memory": 9465, "step": 159108} +{"lr": 0.00028027470262892445, "data_time": 0.0013713836669921875, "grad_norm": 0.05535896783694625, "loss": 0.695352977514267, "time": 0.41545073986053466, "epoch": 14, "memory": 9465, "step": 159208} +{"lr": 0.00028027470262892445, "data_time": 0.0013283014297485352, "grad_norm": 0.028885880671441554, "loss": 0.683948028087616, "time": 0.41185173988342283, "epoch": 14, "memory": 9465, "step": 159308} +{"lr": 0.00028027470262892445, "data_time": 0.0013547182083129884, "grad_norm": 0.03945363105740398, "loss": 0.6870556950569153, "time": 0.42182073593139646, "epoch": 14, "memory": 9465, "step": 159408} +{"lr": 0.00028027470262892445, "data_time": 0.001593613624572754, "grad_norm": 0.05568387564271689, "loss": 0.6910943686962128, "time": 0.44905292987823486, "epoch": 14, "memory": 9465, "step": 159508} +{"lr": 0.00028027470262892445, "data_time": 0.0017818927764892579, "grad_norm": 0.12322857351973653, "loss": 0.6851725339889526, "time": 0.4095428943634033, "epoch": 14, "memory": 9465, "step": 159608} +{"lr": 0.00028027470262892445, "data_time": 0.0017316102981567382, "grad_norm": 0.0301972683519125, "loss": 0.68194180727005, "time": 0.436036491394043, "epoch": 14, "memory": 9465, "step": 159708} +{"lr": 0.00028027470262892445, "data_time": 0.0023550987243652344, "grad_norm": 0.02634110040962696, "loss": 0.6899545848369598, "time": 0.40677814483642577, "epoch": 14, "memory": 9465, "step": 159808} +{"lr": 0.00028027470262892445, "data_time": 0.0015550374984741211, "grad_norm": 0.0231823165435344, "loss": 0.6842324733734131, "time": 0.4056877851486206, "epoch": 14, "memory": 9465, "step": 159908} +{"lr": 0.00028027470262892445, "data_time": 0.0014517068862915038, "grad_norm": 0.024236548971384764, "loss": 0.6827456235885621, "time": 0.39584548473358155, "epoch": 14, "memory": 9465, "step": 160008} +{"lr": 0.00028027470262892445, "data_time": 0.0014163970947265625, "grad_norm": 0.019567030342295767, "loss": 0.6907675683498382, "time": 0.42159667015075686, "epoch": 14, "memory": 9465, "step": 160108} +{"lr": 0.00028027470262892445, "data_time": 0.0014188766479492187, "grad_norm": 0.03959140651859343, "loss": 0.6838397264480591, "time": 0.4625308752059937, "epoch": 14, "memory": 9465, "step": 160208} +{"lr": 0.00028027470262892445, "data_time": 0.002048826217651367, "grad_norm": 0.022645322373136878, "loss": 0.6872423529624939, "time": 0.4821256875991821, "epoch": 14, "memory": 9465, "step": 160308} +{"lr": 0.00028027470262892445, "data_time": 0.001827239990234375, "grad_norm": 0.029500603955239058, "loss": 0.6850732922554016, "time": 0.4751749992370605, "epoch": 14, "memory": 9465, "step": 160408} +{"lr": 0.00028027470262892445, "data_time": 0.002065300941467285, "grad_norm": 0.01995218815281987, "loss": 0.6893764853477478, "time": 0.42381384372711184, "epoch": 14, "memory": 9465, "step": 160508} +{"lr": 0.00028027470262892445, "data_time": 0.0014534950256347655, "grad_norm": 0.0288451686501503, "loss": 0.6832719922065735, "time": 0.39760754108428953, "epoch": 14, "memory": 9465, "step": 160608} +{"lr": 0.00028027470262892445, "data_time": 0.0017058849334716797, "grad_norm": 0.03133925325237215, "loss": 0.6815317809581757, "time": 0.40827987194061277, "epoch": 14, "memory": 9465, "step": 160708} +{"lr": 0.00028027470262892445, "data_time": 0.0015241384506225586, "grad_norm": 0.019520680559799074, "loss": 0.6827360630035401, "time": 0.435684871673584, "epoch": 14, "memory": 9465, "step": 160808} +{"lr": 0.00028027470262892445, "data_time": 0.0020036697387695312, "grad_norm": 0.022928539512213318, "loss": 0.6836845993995666, "time": 0.4341479778289795, "epoch": 14, "memory": 9465, "step": 160908} +{"lr": 0.00028027470262892445, "data_time": 0.0014281749725341796, "grad_norm": 0.019421445624902844, "loss": 0.6835116386413574, "time": 0.47015607357025146, "epoch": 14, "memory": 9465, "step": 161008} +{"lr": 0.00028027470262892445, "data_time": 0.0014080047607421876, "grad_norm": 0.024089456733781843, "loss": 0.6846429884433747, "time": 0.39995200634002687, "epoch": 14, "memory": 9465, "step": 161108} +{"lr": 0.00028027470262892445, "data_time": 0.00597069263458252, "grad_norm": 0.027609765203669666, "loss": 0.6893363296985626, "time": 0.40682644844055177, "epoch": 14, "memory": 9465, "step": 161208} +{"accuracy/top1": 100.0, "data_time": 0.2853405054877786, "time": 0.5453655439264634, "step": 14} +{"lr": 0.0002140463001152259, "data_time": 0.08123340606689453, "grad_norm": 0.01871814049081877, "loss": 0.6883578598499298, "time": 0.4226930856704712, "epoch": 15, "memory": 9465, "step": 161324} +{"lr": 0.0002140463001152259, "data_time": 0.0014652729034423829, "grad_norm": 0.017157578165642916, "loss": 0.6913799524307251, "time": 0.40656795501708987, "epoch": 15, "memory": 9465, "step": 161424} +{"lr": 0.0002140463001152259, "data_time": 0.0014858722686767578, "grad_norm": 0.013531868599238805, "loss": 0.6874639749526977, "time": 0.3904269695281982, "epoch": 15, "memory": 9465, "step": 161524} +{"lr": 0.0002140463001152259, "data_time": 0.0012707233428955079, "grad_norm": 0.016254200006369502, "loss": 0.6860540568828583, "time": 0.41412451267242434, "epoch": 15, "memory": 9465, "step": 161624} +{"lr": 0.0002140463001152259, "data_time": 0.0013164758682250976, "grad_norm": 0.01915219735528808, "loss": 0.6880498588085174, "time": 0.4387094974517822, "epoch": 15, "memory": 9465, "step": 161724} +{"lr": 0.0002140463001152259, "data_time": 0.00133209228515625, "grad_norm": 0.01842809095978737, "loss": 0.6838313162326812, "time": 0.454335355758667, "epoch": 15, "memory": 9465, "step": 161824} +{"lr": 0.0002140463001152259, "data_time": 0.0021929264068603514, "grad_norm": 0.023325519729405643, "loss": 0.6863415241241455, "time": 0.4828128099441528, "epoch": 15, "memory": 9465, "step": 161924} +{"lr": 0.0002140463001152259, "data_time": 0.0015517950057983398, "grad_norm": 0.02742227101698518, "loss": 0.6828200161457062, "time": 0.4705327272415161, "epoch": 15, "memory": 9465, "step": 162024} +{"lr": 0.0002140463001152259, "data_time": 0.0014075756072998047, "grad_norm": 0.02574599109357223, "loss": 0.6864741265773773, "time": 0.4059830904006958, "epoch": 15, "memory": 9465, "step": 162124} +{"lr": 0.0002140463001152259, "data_time": 0.0012774944305419921, "grad_norm": 0.09568184425588697, "loss": 0.691490238904953, "time": 0.46218457221984866, "epoch": 15, "memory": 9465, "step": 162224} +{"lr": 0.0002140463001152259, "data_time": 0.0013579130172729492, "grad_norm": 0.031108124926686288, "loss": 0.6912797629833222, "time": 0.4240349292755127, "epoch": 15, "memory": 9465, "step": 162324} +{"lr": 0.0002140463001152259, "data_time": 0.0013774871826171876, "grad_norm": 0.022236520587466658, "loss": 0.6888606786727905, "time": 0.41772751808166503, "epoch": 15, "memory": 9465, "step": 162424} +{"lr": 0.0002140463001152259, "data_time": 0.0013132333755493165, "grad_norm": 0.01871593832038343, "loss": 0.6861822009086609, "time": 0.42862424850463865, "epoch": 15, "memory": 9465, "step": 162524} +{"lr": 0.0002140463001152259, "data_time": 0.0012571334838867188, "grad_norm": 0.02354892613366246, "loss": 0.6847135305404664, "time": 0.42249486446380613, "epoch": 15, "memory": 9465, "step": 162624} +{"lr": 0.0002140463001152259, "data_time": 0.0013778209686279297, "grad_norm": 0.017027739610057324, "loss": 0.6876058578491211, "time": 0.4112085819244385, "epoch": 15, "memory": 9465, "step": 162724} +{"lr": 0.0002140463001152259, "data_time": 0.0016553640365600587, "grad_norm": 0.021537176356650888, "loss": 0.6872940182685852, "time": 0.44995846748352053, "epoch": 15, "memory": 9465, "step": 162824} +{"lr": 0.0002140463001152259, "data_time": 0.0018907546997070312, "grad_norm": 0.0123816970793996, "loss": 0.6894800066947937, "time": 0.3972836256027222, "epoch": 15, "memory": 9465, "step": 162924} +{"lr": 0.0002140463001152259, "data_time": 0.0020686626434326173, "grad_norm": 0.024162659666035324, "loss": 0.6879245042800903, "time": 0.42353081703186035, "epoch": 15, "memory": 9465, "step": 163024} +{"lr": 0.0002140463001152259, "data_time": 0.001407027244567871, "grad_norm": 0.023543273913674057, "loss": 0.6884262084960937, "time": 0.40325064659118653, "epoch": 15, "memory": 9465, "step": 163124} +{"lr": 0.0002140463001152259, "data_time": 0.0013965606689453126, "grad_norm": 0.0197966959036421, "loss": 0.6834262669086456, "time": 0.4234490394592285, "epoch": 15, "memory": 9465, "step": 163224} +{"lr": 0.0002140463001152259, "data_time": 0.0013370275497436523, "grad_norm": 0.011397153418511153, "loss": 0.6868057489395142, "time": 0.4087916135787964, "epoch": 15, "memory": 9465, "step": 163324} +{"lr": 0.0002140463001152259, "data_time": 0.0018401861190795899, "grad_norm": 0.028872019331902265, "loss": 0.6871312022209167, "time": 0.416003680229187, "epoch": 15, "memory": 9465, "step": 163424} +{"lr": 0.0002140463001152259, "data_time": 0.0015646696090698242, "grad_norm": 0.01595398981589824, "loss": 0.6868103325366974, "time": 0.43205115795135496, "epoch": 15, "memory": 9465, "step": 163524} +{"lr": 0.0002140463001152259, "data_time": 0.0014375209808349609, "grad_norm": 0.01829653845052235, "loss": 0.6877792179584503, "time": 0.44249711036682127, "epoch": 15, "memory": 9465, "step": 163624} +{"lr": 0.0002140463001152259, "data_time": 0.0012975692749023437, "grad_norm": 0.01549928366439417, "loss": 0.6906494081020356, "time": 0.4179201364517212, "epoch": 15, "memory": 9465, "step": 163724} +{"lr": 0.0002140463001152259, "data_time": 0.0012684345245361328, "grad_norm": 0.02421788952779025, "loss": 0.6788218557834625, "time": 0.4103742837905884, "epoch": 15, "memory": 9465, "step": 163824} +{"lr": 0.0002140463001152259, "data_time": 0.0012929677963256837, "grad_norm": 0.026374685857445002, "loss": 0.6880486249923706, "time": 0.42886929512023925, "epoch": 15, "memory": 9465, "step": 163924} +{"lr": 0.0002140463001152259, "data_time": 0.0013208389282226562, "grad_norm": 0.022315075946971775, "loss": 0.6821879923343659, "time": 0.43700494766235354, "epoch": 15, "memory": 9465, "step": 164024} +{"lr": 0.0002140463001152259, "data_time": 0.001446366310119629, "grad_norm": 0.02220885206479579, "loss": 0.6825691878795623, "time": 0.41073644161224365, "epoch": 15, "memory": 9465, "step": 164124} +{"lr": 0.0002140463001152259, "data_time": 0.0013746976852416991, "grad_norm": 0.02066676509566605, "loss": 0.6900253236293793, "time": 0.3928713798522949, "epoch": 15, "memory": 9465, "step": 164224} +{"lr": 0.0002140463001152259, "data_time": 0.0013336658477783204, "grad_norm": 0.024910126929171383, "loss": 0.6860114872455597, "time": 0.43446879386901854, "epoch": 15, "memory": 9465, "step": 164324} +{"lr": 0.0002140463001152259, "data_time": 0.0027292728424072265, "grad_norm": 0.01571215141448192, "loss": 0.6840395569801331, "time": 0.4062842130661011, "epoch": 15, "memory": 9465, "step": 164424} +{"lr": 0.0002140463001152259, "data_time": 0.0014891862869262696, "grad_norm": 0.017370659043081105, "loss": 0.6879391372203827, "time": 0.42487130165100095, "epoch": 15, "memory": 9465, "step": 164524} +{"lr": 0.0002140463001152259, "data_time": 0.0029613494873046873, "grad_norm": 0.017539329291321338, "loss": 0.6863195359706878, "time": 0.39965410232543946, "epoch": 15, "memory": 9465, "step": 164624} +{"lr": 0.0002140463001152259, "data_time": 0.001259589195251465, "grad_norm": 0.032818767335265876, "loss": 0.6848550081253052, "time": 0.42869665622711184, "epoch": 15, "memory": 9465, "step": 164724} +{"lr": 0.0002140463001152259, "data_time": 0.0014023780822753906, "grad_norm": 0.021529263840056954, "loss": 0.6866447985172272, "time": 0.40840222835540774, "epoch": 15, "memory": 9465, "step": 164824} +{"lr": 0.0002140463001152259, "data_time": 0.0013232231140136719, "grad_norm": 0.015167720953468234, "loss": 0.6840157330036163, "time": 0.45548672676086427, "epoch": 15, "memory": 9465, "step": 164924} +{"lr": 0.0002140463001152259, "data_time": 0.002181410789489746, "grad_norm": 0.02375341553706676, "loss": 0.6797486484050751, "time": 0.39582879543304444, "epoch": 15, "memory": 9465, "step": 165024} +{"lr": 0.0002140463001152259, "data_time": 0.0013155460357666016, "grad_norm": 0.027573473216034473, "loss": 0.6872960150241851, "time": 0.41104028224945066, "epoch": 15, "memory": 9465, "step": 165124} +{"lr": 0.0002140463001152259, "data_time": 0.0011765480041503907, "grad_norm": 0.01996021153172478, "loss": 0.6842414379119873, "time": 0.38449184894561766, "epoch": 15, "memory": 9465, "step": 165224} +{"lr": 0.0002140463001152259, "data_time": 0.0012840747833251954, "grad_norm": 0.026236360101029275, "loss": 0.6881677746772766, "time": 0.4240997552871704, "epoch": 15, "memory": 9465, "step": 165324} +{"lr": 0.0002140463001152259, "data_time": 0.001308298110961914, "grad_norm": 0.027417116984724998, "loss": 0.6869114696979522, "time": 0.4011900186538696, "epoch": 15, "memory": 9465, "step": 165424} +{"lr": 0.0002140463001152259, "data_time": 0.0013127326965332031, "grad_norm": 0.018860459560528398, "loss": 0.6868092477321625, "time": 0.4262763261795044, "epoch": 15, "memory": 9465, "step": 165524} +{"lr": 0.0002140463001152259, "data_time": 0.0012764692306518554, "grad_norm": 0.01959541466785595, "loss": 0.6908905982971192, "time": 0.4086129903793335, "epoch": 15, "memory": 9465, "step": 165624} +{"lr": 0.0002140463001152259, "data_time": 0.00145721435546875, "grad_norm": 0.021767177432775498, "loss": 0.6858685433864593, "time": 0.41839349269866943, "epoch": 15, "memory": 9465, "step": 165724} +{"lr": 0.0002140463001152259, "data_time": 0.0018210172653198241, "grad_norm": 0.028958979714661836, "loss": 0.688239985704422, "time": 0.38466658592224123, "epoch": 15, "memory": 9465, "step": 165824} +{"lr": 0.0002140463001152259, "data_time": 0.001928997039794922, "grad_norm": 0.022747311904095114, "loss": 0.6879335582256317, "time": 0.4168570518493652, "epoch": 15, "memory": 9465, "step": 165924} +{"lr": 0.0002140463001152259, "data_time": 0.0012707233428955079, "grad_norm": 0.017367703886702657, "loss": 0.6828688263893128, "time": 0.42338855266571046, "epoch": 15, "memory": 9465, "step": 166024} +{"lr": 0.0002140463001152259, "data_time": 0.0013082027435302734, "grad_norm": 0.022956931637600063, "loss": 0.6796342492103576, "time": 0.4093561887741089, "epoch": 15, "memory": 9465, "step": 166124} +{"lr": 0.0002140463001152259, "data_time": 0.0011998414993286133, "grad_norm": 0.023125998489558697, "loss": 0.6874800384044647, "time": 0.40087432861328126, "epoch": 15, "memory": 9465, "step": 166224} +{"lr": 0.0002140463001152259, "data_time": 0.0021225690841674806, "grad_norm": 0.013915580738103017, "loss": 0.6853613078594207, "time": 0.4467417478561401, "epoch": 15, "memory": 9465, "step": 166324} +{"lr": 0.0002140463001152259, "data_time": 0.0014504432678222657, "grad_norm": 0.01924175429157913, "loss": 0.6847457170486451, "time": 0.4396371841430664, "epoch": 15, "memory": 9465, "step": 166424} +{"lr": 0.0002140463001152259, "data_time": 0.001787424087524414, "grad_norm": 0.01769362259947229, "loss": 0.6888521194458008, "time": 0.4142194986343384, "epoch": 15, "memory": 9465, "step": 166524} +{"lr": 0.0002140463001152259, "data_time": 0.0025406122207641602, "grad_norm": 0.017566913715563716, "loss": 0.6883926689624786, "time": 0.4275252342224121, "epoch": 15, "memory": 9465, "step": 166624} +{"lr": 0.0002140463001152259, "data_time": 0.0015596628189086914, "grad_norm": 0.031176075522671452, "loss": 0.6915314674377442, "time": 0.4031976699829102, "epoch": 15, "memory": 9465, "step": 166724} +{"lr": 0.0002140463001152259, "data_time": 0.0013432025909423829, "grad_norm": 0.02114136191084981, "loss": 0.6833634734153747, "time": 0.3974787473678589, "epoch": 15, "memory": 9465, "step": 166824} +{"lr": 0.0002140463001152259, "data_time": 0.0018229961395263671, "grad_norm": 0.013314515765523538, "loss": 0.6885230422019959, "time": 0.39331295490264895, "epoch": 15, "memory": 9465, "step": 166924} +{"lr": 0.0002140463001152259, "data_time": 0.0013289451599121094, "grad_norm": 0.025387100083753465, "loss": 0.6860304713249207, "time": 0.42376022338867186, "epoch": 15, "memory": 9465, "step": 167024} +{"lr": 0.0002140463001152259, "data_time": 0.0013043403625488282, "grad_norm": 0.020605178590631112, "loss": 0.6824390769004822, "time": 0.41295361518859863, "epoch": 15, "memory": 9465, "step": 167124} +{"lr": 0.0002140463001152259, "data_time": 0.0016310691833496093, "grad_norm": 0.025282455259002745, "loss": 0.686529690027237, "time": 0.4085672855377197, "epoch": 15, "memory": 9465, "step": 167224} +{"lr": 0.0002140463001152259, "data_time": 0.001294088363647461, "grad_norm": 0.012749508069828153, "loss": 0.6860068261623382, "time": 0.39381725788116456, "epoch": 15, "memory": 9465, "step": 167324} +{"lr": 0.0002140463001152259, "data_time": 0.0013279914855957031, "grad_norm": 0.01689598455559462, "loss": 0.67860706448555, "time": 0.43148863315582275, "epoch": 15, "memory": 9465, "step": 167424} +{"lr": 0.0002140463001152259, "data_time": 0.0013501167297363282, "grad_norm": 0.02227825871377718, "loss": 0.6852222859859467, "time": 0.40629117488861083, "epoch": 15, "memory": 9465, "step": 167524} +{"lr": 0.0002140463001152259, "data_time": 0.002580714225769043, "grad_norm": 0.027043653256259858, "loss": 0.6815916717052459, "time": 0.41361634731292723, "epoch": 15, "memory": 9465, "step": 167624} +{"lr": 0.0002140463001152259, "data_time": 0.0024840593338012694, "grad_norm": 0.022307089995592833, "loss": 0.6874937653541565, "time": 0.38384356498718264, "epoch": 15, "memory": 9465, "step": 167724} +{"lr": 0.0002140463001152259, "data_time": 0.0017174482345581055, "grad_norm": 0.02461586700519547, "loss": 0.6910554945468903, "time": 0.4112581729888916, "epoch": 15, "memory": 9465, "step": 167824} +{"lr": 0.0002140463001152259, "data_time": 0.0015367031097412109, "grad_norm": 0.020715971733443438, "loss": 0.6877492725849151, "time": 0.4556956052780151, "epoch": 15, "memory": 9465, "step": 167924} +{"lr": 0.0002140463001152259, "data_time": 0.0015857934951782227, "grad_norm": 0.01240457467501983, "loss": 0.6887055158615112, "time": 0.4200615406036377, "epoch": 15, "memory": 9465, "step": 168024} +{"lr": 0.0002140463001152259, "data_time": 0.0027471303939819334, "grad_norm": 0.025396501854993404, "loss": 0.686190128326416, "time": 0.4084672451019287, "epoch": 15, "memory": 9465, "step": 168124} +{"lr": 0.0002140463001152259, "data_time": 0.001683354377746582, "grad_norm": 0.023855562787503004, "loss": 0.6874362707138062, "time": 0.4132983207702637, "epoch": 15, "memory": 9465, "step": 168224} +{"lr": 0.0002140463001152259, "data_time": 0.0020859479904174806, "grad_norm": 0.022987095126882195, "loss": 0.6872956335544587, "time": 0.3963721990585327, "epoch": 15, "memory": 9465, "step": 168324} +{"lr": 0.0002140463001152259, "data_time": 0.0013257503509521485, "grad_norm": 0.028298271214589478, "loss": 0.6857425749301911, "time": 0.41259739398956297, "epoch": 15, "memory": 9465, "step": 168424} +{"lr": 0.0002140463001152259, "data_time": 0.0013125419616699218, "grad_norm": 0.030274981260299684, "loss": 0.6908557057380676, "time": 0.41288840770721436, "epoch": 15, "memory": 9465, "step": 168524} +{"lr": 0.0002140463001152259, "data_time": 0.0017639636993408204, "grad_norm": 0.02653438232373446, "loss": 0.6908407628536224, "time": 0.41278464794158937, "epoch": 15, "memory": 9465, "step": 168624} +{"lr": 0.0002140463001152259, "data_time": 0.0012291669845581055, "grad_norm": 0.022173001198098063, "loss": 0.6893778920173645, "time": 0.42564690113067627, "epoch": 15, "memory": 9465, "step": 168724} +{"lr": 0.0002140463001152259, "data_time": 0.001319289207458496, "grad_norm": 0.02520542065612972, "loss": 0.6874444842338562, "time": 0.42061102390289307, "epoch": 15, "memory": 9465, "step": 168824} +{"lr": 0.0002140463001152259, "data_time": 0.0014426231384277344, "grad_norm": 0.02720923819579184, "loss": 0.685785847902298, "time": 0.408914852142334, "epoch": 15, "memory": 9465, "step": 168924} +{"lr": 0.0002140463001152259, "data_time": 0.0013175010681152344, "grad_norm": 0.016094760701525955, "loss": 0.6829691350460052, "time": 0.454237961769104, "epoch": 15, "memory": 9465, "step": 169024} +{"lr": 0.0002140463001152259, "data_time": 0.0018984317779541016, "grad_norm": 0.01600051177665591, "loss": 0.6846305847167968, "time": 0.445867919921875, "epoch": 15, "memory": 9465, "step": 169124} +{"lr": 0.0002140463001152259, "data_time": 0.0013016939163208007, "grad_norm": 0.020676075041410513, "loss": 0.6891857981681824, "time": 0.43398048877716067, "epoch": 15, "memory": 9465, "step": 169224} +{"lr": 0.0002140463001152259, "data_time": 0.0014240503311157226, "grad_norm": 0.014566304977051914, "loss": 0.6912540793418884, "time": 0.39125206470489504, "epoch": 15, "memory": 9465, "step": 169324} +{"lr": 0.0002140463001152259, "data_time": 0.0013763904571533203, "grad_norm": 0.023430117953103034, "loss": 0.6863320827484131, "time": 0.4065514087677002, "epoch": 15, "memory": 9465, "step": 169424} +{"lr": 0.0002140463001152259, "data_time": 0.0014593124389648438, "grad_norm": 0.018793767696479335, "loss": 0.6831027686595916, "time": 0.4316640615463257, "epoch": 15, "memory": 9465, "step": 169524} +{"lr": 0.0002140463001152259, "data_time": 0.0013422250747680664, "grad_norm": 0.03210939399432391, "loss": 0.6876102387905121, "time": 0.45074079036712644, "epoch": 15, "memory": 9465, "step": 169624} +{"lr": 0.0002140463001152259, "data_time": 0.0014653682708740234, "grad_norm": 0.024660276668146252, "loss": 0.6877603650093078, "time": 0.4192806243896484, "epoch": 15, "memory": 9465, "step": 169724} +{"lr": 0.0002140463001152259, "data_time": 0.0013684511184692382, "grad_norm": 0.021273887343704702, "loss": 0.6850216567516327, "time": 0.4201963901519775, "epoch": 15, "memory": 9465, "step": 169824} +{"lr": 0.0002140463001152259, "data_time": 0.0016198158264160156, "grad_norm": 0.027001613425090908, "loss": 0.6882172286510467, "time": 0.41813514232635496, "epoch": 15, "memory": 9465, "step": 169924} +{"lr": 0.0002140463001152259, "data_time": 0.0012517929077148437, "grad_norm": 0.022277281119022517, "loss": 0.6907564401626587, "time": 0.4282623291015625, "epoch": 15, "memory": 9465, "step": 170024} +{"lr": 0.0002140463001152259, "data_time": 0.0013345956802368164, "grad_norm": 0.021735110017471016, "loss": 0.686005711555481, "time": 0.4130054235458374, "epoch": 15, "memory": 9465, "step": 170124} +{"lr": 0.0002140463001152259, "data_time": 0.001416921615600586, "grad_norm": 0.024807228142162786, "loss": 0.685529911518097, "time": 0.40968310832977295, "epoch": 15, "memory": 9465, "step": 170224} +{"lr": 0.0002140463001152259, "data_time": 0.0012783050537109376, "grad_norm": 0.024404172063805164, "loss": 0.6864898681640625, "time": 0.4274731159210205, "epoch": 15, "memory": 9465, "step": 170324} +{"lr": 0.0002140463001152259, "data_time": 0.001424098014831543, "grad_norm": 0.02463686892297119, "loss": 0.6869766891002655, "time": 0.4240144968032837, "epoch": 15, "memory": 9465, "step": 170424} +{"lr": 0.0002140463001152259, "data_time": 0.0018577814102172852, "grad_norm": 0.011882529768627137, "loss": 0.67880619764328, "time": 0.4308481216430664, "epoch": 15, "memory": 9465, "step": 170524} +{"lr": 0.0002140463001152259, "data_time": 0.001288437843322754, "grad_norm": 0.02491335610393435, "loss": 0.68327676653862, "time": 0.4180778741836548, "epoch": 15, "memory": 9465, "step": 170624} +{"lr": 0.0002140463001152259, "data_time": 0.0013918161392211914, "grad_norm": 0.011091677920194342, "loss": 0.6901532351970673, "time": 0.4272172212600708, "epoch": 15, "memory": 9465, "step": 170724} +{"lr": 0.0002140463001152259, "data_time": 0.0015508890151977538, "grad_norm": 0.020376018714159727, "loss": 0.6868217885494232, "time": 0.4707669496536255, "epoch": 15, "memory": 9465, "step": 170824} +{"lr": 0.0002140463001152259, "data_time": 0.0017530679702758788, "grad_norm": 0.027968662232160567, "loss": 0.6947459578514099, "time": 0.4309210300445557, "epoch": 15, "memory": 9465, "step": 170924} +{"lr": 0.0002140463001152259, "data_time": 0.0017899513244628907, "grad_norm": 0.021527378889732062, "loss": 0.6858453929424286, "time": 0.4314690828323364, "epoch": 15, "memory": 9465, "step": 171024} +{"lr": 0.0002140463001152259, "data_time": 0.0014004468917846679, "grad_norm": 0.019454246468376367, "loss": 0.687979382276535, "time": 0.43224637508392333, "epoch": 15, "memory": 9465, "step": 171124} +{"lr": 0.0002140463001152259, "data_time": 0.001273179054260254, "grad_norm": 0.018106868490576744, "loss": 0.6844498813152313, "time": 0.4359687328338623, "epoch": 15, "memory": 9465, "step": 171224} +{"lr": 0.0002140463001152259, "data_time": 0.0019911527633666992, "grad_norm": 0.017930368497036397, "loss": 0.6864727973937989, "time": 0.4729830503463745, "epoch": 15, "memory": 9465, "step": 171324} +{"lr": 0.0002140463001152259, "data_time": 0.001212763786315918, "grad_norm": 0.022242502425797284, "loss": 0.6863435804843903, "time": 0.45276756286621095, "epoch": 15, "memory": 9465, "step": 171424} +{"lr": 0.0002140463001152259, "data_time": 0.0013869047164916993, "grad_norm": 0.016562653821893036, "loss": 0.6861638486385345, "time": 0.43400847911834717, "epoch": 15, "memory": 9465, "step": 171524} +{"lr": 0.0002140463001152259, "data_time": 0.0017216920852661133, "grad_norm": 0.022390550188720226, "loss": 0.6856852114200592, "time": 0.4402602195739746, "epoch": 15, "memory": 9465, "step": 171624} +{"lr": 0.0002140463001152259, "data_time": 0.0013005971908569337, "grad_norm": 0.018840618315152824, "loss": 0.6884231984615325, "time": 0.4193726539611816, "epoch": 15, "memory": 9465, "step": 171724} +{"lr": 0.0002140463001152259, "data_time": 0.0012496471405029296, "grad_norm": 0.021178130363114177, "loss": 0.6872885525226593, "time": 0.40828404426574705, "epoch": 15, "memory": 9465, "step": 171824} +{"lr": 0.0002140463001152259, "data_time": 0.0014006614685058594, "grad_norm": 0.024584313973900863, "loss": 0.6844365000724792, "time": 0.45669419765472413, "epoch": 15, "memory": 9465, "step": 171924} +{"lr": 0.0002140463001152259, "data_time": 0.001261591911315918, "grad_norm": 0.017186284391209485, "loss": 0.6858316779136657, "time": 0.42757914066314695, "epoch": 15, "memory": 9465, "step": 172024} +{"lr": 0.0002140463001152259, "data_time": 0.0011881351470947265, "grad_norm": 0.01898417028714903, "loss": 0.6891905903816223, "time": 0.42606825828552247, "epoch": 15, "memory": 9465, "step": 172124} +{"lr": 0.0002140463001152259, "data_time": 0.0014975786209106446, "grad_norm": 0.026393118500709533, "loss": 0.6917441785335541, "time": 0.40974161624908445, "epoch": 15, "memory": 9465, "step": 172224} +{"lr": 0.0002140463001152259, "data_time": 0.0013780593872070312, "grad_norm": 0.0215091391466558, "loss": 0.6942771732807159, "time": 0.42108609676361086, "epoch": 15, "memory": 9465, "step": 172324} +{"lr": 0.0002140463001152259, "data_time": 0.00132443904876709, "grad_norm": 0.018597495416179298, "loss": 0.6829773545265198, "time": 0.42320957183837893, "epoch": 15, "memory": 9465, "step": 172424} +{"lr": 0.0002140463001152259, "data_time": 0.0012312889099121093, "grad_norm": 0.02508988493354991, "loss": 0.6888750612735748, "time": 0.39652304649353026, "epoch": 15, "memory": 9465, "step": 172524} +{"lr": 0.0002140463001152259, "data_time": 0.00141451358795166, "grad_norm": 0.03179497700184584, "loss": 0.6835142970085144, "time": 0.4224919557571411, "epoch": 15, "memory": 9465, "step": 172624} +{"lr": 0.0002140463001152259, "data_time": 0.005132699012756347, "grad_norm": 0.024911126727238296, "loss": 0.6877717256546021, "time": 0.42191898822784424, "epoch": 15, "memory": 9465, "step": 172724} +{"accuracy/top1": 100.0, "data_time": 0.2773989509133732, "time": 0.540760404923383, "step": 15} +{"lr": 0.00015498214331265904, "data_time": 0.04796533584594727, "grad_norm": 0.012402570992708205, "loss": 0.6874396622180938, "time": 0.42837932109832766, "epoch": 16, "memory": 9465, "step": 172840} +{"lr": 0.00015498214331265904, "data_time": 0.0013519287109375, "grad_norm": 0.026468004425987603, "loss": 0.6849124431610107, "time": 0.4147411584854126, "epoch": 16, "memory": 9465, "step": 172940} +{"lr": 0.00015498214331265904, "data_time": 0.0013030290603637696, "grad_norm": 0.021316367224790154, "loss": 0.6918719947338104, "time": 0.43860588073730467, "epoch": 16, "memory": 9465, "step": 173040} +{"lr": 0.00015498214331265904, "data_time": 0.0022736787796020508, "grad_norm": 0.016826165071688593, "loss": 0.6860176026821136, "time": 0.42089903354644775, "epoch": 16, "memory": 9465, "step": 173140} +{"lr": 0.00015498214331265904, "data_time": 0.0016296625137329102, "grad_norm": 0.020871743821771815, "loss": 0.6868066191673279, "time": 0.40884759426116946, "epoch": 16, "memory": 9465, "step": 173240} +{"lr": 0.00015498214331265904, "data_time": 0.0013618230819702148, "grad_norm": 0.031826614792225885, "loss": 0.6908757865428925, "time": 0.43290183544158933, "epoch": 16, "memory": 9465, "step": 173340} +{"lr": 0.00015498214331265904, "data_time": 0.0016141414642333984, "grad_norm": 0.03186066672205925, "loss": 0.6856403648853302, "time": 0.46909973621368406, "epoch": 16, "memory": 9465, "step": 173440} +{"lr": 0.00015498214331265904, "data_time": 0.0013225793838500977, "grad_norm": 0.025826570627395996, "loss": 0.6869646489620209, "time": 0.44034247398376464, "epoch": 16, "memory": 9465, "step": 173540} +{"lr": 0.00015498214331265904, "data_time": 0.0019741058349609375, "grad_norm": 0.01851236880174838, "loss": 0.6882412612438202, "time": 0.41535918712615966, "epoch": 16, "memory": 9465, "step": 173640} +{"lr": 0.00015498214331265904, "data_time": 0.0015933990478515625, "grad_norm": 0.029405278526246546, "loss": 0.684426361322403, "time": 0.3884117841720581, "epoch": 16, "memory": 9465, "step": 173740} +{"lr": 0.00015498214331265904, "data_time": 0.0013489246368408204, "grad_norm": 0.024283888889476656, "loss": 0.6869985878467559, "time": 0.3578831672668457, "epoch": 16, "memory": 9465, "step": 173840} +{"lr": 0.00015498214331265904, "data_time": 0.0014805555343627929, "grad_norm": 0.02539905053563416, "loss": 0.6883923649787903, "time": 0.3797483444213867, "epoch": 16, "memory": 9465, "step": 173940} +{"lr": 0.00015498214331265904, "data_time": 0.0012373924255371094, "grad_norm": 0.01833083345554769, "loss": 0.6885931849479675, "time": 0.37533481121063234, "epoch": 16, "memory": 9465, "step": 174040} +{"lr": 0.00015498214331265904, "data_time": 0.0011394739151000977, "grad_norm": 0.016121975146234035, "loss": 0.6888462603092194, "time": 0.4132434368133545, "epoch": 16, "memory": 9465, "step": 174140} +{"lr": 0.00015498214331265904, "data_time": 0.0015220403671264648, "grad_norm": 0.02277687704190612, "loss": 0.6871201574802399, "time": 0.36899352073669434, "epoch": 16, "memory": 9465, "step": 174240} +{"lr": 0.00015498214331265904, "data_time": 0.0019240856170654296, "grad_norm": 0.01913366199587472, "loss": 0.6838581979274749, "time": 0.411507248878479, "epoch": 16, "memory": 9465, "step": 174340} +{"lr": 0.00015498214331265904, "data_time": 0.0014511585235595704, "grad_norm": 0.015426935977302492, "loss": 0.6843148469924927, "time": 0.3962219476699829, "epoch": 16, "memory": 9465, "step": 174440} +{"lr": 0.00015498214331265904, "data_time": 0.0014513731002807617, "grad_norm": 0.025891577964648603, "loss": 0.6853933334350586, "time": 0.4263045310974121, "epoch": 16, "memory": 9465, "step": 174540} +{"lr": 0.00015498214331265904, "data_time": 0.0012848854064941406, "grad_norm": 0.016526716179214418, "loss": 0.6849218487739563, "time": 0.3768187999725342, "epoch": 16, "memory": 9465, "step": 174640} +{"lr": 0.00015498214331265904, "data_time": 0.001394510269165039, "grad_norm": 0.019196415692567824, "loss": 0.6875948488712311, "time": 0.38160154819488523, "epoch": 16, "memory": 9465, "step": 174740} +{"lr": 0.00015498214331265904, "data_time": 0.0012467145919799806, "grad_norm": 0.019742982322350146, "loss": 0.68916255235672, "time": 0.41026127338409424, "epoch": 16, "memory": 9465, "step": 174840} +{"lr": 0.00015498214331265904, "data_time": 0.001456284523010254, "grad_norm": 0.032116131024667996, "loss": 0.6809656322002411, "time": 0.3904338598251343, "epoch": 16, "memory": 9465, "step": 174940} +{"lr": 0.00015498214331265904, "data_time": 0.0020202875137329103, "grad_norm": 0.023475692491047084, "loss": 0.6892821431159973, "time": 0.4652515411376953, "epoch": 16, "memory": 9465, "step": 175040} +{"lr": 0.00015498214331265904, "data_time": 0.001911473274230957, "grad_norm": 0.025645952252671123, "loss": 0.6807098090648651, "time": 0.39325723648071287, "epoch": 16, "memory": 9465, "step": 175140} +{"lr": 0.00015498214331265904, "data_time": 0.0021973848342895508, "grad_norm": 0.019104114378569646, "loss": 0.6858468115329742, "time": 0.4022464036941528, "epoch": 16, "memory": 9465, "step": 175240} +{"lr": 0.00015498214331265904, "data_time": 0.002050304412841797, "grad_norm": 0.018378050858154894, "loss": 0.6855072259902955, "time": 0.38553872108459475, "epoch": 16, "memory": 9465, "step": 175340} +{"lr": 0.00015498214331265904, "data_time": 0.001916360855102539, "grad_norm": 0.02121393126435578, "loss": 0.6814954698085784, "time": 0.3946491479873657, "epoch": 16, "memory": 9465, "step": 175440} +{"lr": 0.00015498214331265904, "data_time": 0.0015674114227294921, "grad_norm": 0.022905727766919882, "loss": 0.6902084887027741, "time": 0.3887176036834717, "epoch": 16, "memory": 9465, "step": 175540} +{"lr": 0.00015498214331265904, "data_time": 0.0019622564315795897, "grad_norm": 0.02395902179996483, "loss": 0.6900645136833191, "time": 0.3686830997467041, "epoch": 16, "memory": 9465, "step": 175640} +{"lr": 0.00015498214331265904, "data_time": 0.0019403457641601562, "grad_norm": 0.0249115219572559, "loss": 0.6885612308979034, "time": 0.38416781425476076, "epoch": 16, "memory": 9465, "step": 175740} +{"lr": 0.00015498214331265904, "data_time": 0.0014475345611572265, "grad_norm": 0.0305089369736379, "loss": 0.6847354888916015, "time": 0.40173022747039794, "epoch": 16, "memory": 9465, "step": 175840} +{"lr": 0.00015498214331265904, "data_time": 0.0013242483139038086, "grad_norm": 0.030720880068838597, "loss": 0.6836639523506165, "time": 0.4245003700256348, "epoch": 16, "memory": 9465, "step": 175940} +{"lr": 0.00015498214331265904, "data_time": 0.0015934944152832032, "grad_norm": 0.021535947010852395, "loss": 0.6820919990539551, "time": 0.3820359230041504, "epoch": 16, "memory": 9465, "step": 176040} +{"lr": 0.00015498214331265904, "data_time": 0.002273297309875488, "grad_norm": 0.027006560820154846, "loss": 0.6867980062961578, "time": 0.37694535255432127, "epoch": 16, "memory": 9465, "step": 176140} +{"lr": 0.00015498214331265904, "data_time": 0.0013098716735839844, "grad_norm": 0.01856318542268127, "loss": 0.689912486076355, "time": 0.397737455368042, "epoch": 16, "memory": 9465, "step": 176240} +{"lr": 0.00015498214331265904, "data_time": 0.0013541936874389648, "grad_norm": 0.020981867541559042, "loss": 0.6854382276535034, "time": 0.3872718095779419, "epoch": 16, "memory": 9465, "step": 176340} +{"lr": 0.00015498214331265904, "data_time": 0.0013765335083007813, "grad_norm": 0.026773387455614285, "loss": 0.6820409655570984, "time": 0.37392315864562986, "epoch": 16, "memory": 9465, "step": 176440} diff --git a/swin_tiny_4xb256_1e-3lr_5m/20230531_183311/vis_data/vis_image/0000000.png_1.png b/swin_tiny_4xb256_1e-3lr_5m/20230531_183311/vis_data/vis_image/0000000.png_1.png new file mode 100644 index 0000000000000000000000000000000000000000..b1d5db9b91db71ddcbbba2e61273dd6141d70f3d Binary files /dev/null and b/swin_tiny_4xb256_1e-3lr_5m/20230531_183311/vis_data/vis_image/0000000.png_1.png differ diff --git a/swin_tiny_4xb256_1e-3lr_5m/20230531_183311/vis_data/vis_image/0000000.png_10.png b/swin_tiny_4xb256_1e-3lr_5m/20230531_183311/vis_data/vis_image/0000000.png_10.png new file mode 100644 index 0000000000000000000000000000000000000000..934769c1f2bb2402694533d16138c8dbc1ce4a97 Binary files /dev/null and b/swin_tiny_4xb256_1e-3lr_5m/20230531_183311/vis_data/vis_image/0000000.png_10.png differ diff --git a/swin_tiny_4xb256_1e-3lr_5m/20230531_183311/vis_data/vis_image/0000000.png_11.png b/swin_tiny_4xb256_1e-3lr_5m/20230531_183311/vis_data/vis_image/0000000.png_11.png new file mode 100644 index 0000000000000000000000000000000000000000..ae323dfb2f99afa256eed926b072e5d438f42b32 Binary files /dev/null and b/swin_tiny_4xb256_1e-3lr_5m/20230531_183311/vis_data/vis_image/0000000.png_11.png differ diff --git a/swin_tiny_4xb256_1e-3lr_5m/20230531_183311/vis_data/vis_image/0000000.png_12.png b/swin_tiny_4xb256_1e-3lr_5m/20230531_183311/vis_data/vis_image/0000000.png_12.png new file mode 100644 index 0000000000000000000000000000000000000000..934769c1f2bb2402694533d16138c8dbc1ce4a97 Binary files /dev/null and b/swin_tiny_4xb256_1e-3lr_5m/20230531_183311/vis_data/vis_image/0000000.png_12.png differ diff --git a/swin_tiny_4xb256_1e-3lr_5m/20230531_183311/vis_data/vis_image/0000000.png_13.png b/swin_tiny_4xb256_1e-3lr_5m/20230531_183311/vis_data/vis_image/0000000.png_13.png new file mode 100644 index 0000000000000000000000000000000000000000..934769c1f2bb2402694533d16138c8dbc1ce4a97 Binary files /dev/null and b/swin_tiny_4xb256_1e-3lr_5m/20230531_183311/vis_data/vis_image/0000000.png_13.png differ diff --git a/swin_tiny_4xb256_1e-3lr_5m/20230531_183311/vis_data/vis_image/0000000.png_14.png b/swin_tiny_4xb256_1e-3lr_5m/20230531_183311/vis_data/vis_image/0000000.png_14.png new file mode 100644 index 0000000000000000000000000000000000000000..934769c1f2bb2402694533d16138c8dbc1ce4a97 Binary files /dev/null and b/swin_tiny_4xb256_1e-3lr_5m/20230531_183311/vis_data/vis_image/0000000.png_14.png differ diff --git a/swin_tiny_4xb256_1e-3lr_5m/20230531_183311/vis_data/vis_image/0000000.png_15.png b/swin_tiny_4xb256_1e-3lr_5m/20230531_183311/vis_data/vis_image/0000000.png_15.png new file mode 100644 index 0000000000000000000000000000000000000000..934769c1f2bb2402694533d16138c8dbc1ce4a97 Binary files /dev/null and b/swin_tiny_4xb256_1e-3lr_5m/20230531_183311/vis_data/vis_image/0000000.png_15.png differ diff --git a/swin_tiny_4xb256_1e-3lr_5m/20230531_183311/vis_data/vis_image/0000000.png_2.png b/swin_tiny_4xb256_1e-3lr_5m/20230531_183311/vis_data/vis_image/0000000.png_2.png new file mode 100644 index 0000000000000000000000000000000000000000..934769c1f2bb2402694533d16138c8dbc1ce4a97 Binary files /dev/null and b/swin_tiny_4xb256_1e-3lr_5m/20230531_183311/vis_data/vis_image/0000000.png_2.png differ diff --git a/swin_tiny_4xb256_1e-3lr_5m/20230531_183311/vis_data/vis_image/0000000.png_3.png b/swin_tiny_4xb256_1e-3lr_5m/20230531_183311/vis_data/vis_image/0000000.png_3.png new file mode 100644 index 0000000000000000000000000000000000000000..934769c1f2bb2402694533d16138c8dbc1ce4a97 Binary files /dev/null and b/swin_tiny_4xb256_1e-3lr_5m/20230531_183311/vis_data/vis_image/0000000.png_3.png differ diff --git a/swin_tiny_4xb256_1e-3lr_5m/20230531_183311/vis_data/vis_image/0000000.png_4.png b/swin_tiny_4xb256_1e-3lr_5m/20230531_183311/vis_data/vis_image/0000000.png_4.png new file mode 100644 index 0000000000000000000000000000000000000000..934769c1f2bb2402694533d16138c8dbc1ce4a97 Binary files /dev/null and b/swin_tiny_4xb256_1e-3lr_5m/20230531_183311/vis_data/vis_image/0000000.png_4.png differ diff --git a/swin_tiny_4xb256_1e-3lr_5m/20230531_183311/vis_data/vis_image/0000000.png_5.png b/swin_tiny_4xb256_1e-3lr_5m/20230531_183311/vis_data/vis_image/0000000.png_5.png new file mode 100644 index 0000000000000000000000000000000000000000..b1d5db9b91db71ddcbbba2e61273dd6141d70f3d Binary files /dev/null and b/swin_tiny_4xb256_1e-3lr_5m/20230531_183311/vis_data/vis_image/0000000.png_5.png differ diff --git a/swin_tiny_4xb256_1e-3lr_5m/20230531_183311/vis_data/vis_image/0000000.png_6.png b/swin_tiny_4xb256_1e-3lr_5m/20230531_183311/vis_data/vis_image/0000000.png_6.png new file mode 100644 index 0000000000000000000000000000000000000000..934769c1f2bb2402694533d16138c8dbc1ce4a97 Binary files /dev/null and b/swin_tiny_4xb256_1e-3lr_5m/20230531_183311/vis_data/vis_image/0000000.png_6.png differ diff --git a/swin_tiny_4xb256_1e-3lr_5m/20230531_183311/vis_data/vis_image/0000000.png_7.png b/swin_tiny_4xb256_1e-3lr_5m/20230531_183311/vis_data/vis_image/0000000.png_7.png new file mode 100644 index 0000000000000000000000000000000000000000..934769c1f2bb2402694533d16138c8dbc1ce4a97 Binary files /dev/null and b/swin_tiny_4xb256_1e-3lr_5m/20230531_183311/vis_data/vis_image/0000000.png_7.png differ diff --git a/swin_tiny_4xb256_1e-3lr_5m/20230531_183311/vis_data/vis_image/0000000.png_8.png b/swin_tiny_4xb256_1e-3lr_5m/20230531_183311/vis_data/vis_image/0000000.png_8.png new file mode 100644 index 0000000000000000000000000000000000000000..934769c1f2bb2402694533d16138c8dbc1ce4a97 Binary files /dev/null and b/swin_tiny_4xb256_1e-3lr_5m/20230531_183311/vis_data/vis_image/0000000.png_8.png differ diff --git a/swin_tiny_4xb256_1e-3lr_5m/20230531_183311/vis_data/vis_image/0000000.png_9.png b/swin_tiny_4xb256_1e-3lr_5m/20230531_183311/vis_data/vis_image/0000000.png_9.png new file mode 100644 index 0000000000000000000000000000000000000000..b1d5db9b91db71ddcbbba2e61273dd6141d70f3d Binary files /dev/null and b/swin_tiny_4xb256_1e-3lr_5m/20230531_183311/vis_data/vis_image/0000000.png_9.png differ diff --git a/swin_tiny_4xb256_1e-3lr_5m/epoch_1.pth b/swin_tiny_4xb256_1e-3lr_5m/epoch_1.pth new file mode 100644 index 0000000000000000000000000000000000000000..8a93fac8522d1c839c5a1c77f863438c40ca2b86 --- /dev/null +++ b/swin_tiny_4xb256_1e-3lr_5m/epoch_1.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:d3ab642cb3b0a342a4828547a8ae2034f7db7d15fa20908d3d8e7d5f36141c15 +size 331774401 diff --git a/swin_tiny_4xb256_1e-3lr_5m/epoch_10.pth b/swin_tiny_4xb256_1e-3lr_5m/epoch_10.pth new file mode 100644 index 0000000000000000000000000000000000000000..a54a51034f60ffa2d63a11b01a865a99c81bbaaa --- /dev/null +++ b/swin_tiny_4xb256_1e-3lr_5m/epoch_10.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:12e118fa000fc0d37da59cbc4e59b0d49d6ed08fd6ef9e5ef133292e5828b206 +size 341169409 diff --git a/swin_tiny_4xb256_1e-3lr_5m/epoch_11.pth b/swin_tiny_4xb256_1e-3lr_5m/epoch_11.pth new file mode 100644 index 0000000000000000000000000000000000000000..24e53327ff36032258f7bbc6533fc8a939e7005c --- /dev/null +++ b/swin_tiny_4xb256_1e-3lr_5m/epoch_11.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:7ec30ce5cbd8311ed7433728c63f8877f363dc9e8e3ec34875340f4f818c6ce1 +size 342199105 diff --git a/swin_tiny_4xb256_1e-3lr_5m/epoch_12.pth b/swin_tiny_4xb256_1e-3lr_5m/epoch_12.pth new file mode 100644 index 0000000000000000000000000000000000000000..404d40b7e669512ba0f152a7fd548cc72d63e9c5 --- /dev/null +++ b/swin_tiny_4xb256_1e-3lr_5m/epoch_12.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:3ac37debc42098112026ae04e34dbdedb1d27b6ec488f5618fd3ba3f9d17a8e7 +size 343241665 diff --git a/swin_tiny_4xb256_1e-3lr_5m/epoch_13.pth b/swin_tiny_4xb256_1e-3lr_5m/epoch_13.pth new file mode 100644 index 0000000000000000000000000000000000000000..76ee7b5a68744bbc40171cb173c115543d551db3 --- /dev/null +++ b/swin_tiny_4xb256_1e-3lr_5m/epoch_13.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:37aa29d32adfeaffd426dffb9fb753b4f7ab3a928b4e3d034a2747727eb65a21 +size 344296193 diff --git a/swin_tiny_4xb256_1e-3lr_5m/epoch_14.pth b/swin_tiny_4xb256_1e-3lr_5m/epoch_14.pth new file mode 100644 index 0000000000000000000000000000000000000000..439382e1d78ebab7196d6df84d985968fc18670b --- /dev/null +++ b/swin_tiny_4xb256_1e-3lr_5m/epoch_14.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:576dc8874c9ef5e9c77aafec0ed96975368713eddd8a147c31bb68c2331b3c71 +size 345328449 diff --git a/swin_tiny_4xb256_1e-3lr_5m/epoch_15.pth b/swin_tiny_4xb256_1e-3lr_5m/epoch_15.pth new file mode 100644 index 0000000000000000000000000000000000000000..5140c47dcfdd9b353bffe07380516cc7ace6a9bb --- /dev/null +++ b/swin_tiny_4xb256_1e-3lr_5m/epoch_15.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:35783257385e5caf28eb207d7a44ff80e4e766640da081d7911bcca9bc9e1ca3 +size 346382209 diff --git a/swin_tiny_4xb256_1e-3lr_5m/epoch_2.pth b/swin_tiny_4xb256_1e-3lr_5m/epoch_2.pth new file mode 100644 index 0000000000000000000000000000000000000000..c16a07be233f3d02e63ffe0e3a37518d08d40e04 --- /dev/null +++ b/swin_tiny_4xb256_1e-3lr_5m/epoch_2.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ca7dd3cc9f0238af4d06b296aaaf4bfa841c8a269a1cd9c586caabed3a9c3a34 +size 332795137 diff --git a/swin_tiny_4xb256_1e-3lr_5m/epoch_3.pth b/swin_tiny_4xb256_1e-3lr_5m/epoch_3.pth new file mode 100644 index 0000000000000000000000000000000000000000..df51f6fc5e1c209e550524e33077b0c5913d6475 --- /dev/null +++ b/swin_tiny_4xb256_1e-3lr_5m/epoch_3.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:6ee1ba084875df83160521f3f9d43184b1f5a19945ae4491a089d8cd44a88ba9 +size 333837889 diff --git a/swin_tiny_4xb256_1e-3lr_5m/epoch_4.pth b/swin_tiny_4xb256_1e-3lr_5m/epoch_4.pth new file mode 100644 index 0000000000000000000000000000000000000000..9ca75f1cc2e7afd33438fd04f93d2877fd937939 --- /dev/null +++ b/swin_tiny_4xb256_1e-3lr_5m/epoch_4.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:f595b5a63cce9ff9ed40db9fbd5ba31c6208d09d3f9547a5e1dd4b0219db5ae9 +size 334857217 diff --git a/swin_tiny_4xb256_1e-3lr_5m/epoch_5.pth b/swin_tiny_4xb256_1e-3lr_5m/epoch_5.pth new file mode 100644 index 0000000000000000000000000000000000000000..b72ddf4744aafcce1d25855d5e4fa1ae2478ce73 --- /dev/null +++ b/swin_tiny_4xb256_1e-3lr_5m/epoch_5.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:53824d043dc35f9e74c338f6b638cf81f80fdf18ad1a2e6af135e6b3ff81bcac +size 335910785 diff --git a/swin_tiny_4xb256_1e-3lr_5m/epoch_6.pth b/swin_tiny_4xb256_1e-3lr_5m/epoch_6.pth new file mode 100644 index 0000000000000000000000000000000000000000..84a386852ebe132520b01f758d2953404bfaa01d --- /dev/null +++ b/swin_tiny_4xb256_1e-3lr_5m/epoch_6.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:547447ecb5308b51c16109cae9cb0b4e8c0e926615c67b095956150b375f3d9a +size 336952705 diff --git a/swin_tiny_4xb256_1e-3lr_5m/epoch_7.pth b/swin_tiny_4xb256_1e-3lr_5m/epoch_7.pth new file mode 100644 index 0000000000000000000000000000000000000000..bde1fed6379d521a16b1ac67650f9be7d304163b --- /dev/null +++ b/swin_tiny_4xb256_1e-3lr_5m/epoch_7.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:487b77442a086527a9d6c7076a5b2fd79085365cce7ad592ff83bf9d633e91f4 +size 338006913 diff --git a/swin_tiny_4xb256_1e-3lr_5m/epoch_8.pth b/swin_tiny_4xb256_1e-3lr_5m/epoch_8.pth new file mode 100644 index 0000000000000000000000000000000000000000..15828aba73d3d315d0e079792867c9d1a2b05869 --- /dev/null +++ b/swin_tiny_4xb256_1e-3lr_5m/epoch_8.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:6ecbf62e36c739dd98599b8c5949c1fc83e8c356053a3d4cbd286e4bfee3195f +size 339049409 diff --git a/swin_tiny_4xb256_1e-3lr_5m/epoch_9.pth b/swin_tiny_4xb256_1e-3lr_5m/epoch_9.pth new file mode 100644 index 0000000000000000000000000000000000000000..b66582e0db008076e04f2f8c115dff0e378388b1 --- /dev/null +++ b/swin_tiny_4xb256_1e-3lr_5m/epoch_9.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:bd355768e851b08131c046cc65dbaca13298c7b995bf1b91f9feaab4a38fbda9 +size 340103553 diff --git a/swin_tiny_4xb256_1e-3lr_5m/last_checkpoint b/swin_tiny_4xb256_1e-3lr_5m/last_checkpoint new file mode 100644 index 0000000000000000000000000000000000000000..a764d0af1406c97c6be62e754e503372964f9690 --- /dev/null +++ b/swin_tiny_4xb256_1e-3lr_5m/last_checkpoint @@ -0,0 +1 @@ +/mnt/petrelfs/luzeyu/workspace/fakebench/mmpretrain/workdir/swin_tiny_4xb256_1e-3lr_5m/epoch_15.pth \ No newline at end of file diff --git a/swin_tiny_4xb256_1e-3lr_5m/swin_tiny_4xb256_fake5m.py b/swin_tiny_4xb256_1e-3lr_5m/swin_tiny_4xb256_fake5m.py new file mode 100644 index 0000000000000000000000000000000000000000..2f2d04453ec2c0ed6c5c62055a8b360ce00c32d5 --- /dev/null +++ b/swin_tiny_4xb256_1e-3lr_5m/swin_tiny_4xb256_fake5m.py @@ -0,0 +1,174 @@ +optim_wrapper = dict( + optimizer=dict( + type='AdamW', + lr=0.001, + weight_decay=0.05, + eps=1e-08, + betas=(0.9, 0.999), + _scope_='mmpretrain'), + paramwise_cfg=dict( + norm_decay_mult=0.0, + bias_decay_mult=0.0, + flat_decay_mult=0.0, + custom_keys=dict({ + '.absolute_pos_embed': dict(decay_mult=0.0), + '.relative_position_bias_table': dict(decay_mult=0.0) + })), + type='AmpOptimWrapper', + dtype='bfloat16', + clip_grad=dict(max_norm=5.0)) +param_scheduler = [ + dict(type='CosineAnnealingLR', eta_min=1e-05, by_epoch=True, begin=0) +] +train_cfg = dict(by_epoch=True, max_epochs=20, val_interval=1) +val_cfg = dict() +test_cfg = dict() +auto_scale_lr = dict(base_batch_size=1024) +model = dict( + type='ImageClassifier', + backbone=dict( + type='SwinTransformer', arch='tiny', img_size=224, drop_path_rate=0.2), + neck=dict(type='GlobalAveragePooling'), + head=dict( + type='LinearClsHead', + num_classes=2, + in_channels=768, + init_cfg=None, + loss=dict( + type='LabelSmoothLoss', label_smooth_val=0.1, mode='original'), + cal_acc=False), + init_cfg=[ + dict(type='TruncNormal', layer='Linear', std=0.02, bias=0.0), + dict(type='Constant', layer='LayerNorm', val=1.0, bias=0.0) + ], + train_cfg=None) +dataset_type = 'CustomDataset' +data_preprocessor = dict( + num_classes=2, + mean=[123.675, 116.28, 103.53], + std=[58.395, 57.12, 57.375], + to_rgb=True) +bgr_mean = [103.53, 116.28, 123.675] +bgr_std = [57.375, 57.12, 58.395] +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') +] +train_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=128, + num_workers=10, + dataset=dict( + type='CustomDataset', + data_root='', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/train/stablediffusionV1-5R2-dpmsolver-25-5m.csv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='RandomResizedCrop', + scale=224, + backend='pillow', + interpolation='bicubic'), + dict(type='RandomFlip', prob=0.5, direction='horizontal'), + dict(type='PackInputs') + ]), + sampler=dict(type='DefaultSampler', shuffle=True)) +val_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +val_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +test_dataloader = dict( + pin_memory=True, + persistent_workers=True, + collate_fn=dict(type='default_collate'), + batch_size=256, + num_workers=10, + dataset=dict( + type='CustomDataset', + data_root='/mnt/petrelfs/luzeyu/workspace/fakebench/dataset', + ann_file= + '/mnt/petrelfs/luzeyu/workspace/fakebench/dataset/meta/val/stablediffusionV1-5R2-dpmsolver-25-1w.tsv', + pipeline=[ + dict(type='LoadImageFromFile'), + dict( + type='ResizeEdge', + scale=256, + edge='short', + backend='pillow', + interpolation='bicubic'), + dict(type='CenterCrop', crop_size=224), + dict(type='PackInputs') + ]), + sampler=dict(type='DefaultSampler', shuffle=False)) +test_evaluator = [ + dict(type='Accuracy', topk=1), + dict(type='SingleLabelMetric', average=None) +] +default_scope = 'mmpretrain' +default_hooks = dict( + timer=dict(type='IterTimerHook'), + logger=dict(type='LoggerHook', interval=100), + param_scheduler=dict(type='ParamSchedulerHook'), + checkpoint=dict(type='CheckpointHook', interval=1), + sampler_seed=dict(type='DistSamplerSeedHook'), + visualization=dict(type='VisualizationHook', enable=True)) +env_cfg = dict( + cudnn_benchmark=True, + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), + dist_cfg=dict(backend='nccl')) +vis_backends = [dict(type='LocalVisBackend')] +visualizer = dict( + type='UniversalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + dict(type='TensorboardVisBackend') + ]) +log_level = 'INFO' +load_from = None +resume = False +randomness = dict(seed=None, deterministic=False) +launcher = 'slurm' +work_dir = 'workdir/swin_tiny_4xb256_1e-3lr_5m'